Docstoc

Insect Repellents-Principles, Method, and Use

Document Sample
Insect Repellents-Principles, Method, and Use Powered By Docstoc
					q 2006 by Taylor & Francis Group, LLC
q 2006 by Taylor & Francis Group, LLC
q 2006 by Taylor & Francis Group, LLC
q 2006 by Taylor & Francis Group, LLC
                                         Dedicated to
My dear four brothers and four sisters; my beautiful and loving wife, Natalie, and our wonderful
        children: Ameena, Adam and David; and to the memory of my beloved parents


                 Philip Granett, Rutgers University, New Brunswick, New Jersey, USA

               Carroll N. Smith, Harry K. Gouck, T. P. McGovern, and Carl E. Schreck,
               Agricultural Research Service, U.S. Department of Agriculture; Orlando,
                     Florida; Beltsville, Maryland; and Gainesville, Florida, USA

           Robert N. McCulloch and Douglas F. Waterhouse, Commonwealth Scientific
     and Industrial Research Organisation, Canberra, Australian Capital Territory, Australia


                 Pioneers, leaders, intellects, and good friends of insect repellent science




q 2006 by Taylor & Francis Group, LLC
q 2006 by Taylor & Francis Group, LLC
Preface

The use of repellent products to prevent insect and arthropod bites is probably proportional to the public
perception of the threat from biting arthropods, whether the threat is from annoyance or from the risk of
disease. The connection between perception and use is logical when one considers that repellents are
generally used as personal protection. It is the individual who usually decides whether or not to use a
repellent, what kind of repellent to use, and how much to apply. The application to the individual makes
entomological sense, in that the countermeasure is applied exactly where it is needed. On the other hand,
the application by the individual presents a challenge to the professional who must educate and inform
people with widely varied backgrounds on how to best protect themselves from biting arthropods.
   Insect and arthropod repellents are usually the first line of defense because they require no large
equipment, no organized effort of community vector control, and they distribute the responsibility for
protection to the individual. Today, there is great public concern throughout the world about vector-
borne pathogens as human ecology seems to favor outbreaks of diseases as varied as West Nile fever,
dengue, Lyme disease, malaria, leishmaniasis, and tick-borne encephalitis. In contrast to community
vector control programs, insect and arthropod repellents give the individual control over exposure to
biting arthropods. Professional researchers in public health are also interested in the development and use
of repellents given the increasingly complicated requirements for pesticide use, the high cost of
developing effective prophylactic vaccines and drugs, and the increase in incidences of arthropod-
borne diseases.
   All of the insect and arthropod repellent literature is in the form of individual articles, reviews,
symposia, commercial literature, book chapters in medical entomology texts, etc. We are only aware of
two volumes dedicated solely to repellents, and those were handbooks published by the U.S. Department
of Agriculture (USDA) in the 1950s. Our objective was to provide a one-volume source for most aspects
of the development and use of repellent products designed to protect people from biting arthropods. As
the title implies, parts of this book cover the theory and science (principles), the means for advancing the
particular area (use of standard methods for future product development and testing), and the
implications for effective protection of people from biting arthropods.
   Although most of the writing is technical, the informed public, physicians, public health officials, and
other nonspecialists will find this book easily comprehensible. We hope that the following groups will get
specific benefit from the book: The public will be able to choose the proper repellent product for their
situation and use it more effectively. Medical professionals will be able to make better recommendations
to patients who are seeking safe and effective means of preventing arthropod bites and arthropod-borne
diseases in particular situations. Public health personnel will be able to integrate repellents more
effectively into programs to limit arthropod-borne diseases and better inform travelers about protection in
unfamiliar parts of the world. Medical entomologists will be able to perform evaluations with greater
knowledge of theoretical concerns and using more standard techniques. Hopefully, the two appendices
will be useful to those who want to identify common chemicals and organisms used in this line of
research.
   While editing this book, we found that the very word “repellent” was used with many different
meanings. Although there is a rich vocabulary of terms for chemicals affecting arthropod behavior, none
combined the specificity about the feeding process while retaining the general application to the many
different aspects of that process. With some reluctance, we introduce the term “phagomone” to fill this
gap and we suggest that the term “repellent” be restricted to products, rather than to any technical
description of a particular chemical that affects behavior. Phagomone designates any chemical that
influences the arthropod feeding process, whether on the side of facilitating feeding (attractants,
phagostimulants, host-recognition factors, etc.) or on the side of discouraging feeding (irritants,




q 2006 by Taylor & Francis Group, LLC
disorienting toxicants, blockers of receptors, compounds with gustatory effect or with olfactory effect,
etc.). People have been creative in the use of phagomones but much less active in performing the science
to understand how these chemicals work. Use of the term should allow researchers to discuss their work
more accurately, particularly when they are at the stage when they do not completely understand the
exact aspect of feeding behavior disrupted or enhanced by the phagomone.
   The first part of the book treats some of the basic principles behind the use of repellent products.
Starting with a history of repellent product development that has led to the formulations in use today and
proposed for the future, this part proceeds to a discussion of terminology that attempts to specify what has
been a confusing vocabulary used in the field. Some of the biological variety of phagomones as they
function in nature are presented in two chapters on naturally occurring compounds in vertebrates,
including humans.
   The chapters in the second part address the methods used to assess the activity of phagomones and
repellent products. The authors of these chapters present many careers’ worth of experience in this field,
but the reader will soon see that the experts’ opinions do not always correspond to each other. What seem
like standard methods to one author might lack sufficient statistical rigor for another. Whereas one school
of thought might advocate the use of human trials in the field, another advocates the use of animal models
under controlled conditions. The reader should be able to gain a good appreciation of the variety of
purposes served by repellent bioassays and then be better prepared to evaluate data and design new tests.
Apart from traditional bioassays of biting behavior, this part also presents what may be the cutting edge
of repellent testing: automated tests of the arthropods’ responses, the use of computer models of
chemistry, and the use of molecular biology methods.
   The third part of the book concentrates on repellent products that have come to market at one time or
another. Following a thoughtful discussion of the process of testing formulated products, the part
includes comprehensive reviews of both natural and synthetic active ingredients. Currently, the most
important active ingredients are deet, Picaridin, PMD, DEPA, and IR3535, but older active ingredients
were very useful in their time. Some of the older active ingredients are still used as additives that have a
somewhat synergistic effect on product performance. The botanical active ingredients are very
interesting in their variety and origins, illustrating Gene Gerberg’s wise comment that when it comes
to repellents, “one size does not fit all.”
   The fourth and final part of the book deals with the concept of the use of insect repellents. User
acceptability and public perceptions of insect repellents are discussed at length because a repellent
product cannot perform well if it is not used. The great variety of commercially available products shows
how the business community continues to try to perfect its efforts and satisfy what the public considers its
needs. The review of global regulatory procedures is just an introduction to this process, but it
emphasizes the supreme importance of regulation of products that are applied directly to the skin. We
conclude the volume with an epilogue in which we indulge ourselves in some speculation on where the
fields of phagomone and repellent product research have been and where they are going.
   As the editors of this volume, we thank the authors and their employers, who generously donated their
time and centuries of accumulated professional experience. We also thank Jill Jurgensen and
John Sulzycki of CRC/Taylor & Francis, who patiently guided us through the process of translating
an idea into a book.

                                                         Mustapha Debboun, San Antonio, Texas, USA
                                                   Stephen P. Frances, Enoggera, Queensland, Australia
                                                          Daniel Strickman, Beltsville, Maryland, USA




q 2006 by Taylor & Francis Group, LLC
Acknowledgments

Major (Dr.) Frances would like to thank Professor Karl Rieckmann (former Director) and Lieutenant
Colonel (Dr.) Robert Cooper (Commanding Officer), Australian Army Malaria Institute, for their support
over many years. The opinions expressed in his chapters are his alone, and do not reflect those of the
Australian Defence Health Service or any extant defence policy.




q 2006 by Taylor & Francis Group, LLC
q 2006 by Taylor & Francis Group, LLC
Editors

                               Dr. Mustapha Debboun is a medical and veterinary entomologist in the
                               United States (U.S.) Army Medical Department. He was born in Tangier,
                               Morocco and commissioned as an officer in the U.S. Army Medical
                               Service Corps in 1989 after receiving his doctoral degree. He received
                               his B.A. degree in cellular and molecular biology from Skidmore College
                               in Saratoga Springs, New York, a master of science degree in medical
                               entomology from the University of New Hampshire, Durham, New
                               Hampshire, and a doctor of philosophy degree in medical and veterinary
                               entomology from the University of Missouri–Columbia.
                                  Dr. Debboun joined the U.S. Army in 1989 and served in assignments at
the 44th Medical Brigade, Fort Bragg, North Carolina, Operations “Desert Shield” and “Desert Storm,”
Saudi Arabia; Academy of Health Sciences, U.S. Army Medical Department Center and School, Fort
Sam Houston, Texas; U.S. Army Europe, Commander, 255th Medical Detachment, Vicenza, Italy,
“Operation Joint Endeavor,” Bosnia; Office of the Assistant Secretary of the Army for Research,
Development, and Acquisition, Pentagon, Arlington, Virginia; Walter Reed Army Institute of Research,
Washington, D.C.; 3rd Medical Command, Operations “Enduring Freedom” and “Iraqi Freedom,” Camp
Arifjan, Kuwait and Camp Bucca, Iraq; and Commander, U.S. Army Center for Health Promotion and
Preventive Medicine–South, Fort McPherson, Georgia.
   During 17 years in the military, Dr. Debboun has worked in preventive medicine operations, research
and development of arthropod repellents, and field personal protective measures. This work has taken
him to 20 different countries in Africa, Asia, Australia, Europe, and Latin America. He is now deputy
chief of the Department of Preventive Health Services and chief of the Medical Zoology Branch at the
Academy of Health Sciences. He is a board certified medical and veterinary entomologist, and a member
of the Entomological Society of America, American Mosquito Control Association, and the Society of
Vector Ecology. He serves as adjunct faculty of the Non-Resident Command and General Staff College,
and as chair of the Repellents Committee and vice-chair of the Education and Training Committee,
Armed Forces Pest Management Board, Silver Spring, Maryland. Dr. Debboun has authored or co-
authored more than 50 research publications.

                              Dr. Stephen P. Frances is a medical entomologist at the Australian Army
                              Malaria Institute, Royal Australian Army Medical Corps, in Brisbane,
                              Queensland, Australia. He was born in Sydney, Australia, and attended the
                              University of Sydney, graduating with a bachelor of science in agriculture
                              with honors in 1980. From 1981 to 1984 he worked at the Commonwealth
                              Institute of Health, Sydney, as a research assistant conducting studies on a
                              fungal pathogen of mosquito larvae as a biocontrol agent under the
                              supervision of Professor Richard Russell. He completed a master of
                              science in agriculture in 1985.
                                 In 1985 he was commissioned as an officer in the Australian Defence
Force in the Army Malaria Research Unit, Ingleburn, New South Wales, and commenced work on
evaluating insect repellent formulations and other physical and chemical barriers to protect soldiers
against medically important arthropods, especially mosquitoes and mites. From 1992 to 1994 he served
in the Department of Medical Entomology, Armed Forces Research Institute of Medical Sciences,
Bangkok, Thailand, continuing his work evaluating repellents. He commenced PhD studies in 1991 on a
part time basis under the supervision of Dr. (Lieutenant Colonel) Anthony (Tony) Sweeney, and was




q 2006 by Taylor & Francis Group, LLC
awarded his doctor of philosophy from the Faculty of Medicine at the University of Sydney in 1999. His
dissertation was written on aspects of the transmission of the rickettsia that causes scrub typhus in two
regions of Thailand. He has also been involved in vector surveillance in northern Australia, Papua New
Guinea, and Timor Leste. He is a member of the American Mosquito Control Association, the Mosquito
Control Association of Australia, the Australian Entomological Society, and the Entomological Society
of Queensland. Dr. Frances has authored or co-authored 60 research publications.

                            Dr. Daniel Strickman attended Dartmouth College from 1971 to 1973,
                            transferring to the University of California at Riverside to pursue his interest
                            in entomology. He received his bachelor of arts degree in biology from UCR
                            in 1974 and began graduate studies at the University of Illinois, Champaign–
                            Urbana, in the same year. Studying under the late Dr. William Horsfall, he
                            worked on oviposition habits of Midwestern mosquitoes, completing his
                            master of science degree in 1976 and his doctor of philosophy degree in 1978.
                            Dr. Strickman and his wife, Linda, joined the Peace Corps and served in
                                                                                          ´
                            Paraguay on the staff of the National University of Asuncion for two years.
                            During that time, they taught environmental education and field entomology,
and completed natural history studies of mosquitoes, horse flies, and dragonflies. Dr. Strickman joined
the U.S. Air Force in 1981 and served as a consultant on toxicological issues throughout the United
States. He transferred to the U.S. Army in 1984, completing assignments at the Smithsonian Institution,
the Armed Forces Research Institute of Medical Science in Bangkok, the 5th Preventive Medicine Unit in
Seoul, and Walter Reed Army Institute of Research in Washington, D.C. During 22 years in the military,
Dr. Strickman worked in operations and research dealing with toxicology, taxonomy, repellents,
rickettsial diseases, dengue, malaria, and insect control. This work took him to nine different countries
in Latin America, Africa, and Asia. Following his retirement from the military in 2003, Dr. Strickman
worked as the vector ecologist for the Santa Clara County (California) Vector Control District. He is now
national program leader for veterinary, medical, and urban entomology for the Agricultural Research
Service, U.S. Department of Agriculture. Dr. Strickman is a member of the Entomological Society of
America, the American Mosquito Control Association, the American Society of Tropical Medicine and
Hygiene, the National Cattlemen’s Beef Association, and the U.S. Animal Health Association. He is an
author of 90 scientific publications and his main research interest is the integration of entomology with
other operational fields to provide efficient, sustainable management of disease to protect humans and
animals.




q 2006 by Taylor & Francis Group, LLC
Contributors

Arshad Ali, Ph.D.                             Bldg. 970, Natural Area Drive
University of Florida, IFAS                   P.O. Box 110620
Department of Entomology and                  Gainesville, FL 32611-0620
  Nematology                                  Phone: (352) 392-1930 ext. 152
Mid-Florida Research Education Center         FAX: (352) 392-0190
2725 Binion Road                              E-mail: jfb@ifas.ufl.edu
Apopka, FL 32703
Phone: (407) 884-2034                         John F. Carroll, Ph.D.
FAX: (407) 814-6186                           USDA/ARS
E-mail: aali@mail.ifas.ufl.edu                 Animal Parasitic Diseases Laboratory
                                              BARC-East, Bldg. 1040
Donald R. Barnard, Ph.D.                      Beltsville, MD 20705
USDA/ARS                                      Phone: (301) 504-9017
Center for Medical, Agricultural and          FAX: (301) 504-5306
  Veterinary Entomology                       E-mail: jcarroll@anri.barc.usda.gov
P.O. Box 14565
Gainesville, FL 32604                         Scott P. Carroll, Ph.D.
Phone: (352) 374-5930                         University of California–Davis
FAX: (352) 374-5870                           Department of Entomology
E-mail: dbarnard@gainesville.usda.ufl.edu      Center for Population Biology
                                              Davis, CA 95616
Ulrich R. Bernier, Ph.D.                      E-mail: spcarroll@ucdavis.edu
USDA/ARS/CMAVE
Mosquito and Fly Research Unit                Jonathan F. Day, Ph.D.
1600 SW 23rd Drive                            University of Florida, IFAS
Gainesville, FL 32608                         Florida Medical Entomology Laboratory
Phone: (352) 374-5917                         2009th Street, SE
FAX: (352) 374-5922                           Vero Beach, FL 32962
E-mail: ubernier@gainesville.usda.ufl.edu      Phone: (772) 778-7200
                                              E-mail: jfday@ifas.ufl.edu
Apurba K. Bhattacharjee, Ph.D.
Walter Reed Army Institute of Research        LTC Mustapha Debboun, Ph.D., BCE
Department of Medicinal Chemistry             U.S. Army Medical Department Center
Division of Experimental Theraputics            and School
Silver Spring, MD 20910                       Academy of Health Sciences
Phone: (301) 319-9043                         Department of Preventive
E-mail: apurba.bhattacharjee@na.amedd.          Health Services
army.mil                                      Medical Zoology Branch
                                              3151 Scott Road, Suite 0408A
Jerry F. Butler, Ph.D.                        Fort Sam Houston, TX 78234-6142
University of Florida                         Phone: (210) 210-7649
Medical-Veterinary Entomology                 FAX: (210) 210-8332
Entomology and Nematology Department          E-mail: mustapha.debboun@us.
Institute of Food and Agricultural Sciences   army.mil




q 2006 by Taylor & Francis Group, LLC
David N. Durrheim, DrPH, MPHandTM,       James Ha, B.A.
MBChB, FAFPHM, FACTM                     University of Illinois
Hunter New England Population Health     College of Medicine-Rockford
Health Protection                        1601 Parkview Avenue
Locked Bag 10                            Rockford, IL 61107
Wallsend New South Wales 2287            Phone: (815) 621-0800
Australia                                FAX: (815) 395-5887
Phone: 61-2-49246473                     E-mail: jha2@uic.edu
FAX: 61-2-49246048
E-mail: david.durrheim@hnehealth.        Yi-Xun He, Ph.D.
nsw.gov.au                               University of Illinois
                                         College of Medicine-Rockford
Major Stephen P. Frances, Ph.D.          1601 Parkview Avenue
Australian Army Malaria Institute        Rockford, IL 61107
Vector Surveillance and Control          Phone: (815) 395-5694
Weary Dunlop Drive                       FAX: (815) 395-5666
Gallipoli Barracks Enoggera QLD 4051     E-mail: yixunhe@uic.edu
Australia
Phone: 61 7 3332 4807                    Nigel Hill, Ph.D.
FAX: 61 7 3332 4800                      London School of Hygiene and Tropical
E-mail: steve.frances@defence.gov.au       Medicine
                                         Department of Infectious and Tropical
Eugene J. Gerberg, Ph.D.                   Diseases
University of Florida, IFAS              Vector Biology and Disease Control Unit
Entomology and Nematology Department     Keppel Street
P.O. Box 110620                          London WC1E 7HT
Gainesville, FL 32611-0620               United Kingdom
Phone: (home) (352) 373-7384             Phone: 020 7927 2646
E-mail: genejg2@aol.com                  FAX: 020 7636 8739
                                         E-mail: nigel.hill@lshtm.ac.uk
John M. Govere, Ph.D.
World Health Organization for African    Daniel L. Kline, Ph.D.
  Region (WHO/AFRO)                      USDA/ARS/CMAVE
Parirenyatwa Hospital                    Mosquito and Fly Research Unit
Mazoe Street                             1600 SW 23rd Drive
P.O. Box CY384                           Gainesville, FL 32608
Harare, Zimbabwe                         Phone: (352) 374-5917
Phone: 263 4 253724-30                   FAX: (352) 374-5922
FAX: 263 4 253731-2                      E-mail: dkline@gainesville.usda.ufl.edu
E-mail: goverej@zw.afro.who.int
                                         Glenn J. Leach, Ph.D.
Col. Raj K. Gupta, Ph.D.                 U.S. Army Center for Health Promotion
Walter Reed Army Institute of Research     and Preventive Medicine
Office of the Science Director            Directorate of Toxicology
503 Robert Grant Avenue                  Toxicity Evaluation
Silver Spring, MD 20910                  5158 Blackhawk Road
Phone: (301) 619-7732                    Aberdeen Proving Ground, MD 21010
FAX: (301) 619-2982                      FAX: (410) 436-6710
E-mail: raj.gupta@amedd.army.mil         E-mail: glenn.leach1@us.army.mil




q 2006 by Taylor & Francis Group, LLC
Walter S. Leal, Ph.D.                       1600 SW 23rd Drive
University of California–Davis              Gainesville, FL 32608
Department of Entomology                    Phone: (352) 374-5917
308D Briggs Hall                            FAX: (352) 374-5922
One Shield Ave                              E-mail: kposey@gainesville.usda.ufl.edu
Davis, CA 95616-8584
Phone: (530) 752-7755                       Shri Prakash, Ph.D.
FAX: (530) 754-8682                         Defence Research and Development
E-mail: wsleal@ucdavis.edu                    Establishment
                                            Division of Entomology
Annick Lenglet                              Jhansi Road
London School of Hygiene and Tropical       Gwalior-474002
  Medicine                                  M.P., India
Vector Biology and Disease Control Unit     E-mail: shripra2004@yahoo.co.in
Keppel Street
London WC1E 7HT                             Germain Puccetti, Ph.D.
United Kingdom                              EMD Chemicals, Inc.
E-mail: annick@thelenglets.com              7 Skyline Drive
                                            Hawthorne, NY 10532
Wilfred C. McCain, Ph.D.
                                            Phone: (914) 592-4660 ext 489
U.S. Army Center for Health Promotion and
                                            FAX: (914) 785-5889
  Preventive Medicine
                                            E-mail: gpuccetti@emdchemicals.com;
Directorate of Toxicology
                                            gp1000@mail.solgel.com
5158 Blackhawk Road
Aberdeen Proving Ground, MD 21010
                                            Kalyanasundaram Ramaswamy, Ph.D.
Phone: (410) 436-2201
                                            University of Illinois
FAX: (410) 436-6710
                                            College of Medicine–Rockford
E-mail: wilfred.mccain@us.army.mil
                                            Department of Biomedical Sciences
Sarah J. Moore, Ph.D.                       1601 Parkview Avenue
London School of Hygiene and Tropical       Rockford, IL 61107
  Medicine                                  Phone: (815) 395-5696
Keppel Street                               FAX: (815) 395-5666
London WC1E 7HT                             E-mail: ramswamy@uic.edu
United Kingdom
Phone: 020 763 68636                        Louis C. Rutledge
E-mail: sarah.moore@sjmoore.net             United States Army, Retired
                                            11 Circle Way
Robert J. Novak, Ph.D.                      Mill Valley, CA 94941-3420
University of Illinois Urbana–Champaign     Phone: (415) 388-2937
Illinois Natural History Survey             E-mail: louiscrutledge@msn.com
Medical Entomology Program
607 E. Peabody Drive                        Buz Salafsky, Ph.D.
Champaign, Illinois 61820                   University of Illinois
Phone: (217) 333-1186                       College of Medicine–Rockford
FAX: (217) 333-2359                         Department of Pharmacology
E-mail: rjnovak@uiuc.edu                    1601 Parkview Avenue
                                            Rockford, IL 61107
Kenneth H. Posey                            Phone: (815) 395-5697
USDA/ARS/CMAVE                              FAX: (815) 395-5887
Mosquito and Fly Research Unit              E-mail: buzs@uic.edu




q 2006 by Taylor & Francis Group, LLC
K. Sekhar, Ph.D.                            R. Vijayaraghavan, Ph.D.
Defence Research and Development            Defence Research and Development
  Establishment                               Establishment
Jhansi Road                                 Jhansi Road
Gwalior-474 002                             Gwalior-474 002
M.P., India                                 M.P., India

Takeshi Shibuya, Ph.D.
University of Illinois                      Paul J. Weldon, Ph.D.
College of Medicine–Rockford                Smithsonian Institution
Department of Pharmacology                  Conservation and Research Center
1601 Parkview Avenue                        1500 Remount Road
Rockford, IL 61107                          Front Royal, VA 22630
Phone: (815) 395-0600                       Phone: (410) 732-1539
FAX: (815) 395-5887                         E-mail: weldonp@si.edu

Daniel Strickman, Ph.D.
Veterinary, Medical, and Urban Entomology   Graham B. White, Ph.D.
USDA/ARS, National Program Staff            University of Florida
5601 Sunnyside Avenue                       Entomology and Nematology Department
Beltsville, MD 20705-5134                   P.O. Box 14565
Phone: (301) 504-5771                       Gainesville, FL 32604-2565
FAX: (301) 504-5467/4725                    Phone: (352) 374-5968
E-mail: daniel.strickman@ars.usda.gov       FAX: (352) 374-5922
                                            E-mail: gbwhite@ufl.edu
Kevin J. Sweeney
U.S. Environmental Protection Agency        Rui-de Xue, Ph.D.
Office of Pesticide Programs                 Anastasia Mosquito Control District
Registration Division (7505C)               500 Old Beach Road
1200 Pennsylvania Avenue, NW                St. Augustine, FL 32080
Washington, DC 20460-0001                   Phone: (904) 471-3107
Phone: (703) 305-5063                       FAX: (904) 471-3189
E-mail: sweeney.kevin@epamail.epa.gov       E-mail: xueamcd@bellsouth.net




q 2006 by Taylor & Francis Group, LLC
Table of Contents

PART 1 Principles

Chapter 1 History of Insect Repellents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Sarah J. Moore and Mustapha Debboun


Chapter 2 Terminology of Insect Repellents. . . . . . . . . . . . . . . . . . . . . . . . . . 31
Graham B. White


Chapter 3 Vertebrate Chemical Defense: Secreted and Topically
                     Acquired Deterrents of Arthropods . . . . . . . . . . . . . . . . . . . . . . . 47
Paul J. Weldon and John F. Carroll


Chapter 4 Human Emanations and Related Natural Compounds
                     That Inhibit Mosquito Host-Finding Abilities. . . . . . . . . . . . 77
Ulrich R. Bernier, Daniel L. Kline, and Kenneth H. Posey


PART 2 Methods

Chapter 5 Standard Methods for Testing
                     Mosquito Repellents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Donald R. Barnard, Ulrich R. Bernier, Rui-de Xue, and Mustapha Debboun


Chapter 6 Biometrics and Behavior in Mosquito
                     Repellent Assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Donald R. Barnard and Rui-de Xue


Chapter 7 Animal Models for Research and Development of
                     Insect Repellents for Human Use . . . . . . . . . . . . . . . . . . . . . . . . 125
Louis C. Rutledge and Raj K. Gupta


Chapter 8 Techniques for Evaluating Repellents. . . . . . . . . . . . . . . . . . . . 147
John M. Govere and David N. Durrheim




q 2006 by Taylor & Francis Group, LLC
Chapter 9 Use of Olfactometers for Determining Attractants
                     and Repellents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
Jerry F. Butler


Chapter 10 Discovery and Design of New Arthropod/Insect
                       Repellents by Computer-Aided Molecular
                       Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
Raj K. Gupta and Apurba K. Bhattacharjee


Chapter 11 Molecular-Based Chemical Prospecting of Mosquito
                       Attractants and Repellents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
Walter S. Leal


PART 3 Products and Active Ingredients

Chapter 12 Evaluation of Topical Insect Repellents and Factors
                       That Affect Their Performance . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Scott P. Carroll


Chapter 13 Repellents Used in Fabric: The Experience of the
                       U.S. Military. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
Wilfred C. McCain and Glenn J. Leach


Chapter 14 Plant-Based Insect Repellents . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
Sarah J. Moore, Annick Lenglet, and Nigel Hill


Chapter 15 Considerations on the Use of Botanically-Derived
                       Repellent Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305
Eugene J. Gerberg and Robert J. Novak


Chapter 16 Efficacy and Safety of Repellents Containing Deet . . . . 311
Stephen P. Frances


Chapter 17 Lipodeet: An Improved Formulation for a Safe,
                       Long-Lasting Repellent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
Buz Salafsky, Takeshi Shibuya, Yi-Xun He, James Ha, and
Kalyanasundaram Ramaswamy



q 2006 by Taylor & Francis Group, LLC
Chapter 18 Picaridin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
Stephen P. Frances

Chapter 19 DEPA: Efficacy, Safety, and Use of N,N-Diethyl
                             Phenylacetamide, a Multi-Insect Repellent . . . . . . . . . . . . . 341
Shri Prakash, R. Vijayaraghavan, and K. Sekhar

Chapter 20 PMD (p-Menthane-3,8-Diol) and Quwenling . . . . . . . . . . . . . . 347
Daniel Strickman

Chapter 21 IR3535 (Ethyl Butylacetylaminopropionate) . . . . . . . . . . . . 353
Germain Puccetti

Chapter 22 Older Synthetic Active Ingredients and
                             Current Additives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361
Daniel Strickman

Chapter 23 Area Repellent Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
Daniel Strickman

PART 4 Uses

Chapter 24 User Acceptability: Public Perceptions of
                             Insect Repellents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 397
Stephen P. Frances and Mustapha Debboun

Chapter 25 Commercially Available Insect Repellents and Criteria
                             for Their Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405
Rui-de Xue, Arshad Ali, and Jonathan F. Day

Chapter 26 Global Regulatory Perspective on Insect Repellent
                             Development and Registration . . . . . . . . . . . . . . . . . . . . . . . . . . 417
Kevin J. Sweeney

Epilogue: Prospects for the Future . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425
Daniel Strickman, Stephen P. Frances, and Mustapha Debboun

Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 429




q 2006 by Taylor & Francis Group, LLC
q 2006 by Taylor & Francis Group, LLC
1
History of Insect Repellents


Sarah J. Moore and Mustapha Debboun



CONTENTS
Historical Review..............................................................................................................................3
Traditional Repellent Use Today......................................................................................................5
Pyrethrum, Mosquito Coils, and Area Repellents............................................................................5
The Development of Modern Synthetic Repellents .........................................................................6
Deet—A Breakthrough in Repellents ...............................................................................................8
Recent Repellent Discoveries ...........................................................................................................8
  DEPA .............................................................................................................................................8
  IR 3535 ..........................................................................................................................................9
  Piperidine Compounds ..................................................................................................................9
  KBR 3023 ......................................................................................................................................9
  AI3-35765 and AI3-37220 ..........................................................................................................10
  SS220 ...........................................................................................................................................10
Repellent Delivery Methods ...........................................................................................................11
  Area Repellents............................................................................................................................13
The Evolution of Repellent Testing ...............................................................................................14
  Kairomones ..................................................................................................................................14
  Choice ..........................................................................................................................................14
  In Vitro and Animal Tests...........................................................................................................15
  Test Standardization ....................................................................................................................15
References .......................................................................................................................................17




Historical Review
It is likely that the use of repellents against biting arthropods developed thousands—possibly even
millions—of years ago. Several species of primate have been observed anointing their pelage by rubbing
it with millipedes and plants including Citrus spp., Piper marginatum, and Clematis dioica.1–4 Wedge-
capped capuchins (Cebus olivaceus) were observed rubbing the millipede Orthoporus dorsovittatus
onto their coat during the period of maximum mosquito activity.5 The O. dorsovittatus species contains
insect-repellent chemicals called benzoquinones, and it was hypothesized that the anointing behavior
was designed to deter biting insects. Laboratory studies went on to show a significant repellent effect of


                                                                                                                                                   3

q 2006 by Taylor & Francis Group, LLC
4                                                            Insect Repellents: Principles, Methods, and Uses


benzoquinones against Aedes (Stegomyia) aegypti (the yellow fever mosquito)6 and Amblyomma
americanum (the lone star tick).7 Anointing behavior to deter blood-feeding arthropods is also
common among birds,8 and it may be genetically expressed as an “extended phenotype” because it
has obvious adaptive benefit.9 Evidence for this lies in the fact that benzoquinones applied to filter paper
elicited anointing activity among captive-born capuchins.6
   The first recorded use of repellents may be found among the writings of Herodotus (484 BCE—ca. 425
BCE), who observed Egyptian fishermen.10 Herodotus stated that,

        The Egyptians who live in the marsh-country use oil extracted from the castor-oil plant. This
        plant, which grows wild in Greece, they call Kiki, and the Egyptian variety is very prolific and has
        a disagreeable smell. Their practice is to sow it along the banks of rivers and lakes, and when the
        fruit is gathered it is either bruised and pressed, or else boiled down, and the liquid thus obtained
        is of an oily nature and quite as good as olive oil for burning in lamps, although the smell is
        unpleasant.

   It was argued that the oils acted as an area repellent because high densities of nuisance mosquitoes are
active in the evenings in this region. This would have driven the Egyptians to their beds (where
Herodotus also observed that they slept under rudimentary bed nets) had the lamp not provided
protection from biting insects.11
   The Romans also recorded methods of repelling flying insects (gnats) that would have included
mosquitoes, as much of Italy was once swampland where the malaria vectors Anopheles labranchiae,
Anopheles sacharovi, and Anopheles superpictus were abundant prior to the malaria eradication program
of 1947.12 The Geoponika is a collection of Roman agricultural lore, compiled during the tenth century
for the Byzantine emperor Constantine VII Porphyrogenitus, that was heavily based upon the writings of
Vindonius Anatolius (fourth century), as well as earlier writers, including Pliny.13 The text suggests
rubbing a concoction of vinegar, manna, and oil on the body, especially the head and feet, to repel
gnats.13 This may have had an effect on nuisance insects, especially mosquitoes, as natural vinegars
contain acetic acid and smaller amounts of tartaric and citric acids. These acids may have had a mild
antibacterial effect on the skin and therefore reduced the production of bacterial metabolites that
mosquitoes use to locate human hosts,14 particularly those produced by the feet.15 In addition, some oils
have a mild repellent action,16 perhaps by reducing the emanation of host odor. In addition, Geoponika
describes burning herbs such as black cumin (Nigella sativa), bay (Laurus nobilis), galbanum (Ferula
gummosa), and oregano (Origanum vulgare) to drive away nuisance insects.13 Writings (ca. seventeenth
century) derived from the ancient Sanskrit Yoga Ratnakara also contain references to the burning of
plants to repel biting insects, including Vaca (Acorus calamus), Marica (Piper nigrum), asafoetidia
(Ferula asafoetida), and Nimba or Neem (Azadirachta indica).17 Other remedies suggested in
Geoponika and Yoga Ratnakara included burning fish, shells, various bones, dung, snakeskin, and
peacock feather. This would have created a thick noxious smoke, as would have burning asafoetidia that
has the colloquial name of Devil’s Dung in old French. This may have been perceived to work, as the
smoke generated was thick and noxious to humans, although smoke does have some repellent effects on
mosquitoes.18 The smoke may mask human kairomones, particularly carbon dioxide, and the convection
currents that mosquitoes need for short-range host location. Smoke production also lowers humidity by
reducing the moisture-carrying capacity of the air. This makes mosquitoes susceptible to desiccation and
reduces sensory input because mosquito chemoreceptors are more responsive in the presence of
moisture.19
   In North America, native cultures relied heavily on plants, and many used plants to repel biting
insects.20 The Southern Carrier Tribe, or Dakelh, meaning “people who go around in boats,” live near
rivers in British Columbia where mosquito densities are extremely high. This group used an infusion of
common cow parsnip blossoms (Heracleum maximum) rubbed on the body to repel flies and mosquitoes;
however, the more common mode of use was burning. For instance, the Colville Indians based around the
Columbia River used leaves and stems of Common Yarrow (Achillea millefolium) as a smudge to keep


q 2006 by Taylor & Francis Group, LLC
History of Insect Repellents                                                                                  5


away mosquitoes. The Blackfoot tribe, whose territory stretched along the Saskatchewan River, put the
fringed sagewort (Artemesia frigida) plant on campfire coals to drive away mosquitoes. Apparently, it
was so effective that wild horses sheltered from insect pests in the smoke; consequently, the Indians used
it to attract horses. The use of smoke against biting insects was carried on by European settlers as
recorded by settlers to The Black Swamp in Ohio21:

        They [first settlers in Wood County] were subject to all kinds of deprivations. The most
        distressing of all the rest was their being subject to epidemics that swept through the country
        every summer and fall in the shape of malarial fevers.

        The warm months gave way to unrelenting swarms of gnats and mosquitoes.

        The most effective tool available [to fight the mosquito] was the smudge pot. These pots and their
        accompanying clouds of dark smoke discouraged the insects and were useful throughout most of
        the day; they were next to the cow while milking, under the table while eating, and even beside the
        bed while sleeping.




Traditional Repellent Use Today
Smoke is still the most widely used means of repelling mosquitoes utilized throughout the rural tropics.
Waste plant materials are frequently burned in Sri Lanka as a mosquito repellent, even though indoor
residual spraying has been carried out by the government for many years.22 In rural Guinea-Bissau, 86%
of residents used an unimpregnated bednet in conjunction with mosquito coils or plant-based smoke.23 In
the Solomon Islands, a recent survey revealed that fire with coconut husks and papaya leaves was the
most prevalent form of personal protection from mosquitoes, being used by 52% of residents.24 Surveys
from South America found that 69 and 90% of respondents from Mexico25 and Guatemala,26
respectively, burned waste materials to drive away mosquitoes. Smoke is also used to drive away
biting insects in Southeast Asia: wood-fires and smudge pots are used in Myanmar,27 whereas herbs are
thrown on the fire in Yunnan, China.28
   Although these methods are crude, many traditional repellents do have a repellent effect. A recent
controlled field trial showed a comparable repellent effect produced by a 0.2% pyrethrin mosquito coil
and lemon gum (Corymbia citriodora) volatiles expelled by heating on metal plates.29 Several field
evaluations, where plants were burned to repel mosquitoes, have shown good reduction in mosquito
landings.23,30,31 One well-designed study in Papua New Guinea showed that burning local wood and
leaves (mango wood, coconut husks, wild ginger leaves, and betelnut leaves) repelled between 57
and 75% of mosquitoes.31 Smoke also reduced indoor sand fly density by 1.7 times in East Africa.32
   The use of smoke, although effective, requires continuous production in order to repel biting insects
when used as an area repellent outdoors.33 Although smoke does have a residual repellent effect when
used within houses,29 the indoor combustion of biomass has severe health consequences.34 Therefore,
safer and more modern methods of repelling mosquitoes are desirable.




Pyrethrum, Mosquito Coils, and Area Repellents
Pyrethrum is natural plant oil that occurs in the two species of pyrethrum daisy: Chrysanthemum
cinerariifolium from the Dalmatian region and Chrysanthemum coccineum of Persian origin. The
insecticidal component, comprising six esters (pyrethrins), is found in tiny oil-containing glands on


q 2006 by Taylor & Francis Group, LLC
6                                                        Insect Repellents: Principles, Methods, and Uses


the surface of the seed case in the flower head. It is a highly effective insecticide that, although it has been
used for centuries against all manner of insect pests, is relatively harmless to mammals.35
   Pyrethrum is thought to have originally been used in China and was introduced to the Middle East
along the trade routes through Central Asia,36 from where it was introduced into Europe during the
nineteenth century.37 It is currently incorporated into mosquito coils to repel insects, and this practice
probably derived from the incense used in religious ceremonies by Hindus, Buddhists, and the followers
of Confucius. In Java today, the same incense used in ceremonies to honor ancestors is also used on a
daily basis to repel mosquitoes.38
   Pyrethrum powders were used by armies from the time of Napoleon to World War II to combat head
and body lice. Before World War II, Japan was the major growing area,37 and exported pyrethrum
powder that was mainly used directly in its unrefined form as a powder for killing fleas. At that time in
Japan, people usually mixed pyrethrum powder with sawdust and burned it in a brazier or incense burner
to repel mosquitoes. Around 1890, the businessman Eiichiro Ueyama improved the pyrethrum powder
and successfully developed a spiral-shaped mosquito repellent.39 He formulated that idea when he met
the son of an incense dealer at an inn in Tokyo. While talking with him, he came up with the idea of
mixing starch powder with pyrethrum powder, then kneading it into the shape of stick incense. After
several failures, Mr. Ueyama employed the workers of incense makers in Sakai, and thereby succeeded in
creating a viable commercial product: a mixture of starch powder, dried mandarin orange skin powder,
and pyrethrum powder. It was thoroughly mixed and kneaded, placed into a wooden mortar, extruded,
and cut into the form of stick incense. Ueyama then replaced the wooden mortar with a compressing
machine and was able to realize mass production.
   However, the bar-shaped mosquito stick burned rapidly, and several sticks had to be burned at once to
obtain sufficient smoke to repel insects. In 1895, Yuki, the wife of Eiichiro, proposed making the stick
thicker and longer, and curling it into a spiral shape. Eiichiro acted immediately on her proposal, but it
was not until 1902, after years of experimentation, that he was finally able to complete a mosquito
repellent with a spiral shape that was worthy of marketing. The final method involved cutting a thick bar
of incense to a certain length and manually winding it. This same method continued to be used until 1957,
when it was improved through machine punching, making mass production possible on a far larger
scale.39 Mosquito coils are widely used today: 29 billion mosquito coils are sold each year, 95% of them
in Asia,40 and household expenditure on these methods in the developing countries is substantial.41,42
There is ample evidence that mosquito coils effectively repel mosquitoes.43
   Pyrethrum affects the central nervous systems of all types of flying and crawling insects, blocking
sodium-gated nerve junctions so that nervous impulses fail,44 and the insect is knocked down and may
die. In the lowest concentrations, pyrethrum affects insect behavior, producing a so-called “avoidance
reaction” or “excito-repellency” that results in the insect fleeing the source of the chemicals.45
Synthetic analogues of pyrethrum were developed from the 1940s onwards. They exhibit a similar
mode of action to pyrethrum, but are more potent and photostable.46,47 The insecticides broadly act in
two ways: (1) the choreoathetosis/salivation (CS) pathway, and (2) the tremor (T) pathway.48
Importantly, these effects result in deterrency from entering a room where coils are burning, expellency
of mosquitoes from within, interference with host-finding, bite inhibition, knockdown, and kill.49 These
repellence and bite-inhibition effects have been exploited to produce highly-efficacious repellents that
combine permethrin (a synthetic pyrethroid) and deet, a synthetic repellent discussed extensively in this
volume.50,51




The Development of Modern Synthetic Repellents
The military has conducted significant research into modern repellents to protect their troops from
insect-borne disease. The first military repellents contained essential oils derived from plants.


q 2006 by Taylor & Francis Group, LLC
History of Insect Repellents                                                                                    7




FIGURE 1.1 Repellents distributed to U.S. troops, Bombay 1945 (qOffice of the Army Surgeon General, Public Affairs,
and the Directorate of Information Management, Fort Detrick, MD, USA.).




For instance, the Indian Army was issued a repellent comprised of citronella, camphor, and
paraffin.52 However, these repellents had limited duration, and intensive research began during
World War II to find long-lasting repellents. The enormous burden of disease suffered by troops
fighting in endemic areas motivated this research. For instance, 821,184 cases of malaria were
recorded among U.S. troops involved in overseas campaigns, resulting in 302 deaths,53 and over 12
million lost duty days.54 With the advent of large-scale jungle warfare, chigger-borne scrub typhus
became an important medical problem for troops in the Far East. Indeed, approximately 6,000 cases
were to appear in U.S. forces alone during the campaigns that followed the outbreak of war with
Japan.55 Chiggers were also the cause of considerable discomfort for soldiers training in the U.S.;
this resulted in the Surgeon General requesting the Orlando laboratory of the United States
Department of Agriculture (USDA) to study means and methods for controlling chiggers by
repellents or insecticides in the summer of 1941.56 Between 1942 and 1945, over 7,000 potentially
repellent compounds were tested by the USDA.57 One of the first chemical repellents to be
developed was dimethyl phthalate (DMP; it was patented in 1929 as a fly repellent), followed by
Indalonew (butyl-3,3-dihydro-2,2-dimethyl-4-oxo-2H-pyran-6-carboxylate; patented in 1937), and
ethyl hexanediol (2-ethyl-1,3-hexanediol), also called Rutgers 612, that became available in
1939.58 In 1942, DMP and Indalone demonstrated significant protection against chiggers when
tested by Madden, Lindquist, and Knipling of the Orlando laboratory for troops in Louisiana.59 The
result was corroborated by field trials in New Guinea against scrub chiggers (Leptotrombidium)60
and a range of other species.61 The introduction of chemical repellents dramatically lowered
incidence of scrub typhus,55,60 and allowed less-restrictive battle uniforms. Prior to the introduction
of DMP and Indalone for impregnation of uniforms and application to exposed skin, prevention of
insect bites had relied on long clothing plus head nets and mosquito gloves.56 The head nets were
uncomfortably hot and restricted vision, making them unpopular with troops and therefore rarely
used. The introduction of repellents for exposed parts of the body proved more popular
(Figure 1.1).56
   After the war, a repellent known as 6-2-2 or M-250, containing 6 parts DMP, 2 parts Indalone, and
2 parts ethyl hexanediol, became popular in the U.S.A. However, products containing ethyl hexanediol
were voluntarily removed from the U.S. and Canadian markets in 1991 in response to an unpublished
study by a manufacturer showing poor lung expansion in the offspring of exposed animals.62
Additional studies showed mild developmental toxicity after cutaneous administration to pregnant
rats.63


q 2006 by Taylor & Francis Group, LLC
8                                                      Insect Repellents: Principles, Methods, and Uses




Deet—A Breakthrough in Repellents
Prior to the removal of 6-2-2 from the marketplace, its use was eclipsed by the discovery of deet
(N,N-diethyl-3-methlybenzamide or N,N-diethyl-m-toluamide) in 1953.64 This was perhaps the single
most important event in the evolution of repellents, and deet remains the principal, and the most effective
repellent in use today65—more than 50 years after its discovery. Deet is a broad-spectrum repellent that is
highly effective against all mosquitoes: Aedes spp.,66–69 including the dengue vectors Aedes aegypti70,71
and Aedes albopictus72,73; Culex spp.71,73–76; Mansonia spp.71,74,77; and Anopheles malaria vectors,
including the Afrotropical Anopheles gambiae,74,78,79 and Anopheles arabiensis,80–82 Southeast Asian
Anopheles dirus,71,83,84 and Anopheles minimus30; South American Anopheles darlingi,85 and Western
Pacific Anopheles farauti.86,87 Other insects of medical importance repelled by deet include sand flies
(Psychodidae, both Old World and New World)88–90; black flies (Simulidae)67,91; chiggers (Trombicu-
lidae)92–94; hard and soft ticks (Ixodidae)95–98; bedbugs (Cimex hemipterus) 99 ; and fleas
(Siphonaptera).100 It is, therefore, now used as the “gold-standard” repellent against which other
substances are compared in laboratory and field trials.
   An estimated 15 million people in the United Kingdom, 78 million people in the United States of
America.,101 and 200 million people globally use deet each year.102 There has been much speculation
on the safety of deet following reports linking it to seizures and encephalopathy, particularly in
children,103–106 as well as neurotoxicity,107 especially in combination with other pesticides.108
However, deet has been used for 50 years with a tiny number of reported adverse effects, many of
which had a history of excessive or inappropriate use of repellent.109,110 Nonetheless, its toxicology has
been more closely scrutinized than any other repellent, but it has been deemed safe for human use,101,111
including use on children106 and pregnant women.112 The use of a deet/permethrin repellent has recently
been proven to reduce malaria incidence amongst users.113




Recent Repellent Discoveries
DEPA
Recently, DEPA (N,N-diethyl phenyl acetamide), a compound developed around the same time as deet,64
has received renewed attention. It has similar cosmetic properties to deet, similar dermal absorption and
excretion, plus the symptoms of acute poisoning with DEPA are similar to deet.114 However, its dermal
toxicity to rats has been reported as LD50 1.7–2.1 g/kg,114 and 3–4 g/kg,115 which may require
further clarification.
   In a field study, 0.3 mg/cm2 DEPA in alcohol provided complete protection against Culex
quinquefasciatus mosquitoes at a mean landing rate of 9.22 mosquitoes/person/h.116 Another field test
of DEPA with Culex quinquefasciatus, Simulium himalayense, and the leech Haemadipsa zeylanica
showed 1.5, 2, and 1.5 h of complete protection, respectively.117 However, control numbers were not
given in this publication. Laboratory tests using rabbits showed that there was no significant difference in
the response of the sand fly Phlebotomus papatasi to DEPA or deet.90 Furthermore, in vitro application of
repellents to a membrane blood feeding system, for Aedes aegypti, has shown that two analogues
of DEPA, DM156 and DM34, show promising repellency and low toxicity, warranting further
evaluation.118 DEPA is an extremely cheap repellent, costing Rs. 1140 (U.S. $25.40) per kg compared
to Rs. 2170 (U.S. $48.40) for deet.116 This is because one of the precursors of deet, (3-methylbenzoic
acid) is not readily available in India.119 DEPA has now been formulated in a commercial preparation by
the Defence Research and Development Establishment (DRDE) and has been granted approval by the


q 2006 by Taylor & Francis Group, LLC
History of Insect Repellents                                                                               9


Drug Controller of India.120 This repellent may prove useful, particularly among residents of the
developing world, for whom cost is the main motivator in personal repellent choice.121


IR 3535
Insect repellent 3535 (IR 3535), [3-(N-acetyl-N-butyl) aminopropionic acid ethyl ester], also known as
MERCK 3535, was developed in 1975 by Merck,122 and has been on the market in Europe for the past
twenty years. It has low toxicity, although it is irritating to the eyes and sometimes the skin.123 It became
available in the U.S. in 1999 after being passed by the EPA, classified as a biopesticide, as it is a
substituted B-amino acid, structurally similar to naturally occurring B-alanine.124
   Efficacy data for IR 3535 is variable, but it is generally comparable with deet. Data from the laboratory
showed IR 3535 to be equal to deet against Aedes aegypti, Culex quinquefasciatus,69,125 and Culex
taeniorhynchus, but not Anopheles dirus.125 However, another laboratory study with Aedes aegypti and
Anopheles maculatus showed IR 3535 to be significantly inferior to deet.126 Field trials in Southeast Asia
against Aedes albopictus, and Culex gelidus125; and Aedes albopictus, Culex quinquefasciatus, and Culex
bitaeniorhynchus127 found that IR 3535 and deet offered similar protection. However, a test against
Aedes cantans and Aedes annulipes under initial biting pressures of 714 landings/person/h produced data
that indicated that deet had a duration twice that of IR 3535 (4.8 vs 9.7 h).128 A further test against Aedes
(Ochlerotatus) taeniorhynchus in the Everglades, also under high biting pressure, measured no
significant difference between the protection offered by deet and IR 3535.68 A comprehensive field
test against Anopheles gambiae showed that IR 3535 decayed at a similar rate to deet,78 and the World
Health Organization (WHO) has declared it a safe and effective repellent for human use.123 In fact, there
is not a single recorded incidence of an adverse reaction to this compound.


Piperidine Compounds
There has been a flurry of renewed interest in the piperidine-based compounds, leading to the discovery
of several new and highly effective repellents. Piperidines, as a chemical class, are cyclic amines. The
piperidine skeleton is present in piperine, the main active chemical agent in pepper (Piper sp.). During
the 1970s, approximately 600 synthetic compounds related to piperidines were developed by scientists at
the Gainesville and Beltsville research centers of the USDA. The data from these experiments is now
being re-examined using new, high-tech methodologies coupled with rapid-screening bioassays. This
interest in finding deet alternatives has been motivated by the controversy around the safety of deet, its
low user acceptability, and its plasticizing effect.


KBR 3023
The repellent 1-piperidine carboxylic acid-2(2-hydroxyethyl)-1-methylpropylester was developed by
Bayer in the 1980s using molecular modelling.129 It has several synonyms: Picaridin is its common
name, Bayrepelw is its Bayer trademark name, Icaridin was used by WHO, and KBR 3023 is another
trade name. This compound, the most recent piperidine derivative, is registered in many European, South
American, Asian and African countries as well as Japan, Canada, and the U.S. Its most important new
feature is its very low toxicity (EPA Grade III). Most importantly, it elicits practically no dermal or eye
irritation (EPA Grade IV) nor skin sensitization.130 Furthermore, it does not have a significant
plasticizing effect, which is a major drawback of deet. Cosmetically, it is superior to deet as it is
colorless, odorless and has a pleasant feel on the skin.131 A user acceptability study showed a distinct
preference for KBR 3023 among Australian troops when compared to deet, which was uncomfortably
oily or caused irritation to half of respondents.132
   The efficacy of Picaridin is excellent, and it is generally superior to deet in terms of longevity. In a
carefully designed field evaluation against Anopheles gambiae and Anopheles funestus, KBR 3023 in


q 2006 by Taylor & Francis Group, LLC
10                                                      Insect Repellents: Principles, Methods, and Uses


ethanol outperformed deet after a 10-h exposure, and the half-life of the repellent was one hour longer
than that calculated for deet and IR 3535,78 using modelling first used by Rutledge et al. in 1985.133 This
is because KBR 3023 evaporates at a slower rate than deet. Were it not for the lower volatility of
Picaridin, it would probably be less effective, because dose for dose it is less repellent than deet when
freshly applied.134 Other studies have shown a similar performance when compared to deet in field trials:
against Anopheles spp.76 and Verrallina lineata87 in Australia, Aedes (Ochlerotatus) taeniorhynchus in
U.S.,68 Aedes albopictus, Culex quinquefasciatus and Anopheles spp. in Malaysia,135 as well as one field
trial under biting pressures of 1,200–2,400 Aedes cantans and Aedes annulipes landings/person/h.128
KBR 3023 has also shown similar efficacy to deet against Aedes aegypti, Anopheles gambiae,136 and
Amblyomma hebraeum98 in laboratory tests.
   It is this combination of efficacy, safety, and cosmetic appeal that has led to the WHO designating
KBR 3023 as its “repellent of choice for malaria prevention.”137 In addition, the Centers for Disease
Control and Prevention (CDC) recommended both deet and KBR 3023 for West Nile virus and malaria
prevention.138 It is also being investigated for incorporation into military repellents after outperforming
the standard Australian Defence Force formulation of 33% deet.87,132


AI3-35765 and AI3-37220
The piperidine compounds 1-[3-cyclohexen-1-ylcarbonyl] piperidine, called AI3-35765, and
1-[3-cyclohexen-1-ylcarbonyl]-2-methylpiperidine, also known as AI3-37220, were first synthesized
by the USDA in 1978.139 It should be noted, however, that neither of these compounds is
available commercially.
   Research on AI3-35765 showed it to have similar efficacy as deet against Anopheles albimanus,
Anopheles freeborni, Anopheles gambiae, Anopheles stephensi, and Phlebotomus papatasi88; Prosimu-
lium mixtum, and Prosimulium fuscum140; Anopheles stephensi and Culex quinquefasciatus141; as well as
Culex pipiens, both in the laboratory and the field.88 A13-35765 was dropped from the Army research
program, despite its impressive efficacy, because it caused an uncomfortable liniment-like warming
reaction on some peoples’ skin (Dan Strickman, pers. com.). However, recent interest has focused on
AI3-37220, a compound consisting of a racemic mixture of four isomers.142 This mix has proven highly
effective against a variety of blood-feeding arthropods, including Anopheles albimanus, Anopheles
freeborni, Anopheles gambiae, Anopheles stephensi, and Phlebotomus papatasi88; Prosimulium mixtum
and Prosimulium fuscum140; and Aedes communis and Simulium venustum.67 In fact, its longevity was
shown to be superior to that of an equivalent concentration of deet in field trials with Anopheles farauti in
Australia143 and Papua New Guinea,86 Anopheles dirus,144 Anopheles funestus and Anopheles
arabiensis,80 Leptoconops americanus,145 Amblyomma americanum96 and a laboratory trial with
Anopheles stephensi.88 It has undergone extensive toxicology testing and has been deemed safe.146,147
However, it should be noted that it has not yet undergone all of the necessary toxicological testing to
support registration.


SS220
The latest development in synthetic skin repellents is optically active (1S,2S)-2-methylpiperidinyl-3-
cyclohexen-1-carboxamide, discovered by the USDA and dubbed SS220. It is derived from AI3-
37220—insomuch as it is the most repellent of the four isomers that comprise racemic AI3-37220, and is
2.5 times as effective as the racemic mixture against Aedes aegypti.148 Laboratory tests showed SS220 to
be equal to deet against Anopheles stephensi and Aedes aegypti, and better than KBR 3023 against Aedes
aegypti.134 Against the tick, Ixodes scapularis, SS220 outperformed deet and was as good as deet against
Amblyomma americanum.149 However, SS220 is less effective than deet against Anopheles albimanus.150
To date, no field studies have been published, although a USDA report stated that SS220 equals the
effectiveness of 33% deet.151 It has been reviewed by the U.S. military as the new active ingredient to


q 2006 by Taylor & Francis Group, LLC
History of Insect Repellents                                                                               11


replace deet. Extensive toxicological tests have shown low irritation and toxicity.152–156 In addition,
SS220 has a low rate of evaporation that will improve longevity. Using a smaller amount of long-lasting
repellent makes for a more cost-effective and safe product because potential dermal absorption will be
reduced. User acceptability is also likely to be higher because it has a slightly fruity odor, does not have
an oily consistency, and has little plasticizing effect.157 The disadvantage of SS220 lies in the fact that it
is a single stereoisomer, and will, therefore, be more costly to produce than a racemic mixture.
Furthermore, SS220 has not yet been registered, and the huge costs associated with this process,
although necessary, mean that many promising new compounds may never be realized, as developers
need to consider the potential financial benefits of registering a compound versus the initial large
monetary outlay.




Repellent Delivery Methods
Most insect repellents are effective in the vapour phase, defined as vapour or olfactory repellents by
Garson and Winnike as “those materials which are sufficiently volatile to keep an insect at a distance.”158
As the repellent molecules are volatile, temperature, humidity and wind affect evaporation of the
repellent and therefore its longevity.159–161 Perspiration and abrasion will also reduce the longevity of the
repellent.162,163
   Many effective repellents have a high vapor pressure and are therefore volatile. At high mosquito
densities, a heavy dose of a low vapor pressure repellent may be necessary to repel mosquitoes initially,
whereas repellents with high vapor pressures may offer protection at low concentration. Subsequently,
the lower evaporation rate of a repellent with less volatility means that it will continue to repel for a
longer time period.164 For instance, citronella (Cymbopogon nardus) essential oil and pure deet have
similar ED90 values of 112.8 and 95.5 nL/cm2, respectively, for Anopheles gambiaedes.74 However, pure
citronella oil at a dose of 3.33!10K3 mL/cm2 provides protection for only 2 h.165 Citronella contains
actives such as citronellol that has a vapor pressure of 0.009 kPa vs. 0.003 kPa at 208C for deet.166,167
   The high rate of loss of repellents was overcome by using extremely high concentrations of deet,
especially among military personnel. Standard deet concentrations for military repellents were 75%
(U.S.) and 95% (Australia).132 However, repellents may also be lost through dermal absorption.
Absorption of deet is generally high, at 0.8%/h in humans.168 The high rate of dermal absorption
raised safety concerns for adverse side effects associated with using high concentrations of deet. This
prompted several collaborative research studies that eventually resulted in the development of slow-
release formulations based on creams, polymer mixtures, or microcapsules that are available on the
market today. Increased repellent longevity may be achieved in one of three ways: (1) controlling release
or lowering vapor pressure, (2) preventing or reducing repellent absorption, or (3) improving resistance
to abrasion and sweat.
   Formulations can prolong the effect of repellents. Initially, additives such as olive oil169 and mineral
oil170 were used. They may improve repellent longevity by inhibiting loss of repellent volatiles and loss
through sweating and abrasion.171 Additives such as perfume fixatives were also researched. Fixatives
are large branching molecules that lower the vapor pressure of repellents. These included Tibetene and
vanillin, both of which have a significant effect on repellent longevity, increasing it by 29 and 95%,
respectively, when used with deet at a 1:1 ratio.172,173
   In the early 1970s, an intense research program involving military, federal, academic, institutional,
and industrial investigators began with the aim of providing a non-toxic, cosmetically-acceptable, and
effective repellent system that would repel insects for 12 h under tropical conditions. They aimed to
develop a repellent that would provide 24-hour protection under conditions that induced sweating
through a two-pronged approach: (1) searching for agents with higher intrinsic repellency and


q 2006 by Taylor & Francis Group, LLC
12                                                           Insect Repellents: Principles, Methods, and Uses


(2) enhancing repellent protection time of deet using hydrophobic agents to maintain the repellent on
sweating skin.174
   The research on development of a “binding agent” was carried out by the Letterman Army Institute of
Research (LAIR). The formulation of deet with film-forming polymer resins was aided by enlisting the
help of industrial and institutional laboratories. Evaluation of cosmetic properties, dry-skin protection
time, and wash resistance was carried out at LAIR using radio-labelled deet formulated with polymer-
film formers to study evaporation, skin penetration, and wash resistance.174 During the first year of skin
testing, several formulations were developed that were far superior to deet in both dry protection time and
wash resistance, but few were cosmetically acceptable. However, the basic premise that film formers are
extremely effective in enhancing protection time was confirmed. Then, almost a year was spent
attempting to upgrade the cosmetic properties of those formulations that had superior wash resistance.
The majority of research used silicone and carboset acrylic polymers and showed dramatically enhanced
protection times. One example was the use of silicone that improved the dry protection time of deet by a
factor of two, although it did not impart appreciable wash resistance. Tests with carboset acrylic
polymers enhanced the dry protection time of deet and significantly improved wash resistance. Over 150
reformulations were prepared, examined, and about half were studied for wash resistance and cosmetic
appeal on volunteers. However, little success was realized: cosmetically-elegant formulations had
inferior wash resistance, whereas systems having superior wash resistance were sticky or brittle on the
skin. A further year was spent expended in attempting to reformulate carboset polymers to improve their
cosmetic appeal without sacrificing their excellent wash resistance. Formulations of carboset/deet were
combined in increments with silicone polymer (decreasing carboset content in each member of the
series) trying to upgrade cosmetic acceptability without losing the excellent wash resistance.174
   This research at LAIR in the late 1970s and early 1980s established the physical parameters and
theoretical framework that demonstrated the feasibility of polymer and microcapsule mechanisms to
release deet at a predetermined rate. The formulations tested in those early studies utilized microcapsule
and polymer systems designed to provide continuous long-term release of the active ingredient. In
microcapsule formulations, the active ingredient is contained in tiny capsules produced by coacervation,
interfacial polymerization, extrusion, and other processes. The release rate is determined by the size and
number of the microcapsules, the composition and thickness of the microcapsule walls, the concentration
and properties of the excipient, and other additives used. These formulations may also contain free active
repellent in addition to that contained in the microcapsules. In polymer systems, the active ingredient is
formulated with a polymer that will form a thin film over the skin. This film acts as a reservoir for the
active ingredient and slows its absorption and evaporation. In microparticulate controlled-release
systems, the active ingredient is absorbed on the surface of microparticles and released slowly over
time.70 Further research was conducted that looked at formulations based on hydrophilic vinyl polymer,
polyvinylpyrrolidone (PVP),133 before the 3M Corporation’s proprietary polymer formulation was
finally devised. The polymers and microcapsules in the formulations slow the absorption and evaporation
of deet, thereby holding it on the surface of the skin, where it can continue to repel arthropods for an
extended period of time.171
   Ultrathone* (3M) has been the military topical repellent of choice since 1990, when it first became
available in the military supply system. The product contains 33% deet in a controlled-release polymer
base, and is a nongreasy, white lotion with a mild, pleasant odor.175 It was validated by the USDA176 and
chosen as a result of tests against a variety of mosquito species under three climatic regimes: (1) 248C
and 98% relative humidity (RH), (2) 308C and 78% RH, (3) 378C and 31% RH.70 In these tests, the
polymer formulation performed as well as a microparticulate formulation of 42% deet (Biotek) or 75%
deet in alcohol (former standard of the U.S. Army). Field trials in the Philippines with Anopheles
flavirostris showed that 3M was significantly more effective than 71% deet in ethanol for between 6 and
12 h after application.177 However, in Australian field tests against Anopheles farauti, 3M and Biotek

*
    A registered trademark of 3M Corporation, Minneapolis, MN.


q 2006 by Taylor & Francis Group, LLC
History of Insect Repellents                                                                                    13


performed as well as the Army deet formulation.176 In this study, the volunteers applied repellents
themselves according to label instructions, therefore reflecting normal use conditions. There was a
significant difference in the amount of product applied by the individuals, but due to differences in the
deet concentration of the three formulations, the amount of deet applied was fairly consistent.176 The 3M
33% deet polymer formulation was found to be just as effective in repelling mosquitoes in field tests by
the Australian military,87,178 and a 35% deet formulation in cellulose gel that lowered dermal absorption
and evaporation of deet was placed into service in Australia in 1992.132 The British military now also
uses 3M Ultrathon.179 Importantly, the slow-release formulations have significantly lowered dermal
absorption, compared to ethanol formulations with deet at comparable concentrations.180 The addition of
polymers also improves the cosmetic appeal of repellents by lowering the amount of deet available, thus
reducing odor, stickiness, and plasticization, as well as improving abrasion and wash resistance.181
Several polymer and microencapsulated formulations are available on the market, including Sawyer
Controlled Releasew, HourGuardw, Skedaddlew, and Ultrathon.110*
   Gel-based slow-release deet formulations, such as Ultrathone, have low acceptability among troops.
Recently, 10% of American soldiers serving in Kuwait, Haiti, and Bosnia used the U.S. Army repellent
containing 33% deet alone, 29% used commercial formulations, 34% used both types, and 27% used
neither.182 A similar situation was witnessed among Australian troops. Only 26 out of 955 soldiers
interviewed used the standard issue 35% deet formulation in gel base.183 The main reason given for
nonuse is the sticky feel of the repellent on the skin.183 Soldiers do not use military-issued repellents for
several reasons, including their previous familiarity with nonmilitary products before joining the
military, availability of commercial options, and advertising of repellents in various commercial
media.182 Additionally, soldiers’ perceptions of what is acceptable or good has been demonstrated by
Gambel et al. (1998),182 who observed American soldiers declining free military issue (33% deet
formulation) repellent in an olive-green tube for a commercial product that was identical to the military
issue, except that it was packaged using a different name and supplied in a brightly colored tube.


Area Repellents
There has been a recent increase in interest in area repellents that repel all biting insects within a set
distance of the source of repellent molecules. Mosquito coils that are area repellents continue to be the
most popular form of personal protection in use today.40 In addition, citronella candles are commonly
used as insect repellents in backyards and can provide 42% protection.184 Spatial repellents have been
defined as “an inhibiting compound, dispensed into the atmosphere of a three dimensional space which
inhibits the ability of mosquitoes to locate and track a target such as a human or livestock.”185 As
repellents act in the vapor phase, they may potentially have a long-range effect through toxicity or
confusing signals that indicate the presence of a host, established by saturating a zone or space with the
spatial repellent.186 One important concern with area repellents is the fact that they may only be used
under conditions where air flow is minimal—for instance, in forests—as the repellent volatiles may be
diluted with significant air flow.28
   A new development in spatial repellent technology is the ThermaCELLw† Mosquito Repellent system,
consisting of a butane-fueled generator that heats a metal plate to volatilize cis or trans-allethrin from an
impregnated pad. The repellent is effective over a distance of 7 m and provided O90% protection over
6 h from sand flies and mosquitoes in a field trial in Turkey.187 The system vaporizes the active ingredient
from paper mats that are heated. These are highly effective, even under drafty conditions, as shown by
laboratory188 and field trials.189 The various devices that heat such impregnated mats are second only to
mosquito coils in global consumption.40 The ThermaCELL system is an excellent development in

*
  Sawyer Controlled Release is a registered trademark of Spectrum Brands; HourGuard is a registered trademark of 3M
Corporation; and Skedaddle is a registered trademark of Multicrop International Pty. Ltd.
†
  A registered trademark of Schawbel Corportation, MA.


q 2006 by Taylor & Francis Group, LLC
14                                                      Insect Repellents: Principles, Methods, and Uses


repellent research, as it allows several people to be protected at once. However, it costs approximately
U.S. $1.00 per hour to run, making it too expensive for use in the developing world.
   Several experiments with y-port olfactometers have recently shown the repellent effect of plant-based
components such as the essential oil of catnip.190 Plant-based repellents usually have a short longevity
when applied to the skin, as they have high vapor pressure.166,191 However, it is this feature that makes
them excellent spatial repellents. In a field test, mint (Mentha arvensis) oil volatilized using a kerosene
lamp significantly protected volunteers from Mansonia titillans,28 and field experiments have demon-
strated the spatial repellent effect of volatilization of plant oils using heated plates against Anopheles
gambiae.29




The Evolution of Repellent Testing
Kairomones
Progress in the development of new repellents has been slow until the recent breakthroughs, perhaps due
to improved understanding of the repellents’ modes of action on the target organisms. However, this is
now changing, and many papers have now been published on the mode of action of host kairomones on
host-seeking insects.14,192–202 Delicate methods, such as electroantennogram readings of the response of
sensory neurons in insect antennae to attractive and repellent compounds, have allowed greater
understanding of insects’ sensory systems.19,203–205 Y-port olfactometers have also proved very useful
in discriminating the effect of kairomones,206–213 as well as insect repellents and inhibitors. They have
shown the importance of the interplay between whole host odor and repellents. Olfactometer experiments
with deet have shown that it is not a true repellent insomuch as it causes insects to make oriented
movements away from its source, but it is an inhibitor that prevents insects from feeding on a host in its
presence (in this case a host-derived odor blend).186,214 Many groups are working on quantifying what
elements of human skin and breath are actually attractive to host-seeking insects, and, in particular,
highly-anthropophilic disease vectors. Cork and Park (1996)215 chemically fractionated human sweat
samples into acid and nonacid components. They measured the electrical response of sensillae in the
antennae of mosquitoes, and found that short-chained aliphatic acids (C2–C8) elicited significantly
greater responses than the longer-chained acids. These acids elicit a landing response198 and they have a
significant effect on mosquito host-seeking behavior.203 There is a growing body of evidence indicating
short-chained fatty acids are reliable cues; however, these require complex blends, including synergists
such as ammonia and lactic acid.206,216 Therefore, the potential for an olfactometer with a reliable
synthetic lure for repellent testing is some way off.


Choice
Should olfactometers become used regularly in the future for repellent/inhibitor screening, they will only
ever be suitable for preliminary screening because olfactometers allow insects to choose between one or
more targets. This is a disadvantage, as it causes an inflation of repellent efficacy: it “shifts the point of
reference for the ED50 to a lower level.”217 It was argued that “free choice” between repellent-treated and
untreated areas more accurately reflects use conditions where mosquitoes will feed on untreated areas of
a repellent user, or their untreated companions.218 This, however, is not a useful scenario, as a single
infected bite is sufficient to transmit vector-borne pathogens. Therefore, recent publications have stressed
the importance of high (O95%) protection, where the mosquito has no choice but to feed on repellent-
treated skin if they wish to feed at all.78,219 In addition, several experiments have demonstrated that
offering mosquitoes free choice in laboratory overestimates repellency.74,220 A free-choice test
calculated the ED50 of deet as 0.024–0.042 mg/cm2, 221 whereas a similar test with no choice calculated
it as 0.35 mg/cm2. 220 In field tests, Barnard et al.68 showed that the application of a repellent to one limb


q 2006 by Taylor & Francis Group, LLC
History of Insect Repellents                                                                              15


where the other was used as a control inflates repellency, while the use of repellent using a repellent-
wearing “bait” individual with an untreated “collector” also overestimates repellency.222 A study in
Vietnam was performed with Anopheles dirus using one pair (one wearing deet, and one solvent control)
and one individual (wearing deet) sitting alone.223 In this case, the repellent wearer that was sitting alone
received 3.5 times more mosquito bites than the repellent wearer that was sitting close to an alternate
blood source. This is because the mosquitoes will always feed on the “easiest option” with least repellent,
be that an adjacent area of skin, an alternate limb, or another individual. When the protection afforded by
the repellent wanes, mosquitoes will start to feed through the repellent. However, if there is an
unprotected alternative, they will be diverted and feed upon it. This also applies in field tests if
individuals are less than 10 m apart because this is the limit of short-range attraction.224


In Vitro and Animal Tests
Tests on repellents, from the 1920s until recently, were often performed on shaven animals including
rabbits, dogs,225 guinea pigs,226 and chicks.227 This method, despite questions regarding the ethical
treatment of animals, may distort the results of repellent tests. Nicolaides et al. (1968)228 compared the
skin of humans and other domestic animals. They concluded that humans excrete mainly triglycerides and
are, therefore, unique in having fatty acids as breakdown products on the skin surface. This means that
short-chained aliphatic acids are reliable host cues for anthropophilic mosquitoes, and, therefore, testing
repellents on animals will not give representative data of how the repellent will perform when applied to
human skin. In addition, the most efficient malaria vectors are extremely anthropophilic, and will be less
attracted to nonhuman hosts,212 possibly due to genetically mediated innate preferences.229 Thus, this
method gave a distorted measure of repellency. Indeed, Rutledge et al. directly compared measurements of
repellent efficacy obtained using rabbits171 and mice230 with that obtained using human arms. In both
cases, repellents showed greater variability and greater persistence when applied to animals than humans.
   Other studies have utilized membrane blood feeders, commonly used for feeding mosquitoes in
insectaries, to measure repellency.217 Although the data obtained using this method roughly correspond
to data obtained with human-arm tests,220 this method should be used only for rapidly screening large
numbers of repellents to narrow down candidates for further testing. This is because membrane feeder
tests differ from human-arm tests because mosquitoes do not respond as enthusiastically to a feeder as
they do to a living host, and there is much interspecific variation in readiness to feed from membrane
feeders.231 Another testing method employs disks of paper impregnated with a test repellent, and the
numbers of insect landings on impregnated and unimpregnated control disks are counted. This was
shown to be an excellent method for testing irritancy of a chemical, but is not a measure of repellency.232
In vitro methods are cheap, and yield many results rapidly with no risk to human subjects, but they do not
accurately mimic the conditions of repellent usage. Thus, different methodologies cannot be compared,
nor can their results be directly extrapolated to the end user. This is particularly important since the
discovery that deet, the leading insect repellent, is an inhibitor and not a true repellent.214


Test Standardization
Other recent developments in repellents research have followed after the call of the WHO in 2000233 to
standardize repellent testing protocols. It would appear that tests are becoming far more stringent
and standardized.
   It is always preferable to conduct tests on human volunteers for greatest accuracy, provided that
laboratory-reared mosquitoes are used to eliminate the risk of pathogen transmission, and the selected
volunteers show mild or no allergic reaction to mosquito bites.234 It is conventional to use Aedes aegypti
mosquitoes for repellent testing, but people generally show milder reactions to Anopheles bites. Aedes
aegypti are commonly used, as they are easy to rear under laboratory conditions, and are avid biters.
However, several other species also fulfill these criteria, including Anopheles stephensi, Anopheles


q 2006 by Taylor & Francis Group, LLC
16                                                      Insect Repellents: Principles, Methods, and Uses


gambiae s.s., Anopheles arabiensis, and Anopheles albitarsis. The U.S. EPA now recommends using
Aedes aegypti along with a representative human biting species from both the Anopheles and Culex
genera for laboratory studies of repellent efficacy.235 It is preferred to perform bioassays on the vectors in
the region for which the repellent is to be used233 because the sensitivity of different mosquito species to
repellents varies.74 In addition, as deet is the active ingredient of most commercially available skin
repellents and is the most effective and well-researched insect repellent available at this time, it is now
considered a useful standard against which the effectiveness of alternative repellents may be judged.234
   Laboratory tests are generally conducted with mosquitoes held in large laboratory cages into which the
forearm(s) of the volunteer is introduced with the hand protected by a glove. The whole forearm may be
exposed, or a 25 cm2 area of skin, the remainder being covered with a rubber sleeve. In some tests, the
repellent is applied to a cotton stocking, as repellents are much more persistent on fabric than on skin
because loss of repellent through abrasion, skin absorption, evaporation and sweating is reduced. The
stocking is drawn over another stocking that has been drawn over the arm to prevent skin contact with a
repellent compound; this is particularly important for volunteers involved in regular testing, or where
compounds have not been screened for toxicity or dermal absorption. However, it was shown that this
method does not correlate well with results from tests where repellent is applied directly to the skin,236
and further studies need to be performed when the substance is deemed safe for use on the skin after
toxicological evaluation.
   A major source of variation in laboratory tests is caused by differences in mosquito avidity related to
their physiological state. The team from the USDA Agricultural Research Service’s Mosquito and Fly
Research Unit have made excellent progress toward standardization of repellent testing methodology.
They have published a series of important papers showing that mosquito attack rates, and consequently
repellent protection times, are significantly influenced by mosquito body size (hence larval nutrition), the
age and parity of the mosquitoes, as well as the time of day.237–240 Allowing the mosquitoes access to
sugar solution or blood will also decrease their avidity, and subsequently, the measured repellency of a
chemical because they will be at least partially engorged.241 Also of importance is the density of the
mosquitoes in the cage. For “time to first bite,” it was shown that the most rigorous tests required
densities of mosquitoes where each mosquito had 49 cm3. 239 These experiments have drawn the
conclusion that mosquitoes used for repellent testing should therefore be nulliparous, aged between 3
and 10 d, and denied access to sugar prior to testing repellents.
   In addition, the EPA’s FIFRA (Federal Insecticide and Rodenticide Act) Scientific Advisory Panel has
advised that the commonly used “time to first bite” test should no longer be utilized.219 This test was
commonly used where one arm is treated with 1 mL of a 25% solution of the test compound in ethanol.
The arm is exposed for 3 min in every 30 min and the first time after treatment noted at which a bite
occurs followed by a “confirmatory” bite in the same or the following exposure period. However, it has
been concluded that the time to first bite method has not been developed using a statistically valid
approach because its result depends on the behavior of a few individuals in the upper distribution of
tolerance and does not reflect the behavior of the population as a whole. Therefore, it is increasingly
recognized that methods that measure 95% reduction in bites are preferable because all bites are counted
and the method provides a more “real-world” assessment of insect repellent efficacy.219 This method
requires sequential exposure of an arm with zero, and then progressively higher, doses of repellent for
30 s to cages containing approximately 50 hungry Anopheles gambiae (or 45 s with Anopheles
stephensi). The number biting at the end of the short exposure is quickly counted (preferably with the
help of an assistant) and the mosquitoes are then shaken off before they can imbibe any blood. Hence,
the same mosquitoes can be used for testing each dose, and their continued hunger can be checked by
exposing the other untreated arm. Probit analysis is used to calculate the ED50, ED90, or ED95. After
reaching a dose that gives 100% repellency, the arm is re-exposed hourly until repellency declines to
50% compared with contemporary counts on the untreated arm to measure the duration of this protection.
This is a labor-intensive method of repellent evaluation, but it allows the direct comparison of repellents
via the ED values. It also measures the relative tolerance of different species to repellents.


q 2006 by Taylor & Francis Group, LLC
History of Insect Repellents                                                                                    17


   A new, less labor-intensive method that may be used to calculate the probit curve of a repellent is the
K&D module.242 The K&D module is made of plexiglas and has six cells. Each cell has a stoppered
access hole for transfer of mosquitoes to the cell, and a bottom with a rectangular 3 cm ! by 4 cm hole
that is opened and closed by a sliding door.
   The concave bottom conforms to the curvature of a human thigh and a separate bottom section with the
same dimensions serves as a skin-marking template. A human test subject wearing shorts, seated with
legs horizontally extended, uses the template and a water-soluble marker to denote the areas to which
repellent or control is applied that correspond to the openings of the module. The test compounds are
exposed to mosquitoes by placing the K&D module over it and opening the cell doors. The number of
insects biting in each cell within the 2-min exposure is recorded, after which the doors are closed. At the
conclusion of each assay, mosquitoes are freed by opening cells of the K&D module in a sleeved,
screened cage. The method can be used to calculate the ED90 of a compound by applying incremental
doses of repellent, and may prove to be a simple and efficient rapid screening tool.
   Although laboratory tests are extremely useful, it is now generally agreed that field tests are the
definitive test of a repellent, as they allow the evaluation of a substance under representative user
conditions.219,233,234 Field tests allow the evaluation of a repellent with the desired test species, under
the environmental conditions that it will be required to perform. Tests that use a Latin square or “round
robin” design are favored to ensure that an adequate number of replicates are employed, as individual
attractiveness to hematophagous insects, as well as their ability to capture them in tests, varies
widely.219
   Repellent science has advanced greatly in the last decade and will continue to progress in the future.
New methods such as the molecular modelling and characterization of repellent molecules that attempt to
explain the structure–activity relationship of repellent molecules, especially their stereochemical activity
relationships, are beginning to emerge.243 In the future, new repellents may be discovered based on their
molecular structure and tested in the field using host-odor-baited traps. This will remove the risk
associated with testing repellents against the insect vectors of pathogens, the infection with which they
are designed to prevent. Until then, the recent advances in laboratory science mean that the rapid and
accurate screening of candidate repellent compounds that closely represent field conditions is becoming
more and more attainable.

References
      1. M. Baker, Fur rubbing: Use of medicinal plants by capuchin monkeys (Cebus capucinus), American
         Journal of Primatology, 38, 263, 1996.
      2. C. R. Birkinshaw, Use of millipedes by black lemurs to anoint their bodies, Folia Primatologica, 70,
         170, 1999.
      3. D. J. Overdorff, Similarities, differences, and seasonal patterns in the diets of Eulemur fulvus rufus and
         Eulemur rubiventer in the Ranomafana National Park, Madagascar, International Journal of
         Primatology, 14, 721, 1993.
      4. M. Zito, S. Evans, and P. J. Weldon, Owl monkeys (Aotus spp.) self anoint with plants and millipedes,
         Folia Primatologica, 74, 158, 2003.
      5. X. Valderrama, J. G. Robinson, A. B. Attygalle, and T. Eisner, Seasonal anointment with millipedes in
         a wild primate: A chemical defense against insects?, Journal of Chemical Ecology, 26, 2781, 2000.
      6. P. J. Weldon, J. R. Aldrich, J. A. Klun, J. E. Oliver, and M. Debboun, Benzoquinones from millipedes
         deter mosquitoes and elicit self-anointing in capuchin monkeys (Cebus spp.), Naturwissenschaften,
         90, 301, 2003.
      7. J. F. Carroll, M. Kramer, P. J. Weldon, and R. G. Robbins, Anointing chemicals and ectoparasites:
         Effects of benzoquinones from millipedes on the lone star tick Amblyomma americanum, Journal of
         Chemical Ecology, 31, 63, 2005.
      8. K. C. Parkes, P. J. Weldon, and R. L. Hoffman, Polydesmidan millipede used in self-anointing by a
         strong-billed woodcreeper (Xiphocolaptes promeropirhyncus) from Belize, Ornitologia Neotropica,
         14, 285, 2003.


q 2006 by Taylor & Francis Group, LLC
18                                                        Insect Repellents: Principles, Methods, and Uses


      9. P. J. Weldon, Defensive anointing: Extended chemical phenotype and unorthodox ecology,
         Chemoecology, 14, 1, 2004.
     10. Herodotus, The Histories, London: Penguin, 1996 (reprint).
     11. D. Charlwood, Did Herodotus describe the first airborne use of mosquito repellents?, Trends in
         Parasitology, 19, 555, 2003.
     12. R. Romi, G. Sabatinelli, and G. Majori, Could malaria reappear in Italy?, Emerging Infectious Disease,
         7, 915, 2001.
     13. T. Owen, Geoponika: Agricultural Pursuits, http://www.ancientlibrary.com/geoponica/index.html,
         1805.
     14. M. A. Braks, R. A. Anderson, and B. G. Knols, Infochemicals in mosquito host selection: Human skin
         microflora and Plasmodium parasites, Parasitology Today, 15, 409, 1999.
     15. R. de Jong and B. G. J. Knols, Selection of biting sites on man by two malaria mosquito species,
         Experientia, 51, 80, 1995.
     16. Y. L. Konan, M. S. Sylla, J. M. Doannio, and S. Traore, Comparison of the effect of two excipients
         (karite nut butter and vaseline) on the efficacy of Cocos nucifera, Elaeis guineensis and Carapa
         procera oil-based repellents formulations against mosquitoes biting in Ivory Coast, Parasite, 10, 181,
         2003.
     17. D. M. S. S. Babu (Ed.), Yoga Ratnakara Visama Jvara Cikitsa-1, in Yoga Ratnakara: Vol. I—
         Purvardham (Sanskrit text with English translation and explanatory notes), Chowkhamba: Sanskrit
         Series Office, p. 698, 2005.
     18. M. J. Bockarie et al., The effect of woodsmoke on the feeding and resting behavior of Anopheles
         gambiae s.s, Acta Tropica, 57, 337, 1994.
     19. E. E. Davis and M. F. Bowen, Sensory physiological basis for attraction in mosquitoes, Journal of the
         American Mosquito Control Association, 10, 316, 1994.
     20. D. E. Moerman, Native American Ethnobotany, Portland, OR: Timber Press, 1998.
     21. K. Kinney, Black Swamp once ruled the land and people, in Sentinel Tribune, Bowling Green, OH,
         1999.
     22. K. T. Silva, Ayurveda, malaria and the indigenous herbal tradition in Sri Lanka, Social Science and
         Medicine, 33, 153, 1991.
     23. K. Palsson and T. G. Jaenson, Comparison of plant products and pyrethroid-treated bed nets for
         protection against mosquitoes (Diptera: Culicidae) in Guinea Bissau, West Africa, Journal of Medical
         Entomology, 36, 144, 1999.
     24. J. M. Dulhunty, K. Yohannes, C. Kourleoutov, V. T. Manuopangai, M. K. Polyn, W. J. Parks, and
         J. H. Bryan, Malaria control in central Malaita, Solomon Islands 2. Local perceptions of the disease
         and practices for its treatment and prevention, Acta Tropica, 75, 185, 2000.
     25. A. D. Rodriguez, R. P. Penilla, M. Henry-Rodriguez, J. Hemingway, A. Francisco Betanzos, and
         J. E. Hernandez-Avila, Knowledge and beliefs about malaria transmission and practices for vector
         control in southern Mexico, Salud Publica de Mexico, 45, 110, 2003.
     26. R. E. Klein, S. C. Weller, R. Zeissig, F. O. Richards, and T. K. Ruebush II, Knowledge, beliefs, and
         practices in relation to malaria transmission and vector control in Guatemala, American Journal of
         Tropical Medicine and Hygiene, 52, 383, 1995.
     27. P. T. Tin-Oo, Pe-Thet-Htoon, Khin-Thet-Wai, W. Parks, and J. Bryan, Gender, mosquitos and
         malaria: Implications for community development programs in Laputta, Myanmar, Southeast Asian
         Journal of Tropical Medicine and Public Health, 32, 588, 2001.
     28. S.J. Moore, A methodology for developing plant-based products for use against Anopheles
         mosquitoes, Ph.D Thesis, London: London School of Hygiene and Tropical Medicine, 2005.
     29. A. Seyoum, G. F. Killeen, E. W. Kabiru, B. G. Knols, and A. Hassanali, Field efficacy of thermally
         expelled or live potted repellent plants against African malaria vectors in western Kenya, Tropical
         Medicine and International Health, 8, 1005, 2003.
     30. S. W. Lindsay, J. A. Ewald, Y. Samung, C. Apiwathnasorn, and F. Nosten, Thanaka (Limonia
         acidissima) and deet (di-methyl benzamide) mixture as a mosquito repellent for use by Karen Women,
         Medical and Veterinary Entomology, 12, 295, 1998.


q 2006 by Taylor & Francis Group, LLC
History of Insect Repellents                                                                                  19


    31. R. Vernede, M. M. van Meer, and M. P. Alpers, Smoke as a form of personal protection against
        mosquitoes, a field study in Papua New Guinea, Southeast Asian Journal of Tropical Medicine and
        Public Health, 25, 771, 1994.
    32. J. Stevenson, Zooprophylactic effect of goat herds on human exposure to Phlebotomus martini
        sandflies in East Africa, Transactions of the Royal Society of Tropical Medicine and Hygiene,
        submitted for publication, 2005.
    33. E. J. Hoffman and J. R. Miller, Reduction of mosquito (Diptera: Culicidae) attacks on a human subject
        by combination of wind and vapor-phase DEET repellent, Journal of Medical Entomology, 39, 935,
        2002.
    34. K. R. Smith and S. Mehta, The burden of disease from indoor air pollution in developing
        countries: Comparison of estimates, International Journal of Hygiene and Environmental Health,
        206, 279, 2003.
    35. R. A. Cloyd, Natural indeed: Are natural insecticides safer better than conventional insecticides?,
        Illinois Pesticide Review, 17, 2004 (http://www.pesticidesafety.uiuc.edu/newsletter/html/v17n304.
        pdf).
    36. R. Eisner, ‘Natural’ insecticide research: Still working out the bugs, The Scientist, 5, 14, 1991.
    37. J. E. Casida and G. B. Quistad, Pyrethrum Flowers: Production, Chemistry, Toxicology and Uses,
        New York: Oxford University Press, Inc., 1995.
    38. H. Sangat-Roemantyo, Ethnobotany of Javanese incense, Economic Botany, 44, 413, 1990.
    39. M. Uemura, Eiichiro Ueyama: Developing and promoting insecticide together with pyrethrum, Osaka
        Business Update, 4, 2004 (http://www.ibo.or.jp/e/2004_4/01_1/1_1.html).
    40. WHO, Draft guideline specifications for household insecticide products—mosquito coils, vaporising
        mats, liquid vaporisers, aerosols. Report of the WHO Informal Consultation, Geneva: World Health
        Organization, February 3–6, 1998.
    41. M. S. Mulla, U. Thavara, A. Tawatsin, W. Kong-Ngamsuk, and J. Chompoosri, Mosquito burden and
        impact on the poor: Measures and costs for personal protection in some communities in Thailand,
        Journal of the American Mosquito Control Association, 17, 153, 2001.
    42. M. Ettling, D. A. McFarland, L. J. Schultz, and L. Chitsulo, Economic impact of malaria in Malawian
        households, Tropical Medicine and Parasitology, 45, 74, 1994.
    43. C. E. Lawrance and A. M. Croft, Do mosquito coils prevent malaria? A systematic review of trials,
        Journal of Travel Medicine, 11, 92, 2004.
    44. T. Narahashi, Neuroreceptors and ion channels as the basis for drug action: Past, present, and future,
        Journal of Pharmacology and Experimental Therapeutics, 294, 1, 2000.
    45. D. R. Roberts, T. Chareonviriyaphap, H. H. Harlan, and P. Hshieh, Methods of testing and analyzing
        excito-repellency responses of malaria vectors to insecticides, Journal of the American Mosquito
        Control Association, 13, 13, 1997.
    46. M. Elliott, N. F. Janes, and C. Potter, The future of pyrethroids in insect control, Annual Review of
        Entomology, 23, 443, 1978.
    47. M. Elliott, Properties and applications of pyrethroids, Environmental Health Perspectives, 14, 1, 1976.
    48. D. M. Soderlund and J. R. Bloomquist, Neurotoxic actions of pyrethroid insecticides, Annual Review of
        Entomology, 34, 77, 1989.
    49. P. R. Chadwick, The activity of some pyrethroids, DDT and lindane in smoke from coils for biting
        inhibition, knockdown and kill of mosquitoes (Diptera, Culicidae), Bulletin of Entomological Research,
        65, 97, 1975.
    50. S. P. Frances, Effectiveness of DEET and permethrin, alone, and in a soap formulation as skin and
        clothing protectants against mosquitoes in Australia, Journal of the American Mosquito Control
        Association, 3, 648, 1987.
    51. M. Rowland, T. Freeman, G. Downey, A. Hadi, and M. Saeed, DEET mosquito repellent sold through
        social marketing provides personal protection against malaria in an area of all-night mosquito biting and
        partial coverage of insecticide-treated nets: A case–control study of effectiveness, Tropical Medicine
        and International Health, 9, 343, 2004.
    52. G. Covell, Anti-mosquito measures with special reference to India, Health Bulletin, 11, pp. 1–60,
        1943.


q 2006 by Taylor & Francis Group, LLC
20                                                          Insect Repellents: Principles, Methods, and Uses


     53. F. H. Mowrey, Statistics of malaria, in Internal Medicine in World War II, Volume II: Infectious
         Diseases, L. D. Heaton, J. B. Coates, and W. P. Havens (Eds.), Washington, DC: Medical
         Department, United States Army, pp. 449–463, 1963, http://history.amedd.army.mil/booksdocs/
         wwii/infectiousdisvolii/frameindex.html.
     54. R. W. Bunn, K. L. Knoght, and W. J. Lacasse, The role of entomology in the preventive medicine
         program of the Armed Forces, Military Medicine, 116, 119, 1955.
     55. C. J. D. Zarafonetis and M. P. Baker, Scrub typhus, in Internal Medicine in World War II, Volume
         II: Infectious Diseases, L. D. Heaton, J. B. Coates, and W. P. Havens (Eds.), Washington, DC:
         Medical Department, United States Army, pp. 111–142, 1963, http://history.amedd.army.mil/
         booksdocs/wwii/infectiousdisvolii/frameindex.html.
     56. T. F. Whayne, Clothing, in Personal Health Measures and Immunization, J. B. Coates, E. C. Hoff,
         and P. M. Hoff, (Eds.), Office of the Surgeon General Department of the Army, Washington, DC,
         pp. 31–83, 1955, http://history.amedd.army.mil/booksdocs/wwii/PrsnlHlthMsrs/frameindx.html.
     57. U.S. Department of Agriculture, Results of Screening Tests with Materials Evaluated as
         Insecticides, Miticides and Repellents at the Orlando, Orlando: Florida Laboratory E-733,
         Bureau of Entomology and Plant Quarantine United States Department of Agriculture, 1947.
     58. C. Peterson and J. Coats, Insect repellents—past, present and future, Pesticide Outlook, 12, 154, 2001.
     59. A. H. Madden, A. W. Lindquist, and E. F. Knipling, Tests of repellents against chiggers, Journal of
         Economic Entomology, 37, 283, 1944.
     60. R. N. McCulloch, Studies in the control of scrub typhus, Medical Journal of Australia, 1, 717, 1946.
     61. R. N. McCulloch and D. F. Waterhouse, Laboratory and field tests of mosquito repellents, in CSIR
         Bulletin No. 213, Melbourne: Australian Government Printer, 1947, p. 1.
     62. Fisher Scientific, Material safety data sheet: 2-ethyl-1,3-hexanediol, mixture of isomers, 99%, in http://
         fscimage.fishersci.com/msds/71746.htm, 1998.
     63. T. L. Neeper-Bradley, L. C. Fisher, B. L. Butler, and B. Ballantyne, Evaluation of the developmental
         toxicity potential of 2-ethyl-1,3-hexanediol in the rat by cutaneous application, Journal of Toxicology:
         Cutaneous and Ocular Toxicology, 13, 203, 1994.
     64. E. T. McCabe, W. F. Barthel, S. I. Gertler, and S. A. Hall, Insect repellents, III, N,N-diethylamides,
         Journal of Organic Chemistry, 19, 493, 1954.
     65. J. R. Roberts and J. R. Reigart, Does anything beat DEET?, Pediatric Annual, 33, 443, 2004.
     66. C. E. Schreck and T. P. McGovern, Repellents and other personal protection strategies against Aedes
         albopictus, Journal of the American Mosquito Control Association, 5, 247, 1989.
     67. M. Debboun, D. Strickman, V. B. Solberg, R. C. Wilkerson, K. R. McPherson, C. Golenda, L. Keep,
         R. A. Wirtz, R. Burge, and T. A. Klein, Field evaluation of DEET and a piperidine repellent against
         Aedes communis (Diptera: Culicidae) and Simulium venustum (Diptera: Simuliidae) in the Adirondack
         Mountains of New York, Journal of Medical Entomology, 37, 919, 2000.
     68. D. R. Barnard, U. R. Bernier, K. H. Posey, and R. D. Xue, Repellency of IR3535, KBR3023, para-
         menthane-3,8-diol, and DEET to black salt marsh mosquitoes (Diptera: Culicidae) in the Everglades
         National Park, Journal of Medical Entomology, 39, 895, 2002.
     69. J. E. Cilek, J. L. Petersen, and C. E. Hallmon, Comparative efficacy of IR3535 and DEET as repellents
         against adult Aedes aegypti and Culex quinquefasciatus, Journal of the American Mosquito Control
         Association, 20, 299, 2004.
     70. R. K. Gupta and L. C. Rutledge, Controlled release repellent formulations on human volunteers under
         three climatic regimens, Journal of the American Mosquito Control Association, 7, 490, 1991.
     71. Y. Trongtokit, Y. Rongsriyam, N. Komalamisra, P. Krisadaphong, and C. Apiwathnasorn, Laboratory
         and field trial of developing medicinal local Thai plant products against four species of mosquito
         vectors, Southeast Asian Journal of Tropical Medicine and Public Health, 35, 325, 2004.
     72. B. Annis, Comparison of the effectiveness of two DEET formulations against Aedes albopictus in the
         Philippines, Journal of the American Mosquito Control Association, 7, 543, 1991.
     73. D. R. Barnard and R. D. Xue, Laboratory evaluation of mosquito repellents against Aedes albopictus,
         Culex nigripalpus, and Ochierotatus triseriatus (Diptera: Culicidae), Journal of Medical Entomology,
         41, 726, 2004.


q 2006 by Taylor & Francis Group, LLC
History of Insect Repellents                                                                                 21


    74. C. F. Curtis, J. D. Lines, J. Ijumba, A. Callaghan, N. Hill, and M. A. Karimzad, The relative
        efficacy of repellents against mosquito vectors of disease, Medical and Veterinary Entomology, 1,
        109, 1987.
    75. R. E. Coleman, A. L. Richards, G. J. Magnon, C. S. Maxwell, M. Debboun, T. A. Klein, and
        R. A. Wirtz, Laboratory and field trials of four repellents with Culex pipiens (Diptera: Culicidae),
        Journal of Medical Entomology, 31, 17, 1994.
    76. S. P. Frances, N. Van Dung, N. W. Beebe, and M. Debbboun, Field evaluation of repellent
        formulations containing DEET and picaridin against mosquitoes in Northern Territory, Australia,
        Journal of Medical Entomology, 41, 414, 2004.
    77. G. L. Chiang, S. L. Tay, K. L. Eng, and S. T. Chan, Effectiveness of repellent/insecticidal bars against
        malaria and filariasis vectors in peninsular Malaysia, Southeast Asian Journal of Tropical Medicine
        and Public Health, 21, 412, 1990.
    78. C. Costantini, A. Badolo, and E. Ilboudo-Sanogo, Field evaluation of the efficacy and persistence of
        insect repellents DEET, IR3535, and KBR 3023 against Anopheles gambiae complex and other
        Afrotropical vector mosquitoes, Transactions of the Royal Society of Tropical Medicine and Hygiene,
        98, 644, 2004.
    79. J. K. Trigg, Evaluation of a eucalyptus-based repellent against Anopheles spp. in Tanzania, Journal of
        the American Mosquito Control Association, 12, 243, 1996.
    80. T. W. Walker, L. L. Robert, R. A. Copeland, A. K. Githeko, R. A. Wirtz, J. I. Githure, and
        T. A. Klein, Field evaluation of arthropod repellents, DEET and a piperidine compound, AI3-37220,
        against Anopheles funestus and Anopheles arabiensis in western Kenya, Journal of the American
        Mosquito Control Association, 12, 172, 1996.
    81. J. Govere, D. N. Durrheim, L. Baker, R. Hunt, and M. Coetzee, Efficacy of three insect repellents
        against the malaria vector Anopheles arabiensis, Medical and Veterinary Entomology, 14, 441, 2000.
    82. J. Govere, L. E. Braack, D. N. Durrheim, R. H. Hunt, and M. Coetzee, Repellent effects on Anopheles
        arabiensis biting humans in Kruger Park, South Africa, Medical and Veterinary Entomology, 15, 287,
        2001.
    83. S. P. Frances, N. Eikarat, B. Sripongsai, and C. Eamsila, Response of Anopheles dirus and Aedes
        albopictus to repellents in the laboratory, Journal of the American Mosquito Control Association, 9,
        474, 1993.
    84. S. P. Frances, T. A. Klein, D. W. Hildebrandt, R. Burge, C. Noigamol, N. Eikarat, B. Sripongsai, and
        R. A. Wirtz, Laboratory and field evaluation of DEET, CIC-4, and AI3-37220 against Anopheles
        dirus (Diptera: Culicidae) in Thailand, Journal of Medical Entomology, 33, 511, 1996.
    85. S. J. Moore, A. Lenglet, and N. Hill, Field evaluation of three plant-based insect repellents against
        malaria vectors in Vaca Diez Province, the Bolivian Amazon, Journal of the American Mosquito
        Control Association, 18, 107, 2002.
    86. S. P. Frances, R. D. Cooper, S. Popat, and A. W. Sweeney, Field evaluation of the repellents DEET,
        CIC-4, and AI3-37220 against Anopheles in Lae, Papua New Guinea, Journal of the American
        Mosquito Control Association, 15, 339, 1999.
    87. S. P. Frances, N. Van Dung, N. W. Beebe, and M. Debboun, Field evaluation of repellent
        formulations against daytime and nighttime biting mosquitoes in a tropical rainforest in northern
        Australia, Journal of Medical Entomology, 39, 541, 2002.
    88. R. E. Coleman, L. L. Robert, L. W. Roberts, J. A. Glass, D. C. Seeley, A. Laughinghouse,
        P. V. Perkins, and R. A. Wirtz, Laboratory evaluation of repellents against four anopheline
        mosquitoes (Diptera: Culicidae) and two phlebotomine sand flies (Diptera: Psychodidae), Journal
        of Medical Entomology, 30, 499, 1993.
    89. B. Alexander, H. Cadena, M. C. Usma, and C. A. Rojas, Laboratory and field evaluations of a
        repellent soap containing diethyl toluamide (DEET) and permethrin against phlebotomine sand flies
        (Diptera: Psychodidae) in Valle del Cauca, Colombia, American Journal of Tropical Medicine and
        Hygiene, 52, 169, 1995.
    90. M. Kalyanasundaram, R. Srinivasan, S. Subramanian, and K. N. Panicker, Relative potency of DEPA
        as a repellent against the sand fly Phlebotomus papatasi, Medical and Veterinary Entomology, 8, 68,
        1994.


q 2006 by Taylor & Francis Group, LLC
22                                                         Insect Repellents: Principles, Methods, and Uses


    91. L. L. Robert, R. E. Coleman, D. A. Lapointe, P. J. Martin, R. Kelly, and J. D. Edman, Laboratory and
        field evaluation of five repellents against the black files Prosimulium mixtum and P. fuscum (Diptera:
        Simuliidae), Journal of Medical Entomology, 29, 267, 1992.
    92. T. M. Ho and M. K. Fauziah, Laboratory evaluation of two commercial repellants against
        Leptotrombidium fletcheri (Acari: Trombiculidae), Southeast Asian Journal of Tropical Medicine
        and Public Health, 24, 165, 1993.
    93. S. P. Frances, Response of a chigger, Eutrombicula histri (Acari: Trombiculidae) to repellent and
        toxicant compounds in the laboratory, Journal of Medical Entomology, 31, 628, 1994.
    94. C. Eamsila, S. P. Frances, and D. Strickman, Evaluation of permethrin-treated military uniforms for
        personal protection against malaria in northeastern Thailand, Journal of the American Mosquito
        Control Association, 10, 515, 1994.
    95. S. Kumar, S. Prakash, and K. M. Rao, Comparative activity of three repellents against the ticks
        Rhipicephalus sanguineus and Argas persicus, Medical and Veterinary Entomology, 6, 47, 1992.
    96. V. B. Solberg, T. A. Klein, K. R. McPherson, B. A. Bradford, J. R. Burge, and R. A. Wirtz, Field
        evaluation of DEET and a piperidine repellent (AI3-37220) against Amblyomma americanum (Acari:
        Ixodidae), Journal of Medical Entomology, 32, 870, 1995.
    97. B. Salafsky, Y. X. He, J. Li, T. Shibuya, and K. Ramaswamy, Short report: Study on the efficacy of a
        new long-acting formulation of N,N-diethyl-m-toluamide (DEET) for the prevention of tick
        attachment, American Journal of Tropical Medicine and Hygiene, 62, 169, 2000.
    98. A. M. Pretorius, M. Jensenius, F. Clarke, and S. H. Ringertz, Repellent efficacy of DEET and KBR 3023
        against Amblyomma hebraeum (Acari: Ixodidae), Journal of Medical Entomology, 40, 245, 2003.
    99. S. Kumar, S. Prakash, and K. M. Rao, Comparative activity of three repellents against bedbugs Cimex
        hemipterus (Fabr.), Indian Journal of Medical Research, 102, 20, 1995.
   100. Z. A. Mehr, L. C. Rutledge, and J. L. Inase, Evaluation of commercial and experimental
        repellents against Xenopsylla cheopis (Siphonaptera: Pulicidae), Journal of Medical Entomology,
        21, 665, 1984.
   101. L. Goodyer and R. H. Behrens, Short report: The safety and toxicity of insect repellents, American
        Journal of Tropical Medicine and Hygiene, 59, 323, 1998.
   102. U.S. Environmental Protecton Agency, Pesticide Registration Standard for N,N-diethyl-m-toluamide
        (DEET), Washington, DC: Office of Pesticides and Toxic Substances Special Pesticides Review
        Division, United States Environmental Protection Agency, 1980.
   103. E. H. Roland, J. E. Jan, and J. M. Rigg, Toxic encephalopathy in a child after brief exposure to insect
        repellents, Canadian Medical Association Journal, 132, 155, 1985.
   104. T. G. Osimitz and J. V. Murphy, Neurological effects associated with use of the insect repellent
        N,N-diethyl-m-toluamide (DEET), Journal of Toxicology: Clinical Toxicology, 35, 435, 1997.
   105. G. Briassoulis, M. Narlioglou, and T. Hatzis, Toxic encephalopathy associated with use of DEET
        insect repellents: a case analysis of its toxicity in children, Human and Experimental Toxicology, 20, 8,
        2001.
   106. D. L. Sudakin and W. R. Trevathan, DEET: A review and update of safety and risk in the general
        population, Journal of Toxicology: Clinical Toxicology, 41, 831, 2003.
   107. T. G. Osimitz and R. H. Grothaus, The present safety assessment of DEET, Journal of the American
        Mosquito Control Association, 11, 274, 1995.
   108. M. B. Abou-Donia, K. R. Wilmarth, K. F. Jensen, F. W. Oehme, and T. L. Kurt, Increased
        neurotoxicity following concurrent exposure to pyridostigmine bromide, DEET, and chlorpyrifos,
        Fundamental and Applied Toxicology, 34, 201, 1996.
   109. J. C. Veltri, T. G. Osimitz, D. C. Bradford, and B. C. Page, Retrospective analysis of calls to poison
        control centers resulting from exposure to the insect repellent N,N-diethyl-m-toluamide (DEET) from
        1985–1989, Journal of Toxicology: Clinical Toxicology, 32, 1, 1994.
   110. M. S. Fradin, Mosquitoes and mosquito repellents: A clinician’s guide, Annals of Internal Medicine,
        128, 931, 1998.
   111. U.S. Environmental Protection Agency, Reregistration Eligibility Decision (RED): DEET, EPA738-
        R-98-010, in Prevention, Pesticides and Toxic Substances, Washington, DC: United States
        Environmental Protection Agency, 1998.


q 2006 by Taylor & Francis Group, LLC
History of Insect Repellents                                                                               23


   112. R. McGready, K. A. Hamilton, J. A. Simpson, T. Cho, C. Luxemburger, R. Edwards, S. Looareesuwan,
        N. J. White, F. Nosten, and S. W. Linds, Safety of the insect repellent N,N-diethyl-m-toluamide (DEET)
        in pregnancy, American Journal of Tropical Medicine and Hygiene, 65, 285, 2001.
   113. M. Rowland, K. A. Hamilton, J. A. Simpson, T. Cho, C. Luxemburger, R. Edwards,
        S. Looareesuwan, N. J. White, F. Nosten, and S. W. Lindsay, DEET mosquito repellent provides
        personal protection against malaria: A household randomized trial in an Afghan refugee camp in
        Pakistan, Tropical Medicine and International Health, 9, 335, 2004b.
   114. S. S. Rao, S. Prakash, S. Kumar, U. Kaveeshwar, B. K. Bhattacharya, D. K. Jaiswal, and
        K. M. Rao, Toxicologic studies of an insect repellent N,N-diethylphenylacetamide, Indian Journal
        of Medical Research, 85, 626, 1987.
   115. S. S. Rao, R. V. Swamy, and P. K. Ramachandran, Toxicity and metabolism of a new insect repellent
        N,N-diethylphenylacetamide in mice, rats and guinea pigs on cutaneous application, Toxicology, 58,
        81, 1989.
   116. C. Sadanandane, N. Mathew, P. Jambulingam, and K. Kalyanasundaram, Laboratory and field
        evaluation of controlled release formulation of the insect repellents N,N-diethyl-m-toluamide (DEET)
        and N,N-diethyl phenylacetamide (DEPA) against mosquito vectors, Indian Journal of Medical
        Research, 113, 108, 2001.
   117. S. Kumar, S. Prakash, R. K. Sharma, S. K. Jain, M. Kalyanasundaram, R. V. Swamy, and K. M. Rao,
        Field evaluation of three repellents against mosquitoes, blackflies and land leeches, Indian Journal of
        Medical Research, 80, 541, 1984.
   118. M. Debboun and J. Wagman, In vitro repellency of N,N-diethyl-3-methylbenzamide and N,N-diethyl-
        phenylacetamide analogs against Aedes aegypti and Anopheles stephensi (Diptera: Culicidae),
        Journal of Medical Entomology, 41, 430, 2004.
   119. S. S. Rao and K. M. Rao, Insect repellent N,N-diethylphenylacetamide: An update, Journal of
        Medical Entomology, 28, 303, 1991.
   120. DRDO, Multi-insect repellent diethylphenylacetamide spray, Technology Focus, 12, 2004 (http://
        www.drdo.com/pub/techfocus/apr04/welcometfapril04.htm).
   121. E. Worrall, S. Basu, and K. Hanson, The relationship between socio-economic status and malaria: A
        review of the literature, in Ensuring that Malaria Control Interventions Reach the Poor,
        London: The World Bank, p. 46, 2002, http://siteresources.worldbank.org/INTMALARIA/
        Resources/SESMalariaBackgroundPaper.pdf.
   122. M. Klier and F. Kuhlow, Neue Insektenabwehrmittel—am Stickstoff disubstituierte beta-Alaninder-
        ivate, Journal of the Society of Cosmetic Chemists, 27, 141, 1976.
   123. WHO, IR 3535 ethyl butylacetylaminoproprionate information: Interim specification WHO/IS/TC/
        667/2001, Geneva: World Health Organization (http://www.who.int/whopes/quality/en/Icaridin_
        spec_eval_Oct_2004.pdf).
   124. EPA, 3-[N-Butyl-N-acetyl]-aminopropionic acid, Ethyl Ester (113509) Technical Document,
        Washington, DC: Environmental Protection Agency, Biopesticides and Pollution Prevention Division
        1999, (http://www.epa.gov/oppbppd1/biopesticides/ingredients/tech_docs/tech113509.htm).
   125. U. Thavara, A. Tawatsin, J. Chompoosri, W. Suwonkerd, U. R. Chansang, and P. Asavadachanukorn,
        Laboratory and field evaluations of the insect repellent 3535 (ethyl butylacetylaminopropionate) and
        DEET against mosquito vectors in Thailand, Journal of the American Mosquito Control Association, 17,
        190, 2001.
   126. H. H. Yap, Laboratory Repellent Test on IR3535 and DEET at 25% w/v Formulations Against Aedes
        Albopictus and Anopheles Maculatus, Penang, Malaysia: Universiti Sains Malaysia, Vector Control
        Research Unit, 1998a.
   127. H. H. Yap, Field Efficacy Test on Insect Repellent Samples with IR3535 or DEET as Active Ingredients
        Against Aedes Albopictus and Culex quinquefaciatus in the tropical environment, Penang, Malaysia:
        Universiti Sains Malaysia, Vector Control Research, 1998b.
   128. F. Rettich, Laboratory and field evaluation of two new mosquito repellents, in 13th European SOVE
        Meeting, Ankara: European Society of Vector Ecology, 1999.
   129. B. W. Kruger et al., Agents for repelling insects and mites European Patent, 2.3, Bayer AG, 281, 903,
        1998.


q 2006 by Taylor & Francis Group, LLC
24                                                          Insect Repellents: Principles, Methods, and Uses


   130. EPA, New pesticide fact sheet picaridin, Washington, DC: Prevention, Pesticides and Toxic
        Substances, United States Environmental Protection Agency 2005, (http://www.epa.gov/opprd001/
        factsheets/picaridin.pdf).
   131. G. Nentwig, J. Boeckh, F. P. Hoever, B. W. Kruger, K. Roder, and A. G. Bayer, Bayrepel (KBR 3023), a
        new mosquito repellent: from laboratory synthesis to a worldwide commercial product, in 3rd European
        Conference on Travel Medicine, Florence: World Health Organization Collaborating Centre for Travel
        Medicine, 2002.
   132. S. P. Frances and R. D. Cooper, Personal protection measures against mosquitoes: A brief history and
        current use of repellents by the Australian Defence Forces, ADF Health, 3, 58, 2002.
   133. Z. A. Mehr, L. C. Rutledge, E. L. Morales, V. E. Meixsell, and D. W. Korte, Laboratory evaluation of
        controlled-release insect repellent formulations, Journal of the American Mosquito Control Associ-
        ation, 1, 143, 1985.
   134. J. A. Klun, A. Khrimian, A. Margaryan, M. Kramer, and M. Debboun, Synthesis and repellent efficacy of
        a new chiral piperidine analog: Comparison with DEET and Bayrepel activity in human-volunteer
        laboratory assays against Aedes aegypti and Anopheles stephensi, Journal of Medical Entomology, 40,
        293, 2003.
   135. H. H. Yap, K. Jahangir, A. S. Chong, C. R. Adanan, N. L. Chong, Y. A. Malik, and B. Rohaizat, Field
        efficacy of a new repellent, KBR 3023, against Aedes albopictus (Skuse) and Culex quinquefasciatus
        (Say) in a tropical environment, Journal of Vector Ecology, 23, 62, 1998.
   136. A. Badolo, E. Ilboudo-Sanogo, A. P. Ouedraogo, and C. Costantini, Evaluation of the sensitivity of
        Aedes aegypti and Anopheles gambiae complex mosquitoes to two insect repellents: DEET and KBR
        3023, Tropical Medicine and International Health, 9, 330, 2004.
   137. WHO, Report of the 4th WHOPES Working Group meeting 4–5th December 2000: Review of IR3535,
        KBR3023, (RS)-methoprene 20% EC, pyriproxyfen 0.5% GR and lambda-cyhalothrin 2.5% CS. WHO/
        CDS/WHOPES/2001.2, Geneva: World Health Organization, p. 102, 2001.
   138. CDC, Updated Information regarding mosquito repellents: April 22, 2005, http://www.cdc.gov/elcosh/
        docs/d0600/d000695/d000695.pdf, Atlanta: Centers for Disease Control and Prevention, 2005.
   139. T. P. McGovern, C. E. Schreck, and A. Jackson, Mosquito repellents: Alicyclic amides as repellents for
        Aedes aegypti and Anopheles quadrimaculatus, Mosquito News, 38, 346, 1978.
   140. L. L. Robert, R. E. Coleman, D. A. Lapointe, P. J. Martin, R. Kelly, and J. D. Edman, Laboratory and
        field evaluation of five repellents against the black flies Prosimulium mixtum and P. fuscum (Diptera:
        Simuliidae), Journal of Medical Entomology, 29, 267, 1992.
   141. M. Debboun, D. Strickman, T. A. Klein, J. A. Glass, E. Wylie, A. Laughinghouse, R. A. Wirtz, and
        R. K. Gupta, Laboratory evaluation of AI3-37220, AI3-35765, CIC-4, and DEET repellents against
        three species of mosquitoes, Journal of the American Mosquito Control Association, 15, 342, 1999.
   142. J. A. Klun, D. Ma, and R. Gupta, Optically active arthropod repellents for use against disease vectors,
        Journal of Medical Entomology, 37, 182, 2000.
   143. S. P. Frances, R. D. Cooper, and A. W. Sweeney, Laboratory and field evaluation of the repellents
        DEET, CIC-4, and AI3-37220 against Anopheles farauti (Diptera: Culicidae) in Australia, Journal of
        Medical Entomology, 35, 690, 1998.
   144. S. P. Frances, T. A. Klein, D. W. Hildebrandt, R. Burge, C. Noigamol, N. Eikarat, B. Sripongsai, and
        R. A. Wirtz, Laboratory and field evaluation of deet, CIC-4, and AI3-37220 against Anopheles dirus
        (Diptera: Culicidae) in Thailand, Journal of Medical Entomology, 33, 511, 1996.
   145. M. J. Perich, D. Strickman, R. A. Wirtz, S. A. Stockwell, J. I. Glick, R. Burge, G. Hunt, and P. G. Lawyer,
        Field evaluation of four repellents against Leptoconops americanus (Diptera: Ceratopogonidae) biting
        midges, Journal of Medical Entomology, 32, 306, 1995.
   146. H. L. Snodgrass, J. T. Houpt, and P. A. Beall (Eds.), Dermal Penetration of the Candidate Insect
        Repellent AI3-37220 in Swine and Rabbits, Baltimore, MD: U.S. Army Center for Health Promotion and
        Preventative Medicine, Aberdeen Proving Ground, 1991.
   147. H. L. Snodgrass and H. G. Harvey, Toxicology Study No. 86-8104-98, The Acute Toxicity of the
        Diastereomers of the Insect Repellent AI3-37220, Baltimore, MD: U.S. Army Center for Health
        Promotion and Preventative Medicine, Aberdeen Proving Ground, MD, 1998.


q 2006 by Taylor & Francis Group, LLC
History of Insect Repellents                                                                                 25


   148. J. A. Klun, W. F. Schmidt, and M. Debboun, Stereochemical effects in an insect repellent, Journal of
        Medical Entomology, 38, 809, 2001.
   149. J. F. Carroll, J. A. Klun, and M. Debboun, Repellency of DEET and SS220 applied to skin involves
        olfactory sensing by two species of ticks, Medical and Veterinary Entomology, 19, 101, 2005.
   150. J. A. Klun, D. Strickman, E. Rowton, J. Williams, M. Kramer, D. Roberts, and M. Debboun,
        Comparative resistance of Anopheles albimanus and Aedes aegypti to N,N-diethyl-3-methylbenzamide
        (DEET) and 2-methylpiperidinyl-3-cyclohexen-1-carboxamide (AI3-37220) in laboratory human-
        volunteer repellent assays, Journal of Medical Entomology, 41, 418, 2004.
   151. J. Core, R. Bliss, and A. Flores, ARS partners with defense department to protect troops from insect
        vectors, Agricultural Research, 25, 2005, (http://www.afpmb.org/bulletin/vol25/vector0905.pdf).
   152. J. T. Houpt, Toxicology Study No. 85-XC-3929-02, Protocol No. 3929-13b-01-04-01, The Skin
        Sensitization of the Repellent SS-220 in Guinea Pigs, Baltimore, MD: U.S. Army Center for Health
        Promotion and Preventative Medicine, Aberdeen Proving Ground, 2002.
   153. H. L. Snodgrass and J. T. Houpt, Toxicology Study No. 85-XC-3929-02, Protocol No. 3929-23-01-07-
        02, The Acute Dermal Toxicity of the Repellent SS-220 in Guinea Pigs, Baltimore, MD: U.S. Army
        Center for Health Promotion and Preventative Medicine, Aberdeen Proving Ground, 2002.
   154. H. L. Snodgrass, Toxicology Study No. 85-XC-3929-02, Protocol No. 3929-10-99-06-01, The Primary
        Eye Irritation of the Repellent SS-220 in Rabbits, Baltimore, MD: U.S. Army Center for Health
        Promotion and Preventative Medicine, Aberdeen Proving Ground, 2002.
   155. H. L. Snodgrass, Toxicology Study No. 85-XC-3929-02, Protocol No. 3929-11-01-07-04, The Primary
        Skin Irritation of the Repellent SS-220 in Rabbits, Baltimore, MD: U.S. Army Center for Health
        Promotion and Preventative Medicine, Aberdeen Proving Ground, 2002.
   156. H. L. Snodgrass, Toxicology Study No. 85-XC-3929-02, Protocol No. 3929-18-01-07-01, The Acute
        Oral Toxicity of the Repellent SS-220 in Rats, Baltimore, MD: U.S. Army Center for Health Promotion
        and Preventative Medicine, Aberdeen Proving Ground, 2002.
   157. Leahy, S., Building a better bug beater, in Wired News, http://www.wired.com/news/medtech/
        0,1286,59363,00.html, 2003.
   158. L. Garson and M. E. Winnike, Relationship between insect repellency and physical parameters—
        a review, Journal of Medical Entomology, 5, 339, 1968.
   159. M. L. Gabel, I. S. Spencer, and W. A. Akers, Evaporation rates and protection times of repellents,
        Mosquito News, 36, 141, 1976.
   160. A. A. Khan, H. I. Maibach, and D. L. Skidmore, A study of insect repellents. 2. Effect of temperature on
        protection time, Journal of Economic Entomology, 66, 437, 1973.
   161. P. V. Wood, The effect of ambient humidity on the repellency of ethylhexanediol to Aedes aegypti,
        Canadian Entomologist, 100, 1331, 1968.
   162. A. Kroeger, A. Gerhardus, G. Kruger, M. Mancheno, and K. Pesse, The contribution of repellent soap to
        malaria control, American Journal of Tropical Medicine and Hygiene, 56, 580, 1997.
   163. L. M. Rueda, L. C. Rutledge, and R. K. Gupta, Effect of skin abrasions on the efficacy of the repellent
        DEET against Aedes aegypti, Journal of the American Mosquito Control Association, 14, 178, 1998.
   164. T. S. Spencer, R. K. Shimmin, S. Bayles, and W. A. Akers, Consideration of repellent screening
        standards report no. 20, U.S. Research and Development Command, San Francisco: Letterman Army
        Institute, p. 23, 1974.
   165. Y. Trongtokit, Y. Rongsriyam, N. Komalamisra, and C. Apiwathnasorn, Comparative repellency of 38
        essential oils against mosquito bites, Phytotherapy Research, 19, 303, 2005.
   166. Terpine Consortium, Test Plan for Terpenoid Primary Alcohols and Related Esters, Washington, DC:
        The Flavor and Fragrance High Production Volume Consortia, http://www.epa.gov/chemrtk/terprial/
        c12965tp.pdf, 1999.
   167. ATSDR, Chemical technical summary for public health and public safety professionals, Atlanta, GA:
        Agency for Toxic Substances and Disease Registry, http://www.atsdr.cdc.gov/consultations/DEET/
        index.html, 2005.
   168. R. J. Feldmann and H. I. Maibach, Absorption of some organic compounds through the skin in man,
        Journal of Investigative Dermatology, 54, 399, 1970.
   169. C. G. MacNay, Studies repellents for biting flies, Canadian Entomologist, 71, 38, 1939.


q 2006 by Taylor & Francis Group, LLC
26                                                          Insect Repellents: Principles, Methods, and Uses


   170. A. J. S. Weaving and N. K. Sylvester, Pyrethrum as an insect repellent. Part II: A laboratory technique for
        its evaluation as a mosquito repellent, and the influence of formulation on persistence introduction,
        Pyrethrum Post, 9, 31, 1967.
   171. L. C. Rutledge, R. K. Gupta, Z. A. Mehr, M. D. Buescher, and W. G. Reifenrath, Evaluation of
        controlled-release mosquito repellent formulations, Journal of the American Mosquito Control
        Association, 12, 39, 1996.
   172. A. A. Khan, H. I. Maibach, and D. L. Skidmore, Addition of perfume fixatives to mosquito repellents to
        increase protection time, Mosquito News, 35, 23, 1975.
   173. A. A. Khan, H. I. Maibach, and D. L. Skidmore, Insect repellents: Effect of mosquito and repellent-
        related factors on protection time, Journal of Economic Entomology, 68, 43, 1975.
   174. A. P. Kurtz, J. A. Logan, and W. A. Akers, More Effective Topical Repellents Against Malaria-bearing
        Mosquitoes: Review of Volunteer Tests of Mosquito Repellent Formulations October 1969–September
        1971, San Francisco, CA: Letterman Army Institute of Research, 1973.
   175. AFPM Board (Ed.), Personal Protective Measures Against Insects and Other Arthropods of Military
        Significance Technical Guide 36, Washington, DC: Defense Pest Management Information Analysis
        Center Forest Glen Section, Walter Reed Army Medical Center, 2002.
   176. R. K. Gupta, A. W. Sweeney, L. C. Rutledge, R. D. Cooper, S. P. Frances, and D. R. Westrom,
        Effectiveness of controlled-release personal-use arthropod repellents and permethrin-impregnated
        clothing in the field, Journal of the American Mosquito Control Association, 3, 556, 1987.
   177. B. Annis, Comparison of the effectiveness of two formulations of DEET against Anopheles flavirostris,
        Journal of the American Mosquito Control Association, 6, 430, 1990.
   178. S. P. Frances, R. D. Cooper, S. Popat, and N. W. Beebe, Field evaluation of repellents containing DEET
        and AI3-37220 against Anopheles koliensis in Papua New Guinea, Journal of the American Mosquito
        Control Association, 17, 42, 2001.
   179. A. M. Croft, D. Baker, and M. J. Von Bertele, An evidence based vector control strategy for military
        deployments: The British Army experience, Medicine in the Tropics, 91, 2001.
   180. H. Qui, H. W. Jun, M. Dzimianski, and J. McCall, Reduced transdermal absorption of N,N-diethyl-m-
        toluamide from a new topical insect repellent fomulation, Pharmaceutical Development and Tech-
        nology, 2, 33, 1997.
   181. A. A. Khan, H. I. Maibach, and D. L. Skidmore, Increased abrasion and wash resistance of repellents
        with the addition of polymers, Mosquito News, 37, 123, 1977.
   182. J. M. Gambel, J. F. Brundage, R. A. Kuschner, and P. W. Kelley, Deployed US Army soldiers’
        knowledge and use of personal protection measures to prevent arthropod-related casualties, Journal of
        Travel Medicine, 5, 217, 1998.
   183. S. P. Frances, A. M. Auliff, M. D. Edstein, and R. D. Cooper, Survey of personal protection measures
        against mosquitoes among Australian defense force personnel deployed to East Timor, Military
        Medicine, 168, 227, 2003.
   184. L. R. Lindsay, G. A. Surgeoner, J. D. Heal, and G. J. Gallivan, Evaluation of the efficacy of 3% citronella
        candles and 5% citronella incense for protection against field populations of Aedes mosquitoes, Journal
        of the American Mosquito Control Association, 12, 293, 1996.
   185. J. A. Nolen, R. H. Bedoukan, R. E. Maloney, and D. L. Kline, Method apparatus and compositions for
        inhibiting the human scent tacking ability of mosquitoes in environmentally defined three dimensional
        spaces, in US Patent 6,362,235, Alexandria, VA: US Patent Office, 2002.
   186. D. L. Kline, U. R. Bernier, K. H. Posey, and D. R. Barnard, Olfactometric evaluation of spatial repellents
        for Aedes aegypti, Journal of Medical Entomology, 40, 463, 2003.
   187. B. Alten, U. R. Bernier, K. H. Posey, and D. R. Barnard, Field evaluation of an area
        repellent system (Thermacell) against Phlebotomus papatasi (Diptera: Psychodidae) and Ochlerotatus
        caspius (Diptera: Culicidae) in Sanliurfa Province, Journal of Medical Entomology, Turkey, 40, 930,
        2003.
   188. C. F. Curtis and N. Hill, Comparison of methods of repelling mosquitoes, Entomologia Expermentalis et
        Applicata, 49, 175, 1988.
   189. S. E. Hewitt, M. Farhan, H. Urhaman, N. Muhammad, M. Kamal, and M. W. Rowland, Self-protection
        from malaria vectors in Pakistan: An evaluation of popular existing methods and appropriate new


q 2006 by Taylor & Francis Group, LLC
History of Insect Repellents                                                                                    27


          techniques in Afghan refugee communities, Annals of Tropical Medicine and Parasitology, 90, 337,
          1996.
   190.   U. R. Bernier, K. D. Furman, D. L. Kline, S. A. Allan, and D. R. Barnard, Comparison of contact and
          spatial repellency of catnip oil and N,N-diethyl-3-methylbenzamide (DEET) against mosquitoes,
          Journal of Medical Entomology, 42, 306, 2005.
   191.   S. J. Moore and A. Lenglet, An overview of plants as insect repellents, in Traditional Medicine,
          Medicinal Plants and Malaria, G. Bodeker (Ed.), London: Taylor and Francis, pp. 343–363, 2004.
   192.   U. R. Bernier, M. M. Booth, and R. A. Yost, Analysis of human skin emanations by gas chromatography/
          mass spectrometry. 2. Identification of volatile compounds that are candidate attractants for the yellow
          fever mosquito (Aedes aegypti), Analytical Chemistry, 72, 747, 2000.
   193.   U. R. Bernier, D. L. Kline, K. H. Posey, M. M. Yost, and D. R. Barnard, Synergistic attraction of Aedes
          aegypti (L.) to binary blends of L-lactic acid and acetone, dichloromethane, or dimethyl disulfide,
          Journal of Medical Entomology, 40, 653, 2003.
   194.   M. F. Bowen, The sensory physiology of host-seeking behavior in mosquitoes, Annual Review of
          Entomology, 36, 139, 1991.
   195.   M. F. Bowen, Sensory aspects of host location in mosquitoes, CIBA Foundation Symposia, 200, 197,
          1996.
   196.   R. de Jong and B. G. J. Knols, Olfactory responses of host-seeking Anopheles gambiae s.s. Giles
          (Diptera: Culicidae), Acta Tropica, 59, 333, 1995.
   197.   A. E. Eiras and P. C. Jepson, Responses of female Aedes aegypti (Diptera: Culicidae) to host odors
          and convection currents using an olfactometer bioassay, Bulletin of Entomological Research, 84, 207,
          1994.
   198.   T. P. Healy, M. J. Copland, A. Cork, A. Przyborowska, and J. M. Halket, Landing responses of
          Anopheles gambiae elicited by oxocarboxylic acids, Medical and Veterinary Entomology, 16, 126,
          2002.
   199.   B. G. Knols, R. de Jong, and W. Takken, Differential attractiveness of isolated humans to mosquitoes in
          Tanzania, Transactions of the Royal Society of Tropical Medicine and Hygiene, 89, 604, 1995.
   200.   B. G. Knols, On human odor, malaria mosquitoes, and limburger cheese, Lancet, 348, 1322, 1996.
   201.   W. Takken, The role of olfaction in host-seeking of mosquitoes: A review, Insect Science and its
          Applications, 12, 287, 1991.
   202.   W. Takken, B. G. J. Knolls, and H. Otten, Interactions between physical and olfactory cues in the host-
          seeking behavior of mosquitoes: The role of relative humidity, Annals of Tropical Medicine and
          Parasitology, 91, S119, 1997.
   203.   C. Costantini, M. A. Birkett, G. Gibson, J. Ziesmann, N. F. Sagnon, H. A. Mohammed, M. Coluzzi, and
          J. A. Pickett, Electroantennogram and behavioral responses of the malaria vector Anopheles gambiae to
          human-specific sweat components, Medical and Veterinary Entomology, 15, 259, 2001.
   204.   Y. T. Qiu, R. C. Smallegange, S. Hoppe, J. J. A. Van Loon, E. Bakker, and W. Takken, Behavioral and
          electrophysiological responses of the malaria mosquito Anopheles gambiae Giles sensu stricto (Diptera:
          Culicidae) to human skin emanations, Medical and Veterinary Entomology, 18, 429, 2004.
   205.   E. E. Davis, Insect repellents: Concepts of their mode of action relative to potential sensory mechanisms
          in mosquitoes (Diptera: Culicidae), Journal of Medical Entomology, 22, 237, 1985.
   206.   R. C. Smallegange, Y. T. Qiu, J. J. van Loon, and W. Takken, Synergism between ammonia, lactic acid
          and carboxylic acids as kairomones in the host-seeking behavior of the malaria mosquito Anopheles
          gambiae sensu stricto (Diptera: Culicidae), Chemical Senses, 30, 145, 2005.
   207.   Y. T. Qiu, R. C. Smallegange, S. Hoppe, J. J. van Loon, E. J. Bakker, and W. Takken, Behavioral and
          electrophysiological responses of the malaria mosquito Anopheles gambiae Giles sensu stricto (Diptera:
          Culicidae) to human skin emanations, Medical and Veterinary Entomology, 18, 429, 2004.
   208.   W. R. Mukabana, W. Takken, G. F. Killeen, and B. G. Knols, Allomonal effect of breath contributes to
          differential attractiveness of humans to the African malaria vector Anopheles gambiae, Malaria
          Journal, 3, 1, 2004.
   209.   W. R. Mukabana, W. Takken, R. Coe, and B. G. Knols, Host-specific cues cause differential
          attractiveness of Kenyan men to the African malaria vector Anopheles gambiae, Malaria Journal, 1,
          17, 2002.


q 2006 by Taylor & Francis Group, LLC
28                                                         Insect Repellents: Principles, Methods, and Uses


   210. U. R. Bernier, D. L. Kline, C. E. Schreck, R. A. Yost, and D. R. Barnard, Chemical analysis of human
         skin emanations: Comparison of volatiles from humans that differ in attraction of Aedes aegypti
         (Diptera: Culicidae), Journal of the American Mosquito Control Association, 18, 186, 2002.
   211. T. Dekker, B. Steib, R. T. Carde, and M. Geier, L-Lactic acid: A human-signifying host cue for the
         anthropophilic mosquito Anopheles gambiae, Medical and Veterinary Entomology, 16, 91, 2002.
   212. T. Dekker, W. Takken, and M. A. Braks, Innate preference for host-odor blends modulates degree of
         anthropophagy of Anopheles gambiae sensu lato (Diptera: Culicidae), Journal of Medical Entomology,
         38, 868, 2001.
   213. H. V. Pates, W. Takken, K. Stuke, and C. F. Curtis, Differential behavior of Anopheles gambiae sensu
         stricto (Diptera: Culicidae) to human and cow odors in the laboratory, Bulletin of Entomological
         Research, 91, 289, 2001.
   214. E. Barc, I. N. Dogan, and P. A. Rossignol, An olfactometer for discriminating between attraction,
         inhibition, and repellency in mosquitoes (Diptera: Culicidae), Journal of Medical Entomology, 36, 788,
         1999.
   215. A. Cork and K. C. Park, Identification of electrophysiologically active compounds for the malaria
         mosquito Anopheles gambiae, in human sweat extracts, Medical and Veterinary Entomology, 10, 269,
         1996.
   216. O. J. Bosch, M. Geier, and J. Boeckh, Contribution offatty acids to olfactory host finding offemale Aedes
         aegypti, Chemical Senses, 25, 323, 2000.
   217. L. C. Rutledge, M. A. Moussa, and C. J. Belletti, An in vitro blood-feeding system for quantitative
         testing of mosquito repellents, Mosquito News, 36, 283, 1976.
   218. L. C. Rutledge, R. A. Wirtz, M. D. Buescher, and Z. A. Mehr, Mathematical models of the effectiveness
         and persistence of mosquito repellents, Journal of the American Mosquito Control Association, 1, 56,
         1985.
   219. FIFRA, Insect repellent product performance testing guideline evaluation SAP report no. 02-02B, in
         Federal insecticide, fungicide and rodenticide act scientific advisory panel meeting, M. A. Thrall (Ed.),
         Arlington, VA: Environmental Protection Agency, p. 46, 2002, http://www.epa.gov/scipoly/sap/2000/
         april/freportapril572000.pdf.
   220 A. Cockcroft, J. B. Cosgrove, and R. J. Wood, Comparative repellency of commercial formulations of
        DEET, permethrin and citronellal against the mosquito Aedes aegypti, using a collagen membrane
        technique compared with human arm tests, Medical and Veterinary Entomology, 12, 289, 1998.
   221. L. C. Rutledge, M. A. Moussa, C. A. Lowe, and R. K. Sofield, Comparative sensitivity of mosquito
         species and strains to the repellent diethyl toluamide, Journal of Medical Entomology, 14, 536, 1978.
   222. S. J. Moore, C. R. Davies, and M. M. Cameron, Research in Progress: Quantifying diversion of malaria-
         vector mosquitoes from repellent treated ‘bait’ to unprotected ‘collectors’ in field tests in the Bolivian
         Amazon, Transactions of the Royal Society of Tropical Medicine and Hygiene, 97(6), 629, 2003.
   223. S. H. Nguyen, T. V. Nguyen, D. Phan Chau, and R. P. Marchand, Repellent cream with DEET: An
         effective and cheap method to reduce bites of An. dirus species A in the forest, First ASEAN Congress of
         Parasitology and Tropical Medicine, Kuala Lumpur: Malaysian Society of Parasitology and Tropical
         Medicine, 2004.
   224. M. T. Gillies and T. J. Wilkes, Range of attraction of single baits for some West-African mosquitoes,
         Bulletin of Entomological Research, 60, 225, 1970.
   225. J. A. Hill, P. B. Robinson, D. L. Mcvey, W. A. Akers, and W. G. Reifenrath, Evaluation of mosquito
         repellents on the hairless dog, Mosquito News, 39, 307, 1979.
   226. S. Kasman, L. A. O. Roadhouse, and G. F. Wright, Studies in testing insect repellents, Mosquito News,
         13, 116, 1953.
   227. K. Watanabe, Y. Takada, N. Matsuo, and H. Nishimura, Rotundial, a new natural mosquito repellent
         from the leaves of Vitex rotundifolia, Bioscience, Biotechnology and Biochemistry, 59, 1995.
   228. N. Nicolaides, H. C. Fu, and G. R. Rice, The skin surface lipids of man compared with those of eighteen
         species of animals, Journal of Investigative Dermatology, 51, 83, 1968.
   229. Z. X. Li, J. A. Pickett, L. M. Field, and J. J. Zhou, Identification and expression of odorant-binding
         proteins of the malaria-carrying mosquitoes Anopheles gambiae and Anopheles arabiensis, Archives of
         Insect Biochemistry and Physiology, 58, 175, 2005.


q 2006 by Taylor & Francis Group, LLC
History of Insect Repellents                                                                                29


   230. L. C. Rutledge, R. K. Gupta, R. A. Wirtz, and M. D. Buescher, Evaluation of the laboratory mouse
        model for screening topical mosquito repellents, Journal of the American Mosquito Control
        Association, 10, 565, 1994.
   231. M. G. Novak, W. G. Berry, and W. A. Rowley, Comparison of four membranes for artificially
        bloodfeeding mosquitoes, Journal of the American Mosquito Control Association, 7, 327, 1991.
   232. L. C. Rutledge, N. M. Echano, and R. K. Gupta, Responses of male and female mosquitoes to repellents
        in the World Health Organization insecticide irritability test system, Journal of the American Mosquito
        Control Association, 15, 60, 1999.
   233. D. R. Barnard, Global Collaboration for Development of Pesticides for Public Health: Repellents and
        Toxicants for Personal Protection, Geneva: World Health Organization, 2000.
   234. World Health Organization, Report of the WHO informal consultation on the evaluation and testing of
        insecticides, Geneva: World Health Organization, 1996.
   235. U.S. Environmental Protection Agency, Product performance test guidelines OPPTS 810.3700 insect
        repellents for human skin and outdoor premises “Public Draft”, Washington, DC: United States
        Environmental Protection Agency Prevention, Pesticides and Toxic Substances, http://www.epa.gov/
        opptsfrs/publications/OPPTS_Harmonized/810_Product_Performance_Test_Guidelines/Drafts/
        810, 1999.
   236. R. K. Gupta and L. C. Rutledge, Laboratory evaluation of controlled-release repellent formulations on
        human volunteers under three climatic regimens, Journal of the American Mosquito Control
        Association, 5, 52, 1989.
   237. R. D. Xue, D. R. Barnard, and C. E. Schreck, Influence of body size and age of Aedes albopictus on
        human host attack rates and the repellency of DEET, Journal of the American Mosquito Control
        Association, 11, 50, 1995.
   238. R. D. Xue and D. R. Barnard, Human host avidity in Aedes albopictus: Influence of mosquito body size,
        age, parity, and time of day, Journal of the American Mosquito Control Association, 12, 53, 1996.
   239. D. R. Barnard, K. H. Posey, D. Smith, and C. E. Schreck, Mosquito density, biting rate and cage size
        effects on repellent tests, Medical and Veterinary Entomology, 12, 39, 1998.
   240. D. R. Barnard, Mediation of DEET repellency in mosquitoes (Diptera, Culicidae) by species, age and
        parity, Journal of Medical Entomology, 35, 340, 1998.
   241. R. D. Xue and D. R. Barnard, Effects of partial blood engorgement and pretest carbohydrate availability
        on the repellency of DEET to Aedes albopictus, Journal of Vector Ecology, 24, 111, 1999.
   242. J. A. Klun and M. Debboun, A new module for quantitative evaluation of repellent efficacy using human
        subjects, Journal of Medical Entomology, 37, 177, 2000.
   243. R. Natarajan, S. C. Basak, A. T. Balaban, J. A. Klun, and W. F. Schmidt, Chirality index, molecular
        overlay and biological activity of diastereoisomeric mosquito repellents, Pest Management Science, 61,
        1193, 2005.




q 2006 by Taylor & Francis Group, LLC
q 2006 by Taylor & Francis Group, LLC
2
Terminology of Insect Repellents


Graham B. White



CONTENTS
Basic Repellent Terminology .........................................................................................................31
Glossary and Definitions for Repellent Science.............................................................................32
References .......................................................................................................................................43



        Smell is fatal for repellents intended to be used in jungle warfare but, provided it is pleasant, it may
        even be an advantage in civilian use. Owing to the importance attached to long duration of
        effectiveness for military purposes, research on repellents during the war has tended to develop a
        type of repellent with very high boiling-point and hence, almost as a corollary, less effective at
        a distance than some more volatile repellents.
                                                                                        (Christophers, 1947)1




Basic Repellent Terminology2–4
The English word repellent is a noun (the repellent material) or an adjective (repellent effect), derived
from the Latin verb repellere, meaning “to drive back,” the movement away being repulsion. The
alternative spelling repellant, with an “a,” comes from –antem meaning “an agent of action.”
Attractant has the opposite meaning, based on the Latin attractum for being pulled towards
something. The word attractant is a noun (something that attracts) or an adjective (being attractive),
depending on the context and syntax, etymologically derived from the Latin verb trahere, meaning
“to draw or pull.” Therefore, anything that attracts or repels particular insects is either an insect
attractant or an insect repellent. Generally, for chemicals affecting feeding behavior negatively or
positively, by any mode of action, this book introduces the new term phagomone, as discussed in the
Introduction and the Epilogue. Some materials and physical factors (e.g., heat and light) can elicit
either repellent or attractant effects, depending on quantitative factors (Chapter 9) and circumstances.
   To help foster scientific perceptions, Dethier5–7 defined repellents as “any stimulus which elicits an
avoiding reaction” and made a further distinction, in terms of the physical state of the chemical, by
recognizing contact repellents and vapor repellents, i.e., those that have to be touched by the insect or
simply detected in the air. Differentiating these modes of exposure remains challenging, as discussed in
Part 2 of this book, because the treatment distinction may not be absolute. Generally, to achieve personal
protection with some duration of effectiveness, repellents are applied ad libitum to chosen parts of the
skin and clothes; due to this topical treatment (derived from the Greek word topos, meaning “limited


                                                                                                                                                31

q 2006 by Taylor & Francis Group, LLC
32                                                     Insect Repellents: Principles, Methods, and Uses


location”) they are commonly known as topical repellents. Some devices, e.g., mosquito coils and
repellent vaporizers, are designed to protect an area outdoors or volume of space indoors by releasing
spatial repellent8,9 vapor for as long as the device operates (Chapter 23), but their effectiveness quickly
fades when emission stops and the repellent dissipates.
   Commercially, insect repellents are consumer products marketed in every society through suitable
retailers (e.g., camping and travel shops, pharmacies, supermarkets) and by mail order. The traditional
repellent business became more scientifically rigorous when synthetic chemicals began to replace
botanicals as the products-of-choice during the 1940s and 1950s. Previously, so-called “culicifuges”
and repellents to ward off noxious arthropods comprised a wide variety of popular natural products
(Chapter 14 and Chapter 15), few of which had been evaluated entomologically or standardized for
efficacy. The repellent market grew and evolved rapidly following the 1939–1945 World War II
period, thanks to results of intense research efforts to discover and develop repellents for military use,
as described in Chapter 1. Hence the technical foundations of repellent science were mainly
established by three loosely coordinated groups: working in Rutgers,10,11 New Jersey, USA,
Cambridge,12,13 UK, and Orlando,14–17 Florida, USA, continuing to this day at Gainesville,18
Florida, USA. They developed standardized testing methods with mosquitoes (Aedes aegypti) and
ticks (Amblyomma americanum) that still provide the basis of screening procedures and comparative
assessment of repellents (Chapters 5–9).
   The following glossary attempts to explain the meanings of a wide range of terms needed to
understand repellent science and associated research. This list and supporting references augment the
greater attention given to the major topics in successive chapters of this book. Included here are the
acronyms for relevant organizations and regulatory statutes. The Index provides further reference to key
words and Appendix 2 provides details on the appropriate chemical designations for many of the
active ingredients.




Glossary and Definitions for Repellent Science
abiotic factors Pertaining to repellents: non-biological variables that may influence repellency,
         e.g., air quality, humidity, light, temperature, wind (discussed in Chapters 5, 8, and 12);
         c.f. biotic factors.
absorb; absorption The process by which repellent enters a substrate, e.g., skin (c.f. adsorb).
acidity pH ! 7.
activator Something (e.g., heat, synergist, volatile solvent) that, when added to or combined with a
         repellent, increase its availability or activity (c.f. synergist).
active ingredient (a.i.); active material See below, under ingredient.
adjuvant Inert chemical added to repellent formulation to enhance its effectiveness.
adsorb; adsorption The process by which repellent is bound to the surface of a particle or
         absorbent substance.
aerosol Extremely fine spray droplets suspended in air. The WHO19 classifies spray droplets as fine
         aerosols ! 25 mm, coarse aerosols 25–50 mm, mists 50–100 mm, fine sprays 100–200 mm,
         medium sprays 200–300 mm and coarse sprays O 300 mm.
aggregate To gather together, assemble.
alkalinity pH O 7.
allelochemicals Non-nutritional semiochemicals used by one species to affect (behavior, feeding,
         growth, health, breeding of) another species.
allomone Chemical substance (produced or acquired by an organism) that, when contacting an
         individual of another species, evokes in the receiver a behavioral or physiological reaction
         adaptively favorable to the emitter (opposite of kairomone).
antagonism; antagonist Reduction of the potency of a repellent; that which causes antagonism.


q 2006 by Taylor & Francis Group, LLC
Terminology of Insect Repellents                                                                               33


anthropophagous; anthropophagy Feeding on humans (c.f. Zoophagy).
anthropophilic; anthropophily Tendency of hematophagous anthropods to prefer human hosts.
AOAC Association of Official Analytical Chemists International (http://www.aoac.org), founded
          1884, oversees the most extensive program for validation of Official Methods of Analysis
          (OMAs), but none specifically for repellents (c.f. CIPAC).
aqueous Dilution in water.
arrestant Chemical that causes insects to aggregate in contact with it, the mechanism of
          aggregation being kine (by movement) or having a kinetic component.7 An arrestant may
          slow the linear progression of the insects by reducing actual speed of locomotion or by
          increasing turning rate (c.f. locomotor stimulus). The –ant form of this word is
          etymologically correct (not arrestent) because arrest is derived, through Old French, from
          the vulgar Latin arrestare.
arthropods Invertebrate Phylum Arthropoda. Creatures with exoskeleton (consisting of chitin) and
          jointed legs. The blood-feeding arthropods are either insects (Class Insecta) or mites/ticks
          (Class Arachnida, Order Acari). Numerous other groups of animals affect humans directly
          through bites or envenomation (e.g., snakes, scorpions, spiders, and wasps).
attractant For insects, something that causes (attraction) insects to make oriented movements
          towards its source7—i.e., the opposite of repellent (Chapter 9). Associated terms: (verb) to
          attract; (nouns) attractance, the quality of attracting; attraction, the act of attracting or the
          state of being attracted; (adjective) attractive, serving to attract. Sex attractant, substance or
          mixture of substances released by an organism to attract members of the opposite sex of the
          same species for mating.
behavioristic avoidance 20 Also known as behavioristic resistance or protective avoidance—
          modified behavior whereby endophilic mosquito populations sometimes adapt to exophily
          in response to pressure of indoor residual spraying with excitorepellent insecticide.
bioassays Standard methods and procedures for replicated comparative testing of effects on
          biological materials.21–23 Chapter 6 and Chapter 9 describe bioassays for attractants
          and repellents.
Biocidal Products Directive of the European Commission Regulatory law for pesticides in all
          countries of the European Union, implemented by national governments and the E.U.
          Environment Directorate (http://europa.eu.int/comm/environment/biocides/index.htm).
          This Directive 98/8/EC of the European Parliament and of the Council on the placing on
          the market of biocidal products was adopted in 1998. According to the directive, member
          states had to transpose the rules before 14 May 2000 into national law. The Biocidal Product
          Directive aims to harmonize the European market for biocidal products and their active
          substances. At the same time, it aims to provide a high level of protection for humans,
          animals, and the environment. The Commission adopted the original proposal for the
          Biocidal Products Directive in 1993, following the model established by Directive 91/414/
          EEC on plant protection products, adopted in 1991.
biotic factors Pertaining to repellents. Biological variables that may influence repellency, such as
          physiological condition of the insect (e.g., level of hunger, activity cycle) or the host (e.g.,
          rates of exhalation and sweating), as discussed in Chapter 5; c.f. abiotic factors.
biting rate The number of bites/person/time period (e.g., 12 bites/hour), as a measure of population
          density in relation to humans, for any given species of biting arthropods, or group of species
          at a particular place and time. For ethical reasons, especially where vector-borne disease
          risks must be considered, it is customary to intercept the attacking insects before they
          actually bite (possibly increasing catch efficiency); the results are therefore reported in terms
          of the “landing rate” rather than the biting rate. The coefficient of protection24 (CP), is given
          by [(AKB)/A]!100, where A is the average number biting the untreated person per hour
          and B is the average number biting the experimentally treated subject during the same
          exposure period and conditions; CP is commonly used to assess the relative effectiveness of


q 2006 by Taylor & Francis Group, LLC
34                                                     Insect Repellents: Principles, Methods, and Uses


         candidate materials compared to deet. Other criteria for repellent testing under field
         conditions are the period of time to first bite, or first confirmed bite, or duration of a
         reduction in biting—the choice of criterion depending inter alia on the local biting rate
         pressure.25 Considerable disagreement exists on the appropriate measurement of repellent
         product efficacy, as discussed throughout this volume.
botanical Pertaining to green plants (Embryophytes): plant sources of repellent natural products.
butyl carbitol acetate Also known as diethelene glycol monobutyl ether acetate. This compound
         was the standard of comparison adopted by Granett11,12 (1940) for screening repellents at
         the Orlando Institute, Florida, precursor of the Insects Affecting Man and Animals Research
         Laboratory at Gainesville, Florida, now the Center for Medical, Agricultural and Veterinary
         Entomology, of the Agricultural Research Service of the U.S. Department of Agriculture.
carrier Inert solid or liquid material used to prepare repellent formulation.
CAS numbers Unique numerical identifiers for chemical compounds, polymers, mixtures and
         biological sequences. Chemical Abstracts Service (CAS), a division of the American
         Chemical Society, assigns these identifiers to every chemical described in the literature.
         They are also called CAS registry numbers (CAS RNs). Substances also receive unique CA
         index names, constructed using rigid nomenclature rules. In an effort to facilitate searching
         for related compounds, the most important functional groups of a substance are named first,
         followed by their modifications (c.f. IUPAC names). http://www.cas.org/EO/regsys.html.
CDC Centers for Disease Control and Prevention, U.S. Department of Health and Human Services.
         CDC policy and guidelines26,27 for repellents are issued by the Division of Vector-
         Borne Infectious Diseases, and implemented by the National Center for Infectious Diseases
         (based at Fort Collins, Colorado, USA), and by the Entomology Branch (based at Atlanta,
         Georgia, USA).
CFR Code of Federal Regulations of the United States of America (USA): http://www.gpoaccess.
         gov/cfr/index.html. Concering pesticides, including repellents, Title 21 deals with FDA,
         including GRAS materials; Title 40 deals with EPA including FIFRA and FQPA
         (Chapter 26).
CIPAC Collaborative International Pesticides Analytical Council (http://www.cipac.org). The reco-
         gnized international, nonprofit, and non-governmental organization, promotes international
         agreement on methods for the analysis of pesticides and physico-chemical test methods for
         formulations. Methods are proposed by manufacturers (Companies) and are tested
         internationally by the inter-laboratory program for evaluation of test methods. After
         validation of analytical results and adoption, the methods are published in CIPAC
         Handbooks.
compatible Ingredients that retain their individual properties when mixed together.
concentrate Chemical formulation containing a high percentage of active ingredient (a.i.).
concentration Proportion of a given ingredient in a formulation or solution, e.g., oz/gal, mg/L.
cosmetic (adj.): Serving to beautify, or (n.): a preparation for beautifying the face, hair, skin, etc.
         Chapter VI of the U.S. Federal Food, Drug, and Cosmetic Act (FD&C Act of 1906, Title 21
         of the U.S. Code, plus amendments, currently administered by the FDA) defines cosmetics
         as articles intended to be applied to the human body for cleansing, beautifying, promoting
         attractiveness, or altering the appearance without affecting the body’s structure or functions.
         Included in this definition are products such as skin creams, lotions, perfumes, lipsticks,
         fingernail polishes, eye and facial make-up preparations, shampoos, permanent waves, hair
         colors, toothpastes, deodorants, and any material intended for use as a component of a
         cosmetic product. Soap products consisting primarily of an alkali salt of fatty acid and
         making no label claim other than cleansing of the human body are not considered cosmetics
         under U.S. law. Likewise, insect repellents are not cosmetic products, although it would be
         possible to include repellent active ingredients in particular cosmetics, as done with some
         “sun screen” anti-UV preparations combined with deet (that enhances absorption, raising


q 2006 by Taylor & Francis Group, LLC
Terminology of Insect Repellents                                                                                           35


         systemic toxicity)28 marketed for giving skin protection against both sunburn and biting
         insects.* The term cosmetic properties of a repellent product is often used to describe the
         properties of the formulation that do not affect performance, but that alter the subjective
         perception of the product (e.g., fragrance, oiliness, color).
CSPA Consumer Specialty Products Association represents the interests of the consumer specialty
         products industry in the U.S., providing households, institutions and industrial customers
         with products for a cleaner and healthier environment (http://www.cspa.org).
CTFA Cosmetic, Toiletry, and Fragrance Association (http://www.ctfa.org), publisher of the
         International Cosmetic Ingredient Dictionary and Handbook,29 giving International Nomen-
         clature Cosmetic Ingredient (INCI) names for cosmetics and personal care products, e.g.,
         EBAAP for IR3535.
culicifuge 30,31 Repellent for use against mosquitoes (Culicidae), the suffix based on the Latin verb
         fugere, meaning “to flee.”
deet N,N-diethyl-3-methylbenzamide (originally known as N,N-diethyl-meta-toluamide), usually
         abbreviated to deet or deet in literature. It is the dominant repellent used worldwide since the
         1960s. Globally it is the leading active ingredient of insect repellent products, being
         effective against all groups of biting arthropods and even leeches. Formulations containing
         from 4% to 100% deet are registered by the EPA for direct skin application to repel insects,
         rather than kill them. Deet is registered for use by consumers, plus a few veterinary uses, but
         is not used on food. Market surveys in the U.S. show that about a third of the population use
         deet-based products, currently available to the public in a variety of liquids, lotions, sprays,
         and impregnated materials (e.g., wipes and wrist bands). After it was discovered by the
         USDA Agricultural Research Service and developed by the U.S. Army in 1946, deet was
         introduced for use by the general public in 1957. More than 230 products containing deet
         (CAS# 134-62-3) are currently registered with EPA by more than 70 companies (http://deet.
         com and http://www.deetonline.org). Further details on deet are given in Chapter 16.
deterrent (n. or adj.) In the repellent context, something that inhibits feeding or oviposition when
         present in a place where insects would, in its absence, feed or oviposit.7 In the biological
         context, something that protects against bodily harm: see Chapter 3 and Berenbaum (1995)
         for deterrent chemicals.32 Associated terms include deter (v.): to discourage or prevent, and
         deterrence (n.): the act of deterring. These terms fit the way that permethrin-impregnated
         materials (e.g., clothes or bednets) deter blood-thirsty female mosquitoes, etc. from biting,
         or even from entering a house33; whereas, other pyrethroid treatments are more insecticidal
         than deterrent or repellent (c.f. excitorepellent).
diluent Material used to reduce concentration of an active ingredient in a formulation, e.g.,
         dilution of concentrate to make the operational concentration.
DMP Dimethyl phthalate (CAS# 131-11-3), an insect repellent with many other uses as a plasticizer
         and in solid rocket propellants. Commercially, DMP was superseded by deet, DEPA, PMD,
         and others (chapters 1 and 22) for repellent markets.
dispersing agent Material that reduces the attraction between particles.
dosage Quantity of active ingredient applied per unit of time (e.g., 10 oz/day) or area (1 cm/m2) or
         volume (e.g., 1 mg/L) or personal application (e.g., 1 mL/arm/day). See Chapters 6, 8, 12 for
         dosage criteria employed for comparative evaluation of repellents, including the effective
         dose (actual concentration) giving 50% or 90% reduction of biting (ED50 and ED90) and the
         minimum effective dose to prevent biting completely, these bioassay parameters are mostly
         employed for comparative studies in the laboratory; see biting rate for field criteria,
         discussed in Chapters 6, 8 and 12.

*
  Table 25.1 includes one such product marketed in the U.S., whereas the Canadian Pest Management Regulatory Agency
ruled (RRD2002-01) against their acceptability for registration, due to incompatible rates of application (i.e., deet should be
applied sparingly, whereas sunscreens should be applied liberally: www.pmra-arla.gc.ca/english/pdf/rrd/rrd2002-01-e.pdf).


q 2006 by Taylor & Francis Group, LLC
36                                                          Insect Repellents: Principles, Methods, and Uses


EBAAP       Ethyl butyl acetyl aminopropionate (INCI name); chemical description 3-(N-n-butyl-N-
          acetyl)-aminopropionic acid ethyl ester; derived synthetically from b-alanine (a natural
          amino acid); commercially known as IR3535w.* Approved by the WHOPES34 and interim
          specifications issued.
ECB European Chemicals Bureau, responsible inter alia for the Biocidal Products Directive (q.v.)
          of the European Commission (Chapter 26 and http://ecb.jrc.it/biocides/).
EDTIAR Extended Duration Topical Insect and Arthropod Repellent (deet-based slow-release
          formulation) introduced in 1990 for U.S. military use; commercially marketed as
          Ultrathone† (http://www.ultrathon.com).
emulsifier A chemical that aids in the suspension of one liquid in another.
endophagic; endophagy Feeding indoors by endophilic mosquitoes etc.
endophilic; endophily Tendency of insects (especially female Anopheles mosquitoes of some
          species) to come into houses for biting nocturnally and resting diurnally (opposite of
          exophily).
entomology The study of insects; commonly assumed to include other arthropods (q.v.).
EPA U.S. Environmental Protection Agency, see USEPA below and Chapter 26.
essential oils Terpenes and other volatiles obtained from plants by steam distillation or pressing,
          they are hydrophobic and mostly aromatic. Many are repellent to insects and some are
          potent insecticides; traditionally they have been employed as pesticides around the world.35
          Encouraged by the EPA 1996 exemption to FIFRA for minimum risk pesticides, many have
          recently been developed and commercialized as pesticides in the USA. Among the most
          effective36 as repellents are white cedar oil (CAS# 8000-34-8), peppermint oil (CAS# 806-
          90-4), red thyme oil (CAS# 8007-46-3), bourbon geranium oil (CAS# 8000-46-2), linalool
          (Appendix 2), and dehydrolinalool.37 However, as indicated in Chapter 14 (pp. 292-293),
          toxicological risk assessment is necessary to establish safety and tolerance levels for
          essential oils used as repellents or in foodstuffs [FDA category (CFR 21: 170) Generally
          Regarded as Safe (GRAS)].
EU The European Union (EU) of twenty-five countries (2006) with 20 official languages, formerly
          known as the European Community (EC), originally the European Economic Community
          (EEC). The Biocidal Products Directive (q.v.) determines pesticide regulatory status
          throughout the EU (http://europa.eu.int/).
evaporate (v.) To change from solid or liquid to vapor (U.S.) or vapour (U.K.), synonymous with
          vaporize (U.S.) or vapourise (U.K.); evaporation (n.): The process of evaporating;
          evaporate to dryness.
excitorepellency The power of DDT and some pyrethroids, especially through tarsal exposure of
          insects, to irritate them sufficiently that they fly away before knockdown, even from
          sublethal exposure;20,38–40 thereby adult female mosquitoes become more exophilic instead
          of endophilic and this contributes to greater reduction of their vectorial capacity than from
          simply killing a lesser proportion of the vector population.41
exophagous; exophagy Behavioral tendency of female mosquitoes etc. to bite hosts outdoors.
exophilic; exophily Tendency of most insects to stay outside buildings (contrasts with endophily
          for malaria vector Anopheles females that enter houses to bite and take shelter).
FDA Food and Drug Administration of the U.S. Department of Health and Human Services, having
          regulatory responsibility for cosmetics and medicines etc., but not for insect repellents
          (http://www.fda.com/).
FIFRA The U.S. Federal Insecticide, Fungicide, and Rodenticide Act (1947, 1972 and amend-
          ments) for pesticides regulation (40 CFR), administered by the EPA.

*
    A registered trademark of Merck KGaA, Darmstadt, Germany.
†
    A registered trademark of 3M Corporation, St. Paul, MN.


q 2006 by Taylor & Francis Group, LLC
Terminology of Insect Repellents                                                                           37


formulation (n.) Defined chemical product mixture, usually meaning the commercialized version
          of a special formula, sometimes requiring dilution before use.
FQPA Food Quality Protection Act (U.S. Public Law 104–170, 1996: http://www.epa.gov/
          oppfead1/fqpa/backgrnd.htm), augmenting FFDCA and FIFRA, administered by the
          USEPA (q.v.): intensifies regulatory controls on pesticides for reasons of human and
          environmental health (Chapter 26 summarizes EPA role under Title 40, parts 150–189, of
          the U.S. code of Federal Regulations).
GFP and GLP Good Field Practice and Good Laboratory Practice, internationally recognized
          standards of conduct and procedure, administered by the Organisation for Economic
          Co-operation and Development (OECD), to ensure the generation of high quality and
          reliable test data related to the safety of industrial chemical substances and preparations in
          the framework of harmonizing testing procedures for the Mutual Acceptance of Data
          (MAD) (http://www.oecd.org/document/63/0,2340,en_2649_34381_2346175_1_1_1_1,00.
          html).
GRAS Generally Regarded as Safe, classification by FDA, www.cfsan.fda.gov/~lrd/cfr17030.html
          !http://www.cfsan.fda.gov/~lrd/cfr17030.htmlO, similar to minimum risk classification
          by EPA (Chapter 26, p. 420).
hazard Potential source of harm. For repellents and other pesticides, the World Health Organiz-
          ation (WHO) classification68 based on the rat LD50 by weight, following oral or dermal
          exposure, assuming solids are four-fold more hazardous than liquids, recognizes the
          following categories: class Ia, extremely hazardous; class Ib, highly hazardous; class II,
          moderately hazardous (e.g., DDT, permethrin, pyrethrins); class III, slightly hazardous
          (e.g., deet); plus active ingredients unlikely to cause acute hazard in normal use.
hematophagous arthropods Blood-feeding insects, ticks and mites. English spelling: haemato-
          phagous. Commonly referred to as “biting insects.”
hydrogen-ion concentration Usually expressed as the negative log (pH), a measure of acidity–
          alkalinity.
immiscible Liquids that cannot mix to form homogeneous solution.
INCI See CTFA.
incompatible Ingredients that do not retain their individual properties when mixed together.
ingredient That which goes into a compound, formulation, preparation, or mixture; active
          ingredient (a.i.), the key ingredient with intended activity.
inhibition As discussed in Chapter 4, activity-inhibitors cause a neutral reaction, neither attraction
          nor repulsion, whereby an insect fails to proceed questing purposefully, but is not
          anaesthetized nor narcotized. Dogan and Rossignol describe an olfactometer42 for
          discriminating between attraction, inhibition, and repellency in mosquitoes.
insect Any member of the arthropod Class Insecta. The name derived from the Latin insectum for
          having been cut, referring to the articulated body; adults typically with three pairs of legs
          (hexapod).
insecticide Chemical agent used to kill insects; mostly suitable for use also as acaricides.
insoluble Inability of a substance to dissolve in a particular liquid solvent.
irritancy The power of DDT and some pyrethroids (especially those with alpha-cyano moiety) to
          irritate arthropods, causing excitorepellency (q.v.).
IR3535 See EBAAP: insect repellent.
IUPAC International Union of Pure and Applied Chemistry (an international non-profit, non-
          governmental organization for the advancement of chemistry, consisiting of national
          chemistry societies (http://www.iupac.org). IUPAC is the recognized authority in develop-
          ing standards for naming the chemical elements and their compounds, through its Inter-
          Divisional Committee on Nomenclature and Symbols (IUPAC nomenclature). c.f.
          CAS, CIPAC.


q 2006 by Taylor & Francis Group, LLC
38                                                     Insect Repellents: Principles, Methods, and Uses


kairomone A substance released by one species that benefits members of another (e.g., parasites
         detect host kairomones), by being a signal or attractant to them (opposite of allomone).
kinesis non-oriented movement of an organism; c.f. taxis (oriented movement); kinetic (adj.).
knockdown Sublethal incapacitation; early symptom of an insect responding to a pesticide; not
         necessarily lethal because metabolic recovery may occur. Hence the rates of knockdown and
         mortality are scored separately, usually 1 hour and 24 hours post-treatment in bioassays.
         Knockdown has another meaning in molecular biology, for gene incapacitation.
locomotor stimulant A chemical that causes, by a kinetic mechanism, insects to disperse from a
         region more rapidly than if the area did not contain the chemical. The effect may be to
         increase the speed of locomotion, to cause the insects to carry out avoiding reactions, or
         to decrease the rate of turning.43
market value Globally and locally, the price of repellent products is determined by market forces,
         whereas the sale cost (over-the-counter price) of each repellent unit (pack) includes the
         values of active ingredients, formulation ingredients, manufacturing and labor, packaging,
         distribution, promotion, sales and profit margins, plus taxes and tariffs. World-wide the
         global market value70 of repellents was estimated at $2 billion in 2002.
mortality rate Proportion of sample killed in a test (usually scored 24 hours after treatment) by
         exposure to a lethal dose causing fatality; those surviving treatment have experienced only a
         sub-lethal dose, that may affect their bionomics and behaviour, e.g., inhibition, deterrence,
         and repulsion.
natural products Exploitable materials formed by nature, including foodstuffs and natural fibres
         used for weaving fabric, e.g. cotton. Natural repellent products from plants (botanically-
         derived) are reviewed in Chapters 14 and 15: those from non-woody plants are herbal-based
         (Chapter 9). Natural pyrethrins comprise important insecticides and repellents.
organic Strictly, chemical compounds derived from plants or animals, plus other carbon-based
         materials. Essential oils from plants (Chapter 14) include many useful organic repellents. In
         the terminology of modern farming and horticulture, so-called “organic” vegetables and
         other agricultural produce are defined as those grown and marketed without application of
         synthetic pesticides.
personal protective measures 44 (PPM) Protective measures against biting arthropods, such as the
         personal use of repellents, bednets and clothing.
pesticides Chemicals for killing pests, classified by EPA as follows: algicides, antifouling agents,
         antimicrobials, biocides, biopesticides, defoliants, desiccants, disinfectants and sanitizers,
         fungicides, fumigants, herbicides, insect growth regulators, insecticides, acaricides
         (including miticides), molluscicides, nematicides, ovicides, pheromones, plant growth
         regulators, rodenticides, and repellents (http://www.epa.gov/pesticides/about/types.htm).
pH Number expressing degrees of acidity (pH!7) and alkalinity (pHO7) in solutions; pH 7 is
         neutral. Mathematically, pH is the log10 of the reciprocal of the hydrogen ion concentration;
         it is usually measured by comparison with a standard solution of potassium hydrogen
         phthalate, with a pH of 4 at 158C.
phagomone Chemical that affects feeding behavior, negatively or positively, by any mode of action
         (see Preface and Epilogue).
pheromone A chemical compound, emitted by an organism, that influences the behavior and
         development of other members of the same species.
phytotoxicity Pathological effect on plant (Greek: phytos) vegetation.
picaridin (KBR 3023) Insect repellent developed and commercialized as Bayrepelw; approved by
         WHOPES34 and interim specifications issued under the proposed ISO name icaridin [sic].
piperamides and piperidine alkaloids A series of compounds and analogs that includes many
         useful repellents, some being also insecticidal, e.g., deet, SS220 and pipernonaline.45–48 The
         amides have a carbonyl (CaO) group linked to a nitrogen, N–(CaO), while
         the nitrogen’s other two bonds are linked with hydrogens (Figure 2.1) or other groups,


q 2006 by Taylor & Francis Group, LLC
Terminology of Insect Repellents                                                                                          39


                                         O                                        O

                                              N       CH3                              N

                                                  CH3
                                   CH3                                    CH3
                            N,N-diethyl-3-methylbenzamide         1-(3-methylbenzoyl)piperidine

FIGURE 2.1 Deet (on the left) has a benzene ring linked by a carbonyl group (CaO) to an amide (piperamide) with two
CH3 methyl groups; the piperidine analog (on the right) has a saturated carbon ring that includes the nitrogen from the amide.


        e.g., N,N-diethyl-3-methylbenzamide (deet). When the nitrogen joins a saturated hetero-
        cyclic ring with five carbons, the compound constitutes a piperidine—the chemical name
        derived from plants of the pepper family (Piperaceae) that contain many such natural
        compounds, sometimes used as repellents (Chapter 14). Natural piperidine (CAS# 110-89-
        4) is the noxious ingredient of poison hemlock (Conium maculatum) in the carrot family
        Apiaceae. Among more than 200 such compounds identified in Piper,49 the relatively simple
        amides provide much of the “hot pungent spice” taste as well as the biological activity in
        many species.44 The piperamides commonly found in the genus Piper are bifunctional; an
        isobutyl amide functionality is combined with a methylenedioxyphenyl (MDP) moiety, as
        seen in piperine of Piper nigrum fruit and 4,5-dihydropiperlonguminine in foliage of the
        Central American Piper tuberculatum. The most active piperamide discovered to date is
        pipercide, approximately 100-fold more active than piperine.50–52 The piperamides are also
        unusual because of their dual biological activities: the amide functionality is neurotoxic and
        the MDP group is an inhibitor of cytochrome P450 enzymes. Scott et al.53 demonstrated that
        combinations of piperamides in binary, tertiary, and quarternary mixtures had successively
        higher toxicity at equimolar concentrations. This combination of useful traits suggests that
        Piper extracts may be good candidate pesticides with a rich range of insecticidal and
        repellent properties.
PMD, p-menthane-3,8-diol Occurs naturally in leaves of the Australian lemon-scented gum tree
        (Corymbia citriodora), commonly called lemon eucalyptus.54 This monoterpene, structu-
        rally similar to menthol (CAS# 42822-86-6), remains as a spent product after distillation of
        essential oils from leaves and twigs of Corymbia citriodora. Whereas natural PMD-based
        repellents have long been popular in China and elsewhere,55 and registered in Europe for
        over a decade, synthetic PMD is used as the active ingredient for some of the repellents
        marketed as “lemon eucalyptus” in the U.S. As described in Chapter 20, PMD exerts
        repellency of the highest order against a wide range of hematophagous arthropods.
        Formulations registered in the USA include liquids that are sprayed on skin or clothing,
        or lotions that are rubbed on skin. Not yet submitted for WHOPES evaluation.
PPM Acronym for Personal Protective Measures44 against biting arthropods, such as the use of
        topical repellents and clothing (c.f. ppm expressing dilution in terms of parts per million).
pyrethrins Oily esters extracted from cultivated pyrethrum flowers, Chrysanthemum cinerariae-
        folium Benth. & Hook., syn. Tanacetum cinerariifolium (Trevir); also found in pyrethrum
        daisies: Chrysanthemum roseum Web. & Mohr., syn. C. coccineum Willd. (Asteraceae).
        Crude pyrethrin extract contains three esters of chrysanthemic acid (chrysanthemates:
        pyrethrin I, cinerin I, jasmolin I) plus three esters of pyrethrin acid (pyrethrates: pyrethrin II,
        cinerin II, jasmolin II), combined ratio 71:21:7, generally known as pyrethrins. Being
        lipophilic but having low aqueous solubility, pyrethrins are readily absorbed via
        arthropod cuticle but not via the skin of vertebrates. Pyrethrins are very potent insecticidal
        knockdown agents, causing excitorepellency at sublethal doses, due to disruption of


q 2006 by Taylor & Francis Group, LLC
40                                                      Insect Repellents: Principles, Methods, and Uses


         sodium channel gating in myelinated nerves. Commercially, 25% to 50% pyrethrin
         concentrates are very stable in darkness at ambient temperatures, but degrade rapidly in
         sunlight (DT50 10K12 mins).
pyrethroids Numerous synthetic organic compounds, mostly based on the chrysanthemate moiety
         of pyrethrum, having analogous neurotoxic modes of action causing rapid knockdown and
         insecticidal effects. Discovery and development of synthetic pyrethroids, during 1960s and
         70s, accomplished several goals: more economical and consistent production than with
         natural pyrethrins; photostable products with residual efficacy but limited bioaccumulation.
         After early progress with allethrins (transient space sprays and vaporizers), the first truly
         stable pyrethroids were fenvalerate and permethrin; their relative safety and potency greatly
         surpassed other classes of insecticides. Wide variations in potency occur between cis and
         trans isomers, and among enantiomers of pyrethroids, allowing much diversity of pyrethroid
         products, providing manufacturers and users with choices between knockdown versus
         insecticidal potency, and degrees of residual stability. As hundreds of pyrethroids became
         commercialized, this class of compounds has dominated the insecticide industry during
         recent decades. Permethrin remains one of the favorites for its versatility as an insecticide
         with repellent and deterrent properties (Chapters 5 and 6). Other pyrethroids mentioned in
         this book include allethrins, alpha-cypermethrin, beta-cyfluthrin deltamthrin, esbiothrin,
         lambda-cyhalothrin, metofluthrin, prallethin, tetramethrin and transfluthrin (Appendix 2).
repellent, repellant For insects, something that causes insects to make oriented movements away
         from its source.7 Associated terms: (verb) to repel; (nouns) repellency (repellancy), the
         quality of repelling; repeller, device for repelling (invalid for electronic56 so-called
         “mosquito repellers”); repulsion, the act of repelling or the state of being repelled;
         (adjective) repulsive, serving to repel. The term repellent has received such general usage
         as a formulated product or as a chemical with a specific behavioral effect that it has lost
         much of its technical meaning. The editors of this volume advocate that the term repellent
         be restricted to the designation of products intended to reduce the rate of biting from
         hematophagous arthropods (French: insectifuges corporels). In this way, the technical
         literature will tend to use more precise terms that describe the effects of chemicals on
         specific behaviors. The introduction of the term phagomone is, in part, an attempt to
         facilitate this transition by providing the technical literature with an alternative to the term
         repellent used generally.
resistance Defined by the WHO (1957)57 as “the development of an ability in a strain of some
         organism to tolerate doses of a toxicant that would prove lethal to a majority of individuals
         in a normal (susceptible) population of the same species,” various types of insecticide
         resistance are well known in many species of flies, mosquitoes and other vectors and pests of
         public health importance.58 For an increasing number of species, diagnostic and discrimi-
         nating dosages have been determined59,60 for distinguishing between susceptible and
         resistant individuals. Selection for resistance against repellents might be expected, due to
         their ubiquitous usage and environmental persistence61,62 of deet. Because no effort is made
         to monitor the sensitivity of wild populations of the many arthropod species that repellents
         are employed against, the possibilities of behavioral or physiological resistance to repellents
         remain unexplored. However, studies with laboratory strains of mosquitoes63,64 and
         Drosophila65 demonstrate genetic selection of insensitivity and tolerance, indicating the
         potential for resistance to deet and other repellents.
risk assessment In context of human health, estimating the probability of adverse effects resulting
         from defined exposure to known chemical hazard68 - see (Chapter 14, pp. 292-293) and
         (Chapter 26, p. 422) for repellents.
Rutgers 612 The original proprietary name for ethyl hexanediol (CAS# 94-96-2) when used as
         a repellent product; withdrawn 1991 for toxicological reasons (Chapter 1).


q 2006 by Taylor & Francis Group, LLC
Terminology of Insect Repellents                                                                          41


Rutgers 6-2-2 A repellent mixture consisting of 6 parts dimethyl phthalate, 2 parts Rutger’s 612,
         and 2 parts indalone), optimized for military4,11 use during World War II as M-250.
semiochemicals Chemicals involved in communication among organisms.66
SI units International System of Units (http://physics.nist.gov/cuu/Units/index.html).
soluble Ability to dissolve in a given solvent, such as acetone, alcohol, water.
solute That which dissolves.
solution Solvent plus solute.
solvent Liquid in which solute dissolves to form solution.
specifications Standard descriptions of products for quality control purposes. For repellents and
         other pesticides, international specifications are prepared by the FAO and/or WHO, then
         adopted by the FAO/WHO Joint Meeting on Pesticide Specifications (JMPS, http://www.
         who.int/whopes/quality/en/), in conjunction with CIPAC analytical methods. Joint FAO/
         WHO specifications are issued by the World Health Organization Pesticides Evaluation
         Scheme (WHOPES), available only in electronic format from http://www.who.int/whopes/
         quality/en, providing a qualitative basis for production and procurement.
spreader A chemical that increases the area that a certain volume of liquid will cover.
sticker Something increasing adherence; formulation ingredient to enhance adherence of the
         active ingredient.
stimulants Substances that cause insects to begin moving, copulating, feeding or laying eggs,6
         hence qualified terms such as locomotor stimulant, mating stimulant and oviposition
         stimulant. The term feeding stimulant is synonymous with phagostimulant.67
substrate For purposes of repellents and other pesticides, the substrate is a treated surface (c.f.
         biochemical substrate—molecule acted upon by an enzyme; bioecological substrate—
         environment in which an organism lives).
surfactant Chemical agent that increases the emulsifying, dispersing, spreading and/or wetting
         properties of another chemical when contacting a surface.
suspension Finely divided solid particles mixed in liquid, in which they not soluble.
synergist A substance that, when combined with another substance, gives effect that is greater than
         the sum of their individual effects.
synomone Mutually beneficial signal chemical, released by members of one species, that affects the
         behavior of another species and benefits individuals of both species.
synthetic Chemical compounds made by human directed process, as opposed to those of natural
         origin; the same material may be produced naturally or synthetically (e.g. PMD,
         Chapter 20). Since the 1940s (Chapter 1), most commercial repellents are synthetic
         compounds. Synthetic pyrethroids (q.v.) are important insecticides and irritant repellents,
         usually including a chrysanthemic moiety homologous to natural pyrethrins (q.v.).
taxis Directional response to stimulus: movement towards the source being positive taxis; move-
         ment away from the source being negative taxis; c.f. kinesis. Chemotaxis (n.), chemotactic
         (adj.): movement responding to chemical (attractant or repellent).
tolerance Having low susceptibility, due to high fitness of the individual or population; usually
         attributable to presence of some robust or resistant individuals from which a more obviously
         resistant population can be selected (due to increased frequency of resistant genotypes when
         successive generations are subjected to Darwinian selection). In many countries, regulatory
         systems set pesticide tolerances as maximum permissible levels of residues in foodstuffs etc.
         (established by EPA in the USA and by the ECB in the EU). Tolerance has special meaning
         for quality control purposes, whereby the permissible range of variation is defined in product
         specifications with respect to the active ingredient, e.g., meanG10%, possibly expressed
         as variance. Using this mathematical concept, Rutledge63,64 assessed repellent tolerances of
         mosquito populations in order to compare ranges of responses and resistance potential. For
         pesticides generally, tolerance is recognized when the LC50 of a population rises upto 5
         times greater than normal for a standard susceptible strain of the same species; higher ratios


q 2006 by Taylor & Francis Group, LLC
42                                                    Insect Repellents: Principles, Methods, and Uses


        (dose-efficacy comparisons between populations of the same species) indicate resistance
        (q.v.).
toxics Based on the Greek word toxikon for arrow-poison, toxicology is the study of poisons
        biologically harmful substances and their effects; dose-dependent criteria allow any material
        to be toxic, serving as a toxicant or toxin for sensitive tissues or organisms, although this
        term is normally applied to hazardous pathogens, pesticides68 and other chemicals; toxicity
        of pesticides is commonly measured (for each species) in terms of lethal concentrations or
        dosages at the 50% level (LC50 or LD50) and the 99% level (LC99 or LD99) for comparative
        purposes when dealing with target insects and non-target species. The Toxics Release
        Inventory (TRI) is a publicly available EPA database (http://www.epa.gov/tri/) that contains
        information on toxic chemical releases and other waste management activities reported
        annually by industry and U.S. federal facilities. For chemical safety purposes, in setting
        tolerances (as above), toxicologists determine the ‘no observed adverse effect level’
        (NOAEL) for laboratory animals. Mammalian toxicity values, required by regulatory
        authorities (such as the USEPA, Chapter 26) for assessing pesticides for regulatory
        approval, are based on effects of short-term (acute), long-term (chronic) and intermediate
        (sub-chronic) periods of exposure, as well as effects on development and reproduction,
        including mutagenicity and carcinogenicity, to establish dose-response relationships. For
        example, acute tests (so-called 6-pack) comprise oral, dermal and inhalation LDs,
        neurotoxicity, eye irritation, dermal irritation and sensitization (www.epa.gov/oppfead1/
        trac/a-toxreq.htm). The human equivalency potency factor (Q) is usually based on the oral
        exposure route, designated Q* when considered carcinogenic (www.epa.gov/pesticides/
        carlist). The so-called Reference Dose (RfD) is the average daily oral exposure that is
        estimated to be unlikely to cause harmful effects during a lifetime. RfDs are generally used
        by the EPA for health effects that are thought to have a low threshold (dose limit) for
        producing effects. The International Programme on Chemical Safety71 (IPCS) emphasizes
        the Acceptable Daily Intake71 (ADI) for each chemical, aggregated from all sources of
        exposure, whereas the USEPA increasingly considers cumulative risk (www.epa.gov/
        oppsrrd1/cumulative/) from exposure to groups of pesticides with an equivalent mode of
        action (e.g. organophosphates). Whereas the mode of action of insect repellents is not well
        understood (Chapter 11), the toxicology of repellent compounds is not difficult to assess by
        standard methods.
U.K. United Kingdon of Great Britain and Northern Ireland, one of 25 Member States of the
        European Union, therefore subject to the Biocidal Products Directive (q.v.) for regulation
        of pesticides.
USDA United States Department of Agriculture has a variety of Agencies, Offices and Services,
        notably the Agricultural Research Service (ARS) with long-term research on insect
        attractants and repellents.
U.S. EPA, OPP United States Environmental Protection Agency, Office of Pesticide Programs
        (http://www.epa.gov/pesticides/), comprises several operating divisions, currently named:
        Antimicrobials, Biological and Economic Analysis, Biopesticides and Pollution Prevention,
        Environmental Fate and Effects, Field and External Affairs, Health Effects Division,
        Information Technology and Resources Management, Registration, Special Review and
        Reregistration. Collectively they are responsible for pesticide regulatory management in
        the USA.
vapor pressure The property causing a chemical to evaporate, defined as the pressure of the vapor
        in equilibrium with the liquid or solid state; measured in joules, SI units of energy
        (International System of Units, http://physics.nist.gov/cuu/Units/index.html).
vector Carrier of infection. Vector-borne pathogens cause disease; e.g., Plasmodium causes
        malaria, transmitted by vector Anopheles mosquito.


q 2006 by Taylor & Francis Group, LLC
Terminology of Insect Repellents                                                                           43


viscosity   The property of liquids to resist flow, due to forces acting between the molecules. The SI
          physical unit of dynamic viscosity (greek symbol: m) is the pascal-second (Pa s), identical
          to 1 Ns/m2 or 1 kg/(ms).
volatility Rate of evaporation of liquid or solid.
wetting agent A chemical that increases the liquid contact of dry material.
WHOPES World Health Organization Pesticides Evaluation Scheme, responsible for assessments,
          specifications and recommendations for pesticides (including repellents) used for public
          health pest and vector control,69 on behalf of Member States of the United Nations (U.N.).
          (http://www.who.int/whopes/en/).
zoophagy; zoophily Tendency of hematophagous insects to bite or prefer hosts other than humans
          (c.f. anthropophagy, anthropophily).



References
     1. S. R. Christophers, Mosquito repellents, being a report of the work of the mosquito repellent inquiry,
        Cambridge 1943–5, J. Hyg., 45, 176, 1947.
     2. Webster’s Encyclopaedic Unabridged Dictionary of the English Language, New York: RHR
        Publishing, 2001.
     3. Chambers Twentieth Century Dictionary, Edinburgh: W & R Chambers, 1972.
     4. G. Gordh and D. Headrick, A Dictionary of Entomology, Wallingford: CABI Publishing, 2001.
     5. V. G. Dethier, Chemical Insect Attractants and Repellents, Philadelphia: Blakiston, 1947.
     6. V. G. Dethier, Man’s plague? Insects and Agriculture, Princeton, NJ: Darwin Press, 1976.
     7. V. G. Dethier et al., The designation of chemicals in terms of the responses they elicit from insects,
        J. Econ. Entomol., 53, 134, 1960.
     8. H. K. Gouck, T. P. McGovern, and M. Beroza, Chemicals tested as space repellents against yellow-
        fever mosquitoes. I. Esters, J. Econ. Entomol., 60, 1587, 1967.
     9. C. E. Schreck, Spatial action of mosquito repellents, J. Econ. Entomol., 63, 1576, 1970.
    10. P. Granett, Studies of mosquito repellents, I. Test procedure and methods of evaluating test data,
        J. Econ. Entomol., 33, 563, 1940.
    11. P. Granett, Studies of mosquito repellents, II. Relative performance of certain chemicals and
        commercially available mixtures as mosquito repellents, J. Econ. Entomol., 33, 566, 1940.
    12. S. R. Christophers, Insect repellents, Brit. Med. Bull., 3, 222, 1945.
    13. S. R. Christophers, Mosquito repellents, Rev. Med. Mex., 26, 213, 1946.
    14. W. V. King, Repellents and insecticides for use against insects of medical importance, J. Econ.
        Entomol., 44, 338, 1951.
    15. W. V. King, Chemicals Evaluated as Insecticides and Repellents at Orlando, Florida, Bureau of
        Entomology and Plant Quarantine, Agriculture Research Service, United States Department
        of Agriculture. Agriculture Handbook 69, Washington, DC: U.S. Government Printing Office, 1954.
    16. USDA, Results of Screening Tests with Materials Evaluated as Insecticides, Miticides, and Repellents
        at the Orlando, Florida, Laboratory, April, 1942–April, 1947, Bureau of Entomology and Plant
        Quarantine, Agriculture Research Service, United States Department of Agriculture. Publication
        E-733, Washington, DC: U.S. Government Printing Office, 1947.
    17. USDA, Materials Evaluated as Insecticides, Repellents, and Chemosterilants at Orlando and
        Gainesville, FL, 1952–1964, Entomology Research Division, Agricultural Research Service, United
        States Department of Agriculture. Agriculture Handbook No. 340, Washington, DC: U.S. Government
        Printing Office, 1967.
    18. C. E. Schreck, K. Posey, and D. Smith, Repellent Activity of Compounds Submitted by Walter Reed
        Army Institute of Research, Part I. Protection Time and Minimum Effective Dosage against Aedes
        aegypti Mosquitoes, Agricultural Research Service, Technical Bulletin No. 1549, Washington, DC:
        United States Department of Agriculture in cooperation with Walter Reed Army Institute of Research,
        1977.


q 2006 by Taylor & Francis Group, LLC
44                                                         Insect Repellents: Principles, Methods, and Uses


     19. WHO, Equipment for Vector Control, 3rd ed., Geneva: World Health Organization, 1990.
     20. R.C. Muirhead-Thomson, The significance of irritability, behaviouristic avoidance and allied
         phenomena in malaria eradication. Bull. Wld. Hlth. Org. 22, 721, 1960.
     21. J. S. Kennedy, Behaviorally discriminating assays of attractants and repellents, in Chemical Control of
         Insect Behavior: Theory and Application, H. H. Shorey and J. J. McKelvey (Eds.), Chichester: Wiley,
         1977, chap. 13.
     22. J. L. Robertson and H. K. Preisler, Pesticide Bioassays with Arthropods, Boca Raton, FL: CRC Press,
         1992.
     23. WHO, Report of the WHO Informal Consultation on the Evaluation and Testing of Insecticides, Control
         of Tropical Diseases, Pesticide Evaluation Scheme, Informal Consultation, unpublished document 96.1,
         Geneva: World Health Organization, 1996.
     24. C. E. Schreck et al., Evaluation of personal protection methods against phlebotomine sand flies
         including vectors of leishmaniasis in Panama, Am. J. Trop. Med. Hyg., 31, 1046, 1982.
     25. EPA, Product Performance Test Guidelines, OPPTS 810.3700, Insect Repellents For Human Skin
         and Outdoor Premises, document EPA 712-C-99-369, December 1999, public draft issued 8th June
         2006, following Scientific Advisory Panel (SAP) Meeting, April 7, 2000: Evaluation of Product
         Performance Testing Guidelines for Insect Repellents, SAP unpublished Report No. 00-02B, pp. 44-
         56; Federal Insecticide, Fungicide, and Rodenticides Act, Washington, DC, U.S. Environmental
         Protection Agency, SAP Report No. 2000-02, August 2, 2000. (http://www.epa.gov/OSA/hsrb/
         backgrounddocuments.htm)
     26. P. M. Arguin, P. E. Kozarsky, and A. W. Navin, Health Information for International Travel, 2005–
         2006. Travelers’ Health: Yellow Book, Atlanta, GA: U.S. Centers for Disease Control and
         Prevention, 2005, Chap. 2. (http://www.cdc.gov/travel/yb/about.htm)
     27. CDC, Updated Information Regarding Mosquito Repellents, Fort Collins, CO: U.S. Department of
         Health and Human Services, Centers for Disease Control and Prevention, National Center for
         Infectious Diseases, Division of Vector-Borne Infectious Diseases, 2005. (http://www.cdc.gov/
         ncidod/dvbid/westnile/RepellentUpdates.htm)
     28. E. A. Ross et al., Insect repellant [sic] interactions: Sunscreens enhance DEET (N,N-diethyl-m-
         toluamide) absorption, Drug Metab. Dispos., 32, 783, 2004.
     29. CTFA, International Cosmetic Ingredient Dictionary and Handbook, 10th ed., Washington, DC:
         Cosmetic, Toiletry, and Fragrance Association, 2004. (http://www.ctfa.org/Content/Navigation-
         Menu/CTFA_Books_CDs/Dictionary.htm)
     30. D. N. Roy, S. H. Ghosh, and R. N. Chopra, Comparative efficacy of different culicifuges under
         laboratory conditions, Parasitology, 34, 152, 1942.
     31. D. N. Roy and S. H. Ghosh, Further work on the comparative efficacy of different culicifuges under
         laboratory conditions, Parasitology, 34, 291, 1942.
     32. M. R. Berenbaum, The chemistry of defense: Theory and practice, Proc. Natl. Acad. Sci. U.S.A., 92,
         2, 1995.
     33. C. F. Curtis, J. Myamba, and T. J. Wilkes, Comparison of different insecticides and fabrics for anti-
         mosquito bednets and curtains, Med. Vet. Entomol., 10, 1, 1996.
     34. World Health Organization. IR3535; KBR3023; (RS)-methoprene 20%; pyriproxyfen 0.5% GR; and
         lambda-cyhalothrin 2.5% CS, Report of the Fourth WHOPES Working Group Meeting IR3535;
         KBR3023; (RS)-methoprene 20%; pyriproxyfen 0.5% GR; and lambda-cyhalothrin 2.5% CS,
         Document WHO/CDS/WHOPES/2001.2. Geneva: WHO, 2001.
     35. Y. Trongtokit et al., Comparative repellency of 38 essential oils against mosquito bites,
         Phytotherapy Res., 19, 303, 2005.
     36. D. Barnard, Repellency of essential oils to mosquitoes (Diptera: Culicidae), J. Med. Entomol., 36,
         625, 1999.
     37. D. L. Kline et al., Olfactometric evaluation of spatial repellents for Aedes aegypti, J. Med. Entomol.,
         40, 463, 2003.
     38. J. S. Kennedy, The excitant and repellent effects on mosquitoes of sub-lethal contacts with DDT,
         Bull. Entomol. Res., 37, 593, 1947.
                                      `
     39. M. Coluzzi, Sulla irritabilita al DDT in Anopheles, Riv. Malariol., 42, 208, 1963.


q 2006 by Taylor & Francis Group, LLC
Terminology of Insect Repellents                                                                               45


    40. E. J. Pampana, A Textbook of Malaria Eradication, 2nd ed., London: Oxford University Press, 1969,
        pp. 183–192.
    41. D. R. Roberts et al., A probability model of vector behavior: Effects of DDT repellency, irritancy,
        and toxicity in malaria control, J. Vector Ecol., 25, 48, 2000.
    42. E. B. Dogan and P. A. Rossignol, An olfactometer for discriminating between attraction, inhibition,
        and repellency in mosquitoes (Diptera: Culicidae), J. Med. Entomol., 36, 788, 1999.
    43. G. S. Fraenkel and D. L. Gunn, The Orientation of Animals, Oxford: Clarendon Press, 1940.
    44. Committee to Advise on Tropical Medicine and Travel (CATMAT), Advisory Committee Statement
        13, Public Health Agency of Canada, Statement on Personal Protective Measures to Prevent
        Arthropod Bites, Can. Commun. Dis. Rep., 31, ACS-13, 1–20, 2005 (http://www.phac-aspc.gc.ca/
        publicat/ccdr-rmtc/05vol31/asc-dc-13/index.html).
    45. J. P. Grieco, A novel high throughput screening system to evaluate the behavioral response of adult
        mosquitoes to chemicals, J. Am. Mosq. Control Assoc., 21, 404, 2005.
    46. S. E. Lee, Mosquito larvicidal activity of pipernonaline, a piperidine alkaloid derived from long
        pepper, Piper longum. J. Am. Mosq. Control Assoc., 16, 245, 2000.
    47. I. M. Scott et al., Botanical insecticides for controlling agricultural pests: Piperamides and the
        Colorado Potato Beetle Leptinotarsa Decemlineata Say (Coleoptera: Chrysomelidae), Arch. Insect
        Biochem. Physiol., 54, 212, 2003.
    48. I. M. Scott et al., Efficacy of Piper (Piperaceae) extracts for control of common home and garden
        insect pests, J. Econ. Entomol., 97, 1390, 2004.
                                                        `
    49. J. T. Arnason, T. Durst, and B. J. R. Philogen, Prospection d’insecticides phytochimiques de plantes
              ´ ´                                                                   ´ ´
        temperees et tropicales communes ou rares, in Biopesticides d’origine vegetale, C. Regnault-Roger,
                         `
        B. J. R. Philogene, and C. Vincent (Eds.), Paris: Editions TEC and DOC, 2002, pp. 37–51.
    50. M. Miyakado et al., The Piperaceae amides, I: Structure of pipercide, a new insecticidal amide from
        Piper nigrum L, Agric. Biol. Chem., 43, 1609, 1989.
    51. M. Miyakado, I. Nakayama, and H. Yoshioka, Insecticidal joint action of pipercide and co-occurring
        compounds isolated from Piper nigrum L, Agric. Biol. Chem., 44, 1701, 1980.
    52. S. Dev and O. Koul, Insecticides of natural origin, Amsterdam: Hardwood Academic, 1997.
    53. I. M. Scott et al., Insecticidal activity of Piper tuberculatum Jacq. extracts: synergistic interaction of
        piperamides, Agric. Forest Entomol., 4, 137, 2002.
    54. S.P. Carroll and J. Loye, A registered botanical mosquito repellent with deet-like efficacy., J. Am.
        Mosq. Control Assoc., 21, in press.
    55. C. F. Curtis, Control of Disease Vectors in the Community, London: Wolfe, 1990, pp. 79–80.
    56. F. Coro and S. Suarez, Review and history of electronic mosquito repellers, Wing Beats, 11, 6, 2000.
        http://www.floridamosquito.org/WING/WBindex.html
    57. WHO, 7th Report of WHO Expert Committee on Insecticides, Geneva: World Health Organization.
        Technical Report Series, No. 125, 1957.
    58. J. Hemingway and H. Ranson, Insecticide resistance in insect vectors of human disease, Annu. Rev.
        Entomol., 45, 371, 2000.
    59. WHO, Vector Resistance to Pesticides, 15th Report of the WHO Expert Committee on Vector
        Biology and Control. World Health Organization, Technical Report Series, No. 818, 1992.
    60. WHO, Test Procedures for Resistance Monitoring in Malaria Vectors, Bio-efficacy and Persistence
        of Insecticides on Treated Surfaces, Report of the WHO Informal Consultation, document WHO/
        CDS/CPC/MAL/98.12, Geneva: World Health Organization, 1998.
    61. D. W. Kolpin, Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S.
        streams, 1999–2000: a national reconnaissance, Environ. Sci. Technol., 36, 1202, 2002.
    62. M. W. Sandstrom et al., Widespread detection of N,N-diethyl-m-toluamide in U.S. streams:
        Comparison with concentrations of pesticides, personal care products, and other organic wastewater
        compounds, Environ. Toxicol. Chem., 24, 1029, 2005.
    63. L. C. Rutledge et al., Studies on the inheritance of repellent tolerances in Aedes aegypti, J. Am.
        Mosq. Control Assoc., 10, 93, 1994.
    64. L. C. Rutledge, R. K. Gupta, and Z. A. Meher, Evolution of repellent tolerances in representative
        arthropods, J. Am. Mosq. Control Assoc., 13, 329, 1997.


q 2006 by Taylor & Francis Group, LLC
46                                                       Insect Repellents: Principles, Methods, and Uses


     65. H. J. Becker, The genetics of chemotaxis in Drosophila melanogaster: selection for repellent
         insensitivity, Mol. Gen. Genet., 107, 194, 1970.
     66. D. A. Nordlund, R. L. Jones, and W. J. Lewis (Eds.), Semiochemicals: Their Role in Pest Control,
         New York: Wiley, 1981.
     67. A. J. Thorsteinson, The experimental study of the chemotactic basis of host specificity in
         phytophagous insects, Canadian Entomologist, 87, 49, 1955.
     68. The WHO Recommended Classification of Pesticides by Hazard and Guidelines to Classification,
         International Programme on Chemical Safety. Geneva: World Health Organization, Geneva, 2004.
         http://www.inchem.org/pages/pds.html
     69. WHO, Pesticides and their Application for the Control of Vectors and Pests of Public Health
         Importance, 6th ed., WHO Department of Control of Neglected Tropical Diseases, and WHO
         Pesticides Evaluation Scheme (WHOPES), document WHO/CDS/NTD/WHOPES/GCDPP/2006.1,
         Geneva: World Health Organization, 2006.
     70. A. N. Gilbert and S. Firestein, Dollars and scents: commercial opportunities in olfaction and taste,
         Nat. Neurosci., 5(11), Supplement, Beyond the Bench: The Practical Promise of Neuroscience, 1045,
         2002.
     71. International Programme on Chemical Safety, Inventory of IPCS and other WHO pesticide
         evaluations and summary of toxicological evaluations performed by the Joint Meeting on Pesticide
         Residues (JMPR), Evaluations through 2005, document WHO/PCS/06.2, World Health Organiz-
         ation, Geneva, 2005.




q 2006 by Taylor & Francis Group, LLC
3
Vertebrate Chemical Defense: Secreted and Topically
Acquired Deterrents of Arthropods


Paul J. Weldon and John F. Carroll



CONTENTS
Introduction .....................................................................................................................................47
Arthropod Deterrents from Tetrapods ............................................................................................48
  Amphibians ..................................................................................................................................48
  Snakes ..........................................................................................................................................51
  Birds.............................................................................................................................................51
  Mammals .....................................................................................................................................54
     Ungulates .................................................................................................................................54
     Humans ....................................................................................................................................59
Anointing.........................................................................................................................................61
Fumigation.......................................................................................................................................64
Discussion........................................................................................................................................65
Acknowledgments ...........................................................................................................................67
References .......................................................................................................................................67



         . for we know that the distribution and existence of cattle and other animals in South America
         absolutely depends on their power of resisting the attacks of insects: so that individuals which
         could by any means defend themselves from these small enemies, would be able to range into new
         pastures and thus gain a great advantage.
                                                                                          (Darwin, 1857)1




Introduction
Arthropods profoundly affect the fitness of terrestrial vertebrates. Arachnids, centipedes, and insects
opportunistically prey on small tetrapods. Some social hymenopterans—ants, bees, and wasps—fiercely
defend their colonies against intruders, including a host of foraging vertebrates, via multiple stinging or
biting attacks. Pelage- or plumage-degrading arthropods, such as lice, imperil their hosts by
compromising the insulative and other qualities of the integument. The feeding activities of
hematophagous insects, mites, and ticks irritate, weaken, and exsanguinate their victims. These


                                                                                                                                                  47

q 2006 by Taylor & Francis Group, LLC
48                                                     Insect Repellents: Principles, Methods, and Uses


ectoparasites also transmit debilitating or lethal pathogens ranging from viruses to parasitic arthropods
and inflict wounds vulnerable to microbial infection. Predatory, aggressive, and ectoparasitic arthropods
are pervasive and potent agencies of natural selection that have forged in vertebrates an array of
defensive adaptations.
   This chapter reviews evidence and suggestions that amphibians, snakes, birds, and mammals use
chemicals in defense against arthropods. We focus on semiochemical effects rather than on mechanical
protection afforded by secreted cuticles or adhesives. We also examine self anointing, where scent-laden
materials are rubbed against the integument, and fumigation, which involves exposure of the integument
to volatile compounds, as mechanisms by which tetrapods combat arthropods.2–7 A number of functions
have been proposed for topically acquired chemicals; most authors suggest that these substances thwart
predators, ectoparasites, and/or pathogenic microbes.7
   A variety of terms have been used to denote chemicals that affect nuisance arthropods (Chapter 2).
A repellent, as Dethier et al.8 proposed, denotes a chemical that elicits orientation away from its source
(cf. Barton Browne9). Thus, we confine our use of this term to cases in which arthropods avoid chemicals.
We use “deterrent” to refer to any defensive chemical, including biocides, that reduces the risk of bodily
harm.10




Arthropod Deterrents from Tetrapods
Amphibians
Amphibians harbor an array of bioactive compounds in their skin, including alkaloids, bufadienolides,
and peptides.11 The skin chemicals of a number of amphibians are believed to deter predatory and/or
ectoparasitic arthropods.
   Field experiments in Costa Rica examined the acceptability of the poison frog Dendrobates pumilio
and frogs of the genus Eleutherodactylus (presumed to lack skin toxins) as prey to the large ant
Paraponera clavata12 and the ctenid spider Cupiennius coccineus.13 Ants typically refused to eat
Dendrobates pumilio, retreating after touching frogs with their antennae, or releasing them after biting.12
Ants that bit Dendrobates pumilio typically wiped their jaws on their forelegs or tree bark. In contrast,
ants readily attacked Eleutherodactylus spp., usually fatally. Similarly, the spider Cupiennius coccineus
grasped and bit both kinds of frogs, but released Dendrobates pumilio and ate the Eleutherodactylus
spp.13 The tendency of these arthropods to reject Dendrobates pumilio after biting or antennating it
suggests that they rely upon contact chemoreception to recognize this frog.
   Aquatic insects are major predators of amphibian larvae. Brodie and colleagues14,15 tested predatory
larvae of the diving beetle (Dytiscus verticalis), nymphs of the giant water bug (Lethocerus americanus),
and crayfishes (Cambarus diogenes and Orconectes propinquus) for responses to larval and metamorphic
anurans (Bufo americanus, Hyla crucifer, Rana clamitans, Rana palustris, and Rana sylvatica) and
urodeles (Ambystoma maculatum and Notophthalmus viridescens) from the eastern United States. All
larvae were attacked and consumed. Metamorphic stages of Bufo americanus, Rana palustris, Rana
sylvatica, Ambystoma maculatum, and Notophthalmus viridescens, on the other hand, generally were
rejected by the predators and survived.
   Histological studies of the skin of Rana sylvatica revealed that the unpalatability of this frog’s
metamorphic stage correlates with the development of epidermal granular glands.15 The glands are
sparsely distributed in the larval stages and become abundant, larger, and appear active in late
metamorphic stages. The greater immunity to predation of adults of this frog and other amphibians
was attributed to the elaboration of defensive skin chemicals during maturation.
   Peterson and Blaustein16 tested diving beetles (Dytiscus sp.) and giant water bugs (Lethoceros
americanus) for feeding responses to various developmental stages of a toad, Bufo boreas, and two frogs,
Hyla regilla and Rana cascadae. These investigators failed to find evidence of stage-specific or species


q 2006 by Taylor & Francis Group, LLC
Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods                     49


differences in the palatabilities of these anurans. Peterson and Blaustein attribute the discrepancies of
their results with those obtained in other studies to possible species differences in chemical defenses or
predator tolerances, or to problems in the design of previous investigations.
   Predators experienced with noxious amphibians often develop an aversion to them. This effect was
demonstrated in the diving beetle (Dytiscus verticalis).17 Wild-caught diving beetle larvae were confined
with eastern red-spotted newts (Notophthalmus viridescens), which are unpalatable to vertebrate
predators18 and leeches.19 When tested later, beetle larvae rejected newts’ tails or aqueous extracts of
newts absorbed onto cotton swabs. Thus, newts’ skin chemicals acted as an aposematic cue. Starved
beetles, however, seized newts more often than did beetles that had been fed meat.20 Thus, although
beetles were averse to (water-borne) chemicals from newts, their tendency to attack newts increased
with hunger.
   Some frogs reside in ant nests, spider burrows or scorpion retreats where they access humid estivation
cavities, refuge from predators, and/or prey (see Rodel and Braun21). Skin chemicals are believed to
                                                        ¨
protect these anurans from the arthropods with which they associate. Rodel and Braun,21 for example,
                                                                              ¨
studied the frogs Kassina fusca and Phrynomantis microps, which live in nests of the ponerine ants
Megaponera foetens and Pachycondyla tarsatus in the savanna of West Africa; Pachycondyla tarsatus is
the largest African ant, and a hunter and scavenger. Humans were attacked when ant nests were
excavated, but the frogs, whose burrows occur deeply within the nests, were unmolested.
   Encounters were staged between Pachycondyla ants and five frog species: the two species mentioned
above that reside in ant nests; Hemisus marmoratus, which estivates underground (and may encounter
ants); and Phrynobatrachus latifrons and Ptychadena maccarthyensis, both of which estivate above
ground. Phrynobatrachus and Ptychadena were stung by ants and killed immediately, but the other frogs
were unharmed. Phrynomantis microps assumes a crouched posture in the presence of ants and allows
them to antennate, lick, and climb on its body (Figure 3.1). Ants that bit this frog wiped their mouthparts
and antennae and the results of the encounter were fatal for one ant.
   The possible ant-deterrent properties of frog skin chemicals were examined by rubbing the ant-
susceptible Phrynobatrachus latifrons against a live ant inquiline, Phrynomantis microps, and comparing
ants’ responses to skin-rubbed versus untreated frogs. Untreated frogs generally were stung and killed;
whereas only one of four Phrynomantis-treated frogs was stung. In another experiment, termites
(Macrotermes bellicosus) were dipped into water that had contained Phrynomantis frogs, and were
then confined with ants. Ants stung both Phrynomantis-treated and untreated termites, but they attacked
the latter more readily. Rodel and Braun21 postulated that frogs are tolerated by ants because their skin
                           ¨
contains a “stinging inhibitor.” Similarly, in Cameroon, the frog Kassina senegalensis resides in nests of
the ant Megaponera foetens, and may rely upon mollifying chemicals, possibly mimics of ant
pheromones, to do so.22
   The “ant frog” (Lithodytes lineatus), which ranges from the Peruvian Amazon to Surinam, typically
inhabits nests of the leaf-cutting ant Atta cephalotes.23 Wild-caught frogs emit a scent similar to that of
the “maggi plant” (Levisticum officinale), but captive-reared individuals and those held captive for
several months lacked this scent. When placed near ant colonies, unscented frogs were attacked and
killed, even by the colonies from which they were obtained. These observations imply that Lithodytes
lineatus must emit special chemicals to enter ant nests unharmed.
   Williams et al.24 tested the Australian sheep blowfly (Lucilia cuprina) and the eastern goldenhaired
blowfly (Calliphora stygia) with skin secretions of the brown tree frog (Litoria ewingi). These blowflies
are carrion feeders and do not pose a threat to frogs, but related flies parasitize frogs. An aqueous solution
of Litoria skin secretions topically applied to third-instar larvae of Lucilia cuprina increased their
mortality, but did not affect their behavior. When the skin solution was poured into glass vials containing
adult Calliphora stygia, immersing their tarsi, flies vigorously groomed, exhibited “frantic” uncoordi-
nated behavior, and became inverted. All flies died within 15 minutes. Calliphora stygia did not refuse
food treated with frog secretions, but died after eating it. Williams et al. suggested that both physical and
chemical properties contribute to the toxicity of Litoria skin secretions. Contact toxicity may be due to
the occlusion of flies’ spiracles and the obstruction of respiration.


q 2006 by Taylor & Francis Group, LLC
50                                                    Insect Repellents: Principles, Methods, and Uses




FIGURE 3.1 A Phrynomantis microps from Africa adopts a crouched posture while being examined by ants
                                      ¨
(Pachycondyla tarsatus). (From M.-O. Rodel and U. Braun, Biotropica, 31, 178, 1999. With permission.)




   Many frogs and toads emit distinctive scents, especially when disturbed. Smith et al.25 suggested that
volatile compounds discharged by anurans deter predators and ectoparasites. Williams et al.26
investigated whether the skin secretions of five Australian frogs—Litoria caerulea, Litoria splendida,
Litoria rubella, Litoria rothi, and Uperoleia mjobergi—repel the mosquito, Culex annulirostris.
A previous analysis of the parotoid gland secretions of Litoria caerulea had revealed b-caryophyllene,
a presumed mosquito repellent.27 Mosquitoes were confined in a choice chamber into which airsteams
drawn over frogs’ skin secretions or water (control) were introduced, and the distribution of mosquitoes
on either side of the apparatus was monitored. Mosquitoes avoided the frog-scented side the chamber
only in response to the scents of Litoria rubella and Uperoleia mjobergi. The scent of Litoria caerulea
was marginally repellent.
   To examine whether the skin chemicals of Litoria caerulea deter biting by mosquitoes, Williams
et al.26 applied an aqueous rinse of this frog’s skin secretions to the tails of mice (Mus musculus) and
allowed mosquitoes to bite them; plain water was applied to control mice. Mosquitoes exhibited a greater
latency (up to 50 minutes) to bite the secretion-treated tails than the controls. Analogous results were
obtained with Litoria ewingi, the skin of which emits eucalyptol, limonene, and a-pinene.28 The
secretions of this frog applied to a human forearm delayed landing by Culex annulirostris for up to
approximately 10 minutes.


q 2006 by Taylor & Francis Group, LLC
Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods                                 51


Snakes
A number of reptiles, including snakes, lizards, and amphisbaenians, take shelter, feed, or deposit their
eggs within insect colonies, somehow avoiding sustained attack. Chemicals seem likely to mediate these
interactions, but explicit suggestions for this are rare.
   The Texas blind snake (Leptotyphlops dulcis) is a burrower that travels in columns of foraging army
ants (Neivamyrmex nigrescens) and feeds within ant colonies. When Leptotyphlops dulcis is manually
placed into raiding columns of ants, it is attacked.29 In response, it smears itself with feces and a clear
viscous fluid discharged from its cloaca.30 This exudate includes secretions from the cloacal scent glands,
which contain a glycoprotein and C12–C20 free fatty acids.31
   The deterrent properties of the cloacal exudate of Leptotyphlops dulcis against ants were examined by
wiping fluids off of a snake’s body, dissolving them in ethanol, and presenting them on the floor of an
arena to Neivamyrmex nigrescens and other ants.30 Ants placed at the juncture of areas treated with either
snake cloacal extract or ethanol spent more time in the ethanol-treated area. Ants’ aversion to snake
secretions was thought to be due to free fatty acids from the scent glands.31 Gehlbach et al.30 also
observed that ants failed to attack Leptotyphlops dulcis maintained in laboratory ant colonies, suggesting
that ant-derived chemicals acquired by snakes curtail attack.
   The Mexican short-tailed snake (Sympholis lippiens) is a fossorial, insectivorous species from western
Mexico that resides in colonies of leaf-cutting ants, Atta mexicana, which it eats. Sympholis possesses a
tough skin thought to protect it from biting ants and an oily epidermis that is believed to deter them.32


Birds
During the 1940s, Cott33 undertook a series of field experiments in Egypt and Lebanon on chemical
defense in birds, examining their acceptability to mammalian and insect predators (see Dumbacher and
Pruett-Jones34). While collecting bird specimens and preparing their skins, Cott noticed that the
discarded carcasses of the laughing dove (Streptopelia senegalensis) were vigorously consumed by
Oriental hornets (Vespa orientalis), whereas pied kingfisher (Ceryle rudis) carcasses were ignored
(Figure 3.2). Further comparisons of hornets’ responses to bird carcasses culminated in more than 140




FIGURE 3.2 Oriental hornets (Vespa orientalis) attack a freshly skinned carcass of a laughing dove (Streptopelia
senegalensis) (left), while ignoring the carcass of a pied kingfisher (Ceryle rudis). (From H. B. Cott, Proceedings of the
Zoological Society of London, 116, 371, 1947. With permission.)


q 2006 by Taylor & Francis Group, LLC
52                                                      Insect Repellents: Principles, Methods, and Uses


experiments involving 38 bird species. Cott deemed hornets suitable as subjects because they convey
what they consume back to their colony, thus their hunger state was presumed not to have varied
substantially between experiments.
   Cott typically placed skinned pieces of two bird species side-by-side and observed hornets feeding on
them until one was consumed. Tissues from different parts of birds varied greatly in their acceptability to
hornets. The breasts and wings of Ceryle rudis, for example, were more acceptable than were the legs and
visceral tissues. The breasts of other species also generally were preferred, in contrast to their legs which
were consumed last or remained uneaten. Adipose tissue was uniformly avoided. Because different
tissues varied in their acceptability, Cott compared responses by hornets to homologous tissues from
different bird species.
   In addition to Ceryle rudis, the following birds were deemed unpalatable to hornets: the kingfisher
(Alcedo atthis), the blackcap (Sylvia atricapilla), the golden oriole (Oriolus oriolus), the hoopoe (Upupa
epops), chats (Oenanthe spp.), shrikes (Lanius spp.), and swallows (Hirundo spp.). Interestingly, hornets
and domesticated cats (Felis catus) exhibited similar feeding preferences for bird flesh. The
unpalatability of these birds in Cott’s study implies that distasteful chemicals occur systemically
because the specimens he presented were skinned. The responses of consumers to intact birds need to
be assessed to evaluate this as a mechanism of avian defense.
   Thiollay35 reported that the red-throated caracara (Daptrius americanus), an insectivorous, forest-
dwelling raptor ranging from Mexico to Brazil, emits volatiles that deter hymenopterans. He provided
the following account of this bird attacking wasp nests: “As soon as one bird reached a nest, all the
insects abandoned it and never attacked the raider, nor followed it when it carried the nest away.
The wasps flew at a distance around the bird, rarely coming nearer than 1 m as long as it was on the nest.
They returned to the remains of the nest shortly after the caracara left, sometimes within a few seconds
.. The fact that wasps never attacked, nor even closely approached the caracaras raiding their nests,
suggests the involvement of some powerful chemical repellent.” Daptrius possesses a bare, thin-skinned
face and throat, a possible source of the putative wasp deterrents.36
   Daptrius americanus is not consumed by humans in Guiana due to its disagreeable odor and taste.35 In
Mexico, this bird is avoided because a bluish dust on its feathers is believed to be poisonous.37 Whether
the chemicals from Daptrius thought by humans to be noxious are related to those that putatively deter
wasps is unknown. Nonetheless, many birds are regarded by humans as characteristically malodorous or
unpalatable, which may reflect their possession of deterrents against natural enemies.38
   Tribespeople and field biologists in New Guinea consider the hooded pitohui (Pitohui dichrous) to be
unpalatable and noxious because individuals who handle it typically sneeze and experience numbness
and a burning sensation. Dumbacher et al.39 analyzed the feathers and other tissues of Pitohui dichrous,
the variable pitohui (Pitohui kirkocephalus), and the rusty pitohui (Pitohui ferrugineus). They discovered
that the feathers and, to a lesser extent, the muscle tissue of these birds contain homobatrachotoxin
(HBTX), a steroidal alkaloid first characterized from the skin of neotropical dendrobatid frogs (genus
Phyllobates). HBTX is a potent neurotoxin that binds sodium channels and depolarizes electrogenic
membranes. Further analyses demonstrated HBTX, batrachotoxinin-A, and other batrachotoxins in
additional pitohui species and in an unrelated New Guinean bird, the blue-capped ifrita (Ifrita kowaldi).40
HBTX and other batrachotoxins have since been discovered in New Guinean melyrid beetles (Choresine
spp.) that are consumed by pitohuis.41 The Melyridae is a cosmopolitan family thought to be the dietary
source of alkaloid toxins in both neotropical frogs and New Guinean birds. Alkaloids also occur in the
feathers of the red warbler (Ergaticus ruber) from Mexico, an insectivorous bird reputed to be inedible to
humans.42
   Dumbacher et al.39 suggested that HBTX protects birds against predatory vertebrates, such as snakes.
Other authors have postulated that this toxin combats ectoparasites.43,44 Poulsen43 estimated the amount
of HBTX in the skin of Pitohui dichrous, the most toxic pitohui, to be several orders of magnitude lower
than that of the most toxic dendrobatid frog, Phyllobates terribilis. He surmised that higher toxin levels
are necessary to deter vertebrates as formidable as those that threaten pitohuis, leaving ectoparasites as
more likely targets. Mouritsen and Madsen44 noted that batrachotoxins are toxic to a wide variety of


q 2006 by Taylor & Francis Group, LLC
Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods                   53


insects, and they inferred that these compounds affect a spectrum of ectoparasitic arthropods. These
authors also cited a survey of New Guinean birds in which pitohuis were found to harbor among the
lowest tick loads of 30 passerine genera examined.45
   Dumbacher46 tested chewing lice for responses to the feathers of pitohuis and other birds. Lice were
collected from 17 species of free-ranging birds in New Guinea, including pitohuis, and then confined in
Petri dishes where they were given a choice between two contour feathers from different species. Lice
presented with the feathers of pitohuis versus nonpitohuis, or the feathers of Pitohui dichrous versus
those of the less toxic Pitohui cristatus, avoided the feathers of Pitohui dichrous in all cases. When
exposed to the contour feathers from a pitohui and a nontoxic bird, louse mortality was higher with the
pitohui feathers. For example, lice confined with feathers from Colluricincla megarhyncha, which was
presumed to be nontoxic (cf. Dumbacher et al.40), survived an average of 417 hours, but they did so for
only 37 hours with the feathers of Pitohui dichrous. Louse species differed greatly in their tolerance of
pitohui feathers. An undetermined species of Brueelia was the most sensitive; Neopsittaconirmus
circumfasciatus was the most resistant. Dumbacher46 postulated that lice that have coevolved with
pitohuis are more tolerant of HBTX.
   Burtt47 pointed out that the choice of feathers by lice in Dumbacher’s study, and the consequences of
that choice on their survivorship, may have been influenced by feather microstructure peculiar to each
bird species. He suggested that tests of lice from a single bird species would have improved Dumbacher’s
experimental design. In addition to this concern, subsequent studies have revealed highly variable or
undetectable levels of HBTX and the presence of other toxic compounds in pitohui feathers.40 Thus it is
unclear whether Dumbacher’s46 results can be attributed solely to HBTX. Nonetheless, his study
indicates that natural concentrations of chemicals from pitohui feathers adversely affect some lice.
   The effects of pitohui toxins on other arthropods need to be examined. Hippoboscid flies, which are
hematophagous, often occur on pitohuis in nature.46 Hippoboscids, along with mosquitoes and biting
midges, transmit the pathogens that cause avian malaria, Haemoproteus, Leucocytozoon, and Plasmo-
dium. One survey of these malarial parasites among birds of Australia and New Guinea revealed an
infection rate for pitohuis that was similar to both the overall infection rate for their family, the
Pachycephalidae, and the average prevalence of these parasites in New Guinea.48 These results do not
support the notion that toxins protect pitohuis against vectors of malarial parasites. Another survey of 45
bird species from southeastern New Guinea, on the other hand, revealed that the Pachycephalidae,
including pitohuis, had the lowest hematozoan loads, with no mature gametocytes in 21 individuals
examined.49
   The crested auklet (Aethia cristatella) and the whiskered auklet (Aethia pygmaea) are planktivorous,
colonial seabirds from the Bering Sea and North Pacific Ocean. These birds emit a citrus-like aroma from
their plumage that humans can detect emanating from their colonies.50 Douglas et al.50,51 identified the
following volatiles associated with this scent from the feathers of Aethia cristatella: n-octanal and, in
lesser amounts, n-hexanal, n-decanal, (Z)-4-decenal, (Z)-4-dodecenal, (Z)-6-dodecenal, and hexanoic
and octanoic acids. Hexadecanol, heptanal, nonanal, and decanal were identified from Aethia pygmaea.
Douglas et al.50 noted that similar aldehydes are used by heteropteran insects to repel predators, and they
suggested that these compounds serve auklets as repellents of ectoparasitic arthropods. In addition,
auklet volatiles were hypothesized to act as signals of mate quality related to the enhanced fitness
associated with ectoparasite deterrence.
   Douglas et al.51 tested the ticks Ixodes uriae (which parasitizes auklets) and Amblyomma americanum
for responses to volatiles from the crested auklet. Laboratory tests of Amblyomma americanum nymphs
were conducted by applying compounds to filter papers and attaching them to a heated rotating drum that
served as an artificial host. Fewer nymphs transferred to papers treated with octanal or the mixture of
auklet-derived volatiles than to control papers, and they spent less time on them. A field test indicated
that Ixodes uriae nymphs were deterred by octanal and, to lesser extent, by decanal and a mixture of
aldehydes. Ixodes uriae nymphs became moribund within 15 minutes after confinement with octanal;
adults did so within one hour.


q 2006 by Taylor & Francis Group, LLC
54                                                     Insect Repellents: Principles, Methods, and Uses


   The suspicion that crested auklets in colonies near tundra habitats are vulnerable to mosquitoes
prompted Douglas et al.52 to test the effects of auklet volatiles on the yellow fever mosquito (Aedes
aegypti). Hexanal, octanal, (Z)-4-decenal, and hexanoic acid were applied separately or blended to filter
papers laid on human hands that were then inserted into a chamber with mosquitoes; ethanol served as a
control. All auklet-derived compounds deterred landing by mosquitoes. Octanal alone was as effective as
the blend of volatiles and hexanal was as effective as hexenoic acid. Douglas et al.52 postulated that
volatiles emitted by auklets act against a broad spectrum of ectoparasitic arthropods.
   Investigations of the effects of auklet volatiles on lice, however, have yielded ambiguous or negative
results. Lice of the genera Austromenopon and Quadraceps from wild-caught birds in Alaska became
moribund within seconds after being confined with feathers treated with octanal or (Z)-4-decenal.51 The
louse loads of free-ranging crested auklets, however, were higher than those of the least auklet (Aethia
pusilla), which lacks a noticeable odor.53 Moreover, when lice from the rock dove (Columba livia),
Campanulotes compar and Columbicola columbae, were confined with feathers from the crested auklet,
least auklet, or rock dove, or were placed into beakers containing the carcasses of these birds, their
survivorship did not differ between treatments. Douglas et al.53 concluded that natural concentrations of
crested auklet volatiles are not lethal to lice, but may repel or otherwise deter them.
   Most extant birds possess on their rump a uropygial gland which secretes an oil that is spread through
the plumage during preening. Uropygial gland lipids inhibit microbial growth and thus retard feather
degradation (see Moyer et al.54). Poulsen43 suggested that uropygial gland secretions also repel
ectoparasitic arthropods.
   Moyer et al.54 examined the effects of uropygial gland secretions from the rock dove on its host-
specific feather lice, Campanulotes compar and Columbicola columbae. Uropygial gland oils applied to
feathers obtained from glandectomized birds doubled the mortality of lice confined in jars, but louse
loads on glandectomized versus intact birds did not differ. In fact, one dove that lacked a uropygial gland
had among the lowest observed louse loads. Moyer et al.54 discussed reasons for the disparate results of
their in vitro and in vivo tests, including the possibility that excessive amounts of oil in their in vitro
experiment killed lice by clogging their spiracles. These investigators entertained the prospect that
uropygial gland secretions normally do not affect feather lice.


Mammals
Ungulates
Investigations of the chemosensory responses by arthropods to mammals focus on attraction by
ectoparasites to hosts. However, some studies reveal that chemicals from mammals deter these pests.
   Tsetse flies (Glossina spp.) are hematophagous vectors of African trypanosomiasis. Blood-meal
analyses of Glossina spp. in various parts of Africa reveal that they prefer particular mammalian hosts
(see Galun55 and Gikonyo et al.56). Tsetse flies obtain a preponderance of blood-meals from a few
ungulates, such as the warthog (Phacochoerus aethiopicus), the bushpig (Potamochoerus porcus), the ox
(Bos taurus), and the bushbuck (Tragelaphus scriptus), but rarely attack other species, including the
waterbuck (Kobus defassa), the hartebeest (Alcelaphus buselaphus), and the impala (Aepyceros
melampus), even when these mammals are abundant. Nash57 discussed why some mammals may be
rejected as hosts by tsetse flies. He suggested that Glossina morsitans is averse to the scent of
the hartebeest.
   Vale et al.58 studied the responses of Glossina morsitans and Glossina pallidipes in Zimbabwe to ox
urine and urinary phenols. Whole urine and some phenols attracted flies to traps or visual targets.
2-Substituted phenols, on the other hand, suppressed attraction, an observation confirmed with Glossina
pallidipes by Torr et al.59 Species differences were observed in the synergistic effects of the phenols as
attractants or deterrents.58 4-Methylphenol (p-cresol) added to 3-n-propylphenol, for example, increased
the trap captures of Glossina pallidipes, but reduced those of Glossina morsitans.


q 2006 by Taylor & Francis Group, LLC
Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods                   55


   Madubunyi et al.60 investigated the responses of Glossina longipennis and Glossina pallidipes in
Kenya to the urine of two preferred hosts, the African buffalo (Syncerus caffer) and domesticated
cattle (Bos taurus), and a nonpreferred host, the waterbuck. These investigators observed no differences
in the trap catches with the urine of these animals, and they deemed it unlikely that tsetse flies locate
hosts, or distinguish between hosts and nonhosts, by the scent of urine. Madubunyi et al. postulated
that the chemicals used by tsetse flies to locate and distinguish among mammalian hosts arise from
skin glands.
   Vale61 conducted field studies in Zimbabwe to assess the responses of Glossina morsitans and
Glossina pallidipes to various ketones, aldehydes, alcohols, carboxylic acids, and other compounds,
some of which occur on the integument of mammals. A compound was characterized as “repellent” if it
suppressed the number of flies attracted to a black cylinder model, which served as a visual target, placed
near the release site of attractants, such as carbon dioxide or the scent of an ox. Acetophenone, which is
emitted by bats, elephants, and other mammals,62 and several related phenols; methyl ketones;
aldehydes; and carboxylic acids, especially caproic acid, suppressed attraction. Glossina
pallidipes was more effectively deterred by acetophenone and caproic acid than was Glossina morsitans
(see also Torr et al.59 on Glossina pallidipes). A greater proportion of females than males of both species
was deterred by acetophenone.61 Acetic acid deterred both (biting) stomoxyine and nonbiting
muscid flies.
   Gikonyo et al.56 tested Glossina morsitans for responses to a preferred host, the ox, and a nonpreferred
host, the waterbuck, in encounters and in feeding experiments using silicone membranes treated with the
pelage extracts of these ungulates. No differences were observed in the tendencies of tsetse flies to land
on live subjects or on extract-treated versus control membranes. However, flies that landed on a live
waterbuck or membranes treated with its sebum changed probing sites more often, probed for longer
periods, fed less frequently, and flew off sooner than did those landing on a live ox or membranes treated
with ox sebum. The compounds from waterbuck to which tsetse flies attend were postulated to exhibit
low volatility because the flies avoided waterbuck sebum after landing on treated membranes. However,
tsetse flies that landed near but not on membrane zones treated with high doses of waterbuck sebum
exhibited decreased feeding, thus indicating that they are capable of detecting volatile cues.
   Gikonyo et al.63 examined the electroantennogram (EAG) responses of Glossina morsitans and
Glossina pallidipes to volatiles extracted from absorbent pads impregnated with pelage chemicals from
two preferred hosts, an African buffalo and an ox, and from the nonpreferred waterbuck. Glossina
pallidipes reared in an insectary failed to respond to waterbuck volatiles, but field-trapped flies responded
with EAG activity to 13 gas chromatographic peaks. Insectary-reared Glossina morsitans exhibited EAG
responses to 14 components from waterbuck, and to 10 and 11 components from the ox and the buffalo,
respectively. The following EAG-active compounds were unique to waterbuck (if also present in buffalo
and ox, they occurred in trace amounts): d-octalactone, 2-methoxyphenol (guaiacol); 3-isopropyl–6-
methylphenol (carvacrol); 2-octanone; 2-nonanone; 2-decanone; 2-undecanone; 2-dodecanone; and
(E)-6,10-dimethyl-5,9-undecadien-2-one.
   Field studies by Vale and colleagues58,61 in Zimbabwe demonstrated that guaiacol and C4–C6
methylketones reduced the trap captures of Glossina morsitans and Glossina pallidipes. Paradoxically,
although a series of C5–C9 straight-chain carboxylic acids unique to the waterbuck failed to elicit
discernible EAG responses by tsetse flies in the study of Gikonyo et al.,63 two such compounds—
pentanoic and hexanoic acids—suppressed attraction of flies to hosts in the field.59 The various
compounds present in the waterbuck blend, including the carboxylic acids, were postulated to
differentially influence the distant- and close-range responses of tsetse flies to nonpreferred hosts.63
A notable result emerged from an interspecific comparison of flies’ EAG responses to waterbuck
volatiles: Glossina pallidipes, which attacks waterbuck more readily than does Glossina morsitans,
detected fewer methyl ketones of the repellent blend.
   Gikonyo et al.64 tested Glossina morsitans to blends of EAG-active compounds from the ox, buffalo,
and waterbuck in a choice wind tunnel. Compounds from the waterbuck included d-octalactone,
carvacrol, m-cresol, C7–C10 aldehydes, and C8–C13 methlyketones. Tsetse flies exposed to waterbuck


q 2006 by Taylor & Francis Group, LLC
56                                                     Insect Repellents: Principles, Methods, and Uses


volatiles failed to choose a consistent flight direction in the apparatus, as opposed to flies that embarked
directly toward the scent of preferred hosts, and they came to rest in the odorless (control) arm of the
tunnel. Flies exposed to waterbuck volatiles also flew shorter distances, made frequent and sharp in-flight
turns, and fanned their wings against the tunnel walls—ostensibly attempting to fly out of the apparatus.
   Bett et al.65 tested the efficacy of a synthetic blend of waterbuck volatiles in protecting oxen from
Glossina pallidipes in Kenya. The artificial blend was applied to sachets and attached to oxen tethering
posts. Flies were captured using electrified screens surrounding each bait animal. Feeding by Glossina
pallidipes was reduced nearly 95% by the waterbuck blend. This blend also deterred Glossina
swynnertoni, a species for which chemical control agents have been sought.65
   Some mosquitoes also are deterred by the skin chemicals of bovids. Weldon66 examined the responses
of Aedes aegypti to pelage extracts of more than 15 ungulates, primarily artiodactyls. Mosquitoes were
tested in a Plexiglas module using silicone feeding membranes, as described by Weldon et al.67 Wells in
the base of the module were filled with a 10% sucrose solution with added adenosine triphosphate and
green food coloring. Mosquitoes were confined in chambers, the floors of which opened to allow them
access to membranes placed over the wells. The number of mosquitoes landing on membranes treated
with acetone or hair extracts was monitored for five-minute trials. The number of mosquitoes feeding was
determined after each trial by crushing them on white paper towels and examining their remains for
green fluid.
   Extracts of hair and sebum from the gaur (Bos frontalis), a large (650–1,000 kg) bovid with a greasy
pelage that ranges from India to Indochina and the Malay Peninsula, significantly reduced landing and
feeding by mosquitoes. Fraction-directed bioassays and analyses of gaur pelage extracts suggested that
(6R, 9S, 10S)-10-hydroxy-6,9-oxidooctadecanoic acid deterred landing and feeding by these mosqui-
toes.68 Ishii et al.69 however, found that this compound, named 18-bovidic acid, exhibits the opposite
stereochemical configuration, namely 6S, 9R, 10R.
   A three-dimensional structure-activity model designed to identify potential insect repellents indicated
18-bovidic acid as a candidate compound.70 Feeding studies using the silicone membrane feeding system
described above confirmed that 18-(6S, 9R, 10R)-bovidic acid, purified from the sebum of a gaur, deters
landing and feeding by Aedes aegypti (Figure 3.3).71 This compound contains a tetrahydrofuranoid ring
that is adjacent to a hydroxyl group and is flanked by saturated hydrocarbon chains. This structure is
reminiscent of that of acetogenins from custard apples (Annonaceae), which possess anti-feeding and
biocidal properties against insects and other arthropods.72
   Costantini et al.73,74 studied mosquito attraction in Sudan using odor-baited entry traps, where air
drawn over humans or cattle confined in tents was conveyed through tubing and released near a trap;
carbon dioxide released in amounts comparable to those emitted by bait animals was released from
control traps. Anopheles gambiae, an anthropophilic mosquito and a significant vector of malaria in
Africa, exhibited an approximate 2:1 preference for human scent when compared to a human-equivalent
of carbon dioxide,73 but this species showed a nearly 20:1 preference for human scent when
tested against the scent of cattle74 (cf. Gillies75). A similar effect was observed with Anopheles
pharoensis. Costantini et al.74 stated that these results may reflect mosquitoes’ avoidance of cattle as
unsuitable hosts.
   Dekker and Takken76 conducted a field study in South Africa on the attraction of mosquitoes to a
human, a cattle calf, and carbon dioxide. Mosquitoes were captured in tents emitting the scents of these
animals or carbon dioxide released in amounts comparable to those exhaled by them. The zoophilic
mosquitoes Aedes mcintoshi, Anopheles coustani, Anopheles pretoriensis, and Anopheles rufipes were
less attracted to humans than to a human-equivalent of carbon dioxide. Similarly, Culex quinquefas-
ciatus, a species known to attack humans, preferred carbon dioxide over a live calf when this gas was
released at a rate equivalent to that emitted by the calf. None of the Culex quinquefasciatus caught with
the calf had fed, in contrast to those trapped in human-baited tents. These results imply that the
chemosensory basis of host selection by mosquitoes involves not only attraction to preferred hosts, but
avoidance of nonhosts.


q 2006 by Taylor & Francis Group, LLC
Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods                                   57




                           1.0
                                             O                    HO       H
                                                          H
                                                              O
                                        HO
                                                                       H
                           0.8
                           0.6
              Proportion
                           0.4
                           0.2
                           0.0




                                 1 mM            4 mM         11 mM               33 mM              acetone

FIGURE 3.3 Mean percentages and 95% confidence intervals of Aedes aegypti landing (!) and feeding (C) on silicone
membranes (9.6 cm2) covering wells containing 10% sucrose solution and treated with 50 mL of acetone or 1, 4, 11, and
33 mM solutions of 18-6S, 9R, 10R-bovidic acid (structure shown) in acetone. Fifty mosquitoes, five per trial, were observed
for 5 min with each treatment. Values for landing were backtransformed, as described in Weldon et al.67 An asterisk (*)
indicates values significantly different from the control. (P. A. Evans, W. J. Andrews, and P. J. Weldon, Unpublished.)




   Dekker et al.77 conducted wind-tunnel experiments on host-odor responses by Anopheles gambiae,
which is anthropophilic, and Anopheles quadriannulatus, which feeds predominantly on bovids
(cf. Pates et al.78). Mosquitoes were given a choice of airstreams laden with either human or cattle
scents emanating from nylon socks that had absorbed the skin secretions of these mammals, and they
were captured in traps upwind. The resulting trap captures reflected mosquitoes’ typical host
preferences: fewer Anopheles gambiae and Anopheles quadriannulatus entered airstreams containing
cattle and human scents, respectively. Although cattle scent contains ammonia, which attracts
Anopheles gambiae over a range of concentrations, more mosquitoes were caught in plain air traps
than in those releasing cattle scent. Dekker et al. inferred that one or more compounds from cattle
reduced the attractiveness of ammonia and possibly other cattle-derived volatiles. The deterrent effect
of cattle scent on Anopheles gambiae also was observed in olfactometry tests by Pates et al.79
   Observations of domestic cattle reveal variation within and between breeds in their attractiveness to
ectoparasitic flies (see Jensen et al.80). Jensen et al.80 studied interactions between the horn fly
(Haematobia irritans), an obligate, blood-feeding pest of pastured cattle in many parts of the world,
and herds of Holstein-Friesian heifers in Denmark. Some individual heifers were highly attractive to
flies, whereas others were fly-resistant. Heifers maintained their status with respect to fly-attractiveness
over the two-year duration of the study. The exchange of three or four fly-resistant heifers for fly-
susceptible individuals between herds of up to 17 cattle reduced overall fly loads for the herd, a pattern
that was reversed when heifers were returned to their original herd. Thus, the degree to which flies
menace a cattle herd depends upon the number of fly-susceptible and fly-resistant individuals it contains.
The propensity of some individual cattle to draw fewer flies was attributed to their ability to emit
chemicals that mask attractive cues.81
   Birkett et al.82 studied the role of volatiles from heifers in host selection by horn flies and face flies
(Musca autumnalis) using gas chromatography-electroantennography (GC–EAG), GC-mass


q 2006 by Taylor & Francis Group, LLC
58                                                       Insect Repellents: Principles, Methods, and Uses


spectrometry, EAG, and behavioral studies in the laboratory. Volatiles from heifers that exhibited high
and low fly loads were analyzed and tested for EAG and behavioral activities with flies. Both fly species
exhibited EAG responses to polar aromatic compounds, including phenol and o-cresol, m-cresol, and
p-cresol, and to nonpolar aromatics, such as naphthalene and acenaphthene. In addition, horn flies
exhibited EAG responses to the following volatiles from unattractive heifers: propylbenzene, styrene,
camphene, 2-heptanone, and propyl butanoate. Phenol, m-cresol, and p-cresol, which were present in
both attractive and unattractive samples, also were EAG-active.
   Compounds that exhibited EAG activity with an array of dipteran pests of cattle were presented to face
flies in wind-tunnel experiments.82 Propyl butanoate, a component of the least attractive heifer,
suppressed attraction, as did naphthalene and, from urine, linalool. Interestingly, 1-octen-3-ol and
6-methyl-5-hepten-2-one, which were characterized in laboratory tests as attractants, tended to deter horn
flies when artificially dispensed from free-ranging heifers. This result was attributed to the release of
abnormally high levels of these compounds. The study of Birkett et al. demonstrates the importance of
volatile cues in the differential attraction of hematophagous insects to individual bovids. Pickett and
Woodcock83 postulated that flies avoid individual cattle because they detect chemicals from them that
reflect their immunological competence, thus rendering them less suitable hosts.
   Breeds of domestic cattle differ markedly in their vulnerability to ticks. Pan84 suggested that reduced
tick loads of Sahiwal versus Jersey cattle are related to the greater production of sebum by the former
breed. Other authors also have noted a possible correlation between sebum production and tick resistance
among cattle breeds, e.g., Bonsma,85 but it is unclear whether semiochemicals are involved.
   The brown ear tick (Rhipicephalus appendiculatus) and the red-legged tick (Rhipicephalus evertsi)
from Africa feed chiefly inside the ears and in the anal region of ungulates, respectively. Wanzala et al.86
observed that these ticks locate their characteristic feeding sites when placed on different parts of a host’s
body. These investigators postulated that this ability involved concurrent responses to repulsive (from
distant sites) and attractive (from feeding sites) cues. When tested in a climbing bioassay with extracts
from different body regions of domestic cattle, brown ear ticks crawled toward ear volatiles and away
from volatiles from the anal region. Conversely, red-legged ticks were attracted to anal volatiles and
repelled by ear volatiles. The contrasting effects on ticks of chemicals from different body regions of their
host represent a “push–pull” system of feeding site location, a phenomenon that may be widespread
among organisms specializing on particular host microenvironments.86
   The repulsive effects of host chemicals described by Wanzala et al.86 facilitate the ticks’ location of
their characteristic feeding sites, thus it is not clear if host chemical defenses are involved. Nonetheless,
Sika87 observed that when the anal scent of cattle was artificially applied to the area around their ears,
brown ear ticks became disoriented, resulting in most subjects failing to locate their preferred
feeding site.
   Many ticks exhibit an arrestant response (akinesis) to mammalian skin chemicals, a normal questing
reaction in which ticks cease locomotion at ambush vantage points. Carroll and colleagues88–90 observed
that adult blacklegged ticks (Ixodes scapularis) and American dog ticks (Dermacentor variabilis)
generally became akinetic in response to secretions from skin glands (tarsal and metatarsal) on the legs of
the white-tailed deer (Odocoileus viginianus), but failed to do so in laboratory assays with some samples,
avoiding them. These results were suspected to have been due to the contamination of some samples by
urine, which mixes with glandular products during the deer’s rub-urination scent marking behavior. The
tendency of blacklegged ticks to avoid interdigital gland secretions from the hindlegs of deer, but not
those from the forelegs, is consistent with the hypothesized deterrent effect of urine.90
   Host-seeking blacklegged ticks tested under conditions of high humidity (ca. 95% RH) avoided urine
from bucks and nonestrous females of Odocoileus viginianus, but failed to do so at 50% RH.91 Urine
collected from the urinary bladder of a buck deterred ticks down to a 10,000-fold dilution. A subsequent
study examining ticks’ responses to urine from immature and adult male and female deer revealed that
only buck urine repelled them.92 Carroll91 suggested that fresh buck urine counteracts the arrestant
properties of the glandular residues to which ticks ordinarily attend in identifying hosts.


q 2006 by Taylor & Francis Group, LLC
Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods                  59


Humans
Vale93–96 and Hargrove97 studied responses by tsetse flies (Glossina morsitans and Glossina pallidipes)
in Rhodesia to humans and other mammals. These investigators observed that flies were averse to the
scent of humans and less attracted to preferred hosts, such as the ox, when humans were nearby. This
effect was especially pronounced in female flies, and thus could bias efforts to monitor tsetse fly
populations when hand-netting was used to collect them. Vale95 suggested that tsetse flies tend to avoid
humans and other primates because they are vigilant and dexterous hosts that can capture and kill flies
when they land.
   Vale98 investigated the source and nature of human-derived deterrents of tsetse flies. Human body
odor, but not human breath, significantly reduced fly catches in the field. When lactic acid, which recent
studies show is distinctively abundant on human skin,99 was applied to cloth, the catches of both Glossina
morsitans and Glossina pallidipes were reduced. To further evaluate lactic acid as a deterrent, the
number of flies attracted to an ox treated with 4.5 l of a 1% solution of this compound were monitored;
flies were examined to ascertain if they previously had fed. The catches of female Glossina morsitans and
both sexes of Glossina pallidipes were reduced 50–66% for fed flies, but the catches of unfed flies were
unaffected; too few male Glossina morsitans were caught to evaluate their response. With the exception
of male Glossina morsitans, fewer flies engorged on lactic acid-treated versus untreated oxen.
   A number of investigators have reported that human skin secretions suppress attraction by mosquitoes
or repel them.100–104 Maibach et al.103 observed that lipids from the elbow more effectively deterred
Aedes aegypti than did lipids from the scalp. They inferred from this result that deterrent chemicals
originate in the epidermis rather than in sebaceous glands, presumably because these glands are scarce or
absent in the elbow region and abundant on the scalp. Thompson and Brown,101 on the other hand,
suggested that volatile acids released from the esterified components of sebum decrease the attractive-
ness of human sweat to Aedes aegypti. Muller104 observed that Aedes aegypti was attracted to sweat from
                                           ¨
the axilla, but avoided sweat from the trunk.
   Skinner et al.105 used a dual-port olfactometer to compare the responses by Aedes aegypti to clean air
or air laden with the extracts of human hands and elbows. Fewer mosquitoes landed near entry ports
releasing volatiles from whole skin extracts and extract fractions obtained by thin-layer chromatography.
Several nonpolar bands deterred mosquitoes, including one containing hydrocarbons;106 the unsaturated
components were deemed active. Squalene and a number of straight-chain alkanes and alkenes were
presented to mosquitoes, but only 1-eicosene significantly deterred them. Neither this compound nor
others tested singly, however, were as effective as the composite fraction.
   Another fraction from human skin extracts contained free fatty acids,107 including saturated
compounds that weakly deterred mosquitoes and at least two fractions containing unsaturated acids
that were highly deterrent. Comparisons of mosquitoes’ responses to authentic saturated (C5–C18) and
unsaturated (C9–C20) compounds revealed a greater aversion to the latter. Further studies indicated three
deterrent components in human skin extracts, the most abundant of which was lactic acid.108 The other
two components were tentatively characterized as hydroxy carbonyl compounds.
   Olfactometry studies by Bosch et al.109 demonstrated that C1 to C18 n-aliphatic carboxylic acids
generally attract Aedes aegypti when combined with lactic acid, but this attractiveness was reduced with
nonanoic and undecanoic acids; similarly, undecanoic and tetradecanoic acids reduced the attractiveness
of lactic acid when combined with other carboxylic acids. Carboxylic acids generally attract Anopheles
gambiae, as well,110 but Smallengange et al.111 observed that a mixture of twelve C2–C16 carboxylic
acids presented in an olfactometer repelled these mosquitoes. These results may have been due to
impurities in the chemical samples.
   Lactic acid has been implicated in mosquitoes’ responses to humans as both an attractant and repellent,
depending upon its concentration and the responding species (see Steib et al.99 and Shirai et al.112).
Smith et al.113 applied lactic acid to a cotton stocking at 3.56 mg/cm.2 The stocking was placed on a
human subject’s arm and inserted into a cage containing Aedes aegypti. Substantially fewer mosquitoes
landed on the treated stocking than on an untreated one, and there were fewer bites. Subsequent


q 2006 by Taylor & Francis Group, LLC
60                                                     Insect Repellents: Principles, Methods, and Uses


quantitative analyses of lactic acid on human skin revealed that the amount of this compound that had
been applied to the stocking was 1,000–8,000 times greater than that normally present on human hands.
Smith et al. concluded that the amount of lactic acid usually present on human hands attracts Aedes
aegypti and probably never attains a concentration high enough to repel them.
   Shirai et al.112 tested intact and proboscis-amputated Aedes albopictus with serial dilutions of lactic
acid (1–10,000 ppm) applied to human forearms and hairless mice. Mosquitoes landed less frequently on
humans and mice treated with both high and low concentrations of lactic acid than on water-treated
controls. Intermediate concentrations (10–3,000 ppm) did not deter landing. Shirai et al. stated that the
range of lactic acid concentrations they presented to mosquitoes was within that observed in human
sweat, and that the minimum concentration of this compound found to deter mosquitoes was lower than
that typically present on human skin. Shirai et al. suggested that the amount of lactic acid emitted by
human sweat may reflect the circulating titers of this acid, and that mosquitoes shun hosts emitting large
amounts of this compound to avoid imbibing it in blood-meals.
   A field study in Sudan by Costantini et al.73 compared the attraction of mosquitoes to traps emitting
human scent or carbon dioxide in amounts comparable to those emitted by a human subject. The number
of Anopheles gambiae entering traps releasing human scent was twice that trapped with carbon dioxide
alone, thus indicating that volatiles other than carbon dioxide are attractive to them. However, twice as
many Anopheles pharoensis were captured in carbon dioxide-releasing traps as in the human-baited
traps. This result might reflect the presence of human-derived inhibitors or repellents of this mosquito.
Costantini et al., however, suggested that this may have been a spurious result, arising from variation in
mosquitoes’ responses at the detection threshold of carbon dioxide used in their study. However, similar
studies by Dekker and colleagues76,77 on mosquitoes from South Africa have demonstrated that several
zoophilic species avoid the scent of humans as nonhosts.
   A number of human-derived volatiles arise via microbial degradation of secretions from the apocrine
glands of the axilla. Costantini et al.114 investigated responses by strains of the mosquito Anopheles
gambiae originating from East and West Africa to the axillary components, (E)- and (Z)-3-methyl-2-
hexenoic and 7-octenoic acids, using EAG and wind-tunnel and field-trapping bioassays in Burkina Faso.
Both acids elicited EAG responses. In wind-tunnel experiments, fewer females entered chambers
releasing carbon dioxide and a combination of a (E/Z)-3-methyl-2-hexenoic acid isomeric mixture and
7-octenoic acid when these compounds were presented in a range of doses. Field tests demonstrated that
7-octenoic acid increased the catches of traps releasing carbon dioxide. However, when 7-octenoic acid
was presented with (E/Z)-3-methyl-2-hexenoic acid isomers, or when these isomers were presented
singly or combined, fewer mosquitoes were captured in otherwise attractive traps. Costantini et al.114
stated that these acids may repel Anopheles gambiae or mask the attractiveness of carbon dioxide or other
human scents. Alternatively, these investigators suggested that reduced captures with human-specific
acids reflect the activity of these compounds in arresting the upwind flight of mosquitoes arriving at a
scent source, a response that normally prevents them from overshooting their hosts. Thus, the inhibition
of particular behaviors in the host-seeking repertoire of mosquitoes may, under some circumstances,
facilitate host location.
   Humans exhibit individual variation in their attractiveness to mosquitoes (see Mukabana et al.115).
McKenzie116 tested Aedes aegypti in an olfactometer to skin substances from individual subjects
absorbed onto silicone membranes. The skin emanations from scent donors differed significantly in their
attractiveness to mosquitoes. Some subjects were designated as “repellent.” Interestingly, the presence of
highly attractive subjects appeared to render otherwise acceptable individuals less attractive.
   Bernier et al.117 also tested Aedes aegypti in an olfactometer to chemicals from humans differing in
mosquito attractiveness. An analysis of volatiles desorbed from glass beads handled by a less preferred
subject revealed a greater abundance of the following compounds: 2-nonene; nonane; methyl undecane;
pentacosane; decanoic acid; heptanal; 2,4-nonadienal; and nonanal. Olfactometry experiments demon-
strated that these aldehydes inhibit mosquitoes’ normal attraction to lactic acid (Chapter 4).
   Mukabana et al.115 investigated the involvement of volatiles from human breath in the ability of
Anopheles gambiae from Tanzania to distinguish between individual hosts. Mosquitoes were tested in


q 2006 by Taylor & Francis Group, LLC
Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods                         61




FIGURE 3.4 A nest of wasps (Polybia diguetana) from Costa Rica is handled by a male farmer whose hand was covered
with sweat from his underarm. (From A. M. Young, Biotropica, 10, 73, 1978. With permission.)

an olfactometer with the breath, body odor, and total body emanations (both breath and body odor) of
male subjects who differed in their attractiveness to mosquitoes. Breath was separated from body odors
by requiring subjects confined in tents to mouth-breathe through a one-way valve, which diverted their
exhalants away from other body effluents; air from an empty tent served as a control. More mosquitoes
were captured in traps emitting body odors, total body emanations, and control odors than human
breath, suggesting that breath contains deterrents. Moreover, when breath was removed from the scents
of the subjects, they no longer differed in their attractiveness to mosquitoes. Mukabana et al. concluded
that the body odors and total body emanations of humans have a kairomonal (attractive) effect on
Anopheles gambiae and that human breath has an allomonal (repellent) effect on them. These authors
suggested that the differential attractiveness of individual humans to mosquitoes is due to the effect of
breath volatiles.
   The use of chemicals by mammals to deter hymenopterans has rarely been considered (but see
Kingdon118 on the African ratel, Mellivora capensis). Young119 reported that the highly aggressive wasp
(Polybia diguetana) from Costa Rica failed to sting and “seemed drugged” when its nest was handled by
a male farmer who had coated his hand with his axillary secretions (Figure 3.4). The axillary odors of
other farmers in Costa Rica and the United States also reportedly mollify wasps. Young119 suggested that
human sweat generally deters wasps.




Anointing
Many birds, primarily passerines,120–133 and mammals, including insectivores,134,135 primates,4,5,136–143
rodents,144 carnivores,3,4,145–150 and ungulates,151 apply scent-laden materials to their integument. Birds
hold objects in their beak and streak them through their plumage, primarily the wing feathers. Some


q 2006 by Taylor & Francis Group, LLC
62                                                      Insect Repellents: Principles, Methods, and Uses


mammals, such as primates, rub materials into their fur with their forepaws. Others, like canids and
felids, roll directly onto scent sources. Ungulates may acquire the scent of leaves by thrashing their horns
in vegetation.
   Dolichoderine and formicine ants are used extensively in anointing by free-ranging birds.123,124,127 In
addition, birds rub themselves with other insects (see Whitaker123), millipedes (see Parkes et al.132),
gastropods,133 citrus121,130,131 and other fruits,124,133 onions,122 resin,120 and fresh124,129 and smoking
vegetation.123–125 Mammals also anoint with ants,138,143 citrus fruits,4,5,141,145 onions,127,142 milli-
pedes,139,140,142 and smoking vegetation,136 in addition to noncitrus fruits, 137 leaves;4,5,141,147
resins;149 toads;135 carrion (see Reiger146); and the feces,134,146 urine,134,146 and skin gland secretions
of heterospecifics.144,146,148
   The presence of insecticides and other deterrents in the materials used in feather- and fur-rubbing
points to these behaviors as mechanisms by which anti-arthropod compounds are acquired.7 The
increased incidence of anointing by animals confronted with heightened ectoparasite infestations also
accords with this hypothesis. Free-ranging neotropical capuchin monkeys (Cebus spp.), for example, rub
their fur with leaves, fruits or millipedes most frequently during the wet season, when nuisance
arthropods are abundant.5,140 Similarly, many North American birds rub themselves with ants during
molting periods that coincide with heavy rainfall;128 the timing of this activity also is consistent with
anti-microbial defense.152
   Formic acid is believed to be the main compound appropriated by birds and mammals that anoint with
formicine ants. In the 1940s, Dubinin observed that steppe pipits (Anthus godlewskii) in Russia rubbed
themselves with wood ants (Formica rufa) (summarized in Kelso and Nice126). Birds that rubbed
themselves with ants, and those not observed to do so, were examined for feather mites (Pterodectes
bilobatus). The anointing birds harbored 87 dead mites and 612 mites that crawled over their feathers,
171 of which died within 21 hours. Mites on nonanointing birds, on the other hand, remained attached to
feathers. Of 758 mites collected from these birds, only seven died within 21 hours. Observations of
hoopoes (Upupa epops) in Tadzhikistan also suggested that mites (Pterodectes cuculi) are induced to
crawl over feathers following bouts of anointing, but after 12 hours, only 1.7% of them died compared
to up to 1.2% of the controls. Kelso and Nice126 concluded that anointing with ants agitates ectoparasites,
if it does not kill them.
   Eichler153 placed lice that infested the feathers of domestic chickens (Gallus domesticus), principally
Eomenacanthus stramineus, into glass jars and sprayed them with a 50% solution of formic acid; controls
were sprayed with water. Not surprisingly, all acid-sprayed lice died within a few minutes, whereas those
treated with water survived. Wilson and Hillgarth154 also observed that formic acid vapor kills lice and
feather mites. Field observations of North American passerine birds128 and studies under semi-natural
conditions,155,156 however, have failed to indicate that avian ectoparasites are affected by ant-derived
fluids expressed during anointing.
   Millipedes, primarily those that secrete noxious benzoquinones, also are used for anointing by
birds,132 capuchin (Cebus spp.)140 and owl monkeys (Aotus spp.),142 and lemurs.139 The benzoquinones,
which millipedes release when disturbed, elicit fur-rubbing behaviors in capuchin67 and owl monkeys.142
Valderrama et al.140 observed that wedge-capped capuchin monkeys (Cebus olivaceous) in Venezuela
rub themselves with the millipede, Orthoporus dorsovittatus, which secretes 2-methyl-1,4-benzoquinone
(toluquinone) and 2-methoxy-3-methyl-1,4-benzoquinone (MMB). These investigators hypothesized
that benzoquinones acquired by monkeys repel mosquitoes. To test the plausibility of this hypothesis,
Weldon et al.67 presented toluquinone and MMB, individually and in a 1:1 mixture, to the mosquito
Aedes aegypti on silicone feeding membranes placed over wells of human blood, a highly preferred
food. Fewer mosquitoes landed on or fed through benzoquinone-treated membranes than did on solvent-
treated membranes. Mosquitoes also exhibited higher flying scores when exposed to these compounds,
a possible indication that they were repelled.
   Carroll et al.158 tested lone star ticks (Amblyomma americanum) for responses to 1,4-benzoquinone,
toluquinone, and MMB to examine the possible effects of these compounds in anointing by birds and
mammals. Ticks typically spend hours or days wandering or feeding on their hosts, and thus are exposed


q 2006 by Taylor & Francis Group, LLC
Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods                                63


over long periods to chemicals on their hosts’ integument. Carroll et al. observed that at low
concentrations benzoquinones impaired ticks’ ability to climb or right themselves, behavioral deficits
that could affect their ability to access hosts. MMB also was mildly repellent. Ticks died when exposed to
higher concentrations of these compounds. Thus, studies on mosquitoes and ticks show that the topical
appropriation of benzoquinones may deter ectoparasites. However, key questions remain on the amounts
of benzoquinones typically available to and appropriated by free-ranging animals.
   Citrus fruits are used in anointing by birds,121,130,131 monkeys (Cebus spp.),4,5 and canids.145 After
observing a grackle (Quiscalus quiscula) rub its plumage with a slice of lime (Citrus aurantifolia), Clayton
and Vernon130 tested the effects of volatiles from this fruit on feather lice (Columbicola columbae) from the
rock dove. Lice confined with lime slices experienced higher mortality than did control lice with water.
Tests of extracts from different parts of limes revealed that the biocidal compounds occur in the peel.
   The concentrated peel oils of citrus fruits are known to repel or kill a variety of insects,159 including
mosquitoes, such as Aedes aegypti,160 Anopheles stephensi,161 and Culex pipiens.162 Weldon163
examined the effects of fresh unconcentrated lemon (Citrus limon) peel extracts on Aedes aegypti
using the membrane feeding system described above. Shallow incisions were made in the peel of
organically grown lemons and the cut areas were lightly pressed against silicone membranes.
The membranes were placed over wells containing a green-dyed sugar solution, and mosquitoes were
allowed to land on and feed through them. Control membranes were treated with extracts of sliced kale
(Brassica oleracea) leaves or were left untreated (blank). The results of this experiment showed that
mosquitoes landed and fed less on lemon-treated membranes than on either of the control membranes
(Figure 3.5). Mosquitoes also flew more in chambers exposed to the lemon extract, suggesting that they
were repelled by it.
   The peels of Citrus spp. are rich in volatiles, such as limonene, linalool, and citral, and nonvolatiles,
such as coumarins and furanocoumarins.159,164 These compounds are known to repel165 or kill166
                                  1.0
                                  0.8
                                  0.6
                     Proportion
                                  0.4
                                  0.2
                                  0.0




                                        lemon                 kale                          blank

FIGURE 3.5 Mean percentages and 95% confidence intervals of Aedes aegypti landing (!), flying (:), and feeding (C)
on silicone membranes (9.6 cm2) covering wells containing 10% sucrose solution and treated with a sliced lemon (Citrus
limon) peel or kale (Brassica oleracea) leaves or left untreated (blank). Sixty mosquitoes, five per trial, were observed
for 5 min with each treatment. Values for landing and flying were backtransformed, as described in Weldon et al.67
The values of all measures for lemon differ significantly from those for kale and blank conditions. (P. J. Weldon,
Unpublished.)


q 2006 by Taylor & Francis Group, LLC
64                                                      Insect Repellents: Principles, Methods, and Uses


a variety of arthropods, including vertebrate ectoparasites such as fleas,167,168 lice,169 and mites.170
Studies are needed to assess whether birds and mammals are protected from nuisance arthropods by
topically appropriating these compounds.
   The propensity of domestic cats to rub and roll on the leaves of catnip (Nepeta cataria) is well known
(see Tucker and Tucker147). This response is elicited by nepetalactone, a cyclopentanoid monoterpene
that protects Nepeta spp. from phytophagous insects.171 Nepetalactone also repels other types of insects,
including mosquitoes. Petersen172 tested Aedes aegypti in a static-air choice test apparatus with catnip
essential oil and Z,E-nepetalactone, both of which were avoided. Similar results have been reported in
other studies with catnip essential oil173,174 and nepetalactone.174 E,Z- and Z,E-nepetalactone, presented
singly or in a 1:1 mixture, reduced feeding by Aedes aegypti in tests with both artificial feeding
membrane and human subjects.174 The deterrent properties of nepetalactone demonstrated in these
studies support an anti-ectoparasite function for the catnip (anointing) response, thus providing a
plausible alternative, if not more compelling, interpretation of this felid behavior than that of displaced
socio-sexual or hallucinogenic responses (see Tucker and Tucker147), or as a nonfunctional activity.175
   Kodiak bears (Ursus arctos) chew “bear root” (Ligusticum wallichii) and rub themselves with a
mixture of the root juice and saliva.150 Passreiter et al.176 observed that the essential oil of Ligusticum
mutellina from central and southern Europe kills third instar armyworms (Pseudaletia unipunctata) when
topically applied to them, and that it contains phenylpropanoid insecticides, such as ligustilide. The
effects of chemicals from Ligusticum spp. on the ectoparasites of ursids should be examined.




Fumigation
In fumigation, animals are exposed to volatile (usually plant-derived) chemicals that deter microbial
pathogens and/or ectoparasites. Fumigation has been most extensively studied in birds that add fresh
aromatic leaves to their nests, usually in amounts small relative to the structurally supportive nest matrix.
Phytochemicals involved in nest fumigation are thought to repel ectoparasites or disrupt their
reproduction or development (see Clark2; cf. Fauth et al.177 and Gwinner et al.178).
   Surveys of falconiforms and passerines reveal that birds that re-use nests,179,180 or build nests in
enclosed spaces180—situations where the risks of parasite or pathogen loads are high—are more likely to
use fresh nest vegetation than are those that infrequently re-use nests or use open, cup-like nests.
Comparisons of the plants locally available with those incorporated into nests reveal that birds select
particular aromatic species for fumigation. House sparrows (Passer domesticus) in India, for example,
use the leaves of the margosa (neem) tree (Azadirachta indica), ignoring many other available plants.181
The leaves of this tree contain numerous biocidal compounds, including b-sitosterol, a repellent and
oviposition inhibitor of mites.
   European starlings (Sturnus vulgaris) in the eastern United States select the following plants for their
nests: agrimony (Agrimonia parviflora), wild carrot (Daucus carota), fleabane (Erigeron philadelphicus),
yarrow (Achillea millefolium), purple dead-nettle (Lamium purpureum), and goldenrod (Solidago
rugosa).180,182 Chemical analyses of these plants reveal mono- and sesquiterpenes, including carene,
cymene, limonene, myrcene, ocimene, a- and b-pinene, a-phellandrene, sabinene, a-terpineol, and
a-terpinoline, many of which are known to repel or kill insects and other arthropods.165,166 Goldenrod
contains 2-bornyl acetate and farnesol, which are analogs of juvenile hormone. These compounds suppress
molting in arthropods, and thus may interfere with the growth and reproduction of ectoparasites.
   Clark and Mason180 tested the effects of nest fumigation by starlings on the hatching success of lice
(Menacanthus sp.) and on the survival of adult northern fowl mites (Ornithonyssus sylviarum). Volatiles
from plants preferred by starlings more effectively retarded the hatching of lice than did volatiles from a
random sample of nonpreferred plants, but mite survival was not affected. Nests devoid of wild carrots,
however, possessed larger mite populations than those containing this plant.180 In the laboratory, the


q 2006 by Taylor & Francis Group, LLC
Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods                     65


emergence of feeding mite instars from nests containing wild carrot and fleabane, both preferred by
starlings, was suppressed, whereas garlic mustard (Alliaria officinalis), a commonly available
nonpreferred plant, had no effect.
   Behavioral and electrophysiological studies demonstrate that birds use olfaction to identify fumigation
plants.183–185 Clark and Smeraski185 showed that the olfactory sensitivity of European starlings varies
seasonally, peaking during spring, when birds are breeding and building nests, and waning when birds
enter a nonbreeding condition. Seasonal differences were demonstrated in birds’ responses to a variety of
olfactory stimuli, including both preferred (dead-nettle and wild carrot) and nonpreferred (garlic
mustard) nest plants. The seasonal expression of heightened olfactory sensitivity in starlings may
allow them to identify fumigating plants. Clark and Mason182 hypothesized that birds select plants based
upon of their aromaticity and chemical complexity, rather than upon the presence of
particular compounds.
   Cavity-nesting blue tits (Parus caeruleus) on the island of Corsica build nests that contain aromatic
plants, including the leaves of Achillea ligustica and Lavandula stoechas.184 Birds regularly replace
these leaves in their nests between the onset of egg laying and chick fledging. The postulated antiseptic
and insect repellent function of leaf placement was corroborated by the presence of leaf volatiles such as
camphor, eucalyptol, limonene, linalool, myrcene, piperotenone, and terpin-4-ol, which are bacterio-
static and/or insect deterrents.165,166
   Nestlings lack protective feathers and are thus vulnerable to hematophagous insects. Lafuma et al.186
investigated whether aromatic plants in the nests of blue tits repel the mosquito, Culex pipiens, a vector of
avian malaria. Mosquitoes were placed into boxes containing leaves and allowed to escape through tubes.
The mixture of plants repelled them, but only the leaves of Achillea or Lavandula did so alone.
Mosquitoes allowed to choose between domestic chicks confined in boxes with or without aromatic
leaves fed less on chicks with the leaves of Achillea or Lavandula. Lafuma et al. suggested that leaf
volatiles repel mosquitoes or mask the odors of their hosts.
   Some mammals construct nests that include aromatic leaves or bark. Hemmes et al.187 observed that
the dusky-footed wood rat (Neotoma fuscipes) in the western United States deposits fresh foliage,
including the leaves of California bay (Umbellularia californica), oak (Quercus spp.), and toyon
(Heteromeles arbutifolia), in or near their stickhouses. Bay leaves, in contrast to other foliage, were
deposited by wood rats more often near nests than away from them, and were nibbled, leaving shallow,
sporadic lacerations along leaf margins. This unique nibbling pattern suggests that bay leaves are not
used for food, but for the release of volatiles. When larvae of the cat flea (Ctenocephalides felis) were
incubated with intact or torn leaves from bay, oak, or toyon, or without foliage, their survival was more
severely reduced with torn bay leaves (27%) than with the other conditions (88–94%). This study
suggests that volatiles from bay leaves reduce nest-borne ectoparasites of Neotoma fuscipes. Studies are
needed of the effects of volatiles from mammalian nest materials on other ectoparasites.
   Fumigation may be the mechanism by which animals are tolerated entering or nesting close to social
insect colonies. The Tui parakeet (Brotogeris sanctithomae), the cobalt-winged parakeet (Brotogeris
cyanoptera), and the black-tailed trogon (Trogon melanurus) in the Peruvian Amazon, for example, nest
in arboreal termitaria inhabited by both termites and an aggressive biting ant, Dolichoderus sp.188
Brightsmith188 suggested that birds acquire the strong smell of Dolichoderus, thereby chemically
camouflaging their nest and avoiding attack. Similarly, Janzen190 suggested that the nests of birds
built in swollen-thorn acacias of Central America and northern Colombia acquire the scent of ants,
probably trail substances, and thus permit birds to avoid being evicted by ants.




Discussion
A taxonomically and ecologically diverse array of tetrapods produces or appropriates arthropod
deterrents, borne principally on the integument. The skin of vertebrates has long been known as


q 2006 by Taylor & Francis Group, LLC
66                                                      Insect Repellents: Principles, Methods, and Uses


a source of novel natural products featuring diverse chemical classes and structures, e.g., unusual
branching patterns and unsaturation sites, that contrast with those of compounds from internal tissues.190
The unique chemistry of the skin has variously been attributed to the production of antibiotics,
pheromones, or water retardants. We propose that the chemical diversity of the vertebrate integument,
and interspecific variation in skin chemical profiles (see Weldon191), are related, in part, to the
elaboration of arthropod deterrents. Indeed, some vertebrate integumentary glands may function to
secrete compounds that combat arthropod pests. The deterrence of biting flies by sebum,56,63,71,101 for
example, suggests an important but underappreciated role for mammalian sebaceous glands, the function
of which has long been unclear.192
   Many compounds identified from the skin of tetrapods are known to adversely affect arthropods. The
odoriferous pelage of the reticulated giraffe (Giraffa camelopardalis reticulata), for example, emits
indole and skatole,193 which deter some mosquitoes (Aedes spp.).194,195 Another giraffe pelage
compound, p-cresol, is a pheromone for many ticks,196 but in high concentrations (comparable to
those arising from giraffes) it repels them.193 Tests of chemicals known to occur on the skin of tetrapods
have revealed additional compounds that deter ectoparasites.51,52,61,109,114 Even some free fatty acids,
which are ubiquitous tetrapod skin products, are insecticidal,166 and thus may discourage
some ectoparasites.
   Anointing and fumigation enable tetrapods to acquire chemicals from a variety of exogenous sources.
Key questions, however, remain regarding the defensive significance of topically appropriated
chemicals. Although laboratory studies have demonstrated that compounds present in anointing and
fumigation materials deter some ectoparasites, studies are needed on the quantities of chemicals that are
topically acquired and the extent to which they deter arthropods under natural conditions. Formic acid,
for example, kills feather lice in laboratory tests,153,154 but neither controlled aviary experiments155,156
nor observations of free-ranging North American birds128 support the hypothesis that ant secretions
topically applied to the plumage reduce louse loads. These results have prompted some authors to dismiss
anointing as a mechanism of ectoparasite control.128 However, taxon-specific ectoparasites, such as
feather lice, may be more tolerant of their host’s defensive chemicals if they and their hosts have
coevolved.34,46 Hypotheses on the defensive function of chemicals should be tested with a spectrum of
actual and potential ectoparasites to consider possible species differences in chemical tolerance, host
dependence, and other variables.
   An arsenal of chemicals, including fumigant repellents, contact irritants, neurotoxins, masking agents,
and hormonal antagonists/mimics, are thought to be deployed against arthropods by amphibians, reptiles,
and mammals. The nature of the responses elicited by these compounds depends upon their
concentrations,56,58,64,112,113 the duration of exposure arthropods receive to them,24,46,167,168 synergism
of components of a blend,58,64 and the sex,58,93–96,98 rearing conditions,63,64 and hunger state of the
arthropods,20,56,96,98 to name a few variables. Future studies examining species and populational
differences in the ectoparasites’ responses to vertebrate-derived chemicals, including deterrents, may
provide useful information for integrated pest management, in addition to elucidating microevolutionary
variation in host-chemical preferences.
   Pickett and Woodcock83 suggested that hematophagous insects avoid individual cattle because they
emit volatile chemicals that denote their immunological competence. Biting flies also avoid the scents of
nonhost species,73,74,76,77,93–98 and it may be commonplace for foraging arthropods to do so.197 Aside
from circumventing immunological defenses, ectoparasites should be selected to avoid chemicals
denoting organisms that are dangerous, noxious or otherwise unprofitable as hosts. Conversely, would-
be hosts should be selected to signal their status as an inappropriate resource, thus discouraging encounters
with potentially injurious ectoparasites. Eisner and Grant198 suggested that “olfactory aposematism,”
where chemicals advertise an organism’s distasteful or other undesirable qualities to predators, is
commonplace. Chemical aposematism vis-a-vis ectoparasites may also be widespread.
   Amphibians, reptiles, and mammals exhibit a variety of behavioral, physiological and anatomical
defenses against nuisance arthropods. We have focused on their potential use of chemical defenses and
described how these agents may reduce vulnerability to predators, ectoparasites, and disease vectors in an


q 2006 by Taylor & Francis Group, LLC
Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods                    67


attempt to frame questions on this expanding area of interest. What roles do chemicals play in the
symbiotic and other intimate associations between tetrapods and social insects?21–23,30,188,189 Can the
reduced intensity of ectoparasitic infestations44,45,55–57,80,82 and the lowered incidence of vector-borne
diseases among free-ranging animals49 point to organisms profitably investigated for arthropod
deterrents, or be used to assess the efficacy of these defenses in nature?48 How have tetrapods converged
with plants and invertebrates in the chemicals they use to counter arthropods?50,51,165 Collaborations
between chemists and biologists—a hallmark of the discipline of chemical ecology—are needed
to address these questions, ideally, with the combined participation of arthropod and
vertebrate specialists.


Acknowledgments
W. J. Andrews; U. R. Bernier; M. S. Blum; J. F. Butler; E. D. Brodie, Jr.; R. L. Chazdon; D. H. Clayton;
P. Coon; M. Debboun; H. D. Douglas, III; T. Eisner; J. A. Endler; P. A. Evans; T. Falotico; R. C.
Fleischer; H. W. Greene; A. Hassanali; R. Heyer; P. Holm; M. B. Isman; A. F. Jahn; S. Krane; M.
                                                                ¨
Kramer, N. Kreiter; P. Manly; J. Millar; J. A. Pickett; M.-O. Rodel; B. P. C. Smith; T. F. Spande; D.
Strickman; J.-M. Thiollay; R. K. Vander Meer; G. B. White; and A. M. Young provided materials,
information, or comments on the manuscript. The Association for Tropical Biology and Conservation
and the Zoological Society of London granted permission to reproduce figures. J. Bogard (American
Council on Education, Washington, D. C.); A. Hutchinson; P. Lasker; M. Rosen; D. T. Steere, Jr.
(Smithsonian Libraries, Washington, D. C.); M. Lothers; W. Olson; W. Thompson; C. Toefield Keen; F.
Tyler (National Agricultural Library, Beltsville, Maryland); C. Twose (William H. Welch Medical
Library, Johns Hopkins University, Baltimore, Maryalnd); and H. Brooks (Gorgas Memorial Library,
Walter Reed Army Institute of Research, Silver Spring, Maryland) supplied references. L. Morel and M.
Olshausen provided translations. Preliminary studies of mosquitoes by P. J. W. were supported by the
Chemicals Affecting Insect Behavior Laboratory, USDA, Beltsville, Maryland. J. P. Benante; R.
Coleman, W. Dheranetra, M. Dowler, S. Gordon, L. Jones, N. McLean-Cooper, E. Rowton, and J.
Williams facilitated mosquito studies in the Department of Entomology, Walter Reed Army Institute of
Research, Silver Spring, Maryland. D. L. Armstrong (Henry Doorly Zoo, Omaha, Nebraska), J. Chatfield
(Gladys Porter Zoo, Brownsville, Texas), and J. Zeliff (Silver Springs Animal Park, Silver Springs,
Florida) provided samples from animals in their care. This chapter was written while P. J. W. was
supported by Bedoukian Research Inc., Danbury, Connecticut.


References
      1. C. Darwin, The Origin of Species by Means of Natural Selection, or the Preservation of Favoured
         Races in the Struggle for Life, London: John Murray, 1859.
      2. L. Clark, The nest protection hypothesis: The adaptive use of plant secondary compounds by European
         starlings, in Bird–Parasite Interactions, J. E. Loye and M. Zuk (Eds.), Oxford: Oxford University
         Press, 1991, p. 205
      3. D. H. Clayton and N. D. Wolfe, The adaptive significance of self-medication, Trends in Ecology and
         Evolution, 8, 60, 1993.
      4. J. P. Berry, M. A. McFerren, and E. Rodriguez, Zoopharmacognosy: A “biorational” strategy for
         phytochemical prospecting, in Phytochemicals and Health, D. L. Gustine and H. E. Flores (Eds.),
         Rockville, MD: American Society of Plant Physiologists, 1995, p. 165.
      5. M. Baker, Fur rubbing: Use of medicinal plants by capuchin monkeys (Cebus capucinus), American
         Journal of Primatology, 38, 263, 1996.
      6. B. L. Hart, Behavioural defence, in Host–Parasite Evolution: General Principles and Avian Models,
         D. H. Clayton and J. Moore (Eds.), Oxford: Oxford University Press, 1997, p. 59.
      7. P. J. Weldon, Defensive anointing: Extended chemical phenotype and unorthodox ecology,
         Chemoecology, 14, 1, 2004.


q 2006 by Taylor & Francis Group, LLC
68                                                           Insect Repellents: Principles, Methods, and Uses


      8. V. G. Dethier, L. Barton Browne, and C. N. Smith, The designation of chemicals in terms of the
         responses they elicit from insects, Journal of Economic Entomology, 53, 134, 1960.
      9. L. Barton Browne, Host-related responses and their suppression: Some behavioral considerations, in
         Chemical Control of Insect Behavior: Theory and Application, H. H. Shorey and J. J. McKelvey, Jr.,
         (Eds.), New York: John Wiley, 1977, p. 117.
     10. M. R. Berenbaum, The chemistry of defense: Theory and practice, Proceedings of the National
         Academy of Sciences USA, 92, 2, 1995.
     11. V. Erspamer, Bioactive secretions of the amphibian integument, in Amphibian Biology,
         H. Heatwole and G. T. Barthalmus (Eds.), Vol. 1, Chipping Norton, NSW: Surrey Beatty &
         Sons, 1994, p. 178.
     12. G. Fritz, A. S. Rand, and C. W. de Pamphilis, The aposematically colored frog, Dendrobates
         pumilio, is distasteful to the large, predatory ant, Paraponera clavata, Biotropica, 13, 158, 1981.
     13. W. A. Szelistowski, Unpalatability of the poison arrow frog Dendrobates pumilio to the ctenid
         spider Cupiennius coccineus, Biotropica, 17, 345, 1985.
     14. E. D. Brodie, Jr., D. R. Formanowicz, Jr., and E. D. Brodie III, The development of noxiousness
         of Bufo americanus tadpoles to aquatic insect predators, Herpetologica, 34, 302, 1978.
     15. D. R. Formanowicz, Jr., and E. D. Brodie, Jr., Relative palatabilities of members of a larval
         amphibian community, Copeia, 1982(1), 91, 1982.
     16. J. A. Peterson and A. R. Blaustein, Relative palatabilities of anuran larvae to natural aquatic insect
         predators, Copeia, 1992(2), 577, 1992.
     17. E. D. Brodie, Jr., and D. R. Formanowicz, Jr., Larvae of the predaceous diving beetle Dytiscus
         verticalis acquire an avoidance response to skin secretions of the newt Notophthalmus viridescens,
         Herpetologica, 37, 172, 1981.
     18. E. D. Brodie, Jr., R. T. Nowak, and W. R. Harvey, The effectiveness of antipredator secretions and
         behavior of selected salamanders against shrews, Copeia, 1979(2), 270, 1979.
     19. F. H. Pough, Leech-repellent property of eastern red-spotted newts, Notophthalmus viridescens,
         Science, 174, 1144, 1971.
     20. K. S. Hileman, E. D. Brodie, Jr., and D. R. Formanowicz, Jr., Avoidance of unpalatable prey by
         predaceous diving beetle larvae: The role of hunger level and experience (Coleoptera: Dytiscidae),
         Journal of Insect Behavior, 8, 241, 1995.
                    ¨
     21. M.-O. Rodel and U. Braun, Associations between anurans and ants in a West African savanna
         (Anura: Microhylidae, Hyperoliidae, and Hymenoptera: Formicidae), Biotropica, 31, 178, 1999.
                                                              ´
     22. A. Dejean and J.-L. Amiet, Un cas de myrmecophilie inattendu: La cohabitation de l’anoure
         Kassina senegalensis avec la fourmi Megaponera foetans, Alytes, 10, 31, 1992.
                   ¨                 ¨
     23. A. Schluter and J. Regos, The tadpole of Lithodytes lineatus—with note on the frogs [sic]
         resistance to leaf-cutting ants (Amphibia: Leptodactylidae), Stuttgarter Beitrage zur Naturkunde,
         Series A, 536, 1, 1996.
     24. C. R. Williams, J. F. Wallman, and M. J. Tyler, Toxicity of green tree frog (Litoria caerula) skin
         secretion to the blowflies Calliphora stygia (Fabricus) and Lucilia cuprina (Wiedemann) (Diptera:
         Calliphoridae), Australian Journal of Entomology, 37, 85, 1998.
     25. B. P. C. Smith, C. R. Williams, M. J. Tyler, and B. D. Williams, A survey offrog odorous secretions, their
         possible functions and phylogenetic significance, Applied Herpetology, 2, 47, 2004.
     26. C. R. Williams, B. P. C. Smith, S. M. Best, and M. J. Tyler, Mosquito repellents in frog skin, Biology
         Letters, 2, 242, 2006.
     27. B. P. C. Smith, Y. Hayasaka, M. J. Tyler, and B. D. Williams, ß-Caryophyllene in the skin secretion of
         the Australian green tree frog, Litoria caerulea: An investigation of dietary sources, Australian Journal
         of Zoology, 52, 521, 2004.
     28. B. P. Smith, M. J. Tyler, C. R. Williams, C. Alcaraz Zini, J. Pawliszyn, B. D. Williams, Y. Hayasaka, and
         J.M. Griffiths, First demonstration of the function of frog odour as a defensive mechanism, Unpublished
         manuscript.
     29. J. F. Watkins II, F. R. Gehlbach, and R. S. Baldridge, Ability of the blind snake, Leptotyphlops dulcis, to
         follow pheromone trails of army ants, Neivamyrmex nigrescens and N. opacithorax, The Southwestern
         Naturalist, 12, 455, 1967.


q 2006 by Taylor & Francis Group, LLC
Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods                          69


    30. F. R. Gehlbach, J. F. Watkins II, and H. W. Reno, Blind snake defensive behavior elicited by ant attacks,
        BioScience, 18, 784, 1968.
    31. M. S. Blum, J. B. Byrd, J. R. Travis, J. F. Watkins II, and F. R. Gehlbach, Chemistry of the cloacal sac
        secretion of the blind snake Leptotyphlops dulcis, Comparative Biochemistry and Physiology, 38B, 103,
        1971.
    32. Holm, P, Personal communication, 2002.
    33. H. B. Cott, The edibility of birds: Illustrated by five years’ experiments and observations
        (1941–1946) on the food preferences of the hornet, cat and man; and considered with special
        reference to the theories of adaptive coloration, Proceedings of the Zoological Society of London,
        116, 371, 1947.
    34. J. P. Dumbacher and S. Pruett-Jones, Avian chemical defense, in Current Ornithology, V. Nolan,
        Jr. and E. D. Ketterson (Eds.), Vol. 13, New York: Plenum, 1996, p. 137.
    35. J.-M. Thiollay, Foraging, home range use and social behaviour of a group-living rainforest raptor,
        the red-throated caracara Daptrius americanus, Ibis, 133, 382, 1991.
    36. J.-M. Thiollay, Personal communication, 1996.
    37. G. H. Lowery, Jr. and W. W. Dalquest, Birds from the state of Veracruz, Mexico, University of
        Kansas Publications, Museum of Natural History, 3, 531, 1951.
    38. P. J. Weldon and J. H. Rappole, A survey of birds odorous or unpalatable to humans: Possible
        indications of chemical defense, Journal of Chemical Ecology, 23, 2609, 1997.
    39. J. P. Dumbacher, B. M. Beehler, T. F. Spande, H. M. Garraffo, and J. W. Daly, Homo-
        batrachotoxin in the genus Pitohui: Chemical defense in birds?, Science, 258, 799, 1992.
    40. J. P. Dumbacher, T. F. Spande, and J. W. Daly, Batrachotoxin alkaloids from passerine birds: A
        second toxic bird genus (Ifrita kowaldi) from New Guinea, Proceedings of the National Academy
        of Science USA, 97, 12970, 2000.
    41. J. P. Dumbacher, A. Wako, S. R. Derrickson, A. Samuelson, T. F. Spande, and J. W. Daly,
        Melyrid beetles (Choresine): A putative source for batrachotoxin alkaloids found in poison-dart
        frogs and toxic passerine birds, Proceedings of the National Academy of Science USA, 101,
        15857, 2004.
    42. P. Escalante, V. Emanuel, and J. W. Daly, Alkaloids in extracts of feathers of the red warbler,
        Journal of Ornithology, 135, 410, 1994.
    43. B. O. Poulsen, Poison in Pitohui birds: Against predators or ectoparasites?, Emu, 94, 128, 1994.
    44. K. N. Mouritsen and J. Madsen, Toxic birds: Defence against parasites?, Oikos, 69, 357, 1994.
    45. M. Pruett-Jones and S. Pruett-Jones, Analysis and ecological correlates of tick burdens in a New
        Guinea avifauna, in Bird–Parasite Interactions, J. E. Loye and M. Zuk (Eds.), Oxford: Oxford
        University Press, 1991, p. 154.
    46. J. P. Dumbacher, Evolution of toxicity in pitohuis, I. Effects of homobatrachotoxin on chewing
        lice (order Phthiraptera), Auk, 116, 957, 1999.
    47. E. H. Burtt, Jr., Think small, Auk, 116, 878, 1999.
    48. J. S. Beadell, E. Gering, J. Austin, J. P. Dumbacher, M. A. Pierce, T. K. Pratt, C. T. Atkinson,
        and R. C. Fleischer, Prevalence and differential host-specificity of two avian blood parasite
        genera in the Australo-Papuan region, Molecular Ecology, 13, 3829, 2004.
    49. H. L. Jones, Hematozoa from montane forest birds in Papua New Guinea, Journal of Wildlife
        Diseases, 21, 7, 1985.
    50. H. D. Douglas III, J. E. Co, T. H. Jones, and W. E. Connor, Heteropteran chemical repellents
        identified in the citrus odor of a seabird (crested auklet: Aethia cristatella): Evolutionary
        convergence in chemical ecology, Naturwissenschaften, 88, 330, 2001.
    51. H. D. Douglas III, J. E. Co, T. H. Jones, and W. E. Connor, Interspecific differences in Aethia
        spp. auklet odorants and evidence for chemical defense against ectoparasites, Journal of Chemical
        Ecology, 30, 1921, 2004.
    52. H. D. Douglas III, J. E. Co, T. H. Jones, W. E. Connor, and J. F. Day, Chemical odorant of
        colonial seabird repels mosquitoes, Journal of Medical Entomology, 42, 647, 2005.
    53. H. D. Douglas III, J. R. Malenke, and D. H. Clayton, Is the citrus-like plumage odorant of crested auklets
        (Aethia cristatella) a defense against lice?, Journal of Ornithology, 146, 111, 2005.


q 2006 by Taylor & Francis Group, LLC
70                                                         Insect Repellents: Principles, Methods, and Uses


     54. B. R. Moyer, A. N. Rock, and D. H. Clayton, Experimental test of the importance of preen oil in rock
         doves (Columba livia), Auk, 120, 490, 2003.
     55. R. Galun, Responses of blood-sucking arthropods to vertebrate hosts, in Chemical Control of Insect
         Behavior: Theory and Application, H. H. Shorey and J. J. McKelvey, Jr., (Eds.), New York: Wiley, 1977,
         p. 103.
     56. N. K. Gikonyo, A. Hassanali, P. G. N. Njagi, and R. K. Saini, Behaviour of Glossina morsitans morsitans
         Westwood (Diptera: Glossinidae) on waterbuck Kobus defassa Ruppel and feeding membranes smeared
         with waterbuck sebum indicates the presence of allomones, Acta Tropica, 77, 295, 2000.
     57. T. A. M. Nash, Africa’s Bane: The Tsetse Fly, London: Collins, 1969.
     58. G. A. Vale, D. R. Hall, and A. J. E. Gough, The olfactory responses of tsetse flies, Glossina spp.
         (Diptera: Glossinidae), to phenols and urine in the field, Bulletin of Entomological Research, 78,
         293, 1988.
     59. S. J. Torr, T. N. C. Mangwiro, and D. R. Hall, Responses of Glossina pallidipes (Diptera: Glossinidae)
         to synthetic repellents in the field, Bulletin of Entomological Research, 86, 609, 1996.
     60. L. C. Madubunyi, A. Hassanali, W. Ouma, D. Nyarango, and J. Kabii, Chemoecological role of
         mammalian urine in host location by tsetse, Glossina spp. (Diptera; Glossinidae), Journal of
         Chemical Ecology, 22, 1187, 1996.
     61. G. A. Vale, Field studies of the responses of tsetse flies (Glossinidae) and other Diptera to carbon
         dioxide, acetone and other chemicals, Bulletin of Entomological Research, 70, 563, 1980.
     62. W. F. Wood, A. Walsh, J. Seyjagat, and P. J. Weldon, Volatile compounds in shoulder gland
         secretions of male flying foxes, genus Pteropus (Pteropodidae, Chiroptera), Zeitschrift fur           ¨
         Naturforschung C, 60, 779, 2005.
     63. N. K. Gikonyo, A. Hassanali, P. G. N. Njagi, P. M. Gitu, and J. O. Midiwo, Odor composition of
         preferred (buffalo and ox) and nonpreferred (waterbuck) hosts of some savanna tsetse flies,
         Journal of Chemical Ecology, 28, 969, 2002.
     64. N. K. Gikonyo, A. Hassanali, P. G. N. Njagi, and R. K. Saini, Responses of Glossina morsitans
         morsitans to blends of electroantennographically active compounds in the odors of its preferred
         (buffalo and ox) and nonpreferred (waterbuck) hosts, Journal of Chemical Ecology, 29, 2331,
         2003.
     65. M. Bett, R. K. Saini, and A. Hassanali, In preparation.
     66. P. J. Weldon, Unpublished.
     67. P. J. Weldon, J. R. Aldrich, J. A. Klun, J. E. Oliver, and M. Debboun, Benzoquinones from
         millipedes deter mosquitoes and elicit self-anointing in capuchin monkeys (Cebus spp.),
         Naturwissenschaften, 90, 301, 2003.
     68. J. E. Oliver, P. J. Weldon, K. S. Petersen, W. F. Schmidt, and M. A. Debboun, Mosquito-
         deterring acid from pelage of gaur, Bos gaurus, Abstracts of Papers, 225th National Meeting of
         the American Chemical Society, New Orleans, LA, March 23–27. Washington, DC: American
         Chemical Society, 2003.
     69. H. Ishii, S. Krane, Y. Itagaki, N. Berova, K. Nakanishi, and P. J. Weldon, Absolute configuration
         of a hydroxyfuranoid acid from the pelage of the genus Bos, 18-(6S,9R,10R)-bovidic acid,
         Journal of Natural Products, 67, 1426, 2004.
     70. A. K. Bhattacharjee, W. Dheranetra, D. A. Nichols, and R. K. Gupta, 3D Pharmacophore model
         for insect repellent activity and discovery of new repellent candidates, QSAR & Combinatorial
         Science, 24, 593, 2005.
     71. P. A. Evans, W. J. Andrews, and P. J. Weldon, Unpublished.
     72. F. Q. Alali, X.-X. Liu, and J. L. McLaughlin, Annonaceous acetogenins: Recent progress, Journal
         of Natural Products, 62, 504, 1999.
     73. C. Costantini, G. Gibson, N’F. Sagnon, A. della Torre, J. Brady, and M. Coluzzi, Mosquito
         responses to carbon dioxide in a West African Sudan savanna village, Medical & Veterinary
         Entomology, 10, 220, 1996.
     74. C. Costantini, N’F. Sagnon, A. della Torre, M. Diallo, J. Brady, G. Gibson, and M. Coluzzi,
         Odor-mediated host preferences of West African mosquitoes, with particular reference to malaria
         vectors, American Journal of Tropical Medicine and Hygiene, 58, 56, 1998.


q 2006 by Taylor & Francis Group, LLC
Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods                     71


    75. M. T. Gillies, Experiments on host selection in the Anopheles gambiae complex, Annals of
        Tropical Medicine and Parsitology, 61, 68, 1967.
    76. T. Dekker and W. Takken, Differential responses of mosquito sibling species Anopheles
        arabiensis and An. quadriannulatus to carbon dioxide, a man or a calf, Medical & Veterinary
        Entomology, 12, 136, 1998.
    77. T. Dekker, W. Takken, and M. A. H. Braks, Innate preference for host-odor blends modulates
        degree of anthropophagy of Anopheles gambiae sensu lato (Diptera: Culicidae), Journal of
        Medical Entomology, 38, 868, 2001.
    78. H. V. Pates, W. Takken, C. F. Curtis, P. W. Huisman, O. Akinpelu, and G. S. Gill, Unexpected
        anthropophagic behaviour in Anopheles quadriannulatus, Medical & Veterinary Entomology, 15,
        293, 2001.
    79. H. V. Pates, W. Takken, K. Stuke, and C. F. Curtis, Differential behaviour of Anopheles gambiae
        sensu stricto (Diptera: Culicidae) to human and cow odours in the laboratory, Bulletin of
        Entomological Research, 91, 289, 2001.
    80. K. M. V. Jensen, J. B. Jespersen, M. A. Birkett, J. A. Pickett, G. Thomas, L. J. Wadhams, and
        C. M. Woodcock, Variation in the load of the horn fly, Haematobia irritans, in cattle herds is
        determined by the presence or absence of individual heifers, Medical & Veterinary Entomology,
        18, 275, 2004.
    81. G. Thomas, cited in Pickett and Woodcock, 1996 (Ref.83).
    82. M. A. Birkett, N. Agelopoulos, K. M. V. Jensen, J. B. Jespersen, J. A. Pickett, H. J. Prijs,
        G. Thomas, J. J. Trapman, L. J. Wadhams, and C. M. Woodcock, The role of volatile
        semiochemicals in mediating host location and selection by nuisance and disease-transmitting
        cattle flies, Medical & Veterinary Entomology, 18, 313, 2004.
    83. J. A. Pickett and C. M. Woodcock, The role of mosquito olfaction in oviposition site location
        and in the avoidance of unsuitable hosts, in Olfaction in Mosquito–Host Interactions, G. R. Bock
        and G. Cardew (Eds.), New York: Wiley, Ciba Foundation Symposium 200, 1996, p. 109.
    84. Y. S. Pan, Breed and seasonal differences in quantities of lipids on skin surface and hair in cattle,
        Journal of Agricultural Science, 75, 41, 1970.
    85. J. C. Bonsma, Breeding tick-repellent cattle, in Tick Biology and Control: Proceedings of an
        International Conference, G. B. Whitehead and J. D. Gibson (Eds.), Tick Research Unit, Rhodes
        University: Grahamstown, 1981, p. 67.
    86. W. Wanzala, N. F. K. Sika, S. Gule, and A. Hassanali, Attractive and repellent host odours guide
        ticks to their respective feeding sites, Chemoecology, 14, 229, 2004.
    87. N. F. K. Sika, Behavioural responses of Rhipicephalus appendiculatus Neumann 1901 and
        Rhipicephalus evertsi Neumann 1897 (Acari: Ixodidae) to host and non-host semiochemicals.
        Ph.D. dissertation, Kenyatta University, Kenya, 1996.
    88. J. F. Carroll, J. A. Klun, and E. T. Schmidtmann, Evidence for kairomonal influence on selection
        of host-ambushing sites by adult Ixodes scapularis (Acari: Ixodidae), Journal of Medical
        Entomology, 32, 119, 1995.
    89. J. F. Carroll, Responses of three species of adult ticks (Acari: Ixodidae) to chemicals in the coats
        of principal and minor hosts, Journal of Medical Entomology, 36, 238, 1999.
    90. J. F. Carroll, Interdigital gland substances of white-tailed deer and the response of host-seeking
        ticks (Acari: Ixodidae), Journal of Medical Entomology, 38, 114, 2001.
    91. J. F. Carroll, Notes on responses of blacklegged ticks (Acari: Ixodidae) to host urine, Journal of
        Medical Entomology, 36, 212, 1999.
    92. J. F. Carroll, Responses of adult Ixodes scapularis (Acari: Ixodidae) to urine produced by
        white-tailed deer of various reproductive conditions, Journal of Medical Entomology, 37, 472,
        2000.
    93. G. A. Vale, Mobile attractants for tsetse flies, Arnoldia, 4, 1, 1969.
    94. G. A. Vale, New field methods for studying the responses of tsetse flies (Diptera, Glossinidae) to
        hosts, Bulletin of Entomological Research, 64, 199, 1974.
    95. G. A. Vale, The responses of tsetse flies (Diptera, Glossinidae) to mobile and stationary hosts,
        Bulletin of Entomological Research, 64, 545, 1974.


q 2006 by Taylor & Francis Group, LLC
72                                                         Insect Repellents: Principles, Methods, and Uses


    96. G. A. Vale, Feeding responses of tsetse flies (Diptera: Glossinidae) to stationary hosts, Bulletin of
        Entomological Research, 67, 635, 1977.
    97. J. W. Hargrove, The effect of human presence on the behaviour of tsetse (Glossina spp.) (Diptera,
        Glossinidae) near a stationary ox, Bulletin of Entomological Research, 66, 173, 1976.
    98. G. A. Vale, Field responses of tsetse flies (Diptera: Glossinidae) to odours of men, lactic acid and
        carbon dioxide, Bulletin of Entomological Research, 69, 459, 1979.
    99. B. M. Steib, M. Geier, and J. Boeckh, The effect of lactic acid on odour-related host preference of
        yellow fever mosquitoes, Chemical Senses, 26, 523, 2001.
   100. A. W. A. Brown, D. S. Sarkaria, and R. P. Thompson, Studies on the responses of the female Aedes
        mosquito, Part I. The search for attractant vapours, Bulletin of Entomological Research, 42, 105, 1951.
   101. R. P. Thompson and A. W. A. Brown, The attractiveness of human sweat to mosquitoes and the role of
        carbon dioxide, Mosquito News, 15, 80, 1955.
   102. M. R. Smart and A. W. A. Brown, Studies on the responses of the female Aedes mosquito. Part VII. The
        effect of skin temperature, hue and moisture on the attractiveness of the human hand, Bulletin of
        Entomological Research, 47, 89, 1956.
   103. H. I. Maibach, W. A. Skinner, W. G. Strauss, and A. A. Khan, Factors that attract and repel mosquitoes in
        human skin, Journal of the American Medical Association, 196, 173, 1966.
               ¨                                                            ¨
   104. W. Muller, Die Distanz- und Kontakt-Orientierung der Stechmucken (Aedes aegypti) (Wirtsfindung,
                                                         ¨
        Stechverhalten und Blutmahlzeit), Zeitschrift fur Vergleichende Physiologie, 58, 241, 1968.
   105. W. A. Skinner, H. Tong, H. Maibach, A. A. Khan, and T. Pearson, Repellency of skin-surface lipids of
        humans to mosquitoes, Science, 149, 305, 1965.
   106. W. A. Skinner, H. C. Tong, T. R. Pearson, and H. I. Maibach, Repellency of human skin-surface lipid
        hydrocarbons to the yellow-fever mosquito, Journal of Economic Entomology, 60, 927, 1967.
   107. W. A. Skinner, H. C. Tong, H. I. Maibach, and D. Skidmore, Human skin-surface lipid fatty acids—
        mosquito repellents, Experientia, 26, 728, 1970.
   108. W. A. Skinner, H. Tong, H. Johnson, H. Maibach, and D. Skidmore, Human sweat components—
        attractancy and repellency to mosquitoes, Experientia, 24, 679, 1968.
   109. O. J. Bosch, M. Geier, and J. Boeckh, Contribution of fatty acids to olfactory host finding offemale Aedes
        aegypti, Chemical Senses, 25, 323, 2000.
   110. B. G. J. Knols, J. J. A. van Loon, A. Cork, R. D. Robinson, W. Adam, J. Meijerink, R. De Jong, and
        W. Takken, Behavioural and electrophysiological responses of the female malaria mosquito Anopheles
        gambiae (Diptera: Culicidae) to Limburger cheese volatiles, Bulletin of Entomological Research, 87,
        151, 1997.
   111. R. C. Smallegange, Y. T. Qiu, J. J. van Loon, and W. Takken, Syngergism between ammonia, lactic acid
        and carboxylic acids as kairomones in the host-seeking behaviour of the malaria mosquito Anopheles
        gambiae sensu stricto (Diptera: Culicidae), Chemical Senses, 30, 145, 2005.
   112. Y. Shirai, K. Kamimura, T. Seki, and M. Morohashi, L-lactic acid as a mosquito (Diptera: Culicidae)
        repellent on human and mouse skin, Journal of Medical Entomology, 38, 51, 2001.
   113. C. N. Smith, N. Smith, H. K. Gouck, D. E. Weidhaas, I. H. Gilbert, M. S. Mayer, B. J. Smittle, and
        A. Hofbauer, L-lactic acid as a factor in the attraction of Aedes aegypti (Diptera: Culicidae) to human
        hosts, Annals of the Entomological Society of America, 63, 760, 1970.
   114. C. Costantini, M. A. Birkett, G. Gibson, J. Ziesmann, N. F. Sagnon, H. A. Mohammed, M. Coluzzi, and
        J. A. Pickett, Electroantennogram and behavioural responses of the malaria vector Anopheles gambiae
        to human-specific sweat components, Medical & Veterinary Entomology, 15, 259, 2001.
   115. W. R. Mukabana, W. Takken, G. F. Killeen, and B. G. J. Knols, Allomonal effect of breath contributes to
        differential attractiveness of humans to the African malaria vector Anopheles gambiae, Malaria
        Journal, 3, 1, 2004.
   116. K. E. McKenzie, The occurrence of preferential host selection by Aedes aegypti (Linnaeus 1762)
        (Diptera: Culicidae), M.S. thesis, University of Florida, Gainesville, 2000.
   117. U. R. Bernier, D. L. Kline, C. E. Schreck, R. A. Yost, and D. R. Barnard, Chemical analysis of human
        skin emanations: Comparison of volatiles from humans that differ in attraction of Aedes aegypti
        (Diptera: Culicidae), Journal of the American Mosquito Control Association, 18, 186, 2002.
   118. J. Kingdon, East African Mammals: Vol. IIIA, Carnivores, Chicago: University of Chicago Press, 1977.


q 2006 by Taylor & Francis Group, LLC
Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods                     73


   119. A. M. Young, A human sweat-mediated defense against multiple attacks by the wasp Polybia diguetana
        in northeastern Costa Rica, Biotropica, 10, 73, 1978.
   120. L. E. Sedgwick, Magpies “anting”, Western Australian Naturalist, 1, 21, 1947.
   121. A. R. Laskey, Bronzed grackle anointing plumage with orange-skin, The Wilson Bulletin, 60, 244,
        1948.
   122. A. C. Govan, “Anting”—outside and inside the windowpane, Nature Magazine, January, 41, 1954.
   123. L. M. Whitaker, A resume of anting, with particular reference to a captive orchard oriole, The Wilson
        Bulletin, 69, 195, 1957.
   124. A. H. Chisholm, The history of anting, Emu, 59, 101, 1959.
   125. J. M. Goodman, Aves incendiaria, The Wilson Bulletin, 72, 400, 1960.
   126. L. Kelso and M. M. Nice, A Russian contribution to anting and feather mites, The Wilson Bulletin,
        75, 23, 1963.
   127. K. E. L. Simmons, Anting and the problem of self-stimulation, Journal of the Zoology (London), 149,
        145, 1966.
   128. E. F. Potter and D. C. Hauser, Relationship of anting and sunbathing to molting in wild birds, Auk,
        91, 537, 1974.
   129. J. V. Dennis, Commentary on grackles anting with marigold blossoms, Blue Jay, 43, 175, 1985.
   130. D. H. Clayton and J. G. Vernon, Common grackle anting with lime fruit and its effect on
        ectoparasites, Auk, 110, 951, 1993.
   131. H. Sick, Birds in Brazil, Princeton, NJ: Princeton University Press, 1993.
   132. K. C. Parkes, P. J. Weldon, and R. L. Hoffman, Polydesmidan millipede used in self-anointing by a
        strong-billed woodcreeper (Xiphocolaptes promeropirhyncus) from Belize, Ornitology Neotropical,
        14, 285, 2003.
   133. E. A. Vander Werf, ‘Elepaio “anting” with a garlic snail and a Schinus fruit, Journal of Field
        Ornithology, 76, 134, 2005.
   134. E. Gould and J. F. Eisenberg, Notes on the biology of the Tenrecidae, Journal of Mammalogy, 47,
        660, 1966.
   135. E. D. Brodie, Jr., Hedgehogs use toad venom in their own defence, Nature, 268, 627, 1977.
                                    ¨
   136. A. Nolte, Beobachtungen uber das Instinktverhalten von Kapuzineraffen (Cebus apella L.) in der
        Gefangenschaft, Behaviour, 13, 189, 1958.
   137. Oppenheimer, J. R. Behavior and ecology of the white-faced monkey, Cebus capucinus, on Barro
        Colorado Island, C.Z., Ph.D. dissertation, University of Illinois at Urbana-Champaign, 1968.
   138. J. Longino, True anting by the capuchin, Cebus capucinus, Primates, 25, 243, 1984.
   139. D. J. Overdorff, Similarities, differences, and seasonal patterns in the diets of Eulemur rubriventer
        and Eulemur fulvus rufus in the Ranomafana National Park, Madagascar, International Journal of
        Primatology, 14, 721, 1993.
   140. X. Valderrama, J. G. Robinson, A. B. Attygalle, and T. Eisner, Seasonal anointment with
        millipedes in a wild primate: A chemical defense against insects?, Journal of Chemical Ecology,
        26, 2781, 2000.
   141. M. DeJoseph, R. S. Taylor, M. Baker, and M. Aregullin, Fur-rubbing behavior of capuchin monkeys,
        Journal of the American Academy of Dermatology, 46, 924, 2002.
   142. M. Zito, S. Evans, and P. J. Weldon, Owl monkeys (Aotus spp.) self-anoint with plants and
        millipedes, Folia Primatologica, 74, 159, 2003.
   143. M. P. Verderane, T. Falotico, B. D. Resende, M. B. Labruna, P. Izar, and E. Ottoni, Anting in a
        semifree-ranging group of Cebus apella, International Journal of Primatology, in press.
   144. Z. Xu, D. M. Stoddart, H. Ding, and J. Zhang, Self-anointing behavior in the rice-field rat, Rattus
        rattoides, Journal of Mammalogy, 76, 1238, 1995.
   145. E. Zimen, The Wolf: A Species in Danger, New York: Delacourte, 1981.
   146. I. Reiger, Scent rubbing in carnivores, Carnivore, 2, 17, 1979.
   147. A. O. Tucker and S. S. Tucker, Catnip and the catnip response, Economic Botany, 42, 214,
        1988.
   148. P. J. Weldon and D. B. Fagre, Responses by canids to scent gland secretions of the western
        diamondback rattlesnake (Crotalus atrox), Journal of Chemical Ecology, 15, 1589, 1989.


q 2006 by Taylor & Francis Group, LLC
74                                                         Insect Repellents: Principles, Methods, and Uses


   149. M. E. Gompper and A. M. Hoylman, Grooming with Trattinnickia resin: Possible pharma-
        ceutical plant use by coatis in Panama, Journal of Tropical Ecology, 9, 533, 1993.
   150. S. Sigstedt, cited in Clayton and Wolfe, 1993 (Ref.3).
   151. L. M. Gosling and H. V. McKay, Scent-rubbing and status signalling by male mammals,
        Chemoecology, 1, 92, 1990.
   152. P. R. Ehrlich, D. S. Dobkin, and D. Wheye, The adaptive significance of anting, Auk, 103, 835, 1986.
                                                             ¨
   153. W. Eichler, Die Biologie der Federlinge, Journal fur Ornithologie, 84, 471, 1936.
   154. S. Wilson and N. Hillgarth, cited in Clayton and Wolfe, 1993 (Ref.3).
   155. A. Bennett and D. Clayton, cited in Clayton and Wolfe, 1993 (Ref.3).
   156. A. T. D. Bennett, N. Hillgarth and J. Vernon cited in Judson and Bennett, 1992 (Ref.157).
   157. O. P. Judson and A. T. D. Bennett, Anting as food preparation: Formic acid is worse on an empty
        stomach, Behavioral Ecology and Sociobiology, 31, 437, 1992.
   158. J. F. Carroll, M. Kramer, P. J. Weldon, and R. G. Robbins, Anointing chemicals and ectoparasites:
        Effects of benzoquinones from millipedes on the lone star tick, Amblyomma americanum, Journal of
        Chemical Ecology, 31, 63, 2005.
   159. G. Bisignano and A. Saija, The biological activity of citrus oils, in Citrus: The Genus Citrus, G. Dugo
        and A. Di Giacomo (Eds.), New York: Taylor & Francis, 2002, p. 602.
   160. Y. Trongtokit, Y. Rongsriyam, N. Komalamisra, and C. Apiwathnasorn, Comparative repellency of 38
        essential oils against mosquito bites, Phytotherapy Research, 19, 303, 2005.
   161. M. A. Oshaghi, R. Ghalandari, H. Vatandoost, M. Shayeghi, M. Kamali-Nejad, H. Tourabi-Khaledi,
        M. Abolhassani, and M. Hashemzadeh, Repellent effect of extracts and essential oils of Citrus limon
        (Rutaceae) and Melissa officinalis (Labiatae) against main malaria vector, Anopheles stephensi
        (Diptera: Culicidae), Iranian Journal of Public Health, 32, 47, 2003.
   162. A. A. Shalaby, K. A. Allam, A. A. Mostafa, and S. M. Fahmy, Insecticidal properties of citrus oils against
        Culex pipiens and Musca domestica, Journal of the Egyptian Society of Parasitology, 28, 595, 1998.
   163. P. J. Weldon, Unpublished.
   164. G. Dugo, A. Cotroneo, A. Verzera, and I. Bornaccorsi, Composition of the volatile fraction of cold-
        pressed citrus peel oils, in Citrus: The Genus Citrus, G. Dugo and A. Di Giacomo (Eds.), New York:
        Taylor & Francis, 2002, p. 201.
   165. D. M. Norris, Repellents, in CRC Handbook of Natural Pesticides, Vol. VI, Insect Attractants
        and Repellents, E. D. Morgan and N. B. Mandava (Eds.), Boca Raton: CRC Press, 1996, p. 135.
   166. S. Dev and O. Koul, Insecticides of Natural Origin, Amsterdam: Harwood, 1997.
   167. W. F. Hink and B. J. Fee, Toxicity of D-limonene, the major component of citrus peel oil, to all life
        stages of the cat flea, Ctenocephalides felis (Siphonaptera: Pulicidae), Journal of Medical
        Entomology, 23, 400, 1986.
   168. W. F. Hink, T. A. Liberati, and M. G. Collart, Toxicity of linalool to life stages of the cat flea,
        Ctenocephalides felis (Siphonaptera: Pulicidae), and its efficacy in carpet and on animals, Journal
        of Medical Entomology, 25, 1, 1988.
   169. K. Y. Mumcuoglu, R. Galun, U. Bach, J. Miller, and S. Magdassi, Repellency of essential oils and
        their components to the human body louse, Pediculus humanus humanus, Entomologia Experi-
        mentalis et Applicata, 78, 309, 1996.
   170. J. F. Carroll, Feeding deterrence of northern fowl mites (Acari: Macronyssidae) by some naturally
        occurring plant substances, Pesticide Science, 41, 203, 1994.
                                          ˆ
   171. T. Eisner, Catnip: Its raison d’etre, Science, 146, 1318, 1964.
   172. C. J. Petersen, Insect repellents of natural origin: Catnip and osage orange, Ph.D. dissertation, Iowa
        State University, Ames, 2001.
   173. U. R. Bernier, K. D. Furman, D. L. Kline, S. A. Allan, and D. R. Barnard, Comparison of contact
        and spatial repellency of catnip oil and N,N-diethyl-3-methylbenzamide (deet) against mosquitoes,
        Journal of Medical Entomology, 42, 306, 2005.
   174. K. R. Chauhan, J. A. Klun, M. Debboun, and M. Kramer, Feeding deterrent effects of catnip oil
        components compared with two synthetic amides against Aedes aegypti, Journal of Medical
        Entomology, 42, 643, 2005.
   175. W. Wickler, Mimicry in Plants and Animals, New York: McGraw-Hill, 1968.


q 2006 by Taylor & Francis Group, LLC
Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods                      75


   176. C. M. Passreiter, Y. Akhtar, and M. B. Isman, Insecticidal activity of the essential oil of
                                                 ¨
        Ligusticum mutellina roots, Zeitschrift fur Naturforschung C, 60, 411, 2005.
   177. P. T. Fauth, D. G. Krementz, and J. E. Hines, Ectoparasitism and the role of green nesting material
        in the European starling, Oecologia, 88, 22, 1991.
   178. H. Gwinner, M. Oltrogge, L. Trost, and U. Nienaber, Green plants in starling nests: Effects on
        nestlings, Animal Behaviour, 59, 301, 2000.
   179. P. H. Wimberger, The use of green plant material in bird nests to avoid ectoparasites, Auk, 101,
        615, 1984.
   180. L. Clark and J. R. Mason, Use of nest material as insecticidal and anti-pathogenic agents by the
        European starling, Oecologia, 67, 169, 1985.
   181. S. Sengupta, Adaptive significance of the use of margosa leaves in nests of house sparrows Passer
        domesticus, Emu, 81, 114, 1981.
   182. L. Clark and J. R. Mason, Effect of biologically active plants used as nest material and the derived
        benefit to starling nestlings, Oecologia, 77, 174, 1988.
   183. J. R. Mason and L. Clark, Chemoreception and the selection of green plants as nest fumigants by
        starlings, in Chemical Signals in Vertebrates 4: Ecology, Evolution, and Comparative Biology,
                         ¨
        D. Duvall, D. Muller-Schwarze, and R. M. Silverstein (Eds.), New York: Plenum Press, 1986, p. 369.
   184. C. Petit, M. Hossaert-McKey, P. Perret, J. Blondel, and M. M. Lambrechts, Blue tits use selected
        plants and olfaction to maintain an aromatic environment for nestlings, Ecology Letters, 5, 585,
        2002.
   185. L. Clark and C. A. Smeraski, Seasonal shifts in odor acuity by starlings, Journal of Experimental
        Zoology, 255, 22, 1990.
   186. L. Lafuma, M. M. Lambrechts, and M. Raymond, Aromatic plants in bird nests as a protection
        against blood-sucking flying insects?, Behavioural Processes, 56, 113, 2001.
   187. R. B. Hemmes, A. Alvarado, and B. L. Hart, Use of California bay foliage by wood rats for possible
        fumigation of nest-borne ectoparasites, Behavioral Ecology, 13, 381, 2002.
   188. D. J. Brightsmith, Use of arboreal termitaria by nesting birds in the Peruvian Amazon, Condor, 102,
        529, 2000.
   189. D. H. Janzen, Birds and the ant!acacia interaction in Central America, with notes on birds and other
        myrmecophytes, Condor, 71, 240, 1969.
   190. N. Nicolaides, Skin lipids: Their biochemical uniqueness, Science, 186, 19, 1974.
   191. P. J. Weldon, Thin-layer chromatography of the skin secretions of vertebrates, in Practical Thin-layer
        Chromatography: A Multidisciplinary Approach, B. Fried and J. Sherma (Eds.), Boca Raton: CRC
        Press, 1996, p. 105.
   192. A. M. Kligman, The uses of sebum?, in The Sebaceous Glands, W. Montagna, R. A. Ellis, and
        A. F. Silver (Eds.), New York: Pergamon Press, 1963, p. 110.
   193. W. F. Wood and P. J. Weldon, The scent of the reticulated giraffe (Giraffa camelopardalis
        reticulata), Biochemical Systematics and Ecology, 30, 913, 2002.
   194. W. Rudolfs, Chemotropism of mosquitoes, New Jersey Agricultural Experiment Station Bulletin,
        367, 4, 1922.
   195. W. Rudolfs, Effects of chemicals upon the behavior of mosquitoes, New Jersey Agricultural
        Experiment Station Bulletin, 496, 2, 1930.
   196. W. F. Wood, M. G. Leahy, R. Galun, G. D. Prestwich, J. Meinwald, R. E. Purnell, and R. C. Payne,
        Phenols as pheromones of ixodid ticks: A general phenomenon?, Journal of Chemical Ecology, 1,
        501, 1975.
   197. J. A. Pickett, L. J. Wadhams, and C. M. Woodcock, Insect supersense: Mate and host location by
        insects as model systems for exploiting olfactory interactions, The Biochemist, August, 8, 1998.
   198. T. Eisner and R. P. Grant, Toxicity, odor aversion, and “olfactory aposematism”, Science, 213, 476,
        1981.




q 2006 by Taylor & Francis Group, LLC
q 2006 by Taylor & Francis Group, LLC
4
Human Emanations and Related Natural Compounds
That Inhibit Mosquito Host-Finding Abilities


Ulrich R. Bernier, Daniel L. Kline, and Kenneth H. Posey



CONTENTS
Concepts and Terminology Used in This Chapter .........................................................................78
  Attraction Antagonists and Anti-Attractants...............................................................................78
  Spatial Repellents and Irritants ...................................................................................................79
  Attraction-Inhibitors ....................................................................................................................80
Spatial Repellency and Attraction-Inhibition Research .................................................................80
  Early History of Spatial Repellents Testing ...............................................................................80
  Spatial Repellency and Attraction-Inhibition of Catnip Oil.......................................................81
  Spatial Repellency and Attraction-Inhibition of Deet ................................................................81
  Attraction-Inhibition by Linalool and Related Compounds .......................................................82
  Human-Produced Compounds That Affect Host-Seeking ..........................................................82
     Attraction-Inhibition by Carboxylic Acids .............................................................................85
     Attraction-Inhibition by Aldehydes.........................................................................................85
     Attraction-Inhibition by Ketones.............................................................................................86
     Attraction-Inhibition by Alcohols ...........................................................................................86
     Attraction-Inhibition by Compounds of Other Classes ..........................................................87
Identification of Host-Produced Allelochemicals...........................................................................87
  Analysis of Human Emanations..................................................................................................88
  Merging Chemistry and Sensory Physiology..............................................................................88
  Current State and Future Directions of Host Odor Research .....................................................89
Laboratory Bioassays of Spatial Repellents and Attraction-Inhibitors..........................................89
  Olfactometers for the Assessment of Spatial Repellents............................................................89
  Olfactometers for the Assessment of Attraction-Inhibitors........................................................90
     Considerations in the Experimental Design............................................................................90
     Correlating Small- and Large-Scale Laboratory Results to Field Experiments ....................91
Field Tests and Use of Spatial Repellents and Attraction-Inhibitors ............................................92
  Experimental Design of Field Tests............................................................................................92
     Use of Large-Cage Experiments and Laboratory-Reared Colony Mosquitoes......................92
     Experiments with Wild Mosquitoes in the Field ....................................................................92
  Use of Stand-Alone Inhibitor-Delivery Technology ..................................................................93
Potential Applications of Spatial Repellents and Attraction-Inhibitors.........................................93
  Species-Specific or Species-Exclusive Trapping ........................................................................93


                                                                                                                                          77

q 2006 by Taylor & Francis Group, LLC
78                                                                           Insect Repellents: Principles, Methods, and Uses


  Local Area Host-Finding Reduction ...........................................................................................93
  Local Control Using a Push–Pull Strategy with Attractant-Baited Surveillance Traps ............93
  Use of Structure–Activity Relationships to Benefit Development of Attraction-Inhibitors......94
References .......................................................................................................................................94



        The beginning of knowledge is the discovery of something we do not understand.1




Concepts and Terminology Used in This Chapter
One mechanism by which the action of semiochemicals can be classified is based on the behavioral
impact within or external to the species of interest. As such, one can classify a chemical as one of the
following2,3:

      1. Pheromone, if it results in response between insects of the same species
      2. Kairomone, if it results in response in another species that benefits the species receiving the
         chemical cue
      3. Allomone, if it results in response in another species that benefits the species releasing the
         chemical cue

   However, the distinctions can be more specific by classification of chemical cues through the imparted
behavioral effect: attractant; repellent; arrestant; locomotory stimulant; feeding, mating, or oviposition
stimulant; and feeding, mating, or oviposition deterrent.2,4 Karlson and Luscher first proposed the term
                                                                             ¨
“pheromone” to describe chemicals with instraspecific species activity.5,6 Chemicals with interspecific
species activity are allelochemicals.3 Allelochemicals can be separated further into kairomones, of which
attractants are a category of, and allomones, which are the primary focus of both this book and chapter,
and the class that repellents are a part of. Furthermore, attraction-inhibitors, are classified by us as a
category of repellents. Ironically, many of the attraction-inhibitors have been discovered in a search for
kairomones used by mosquitoes to locate human hosts. Attraction-inhibitors may not repel by the
traditional mechanisms, but they do interfere, or act as an antagonist to the normal attraction response of
an insect to attractive odor(s).
   The proper name for the behavioral actions that are described in this chapter can be debated
extensively and additional discussion of terminology is found in Chapter 2 by White. In this short
prequel to the main body of our contribution on human and other compounds that interfere with mosquito
host-finding, we put forth our rationale supporting the terms used to describe behaviors reported in
this chapter.


Attraction Antagonists and Anti-Attractants
Attraction antagonist is an appropriate term to describe compounds that interrupt the blood-feeding
process in bioassays. The term “antagonism” has been used to describe a phenomenon between two
toxicants that is the opposite effect of synergism.7,8 Applying this by analogy to attraction, a synergistic
response is then a response where the combination of chemicals in a blend produces a level of attraction
greater than the sum of attraction response levels to the single substances. Thus, antagonism describes a
situation where the attraction to a combination is less than the sum of individual attraction levels.


q 2006 by Taylor & Francis Group, LLC
Natural Compounds That Inhibit Mosquito Host-Finding Abilities                                             79


Wright et al.9 stated that, in principle, these compounds also function as “anti-attractants” in that they
disrupt the function of naturally occurring attractants. Wright et al. noted furthermore that repellents and
anti-attractants should be considered as separate functional classes of compounds based on their different
modes of action.
   It is commonly accepted that a volatile chemical attractant is a substance that produces oriented insect
movement (positive taxis) or upwind movement (anemotaxis) toward a source by following a
concentration gradient of gas-phase molecules distributed in plumes.10 Therefore, a broad term is
needed to describe compounds that prevent host finding by interference of the positive anemotaxis
without too much reliance on characterizing the mechanism of action on the insect. Bearing this mind,
compounds that repel using the criteria of Dethier et al.4 would be classified as antagonists because the
repellent substances prevent host finding by an oriented movement away from the source. However,
compounds that cloak or hide the host from mosquitoes that would otherwise be able to locate the host for
a blood meal would not fit the strict definition of repellents in the sense of Dethier et al. because the
mosquitoes would not necessarily exhibit oriented movement away from the source.4


Spatial Repellents and Irritants
Spatial repellency and irritancy can involve more than simple concealment of host location or attractant
odor source. As noted above, a rigorous definition of “repellent” requires movement away from the
source. Barton Browne later suggested that “movement away” is not necessarily a suitable criterion, and
that “a repellent is almost always assessed in terms of its ability to inhibit the insect’s response to
chemical attractants.”10 This led to the proposal that a repellent is a chemical that, acting in the vapor
phase, prevents an insect from reaching a target to which it would otherwise be attracted. However, it
should be noted that vapor-phase activity might not be necessary to repel. If contact is made with a
surface that contains repellent, then the mosquito chemoreceptors can detect this repellent if it is present
at the required threshold concentration to cause repellency. Additionally, it should be apparent that
chemical compounds have a vapor phase concentration that is dependent upon their volatilities, and
that this concentration falls off as the distance from the source increases. Therefore, the true criterion is
linked to the threshold level of chemoreception of (repellent) molecules by the mosquito to elicit the
desired behavior (repellency). Obviously, this can occur in space if the mosquito has high sensitivity to
the chemical and the chemical has a high vapor phase concentration. Similarly, repellent chemicals that
the mosquito is less sensitive to, or that have low vapor phase concentration from low evaporative loss,
will result in repellency closer to the surface, or perhaps even by contact.
   The use of the word “spatial” to classify repellents was defined by Gouck et al.11 as a compound or
agent that can produce repellency at a distance. Furthermore, spatial repellents have been described as
repellents that inhibit the ability of mosquitoes to locate a target host.12 Thus, topical repellents with low
vapor pressure, such as N,N-diethyl-3-methylbenzamide (deet), and highly volatile spatial repellents
(attraction-inhibitors) are repellents, even though their modes of action may be radically different.
   Another possible source of confusion arises from the term “area repellent.” Although spatial repellents
should ideally prevent biting in a defined local area, an area repellent does not necessarily require a
significant vapor phase spatial repellent effect. An example of this is the use of a repellent that is
normally applied topically, such as deet, on a treated net to form a barrier around a perimeter.13 For more
discussion of area repellents, see Chapter 23 by Strickman.
   At times, the term “spatial repellent” is used to describe the action of some pyrethroids.14,15 It should
be noted that pyrethroids can produce excito-repellency with possible mortality as a result of the
exposure.16 A pyrethroid with sufficiently high vapor phase concentration, e.g., metofluthrin,15,17 can
result in a spatial repellent (barrier) effect regardless of knockdown and mortality of insects. In this
chapter, the discussion is mostly confined to the natural compounds that impact mosquito behavior by a
means of masking attractive odors while minimizing concerns over the mode of action, such as the
pyrethroids that exhibit excito-repellency and insecticidal properties.


q 2006 by Taylor & Francis Group, LLC
80                                                      Insect Repellents: Principles, Methods, and Uses


Attraction-Inhibitors
The term ‘‘inhibition’’ has been used to describe a net behavioral effect from a particular mechanism,
such as ‘‘distension-induced inhibition.’’18 Simpson and Wright described the use of low-level
continuous emission of a chemical, e.g. Rutgers 612 (2-ethyl-1,3-hexanediol) as a means to ‘‘inhibit
the normal response’’ of mosquitoes to an increase in the carbon dioxide gradient.19 Although the normal
response to carbon dioxide can range from flight activation to oriented positive anemotaxis to the odor
source, it is assumed that in this case, the authors expected the normal response to be that of attraction.
The term ‘‘inhibitor’’ also denotes a compound that imparts a reduction in trap catches for traps baited
with a pheromone.20,21 Davis linked a decrease in sensitivity of lactic acid receptor neurons to the
inhibition of host-seeking behavior following a blood meal.22 If we adopt and apply ‘‘inhibitor’’ in an
analogous way to describe these allomones that inhibit the activity of kairomones, these compounds are
then inhibitors of attractants (i.e. attraction-inhibitors analogous to pheromone inhibitors described by
Roelofs and Comeau,20 and Kennedy).22 Torr et al.23 later expounded on the work of Davis and discussed
the manner in which these ‘‘attractant-inhibitors’’ may affect the insects. We have shifted away from
calling human-produced masking chemicals ‘‘spatial repellents’’ in recent years and adopted the term
‘‘attraction-inhibitors.’’ We believe this term to be a logical choice to describe the observed behavioral
effect (inhibition) in bioassays.24,25




Spatial Repellency and Attraction-Inhibition Research
In the mid 1960s, Skinner et al.26 collected lipid fractions from human skin exudates and reported that
some of these lipid fractions were “repellent” to mosquitoes. They hypothesized that the attraction of
mosquitoes to humans was more complex than simply locating a host using kairomones only. It was
speculated that the combination of human-produced kairomones and allomones resulted in the overall
measured attractiveness of an individual to mosquitoes.26,27 Further investigation of the lipid fractions
implicated unsaturates as the repellent allomones.28 Moreover, measured attraction increased when these
lipids were removed from sweat. Skinner et al.29 later identified the most repellent of these acids as
a-linolenic (9,12,15-octadecatrienoic), 2-decenoic, 2-nonenoic, arachidonic (5,8,11,14-octadecatetrae-
noic), and 10-undecenoic acids. Some saturated fatty acids, e.g., caproic (hexanoic), enanthic
(heptanoic), and pelargonic (nonanoic) acids also exhibited high repellency. The carboxylic acids and
their effect on host-seeking will be examined more in-depth in Section “Attraction-Inhibition by
Carboxylic Acids”.


Early History of Spatial Repellents Testing
The notations of vapor phase, spatial, and area effects from repellents were reported by Christophers in
1945, and this was especially noticed from the action of pyrethrins.30 Christophers also noted a
distinction between “contact” repellents and “vapour” repellents described by McCulloch and Water-
house.31 A concerted search for “spatial” repellents was undertaken by the USDA in 1948.32 Although
we will continue to use “spatial” in place of “vapour” repellents, they are both defined by the respective
authors as repellents that work “at a distance.”11,31 The USDA effort came about as an offshoot from the
established program of topical repellent testing. In their first report of spatial repellents, some of the 110
chemicals tested exhibited repellency, but none were deemed to be outstanding. Results based upon this
USDA effort were first described in the literature by Gouck et al.11; who reported on the spatial
repellency of various esters using time of protection from bites as the means of quantifying the
differences in repellency (see Section “Olfactometers for the Assessment of Spatial Repellents” for a
description of their bioassay system). Of the esters that were tested, it was reported that the spatial
repellency for the mandelates increased as a function of the carbon chain length from C3 to C8, with an


q 2006 by Taylor & Francis Group, LLC
Natural Compounds That Inhibit Mosquito Host-Finding Abilities                                             81


optimum that occurred at C5. McGovern et al.33 used a similar assay system to study other compound
classes, and in particular noted that deet was an effective topical repellent, but performed only weakly as
a spatial repellent. Maximum spatial repellency occurred in the C5 to C9 range for most of the compound
classes tested. Later, other common topical repellents were examined and dimethyl phthalate (DMP) was
reported as one of the best spatial repellents against Aedes aegypti34; however, Khan and Maibach found
deet to be better than DMP using their own biossay methods.35 Other noteworthy substances that have
been shown to produce spatial repellency of Aedes aegypti include essential oils like citronellal
(3,7-dimethyl-6-octen-1-al) and geraniol (E-3,7-dimethyl-2,6-octadien-1-ol), pyrethrums and pyre-
throids, and common topical repellents.14 Recent efforts to find inhibitory chemicals are directed at
natural compounds by examining differences among individual humans of variable attractiveness to
biting arthropods.


Spatial Repellency and Attraction-Inhibition of Catnip Oil
It has been known for some time that volatiles produced from catnip, specifically the isomers of
nepetalactone, repel phytophagous insects.36,37 Peterson and Coats examined catnip oil and nepetalac-
tone isomers as alternatives to deet for protection from mosquitoes and found these to be more repellent
than deet in their bioassay system.38,39 Recently, catnip was examined for its ability to inhibit the host-
seeking of mosquitoes and was found to be a better attraction-inhibitor than deet, but the less effective
repellent of the two based on mean complete protection time (CPT) on a treated cloth affixed to a card
above the skin surface.40 Further examination of catnip oil and its constituents to deter biting was
conducted by Chauhan et al.41 The results of their in vivo and in vitro studies were similar in that the
biting deterrency of each of the two nepetalactone isomers (Z,E- and E,Z-) and of the racemic mixture
were all significant compared to the control, but not different from each other. Tested in vitro, these
compounds did not deter biting as well as deet or the repellent, (1S,2S 0 )-2-methylpiperidinyl-3-
cyclohexene-1-carboxamide (SS220). Further discussion of natural plant and botanical insect repellents
are the topics of Chapter 14 by Moore and Hill, and Chapter 15 by Gerberg and Novak, respectively.


Spatial Repellency and Attraction-Inhibition of Deet
Deet has produced mixed results as a spatial repellent as was mentioned briefly in Section “Early History
of Spatial Repellents Testing” and Section “Spatial Repellency and Attraction-Inhibition of Catnip Oil”.
In some cases, it is weak or less effective than other compounds,11,33,34,40,42 and in others it is more
effective.14,35,43 One possible explanation is that the concentration of deet needs to reach a specific
threshold in the vapor phase so that the concentration is sufficiently high enough to affect the mosquito
chemosensilla. Otherwise, vapor phase concentrations below this level require landing on a topically
treated surface to result in contact with deet at sufficient concentration to act as a biting deterrent and
therefore be repellent. Dogan et al.25 concluded that deet inhibited the action (attraction) of L-lactic acid,
but did not act as a repellent. Dogan and Rossignol noted that just after topical application of deet,24 test
subjects were still attractive to Aedes aegypti. The results of Bernier et al.40 showed that deet inhibits the
attraction of mosquitoes, but when compared directly at equivalent dosages, it did not function as an
attraction-inhibitor as effectively as catnip oil. It merits mentioning that the individual volunteer whose
odors were used in the Bernier et al. study was relatively less attractive to Aedes aegypti than most
individuals and this may have produced atypical results. Normally, the mixing of deet into the air stream
of a port with human odors does produce a small decrease in the percentage of mosquitoes collected in
the olfactometer trap.44
   Dogan et al.25 reported deet to be attractive in the absence of L-lactic acid; this has been reported
previously for low doses of deet and Rutgers 612.45 We have also observed this in bioassays with our
olfactometer.46 In the absence of attractive odors, the clean airstream in our system produces no response
(no flight activation nor positive anemotaxis) by the mosquitoes. However, with the release of a


q 2006 by Taylor & Francis Group, LLC
82                                                       Insect Repellents: Principles, Methods, and Uses


chemical, upwind anemotaxis and subsequent trapping of a few mosquitoes is observed at times, even
when a compound does not attract mosquitoes when tested in competition against a potent attractant
(e.g., host odors or chemicals based on host odors).44 What appears to be important, at least in the case of
deet, is that wind movement contributes significantly to the ability of a compound to repel (or perhaps
more appropriately, inhibit host finding) in the vapor phase in both a controlled setting and in the
field.43,44,47


Attraction-Inhibition by Linalool and Related Compounds
Alcohols are widely known to repel mosquitoes. For example, citronellol (3,7-dimethyl-6-octen-1-ol),
and it’s related aldehyde analog, citronellal mentioned in Section “Early History of Spatial Repellents
Testing”, exhibit spatial repellency of Aedes aegypti in laboratory bioassays.14 In fact, essential oils, e.g.,
citronella Cymbopogon nardus, were the most commonly used repellents prior to the 1940s.37
Interestingly, citronella oil contains primarily geraniol; however, citronellol and citronellal were reported
as the active ingredients leading to repellency.48,49 Linalool (3,7-dimethyl-1,6-octadien-3-ol) is a water-
insoluble alcohol that is a colorless liquid and is used commonly by the perfume and cosmetics industry
because of its appealing flowery odor. It can be found naturally in such sources as apricots, carrots,
lavender, cardamom and marjoram. Human inhalation of this compound is known to produce sedation,
and it has been shown to suppress the voltage-gated currents in newt olfactory receptor cells.50 Birkett et
al.51 reported that linalool produced significant electroantennogram (EAG) responses in four species of
biting flies, and reduced the upwind (positive) anemotaxis in laboratory wind tunnel studies.
   Linalool has two optically active isomers; researchers have found the (S)-(C)-enantiomer to be the
better attraction-inhibitor.52 Using a dual-port triple-cage olfactometer,46 Kline et al. examined the
impact of linalool, dehydrolinalool (3,7-dimethyl-6-octen-1-yl-3-ol), and deet on the host-seeking ability
of laboratory-reared Aedes aegypti.53 Compared to dehydrolinalool and deet in competitive bioassays,
linalool was the most potent inhibitor (competitive bioassays are defined in Section “Considerations in
the Experimental Design”). An important finding of this work was that the release of linalool resulted in
two observable effects on mosquito behavior. The first effect was that fewer mosquitoes in the cage were
activated to flight from concomitant release of attractant plus linalool in the airstreams of separate ports
of the dual-port olfactometer. This indicated that vapor phase linalool acted as an attraction-inhibitor by
preventing some of the mosquitoes from detecting the normally attractive odors. The second observable
effect was that of the mosquitoes that were activated to flight, fewer than normal numbers of these were
able to locate the odor source. This indicated that even though some mosquitoes could detect the presence
of attractive odors, they were not as capable of orienting towards and, thus, locating the odor source.


Human-Produced Compounds That Affect Host-Seeking
The skin surface of humans differs greatly from that of other animals. Except for a few specific localized
areas, human skin normally ranges from pH 4.2–6.0 due to the abundance of fatty acids that are
present.54,55 In addition to carboxylic acids, skin also has high levels of triglycerides and squalene;
however, it is the acids that contribute largely to the types of microbes that can exist on skin.56,57 Humans
are the only animal to exhibit acne vulgaris, and within the comedo of acne, there are high levels of fatty
acids.58 The distribution of saturated fatty acid molecular sizes are clustered in the C12–C20 and C21–C30
ranges, of which C16 and C18 in the former and C24 in the latter are present in the highest relative
abundance.59 The most abundant unsaturated fatty acids are palmitoleic (9-hexadecenoic), oleic
(9-octadecenoic) and linoleic (9,12-octadecadienoic) acids.
   While some studies have focused on endogenous lipid production, others have focused more on the
end products, or volatiles that are released by metabolic activity via respiration through the skin, or from
degradation of skin surface compounds by microbial action. Sastry et al.60 assembled a comprehensive
treatise of human-produced compounds, covering how these compounds can be used in the diagnosis of


q 2006 by Taylor & Francis Group, LLC
Natural Compounds That Inhibit Mosquito Host-Finding Abilities                                                      83


diseases and in the interpretation of human metabolism. In their review of the subject, they highlighted
studies that identified chemically underivatized compounds, such as that of Ellin et al. who reported the
identification of over 130 compounds in a study of total human body emanations.61 Among the high
levels of acids, there were also significant volatile constituents identified that consisted of alcohols,
ketones, aldehydes and other chemical compound classes. For example, acetone and isoprene (2-methyl-
1,3-butadiene) were two of the most abundant components, emitted at rates of 240–470 and 251–425 mg/h,
respectively, for the three subjects that were examined by Ellin et al.61 In a later study, Naitoh et al.
determined the release rate of acetone from human skin and reported a range as 80–800 pg/cm2 min.62 In
our own studies, we knew of one individual (volunteer A in Table 4.1) who consumed alcohol regularly.
This subject was consistently the most attractive to Aedes aegypti of the six subjects who participated in
this study. Of the most volatile emanations quantified, the most attractive individual (A) produced the
highest level of acetone, ethanol, and methanol. Shirai et al.63 reported that landings of the Asian tiger
mosquito, Aedes albopictus increased after consumption of a beverage containing ethanol. They
measured both skin temperature and the ethanol in the perspiration of human subjects, but they did
not find a relationship between either of these two variables and the landing rates.
   Similar to Sastry et al.60 we also prefer the analysis of volatiles without chemical derivatization for the
identification of human skin (and other host) emanations that may affect mosquito host-seeking.64
Mosquitoes detect volatile host-finding cues in the gas phase, so we believe that minimization of
complexity in the sampling process will tend to cause the least change or bias toward the compound
classes and proportions of each chemical detected. The relative abundances of many of the volatile
compounds have a significant impact on the overall attraction process. Additionally, it is important to
avoid comparing too closely human and mosquito olfaction. The odor of human perspiration that we
smell is due inpart to saturated and unsaturated C6–C11 acids and one of the most abundant odiferous
compounds is (E)-3-methyl-2-hexenoic acid.60,65,66 The sensitivity of humans to the odor of these
compounds does not necessarily imply that these same compounds have a role in mosquito host finding.



TABLE 4.1
Comparison of Volatile Compounds Emanated from Six Different Humans
                                                           Human Subject
Compound
(Class)                      A             B                C                 D                 E                F

(Aldehydes)
Acetaldehyde              160            83                74                52               190             172
2-Methyl-2-                 2.2           1.4               6.5               1.5               8.3             7.8
   propenal
2-Methylbutanal              2.5          1.3               0.92              0.87              4.6             4.7
Hexanal                      5.4          6.2               8.1               6.4              29              38
(Ketones)
Acetone                   900            50                24                45               168             200
2-Butanone                  1.3           0.30              0.39              0.33              1.0             0.82
2,3-Butanedione             2.2           1.4               6.5               1.5               8.3             7.8
(Alcohols)
Methanol                  638             6.9               4.1               8.1              13              13
Ethanol                   638           219                 4.1             117                18              45
(Sulfides)
Carbon disulfide              0.38         0.12              0.13              0.33               1.7            0.55

The headspace of forearm emanations was collected in a Tedlar bag and analyses conducted by microscale purge and trap
GC/MS. Values are in parts per billion by volume (ppbv).
Source: From M. M. Booth, Unpublished results, 1997. With permission.


q 2006 by Taylor & Francis Group, LLC
84                                                       Insect Repellents: Principles, Methods, and Uses


   Bernier et al.67 used gas chromatography/mass spectrometry (GC/MS) of compounds adsorbed and
then thermally desorbed from glass beads to identify 277 compounds present on the skin of humans.
They used columns with different stationary phase polarities to perform the chemical separation of
samples collected from four subjects (males ranging from 26 to 61 years in age). The composition of
emanations was qualitatively similar for all subjects, but quantitative differences were readily observed.
This study provided the groundwork to explore chemical differences between individuals who
represented the extremes of low and high attractiveness to biting mosquitoes. The same study
also examined day-to-day chemical changes correlated to changes in laboratory measured mosquito
attraction for a single individual.68 In the comparison of two different subjects, the individual who was
more attractive to Aedes aegypti had, on average, higher levels of lactic acid, butanone, 2-pentanone,
3-pentanone, and 6-methyl-5-hepten-2-one. The less attractive host had a higher level of methylpentanol,
1,3-butanediamine, capric acid (decanoic acid), lauric acid (dodecanoic acid), heptanal, and pelargo-
naldehyde (nonanal). From studies of a single individual, nonanal, 6-methyl-5-hepten-2-one, and
benzaldehyde were less abundant in the emanations on the day that the residuum was more attractive
to mosquitoes. Those individuals who were less attractive to Aedes aegypti tended to have the highest
concentrations of aldehydes, particularly nonanal, on their skin.68,69 Thus, aldehydes appear to have an
important role in the balance of attraction and inhibition.
   Human fingerprint residues have been examined to identify gender-specific and age differences in the
lipids.70 Hexadecenoic, palmitic (hexadecanoic), and octadecenoic acids were among the most abundant
acids observed, in agreement with Ansari et al.59 Although these three acids occur at higher relative
abundances in males compared to females, the differences were not statistically significant. Curran
et al.71,72 examined male and female odors over time and after exercise. They described a classification of
detected compounds based upon the origin of the odors. “Primary odors” were comprised of emanations
that were present regardless of sampling date or time. Compounds that originated from dietary or
environmental factors were considered secondary odors. Tertiary odors were those attributable to
exogenous factors that resulted in adherence of a chemical to the outer layer of the skin. Using this
terminology, the base attraction of mosquitoes to human hosts would be associated with the primary odor
components, with some differences possibly found in the secondary odors and less likely in the tertiary
odors. Finally, before focusing on the specific compound classes in human emanations, it is interesting to
note the similarities of constituents for skin compounds compared to those found in the oral cavity, urine
and alveolar breath.
   Oral odors are comprised primarily of sulfides, ethanol, diacetyl (2,3-butanedione), acetone,
acetaldehyde, and methyl mercaptan (methanethiol).60 Acetaldehyde and other aldehydes are also
detectable in blood and breath.73,74 Many of the short-chain ketones, acids, and hydroxy acids, such as
L-lactic acid, are also present in human urine. Breath has been reported to contain hundreds of detectable
compounds,75,76 and many of these constituents overlap with those present in blood,60,77 urine,60 and on
the skin.67,72 It is fairly obvious that exhaled breath contains large quantities of carbon dioxide and this is
one of the most universally known behavioral activators and trap attractants for mosquitoes.78–82
However, breath also contains compounds that inhibit the host-seeking response, as was shown for
Anopheles gambiae.83 Therefore, in addition to known Aedes aegypti attractants (e.g., acetone, dimethyl
disulfide (DMDS), and 2-pentanone) in human breath, there are also attraction-inhibitors, e.g.,
nonanal.75,76 From attraction studies, it was evident that certain combinations of chemicals and
classes of chemicals when combined with L-lactic acid resulted in blends with much lower than expected
attraction of mosquitoes in laboratory bioassays.64,84,85 For example, in Bernier et al.84 some branched
ketones and aldehydes that were combined with L-lactic acid resulted in attraction responses that were
less than that of L-lactic acid alone (26% in this study). Some of these specific compounds and functional
groups are discussed in greater detail in the next few subsections. While generalizations can be made
about specific compounds and their ability to attract and inhibit, it is crucial to keep in mind that the
reported behavioral effect is heavily dose-dependent. Specifically, some compounds that attract at low
vapor-phase concentrations may inhibit, arrest, or repel insects at higher concentrations [viz. the response
of deet described in Section “Spatial Repellency and Attraction-Inhibition of Deet”].


q 2006 by Taylor & Francis Group, LLC
Natural Compounds That Inhibit Mosquito Host-Finding Abilities                                           85


Attraction-Inhibition by Carboxylic Acids
Saturated and unsaturated fatty acids are abundant in skin emanations. Other substituted acids such as
L-lactic acid are also present at relatively high levels, and dicarboxylic acids can be detected as some of
the constituents deposited on handled glass beads.67 L-lactic acid is a hydroxy acid that is expected to be
present at substantial levels because it is formed in the body from metabolism of proteins and
carbohydrates under anaerobic conditions. Ellin et al. also detected another important metabolic
product, pyruvic acid, which is an oxo acid that plays a vital role in human metabolism under aerobic
conditions.61
   The initial discovery that fatty acids resulted in “repellency” (inhibition of host-seeking in
bioassays) was reported by Skinner et al. as noted in Section “Spatial Repellency and Attraction-
Inhibition Research”26 Examination of the volatile acids used in blends developed to attract Aedes
aegypti led to discoveries about the compounds that inhibit host-seeking, specifically that addition of
some saturated acids to blends decreased the attraction. Bosch et al. used a Y-tube olfactometer to
demonstrate that combinations of L-lactic acid and either butanoic (C4) or any of the C9–C12 acids
resulted in a composition that did not produce a significant increase in the attraction of female Aedes
aegypti compared to the attraction to L-lactic acid alone.86 For binary blends of L-lactic acid and either
propanoic (C3) or pentanoic (C5) acids, they observed that addition of either undecanoic (C11) or
myristic (tetradecanoic) (C14) acids to this blend resulted in a significant decrease in attraction.
Smallegange et al. reported that a blend of 12 carboxylic acids was repellent against Anopheles
gambiae when tested alone or with L-lactic acid.87
   Constantini et al. reported that the electrophysiologically active acids that produce the odor in human
perspiration, such as Z- and E-3-methyl-2-hexenoic acid, and 7-octenoic acid, repelled or masked the
presence of attractants, and that these may be involved in the avoidance of nonpreferred individuals for
blood meals.88 These findings and those of Bosch et al.86 provide compelling support to the view that as
the concentration of constituents in the human odor profile is perturbed greatly, it can result in host-
avoidance behavior by mosquitoes.
   Reifenrath indicated that acids in the C6 to C8 range coupled with C8–C12 acids were repellent to
arthropods, and that binary combinations of octanoic (C8) and nonanoic (C9) acids, or the tertiary
combination of C8–C10 acids effectively prevented host location.89 Reifenrath examined repellency of
Aedes aegypti by treating gauze or polyester film with each acid applied at 0.3 mg/cm2. These
experiments indicated that 2-pentenoic, 2-octenoic, 3-methyl-2-octenoic, nonanoic, decanoic, and
undecanoic acids were the most effective. Topical tests on human skin showed that the most
repellent compounds were 4-methyloctanoic, 3-methyl-2-octenoic and nonanoic acids, implicating the
most repellent compounds as those that contain 9 carbons and to some extent 8 and 10 carbons
[viz. nonanal discussed throughout this chapter, but also the C8 and C10 carbon compounds such as
linalool, citronellol, citronellal, dehydrolinalool in Section “Spatial Repellency and Attraction-
Inhibition of Deet” and Section “Attraction-Inhibition by Linalool and Related Compounds”, geraniol
in Section “Early History of Spatial Repellents Testing”, and Z-4-decenal and octanal in Section
“Attraction-Inhibition by Aldehydes”].

Attraction-Inhibition by Aldehydes
Aldehydes have received attention recently because they have been identified as the repellent compounds
in the emanations of the crested auklet (Aethia cristatella).90–93 Three of the four reported repellents are
aldehydes, hexanal, octanal, and Z-4-decenal, and one is an acid, hexanoic acid. Additional discussion of
these compounds and chemical defenses of birds and other vertebrates is found in Chapter 3 by Weldon
and Carroll. Although nonanal was not identified in emanations of the crested auklet, it has been reported
as a major constituent of emanations from the whiskered auklet (Aethia pygmaea).91 Nonanal appears to
be detected not only by mosquitoes, but other blood-feeding arthropods as well. Steullett and Guerin
demonstrated that numerous aldehydes, including hexanal, heptanal, nonanal, benzaldehyde, and
methyl-substituted benzaldehydes stimulated tarsal chemoreceptors of the tick Amblyomma variegatum,


q 2006 by Taylor & Francis Group, LLC
86                                                       Insect Repellents: Principles, Methods, and Uses


another arthropod that relies at least in part on chemical cues for host location.94 Guerenstein and Guerin
identified nonanal as the compound that elicited an electrophysiological response from a receptor on the
basiconic sensillum of triatomine bugs (Triatoma infestans).95 In that study, nonanal was also identified
chemically by GC/MS in the extracts of sheep wool and chicken feathers. The unsaturated and
diunsaturated aldehydes tested in their study did not produce an electrophysiological response, nor
did other C9 compounds, including nonanoic acid, 2-nonanone, and nonanol. Heptanal and octanal also
produced linear responses in the sensillum cells, but other saturated aldehydes (C6, C10–C12) did not.
Interestingly, researchers have previously observed a linear correlation of attraction and repulsion to the
concentration of aliphatic aldehydes in blowflies.37,96
   Aldehydes are commonly reported in residue from human skin; these are predominantly in the C6–C10
range.72 Haze et al. documented that the concentration of 2-nonenal, an unsaturated analog of nonanal, is
related to the age of an individual with higher levels observed in males over 40-year-old and that all
subjects produced detectable quantities of C6–C10 saturated aldehydes in this study.97 In contrast, Curran
et al. was able to detect 2-nonenal in females and in individuals less than 25-year-old.71,72 Curran et al.
reported that the C8–C10 aldehydes were detectable in 88% of their subjects,71 and Zhang et al. also
reported these C8–C10 aliphatic aldehydes.98 A better understanding about the role of C8–C10 aldehydes
in the mosquito host-finding process may benefit from experiments comparing the relative attractiveness
of subjects who have high or low concentrations of these compounds on their skin.
   Bernier et al. used microscale purge and trap GC/MS to identify aldehydes from butanal to
undecanal, with nonanal as the most abundant in this series.64 The cryo-focused GC/MS analysis of
glass beads allowed the detection of propanal (C3) to nonanal (C9), including branched and
unsaturated analogs of these compounds. The more volatile aldehydes are partly responsible for off
odors in spoiled meat,99 while the less volatile, such as octanal, nonanal, and benzaldehyde have a
more pleasant floral aroma. Endogenous aldehydes that are oxidized from their respective acids are
hexanal from linoleic and arachidonic acids; heptanal from palmitoleic acid; and nonanal from oleic
acid.74,100 As noted earlier in this chapter, these acids are the some of the most abundant in human
emanations.59,67,70 By analogy, this may partly explain the abundance of these specific aldehydes in
human emanations.64

Attraction-Inhibition by Ketones
Acetone is the most abundant ketone in human odors (see Table 4.1).61,62 One mechanism for
endogenous production of this compound is from fat metabolism.62 In addition to acetone, numerous
2- and 3-substituted ketones, as well as cyclohexanone, have been reported in human odors.61
Unsaturated ketones have also been found in the residue of more than 50% of human subjects.72
Birkett et al.51 reported that when the unsaturated branched ketone, 6-methyl-5-hepten-2-one, was
applied to cattle, it reduced the attraction to biting flies.
   Saturated ketones, particularly in the C7–C12 range have been found to inhibit mosquitoes.101 The
combination of L-lactic acid with either acetone or butanone, the smallest and most volatile of the
saturated ketones, produced synergistic attractant blends for Aedes aegypti.101,102 However, as larger
saturated ketones within the series, like pentanone (C5) and hexanone (C6), are blended with L-lactic acid,
the attraction drops from synergistic to additive, and then results in inhibition of attraction for blends with
heptanone (C7) through dodecanone (C12). When chain lengths exceed C12 in the ketones (C10 in acids
and aldehydes) it is expected that the volatility decreases below a threshold level such that the vapor
phase concentration is so low that the impact on host-seeking disappears. This effect was also evident
when researchers examined the repellency of alcohols larger than decanol.37

Attraction-Inhibition by Alcohols
Bernier et al.67 identified unsaturated and saturated alcohols from butanol to heptadecanol were in human
skin. Ellin et al. also observed a number of these alcohols and ethylene glycol.61 Glycerol also was


q 2006 by Taylor & Francis Group, LLC
Natural Compounds That Inhibit Mosquito Host-Finding Abilities                                           87


reported in both studies; it is a major breakdown product of bacterial action on triglycerides.56 Phenol
was produced by all human subjects in the study of Curran et al.72 In addition to amides like deet,
aliphatic alcohols have been popular historically as insect repellents, e.g., the series of decanol (C10)
through tetradecanol (C14),103 and Rutgers 612.104,105 Dogan and Rossignol examined various fragrances
and compositions that contained alcohols such as geraniol and dimethyl cyclormol (hexahydrodimethyl
methanoinden-5-ol) and found these to either inhibit or repel mosquitoes in a modified Feinsod–
Spielman olfactometer.24
   In contrast to the well known attractant 1-octen-3-ol,106,107 several related, more volatile unsaturated
alcohols, including linalool will inhibit attraction by Aedes aegypti in laboratory bioassays.53,108 Yet,
other unsaturated alcohols, such as geraniol,24 or diols that are similar in structure, such as 7-octen-1,2-
diol, have little or no effect on the host-seeking of Aedes aegypti.108 The examination of compounds from
cattle to identify compounds that affect host location by five species of biting flies revealed that 1-octen-
3-ol and 3-octanol were attractants in wind tunnel studies. In contrast, these compounds reduced the
number of biting flies on cattle in the field.51 This may be a case where the normal host odor profile is
perturbed so greatly by the added volatiles that host avoidance by the insects is the net result.

Attraction-Inhibition by Compounds of Other Classes
Researchers have documented ammonia and a series of amines from methylamine to butylamine in
human emanations.61,109 Ammonia is formed through amino acid catabolism, and along with urea and
uric acid are the three main nitrogen-containing compounds excreted by animals.110 Ammonia has been
demonstrated to attract Aedes aegypti and Anopheles gambiae at low concentrations,111,112 and to deter
feeding at higher doses.87,113 In addition to these alkaline substances, Bernier et al. also reported a
substantial number of hydrocarbons and heterocyclic compounds present in human emanations.67 Some
of these are currently being tested in our laboratory to determine if they play a role in the host-seeking
behavior of mosquitoes. Bernier et al. identified some sulfides and some 1-chloroalkanes in human skin
emanations.67 Sulfides and chlorides have not been observed to inhibit the host-seeking of Aedes
aegypti101; however, larger sulfides, chlorides and other alkyl halides have not yet been tested as
attraction-inhibitors.
   If we attempt to make a general statement regarding compounds capable of attraction-inhibition, then
we could base this upon the presence of oxygen in the molecule, as Bunker and Hirshfelder noted for
“good” repellents in 1925.114 Roadhouse later noted that many effective repellents contain nitrogen.115
However, this should be kept in perspective because many compounds contain oxygen, nitrogen, or both
and do not show effective repellency or inhibition of mosquito host-seeking.115,116




Identification of Host-Produced Allelochemicals
Numerous techniques exist to sample, collect, concentrate, chemically separate, and identify compounds
in host emanations. There are benefits and drawbacks to each choice. One needs to consider all of these
factors carefully when selecting the approaches to solve a complex problem, such as the identification of
chemicals that affect mosquito host-seeking behavior. It is important to realize that a single method in
any of these processes is likely to prove inadequate for the resolution of a complex situation involving
potentially numerous compounds that can span a wide range of differing compound polarities and
volatilities. For example, multiple preconcentration techniques may be needed to provide comp-
lementary information, and multiple chromatography columns with stationary phases of different
polarities may need to be used to resolve all of the compounds.64,67 By combining information from
different types of analyses, the total chemical profile will be more complete. Some of the more recent
techniques applied to the analysis of human emanations have either involved solvent extraction,


q 2006 by Taylor & Francis Group, LLC
88                                                       Insect Repellents: Principles, Methods, and Uses


deposition onto glass beads, or the use of solid phase microextraction (SPME) fibers as noted in Section
“Analysis of Human Emanations”.


Analysis of Human Emanations
The analytical method of choice for almost all comprehensive chemical analyses of volatile human body
emanations has involved chromatographic separation followed by mass spectrometric detection, e.g.,
GC/MS, whether the emphasis is on skin emanations, breath, urine, blood, oral cavity, or the total
composite of emanations from an entire person.60,61,64,67,68,70–72,75,76,98,109,117,118 Mass spectrometry
allows for the identification of compounds based on the fragmentation pattern of compounds. These
patterns consist of differing intensities of ions (technically, as a ratio of mass to charge, m/z) that result
from bombardment of sample molecules by electrons. There are various types of mass analyzers for mass
spectrometers, but the most common for these studies are either magnetic/electric sector or quadrupole
instruments because they provide mass spectra that is most similar, and therefore the most easily
matched, to mass spectra in existing computerized mass spectral libraries.
   In many of these analyses, hundreds of compounds are present. Therefore, separation must be
effected prior to mass spectral analysis. This is accomplished by column chromatography. Over the
last few decades, the columns employed for this purpose have improved greatly. They are more stable
due to better phase bonding, allow greater sample capacity, and are capable of better resolution.
Despite all of these improvements, exposure to air and/or extreme hot or cold temperatures still easily
degrade the GC column stationary phase. In general, the more polar that the column phase is, the more
constrained that it will be with respect to temperature limits than a column that has a relatively
nonpolar stationary phase.
   Soxhlet extraction, commonly used for fat and oil extraction, followed by GC/MS was used to
concentrate and identify volatiles from foot stockings.119 Bernier et al. used glass beads to collect
emanations for subsequent thermal desorption into a GC injection port.64,67,68 In doing so, the problems
from the high water content of perspiration was avoided. This remedy is significant because loading
water onto gas chromatography columns is detrimental to the stationary phase. Asano et al. used glass
beads followed by solvent extraction of compounds from the beads to study fingerprint residues.70
Headspace GC/MS was used to analyze age-specific male individual odor differences,97 and SPME has
been used to collect and concentrate skin volatiles for subsequent identification and quantitation by GC/
MS.71,72,98,118 In the work of Curran,72 supercritical fluid extraction (SFE) was used as a pretreatment to
reduce or eliminate some of the background compounds in the gauze, which was necessary for
quantitation of human emanations because a number of human emanations also are measurable in the
background contaminants from the gauze. This innovative pretreatment reduced exogenous compounds
and allowed them to achieve accurate quantitative results.


Merging Chemistry and Sensory Physiology
One of the earliest reports of detection of electrical impulses along the nerves was that of Adrian, who in
1930 recorded the discharge of the caudal nerve in the caterpillar.120 Electrophysiological studies of
these impulses based upon selection of innervated nerve has contributed significantly toward an
understanding of which compounds and which sensory organs may factor into the process of host
attraction or other behavioral responses. Electroantennograms provide an ideal screening tool for
compounds that insects detect, although it does not reveal whether this detection may lead to attraction,
avoidance, repellency, or other behaviors. Single-cell recording can determine precisely which receptor
organ a compound stimulates. In the early days of these techniques, Roelofs used GC to separate
compounds and coupled the resulting sample stream with EAG to identify pheromones and compounds
that are synergists and inhibitors for pheromones.20,121,122


q 2006 by Taylor & Francis Group, LLC
Natural Compounds That Inhibit Mosquito Host-Finding Abilities                                         89


   Combination of these techniques with gas chromatographic separation is a powerful approach to
analyze complex samples for the compound (peaks) that produce bioactivity. By either routing the
sample via column splitting to both instruments (GC-EAD and GC/MS), or simply injecting the same
sample on separate instruments with the GCs configured similarly, it is possible to identify and thus focus
on a smaller set of compounds that are bioactive in a sample that may contain hundreds of compounds.
Recent applications of this technique can be found in the report of Cork et al.123 and related studies
involving mosquitoes, such as Anopheles gambiae.117,124–126 Working with Anopheles gambiae
antennae, Cork and Park examined extracted human skin compounds and identified the most abundant
acids as acetic, heptanoic, and hexadecanoic acids, whereas the EAG responses were greatest for formic,
pentanoic, butanoic, propanoic, acetic, and hexanoic acids, all of which were more intense than the
response to the 1-octen-3-ol standard.117 Constantini et al.88 examined EAG responses of common
human-produced odiferous compounds in sweat and evaluated their impact on host-seeking using a wind
tunnel for bioassay as reported in Section “Attraction-Inhibition by Carboxylic Acids”. Other successful
recent electrophysiological studies with additional arthropods have been reported for tsetse flies,106,127
ticks,94 and the New World screwworm.128


Current State and Future Directions of Host Odor Research
Section “Attraction-Inhibition by Carboxylic Acids” described a recent example of the application of
allomonal odors in which Reifenrath added carboxylic acids to host emanations to make the normally
attractive host appear to have a different chemical profile.89 The result was that the host was much less
attractive to biting insects. At present, host-odor research continues with increased emphasis on
understanding how kairomones and allomones function together to mediate the overall behaviors in
the host-seeking process of arthropods. Some of the studies involve human hosts for anthrophophilic
species that transmit malaria, such as Anopheles gambiae and Anopheles albimanus, or for those that
transmit dengue and yellow fever, such as Aedes aegypti. Other studies center on birds, the preferred
hosts of ornithophilic species such as Culex tarsalis, Culex pipiens quinquefasciatus and Culex
nigripaplus, which are vectors of West Nile Virus (WNV) in North America. Studies involving
animals as sources of chemicals that may attract arthropods, repel them, or inhibit the attractive
emanations of a host is the subject of Chapter 3 by Weldon and Carroll.




Laboratory Bioassays of Spatial Repellents and Attraction-Inhibitors
The information derived from a particular study depends heavily upon the bioassay because the
construction design of the device and the procedure used determine the behaviors that are assessed.
The subject of this section is the common laboratory bioassay devices that have been used to produce
many of the results described in this chapter. Additional coverage of olfactometer design and usage can
be found in Chapter 9, written by Butler.


Olfactometers for the Assessment of Spatial Repellents
One can trace the design of dual-port olfactometers back to the 1930s.129–131 Early USDA spatial
repellency studies employed a similar style single-cage olfactometer modified to hold mesh netting in the
trap ports.132,133 Researchers conducted tests by passing air over a human arm and through a trap into the
cage where 100 mosquitoes were located. The mesh cotton netting within the traps was either treated or
untreated (as the control) with candidate spatial repellents. The test period was 5 min and netting was
tested every other day until two successive trials resulted in O10% of the test mosquitoes trapped in the


q 2006 by Taylor & Francis Group, LLC
90                                                       Insect Repellents: Principles, Methods, and Uses


port with human odors. Thus, effectiveness of compounds was evaluated based on days of duration
of repellency.
   Skinner and colleagues also used a dual-port olfactometer, operated in noncompetitive and
competitive modes (see Section “Considerations in the Experimental Design” for a description of
these modes) to compare two treatments consisting of human lipid fractions.26–29 Researchers also
quantified the repellency by the location of mosquitoes in the test and control ports after allowing the
insects to fly upwind and select a port. In the noncompetitive mode, they compared the ratio of
mosquitoes captured in the control port to the number in the sample port and the greater the ratio, the
higher “repellency” according to this index. Because this experiment did not allow contact between
mosquito and attractant, we believe that operation of the olfactometer in this way measured the
attraction-inhibition of specific compounds. Dogan and Rossignol modified a Feinsod–Spielman
olfactometer by constructing an additional chamber to allow measurement of “repellent” response
based on insects moving away from the treatment.24 Recently, Grieco et al.134 designed a modular
bioassay device which can be assembled to provide a system to screen contact irritancy of candidate
chemicals, and reconfigured in a manner to allow assessment of spatial repellency. The movement of
chemical inside each of these olfactometers is accomplished by convection and diffusion, without
supplementation of a stream of air.


Olfactometers for the Assessment of Attraction-Inhibitors
Barrows first used the Y-tube olfactometer in studies of flies.135 Geier et al. and Bosch et al. have
used recent models to test mosquito responses.86,136 The triple-cage dual port olfactometer constructed
by Posey et al.46 and used in our laboratory is based on older designs described in Section
“Olfactometers for the Assessment of Spatial Repellents”.132,133 Because all of these olfactometer
designs employ two ports, they can be used to measure attraction response to either a single treatment
versus a control, or to two individual treatments in competition. Reifenrath used a Feinsod–Spielman
olfactometer to measure the repellent effect imparted by carboxylic acids on human odors.89 The
design of this olfactometer allowed odors to pass through a linear arrangement (similar to Grieco
et al.134) of chambers by (in this case) a fan that drew the odors upward into the top chamber. Prior to
conducting a test with human odors, mosquitoes were released in the top chamber, and allowed to
distribute between the two chambers. After human odors were introduced through the bottom of the
olfactometer, the mosquitoes that flew from the upper chamber down to the lower chamber were
counted as responding to an attractive stimulus. Those remaining in the upper chamber were
considered “repelled.” Again, this may not be truly indicative of repellency—it can be reasonably
argued that mosquitoes that remain in the top part could be inhibited from detection of potentially
attractive odors, or simply nonresponding. Provided that a standard is assessed with this design, then a
reduction in attraction can be attributed to either the effect of a spatial repellent or attraction-inhibitor.
   As noted above, the standard design of the Y-tube, or dual-port olfactometer (without modification
inside the traps) is perhaps not the best bioassay system to measure spatial repellency because one
cannot discern whether mosquitoes left in the original position were nonresponding or truly
“repelled.” Additionally, it remains unclear how to characterize mosquito behavior response to a
treatment when they respond by positive anemotaxis into the clean air (control) port. For occasions
that we observe this phenomenon, we always follow the test by examining the response to the
individual control apparatus with no treatment in the opposite port to test for contamination of either
the port or apparatus.46

Considerations in the Experimental Design
In a dual-port olfactometer, there are two common modes by which the device can be operated and this is
based on the number of treatments. A noncompetitive assay is arranged so that there is a treatment in one


q 2006 by Taylor & Francis Group, LLC
Natural Compounds That Inhibit Mosquito Host-Finding Abilities                                             91


port compared to a second port containing the blank control (all apparatus used to hold the treatment, but
with the treatment absent).53 The advantage of this mode of operation is that it allows a means to compare
attraction to treatments based upon a measurement of “inherent” or “absolute” attraction level, without
possible interference or complications in mosquito behavior that may arise from interaction with odor
released from a treatment in the second port. This approach is commonly used in our laboratory to screen
for attractants and inhibitors.
   In a competitive assay, one treatment chemical is compared simultaneously to another to provide
information on the relative attraction of one treatment to another.53 It also can provide information on the
interfering effects from an inhibitor released in the opposite port and provide an indication about whether
the inhibitor functions best when released at close range to the attractants, or if it can be released from
another location and still be effective. The advantage of this technique is that it may provide a closer
approximation to field situations where attractants or inhibitors must function in a complex situation
against mosquitoes in competition with many other odors. Olfactometers that are used to assess
the biological activity of candidate attractants have allowed the development of the human odor
blends,68,84–86,101,102,111,136,137 such as L-lactic acid and carbon dioxide,138 L-lactic acid and ammonia,111
                                                86
L-lactic acid and specific carboxylic acids, and a three-component blend of L-lactic acid, acetone, and
                     84,85,102,137
dimethyl disulfide.
   The development and use of a standard that has high attraction efficiency, reproducibility, and
stability is important when conducting experiments to identify attraction-inhibitors. The use of such a
blend has applicability to in vitro repellent experiments by obviating the need for volunteers to
participate in in vivo studies. A standard chemical blend of attractants removes the variability inherent
in the use of live hosts. Not only do individuals vary in their attractiveness and compound abundances
detected, but a single human can vary substantially in both biological activity and compound
abundances in their profile from day-to-day.69 However, caution must be exercised in the interpretation
of results from trials in which blends of attractant chemicals are used because they represent an
approximation of a host. These mixtures consist of only a small number of kairomones and it is
reasonably certain that of the hundreds of compounds emanated from human skin and some of the
important attractants still remain unidentified. Most humans and skin extracts are still more attractive
than our best synthetic blends when tested competitively in laboratory bioassays.136,137 One of our
bioassay protocols for attraction-inhibition involves comparing the response of a standard blend to the
response of the same blend, delivered at the same dose but with a candidate attraction-inhibitor added to
it. In other cases, the response of the candidate plus another known attractant like L-lactic acid is
compared to the response to L-lactic acid alone when looking for synergism. Again, this method of
testing attraction-inhibition may be even further removed from reality than using human odors or a more
complex blend with higher attractiveness because as noted above, the human odor profile is significantly
much more complex.

Correlating Small- and Large-Scale Laboratory Results to Field Experiments
One concern with results from laboratory bioassays is that they may not correlate well to the performance
in the field. Laboratory bioassays are conducted under well-controlled conditions with the temperature,
humidity, wind speed, and other variables controlled as needed. Although bioassays can involve
movement in space, this movement is often confined. At best, the movement is in essence two
dimensional, if not actually closer to a one-dimensional situation in which the mosquito travels linearly
upwind through a tube. Additionally, bioassays in the laboratory may only examine a subset of all factors
involved in host location, even though this may be intended partly by design. Laboratory olfactometers
have a finite length or depth, and thus can best assess only the medium- to close-range stimuli. Finally,
bioassays of this nature are considered to be undiscriminating assays in the treatise of Kennedy because
the overall result, e.g., attraction, is analyzed as a complex of responses, rather than the individual
isolated responses, as would be done in a discriminating assay.21


q 2006 by Taylor & Francis Group, LLC
92                                                      Insect Repellents: Principles, Methods, and Uses




Field Tests and Use of Spatial Repellents and Attraction-Inhibitors
Experimental Design of Field Tests
Use of Large-Cage Experiments and Laboratory-Reared Colony Mosquitoes
Researchers have conducted large cage (9.1 m wide!18.3 m long!4.9 m high, gabled to 5.5 m) studies
to simulate environmental conditions that might be encountered in field studies against natural
populations of mosquitoes. Traps releasing known attractants at specified release rates are placed in
the center of the cage.139 We choose to test with a 2.4 m!2.4 m designed perimeter around the trap. A
wooden stake with an attached attraction-inhibitor releasing device is located at each corner of the
perimeter. An inhibitor release device is attached 0.6 m above ground level to each stake. Both the
inhibitor release device and trap are activated at least 30 min before mosquitoes were released into
the cage, and operated for a specified time period, typically 12 h. At the conclusion of the test period, the
trap collection device is retrieved and landing rates on humans are conducted within the cage at several
established locations outside the 2.4 m!2.4 m perimeter. The landing rate counts are performed in
addition to trap collections to provide a more comprehensive indication of the effectiveness of the
candidate attraction-inhibitor being tested. The benefits to using a large cage, similar to the benefits of
laboratory studies, is that they provide a controlled setting with mosquitoes of known species
composition, physiological and chronological age, and quantity. Furthermore, the escape of mosquitoes
is minimized. However, the environmental conditions inside the cage are similar to those outside, as is
the landscaping within the space. The drawback is that the mosquitoes are not allowed to migrate beyond
the enclosure, as they would be able to do in the wild.

Experiments with Wild Mosquitoes in the Field
One concern with using colony-reared mosquitoes is whether or not they will behave similarly to those in
the wild. Additionally, there are a variety of mosquito species and this composition can vary significantly
during the course of a study. Conduction of field tests against natural populations of mosquitoes is
performed in a similar experimental setup as that used in the large cage studies. A series of 2.4 m!2.4 m
plots can be established with traps, similarly baited as in the large cages, located in the center with the
inhibitor dispensing devices placed on the four corners. A Latin square design can be used with days as
replicates.140 Initially, treatments and controls should be randomly assigned to each plot. The plots
should be located far enough apart to prevent interactions among treatments. The treatments are then
moved to new stations each day until all treatments have been evaluated in each plot at least once. Jensen
et al. has used a variation of this design to evaluate citronella candles in Illinois.141 At each sampling
station in their study, the candles were arranged into an equilateral triangle, 3 m apart, with an individual
measuring efficacy sitting in the center, about 1.5 m from each candle. The individual aspirated
mosquitoes trying to bite exposed legs during four 15-min collection periods using a
mechanical aspirator.
   Another study conducted by Lindsay et al.142 in Canada used eight sampling stations arranged in a grid
separated by at least 10 m. Two of each kind of dispenser were placed at each sampling period on top of
35-cm-high plastic stands 1 m apart. A plastic lawn chair was placed between the plastic stands and
subjects conducted biting counts while seated on the lawn chairs. The subjects were assigned to one of
the eight sampling stations at the beginning of each evening and then rotated through all eight positions
twice each night. Treatments were assigned to positions on the grid such that each treatment was at each
position during the eight-night evaluation. It is important to evaluate each candidate product under a wide
range of field conditions against a diversity of mosquito species, comparing their effectiveness to both
negative (untreated) and positive (deet-treated individual) controls. Recently, Webb et al. used carbon
dioxide-baited light traps and dispersed candidate inhibitors in a 4 unit!4 unit grid with each of the


q 2006 by Taylor & Francis Group, LLC
Natural Compounds That Inhibit Mosquito Host-Finding Abilities                                            93


16 dispensers about 1.5 m above ground.143 Significant repellency was noted for catnip oil, deet, and the
E,Z-dihydronepetalactone isomer from catnip oil.

Use of Stand-Alone Inhibitor-Delivery Technology
Currently, there is a commercial device that is on the market using inhibitor technology based on linalool.
The active ingredient is primarily the (S)-(C)-linalool isomer [as opposed to (R)-(K)-linalool] in
candles and sold under the trade name Concealw.* However, not all chemicals may be amenable to
delivery by candle, and therefore devices similar to another commercial device, the Mosquito Cognitow,*
may be an alternative approach to disperse low levels of inhibitor aerosols into the environment. The
active ingredient is contained in cartridges and is used in a battery-powered device.




Potential Applications of Spatial Repellents and Attraction-Inhibitors
Species-Specific or Species-Exclusive Trapping
At the present time, not enough is known about the concentration-dependent effects of attractants and
attraction-inhibitors and how these chemicals may work or not work on many different mosquito species.
Some inhibitors negatively affect the female mosquito at all concentrations tested, and against all species
we have tested in the laboratory (Anopheles quadrimaculatus, Anopheles abimanus, Aedes aegypti,
Aedes albopictus, and Culex nigripalpus).108 The rationale behind species-exclusive trapping would
likely involve the use of odors based on avian emanations to selectively lure ornithophilic species of
mosquitoes away from opportunistic feeding on a lesser-preferred host, such as humans. There is some
basis for exploring this avenue of research because it has been shown that high (and/or low) levels of
                                                             78,138,144
L-lactic acid are repellent for some species of mosquitoes,             and that specific species exhibit a
                                                 145
strong host-preference based on emanated odors.

Local Area Host-Finding Reduction
One application of inhibitors has already been discussed in Section “Use of Stand-Alone Inhibitor-
Delivery Technology”, i.e., the Mosquito Cognito/Conceal technology. The range of reduction in host
finding is 50–95% with an average of 65% reduction based on tests in Sarasota, Vero Beach, and
Loxahatchee (candles) and Lower Suwannee (candles) wildlife refuges in Florida.146 It is possible that
additional reduction might be achieved with the discovery of additional attraction-inhibitors. Also, it may
be possible to design blends of inhibitors that may function synergistically in their effect, similar to that
observed for chemicals used in kairomone blends that are derived from human odorants.


Local Control Using a Push–Pull Strategy with Attractant-Baited Surveillance Traps
Perhaps one of the greatest benefits to the development of potent inhibitors is the use of these compounds
at a slow release rate to conceal host attractive odors in conjunction with surveillance traps to lure and
trap or kill as a means of a barrier-forming push–pull strategy.147 There are isolated situations, such as
was shown in the work of Kline on Atsena Otie Key in Florida, where a reduction in mosquito biting
incidence can be obtained using traps with attractants only.148 This success is not expected to be possible
in an area where competing host odors are constantly present. However, it is believed that even if there is
a trap containing an attractant lure that is inferior to host odors, a push–pull strategy may overcome this
and allow for local control in small areas.

*
    Registered trademark of BioSensory, Inc., Putnam, CT, USA.


q 2006 by Taylor & Francis Group, LLC
94                                                         Insect Repellents: Principles, Methods, and Uses


Use of Structure–Activity Relationships to Benefit Development of Attraction-Inhibitors
Scientists are exploring the use of quantitative structure–activity relationships (QSAR) as a means to
examine repellents and to discover the structural basis that results in their biological activity.149,150
Furthermore, this approach can be used as a means to predict novel molecular structures that are likely to be
repellent. As attraction-inhibition becomes a more precisely characterized phenomenon, with increased
numbers of inhibitors, dose response studies, and experiments designed to accurately assess inhibition
level, these data should be amenable to QSAR studies. Through QSAR, researchers may also be able to
predict the molecular and electronic properties of chemicals that result in attraction-inhibition. A
comprehensive understanding of the chemicals could, in time, lead to a better understanding of the
function of the odorant receptors. Extensive coverage of approaches to modeling repellents is found in
Chapter 10 by Gupta and Bhattacharjee.


References
      1. F. Herbert, October 8, 1920-February 11, 1986.
      2. H. H. Shorey, Interaction of insects with their chemical environment, in Chemical Control of Insect
         Behavior: Theory and Application, H. H. Shorey and J. J. McKelvey (Eds.), New York: Wiley, 1977,
         pp. 1–5.
      3. P. E. Howse, O. T. Jones, and I. D. R. Stevens, Pheromones and Behavior in Insect Pheromones and
         their Use in Pest Management, London: Chapman & Hall, 1998, pp. 3–6.
      4. V. G. Dethier, L. Barton Browne, and C. N. Smith, The designation of chemicals in terms of the
         responses they elicit from insects, J. Econ. Entomol., 53, 134, 1960.
                               ¨
      5. P. Karlson and M. Luscher, “Pheromones”: A new term for a class of biologically active substances,
         Nature, 183, 55, 1959.
                               ¨
      6. P. Karlson and M. Luscher, The proposed term “pheromone”, Nature, 183, 1835, 1959.
      7. C. I. Bliss, The toxicity of poisons applied jointly, Ann. Appl. Biol., 585, 26, 1939.
      8. R. L. Plackett and P. S. Hewlett, Quantal responses to mixtures of poisons, J. Royal. Stat. Soc., 14B,
         141, 1952.
      9. R. H. Wright, D. L. Chambers, and I. Keiser, Insect attractants, anti-attractants, and repellents, Can.
         Entomol., 103, 627, 1971.
     10. L. B. Browne, Host-related responses and their suppression: Some behavioral considerations, in
         Chemical Control of Insect Behavior: Theory and Application, H. H. Shorey and
         J. J. McKelvey (Eds.), New York: Wiley, 1977, pp. 117–127.
     11. H. K. Gouck, T. P. McGovern, and M. Beroza, Chemicals tested as space repellents against yellow-
         fever mosquitoes, I. Esters, J. Econ. Entomol., 60, 1587, 1967.
     12. J. A. Nolen et al., Method, Apparatus, and Compositions for Inhibiting the Human Scent Tracking
         Ability of Mosquitoes in Environmentally Defined Three Dimensional spaces, Washington, D.C.: U.S.
         Patent and Trademark Office, 2002 (U.S. Patent No. 6,362,365).
     13. C. E. Schreck and D. L. Kline, Area protection by use of repellent-treated netting against Culicoides
         biting midges, Mosq. News, 43, 338, 1983.
     14. R. A. Wirtz, J. D. Turrentine, Jr., and L. C. Rutledge, Mosquito area repellents: Laboratory testing of
         candidate materials against Aedes aegypti, Mosq. News, 40, 432, 1980.
     15. H. Kawada et al., Field evaluation of spatial repellency of metofluthrin impregnated plastic strips
         against mosquitoes in Hai Phong City, Vietnam, Am. J. Trop. Med. Hyg., 73, 350, 2005.
     16. D. R. Roberts et al., Methods of testing and analyzing excito-repellency responses of malaria vectors
         to insecticides, J. Am. Mosq. Control Assoc., 13, 13, 1997.
     17. K. Ujihara et al., Metofluthrin: A potent new synthetic pyrethroid with high vapor activity against
         mosquitoes, Biosci. Biotechnol. Biochem., 68, 170, 2004.
     18. M. J. Klowden, The physiological control of mosquito host-seeking behavior, in Current Topics in
         Vector Research, K. F. Harris (Ed.), Vol. 1, Westport, Connecticut: Praeger Scientific, 1983,
         pp. 93–116.


q 2006 by Taylor & Francis Group, LLC
Natural Compounds That Inhibit Mosquito Host-Finding Abilities                                                 95


    19. J. E. Simpson and R. H. Wright, Area treatment to combat mosquitoes, Nature, 214, 113, 1967.
    20. W. L. Roelofs and A. Comeau, Sex pheromone perception: Synergists and inhibitors for the
        red-banded leaf roller attractant, J. Insect Physiol., 17, 435, 1971.
    21. J. S. Kennedy, Behaviorally discriminating assays of attractants and repellents, in Chemical Control of
        Insect Behavior: Theory and Application, H. H. Shorey and J. J. McKelvey (Eds.), New York: Wiley,
        1977, pp. 215–229.
    22. E. E. Davis, Structure-response relationship of the lactic-acid excited neurons in the antennal grooved-
        peg sensilla of the mosquito Aedes aegypti, J. Insect Physiol., 34, 443, 1988.
    23. S. J. Torr, T. N. C. Mangwiro, and D. R. Hall, Responses of Glossina Pallidipes (Diptera: Glossinidae)
        to synthetic repellents in the field, Bull. Entomol. Res., 86, 609, 1996.
    24. E. B. Dogan and P. A. Rossignol, An olfactometer for discriminating between attraction, inhibition,
        and repellency in mosquitoes (Diptera: Culicidae), J. Med. Entomol., 36, 788, 1999.
    25. E. B. Dogan, J. W. Ayres, and P. A. Rossignol, Behavioural mode of action of deet: Inhibition of lactic
        acid attraction, Med. Vet. Entomol., 13, 97, 1999.
    26. W. A. Skinner et al., Repellency of skin-surface lipids of humans to mosquitoes, Science, 149, 305,
        1965.
    27. H. I. Maibach et al., Factors that attract and repel mosquitoes in human skin, J. Am. Med. Assoc., 196,
        173, 1966.
    28. H. I. Maibach et al., Human skin in relationship to mosquito attraction and repulsion, Conn. Med., 33, 23,
        1969.
    29. W. A. Skinner et al., Human skin-surface lipid fatty acid—Mosquito repellents, Experientia, 26, 728,
        1970.
    30. S. R. Christophers, Mosquito repellents, Being a report of the work of the mosquito repellent inquiry,
        Cambridge (1943–5), J. Hygiene., 45, 176, 1947.
    31. R. N. McCulloch and D. F. Waterhouse, Laboratory and field tests of mosquito repellents, Bull. Council.
        Sci. Indust. Res. Aust., 213, 28, 1947.
    32. USDA, Report of Work Conducted Under Funds Allotted by the Department of National Defense,
        Fourth Quarterly Progress Report, Orlando, FL, 1948.
    33. T. P. McGovern, M. Beroza, and H. Gouck, Chemicals tested as space repellents against yellow-fever
        mosquitoes, II. Carbanilates, benzamides, aliphatic amides, and imides, J. Econ. Entomol., 60, 1591,
        1967.
    34. C. E. Schreck et al., Spatial action of mosquito repellents, J. Econ. Entomol., 63, 1576, 1970.
    35. A. A. Khan and H. I. Maibach, A study of insect repellents, 1. Effect on flight and approach by Aedes
        aegypti, J. Econ. Entomol., 65, 1318, 1972.
    36. T. Eisner, Catnip: Its raison d’etre, Science, 146, 1318, 1964.
    37. L. R. Garson and M. E. Winnike, Relationships between insect repellency and chemical and physical
        parameters—A review, J. Med. Entomol., 5, 339, 1968.
    38. Peterson, C. J. Insect repellents of natural origin: Catnip and osage orange, Ph.D. dissertation, Iowa
        State University, Ames, IA, 2001.
    39. C. Peterson and J. Coats, Insect repellents—past, present, and future, Pestic. Outlook., 12, 154, 2001.
    40. U. R. Bernier et al., Comparison of contact and spatial repellency of catnip oil and N,N-diethyl-3-
        methylbenzamide (deet) against mosquitoes, J. Med. Entomol., 42, 306, 2005.
    41. K. R. Chauhan et al., Feeding deterrent effects of catnip oil components compared with two synthetic
        amides against Aedes aegypti, J. Med. Entomol., 42, 643, 2005.
    42. J. Boeckh et al., Acylated 1,3-aminopropanols as repellents against bloodsucking arthropods, Pestic.
        Sci., 48, 359, 1996.
    43. E. J. Hoffman and J. R. Miller, Reduction of mosquito (Diptera: Culicidae) attacks on a human subject by
        combination of wind and vapor-phase deet repellent, J. Med. Entomol., 39, 935, 2002.
    44. U. R. Bernier, Unpublished data, 1997.
    45. Z. A. Mehr et al., Attraction of mosquitoes to diethyl methylbenzamide and ethyl hexanediol, J. Am.
        Mosq. Control Assoc., 6, 469, 1990.
    46. K. H. Posey, D. R. Barnard, and C. E. Schreck, Triple cage olfactometer for evaluating mosquito
        (Diptera: Culicidae) attraction responses, J. Med. Entomol., 35, 330, 1998.


q 2006 by Taylor & Francis Group, LLC
96                                                         Insect Repellents: Principles, Methods, and Uses


     47. E. J. Hoffman and J. R. Miller, Reassessment of the role and utility of wine in suppression of mosquito
         (Diptera: Culicidae) host finding: Stimulus dilution supported over flight limitation, J. Med. Entomol.,
         40, 607, 2003.
     48. R. R. Painter, Repellents, in Pest Control: Biological, Physical and Selected Chemical Methods,
         W. W. Kilgore and R. R. Painter (Eds.), New York: Academic Press, 1967.
     49. G. F. Shambaugh, R. F. Brown, and J. J. Pratt, Jr., Repellents for biting arthropods, in Advances
         in Pest Control Research, Vol. 1, R. L. Metcalf (Ed.), New York: Interscience Publishers, 1957,
         pp. 277–303.
     50. K. Narusuye et al., Linalool suppresses voltage-gated currents in sensory neurons and cerebellar
         purkinje cells, J. Neural Trans., 112, 193, 2005.
     51. M. A. Birkett et al., The role of volatile semiochemicals in mediating host location and selection
         by nuisance and disease-transmitting cattle flies, Med. Vet. Entomol., 18, 313, 2004.
     52. R. H. Bedoukian, Method and Compositions for Inhibiting the Scent Tracking Ability of
         Mosquitoes and Biting Midges, Washington D.C.: U.S. Patent and Trademark Office, 2005.
         Published U.S. Patent Application No. 20050090563.
     53. D. L. Kline et al., Olfactometric evaluation of spatial repellents for Aedes aegypti, J. Med.
         Entomol., 40, 463, 2003.
     54. M. Markowitz, Practical Survey of Chemistry and Metabolism of the Skin, Philadelphia: Blakiston
         Company, 1942.
     55. H. C. Koring et al., Changes in skin pH and resident flora by washing and synthetic detergent
         preparations at pH 5.5 and 8.5, J. Soc. Cosmet. Chem., 42, 147, 1991.
     56. N. Nicolaides, H. C. Fu, and G. R. Rice, The skin surface lipids of man compared with those of
         eighteen species of animals, J. Invest. Dermatol., 51, 83, 1968.
     57. N. Nicolaides, H. C. Fu, and M. N. A. Ansari, Diester waxes in surface lipids of animal skin,
         Lipids, 5, 299–307, 1970.
     58. N. Nicolaides et al., Lipid composition of comedomes compared with that of human skin surface
         in acne patients, J. Invest. Dermatol., 54, 487, 1970.
     59. M. N. A. Ansari, N. Nicolaides, and H. C. Fu, Fatty acid composition of the living layer and
         stratum corneum lipids of human sole skin epidermis, Lipids, 5, 838, 1970.
     60. S. D. Sastry et al., Volatiles emitted by humans, in Biochemical Applications of Mass
         Spectrometery, First Supplementary Volume, G. R. Waller and O. C. Dermer et al. (Eds.),
         Chichester: Wiley, 1980, pp. 1085–1129.
     61. R. I. Ellin et al., An apparatus for the detection and quantitation of volatile human effluents,
         J. Chromatogr., 100, 137, 1974.
     62. K. Naitoh et al., New measurement of hydrogen gas and acetone vapor in gases emanating from
         human skin, Instr. Sci. Technol., 30, 267, 2002.
     63. Y. Shirai et al., Alcohol ingestion stimulates mosquito attraction, J. Am. Mosq. Control Assoc.,
         18, 91, 2002.
     64. U. R. Bernier, M. M. Booth, and R. A. Yost, Analysis of human skin emanations by gas
         chromatography/mass spectrometry, 1. Thermal desorption of attractants for the yellow fever
         mosquito (Aedes aegypti) from handled glass beads, Anal. Chem., 71, 1, 1999.
     65. X.-N. Zeng et al., Analysis of characteristic odors from human male axillae, J. Chem. Ecol., 17,
         1469, 1991.
     66. X.-N. Zeng et al., An investigation of human apocrine gland secretion for axillary odor
         precursors, J. Chem. Ecol., 18, 1039, 1992.
     67. U. R. Bernier et al., Analysis of human skin emanations by gas chromatography/mass
         spectrometry, 2. Identification of volatile compounds that are candidate attractants for the
         yellow fever mosquito (Aedes aegypti), Anal. Chem., 72, 747, 2000.
     68. U. R. Bernier et al., Chemical analysis of human skin emantions: Comparison of volatiles from
         humans that differ in attraction of Aedes aegypti (Diptera: Culicidae), J. Am. Mosq. Control
         Assoc., 18, 186, 2002.
     69. C. E. Schreck, D. L. Kline, and D. A. Carlson, Mosquito attraction to substances from the skin of
         different humans, J. Am. Mosq. Control Assoc., 6, 406, 1990.


q 2006 by Taylor & Francis Group, LLC
Natural Compounds That Inhibit Mosquito Host-Finding Abilities                                          97


    70. K. G. Asano et al., Chemical composition of fingerprints for gender determination, J. Forensic
        Sci., 47, 805, 2002.
    71. A. M. Curran, S. I. Rabin, and K. G. Furton, Analysis of the uniqueness and persistence of
        human scent, Forensic Sci. Commun., 72, 1, 2005.
    72. A. M. Curran et al., Comparison of the volatile organic compounds present in human odor using
        spme-gc/ms, J. Chem. Ecol., 31, 1607, 2005.
    73. W. Ma and W. R. Klemm, Determination of acetaldehyde in blood by solid phase extraction and
        high performance liquid chromatography, Alcohol, 14, 469, 1997.
    74. R. Andreoli et al., Determination of patterns of biologically relevant aldehydes in exhaled breath
        condensate of healthy subjects by liquid chromatography/atmospheric chemical ionization tandem
        mass spectrometry, Rapid Commun. Mass Spectrom., 17, 637, 2003.
    75. B. Krotoszynski, G. Gabriel, and H. O’Neill, Characterization of human expired air: A promising
        investigative and diagnostic technique, J. Chromatogr. Sci., 15, 239–244, 1977.
    76. M. Phillips, Method for the collection and assay of volatile organic compounds in breath, Anal.
        Biochem., 247, 272, 1997.
    77. D. L. Ashley et al., Determining volatile organic compounds in human blood from a large sample
        population by using purge and trap gas chromatography/mass spectrometry, Anal. Chem., 64,
        1021, 1992.
    78. W. Rudolfs, Chemotropism of mosquitoes, Bull. N.J. Agric. Exp. Sta., 367, 4, 1922.
    79. M. Bar-Zeev, H. I. Maibach, and A. A. Khan, Studies on the attraction of Aedes aegypti (Diptera:
        Culicidae) to man, J. Med. Entomol., 14, 113, 1977.
    80. M. T. Gillies, The role of carbon dioxide in host-finding by mosquitoes (Diptera: Culicidae): A
        review, Bull. Ent. Res., 70, 525, 1980.
    81. W. Takken, The role of olfaction in host-seeking of mosquitoes: A review, Insect Sci. Applic., 12,
        287, 1991.
                                               ´
    82. T. Dekker, M. Geier, and R. T. Carde, Carbon dioxide instantly sensitizes female yellow fever
        mosquitoes to human skin odors, J. Exp. Biol., 208, 2963, 2005.
    83. W. R. Mukabana et al., Allomonal effect of breath contributes to differential attractiveness of
        humans to the African malaria vector Anopheles gambiae, Malar. J., 3, 1, 2004. (http://www.
        malariajournal.com/content/3/1/1).
    84. U. R. Bernier et al., Chemical Composition that Attract Arthropods, Washington D.C.: U.S.
        Patent and Trademark Office, 2001, U.S. Patent No. 6,267,953.
    85. U. R. Bernier et al., Chemical Composition that Attract Arthropods, Washington D.C.: U.S.
        Patent and Trademark Office, 2004, U.S. Patent No. 6,800,279.
    86. O. J. Bosch, M. Geier, and J. Boeckh, Contribution of fatty acids to olfactory host finding of
        female Aedes aegypti, Chem. Senses., 25, 323, 2000.
    87. R. C. Smallegange et al., Synergism between ammonia, lactic acid and carboxylic acids as
        kairomones in the host-seeking behavior of the malaria mosquito Anopheles gambiae sensu stricto
        (Diptera: Culicidae), Chem. Senses., 30, 145, 2005.
    88. C. Constantini et al., Electroantennogram and behavioral response of the malaria vector
        Anopheles gambiae to human-specific sweat components, Med. Vet. Entomol., 15, 259, 2001.
    89. W. G. Reifenrath, Natural Insect Repellant, Washington D.C.: U.S. Patent and Trademark Office,
        2005, U.S. Patent No. 6,953,814.
    90. H. D. Douglas III, et al., Heteropteran chemical repellents identified in the citrus odor of a
        seabird (crested auklet: Aethia cristatella): Evolutionary convergence in chemical ecology,
        Naturwissenschaften, 88, 330, 2001.
    91. H. D. Douglas III, et al., Interspecific differences in Aethia spp. auklet odorants and evidence for
        chemical defense against ectoparasites, J. Chem. Ecol., 30, 1921, 2001.
    92. H. D. Douglas III, et al., Chemical odorant of colonial seabird repels mosquitoes, J. Med.
        Entomol., 42, 647, 2005.
    93. J. C. Hagelin, I. L. Jones, and L.E L. Rasmussen, A tangerine-scented social odour in a
        monogamous seabird, Proc. R. Soc. Lond., 270B, 1323, 2003.


q 2006 by Taylor & Francis Group, LLC
98                                                        Insect Repellents: Principles, Methods, and Uses


    94. P. Steullet and P. M. Guerin, Identification of vertebrate volatiles stimulating olfactory receptors on
        tarsus I of the tick Amblyomma Variegatum Fabricus (Ixodidae), J. Comp. Physiol., 174A, 27,
        1994.
    95. P. G. Guerenstein and P. M. Guerin, Olfactory responses of the blood-sucking bug Triatoma
        infestans to odours of vertebrate hosts, J. Exp. Biol., 204, 585, 2001.
    96. V. G. Dethier, Olfactory responses of blowflies to aliphatic aldehydes, J. Gen. Physiol., 37, 743,
        1954.
    97. S. Haze et al., 2-Nonenal newly found in human body odor tends to increase with aging, J. Invest.
        Dermatol., 116, 520, 2001.
    98. Z.-M. Zhang et al., The study of fingerprint characteristics of the emanations from human arm skin
        using the original sampling system by spme-gc/ms, J. Chromatogr., 822B, 244, 2005.
    99. Y. K. Takahashi, S. Nagayama, and K. Mori, Detection of spoiled food smells by odor maps in the
        olfactory bulb, J. Neurosci., 24, 8690, 2004.
   100. M. Corradi et al., Aldehydes in exhaled breath condensate of patients with chronic obstructive
        pulmonary disease, Am. J. Respir. Crit. Care Med., 167, 1380, 2003.
   101. U. R. Bernier et al., Unpublished data, 2005.
   102. U. R. Bernier et al., Synergistic attraction of Aedes aegypti (L.) to binary blends of L-lactic acid and
        acetone, dichloromethane, or dimethyl disulfide, J. Med. Entomol., 40, 653, 2003.
   103. A. W. Ralston and J. P. Barrett, Insect Repellent Containing Aliphatic Alcohols, Washington D.C.:
        U.S. Patent and Trademark Office, 1941, U.S. Patent No. 2,254,665.
   104. B. G. Wilkes, Insect Repellents, Washington D.C.: U.S. Patent and Trademark Office, 1946, U.S.
        Patent No. 2,407,225.
   105. P. Granett and H. L. Haynes, Insect-repellent properties of 2-ethylhexanediol-1,3, J. Econ.
        Entomol., 38, 671, 1945.
   106. D. R. Hall et al., 1-Octen-3-ol: A potent olfactory stimulant and attractant for tsetse isolated from
        cattle odors, Insect Sci. Applic., 7, 5, 1984.
   107. W. Takken and D. L. Kline, Carbon dioxide and 1-octen-3-ol as mosquito attractants, J. Amer.
        Mosq. Control Assoc., 5, 311, 1989.
   108. U. R. Bernier, Unpublished data, 2000.
   109. J. P. Conkle et al., Detailed study of contaminant production in a space cabin simulator at 760 mm
        of mercury, Aerospace Med., 38, 491, 1967.
   110. P. A. Wright, Nitrogen excretion: Three end products, many physiological roles, J. Exper. Biol.,
        198, 273, 1995.
   111. M. Geier, O. J. Bosch, and J. Boeckh, Ammonia as an attractive component of host odour for the
        yellow fever mosquito, Aedes aegypti, Chem. Senses., 24, 647, 1999.
   112. M. A. H. Braks, J. Meijerink, and W. Takken, The response of the malaria mosquito, Anopheles
        gambiae, to two components of human sweat, ammonia and L-lactic acid, in an olfactometer,
        Physiol. Entomol., 26, 142, 2001.
   113. U. R. Bernier, Unpublished data, 1999.
   114. C. W. O. Bunker and A. D. Hirshfelder, Mosquito repellents, Am. J. Trop. Med., 5, 359, 1925.
   115. L. A. O. Roadhouse, Laboratory studies on insect repellency, Can. J. Zool., 31, 535, 1953.
   116. V. G. Dethier, Repellents, Ann. Rev. Entomol., 1, 181, 1956.
   117. A. Cork and K. C. Park, Identification of electrophysiologically-active compounds for the
        malaria mosquito, Anopheles gambiae, in human sweat extracts, Med. Vet. Entomol., 10, 269,
        1996.
   118. A. Ostrovskaya et al., Study and identification of volatile compounds from human skin,
        J. Cosmetic Sci., 53, 147, 2002.
   119. F. Kanda et al., Elucidation of chemical compounds responsible for foot malodour, Brit.
        J. Dermatol., 122, 771, 1990.
   120. E. D. Adrian, The activity of the nervous system in the caterpillar, J. Physiol., 70, 34, 1930.
   121. W. Roelofs et al., Sex attractant of the colding moth: Characterization with electroantennogram
        technique, Science, 174, 291, 1971.


q 2006 by Taylor & Francis Group, LLC
Natural Compounds That Inhibit Mosquito Host-Finding Abilities                                        99


   122. W. C. Agosta, Chemical Communication: The Language of Pheromones, Scientific American
        Library, New York: W.H. Freeman & Co., 1992, p. 66.
   123. A. Cork et al., Gas chromatography linked to electroantennography: A versatile technique
        for identifying insect semiochemicals, in Chromatography and Isolation of Insect Hormones and
        Pheromones, A. R. McCaffery, D. Wilson et al. (Eds.), London: Plenum Press, 1990.
   124. A. Cork, Olfactory basis of host-finding in haemotophagous arthropods, in Olfaction in
        Mosquito–Host Interactions, Ciba Foundation Symposium No. 200, G. R. Cardew (Ed.),
        London: John Wiley & Sons, 1996.
   125. J. Meijerink et al., Identification of olfactory stimulants for Anopheles gambiae from human
        sweat samples, J. Chem. Ecol., 26, 1367, 2000.
   126. Y. T. Qiu et al., Behavioural and electrophysiological responses of the malaria mosquito
        Anopheles gambiae Giles sensu stricto (Diptera: Culicidae) to human skin emanations, Med.
        Vet. Entomol., 18, 429, 2004.
   127. E. Bursell et al., Identification of compounds of cattle urine attractive to tsetse flies. Glossina
        spp. (Diptera: Glossinidae), Bull. Entomol. Res., 78, 281, 1988.
   128. A. Cork, Identification of electrophyisologically-active compounds for new world screwworm,
        Cochliomyia hominovorax, in larval wound fluid, Med. Vet. Entomol., 8, 151, 1994.
   129. W. M. Hoskins and R. Craig, The olfactory responses of flies in a new type of insect
        olfactometer, I. Theory and design of the olfactometer, J. Econ. Entomol., 27, 1029, 1934.
   130. J. O. G. Wieting and W. M. Hoskins, The olfactory responses of flies in a new type of insect
        olfactometer, II. Responses of the housefly to ammonia, carbon dioxide and ethyl alcohol,
        J. Econ. Entomol., 32, 24, 1939.
   131. E. R. Willis, The olfactory responses of female mosquitoes, J. Econ. Entomol., 40, 769,
        1947.
   132. H. K. Gouck and C. E. Schreck, An olfactometer for use in the study of mosquito
        attractants, J. Econ. Entomol., 58, 589, 1965.
   133. C. E. Schreck, H. K. Gouck, and N. Smith, An improved olfactometer for use in studying
        mosquito attractants and repellents, J. Econ. Entomol., 60, 1188, 1967.
   134. J. P. Grieco et al., A novel high-throughput screening system to evaluate the behavioral
        response of adult mosquitoes to chemicals, J. Am. Mosq. Control Assoc., 21, 404, 2005.
   135. W. M. Barrows, The reactions of the pomace fly, Drosophila ampelophila Loew, to odorous
        substances, J. Exper. Zool., 4, 515, 1907.
   136. M. Geier, H. Sass, and J. Boeckh, A search for components in human body odour that
        attract females of Aedes aegypti, in Olfaction in Mosquito–Host Interactions, Ciba Foundation
        Symposium No. 200, G. R. Cardew (Ed.), London: John Wiley & Sons, 1996.
   137. U. R. Bernier et al., Laboratory comparison of Aedes aegypti (L.) attraction to human odors
        and to synthetic human odor compounds and blends, J. Am. Mosq. Control Assoc., in press.
   138. C. N. Smith et al., L-lactic acid as a factor in the attraction of Aedes aegypti (Diptera
        Culicidae) to human hosts, Ann. Entomol. Soc. Am., 63, 760, 1970.
   139. D. L. Kline, Unpublished data, 2000.
   140. W. G. Cochran and G. M. Cox, Experimental Design, New York: Wiley, 1957.
   141. T. Jensen et al., Field efficacy of commercial antimosquito products in Illinois, J. Am. Mosq.
        Control Assoc., 16, 148, 2000.
   142. L. R. Lindsay et al., Evaluation of the efficacy of 3% citronella candles and 5% citronella
        incense for protection against field populations of Aedes mosquitoes, J. Am. Mosq. Control
        Assoc., 12, 293, 1996.
   143. R. E. Webb et al., Evaluation of spatial repellents against mosquitoes in the field, J. Am.
        Mosq. Control Assoc., in press.
   144. Y. Shirai et al., L-lactic acid as a mosquito (Diptera: Culicidae) repellent on human and
        mouse skin, J. Med. Entomol., 38, 51, 2001.
   145. S. A. Allan, U. R. Bernier, and D. L. Kline, Laboratory evaluation of avian odors for
        mosquito (Diptera: Culicidae) attraction, J. Med. Entomol., in press.
   146. J. A. Nolen, Personal communication, 2006.


q 2006 by Taylor & Francis Group, LLC
100                                                  Insect Repellents: Principles, Methods, and Uses


   147. J. R. Miller and R. S. Cowles, Stimulo-deterrent diversion: A concept and its possible
        application to onion maggot control, J. Chem. Ecol., 16, 3197, 1990.
   148. D. L. Kline, Unpublished data, 2005.
   149. C. Hansch, A. Leo, and D. H. Hoekman, Exploring QSAR., ACS Professional Reference
        Book, Washington D.C.: American Chemical Society, 1995.
   150. D. Ma et al., Predicting mosquito repellent potency of N,N-diethyl-m-toluamide (deet) analogs
        from molecular electronic properties, Am. J. Trop. Med. Hygiene., 60, 1, 1999.




q 2006 by Taylor & Francis Group, LLC
5
Standard Methods for Testing Mosquito Repellents


Donald R. Barnard, Ulrich R. Bernier, Rui-de Xue, and Mustapha Debboun


CONTENTS
Introduction ...................................................................................................................................103
Laboratory Repellent Bioassay Methods......................................................................................104
  World Health Organization Method..........................................................................................104
  American Society for Testing and Materials Method E951-94 (Revised 2000) .....................105
  Screened Cage Method..............................................................................................................105
  K&D Module Method ...............................................................................................................106
Field Repellent Bioassay Methods ...............................................................................................107
  World Health Organization Method..........................................................................................107
  American Society for Testing and Materials Method E939-94 (Revised 2000) .....................107
U.S. Environmental Protection Agency Test Guidelines .............................................................107
Sources of Variation in Repellent Bioassays ...............................................................................108
  Abiotic Factors ..........................................................................................................................108
  Biotic Factors.............................................................................................................................108
Conclusions ...................................................................................................................................109
References .....................................................................................................................................109




Introduction
Testing of mosquito repellents, whether in the laboratory or the field, is performed using a process called
biological assay (bioassay for short).1 Bioassays can be used to answer three questions about repellents:

      1. Is the candidate material repellent?
      2. What quantity of material is required for repellency?
      3. How long does repellency last?

  There are three repellent bioassay procedures documented as standard methods in the literature and
one set of repellent testing guidelines available on the internet:

      1. World Health Organization (WHO), WHO/Control of Tropical Disease/WHO Pesticide
         Evaluation Scheme/Informal Consultation (WHO/CTD/WHOPES/IC 96.1). Report of
         WHOPES Informal Consultation on the Evaluation and Testing of Insecticides.2


                                                                                                                                             103

q 2006 by Taylor & Francis Group, LLC
104                                                     Insect Repellents: Principles, Methods, and Uses


      2. American Society for Testing and Materials (ASTM) E951-94 (revised 2000). Laboratory
         Testing of Non-Commercial Repellant Formulations on the Skin.3
      3. American Society for Testing and Materials (ASTM) E939-94 (revised 2000). Field Testing
         Topical Applications of Compounds as Repellents for Medically Important and Pest
         Arthropods. 1. Mosquitoes.4
      4. U.S. Environmental Protection Agency (EPA) Office of Prevention, Pesticides, and Toxic
         Substances (OPPTS) 810.3700. Product Performance Test Guidelines. Insect Repellents for
         Human Skin and Outdoor Premises. EPA #712-C-99-369, December 1999 (available in
         public draft on the internet; see below for URL).

   A fifth repellent bioassay system, the screened cage method, is frequently cited in the literature,5–8 and
a sixth system, modified from ASTM E951-94 and adaptable to both in vivo and in vitro testing, has
recently been published.9–11




Laboratory Repellent Bioassay Methods
World Health Organization Method
Laboratory repellent bioassays based on the WHO protocol2 require a mosquito-filled, screened cage and
use deet as a positive control. Human test subjects are preferred over laboratory animals or artificial
membranes. Aedes aegypti, the normal test species, is used in variable numbers, but other mosquito
species can be substituted depending on the needs of the experiment. An area of skin ranging from that
covering the entire forearm to as little as 25 cm2 is treated with repellent and exposed to caged
mosquitoes. Untreated skin is covered with a glove or other protective material. For compounds of
unknown toxicology, the repellent may be applied to a cotton stockinette sleeve, and the treated sleeve
may be pulled over a second untreated stockinette on the arm to prevent skin contact with the repellent.
   At least five variations of the WHO method have been developed to meet the testing needs of different
institutions.2 These meathods emphasize either the determination of protection time after treatment with
a single repellent dose or the percent protection in relation to repellent dose. The protocols are as follow:

      1. A 25 cm2 area on a subject’s forearm is treated with an ethanolic solution of repellent
         (treatment), and the same-sized area on the adjacent forearm is treated with alcohol (negative
         control). Both arms are simultaneously introduced into one cage, and the numbers of
         mosquitoes biting each arm in 5 min is recorded. Percent protection is calculated by
         comparing biting rates on the treatment and control arms.
      2. A subject’s feet and legs are treated with repellent, exposed to 25 female mosquitoes in a
         mosquito-proof enclosure (1 m!1 m!3 m high), and the number of bites in 10 min
         is recorded.
      3. A subject’s forearm is treated with 1 mL of a 25% ethanolic repellent solution and introduced
         into a mosquito-filled cage for 3 min once every 30 min. Protection time is that elapsed
         between repellent application and the first mosquito bites followed by a confirmatory bite in
         the same, or next, exposure period.
      4. One gram of repellent is dissolved in sufficient acetone to saturate 280 cm2 of cotton
         stockinette. The stockinette is drawn over the arm of a subject and exposed to 1,500 caged
         female mosquitoes for 1 min, at daily or weekly intervals, until 5 bites are obtained.
      5. A subject’s untreated arm is exposed to 50 caged female mosquitoes, followed by repeated
         exposures of the same arm with increasingly high doses of repellent. In each exposure, the


q 2006 by Taylor & Francis Group, LLC
Standard Methods for Testing Mosquito Repellents                                                       105


          arm is withdrawn before the mosquitoes can imbibe blood. Probit analysis is used to calculate
          the ED50. When the dose giving 100% repellency is identified, the arm is re-exposed at 60 min
          intervals until repellency declines to 50%.

American Society for Testing and Materials Method E951-94 (Revised 2000)
This method3 comprises the use of a rectangular (18 cm length!5 cm width!4 cm height) clear plastic
test cage with five 29-mm-diameter openings in the bottom. A template is used to place four repellent
dosages and a control on the skin of a human volunteer in a pattern that matches the openings on the test
cage bottom. The cage is strapped to the arm or leg of a volunteer, bottom-side to the skin, with 10–15
nulliparous, 5–15-day-old female mosquitoes placed into the cage through a 13-mm opening at one end.
A test commences when the plastic slide (0.3-mm thick) that blocks the openings in the test cage bottom
is withdrawn, allowing mosquitoes access to the repellent treated skin. The number of mosquitoes that
land on and probe the skin in 2.5 min is recorded. The dose-response data obtained with ASTM E951-94
has been used to calculate median (ED50) and 95% effective doses (ED95)12,13 and to describe functional
responses, in time, of mosquitoes to topical repellents.12


Screened Cage Method
The screened cage bioassay method employs a 40 cm3 aluminum-frame cage with a metal bottom,
screened top and back, clear acrylic sides (for viewing), and a front stockinette sleeve for access. Two
hundred human host-seeking14 nulliparous, 7–8-day-old female mosquitoes are placed in the cage 1 h
before the test. Treatment consists of a 25% ethanolic solution of repellent active ingredient applied to
the forearm of a volunteer (between the wrist and elbow) at the rate of 1 mL/650 cm2 of skin surface area.
The treated forearm is inserted into the cage (a glove is used to protect the hand from mosquito bites,
Figure 5.1) and the number of mosquitoes that land and probe the skin in 3 min is observed and recorded.
The observations are repeated every 30 or 60 min. Two bites in one 3 min test or one bite in one 3 min
test, followed by one or more bites in a second test 30 min later ends the test for the repellent. A second
cage of mosquitoes is used as a positive or negative control. Depending upon the requirements of the
experiment, protection time is calculated as either the time elapsed between repellent application and the




FIGURE 5.1 The screen cage method of testing trial repellent formulations.


q 2006 by Taylor & Francis Group, LLC
106                                                        Insect Repellents: Principles, Methods, and Uses


first confirmed mosquito bite, or the time between repellent application and the observation period
immediately preceding the first confirmed bite. Data obtained with this bioassay method can be used to
calculate complete protection time (CPT).


K&D Module Method
An extension of ASTM E951-94 is the “K&D module”.9 This apparatus is reported to minimize the
likelihood of treatment interactions. It increases the number of possible treatments per replicate and
permits large numbers of replicated observations for each human test subject. The module can be used to
test the responses of more than one mosquito species at a time to one dose of repellent or to evaluate
repellent responses in the same species using specimens from geographically distinct locations.
   In vivo bioassays using the K&D module (Figure 5.2) are conducted in a walk-in incubator (278C and
80% RH) under fluorescent light. A template is used to delineate 3 cm!4 cm areas on the skin that
correspond to each of the six cell openings on the bottom of the module. A treatment is administered by
pipette onto a 4 cm!5 cm rectangular area of skin centered over one of the individual template marks on
a human volunteer and consists of 55 mL of ethanol containing 8.73 nmol of candidate repellent per
microlitre of ethanol. This process results in a 24 nmol dose of the treatment on 1 cm2 of skin. Skin
treated with ethanol serves as the control. The module, with five mosquitoes in each cell, is then
positioned over the treated skin area, each cell door is opened, and the number of mosquitoes that bite the
skin or become blood-engorged in 2 min is recorded.
   The in vitro system10,11 (Figure 5.3) consists of six reservoirs (3 cm!4 cm) warmed to 388C by a
water bath. Each reservoir is filled with 6 mL of outdated human blood and covered with a Baudruche or
collagen membrane. Trial repellent compounds dissolved in 110 mL of ethanol are applied in random
order to six 4 cm!5 cm pieces of organdy cloth. Each cloth is allowed to dry and then placed over one of
the membrane-covered, blood-filled cells. The module, with five mosquitoes in each cell, is positioned
over the treated cloth, and the doors are opened. The number of mosquitoes with their probosces inserted
through the cloth into the Baudruche or collagen membrane into the blood after 2 min is recorded.




FIGURE 5.2 The in vivo K&D module apparatus for bioassay of repellent active ingredients and formulations.


q 2006 by Taylor & Francis Group, LLC
Standard Methods for Testing Mosquito Repellents                                                              107




FIGURE 5.3 The in vitro K&D module apparatus for bioassay of repellent active ingredients and formulations.




Field Repellent Bioassay Methods
World Health Organization Method
When using the WHO field method,2 repellent tests are made in the vicinity of human domiciles.
Mosquito biting rate and the assessment of repellency is based on the capture of mosquitoes attacking
human volunteers; thus, tests are timed to exploit the biting cycle of the target mosquito species. Test
subjects are spaced 10 m apart and rotated in a randomized manner throughout the experiment to
minimize positional errors. Appropriate criteria for repellency include 80% reduction in biting rate for
6–8 h without adverse user side effects.


American Society for Testing and Materials Method E939-94 (Revised 2000)
In this method,4 1.5 mL of repellent solution is applied to the forearm (between the wrist and elbow) or
lower leg (between the knee and ankle) and the treated limb is exposed continuously to biting mosquitoes
as the subject moves through mosquito-infested habitat. Biting mosquitoes are collected from treated and
untreated skin (usually an exposed forearm) at regular intervals to determine mosquito biting rates and
for species identification. This procedure is used to determine CPT, but percent repellency can also be
calculated when a negative control is used.




U.S. Environmental Protection Agency Test Guidelines
The OPPTS guidelines have been developed for laboratory and field evaluation of pesticides and
toxic substances and for acquiring test data submitted to the EPA for review under the Toxic


q 2006 by Taylor & Francis Group, LLC
108                                                     Insect Repellents: Principles, Methods, and Uses


Substances Control Act (TSCA) (15 U.S.C. 2601) and the Federal Insecticide, Fungicide, and
Rodenticide Act (FIFRA) (7 U.S.C. 136 et seq.). The product performance test guidelines contained
in OPPTS 810.3700 describe specific methods for evaluating insect repellents and reflect the EPA’s
minimum recommendations for developing reliable repellent product performance data. The draft
guidelines are available electronically in portable document format (pdf) at http://www.epa.gov/
opptsfrs/publications/OPPTS_Harmonized/810_Product_Performance_Test_Guidelines/Drafts/810-
3700.pdf




Sources of Variation in Repellent Bioassays
Abiotic Factors
Many factors influence the outcome and interpretation of repellent bioassays. Skin-mediated effects
comprise absorption and penetration of repellent on skin, but evaporation, abrasion (contact with
clothing), washing or rinsing of treated surfaces, and perspiration also result in repellent loss.15–18 These
physical factors are difficult to control in a bioassay, but their contribution to experimental error can be
minimized by random selection of test subjects, the use of appropriate sample sizes in bioassays, and by
recognizing and avoiding pseudo replication. Loss of repellent by abrasion or by washing or rinsing from
treated skin can be minimized by rigorous oversight of the test proceedings and by diligence on the part
of the test subject.
   Light, temperature, humidity, and air quality at the testing venue are important environmental
influences in repellents bioassays.17,19,20 These factors can be manipulated to desired levels in the
laboratory, but in nature their variation profoundly affects mosquito responses to repellent stimuli.
Therefore, field bioassays should be standardized with respect to season, geographic location, and the
time within the diel period in which observations are made. When this is not possible, tests should be
designed so that estimates of important physical and climatic parameters are included as treatment
variables in the statistical analysis.
   Additional environmental sources of variation in bioassays are repellent dose and exposure time15 and
test cage configuration.7,8,21 In the latter case, research suggests relationships between protection time,
mosquito test population size, and the mosquito biting rate. However, investigations using different test
cage configurations and mosquito population sizes19,21–24 have not led to a consensus regarding the
optimal mosquito biting rate and density for repellency tests. One reason is that test cage shape and size
and mosquito density effects vary between mosquito species. For Aedes aegypti, for example, repellent
protection time is inversely related to cage size but is not affected by mosquito density; whereas, for
Anopheles quadrimaculatus, protection time is short in large (125-L).cages with high mosquito densities
(49 cm3 per mosquito) and long in medium (65-L) cages with low mosquito densities (640 cm3 per
mosquito).8



Biotic Factors
Biological factors in repellent bioassays consist of larval nutrition, carbohydrate availability to adult
mosquitoes, age and parity in female mosquitoes, partial blood engorgement, and innate differences
among repellent-treated test subjects.8,22,25,26 An important behavioral factor that affects bioassay results
is the timing and intensity of mosquito biting activity.5,27 Ignorance of temporal feeding patterns can
compromise estimates of protection time for repellents that have extended activity, as can poor
knowledge of biting rates. In screened cage tests, biting patterns can vary with the size of the cage,
and this factor can affect the determination of repellency.8


q 2006 by Taylor & Francis Group, LLC
Standard Methods for Testing Mosquito Repellents                                                           109




Conclusions
A comprehensive understanding of the parameters that affect repellent bioassays can minimize false
positive responses in the early stages of repellent screening. Rigorous bioassay standards in the later
stages of testing facilitate identification of the most promising new repellents and provide a sound basis
for selecting new repellents for toxicology testing and evaluation in field tests. The selection of a
repellents bioassay procedure should always be based on the biological relevance of the method and its
capacity to yield precise experimental data. When these two outcomes are achieved, one can correlate the
results from different bioassay techniques to obtain an accurate estimate of the repellency of
any compound.


References
     1. J. L. Robertson and H. K. Preisler, Pesticide Bioassays with Arthropods, Baca Raton: CRC Press,
        1992.
     2. World Health Organization [WHO], Report of the WHO informal consultation on the evaluation and
        testing of insecticides, World Health Organization, Control of Tropical Diseases, Pesticide Evaluation
        Scheme, Informal Consultation 96.1, Geneva, 1996.
     3. American Society for Testing and Materials [ASTM], Laboratory testing of non-commercial repellant
        formulations on the skin, ASTM-E951-94, 2000.
     4. American Society for Testing and Materials [ASTM], Field testing topical applications of compounds
        as repellents for medically important and pest arthropods, 1. Mosquitoes, ASTM-E393-94, 2000.
     5. H. K. Gouck and C. N. Smith, The effect of age and time of day on the avidity of Aedes aegypti, Fla.
        Entomol., 45, 93, 1962.
     6. C. N. Smith et al., Factors affecting the protection period of mosquito repellents, USDA Tech. Bull.,
        1258, 36, 1963.
     7. C. E. Schreck, Techniques for the evaluation of insect repellents: A critical review, Annu. Rev.
        Entomol., 22, 101, 1977.
     8. D. R. Barnard, Mediation of deet repellency in mosquitoes (Diptera: Culicidae) by species, age, and
        parity, J. Med. Entomol., 35, 340, 1998.
     9. J. A. Klun and M. Debboun, A new module for quantitative evaluation of repellent efficacy using
        human subjects, J. Med. Entomol., 37, 177, 2000.
    10. P. J. Weldon et al., Benzoquinones from millipedes deter mosquitoes and elicit self-anointing in
        capuchin monkeys (Cebus spp.), Naturwissenshaften, 90, 301, 2003.
    11. A. J. Klun et al., A new in vitro bioassay system for discovery of novel human-use mosquito repellents,
        J. Am. Mosq. Control Assoc., 21, 64, 2005.
    12. M. D. Buescher et al., The dose-persistence relationship of deet against Aedes aegypti, Mosq. News,
        43, 364, 1983.
    13. L. C. Rutledge et al., Comparative sensitivity of representative mosquitoes (Diptera: Culicidae) to
        repellents, J. Med. Entomol., 20, 506, 1983.
    14. K. Posey and C. E. Schreck, An airflow apparatus for selecting female mosquitoes for use in repellent
        and attraction studies, Mosq. News, 41, 566, 1981.
    15. M. L. Gabel et al., Evaporation rates and protection times of mosquito repellents, Mosq. News, 36, 141,
        1976.
    16. L. C. Rutledge et al., Mathematical models of the effectiveness and persistence of mosquito repellents,
        J. Am. Mosq. Control Assoc., 1, 56, 1985.
    17. R. Gupta and L. C. Rutledge, Laboratory evaluation of controlled release repellent formulations on
        human volunteers under three climatic regimens, J. Am. Mosq. Control Assoc., 5, 52, 1989.
    18. L. M. Rueda et al., Effect of skin abrasions on the efficacy of the repellent deet against Aedes aegypti,
        J. Am. Mosq. Control Assoc., 14, 178, 1998.


q 2006 by Taylor & Francis Group, LLC
110                                                        Insect Repellents: Principles, Methods, and Uses


    19. S. P. Frances et al., Laboratory and field evaluation of deet CIC-4, and AI3-37220 against Anopheles
        dirus (Diptera: Culicidae) in Thailand, J. Med. Entomol., 33, 511, 1996.
    20. W. G. Reinfenrath and T. S. Spencer, Evaporation and penetration from the skin, in Percutaneous
        Absorption: Mechanisms—Methods—Drug Delivery, R. L. Bronaugh and H. I. Maibach (Eds.),
        2nd ed., New York: Marcel Dekker, 1989, pp. 313–334.
    21. L. L. Lomax and P. Granett, Current laboratory procedures for the development of improved insect
        repellents at Rutgers-The State University, Proc. NJ Mosq. Exterm. Assoc., 58, 41, 1971.
    22. M. Bar-Zeev and D. Ben-Tamar, Evaluation of mosquito repellents, Mosq. News, 31, 56, 1971.
    23. A. A. Khan et al., Insect repellents: Effect of mosquito and repellent related factors on protection time,
        J. Econ. Entomol., 68, 43, 1975.
    24. S. P. Frances et al., Response of Anopheles dirus and Aedes albopictus to repellents in the laboratory,
        J. Am. Mosq. Control Assoc., 9, 474, 1993.
    25. P. V. Wood, The effect of ambient humidity on the repellency of ethylhexanediol (‘6–12’) to Aedes
        aegypti Can. Entomol., 100, 1331, 1968.
    26. R. D. Xue and D. R. Barnard, Effects of partial blood engorgement and pretest carbohydrate
        availability on the repellency of deet to Aedes albopictus, J. Vector Ecol., 24, 111, 1999.
    27. R. D. Xue and D. R. Barnard, Human host avidity in Aedes albopictus: Influence of mosquito body
        size, age, parity, and time of day, J. Am. Mosq. Control Assoc., 12, 58, 1996.




q 2006 by Taylor & Francis Group, LLC
6
Biometrics and Behavior in Mosquito
Repellent Assays


Donald R. Barnard and Rui-de Xue


CONTENTS
Introduction ...................................................................................................................................111
Physical and Biological Influences in Repellent Bioassays.........................................................112
  Mosquito Taxon.........................................................................................................................112
  Larval Rearing and Nutrition ....................................................................................................112
  Adult Age, Oviparity, and Body Size .......................................................................................113
  Carbohydrate Availability .........................................................................................................113
  Blood Feeding Patterns in Mosquitoes .....................................................................................113
  Mosquito Density, Landing Rate, and Repellency ...................................................................113
  Attraction of Mosquitoes to Human Hosts ...............................................................................115
Minimizing Variation in Repellent Bioassays..............................................................................117
  Selection of Mosquito Taxa ......................................................................................................117
  Selection of Human Test Subjects ............................................................................................117
  Selection of Mosquito Specimens .............................................................................................117
  Selection of Test Arena Configuration .....................................................................................118
Management of the Repellent Bioassay Process..........................................................................119
  Quantification of Repellency Responses...................................................................................119
  Experimental Design .................................................................................................................120
  Laboratory Bioassays ................................................................................................................121
  Field Bioassays ..........................................................................................................................121
Conclusion.....................................................................................................................................122
References .....................................................................................................................................122




Introduction
Humans use a variety of techniques for protection from arthropod bites. In the simplest case, one can
avoid entering habitat that is infested with arthropod pests or disease vectors. Conversely, biting
arthropods can be excluded from human living space by physical barriers, such as screens and nets, and
by the use of building construction methods that prevent arthropod entry.


                                                                                                                                             111

q 2006 by Taylor & Francis Group, LLC
112                                                     Insect Repellents: Principles, Methods, and Uses


   Tests for insect repellency are performed using a process called biological assay (bioassay for short).1
Repellent bioassays typically involve mosquitoes. In vitro repellent bioassays measure mosquito
response to repellent on an inanimate surface, such as repellent-treated cloth, filter paper, and animal
membrane, or to airborne repellents in an olfactometer.2,3 In vivo systems measure mosquito response to
animal and human subjects that have been treated with a repellent.4–8
   With both in vitro and in vivo bioassay systems, a stimulus is applied and a response to the stimulus by
mosquitoes is observed. This process is repeated until an average response for the test population can be
estimated with a desired level of precision. The stimulus can be a dose of repellent applied to human or
animal skin or to an inanimate object. Typical responses comprise the number of mosquitoes that
approach; land; land and probe; or land, probe, and bite the repellent-treated object.
   Procedural standards for in vivo evaluation of arthropod repellents in the laboratory and field have
been published by the World Health Organization (WHO),9 the American Society for Testing and
Materials (ASTM),10,11 and the U.S. Environmental Protection Agency (EPA).12 Two other method-
ologies, the large screened-cage 8,13 and the K&D module,14 are commonly cited in the
scientific literature.




Physical and Biological Influences in Repellent Bioassays
The results obtained in preliminary bioassays of a candidate repellent provide the basis for further study
of the repellent, including toxicological evaluation and field testing. For this reason, it is important to
minimize variation in repellency responses during the early stages of testing. Minimizing variation
requires knowledge of the characteristics and limitations of the bioassay method that is used, as well as
the capacity of the method to yield precise experimental data.
   Many physical and biological factors affect the outcome of a repellent bioassay. Some of this variation
cannot be controlled and becomes part of the experimental error. Other sources of error can be identified
and managed satisfactorily, particularly in the laboratory setting; known sources of variation can also be
accounted for by blocking or the use of other experimental designs.15



Mosquito Taxon
Species and genera of mosquitoes differ significantly in their responses to insect repellents. These
differences appear to be independently inherited and unrelated to taxonomic distance.16 Median effective
dose (ED50) values for deet vary by more than 300% among species in the same genus and by more than
600% for species in different genera16; similarly, intergeneric and intrageneric variation in the responses
of different mosquito species to a wide range of repellents is not significantly different.17 This means that
repellency responses for one species of mosquito cannot be reliably inferred from those of another
species, even among closely related taxa.



Larval Rearing and Nutrition
Overcrowding of mosquito larvae results in slow growth, small and/or irregular-sized adults, and low
fecundity in females.18–20 Khan21 observed similar protection times for deet against adult mosquitoes
from the same larval population (cohort), whereas adults from different larval populations varied
significantly in their responses to deet. In this regard, the most robust estimates of deet repellency will be
obtained from bioassays that use adult mosquitoes from different larval cohorts.


q 2006 by Taylor & Francis Group, LLC
Biometrics and Behavior in Mosquito Repellent Assays                                                      113


Adult Age, Oviparity, and Body Size
In terms of repellent protection time, that provided by 25% deet against host-seeking female Aedes
aegypti is not significantly affected by age or parity status,13 nor do these variables interact to affect deet
repellency.22 Thus, one may expect consistent protection time responses to deet when considering the
age and parity structure of the biting mosquito population.
   In contrast, mosquito age, oviparity, and body size (and the interactions of these factors) influence host
attack rates20,23 and, in some cases, repellency.24 The interplay of host attack rates and repellent
protection time8 affects the risk of exposure to mosquito-borne disease agents. In Anopheles albimanus,
the proportion of the population that is biting repellent-treated skin at the time of repellent failure is
highest in 20-day-old parous females.22 For nulliparous Aedes albopictus, host attack rates are higher in
15- and 20-day-old (post-emergence) females than in 5- and 10-day-old females, regardless of body size,
and in large compared with smaller females, regardless of age (Figure 6.1).24 Deet (25%) repellency to
large female Aedes albopictus is 2 h less than to small females.24


Carbohydrate Availability
Sugar availability affects host-seeking behavior and blood feeding in Aedes aegypti.25,26 Repellent
protection times are 4.5 h against this species when females are provided sugar water ad libitum and 3.3 h
when starved.27 In Aedes albopictus, the pretest availability of 10% sucrose solution in screened cages
increases host attack rates and the complete protection time (CPT) for 25% deet compared with females
provided water only for 12 h before tests using the same repellent treatment (Table 6.1).28


Blood Feeding Patterns in Mosquitoes
Repellent bioassays may require eight or more hours to complete. During this time, repeated observations
for mosquito landing/biting activity are made at successive intervals within the diel (24 h) period.
Mosquito host attack rates during such times can vary significantly,24,29,30 depending on the
mosquito species.
   In afternoon tests against Aedes aegypti, CPT exceeds that in morning tests by 1,000%.29 The
differences are related to variations in body size, age, and oviparity and result in higher mean attack rates
by large nulliparous females than by small nulliparous females.31 Parous females 15 or more days old are
more likely to attack human hosts than parous females less than 15 days old, whereas large-bodied, old,
parous females exhibit the highest host attack rates overall.
   Five-day-old, large-bodied, nulliparous Aedes albopictus exhibit consistent host attack rate responses
throughout the diel period.24 This fact likely contributes to their widespread use in repellent
bioassays.8,17 Mosquitoes otherwise categorized according to parity or age can provide added rigor to
tests of repellent effectiveness because of their high host attack rates. Fifteen- and twenty-day-old female
Aedes albopictus tested between 1400 and 2000 h are one example. When calculated as a percentage of
the mean daily host attack rate for 5-day-old females, attack rates for 15- and 20-day-old females range
from K18% to C148%, depending on the time of day, and are lower (negative) only between 1000 and
1200 h. This means that repellent bioassays commenced early in the day using 5-day-old female Aedes
albopictus, and that last 6 h or longer, will overestimate CPT for 15- and 20-day-old females.


Mosquito Density, Landing Rate, and Repellency
Conventional test methods for mosquito repellents9–11 assume a linear response to repellent dose and a
constant level of mosquito biting activity. Despite these assumptions, repellency responses can be highly
variable.27 Differences between humans in their attractancy to mosquitoes32 accounts for some of the
variation, as do changed biting/landing rates caused by fluctuations in mosquito density and/or


q 2006 by Taylor & Francis Group, LLC
114                                                                                 Insect Repellents: Principles, Methods, and Uses


                                           15           Large females                               Small females

                                                           5 Days                                        Nulliparous
                                           10                                                            Parous



                                           5


                                           0

                                           15
                                                           10 Days


                                           10
              Mean mosquito landing rate




                                           5


                                           0

                                           15              15 Days



                                           10


                                           5


                                           0

                                           20              20 Days


                                           15


                                           10


                                           5


                                           0
                                                0   6       12       18        24         0     6       12        18   24


                                                            Light                                      Light

                                                                        Time in diel period

FIGURE 6.1 Mean landing rates on a human host by Aedes albopictus according to time in the diel (24 h) period. Vertical
bar represents one standard deviation.




q 2006 by Taylor & Francis Group, LLC
Biometrics and Behavior in Mosquito Repellent Assays                                                      115


                TABLE 6.1
                Mean Landing Rates (in 60 s) of, and Repellent (25% deet) Protection Times
                Against, Nonblood-Fed Female Aedes albopictus Provided Water or Sucrose
                Solution (10% in Water) for 12 h Before Testing in Screened Cages
                                                              Water                    Sucrose Solution

                Mosquito landing rate (GSE)                 24.7 (G1.2)                  14.0 (G3.5)
                Repellent protection time (hGSE)             6.2 (G0.3)                   8.2 (G0.3)




endogenous cycles in the population. 24,29 An additional factor is repellent dose, the response to which by
mosquitoes depends on taxon, testing arena, and mosquito density/biting pressure. In screened cage
bioassays, Aedes aegypti responses to 25% deet are not significantly affected by mosquito landing rate;
whereas, for Anopheles quadrimaculatus protection times are short when the rates are high and long
when the rates are low.8
   The protection time of deet against Aedes aegypti in screened cages varies with differences in
mosquito density and repellent concentration.8 The relationship between protection time and deet
concentration (Table 6.2) shows that increases in repellent dose increase protection time, regardless of
mosquito density. A high (45%) deet concentration provided longer repellency than expected, given the
increase in repellency observed between 15% and 30% deet. In contrast, changes in protection time
associated with increasing mosquito density are linear, irrespective of deet concentration (Table 6.2).
Successive increases in mosquito density (up to and including 64 cm3 cage volume per female mosquito)
result in an approximately 50% reduction in protection time.
   In screened cages, there is a negative correlation between landing rates for Aedes aegypti and
Anopheles quadrimaculatus on a human subject and repellent (25% deet) protection time.8 Regression
analysis indicates that a significant portion of the repellency responses for both species can be explained
on the basis of mosquito landing rate8 and that estimated protection times (Figure 6.2) range from 4.6 to
6.2 h, when Aedes aegypti landing rates are 62 to 6 per half minute, respectively, and from 1.8 to 6.5 h
when Anopheles quadrimaculatus landing rates are from 55 to 2 per half minute, respectively.


Attraction of Mosquitoes to Human Hosts
Mosquitoes use vision, heat, and host emanations to locate their prey.33–35 Human emanations that attract
hungry mosquitoes include carbon dioxide,36–38 carboxylic acids, and lactic acid.39–42 Mixtures of



            TABLE 6.2
            Mean (GSE) Protection Time from Bites of Aedes aegypti Using Three Concentrations
            of Repellent (Deet) and Three Densities of Mosquitoes in Screened Cages
                                                    Mean (GSE) Repellent Protection Time (min)
            Deet                                               Mosquito Densitya
            Concentration (%)                      Low                    Medium                 High

            15                               290 (G40)                230 (G40)              130 (G20)
            30                               320 (G10)                260 (G26)              160 (G10)
            45                               530 (G20)                360 (G12)              290 (G52)
            a
              Number of female mosquitoes per test cage (unit of test cage volume per female). Low: 200
            (640 cm3 per female); medium: 1,000 (128 cm3 per female); high: 2,600 (49 cm3 per female).



q 2006 by Taylor & Francis Group, LLC
116                                                                                Insect Repellents: Principles, Methods, and Uses


                                                        10
                                                                              Aedes aegypti

                                                         8


                                                         6


                                                         4


                                                         2


                                                         0
                        Repellent protection time (h)




                                                             0   10      20       30          40   50      60




                                                        12
                                                                       Anopheles quadrimaculatus
                                                        10


                                                         8

                                                         6


                                                         4


                                                         2


                                                         0
                                                             0   10      20       30          40   50      60
                                                                      Mosquito landings in 30 seconds

FIGURE 6.2 Observed and estimated repellent protection times against Aedes aegypti and Anopheles quadrimaculatus in
relation to mosquito landing rates in screened cages. Broken lines are the upper and lower 95% confidence limits for
estimated repellent protection time.




emanations, such as carbon dioxide and L-lactic acid, are highly attractive to Aedes albopictus, as are
combinations of these substances with a variety of sulfides, ketones, and halogenated compounds.43
   Mosquito attraction responses can vary widely among the human subjects participating in a
repellents bioassay. They also depend on the species of mosquito that is being used.32 Human hosts,
for example, are highly or moderately attractive to Anopheles quadrimaculatus, Anopheles
freeborni, and Culex salinarius, but less attractive to Anopheles crucians and Culex nigripalpus.
For a given human subject, attraction responses vary depending on the body region to which hungry
mosquitoes are exposed.32 Early workers sought to explain these differences on the basis of skin
temperature, gender, age, and other simple effects.44,45 We know now that host-finding in
mosquitoes involves a complex of host-mediated and mosquito-based behavioral events and physio-
logical processes.46


q 2006 by Taylor & Francis Group, LLC
Biometrics and Behavior in Mosquito Repellent Assays                                                       117




Minimizing Variation in Repellent Bioassays
Selection of Mosquito Taxa
As noted earlier, species and genera of mosquitoes differ in their responses to repellents and human test
subjects in a manner unrelated to taxonomic distance.16,32 The accurate determination of repellency thus
requires that candidate compounds be tested against the target species.16 In this regard, practical
recommendations for the use of mosquito repellents should be based on laboratory and field bioassays
that use species of known pest or vector importance.47 This can be accomplished by collecting, rearing,
and testing field specimens in the laboratory, or by testing wild populations in the field, although outdoor
testing of repellents in areas with endemic mosquito-borne pathogens is accompanied by the risk of
human infection.
   In general, the probability of detecting repellency is increased by using repellent-sensitive species,
such as Aedes taeniorhynchus or Culex pipiens.17 In contrast, the identification of broad-spectrum
repellency requires the use of mosquito species, such as Aedes aegypti and Anopheles albimanus, that
have a low sensitivity to repellents.17


Selection of Human Test Subjects
Gilbert et al.44 showed that female subjects, on average, were less attractive to Aedes aegypti than male
subjects. These workers observed an inverse (but not significant) correlation between attractiveness to
mosquitoes and protection time in male subjects, when using 5% deet, but not in female subjects. Neither
body weight, age, nor skin color/temperature affected repellent protection time, regardless of gender,
although low skin moisture production in females was significantly related to increased repellent
protection time. The results of Gilbert et al.44 generally agreed with those of earlier studies,48,30 but later
analysis of their data by Rutledge49 suggested that differences in deet protection times for men and
women were not proved.
   One concern in field bioassays of repellents is the variance of estimates of mean mosquito biting rate.
Typically, the innate attractiveness of human subjects to mosquitoes ranges from 30% to 70%,32 thus,
estimates of the biting rate can be imprecise, particularly when based on small sample size. Increasing the
numbers of test subjects improves precision but the resources required to do so quickly exceed practical
limits. As an alternative to large sample sizes, Barnard et al.50 suggested that test subjects be selected
according to their comparative attractiveness to mosquitoes. This factor can be determined with an
olfactometer,51–53 or by other means. Subjects selected for use in repellents bioassays would be those
individuals with an attractiveness index within 1 or 2 standard deviations of the mean mosquito
attractiveness index for the test population.


Selection of Mosquito Specimens
Mosquitoes selected for testing in laboratory repellent bioassays should be of equal age, sex, size, and
vigor, and should be otherwise manipulated as little as possible prior to testing. Techniques for this
purpose that involve the use of carbon dioxide, low temperature, anesthesia, or aspiration54–56 subject
mosquitoes to chemical exposure, drying, and temperature extremes and can induce morbidity-related
behavioral changes in the test population with resultant variation in bioassay results.
   One method for minimizing trauma to the mosquitoes used for a bioassay is to attract and capture host-
seeking females. This can be accomplished with the apparatus described by Posey and Schreck57 that
encloses a stock cage of mosquitoes and combines air flow and human odor to attract hungry females into
a transfer module. Adjusting the flow rate of air through the apparatus allows one to count mosquitoes as


q 2006 by Taylor & Francis Group, LLC
118                                                     Insect Repellents: Principles, Methods, and Uses


they exit the stock cage and enter the transfer module. The module is then placed inside a test cage,
opened, and the mosquitoes released. Because this apparatus extracts only host-seeking female
mosquitoes from the stock cage mosquito population, one can pre-select the size of the biting mosquito
population before a test is made. This process can be repeated to produce equivalent mosquito biting
pressures in subsequent bioassays.


Selection of Test Arena Configuration
One advantage of the dose-response (small cage) testing method is that it provides a measure of
repellency at the level of the mosquito population median and/or other percentile(s) of interest. These
measures are essentially independent of the size of the mosquito populations tested.3 However, skill is
required in the design of experiments that use small cage testing methods as treatment effects can be
confounded with “edge effects,”58 the latter as a consequence of position (of a feeding port or module). In
addition, multiple replicates of treatments on the same human subject do not provide a basis for
comparison of treatments among different subjects,59 the attractiveness or repellency of which, to
mosquitoes, can be highly variable.32
   The determination of protection times using the screened cage method is based on the responses
of mosquitoes in the upper extreme of the frequency distribution for repellent tolerance rather than
on the mean response of the population. This technique does not measure the ED50 of the test
repellent by the mosquito population or other percentiles of interest.16 Additionally, it confounds
variation in repellent activity (per unit concentration applied) with the rate of repellent loss from the
skin.60,61
   In screened cage tests, test cage configuration affects repellency responses.8,13 However, studies of test
cage configuration4,13,27,62,63 have not led to a consensus regarding an optimal configuration. This is
because the effects of test cage shape and size vary depending on the species of mosquito under study.
For Aedes aegypti, CPT is inversely related to cage size, whereas for Anopheles quadrimaculatus,
protection time is shortest in 125-L cages with 49 cm3 of cage volume per female and longest in 65-L
cages with 640 cm3 of cage volume per mosquito.8
   Barnard et al.8 used regression analyses to identify combinations of cage size and mosquito density
that posed a range of challenges (from least to most rigorous) to the repellency of 25% deet in a
laboratory test with Aedes aegypti and Anopheles quadrimaculatus. For both species, estimated CPTs
were proportional to mosquito density, but showed a curvilinear relationship to cage size.
Accordingly, the shortest protection times against Aedes aegypti were observed in large cages with
high mosquito densities, where longer CPTs were associated with small cages and low mosquito
densities (640 cm3 of cage volume per female) (Figure 6.3). For Anopheles quadrimaculatus, large
cages with high mosquito densities resulted in short CPTs; medium cages with low mosquito densities
resulted in long CPTs.
   An important consideration when accepting or rejecting cage size and mosquito density parameters
is temporal variation in the host avidity pattern.24 In an attempt to characterize this phenomenon,
Barnard et al.8 calculated deviations in the mosquito biting rate in three different cage sizes at 0800,
1200, and 1600 h as a percentage of the mean biting rate, and used the deviations to select or reject
cage size and mosquito density conditions (Table 6.3). Based on a G25% deviation from the mean
biting rate, large cages with low mosquito densities, medium cages with medium mosquito densities,
and small cages with high mosquito densities would not be used in repellent tests with Aedes aegypti
or Anopheles quadrimaculatus. The G25% deviation also excluded the use of medium cages with
low densities of Aedes aegypti or small cages with low densities of Anopheles quadrimaculatus. A
deviation of G10% indicates that large cages with high mosquito densities and small cages with
medium mosquito densities would be acceptable for use in repellent bioassays with Aedes aegypti and
Anopheles quadrimaculatus, as would medium cages with high mosquito densities in assays with
Anopheles quadrimaculatus.


q 2006 by Taylor & Francis Group, LLC
Biometrics and Behavior in Mosquito Repellent Assays                                                             119


                                                                       Female density
                                                                             High
                                                                  8                     Aedes aegypti
                                                                             Medium
                                                                             Low

                                                                  6



                                                                  4



                                                                  2
                        Estimated repellent protection time (h)




                                                                  0
                                                                             Large        Medium        Small




                                                                  8

                                                                         Anopheles
                                                                      quadrimaculatus
                                                                  6



                                                                  4



                                                                  2



                                                                  0
                                                                             Large        Medium        Small

                                                                                         Cage size

FIGURE 6.3 Estimated repellent (25% deet) protection time in relation to cage size (large, medium, smallZ125, 65, and
27 L, respectively) and the density (high, medium, and lowZ49, 128, and 640 cm3 of cage volume per female mosquito) of
female mosquitoes in screened cages.




Management of the Repellent Bioassay Process
Quantification of Repellency Responses
Repellency responses in mosquitoes can be quantified in terms of the effective dose (ED), CPT, and/or
percent repellency (%R). The ED method is based on the dose-response data obtained according to
ASTM E951-9410 and is used to calculate the median (ED50) and other ED percentiles of interest for a
repellent.17,64 The CPT from mosquito bite is that time elapsed between application of a repellent on the


q 2006 by Taylor & Francis Group, LLC
120                                                     Insect Repellents: Principles, Methods, and Uses


TABLE 6.3
Deviations from Mean Landing Rate on Human Subjects by Aedes aegypti (Aa) and Anopheles
Quadrimaculatus (Aq) at Three Times of Day, in Three Different Sized Cages, Using Three
Densities of Female Mosquitoes
                                                    Cage Size (volume)
Mosquito
Density and                  Large (125 L)           Medium (65 L)                      Small (27 L)
Time (h)                  Aa             Aq        Aa             Aq             Aa                    Aq
              3
Low (640 cm cage volume per female)
0800                 K27             K37           11              6               6                   K48
1200                    5             34          K57            K13             K12                     4
1600                   22               3          46              7               6                    44
Medium (128 cm3 cage volume per female)
0800                  K4             K5           K19             25              K7                    0
1200                 K15             K4           K32              4               8                    8
1600                   19               9          51            K29              K1                   K8
High (49 cm3 cage volume per female)
0800                    7            K4           K11              7              K5                    28
1200                  K9             K2           K13             K6             K25                     1
1600                    2               6          24             K1               30                  K29


skin and the first mosquito bite on the treated skin, or the time between repellent application and the
observation time immediately preceding the first bite. Percent repellency is a quotient, comprising the
difference (at the same point in time) between mosquito biting rates on untreated and repellent treated
skin, divided by the biting rate on untreated skin, multiplied by 100. A negative control is required for
calculation of %R.
   The ED method is used to characterize repellency in insectary-reared and wild mosquitoes; however,
the data for this purpose are acquired in the laboratory. Complete protection time and %R can be used to
describe repellency responses in laboratory and field bioassays. Evaluations based on ED measure the
inherent repellency of a chemical with no consideration of how long the chemical will produce
repellency. CPT and some uses of %R measure what is essentially a combined statistic for inherent
repellency and duration. Therefore, a hypothetically good repellent might have a higher ED, but be a
powerful active ingredient because its volatility and skin absorption are low. On the other hand, a volatile
repellent with a low ED might be a poor product because it disappears from the skin too quickly.


Experimental Design
Pre-test conditions that favor an objective outcome in repellent bioassays include the randomization of
test subjects and of treatments among test subjects, adequate replication, and the use of negative
(untreated) and/or positive (treated) controls. A negative control can be the biting rate observed on the
untreated forearm, leg, or other body part of a subject when exposed to a population of mosquitoes. In
laboratory tests, a positive (treated) control can be the biting rate on one forearm of a subject that has
been treated with a known repellent (25% deet in ethanol), compared with the biting rate on the same
subject’s remaining forearm that has been treated with a repellent of unknown efficacy, and each arm
exposed separately to a population of mosquitoes. Positive controls are used to determine the
comparative efficacy of two repellents on the same test subject but for the reasons described below
should not be used in field bioassays. If positive and negative controls are used in the same bioassay, they
should be allocated to separate test subjects.
   The effects of skin-mediated sources of variation (absorption, penetration, evaporation, perspiration)
in bioassays can be minimized by random selection of test subjects, the use of an adequately-sized test
population, and by proper replication of treatments. Loss of repellent by abrasion or by washing/rinsing


q 2006 by Taylor & Francis Group, LLC
Biometrics and Behavior in Mosquito Repellent Assays                                                   121


from treated skin can be minimized by careful oversight and management of the test proceedings and by
diligence on the part of the test subject.
   Rutledge49 noted three shortcomings of the methods described in ASTM 939-9411 for data obtained in
field repellent bioassays using the incomplete block design (IBD). The first concerned the design itself,
correction of which involves analysis of data from a “resolvable balanced IBD,” rather than the balanced
IBD. The second shortcoming concerned inefficient evaluation of inter-block information, and the third,
improper use of adjusted means for estimating treatment means. Rutledge49 provides details for
correcting each problem.


Laboratory Bioassays
When determining repellency in the laboratory, at least two techniques can be used to address time-
of-observation-based systematic errors.24 The first involves the use of 5-day-old nulliparous females in
bioassays and the application of test results and inferences to only this group of mosquitoes. The second
technique is based on the assumption of equivalent host avidity throughout a bioassay and requires
commencement and completion of the test between 1400 and 2000 h (when sunrise is at 0600 h). When
this is not possible (for example, when bioassay times exceed 6 h), a single test can be divided into two
phases (early and late) of 3–6 h duration each, with the order of the phases in each test selected
at random.


Field Bioassays
ASTM E939-9411 prescribes the side-by-side comparison of repellents on the same test subject. Results
obtained in this manner can be misleading because the presence of repellent on one arm of a subject
affects the mosquito landing rate (and apparent repellency) on the opposite arm of the same subject.50
When the landing rate on negative controls is used as the reference point, this effect is manifested as a
lower landing rate on repellent treated subjects early in tests and a higher landing rate on the same
subjects late in tests. An example is shown for para-menthane-3,8-diol (PMD) (against Aedes
taeniorhynchus in the Everglades National Park [U.S.A.]), the landing rate for which, on repellent
treated subjects, differs from the landing rate on control subjects in periods 3 and 7 (approximately 3 and
7 h after repellent application) by K8% and C22%, respectively (Table 6.4). The difference equates to
an actual %R of 95.6% and 60.0%, respectively, compared with calculated %Rs of 85% and 79%, and
leads to the underestimation of PMD repellency (by 11%) in period 3 and overestimation of repellency
(by 19%) in period 7. Such interactions result in confounding of treatment (repellent) effects, biased
estimates of mosquito biting rate, and faulty estimation of repellency responses. One solution to this
problem is to use one repellent per test subject and one or more negative controls in each bioassay.9


TABLE 6.4
Mean Landing Rates for Aedes Taeniorhynchus on the Untreated Forearms of Human Subjects Whose
Opposite Forearms Had Been Treated with Deet, KBR3023, IR3535, or para-menthane-3,8-diol (PMD),
Calculated as a Percent of the Mean Landing Rate on the Forearms of (Ethanol only) Control Subjects

Approximate Number of Hours                                   Percent of Control
Following Repellent Application             Deet               KBR3023             IR3535             PMD

1                                            63                    72                76                60
3                                            85                    82                78                92
5                                            83                    91               102                87
7                                            84                   103                98               122



q 2006 by Taylor & Francis Group, LLC
122                                                        Insect Repellents: Principles, Methods, and Uses


   Field tests should be made comparable with respect to season, geographic location, and the time of the
diel period in which observations are made. When this is not possible, estimates of the important physical
and climatic parameters need to be included as treatment variables in the experimental design. In this
same regard, when mosquito biting rate and the assessment of repellency is based on the capture of
mosquitoes attacking human volunteers, bioassays should be timed to exploit the biting cycle of the
target mosquito species. Test subjects should be spaced at least 10 m apart and rotated in location in a
randomized manner throughout the experiment to minimize positional bias.9




Conclusion
Unfortunately, no single repellent bioassay system provides a definitive measure of repellent
effectiveness against mosquitoes. In fact, there is probably no “best” repellent bioassay system.
Nevertheless, it is important to know the suitability of a given system to the natural history and behavior
of the taxon under study, as well as the biological meaning of the species’ response to repellent within the
physical context of the bioassay system. Given these conditions, and the minimization of external sources
of variation, the repellent bioassay system should provide precise and repeatable measurements of
repellency. A reliable judgment of repellent effectiveness can be made on this basis and verified by
comparison with results from other repellent bioassay systems.

References
     1. J. L. Robertson and H. K. Preisler, Pesticide Bioassays with Arthropods, Boca Raton, FL: CRC Press,
        1992.
     2. H. S. Lal, S. Ginocchio, and E. J. Hawrylewicz, Procedure for bioassaying mosquito repellents in
        laboratory animals, Proc. Soc. Exper. Biol. Med., 113, 770, 1963.
     3. L. C. Rutledge, M. A. Moussa, and C. J. Belletti, An in vitro blood-feeding system for quantitative
        testing of mosquito repellents, Mosq. News, 36, 283, 1976.
     4. S. P. Frances et al., Response of Anopheles dirus and Aedes albopictus to repellents in the laboratory,
        J. Am. Mosq. Control Assoc., 9, 474, 1993.
     5. J. T. Chou, P. A. Rossignal, and J. W. Ayres, Evaluation of commercial insect repellents on human
        skin against Aedes aegypti (Diptera: Culicidae), J. Med. Entomol., 34, 624, 1997.
     6. J. A. Hill et al., Evaluation of mosquito repellents on the hairless dog, Mosq. News, 39, 307, 1979.
     7. L. C. Rutledge et al., Evaluation of the laboratory mouse model for screening topical mosquito
        repellents, J. Am. Mosq. Control Assoc., 10, 565, 1994.
     8. D. R. Barnard et al., Mosquito density, biting rates and cage size effects on repellents tests, Med. Vet.
        Entomol., 12, 39, 1998.
     9. World Health Organization (WHO), Report of the WHO informal consultation on the evaluation and
        testing of insecticides, World Health Organization, Control of Tropical Diseases, Pesticide Evaluation
        Scheme, Informal Consultation 96.1, Geneva, 1996.
    10. American Society for Testing and Materials (ASTM), Laboratory testing of non-commercial repellant
        formulations on the skin, ASTM-E951-94, 2000.
    11. American Society for Testing and Materials (ASTM), Field testing topical applications of
        compounds as repellents for medically important and pest arthropods. 1. Mosquitoes, ASTM-E393-
        94, 2000.
    12. United States Environmental Protection Agency (USEPA), Office of Prevention, Pesticides, and Toxic
        Substances, Product performance test guidelines, Insect repellents for human skin and
        outdoor premises, 810.3700, 1999, http://www.epa.gov/opptsfrs/publications/OPPTS_Harmonized/
        810_Product_Performance_Test_Guidelines/Drafts/810-3700.pdf (accessed 2006).
    13. C. E. Schreck, Techniques for the evaluation of insect repellents: a critical review, Ann. Rev. Entomol.,
        22, 101, 1977.


q 2006 by Taylor & Francis Group, LLC
Biometrics and Behavior in Mosquito Repellent Assays                                                          123


    14. J. A. Klun and M. Debboun, A new module for quantitative evaluation of repellent efficacy using
        human subjects, J. Med. Entomol., 37, 177, 2000.
    15. W. G. Cochran and G. M. Cox, Experimental Designs, 2nd ed., New York: Wiley, 1957.
    16. L. C. Rutledge et al., Comparative sensitivity of mosquito species and strains to the repellent diethyl
        toluamide, J. Med. Entomol., 14, 536, 1978.
    17. L. C. Rutledge et al., Comparative sensitivity of representative mosquitoes (Diptera: Culicidae) to
        repellents, J. Med. Entomol., 20, 506, 1983.
    18. T. Ikshoji, The influence of larval breeding conditions on fecundity of Culex pipiens fatigans Weid,
        WHO Vector Contrib., 135, 65, 1965.
    19. A. A. Khan et al., Increased attractiveness of man to mosquitoes with increased eccrine sweating,
        Nature, 223, 859, 1969.
    20. M. J. Klowden, Factors influencing multiple host contacts by mosquitoes during a single gonotrophic
        cycle, Misc. Public. Entomol. Soc. Am., 68, 29, 1988.
    21. A. A. Khan, Gross variation in the response to man among yellow fever populations in the laboratory,
        J. Econ. Entomol., 62, 96, 1969.
    22. D. R. Barnard, Mediation of deet repellency in mosquitoes (Diptera: Culicidae) by species, age, and
        parity, J. Med. Entomol., 35, 340, 1988.
    23. M. M. Klowden, J. L. Blackmer, and G. M. Chambers, Effects of larval nutrition on the host-seeking
        behavior of adult Aedes aegypti mosquitoes, J. Am. Mosq. Control Assoc., 4, 73, 1988.
    24. R. D. Xue and D. R. Barnard, Human host avidity in Aedes albopictus: influence of mosquito body
        size, age, parity, and time of day, J. Am. Mosq. Control Assoc., 12, 58, 1996.
    25. W. A. Foster and F. Eischen, Frequency of blood-feeding in relation to sugar availability in Aedes
        aegypti and Anopheles quadrimaculatus (Diptera: Culicidae), Ann. Entomol. Soc. Am., 80, 103, 1987.
    26. M. M. Klowden et al., Effects of carbohydrate ingestion on the pre-oviposition behavior of the
        mosquito Aedes aegypti (L.), Bull. Soc. Vector Ecol., 15, 59, 1990.
    27. A. A. Khan, H. I. Maibach, and D. L. Skidmore, Insect repellents: effect of mosquito and repellent
        related factors on protection time, J. Econ. Entomol., 68, 43, 1975.
    28. R. D. Xue and D. R. Barnard, Effects of partial blood engorgement and pretest carbohydrate
        availability on the repellency of deet to Aedes albopictus, J. Vector Ecol., 24, 111, 1999.
    29. H. K. Gouck and C. N. Smith, The effect of age and time of day on the avidity of Aedes aegypti, Fla.
        Entomol., 45, 93, 1962.
    30. C. N. Smith et al., Factors affecting the protection period of mosquito repellents, USDA Tech. Bull.
        1258.
    31. R. Xue, D. R. Barnard, and C. E. Schreck, Effect of body size, Parity, and age of Aedes albopictus on
        human host attack rates and the repellency of deet, J. Am. Mosq. Control. Assoc., 11, 50, 1995.
    32. C. E. Schreck, D. L. Kline, and D. A. Carlson, Mosquito attraction to substances from the skin of
        different humans, J. Am. Mosq. Control Assoc., 6, 406, 1990.
    33. G. D. Price, N. Smith, and D. A. Carlson, The attraction of female mosquitoes (Anopheles
        quadrimaculatus Say) to stored human emanations in conjunction with adjusted levels of relative
        humidity, temperature, and carbon dioxide, J. Chem. Ecol., 5, 383, 1979.
    34. W. Takken, The role of olfaction in host-seeking of mosquitoes: a review, Ins. Sci. Applic., 12, 287, 1991.
    35. W. Takken and B. G. J. Knols, Odor-mediated behavior of Afrotropical malaria mosquitoes, Ann. Rev.
        Entomol., 44, 131, 1999.
    36. C. E. Schreck, H. K. Gouck, and C. N. Smith, An improved olfactometer for use in studying mosquito
        attractants and repellents, J. Econ. Entomol., 60, 1188, 1967.
    37. M. T. Gillies, The role of carbon dioxide in host-finding by mosquitoes (Diptera: Culicidae) a review,
        Bull. Entomol. Res., 70, 525, 1980.
    38. M. Geier, O. J. Bosch, and J. Boeckh, Ammonia as an attractive component of host odour for the yellow
        fever mosquito Aedes aegypti, Chem. Sens., 24, 647, 1999.
    39. F. Acree et al., L-Lactic acid: a mosquito attractant isolated from humans, Science, 161, 1346, 1968.
    40. O. J. Bosch, M. Geier, and J. Boeckh, Contribution of fatty acids to olfactory host-finding of female
        Aedes aegypti, Chem. Sens., 25, 323, 2000.


q 2006 by Taylor & Francis Group, LLC
124                                                        Insect Repellents: Principles, Methods, and Uses


    41. T. Dekker et al., L-Lactic acid: a human-signifying host cue for the anthropophilic mosquito Anopheles
        gambiae, Med. Vet. Entomol., 16, 91, 2002.
    42. U. R. Bernier et al., Analysis of human skin emanations by gas chromatography/mass spectrometry, 2.
        Identification of volatile compounds that are candidate attractants for the yellow fever mosquito, Anal.
        Chem., 72, 7474, 2000.
    43. U. R. Bernier et al., Chemical composition that attract arthropods, U.S. Patent 6,800,279, 2004.
    44. I. H. Gilbert, H. K. Gouck, and N. Smith, Attractiveness of men and women to Aedes aegypti and relative
        protection time obtained with deet, Fla. Entomol., 49, 53, 1966.
    45. A. A. Khan, Mosquito attractants and repellents, in Chemical Control of Insect Behavior, H. H. Shorey
        and J. J. McKelvey (Eds.), New York: Wiley, pp. 155–176, 1977.
    46. A. N. Clements, Sensory Reception and Behavior, The Biology of Mosquitoes, Vol. 2, Wallingford:
        CABI Publishing, 1999.
    47. D. R. Barnard and R. D. Xue, Laboratory evaluation of mosquito repellents against Aedes albopictus
        Culex nigripalpus, and Ochlerotatus triseriatus, J. Med. Entomol., 41, 726, 2004.
    48. H. K. Gouck and M. C. Bowman, Effect of repellents on the evolution of carbon dioxide and moisture
        from human arms, J. Econ. Entomol., 52, 1157, 1959.
    49. L. C. Rutledge, Some corrections to the record on insect repellents and attractants, J. Am. Mosq. Control
        Assoc., 4, 414, 1988.
    50. D. R. Barnard et al., Repellency of IR3535, KBR3023, para-menthane-3,8-diol, and deet to Black Marsh
        mosquitoes (Diptera: Culicidae) in the Everglades National Park, J. Med. Entomol., 39, 895, 2002.
    51. A. A. Khan, Effects of repellents on mosquito behavior, Quaest. Entomol., 1, 1, 1965.
    52. K. H. Posey, D. R. Barnard, and C. E. Schreck, Triple cage olfactometer for evaluating mosquito
        (Diptera: Culicidae) attraction responses, J. Med. Entomol., 35, 330, 1998.
    53. D. J. Mauer and W. A. Rowley, Attraction of Culex pipiens (Diptera: Culicidae) to flower volatiles,
        J. Med. Entomol., 36, 503, 1999.
    54. R. I. Harris, R. A. Hoffman, and E. D. Frazer, Chilling vs. other methods of immobilizing flies, J. Econ.
        Entomol., 58, 379, 1965.
    55. C. M. Gjullin and V. D. Bevill, Insect chilling table, J. Med. Entomol., 9, 266, 1972.
    56. L. A. Magnerelli, A portable battery powered aspirator for mosquito collection, J. Med. Entomol., 12,
        308, 1975.
    57. K. H. Posey and C. E. Schreck, An airflow apparatus for selecting female mosquitoes for use in repellent
        and attraction studies, Mosq. News, 41, 566, 1981.
    58. T. R. E. Southwood and P. A. Henderson, Ecological Methods, 3rd Ed., Oxford: Blackwell Science,
        2000.
    59. R. Mead, The Design of Experiments, Statistical Principles for Practical Application, Cambridge:
        Cambridge University Press, 1988.
    60. C. F. Curtis et al., The relative efficacy of repellents against mosquito vectors of disease, Med. Vet.
        Entomol., 1, 109, 1987.
    61. L. C. Rutledge et al., Mathematical models of the effectiveness and persistence of mosquito repellents,
        J. Am. Mosq. Control Assoc., 1, 56, 1985.
    62. M. Bar-Zeev and D. Ben-Tamar, Evaluation of mosquito repellents, Mosq. News, 31, 61, 1971.
    63. L. L. Lomax and P. Granett, Current laboratory procedures for the development of improved insect
        repellents at Rutgers—The State University, Proc. NJ Mosq. Exterm. Assoc., 58, 41, 1971.
    64. M. D. Buescher et al., The dose-persistence relationship of deet against Aedes aegypti, Mosq. News, 43,
        364, 1983.




q 2006 by Taylor & Francis Group, LLC
7
Animal Models for Research and Development of
Insect Repellents for Human Use


Louis C. Rutledge and Raj K. Gupta



CONTENTS
Introduction ...................................................................................................................................126
History and Development .............................................................................................................127
Arthropod Target Species .............................................................................................................127
Vertebrate Species.........................................................................................................................128
Interspecific Differences................................................................................................................129
  Skin Temperature ......................................................................................................................129
  Skin Permeability ......................................................................................................................129
  Blood Content and Circulation..................................................................................................129
  Eccrine and Apocrine Sweat Glands.........................................................................................130
  Sebaceous Glands ......................................................................................................................130
  Hair ............................................................................................................................................130
Technique ......................................................................................................................................131
  Humane Treatment of Experimental Animals ..........................................................................131
  Preparation and Treatment ........................................................................................................131
  Test Methods .............................................................................................................................132
  Test Population Size ..................................................................................................................132
  Observing and Recording Test Data .........................................................................................133
     Methods of Observing and Recording ..................................................................................133
     Recording Protection Time ...................................................................................................133
Experimental Design and Data Analysis ......................................................................................134
  Null Treatment (Control Experiment).......................................................................................134
  Protection Time Models ............................................................................................................134
     Protection Time as a Random Variable from a Normal Distribution ..................................134
     Observations from Truncated and Censored Distributions ..................................................135
     Failure Time Data..................................................................................................................135
     Singularities in Catastrophe Data..........................................................................................135
  Analysis of Variance .................................................................................................................136
     Paired Observations ...............................................................................................................136
     One-Way and Multi-Way Experimental Designs .................................................................136
     Balanced Incomplete Block Designs ....................................................................................136




                                                                                                                                                125

q 2006 by Taylor & Francis Group, LLC
126                                                                         Insect Repellents: Principles, Methods, and Uses


  Bioassay Methods......................................................................................................................136
    No-Choice and Free-Choice Designs....................................................................................137
    Effective Dose........................................................................................................................138
    Persistence .............................................................................................................................138
Extrapolation to Humans ..............................................................................................................139
  Material Standards and Comparative Observations..................................................................139
  Statistical Adjustment of Data ..................................................................................................140
    Correction Terms ...................................................................................................................140
    Correction Factors .................................................................................................................141
    Curve Fitting..........................................................................................................................141
Conclusion.....................................................................................................................................141
References .....................................................................................................................................142




Introduction
Experimental animals have been widely used in basic and applied research leading to the commercial
production and sale of insect repellents (Figure 7.1). The present review is concerned with the use of
animal subjects in research and development of repellents intended for human use, but studies using
human subjects and studies of repellents intended for veterinary use are cited where relevant.
Toxicological studies have been excluded from consideration because toxicological tests of repellents




FIGURE 7.1 Laboratory test of repellents on infant mice in progress. Five treated mice were confined in hardware cloth
boxes inside a test cage containing 100 female mosquitoes, and biting activity was recorded at 2-min intervals for 20 min.
(From Letterman Army Institute of Research, San Francisco, CA.)


q 2006 by Taylor & Francis Group, LLC
Animal Models for Research and Development of Insect Repellents for Human Use                            127


on animals do not involve the target insects. Studies in which tissues or organs such as blood or skin are
used in lieu of the whole animal and studies of repellents intended for use against nonarthropod forms
such as schistosome cercariae or leeches are also excluded.
   The preparation of the review included the identification and assembly of 53 scientific papers published
over the period 1926–2004 reporting studies utilizing experimental animals in research and development
of repellents. In addition, the review used 11 scientific papers published by the authors and colleagues of
the former Letterman Army Institute of Research, Presidio of San Francisco, California, over the period
1980–1997 in an unofficial program of research on the use of laboratory animals in repellent testing.




History and Development
Perhaps the first recorded use of animals to test repellents intended for human use was that of Kawamura
in 1926. Kawamura used monkeys, guinea pigs, and rabbits in tests of repellents against Leptotrombi-
dium akamushi (Brumpt) (Acari: Trombiculidae) in Japan.1 In this survey, no subsequent example was
found until that of Lindquist et al. in 1944, who used chickens in tests of repellents against
Ctenocephalides felis (Bouche), Ctenocephalides canis (Curtis), and Echidnophaga gallinacea
(Westwood) (Siphonaptera: Pulicidae).2
   The number of studies using animals in research and development of repellents for human use
increased slowly through the 1940s (5 studies found), 1950s (7 studies found), 1960s (9 studies found),
1970s (9 studies found), and 1980s (12 studies found), decreasing thereafter in the 1990s (7 studies
found), and 2000s (6 studies found). The figure given for the 2000s is a projection, being prorated from
the five years 2000–2004 to the whole decade. To avoid bias, 11 papers published by the authors and
colleagues of the Letterman Army Institute of Research from 1980 to 1997 were excluded. The rise of the
modern animal rights movement from the 1970s to the present3 is a possible contributing factor to the
decline in the number of studies conducted in the 1990s and 2000s. It is not possible now to know either if
the decline will continue or for how long it will decline. In any case, a variety of repellent test systems
utilizing a variety of animal species in tests of a variety of repellents against a variety of arthropods had
been described in the scientific literature by 2004. The sections that follow present the principles and
procedures demonstrated in this body of literature as currently understood.




Arthropod Target Species
Species belonging to eight arthropod families were targeted in the 53 studies reviewed: Argasidae (soft
ticks), Ixodidae (hard ticks), Trombiculidae (chigger mites), Reduviidae (assassin bugs), Pulicidae
(pulicid fleas), Psychodidae (sand flies), Culicidae (mosquitoes), Muscidae (stable flies), and
Glossinidae (tsetse flies). This diversity of arthropod species used in the studies indicates that animal
test systems can be adapted for use with all, or nearly all, species of interest in repellent research and
development. Of the major families of human-biting arthropods, only the Pediculidae (human lice),
Ceratopogonidae (biting midges), and Tabanidae (horse flies and deer flies) were not represented in the
studies reviewed. Eventually, animal test systems may also be developed for use in research and
development of repellents for use against nonbiting species such as the bush fly, Musca vetustissima
Walker (Diptera: Muscidae), and the eye gnats, Hippelates colusor (Townsend) and Hippelates pusio
Loew (Diptera: Chloropidae).
   A total of 21 species of mosquitoes were represented in the studies reviewed, more than half of the 39
species of all arthropods tested. The number of mosquito species used reflects the generally recognized
status of the Culicidae as the single most important family of medically important arthropods. Twenty of


q 2006 by Taylor & Francis Group, LLC
128                                                     Insect Repellents: Principles, Methods, and Uses


the 53 studies reviewed used the yellow fever mosquito, Aedes aegypti (L.) (Diptera: Culicidae). Its use
reflects the longstanding status of Aedes aegypti as a preferred species for laboratory studies.4,5 No other
arthropod species was used in more than six studies.




Vertebrate Species
The first recorded use of animals in biomedical research was by the Greek anatomist Galen (129–200
ACE).3,6According to Locy,7 the animals used by Galen included Barbary apes (a species of macaque),
dogs, swine, and cattle. Today the common laboratory animals are rhesus monkeys, dogs, golden
hamsters, Mongolian gerbils, house mice, Norway rats, guinea pigs, and European rabbits. Most of the
studies surveyed for the present review used one or more common laboratory species (Table 7.1). This
circumstance reflects the experience and tradition of the biomedical research community cumulated over
the course of many years. One of the primary advantages of using the common laboratory species in
research is the extensive data available on those species in the scientific literature, much of which is
related in one way or another to the present subject.
   Biomedical research uses the guinea pig so often that the guinea pig has become a metaphor for any
subject of research, experimentation, or testing. Guinea pigs were used in 18 of the studies surveyed; 17
studies used mice; and 16 studies used rabbits. Mice have been used in biomedical research for centuries,
and the mouse has been called “the instrument of biomedical research par excellence.”8 Besides the
guinea pig, mouse, and rabbit, no other common laboratory species was used in more than four studies.
A few studies in the literature surveyed employed poultry and/or livestock as experimental animals.
Because there are no obvious scientific or technical advantages in using birds and ungulates in lieu of the
common laboratory species, their use may have been simply a matter of availability.
   Primates are regarded as the most appropriate animals for biomedical research because of their close
relationship to humans. Because use of the chimpanzee, the closest living relative of humans, is highly
restricted, the rhesus monkey is the preferred primate for biomedical research. Only two of the studies
surveyed for the present review employed primates: Kawamura1 identified the species used in his study


            TABLE 7.1
            Animal Species Used in Research and Development of Repellents for Human Use
            Species                                                  References

            Birds
            Chicken (domestic fowl)              2,35
            Pigeon (rock dove)                   42
            Canary                               48
            Mammals
            “Monkey”                             1
            Rhesus monkey                        9
            Dog                                  24
            Horse                                48
            Pig                                  34
            Ox (cow)                             34,36,49
            Golden hamster                       45
            Mongolian gerbil                     40
            House mouse                          36,45,51,54,67,68,69,70,71,72,74,78,83,90,99,100,101
            Norway rat                           23,37,45,38
            Guinea pig                           1,9,32,33,34,39,41,42,45,48,52,55,58,62,63,72,102,103
            European rabbit                      1,34,38,43,44,45,46,47,50,53,56,59,61,73,75,91



q 2006 by Taylor & Francis Group, LLC
Animal Models for Research and Development of Insect Repellents for Human Use                              129


only as “monkeys,” and, given the time and place of the study, it seems likely that the species used was
the Japanese macaque, not the rhesus monkey. Coulston and Korte9 used rhesus monkeys and guinea pigs
in tests of bicyclic lactones against Aedes aegypti. The relative disuse of rhesus monkeys in repellent
research and development reflects the high costs of procurement, care, and feeding of primates and the
cost of the special training required for experimental use of primates.




Interspecific Differences
For reasons of economy and human safety, repellents intended for human use are frequently tested on
a surrogate species, most often the guinea pig, mouse, or rabbit. However, due to specific difference
between test animals and humans, results obtained in such tests cannot be directly equated to results that
would be expected in comparable tests on humans. Valid interpretation of results obtained in tests on
experimental animals depends on the recognition and evaluation of relevant differences between the
surrogate species and humans. In the case of repellents intended for topical use, the site of interaction of
the repellent and the target arthropod is the skin. The skin is the largest organ of the body, and its
structure and function are among the most complex of all organs. It is composed of the epidermis, which
produces the stratum corneum and the skin pigments, and the dermis, a connective tissue containing
blood vessels, lymph vessels, nerve endings, the hair follicles, and the skin glands.


Skin Temperature
Using data reported by MacNay10 and Khan et al.11 Rutledge and Gupta12 quantified the importance of
temperature on the effectiveness and persistence of repellents. The MacNay data were collected in field
tests of repellents against a natural association of Aedes sticticus (Meigen), Aedes stimulans (Walker),
Aedes vexans (Meigen), and Aedes trichuris (Dyar). Protection periods of the repellents tested decreased
by an average of 7.6 min for each increase of 18C in ambient temperature.12 The data of Khan et al. were
collected in laboratory tests of deet against Aedes aegypti. The protection period of deet decreased by an
average of 2.4 min for each increase of 18C in ambient temperature.12
   As a rule, skin temperature equilibrates between the body temperature, which is nearly constant in
healthy individuals, and the ambient temperature, which is variable. Typical skin temperatures of humans
are in the range 308C to 328C,13 compared with the normal body temperature of 378C. Normal body
temperatures of common laboratory animals do not vary more than 38C from that of humans: rhesus
monkeys, 388C; dogs, 398C; hamsters, 378C; mice, 358C; rats, 378C; guinea pigs, 398C; and rabbits, 408C.14


Skin Permeability
The skin protects the organism from physical injury and acts as a barrier to the penetration of foreign
substances. The efficiency of the skin as a barrier to foreign substances depends largely on its thickness.
The thickness of the skin varies in different species and also on different parts of the body.15 The skin is
generally thicker on the dorsal than on the ventral parts of the body, except for the hands and feet, where
it is thicker on the palms and soles than on the dorsal surface. In humans the thickness of the skin varies
from about 0.1 mm on the eyelids to as much as 6 mm on the soles of the feet. Human studies have
resulted in penetration of a topically applied dose of deet into the skin from 9% to 56%.16


Blood Content and Circulation
The network of blood vessels in the skin varies in different species and also on different parts of the body.15
The blood content of human skin is greater than the blood content of the skin of any other mammal.17


q 2006 by Taylor & Francis Group, LLC
130                                                    Insect Repellents: Principles, Methods, and Uses


A continuous venous plexus supplied by inflow from the skin capillaries is especially important for the
supply of blood to the skin. The rate of blood flow into the venous plexus can vary from nearly zero to as
much as 30% of the total cardiac output.18 The blood vessels of the dermis affect the effectiveness and
persistence of repellents applied to the skin by dissipation of body heat through the skin and by the
absorption and removal of foreign substances from the skin for metabolism and/or excretion elsewhere in
the body. As indicated above in connection with skin temperature, heat flux at the surface of the skin
promotes the loss of repellents from the skin by promoting evaporation and convection of fluids at the skin
surface. In addition, human studies have shown that approximately 17% of a topically applied dose of deet
is absorbed from the skin into the circulatory system.16



Eccrine and Apocrine Sweat Glands
Active sweat glands wash away foreign substances with sweat. Even minimally functioning sweat glands
are effective in removing foreign substances from the skin.19 Two types of sweat glands occur in
mammals. Eccrine sweat glands secrete a water-and-salt filtrate derived from the blood plasma. Apocrine
sweat glands secrete organic substances derived from secretory cell cytoplasm in addition to water and
salt. Eccrine sweat glands predominate in the higher primates, and apocrine sweat glands are restricted to
small areas such as the armpits and inguinal areas. In lower primates, including the rhesus monkey, and in
all other mammals, eccrine sweat glands, when present, are restricted to areas of thickened epidermis
such as the soles; apocrine sweat glands, when present, occur elsewhere on the body. Many species,
including hamsters, gerbils, mice, rats, guinea pigs, and rabbits have neither eccrine nor apocrine sweat
glands.15


Sebaceous Glands
The number and size of the sebaceous glands vary in different species and on different parts of the
body.15 The sebaceous glands secrete an oily film, sebum, which inhibits penetration of water-soluble
substances. A variety of lipids, including saturated and unsaturated hydrocarbons and fatty acids,
compose sebum. In humans, the composition of sebum is known to vary significantly among individuals.
Correlations have been found between total skin lipid content and the duration of protection of deet and
between certain fatty acid concentrations and the duration of protection of deet.20 Studies have also
shown that a number of skin-surface lipids of humans are repellent to Aedes aegypti.21



Hair
The pelage of mammals varies in different species and on different parts of the body. Humans have
neither the greatest nor the least amount of hair among mammals. Variably coarse and pigmented
terminal hairs cover areas of visible hairiness on humans. Seemingly bare areas are covered by fine,
mostly invisible, vellus hairs, with the exception of on the palms, soles, lips, and a few other small
areas.17 As used, topical repellents are applied to exposed, relatively bare areas of skin, excluding the
scalp where the hair is long. But the skin of most animals is covered with a dense pelage comparable to
the hair of the human scalp and is therefore not a good model for the relatively bare skin of humans.
Three methods are used to circumvent this difficulty: (1) infants of species that bear hairless young are
used in lieu of adult animals22,23; (2) hairless strains or breeds are used in lieu of normal animals24
(hairless strains of dogs, mice, and rats are commercially available); and (3) test animals are shaved on
the part of the body that is to be treated and exposed to the test insects. Shaving is the most widely
accepted and frequently used method.


q 2006 by Taylor & Francis Group, LLC
Animal Models for Research and Development of Insect Repellents for Human Use                             131




Technique
Humane Treatment of Experimental Animals
The origin of the antivivisectionist movement is associated with the experiments of the French
physiologists Francois Magendie (1783–1855) and Claude Bernard (1813–1878).3,6 From France, the
                     ¸
movement spread to Britain and then to the U.S. The Royal Society for Prevention of Cruelty to Animals
was founded in 1824. The Cruelty to Animals Act was enacted in 1876, remaining in force until replaced
with new legislation in 1986. The American Society for Prevention of Cruelty to Animals was founded in
1866, and the Animal Welfare Act was enacted in 1966 with amendments in 1970 and 1985. The
touchstone documents of modern animal protection movements are Animal Liberation by Peter Singer25
and The Case for Animal Rights by Tom Regan,26 in which both authors promote a concept of animal
rights that are collateral to human rights. These works have inspired a large body of literature on animal
rights and have given rise to a number of organizations promoting this concept, including the radical
Animal Liberation Front and People for the Ethical Treatment of Animals.3,27
   Investigators using animal subjects in research must comply with all current laws and regulations
governing the care and use of experimental animals. The Animal Welfare Act has had the general effect
of increasing the costs of animal research and reducing the number of experimental animals used in the
United States.27 However, requirements for use of animal subjects are generally less burdensome than
the requirements for use of human subjects, this being a primary advantage of animal research.
   In the U.S., the Animal Welfare Act establishes standards of animal care and use and provides for regular
inspections by the U.S. Department of Agriculture. Researchers must prepare written protocols for
approval by an institutional animal care and use committee and consider alternatives to moderate the
number and/or kind of animals used, including use of in vitro methods in lieu of animals, use of lower rather
than higher vertebrates, use of advanced statistical techniques to reduce the number of animals needed, and
refinement of experimental procedures to reduce pain and/or stress. Institutional animal care and use
committees have power to require changes in protocol to comply with animal care and use standards and to
stop research projects that do not adhere to an approved protocol. Mroczek28 has presented a framework for
understanding pain and suffering in laboratory animals. Also, Toth and Olson29 have presented strategies
for minimizing pain and distress in laboratory animals. Two publications useful to researchers at the
practical level are The Principles of Humane Experimental Technique by Russell and Burch30 and the
Guide for the Care and Use of Laboratory Animals published by the National Research Council.31


Preparation and Treatment
Miller and Gibson32 tested treated netting for irritancy to mosquitoes in a wind tunnel using a guinea pig
as an attractant without special preparation or treatment. However, most procedures in repellent research
and development require prior preparation and treatment of the subject. Mechanical restraint may be
needed to maintain the animal in a position that permits access to the part of the body to which the test
materials are to be applied and/or to prevent grooming and anti-insect behavior during the procedure.
Restraining devices include handmade tables,33 cradles,34 and boards35 for restricting movement, and
hardware cloth,36–38 sausage casings,39 or knitted stockinette cloth40 for close confinement. Restrainers
for the common laboratory animal species are also available from laboratory equipment and veterinary
supply firms. Some experimental procedures may require only minor mechanical restraints such as
collars or stanchions. The procedures of Hill et al.24 using standing dogs, Kelkar et al.41 using guinea pigs,
Rutledge et al.22 using infant mice, and Mathur et al.23 using infant rats, did not require use of restraints.
   In practice, mechanical restraint is usually used in combination with chemical restraint (anesthesia).
With a few exceptions, notably the tabanid flies and the stable fly, Stomoxys calcitrans (L.) (Diptera:
Muscidae), insect bites are not very painful. The purpose of anesthesia is not only to relieve the discomfort


q 2006 by Taylor & Francis Group, LLC
132                                                      Insect Repellents: Principles, Methods, and Uses


and stress of insect bites but also to relieve the discomfort and stress of mechanical restraint. Anesthesia is
particularly important if the subject is to be restrained for several hours, as in some protocols to determine
the protection periods of repellents. Some anesthetics frequently used in work with laboratory animals are
acepromazine, ketamine, pentobarbital, and thiopental sodium. Investigators should consult the insti-
tutional veterinary staff for guidance on the use of anesthetics.
   In escape-box methods42,43 and olfactometer methods,32 the test materials are not usually applied to
the animal subject, and the shaving and marking of the animal for treatment is not required. But, except in
the case of hairless animals, it is usually necessary to shave the part of the body to which the test
materials will be applied. Then, except in the case of systemic treatments41,44–45 or of whole-body
treatments such as dips22–23 and sprays,1,36–37 a template, cutout, or pen is used to enclose or mark off the
shaved treatment area(s).24,39,46,47 Test materials are applied generally to the treatment area(s) by pipette
as ethanol solutions in concentrations and volumes calculated to provide the desired application rates in
mg/cm2 or mg/cm2 of skin. For planning purposes, the maximum rate of application of a liquid that can be
accomplished without runoff is about 2 mg/cm2. Fully formulated products are applied undiluted at the
rate specified in the “Directions for Use” section of the product label.


Test Methods
In four of the studies reviewed, the treated animals were exposed to natural populations of insects in the
field rather than to caged insects in the laboratory.1,37,48,49 Field studies have the advantage that the
environmental conditions of the study are closer to those under which the end-use product will be used,
while laboratory studies have the advantage that the conditions of the study can be more closely
controlled to reduce experimental error.
   In small-cage methods of exposure, a cage containing a known number of test insects is applied to the
treated part of the animal’s body. The test insects are allowed access to the treatment(s) by withdrawing a
slide or by some other means.47,50 In large-cage methods, one or more treated animals are placed inside
a cage containing a known number of test insects.51,52 Except in the case of whole-body treatments,
untreated parts of the animal’s body are excluded from the test insects. The experimental design may also
be such that each insect test population has access to one treatment only (no-choice method of
exposure)53,54 or that each insect test population has access to any of two or more treatments (free-
choice method).55,56 Curtis et al.57 have shown that no-choice methods and free-choice methods do not
provide equivalent results.


Test Population Size
Although natural insect population densities vary widely, densities of insect test populations are normally
standardized in laboratory studies. The procedure in which the behavior of individual animals is observed
and recorded is called focal sampling.58 For example, Miller and Gibson,32 Dethier,59 and Galun et al.60
observed and recorded the responses of individual mosquitoes and tsetse flies to various test materials.
Where larger numbers have been used, there has been little or no coordination or agreement among
investigators on test population size or test cage size. For 17 reports in which insect test population sizes
were stated for the small-cage method, the range was from 5–1061 to 2,000,47 and the median was 20. For
13 reports in which insect test population size was stated for the large-cage method, the range was from
2036,38,51 to 1,000–2,000,40 and the median was 94.
   Similarly, for 15 reports in which test cage shape and size were stated for the small-cage method,
the range of the computed volume was from 5.0!101 cm3 to 1.6!104 cm3, and 1.6!102 cm3 was the
median.33,43,62 For 11 reports in which test cage shape and size were stated for the large-cage method,
the range of the computed volume was from 1.3!103 cm3 to 1.6!107 cm3, and 3.0!104 cm3 was
the median.36 The value 1.6!107 cm3 was not representative of the set; it refers to a large cage made
to enclose cattle into which 100 Stomoxys calcitrans were released.


q 2006 by Taylor & Francis Group, LLC
Animal Models for Research and Development of Insect Repellents for Human Use                              133


   For 12 reports in which both insect test population size and test cage shape and size were stated for the
small-cage method, the range of the computed insect population density was from 1.9 insects per
1,000 cm3 to 5.6!102 insects per 1,000 cm3, and 5.9!101 insects per 1,000 cm3 was the median.43,63
For eight reports in which both insect test population size and test cage shape and size were stated for the
large-cage method, the range of the computed insect population density was from 6.2!10K3 insects per
1,000 cm3 to 2.6!102 insects per 1,000 cm3, and 3.6 insects per 1,000 cm3 was the median.36,40 The
value 6.2!10K3 insects per 1,000 cm3 was not representative of the set, referring to the experiment
described above involving a large cage made to enclose cattle.
   Considerations of economy in rearing and handling of test insects and considerations of animal welfare
favor the use of small numbers of insects, while considerations of statistical precision favor the use of large
numbers of insects. Khan et al.64 and Barnard et al.65 have shown that insect test populations of different
density do not provide equivalent estimates of protection time when the test method used depends on a
fixed endpoint for protection time such as the first (or second) observed bite. This is because the first (or
second) insect to bite in a small test population represents a less extreme position in the tolerance
distribution than the first (or second) insect to bite in a large test population.66


Observing and Recording Test Data
Methods of Observing and Recording
Preliminary, informal observations aimed at clarifying and finalizing details of technique, experi-
mental design, and data analysis are almost always necessary in repellent research, because the test
insects, test subjects, materials, equipment, facilities, and personnel involved form a complex and
variable system that may not provide accurate and definitive results as anticipated in planning the
study. In experiments on animals, scoring is usually based on observations of biting, full or partial
feeding, or attachment of the test arthropods. Most of the studies that were reviewed relied on visual
observation of the data to be recorded. However, Lal et al.67 demonstrated radioactive tracer and
fluorescent dye techniques in tests of repellents on mice against Aedes aegypti. Also Kashin and
Kardatzke68,69 and Kashin and Arneson70 electronically recorded the time of each bite in tests of
repellents on mice against Aedes aegypti.
   In the terminology of behavior studies,60 procedures for recording bites in a repellent test or experiment
may be either continuous recording, meaning that the observer records the occurrence and time of each bite
from the beginning to the end of the procedure, or time sampling, meaning that the observer records the
occurrence of biting periodically during the procedure (for examples of continuous recording, see Kashin
and Arneson,68 Kashin and Kardatzke,69,70 Sachdeva et al.71 and Abu-Shady et al.72).
   Time sampling may be either instantaneous sampling or one-zero sampling. In instantaneous sampling
the observation session is divided into successive periods of time called sample intervals. The instant of
time at the end of each sample interval is called a sample point, and the observer records the biting
activity of the test insects “instantaneously” at each sample point (for examples of instantaneous
sampling, see Wirtz et al.73 and Choi et al.74). In one-zero sampling, the observation session is similarly
divided into sample intervals, but at each sample point the observer records all bites that have occurred
during the preceding sample interval. In repellent studies, there may be only one sample interval that
extends from the beginning to the end of the observation session, and the number of insects that have fed
on (argasid ticks, reduviid bugs, mosquitoes, biting flies, fleas) or attached to (mites, ixodid ticks,
sticktight fleas) the animal subject is determined at the end (for examples of one-zero sampling see
Tripathi et al.43 and Fryauff et al.61).

Recording Protection Time
Both continuous and time sampling methods are used in tests to determine the protection times of
repellents. In a common procedure the treatment is exposed to the test insects continuously or at intervals


q 2006 by Taylor & Francis Group, LLC
134                                                        Insect Repellents: Principles, Methods, and Uses


until a specified endpoint such as the first (or second) observed bite is reached. In an alternative procedure
demonstrated by Hill et al.24 in tests of repellents on dogs, the treatment is applied at intervals to different
test subjects or at intervals to different treatment areas on the same subject. All subjects or areas are
exposed subsequently to the test insects at the same time, and the number of bites received is recorded at
that time.
   Researchers have overestimated consistently protection times determined by time sampling to a fixed
endpoint such as the first (or second) observed bite because protection times have been recorded
traditionally as the period from the time of application of the treatment to the time of the sample point at
which the first (or second) bite is observed, ignoring the possibility that the first (or second) bite might
have occurred at another time during the preceding sample interval if the test subject had been available
(for examples of sample interval error, see Bar-Zeev and Ben-Tamar75 and Bar-Zeev and Gothilf62).
Kasman et al.39 reported a mathematical correction for sample interval error. Rutledge76 recommended
recording the midpoint of the sample interval preceding the sample point at which the endpoint of the test
was observed.




Experimental Design and Data Analysis
Null Treatment (Control Experiment)
The use of controls in biological experiments is universally accepted,77 but some investigators using
human subjects in repellent studies have limited the number of bites received by control subjects: (1) by
restricting the control subjects’ exposure to a smaller area of skin than the treated subjects; (2) by
restricting the control subjects’ exposure to the beginning and end of the test only or to short periods of
time during the test; (3) by preventing actual bites and counting landings instead; (4) by substituting a
standard treatment for the null treatment; or (5) by dispensing with the null treatment altogether. Such
shortcuts and substitutions inevitably lead to ambiguity in the data obtained because the biting activity of
the insect test population is in a constant state of flux in the course of any repellent test. A bona fide
control experiment, defined as an experiment that duplicates the primary experiment in every way except
for inclusion of the test material, is an essential part of any repellent test procedure. In this regard, the use
of animal subjects offers a distinct advantage over the use of human subjects, who are not anesthetized
during repellent test procedures.


Protection Time Models
Traditionally, estimates of protection time have been assumed to be random variables from a normal
distribution for purposes of analysis. The purpose of this section is to point out some implications of this
assumption and to suggest some alternative models used in science and technology to analyze
analogous data.

Protection Time as a Random Variable from a Normal Distribution
As stated above in connection with insect test population size, the first (or second) individual to bite in
a population of insects is the individual that occupies the most (or next to most) extreme position in the
tolerance distribution of that population, i.e., the individual whose position in the tolerance distribution is
most (or next to most) distant from the mean of the test population. It is, in other words, the individual
that is the least (or the next least) representative of the population as a whole with respect to tolerance for
the test material. Statistically, the consequence of using the first or second bite as the endpoint for
determination of protection time is that the variance of the estimate obtained is large compared with the


q 2006 by Taylor & Francis Group, LLC
Animal Models for Research and Development of Insect Repellents for Human Use                         135


variance of estimates obtained from observations of less extreme individuals. Rutledge et al.66 have
demonstrated this point in terms of the standard normal distribution.

Observations from Truncated and Censored Distributions
Traditional protection time methods, such as the method used by Smith et al.34 in tests of repellents
on guinea pigs, rabbits, swine, and cattle against Aedes aegypti and the method used by Li et al.78 in
tests of repellents on mice against Aedes aegypti, depend on a fixed endpoint such as the first (or
second) observed bite. In other words, the protection time of the test material is observed for only
that individual that is the first to bite (or for only those individuals that are the first and second to
bite). The protection time of the test material is not determined for any other member of the insect
test population. Samples obtained when observation is restricted over a portion of the sample space
are known as truncated and censored samples.79 In terms of protection time, truncated samples are
those samples in which the number of test insects with protection times that lie within the restricted
area is not known, as in a field test; censored samples are those samples in which the number of test
insects with protection times that lie within the restricted area is known, as in a laboratory test.
Because all the protection times that are not observed in the test are known to be longer than those
for the first and second individuals to bite, the samples are said to be right singly truncated or right
singly censored. Because the values of protection time obtained in this kind of test method are
extreme values, the extreme value distribution may be more appropriate than the normal distribution.
Cohen79 provides examples of the analysis of truncated and censored samples from both the normal
and the extreme value distribution.


Failure Time Data
Data obtained in tests in which the time to occurrence of a specified event is observed and recorded
are called failure time data.80 In failure time terminology, a bite recorded in a protection time test,
such as the first or second observed bite, is regarded as a “failure” with respect to the test material,
and the time to occurrence of the bite is called the failure time. As stated in the preceding section,
data obtained in protection time procedures that depend on a fixed endpoint such as the first (or
second) observed bite are truncated or censored. In failure time terminology, truncation or censoring
is called order statistic, or type II, truncation or censoring because the test is discontinued when the
first (or second bite) has been observed. Kalbfleisch and Prentice80 discuss the analysis of failure time
data in terms of various distributions, including the Weibull distribution. The Weibull distribution is
related to the extreme value distribution, and results obtained in terms of either can be transferred to
the other.


Singularities in Catastrophe Data
Mathematically, a sudden change caused by gradual alteration of circumstance is termed
a catastrophe. Accordingly, the occurrence of the first, second, and succeeding bites in the course
of a protection time test can be interpreted in terms of catastrophe theory. In this interpretation, the
successive bites inflicted by members of the insect test population are catastrophic events, or
singularities, induced by smooth changes in the various factors modulating biting behavior in the test
environment, of which the factor of primary interest and importance is the test material itself. It has
been demonstrated that the gradual loss of effectiveness of a repellent on the skin conforms to the
half-life law.81 When the deposit has dissipated to the level of tolerance of the most tolerant
individual in the insect test population, the first bite will occur, when the deposit has dissipated to the
level of tolerance of the next most tolerant individual in the insect test population, the second bite
will occur, and so on for succeeding bites. Poston and Stewart82 give an integrated treatment of the
main ideas of catastrophe theory.


q 2006 by Taylor & Francis Group, LLC
136                                                      Insect Repellents: Principles, Methods, and Uses


Analysis of Variance
This section addresses the experimental designs for analysis of variance that are or have been in common
use in repellent research and development.

Paired Observations
The familiar t test has been applied in a variety of repellent studies. For example, Wood55 determined the
relative effectiveness of ethyl hexanediol at 35% and 82% RH in paired observations using guinea pigs
and Aedes aegypti. Similarly, Shirai et al.83 determined the relative effectiveness of various concen-
trations of L-lactic acid with null treatments in paired observations using mice and Aedes albopictus
(Skuse) (Diptera: Culicidae). The method of paired observations is also used to compare the effectiveness
and/or persistence of new or candidate repellents with that of a material standard.

One-Way and Multi-Way Experimental Designs
One-way, or completely random, designs are relatively simple to design, execute, analyze, and interpret.
The variance components are those for treatments, error, and sampling units, if subsampling is included
in the experimental design. For example, Miller and Gibson32 used the one-way analysis of variance to
analyze responses of Anopheles gambiae Giles and Culex quinquefasciatus Say (Diptera: Culicidae)
to permethrin, pirimiphosmethyl, and lambdacyhalothrin in a wind tunnel baited with a guinea pig.
   Multi-way designs include randomized complete block, Latin square, factorial, splitplot, and many
other designs. Such designs, in which the observations are cross-classified by blocks, plots, factors, etc.,
are used to reduce the variance of treatment means and to increase the scope of inference of the
experiment.84 For example, Fryer et al.85,86 used multi-way analysis of variance designs in tests of fly
repellents on cattle.

Balanced Incomplete Block Designs
In 1945, F.A. Morton introduced the balanced incomplete block design into repellent research and
development as a way to reduce the variance of treatment means by segregating the variance attributable
to differences among test subjects.87 Ten years later, Altman and Smith88 introduced a mathematical
formula for computing adjusted treatment means in balanced incomplete block design experiments. This
formula has been widely used in repellent research and development since 1955 and has been included in
a standard method published by the American Society for Testing and Materials.89 However, the formula
of Altman and Smith88 is erroneous, and treatment means computed with it are inaccurate.76 In some
cases, adjusted treatment means computed with this formula do not lie within the range of observed
values for the treatment. Additionally, in some cases the adjusted treatment mean may even be negative.
For example, in Table 1, a report of tests of repellents on rabbits against Ornithodoros tholozani
Laboulbene & Megnin (Acari: Argasidae), Bar-Zeev and Gothilf62 gave the adjusted mean protection
time of compound 14458 as 2.44 h and the range of observed protection times as 1–2 h. Obviously, use of
the formula of Altman and Smith for adjusting treatment means of balanced incomplete block
experiments is misuse of the balanced incomplete block design.


Bioassay Methods
In bioassay test methods the responses of the test population to the test material are determined over
a range of doses (application rates). The results of testing are analyzed as the linear regression of response
(probit transformation) on dose (logarithmic transformation) (Figure 7.2) to obtain estimates of the
median effective dose (ED50), the 95% or 99% effective dose (ED95 or ED99), and their respective
confidence limits. The basic method of bioassay and its variations are indispensable in modern science
and technology, particularly in the fields of pharmacology and toxicology, including insect toxicology.


q 2006 by Taylor & Francis Group, LLC
Animal Models for Research and Development of Insect Repellents for Human Use                                           137


                                                                                                            Probit    %

                                                                                                             7.5     99.38

  Probit        %
               100                                                                                           7.0     97.7


    6.28        90                                                                                           6.5     93.3

    5.84        80
                                                                                                             6.0     84.1
    5.52        70
                                                                                                             5.5     69.1
    5.25        60

                50                                                                                           5.0     50.0
    5.00

    4.75        40                                                                                           4.5     30.9

    4.48         30
                                                                                                             4.0     15.9
    4.16        20
                                                                                                             3.5      6.7
    3.72         10

                  0                                                                                          3.0       2.3
                      0.4               0.6     0.8           1.0          1.2           1.4          1.6
                                                      Log concentration

FIGURE 7.2 Illustrating the probit transformation: The sigmoid curve of per cent response (vertical axis) on repellent
concentration (horizontal axis) becomes linear when the percent response is transformed to the probit scale and the repellent
concentration is transformed to the logarithmic scale. (From D. J. Finney, Probit Analysis, 3rd ed., London: Cambridge
University Press, 1971.)

No-Choice and Free-Choice Designs
The statistical methods employed in bioassay differ in no-choice and free-choice experimental designs.
The data obtained in a no-choice test are all-or-nothing, or quantal, data. Specifically, the test insects in a
no-choice repellent test may either feed or not feed. There is no other alternative. On the other hand, data
obtained in free-choice repellent tests are quantitative, or nonquantal, data because a test insect may either
feed or not feed on any of two or more alternative treatments. In addition, free-choice repellent tests usually
employ an instantaneous sampling method because it is usually not possible or practicable to determine on
which treatment each insect has fed when the test is terminated. For example, Yeoman et al.90 tested butyl
3-methylcinchoninate on mice against Stomoxys calcitrans and Galun et al.33 tested microencapsulated
pyrethrum on guinea pigs against Glossina morsitans Westwood (Diptera: Muscidae) and Ornithodoros
tholozani by the no-choice repellent bioassay method. Robert et al.91 tested five repellents on rabbits
against four species of Anopheles. Choi et al.74 tested several repellents on mice against Culex pipiens L.
(Diptera: Culicidae) by the free-choice repellent bioassay method.
   Probit Analysis by D.J. Finney92 is the classical reference on the statistical methods of bioassay.
Earlier editions of this book included the method for probit analysis of quantitative data, but these were
eliminated in the third (1971) edition because, according to the author, “The problem is not very
common.” The method for probit analysis of quantitative data is also available in Goldstein.93 However,
the no-choice method is heavily favored in research and development and is the standard laboratory
method in nearly all insecticide studies. Curtis et al.57 have demonstrated use of the logit transformation,


q 2006 by Taylor & Francis Group, LLC
138                                                     Insect Repellents: Principles, Methods, and Uses


which is based on the logistic (autocatalytic) growth function, in lieu of the probit transformation, which
is based on the binomial distribution of probabilities, in the analysis of data obtained in free-choice
bioassays. Nonetheless, it is known that there is no practical difference in results of analyses using the
logit and probit transformations.94

Effective Dose
Although the effective dose of the test material can be computed from bioassay data for any desired
fraction of the insect test population except 0% and 100%, the doses normally computed are the ED50, for
statistical use, and the ED95 or ED99, for practical use. Unless otherwise indicated, the term “effective
dose” is understood to mean the dose that is effective at the stated level (50%, 95%, or 99%) at the time of
application. An effective dose determined for a longer period (for example, 4 h after application) is so
designated (for example, the 4 h effective dose). A standard method for determining the effective dose
and the 4 h effective dose of repellents on humans against mosquitoes has been published by the
American Society for Testing and Materials.95
   Smith et al.34 defined the “minimum effective dosage” as “the minimum amount [of the test material]
per unit of surface required to protect against the given population of insects.” That would be the 100%
effective dose or ED100, and, as Finney92 has stated, “Even a very large experiment could scarcely
estimate [the ED100] with any accuracy.” An additional source of confusion is that the abbreviation MED
has been used variously to designate the median effective dose, minimal effective dose (terminology of
Finney92), and “minimum effective dosage” (terminology of Smith et al.34).


Persistence
Two primary attributes of a topical repellent are its effectiveness, i.e., its ED50 and its ED95 or ED99,
and its persistence on the skin.81 Traditionally, persistence has been defined in terms of protection
time, i.e., the time elapsing from the time application of the test material to the time at which the first
(or second) bite is obtained from the insect test population. When defined in this way, the observed
protection time varies with the dose applied and the density of the insect test population. Bioassay
methods minimize this uncertainty by testing a range of doses over time and by substituting a
proportional endpoint (95% or 99% effectiveness) for the absolute endpoint (the first or second bite)
traditionally employed.66,81
   Probit Plane Model. In the probit plane method for bioassay of repellents, the number of bites
permitted by each of several doses of the test material (for example, 0.0, 0.2, 0.4, 0.8, and 1.6 mg/cm2) is
determined at each of several different times after application (for example, at 0, 2, 4, 6, and 8 h). The
data obtained are analyzed as the multiple regression of response (probit transformation) on dose
(logarithmic transformation) and time (Figure 7.3). Estimates provided by the multiple regression
equation include the median, 95%, and 99% protection time for any desired dose within the range of
doses tested; the median, 95%, and 99% effective dose for any desired time within the range of times
tested; and the confidence limits these estimates at any desired level of confidence. The basic reference
on probit plane bioassay is that of Finney.92 Rutledge et al.66,81,96 have demonstrated the probit plane
method in tests of deet and ethyl hexanediol on humans against Aedes aegypti in the laboratory and Aedes
dorsalis (Meigen) in the field.
   Effective Half-Life. Rutledge et al.81 have suggested that the effective half-life could be an
alternative to protection time as a measure of repellent persistence. Computation of effective half-life
was demonstrated with data obtained in probit plane bioassays of deet and ethyl hexanediol on
humans against Aedes aegypti (Figure 7.4). When the effective halflife and the effective dose of a
repellent are known, it is possible to estimate the initial dose required to provide a given level of
protection (for example, 95%) for a given time (for example, 4 h) after application and to estimate the
time that a given initial dose (for example, 2 mg/cm2) will remain effective at a given level of
protection (for example, 95%).


q 2006 by Taylor & Francis Group, LLC
Animal Models for Research and Development of Insect Repellents for Human Use                                                                    139


                                                                             y                                                    (+1,0,10.39)

                                                                        10

                                                                         9




                                               Response (Probits)
                                                                         8
                                                                                                4
                                                                         7
                                                                                           3
                                                                         6                                    13
                                                                                      2
                                                                                                         9
                                                                         5
                                                                                     1
                                       (-3,0,4.16)

                                                                         3
                                                                                                10       14
                                                                         2
                                                                                               15
                                                                         1
                                                                                                                   (+1,4,7.60 )
                                             (-3,0,0 )                                                                                 x1
                                                                                                              1
                                                                                          11         8
                                                                                                              -1        0         +1
                                                                                          6
                                                                                     16                        Dose (log mg / cm2)
                                              )
                                             urs




                                                                                 5              7
                                                                    1
                                         (Ho
                                      riod




                                                                                          12
                                  t pe




                                         2
                                 Tes




                                  3
                                                                             Deet
                 (-3,4, 1.37)                                                 Y = 8.83090 +1.55781 x1 -0.69605 x2

                           4
                            x2

FIGURE 7.3 Illustrating the probit plane model: The response of Aedes aegypti to deet (Y axis, in probit values) is
represented as a plane plotted on the applied dose (X1 axis, logarithmic scale) and the elapsed time from the time of
application (X2 axis, hours). Dark circles and associated vertical lines show the observed values and their respective
deviations from the probit plane. (From L. C. Rutledge et al., Mathematical models of the effectiveness and persistence of
mosquito repellents, J. Am. Mosq. Control Assoc., 1, 56, 1985.)




Extrapolation to Humans
Material Standards and Comparative Observations
A material standard is a standardized material to which other materials can be compared in paired
observations. For example, Yeoman et al.90 determined the effectiveness of six doses of butyl
3-methylcinchoninate (the test material) and deet (the material standard) against Stomoxys calcitrans
in paired observations on mice. The ED50’s of the test material (0.002 mg/cm2) and material standard
(0.01 mg/cm2) were estimated graphically. If the ED50 of the material standard in comparable tests on
humans were known, and if the difference between the ED50’s of test materials and the material standard
(K0.008 mg/cm2) were known to be the same in comparable tests on mice and humans, then an


q 2006 by Taylor & Francis Group, LLC
140                                                                                     Insect Repellents: Principles, Methods, and Uses


                                           0                                                                                1
                                                    X1                                              Deet
                                                                                                    λ = 1.04 hours-1
                                         −0.7                                                       t1/2 = 0.67 hours       .2
                                          −1                                                                                .1
                                                                      ED95 = 0.040 mg/cm2
                  Residue (log mg/cm2)




                                                                                                                                   Residue (mg/cm2)
                                                                                   Z=
                                                                                        -0.7
                                                                                            0-0
                                                                                                   .45
                                          −2                                                             X                  .01
                                                                                                         2
                                                         ED50=0.003   mg/cm2



                                          −3                                                                                .001




                                          −4                                                                                .0001
                                                0          1    1.6            2               3             3.9 4      5
                                                                      Test period (Hours)

FIGURE 7.4 Illustrating the half-life of deet on the human forearm: The applied dose was found to decay at a rate (l) of
1.04 log mg/cm2 per hour, from which the half-life (t ⁄ ) was computed to be 0.67 hours. X1 is the applied dose (0.2 mg/cm2)
                                                                       1
                                                                           2

and Z is the residue remaining at X2 hours after application. The dashed lines indicate the computed ED50 and ED95. (From
Letterman Army Institute of Research, San Francisco, CA.)


extrapolation from the value obtained in tests on mice to the value that would be expected in tests on
humans could be made (See section on correction terms).
   Similarly, Bar-Zeev and Gothilf 63 determined the protection times of 538 organic compounds (the test
materials) and deet (the material standard) against Xenopsylla cheopis (Rothschild) (Siphonaptera:
Pulicidae) in paired observations on guinea pigs. The ratios of the mean protection times of the test
materials to the respective mean protection times of deet were computed for purposes of comparison. If
the protection time of the material standard in comparable tests on humans were known, and if the ratios of
the protection times of the test materials to the material standard were known to be the same in comparable
tests on guinea pigs and humans, then extrapolations from values obtained in tests on guinea pigs to values
that would be expected in tests on humans could be made (See section on correction factors).


Statistical Adjustment of Data
Because animal models do not precisely simulate the human standard, it is necessary to adjust values
obtained in experiments on animals statistically to estimate the corresponding values that would be
obtained in comparable experiments on humans. In this context, an adjusted value is defined as a derived
value that can be used for an intended purpose.97 The problem of adjusting experimental data occurs in
many different fields of science and technology and particularly in the fields of pharmacology
and toxicology.

Correction Terms
In the present context a correction term, or additive correction, is a derived value that can be added to a
value obtained in an animal test system to estimate the corresponding value that would be obtained in
a comparable human test system. The correction term is computed as the mean difference between values
obtained in the animal and the human test system, using data obtained by testing a series of materials in


q 2006 by Taylor & Francis Group, LLC
Animal Models for Research and Development of Insect Repellents for Human Use                           141


both systems. Corrected values of estimates subsequently obtained in the animal test system, i.e.,
adjusted values that estimate corresponding values in the human test system, are computed by adding the
correction term to estimates obtained in the animal test system. The variance of a corrected value is equal
to the variance of the value obtained in the animal test system plus the variance of the correction term.
   For example, Rutledge et al.98 tested eight repellents on mice and humans against Aedes aegypti using
the probit plane method. Correction terms for the Y intercept and each of the two regression coefficients
were computed from data obtained on four of the repellents and verified with data obtained on the other
four. Corrected values computed from tests on mice did not differ significantly from values obtained in
tests on humans. Similarly, Kasman et al.39 determined graphically that the protection times of a number
of repellents obtained in tests on humans against Aedes aegypti were approximately 20 min longer than
those obtained in tests on guinea pigs. In this study, then, the value of the correction term was C20 min.

Correction Factors
In the present context a correction factor, or multiplicative correction, is a derived value that can be
multiplied by a value obtained in an animal test system to estimate the corresponding value that would be
obtained in a comparable human test system. The correction factor is computed as the mean ratio of the
values obtained in the animal and the human test system, using data obtained by testing a series of
materials in both systems. Corrected values of estimates subsequently obtained in the animal test system,
i.e., adjusted values that estimate corresponding values in the human test system, are computed by
multiplying estimates obtained in the animal test system by the correction factor. The variance of a
corrected value is equal to the variance of the value obtained in the animal test system times its mean plus
the variance of the correction term times its mean, if the variances are small compared to the means.
   No exact example of the use of correction factors was found in the literature reviewed. However, the
practice of converting animal test data to ratios of the value obtained on the test material to the value
obtained on a material standard is suggestive of the correction factor approach. Also Hill et al.24
computed the linear regression of the protection time of repellents on guinea pigs against Aedes aegypti
on the protection time of the same repellents on humans against Aedes aegypti. In this case, an additive
correction (the Y intercept) and a multiplicative correction (the regression coefficient) were used in
conjunction. In this study, however, the regression line fitted was inverted (i.e., guinea pig values were
the dependent variable). Values of the Y intercept and the regression coefficient were not reported.

Curve Fitting
In the graphical method of curve fitting, values obtained in the human test system (the dependent
variable) are plotted against values obtained in the animal test system (the independent variable), using
data obtained by testing a series of materials in both systems. Values subsequently obtained in the animal
test system can then be converted to estimates of values to be expected in the human test system by
reading the latter from the graph.
   Statistical methods of curve fitting are more precise and provide confidence limits for the values
estimated. In the study of Hill et al.24 the relation between protection times of repellents on guinea pigs
and humans against Aedes aegypti was linear. If the relation between the variables studied is nonlinear,
the data may be either transformed to yield a linear relation or analyzed by the methods of curvilinear
regression.84




Conclusion
History shows that nearly all significant advances in biomedical science are made first in experiments on
animals. Some familiar examples are the development of physiology and biochemistry through


q 2006 by Taylor & Francis Group, LLC
142                                                      Insect Repellents: Principles, Methods, and Uses


experiments done primarily on mice and rats, toxicology and endocrinology in experiments done on
rabbits, and genetics in experiments done on fruit flies. The twentieth century advances in pharmacology
and toxicology can be regarded as prototypical of the advances in repellent science currently in progress
because the basic problem of quantifying and measuring the response of the organism to the test material
is the same in each case. Because no animal test system for repellents has come into general use to date,
the researchers cited in the present paper are best regarded as pioneers of a methodology that has not yet
reached maturity. Collectively, however, they have demonstrated the principles and procedures that will
shape the mature methodology that is eventually standardized and adopted for general use.
   The standardization and general adoption of animal repellent test systems will depend on the
development and refinement of available techniques for increased precision and accuracy and for
accurate extrapolation to comparable human test systems. However, the twin problems of precision
and accuracy104 apply equally to both animal and human test systems, and the problem of accurate
extrapolation from one system to the other can be resolved only by improving the precision and
accuracy of both. Although animal test systems have been developed, or invented, entirely on an ad
hoc basis to date, standardization of animal test systems for general use will require funding of a
research and development project, including a program of interlaboratory trials, dedicated to that
specific end. In computing cost-benefit figures for the project, account should be taken not only of the
potential savings in the long-term costs of repellent research and development but also the projected
increase in human safety resulting from deferral of tests on humans to the late stages of
repellent development.
   The potential of experiments on animals in repellent science is even greater in the area of basic
research than it is in the area of applied research. Chemoreception, structure-activity relationships, mode
of action, and other basics of biochemistry, physiology, and ethology bearing on the interaction of
repellent, arthropod, vertebrate host, and environment are still little known. Basic principles of
combining and formulating repellent compounds for controlled release, reduced absorption, abrasion
resistance, synergism, user acceptance, and other desirable properties are similarly little known.
Advances in repellent science can not be achieved through the development of precise, accurate test
methods alone. Advances in basic knowledge also are needed, and, as the history of biomedical science
has shown, the most productive approach to advances in basic knowledge is that of basic research
on animals.


References
     1. R. Kawamura, Studies on tsutsugamushi disease (Japanese flood fever), Med. Bull. Coll. Med. Univ.
        Cincinnati, 4(Suppl.), 1, 1926.
     2. A. W. Lindquist, A. H. Madden, and C. N. Watts, The use of repellents against fleas, J. Econ.
        Entomol., 37, 485, 1944.
     3. A. Guerrini, Experimenting with Humans and Animals: From Galen to Animal Rights, Baltimore: The
        Johns Hopkins University Press, 2003.
     4. S. R. Christophers, Aedes aegypti (L.), the Yellow Fever Mosquito: Its Life History, Bionomics and
        Structure, London: Cambridge University Press, 1960.
     5. G. B. Craig, The contribution of Aedes aegypti research to the advancement of biological science, Am.
        J. Trop. Med. Hyg., 14, 904, 1965.
     6. E. Gasking, The Rise of Experimental Biology, New York: Random House, 1970.
     7. W. A. Locy, The Story of Biology, Garden City, NY: Garden City, 1925.
     8. R. Bud and D. J. Warner (Eds.), Instruments of Science: An Historical Encyclopedia, London: The
        Science Museum, 1998.
     9. Coulston, F. and Korte, F. W. A. G. K., Insect Repellent, U.S. Patent 4,663,346, 1987.
    10. C. G. MacNay, Studies on repellents for biting flies, Can. Entomol., 71, 38, 1939.
    11. A. A. Khan, H. I. Maibach, and D. L. Skidmore, A study of insect repellents, 2. Effect of temperature
        on protection time, J. Econ. Entomol., 66, 437, 1973.


q 2006 by Taylor & Francis Group, LLC
Animal Models for Research and Development of Insect Repellents for Human Use                                 143


    12. L. C. Rutledge and R. K. Gupta, Reanalysis of the C.G. MacNay mosquito repellent data, J. Vector
        Ecol., 21, 132, 1996.
    13. B. G. J. Knols, W. Takken, and R. De Jong, Influence of human breath on selection of biting sites by
        Anopheles albimanus, J. Am. Mosq. Control Assoc., 10, 423, 1994.
    14. W. S. Spector (Ed.), Handbook of Biological Data, Philadelphia: W.B. Saunders, 1956.
    15. V. E. Sokolov, Mammal Skin, Berkeley: University of California Press, 1982.
    16. P. J. Robbins and M. G. Cherniak, Review of the biodistribution and toxicity of the insect repellent
        N,N-diethyl-m-toluamide (deet), J. Toxicol. Environ. Health, 18, 503, 1986.
    17. W. Montagna, The Structure and Function of Skin, 3rd ed., San Diego: Academic Press, 1974.
    18. A. C. Guyton and J. E. Hall, Human Physiology and Mechanisms of Disease, 6th ed., Philadelphia:
        W.B. Saunders, 1997.
    19. R. L. Rietschel and T. S. Spencer, Correlation between mosquito repellent protection time and
        insensible water loss, J. Invest. Dermatol., 65, 385, 1975.
    20. W. A. Skinner et al., Influence of human skin surface lipids on protection time of topical mosquito
        repellent, J. Pharm. Sci., 66, 1764, 1977.
    21. W. A. Skinner et al., Human skin-surface lipid fatty acids—mosquito repellents, Experientia, 26, 728,
        1970.
    22. L. C. Rutledge, M. A. Lawson, and L. L. Young, Tests of repellents against Diamanus montanus
        (Siphonaptera: Ceratophyllidae), J. Med. Entomol., 19, 361, 1982.
    23. K. K. Mathur et al., Laboratory evaluation of diethyl phenyl acetamide & other insect repellents
        against rat flea, Xenopsylla cheopis (Rothschild), Indian J. Med. Res., 83, 466, 1986.
    24. J. A. Hill et al., Evaluation of mosquito repellents on the hairless dog, Mosq. News, 39, 307, 1979.
    25. P. Singer, Animal Liberation: A New Ethics for Our Treatment of Animals, New York: New York
        Review of Books, 1975.
    26. T. Regan, The Case for Animal Rights, Berkeley: University of California Press, 1983.
    27. M. Mukerjee, Trends in animal research, Sci. Am., 276(2), 86, 1997.
    28. N. S. Mroczek, Recognizing animal suffering and pain, Lab Animal., 21, 27, 1992.
    29. L. A. Toth and G. A. Olson, Strategies for minimizing pain and distress in laboratory animals, Lab
        Animal, 20, 33, 1991.
    30. W. L. M. Russell and R. L. Burch, The Principles of Humane Experimental Technique, London:
        Methuen, 1959.
    31. National Research Council, Guide for the Care and Use of Laboratory Animals, Washington, DC:
        National Academies Press, 1996.
    32. J. E. Miller and G. Gibson, Behavioral response of host-seeking mosquitoes (Diptera: Culicidae) to
        insecticide-impregnated bed netting: A new approach to insecticide bioassays, J. Med. Entomol., 31,
        114, 1994.
    33. R. Galun et al., Long-term protection of animals from tsetse bites through controlled-release
        repellents, Isotope and Radiation Research on Animal Diseases and their Vectors, Vienna:
        International Atomic Energy Agency, 1980, p. 207.
    34. C. N. Smith et al., Factors affecting the protection period of repellents, U.S. Dept. Agric. Tech. Bull.,
        1285, 1963.
    35. K. Watanabe et al., New mosquito repellent from Eucalyptus camaldulensis, J. Agric. Food Chem., 41,
        2164, 1993.
    36. G. W. Eddy and W. S. McGregor, Use of white mice for testing materials used as repellents and
        toxicants for stable flies, J. Econ. Entomol., 42, 461, 1949.
    37. M. Bar-Zeev and S. Gothilf, Field evaluation of repellents against the tick Ortithodoros tholozani
        Labou & Megn., J. Med. Entomol., 11, 389, 1974.
    38. M. Bar-Zeev et al., Studies of repellents against Panstrongylus megistus (Hemiptera: Reduviidae),
        J. Med. Entomol., 1, 70, 1980.
    39. S. Kasman, L. A. O. Roadhouse, and G. F. Wright, Studies in testing insect repellents, Mosq. News, 13,
        116, 1953.
    40. J. A. Klocke, M. V. Darlington, and M. F. Balandrin, 1, 8-Cineole (eucalyptol), a mosquito feeding and
        ovipositional repellent from volatile oil of Hemizonia fitchii (Asteraceae), J. Chem. Ecol., 13, 2131, 1987.


q 2006 by Taylor & Francis Group, LLC
144                                                         Insect Repellents: Principles, Methods, and Uses


    41. V. V. Kelkar, A. P. Pandya, and N. R. Metha, Biting activity by Aedes aegypti mosquitoes in guinea pigs:
        an experimental method for screening the effect of some systemically administered compounds, Trop.
        Geogr. Med., 31, 415, 1979.
    42. J. C. P. Dias and J. C. da Silva, Some aspects of defensive prophylaxis in Chagas’ disease [in Portuguese],
        Rev. Inst. Med. Trop. Sao Paulo, 11, 236, 1969.
    43. A. K. Tripathi et al., Piperitenone oxide as toxic, repellent, and reproduction retardant toward malarial
        vector Anopheles stephensi (Diptera: Anophelinae), J. Med. Entomol., 41, 691, 2004.
    44. M. Bar-Zeev and C. N. Smith, Action of repellents on mosquitoes feeding through treated membranes or
        on treated blood, J. Econ. Entomol., 52, 263, 1959.
    45. Kingscote, A. A., Orally administered insect repellents: approaches and problems related to the search,
        Proc.III Int. Congr. Entomol, 799, 1956.
    46. M. Bar-Zeev and S. Sternberg, Testing space repellents against mosquitoes, Mosq. News, 30, 27, 1970.
    47. R. A. Wirtz et al., Laboratory testing of repellents against the sand fly Phlebotomus papatasi (Diptera:
        Psychodidae), J. Med. Entomol., 23, 64, 1986.
    48. R. Wasicky, O. Unti, and A. da Silva Ramos, Repellents for mosquitoes [in Portuguese], Anais Fac. Farm.
        e Odont. Univ. Sao Paulo, 7, 297, 1949.
    49. S. K. Sharma, V. K. Dua, and V. P. Sharma, Field studies on the mosquito repellent action of neem oil, S.E.
        Asian J. Trop. Med. Public Health, 26, 180, 1995.
    50. L. L. Robert, I. Schneider, and R. A. Wirtz, Deet and permethrin as protectants against malaria-infected
        and uninfected Anopheles stephensi mosquitoes, J. Am. Mosq. Control Assoc., 7, 304, 1991.
    51. Y. Okada et al., Amino acids and peptides, XI. Synthesis of attractant and repellent peptides for Aedes
        aegypti and Blatella germanica, Chem. Pharm. Bull., 32, 4608, 1984.
    52. W. Neng et al., Laboratory evaluation of efficacy of bednets impregnated with pyrethroids, J. Am. Mosq.
        Control Assoc., 7, 294, 1991.
    53. E. B. Starnes and P. Granett, A laboratory method for testing repellents against biting flies, J. Econ.
        Entomol., 46, 420, 1953.
    54. M. J. R. Healy, A method for comparing fly-repellant sprays, Biometrics, 9, 290, 1953.
    55. P. W. Wood, The effect of ambient humidity on the repellency of ethylhexanediol (“6–12”) to Aedes
        aegypti, Can. Entomol., 100, 1331, 1968.
    56. M. Kalyanasundarum, A preliminary report on the synthesis and testing of mosquito repellents, Indian
        J. Med. Res., 76, 190, 1982.
    57. C. F. Curtis et al., The relative efficacy of repellents against mosquito vectors of disease, Med. Vet.
        Entomol., 1, 109, 1987.
    58. P. Martin and P. Bateson, Measuring Behaviour, an Introductory Guide, 2nd ed., Cambridge: Cambridge
        University Press, 1993.
    59. V. G. Dethier, Notes on the biting response of tsetse flies, Am. J. Trop. Med. Hyg., 3, 160, 1954.
    60. R. Galun, E. M. Kosower, and N. S. Kosower, Effect of methyl phenyldiazencarboxylate (azoester) on the
        feeding behaviour of blood sucking invertebrates, Nature, 224, 181, 1969.
    61. D. J. Fryauff et al., Contact toxicity of permethrin-impregnated fabric to Hyalomma anatolicum
        excavatum (Acari: Ixodidae): Effects of laundering and exposure and recovery times, J. Med.
        Entomol., 35, 335, 1998.
    62. M. Bar-Zeev and S. Gothilf, Laboratory evaluation of tick repellents, J. Med. Entomol., 10, 71, 1973.
    63. M. Bar-Zeev and S. Gothilf, Laboratory evaluation of flea repellents, J. Med. Entomol., 9, 215, 1972.
    64. A. A. Khan, H. I. Maibach, and D. L. Skidmore, Insect repellents: Effect of mosquito and repellent-related
        factors on protection time, J. Econ. Entomol., 68, 43, 1975.
    65. D. R. Barnard et al., Mosquito density, biting rate and cage size effects on repellent tests, Med. Vet.
        Entomol., 12, 39, 1998.
    66. L. C. Rutledge et al., Efficacy of diethyl methylbenzamide (deet) against Aedes dorsalis and a comparison
        of two end points for protection time, J. Am. Mosq. Control Assoc., 5, 363, 1989.
    67. H. Lal, S. Ginocchio, and E. J. Hawrylewicz, Procedure for bioassaying mosquito repellents in laboratory
        animals, Proc. Soc. Exp. Biol. Med., 113, 770, 1963.
    68. P. Kashin and M. L. Kardatzke, An automated repellency assay system, I. Statistical methods, J. Econ.
        Entomol., 62, 10, 1969.


q 2006 by Taylor & Francis Group, LLC
Animal Models for Research and Development of Insect Repellents for Human Use                                   145


    69. P. Kashin and M. L. Kardatzke, Diurnal rhythm, age, and other variables affecting yellowfever mosquito
        avidity and the laboratory assay of repellents, J. Econ. Entomol., 68, 766, 1975.
    70. P. Kashin and B. E. Arneson, An automated repellency assay system, II. A new electronic “bitometer-
        timer”, J. Econ. Entomol., 62, 200, 1969.
    71. N. L. Sachdeva et al., Laboratory evaluation of some repellents against the chiggers of trombiculid mites,
        Med. J. Armed Forces India, 39, 39, 1983.
    72. O. M. Abu-shady et al., Some compounds as repellents against insescts, J. Egypt. Soc. Parasitol., 13, 223,
        1983.
    73. R. A. Wirtz et al., Laboratory testing of repellents against the tsetse Glossina morsitans (Diptera:
        Glossinidae), J. Med. Entomol., 22, 271, 1985.
    74. W. S. Choi et al., Repellent activities of essential oils and monoterpenes against Culex pipiens pallens,
        J. Am. Mosq. Control Assoc., 18, 348, 2002.
    75. M. Bar-Zeev and D. Ben-Tamar, Evaluation of mosquito repellents, Mosq. News, 31, 56, 1971.
    76. L. C. Rutledge, Some corrections to the record on insect repellents and attractants, J. Am. Mosq. Control
        Assoc., 4, 414, 1988.
    77. E. B. Wilson, An Introduction to Scientific Research, New York: Dover Publications, 1990.
    78. Z. Li et al., Studies on the repellent quwenling [in Chinese], Malaria Res., 6, 1974.
    79. A. C. Cohen, Truncated and Censored Samples: Theory and Application, New York: Marcel Dekker,
        1991.
    80. J. D. Kalbfleisch and R. L. Prentice, The Statistical Analysis of Failure Time Data, 2nd ed., Hoboken:
        Wiley, 2002.
    81. L. C. Rutledge et al., Mathematical models of the effectiveness and persistence of mosquito repellents,
        J. Am. Mosq. Control Assoc., 1, 56, 1985.
    82. T. Poston and I. Stewart, Catastrophe Theory and its Applications, Mineola, NY: Dover Publications,
        1996.
    83. Y. Shirai et al., L-Lactic acid as a mosquito (Diptera: Culicidae) repellent on human and mouse skin,
        J. Med. Entomol., 38, 51, 2001.
    84. R. G. D. Steel and J. H. Torrie, Principles and Procedures of Statistics: A Biometrical Approach, 2nd ed.,
        New York: McGraw-Hill, 1980.
    85. H. C. Fryer et al., Techniques for conducting fly-repellency tests on cattle, J. Econ. Entomol., 36, 33, 1943.
    86. H. C. Fryer, F. W. Atkeson, and R. C. Smith, Comparison of methods for testing repellent type fly sprays,
        J. Econ. Entomol., 41, 80, 1948.
    87. F. M. Wadley, Incomplete-block design adapted to paired tests of mosquito repellents, Biometrics Bull., 2,
        30, 1946.
    88. R. M. Altman and C. N. Smith, Investigations of repellents for protection against mosquitoes in Alaska,
        1953, J. Econ. Entomol., 48, 67, 1955.
    89. American Society for Testing and Materials, Standard Method of Field Testing Topical Applications of
        Compounds as Repellents for Medically Important and Pest Arthropods (Including Insects, Ticks, and
        Mites), Philadelphia: American Society for Testing and Materials, 1983.
    90. G. H. Yeoman et al., Repellency of the butyl ester of 3-methylcinchonic acid against Stomoxys calcitrans,
        the stable fly, Vet. Rec., 83, 131, 1968.
    91. L. L. Robert et al., Comparative sensitivity of four Anopheles (Diptera: Culicidae) to five repellents,
        J. Med. Entomol., 28, 417, 1991.
    92. D. J. Finney, Probit Analysis, 3rd ed., London: Cambridge University Press, 1971.
    93. A. Goldstein, Biostatistics: An Introductory Text, New York: Macmillan, 1964.
    94. A. F. Naylor, Comparisons of regression constants fitted by maximum likelihood to four common
        transformations of binomial data, Ann. Hum. Genet., 27, 241, 1964.
    95. American Society for Testing and Materials, Standard Test Methods for Laboratory Testing of
        Noncommercial Mosquito Repellant Formulations on the Skin, Philadelphia: American Society for
        Testing and Materials, 1983.
    96. L. C. Rutledge et al., A field trial of ethyl hexanediol against Aedes dorsalis in Sonoma County, California,
        J. Am. Mosq. Control Assoc., 5, 374, 1989.
    97. W. E. Deming, Statistical Adjustment of Data, New York: Dover Publications, 1964.


q 2006 by Taylor & Francis Group, LLC
146                                                        Insect Repellents: Principles, Methods, and Uses


    98. L. C. Rutledge et al., Evaluation of the laboratory mouse model for screening topical mosquito repellents,
        J. Am. Mosq. Control Assoc., 10, 565, 1994.
    99. Von R. Wiesmann and R. Lotmar, Observations and investigations on the effectiveness of the new “Kik-
        Geigy” repellent [in German], Acta Trop., 6, 292, 1949.
   100. P. I. Rakhmanova, A method of initial selection of mosquito repellents using animals under laboratory
        conditions [in Russian], Med. Parazitol. Parazitar. Bolezni., 28, 620, 1959.
   101. O. V. Smirnov, V. S. Suvorov, and H. P. Bocharov, New tests of certain repellents [in Russian], Med.
        Parazitol. Parazitar. Bolezni, 30, 613, 1961.
   102. N. V. Troitskii, Experimental use of carbon disulfide, chloropicrin, K-soap and other insecticides to
        destroy and repel Ornithodoros papillipes [in Russian], Med. Parazitol. Parazitar. Bolezni, 14, 75, 1945.
   103. M. Y. Liu, Mosquito repellent activity of a series of 2-(N-alkylamino) acetaldehyde dialkyl acetals and
        their N-acyl derivatives [in Chinese], Sinozoologica, 1, 205, 1981.
   104. J. Mandel, The Statistical Analysis of Experimental Data, New York: Dover Publications, 1964.




q 2006 by Taylor & Francis Group, LLC
8
Techniques for Evaluating Repellents


John M. Govere and David N. Durrheim



CONTENTS
Introduction ...................................................................................................................................147
Important Variables When Evaluating Repellents .......................................................................149
Evaluating Repellents Against Blood-Feeding Arthropods .........................................................151
  Ethical Considerations...............................................................................................................151
  Efficacy Testing .........................................................................................................................152
  Amount of Repellent .................................................................................................................153
  Data Recording ..........................................................................................................................153
Evaluating Compounds as Repellents of Mosquitoes and Biting Flies.......................................153
Evaluating Compounds as Repellents of Crawling Arthropods ..................................................154
  Fleas ...........................................................................................................................................154
  Ticks...........................................................................................................................................155
  Chigger Mites ............................................................................................................................156
General Considerations for Evaluating Repellent Candles, Coils, and Vaporizing Mats...........156
General Considerations for Evaluating Treated Articles or Clothing..........................................156
General Principles for Obtaining Valid and Reliable Results .....................................................157
References .....................................................................................................................................157


Editors’ note: Drs. Govere and Durheim have produced a comprehensive guide to repellent product
testing based on their own experiences in the field, including a justification of the process based on a
partial review of medical entomology. The chapter is presented in the form of definite steps for a
successful test based on specific assumptions. Some of the procedures recommended by the authors are
controversial (e.g., the use of a subject as his own control or rinsing the skin with alcohol prior to
application of the repellent product). The editors realize that, in reality, the assumptions are often
violated and that every technique is a compromise between practicality and precision. Nonetheless, the
chapter should be valuable as a foundation for considerations necessary for reliable evaluations of
repellent products.




Introduction
Arthropod-transmitted pathogens remain a major source of morbidity and mortality worldwide.1
The wide array of arthropod-borne pathogens constitute an enormous public health burden, particularly


                                                                                                                                                147

q 2006 by Taylor & Francis Group, LLC
148                                                     Insect Repellents: Principles, Methods, and Uses


TABLE 8.1
Medically Important Biting Diptera and Other Arthropods
Vectors                                                                      Diseases

Mosquitoes (Culicidae)
Anopheles                                       Malaria, lymphatic filariasis
Culex                                           Lymphatic filariasis, Japanese encephalitis, certain arboviruses
Aedes                                           Yellow fever, Lympatic filariasis, dengue fever, certain arboviruses
Mansonia                                        Lymphatic filariasis
Other biting Diptera
Tsetse flies (Glossina)                          African sleeping sickness
Blackflies (Simulium)                            River blindness (onchocerciasis)
Sandflies (Phlebotomus, Lutzomyia)               Leishmaniasis, sandfly fever
Horseflies (Tabanidae)                           Loiasis, tularaemia
Stable flies (Stomoxys)
Biting midges (Ceratopogonidae)                 Mansonellosis
Other biting Arthropods
Fleas (Siphonaptera)                            Plague, flea-borne typhus
Chigoe Fleas (Tunga penetrans)                  Jigger infection
Head louse (Pediculus humanus capitis)          Pediculosis
Body louse (Pediculus humanus humanus)          Trench fever, louse-borne relapsing fever, epidemic typhus
Pubic louse (Pthirus pubis)
Hard ticks (Ixodidae)                           Arboviral encephalitis and fevers, Lyme disease
Soft ticks (Argasidae)                          Tick-borne relapsing fever
Biting mites (Trombiculidae)                    Scrub typhus


in developing tropical countries (Table 8.1). Mosquitoes transmit pathogens to more than 700 million
people annually and the diseases they cause are estimated to be responsible for one out of every 17 deaths
globally.2 Malaria is undoubtedly the most important of the diseases caused by mosquito-transmitted
pathogens; it is responsible for as many as three million deaths and 500 million episodes of illness each
year.3 Mosquitoes also transmit the arboviruses responsible for yellow fever, dengue hemorrhagic fever,
epidemic polyarthritis (including Ross River and Barmah Forest viruses), and several forms of
encephalitis. The filarial nematodes are another group of pathogens transmitted by mosquitoes. These
parasites cause lymphatic filariasis, the second most common cause of chronic disability world-wide.4
   Ticks are vectors of a large number of diseases of animal and humans. Human diseases associated with
pathogens transmitted by ticks include tick-borne relapsing fever, Rocky Mountain spotted fever, Q fever,
Lyme disease, and others. Chiggers are both the reservoir and the vector for the pathogen that causes scrub
typhus, a disease that accounts for up to 20% of all fever presentations in parts of Asia.5 Fleas are also
important vectors of disease, including bubonic plague and flea-borne endemic typhus.4 In addition to
disease, many insects, including mosquitoes, biting flies, fleas, and ticks, are a source of mental
anguish—causing intense annoyance and sleep disturbance. Although most biting Diptera also feed on
plant juices, females generally need a blood meal for egg development.6
   Given the enormous burden of disease resulting from arthropods it is not surprising that humankind
has invested in a vast assortment of methods for controlling insect pests and vectors. Personal protection
has become an increasingly popular method for preventing contact with arthropods, as community vector
control is not always available. There is an understanding that even if personal protection is not
completely effective in preventing exposure to infectious biting arthropods, it can significantly reduce
personal risk by decreasing the chance of being bitten by an infectious vector, complementing other
strategies to reduce risk.
   Unfortunately, many commercially available products, as well as traditional home remedies that are
more affordable in some regions, are not very effective. As a false sense of security based on application
of an ineffective repellent may have devastating consequences, it is crucial that all products alleged to


q 2006 by Taylor & Francis Group, LLC
Techniques for Evaluating Repellents                                                                       149


have repellent properties are rigorously evaluated, preferably in a standardized fashion that facilitates
comparisons between studies.
   An ideal insect repellent should be active against multiple species of biting arthropods, remain
effective for at least eight hours, cause no topical irritation of the skin or mucous membranes, cause no
systemic toxicity, resist physical removal by rubbing, and be greaseless and odorless. If a repellant is to
be used on surfaces or materials, it should not be abrasive or damaging to the target material. This chapter
introduces approaches that have proven valuable in evaluating repellents against a number of arthropod
pests and vectors.



Important Variables When Evaluating Repellents
The attractiveness of different persons to insects varies substantially.7–9 Adults are more likely to be
bitten than children.10,11 Men are more readily bitten than women and larger persons attract more insects,
perhaps because of greater relative heat and carbon dioxide output.12,13 Recent research suggests that
malaria-infected children with gametocytes are more attractive to malaria vectors.14
   Biting insects are attracted to dark clothing, carbon dioxide, lactic acid, floral or fruity fragrances, skin
temperature, and moisture.15–17 Mosquitoes become restless when there is an increase of carbon dioxide
in their vicinity; they tend to fly in the direction of the carbon dioxide. Sensory hairs on the mosquito’s
antennae detect changes in carbon dioxide content, humidity and temperature of the air. If the sensors
detect a decrease in carbon dioxide, humidity, or temperature, the mosquito turns aside. Repellent
molecules block receptor sites in mosquito sensory hairs causing a mosquito to avert their potential
human target.18
   The environment, and particularly climatic conditions, can affect biting behavior. This is a significant
issue during field trials when wind, cooler temperature, and rainfall can markedly decrease feeding.
Seasonal fluctuation in vector abundance and population fluctuations in response to environmental
factors are also important considerations in attempting to conduct and interpret field trials.
   Additionally, an understanding of vector feeding preferences is notable. This includes preferences
for feeding indoors or outdoors as well as preferences for feeding at certain times of day or night
(Table 8.2).
   Insect bites are not randomly distributed on the human body (Table 8.3). This concept has been most
thoroughly studied in relation to mosquito disease vectors. Anopheles gambiae s.s. prefers to bite the feet
of seated humans and Anopheles arabiensis mosquitoes prefer to bite the ankles and feet of motionless
humans.9,19–21 The absolute number of Anopheles arabiensis mosquitoes biting motionless humans
drops dramatically when their feet are covered with shoes without a significant shift to other parts of the
leg or remainder of the body.22 Simulium damnosum in West Africa predominantly bites on the leg.23
Aedes simpsoni prefers to bite the face of naked humans, while Eretmopodites chrysogaster prefers to
bite ankles and feet.24 Sabethes belisarioi appears to exclusively bite human noses.25 Anopheles
albimanus mosquitoes prefer to bite the head and neck, Anopheles atroparvus prefers to bite the head
and shoulders, and Culex pipiens and Culex quinquefasciatus bite on the lower half of the body.26–28
Anopheles farauti prefers to bite near the ground and Anopheles gambiae s.s has a strong preference for
the feet of seated humans.29 Anopheles arabiensis has a strong biting preference for the ankles and feet of
individuals sitting on camp chairs.9,30,31
   An understanding of the various preferences involved in mosquito feeding behavior should be woven
into field trials, as these characteristics can provide opportunities for targeted prevention efforts. For
example, in a South African randomized cross-over study of Anopheles arabiensis it was demonstrated
that mosquito bites could be reduced by almost 70% when only the ankles and feet were treated with deet
repellent (Table 8.4a and Table 8.4b).9 The application of affordable repellents could certainly
complement the use of other techniques for control of a number of vector-borne pathogens, particularly
those in which human behavior and mosquito feeding temporally intersect in an outdoor environment.


q 2006 by Taylor & Francis Group, LLC
150                                                               Insect Repellents: Principles, Methods, and Uses


TABLE 8.2
Typical Biting Behavior, Day/Night and Indoor/Outdoor Preference of Biting Diptera and Other Arthropods
Pest/Vector                             Blood-Feeding Stage          Indoor/Outdoor Biting         Day/Night Biting

Anopheles                Female adult                                     Indoor                    Night
Culex                    Female adult                                     Indoor                    Night
Aedes                    Female adult                                     Outdoor/indoor            Day
Mansonia                 Female adult                                     Indoor                    Night
Tsetse flies              Female and male adults                           Outdoor                   Day
Triatomine bugs          Female and male nymphs and adults                Indoor/outdoor            Night
Bedbugs                  Female and male nymphs and adults                Indoor                    Night
Black flies               Female adult                                     Outdoor                   Day
Sand flies                Female adult                                     Outdoor/indoor            Night
Horse flies               Female adult                                     Outdoor                   Day
Stable flies              Female and male adults                           Outdoor/indoor            Day
Fleas                    Female and male nymphs and adults                Indoor/outdoor            Day or night
Chigoe fleas              Female and male nymphs and adults                Indoor/outdoor            Day or night
Head louse               Female and male nymphs and adults                Indoor/outdoor            Day or night
Body louse               Female and male nymphs and adults                Indoor/outdoor            Day or night
Pubic louse              Female and male nymphs and adults                Indoor/outdoor            Day or night
Hard ticks               Female and male nymphs and adults                Indoor/outdoor            Day or night
Soft ticks               Female and male nymphs and adults                Indoor/outdoor            Day or night
Biting mites             Female and male nymphs and adults                Indoor/outdoor            Day or night


This knowledge has also been successfully applied, at least on one occasion, to control a focal malaria
epidemic by topical application of 15% deet to the feet and ankles of an affected community.32
   The variety of factors that influence the field effectiveness of a repellent—including the frequency and
uniformity of application, number and species of insects attempting to bite, an individual’s attractiveness
to blood-sucking arthropods, an individual’s level of physical activity. amount of abrasion of treated skin
by clothing, evaporation and absorption from the skin surface, wash-off from rain or sweat, prevailing
temperature, and degree of wind disturbance—must be controlled during evaluation to allow comparison.
Certain factors are easier to control, like abrasion, which can be reduced by limiting movement of the
experimental subjects. Others, including interpersonal differences, may be controlled by using the
individual as their own control or through randomized cross-over study designs. Climate variables
should be carefully recorded and described, and repeated measures made under different conditions to
ensure that findings are reproducible.

            TABLE 8.3
            Anatomical Biting Preferences of Selected Mosquitoes on Humans
            Mosquito Species                                       Site                         Reference

            Eretmopodites chrysogaster                        Ankles & feet                        24
            Aedes simpsoni                                    Face                                 24
            Sabethes belisarioi                               Human noses                          25
            Anopheles albimanus                               Head & neck                          26
            Anopheles atroparvus                              Head & shoulders                     23
            Aedes aegypti                                     Head & shoulders                     23
            Culex pipiens                                     Lower half of body                   28
            Culex quinquefasciatus                            Lower half of body                   28
            Anopheles farauti                                 Near ground level                    29
            Anopheles gambiae s.s.                            Near ground level                    20
            Anopheles arabiensis                              Ankles & feet                        31
            Anopheles arabiensis                              Ankles & feet                        32



q 2006 by Taylor & Francis Group, LLC
Techniques for Evaluating Repellents                                                                           151


TABLE 8.4a
Distribution of Mosquito Bites on the Human Body when Ankles and Feet Were Untreated
                                                                 Nights
Body Part                1                  2           3                 4             5                Total (%)

Ankle/feet         48                   54          145              102            72                  421 (81.1)
Legs                5                    8           30               31            11                   85 (16.4)
Arms                1                    0            1                4             1                    7 (1.3)
Body                1                    1            3                1             0                    6 (1.2)
Total (%)          55 (10.6)            63 (12.1)   179 (34.5)       138 (26.6)     84 (16.2)           519 (100.0)

Source: J. M. Govere, L. E. O. Braack, D. N. Durrheim, R. H. Hunt, and M. Coetzee Med. Vet. Entomol., 15, 287, 2001.



TABLE 8.4b
Distribution of Mosquito Bites on the Human Body when Ankles and Feet Were Treated with Deet
                                                                 Nights
Body Part                1                  2           3                 4              5               Total (%)

Ankle/feet          0                    0           0                0              0                    0 (0.0)
Legs               32                   15          28               39             27                  141 (88.1)
Arms                0                    0           2                2              0                    4 (2.5)
Body                3                    0           9                1              2                   15 (9.4)
Total (%)          35 (21.8)            15 (9.4)    39 (24.4)        42 (26.3)      29 (18.1)           160 (100.0)
% Protection       36.4                 76.2        78.2             69.6           65.5                 69.2

Source: J. M. Govere, L. E. O. Braack, D. N. Durrheim, R. H. Hunt, and M. Coetzee Med. Vet. Entomol., 15, 287, 2001.




Evaluating Repellents Against Blood-Feeding Arthropods
Repellent testing procedures are conducted through both laboratory evaluation and field testing;
although a myriad of testing procedures have been described, few have been widely used. Most
validated procedures relate to medically important arthropods with a special emphasis on mosquito
repellent testing. This chapter will focus on generally accepted, reliable and practical approaches
for testing the performance of compounds that purport to repel mosquitoes, biting flies, fleas,
chiggers, and ticks from human skin or from the environment near people. As a variety of
formulations are possible, including liquid or pressurized products for spray treatments, material or
article impregnation, lotions, coils, candles or vaporizing mats, testing must include the end-use
product formulation.



Ethical Considerations
It is generally accepted as unethical to expose a person to an experimental chemical compound without
fully informed consent from that subject. Information provided to the volunteer subject must include a
full description of the nature and purpose of the test, and any physical or mental health consequences that
are reasonably foreseeable. The subject must be guaranteed the option to withdraw at any stage without
prejudice. It is essential to ensure that the insects used are not infected with known human pathogens.
Some of the arguments against using humans as test subjects in laboratory and field tests include


q 2006 by Taylor & Francis Group, LLC
152                                                      Insect Repellents: Principles, Methods, and Uses


concerns regarding ethical considerations, sensitization reactions following tick, flea or mite bites, and
poorly established toxicity profiles of chemicals. It is therefore important that the toxicokinetics
(absorption, biodistribution, metabolism, and excretion) of test compounds be determined in animal
models. However, products developed for human use should, whenever possible in the course of
development and evaluation, be tested on people.
   Although it may be considered preferable to fully assess the safety profile of a product, including
potential for topical irritation, prior to exploring its repellency, initial laboratory screening of promising
candidates on a human arm is often conducted without prior skin irritation studies. Treated cotton
stocking is worn over the arm or a treated cloth is tested over untreated cloth that covers the skin surface
so that chemicals are not in direct contact with the skin surface.33,34



Efficacy Testing
The number of test subjects required depends on the purported duration of effect. For a product with
1–4 h of repellency, at least five treated test subjects should be used. For a label claim of 5 or more
hours of repellency, at least 10 treated test subjects should be used. Similar numbers of adult male
and female test subjects are preferable. Test subjects should not ingest alcohol or caffeine and avoid
applying fragrant products (e.g., perfumes, colognes, hair sprays, and lotions) for at least 12 h before
the testing.
   The behavior of the species of biting arthropods that are the subject of the study should be examined
during the trial. Biting frequency on untreated skin is used to determine avidity of flying insects. Tick
drags made of white flannel cloth can be pulled over the ground and low vegetation to identify heavy tick
infestations. Chigger mites are located by laying black plates on the ground.
   Ideally, untreated subjects should be used as controls. Test subjects should be at least 3 m apart during
the test and may engage in usual outdoor activity, including non-vigorous movements like intermittent
slow walking, standing, squatting, sitting, and raising or lowering arms. Tobacco should not be used
during testing.
   Many studies have been based on treatment of the test subject’s forearm (wrist to elbow), but the
lower leg may also be used. The exposed surface area (in cm2) of each test subject should be carefully
calculated, by measuring the circumference of the arm at the wrist, the elbow, and three to four equally
spaced points in-between, and then multiplying the average of these circumference measures by the
distance from the wrist to elbow; the same method for calculating surface area can also be applied to
the lower leg, measuring from the ankle to the knee. The upper arm or leg and hand or foot should be
covered with a material impenetrable to the insect’s proboscis. Dark colors should be avoided and latex
gloves may be used to cover the subject’s hands. The test area should be washed with unscented soap,
rinsed first with water and then with a solution of 70% ethanol in water, and finally dried with
a clean towel.
   A test subject should receive no more that one treatment per test, potentially replicated on each limb.
Test subjects should avoid exertion, which might increase perspiration or abrasion. The treated area
should also not be rubbed, touched, or wetted. Other body parts, including the face, back, and non-test
limbs should be adequately protected with gloves, head net, and protective clothing so that biting
pressure is concentrated on the exposed treated skin. Field evaluation of repellents on skin should only be
conducted after favorable toxicology has been established for the test chemicals.
   A subject’s forearms can be used in paired tests to determine protection time, which is calculated
to the first bite, and confirmed by second and third bites within 5 min. A number of factors can affect
results, including the species being evaluated, the density of insects, age and gonotrophic state of the
insect population, age of host, time of day, and temperature and humidity. To effectively deal with
multiple factors in the analysis, a larger number of test subjects and more test replications
are necessary.


q 2006 by Taylor & Francis Group, LLC
Techniques for Evaluating Repellents                                                                    153


Amount of Repellent
The test formulation should be stored at room temperature and ambient humidity before testing. The
time since production of the test formulation should be recorded. Generally, products should be less
than 12 months old. Standard application quantities are 1 g of liquid aerosol or pump spray test
material, or 1–1.5 g of cream, lotion or stick, per 600 cm2 of test area. This should be evenly applied to
the forearm or lower leg. It is important to confirm that the quantity per unit area does not vary by more
than 5% for all subjects or replicates.


Data Recording
The number of bites and probes should be recorded by the investigator rather than the test subject. The
duration of repellent protection should be recorded for each test subject and for each test site.
Traditionally test results are reported for complete protection time and 95% repellency.
   Complete protection time is a measure of the duration of repellent protection until the time of first bite
for each test subject, usually confirmed by a second bite occurring within the same time interval. The
mean protection time and standard error are then calculated for each test species across subjects.
Statistical testing should be used to examine variability between repetitions and between means.
   The duration of repellent protection based on the period with 95% reduction in bites for each
test subject is referred to as 95% repellency. The mean protection time and standard error based on a
95% reduction in bites for each test species should be reported. Statistical testing may be used for
examining variability between repetitions and means. Survival analysis may also be used when
comparing multiple products. It is important that the choice of statistical method selected be
clearly explained.




Evaluating Compounds as Repellents of Mosquitoes and Biting Flies
There is a rich tradition of testing mosquito repellents, with the first well-planned laboratory evaluation
conducted in 1919.35 The principal mosquito specie used in tests for mosquito repellency is Aedes
aegypti, which is relatively easily reared and maintained, and an avid blood feeder even in the laboratory.
However, compounds have differential effectiveness against other vector species. For example,
a repellent considered poorly effective against Aedes aegypti was found to be highly effective against
deer flies (Chrysops spp.)36 It is therefore important to test compounds against the specific target species
of interest.
   Generally, laboratory testing of mosquitoes should include at least three genera of human biters;
Aedes aegypti, an Anopheles species, and a Culex species. When reporting on either laboratory or
field testing, it is important to identify test insects by genus and species, and by subspecies or strain,
particularly with mosquitoes. With field testing, identity of the insect should be confirmed by
aspirating specimens into a vial for laboratory identification before testing begins. Biting pressure
should be periodically determined throughout the test. Laboratory mosquitoes should be adult females
5–10 days old. Stable flies should be 3 days old. The age or age range of the test insects should
be reported.
   Larvae should be reared in the laboratory under optimal conditions for the particular species. As
a general guide, most species should be reared at 27G38C, with a relative humidity of 80G10%, and a
photoperiod of 16:8 h (light:dark). Other conditions may be used where appropriate for a particular
species, with any alternative rearing techniques justified in the study summary. Adults should be fed 10%
sucrose and no blood meal should be offered before the test. Test insects should be starved for 12 h
immediately before the test, used for only a single test, and destroyed after the trial.


q 2006 by Taylor & Francis Group, LLC
154                                                     Insect Repellents: Principles, Methods, and Uses


   It is common practice that at least one mosquito should be introduced for every 100 cm3 of cage space
with at least 200 mosquitoes introduced in each test cage. Similarly, for stable flies, at least one stable fly
is commonly introduced for each 500 cm3 of cage space with at least 45 stable flies introduced in each
test cage.
   Test cages should be at least 20,000 cm in volume, square or rectangular in shape, and feature one
sleeved opening for the subject’s arm. Each cage should be used for only one test subject and treatment at
a time. Temperature should be maintained at 22–278C, relative humidity at 50% to 80%, and lights
should be kept on during testing.
   An untreated (negative) control is recommended to verify biting pressure. There is some debate as to
the most appropriate control, with protagonists arguing in favor of either using the untreated forearm or
lower leg of the same test subject, or another untreated individual. Both approaches have advantages and
disadvantages. The control forearm or lower leg should be prepared, washed, rinsed, and dried in
precisely the same manner as the treated forearm or lower leg. Before the test begins, subjects should
expose their untreated forearm to the mosquitoes or stable flies in the test cage to establish their
attractiveness. It is recommended that a minimum of 10 mosquitoes land and probe within 30 s, or five
stable flies land and probe within 60 s, for a subject to participate. Every hour, an untreated forearm or
lower limb should be inserted through the sleeve into the cage and exposed to mosquitoes for up to 30 s,
or to stable flies up to 60 s, to verify biting pressure. The forearm or lower limb should be removed from
the test cage as soon as it has received the necessary number of probes. Probing is preferred to biting so
that a subject’s discomfort is limited.
   Thirty minutes after treatment with repellent, the treated area should be inserted through the sleeve
into the cage for 5 min. This allows sufficient time for the repellent to dry and still tests the minimum
reasonable protection time that might be of practical value. The number of bites or probes in each
exposure period should be recorded. The treated area should be exposed for 5 min every 30 min while
biting pressure lasts, i.e., until the control area no longer receives 10 mosquito landings in 30 s or five
stable flies landings in 60 s. Test subjects should avoid rubbing their arm or leg when introducing or
removing it from the cage and between exposure periods.
   An alternative approach, particularly when comparing products, is to insert the treated limb into the
cage for one minute, and if not bitten, to reinsert the limb for one minute every 5 min, up to 2 h. If biting
still does not occur, then the interval can be extended to 15 min. If at any point mosquitoes begin landing
but not biting (a behavior that occurs when the efficacy of a repellent begins to wane), then the intervals
between insertions could be reduced to 5 min.
   Field testing should be conducted at a minimum of two field sites in environmentally distinctive
habitats (e.g., forest, grassland, salt marsh, wet land, beach, barns, or an urban environment) suitable for
the target insect. For mosquitoes, different species prefer various habitats. Habitats where biting pressure
is below the levels described previously are unlikely to provide reliable and reproducible results. It is
important to record details of weather conditions during the test, including temperature, relative
humidity, cloud cover, precipitation, light intensity, and wind speed, allowing 90 s of observation for
each exposure period. It is important that wind speed does not exceed 10 mph as windy conditions cause
diminished probing.




Evaluating Compounds as Repellents of Crawling Arthropods
Fleas
The cat flea, Ctenocephalides felis, is the preferred model flea for repellent testing. Adult male and/or
female fleas that are 5–10 days old, reared at 27G38C, with a relative humidity of 80G10%, and a
photoperiod of 16:8 h (light:dark) should be used. The adult fleas should not be blood-fed, and after one
trial they should be destroyed. There should be one flea per 9 cm3 and at least 100 fleas in each test cage.


q 2006 by Taylor & Francis Group, LLC
Techniques for Evaluating Repellents                                                                     155


Twenty-five fleas should be added to the test cage after each exposure period. Cages should be at least
900 cm3 in volume, square, circular, or rectangular in shape, and made of plastic or glass with an opening
on top to insert the test subject’s limb. The cages should have a rough floor utilizing a material such as
clean sand. Replications should be limited to one test subject and treatment at a time for each cage.
The temperature during testing should be maintained at 22–278C, with relative humidity at 50–80%,
and the lights should be on.
   A negative (untreated) control is recommended to verify biting pressure, with the negative control
being either the test subject’s untreated forearm or lower leg, or an untreated subject. The control limb
should be washed, rinsed, and dried in exactly the same way as the treated area. Before testing begins,
subjects should expose their forearms to the fleas in the test cage to establish their attractiveness. It is
recommended that to qualify as a participant, a subject should experience a minimum of 10 flea landings
or probes within 30 s. Every hour, a control limb should be inserted through the sleeve into the cage and
exposed to the fleas for up to 30 s to verify biting pressure, with the limb removed as soon as 10 landings
have occurred.
   Thirty minutes after treatment with repellent, the test subject’s forearm or lower leg should be inserted
through the sleeve into the cage for 5 min and the number of landings recorded for each exposure period.
This should be repeated every 30 min while biting pressure lasts, that is, until the control no longer
receives 10 flea landings in 30 s. The duration of repellent protection for each test subject should be
recorded and the mean protection time and standard error reported.
   An alternative method for testing compounds against fleas requires use of two strips of fabric—one
impregnated with a chemical and one untreated control. These are lowered into a container into which
fleas have been added.37 After a predetermined time, the strips are removed, the fleas remaining on the
cloth strips are counted, and the percentage of repellency is calculated. Use of an olfactometer for
comparative testing of new repellents against fleas has also been described.38


Ticks
Animals, such as gerbils, have been used for evaluating repellents against crawling insects.39 The
animal may be immobilized in a stanchion with its shaved abdomen exposed and two identically sized
areas treated with a candidate repellent and a standard. An alternative strategy is to dip or spray the
caged and restrained animal and then place it on the periphery of a tick or mite-infested area.
Following a specific exposure time, the engorged and attached ticks or mites are counted to determine
protection afforded by the repellent. Results from laboratory testing may not always correlate to field
evaluation. Similarly, performance in animal testing may not fully correlate to performance
on humans.
   Tick species used for evaluation should be disease free and represent both ixodid (hard) and argasid
(soft) ticks. Adult and nymphal ticks should be tested, since both life stages can be involved in pathogen
transmission. Ticks should be reared at 22G38C, with a relative humidity of 50–80%, and a photoperiod
of 16:8 h (light:dark). Ticks used for testing should be destroyed after a single trial. Five ticks should be
exposed to the treated forearm or lower leg in each exposure period, keeping the temperature during the
test at 22–278C, with a relative humidity of 50–80%, and the light on. The duration of repellent protection
for each subject should be recorded.
   As an alternative to direct testing on humans, treated cloth patches may be placed on a paddle and
touched to the bottom of a pen infested with ticks or mites. The number of ticks or mites crawling from
the untreated part of the paddle to a point midway up the treated patch allows evaluation of repellency
when compared with an untreated control paddle.
   A means for determining the minimum effective dosage of a repellent against ticks has also been
described.39 The candidate material is applied in horizontal stripes of progressively increasing
concentrations. Ticks then climb the vertically positioned fabric until they reach a concentration they
cannot tolerate, as indicated by the ticks dropping off of the surface.


q 2006 by Taylor & Francis Group, LLC
156                                                      Insect Repellents: Principles, Methods, and Uses


Chigger Mites
Chigger mites include a number of genera in the family Trombiculidae, but most tests have involved
either Eutrombicula or Leptotrombidium. Larval, unfed chigger mites reared at 22G38C, at a relative
humidity of 50–80%, and a photoperiod of 16:8 h (light:dark) should be used. The age of the mites should
be recorded. Mites used in a trial should be destroyed afterwards. Five mites should be exposed to the
treated area of skin during each exposure period, while keeping the temperature at 22–278C, relative
humidity at 50–80%, and the light on.
   The area above and below the test area should be covered with material that chigger mite mouthparts
cannot penetrate. A negative (untreated) control, usually an untreated surface on the same subject, is
recommended to verify biting pressure. Test organisms should be picked up with a soft artist’s paintbrush
and placed on the test subject about 2 cm from the area of the forearm where the repellent has been
applied, near the wrist, with a new tick or mite placed 2 cm below the test area once the previous
specimen has crossed onto the test area. After moving toward the margin of the test area, chigger mites
should be allowed 5 min to cross the margin onto the test material (toward the elbow). Once the chigger
mite has been recorded as not repelled, it should be replaced with one that was not previously tested.
A new group of chigger mites should be exposed to the test material every 30 min. No test arthropods
should be reused.




General Considerations for Evaluating Repellent Candles, Coils,
and Vaporizing Mats
The species and biting pressure should be determined, test sites prepared and testing conditions recorded
as described above. If more than one test subject is exposed to the same candle, coil, or mat, then the
number of bites should be averaged. The number and placement of the intervention(s) should be
consistent with label directions or proposed use. Test subjects should be located at the maximum distance
of usefulness proposed or described. If the product description states that the candle, coil, or mat should
be placed upwind, then test subjects should remain downwind; otherwise, test subjects should move
around the circumference of the test area periodically with the time interval of movements reported in the
study results.
   A negative (untreated) control of the same size as the test area is desirable to determine biting pressure.
Control subjects should remain upwind, far enough from the treatment area as is necessary so as not to be
affected by the repellent. They should be exposed for the full period of activity of the candle, coil, or mat.
For coils, protection time should be the same as burning time.
   An investigator or study partner (not the test subject) should record the number of bites and probes.
When compared to the negative control, at least 50% of insects should be repelled. The duration of
repellent protection and mean time to 50% reduction in bites, with standard error, should be reported.
   An alternative approach that has merit for evaluation of area repellent systems is the use of carbon
dioxide-baited traps. These can be placed in the treated and control areas simultaneously and collections
made for equivalent time periods to allow for comparison. It is important that catches are correctly
identified, as the repellent method may have a differential effect against different species.




General Considerations for Evaluating Treated Articles or Clothing
Evaluations of repellent impregnated clothing or treated articles should report the repellent used,
impregnation formulation, method of impregnation, type of garment treated, and amount of repellent


q 2006 by Taylor & Francis Group, LLC
Techniques for Evaluating Repellents                                                                    157


absorbed per unit area of textile. Repellents may be used to treat a variety of materials, including those
used for making bed nets, tablecloths, loose jackets, and other clothing items.
   Reports of field tests should include details on type of material treated, mesh size for nets, weight
per unit area of material, impregnation formulation, method of impregnation, amount of repellent
absorbed per unit area, and method of exposure. The degree of protection between subjects using treated
articles or clothing should be compared to subjects in the same environment using the same but untreated
articles or clothing during a standard exposure period. A bite or probe should be recorded whenever an
arthropod proboscis penetrates the treated material.




General Principles for Obtaining Valid and Reliable Results
Evaluations should be as simple and practical as possible to encourage standard comparison and
universal acceptance by being easily understood and performed.40 Although candidate repellents should
be taken to the field to determine protection time and effectiveness under field conditions, it is preferable
to first conduct evaluations in a laboratory environment as it is easier to control for potential confounding
factors.41 Creative design may permit imitation of important individual field conditions in the laboratory.
A good example is the use of extractor fans to create draught.42 Testing should be performed on human
subjects to evaluate actual performance against the target host species. Standardized amounts of test
compounds should be applied and uniform coverage ensured.
   Where a “gold standard” repellent exists, it is sensible to compare promising candidates to this
product. It is important that the test subjects and the person recording probes or bites do not know which
compound has been applied (i.e., a double-blinded trial). Cross-over designs are useful to take account of
other potential confounding factors, but persistent and longer range effects of repellents must be factored
into the study design. The sequential application of all repellents to each individual is a preferred
strategy.43 It is then important that the sequence of application is randomized and that there is careful
cleaning of the test area after each application. Ideally, different products should be tested on different
days on the same individual. Meticulous recording of experimental conditions should allow easy
replication by other investigators. Biting rates on untreated skin should be recorded to assure adequate
biting pressure. Three to six replications, preferably on multiple subjects, should be conducted to
determine interpersonal variability and provide a mean protection time. This variability should also be
factored into study design and analysis as the differences in interpersonal attraction may be profound.44
   The potential for repellents to contribute to the integrated control of arthropod borne diseases of
humans has not yet been fully realized.45 The availability of affordable alternatives, including mosquito-
repellent plants or plant-derived natural products may make this complementary strategy more
feasible.46,47 However, before widespread use can be encouraged to prevent potentially life-threatening
diseases, it is essential that the efficacy and duration of effect is determined reliably.

References
      1. J. R. Roberts and J. R. Reigart, Does anything beat DEET?, Pediat. Ann., 33, 444, 2004.
      2. C. P. McHugh, Arthropods: Vectors of disease agents, Lab. Med., 25, 429, 1994.
      3. World Health Organization, International Travel and Health Vaccination Requirement and Health
         Advice, Geneva: WHO, 1995.
                  ¨
      4. D. N. Durrheim et al., Lymphatic filariasis endemicity—an indicator of poverty?, Trop. Med. Int.
         Health, 9, 843, 2004.
      5. D. Strickman, Scrub typhus, in The Encyclopedia of Arthropod-Transmitted Infections of Man and
         Domesticated Animals, M. W. Service (Ed.), Wallingford, U.K., CAB International Publishing, 2001,
         p. 456.
      6. A. N. Clements, The Physiology of Mosquitoes, Oxford: Pergamon Press, 1963.


q 2006 by Taylor & Francis Group, LLC
158                                                        Insect Repellents: Principles, Methods, and Uses


     7. C. F. Curtis, Personal protection methods against vectors of disease, Rev. Med. Vet. Entomol., 80, 543,
        1992.
     8. A. A. Khan, Mosquito attractants and repellents, in Chemical Control of Insect Behavior, H. H. Shorey
        and J. J. McKelvey (Eds.), New York: Wiley, 1977, p. 305.
     9. J. M. Govere et al., Effect of repellent on Anopheles arabiensis biting humans in Kruger Park, South
        Africa. Med. Vet. Entomol., 15, 287, 2001.
    10. R. C. Muirhead-Thomson, The distribution of anopheline mosquito bites among different age groups,
        Br. Med. J., 1, 1114, 1951.
    11. I. H. Gilbert, H. K. Gouck, and N. Smith, Attractiveness of men and women to Aedes aegypti and
        relative protection time obtained with DEET, Fla. Entomol., 49, 53, 1966.
    12. G. R. Port and P. F. L. Boreham, The relationship of host size to feeding by mosquitoes of the
        Anopheles gambiae Giles complex (Diptera: Culicidae), Bull. Entomol. Res., 70, 133, 1980.
    13. C. M. Gjullin, Effect of clothing color on the rate of attack of Aedes mosquitoes, J. Econ. Entomol., 40,
        326, 1947.
    14. R. Lacroix et al., Malaria infection increases attractiveness of humans to mosquitoes, PLoS Biol., 3,
        e298, 2005.
    15. C. E. Schreck, D. L. Kline, and D. A. Carlson, Mosquito attraction to substances from the skin of
        different humans, J. Am. Mosq. Control Assoc., 6, 406, 1990.
    16. R. Brouwer, Variations in human body odour as a cause of individual differences of attraction for
        malaria mosquitoes, Trop. Geogr. Med., 12, 186, 1960.
    17. M. Geier, H. Sass, and J. Boeckh, A search for components in human body odour that attract females
        of Aedes aegypti, in Olfaction in Mosquito-host Interaction, G. R. Bock and G. Cardew (Eds.), New
        York: Wiley, 1996, p. 132.
    18. E. E. Davis and M. F. Bowen, Sensory physiological basis for attraction in mosquitoes, J. Am. Mosq.
        Control Assoc., 10, 316, 1994.
    19. T. R. Burkot, Non random host selection by anopheline mosquitoes, Parasitol. Today, 4, 156, 1988.
    20. C. F. Curtis et al., The relative efficacy of repellents against mosquito vectors of disease, Med. Vet.
        Entomol., 1, 109, 1987.
    21. R. De Jong and B. G. J. Knols, Selection of biting sites on man by two malaria mosquito species,
        Experientia, 51, 80, 1995.
                 ¨
    22. D. N. Durrheim and J. M. Govere, Targeting vector behaviour and characteristics for effective malaria
        control in southern Africa, Appl. Env. Sci. Public Hlth., 1, 73, 2003.
    23. B. G. J. Knols, Odour-mediated host-seeking behaviour of the Afrotropical malaria vector An.
        gambiae Giles, Ph.D diss., University of Wageningen, The Netherlands, 1996.
    24. A. J. Haddow, Observation on the biting habits of African mosquitoes in the genus Eretmapodites
        Theobold, Bull. Entomol. Res., 14, 761, 1956.
    25. M. T. Gillies, The role of carbon dioxide in host-finding of mosquitoes (Diptera: Culicidae): A review,
        Bull. Entomol. Res., 70, 525, 1980.
    26. B. G. J. Knols and R. De Jong, Limburger cheese as an attractant for the malaria mosquito An. gambiae
        s.s., Parasitol. Today, 12, 159, 1996.
    27. B. G. J. Knols, R. De Jong, and W. Takken, Differential attractiveness of isolated humans to
        mosquitoes in Tanzania, Trans. R. Soc. Trop. Med. Hyg., 89, 604, 1995.
    28. A. S. Self, M. H. M. Abdulcader, and M. M. Tun, Preferred biting sites of Culex pipiens fatigans on
        adult Burmese males, Bull. World Health Org., 40, 324–327, 1969.
    29. J. D. Charlwood, R. Paru, and H. Dagaro, Raised platforms reduce mosquito bites, Trans. R. Soc. Trop.
        Med. Hyg., 78, 141, 1984.
    30. J. S. Keystone, Of bites and body odour, Lancet, 347, 1423, 1996.
    31. L. E. O. Braack et al., Biting pattern and host seeking behaviour of An. arabiensis (Diptera: Culicidae)
        in North-eastern South Africa, J. Med. Entomol., 31, 333, 1994.
    32. D. N. Durrheim and J. M. Govere, Malaria outbreak control in an African village by community
        application of “deet” mosquito repellent to ankles and feet, Med. Vet. Entomol., 16, 112, 2002.
    33. P. Granett, Studies of mosquito repellents I. Test procedure and method of evaluating test data,
        J. Econ. Entomol., 33, 563, 1940.


q 2006 by Taylor & Francis Group, LLC
Techniques for Evaluating Repellents                                                                       159


    34. J. K. Trigg, Evaluation of a eucalyptus-based repellent against Anopheles spp. in Tanzania, J. Am.
        Mosq. Control Assoc., 12, 243, 1996.
    35. A. Bacot and G. Talbot, The comparative effectiveness of certain culicifuges under laboratory
        conditions, Parasitology, 11, 221, 1919.
    36. C. E. Schreck, N. Smith, and H. K. Gouck, Repellency of N,N-diethyl-m-toluamide (deet) and
        2-hydroxyethyl cyclohexanecarboxylate against the deer fly Chrysops flavidus Wiedemann, J. Med.
        Entomol., 13, 115, 1976.
    37. M. L. Fedder, On the method of laboratory testing of insect repellents for different insects, Med.
        Parazitol., 30, 730, 1961.
    38. A. A. Potapov, Olfactometer for repellent testing on fleas, Med. Parazitol., 1, 97, 1968.
    39. M. Bar-Zeev and S. Gothilf, Laboratory evaluation of tick repellents, J. Med. Entomol., 10, 71, 1973.
    40. C. E. Schreck, Techniques for the evaluation of insect repellents: A critical review, Annu. Rev.
        Entomol., 22, 101, 1977.
    41. H. H. Yap, K. Jahangir, and J. Zairi, Field efficacy of four insect repellent products against vector
        mosquitoes in a tropical environment, J. Am. Mosq. Control Assoc., 16, 241, 2000.
    42. C. F. Curtis and N. Hill, Comparison of methods of repelling mosquitoes, Entomol. Exp. Appl., 49,
        175, 1998.
    43. M. S. Fradin and J. F. Day, Comparative efficacy of insect repellents against mosquito bites, N. Engl.
        J. Med., 347, 13, 2002.
    44. B. G. J. Knols, R. De Jong, and W. Takken, Differential attractiveness of isolated humans to
        mosquitoes in Tanzania, Trans. R. Soc. Trop. Med. Hyg., 89, 604, 1995.
    45. R. I. Rose, Pesticides and Public Health: Integrated methods of mosquito management, Emerg. Infect.
        Dis., 7, 17, 2001.
    46. A. Seyoum et al., Traditional use of mosquito-repellent plants in western Kenya and their evaluation in
        semi-field experimental huts against Anopheles gambiae: Ethnobotanical studies and application by
        thermal expulsion and direct burning, Trans. R. Soc. Trop. Med. Hyg., 96, 225, 2002.
    47. J. M. Govere et al., Local plants as repellents against Anopheles arabiensis, in Mpumalanga Province,
        South Africa, Cent. Afr. J. Med., 46, 213, 2000.




q 2006 by Taylor & Francis Group, LLC
q 2006 by Taylor & Francis Group, LLC
9
Use of Olfactometers for Determining Attractants and
Repellents


Jerry F. Butler



CONTENTS
Introduction ...................................................................................................................................161
Review of Literature .....................................................................................................................162
Olfactometer Test Systems ...........................................................................................................163
Research Objectives ......................................................................................................................165
Laboratory Olfactometer ...............................................................................................................165
Summary of Olfactometer Development......................................................................................167
Summary of Olfactometer Market Sample Tests .........................................................................168
Market Sample Results .................................................................................................................168
Summary of Olfactometer Research.............................................................................................169
References .....................................................................................................................................191




Introduction
The development of attractants and repellents for future management of blood-feeding flies, mosquitoes,
and other arthropods requires that behavioral regulating compounds (semiochemicals) first be identified.
Preferably, these materials should be noninsecticidal, so that they may reduce selective resistance
problems. When possible, semiochemical compounds should also be obtained from natural self-
protective systems that are developed through plant and animal evolution, as these materials will be
long term products of natural selection. These materials can be developed as attractants for baits and
traps, as repellents for individual hosts, and as area treatments for exclusion of pests. They would most
desirably be used to develop a push-pull system with attractant traps on the perimeter and repellents
placed on or near the hosts to first capture and then exclude blood feeders.
   Identification procedures for repellent and attractant semiochemicals have historically been based on
several approaches. “Trial and error” is the most common. In trial and error, test mixtures are placed into
traps as baits or on skin to see if they will work. “Ask the insect to choose its preference” is a second
approach. Here, semiochemicals within test systems where insects select or avoid points of treatment are
exposed and the insects choose the semiochemical that they prefer. The third system is called “ask the
insect to use its sensors to detect semiochemicals.” In this system, neural pathways to the insect’s brain


                                                                                                                                             161

q 2006 by Taylor & Francis Group, LLC
162                                                      Insect Repellents: Principles, Methods, and Uses


are probed to monitor detection. Neural activity reflects the insect both detecting and processing
information and indicates the importance of the compound to the insect. However, monitoring neural
activity requires skill in not only obtaining neural signals but also decoding their output to identify
whether or not the signal has a positive or negative identity. Whichever system is used, it is imperative to
make field evaluations that can determine the activity of semiochemicals under the proposed usage.




Review of Literature
Arthropods’ responses to semiochemical treatments can be identified based on physiological processes
that are triggered by chemoreception and result in either attractancy or repellency behavior.1–4 Chemical
cues may be derived from food sources found in host plants or animals; larval habitats such as decaying
organic material; oviposition sites in manure, soil, or water; and from the insects themselves. Selection
by the arthropod requires that they detect key elements by olfactory senses, located on the antennae and
maxillary palps, or visual senses.5–8
   The complex behavioral sequence that results in host seeking and host-location by hematophagous
arthropods involves an array of both chemical and physical cues.9–16 These clues originate from the
environment and are modified by visual cues, thermal effects, air movement, relative humidity, and
chemical stimuli.2,10,14,17–25 Host location at a distance is thought to be regulated by stimuli generated by
the host, such as an increase in temperature (detected as infrared radiation), carbon dioxide (CO2),
and other host generated gases.26 As the biting arthropod approaches the host, such stimuli orient
the arthropod to the final landing site.16,27–31 Vision has also been found to be important in host
location.15,21–23,32–40
   Presumably, insects detect odors because odors cause changes in the electrical activity of primary
olfactory receptor neurons contained within the antenna and maxillary palpal sensillae. Such nervous
signals can be measured in the laboratory at a cellular level using probes placed at the neural receptors
and the olfactory lobe of the insect’s brain.7,8,41
   In blood-feeding insects, olfactory components in host finding are regulated in part by the plume of
CO2 that serves as a primary attractant along with other gases, including blood gases.14,29,42
   Host location for parasitic species is complex. It requires that the parasite integrate all of the arthropod
senses while it simultaneously gathers momentum during the process of host location.26 The sequences
of behaviors involved in host searching are susceptible to manipulation or interference by humans.43,44
For example, host odor and the odors of excretory products have been found to be highly attractive to
tsetse flies (Glossina spp.).45–47 Takken14 reviewed the active odors influencing host location by
mosquitoes. Whether for mosquitoes or tsetse, identification of the components of such odors has
led to the development of effective baits for sampling or controlling populations.48 The chemical,
1-octen-3-ol, isolated from cattle (ox breath), has also been found to be a potent olfactory stimulant and
attractant for tsetse and mosquitoes.49–51
   Krijgman,52 conducting experiments with the stable fly, Stomoxys calcitrans, reported orientation in a
simple olfactometer to the odor of fresh blood.53 Tests of various components and fractions of blood as
attractants resulted in the discovery of an extremely volatile attractant constituent for Culex mosquitoes
and Stomoxys. It is believed that this volatile fraction of blood diffuses through the skin of the host and is
an important factor in attracting mosquitoes and biting flies to the host.54 McKenzie16 demonstrated that
materials collected from human skin were attractive to host-seeking mosquitoes.
   Among the olfactory stimuli implicated in host location to date are carbon dioxide, lactic acid, acetone,
butanone, octenol, phenolic components of urine, and oils on the skin.14,16,28,29,55,56 Similar kinds of
materials are probable cues utilized for finding oviposition sites by house flies, stable flies, and horn flies
and include products of decaying plant and animal matter.


q 2006 by Taylor & Francis Group, LLC
Use of Olfactometers for Determining Attractants and Repellents                                                  163




Olfactometer Test Systems
Searching for semiochemicals that are detectable by blood-feeding flies and mosquitoes requires “asking
the insect to choose its preference” to identify their detection of odors and components that can be
isolated from the environment. This entails presenting choices to insects within test systems that isolate
identified cues.24,25


                                              Exhaust fan




                   Side ring                                                    Aqar based
                                                                               artificial hosts




                     Cover

                                                                                         Electric
                                                                                       power source




                                                                                                     (+)
                                                                                                     (−)
  (+)                                                                                                      (+)
                                                                                                           (−)
  (−)

                                                                                                  Base plate
      (+)                                                                                              (+)
                                                                                                       (−)
      (−)


        Electric power
           supply for                                                                        Multichannel
         light source                                                                        A-C converter



                                                                              Electically
                       Fiber optic                                              biased
                      light source                                            differential
                                                  Mixing                       amplifier
                                                  station


                         Air supply                                           Treatment
(a)                       source                                                agent

FIGURE 9.1 (a) Laboratory olfactometer top-view patent drawing. (b) Laboratory olfactometer side-view drawing
(U.S. Patent 4,759,228).


q 2006 by Taylor & Francis Group, LLC
164                                                           Insect Repellents: Principles, Methods, and Uses




                                                                          Exhaust fan




                                                                                                 Aqar based
                                                                                                artificial hosts




                                                                                                      Sensor

                                                                                                            (+)
                                                                                                            (−)




                                   Light
                                                              Electric
                                  source                                                   Program
                                                             power for
                                                           light source                 digital storage




                                           Air supply
                                            source                                       Multichannel
                                                                                        A−C Convertor




                      Electic bias
                    for pseudo host                                                         Printer




                                            Differential                                  Computer
                                             amplifier

   (b)

FIGURE 9.1 Continued.




q 2006 by Taylor & Francis Group, LLC
Use of Olfactometers for Determining Attractants and Repellents                                          165


   Olfactometer test systems, effectively developed by Dethier, use the insect’s behavioral choices to
place numbers on preference activities.3 Optimum test systems present common sources of temperature,
relative humidity, light, air quality, air flow, and when possible include the use of insect behavioral
geotaxis to aid in triggering choices. The test cells are adjusted to allow free behavioral choices for the
test species. Initial test chambers were observable, simple activity chambers with air flows, in which
mosquitoes and other insects were allowed to move onto exposed, treated surfaces or host arms.3,24,57–59
A U.S. Department of Agriculture (USDA) olfactometer at the Mosquito and Vector Research
Laboratory, Gainesville, Florida, U.S.A., based on the Dethier3 model, has been in use as a screening
device for many years.56,60
   A two-choice “Y” tube olfactometer has been used to identify the pheromonal activity for horn and
stable flies. It offers insects two choices.61–63 This system has been further developed into our current test
system with choices presented like wedges in a pie (hereafter: pie-type) for evaluating the response of
arthropods to multiple odors or choices on treated artificial hosts.24,25




Research Objectives
Our research objectives were to develop an arthropod olfactometer for rapid screening of semiochem-
icals against a wide range of insects, mites, and ticks to evaluate the responses of arthopods to certain
attractants, repellents, feeding stimuli, and oviposition stimuli. The system was adapted to measure
arthropod preference by mapping arthropod movement in relation to a treated air stream, a treated
surface, or a suitable food or oviposition source. The insect was then used to select potential compounds
with strong semiochemical activity.




Laboratory Olfactometer
A multiport pie-type olfactometer was developed to electronically quantify insect feeding activity on ten
compounds simultaneously for a set time period.25,64–67 This olfactometer was developed to rapidly
screen and compare large numbers of materials through a supporting grant from International Flavors and
Fragrances, Inc. (IFF, Union Beach, New Jersey, U.S.A.). The multiport system was made possible by
recent technological advancement in electronics and computers that are capable of both regulating test
conditions and data-logging the results.24,25,68
   The olfactometer integrated both computer and electronic detection to measure insect position and the
act of feeding within a pie-type choice chamber. This design allowed the olfactometer to present testable
materials and measure arthropod response to multiple chemical stimuli, with electronic monitoring of
insect choice contact for eight hours or more. Comparisons were made by offering individual insects up
to ten different choices presented simultaneously on ten artificial hosts or air streams (Figure 9.1a, b and
Figure 9.2).
   The artificial hosts were made up of an agar base containing normal saline, feeding attractants, and
cow blood covered by a silicone membrane.16,25,69,70 This artificial host was attractive to blood-feeding
insects, but they were unable to complete feeding due to the gel media. They continued to attempt
feeding for more than eight hours.16 The olfactometer created distinct and contiguous odor fields that
could be easily entered, left, and reentered by the arthropods seeking a source. An electrical signal
generated by the insect contacted with the gel medium was differentially amplified up to 10,000 times
and data logged to a computer file for analysis. The feeding contact signal amplification with the
differential amplifiers required the olfactometer to be housed in a temperature controlled, light-proof,
Faraday-cage room (Lindgren Enclosures, Model No. 18-3/5-1, Glendale Heights, Illinois, U.S.A.) to


q 2006 by Taylor & Francis Group, LLC
166                                                        Insect Repellents: Principles, Methods, and Uses




FIGURE 9.2 University of Florida multiport olfactometer in Faraday cage: paired T configuration (top), open 10-port
configuration (bottom).




q 2006 by Taylor & Francis Group, LLC
Use of Olfactometers for Determining Attractants and Repellents                                        167


control extraneous electrical noise and test parameters.16,25,66,69 This was required to maintain an
acceptable signal to noise ratio (Figure 9.1a,b, and Figure 9.2).
   The air supply was augmented with moist CO2 that was introduced into the air-stream by a computer-
controlled valve at a rate of 250 mL/min (4 min on and 6 min off). This was used to expose treatments
and mosquitoes to expected CO2 near the artificial host, in case host CO2 production affected the material
being evaluated. The comparisons were made under identical light, temperature, air flow, and relative
humidity, with exposure for up to 8 h so changes in activity and treatment could be monitored.
   The large data sets were manipulated with computer programs specifically written for processing these
data. The activity was recorded as feeding contact, summed as bite-second totals on 10-min or 1-h
intervals, and data-logged via computer. The test series were randomized and replicated over time for
statistical analysis, with comparisons made to standard attractants and repellents within each trial.
Optimum trials included an untreated standard and both attractants and repellents to statistically separate
choices. The trials were conducted as replicated tests with analysis of variance utilized to evaluate the
overall test significance. The significant tests (P!0.05) were then compared at a fixed time interval for
differences within the trial. Data was summarized by time intervals of 4 or 8 h or as an 8-h time series of
activity in a raster graph to show treatment formulation change over time.
   Standard statistical analysis was designed as randomized multi-choice tests, with ten randomly placed
artificial hosts. Software designed to log data on contact seconds of feeding and biting (Medusa 2.1.2
F&B designed by Nick Hostettler, Gainesville, Florida, U.S.A.) was used to consolidate and analyze the
number of bite-seconds per sample over an 8-h period. Normalized data were compared to the standard
using a one-tailed t test to determine whether there was any significant difference between samples.
ANOVA was used to separate interaction and independent error term to avoid misrepresenting actual
significance. Data were normalized by transformation using the square root of (nC1).
   The trial consisted of 8-h replications as noted in Figure 9.3 through Figure 9.8. The replications
were used in the final statistical analysis if they had no mechanical, electrical, or behavioral
complications (e.g., amplifiers not communicating with the computer, sensors shorting, insects
trapped behind sensors, no activity recorded for the standard untreated host during a replication).




Summary of Olfactometer Development
The laboratory olfactometer trials were conducted on a large number (greater than 3,999) of
semiochemicals supplied by the granting agency, International Flavors and Fragrances, and other
sources. The majority of tests were conducted on Aedes aegypti mosquitoes, house flies (Musca
domestica), and horn flies (Haematobia irritans). The semiochemicals were principally selected extracts
of plants and animals similar to those in Mookherjee et al.24 Seventy-seven patents have been issued on
attractants, repellents and test equipment as a result of this project.
   Table 9.1 lists the number of tests conducted, Table 9.2 the materials patented as attractants, and
Table 9.3 the materials patented as repellents from the system through 1999. This list includes 28
attractants and 89 repellents that were “new to science.” It should be noted that some of these
compounds are dosage dependent. Deet was listed here as an attractant, although it is presented in the
literature as a repellent. Deet is actually considered an inhibitor. It works by inhibiting the mosquito’s
ability to sense L-lactic acid, effectively blocking antennal receptors. However, there are studies that
suggest that at some levels, deet appears to act as an attractant instead of an inhibitor.71 Acting as an
activator, deet increased mosquito and house fly catch rates when added at low rates to attractant bait
traps. At high rates (0.005 g/cm2) for house flies and mosquitoes in the olfactometer, it acted as a
repellent; at low rates (less than 0.0025 g/cm2), it was often seen as an attractant.72–74 Our data
indicated that this and some other materials were variable depending on the concentration, reversing
the insect biological response with either high or low concentrations.


q 2006 by Taylor & Francis Group, LLC
168                                                                                                       Insect Repellents: Principles, Methods, and Uses


                           2500




                           2000
    Average bite sec / h




                           1500




                           1000




                           500




                             0
                                  Std. att.



                                              Deet 31% cream



                                                               Deet 26% spray



                                                                                SSS rep.



                                                                                           SSS bath oil


                                                                                                              skintastic™
                                                                                                              Deet 6.65%


                                                                                                                            Absorbine Jr.®



                                                                                                                                             Water babies®



                                                                                                                                                             Green ban (herbal)



                                                                                                                                                                                  Buz away (herbal)
FIGURE 9.3 Horn fly (Haematobia irritans) feeding response to artificial host skin treatment of 0.005 g with market
sample repellents and conditioners (5 rep, 8-h consolidated-feeding assay).




Summary of Olfactometer Market Sample Tests
The market sample repellents and skin conditioners were obtained from commercial sources. These
market samples (Table 9.4) were evaluated in the multiport olfactometer to determine their comparative
value in protecting the artificial host from arthropod bites. Applied directly to the silicone membrane of
the artificial host or volatilized in an air stream directly above the membrane, these were evaluated by
summing the insect feeding activity for either 4 or 8 h. Data were normalized using the square root of
(nC1). Evaluations were also made as a time series to determine changes in activity over the test period.
Formulations as tested are presented in Table 9.4. These were evaluated against horn flies (Haematobia
irritans), stable flies (Stomoxys calcitrans), and the yellow fever mosquito (Aedes aegypti). Other
mosquitoes were evaluated against herbal repellents compared to 6.5% deet products.




Market Sample Results
The mean comparisons for market sample tests are presented to give a general relationship to the activity
when compared to the untreated hosts (code #3776 std host). The three most effective market sample


q 2006 by Taylor & Francis Group, LLC
Use of Olfactometers for Determining Attractants and Repellents                                                                                            169


                 1200


                 1000


                  800
    Bite sec/h




                  600


                  400


                  200


                    0
                        Std. att.


                                    cream
                                    Deet 31%


                                               spray
                                               Deet 26%


                                                          SSS rep.



                                                                     SSS bath oil



                                                                                    Deet 6.65%


                                                                                                  Absorbine Jr.®



                                                                                                                   Water babies®


                                                                                                                                   (herbal)
                                                                                                                                   Green ban


                                                                                                                                               Alsenite®
                                                                                    skintastic™


FIGURE 9.4 Stable fly (Stomoxys calcitrans) feeding response to artificial host skin treatment of 0.005 g with market
sample repellents and conditioners (6 rep, 8-h consolidated-feeding assay).



repellents in these trials on horn flies were #3902 UltraThone (31.5% deet), #3903 Cutterw (21% deet),
and #3907 Absorbine Jr.w (Figure 9.3). The four most effective repellents on stable flies were #3905
Skin-So-Softw Bath Oil, #3909 Green Banw (10% citronella, 2% peppermint, and other plant extracts),
#3902 Ultrathon, and #3907 Absorbine Jr. (Figure 9.4).* The three most effective repellents on Aedes
aegypti were #3902 Ultrathon, #3909 Green Ban, and #3905 Skin-So-Soft Bath Oil (Figure 9.5).
Additional trials were conducted comparing the present market standard OFF! Skintastice
Insect Repellent (6.65% deet) to the herbal repellents MosquitoSafee and TickSafee (30% geraniol).
In all trials when geraniol based herbal repellents were tested with various arthropod species, geraniol
based repellents had equal or significantly better repellent activity as skin treatments (Figure 9.6 through
Figure 9.8).




Summary of Olfactometer Research
A laboratory olfactometer developed at the University of Florida is a system that electronically
monitors treated artificial hosts or air-streams, with the choices arranged like wedges in a pie. It can
be partitioned as a one, two, five or ten-choice system so comparative samples can be evaluated.
The olfactometer can also be configured to evaluate dosage levels of the same compounds so that
threshold activities can be evaluated for materials that change from attractants at low concentrations

*
  Ultrathon is a registered trademark of 3M Corp., Minneapolis, MN; Cutter is a registered trademark of United Industries
Corp., St. Louis, MO; Absorbine Jr. is a registered trademark of W.F. Young, Inc., East Longmeadow, MA; Skin-So-Soft is a
registered trademark of Avon Products, Inc., New York, NY; Green Ban is a registered trademark of Mulgum Hollow Farm,
Brookvale, New South Wales, Australia; OFF! Skintastic is a registered trademark of S.C. Johnson and Son, Inc., Racine, WI.


q 2006 by Taylor & Francis Group, LLC
                                                                                                                                                                                                                                                                                                                                            FIGURE 9.6 The mosquito (Aedes aegypti) mean 8-h exposure as bite second response to treated artificial host skin (S) at
Insect Repellents: Principles, Methods, and Uses




                                                                                                                 FIGURE 9.5 Aedes aegypti air-skin comparisons as bite second counts on treated artificial host skin (S) or air (A) at 0.005 g




                                                                                       Bull frog body gel                                                                                                                                                                                  3970 Plastic N,N-dethyl-m-touluamide 9.5%
                                                                                       Green ban (herbal)                                                                                                                                                                                  3969 Plastic geraniol 77%/1449-23% 30% loading
                                                                                       Water babies                                                                                                                                                                                        3967new TickSafe perfume formulation
                                                                                       Absorbine Jr.                                                                                                                                                                                       3966 TickSafe
                                                                                       Deet 6.65% (skintastic)
                                                                                                                                                                                                                                                                                           3965 Base without active repellents




                                                                                                                                                                                                                                                                                                                                            0.002 g or air (A) at 0.005 g; results as average sqrt(S) of bite sec/h.
                                                                                       SSS bath oil                                                                                                                                                                                        3964 Base with deet
                                                          Skin




                                                                                       SSS repellent
                                                                                                                                                                                                                                                                                           3962 Geraniol couer 18%
                                                          Air




                                                                                       Deet 26% (spray)
                                                                                                                                                                                                                                                                                           3928A W59576-139028 MUSK 781




                                                                                                                                                                                                                                                                                                                                                                                                                                                                      q 2006 by Taylor & Francis Group, LLC
                                                                                       Deet 31% (cream)                                                                                                                                                                                    3906 Off skintastic™ Deet (N,N-dethyl-m-
                                                                                                                                                                                                                                                                                           touluamide 6.65% Other Isomers .35%)
                                                                                       Std att
                                                                                                                                                                                                                                                                                           3776 Std host




                                                                                                                 (8-h exposure).
                                                   250

                                                         200

                                                                  150

                                                                        100

                                                                              50

                                                                                   0




                                                                                                                                                                                                                                                60



                                                                                                                                                                                                                                                     50



                                                                                                                                                                                                                                                            40



                                                                                                                                                                                                                                                                   30



                                                                                                                                                                                                                                                                          20



                                                                                                                                                                                                                                                                                  10



                                                                                                                                                                                                                                                                                       0
                                                                 Bite sec/h                                                                                                                                                                           Average sqrt(Σ) of bite sec/h
170
Use of Olfactometers for Determining Attractants and Repellents                                                                                                                                                                                                                                     171


                                           20
                                           18
      Average sqrt(Σ) of bite sec/h


                                           16
                                           14
                                           12
                                           10
                                            8
                                            6
                                            4
                                            2
                                            0
                                                 3776 Std host



                                                                      3905S SSSB oil



                                                                                        3906A Skintastic®



                                                                                                              3906S Skintastic®



                                                                                                                                  3962A RepelIo geraniol



                                                                                                                                                           3962S RepelIo geraniol



                                                                                                                                                                                    3964A Repello base with



                                                                                                                                                                                                              3964S Repello base with


                                                                                                                                                                                                                                            3965A Repello base only



                                                                                                                                                                                                                                                                          3965S Repello base only
                                                                                                                                         couer



                                                                                                                                                                  couer



                                                                                                                                                                                            deet



                                                                                                                                                                                                                      deet
FIGURE 9.7 The mosquito (Anopheles quadrimaculatus) mean 8-h exposure as bite second response to treated artificial
hosts: skin (S) at 0.002 g, air (A) at 0.005 g; results as average sqrt(S) of bite sec/h.




                                           120


                                           100
           Average sqrt(Σ) of bite sec/h




                                           80


                                           60


                                           40


                                           20


                                            0
                                                     3776 Std host




                                                                     3905S SSSB


                                                                                       Skintastic®



                                                                                                            Skintastic®




                                                                                                                                        geraniol coeur
                                                                                                                                        3962A Repelo


                                                                                                                                                                geraniol coeur
                                                                                                                                                                3962S Repelo


                                                                                                                                                                                         base with deet
                                                                                                                                                                                         3964A Repelo


                                                                                                                                                                                                                   base with deet
                                                                                                                                                                                                                   3964S Repelo

                                                                                                                                                                                                                                        base only
                                                                                                                                                                                                                                        3965A Repelo


                                                                                                                                                                                                                                                                      base only
                                                                                                                                                                                                                                                                      3965S Repelo
                                                                                         3906A



                                                                                                              3906S
                                                                         oil




FIGURE 9.8 Culex pipiens 8-h average bite second response as comparative feeding rates: (S) skin at 0.002 g, (A) air
treatment at 0.005 g; results as average sqrt(S) of bite sec/h.



q 2006 by Taylor & Francis Group, LLC
172                                                            Insect Repellents: Principles, Methods, and Uses


                 TABLE 9.1
                 Insects Studied and Number of Semiochemicals Evaluated in the Patented
                 Laboratory Multiport Olfactometer
                                              Number of               Number of             Number of
                 Insect                         Tests                 Attractants           Repellents

                 Mosquito                        5,626                    189                   688
                 House Fly                       1,400                    173                   423
                 Horn Fly                          568                     28                    89
                 # Patents Issued                   77                     28                    89
                 # Compounds                                               32                   120

                 Source: From B. D. Mookherjee et al., Bioactive Volatile Compounds from Plants; 203rd
                 National Meeting of the American Chemical Society, R. Teransishi, R. G. Buttery, and
                 H. Sugisawa (Eds.), ACS Symposium Series, Vol. 525, Washington, DC: American Chemical
                 Society, 1993, p. 35; J. F. Butler and J. S. Okine, Nuisance Concerns in Animal Manure
                 Management: Odors and Flies, Proceedings of a Workshop, H. H. Van Horn (Ed.), Vol. 117,
                 Gainesville, FL: University of Florida and Georgia Agriculture Cooperative Extension
                 Station, 1995. p. 1.




to repellents at high concentrations. The components of repellent formulations were evaluated with
the system to determine their compatibility and loss rates over eight-hours time series evaluations.
The system can also be configured to evaluate insect response to light wave (color) and frequencies
of light (Hz). Arthropod activity is detected by touch or feeding contact sensors using differential
amplifiers that increase the signal by 10,000 times so that contact seconds of activity can be data
logged to a computer for analysis. The assays of first to tenth bite time intervals can also be
determined with selected computer programs. Measurements are obtained as feeding or contact
response in time and summed as contact bite-seconds/hour. Replications are made with treatment
positions randomized within each trial to eliminate position effect and determine interaction that
may occur. Overall, treatment significance is evaluated with comparisons between choices for the
sum of activity for an eight-hours exposure. Individual one-tailed t tests are used to determine
significant differences between two choice treatments.
   The laboratory olfactometer has been used as a rapid screening system of semiochemicals against
a wide range of arthropods to evaluate their potential activity as attractants, repellents, and feeding
stimuli. “New to science” repellents and attractants were identified. The semiochemicals with
attractant and repellent activities were identified as new to science with 77 patents issued to date
covering a total of 139 compounds. Presented in this chapter are the new-to-science attractants (33)
and repellents (87) that have been identified and patented at the University of Florida using this test
system. The standard market sample repellents were compared demonstrating the effectiveness of
high rate deet products compared to low rate deet standards. When herbal repellents (geraniol) were
compared to the lower rate deet products, several formulations demonstrated repellent activity equal
to or better than the deet products. Research is underway to evaluate factors of human attractant
and repellent activities. The results have demonstrated individuals with repellent and attractant
characteristics.16 The laboratory olfactometer was adapted to measure fly, mosquito, cockroach, tick,
and flea responses to treated hosts or air streams. The tested mosquitoes and fly species evaluated in
the olfactometer include mosquitoes (Culex spp., Anopheles quadrimaculatus, Aedes aegypti), the
horn fly (Haematobia irritans), the house fly (Musca domestica), and the stable fly (Stomoxys
calcitrans).


q 2006 by Taylor & Francis Group, LLC
Use of Olfactometers for Determining Attractants and Repellents                                         173


TABLE 9.2
Attractants Identified and Patented from the Olfaction Research Program at the University of Florida,
Including 32 Attractant Compounds Cited in U.S. and Foreign Patents
1-(2-Butenoyl)-2,6,6-trimethyl-1-1-3-cycohexadiene
                                                                                    O




2,3-Dimethyl-3-hexanol



                                                                      HO



2-Methyl-3-pentenoic acid
                                                                           O


                                                                           O



3-Methyl-3-buten-1-ol
                                                                               OH




3-Ethyl-3-hexanol

                                                                      HO




3-Ethyl-2-methyl-3-pentanol
                                                                 HO




9-Decen-1-ol having the structure                                                                      OH




10-Undecen-1-ol having the structure                                                                        OH


                                                                           O

Alpha-damascone




                                                                                                (continued)

q 2006 by Taylor & Francis Group, LLC
174                                     Insect Repellents: Principles, Methods, and Uses


TABLE 9.2 (continued)


Alpha-terpineol



                                                  OH




Aryl moiety compound (a)                                 OH




                                                              O




Aryl moiety compound (b)
                                                                  O

                                                                      O




                                                                  O
Benzyl formate
                                                              O       H




                                                         O
Beta-damascone




d-Carvone                                               O




                                              O

                                                   O

Dibutyl succinate
                                                   O


                                              O
                                                                              (continued)




q 2006 by Taylor & Francis Group, LLC
Use of Olfactometers for Determining Attractants and Repellents                                    175


TABLE 9.2 (continued)




d-Limonene structure




                                                      CH3

Dimethyl disulfide                                      S          CH3
                                                            S



                                                                                     O


                                                                                 O       O
Dimethyl substituted oxymethyl cyclohexene (1a)




d-Pulegone


                                                                   O




Ethyl ester of 2-methyl-3-pentenoic acid
                                                                            O


                                                                            O




                                                                        O
Isobutyric acid
                                                                                OH




Jasmine absolute; racemic borneol

                                                                        OH

                                                                                             (continued)




q 2006 by Taylor & Francis Group, LLC
176                                     Insect Repellents: Principles, Methods, and Uses


TABLE 9.2 (continued)

Marigold absolute


                                              O                    OH




                                              OH                   OH          O




                                         OH                    O          OH



                                                   O
Methyl-isoeugenol
                                                           O




                                                                        OH
n-Dodecanol




                                              O        N


N,N-Diethyl-m-toluamide




                                                           O


trans,trans-Delta-damascone




q 2006 by Taylor & Francis Group, LLC
Use of Olfactometers for Determining Attractants and Repellents                                            177


TABLE 9.3
Repellents Identified and Patented from the Olfaction Research Program at the University of Florida,
Including 89 Repellents Compounds Cited in U.S. and Foreign Patents
1-Octen-4-ol


                                                                             OH




2,4-Dimethyl-4-phenyl-1-                                                                              OH
   butanol




2-Norbornylidene-ethanol-1                                                                   OH




                                                                                  C    N
6-Octenenitrile




                                                                     [Z]
                                                                             R2         R5
Acyclic and carbocyclic
  ketones, alcohols, alde-
                                                                                             OR1
  hydes, nitriles and esters
  and uses thereof (1)
                                                                      R6               R7
                                                                                  R8

                                                                                                     O


Acyclic and carbocyclic
  ketones, alcohols, alde-                                                                     O
  hydes, nitriles and esters
  and uses thereof (2.1)


Acyclic and carbocyclic
  ketones, alcohols, alde-                                                                     OH
  hydes, nitriles and esters
  and uses thereof (2.2)

                                                                                                    (continued)




q 2006 by Taylor & Francis Group, LLC
178                                     Insect Repellents: Principles, Methods, and Uses


TABLE 9.3 (continued)

Acyclic and carbocyclic                                                      OH
  ketones, alcohols, alde-
  hydes, nitriles and esters
  and uses thereof (2.3)


Acyclic and carbocyclic
  ketones, alcohols, alde-
  hydes, nitriles and esters                                                 OH
  and uses thereof (2.4)




Acyclic and carbocyclic                                        OH
  ketones, alcohols, alde-
  hydes, nitriles and esters
  and uses thereof (2.5)


                                                                                   O


Acyclic and carbocyclic
  ketones, alcohols, alde-                                                O            O
  hydes, nitriles and esters
  and uses thereof (2.6)




Acyclic and carbocyclic
  ketones, alcohols, alde-
  hydes, nitriles and esters
  and uses thereof (3)                                                   O
                                                                H




Alkyl cyclopentanone and
  phenyl alkanol derivative-
  containing composition (a)
                                                         O


                                                                                  OH




Alkyl cyclopentanone and
  phenyl alkanol derivative-
  containing composition (a)



                                                         O

                                                                                  (continued)




q 2006 by Taylor & Francis Group, LLC
Use of Olfactometers for Determining Attractants and Repellents                   179


TABLE 9.3 (continued)


Alkyl cyclopentanone,
  cycloalkanal and phenyl
                                                                  OH
  alkanol derivative-
  containing (b)




Alkyl cyclopentanone,
  cycloalkanal
                                                                  HO
  and phenyl alkanol deriva-
  tive-containing (c)




Alkyl cyclopentanone,
  cycloalkanal and phenyl
  alkanol derivative-
                                                                  HO
  containing (d)


Alkyl cyclopentanone,
  cycloalkanal and phenyl
  alkanol derivative-
                                                                  HO
  containing (e)


                                                                           OH

Bisabolene isomer (a)




Bisabolene isomer (b)




Bisabolene isomer (c)                                                  E




                                                                            (continued)



q 2006 by Taylor & Francis Group, LLC
180                                     Insect Repellents: Principles, Methods, and Uses


TABLE 9.3 (continued)


Bisabolene isomer (d)




Bisabolene isomer (e)




Bisabolene isomer (f)




Bisabolene isomer (g)




                                                        OH
C12 branched alcohol




C12 unsaturated ketone                                   O

  mixture




Carbocyclic compounds

                                                                       O
                                                                 O


                                                                              (continued)




q 2006 by Taylor & Francis Group, LLC
Use of Olfactometers for Determining Attractants and Repellents                                             181


TABLE 9.3 (continued)

                                                                         HO
Camekol dh structure




                                                                          O        O
Carbonate esters (a)
                                                                              O



                                                                                   O


Carbonate esters (b)                                                          O         O




                                                                         H
                                                                         C1
Carbonate esters (c)                                                                        CH3
                                                                  H2C6             2C
                                                                                            CH3
                                                                         CH2

                                                                  HC5              3
                                                                                       CH
                                                    HO                                       CH3
                                                                         C4
                                                                         H




Citronellol


                                                                                  OH




                                                                                            O

Cyclemonew (a)                                                                                    H




Cyclemonew (b)

                                                                                                  H


                                                                                             O

                                                                                                      (continued)




q 2006 by Taylor & Francis Group, LLC
182                                     Insect Repellents: Principles, Methods, and Uses


TABLE 9.3 (continued)


Cyclemonew (c)




                                               O                      O




Cycloalkanol derivative-                                  HO
  containing composition (a)




Cycloalkanol derivative-
                                                          HO
  containing composition (b)




Cycloalkanol derivative-                                  HO
  containing composition (c)




Cycloalkanol derivative-
  containing composition (a)




Cycloalkanol derivative-
  containing composition (b)




Cycloalkanol derivative-
  containing composition (c)




Cycloalkanol derivative-
  containing composition (b)
                                                     HO



                                                                              (continued)




q 2006 by Taylor & Francis Group, LLC
Use of Olfactometers for Determining Attractants and Repellents                               183


TABLE 9.3 (continued)


Cycloalkanol derivative-
  containing composition (c)                                      HO




                                                                  and




                                                                  HO




Cycloalkanol derivative-
  containing composition (d)




Cycloalkanol derivative-
  containing composition
  (e2)
                                                  HO                         HO




                                                             HO




Cycloalkanol derivative-
  containing composition (b)                                            HO




                                                                        HO
Cycloalkanol derivative-
  containing composition (c)
                                                                                  and




                                                                        HO




Cycloalkanol derivative-
                                                                                            OH
  containing composition (d)



                                                                                        (continued)




q 2006 by Taylor & Francis Group, LLC
184                                     Insect Repellents: Principles, Methods, and Uses


TABLE 9.3 (continued)


                                                          HO
Cycloalkanol derivative-
  containing composition (e)



                                                          HO




                                                          HO




Dihydrofloralol                                                                  OH




                                                                            OH
Dihydrofloralol




                                                                R2

Dimethyl substituted
  oxymethyl cyclohexane                                                     OH
  derivative (1a)




                                                                            O

Dimethyl substituted
  oxymethyl cyclohexane                                                O
  derivative




Geraldehyde



                                                                        H       O




                                                                                 (continued)




q 2006 by Taylor & Francis Group, LLC
Use of Olfactometers for Determining Attractants and Repellents                                             185


TABLE 9.3 (continued)

Geraniol precursor


                                                                            O           O



                                                                                    O




Geraniol

                                                                                         H        O




Geraniol Coeur: nerol

                                                                                        OH




Geraniol Coeur: citronellol

                                                                                        OH




Geraniol Coeur: geraniol                                                                     OH




                                                  O
                                                                                    O
Hedione
                                             O                    and           O



                                        O                               O

                                                                                                      (continued)




q 2006 by Taylor & Francis Group, LLC
186                                     Insect Repellents: Principles, Methods, and Uses


TABLE 9.3 (continued)
                                                                                 O

Isocyclogeraniol (a)
                                                                          O          O




Isocyclogeraniol (b)
                                                                          OH




                                                                      O
Karismal
                                                              O



                                                         O

                                                                  O


                                                              O                and



                                                         O

                                                                      O


                                                              O



                                                         O



Ketones
                                                             OH




                                                     O
Ketones, aldehydes, and
  esters (a)
                                                     H




Ketones, aldehydes, and
  esters (b)
                                                     O

                                                         H


                                                                                (continued)




q 2006 by Taylor & Francis Group, LLC
Use of Olfactometers for Determining Attractants and Repellents                                         187


TABLE 9.3 (continued)

                                                                      O
Ketones, aldehydes, and                                           H
  esters (c)




                                                                              OH

Ketones, aldehydes, and
  esters (d)




Ketone and Schiff base-
  containing compositions                                  OH
  (a)                                                                         N



                                                                                   O          O




Ketone and Schiff base-
  containing compositions
                                                          OH
  (b)


                             H
                                  N
                                            OCH3                          C   N

                                        O

                            O                                                 O

                                                                                       OCH3



Kovone                                         O


                                                                                   O
                                                                                        and
                                                      O




Kovone                                                                    O




                                                                                                  (continued)




q 2006 by Taylor & Francis Group, LLC
188                                     Insect Repellents: Principles, Methods, and Uses


TABLE 9.3 (continued)
                                                                                 O

Lavonax




                                                                 OH                      O

Lyral (1)
                                                                                             H




                                                                 OH
Lyral (2)



                                                                                             H; and

                                                                                         O


                                                OH
                                                                     N



                                                                             O       O




                                                                              OCH3
Lyrame                                               H   N
                                                                         O


                                                  O




                                           OH




                                                             C   N



                                                                 O
                                                                             OCH3



                                                                                         (continued)




q 2006 by Taylor & Francis Group, LLC
Use of Olfactometers for Determining Attractants and Repellents                                         189


TABLE 9.3 (continued)


                                                                             O
Melozone (a)
                                                                     H




Melozone (b)
                                                                     O

                                                                             H




                                                                                 O
Melozone (c)                                                         H




Melozone, wherein 60–40                                                          O
 mole percent is the               O                                     H
 compounds                                     O
                               H

                                                   H




                                                                                              O
                                                               O

Methyl jasmonate                                           O       and                    O



                                                       O                             O




Nerol


                                                                                         OH




                                                                                                  (continued)




q 2006 by Taylor & Francis Group, LLC
190                                     Insect Repellents: Principles, Methods, and Uses


TABLE 9.3 (continued)


Orange flower ether




                                                                 OCH3



Organo-boron derivative


                                                             O           O
                                                                     B

                                                                     O




Schiff base of ethyl vanillin
                                                                                 CH3
  and methyl anthranilate

                                                                             N


                                                                         O


Substituted silane,
  digeranyloxy-dimethyl-                                O   Si   O
  silane structure




q 2006 by Taylor & Francis Group, LLC
Use of Olfactometers for Determining Attractants and Repellents                                                  191


TABLE 9.4
Market Sample Repellents, Skin Conditioners, and Herbal Based Repellents used in Olfactometer Trials
Sample                                  Name                                          Manufacturer

3776                Standard Artificial Host Attractant                N/A
3902                UltrathonTM 3M EPA 58007-1 31.5% Deet             3M St. Paul, MN 55144-1000
3903                Cutter 21% Deet EPA 121–129                       Miles Inc, Consumer Household Products Div.
                                                                        7123 W. 65th St., Chicago, IL 60638
3904                Skin So Soft Mosquito Flea and Deer Tick          Avon Products, Inc. New York, N.Y. 10019
                      Repellent EPA 65233.1.806
3905                Skin-So-Soft Bath Oil Spray                       Avon Products, Inc. New York, N.Y. 10019
3906                Off! SkintasticTM Insect Repellent                S.C. Johnson & Son, Inc Racine, WI 53403
3907                Absorbine Jr.w Antiseptic Liniment                W.F. Young, Inc. Springfield, MA 01103
3908                Water Babiesw SPF 30                              Schering Plough Healthcare Products, Inc
                                                                        Memphis, TN 38151
3909                Green Ban Citronella With Calendula and Cajuput   Mulgum Hollow Farm P.O. Box 225, Brookvale
                                                                        NSW 2100 Australia
3910                Bull Frogw Sunblock SPF 36                        Chattem, Inc. Chattanooga, TN 37409
3911                Quantum Buzz Away                                 Quantum, Inc. P.O. Box 2791 Eugene, OR 97402
3912                TickSafeTM                                        Naturale LTD (now Fasst Products Inc., Rockville
                                                                        Center, NY 11570) 9 Park Place, Great Neck,
                                                                        NY 11021
3913                MosquitoSafeTM                                    Naturale LTD (now Fasst Products Inc., Rockville
                                                                        Center, NY 11570) 9 Park Place, Great Neck,
                                                                        NY 11021
3923                Alsenitew                                         Imperial Builders Apopka, FL
3928                Experimental Attractant                           N/A
3962                Experimental Repellent                            N/A
3964                Repello Base with Deet                            Naturale LTD (now Fasst Products Inc., Rockville
                                                                        Center, NY 11570) 9 Park Place, Great Neck,
                                                                        NY 11021
3965                Experimental Base                                 N/A
3966                Experimental Repellent                            N/A
3967                Experimental Repellent                            N/A
3969                Geraniol Wristband                                Naturale LTD (now Fasst Products Inc., Rockville
                                                                        Center, NY 11570) 9 Park Place, Great Neck,
                                                                        NY 11021
3970                Experimental Deet Wristband                       Naturale LTD (now Fasst Products Inc., Rockville
                                                                        Center, NY 11570) 9 Park Place, Great Neck,
                                                                        NY 11021




References
       1. V. G. Dethier, Chemoreception, in Insect Physiology, D. Kenneth Roeder (Ed.), New York: Wiley,
          1953, p. 544.
       2. V. G. Dethier, The sensory physiology of blood-sucking arthropods, Exp. Parasitol., 6, 68, 1957.
       3. V. G. Dethier, The Physiology of Insect Senses, New York: Wiley, 1963, p. 266.
       4. V. G. Dethier, The Hungry Fly: A Physiological Study of the Behavior Associated with Feeding,
          Cambridge: Harvard University Press, 1976.
       5. E. Stadler, Contact chemoreseption, in Chemical Ecology of Insects, W. J. Bell and T. Carde (Eds.),
          Sunderland, MA: Sinauer Associates, 1984.
       6. H. Mustaparta, Olfaction, in Chemical Ecology of Insects, W. J. Bell and T. Carde (Eds.), Sunderland,
          MA: Sinauer Associates, 1984, pp. 35–70.


q 2006 by Taylor & Francis Group, LLC
192                                                         Insect Repellents: Principles, Methods, and Uses


     7. N. E. Kiwia, Electrophysiological brain activity recorded from the house fly, Musca domestica (L.)
        (Diptera: Muscidae) stimulated by attractant and repellent semiochemicals, MS thesis, University of
        Florida, 1989, p. 126.
     8. A. J. Grant et al., Electrophysiological responses of receptor neurons in mosquito maxillary palp
        sensilla to carbon, J. Comp. Physiol. A., 177, 389, 1995.
     9. R. F. Chapman, Some experiments to determine the methods used in host finding in tsetse flies,
        Glossina medicorum, Bull. Entomol. Res., 52, 83, 1961.
    10. B. Hocking, Blood sucking behavior of terrestrial arthropods, Ann. Rev. Entomol., 16, 1, 1971.
    11. W. G. Friend and J. J. B. Smith, Factors affecting feeding by blood-sucking insects, Ann. Rev.
        Entomol., 22, 309, 1977.
    12. D. E. Kline, J. R. Wood, and J. A. Cornell, Evaluation of 1-octen-3-ol as an attractant for Coquillettidia
        perturbans, Mansonia spp. and Culex spp. Associated with phosphate mining operations, J. Am. Mosq.
        Control Assoc., 6, 605, 1990.
    13. M. Jacobson, Glossary of Plant Derived Insect Deterrents, Boca Raton, FL: CRC Press, 1990,
        pp. 7–8.
    14. W. Takken, The role of olfaction in host-seeking of mosquitoes, Insect Sci. Applic., 12, 287, 1991.
    15. D. A. Burkett, Light color attraction and dietary sugar composition for several mosquito (Diptera:
        Culicidae) species found in north central Florida, Ph.D. diss., University of Florida, 139, 1998.
    16. K. E. McKenzie, Determining factors of preferential host selection by Aedes aegypti (Diptera:
        Culicidae), Ph.D diss., University of Florida, 148, 2003.
    17. J. J. Laarman, The host seeking behavior of the malaria mosquito, Anopheles maculipennis atroparvus,
        Acta Leinden., 25, 1, 1975.
    18. A. W. A. Brown, The attraction of mosquitoes to hosts, J. Am. Med. Assoc., 196, 249, 1966.
    19. M. T. Gillies and T. J. Wilkes, A comparison of the range of attraction of animal baits and of carbon
        dioxide for some West African mosquitoes, Bull. Entomol. Res., 59, 441, 1969.
    20. G. A. Vale, The flight of tsetse flies (Diptera: Glossinidae) to and from a stationary ox, Bull. Entomol.
        Res., 67, 297, 1977.
    21. L. W. Dalton et al., Host location in the horn fly: Roles of heat, CO2, water vapor, cow produced odors in
        attraction, Southwest Entomol., 3, 147, 1978.
    22. H. G. Kinzer, J. M. Reeves, and J. W. Atmar, Host location by the horn fly: Field evaluation of an
        artificial devise for measuring attraction to various stimuli, Environ. Entomol., 7, 375, 1978.
    23. J. R. Agee and R. S. Patterson, Spectral sensitivity of stable, face, and horn flies and behavioral responses
        of stable flies to visual traps (Diptera, Muscidae), Environ. Entomol., 12, 1823, 1983.
    24. B. D. Mookherjee et al., Semio activity of flavor and fragrance molecules on various insect species, in
        Bioactive Volatile Compounds from Plants; 203rd National Meeting of the American Chemical Society,
        R. Teransishi, R. G. Buttery, and H. Sugisawa (Eds.), ACS Symposium Series, Vol. 525, Washington,
        DC: American Chemical Society, 1993, p. 35.
    25. J. F. Butler and J. S. Okine, Development of attractants and repellents for future management of pest fly
        populations, in Nuisance Concerns in Animal Manure Management: Odors and Flies, Proceedings of a
        Workshop, H. H. Van Horn (Ed.), Vol. 117, Gainesville, FL: University of Florida and Georgia
        Agriculture Cooperative Extension Station, 1995, p. 1.
    26. M. J. Lehane, Biology of Blood-Sucking Insects, London: Harper Collins Academic, 1991.
    27. W. Takken and D. L. Kline, Carbon dioxide and 1-octen-3-ol as mosquito attractants, J. Am. Mosq.
        Control Assoc., 5, 311, 1989.
    28. J. F. Sutcliffe, Distance orientation of biting flies to their hosts, Insect Sci. Appl., 8, 611, 1887.
    29. L. P. M. Willemse and W. Takken, Odor-induced location in Tsetse flies (Diptera: Glossinidae), J. Med.
        Entomol., 31(6), 775, 1994.
    30. R. Garcia, Carbon dioxide as an attractant for certain ticks (Acarina: Argasidae and Ixodidae), Ann.
        Entomol. Soc. Am., 55, 605, 1962.
    31. R. Garcia, Collection of Dermacentor andersonii (Stiles) with carbon dioxide and its application in
        studies of Colorado tick fever virus, Ann. J. Trop. Med. Hyg., 14, 1090, 1965.
    32. H. C. M. Parr, Studies of Stomoxys calcitrans (L.) in Uganda, East Africa. 2. Notes on life history and
        behavior, Bull. Entomol. Res., 53, 437, 1962.


q 2006 by Taylor & Francis Group, LLC
Use of Olfactometers for Determining Attractants and Repellents                                                193


    33. J. Brady, The visual responsiveness of the tsetse fly Glossina morsitans Westw, (Glossinidae) to moving
        objects: The effects of hunger, sex, host odor and stimulus characteristics, Bull. Entomol. Res., 62, 257,
        1972.
    34. G. A. Vale, New field methods for studying the response of tsetse flies (Diptera: Glossinidae) to hosts,
        Bull. Entomol. Res., 64, 199, 1974.
    35. L. T. Hargett, Visual and olfactory responses of the horn fly, Haematobia irritans, Ph.D dissertation
        abstr., Oregon State University, 1962.
    36. L. T. Hargett and R. L. Goulding, Rearing the horn fly, Haematobia irritans, J. Econ. Entomol., 55, 565,
        1962.
    37. E. G. Milstrey, Variation in horn fly, Haematobia irritans (L.) hydrocarbons and the effect of alternating
        pheromone levels in the field on wild horn fly population, 289. MS thesis, University of Florida, 1983.
    38. W. G. Bruce, The history and biology of the horn fly, Haematobia irritans (Linnaeus), with comments
        on control, N.C. Exp. Sta. Tech. Bull., 157, 1964.
    39. W. L. Bidlingmayer and D. G. Hem, The range of visual attraction and the effect of competitive
        attractants on mosquito (Diptera; Culicidae) flight, Bull. Entomol. Res., 70, 321, 1980.
    40. W. C. Smith and J. F. Butler, Ultrastructure of the Tabanidae compound eye: Unusual features for
        Diptera, J. Insect Physiol., 37(4), 287, 1991.
    41. A. J. Grant and R. J. O’Connell, Neurophysiological and morphological investigations of pheromone-
        sensitive sensilla on the antenna of male Trichoplusia ni, J. Insect Physiol., 32(6), 503, 1986.
    42. R. Galun, Diversity of Phagostimulants used for recognition of blood meal by haematophagus
        arthropods, in Host-Regulated Development Mechanisms in Vector Arthropods, D. Borovsky and
        A. Spielman (Eds.), Gainesville, FL: IFAS, 1986, p. 1.
    43. R. F. Harwood and M. T. James, Entomology in Human and Animal Health, New York: Macmillan,
        1979, p. 80.
    44. J. Colvin and G. Gibson, Host seeking behavior and management of Tsetse, Annu. Rev. Entomol., 37, 21,
        1992.
    45. J. W. Hargrove and G. A. Vale, The effect of host odor concentrations on catches of tsetse flies
        (Glossinidae) and other Diptera in the field, Bull. Entomol. Res., 68, 607, 1978.
    46. M. L. A. Owaga, Preliminary observations on the efficacy of olfactory attractants derives from wild
        hosts of tsetse, Insect Sc. Appl., 5, 87, 1984.
    47. M. L. A. Owaga, Observations on the efficacy buffalo urine as a potent olfactory attractant for Glossina
        pallidipes Austen, Insect Sc. Appl., 6, 561, 1985.
    48. G. A. Vale, S. Flint, and D. R. Hall, The field responses of tsetse flies, Glossina spp. (Diptera:
        Glossinidae), to odors of host residues, Bull. Entomol. Res., 76, 685, 1986.
    49. D. R. Hall et al., I-Octen-3-ol: A potent olfactory stimulant and attractant for tsetse flies isolated from
        cow odor, Insect Sc. Appl., 5, 335, 1984.
    50. D. E. Kline, Olfactory attractants for mosquito surveillance and control: 1-octen-3-ol, J. Am. Mosq.
        Control Assoc., 10, 280, 1994.
    51. D. E. Kline et al., Field studies on the potential of butanone, carbon dioxide, honey extract, 1-octen-3-ol,
        L-lactic acid and phenols as attractants for mosquitoes, Med. Vet. Entomol., 4, 383, 1990.
    52. B. J. Krijgsman, Reizphysiologische Untersuchungen an blutsaugenden Arthropoden im Zusammen-
        hang mit ihrer Nahrungswahl. I. Teil. Stomoxys calcitrans, Z. vergl. Physiol., 11, 702, 1930.
    53. A. G. Gatehouse and C. T. Lewis, Host location behavior of Stomoxys calcitrans, Entomol. Exp. Appl.,
        16, 275, 1973.
    54. R. Galun, The role of host blood in feeding behavior of Ectoparasites, in Dynamic Aspects of Host-
        Parasite Relationships, A. Zuckerman (Ed.), Vol. 2, New York: Wiley, 1975, p. 132.
    55. G. Nicolas and D. Sulilans, Immediate and latent effects of carbon dioxide on insects, Ann. Rev.
        Entomol., 34, 97, 1989.
    56. U. R. Bernier et al., Chemical analysis of human skin emanations: Comparison of volatiles from
        humans that differ in attraction of Aedes aegypti (Diptera: Culicidae), J. Am. Mosq. Control Assoc.,
        18, 186, 2002.
    57. F. M. Feinsod and A. Spielman, An olfactometer for measuring host-seeking behavior of female
        Aedes aegypti (Diptera: Culicidae), J. Med. Entomol., 15, 282, 1979.


q 2006 by Taylor & Francis Group, LLC
194                                                      Insect Repellents: Principles, Methods, and Uses


    58. T. G. Floore et al., Mosquito trapping studies to determine the efficacy of two models of the
        Flowtron mosquito luring device, J. Fla. Anti-Mosq. Assoc., 56, 13, 1985.
    59. A. E. Eiras and P. C. Jepson, Responses of female Aedes aegypti (Diptera: Culicidae) to host odors
        and convection currents using an olfactometer bioassay, Bull Ent. Res., 84, 207, 1994.
    60. D. A. Carlson and C. D. Grant, Attraction of female mosquitoes (Aedes aegypti and Anopheles
        quadrimaculatus Say) to stored human emanations: Effect of CO2 H2O and temperature adjustments,
        Proceedings and Papers of the Forty-seventh Annual Conference of the California Mosquito and
        Vector Control Association, Inc., Burlingame, CA: California Mosquito and Vector Control
        Association, 1979.
    61. H. T. Bolton, J. F. Butler, and D. A. Carlson, A mating stimulant pheromone of the horn fly,
        Haematobia irritans (L.) demonstration of biological activity in separated cuticular components,
        J. Chem. Ecol., 6, 951, 1981.
    62. S. Muhammed, J. F. Butler, and D. A. Carlson, Stable fly sex attractant and mating pheromones
        found in female body hydrocarbons, J. Chem. Ecol., 1, 387, 1975.
    63. J. W. Mackley, D. A. Carlson, and J. F. Butler, Identification of the cuticular hydrocarbons of the
        horn fly and assay for attraction, J. Chem. Ecol., 7, 669, 1981.
    64. J. F. Butler and I. Katz, Process for determination of repellency and attractancy. U.S. Patent
        4,759,228, 1987.
    65. A. B. Marin, C. B. Warren, and J. F. Butler, Method for repelling Aedes aegypti using 3, 7 dimethyl-
        6- octenenitrile and/or 2(3,3-dimethyl-2-norbornylidene) ethanol-1. U.S. Patent 5,734,892, 1991.
    66. R. A. Wilson, B. D. Mookheijee, and J. F. Butler, Electronic insect repellency and attractancy tester,
        U.S. Patent 5,134,892, 1991.
    67. J. S. Okine, Aspects of oogenesis in the horn fly, Haematobia irritans (Linnaeus) (Diptera:
        Muscidae), Ph.D diss., University of Florida, 30, 1994.
    68. R. J. Symonds and D. M. Unwin, The use of a microcomputer to collect activity data, Physiol.
        Entomol., 7, 91, 1982.
    69. J. F. Butler et al., In vitro feeding of Ornithodoros ticks for rearing and assessment of disease
        transmission, in Acarology VI, D. A. Griffiths and C. E. Bowman (Eds.), Vol. 2, Ellis Horwood:
        Chichester, 1984, p. 1075.
    70. E. L. Davis et al., Laboratory blood-feeding of Culicoides mississippiensis Hoffman through a
        reinforced silicone membrane, J. Med. Entomol., 20, 177, 1983.
    71. E. J. Hoffmann and J. R. Miller, Reduction of mosquito attacks on a human subject by combination
        of wind and vapor-phase DEET repellent, J. Med. Entomol., 39, 935, 2002.
    72. R. A. Wilson, et al., Use of N,N-diethyl-M-toluamide and/or 2-methyl-3-pentenoic acid as insect
        attractants. U.S. Patent 4,876,087, 1989.
    73. R. A. Wilson, et al., Use of N,N-diethyl-M-toluamide and/or 2-methyl-3-pentenoic acid as insect
        attractants. U.S. Patent 4,880,625, 1989.
    74. R. A. Wilson, et al., Use of N,N-diethyl-M-toluamide and/or 2-methyl-3-pentenoic acid as insect
        attractants. U.S. Patent 4,959,209, 1990.




q 2006 by Taylor & Francis Group, LLC
10
Discovery and Design of New Arthropod/Insect
Repellents by Computer-Aided Molecular Modeling


Raj K. Gupta and Apurba K. Bhattacharjee


CONTENTS
Background....................................................................................................................................195
Historical Development of Arthropod Repellents ........................................................................197
Chemical Functional Requirements for Arthropod Repellent Compounds .................................198
  Electronic and Stereoelectronic Considerations .......................................................................199
  Molecular Mechanism of Arthropod Repellent Activity..........................................................202
  Computational Procedure ..........................................................................................................204
  Results and Discussion ..............................................................................................................204
    Conformational Analysis .......................................................................................................204
    Molecular Similarity Analysis of JH-Mimic and Deet Compounds ....................................205
    Correlation of Molecular Orbital Properties in JH-Mimic and in Deet and Its Analogs ....208
    Molecular Electronic Properties of JH..................................................................................209
Development of a New Model for Repellent Research ...............................................................209
  Chemical-Feature Based Considerations ..................................................................................209
  Significance and Uniqueness of the Methodology ...................................................................210
  Computational Methods and Materials .....................................................................................210
    Procedure for Development of the 3D-QSAR Pharmacophore Model ................................210
    Bioassay for Mosquito Repellency .......................................................................................211
  Results and Discussion ..............................................................................................................212
Concluding Remarks and Future Perspectives .............................................................................221
Acknowledgments .........................................................................................................................225
References .....................................................................................................................................225




Background
The goal of this chapter will be to focus on new, next-generation computer techniques of molecular
modeling to illustrate to researchers in the field of arthropod repellents how information on the three-
dimensional structure of small molecules can facilitate the identification, design, and synthesis of
repellents. The emphasis is primarily on understanding the quantitative structure activity relationships


                                                                                                                                             195

q 2006 by Taylor & Francis Group, LLC
196                                                    Insect Repellents: Principles, Methods, and Uses


(QSAR) and mechanisms of action, enabling early planning for structural design, synthesis, and
further development.
   Computer-assisted molecular modeling (CAMM) has been used to make remarkable advances in
mechanistic drug design and in the discovery of new potential bioactive chemical entities in recent
years.1–3 CAMM techniques can provide five major types of information that are crucial for mechanistic
design of drugs and potent new chemical compounds. They are:

      † Three-dimensional structure of a molecule
      † Chemical and physical characteristics of a molecule
      † Comparison of the structure of one molecule with other molecules
      † Graphical visualization of complexes formed between the modeled compound and proteins or
        other molecules
      † Predictions about how related molecules match the modeled ones, along with an estimate
        of potency

   With the advent of modern computers and graphic techniques, computations and visualization of
structures ranging from small to large biomolecules, such as proteins, can be accomplished with greater
speed and precision. The graphic tools in modern computers have made it possible not only to visualize
the three-dimensional structures of large protein molecules, but also to perform interactive, virtual
docking experiments between potential drug molecules and the binding sites of proteins.
   Molecular modeling has now become an inseparable part of research activities that require an
understanding of molecular bases of environmental, biochemical, and biological processes. Compu-
tational methodologies are routinely being used to make decisions about chemical development and also
to perform direct experimental investigations. The current advances in these methodologies allow direct
applications ranging from accurate ab initio quantum chemical calculations of stereoelectronic proper-
ties, generation of three-dimensional pharmacophores, and performance of database searches to identify
potent bioactive agents.
   Discovery of new insect repellent active ingredients is a complex process with ever-changing new
technologies. For example, it still takes about 10 years and, on average, approximately $30 million to
bring a new insect repellent to market. Thus, historically, any technology that can improve the efficiency
of the process is highly valuable to the commercial industry. In silico technologies are relatively new and
have shown remarkable success in recent years, particularly in virtual screening of compound databases.
These technologies are primarily driven by both cost- and time-effectiveness of new active ingredient
discovery. Although no model is perfect, regardless of whatever it represents, the ability to virtually
screen hundreds of compounds in a few hours and to construct simulations of three-dimensional protein
structures in a computer has pushed these technologies to the cutting edge of discovery of new insect
repellent active ingredients.
   The ability of a bioactive molecule to interact with the recognition sites in receptors results from a
combination of steric and electronic properties. Therefore, the study of stereoelectronic properties of
these molecules can provide valuable information, not only to better understand the mechanism of action,
but also to develop a reliable pharmacophore to aid in the design of more efficient analogues. Quantum
chemical computations in modern computers can provide accurate estimates of the stereoelectronic
properties of molecules, and thus can be used to assess interaction of potential repellent active
ingredients with the receptor.
   Developing in silico three-dimensional pharmacophore models and using them selectively as
templates for three-dimensional multi-conformer database searches to identify new potent compounds
are a few of the many other remarkable successes of computational methodologies in recent years.4
A three-dimensional pharmacophore may be perceived as a geometric distribution of chemical features,
such as a hydrogen bond acceptor, hydrogen bond donor, aliphatic and aromatic hydrophobic moieties,
ring aromatic hydrophobicity, etc., in the three-dimensional space that defines the specific biological


q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling                            197


activity of a molecule. Pharmacophores are generated by multiple conformations from a set of
structurally diverse molecules. The generated pharmacophore enables rapid screening of virtual
molecules/libraries to identify potent and non-potent bioactive agents.




Historical Development of Arthropod Repellents
The occupational hazards of the military with respect to exposure to arthropod-borne diseases is in some
ways very representative of the worst hazards presented to the public in any region. Military personnel
generally come from outside the region and, therefore, may not have any more immunity to local
pathogens than a newborn baby. Because of their extensive exposure outdoors during all times of day and
in all kinds of weather, military personnel tend to receive the maximum number of arthropod bites.
Lessons learned from the military experience with arthropod-borne diseases are, therefore, widely
applicable to the public in general.
   Arthropods continue to be important to military operations when they act as vectors of disease.5
Arthropods serve as vectors in a number of different ways, from simple mechanical transmission of
pathogenic organisms on the arthropod body—for instance, when house flies carry dysentery bacilli from
infected feces to food—to the more complicated process of biological transmission, where the pathogens
must spend part of their life cycle in the body of the arthropod before humans can be infected.6–10
Regardless of the specifics of the association, a vector is an organism that transmits a pathogen to a
susceptible host.11,12 Arthropod-borne diseases are extensive both in terms of variety and public health
impact, but few effective economical vaccines are currently available.13–18
   The increase in the U.S. military’s operations will continue to expose its personnel to region-specific
biting arthropods and the vector-borne diseases that they carry. The degree of exposure will largely
depend on environmental factors and operational intensity. Success of high-intensity field operations in
regions of significant arthropod infestations may be associated with, or even depend on, a safe and
effective repellent and service members’ adherence to its proper application. Because concerns have
been raised in recent years regarding the safety of N,N-diethyl-m-toluamide (deet), one of the most
widely used and reliable insect repellents available, the search for an alternative form of deet is also an
important research goal for the U.S. Army.19
   A fundamental activity of military medical entomologists is to establish the role that certain arthropod
species or populations play in the transmission of a particular infectious disease to service members.20
Primary vectors are those that are mainly responsible for transmitting a pathogen to humans or animals;
secondary vectors are those that play a supplementary role in transmission but would be unable to
maintain disease transmission in the absence of the primary vector.6 Mosquitoes are arthropods of special
significance. They cause more human suffering than any other organism, with over two million people
dying of mosquito-borne diseases every year. Not only can mosquitoes carry diseases that afflict humans,
but they also transmit several diseases and parasites to which dogs and horses are very susceptible. These
include dog heartworm, west Nile virus (WNV), and eastern equine encephalitis virus (EEEV). In
addition, mosquito bites can cause severe skin irritation through an allergic reaction to the mosquito’s
saliva. Mosquito-vectored pathogens include the protozoa that cause diseases such as malaria, nematodes
that cause filarial diseases such as dog heartworm, and viruses that cause diseases such as dengue, many
encephalitides, and yellow fever.
   Deet has been regarded as the standard mosquito repellent for the past several decades. However, as a
repellent for human use, deet is not equally effective against all insects and arthropod vectors of
diseases.21–23 Furthermore, in most formulations, it has a short duration of action (no more than several
hours) and several disagreeable cosmetic effects, such as unpleasant odor. Of greater concern is that
when it is used in higher concentrations, the deeper skin penetration can cause potential toxicity. In
addition, deet is a plasticizer that reacts with certain plastics and synthetic rubber.


q 2006 by Taylor & Francis Group, LLC
198                                                     Insect Repellents: Principles, Methods, and Uses


   With increased international travel, illnesses caused by mosquito-borne pathogens, such as malaria,
yellow fever, dengue fever, filariasis, and viral encephalitis, are flaring up all over the globe.19 One
mosquito species that easily adapts to urban conditions, Culex pipiens, caused the epidemic of west Nile
viral encephalitis in New York City in 1999 that has since spread up and down the eastern seaboard,24 as
well as the rest of North America. Insect repellents that are completely safe and more effective than
current products would be important additions to the armamentarium of tools available to prevent
transmission of arthropod-borne pathogens.
   Deet (the structural chemical name was previously N,N-diethyl-m-toluamide, but now designated N,N-
diethyl-3-methylbenzamide) remains the gold standard of currently available insect repellents. This
substance was discovered and developed by scientists at the U.S. Department of Agriculture in 1946
during a program to develop better repellents for the U.S. Army. It was subsequently registered in 1957
for use by the general public. It is a broad-spectrum repellent that is effective against mosquitoes, biting
flies, chiggers, fleas, ticks, and other many other biting organisms. Twenty years of empirical testing of
more than 20,000 other compounds has not resulted in another marketed chemical product with the
duration of protection and broad-spectrum effectiveness of deet,20 though recently introduced active
ingredients may equal or exceed the effectiveness of deet (see Chapters 18–21). The U.S. Environmental
Protection Agency (EPA) estimates that more than 38% of the U.S. population uses a deet-based insect
repellent every year and that worldwide use exceeds 200,000,000 people annually. However, because it
does not protect against all arthropod-borne diseases, a rational search for an alternative effective broad-
spectrum repellent is needed.
   Despite the obvious desirability of finding an effective oral, systemic mosquito repellent, no such agent
has been identified.20,23 Thus, the search for the perfect topical insect repellent continues. This ideal
agent would repel multiple species of biting arthropods, remain effective for at least eight hours, cause no
irritation to the skin or mucous membranes, cause no systemic toxicity, resist abrasion and ruboff, and
integrate into a greaseless and odorless formulation.
   Efforts to find such a compound have been hampered by the numerous variables that affect the inherent
repellency of any chemical. Repellents do not all share a single mode of action, and surprisingly little is
known about how repellents act on their targets.22 Moreover, different species of mosquitoes may react
differently to the same repellent. To be effective, a repellent must show an optimal degree of volatility,
making it possible for an effective repellent vapor concentration to be maintained at the skin surface
without evaporating so quickly that it loses its effectiveness. Many factors play a role in how effective
any repellent is, including the frequency and uniformity of application, the number and species of the
organisms attempting to bite, the user’s inherent attractiveness to blood-sucking arthropods, and the
overall activity level of the potential host.23 Abrasion from clothing, evaporation and absorption from
the skin surface, wash-off from sweat or rain, higher temperatures, or a windy environment all decrease
repellent effectiveness.23 Each 108C increase in temperature can lead to as much as a 50% reduction in
protection time. The repellents currently available must be applied to all exposed areas of skin;
unprotected skin a few centimeters away from a treated area can be attacked by hungry mosquitoes.




Chemical Functional Requirements for Arthropod Repellent Compounds
A number of studies have shown that chemical compounds containing specific functional groups or
features are more effective arthropod repellents as measured by duration of protection.25,26 Recently, we
have reported a study27 of similarity analysis of stereoelctronic properties (steric and intrinsic electronic
properties) between natural insect juvenile hormone (JH), a synthetic insect juvenile hormone mimic
(JH-mimic, undecen-2-yl carbamate), and deet and its analogues. Structure-activity studies on juvenile
hormones have resulted in the discovery of JH-like compounds that mimic the morphogenetic activity of


q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling                                199


JH with the aim of controlling insect populations. However, no attempt has thus far been made to design
insect repellents rationalizing the pharmacophores obtained from these studies.
   Understanding the mechanism of arthropod repellent activity is a major goal of chemists for designing
more effective repellents. Because the biochemical steps leading to the desired repellent effect, especially
the interaction with the three-dimensional molecular structure of the receptor(s), are still unknown,
various efforts are being made to develop a general structural framework with high probability for
repellent activity to guide the synthesis work.28 The ability of the insect repellents to interact with the
recognition sites in receptors results from a combination of steric and electronic properties. Therefore,
the study of stereoelectronic properties of insect repellents can provide valuable information, not only to
better understand the mechanism of repellent action, but also to develop a reliable pharmacophore to aid
in the design of more efficient analogues. In addition, a three-dimensional (3D) pharmacophore model
would be useful to identify the structural requirements for repellent activity that, in turn, could be utilized
for 3D database queries to search for proprietary and/or commercially available compounds.
   Strategies for reducing the abundance and longevity of arthropod vectors of pathogens have been two-
pronged, centering around habitat control (through chemical, physical, engineering, and biological
means) and the use of personal protection in the form of insect or arthropod repellents. This chapter also
reviews the quantitative structure activity relationships from currently available scientific data on
synthetic and plant-derived insect repellents, and how new and effective repellents can be developed
using computational methodologies.
   Few attempts have previously been made to apply QSAR modeling to repellent activities. This deficiency
may be primarily due to availability of only semi-quantitative data on most of the extensive testing that was
carried out earlier.29 One of the first quantitative attempts for measuring molecular properties such as
lipophilicity, vapor pressure, and molecular chain lengths was by Suryanarayana et al.30 Working with 31
insect repellent compounds, these researchers proposed a QSAR relationship in the form of

                                        PT Z a log P C b log Vp C c log ML                              ð10:1Þ

where PT is the protection time provided by repellent activity, P is lipophilicity, Vp is vapor pressure, ML is
molecular length, and a, b, c, and d are constants.
   Taking into account the paucity of quantitative data on insect repellents and the objectives discussed
above, repellent structural and electronic properties were initially investigated using quantum-chemical
methods to determine any functional dependence with protection time as measured by Suryanarayana
et al.30 The goal of their study was to provide predictive discriminators of insect repellency and a better
understanding of the structure and repellency properties of these compounds. Although the authors’
initial study specifically addresses repellent efficacy, the technique of linking specific molecular
electronic properties to biological activity is generally applicable to both efficacy and toxicity studies.
The authors’ developmental model for structure-activity relationships and generation of pharmacophores
was based on the following two approaches:

     † Consideration of electronic and stereoelectronic chemical properties of the known arthropod
       repellents to identify three-dimensional molecular-interaction pharmacophores.
     † Consideration of pharmacophores or chemical features of known arthropod repellents to
       identify three-dimensional pharmacophores with potential repellent activity.



Electronic and Stereoelectronic Considerations
Because physical-chemical properties of repellents play a significant role in their effectiveness, the role
of molecular electronic properties in relation to repellent protection time was also assessed, using a series
of deet analogues. 30 Using quantum chemical methods, lowest energy conformations and


q 2006 by Taylor & Francis Group, LLC
200                                                                  Insect Repellents: Principles, Methods, and Uses


molecular electronic properties were calculated for 31 amides divided into five different types: (1) N,N-
dimethylamide, (2) N,N-diethylamide, (3) N,N-diisopropylamide, (4) N-ethyl amides, and (5) piperidi-
neamides (Table 10.1). Biological testing of the compounds was performed as reported by
Suryanarayana et al.30 Briefly, a dose of 1 mg/cm2 was applied onto the external surface of a human
fist, followed by exposure for 5 min to 200 female Aedes aegypti (aged 5–7 days). Exposure was repeated
every 30 min until two consecutive bites were observed, defining the protection time as the time up to the
period before the bites.31


TABLE 10.1
Structure and Mosquito Repellent Protection Time of Deet and Its Analogs Organized According to Their
Amide Substituents
                                        O                                                         O
                             2                                                            2
                                  1              R1                                           1              R1
                                         7   N                                                    7    N
                            a                    R2             or                    s                     R2
                    R              6                                        R
                                                                                              6

                           a = aromatic ring                                        s = saturated ring

                                                      Protection
Compound                     Structure                 Time (h)              Ring                       R          R1ZR2

1a                o-Chlorobenzamide                      5                      a                   2-Cl          CH3
1b                Cyclohexamide                          3                      s                    H            CH3
1c                m-Toluamide                            3                      a                  3-CH3          CH3
1d                o-Ethoxylbenzamide                     2.83                   a                 2-OC2H5         CH3
1e                Benzamide                              1.67                   a                    H            CH3
1f                p-Anisamide                            1                      a                 4-OCH3          CH3
2a                m-Toluamide                            5                      a                  3-CH3          C2H5
2b                Benzamide                              4                      a                    H            C2H5
2c                Cyclohexamide                          4                      s                    H            C2H5
2d                o-Ethoxylbenzamide                      3.5                   a                 2-OC2H5         C2H5
2e                p-Toluamide                             2.8                   a                  4-CH3          C2H5
2f                p-Anisamide                            1                      a                 4-OCH3          C2H5
3a                Benzamide                              3                      a                    H            iC3H7
3b                m-Toluamide                            2.67                   a                  3-CH3          iC3H7
3c                Cyclohexamide                          2                      s                    H            iC3H7
3d                p-Anisamide                            1.17                   a                 4-OCH3          iC3H7
3e                o-Ethoxylbenzamide                     1.08                   a                 2-OC2H5         iC3H7
3f                o-Chlorobenzamide                      1                      a                   2-Cl          iC3H7
3g                p-Toluamide                            0.5                    a                  4-CH3          iC3H7
                                                                                                                  R1 R2
4a                m-Toluamide                            0.67                   a                  3-CH3          H C2H5
4b                Benzamide                              0.58                   a                    H            H C2H5
4c                Cyclohexamide                          0.5                    s                    H            H C2H5
4d                p-Toluamide                            0.08                   a                  4-CH3          H C2H5
4e                p-Anisamide                            0.08                   a                 4-OCH3          H C2H5
4f                o-Ethoxylbenzamide                     0.08                   a                 2-OC2H5         H C2H5
                                                                                                                  N, R1, R2
5a                Benzamide                              3                      a                        H        Piperidine
5b                Cyclohexamide                          2                      s                        H        Piperidine
5c                m-Toluamide                            1.42                   a                      3-CH3      Piperidine
5d                o-Chlorobenzamide                      1                      a                       2-Cl      Piperidine
5e                p-Toluamide                            1                      a                      4-CH3      Piperidine
5f                p-Anisamide                            0.75                   a                     4-OCH3      Piperidine



q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling                                             201




FIGURE 10.1 (See color insert following page 204.) Optimized geometry and electrostatic potential profiles of three
repellents that have good (top row, PTZ5 h), moderate (middle row, PTZ1.4 h) and poor (bottom row, PTZ0.08 h)
protection times. First column: optimized geometry; Second column: electrostatic potential onto surface of constant electron
density (0.002 e/au3); Third column: isoelectrostatic potential surface at K10 kcal/mol. Atoms are colored black for carbon,
red for oxygen, blue for nitrogen, and gray for hydrogen. The deepest blue surface in the second column is the most positive,
and the deepest red surface is the most negative. (From D. Ma, K. Bhattacharjee, R. K. Gupta, and J. M. Karle, American
Journal of Tropical Medicine and Hygine, 60, 1, 1999.)

   An examination of the electrostatic potential maps of the repellents at K10 kcal/mol (Figure 10.1),
which roughly correspond to the electronic features beyond the van der Waals surface of the molecules,
indicated that all repellents have a large extended negative potential region extending out from the
carbonyl group. The electrostatic potential profiles of molecules are considered to be key features
through which a molecule fits into a receptor at longer distances, and accordingly, promotes interaction
between complementary sites with the receptor.32 Although this potential characterizes the primary level
of interaction with the receptor, there is no apparent relationship with the size or shape of these surfaces
to protection time. Regions of positive potentials, the blue-colored regions in Figure 10.1, at the van der
Waals surface indicate the electrophilic or acidic sites. Although the location of the most positive
potential (deepest blue color) in the repellent molecules is found to be located adjacent to different
hydrogen atoms on different molecules, the magnitude of the most positive potentials appears to be
related to protection time. All compounds that provided protection for at least 2.8 h have a maximum
positive potential in the range of 16.2–21.1 kcal/mol, whereas all compounds with a positive potential
higher than 21.1 kcal/mol provided protection for no more than one hour.33 Thus, the intrinsic
electrophilicity of the repellent amides appears to play a role in the repellency of a compound.
   The dipole moment is another interesting electronic property that seems to have a role in repellent
activity. This property is the intrinsic polarity of a molecule. Its magnitude is a good indicator of intrinsic
lipophilicity or hydrophobicity. In general, the larger the magnitude, the more likely the compound is
hydrophilic. In study conducted by the authors with 31 repellents,33 the magnitude of the dipole moment
for the most active repellents (PTO3.5 h) was found to be ranging between 3.25 and 3.82 Debye, an


q 2006 by Taylor & Francis Group, LLC
202                                                                 Insect Repellents: Principles, Methods, and Uses


indication that an optimal lipophilicity or hydrophobicity for this class of compounds is necessary for a
molecule to be an active repellent. Although the orientation of the dipole moment of the repellents does
not seem to have any link to protection time, the negative end of the dipole in these compounds was
always observed to be pointing toward the oxygen atom of the carbonyl functional group.
   Atomic charges of the compounds seem to have a significant role in repellency. These charges indicate
the intrinsic reactive character of the individual atoms constituting the molecules. The magnitude of
negative charge on an atom characterizes the nucleophilic nature of the atom, whereas the magnitude of
the positive charge correspondingly characterizes the electrophilic nature of the atom. In the data set
from the above study of repellents,33 a low atomic charge on the amide nitrogen atom in compounds
having low PT values was observed. In general, it was observed that the more negative the charge on the
amide nitrogen atom, the less protection time provided by the compound containing the atom.

Molecular Mechanism of Arthropod Repellent Activity
In the authors’ next study,27 the stereoelectronic features of 15 of the 31 arthropod repellents (Table 10.2)
reported by earlier workers were assessed,30 by identifying both electronic and steric requirements for
repellent activity. In addition, these profiles were compared with JH, not only to identify the
stereoelectronic requirement, but also the probable mechanism of repellent action of the compounds.


TABLE 10.2
Structure and Mosquito Repellent Protection Time of Deet and Its Analogs Organized According to Their
Amide Substituents
                                        O                                                          O
                             2                                                         2
                                  1              R1                                        1                R1
                                         7   N                                                     7   N
                            a                    R2            or                  s                       R2
                    R              6                                         R
                                                                                               6

                           a = aromatic ring                                     s = saturated ring

                                             Protection Time
Compound                   Structure               (h)                 Ring                R                      R1ZR2

1a                      m-Toluamide                   5                  a          3-CH3                        C2H5
                          (deet)
1b                      Cyclohexamide                 4                  s          H                            C2H5
1c                      p-Anisamide                   1                  a          4-OCH3                       C2H5
2a                      o-Chlorobenza-                5                  a          2-Cl                         CH3
                          mide
2b                      m-Toluamide                   3                  a          3-CH3                        CH3
2c                      p-Anisamide                   1                  a          4-OCH3                       CH3
3a                      Benzamide                     3                  a          H                            iC3H7
3b                      p-Anisamide                   1.17               a          4-OCH3                       iC3H7
3c                      p-Toluamide                   0.5                a          4-CH3                        iC3H7
                                                                                                                  R1 R2
4a                      m-Toluamide                   0.67               a          3-CH3                        H C2H5
4b                      Cyclohexamide                 0.5                s          H                            H C2H5
4c                      o-Ethoxylbenza-               0.08               a          2-OC2H5                      H C2H5
                          mide
                                                                                                                 N, R1, R2
5a                      Benzamide                     3                  a          H                            Piperidine
5b                      m-Toluamide                   1.42               a          3-CH3                        Piperidine
5c                      p-Anisamide                   0.75               a          4-OCH3                       Piperidine



q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling                                     203


                                            R    R'                  O


                                        O                                 OMe
                                                 JH
                                                             (1) R = R' = Et
                                                             (2) R = Et, R' = Me
                                                             (3) R = R' = Me

                                                                          O


                                                                      O       N
                                                                              H
                                                JH - mimic

FIGURE 10.2 Structures of JH and JH-mimic. (From A. K. Bhattacharjee, R. K. Gupta, D. Ma, and J. M. Karle, Journal of
Molecular Recognition, 13, 213, 2000.)


Juvenile hormones (Figure 10.2) are ubiquitous growth regulators among insects and serve as a rational
source for the design of synthetic insect growth regulators.24,34,35 Structure-activity studies on JH have
resulted in the discovery of JH-like compounds25,26 that mimic the morphogenic activity of the natural
compound and are used commercially in insect control. However, to the knowledge of the authors, no
attempt has been made thus far to design insect repellents based on JH chemical structure and activity.
   Although considerable research efforts have focused on why humans are attractive to insects,36
especially mosquitoes, and many chemicals have been discovered to have repellent activity, the mode of
action of repellents remains poorly understood. In recent years, a tentative model of physical properties
required for potent repellency of the two well-known insect repellents,20 deet and N,N-diethylpheny-
lacetamide (DEPA), against Aedes aegypti mosquitoes has been proposed on the basis of their
lipophilicity, vapor pressure, and molecular length. However, it is now widely believed37–39 that a
repellent must impact insects’ olfactory sense and that the olfactory sensation is primarily controlled by
JH responses or activity. Therefore, an ideal repellent must be volatile, must come in contact with the
mosquito’s olfactory organ, and have some degree of lipid solubility to trigger the olfactory sensation.
The gas-phase molecular properties of the repellents thus should be an important aspect of studies on the
mechanism of this interaction process with the olfactory organ. Quantum-chemically calculated
stereoelectronic properties can provide an accurate estimate of gas-phase properties of molecules.
Calculating and assessing these properties should be important objectives of chemists before the design
and synthesis of new repellents.
   The mechanism of olfactory sensation may be viewed as an interaction of fundamental molecular
forces between the repellents and the JH receptor of the insects, from the point of view of the century-old
lock-and-key hypothesis of Emil Fischer.40 According to the lock-and-key hypothesis, the biological
activity of a compound may be accounted for through a molecular recognition mechanism between the
biomolecule (lock) and the active compound (key). Because the JH receptor recognizes the stereo-
electronic features of the active compound (repellent) that resemble those of JH itself, and not of atoms
per se,41–43 a comparative analysis of these features should provide a wealth of molecular level
information that would not only aid in the design of new repellents, but also illuminate more completely
the fundamental forces that affect the function and utility of the compounds.
   In recent years, several other studies based on quantitative derivations44–46 have shown that this
recognition process can be analyzed from three types of three-dimensional molecular similarity studies:
(1) steric, (2) electrostatic, and (3) hydrophobic. It is well documented that bioactive compounds
(ligands) will bind to a receptor in a similar manner by aligning their common molecular field or property
characteristics to the receptor.44 This concept is known as bioisosterism, wherein atoms or functional
groups with similar properties are used for ligand design.47 The study of bioisosterism has been one of the


q 2006 by Taylor & Francis Group, LLC
204                                                     Insect Repellents: Principles, Methods, and Uses


most common means of discovery of new leads in pharmaceutical research. This method of selection of
pharmacophores is mainly based on simple superposition principles using the analogy of complemen-
tarity. The degree of complementarity between the molecular fields of the bioactive agent and its receptor
should be directly related to the binding strength and relative activity of the agent. Similarity may also be
determined by comparing the molecular graphs.47 The word “similarity” in the present study means that
two molecules have a common bioisosteric group.
   Accordingly, the authors’ study27 assessed the similarity of the stereoelectronic properties of deet and
its analogs to natural JH (Table 10.2, Figure 10.2 where RZmethyl), and to a synthetic JH-mimic
terpenoid. This study was an attempt to gain a better understanding of the mechanism of action of
the deet-type insect repellents and to aid in the design and synthesis of more efficacious repellents.
Undecen-2-yl carbamate, the JH-mimic, is a potent inhibitor of metamorphosis of the common mosquito,
Culex pipens.25 Structurally, deet, its analogs, and the JH-mimic all have an ON–CaO fragment
(Table 10.2, Figure 10.2), making it likely that a similar recognition interaction with the receptor will
take place. Juvenile hormone is structurally different, containing an –O–CaO fragment. However,
similarity in molecular electronic shape, not solely the similarity in chemical structure, has long been
recognized as the dominant factor for olfactory sensations.48 Thus, using data from the earlier study33 on
predicting mosquito repellent activity from calculated stereoelectronic properties on 31 deet analogs, the
authors carried out a computational study based mainly on the similarity analysis of the stereoelectronic
properties of deet and 14 of its analogs with JH and JH-mimic using the semi-empirical AM1 quantum
chemical method.


Computational Procedure
Computational calculations were performed using SPARTAN version 5.049 running on a Silicon
Graphics Indigo Extreme R4000 workstation. A detailed conformational search of JH and JH-mimic
was performed by multiple rotation of single bonds in the compounds, thereby generating several
low-energy conformers with varying population densities. The most abundant and the lowest energy
conformers were identified. The geometry of these conformers was optimized, and the electronic
properties were calculated using the optimized geometry. Geometry optimization and energy
calculations were performed on the compounds in the gaseous phase at the semi-empirical AM1
quantum chemical level using the method as implemented in SPARTAN. Three-dimensional
molecular electrostatic potential (MEP) maps for all compounds were calculated using the
SPARTAN calculations of the AM1-optimized geometry of the molecules. The MEPs were
sampled over the entire accessible surface of a molecule (corresponding roughly to the van der
Waals contact surface) and into space extending beyond the molecular surface, providing a measure
of charge distribution from the point of view of an approaching reagent. The regions of negative
potential indicated areas of excess negative charges and, therefore, suitable attraction sites in the
molecule for the positively charged test probe.


Results and Discussion
Conformational Analysis
The lowest energy conformers of JH-mimic were identified by systematic rotation of the single bonds.
This procedure generated 256 conformers of JH-mimic, identifying the low energy conformers along
with their corresponding Boltzmann population densities. The lowest energy conformer of JH-mimic has
a 75.6% population density, whereas the other conformers were present in varying population densities
ranging from 6 to 0.05%, with energies more than 5.0 kcal/mol greater than the lowest energy conformer.
In the lowest energy conformers, the amide moiety in both JH-mimic and the deet compounds33 was
planar and superimposable on each other. The nonbonded distances, N–O and CaO–CR1, were within
    ˚
0.1 A of each other (Table 10.3).


q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling                          205


        TABLE 10.3
        Selected Nonbonded Distances and Total Surface Areas Containing the C7, O, N, and CR1
        Atoms
                                                                               Surface Area,
        Compound                               ˚
                                        N–OaC, A              ˚
                                                     CaO–CR1, A            ˚
                                                                           A2 C7, O, N, CR1 Atoms

        JH-mimic                          2.345         2.871                       11.5
        1a                                2.296         2.751                       13.6
        1b                                2.282         2.707                       13.6
        1c                                2.296         2.746                       13.6
        2a                                2.293         2.761                       13.6
        2b                                2.286         2.749                       13.6
        2c                                2.285         2.744                       13.6
        3a                                2.297         2.810                       11.7
        3b                                2.296         2.806                       11.7
        3c                                2.296         2.809                       11.7
        4a                                2.296         2.811                       11.7
        4b                                2.292         2.815                       11.7
        4c                                2.298         2.803                       11.7
        5a                                2.293         2.761                       13.6
        5b                                2.293         2.761                       13.6
        5c                                2.289         2.757                       13.6




   Thus, although JH, JH-mimic, and the deet molecules have many degrees of conformational freedom
in their structure, the main bioactive pharmacophore (the ester, carbamate, or the amide group) was
found to be superimposable. This ensures the steric similarity of the pharmacophore.27
   Conformational search calculations on the structure of the natural JH molecule where RZmethyl
identified three conformers of significant abundance, with a relatively small energy difference of
3.9 kcal/mol between the maximum and the minimum energy conformer. An energy barrier
of 3.9 kcal/mol can easily be surmounted in biological systems, and, statistically, the distribution of
all three conformers cannot be ruled out. The ester moiety is flat in all three conformers, and the
                                                                           ˚
nonbonded distances OaO and CaO–CR1 are equal to 2.23 and 2.58 A, respectively. The three
conformers differ in the conformation of the alkyl chain, due to rotations about the single bonds.27
Although JH has an epoxide moiety at one end of the molecule, this functionality does not appear to be
important for growth regulator activity, because mimics lacking this functionality are potent growth
inhibitors.26

Molecular Similarity Analysis of JH-Mimic and Deet Compounds
The analysis of molecular recognition process in this investigation was based on the strategy of
superimposition of the amide fragments and analysis of steric, electrostatic, and hydrophobic properties.
JH-mimic is more structurally similar to the deet compounds than to the natural JH, as JH has an ester
rather than an amide moiety. Therefore, this section concentrates on the comparison of the deet
compounds to JH-mimic.
   The surface area and volume of the amide-containing portion of JH-mimic and the deet compounds
show considerable similarity. The surface area containing the amide C7, O, N, and CR1 atoms of
                                                                ˚
JH-mimic and the deet compounds ranges from 11.5 to 13.6 A2 (Table 10.3), whereas the calculated
                                                        ˚                       ˚
steric bulk of this amide portion in JH-mimic is 8.7 A3 and the deet 0.2 A3 compounds is 8.3 A3,      ˚
respectively. Because steric complementarity is a prerequisite for ligand-receptor recognition, active
biological agents with a common receptor binding site should possess sterically similar binding surfaces.
This contribution reflects molecular size and overall shape, features important to the steric


q 2006 by Taylor & Francis Group, LLC
206                                                         Insect Repellents: Principles, Methods, and Uses


complementarity of the ligand at the binding site.50 Bond distances and angles of the amide moiety in
deet and its analogs are also similar to those in the JH-mimic (Table 10.4). The bond distances of
                                                                       ˚
JH-mimic, deet, and deet’s most repellent analogs are within 0.007 A of each other. The bond angles and
the dihedral angles differ from each other by up to 88, and 78 to about 198, respectively, a reasonable
variation keeping in mind the large intrinsic differences in the geometry of the molecules.
   The similarity of electrostatic characteristics of the deet compounds with JH-mimic is likely to result
in a similar recognition interaction with the JH receptor to promote binding interactions. The electrostatic
characteristics include Mulliken charges, electrostatic potentials at essentially the van der Waals surface,
dipole moment, and the profiles of electrostatic potential beyond the van der Waals surface. Electrostatic
complementary interactions are believed to be long-range interactions between a ligand and its binding
site, and are considered to be a very important contributing factor for a ligand/protein binding
mechanisms.51–54 This complementarity essentially means that the charge distribution of a substrate
has to find its counterpart at the binding sites to allow maximum interaction with the receptor.55 It works
like a magnet between them and, thereby, contributes to the binding affinity. The calculated Mulliken
charges and the electrostatic potential at the amide atoms in JH-mimic and deet and its analogs are
presented in Table 10.5.
   The charge of the carbonyl oxygen atom of JH-mimic is K0.02 electrons more negative than for deet
and its analogs, while the negative potential by the carbonyl oxygen atom, K73.2 kcal/mol, falls in the
same range as calculated for deet and its analogs, K73.1 to K77.3 kcal/mol. The carbonyl oxygen atom
is also the site for the most negative potential27 (see Figure 10.3) in deet, its analogs, and JH-mimic,
making the carbonyl oxygen atom the most nucleophilic site in all the molecules, as this site has the
maximum localized electron density. Although the calculated dipole moment is somewhat lower in
JH-mimic than in the deet compounds, the dipole moment is pointing toward the carbonyl oxygen atom
in all the compounds. Thus, the carbonyl oxygen atom seems to be the most reactive site in both
JH-mimic and the deet compounds.
   Conversely, the site for the most positive potential is considered to be most electrophilic or acidic
because of minimum electron density. In JH-mimic, the most positive potential is located by the amide
hydrogen atom with a value of 37.3 kcal/mol, whereas in deet and its analogs, it is scattered around different


TABLE 10.4
Selected Structural Parameters of JH-Mimic and Deet Compounds
                                              ˚
                               Bond Distance, A              Bond Angle, 8        Dihedral       CR1-N-C7-C1
                                                                                  Angle, 8           or
Compound              C7aO              N–C7      N–CR1   N–C7aO     CR1-N-C7   CR1-N-C7ZO       CR1-N-C7-O

JH-mimic               1.241            1.388     1.442    127.8       119.1        23.0            K161.1
1a                     1.247            1.392     1.447    120.8       119.5         2.6             179.9
1b                     1.248            1.393     1.448    119.4       118.5         4.6            K178.6
1c                     1.247            1.393     1.447    120.7       120.7         2.2             179.5
2a                     1.246            1.387     1.437    120.9       120.2         7.8            K174.9
2b                     1.248            1.387     1.436    120.2       120.4         3.2            K176.9
2c                     1.248            1.388     1.437    120.0       120.3        K3.2            K176.6
3a                     1.248            1.384     1.436    121.4       122.2         8.1            K173.1
3b                     1.248            1.385     1.436    121.2       122.1         8.0            K173.2
3c                     1.248            1.385     1.436    121.3       122.2         8.1            K173.1
4a                     1.248            1.384     1.437    121.3       122.3         7.5            K173.5
4b                     1.247            1.380     1.437    121.3       122.8         7.0            K174.9
4c                     1.249            1.384     1.439    121.4       121.5        11.6            K173.5
5a                     1.247            1.396     1.450    120.5       119.7       K10.7             173.0
5b                     1.247            1.391     1.494    120.6       119.7       K10.4             173.3
5c                     1.248            1.391     1.492    120.2       119.7       K12.5             171.2



q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling                                           207


TABLE 10.5
Comparison of Dipole Moments, Mulliken Charges (Electrons), and MEPs
                          Carbonyl O Atom                 Amide N Atom                 C7 Atom          Dipole Moment,
Compound              Charge            MEPa          Charge            MEPa           Charge               Debye

JH-mimic               K0.39            K73.2          K0.32            K37.3             0.38                 2.02
1a                     K0.35            K75.0          K0.33            K23.2             0.34                 3.68
1b                     K0.37            K74.5          K0.33            K25.1             0.30                 3.25
1c                     K0.36            K75.5          K0.33            K30.0             0.35                 3.55
2a                     K0.35            K73.1          K0.33            K22.8             0.35                 3.82
2b                     K0.36            K76.2          K0.34            K17.0             0.35                 3.45
2c                     K0.37            K75.7          K0.34            K18.7             0.35                 4.37
3a                     K0.36            K75.9          K0.37            K20.3             0.34                 3.63
3b                     K0.37            K76.7          K0.37            K27.8             0.34                 3.25
3c                     K0.36            K77.3          K0.37            K22.1             0.34                 3.74
4a                     K0.36            K74.5          K0.37            K25.8             0.34                 3.22
4b                     K0.37            K75.4          K0.38            K26.7             0.30                 3.46
4c                     K0.37            K75.7          K0.35            K38.6             0.35                 3.55
5a                     K0.35            K73.8          K0.31            K33.0             0.34                 3.52
5b                     K0.35            K75.6          K0.31            K33.1             0.34                 3.56
5c                     K0.36            K75.7          K0.32            K30.2             0.35                 3.27
a
  Most negative electrostatic potential located by indicated atom expressed in kcal/mol superimposed on the isodensity
surface (0.002 e/au3) of the molecule.




FIGURE 10.3 (See color insert following page 204.) Molecular electrostatic potential plotted onto the total electron
density surface defined as 0.002 e/au3 (essentially the van der Waals surface) of the maximum energy conformer of JH
where R Z R 0 Z methyl, JH-mimic, and deet (compound 1a) (Table 10.2). The surface is color-coded according to the
magnitude of its potential in units of kcal/mol. The red region is by the carbonyl oxygen atom. (From A. K. Bhattacharjee,
R. K. Gupta, D. Ma, and J. M. Karle, Journal of Molecular Recognition, 13, 213, 2000.)


q 2006 by Taylor & Francis Group, LLC
208                                                      Insect Repellents: Principles, Methods, and Uses


hydrogen atoms in the molecules, in the range of 15.4–35.2 kcal/mol. The calculated charge densities on
the amide nitrogen atom and the C7 atom are also found to be quite close between JH-mimic and the deet
compounds, keeping in mind the diverse nature of the substituents in other parts of the molecules.
   Furthermore, the profiles of electrostatic potential beyond the van der Waals surface at a constant
potential of K10.0 kcal/mol are comparable to a large negative potential region localized by the amide.27
Electrostatic potential characteristics beyond the van der Waals surface of the molecules are believed to
be the key features primarily responsible for recognition interaction between an approaching molecule
and its receptor at longer distances of separation.32 It is through this potential that a molecule reacts with
any other system in its vicinity, recognizes its receptor, and accordingly promotes interaction between
the complimentary sites.
   The electrostatic potential of functional groups that are commonly found in diphenylether and
terpenoid JH mimics are similar to each other in terms of their electrostatic potential characteristics.
This electrostatic bioisosterism has led to the understanding of the universality of active structures and
aided in the design of new active analogs.52 Therefore, the present investigation indicates that the
electrostatic characteristics of the deet compounds are likely to cause similar recognition interactions
with the JH receptor as the JH of the insects at a distance to promote binding interactions.
   It is interesting to note that the localized negative potential region by the amide moiety in the deet
compounds is qualitatively linked to their potent repellent activity, with the less-potent repellent
compounds having a more extended, and therefore a more diffuse, negative potential zone.27 It
appears that a more localized negative potential region in the amide group, as seen with JH-mimic, is
consistent with higher protection times. Because the similarity of the negative potential profiles at
K10.0 kcal/mol seems to play a role in the repellent potency of deet analogs, this observation should aid
in the design of potent analogs of this class of insect repellents.
   Large hydrophobic regions in the molecule appear to be necessary for both recognition and potent
repellent activity. Hydrophobic effects are the result of averaged electrostatic interaction of the molecule
with its surroundings, solvent, and protein environment. Sites of nonpolar or weakly polar regions in
different molecules tend to come together to escape contact with water and to minimize the dehydration
free energies.45 Thus, matching the nonpolar regions of ligands with the receptor sites gives a reasonable
measure of hydrophobic complementarity, and also represents the stabilization of the enzyme–substrate
or ligand–receptor complex.
   Polarity of a certain region in the molecule can be regarded as proportional to the electrostatic field.
A strong electrostatic field of a molecule attracts molecules having large dipoles, such as water, while the
weak electrostatic field regions of the molecule do not attract water molecules and are, therefore,
hydrophobic.45,46 Different approaches have recently appeared to theoretically represent hydrophobic
interactions in terms of local solute–solvent electrostatics. 56 However, a simple assessment of
hydrophobic similarity may be carried out by determining the distribution of charges or electrostatic
potentials at different regions on the van der Waals surface of the molecule. The observed low-dipole
moments (!4 Debye) of JH-mimic and the deet analogs also correspond to the lipophilic nature of the
compounds. Because olfactory sensations of the insects require some degree of lipid solubility,36
hydrophobicity of the repellents is likely to be an important factor for potent repellent activity.

Correlation of Molecular Orbital Properties in JH-Mimic and in Deet and Its Analogs
The deet compounds have similar highest occupied molecular orbital (HOMO)–lowest unoccupied
molecular orbital (LUMO) energy gaps as shown by the h values in Table 10.6, an index of intrinsic
reactivity.57 These values indicate that the deet compounds are nearly similar in intrinsic reactivity. The
energy of HOMO and LUMO orbitals plays a major role in governing chemical reactions. The energy
difference between the orbitals is known as the electronic band gap, and is often responsible for the
formation of many charge-transfer complexes.57 Table 10.6 shows a relatively constant EHOMO with a
large negative value and a more variable and much smaller magnitude ELUMO, implying a greater role of
LUMO or electron-acceptor ability of the compounds than their electron donating. Therefore, the


q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling                           209


            TABLE 10.6
            HOMO and LUMO Eigenvalues
                                                 Eigenvalues, eV
            Compound                    HOMO               LUMO           hZ(ELUMO – EHOMO)/2

            JH-mimic                    K9.847              0.98                   5.41
            1a                          K9.542              0.146                  4.84
            1b                          K9.555              1.514                  5.53
            1c                          K9.207              0.137                  4.67
            2a                          K9.589             K0.187                  4.70
            2b                          K9.518              0.027                  4.77
            2c                          K9.256              0.011                  4.63
            3a                          K9.854             K0.090                  4.88
            3b                          K9.274             K0.063                  4.60
            3c                          K9.599             K0.105                  4.74
            4a                          K9.593             K0.051                  4.77
            4b                          K9.889              1.536                  5.71
            4c                          K9.319             K0.047                  4.63
            5a                          K9.514              0.111                  4.81
            5b                          K9.495              0.151                  4.82
            5c                          K9.236              0.111                  4.67


electron transfer from a suitable receptor molecular orbital to the LUMO of the deet compounds, rather
than a donation of electrons from the deet compounds, seems a more plausible mechanism for
the compounds.

Molecular Electronic Properties of JH
Stereoelectronically, the maximum energy conformer has features adjacent to the carbonyl oxygen atom
most similar to the deet analogs and JH-mimic. The carbonyl oxygen atom of all the JH conformers is
the most nucleophilic site, being the most negative potential site on the van der Waals surface in the
molecule. It varies from K66.1 to K71.3 kcal/mol for the maximum energy conformer. The electrostatic
potential feature beyond the van der Waals surface generated by the carbonyl oxygen atom of the
maximum energy conformer is most similar to the deet analogs and JH-mimic, as it has the largest
K10 kcal/mol potential surface of the three conformers.27
   Other electronic features of the JH conformers are similar to the deet analogs and JH-mimic. Again,
the surface of JH has large hydrophobic regions (light green regions of potentials ranging between 11.7
and 12.1 kcal/mol, Figure 10.3). The HOMO, LUMO, and reactivity indices range from K9.574 to
K9.602 eV, K0.038 to 0.025 eV, and 4.76 to 4.81 kcal/mol, respectively. The dipole moment of the JH
conformers varies from 1.83 to 4.45 Debye. Thus, clearly there exists an electronic similarity between the
natural juvenile hormone molecule, the JH-mimic, and the deet analogs that most likely implies an
electrostatic bioisoterism between all the molecules.




Development of a New Model for Repellent Research
Chemical-Feature Based Considerations
The factors involved in attracting mosquitoes to a host are complex and are not fully understood.58
Mosquitoes use, at the very least, visual, thermal, and olfactory stimuli to locate a host. Of these,
olfactory cues are probably the most important. It has been estimated that 300–400 compounds are


q 2006 by Taylor & Francis Group, LLC
210                                                       Insect Repellents: Principles, Methods, and Uses


released from a human body as by-products of metabolism and that more than 100 volatile compounds
can be detected in human breath (see Chapter 4). Of these odors, only a fraction have been isolated and
fully characterized. Carbon dioxide and lactic acid are the two best-studied mosquito attractants. Carbon
dioxide, released mainly from breath but also from skin, serves as a long-range airborne attractant and
can be detected by mosquitoes at distances of up to 40 m. Lactic acid, in combination with carbon
dioxide, and uric acid are also highly attractive.
   It is also believed that mosquitoes can sense which host is the richest source of cholesterol and
B vitamins, nutrients that mosquitoes cannot synthesize. Mosquitoes have chemo-receptors on their
antennae that are stimulated by lactic acid.59 It is also speculated that the same receptors may be inhibited
by deet-based insect repellents.60
   In a continuation of the efforts to design and discover new insect repellents from structure–activity
relationship studies27,33 and to better understand the mechanism of insect repellency, the authors have
developed61 a three-dimensional chemical function-based pharmacophore model for potent arthropod
repellent activity to provide a foundation for compound database searches to aid the discovery of new
repellent candidates. We have utilized 3D QSAR-CATALYSTw* methodology on a training set of
eleven known structurally diverse insect repellent compounds, including deet, to develop the model
whose validity applies to a variety of other arthropod repellents beyond that of the training set.


Significance and Uniqueness of the Methodology
Thus far, no attempt has been made to design insect repellents rationalizing the pharmacophores obtained
from the similarity analysis of studies on stereoelectronic properties. The authors developed a
pharmacophore from a training set of deet and its eleven analogues using 3D QSAR. This was
accomplished by utilizing the existing expertise and CATALYST computer software62 at Walter Reed
Army Institute of Research (WRAIR), Silver Spring, Maryland, U.S.A. The prerequisite for developing a
reliable 3D-QSAR model for a novel insect repellent compound is the correlation of a characteristic and
reproducible biological activity to structural information of the respective compound. The confor-
mational model of the compound in the training set has enabled us to use the best three-dimensional
arrangement of chemical functions predicting the repellent activity variations among the compounds in
the training set. The pharmacophore has also facilitated the search for compound databases to identify
new repellent compounds.


Computational Methods and Materials
Procedure for Development of the 3D-QSAR Pharmacophore Model
The 3D-QSAR study was performed using CATALYST 4.8 software.62 The algorithm treats molecular
structures as templates composed of chemical functions localized in space that will bind effectively with
complementary functions on the respective binding proteins. The most relevant biological features are
extracted from a small set of compounds that cover a broad range of activity.63
   This process makes it possible to use structure and activity data for a set of lead compounds to generate
a pharmacophore representative of the activity of the lead set. At the heart of the software is the HypoGen
algorithm that allows identification of pharmacophores that are common to the “active” molecules in the
training set but are absent in the “inactives.”64 Structures of the arthropod repellent compounds
(Table 10.7) were edited within CATALYST and energy minimized to the closest local minimum
using the generalized CHARMM-like forcefield as implemented in the program. Molecular flexibility
was taken into account by considering each compound as an ensemble of conformers representing
different accessible areas in a three-dimensional space. The “best searching procedure” was applied to
select representative conformers within 10 kcal/mol of the global minimum.65

* Registered trademark of Accelrys Inc., San Diego, CA.


q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling                                    211


              TABLE 10.7
              Names and Activities of the Repellents Used to Create the Training Set
              Compound                           Name                       Repellent Activity (in h of PTa)

              1                     Deet (N,N-diethyl-m-toluamide)                        1.0
              2                     N,N-Diethyl-2-ethoxybenzamide                         0.5
              3                     N,N-Dipropyl-2-benzyloxyacetate                       0.5
              4                     1-butyl-4-methylcarbostyril                           2.0
              5                     N,N-Dipropyl-2-ethoxybenzamide                        0.3
              6                     2-butyl-2-ethyl-1,3-propanediol                       1.7
              7                     1,3-bisbutoxymethyl-2-imidazol                        0.6
              8                     N,N-Diethyl-2-chlorobenzamide                         1.2
              9                     Hexachlorophenol                                      0.2
              10                    1,3-propanediolmonobenzoate                           7.5
              11                    Diisobutylmalate                                      2.5
              a
                  PT is protection time in hours provided by the repellent compounds.


   Conformational models of the training set of 11 repellents were generated that emphasize
representative coverage within a range of permissible Boltzmann population with significant abundance
(within 10.0 kcal/mol) of the calculated global minimum. This conformational model was used for
pharmacophore generation within CATALYST, which aims to identify the best three-dimensional
arrangement of chemical functions, such as hydrophobic regions, hydrogen bond donor, hydrogen bond
acceptor, and positively or negatively ionizable sites, distributed over a three-dimensional space
explaining the activity variations among the compounds in the training set. The hydrogen bonding
features are vector functions, whereas all other functions are points.
   Pharmacophore generation was carried out by setting the default parameters in the automatic generation
procedure in CATALYST (function weightZ0.302, mapping coefficientZ0, resolutionZ260 pm,
andactivity uncertaintyZ3). An uncertainty “D” in the CATALYST paradigm indicates an activity
value lying somewhere in the interval from “activity divided by D” to “activity multiplied by D.” The
statistical relevance of the obtained pharmacophore is assessed on the basis of the cost relative to the null
hypothesis and the correlation coefficient.62,64 The pharmacophores are then used to estimate the
activities of the training set. These activities are derived from the best conformation generation model of
the conformers displaying the smallest root-mean square (RMS) deviations when projected onto the
pharmacophore. HypoGen considers a pharmacophore to be one that contains features with equal
weights and tolerances. Each feature (e.g., hydrogen-bond acceptor, hydrogen-bond donor, hydrophobic
regions, positive ionizable group, etc.) contributes equally to estimate the activity. Similarly, each
chemical feature in the HypoGen pharmacophore requires a match to a corresponding ligand atom to be
within the same distance of tolerance.64 The method has been documented to perform better than a
structure-based pharmacophore generation.63

Bioassay for Mosquito Repellency
The new arthropod repellent candidates were tested for repellent efficacy against Aedes aegypti using an
in vitro blood feeding system. The in vitro test system provided an estimate of the amount of repellent
that must be applied to produce a given level of effectiveness against an arthropod test population
(i.e., the compound’s inherent repellency). Mosquitoes were reared under standardized conditions and
held in a cage at 278C and 75% RH until testing. This test system consisted of a mosquito blood feeder, a
constant-temperature water circulator, and a specially designed cage. The mosquito blood-feeder
contained five circular blood reservoirs, each of which was filled with outdated human blood and
covered with the candidate repellent-treated Baudruche membrane. In the beginning, the candidate
repellents were diluted in ethanol to provide concentrations of 0.02, 0.04, 0.08 and 0.16 mg/cm2. The test


q 2006 by Taylor & Francis Group, LLC
212                                                         Insect Repellents: Principles, Methods, and Uses


materials, including the control, were applied randomly to the five separate membrane positions. Then,
250 female mosquitoes (5–15 days old) were given access to the blood reservoirs on a “free choice” basis
by sliding back a door in the floor of the test cage. The number of mosquitoes probing and feeding on
each well was noted at 2-min intervals. The test was terminated at the end of 20 min. The test results were
expressed as the total of ten feeding counts. The effective dose was then calculated from a probit analysis
of the feeding count obtained in the respective tests. The statistical distribution of tested chemical
sensitivity levels for Aedes aegypti was calculated from the dose-response regression equation.66–68


Results and Discussion
The three-dimensional chemical function or feature-based pharmacophore for arthropod repellent
activity of a compound developed in the present study was found to contain two aliphatic hydrophobic
functions, one aromatic hydrophobic (aromatic ring) function and one hydrogen bond acceptor function
in specific geometric locations surrounding the molecular space (Figure 10.4). This implies that an insect
repellent compound needs to have the physico-chemical characteristics described above to have
potent activity.
   The pharmacophore model was generated by creating a training set of 11 structurally diverse known
arthropod repellent compounds having a broad range of repellent activities as shown in Figure 10.5
(diagrams of the structures found in Table 10.7). The repellent activity of the 11 repellent compounds in
the training set that includes deet covers a broad range of activity, from an ED50 of about 1 mg/cm2 to
about 50 mg/cm2 (Table 10.8 and Table10.9). CATALYST methodology62 was used to develop the
model by placing suitable constraints on the number of available chemical features, such as aromatic
hydrophobic or aliphatic hydrophobic interactions, hydrogen bond donors, hydrogen bond acceptors,
hydrogen bond acceptors (lipid), and ring aromatic sites, to describe the arthropod repellent activity of
the compounds. Earlier reported27 results of quantum chemical calculations and the stereoelectronic
properties of these compounds provided guidance for selection of these physico-chemical features.




FIGURE 10.4 Pharmacophore model for insect repellent activity. It is characterized by two hydrophobic aliphatic
functions, one aromatic function, and one hydrogen bond acceptor function. The hydrogen bonding feature is a vector;
whereas, all other functions are points. The sphere indicates the tolerance area under the specific function. (From
A. K. Bhattacharjee, W. Dheranetra, D. A. Nichols, and R. K. Gupta, QSAR and Combinatorial Science, 24, 593, 2005.)


q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling                                             213




FIGURE 10.5 The correlation diagram between the protection time (insect repelling time) conferred by the compounds in
the trainning set and their predicted protection time (R Z 0.9). (From A. K. Bhattacharjee, W. Dheranetra, D. A. Nichols, and
R. K. Gupta, QSAR and Combinatorial Science, 24, 593, 2005.)

   During the pharmacophore development, molecules were mapped to the features with pre-determined
conformations generated using the “fast fit” algorithm in CATALYST. The conformational energy for
developing the set of three-dimensional conformers ranged between 0 and 20 kcal/mol. The procedure
resulted in the generation of 10 alternative pharmacophores for repellent activity of the compounds and

                          TABLE 10.8
                          Predicted and Experimentally Determined Protection Times of
                          the Repellents in the Training Set
                                            Experimental               Predicted
                          Compd.               PT (h)                   PT (h)                Error

                          1                       1.0                      1.4                 1.4
                          2                       0.5                      0.73                1.5
                          3                       0.5                      0.52                1.0
                          4                       2.0                      1.7                K1.2
                          5                       0.3                      0.17               K1.8
                          6                       1.7                      0.68               K2.5
                          7                       0.6                      0.44               K1.4
                          8                       1.2                      2.1                 1.7
                          9                       0.2                      0.3                 1.5
                          10                      7.5                     14.0                 1.9
                          11                      2.5                      2.6                 1.0



q 2006 by Taylor & Francis Group, LLC
214                                                                Insect Repellents: Principles, Methods, and Uses


                TABLE 10.9
                Description of Cost Analysis of the Pharmacophoresa
                Hypothesis              Total Cost         Fixed Cost         RMS           Correlation

                1st                       44.9356            40.0875         0.82985         0.918127
                2nd                       45.4925            40.0875         0.84233         0.908573
                3rd                       45.8306            40.0875         0.84894         0.895599
                4th                       46.8375            40.0875         0.88776         0.881202
                5th                       47.6982            40.0875         0.89736         0.879984
                6th                       48.9341            40.0875         0.90091         0.872108
                7th                       49.0084            40.0875         0.92756         0.866847
                8th                       49.8163            40.0875         0.91329         0.852024
                9th                       50.2185            40.0875         0.98337         0.833532
                10                        51.1013            40.0875         0.98973         0.807611
                Null cost                110.8533            40.0875         1.00971         0.0
                a
                    Log output file showing the calculated statistics.




appeared to perform quite well for the training set. The correlation coefficients ranged from 0.91 to 0.87
for six of the ten models. The total costs of the pharmacophores varied over a narrow range and the
difference between the fixed cost and the null cost was 71 bits, satisfying the acceptable range as
recommended in the cost analysis of the CATALYST procedure.62,64
   Significantly, the best pharmacophore, characterized by two hydrophobic aliphatic functions, one
aromatic ring function, and one hydrogen bond acceptor function (Figure 10.4), was also statistically
the most relevant pharmacophore. The predicted arthropod repellent activity values, along with the
experimentally determined protection time (in hours) for repellent activity, of the compounds are
presented in Table 10.2. A plot of the protection time conferred by the compounds in the training set and
their predicted protection time demonstrated a good correlation (RZ0.91), indicating the predictive
power of the pharmacophore (Figure 10.5). The highly potent analogues of the series mapped all the
functional features of the best hypothesis with high scores (e.g., Compd. 1, Figure 10.6a), whereas the
less-potent compounds mapped fewer of the features (e.g., Compd. 9, Figure 10.6b). In order to further
cross-validate the model, it was mapped onto four other earlier studied repellent candidates (Figure 10.7)
in the authors’ laboratory: (1) N,N-diethyl-2-(3-trifluoromethyl-phenyl)-acetamide (PTZ0.14 h),
(2) 2-cyclohexyl-N,N-diethylacetamide (PTZ0.24 h), (3) N,N-diethyl-2-(3-bromo-phenyl)-acetamide
(PTZ0.63 h), and (4) N,N-diethyl-3-trifluromethyl-benzamide (PTZ0.5 h). All of these compounds
map the pharmacophore in varying degrees (Figure 10.7a through d).
   To further examine the validity of the pharmacophore, it was mapped on a compound recently reported
in the literature, a novel 18-carbon acid, isolated from samples of greasy gaur hair, that was found to
have insect-repellent activity and can function as a landing and feeding deterrent to mosquitoes (see
Chapter 3).69 Surprisingly, the pharmacophore mapped extremely well on this molecule, proving the
consistency in the predictive power for insect repellent activity of the model (Figure 10.7e).61
   The pharmacophore allowed the authors to screen the in-house WRAIR-Chemical Information System
(WRAIR-CIS) database70 to search for candidate arthropod repellent compounds. The WRAIR-CIS
database has over 290,000 compounds; it was transformed into a multi-conformer database in
CATALYST using the catDB utility program as implemented in the software. 62 The catDB
format allows a molecule to be represented by a limited set of conformations, thereby permitting
conformational flexibility to be included during the search of the database. The authors have utilized the
best fit mapping of the pharmacophore in potent analogues by using a fast-fit algorithm, a principle
component analysis, a partial least squares technique, a linear regression technique, or a non-linear
regression technique.


q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling                                   215




FIGURE 10.6 (See color insert following page 204.) Pharmacophore mapping onto (a) Compd. 1, deet (a highly potent
repellent) and (b) Compd. 9 (a less potent repellent). (From A. K. Bhattacharjee, W. Dheranetra, D. A. Nichols, and
R. K. Gupta, QSAR and Combinatorial Science, 24, 593, 2005.)

q 2006 by Taylor & Francis Group, LLC
216                                                           Insect Repellents: Principles, Methods, and Uses




FIGURE 10.7 (See color insert following page 204.) Pharmacophore mapping on other known repellents showing the
cross-validation of the pharmacophore. Protection time of these agents are shown in the parentheses: (a) N,N-diethyl-2-
(3-trifluoromethyl-phenyl)-acetamide (PT Z 0.14 h), (b) 2-cyclohexyl-N,N-diethylacetamide (PT Z 0.24 h), (c) N,N-
diethyl-2-(3-bromo-phenyl)-acetamide (PT Z 0.63 h), (d) N,N-diethyl-3-trifluoromethyl-benzamide (PT Z 0.5 h), and
(e) onto 5-[5-(1-hydroxy-nonyl)-tetrahydro-furan-2-yl]-pentanoic acid. (From A. K. Bhattacharjee, W. Dheranetra, D. A.
Nichols, and R. K. Gupta, QSAR and Combinatorial Science, 24, 593, 2005.)

q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling   217




FIGURE 10.7 Continued.




q 2006 by Taylor & Francis Group, LLC
218                                                     Insect Repellents: Principles, Methods, and Uses




FIGURE 10.7       Continued.



   The pharmacophore was finally converted into a 3D shape-based template containing all the chemical
features necessary for potent arthropod repellent activity and used for WRAIR-CIS database searches.
After each compound in the WRAIR-CIS was converted into 3D multi-conformations, with an energy
range of 0–20 kcal/mol using the catDB algorithm of CATALYST, the full data set was stored in an SGI
Octane workstation. The result of the search led us to identify 138 compounds for repellent activity. The
down selection of the identified compounds was carried out by evaluating the in silico ADME/Toxicity
properties and choosing only those compounds that had favorable properties. ADME/Toxicity
evaluations were carried out by using Cerius2 and TOPKAT methodology,71,72 as implemented in
these software applications.
   The overall procedure of compound identification and selection was carried out in an iterative manner
by generating several shape-based pharmacophore templates on a few potent repellent compounds.
Ultimately, it was possible to shortlist four compounds (Figure 10.8 and Figure 10.11) that were found to
exhibit remarkable repellent activity, fulfilling the important goals of an ideal repellent. One of these four
compounds, 2-methyl-1-(2,3,5,6-tetramethyl-phenyl)-propan-1-one appears to fulfill most of the goals
for developing an ideal repellent. The four compounds are presented with the protection time in the
parentheses of each of them: (1) 2-bromo-1-(2,5-dimethoxy-phenyl)-ethanone (PTZ2.6 h);
(2) 2-methyl-1-(2,3,5,6-tetramethyl-phenyl)-propan-1-one (PTZ9.3 h); (3) 2-allylsulfanyl-3-methyl-
pyrazine (PTZ1.6 h); and (4) 2-(2-chloro-phenoxy)-2-methyl-propionamide (PTZ2.65 h). Mappings
of the pharmacophore on these four compounds are shown in Figure 10.8a through d. Although all of the
compounds possess outstanding in vitro arthropod repellent activity and have reasonably well-tolerated
properties for promising repellent candidates, the compounds have yet to be tested for in vivo efficacy
and toxicity.
   Thus, the pharmacophore model for repellent activity allowed the authors to successfully search for
compounds in databases and identify four lead repellent candidates that are currently under further
investigation. Although no model is perfect, regardless of what it represents, virtual screening of


q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling                                         219




FIGURE 10.8 (See color insert following page 204.) Pharmacophore mapped onto four new insect repellents discovered
through database searches by using the pharmacophore: (a) 2-bromo-1-(2,5-dimethyoxy-phenyl)-ethanone, (b) onto 2-
methyl-1-(2,3,5,6-tetrahmethyl-phenyl)-propan-1-one, (c) onto 2-allylsufanyl-3-methyl-pyrazine, and (d) onto 2-(2-chloro-
phenoxy)-2-methyl-propionamide. (From A. K. Bhattacharjee, W. Dheranetra, D. A. Nichols, and R. K. Gupta, QSAR and
Combinatorial Science, 24, 593, 2005.)


q 2006 by Taylor & Francis Group, LLC
220                                     Insect Repellents: Principles, Methods, and Uses




FIGURE 10.8       Continued.




q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling                                    221




                                                                      N
                               O                                               O

                                   N

                                                                               O



                           1                                              2




                                        O       N

                                                                               N
                                                O
                                                                                     OH



                                            3                                       4

FIGURE 10.9 Structure of compounds in the training set. (From A. K. Bhattacharjee, W. Dheranetra, D. A. Nichols, and
R. K. Gupta, QSAR and Combinatorial Science, 24, 593, 2005.)


hundreds of compounds from databases to identify potential hits in a relatively short period of time has
opened a new dimension for the search for new arthropod repellent candidates. Even though it is a
complex, expensive, and time-consuming path to develop a perfect repellent from the discovery stage to
the shelf for over-the-counter sale, scientists are continuing to explore new strategies to efficiently
minimize the amount of effort required and to translate effectively the potent intrinsic physico-chemical
characteristics of the compounds into new candidates with superior properties for suitable skin
application to protect against arthropod biting.



Concluding Remarks and Future Perspectives
The stereoelectronic properties, similarity analysis, and the 3D pharmacophore models in the above
studies could satisfactorily explain the insect repellent properties of the compounds. The pharmacophore
model made it possible to search compound databases to identify new repellent candidates.
   The first investigation on the electronic properties of 31 repellents suggests that the properties
of the amide group (N–CaO atoms) in these compounds play a key role in determining the duration of
the protection against mosquito bites. The substituents attached to carbon and nitrogen atoms of the amide
group together influence the electronic properties of the amide group. Thus, a balance of polarity between
the two parts of the molecule seems to be an important contributing factor for potent repellent activity.


q 2006 by Taylor & Francis Group, LLC
222                                                          Insect Repellents: Principles, Methods, and Uses




                                                N



                                                O                           HO

                                    O



                                                                    HO
                                    5                                       6




                                             N

                                    O   N
                                                     N
                                                         O                O             Cl

                                             O
                                                                      N



                                            7                                       8




                                        Cl

                              ClO                   Cl

                                                              HO                O

                               Cl                   Cl
                                        Cl                                      O



                                        9                                                    10

FIGURE 10.9       Continued.

  The investigation of the comparison of stereoelectronic properties of deet compounds with JH-mimic
and JH unraveled a few important facts about these compounds at a molecular level, not only providing a
better insight into the mechanism of action of the deet repellents, but also facilitating the design of more
efficacious deet-like compounds. The results of the study indicate a model for similar molecular
recognition of the deet compounds and the JH-mimic where the three crucial factors appear to be:

      † Considerable steric similarity between the amide moiety of the compounds
      † Similarity of electrostatic properties and profiles beyond the van der Waals surface
      † Similarity of a large distribution of a weak electrostatic field on the van der Waals surface


q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling                            223




                                                    O
                                         O




                                             HO

                                                               O
                                                    O




                                                                       11

FIGURE 10.9 Continued.


   Although electrostatic similarity beyond the van der Waals surface is considered to be the primary
index for the molecular recognition between compounds, the similarity of the steric components,
Mulliken charges, and negative potential adjacent to the oxygen and nitrogen atoms of the amide moiety
may all contribute significantly to the overall mechanism of repellent action of the deet compounds. On a
molecular level, the repellent action of the deet compounds may be attributed to avoiding a host-guest
complementarity conflict with the receptor.
   The stereoelectronic property study provides three important guidelines to effectively design this class
of insect repellents. Specifically, there needs to be:

     † An amide moiety on one end of the molecule that contains a charge separation between the
       oxygen and nitrogen atoms to facilitate a strong electronic interaction with the receptor
     † Electrostatic similarity to JH or its mimic molecules
     † A large, weakly charged region to facilitate optimum hydrophobic interaction with the receptor
       that may be a long-chain hydrocarbon and need not be an aromatic ring

   Thus, the study has illustrated: (1) the electrostatic bioisosterism of juvenile hormone, its mimic, and
the repellents; (2) a probable mode of action of insect repellent activity; and (3) additional
stereoelectronic features that may be added to the previous report on the design of insect repellents.62
However, possible pitfalls for designing potent repellents based on these criteria may not be ruled out if
the designed molecules are either too volatile or insufficiently volatile with R groups too bulky for fitting
into the receptor site.
   The 3D-QSAR pharmacophore study on repellents demonstrated a new computational approach for
organizing the molecular characteristics of a set of structurally diverse arthropod repellents to a model
that may be both statistically and mechanistically significant for potent repellent activity and may have
applicability beyond the bounds of known repellents. The resulting model can also be used to unravel a
possible rationale for the target-specific arthropod repellent activity of these compounds. The chemically
significant molecular characteristics disposed on a three-dimensional space generated a pharmacophore
that is found to be quite satisfactory in correlating experimental repellent activity with the predicted


q 2006 by Taylor & Francis Group, LLC
224                                                                     Insect Repellents: Principles, Methods, and Uses



                                                        F
                                                                F
                       N                                                                                    N
                                                            F
                                                                                                O
                             O

                                                                               2-Cyclohexyl-N,N-diethylacetamide
             N,N-Diethyl-2-(3-trifluoromethyl-phenyl)-acetamide
                                                                                          (PT = 0.24 hr)
                                 (PT = 0.14 hr)
                                                                                      O                     F
                                                                                                                    F

                                               B            r                    N
                         N
                                                                                                                F

                              O

                N,N-Diethyl-2-(3-bromo-phenyl)-acetamide                   N,N-Diethyl-3-trifluromethyl-benzamide
                          (PT = 0.63 hr)                                              (PT = 0.5 hr)




                                          HO                        O    OH

                                                   O

                                        5-[5-(1-Hydroxynonyl)tetrahydrofuran-2-yl]pentanoic acid

FIGURE 10.10 Structure of known insect repellents used for validating the pharmacophore. (From A. K. Bhattacharjee,
W. Dheranetra, D. A. Nichols, and R. K. Gupta, QSAR and Combinatorial Science, 24, 593, 2005.)



                                          O
                             O                     Br                                               N
                                                                                                            N
                                                                                                S
                                          O


                  2-Bromo-2',5'-dimethoxyacetophenone
                                                                              2-Allylsulfanyl-3-methyl-pyrazine
                                (PT=2.6hr)
                                                                                          (PT=1.6hr)



                                                                                           O            O

                                                                                                        NH2
                                          O                                                Cl


                       2-Methyl-1-(2,3,5,6-tetramethyl                         2-(2-Chlorophenoxy)-2-methyl
                           phenyl)propan-1-one                                         propionamide
                                    (PT=9.3hr)                                            (PT=2.7hr)

FIGURE 10.11 Structure of four new insect repellents discovered using the pharmacophore. (From A. K. Bhattacharjee,
W. Dheranetra, D. A. Nichols, and R. K. Gupta, QSAR and Combinatorial Science, 24, 593, 2005.)


q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling                            225


activity of the compounds (RZ0.9). Potent repellent activity appears to be favored by two aliphatic
hydrophobic functions, one aromatic hydrophobic function (aromatic ring) and one hydrogen bond
acceptor function in specific geometric locations surrounding the molecular space.
   The validity of the pharmacophore, which extends to structurally different classes of compounds,
allowed us to discover new repellent candidates, and thereby provides a powerful template for
identification of novel arthropod repellent candidates. Because the identity of the biological target
for arthropod repellent activity remains unknown, this 3D-QSAR pharmacophore should aid in the
design of well-tolerated, target-specific arthropod repellent active ingredients. The success of discovery
of new repellent candidates in this study suggests that the 3D-QSAR studies on repellents cannot only
facilitate the examination of databases to identify new candidates, but also could be a great benefit in
synthetic efforts to discover better repellents for practical use.
   Although the process of arthropod repellent discovery and development is a long and continuous
endeavor, in silico technologies can undoubtedly help in reducing the rapidly increasing costs of
developing new active ingredients. Molecular modeling techniques using in silico tools are uniquely
suitable for integrating new knowledge about molecular structure with new knowledge about
repellent activity.

Acknowledgments
Material has been reviewed by the Walter Reed Army Institute of Research. There is no objection to its
presentation and/or publication. The opinions or assertions contained herein are the private views of the
author, and are not to be construed as official, or as reflecting true views of the Department of the Army or
the Department of Defense. Research was conducted in compliance with the Animal Welfare Act and
other federal statutes and regulations relating to animals and experiments involving animals and adheres
to principles stated in the Guide for the Care and Use of Laboratory Animals, NRC Publication,
1996 edition.
  We also wish to express our profound thanks to Ms. Linette Sparacino, from ANTEON, Ft. Detrick,
MD 21702-5012, for reading and providing invaluable suggestions for improving the manuscript.

References
     1. P. Buchwald and N. Bodor, Computer-aided drug design: The role of quantitative structure-property,
        structure-activity and structure-metabolism relationships (QSPR, QSAR, QSMR), Drug Future, 27,
        577, 2002.
     2. D. Janseen, The power of prediction, Drug Discovery and Development, 5(1), 38, 2002.
     3. B. L. Podlogar, I. Muegge, and L. J. Brice, Computational methods to estimate drug development
        parameters, Current Opinion in Drug Discovery and Development, 12, 102, 2001.
     4. O. Guner (Ed.), Pharmacophore Perception, Development and Use in Drug Design, La Jolla, CA:
        International University Line Biotechnology Series, 2000.
     5. R. K. D. Peterson, Insects, disease, and military history, American Entomologist, 41, 147, 1995.
     6. R. F. Harwood and M. T. James (Eds.), Entomology in human and animal health, 7th Ed., New York:
        MacMillan, 1979.
     7. B. F. Eldridge and J. E. Edman (Eds.), Medical Entomology: A Textbook on Public Health and
        Veterinary Problems Caused by Arthropods, Boston: Kluwer Academic Publishers, 2000.
     8. A. N. Clements, Development, nutrition and reproduction, The Biology of Mosquitoes. Vol. 1, London:
        Chapman and Hall, 1992.
     9. J. Goddard, Physician’s Guide to Arthropods of Medical Importance, 2nd Ed., Boca Raton, FL: CRC
        Press, 1996.
    10. M. W. Service, Medical Entomology for Students, London: Chapman and Hall, 1996.
    11. M. Bates, The Natural History of Mosquitoes, New York: MacMillan, 1949.
    12. S. E. Christophers, Aedes Aegypti (L.) The Yellow Fever Mosquito: Its Life History, Bionomics, and
        Structure, London: Cambridge University Press, 1960.


q 2006 by Taylor & Francis Group, LLC
226                                                        Insect Repellents: Principles, Methods, and Uses


    13. A. N. Clements, The Physiology of Mosquitoes, Oxford: Pergamon Press, 1963.
    14. J. D. Gillett, Mosquitoes, London: Weidendield and Nicolson, 1971.
    15. J. D. Gillett, The Mosquito: Its Life, Activities, and Impact on Human Affairs, Garden City, NY:
        Doubleday, 1972.
    16. W. R. Horsfall, Mosquitoes: Their Behavior and Relation to Disease, New York: Ronald Press, 1955.
    17. S. C. Oaks, V. S. Mitchell, G. W. Pearson, and C. C. J. Carpenter (Eds.), Malaria: Obstacles and
        Opportunities, Washington, DC: Institute of Medicine, National Academy Press, 1991.
    18. World Health Organization, World malaria situation in 1992, Weekly Epidemiological Record, 69,
        309, 1994.
    19. R. K. Gupta et al., Personal protection measures against arthropods, in Textbook of Military Medicine,
        Vol. 1, D. E. Lounsbury (Ed.), Washington, D.C.: Borden Institute, Walter Reed Army Medical
        Center, 2003, p. 503.
    20. E. T. McCabe, W. F. Barthel, S. I. Gertler, and S. A. Hall, Insect repellents. III. N,N-diethylamides,
        Journal Organic Chemistry, 19, 493, 1954.
    21. I. H. Gilbert, H. K. Gouck, and C. N. Smith, New mosquito repellents, Journal of Economic
        Entomology, 48, 741, 1955.
    22. R. K. Gupta and L. C. Rutledge, Role of repellents in vector control and disease prevention, American
        Journal of Tropical Medicine and Hygiene, 50(6), 82, 1994.
    23. M. S. Fradin, Mosquitoes and mosquito repellents: A clinician’s guide, Annals of Internal Medicine,
        128, 931, 1998.
    24. L. A. Thomas, Distribution of the virus of western equine encephalomyelitis in the mosquito vector,
        Culex tarsalis, American Journal of Hygiene, 78, 150, 1963.
    25. A. Nakayama, H. Iwamura, A. Niwa, Y. Nakagawa, and T. Fujita, Development of insect juvenile
        hormone active oxime o-ethers and carbamates, Journal of Agricultural and Food Chemistry, 33,
        1034, 1985.
    26. A. Nakayama and W. G. Richards, A quantum chemical study of insect juvenile hormone mimics: The
        active conformation and the electrostatic similarities, Quantitative Structure-Activity Relationships, 6,
        153, 1987.
    27. A. K. Bhattacharjee, R. K. Gupta, D. Ma, and J. M. Karle, Molecular similarity analysis between insect
        juvenile hormone and N,N-diethyl-m-toluamide (deet) analogs may aid design of novel insect
        repellents, Journal of Molecular Recognition, 13, 213, 2000.
                                                                        ¨
    28. J. Boeckh, H. Breer, M. Geier, F. P. Hoever, and B. W. Kruger, Acylated 1,3-aminopropanola as
        repellents against bloodsucking arthropods, Pesticide Science, 48, 359, 1996.
    29. D. Mackay, J. Hubbarde, and E. Webster, The role of QSARs and fate models in chemical hazard and
        risk assessment, in Encyclopedia of Agrochemicals, J. Plimmer (Ed.), New York: Wiley-Interscience,
        2003.
    30. M. V. S. Suryanarayana, K. S. Pandey, S. Prakash, C. D. Raghuveeran, R. S. Dangi, R. V. Swamy, and
        K. M. Rao, Structure-activity relationship studies with mosquito repellent amides, Journal of
        Pharmacological Sciences, 80, 1055, 1991.
    31. R. K. Sharma, S. K. Jain, S. Kumar, and K. M. Rao, Evaluation of some insect repellent formulations.
        Part I—Water soluble ointment bases, Indian Journal Hospital Pharmacy, 21, 26, 1984.
    32. J. S. Murray, B. A. Zilles, K. Jayasuriya, and P. Politzer, Comparative analysis of the electrostatic
        potentials of dibenzofuran and some dibenzo-p-dioxins, Journal of the American Chemical Society,
        108, 915, 1986.
    33. D. Ma, K. Bhattacharjee, R. K. Gupta, and J. M. Karle, Predicting mosquito repellent potency of
        N,N-diethyl-m-toluamide (deet) analogs from molecular electronic properties, American Journal
        of Tropical Medicine and Hygiene, 60, 1, 1999.
    34. K. J. Judy, D. A. Schooley, L. L. Dunham, M. S. Hall, and B. J. Bergot, Isolation, structure, and
        absolute configuration of new natural insect juvenile hormone from Manduca sexta, Proceedings
        of the National Academy of Sciences, U.S.A., 70, 1509, 1973.
    35. B. J. Bergot, G. C. Jamieson, M. A. Ratcliff, and D. A. Schooley, JH zero: New naturally
        occurring insect juvenile hormone from developing embryos of the tobacco hornworm, Science,
        210, 336, 1980.


q 2006 by Taylor & Francis Group, LLC
Discovery and Design of New Repellents by Computer-Aided Molecular Modeling                                  227


    36. W. A. Skinner and H. L. Johnson, The design of insect repellents, Drug Design, 10, 277, 1980.
    37. H. L. Johnson et al., Repellent activity and physical properties of ring-substituted N,N-
        diethylbenzamides, Journal of Economic Entomology, 60, 173, 1967.
    38. S. B. McIver, A model for the mechanism of action of the repellent deet on Aedes aegypti
        (Diptera: Culicidae), Journal of Medical Entomology, 18, 357, 1981.
    39. S. S. Rao and K. M. Rao, Insect repellent N,N-diethylphenylacetamide: An update, Journal of
        Medical Entomology, 28, 303, 1991.
    40. E. Fischer et al., Einfluss der configuration auf die Wirkung der Enzyme, Berichte Deutsch
        Chemistry Gesellschaft, 27, 2985, 1894.
    41. H. Weinstein, M. N. Liebman, and C. A. Venazi, in New Methods in Drugs Research,
        A. Makriayannis (Ed.), Vol. I, Barcelona: Prous Science, 1985, p. 233.
    42. A. K. Bhattacharjee, D. Majumdar, and S. Guha, Theoretical studies on the conformational
        properties and pharmacophoric pattern of several bipyridine cardiotonics, Journal of the Chemical
        Society, Perkin Transactions, 2, 805, 1992.
    43. A. K. Bhattacharjee and J. M. Karle, Molecular electronic properties of a series of 4-quinolinecarbi-
        nolamines define antimalarial activity profile, Journal of Medicinal Chemistry, 39, 4622, 1996.
    44. J. Mestres, D. C. Rohrer, and G. M. Maggiora, MIMIC: A molecular-field matching program: exploiting
        applicability of molecular similarity approaches, Journal of Computational Chemistry, 18, 934, 1997.
    45. G. Naray-Szabo and T. Balogh, Viewpoint 7—the average molecular electrostatic field as a QSAR
        descriptor. Part 4. Hydrophobicity scales for amino acid residues, Journal of Molecular Structure:
        THEOCHEM, 284, 243, 1993.
    46. G. Naray-Szabo and T. Balogh, Application of the average molecular electrostatic field in quantitative
        structure-activity relationships, Croatica Chemica Acta, 66, 129, 1993.
    47. A. R. Leach, Molecular Modelling, Principles and Applications, Essex: A.W. Longman Ltd, 566, 1998.
    48. M. Randic and G. Krilov, On characterization of molecular surfaces, International Journal of Quantum
        Chemistry, 65, 1065, 1997.
    49. SPARTAN SGI Version 5.1.2., SPARTAN user’s manual, version 5.0. Wavefunction, Inc., Irvine, CA,
        1998.
    50. P. G. Mezey, Shape in Chemistry: An Introduction to Molecular Shape and Topology, New York: VCH,
        1993.
    51. J. D. Petke, Cumulative and discrete similarity analysis of electrostatic potentials and fields, Journal of
        Computational Chemistry, 14, 928, 1993.
    52. A. M. Richard, Quantitative comparison of molecular electrostatic potentials for structure-activity
        studies, Journal of Computational Chemistry, 12, 959, 1991.
    53. W. G. Richards, Quantum Pharmacology, 2nd (Ed.), London: Butterworth, 1983.
    54. A. C. Good, S. S. So, and W. G. Richards, Structure-activity relationships from molecular similarity
        matrices, Journal of Medicinal Chemistry, 36, 433, 1993.
    55. J. Avery, A model for biological specificity, International Journal of Quantum Chemistry, 26, 843, 1984.
    56. R. G. A. Bone and H. O. Villar, Discriminating D1 and D2 agonists with a hydrophobic similarity index,
        Journal of Molecular Graphing, 13, 201, 1995.
    57. R. G. Pearson, The principle of maximum hardness, Accounts of Chemical Research, 26, 250, 1993.
    58. G. R. Bock and G. Cardew (Eds.), Olfaction in Mosquito-Host Interactions, New York: Wiley, 1996.
    59. E. E. Davis and M. F. Bowen, Sensory physiological basis for attraction in mosquitoes, Journal of the
        American Mosquito Control Association, 10, 316, 1994.
    60. E. E. Davis, Insect repellents: Concepts of their mode of action relative to potential sensory mechanisms
        in mosquitoes (Diptera: Culicidae), Journal of Medical Entomology, 22, 237, 1985.
    61. A. K. Bhattacharjee, W. Dheranetra, D. A. Nichols, and R. K. Gupta, 3D pharmacophore model for insect
        repellent activity and discovery of new repellent candidates, QSAR and Combinatorial Science, 24, 593,
        2005.
    62. CATALYST Version 4.8 software, San Diego: Accelrys, Inc., 2003.
    63. M. Grigorov, J. Weber, J. M. J. Tronchet, C. W. Jefford, W. K. Milhous, and D. Maric, A QSAR study of
        the antimalarial activity of some synthetic 1,2,4-trioxanes, Journal of Chemical Information and
        Computer Science, 35, 285, 1995.


q 2006 by Taylor & Francis Group, LLC
228                                                     Insect Repellents: Principles, Methods, and Uses


    64. O. A. Gunner, in Pharmacophore, Perception, Development, and Use in Drug Design, University
        International Line (Ed.), San Diego: International University Line, 2000, pp. 17–20.
    65. P. A. Greenridge and J. Weiser, A comparison of methods for pharmacophore generation with the
        catalyst software and their use for 3D-QSAR: Application to a set of 4-aminopyridine thrombin
        inhibitors, Mini-Reviews in Medicinal Chemistry, 1, 79, 2001.
    66. M. Debboun, D. Strickman, T. A. Klein, J. A. Glass, E. Wylie, A. Laughinghouse, R. A. Wirtz, and
        R. K. Gupta, Laboratory evaluation of AI3-37220, AI3-35765, CIC-4, and deet repellents against three
        species of mosquitoes, Journal of the American Mosquito Control Association, 15, 342, 1999.
    67. R. K. Gupta and L. C. Rutledge, Controlled-release repellent formulations on human volunteers under
        three climatic regimens, Journal of the American Mosquito Control Association, 7, 490, 1991.
    68. A. Goldstein, Biostatistics: An Introductory Text, New York: MacMillan, 1964.
    69. J. E. Oliver and K. S. Patterson, Wild ox bugs mosquitoes, Chemical and Engineering News, 49, 2003.
    70. The Chemical Information System, Division of Experimental Therapeutics, Walter Reed Army Institute
        of Research, 503 Robert Grant Avenue, Silver Spring, MD 20910-7500, U.S.A.
    71. Cerius2 4.9, San Diego: Accelrys, Inc., 2003.
    72. TOPKAT 6.1, San Diego: Accelrys, Inc., 2003.




q 2006 by Taylor & Francis Group, LLC
11
Molecular-Based Chemical Prospecting of Mosquito
Attractants and Repellents


Walter S. Leal



CONTENTS
Molecular Basis of Insect Olfaction .............................................................................................229
  Choosing Functional Molecular Targets ...................................................................................234
Screening Techniques ...................................................................................................................235
  Receptor-Based Approach .........................................................................................................235
  Binding Assay-Based Approach................................................................................................236
  Validating Molecular Targets....................................................................................................237
Concluding Remarks .....................................................................................................................239
Acknowledgments .........................................................................................................................239
References .....................................................................................................................................240




Molecular Basis of Insect Olfaction
Most insects are primarily reliant on chemical communication to guide their essential behaviors. In
natural settings, female mosquitoes undoubtedly use airborne chemical signals (semiochemicals)
integrated with other sensory modalities to find and determine the suitability of hosts for blood
feeding, sites for oviposition, etc. Female moths, on the other hand, advertise their readiness to mate
and reproduce by releasing sex pheromones, which are utilized by male moths in odorant-mediated
navigation toward females. Reception of the semiochemicals by specialized structures in the periphery,
such as antennae and maxillary palps, is a sine qua non step prior to integration with other stimulus
modalities in the brain and subsequent translation into behavior. Insect communication, be it host-finding
in mosquitoes or mate-finding in moths, is a feasible target to disrupt important behaviors. While
chemical communication-based strategies for monitoring and controlling populations of insects of
agricultural importance have been extensively used in integrated pest management (IPM) programs,
similar approaches for reducing mosquito populations and contact between disease vector and host have
not been fully exploited.
   In general, chemical communication is achieved with airborne semiochemicals that are typically
available at very low concentrations and buried in complex mixtures of physiologically irrelevant
compounds. To cope with this, the olfactory system in insects evolved to be highly selective and


                                                                                                                                             229

q 2006 by Taylor & Francis Group, LLC
230                                                              Insect Repellents: Principles, Methods, and Uses


sensitive. To find a semiochemical source; such as blood meal, mate, oviposition site, etc.; insects take
odorant-mediated flights, which also require a dynamic process for odorant detection. While flying en
route to a source, insects encounter pockets of semiochemicals separated by clean air spaces. They have
only a few milliseconds to reset the olfactory system while navigating through clean air.1 Three major
groups of proteins play pivotal roles in the dynamics, selectivity, and sensitivity of the insect olfactory
system.2,3 They are the odorant receptors (ORs), odorant-binding protein (OBPs), and odorant-degrading
enzymes (ODEs), which are feasible molecular targets for the development of mosquito attractants and
novel strategies to reduce mosquito bites.
   Semiochemicals reach the aqueous sensillar lymph through pore tubules (Figure 11.1), but relative
solubility prevents these hydrophobic molecules from reaching the membrane-bound ORs. The
semiochemicals are hydrophobic and the sensillar lymph is an aqueous barrier.2,3 Biochemical and
structural evidence suggest that OBPs selectively bind the physiologically relevant chemical
compounds2,4,5 and solubilize them in the form of a ligand-protein complex.6 While encapsulated by




FIGURE 11.1 (See color insert following page 204.) Schematic representation of a proposed model for perireceptor
events in insect olfaction. Odorants enter the sensillar lymph through pore tubules in the cuticle (sensillar wall), are
solubilized upon being encapsulated by odorant-binding proteins (OBP), and transported to the olfactory receptors. Bound
pheromone molecules are protected from odorant-degrading enzymes (ODE). Upon interaction with negatively-charged sites
on the dendritic membrane, the OBP-ligand complex undergoes a conformational change that leads to the ejection of
pheromone. In BmorPBP, this is achieved by the formation of a C-terminal a-helix in BmorPBPA that blocks the cavity that
serves as the binding site in BmorPBPB. After releasing the odorant, the C-terminus may remain in the cavity of BmorPBPB
until another odorant is picked up.56 Note that in this model, the pheromone molecule (not the complex) activates the odorant
receptor. The signal is terminated by chemical inactivation of the odorant by an odorant-degrading enzyme (ODE).


q 2006 by Taylor & Francis Group, LLC
Molecular-Based Chemical Prospecting of Mosquito Attractants and Repellents                                                231


OBPs, semiochemicals are not only soluble, but also protected from aggressive ODEs. Then, OBPs carry
the odorants through the sensillar lymph to the ORs (Figure 11.1). Interaction with negatively charged
sites on the dendritic membranes4,7,8 leads to a unique intramolecular rearrangement of the ligand-
protein complex9 (Figure 11.1) resulting in the release of odorants. Stopped-flow fluorescence
measurements show that uptake of pheromones is a rapid process (in the timescale of milliseconds).
The release of the ligand would be a slow process (half life on the order of 100 s) if not for the
pH-mediated conformational change that speeds up the delivery of odorants by 10,000-fold.10
   Although biochemical and structural biology indicate that mosquito OBPs may not undergo the same
type of intramolecular rearrangement as moth OBPs, the delivery of odorants to the receptors is also
mediated by a pH-dependent conformational change with a different molecular mechanism. As
suggested by the crystal structure of an OBP from the malaria mosquito, Anopheles gambiae,
AgamOBP1,11 mosquito OBPs possess an overall fold of six helices connected by loops and knitted
together by three disulfide bridges (Figure 11.2; see also the cover of this book). Although the C-terminus
of AgamOBP1 is too short to form a helix that would occupy the binding pocket at low pH as in the
silkworm moth’s OBP, BmorPBP,9 it does form a wall of the binding pocket (Figure 11.2). The
C-terminus wall is held in place by acid labile hydrogen bonding involving the surrounding helices and
the N-terminus. The C-terminal carboxylate of valine, Val-125, are within hydrogen bonding distance of
the hydroxyl of tyrosine, Tyr-54, and of the d nitrogen of histidine, His-23. In addition, there are three
aspartic acid residues, Asp-7, Asp-42, and Asp-118, that interact with either arginine, Arg-5 and Arg-6,
histidine, His-121, or the backbone nitrogen of Tyr-10. These interactions are likely acid-labile and
would be disrupted at lower pH, causing both the C- and N-termini to separate.11 Unlike the formation of
a C-terminus helix that fits like a piston in the binding pocket of BmorPBP,9 the C-terminus of
AgamOPB1 might move away from the binding pocket.11 This pH-mediated “unbuckling of the seat
belt” would expose the ligand to the solvent and, consequently, lower binding affinity at low pH.
   OBPs contribute to the sensitivity of the olfactory system by increasing the capture of molecules
reaching the sensillar lymph. More importantly, OBPs participate in the selectivity of the olfactory
system as the conduit between the external environment and the receptors. The remarkable selectivity of
insect olfactory system is likely to be achieved by “layers of filters,” i.e., by the participation of
compartmentalized OBPs and olfactory receptors.3 As suggested by binding assays,4,5 OBPs transport
only a subset of compounds that reach the pore tubules. The odorant receptors, on the other hand, are
activated also by a subset of compounds, as indicated by studies in Drosophila, showing that a single OR




FIGURE 11.2 (See color insert following page 204.) Three-dimensional structure of AgamOBP1. In the left, three
disulfide bridges that knit together the scaffold of a-helices are highlighted in cyan. In the right, the acid-labile hydrogen
bonding involving the C-terminal carboxylate of Val-125 with the hydroxyl of Tyr-54 and the d nitrogen of His-23 are shown
in blue. Although the C-terminus (green) is flexible, Tyr-54 and His-23 are in rigid positions as part of a helices held in place
by disulfide bridges.


q 2006 by Taylor & Francis Group, LLC
232                                                     Insect Repellents: Principles, Methods, and Uses


is fired in response to multiple compounds.12 Even if neither OBPs nor ORs are extremely specific, the
detection of semiochemicals at the periphery (antennae or maxillary palps) can show remarkable
selectivity if they function as a two-step filter with only one or very few common ligands. It is worth
mentioning that OBPs may seem to be selectively less stringent when tested in non competitive binding
assays. Under physiological conditions, however, OBPs encounter complex mixtures of compounds
bombarding the sensilla. Therefore, filtering by OBPs may be achieved by selective binding to the key
stimulus or kinetic competition.
   Semiochemical-OBP interactions are better understood in moths than in mosquitoes. The main
pheromone-binding protein (PBP is an OBP involved in the reception of a pheromone) from the wild
silkmoth, Antheraea polyphemus, ApolPBP1, shows apparent high affinity to all three constituents of the
female-produced sex pheromones: (E,Z)-6,11-hexadecadienyl acetate (E6,Z11-16Ac), (E,Z)-6,11-
hexadecadienal (E6,Z11-16Ald), and (E,Z)-4,9-tetradecadienyl acetate (E4,Z9-14Ac). However,
ApolPBP1 shows considerable preference for the major constituent, E6,Z11-16Ac, shows lower affinity
for the shorter acetate, E4,Z9-14Ac, and no affinity for the aldehyde, E6,Z11-16Ald, when the protein is
incubated with equal amounts of the three sex pheromones.5
   Earlier experiments based on electroantennogram (EAG) and single sensillum recordings (SSR)
highlighted the extraordinary specificity and sensitivity of the insect olfactory system. It has been clearly
demonstrated that each olfactory receptor neuron in a sensillum is highly tuned to a key stimulus (e.g., a
pheromone constituent) such that minimal structural modification to a pheromone molecule renders it
inactive.13 The large number of sensilla distributed over the surface of the antennae and maxillary palps
most likely contributes to the sensitivity of the insect olfactory system, but selectivity is mediated by
molecular recognition at the periphery. Selectivity does not have to rely entirely on the odorant receptors,
if odorant-binding proteins filter out some of the potential receptor ligands.
   Insects have evolved molecular mechanisms for the rapid inactivation or deactivation of chemical
signals. To clear up the ORs and avoid continuous activation by “stray” semiochemicals, the signal
from molecules that have already been conveyed must be terminated immediately. It has been
suggested that the process is so rapid that it requires a hitherto unknown molecular mechanism for
trapping the signal-carrying molecules (odorants).14 On the other hand, it has also been demonstrated
that inactivation can be accomplished by antennae-specific odorant-degrading enzymes.15,16 Indeed,
inhibition of a pheromone-degrading enzyme in vivo led to the complete desensitization of highly
sensitive, pheromone-specific olfactory receptor neurons in male antennae of a scarab beetle.17
It seems that localized low-pH environments generated by negatively charged surfaces on the
dendrites are also essential to prevent “premature inactivation” of odorants. As demonstrated with
ApolPDE, the pheromone-degrading enzyme of the wild silkmoth, Antheraea polyphemus, ODEs are
fast at the bulk pH of sensillar lymph, but sluggish at the low pH environments where odorant
“undocking” takes place.16 The generalization of this finding must await further experiments given the
diverse nature of insect ODEs.
   Because no odorant comes to the ORs except through OBPs, functional OBPs can be utilized as
molecular targets for the screening of mosquito attractants and repellents in an approach similar to
receptor-based drug discovery. While we have gained a better understanding of the molecular basis of
attractant reception. The mode of action of deet and other mosquito repellents is not yet known. It is
known, however, that deet has an olfactory-based repellent effect18,19 as well as feeding-deterrent
effect.19 In addition, it has been demonstrated by single sensillum recordings,20 gas chromatography
coupled to antennographic detection21 (Figure 11.3), and electroantennograms22 that mosquito possess
deet-detecting olfactory receptor neurons. These findings suggest that deet may be bound and transported
by OBPs as has been shown for other odorants. Therefore, mosquito OBPs are likely suitable molecular
targets for the development of repellents.
   Because ODE inhibitors desensitize the olfactory system,17 an untapped strategy could exploit the
design of anosmia-inducing ODE inhibitors to reduce mosquito bites. To inhibit ODEs in vivo, a
compound must be assisted by odorant-binding protein to penetrate the sensillar lymph and reach the
molecular target. Environmentally safe ODE inhibitors could be designed by a rational approach


q 2006 by Taylor & Francis Group, LLC
Molecular-Based Chemical Prospecting of Mosquito Attractants and Repellents                                            233




FIGURE 11.3 Separation of a sample of deet by gas chromatography with simultaneous recording with a flame ionization
detector (upper trace) and a mosquito antennae-based biosensor (lower trace). Antennae of blood-fed female Culex pipiens
pallens were used for electroantennography (lower trace). The sample (10 mg) was injected in splitless mode and separated
on a HP-5MS capillary column (30 m!0.25 mm; 0.25 mm; Agilent Technologies, Palo Alto, CA) that was operated at 1008C
for 1 min, increased to 2708C at a rate of 158C/min, and held at this temperature for 10 min. Deet appeared at 8.43 min with
an unambiguous EAD response. Smaller peaks at 7.4, 7.6, and 9.3 min are EAD-inactive chemical impurities.



q 2006 by Taylor & Francis Group, LLC
234                                                     Insect Repellents: Principles, Methods, and Uses


considering structural features of the binding cavities of OBPs and catalytic sites of ODEs. By decreasing
mosquito bites with novel and user-friendly repellents, disease transmission could also be reduced.


Choosing Functional Molecular Targets
The number of functional OBPs that exist for a single insect species is unknown, but to date only one
pheromone-binding protein23,24 (and a few general odorant-binding proteins) have been identified from
the silkworm moth, Bombyx mori; whereas the malaria mosquito, Anopheles gambiae, for example, has
potentially as many as 55 OBPs.25 (Initially, 57 were suggested, but Biesmann and collaborators26 found
that OBP34 and OBP37 genes encode the same OBP and the proteins predicted from OBP35 and OBP36
are identical.) This huge discrepancy in number of OBPs per species may be related to the method of
“identification” of OBPs. Protein-based approaches are aimed at the isolation and identification of OBPs,
followed by the cloning of the genes (or cDNAs) encoding these proteins. On the other hand, the gene-
based approaches provide little data on expression and functions of proteins. While minor OBPs may be
expressed at levels below the detection limits of the protein-based methods, the gene-based approach
may lead to putative proteins which may not even be expressed in the sensillar lymph of insect antennae
or maxillary palps. Even if a single OBP is involved in the detection of multiple compounds, one would
expect that the insect antennae possess multiple OBPs since insects can detect a number of
physiologically relevant compounds with diverse chemical structures, derived from conspecifics
(pheromones), hosts, or potential ovipostion sites. However, it is highly unlikely that all OBPs predicted
from an insect’s genome are indeed olfactory proteins. As an example, it has been suggested that the
OBP-gene family of Drosophila melanogaster comprises as many as 51 putative OBPs,27 but only seven
of them have been demonstrated to be expressed specifically in olfactory organs of adults (antennae only
or antennae and maxillary palps): Obp19a, Obp57a, Obp69a (formerly named PBPRP-1), Obp83a
(PBPRP-3, OS-F), Obp83b (OS-E), Obp84a (PBPRP-4), and Obp99d. Two other putative OBPs namely,
Obp28a (PBPRP-5) and Obp76a (LUSH), are detected in the antennae of adults as well as in larval
chemosensory organs.28 The same is true for mosquito “OBPs” whose genes are more broadly
expressed.29 For example, out of 20 OBP genes, Li et al.29 found three genes expressed in all tissues
and three that are either expressed at low level or not expressed at all in adults. Biessmann and
collaborators26 employed microarray and real-time quantitative RT-PCR in an attempt to obtain a better
understanding of the expression patterns of the genes possibly involved in reception of host odorants in
females of Anopheles gambiae. Twenty-four “typical” OBP genes were detected above background
levels, some with higher expression levels in female or male antennae; whereas, others were detected in
antennae but not in maxillary palps, and others in both olfactory tissues. Conversely, RT-PCR analysis
showed that genes suggested by microarray analysis to be expressed predominantly in antennae were also
expressed in nonolfactory tissues, as well as in mosquito larvae.26
   Olfactory and nonolfactory proteins from the OBP-gene family appear to belong to the same structural
family. Their helix-rich structures suggest that these proteins encapsulate hydrophobic odorants and
other ligands, with the ability to transport them in aqueous environments.2 Therefore, proteins of this
group should be named encapsulins to imply the common role of encapsulating small ligands.2 The
labeling “OBP” then should be restricted to olfactory odorant-binding proteins. It is possible that a large
number of genes annotated from insect genomes as putative OBPs are merely encapsulins. One of the
criteria widely utilized to annotate/identify putative OBPs is the occurrence of six well-conserved
cysteine residues. The spacing patterns are also structurally important. While the six-cysteine pattern is a
hallmark for most moth OBPs identified to date, it is not limited to OBPs. Insect defensins, for example,
share the same feature. Conversely, not all odorant-binding proteins and antennae-specific proteins
(putative OBPs) possess only six cysteine residues. For example, pheromone-binding proteins with
seven-cysteine residues have been isolated from a silkmoth, Samia cynthia ricini (Leal, unpublished
data) and predicted from the genome of Drosophila27 and Anopheles gambiae.30 We have also isolated
and cloned an eight-cysteine, antennae-specific, putative OBP from the yellow fever mosquito,


q 2006 by Taylor & Francis Group, LLC
Molecular-Based Chemical Prospecting of Mosquito Attractants and Repellents                             235


Aedes aegypti.31 On the other hand, it is unlikely that putative OBPs with 12-cysteine residues deduced
from the genomes of Drosophila27 (Obp58b, Obp58c, Obp58d, Obp83c, Obp93a) and Anopheles
gambiae30 would bind, transport, and release ligands in the same way as described for pheromone-
binding proteins.
   When using olfactory proteins as molecular targets in studies aimed at the design of potential mosquito
attractants or repellents, it is essential to focus on functional OBPs. A solid literature on pheromones and
sensory physiology has laid the foundation, in that it has revealed that moth pheromone-binding proteins
are expressed specifically in male antennae and are restricted to pheromone-detecting sensilla. For
example, long sensilla trichodea are present in both male and female antennae of the silkworm moth,
Bombyx mori. In males, these sensilla respond to bombykol and bombykal32,33; whereas, in females they
respond to benzoic acid and linalool.34 By immunolocalization of different OBPs with specific antisera,
Steinbrecht and collaborators showed that BmorPBP is expressed in pheromone-detecting sensilla
trichodea.35 In contrast, general odorant-binding proteins are detected in most sensilla basiconica,36 a
detector for plant-derived compounds. In addition, the female long sensilla trichodea, morphologically
identical to the male pheromone detectors, express a general odorant-binding protein37 and detects only
nonpheromonal compounds.
   Mapping of mosquito sensilla on antennae and maxillary palps, and the identification of all
physiologically relevant ligands (semiochemicals) are yet to be completed. Thus, mosquito OBPs can
not be identified on the basis of the same gold standards as for moth OBPs. However, it is reasonable to
assume that OBPs expressed in olfactory (antennae and/or maxillary palps), and not in nonolfactory
tissues, are functional proteins (olfactory OBPs). In a recent structural study (see above),11 we focused on
an OBP from the malaria mosquito (AgamOBP1), which is expressed in antennae, but not in legs (control
tissue) (Figure 11.4). Following the protocol that led to the isolation of the first OBPs from mosquito
species38,39 we isolated four antennae-specific proteins from Anopheles gambiae (Mopti strain)
(Figure 11.4). MALDI-TOF (matrix-assisted laser desorption/ionization-time of flight) mass spectrometry
and tandem (LC-MS-MS) mass spectral analysis of the protein bands (Ag1–4) led to isolation of OBPs
whose genes have been previously identified40 and one putative odorant-degrading enzyme. Therefore, we
focused on the main antennae-specific protein, AgamOBP1, in our structural biology studies (see above).




Screening Techniques
Prospecting for novel mosquito attractants or repellents can be based on molecular interactions of
candidate compounds with olfactory proteins. Regardless of having odorant receptors or odorant-binding
proteins as molecular targets, these screening techniques do not completely replace behavioral or field
studies. However, large numbers of test compounds can be eliminated if they can not be transported to
odorant receptors or do not activate these receptors. Provided that the appropriate molecular targets have
been selected, these strategies can shorten the list of test compounds for further in-depth evaluations.


Receptor-Based Approach
In principle, the screening of potential mosquito attractants and repellents can be based on receptor-
ligand interactions. This could be achieved either by heterologous expression of putative odorant
receptors or in vitro binding assays. The latter is technically challenging, but the former could be
performed, for example, with candidate receptors expressed in Xenopus oocytes.12 Another promising
avenue is the expression of target ORs in a mutant of Drosophila containing an “empty neuron.” The
odorant receptors Or22a and Or22b of Drosophila were shown to be co-expressed specifically in the
ab3A antennal neuron41 and a mutant (Dhalo) lacking these genes has been utilized for functional
analysis of odorant receptors of Drosophila.42 Two odorant receptors from the malaria mosquito,


q 2006 by Taylor & Francis Group, LLC
236                                                          Insect Repellents: Principles, Methods, and Uses


                                            ANT                            LEG




                                   Ag4




                                   Ag3


                                  Ag2

                                   Ag1




FIGURE 11.4 Gel electrophoresis (15% native) analysis of female antennae-specific proteins from Anopheles gambiae.
Antennal (ANT; 700 antennae) and leg (100 hindlegs) extracts from 7–10 day-old Anopheles gambiae females (Mopti). The
migration of a moth OBP (BmorPBP) is indicated by a bar above Ag2. Ag1 was identified as AgamOBP1.


An. gambiae, have been expressed in the Dhalo mutant, with the response of the ab3A neuron of
Drosophila being analyzed by single sensillum recordings.43 Based on the response of AgOr1 to
4-methylphenol and AgOr2 to 2-methylphenol, these then putative odorant receptors were identified as
the first odorant receptor genes from the malaria mosquito.43
    It is possible that mutants of Drosophila engineered with malaria mosquito odorant receptors could be
used to screen for other potential ligands of the AgOR1 and AgOR2 receptors. It is not certain, however,
if the OBPs compartmentalized in the sensillum-housing the ab3A neuron are necessary and sufficient to
mimic the olfactory system of the malaria mosquito. Further experiments may clarify if host-finding-
related receptors in the mosquito will be functional in an engineered mutant of Drosophila.


Binding Assay-Based Approach
Binding of a semiochemical to an OBP can be investigated by incubation of a recombinant protein
and test ligands, with binding affinity being assessed, for example, by fluorescence, calorimetry, or by


q 2006 by Taylor & Francis Group, LLC
Molecular-Based Chemical Prospecting of Mosquito Attractants and Repellents                            237


measuring the amount of bound ligand. Intrinsic protein fluorescence can be very sensitive and
requires low amounts of protein and ligand,4 but signal-to-noise may be too small if there is no ligand-
induced conformational change leading to a change in the environment of tryptophan residues. This
difficulty can be overcome by employing a reporter group either as a ligand44 or by attaching a
fluorophore to the protein in a sensitive environment.45 The use of extrinsic reporter groups may
enhance sensitivity dramatically, but it may be difficult to determine if the labeling is not affecting the
normal function of the protein. This is particularly problematic for OBPs whose structures and cognate
ligands are not yet known, which is the case for mosquito OBPs. With a noncovalent fluorescent probe
the protein may be tested in its native conformation. These probes are normally nonfluorescent in
water, but their emission spectra are modified when bound to a protein normally undergoing a blue
shift with a marked increase in intensity. Binding of a test ligand to an OBP can be determined by
quenching the extrinsic fluorescence. The most widely used probes are 1-aminoanthracene (1-AMA)
and N-phenyl-1-naphthylamine (1-NPN), which were initially employed in binding experiments with
vertebrate OBPs.46
   Isothermal titration calorimetry (ITC) has been employed to demonstrate binding of 2-isobutyl-3-
methoxypyrazine to an OBP (ASP2) from the honeybee.47 Despite several attempts, we were unable to
measure binding of bombykol to BmorPBP by ITC (Leal, unpublished data). Preliminary attempts to
employ surface plasmon resonance (BIACOREw*) were also unrewarding, probably because of both the
low solubility and small size of the ligand (analyte in BIACORE jargon) (Leal, unpublished data).
   In early work to identify pheromone-binding proteins, radiolabeled pheromones were employed in
qualitative binding assays. With the availability of recombinant proteins, radiolabeled pheromones can
be used in quantitative assays in which free ligands are separated by gel filtration from bound ligands.48
Although a valuable tool for studies of pheromone-PBP interactions, this type of binding assay has
limited application in screening programs because radioactive test ligands are required. If a library of
radioactive test compounds were available, one might as well employ a high-throughput screening, such
as the scintillation proximity assay.49
   Recently, we have developed a low-throughput screening protocol named the cold binding assay10
(Figure 11.5) that does not require radioactive (hot) ligands. After incubation of test compound(s) with an
OBP, the free ligand is removed by filtration, whereas the protein bound ligand is retained in the
centrifugal device and extracted with an internal standard containing organic solvent. Binding is
quantified by gas chromatography and the identity of the recovered ligand is confirmed by gas
chromatography-mass spectrometry. As a negative control, binding is also investigated at low pH.
Among other advantages, this protocol allows competitive binding assays5 in which the best ligand can
be determined in a single assay.
   A promising strategy for online screens is the covalent immobilization of OBP to a liquid
chromatography stationary phase.50 We have tested this principle with two odorant-binding proteins,
the PBP from the silkworm moth, BmorPBP,23,24 and an OBP from Culex quinquefasciatus,
CquiOBP1.38 The BmorPBP column distinguished four compounds, with bombykol showing the
highest affinity, followed by bombykal, 1-hexadecanol, and (Z,E)-5,7-dodecadien-1-ol.50 Zonal
chromatographic studies using D-, L-, and D/L-lactic acid showed that the CquiOBP1 column separated
the two isomers of lactic acid, with L-lactic acid having higher affinity.


Validating Molecular Targets
The reverse chemical ecology approach described above has already been employed for the development
of better lures for the Navel Orangeworm moth, Amyelois transitella.51,52 Despite the tremendous effort
by leading scientists in the field of chemical ecology, only one constituent of the pheromone system of
this species was known until recently.53 With a multidisciplinary approach, including OBP-based

*
    A registered trademark of Biacore International AB, Uppsala, Sweden.


q 2006 by Taylor & Francis Group, LLC
238                                                             Insect Repellents: Principles, Methods, and Uses


                                                     Cold Binding Assay

                                1. Incubation                                         2. Separation




             A                      B                 C                           D                   E


                        3. Extraction                                       4. Chemical Analysis




                            F
                                                G                      H                       I


FIGURE 11.5 Schematic view of the four steps of a cold binding assay. A glass insert A deactivated by Silcote CL7
treatment (Kimble Chromatography, Vineland, NJ) is used to incubate protein and test ligand B. The reaction mixture is
shaken (100 rpm) at 25G28C for 1 h C. For separation of the bound and free ligands, the reaction mixture is transferred to a
washed Microcon YM-10 (Millipore) D and centrifuged (12,000!g, 48C) for 5 min E. The retentate is transferred to a 100 ml
V-vial (Wheaton, Millville, NJ) F along with a hexane containing an internal standard (eicosyl acetate, Fuji Flavor Co.,
Tokyo, Japan). The vial is capped G, vortexed for 1 min, and then centrifuged (2,500!g, 48C) for 5 min. The hexane fraction
(upper layer) is recovered H and analyzed by gas chromatography (GC) for quantification I. The extract can be analyzed by
gas chromatography-mass spectrometry (GC-MS) to confirm identification of the ligand extracted from the OBP-ligand
complex.


screening of potential attractants, we discovered a complex pheromone system.51,52 In addition, a
mosquito OBP-based screening program aimed at the development of oviposition attractants is underway
in my laboratory.
   The success of these programs depends heavily on the utilization of appropriate molecular targets, i.e.,
functional olfactory proteins. Our strategy is to select major antennae-specific olfactory proteins based on
the assumption that the most abundant OBPs in mosquito antennae play critical roles in chemical
communication as PBPs do in moths. Two lines of evidence substantiate this hypothesis. Based on the
binding of the mosquito oviposition pheromone (MOP)54 to an OBP, previously isolated from Culex
quinquefasciatus,38 we now have evidence that a major female antennae-specific protein in Culex
quinquefasciatus plays a critical role in insect olfaction (Leal et al., unpublished data). In addition,
sensilla in female antennae that are tuned to MOP house other olfactory receptor neurons sensitive to
chemical cues used for attraction to oviposition sites (Syed and Leal, unpublished data). A second line of
evidence comes from binding studies with an OBP previously isolated from the antennae of the yellow
fever mosquito, Aedes aegypti, AaegOBP1.31 Binding assays (Figure 11.6) indicate that at pH 7
AaegOBP1 bound nonanal, whereas no binding was found at low pH. Nonanal is the active ingredient


q 2006 by Taylor & Francis Group, LLC
Molecular-Based Chemical Prospecting of Mosquito Attractants and Repellents                                  239




                                                     15

                   Relative Binding (ng of ligand)


                                                     10




                                                      5




                                                      0
                                                          pH7   pH5       Buffer

FIGURE 11.6 Nonanal binds with high affinity to AaegOBP1 at pH 7. The amount of ligand recovered at low pH 5 is not
significantly different from the amount detected in buffer (NZ5), indicating no affinity at low pH.


of a commercially available lure (AtrAedesw*) utilized for monitoring populations of gravid females of
Aedes aegypti.55




Concluding Remarks
The state-of-the-art screening programs described here should not be oversold as the panacea for
controlling mosquito-borne diseases. To generate practical applications, these molecular-based
screening strategies have to be integrated with sensory physiology, behavioral bioassays, and field
studies. Interaction of a ligand with an olfactory protein does not necessarily imply full physiological
function, behavioral response, mosquito trapping, or reduced biting. While we hope that these molecular-
based programs may ultimately lead to the decrease of mosquito-transmitted diseases, the discovery of
new repellents and attractants is just a stepping stone towards the ultimate goal.

Acknowledgments
This work was supported by the NIH-National Institute of Allergy and Infectious Diseases
(1U01AI058267-01), a Specific Cooperative Agreement (No. 58-1275-1-042) with Chemical Affecting
Insect Behavior Laboratory, Agricultural Research Service, U.S. Department of Agriculture, and a
Research Agreement with Bedoukian Research, Inc. I benefited greatly from discussions with past and
current undergraduate, graduate, postdoctoral students, and visiting scientists in my laboratory as well as
with various collaborators over the years. I thank Yuko Ishida, Zainulabeuddin Syed, and Wei Xu for
their suggestions to improve an earlier draft of the manuscript; Wei Xu for running the binding assay with
AaegOBP1, and Dr. Greg Lanzaro for providing Anopheles gambiae mosquitoes for protein extraction.
Helpful comments and suggestions by Dr. George Kamita, Dr. Mark Wogulis and my department
colleague and collaborator, Dr. Anthon Cornel, are also highly appreciated.


*
    Registered trademark of Ecovec, Ltd, Brazil.


q 2006 by Taylor & Francis Group, LLC
240                                                       Insect Repellents: Principles, Methods, and Uses


References
     1. J. Murlis, M. A. Willis, and R. T. Carde, Spatial and temporal structures of pheromone plumes in fields
        and forests, Physiol. Entomol., 25, 211–222, 2000.
     2. W. S. Leal, Proteins that make sense, in Insect Pheromone Biochemistry and Molecular Biology, The
        Biosynthesis and Detection of Pheromones and Plant Volatiles, G. J. Blomquist and R. G. Vogt (Eds.),
        London: Elsevier Academic Press, 2003, pp. 447–476.
     3. W. S. Leal, Pheromone reception, Top. Curr. Chem., 240, 1–36, 2005.
     4. H. Wojtasek and W. S. Leal, Conformational change in the pheromone-binding protein from Bombyx
        mori induced by pH and by interaction with membranes, J. Biol. Chem., 274, 30950–30956, 1999.
     5. W. S. Leal, A. M. Chen, and M. L. Erickson, Selective and pH-dependent binding of a moth
        pheromone to a pheromone-binding protein, J. Chem. Ecol., 31, 2493–2499, 2005.
     6. B. H. Sandler et al., Sexual attraction in the silkworm moth: Structure of the pheromone-binding-
        protein-bombykol complex, Chem. Biol., 7, 143–151, 2000.
     7. F. Damberger et al., NMR characterization of a pH-dependent equilibrium between two folded
        solution conformations of the pheromone-binding protein from Bombyx mori, Protein Sci., 9,
        1038–1041, 2000.
     8. W. S. Leal, Duality monomer-dimer of the pheromone-binding protein from Bombyx mori, Biochem.
        Biophys. Res. Commun., 268, 521–529, 2000.
     9. R. Horst et al., NMR structure reveals intramolecular regulation mechanism for pheromone binding
        and release, Proc. Natl. Acad. Sci., USA, 98, 14374–14379, 2001.
    10. W. S. Leal et al., Kinetics and molecular properties of pheromone binding and release, Proc. Natl.
        Acad. Sci., USA, 102, 5386–5391, 2005.
    11. M. Wogulis et al., The crystal structure of an odorant binding protein from Anopheles gambiae:
        Evidence for a common ligand release mechanism, Biochem. Biophys. Res. Commun., 339, 157–164,
        2006.
    12. C. H. Wetzel et al., Functional expression and characterization of a Drosophila odorant receptor in a
        heterologous cell system, Proc. Nat. Acad. Sci., USA, 98, 9377–9380, 2001.
    13. K. -E. Kaissling, R.H. Wright Lectures on Insect Olfaction, British Columbia: Simon Fraser University,
        1987.
    14. K. -E. Kaissling, Olfactory perireceptor and receptor events in moths: A kinetic model, Chem. Senses,
        26, 125–150, 2001.
    15. R. G. Vogt, L. M. Riddiford, and G. D. Prestwich, Kinetic properties of a sex pheromone-degrading
        enzyme: The sensillar esterase of Antheraea polyphemus, Proc. Natl. Acad. Sci., USA, 82, 8827–8831,
        1985.
    16. Y. Ishida and W. S. Leal, Rapid inactivation of a moth pheromone, Proc. Natl. Acad. Sci., USA, 102,
        14075–14079, 2005.
    17. M. Maibeche-Coisne et al., Pheromone anosmia in a scarab beetle induced by in vivo inhibition of a
        pheromone-degrading enzyme, Proc. Natl. Acad. Sci., USA, 101, 11459–11464, 2004.
    18. E. J. Hoffmann and J. R. Miller, Reassessment of the role and utility of wind in suppression of mosquito
        (Diptera: Culicidae) host finding: Stimulus dilution supported over flight limitation, J. Med. Entomol.,
        40, 607–614, 2003.
    19. J. A. Klun, A. Khrimian, and M. Debboun, Repellent and deterrent effects of SS220, Picaridin, and
        DEET suppress human blood feeding by Aedes aegypti, Anopheles stephensi, and Phlebotomus
        papatasi, J. Med. Entomol., 43, 34–39, 2006.
    20. J. Boeckh et al., Acylated 1,3-aminopropanols as repellents against bloodsucking arthropods, Pest. Sci.,
        48, 359–373, 1996.
    21. W. S. Leal and K. Uchida, Application of GC-EAD to the determination of mosquito repellents derived
        from a plant, Cymbopogon citratus, J. Asia-Pacific Entomol., 1, 217–221, 1998.
    22. C. Constantini et al., Electroantennogram and behavioural responses of the malaria vector Anopheles
        gambiae to human-specific sweat components, Med. Vet. Entomol., 15, 259–266, 2001.
    23. R. Maida et al., The pheromone binding protein of Bombyx mori: Purification, characterization and
        immunocytochemical localization, Insect Biochem. Molec. Biol., 23, 243–253, 1993.


q 2006 by Taylor & Francis Group, LLC
Molecular-Based Chemical Prospecting of Mosquito Attractants and Repellents                                241


    24. J. Krieger et al., Binding proteins from the antennae of Bombyx mori, Insect Biochem. Molec. Biol., 26,
        297–307, 1996.
    25. P. X. Xu, L. J. Zwiebel, and D. P. Smith, Identification of a distinct family of genes encoding atypical
        odorant-binding proteins in the malaria vector mosquito, Anopheles gambiae, Insect Mol. Biol., 12,
        549–560, 2003.
    26. H. Biessmann et al., Microarray-based survey of a subset of putative olfactory genes in the mosquito
        Anopheles gambiae, Insect Mol. Biol., 14, 575–589, 2005.
    27. D. S. Hekmat-Scafe et al., Genome-wide analysis of the odorant-binding protein gene family in
        Drosophila melanogaster, Genome Res., 12, 1357–1369, 2002.
    28. K. Galindo and D. P. Smith, A large family of divergent Drosophila odorant-binding proteins expressed
        in gustatory and olfactory sensilla, Genetics, 159, 1059–1072, 2001.
    29. Z. X. Li et al., Identification and expression of odorant-binding proteins of the malaria-carrying
        mosquitoes Anopheles gambiae and Anopheles arabiensis, Arch. Insect. Biochem. Physiol., 58,
        175–189, 2005.
    30. J. J. Zhou et al., “Plus-C” odorant-binding protein genes in two Drosophila species and the malaria
        mosquito Anopheles gambiae, Gene, 327, 117–129, 2004.
    31. Y. Ishida et al., Intriguing olfactory proteins from the yellow fever mosquito, Aedes aegypti,
        Naturwissenschaften, 91, 426–431, 2004.
    32. K.-E. Kaissling et al., A new pheromone of the silkworm moth Bombyx mori. Sensory pathway and
        behavioral effect, Naturwissenschaften, 65, 382–384, 1978.
    33. K. -E. Kaissling and E. Priesner, Die Riechschwelle des Seidenspinners, Naturwissenschaften, 57,
        23–28, 1970.
    34. E. Priesner, Progress in the analysis of pheromone receptor systems, Ann. Zool. Ecol. Anim., 11,
        533–546, 1979.
    35. R. A. Steinbrecht, M. Ozaki, and G. Ziegelberger, Immunocytochemical localization of pheromone-
        binding protein in moth antennae, Cell Tissue Res., 270, 287–302, 1992.
    36. M. Laue, R. A. Steinbrecht, and G. Ziegelberger, Immunocytochemical localization of general odorant-
        binding protein in olfactory sensilla of the silkmoth Antheraea polyphemus, Naturwissenschaften, 81,
        178–180, 1994.
    37. M. Laue and R. A. Steinbrecht, Topochemistry of moth olfactory sensilla, Int. J. Insect Morphol.
        Embryol., 26, 217–228, 1997.
    38. Y. Ishida, A. J. Cornel, and W. S. Leal, Identification and cloning of a female antenna-specific
        odorant-binding protein in the mosquito Culex quinquefasciatus, J. Chem. Ecol., 28, 867–871,
        2002.
    39. Y. Ishida, A. J. Cornel, and W. S. Leal, Odorant-binding protein from Culex tarsalis, the most
        competent vector of West Nile Virus in California, J. Asia-Pacific Entomol., 6, 45–48, 2003.
    40. H. Biessmann et al., Isolation of cDNA clones encoding putative odourant binding proteins from
        the antennae of the malaria-transmitting mosquito, Anopheles gambiae, Insect Mol. Biol., 11,
        123–132, 2002.
    41. A. A. Dobritsa et al., Integrating the molecular and cellular basis of odor coding in the
        Drosophila antenna, Neuron, 37, 827–841, 2003.
    42. E. A. Hallem, M. G. Ho, and J. R. Carlson, The molecular basis of odor coding in the
        Drosophila antenna, Cell, 117, 965–979, 2004.
    43. E. A. Hallem et al., Mosquito receptor for human-sweat odorant, Nature, 427, 212–213, 2004.
    44. L. Ban et al., Binding properties of a locust’s chemosensory protein, Biochem. Biophys. Res.
        Commun., 293, 50–54, 2002.
    45. N. Honson et al., Structure—activity studies with pheromone-binding proteins of the gypsy moth,
        Lymantria dispar, Chem. Senses, 28, 479–489, 2003.
    46. P. Pelosi, et al., Soluble proteins in insect chemical communication, Cell. Mol. Life Sci., in press.
    47. L. Briand et al., Ligand binding and physico-chemical properties of ASP2, a recombinant
        odorant-binding protein from honeybee (Apis mellifera L.), Eur. J. Biochem., 268, 752–760, 2001.
    48. E. Plettner et al., Discrimination of pheromone enantiomers by two pheromone binding proteins
        from the Gypsy moth Lymantria dispar, Biochemistry, 39, 8953–8962, 2000.


q 2006 by Taylor & Francis Group, LLC
242                                                     Insect Repellents: Principles, Methods, and Uses


    49. N. D. Cook, Scintillation proximity assay: A versatile high-throughput screening technology,
        Drug Discov. Today, 1, 287–294, 1996.
    50. A. Margaryan, et al., Synthesis of an immobilized Bombyx mori pheromone binding protein liquid
        chromatography stationary phase, Talanta, in press.
    51. W. S. Leal et al., Unsual pheromone chemistry in the navel orangeworm: Novel sex attractants
        and a behavioral antagonist, Naturwissenschaften, 92, 139–146, 2005.
    52. W. S. Leal et al., Molecular-based screening of attractants for the Navel Orangeworm, in Proc.
        33rd Almond Industry Conference, Modesto, CA, 27–35, 2005.
    53. J. A. Coffelt et al., Isolation, identification, and synthesis of a female sex pheromone of the navel
        orangeworm, Amyelois transitella (Lepidoptera: Pyralidae), J. Chem. Ecol., 5, 955–966, 1979.
    54. B. R. Laurence and J. A. Pickett, Erythro-6-Acetoxy-5-hexadecanolide, the major component of a
        mosquito oviposition attractant pheromone, J. Chem. Soc. Chem. Commun., 59–60, 1982.
                                                ¸˜                   ´              ¸˜       ˆ
    55. R. A. Roque and A. E. Eiras, Avaliacao de atraente sintetico de oviposicao para femeas gravidas
        de Aedes aegypti (Diptera: Culicidae), in IV Encontro Brasileiro de Ecologia Quımica,         ´
        Piracicaba: SP-Brazil, 156, 2005.
    56. C. Lautenschlager, W. S. Leal, and J. Clardy, Coil-to-helix transition and ligand relase of Bombyx
        mori pheromone-binding protein, Biochem. Biophys. Res. Commun., 335, 1044–1050, 2005.




q 2006 by Taylor & Francis Group, LLC
12
Evaluation of Topical Insect Repellents and Factors
That Affect Their Performance


Scott P. Carroll



CONTENTS
Introduction ...................................................................................................................................245
History ...........................................................................................................................................246
Types of Tests—Background........................................................................................................247
Factors Affecting Repellent Performance.....................................................................................248
  Mosquito Taxonomy and Genetics ...........................................................................................248
  Individual Human-Subject Differences .....................................................................................249
  Conditions of Use ......................................................................................................................250
  Formulation Chemistry..............................................................................................................251
  Active Ingredients and Their Efficacy Assessment ..................................................................252
     Laboratory Efficacy Comparisons .........................................................................................252
     Field Efficacy Comparisons...................................................................................................253
Conclusions ...................................................................................................................................255
Acknowledgments .........................................................................................................................256
References .....................................................................................................................................256




Introduction
Personally-applied topical insect repellents are a flexible and relatively affordable means of gaining
protection from biting arthropods and the disease-causing pathogens they sometimes carry.1,2 Although a
number of useful repellents have been developed, a variety of factors limits their effectiveness in
application. The purpose of this chapter is to review those factors, consider their importance, and discuss
means of overcoming them. The majority of investigations have been conducted against mosquitoes that
are vectors of important disease agents: the yellow fever mosquito, Aedes aegypti, and the Anopheles
species that transmit malaria pathogens. Although this chapter emphasizes results from studies of
mosquitoes, data from other biting arthropods are included when helpful or relevant.
   For a repellent to be successful, it must first have a high percentage of effectiveness against the biting
arthropods of concern for the entire period of likely use. Second, it should be toxicologically safe at the
rate of application for which it is intended. Third, it should be easy to apply and pleasant on the skin in


                                                                                                                                                245

q 2006 by Taylor & Francis Group, LLC
246                                                     Insect Repellents: Principles, Methods, and Uses


terms of residual feeling and odor. Finally, the entire spectrum of costs involved in production and
marketing of the repellent should result in a product that is reasonably priced for the consumer. Among
the repellent active ingredients formulated over the last half century, deet (N,N-diethyl-3-methylbenza-
mide) has been included in numerous products that come remarkably close to approaching that ideal, and
it is estimated that deet is employed at least 200 million times per year around the globe.3 Persistent
public concerns about its safety (some based on hearsay) have been aggravated by its cosmetic
shortcomings and plasticizing (i.e., tendency to soften plastics) effects. Cosmetic improvements have
been achieved mainly by limiting deet concentrations to 10% or lower, resulting in formulations with
efficacy of limited duration. In addition, while high-concentration deet formulations often remain
efficacious for eight or more hours, attempts to enhance duration by manipulating carrier formulations
have not resulted in substantial improvements. This suite of concerns has helped to fuel the search for
suitable alternatives for both civilian and military applications.
   Little is known about how insect repellents function.4 Such knowledge would promote the
development of more effective repellents based on biochemical and neurophysiological principles. In
the absence of real knowledge about mechanisms, we may instead progress inferentially through the
collation and analysis of natural history data on factors that influence success. Interactions between
parasites and hosts are biologically complex and therefore inherently dynamic and challenging to
control. Among the many factors likely to influence the effectiveness of a repellent are those involving
the active ingredient and formulation, biology of the arthropod, the conditions and mode of use, and
lastly, individual user traits. The diversity of variables and their interactions makes the precise
measurement of performance difficult, requiring a great deal of empirical effort. Organized testing
schemes that control variables systematically are therefore especially useful. Nonetheless, the
complexity of host-parasite interplay suggests a priori that protection afforded by even the best active
ingredient in an ideal formulation is likely to differ among arthropod taxa and among individual human
subjects. Accordingly, comparative studies that examine such interactions should be especially valuable
for advancing repellent science.
   In spite of these challenges, a number of promising active ingredients and formulation technologies
have recently been developed. By identifying the liabilities that influence repellent performance, chances
are now better than in the past to integrate the new resources to create superior, longer-lasting, more
universally acceptable insect repellents. Laboratory tests are effective for screening purposes and for
making comparisons under controlled conditions. Field tests give a better picture of repellent
performance in actual use, and highlight the importance of the environment and other conditions of
use. Accordingly, this chapter first reviews studies that describe the action and importance of factors that
influence repellent performance. It then considers those factors in evaluating recent performance tests of
promising deet alternatives. The goal is to present information that is directly relevant to issues faced by
contemporary decision makers and to emphasize the importance of recognized variables, the better
understanding of which may improve development prospects.




History
Insect repellents have been examined systematically in the U.S. since World War II, when military
initiatives, in response to outbreaks of malaria in American soldiers in tropical theaters, were taken up by
the U.S. Department of Agriculture (USDA).5 That work mainly involved the screening of novel active
ingredients against caged laboratory populations of Aedes aegypti and Anopheles albimanus.6
Ultimately, however, substantial work also assessed factors that influenced the performance of known
repellents (principally dimethyl phthalate and deet), particularly with regard to the duration of
repellency.7 Those pioneering studies established the fundamental importance of dosage and rate of
loss for determining the period of protection. Among the chief factors they identified as influencing loss


q 2006 by Taylor & Francis Group, LLC
Evaluation of Topical Insect Repellents and Factors That Affect Their Performance                                  247


were rates of evaporation and absorption that differed among individuals, and abrasion by clothing.
Individual attractiveness to a biting arthropod was also important, but gender, hairiness, sweat, and
chemical deterioration were thought not to influence repellency.7 While conceptually robust and
comprehensive, most early studies had five or fewer subjects and probably served later researchers
more in terms of intellectual guidance than through the specific applications of the results.
   In the succeeding four decades, basic research on repellents in the U.S. has continued to be sponsored
heavily by the military and the USDA, with emphasis on extending duration. Industrial research over this
period has stressed user acceptability and marketing appeal, whereas in Europe the market has more
frequently addressed safety. Developing countries seem to stress cost (including searches for natural
products). The majority of military work has been conducted with deet and laboratory strains of Aedes
aegypti, although more recent work includes significant field studies and tests of experimental active
ingredients. That initiative includes several studies that compared the original U.S. Army Insect
Repellent (75% deet in ethanol, hereinafter “Army 75% deet”) to two polymerized deet lotions,
specifically the 3M 34% deet formulation currently known as EDTIAR (extended duration topical
insect and arthropod repellent) and marketed to the public as 3M Ultrathonw, and the Biotekw 42% deet
formulation.* Such work is discussed in detail later in this chapter when the influence of formulation
is considered.




Types of Tests—Background
Performance evaluations of repellents fall into two basic classes or design types. In the first approach,
developed for field testing, a treated surface is exposed until a conservative, predefined failure event
occurs, e.g., the time of the first bite, or the “first confirmed bite” (defined as the first bite that is followed
by another bite within 30 min). This approach has the practical advantage of minimizing subject risk
from wild mosquito bites. However, its scientific disadvantages include that the data set is truncated and
minimized in size, and offers no basis for analyzing or comparing the period of partial protection after the
onset of biting. In addition, truncation may inherently oversample that portion of the mosquito
population that is most insensitive to the repellent. As pointed out by Rutledge in a number of
publications,8 measurements made of extreme individuals will be less reliable than those taken closer
to the center of the population distribution. Depending on biting rates, some of these problems may be
partially ameliorated by instead defining effective repellency as the duration of some percentage of
protection (e.g., 90 or 95%) relative to the control.
   In field studies, an important factor influencing protection time is therefore likely to be the population
size of the arthropod.8,9 Khan et al.10 and Barnard et al.11 reached similar conclusions based on
experimental manipulations of mosquito numbers in cages. The probability that a test subject will
encounter extremely insensitive arthropods will be higher in large parasite populations. Based on these
statistical observations, Rutledge et al.8 recommended the adoption of dose-response test design focused
at the more typical portion of the mosquito population. At the median dose (i.e., the quantity required to
repel 50% of the test arthropods), the result is essentially independent of the population size. Known as
ED50 (the minimum effective dosage to repel half of the arthropods), this test design allows much greater
precision in the generation of a true estimate of repellent performance because of the inherent
mathematics of error around a log-dose/probit curve. It also permits measurement of the sensitivity of
different percentiles to population size, and focuses on percentiles of specific interest.
   “Minimum effective dosage” design and analysis is employed in laboratory evaluations of inherent
repellency where the size of the test population is known. The resulting precision may be especially

*
  Ultrathon is a registered trademark of the 3M Corporation, Minneapolis, MN; Biotek is a registered trademark of Biotek
Corporation, Woburn, MA.


q 2006 by Taylor & Francis Group, LLC
248                                                       Insect Repellents: Principles, Methods, and Uses


valuable for comparing active ingredients and formulations. To bolster data quality and information
content, field evaluations would likewise benefit from more extended records of biting events (i.e.,
extending the trial past the time of the first confirmed bite). For field testing, an important corollary of
the foregoing is that the number of study subjects will directly influence the number of mosquitoes
sampled, and thus the effective population size of mosquitoes from which data are collected. The
common practice of employing just a few subjects per formulation (below) may therefore give a poor
indication of the range of experiences that would characterize a larger sample of subjects. In other
words, while analytical precision is gained from ED50 laboratory studies by reducing the influence of
rare insensitive mosquitoes, field evaluations of effective repellency benefit from the inclusion of
exceptional mosquitoes, the avidity of which exceeds the capacity of the repellent to stop them from
biting. It is important to sample with sufficient intensity to gauge performance against a large number
of potentially biting individuals.




Factors Affecting Repellent Performance
Mosquito Taxonomy and Genetics
The first comprehensive study of the interaction between repellency and mosquito taxonomy was
conducted by Travis,12 who showed that the ranking of protection provided by four repellents was not the
same among two Aedes and two Anopheles species. Rutledge and colleagues conducted both intensive8
and extensive13 studies of such interaction. In a study examining deet alone against Anopheles, Aedes,
and Culex, the range in ED50 was seven-fold.8 Three species of Anopheles ranged from nearly the most,
to the least easily repelled as a function of dosage. Even within a species (among ten strains of Aedes
aegypti), they found significant variation in efficacy. Later, in a comparison of 31 repellent compounds,
there was little or no predictability in performance rank across species.13 Variation in observed
repellency between species within a genus was as great as variation between species in different
genera. Performance against Aedes aegypti was a poor predictor of performance against other
mosquitoes, especially Anopheles species.
   In a series of incisive analyses, Curtis et al.14 considered the interactions of mosquito species,
repellents, and individual subject effects. Six species of mosquitoes from Anopheles, Aedes, and Culex
were exposed to six repellents. The ED50 of the repellents varied within and among genera by a factor
ranging from three to 20-fold. Subjects differed in attractiveness, but not consistently across species of
mosquitoes (assessed in the next section). Performance depended on the interaction of subject, repellent,
and mosquito taxon. Similarly, Badolo et al.15 found a repellent-by-taxon interaction in effective dosage
of deet and Picaridin against native West African strains of caged Aedes aegypti and Anophles gambiae.
Results from studies such as these discourage the notion that accurate performance generalizations are
possible from tests with small numbers of subjects against a limited set of target species.
   Finally, Coleman et al.16 broadened systematic comparisons further when considering the influence of
deet, a lactone, and two piperidines against four Anopheles species, and two phlebotomines, Phlebotomus
papatasi and Lutzomyia longipalpus. In general, Anopheles stephensi and the phlebotomines were the
most susceptible to the repellents, and Anopheles albimanus was the least susceptible. Beyond those
patterns, however, the relationship of performance among all the taxa was highly variable. Note also that
deet is not always a superior repellent for phlebotomines.17
   Given the high intergeneric, interspecific, and intraspecific variation in response to repellents observed
in controlled laboratory settings, it is not surprising that the response has a genetic basis. Rutledge et al.18
established that repellent tolerance in Aedes aegypti is heritable, and in the case of deet involves partial
dominance (one or a few genes of major effect). Such genetic control could result in an initially rapid
phenotypic response to selection for deet tolerance.


q 2006 by Taylor & Francis Group, LLC
Evaluation of Topical Insect Repellents and Factors That Affect Their Performance                        249


Individual Human-Subject Differences
Bernier et al. (Chapter 4) reviewed the influence of human skin emanations on mosquito host location.
Gilbert et al.19 examined the influence of ten “subject variables” on attractiveness and repellency:
gender, age, weight, skin temperature, skin moisture production, menses (females), and race, plus hair
color and complexion within Caucasians. A remarkable sample size—50 adults of each gender—gave
unusual statistical power to analyze subtle effects. The attractiveness tests were conducted in
“olfactometer cages,” in which Aedes aegypti were exposed to air pulled across the surface of the
repellent-treated arms of the subjects. The mosquitoes had the option of moving toward the arm and
becoming trapped (and counted) as they approached it. Repellency was scored using 5% deet with
exposure to mosquitoes at intervals.
   Only the effect of gender was clearly and strongly significant. On a proportional scale, the
attractiveness of women was just 73% that of men. Only about 5% were more attractive than the
male median. So while a few women were highly attractive to the mosquitoes (two of the ten most
attractive subjects), all ten of the ten least attractive subjects were female. In terms of repellency, the
lower female attractiveness was reflected in a 37% greater mean protection time for females as a group.
Nonetheless, there was no significant correlation between individual attractiveness and protection time in
either gender, suggesting that other factors are involved in repellent performance.
   Among the other factors investigated, subjects with the highest skin temperatures were more
attractive or more poorly protected than those at the opposite extreme. Women with the highest
moisture production from the skin were also more attractive than the opposite extreme, but that
comparison yielded the reverse in men. Neither of these variables correlated with attractiveness or
repellency across all subjects in a gender, however. Age, weight, menses, hair color and complexion
were all inconsequential,19 and the number of non-Caucasians tested was insufficient for meaningful
interpretation of racial effects. No formal multivariate analyses of the dependent variables
were conducted.
   Given the clarity of that study’s conclusion that women were less attractive and better protected
from Aedes aegypti by deet, it is intriguing that a recent major study with Anopheles stephensi
reported the opposite result. Golenda et al.20 examined the duration of protection by EDTIAR to
caged Anopheles stephensi in 60 female and 60 male volunteers. Self-dosing was performed by
subjects in accordance with product label directions, and the mean rate of application was slightly
higher in females (6%), but not significantly different from males. Biting rates on untreated arms were
also the same between the sexes. Protection rates (relative to the untreated arms) are shown for each
3-h sample interval in Table 12.1. Women experienced significantly less protection over time than
did men.
   Examining an additional aspect of subject variation, Curtis et al.14 reported that each subject’s relative
attractiveness to mosquitoes is species-specific. Using caged Anopheles coustani, Culex quinquefas-
ciatus, and Mansonia species, they found no predictable relationship between how the biting rate



                    TABLE 12.1
                    Comparative Repellency ((1—Biting Rate Treated)/(Biting Rate Control)!
                    100) of U.S. Military EDTIAR (34% deet) on Male and Female Subjects
                                                        Mean Repellency (%)
                    Gender              0h           3h             6h            9h          12 h

                    Females             100          99.3          92.8           79.7        66.3
                    Males               100         100            97.6           91.9        77.5

                    Source: From C. F. Golenda, V. B. Solberg, R. B. Burge, J. M. Gambel, and R. A.
                    Wirtz, American Journal of Tropical Medicine and Hygiene, 60, 654–657, 1999.


q 2006 by Taylor & Francis Group, LLC
250                                                    Insect Repellents: Principles, Methods, and Uses


individuals experienced ranked from one species versus another. In addition to the possible effects of skin
temperature and moisture,19,21 or their correlates, such inter-individual variation in attractancy may be
influenced by differences in skin surface lipids.22 Subjects may also vary in repellent performance due to
differences in dermal absorption of the active ingredient, which in one study ranged from four to 14% of
deet applied in a 15% ethanol solution.23


Conditions of Use
Insect repellents are used in nature, where conditions may interact with user activity to influence
repellency. It is well known that mosquitoes are most active under particular environmental conditions,
and while optima vary among species, warm humid conditions with moderate to low light levels and low
wind generally enhance mosquito foraging activity. Within the range of conditions appropriate for
mosquito foraging, variation in temperature and humidity may not strongly influence biting rate and
repellent performance.24 Comparatively less is known about the state-dependence of mosquito foraging
decisions beyond basic effects of age and parity.25 For example, nutritional status, as determined by
either the larval or adult environment, could influence foraging decisions. In addition, social facilitation
(i.e., stimulation to feed by the presence of foraging conspecifics)26 could in theory increase tolerance to
a repellent.
   Biting pressure, also known as the “ambient biting rate,” is a condition basic to the measurement of
repellent performance. This value may be measured in untreated subjects exposed to foraging
ectoparasites. Higher biting pressures should correspond, in general, to greater parasite densities
and, in nature, larger local population sizes and relatively fewer alternate sources of blood meals.
Under high biting pressure conditions, repellents are likely to fail sooner because the encounter rate
with the least sensitive foragers in the population will be great enough to cause failure based on
absolute (e.g., first confirmed bite) rather than relative (percent biting reduction) criteria.8 Similarly,
efficacy tests with large numbers of subjects may sample more such insensitive mosquitoes, and
perhaps even more on a per capita basis should group size enhance the detectability of hosts to
parasites. Moreover, the availability of alternative host individuals may affect mosquito biting
behavior and thus repellent performance. Repellents may be more effective when mosquitoes have
the simultaneous option of choosing a more attractive host.14 All of these basic factors should
influence test design and conduct, but their importance may differ across mosquito species
and conditions.
   Studies have also shown a number of more specific, user-mediated, proximate conditions that
influence repellent performance. As is typical, most experimental data available are for deet
formulations. Conditions of actual use that may reduce the duration of protection include contact with
water, sweating, and abrasion by clothing or vegetation.7,27–29 Rueda et al.29 reached two main
conclusions regarding the interaction of repellents and clothing. First, abrasion of treated skin by
clothing fabric can significantly lower the protection afforded by a repellent. Second, the amount of
friction between skin and fabric was increased by the presence of a repellent on the skin. This increase in
friction likely aggravated the rate of its loss to the fabric. This study was conducted using the U.S.
military polymer based extended duration deet formulation (EDTIAR). The generality of the results has
not been explored with other formulations or active ingredients.
   Volatilization may be one of the most important variables, as it accounts for a major fraction of
repellent loss from the skin.30 In consequence, subject-caused differences in the rate of volatilization
(whether related to physiological or activity differences) should be an important determinant of
individual variation in repellent efficacy. However, no research appears to have directly examined the
relationship between volatilization and repellency beyond the basic studies of Rutledge et al.30 Costantini
et al.31 used the Rutledge method to model evaporation differences among repellents based on efficacy
data, but they did not measure volatilization directly. Likewise, the extent to which conditions of use
influence dermal absorption appears not to have been quantified.


q 2006 by Taylor & Francis Group, LLC
Evaluation of Topical Insect Repellents and Factors That Affect Their Performance                       251


Formulation Chemistry
Even within the standard test model of deet and Aedes aegypti, substantial variation in protection has
been reported for decades.19,32,33 Given the many variables likely to underlie unexplained performance
variation, Buescher et al.33 reasoned that illuminating basic physical properties of repellent persistence
could provide an important baseline for sensible repellent design. Using deet at a series of dilutions, they
computed a dose-response curve describing the influence of concentration on the duration of 95%
protection against caged Aedes aegypti. The curve is negatively exponential, meaning that each increase
in concentration provides a progressively smaller increment in protection. Their main conclusion was
that the Army 75% deet formulation achieved little added protection compared to, for example, a 50%
concentration. This is a significant finding because use of lower concentrations would reduce deet’s
plasticizing effects and toxicological risk values.
   While the importance of volatilization in limiting repellency duration was understood when the
Buescher et al.33 report appeared in the mid-1980s, it is likely that formalizing the dose-response
relationship laid the foundation for a more analytical approach to designing extended-duration
formulations that would deliver sufficient molecules for repellency over a predictable time span.
Nonetheless, attempts to manipulate the chemistry of repellent carriers, whether through blending
with a polymer or microencapsulation, to control volatilization (and dermal absorption at the same time)
have met with mixed success.
   High volatility is likely to both enhance repellency and evaporation, leading to ephemeral protection.
In the face of this tradeoff, Reifenrath and Rutledge34 investigated the impact of numerous silicone
polymers on the efficacy or protection time of deet against Aedes aegypti using dogs and mice. There
was little influence in the dogs, and while 40% of the polymers increased performance in the mice, the
changes were not large. Mehr et al.35 examined controlled release polymers and starch microencapsula-
tion of deet using the same mosquito species on white rabbits. Some increased duration of efficacy
significantly, but none achieved better than 80% protection at 12 h. The efficacy results of a field test by
Gupta et al.36 that compared the Army 75% deet repellent with two candidate extended duration polymer
formulations (Biotek with 42% deet and EDTIAR with 34% deet) are not interpretable for our purposes
here, but important information on dosing did emerge. Ad libitum self-application resulted in an inverse
relationship between deet concentration and the total amount of each formula applied, so that the mean
quantity of deet applied differed little between the three products.
   This same inverse dosing relationship characterized a laboratory test of the same formulations against
Aedes aegypti, Aedes taeniorhynchus, Anopheles stephensi, and Anopheles albimanus by Gupta and
Rutledge.24 With a total of three subjects in three simulated climates, Biotek provided 94.9% protection,
and EDTIAR 94.8% protection, from bites of all mosquito species in a series of exposures over 12 h.
These values were superior to the 82% protection afforded by the Army 75% deet in ethanol. Enhanced
performance in the polymerized formulations may stem from a combination of reduced volatilization and
skin penetration.37 Interestingly, Gupta and Rutledge24 concluded that the EDTIAR was the best
formulation because the performance of Biotek was “at best similar or less than that provided by the 3M
formulation,” an assessment not consistent with the means they reported (above). In addition they cited
the advantage of EDTIAR having the lowest deet concentration, but given the observed dosing (mean
Biotek 0.9 mg/cm2, mean EDTIAR 1.1 mg/cm2), more deet was actually delivered when the EDTIAR
was applied. Overall, in spite of the excellent general design of this study, the use of only three study
subjects limits the value of assessing the results at any greater level of detail or generalizing strongly
from them.
   Two more recent studies, using laboratory rabbits and deet, have yielded clearer and more positive
results concerning formulation and duration. Rutledge et al.37 tested eight polymer and nine
microencapsulated formulations. Against Aedes aegypti and Anopheles albimanus, several were more
effective than unformulated deet at the same concentration for periods of up to 24 h. The best
performance was with a polymer containing a high molecular weight fatty acid, and with micro-
encapsulated formulations containing a diversity of large molecules, including lanolin, gums, acids, and


q 2006 by Taylor & Francis Group, LLC
252                                                     Insect Repellents: Principles, Methods, and Uses


polypropylene glycol. In a study with argasid ticks, Salafsky et al.38 reported that a liposomal
formulation designed to reduce volatilization and dermal absorption extended the duration of repellent
protection. In a three-day trial, attachment to a finger treated with liposomal deet was absent or
significantly reduced compared to an equal concentration of deet in isopropanol, sampled at 24, 48, and
72 h. Given the difficulty of preparing stable polymer formulations of deet, refined alternatives, including
microcapsules and liposomes, should be considered for tests with other active ingredients and biting
arthropods as well.


Active Ingredients and Their Efficacy Assessment
Active ingredients are the focus of most repellent development programs, and their efficacy is assessed
through cage and field testing. The history of deet and other prominent repellents such as dimethyl
phthalate is treated by Moore and Debboun in Chapter 1 and Strickman in Chapter 22. While it is
accurate to state that a variety of subject factors and their interactions with other variables influence
repellent performance, the review in the foregoing sections shows that the precise nature of those factors
is poorly understood. At present, the chief manner in which the influence of such uncontrolled variation
can be moderated (and studied) is by conducting tests with large numbers of subjects.
   This section reviews recent laboratory and field performance trials of promising non-deet repellents
currently marketed in the U.S. and Europe. The goal is to apply insights gained from the foregoing review
to evaluate how factors that influence repellent efficacy have been controlled and coordinated. Studies
considered are mainly those treating Merck IR3535 (3-[N-butyl-N-acetyl]-amino proprionic acid, ethyl
ester), Lanxess Picaridin (aka KBR3023, (1-(1-methyl-propoxycarbonyl)-2-(2-hydroxy-ethyl)-piper-
idine), and PMD (para-menthane-3,8-diol, which is the prime repellent constituent of the U.S. EPA-
registered active ingredient “oil of lemon eucalyptus,” from the tree Corymbia citriodora). These active
ingredients were developed much more recently than deet; all are registered by the U.S. EPA. Most
studies compare them to some type of deet standard. Given the variety of contingencies that apply to the
performance of deet even under controlled conditions against well known mosquito taxa, it is worth
examining how well conditions have been accounted for in tests of active ingredients that are less well
studied. Frances (Chapter 18), Strickman (Chapter 20), and Puccetti (Chapter 21) also treat these three
active ingredients in detail.

Laboratory Efficacy Comparisons
The most widely referenced recent study of comparative mosquito repellent efficacy was conducted with
caged Aedes aegypti by Fradin and Day.39 Their goal was to compare commercial deet products at
various concentrations with plant-based repellents and IR3535 at 7.5%. Two lotions with at least 20%
deet protected subjects for an average of 4–6 h (time to first bite), and most other formulations provided
protection for well under 1 h. The authors concluded that “only products containing deet offer long-
lasting protection.” The design was comparatively strong in terms of the number of test subjects (15), but
the study had at least two apparent weaknesses. First, dosage was not reported and perhaps not closely
controlled. Second, repellents that performed well in a subject’s first exposure were tested at less frequent
intervals in the second and third exposures (apparently for convenience), adding a bias that probably
exaggerated true differences among the products. Despite those shortcomings, the performance
differences were large enough to suggest that conclusions were generally accurate.
   A substantially different picture emerged in the next broad-based cage study,40 which included more
effective commercial deet alternatives. Three mosquito species were tested separately: Culex nigri-
palpus, Aedes albopictus, and Aedes triseriatus. Results for the four most effective products are
highlighted in Table 12.2. Most remarkably, given deet’s five decade reign of superiority in such
testing, overall repellency of the non-deet active ingredients was either consistently slightly greater (in
the case of PMD), or equivalent to, 15% deet. However, for comparative purposes it is unfortunate that
the highest deet concentration tested was only 15%.


q 2006 by Taylor & Francis Group, LLC
Evaluation of Topical Insect Repellents and Factors That Affect Their Performance                                 253


TABLE 12.2
Mean Protection Timea (SE) (hours) for the Four Most Effective Repellents Studied in the Laboratory by
Barnard and Xue40
Product                                 Aedes albopictus              Culex nigripalpus               Aedes triseriatus

Repelw (19.5%b PMD)                         7.8 (0.2)                      7.3 (0.7)                      7.8 (0.2)
Bite Blockerw (2% soy oil)c                 5.5 (1.3)                      8.3 (0.2)                      7.8 (0.2)
Autanw (10% Picaridin)d                     5.7 (0.9)                      8.0 (0.0)                      7.8 (0.2)
Off!w (15% deet)d                           7.2 (0.8)                      7.0 (0.6)                      7.3 (0.3)
a
    Time to second bite in one or two sequential periods.
b
    Corrected from Barnard and Xue40; a registered trademark of Wisconsin Pharmacol Co., Inc., Jackson, WI.
c
    Methylated soy bean oil; a registered trademark of HOMS, LLC, Clayton, NC.
d
    Registered trademarks of S.C. Johnson and Son, Inc., Washington, DC.
Source: From D. R. Barnard, and R. D. Xue, Journal of Medical Entomology, 41(4), 726–730, 2004.



   Strengths of that study include that the repellents were applied at a standard dosage (1 mL/650 cm2 of
skin surface), and tested against a high density of avid mosquitoes. However, an important weakness was
that only two subjects tested each repellent, out of a total of five subjects. Because individuals differ
inherently in their attractiveness to mosquitoes and dermal interaction with repellents, and both factors
interact with mosquito taxon, a substantial portion of the variation reported may be from uncontrolled
subject error.
   Cage studies against Anopheles vectors of Plasmodium (malaria) likewise showed PMD41–43 and
Picaridin15 to be at least as effective as deet formulations. The first three tests had six or fewer subjects
and uncontrolled or unspecified dosing.41 Badolo et al.15 also found Picaridin to be more effective than
deet against an African strain of Aedes aegypti, but the number of subjects and biting pressure were not
reported. Data in Carroll and Loye44 suggested that 19.5% PMD was intermediate in performance
between ten and 30% deet products against Aedes aegypti over an eight hour period (eight PMD subjects,
one subject for each deet formulation, with equivalent dosing and biting pressure of 50 bites/min on
untreated arms). All of these studies would benefit from larger samples or more complete reporting. One
major benefit from more replication would be more realistic comparisons between separate studies.
   There have been fewer studies of IR3535 at higher concentrations than the basic 7.5% Avon formula
(above), but there is an indication that efficacy improves. At 20% IR3535, a study of three subjects at
high biting pressures by Thavara et al.45 found IR3535 comparable to 20% deet against two Culex and
one Aedes species, but less repellent against an Anopheles species.

Field Efficacy Comparisons
Most field efficacy trials share problems common in laboratory trials, including small numbers of
subjects, lack of repetition, uncontrolled dosing, and unclear ambient biting rates. As a result,
characterizing the repellency of a given active ingredient across taxa, and comparing it with other
active ingredients, is difficult to do at a suitable level of precision.
   One of the most thorough and thoughtful studies of contemporary active ingredients was conducted by
Costantini et al.,31 measuring dose-response curves of deet, Picaridin and IR3535 against Anopheles
gambiae complex mosquitoes in Burkina Faso. Eight male subjects tested a series of dosages of the
technical grade repellents diluted in ethanol. Apparently each repellent was tested on 96 nights (12 times by
each subject). Testing was performed over the ten hour period 18:00–04:00 with a two hour break from
22:00–00:00. Four dosages (in ethanol) were tested, specifically 0.1, 0.3, 0.6, and 0.8 mg/cm2 of each
active ingredient. For comparison, standard volume for efficacy testing in the U.S. is ca. 1.54 mg/cm2, so
that for a 30% (high) concentration active ingredient, dosing would equal about 0.5 mg/cm2 of active
ingredient. The two higher doses in this study were thus greater than those intended for most military or


q 2006 by Taylor & Francis Group, LLC
254                                                    Insect Repellents: Principles, Methods, and Uses


popular commercial formulations in the U.S. Picaridin performed best against the anophelines in this
study, with an estimated 95% or more repelled for at least 8 h at the three higher dosages. Deet’s
performance was intermediate, and IR3535 was the least repellent at all dosages. These results are
important because even though deet is historically the best repellent against anophelines, public health
professionals have long recognized the need for a better repellent against these important vectors of the
pathogens causing malaria. The 0.3 mg/cm2 dosage corresponded to a 20% Picaridin formulation, the
maximum concentration that is registered for use in Europe and Australia. Costantini et al.31 provide some
of the first evidence of a repellent lasting for such a long period against Anopheles gambiae (see also
Trigg46 below for PMD performance).
   As in other studies, however, caution is in order. First, in spite of the unusually long duration of the
study (six months in total), which yielded an unusually large data set, just eight subjects were involved,
and only local populations of Anopheles gambiae. Second, although samples for other mosquito taxa
were small, Picaridin did not repel Aedes species better than the other repellents. Third, while control
subjects collected a large number (27,231) of alighting Anopheles gambiae during the study, arithmetic
shows this to be a low ambient biting rate for the study: less than 0.3 per minute (27,231
bites/92,160 min). For perspective, current U.S. EPA guidelines call for a minimum of 1 bite/min on
a lower limb (feet and hands excluded), more than three times greater than the observed rate. So while
the strength of this study is that it was conducted under representative (long-term) conditions, and low
biting rates may be medically important when infection rates in mosquitoes are high, it would still be
valuable to have performance data at higher biting rates. Lastly, data from women are clearly merited.
   Even at such low biting rates, Picaridin may fail quickly against anophelines. Frances et al.50 tested
19.2% Picaridin (Autan Repel Army 20) against 20% deet in ethanol and 35% deet in a gel (the repellent
issued by the Australian Defense Force) against Anopheles meraukensis and Anopheles bancroftii in
Australia’s Northern Territory. At control biting rates slightly under 0.5 bites/min, 35% deet and
Picaridin protected at more than 95% over the first hour, but by the second hour repellency dropped to
78% for Picaridin, and declined variably in all three repellents thereafter. Those data were collected by
four subjects, all male, with each testing a repellent or ethanol control twice over eight consecutive
nights. Dosage appears to have been ad libitum, determined by the subjects at the time of application. By
weight, one can calculate that Picaridin was applied at an average rate 31% higher than the 20% deet, and
45% higher than the 35% deet. In this latter case, only about 25% more deet than Picaridin was actually
administered (estimated from Table 12.1 of Frances et al.50). The rate at which formulated Picaridin was
applied averaged 13% higher than standard procedure for a U.S. repellent efficacy test (1 mL/650 cm2 of
skin surface).
   Other field tests of Picaridin against anophelines are similarly plagued by small samples or low
ambient biting rates (!0.5/min, e.g., Yap et al.47,48), but still suggest its promise as a broad-spectrum
mosquito repellent. In the single test conducted at high ambient biting rates, Barnard et al.49 compared
25% ethanol solutions of technical deet and IR3535, and Picaridin, and PMD at 19.5% in a commercial
lotion (not 40% PMD as indicated in the source publication; see Carroll and Loye44). Five males exposed
treated limbs for 3 min each hour for 6 h, beginning 15 min after application. The test was repeated five
times over three days so that each subject tested each repellent and served as a control (25% water in
ethanol) once. Black salt marsh mosquitoes (Aedes taeniorhynchus) attacked control subjects at a high
average rate of 19.5G13.7 bites/min. Given the small number of subjects, statistical power was low, but
Picaridin and deet appeared to be the most repellent, followed by PMD and then IR3535. Only Picaridin
repelled at greater than 95% through hour five.
   The efficacy of PMD against anophelines appears noteworthy. Using six self-dosed subjects exposed
to Anopheles gambiae in rural Tanzania, Trigg46 compared 50% PMD to 50% deet under low ambient
biting rate conditions (apparently 0.13/min, calculated from grand mean of controls over the 240 min
exposure period, Trigg’s Table 1). Repellents were applied 5 h before the onset of exposure. Deet
prevented all biting on six subjects for close to 7 h, and PMD for 6–8 h, depending on formulation.
Moore et al.50 collected similar data for Anopheles darlingi in Bolivia, but tested only 2–4 h after
application. PMD (30%) reduced biting on five subjects by a mean of 97%, while 15% deet in ethanol


q 2006 by Taylor & Francis Group, LLC
Evaluation of Topical Insect Repellents and Factors That Affect Their Performance                       255


gave just 85% protection. Compared to other studies of anophelines, ambient biting pressure was
respectably high, greater than or equal to 1 bite/min (estimated from the mean percentage biting rate
reductions of the test products, including 0% for the control, and the total number of mosquitoes captured
landing). Variation in the performance of Picaridin among anophelines (e.g., Frances et al.51 above)
suggests that PMD, too should be tested against more anopheline species, using controlled dosing on
more study subjects than in the foregoing studies.
   In a six hour field study of PMD with a large number (20) of adult male and female subjects exposed to
Aedes melanimon and Aedes vexans in California, Carroll and Loye44 found excellent protection with
continuous exposure of lower arms and legs at mean biting pressures of approximately 1.5 and 3 per
minute, respectively. Subjects tested lotion (19.5% PMD) and spray (26% PMD) formulations at dosages
of either 1.6 or 2.4 mg/cm2. Mean biting rate reduction for all treatments over the 6 h was 99.9%.
Protection provided by 20% deet lotion was similar, but only two subjects tested deet.
   Other than Barnard et al.49 field studies of IR3535 at higher concentrations are rare. Thavara et al.45
compared IR3535 and deet at a rate of 20% in ethanol with six subjects against several mosquito species at
low biting rates. In two 8 h field studies of Aedes albopictus at ambient biting rates of about 0.35 bites/
min, there were no bites from this species on subjects using either repellent. IR3535 reduced biting by a
mean of 98.4%. Deet reduced biting by 97.4%. The authors’ claim that the difference, statistically
significant at P!0.05, is inconsequential, however, given the similarity of the means (see Table 12.1 of
referenced study). Protection in similar five hour studies against night-biting Culex, Mansonia and several
Anopheles species (ambient biting pressure 0.15–0.25 in the last genus) averaged 98% and greater for
both repellents.45 Doses were approximately double the standard. Like studies of other promising
repellents, work on IR3535 would benefit from greater standardization of protocols, more subjects, and
higher biting rates.




Conclusions
The task of generating predictable, generalized results from insect repellent efficacy tests is challenging.
The basic difficulty is in the effort to generate and deliver chemicals that will interrupt the feeding
behavior of highly diverse and refined biting arthropods without harming the user. Even among
apparently safe and effective repellent candidates, however, this review demonstrates that the interplay
of host, arthropod, environment, and utility significantly controls performance. We have a better idea of
what classes of variables are influential than we do of how to predict the impact of a particular variable in
any given case.
   For each active ingredient, the basic three-way interplay between subject, formulation, and mosquito
taxon appears to be the principal source of variation in the outcomes of efficacy tests. It is typically an
uncontrolled source of error that hinders all attempts to analyze additional conditional factors (e.g.,
environment, use). Because this axis of interaction is poorly described, the precision of performance
estimates is generally questionable.
   For practical reasons, most studies have attempted to assess variables one or two at a time. From these
we can begin to list factors that should be included in improved models of repellent performance. For
example, it is likely that variation in skin emanations, including lipids,22 moisture and heat19 affect
attractiveness, and that variation in dermal absorbency (three-fold for deet23) affects protection time.
These results can be linked to build a nascent picture of subject variation. At the same time, however, two
large-scale studies of gender that both reported highly significant and substantial effects produced
strikingly dissimilar results: Gilbert, Gouch, and Smith19 found that females were clearly less susceptible
to biting by Aedes aegypti in a test of 5% deet in alcohol, while Golenda et al.20 found that females were
clearly more susceptible to biting by Anopheles stephensi in a test of EDTIAR. So while we can state that


q 2006 by Taylor & Francis Group, LLC
256                                                       Insect Repellents: Principles, Methods, and Uses


gender is an important consideration for repellent performance, at present, uninvestigated interactions
between gender, repellent formulation, and mosquito taxon prevent us from offering further direction.
   The problems of inconsistency in the design, execution, and reporting of efficacy studies likewise
hinder the effort to evaluate repellent performance. In all of the studies reviewed that compare active
ingredients, sample sizes are too small to permit confident distinction among treatments with moderately
close performance values. Even when statistical significance is shown, differences cannot necessarily be
attributed to the repellents alone, and peculiarities of individual interactions may be paramount. Note, for
example, that many studies have used only male subjects. Dosage is another factor of obvious
importance33 that is too often uncontrolled.
   The use of limited numbers of subjects to test repellents probably has its justifications in the desire to
minimize risk, the difficulties associated with recruiting people for this type of work, the use of the first
confirmed bite criterion (a threshold measure), and perhaps also in the history of testing deet, for which
relative variation was apparently regarded as inconsequential due to its outstanding comparative efficacy.
This tradition is reflected in guidelines for efficacy testing proposed by the U.S. EPA, requiring just six
subjects for the generation of registration data. We have entered a new era in which there is for the first
time an interest in comparing several repellent active ingredients, all of high efficacy. How shall they be
distinguished?
   Rutledge and Gupta52 determined by meta-analysis of published studies that the standard deviations of
protection times are a linear function of the means. As a result, the statistical differentiation of long-acting
formulations in particular will probably require especially large samples. It is unfortunate that nZ20, the
minimum acceptable sample size for parametric hypothesis testing at alpha-levels of 0.05, is not the norm
for repellent studies. While the Rutledge and Gupta52 estimate of required n’s is likely inflated by
interstudy variation beyond that relevant to any given comparison (e.g., of two formulations tested
simultaneously), their study does give the impression that even 20 subjects per formula might be too few.
Nonetheless, an agenda to deploy large, balanced groups of subjects to test various repellents against
various mosquitoes would likely advance repellent science substantially.
   The complementary perspective is to accept that separating the performances of candidate deet-
replacements is a futile exercise. A positive outcome of that view might be to open the door more readily
to inclusive strategies, such as combining active ingredients to see if, for example, variance in
performance can be limited. Reducing variance is important because we tend to rely on the mean
protection period when evaluating performance. Any subgroup of people that is less protected than
average will be systematically less protected than is otherwise assumed.
   In addition, it is sensible to make inferences from the results of many different studies that involve the
same repellents. For example, Picaridin seems to be especially efficacious in many of the studies
reviewed here. Meta-analyses of such data sets potentially have the added advantage of treating data
from tests conducted in a variety of conditions against a variety of mosquito taxa. At the same time, the
present suite of studies available seems to share inherent biases (male subjects) and serious
inconsistencies (very unequal dosing) that obscure their value to objective analysis. If we coordinate
and more thoroughly standardize the conduct and reporting of repellent studies in the future, interested
scientists, health professionals, and the public will all benefit from the resulting increase in
available knowledge.

Acknowledgments
For guidance and assistance at UC-Davis, I thank R. Washino, B. Eldridge, T. Scott, J. Loye, S. Lawler,
and S. Minnick. Additional insight has come from D. Barnard, W. Reifenrath, L. Rutledge, M. Schneider,
W. Wakesa, G. White, M. Wundrock, and the editors of this book.

References
      1. R. K. Gupta and L. C. Rutledge, Role of repellents in vector control and disease prevention, American
         Journal of Tropical Medicine and Hygiene, 50, 82–86, 1994.


q 2006 by Taylor & Francis Group, LLC
Evaluation of Topical Insect Repellents and Factors That Affect Their Performance                          257


     2. M. Debboun and D. Strickman, Repellents and the military: Our first line of defense, Wing Beats, 14,
        30–31, 2003, see also pages 33–34.
     3. U.S. Environmental Protection Agency. Office of Pesticides and Toxic Substances, Special Pesticide
        Review Division, N,N-diethyl-m-toluamide (DEET) Pesticide Registration Standard (EPA-540/RS-
        81-004), U.S. Environmental Protection Agency, (PB81-207722), Washington, D.C., 1980.
     4. E. E. Davis, Insect repellents: Concepts of their mode of action relative to potential sensory
        mechanisms in mosquitoes (Diptera: Culicidae), Jounal of Medical Entomology, 22, 237, 1985.
                                                                                                       ¨
     5. W. A. Skinner and H. L. Johnson, The design of insect repellents, in Drug Design, E. J. Ariens (Ed.),
        New York: Academic Press, 1980, pp. 277–305.
     6. W. V. King, Chemicals Evaluated as Insecticides and Repellents at Orlando, Florida, Entomology
        Research Branch, Department of Agriculture, Washington, DC, 1954.
     7. C. N. Smith, I. H. Gilbert, H. K. Gouck, M. C. Bowman, F. Acree, and C. H. Bowman, Factors
        affecting the protection periods of mosquito repellents, United States Department of Agriculture
        Technical Bulletin, 1285, 1–36, 1963.
     8. L. C. Rutledge, M. A. Moussa, C. A. Lowe, and R. K. Solfield, Comparative sensitivity of mosquito
        species and strains to the repellent diethyl toluamide, Journal of Medical Entomology, 14, 536–541,
        1978.
     9. P. Granett, Studies of mosquito repellents. I. Test procedure and method for evaluating data, Journal of
        Economic Entomology, 33, 563–565, 1940.
    10. A. A. Khan, H. I. Maibach, and D. L. Skidmore, Insect repellents: Effect of mosquito and repellent-
        related factors on protection time, Journal of Economic Entomology, 68, 43–45, 1975.
    11. D. R. Barnard, K. H. Posey, D. Smith, and C. E. Schreck, Mosquito density, biting rate and cage size
        effects on repellent tests, Medical and Veterinary Entomology, 12, 39–45, 1998.
    12. B. V. Travis, Relative effectiveness of various repellents against Anopheles farauti Laveran, Journal
        of the National Malaria Society, 6, 180–183, 1947.
    13. L. C. Rutledge, D. M. Collister, V. E. Meixsell, and G. H. G. Eisenberg, Comparative sensitivity of
        representative mosquitoes (Diptera: Culicidae) to repellents, Journal of Medical Entomology, 20,
        506–510, 1983.
    14. C. F. Curtis, J. D. Lines, J. Ijumba, A. Callaghan, N. Hill, and M. A. Karimzad, The relative efficacy of
        repellents against mosquito vectors of disease, Medical and Veterinary Entomology, 1, 109–119, 1987.
    15. A. Badolo, E. Ilboudo-Sanogo, A. P. Ouedraogo, and C. Costantini, Evaluation of the sensitivity of
        Aedes aegypti and Anopheles gambiae complex mosquitoes to two insect repellents: DEET and KBR
        3023, Tropical Medicine & International Health, 9(3), 330–334, 2004.
    16. R. E. Coleman, L. L. Robert, L. W. Roberts, J. A. Glass, D. C. Seeley, A. Laughinghouse, P. V. Perkins,
        and R. A. Wirtz, Laboratory evaluation of repellents against four anopheline mosquitoes
        (Diptera: Culicidae) and two phlebotomine sand flies (Diptera: Psychodidae), Journal of Medical
        Entomology, 30, 499–502, 1993.
    17. R. A. Wirtz, E. D. Rowton, J. A. Hallam, P. V. Perkins, and L. C. Rutledge, Laboratory testing of
        repellents against the sand fly Phlebotomus papatasi (Diptera: Psychodidae), Journal of Medical
        Entomology, 23, 64–67, 1986.
    18. L. C. Rutledge, R. K. Gupta, G. N. Piper, and C. A. Lowe, Studies on the inheritance of
        repellent tolerance in Aedes aegypti, Journal of the American Mosquito Control Association, 10,
        93–100, 1994.
    19. I. H. Gilbert, H. K. Gouck, and N. Smith, Attractiveness of men and women to Aedes aegypti and relative
        protection time obtained with deet, Florida Entomologist, 49, 53–66, 1966.
    20. C. F. Golenda, V. B. Solberg, R. B. Burge, J. M. Gambel, and R. A. Wirtz, Gender-related efficacy
        difference to an extended duration formulation of topical N,N-diethyl-m-toluamide (deet), American
        Journal of Tropical Medicine and Hygiene, 60, 654–657, 1999.
    21. R. L. Reitschel and P. S. Spencer, Correlation between mosquito repellent protection time and insensible
        water loss from the skin, Journal of Investigative Dermatology, 65, 385–387, 1975.
    22. W. A. Skinner, H. C. Tong, H. Johnson, R. M. Parkhurst, D. Thomas, T. Spencer, W. Acres,
        D. Skidmore, and H. Maibach, Influence of human skin surface lipids on protection time of topical
        mosquito repellent, Journal of Pharmaceutical Science, 66, 1764–1766, 1977.


q 2006 by Taylor & Francis Group, LLC
258                                                       Insect Repellents: Principles, Methods, and Uses


    23. S. Selim, R. E. Hartnagel, T. G. Osimitz, K. L. Gabriel, and G. P. Schoenig, Absorption, metabolism, and
        excretion of N,N-diethyl-m-toluamide following dermal application to human volunteers, Fundamental
        and Applied Toxicology, 25, 95–100, 1995.
    24. R. K. Gupta and L. C. Rutledge, Controlled release repellent formulations on human volunteers under
        three climatic regimens, Journal of the American Mosquito Control Association, 7, 490–493, 1991.
    25. D. R. Barnard, Mediation of deet repellency in mosquitoes (Diptera: Culicidae) by species, age, and
        parity, Journal of Medical Entomology, 35(3), 340–343, 1998.
    26. A. Ahmadi and G. H. A. McLelland, Mosquito-mediated attraction of female mosquitoes to a host,
        Physiological Entomology, 10, 251–255, 1985.
    27. L. C. Rutledge, W. G. Reifenrath, and R. K. Gupta, Sustained release formulations of the U.S. Army
        insect repellent, Proceedings of the Army Sciences Conference, 3, 343–357, 1986.
    28. V. B. Solberg, T. A. Klein, K. R. McPherson, B. Bradford, J. R. Burge, and R. A. Wirtz, Field evaluation
        of deet and a piperidine repellent (A13-37220) against Amblyomma americanum (Acari: Ixodidae),
        Journal of Medical Entomology, 32, 870–875, 1995.
    29. L. M. Rueda, L. C. Rutledge, and R. K. Gupta, Effect of skin abrasions on the efficacy of the repellent
        deet against Aedes aegypti, Journal of the American Mosquito Control Association, 14, 178–182, 1998.
    30. L. C. Rutledge, R. A. Wirtz, M. D. Buesher, and Z. A. Mehr, Mathematical models of the effectiveness
        and persistence of mosquito repellents, Journal of the American Mosquito Control Association, 1,
        56–62, 1985.
    31. C. Costantini, A. Badolo, and E. Ilboudo-Sanogo, Field evaluation of the efficacy and persistence of
        insect repellents DEET, IR3535, and KBR 3023 against Anopheles gambiae complex and other
        Afrotropical vector mosquitoes, Transactions of the Royal Society of Tropical Medicine and
        Hygiene, 98(11), 644–652, 2004.
    32. C. E. Schreck, Techniques for the evaluation of insect repellents: A critical review, Annual Review of
        Entomology, 22, 101–119, 1977.
    33. M. D. Buesher, L. C. Rutledge, R. A. Wirtz, and J. H. Nelson, The dose-persistence relationship of deet
        against Aedes aegypti, Mosquito News, 43, 364–366, 1983.
    34. W. G. Reifenrath and L. C. Rutledge, Evaluation of mosquito repellent formulations, Journal of
        Pharmaceutical Science, 73, 169–173, 1983.
    35. Z. A. Mehr, L. C. Rutledge, E. L. Morales, V. E. Meixsell, and D. W. Corte, Laboratory evaluation of
        controlled-release insect repellent formulations, Journal of the American Mosquito Control Associ-
        ation, 1, 143–147, 1985.
    36. R. K. Gupta, A. W. Sweeney, L. C. Rutledge, R. D. Cooper, S. P. Frances, and D. R. Westrom,
        Effectiveness of controlled-release personal-use arthropod repellents and permethrin impreg-
        nated clothing in the field, Journal of the American Mosquito Control Association, 3, 556–560, 1987.
    37. L. C. Rutledge, R. K. Gupta, Z. A. Mehr, M. D. Buesher, and W. G. Reifenrath, The evaluation of
        controlled-release mosquito repellent formulations, Journal of the American Mosquito Control
        Association, 12, 39–44, 1996.
    38. B. Salafsky, Y. He, J. Li, T. Shibuya, and K. Ramaswami, Short report: Study on the efficacy of a new
        long-acting formulation of N,N-diethyl-m-toluamide (deet) for the prevention of tick attachment,
        American Journal of Tropical Medicine and Hygiene, 62, 169–172, 2000.
    39. M. S. Fradin and J. F. Day, Comparative efficacy of insect repellents against mosquito bites, New
        England Journal of Medicine, 347, 13–18, 2002.
    40. D. R. Barnard and R. D. Xue, Laboratory evaluation of mosquito repellents against Aedes albopictus,
        Culex nigripalpus, and Ochlerotatus triseriatus (Diptera: Culicidae), Journal of Medical Entomology,
        41(4), 726–730, 2004.
    41. J. K. Trigg and N. Hill, Laboratory evaluation of a eucalyptus-based repellent against four biting
        arthropods, Phytotherapy Research, 10, 313–316, 1996.
    42. N. Hill, Insect repellent test report, unpublished document, London School of Hygiene and Tropical
        Medicine, 1998.
    43. J. Govere, D. N. Durrheim, L. Baker, R. Hunt, and M. Coetze, Efficacy of three insect repellents
        against the malaria vector Anopheles arabiensis, Medical and Veterinary Entomology, 14,
        441–444, 2000.


q 2006 by Taylor & Francis Group, LLC
Evaluation of Topical Insect Repellents and Factors That Affect Their Performance                       259


    44. S. P. Carroll and J. E. Loye, A commercially-available botanical insect repellent as effective as
        deet, Journal of the American Mosquito Control Association, in press.
    45. U. Thavara, A. Tawatsin, J. Chompoosri, W. Suwonkerd, U. Chansang, and
        P. Asacadachanukorn, Laboratory and field evaulations of the insect repellent IR3535 (ethylbu-
        tylacetylaminopropionate) and deet against mosquito vectors in Thailand, Journal of the American
        Mosquito Control Association, 17, 190–195, 2001.
    46. J. K. Trigg, Evaluation of a eucalyptus-based repellent against Anopheles spp. in Tanzania,
        Journal of the American Mosquito Control Association, 12, 243–246, 1996.
    47. H. H. Yap, K. Jahangir, A. S. C. Chong, C. R. Adanan, N. L. Chong, Y. A. Malik, and
        B. Rohaizat, Field efficacy of a new repellent, KBR 3023, against Aedes albopictus (SKUSE) and
        Culex quinquefasciatus (SAY) in a tropical environment, Journal of Vector Ecology, 23, 62–68,
        1998.
    48. H. H. Yap, K. Janangir, and J. Zairi, Field efficacy of four insect repellent products against vector
        mosquitoes in a tropical environment, Journal of the American Mosquito Control Association, 16,
        241–244, 2000.
    49. D. R. Barnard, U. R. Bernier, K. H. Posey, and R. D. Xue, Repellency of IR3535, KBR3023,
        para-menthane-3,8-diol, and deet to black salt marsh mosquitoes (Diptera: Culicidae) in the
        Everglades National Park, Journal of Medical Entomology, 39(6), 895–899, 2002.
    50. S. J. Moore, A. Lenglet, and N. Hill, Field evaluation of three plant-based insect repellents
        against malaria vectors in Vaca Diez province, the Bolivian Amazon, Journal of the American
        Mosquito Control Association, 18, 107–110, 2002.
    51. S. P. Frances, N. V. Dung, N. W. Beebe, and M. Debboun, Field evaluation of repellent
        formulations against daytime and nighttime biting mosquitoes in a tropical rainforest in northern
        Australia, Journal of Medical Entomology, 39, 541–544, 2002.
    52. L. C. Rutledge and R. K. Gupta, Variation in the protection periods of repellents on individual
        human subjects: An analytical review, Journal of the American Mosquito Control Association, 15,
        348–355, 1999.




q 2006 by Taylor & Francis Group, LLC
q 2006 by Taylor & Francis Group, LLC
13
Repellents Used in Fabric: The Experience
of the U.S. Military


Wilfred C. McCain and Glenn J. Leach



CONTENTS
Introduction ...................................................................................................................................261
Repellents and Fabric Treatment ..................................................................................................262
Repellent Application to Fabric....................................................................................................265
  Hand Application.......................................................................................................................266
  Barrier Method ..........................................................................................................................266
  Spray Method.............................................................................................................................267
  Dust Method ..............................................................................................................................267
  Immersion Method ....................................................................................................................267
  Factory Pretreatment..................................................................................................................267
  Long-Lasting Surface Treatment of Fabrics with Insecticides ................................................268
  Olysetw Technology and Production ........................................................................................269
  Industrial Production of Long-Lasting Insecticide-Treated Yarns...........................................269
  Safety of Clothing Treatments ..................................................................................................269
References .....................................................................................................................................271




Introduction
Humans are the only mammals that routinely wear clothing. The obvious functional comparison is
between our clothing and the pelage of mammals or birds. Just as our use of clothing varies with our
intended activity, the nature of fur or feathers can vary with specific adaptation of a species and from
season to season for a single species. Considering this comparison, it is perhaps not surprising that many
of the arthropods that bite humans interact with clothing. Chigger mites and tropical rat mites tend to bite
where clothing is most closely pressed to the skin. Mosquitoes and stable flies commonly bite right
through clothing, especially when the weave creates openings and the cloth lays flat on the skin. Body
lice actually require clothing for resting and oviposition sites. Complete systems for personal protection
from biting arthropods must consider the means to enhance clothing as a barrier, as well as protect
exposed skin by the use of topical repellent products.



                                                                                                                                             261

q 2006 by Taylor & Francis Group, LLC
262                                                     Insect Repellents: Principles, Methods, and Uses




Repellents and Fabric Treatment
In May of 1942, Philip Granett, working under a National Carbon Company Fellowship at Rutgers
University, found that a cheesecloth sleeve impregnated with the repellent butyl carbitol acetate
protected an untreated arm against mosquitoes for 24 h. In progress reports he sent to the company,
he stated that some materials were effective for longer periods of time on cloth than on skin. Preliminary
laboratory and field tests conducted in Florida in 1942 demonstrated that mosquitoes avoided some
treated fabrics for several days; whereas, repellents applied to the skin failed to give complete protection
after a few hours.1 This finding started a new era in the field of personal protection from arthropods of
public health importance. Madden and Lindquist2 in 1944 and by McCullough and Jones3 in 1945
summarized studies showing the relative effectiveness of several repellents when applied to clothing.
   Prior to World War II, reports from the Panama Canal and other tropical areas prompted the
Quartermaster General’s Office to review the entire problem of “adaptation of clothing to the physiologic
requirements of the soldier.”4 In 1941, The National Research Council approved support for the Harvard
Fatigue Laboratory to investigate “clothing, fatigue, and supplementary substances.”
   Chemical substances that were to be impregnated into clothing materials to protect against chemical
warfare agents, fire, and insects had to be effective, nonirritating and nontoxic. The Surgeon General
tasked the Orlando Laboratory of the United States Department of Agriculture (USDA) to develop
methods to control arthropods of medical interest. This resulted in a massive effort to quickly identify and
appraise the potential of chemical compounds and mixtures. The USDA in conjunction with a number of
other agencies including the U.S. Army, U.S. Navy, Rockefeller Foundation, Tennessee Valley
Authority, the U.S. Public Health Service, and various groups working under contract to the Office of
Scientific Research and Development entered into this effort. Collaborative research with other allied
nations was also conducted and a vast amount of information was exchanged. Industry submitted most of
the chemicals tested for their effectiveness against a variety of insects and arthropods.
   During World War II, scientists evaluated more than 10,000 such materials. By 1950, more than
20,000 chemicals and mixtures had been tested for efficacy against arthropods and more than 6,000 had
been tested as clothing treatments.5 Of these, only a handful met the criteria of being both effective and
safe. The Army selected some of these for general use.
   During this period, the USDA had scientists working under a fellowship that Rutgers University
organized with Philip Granett in charge. The fellowship funded the development of an economic
mosquito repellent that would be effective, safe, and acceptable for application to the skin. This
collaboration resulted in the development of Insect Repellent 612 (2-ethyl-1,3-hexanediol) that was
seven times more effective than citronella and superior to it in other respects. The repellent properties of
another compound, dimethyl phthalate, had been discovered by another Rutgers Fellowship sponsored
by the Standard Oil Company of New Jersey in the late 1930s and a third compound, indalone (2,2-
dimethyl-6-carbobutoxy-2,3-dihydro-4-pyrone), was originally tested by the fellowship in 1937.
Researchers found these agents effective against mites as well as several other types of arthropods.6
   Extensive testing both in the laboratory and in the field followed the development of dimethyl
phthalate, indalone, Rutgers 612, as well as dibutyl phthalate. Madden and his colleagues2 reported the
findings of field studies conducted in Orlando in 1942, using 28 different materials or combination of
materials including 10 kinds of dusts, 1 ointment, and 3 liquid formulations as skin and clothing
repellents. Skin testing was discontinued when it was demonstrated that liquid repellents applied to
clothing were far more effective for protection from mites, considered key vectors at the time because of
the deadly scourge of scrub typhus in the Pacific Theater. Scientists found indalone, Rutgers 612, and
dimethyl phthalate all effective in protecting individuals against mites. The application of any of these
three compounds in solvents to fabrics gave protection for up to 39 d and undiluted dimethyl phthalate
protected people for 59 d, especially when applied around the openings of clothing. This led to the army’s
adoption of dimethyl phthalate to control scrub typhus. It was applied as a spray to the clothing or used as


q 2006 by Taylor & Francis Group, LLC
Repellents Used in Fabric: The Experience of the U.S. Military                                           263


an emulsion in which clothing was dipped and then dried. Bulk supply of dimethyl phthalate in gallon
containers for application to clothing was recommended in the fall of 1943 for certain of the overseas
theaters. In December 1944, it was recommended that an emulsifier be incorporated with the dimethyl
phthalate to facilitate the preparation of emulsions in the field.7 In 1945, members of the United States of
America Typhus Commission working in New Guinea developed a field method for the impregnation of
clothing with a 5% emulsion of dimethyl phthalate in soapy water. This treatment was used in endemic
areas for protection of both combat and staging troops. Research found this more practicable than
previously used methods of applying repellents to clothing and was widely employed as a preventive
measure to protect troops from scrub typhus in the western Pacific.
   Because chemicals to manufacture dimethyl phthalate were in short supply during World War II,
benzyl benzoate was used as a miticide and eventually became the standard by which researchers
evaluated future clothing repellents. As investigations of repellents continued, scientists discovered a
number of substances that were superior to dimethyl phthalate for impregnation of clothing to protect
against larval mites. They selected benzyl benzoate because of its rapid action against mites and its
persistence in clothing after laundering. It was used in various mixtures for impregnation of clothing to
repel mosquitoes, gnats, ticks and flies).
   In March 1945, a recommendation was made to the Office of the Quartermaster General that benzyl
benzoate, together with an emulsifier, be substituted for dimethyl phthalate in the bulk issue of insect
repellent. Because of difficulties in procurement, supplies of benzyl benzoate did not reach the field in
time to be of use before the end of the war. It was used through the action in Vietnam to control and repel
certain ticks and mites8 and was available for clothing application as 90% benzyl benzoate.
   With the exception of occasional cases of skin irritation, few adverse effects have been reported from
the use of benzyl benzoate. The efficiency of skin absorption is not known. Absorbed benzyl benzoate is
rapidly biotransformed to hippuric acid that is excreted in the urine. When given in large doses to
laboratory animals, benzyl benzoate causes excitement, lack of coordination, paralysis of the limbs,
convulsions, respiratory paralysis, and death. No human poisonings have been reported.9 The U.S.
Environmental Protection Agency (EPA) is currently reviewing data submitted by the producers
regarding benzyl benzoate’s human health and environmental effects. This data will be used to determine
the pesticide’s eligibility for reregistration and will result in a Reregistration Eligibility Decision (RED)
document.10
   To develop a repellent with efficacy against a greater variety of arthropods, scientists used a
combination of repellents. Formula 6-2-2 (M-250), a mixture containing 60% dimethyl phthalate,
20% Rutgers 612, and 20% indalone, proved to be a highly effective repellent. The 6-2-2 formula was
tested as a clothing treatment and compared with both dimethyl phthalate and Rutgers 612 in 1942–
1943.1 The testing showed that there was virtually no difference in protection among the three
products tested.
   Another mixture of several effective repellents was developed specifically as a clothing repellent at the
Orlando laboratory around this time and given the code number M-1960. This mixture contained 30%
each of 2-butyl-2-ethyl-1,3-propanediol, benzyl benzoate, N-butylacetanilide and 10% of a surfactant,
Tween 80. This combination proved to be effective against mosquitoes, biting flies, fleas, mites, and ticks
when applied to clothing at 3.9 mg/cm2. M-1960 was applied to clothing in the Pacific Theater to prevent
the devastating effects of scrub typhus. The use of M-1960 as a clothing impregnant continued
throughout the Korean and Viet Nam wars (repellent, clothing application, M-1960, 1-gal can) on a
limited basis. Because of its tendency to irritate the skin and other reasons, M-1960 and benzyl benzoate
were not well accepted by personnel. M-1960 also had a disagreeable odor that many soldiers and their
commanders did not like.11 Their effectiveness as repellents was only useful if the uniforms were
impregnated. For this reason efforts were undertaken to replace M-1960 in 1974. A new effective
clothing repellent was not fielded until 1991, when permethrin came into use.
   The U.S. Military recognized in the late 1980s that it needed to replace M-1960 and that the standard
topical repellent (75% alcohol solution of deet) had drawbacks. A major research program centered at the
Letterman Army Institute of Research in San Francisco raised new possibilities for repellent formulation


q 2006 by Taylor & Francis Group, LLC
264                                                      Insect Repellents: Principles, Methods, and Uses


and generated greater expectations for personal protection products among the military entomological
community. These expectations were translated in 1987 to a powerful document known as Joint Service
Operational Requirement (JSOR). This document listed 11