Document Sample
_Fish_Physiology_12_Author_Unknown-The_Cardiovascular_System__The_Cardiovascular_System._Part_B-Academic_Press_1992_ Powered By Docstoc

The Cardiovascular System

Edited by
W. S. H O A R

D. J. R A N D A L L

A. P. F A R R E L L

The Cardiovascular System

Harcourt Brace Jovanovich, Publishers
San Diego New York Boston London        Sydney   Tokyo Toronto
This book is printed on acid-free paper. @

Copyright 0 1992 by ACADEMIC PRESS, INC.
All Rights Reserved.
No part of this publication may be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopy, recording, or any information
storage and retrieval system, without permission in writing from the publisher.

Academic Press, Inc.
1250 Sixth Avenue, San Diego, California 92101431 1

United Kingdom Edition published by
Academic Press Limited
24-28 Oval Road, London NWl 7DX

Library of Congress Cataloging-in-PublicationData
(revised for vol. 12)

Hoar, William Stewart, date
  Fish physiology.

   Vols.8-12 edited by W.S. Hoar [et al.]
   Includes bibliographies and index.
   Contents: v. 1. Excretion, ionic regulation, and
metabolism -- [etc.] -- v. 11. The physiology of
developing fish. pt. B. Viviparity and posthatching
juveniles -- v. 12, pt. A-B. The cardiovascular system.
   1. Fishes--Physiology--Collectedworks. I. Randall,
David J,. date. 11. Conte, Frank P., date.
111. Title.
QL639.1.H6         597.01       76-84233
ISBN 0-12-350436-8 (v. 12B)

92 93 94     95 96 91         MV   9   8 1 6 5 4 3 2         1

       OF    A                                     ix

CONTRIBUTORS                                      xi
       OF     VOLUMES                            xvii

1.     Fish Blood Cells
       Ragnar Fange
     I. Introduction                               2
     11. Red Cells: Morphology                     3
 111. Red Cells: Physiology and Biochemistry       8
  IV. White Cells: Morphology                     14
   V. White Cells: Physiology and Biochemistry    26
  VI. Lymphomyeloid Tissues                       36
 VII. Hemopoiesis                                 39
VIII. Future Research                             42
      References                                  46

2.      Chemical Properties of the Blood
        D. G. McDonald and C . L. Milligan
    I. Introduction                                56
   11. Hormones: Teleosts                          60
 111. Hormones: Cyclostomes and Chrondricthyes     74
  IV. Metabolites                                  76
   V. Nonprotein Nitrogenous Compounds             80
  VI. Plasma Proteins                              87
 VII. Lipids                                       96
VIII. Electrolytes                                106
       References                                 113
vi                                                               CONTENTS

3.      Blood and Extracellular Fluid Volume Regulation: Role
        of the Renin-Angiotensin System, Kallikrein-Kinin System,
        and Atrial Natriuretic Peptides
        Kenneth R. Olson
      I. Introduction                                                 136
     11. Fluid Compartments                                           136
 111.    Renin-Angiotensin System                                     193
 IV.     Kallikrein-Kinin System                                      213
  V.     Atrial Natriuretic Peptides                                  217
 VI.     Summary                                                      23 1
         References                                                   232

4. Catecholamines
   D. J. Randall and S. F . Perry
       I. Catecholamine Metabolism                                    255
      11. Control of Blood Catecholamine Levels                       263
     1 1 Actions of Circulating Catecholamines
      1.                                                              275
     IV. Factors Influencing Actions of Catecholamines                287
          References                                                  290

5.      Cardiovascular Control by Purines, 5-Hydroxytryptamine,
        and Neuropeptides
        Stefan Nilsson and Susanne Holmgren
       I. Introduction                                                 30 1
      11. Origin of Vasomotor and Cardiac Nerves                       303
     111. Purines                                                      307
     IV. 5-Hydroxytryptamine (Serotonin)                               311
      V. Neuropeptides                                                 317
     VI. Endothelial Factors                                           33 1
          References                                                   333

6.      Nervous Control of the Heart and
        Cardiorespiratory Interactions
        E . W. Taylor
       I. Introduction                                                343
      11. Innervation of the Heart                                    344
     111. The Central Location of Vagal Preganglionic Neurones        3.50
CONTENTS                                                            vii

 IV. Control of the Heart and Branchial Circulation                 360
  V. Cardiorespiratory Interactions                                 37 1
 VI. Cardiorespiratory Synchrony                                    375
     References                                                     38 1

7.    Afferent Inputs Associated with Cardioventilatory Control
      in Fish
      Mark L. Burleson, Neal J . Smatresk, and William K . Milsom
   I. Introduction                                                  390
  11. Mechanoreceptors                                              390
 111. Chemoreceptors                                                404
 IV. Nociceptors                                                    416
  V. Central Projections of Sensory Neurons                         419
      References                                                    420

AUTHORINDEX                                                         427
         INDEX                                                      453
      INDEX                                                         463
This Page Intentionally Left Blank
                     CONTENTS OF PART A

The Heart
  Anthony P . Farrell and David R . Jones
The Arterial System
   P . G . Busnell, David R.Jones, and Anthony P . Farrell
The Venous System
   Geoffrey H. Satchel1
The Secondary Vascular System
  J. F . Steffensen andJ. P. Lornholt
Cardiac Energy Metabolism
   William R. Driedzic
Excitation-Contraction Coupling in the Teleost Heart
   Glen F . Tibbits, Christopher D . Moyes, and Leif Hove-Madsen
Author Index-Systematic     Index-Subject    Index

This Page Intentionally Left Blank

Numbers in parentheses indicate the pages on which the authors’ contributions begin.

Mark L. Burleson (390), Department of Biology, University of Texas
  at Arlington, Arlington, Texas 76019
Ragnar Fange (2),Department of Zoophysiology, University of Gote-
  borg, S-40031, Goteborg, Sweden
Susanne Holmgren (301),Department of Zoophysiology, University of
   Goteborg, S-40031, Goteborg, Sweden
D. G. McDonald (56), Department of Biological Sciences, McMaster
   University, Hamilton, Ontario, Canada L8S 4 K 1
C. L. Milligan (56), Department of Zoology, University of Western
   Ontario, London, Ontario, Canada N6A 5B7
William K. Milsom (390), Department of Zoology, University of Brit-
   ish Columbia, Vancouver, British Columbia, Canada V6T 2A9
Stefan Nilsson (301), Department of Zoophysiology, University of
   Goteborg, S-40031, Goteborg, Sweden
Kenneth R. Olson (136), South Bend Center f o r Medical Education,
  lndiana University School of Medicine, University of Notre Dame,
  Notre Dame, Indiana 46556
S. F. Perry (255), Department of Biology, University of Ottawa, Ot-
   tawa, Ontario, Canada K 1 N 6N5
D. J. Randall (255), Department of Zoology, University of British
   Columbia, Vancouver, British Columbia, Canada V6T 2A9
Neal J . Smatresk (390),Department of Biology, University of Texas at
  Arlington, Arlington, Texas 76019
E . W. Taylor (343), Department of Zoology and, Comparative
    Physiology, University of Birmingham, Birmingham B15 2TT,
    United Kingdom
This Page Intentionally Left Blank

   A considerable amount of new information has accumulated in
recent years concerning the cardiovascular system of fishes. As a result
we now have a better understanding of the cardiovascular diversity
among fishes, and a number of unifying concepts have emerged re-
garding both design and function. Our present understanding of the
cardiovascular system is presented in Volumes XIIA and XIIB.
    Fish are the most successful vertebrate group both in terms of
biomass and number of species. They also occupy a wide range of
environments. As a result, the basic cardiovascular design shows a
multiplicity of modifications. As in all vertebrates, it appears that the
underlying tenet is that the design of the cardiovascular system pri-
marily reflects the need for oxygen transfer. In fact, the influence of
activity pattern on cardiovascular design is such that correlations often
transcend phylogeny. Thus, we find that fish with a higher oxygen
consumption also have hearts that are bigger, have a more complex
anatomy, beat faster, and generate higher blood pressures (Chapters
1A and 2A). Cardiac metabolism and excitation-contraction coupling
are correspondingly fine-tuned to these overall demands (Chapters 5A
and 6A).
    Earlier studies were essentially descriptive and drew from our
knowledge of mammalian cardiovascular systems. The inherent dan-
ger of this approach is that similarities between systems tend to be
emphasized and what is special is often ignored. Fish live in a different
environment from most mammals, one in which the effects of gravity
are relatively minor because the fish has a density similar to the me-
dium. Instead fish must meet the challenge of moving through this
viscous medium.
   The circulatory system of fish is divided into primary and secondary
circulations (Chapters 2A, 3A, and 4A). Fish do not have a lymphatic
system. The primary circulation consists of branching arterial, capil-
lary, and venous networks. The secondary circulation arises from nar-
row vessels that connect with primary arteries. This secondary circu-
xiv                                                            PREFACE

lation is a low-pressure and low-hematocrit system serving a primarily
nutritive rather than respiratory function to surface structures that
exchange gases directly with the water (Chapter 3A). Consequently,
the secondary circulation is particularly prone to the hydrodynamic
forces acting on the body surface.
    As a fish moves forward, pressure waves pass backward down the
body squeezing blood beneath the skin toward the tail. This is a major
problem for the design of the venous system, analogous to gravitational
effects on the circulation of terrestrial vertebrates. Thus, the venous
system, into which the secondary circulation empties, incorporates a
number of accessory hearts to aid in the return of blood to the branchial
heart via the central core of the body, which is less influenced by
surface pressure waves (Chapter 4A).
    The cellular components of fish blood are well established but the
mechanisms involved in blood cell production, differentiation, and
release are still being defined (Chapter 1B). With respect to plasma, its
ionic composition is relatively well documented (Chapter 2B).
However, fish must cope with periodic changes in their environment,
especially light and temperature, and in some species, salinity. Many
of the mechanisms for responding to these changes involve the endo-
crine system. Thus there are circadian and seasonal variations in blood
hormone levels, as well as many other components. We are in the
process of describing these variations (Chapter 2B), but the nature of
the control systems governing these circadian and seasonal rhythms is,
in most cases, vague.
    The control of blood volume and its effect on venous return to the
heart are intriguing questions. Some fish are tight skinned, such as tuna
and flat fish. Others, such as the sea raven, are baggy skinned, probably
so they can gorge meals that are about 50% of their body weight,
presumably without raising intraperitoneal pressure which would af-
fect venous return to the heart. Whether or not the tight skin of, for
example, tuna has a functional parallel with the encapsulation of the
mammalian kidney is not clear. That is, the volume of a fish may be
limited by the lack of distensibility of the skin, requiring only systems
that keep the body inflated. It appears, however, that fish do have
mechanisms for monitoring venous pressure (Chapter 7B), but it is not
 known if such mechanisms are linked to the control of blood volume.
 Fish also possess a renin-angiotensin system and atrial natriuretic
 peptides, but again exactly what role they play in volume regulation is
 not known (Chapter 3B). The whole question of fluid exchange across
 capillary walls and the regulation of blood volume in fish remains
 largely unanswered.
PREFACE                                                               xv

    Cardiovascular regulation centers around control of the heart’s ac-
tivity, modulation of central blood pressure, and alterations to vascular
resistance to effect regional control ofblood flow (Chapters 1A and 2A).
Very little is known about the control of blood flow through capillaries
with the exception of gill lamellae. Gill blood flow was reviewed
recently in Volume X and, therefore, is not discussed in the present
volume. Because of the paucity of information on other capillary beds
in fish, we have not reviewed the subject in this volume. The emerging
and complex area of vasoactive peptides and their associated nerves,
however, has been reviewed (Chapter 5B).
    Respiratory and cardiac control are intimately coupled in verte-
brates, and perhaps most obviously in fish. There is a sequential
grouping of neurons in the central nervous system driving ventilation,
with the most posterior neurons involved in cardiac control. This lin-
ear arrangement of neurons may allow coupled rhythm generation to
be more easily studied than in other vertebrate groups. An understand-
ing of peripheral receptors involved in the control of respiration and
circulation is gradually evolving. Both peripheral and central nervous,
as well as humoral, control are reviewed in this volume (Chapters 4B,
5B, 6B, and 7B).
    We hope that this volume of Fish Physiology sheds some light on
these problems.
                                                    W. S. HOAR
                                                    D. J. RANDALL
                                                    A. P. FARRELL
This Page Intentionally Left Blank

Volume I
The Body Compartments and the Distribution of Electrolytes
   W. N . Holmes and Edward M . Donaldson
The Kidney
   Cleveland P . Hickman, Jr., and Benjamin F . Trump
Salt Secretion
    Frank P . Conte
The Effects of Salinity on the Eggs and Larvae of Teleosts
   F . G . T . Holliday
Formation of Excretory Products
   Roy P . Forster and Leon Goldstein
Intermediary Metabolism in Fishes
    P . W. Hochachka
Nutrition, Digestion, and Energy Utilization
   Arthur M . Phillips,Jr.

Volume I1
The Pituitary Gland: Anatomy and Histophysiology
  J . N . Ball and Bridget I . Baker
The Neurohypophysis
  A. M . Perks
Prolactin (Fish Prolactin or Paralactin) and Growth Hormone
   J . N . Ball
                                               CONTENTS OF OTHER VOLUMES

Thyroid Function and Its Control in Fishes
   Aubrey Gorbman
The Endocrine Pancreas
  August Epple
The Adrenocortical Steroids, Adrenocorticotropin and the Corpuscles
of Stannius
    I . Chester Jones, D. K . 0 .Chan, I . W. Henderson, and J . N . Ball
The Ultimobranchial Glands and Calcium Regulation
   D. Harold Copp
Urophysis and Caudal Neurosecretory System
   Howard A . Bern

Volume 111
   William S. Hoar
Hormones and Reproductive Behavior in Fishes
   N . R. Liley
Sex Differentiation
   Toki-o Yamamoto
Development: Eggs and Larvae
  J . H . S. Blaxter
Fish Cell and Tissue Culture
   Ken Wolf and M . C . Quimby
Chromatophores and Pigments
   Ryozo Fujii
   J . A. C. Nicol
 Poisons and Venoms
    Findlay E. Russell

Volume IV
Anatomy and Physiology of the Central Nervous System
  Jerald J . Berstein
CONTENTS OF OTHER VOLUMES                               xix

The Pineal Organ
  James Clarke Fenwick
Autonomic Nervous System
   Graeme Campbell
The Circulatory System
   D . J . Randall
Acid-Base Balance
   C . Albers
Properties of Fish Hemoglobins
   Austen R i g s
Gas Exchange in Fish
   D . J . Randall
The Regulation of Breathing
   G . Shelton
Air Breathing in Fishes
    Kjell Johansen
The Swim Bladder as a Hydrostatic Organ
  Johan B . Steen
Hydrostatic Pressure
  Malcolm S . Gordon
Immunology of Fish
  John E . Cushing

 Volume V
 Vision: Visual Pigments
     F . W. Munz
 Vision: Electrophysiology of the Retina
     T . Tomita
 Vision: The Experimental Analysis of Visual BehavioI
    David lngle
     Toshiaki J . Hara
  Temperature Receptors
    R . W. Murray
xx                                          CONTENTS OF OTHER VOLUMES

Sound Production and Detection
   William N . Tavolga
The Labyrinth
   0. Lowenstein
The Lateral Organ Mechanoreceptors
   Ake Flock
The Mauthner Cell
  J. Diamond
Electric Organs
   M . V. L. Bennett
   M . V. L. Bennett

Volume VI
The Effect of Environmental Factors on the Physiology of Fish
   F. E .J . Fry
Biochemical Adaptation to the Environment
   P . W. Hochachka and G . N . Somero
Freezing Resistance in Fishes
   Arthur L. DeVries
Learning and Memory
   Henry Gleitman and Paul Rozin
The Ethological Analysis of Fish Behavior
   Gerard P . Baerends
Biological Rhythms
   Horst 0. Schwassmann
Orientation and Fish Migration
   Arthur D. Hasler
 Special Techniques
    D. J. Randall and W. S. Hoar

Volume VII
Form, Function, and Locomotory Habits in Fish
   C. C. Lindsey
Swimming Capacity
  F . W . H . Beamish
Hydrodynamics: Nonscombroid Fish
  Paul W .Webb
Locomotion by Scombrid Fishes: Hydromechanics, Morphology,
and Behavior
   JohnJ . Magnuson
Body Temperature Relations of Tunas, Especially Skipjack
   E . Don Stevens and William H . Neil1
Locomotor Muscle
   Quentin Bone
The Respiratory and Circulatory Systems during Exercise
  David R. Jones and David J . Randall
Metabolism in Fish during Exercise
  William R . Driedzic and P . W . Hochachka

Volume VIII
   C. B . Cowey and J . R. Sargent
Feeding Strategy
   Kim D . Hyatt
The Brain and Feeding Behavior
   Richard E . Peter
   Ragner Fange and David Groue
Metabolism and Energy Conversion during Early Development
   Charles Terner
Physiological Energetics
   J . R . Brett and T . D. D. Groves
xxii                                        CONTENTS OF OTHER VOLUMES

   J. R. Gold
Population Genetics
   Fred W. Allendorfand Fred M . Utter
Hormonal Enhancement of Growth
   Edward M . Donaldson, Ulf H . M . Fagerlund, David A. Higgs,
   a n d J . R . McBride
Environmental Factors and Growth
  1.R. Brett
Growth Rates and Models
   W. E . Ricker

Volume IXA
Reproduction in Cyclostome Fishes and Its Regulation
   Aubrey Gorbman
Reproduction in Cartilaginous Fishes (Chondrichthyes)
  J . M . Dodd
The Brain and Neurohormones in Teleost Reproduction
   Richard E . Peter
The Cellular Origin of Pituitary Gonadotropins in Teleosts
   P . G . W.]. van Oordt and]. Peute
Teleost Gonadotropins: Isolation, Biochemistry, and Function
   David R . Idler and T . Bun N g
The Functional Morphology of Teleost Gonads
   Yoshitaka Nagahama

The Gonadal Steroids
  A . Fostier, B.]alabert, R. Billard, B. Breton, and Y. Zohar
Yolk Formation and Differentiation in Teleost Fishes
   T . Bun N g and David R. Idler
An Introduction to Gonadotropin Receptor Studies in Fish
   Glen V a n Der Kraak

CONTENTS OF OTHER VOLUMES                                             xxiii

Volume IXB
Hormones, Pheromones, and Reproductive Behavior in Fish
   N . R. Liley and N . E . Stacey
Environmental Influences on Gonadal Activity in Fish
   T .J . Lam
Hormonal Control of Oocyte Final Maturation and Ovulation in Fishes
   Fredrick W. Goetz
Sex Control and Sex Reversal in Fish under Natural Conditions
   S . T. H . Chan and W. S . B . Yeung
Hormonal Sex Control and Its Application to Fish Culture
   George A. Hunter and Edward M . Donaldson
Fish Gamete Preservation and Spermatozoan Physiology
   Joachim Stoss
Induced Final Maturation, Ovulation, and Spermiation in Cultured Fish
   Edward M . Donaldson and George A. Hunter
Chromosome Set Manipulation and Sex Control in Fish
   Gary H . Thorgaard

Volume XA
General Anatomy of the Gills
  George Hughes
Gill Internal Morphology
    Pierre Laurent
Innervation and Pharmacology of the Gills
   Stefan Nilsson
Model Analysis of Gas Transfer in Fish Gills
  Johannes Piiper and Peter Scheid
Oxygen and Carbon Dioxide Transfer across Fish Gills
  David Randall and Charles Daxboeck
Acid-Base Regulation in Fishes
   Norbert Heisler
xxiv                                            CONTENTS OF OTHER VOLUME!

Physicochemical Parameters for Use in Fish Respiratory Physiology
   Robert G . Boutilier, Thomas A. Heming, and George K . Iwama

Volume XB
Water and Nonelectrolyte Permeation
  Jacques lsaia
Branchial Ion Movements in Teleosts: The Role of Respiratory and
Chloride Cells
   P. Payan,]. P. Girard, and N . Mayer-Gostan
Ion Transport and Gill ATPases
   Guy de Renzis and Michel Bornancin
Transepithelial Potentials in Fish Gills
   W. T . W. Potts
The Chloride Cell: The Active Transport of Chloride and the
Paracellular Pathways
   1.A. Zadunaisky
Hormonal Control of Water Movement across the Gills
  J . C . Rankin and Liana Bolis
Metabolism of the Fish Gill
  Thomas P. Mommsen
The Roles of Gill Permeability and Transport Mechanisms in Euryhalinity
   David H . Evans
The Pseudobranch: Morphology and Function
   Pierre Laurent and Suzanne Dunel-Erb
Perfusion Methods for the Study of Gill Physiology
   S . F . Perry, P. S . Davie, C . Daxboeck, A . G. Ellis, and D. G. Smith

Volume XIA
Pattern and Variety in Development
   J . H. S . Blaxter
 Respiratory Gas Exchange, Aerobic Metabolism, and Effects of Hypoxia
 during Early Life
    PeterJ. Rombough
CONTENTS OF OTHER VOLUMES                                         xxv

Osmotic and Ionic Regulation in Teleost Eggs and Larvae
  D. F . Alderdice
Sublethal Effects of Pollutants on Fish Eggs and Larvae
   H . von Westernhagen
Vitellogenesis and Oocyte Assembly
   Thomas P . Mommsen and Patrick]. Walsh
Yolk Absorption in Embryonic and Larval Fishes
   Thomas A . Heming and Randal K . Buddington
Mechanisms of Hatching in Fish
  Kenjiro Yamagami

Volume XIB
The Maternal-Embryonic Relationship in Viviparous Fishes
  John P . Worums, Bryon D. Grove, and Julian Lombardi
First Metamorphosis
   John H . Youson
Factors Controlling Meristic Variation
   C. C. Lindsey
The Physiology of Smoking Salmonids
   W. S. Hoar
Ontogeny of Behavior and Concurrent Developmental Changes in Sensory
Systems in Teleost Fishes
   David L. G. Noakes andlean-Guy].Godin
This Page Intentionally Left Blank

Department of Zoophysiology
University of Goteborg
Goteborg, Sweden

   I. Introduction
  11. Red Cells: Morphology
      A. Hematocrit
       B. Shape and Size
      C. Cytoplasmic Structures
      D. Immature and Senescent Cells
 111. Red Cells: Physiology and Biochemistry
       A. Erythrocyte Homeostasis
       B. Metabolism
      C. Proteins, Phosphates, and Nitrogen Metabolites
      D. Membrane Properties
       E. Osmotic Fragility
       F. Gas Transport
 IV. White Cells: Morphology
       A. Occurrence
       B. Staining Methods, Classification
       C. Granulocytes
       D. Mast Cells, PAS-Positive Granulocytes
       E. Lymphocytes, Plasma Cells
       F. Monocytes and Macrophages
       G. Thrombocytes or Spindle Cells
       H. Blast Cells
   V. White Cells: Physiology and Biochemistry
       A. Leucocyte Homeostasis
       B. Phagocytosis
       C. Granulocytic Defense Mechanisms
       D. Lymphocytic Functions, Immune Responses
       E. Pathology, Inflammation
       F. Thrombocytes and Blood Coagulation
  VI. Lymphomyeloid Tissues
       A. Thymus
       B. Spleen
       C. Kidney
FISH PHYSIOLOGY, VOL. XlIB                            Copyright 0 1992 by Academic Press, Inc.
                                                 All rights of reproduction in any form reserved.
2                                                       RAGNAR FANGE

      D. Lymphocytic Infiltrations
      E. Granulo(cyto)poieticTissues
      F. Melanomacrophage Centers
VII. Hemopoiesis
      A. Stem Cells
      B. Tissue Microenvironment
      C. Factors Stimulating Hemopoiesis
      D. Erythropoiesis in the Peripheral Blood
      E. Toxic Effects on Erythropoiesis
VIII. Future Research
      A. Gaps in Knowledge
      B. Hemopoiesis
      C. Lymphocyte Functions
      D. Blood Coagulation
      E. Granulocytic Function
      F. Electron Microscopy
      C. Immune System of Long-Lived Fishes
      H. Microcirculation of Hemopoietic Tissues
       I. Cell Interactions


   Aquatic vertebrates originated hundreds of millions of years ago
and evolved in different directions (Jarvik, 1980; Bjerring, 1985). In
spite of systematic diversity (Nelson, 1984) all fishes possess two main
types of blood cells, erythrocytes (red cells) and leucocytes (white
cells), a property shared by the land-living vertebrates, which are
derived from early fishlike ancestors. Detailed studies, however, re-
veal considerable variations in the structure and function of the blood
cells between different groups of fishes.
    Drzewina’s (1911) thorough description of fish granulocytes was
based on her own microscopic studies of68 species, but she and Jordan
(1938), and Grodzinski and Hoyer (1938) in their article on compara-
tive hematology, refer to many previous authors. A bibliography on fish
hematology was assembled by Hawkins and Mawdesley-Thomas
(1972). Rowley et al. (1988) have put together an extensive work on fish
blood cells illustrated by numerous electron microscopic photos. Com-
parative hematology is discussed by Andrew (1965)and Ratcliffe and
Millar (1988).Yasutake and Wales (1983),in an atlas of the microscopic
anatomy of salmonids, include a chapter on blood cells. Ivanova (1983)
has illustrated the microscopic morphology of different stages of fish
blood cells, and Golovina and Trombitsky (1989) have treated some
hematological parameters of freshwater fishes. Methods in fish hema-
1. FISH BLOOD   CELLS                                                  3

tology are described by Blaxhall and Daisley (1973) and Houston
(1990). Ellis (1977) has reviewed works on fish leucocytes dealing with
morphology, staining characteristics, physiology, biochemistry, immu-
nology, and the relationships of mast cells and eosinophils. Hine et al.
(1987) describe enzyme cytochemistry of leucocytes of different spe-
cies of fishes. Nikinmaa (1990)has published an excellent monograph
on functional aspects of vertebrate erythrocytes that includes material
on fish red blood cells.
    Articles or books on fish immunology are written by Cushing
(1970), Anderson (1974), Corbel (197~4,    Marchalonis (1977), Ingram
(1980),Manning and Tatner (1985), and Ellis (1989). Cooper (1983), in
a textbook on general immunology, also treats fishes. Litman et al.
(1990) discuss the function of immunoglobulin genes in lower verte-


A. Hematocrit
   The concentration of erythrocytes in the blood can b e expressed as
hematocrit or as the number of red cells per volume blood. The hemat-
ocrit values of fish blood range from almost 0 to more than 50% in
actively swimming, surface-feeding species (Table I). I n most teleosts
the hematocrit is between 20 and 40%, but some members ofthe family
Chaenichthyidae (Antarctic icefishes) have colorless blood containing
extremely few erythrocytes, which are fragile and lack hemoglobin. In
Nototheniidae, another group of Antarctic fishes, the blood contains
0.38-1.2 x lo6 erythrocytes per pl, less than normal values for teleosts
(Hureau, 1966; Barber et al., 1981; D’Avino et al., 1990). Average
number of red blood cells in a number of marine teleosts of Puerto Rico
was 2.3-5.3 x 106/pl (Saunders, 1966a).Low amounts of erythrocytes
have been noted in stomiiforms and alepocephalids (Hine et al., 1987).
Leptocephalus larvae (Anguilla, Conger) may lack erythrocytes. No
data on the erythrocyte number of the Asiatic glassfish, Chanda (Am-
bassis) (Nelson, 1984), or other transparent fishes seem to be available.
Elasmobranchs in general show lower erythrocyte number in the
blood than teleosts (0.1-0.4 x 106/p1:Grodzinski and Hoyer, 1938).
This may partly be related to the large size of elasmobranch red cells.
However, in the North Atlantic region chondrichthyans (sharks, rays)
and holocephalans (Chimaera) show not only lower hematocrits but
also lower hemoglobin values than the teleosts (Fange, 1978).
 4                                                                  RAGNAR F A N G E

                                      Table I
          Data on Erythrocytes and Hemoglobin Concentrations in Fish Blood"

                                                            Hemoglobin     Erythrocyte
                          Hematocrit    Erythrocyte NO.^                     lengthb
          Fish                (70)           (103/4)            (g%)           (ym)

C yclostomes
  Myxine glutinosa          19.1 (5)      0.12-0.19 (8)         4.1 (5)    15-35 (6)
  Squalus acanthias         16.3 (5)      0.06-0.07 (8)
                                          0.24 (4)              3.2 (8)    2U4)
  Somniosus                 20 (5)              -               3.2 (5)    31-38 (1)
  Prionace glauca           25 (2)        0.76 (2)             10.0 (2)    15.7 (2)
Teleosts, slow
  Lophius piscatorius       17.2 (5)      0.97-1.30 (8)         3.2 (5)    13-15 (8)
  Cyclopterus lumpus        19.3 (5)      1.09 (4)              3.3 (5)    10-12 (1)
Teleosts, fast swimmers
  Thunnus thynnus           52.4 (3)      2.15 (3)             15.4 (3)     13.1 (3)
  Scomber scombrus          52.5 (5)      3.87 (4)             15.2 (5)     12.0 (7)
  Clupea harengus           51.2 (5)            -              14.0 (5)     12.0 (7)
                                                                             9.0-14 (9)

     (Only a few species and groups have been included. Examples from other fishes may

be found in the text.
      Numbers in parentheses: (1)own estimation; (2) Glazova (1977);(3) CutiCrrez (1967);
(4) Kisch (1951);(5) Larsson et al. (1976);(6) Mattisson and Fange (1977);(7) Wilkins and
Clarke (1974);(8)Wintrobe (1933);(9) Sherburne (1973).

     The hematocrit, hemoglobin concentration, and other hematologi-
 cal parameters are highly sensitive to physiological changes, for in-
 stance those occurring due to stress (Soivio and Oikari, 1976). Contrac-
 tion of splenic vessels may force stored red cells into the general
 circulation (Fange and Nilsson, 1985).

 B. Shape and Size
     Mature fish red cells usually are oval and disk-shaped with a com-
 pact nucleus. The erythrocytes of the lamprey, Lampetra fluuiatilis,
 are biconcave with an almost circular outline, rather similar to mamma-
 lian erythrocytes (Potter et al., 1982). Round disc-shaped red cells
 occur in some teleost species (Siphonostoma typhle: Undritz, 1963).
     The average red cell size differs between systematic groups of
 fishes. Teleost red cells usually measure between about 8 and 15 pm.
 Shrivastava and Griffith (1974) recognized a tendency for brackish
1. FISH   BLOOD CELLS                                                               5

water species of Fundulus to have slightly smaller blood cells than
freshwater species. Glazova (1977) noted that the erythrocytes are
slightly smaller in active species than in nonactive. Probably small
dimensions facilitate physiological exchanges by a favorable cellular
surface/volume ratio. Erythrocytes larger than normal for teleosts are
found in certain deep sea forms: Alepocephalidae, Halosauridae, Sto-
miiformes (Hine et al., 1987), and Saunders (1966a) observed large
erythrocytes (17 pm) in a muraenid teleost, Gymnothorax funebris.
Exceptionally large red cells, about 40 pm in length, occur in dipnoans
(lungfishes) (Parker, 1892). The dipnoan cell sizes almost approach
those of some urodelan amphibians (Amphiuma,Necturus). Elasmo-
branchs, holocephalans (chimaeroid fishes), and myxinoids (hagfishes)
also possess relatively large blood cells (Wintrobe, 1933). The physio-
logical consequences of the enormous cell sizes in lungfishes and
amphibians are not well understood.
    Nonnucleated red cells (erythroplastids, hemoglobin packets) in
marine deep water teleosts of the family Gonostomidae (Maurolicus
mulleri, Valencienellus tripunctatus, and Vinciguerria sp.) have di-
mensions of 5.5 x 2.5 pm. The presence of nonnucleated red cells is
associated with unusually small dimensions of blood vessels (2 pm

                                         .*... .
               A0                      B                         C


           D                                                 F

                                                                     1 Am

    Fig. 1. Camera lucida drawings at the same scale of cells in MGG-stained blood
smears of a teleost, Gadus morhua (cod) (A, B, C), and a lungfish (Protopterus aethio-
picus (D, E, F). A and D, erythrocytes; B and E, polymorphomucleated granulocytes
(B, neutrophil; E, eosinophil); C and F, lymphocytes. The lungfish cells are several
times larger than the corresponding teleost cells.
6                                                        RAGNAR FANGE

internally) (Hansen and Wingstrand, 1960; Hine et al., 1987).
    The sizes of blood and other cells are correlated to the content of
DNA (Ohno and Atkin, 1966; Pedersen, 1971; Hinegardner, 1976). In
a general sense the amount of nuclear DNA increases with biological
complexity. However, within a systematic group of animals, such as
fishes, variations of the amount of nuclear DNA depend mainly on the
amount of repeated DNA (Schmidtke et al., 1978) and the polyploidy
level. Erythrocyte nuclear measurements have been used to distin-
guish between diploid and triploid individuals of fish species (Wolters
et al., 1982). The diploid cellular DNA content in teleosts is around
2 pg, in elasmobranchs 5.6-18.6 pg, and values as high as 248 pg are
noted in dipnoans (Hinegardner, 1976). The average size of the red
blood cells seems to vary in parallel with this series (Table I: erythro-
cyte length).

C. Cytoplasmic Structures
    Ultrastructurally the cytoplasm of mature fish red cells has few
organelles other than pinocytotic vacuoles (Fig. 2A,B) and microtu-
bules (marginal bands). Single small mitochondria may be present
(Keen et al., 1989). Kreutzmann and Jonas (1978) describe a Golgi
complex and a so-called segregation apparatus in adult red cells of the
eel (Anguilla anguilla) and the rainbow trout (Oncorhynchus mykiss).
The segregation apparatus appears to be associated with the hemoglo-
bin formation (Keen et al., 1989).
    Cell elongation during maturation of red cells in rainbow trout
(Salmo gairdneri) is correlated with the appearance of a marginal band
system (Keen et al., 1989). The peripheral bundles of microtubules
presumably protect the shape of the erythrocytes and hinder deforma-
tion of red cells during passage through capillaries ( Joseph-Silverstein
and Cohen, 1984). The number of microtubules in teleost red cells
varies with species from 6-10 to 27 (Kreutzmann and Jonas, 1978).The
microtubules connect with other types of filaments of the cytoskeleton
(Nikinmaa, 1990). Centrioles participate in the biogenesis of the mar-
ginal bands as concluded from ultrastructural studies of the erythro-
cytes of the skate, Raja erinacea (Cohen, 1986). Microtubules of the
marginal bands of the red cells of the smooth dogfish (Mustelus canis)
have been isolated and examined after lysis of the cells with detergents
in the presence of protease inhibitors (Sanchez and Cohen, 1988).
    Intracellular crystallization of hemoglobin commonly is observed
in teleostean red cells (Yoffey, 1929: whiting, Gadus merlangus; Daw-
son, 1932: the pipefish, Syngnathus fuscus; Kisch, 1949: the eel,
1. FISH   BLOOD CELLS                                                             7

    Fig. 2. Erythrocytes. ( 4 Cyclostome: Myxine glutinosa. The cytoplasm contains
pinocytotic vacuoles (rhopheocytosis), canaliculi, and minute granules. (B) Teleost:
Gadus rnorhua (blood cells inside a vessel). The cytoplasm more homogenous and free
from organelles. Electron micrographs. Bar = 1 pm.

Anguilla rostrata; Hansen and Wingstrand, 1960: stomatids and myc-
tophids; and Mattisson, pers. comm.: Aphanopus carbo). It is con-
sidered as a postmortal phenomenon (Hansen and Wingstrand, 1960),
but according to Thomas (1971) the hemoglobin already in the
erythrocytes of the living normal whiting (Gadus merlangus) may exist
in a paracrystalline state as bundles of parallel tubules or filaments.
The tendency to form hemoglobin crystals in the red cells of certain
teleosts may be a phenomenon somewhat similar to the propensity for
crystal formation inside red cells of human beings suffering from sickle
cell disease.
8                                                         RAGNAR F A N G E

   Virus particles or sporozoan parasites often occur in fish blood cells.
Occasionally it may be hard to decide if inclusions represent parasites
or organelles (Cenini, 1984; Rodger et al., 1991).

D. Immature and Senescent Cells
    The blood of just-hatched larva of the rainbow trout (Oncorhynchus
mykiss) contains round disc-shaped larval erythrocytes with special
embryonic hemoglobin (Iuchi, 1973). Bielek (1975) described similar
primary or larval erythrocytes in the pike, Esox lucius, and the gray-
ling, Thymallus thymallus.
    In adult teleosts, the blood normally contains a certain percentage
of immature (juvenile) red cells (proerythrocytes, reticulocytes) (Daw-
son, 1933, Hardig, 1978, Boomker, 1980; Keen et al., 1989, Houston,
1990). Sherburne (1973) found between 6 and 38% immature erythro-
cytes (average 21%) in the herring (Clupea harengus). Immature red
cells differ from the mature ones by a more circular outline, by
presence of mitochondria, polyribosomes and other organelles, and by
less hemoglobin (Lane and Tharp, 1980). When maturing they lose
organelles, get loaded with hemoglobin, and elongate and develop
marginal bands (Keen et al., 1989). Hemoglobin-free erythroblasts
with a basophilic cytoplasm and a large nucleus are abundant in elas-
mobranch embryos (Saunders, 1966b) and in the blood of the hagfish,
Myxine glutinosa (Mattisson and Fange, 1977). Small numbers of
erythroblasts normally occur in the circulation of adult elasmobranchs,
teleosts and dipnoans (Saunders 1966a; Boomker, 1980; see Section on
Blast Cells).
    With senescent red cells a part of the circulating erythrocytes are
aged or effete. “Nuclear shadows” in blood smears may consist of
disintegrated red cells (Weinberg et al., 1972). Seven and one-half
percent of senescent red cells occur in the blood of the rainbow trout
(Keen et al., 1989).


A. Erythrocyte Homeostasis
   The hematocrit values are relatively constant within species but
vary between species. The amount of erythrocytes (evaluated as he-
matocrit or erythrocyte count) and the total hemoglobin concentration
1.   FISH BLOOD CELLS                                                   9

in the blood vary in accordance to life habits. Fast swimming species of
fishes on average have more erythrocytes, larger hematocrits, and more
hemoglobin than less mobile forms (Table I). A hemoglobin content of
more than 20 g/100 ml is found in tropical scombrids (Klawe et al.,
1963).Low oxygen in the environment stimulates erythropoiesis. Thus
values for hematocrit and hemoglobin increase in freshwater fishes
(Cottus poecilopus and C . gobio),which were transferred to water with
low oxygen concentration (Starmach, 1970).Seasonal variations in the
hematocrits of the winter flounder, Pseudopleuronectes americanus,
may depend on nutritional or hormonal factors (Bridges et al., 1976).
However, under normal conditions fishes usually are able to keep the
concentration of red cells in the blood relatively constant. Such a red
cell homeostasis results from a dynamic equilibrium between new
formation (erythropoiesis) and destruction of erythrocytes. New eryth-
rocytes are continuously entering the circulation, and effete erythro-
cytes are destroyed at the same rate. Destruction and elimination of
aged or damaged erythrocytes are brought about by macrophages in the
kidney and the spleen. Whereas the anucleate mammalian red cells
have a life span of about 120 days, the less differentiated fish red cells
presumably live somewhat longer. Hevesy et al. (1964), by isotope
labeling, estimated a life span of more than 150 days for erythrocytes of
a teleost fish, the tench (Tinca tinca).
    Youson (1971) has observed destruction of red blood cells in the
kidney of the sea lamprey (Petromyxon marinus). Binding of antibod-
ies to aged red cells or exposure of fish to pollutants may increase the
rate of destruction of red cells in salmonids (Nikinmaa, 1990). The
hemoglobin from decomposed erythrocytes is transformed into bile
pigments and iron. The iron, stored as ferritin or hemosiderin, may be
reused in erythropoiesis. Main iron stores are found in lymphomyeloid
tissues and in the liver. In the tench (Tinca tinca) the spleen contains
10-15 times more iron per gram of tissue than the liver (Dijk et al.,
1975). Probably the iron in the spleen occurs largely in melanomacro-
phage centers (Walker and Fromm, 1976; Agius, 1985).

B. Metabolism
    The metabolism of vertebrate red cells generates energy for the
maintenance of cell shape and for transport of substances across the
cell membrane (Nikinmaa, 1990). The sodium pump and phosphoryla-
tion processes consume about 50% of the total energy. Fish erythro-
cytes are metabolically more active than mammalian erythrocytes and,
like unripe mammalian red cells or reticulocytes, consume oxygen by
10                                                         RAGNAR FANGE

respiration (Bushnell et al., 1985). In mammals a large part of the
erythrocytic energy production is based on anaerobic glycolysis and
lactate formation. A fraction of glucose is metabolized along the pen-
tose phosphate pathway producing reduced nicotiamide adenine di-
nucleotide phosphate (NADPH), reduced glutathione and sulfhydryl
groups being needed for the detoxification of free radicals. The impor-
tance of the pentose phosphate pathway in fish erythrocytes is not
known. Although glycolytic enzymes are present, glycolysis may be
limited by low permeability to glucose of the red cells in some fish
species (Bolis et al., 1971; Bachand and Leray, 1975).

C. Proteins, Phosphates, and Nitrogen Metabolites
   The hemoglobins of vertebrates with the exceptions of cyclostomes
(hagfishes and lampreys) are tetrameric, built of four peptide chains,
each with one heme and a molecular weight of about 17,000. Mono-
meric and dimeric hemoglobins are found in the cyclostome red cells.
The hemoglobin of the coelacanth, Latimeria, shows features of fish as
well as tetrapod relationships. Sequence analyses of the hemoglobin
indicate that Latimeria is more closely related to tetrapods than are
dipnoans (Weber et al., 1973; Gorr et al., 1991). Fish embryos contain
special kinds of hemoglobin, which are electrophoretically different
from adult hemoglobins (Iuchi, 1973).
   Multiple hemoglobin systems in many fishes may be important in
physiological adaptations to variable environments. Pterygoplichthys
multiradiatus, a catfish of the Amazon basin, has multiple erythrocytic
hemoglobin. It is a facultative air breather adapted for periods with low
oxygen availability (Val et al., 1990). But even fishes living in constant
oceanic environment often possess many forms of hemoglobin.
   In addition to hemoglobin, erythrocytes contain other proteins such
as glykolytic metabolic enzymes, but mitochondria1 enzymes of the
aerobic cell metabolism are poorly represented. Vislie (1978) found
lysosomal enzymes such as p-N-acetylglucosaminidase in mature
erythrocytes of the flounder (Platichthys Jesus).
   The enzyme carbonic anhydrase occurs in red cells of all verte-
brates, but the activity in the blood of the flounder (P. f l e w s ) is low
(Mashiter and Morgan, 1975). It catalyzes the reversible hydration/
dehydration of carbon dioxide and functions in the transport of carbon
dioxide. Carbonic anhydrase from red cells of hagfish (Myxine gluti-
nosa) has been investigated by Carlsson et al. (1980), who found its
chemical properties to resemble those in other vertebrates. Superox-
ide dismutase has been isolated from the red cells of salmonids (Scott
and Harrington, 1990).
1. FISH   BLOOD CELLS                                                   11

     Nucleoside triphosphates are physiologically important in fish red
cells because they function as allosteric modifiers controlling the oxy-
gen affinity of the hemoglobin. In many fish species adenosine triphos-
phate (ATP) is the main compound acting in this way (Clupea:
Everaarts, 1978; Scomber: Bartlett, 1982; Cichlasomu: Gillen and
Riggs, 1971), but the erythrocytes of some species possess both ATP
and quanosine triphosphate (GTP) (smooth dogfish, Mustelus canis,
and the American eel, Anguilla rostrata: Peterson and Poluhowich,
1976; Bartlett, 1982; the Australian lungfish, Neocerutodus: Isaacks
and Kim, 1984). Quanosine triphosphate is more efficient than ATP in
lowering the oxygen affinity in certain species. Other phosphates
present in the red cells of some air breathing species are inositol-
pentaphosphate in Arapaima gigas, inositol-diphosphate in the South
American lungfish (Lepidosiren), and 2,3-diphosphoglycerate (DPG)
in Pterygoplichthys sp. (Isaaks and Kim, 1984).A considerable amount
of DPG is found in the red cells of the lamprey (Entosphenus tridenta-
t u s ) but the erythrocytes of hagfish (Eptatretus) contain varying
amounts of ATP (Bartlett, 1982).
      In fishes a major part of nitrogenous excretory products are excreted
as ammonia via the gills. Experiments on the carp show that ammonia
is transported in the blood in about equal amounts by red cells and by
plasma (Ogata and Murai, 1988).

D. Membrane Properties
    Red cell membranes consist of a bimolecular lipid layer associated
with carbohydrate-containing antigens intermingled with proteins,
among which are contractile proteins and enzymes. Spectrin, actin,
and other proteins form an intricate membrane skeleton interacting
with hemoglobin, membrane transport proteins, and tubulin of the
marginal bands. Spectrin and actin occur in the red cell membrane of
elasmobranchs (Cohen e t ul., 1982), but in cyclostomes spectrin is
lacking (Ellory e t ul., 1987; Nikinmaa, 1990).
    A few studies have been made in order to find blood group systems
in fish (Cushing, 1970). Studies of erythrocyte antigens in cod, Gadus
morhua (Moller, 1967), and in the American eel, Anguilla rostrata
(Sindermann and Krantz, 1968), are of genetic interest. The American
eel shows complex isoagglutinin-isoantigen systems. A number of cod
blood group antigens were discovered by using antiserum from
    Somewhat less than half of the erythrocyte membrane consists of
lipids. The membrane lipids composition have been investigated in a
few species. Phospholipids make up about 80% of the total lipids.
12                                                       RAGNAR FANGE

Polyunsaturated fatty acids constitute a high percentage of the fatty
acids (Bolis and Fange, 1979). The function of the membrane lipid
components is influenced by environmental factors such as diet and
temperature (Bly and Clem, 1988). Low environmental temperature
induces increased degree of unsaturation in fish lipids (Cowey and
Sargent, 1977; see the review by Nikinmaa, 1990).
    The red cell membrane contains transport proteins for both anions
and cations. The exchange of cations may be brought about by a so-
dium pump. Little is known about molecules responsible for cation
transport across the erythrocyte membrane in fishes. Chloride ions
pass extremely slowly through the red cell membrane of the hagfish,
Myxine, and in lampreys (Larnpetra, Petrornyzon) the red cells are
impermeable to bicarbonate. This makes exchange between chloride
and bicarbonate difficult, and cyclostome erythrocytes seem incapa-
ble of intracellular buffering during transport of carbon dioxide
(Nikinmaa, 1990).
    The erythrocytes from the Japanese lamprey, Entosphenus japoni-
cus, show 50 or 100 times higher adenosine triphosphatase (ATPase)
activity than mammalian erythrocytes. The ATPase is activated by Mg
and Ca ions, but not by Na and K ions (Asai et al., 1976).
    Many vertebrate erythrocyte membranes possess P-adrenergic
receptors for catecholamines. Under influence of catecholamines
adenylate cyclase is activated catalyzing the conversion of ATP to
cyclic adenosine monophosphate (CAMP). However, according to
Tufts and Randall (1988) the erythrocytes of elasmobranchs and cy-
clostomes, in contrast to those of other vertebrates, are not influenced
by catecholamines, and adrenergic receptors may not occur in the
erythrocyte membranes of cyclostomes (Nikinmaa, 1990).
    Erythrocytes are frequently used to investigate the permeability of
organic substances through cell membranes. The observations may be
based on hematocrit estimations, hemolysis experiments ( Jacobs,
 1931), or tracer technique. Relatively few studies concern fish erythro-
cytes. The permeability of fish red cells to glucose and other carbohy-
drates varies considerably between species. The red cells of armored
catfish (Pterygoplichthys)and brown trout (Salmo trutta) are imper-
meable to glucose, those of arawana (Osteoglossum)slightly perme-
able, and those of the electric eel (Electrophorms)and lungfish (Lepid-
osiren) show the greatest permeability (Bolis et aZ., 1971; Kim and
 Isaacks, 1978). In a series of South American fish (lungfish, Lepido-
siren; electric eel, Electrophorus; arawana, Osteoglossum; armored
catfish, Pterygoplichthys; piraruca, Arapaima) the red cells proved to
be permeable to urea in decreasing order (Kim and Isaacks, 1978).The
1. FISH   BLOOD CELLS                                                13

effects of urea on red cells do not seem to have been specifically
studied in those fish that use urea retention for their osmotic balance
(elasmobranchs, holocephalans, the coelacanth, Latimeria).

E. Osmotic Fragility
    Hemolysis is caused by the addition of distilled water or hypotonic
sodium chloride solutions to blood. The osmotic fragility of the eryth-
rocytes differs between species. The erythrocytes of euryhaline spe-
cies, such as the rainbow trout (Oncorhynchus mykiss) (Hughes et al.,
1986a,b)and gars (Lepisosteus osseus and L. productus), show higher
tolerance against hypotonic salt solutions than those of exclusively
marine teleosts (Ezell et al., 1969). Pitombeira et al. (1971) studied
osmotic fragility in tuna (Thynnus thynnus) and Spanish mackerel
(Scomberomorus maculatus) before and after spawning. The osmotic
fragility of the red cells may be expressed as a percentage of NaCl
causing 50% hemolysis (time and temperature standardized). Erythro-
cytes of the brook trout (Saluelinusfontinalis) show an osmotic fragil-
ity of 0.30-0.40% NaCl (Christensen et al., 1978: O"C,60 min). In Esox
Zucius a value of 0.32% NaCl is measured (Mulcahy, 1970). The eryth-
rocytes of the hagfish (Myxine glutinosa) seem more resistant against
hypotonic salt solutions than those from an elasmobranch. The eryth-
rocytes from the hagfish and marine teleosts (Labrus berggylta, Gadus
morhua) hemolyze in 0.07-0.1 M NaCl solution, while those of the
greenland shark (Somniosus microcephalus) hemolyze at sodium
chloride concentration of 0.13-0.21. M (Fange, 1985). Taurine and
amino acids are important in the intracellular osmotic volume regula-
tion of the erythrocytes of teleosts and elasmobranchs (Fugelli, 1967;
Goldstein and Boyd, 1978; Bedford, 1983).
    Studies on amphibian blood show that the capacity of red cells to
resist osmotic swelling is affected by the cell metabolism. Increased
metabolism, resulting in higher ATP levels, increases the resistance
against hypotonic NaCl solutions (Goniakowska-Witalinska, 1974).

F. Gas Transport
   The oxygen binding capacity of the blood is dependent on the
concentration and properties of the hemoglobin containing red cells.
Fish in different ecological environments need different functioning
hemoglobins. The oxygen binding capacity of the hemoglobin mole-
cules varies with the chemical structure of the protein moiety of the
molecule and with the intraerythrocytic content of certain substances
14                                                        RAGNAH FdNGE

such as phosphates, which modulate the properties ofthe hemoglobin
molecule. The oxygen binding properties of the blood are controlled by
changes in the physicochemical environment of the erythrocytes. The
Bohr and Root effects on fish hemoglobins, the effects of urea on the
oxygen affinity in shark blood, and temperature effects on the blood gas
binding properties are discussed by Nikinmaa (1990). The oxygen
affinity of the blood of water breathing fishes generally is higher than
in air-breathing fishes due to different intrinsic properties of the hemo-
globins (Johansen and Lenfant, 1972; Johansen et al., 1978).
    The efficiency of the transport of carbon dioxide in vertebrate circu-
lation depends on a rapid exchange of chloride and bicarbonate
through the erythrocyte cell membrane (Deuticke, 1970). The enzyme
carbonic anhydrase facilitates the anion exchange through the erythro-
cytic membrane. However, in fish the anion permeability of the red
cells varies. In certain teleosts (carp, Cyprinus carpio; pikeperch,
Stizotedion lucioperca) anion exchange of the red cells is slow, and in
cyclostomes-due to nonexistent intracellular buffering of the blood
(Nikinmaa, 199O)-the capacity for carbon dioxide excretion probably
is lower than in any other vertebrates.


A. Occurrence
    Except for a few hours during which they are transported b y the
blood, vertebrate leucocytes function outside the circulation (Tavas-
soli and Yoffey, 1988). In mammals there are about 60 times more
mature granulocytes and perhaps 400 times more lymphocytes in the
tissues than inside the blood vessels (Antonioli, 1961). In fishes large
amounts of leucocytes, in all phases of development, occur in specific
lymphomyeloid tissues and organs and infiltrate the skin, muco-
sal membranes, and connective tissue areas all over the organism
(Drzewina, 1905; Kanesada, 1956; Fange, 1984, 1987).
    Fish blood is remarkably rich in leucocytes (Parker, 1892; Drze-
wina, 1911;Wintrobe, 1933; Reznikoff and Reznikoff, 1934; Weinberg
et d., 1972). In teleosts and elasmobranchs the blood contains 15-135
x lo3 respectively 22-57 x lo3 white cells per p (Kisch, 1951), as
compared with about 7 x lo3 leucocytes per p1 in human blood. The
high values may be explained by nucleated thrombocytes being
counted as leucocytes.
1.   FISH BLOOD CELLS                                                15

B. Staining Methods, Classification
    Since Ehrlich (1879) leucocytes are classified according to their
affinities to acid and basic dyes. Application of differential stain-
ing methods (May-Griinwald-Giemsa, Romanowsky, Pappenheim,
Wright, etc.) on fish blood has resulted in a nomenclature based on
human hematology (Jakowska, 1956; Cenini, 1984; Rowley et al.,
1988). Thus fish leucocytes may be divided into granulocytes (neutro-
phil, eosinophil, and basophil) and nongranulocytes (lymphocytes,
monocytes, and thrombocytes). The thrombocytes are often regarded
as forming their own cell line. Blast cells, early undifferentiated blood
cells with a large nucleus and basophilic cytoplasm, may be included
in the nongranulate fraction of circulating leucocytes. Stem cells are in
part identical with blast cells. However, sometimes differential stain-
ing methods seem to work less well on fish blood cells than on mamma-
lian cells.
    When possible microscopy studies of fixed and stained smears of
fish blood should be complemented with observations on living cells,
either illuminated by phase contrast, dark field, polarized or ordinary
light, or supravitally stained with neutral red or fluorescent dyes like
acridine orange. Cytoplasmic particles containing hydrolytic enzymes,
such as lysosomes, take up stains by a vital process (Koehring, 1930;
Allison, 1968). As an example shark granulocytes, probably due to
the lysosomal nature of the cytoplasmic granules, accumulate supra-
vital stains very rapidly (Fange, 1968). Phagocytosis may be demon-
strated by injections of latex particles or yeast cell membranes (zy-
     Density gradient centrifugations with Percoll or Ficoll-Paque are
 used to separate types of fish leucocytes (Braun-Nesje et al., 1981;
 Fujii, 1981; Savage, 1983; Blaxhall and Sheard, 1985; Mainwaring and
 Rowley, 1985a,b; Fange, 1987; Suzuki, 1988; Plytycz et al., 1989).
 Leucocyte types can also be separated by means of their different
adherence to glass (Mainwaring and Rowley, 1985a,b) and by flow
cytometry (Ellsaesser et al., 1985). Electron microscopic, cyto-
 chemical, and immunological methods are increasingly used to study
 fish blood cells. Difficulties are involved in overbridging the gap be-
tween light microscopic and electron microscopic observations. Fish
hematology is especially complicated because of differences between
 systematic groups. Combinations of many methods are needed if
 classification of fish blood cells shall improve its present provisional
16                                                        HAGNAR FANGE

C. Granulocytes
   The majority of granulocytes are mobile, phagocytically active
cells. The cytoplasm contains lysosomal granules, vacuoles, mitochon-
dria, and other organelles or particles. Contractile vacuoles occur in
elasmobranch granulocytes (Fey, 196613).The properties of the granu-
locytes vary extremely, especially between systematic groups of fishes,
and Rowley et al. (1988, p. 41) rightly remark that “no other leucocyte
type has caused as much confusion in the fish literature.”

    In teleosts, granulocytes constitute 4.5-18% or more of the leuco-
cytes in the blood (Duthie, 1939; Watson et al., 1963; Wardle, 1971;
Hines and Spira, 1973). They measure about 9-12 pm in diameter in
blood smears, less in the living state, and resemble mammalian granu-
locytes in appearance. Ellis et al. (1976) and Ellsaesser et al. (1985)
distinguish only one type of granulocyte in the plaice (Pleuronectes
platessa) and the channel catfish (lctalurus punctatus) respectively.
Other authors describe several categories: (a) neutrophils (hetero-
phils), (b) eosinophils, and (c) basophils.

    a . Neutrophils (Heterophils).The predominating, sometimes only
existing, type of granulocyte in teleost blood is named neutrophil due
to its similarity with neutrophils of human blood. Neutrophils consti-
tute 5.9% of the total leucocytes in the brook trout, Salvelinus fonti-
nalis (Christensen et al., 1978). They are phagocytic in most species
(Phromsuthirak, 1977). Neutrophils is the usual denomination (Catton,
1951; Weinreb, 1963; Watson et al., 1963, Kelknyi and Nkmeth, 1969;
Wardle, 1971; Javaid and Lone, 1973; Lester and Desser, 1975; Ellis,
1977; Cannon et al., 1980; Bielek, 1981; Garavini et al., 1981; Roubal,
1986). But many synonymous terms are used, such as heterophils (Fey,
1966b; Barber and Westermann, 1978), “fine” or “specific” leucocytes,
or “polymorphonuclears.” In many teleosts, for instance salmonids
and cyprinids, the neutrophils possess polymorph (segmented or mul-
tilobed) nuclei (Fey, 1966b; Rowley et al., 1988),but in other species
the nuclei are round or bilobed (the eel, Anguilla anguilla: Fey, 1966b;
the plaice, Pleuronectes platessa: Ellis, 1976; Tilapia sp.: Ezzat et al.,
 1974; and the channel catfish, lctalurus punctatus: Elssaesser et al.,
    The cytoplasm of neutrophils contains numerous fine granules
(Phromsuthirak, 1977). These stain faintly red, pink, or violet in
blood smears (Catton, 1951; Gardner and Yevish, 1969; Ezzat et al.,
1.   FISH BLOOD CELLS                                                 17

1974; Lester and Desser, 1975; Ellis, 1977; Hightower et al., 1984;
Roubal, 1986), or azurophil (Haider, 1967). But often the granules are
unstained by ordinary staining methods or need extra long staining
time (Durand, 1950; Catton, 1951). Finn and Nielsen (1971)observed
granulocytes in the rainbow trout, Oncorhynchus mykiss, that did not
stain. Granules of the granulocytes in the cod, Gadus morhua, are
unstained by Giemsa but show intense peroxidase response and are
visible with phase contrast or dark field (Fange and Koskinen, 1984,
    The teleost neutrophil granules generally are peroxidase positive
and show acid phosphate reactions and affinity to Sudan black B (Can-
non et al., 1980 and Garavini et al., 1981: Ictnlurus; Bielek, 1981:
Cyprinus carpio, Tinca tinca, Salmo gairdneri; Hine et al., 1987:
various species). But peroxidase negative neutrophils occur in certain
species of eels (Hine et al., 1987). Neutrophils ofthe plaice, Pleuronec-
tes platessa, are stained with PAS indicating presence of glycogen or
other polysaccharides (Ellis, 1976). However, Barber et al. (1981) de-
scribe a special type of PAS-positive granulocytes in freshwater fish. In
the main cytochemical responses of teleost neutrophils resemble those
of similarly named mammalian cells.
    In the electron microscope the neutrophils show Golgi apparatus,
mitochondria, ribosomes, endoplasmic reticulum, vacuoles, glycogen
particles, and specific granules (Weinreb, 1963, Cenini, 1984, and
Fujiinaki and Isoda, 1990: cyprinids; Bielek, 1980: salmonids and
cyprinids; Ferguson, 1976: plaice, Pleuronectes platessa; Savage,
 1983: Esox lucius; Cannon et al., 1980: channel catfish, Ictalurus
punctatus; Lester and Desser, 1975: white sucker, Catastomus com-
mersoni; Ishizeki et al., 1984: loach, Misgurnus sp.). The granules are
round or elongate with dimensions of 0.1-0.5 pm. The interior of the
granules is either homogenous or contains fibrillary or rodlike inclu-
sions (Ferguson, 1976). Subpopulations of granules may exist. Small
and large granules in heterophils of the pike (Esox lucius) struc-
turally resemble primary and secondary granules of mammalian
neutrophils (Savage, 1983). Fujimaki and Isoda (1990) distinguish
three kinds of granules in the neutrophils of the goldfish, Carassius

    b. Eosinophils. Eosinophils contain cytoplasmic granules that
stain by acid stains. In most teleosts they are scarce or lacking in
the circulation (Clupea harengus: Sherburne, 1973; Acanthopagrus:
Roubal, 1986), but in labrids and a few other groups they are relatively
abundant (Drzewina, 1911).Unique eosinophils containing one large
18                                                        RAGNAR FANGE

granule (2.5-2.8 pm) occur in Misgurnus anguillicaudatus (Ishizeki et
al., 1984).
    Teleosts lacking eosinophils in the blood may possess such cells in
the tissues and the peritoneum. The abundance of eosinophils in the
intestinal mucosa of many teleosts led Jordan and Speidel (1924) to
assume that eosinophils are responsible for immunity against bacteria.
In the stickleback, Gasterosteus aculeatus, eosinophils are a rare com-
ponent among other leucocytes infiltrating the skin (Phromsuthirak,
1977). Lester and Daniels (1976) found eosinophils in histologi-
cal sections of inflammatory tissue of the white sucker (Catastomus
commersoni) affected by parasites. The sections were stained with
hematoxylin-eosin. However, Romanowsky stained blood smears
showed no eosinophils. Eosinophilic granule cells in peritoneal exu-
dates of the eel, Anguilla australis, give peroxidase positive reaction
(Hine and Wain, 1989). Electron microscopic studies show that teleost
eosinophil granules, in contrary to those of mammals, do not possess
crystalline inclusions.

   c . Basophils. The granules of basophils stain by basic dyes. These
cells scarcely occur in the blood of some teleosts such as carp (Cypri-
nus carpio), tench (Tinca tinca), and perch (Perca fluuiatilis) (Fey,
l966b; Haider, 1968).
    The blood of chondrosteans (sturgeons, paddlefish) contains neu-
trophils (heterophils) and eosinophils (Good et al., 1966; Kelenyi,
1972; Ivanova, 1983; Fange, 1986b). Neutrophils with 1-5 lobed nu-
clei predominate in the sturgeon, Acipenser brevirostris. The neutro-
phi1 granules are of two kinds, large, elongate, and uniformly electron
dense, and small with a fibrillary interior. Eosinophils, also with lobed
nuclei, are scarce. The eosinophils of sturgeons possess large, round or
oval homogenous granules, which are peroxidase-positive (Hine and
Wain, 1988b).Tissue infiltration with eosinophils is observed in pad-
dlefish, Polyodon spathuZa, infected by nematode larvae (Miyazaki et
ul., 1988).
   The elasmobranch blood is exceptionally rich in granulocytes with
distinct granules, predominantly showing various grades of eosino-
philia. Already in 1846 Wharton Jones (p. 63) described such a cell in
the blood of the skate (Raja batis) as “composed of an agglomerulation
of granules surrounded by a cell-membrane. The granules are clear
and strongly refract the light.” There is little agreement on the number
1.   FISH BLOOD CELLS                                                  19

of existing types of elsasmobranch granulocytes. Some investigators
distinguish two types (Drzewina, 1911; Jordan, 1938; Fange, 1968;
Johansson-Sjobeck and Stevens, 1976). Pica et al. (1983)and Saunders
(1966a,b) describe eosinophils and heterophils but state that in some
species heterophils are replaced by neutrophils. Other authors find
three types of elasmobranch granulocytes (Fey, 1966a; Stokes and
Firkin, 1971). In the Greenland shark, Somniosus microcephalus, the
blood contains: (a) cells with large, intensely eosinophilic granules,
(b) cells with small, weakly eosinophilic granules, and (c) cells con-
taining round granules of high density, which do not stain by Giemsa
(Fange, 1987). The “heavy granulocytes” may correspond to the type
I1 cells of Morrow and Pulsford (1980) or G3 cells of the type I1 cells of
Morrow and Pulsford (1980) or G3 cells of Parish et al. (1986)found in
the blood of the dogfish (Scyliorhinus canicula) by electron micro-
scopy (see Section IV,D).
     Electron microscopic studies confirm and extend results from light
microscopic observations (Fig. 3A,B). Two types of granulocytes are
distinguished by their ultrastructure in nurse shark, Ginglymostoma
cirriitum (Hyder et al., 1983), and in rays (Torpedo spp.: Pica et al.,
 1983; Raja spp.: Mainwaring and Rowley, 1985a). But in the blood of
the dogfish, Scyliorhinus canicula, investigators found a large diver-
 sity of granulocyte ultrastructure (Morrow and Pulsford, 1980; Main-
waring and Rowley, 1985a; Parish et al., 1986). The most abundant
granulocytes have relatively large (0.8 pm in diameter) membrane
bound eosinophil granules; in addition three or four less common
types are found. Electron microscopic and cytochemical observations
on a series of other elasmobranchs (Hine and Wain, 1987a,b,c; Hine et
al., 1987) show granulocyte populations difficult to put into categories
 corresponding to those of other vertebrates.
     At present there is no generally accepted nomenclature for elasmo-
branch granulocytes. The cytoplasmic granules are mostly eosino-
 philic and have different dimensions and structure. A scarce type of
 granulocytes contains high density granules, which do not stain by
     The granulocytes of rays and sharks as a rule are peroxidase nega-
 tive but contain acid phosphatase and esterases (Hine et al., 1987).
   Microscopic and ultrastructural studies of the blood of ratfishes
such as Chimaera rnonstrosa show two types of granulocytes, one with
fine, faintly red granules, the other with coarse, brightly red granules.
In the adults the fine granulocytes predominate (Fange and Sundell,
1968; Mattisson et al., 1990). Obvious species differences exist. The
 20                                                                   KAGNAR FANGE

    Fig. 3. Leucocytes. Elasmobranch: Raja radiata. (A) Granulocyte (from Leydig
organ). Electron-dense granules with rod-shaped inclusions are grouped around a Golgi
apparatus ( G ) .Part of the nucleus (N) has been sectioned. (B) Granulocyte containing
large, round, electron-dense granules. The cytoplasm is highly vacuolated. (C) Small
lymphocyte. The nucleus is surrounded by a narrow rim of nongranulated cytoplasm.
Electron micrographs. Bar = 1 wm.
1.   FISH BLOOD CELLS                                                  21

holocephalan granulocytes are peroxidase negative but contain acid
phosphatase and esterases (Hine and Wain, 1988a,b).
    Dipnoan white blood cells are exceptionally abundant and promi-
nent. The leucocytes of the Australian lungfish, Neoceratodus forsteri,
measure 25 pm in diameter (probably measured in histological sec-
tions: Ward, 1969). According to Parker (1892)the largest leucocytes of
the African lungfish, Protopterus annectens, may exceed the erythro-
cytes (length 40-46 pm) in size. The proportion of leucocytes to eryth-
rocytes in the blood was about 1: 3, but probably erythroblasts were
counted as leucocytes. In the blood of Neoceratodus, about 16% of the
cells of the erythrocytic line are in the blast stage (Ward, 1969; Hine et
al., 1990a,b). Parker (1892) found that lungfish leucocytes may con-
tinue amoeboid movements for hours under the cover glass and re-
garded the cells as “admirably adapted for examination in the living
condition” (p. 169). Further studies have shown that the white blood
cell pattern of lungfishes is relatively complex (Protopterus: Jordan,
1938; DeLaney et al., 1976; Neoceratodus: Ward, 1969; Hine et al.,
1990a,b). Heterophils, or small eosinophils, which constitute 69% of
the leucocytes (Ward, 1969), have find eosinophilic granules. Large
eosinophils have an ovoid or bilobed nucleus and coarse intensely
eosinophilic granules with a rodlike internal structure. Neutrophils
possess azurophilic granules and segmented (polymorph) nuclei.
As in chondrosteans (Hine and Wain, 1988b) eosinophil granules
are peroxidase-positive, while heterophil and neutrophil granules con-
tain no peroxidase.
   Granulocytes constitute the majority of leucocytes in the blood of
Latimeria chalumnae. Most common are neutrophils (pseudoeosino-
phils) with large granules and eosinophils. However, the blood cells of
the coelacanth have not been much investigated (Millot et al., 1978;
Locket, 1980).
   Granulocytes constitute about 50% of the total leucocytes in the
blood of the hagfish, Myxine glutinosa. Only one granulocyte type is
recognized. The nuclei are polymorph with one to three segments.
The granules resemble ultrastructurally the primary or azurophil gran-
ules of mammalian neutrophils (Mattisson and Fange, 1977) but are
peroxidase-negative ( Johansson, 1973). The hagfish granulocytes are
22                                                        RAGNAR F A N G E

 phagocytically active. Very large and active phagocytic granulocytes
 occur in the peritoneal cavity together with nongranulated macro-
 phages (Fange and Gidholm, 1968; Mattisson and Fange, 1977).
    The granulocytes of the lamprey (Lampetra fluzjiatilis) resemble
those of the hagfish but are smaller. As in the hagfish, they constitute
about 50% of the leucocytes and have nonsegmented or 2-3 lobed
nuclei. The granules are heterogenous in size (0.07-0.40 pm) and
peroxidase negative (Page and Rowley, 1983). The granulocytes of the
sea lamprey (Petromyzon marinus) resemble teleostean neutrophils
ultrastructurally (Potter et al., 1982). They are phagocytically active
and are opsonized by antiserum (Page and Rowley, 1984; Fujii, 1982).
Blood eosinophils are found in the ammocoetes stage (Potter et al.,

D. Mast Cells, PAS-Positive Granulocytes
    Mast cells are supposed to belong to the same cell line as the blood
basophils. The cytoplasmic granules are basophilic and stain meta-
chromatically with toluidin blue. The granules contain sulfated
polysaccharides (heparin), which explains metachromacia, and sub-
stances such as histamine, which are released during anaphylaxia. The
mast cells are found in mammalian connective tissue, especially
around blood vessels, but it has been debated if similar cells occur in
fish (Ellis, 1977).
    In teleosts basophilic (or eosinophilic) granulated cells form a stra-
tum granulosum in the stomach and intestine (Drzewina, 1911; Bolton,
1933; Bielek, 1975).The cells have been regarded as mast cells but are
not metachromatic and do not react to the histamine liberator 48/80
(Arvy, 1955; Weinreb and Bilstad, 1955). Also to some extent resem-
bling mast cells are PAS-positive granular leucocytes (PAS-GL) in
certain freshwater fishes (Barber and Westermann, 1975, 1978), and
eosinophilic granular cells (EGC) in the plaice Pleuronectes platessa
(Ellis, 1977). T h e PAS-GL are considered as “forerunners to mast
cells.” The granules contain nonsulfated, neutral polysaccharides,
which may be experimentally sulfated to produce a metachromatic
reaction (Barber and Westermann, 1978). Davina et uZ. (1980) describe
a PAS-positive granulocyte in the intestine of a barb (Barbus); the
nucleus is eccentric, the granules do not stain with Giemsa but are
intensively red with PAS. The “heavy granulocytes” that constitute a
minor fraction of the granulocytes in the blood of the Greenland shark,
Soinniosus microcephalus, also possess granules that do not stain by
Giemsa. Due to the density of the granules the heavy granulocytes can
be isolated by centrifugation. The granules may contain polysac-
1. FISH   BLOOD CELLS                                                 23

charides, but PAS staining was not tried (Fange, 1987). Heavy granulo-
cytes of teleosts (blood of puffers, and pronephros of carp: Suzuki,
1986, 1988), “secretory granulocytes” of the buffalofish, Ictiobus, and
the paddlefish, Polyodon ( Jordan, 1938), and “finely reticular cells”
(Hines and Spira, 1973) show similarities to mast cells. But it is doubt-
ful if any mast cell-like cells in fishes represent true mast cells; they
usually lack metachromacia, and there is no direct evidence that they
contain histamine and heparin (Ellis, 1982).

E . Lymphocytes and Plasma Cells
    Lymphocytes may be defined either morphologically or function-
ally (Ellis, 1977). Lymphocytes in the morphological sense are rela-
tively small cells with a round or oval nucleus (Fig. 2C). The cytoplasm
is nongranulated or contains few minute granules and usually stains
blue in routinely stained smears. Teleost lymphocytes measure be-
tween 4.5 and 8.2 pm (Ellis, 1977),but some authors distinguish small,
medium, and large lymphocytes. Remarkably large lymphocytes are
found in lungfishes and certain elasmobranchs that possess large cells
generally. Lymphocytes are mobile but usually nonphagocytic. Ultra-
structurally the cytoplasm shows mitochondria, rough and smooth en-
doplasmic reticulum, ribosomes, and a Golgi system. They constitute
from 50 to 80% of the leucocytes, but sometimes thrombocytes are
reported to be more frequent than lymphocytes in fish blood. Verte-
brate lymphocytes are supposed to be the predominating cells of
lymph, but all kinds of leucocytes, not exclusively lymphocytes, are
found in the supraspinal fluid of the plaice (Pleuronectes flesus)
(Wardle, 1971). However, so-called lymph collected from fishes is
claimed to originate from plasma skimming and does not correspond to
the lymph of mammalian lymph vessels (Vogel and Claviez, 1981).
    There are few if any morphological criteria that can be used to
distinguish lymphocytes from other types of nongranulated cells such
as circulating stem cells, blast cells, monocytes, or thrombocytes.
Phase-contrast microscopy, vital and fluorescence staining, and elec-
tron microscopy may solve part of the problem, but experimental
methods, preferably immunological, are needed. The matter is
discussed by Ellis (1977) and Rowley et al. (1988). In practice all
nongranulated white blood cells, not identified morphologically
as thrombocytes, may be provisionally described as lymphocytes or
“lymphocyte-like” cells.
    Lymphocytes of hagfish ( M y x i n e glutinosa) cannot be morphologi-
cally distinguished from early stages of erythrocytes, the “lymphoid
24                                                       KAGNAR FANGE

hemoblasts” of Jordan (1938) and Good et al. (1966), or from varieties
of spindle cells (Mattisson and Fange, 1977). In fact lymphocytes and
spindle cells in Myxine may belong to the same cell line. Destruction
of microtubules by incubation with vinblastine (1pg/ml) or colchicine
causes spindle cells to transform into round lymphocyte-like cells
(Fange et al., 1974).
    Plasma cells have a basophil, usually nongranulated cytoplasm, and
an eccentric spherical nucleus. They occur in connective tissue, rarely
in the blood, and may originate from blast transformed antigen-
activated lymphocytes (B cells). In the electron microscope the cy-
toplasm exhibits numerous ribosomes in a rough endoplasmic reticu-
lum that forms flat or irregular cisternae.
    Downey (1911) studied plasma cells in the renal lymphomyeloid
tissue of the paddlefish, Polyodon spathula, before the immunological
function of these cells was known. The plasma cells in the paddlefish
resemble mammalian plasma cells except in nuclear structure. Good et
al. (1966) observed large numbers of plasma cells in the spleen and the
pericardial lymphomyeloid tissue of paddlefish immunized against
Brucella microorganisms. T h e cells were identified by their ultrastruc-
ture. Plasma cells in lymphomyeloid tissues of elasmobranchs, teleosts,
lampreys, and hagfishes have been identified electron microscopi-
cally (Fujii, 1982; Zapata et al., 1984).The plasma cells are considered
as the main producers of immunoglobulins, but lymphocytes and
other cells probably also produce antibodies (see Section V,D).

F. Monocytes and Macrophages
   Monocytes are mobile, phagocytic cells, usually slightly larger than
other leucocytes. They have a vacuolated, weakly basophilic cy-
toplasm, which lacks distinct granules. The nucleus is generally oval
or kidney shaped. Monocytes constitute only a minute fraction of
the blood leucocytes in fishes and may be lacking in some species.
Sherburne (1973) found no monocytes in the blood of the herring,
Clupea harengus, but Ellis et al. (1976) noted 0.1-0.2% monocytes
among the blood leucocytes in the plaice, Pleuronectes platessa. The
plaice monocytes increased in number after injection of carbon
   Macrophages, or phagocytically active cells in tissues and body
cavities, are supposed to belong to the same cell line as monocytes, but
occasionally macrophages may be related to granulocytes and contain
cytoplasmic granules or may be derived from connective tissue cells.
1.   FISH BLOOD CELLS                                                  25

Tissue bound macrophages are often collectively termed the reticulo-
endothelial system (RES) (Mc Cumber et al., 1982; Page and Rowley,
1984). In teleosts, such as the rainbow trout (Oncorhynchus mykiss),
macrophages are especially abundant in the spleen and the renal lym-
phomyeloid tissue (“pronephros”; Bielek, 1980) but they also occur in
other tissues, for instance the olfactory mucosa (migrating Baltic sal-
monids: Bertmar, 1980). Macrophages that ingest India ink and sheep
red cells are described from the pronephros of the carp, Cyprinus
carpio (Smith et al., 1970).Weak phagocytic responses were observed
when killed bacteria (Bacillus cereus) were injected into the perito-
neum of a teleost, the striped bass (Morone saxatilis).The weakness of
the reaction might have been due to absence of opsonizing effects
because no immunization had taken place (Bodammer, 1986). Melano-
macrophages are pigment containing cells that form nodule-like accu-
mulations in lymphomyeloid tissues (Agius, 1985).
    Yamaguchi et al. (1979) found macrophages in the cavernous bodies
at the base of the gill filaments of lampreys (Larnpetrareissneri). These
phagocytic cells are considerably larger, about 50 pm in diameter, than
the blood granulocytes. The cavernous bodies constitute unique filtra-
tion organs, the extensive phagocytic capabilities of which may keep
the blood free from infections more efficiently than the circulating
granulocytes (Page and Rowley, 1984).
    Granulated and nongranulated macrophages in the peritoneal cav-
ity of hagfishes ingest intraperitoneally injected heat-killed yeast cells
(Myxine: Fange and Gidholm, 1968; Mattisson and Fange, 1977).
Thoenes and Hildemann (1970) used thioglycolate to activate the peri-
toneal macrophages in the California hagfish, Eptatretus.

G. Thrombocytes or Spindle Cells
   Thrombocytes, or spindle cells, usually are oval or spindle shaped.
Contrary to the mammalian platelets (or thrombocytes), they are nucle-
ated and occasionally consist of almost naked nuclei (Ellis, 1977).
Teleostean thrombocytes look like hemoglobin-free, slightly de-
formed erythrocytes or are difficult to distinguish by light microscopy
from lymphocytes. The thrombocytes constitute up to 80.2% of the
white cells in the herring (Clupea harengus) (Sherburne, 1973) but
only 0-7% in other teleost species (Boyar, 1962). The large variation in
thrombocyte numbers probably reflects difficulties in identifying the
thrombocytes and the tendency of these cells to aggregate, adhere to
surfaces, and disappear from the blood samples. In stained smears the
cytoplasm of thrombocytes is nongranulated, grayish blue, or un-
stained. Ultrastructurally it shows mitochondria, ribosomes, glycogen
26                                                      RAGNAR F A N G E

granules, and bundles of microtubules (Ferguson, 1976: plaice, Pleu-
ronectes platessa; Savage, 1983: pike, Esox Zucius). It contains vacu-
oles or vesicles, sometimes arranged like a string of pearls and con-
nected with the surface (Bielek, 1979; Zapata and Carrato, 1980; Hyder
et al., 1983; Cenini, 1984: Cyprinus carpio). Granules similar to
a-granules of mammalian platelets occur in dogfish thrombocytes
(Cannon et al., 1980). Lamellated inclusion bodies, probably phospho-
lipids, are observed in lungfish thrombocytes (Neoceratodus, Tanaka
and Saito, 1981). Dogfish thrombocytes show cytochemical responses
for PAS, p-glucuronidase, and aliesterase (D’Ippolito et al., 1985).
    In the lamprey (Larnpetra fluviatilis) thrombocytes constitute
27.4% of the leucocytes (Page and Rowley, 1983). Spindle shaped cells
constitute about 50% of the leucocytes in the blood of hagfish (Myxine).
They contain mitochondria and microtubules. The spindle cells of
Myxine may easily transform into round lymphocyte-like cells (Fange
et al., 1974; Mattisson and Fange, 1977; see Section on Lymphocytes).

H. Blast Cells
    Blast cells represent early stages of red or white blood cells that
arise by transformation of hemopoietic stem cells. Blast cells have
increased nuclear volume, and the cytoplasm is intensely basophilic
due to numerous ribosomes. Blast cells have the capacity to divide and
form new cells. Rapid synthesis of DNA and RNA is indicated by
intense cellular uptake of [3H]thyn~idine [3H]~ridine        (Fange and
Edstrom, 1973), but few experimental studies have been made on blast
cells in fish blood. Although blast cells are regularly present in small
numbers in fish circulation (Saunders, 1966b; Rubashev, 1969;
Boomker, 1980; Hine et al., 1990a,b),they are ignored by many investi-
gators, probably due to difficulties in distinguishing them from lym-
phocytes and monocytes in microscopic studies. In teleosts the blast
cells, to a great extent, may remain in the hemopoietic tissues, but in
the blood of the cyclostome, hagfish (Myxine glutinosa), erythroblasts,
and leucocytoblasts are abundant (Mattisson and Fange, 1977).


A. Leucocyte Homeostasis
   The frequency of leucocytes in the blood of fishes is influenced by
physiological conditions. The number of leucocytes is affected by
1. FISH   BLOOD CELLS                                                 27

hormones in similar ways as in other vertebrates. In the rainbow trout
(Salmo gairdneri), cortisone and adrenocorticotropic hormone
(ACTH) produce lymphopenia and thrombocytopenia, whereas exper-
imental injections of small amounts of turpentine cause inflammation
and increase of granulocytes (Weinreb, 1958). The amount of circu-
lating white blood cells increases with infection and parasitic infesta-
tion. Drzewina (1911)found high leucocyte counts in the blood of eel
(Anguilla)infected with trypanosomes, and leucocytosis was noted by
Murad and Kustafa (1988) in the catfish, Heteropneustes fossilis, para-
sitized by metacercariae. In higher vertebrates the granulocytes are
relatively short-lived and must be replaced by continuous granulocy-
topoiesis. Conditions in fishes are unknown, but the granulocytes
probably have a short life span in fishes too.

B. Phagocytosis
    Phagocytically active cells in fish blood are mainly neutrophil (het-
erophil) granulocytes and monocytes (mononuclear phagocytes). Mac-
rophages are the main cells responsible for phagocytosis in the perito-
neum, spleen, kidney, liver, gills, and other tissues. Usually the
granulocytes are the most efficient blood phagocytes, but in a holo-
stean, the gar (Lepisosteus platyrhincus) monocytes dominate (McKin-
ney et al., 1977). The phagocytes destroy ingested microorganisms by
chemical mechanisms. Deficiency of phagocytic killing mechanisms
may explain why microorganisms and parasites cause disease in fishes
and other animals (MacArthur and Fletcher, 1985).Phagocytic uptake
of antigens is an important step in the initiation of humoral immune
responses. Special antigen-trapping or antigen-presenting cell systems
may be distinguished. The phagocytically very active cells in the
blood and peritoneal cavity of hagfish, Myxine glutinosa, attack foreign
material such as heat-killed yeast cells without previous immunization
(Fange and Gidholm, 1968).Phagocytically active polymorphonuclear
granulocytes isolated from the blood of the lamprey, Lampetrajapon-
ica, by the Ficoll-Paque technique were studied by electron microscope
(Fujii, 1981). In the lamprey antigenic material (sheep red blood cells)
was more easily ingested by phagocytes if it was bound to specific
lamprey immunoglobulins.
    Braun-Nesje et al. (1981) isolated and cultivated macrophages from
the pronephros of the rainbow trout (Oncorhynchus mykiss). Ferguson
(1984) examined the kidney of rainbow trout in the electron micro-
scope after inoculation with killed bacteria. The bacteria were
phagocytosed by macrophages associated with the endothelium of the
28                                                      RAGNAR FANCE

renal portal veins. The venous renal portal system is a large area for
antigen trapping. Datta Munshi et al. (1990) in cytochemical studies
have investigated damage on fish macrophages caused by heavy
   Phagocytes are affected by nerve transmitters. a-adrenergic and
cholinergic receptor agonists enhance the ability of stimulated
pronephric macrophages and granulocytes to produce reactive oxy-
gen species. Adrenaline and P-adrenergic agonists suppress the
chemiluminescent response of pronephric phagocytes from the rain-
bow trout, Oncorhynchus mykiss (Flory and Bayne, 1991). Hydrocorti-
sone depresses the activity (chemiluminescense response) of striped
bass phagocytes (Stave and Roberson, 1985)indicating that phagocytes
may be controlled by stress hormones.

C. Granulocyte Defense Mechanisms
    Phagocytically active granulocytes destroy bacteria and fungal par-
asites by using enzymes localized in the cytoplasmic granules. Several
antibacterial peptides are isolated from mammalian neutrophils
(Cline, 1981; Lehrer and Ganz, 1990). Similar peptides are present in
fish granulocytes.
   The bacteriolytic enzyme lysozyme (N-acetylmuramylhydrolase) is
active against gram-positive bacteria. It occurs in granulocytes and
peritoneal macrophages of the teleost plaice (Pleuronectes platessa)
(Murray and Flechter, 1976) and may occur in leucocytes of other
species of teleosts. High activities of lysozyme, chitinase, and other
glycosidases are found in granulopoietic tissues of elasmobranchs and
other lymphomyeloid tissues of fishes (Fkinge et al., 1980). Cyto-
chemical observations confirm that fish granulocytes generally are rich
sources of hydrolytic enzymes (Hine et al., 1987). Leucocytic enzymes
besides those mentioned are acid and alkaline phosphatase, sulfatases,
and various esterases. The molecular aspects of fish leucocytic en-
zymes and their biological importance are virtually unknown, but one
might feel tempted to assume that several of the enzymes may play a
role in defense against parasites and microbes, for example by dissolv-
ing cell walls of phagocytically ingested microbes. Other enzymes may
act in cellular production of chemotactic substances or other biologi-
cally active products.
1. FISH   BLOOD CELLS                                                   29

    Peroxidase (myeloperoxidase: MPO) is a component of mammalian
granulocytes and is located in primary or azurophil granules. Peroxi-
dase is demonstrated cytochemically in the neutrophils of most tele-
osts but is lacking in the leucocytes of certain teleost families
(Johansson, 1973; Hine et al., 1987; Hine and Wain, 1 9 8 8 ~ )In the
Australian lungfish, Neoceratodus, and the sturgeon, Acipenser bre-
virostrum, eosinophils but not neutrophils show peroxidase reaction
(Hine and Wain, 1988b; Hine et al., 1990a,b). The enzyme is poorly
represented in elasmobranch granulocytes, and hagfishes lack leuco-
cytic peroxidase (Johansson, 1973).
    Leucocytic peroxidase is supposed to function in a bacteria killing
system of phagocytic granulocytes. During phagocytosis the cells show
an intense increase of oxygen consumption, a “respiratory burst.” This
results in production of superoxide anion ( 0 2 - ) , hydrogen peroxide
(HZOZ),and hydroxyl radicals (OH‘).Under catalytic influence of per-
oxidase hydrogen peroxide reacts with chloride ions producing anti-
bacterial substances. The respiratory burst can be shown by the sub-
stance luminol, which produces light in the presence of oxygen
superoxide. Chemiluminescent phagocytes, probably neutrophils, oc-
cur in the blood and the pronephros of teleosts: striped bass, Morone
saxatilis (Stave et al., 1984); plaice, Pleuronectes platessa (Nash et al.,
1987); rainbow trout, Oncorhynchus mykiss (Plytycz et al., 1989); and
the channel catfish, Lctalurus punctatus (Dexiang and Ainsworth,
1991).Apparently teleosts possess similar oxygen-dependent, bacteria
killing leucocytic mechanisms as those found in mammals.

D. Lymphocyte Functions, Immune Responses
    Vertebrate lymphocytes were long regarded as mature cells, end
products of hemopoietic cell differentiating processes. However, they
have the capacity to develop and undergo blast transformation when
appropriately stimulated. Then the nucleus enlarges and the amount of
ribosomes increases in the cytoplasm. Stimulating substances are lec-
tins or mitogens usually of plant origin. Phyto-haemagglutinin (PHA)
stimulates T lymphocytes to enlarge and divide, and lipopolysaccha-
ride (LPS) stimulates B cells. Formation of plasma cells may be re-
garded as a blast transformation of lymphocytes (B cells) when stimu-
lated by foreign antigens (Cooper, 1983).
   Immunocompetent lymphocytes constitute the basis of immune
30                                                      RAGNAR FANCE

reactions. Fishes, like all other vertebrates, show both cell mediated
and humoral immune responses and may possess functional equiva-
lents to T and B cells of higher vertebrates. Teleostean lymphocytes
respond to mitogens such as PHA, concanavalin A (Con A), and LPS,
which are considered as specific to mammalian subclasses of lympho-
cytes. However, subpopulations of fish lymphocytes may not necessar-
ily be analogous to the mammalian T and B lymphocytes (Warr and
Simon, 1983; Caspi et aZ., 1984). Gilbertson et al. (1986), by flow
cytometry of hagfish (Eptatretus) peripheral blood, separated a frac-
tion of granulocytes and macrophages from a fraction of small cells,
mainly lymphocytes. Heterogeneity of the hagfish leucocytes was
noted using monoclonal antibodies against cell surface antigens. Small
leucocytes reacted in mixed lymphocyte tests as T cells.
    Nonspecific cytotoxic cells or natural killer cells (NK cells) are
thought to play a role in natural immunity of higher vertebrates. They
are cytolytic for cultured tumor cell lines and virus infected cells.
Natural killer cells are also produced in lymphomyeloid tissues of
freshwater teleosts. Thus the channel catfish (Ictalurus punctatus) is
protected against the protozoan disease ichthyophthiriasis by the com-
bined effects of nonspecific cytotoxic cells and a humoral immune
response. Immobilization of the protozoans by immunoglobulins facil-
itates adherence of NK cells to the parasites (Hinuma et al., 1980;
Graves et al., 1985; Greenlee et al., 1991). Cells in the blood of the
nurse shark, Ginglymostoma cirratum, react unspecifically cytotoxic
to red cells of other species after exposure to the mitogens PHA, Con A,
or LPS (McKinney et al., 1977; Mc Cumber et al., 1982; Haynes and
McKinney, 1991).
   Antibody producing cells are assumed to derive from lymphocytes
(B cells), which are influenced by antigens to transform into blast-like
cells, the plasma cells. Antigen presenting macrophages cooperate in
the process. However, it is not clear if plasma cells are the only pro-
ducers of immunoglobulins in fish.
   Antibody forming cells have been demonstrated in the anterior
kidney (pronephros) and the spleen of immunized teleosts by the
Jerne hemolytic plaque assay (Anderson et al., 1979; Smith et al., 1967,
1970), and by rosette test (immunocytoadherence test) (Chiller et al.,
1969a,b).According to Chiller et al. (1969a) most of the rosette forming
cells of the pronephros and spleen of the teleost, the rainbow trout
(Salmo gairdneri), have the appearance of small, medium, or large
1. FISH   13LOOD CELLS                                               31

lymphocytes. More seldom they are plasma cells. Plasma cells are
lymphocyte-like cells involved in especially intense synthesis of im-
munoglobulins, but other kinds of leucocytes may produce antibod-
ies too.
    Plasma cells have been demonstrated by light and electron micro-
scopy in several main groups of fishes, including cyclostomes (Zapata
et al., 1984). In cyclostomes such as hagfish (Myxine, Eptatretus) and
lampreys (Lampetra),the amount of immunoglobulins produced is
estimated at about 1/50ofthat found in mammals. In elasmobranchs up
to 60% of the plasma proteins consist of immunoglobulins, in a chon-
drostean, the paddlefish (Polyodon)about 40% (Litman et al., 1990;
Legler et al., 1971).
    The immunoglobulins of cyclostomes have properties that are in-
termediate between immunoglobulins of vertebrates and invertebrate
lectins. The immune responses of lampreys are weak and slow acting
(Corbel, 1975; Marchalonis, 1977; Fujii et al., 1979). In the Pacific
hagfish, Eptatretus stoutii, about half of the blood leucocytes show
surface immunoglobulins as revealed by radioimmunoassay (Raison
and Hildemann, 1984).
    Lymphocytes are considered to be extraordinarily sensitive to irra-
diation. However, high doses of irradiation (1200, 5000, or 10,000 R)
caused little change in leucocyte counts in the hagfish (Myxine gluti-
nosa), while the white blood cells of the sea lamprey, Petromyzon
marinus, proved very susceptible. The reason for the unusual re-
sistancy of Myxine leucocytes is not known (Finstad et al., 1969).
   Several observations indicate that phagocytosis in fishes is accel-
erated by the presence of antibodies. Michel et al. (1990) reported
opsonizing properties of natural antibodies of rainbow trout (On-
corhynchus mykiss). Scott et al. (1985) showed an opsonin effect on
peripheral blood phagocytes from channel catfish.
   Chiller et al. (1969a,b) showed that lymphocytes and macrophages
from the spleen and kidney of rainbow trout immunized against sheep
red cells form rosette complexes with the sheep cells.
                            ON            SYNTHESIS
    It has been known for a long time that temperature affects synthesis
of antibody in fishes. Although production of antibodies takes place at
32                                                        RAGNAR FANGE

low temperature, it takes a longer time, and it increases if the tempera-
ture goes up (Harris, 1973). Probably antibody production is greater
and biologically more important in species living in warm tropical
water than in fish living in cold water.

    Teleosts, holosteans ( A d a ,Lepisosteus),and polypterids have tet-
rameric immunoglobulins of the IgM type with a molecular weight of
about 600,000 (Warr, 1983; Acton et al., 1971; Bradshaw et al., 1971).
    Elasmobranchs possess pentameric and monomeric immunoglobu-
lins that resemble the mammalian immunoglobulin (Ig)type IgM. The
pentameric IgM has a molecular weight of about 900,000 (Bradshaw et
al., 1971; Acton et al., 1971; Corbel, 1975). Dipnoans (lungfishes)
(Corbel, 1975; Marchalonis, 1977)possess Igs oftwo classes, analogous
to IgM and igG types in mammals. Among elasmobranchs, the spiny
rasp skate, Raja kenojei, and the Aleutian skate, Bathyraja aleutica,
possess two kinds of Igs, a high molecular weight Ig analogous to
mammalian IgM and a low molecular weight Ig. Two distinct popula-
tions of Ig-forming plasma cells were found in the spleen and the
intestinal mucosa. In the Leydig organ, the epigonal organ, and the
liver plasma cells were less frequent (Tomonaga et al., 1985, 1986;
Kobayashi et al., 1985).
    Immunoglobulins do not occur only in blood plasma and mucus
secretions but are also associated with cell membranes of lymphocytes
acting as antigen-specific receptors. Immunoglobulins from splenic
lymphocytes of the goldfish, Carassius auratus, differ from the Ig of'
the plasma (Warr and Marchalonis, 1977).

    Secretory Igs are found in skin mucus of teleosts and holosteans
(Corbel, 1975; Ourth, 1980). The mucous antibodies are of the IgM
type as those of the plasma but further molecular properties are not
known. In the Amazonian discus fish, Symphysodon, young fishes feed
from their parent's epidermal mucus, probably acquiring immunity in
this way. The cells producing secretory antibodies may be leucocytes
infiltrating the skin (Hildemann, 1962).

   The reaction between antibodies and antigen-carrying foreign cells
stimulate proteins of the complement system in the plasma to start a
cascade of enzyme reactions that damage and ultimately destroy cells
covered by an antibody. In mammals the C system consists of nine
1. FISH   BLOOD CELLS                                                 33

components. The C factors have been studied in some fish groups
(Marchalonis, 1977). Hemolytic Cs, extremely labile to freezing, are
found in the blood of elasmobranchs (Legler and Evans, 1967). Jensen
et al. (1981) investigated the C system of the nurse shark, Gingly-
mostoma cirratum. It consists of only six components, but in a
mammalian-like way it forms holes in the stroma of antibody-
sensitized sheep erythrocytes. Teleost complement is species spe-
cific (bluegills, salmonids: Smith et d., 1967, Chiller et d., 1969b).
Complement-like activity is detected in lampreys (Lampetra japon-
ica). However, information on structure and function of C systems in
lower vertebrates are fragmentary (Fujii and Murakawa, 1981). The C
factors are supposed to be produced by liver cells and by leucocytes.
   Fish blood plasma contains nonspecific lectin-like proteins or gly-
coproteins, different from immunoglobulins, which provide some
protection against infections. These substances may act together with
the immune mechanisms (Ingram, 1980).
                       ON          PROCESSES
    The teleost spleen has a rich adrenergic innervation, which may
affect the antibody producing cell system. According to Flory and
Bayne (1991), damage of the adrenergic nerves increases secretion of
antibodies. In the rainbow trout, Oncorhynchus mykiss, a-adrenergic
and cholinergic agents increase the number of antibody secreting cells
while P-adrenergic substances have the opposite effect.
   Chemical pollutants in the environment are thought to depress fish
immune mechanisms leading to an increase of infections or tumors
(Vos et al., 1989). Production of stress hormones such as corticosteroids
may also unfavorably affect immune functions (Grimm, 1985).
    Anaphylactic reactions are thought to be caused by substances that
are released from mast cell granules, when Igs of IgE type react with
specific antigens at the cell surface. A few observations indicate that
anaphylaciic-like hypersensitivity reactions exist in fishes (Goven et
al., 1980; Ellis, 1982). Cells responsible could be PAS-positive granu-
locytes or EGC, but they lack histamine and Igs of the IgE type are
unknown in fishes. The question of anaphylactoid responses in fishes
is unsolved (Ellis, 1977, 1982).
34                                                       KAGNAH FANGE

   Several methods of vaccination against infections of fish in aquacul-
ture have been developed. Intraperitoneal injection of vaccines is
superior to other methods in experiments but may also be used on a
commercial scale. Other methods build on adding the vaccines to the
food or to the water (Horne et al., 1984; Ellis, 1988).

E. Pathology ZnAammation
    Estimation of erythrocyte sedimentation rate (ESR) is an unspecific
method used in human medicine to detect infections. The method has
been tried in hematological studies on salmonids but probably is of
little practical value (Blaxhall and Daisley, 1973). Schumacher e t al.
(1956) reported increased ESR in brook trout infected by furunculosis,
and Murad and Mustafa (1988) found a similar response in catfish
(Heteropneustes fossilis) parasitized by metacercariae and also re-
ported low hematocrit and increased white cell counts in the diseased
    Pathological processes caused by infections or parasites are rela-
tively rarely examined in fishes, but leucocytes are known to be en-
gaged. The role of lymphocytes in inflammatory processes in fishes is
described b y Ellis (1976). In mirror carp (Cyprinus carpio L.) infection
with lchthyophthirius causes a rise in the percentage of neutrophils
and granuloblasts and a temporary drop in lymphocytes in the blood
(Wines and Spira, 1973). Appearance of immature neutrophils with
intensely basophilic cytoplasm and reduced numbers of granules are
seen in inflammatory conditions in the eel, A n g u i h australis (Hine
and Wain, 1988a).
    Suzuki (1986) obtained peritoneal exudate cells b y injection of
liquid paraffin into tilapia (Oreochromis niloticus) and carp (Cyprinus
curpio). Eosinophils appeared later in the exudate and were less
phagocytic than macrophages-monocytes and neutrophils. Hyder
Smith e t al. (1989) found that in the nurse shark (Ginglymostoma
cirrhatum) granulocytes and mononucleate macrophages, like mam-
malian neutrophils and monocytes-macrophages, react chemotac-
tically to endotoxin-activated rat serum.
    Infection of paddlefish, Polyodon spathula, with larval nematodes
causes ulcers of the gastric mucosa with accompanying infiltration with
eosinophils. The lymphomyeloid tissue of'the epicardium reacts with
1. FISH BLOOD   CELLS                                                   35

extensive proliferation of macrophages (Miyazaki et al., 1988). Phrom-
suthirak (1977) followed leucocytic responses at healing of an incision
in the skin of Gasterosteus aculeatus. Macrophages, neutrophils, eo-
sinophils, and lymphocytes accumulated in the skin. Leucocytes mi-
grated into the skin from the blood. The number of neutrophils
reached a peak after 1 day, that of macrophages after 3 days.

F. Thrombocytes and Blood Coagulation
    In cyclostomes and elasmobranchs blood clotting is relatively slow,
in teleosts and lungfishes very fast (Doolittle and Surgenor, 1962;
Ward, 1969). Addition of seawater accelerates blood clotting in hagfish
( M y x i n e )(Fange and Gidholm, 1973) and in an elasmobranch (Hetero-
dontus) (Stokes and Firkin, 1971).
    In mammals hemostasis (i,e., stoppage of bleeding from damaged
vessels) results from interaction of vasoconstriction and processes tak-
ing place in the blood. The latter involve (a) aggregation of platelets
leading to plug formation and ( b )blood coagulation. Hemostatic re-
sponses of fishes have not been much investigated, but evidences are
that similar mechanisms are at work as in other vertebrates. Vertebrate
blood coagulation results from a cascade of reactions producing an
enzyme, thrombin, that splits fibrinogen into insoluble fibrin and pep-
tides. Two pathways or systems are distinguished. The extrinsic sys-
tem is initiated by tissue factors, the intrinsic by factors from platelets
or other blood cells. The nucleated thrombocytes of fish blood partici-
pate in the coagulation process in the conversion of prothrombin to
thrombin and in clot retraction (Doolittle and Surgenor, 1962; Bela-
marich et al., 1962; Fey, l966a; Gardner and Yecish, 1969; Stokes and
Firkin, 1971; Rowley et al., 1988),but detailed knowledge on the roles
played b y thrombocytes and other cells in fish blood coagulation is
    Stobbe (1963) observed in phase-contrast microscope that the re-
markably large thrombocytes of the salamander, Amphiuma, have a
tendency to aggregate and undergo “viscous metamorphosis.” Similar
observations by Wardle (1971) on thrombocytes in lymph from the
supraneural duct of the plaice (Pleuronectesplatessa) show that, in the
absence of anticoagulants, thrombocytes send out filaments that attach
to the glass, after which radiating fibrin threads link the thrombocytes
together. Boomker (1980)briefly reports that reactive stages of throm-
bocytes, similar to those in avian blood, are observed in the catfish,
Clarias. Phase-contrast microscopy of coagulating blood of the hagfish
36                                                         RAGNAR FAKE

(Myxine glutinosa) blood show disintegrating leucocytes to form cen-
ters of clot retraction. The exact morphological type of leucocytes
initiating the process could not be identified but probably spindle cells
or lymphocyte-like cells were involved (Fange and Gidholm, 1973;
Mattisson and Fange, 1977).
    Small amounts of ADP cause mammalian platelets to aggregate,
whereas serotonin (5-HT) has a similar effect on the nucleated throm-
bocytes in avian blood (Stiller et al., 1975). The factors that induce
aggregation of thrombocytes in fish blood are unknown. Belamarich et
al. (1968) found no aggregating effect of adenosine diphosphate (ADP)
on thrombocytes in any nonmammals. and contrary to mammalian
platelets the thrombocytes of dogfish (Mustelus canis) do not accumu-
late or produce serotonin (Belamarich et al., 1962).
   Inclusion bodies, probably consisting of phospholipid membranes,
occur in thrombocytes of lungfishes (Lepidosiren, Protopterus: Tanaka
and Saito, 1981). In annelid coelomocytes somewhat similar inclusions
form myelin figures during a clotting-like process (Enchytraeus:
Fange, 1951). In mammalian platelets lamellar bodies are visualized
by the use of tannic acid (Baker et al., 1982). The “viscous metamor-
phosis” of activated platelets and thrombocytes may be explained by
myelin figures developing from intracellular phospholipid particles.
The “platelet factor 3” functioning in mammalian blood coagulation
probably is a phospholipid.
    Anticoagulants used in hematological studies on fishes are EDTA,
citrates, oxalates, and heparin. Smit and Hattingh (1980) found heparin
to be the most suitable anticoagulant in experiments on freshwater
fishes. The anticoagulatory effect of heparin is assumed to be caused
by activation of antithrombin, a protein that inhibits coagulation en-
zymes. Curiously, Jordan (1983) was unable to isolate antithrombin
from the blood of various fish species. It seems as if in fishes the plasma
contains heparin dependent coagulation inhibitory factors with
properties different from those of the antithrombins of other verte-


   Lymphocytes and other blood cells originiate from and are stored in
so-called lymphomyeloid tissues. These do not always form distinct
1. FISH   BLOOD CELLS                                                 37

organs but may consist of temporary accumulations of proliferat-
ing blood cells. A continuous migration of lymphocytes unites the
different structures into a “lymphomyeloid complex” (Yoffey and
Courtice, 1970; Yoffey, 1985). Although lymph nodes and hemo-
poietic bone marrow are lacking, fishes have a rich array of lympho-
myeloid structures. The thymus and the spleen belong to the
lymphomyeloid complex of tissues. In total, the lymphomyeloid
tissues of fishes compose 0.5-1.5% of the body weight (Fange,
1987), or about half the percentual weight of analogous tissues in

A. Thymus
    A thymus exists in all fishes except cyclostomes, but diffuse accu-
mulations of lymphocytes in the gill region of larval lampreys are
supposed to represent a “protothymus.” In holocephalans and chon-
drosteans the thymus is well developed, lobated, and organized in
cortex and medulla as in mammals, but no Hassall’s corpuscles
are found (Fange and Sundell, 1968; Fange, 1986b). The teleost
thymus has an intraepithelial position inside the epithelium of the
gill chamber (rainbow trout, Oncorhynchus mykiss: Chilmonczyk,

B. Spleen
   A spleen is present in all fishes, but in cyclostomes it is represented
by lymphomyeloid aggregations in the typhlosole (lamprey) and the
intestinal submucosa (hagfish). The structures are unlike real spleens
(hagfish: Tomonaga et al., 1973; Fujii, 1982)but they may serve similar
functions. In elasmobranchs, teleosts, and chondrosteans (sturgeons)
the spleen consists of red and white pulp, although the boundary
between these is diffuse. In elasmobranchs (sharks, rays) the lymphoid
white pulp areas are more distinct than in teleosts, and in the white
pulp of the spleen of rays (Dasyatis, Myliobatis) follicle-like structures
occur (Tomonaga et al., 1986). However, it may be difficult to distin-
guish between lymphoid and erythropoietic centers. In elasmo-
branchs as in holocephalans (rabbit fish, Chimaera) and a series of
teleost species, the spleen is the primary hemopoietic organ (Fange
and Nilsson, 1985). It is also an important immune organ contain-
ing plasma cells and phagocytes and has a blood filtering, antigen-
capturing function.
38                                                       RAGNAR FANGE

C. Kidney
    The kidney of most fishes contains lymphomyloid tissue, but holo-
cephalans (chimaeroids, rabbit fishes) and many elasmobranchs are
exceptions. The pronephros or head kidney, the anterior part of the
kidney, is a complex tissue found in many teleosts. It contains lympho-
myeloid, renal and endocrine components richly supplied with blood
from arteries and caudal portal veins and innervated from the sympa-
thetic ganglionic chain. It serves as the analog of bone marrow, lymph
nodes, and, in part, the adrenal gland of higher vertebrates. It is the
main hemopoietic organ in many teleosts producing erythrocytes,
granulocytes, lymphocytes, macrophages, thrombocytes, and plasma
cells, and it is a main source of antibodies. Both granulopoietic and
erythropoietic areas can be identified histologically in the pronephros
of teleosts. Smith et al. (1970)describe the histology of the pronephros
of the carp as resembling the subcortical region of mammalian lymph
nodes, while Zapata (1981) emphasizes the similarity between tele-
ostean pronephros and mammalian red bone marrow. In some teleosts
lymphomyeloid tissue extends into areas of the ordinary kidney, the
mesonephros. Like the spleen the lymphomyeloid structures of the
kidney are rich in macrophages and are supposed to have a blood
filtering, antigen-capturing function.

D. Lymphocytic Infiltrations
    Lymphocytes occur in all lymphomyeloid tissues, and these may
contain follicle-like accumulations of lymphocytes, although real folli-
cles or germinal centers, such as found in avian and mammalian lym-
phoid tissues, probably never occur. Lymphocytes and other types of
leucocytes infiltrate mucosal and other membranes all over the organ-
isms. They occur regularly in the intestinal mucosa and submucosa
giving rise to voluminous accumulations in the spiral intestinal valve
of elasmobranchs (bullhead shark, Heterodontus: Tomonaga et al.,
1985; rays, Dasyatis, Myliobatis: Tomonaga et al., 1986) and chondro-
steans (sturgeons, and paddlefish, Polyodon: Weisel, 1973). Rich de-
velopment of intestinal lymphoid tissue as in the paddlefish may be
related to the presence of parasites. In the massive lymphoid accumu-
lations in the spiral valve of rays, follicle-like structures have been
    Lymph node-like cell masses in the pericardium of the heart of
sturgeons (Acipenser) and paddlefish (Polyodon) contain large num-
bers of lymphocytes that migrate through the endothelium of
venous sinuses (Fange, 1986a). Interactions between immunocompe-
1. FISH   BLOOD CELLS                                                 39

tent lymphocytes and endothelial cells may have physiological impli-
cations (Baldwin, 1982).

E. Granulo(cyt0)poietic Tissues
    A whitish, mainly granulopoietic bone marrow-like tissue is found
in the esophagus (Leydig organ), the gonads (epigonal organ), and
occasionally in the kidney of elasmobranchs. Similar tissue, analogous
to the red bone marrow of terrestrial vertebrates but producing leuco-
cytes only, forms a “cartilage marrow” in recesses and canals of the
skeleton of holocephalans (Chimaera)(Fange, 1987; Mattisson et al.,
1990). Granulopoiesis also takes place in the meninges of chondro-
steans (sturgeon, Acipenser, paddlefish, Polyodon) and holosteans
(Amia,Lepisosteus). In lungfish (Lepidosiren,Neoceratodus) granulo-
cyte producing lymphomyeloid tissue is distributed around the kid-
ney, gonads, spleen, and pancreas, but histological investigations are
impeded by the abundance of large pigmented cells (Bargmann, 1934;
Rafn and Wingstrand, 1981). Extensive lymphomyeloid masses are
found in the viscera of the coelacanth, Latimeria chalumnae, but com-
plete anatomical data are missing (Millot et al., 1978; Locket, 1980).

F. Melanomacrophage Centers
   Accumulations of macrophages and pigment into melanomacro-
phage centers are found in lymphomyeloid structures in many fishes,
especially in the spleen. According to one theory these centers are
primitive analogues of the germinal centers or secondary follicles in
avian and mammalian lymphoid tissues. However, there are major
differences between melanomacrophage centers and germinal centers
(Agius, 1985), and their importance is unclear. They are involved in the
metabolism and store iron as hemosiderin, but they may have other
functions as well.


   Hemopoiesis is the production of cells and fluid of the blood, but
usually the term is restricted to cells.

A. Stem Cells
   According to current views (Cline and Golde, 1979; Hoffbrand and
Pettit, 1980) vertebrate blood cells arise from pluripotent stem cells in
40                                                      RAGNAR FANGE

hemopoietic tissues. The concept of stem cells in mammalian hematol-
ogy is founded on experiments in which cells from nonirradiated
animals form colonies in the spleen when injected into irradiated
animals (Weiss, 1981). Such experiments have not been made on
fishes. Present knowledge about piscine stem cells is founded on
analogy with mammalian hemopoiesis and indirect morphological evi-
    Yoffey (1985)considers hemopoietic stem cells (mammalian) to be
identical with transitional cells. These are lymphocyte-like with a
basophilic cytoplasm, a leptochromatic nucleus, and a high tendency
to incorporate tritiated thymidine. Attempts have been made to find
and characterize hemopoietic stem cells in fishes. The descriptions by
Jordan and Speidel (1924) and Jordan (1938) of “lymphoid hemo-
blasts” or “hemocytoblasts” indicate that these function as stem cells,
but the conception of lymphoid hemoblasts is criticized by Ellis
(1977). Zapata (1981) in an electron microscopic study found cells in
the teleost kidney similar to transitional cells of mammalian bone

B. Tissue Microenvironment
    Comparative studies led Jordan (1933) to assume that erythropoie-
sis occurs in tissues with a sluggish or stagnant sinusoidal venous
circulation, because blood with a high carbon dioxide concentration
was needed for the synthesis of hemoglobin. In fishes the portal circu-
lation of the kidney and the sinusoidal venous blood flow of the splenic
red pulp seem to favor erythropoiesis. Thus in most teleosts (bony
fish), in chondrosteans (sturgeons, paddlefish), and holosteans (bow
fin, gar) erythropoiesis occurs in the kidney. And in other teleosts
(Tautoga and Stenotomus: Jordan and Speidel, 1924; Perca: Catton,
1951; Scorpaena: Fey, 1965), in elasmobranchs (sharks, rays), and in
holocephalans (rabbit fish: Chimaera) the spleen is the main erythro-
poietic organ. In larval lamprey (Lampetra) the typhlosole (primitive
spleen), together with the kidney, is the major site of hemopoiesis
(Fujii, 1982).
    According to Jordan (1933)erythrocytes and thrombocytes (spindle
cells) form inside and granulocytes form outside the blood vascular
system. The formation of leucocytes needs a sparse blood supply.
Granulocytes differentiate within sparsely vascularized mesenchymal
connective tissue. In accordance with these theories the granulopoie-
tic tissues of holocephalans and elasmobranchs receive only sparse
arterial supply. However, the tissues are associated with prominent
venous sinuosities (Stahl, 1967; Fange, 1986a). T h e granulopoietic
1.   FISH BLOOD CELLS                                                 41

areas of the teleost pronephros would seem to have a rich blood supply,
which does not fit well with the theories given. It is of interest to
examine in greater detail the microcirculation and microenvironment
of hemopoietic tissues in fish.
    Accumulation and proliferation of lymphocytes seem to occur at
anatomical sites where fluids are filtered such as mucosal membranes,
renal tissue, and the pericardium (chondrosteans: sturgeons, pad-

C. Factors Stimulating Hemopoiesis
    The stem cells grow, multiply, and develop into different lines of
blood cells if adequately influenced. Stimulatory factors may include
hormones and microenvironmental factors (suitable concentrations of
oxygen, carbon dioxide, nutritional substances, metabolites). The
developing blood cells seem to possess receptors for hormones
such as erythropoietin, which stimulate differentiation into specific
cell lines. However, whether or not a specific “leucocytopoietin”
exists is an open question. Certain observations indicate that
erythropoietin mechanisms exist in fishes. Anemia after bleed-
ing or phenylhydrazine-induced hemolysis stimulates new formation
of red cells and hemoglobin (Cameron and Wohlschlag, 1969; Smith et
al., 1971). A few days after withdrawal of blood the number of erythro-
blasts and proerythrocytes increased to 55% of the total number of
red cells in the blood of the eel (Anguilla anguilla) (Kreutzmann,
1976a), and in the gar (Lepisosteus platyrhincus) erythroblasts in-
creased to 26%. In the latter case, hemopoiesis was stimulated by
anemia but not by hypoxia (McLeod et al., 1978). Blood plasma from
experimentally anemic fishes stimulates erythropoiesis and hemoglo-
bin formation in the tropical fish Blue Gourami (Trichogaster trichop-
terus) (Zanjani et al., 1969; Yu et al., 1971), and mammalian urinary
erythropoietin, although in high doses, increases erythropoiesis in the
Blue Gourami (Zanjani et al., 1969)and in the tropical teleosts, Clarias
batrachus and Heteropneustes fossilis (Pradhan et al., 1989). But fish
erythropoietins have not been isolated and analyzed.

D. Erythropoiesis in the Peripheral Blood
    The late phase of fish erythropoiesis, including hemoglobin synthe-
sis, occurs in the circulating blood. The earliest red cells in the blood
are erythroblasts, spherical cells with a cytoplasm, which is rich in
RNA and stains intensely blue with Giemsa. The erythroblasts possess
receptors for transferrin, the iron-transporting protein of the blood
42                                                       RAGNAR F A N C E

(Fletcher and Huehns, 1968). Uptake of iron into immature red cells
goes on in the blood (Tinca:Hevesy et al., 1964; Lepisosteus: McLeod
et al., 1978). The immature cells continue to synthesize hemoglobin
until synthesis ends, and they lose RNA and transform into adult
    The blood of hagfish (Myxine)is the site of both the early prolifera-
tive and the late differentiating phase of erythropoiesis. It may be
regarded as a fluid red bone marrow. It contains mitotically dividing
erythroblasts (Mattisson and Fange, 1977). These show strong nuclear
uptake of tritiated thymidine, indicating synthesis of DNA, while spin-
dle cells show intense nuclear uptake of uridine (Fange and Edstrom,
1973). Undifferentiated spindle cells or lymphocytes may b e stem cells
in the sense of Jordan’s (1938) lymphoid hemoblast theory. The spin-
dle cells easily transform into lymphocytes (Fange et al., 1974). When
stimulated by the mitogen PHA spindle cells of Myxine blood trans-
form into erythroblast-like cells (Fange and Zapata, 1985). Several
observations, although preliminary, indicate that spindle cells and
lymphocytes of hagfish are undifferentiated cells with a considerable
capacity of growth and differentiation. Concerning late hemopoiesis
Tomonaga et al. (1973) have shown by autoradiography that in hagfish
(Eptatratus burgeri) incorporation of iron into erythroblasts does not
take place in the hemopoietic tissues but probably in the blood. A
transferrin in the blood plasma, similar to that of other vertebrates
(Aasa, 1973), probably supplies iron.

E. Toxic Effects on Erythropoiesis
      Chronic exposure of fish to sublethal concentrations of cadmium
impedes erythropoiesis and hemoglobin formation ( Johansson-
Sjobeck and Larsson, 1978: Carassius; Houston and Keen, 1984). Lead,
absorbed from very low concentrations in the surrounding water, in-
hibits the enzyme S-aminolevulinic acid dehydratase, which is in-
volved in hemoglobin synthesis ( Johansson-Sjobeck and Larsson,
1979). Copper significantly changes hemoglobin values in the blood of
freshwater fish (McKim et al., 1970; Christensen et al., 1972: Salmo,
I c t a lurus).


A. Gaps in Knowledge
  The hematology of teleostean fishes is better known than that of
most other fish groups, but many dark points remain. Variations in
1. FISH   BLOOD CELLS                                                  43

terminology have created minor confusions, which ought to be over-
come. Much work is required to reach a better understanding of the
function and the hemopoietic origin of fish blood cells. The concept of
cellular migration streams (Yoffey, 1985) may be useful in investiga-
tions on leucocytic function.
    The blood cells of phylogenetically important fishes are rarely stud-
ied. Although some aspects of the blood of the coelacanth, Latimeria
chalumnae, are known (Weber et al., 1973; Gorr et al., 1991),the blood
cytomorphology of our closest relative among fishes living now (Gorr
et al., 1991) is practically unknown. Also the blood cells of dipnoans
(lungfishes), holosteans (gars, bowfin), chondrosteans (sturgeons, pad-
dlefish), and brachiopterygians (bichirs, Polypterus) have been insuf-
ficiently investigated.

B. Hemopoiesis
    In elasmobranchs, erythrocytes and granulocytes are produced in
different tissues, whereas in teleosts both kinds of blood cells originate
in the same tissue, usually the head kidney (pronephros). The factors
that lead to accumulation of multipotent stem cells in certain tissues
and to differentiation of these cells into different lines of blood cells
are poorly known. Investigations are needed on the importance of
endocrine factors (erythropoietin) and of tissue hypoxia (or hypercap-
nia?) for red cell formation.

C. Lymphocyte Functions
    Few vertebrate cells are more important than the lymphocytes.
Hagfishes (myxinoids) seem central in understanding the evolution of
lymphocyte functions. The conspicuous lymphocytes/spindle cells of
Myxine glutinosa may play roles in both hemopoiesis and immune
functions. Responses of myxinoid lymphocytes to various mitogens are
important to investigate. Vertebrate lymphocytes generally are very
sensitive to irradiation. The remarkably low susceptibility of hagfish
lymphocytes (Finstad et al., 1969) is still waiting for explanation.
    Plasma cells are supposedly derived from lymphocytes (B cells)
that have been activated by immunization processes. Plasma cells are
found in most major groups of fishes. It is not known if cells, identified
as plasma cells on morphological criteria, are the only source of Ig in
    Investigations on functional categories of fish lymphocytes are in a
preliminary stage. A thymus probably exists in all fishes except cyclos-
tomes, but it is not clear if it functions in maturation of T lymphocytes
44                                                      RAGNAR FANGE

as in mammals. The existence of cytotoxic lymphocytes, NK cells, and
similar cells in fishes has to be further investigated.

D. Blood Coagulation
   The importance of activation of thrombocytes (spindle cells) or
other types of cells during fish blood coagulation is poorly under-
stood. The number and chemical nature of the coagulation factors
are imperfectly known, and the clotting mechanisms may differ be-
tween systematic groups of fishes. Facts are few and need to be

E. Granulocytic Function
    The abundant and prominent granulated white blood cells of elas-
mobranchs and dipnoans may have functions that are still unknown.
Probably, together with lymphocytes and macrophages, they partici-
pate in the defense against parasites and microbes. The great diversity
of granulocyte types may indicate a variety of functions. Enzymes
released from granulocytes could influence growth or repair in the
tissues, blood coagulation, and so on. In spite of some histochemical
work having been done, the composition and function of substances,
which may be released from the leucocytic granules, are practically
unknown. The blood and the granulocyte-producing lymphomyeloid
tissues of large sharks may be rich sources for investigations of the
properties of leucocytic granules. In this connection it is well to re-
member that sharks, according to some popular views, are remarkably
free from cancer diseases. However, scarce tumors, probably benign,
have been observed (Harshbarger, 1981).
    Certain types of teleost granulocytes may kill ingested bacteria by
mechanisms similar to those of mammalian neutrophils. It is not
known if the eosinophils of elasmobranchs and labrids (and some other
teleosts) resemble the mammalian eosinophils in showing activities
directed against parasites. The “heavy granulocytes” and other odd
types of fish granulocytes require further investigations.

F. Electron Microscopy
    The results of electron microscopical studies need to be better
correlated with those obtained by light microscopy. Probably electron
microscopy, in combination with experimental methods, will be of use
to improve the complicated classification of fish granulocytes.
1.   FISH BLOOD CELLS                                                             45

G. Immune System of Long-Lived Fishes
   Large fishes with a long life presumably have evolved more effi-
cient immune mechanisms than small short-lived forms. Long-lived
species are found among chondrosteans (sturgeons), elasmobranchs
(sharks), dipnoans (lungfishes), and certain teleosts (pike, Esox Zucius;
the wels, Silurus glanis). Certain chondrosteans such as the American
white sturgeon, Acipenser montanus, get sexually mature at around
20 years of age and may live more than 80 years (Gahlbreath,
1979). During their long, life sturgeons must resist influences from
bacteria, viruses, parasites, and cancerogenic agents. However, few
works have been done on the hematology and immunology of stur-

H. Microcirculation of Hemopoietic Tissues
    The distribution of lymphomyeloid tissues at different anatomical
sites of the fish organism is not only a comparative anatomical problem.
The fact that erythropoiesis predominates in some tissues and lympho-
poiesis or granulopoiesis in others may partly be due to local differ-
ences in circulation. Studies of the tissue microenvironment of
proliferating and differentiating stem cells is an important research

I. Cell Interactions
    In red bone marrow erythroblasts agglutinate around macrophages
forming erythroblastic islands (Bernard, 1991), nurse cells and lym-
phocytes interact in the thymus (Wekerle and Ketelsen, 1980), and
something goes on between lymphocytes and capillary endothelial
walls, perhaps production of substances causing lymphocyte accumu-
lation (Baldwin 111,1982).These examples are from mammals. Studies
of analogous phenomena in fishes may enlighten general biological


   I thank A. Mattisson for electron micrographs and for reading the manuscript, Inger
Holmqvist for technical work, and D. I. Randall for constructive advice.
46                                                                   RAGNAR FANGE


Aasa, R. (1973).Acta Reg. Soc. Sci. Litt. Gothoburg., Zool. 8,46.
Acton, R. T., Weinheimer, P. F., Hall, S. J., Niedermeier, W., Shelton, E., and Bennett,
    J. C. (1971).Proc. N t l . Acad. Sci. 68, 107-111.
Agius, C. (1985).In “Fish Immunology” (M. J. Manning, and M. F. Tatner, eds.), pp.
    85-105. Academic Press, New York.
Allison, A. (1968). In “Advances in Chemotherapy” (A. Goldin et a/., eds.), VO].3, pp.
    253-302. Academic Press, New York, London.
Anderson, D. P. (1974). Fish immunology. In “Diseases of Fishes” (S. F. Snieszko, and
     H. R. Axelrod, eds.), Book 4. T. F. H. Publ., Hong Kong.
Anderson, D. P., Roberson, B. S., and Dixon, 0.W. (1979).]. Fish Bid. 15,317-322.
Andrew, W. (1965). “Comparative Hematology.” Grune and Stratton, New York,
Antonioli, A. (1961). In “Biological Activity of the Leucocyte” (G. E. W. Wolstenholme,
     and M. O’Connor, eds.) pp. 92-99. Ciba Foundation, Churchill, London.
Arvy, L. (1955).C. R . SOC. Biol. Paris 149,2131-2134.
Asai, H., Takagi. H., and Tsunoda, S. (1976). Comp. Biochem. Physiol. B 55,69-75.
Bachand, L., and Leray, C. (1975).Comp. Biochem. Physiol. B 50,567-570.
Baker, N. J., De, C., and Reid, C. (1982).J.Anat. 135,539-548.
Baldwin, W. M., 111 (1982). Immunol. Today 3,267-268.
Barber, D. L., and Westermann, J. E. M. (1975).Am. J. Anat. 142,205-220.
Barber, D. L., and Westermann, J. E. M. (1978).J. Fish Biol. 12,35-43.
Barber, D. L., Westermann, J. E. M., and White, M. G. (1981).J.Fish Biol. 19, 11-28.
Bargmann, W. (1934). Zeitschr. f.Zellforsch. 21,388-411.
Bartlett, G. R. (1982).Comp. Biochem. Physiol. A 73, 135-140.
Bedford, J. J. (1983). Comp. Biochem. Physiol. A 7 6 , 8 1 4 4 .
Belamarich, F. A., Doolittle, R. F., and Surgenor, D. M. (1962). Biol. Bull. 123,479.
Belamarich, F. A., Shepro, D., and Kien, M. (1968).Nature 220,509.
Bernard, J. (1991). Blood Cells 17,5-14.
Bertmar, G. (1980).Olfaction and Taste 7, 159-162.
Bielek, E. (1975).ZooLJb. Anat. 95, 193-205.
Bielek, E. (1979).Zool.Jb. Anat. 101, 19-26.
Bielek, E. (1980). Z o o l . / b . Anat. 103,498-509.
Bielek, E. (1981). Cell Tissue Res. 220, 163-180.
Bjerring, H. C. (1985). Zn “Evolutionary Biology of Primitive Fishes” (R. E. Foreman et
     a/., eds.), pp. 31-57. Plenum Press, New York, London.
Blaxhall, P. C. (1972).J.Fish Biol. 4, 593-604.
Blaxhall, P. C., and Daisley, K. W. (1973).J.Fish Biol. 5, 771-782.
Blaxhall, P. C., and Sheard, P. R. (1985).J. Fish Biol. 26,209-216.
Bly, J. E., and Clem, L. W. (1988).Comp. Biochem. Physiol. A 91,481-485.
Bodammer, J. E. (1986).Vet. Immunol. Zmmunopato/. 12,127-140.
Bolls, L., and Fange, R. (1979).Comp. Biochem. Physiol. B 62,345-348.
Bolis, L., Luly, P., and Baroncelli, V. (1971).J.Fish Biol. 3,273-275.
Bolton, L. L. (1933).J . Morph. 54, 549-592.
Boomker, J. (1980).Onderstepoort J. Vet. Res. 47,95-100.
Boyar, H. C. (1962). Copeia (2),463-465.
Bradshaw, C. M., Clem, L. W., and Sigel, M. M. (1971).]. Zmmunol. 106,1480-1487.
Braun-Nesje, R., Bertheussen, K., Kaplan, K., and Seljelid, R. (1981). J. Fish Dis. 4,
1. FISH BLOOD CELLS                                                                     47

Bridges, D. W., Cech, J. J. Jr., and Pedro, D. N. (1976). Trans. Am. Fish. S O C . 105,
Bushnell, P. G., Nikinmaa, M., and Oikar, A. (1985). Comp. Biochem. Physiol. C 81,
    39 1-394.
Cameron, J. N., and Wohlschlag, D. E. (1969).J . E x p . Biol. 50,307-317.
Cannon, M. S., Mollenhauer, H. H., Cannon, A. M., Eurell, T. E., and Lewis, D. H.
    (1980).Can.J.Zool. 58, 1139-1143.
Carlsson, U., Kjellstrom, B., and Antonsson, B. (1980). Biochim. Biophys. Acta 612,
    160- 170.
Caspi, R. R., Shahrabani, R., Kehati-Da, T., and Avtalion, R. R. (1984). Deo. Cornp.
    Immunol. 8,61-70.
Catton, W. T. (1951).Blood cell formation in certain teleost fishes. Blood 6 , 3 9 4 0 .
Cenini, P. (1984).J . Zool. (London)204,509-520.
Chiller, J. M., Hodgkins, H. O., Chambers, V. C., and Weiser, R. S. (1969a).J. Zmmunol.
Chiller, J. M., Hodgkins, H. O., and Weiser, R. S. (1969b).J.Immunol. 102, 1202-1207.
Chilmonczyk, S. (1985).In “Fish Immunology” (M. J. Manning, and M. F. Tatner, eds.).
    Academic Press, London, New York.
Christensen, G . M., Fiandt, J. T., and Poeschl, B. A. (1978).J.Fish Biol. 12,51-60.
Christensen, G. M., McKim, J. M., Brungs, W. A., and Hunt, E. P. (1972). Toxic. Appl.
    Pharmacol. 23,547-551.
Clem, L. W., and Small P. A., Jr (1967).J.E x p . Med. 125,893-920.
Cline, M. J. (1981). “Leukocyte Function.” Churchill Livingstone, New York,
Cline, M. J . , and Golde, D. W. (1979).Nature 277, 177-181.
Cohen, W. D. (1986).Biol. Bull. 171,338-349.
Cohen, W. D., Bartelt, D., Jaeger, R., Langford, G., and Nemhauser, I. (1982).J.  Cell Biol.
Cooper, E. L. (1983).“General Immunology.” Pergamon Press, Oxford, New York.
Corbel, M. J. (1975).J.Fish Biol. 7, 539-563.
Cowey, C. B., and Sargent, J . R. (1977).Ado. Mar. Biol. 10,383-492.
Cushing, J. E. (1970).In “Fish Physiology” (W. S. Hoar, and D. J. Randall, eds.), Vol. 4,
    pp. 465-500. Academic Press, London.
Datta Munshi, J. S., Singh, N. K., Mishra, N., and Ojka, J. (199O).J.Fish Biol.37,617-622.
Davina, J. H. M., Rijkers, G. T., Rombout, J. H . M., Timmermans, L. P. M., and van
    Muiswinkel, W. B. (1980). In “Development and Differentiation of Vertebrate Lym-
    phocytes” ( J . D. Horton, ed.). Elsevier/North Holland Biomed. Press, Amsterdam.
D’Avino, R., Caruso, C., and Schinina, M. E. (1990). Comp. Biochem. Physiol. B 96,
Dawson, A. B. (1932). Biol. Bull. 63,492-495.
Dawson, A. B. (1933). Biol. Bull. 64,33-43.
DeLaney, R. G., Shub, C., and Fishman, A. P. (1976). Copeia (3), 423-434.
Deuticke, B. (1970). Naturw. 57, 172-179.
Dexiang, C., and Ainsworth, A. J. (1991).Deo. Comp. Zmmunol. 15,201-208.
Dijk, J. P. van, Lagerwerf, A. J., Eijk, H. G. van, and Leijsne, B. (1975).J.Comp. Physiol.
D’Ippolito, S., Grimaldi, M. C., Pica, A., and Della Corte, F. (1985). Arch. Ztal. Anat.
    Embriol. 90,31-46.
Doolittle, R, F., and Surgenor, D. M. (1962).Arner.J.Physiol. 203,964-970.
Downey, H. (1911).Folia Haematol. (Lpz).11,275-313.
48                                                                    RAGNAR FANGE

Drzewina, A. (1905).Archs. Zool. E x p . Gen. 3, 145-338.
Drzewina, A. (1911). Arch. D’Anat. Microsc. 13,319-376.
Durand, J. (1950).Ann. lnst. Ocdanogr. 25,110-206.
Duthie, E. S. (1939).J.Anat. 73,396-412.
Ehrlich, P. (1879). Arch. Anat. Physiol. (Physiol.Abt.), 571.
Ellis, A. E. (1976).J.Fish Biol. 8, 143-156.
Ellis, A. E. (1977).J.Fish Biol. 11,453-491.
Ellis, A. E. (1982).De n Comp. Immunol. Suppl. 2, 147-155.
Ellis, A. E. (1988).“Fish Vaccination.” Academic Press, London.
Ellis, A. E. (1989). In “Fish Pathology” (R. J. Roberts, ed.). Bailliere-Tindall, London.
Ellis, A. E., Munroe, A. L., and Roberts, R. J . (1976).J.Fish Biol. 8,67-78.
Ellory, J. C., Wolowyk, M. W., and Young, J. D. (1987).J.E r p . Biol. 129,377-383.
Ellsaesser, C. F., Miller, N. W., Cuchens, M. A., Lobb, C. J., and Clem, L. W. (1985).
    Trans. Am. Fish. SOC. 114,279-285.
Everaarts, J. (1978).Neth. J. Sea Res. 12, 1-57.
Ezell, G. H., Sulya, L. L., and Dodgen, C. L. (1969). The osmotic fragility of some fish
    erythrocytes in hypotonic saline. Comp. Biochem. Physiol. 28,409-415.
Ezzat, A. A., Shabana, M. B., and Farghaly, A. M. (1974).J.Fish Biol. 6,l-12.
Fange, R. (1951).Ark. f.Zool. (Ser.2) 1,259-264.
Fange, R. (1968).Acta Zool. (Stockh.)   49,155-161.
Fange, R. (1978).J.Mar. Biol. Ass. U . K . 58,727-730.
Fange, R. (1984).Vidensk. Meddr. Dansk Naturh. Foren. 145,143-162.
Fange, R. (1985). In “Evolutionary Biology of Primitive Fishes” (R. E. Foreman et al.,
    eds.), pp. 253-273. Plenum Press, New York, London.
Fange, R. (1986a).In “Fish Physiology: Recent Advances” (S. Nilsson and S. Holmgren,
    eds.), pp. 1-23. Croom Helm, London, Sydney.
Fange, R. (198613).Vet. Immunol. Immunopathol. 12,153-161.
Fange, R. (1987). Arch. Biol. (Bruxelles)98, 187-208.
Fange, R., and Edstrom, A. (1973).Acta Regiae SOC. Sci. Litt. Gothob., Zool. 8,49-50.
Fange, R., and Gidholm, L. (1968). Naturwiss. 55,44.
Fange, R., and Gidholm, L. (1973).Acta Regiae SOC. Sci. Litt. Gothob., Zool. 8,51-52.
Fange, R., and Koskinen, E. (1984).Unpublished observations from exam work. Univ. of
     Goteborg, Sweden.
Fange, R., and Nilsson, S. (1985).Experientia 41, 152-158.
Fange, R., and Sundell, G. (1968). Acta Zool. (Stockh.)50, 155-168.
Fange, R., and Zapata, A. (1985). I n “Ontogeny and Phylogeny of the Immune System”
     (R. A. Good, ed.), Vol. 4, pp. 59-76. Iwanami Shoten Immune System, Tokyo (In
Fange, R., Johansson-Sjobeck, M.-L., and Kanje, M. (1974). Acta Physiol. Scand. 91,
Fange, R., Lundblad, G., Slettengren, K., and Lind, J. (1980).Comp. Biochem. Physiol. B
Ferguson, H. W. (1976). The ultrastructure of plaice (Pleuronectes platessa) leucocytes.
    J. Fish Biol. 8, 139-142.
Ferguson, H. W. (1984). Can.3.Zool. 62,2505-2511.
Fey, F. (1965).Folio Haematol. 84, 122-146.
Fey, F. (1966a).Folia Haematol. 85,205-217.
Fey, F. (1966b). Folia Haematol. 86, 1-20,
Finn, J. P., and Nielsen, N. 0. (1971).J.Fish Biol. 3,393-422.
Finstad, J., Fange, R., and Good, R. A. (1969). In “Lymphatic Tissue and Germinal
     Centers in Immune Response,” pp. 21-31. Plenum Press, New York.
1. FISH   BLOOD CELLS                                                                 49

Fletcher, J., and Huehns, E. R. (1968).Nature 218, 1211-1214.
Flory, C. M., and Bayne, C. J . (1991).Deo. Comp. Immunol. 15, 135-142.
Fugelli, K. (1967). Comp. Biochem. Physiol. 22,253-260.
Fujii, T. (1981). Cell. Tissue Res. 219,41-51.
Fujii, T. (1982).J. Morphol. 173,87-100.
Fujii, T., and Murakawa S. (1981). Dev. Comp. Immunol. 5,251-259.
Fujii, T., Nakagawa, H., and Murakawa, S. (1979).Dev. Comp. Immunol. 3,441-451.
Fujimaki, Y., and Isoda, M. (199O).J.Fish B i d . 36, 821-831.
Galbreath, J. (1979).Oregon Wildlife March, 3-8.
Garavini, C., Martelli, P., and Borelli, B. (1981). Histochemistry 72,75-81.
Gardner, G. R., and Yevich, P. P. (1969).J . Fish Res. Bd. Can. 26,433-447.
Gilbertson, P., Wotherspoon, J., and Raison, R. L. (1986). Deo. Comp. Immunol. 10,
Gillen, R. G., and Riggs, A. (1971).Comp. Biochem. Physiol. B 38,585-595.
Glazova, T. N. (1977). Ichthyologia (Beograd) 9,65-73.
Goldstein, L., and Boyd, T. A. (1978).Comp. Biochem. Physiol. 60,319-325.
Golovina, N. A., and Trombitsky, I. D. (1989).“Haematology of Pond Fishes.” Kishinev
    “Shtiintsa” (In Russ.).
Goniakowska-Witalinska, L. (1974). Bull. de l’Acad. Polon. des Sci. Sdr. Sci. Biol. Cl. V.
Good, R. A., Finstad, J., Pollara, B., and Gabrielsen, A. E. (1966). In “Phylogeny of
    Immunity” (R. T. Smith, P. A. Miescher, and R. A. Good, eds.), pp. 149-170. Univer-
    sity of Florida Press, Gainesville.
Gorr, T., Kleinschmidt, T., and Friche, H. (1991).Nature 351,394-397.
Goven, B. A,, Dawe, D. L., and Gratzek, J. B. (1980). Deu. Comp. Immunol. 4,55-64.
Graves, S. S., Evans, D. L., and Dawe, D. L. (1985).J.Immunol. 134,34-44.
Greenlee, A. R., Brown, R. A., and Ristow, S. S. (1991). Deo. Comp. Immunol. 15,
    153- 164.
Grimm, A. S. (1985). In “Fish Immunology” (M. J. Manning, and M. F. Tatner, eds.), pp.
    263-271. Academic Press, London, New York.
Grodzinski, H., and Hoyer, H. (1938).In “Dr. H. G. Bronns K1. u. Ordn. d. Tierreichs”6.
    Bd., 1. Abt., 2. Buch, Teil2, pp. 63-77. Akademische Verlagsgesellschaft, Leipzig.
GutiCrrez, M. (1967).Ino. Pesq. (Barcelona)3 1 , 5 3 4 0 .
Haider, G. (1967).Zool. Ans. 179,384-409.
Haider, G. (1968).Zool. Ans. 180, 110-130.
Hansen, V. K., and Wingstrand, K. G. (1960).Dana-report No. 54,l-15. A. F. Host & Son,
Hardig, J . (1978).Acta Physiol. Scand. 60,256-266.
Harris, J. E. (1973).J.Fish. Biol. 5,261-276.
Harshbarger, J. C. (1981).“Activities report: Registry of tumors in lower animals, 1981
    supplement.” Smithsonian Institution, Washington.
Hawkins, R. I., and Mawdesley-Thomas, L. E. (1972).]. Fish Biol. 4, 193-232.
Haynes, L., and McKinney, E. C. (1991).Del;. Comp. Immunol. 15,123-134.
Hevesy, G., Lockner, D., and Sletten, K. (1964).Acta Physiol. Scand. 60,256-266.
Hightower, J. A., McCumber, L. J., Welsh, M. G., Whatley, D. S., Hartvigsen, R. E., and
    Sigel, M. M. (1984).J.Fish Biol. 24,587-598.
Hildemann, W. H. (1962).Am. Naturalist 96, 195-204.
Hine, P. M., and Wain, J . M. (1987a).J.Fish Biol. 30,547-556.
Hine, P. M., and Wain, J . M. (1987b).J. Fish Biol 30,557-565.
Hine, P. M., and Wain, J . M. (1987c).J.Fish B i d . 30, 567-576.
Hine, P. M., and Wain, J. M. (1988a).J.Fish B i d . 32, 579-592.
                                                                      HAGNAR FANCE

Hine, P. M., and Wain, J. M. (1988b).j.Fish Biol. 33,235-245.
Hine, P. M., and Wain, J . M. ( 1 9 8 8 ~ J . Fish Biol. 33,247-254.
Hine, P. M., and Wain, J. M. (1989).]. Fish Biol. 34,841-853.
Hine, P. M., Wain, J. M., and Boustead, N. C. (1987). New Zealand Fisheries Research
    Bulletin No. 28, 74 pp.
Hine, P. M., Lester, R. J. G., and Wain, J. M.(l99Oa).Aust.]. Zool. 38, 131-144.
Hine, P. M., Wain, J. M., and Lester, R. J. G. (1990b).Aust.]. Zool. 38, 145-154.
Hinegardner, R. (1976). Comp. Biochem. Physiol. B 55,367-370.
Hines, R. S., and Spira, D. T. (1973).]. Fish Biol. 5,527-534.
Hinuma, A., Abo, T., Kumagai, K., and Hata, M. (1980). Deo. Comp. Immunol. 4,653-
Hoffbrand, A. V., and Pettit, J. E. (1980).“Essential Haematology.” Blackwell, Oxford.
Horne, M. T., Tatner, M. F., and Ward, P. D. (1984).Vaccination of fish: A practical view.
    The Veterinary Record June 2,537-539.
Houston, A. H. (1990).In “Methods for Fish Biology” (C. B. Schreck, and P. B. Moyles,
    eds.), pp. 273-334. Am. Fish. Soc., Bethesda, Maryland.
Houston, A. H., and Keen, J. E. (1984).C a n . ] . Fish Aquat. Sci. 41,1829-1834.
Hughes, G. M., Kikuchi, Y., and Barrington, J. (1986a).]. Fish Biol. 29,393-402.
Hughes, G. M., Martinex, I., and Boeuf, G. (1986b).]. Fish Biol. 28,403-406.
Hureau, J.-C. (1966).Bull. Soc. Zool. France 91,735-751.
Hyder, S. L., Cayer, M. L., and Pettey, C. L. (1983). Tissue Cell 15,437-455.
Hyder Smith, S., Obenauf, S. D., and Smith, D. S. (1989). Tissue Cell 21,47-58.
Ingram, G . A. (1980).]. Fish B i d . 16,23-60.
Iorio, R. J. (1969). Cell Tissue Kinetics 2,319-331.
Isaacks, R. E., and Kim, H. D. (1984).Comp. Biochem. Physiol. A 79,667-671.
Ishizeki, K., Nawa, T., Tachibana, T., Sakakura, Y., and Iida, S. (1984).Cell Tissue Res.
Iuchi, I. (1973).J.E x p . Zool. 184,383-396.
Ivanova, N. T. (1983). “Atlas of Fish Blood Cells (Atlas Kletok Krovi Ryb). Comparative
    Morphology and Classification of the Formed Elements of the Blood of Fishes.”
    “Light and Food Industry.” Moscow, 151 pp., 58 pl. (In Russian).
Jacobs, M. H. (1931). Ergebnisse der Biologie 7, 1-55.
Jakowska, P. (1956). Reoue Hemat. 11,519-539.
Jarvik, E. (1980). “Basic Structure and Evolution of Vertebrates.” Academic Press,
Javaid, M. Y., and Lone, K. P. (1973).Actu Physiol. Latin. Amer. 23,396-412.
Jensen, J . A,, Festa, E., Smith, D. S., and Cayer, M. (1981).Science 214,566-569.
Johansen, K., and Lenfant, C. (1972).In “Oxygen affinity ofhemoglobin and red cell acid
    base status” (M. Rorth, and P. Astrup, eds.), pp. 750-780. Alfred Benzon Symposium
    IV. Munksgaard, Copenhagen.
Johansen, K., Mangum, C. P., and Weber, R. E. (1978).Can.]. Zoo!. 56,891-897.
Johansson, M.-L. (1973).Acta Regiae Soc. Sci. Litt. Gothob., Zool. 8,53-56.
Johansson-Sjobeck, M.-L., and Larsson, A.(1978).Enuiron. Res. 17, 191-204.
Johansson-Sjobeck, M.-L., and Larsson, A. (1979).Arch. Enoiron. Control Toxicol. 8,
     4 19-43 1.
Johansson-Sjobeck, M.-L., and Stevens, J. D. (1976).]. Mar. Biol. Ass. U . K . 56,237-240.
Jordan, H. E. (1933).Quart. Reo. Biol. €458-76.
Jordan, H. E. (1938). In “Handbook of Hematology” (H. Downey, ed.), Vol. 2, pp.
     715-744. Hamilton Medical Books, London.
Jordan, H. E., and Speidel, C. C. (1924).]. Morphol. 38,355-391.
Jordan, R. E. (1983).Arch. Biochem. Biophys. 227,587-595.
1. FISH BLOOD      CELLS                                                               51

Joseph-Silverstein, J., and Cohen, W. D. (1984).J.Cell. Biol. 98,2118-2125.
Kanesada, A. (1956). Bull. Yamaguchi Med. School 4,l-35.
Keen, J. E., Steele, A. M. C., and Houston, A. H. (1989).Comp. Biochem. Physiol. A 94,
    699-71 1 ..
KelBnyi, G. (1972). Experirntia 28, 1094-1096.
KelBnyi, G., and Nemeth, A. (1969).Acta Biol. Acad. Sci. Hung. 20,405-422.
Kim, H . D., and Isaacks, R. E. (1978).Can. J . Zool.56,863-869.
Kisch, B. (1949). E x p . Med. Surg. 7,318-320.
Kisch, B. (1951).E x p . Med. Surg. 9, 125-137.
Klawe, W. L., Barrett, I., and Klawe, B. M. (1963).Nature 198, 196.
Kobayashi, K., Tomonaga S., Teshima K., and Kajii, T. (1985). Eur. J . Immunol. 15,
Koehring. V. (1930).]. Morphol. 49,45-137.
Kreutzmann, H.-L. (1976). Folia Haematol. 105,239-247.
Kreutzmann, H.-L., and Jonas, L. (1978).Acta Histochem. 62,282-292.
Lane, H. S.,   and Tharp, T. P. (1980).J.Fish B i d . 1 7 , 7 5 8 1 .
Larsson, A., Johansson-Sjobeck, M.-L., and Fange, R. (1976).1 Fish. Biol. 9,425-440.
Legler, D. W., and Evans, E. E. (1967).Proc. SOC. E x p . Biol. Med. 124,30-34.
Legler, D. W., Weinheimer, P. F., Acton, R. T., Dupree, H. K., and Russel, T. R. (1971).
    Comp. Biochem. Physiol. 38,523-527.
Lehrer, R. I., and Ganz, T. (1990).Blood 76,2169-2181.
Lester, R. J. G., and Daniels, B. A. (1976).]. Fish. Res. Bd. Can. 33, 139-144.
Lester, R. J. G., and Desser, S. S. (1975).C a n . J . Zool. 53, 1648-1657.
Litman, G. W., Hinds, K., and Kokubu, F. (1990). In “Immunoglobulin genes” (T.
    Hoinjo, F. W. Alt, and T. H. Rabbitts, eds.), pp. 163-180. Academic Press, London,
    New York.
Locket, N. A. (1980).Proc. R . Soc. Lond. B 208,265-307.
MacArthur, J. I., and Fletcher, T. C. (1985).In “Fish Immunology” (M. J. Manning, and
    M. F. Tatner, eds.), pp. 29-46. Academic Press, New York.
Mainwaring, G . , and Rowley, A. F. (1985a).Cell Tissue Res. 241,283-290.
Mainwaring, G., and Rowley, A. F. (1985b). C o m p . Biochem. Physiol. A 80,85-91.
Manning. M. J., and Tatner, M. F. (eds.) (1985). “Fish Immunology.” Academic Press,
    London, New York.
Marchalonis, J. J. (1977). “Immunity in Evolution.” Harvard University Press, Cam-
    bridge, Mass.
Mashiter. K. E., and Morgan, M. R. (1975). Comp. Biochem. Physiol. A 52,713-717.
Mattisson, A., and Fange, R. (1977).Acta Zool., Stockh. 58,205-221.
Mattisson, A., Fange, R., and Zapata, A. (1990).Acta Zool., Stockh. 71,97-106.
Mc Cumber, L. J., Sigel, M. M., Trauger, R. J . , and Cuchens, M. A. (1982).In “Phylogeny
    and Ontogeny” (N. Cohen, and Sigel, M. M., eds.), Vol. 3, pp. 393-422. Plenum
    Press, New York.
McKim, J., Christensen, G. M., and Hunt, E. (1970).J.      Fish Res. Bd. Can. 27,1883-1889.
McKinney, E. C., Smith, S. B., Haines, H. G., and Sigel, M. M. (1977).J . Reticuloen-
    dothel. Soc. 21, 89-95.
McLeod, T. V., Sigel, M. M., and Yunis, A. A. (1978). Comp. Biochem. Physiol. A 60,
    145- 150.
Michel, C . , Gonzales, R., and Avrameas, S. (199O).J.Fish B i d . 37,617-622.
Millot, J.. Anthony, J., and Robineau, D. (1978).“Anatomie de Latimeria chalumnae.”
    111. Editions cle Centre National de Rech. Scient., Paris.
Miyazaki, T., Rogers, W. T., and Semmens, K. J. (1988).J.Fish Dis. 11,245-250.
Moller, D. (1967).Sarsia 29,413-430.
                                                                       RAGNAR FANCE

Morrow, W. J. W., and Pulsford, A. (1980).J.Fish Biol. 17,461-475.
Mulcahy, M. F. (1970).J.Fish B i d . 1,333-338.
Murad, A., and Mustafa, S. (1988).J.Fish Dis. 11,365-368.
Murray, C. K., and Fletcher, T. C. (1976).J.Fish Biol 9,329-334.
Nash, K. A., Fletcher, T. C., and Thomson, A. W. (1987). Comp. Biochem. Physiol. B 86,
Nelson, J. S. (1984). “Fishes of the World,” 2nd ed. John Wiley & Sons, New York.
Nikinmaa, M. (1990). “Vertebrate Red Blood Cells.” (Zoophysiology 28) Springer-
    Verlag, Berlin-Heidelberg.
Ogata, H., and Murai, T. (1988).]. Fish Biol. 33,471-479.
Ohno, S., and Atkin, N. B. (1966).Chromosorna 18,455-466.
Ourth, D . D. (1980). Deo. Comp. Zmmunol. 4,65-74.
Page, M., and Rowley, A. F. (1983).A cytochemical, light and electron microscopical
    study of the leucocytes of the adult river lamprey, Lampetrafluviatilis (L. Gray).
    J. Fish Biol. 22,502-517.
Page, M., and Rowley, A. F. (1984).J.Fish Dis. 7,339-353.
Parish, N., Wrathmell, A., Hart, S., and Harris, J. E. (1986).J.Fish Biol. 28,545-561.
Parker, W. N. (1892). Trans. Roy. Zrish Acad. (Dublin)30, Part 111, 109-130.
Pedersen, R. A. (1971).J.E x p . Zool. 177,65-78.
Peterson, A. J., and Poluhowich, J. J. (1976). Comp. Biochern. Physiol. A 55,351-354.
Phromsuthirak, P. (1977).J. Fish Biol. 11,193-206.
Pica, A., Grimaldi, M. C., and Della Corte, F. (1983). Monitore Zoo/ Ztal ( N . S . ) 17,
Pitombeira, M. S., Barrets Gomes, F. V., and Martins, J. M. (1971). Mar. B i d . (Berlin)9,
Plytycz, B., Flory, C. M., Galvan, I., and Bayne, C. J. (1989).Dec. Comp. Zmmunol. 13,
Potter, I. C., Percy, L. R., Barber, D. L., and Macey, D. J. (1982). In “The Biology of
    Lampreys” (M. W. Hardisty, and I. C. Potter, eds.), Vol. 4A, pp. 233-292. Academic
    Press, London.
Pradhan, A. K., Saini, S. K., Biswas, J., and Pati, A. K. (1989).Gen. Comp. Endocrinol. 76,
Rafn, S., and Wingstrand, K. G. (1981). Zool. Scripta 10,223-239.
Raison, R. L., and Hildemann, W. H. (1984).Dev. Comp. Immunol. 8,99-108.
Ratcliffe, N. A., and Millar, D. A. (1988).In “Vertebrate Blood Cells” (A. F. Rowley and
    N. A. Ratcliffe, eds.), pp. 1-17. Cambridge University Press, Cambridge.
Reznikoff, P., and Reznikoff, D. G. (1934). Biol. Bull. 66, 115-123.
Rodger, H. D., Deinan, E. M., Murphy, T. M., and Lunder, T. (1991).Bull. Eur. Ass. Fish
    Pathol. 11, 108-111.
Roubal, F. R. (1986).J . Fish Biol. 28,573-593.
Rowley, A. F., Hunt, T. C., Page, M., and Mainwaring, G. (1988).In “Vertebrate Blood
     Cells” (A. F. Rowley, and N. A. Ratcliffe, eds.), pp. 19-127. Cambridge University
     Press, Cambridge.
Rubashev, S. I. (1969). Trydy Borodin. Biol. S t . (Proc. Biol. Station, Borodino) 9(1),
     61-69; (Fisheries Res. Bd. Canada. Translation Series No. 1312).
Sanchez, I., and Cohen, W. D. (1988).B i d . Bull. 175,302.
Saunders, D. C. (1966a).Trans. Amer. Microsc. Soc. 85,427-449.
Saunders, D. C. (1966b).Elasmobranch blood cells. Copeia (2), 348-351.
Savage, A. G. (1983).J . Morphol. 178,187-206.
Schmidke, J., Schmitt, E., and Engel, W. (1978).Comp. Biochem. Physiol. B 64,117-120.
1.   FISH BLOOD CELLS                                                                 53

Schumacher, R. E., Hamilton, C. H., and Longtin, E. J. (1956). Progr. Fish Cult. 18,
     147- 148.
Scott, A. L., Rogers, W. A,, and Klesius, P. H. (1985).Deu. Comp. Immunol. 9,241-250.
Scott, E. M., and Harrington, J. P. (1990). Comp. Biochem. Physiol. B 95,91-93.
Sherburne, S . W. (1973).Fishery Bull. 71, 1011-1017.
Shrivastava, A. K., and Griffith, R. W. (1974). Copeia (l),136-141.
Sindermann, C. J., and Krantz, G. E. (1968). Chesapeake Science 9,94-98.
Smit, G . L., and Hattingh, J. (1980).J . Fish B i d . 17,337-341.
Smith, A. M., Potter, M., and Merchant, E. B. (1967).J . Immunol. 99,876-882.
Smith, A. M., Wivel, N. A., and Potter, M. (1970).Anat. Rec. 167,351-356.
Smith, C. E., McLain, L. R., and Zaugg, W. S. (1971). Toxicol. Appl. Pharmacol. 20,
Soivio, A., and Oikari, A. (1976).J.Fish B i d . 8,397-411.
Stahl, B. (1967). Bull. Mus. Comp. Zool. Hart;ard 135, 141-213.
Starmach, J . (1970).Acta Biol. Cracooiensia Series: Zoologia 13,243-249.
Stave, J . W., and Roberson, B. S. (1985). Det;. Comp. Immunol. 9 , 7 7 4 4 .
Stave, J. W., Roberson, B. S., and Hetrick, F. M. (1984).J.Fish Biol. 25, 197-206.
Stiller, R. A., Belamarich, F. A,, and Shepro, D. (1975).A m . J . Physiol. 229,206-210.
Stobbe, H. (1963). Folia Haematol. 80,452-462.
Stokes, E. E., and Firkin, B. G. (1971).Brit.J.Haematol. 20,427-435.
Suzuki, K. (1986).J . Fish Biol. 29,349-364.
Suzuki, Y. (1986). Bull.Jap. Soc. Sci. Fish. 52,1895-1899.
Suzuki, Y. (1988). Bull. Jap. Soc. Sci. Fish. 54, 1257.
Tanaka, Y., and Saito, Y. (1981).J.Electron Microsc. 30,253-273.
Tavassoli, M., and Yoffey, J. M. (1988).“Bone Marrow. Structure and Function.” Alan R.
     Liss, Inc., New York.
Thoenes, G. H., and Hildemann, W. H. (1970). In “Developmental Aspects ofAntibody
     Formation and Structure” (Sterz et al., eds.), Vol. 2, pp. 711-726. Czechoslovak
     Akad. Sci./Academic Press, Prague, New York.
Thomas, N. W. (1971).J.Cell Sci. 8,407-412.
Tomonaga, S., Hirikane, T., Shinohara, H., and Awaya, K. (1973).2002.       Mag. (Tokyo)82,
Tomonaga, S., Kobayashi, K., Hagiwara, K., Sasaki, K., and Sezaki, K. (1985).Det;. Comp.
     Immunol. 9,617-626.
Tomonaga, S., Kobayashi, K., Hagiwara, K., Yamaguchi, K., and Awaya, K. (1986). 2001.
     Science (lapan) 3,453-458.
Tufts, B. L., and Randall, D. J. (1988). Can.1.Zoo/. 67,235-238.
Undritz, E. (1963). The Physician’s Panorama (Sandoz) Febr., 4-5.
Val, A. L., D e Almeida-Val, V. M., and Affonso, E. G. (1990). Comp. Biochem. Physiol. B
Vendrely, R. (1955).In “The Nucleic Acids” (E. Chargaff, and J. N. Davidson, eds.), Vol.
     2, pp. 155-180. Academic Press, New York.
Vislie, T. (1978). Comp. Biochem. Physiol. B 60, 35-40.
Vogel, V. 0. P., and Claviez, M. (1981).Z. Naturforsch. 36,490-492.
Vos, J . , Van Loveren, H., Wester, P., and Vethaak, D. (1989).TZPS 10,289-292.
Walker, R. L., and Fromm, P. 0. (1976).Comp. Biochem. Physiol. A 55,311-318.
Ward, J . W. (1969). Copeia (3), 633-635.
Wardle, C. S. (1971).J.Mar. Biol. Ass. U . K . 51,977-990.
Warr, G. W. (1983). Comp. Biochem. Physiol. B 76,507-514.
Warr, G. W., and Marchalonis, J. J. (1977). Deo. Comp. Immunol. 1, 15-22.
54                                                                     RAGNAR FANCE

Warr, G. W., and Simon, R. C. (1983).Det;. Comp. Immunol. 7,379-384.
Watson, L. J., Shechmeister, I. L., and Jackson, L. L. (1963).Cytologia 28, 118-130.
Weber, R. E., Bol, J . F., Johansen, K., and Wood, S. C. (1973).Arch. Biochem. Biophys.
Weinberg, S. R., Siegel, C. D., Nigrelli, R. F., and Gordon, A . S. (1972).Zoologica 71-78.
Weinreb, E . L. (1958).Zoologica 43, 145-154.
Weinreb, E. L. (1963).Anat. Rec. 147,219-238.
Weinreb, E. L., and Bilstad, N. M. (1955). Copeia (3), 194-204.
Weisel, G. F. (1973).]. Morphol. 140,243-255.
Weiss, L. (1981). In “Microenvironments in Haemopoietic and Lymphoid Differentia-
    tion” (Ciba Found. Symp. 84), pp. 5-21.
Wekerle, H., and Ketelsen, U. P. (1980).Nature 283,402-404.
Wharton Jones, T. (1846).Phil. Trans. Roy. Soc. London, Part 11,63-87.
Wilkins, N. P., and Clarke, F. D. (1974). Comp. Biochem. Physiol. A 47,341-345.
Wintrobe, M. M. (1933). Folia Haematol. 51,32-49.
Wolters, W. R., Chrisman, C. L., and Libey, G. S. (1982).]. Fish Biol. 20,253-258.
Yamaguchi, K., Tomonaga, S., Ihara, K., and Awaya, K. (1979).], Electron Microsc. 28,
Yasutake, W. T., and Wales, J. H. (1983).“Microscopic Anatomy ofSalmonids: An Atlas.”
    Fish and Wildlife Service. U.S. Dept. of the Interior, Washington, D.C.
Yoffey. J. M. (1929).]. Anat. 63,314-344.
Yoffey, J. M. (1985).Lymphology 18, 5-21.
Yoffey, J. M., and Courtice, F. C. (1970).“Lymphatics, Lymph and the Lymphomyeloid
    Complex.” Academic Press, New York.
Youson, J. H. (1971).C a n . ] . Zool. 49,962-963.
Yu, M. L., Kiley, C. W., Sarot, D. A,, and Perlmutter, A. (1971).J.Fish. Res. Bd. Can. 28,
Zanjani, E. D., Yu, M.-L., Perlmutter, A., and Gordon, A. S. (1969).Blood 33,573-581.
Zapata, A. (1981).Det;. Comp. Endocrinol. 5,685-690.
Zapata, A., and Carrato, A. (1980).Actu 2001.     (Stockh.)61, 179-182.
Zapata, A., Fange, R., Mattisson, A,, and Villena, A. (1984).Cell Tissue Res. 235, 691-

Department of Biology
McMaster University
Hamilton, Ontario, Canada

Department of Zoology
University of Western Ontario
London, Ontario, Canada

    I. Introduction
       A. Effects of Sampling Method on Blood Chemistry
       B. Die1 Cycles in Blood Chemistry
   11. Hormones: Teleosts
       A. Gonadotropins and Sex Steroids
        B. Growth Hormone
       C. Prolactin
       D. Arginine Vasotocin
        E. Melatonin
        F. Thyroid Hormones
       G. Calcitonin
        H. Cortisol
         I. Catecholamines
         J. Pancreatic Hormones
        K. Stanniocalcin
        L. Urotensins
  111. Hormones: Cyclostomes and Chrondricthyes
  IV. Metabolites
        A. Glucose
        B. Lactate
        C. Ethanol
        D. Ketone Bodies
        E. Bile Pigments
    V. Nonprotein Nitrogenous Compounds
        A. Amino Acids
        B. Ammonia
        C. Urea and Uric Acid
56                                   D. G . MCDONALD AND C. L. MILLIGAN

      D. Trimethylamine Oxide
      E. Creatine and Creatinine
  VI. Plasma Proteins
      A. Total Plasma Protein
      B. Albumin
      C. Immunoglobulins
      D. Hormone-Binding Proteins
      E. Antifreeze Proteins
      F. Plasma Enzymes
 VII. Lipids
      A. Total Lipids
      B. Lipoproteins
      C. Cholesterol
      D. Nonesterified Fatty Acids
      E. Triglycerides
VIII. Electrolytes
      A. Na+, C1-, and Osmolarity
      B. Calcium
      C. Magnesium
      D. Potassium
      E. Phosphate and Sulfate


    In this Chapter our objective is to compile normal values for such
blood variables as hormones, metabolites, proteins, lipids, and electro-
lytes as well as variations resulting from such factors as temperature,
hypoxia, exercise, salinity, feeding, development, and reproductive
state. To make this review comprehensive is essentially an impossible
task as there are over 25,000 species of fish in 6 major groups (cyclo-
stomes, chrondricthyes, dipnoans, chondrosts, holosts, and teleosts)
and there is a considerable bias toward species that are readily avail-
able and of economic importance. Most data are for salmonids,
and there is a considerable bias toward species that are readily avail-
able and of economic importance. Most data are for salmonids,
al., 1988; Bergheim et al., 1990).
    Thus, we have not attempted to compile all available data but rather
report the “best” or “most representative” available value for each
variable, some estimate of its maximum range in the healthy organism,
and the condition(s) under which the maxima are reached. To this end,
we have used our judgment as to what are representative values.
Furthermore, we have tended to emphasize those areas of blood chem-
istry that have previously been neglected. Consequently, our empha-
sis is on plasma levels of hormones and certain key metabolites, while
2.   CHEMICAL PROPERTIES OF THE BLOOD                                    57

our treatment of such topics as electrolytes and acid-base chemistry is
highly selective because this subject has already been comprehen-
sively reviewed (see, in particular, Chapters by Holmes and Donald-
son in Volume 1and Heisler in Volume XA).
    Irrespective of the endogenous or exogenous factor of interest,
attention must first be paid to two factors that have profound influence
on blood chemistry: the method employed to obtain the blood sample
for analysis and the time of day the samples are collected.

A. Effects of Sampling Method on
   Blood Chemistry
    It is now well established that blood chemistry is extremely sensi-
tive to sampling procedure (e.g., Railo et al., 1985; Laidley and Leath-
erland, 1988a; Houston, 1990). Two approaches to blood sampling are
in routine use: acute sampling of stunned or anesthetized fish via
either cardiac or caudal puncture, or caudal severence; and chronic
sampling via an indwelling catheter, usually implanted in the dorsal
aorta. Acute sampling procedures have often been successful in ob-
taining resting or routine levels of blood variables but potentially alter
blood chemistry more than chronic procedures. The trauma associated
with capture, handling, and sampling can activate physiological stress
responses that will have an effect on blood chemistry. This can become
compounded if the sampling duration on each fish is greater than about
30 sec and if the sampling protocol involves serial removal of individ-
ual fish from the same tank (Laidley and Leatherland, 1988a). The
initial physiological effect of primary importance is the mobilization of
catecholamines, particularly adrenaline. Adrenaline will cause mobili-
zation of red cells from the spleen and red cell swelling (Nikinmaa and
Heustis, 1984) thus elevating hematocrit. It will also increase blood
pressure and gill blood flow, which will, in effect, increase electrolyte
permeability of the gills and can quickly cause a depression in plasma
Na+ and C1- in freshwater teleosts (Gonzalez and McDonald, 1992) or
a net gain in seawater teleosts (Boutilier et al., 1984; Wells et al., 1986).
If the agitation of the remaining fish is prolonged beyond 5 min or so,
then significant elevation in white muscle lactate levels will occur.
This will cause acid-base disturbances, elevation in plasma lactate,
and a shift of fluid to the intracellular compartment, concentrating most
plasma constituents. While the amount of the disturbance can vary
from species to species and with sampling methodology, it is generally
agreed that repetitive sampling by acute methodology is inappropriate
because of the long-term effects of the stress on a single sample. For
 example, Pickering et al. (1982) found in brown trout, Salmo trutta,
58                                D. G. MCDONALD A N D C. L. MILLIGAN

that a minimum of 2 wk was required for complete recovery of all blood
parameters from a 2-min bout of handling stress.
    Chronic indwelling catheters are the method of choice for obtain-
ing resting levels of blood parameters (in particular for catechol-
amines), but they are also not entirely free from stress. A certain
amount of blood loss is inevitable during surgery, and these losses will
increase with repetitive blood sampling and with procedures required
to keep the catheters clear and working. Hematocrits are typically
below average as a result. Although some splenic compensation for
blood loss can be expected (Pearson and Stevens, 1991),the reduction
in blood volume will produce changes in physiological state such as
renin release and activation of angiotensin I1 (Bailey and Randall,
 1981). The procedure of anesthesia and surgery is stressful in itself
and, at least in salmonids, up to 4 days is required for plasma electro-
lytes, metabolites, and acid-base status to return to normal (Heisler,
 1984). Also, chronic catheterization is impractical for animals much
smaller than 100 g. The effects of anesthesia on blood chemistry have
been studied extensively; see Laidley and Leatherland (1988a),
 Iwama et al. (1989), and Summerfelt and Smith (1990) for discus-

B. Die1 Cycles in Blood Chemistry
    The levels of many plasma constituents exhibit daily (diel) varia-
tions that can modify and complicate any analysis of the influence of
environmental factors. These diel fluctuations are often considered
endogenous. This designation, however, requires that they be shown
to free-run under constant conditions, and this has not been demon-
strated in most instances. Hence their designation as diel cycles is
more appropriate (Laidley and Leatherland, 198813).
    Daily fluctuations in plasma cortisol concentration are one of the
most studied and reproducible of the cyclic phenomena (Peter et al.,
1978; Spieler, 1979; Bry, 1982; Pickering and Pottinger, 1983; Thorpe
et al., 1987; Laidley and Leatherland, 1988b; Planas et al., 1990). This
has led to the suggestion that plasma cortisol might be one of the
endogenous variables to which other endocrine and metabolic
rhythms are tied (Meier, 1984). Entraining stimuli include the day/
night cycle and time of feeding. The cortisol peak typically occurs just
before the onset of light and precedes an increase in locomotory activ-
ity (Spieler, 1979) although there is a wide variation in the frequency,
amplitude, and phasing of such rhythms (Pickering and Pottinger,
2.   CHEMICAL PROPERTIES OF THE BLOOD                                  59

    In fishes, other hormones reported to show diel cycling include
prolactin (Spieler, 1979), growth hormone (Bates et al., 1989), thyroid
hormones (White and Henderson, 1977; Spieler and Noeske, 1979;
Cook and Eales, 1987), gonadotropin (Hontela, 1984; Zohar and
Billard, 1984), gonadal steroids (Lamba et al., 1983; Zohar and Billard,
1984), insulin (Gutikrrez et al., 1984), and melatonin (Gern et al.,
    It is significant, however, that in many studies, diel endocrine
rhythms have been looked for but not found. Two reasons can be
suggested for their apparent absence. In some instances the rhythms
may be present but masked by the method of blood sampling. In those
studies where blood has been sampled acutely and each animal
sampled only once, rhythmicity would not be detected unless the
population was synchronized. In other studies where animals are re-
petitively sampled by means of indwelling catheters, the rhythms may
be masked by the effects of the surgery and serial sampling on the
physiology and behavior of the fish. In other instances, however, the
absence of the rhythms may be real. For example, the intensity of the
cortisol rhythm fluctuates with the season and may disappear in winter
months under short photoperiod conditions (Rance et aZ., 1982) and
may also fluctuate with the age and sex of fish (Peter et al., 1978).Also,
Marchant and Peter (1986) were unable to find a reproducible daily
rhythm in circulating growth hormone in goldfish (Carassius auratus)
at any time of the year even though Bates et al. (1989), using a similar
methodology, found a pronounced nocturnal peak in growth hormone
in coho salmon (Oncorhynchus kisutch).
    The frequent appearance of diel fluctuations in many of the circu-
lating hormones are often, not surprisingly, accompanied by fluctua-
tions in other plasma constituents. For example, daily fluctuations in
plasma glucose and plasma lipids in goldfish appear related to cortisol
and thyroid hormone fluctuations (Delahunty et al., 1978). Similarly, in
sea bass (Dicentrarchus Zabrax) there were significant and inverse
rhythms in glucose and insulin with glucose peaking during the day
and insulin peaks during the dark period (Gutikrrez et al., 1984).
Although these fluctuations were tied to feeding times, their continua-
tion during a fast of 7 days indicates an endogenous rhythm. Diel
fluctuations in plasma protein in rainbow trout (Oncorhynchus mykiss)
were significantly correlated with fluctuations in thyroid hormone lev-
els (Laidley and Leatherland, 1988b).
    Diel fluctuations in plasma electrolytes (Na+,Ca2+,Mg2+,K+ ) have
also been reported (Toews and Hickman, 1969; Houston and Koss,
1982; Carillo et al., 1986; Laidley and Leatherland, 1988b; Peterson
60                                 D. G . MCDONALD AND C. L. MlLLIGAN

and Gilmore, 1988) although, to this point, they have not been consis-
tently linked to hormonal fluctuations (Kiihn et al., 1986; Laidley and
Leatherland, 1988b). Daily temperature fluctuations appear to be an
important entraining stimulus for electrolyte fluctuations (Toews and
Hickman, 1969; Houston and Koss, 1982), but the rhythms persist in a
constant temperature environment indicating their circadian nature
(Houston and Koss, 1982). A recent report suggests that electrolyte
rhythms are more pronounced in euryhaline species inhabiting estu-
arine environments than in stenohaline freshwater species (Peterson
and Gilmore, 1988).
    Interpreting daily endogenous variations in plasma constituents is
further complicated by the presence of additional cycles with longer
period lengths. These include cycles related to the tides, phases of the
moon (i.e., lunar cycles), and seasonal variations in photoperiod and/or
temperature, particularly in temperate zone animals. The latter varia-
tions have prominent effects on growth rate and on reproductive status,
and thus it is not surprising to find significant seasonal fluctuations in
virtually all plasma constituents. For example, lunar cycles have been
reported for thyroid hormones and some related metabolites such as
plasma glucose and triglycerides in salmonids (Grau et al., 1981; Hop-
kins and Sadler, 1987; Farbridge and Leatherland, 1987) and for go-
nadal steroids in the semilunar spawning mummichog, Fundulus het-
eroclitus (Taylor, 1984). Other examples include seasonal changes in
growth hormone in goldfish related to changes in day length (Marchant
and Peter, 1986), in cortisol in trout and sea bass related to water
temperature (Pickering and Pottinger, 1983; Thorpe et al., 1987;
Planas et al., 199O), in gonadotropins and gonadal steroids related to
either water temperature or photoperiod (Peter, 1981; Crim, 1982;
Zohar and Billard, 1984), and in electrolytes associated with vitello-
genesis (e.g., Carillo et al., 1986).
    In the following text we have avoided reporting data where the
effects of sampling may, in our judgment, have been prominent. As for
the circadian fluctuations, we report the amplitude wherever possible.
 However, it is common practice to sample fish at the same time of the
day so as to minimize contributions from die1 fluctuations.


    There has been a considerable increase in information concerning
the plasma levels of hormones in fishes over the last decade associated
in large part with improvements in measurement technologies. Com-
2.   CHEMICAL PROPERTIES OF THE BLOOD                                 61

mercial kits developed for mammalian plasma have been exploited,
particularly in those instances where the hormones are identical be-
tween fish and mammals (e.g., steroid hormones, thyroid hormones,
catecholamines, melatonin). There have also been a number of sensi-
tive homologous radioimmunoassays (RIAs) developed for various fish
hormones, although such assays apply to relatively few species, almost
exclusively teleosts, and mostly salmonids and cyprinids. This discus-
sion is restricted mainly to the hormone measurements that have been
validated for a particular fish species. Table I summarizes the teleost
hormones reviewed here. Hormones of cyclostomes and elasmo-
branchs are summarized briefly later. Also, there are a number of real
or putative hormones that we have chosen to leave out because only
limited information is available. Fish endocrinology is a rapidly chang-
ing field and rather than report information that may soon be out of date
we have restricted ourselves to those hormones for which there is
already extensive literature. The group not covered includes the
hypothalamic releasing hormones, some of the trophic hormones
(adrenocorticotropic hormone [ACTH], thyroid stimulating hormone),
melanocyte stimulating hormone, other neuropeptides, and the gut
peptides (e.g., cholecystokinin and gastrin). We have also not covered
the renin-angiotensin system (RAS) or atrial natriuretic factor (ANF or
atriopeptin) as these are comprehensively reviewed by Olson (see
Chapter 3).
    There is no universal standard for reporting plasma hormone con-
centrations at the present time so we have adopted the common prac-
tice of reporting nanogram or picogram quantities for most hormones
except for the catecholamines, which are reported in nanomolar quan-
tities. Table I lists molecular weights (where known) for teleost hor-

A. Gonadotropins and Sex Steroids
   Studies on gonadotropins (GtH) have been largely confined to sal-
monid and cyprinid species (see Billard et al., 1978; Peter, 1981; Idler
and Ng, 1983; Hontela, 1984; Zohar and Billard, 1984; Donaldson,
1990 for reviews); species for which there are homologous RIAs (see
Table I for references). The sex steroids have been examined in a much
wider range of species (see Fostier et al., 1983 in Volume IXA for an
extensive review), but comments here are largely confined to salmo-
nids and cyprinids.
   There has been a considerable controversy over whether there are
one or two GtHs in teleosts. Although initially it was believed that
                                                                     Table I
                                                  Partial Survey of Hormones of Teleost Fishes

     Glandular tissue          Hormone                Nature           MW                Measurement                        References

   Pars distalis        Gonadotropin(s) (GtH)    Glycoprotein         36-56 K   RIAs for salmonids',2. 3 ;        1. Crim et ul., 1973; 2. Breton
                                                                                cyrinids4, 5 , ' eel7, catfish*   and Billard, 1977; 3. Suziiki et
                                                                                                                  al., 1988; 4. Breton et al.,
                                                                                                                  1971; 5. Crim et al., 1976;
                                                                                                                  6. Hontela and Peter, 1978;
                                                                                                                  7 . Dufour et al., 1983; 8. Goos
                                                                                                                  et al., 1986
                        Growth hormone (GH)      Protein                27 K    RIAs for carp', Pacific           1. Cook et al., 1983; 2 . Wagner
                                                                                salmon', 3, cod4                  and McKeown, 1986;
                                                                                                                  3. Bolton et al., 1986; 4. Rand-
                                                                                                                  Weaver et al., 1989
                                                                                                                  1. Nicoll et d.,1981;
                        Prolactin (Prl)          Protein                25 K    RIAs for tilapia', Pacific        2. Hirano et al., 1985;
                                                                                salmon2, eel3                     3. Suzuki and Hirano, 1991
                                                                                                                  Sumpter and Donaldson, 1986
                        Adrenocorticotropic      Peptide                45 K    Mammalian RIA validated for
                        hormone (ACTH)                                          salmonids
                        Thyrotropic hormone      Glycoprotein         -30 K     In oioo bioassay                  Swanson et al., 1989
      Pars intermedia   Melanocyte stimulating   Peptide               13 aas   Mammalian RIA validated for       Rodrignes and Sumpter, 1984
                        hormone (aMSH)                                          salmonids
      Pars nervosa      Arginine vasotocin       Octapeptide          1050      RIA                               Holder et al., 1982; Hontela
                        (AVT)                                                                                     and Lederis, 1985
 Pineal                 Melatonin (MLT)          Indoleamine           232      RIA                               Gern et al., 1978; Kezuka et
                                                                                                                  al., 1988
 Thyroid                Triiodothyrouine (T,)    Iodiuated tyrosine    651      KIA                               Commercial kits available
                        Thyroxine (T4)           derivatives           777
Ultimobrancial      Calcitonin (CT)         Peptide         3432     RIAs for salmon’ and eel”     1. Deftos et al., 1974;
bodies                                                                                             2. Orimo et al., 1977
Interrenal tissue   Cortisol                C21 steroid     362.5    RIA                           Commercial kits available
Chromaffin tissue   Adrenalin (-4)          Catecholamine   183.2    HPLC’, fluorimetric2 or       1 Woodward, 1982; 2. Nakano
                                                                     radio-enzymatic methods”      and Tomlinson, 1967,
                                                                                                   3. Peuler and Johnson, 1977
                    Noradrenaline (NA)      Catecholamine    169.2
 Pancreas           Insulin (INS)           Peptide         5784     RIAs for chum salmon’,        1. Plisetskaya et al., 1986a;
                                                                     bonito (tuna)’, cod”          2. Gutierrez et al., 1984;
                                                                                                   3. Thorpe and Ince, 1976
Q,                  Glucagori (GLU)         Peptide         3508     Very similar to inanimalian   1. Gutierrez et al., 1986;
                                                                     glucagon’, RIA for salmon2    2. Plisetskaya et al., 1989
                    Glucagon-like peptide   Peptide         31 aas   RIA for salmon                Plisetskaya et al., 1989
                    Somatostatin (SST-25)   Peptide         25 aas   RIA for salmon                Plisetskaya et al., 198613
 Corpuscles of      Stanniocalcin (STC,     Glycoprotein     52 K    ELISA for salmon              Mayer-Gostan et al., 1992
 Stannius           formerly hypocalcini
  Ovary             Estrogen (17p-          C19 steroid      272     RIA                           Commercial kits available
  Testes            estradiol)              C19 steroid      302     RIA                           Commercial kits available
                    Androgens (11-keto
 Caudal urophysis   Urotensin I (UI)        41 aas          4864     RIA for white sucker          Suess et al., 1986
                    Urotensin I1 (UII)      12 aas          1351     RIA for white sucker          Kobayashi et al., 1986
64                                 D. G . MCDONALD AND C. L. MILLIGAN

there was only one GtH (see review by Peter and Crim, 1979), Idler
and N g (1979) demonstrated the presence of two GtHs, one rich in
carbohydrate (Con A-I1 or maturational GtH) and one low in carbohy-
drate content (vitellogenic GtH). Further research has provided evi-
dence that the two active gonadotropins in teleosts are, in fact, both
glycoproteins (GtH I MW 50,000, GtH 11, MW 36,000 in chum salmon
[Oncorhynchus keta]; Suzuki et al., 1988; 56,000 and 53,000 in the
grass carp [Ctenopharyngodon idell]; Yu and Shen, 1989). GtH I is
mainly secreted during early gonadal development, whereas GtH I1 is
secreted at the time of spermiation and ovulation (Kawauchi et al.,
1989). GtH I1 is comparable to the previously isolated maturational
GtH and is biochemically similar to mammalian leutinizing hormone
(Kawauchi et al., 1989) whereas GtH I (similar to mammalian follicle
stimulating hormone) is newly identified and, therefore, its plasma
concentrations are not widely reported (Yu and Shen, 1989).
    T h e principal circulating sex steroids in male teleosts are the an-
drogens, 11-keto-testosterone (KT) and testosterone (T) (in order of
importance, Zohar and Billard, 1984; Barry et al., 1990), while in female
teleosts it is 17p estradiol (Ez).Final maturation in males and females
in many teleosts is brought about by the progestogen, 17a, 200-
dihydroxyprogesterone (P) (Donaldson, 1990). Significant circulating
levels of T are also found in females because it is the immediate
precursor to Ez.
    In sexually immature animals the plasma levels of gonadotropins
and sex steroids are either very low or undetectable; less than 1 ng
ml-' for GtHs, 0.2-0.3 ng ml-' for the sex steroids. Sexual maturation
is associated not only with an increase in the average plasma concen-
trations of gonadotropins and sex steroids but also with an increase in
the frequency and amplitude of daily fluctuations in plasma levels
(Hontela, 1984; Zohar and Billard, 1984).
    Timing of gonadrotropin release and steroidogenesis in temperate
zone teleosts is primarily controlled by variations in temperature and
photoperiod (Peter, 1981;Crim, 1982). Salmonids, spawning mainly in
autumn and winter, are cued primarily by decreasing day length
whereas cyprinids, spawning in spring and summer are more depen-
dent on increasing temperature (Billard et d.,1978).
    In all fish, plasma GtH levels (GtH 11) increase first gradually
 during the major part of gonad development in both sexes (i.e., vitello-
 genesis and spermatogenesis) and then sharply around the time of
 oocyte maturation and ovulation and before the start of spermiation. In
 salmonids the concentration of maturation GtH is elevated for several
 days before, during, and after ovulation whereas in cyprinids plasma
 GtH changes tend to be restricted to a brief surge associated with final
2.   CHEMICAL PROPERTIES OF TIIE BLOOD                                65

maturation and ovulation (Stacey et al., 1979; Donaldson, 1990).
    In salmonids there are considerable interspecific and strain differ-
ences in GtH levels. Domesticated species show less of an elevation
than natural spawning wild strains (Billard et al., 1978), and species
that spawn only once have substantially higher GtH levels during this
period than repeat spawners (Suzuki et al., 1988).The maximum daily
average level may be as little as 10 ng ml-' in domesticated species
and up to 10 times higher in some other salmonid species allowed to
spawn naturally (Zohar and Billard, 1984; Dye et al., 1986).
    In cyprinids the GtH peaks range from 30 to 160 ng ml-' with no
major differences apparent between males and females (Billard et al.,
1978; Stacey et al., 1979; Hontela and Peter, 1983; Barry et al., 1990).
Considering the cyclic nature of plasma GtH and the effects of such
exogenous factors as water temperature on GtH levels (Peter, 1981),
this amount of variation in GtH levels among studies is not surprising.
    In female teleosts plasma Ez and T gradually increase during vitel-
logenesis, followed, in many cases, by a decrease in Ez before and
during oocyte maturation. There is also a temporary increase in T
before P increases, which induces oocyte maturation. In males, KT and
T rise during spermatogenesis with KT reaching a maximum just prior
to the start of spermiation. The concentrations of both androgens de-
crease during spermiation, whereas P shows an important rise related
to its putative role in regulation of spermiation and control of male
spawning behavior (see Barry et al., 1990 for a discussion). However, P
is not the final maturation inducing steroid in all teleosts, and absolute
plasma steroid concentrations vary dramatically among families. For
example, salmonids tend to have high steroid concentrations and
cyprinids low concentrations (Donaldson, 1990). In male salmonids,
maximum values for the androgens approach 45 ng ml-' (Billard et al.,
1978; Zohar and Billard, 1984) while in females the peak estrogen
levels are around 30 ng ml-' (Dickhoff et al., 1989) and progestogen
levels are very high; 300-600 ng ml-' at the peak in the preovulatory
period (e.g., Scott et al., 1982; Dye et al., 1986). In male cyprinids
androgen levels are lower than in male salmonids. For example, peak
T and KT levels were 12 and 25 ng ml-', respectively, in the male
cyprinids (Barry et al., 1990) whereas in female cyprinids E and P
levels rarely exceed 10 ng ml-' (Peter et al., 1984; Venkatesh et al.,

B. Growth Hormone
  Growth hormone (GH) is a protein hormone molecular weight
(MW) of Atlantic salmon [Salmo salar] GH is 23,000; Skibelli et al.,
66                                 D. G. MCDONALD A N D C. L. MILLIGAN

1990) released by GH-specific cells in the rostra1 pars distalis (RPD) of
the pituitary. Homologous RIAs have now been developed for carp,
salmon, and cod (see Table I for references). Detection limits vary from
<1 to 3 ng m1-l.
    Growth hormone levels in resting fish in freshwater can be as low as
5 ng ml-'. However, levels can increase from this baseline with star-
vation (6-fold elevation after 30 days starvation in rainbow trout, Bar-
rett and McKeown, 1988), exercise (&fold elevation after 24 h at 1.5
body lengths sec-'; Barrett and McKeown, 1988), seawater transfer
(2- to 3-fold elevation after 2 days in seawater (SW); rainbow trout,
Sakamoto et al., 1990; coho salmon, Sweeting and McKeown, 1987),
smoltification @-fold elevation during parr-smolt transformation in At-
lantic salmon reared in freshwater; Prunet et al., 1989); rearing at a
higher temperature (&fold higher in rainbow trout at 16 versus 5°C;
Barrett and McKeown, 1989), or at different times of the day (8-fold
difference in juvenile coho salmon between daytime minimum and
midnight maximum; Bates et al., 1989). There is also a consistent
relationship between GH levels and the age (size) of the fish (B. A.
McKeown, personal communication) with small fish invariably show-
ing much higher GH circulating levels. In addition to daily variations,
there are variations associated with changing season and daylength. In
goldfish, the seasonal peak in plasma GH levels occurs in spring to late
summer where levels of GH are about 2-fold higher than seasonal
mimima in November (Marchant and Peter, 1986).

C . Prolactin
    Prolactin (PRL) is a protein hormone (MW of chum salmon PRL
-23,000; Kawauchi et al., 1983) released by prolactin specific cells in
the RPD of the pituitary. It is very similar in structure to GH from
which it is thought to have arisen by gene duplication (Hirano et al.,
1987). Homologous RIAs are available for a few species, all euryhaline
teleosts (tilapia, salmon, and eel; see Table I for references).
    All measurements point clearly to high PRL levels in freshwater
fish and low levels in seawater fish. In all seawater fish, PRL levels are
at or below detection limits of the RIAs (0.1-1.0 ng ml-l) and rise on
transfer to fresh water; to 5-15 ng ml-' in salmon and eels (Hirano,
1986;Ogasawara et al., 1989;Fargher and McKeown, 1990; Suzuki and
Hirano, 1991)and to 60 ng ml-' in tilapia (Sarotherodonmossambicus;
Nicoll et al., 1981).The higher PRL levels in tilapia may be related to
the more critical dependence of this species on PRL for freshwater
survival. Salmon and eels survive in fresh water without prolactin (i.e.,
after hypophysectomy) while tilapia do not (Hirano, 1986; Bern, 1990).
2.   CHEMICAL PROPERTIES O F THE BLOOD                                 67

    The major stimulus to prolactin release appears to be a reduction of
plasma osmolarity, and/or specific ions such as Ca2+, or both (see
Nishioka et al., 1988 for detailed review). In SW-adapted fish, depres-
sion of plasma Ca2+ stimulates an increase in PRL levels in plasma; for
example, injection of EGTA into SW adapted coho salmon lowers
plasma Ca2+ and causes PRL levels to double (Fargher and McKeown,
1989). The latter reflects the now well-established hypercalcemic ac-
tion of PRL in teleosts (Clark, 1983; Flik et al., 1989). Nonetheless,
PRL levels can change independently of changes in plasma osmolarity
or electrolytes. Prolactin levels rise in response to stress in either
fresh water or seawater (Avella et al., 1991) and decline with smolti-
fication in fresh water (Prunet et al., 1989; Younget al., 1989).The high
variation in plasma PRL among individual fish (all collected at similar
times of the day) suggests pulsatile release of the hormone (Nicoll e t
al., 1981).There also is some evidence of a marked difference between
the sexes in PRL release. Plasma PRL levels rose about 4-fold higher in
mature female chum salmon compared to males when both were trans-
ferred from seawater to fresh water (Hirano et al., 1985).

D. Arginine Vasotocin
    Arginine vasotocin (AVT) from the pars nervosa circulates in the
blood of teleosts. It has cardiovascular effects although its status as a
hormone and its role in osmoregulation are uncertain (Sawyer, 1987).
It has been measured in teleosts by RIA, but appears to be present in
body fluids at or near the limit of sensitivity of the assay, 5-10 pg ml-'
(Holder et id.,1982; Hontela and Lederis, 1985). Perrot et al. (1986),
however, reported somewhat higher levels of AVT in rainbow trout
and flounder in fresh water, 88 and 192 pg ml-', respectively, and
significantly lower levels in seawater, 35 and 82 pg ml-', thus sug-
gesting a role for AVT in freshwater adaptation. While unable to mea-
sure plasma AVT levels in freshwater brook trout (Saluelinus fonti-
nalis),Hontela et al. (1991) showed an increase in pituitary AVT levels
in acid stressed animals. Acid stress produces ion losses in fresh water
similar to those experienced by euryhaline teleosts acutely transferred
from seawater to fresh water, suggesting a role for AVT in hyperos-
motic regulation.

E, Melatonin
   Melatonin (MLT) is an indole-amine secreted by the pineal gland.
A RIA based on a rabbit antiserum to a melatonin-BSA (bovine serum
68                                 D. G. MCDONALD AND C. L. MILLIGAN

albumin) conjugate has been validated for use on teleost plasma (Gern
et al., 1978; Kezuka et al., 1988)and has detection limits of about 50 pg
ml-' (Kezuka et al., 1988).The most prominent feature of plasma MLT
in teleosts is marked day-night fluctuations. In common carp (Cypri-
nus carpio),MLT levels were high during darkness (220-540 pg ml-')
and low (23-104 pg ml-l) in the light phase (Kezuka et al., 1988). I n
rainbow trout the daily fluctuations were similar but the amplitude
was less; 81 t 19 pg ml-' in the light and 153 +- 16 pg ml-' in the dark
(Gern et al., 1978). Studies indicate that MLT secretion is light sensi-
tive as well as having a circadian or endogenous rhythm (Iigo et al.,
1991). Other than strictly day/night fluctuations, plasma MLT levels
increase with seawater adaptation in salmonids (Gern et al., 1984) and
cyclic MLT levels in plasma have been implicated in sexual matura-
tion by altering daily cycles in plasma GtH concentrations (Hontela,

F. Thyroid Hormones
    The thyroid hormones are thyroxine (T4) and triiodothyronine (TJ).
While both hormones circulate in the blood, T4 is the primary, if not
the only, substance released by the thyroid gland, at least in teleosts
(Eales and MacLatchy, 1989), while T3 is mainly formed by de-
iodination of T4 in peripheral tissues. Triiodothyronine is regarded as
the active hormone and T4 a prohormone; nuclear receptors have a
much higher binding affinity for T3 than T4 (at least fourfold higher;
Eales, 1985). Triiodothyronine and Tq levels are readily measured in
plasma with commercial kits that have detection limits (-0.1 ng ml-')
appropriate for measuring routine minimum plasma levels in fish
(0.1-1.0 ng ml-'). It has become fairly common practice to measure
both substances simultaneously in plasma, T4 levels because they are
more sensitive to activation of the hypothalamic-pituitary-thyroid
axis and thus to environmental influences (Leatherland, 1988), and Tx
levels because this is the biologically active molecule. The interpreta-
tion of the meaning of changing plasma hormone levels is, however,
particularly difficult in the case of the thyroid hormones. For example,
a decrease in plasma T4 levels could reflect a decrease in secretion by
the thyroid gland or an increase in conversion of T4 to T3 in peripheral
tissues. Since T3 levels are a function not only of T4 secretion rates but
also of the rate of extrathyroidal deiodination, there is no universally
established relationship between plasma T3 and T4 levels. For exam-
ple, in brook trout, plasma T3 levels are usually consistently higher
than T4 levels (White and Henderson, 1977), while in coho salmon the
2.   CHEMICAL PROPERTIES OF THE BLOOD                                 69

reverse is true (Dickhoff and Darling, 1983), and in still others, the
levels of the two are similar,
    Most of the circulating TBand T4 are bound to plasma proteins (Eales
and Shostak, 1985) leaving less than 1% of the total hormones in the
free or diffusible form. Variations in plasma protein concentration can
then explain some of the fluctuations in plasma thyroid hormone levels
(Laidley and Leatherland, 1988a).
    In general, thyroid hormones are elevated when fish are under
conditions favorable for somatic growth (after food intake or treatment
with androgens or growth hormone) and lowered when conditions are
unfavorable (starvation, stress, or high estradiol levels associated with
vitellogenesis) (Dickhoff and Darling, 1983; Eales and MacLatchy,
1989). In addition, thyroid hormones can show diurnal (increasing
during daylight hours) and annual variations (White and Henderson,
1977) and are elevated during larval development, smoltification (parr-
smolt transformation), and during reproductive maturation (Specker,
1988; Dickhoff and Darling, 1983). These latter elevations indicate an
important role in preparing animals to exploit new habitats; a switch to
active feeding in the case of larvae and increased adaptability to sea-
water in the case of smolts. Environmental factors such as daylength,
temperature, lunar cycle events, pH, salinity, and nutrition are all
implicated in stimulating thyroid activity (Grau, 1988).
    For salmonids at least, the elevation of thyroid hormones during
smoltification is apparently the largest observed. Peak levels around
10-15 ng ml-' are usually reported for salmonids smolts (Prunet et al.,
1989; Young et al., 1989), but levels as high as 40 ng ml-' (T4)have
been reported in coho salmon (Dickhoff and Darling, 1983). In other
circumstances the elevation in thyroid hormones is less, with an increase
from 1 to 4 ng ml-I being fairly typical (e.g., due to die1 variation
[rainbow trout, Cook and Eales, 19871 and in response to feeding
[Leatherland and Hilton, 1988; Himick and Eales, 19901).

G. Calcitonin
    Calcitonin (CT) is a peptide hormone (MW of salmon C T is 3452)
produced by the ultimobranchial bodies. Calcitonin function in fishes
is reviewed in Copp (1969) in Volume 11, Clark (1983);and Pang and
Pang (1986). Homologous RIAs have been developed for salmon and
for Japanese eel (Anguilla japonica; see Table I for references) with
detection limits <1 ng ml-'.
    Although CT is present in large amounts in the gland and can be
greatly elevated in the circulation, it does not appear to play a promi-
70                                 D. G. MCDONALD AND C. L. MILLIGAN

nent role, at least directly, in Ca2+ regulation (Clark, 1983). This is in
contrast to tetrapods where CTs role is to correct excessive calcium
levels, although in this context it is antihypercalcemic rather than
hypocalcemic (Talmage et al., 1980). Hypercalcemia in coho salmon,
brought about by either netting and confinement stress in freshwater
or b y seawater transfer, does not elevate plasma C T levels (Bjornsson
et al., 1989), nor do C T levels change in Atlantic cod (Gadus morhua)
exposed to Ca2+-enriched (100 mM) seawater (Bjornsson and Deftos,
1985).In contrast, hypercalcemia in mud skipper (Periophthalmodon
schlosseri) induced by air exposure could be corrected by daily CT
injections (Fenwick and Lam, 1988).
    Fish have very precise plasma Ca2+ homeostasis and most, if not
all, of the major changes in plasma C T levels and in plasma [Ca"] are
associated with gonadal maturation. I n rainbow trout C T levels are
around 1 ng ml-' in immature males and females but rise to 23 and
47 ng ml-', respectively, with sexual maturation (Fouchereau-Peron et
al., 1990). Female Japanese eels exhibit a similar increase (from 1 to
40 ng ml-l) with sexual maturation (Yamauchi et al., 1978). Accom-
panying maturation in females are major increases in plasma Ca2+ as-
 sociated with vitellogenesis. However, the Ca2+ increase, by itself, is
not responsible for the elevation in CT (Yamauchi et al., 1978;
Bjornsson et al., 1986).Rather, the rise in C T in both sexes is probably
most closely associated with the rise in plasma concentrations of the
 gonadal steroids, GtH, or both. However, it is still possible that C T
 may play a role in Ca2+ homeostasis by protecting the skeleton during
periods of the increased Ca2+ demand of oogenesis (Bjornsson et al.,

H. Cortisol
    Cortisol is the main (380%) circulating corticosteroid hormone in
teleosts (Donaldson, 1981). It is probably the most frequently mea-
sured hormone in fishes. Commercial kits are available and are valid
for measuring teleost cortisol levels, providing the protein levels of
standards and unknowns are similar. Detection limits are 3-8 ng ml-'
for commercial kits, but improvements in assay procedures have per-
mitted detection limits of 0.5-2 ng ml-' (Bry, 1982; Pickering and
Pottinger, 1983).Resting levels as low as 5 ng ml-' have been reported
for salmonids by Pickering and co-workers, although studies on other
teleosts (Peter et al., 1978; Lamba et al., 1983; Venkatesh et aZ., 1989)
report minimum levels that are higher (10-50 ng ml-'). The most
usual reason for an increase in plasma cortisol levels is stress. Indeed,
2.   CHEMICAL PROPERTIES OF THE BLOOD                                   71

cortisol measurements are the method of choice for quantifying stress
in fishes (Donaldson, 1981). Stressful circumstances include handling,
close confinement, transport or other physical disturbances, and rapid
changes in water quality (e.g., pH, salinity, or temperature).
    The response to stress is mediated by ACTH and can be very rapid.
For example, in salmonids exposed to handling stress, ACTH levels
were significantly elevated in 2 min and cortisol levels within 10 min
of the start of the stress (Sumpter et al., 1986). For a given stress there
are marked interspecific differences in the amount of cortisol elevation
(Davis and Parker, 1983,1986).The greatest elevations appear to be in
salmonids (Davis and Parker, 1986), although the response is greater in
wild than in cultured (i.e., domesticated) salmonids (Woodward and
Strange, 1987) and the increase is larger and faster at higher tempera-
tures (Sumpter et al., 1985). Generally speaking, there appears to be an
approximate correlation between metabolic scope and cortisol eleva-
tion in response to stress. In metabolically active species such
as salmonids and cyprinids, cortisol elevations as high as 400-
600 ng ml-' have been reported, while species of intermediate
activity such as largemouth bass (Micropterus salmoides) have lev-
els around 200 ng ml-' and inactive species such as the holosts, gar
(Lepisosteus sp.), and bowfin (Amia calua) have peaks around 50-
60 ng ml-' (Davis and Parker, 1986).
    Other than strictly stressful conditions, there are other circum-
stances under which cortisol rises. These include die1 cycles (see
earlier discussion), smoltification in anadromous salmonids (Young et
aZ., 1989), feeding (Bry, 1982; Pickering and Pottinger, 1983), and
sexual maturation (Pickering and Christie, 1981).

I. Catecholamines
    The main circulating catecholamines (CAs) in teleosts are adrena-
line (A) and noradrenaline (NA), which originate predominantly from
the chromaffin tissue, mostly in the head kidney and postcardinal
veins. Plasma levels have been extensively measured but only in a few
species (e.g., rainbow trout, eel, cod, carp, flounder, dogfish, and lam-
prey; see summary tables in Mazeaud and Mazeaud, 1981; Milligan
and Wood, 1987; Tang and Boutilier, 1988; Perry et al., 1989; Thomas
and Perry, 1991; Randall and Perry, 1992). Measurement techniques
include fluorimetric, radioenzymatic, and HPLC methods, with the
latter being the most popular. Detection limits are 51 nM.
    Resting levels of A and NA are usually similar and are typically less
than 5 nM but such low levels are achievable only by blood sampling
72                                D . G . MCDONALD AND C. L. MILLIGAN

via indwelling catheters. Acute sampling methods (see earlier discus-
sion) cause rapid and massive elevation of CA levels in plasma, at least
in salmonids. The degree of elevation depends on the method of
immobilizing the animal and the rapidity of sampling; however,
plasma levels of A reach 700 nM in rainbow trout stunned by a
blow to the head, and blood collected by caudal severence, all
within 30 sec (McDonald and Goldstein, 1992). Even higher levels
(>1000 nM) have been reported for Pacific salmon (Oncorhynchus
tshawytscha) after struggling in air, handling, and hauling in the net
(Mazeaud and Mazeaud, 1981). Treatments that cause CA elevation in
cannulated animals are reviewed in Randall and Perry (Chapter 4)and
include hypoxia, hypercapnea, air exposure, exhaustive or violent ex-
ercise, metabolic acidosis, and anemia. Violent, enforced exercise and
deep hypoxia appear to be the most effective means to elevate CAs
with levels in cannulated animals reaching as high as >200 and >100
nM respectively for A and NA in rainbow trout after enforced exercise
(Butler et al., 1986), and A > 400 nM in rainbow trout during exposure
to severe hypoxia (PO, < 20 mmHg = 3 KPa; Thomas and Perry, 1992).
In salmonids, the elevation in A is typically greater than NA reflecting
the higher A content of the chromaffin tissue. Although only a limited
number of teleosts have been examined there does appear to be
marked species differences in CA mobilization that may in part be
related to the activity level of the species. For example, Milligan and
Wood (1987) showed that A levels were marginally elevated in starry
 flounder (Platichthys stellatus) after enforced exhaustive exercise,
 whereas a similar treatment in rainbow trout produced a 10-to 100-fold
 elevation in A. Similarly, deep hypoxia in the American eel (Anguilla
 rostrata) produced A levels of only about 10 nM while a comparable
 treatment produced 400 nM in the rainbow trout and in Salmo fario
 (Thomas and Perry, 1991). In addition to marked species differences
there are seasonal and water temperature related variations in cate-
 cholamine release (Mazeaud and Mazeaud, 1981; Randall and Perry,
 Chapter 4 .

J. Pancreatic Hormones
    The pancreas of teleosts contain the following peptide hormones
(listed in order of decreasing content in coho salmon endocrine pan-
creas): insulin (INS), somatostatin-25, glucagon-like peptide (GLP),
glucagon (GLU), somatostatin-14, and pancreatic peptide (PP) (see
Plisetskaya, 1989, 1990 for reviews). Homologous RIAs have been
developed for INS, GLU, GLP, and somatostatin-25 in coho salmon
2.   CHEMICAL PROPERTIES OF THE BLOOD                                 73

 and for INS in cod and tuna (see Table I for references). Although
 structurally similar to mammalian INS, fish INS are very different
 immunologically, hence mammalian RIAs are unsuitable (Plisetskaya,
 1989). In contrast, fish GLU can be measured with mammalian RIAs
 (Gutikrrez et aZ., 1986).
    Routine INS levels in regularly fed teleosts range from 5 to
20 ng ml-' with no consistent interspecific differences (Gutikrrez et
aZ., 1984; Moon et aZ., 1989; Dickhoff et al., 1989; Hemre et al., 1990).
With prolonged starvation (3-6 wk) INS drops to very low levels
(0.2-1.0 ng ml-'; Moon et al., 1989; Hemre et al., 1990). Feeding
typically increases INS secretion. Amino acids are the most powerful
stimulators of INS secretion in fish with arginine being the most potent
of the amino acids tested (Plisetskaya, 1989).Nonetheless, high carbo-
hydrate diets are more effective at stimulating INS release than low
carbohydrate diets (Hilton et aZ., 1987; Hemre et al., 1990). In one
extreme case, INS levels as high as 48 ng ml-I were reported in
rainbow trout 3 h after a high carbohydrate meal (Hilton et al., 1987).
    Insulin levels also increase with the parr-smolt transition in salmo-
nids (Dickhoff et al., 1989; Plisetskaya, 1990), decrease during spaw-
ning and with a drop in water temperature, and show seasonal fluctua-
tions (GutiCrrez et al., 1987a). Starvation may be responsible for some
of the drops in INS levels.
    Glucagon and GLP levels fluctuate to a smaller extent than INS.
For example, both are in the range of 0.5-2.0 ng ml-l in coho salmon
and rainbow trout (Plisetskaya, 1990).Glucagon and GLP levels de-
crease with starvation (Gutikrrez et al., 1990; Moon et al., 1989; sea
bass, rainbow trout) but the ratio of GLU (and GLP) to INS increases;
this change is consistent with activation of gluconeogenesis (Moon et
al., 1989).

K. Stanniocalcin
   Stanniocalcin (STC, formerly called hypocalcin or teleocalcin, Flik
et al., 1990) is a glycoprotein hormone released from the Stannius
corpuscles (CS, unique to teleost and holostean fish) and is the pre-
dominant hypocalcemic hormone in fish. Removal of the glands results
in prolonged hypercalcemia, particularly in seawater fish. A homol-
ogous enzyme-linked immunosorbent assay (ELISA) has been devel-
oped for Atlantic salmon STC (MW 54,000; Mayer-Gostan et al., 1992)
that has detection limits of 0.2 ng ml-'. In Atlantic salmon, STC levels
were 40 and 150 ng ml-' in freshwater- and seawater-adapted forms,
respectively. A survey of STC levels in 11 marine teleost species
yielded values of 76-186 ng ml-'.
74                                 D. G . MCDONALD A N D C. L. M I L L I G A N

L. Urotensins
    Urotensin I (UI) and urotensin II (UII) are neuropeptide hormones
released from the caudal urophysis in teleosts. Urotensin I is a homol-
ogue of mammalian corticotropin releasing factor (CRF) and has a
molecular weight of 4864 while UII is a partial homologue of mamma-
lian somatostatin and is a 12-amino-acid peptide of MW 1351. Homol-
ogous RIAs have been developed for urotensins I and I1 extracted from
white suckers, Catastomus commersoni (Suess et al., 1986; Kobayashi
et al., 1986).Blood levels are low (74 and 55 pg ml-’, respectively) and
difficult to measure (Hontela et al., 1989). Nevertheless, measure-
ments of urotensins in urophysial tissue indicate a role in osmoregu-
lation. Tissue levels increase in brook trout in response to low p H
exposure (Hontela et al., 1989) and increase in flounder, Platichthys
flesus, while adapting to seawater (Arnold-Reed et al., 1991).


     At the present time neither the presence or absence nor the identity
of most of the peptide and protein hormones of cyclostomes and chron-
dricthyes is known and no homologous RIAs have, therefore, been
developed. However, there has been some limited use of heterologous
RIAs (e.g., insulin levels in spotted dogfish [Scyliorhinus canicula;
Gutierrez et al., 19881). Consequently, the following discussion is
confined to nonprotein hormones: steroids, thyroid hormones, and
    In cyclostomes the steroidogenic tissues have not been localized
(Gorbman, 1989) and the steroid hormones not precisely identified. A
variety of steroids have been detected as circulating in the plasma
(Katz et al., 1982; Kime and Larsen, 1987) but the active sex steroids
are thought to be hydroxylated derivatives of T and Ez in males and
females respectively (Larsen, 1990).
    In chondrichthyes, the sex steroids are similar to those of teleosts.
In females, the important gonadal steroids are Ez, T, and progesterone.
In immature animals steroid levels are 51 ng ml-’. Ez and T are the
principal steroids during the preovulatory phase of the cycle, while P
is elevated approximately 24-48 h prior to ovulation. Peak levels of sex
steroids vary depending on the steroid and the species but most peaks
fall within the range of 10-40 ng ml-’ (Dodd, 1983; Callard and
Klosterman, 1988). I n males, there are less data available but, like
2.   CHEMICAL PROPERTIES OF THE BLOOD                                  75

teleosts, T is apparently the main sex steroid. Testosterone levels
are around 10 ng ml-' in mature male spotted dogfish (Dobson and
Dodd, 1977) but very high levels have been reported in skate, Raja sp.
(>2000 ng ml-') associated with very high levels ofa plasma testoster-
one binding protein (Callard, 1988).
    Cortisol is riot the steroid stress hormone in either cyclostomes
or chrondrichthyes. In the sea lamprey, Petromyzon marinus, an-
drostenedione may be the cortisol analog. It increases from 4 to
14 ng ml-' with agitation stress (Katz et al., 1982). In elasmobranchs
the main interrenal steroid is 1a-hydroxycorticosterone, but its precise
physiological function remains unclear, although mineralocorticoid
rather than glucocorticoid activity has been reported. Resting levels in
spotted dogfish are around 4 ng ml-' and increase three- to fourfold
when salinity is reduced (Balment et al., 1987). In general, it appears
that the corticosteroid stress responses in cyclostomes and chron-
dricthyes are small relative to those of teleosts.
    The physiological role of thyroid hormones in cyclostomes is not
precisely known (Plisetskaya et al., 1983; Specker, 1988), and the
thyroid does not appear to be under pituitary control (Dickhoff and
Darling, 1983). Most of the studies have been done on lampreys.
Thyroid hormone levels are particularly high in larvae (ammocoetes),
Tq is 70 ng ml-' and T3 is 16 ng ml-' (Wright and Youson, 1977) but
decrease (50-90%) at the onset of metamorphosis to adults. The levels
in ammocoetes are as much as 10 times higher than those of most other
vertebrates but levels in juveniles are within the normal vertebrate
range (Youson, 1988). Larval and metamorphosing forms also have a
much higher liver T3 binding capacity than adults (Lintlop and
Youson, 1983).
    The circulating catecholamines in cyclostomes and chron-
drichthyes are identical to those of teleosts (Dashow et al., 1982).
Resting levels in lampreys, measured on cannulated animals, are sub-
stantially higher than those measured in resting teleosts but the re-
sponse to stress has a much lower amplitude. Adrenaline and nor-
adrenaline were 20 and 8 nM, respectively, and rose only 3- and %fold,
respectively, after 5 min of agitation stress (Dashow et i . 1982). In the
spotted dogfish resting catecholamine levels were similar to those of
the lamprey (A and NA were 6 and 14 nM, respectively) but the re-
sponse to agitation stress (repeated burst swimming for 2-3 min) was
much larger (16-and y-fold, respectively; Butler et al., 1986).Nonethe-
less, in comparison to active teleost species, adrenaline mobilization is
still relatively small. In rainbow trout, treated comparably, A and NA
elevations were 150- and g-fold, respectively (Butler et al., 1986).
76                                   D. G. MCDONALD AND C . L. MILLIGAN


A. Glucose
   Glucose is one of the most frequently measured blood metabolites
and is, perhaps, the most variable. Typically, whole blood rather than
plasma levels are reported. Free glucose in the red cells is quite low
(<1mM) and small variations in hematocrit have little impact on whole
blood glucose levels. Resting blood glucose levels range from lows of
0.2 mM in the African lungfish (Proptopterus acthiopicus; Dunn et al.,
1983) to highs of 15 mM in tuna (Katsuwonus pelamis; Arthur et al.,
1991).Interspecific variation in whole blood glucose levels has been
attributed to differences in activity levels of fish: More sluggish,
benthic species tend to have lower blood glucose levels than d o more
active, pelagic species (Umminger, 1977). Glucose levels also vary
among individuals within a species. In the extreme, blood [glucose]
ranged from 0.3 to 8.8 mM in 26 individual kelp bass (Paralabrax sp.,
Bever et al., 1977) caught from the wild and held in the laboratory.
Some of this individual variation can be attributed to differences in
size, age, and the nutritional and reproductive states of the fish
(Fletcher, 1985).
    Blood glucose levels are often cited as being a sensitive physiologi-
cal indicator of stress in fish (Wedermeyer and McLeay, 1981). In
general, stress (e.g., hypoxia, crowding, handling, forced exercise, dis-
ease, angling, captivity) causes a relative hyperglycemia of variable
extent and duration. The severity of the stress in large part dictates the
degree and duration of the hyperglycemic response. For example, in
pike (Esox lucius),angling stress resulted in a threefold increase (from 2 to
6.2 mM) in blood glucose, whereas mild chasing was without effect
(Schwalme and McKay, 1985).
    The source of the elevated blood glucose is hepatic glycogen,
which is mobilized in response to catecholaminergic stimulation
(Mazeaud and Mazeaud, 1981). The intra- and interspecific variability
in glycemic response to stress may reflect differences in catecholamine
secretion (Mazeaud and Mazeaud, 1981),hepatic glycogen reserves, or
both. The latter may explain why, in some instances, chronic stress
results in hypoglycemia (Fletcher, 1985).To further complicate mat-
ters, in some species (e.g., the European eel, Anguilla anguilla; Soko-
lowska and Bieniarz, 1981) the time of day the stress is applied influ-
ences the glycemic response.
    Nutritional status also influences blood glucose levels. All species
examined show a postprandial hyperglycemia, the duration and mag-
nitude of which is dependent on the dietary carbohydrate level (Hilton
2.   CHEMICAL PROPERTIES OF THE BLOOD                                77

and Atkinson, 1982; Fletcher, 1985; Suarez and Mommsen, 1987;
Moon, 1988). The effect of starvation on blood glucose levels is
species- and time-dependent. Mild hypoglycemia (i.e., blood glucose
rarely drops by more than 25-30% of resting levels) is observed after 6
weeks starvation in rainbow trout (Moon et al., 1989) and sea bass
(Zammit and Newsholme, 1979; Gutikrrez et al., 1990)and after 22 but
not 7 days starvation in the sand dab (Limanda limanda; Fletcher,
    Several species show blood glucose homeostasis with prolonged
starvation; after 150 days starvation, blood glucose was unaltered in
Squalus acanthias (deRoos et al., 1985), the spotted dogfish (Zammit
and Newsholme, 1979), and the kelp bass (Bever et al., 1977). Perhaps
the record for tolerance to food deprivation belongs to the American
eel; after 36 months of starvation, blood glucose levels remained con-
stant at about 9 mM (Cornish and Moon, 1985). The mechanism for the
maintenance of blood glucose during prolonged starvation is not clear.
Metabolic depression, resulting in a decrease in glucose utilization,
may be involved (Cornish and Moon, 1985) or, alternatively, gluco-
neogenesis from amino acids and fatty acids (Suarez and Mommsen,
1987) may be activated.
    In those species that show seasonal variations in blood glucose
levels, it is associated with the reproductive cycle. In rainbow trout
(Miller et al., 1983), the spotted dogfish (Gutikrrez et al., 1988), and
Spicara chryselis (FernBndez and Planas, 1980), the lowest blood glu-
cose levels (3.9, 0.5, and 2.7 mM, respectively) were associated with
peak gonadal development and spawning, whereas the highest levels
(11.7, 1.0, and 8.3 mM, respectively) were associated with post-
spawning feeding activity.
    Although blood glucose levels vary considerably among species
(Fletcher, 1985),within a species, and over time within an individual
(Bever et al., 1977),severe hypoglycemia (i.e., >50% decline in blood
glucose) does not occur. Suarez and Mommsen (1987) argue that be-
cause glucose turnover is directly related to blood [glucose], mainte-
nance of a critical blood glucose level, albeit over a wide range, is
important in ensuring that glucose turnover continues in the face of
starvation. This strategy appears to be important in maintaining fuel to
the nervous system, particularly in species that experience prolonged
starvation as part of their life cycle.

B. Lactate
   Lactate is the major end product of anaerobic metabolism in verte-
brates and is routinely found in low levels, <1 mM, in fish blood. Its
78                                   D. G. MCDONALD AND C . L. hlILLIGAN

accumulation in the blood is usually indicative of oxygen limitation.
Blood lactate levels increase in response to environmental hypoxia
(low inspired P o , ) , internal hypoxemia (e.g., anemia, gill dysfunction),
or strenuous activity. The elevation of blood lactate in response to
environmental hypoxia is directly related to the degree of hypoxia: the
lower the Po,, the higher the blood lactate (Boutilier et al., 1988).The
P o , at which anaerobic metabolism is activated and lactate appears in
the blood, the critical P o , , varies among species (Boutilier et al., 1988).
In general, the more active the species, the higher the critical Po,.
     Similarly, the degree to which lactate accumulates in the blood
following a bout of strenuous activity (e.g., enforced swimming,
angling) is species specific and appears to be correlated with life-style
and habitat. In active, pelagic species, including pike, salmonids, tu-
nas ( K . pelamis, Thunnus albacares),marlins (Tetaptusrus audas, Ma-
kaira nigricans, M. indica), and a number of sharks (Zsurus oxyr-
hinchus, S . canicula; see Wood and Perry [1985] for a review;
Schwalme and McKay l19851; Wells et al. [1986]), blood lactate levels
are often in the range of 15-20 mM following a bout of activity. In
contrast, in benthic, more sluggish species, including flatfish (e.g., P .
stellatus, Hippoglossoides elassodon, Pseudopleuronectes ameri-
canus), toadfish (Opsanus beta), sea raven (Hemitripterus ameri-
canus), and skates (Raja ocellata) (see Wood and Perry [1985] for a
review; Walsh [1989]), strenuous activity rarely raises blood lactate
levels in excess of 1-2 mM. Though well-documented, the reasons for
the differences in postexercise blood lactate levels between active,
pelagic, and benthic, sluggish species are unknown.

C. Ethanol
   While lactate is considered the main anaerobic end product in
vertebrates, there are a few fish species in which ethanol has been
found to accumulate in significant amounts under anaerobic condi-
tions. Goldfish (Shoubridge and Hochachka, 1980; van den Thillart
and Verbeek, 1982), crucian carp (Carassius carassius; Johnston and
Bernard, 1983), and bitterling (Rhodeus amarus; Wissing and Zebe,
1988) produce ethanol only in response to severe (Po, < 1.33 kPa,
1 mmHg = 0.133 kPa) prolonged hypoxia or anoxia and can reach
levels of 2-4 mM in the blood (versus resting levels of <1 mM; Shou-
bridge and Hochachka, 1980; Johnston and Bernard, 1983).

D. Ketone Bodies
     Ketone bodies, acetoacetate and P-hydroxybutyrate, are produced
2.   CHEMICAL PROPERTIES OF THE BLOOD                                  79

primarily in the liver from nonesterified fatty acids. The abundance of
ketone bodies in elasniobranch blood and their virtual absence in that
of teleosts may be linked to the absence of free fatty acids in elasmo-
branch plasma (see Section VII). Elasmobranchs may rely upon ketone
bodies generated in liver as a fuel substitute for fatty acids (Zammit
and Newsholme, 1979). Elasmobranch blood contains both P-hydroxy-
butyrate and acetoacetate, although the former predominates (0.08-
9.6 mM versus 0.03-055 mM; Zammit and Newsholme, 1979; deRoos
et al., 1985; Gutikrrez et al., 1988). Teleost blood, however, contains
only acetoacetate and at much lower levels (0.05-0.1 mM; Zammit and
Newsholme, 1979; Fernandez et al., 1989).Plasma ketone levels are
influenced by both starvation and reproductive status. In elasmo-
branchs, plasma levels of both acetoacetate and P-hydroxybutyrate
increase in response to long-term starvation (weeks), although the
latter increases to a much greater extent (2O-foldversus 6-fold increase;
Zammit and Newsholme, 1979; deRoos et al., 1985). In teleosts, star-
vation was without effect on plasma acetoacetate levels (Zammit and
Newsholme, 1979; FernBndez et al., 1989). In both elasmobranchs and
teleosts, the highest plasma levels of ketones (8-9 mM and 0.1-
0.5 mM, respectively) were observed in prespawning individuals, dur-
ing the period of gonadal development ( F e r n h d e z and Planas, 1980;
Gutierrez et al., 1987a; Fernandez et al., 1989).

E. Bile Pigments
   Biliverdin and bilirubin are common bile pigments formed in the
liver from the breakdown of heme and other porphyrine proteins. In
most fish, they are stored in the gall bladder and are removed from the
body when food moves along the gut. Thus, in most species, the plasma
levels ofthese pigments are quite low (e.g., 1-5 p M ; Wells et al., 1986).
The lamprey (Lampteru lamottenii), however, is a notable exception.
Upon metamorphosis from the ammocoete to adult form, the liver
undergoes extensive reorganization; the bile ducts degenerate and
the gall bladder is lost (Sidon and Youson, 1983). Bile pigments have
not been detected in the plasma of ammocoetes, but both bilirubin
(-20 p M ) and biliverdin (-2 p M ) are found in juvenile and adult
upstream migrant fish (Makos and Youson, 1987).
    A number of cottids and anguillids have blue-green, lavender, and
purple colored plasma (Yamaguchi, 1973; Low and Bada, 1974; Ellis
and Poluhowich, 1981; Makos and Youson, 1987; Eng and Youson,
 1991). Also, the plasma of female lumpfish (CycZopterus Zumpus) is a
deep blue-green whereas that of males is reddish in color (Mudge and
Davenport, 1986).The blue-green coloration is due to the presence of
80                                D. G. MCDONALD AND C. L. MILLIGAN

biliverdin and to a lesser extent, bilirubin. It is not clear why these
pigments accumulate in the plasma and why they are not toxic since
biliverdin levels in these species are in excess of 100 p M ; well above
the levels considered toxic in mammals (25 pM;Low and Bada, 1974;
Mudge and Davenport, 1986). The purpose (if any) of the serum pig-
mentation is obscure, although Mudge and Davenport (1986) have
suggested that plasma pigmentation may be involved in sexual signal-
ling in the lumpfish.


A. Amino Acids
    Amino acids are transported within erythrocytes and the plasma,
although the relative importance of each pool is a matter of some debate
(Dabrowski, 1982; Murai and Ogata, 1990; van der Boon et al., 1991).
The distribution of free amino acids between erythrocytes and plasma
differs among species (see Table 11), and there are no obvious trends.
Furthermore, in some instances, the response of red cell free amino
acids is independent of that of the plasma. For example, in rainbow
trout after exhaustive exercise, some free amino acids in red cells
decrease whereas other amino acids in the plasma increase (S. Eros
and L. Milligan, unpublished observations).
    Although the data are sparse, it appears that plasma amino acid
levels are higher in marine elasmobranchs than in either marine or
freshwater teleosts or marine agnathids (Boyd et aZ., 1977).This is most
likely a consequence of the high extracellular osmotic pressure main-
tained by organic osmolytes in elasmobranchs. When skate and sting-
ray (Dasyatis sabina), euryhaline elasmobranchs, are moved into a
dilute environment (15%)there is a decrease in the level of several
amino acids, particularly taurine, in the red cells (Boyd et aZ., 1977).
There is, however, no effect of salinity changes on plasma amino acid
levels, further illustrating the independence of free amino acids in red
cells and plasma.
    Fish show minimal dietary requirements for certain amino acids.
The essential amino acids requirements have been determined for
only a few species, but these show little interspecific variation. Whole
blood and plasma levels of essential amino acids (EAA) are mainly a
function of dietary protein intake and are positively correlated with
their levels in the diet (e.g., carp, Dabrowski, 1982; Murai and Ogata,
2.   CHEMICAL PROPERTIES OF THE BLOOD                                   81
1990; trout, Walton and Wilson, 1986). Nonessential amino acids, in
contrast, are independent of dietary levels (Plakas et al., 1980;Walton
and Wilson, 1986).
    Although the plasma levels of EAA are similar among species (at
least in the few examined),the abundance of nonessential amino acids
in the plasma varies considerably. For example, in the common carp
(Dabrowski, 1982) and tilapia (Yamada et al., 1982) plasma is abundant
in proline and threonine, whereas in sockeye salmon (0. nerka;
Mommsen et al., 1980) and rainbow trout (Nose, 1972; Kaushik and
Luquet, 1977), glycine, alanine, and lysine are most abundant, and
serine and lysine are abundant in channel catfish (Zctalurus punctatus;
Wilson et al., 1985).
    By and large, the majority of studies on plasma amino acids in fish
have examined the influence of food deprivation and various diets,
with the goal being to define EAA requirements. In all species exam-
ined, the highest whole blood and plasma amino acid levels are seen
within hours after a meal. The response of plasma amino acid levels to
fasting is complex; dependent on, among other things, species and
duration of starvation. In general, plasma levels of the EAA tend to
decline at the start of starvation (i.e., within days), and subsequently
increase to prestarvation levels, while there is little change in the level
of nonessential amino acids (Timoshina and Shabalina, 1972; Blasco et
al., 1991). A striking example of the preservation of plasma amino acid
levels during prolonged starvation is seen in sockeye salmon. The
long-term starvation associated with spawning migration caused only
very modest changes in plasma amino acid levels (Mommsen et al.,
1980), despite significant muscle proteolysis. Some amino acids that
are poor metabolic substrates (e.g., lysine and valine) tend to accumu-
late in the plasma as a result of muscle proteolysis.
    Few studies have examined the influence of seasonality or photo-
period on circulating levels of free amino acids. To our knowledge, the
influence of photoperiod on circulating amino acid levels has been
assessed in only two species: goldfish (Carrillo et al., 1980) and sea
bass (Carrillo et al., 1982). The story emerging is complex: Only some
amino acids show a circadian rhythm and, at least in the goldfish, the
rhythmicity is dependent on duration of the light phase. For example,
plasma serine showed a circadian rhythm when fish were exposed to
short day lengths (9L : 13D) and natural day lengths (1OL : 14D) but no
rhythmicity was observed in fish exposed to long day lengths
(15L : 9D) (Carrillo et al., 1980). The explanation for this is not at all
clear but is undoubtedly linked to the complex hormonal changes
accompanying changes in day length.
                                                                 Table I1
                      Whole Blood, Plasma, and Erythrocyte Amino Acid Levels in an Agnathid, Elasmobranch, and Teleost

M                                                                     Dogfish
                                          Pacific hagfish           Scyliorhinus                        Common carp
                                         Eptatretus stoutin          caniculab                         C yprinus carpio"
                                                                                                             Calculated     Measured
                                      Erythrocyte                Whole                  Whole               erythrocyte'   erythrocytef
                                       (mM cell       Plasma     blood'/    Plasma      blood'   Plasma     (mM whole      (mM whole
             Amino acid                 water)         (mM)       (mM)       (mM)       (mM)      (mM)         blood)         blood)

     Arginine                             3.36         0.265      0.20       0.31       0.256*    0.138*       0.047*        0.088*
     Histidine                            1.22         0.014      0.11       0.08       0.169*    0.092*       0.042*        0.040*
     Isoleucine                           3.18         0.059      0.32       0.28       0.103*    0.068*       0.035*        0.017*
     Leucine                              8.76         0.159      0.52       0.39       0.183*    0.127*       0.056*        0.035*
     Lysine                               4.00         0.352      1.01       0.57       0.451*    0.214*       0.052*        0.067*
     Methionine                           2.72         0.053      0.07       0.12       0.162*    0.069*       0.093*        0.065*
     Phen ylalanine                       4.73         0.075      0.14       0.11       0.103*    0.076*       0.027*        0.013*
     Threonine                            7.58         0.113      0.24       0.07       0.702*    0.304*       0.350*        0.328*
     Valine                                    6.27       0.106        1.03       0.43       0.161*      0.112*   0.048*   0.033*
     Cystine                                   -            -          -           -         0.042       0.021      -      0.034
     Tyrosine                                  4.16       0.058       0.11        0.10       0.068       0.054    0.014    0.008
     Alan i n e                                8.99       0.041       1.25        0.32       0.418       0.218    0.157    0.264
     Aspartate                                 1.63       0.049       0.22        0.04       0.134       0.134    0.153    0.125
     Glutamate                                 3.00       0.064       0.22        0.03       0.101       0.053    0.109    0.105
     Glutamine                                 6.93       0.108       0.22        0.12         -           -        -        -
     Glycine                                   3.72       0.115       3.66        0.19       0.602       0.315    0.264    0.332
     Proline                                  16.8        0.097       0.56        0.36       0.572       0.246    0.167    0.349
     Serine                                    4.15       0.092       0.36        0.15       0.236       0.115    0.115    0.135
     Taurine                                  N            ND         7.44        0.23                              -
     ?-Amino butyric acid (GABA)               1.09       0.020
     a-Amino-n-butyrate                        2.23       0.11

        * Essential amino acid for this species.
          Fincham et al. (1990).
CJ        GutiCrrez et al., (198713). Samples taken in November, 24 h after feeding.
          Dabrowski (1982)Samples taken 72 h after feeding.
          Measured in deproteinized extracts of whole blood. Hematocrits not given.
          Plasma [amino acid] was subtracted from whole blood [amino acid].
        f Erythrocytes were washed twice with 0.65% Ringer solution prior to deproteinizing red cells.
        g Not detectable.
84                                D. G. MCDONALD AND C. L. MILLIGAN

B. Ammonia

    Because of the high pk of ammonia (-9.5; Cameron and Heisler,
1983) relative to the physiological p H of fish plasma (-7.6-8.0; see
Heisler, 1984 for a review of fish acid-base physiology), >99% of
ammonia is present in the ionized form, NH4+. Ammonia, as NH3 and
NH4+,is the primary transport vehicle for nitrogen in plasma of most
fish, and the unionized form, NH3, can be extremely toxic if allowed to
accumulate in the body. Under aerobic conditions most ammonia is
produced in the liver, but under anoxic conditions, liver production is
reduced and muscle proteolysis becomes the main source of ammonia
(van Waarde, 1983). During exhaustive exercise, adenylate deam-
ination in the muscle becomes a major source of ammonia (Dobson and
Hochachka, 1987), although its quantitative importance to total ammo-
nia production depends on the activity level of the animal; increasing
with increasing workload (Driedzic and Hochachka, 1976).
    Plasma total [ammonia] is variable and dependent, among other
things, on the site of blood sampling. In tilapia, Oreochromis nilotica,
plasma ammonia levels in blood sampled via caudal puncture
(prehepatic blood) were greater than that in plasma obtained via car-
diac puncture (posthepatic blood), indicating a significant hepatic am-
monia uptake (Wood et al., 1989). Similarly in rainbow trout, the
plasma total [ammonia] in blood drawn from a dorsal aorta catheter
(posthepatic) was about 30%ofthe plasma total [ammonia] in the blood
sampled by caudal puncture (prehepatic blood; Wood et al., 1989),
although the stress associated with caudal sampling may have contrib-
uted to the higher ammonia levels in the caudal blood. In blood drawn
from the ventral aorta (pregill), the plasma [ammonia] is 1.5-2.0 times
that in dorsal aortic (postgill) blood (Cameron and Heisler, 1983;
Wright and Wood, 1985); the difference representing ammonia ex-
creted at the gills.
    In most fish, plasma total [ammonia] ranges from 0.1 to 0.8 mM
(Watts and Watts, 1974; Wright and Wood, 1985; Perry and Vermette,
1987; Perlman and Goldstein, 1988; MacKenzie and Randall, 1990); fed
fish typically have higher levels than do starved ones. Plasma [ammonia]
increases in response to exhaustive exercise (e.g., Turner et al.,
 1983a,b),air exposure (van Waarde, 1983; Walsh et al., 1990), increases
in temperature and water [ammonia] (Thurston et al., 1984),and expo-
sure to alkaline (pH -9:5; Wright and Wood, 1985) and acidic (pH
-4.5; McDonald, 1983a; H6be et al., 1984) environments (see Randall
and Wright, 1987, for a review). Interestingly, in the lemon sole (Par-
ophrys uetulus), hypercapnia results in a reduction in plasma ammo-
2. CHEMICAL   PROPERTIES OF THE BLOOD                                  85

nia which may be due, in part, to a COZ-mediated suppression of
metabolism (Wright et al., 1988).

C. Urea and Uric Acid
    Elasmobranchs, holocephalans (chimaeras), and the coelocanth
(Latimeria),  whose milieu interieur are isotonic or slightly hypertonic
to seawater, are distinguished by the high levels (250-400 mM) of urea
in their plasma. The source of this urea, which serves as the animal’s
most abundant organic osmolyte, is de novo synthesis by the ornithine-
urea cycle in the liver. In some euryhaline elasmobranchs (e.g., Scyl-
iorhinus africanus; Haywood, 1973; Raja elanteria; Watts and Watts,
1974; Perlman and Goldstein, 1988),plasma urea levels decline when
animals are placed in dilute seawater and, conversely, increase when
fish are exposed to concentrated seawater. In others (e.g., lemon shark,
Negaprion brevirostus; Watts and Watts, 1974), plasma [urea] is con-
stant in the face of changing salinities.
    A reduction in food intake over a period of several weeks can
reduce plasma [urea] in elasmobranchs and, hence, their ability to
hyperosmoregulate. At any given salinity, plasma [urea] was consis-
tently greater in well-fed sharks ( S . africanus fed twice weekly; Hay-
wood, 1973) than in poorly fed (once a month) sharks. Plasma [urea]
increased almost immediately on refeeding. The dependence of
plasma [urea] on food availability may explain, to some extent, the
variation in reported plasma urea levels for marine elasmobranchs.
    Urea is also present in the plasma of marine and freshwater teleosts,
although at much lower levels (1-10 mM), and plays an insignificant
role in osmoregulation. In most teleosts, urea is derived from the
degradation of purines via uric acid, the hydrolysis of arginine, or both.
In some species, e.g., Gulf and oyster toadfish (0.    beta, 0. tau; Walsh
et al., 1990); tilapia [Oreochromis alkalicus grahami; Wood et al.,
1989) and several air-breathing fish from the Indian subcontinent (Het-
eropneustes fossilis, Clarias batrachus, Anabas testuidneus, Amphip-
nous cuchia; Saha and Ratha, 1989), urea is synthesized via the
ornithine-urea cycle. Plasma urea levels in these ureogenic teleosts,
even when air exposed, are generally not much greater than those in
ammoniotelic species. In the marine, ureogenic 0. beta, plasma [urea]
is about 9-10 mM (Walsh et al., 1990) compared to 4-5 mM in the
freshwater ammoniotelic rainbow trout (Wood et al., 1989). Unlike air
breathing teleosts, urea in the African lungfish builds up to high levels
during estivation (200 mM after 13 months). When aquatic, lungfish
86                                 D. G. MCDONALD AND C. L. MILLIGAN

plasma [urea] (1-7 mM) is not unlike that in teleosts (DeLaney et al.,
   For a complete discussion of the evolutionary and physiological
significance of urea synthesis in fish, see Mommsen and Walsh (1989,
    Uric acid is formed by the degradation of purine nucleotides and
protein catabolism via purines, primarily in the liver and white mus-
cle. Uric acid is generally converted to urea for excretion, so blood
levels are typically low. In rainbow trout, plasma uric acid ranges from
40 to 100 pM (Hille, 1982). Plasma uric acid is not commonly measured,
so there is little data available on phylogenetic trends, effects of star-
vation, stress, or seasonal variations.

D. Trimethylamine Oxide
    Trimethylamine oxide [ (CH3)3N-O; TMAO], like urea, occurs in
high concentrations in marine elasmobranchs (22-120 mM; Griffith,
1981)and the coelocanth (Griffith et al., 1974) and contributes substan-
tially to plasma osmotic pressure. Unlike urea, however, plasma
TMAO levels are quite variable both among and within species (Grif-
fith, 1981).Some of this variability might be due to differences in diet,
since diet is thought to be the principle source of TMAO (Watts and
Watts, 1974). Levels of TMAO in the freshwater elasmobranch, Po-
tamotrygon sp., are considerably lower (often <1 mM) than in their
marine counterparts. Trimethylamine oxide in holocephalans are also
quite low, averaging 5 mM (Griffith et al., 1974).
    Trimethylamine oxide is normally considered negligible in the
blood of teleosts (<1 mM; Griffith, 1981); however, several marine
teleosts are reported to have substantial plasma TMAO levels. In sev-
eral shallow water marine teleosts (e.g., Scomber scomber, Morone
saxatilus, 0.tau; Griffith, 1981), plasma TMAO levels range from 1 to
38 m M . In deeper water fish, which are often moribund when sampled,
plasma TMAO levels are considerably higher (often 60-90 mM).

E. Creatine and Creatinine
    Creatine is an amino acid that is an end product of the metabolism
of glycine, arginine, and methionine and is found primarily in the white
muscle. Creatine is a precursor for the high energy phosphate, phos-
phocreatine. The source of creatine in fish is unclear. The necessary
enzymes for de novo synthesis of creatine have been found in carp, but
are absent, or undetectable, in the Pacific hagfish (Eptatertus),a shark
(Prionace sp.), ray (Urolophus sp.), and a teleost, the buffalo fish (Mega-
2.   CHEMICAL PROPERTIES OF THE BLOOD                                  87

stomatobus SP.) (Van Pilsum et al., 1972). Creatine is present at fairly
high concentrations in seawater (-0.1 mM), but only Eptatertus is
known to obtain creatine via absorption from the water (Van Pilsum et
al., 1972). In other fish, it has been concluded that creatine is supplied
in sufficient quantities in the diet (Danulat and Hochachka, 1989).
Creatine in the blood, therefore, is in a dynamic state, representing a
balance between that absorbed via the gut and that released from
muscle. Whole blood creatine levels average 270 pM in starry flounder
(Danulat and Hochachka, 1989) and are unaffected b y 6 wk of star-
vation. Whole blood creatine in trout averages 1 mM and is about half
the level found in extracts of red cells (2.5 mM; Danulat, unpublished,
cited in Danulat and Hochachka, 1989).Danulat and Hochachka (1989)
have suggested that the differences between creatine levels in floun-
der and trout may reflect differences in swimming activity or nutri-
tional status.
    Creatinine is formed by spontaneous (i.e., nonenzymatic) cycli-
zation of creatine. Its levels in plasma are typically low, 10-80 pM
(Sandnes et al., 1988), and appear to be unaffected by stress (Wells et
al., 1986) or, in the African lungfish, estivation (DeLaney et al., 1977).
Creatinine is not metabolized further and is excreted by the kidneys.


    The plasma proteins in fish have not been studied in detail in any
species, and other than electrophoretic mobility patterns there is little
information about many of the plasma proteins. The nomenclature for
fish plasma proteins is adopted from that used for mammals, which is
based on molecular weight and pattern of electrophoretic mobility.
Among the plasma proteins identified, most fish plasma contains
albumin-like proteins, a number of &"-binding        proteins (e.g., ceru-
loplasmin, vitellogenin), blood clotting proteins (e.g., fibrinogen and
prothrombin), metal-binding proteins (e.g., transferrin), immunoglob-
ulins, lipoproteins (e.g., high density lipoproteins [HDL], low density
lipoproteins [LDL]; see Section VII,B), and hormone-binding proteins
(Ts-binding protein; steroid-binding proteins). In Arctic and Antarctic
species, there are specialized antifreeze proteins and glycoproteins.
Aside from albumin or para-albumin, there is very little information on
the phylogenetic distribution of these proteins. Most of the informa-
tion about these proteins is descriptive (e.g., molecular weights, bind-
ing characteristics, electrophoretic mobility), with very little informa-
tion available as to the amounts in plasma. For a thorough discussion of
the evolution of plasma proteins, see Doolittle (1984, 1987).
88                                 D. G . MCDONALD AND C. L. MILLIGAN

A. Total Plasma Protein
    Total plasma protein can be determined relatively simply from the
refractive index of plasma. Since proteins form approximately 80% of
the plasma solutes, it is possible to use the measurement of refractive
index to show changes in total plasma [protein] (Gowenlock, 1988).
There are, however, differences between total protein estimated by
colorimetric techniques (e.g., Lowry method, Biuret reaction; see
Gowenlock, 1988 for a detailed discussion of these methods) and by
refractometry. Alexander and Ingram (1980) and more recently Hunn
and Greer (1990) reported that the plasma [protein] estimated by the
colorimetric methods was lower than that estimated by refractometry.
Nonetheless, there were good correlations between refractometry esti-
mates and colorimetric estimates. This led Alexander and Ingram
(1980)and Hunn and Greer (1990)to suggest that, because of its ease of
use and small sample volume required, refractometry is the method of
choice for estimating plasma [protein], providing the refractometer
measurements are calibrated with a colorimetric method of analysis.
    The total plasma [protein] in fish ranges from 2 to 8 g dl-' (Fletcher,
 1975; Fellows et al., 1980; Miller et al., 1983; Sandnes et al., 1988;
Hunn and Greer, 1990) and appears to be fairly constant within and
among species. Total plasma [protein] is altered mainly by changes in
plasma volume; an increase is caused by a shift of fluid from the plasma
to the intracellular compartment, and a decrease can be caused by
hydration of the plasma. Fluid shifts out of the plasma are caused by an
osmotic imbalance between the extracellular and intracellular com-
partments, and any stress that induces such an imbalance can lead to
increases in plasma [protein] (see Olson, Chapter 3, for a discussion on
the regulation of fluid volume). For example, total plasma [protein]
increases in rainbow trout in response to strenuous exercise (Milligan
and Wood, 1986) and exposure to low environmental pH (Milligan and
Wood, 1982).
     Reductions in plasma [protein] are less common, and are generally
associated with prolonged starvation (Love, 1980) and severe stress
 (Stevens, 1968). Several studies have examined seasonal variations in
 total plasma [protein] with often conflicting results. In rainbow trout,
 Haider (1970) reported peak plasma levels in midwinter, whereas
 Schlotfeldt (1975) found maximal levels at the end of the summer.

B. Albumin
     Albumin has a molecular mass of about 68-70 kDa. It reversibly
2.   CHEMICAL PROPERTIES OF THE BLOOD                                 89

binds with fatty acids, bilirubin, and many other substances and serves
as a major transport protein in the blood. In fish, albumin-like proteins
have been identified in lamprey, hagfish, and several teleost species.
There is, however, some question as to the presence of albumin in
elasmobranchs (Fellows et al., 1980; Fellows and Hird, 1981, 1982).
    The bromocresol green method (see Gowenlock, 1988 for details) is
routinely used to measure plasma albumin. In teleosts, plasma al-
bumin concentration ranges from 1.0 to 2.4 g dl-' and constitutes 25 to
50% of the total protein (Fellows et al., 1980; Miller et al., 1983;
Sandnes et al., 1988).Albumin levels tend to be low in the lamprey,
representing less than 15% of the total protein (Fellows and Hird,
1981). Although the presence of albumin in elasmobranchs has been
questioned (see following discussion), plasma albumin has been de-
tected by the bromocresol green method in several elasmobranchs but
at very low levels: 0.5-0.7 g dl-l (Fellows et al., 1980).When albumin
levels are low, the bromocresol green method tends to overestimate
[albumin] because of the binding of bromocresol green to lipoproteins
and transferrin (Gowenlock, 1988).
    The presence of albumin in fish plasma has been confirmed on the
basis of electrophoretic mobility, fatty acid-binding properties, and
molecular mass (Fellows and Hird, 1981; Davidson et al., 1988). Ag-
nathid, elasmobranch, and teleost plasma all contain a protein with a
molecular mass similar to that of human albumin (68-70 kDa; Fellows
and Hird, 1981; Davidson et al., 1988).In the agnathids and teleosts,
this protein reversibly binds fatty acids (palmitate and oelate; Fellows
and Hird, 1981; Davidson e t al., 1988) and bilirubin (Fellows and
Hird, 1982). In elasmobranchs, however, this albumin-like protein
does not bind fatty acids but does bind bilirubin (Fellows and Hird,
1982). The significance of the absence of a fatty acid-binding protein in
elasmobranchs is not clear but is consistent with the low levels of
plasma free fatty acids found in these species.

C. Immunoglobulins
   Next to albumin, the immunoglobulins are the most abundant
plasma proteins, representing anywhere from 2 to 15%of total protein
(Kobayashi et al., 1982; Olsen and J@rgesen, 1986; Haversein et al.,
1988; Klesius, 1990). Immunoglobulin levels are quite variable in fish
(44-650 mg ml-') and tend to be higher in wild-caught fish (Klesius,
1990). These higher levels in fish from natural environments most
likely reflect exposure to many different antigens. In channel catfish
(Klesius, 1990)and rainbow trout (Olsen and Jqjrgensen, 1986), plasma
90                                 D. G. MCDONALD A N D C. L. MILLIGAN

immunoglobulin levels were independent of temperature. Immuno-
globulin levels significantly increased with increasing size and age of
catfish (Klesius, 1990),which is similar to the pattern found for immu-
noglobulins from several other species, including humans.

D. Hormone-Binding Proteins
   T h e steroid and thyroid hormones are transported in the plasma
bound to proteins. It has been argued that during blood transit through
the capillary, rapid dissociation of the hormone from a large protein-
bound reserve helps keep the free hormone level low and thereby
facilitates tissue uptake (Ekins, 1986).
    Steroid hormones bind to three classes of proteins:
     1. Albumin binds most steroids with low affinity and very high
     2. Corticosteroid-binding globulin binds glucocorticosteroids and
         progesterone with high affinity but binds aldosterone and sex
         steroids with low affinity.
     3 . Sex hormone-binding globulin binds sex steroids with a speci-
         ficity and affinity that is species specific (Wingfield, 1988).
    Two steroid-binding globulins have been demonstrated with elec-
trophoretic techniques in adult sea lamprey (Boffa et al., 1972); one
preferentially binds progesterone and the other has a high affinity for
estradiol. However, these proteins also show a high cross-reactivity
with testosterone and corticosterone, suggesting a low specificity for
steroid binding. In contrast, only very low affinity and nonspecific
binding of sex steroids is observed in the blood of the Pacific hagfish
and the anadromous lamprey (L. tridentata) (Wingfield, 1988),which
may be linked to generally low levels of steroids found in agnathan
    Elasmobranchs appear to have only a single protein system that
binds steroids with low specificity but high affinity. Although this
protein binds both cortisol and corticosteroid, the major corticosteroid
in elasmobranchs, la-hydroxycorticosterone, is bound with low affin-
ity (Martin, 1975).Skate (Raja rudiata) plasma possesses a protein that
binds sex steroids with a rather high affinity but low capacity (Wing-
field, 1988).
    I n teleosts, steroid hormones appear to bind to a single protein
system showing specificity and high affinity for CIS and C19 steroids
(e.g., testosterone, estrogen) and low affinity for Czl steroids (e.g.,
progesterone, corticosteroids). In Atlantic salmon and Atlantic cod a
2.   CHEMICAL PROPERTIES OF THE BLOOD                                91

second protein has been found that binds cortisol with a very high
affinity (Freeman and Idler, 1971).
    In general, the steroid hormone-binding proteins in fish tend to
have a higher affinity and higher capacity than those in mammals.
Wingfield (1988) has argued that this may explain how fish are able to
maintain the very high blood steroid levels typical of most fish.
    Less is known about the thyroid-binding hormones in fish. It is
known that only a small portion of T3 and T4 (<3%) is present in the
circulation in the free form (Cyr and Eales, 1989). I n brook trout
(Faulkner and Eales, 1973)and rainbow trout, T3 appears to bind to an
albumin-like protein with high affinity. Thyroxine appears to bind to
vitellogenin in several cyprinid species (MacKenzie et al., 1987) but
not in trout. In both salmonids and cyprinids, the level of thyroid
hormone binding protein increases in response to estradiol administra-
tion (MacKenzie et al., 1987; Cyr and Eales, 1989).

E. Antifreeze Proteins
    A number of marine teleosts inhabiting polar and subpolar regions
produce antifreeze proteins (AFP) or antifreeze glycoproteins (AFGP).
These proteins act in a noncolligative fashion; they lower the freezing
point of plasma without affecting the melting point. This property of
antifreeze proteins is exploited and used to estimate AFP activity,
which is expressed in terms of thermal hysteresis: The difference be-
tween the melting and freezing points. The greater the difference, the
greater the antifreeze protein activity (DeVries, 1983; Fletcher et al.,
   Antifreeze glycoproteins are found in the Antarctic Nototheniidae
and in a number of cod species (e.g., Gadus ogac, G. morhua, Bareo-
gadus saida, Eleginus garcills, Micragadus tomcod; DeVries, 1983)
from the Arctic and North Atlantic. The AFGP are well conserved and
consist of a repeating structure of alanine-alanine-threonine with a
disaccharide attachment to the threonine side chain. It is estimated
that AFGP constitutes about 3.5% of the total plasma protein in noto-
theniids (see DeVries, 1983 for a complete review).
   Antifreeze proteins have been characterized from a number of spe-
cies found in the North Atlantic and three distinct types are recognized
on the basis of amino acid composition and secondary structure: AFP I,
AFP 11, and A F P 111. Unlike AFGP, AFP contain no sugar moieties.
Antifreeze protein I are a-helical and rich in alanine and have been
found in winter (Pseudopleuronectesamricanus)and yellow tail flounder
(Limanda feruginea) and in shorthorn (Myaxocephalus scorpius) and
92                                 D. G . MCDONALD AND C. L. MILLIGAN

grubby sculpins ( M . awnaeus) (Davies et al., 1988).Antifreeze protein
I1 is alanine poor, half-cysteine rich, and has been found in sea raven
(Hemitripterus americanus), smelt (Osmerus mordax), and Atlantic
herring (Clupea harengus harengus) (Davies et al., 1988; Ewart and
Fletcher, 1990).Antifreeze protein 111is different from the other two in
that no amino acid predominates and, to date, it has been found only in
ocean pout (Macrozoarces americanus) (Davies et al., 1988).
    There is considerable species variation in the regulation of expres-
sion of AFP and AFGP. Winter flounder and shorthorn sculpin pro-
duce AFP immediately prior to environmental freezing, whereas At-
lantic cod produce AFGP only in response to freezing (Davies et al.,
1988). Ocean pout and the Antarctic nototheniids have high levels of
AFP and AFGP, respectively, year round (DeVries, 1983; Davies et al.,
 1988). From studies with winter flounder, it appears that the reduction
in day length is the primary environmental cue responsible for initia-
tion of AFP synthesis. Water temperature appears to be an important
cue for the clearance of AFP from the plasma in spring (see Davies et
al., 1988 for a review).

F. Plasma Enzymes
     Plasma enzymes are broadly classified into two categories:
(a)plasma specific enzymes, having function and purpose in the blood;
and (b) plasma nonspecific enzymes that are derived from moribund
cells (Gowenlock, 1988).Measurement of the activities of the plasma
nonspecific enzymes has diagnostic potential in fish toxicology and
pathology because enzyme activities can often be related to cell dam-
age in specific organs. For example, the liver is rich in glutamic-
oxalacetic transaminase (GOT) and glutamic-pyruvate transaminase
(GPT), and changes in plasma levels of these enzymes may be indica-
tive of liver dysfunction. In trout treated with carbon tetrachloride
(CC14), a known hepatotoxicant, plasma GOT and GPT levels in-
creased significantly (Racicot et al., 1975; see Table 111).
    A few studies have attempted to use plasma enzyme values for the
assessment of general fish health and these are summarized in Table
111. However, there are few data available for fish other than rainbow
trout and, more importantly, information about the impact of nonpatho-
logical factors (e.g., salinity, temperature, reproductive state) is lack-
ing. Furthermore, when available the effects are often complex. For
example, in trout, acclimation temperature has a complex effect on the
activities of plasma creatine phosphokinase (CPK) and alkaline phos-
phatase (AlkPase). Creatine phosphokinase activity in trout plasma
                                                             Table I11
                              The Activity of Nonspecific Plasma Enzymes in Different Fish Species

                                                                          Enzyme activity
                                                                                                                               ~    ~-

                               LDH"          GOT     GPT        AlkPase         CPK         GDH    HBDH        AchE                LAP
                               Liver                             Liver
                              Kidney                            Kidney
                               Heart                         Red blood cells   Muscle    Liver             Red blood cells      Small
     Tissue specificity:   Red blood cells   Heart   Liver   Small intestine   Heart     Kidney    Liver    Nerve tissue      intestine


Rainbow trout
    Fedb                        1111          244     7.2         138                       68.3     373                           66
    Starvedb                     730          157     4.9          68                       45.7     230                           66
    Fresh waterb                 610          171     8.0         136                       40       198                           74
    10% salinityb                49 1         192     7.8         191                       49       185                           88
    20% salinityb                579          267    17.9         111                       38       249                           62
    3.5"C"                       150          125     7.0          72              365      35        50                           50
    10°C"                        825          275    11.5         140            1,032      55       200                           51
    15°C"                        910          310    15.2         122              382      60       225                           80
    Electroshockedd              204                                             1,620                                             10.2
    Seinedd                      222                                               630                                             10.8
    Angledd                      213                                               920                                             11.3
    Caught by dip net?           198                                               790                                             11.9
    Controle                     86 1         196    26.6           4.2                      0.6
    CC14 injectione             1793          347    47             3.6                      1.2
    Aeromonas infection;        1560          167    12.9           3.6          2,940      16.2                             continues
    early stage"
                                                        Table 111 continued

                                                                           Enzyme activity
                                                                               UI - 1

                                LDH"          GOT     GPT        AlkPase          CPK        GDH    HBDH        AchE            LAP

                                Liver                             Liver
                               Kidney                             Kidney
                                Heart                         Red blood cells   Muscle     Liver            Ked blood cells     Small
      Tissue specificity:   Red blood cells   Heart   Liver   Small intestine   Heart     Kidney    Liver    Nerve tissue     intestine


     Aeromonas infection;        2468          450    70.7                       10,147      32.4
     late stage"
     Caudal puncture'            4952          187    85                          2,940              216
$ Cardiac puncturef               882          170    51             7.2            678      33       50          12
 Atlantic salmong
   Salmo salar                                 278     6           853
 Pink salmon"
     Nonspawning                  518          309                                1,091              965
     Prespawning                 2427          500                                                  3387
     Spawning                    2463          797                                2,936             3421
 Pacific herring''
   Culpea pallasii               3358         1778                                2,948             3228
   Ophiodon                        107          28                                                    141
 Red rock fish"
   Sebastodes                       18         389                                                      0
  Squolus acarithias                    149           128                                   111                   0
    Male: prespawning                                 243    18.6                         5,767      60
    Male: spawning                                    702    78                           5,064     195
    Male: postspawning                                756    27                           2,802      94
    Female: prespawning                               248    17                           5,130     178
    Female: spawning                                  526    62                           6,228     236
    Female: postspawning                              455    29                          15,600     164

     'I Total IJDH activity is reported. LDH i s a ubiquitous enzyme and most vcrtcbrates show five isozymes, which can be broadly1 classified into

 three functional groups: heart, spleen, red cell; gill, kidney and liver; white muscle. These groups have distinctive biochemical characteristics
 that can be exploited to determine the contribution of various tissues to circulating LDH activity (for details, see Gaudet et al., 1975).
        Sauer and Haider (1979). Fish (100-150 g) held at 15°C; fed group were fed 2 g fish-' day-'; food was withheld from starved group for 20
 days. Blood collected via cardiac puncture. Fish held at 10 and 20%0salinity for 20 days.
ED      Sauer and Haider (1977). Fish (180-222 g) held at various temperatures for 20 days and fed daily. Blood collected via cardiac puncture.
        Bouck et al. (1978). Fish held at 12°C. Blood collected from anesthetized fish via cardiac puncture. Angled fish struggled for 1 min prior to
 sampling. Fish were electroshocked with 75 V until stunned. Seined fish were netted 2 at a time. Dip netted fish were netted individually,
      ' Racicot et a / . (1975). Fish (150-300 g) held at 15°C. Blood sampled via caudal puncture. CC14, 0.4 mg/kg administered in mineral oil via
 intraperitoneal injection. Aermonas infection: Early stage, fish were starting to lose scales and white spots appeared in caudal region; late stage:
 dorsal and ventral sides of the peduncle were badly affected.
     f Gaudet et al. (1975). Fish (100-250 g) held at 15°C.
        Sandnes et a1. (1988). Fish (1-1.8 kg) held at ambient temperature (2-8°C) at a salinity of 32%. Blood collected from anesthetized fish from
 the ductus cuvieri.
        Marquez (1976). Prespawning fish collected by angling. Spawning and nonspawning fish collected by netting. Lingcod, red rock fish, and
 dogfish were collected via hook and line. Fish placed in holding tanks 12-24 h prior to blood collection via caudal puncture.
        Hlavovi (1989). Fish (220-350 g) were captured via electroshocking, and blood was collected 2-4 min after capture via cardiac puncture.
      Abbreviations: LDH: lactate dehydrogenase (E.C. glutamic-oxalacetic transaminase also known as aspartate aminotransferase
 (E.C.; GPT: glutamic-pyruvate transaminase, also known as alanine aminotransferase (E.C.; AIkP: alkaline phosphatase (E.C.; CPK: creatine phosphokinase (E.C.; GDH: glutamate dehydrogenase (E.C.; HBDH: a-hydroxybutyric dehydrogenase
 (E.C.;AchE: acetylcholinesterase (E.C.; LAP: leucine amino peptidase (E.C.
      AlkP Activity expressed in phosphatase unit: Activity which liberates 1pmol ofp-nitrophenol per 30 min at 25°C. AchE activity expressed in
 Happaport units: Activity that hydrolyzes 1pmol of acetylcholine in 39 min at 25°C. All other enzyme activities expressed in international units:
 1 U = 1pmol of substrate used per minute at 25°C.
96                                 D. G. MCDONALD AND C. L. MILLIGAN

increases when acclimation temperature is increased from 3.5" to 10°C
but then declines at 15°C. Alkaline phosphatase activity decreases
when acclimation temperature is increased from 6"to 19°C (Table 111).
    There are also methodological considerations that can influence the
interpretation of plasma enzyme activities. In rainbow trout, the
method of blood sampling influences the activity of at least two
plasma-nonspecific enzymes: lactate dehydrogenase (LDH) and CPK.
These enzymes, which are abundant in skeletal muscle, are greater in
blood sampled via caudal puncture than via cardiac puncture (Table
111).When catheterization is not possible, Gaudet et aZ. (1975) recom-
mend withdrawing blood via cardiac puncture for enzyme analysis.
    In addition, the methods commercially available for measurement
of plasma enzymes have been developed for mammalian plasma. For
some enzymes (e.g., LDH), mammalian assay systems are applicable to
fish plasma (D'Apollonia and Anderson, 1980).However, for analysis
of enzymes of ammonia metabolism, GTP, GOT, and glutamate dehy-
drogenase (GDH) assay kits designed for analysis of mammalian
plasma may not be directly applicable. D' Apollonia and Anderson
(1980) have indicated that unlike the mammalian system, GDH inter-
feres with the transaminase assays in rainbow trout and if the level of
GDH activity is high, this may lead to overestimation of GPT and GOT
activity. Furthermore, the higher ammonia levels found in fish plasma
relative to mammals can lead to higher transaminase activities. This is
particularly important when attempting to interpret changes in plasma
GOT and GPT activity: An increase in activity may be indicative of
liver dysfunction or reflect changes in plasma [ammonia] (D'Apollonia
and Anderson, 1980).


A. Total Lipids
   Lipids are a heterogeneous class of compounds that are grouped
together by virtue of their solubility in organic solvents (e.g., chloro-
form, acetone, ether) and their relative insolubility in water. The lipids
include fatty acids and some of their esters (e.g., wax esters), choles-
terol and its esters, triglycerides, and phospholipids (Sheridan, 1989).
In fish, triglycerides and phospholipids are the most abundant lipid
    Total plasma lipid levels in most species range from 1200 to
3000 mg dl-' (Dindo and MacGregor 111,1981; Freemont et d.,          1981;
2.   CHEMICAL PROPERTIES O F THE BLOOD                                   97
Sheridan, 1989) and are influenced by many factors, including diet,
stress, and reproductive state. In general, the lowest plasma lipid
levels are associated with spawning activity and the highest are ob-
served within hours after a meal (Sheridan, 1989). Stress in the form of
exhaustive exercise can also deplete plasma lipids; immediately fol-
lowing 5 min of forced swimming, plasma lipids in rainbow trout
declined from 3400 to 1000 mg dl-' (Girard and Milligan, 1992).

B. Lipoproteins
    Plasma lipoproteins facilitate the transport of the otherwise insolu-
ble lipids in plasma from sites of storage (e.g., liver and muscle) to sites
of utilization (Henderson and Torcher, 1987). Lipoproteins are thought
to consist of a core of the more hydrophobic lipids (e.g., triglycerides,
cholesterol esters) surrounded by phospholipids, free cholesterol, and
proteins (Darnel1 et al., 1990).
    Lipoproteins in fish are classified based on their densities accord-
ing to the system devised for mammals: chylomicrons, very low den-
sity lipoproteins (VLDL), LDL, and HDL. The lipid composition of
each density class of plasma lipoproteins is comparable to those in
mammals; chylomicrons, when found, are rich in triglycerides and are
involved in the absorption of lipids from the gut; VLDL is rich in
triglycerides; LDL is rich in cholesterol; and HDL is rich in choles-
terol and phospholipids (McKay et al., 1985). For a comprehensive
review of the subject of plasma lipoproteins in fish and a summary of
plasma levels for numerous fish species, see Babin and Vernier (1989).
    Although plasma lipoprotein levels vary both within and among
species and are affected by such things as nutritional and reproductive
status, a few generalizations can be made. Relative to mammals, fish,
with the exception of elasmobranchs and garpike, are considered hy-
perlipidemic. The distribution of lipids within each class of lipo-
protein shows a strong phylogenetic trend.

   a. Agnathids. In the agnathids, no one lipoprotein predominates;
VLDL, LDL, and HDL are present in more or less equal amounts. In
the lamprey (Mordacia rnordax), plasma levels of VLDL, LDL, and
HDL during a spawning migration are 426, 664, and 507 mg dl-',
respectively (Fellows and McLean, 1982). The LDL fraction is rich in
cholesterol, while HDL is associated mainly with phospholipids.
98                                 13. G. MCDONALD AND C. L. MILLIGAN

While the plasma lipoprotein levels are similar in the Atlantic hagfish
(Myxine glutinosa), they contain very little cholesterol; rather, all
classes of lipoproteins are rich in triglycerides and phospholipids
(Mills and Taylaur, 1973).

   b. Chondrichthyes. Very low density lipoproteins and LDL are the
predominant lipoproteins in elasmobranchs, with levels ranging from
196 to 415 mg dl-' and 120 to 230 mg dl-', respectively, while HDL
levels are quite low; 20-40 mg dl-' (Babin and Vernier, 1989). The
lipoproteins of elasmobranchs are distinct from those of other fish in
that they are high in squalene and alkyldiacylglycerols. These two
constituents are poorly metabolized, have low densities, and tend to
concentrate in the liver, suggesting that they play a role in hydrostatic
equilibrium in fish lacking a swimbladder (Sargent, 1976).

    c. Osteichthyes. High density lipoproteins tend to dominate the
lipoprotein classes in the bony fish, accounting for as much as 50% of
the total lipoprotein. These levels are not only variable among species
but also within species, affected primarily by nutritional and reproduc-
tive status. Most of the available data are for the salmonids. High
density lipoprotein levels range from 238 mg dl-' in juvenile sockeye
salmon to as high as 3300 mg dl-' in prespawning pink salmon (Babin
and Vernier, 1989). The high HDL levels are associated with high
cholesterol levels.
    Very low density lipoprotein and LDL levels in plasma tend to
be lower (167-650 mg dl-' and 225-1189 mg dl-'; Babin and Ver-
nier, 1989) than HDL and are very low or absent from the plasma of
spawning salmonids. Also, prolonged starvation (8 weeks) in rainbow
trout (Black and Skinner, 1986)and channel catfish (McKay et al., 1985)
results in approximately 60-70% and 40-50% reductions in plasma
levels of VLDL and LDL, respectively, but HDL levels were unaf-
    Although the datum is limited to a single observation on plasma
obtained from a moribund specimen, the coelacanth (Latimeria cha-
Zumnae) appears to differ from other bony fish. In coelocanth plasma,
VLDL predominates (1105 mg dl-l) with LDL and HDL levels con-
siderably lower (194 and 127 mg dl-', respectively; Mills and Taylaur,
1973). Also, like the hagfish, the coelocanth lipoproteins are choles-
terol poor.

     One further class of lipoproteins, vitellogenin, is found in large
2.   CHEMICAL PROPERTIES OF THE BLOOD                                   99

quantities in mature oviparous female fish (Dye et al., 1986; Hender-
son and Torcher, 1987). Vitellogenin is a higher density lipoprotein
than H D L and contains approximately 80% protein and 20% lipid,
most of which is phospholipid (Henderson and Torcher, 1987).Vitello-
genesis has been most extensively studied in the salmonids, but simi-
lar trends are observed in other species. Although vitellogenin is found
in both males and females, the plasma level in females is orders
of magnitude greater than in males. For example, in male rainbow
trout (mature and immature), plasma vitellogenin ranges from 0.3 to
0.8 p g dl-', whereas plasma vitellogenin ranges from 115 fig dl-' in
immature females to as high as 5000 mg dl-' in ovulating females
(Scott and Sumpter, 1983; Copeland et al., 1986), representing >50%
of the total plasma protein. After ovulation, vitellogenin levels drop to
around 50-100 mg dl-' (Scott and Sumpter, 1983).

    Chylomicrons are microscopically visible, triglyceride-rich lipo-
protein particles formed in the intestinal mucosa following the absorp-
tion of' dietary fatty acids. As they circulate in the blood, triglycerides
are stripped off by the action of lipoprotein lipases present in plasma
and several tissues. The remaining protein particles are thought to be
metabolized in the liver (Sheridan, 1988).
    Chylomicron particles have been observed in plasma from a num-
ber of elasmobranchs (Mills et al., 1977) and teleosts (Sheridan, 1988).
I n rainbow trout, the particles appear as early as 2-4 h after feeding
and are about 80% triglyceride, with minor amounts of cholesterol,
cholesterol esters, and phospholipids (Sheridan, 1988). The size of the
particles, which ranges from 1000 to 6500 A, is dependent on the lipid
content of the diet; the higher the lipid content, the larger the chylomi-
cron particle size (Sheridan, 1988).

C. Cholesterol
   Most fish, with the exception of elasmobranchs and garpike, are
hypercholesterolemic relative to mammals. In the agnathids, approxi-
mately 50% of the cholesterol is carried by LDL, whereas in elasmo-
branchs and teleosts, 60-90% of the cholesterol is carried by HDL
(Babin and Vernier, 1989).
   Plasma cholesterol levels in agnathids are high (e.g., 400 mg dl-I in
the Atlantic hagfish), though only about 50% of it is associated with
100                                D. G. MCDONALD AND C. L. MILLIGAN

lipoproteins (Larsson and Fange, 1977). As with most fish, the highest
cholesterol levels are associated with spawning and the lowest seen in
food-deprived animals (Babin and Vernier, 1989).
    Elasmobranchs have substantially lower plasma cholesterol
levels than do agnathids or teleosts; typical values range from 86 to
200 mg dl-' (Larsson and Fange, 1977; Babin and Vernier, 1989;
Garcia-Garrido et al., 1990). Plasma cholesterol levels vary during
sexual maturation, and at least in the spotted dogfish (Garcia-Garrido et
al., 1990), there are pronounced sex differences. Females had con-
sistently lower levels of plasma cholesterol (82 mg dl-') than males
(103rng dl-') and, in fact, the lowest were seen in egg-carrying females
(Garcia-Garrido et al., 1990). The highest cholesterol levels (112 mg
dl-') were seen in males during spermatogenesis. There appeared to
be no effect of 3 months of starvation on cholesterol levels in the
spotted dogfish (Garcia-Garrido et al., 1990).
    Plasma cholesterol levels show considerable intra- as well as inter-
specific variability. For example, serum cholesterol in Pacific salmon
(Oncorhynchus tshawytscha) varies from 300 to 1470 mg dl-'
(Robertson et al., 1961) and in the Atlantic cod, values range from 399
to 1598 mg dl-' (Larsson and Fange, 1977). These extreme individual
variations are no doubt linked to differences in diet, activity, and
sexual development. Generally, the highest plasma cholesterol levels
tend to be seen in both male and female prespawning fish and appear
to be linked to the time when fish are actively feeding. Upon spawning,
plasma cholesterol tends to drop, though more so in females than
in males. In female migrating Pacific salmon, plasma cholesterol
reaches highs of 635 mg dl-', but in spawning fish, cholesterol levels
drop precipitously, to 126 mg dl-' (Robertson et al., 1961).Although
this trend is observed in most species examined, the decline in plasma
cholesterol is more severe in fish that undergo an active migration. For
example, following a 1000-km upstream migration, plasma cholesterol
levels in sockeye salmon dropped from 570 to 294 mg dl-' in males and
from 585 to 202 mg dl-' in females (Idler and Tsujuki, 1958). In
contrast, in Atlantic salmon captured at sea and transported to the
spawning site, plasma cholesterol levels did not drop with spawning,
but remained at prespawning levels, 576 mg dl-' (Farrell and Munt,
1983). Furthermore, in Lake Erie coho salmon that undergo only short
migrations, plasma cholesterol is 355 mg dl-' in spawning fish com-
2.   CHEMICAL PROPERTIES OF THE BLOOD                                101

pared to 540 mg dl-l in prespawning fish (Leatherland and Sonstegard,
    Although not as extensively studied, nonsalmonid teleosts tend to
show the same general trends: In prespawning sea bass, plasma cho-
lesterol levels peak at about 900 mg dl-' and fall to about 350 mg dl-'
in spawning fish (FernBndez et al., 1989). Similarly in plaice (White et
al., 1986) and the stripped mullet (Mugil cephalus; Dindo and Mac-
Gregor III,1981), peak plasma cholesterol levels (225 and 300 mg dl-',
respectively) were seen in the summer months when fish were actively
feeding and lowest levels (150 mg d1-l) were seen when fish were
    Whether or not the changes in plasma cholesterol associated with
spawning are a direct consequence of the physiological changes asso-
ciated with spawning per se is not clear. Many species cease feeding
while spawning, and it may be starvation that reduces plasma choles-
terol levels. Few studies have systematically examined the effect of
starvation on plasma cholesterol. Farrell and Munt (1983) reported that
plasma cholesterol levels declined only slightly in immature Atlantic
salmon starved for 6 months (454 mg dl-' in starved, immature females
versus 572 mg dl-' in fed, mature females). However, in salmon that
had spawned and were starved for 11 months, plasma cholesterol
levels were substantially reduced (243 mg d1-l; Farrell and Munt
[ 19831). These observations suggest that spawning and the activity
associated with it may be the more important determinant of plasma
cholesterol levels.

D. Nonesterified Fatty Acids
    Nonesterified fatty acids (NEFA) are the most metabolically active
form of lipid in the blood and are indicative of the extent to which fish
rely on lipid as a fuel. Thus, measurement of plasma levels of NEFA in
response to environmental and physiological changes yields informa-
tion about fuel use.
    A problem that became evident when surveying the literature is
that a variety of methodologies have been used to measure plasma
[NEFA]; colorimetric (Larsson and Fange, 1977; Zammit and News-
holme, 1979; Moon, 1983; Black and Love, 1986), enzymatic (Santulli
et al., 1988), gas-liquid chromatographic (Fellows et al., 1980),and gas
chromatographic (Singer et aZ., 1990; Gong and Farrell, 1990; Singer
and Ballantyne, 1991). Singer et al. (1990) suggested that the
colorimetric, enzymatic, and earlier gas-liquid chromatographic
methods may underestimate plasma [NEFA] in some cases b y as much
102                               D. G. MCDONALD AND C. L. MILLIGAN

as 30%. The colorimetric method tends to miss the longer chain (18-22
carbon) NEFA because they are easily oxidized and antioxidants were
not routinely used. The longer chain NEFA can constitute as much as
30% of the total [NEFA] (Singer et al., 1990). Similarly the enzymatic
assay tends to underestimate the longer chain NEFA because the assay
is designed primarily to measure palmitic acid (16 carbon chain) in
human plasma. In some chromatographic methods, the plasma is ex-
tracted and, as with most extraction procedures, there is a tendency to
lose sample and unless internal standards are added to the plasma, the
loss cannot be quantified.
    Comparisons between studies employing different methodologies
are quite difficult and make not only interspecific comparisons nearly
impossible but even intraspecific comparisons difficult. For example,
in Atlantic cod, [NEFA] ranges from 0.20 mM (Black and Love, 1986),
as determined by one colorimetric method (described by Duncombe,
 1963), to 1.28 mM (Larsson and Fange, 1977) using a different method
(described by Laurel1 and Tibbling, 1967). Thus, there is no advantage
to be gained in citing “typical” values; values reported range from 0.10
to 2.5 mM. For a comprehensive tabulation of plasma [NEFA] in a
variety of fish, see Plisetskaya (1980).
    Despite these methodological problems, plasma [NEFA] deter-
mined using the same methodology can be compared and, in doing so,
 some trends become evident. The lowest [NEFA] are consistently
observed in elasmobranchs and the highest levels in cod (Gadus sp.)
 (Zammit and Newsholm, 1977; Larsson and Fange, 1979; Fellows et
 d., 1980; Black and Love, 1986).Typically, codfish plasma [NEFA] is
 two to three times that of other fish and elasmobranch plasma [NEFA]
is one-tenth that seen in teleosts and cyclostomes. It is not clear why
the cod have such high plasma [NEFA]. However, the striking differ-
ence in plasma [NEFA] between elasmobranchs and teleosts may be a
 direct consequence of the lack of a serum fatty acid-binding protein in
the former (see Section VI1,B). Since the solubility of fatty acids in
plasma is quite low (pmolar range), only very low levels would be
 expected in the absence of a fatty acid-binding protein.
     Many fish species undergo periods of starvation as part of their
 natural life cycles; some in association with migration and spawning
 (e.g., salmonids, eels) and others in association with seasonal low
 temperatures (e.g., eels, bass). Consequently, the response of fish
 plasma [NEFA] to starvation has received wide attention, and the
 variability in response is tremendous. An important variable in deter-
 mining the effect of starvation on plasma [NEFA] is the duration of
 food deprivation. In the American eel, plasma [NEFA] was unaffected
2.   CHEMICAL PROPERTIES OF TffE BLOOD                                103

by 95 days of starvation (Larsson and Lewander, 1973), whereas after 6
months [NEFA] nearly tripled (Moon, 1983). In sea bass, plasma
[NEFAI increased 65% after 40 days of starvation (0.9 to 1.5 mM), but
declined after 150 clays of starvation (0.76-0.44 mM; Zammit and
Newsholme, 1977). In other species, the response of plasma [NEFA]
levels to starvation is more consistent. For example, in the spotted
dogfish starved for 40 and 150 days, plasma [NEFA] decreased by
about 30% (0.13-0.09 mM; Zammit and Newsholme, 1979) and after 85
days of starvation plasma [NEFA] in the Atlantic cod declined from 8.8
to 2.8 mM and to a low of 0.8 mM after 154 days of food deprivation
(Black and Love, 1986).
    Not only are there extreme variations among species in the re-
sponse of plasma [NEFA] to starvation but within species as well. For
example, rainbow trout starved for 1-4 months show either a steady
increase in plasma [NEFA] (Love, 1980)or no change at all (Robinson
and Mead, 1973; Black and Skinner, 1986).Also, in some experiments,
long-term starvation ( 145 days) raised plasma [NEFA] in the European
eel from 0.38 to 0.54 mM, but in other experiments, there was no effect
(Larsson and Lewander, 1973; Dave et al., 1976).
    The interspecific variations of the response of plasma [NEFA] to
starvation may be related to the site of lipid storage. In teleosts with
extrahepatic lipid stores (the so-called "fatty fish," that is, salmonids,
eels), starvation results in a rise in plasma [NEFA] (e.g., Zammit and
Newsholme, 1979), as stored triglycerides are mobilized for hepatic
metabolism. However, in those species where the liver is the major
storage site for lipids (e.g., cod), plasma [NEFA] tend to fall with
starvation since triglycerides are already in the liver and their products
need not enter the blood stream (Black and Love, 1986; Garcia-Garrido
et al., 1990).
    Plasma [NEFA] levels are also affected by spawning and migration.
In all species examined to date, plasnia [NEFA] consistently increases
during spawning. For example, in plaice, the peak plasma [NEFA]
(0.5-0.6 mM; White et ul., 1986)was seen at spawning and during the
period of increased feeding postspawning. Similar trends are seen in
the sand dab (Fletcher, 1985), Spicara chryselis ( F e r n h d e z and
Planas, 1980), the sea bass (FernAndez et al., 1989), spotted dogfish
(Garcia-Garrido et al., 1990), and salmonids (Plisetskaya, 1980). The
peak in plasma [NEFA] associated with spawning is associated with
gametogenesis, especially vitellogenesis, and reflects the mobilization
of lipid reserves required for gondal development (Fletcher, 1985).
The increased activity associated with spawning and migration may
also be a contributing factor.
104                               D. G. MCDONALD AND C. L. MILLIGAN

    The changes in plasma [NEFA] associated with maturation are not
well documented and are limited to observations in salmonids. Gener-
ally, smoltification results in a reduction in plasma [NEFA] (by about
50%; Sheridan, 1989; Gong and Farrell, 1990) and a change in plasma
[NEFA] composition. The change in NEFA composition accompany-
ing smoltification may be explained by differences in diet, water tem-
perature, and salinity, although the physiological significance remains
obscure (see Sheridan, 1989 for a review).
    Although the response of plasma [NEFA] to stress has not received
systematic study, it is apparent that the response is quite variable,
again making generalizations difficult. Stress is often associated with
hyperglycemia, which, in mammals, results in an inhibition of lipid
mobilization and a reduction in plasma [NEFA]. However, this is not
necessarily so in fish. In extensive experiments with common carp,
Mazeaud and co-workers showed that stress both increased and de-
creased plasma [NEFA], independent of changes in blood glucose
(Mazeaud and Mazeaud, 1981). In response to hypoxia and handling
stress, plasma [NEFA] increased in rainbow trout but not until 4-5 h
after the stress (Mazeaud and Mazeaud, 1981). Tench (Tinca tinca)
show a biphasic response to stress: Plasma [NEFA] was elevated im-
mediately following the stress and later declined to below prestress
levels (see Plisetskaya, 1980). Plisetskaya grouped fish into two very
broad categories based on the response of plasma [NEFA] to hypoxic
stress: (a) those that respond to stress by increasing both plasma [glu-
cose] and [NEFA] (e.g., lamprey, trout); and (b)those that show hyper-
glycemia and a lowering of plasma [NEFA] (e.g., carp, pike, perch
(PercaJuviatilis),  bream (Abramis brama).However, given the incon-
sistencies in the type of stress and the influence of such variables as
starvation, sexual maturation, and even time of the day the fish are
stressed (Mazeaud and Mazeaud, 1981) on the stress response, such
generalizations are premature.

E. Triglycerides
   Triglycerides (TG) are the primary storage form of lipid in most fish
species and are readily mobilized in response to physiological de-
mand. Triglycerides released from storage sites (e.g., liver, adipocytes)
are transported in the plasma in association with VLDL and LDL.
Fatty acids are released from TG by the action of extracellular lipases;
the fatty acids are taken up by the tissue, and the glycerol backbone is
transported back to the liver via the plasma.
2.   CHEMICAL PROPERTIES OF THE BLOOD                                  105

   As is the case with the other lipids, the level of plasma TG varies
throughout the life cycle of many fish and is affected by such factors as
sexual maturation, smoltification (in the salmonids), spawning, and
nutritional status. In the salmonids, smoltification generally results in a
depletion of body lipids, reflected in a reduction in plasma TG levels
from 1100 mg dl-' in parr to 700 mg dl-' in smolts (Dannevig and
Norum, 1982; Sheridan, 1989).
   In all teleost species examined, the highest levels of plasma TG are
associated with the period of peak feeding in preparation for spawning.
In feeding prespawning plaice, channel catfish, and Arctic
char (Saluelinus alpinus), plasma TG levels peaked at 100, 600, and
750 mg dl-', respectively (Dannevig and Norum, 1983; McKay et al.,
1985; White et al., 1986). During the period of spawning, plasma TG
levels generally decline by about 20-50%. The extent of the decline
appears to be related to the level of activity during spawning: The
more active fish (e.g., Arctic char) experience a greater decline in TG
levels (McKay et al., 1985; Sheridan, 1989).
   Similarly, starvation results in a loss of TG from the plasma: The
longer the period of starvation, the greater the decline in TG levels.
In sea bass starved for 40 days, plasma TG levels dropped from 560 to
280 mg dl-' and after 150 days, TG levels fell to about 70 mg dl-'
(Zammit and Newsholme, 1979). Similarly, in Arctic char that have
ceased feeding due to low temperature, plasma TG levels fell by about
60% to 200 mg dl-' after 60 days (Dannevig and Norum, 1983).
   The decrease in plasma TG levels is accompanied by an increase in
free glycerol. In sea bass, free glycerol doubled in spawning fish,
increasing from 2.76 to 5.5 mg dl-' (White et ul., 1986). Similarly, after
100 days of starvation, glycerol increased from 0.18 to 6.7 mg dl-'
(Zammit and Newsholme, 1979).There was, however, no further in-
crease after 150 days of starvation.
   Generally, elasmobranchs tend to have lower plasma T G levels
than do teleosts and are less dependent on the physiological and
nutritional state of the animal. In spotted and spiny dogfish, plasma TG
levels ranged from 30 to 120 mg dl-' (Zammit and Newsholme, 1979;
Garcia-Garrido et al., 1990) and, at least in the spotted dogfish, were
unaffected by 100-150 days of starvation (Zammit and Newsholme,
1979) or sexual maturation (Garcia-Garrido et al., 1990). This differ-
ence may be explained b y the fact that elasmobranchs do not mobilize
lipid reserves as triglycerides or NEFA, but rather as ketone bodies,
which is supported by the absence of detectable free glycerol in the
plasma (Zammit and Newsholme, 1979).
106                                D. G. MCDONALD AND C. L. MILLIGAN


   The major trends in electrolyte/osmolyte composition in fish from
different groups is well known and only the highlights will be re-
viewed here. For more detailed information the comprehensive re-
view by Holmes and Donaldson (1969) in Volume I11 should be con-

A. Na+, C1-, and Osmolarity
     Na+ and C1- are the major ions in the blood of all fishes. Na+ con-
centration typically exceeds [Cl-] in all but the hagfishes (Table IV)
where the Na :C1 ratio is 0.95 reflecting the similarity of hagfish plasma
to seawater (seawater Na: C1 ratio = 0.86). In lampreys and marine
elasmobranchs the Na: C1 ratio is 1.04 while it averages 1.1 in most
teleosts (both marine and fresh water). The one major exception is the
freshwater anguillids that have unusually low plasma [Cl-] and where
the ratio averages 1.55 (cf. Farrell and Lutz, 1975). Furthermore, in all
but the chondrichthyes and the coelocanths, NaCl contributes over
75% of the osmolarity of the plasma. In marine cartilaginous fishes,
NaCl makes u p only about 50% of the total osmolarity. Nonprotein
nitrogen compounds, mostly urea, secondarily TMAO, make up most
of the balance. Plasma Na+ and C1- measurements are much more
common than measurements of plasma osmolarity, and the electrolyte
measurements are typically more reliable because of the simplicity
and standardization of the techniques. Hence, we have used the equa-
tion OsmNaCl= (Na+ + C1-) * 0.91 (the NaCl osmotic activity coeffi-
cient; Robertson, 1989) instead of osmolarity in a survey of the electro-
lyte composition (degree of dilution of body fluids) among several
groups (see Table IV).
     AS Lutz (1975) has argued, there are broad trends in the osmolarity
and electrolyte composition of fish plasma that reflect the evolutionary
and environmental history of each group. The basic notion is that, with
the exception of the hagfish, all groups of fishes originated in fresh
water (or brackish water), 400 million years ago in the major radiation
that occurred in the Ordivician and Silurian periods. In the evoiution-
ary ferment of this period, selection pressures, Lutz argues, would
have strongly favored dilution of body fluids to reduce the cost of
osmoregulation. Hence, the lowest electrolyte levels are found in
the lungfishes (OsmN,cl = 172) and the African polypterids (Osm-
N a C l = 172) that have had a long and continuous history of life in fresh
water. Other groups (elasmobranchs, coelocanths, lampreys, chon-
2.   CHEMICAL PROPERTIES OF THE BLOOD                                107

drosts, holosts, and teleosts) have at some point in their evolutionary
history reinvaded seawater, and some representatives of all groups
except coelocanths later returned to fresh water. Ion levels in the fresh
water representatives of these groups are correspondingly higher than
in the lungfishes and reflect, approximately, the duration in evolution-
ary time spent in freshwater habitats.
    Among the freshwater elasmobranchs, the euryhaline species (e.g.,
Lake Nicaragua shark, bull shark; Carcharhinus sp.), which are quite
recent emigrants to fresh water (McFarland et al., 1979), have the
highest OsmNaCl of all freshwater fish (381) and still retain urea at
levels about 50%of that of their marine counterparts (137 versus 374 mM;
Table IV). I n contrast, the freshwater rays, Potarnotrygon sp., which
are stenohaline and have been in fresh water for at least 15 million
years (McFarland et al., 1979), have a much lower OsmNaCl (281),
while urea (0.7 mM) makes a negligible contribution to plasma osmo-
larity. The OsmNaClof the remaining freshwater fish species fall within
the range of 195-252 (lampreys [201], chondrosts [2161, holosts [2521,
and teleosts [ 195-2523). Freshwater teleosts exhibit almost the full
range of OsmNaCl in freshwater fishes but the trends within this group
are not as readily apparent. OsmNaCl in stenohaline freshwater species
such as the white sucker (195)are low relative to the euryhaline salmo-
nids (252),but other euryhaline species such as eel (Anguilla sp.) are
also low (229).
    Among marine teleosts, a clear distinction can be drawn between
euryhaline and stenohaline forms. In stenohaline species (average of
21 species) OsmNaCl averaged 346, while in euryhaline species (aver-
age of 16 species) it was significantly lower, averaging 311 (Table IV).
    Plasma Naf and C1- levels are sensitive to a wide variety of envi-
ronmental and endogenous influences. In addition to circadian fluctu-
ations described earlier, these influences can be divided into three
categories: (a) changes in [NaCI] related to a change in the osmotic
gradient across the gills (i.e., due to a change in external salinity),
(b) changes brought about by effects of external pollutants on gill
function, and (c) stress-related changes.

         OF       CHANGE
    The emphasis in research has been on euryhaline species (salmon,
trout, killifish, eels, flounders, and mullet; Evans, 1984), particularly
on salmonids and on the transfer from fresh water to seawater (see
Evans, 1984 in Vol XB for a review of euryhalinity; Hoar, 1976,1988 in
Vol XIB for reviews on smoltification in salmonids). The adaptability
of salmonids to seawater is both species and age dependent and is also
                                                                  Table IV
                                               Summary of Plasma Electrolytes and Urea" in Fish"

                                         Na+ C1-     NaCl   OSM     K+    Ca2+ Mg2+ PO4 SO4          Urea       Comments         References
    Seawater                             458   535   903    1070   10.5   9.8   42.6    0     29.1     0   6 analyses            1
      Hagfishes                          499   524   930    1069   9.5    5.4   18.2    6.5    5.2     3.3 Eptatretus, Myxine,   1
                                                                                                           Polis t remata
      Lamprey (Petromyzon marinus)       159 153     284     331    5.4   3.5    7.0    -      4.4    -                          2-4
      Marine elasmobranchs               266 259     476     991    8.4   5.6    2.8    4.0    2.3   374   9 species             1,5-7
      Coelacanth (Latimeria chalumnae)   197 187     349     932    5.8   4.9    5.3    5.1    4.8   377    1 living specimen    8
      Stenohaline marine teleosts        20 1 174    346     425    8.9   3.4    4.7    5.8    4.2     8.3 21 species            1,6
      Euryhaline marine teleosts         178 164     311     -      7.3   3.2    -      _     -       - 15 species               1
    Freshwater elasmobranch sp.          22 1 189    381     527    7.0   3.9    1.5    3.8   0.6    137 3 species               1,9
      Potamotrygon sp.                   152 158     281     295    3.1   1.4    4.5    -     -        0.7                       1, 10, 11
        Lamprey (freshwater adults)      117 104     201     256    4     1.9    1.2    5     -            P . marinus and L.    1,3, 12-14
       Dipnoi (lungfish)                 106    83   172     238    2.5   1.8    1.2    1.8    0.9     0.7 5 species             1, 11, 15
       Chondrostei                       130   108   216     -      3.3   2.0    1.9    1.5    0.6    - 2 species                15
       Pol ypteridae                     102    87   172     -      2.2   2.2    1.3    1.5    0.6    -    2 species             15
       Holostei                          147   131   252     -      2.6   2.8    1.5    1.9    1.4     0.5 2 species             1
       Salmonidae (Oncorhynchus              147   122    246      -      3.7    2.5      1.3    2.6    2.8     1.0 17 studies               16,22
       Cyprinidae                            133   118    228      274    3.3    1.8      1.5    0.4    -      -     Goldfish and carp       17-18
       Anguillidae                           151   110    229      293    3.0    2.9      1.7    1.5    -      -     A. rostrata and A.      1, 13, 19
        Catostomidae (Catastomus             120     94    195     245    2.6    2.1      0.8    -      1.0    -                             20-22
        Ictaluridae (Ictaluris punctatus)    142    110   229      259    3.1    -       -       -      -      -                             23

       (’All measurements are in mM.
         Except where noted values are means of 3-21 measurements with sources indicated. The osmolarity contributed by NaCl (OsrnNacl)was
   calculated as the sum of Na+ and C1- multiplied by 0.91 (After Robertson, 1989).
       1. Holmes and Donaldson (1969).2. Mathers and Beamish (1974).3. Beamish et al. (1978).4. Beamish (1980).5. Wells et al. (1986).6. Griffith
   (1981).7. Robertson (1989).8. Griffith et al. (1974).9. Thorson et al. (1973). 10. Griffith et al. (1973). 11. Mangum et al. (1978). 12. Pickering and
   Morris (1970).13. Robertson (1984).14. Tufts and Boutilier (1989).15. Urist et al. (1972). 16. Hille (1982).17. Houston and Koss (1982).18. Fuchs
0 and Albers (1988). 19. Farrell and Lutz (1975).20. Wilkes et al. (1981).21. H6be et al. (1983).22. Hdbe et al. (1987).23. Cameron (1980).
110                                D. G. MCDONALD AND C. L. hlILLIGAN

dependent, in some species, on there being preparatory physiological
changes in fresh water prior to seaward migration (smoltification). The
latter is particularly prominent in those species that migrate to sea only
once (e.g., all Oncorhynchus sp. except 0.mykiss) (Hoar, 1988). Upon
abrupt transfer of Oncorhynchus sp. to seawater there are transient
increases in plasma Na+ and C1-, not normally exceeding 20-40 mM in
the successfully adapting animal; the peak is reached within 12 h of
transfer and thereafter plasma levels decline to stabilize at or near
freshwater values, usually by 24-36 h (Sweeting and McKeown, 1987;
Yada et al., 1991). The corresponding return to fresh water for Oncor-
hynchus sp. is accompanied by a transient depression of plasma elec-
trolytes followed by a more gradual recovery, but still within 7 days
(e.g., Ogasawara et al., 1989). In most other teleosts, including salmo-
nids that have a lower seawater tolerance (e.g., Salvelinus sp.; Finstad
et al., 1989) and other teleosts with similar or higher seawater toler-
ance (e.g., milkfish, Chanos chanos; Ferraris et al., 1988) the response
to salinity transfer is usually more exaggerated. Plasma Na+ and C1-
levels take a much longer time to stabilize after salinity transfer (up to
2 weeks) and typically stabilize at levels substantially higher in sea-
water than in fresh water. For example, in eels (Anguilla sp.), seawater
adapted forms have plasma Na+ and C1- levels 50-60 mM higher than
freshwater forms (Table IV).

    A variety of environmental contaminants in fresh water and sea-
water are toxic to fish, at least in part, because they disrupt gill func-
tion. The primary effect of many of these contaminants is the impair-
ment of gill iono-regulation and, therefore, disruption of plasma
electrolyte balance (see McDonald et al., 1989 for review). Some tox-
ins, such as Cd and Zn, disrupt gill Ca2+ fluxes but the majority (e.g.,
acidic pH, Cu, Al, Hg) act principally by the disruption of NaCl fluxes.
This disruption develops in part through inhibition of NaCl transport,
but mostly through increasing the NaCl permeability of the gills (Mc-
Donald, 1983b; McDonald et al., 1989). Substantial depression of
plasma [NaCl] can result in freshwater fish (or elevation in marine fish;
Bouquegneau and Gilles, 1980) even with sublethal exposures. The
response of freshwater fish is characterized by an initially rapid de-
pression in plasma NaCl, which can be as much as 30%,and is usually
complete within 24 h (in sublethal exposures) followed by a more
gradual recovery where fish may either regain original ion levels or,
more commonly, reach a new steady state with continued depression
of plasma [NaCl]. For example, Audet et al. (1988) showed in rainbow
2.   CHEMICAL PROPERTIES OF THE BLOOD                                  111

trout surviving and feeding through 84 days at p H 4.8 that NaCl levels
were still about 20 mM lower than in neutral p H controls.

    Stress has profound effects on plasma electrolytes, particularly Na+
and C1-. A variety of acutely stressful procedures such as angling or
capture, enforced exhaustive exercise, and blood sampling trauma
provoke catecholamine and cortisol release and niuscle lactacidosis.
Over the time frame of a very few minutes, the latter increases in-
tracellular osmolarity, provoking a shift in fluid from the extracellular
space and elevating plasma [NaCll. For example, in rainbow
trout, 5-6 min of enforced exercise produced a 10-20 mM increase in
plasma Na+ and C1-, and both remained significantly elevated for 2-
4 h (Graham et al., 1982; Holeton et al., 1983). If the stress is prolonged
over the longer term, >0.5 h, then the impact of chronic catecholamine
elevation on branchial and renal ion and electrolyte flows will become
more apparent. For freshwater teleosts this will mean a net loss of NaCl
and for marine teleosts a net gain. For example, 8 h of crowding stress
in lake trout, Salvelinus namaycush, led to a 25-30 mM depression in
Naf and C1+ in the first 4 h, little change over the subsequent 4 h and,
when crowding stress was removed, over 24 h was required for recov-
ery (McDonald and Robinson, 1992).

B. Calcium
    In freshwater fishes total plasma Ca2+ levels are quite uniform,
falling within the narrow range of 2-3 mM, except for the euryhaline
elasmobranchs where plasma Ca2+ is 3.9 mM (Table IV). In marine
fishes, Ca2+ levels are higher (ranging from -3 to 2 5 mM in coelo-
canths, elasmobranchs, and myxinoids, Table IV) but are still one half
or less seawater values (-10 mM). [Ca"] is apparently very tightly
regulated in both freshwater and marine fishes. For example, a variety
of normal stresses produced either no effect on plasma [Ca"] in rain-
bow trout (hypoxia, hypercapnia, sustained exercise) or caused a slight
increase (exhaustive exercise; Andreason, 1985). A much more ex-
treme treatment, elevating [Ca2+]from 10 to 100 mM in seawater, led
to only a 40% increase in plasma [Ca"] in the Atlantic cod (Bjornsson
and Deftos, 1985). Furthermore, circadian fluctuations in plasma
[Ca2' 1 appear to be negligible (Houston and Koss, 1982; Kuhn et al.,
1986; Laidley and Leatherland, 1988b).
    A fraction of total plasma calcium is always protein bound. The
bound fraction in freshwater teleosts ranges from 30 to 48% in males
112                                D. G. MCDONALD AND C. L. MILLIGAN

and nongravid females (Bjornsson and Haux, 1985). Marine fishes are
less well studied but the bound fraction in the cod was 22% (Bjornsson
and Deftos, 1985). Any change in plasma protein will effect total
plasma [Ca" I although, generally speaking, the free [Ca2+I tends to
remain constant. The largest change in plasma protein (>threefold
increase) is that associated with vitellogenesis, where total Ca2+ levels
reach 5 mM in normally ovulating rainbow trout, but can be as high as
9 mM with estrogen injection (Bornsson and Haux, 1985; Bjornsson
et al., 1986). However, the ionic [Ca2+] remained constant at about
1.4 mM.

C . Magnesium
    Plasma M 2 + levels are lower than Ca2+ levels in all freshwater
fishes (except Potamotrygon sp.) and in the marine elasmobranchs
(Table IV). In the other marine fish, Mg2+ levels exceed Ca2+ levels
with the highest M 2 + levels being found in the myxinoids (18 mM)
although this is still less than half that of seawater (43 mM). Mg2+ is
less well studied than Ca2+ with only a few studies, all on freshwater
teleosts. However, the present evidence (van der Velden et al., 1989)
suggests that Mg2+ is as tightly regulated as Ca2+, even though the
mechanisms of Mg2+ regulation are unknown. Like Ca2+, Mg2' is
bound by plasma proteins; about 25% of the total plasma Mg2+ in
rainbow trout (Bjornsson and Haux, 1985). Since Mg2+ levels in eryth-
rocytes are 210-fold higher than plasma levels (Houston, 1985),hemo-
lysis can contribute significantly to erroneous plasma values.

D. Potassium
   Plasma levels are 2-4 mM in freshwater fish and 5-10 mM in
marine fish (Table IV). Less than 2% of the total body K+ is contained
in extracellular fluids (Eddy, 1985). Consequently, plasma K+ levels
are relatively unaffected by any treatments that increase gill electro-
lyte permeability because any branchial influx or efflux would be
readily buffered by transfers to and from the intracellular com-
partment. However, treatments that produce an intracellular acidosis,
such as strenuous exercise, will cause an outward leak of K+ from
muscle cells. For example, in rainbow trout, 6 min of exhaustive exer-
cise led to a progressive increase in plasma [K+ 1, reaching a peak twice
preexercise levels at 4 h postexercise and still significantly elevated at
8 h (Graham et al., 1982). Similar increases in plasma [K)] are seen in
marine teleosts following strenuous exercise (e.g., Boutilier et al.,
2.   CHEMICAL PROPERTIES OF THE BLOOD                                                    113

1984; Wells et al., 1986). There is also the potential for a substantial
error in estimates of plasma K+ levels if hemolysis occurs because of
high K+ levels in erythrocytes (90-130 mM).

E. Phosphate and Sulfate
   Data for both of these anions are sparse relative to other plasma
ions. Available data (Table IV) indicate that the levels of phosphate are
usually higher than sulfate and that both are higher in marine fish
compared to freshwater fish. A fraction of phosphate is protein bound,
about 17% in freshwater rainbow trout (Bjornsson and Haux, 1985),but
the amount bound increases with increasing plasma protein levels so
the free phosphate fraction remains unchanged (Bjornsson and Haux,
1985).Little, however, is known of the regulation of either ion nor of
the factors that affect plasma concentrations (c.f. Bjornsson and Haux,
1985; Hille, 1982; HGbe, 1987).


     We gratefully acknowledge the information and advice contributed by our col-
leagues, in particular, C . Audet, J . Ballantyne, A. P. Farrell, G. Flik, J. Gutibrrez, A. H.
Houston, B. A. McKeown, T. W. Moon, M. Nikinmaa, R. E. Peter, E. M. Plisetskaya, and
C . M. Wood. Any errors, however, are ours.


Alexander, J. B., and Ingram, G. A. (1980).Acomparison offive ofthe methods commonly
    used to measure protein concentrations in fish sera.]. Fish B i d . 16, 115-122.
Andreason, P. (1985). Free and total calcium concentrations in the blood of rainbow
    trout, Salmo gairdneri, during ‘stress’ conditions. J. E x p . B i d . 118, 111-120.
Arnold-Reed, D. E., Balment, R. J., McCrohan, C. R., and Hackney, C . M. (1991). The
    caudal neurosecretory system of Platichthys flesus: General morphology and re-
    sponses to altered salinity. Comp. Biochern. Physiol. 99A, 137-143.
Arthur, P. G., West, T. G., Brill, R. W., Schulte, P. M., and Hochachka, P. W. (1992).
    Recovery metabolism of tuna white muscle: Rapid and parallel changes of lactate
    and phosphocreatine after exercise. Can.J . Zool. (in press.)
Audet, C., Munger, R. S., and Wood, C . M. (1988). Long-term sublethal acid exposure in
    rainbow trout (Salmo gairdneri) in soft water: Effects on ion exchanges and blood
    chemistry. C u n J . Fish. Ayuat. Sci. 45, 1387-1398.
114                                          D. G. MCDONALD AND C. L. MILLIGAN

Avella, M., Schreck, C. B., and Prunet, P. (1991). Plasma prolactin and cortisol concentra-
    tions of stressed coho salmon, Onchorhynchus kisutch, in fresh water or salt water.
    Gen. C o m p . Endocrinol. 81,21-27.
Babin, P. J., and Vernier, J.-M. (1989). Plasma lipoproteins in fish. J . Lipid Res. 30,
Bailey, J . R., and Randall, D. J. (1981). Renal perfusion pressure and renin secretion in
    the rainbow trout, Salmo guirdneri. Can. J . Zoo/. 59, 1220-1226.
Balment, R. J., Hazon, N., and Perrott, M. N . (1987). Control of corticosteroid secretion
    and its relation to osmoregulation in lower vertebrates. In “Comparative Physiology
    of Environmental Adaptations” ( J . Kirsch and B. Lahlou, eds.), Vol. 1, pp. 92-102.
    Karger, Basel.
Barrett, B. A,, and McKeown, B. A. (1988). Sustained exercise augments long-term
     starvation increases in plasma growth hormone in the steelhead trout, Salmo
    gairdneri. C a n . J. 2001. 66,853-855.
Barrett, B. A,, and McKeown, B. A. (1989). Plasma growth hormone levels in Salrno
    gairdneri: Studies on temperature and the exercise intensity/duration relationship.
     C o m p . Biochem. Plzysiol. 94A, 791-794.
Barry, T. P., Santos, A. J. G., Furukawa, K., Aida, K., and Hanyu, I. (1990). Steroid profiles
     during spawning in male common carp. Gen. Comp. Endocrinol. 80,223-231.
Bates, D. J., Barrett, 8 . A., and McKeown, B. A. (1989). Daily variation in plasma growth
     hormone of juvenile coho salmon, Oncorhynchus kisutch. Cun. J . Zool. 67, 1246-
Beamish, F. W. H. (1980). Osmoregulation in juvenile and adult lamprey. C a n . ] . Fish
     Aquat. Sci. 37,1737-1750.
Beamish, F. W. H., Strachan, P. D., and Thomas, E. (1978). Osmotic and ionic perfor-
     mance of the anadromous sea lamprey, Petromyzoti marinus. C o m p . Biochem.
     Physiol. 60A, 435-443.
Bergheim, A., Kroglund, F., Vatne, D. F., and Rosseland, B. 0. (1990). Blood plasma
      parameters in farmed Atlantic salmon (Salmo salar L.) transferred to sea cages at age
      eight to ten months. Aquaculture 84, 159-165.
 Bern, H. A. (1990). The “new” endocrinology: Its scope and impact. Amer. Zoo/. 30,
 Bever, K., Chenoweth, M., and Dunn, A. (1977). Glucose turnover in kelp l ~ a s a    (Purula-
      brux sp.): In t h o studies with [6-‘HH,6-I4C] glucose. Arn.J. Physiol. 232, R66-R72.
 Billard, R., Breton, B., Fostier, A., Jalabert, B., and Weil, C. (1978). Endocrine control of
      the teleost reproductive cycle and its relation to external factors: Salmonid and
      cyprinid models. In “Comparative Endocrinology” (P. J. Gaillard and H. H. Boer,
      eds.), pp. 37-48. Elsevier/North Holland, New York.
 Bjornsson, B. T. and Deftos, L. J. (1985). Plasma calcium and calcitonin in the marine
      teleost, Gadus morhua. C o m p . Biochem. Physiol. 81A, 593-596.
 Bjornsson, B. T.,and Haux, C. (1985). Distribution ofcalcium, magnesium, and inorganic
      phosphate in plasma of estradiol-17p treated rainbow trout.]. C o m p . Physiol. B. 155,
 Bjornsson, B. T., Haux, C., Forlin, L., and Deftos, L. J. (1986). The involvement of
      calcitonin in the reproductive physiology of the rainbow trout. J. Endocritaol. 108,
 Bjornsson, B. T., Young, G., Lin, R. J., Deftos, L. J., and Bern, H. A. (1989). Smoltification
      and seawater adaptation in coho salmon (Oncorhynchus kisutch): Plasma calcium
      regulation, osmoregulation and calcitonin. Gen. Comp. Endocrinol. 74,346-354.
2.   CHEMICAL PROPERTIES OF THE BLOOD                                                 115

Black, D., and Love, R. M. (1986).The sequential mobilization and restoration of energy
    reserves in tissues of Atlantic cod during starvation and refeeding.]. Comp. Physiol.
     156B, 469-479.
Black, D., and Skinner, E. R. (1986). Features of lipid transport system of fish as demon-
     strated by studies on starvation in the rainbow trout. ]. Comp. Physiol. 156B, 497-
Blasco, J., FernLndez, J., and Gutierrez, J. (1991).The effects of starvation and refeeding
    on plasma amino acid levels in carp, Cyprinus carpio L., 1758.1. Fish Biol. 38,
Boffa, G. A,, Martin, B., Winchenne, J. J., and Ozon, R. (1972). Interactions steroids-
    proteines dans le sCrum de l’Homme, d’un Amphibien et d’un cyclostome. Bio-
    chimie 54,1137-1145.
Bolton, J . P., Takahashi, A., Kawauchi, H., Kubota, J., and Hirano, T. (1986). Develop-
    ment and validation of a salmon growth hormone radioimmunoassay. Gen. Comp.
    Endocrinol. 62,230-238.
Bouck, G. R., Cairns, M. A,, and Christian, A. R. (1978).Effect ofcapture stress on plasma
    enzyme activities in rainbow trout (Salmo gairdneri). ]. Fish Res. Board Can. 35,
Bouquegneau, J. M., and Gilles, R. (1980). Osmoregulation and pollution of the aquatic
    medium. I n “Mechanisms of Osmoregulation in Animals” (R. Gilles, ed.), pp. 563-
     580. Springer-Verlag, Berlin.
Boutilier, R. G . , Aughton, P., and SheIton, G. (1984).O2 and COB transport in relation to
     ventilation in the Atlantic mackerel, Scomber scombrus. Can. ]. Zool. 62,546-554.
Boutilier, R. G., Dobson, G., Hoeger, U., and Randall, D. J. (1988). Acute exposure to
     graded levels of hypoxia in rainbow trout (Salmo gairdneri): Metabolic and respira-
     tory adaptations. Respir. Physiol. 71,69-82.
Boyd, T. A., Chung-Ja, C., Forster, R. P., and Goldstein, L. (1977). Free amino acids in
    tissues of the skate Raja erinacea and the stingray Dasyatis sabina: Effects of
    environmental dilution. J. Exp. Zool. 199,435-442.
Breton, R., and Billard, R. (1977). Effects of photoperiod and temperature on plasma
     gonadotropin and spermatogenesis in the rainbow trout, Salmo gairdneri,
     Richardson. Ann. Biol. Anim. Biochem. Biophys. 17,331-340.
Breton, B., Kann, G., Burzawa-Gerard, E., and Billard, R. (1971). Dosage radioimmu-
     nogique d’une hormone gonadotrope de Carpe (Cyprinus carpio L.). C . R . Acad. Sci.
     Paris Ser. D 272,1515-1517.
Bry, C. (1982).Daily variations in plasma cortisol levels of individual female rainbow
     trout, Salmo gairdneri: Evidence for a post-feeding peak in well-adapted fish. Gen.
     Conip. Endocrinol. 48,462-468.
Butler, P. J ~Metcalfe, J. D., andGinley, S. A. (1986).Plasma catecholamines in the lesser
     spotted dogfish and rainbow trout at rest and during different levels of exercise.
    ]. Exp. Biol. 123,409-421.
Callard, G . V. (1988). Reproductive physiology. I n “The Physiology of Elasmobranch
     Fishes” (T. J. Shuttleworth, ed.), Part B, pp. 292-317. Springer-Verlag, Berlin.
Callard, I. P., and Klosterman, L. (1988). Reproductive physiology. I n “The Physiology
    of Elasmobranch Fishes” (T. J. Shuttleworth, ed.), Part A, pp. 277-292. Springer-
    Verlag, Berlin.
Cameron, J. N. (1980). Body fluid pools, kidney function, and acid-base regulation in the
    freshwater catfish Ictalurus punctatus. ]. Exp. Biol. 86, 171-186.
Cameron, J. N., and Heisler, N. (1983). Studies ofammonia in the rainbow trout: Physi-
    cochemical parameters, acid-base behaviour and respiratory clearance.]. Exp. Biol.
     105, 107-125.
116                                        D. G. MCDONALD AND C. L. MILLIGAN

Carrillo, M., Zanuy, S., and Herrera, E. (1980). Daily rhythms of amino acid levels in the
    plasma of goldfish (Carassius auratus). Comp. Biochem. Physiol. 67A, 581-586.
Carrillo, M., Zanuy, S., and Herrera, E. (1982). Growth and diurnal variations in meta-
    bolic parameters in the starved bass, Dicentrarchus labrax, after experimental feed-
    ing. Comp. Biochem. Physiol. 72A, 11-16.
Carillo, M., Zanuy, S., Montoro, R., and Ollevier, F. (1986). Daily patterns of blood
    electrolytes in sea bass adapted to different osmotic media at different times of the
    year. J . Interdisc. Cycle Res. 17,69-80.
Clark, N. B. (1983).Evolution of calcium regulation in lower vertebrates. Amer. Zool. 23,
Cook, R. F., and Eales, J. G. (1987). Effects of feeding and photocycle on die1 changes in
    plasma thyroid hormone levels in rainbow trout, Salmo gairdneri. J Exp. Zool. 242,
Cook, A. F., Wilson, S. W., and Peter, R. E. (1983).Development and validation of a carp
    growth hormone radioimmunoassay. Gen. Comp. Endocrinol. 50,335-347.
Copeland, P. A., Sumpter, J. P., Walker, T. K., and Croft, M. (1986).Vitellogenin levels in
    male and female rainbow trout (Salmo gairdneri Richardson) at various stages ofthe
    reproductive cycle. Comp. Biochem. Physiol. 83B, 487-493.
Copp, H. (1969). The ultimobranchial glands and calcium regulation. I n “Fish Physiol-
    ogy” (W. S. Hoar and D. J. Randall, eds.), Vol. 6, pp. 377-398. Academic Press, New
Cornish, I., and Moon, T. W. (1985). Glucose and lactate kinetics in American eel
    Anguilla rostrata. A m . ] . Physiol. 249, R67-R72.
Crim, L. W. (1982).Environmental modulation of annual and daily rhythms associated
    with reproduction in teleost fishes. Can.]. Fish.Aquat. Sci. 39, 17-21.
Crim, L. W., Peter, R. E., and Billard, R. (1976).Stimulation ofgonadotropin secretion by
    intra-ventricular injection of hypothalamic extracts in the goldfish, Carussius au-
    rutus. Gen. Comp. Endocrinol. 3 0 , 7 7 4 2 .
Crim, L. W., Meyer, R. K., and Donaldson, E. M. (1973).Radioimmunoassay estimation
    of plasma gonadotropin levels in the spawning pink salmon. Gen. Comp. Endocri-
    nol. 21,69-76.
Cyr, D. G., and Eales, J. G. (1989). Effects of short-term 17P-estradiol treatment on the
    properties of T3-binding proteins in the plasma of immature rainbow trout, Salmo
    gairdneri. J. Exp. Zool. 252,245-251.
Dabrowski, K. (1982).Postprandial distribution of free amino acids between plasma and
    erythrocytes of common carp (Cyprinus carpio). Comp. Biochem. Physiol. 72A,
Dannevig, B. H., and Norum, K. R. (1983). Cholesterol esterification and lipids in the
    blood plasma of the char (Saluelinus alpinus L.) during sexual maturation. Conap.
     Biochem. Physiol. 73B, 771-777.
Danulat, E., and Hochachka, P. W. (1989). Creatine turnover in the starry flounder,
     Platichthys stellatus. Fish Physiol. Biochem. 6, 1-9.
D’Apollonia, S., and Anderson, P. D. (1980). Optimal assay conditions for serum and
     liver glutamate oxaloacetate transaminase, glutamate pyruvate transaminase, and
     sorbitol dehydrogenase from the rainbow trout, Salmo gairdneri. Can.]. Fish Aquat.
     Sci. 37, 163-169.
Darnell, J., Lodish, H., and Baltimore, D. (1990). “Molecular Cell Biology.” Scientific
     American Books, Inc., New York.
Dashow, L., Epple, A,, and Nibbio, B. (1982). Catecholamines in adult lampreys: Base-
     line levels and stress-induced changes, with a note on cardiac cannulation. Gen.
     Comp. Endocrinol. 46,500-504.
2.   CHEMICAL PROPERTIES OF THE BLOOD                                                  117

Dave, G., Johansson-Sjobeck, M.-L., Larson, A., Lewander, K., and Lidman, U. (1976).
   Metabolic and hematological effects of starvation in the European eel, Anguilla
   anguilla. I. Carbohydrate, lipid, protein and inorganic ion metabolism. Comp. Bio-
   chem. Physiol. 52A, 423-430.
Davidson, W. S., Birt, V. L., Birt, T. P., and Green, J. M. (1988).Palmitate-binding, serum
   albumin-like proteins in salmonids. FEBS Lett. 233,299-302.
Davies, P. L., Hew, C. L., and Fletcher, G. L. (1988). Fish antifreeze proteins: Physiol-
   ogy and evolutionary biology. C a n . ] . Zool. 66,2611-2617.
Davis, K. B., and Parker, N. C. (1983). Plasma corticosteroid and chloride dynamics in
   rainbow trout, Atlantic salmon and lake trout during and after stress. Aquaculture 32,
Davis, K. B., and Parker, N. C. (1986).Plasma corticosteroid stress response of fourteen
   species of warmwater fish to transportation. Trans. Am. Fish SOC. 115,495-499.
Deftos, L. J., Watts, E. G., Copp, D. H., and Potts, J. T., Jr. (1974).A radioimmunoassay
   for salmon calcitonin. Endocrinology 94, 155-160.
Delahunty, G., Olcese, J., Prack, M., Vodicnik, M. J., Schreck, C. B., and deVlaming, V.
    (1978). Diurnal fluctuations in the physiology of the goldfish, Carassius auratus.
   J. Inter. Cycle Res. 9,75-88.
DeLaney, R. G., Lahiri, S., Hamilton, R., and Fishman, A. P. (1977). Acid-base balance
   and plasma composition in the aestivating lungfish (Protopterus). Am. J. Physiol.
   232, RIO-Rl7.
deRoos, R., deRoos, C. C., Werner, C. S., and Werner, H. (1985). Plasma levels of
   glucose, alanine, lactate and P-hydroxybutyrate in the unfed spiny dogfish shark
   (Squalus acanthias) after surgery and following mammalian insulin infusion. Gen.
   Comp. Endocrinol. 58,28-43.
DeVries, A. L. (1983). Antifreeze peptides and glycopeptides in cold-water fishes. Ann.
   Rev. Physiol. 45,245-260.
Dickhoff, W. W., and Darling, D. S. (1983).Evolution of thyroid function and its control
    in lower vertebrates. Amer. Zool. 23,697-707.
Dickhoff, W. W., Yan, L., Plisetskaya, E. M., Sullivan, C. V., Swanson, P., Hara, A., and
    Bernard, M. G. (1989). Relationship between metabolic and reproductive hormones
    in salmonid fish. Fish Physiol. Biochem. 7, 147-155.
Dindo, J. J., and MacGregor 111, R. (1981). Annual cycle of serum gondal steroids and
   serum lipids in striped mullet. Trans. Am. Fish Soc. 110,403-409.
Dobson, G. P., and Hochachka, P. W. (1987).Role of glycolysis in adenylate depletion
   and repletion during work and recovery in teleost white muscle. /. E x p . Biol. 129,
Dobson, S., and Dodd, J. M. (1977). Endocrine control of the testis in the dogfish,
    Scyliorhinus canicula, L. I. Effects of partial hypohysectomy on gravimetric, hor-
   monal, and biochemical aspects of testis function. Gen. Cornp. Endocrinol. 31,
Dodd, J. M. (1983). Reproduction in cartilaginous fishes. In “Fish Physiology” (W. S.
    Hoar, D. J. Randall, andE. M. Donaldson, eds.), Vol. IX, PartA, pp. 31-95. Academic
    Press, New York.
Donaldson, E. M. (1981). The pituitary-interrenal axis as an indicator of stress in fish. In
    “Stress in Fish” (Pickering, A. D. ed.), pp. 11-47. Academic Press, New York.
Donaldson, E. M. (1990). Reproductive indices as measures of the effects of environ-
    mental stressors in fish. I n “Biological Indicators of Stress in Fish” (S. M. Adams,
    ed.), Symposium 8, pp. 109-122. American Fisheries Society, Bethesda, Maryland.
118                                        D. G. MCDONALD AND C. L. MILLIGAN

Doolittle, R. F. (1984). Evolution of the vertebrate plasma proteins. In “The Plasma
    Proteins” (F. W. Putnam, ed.), 2nd ed., Vol. IV, pp. 317-360. Academic Press, New
Doolittle, R. F. (1987). Evolution of the vertebrate plasma proteins. Biol. Bull. 172,
Driedzic, W. R., and Hochachka, P. W. (1976). Control of energy metabolism in fish
    white muscle. A m . J . Physiol. 230,579-582.
Dufour, S., Belle, N. D. L., and Fontaine, Y. A. (1983). Development of a heterologous
    radioimmunoassay for eel (Anguilla anguilla) gonadotropin. Gen. Comp. Endocri-
    nol. 49, 404-413.
Duncornbe, W. G. (1963).The colorimetric micro-determination of long chain fatty acids.
    Biochem. J . 88, 7-10.
Dunn, J. F., Hochachka, P. W., Davison, W., and Guppy, M. (1983). Metabolic ad-
    justments to diving and recovery in the African lungfish. Am.]. Physiol. 245, R651-
Dye, H. M., Sunipter, J. P., Fagerlund, U. H. M., and Donaldson, E. M. (1986).Changes
     in reproductive parameters during the spawning migration of pink salmon, Oncor-
     hynchus gorbuscha (Walbaum).J . Fish Biol. 29,167-176.
Eales, J. G. (1985).The peripheral metabolisrn of thyroid hormones and regulation of
     thyroidal status in poikilotherms. Can. J . Zool. 63, 1217-1231.
Eales, J. G., and MacLatchy, D. L. (1989).The relationship between T3 production and
     energy balance in salmonids and other teleosts. Fish Physiol. Biochem. 7,289-293.
Eales, J. G., and Shostak, S. (1985). Free T4 and T3 in relation to total hormone, free
     hormone indices, and protein in plasma of rainbow trout and arctic char. Gen. Comp.
     Endocrinol. 58,291-302.
Eddy, E. B. (1985). Uptake and loss of potassium by rainbow trout (Sulmo gairdneri) i n
     fresh water and dilute sea water.]. E x p . Biol. 118,277-286.
Ekins, R. (1986). The free hormone concept. In “Thyroid Hormone Metabolism” (G.
     Hennemann, ed.), pp. 77-106. Marcel Dekker, Inc., New York.
Ellis, M . J., and Poluhowich, J. J . (1981). Biliverdin concentrations in the plasmas of‘
     fresh and brackish water eels, Anguilla rostratu. Comp. Biochem. Physiol. 70A,
Eng, F., and Youson, J. H. (1991). Biliverdin in the serum of ammocoetes of Lumpterci
     lamottenii (Le Sueur). C a n . J . Zoo2 69, 1126-1129.
Eros, S., and Milligan, C. L. (1992). Unpublished observations.
Evans, D. H. (1984). The roles of gill permeability and transport mechanisms in eury-
     halinity. In “Fish Physiology” (W. S. Hoar and D. J. Randall, eds.),Vol. X, Part B, pp.
     239-284. Academic Press, New York.
Ewart, K. V., and Fletcher, G. L. (1990). Isolation and characterization of antifi-ecze
     proteins from smelt (Osrnerus mordax) and Atlantic herring (Clupea haretigus
     harengus). Can.J . Zool. 68, 1652-1658.
Farbridge, K. J,, and Leatherland, J. F. (1987). Lunar cycles of coho salmon, Oncor-
     hynchus kisutch. 11. Scale amino acid uptake, nucleic acids. metabolic reserves and
     plasma thyroid hormones. J . E x p . Biol. 129, 179-189.
Fargher, R. C . ,and McKeown, B. A. (1989).The effect ofprolactin on calcium homeosta-
     sis in coho salmon (Oncorhynchus kisutch). Gen. Comp. Endocrinol. 73,398-403.
Fargher, R. C., and McKeown, B. A. (1990). Differential response of plasma prolactin to
      freshwater transfer of smolts and post-smolts of seawater-adapted coho salmon. Gen.
      Comp. Endocrinol.75, 129-133.
Farrell, A. P., and Lutz, P. L. (1975). Apparent anion imbalance in the fresh water
      adapted eel.]. Comp. Physiol. 102, 159-166.
2.   CHEMICAL PROPERTIES OF THE BLOOD                                                     119

Farrell, A. P., and Munt, B. (1983).Cholesterol levels in the blood of Atlantic salmon.
    Coinp. Biochem. Physiol. 75A, 239-242.
Faulkner, N. W., and Eales, J . G. (1973). Investigations of iodothyonine binding to
    plasma proteins in brook trout, Salvelinus fontinalis, using precipitation, dialysis
    and electrophoretic methods. J. Fish Res. Board Can. 30, 1131-1140.
Fellows, F. C. I., and Hird, F. J. R. (1981). Fatty acid-binding proteins in the serum of
    various animals. Comp. Biochem. Physiol. 68B, 83-87.
Fellows, F. C. I., and Hird, F. J. R. (1982). A comparative study of the binding of
    L-tryptophan and bilirubin by plasma proteins. Arch. Biochem. Biophys. 216, 93-
Fellows, F. C. I., and McLean, R. M. (1982).A study of the plasma lipoproteins and the
    tissue lipid levels of the migrating lamprey Mordacia mordaz. Lipids 17, 741-
Fellows, F. C. I., Hird, F. J. R., McLean, R. M., and Walker, T. J. (1980). A survey of the
    nonesterified fatty acids and binding proteins in the plasmas of selected animals.
    Conip. Biochem. Physiol. 67B, 593-597.
Fenwick, J. C., and Lam, T. J. (1988). Effects of calcitonin on plasma calcium and
    phosphate in the mudskipper, Periophthulmodon schlosseri (Teleostei), in water
    and during exposure to air. Gen. Comp. Eudocrinol. 70,224-230.
Fernandez, J., and Planas, J. (1980). Annual variations of some carbohydrate and lipid
    parameters in the fish Spicara chryselis during captivity. Comp. Biochem. Physiol.
    67A, 383-389.
Fernandez, J . , Gutierrez, J . , Carrillo, M., Zanuy, Z., and Planas, J. (1989).Annual cycle of
    plasma lipids in sea bass, Dicentrarchus labrax L.: Effects of environmental condi-
    tions and reproductive cycle. Comp. Biochem. Physiol. 93A, 407-412.
Ferraris, R. P., Almendras, J . M., and Jazul, A. P. (1988). Change in plasma osmolarity
    and chloride concentration during abrupt transfer of milkfish (Chanos chanos) from
    seawater to different test salinities. Aquaculture 70, 145-157.
Fincham, D. A,, Wolowyk, M. W., and Young, J. D. (1990).Characterization ofaminoacid
    transport of a primitive vertebrate, the pacific hagfish ( Eptatretus stouti). J. E x p .
    Biol. 154,355-370.
Finstad, B., Nilssen, K. J., and Arnesen, A. M. (1989). Seasonal changes in sea-water
    tolerance of Arctic charr (Salvelinus alpinus).J. Corny. Physiol. B. 159,371-378.
Fletcher, D. (1985). Plasma glucose and plasma fatty acid levels ofLimanda limandu (L.)
    in relation to season, stress, glucose loads and nutritional state. J. Fish Biol. 16,
Fletcher, G . L. (1975).The effects of capture, “stress,” and storage of whole blood on the
     red blood cells, plasma proteins, glucose, and electrolytes of the winter flounder
     (Pseudopleuronectes amrricanus). Can. J. Zool. 53, 197-206.
Fletcher, G. L., Kao, M. H., and Haya, K. (1984).Seasonal and phenotypic variations in
    plasma protein antifreeze levels in a population of marine fish, sea revan (Nemitrip-
     t e r m americanus). Can. J . Zool. 41, 819-824.
Flik, G . ,Fenwick, J. C., and Wendelaar Bonga, S. (1989).Calcitropic actions of prolactin
    in freshwater North American eel (Anguilla rostrata LeSueur). A m . J. Physiol. 257,
Flik, G., Labedz, T., Neelissen, J . A. M., Hanssen, R. G. J. M., Wendelaar Bonga, S. E.,
    and Pang, P. K. T. (1990). Rainbow trout corpuscles of Stannius: Stanniocalcin
    synthesis i n oitro. Am. J. Physiol. 258, R1157-Rl164.
Fostier, A., Jalabert, B., Billard, R., Breton, B., and Zohar, Y. (1983). The gonadal
    steroids. In “Fish Physiology” (W. S. Hoar, D. J. Randall, and E. M. Donaldson, eds),
    Vol. IX, Part A, pp. 277-346. Academic Press, New York.
120                                         D. G. MCDONALD A N D C. L. MILLIGAN

Fouchereau-Peron, M., Arlot-Bonnemains, Y., Maubras, L., Milhaud, G., and Moukhtar,
   M. S. (1990). Calcitonin variations in male and female rainbow trout, Salmo
   gairdneri, during the annual cycle. Gen. Comp. Endocrinol. 78, 159-163.
Freeman, H. C., and Idler, D. R. (1971). Binding affinities of blood proteins for sex
   hormones and corticosteroids in fish. Steroids 17,233-250.
Freemont, L., Leger, C., Boudon, M., and Gozzelino, M.-T. (1981). Fatty acid composi-
   tion of lipids in the trout. 11. Fractionation and analysis of plasma lipoproteins.
   Comp. Biochem. Physiol. 63B, 107-113.
Fuchs, D. A., and Albers, C. (1988).Effect of adrenaline and blood gas conditions on red
   cell volume and intra-erythrocytic electrolytes in the carp, Cyprinus carpio. J. E x p .
    Biol. 137,457-476.
Garcia-Garrido, L., Mufioz-Chapuli, R., and d e Andres, A. V. (1990). Serum cholesterol
    and triglyceride levels in Scyliorhinus canicula (L.) during sexual maturation.
   J . Fish Biol. 36,499-509.
Gaudet, M., Racicot, J-G., and Leray, C. (1975).Enzyme activities ofplasmaand selected
    tissues in rainbow trout, Salmo gairdneri. J. Fish Biol. 7,505-512.
Gem, W., Dickhoff, W. W., and Folmar, L. C. (1984).Increases in plasma melatonin titers
    accompanying seawater adaptation of coho salmon (Oncorhynchus kisutch). Gen.
    Comp. Endocrinol. 55,458-462.
Gern, W. A., Owens, D. W., and Ralph, C. L. (1978). Plasma melatonin in trout: Day-
    night change demonstrated by radioimmunoassay. Gen. Comp. Endocrinol. 34,453-
Gerst, J. W., and Thorson, T. B. (1977). Effects of saline acclimation on plasma electro-
    lytes, urea excretion, and hepatic urea biosynthesis in a freshwater stingray, Potamo-
    t y g o n sp. Graman, 1877. Comp. Biochem. Physiol. 56A, 87-93.
Girard, S. S., and Milligan, C. L. (1992). Unpublished observations.
Goldstein, L., and Kleinzeller, A. (1987). Cell volume regulation in lower vertebrates.
    Curr. Topics Membr. Transport 30,181-204.
Gong, B., and Farrell, A. P. (1990).Comparison ofblood and muscle levels ofunsaturated
    fatty acids in parr and mature coho salmon. Comp. Biochem. Physiol. 96B, 483-486.
Gonzalez, R. J . , and McDonald, D. G. (1992).The relationship between oxygen con-
    sumption and ion loss in a freshwater fish. J. E x p . Biol. 163,317-332.
Goos, H. J. T., Leeuw, R. D., Burzawa-Gerard, E., Terlou, M., and Richter, C. J . J. (1986).
    Purification of gonadotropic hormone from the pituitary of the African catfish,
    Clarias gariepinus (burchell), and the development of a homologous radioimmuno-
    assay. Gen. Comp. Endocrinol. 63, 162-170.
Gorbman, A. (1989). Roots of fish endocrinology, aperspective. Fish Physiol. Biochem. 7,
Gowenlock, A. H. (1988).“Varley’s Practical Clinical Biochemistry,” 6th ed. Heinemann
     Medical Books, London.
Graham, M. S., Wood, C. M., and Turner, J. D. (1982).The physiological responses ofthe
    rainbow trout to strenuous exercise: Interactions of water hardness and environmen-
    tal acidity. C a n . ] . Zool. 60,3153-3164.
Gras, J., Gudefin, Y., Chagny, F., and Perrier, H. (1982).Free amino acids and ninhydrin-
     positive substances in fish. 11. Cardio-respiratory system: Plasma, erythrocytes,
     heart and gills ofthe rainbow trout (Salmo gairdneri). Comp. Biochem. Physiol. 73B,
Grau, E. G. (1988). Environmental influences on thyroid function in teleost fish. Amer.
Grau, E. G., Dickhoff, W. W., Nishioka, R. S., Bern, H. A., and Folmar, L. C. (1981).Lunar
2.   CHEMICAL PROPERTIES OF THE BLOOD                                                  121

    phasing of the thyroxine surge preparatory to seaward migration of salmonid fish,
    Science 211,607-609.
Griffith, R. W. (1981).Composition ofthe blood serum ofdeep-sea fishes. Biol. Bull. 160,
Griffith, R. W., Pang, P. K. T., Srivastava, A. K., and Pickford, G. E. (1973). Serum
    composition of freshwater stingrays (Potomotrygonidue)adapted to fresh and dilute
    seawater. Biol. Bull. 144, 304-320.
Griffith, R. W., Umminger, B . L., Grant, B. F., Pang, P. K. T., and Pickford, G. E. (1974).
    Serum composition of the Coelacanth, Latimeria chalumnae Smith. J. E x p . 2001.
GutiCrrez, J . , Carillo, M., Zanuy, S., and Planas, J. (1984).Daily rhythms of insulin and
    glucose levels in the plasma of sea bass Dicentrarchus labrar after experimental
    feeding. Gen. Comp. Endocrinol. 55,393-397.
GutiCrrez, J., Fernandez, J., Blasco, J., Gesse, J. M., and Planas, J. (1986). Plasma
    glucagon levels in different species of fish. Gen. Comp. Endocrinol. 63,328-333.
GutiCrrez, J., Fernandez, J., Carillo, M., Zanuy, S., and Planas, J. (1987a).Annual cycle of
    plasma insulin and glucose of sea bass, Dicentrachus labrux, L. Fish Physiol. Bio-
    chem. 4,137-141.
Gutierrez, J., Fernandez, J., Gesse, J. M., and Blasco, J. (198713). Amino acid levels in
    whole blood and plasma of Scyliorhinus cuniculu. Comp. Biochem. Physiol. 87A,
GutiCrrez, J., Fernandez, J . , and Planas, J. (1988).Seasonal variations ofinsulin and some
    metabolites in dogfish plasma, Scyliorhinus canicula L. Gen. Comp. Endocrinol. 70,
GutiCrrez, J . , PCrez, J., Navarro, I., Zanuy, S., and Carrillo, M. (1990).Changes in plasma
    glucagon and insulin associated with fasting in sea bass (Dicentrurchus labrax). Fish
    Physiol. Biochem. 9, 107-112.
Haider, G. (1970). Alter-und saisonbedingte Veranderungen im Sereiwelpbild der
    Regenbogenforeile (Salnio gairdneri Richardson). Z. Fisch. N R . 18, 107-124.
Haversein, L. S., Aasjord, P. M., Ness, S., and Endersen, C. (1988). Purification and
    partial characterization of an IgM-like serum immunoglobulin from Atlantic salmon
    (Salno salar). Deu. Comp. Immunol. 12,773-785.
Haywood, G. P. (1973).Hypo-osmotic regulation coupled with reduced metabolic urea
    in the dogfish, Poroderma africanum: An analysis of serum osmolarity, chloride and
    urea. Mar. Biol. 23, 121-127.
Heisler, N. (1984).Acid-base regulation in fishes. In “Fish Physiology” (W. S. Hoar and
    D. J. Randall, eds.), Vol. X,Part A, pp. 315-401. Academic Press, New York.
Hemre, G-I., Lie, @., Lambertsen, G., and Sundby, A. (1990). Dietary carbohydrate
    utilization in cod (Cadus morhua). Hormonal response of insulin, glucagon, and
    glucagon-like peptide to diet and starvation. Comp. Bichem. Physiol. 97A, 41-44.
Henderson, R. J., and Torcher, D. R. (1987).The lipid composition and biochemistry of
    freshwater fish. Prog. Lipid Res. 26,281-347.
Hille, S . (1982). A literature review of the blood chemistry of rainbow trout, Salmo
    gairdneri Rich. J . Fish Biol. 20,535-569.
Hilton, J . W., and Atkinson, J. L. (1982). Response of rainbow trout (Salmo gairdneri) to
    increased levels of available carbohydrate in practical trout diets. Br. J. Nutr. 47,
Hilton, J. W., Plisetskaya, E. M., and Leatherland, J. F. (1987). Effects of oral 3,5,3’-
    triodo-L-thyronine on dietary glucose utilization and plasma insulin levels in rain-
    bow trout (Salmo gairdneri). Fish Physiol. Biochem. 4, 113-120.
122                                         D. 6. MCDONALD AND C . L. MILLIGAN

Himick, B. A,, and Eales, J. G. (1990).The acute effect of food and glucose challenge on
    plasma thyroxine and triiodothyronine levels in previously starved rainbow trout
    (Oncorhynchus mykiss). Gen. Comp. Endocrinol. 78,34-41.
Hirano, T. (1986). The spectrum of prolactin action in teleosts. In “Comparative Endo-
    crinology: Developments and Directions” (C. L. Ralph, ed.), pp. 53-74. A. R. Liss,
    New York.
Hirano, T., Ogasawara, T., Bolton, J. P., Collie, N. L., Hasegawa, S., and Iwata, M .(1987).
    Osmoregulatory role of prolactin in lower vertebrates. In “Comparative Physiology
    of Environmental Adaptations” ( J . Kirsch, and B. Lahlou, eds.), Vol. 1, pp. 112-124.
    Karger, Basel.
Hirano, T., Prunet, P., Kawauchi, H., Takahashi, A,, Ogasawara, T., Kubota, J., Nishioka,
    R. S., Bern, H. A., Takada, K., and Ishii, S. (1985). Development and validation of a
    salmon prolactin radioimmunoassay. Gen. Comp. Endocriizol. 59,266-276.
Hlavova, V. (1989). Enzyme activities in the blood plasma of grayling, Thymallus thy-
    mallus (Linn.), in the breeding season. J. Fish Biol. 34,779-789.
Hoar, W. S. (1976). Smolt transformation: Evolution, behavior and physiology. J. F i s h
    Res. B d . Can. 33,1234-1252.
Hoar, W. S. (1988).The physiology of smolting salmonids. I n “Fish Physiology” (W. S.
    Hoar and D. J. Randall, eds.), Vol. XI, Part B, pp. 275-343. Academic Press, New
Hobe, H. (1987). Sulphate entry into soft-water fish (Salmo garidneri, Catostomus
    commersoni) during low ambient pH exp0sure.J. E x p . Biol. 133,87-109.
Hdbe, H., Wilkes, P. R. H., Walker, R. L., Wood, C. M., and McMahon, B. R. (1983).
    Acid-base balance, ionic status, and renal function in resting and acid-exposed white
    suckers (Catostomus commersoni). Can.J . Zool. 61,2660-2668.
Hobe, H., Wood, C . M., and McMahon, B. R. (1984). Mechanisms of acid-base and
    ionoregulation in white suckers (Cutostomus commersoni) in natural softwater.
    I. Acute exposure to low ambient pH.J. C o m p . Physiol. 154B, 35-46.
Holder, F. C., Schroeder, M. D., Pollatz, M., GuernC.,J. M., Vivien-Roels, B., Pkvet, P.,
    Buijs, R. M., Dogterom, J., and Meiniel, A. (1982).A specific and sensitive bioassay
    for arginine-vasotocin: Description, validation, and some applications in lower and
    higher vertebrates. Gen. Comp. Endocrinol. 47,483-491.
Holeton, G . F., Neumann, P., and Heisler, N. (1983). Branchial ion exchange and
    acid-base regulation after strenuous exercise in rainbow trout (Salino gairdtier-i).
    Resp. Physiol. 51,303-318.
Holmes, W. N., and Donaldson, E. M. (1969). The body compartments and the distribu-
     tion ofelectrolytes. In “Fish Physiology” (W. S. Hoar and D. J. Randall, eds.), Vol. I,
     pp. 1-89. Academic Press, New York.
Hontela, A. (1984).Daily cycles ofserum gonadotropin hormone in fish. ?’runs. A m . Fish
     Soc. 113,458-466.
Hontela, A., and Lederis, K. (1985). Die1 variations in arginine vasotocin content of
    goldfish brain and pituitary: Effects of’photoperiod and pinealectomy. Gen. Conrp.
    Endocrinol. 57,397-404.
Hontela, A., and Peter, R. E. (1978). Daily cycles in serum gonadotropin levels in the
     goldfish: Effects of photoperiod, temperature, and sexual condition. Can.J. Zool. 56,
Hontela, A., and Peter, R. E. (1983).Characteristics and functional significance of daily
     cycles in serum GTH levels in goldfish. J. E x p . Zool.228,543-550.
Hontela, A., Rasmussen, J. B., KO, D., Lederis, K., and Chevalier, G. (1991).Arginine
     vasotocin, an osmoregulatory hormone, as a potential indicator of acid stress in fish.
     Can. J. Fish Ayuat. Sci. 48,232-242.
2 . CHEMICAL PROPERTIES            OF THE BLOOD                                            123
Hontela. A., Roy, Y., Van Coillie, R., Lederis, K., and Chevalier, G. (1989). Differential
     effects of low pH and aluminium on the caudal neurosecretory system of the brook
    trout, Saluelinzlsfontinalis. J. Fish Biol. 35,265-273.
Hopkins, C. L., and Sadler, W. A. (1987).Rhythmic changes in plasma thyroxine concen-
    trations in hatchery-reared chinook salmon, Oncorhynchus tshawytscha. New Zeal.
    J . Mar. Fresh. Res. 21,3 1-34.
Houston, A. H. (1985). Erythrocytic magnesium in freshwater fishes. Magnesium 4,
Houston, A. H. (1990). Blood and circulation. In “Methods for Fish Biology” (C. B.
    Schreck and P. B. Moyle, eds.),pp. 273-333. American Fisheries Society, Bethesda,
Houston, A. H., and Koss, T. F. (1982). Water-electrolyte balance in goldfish Carassius
    auratus, under constant and diurnally cycling temperature conditions. J. E x p . Biol.
Hunn, J . B., and Greer, I. E. (1990). Colorimetric and refractometric estimates of total
    plasma protein in striped bass, Morone suxatilis (Walbaum).J . Fish Biol. 36, 617-
Idler, D. R., and Tsuyuki, €1. (1958). Biochemical studies on sockeye salmon during
    spawning migration. I. Physical measurements, plasma cholesterol and electrolyte
    levels. Can.J. Biochem. Physiol. 36,783-791.
Idler, D. R., and Ng, T. B. (1979). Studies on two types of gonadotropins from both
     salmon and carp pituitaries. Gen. Comp. Endocrinol. 38,421-440.
Idler, D. R., and Ng, T. B. (1983). Teleost gonadotropins: Isolation, biocheniistry and
     function. In “Fish Physiology” (W. S. Hoar, D. J. Randall, and E. M. Donaldson,
     eds.), Vol. IX, Part A, pp. 187-221. Academic Press, New York.
Iigo, M., Kezuka, H., Aida, K., and Hanyu, I. (1991). Circadian rhythms of melatonin
     secretion from superfused goldfish (Carussius auratus) pineal glands in uitro. Gen.
     Comp. Endocrinol. 83, 152-158.
Iwama, G. K., McGeer, J. C., and Pawluk, M. P. (1989). The effects of five fish anaes-
    thetics on acid-base balance, heinatocrit, blood gases, cortisol, and adrenaline in
     rainbow trout. Can.J. Zool. 67,2065-2073.
Johnston, I. A,, and Bernard. L. M. (1983). Utilization of the ethanol pathway in carp
     following exposure to anoxia. J. Exp. Biol. 104,73-78.
Katz, Y., Dashow, L., and Epple, A. (1982).Circulating steroid hormones of anadromous
     sea lampreys under various experimental conditions. Gen. Comp. Endocrinol. 48,
Kaushik, S., and Luquet, P. (1977). Study of free amino acids in rainbow trout in relation
     to salinity changes. I. Blood free amino acids during starvation. Ann. H y d r o b i d . 8,
Kawauchi, H., Abe, K., Takahashi, A., Hirano, T., Hasegawa, S., Naito, N., and Nakai, Y.
     (1983).Isolation and properties of chum salmon prolactin. Gen. Comp. Endocrinol.
Kawauchi, H., Suzuki, K., Itoh, H., Swanson, P., Naito, N., Nagahama, Y., Nozaki, M.,
     Nakai, Y., and Itoh, S. (1989).The duality of teleost gonadotropins. Fish Physiol.
     Biocheni. 7,29-38.
Kezuka, H., Furukawa, K., Aida, K., and Hanyu, I. (1988).Daily cycles in plasma melato-
    nin levels under long or short photoperiod in the common carp, Cyprinus carpio.
    Gen. Comp. Endocrinol. 72,296-302.
Kime, D. E., and Larsen, L. 0. (1987). Effect ofgonadectomy and hypophysectomy on
    plasma steroid levels in male and female lampreys (Lampetra juoiatilis, L.). Gen.
    Comp. Endocrinol. 68, 189-196.
124                                         D. G. MCDONALD AND C. L. MILLIGAN

Klesius, P. H. (1990). Effect of size and temperature on the quantity of immunoglobulin
    in channel catfish, lctalurus punctatus. Vet. Immunol. Immunopath. 24, 187-195.
Kobayashi, K., Hara, A,, Takano, K., and Hirai, H. (1982).Studies on subunit components
    of immunoglobulin M from a bony fish, the chum salmon (Oncorhynchus keta).Mol.
    Immunol. 20,811-820.
Kobayashi, Y., Lederis, K., Rivier, J., KO, D., McMaster, D., and Poulin, P. (1986).
    Radioimmunoassays for fish tail neuropeptides. 11. Development of a specific and
     sensitive assay for and the occurrence of immunoreactive urotensin I1 in the central
    nervous system and blood of Catostomus commersoni. J Pharm. Meth. 15,322-333.
Kiihn, E. R., Corneillie, S., and Ollevier, F. (1986). Circadian variations in plasma
    osmolality, electrolytes, glucose, and cortisol in carp (Cyprinus carpio).Gen. Comp.
    Endocrinol. 61,459-468.
Laidley, C. W., and Leatherland, J. F. (1988a). Cohort sampling, anaesthesia and
     stocking-density effects on plasma cortisol, thyroid hormone, metabolite and ion
     levels in rainbow trout, Salmo gairdneri Richardson./. Fish Biol. 3 3 , 7 3 4 8 .
Laidley, C. W., and Leatherland, J. F. (198813). Circadian studies of plasma cortisol,
    thyroid hormone, protein, glucose and ion concentration, liver glycogen concentra-
    tion and liver and spleen weight in rainbow trout, Salmo gairdneri Richardson.
    Comp. Biochem. Physiol. 89A, 495-502.
Lamba, V. J., Goswami, S. V., and Sundararajm B. I. (1983). Circannual and circadian
    variations in plasma levels of steroids (cortisol, estradiol-l7a, estrone and testoster-
    one) correlated with the annual gonadal cycle in the catfish, Heteropneustesfossilis,
    Block. Gen. Comp. Endocrinol. 50,205-225.
Larsen, L. 0. (1990). The role of external and internal factors during the phase of
    reproduction in lampreys, with special regard to the interplay between gonadal and
    extragonadal (interrenal steroids). Fish PhysioZ. Biochena. 8,375-388.
Larsson, A., and Fange, R. (1977).Cholesterol and free fatty acids (FFA)in the blood of
    marine fish. Comp. Biochem. Physiol. 57B, 191-196.
Larsson, A., and Lewander, K. (1973). Metabolic effects of starvation in the European
    eel, Anguilla anguilla. Comp. Biochem. Physiol. 44A, 367-374.
Laurell, S., and Tibbling, G. (1967). Colorimetric micro-determination of free fatty acids
    in plasma. Clin. Chem. Acta 16,57-62.
Leatherland, J. F. (1988). Endocrine factors affecting thyroid economy of teleost fish.
    Amer. Zool. 28,319-328.
Leatherland, J. F., and Hilton, J. W. (1988). Thyroidal compensation in rainbow trout
    (Salmo gairdneri)fed canola meal. Fish Physiol. Biochem. 5, 199-207.
Leatherland, J . F., and Sonstegard, R. A. (1981).Thyroid function, pituitary structure and
    serum lipids in Great Lakes coho salmon, Oncorhynchus kisutch “jacks” compared
    with sexually immature spring salmon. J . Fish Biol. 18,643-654.
Lintlop, S . P., and Youson, J. H. (1983). Concentration of triiodothyronirie in the sera of
    the sea lamprey, Petromyzon marinus, and the brook lamprey, Lampetra Zamottenii,
    at various phases of the life cycles. Gen. Comp. Endocrinol. 49, 187-194.
Love, R. M. (1980).“The Chemical Biology of Fishes,”Vol. 2, Advances 1968-1977,943
    pp. Academic Press, London.
Low, P. S., and Bada, J . L. (1974).Bile pigments in the blood serum offish from the family
    Cottidae. Comp. Biochem. Physiol. 47A, 411-418.
Lutz, P. L. (1975). Adaptive and evolutionary aspects of the ionic content of fishes.
    Copeia 1975,369-373.
Mackenzie, D. J., and Randall, D. J. (1990). Does Amia caloa aestivate? Fish Physiol.
    Biochem. 8,147-158.
2. CHEMICAL PROPERTIES OF THE              BLOOD                                     125

MacKenzie, D. S., Warner, J., and Thomas, P. (1987).Thyroid-reproductive relationships
   in the channel catfish, Zctalurus punctatus: Evidence for estradiol-induced changes
   in plasma thyroid hormone binding. In “Reproductive Physiology of Fish” (D. R.
   Idler, L. W. Crim, and J . M. Walsh, eds.), pp. 205-211. Memorial University Press,
    St. John’s, Newfoundland.
Makos, B. K., and Youson, J. H. (1987). Serum levels ofbilirubin and biliverdin in the sea
   lamprey, Petromyzon marinus L., before and after biliary atresia. Comp. Biochem.
   Physiol. 87A, 761-764.
Manguni, C. P., Haswell, M. S., Johansen, K., and Towle, D. W. (1978). Inorganic ions
   and pH in the body fluids of Amazon animals. Can.J . Zool. 56,907-916.
Marchant, T. A., and Peter, R. E. (1986). Seasonal variations in body growth rates and
   circulating levels of growth hormone in the goldfish, Carassius auratus.J.Exp. Zool.
MBrquez, E. D. (1976).A comparison of glutamic-oxalacetate transaminase, lactate dehy-
   drogenase and creatine phosphokinase activities in non-spawning, prespawning and
   spawning pink salmon. Comp. Biochem. Physiol. 45B, 121-123.
Martin, B. (1975). Steroid-protein interactions in nonmammalian vertebrates. Gen.
   Comp. Endocrinol. 25,42-51.
Mathers, J. S., and Beamish, F. W. H. (1974). Changes in serum osmotic and ionic
   concentration in landlocked Petromyzon marinus. Comp. Biochem. Physiol. 49A,
Mayer-Gostan, N., Flik, G., and Pang, P. K. T. (1992).An enzyme-linked immunosorbent
    assay for stanniocalcin, the major hypocalcemic hormone in fish. Gen. Comp. Endo-
    crinol. (In press.)
Mazeaud, M. M., and Mazeaud, F. (1981). Adrenergic responses to stress in fish. In
    “Stress in Fish” (A. D. Pickering, ed.), pp. 49-75. Academic Press, New York.
McDonald, D. G. (1983a). The interaction of calcium and low pH on the physiology of
   the rainbow trout, Salmo gairdneri. I. Branchial and renal net ion and H + fluxes.
   J. E x p . Biol. 102, 123-140.
McDonald, D. G. (1983b). The effects of H + upon the gills of freshwater fish. Can. J.
     ZOO^. 61,691-703.
McDonald, D. G., and Goldstein, M. D. (1992). Unpublished observations.
McDonald, D. G., Reader, J . P., and Dalziel, T. R. K. (1989). The combined effects of pH
   and trace metals on fish ionoregulation. In “Acid toxicity and Aquatic Animals” (R.
    Morris, E. W. Taylor, D. J. A. Brown, and J. A. Brown ed.), pp. 221-242. Cambridge
    University Press, Cambridge.
McDonald, D. G., and Robinson, J. L. (1992). Unpublished observations.
McFarland, W. N., Pough, F. H., Cade, T. J., and Heiser, J.B. (1979).Vertebrate Life. 875
    pp. Macmillan Publishing, New York.
McKay, M. C., Lee, R. F., and Smith, M. A. K. (1985).The characterization ofthe plasma
    lipoproteins of the channel catfish Ictalurus punctatus. Physiol. Zool. 58,693-704.
Meier, A. H. (1984). Temporal snyergism of circadian neuroendocrine oscillations regu-
    lates seasonal conditions in the Gulf killifish. Trans. Am. Fish. SOC. 113,422-431.
Miller, W. R., Hendricks, A. C., and Cairns, J. (1983). Normal ranges for diagnostically
    important hematological and blood chemistry characteristics of rainbow trout
    (Salmo gairdneri). C a n . ] . Fish.Aquat. Sci. 40,420-425.
Milligan, C. L., and Wood, C. M. (1982). Disturbances in hematology, fluid volume
    distribution and cardiovascular function associated with low environmental pH in
    the rainbow trout, Salmo gairdneri.]. E x p . Biol. 99,397-415.
Milligan, C. L., and Wood, C. M. (1986). Intracellular and extracellular acid-base status
126                                         D. G. MCDONALD AND C. L. MILLIGAN

     and H + exchange with the environment after exhaustive exercise in the rainbow
     trout. J . E x p . Biol. 123, 92-121.
Milligan, C. L., and Wood, C. M. (1987). Regulation of blood oxygen transport and red
     cell ph, after exhaustive activity in rainbow trout (Salmo gairdneri) and starry
     flounder (Platichthys stellatus).J. E x p . Biol. 133,262-282.
hlills, G. L., and Taylaur C. E. (1973). The distribution and composition of serum
     lipoproteins in the coelacanth (Latimeria).Cornp. Biochem Physiol. 44B, 1235-
Mills, C. L., Taylaur, C. E., Chapman, M.J., and Forster, G. R. (1977).Characterization of
      serum lipoproteins of the shark Centrophorus squamus. Biochem. J . 163,455-465.
Mommsen, T. P., French, C. J., and Hochachka, P. W. (1980). Sites and patterns of
      protein and amino acid utilization during the spawning migration of salmon. Can. J .
      ZOO^. 58,1785-1799.
Mommsen, T. P., and Walsh, P. J. (1989). Evolution ofurea synthesis in vertebrates: The
      piscine connection. Science 243,72-75.
Mommsen, T. P., and Walsh, P. J. (1991). Urea synthesis in fishes: Evolutionary and
      biochemical perspectives. In “Biochemistry and Molecular Biology of Fishes”
      (P. W. Hochachka and T. P. Mommsen, eds.). Elsevier. Amsterdam, New York.
Moon, T. W. (1983).Metabolic reserves and enzyme activities with food deprivation in
      immature American eel, Anguilla rostrata (Le Suer). Can. J . Zool. 61,802-811.
Xloon, T . W. (1988). Adaptation, constraint, and the functions of the gluconeogenic
      pathway. Con. J. Zool. 66, 1059-1068.
Moon, T. W., Foster, G. D., and Plisetskaya, E. M . (1989). Changes in pepticle hornmones
      and liver enzymes in the rainbow trout deprived offood for 6 weeks. C a n . ] .2001.67,
Iludge, S. M., and Davenport, J. (1986). Serum pigmentation in Cyclopterus l u m p u s L.
     J . Fish Biol. 29,737-745.
Llurai, T., and Ogata, H. (1990). Changes in free amino acid levels in various tissues of
      common carp in response to insulin injection followed by force-feeding an amino
      acid diet. Br.1. Nutr. 120,711-718.
Nakano, T., and Tomlinson, K. (1967). Catecholamine and carbohydrate concentrations
      in rainbow trout (Salmo gairdneri) in relation to physical disturbance. J. Fish Res.
      Board Can. 24,1701-1715.
Nicoll, C. S., Wilson, S. W., Nishioka, R. S., and Bern, H. A. (1981). Blood and pituitary
      prolactin levels in tilapia (Sarotherodon mossanibicus: Teleostei) from different
      salinities as measured by homologous RIA. Gen. C o m p . Endocrind. 44,365-373.
Nikinmaa, kl.,and Heustis, W. H . (1984).Adrenergic swelling of nucleated erythrocytes:
      Cellular mechanisms in a bird, domestic goose, and two teleosts, striped bass and
      rainbow trout. J . E r p . B i d . 113, 215-224.
Nishioka, R. S., Kelley, K. M., and Bern, H. A. (1988).Control ofprolactin and growth
      hormone secretion in teleost fishes. Zool. Sci. 5,267-280.
Nose, T. (1972). Changes in pattern of free plasma amino acids in rainbow trout after
      feeding. Bull. Freslzwut. Fish Res. Lab. 22, 137-144.
Ogasawara, T., Hirano, T., Akiyanam, T., Arai, S., and Tagawa, M. (1989). Changes in
      plasma prolactin and growth hormone concentrations during freshwater adaption of
      juvenile chum salmon (Oncorhynchus keta) reared in seawater for a prolonged
      period. Fish Physiol. Biochem. 7,309-313.
Olsen, N. J., and J@rgensen, E. (1986). Quantification of serum immunoglobulin in
      rainbow trout, Salmo guirdneri, under various environmental conditions. Dis.
      Aquat. Organisms 1, 183-192.
Orimo, H., Yamauchi, H., Ohyama, T., Matuo, M., and Otani, M. (1977). Radioinmmuno-
2. CHEMICAL      PROPERTIES OF THE BLOOD                                              127

    assay and immunological properties of eel calcitonin. Gen. Comp. Endocrinol. 31,
Pang, P. K. T., and Pang, R. K. (1986).Hormones and calcium regulation in Fundulus
    heteroclitus. Amer. Zool. 26, 225-234.
Pearson, M. P., and Stevens, E. D. (1991). Size and hematological impact ofthe splenic
    erythrocyte reservoir in rainbow trout, Oncorhynchus mykiss. Fish Physiol. Bio-
    chem. 9,39-50.
Perlman, D., and Goldstein, L. (1988).Nitrogen metabolism. In “Physiology of Elasmo-
    branch Fishes” (T. J. Shuttleworth, ed.), pp. 253-275. Berlin, Springer-Verlag.
Perrot, M . N., Carrick, S., and Balment, R. J. (1986). Pituitary and plasma arginine
    vasotocin levels in euryhaline fish. J. Endocrinol. lOB(Supp1)215. [Abstract]
Perry, S. F., and Vermette, 41.G. (1987). The effects of prolonged epinephrine infusion
    on the physiology of the rainbow trout, Salmo gairdneri. I. Blood respiratory, acid-
    base and ionic status. J . E x p . Biol. 128,235-253.
Perry, S. F., Kinkhead, R., Gallaugher, P., and Randall, D. J. (1989). Evidence that
    hypoxemia promotes catecholamine release during hypercapnic acidosis in rainbow
    trout (Salmo gairdneri). Resp. Physiol. 77,351-364.
Peter, R. E. (1981). Gonadotropin secretion during reproductive cycles in teleosts:
    Influences of environmental factors. Gen. Comp. Endocrinol. 45,294-305.
Peter, R. E., and Crim, L. W. (1979). Reproductive endocrinology of fishes: Gonadal
    cycles and gonadotropin in teleosts. Ann. Rev. Physiol. 41,323-335.
Peter, R. E., Hontela, A., Cook, A. F., and Paulencu, C. R. (1978). Daily cycles in serum
    cortisol levels in the goldfish: Effects of photoperiod, temperature, and sexual condi-
    tion. C a n . ] . Zool. 56,2443-2448.
Peter, R. E., Sokowska, M., Truscott, B., Walsh, J., and Idler, D. R. (1984). Secretion of
    progestogens during induced ovulation in goldfish. Can. 1.Zool. 62,231-242.
Peterson, M . S., and Gilmore, R. G. (1988).Hematocrit, osmolality, and ion concentration
    in fishes: Consideration of circadian patterns in the experimental design. 1. E x p .
    Mar. Biol. Ecol. 121,73-78.
Peuler, J. D., and Johnson, G. A. (1977). Simultaneous single isotope radioenzymatic
    assay of plasma norepinephrine, epinephrine and dopamine. Lije Sci. 21,625-636.
Pickering, A. D., and Christie, P. (1981).Changes in the concentrations of plasma cortisol
    and thyroxine during sexual maturation of the hatchery-reared brown trout, Salmo
    trutta L. Cen. Comp. Endocrinol. 44,487-496.
Pickering, A. D., and Morris, R. (1970). Osmoregulation of Lampetra fluuiatilis and
    Petromyzon marinus (Cyclostomata) in hypertonic solutions. J. E x p . Biol. 53,231-
Pickering, A. D., and Pottinger, T. G . (1983). Seasonal and die1 changes in plasma
    cortisol levels of the brown trout, Salmo trutta L. Gen. Comp. Endocrinol. 49,
Pickering, A. D., Pottinger, T. G., and Christie, P. (1982). Recovery of the brown trout,
    Salmo trutta L., from acute handling stress: A time-course study. J. Fish Biol. 20,
Plakas, S. M., Katayama, T., Tanaka, Y., and Deshimaru, 0.(1980).Changes in the levels
    of circulating plasma free amino acids of carp (Cyrpinus carpio) after feeding a
    protein and an amino acid diet of similar composition. Aquaculture 21,307-322.
Planas, J., GutiCrrez, J., Fernandez, J., Carrillo, M., and Canals, P. (1990). Annual and
    daily variations of plasma cortisol in sea bass, Dicentrachus labrax L. Aquaculture
Plisetskaya, E. (1980). Fatty acid levels in the blood of cyclostomes and fish. Enu. Biol.
    Fish. 5,273-290.
128                                       D. G. MCDONALD AND C. L. MILLIGAN

Plisetskaya, E. M. (1989). Physiology of fish endocrine pancreas. Fish Physiol. Biochem.
Plisetskaya, E. M. (1990).Recent studies of fish pancreatic hormones: Selected topics.
    Zool. Sci. 7,335-353.
Plisetskaya, E. M., Dickhoff, W. W., Paquette, T. L., and Gorbman, A. (1986a).The assay
    of salmon insulin in homologous radioimmunoassay. Fish Physiol. Biochem. 1,
Plisetskaya, E. M., Ottolenghi, T., Sheridan, M. A., Mommsen, T., and Gorbman, A.
    (1989). Metabolic effects of salmon glucagon and glucagon-like peptide in coho and
    chinook salmon. Gen. Comp. Endocrinol. 73,205-216.
Plisetskaya, E. M., Pollock, H. G., Rouse, J. B., Hamilton, J. W., Kimmel, J. R., An-
    drews, P. C., and Gorbman, A. (1986b). Characterization of coho salmon (Oncor-
    hynchus kisutch) islet somatostatins. Gen. Comp. Endocrinol. 63, 252-
Plisetskaya, E. M., Woo, N. Y. S., and Murat, J. C. (1983).Thyroid hormones in cyclos-
     tomes and fish and their role in regulation of intermediary metabolism. Comp.
     Biochem. Physiol. 74A, 179-187.
Prunet, P., Boeuf, G., Bolton, J. P., and Young, G. (1989). Smoltification and seawater
     adaptation in Atlantic salmon (Salmo salar): Plasma prolactin, growth hormone, and
     thyroid hormones. Gen. Comp. Endocrinol. 74,355-364.
Racicot, J.-G., Gaudet, M., and Leray, C. (1975). Blood and liver enzymes in rainbow
     trout (Salmo gairdneri Rich.) with emphasis on their diagnostic use: Study of CC14
     toxicity and a case of Aeromonsas infection. J . Fish Biol. 7,825-835.
Railo, E., Nikinmaa, M., and Soivio, A. (1985).Effects of sampiing on blood parameters
     in the rainbow trout, Salmo gairdneri Richardson. J . Fish Biol. 26,725-732.
Rance, T. A., Baker, B. I., and Webley, G. (1982).Variations in plasmacortisol concentra-
     tions over a 24-hour period in the rainbow trout Salmo gairdneri. C e n . Comp.
     Endocrinol. 48,269-274.
Rand-Weaver, M., Carragher, J. C., and Sumpter, J. P. (1989). Development of cod
      growth hormone radioimmunoassay. In “Proceedings, 11th International Sym-
      posium of Comparative Endocrinology, Malaga.” Abstract P-284.
Randall, D. J., and Wright, P. A. (1987).Ammonia distribution and excretion in fish. Fish
     Physiol. Biochem. 3, 107-120.
 Robertson, D. H., Krupp, M. A., Favour, C. B., Hanes, S., and Thomas, S. F. (1961).
      Physiological changes occurring in the blood of Pacific salmon (Oncorhynchus
      tshawytscha) accompanying sexual maturation and spawning. Endocrinology 68,
 Robertson, J. D. (1984).The composition of blood plasma and parietal muscel of Oslo
      Fjord eels (Anguilla anguilla (L.))and the river lamprey (Lampetrafluuiatilis (L.)).
      Comp. Biochem. Physiol. 77A, 431-439.
 Robertson, J. D. (1989).Osmotic constituents of the blood plasma and parietal muscle of‘
      Scyliorhinus canicula (L). Comp. Biochem. Physiol. 93A, 799-805.
 Robinson, J. S., and Mead, J. F. (1973). Lipid absorption and deposition in rainbow trout
      (Salmo gairdneri). Can. J. Biochem. 51, 1050-1058.
 Rodriques, K. T., and Sumpter, J. P. (1984).The radioimmunoassay of a-melanocyte
      stimulating hormone and endorphin in rainbow trout (Salmo gairdneri) and the
      effect ofblinding on the plasma levels ofthese peptides. Gen. Comp. Endocrinol. 54,
 Saha, N., and Ratha, B. K. (1989). Comparative study of ureogenesis in freshwater,
      air-breathing teleosts. J. Exp. Zool. 252, 1-8.
2.   CHEMICAL PROPERTIES OF THE BLOOD                                                 129

Sakamoto, T., Ogasawara, T., and Hirano, T. (1990). Growth hormone kinetics during
    adaptation to a hyperosmotic environment in rainbow trout. J. Comp. Physiol. B . 160,
Sandnes, K., Lie, 2 . and Waagb2, R. (1988). Normal ranges of some blood chemistry
    parameters in adult farmed Atlantic salmon, Salmo salar. J . Fish Biol. 32,129-136.
Santulli, A., Curatolo, A., Modica, A., and D’Amelio, V. (1988). Time-course changes of
    plasma lipid levels and lipoprotein pattern after feeding in cultured sea bass, Dicen-
    trurchus labrax L. J. Fish Biol. 32,859-867.
Sargent, J. R. (1976). The structure, metabolism and function of lipids in marine organ-
    isms. In “Biochemical and Biophysical Perspectives in Marine Biology” (D. C.
    Malins and J. R. Sargent, eds.), Vol. 3, pp. 149-212. Academic Press, London.
Sauer, D. M., and Haider, G. (1977). Enzyme activities in the serum of rainbow trout,
    Salmo gairdneri Richardson: The effects of water temperature. J. Fish Biol. 11,
    605-6 12.
Sauer, D. M., and Haider, G. (1979). Enzyme activities in the serum of rainbow trout,
    Salmo gairdneri Richardson: The effects of nutritional status and salinity. J. Fish
    Biol. 14,407-412.
Sawyer, W. H. (1987).Neurohypophysial hormones and osmoregulatory adaptations. In
    “Comparative Physiology of Environmental Adaptations” ( J. Kirsch and B. Lahlou,
    eds.), pp. 76-82. Karger, Basel.
Schlotfeldt, H. J. (1975). Hematological studies of cultured rainbow trout: Seasonal
    variations of the blood constituents. J. Fish Biol. 9,605-611.
Schwalme, K., and McKay, W. C. (1985). The influence of angling-induced exercise on
    the carbohydrate metabolism of norther pike (Esox lucius L).J. Comp. Physiol. 15GB,
Scott, A. P., and Sumpter, J. P. (1983).A comparison ofthe female reproductive cycles of
    autumn-spawning and winter-spawning strains of rainbow trout (Salmo gairdneri
    Richardson). Cen. Comp. Endocrinol. 52,79-85.
Scott, A. P., Sheldrick, E. L., and Flint, P. F. (1982).Measurement of 17a, SOP-dihydroxy-
    4-pregnen-3-one in plasma of trout (Salmo gairdneri Richardson): Seasonal changes
    and response to salmon pituitary extract. Gen. Comp. Endocrinol. 49, 128-134.
Sheridan, M. A. (1988). Lipid dynamics in fish: Aspects of absorption, transportation,
    deposition and mobilization. Comp. Biochem. Physiol. 90B, 679-690.
Sheridan, M. A. (1989).Alterations in lipid metabolism accompanying smoltification and
    seawater adaptation of salmonid fish. Aquaculture 82,191-203.
Shoubridge, E. A., and Hochachka, P. W. (1980). Ethanol: Novel end product of verte-
    brate anaerobic metabolism. Science 209,308-309.
Sidon, E. W., and Youson, J . H. (1983). Morphological changes in the liver of the sea
    lamprey Petromyzon marinus L. during metamorphosis. I. Atresia of the bile ducts.
   J . Morph. 177, 109-124.
Singer, T. D., and Ballantyne, J. S. (1991). Metabolic organization ofa primitive fish, the
    bowfin Amia calva L. Can. J . Fish Aquat. Sci. 48,611-618.
Singer, T. D., Mahadevappa, V. G., and Ballantyne, J. S. (1990). Aspects of the energy
    metabolism of the lake sturgeon, Acipenser fuloescens, with special emphasis on
    lipid and ketone body metabolism. Can.J. Fish Aquut. Sci. 747,873-881.
Skibelli, V., Andersen, @., and Gautvik, K. M. (1990).Purification and characterization of
    Atlantic salmon growth hormone. Gen. Comp. Endocrinol. 80,333-344.
Sokolowska, W. P., and Bieniarz, K. (1981). Effects of stress on the circadian rhythmicity
    of hypothalamic catecholamine content in the eel (Anguilla anguilla L). Comp.
    Biochem. Physiol. 70C, 135-138.
130                                         1 . G. MCDONALD AND C. L. MILLIGAN

Specker, J. L. (1988). Preadaptive role ofthyroid hormones in larval andjuvenile salmon:
    Growth, the gut and evolutionary considerations. Amer. Zool. 28,337-349.
Spieler, R. E. (1979). Diel rhythms of circulating prolactin, cortisol, thyroxine, and
    triiodothyronine in fishes: A review. Reo. Can. Biol. 38,301-315.
Spieler, R. E., and Noeske, T. A. (1979). Diel variations in circulating levels oftriiodothy-
    ronine and thyroxine in goldfish, Carassius auratus. Can.J . Zool. 57,665-669.
Stacey, N. E., Cook, A. F., and Peter, R. E. (1979). Ovulatory surge of gonadotropin in the
    goldfish, Carassius auratus. Gen. Comp. Endocrinol. 37,246-249.
Stevens, E. D. (1968).The effect ofexercise on the distribution ofblood to various organs
    in rainbow trout. Comp. Biochem. Physiol. 25,615-625.
Suarez, R. K., and Mommsen, T. P. (1987). Gluconeogenesis in teleost fishes. Can. J .
    Zool. 65, 1869-1882.
Suess, U., Lawrence, J., KO, D., and Lederis, K. (1986). Radioimmunoassays for fish tail
    neuropeptides: I. Development of assay and measurement of immunoreactive uro-
    tensin I in Catostomus commersoni brain, pituitary, and plasma./. Pharm. Meth. 15,
Summerfelt, R. C., and Smith, L. S. (1990). Anesthesia, surgery and related techniques.
     In “Methods for Fish Biology” (C. B. Schreck and P. B. Moyle, eds.), pp. 213-272.
    American Fisheries Society, Bethesda, Maryland.
Sumpter, J. P., and Donaldson, E. M. (1986). The development and validation of a
    radioimmunoassay to measure plasma ACTH levels in salmonid fishes. Gen. Cornp.
    Endocrinol. 62,367-376.
Sumpter, J. P., Dye, H. M., and Benfey, T. J. (1986). The effects of stress on plasma
    ACTH, a-MSH and cortisol levels in salmonid fishes. Gen. Comp. Endocrinol. 62,
Sumpter, J. P., Pickering, A. D., and Pottinger, T. G. (1985). Stress-induced elevation of
    plasma a-MSH and endorphin in brown trout, Salmo trutta L. Gen. Comp. Endocri-
     nol. 59,257-285.
Suzuki, K., Kanamori, A., Nagahama, Y., and Kawauchi, H. (1988). Development of
     salmon GtH I and GtH I1 radioimmunoassays. Gen. Comp. Endocrinol. 71,459-467.
Suzuki, R., and Hirano, T. (1991). Development of a homologous radioimmunoassay for
     eel prolactin. Gen. Comp. Endocrinol. 81,403-409.
Swanson, P., Bernard, M., Nozaki, M., Suzuki, K., Kawauchi, H., and Dickhoff, W. W.
     (1987). Gonadotropins I and I1 in juvenile coho salmon. Fish Physiol. Biochem. 7,
     169- 176.
Sweeting, R. M., and McKeown, B. A. (1987). Growth hormone and seawater adaptation
     in coho salmon, Onchorhynchus kisutch. Comp. Biochem. Physiol. 88A, 147-151.
Talmage, R. V., Grubb, S. A. Norimatsu, H., and VanderWeil, C . J. (1980). Evidence for
     an important physiological role for calcitonin. Proc. Natl. Acad. Sci. USA 77,609-
Tang, Y., and Boutilier, R. G. (1988). Correlation between catecholamine release and
     degree of acidotic stress in trout. Am. J . Physiol. 255, R395-R399.
Taylor, M. H. (1984). Lunar synchronization of fish reproduction. Trans. Am. Fish Soc.
Thomas, S., and Perry, S. F. (1992). Control and consequences ofadrenergic activation of
     red blood cell Na+/H+exchange on blood oxygen and carbon dioxide transport in
     fish./. Erp. Zool. (In press).
Thorpe, A,, and Ince, B. W. (1976). Plasma insulin levels in teleosts determined by a
     charcoal-seperation radioimmunoassay technique. Gen. Comp. Endocrinol. 30,
2. CHEMICAL PROPERTIES          OF THE BLOOD                                         131

Thorpe, J. E., McConway, 41.G., Miles, M. S., and Muir, J. S . (1987). Die1 and seasonal
    changes in resting plasma cortisol levels in juvenile Atlantic salmon, Salmo salar L.
    Gen. Comp. Endocrinol. 65, 19-22.
Thorson, T. B., Cowan, C. M., and Watson, D. E. (1973).Body fluid solutes ofjuveniles
    and adults of the euryhaline bull shark, Carcharhinus leucas, from freshwater and
    saline environments. Physiol. Zool. 46, 29-42.
Thurston, R. V., Russo, R. C., Luedtke, R. J., Smith, C. E., Meyn, E. L., Shakoumakos, C.,
    Wang, K. C., and Brown, C. J . D. (1984).Chronic toxicity of ammonia to rainbow
    trout. Trans. Am. Fish SOC. 113,56-73.
Timoshina, L. A., and Shabalina, A. A. (1972).Effect of starvation on the dynamics of
    concentration of amino acid and free fatty acid in rainbow trout. Hydrobiol. I . 8,
Toews, D. P., and Hickman, C. P. (1969).The effect of cycling temperatures on electro-
    lyte balance in skeletal muscle and plasma of rainbow trout Salmo gairdneri. Comp.
    Biochem. Physiol. 29,905-918.
Tufts, B. L., and Boutilier, R. G. (1989). The absence of rapid chlorid/bicarbonate
    exchange in lamprey erythrocytes: Implications for COZ transport and ion distribu-
    tions between plasma and erythrocytes in the blood of Petromyzon marinus. J . E x p .
    Biol. 144,565-576.
Turner, J . D., Wood, C. M., and Clark, D. (1983a).Lactate and proton dynamics in the
    rainbow trout (Sulmo gciirdneri). J. E x p . Biol. 104,247-268.
Turner, J. D., Wood, C. M., and Hdbe, H. (1983b).Physiological consequences ofsevere
    exercise in the inactive benthic flathead sole (Hippoglossoides elussodon): A com-
    parison with the active pelagic rainbow trout (Salmo gairdneri). J. E x p . Biol. 104,
Umminger, B. L. (1977).Relation of whole blood sugar concentrations in vertebrates to
    standard metabolic rate. Cornp. Biochem. Physiol. 56A, 457-460.
Urist, M. R. Uyeno, S., King, E., Okada, M., and Applegate, S. (1972). Calcium and
    phosphorous in the skeleton and blood of the lungfish Lepidosiren parudoxa, with
    comment on humoral factors in calcium homeostasis in the Osteichthyes. Cornp.
    Biochem. Physiol. 42A, 393-408.
van den Thillart, G., and Verbeek, R. (1982). Substrates for anaerobic COz-production by
    goldfish, Carassius auratus (L): Decarboxylation of 14C-labelled metabolites.
    1. Comp. Physiol. 149, 75-81.
van der Boon, J., Verhagen, M.A. W. H., van den Thillart, G. E. E. J. and Addink, A. D. F.
     (1991). Free amino acids in whole blood and plasma of two cyprinids, Cyprinus
     curpio and Carassius auratus (Cyprinidae: Teleostei). Comp. Biochem. Physiol.
     99A, 391-399.
van der Velden, J. A., Kolar, Z., Flik, G., de Goeij, J. J. M., and Wendelaar Bonga,
     S . E. (1989). Magnesium distribution in freshwater tilapia. Magnesium Bull. 11,
 van Pilsum, J. F., Stephens, G. C., and Taylor, D. (1972). Distribution of creatine,
     guanidinoacetate and the enzymes for their biosynthesis in the animal kingdom.
     Implications for phylogeny. Biochem. I . 126,325-345.
van Waarde, A. (1983). Aerobic and anaerobic ammonia production by fish. Comp.
     Biochem. Physiol. 74B, 675-684.
Venkatesh, B., Tan, C. H., and Lam, T. J. (1989). Blood steroids in the goldfish: Measure-
     ment of six ovarian steroids in small volumes of serum by reverse-phase high-
     performance liquid chromatography and radioimmunoassay. Gen. Comp. Endocri-
     nol. 76,397-407.
132                                       D . G. MCDONALD AND C . L. MILLIGAN

Wagner, G. F., and McKeown, B. A. (1986).Development of a salmon growth hormone
   radioimmunoassay. Gen. Comp. Endocrinol. 62,452-458.
Walsh, P. J. (1989).An in oitro model o f post-exercise hepatic gluconeogenesis in the
   gulf toadfish Opsanus beta. J . E x p . Biol. 147,393-406.
Walsh, P. J., Danulat, E., and Mommsen, T. P. (1990). Variation in urea excretion in the
   gulf toadfish Opsanus betu. Mur. Biol. 106,323-328.
Walton, M. J., and Wilson, R. P. (1986). Postprandial changes in plasma and liver free
   amino acids of rainbow trout fed complete diets containing casein. Aquaculture 51,
Watts, R. L., and Watts, D. C. (1974). Nitrogen metabolism in fishes. In “Chemical
   Zoology” (M. Florkin and B. T. Scheer, eds.), Vol. 8, pp. 369-446. Academic Press,
   New York.
Wedermeyer, G. A,, and McLeay, D. J. (1981). Methods for determining the tolerance of‘
   fishes to environmental stressors. In “Stress and Fish” (A. D. Pickering, ed.), pp.
   247-275. Academic Press, London.
Wells, R. M . G., McIntyre, R. H., Morgan, A. K., and Davie, P. S. (1986).Physiological
   stress responses in big gamefish after capture: Observations on plasma chemistry
   and blood factors. Comp. Biochem. Physiol. 84A, 565-571.
White, A., Fletcher, T. C., and Pope, J. A. (1986). Seasonal changes in serum lipid
   composition of the plaice Pleuronectes platessu L. J . Fish Biol. 28,595-606.
White, B. A., and Henderson, N. E. (1977).Annual variations in the circulating levels of
   thyroid hormones in brook trout, Saloelinus~ontinalis, measured by radioimmu-
   noassay. Can. J . Zool. 55,475-481.
Wilkes, P. R. H., Walker, R. L., McDonald, D. G., and Wood, C. M. (1981).Respiratory,
    ventilatory, acid-base, and iono-regulatory physiology of the white sucker, Catosto-
    mils commersoni: The influence of hyperoxia. J. Exp. Biol. 91,239-247.
Wilson, R. P., Gatlin, D. M. 111, and Poe, W. E. (1985). Postprandial changes in serum
    amino acids of channel catfish fed diets containing different levels of protein and
    energy. Aquaculture 49, 101-110.
Wingfield, J . C. (1988).Sex-steroid binding proteins in vertebrate blood. In “Hormones,
   Adaptation and Evolution” (S. Ishiii, T. Hirano, and M. Wada, eds.), pp. 135-144.
   Japan Sci. Soc. Press, Tokyo/Springer-Verlag, Berlin.
Wissing, J . , and Zebe, E. (1988).The anaerobic metabolism ofthe bitterling Rliodeus
    urnurus (Cyprinidae, Teleostei). Comp. Biochem. Physiol. 89B, 299-303.
Wood, C. M., and Perry, S. F. (1985). Respiratory, circulatory and metabolic adjustments
    to exercise in fish. In “Comparative Physiology and Biochemistry, Current Topics
    and Trends” (R. Gilles, ed.), Vol. A, pp. 1-22. Springer-Verlag, Berlin.
Wood, C. M., Perry, S. F., Wright, P. A,, Bergman, H. L., and Randall, D. J. (1989).
    Ammonia and urea dynamics in the Lake Magadi tilapia, a nreotelic teleost fish
    adapted to an extremely alkaline environment. Respir. Physiol. 77, 1-20.
Woodward, J. J . , and Strange, R. J . (1987).Physiological stress responses in wild and
    hatchery-reared rainbow trout. Trans. A n . Fish Soc. 116,574-579.
Woodward, J. J, (1982).Plasma catecholamines in resting rainbow trout (Salmogairdneri
    Richardson) by high pressure liquid chromatography. J . Fish Biol. 21,429-432.
Wright, G. M., and Youson, J . H. (1977).Serum thyroxine concentrations in larval and
    metamorphosing anadromous sea lamprey, Petromyzon marinus L. J . E x p . Zool. 57,
    1808- 1817.
Wright, P. A., and Wood, C. M. (1985).An analysis ofbranchial ammoniaexcretion in the
    freshwater rainbow trout: Effects of environmental pH change and sodium uptake
    b1ockade.J. E x p . B i d . 114,329-353.
2.   CHEMICAL PROPERTIES OF THE BLOOD                                                   133

Wright, P. A., Randall, D. j., and Wood, C. M . (1988). The distribution of ammonia and
    H between tissue compartments in lemon sole (Parophrys netulus) at rest, during

    hypercapnia and following exercise. J . E x p . Biol. 136, 149-175.
Yada, T., Takahashi, K., and Hirano, T. (1991). Seasonal changes in seawater adaptability
    and plasma levels of prolactin and growth hormone in landlocked sockeye salmon
    (Oncorhynchus nerko) and amago salmon (0.rhodurus). Gen. Comp. Endocrinol.
Yamada, S., Tanaka, Y., Katayama, T., Sameshima, M., and Simpson, K. L. (1982).Plasma
    amino acid changes in Tilapia nilotica fed a casein and a corresponding free amino
    acid diet. Bull.Jpn. Soc. Sci. Fish 48, 1783-1787.
Yamaguchi, K. (1973).Studies on a blue-green serum pigment of eel. VI. Physiological
    role as a lipid transporter. BuZl.Jap. Soc. Scient. Fish 49, 191-196.
Yamauchi, H., Orimo, H., Yamaguchi, K., Takano, K., and Takahashi, H. (1978). In-
    creased calcitonin levels during ovarian development in the eel, Anguillajaponica.
    Gen. Comp. Endocrinol. 36,526-529.
Young, G . , Bjornsson, B. T., Prunet, P., Lin, R. J., and Bern, H . A. (1989).Smoltification
    and seawater adaptation in coho salmon (Oncorhynchus kisutch): Plasma prolactin,
    growth hormone, thyroid hormones, and cortisol. Gen. Comp. Endocrinol. 74,335-
Youson, J. H. (1988).First metamorphosis. In “Fish Physiology” (W. S. Hoar and D. J.
    Randall, eds.), Vol. XI, Part B, pp. 135-196. Academic Press, New York.
Yu, J. Y. L., and Shen, S. T. (1989). Isolation ofpituitary glycoprotein gonadotropins from
    the grass carp (Ctenopharyngodon idell). Fish Physiol. Biochem. 7, 177-183.
Zammit, V. A., and Newsholme, E. A. (1979). Activities of enzymes of fat and ketone-
    body metabolism and effects of starvation on blood concentrations of glucose and fat
    fuels in teleost and elasmobranch fish. Biochem. J . 184,313-322.
Zohar, Y., and Billard, R. (1984).Annual and daily changes in plasma gonadotropin and
    sex steroids in relation to teleost gonad cycles. Trans. Am. Fish Soc. 113,444-451.
This Page Intentionally Left Blank

Indiana University School of Medicine, South Bend Center
University of Notre Dame
Notre Dame, Indiana

  1. Introduction
 11. Fluid Compartments
     A. Total Body Water
     B. Intracellular Water
     C. Extracellular Water
     D. Blood Volume
111. Renin-Angiotensin System
     A. Components of the Renin-Angiotensin System
      B. Occurrence and Distribution in Fish
     C. Stimulus for Activation of the Renin-Angiotensin System
     D. Effects of Angiotensins
 IV. Kallikrein-Kinin System
      A. Components of the Kallikrein-Kinin System
      B. Occurrence and Distribution in Fish
      C. Effects of Kinins
  V. Atrial Natriuretic Peptides
      A. Structure of Natriuretic Peptides
      B. Distribution in Fish
      C. Physiological Significance of Natriuretic Peptides
VI. Summary

FISH PHYSIOLOGY, VOL. XIIB                            Copyright 0 1992 by Academic Press, Inc.
                                                 All rights of reproduction in any form reserved.
136                                                 KENNETH R . OLSON


   Osmoregulation has long been a fascinating subject for biologists
and in fish it has probably been more extensively studied than any
other physiological or biochemical process. Volume regulation, on the
other hand, even though it is inexorably interwoven with osmoregula-
tory activities, has been less frequently examined and is still poorly
understood. Why volume regulation has received so little attention is
not clear. Certainly technical difficulty in measuring and monitoring
the various fluid compartments has been a contributory factor. Perhaps
more importantly, however, is the fact that volume regulation in most
vertebrates is poorly understood. The intent of this chapter is to pro-
vide a conceptual basis for volume regulation in fish and to evaluate
the role of the renin-angiotensin system (RAS), kallikrein-kinin sys-
tem (KKS), and atrial natriuretic peptides (ANP) in this process. Ad-
mittedly there is considerable bias in this approach because many of
the concepts regarding fluid compartments and volume regulation
have been developed in mammalian models and may not be directly
applicable to fish.
    Water is not actively transported and, therefore, movement of fluid
between compartments or within subcompartmental spaces is gov-
erned by hydraulic and osmotic forces across compartmental bound-
aries and the permeability barriers between them. Volume regulatory
processes can be considered at various levels from the intact fish to
single cells and even within subcellular compartments. Events that
affect one compartment invariably impact on all others and initiate a
cascade of physiological changes and responses. The emphasis of this
chapter is on extracellular volume regulation. Because water accounts
for around 95% of the mass of the mobile extracellular compartment,
the assumption is made that extracellular volume equals extracellular
water. Furthermore, the importance of intracellular fluids in regulating
extracellular volume is assumed to be minimal because individual
cells attempt to maintain constancy of their internal milieu. Therefore,
while intracellular water can provide a substantial fluid reservoir it
appears to react to, rather than control, the extracellular fluid envi-


   Organization of the various fluid compartments has been described
by Holmes and Donaldson (1969) and is briefly outlined on the follow-
ing page. Only three compartments (*) can be measured directly.

                   The Basic Fluid Compartments
                   I. Total body water*
                      A. Intracellular fluid
                      B. Extracellular fluid*
                         1. Interstitial fluid
                         2. Intravascular fluid*
                             a. Unstressed volume
                            b. Stressed volume
                         3. Lymph
                         4. Transcellular fluid

The lymphatic system is discussed elsewhere in this volume. Trans-
cellular fluids have been described by Holmes and Donaldson (1969)
and will not be considered further.

A. Total Body Water
    Total body water (TBW) is the amount of water in an intact fish.
Total body water is most accurately determined by desiccating the
carcass to a constant weight in an oven, although the terminal nature of
this process may be of limited value in some experiments. Methods
that employ indicator dilution techniques (Isaia and Masoni, 1976;
Kitzman et al., 1990), neutron activation analysis (Talbot et al., 1986),
and freeze drying (Cameron, 1980)have also been used.
    Values for TBW are usually reported as a fraction of total body
weight (bwt) (i.e., g or ml HzO/100 g bwt). Some authors (cf. Assem and
Hanke, 1979; Bittner and Lang, 1980; Loretz, 1979a,b) prefer to ex-
press body water as a fraction of dry cell solids (dcs) (g HzO/g dcs).
Water expressed as a fraction of total tissue weight will be less sensi-
tive to a change in either water or solids content, whereas slight
changes in tissue solids may overemphasize the importance of the
g HzO/g dcs value. With regard to TBW, neither calculation method
seems to have a particular advantage, although Shearer (1984) has
shown in rainbow trout, Oncorhynchus mykiss, that water weight per
total tissue (wet) weight is a better indicator of elemental status than
dry weight concentration. Presumably there is more variability in the
dry weight calculation due to individual differences in body fat. To be
consistent with the majority of values in the literature, volumes and
spaces originally presented as g HzO/g dcs have been converted to
ml/kg total weight using the formula:
 ml HzO/kg bwt    =   [g HzO/g dcs/(l+g HzO/g dcs)] . (1000/1.05), (1)
138                                                              KENNETH R. OLSON

where 1.05 is the assumed density of extracellular fluid. Values re-
ported in the literature as g/100 ml body weight have also been con-
verted to ml/kg by multiplying by 10/1.05.
    Total body water reflects the net water balance between a fish and
its environment and is determined by osmotic and hydraulic com-
ponents. Osmotic forces are affected by active and passive exchange of
electrolytes across the gills, skin, gastrointestinal and renal epithelia,
and nonionic osmolytes generated through metabolic activities. Net
water balance is also affected by mechanical compliance of the body
(total body fluid volume divided by total body fluid pressure). Total
body compliance determines the relationship between total fluid vol-
ume and average internal hydraulic pressure. Whole body compliance
has not been measured in fish.
    The entropic and homeostatic movements of water and osmolytes
across fish epithelia are summarized for typical freshwater (FW) and
saltwater (SW) teleosts in Fig. 1. All freshwater fish counter a volume
expanding environment by actively accumulating ambient and dietary

                       FRESH WATER

                        SALT WATER

                                       H 2 0 NaCl
     Fig. 1. Salt and water metabolism in freshwater (FW) and saltwater (SW) fish. In FW
there is a net accumulation of water and loss of salt that is compensated by active
accumulation of ions by the gills and through dietary sources. Dilute urine excreted by
the kidneys eliminates the water load while conserving electrolytes. Passive water and
salt fluxes are reversed in SW, and here fish drink SW to restore volume. Monovalent
ions, ingested by drinking and through the diet, are excreted by the gills, divalent ions
are excreted by the kidneys in a minimum of urine. Intracellular volume, indicated by
box, is regulated by adjusting intracellular osmolytes (ions, organics, or both). Coiled
spring on dorsum suggests net whole body compliance conditions, expansion in FW and
contraction in SW.
3.   BLOOD AND EXTRACELLULAR FLUID VOLUME REGULATION                                139

sodium and chloride across the gills and gut, respectively. The kidney
in these fish rids the body of excess water with a minimum solute loss.
Renal mechanisms, although not completely understood, appear to
derive much of their efficacy toward pressure and volume regulation
through pressure natriuresisldiuresis responses (Nishimura, 1985a),
similar in principle to those described in mammals (Guyton, 1990;
Guyton et al., 1990; Roman, 1986).
    Not all saltwater fish are hypoosmotic to their environment. Marine
myxinoids (hagfishes) are osmoconformers; that is, they are very close
to being isosmotic with seawater and extracellular electrolytes almost
mirror the environment. However, most studies have shown that these
fish are in fact around 2-10% hyperosmotic to their environment (Alt et
aZ., 1981; Cholette et al., 1970; Robertson, 1974, 1976, 1986). One
might assume that total body volume regulation in myxinoids is sub-
jected to the same constraints and responses as freshwater fish, albeit at
a reduced scale. Marine elasmobranchs maintain body electrolytes
somewhat higher than teleosts but well below that of the environment.
However, elasmobranchs are also slightly hyperosmotic (1-10%) to
seawater due to internal generation and retention of organic osmolytes,
most notably urea and trimethylamine oxide (Holmes and Donaldson,
 1969; Robertson, 1975, 1989). Thus, similar to myxine, elasmobranchs
osmoregulate in a slightly hypotonic environment. Saltwater teleosts
are hypoosmotic to their environment and respond to volume de-
pletion by drinking seawater and excreting the monovalent ions across
the gill. Divalent cations are excreted by the kidney in minimal sol-
vent. It may well be that marine teleosts and osmoregulating marine
 lampreys are the only fish capable of living in a substantially dehydrat-
 ing environment.'
     Representative water contents of whole fish and skeletal muscle,
 and the effect of environmental salinity are shown in Tables I and 11.
 Some variation in the data can be attributed to interspecific differences
 in dry weight composition due to scales (Thorson, 1961), bone
 (Cameron, 1985), and fat (Shearer, 1984; Thorson, 1961).There is also a
 10-15% decrease in TBW with age (Denton and Yuosef, 1976).
 However, as previously observed (Thorson, 1958, 1959, 1961), the
 most notable features of total water volume, or any of the fluid volumes
 for that matter, are the remarkable constancy between species and the
 minimal effect of environmental salinity.

      Lungfish and other species may encounter transient dehydration but restoration of
full physiological function in these fish is not achieved without external hydration in a
hypoosmotic environment.
   140                                                       KENNETH R. OLSON

                                        Table I
                                Total Body Water in Fish

           Species               Environment'     (ml/kg)            Reference

 Myxine glutinosa                  sw              689      Robertson (1976)
 Eptatretus stoutii                sw              746      McCarthy and Conte (1966)
 Petromyzon marinus                sw              720      Thorson (1959)
 Lampetra planeri                  sw              760'     Bull and Morris (1967)
 Hydrolagus colliei                sw              680      Thorson (1958)
  Squalus acanthias                sw              683
 Raja binoculata                   sw              787
 Raja rhina                        SW              78 1
 Squalus acanthias                 sw              712      Robertson (1975)
 Scyliorhinus canicula                             730      Robertson (1989)
 Potamotrygon hystrix              FW              793      Bittner and Lang (1980)
    AcipenserfEuvescens            FW              692      Thorson (1961)
    Polyodon spathula              FW              705
    Lepiosteus patostomum          FW              635      Thorson (1961)
    Amia calaa                     FW              710
    lctiobus cyprinellus           FW              672      Thorson (1961)
    Notropus cornutus              FW              758      Freda and McDonald (1988)
    Cyprinus carpio                FW              680      Thorson (1961)
    lctalurus punctatus            FW              653      Cameron (1980)
    lctalurus punctatus            FW              664      Kitzman et al. (1990)
   Catostomus commersoni           FW              709      Thorson (1961)
   Catostomus commersoni           FW              775      Hdbe (1987)
   Fundulus grandis                FW              737      Spence et a1. (1977)
   Enneacunthus obesus             FW              737      Gonzalez and Dunson (1987)
   Channa punctatus                FW              668      Sinha and Munshi (1979)
   Lepomis gibbosus                FW              768      Gonzalez and Dunson (1987)
   Perca fluuescens                FW              758      Freda and McDonald (1988)
   Pseudoscarus guacamuia          sw              696      Thorson (1961)
   Anguilla anguilla               sw              708"     Isaia and Masoni (1976)
   Epinephelus striatus            sw              683      Thorson (1961)
   Gymnothorax funebris            SW              607
   Sphyraena barracuda             sw              672
   Sarotherodon mossaiizbicu5      sw              742      Loretz (1979b)
   Mycteroperca tigris             sw              677      Thorson (1961)
   Lutianus griseus                sw              689      Thorson (1961)
   Lutianus campechanus            sw              676
   Platichthys stellatus           sw              784      Milligan and Wood (1987b)

3.   BLOOD AND EXTRACELLULAR FLUID VOLUME REGULATION                                  141

                                 Table I (Continued)

        Species                Environmenta      (mllkg)               Reference

Salmo trutta                     FW              735        Madsen (1990)
                                 SW              720
Salmo salar                      FW              694f,r     Talbot et al. (1986)
                                 sw              752"
Oncorhynchus mykiss              FW              7 14       Freda and McDonald (1988)
Oncorhynchus mykiss              FW              731        Milligan and Wood (1982)
Oncorhynchus mykiss              FW              740        Hdbe (1987)
Oncorhynchus mykiss              FW              773        Madsen (1990)
Oncorhynchus mykiss              FW              8 10       Schiffman and Fromm (1959)
Oncorhynchus mykiss              FW              773        Eddy and Bath (1979)
                                 SW              736
Oncorhynchus mykiss              FW              770        Finstad et a / . (1988)
                                 SW              759
Oncorhynchus mykiss              FW              782        Bath and Eddy (1979a)
                                 sw              780
Oncorhynchus mykiss              FW              737        Jackson (1981)
                                 sw              739
Oncorhynchus mykiss              FW              731        Munger et al. (1991)
Fundulus olivaceus               FW              711        Duff and Fleming (1972a)
                                 50% SW          751
Fundulus catenatus               FW              734        Duff and Fleming (1972b)
                                 40% sw          748

a Environment-salt water (SW) or fresh water (FW).
  Ml/kg body weight, determined by dessication unless otherwise indicated.
  Antipyrine space.
  3 H z 0 space.
f Parr.
R Determined by neutron activation.

B. Intracellular Water
    Control of intracellular fluid volume (ICFV) and composition en-
ables solvation and movement of appropriate molecules upon which
cellular functions are achieved. When confronted with hyper- or hypo-
tonic environments, most cells behave initially as near-perfect osmo-
meters. However, in time these cells restore part, or all, of the volume
perturbation by adjusting intracellular osmolytes, thereby producing a
regulatory volume increase or decrease, respectively. Both intracellu-
lar ions (Assem and Hanke, 1979; Lang et al., 1990; Schmidt-Nielsen,
       142                                                       KENNETH R. OLSON

                                           Table I1
                                  Total Water in Fish Muscle
~~            ~~             ~

                   Species       Environment"      (ml/kg)                Reference

C yclostomes
  Myxine glutinosa                  SW          74 1            Robertson (1986)
  Myxine glutinosa                  SW          683             Cholette et al. (1970)
  Potamotrygon hystrix              FW          799             Bittner and Lang (1980)
  Squalus acanthias                 SW          782             Bedford (1983)
  Callorhyncus millii               SW          792             Bedford (1983)
  Anguilla rostrata                 FW          736             Walsh and Moon (1982)
  Catostomus commersoni             FW          79 1            Gras et al. (1971)
  Saloelinus fontinalis             FW          758 (751-768)   Houston et al. (1971)
                                                                Houston et al. (1969)
                                                                Nichols et al. (1985)
     Sarotherodon mossambicus       FW          77 1            Assem and Hanke (1981)
     Perca fluviatilis              FW          764             Lutz (1972)
     Salvelinus alpinus             SW'         782             Finstad et al. (1989)
                                    SWd         783
     Platichthys stellatus          SW          804             Milligan and Wood (l987b)
     Anguilla anguilla              FW          746f            Chan et al. (1967)
                                    sw          737f
                                    FW          742"
                                    sw          734"
     Brecortiu tyranicus            sw          788             Engel et al. (1987)
                                    FW          792
     Cyprinus carpi0                FW          783 (768-804)   Abo Hegab and Hanke (1982)
                                                                Cupta and Hanke (1982)
                                    sw          749"            Abo Hegab and Hanke (1982)
     Oncorhynchus naykiss           FW          769 (728-817)   Eddy and Bath (1979)
                                    SW          744 (706-802)   Bath and Eddy (19794
                                                                Leray et al. (1981)
                                                                Cras et al. (1971)
                                                                Murphy and Houston (1977)
                                                                hlearow and Houston (1980)
                                                                H6be (1987)
                                                                Houston and Mearow (1979)
                                                                hlilligan and Wood (1982)
                                                                Munger et al. (1991)
     Salnio salar                   sw          753"            Parry (1961)
                                    sw          644
                                    FW          610
     Sulmo salar                    sw          773             Talbot and Potts (1989)
                                    FW          779


                                Table I1 (Continued)

         Species            Environment"      (ml/kg)               Reference

Salmo trutta                    FW         742             Madsen (1990)
                                sw         735
Gasterosteus aculeatus          FW         781             Lange and Fugelli (1965)
                                SW         764
Fundulus heteroclitus           FW         755             Schmidt-Nielsen (1977)
                                SW         736
Gillichthys mirabilis           FW         79 1            Loretz (1979,)
                                SW         785
Tilapia mossambica              FW         769             Assem and Hanke (1979)
                                SW         753 (744-763)
Pleuronectes jiesus             FW         773             Lange and Fugelli (1965)
                                SW         768

   Environment-salt water (SW) or fresh water (FW)
   ml/kg tissue wet weight.
  7 glliter NaCI.
   15 glliter NaCl.
 fyellow eel.
   Silver eel.

 1975, 1977) and organic solutes (Abo Hegab and Hanke, 1983; Assem
 and Hanke, 1983; Ballantyne et al., 1987; Goldstein and Kleinzeller,
 1987; King and Goldstein, 1983; Kleinzeller and Ziyadeh, 1990; Mc-
 Connell and Goldstein, 1990) may be regulated. Fluid movement in
 and out of the intracellular compartment appears to be an obligatory
 response to changes in intracellular and extracellular osmolarity, and it
 is probable that the regulatory responses of individual cells are de-
 signed to protect intracellular volume rather than to control extracellu-
 lar fluids. The assumption is made, then, that although fluid shifts
 between intracellular and extracellular compartments may be substan-
 tial, neurohumoral regulation of fluid volume is achieved principally,
 if not entirely, by adjustments in the volume and composition of extra-
 cellular fluids.

 C. Extracellular Water
    Extracellular water is the sum of interstitial, plasma, lymph, and
 transcellular fluids. Regulation of extracellular fluid volume (ECFV)
144                                                    KENNETH R. OLSON

and composition is necessary to achieve a stable, or at least manage-
able, environment around the cells and to serve as an avenue of molec-
ular commerce between cells and tissues. Extracellular fluid is inter-
posed between the intracellular compartment and the environment
and is thereby directly subjugated to hydraulic and osmotic forces at
both the cellular level in the tissues and across the various body
epithelia. If, as indicated in the previous section, transcellular water
movement is dictated initially by extracellular osmolarity, then extra-
cellular fluid volume is largely governed by transepithelial exchange
processes. Tissues responsible for these activities include the gill,
skin, gastrointestinal tract, and kidney.
    Extracellular fluid volumes have been estimated from either Gibbs-
Donnan distribution of endogenous ions (cf. Holmes and Donaldson,
1969) or with indicator dilution methods. Because of a somewhat in-
congruous distribution of the various ions or indicators, it is difficult to
determine the absolute extracellular volume, and the distribution
space of the specific extracellular volume marker is cited. Sodium
spaces probably provide the least reliable estimate of extracellular
fluid volume (Houston and Mearow, 1979; Lutz, 1972; Wood and
Randall, 1971)and will not be described further. In many fish, plasma
ion concentrations can probably be substituted for extracellular con-
centrations because of the relatively limited Gibbs-Donnan effect due
to high protein permeability of fish capillaries (Houston and Mearow,
    Indicator dilution methods are based on the relationship:

where V1 is the unknown volume, V, and Cz are the volume and
concentration of the indicator prior to dilution in the unknown volume
(Vz * Cs is the amount of indicator injected), and C1 is the concentration
of the indicator after it has been evenly distributed within the com-
partment. An ideal indicator is (a) physiologically and biochemically
innocuous, (b)not metabolized or modified in any way that might affect
its distribution or detection, (c)completely mixed and homogeneously
distributed within the compartment in question, (d) confined to the
compartment, and (e)precisely quantified (Feldschuh, 1990).The time
required for adequate distribution of an indicator in the test volume
must also be compatible with the experimental protocol.
    Much of the rationale for selection of suitable indicators for deter-
mining volumes in fish has been based on the behavior of these indica-
tors in mammalian systems, a tenuous and perhaps hazardous suppo-
sition. In addition, indicators must be introduced into the space in

question and that, in turn, may impose some degree of stress on the fish
and thereby affect volume homeostasis (Houston et al., 1971). The
methods employed to measure extracellular space in fish are provided
in the following paragraphs. An excellent summary of the early litera-
ture has been provided by Holmes and Donaldson (1969).
    Over the past 20 years the extracellular space markers most com-
monly used in fish have been inulin, mannitol, sucrose, 4000 molecu-
lar weight polyethylene glycol (PEG), [35S]sulfate, [32Plphosphate,
glofil ( 1251-iothalamate),51Cr-ethylene diamine tetraacetate (EDTA),
1251-albumin,or radionuclides of sodium or chloride. Two methods of
administering the indicator, constant infusion or single injection, may
be employed. In the former, the indicator is infused until compartmen-
tal concentrations reach steady state. The infusion is then stopped, and
the excreted indicator is collected until compartmental concentrations
fall to zero, which may take weeks (Hickman, 1972).The impracticality
of this method has limited its usefulness, and the single injection
procedure has been the method of choice.
    Hickman (1972) compared two methods, kinetic and net retention,
to determine extracellular space following a single inulin injection into
the flounder, Paralichthys lethostigma. With the kinetic method,
blood samples are taken at timed intervals after the indicator is in-
jected and the activity of the indicator in plasma is then plotted as a
function of time on a semilogarithmic paper (Fig. 2). Typically, the
resultant curve has a minimum of two components, a fast component
due to distribution of the indicator in the vascular and interstitial
spaces and a slower component as the indicator is gradually removed
from the extracellular compartment. Extrapolation of the slow com-
ponent back to the injection time provides the theoretical “instan-
taneous” concentration of indicator in the extracellular space. This
concentration is then used in the dilution equation (Equation 2) to
 solve for the indicator space (Hickman, 1972). Nichols (1987) has
 shown that, in trout, any one of several three-compartment models
provides a better description of inulin kinetics than either one- or
two-compartment models, and a three-compartment mammillary
model provides the best estimate of inulin kinetics. Presumably this
model accounts for differential accessibility of inulin into extravascu-
lar spaces of both highly and poorly vascularized tissues; the latter
taking as long as 10.5 h to reach equilibrium in trout (Nichols, 1987)
and 8 h in eels (Chester Jones et al., 1969).
    The net retention method requires quantification of the indicator
lost from the compartment during the mixing period. In these experi-
ments it is assumed that the ‘lost’ indicator is excreted into the urine,
146                                                                 KENNETH H. OLSON

                            10      20        30      40       50      60
                                  HOURS AFTERINJECTION

    Fig. 2. Kinetic method. The decrease in plasma radioactivity (composite curve)
following injection of [3H]methoxy inulin into southern flounder, P . lethostigrna, has
two components, A and B. The fast component (A) represents inulin equilibration with
the interstitial fluid; curve B primarily indicates inulin excretion from the body. The
intercept resulting from extrapolation of curve B to time 0 (C,) is equivalent to inulin
dilution in the extracellular fluid space and is substituted for C1 in Equation 2 (see text).
[Redrawn from Hickman (1972),with permission.]

and/or environment and can thus be accurately measured. The amount
of indicator excreted (AE) is then subtracted from the total amount
injected and Equation 2 becomes:

Only a single blood sample is necessary once the indicator has equili-
brated throughout the compartment in question (Fig. 3 ) .This is advan-
tageous in minimizing the volume disturbance to the fish but requires
preliminary information on the time required for indicator equili-
bration, a point often overlooked in these studies. The net retention
method is also insensitive to internal removal (e.g., cellular binding or
uptake) of indicator from the test compartment and can result in over-
estimation of the space. Extracellular space of individual tissues is
often calculated as the ratio of the concentrations of indicator in tissue
relative to plasma at the time of tissue collection. This method optimis-
tically assumes that the concentration of indicator in interstitial fluids
temporally mirrors plasma concentrations.
3.   BLOOD AND EXTRACELLULAR FLUID VOLUME REGULATION                             147

                a-                    49
                                     1.            53
                                                  1.         59
                                                      _------ -




                  0                 10                20               J

                                 HOURS AFTER INJECTION
    Fig. 3. Net retention method. Extracellular fluid space determined in FW and SW
southern flounder, P . lethostigma, at 12, 18, and 24 h after injection of [3H]methoxy
inulin using the net retention method. Arrows indicate time when composite plasma
inulin disappearance curve (see Fig. 2) begins an exponential decline. Dots on curve
prior to 10 h indicate calculated extracellular space prior to inulin equilibrium and
underestimate the apparent space. [From Hickman (1972), with permission.]

    Probably none of the indicators used to date give an exact represen-
tation of extracellular space. At least 8-12 h are necessary for inulin to
be sufficiently equilibrated with the extracellular fluids (Chester Jones
et al., 1969; Hickman, 1972; Munger et al., 1991; Nichols, 1987), and
studies over shorter intervals may underestimate distribution vol-
umes. Thorson observed that, after 4-5 h circulation, the volume dis-
tribution of saccharides in an elasmobranch (Thorson, 1958) and tele-
ost (Thorson, 1961) was inversely proportional to the indicator's
molecular weight, and he suggested that this was due to the relative
diffusivity of these molecules in water. Certainly a 5-h equilibration
period would seem to limit inulin penetration into certain com-
partments and thus underestimate inulin spaces. However, if equili-
bration periods become prolonged there is a risk of tissue binding or
metabolism. Although neither [3Hl nor [l4C1inulin are metabolized
after 146 h in the circulation (Hickman, 1972), the 3H-labeled com-
pounds often produce larger extracellular spaces, even during 10- to
148                                                  KENNETH R . OLSON

24-h equilibration periods (Hickman et al., 1972; Munger et al., 1991;
Schmidt-Nielsen et al., 1972). When pairs of commonly used indica-
tors, PEG, inulin, and mannitol, are injected, [ 14C]PEG generally
provides the most conservative estimate of whole body and tissue
extracellular volumes. Inulin, especially the tritiated form, is accumu-
lated by kidney tissue (Beyenbach and Kirschner, 1976; Hickman et
al., 1972; Schmidt-Nielsen et al., 1972) and [14C]inulin may be re-
moved by liver (Munger et al., 1991). Schmidt-Nielsen et al. (1972)
also observed elevated [3H]inulin spaces in muscle, gut, and liver
after a 24-h equilibration period while Munger et al. (1991) found
[ 14C]inulin spaces lower in red muscle than [3H]PEG spaces after 13h.
Mannitol spaces are consistently greater than either PEG or inulin in
most tissues, especially after 6 h (Cameron, 1980; Munger et al., 1991).
Mannitol appears to be accumulated by gill (Munger et al., 1991),liver
(Milligan and Wood, 1986a,b, 1987a,b; Munger et al., 1991), and heart
(Munger et al., 1991). Mannitol (Munger et al., 1991), or other small
saccharides such as sucrose (Fenstermacher and Patlak, 1977), may
more readily penetrate the blood-brain barrier and therefore be
preferable to other indicators in this tissue. However, Wood et al.
(1990) have shown that even after 12 h in the circulation, mannitol
penetration into the cerebrospinal fluid space (CSF) of the skate, Raja
ocellata, is only around 26% of that predicted that equilibrium be-
tween CSF and plasma.
     In some tissues, PEG spaces may compare favorably with spaces
calculated by Gibbs-Donnan methods. Houston and Mearow (1979)
used an indicator dilution method to compare the distribution space of
 [14C]PEG to Na+, C1-, and C1-/Kf spaces in tissues of rainbow trout.
Twelve hours after injection, ['*C]PEG spaces were similar to C1- and
 Cl-/K+ spaces in epaxial and cardiac muscle and gut; however,
 [14C]PEG appeared to provide a more realistic estimate than ion
 spaces in spleen and brain. [14C]PEGspaces in liver were only slightly
higher than those determined by other methods (Houston and
 Mearow, 1979).A number of investigators (H6be and McMahon, 1988;
 Mearow and Houston, 1980; Munger et al., 1991; Sinha and Munshi,
 1980) have found that Cl-/K+ spaces are satisfactory in skeletal mus-
 cle, whereas Munger et al. (1991) found Cl-/K+ underestimated white
 muscle space. Chloride and inulin spaces are virtually identical in
 perch, Perca fluviatilis, muscle, whereas C1- space is 60 and 90%
 greater in liver and gut, respectively (Lutz, 1972). However, compared
 to inulin, Gibbs-Donnan methods may be less applicable when fish are
 adapted to different salinities (Eddy and Bath, 1979).
     Tables 111, IV, and V summarize extracellular fluid spaces for whole
                                                              Table 111
                                            Whole Body Extracellular Fluid Volumes in Fish

                             Env"   (nil/kg)         Indicator          Method"          Comments            Reference
 Lampetra planeri            FW      244       [ ''C]Inulin          Inj (1 h)        Ammocoete     Bull and Morris (1967)
                                     223       [ "CC]Sucrose         Inj (1h)
                                     285       c1-                   G-D
 Myxine glutinosa            SW      255       [ "C]Inulin           Inj (14-18 h)                  Cholette et al. (1970)
 Eptatretus stouti           sw      254       [ '"CIInulin          K (104 h )                     McCarthy and Conte (1966)
 Petromyzon marinus          sw      228       Sucrosed              K (40-235 min)                 Thorson (1959)
 H ydrolagus colliei         SW      101       Inulind               K (4-5   h)                    Thorson (1958)
 Squalus acanthias           Sw      121       Inulind               K (4-5   h)                    Thorson (1958)
                                     145       Ramnosed              K (4-5   h)
                                     202       Sucrosed              K (4-5   h)
 Squalus acanthias           sw      122       [ 3H]Inulin                                          Benyajati and Yokota (1989)
 Raja binoculata             SW      126       Inulind               K (4-5 h)                      Thorson (1958)
 Raja rhina                  sw      112       Inulind               K (4-5 h)                      Thorson (1958)
    Acipenserfluvescens      FW       191      Sucrose"              K (3-5 h)                      Thorson (1961)
    Polydon spathula         FW       149      Sucrosed              K (3-5 h)                      Thorson (1961)
    Amia calva               FW       180      Sucrosed              K (3-5 h)                      Thorson (1961)
    Lepiosteus patostomum    FW       130      Sucrosed              K (3-5 h)                      Thorson (1961)
    Carassius auratus        FW      114       C1-                   G-D                            Houston (1962)
    Catastomus commerso ni   FW      228       CI-/K+                G-D                            H6be (1987)

                                                             Table 111 (Continued)

                             Env“   ( m k )           Indicator             Method’        Comments                Reference
    Catastomus commersoni    FW     116       Sucrosed                  K (3-5 h)                       Thorson (1961)
    Cyprinus carpio          FW     148       Sucrosed                  K (3-5 h)                       Thorson (1961)
    lctiobus cyprinellus     FW     126       Sucrosed                  K (3-5 h)                       Thorson (1961)
    Salzjelinus fontinalis   FW     118       [I4C]InuIin               K (4 h)                         Nichols et a1. (1985)
                             sw     112       [ ‘‘C]Inulin              K (4h)
    S ynbranchus             FW     163       [ 3H]Inulin               K (12 h)                        Heisler (1982)
    Pseudoscarus guacamaia   sw     158       Sucrosed                  K (3-5 h)                       Thorson (1961)
                                    137       Raffinosed                K (3-5 h)
                                    109       Inulind                   K (3-5 h)
    Fundulus grundis         sw     236       [ ’‘C]Inulin              Inj (24 h)                      Spence et al. (1977)
g   Mycteroperca tigris      sw     119       Sucrose“                  K (3-5 h)                       Thorson (1961)
O   Gymnothoraxfunebris      sw     151       Sucrosed                  K (3-5 h)                       Thorson (1961)
    Epinephelus striatus     sw     138       Sucrosed                  K (3-5 h)                       Thorson (1961)
    Sphyraena barracuda      sw     1.51      Sucrosed                  K (3-5 h)                       Thorson (1961)
    Lutianus campechanus     sw     133       Sucrosed                  K (3-5 h)                       Thorson (1961)
    Lutianus griseus         sw     133       Sucrosed                  K (3-5 h)                       Thorson (1961)
    Platich thyesflesus      sw     308       36~1-                     Inj (75 min)                    Macfarlane and Maetz (1975)
    Platichthys stellatus    sw     255       F3H]Mannitol              NR (12 h)                       Milligan and Wood (l987a)
                                    259       l3H1Mannitol              NR (12 h)
                                    319       [ 3H]Mannitol             NR (36 h)
    lctalurus punctatus      FW     166       [3H]Inulin                K (?I                           Cameron (1980)
                                    178       [3H]Mannitol              K (9
    lctalurus punctatus      FW     183       SCN”.“                    K (4h)                          Kitzman et al. (1990)
    Oncorhynchus mykiss      sw     181       [ I4C]PEG                 K (50 h)                        Beyenback and
                                    263       [:’H]Inulin               K (50 h)                          Kirschner (1976)
                                    231       [ 3H]Inulin               K (50 h)       different fish
                                    262       ’2”-glofil                K (50 h)
    Oncorhynchus mykiss      sw     280       c1-                       G-D            10 days          Bath and Eddy (19794
                                FU'      460     c1-                 C-D
    Oncorhynchus mykiss         SW       151     c1-                 G-D                 Transfer-100 h      Houston (1959)
                                FW       157     c1-                 G-D                 <40g BWT
                                         124     c1-                 G-D                 70g BWT
                                FW       138     c1-                 G-D
    Oncorhynchus mykiss         FW       273     [ 14C]Mannitol      K (12 h)                                Milligan and Wood (1982)
    Oncorhynchus mykiss         FW       127     [ ''C]Inulin        Inj (6 h)                               Munger et al. (1991)
                                         287     [3H]Mannitol        Inj (6 h)
                                         193     [ 14C]Inulin        Inj (13h)
                                         278     [ 'HIMannitol       Inj (13h)
    Oncorhynchus mykiss         FW       201     [ 14C]Inulin        Inj (24 h)                              Eddy and Bath (1979
                                         216     CI-/K+              G-D
    Oncorhynchus mykiss         FW       216     c1-                 K                                       Habe (1987)
    Oncorhynchus mykiss         FW       280     c1-                 G-D                                     Bath and Eddy (197%)
    Oncorhynchus mykiss         FW       172     [ ''C]Inulin        K (24 h)            3 Pool model        Nichols (1987)
    Oncorhynchus mykiss         FW       241     ['HIMannitol        NR (12 h)                               Milligan and Wood (1986a)
    Channa punctatus            FW       101     Inulind             Inj (30 min)                            Sinha and Munshi (1980)
                                         196     c1-                 G-D
    Paralichthys lethostigma    SW       156     [metho~y-~H]InulinK (50 h)                                  Hickman (1972)
                                         166     [meth~xy-~H]InulinNR (12 h)
                                         178     [rnetho~y-~H]Inulin NR (18 h)
                                         186     [metho~y-~H]InulinNR (24 h)
                                FW       145     [metho~y-~H]InulinK (50 h)
                                         142     [metho~y-~H]InulinNR (12 h)
                                         146     [metho~y-~H]InulinNR (18 h)
                                         151     [metho~y-~H]InulinNR (24 h)

   a   Environment-salt water (SW) or fresh water (FW).
       In mllkg body weight.
       Kinetic method (K), net retention method (NR), Gibbs-Donnan method (C-D), space at time after injection (inj), equilibration times in
    '' SCN, sodium thiocyanate.
                                                        Table IV
                                 Tissue Extracellular Fluid Volume: Nonskeletal Muscle

                          Env“     (ml/kg)         Indicator         Method              Comments          Reference

    Raja erinacea         SW         259        C1-                 G-D             Telencephalon   Cserr et a / . (1983)
                          SW         243        c1-                 G-D             Medulla
    Squalus acanthias     SW          40        [ 14C]Inulin        Inj (20 h)      Telencephalon   Fenstermacher and
                                     105        [3H]Sucrose         Inj (20 h)      Medulla           Patlak (1977)
                                      29        [ 14C]Inulin        Inj (20 h)      Spinal cord
                                      71        [“HISucrose         Inj (20 h)
                                      47        [14C]Inulin         Inj (20 h)
                                     121        [3H]Sucrose         Inj (20 h)
   Squalus acanthias      SW         242        [14C]PEG            Inj (12 h)      Assume tissue   Schmidt-Nielson et
                                                                                    80% H20           al. (1972)
    Squalus acanthias     SW         29 1       [14C]PEG            Inj (12 h)      Assume tissue   Schmidt-Nielson et
                                                                                    80% HzO           al. (1972)
    Squalus acanthias     Sw         208        [ l4C1PEG           Inj (12 h)      Assume tissue   Schmidt-Nielson et
                                                                                    80% HzO           al. (1972)
  Rectal Gland
    Squalus acanthias     SW         182        [ 14C]PEG           Inj (12 h)      Assume tissue   Schmidt-Nielson et
                                                                                    80% H20           a / . (1972)
    S o u a h ucunthius   SW         212        [‘%]PEG             Inj (12 h)      Assume tissue   Schmidt-Nielson et
                                                                                    80% H20           a1. (1972)
    Ictalurus punctutus    FW   190    ['HH]Inulin     Inj (24 h)   Skull       Cameron (198.5)
                                117    [3H]Inulin      Inj (24 h)   Vertebrae
   Ictalurus punctatus     FW    68    [3H]Inulin      Inj (24 h)               Cameron (1985)
   Oncorhynchus mykiss     FW    63    [ 14C]PEG       Inj (12 h)               Houston and Mearow
                                317    c1-             G-D                       (1979)
                                345    Cl-/K+          G-D
    Oncorhynchus mykiss    FW   277    CI-,Cl-/K+      G-D          Summer      Mearow and Houston
                                405                                 Winter       (1980)
   Oncorhynchus mykiss     FW    25    [14C]Inulin     Inj (6 h)                Munger et a1. (1991)
                                 67    13H]Mannitol    Inj (6 h)
                                 30    [3H]PEG         Inj (13 h)
                                101    [3H]Mannitol    Inj (13 h)
                                125    [14C]Mannitol   Inj (13h)
                                 35    ['4C]Inulin     Inj (13h)
   Platichthys stellatus   SW   224    [3H]Mannitol    NR (12 h)                Wood and Milligan
  Ictalurus punctatus      FW   238    [3H]Inulin      Inj (24 h)               Cameron (1985)
  Anguilla rostrata        FW   296    [3H]Inulin      Inj (8 h)                Walsh and Moon
   Oncorhynchus mykiss     FW   240    [3HlMannitol    In vitro                 Farrell and Milligan
                                                       (20 min)                   (1986)
   Oncorhynchus mykiss     FW    80    [3H]lnulin      In vitro                 Nielsen and Gesser
   Oncorhynchus mykiss     FW   190    [I4C]PEG        Inj (12 h)               Houston and Mearow
                                210    c1-             G-D                        (1979)
                                215    CI-/K+          G-D
   Oncorhynchus mykiss     FW   290    CI-,CI-/K+      G-D          Summer      Mearow and Houston
                                22 1                                Winter       (1980)

                                               Table IV (Continued)

                              Env"   (mlk)         Indicator      Method        Comments         Reference

      Oncorhynchus mykiss     FW     148     [ ''C]Inulin       Inj (6 h)                  Munger et al. (1991)
                                     279     ['HIMannitol       Inj (6 h)
                                     170     ['HIPEG            Inj (13h)
                                     364     [3H]Mannitol       Inj (13 h)
                                     285     [ 14C]Mannitol     Inj (13h)
                                     180     ['4C]Inulin        Inj (13h)
      Oncorhynchus mykiss     FW     164     Inulin"            I n uitro                  Hansen and Gesser
                                                                (1h)                         (1987)
      Salmo truttu            FW     166     [ L4C]Inulin       Inj (24 h)                 Fugelli and Vislie
      Platichthys stellatus   sw     224     ['HIMannitol       NR (12 h)                  Wood and Milligan
cn                                                                                           (1987)
       Oncorhynchus mykiss    FW     218     [3H]PEG            Inj (13 h)                 Munger et ul. (1991)
                                     440     ['HIMannitol       Inj (13 h)
                                     344     [ 14C]Mannitol     Inj (13h )
                                     424     CI-/K+             G-D
      Anguilla unguillu       sw     474     C1-                G-D                        Isaia and Masoni
      Tilapia nilotica        FW     130     "CR-EDTA           Inj (27 h)                 Babiker et a1. (1979)
      Oncorhynchus mykiss     FW     420     [ 14C]PEG          Inj (12 h)                 Houston and Mearow
                                     44s     c1-                G-D                         (1979)
                                     472     CI-IK'             G-D
       Oncorhynchus mykiss    FW     277     CI-,CI-/K'         G-D          Summer        Mearow and Houston
                                     405                                     Winter          (1980)
       Perca Juljiatilis      FW     306     Inulin'            Inj (24 h)   Intestine     Lutz (1972)
      Tilupia nilotica            58   51CR-EDTA        Inj (27 h)   Intestine        Babiker et al. (1979)
                                  94                                 Hepatopancreas
      Pseudopleuronectes    SW   200   [ 14C]PEG        Inj (24 h)   Assume tissue    Schmidt-Nielson et
        umericanus                                                   75% HzO            a1. (1972)
      Anguillu rostrata     FW   171   ['4C]PEC         Inj (24 h)                    Schmidt-Nielson et
                            SW   186   [l4C1PEG                                         al. (1972)
      Tilapia nilotica      FW   208   51CR-EDTA        Inj (27 h)                    Babiker et al. (1979)
      Pseudopleuronectes    SW   229   [ 14C]PEG        In vitro                      Hickman et al. (1972)
      americanus                 358   [3H]PEG
                                 367   [ I4C]Inulin
                                 349   ['H]Inulin
     Pseudopleuronectes     SW   357   [ 14C]PEG        Inj (24 h)   Assume tissue    Schmidt-Nielson et
     americanus                                                      75% HzO            al. (1972)
     Anguilla rostrata      FW   492   [ I4C]PEG        Inj (24 h)                    Schmidt-Nielson et
                            SW   635   [14C]PEG                                         al. (1972)
%   Liver
      Anguilla rostrata     FW   256   [14C]Inulin(or   Inj (8 h)                     Walsh and Moon
                                         "-                                            (1982)
      Oncorhynchus mykiss   FW   322   [14C]PEG         Inj (12 h)                    Houston and Mearow
                                 295   c1-              G-D                            (1979)
                                 310   CI-IK'           G-D
      Oncorhynchus mykiss   FW   323   CI-/K+           G-D          18 h light       Murphy and Houston
                                 327   CI-/K+           G-D          6 h light         (1977)
      Oncorhynchus mykiss   FW   140   [ 14C]Inulin     Inj (6 h)                     Munger et al. (1991)
                                 821   i3H] Mannitol    Inj (6 h)
                                 195   [ ''C]Inulin     Inj (13 h)
                                 145   [3H]PEG          Inj (13 h)
                                 862   [3H]Mannitol     Inj (13 h)
                                 790   [ 14C]Mannitol   Inj (13h)
      Perca fluviatilis     FW   220   Inulin           Inj (24 h)                    Lutz (1972)

                                                   Table IV (Continued)

                                   Env"   (ml/kd     Indicator        Method        Comments         Reference

      Tilapia nilotica              FW      101    "CR-EDTA          Inj (27 h)                 Babiker et a1. (1979)
      Platichthys stellatus         SW    -571*    [3H]Mannitol      NR (12 h)    *Unsuitable   Milligan and Wood
                                            170    [ ''C]Inulin      (12 h)       14C = 3H        (1987b)
      Pseudopleuronectes            SW      229    [ 14C]PEG         Inj (24 h)   Assume        Schmidt-Nielson et


      Anguilla rostratu             FW      186    [ "CIPEG
                                                                                    75% H 2 0
                                                                                                  al. (1972)

                                    SW      171    [ 14C]PEC
       Pleuronectes platessa        SW      607    '311-albumin      Inj (50 h)                 Wardle (1971)
       Oncorhynchus mykiss           FW     234    [l4C1PEG          Inj (12 h)                 Houston and Mearow
                                            306    C1-               G-D                         (1979)
                                            336    Cl-/K+            G-D
       Tilapia nilotica              FW     157    "'CR-EDTA         Inj (27 h)                 Babiker et al. (1979)

      " See Table I11 for abbreviations
      ' Unlabeled.
                                                         Table V
                                             Muscle Extracellular Fluid Volume

                              Env"   Wkg)         Indicator          Method        Comments              Reference
     Myxine glutinosa        SW      128       Inulinb            Inj (24 h)                    Robertson (1976)
                                     194       c1-                G-D
     Myxine glutinosa        sw      128       Inulinh            Inj (24 h)                    Robertson (1986)
                                     278       1
                                               c-                 G-D
      Myxine glutinosa       sw      188       [I4C]Inulin        Inj (14-18 h)   Parietal      Cholette et al. (1970)
                                     155       [I4C]Inulin        Inj (14-18 h)     tongue
     Myxine glutinosa        SW      136       [14C]PEG           Inj (24 h)      Assume        Schmidt-Nielson et d .
                                                                                    tissue        (1972)
                                                                                    80% HzO
%   Elasmobranchs
      Chimaera monstrosa     sw      210       Inulin"            In vitro                      Robertson (1976)
      Squalus acanthias      SW      120       Inulin"            Inj (14-22 h)                 Robertson (1975)
                                     177       c1-                G-D
     Squalus acanthias       sw       50       [14C]PEG           Inj (12 h)      Assume        Schmidt-Nielson et al.
                                                                                    tissue        (1972)
                                                                                    80% H 2 0
      Catastomus             FW       47       CI-/K+                                           H6be (1987)
     lctalurus punctatus     FW       55       [3H]Inulin         K (24 h)        White         Cameron (1985)
                                     156       [3H]Inulin         K (24 h)        Red
     Salvelinus fontinalis   FW       70       1
                                               c-                 G-D                           Houston et al. (1971)
     Saluelinus fontinalis   FW       73       c1-                G-D                           Houston et al. (1969)
     Salvelinus fontinalis   FW       52       [ '*C] Inulin      K (12 h)        Epiaxial      Nichols et al. (1985)

                                              Table V (Continued)

                        Env"   (ml/kd      Indicator          Method      Comments               Reference

Oncorhynchus mykiss   FW        44      [14C]PEG           Inj (12 h)   Postopercular   Houston and Mearow
                                41      C1-                G-D                           (1979)
                                37      CI-/K+             G-D
                                37      [14C]PEG           Inj (12 h)   Subdorsal
                                38      c1-                G-D
                                32      Cl-IK+             G-D
                                75      [ 14C]PEG          Inj (12 h)   Caudal
                                66      c1-                G-D
                                63      CI-/K+             G-D
Oncorhynchus mykiss   FW        73      [I4C]Mannitol      K (12 h)     Epiaxial        Milligan and Wood (1982)
Oncorhynchus mykiss   FW        53      CI-/K+                                          Hdhe (1987)
Oncorhynchus mykiss   FW        52      CI-,CI-IK'         G-D          Postopercular   Mearow and Houston
                      Summer    33                                      Subdorsal        (1980)
                                60                                      Caudal
                      Winter    53                                      Postopercular
                                44                                      Subdorsal
                                50                                      Caudal
Oncorhynchus m k s
              yis     FW        45      ['4C]Inulin        Inj (13 h)   White           Munger et d . (1991)
                                54      [3H]PEG            Inj (13 h)   Muscle
                                85      [3H]Mannitol       Inj (13 h)
                                74      ['4CC]Mannitol     Inj (13 h)
                                45      [ ''C]Inulin       Inj (6 h)
                                64      [3H]Ma~initol      Inj (6 h)
                                23      CI-/K+             G-D
                                76             I
                                        [ 14C] n uli n     Inj (6 h)
                               126      [3H]Mannitol       Inj (6 h)
                                92      [ ''C]Inulin       Inj (13 h)   Red
                               140      [3H]PEG            Inj (13 h)   Muscle
                               131    [3H]Mannitol     Inj (13 h)
                               145    [ 14C]Mannitol   Inj (13 h)
Oncorhynchus mykiss     FW      48    CI-/K+           G-D          18 h light      Murphy and Houston
                                46    CI-/K+           G-D          6 h light        (1977)
Oncorhynchus mykiss     FW      73    [ ''C]Inulin     Inj (24 h)                   Eddy and Bath (1979)
                               105    Cl-/K+           G-D
Oncorhynchus mykiss     FW     222    32P04            In vitro     Tissue slices   Gras et al. (1971)
                               224    Inulin           In vitro
Oncorhynchus mykiss     FW      82    [3H]Mannitol     Inj (12 h)                   Milligan and Wood (1986b)
Perca Juvia tilis       FW      74    huhb             Inj (24 h)                   Lutz (1972)
Tilapia nilotica        FW      51    51Cr-EDTA        Inj (27 h)                   Babiker et ( 1 1 . (1979)
Channa punctatus        FU'    157    c1-              G-D          Hi gonad wt'    Sinha and Munshi (1979)
                                99                                  Low gonad
                                 71                                 Hi gonad wtC
Platichthys stellatus   sw .    120   [3HlMannitol     NR (12 h)                    Milligan and Wood (1987b)
Platichthys stellatus   sw       85   [3H]Mannitol     NR (12 h)                    Wood and Milligan (1987)
Pleuronectes platessa   SW     4070   1311-albumin     Inj (50 h)                   Wardle (1971)
Parophrys vetulus       sw      100   [3HlMannitol     Inj (12 h)   White           Wright et ( 1 1 . (1988)
Pla tichthys Jesus      sw      170   [35S1SO4         Inj (8 h)    6°C             Maetz and Evans (1972)
                        sw      213   [35SI SO4        Inj (8 h)    16°C
Pseudopleuronectes      sw       50   [ 14C]PEG        Inj (24 h)   Assume          Schmidt-Nielson et a1
  americanus                                                          tissue          (1972)
                                                                      75% HzO
Anguilla anguilla       FW       46   [35S]S04         Inj (58 h)   Parietal (y)    Chan et a / . (1967)
                                100                                 Tongue (Y)
                                 63                                 Parietal (s)    y = yellow
                                150                                 Tongue (s)      s =   silver
                        sw      147   [35S1SO4         Inj (58 h)   Parietal (y)
                                229                                 Tongue (Y)
                                102                                 Parietal (s)
                                185                                 Toneue(s)
                                                      Table V (Continued)

                                Env"   (ml/kg)       Indicator         Method     Comments             Reference

Anguilla rostrata               FW      72       [3H]Inulin        Inj (8 h)    White         Walsh and Moon (1982)
                                       149       [3H]Inulin        Inj (8 h)    Red
Anguilla rostrata               FW      50       [I4C]PEG
                                sw      79       [ I4C]PEG
Cyprinus carpio                 FW      97       [ '4C]Inulin      Inj ( 3 h)                 Abo Hegab and Hanke
                                sw     176       [ ''C]Inulin                   15 gll NaCl    (1982)
C yprinus carpio                FW     159       [ '*C]Inulin      Inj ( 3 h)                 Gupta and Hanke (1982)
                                        72       [ ''C]Inulin
Fundulus heteroclitus           FW      54       ['4C]PEG          Inj (6 h)                  Schmidt-Nielsen (1977)
                                sw      58       [ 14C]PEG         Inj (6 h)
Gillichthys mirabilis           FW      70       CI-                                          Loretz (1979a)
                                        40       [ ''C]Inulin      Inj (8 h)
                                sw      92       CI -
                                        52       [ ''C]Inulin      Inj (8 h)
Sarotherodon                    FW      49       ['*C]Inulin       Inj (3 h)                  Assem and Hanke (1981)
Sarotherodon                   SW       65       [ ''C]Inulin      Inj (24 h)                 Loretz (1979b)
Tilapia mossambica             FW       59       [ ''C]Inulin      Inj ( 3 h)                 Assem and Hanke (1979)
                               SW       60       ['4C]Innlin                    27 g/l NaCl
                               sw       70       ['*C]Inulin                    35 g/l NaCl

 a   See Table I11 for abbreviations
 " Unlabeled.
 ' Mature fish.

fish, skeletal muscle, and tissues other than skeletal muscle, respec-
tively. It becomes readily apparent that the variability due to different
methods is often as great, if not greater, than actual differences be-
tween fish. Comparatively little work has been done on nonteleostean
fish since the elegant surveys by Thorson (1958,1959,1961). However,
even these studies should be repeated using other indicators and
longer circulation times because many spaces reported by Thorson,
especially for teleosts, are somewhat lower than those determined
since then (see Table 111). Based on sucrose spaces, Holmes and
Donaldson (1959) concluded that extracellular space is lower in the
more advanced fish (teleosts). As is evident from the variability in
these measurements (Table 111),further work is required to substan-
tiate this hypothesis.
    Frequently, tissue extracellular fluid spaces are measured not for
their intrinsic properties, but incidental to analyses of intracellular
composition. This is especially true for teleost muscle (see Tables IV
and V). Because teleost muscle tissue is around 50% of the total body
weight (Gingerich et al., 1987; Heisler, 1982; although Stevens, 1968
reports 66%), it is clear that muscle intracellular fluid and solutes
constitute a major fraction of the total body composition. It is also
tempting to think that muscle extracellular fluid is a major component
of the total extracellular volume although this is probably not the case.
Most estimates of muscle extracellular space (Table V) are around
50-100 ml/kg tissue weight, which is only around 30-40% of the
extracellular spaces reported for most other tissues or for the whole
body (Tables 111 and IV). Thus muscle may contain as little as 20-30%
of the total extracellular fluid and, therefore, may not be the most
quantitative index of total intercompartmental fluid movements.
         OF     SALINITY ON   ECFV
    Few factors pose a greater immediate threat to fish volume regula-
tion than the osmolarity of the environment. Extracellular fluids ap-
pear to be more sensitive than intracellular fluids to an initial change in
transepithelial osmolarity gradients. When ECFV is evaluated in the
context of the change in TBW, both organismal and cellular regulatory
systems can be appreciated. Studies such as these, although limited to
date, are clearly necessary to understand the homeostatic mechanisms
that accompany life in these diverse environments, and they may well
provide the key to an appreciation of volume regulation in vertebrates
in general.
   Water is lost from intracellular stores by transfer of FW fish to media
of higher salinity. This has been observed in the elasmobranch Po-
tamotrygon hystrin: (Bittner and Lang, 1980), the stenohaline FW tele-
162                                                      KENNETH R . OLSON

ost Cyprinus carpio (Abo Hegab and Hanke, 1982),and the euryhaline
teleosts Tilapia mossambica (Asseni and Hanke, 1979, 1981), Pleuro-
nectes flesus (Lange and Fugelli, 1965), Gasterosteus aculeatus
(Lange and Fugelli, 1965), Fundulus heteroclitus (Schmidt-Nielsen,
1977), Oncorhynchus mykiss (Eddy and Bath, 1979; Finstad et al.,
1988), Salvelinus fontinalis (Nichols et al., 1985),and Eugerres plum-
ieri (Plaza-Yglesias et al., 1988). Intracellular water and ambient salin-
ity are inversely related in stenohaline Myrine glutinosa (Cholette et
al., 1970) over the range of salinity tolerance (600-1540 mOsm/liter).
An increase in extracellular osmolarity accompanying transfer of Cy-
prinus carpio to hypertonic mannitol (Gupta and Hanke, 1982) or
hemorrhage in Oncorhynchus mykiss (Duff and Olson, 1989) also ex-
tracts water from the intracellular compartment. Conversely, events
that increase intracellular osmolarity (e.g., exercise) draw extracellular
water into the cells (Milligan and Wood, 1986a,b, 1987a,b; Wright et
al., 1988).
    An osmoconformer such as the hagfish, Myxine glutinosa, is also a
volume conformer. Cholette et al. (1970) have shown that within the
range of salinity tolerance of myxine (600-1500 mOsm) muscle ECFV
and environmental osmolarity (ENVosm)are directly related by the
                   ECFV    =   0.013(ENVoS,)   + 4.87,                 (4)
while total muscle water (TMW) and ENVosm are inversely related
according to
                   TMW    =   -0.017(ENVoSm)+ 90.1.                    (5)
Thus, as ambient salinity increases from 600 to 1500 mOsm, muscle
ECFV increases from 130 to 225 g/kg while TMW decreases from 780
to 640 g/kg. Although this study (Cholette et al., 1970) is limited to
muscle, it can be tentatively concluded that as ambient salinity in-
creases, there is a net loss of water from the fish and a concomitant shift
of fluid from the intracellular to the extracellular compartment, the
latter as a result of the increased extracellular osmolarity.
    Extracellular fluid volume in the dogfish shark, Squalus acanthias,
is dramatically affected by ambient salinity; transfer from 100% SW to
90 or 110% SW changes ECFV from 122 ml/kg to 220 or 82 ml/kg,
respectively (Benyajati and Yokota, 1989; Yokota and Benyajati, 1988).
Progressive increases in salinity from -0 to 460 mOsm/liter do not
affect TBW or muscle or liver water in the FW stingray (Bittner and
Lang, 1980), while 525 mOsm/liter decreases both total and muscle
water. Cserr et al. (1983) have shown that during transient salinity

perturbations, C1- space, but not non-C1- space, is regulated in the
skate, Raja erinacea, telencephalon, whereas all spaces in the medulla
are unregulated.
    Holmes and Donaldson (1969) concluded, primarily from Thor-
son’s survey of three FW and seven SW teleosts (Thorson, 1961), that
SW teleosts have a greater extracellular volume whereas blood volume
and TBW are the same. Examination of reported values (Tables III-V)
clearly illustrates that the variability in ECFV, which may be largely
methodological, obscures definitive confirmation of this hypothesis.
However, there does appear to be a decrease in muscle and TBW and
an increase in ECFV as teleosts are adapted to progressively higher
salinities, although exceptions to this trend are not uncommon.
    Few studies have used indicator dilution methods to compare
whole body ECFV in fish adapted to FW or SW. In the flounder,
Paralichthys lethostigma, ECFV appears to decrease after FW adapta-
tion (Hickman, 1972), although in this study only four fish, one each
adapted to FW for 4 days, 1 week, 2 week, and 3 months were exam-
ined. In rainbow trout, SW adaptation (3-4 weeks) slightly decreases
ECFV (Kellogg et al., 1990; M. D. Kellogg and K. R. Olson, unpub-
lished). Short-term (48 h) transfer of brook trout, Salvelinusfontinalis,
to SW does not significantly affect ECFV (Nichols et al., 1985). These
few studies do not permit a generalized hypothesis regarding the
affects of salinity on ECFV and clearly this is an open area for further
    A number of studies have reported the effects of salinity on ECFV
in individual tissues, and they provide some idea of fluid economy in
different environments. Salinity adaptation increases ECFV of muscle
tissue in Anguilla, Fundulus, Gillichthys, and Tilapia, although these
effects are often not substantial (Table V). Muscle ECFV also increases
in stenohaline Carpio when adapted to 380-400 mOsm NaC1, their
upper limit of salinity tolerance (Abo Hegab and Hanke, 1982). Gut
and kidney ECFV are greater in SW Anguilla, whereas liver ECFV is
greater in the FW fish (Table IV). However, salinity adaptation does
not affect muscle ECFV in the killifish, Fundulus heteroclitus,
(Schmidt-Nielsen, 1977) and, in fact, transfer from FW to 40-50% SW
increased TBW in Fundulus catenatus and F . olivaceus (Duff and
Fleming, 1972a,b). Ten days after SW transfer, TBW of 13-g trout also
returns to FW values (Bath and Eddy, 1979b), whereas eel TBW is 5%
less in SW after 3 weeks (Sokabe et al., 1973). Inulin spaces and TBW
in F . grandis do not temporally correspond to circadian oscillations in
the chloride balance (Spence et al., 1977). Generally, in teleosts, as in
cyclostomes, when ambient salinity increases tissue and whole body
ECFV, it does so at the expense of intracellular volume.
164                                                  KENNETH R . OLSON

    There is a well-known allometric reduction in ECFV with fish age
or size (Holmes and Donaldson, 1969). These are not considered to be
under physiological regulation in the context of this chapter.

   a . Exercise. In an early study, Stevens (1968) reported that severe
exercise is associated with an increase in body weight (1.5g) in 224-g
trout. Later, Milligan and Wood (1986a,b; 1987a,b) found that severe
exercise does not affect tissue or TBW in either FW trout or SW starry
flounder, Platichthys stellatus. However, exercise decreases total and
muscle ECFV ( [3H]mannitol space; -20% in trout and 28% in floun-
der) indicative of fluid influx into the intracellular compartment, pre-
sumably due to increased intracellular osmolarity (Milligan and Wood,
1986a,b, 1987a,b). Fluid spaces in both flounder and trout brain, and
probably heart and liver as well, are unaffected by exercise (Milligan
and Wood, 198613; Wood and Milligan, 1987). Extracellular fluid vol-
ume in both fish recovers in 4-8 h. Similar muscle-specific increases in
ECFV have been observed to accompany exercise in the lemon sole,
Parophrys vetulus (Wright et al., 1988).

    b. Environment: Temperature, Photoperiod, p H , and POS. In eels,
Anguilla rostrata, an increase in ambient temperature from 5 to 20°C
has no affect on total tissue water or inulin spaces in heart, liver, or
either red or white muscle (Walsh and Moon, 1982). Total tissue water
and C1- or Cl-/K+ spaces (Na+ space in liver) in trout tissues are
variously affected by seasonal acclimation and ambient temperatures
between 4 and 20°C (Mearow and Houston, 1980). Muscle water is
greater in summer than winter acclimated fish and is not appreciably
affected b y ambient temperature. Muscle ion spaces increase with
increasing temperature but are not affected by season. Cardiac water is
slightly greater in winter, and it is independent of temperature in both
summer and winter fish, whereas ion spaces are greater in summer and
increase with temperature in both acclimated groups. Gut water and
spaces behave somewhat similarly to muscle, although total water
decreases as temperature increases in summer fish. Liver water is
greater in summer fish and ion spaces increase slightly with tempera-
ture in winter fish. Total water in the brain is unaffected by season or
temperature, whereas C1- space increases with temperature. Murphy
and Houston (1977) have shown an interactive effect between temper-
ature (2 versus 1SOC) and photoperiod (6 versus 18 h light) on C1-/Kt
spaces and total tissue water in muscle and liver of rainbow trout. Low

light increases total water and decreases Cl-/K+ space in both tissues
at 2°C; at 18"C, photoperiod has no effect on either fluid compartment.
High temperature increases muscle water in 6-h light-exposed fish,
increases liver water at both photoperiods, and increases liver Cl-/K+
space in 6-h light-exposed fish. Low temperature (1°C) also disrupts
TBW regulation in trout transferred from FW to SW (Finstad et al.,
1988), and seasonal effects on TBW responses to salinity have been
observed (Finstad et al., 1989). In Channa punctatus muscle C1- space
is correlated with the reproductive cycle, doubling in October when
gonad weight increases (Sinha and Munshi, 1980). Total body water
also increases in spawning FW trout (Parry, 1961).
     Fish exposed to acidic water lose extracellular ions and may also
experience volume disturbances. While exposure of trout to acid water
(pH 4.0-4.5) for 1-3 days does not affect TBW, both ECFV ( L3HIman-
nitol or Cl-/K+ space) and plasma volume decrease, indicative of a
fluid shift into the intracellular compartment (Habe, 1987; Milligan
and Wood, 1982). Similar responses are found in the white sucker,
Catastomus commersoni (Habe and McMahon, 1988), although in
these fish total water in muscle (but not TBW) also increases (Habe,
 1987).Total body water in acid tolerant sunfish, Enneacanthus obesus,
is unaffected by acid exposure, whereas it increases in the less acid
tolerant Lepomis gibbosus; ECFV was not measured in these fish
(Gonzalez and Dunson, 1987).
     Despite the pronounced effects of acidic medium on extracellular
and intracellular fluid volume, related acid-base disturbances appear
to be relatively innocuous. Environmental hypercapnia has no affect
on ECFV ( [3H]mannitol) or total water of catfish, Zctalurus punctatus,
white or red skeletal muscle, heart, or bone (Cameron, 1985), or of sole,
Parophrys vetulus, white muscle, heart, and brain (Wright et al., 1988).
Adaptation of the air breathing teleost, Synbranchus marmoratus, to
 nitrogen-gassed (hypoxic) water for 4-5.5 days decreases [3H]inulin
 space from 163 to 149 ml/kg (Heisler, 1982), although it is not known if
 the response is due to hypoxia per se, or the attendant acidosis. An
 acidosis effect is more likely because iodoacetate, cyanide, or anoxia
 have no effect on total water or inulin space in isolated ventricular
 strips from trout (Nielsen and Gesser, 1984; Hansen and Gesser, 1987).

   c . Other Stress. Surgical stress, but not anesthesia per se, tran-
siently (<6 h) decreases muscle C1- space in Salvelinus fontinalis but
does not affect TMW (Houston et al., 1971). However, dorsal aortic
cannulation depresses muscle C1- space and increases TMW for up to
72 h (Houston et al., 1969).Twelve months of starvation does not affect
166                                                  KENNETH R . OLSON

liver water content in the eel, A. rostrata (Moon, 1983). Catechol-
amines, secreted by fish in response to a variety of stressors, may also
affect fluid volume. Although P-adrenergic blockade has no effect on
ECFV ( [3H]mannitol) or total water content in starry flounder muscle,
heart, and brain (Wood and Milligan, 1987), 5 x          M epinephrine
increases mannitol space in the in situ perfused heart without affecting
total water, indicating a loss in ICFV (Farrell and Milligan, 1986). A
similar increase in inulin or 32P04 space at the expense of intracellular
water has been observed by Gras et al. (1971) in l-mm-thick skeletal
muscle tissue slices. This effect can only be completely inhibited by
combined a and /3 blockade. Catecholamine-induced cellular dehydra-
tion of tissues contrasts with the well-known @mediated hydration of
red blood cells. The mechanisms of catecholamine action at the tissue
level are not known.

D. Blood Volume
    Total blood volume is influenced by the sum of factors that govern
the volume occupied by red cells and the plasma volume. It is assumed
in this chapter that regulation of total blood volume is achieved pri-
marily, if not exclusively, through adjustments in plasma volume, as it
is in mammals (Manning and Guyton, 1982). In this context, blood
volume is regulated by factors that affect ECFV (see Section II,C) and
b y factors that affect fluid movement between intravascular and inter-
stitial compartments.
    In the last 20 years nearly all blood volume measurements on fish
have employed indicator dilution methods. Apart from the require-
ment that the indicator must be confined to the vascular compartment,
these methods are identical to those already described for measuring
ECFV. The major constraints, loss of indicator from the compartment
and incomplete equilibration, apply here as well. In practice, both
kinetic and single-sample net retention techniques are used. Net re-
tention methods are simplified because indicators sufficiently large
enough to be restricted from the interstitial compartment are also
poorly filtered by the kidney. However, some indicators, especially
those diluted in the plasma, may distribute to extravascular com-
    Usually only one space, red cell (RCS) or plasma (PS), is measured
in a single experiment. Total blood volume (TBV) is then corrected for
the fractional volume of cells (hematocrit, Hct) or plasma by the rela-

               TBV   =   RCS/Hct, or TBV   =   PS/(1-Hct).          67)

Correction for plasma trapped between cells is often ignored, which
may result in under estimation of TBV calculated from plasma spaces
by several percentage points if whole blood is counted.
    Hematocrits for the above calculations are usually measured on
blood taken from a large vessel. This may bias TBV if large vessel
hematocrit (LVH) is not identical to the whole body hematocrit
(WBH).That the latter two are dissimilar has been known in mammals
for some time where the ratio WBH/LVH, commonly called Fcell ratio,
is around 0.91 (Albert, 1971). Fcel1 ratios calculated from experiments
on fish in which red cell and plasma volumes have been determined
simultaneously are considerably lower than 0.91. In hagfish, the Fcell
calculated from the data of Forster et al. (1989)on 20 h distribution of
1251-humanserum albumin, 51Cr red cells, and ventral aorta hematocrit
is 0.54. Substitution of subcutaneous sinus hematocrit (Forster et al.,
1989) results in a Fcell of 0.74, still well below mammalian values. Fcell
values in trout, determined from 4 h circulation of '251-bovine serum
albumin, 51Cr red cell spaces, and dorsal hematocrit, are similarly low,
around 0.78 (Gingerich and Pityer, 1989). The comparatively low Fceil
values reported in fish could be due to a proportionally greater plasma
volume in fish tissues, a red cell inaccessible plasma space (i.e., sec-
ondary circulation; see Part A, Chapter 4), or an artifact due to loss of
plasma indicator from the vascular compartment (see following dis-
    It becomes apparent in species where multiple indicators have
been employed that larger TBV are associated with PS indicators such
as Evans blue dye (T1824) or radioiodinated albumins than with radio-
labeled red cells (Conte et al., 1963; Duff and Olson, 1989; Forster et
al., 1989; Gingerich and Pityer, 1989;Gingerish et al., 1990; Huggel et
al., 1969; McCarthy and Conte, 1966; Wardle, 1971; summarized in
Table VI). Conte et al. (1963)also observed that TI824 is lost from the
circulation at a faster rate than 1311-human serum albumin. If T1824 is
allowed to bind with plasma proteins prior to injection into coho
salmon, Oncorhynchus kisutch, less dye is lost from the circulation 6 h
after injection and the estimated blood volume (45 ml/kg) is nearly
30% lower than when unbound dye is injected (62 ml/kg; Smith, 1966).
However, this is still around 30% higher than blood volumes reported
in other salmonids using tagged red cell methods and equivalent circu-
lation times (Table VI). This indicates that dye and protein together are
lost from the primary circulation because once Ti824 is bound to fish
plasma it remains bound for days (Nichols, 1987; Smith, 1966; Takei,
 1988). No dye is excreted in trout urine or bile for at least 24 h after
                                                       Table VI
                                                     Blood Volume
                                                          ~         ~

                                           volume                           Circulation
             Species         Environment   (mllkg)       Method"               time                 Reference
      Eptatretus stouti      sw             206                         K (104 h)         McCarthy and Conte (1966)

      Eptatretus stouti      sw              169                                          McCarthy in Holmes and
                                                                                           Donaldson (1969)

      Eptatretus cirrhatus   sw 16°C        177                         Inj (24 h)        Forster et al. (1989)
QI     Petromyzon marinus    FW              85                         K (30 min)        Thorson (1959)
       Lampetra planeri      FW              80                                           Bull and Morris (1967)
       Hydrolagus colliei    sw              52                         K (40 min)        Thorson (1958)
       Squalus acanthias     SW              68                         K (40 min)        Thorson (1958)
       Raja binoculata       sw              80                         K (40 rnin)       Thorson (1958)
       Raja rhinus           sw              72                         K (40 min)        Thorson (1958)
       Squalus acanthias     SW              66                         K (2 h)           Opdyke et al. (1975)
       Squalus acanthias     sw              79                         K (15 min)        Solomon et al. (1985b)
                                                                        20% Hct assumed
      Acipenser fluvescens   FW              37                         K (40 min)        Thorson (1961)
      Polyodon spathula      FW              30                         K (40 min)        Thorson (1961)
      Lepiosteus             FW              38                         K (40 min)        Thorson (1961)
    A tractosteus           FW         31    ”Cr-RBC   + Hct   Inj (60 min)    Siret et al. (1976)
    Clarias batrachus       FW         47                      Inj (30 min)    Pandey et al. (1978)
    Amia calva              FW         34                      K (40 min)      Thorson (1961)
    Neoceratodus forsteri   FW         49                      K (?)           Sawyer et al. (1976)
    Anguilla japonica       FW         35                      K ( 1 h)        Takei (1988)
    Anguilla rostrata       sw         28                      K (50 min)      Nishimura et al. (1976)
                            FW         29
                            (5 wk)
    Catastomus              FW         22                      K (40 min)      Thorson (1961)
    C yprinus carpio        FW         30                      K (40 min)      Thorson (1961)
    Cyprinus carpio         FW         53                      K (15 min)      Avtalion et al. (1974)
                                       53                      K (15miti)
    Heteromeustes           FW       13-20                     Inj (30 min)    Pandey et al. (1976)
g    fossilis
    Heteropneustes          FW         14                      Inj (30 min)    Pandey (1977)
    Zctiobus cyprinellus    FW         28                      K (40 min)      Thorson (1961)
    Oncorhynchus mykiss     FW         23                      Inj (10 min)    Schiffman and Fromm (1959)
    Oncorhynchus mykiss     FW         35                      K (30-90 min)   Conte et al. (1963)

    Oncorhynchus mykiss     FW         24                                      Smith (1966)
    Oncorhynchus mykiss     FW         35                      K (3h)          Nikinmaa et al. (1981)
    Oncorhynchus mykiss     FW         35                      K (45 min)      Huggel et al. (1969)
    Oncorhynchus mykiss     FW         56                      K (6 h)         Milligan and Wood (1982)

                                                        Table VI (continued)

                                              volume                                   Circulation
             Species            Environment   (mlikg)          Method"                    time                  Reference

I    Oncorhynchus m y k i s s   F W 6°C        34                              K (P)                 Barron et ul. (1987)
 I                              FW 12°C        33
                                FW 18°C        36
     Oncorhynchus mykiss        FW             34         51Cr-RBC+ Hct        Inj (4h)              D u f f e t al. (1987)
     Oncorhynchus mykiss        FW             41         'lCr-RBC + Hct       K (4 h)               Gingerich et al. (1987)
     Oncorhynchus mykiss        FW             32         51Cr-RBC+ Hct        Inj (4 h)             Duff and Olson (1989)
                                               60          lZ5I-A+ Hct
     Oncorhynchus mykiss        FW             35         "Cr-RBC + Hct                              Gingerich and Pityer (1989)
     (Wythville strain)                        51          12'I-A + Hct
                                               4s         "Cr-RBC + "'I-A
     Oncorhynchus mykiss
     (Wythville)                FW             53         51Cr-RBC + '"I-A     K (4 h)               Gingerich et a/. (1990)
     (Kamloops)                 FW             46         "Cr-RBC + 12sI-A
     Oncorhynchus mykiss        FW             41         Ti824 + Hct          K (24 h )             Nichols (1987)
     Oncorhynchus m y k i s s   SW             69         Tie24 + Hct          K (56 h)              Smith (1966)
     Oncorhynchus kisutch       SW             62         Ti824 + Hct          K (56 h)              Smith (1966)
                                               45         "bound" T1H24
     Oncorhynchus nerka         FW             54         Ti824 + Hct          K (56 h )             Smith (1966)
       Salvelinus fontinalis        FW                    55         Ti824 + Hct             K (5 h)                 Houston and DeWilde (1969)
       Salvelinus fontinalis        FW                    45         Inulin + Hct            K (4 h)                 Nichols et al. (1985)
                                    SW (48h)              44
       Sahelinus namaycush          FW                    43         'lCr-RBC + Hct          Inj (15 min)            Hoffert (1966)
       Amphipnous cuchia            FW                    31         Ti824 + Hct             Inj (30 min)            Munshi et al. (1975)
       Channa punctatus             FW                    30         Ti824 + Hct             K (30 min)              Sinha and Munshi (1981)
       Pseudoscarus                 SW                    36         Ti824 + Hct             K (40 min)              Thorson (1961)
       Enophrys bison               SW                    71         Ti824+ Hct              K ( 3 h)                Sleet and Weber (l983a)
       Epinephelus striatus         sw                    26        Ti824 + Hct              K (40 min)              Thorson (1961)
       Gymnothorax                  sw                    22        Ti824 + Hct              K (40 min)              Thorson (1961)
       Lutianus                     sw                    22        Ti824    + Hct           K (40 min)              Thorson (1961)
       Lutianus griseus             sw                    20        Ti824 + Hct              K (40 min)              Thorson (1961)
       Mycteroperca tigris          sw                    33        Tim4 + Hct               K (40 min)              Thorson (1961)
       Sphyraena barracuda          sw                    28        Ti824 + Hct              K (40 min)              Thorson (1961)
       Seriola quinquerdiata        SW                    47        Ti824 + Hct              K (45 h)                Yamamoto et al. (1980)
       Pomatomus saltatrix          sw                    43        Ti824 + Hct              Inj (10 min)            Ogilvy et al. (1988)
       Thunnus alalunga             sw                   132        1311-A + Hct             K (30 min)              Laurs et al. (1978)

          Abbreviations: A, albumin; Dil, red cell dilution; RBC, red blood cell; Hct, hematocrit. Other abbreviations as in Table 111. Blood volume
     expressed as ml/kg body weight.
172                                                       KENNETH R. OLSON

injection (Nichols, 1987). Tagged red cells are also removed from the
circulation by the spleen (Gingerich et al., 1987; Duff et al., 1987),
which may result in overestimation of blood volume by this method as
      It is likely that plasma-borne indicators, such as T1824 and albumin,
are lost from the vascular compartment in addition to slowly equilibrat-
ing with the secondary circulation. Gingerich and Pityer (1989) point
out that there is no direct correlation between the amount of secondary
vascularization of tissues and the albumidred cell space. Tissue Fcell
( F c e l l T ,the ratio of '251-albumin and 'lCr red cell calculated hematocrit
within a tissue to LVH) is high in a tissue with an extensive secondary
circulation, such as gill (0.96), yet low in muscle (0.44) which lacks
a secondary circulation. The opposite would be expected if the sec-
ondary circulation was a significant sink for 12'I-albumin. Further-
more, turnover of plasma in the gill secondary circulation is quite rapid
(-20 min; Olson, 1984; Olson et al., 1989) and inconsistent with the
rate of protein efflux observed in volume determinations. Nichols
(1987) found that in trout, TI824 was distributed to two anatomically
and kinetically distinct compartments that he attributed to plasma and
interstitial lymph. Blood volume calculated from the plasma com-
partment and hematocrit was 41 ml/kg; the interstitial-lymph com-
partment was 41-48 ml/kg (Nichols, 1987).The volume of the intersti-
tial compartment measured by Nichols (1987) is considerably greater
than that predicted from anatomical studies of the secondary circu-
lation (Vogel, 1985a,b; Vogel and Claviez, 1981) but less than that
determined with other indicators (i.e., inulin, T1824, etc.). Extrava-
sation of the albumin is further supported by several reports that sug-
gest some fish capillaries are relatively permeable to proteins (see
following discussion).
      Optimal mixing times for blood volume indicators have not been
established and often there appears to be a trade-off between adequate
mixing throughout the vascular compartment and loss of indicator from
the vasculature. Nichols (1987) reported that T1824 mixes with trout
plasma in 2 min and then slowly exchanges with the extravascular
compartment at the rate of 1% per minute. Adequate mixing of T1824 in
the shark, Squalus acanthias, takes 20-40 min, yet 30% of the dye
leaves the plasma every hour (Opdyke et al., 1975). In these instances,
short equilibration times for plasma indicators may provide better
estimates of intravascular volume as they minimize extravasation,
whereas longer mixing times may be more desirable from a practical
 standpoint, especially for kinetic methods (see Table VI). Volume
markers such as T1824 may not be suitable for use in fish in either

instance. Blood volume calculated from 51Cr red cell distribution is
less sensitive to mixing time; in trout there is a slight but insignificant
increase in calculated blood volume for 4 h after injection of 51Cr red
cells (Duff and Olson, 1989; Duff et al., 1987). However, tagged red
cells are not rapidly distributed to all tissues and at least 4 h is neces-
sary to ensure adequate mixing within individual tissues (Duff et al.,
1987). Longer circulation times (24)    may be preferable for studies of
organ blood volume (Duff and Olson, 1989; Duff et al., 1987; Ginger-
ich and Pityer, 1989; Gingerich et al., 1987, 1990).
    Several alternative methods have also been used to measure blood
volume. In some instances where indicator methods are not possible,
such as the small Lampetra planeri ammocoete, blood volume may b e
estimated from the total body hemoglobin (Bull and Morris, 1967).
Avtalion er al. (1974) estimated blood volume from the change in hemato-
crit after hemorrhage. In this method, an initial hematocrit (Hctl) is deter-
mined, a volume of blood (a) between 20 and 50% of the estimated blood
volume is removed, and 2 h later, when blood and extravascular fluids are
reestablished, a second hematocrit (Hctz) is determined. Blood volume
(BV) is then calculated from the relationship:
                       BV = a * Hctl/(Hctl-Hctz).                        (8)
Blood volumes determined with this method agree with those ob-
tained by TI824 or 1311-albumindilution methods (Table VI). However,
the imposition of hemorrhage stress may not be desirable in many
instances, and the error incurred by splenic contraction or depletion of
extracellular fluid volume needs to be examined.
    Blood volumes for representative fish are shown in Table VI. The
largest volumes are found in the hagfish followed by lampreys and
skates. In the hagfish, Eptatretus cirrhatus, 30% of the blood volume is
located in a subcutaneous sinus that extends over most of the body
(Forster et al., 1989). Hematocrit is lower in the subcutaneous sinus
than in the remaining vasculature (4.3 versus 13.5),and the time neces-
sary for equilibrium between the two compartments is in excess of 8 h
(Forster et al., 1989).The blood volume accounts for around 50%of the
extracellular space in Myxinoids. Brain blood volume in Petromyzon
marinus is also higher than that of other vertebrates (Heisey, 1968).
Blood volumes or spaces of individual organs or tissues have not been
systematically measured with indicator dilution methods in either
agnatha or chondrichthyes.
    Osteichthyes collectively have the lowest blood volumes of any
vertebrate (Table VI). The tunas appear to be an exception (BV = 132
ml/kg), perhaps due to their extensive heat-exchanging rate (Laurs et
174                                                   KENNETH R . OLSON

al., 1978). However, the results of the latter study (Laurs et ul., 1978)
should be confirmed because Brill (unpublished observations) mea-
sured blood volumes around 50 ml/kg in tunas using idocyanine green.
In their survey of the literature, Holmes and Donaldson (1969) con-
cluded that blood volumes of SW and FW teleosts are virtually identi-
cal. This information, summarized in Table VI, does little to refute this
hypothesis, although, as is the case with extracellular fluids, the varia-
tion attributable to methodology may preclude accurate comparison.
     Surprisingly, BV has rarely been measured in euryhaline fish
adapted to different salinities. Nishimura et al. (1976) adapted SW
eels, A. rostrata, to FW and found no change in BV (1251-albumin-
hematocrit) after 1, 2, and 5 weeks in FW (28 versus 29, 28, and
29 ml/kg, respectively); hematocrit was similarly unaffected. Similar
findings have been observed in rainbow trout adapted to SW for 3
weeks or longer (Kellogg and Olson, unpublished). Thus blood volume
appears to be well maintained in both hydrating and dehydrating

    Tissue fluid compartments have seldom been determined in fish,
although they have been perhaps most extensively studied in salmo-
nids. Vascular space and calculated blood volumes for a variety of
salmonid tissues are given in Table VII. Kidney vascular volumes are
the largest followed by gill and liver; muscle volume per unit tissue
weight is around 10-15% of that found in the kidney. Tissue blood
volumes determined with labeled red cells are consistently lower than
volumes calculated from 1251-albumin spaces except in gill and brain
(Duff and Olson, 1989; Gingerich and Pityer, 1989; Gingerich et al.,
1987, 1990) and sometimes kidney (Duff and Olson, 1989; Gingerich
and Pityer, 1989). The blood-brain barrier appears to effectively ex-
clude albumin from the interstitial compartment in trout and probably
all fish. An excellent review of the phylogenetic development of the
blood-brain barrier is provided by Cserr and Bundgaard (1984). In
agnatha (both hagfish and lamprey), holocephalans and teleosts, brain
vascular endothelium restricts macromolecular movement, whereas
glial end feet provide this barrier function in elasmobranchs (Cserr and
Bungaard, 1984).
    Hydrostatic pressures are greater in gills than in any other tissue,
and a tight endothelial and pillar cell barrier undoubtedly serves to
minimize fluid filtration. Anatomical studies have corroborated the
barrier capability of gills. In winter flounder, Pseudopleuronectes
americanus, and Antarctic cod, Pagothenia borchgrevinki, gill endo-
thelium and pillar cells prevent protein extravasation (Boyd et al.,
                                                               Table VII
                           Estimated Tissue Blood Volumes, "Cr Red Cell, and lZ5I-Albumin Spaces in Salmonids

                                       Blood volume                       51Cr RBC space*                       '"I-plasma   space*

                          Mean          Range         (Ref)       Mean          Range       (Ref)     Mean          Range             (Ref)
Brain                       59         18-100                     19            4-34                   39           14-85
Eye                         66         47-110                     15            9-22                   38           35-45
Fat                         37         33-40                       6            3-9                    24           17-31
Gill                       175         97-315                     48           32-63                  113           71-155
Heart (atrium)             625            -                      125              -                     -             -
Heart (ventricle)          125          47-250                    19            7-50                   68           19-181
Large intestine            104          65-175                    10            0-35                   64           11-108
Small intestine             70          23-120                     8            3-24                   46           10-78
Anterior kidney            32 1        157-509                   158           31-777                 260          197-434
Posterior kidney           367         252-525                    62           42-105                 252          144-475
Liver                      150          85-240                    29           17-48                  110           78-162
Red muscle                  24          17-38                      4            3-8                    16            9-22
White muscle                 8           4-12                      1            1-2                     6            3-9
Pseudobranch               332            -                       66              -                     -              -
Pyloric cecae               56          43-79                      5            4-5                    42           13-75
Spleen                    1830          34-6250                  765            7-1410                151          140-161
Stannius corpuscles        285             -                      57              -                     -              -
Stomach                     34          19-50                      4            2- 10                  22             6-39
Swimbladder                120          16-183                    22            3-45                  138              -
    * Values in ml/kg tissue weight.
   a  Oncorhynchus mykiss (Duff and Olson, 1989),30 min.
      Oncorhynchus mykiss (Duff et al., 1987),4 h.
      Oncorhynchus mykiss Wythville (Gingerich and Pityer, 1989), 4 h.
      Oncorhynchus mykiss (Gingerich et al., 1987),Hct=37 (4 h).
      Oncorhynchus mykiss Kamloops (Gingerich et ul., 1990),4 h.
   foncorhynchus mykiss Wythville (Gingerich et al., 1990),4 h.
      Oncorhynchus mykiss (M. D. Kellogg and K. R. Olson, unpublished), 8 h.
   I, Oncorhynchus mykiss (Stevens, 1968), Hct=27 (4 h).
     Saloelinus namaycush (Hoffert, 1966), Hct= 19.8 (15 min).
176                                                   KENNETH R. OLSON

1980) and eight different vital dyes, including T1824, are effectively
excluded from trout gill extravascular spaces (Davie and Daxboeck,
1982). Neither cationized nor native ferritin bind to or cross gill endo-
thelial or pillar cell barriers of winter flounder, Antarctic cod, sea
raven, Hemitripterus americanus, or Antarctic eelpout, Rhigophila
dearborni (Boyd et al., 1990). In contrast, the rete mirabile of the eel
swimbladder is quite permeable to ferritin and dextran (and to some
extent albumin) but is somewhat more restrictive to charged macro-
molecules (Rasio and Goresky, 1985).
    There are conflicting reports on fluid exudation and the importance
of colloid in preventing branchial edema in perfused gill preparations.
Bornancin et al. (1985) reported that in a perfused trout head prepara-
tion the gills become edematous when perfused with saline and that
edema could be prevented by addition of 4 g/liter bovine serum al-
bumin (BSA) and 4 g/liter gelatin to the perfusate. Similar observations
are reported in the eel, Anguilla australis. Profound edema is ob-
served in eel gills perfused with saline, 143% polyvinylpyrrolidone
(PVP; 30,000-40,000 molecular weight; MWT) or 2-3% dextran
(70,000 MWT), but not when gills are perfused with mammalian blood
or plasma (Ellis and Smith, 1983). Addition of 0.2% BSA to the PVP or
dextran perfusate does not prevent edema formation (Ellis and Smith,
 1983). Other studies indicate that perfused trout gills, in both whole
head (Perry et al., 1984) and isolated holobranch preparations (Olson,
 1984), are resistant to edema formation, even in the complete absence
of colloid in the perfusate. Factors other than colloid impermeability
may be involved in preventing fluid extravasation in the trout gill.
    The volume of blood in the central circulation (i.e., conducting
arteries and veins and cardiac chambers) is not known. Gingerich et al.
 (1990) compared the TBV, determined from combined 'lCr red cells
and 12'1-albumin spaces, to the sum of blood volumes in 13 tissues
 (also determined from red cell and plasma spaces) and found that even
though the tissues account for 60% of the total body weight, the com-
bined tissue blood volume is only 27-43% of the total volume. Ginger-
ich et aZ. (1990) suggest that either a portion of the unaccounted vol-
 ume is in tissues not examined (head and fins) or in the central vessels
 and heart. Table VIII extends this analysis using the organ weights
provided by Gingerich et al. (1990) and the average tissue red cell or
 plasma spaces from Table VII. Table VIII shows that although nearly
 90% of the total body weight is accounted for, the total red cell space of
 the tissues examined is only 4.44 ml per 883 g body weight or 5 ml/kg.
 This is only half of the "Cr red cell distribution volume measured in
 the whole fish (approximately 11 ml/kg; Duff et al., 1987; Duff and
 Olson, 1989; Hoffert, 1966; Gingerich and Pityer, 1989; Gingerich et

al., 1987, 1990). Presumably a significant fraction of the volume that
cannot be accounted for represents blood trapped in the heart and
conducting vessels and blood lost from large vessels in the tissues
when the tissues are removed. Thus the large vessel, or “central”
blood volume in salmonids and perhaps many teleosts, represents a
sizable reservoir that may be accessible under appropriate conditions.
It should also be noted that similar calculations based on ‘251-albumin
spaces cannot be made at this time because they grossly overestimate
the TBV. This is due largely to the lack of information on skin albumin
space. The only estimate available of skin albumin space is on the
plaice (607 pllg; Wardle, 1971). This value produces unusually high
skin and fin plasma spaces when used in the calculations for salmonids
(45 and 17 ml/kg fish weight, respectively) and results in calculated
blood volumes for whole fish in excess of 120 ml/kg. As Satchel1
(Chapter 3, Part A) notes, skin and fins are well endowed with a
secondary circulation and an accurate estimation of skin and fin blood
volumes is needed to estimate central blood volume.
    Table VIII also shows the estimated TBV and Hct calculated from
red cell and albumin spaces in each tissue. Red cell space in white
muscle accounts for less than 15% of the total red cell space in salmo-
nids (19% if the head is included). This is consistent with the relatively
low total extracellular volume found in this tissue (see earlier). Inter-
estingly, 51Cr red cell spaces in trout muscle remain constant between
12 and 240 min after injection of tagged cells indicating little, if any,
vasomotor activity (Duff et al., 1987). Splanchnic tissues, including
swimbladder but excluding spleen, account for 8% of the total red cell
space. Because many of the larger vessels, especially veins, are also
located on or near these tissues, it is likely, a considerable fraction of
the TBV is located in the splanchnic region. Kidney contains around
10% of the red cell space and gills around 25%. Splenic tissue, which
readily accumulates radiolabeled red cells (Duff et al., 1987; Duff and
Olson, 1989; Hoffert, 1966; Gingerich and Pityer, 1989; Gingerich et
al., 1987, 199O),is undoubtedly a source for red cells during emergency
situations (Pearson and Stevens, 1991; Yamamoto, 1987; Yamamoto
and Itazawa, 1989; Yamamoto et al., 1980,1983) but is probably similar
to the small intestine, on a per weight basis, in its ability to contribute
to fluid volume. ‘251-albumin space in the spleen (120 pl/kg fish) is
probably more representative of the actual vascular volume; this
equates to a red cell space of -50 pllkg fish. Blood distribution in
mammals, as a percent of TBV, is: muscle, 17%; splanchnic, 33%;
kidney, 1%; and pulmonary, 10% (Greenway and Lautt, 1986). Thus
fish have proportionally more blood in the respiratory organs (gills)
and kidney and less in the splanchnic, although the latter is probably
        178                                                               KENNETH R . OLSON

                                           Table VIII
                 Total 51CrRed Cell and '251-Albumin Spaces in Salmonid Tissues

                           Tissue       "'Cr red     ""I-albumin         Total blood
                          weight"      cell spaceh      space"            volume         Calculated
                         (g/kg fish)   (&kg fish)     (pllkg fish)            (PI)       hematocrit

Bone                        99.7          128'            631'                 759          17
Brain                        1.2           23              47                   70          33
Eye                          7.1          103             270                  373          28
Fat                          8.9           53             211                  264          20
Fins                        28.6          172*         17,630*              17,532           1
Gill                        20.0         1000           3,089                4,089          24
Head                        90.2          234'            997'               1,231          19
Heart                        1.4           13              95                  108          12
Large intestine              3.9           46             249                  295          16
Small intestine              2.7           23             125                  148          16
Anterior kidney              2.4          167             625                  792          21
Posterior kidney             5.7          311           1,435                1,746          18
Liver                        9.7          193           1,064                1,257          15
Ked muscle                  34.3          132             543                  675          20
White muscle               467.0          600           2,958                3,558          17
Pseudobranch                 0.1            7              18                  25           28
Pyloric cecae               10.1           45             424                  469          10
Skin                        74.8          449"         45,404"              45,853           1
Spleen                       0.8          642             120               -               -
Stannius corpuscles          0.01            1             13"                    14         7
Stomach                     13.3           53             293                    346        15
Swimbladder                  1.8           39             197'                   236        17
Total                      883.71        4434          76,168

    a   Tissue weight from Gingerich et al. (1990)for Oncorhynchus mykiss, Wythville Strain.
    " Spaces averaged from Table VII.
        Estimated from muscle.
    * Estimated from skin.
      Estimated from red and white muscle.
   f  Estimated from gill.
      From Wardle, 1971, red cell space = 6 pl/g, albumin space      =   607 p l / g .
      Estimated from kidney.
    ' Estimated from liver.

        artificially low. The proportion of blood volume in muscle is similar in
        both vertebrates.
            The calculated Hct (Table VIII) for tissues other than brain, eye,
        gill, and pseudobranch is lower than the average large vessel hemato-
        crit (28%)from which these data were obtained (Duff et al., 1987; Duff
        and Olson, 1989; Hoffert, 1966; Gingerich and Pityer, 1989; Gingerich
        et al., 1987, 1990). Low Hcts in many tissues are probably due to at
3.   BLOOD A N D EXTRACELLULAR FLUID VOLUME REGULATION                                 179

least two factors: albumin extravasation and the fact that these samples
may be biased toward capillary Hcts if blood is lost from large vessels
during sampling. Capillary Hcts in mammals may be 10% or less
(Johnson, 1971; Klitzman and Duling, 1979). This issue cannot be
resolved until a suitable indicator of plasma space is identified.
    The primary determinant of blood volume is the physical dimen-
sions of the vascular “container.” Within this constraint, blood volume
is regulated by factors that affect ECFV and by factors that affect
change between intravascular and interstitial fluid compartments. Rel-
atively little is known about these parameters in fish and, of necessity,
the general principles of blood volume regulation in mammals are
offered as a convenient, albiet preliminary, starting point. Figure 4
summarizes these variables; details are provided in Sections a-c.

   a. Extravascular Variables of Blood Volume Regulation. These
are factors that affect water movement between a fish and its environ-
ment or between intracellular and extracellular compartments. Extra-

     Cutaneous    + Gill


      Fig. 4. Factors that govern vascular and interstitial fluid volume. See Sections 3,a,b
for details.
180                                                               KENNETH R. OLSON

vascular variables are determinants of ECFV and thereby affect both
intravascular and interstitial compartments. Intracellular osmolyte
generation further affects fluid balance between intracellular com-
partments as described earlier. Regulation of extravascular variables
by the RAS, KKS, and ANP will be described in later sections.

    b. Intravascular-lnterstitial Variables of Blood Volume Regula-
tion. These are factors that govern fluid exchange between intravascu-
lar and interstitial compartments. Intravascular-interstitial variables
can be further divided into primary and secondary determinants. The
two primary determinants of blood volume are the so-called Starling
forces and return of filtered solute and protein via extravascular (e.g.,
lymphatic) pathways. Starling forces directly affect fluid balance across
the capillary and venule and are summarized b y the equation (Taylor,
1981; Taylor and Granger, 19842):

where JV is the net volume of fluid flow across the capillary wall per
unit time; Kf,, is the capillary filtration coefficient in volume per mmHg
pressure (usually also per unit tissue weight); P , and P, are hydraulic
pressures in capillary and tissue (interstitial) spaces, respectively; (T is
the reflection coefficient; and rr, and rr, are colloid osmotic (oncotic)
pressures in capillary and interstitial spaces, respectively. By conven-
tion, a positive value indicates either a pressure gradient or flow from
the capillary to the interstitium. In mammals, moderate increases in Jv
produce corresponding increases in lymph flow without significant
effects on interstitial fluid volume. Further increases inJv, beyond the
capacity of the lymphatics, result in tissue edema, exudation of fluid
into body cavities, or both. The importance, or even existence, of fish
lymphatics in return of ultrafiltrate and protein to the vasculature is not
    The secondary determinants of Starling forces are factors that affect
Kf,,, CT, P,, Pt, rrc, and rrt. Kt;, and (T have not been measured in fish and
although factors that might affect them are unknown, one can assume
that they are similar to those found in mammals. A summary of the
secondary determinants is provided in the following paragraphs.
    Kf,, is the permeability surface area product and is therefore an
index of both the hydraulic conductivity of the exchange vessel and the

      Other models for fluid filtration have been proposed to better explain transcapillary
fluid movement (Bert et al., 1988; Bassingthwaight and Goresky, 1984; Reed et al.,
1989). These are not considered in the present discussion.

exchange surface area. Hydraulic conductivity is directly related to the
number and radius of filtering pores and inversely related to barrier
thickness and viscosity of the filtering fluid (Taylor, 1981). Kf,, may
vary b y as much as 10-fold between tissues. Local or remote vasoactive
stimulants may affect Kf,, by altering barrier permeability or surface
area in addition to their effects on hydraulic pressure.
   The reflection coefficient (a)is an index of the effectiveness of the
wall of the exchange vessel in acting as a barrier to colloid (usually
plasma proteins). In its simplest form a is defined by the equation:

                           a = l-[col~I/[col,],                          (10)

where [ColJ and [Col,] are the concentrations of colloid (usually
protein) in lymph and plasma, respectively. a approaches 1in capillar-
ies that are impermeable to protein and becomes 0 when the protein
barrier is absent. It may also be affected by hormones (i.e., it is reduced
in frogs by atrial natriuretic peptide) (Huxley et aZ., 1987).
    Although a h a s not been directly measured in fish, protein concen-
tration in the subcutaneous sinus and peritoneal fluid of cod, Gadus
morhua, and flounder, Pleuronectes platessa, and in peritoneal fluid
from the wolf fish, Anarhichas Zupis, is similar to that found in plasma
(Hargens et al., 1974; Turner, 1937). This suggests that a may approach
0 in some fish tissues. A low (+ would offset the predicted low P , in
these tissues thereby stabilizing Starling forces. It is important to note,
however, that the fluid sampled by Hargens et al. (1974) and Turner
(1937) is derived from tissues with the highest apparent albumin permeabil-
ity (Table VIII) and may not be representative of all capillary beds. Tissues
that appear leaky to '251-albumin(Table VIII) may have lower (T values,
whereas a in gill, brain, and eye may, in fact, approach 1 (see earlier). a
may also approach 0 in venular endothelium of carp orbital tissue (Suzuki
and Hibiya, 1981b)or fins (Suzuki and Hibiya, 1981a)during inflammatory
reactions. The anatomy of elasmobranch systemic capillaries (Rhodin,
 1972) is suggestive of a leaky endothelium and very low a; this has been
corroborated by physiological studies (see later).
    Capillary hydrostatic pressure (P,) is determined by arterial and
venous pressure ( P a and P,, respectively) as discussed elsewhere in
this volume. At constant Pa and P,, P , will still increase or decrease if
the ratio of precapillary to postcapillary resistance ( R J R , ) decreases or
increases, respectively. Pulse pressure, often ignored in determining
filtration gradients, may also be important. Davie and Daxboeck (1982)
calculated that half of the total fluid flux moves back and forth across
the gill endothelium with each heart beat.
182                                                     KENNETH R . OLSON

   Tissue (interstitial) hydraulic pressure (Pt)opposes capillary filtra-
tion. Interstitial compliance (C,) determines the relationship between
interstitial volume (V,) and pressure according to the compliance rela-
                                Ct   =   AVtIAPt,                    (11)
where AV, and AP, denote the change in interstitial volume and pres-
sure, respectively. Pt is subatmospheric in many mammalian tissues
(- 1 to -6 mmHg; Guyton, 1991). Under normal circumstances Ct is
quite low in mammals, which prevents accumulation of interstitial
fluid. However, if filtration exceeds reabsorption and lymph removal
capabilities, slight accumulation of interstitial fluid greatly increases
Ct and substantial amounts of fluid will accumulate without a corre-
sponding increase in Pt (Aukland, 1984; Bert and Pearce, 1984; Bert et
al., 1988; Reed et al., 1989; Wiig and Reed, 1987).
    Measurements of P, in fish are limited to one elasmobranch, the
smooth dogfish shark, Mustelus canis, and five teleosts: bluefish, Po-
matomis saltatrix; pink salmon, Oncorhynchus gorbuscha; chum
salmon, 0. keta; chinook salmon, 0. tschawytscha; and coho salmon,
0. kisutch. All values were obtained with the cotton wick method
(Hargens and Perez, 1975; Ogilvy and DuBois, 1982). Average P, in
anterior and posterior muscle of seven unanesthetized smooth dogfish
is 1.5 and 2.1 mmHg, respectively, although pressures as low as -1.5
and - 1.0 mmHg, respectively, were observed in these tissues in one
fish. In anesthetized bluefish, anterior and posterior muscle Pt are
0.3mmHg; in one control, -0.7 and -0.2 mmHg were recorded. Head-
up tilting (30")in air, which places a hydrostatic load on the posterior
region, does not affect anterior Pt in the smooth dogfish but greatly
increases posterior Pt (to 10.3 mmHg). Head-up tilt slightly increases
anterior P, in the bluefish and increases posterior P, (to 1.0 mmHg).
Considerable edema and extravasation accompany head-up tilt in the
smooth dogfish, whereas neither are observed in bluefish. Sub-
cutaneous and peritoneal fluid pressures in all four species of Pacific
salmon are negative (--1 to -2 mmHg) when these fish are in sea-
water; however, upon their spawning migration into FW, fluid pres-
sures become positive (- 1 to 2 mmHg) and the fish develop profound
edema (Hargens and Perez, 1975).
    Plasma protein concentration is the primary determinate of rr, and
both are low in most fish. The approximate correlation between rr, and
protein concentration in fish plasma has been provided by Burton

                        rr, =             +
                                0 . 0 3 4 ~ 0.0073c2,                 (12)

where c is protein concentration in g/liter and reis in mmHg. Plasma
protein concentrations in most fish are between 20 and 50 g/liter. The
predicted r,,calculated from Eq. 12, is between 3.5 and 20 mmHg.
These pressures, and those determined by direct measurement of rC,
may overestimate physiological r,because both methods assume that,
as is the case in mammals, plasma albumins are a major component of
rC. Although albumins are found in all fish (Doolittle, 1984,1987), they
probably account for less than 20% of the total plasma protein and, in
subteleostean species, albumin-like molecules may exist as large mo-
lecular weight (150-360 kDa) monomers or dimers (Davidson et al.,
1988; Elger et al., 1987, 1988; Fellows and Hird, 1981; Filosa et al.,
1982,1986; Hilmy et al., 1978; Logan and Morris, 1981; Nagano et al.,
1976; Perrier and Perrier, 1978; Perrier et al., 1977; Yanagisawa and
Hashimoto, 1984). Two problems arise in calculating rC from plasma
protein concentrations. First, mammalian-size albumins (-60 kDa)
may not be retained by fish capillaries and hence they will be os-
motically ineffective. Second, large molecular weight proteins (the
major constituents of fish plasma) are less osmotically active on a per
weight basis. Errors are undoubtedly made in direct measurement of
r, (Table IX) as well, because the molecular cut-off of membranes
used in most oncometers is 10-30 kDa, and this is probably physiologi-
cally inappropriate for fish. This becomes apparent in the study
of polar fish by Hargens (1972) in which rTT,   is reported in excess of
100 mmHg (Table IX). Resolution of this issue awaits characterization
of the permselectivity of fish capillaries and identification of the rela-
tive proportion of physiologically important colloid to total protein.
    There has been some attempt to correlate elevated plasma protein
concentration in salmonids with salinity adaptation (Alexander, 1977)
or smoltification (Bradley and Rourke, 1984, 1988). Table IX also sug-
gests that rcis slightly greater in SW species. However, these relation-
ships have not been corroborated b y protein measurements in eel
(Robertson, 1984), Tilapia (Gupta and Hanke, 1982), or trout (Elger et
d.,  1988),nor by measurement of rein FW and SW eels (Keys and Hill,
1972). rTT, from 13-14 mmHg to 7-11 mmHg during the anadro-
mous spawning migration of Pacific salmon, although this is probably a
consequence of the associated degenerative processes observed in
these species (Hargens and Perez, 1975). Doolittle (1984,1987) claims
that, in view of the variability in size and concentration of proteins in
different fish in the same salinity, a significant, salinity-dependent,
adaptive effect of protein in volume homeostasis seems doubtful. Ad-
ditional measurements of r, and intravascularlinterstitial com-
partment volumes during salinity adaptation are required to resolve
this issue.
    184                                                         KENNETH R. OLSON

                                       Table IX
                                Plasma Oncotic Pressure

             Species                Environment    (mmHg)               Reference

 Eptatretus stouti                  sw                 7.7      Riegel (1978)
 Eptatretus stouti                  sw                10.5      Riegel (1986)
 Lampetra jluviatilis               20% sw             8.9      McVicar and Rankin (1983)
                                    30% SW             8.6
                                    40% SW             7.8
                                    50% SW             9.3
                                    FW female         10.7
                                    FW male            8.5
                                                       9.0      Immature
                                                       4.1      Maturity
                                    FW                 8.8      2-week adaption
  Raja erinacea                     sw                    2.7   Kakiuchi et al. (1981)
  Carcharias taurus                 sw                    3.4
  Mustelus canis                    sw                    3.0
  Dasybatus marinus                 sw                    2.7
  Raja fullonica                    sw                    2.4
  Chimaera monstrosa                sw                    2.0
  Galeus vulgaris                   sw                    2.0
  Squalus acanthias                 sw                    2.0   Turner (1937)
  Pristiurus catulus                sw                    1.9
  Raja oxyrhynchus                  sw                    1.6
  Etmopteris spinax                 sw                    1.3
    Anguilla japonica               FW                14.3      Kakiuchi et al. (1981)
    Carassius carassius             FW                 5.1
    C yprinus carpio                FW                 5.2
    Tinca vulgaris                  FW                 7.7      Keys and Hill (1972)
    Oncorhynchus mykiss             FW                 3.6      Duff and Olson (1989)
    Oncorhynchus gorbuscha          sw                13.3      Hargens and Perez (1975)
    Oncorhynchus keta               sw                14.6
    Oncorhynchus tschawytscha       sw                14.9
    Oncorhynchus kisutch            SW                13.9
    Esox lucis                      FW                 6.6      Keys and Hill (1972)
    Tantoga onitis                  sw                 9.5      Turner (1937)
    Sauda savda                     SW                17.2
    Anarhichas lupus                sw                11.3
    Opsanus tau                     sw                 7.7
    Molua molva                     sw                10.3      Turner (1937)
    Gadus morhua                    SW                 8.2      Hargens et al. (1974)

3.   BLOOD AND EXTRACELLULAR FLUID VOLUME REGULATION                               185

                                  Table I X (Continued)

          Species                   Environment     (mmHg)             Reference
Gadus morhuu                         sw                8.6     Turner (1937)
Gadus pollachius                     sw                6.9
Gadus cirens                         sw                3.0
Eleginus gracilis                    sw 5°C          129.0     Hargens (1972)
                                     SW -1.8"C       158.0
Myxocephalus scorpioides             SW -1.4"C        94.0
Brosmius bromsome                    sw               10.3     Turner (1937)
Lophius piscatoris                   sw                3.1
Belone acus                          SW               12.8
Echene.s naucrates                   sw               15.9
Cyclopterus lumpus                   SW                6.0
Prionotus carolinus                  sw                7.8
Prionotus strigatus                  sw                7.5
Scomber scombrus                     sw               14.6
Pseudopleuronectes americanus        sw                8.9
Pleuronectes platessa                SW                9.0
Pleuronectes platessa                FW                9.1     Hargens et al. (1974)
Anguilla vulgaris                    SW              20.7&     Keys and Hill (1972)
                                     FW               18.8"
                                     sw               10.3'
                                     FW                8.:3'
Chaenocephalus aceratus              sw               21.0     Hargens (1972)
Pseudochaenichthys                   sw                4.0
georgiaiius                          SW 2°C           59.0
Notothenia corriceps                 sw -1.8"c        75.0
                                     sw 2°C           52.0
Notothenia gibberifrons              sw -1.8"C        73.0

 Abbreviations as in Table 111.
 Starved 7 days.
 Starved 60 days.

    There does not appear to be a significant correlation between
protein concentration or rC   (Table IX) and blood pressure or phylog-
eny, although available data are limited. F, is as high, or higher, in
osmoconforming hagfish and osmoregulating lamprey as it is in tele-
osts, even though the hagfish arterial blood pressure is only -30% that
of a teleost (Riegel, 1986; Satchell, 1986). Riegel(1978,1986)found rC
was from 1.4 to 2 times greater than dorsal aortic pressure (7.7 versus
3.9 mmHg) in the Pacific hagfish, Eptatretus stouti, which raises ques-
186                                                   KENNETH R. OLSON

tions regarding the relationship of Starling forces to filtration even in
the kidney. This apparent imbalance may contribute to the dispropor-
tionately high blood volume in hagfish, which may be 50-60% of the
total extracellular fluid. However, it should be pointed out that consid-
erably higher arterial pressures (10.8 mmHg) have been recorded in
Eptatretus cirrhatus (Forster et al., 1988), and these are more in line
with the plasma oncotic pressures described by Riegel (1978; 1986).
Interestingly, injection of even small amounts of saline into hagfish
will raise blood pressure for several hours (Satchell, 1986), whereas
this has only a minimal transient effect in elasmobranchs (Opdyke et
al., 1975) and teleosts (Chester Jones et al., 1966; Duff and Olson,
 1986; Hipkins et al., 1986; Lee and Malvin, 1987). rc in lamprey is
surprisingly high (8-10 mmHg), especially in young fish (Table IX),
considering that plasma protein in the river lamprey, Larnpetra flu-
uiatilis, is only 38.5 g/liter (Robertson, 1984).
      r, is lower in elasmobranchs than teleosts, and this correlates with
slightly (20-30%) lower blood pressure in the former (Satchell, 1991).
In teleosts, there is also some correlation between arterial pressure
and rC,    both of which are notably elevated in tuna (White, 1988;Table
 IX), although pressure and T,have not been simultaneously measured
in individual fish or species. rcin eels (14-20 mmHg) is considerably
greater than would be predicted by dorsal aortic pressure (-20-
30 mmHg, Hipkins et al., 1986), but it is consistent with the measured
 protein concentrations (74 and 71 g/liter in FW and SW, respectively;
 Robertson, 1984). The reason for comparatively high r, and protein
concentration in eels is unknown.
      Little information is available regarding rt in fish. Turner (1937)
collected peritoneal fluid from the wolf fish, Anarhichas lupis, and
 observed that rt was 7.2 mmHg compared to 9.2 mmHg in serum.
 Similar findings have been reported by Hargens et al. (1974). Sub-
 cutaneous, peritoneal and plasma protein concentrations in cod,
 Gadus morhua, are 56, 51, and 57 g/liter, respectively; in flounder,
 Pleuronectes platessa, they are 56, 57, and 57 g/liter, respectively
 (Hargens et al., 1974). I n these tissues, rt virtually offsets rc and,
 assuming Pt is negative or near atmospheric (see previous discussion),
 it is difficult to envisage how intravascular fluid can be retained unless
 P , is exceptionally low (which is doubtful) or lymph return is high. The
 presence of additional factors or forces may be necessary to adequately
 explain transcapillary fluid balance in these vertebrates.
      The relationship among arterial blood pressure, blood volume, and
 capillary fluid balance can be examined when blood pressure is manu-
 ally controlled through an external reservoir connected to the dorsal
3.   B L O O D A N D EXTRACELLULAR F L U I D V O L U M E R E G U L A T I O N         187

aorta. In trout, a decrease in arterial pressure by 15 mmHg produces an
initial rapid loss of blood (-10 ml/kg) due to vascular decompression,
followed by slow, sustained loss of an additional 12 ml/kg fluid from
interstitial and intracellular compartments (Duff and Olson, 1989).
Both 51Crred cell and 1251-albuminspaces contract to the same volume
(16 ml/kg) after the hypotension. If arterial pressure is raised in two
consecutive 5 mmHg steps over 240 min, fish accumulate fluid. The
extent of volume expansion depends on the composition of fluid enter-
ing the fish from the reservoir (Fig. 5; Table X). Neither Ringer nor
Ringer-containing albumin are retained within the red cell (i.e., vascu-
lar compartment) even though 7r of the latter is hyperoncotic to trout
plasma. On the other hand, steelhead trout plasma is partially retained,
doubling the vascular space and resulting in a lower rate of fluid
accumulation. Vascular compliance, computed from the sum of the
rapid fluid intake volume for steelhead trout plasma, is around 1 ml/
mmHg-kg-'. Compliance of the Ringer and albumin distribution vol-
umes is in excess of 3 ml/'. These experiments support the
idea that 7rC in fish is determined by proteins larger than mammalian-

        ''    i

              0         40           80          120          10
                                                               6         200          240
                                             TIME ( m i d
     Fig. 5. Cumulative fluid intake by trout connected via the dorsal aorta to reservoirs
containing Ringer ( 0 ;N = 12).5% bovine serum albumin ( 0 ;N = 5), or steelhead plasma
(0; N = 13).Reservoir height was maintained at a height equivalent to resting arterial
pressure for 3 h then raised at time 0 min to produce a 5-mmHg increase in pressure. At
120 min (small arrow) reservoirs were raised an additional 5 mmHg.
188                                                             KENNETH R . OLSON

                                       Table X
      Effect of Increased Arterial Pressure on Calculated Blood Volume of Trout    a

                                                 Raised pressure

   Blood                                                             Steelhead'
  volumeb           Control        Ringer'        5% Albumin"          plasma

'251-albumin           59.6         140.3*            129.5*            122.7*
                      (k2.3)       (A24.7)           (k48.0)           ( A 12.7)
51Cr red cells         31.7          42.8              28.5                61.7*
                      (A3.0)        (55.7)            ("3.5)             (29.1)
Ratiod                  1.9           3.3*              4.5*                2.0
                     (N=14)        (N=12)            (N=5)             ( N = 13)

      Pressure was increased above resting in two 5-mmHg increments of 120 min each.
'251-albumin and 51Cr red cell spaces were determined after a 30-min circulation (270
min after initial pressure elevation).
      Blood volume (ml/kg) determined from albumin space/(l-hematocrit) or red cell
space/hematocrit; mean SE.
     Reservoir fluid.
      Ratio of albumidred cell blood volume.
    * Significantly ( P 5 0.05) greater than respective control.

size albumins and perhaps even albumin-size molecules in trout
plasma are of little significance in transcapillary fluid balance.

   c. Venous capacitance. Total blood volume (i.e., vascular capaci-
tance) is the sum of the unstressed and stressed vascular volumes
(Rothe, 1983a,b). The unstressed volume is the volume of blood re-
maining in the vasculature when mean circulatory filling pressure is
zero. Mean circulatory filling pressure is the blood pressure at any
point in the vasculature when cardiac output is zero, and after pressure
has equilibrated throughout the vasculature (Guyton et al., 1973).
Mean circulatory filling pressure is probably the major long-term de-
terminant of capillary hydrostatic pressure. The unstressed volume is
physiologically inert, and its removal does not affect cardiovascular
function even though the unstressed volume may be 60-75% of the
TBV (Greenway and Lautt, 1986; Rothe, 1983a,b). The stressed vascu-
lar volume is dependent on blood pressure according to the com-
pliance relationship:

where CI is vascular compliance, A V is the change in vascular vol-
ume, and APV is the change in vascular transmural (intraluminal-

extraluminal) pressure. Pv equates to mean circulatory filling pressure.
In mammals, systemic compliance and capacitance are largely attribut-
able to the venous system because 70-75% of the systemic blood
volume is located in the veins (Guyton, 1991; Hainsworth, 1986) and
venous compliance is around 25 times greater than arterial compliance
(Guyton, 1991). Sympathetic activation in response to exercise, hemor-
rhage, or other factors decreases unstressed volume in mammals by
increasing venous tone without substantially affecting compliance
(Greenway and Lautt, 1986; Rothe, 1983a, 1986). This shifts blood to
the stressed volume and, if other reflexes are prevented, will increase
venous pressure (and therefore capillary hydraulic pressure) according
to the compliance relationship. Thus the unstressed volume, because
of its physical size and the potential for vasoactive regulation, is an
important blood reservoir that is potentially affected by a variety of
vasoactive stimuli. Even changes in arterial resistance affect venous
capacitance through their effects on intravenous pressure; a decrease
in arterial pressure decreases venous pressure, which, at constant com-
pliance, decreases venous volume (Rothe, 1986). Nearly 70% of the
TBV mobilized by sympathetic stimulation in mammals is supplied by
the splanchnic circulation (Greenway and Lautt, 1986), which also has
the major portion of the total body compliance (Hainsworth, 1986). On
the other hand, atrial natriuretic peptide decreases mean circulatory
filling pressure and decreases blood volume; unstressed volume also
decreases due to passive vascular recoil and possibly venoconstriction
(Trippodo et al., 1986).Venous compliance is not affected by inhibit-
ing angiotensin I1 (ANG 11) formation with the converting enzyme
inhibitor, captopril (Ogilvie, 1985).
     Compliance characteristics of isolated rainbow trout vessels are
remarkably similar to their mammalian counterparts (Conklin and
Olson, 1990). Compliance of unstimulated anterior cardinal veins
(5-8.5 mm long) from 300- to 600-g fish is 2.1 k 1.0 pl/mmHg; some
21 times greater than 4- to 7-mm-long efferent branchial arteries
(0.10 t 0.03 pl/mmHg). Epinephrine, norepinephrine (both 10-lo -
lo-‘ M ) , or ANG I1 (lo-” -       M ) do not affect venous compliance,
whereas epinephrine increases, norepinephrine decreases, and ANG
I1 does not affect arterial compliance. Because trout anterior cardinal
veins contract in response to catecholamines (Olson et al., 1991),it is
possible that the major effects of adrenergic stimulation of trout veins
are on the unstressed volume. The efficacy of vasoconstrictors on other
venous reservoirs, especially those with little obvious smooth muscle
(Satchell, 1991; also see Satchell, Part A, Chapter 3 ) , remains to be
190                                                               KENNETH R. OLSON

    Vascular compliance has not been measured in cytostomes, and
only one attempt has been made to measure it in elasmobranchs.
Opdyke et al. (1975) monitored blood pressure in recently killed dog-
fish sharks during consecutive 5- to 10-ml injections of saline or
100,000-200,000 MWT dextran. Compliance calculated from their
data at 1 min after the first injection is -0.6 ml/'. Curiously,
blood pressure during subsequent injections decreases, indicating
large increases in compliance perhaps caused by stress relaxation (Op-
dyke et al., 1975) or LaPlace effects (final compliance after six injec-
tions is 6.3 ml/mmHg-kg-l). Injection of saline, urea, or dextran in-
creases plasma volume by as much as 65% yet has no effect on blood
pressure (Opdyke et al., 1975). Mean circulatory filling pressure, vas-
cular compliance, and unstressed volume have been measured in
freshly killed FW and SW adapted trout after incremental hemorrhage
or volume expansion with trout plasma from -60 to 160% of their
estimated blood volume (Fig. 6; M. D. Kellogg and K. R. Olson, unpub-
lished). With this method, mean circulatory filling pressure for FW
( n = 8) or SW ( n = 5) trout is 2.8 5 0.6 and 4.2 0.6 mmHg, respec-
tively; unstressed volume is 54 -+ 5 and 30 -+ 5% of estimated blood
volume, respectively; and vascular compliance is 5.5 +- 0.6 and 6.9                  *
0.9 ml/'. As shown in Fig. 6, compliance increases greatly
with volume expansion, as is the case in mammals.
    Opdyke et al. (1975) estimated mean circulatory filling pressure of
5.7 mmHg in the shark, Squalus acanthias, by measuring blood pres-
sure in freshly killed fish. Slightly lower values (3.8 5 0.4 mmHg, n =

                                 -   2    0     2  4     6    8
    Fig. 6. Vascular compliance curves for 5ix SW trout. Blood pressure is measured in
freshly killed trout during incremental hemorrhage or volume expansion between 60
and 160%of estimated blood volume. Mean circulatory filling pressure can be estimated
from the compliance curve at 100%(i.e.,control, volume [horizontal line]). Unstressed
volume is the blood volume at 0 mmHg pressure (vertical line). The slope of the curve is
equivalent to compliance.

5) have been determined in anesthetized trout during temporary car-
diac arrest (M. D. Kellogg and K. R. Olson, unpublished). The latter
agree with filling pressures determined from compliance curves above
(Fig. 6).
              THAT          REDISTRIBUTION
    Exhaustive exercise is perhaps the best known effector of apparent
fluid efflux from the vascular compartment of fish (Milligan and Wood,
1987a,b; Randall and Daxboeck, 1982; Stevens, 1968; Pearson and
Stevens, 1991; Yamamoto, 1987, 1988; Yamamoto and Itazawa, 1989;
Yamamoto et al., 1980, 1985). Hemoconcentration, interpreted from an
increase in Hct, may be produced by the addition of red cells to the
circulation through splenic discharge (see earlier discussion), by cate-
cholamine-induced red cell swelling (Milligan and Wood, 1987a;
Pearson and Stevens, 1991; Primmett et al., 1986),as well as by loss of
vascular fluid to the interstitium. In perhaps the most thorough series
of experiments to date, Yamamoto and colleagues (Yamamoto, 1987,
1988; Yamamoto and Itazawa, 1989; Yamamoto et al., 1980, 1985)
determined that 35% of the increase in Hct in exercising SW yellow-
tail, Seriola quinqueradiata, and 20-40% in exercising FW carp, Cy-
prinus carpio, was due to fluid loss from the vascular compartment (a
net decrease in plasma volume of 18 and -lo%, respectively). Pearson
and Stevens (1991) estimate that in trout, 50% of the increase in Hct
associated with strenuous exercise or air exposure is due to loss of
plasma from the circulation. This is equivalent to a 3-ml/kg reduction,
or around 13% of the plasma volume, and is similar to the findings of
Yamamoto's group. The mechanisms involved in capillary fluid efflux
during exercise are complex and probably include changes in P, and
Kf,, as well as branchial and renal responses. Catecholamines could
contribute somewhat to an exercise hypovolemia. In the perfused eel
tail, catecholamines decrease vascular volume by -2% and increase
interstitial (or possibly secondary circulation) volume by -3% (Davie,
 1982). Acid exposure also decreases trout blood volume by -20%
(Milligan and Wood, 1982). It is not known if an exercise induced
acidosis could have similar effects.
    Temperature acclimation of FW trout between 6" and 18°C does not
affect blood volume (Barron et al., 1987; Nikinmaa et al., 1981).
However, 2°C adapted brook trout, Salvelinus fontinalis, have a 20%
lower blood volume then 20°C adapted fish. A slight decrease in blood
volume at lower temperatures is also apparent in the air breathing singi
192                                                   KENNETH R. OLSON

catfish, Heteropneustes fossilis (Pandey et al., 1976). L-Thyroxine and
progesterone decrease and hydrocortisone increases blood volume in
the walking catfish, Clarias batrachus (Pandey et al., 1978).
    Anesthesia decreases TI824 space in trout, presumably due to im-
paired mixing of indicator (Smith, 1966). Conversely, hyperosmotic
hemodilution following dorsal aortic cannulation increases blood vol-
ume in the SW buffalo sculpin, Enophrys bison (Sleet and Weber,
    The relationships between blood pressure and blood volume and
the effects of environmental salinity on these cardiovascular parame-
ters have received surprisingly little attention. Blood pressure of
Squalus acanthias adapted to 70% seawater for 24 h is nearly double
that of seawater sharks (Solomon et al., 1988). The mechanisms medi-
ating the hypertension are unknown, although they might include
stress induced catecholaminergic responses that would increase pe-
ripheral vascular resistance and an increase in cardiac output resulting
from increased venous return. Increased venous return could be the
result of intra- and extravascular factors that increase vascular filling
pressure, including general hydration of the fish. Both SW adapted
eels, A. anguilla (Chan et al., 1967; Chester Jones et al., 1966, 1969),
and rainbow trout (Olson and Duff, 1992) have significantly lower
blood pressures than their FW counterparts. Because blood volume
does not decrease when either eels, A. rostrata, or trout are transferred
to SW (Kellogg et al., 1990; Nishimura et al., 1976), it is apparent that a
dehydration-induced hypovolemia cannot be the cause of hypotension
in these fish. Clearly other, perhaps multiple, effectors of pressure and
volume are involved. It is also possible that under certain circum-
stances there is a trade-off between pressure and volume regulation,
osmoregulation, or both such that one variable is conserved at the
expense of the others.
    The RAS and ANP systems deserve special consideration because
of their well-known involvement as antidrop and antirise effectors of
pressure and volume in mammals and because of their ubiquity in a
variety of fish. The KKS also impacts on fluid balance in mammals and
may have similar activities in fish. Although there is considerable
information regarding the effects of the RAS and ANP on blood pres-
sure, renal function, and to some extent osmoregulation in fish, there is
virtually nothing known about the involvement of these two systems,
or of the KKS, in volume regulation. For that matter, there is no known
mechanism for detecting either ECFV or TBW in mammals (Cowley
and Roman, 1989).The following sections describe how the RAS, KKS,

and ANP may affect these parameters and contribute to volume ho-


A. Components of the Renin-Angiotensin System
    The renin-angiotensin system is an enzyme-activated and peptide-
mediated effector of extracellular electrolyte and fluid balance in many
vertebrates. The RAS has been extensively examined in mammals.
Historically this has formed the basis for comparative studies in fish,
and a number of analogies between osteichthean and mammalian sys-
tems have been established. In nonosteichthean fish these analogies
are less evident, and it has been argued that these fish lack a RAS.
However, it is not yet clear if nonosteichthean fish actually lack a
complete RAS or if the criteria are too stringent or specific for a pri-
mordial system to be identified.
    The mammalian RAS has been described in a number of reviews
(Dzau et al., 1988; Hall and Brands, 1992) and can be summarized as
follows. Renin, a highly specific aspartyl proteinase, hydrolyzes the
decapeptide angiotensin I (ANG I) from the crz-globulin, angioten-
sinogen. A dipeptide is then hydrolyzed from the carboxy terminal end
of ANG I, principally by angiotensin converting enzyme (ACE; E.C., thereby producing ANG 11. In some instances the amino
terminal residue is also hydrolyzed and the heptapeptide des-Asp1-
ANG 11, or angiotensin I11 (ANG 111),is formed. Angiotensins may be
inactivated by a variety of peptidases. ANG I is for the most part
biologically inactive, whereas ANG I1 has potent vascular, renal, adre-
nal, and other effects. Angiotensin 111 is less vasoactive than ANG I1
but may be equipotent in stimulating aldosterone secretion; ANG
III-like homologs have not been reported in fish. Messenger RNAs for
angiotensinogen and renin have been identified in kidney, brain, adre-
nal, heart, vascular, and other tissues, suggesting a local RAS is opera-
tive. Renin may be circulated in the plasma (as much as 90% in the
inactive prorenin form) or bound to membranes, and it has been pro-
posed that in mammals the RAS has autocrine, paracrine, and endo-
crine functions.
    A number of comprehensive reviews have described the RAS in
fish (Henderson et al., 1980, 1985; Malvin, 1984; Nishimura, 1980a,b,
 1985a,b, 1987; Nishimura and Ogawa, 1973; Sokabe and Ogawa, 1974;
194                                                    KENNETH R. OLSON

Taylor, 1977; Wilson, 1984a,b).These provide an excellent summary of
the initial morphological and physiological evidence regarding the
phylogenetic development of the RAS.

B. Occurrence and Distribution in Fish
    A complete juxtaglomerular apparatus is absent in all fish, being
limited in teleosts by the lack of a true macula densa (Sokabe, 1974;
Sokabe and Ogawa, 1974; Nishimura, 1980a). Criteria for the presence
or absence of a RAS in nonmammalian vertebrates has been defined by
Sokabe et al. (1969) and Nishimura et al. (1973; summarized by Nishi-
mura, 1985b)as (a) presence in the kidney of granulated cells morpho-
logically and histochemically resembling juxtaglomerular ( JG) cells;
(b) time dependent production of an angiotensin, or angiotensin-like,
pressor substance on incubation of homologous plasma (substrate, i.e.,
angiotensinogen) with kidney extract (renin source) in the presence of
angiotensinase inhibitors, both plasma and renal extract being heat
labile; and (c)formed product must resemble angiotensin in its pressor
response in the rat, be resistant to a-adrenergic blockade in the rat
bioassay, dialyzable, heat stable, susceptible to a-chymotrypsin diges-
tion, and adsorbed onto Dowex 50W-X2 resin. The major disadvan-
tages of these criteria are the assumptions that angiotensin-like pep-
tides are pressor in all animals and the dependence on mammalian
angiotensin receptors to recognize heterologous peptides. Undoubt-
edly, application of modern molecular biology techniques will greatly
clarify this issue.
    Based on the above criteria there is little evidence to support the
existence of a complete RAS in cyclostomes. Cyclostomes lack granu-
lated JG cells, and angiotensin-like pressor activity has not been gen-
erated from cyclostome plasma by either homologous or heterologous
kidney extracts (Nishimura, 1980a, 198513).Renal renin activity has not
been demonstrated in cyclostomes (Nishimura, 198%) and, although a
pressor substance can be generated by Lampetra jluviatilis kidney
(Henderson et al., 1980),this is not unequivocally attributable to renin
activity (Nishimura, 198513).However, ANG I1 has pressor activity in
M . glutinosa (see later), suggesting some attributes of a receptor activa-
tion process and ACE-like activity has been found in Pacific hagfish,
Eptatretus stouti, plasma and liver (Lipke and Olson, 1988).Thus
some aspects ofthe RAS are present in cyclostomes. How they function
and what their contribution might be to volume homeostasis remain to

be determined. Because blood pressure appears to be the primary, if
not sole determinant of glomerular filtration rate (GFR) and urine
output, and hence fluid volume, in hagfish (Alt et al., 1981), it may be
that some volume regulatory mechanisms such as the RAS have an
extrarenal origin. The phylogenetic development of an association
between “juxtaglomerular” cells, initially with distributing renal ves-
sels in teleosts and ending with the glomerular circulation in mammals
(Sokabe, 1974; Sokabe and Ogawa, 1974; Sokabe et al., 1969), suggests
the evolutionary refinement of an initially systemic homeostatic sys-
tem for service in intrarenal regulation.
    a. Elasmobranchs. Anatomical studies of a variety of marine
sharks, rays, and the FW stingray, Potamotrygon circularis, have failed
to reveal granulated JG or specialized macula densa cells, even though
the distal nephron returns to the glomerulus as it does in mammals (see
Nishimura, 1985b for summary). This, plus the inability to demonstrate
angiotensin formation or renin activity in elasmobranch tissues (Nishi-
mura et al., l970,1985b), led Nishimura (1985b)to conclude that a RAS
was also absent in elasmobranchs. However, Lacy and Reale (1990)
observed a complete JG apparatus, including granulated afferent ar-
teriolar smooth muscle cells, close apposition of the distal tubule with
glomerular arteries and some specialization of macula densa-like cells,
and glomerular mesangial cells, in four marine elasmobranchs:
Squalus acanthias, Mustelus canis, Raja erinacea, and Rhinoptera
bonasus. Lacy and Reale (1990) postulated that the granules they
observed with electron microscopy may be a different form of renin
and thus undetectable using other histochemical methods. Granulated
peripolar cells are also found at the junction between the parietal and
visceral epithelia in the glomerulus of five species of elasmobranchs:
Raja erinacea, Mustelus canis, Rhizoprionodon terraenovae, Sphryna
lewini, and Rhinopetra bonasus (Lacy and Reale, 1989). These cells,
never before observed in fish, have been postulated to be involved in
osmoregulation via kallikrein-kinin or renin-angiotensin mecha-
nisms, although they are not immunoreactive with mammalian anti-
bodies to kallikrein or renin (Lacy and Reale, 1989).
    Other evidence, albeit limited, supports a RAS in elasmobranchs.
Renin-like activity has been reported in kidney extracts from Scylior-
hinus canicula (Henderson et al., 1981) and, in a preliminary report,
immunoreactivity with antibodies to ANG I1 [Val5] has been found in
the plasma, anterior kidney, rectal gland, hypothalamus, brainstem,
and pituitary of the nurse shark, Ginglymostoma cirratum (Galli and
196                                                  KENNETH R . OLSON

Kiang, 1990).However, Hazon et al. (1989)could not detect any immu-
noreactive ANG I1 (irANG 11) in S. canicula using antibodies against
ANG I1 [Ile5]. Abundant ACE-like activity is found in a variety of
tissues, especially brain, gill, heart, kidney, and rectal gland from the
little skate, Raja erinacea, and dogfish shark, Squalus acanthias (Lipke
and Olson, 1988). The ability of angiotensin to produce various physio-
logical responses in elasmobranchs (see following discussion) sug-
gests that the effector end of a RAS loop is also present.

    b. Holocephaluns. Based on current criteria, the holocephalans
appear to have a complete RAS. Granulated JG cells have been re-
ported in Hydrolagus colliei and Chimaera monstrosa (Nishimura et
al., 1973; Oguri, 1978). Modest renin-like activity is evident in
H . colliei kidneys (Nishimura et al., 1973).
   In general, osteichthyes exhibit most characteristics of a complete
RAS. In the few fish where RAS-like attributes are missing, one might
logically question the sensitivity, or selectivity, of the assay or the
experimental conditions.

    u. Subteleosts. A RAS is probably present in chondrosteans, al-
though information is somewhat fragmentary. Granulated JG cells are
not present in chondrosteans (Nishimura et al., 1973), however, there
is evidence for renin in shortnose sturgeon, Acipenser brevirostris,
and for renin and angiotensinogen in Nile bichir, Polypterus senegalus
(Nishimura et al., 1973). In holosteans, renin activity and plasma
angiotensinogen are found in both the longnose gar, Lepisosteus os-
seus, and bowfin, Amia calva; however, only L. osseus have granulated
JG cells (Nishimura et al., 1973; Youson and Butler, 1988). Renin
activity, angiotensinogen, and granulated JG cells are found in two
dipnoians, the African and South American lungfish, Protopterus ae-
thiopicus and Lepidosiren paradoxa (Nishimura et al., 1973); plasma
renin activity is also found in the Australian lungfish, Neoceratodus
forsteri, (Blair-West et al., 1977). Angiotensin converting enzyme-like
activity is present in chondrostean (shovelnose sturgeon, Scaphir-
hynchus platorynchus), holostean (A. calva), and dipnoan ( P . aethio-
picus) tissues and is often concentrated in gill, gut, kidney, heart, and
accessory respiratory organs (Lipke and Olson, 1988; Olson et al.,
 1987).Juxtaglomerular cells are found in the larger renal arteries of the
coelacanth, Latimera chalumnae, but are rare in arterioles and absent
 in afferent arterioles and mesangial tissue (Lagios, 1974).

    b. Teleosts. Granulated cells have been observed in the renal ar-
teries and afferent arterioles of a wide variety of teleosts (Nishimura,
1980b, 1985b; Sokabe, 1974). Granulated cells are often more numer-
ous in the terminal arterioles, decrease in number near the glomerulus,
and are absent in postglomerular vessels (Christensen et al., 1989;
Sokabe, 1974). They also appear to be innervated with adrenergic
nerves (Elger et al., 1984), although monoamine specific fluorescence
is absent in renal vessels of the toadfish, Opsanus tau (Madey et al.,
1984). Antibodies to human, rat, and mouse renins do not react with
salmon tissues (Christensen et al., 1989), although human, and to a
lesser extent murine, renin antibodies react with epithelioid cells in
renal arteriolar networks of the aglomerular lemon sole, Pleuronectes
microcephalus (Christensen et al., 1987). Receptor-type binding of
'251-ANG I1 [Ala', Val5] to glomeruli of FW trout has also been ob-
served (Brown et al., 199Ob).
    Stannius corpuscles contain renin-like activity (Chester Jones and
Henderson, 1965; Chester Jones et al., 1966; Ogawa and Sokabe, 1982;
Pang et al., 1981a,b; Sokabe, 1968; Sokabe et al., 1970)and irANG I1
(Yamada and Kobayashi, 1987).The amount of renin in Stannius cor-
puscles is less than 1% of that found in the kidney (Sokabe et al., 1970),
and it is probable that corpuscular renin-angiotensin has a paracrine
function. However, Stannius corpuscles may contribute to the regula-
tion of systemic blood pressure or volume as stanniectomy in the eel
lowers blood pressure by 33% (Chester Jones et al., 1966).
    To date, neither angiotensinogen nor renin from fish has been
purified and sequenced. Angiotensins from several teleosts have been
sequenced and synthesized, and their structures are compared to a
mammalian angiotensin in Table XI. In this chapter ANG I and ANG I1
denote the human sequence and residue substitutions are indicated in
brackets. There is considerable homology among all peptides with
variations occurring only at positions 1 , 5 , and 9. Some of the variation
in the amino terminal residue of Anguilla rostrata (and perhaps other
fish as well) may be due to a plasma L-asparaginase amidohydrolase
capable of converting asparaginyl angiotensins to aspartyl angioten-
sins (Khosla et al., 1985). Val5 is common to all fish, whereas the
residue in position 9 is species specific. Plasma and renal renin and
plasma angiotensin levels in fish appear to be generally lower than
levels in mammals (Nolly and Fasciolo, 1972; see also Nishimura,
    Angiotensin converting enzyme-like activity is found in tissues of
nearly all teleosts, although enzymatic activity is generally lower in
teleosts than in other vertebrates, including subteleostean species
                                                                  Table XI
                  Fish Angiotensins Formed by Incubation of Homologous Plasma with Renal or Stannius Corpuscle Extracts

                                                                            ANG I

                                                                   ANG I1

                Species                      1       2       3      4       5       6    7      8       9      10           Reference

Human, rat, dog, pig, horse, rabbit,      Asp       Arg    Val    Tyr    Ile     His    Pro    Phe    His     Leu    Akagietal. (1982)
 guinea pig
Lophius litulon (aglomerular goosefish;   Asn                            Val                          His            Hayashi et al. (1978)
  renal or SC*)                                                                                                      Hasegawa et al. (1984)
Oncorhynchus keta (chum salmon;           Asn                            Val                          Asn            Takemoto e t al. (1983)
 renal or SC)                             Asp**                          Val                          Asn
Anguilla rostrata (American eel; renal)   Asn                            Val                          GlY            Khosla et al. (1985)
                                          Asp**                          Val                          GlY

Anguilla japonica (Japanese eel; renal)   Asp**                          Val                          GlY            Hasegawa et al. (1983)
                                          Asn                            Val                          GlY

   Abbreviations: SC, Stannius corpuscles. *, minor ANG in SC not identified. **, may be due to in eitro conversion from Asn (Khosla et a / . ,
1985). Multiple sequences within a species are listed in descending order of' prevalence.

(Lipke and Olson, 1988; Olson et al., 1987; Polanco et al., 1990). Trout
ACE is similar to mammalian pulmonary ACE with respect to K , and
chloride dependence; however, trout ACE has a higher p H optimum
and is active over a greater range of pH, presumably reflecting the
tendency of fish toward higher plasma alkalinity (Lipke et al., 1987).
Teleost gills frequently have higher ACE activity than other tissues
(Galardy et al., 1984; Lipke and Olson, 1988; Olson et al., 1987;
Polanco et al., 1990).In gills, ACE is predominant in vascular endothe-
lium and pillar cells of secondary lamellae (Olson et al., 1989). Because
the entire cardiac output must first traverse the lamellae before enter-
ing the systemic circulation, the gills are in an ideal position to condi-
tion arterial blood; in fact, the isolated perfused gill converts over 60%
ofANG I to ANG I1 in a single pass (Olson et al., 1986).ACE activity in
nonrespiratory tissues may also contribute to circulating ANG I1 or
may indicate the presence of a local RAS.
    Kohama et al. (1988)purified a novel ACE inhibitor from muscle of
the tuna, Neothunnus macropterus, the sequence being: Pro-Thr-His-
Ile-Lys-Trp-Gly-Asp. The physiological significance of this octapep-
tide is not known, although its presence could account for the un-
usually low ACE activity in tuna tissues (Lipke and Olson, 1988)and
the relative insensitivity of tuna blood pressure to '4CE inhibitors
(R. W. Brill and K. R. Olson, unpublished). In contrast, vascular ACE
activity is quite prevalent in trout tissues. Equal doses of ANG I [Asn',
Val5, Asng] or ANG I1 [Asn', Val5] produce an identical increase in
resistance during single-pass perfusion of either skeletal muscle-
kidney or splanchnic circulations, whereas in the presence of an ACE
inhibitor, ANG I is ineffective (K. R. Olson and R. Ferlic, unpub-
    Angiotensins are inactivated b y peptidases, which are abundant in
most tissues including kidney, Stannius corpuscles (Chester Jones et
al., 1966),and the extensive alamellar vasculature in the core of the gill
filament (Olson et al., 1986). Angiotensin metabolites may be selec-
tively removed by the liver and secreted into the bile (Olson et al.,
 1986), although the catabolic process has not been examined in detail.
The half-time (tllz) for circulating ANG I1 in trout, estimated by a
variety of methods, is 3-7 min (Kellogg and Olson, 1990). In fact, based
 on the rate of ANG I1 infusion necessary to maintain blood pressure in
ACE inhibited trout, the rate of ANG I1 inactivation in vivo appears to
be limited as much by tissue perfusion as it is by peptidase activity
 (Kellogg and Olson, 1990). Thus the RAS in fish appears to be a rapid
 effector of blood pressure.
200                                                    KENNETH R. OLSON

C. Stimulus for Activation of the RAS
    Hypotension, hypovolemia, and, to a lesser extent, osmotic pertur-
bation are the best known activators of systemic RAS responses in fish
although there are considerable species and experimentally induced
variations. It is inherently difficult, if not impossible, to separate these
stimuli, especially at the level of the intact animal, and, therefore,
much of the afferent limb of RAS control remains to be described.
    Studies in elasmobranchs indicate that hypotension, hypovolemia,
or both stimulate a RAS-like response. Angiotensin converting enzyme
inhibitors generally have little effect on blood pressure in resting
sharks (Opdyke and Holcombe, 1976; Hazon et al., 1989),which can be
interpreted as lack of evidence for a RAS. However, moderate hypo-
tension produced by the smooth muscle relaxant papaverine in the
dogfish, Scylorhinus canicula, makes captopril an effective hypoten-
sive drug, presumably by blocking the now active RAS (Hazon et al.,
1989). Furthermore, a 30% hemorrhage increases immunoreactive
ANG II [Val5]in plasma, hypothalamus, brainstem, and pituitary of the
nurse shark, Ginglymostoma cirratum; plasma irANG I1 [Val5] is also
increased in G . cirratum after removal of the rectal gland or adaptation
to 50% seawater for 7 days (Galli and Kiang, 1990).
    Hemorrhage, or other perturbations that produce hypotension or
hypovolemia, usually activates the teleost RAS. Consecutive l-ml
hemorrhages increase drinking and plasma ANG I1 in FW eels,
Anguilla japonica (Kobayashi et al., 1989).Transfer of eels to SW after
the first hemorrhage, which presumably exacerbates the hypovolemia,
enhances drinking, and further increases plasma ANG I1 (Kobayashi et
al., 1980).In the aglomerular toadfish, Opsanus tau, either a decrease
in blood pressure (produced by captopril, isoproterenol, or papaver-
ine) or a combined decrease in blood pressure and volume produced
by repeated small volume (-5% of blood volume) or rapid, large vol-
ume hemorrhage increases plasma renin activity (PRA; Fig. 7; Madey
et al., 1984; Nakamura and Nishimura, in: Nishimura and Bailey, 1982;
Nishimura and Madey, 1989; Nishimura et al., 1979). Bailey and Rand-
all (1981) have also observed a positive correlation between the
amount of hemorrhage and PRA in intact trout and between renin
release and perfusion pressure in the perfused kidney. However, Hen-
derson et al. (1985)found that neither 15 nor 30%hemorrhage affected
irANG I1 in the eel, A . anguilla, and they hypothesized that plasma
sodium concentration, but not plasma volume, is the primary stimulus
for activation of the RAS in fish. It should be noted that Henderson et
al. (1985) measured ANG I1 30 min after hemorrhage, which might

                         I          blood pressure
                                                             -0      D

                                                             --I2    ;
                                                             - -16   v,
                                                             --20    -

                               (5) (10) (20) (40) (60)
                                BIOODLOSS ml/kg
                                ( % e s t blood vol)
    Fig. 7. Relationship between plasma renin activity (PRA) and blood pressure after
graded hemorrhage in the unanesthetized toadfish, Opsunus tau. Mean *SE, N = 8;+,
P<0.05; ++, P<O.Ol. [Redrawn from Nishimura er al. (1979) with permission.]

provided sufficient time for pressure and volume restoration (Duff and
Olson, 1989).
    Volume expansion would be expected to produce an inactivation of the
RAS. Expansion of Australian lungfish, Neoceratodus forsteri, with iso-
osmotic NaCl (0.6% at 28 ml/ for 5 h) decreases PRA by 50%
(Blair-West et al., 1977), consistent with the predicted response.
However, infusion of hypoosmotic saline (0.3%)at the same rate in-
creases PRA to 140% above control (Blair-West et al., 1977). In both
experiments dorsal aortic pressure increases by -2 mmHg while
plasma sodium decreased during 0.3%saline infusion (Blair-West et
al., 1977). The mechanism for the elevation in PRA in these fish re-
mains to be determined.
    Transfer of fish between hypo- and hyperosmotic environments has
been used to examine the effects of ion, volume, or pressure perturba-
tion on the RAS. While ionic changes are readily quantified, volume
and pressure measurements have rarely been attempted in the same
experiment, and it is difficult to evaluate cause and effect relationships
202                                                  KENNETH R . OLSON

between cardiovascular parameters and the RAS. If one assumes that
the general effect of a hyperosmotic environment is to increase extra-
cellular ions, increase (or not change) extracellular and blood volume,
and decrease blood pressure and intracellular water (see earlier sec-
tions), then three parameters-plasma ions, blood pressure, and cellu-
lar dehydration-would be predicted to have the greatest stimulatory
effect on the RAS. Correlative relationships between plasma and tissue
renin activity or angiotensin concentration may be further confounded
by different responses in systemic versus local RAS or by concomitant
alteration in turnover rates. Further experiments will need to be care-
fully designed to separate these variables. Known relationships be-
tween salinity transfer and the RAS are summarized in the following
paragraphs, and it is evident that species differences contribute addi-
tional sources of variation.
    The effects of salinity on the RAS have been most extensively
studied in eels. Henderson et al. (1976) found that PRA of eels,
Anguilla anguilla, transferred to SW slowly increases over 3-5 days to
levels more than twice as great as FW fish and is maintained thereafter
at this level. Plasma sodium and osmolarity are also greater in SW fish
(Henderson et al., 1976). Transfer of SW fish to FW rapidly (4-24 h)
lowers eel PRA (Henderson et al., 1976). Plasma irANG I1 in SW
adapted eels, A. anguilla, is nearly 20 times greater than that of eels
adapted to distilled water; values for FW and 50% seawater are be-
tween these extremes (Henderson et d., 1985). Similar qualitative
responses have been reported by Nishimura et al., (1976) and Okawara
et al. (1987). In A. rostrata,transfer from SW to FW (3days to 5 weeks)
decreases PRA, angiotensinogen, renal renin activity (RRA), and
plasma sodium, but does not affect blood volume; RRA returns to SW
level after 2-5 weeks (Nishimura et al., 1976). Because no correlation
between PRA and either plasma sodium or cortisol is apparent in
either eels or toadfish, Nishimura et al. (1976) hypothesized that PRA
may be correlated with blood pressure. In A. japonica, plasma irANG
I1 does not change during the first 2 h after transfer to SW even though
drinking initially increases, whereas over the next 3 days a further
increase in drinking is accompanied by an increase in plasma irANG I1
(Okawara et al., 1987). Captopril prevents the delayed increase in
drinking but does not affect the early response (Okawara et al., 1987).
Conversely, PRA has been shown to only transiently increase in A.
japonica transferred from FW to 30, 50, or 100% SW or air before
returning to FW levels (Sokabe et al., 1973; Takei et al., 1988).
    Effects of salinity transfer on the salmonid RAS are less clear.
 Granulated epithelioid cells in renal arteries of Atlantic salmon, Salmo

salar, decrease in number between l-year FW and 2-year SW fish,
which the authors suggested might be indicative of reduced renin
secretion b y FW fish (Christensenet al., 1989; although see following).
Salinity adaptation of rainbow trout reduces glomerular size, flattens
podocytes, and increases pedicel density, often occluding the filtration
pores (Brown et al., 1983; Gray and Brown, 1987). Angiotensin I1
[Asn', Val5] infusion for 45 min into FW trout, or added to trout glomer-
uli in oitro, produces similar effects except that glomerular diameter is
unaffected (Brown et al., 1990a; Gray and Brown, 1987). However,
PRA in trout adapted to SW for 10 days is not different from FW fish
(1.65versus 1.19 ng ANG I e q / m l / l 6 h; Brown et al., 1980).
    In nonsalmonids, RRA is usually inversely related to environmen-
tal salinity, whereas the effects on PRA are less obvious, perhaps
suggesting that the predominant effect of the RAS is on kidney func-
tion. Renal renin activity is elevated when aglomerular (Opsanus beta
and Hippocampus hudsonia) or glomerular euryhaline (Alosa sapi-
dissmia and Pornolobus pseudoharengus) teleosts are adapted to dilute
media (Capelli et al., 1970). PRA does not change when Tilapia moss-
ambica are transferred from SW to FW for from 1to 7 days (Malvin and
Vander, 1967), although RRA decreases after transfer from FW to SW
(Sokabe et al., 1968,1973). Plasmarenin activity does not change when
aglomerular toadfish, Opsanus tau, are transferred from 50 to 5% SW
(Nishimura et al., 1976). In other fish, RRA is also higher in FW species
(Capelli et al., 1970; Mizogami et al., 1968).
    Other factors that affect fish RAS have received but little attention.
In trout, unionized ammonia increases ir-RRA in a log-dose related
fashion (Arillo et al., 1981); this may, in turn, be the cause of the
hypertension (Smart, 1978) and diuresis (Lloyd and Orr, 1969) associ-
ated with ammonia toxicity.
    Unlike the situation in mammals, renin secretion in fish does not
appear to be regulated by adrenergic mechanisms. The renin response
of the perfused trout kidney to reduced perfusion pressure is not
affected by prior a, p, or combined a + p adrenergic blockade (al-
though basal secretion appeared to be decreased and increased by a
and p blockade, respectively; Bailey and Randall, 1981). A similar
insensitivity of renin secretory mechanisms to the fi adrenergic agonist
or antagonist, isoproterenol or propranolol, respectively or to chemical
sympathectomy is found in intact toadfish, Opsanus tau (Madey et al.,
1984; Nishimura and Madey, 1989). Renin secretion by the toadfish
kidney also appears to be independent of prostanoid regulators
(Madey et al., 1984). Renin secretion from superfused toadfish kidney
slices is inhibited by polarizing concentrations of K+ (50 mM) or the
204                                                  KENNETH H. OLSON

calcium channel agonist Bay K 8644 and is restored by the calcium
channel antagonist nifedipine (Nishimura and Madey, 1989). Further-
more, renin secretion by toadfish kidney slices does not appear to be
regulated by /3 adrenergic, cholinergic, cyclic adenosine monophos-
phate (CAMP),or cyclic guanosine monophosphate (cGMP) mediated
systems (Nishimura and Madey, 1989).
    Reflex inhibition of the RAS in FW eels, A. anguilla, may be medi-
ated in part through arginine vasotocin (AVT) as exogenous AVT re-
duced PRA activity by 60% (Henderson et al., 1985).However, in SW
eels AVT effects are less evident, perhaps due to the elevated plasma
AVT levels in these fish or other factors (Henderson et al., 1985).
Feedback regulation of the RAS by other hormonal signals has not
been critically examined.
    Little is known regarding control of ACE activity in fish tissues.
Polanco et a2. (1990) found that ACE activity in carp, C yprinus carpio,
gill, but not kidney, was reduced by 50%after 1week in hypoxic water.
However, in these experiments hypoxia was produced by raising water
temperature from 10"to 25"C, and it is not clear if the response was due
to hypoxia per se or to temperature. In obligate air breathing fish the
dependence on branchial ACE is reduced commensurate with a reduc-
tion in gill vascularity, and in these species there is a corresponding
increase in ACE activity in the accessory respiratory organs, presum-
ably reflecting their increased vascularity (Olson et al., 1987).

D. Effects of Angiotensins
    T h e RAS appears to serve primarily as an antidrop effector of blood
pressure, fluid volume, or both in fish. Angiotensin-mediated pressor
responses have been demonstrated in virtually all species examined.
Dipsogenic and antidiuretic effects attributable to the € U S have also
been reported in select species suggesting the importance of this sys-
tem in volume homeostasis. Pressor doses of exogenous angiotensin
may produce hypovolemia by increasing fluid filtration across glomer-
ular and other systemic capillaries. Perhaps this is a physiological
response, but until more evidence is available it can be assumed to be a
pharmacological artifact. Surprisingly, the effects of the RAS on vascu-
lar or extracellular fluid volume have not been directly examined.
    Information on RAS actions in cyclostomes is limited. Angiotensin
11 is pressor in hagfish, Myxine glutinosa, but its effects appear to b e
mediated exclusively by catecholamines as angiotensin responses are

abolished by prior a-adrenergic blockade (Carroll and Opdyke, 1982).
Drinking cannot be induced in either the marine hagfish, Eptatretus
burgeri, or FW arctic lamprey, Lampetra japonica japonica, by intra-
peritoneal injection of high (50-100 mg/kg) doses of ANG I1 [Am',
Val5] (Kobayashi et al., 1983).
    a. Cardiovascu1ar.A variety of angiotensins including ANG I and
11,ANG I [Asp', Val5, Serg], ANG I1 [Asn', Val5], and ANG I11 produce
phentolamine sensitive pressor responses in Squalus acanthias (Car-
roll, 1981; Carroll and Opdyke, 1982; Khosla et d.,1983; Opdyke and
Holcombe, 1976), and ANG I1 [Asp', Ile'] is pressor in Scyliorhinus
canicula (Hazon et al., 1989). Angiotensin I1 [Asn', Val'] (5 pg/kg)
injection increases plasma epinephrine and norepinephrine in S.
acanthias by -10-fold (Opdyke et al., 1981), and this probably ac-
counts for the entire pressor effect. Angiotensin converting enzyme
inhibitors SQ 20881 or captopril inhibit pressor responses to any form
of ANG I, but do not affect resting blood pressure in either S. acanthias
or S. canicula (Hazon et al., 1989; Khosla et al., 1983; Opdyke and
Holcombe, 1976), suggesting that the RAS is not important in resting
blood pressure regulation in elasmobranchs. However, in S. canicula,
captopril is hypotensive if administered after the smooth muscle relax-
ant papaverine (Hazon et al., 1989), indicating that a RAS-like re-
sponse may be activated under certain circumstances. Classical mam-
malian angiotensin antagonists, ANG I or I1 [Ile'], ANG I1 [Sar', Ile']
or ANG I1 [Sar', Thr8] do not block S. acanthias receptors but are often
pressor as well (Khosla et al., 1983).Angiotensin I1 does not affect
cardiac contractility in S. acanthias (Opdyke et al., 1982).
    In vitro studies confirm the dependence of the angiotensin re-
sponse on catecholamines. ANG I or I1 has no effect on vascular
resistance in the perfused gut (Opdyke and Holcombe, 1978) or on
systemic or gill resistance of whole-body perfused dogfish, S.
acanthias (Opdyke et al., 1982). Topical application of ANG I1 [Asn',
Val5] to mesenteric circulation in zjivo or ANG I1 [Asn', Val5] added to
celiac artery or anterior intestinal vein strips in vitro is likewise inef-
fective. However, if chromaffin tissue is added to the medium in which
arterial strips are incubated, then ANG I1 [Am', Val5] stimulates cate-
cholamine release and contracts the muscle (Carroll, 1981).

   b. Renal. Much of the control of GFR in sharks appears to be
regulated by catecholamines. Epinephrine, norepinephrine, and
phenylephrine increase GFR and urine formation, whereas both vari-
206                                                 KENNETH R. OLSON

ables are inhibited by isoproterenol and phenoxybenzamine (Yokota
and Benyajati, 1986).Subpressor doses of ANG I1 (18-36 ng/')
do not affect GFR, urine flow, or sodium excretion in elasmobranchs
(Churchill et al., 1985a).

   c. Drinking. Drinking is not stimulated by ANG I1 [Asn', Val5] in
the banded dogfish, Triakis scyllia, or bull-head shark, Heterodontus
japonicus (Kobayashi et al., 1983). Studies by Hazon et al. (1989)
indicate that some angiotensin-dependentdipsogenic effectors may be
present in elasmobranchs. ANG I1 [Asp', Ile5] increases drinking more
than 2-fold in S. canicula (Hazon et al., 1989).Captopril has no effect
on drinking in otherwise untreated S. canicula (Hazon et al., 1989).
 However, papaverine stimulates drinking in S. canicula by 30-fold,
and this dipsogenic response can then be 80% inhibited by captopril
(Hazon et al., 1989).These studies also suggest that a RAS may func-
tion in elasmobranchs under certain circumstances.

   d . Other Effects. Infusion of homologous renal extracts (0.3 mg/') or ANG I1 (36 pglhakg-') increases the secretion rate and
plasma levels of la-hydroxycorticosterone in Scyliorhius canicula but
does not affect plasma osmolarity or sodium (Hazon and Henderson,
1985). However, ANG I1 [Asp', Ile5] decreases plasma osmolarity and
sodium in this species (Hazon et al., 1989).
   Effects of angiotensins have not been reported in holocephalans.
   T h e effects of angiotensins in chondrosteans or holosteans have not
been reported.

    a. Dipnoians. ANG I1 [Val5] is pressor in both African and Austra-
lian lungfish, Protopterus aethiopicus and Neoceratodus forsteri, re-
spectively, but produces slight to negligible diuresis and natriuresis;
heart rate does not change (Blair-West et al., 1977; Sawyer, 1970;
Sawyer et al., 1976).Infusion of 0.2-4 pg/' ANG I1 [Val5] into N .
forsteri for 2-4 h has no effect on plasma aldosterone or deoxycorticos-
terone (Blair-West et al., 1977).

   b. Teleosteans.
   ( i ) Cardiovascular. Exogenous angiotensins produce pressor re-
sponses in all teleosts thus far examined including SW lumpfish, Cy-
clopterus lumpus (ANG 11; Carroll and Opdyke, 1982), 50% SW
flounder, Platichthyes Jesus (ANG I1 [Asp', Val']; Perrott and Bal-

ment, 1990), and both the aglomerular toadfish, Opsanus tau (ANG I1
[Asn', Val5]; Zucker and Nishimura, 1981), and goosefish, Lophis
americanus (Churchill et al., 1979). Angiotensins are also pressor in
FW eels, Anguilla rostrata (ANG I [Val5, Ser'], partially purified eel
ANG I, ANG I1 [Am', Val5]; Nishimura and Sawyer, 1976; Nishimura
et al., 1978), A . japonica (ANG I, 11; Hirano and Hasegawa, 1984)
intact, hypophysectomized, or stanniectomized A. anguilla (ANG I1
[Asp', Val5]; Henderson et al., 1976; ANG I1 [Val5]; Chester Jones et
al., 1966),and in rainbow trout, Oncorhynchus mykiss (ANG I1 [Asn',
Val5]; Gray and Brown, 1985; ANG I [Asn-l, Val', Asn'l, ANG I1
[Asn', Val']; H. Xu and K. H. Olson, unpublished). Angiotensin I11 is
pressor in A.japonica but it has only one-tenth of the activity of either
ANG I or I1 (Hirano and Hasegawa, 1984). Angiotensin converting
enzyme hydrolysis of decapeptide angiotensins to the octapeptide is
necessary for pressor activity (Hirano and Hasegawa, 1984; Madey et
al., 1984; Nishimura et al., 1978).
    The hypotensive effect of ACE inhibition in SW toadfish, Opsanus
tau (Madey et al., 1984; Nakamura and Nishimura, in: Nishimura and
Bailey, 1982), and FW rainbow trout (Galardy et al., 1984; Lipke and
Olson, 1990) indicates that the RAS is active in blood pressure regula-
tion in these fish even under resting conditions. In other species such
as SW or FW flounder, Platichthyes Jesus, ACE inhibition has no
effect on resting blood pressure (Balment and Carrick, 1985), but ACE
inhibition prevents the gradual recovery of pressure following papav-
erine injection into 50% SW flounder (Perrott and Balment, 1990).
Curiously, infusion of captopril into FW (40 pg/'- or SW
(160 pg/min-kg-')-adapted trout does not affect blood pressure (Ken-
yon et al., 1985)even though the total dose infused is greater than that
which produces hypotension in the same species after a single injec-
tion (1mg/kg, Galardy et al., 1984; Lipke and Olson, 1990).
     Unlike elasmobranchs, only a fraction of the pressor effects of
angiotensins are mediated by catecholamine release in teleosts even
though angiotensin may increase circulating catecholamines two- to
threefold (Carroll and Opdyke, 1982). a-Adrenergic blockade reduces
ANG I1 effects by 10% in lumpfish, Cyclopterus lumpus (Carroll and
Opdyke, 1982), and by 30-40% in A . rostrata (Nishimura et al., 1978),
although reserpine produces 70% inhibition in eels (Nishimura et al.,
1978). The in vitro vascular effects of angiotensins are consistent with
their in vivo actions. Ventral aortic strips from longhorn sculpin, Myo-
xocephalus octodecirnspinosus, are directly stimulated by ANG I1
[Asn', Val'] independent of a-adrenergic inhibition with phentol-
amine (Carroll, 1981). Topical application of ANG I1 [Asn', Val'] also
208                                                      KENNETH R. OLSON

contracts sculpin mesenteric circulation (Carroll, 1981). Although ma-
jor arteries in trout, such as ventral aorta and celiacomesenteric artery,
are not contracted by mammalian or salmonid angiotensins, these pep-
tides are vasoconstrictory in perfused systemic tissues (Olson and
Meisheri, 1989; K. R. Olson, Ferlic and T. Kne, unpublished) and in
perfused gills of both trout (K. R. Olson, unpublished) and eels (Fen-
wick and So, 1981). Classical angiotensin antagonists, such as ANG I1
 [Sar', Thr'l, [Sar', Ile'l, and [Sar', Val5, Ala'], are ineffective in fish in
vivo and in vitro, and, in fact, they often have slight agonistic proper-
ties (Carroll, 1981; Churchill et al., 1985b; Khosla et al., 1983; Nishi-
mura et al., 1978). Lack of suitable angiotensin antagonists has se-
verely limited further examination of the RAS in fish cardiovascular
     ( i i ) Renal. In glomerular fish the renal effects of exogenous
 angiotensins are often obscured by a concomitant systemic hyperten-
 sion and it is difficult to identify specific intrarenal RAS actions. In the
 eel, Anguilla rostrata, ANG I1 [Val5], and partially purified eel ANG I,
 increased GFR and urine formation and, at high doses (100 ng/
 minakg-'), decreased fractional sodium reabsorption (Nishimura aiid
 Sawyer, 1976). These responses were only seen after pressor doses of
 angiotensin, suggesting a primary extrarenal effect. In order to sta-
 bilize blood pressure, Brown et aZ. (1978, 1980) continuously infused
 pressor doses of norepinephrine into anesthetized trout. In these ex-
 periments Brown et al. (1978, 1980) found that FW trout have more
 filtering nephrons, but the single nephron filtration rate (SNGFR) is
 lower in FW than SW trout; the percent filtering, nonfiltering, aiid
 nonperfused nephrons in FW fish is 45,40, and 13%, respectively, and
 in SW fish it is 5,40, and 51%, respectively. After infusion of ANG I1
 (Asn', Val5] in FW fish the percent filtering and nonfiltering nephrons
 is 9 and 46%, respectively, and in SW trout it is 6 and 12%, respec-
 tively. The single nephron filtration rate is unaffected. Thus ANG I1 in
 FW fish appears to reduce GFR b y reducing the number of filtering
 nephrons to the nonperfused type. In SW trout, ANG I1 [Asn', Val5]
 reduces the SNGFR to FW levels without affecting the number of
 filtering nephrons while nonfiltering nephrons are converted to the
 nonperfused type (Brown et al., 1980). Because ANG I1 reduced GFR
 and urine formation by 50% in FW fish (GFR from 140 to 61 and urine
 formation from 76 to 30; all in pllmin-kg-'), ANG I1 appears to exert a
 direct intrarenal antidiuretic effect (Brown et al., 1980). Much higher
 infusion rates of ANG I1 (600 versus 150 ng/min-kg-') are needed in
 SW than FW fish for an equivalent reduction in urine formation, sug-
 gesting that endogenous angiotensins are already elevated in the

former (Brown et al., 1980). In subsequent experiments, ANG I1 [Asn',
Val'] effects were examined in anesthetized FW trout without concom-
itant norepinephrine infusion, and these experiments confirm the anti-
diuretic nature ofangiotensin (Gray and Brown, 1985). In these experi-
ments (Gray and Brown, 1985),ANG I1 [Asn', Val5] infusion increased
blood pressure and heart rate, caused an initial antidiuresis, reduced
GFR, and slightly decreased free water clearance (CH,O); all parame-
ters returned to control levels over 45 min of continuous ANG I1
infusion (CH,O slightly increased) and remained at control even after
cessation of ANG I1 infusion (Gray and Brown, 1985).Tubular maxi-
mum for glucose (TMG) was reduced by ANG I1 and did not fully
recover until after ANG infusion stopped; ANG had no effect on frac-
tional sodium excretion but during the recovery in urine flow, in the
latter stages of infusion, fractional osmolyte excretion increased (Gray
and Brown, 1985). The only partial recovery of TMG, despite complete
recovery of urine formation rate and GFR, was taken to indicate that
there are still fewer filtering nephrons in the later stages of ANG
infusion but that GFR in the remaining nephrons is now greater,
perhaps due to the hypertension. Angiotensin 11 [Asn', Val5] infusion
(45 min) into FW trout, or added to glomeruli in vitro, flattens po-
docytes and obliterates pedicels, consistent with an antidiuretic effect
in vivo (Gray and Brown, 1987).
    Angiotensin converting enzyme inhibition provides further evi-
dence of an endogenous renal RAS in fish. Captopril infusion at a
nonhypotensive rate into FW or SW trout increases urine flow rate,
GFR, and TMG, indicating an increase in the number of functioning
tubules (Kenyon et al., 1985).Captopril also increases urinary electro-
lyte concentration, fractional sodium excretion, solute concentration,
and free water excretion in FW fish (Kenyon et al., 1985). In SW fish,
captopril decreases urinary sodium concentration, but because urine
formation rate increases, sodium excretion rate does not change (Ken-
yon et al., 1985). In both FW and SW trout, urine formation rate and
GFR are linearly correlated; because this is not affected by captopril,
changes in urine production are not attributable to increase in tubular
water reabsorption (Kenyon et al., 1985). The ability of captopril to
increase GFR in SW trout to FW levels indicates that angiotensin may
be an important factor in regulating renal function in SW adaptation of
trout (Kenyon et al., 1985). In eels this effect is less clear. Captopril
injected into eels prior to abrupt seawater transfer greatly increases the
cortisol response (from a 2 x increase without captopril to 10x with  -
captopril), enhances the fall in plasma potassium, but does not affect
the increase in plasma sodium (Kenyon et al., 1985). Furthermore,
210                                                 KENNETH R. OLSON

captopril injected into either FW or SW adapted eels does not affect
plasma sodium or potassium or cortisol in SW eels and only transiently
decreases cortisol in FW fish (Kenyon et al., 1985).
    Aglomerular fish have been used to examine angiotensin responses
in the absence of GFR-mediated effects with mixed results. In goose-
fish, Lophius americanus, ANG I1 increases urine flow rate, urinary
"a+], and urinary Na+ and K+ excretion but has no effect on Mg2+ or
Ca2+excretion, and it has been proposed that pharmacological doses of
ANG I1 inhibit Na+ and K+ reabsorption in the distal segment
(Churchill e t al., 1979). However, in later experiments, saralasin ANG
I1 [Sar', Val5, Alas] appeared to decrease urine formation and excretion
of K+, Mg2+, and Ca2' indicating an inhibition of the effect of physio-
logical levels of ANG I1 on divalent ion secretion (Churchill et al.,
1985b). In view of the ineffectiveness of angiotensin antagonists on
cardiovascular responses (see previous discussion) and on 1251-ANGI1
[Ala', Val5] binding to glomeruli of FW trout (Brown et al., 1990b), it
remains to be determined if saralasin actually inhibited goosefish renal
angiotensin mechanisms or was itself agonistic. In contrast, ANG I1
[Am', Val5] has no effect on urine flow or electrolyte excretion in the
toadfish, Opsanus tau, even at pressor doses (Zucker and Nishimura,
    (iii) Drinking. Kobayashi et al. (1983) observed a dipsogenic re-
sponse to ANG I1 [Asnl, Val'] in 10 of 19 FW and 6 of 14 SW teleosts
and hypothesized that the ability of angiotensin to stimulate drinking
in FW fish is correlated with an estuarine habitat or an ability to
survive hypersaline water; in SW fish angiotensin mediated drinking is
commonly associated with fish inhabiting tide pools or brackish water.
Angiotensin I1 responses occurred irrespective of whether the fish
were euryhaline or stenohaline (Koyabashi et al., 1983).Kobayashi et
al. (1983) also proposed that angiotensins do not usually affect fish
inhabiting solely FW or SW. High doses (10-50 pg/kg) of ANG I1
 [Asn', Val'] frequently inhibited drinking (Kobayashi et al., 1983),
perhaps due to secondary hypertensive effects. Beasley et al. (1986)
 found that drinking is not stimulated by angiotensins in fish that in-
habit only FW, such as goldfish, Crassius auratus, common shiner,
Natropis cornutus, and mottled sculpin, Cottus bairdi, whereas in fish
that inhabit only SW (e.g., winter flounder, Pseudopleuronectes ameri-
 canus, and longhorn sculpin, Myoxocephalus octodecemspinosus),
ANG I1 increases drinking two- threefold. Additional species need to
be examined to further define these relationships. Interestingly, there
 are some apparent intraspecies differences; ANG I1 [Val5] stimulates

drinking in a hypersaline tolerant Japanese strain of Carassius auratus
(Kobayashi et al., 1983; Okawara and Kobayashi, 1988) but not in
hypersaline intolerant C . auratus raised in North America (Beasley et
al., 1986).
    Angiotensin effects on drinking are usually most pronounced
in euryhaline fish adapted to seawater. In euryhaline flounder,
Platichthyes Pesus, SW adaptation increases drinking, plasma osmo-
larity, and [Cl-1; ANG I1 [Asp', Val5] infusion, even at hypertensive
doses, is a greater stimulus for drinking in SW fish (Balment and
Carrick, 1985; Carrick and Balment, 1983). In FW flounders, papaver-
ine lowers blood pressure, plasma osmolarity, and [Cl-] yet increases
drinking (Balment and Carrick, 1985). Captopril alone has no affect on
drinking in FW fish but prevents the dipsogenic response to papaver-
ine and decreases drinking in SW fish (Balment and Carrick, 1985;
Carrick and Balment, 1983). Hypertonic saline (10%) infusion plus
papaverine and captopril further increases drinking (Balment and Car-
rick, 1985). These studies demonstrate the involvement of the RAS in
drinking and indicate that hypovolemia may be a potent stimulus for
RAS activation during cellular hydration. The fact that cellular dehy-
dration plus hypovolemia produce an augmented response suggests
that these effectors act through independent mechanisms.
    Volume or pressure dependent drinking, and multiple dipsogenic
foci have also been shown in Anguilla japonica (Hirano and Hase-
gawa, 1984; Hirano et al., 1978; Okawara et al., 1987; Takei et al.,
1988),and some of these may be independent of systemic angiotensin.
Volume expansion and concomitant cellular dehydration produced by
infusion of 7 or 14% saline or 65% sucrose increases plasma irANG 11,
yet decreases drinking in esophageal cannulated FW or 113 SW eels
(Takei et al., 1988). In FW or SW eels, Anguilla japonica, intraarterial
ANG I1 [Asn', Val5] stimulates drinking; however, intracerebro-
ventricular ANG I1 [Asn', Val5] is a more potent dipsogen than intraar-
terial injection (Takei et al., 1979). Removal of the telencephalon,
diencephalon, and part of the mesencephalon does not affect ANG
II-stimulated drinking, but drinking is inhibited by vagotomy, impli-
cating the medulla as a dipsogenic center (Hirano et al., 1978; Takei et
al., 1979). Angiotensin I11 does not have a dipsogenic function in eels
(Hirano and Hasegawa, 1984). Okawara et al. (1987) found a transient
plasma irANG II-independent increase in drinking in A . japonica
during the first 2 h after transfer to SW, whereas over the next 3 days a
further increase in drinking was accompanied by an increase in plasma
irANG 11. Captopril prevented the delayed increase in drinking at 3
212                                                    KENNETH R. OLSON

and 10 days after transfer but did not affect the early response
(Okawara et al., 1987), suggestive of multiple dipsogenic mechanisms.
    Angiotensin converting enzyme inhibitors consistently inhibit
ANG I-stimulated drinking in fish (Carrick and Balment, 1983; Hirano
and Hasegawa, 1984; Okawara and Kobayashi, 1988). Both ACE inhi-
bition and saralasin decrease drinking by SW killifish, Fundulus het-
eroclitus (Malvin et al., 1980), and flounder, Platichthysflesus (Carrick
and Balment, 1983), whereas neither affect drinking by control or
hemorrhaged SW winter flounder, Pseudopleuronectes americanus, or
by longhorn sculpin, M. octodecemspinosus (Beasley et al., 1986).
Okaware and Kobayashi (1988) found that in goldfish, C . auratus, low
doses of captopril (0.4 and 4.0 pg/fish, -0.1-1 mg/kg) stimulated
drinking, whereas higher doses were ineffective. The authors
(Okawara and Koyabashi, 1988) proposed that during low dose cap-
topril treatment ANG I levels increased, and then as captopril was
metabolized proportionally more ANG I1 was formed, thereby stimu-
lating drinking.
    ( i v ) Other effects. An ability of the RAS to stimulate corticosteroid
production in fish has received attention because of the well-known
stimulatory response in mammals and because of a somewhat temporal
correlation between the RAS and cortisol during salinity adaptation.
The results to date are equivocal. With in vitro preparations using
interrenal tissue or isolated cells, angiotensins may (Hanke, 1990) or
may not (Decourt and Lahlou, 1987; Hanke, 1990; Takahashi et al.,
1985; Vetter and Hanke, 1985) stimulate corticosteroid release.
However, ANG I, ANG I [Asn', Val5, Asn'l, and ANG I1 [Asp', Ile5]
stimulate ACTH release from dispersed goldfish anterior pituitary
cells independent of ovine corticotropin releasing factor or urotensin I
(Weld and Fryer, 1987, 1988). Angiotensin may also act indirectly on
trout interrenal tissue in vitro by interacting synergistically with other
secretagogues (Decourt and Lahlou, 1987). The in vivo response is also
species specific. Angiotensin I1 [Asp', Val5] injection increases plasma
cortisol in 50% SW flounder, Platichthysflesus,as does ANG I1 [Asp',
Val5] and renal extracts in intact or hypophysectomized eels, Anguilla
anguilla (Henderson et al., 1976). However, SW adaptation only tran-
siently (24-48 h) increases plasma cortisol in eels, whereas PRA in-
creases slowly and steadily over 3-5 days (Henderson et al., 1976) and
ACE inhibition does not diminish the cortisol response (Kenyon et al.,
1985). Furthermore, immunoreactive plasma aldosterone in large-
scale suckers, Catostomus macrocheilus, is not affected by injection of
homologous renin extracts or ANG I1 but does increase in response to
hemorrhage (Reinking, 1983).
3.   BLOOD AND EXTRACELLULAR FLUID VOLUME REGULATION                             213


A. Components of the Kallikrein-IGnin System
    The kallikrein-kinin system is the physiological enantiomer of the
RAS (Fig. 8). In its simplest form, the enzyme kallikrein cleaves a
biologically active peptide (kinin) from an inactive precursor (ki-
ninogen). Kinins are inactivated by a variety of peptidases, most nota-
bly ACE. (The latter was originally thought to be a separate enzyme,
kininase 11.) In mammals, the KKS is divided into plasma and glandu-
lar forms based on the location of kallikrein and kininogen (Nustad et
al., 1980; Schachter, 1980). T h e plasma KKS consists of a circulating
inactive “prekallikrein” and kininogen, a high molecular weight kinin-
transporting a-globulin. Upon activation of kallikrein by Hageman
factor, or other stimuli, the nonapeptide bradykinin is cleaved from
kininogen. Glandular KKS is found in the pancreas, salivary glands,
and kidney. Glandular kallikrein cleaves the pentapeptide lysyl-
bradykinin (kallidin) from a low molecular weight tissue kininogen,
although it can also form kinins from plasma kininogens (Rabito et al.,
1972; Claeson et al., 1978).
    The physiological significance of the KKS in mammals is unclear.
The KKS has been shown to have potent effects on renal salt and water

          RAS                                         KKS

             C i v e )
                                     --r-’l-...        Kininogens

                                                 Bradyldnin  Kallidin

                1                                             1
                 -4             converting
                                (Kininase II)

          Angiotensin II                               Metabolites
              (active)                                  (inactive)
    Fig. 8. Activation steps in the renin-angiotensin (RAS) and kallikrein-kinin (KKS)
systems. See text for details.
214                                                   KENNETH R. OLSON

balance in the kidney (Margolius, 1984; Scicli and Carretero, 1986).
Kinins are among the most potent vasodilators known, and many of
their functions are undoubtedly related to this vasorelaxant activity.
Kinins have potent effects on capillary permeability and are often
involved in inflammatory responses; in pathological situations, they
may produce considerable fluid loss from the vasculature (Margolius,
1989; Nustad et al., 1980; Sander and Huggens, 1972; Schacter, 1980).
A complete KKS has also been identified in the brain, and kinins
appear to be involved in the central control of the cardiovascular
system (Diz, 1985).

B. Occurrence and Distribution in Fish
   Evidence for a complete KKS in fish should include the presence of
(a) a kinin precursor; (b) kallikrein-like enzymatic activity; (c) kinin
receptors, which when stimulated mediate a physiological response;
and (d) a mechanism for inactivating kinins. Kinin-like substances
were identified in neural tissue from carp as early as 1961 (Inouye et
al., 1961). However, subsequent reports, indicating that the KKS is
absent in anamniotic vertebrates (Rabito et al., 1972; Seki et al., 1973),
did much to quell enthusiasm for further study. Surprisingly, the dem-
onstration of the ability of mammalian glandular kallikrein to produce
a kinin-like substance from rainbow trout plasma (Dunn and Perks,
1975) did little to promote further investigation.
    In an examination of the phylogenetic expression of ACE, Lipke
and Olson (1988) observed that this enzyme is found in virtually all
vertebrate classes and thereby may predate the supposed origin of the
RAS. They (Lipke and Olson, 1988) hypothesized that perhaps ACE
was originally important in other, non-RAS, proteolytic functions. Be-
cause the Michaelis-Menten constant (K,) for bradykinin is lower than
for ANG I (8 x        M versus 3.3 x lop5M , respectively; Ryan, 1983),
the KKS seemed to be a possible candidate. This formed the basis for
the reexamination of the KKS in the rainbow trout (see next section)
and, although these studies are far from definitive, they provide a
general framework for future investigations.
          FOR       IN

    In vivo and in vitro studies indicate that a biologically active prod-
uct can be generated from trout plasma by porcine glandular kallikrein
(GK; Dunn and Perks, 1975; Lipke and Olson, 1990; Lipke et al.,
1990a). Injection of GK into intact trout pretreated with the ACE
inhibitor captopril increases dorsal aortic blood pressure, whereas in

untreated trout, GK has no effect (Lipke and Olson, 1990). This indi-
cates that the pressor substance generated by GK is rapidly inactivated
by endogenous ACE, consistent with the kininolytic function of the
latter. Incubation of trout plasma for 1 h with GK likewise generates a
product that is pressor in intact trout (Lipke et al., 1980a). This product,
termed T60K (trout plasma incubated 60 min with kallikrein) is a heat
stable, low molecular weight (<10,000 kDa), peptide (Lipke et al.,
1990a). T60K is either very potent or trout plasma contains substantial
quantities of precursor, because as little as 10-50 p l of activated
plasma has pressor activity when injected into 300-g trout or rats
(Lipke et al., 1990a).
    It is not clear whether T60K is a kinin- or angiotensin-like molecule
as it has properties of both peptides yet it is not identical to either.
T60K is similar to angiotensin in that it is pressor in both trout and rats
and that the response in trout is sensitive a-adrenergic blockade,
whereas, in rats, the effect is blocked by the angiotensin receptor
agonist saralasin [Sar', Alas] ANG 11. T60K also competes with mam-
malian ANG I1 in homogenized rat adrenal tissue (Lipke e t d.,       1990a).
T60K is dissimilar to angiotensin in that (a) it appears to be inactivated
by ACE in uiuo, (b)it contracts isolated trout arteries that are refractory
to angiotensions, (c)the time course for contraction of rabbit arteries is
much longer for T60K than ANG 11, and (d) T60K is not detected in a
radioimmunoassay that is 70 and 100% cross reactive to trout ANG I
and ANG 11, respectively. T60K is similar to bradykinin in that both are
pressor in trout and both are inactivated by the arterioarterial pathway
of the isolated, perfused gill (Lipke et al., 1990a,b). Dissimilarities
between T60K and bradykinin include the pressor effect of T60K in
rats and sensitivity of the T60K, but not bradykinin, pressor response to
a-adrenoceptor blockade in trout.
    The limited information available suggests that kallikrein is
present in trout tissues, although definitive studies have yet to be
performed. Results from two spectrophotometric assays, based on the
hydrolytic activity of kallikrein on synthetic substrate, indicate that
kallikrein-like activity is present in trout gill and kidney, but not in
liver, muscle, or plasma (Lipke and Olson, 1990). However, these
results could not be confirmed with a third method that employed a
different substrate (Lipke and Olson, 1990). Further examination of
gill and kidney tissue, in the presence of a specific mammalian kalli-
krein inhibitor, Phe-Phe-Arg-chloromethyl    ketone (PPAMCK), showed
that a fraction of the gill estrolytic activity was indeed due to a
216                                                  KENNETH R. OLSON

kallikrein-like enzyme (Lipke and Olson, 1992). PPAMCK did not affect
kidney estrolytic activity (Lipke and Olson, 1992). Other kallikrein-
like enzymes, undetectable by these methods, may also be present in
fish as, even in mammals, there are a large number of genes that
encode kallikrein (MacDonald et al., 1988).

    Angiotensin converting enzyme, present in many fish tissues, espe-
cially gills, may be important in kinin inactivation, although other
peptidase are undoubtedly involved as well. In the mammalian lung,
ACE is located in the lumenal surface of the pulmonary endothelium,
and there it efficiently (-80%) inactivates kinins and effectively
prevents these vasodilators from entering the systemic circulation.
Inactivation of bradykinin by the gill is less efficient (-20-40%) and
appears to involve incorporation of the peptide into the endothelial/
pillar cell prior to metabolism (Lipke et al., 1990b). Because kinins are
not hypotensive in fish (see later), there is apparently no immediate
need for their rapid inactivation. The comparatively low rate of kinin
inactivation by the gill may indicate that circulating kinins have sys-
temic functions in these vertebrates. Interestingly, as noted above,
gills and lungs are similar in their ability to activate angiotensins
(Olson et al., 1986).

C. Effects of Kinins
    Unlike the potent depressor action of kinins in mammals, bradyki-
nin and other kinin analogs are pressors in rainbow trout (Lipke et al.,
1990b),and bradykinin is a pressor in the eel, Anguilla japonica (Chan
and Chow, 1976). In trout, bradykinin also fails to relax endothelium-
intact arterial and venous rings or to lower resistance in perfused
organs (Olson and Villa, 1991). Similar results are obtained with acetyl-
choline. These results are taken to indicate that trout, and perhaps
other fish as well, lack nonprostanoid endothelium derived relaxing
factors (EDRF; Olson and Villa, 1991). The pressor effects of bradyki-
nin in vivo are unaffected by a-adrenoceptor blockade and are there-
fore unlike angiotensin in this regard (Lipke et al., 199Ob).
    Identification of the physiological function(s) of the KKS awaits
further characterization of this system in fish tissues and the develop-
ment of suitable antagonists to block endogenous KKS activity. The

apparently high level of circulating kallikrein substrate in trout plasma
(Lipke and Olson, 1990; Lipke et al., 1990a) suggests a systemic func-
tion of the KKS (see previous discussion). Yet injection of mammalian
kallikrein inhibitor, PPAMCK, has no overt effects on trout blood
pressure in vivo (Lipke and Olson, 1992) and a circulating kallikrein
has not been detected, both of which argue against the KKS as a tonic
pressor system. Certainly, local regulation (or remote regulation
through a central nervous system KKS) is possible and the effects ofthe
KKS on salt and water balance, which have yet to be measured, could
exert a powerful, but less immediately obvious, role in cardiovascular


    In the early 1 9 8 0 d e~Bold and co-workers demonstrated that pep-
tides with diverse biological activity were synthesized and released by
mammalian cardiac tissues (de Bold et al., 1981; d e Bold, 1982). Since
then the genetic code and biochemical processing from the 151 or 152
amino acid (human or rat) preprohormone through 126-128 residue
prohormone to an active 28-residue peptide has been well character-
ized, as have a number of other important peptide variants (see re-
views: Brenner et al., 1990; Needleman et al., 1989; summaries edited
by Needleman, 1988; Samson and Quirion, 1990). Three families of
naturetic peptides (NP), A-, B-, and C-type (ANP, BNP, and CNP,
respectively), have been identified so far, and more are possible.
These peptides were originally named after the tissues from which
they were first isolated (e.g., atrium [ANP] and brain [BNP]). Most
peptides have since been found in multiple sites and, in fact, may be
more prevalent in tissues other than those in which they were first
    In spite of the intense activity focused on NP, and the variety of
pharmacological responses examined, there is as yet little consensus
regarding their physiological role (cf. Blaine, 1990; Goetz, 1990). In
fish the situation is exacerbated by the great physiological diversity of
these vertebrates, a limited appreciation of their homeostatic attri-
butes, and, until recently, the lack of native peptides with which to
study them. In spite of these limitations a growing amount of evidence
indicates that these peptides are ubiquitous among fish, and they have
somewhat stereotypic effects. Here too, however, the challenge re-
mains to delineate their physiological function.
218                                                               KENNETH R . OLSON

A. Structure of Natriuretic Peptides
    Figure 9 compares the amino acid sequence of rat ANP and fish
NPs. By convention, the lowercase letter preceding the peptide family
indicates the animal from which the peptide was isolated: h, human; p,
porcine; r, rat; e, eel, Anguilla japonica; k, killifish, Fundulus hetero-
clitus; and s, shark, Scyliorhinus canicula. The obvious confusion
generated by this practice will, undoubtedly, limit its value in the
future as more species are examined. Sequence homology of eANP to
rat, fowl, and frog ANP is 59, 52, and 46%, respectively (Takei, et al.,
1989); eCNP is 95 and 86% homologous to kCNP and pCNP, respec-
tively (Takei, 1990a). sCNP is 82% homologous to rCNP (Suzuki et al.,
1991). Residues between the cysteine disulfide bridge of all peptide
families are highly conserved, this is especially true for fish.
    Three NP receptors have been identified in mammals. The ANPR-
A and ANPR-B (also known collectively as B-ANP receptors, B denot-
ing "biologically active") is a 130-kDa transmembrane protein. ANPR-
A receptors are stimulated by ANP and to a lesser extent by BNP; CNP
is the most efficacious agonist of ANPR-B receptors (Koller et al.,
1991). The intracellular segments of both ANPR-A and ANPR-B recep-
tors contains a guanylate cyclase tail through which physiological ac-
tivity is conferred (Brenner et al., 1990; Needleman et al., 1989). The
ANPR-C receptor (also known as the C-ANP receptor; C denoting
"clearance," but confusing relative to the active CNP peptide) is a
-60-kDa protein that lacks the guanylate cyclase tail and has been
proposed to clear NPs from the circulation and thus provide a non-

     Fig. 9. Amino acid sequence for atrial natriuretic peptide family. Diagram shows
full sequence for rat ANP plus substitutions for shark CNP ((sCNP)),killifish CNP
[kCNP], eel CNP {eCNP}, and eel ANP (eANP). * denotes amino acid deletion; + de-
notes amino acid insertion. References: a, Suzuki et al. (1991);b, Price et al. (1990); C,
Takei et al. (1990a);d, Takei et al. (1989); e, Brenner et al. (1990).

proteolytic pathway for NP inactivation (Maack et al., 1987). The ring
heptapeptide residues 10-16 in Fig. 9 are essential for ANPR-C recep-
tor binding. ANPR-C receptors are otherwise relatively insensitive to
residue deletion or substitution (Bovy, 1990). On the other hand,
ANPR-A and ANPR-B receptors are quite sensitive to the structural
integrity of the peptide; minor substitutions variously affect vasodi-
lator or natriuretic responses in different tissues (Bovy, 1990). Substi-
tutions in the eANP undoubtedly account for the 110-fold increase in
potency of this peptide compared to mammalian peptides when in-
jected into eels (Takei et al., 1989) and the decrease in potency of
eANP in trout (Olson and Duff, 1992),toadfish (Evans, 1991),and quail
or rat tissues (Takei et al., 1989,199Oa,b).sCNP is a considerably more
potent agonist of rectal gland secretion than ANP-related peptides or
nonshark CNPs, and the amino acid at position 4 on sCNP appears
functionally important (Solomon et al., 1992). A concomitant interspe-
cific selectivity of NP receptors is also implied by these studies.
    Natriuretic peptides can be removed from the circulation by ANPR-
C receptors prior to proteolysis or by receptor independent proteolysis
through vascular and tissue peptidases. One of the most significant
enzymes in this regard is the neutral endopeptidase E C 24.11 ( E C; Erdos and Skidgel, 1989). One, or both, ofthese inactivation
mechanisms appear operative in trout. Simultaneous injection of a
competitive ANPR-C receptor inhibitor (SC-46542) and an inhibitor of
neutral endopeptidase (phosphoramidon) into trout produces diuresis,
saliuresis, and decreases pulse pressure, consistent with elevation of
endogenous NP (Duff and Olson, 1992). 1251-ratANP(lZ5I-rANP)ex-
traction by the perfused gill is also reduced by ANPR-C receptor
inhibition (Olson and Duff, submitted). These results indicate that the
mechanisms of NP inactivation in fish are qualitatively similar to those
found in mammals. The observation that eANP effects last longer than
hANP in eels, quail, and rats (Takei et al., 1989) may indicate quantita-
tive differences in peptide inactivation mechanisms.

B. Distribution in Fish
    Atrial and often ventricular cardiocytes of virtually all fish exam-
ined contain -0.2- to 0.4-pm-diameter dense core granules, somewhat
smaller in size and fewer in number, but otherwise similar to ANP
secretory granules found in mammalian cardiocytes (Chapeau et al.,
1985; Hirohama et al., 1988;Reinecke et al., 1985,1987a,b; Solomon et
al., 1985a; Uemura et al., 1990; Westenfelder et al., 198813).Additional
evidence for ANP distribution in fish has been obtained largely
220                                                   KENNETH R. OLSON

through immunohistochemical and radioimmunoassay techniques
using antibodies to mammalian ANP, and through bioassay of fish
heart extracts. To date, natriuretic peptides have been isolated from
only one elasmobranch and two teleosts (Fig. 9; Price et al., 1990;
Suzuki et al., 1991;Takei et al., 1989,1990a). A summary ofANP levels
in fish tissues and plasma is provided in Table XII.
    Immunoreactive ANP (ir-ANP) granules are distributed throughout
atrial tissue of Myxine glutinosa (Reinecke et al., 1987a) and atrial
subepicardium of Eptatretus burgeri (Uemura et al., 1990). Immuno-
reactive ANP is rare in M . glutinosa and absent in E. burgeri ventricles
but very prevalent in the portal vein heart of M . glutinosa (Reinecke et
al., 1987a). Immunoreactive ANP has also been demonstrated in vari-
ous regions of M . glutinosa brain, including primordium hippocampi,
pars ventralis thalami, hypothalamus, medulla, and spinal cord (mainly
somato motor tracts; Reinecke et al., 1987a). lZ5I-rANP binding has
been identified in myxine glomerular arterioles, capsular epithelia,
neck cells, and smooth muscle of the archinephric duct as well as aortic
endothelium and smooth muscle (Kloas et al., 1988). Myxine heart
extracts also relax mammalian arteries (Reinecke et al., 1987b). Al-
though granules are present in lamprey, Lampetra japonica, hearts,
they do not bind antibody directed against hANP (Uemura et al., 1990).
     Immunoreactive ANP has been demonstrated in hearts of Chimera
monstrosa, Squalus acanthias, Scyliorhinus canicula, Raja clavala,
and Triakis scyllia (Hirohama et al., 1988; Reinecke et al., 1987b;
Uemura et al., 1990)but not in the ray, Narkejaponica (Uemura et al.,
1990). Plasma concentrations of ir-ANP in chondrichthyes are equiva-
lent to those found in osteichthyes (Table XII). A high-molecular-
weight (1 residue) C-type (presumably prohormone) has been iso-
lated from the heart of the shark, Scyliorhinus canicula (Suzuki et al.,
1991). The putatively active sCNP fragment is shown in Fig. 9. This
study (Suzuki et al., 1991) is noteworthy in two regards; it demon-
strates that the peptide ring is highly conserved phylogenetically, and
it is also the first instance in which a CNP has been isolated from tissue
other than brain.
    Immunohistochemical and immunoassay techniques have demon-
strated ir-ANP in hearts of nearly all teleosts thus far examined (Galli et

al., 1988; Chapeau et al., 1985; Hirohama et al., 1988; Kim et al., 1989;
Reinecke et al., 1985, 198713; Uemura et al., 1990); only rarely is
ir-ANP absent (Chapeau et al., 1985; Hirohama et al., 1988).Ventricu-
lar concentrations of ANP are usually lower than those in the atria,
although in some species such as Zauo platypus and Pelteobagrus
fulvidraco ventricular ir-ANP concentrations are similar (Kim et al.,
 1989), and in Oncorhynchus mykiss and Opsanus beta ventricular
extracts have equal or greater potency in bioassay experiments (Duff
and Olson, 1986; Evans et al., 1989). Considering the relative mass of
the two cardiac chambers, it is probable that the ventricle contributes
significantly to circulating ANP in some fish. Similar ir-ANP secretory
rates by primary cultures of Gila atraria atria and ventricles, 3.9 and
2.8 ng/105 cells, respectively (Baranowski and Westenfelder, 1989),
support this contention.
    Natriuretic peptides have been purified from eel, Anguilla japon-
ica, heart (eANP; Takei et al., 1989) and brain (eBNP, Takei et al.,
 1990a) and killifish, Fundulus heteroclitus, brain (kBNP; Price et al.,
 1990).Truncation of both eBNP and kBNP at the carboxy terminus of
the ring (Fig. 9) indicates that these peptides are C- rather than B-type,
and eCNP and kCNP notation will be followed in the present dis-

C. Physiological Significance of
   Natriuretic Peptides
    Inferences regarding the physiological significance of NP in fish are
based on two types of experiments: (a) those in which circulating and
tissue NP concentrations (more appropriately ANP because most anti-
bodies used to date are directed against this peptide) are correlated
with physiological perturbation, and (b)those that examine the physio-
logical responses to NP administration. Many of these maneuvers di-
rectly or indirectly affect ion and volume regulatory processes and the
associative link between NPs and volume regulation in fish is probably
more than coincidental.
    NP (ANP)
    a. Salinity. Depending on species, environmental salinity may
have either no effect on, or it may be directly correlated with, ANP
synthesis and secretion. In an elasmobranch and several teleosts there
is no apparent correlation between salinity and ANP. Abrupt transfer
or 24-h adaptation of the dogfish shark, Squalus acanthias, to 70%
                                                           Table XI1
                             Distribution of ANP-like Immiinoactivity (IA) in Fish Tissue and Plasma

              Species        Environment             Tissue                 IA            Antibody             Reference

      Myxine glutinosa       sw                    Plasma*            187                hANP          Evans et al. (1989)
      Eptatretus burgeri     sw                    Heart a/v          0.3610.01          hANP          Uemura e t nl. (1990)
                                                   Plasma*             38.7
      Narke juponicu         SW                    Heart alv          0.3310.01          hANP          Uemura et al. (1990)
                                                   Plasma*             21.5
       Triakis scyllia       SW                    Heart alv          0.0210.01          hANP
                                                   Plasma*              2.7
      Dosyatis sabina        SW                    Plasma*            208                hANP          Evans et al. (1989)
      Squalus acanthias      SW                    P 1asm a *         129                hANP
at    Squalus acanthias      SW                    Plasma*             60                hANP          Epstein et al. (1988)
      Anguilla rostrata      SW                    Plasma*            154                hANP          Epstein et al. (1989)
                             FW                                       127
      Anguilla japonicu      FW                    Heart a h          0.081ND            AP 111        Kim et al. (1989)
      Anguilla japonica      FW                    Heart a1v          5.210.04           hANP          Uemura et al. (1990)
                                                   Plasma              45.7              hANP
       Cyprinus carpio       FW                    Heart aiv          44.612.6           hANP
                                                   Plasma              90.5              hANP
      Gila atruria           FW anesth             Plasma             197                hANP          Westenfelder e t al.
                             Awake                                    218
                             Pithed                                   207
                             FW                    Plasma             146
                             1%NaCI                                   347
       Oncorhynchus mykiss   FW                    Plasma             184                hANP          Westenfelder et ul.
      Oncorhynchus mykiss              FW                    Heart a/v          0.3510.01           hANP               Uemura et   (11.   (1990)
                                                             Plasma*               8.9              hANP
      Levomis macrochirus              FW                    Heart a/v          0.23/0.05           hANP
                                                             Plasma*                31.0            hANP

      Channa maculata                  FW                    Heart a h          0.29/0.03           hANP
                                                             Plasma*             70.3               hANP
      Pelteobagrus fuluidraco          FW                    Heart a h          0.1610.21           AP 111             Kim et al. (1989)
      Zauo platypus                    FW                    Heart a/v          0.2110.47           AP I11
      Mugil cephalus                   sw                    Plasma              33                 1-28 ANP           Galli et al. (1988)
                                       50% SW, 7d            Plasma              16
                                       FW,7d                 Plasma                5
      M yoxocephalus                   SW                    P 1asm a *         102                 hANP               Evans et al. (1989)
      Pseudo pleuronectes              SW                    Plasma*             30                 hANP
1     Opsanus beta                     sw                    Plasma              61                                    Epstein et al. (1989)
>                                                            Brain               47
                                                             Heart               49
                                       5% sw                 Plasma              23
                                                             Brain               20
                                                             Heart               29
      Conger myriaster                 SW                    Heart a/v          47.3/7.03           hANP               Uemura et al. (1990)
                                                             P 1asm a *          69.1               hANP
      Oplegnathus fasciatus            SW                    Heart a/v          0.53/0.02           hANP
                                                             Plasma*             45.5               hANP
      Pagrus major                     SW                    Heart a/v          0.05/0.01           hANP
                                                             Plasma*             11.3               hANP
      Trachurusjaponicus               SW                    Heart aiv          0.04/0.01           hANP
                                                             Plasma*             11.1               hANP
      Hexagrammos otakii               SW                    Heart a/v          0.08/0.01           hANP
                                                             Plasma*             13.2               hANP

       Abbreviations: a, atrium; v, ventricle; APIII, atriopeptin 1 1 (1-26); hANP, human (1-28ANP); *, extracted plasma. Plasma concentrations in
    pg/ml; tissue concentrations in ng/g.
224                                                 KENNETH R. OLSON

seawater does not change plasma ir-ANP but enhances the renal re-
sponse to exogenous ANP (Solomon et al., 1988). Long-term adaptation
or rapid transfer of eels, Anguilla rostrata, or rainbow trout between
FW and SW does not affect plasma ir-ANP (Duff and Olson, 1990a;
Epstein et al., 1989; Y. Takei, unpublished observation). Similarly,
Uemura et al. (1990) did not find a significant difference in plasma
ir-ANP when they examined five species of FW and five species of SW
teleosts. However, it should be noted that caution must b e exercised
when interpreting “no response” results until these studies are con-
firmed with assays using antibodies directed against homologous ANP.
This is underscored by the findings of Epstein et al. (1988) in which
only one of four commercial radioimmunoassay kits (two anti-rANP
and two anti-hANP) indicated a change in plasma ANP in volume
expanded sharks.
    In other teleosts there is a direct correlation between salinity and
ANP secretion (Table XII). Evans et al. (1989) observed that 1 week
after transfer from seawater to dilute media (-200 mOsm), circulating
ir-ANP is reduced by 90% in longhorn sculpin, Myoxocephalus octode-
cimspinosus, and winter flounder, Pseudopleuronectes americanus.
Plasma ir-ANP is also lower in toadfish, Opsanus beta, mullet, Mugil
cephalus, and catfish adapted to dilute media; mullet heart and brain
ir-ANP concentrations decline as well (Galli et al., 1988). In Gila
atraria, a fish with somewhat limited osmoregulatory capability, an
increase in medium osmolarity from 106 to 491 mOsmol/kg increases
plasma ir-ANP from 213 to 691 pg/ml and increases plasma osmolarity
from 272 to 486 mOsmol/kg; plasma sodium is highly correlated with
ANP (Westenfelder et al., 1988b). Uemura et al. (1990) observed that
cytochemical immunoreactivity in cardiocytes is greater in FW than
SW teleosts and claimed (as an unpublished observation) that FW to
SW transfer stimulates ANP synthesis in eel, Anguilla japonica.

    b. Volume Expansion. The effect of volume expansion on ANP
secretion by fish has received scant attention despite the fact that an
increase in central blood volume is thought to be the primary stimulus
for ANP secretion in mammals. In the dogfish shark, S . acanthias,
volume expansion with shark Ringer increases rectal gland volume
excretion and nearly doubles plasma ir-ANP (Epstein et al., 1988).
Because the shark rectal gland is involved in regulating intravascular
volume rather than osmoregulation (Solomon et al., 1985b), and vol-
ume expansion (but not salinity transfer) is a stimulus for ANP secre-
tion, a strong case can be made for ANP as a volume regulating hor-
mone in elasmobranchs.

    The effects of volume expansion on ANP secretion in teleosts are
not clear. In a preliminary report Duff and Olson (1990a) found that
trout plasma ir-ANP is unaffected by volume expansion with saline,
albuminated saline, or steelhead plasma. Whether trout are unrespon-
sive to volume expansion or whether the antibody used (hANP) is
insensitive to trout NPs is not known. However, Westenfelder et al.
(1988a,b) reported that trout ANP is immunologically similar to hANP.
Thus, the study by Duff and Olson (1990a) suggests that trout lack
ANP-mediated volume regulatory responses. Additional work with
trout and other teleosts is clearly needed to resolve this issue.
    Volume, electrolyte load, or both could stimulate salinity-mediated
ANP responses. There is clearly a correlative relationship between
plasma ir-ANP and plasma electrolytes, suggestive of ANP’s role in
osmoregulation (summarized by Evans, 1990). Whether the ANP re-
sponse is directed toward ionic or volume imbalance is unknown.
Evans (1990) argued against volume stimulation as a factor because of
the hypovolemic condition of marine teleosts. However, as previously
indicated in this chapter, SW teleosts may not be hypovolemic relative
to their FW counterparts. An even better case can be made for ANP as a
volume regulating hormone in hagfish and elasmobranchs as both fish
are slightly hyperosmotic to their environment and they may continu-
ally have to deal with volume expansion. If this is the case then the
lineage of ANP and volume regulation is indeed ancient and persis-
tent, perhaps explaining the conservatism between structure and func-
tion of this peptide in vertebrates.

    a . Fluid Compartments. Continuous infusion of trout with
300 ng/kg min-’ rANP [Ile26](versus an equal rate of saline infusion)
for 8 h lowers blood volume (51Cr red cell method) from 29.8 2 1.9 to
22.9 0.97 ml/kg and ECFV (“Co-EDTA method) from 244 t 11 to
156 2 10 ml/kg (D. W. Duff and K. R. Olson, unpublished). The
reduction in extracellular fluid volume (36%) is greater than the reduc-
tion in blood volume (23%), indicative of reabsorption of interstitial
fluid to maintain plasma volume. There have not been any other re-
ports of ANP on fluid compartments in fish.

   b. Cardiovascular Effects in Vivo. Hypotension is the most com-
mon response to ANP injection into intact fish. Atriopeptin I1 (AP 11)is
depressor in Squalus acanthias (Benyajati and Yokota, 1990; Solomon
226                                                     KENNETH R. OLSON

et al., 1988).In eels, Anguilla japonica, a-rANP, a-hANP, a-hANP(5,25),
eANP and BNP, and eel heart extracts all lower arterial pressure (Takei
et al., 1989, 1990a,b). Similar responses are found in cod, Gadus mor-
hua, to a-rANP (Acierno et al., 1991) and toadfish, Opsanus tau, to
homologous heart extracts (Lee and Malvin, 1987). hANP (10 pg/kg)
produces only a modest drop in blood pressure (1-2 mmHg) in the
flounder, Pleuronectes platessa, although the hypotension persists for
several hours (Arnold-Reed et al., 1991).
    Atriopeptin 111(AP 111)may lack cardiovascular effects in fish even
though it is less truncated than the depressor AP I1 (AP 11 lacks the
C-terminal tyrosine of AP 111).In both shark (Solomon et al., 1988)and
toadfish (Lee and Malvin, 1987),AP 111has no effect on blood pressure
even though substantial effects on renal function are observed in both
species. It is not clear if this insensitivity to AP I11 is a general feature
of piscine cardiovascular systems. Human cardiodilantin 1-16 (the
1-16 N-terminal residues of pro-ANP) is also without effect in cod
(Acierno et al., 1991).
    Intraarterial injection of ANP into trout may produce a biphasic
pressor-depressor response (Duff and Olson, 1986), no response
(Eddy et al., 1990), or a depressor response (Olson and Duff, 1992).
The variability of this response is probably attributable to the rate of
ANP administration. Injection of 10 pglkg of either rANP[IleZ6] or
eANP as a single bolus produces a pressor-depressor response (Duff
and Olson, 1986,1990b). If injected over 10 min, 10 pg/kg a-hANP has
no effect on blood pressure in either control trout or trout fed a high salt
diet (Eddy et al., 1990). Constant infusion of rANP[Ilez61 (18 pg/
hekg-') produces a steady decline in blood pressure (Olson and Duff,
1992).The reason for this variability is not known, although it probably
is related to the different plasma concentrations of the peptide follow-
ing injection and the resultant titers of ANP at different receptor/
effector sites. The implications from these studies are that there are
multiple foci for ANP action in fish and that perhaps some of these are
dependent on extravascular sources of the peptide for their activation.
    The ability of a-adrenergic antagonists to inhibit the rANP[Ile26]-
mediated pressor response in trout (Duff and Olson, 1990b; Olson and
Duff, 1992) suggests that the sympathetic nervous system is involved.
Both the nonspecific a-receptor antagonist, phenoxybenzamine, and
the specific al-receptor antagonist, prazosine, prevent the pressor re-
sponse; in fact, the former unmasks an ANP-mediated drop in arterial
pressure (Duff and Olson, 1990b; Olson and Duff, 1992). Both antago-
nists prevent ANP diuresis and saliuresis.
3.   BLOOD AND EXTRACELLULAR FLUID VOLUME REGULATION                                 227

    c. Cardiovascular Effects in Vitro:Perfused Tissues. Studies of the
effects of ANP on perfused tissues are limited to three species. ANP
vasodilates tissues precontracted with either carbachol, such as the
isolated head of Opsanus beta (Evans et al., 1989), or with epineph-
rine, such as gill, muscle-kidney, and splanchnic bed of trout, 0. my-
kiss (Olson and Meisheri, 1989). Tissues not precontracted, such as
trout muscle-kidney and splanchnic (Olson and Meisheri, 1989) and
Squalus acanthias rectal gland (Solomon et al., 1985a), do not respond
to ANP. ANP-mediated vasodilation of the trout gill occurs even if the
gill vessels are not precontracted (Olson and Meisheri, 1989) and
favors perfusion of the arterioarteriolar (lamellar) pathway. The effec-
tive ANP concentration producing half maximal response (ECJO) the in
perfused toadfish head is 3 x lO-'M (Evans et al., 1989). ECsos have
not been reported for other tissues. This ECso is similar to those
reported for isolated vessels which indicates that the microcirculation
of the head is neither more or less sensitive to ANP than are large

   d . Cardiovascular Effects in Vitro: Isolated Vessels. Atrial pep-
tides relax all isolated fish vessels examined to date including ventral
aortas from agnatha, elasmobranchs, and teleosts (Fig. 10).Relaxation

                              I       I

    Fig. 1 . Dose response curves for ANP-mediated relaxation ofhagfish, M . glutinosa,
ventral aorta (B; Evans, 1991), shark, S. acanthias, ventral aorta ( 0 ; Evans, 1991), and
trout, 0. mykiss, celiacomesenteric artery (X;Olson and Meisheri, 1989). Approximate
concentrations for half-maximal response (EC,) are hagfish, 4 x lo-”; shark, 7 x lo-’;
and trout, 1.5 x lo-’.
228                                                    KENNETH R. OLSON

can be demonstrated in unstimulated vessels (Evans et al., 1989;Olson
and Meisheri, 1989)and in vessels stimulated with a variety of agonists
such as carbachol (Solomon et al., 1985a),acetylcholine, arginine vaso-
pressin, epinephrine, norepinephrine, serotonin, and a thromboxane
A2 agonist (Olson and Meisheri, 1989). With relatively few exceptions
the ECsos for relaxation by mammalian ANPs are in the 1-10 nanomo-
lar range; these concentrations are two orders of magnitude greater
than plasma ir-ANP levels. However, the potency of homologous ANP
may be considerably greater, as it is in the eel in vivo (Takei et al.,
1989, 1990a) and perfused shark rectal gland (Solomon et al., 1992),
and more in line with circulating titers.
    Nearly all isolated vessel studies have used rings from ventral
aortas and in some species (e.g., dogfish shark and trout) these vessels
are refractory to common agonists, such as catecholamines (Olson and
Villa, 1991; Solomon et al., 1985a). Trout celiacomesenteric and epi-
branchial arteries demonstrate sensitivity to other agonists including
catecholamines and these vessels are relaxed by ANP independent of
agonist type (Olson and Meisheri, 1989).Additional studies on other
postgill vessels and vessels from a variety of species are needed to
better interpret the systemic effects of these peptides.
    ANP also relaxes trout anterior cardinal veins (Olson et al., 1991).
Thus, some of the depressor response to ANP in vivo could be medi-
ated through changes in vascular filling pressure and venous return.
    Evans et al. (1989) found that ventral aorta rings from toadfish
adapted to 5% seawater are more sensitive to rat (101-126) ANP than
rings from seawater adapted toadfish (EC50 of 3 x lo-'' and 4 X lo-'
M , respectively). Evans et al. (1989) proposed that this is likely due to
an up-regulation of ANP receptors secondary to the fall in circulating
ANP titers. The physiological significance of this response is unclear
as ANP sensitivity of the perfused toadfish head does not change with
salinity adaptation (Evans et al., 1989).

    e. Cardiac Effects. In vivo, rANP increases heart rate in cod,
Gadus morhua, and trout (Acierno et al., 1991; Olson and Duff, 1992),
whereas neither heart extract nor AP I11 affects heart rate in Opsanus
tau (Lee and Malvin, 1987). hANP has no effect on blood pressure but
profoundly reduces pulse pressure when slowly (10 min) injected into
normal trout but not fish fed a high salt diet (Eddy et al., 1990),an effect
attributed to an ANP-mediated decrease in gill vascular resistance.
Continuous infusion of rANP [IleZ6]into trout also reduces pulse pres-
sure; however, this appears to be associated with an increase in heart
rate; both effects precede, and appear to be temporally dissociated

from, the depressor response (Olson and Duff, 1992). ANP-induced
tachycardia is partially inhibited by propranolol in trout (Olson and
Duff, 1992) and by atropine in cod (Acierno et al., 1991) suggesting a
reflexogenic response. Neither eANP nor rANP affect heart rate or
cardiac power output in the denervated, spontaneously rhythmic, in
situ perfused trout heart (J. Keen, H. Thoranensen, A. P. Farrell, and
K. R. Olson, unpublished observation). Thus the changes in pulse
pressure and heart rate observed in fish in vivo probably result from
extracardiac events.

    f . Renal Effects. The effects of ANP on renal function in cyclo-
stomes have not been examined. In the elasmobranch, Squalus
acanthias, AP I1 and A P I11 reduce GFR, urine flow, and sodium,
chloride, and osmolar excretion b y 40-50% (Benyajati and Yokota,
1990; Solomon et al., 1988).Potassium excretion is reduced by AP I1
but not AP I11 (Benyajati and Yokota, 1990). The renal response ap-
pears to b e mediated primarily through changes in GFR because frac-
tional electrolyte excretion is unaffected. Renal responses are also
independent of systemic cardiovascular effects; inhibition of renal
function temporally lags behind, and persists several hours after, the
depressor effect of AP I1 and occurs without a concomitant cardiovas-
cular response to AP I11 (Benyajati and Yokota, 1990). If sharks are
placed in 70% SW, urine formation rate increases nearly six times and
chloride excretion triples (Solomon et al., 1988); in 90% SW, GFR and
urine formation increases nearly twofold (Benyajati and Yokota, 1988).
Under these circumstances AP I1 or I11 increases urine formation and
GFR and has a synergistic stimulatory effect with norepinephrine
(Benyajati and Yokota, 1989; Solomon et al., 1988; Yokota and Benya-
jati, 1988).
     ANP-mediated diuresis and saliuresis are found in teleosts with
(trout) or without (toadfish, Opsanus tau) glomeruli. In trout, rANP
 [Ilez61increases urine formation and electrolyte excretion irrespective
of a hypertensive (Duff and Olson, 1986) or hypotensive (Olson and
Duff, 1992) cardiovascular response. rANP [Ile2"] increases urinary
excretion of sodium and chloride three to four times, whereas urine
formation and potassium excretion are doubled (Duff and Olson, 1986).
This suggests that both GFR and fractional excretion of sodium and
chloride contribute to the response. Homologous heart extracts and AP
I11 increase urine formation and sodium excretion in the aglomerular
toadfish but do not affect potassium or divalent cation excretion (Lee
and Malvin, 1987), indicative of a direct stimulatory effect of ANPs on
renal tubular cells, although local changes in renal perfusion cannot be
230                                                   KENNETH R. OLSON

excluded. A potassium-sparing effect of heart extracts and eANP or
BNP is also observed when fish extracts or peptides are injected into
mammals (Takei et al., 1989,1990a; Westenfelder et al., 198813);in fact,
high doses of eel peptides may be antikaliuretic (Takei et al., 1989,

    g. ANP Effects on Other Zonoregulatory Organs. Although infor-
mation is quite limited, there is some evidence to indicate that ANP
affects several nonrenal volume or osmoregulatory organs. ANP acts as
a hypovolemic hormone in the shark rectal gland. Synthetic ANP and
shark heart extracts increase rectal gland chloride and volume excre-
tion (Solomon et al., 1985a) by releasing vasoactive intestinal peptide
within the gland (Silva et al., 1987). Shark CNP is a potent stimulus for
rectal gland secretion; during continuous perfusion as little as         M
sCNP increases secretion over threefold (Solomon et al., 1992). Unlike
ANP and BNP, sCNP appears to have a direct effect on glandular
secretion as it alone will bind to rectal gland membranes and stimulate
guanylate cyclase (Solomon et aZ., 1992).
    AP I1 stimulates short-circuit current and conductance by the
chloride cell-rich opercular epithelium of either FW- or SW-adapted
killifish, Fundulus heteroclitus (Scheide and Zadunaisky, 1988). The
AP 11 effect is independent of neural activity or P-adrenergic stimula-
tion. AP I11 does not affect salt transport by the opercular epithelium of
the SW adapted winter flounder, Pseudopleuronectes arnericanus
(O’Grady et al., 1985). In contrast to the apparent stimulatory effects
of ANPs on renal, rectal gland, and opercular epithelial ion transport,
A P I and I11 inhibit salt transport by the intestine of SW winter floun-
der, P . americanus (O’Grady, 1989;O’Grady et al., 1985).The intestine
is more sensitive to AP I11 than API; ECsos of 7 x lO-’M and 7 x lops
M , respectively (O’Grady et al., 1985). The efficacy of AP I11 is similar
to ECSOSreported for fish vessels (see earlier discussion).
    The physiological significance of ANP effects on opercular and gut
epithelia are not known, although these studies suggest that NPs de-
crease salt intake and promote salt excretion. In uliuo, a single injection
of hANP into either FW trout or trout on a high salt diet has no effect on
ionoregulation or branchial sodium exchange (Eddy et al., 1990).
However, hANP increases sodium efflux when it is injected into SW
flounder, Platichthys jlesus, dab, Limanda limanda, and plaice, Pleu-
ronectes platessa (Arnold-Reed et al., 1991). Furthermore, hANP in-
creased plasma cortisol in flounder (Arnold-Reed et al., 1991) indicat-
ing that NPs may have both direct short-term and indirect long-term
effects on osmoregulatory organs. The effects of long-term ANP admin-
istration on osmoregulatory processes remain to be determined.

    h. Mechanisms ojANP Action: Interaction with Other Endocrine
Systems. The relationship of atrial peptides to other endocrine systems
in fish, especially those with known osmoregulatory functions, is just
beginning to be evaluated. ANP stimulates release of l-a-
hydroxycorticosterone by perifused elasmobranch interrenal tissue
(O’Tool et al., 1990). I n teleosts, hANP increases plasma cortisol in SW
flounder, Platichthys flesus, and stimulates cortisol release by peri-
fused interrenal tissue from SW but not FW trout (Amold-Reed and
Balment, 1991; Arnold-Reed et al., 1991). These stimulatory effects of
ANP are opposite to the well-known inhibitory effect of ANP on aldo-
sterone release in mammals but support the general concept of an
interactive mechanism of action of ANP on other endocrine systems. It
is likely that many of the physiological effects of ANP and related
peptides are mediated through similar indirect processes.

    i. Subcellular Mechanisms. Relatively little is known about the
subcellular mechanism(s) of action of atrial peptides in fish tissues,
however, in most instances they appear similar to those found in mam-
mals. Olson and Meisheri (1989)have shown that rANP [Hez6]inhibits
norepinephrine stimulated contractions of celiacomesenteric arteries
in Ca”-free medium. Thus, as is the case in mammals, the peptide acts
at a level central to receptor activated release of intracellular calcium.
The effects of NP on guanylate cyclase activity in all fish tissues exam-
ined to date are consistent with those observed in mammals. Incuba-
tion of trout arteries with rANP [Ilez6]increases tissue levels of cGMP
(D. W. Lipke and K. R. Olson, unpublished observation) and hANP
increases cGMP in gills of SW, but not FW rainbow trout (Balment and
Lahlou, 1987). AP 111 also increases cGMP in the winter flounder, P .
americanus, intestine (O’Grady, 1989; O’Grady et al., 1985). Thus it
would appear that the basic intracellular response of fish vessels to
ANP is similar to that found in mammals.


    Fluid volume, fluid hydraulic pressure and osmolarity are inti-
mately related; in fact they are probably different attributes of a more
general physiological phenomenon. Although osmoregulation and
blood pressure regulation have been intensively examined in fish,
fluid compartments and their control have received surprisingly little
attention. Blood and extracellular fluid volume are governed by a
variety of physical and physiological processes. Factors that affect
232                                                              KENNETH R. OLSON

water and salt movement across the gills, skin, gut, and kidney deter-
mine overall fluid balance. Fluid distribution between intravascular
and extravascular compartments is further subjugated to hydraulic,
osmotic, and mechanical forces in the microcirculation. The remark-
able constancy of compartmental volumes, even in the face of adverse
circumstances, emphasizes that homeostatic reflexes, especially in
bony fish, are highly developed and efficient.
    Appreciation of volume regulatory mechanisms in fish has suffered
from a lack of information on the nature and disposition of the fluid
compartments themselves, the relative importance of various factors
that affect intercompartmental fluid movements, and the regulatory
processes involved. T h e afferent limb of volume regulatory reflexes in
fish, or for that matter vertebrates in general, has not been character-
ized. Efferent control of appropriate effectors of fluid balance is under
investigation. Two systems, the renin-angiotensin system and the
atrial natriuretic peptides, have been shown to affect a variety of vascu-
lar and extravascular effectors and these systems may play an impor-
tant role in volume homeostasis. The kallikrein-kinin system may also
prove to have important systemic as well as intraorgan cardiovascular
functions. Undoubtedly, other central and peripheral systems are also
    Of all vertebrates, fish are probably the most versatile and manipu-
lative model, from both phylogenetic and physiological perspectives,
in which fluid compartments and volume regulation can be examined.
I hope this chapter provides a convenient starting point for such


    This chapter is dedicated to the colleagues and students that I have had the privilege
to work with and learn from. I would like to express my sincere appreciation to Connie
Gordon and Kathy Drajus for their excellent secretarial and librarian help and to Dixie
Kullman for her technical assistance. Unpublished experiments were supported by
National Science Foundation Grants DCB-8616028, DCB-9004245, and DCB-9105247.


Abo Hegab, S. A., and Hanke, W. (1982). Electrolyte changes and volume regulatory
   processes in the carp (Cyprinus carpio) during osmotic stress. Comp. Biochem.
   Physiol. [A] 71, 157-164.
3. BLOOD AND EXTRACELLULAR FLUID VOLUME REGULATION                                    233

Abo Hegab, S. A,, and Hanke, W. (1983). The significance of the amino acids during
     osmotic adjustment in teleost fish 11. Changes in the stenohaline Cyprinus carpio.
     Comp. Biochem. Physiol. [A] 74,537-543.
Acierno, R., Axelsson, M., Tota, B.. and Nilsson, S. (1991).Hypotensive effect of atrial
     natriuretic factor (ANF) in the Atlantic cod, Gadus morhua. Comp. Biochem.
     Physiol. [C] 99, 11-14.
Akagi, H., Hayashi, T., Nakayama, T., Nakajima, T., Watanabe, T. X., and Sokabe, H.
     (1982). Comparative studies on angiotensins. VI. Structure of angiotensin I pro-
     duced by renal renin of the dog, guinea pig, and rabbit, and re-examination of the
     peptides of the pig, horse, and ox using homologous renin sources. Chem. Pharm.
     Bull. 30,2498-2510.
Albert, S. N. (1971).“Blood Volume and Extracellular Fluid Volume,” 2nd ed. Charles
     C. Thomas, Springfield, Mass.
Alexander, J. B. (1977).Protein concentration in the serum ofthe Atlantic salmon (Salmo
     salar) in Northwest England and Northern 1reland.J. Fish. Biol. 11,673-679.
Alt, J . M., Stoke, H., Eisenbach, G. M., and Walvig, F. (1981).Renal electrolyte and fluid
     excretion in the Atlantic hagfish Myxine glutinosa. J. E x p . Biol. 91,323-330.
Arillo, A., Uva, B., and Vallarino, M. (1981). Renin activity in rainbow trout (Salmo
     gairdneri Rich) and effects of environmental ammonia. Comp. Biochem. Physiol. [A]
     68,307-31 1.
Arnold-Reed, D. E., and Balment, R. J. (1991).Atrial natriuretic factor stimulates in uioo
     and in uitro secretion of cortisol of teleosts. J . Endocrinol. 128, R17-R20.
Arnold-Reed, D., Hazon, N., and Balment, R. J. (1991). Biological actions of atrial
     natriuretic factor in flatfish. Fish Physiol. Biochem. 9,271-277.
Assem, H., and Hanke, W. (1979). Volume regulation of muscle cells in the euryhaline
     teleost, Tilapia mossambica. Comp. Biochem. Physiol. [A] 64,17-23.
Assem, H., and Hanke, W.(1981). Cortisol and osmotic adjustment of the euryhaline
     teleost, Sarotherodon mossambicus. Gen. Comp. Endocrinol. 43,370-380.
Assem, H., and Hanke, W. (1983). The significance of the amino acids during osmotic
     adjustment in teleost fish I. Changes in the euryhaline Sarotherodon mossambicus.
     Comp. Biochem. Physiol. [A] 74,531-536.
Aukland, K. (1984).Distribution of body fluids: Local mechanisms guarding interstitial
     fluid volume. J. Physiol. (Paris) 79, 395-400.
Avtalion, R. R., Gorlin, A., Gutwirth, E., and Wojdani, A. (1974).Determination ofblood
     volume in fish. A new method. State of Israel, Ministry of Agriculture-Fisheries 26,
Babiker, M. M., Rankin, J. C., and Eido, G. (1979).Renal clearance, tissue distribution
     and binding by plasma proteins of 51Cr EDTA in Tilapia nilotica (L). Comp. Bio-
     chem. Physiol. [A] 63,71-74.
Bailey, J . R., and Randall, D. J. (1981). Renal perfusion pressure and renin secretion in
     the rainbow trout, Salmo gairdneri. Can.J . Zool. 59, 1220-1226.
Ballantyne, J. S., Moyes, C. D., and Moon, T. W. (1987). Compatible and counteracting
     solutes and the evolution of ion and osmoregulation in fishes. Can. J. Zool. 65,
Balment, R. J., and Carrick, S. (1985). Endogenous renin-angiotensin system and drink-
     ing behavior in flounder. Am. J. Physiol. 248, R157-R160.
Balment, R. J., and Lahlou, B. (1987). Atrial natriuretic peptide (ANP) modulation of
     second messenger activity in isolated trout gill cells. J . Endocrinol. 115s .[Abstract
Baranowski, R. L., and Westenfelder, C. (1989).Secretion of atrial natriuretic peptide
     from fish atrial and ventricular myocytes in tissue culture. Life Sci. 44, 187-191.
234                                                                 KENNETH R. OLSON

Barron, M. G., Tarr, B. D., and Hayton, W. L. (1987). Temperature-dependence
    of cardiac output and regional blood flow in rainbow trout, Salmo gairdneri
    Richardson. J. Fish Biol. 31,735-744.
Bassingthwaighte, J. B., and Goresky, C. A. (1984). Modeling in the analysis ofsolute and
    water exchange in the microvasculature. In “Handbook of Physiology. The Cardio-
    vascular System” (E. M. Renkin and S. R. Michel, eds.), Vol. IV, pp. 549-626.
    Williams and Wilkins, Baltimore, Maryland.
Bath, R. N., and Eddy, F. B. (1979a). Ionic and respiratory regulation in rainbow trout
    during rapid transfer to seawater. J . Comp. Physiol. [ B ] 134,351-357.
Bath, R. N., and Eddy, F. B. (1979b). Salt and water balance in rainbow trout (Salmo
    gairdneri) rapidly transferred from fresh water to sea water.J. E x p . Biol. 83,193-202.
Beasley, D., Shier, D. N., Malvin, R. L., and Smith, G. (1986). Angiotensin-stimulated
    drinking in marine fish. Am. J . Physiol. 250, R1034-R1038.
Bedford, J. J. (1983). The composition ofthe fluid compartments oftwo chondrichthyans,
    Callorhyncus millii and Squalus acanthias. Comp. Biochem. Physiol. [A] 76,75-80.
Benyajati, S., and Yokota, S. D. (1988). Hormonal regulation of glomerular filtration rate
    in a marine elasmobranch (Squalusacanthias).Bull. Mt. Desert Island Biol. Labora-
    tory. 27,56-58.
Benyajati, S., and Yokota, S. D. (1989). Hormonal regulation of renal function during
    environmental dilution and volume loading in the spiny dogfish. Xlth Int. Symp.
    Comp. Endocrinol. p. 31. [Abstract]
Benyajati, S., and Yokota, S. D. (1990). Renal effects of atrial natriuretic peptide in a
    marine elasmobranch. Am. J. Physiol. 258, R1201-Rl206.
Bert, J . L., and Pearce, R. H. (1984). The interstitium and microvascular exchange. I n
    “Handbook of Physiology. The Cardiovascular System” (E. M. Renkin and S. R.,
    Michel, eds.), Vol. IV, pp. 521-547. Williams and Wilkins, Baltimore, Maryland.
Bert, J. L., Bowen, B. D., and Reed, R. K. (1988). Microvascular exchange and interstitial
    volume regulation in the rat: Model validation. Am. J . Physiol. 254, H384-H399.
Beyenbach, K. W., and Kirschner, L. B. (1976).The unreliability ofmammalian glomeru-
    lar markers in teleostean renal studies. J . E x p . Biol. 64,369-378.
Bittner, A., and Lang, S. (1980). Some aspects of the osmoregulation of Amazonian
    freshwater stingrays (Potamotrygon hystrix) I. Serum osmolality, sodium and
    chloride content, water content, hematocrit and urea level. Comp. Biochem. Physiol.
    [A] 67,9-13.
Blaine, E. H . (1990). Atrial natriuretic factor plays a significant role in body fluid homeo-
    stasis. Hypertension 15,2-8.
Blair-West, J. R., Coghlan, J. P., Denton, D. A., Gibson,A. P., Oddie,C. J., Sawyer, W. H.,
    and Scoggins, B. A. (1977). Plasma renin activity and blood corticosteroids in the
    Australian lungfish Neoceratodus forsteri. J . Endocrinol. 74, 137-142.
Bornancin, M., Isaia, J., and Masoni, A. (1985). A re-examination of the technique of
    isolated, perfused trout head preparation. Comp. Biochem. Physiol. [A] 81,35-41.
Bovy, P. R. (1990). Structure activity in the atrial natriuretic peptide (ANP) family. Med.
    Res. Reu. 10, 115-142.
Boyd, R. B., Atkin, J., Thompson, V. W., and DeVries, A. L. (1990). absence of binding
    and impermeability to ferritins of gill endothelium in marine teleosts. Fish Physiol.
    Biochem. 8,53-60.
Boyd, R. B., DeVries, A. L., Eastman, J. T., and Pietra, G. G. (1980). Permeability to
    protein tracers of gill secondary lamellae from high latitude fishes. Anatomical
    Record 196,22A. [Abstract]
3.   BLOOD A N D EXTRACELLULAR FLUID VOLUME REGULATION                               235

Bradely, T. M., and Rourke, A. W. (1984).An electrophoretic analysis of plasma proteins
   from juvenile Oncorhynchus tshawytscha (Walbaum).J . Fish Biol.24,703-709.
Bradley, T. M., and Rourke, A. W. (1988). Smoltification in steelhead trout (Salmo
   gairdneri): Developmental aspects of plasma constituents. Fish Physiol. Biochem. 5,
Brenner, B. M., Ballermann, B. J., Gunning, M. E., and Zeidel, M. L. (1990). Diverse
   biological actions of atrial natriuretic peptide. Pharmacol. Rev. 70,665-669.
Brown, J. A., Gray, C. J., and Taylor, S. M. (1990a). Direct effects of angiotensin I1 on
    glomerular ultrastructure in the rainbow trout, Salmo gairdneri. Cell Tissue Res.
Brown, J. A,, Jackson, B. A,, Oliver, J. A., and Henderson, I. W. (1978). Single nephron
   filtration rates (SNGFR) in the trout, Salmo gairdneri. Pfluegers Arch. Eur. J.
   Physiol. 377, 101-108.
Brown, J. A,, Oliver, J. A., Henderson, I. W., and Jackson, B. A. (1980). Angiotensin and
    single nephron glomerular function in the trout Salmo gairdneri. Am.J . Physiol. 239,
    R509-R5 14.
Brown, J. A., Taylor, S. M., and Gray, C. J. (1983). Glomerular ultrastructure ofthe trout,
    Salmo gairdneri: Glomerular capillary epithelium and the effects of environmental
    salinity. Cell Tissue Res. 230,205-218.
Brown, J. A,, Taylor, S. M., and Gray, C. J. (1990b).Glomerular binding ofangiotensin I1
    in the rainbow trout, Salmo gairdneri. Cell Tissue Res. 259,479-482.
Bull, J. M., and Morris, R. (1967).Studies on fresh water osmoregulation in the ammo-
    coete larva of Lampetra planeri (Bloch). I. Ionic constituents, fluid compartments,
    ionic compartments and water balance. J. E x p . Biol. 47,485-494.
Burton, R. F. (1988). The protein content of extracellular fluids and its relevance to the
    study of ionic regulation: Net charge and colloid osmotic pressure. Comp. Biochem.
    Physiol. [A] 90, 11-16.
Cameron, J. N. (1980). Body fluid pools, kidney function, and acid-base regulation in the
    freshwater catfish lctalurus punctatus. J. Erp. Biol. 86, 171-185.
Cameron, J. N. (1985). The bone compartment in a teleost fish, Ictalurus punctatus:
    Size, composition and acid-base response to hypercapnia.J. E x p . Biol. 117,307-318.
Capelli, J. P., Wesson, L. G., Jr., and Aponte, G. E. (1970). A phylogenic study of the
    renin-angiotensin system. Am. J. Physiol. 218, 1171-1178.
Carrick, S., and Balment, R. J. (1983).The renin-angiotensin system and drinking in the
    euryhaline flounder, Platichthysflesus. Gen. Comp. Endocrinol. 51,423-433.
Carroll, R. G. (1981).Vascular response of the dogfish and sculpin to angiotensin 11. Am.
   J . Physiol. 240, R139-Rl43.
Carroll, R. G., and Opdyke, D. F. (1982). Evolution of angiotensin 11-induced catechol-
    amine release. Am. J. Physiol. 243, R65-R69.
Chan, D. K. O., and Chow, P. H. (1976).The effects of acetylcholine, biogenic amines
    and other vasoactive agents on the cardiovascular functions of the eel, Anguilla
   japonica. J. E x p . Zool. 196, 13-26.
Chan, D. K. O., Jones, I. C., Henderson, I. W., and Rankin, J. C. (1967).Studies on the
    experimental alteration of water and electrolyte composition of the eel (Anguilla
    anguilla L.).J. Endocrinol. 37,297-317.
Chapeau, C., Gutkowska, J., Schiller, P. W., Milne, R. W., Thibault, G., Garcia, R.,
    Genest, J., and Cantin, M. (1985). Localization of immunoreactive synthetic atrial
    natriuretic factor (ANF) in the heart of various animal species. J. Histochem. Cy-
    tochem. 33,541-550.
Chester Jones, I., Chan, D. K. O., and Rankin, J. C. (1969). Renal function in the
236                                                                    KENNETH R . OLSON

    European eel (Anguilla anguilla L.): Changes in blood pressure and renal function
    of the freshwater eel transferred to seawater. J. Endocrinol. 43,9-19.
Chester Jones, I., and Henderson, I. W. (1965). Electrolyte changes in the European eel
    (Anguilla anguilla L.). J . Endocrinol. 32, iii-iv.
Chester Jones, I., Henderson, I. W., Chan, D. K. O., Rankin, J. C., Mosley, W., Brown,
    J. J., Lever, A. F., Robertson, J. I. S . , and Tree, M. (1966). Pressor activity in extracts
    of the corpuscles of Stannius from the European eel (Anguilla anguilla L.).J. Endo-
    crinol. 34,393-408.
Cholette, C., Gagnon, A., and Germain, P. (1970).Isosmotic adaptation in Myxine gluti-
    nosa L.-I. Variations of some parameters and role of the amino acid pool of the
    muscle cells. Comp. Biochem. Physiol. 33,333-346.
Christensen, J. A., Mikeler, E., and Bohle, A. (1989). Granular epithelioid cells of the
    kidneys in salmon adapted to fresh- and seawater. Anat. Rec. 223,21-26.
Christensen, J. A., Taugner, R., Meyer, D. S., and Bohle, A. (1987).The granular epithe-
    lioid cells in the kidney of the lemon sole (Pleuronectes microcephalus Donovani).
    Cell Tissue Aes. 249, 137-143.
Churchill, P. C., Malvin, R. L., and Churchill, M. C. (1985a). Lack of renal effects of
    DOCA, ACTH, spironolactone, and angiotensin I1 in Squalus acanthias. J. E x p .
    Zool. 234, 17-22.
Churchill, P., Malvin, R., Churchill, M., Beasley, D., and Shier, D. (1985b). Antidiuretic
    effect of [Sar', Val5, Alas] angiotensin I1 in Lophius americanus. J. E x p . Zool. 233,
Churchill, P. C., Malvin, R. L., Churchill, M. C., and McDonald, F. D. (1979). Renal
    function in Lophius americanus: Effects of angiotensin 11. Am. J. Physiol. 236,
    R297-Fi30 1.
Claeson, G., Aurell, L., Friberger, P., Gustavsson, S . , and Karlsson, G. (1978). Designing
    of peptide substrates. Haemostasis 7 , 6 2 4 8 .
Clark, S., Whitmore, D. H., Jr., and McMahon, R. F. (1979). Considerations of blood
    parameters of largemouth bass, Micropterus salmoides. J. Fish Biol. 14,147-158.
Conklin, D. J., and Olson, K. R. (1990).Vascular compliance oftrout vessels. Physiologist
    33, A113. [Abstract]
Conte, F. P., Wagner, H. H., and Harris, T. 0. (1963). Measurement of blood volume in
    the fish (Salmo gairdneri). Am. J. Physiol. 205,533-540.
Cowley, A. W., Jr., and Roman, R. J. (1989). Control of blood and extracellular volume.
    Bailliere's Clin. Endocrinol. Met. 3,331-369.
Cserr, H. F., and Bundgaard, M. (1984). Blood-brain interfaces in vertebrates: A compar-
    ative approach. Am. J. Physiol. 246, R277-R288.
Cserr, H. F., Bradbury, M. W. B., Mackie, K., and Moody, E. J. (1983). Control of
    extracellular ions in skate brain during osmotic disturbances. Am. J. Physiol. 245,
Davidson, W. S . , Birt, V. L., Birt, T. P., and Green, J. M. (1988).Palmitate-binding, serum
    albumin-like proteins in salmonids. F E B S Lett. 233,299-302.
Davie, P. S. (1982). Changes in vascular and extravascular volumes of eel muscle in
    response to catecholamines: The function of the caudal lymphatic heart. J. E x p . B i d .
Davie, P. S., and Daxboeck, C. (1982). Effect of pulse pressure on fluid exchange
    between blood and tissues in trout gills. Can.J. Zool. 60, 1000-1006.
deBold, A. J. (1982).Atrial natriuretic factor in the rat heart. Proc. Soc. E x p . Biol. Med.
3.   BLOOD AND EXTRACELLULAR FLUID VOLUME REGULATION                                 237

deBold, A. J., Borenstein, B., Veress, A. T., and Sonnenberg, H. (1981). A rapid and
    potent natriuretic response to intravenous injection of atrial myocardial extract in
    rats. Life Sci. 28,89-94.
Decourt C., and Lahlou, B. (1987). Evidence for the direct intervention of angiotensin I1
    in the release of cortisol in teleost fishes. Life Sci. 41, 1517-1524.
DeLaney, R. G., Lahiri, S.. Hamilton, R., and Fishman, A. P. (1977). Acid-base balance
    and plasma composition in the aestivating lungfish (Protopterus).Am. J. Physiol.
    232, R10-Rl7.
Denton, J. E., and Yousef, M. K. (1976). Body composition and organ weights ofrainbow
    trout, Salmo gairdneri.]. Fish Biol. 8,489-499.
Diz, D. I. (1985). Bradykinin and related peptides in central control ofthe cardiovascular
    system. Peptides 6,57-64.
Doolittle, R. F. (1984). Evolution of the vertebrate plasma proteins. In “The Plasma
    Proteins” (F. W. Putnam, ed.), pp. 317-360. Academic Press, New York.
Doolittle, R. F. (1987). The evolution of the vertebrate plasma proteins. Biol. Bull. 172,
Duff, D. W., and Fleming. W. R. (1972a). Sodium metabolism of the freshwater cypri-
    nodont, Fundulus catenatus. J . Comp. Physiol. 80,179-189.
Duff, D. W., and Fleming, W. R. (197213). Some aspects of sodium balance in the
    freshwater cyprinodont Fundulus olioaceus. J. Comp. Physiol. 80,191-199.
Duff, D. W., and Olson, K. R. (1986). Trout vascular and renal responses to atrial
    natriuretic factor and heart extracts. Am. J . Physiol. 251, R639-R642.
Duff, D. W., and Olson, K. R. (1989). Response of rainbow trout to constant-pressure and
    constant-volume hemorrhage. Am. J. Physiol. 257, R1307-R1314.
Duff, D. W., and Olson, K. R. (1990a). Effects of a-adrenergic blockade on ANF response
    in rainbow trout. FASEBJ. 4, A1090. [Abstract]
Duff, D. W., and Olson, K. R. (1990b). Effects of salinity and volume expansion on trout
    plasma ANF. Physiologist 33, A38. [Abstract]
Duff, D. W., and Olson, K. R. (1992). Atrial natriuretic peptide clearance receptors in
    trout: Effects of receptor inhibition in oioo.]. E x p . Zool. (in press.)
Duff, D. W., Fitzgerald, D., Kullman, D., Lipke, D. W., Ward, J., and Olson, K. R. (1987).
    Blood volume and red cell space in tissues of the rainbow trout, Salmo gairdneri.
    Comp. Biochem. Physiol. 87A, 393-398.
Dunn, R. S., and Perks, A. M. (1975). Comparative studies of plasma kinins: The kal-
    likrein-kinin system in poikilotherm and other vertebrates. Gem Comp. Endocrinol.
    26, 165-178.
Dzau, V. J., Burt, D. W., and Pratt, R. E. (1988). Molecular biology of the renin-
    angiotensin system. Am. J . Physiol. 255, F563-F573.
Eddy, F. B., and Bath, R. N. (1979). Ionic regulation in rainbow trout (Salmo gairdneri)
    adapted to fresh water and dilute sea water.J. E x p . Biol. 83, 181-192.
Eddy, F. B., Smith, N. F., Hazon, N., and Grierson, C. (1990). Circulatory and iono-
    regulatory effects of atrial natriuretic peptide on rainbow trout (Salmo gairdneri
    Richardson) fed normal or high levels of dietary salt. Fish Physiol. Biochem. 8,
Elger, E., Elger, B., Hentschel, H., and Stolte, H. (1987). Adaptation of renal function to
    hypotonic medium in the winter flounder (Pseudopleuronectes americanus).
   J. Comp. Physiol. [ B ] 157,21-30.
Elger, B., Ruhs, H., and Hentschel, H. (1988). Glomerular permselectivity to serum
    proteins in rainbow trout (Salmo gairdneri). Am. J. Physiol. 255, R418-R423.
238                                                               KENNETH R. OLSON

Elger, M., Wahlqvist, I., and Hentschel, H. (1984). Ultrastructure and adrenergic inner-
    vation of preglomerular arterioles in the euryhaline teleost, Salmo gairdneri. Cell
    Tissue Res. 237,451-458.
Ellis, A. G., and Smith, D. G. (1983). Edema formation and impaired O2 transfer in
    Ringer-perfused gills of the eel, AnguilEa australis. J. E x p . Zool. 227,371-380.
Engel, D. W., Hettler, W. F., Coston-Clements L., and Hoss, D. E. (1987). The effect of
    abrupt salinity changes on the osmoregulatory abilities of the Atlantic menhaden
    Breooortia tyrannus. Comp. Biochem. Physiol. [A] 86,723-727.
Epstein, F. H., Clark, B., Taylor, M., Silva, P., Dick, L., Silva, M., and Solomon, R. J.
    (1988). Atrial natriuretic peptide (ANP) assayed in plasma of Squalus acanthias.
    Bull. Mt. Desert lsland Biological Laboratory 27,72-73.
Epstein, F. H., Clark, B., Taylor, M., and Silva, P. (1989).Plasma levels of atrial natri-
    uretic peptide in Anguilla rostrata during adaptation to freshwater (FW) and sea-
    water (SW). Bull. Mt. Desert lsland Biological Laboratory 28, 16.
Erdos, E. G., and Skidgel, R. A. (1989). Neutral endopeptidase 24.11 (enkephalinase)
    and related regulators of peptide hormones. FASEBJ. 3, 145-151.
Evans, D. H. (1990). An emerging role for a cardiac peptide hormone in fish osmoregu-
    lation. Annu. Reo. Physiol. 52,43-60.
Evans, D. H. (1991).Rat atriopeptin dilates vascular smooth muscle of the ventral aorta
    from the shark (Squulus acanthias) and the hagfish (Myxine glutinosa).J. E x p . Biol.
Evans, D. H., and Weingarten, K. E. (1989). Vasoactive effects of adenosine, vasoactive
    intestinal peptide, and atriopeptin on ventral aortic rings from the shark, Squalus
    acunthias. Bull. Mt. Desert lsland Biological Laboratory 28,4-5.
Evans, D. H., Chipouras, E., and Payne, J. A. (1989). Immunoreactive atriopeptin in
    plasma of fishes: Its potential role in gill hemodynamics. Am.J. Physiol. 257, R939-
Fange, R. (1985). Regulation ofblood and body fluids in primitive fish groups. Eoolution.
    Biol. Primitive Fishes 103,253-273.
Farrell, A. P., and Milligan, C. L. (1986). Myocardial intracellular pH in a perfused
     rainbow trout heart during extracellular acidosis in the presence and absence of
     adrenaline. J. E x p . Biol. 125,347-359.
Feeney, R. E., and Brown, W. D. (1974). Plasma proteins in fishes. In “Chemical
     Zoology” (M. Florkin and B. T. Scheer, eds.), pp. 307-329.
Feldschuh, J . (1990). Blood volume measurements in hypertensive disease. In “Hyper-
     tension: Pathophysiology, Diagnosis, and Management” ( J . H. Laragh and B. M.
     Brenner, eds.), pp. 339-347. Raven Press, New York.
Fellows, F. C. I., and Hird, F. J . R. (1981). Fatty acid binding proteins in the serum of
     various animals. Comp. Biochem. Physiol. [Bl68,83-87.
Fenstermacher, J. D., and Patlak, C. S. (1977). CNS, CSF, and extradural fluid uptake of
     various hydrophilic materials in the dogfish. Am. J. Physiol. 232, R45-R53.
Fenwick, J. C., and So, Y. P. (1981). Effect of an angiotensin on the net influx of calcium
     across an isolated perfused eel gill. C a n . ] . Zool. 59, 119-121.
Filosa, M. F., Sargent, P. A,, Fisher, M. M., and Youson, J. H. (1982).An electrophoretic
     and immunoelectrophoretic characterization of the serum proteins of the adult lam-
     prey, Petromyzon marinus L. Comp. Biochem. Physiol. [ B ] 72,521-530.
Filosa, M. F., Sargent, P. A., and Youson, J . H. (1986). An electrophoretic and immuno-
     electrophoretic study of serum proteins during the life cycle of the lamprey, Petro-
     myzon marinus L. Comp. Biochem. Physiol. [ B ] 83, 143-149.
Finstad, B., Nilssen, K. J., and Arnesen, A. M. (1989). Seasonal changes in sea-water
     tolerance of Arctic charr (Saloelinus alpinus).J. Comp. Physiol. [ B l 159,371-378.
3.   BLOOD AND EXTRACELLUL,AR FLUID VOLUME REGULATION                                  239

Finstad, B., Staurnes, M., and Riete, 0.B. (1988).Effect oflow temperature on sea-water
    tolerance in rainbow trout, Salmo gairdneri. Aquaculture 72,319-328.
Fletcher, G. L. (1975).The effects of capture, “stress,” and storage ofwhole blood on the
    red blood cells, plasma proteins, glucose, and electrolytes of the winter flounder
    (Pseudopleuronectes americanus). Can.J . Zool. 53, 197-206.
Fletcher, G. L., and Haedrich, R. T. (1987). Rheological properties of rainbow trout
    blood. Can.J. Zool. 65,879-883.
Forster, M. E., Davie, P. S., Davison, W., Satchell, G. H., and Wells, R. M. G. (1988).
    Blood pressures and heart rates in swimming hagfish. Comp. Biochem. Physiol. [A]
Forster, M. E., Davison, W., Satchell, G. H., and Taylor, H. H. (1989).The subcutaneous
    sinus of the hagfish, Eptatretus cirrhatus and its relation to the central circulating
    blood volume. Comp. Biochem. Physiol. [A] 93,607-612.
Freda, J., and McDonald, D. G. (1988).Physiological correlates of interspecific variation
    in acid tolerance in fish. J. E x p . Biol. 136,243-258.
Fugelli, K., and Vislie, T. (1980). Physiological responses to acid water in fish 1.Effects
    from changes in plasma osmolality and ion concentration on heart ventricle water
    content and intracellular concentrations of K+ and taurine in the brown trout (Salmo
    trutta L.). Proc. Int. Conf. Ecol. Impact Acid Precip., Norway 346-347.
Galardy, R., Podhasky, P., and Olson, K. R. (1984). Angiotensin-converting enzyme
    activity in tissues of the rainbow trout. J. E x p . Zool. 20, 155-158.
Galli, S. M., and Kiang, K. (1990). Angiotensin I1 like immunoreactivity in sharks:
    Response to hemorrhage and sodium depletion. FASEB J. 4, A955. [Abstract]
Galli, S. M., Evans, D. H., Kimura, B., and Phillips, M. I. (1988).Changes in plasma and
    brain levels of atrial natriuretic peptide in fish adapting to fresh water and sea water.
    FASEB J. 2, A524. [Abstract]
Gingerich, W. H., and Pityer, R. A. (1989).Comparison of whole body and tissue blood
    volumes in rainbow trout (Salmo gairdneri) with lZsI bovine serum albumin and
    51Cr-erythrocyte tracers. Fish Physiol. Biochem. 6,39-47.
Gingerich, W. H., Pityer, R. A,, and Rach, J . J. (1987).Estimates of plasma, packed cell
    and total blood volume in tissues of the rainbow trout (Salmo gairdneri). Comp.
    Biochem. Physiol. [A] 87,251-256.
Gingerich, W. H., Pityer, R. A,, and Rach, J . J. (1990). Whole body and tissue blood
    volumes of two strains of rainbow trout (Oncorhynchus mykiss). Comp. Biochem.
    Physiol. [A] 97,615-620.
Goetz, K. L. (1990). Evidence that atriopeptin is not a physiological regulator of sodium
    excretion. Hypertension 15,9- 19.
Goldstein, L., and Kleinzeller, A. (1987).Cell volume regulation in lower vertebrates.
    Curr. Topics Memb. Transp. 30, 181-203.
Gonzalez, R. J., and Dunson, W. A. (1987).Adaptations of sodium balance to low pH in a
    sunfish (Enneacanthus obesus) from naturally acidic waters. I. Comp. Physiol. [ B l
Gras, J., Perrier, H., Perrier, C., and Gudefin, Y. (1971).Adrenaline activity in extracellu-
    lar fluid compartments of the lateral muscle of the rainbow trout (Salmo gairdneri
    Richardson): Effect of adrenergic blocking agents. Comp. Biochem. Physiol. [A] 39,
Gray, C. J., and Brown, J. A. (1985).Renal and cardiovascular effects of angiotensin I1 in
    the rainbow trout, Salnio gairdneri. Gen. Comp. Endocrinol. 59, 375-381.
Gray, C. J., and Brown, J. A. (1987). Glomerular ultrastructure of the trout, Salmo
    gazrdneri: Effects of angiotensin I1 and adaptation to seawater. Cell Tissue Res. 249,
240                                                             KENNETH R. OLSON

Greenway, C. V., and Lautt, W. W. (1986). Blood volume, the venous system, preload,
   and cardiac output. Can.J. Physiol. Pharmacol. 64,383-387.
Gupta, 0. P., and Hanke, W. (1982). The effects of osmotic stressors on the stenohaline
   carp (Cyprinus carpio). Comp. Biochem. Physiol. 71, 165-173.
Guyton, A. C. (1990). Renal function curves and control of body fluids and arterial
   pressure. Acta Physiol. Scand. 139, 107-113.
Guyton, A. C. (1991). “Textbook of Medical Physiology,” 8th ed. W.B. Saunders Co.,
Guyton, A. C., Hall, J. E., Coleman, T. G., and Manning, R. D., Jr. (1990).The dominant
   role of the kidneys in the long-term regulation of arterial pressure in normal and
   hypertensive states. In “Hypertension: Pathophysiology, Diagnosis, and Manage-
   ment” ( J . H. Laragh and B. M. Brenner, eds.), pp. 1029-1052. Raven Press, New
Guyton, A. C., Jones, C. E., and Coleman, T. G. (1973).“Circulatory Physiology: Cardiac
    Output and Its Regulation”, 2nd ed., W.B. Saunders Co., Philadelphia.
Hainsworth, R. (1986).Vascular capacitance: Its control and importance. In “Rev. Physi-
    ology, Biochemistry and Pharmacology”, pp. 101-173. Springer-Verlag, New York.
Hall, J. E., and Brands, M. W. (1992).The renin-angiotensin-aldosterone systems: Renal
    mechanisms and circulatory homeostasis. In “The Kidney: Physiology & Pathophys-
    iology” (D. W. Seldin and G. Giebisch, eds.), 1455-1504. Raven Press, New York.
Hanke, W. (1990).Corticosteroid function: Evolutionary aspects. In “Progress in Com-
    parative Endocrinology” (A. Epple, C. G. Scanes, and M. H. Stetson, eds.), pp.
    445-452. Wiley-Liss, Inc., New York.
Hansen, S. P., and Gesser, H. (1987). Extracellular Ca”’, force, and energy state in
    cardiac tissue of rainbow trout. Am. J . Physiol. 253, R838-R847.
Hargens, A. R. (1972).Freezing resistance in polar fishes. Science 176, 184-186.
Hargens, A. R., Millard, R. W., and Johansen, K. (1974). High capillary permeability in
    fishes. Comp. Biochem. Physiol. [A] 48,675-680.
Hargens, A. R., and Perez, M. (1975). Edema in spawning salmon. J . Fish Res. Board
    Can. 32,2538-2541.
Hasegawa, Y., Nakajima, T., and Sokabe, H. (1983).Chemical structure of angiotensin
    formed with kidney renin in the Japanese eel, Anguillajaponica. Biomed. Res. 4,
    4 17-420.
Hasegawa, Y., Watanabe, T., Nakajima, T., and Sokabe, H. (1984).Chemical structure of
    angiotensin formed by inculating plasma with the corpuscles of stannius i n the
   Japanese goosefish, Lophius litulon. Gen. Comp. Endocrinol. 54,264-269.
Hayashi, T., Nakayama, T., Nakajima, T., and Sokabe, H. (1978).Comparative studies on
    angiotensins. V. Structure of angiotensin formed by the kidney of Japanese goosefish
    and its identification by Dansyl method. Chern. P h a m . Bull. 26,215-219.
Hazon, N., and Henderson, I. W. (1985). Factors affecting the secretory dynamics of 1
    a-hydroxycorticosterone in the dogfish, Scyliorhinus canicula. Gen. Comp.Endocri-
    1101. 59,50-55.
Hazon, N., Balment, R. J., Perrott, M., and O’Toole, L. B. (1989).The renin-angiotensin
    system and vascular dipsogenic regulation in elasmobranchs. Gen. Comp. Endocri-
    nol. 74,230-236.
Heisey, S. R. (1968). Brain and choroid plexus blood volumes in vertebrates. Comp.
    Biochem. Physiol. 26,489-498.
Heisler, N. (1982). Intracellular and extracellular acid-base regulation in the tropical
    fresh-water teleost fish Synbranchus marmoratus in response to the transition from
    water breathing to air breathing. J. E x p . Biol. 99,9-28.
3.   BLOOD AND EXTRACELLULAR FLUID VOLUME REGULATION                                  24 1

Henderson, I. W., Hazon, N., and Hughes, K. (1985). Hormones, ionic regulation and
   kidney function in fishes. In “Physiological Adaptations of Marine Animals” (M. S.
   Lauerack, ed.), pp. 245-265. Pindar Croup, Scarborough.
Henderson, I. W., Jotisankasa, V., Mosley, W., and Oguri, M. (1976). Endocrine and
   environmental influences upon plasma cortisol concentrations and plasma renin
   activity of the eel, Anguilla anguilla L. J . Endocrinol. 70,81-95.
Henderson, I. W., Oliver, J. A,, McKeever, A,, and Hazon, N. (1980). Phylogenetic
   aspects of the renin-angiotensin system. In “Advances in Animal and Comparative
   Physiology” (G. Pethes and V. L. Frenyo, eds.), pp. 355-363. Pergamon Press,
Henderson, I. W., Oliver, J. A., McKeever, A,, and Hazon, N. (1981). Phylogenetic
   aspects of the renin-angiotensin system. In “Advances in Animal and Comparative
   Physiology” (C. Pethes and V. L. Frenyo, eds.), p. 355. Pergamon Press, Oxford.
Hickman, C. P., Jr. (1972). Determination of the extracellular fluid volume of a eury-
   haline flounder by kinetic and net retention methods using tritium-labeled inulin.
   C a n . J .Zool. 50,1663-1671.
Hickman, C. P., Jr., McNabb, R. A,, Nelson, J . S., Van Breeman, E. D., and Comfort, D.
   (1964). Effect of cold acclimation on electrolyte distribution in rainbow trout. Can.J .
   Zool. 42,577-597.
Hickman, C. P., Jr., Newcomb, E. W., 111, and Kinter, W. B. (1972).Anomalous behavior
   of 3H- and 14C-labelled inulin and 3H-labelled polyethylene glycol in incubated
   kidney tissue of the winter flounder, Pseudopleuronectes americanus. Bull. Mt.
   Desert Island Biological Laboratory 23,47-50.
Hilmy, A. M., Badawi, H. K., and El-Domiaty, N. A. (1978).Seasonal changes of total
    serum proteins and nonprotein nitrogen in the blood of two fresh water teleosts,
   Cyprinus carpio L. and Bagrus bayad Rupp. Comp. Biochem. Physiol. [A] 59,
Hipkins, S. F., Smith, D. C., and Evans, B. K. (1986).Lack ofadrenergic control ofdorsal
   aortic blood pressure in the resting eel, Anguilla australis. I . E x p . Zool. 238, 155-
Hirano, T., and Hasegawa, S. (1984).Effects of angiotensins and other vasoactive sub-
    stances on drinking in the eel, Anguillajaponica. Zool. Sci. 1, 106-113.
Hirano, T., Takei, Y., and Kobayashi, H. (1978).Angiotensin and drinking in the eel and
    the frog. In “Osmotic and Volume Regulation” (C. B. Jorgensen and E. Skadhange,
    eds.), pp. 123-134. Alfred Benzon Symposium XI, Munksgaard.
Hirohama, T., Uemura, H., Nakamura, S., and Aoto, T. (1988). Atrial natriuretic peptide
    (ANP)-immunoreactivity and ultrastructures of cardiocytes in fish. Zool. Sci. 5,833-
HBbe, H. (1987). Sulphate entry into soft-water fish (Salmo gairdneri, Catostomus
    commersoni) during low ambient pH exposure. J. E x p . Biol. 133,87-109.
HBbe, H., and McMahon, B. R. (1988).Mechanisms of acid-base and ionoregulation in
    white suckers (Catostomus commersoni) in natural soft water. 11. Exposure to a
    fluctuating ambient pH regime. J. Comp. Physiol. [B] 158,67-79.
Hoffert, J. R. (1966).Observations on ocular fluid dynamics and carbonic anhydrase in
   tissues of lake trout (Saloelinus namaycush). Comp. Biochem. Physiol. 17,107-114.
Holmes, W. N., and Donaldson, E. M. (1969).The body compartments and the distribu-
   tion of electrolytes. In “Fish Physiology” (W. S. Hoar and D. J . Randall, eds.), pp.
    1-89. Academic Press, New York.
Houston, A. H. (1959).Osmoregulatory adaptation of steelhead trout (Salmo gairdneri
    Richardson) to sea water. Can.J . Zool. 37,729-748.
242                                                                K E N N E T H K. OLSON

Houston, A. H. (1962). Some observations on water balance in the goldfish Carassius
    auratus L. during cold death. C a n . J . Zool. 40, 1169-1174.
Houston, A. H., and DeWilde, M. A. (1969). Environmental temperature and the body
    fluid system of the fresh-water teleost 111. Hematology and blood volume of ther-
    mally acclimated brook trout, Saloelinus fontinalis. Comp. Biochem. Physiol. 28,
Houston, A. H., and Mearow, K. M. (1979). [14C]PEG-4000, chloride, chloride/
    potassium and sodium spaces as indicators of extracellular phase volume in the
    tissues of rainbow trout, Salmo gairdneri Richardson. Comp. Biochem. Physiol. [A]
Houston, A. H., DeWilde, M. A., and Madden, J. A. (1969).Some physiological conse-
    quences of aortic catheterization on the brook trout (Saloelinus fontinalis).J. Fish
    Res. Board Can. 26,1847-1856.
Houston, A. H., Madden, J. A,, Woods, R. J., and Miles, H. M. (1971).Variations in the
    blood and tissue chemistry of brook trout, Saloelinus fontinalis, subsequent to
    handling, anesthesia, and surgery. J. Fish Res. Board Can. 28,635-642.
Huggel, H. J., Lane, H. C., and Ducret, C. G. (1969). Determination de la courbe
    dhomogeneisation et du volume sanguin circulant de la truite Salmo gairdneri
    Rich. par la methode de la dilution isotopique de 1' I3'I et du 51Cr.J, Physiol. (Paris)
    61, 145-154.
Huxley, V. H., Tucker, V. L., Verburg, K. M., and Freeman, R. H. (1987). Increased
    capillary hydraulic conductivity induced by atrial natriuretic peptide. Circ. Res. 60,
Inouye, A., Kataoka, K., and Tsujioka, T. (1961).On a kinin-like substance in the nervous
    tissue extracts treated with trypsin. Jpn.J. Physiol. 11,319-324.
Isaia, J., and Masoni, A. (1976).The effects ofcalcium and magnesium on water and ionic
    permeabilities in the sea water adapted eel, Anguilla anguilla L. J . Comp. Physiol.
    [B] 109,221-233.
Jackson, A. J. (1981).Osmotic regulation in rainbow trout (Salmo gairdneri) following
    transfer to sea water. Aquaculture 24, 143-151.
Johansen, K., Fange, R., and Johannessen, M. W. (1962).Relations between blood, sinus
    fluid and lymph in Myxine glutinosa L. Comp. Biochem. Physiol. 7,23-28.
Johnson, P. C. (1971). Red cell separation in the mesenteric capillary network. Am. J.
    Physiol. 221, 105-111.
Kakiuchi, Y., Hughes, G. M.,Arai, T., Horimoto, M., Kikuchi, Y ., and Koyama, T. (1981).
    A microanalysis of colloid osmotic pressure. Experientia 37,376-377.
Kellogg, M. D., and Olson, K. R. (1990). Inactivation of angiotensin 11-induced pressor
    response in trout. FASEB J. 4, A704. [Abstract]
Kellogg, M. D., Duff, D. W., Kullman, D. E., and Olson, K. R. (1990). Simultaneous
    determination of blood volume and extracellular fluid spaces in fish with two
     gamma-emitting nuclides. Physiologist 33, A112. [Abstract]
Kenyon, C. J., McKeever, A,, Oliver, J. A,, and Henderson, I. W. (1985). Control of renal
     and adrenocortical function by the renin-angiotensin system in two euryhaline
     teleost fishes. Gen. Comp. Endocrinol. 58,93-100.
Keys, A., and Hill, R. M. (1972).The osmotic pressure of the colloids in fish sera. Comp.
     Biochem. Physiol. [A] 44,28-34.
Khosla, M. C., Bumpus, F. M., Nishimura, H., Opdyke, D. F., and Coviello, A. (1983).
     Synthesis of nonmammalian angiotensins and their comparative pressor properties
     in dogfish shark, domestic chicken, and rat. Hypertension 5, V22-V28.
Khosla, M. C., Nishimura, H., Hasegawa, Y., and Bumpus, F. M. (1985). Identification
3.   BLOOD AND EXTRACELLULAR FLUID VOLUME REGULATION                                  243

    and synthesis of [ 1-asparagine, 5-valine, 9-glycine] angiotensin I produced from
    plasma of American eel Anguilla sostrata. Gen. Comp. Endocrinol. 57,223-233.
Kim, S. H., Cho, K. W., Koh, G. Y., Seul, K. H., So, J . N., and Ryu, H. (1989).Phylogenetic
    study on the immunoreactive atrial natriuretic peptide in the heart. Gen. Comp.
    Endocrinol. 74,127-135.
King, P. A., and Goldstein, L. (1983). Organic osmolytes and cell volume regulation in
    fish. Mol. Physiol. 4,53-66.
Kitzman, J. V., Martin, J. F., Holley, J. H., and Huber, W. G. (1990). Determinations of
    plasma volume, extracellular fluid volume, and total body water in channel catfish.
    Progressioe Fish-Culturist 52,261-265.
Kleinzeller, A., and Ziyadeh, F. N. (1990). Cell volume regulation in epithelia-With
    emphasis on the role of osmolytes and the cytoskeleton. In “Cell Volume Regula-
    tion” (K. W. Beyenbach, ed.), pp. 59-86. Karger, Farmington, Connecticut.
Klitzman, B., and Duling, B. R. (1979). Microvascular hematocrit and red cell flow in
    resting and contracting striated muscle. Am. J . Physiol. 237, H481-H490.
Kloas, W., Flugge, G., Fuchs, E., and Stolte, H. (1988).Binding sites for atrial natriuretic
    peptide in the kidney and aorta of the hagfish (Myrine glutinosa). Comp. Biochem.
    Physiol. [A] 91,685-688.
Kobayashi, H., Uemura, H., and Takei, Y. (1980). Physiological role of the renin-
    angiotensin system during dehydration. In “Avian Endocrinology” (A. Epple and
    M. H. Stetson, eds.), pp. 319-3:30. Academic Press, New York.
Kobayashi, H., Uemura, H., Takei, Y., Itatsu, N., Ozawa, M., and Ichinohe, K. (1983).
    Drinking induced by arigiotensin I1 in fishes. Gen. Comp. Endocrinol. 49,295-306.
Kohama, Y., Matsumoto, S., Oka, H., Teramoto, T., Okabe, M., and Mimura, T. (1988).
    Isolation of angiotensin-converting enzyme inhibitor from tuna muscle. Biochem.
    Biophys. Res. Commuri. 155,332-337.
Koller, K. I., Lowe, D. G., Bennett, G. L., Minamino, N., Kangawa, K., Matsuo, H., and
    Goeddel, D. V. (1991). Selective activation of the B natriuretic peptide receptor by
    C-type natriuretic peptide (CNP). Science 252, 120-123.
Lacy, E. R., and Reale, E. (1989). Granulated peripolar epithelial cells in the renal
    corpuscle of marine elasmobranch fish. Cell Tissue Res. 257,61-67.
Lacy, E. R., and Reale, E. (1990). The presence of a juxtaglomerular apparatus in
    elasmobranch fish. Anat. Embryol. 182,249-262.
Lagios, M. D. (1974).Granular eipthelioid (iuxtaglomerular) cell and renovascular mor-
    phology of the coelacanth Latimeria chalumnae Smith (Crossopterygii) compared
    with that of other fishes. Gen. Comp. Endocrinol. 22,296-307.
Lang, F., Volkl, H., and Haussinger, D. (1990). General principles in cell volume
    regulation. In “Cell Volume Regulation” (K. W. Beyenbach, ed.), pp. 1-25. Karger,
     Farmington, Connecticut.
Lange, R., and Fugelli, K. (1965).The osmotic adjustment in the euryhaline teleosts, the
     flounder, Pleuronectes Jesus I,. and the three-spined stickleback, Gasterosteus
    aculeatus L. Comp. Biochem. Physiol. 15,283-292.
Laurs, R. M., Ulevitch, R., and Morrison, D. C. (1978). Estimates ofblood volume in the
    albacore tuna. In “The Physiological Ecology of Tunas” (G. D. Sharp and A. E.
     Dizon, eds.), pp. 135-139. Academic Press, New York.
Lee, J . , and Malvin, R. L. (1987). Natriuretic response to homologous heart extract in
    aglomerular toadfish. Am. J . Physiol. 252, R1055-R1058.
Leray, C., Colin, D. A., and Florentz, A. (1981).Time course of osmotic adaptation and
     gill energetics of rainbow trout (Salmo gairdneri R.) following abrupt changes in
    external salinity.]. Conip. Physiol [Bll44, 175-181.
244                                                                KENNETH R . OLSON

Lipke, D. W., and Olson, K. R. (1988).Distribution of angiotensin-converting enzyme-
   like activity in vertebrate tissues. Physiol. Zool. 61,420-428.
Lipke, D. W., and Olson, K. R. (1990).Enzymes ofthe kallikrein-kinin system in rainbow
   trout. Am. J. Physiol. 258, R501-R506.
Lipke, D. W., and Olson, K. R. (1992). A specific inhibitor of mammalian kallikrein,
   Phe-Phe-Arg-chloromethyl ketone, inhibits the production of vasoactive substances
   from trout plasma by kallikrein and blocks endogenous kallikrein activity in trout
   gills. Fish Physiol. Biochem. (in press).
Lipke, D. W., Meisheri, K. D., and Olson, K. R. (1990a). Generation of vasoactive
   substances in trout plasma by trypsin and kallikrein. Am.J. Physiol. 258, R507-R514.
Lipke, D. W., Oparil, S., and Olson, K. R. (1990b).Vascular effects of kinins in trout and
   bradykinin metabolism by perfused gill. Am. J. Physiol. 258, R515-R522.
Lipke, D. W., Thomas, R. L., and Olson, K. R. (1987).Characterization of angiotensin-
    converting enzyme in the gills of rainbow trout, Salmo gairdneri (Richardson). Fish
    Physiol. Biochem. 3,91-97.
Lloyd, R., and Orr, L. D. (1969). The diuretic responses by rainbow trout to sub-lethal
    concentrations of ammonia. Water Res. 3,335-344.
Logan, A. G., and Morris, R. (1981).The handling ofmacromolecules by the kidney ofthe
   river lamprey, Lampetrafluviatilis. J. E x p . B i d . 93,303-316.
Loretz, C. A. (1979a). Osmotic and cell volume regulation in the goby, Gillichthys
    mirabilis. J. E x p . Zool. 210,237-244.
Loretz, C. A. (197913).   Some effects ofovine prolactation on body fluid composition in the
    cichlid teleost Sarotherodon mossambicus acclimated to seawater. Gen. Comp.
    Endocrinol. 38,38-42.
Lutz, P. L. (1972). Extracellular spaces and composition of various tissues of perch.
    Comp. Biochem. Physiol. [A] 4 1 , 7 7 4 8 .
Maack, T., Suzuki, M., Almeida, F. A,, Nussenzveig, D., Scarborough, R. M., McEnroe,
    G. A,, and Lewicki, J. A. (1987). Physiological role of silent receptors of atrial
    natriuretic factor. Science Wash. DC 238,675-678.
Madey, M. A,, Nakamura, Y., Nishimura, H., Cagen, L. M., and Barajas, L. (1984).Control
    of renin release in the aglomerular toadfish. Fed. Proc. 43,4628.
Madsen, S. S. (1990).The role ofcortisol and growth hormone in seawater adaptation and
    development of hypoosmoregulatory mechanisms in sea trout parr (Salmo trutta
    trutta). Gen. Comp. Endocrinol. 79, 1-11.
Maetz, J . , and Evans, D. H. (1972).Effects oftemperature on branchial sodium-exchange
    and extrusion mechanisms in the seawater-adapted founder Platichthys Jesus L.
   J. E x p . Biol. 56,565-585.
Malvin, R. L. (1984). Some comparative aspects of the renin-angiotensin system.
    Physiologist 27,367-373.
Malvin, R. L., and Vander, A. J. (1967). Plasma renin activity in marine teleosts and
    cetacea. Am. J. Physiol. 214, 187-191.
Malvin, R. L., Katz, S. A., Shier, D. N., and Sessler, F. M. (1984).Functional significance
    of multiple forms of renin. Clin. E x p . Hypertens. Part A Theory Pract. A6(10 & 111,
Malvin, R. L., Schiff, D., and Eiger, S . (1980).Angiotensin and drinking rates in the
    euryhaline killifish. Am. J. Physiol. 239, R31-R34.
Manning, R. D., Jr., and Guyton, A. C. (1982).Control ofblood volume. In “Rev. Physiol.
    Biochem. Pharmacol,” pp. 69-1 14. Springer-Verlag, Berlin.
Margolius, H. S. (1984).The kallikrein-kinin system and the kidney. Annu. Rec. Physiol.
3.   BLOOD AND EXTRACELLULAR FLUID VOLUME REGULATION                                 245

Margolius, H. S. (1989). Tissue kallikreins and kinins: Regulation and roles in hyperten-
   sive and diabetic diseases. Annu. Rev. Pharmacol. Toxicol. 29,343-364.
McCarthy, J . E., and Conte, F. P. (1966). Determination of the volume of vascular and
    extravascular fluids in the Pacific hagfish (Polistotrema stouti). Amer. Zool. 6,605.
McConnell, F. M., and Goldstein, L. (1990). Volume regulation in elasmobranch red
   blood cells. In “Cell Volume Regulation” (K. W. Beyenbach, ed.), pp. 114-131.
   Karger, Farmington, Connecticut.
MacDonald, R. J., Margolius, H. S., and Erdos, E. G . (1988). Molecular biology oftissue
   kallikrein. Biochem. J . 253,313-321.
Mcfarlane, N. A. A,, and Maetz, J. (1975). Acute response to salt load of the NaCl
   excretion mechanisms of the gill of Platichthysflesus in sea water. J. Comp. Physiol.
McVicar, A. J., and Rankin, J. C. (1983). Studies on the plasma colloid osmotic pressure of
   the river lamprey, Lampetrafluviatilis, L. J. E x p . Biol. 104,299-302.
Mearow, K. M., and Houston, A. H. (1980). Temperature-related changes in the content
    and distribution of water in the tissues of rainbow trout, Salmo gairdneri. Comp.
    Biochem. Physiol. [A] 67,117-121.
Milligan, C. L., and Wood, C. M. (1982). Disturbances in haematology, fluid volume
    distribution and circulatory function associated with low environmental pH in the
    rainbow trout, Salmo gairdneri. J. E x p . Biol. 99,397-415.
Milligan, C. L., and Wood, C. M. (1986a). Intracellular and extracellular acid-base status
    and H+ exchange with the environment after exhaustive exercise in the rainbow
    trout.]. E x p . Biol. 123,93-121.
Milligan, C. L., and Wood, C. M. (1986b). Tissue intracellular acid-base status and the
    fate of lactate after exhaustive exercise in the rainbow trout. J. E x p . Biol. 123,
Milligan, C. L., and Wood, C. M. (1987a). Effects of strenuous activity on intracellular
    and extracellular acid-base status and H+ exchange with the environment in the
    inactive, benthic starry flounder Platichthys stellatus. Physiol. Zool. 60,37-53.
Milligan, C. L., and Wood, C. M. (1987b). Muscle and liver intracellular acid-base and
    metabolite status after strenuous activity in the inactive, benthic starry flounder
    Platichthys stellatus. Physiol. Zool. 60,54-68.
Mizogami, S., Oguri, M., Sokabe, H., and Nishimura, H. (1968). Presence of renin in the
    glomerular and aglomerular kidney of marine teleosts. A m . J .Physiol. 215,991-994.
Moon, T. W. (1983). Changes in tissue ion contents and ultrastructure of food-deprived
    immature American eels, Anguilla rostrata.Can.J. Zool. 61,812-821.
Munger, R. S., Reid, S. D., and Wood, C. M. (1991). Extracellular fluid volume measure-
    ments in tissues ofthe rainbow trout (Oncorhynchus mykiss) in wiwo and their effects
    on intracellular pH and ion calculations. Fish Physiol. Biochem. 9,313-323.
Munshi, J. S. D., Pandey, P. K., and Mishra, N. (1975). Measurement ofblood volume in
    the Indian mud eel Amphipnous cuchia (Ham.) (Amphipnoidae; Pisces). Folia Hae-
    matol., Leipzig 102,438-447.
Murphy, P., and Houston, A. H. (1977). Temperature, photoperiod, and water-
    electrolyte balance in rainbow trout, Salmo gairdneri. Can. J . Zool. 55, 1377-1388.
Nagano, H., Hosaka, K., and Shukuya, R. (1976). The fastest migrating component of
    lamprey serum in zone electrophoresis. Comp. Biochem. Physiol. [ B ] 54,369-374.
Needleman, P. (1988). “Biological and Molecular Aspects of Atrial Factors.” Alan R.
    Liss, Inc., New York.
Needleman, P., Blaine, E. H., Greenwald, J. E., Michener, M. L., Saper, C. B.,
246                                                               KENNETH R. OLSON

    Stockmann, P. T., and Tolunay, H. E. (1989).The biochemical pharmacology ofatrial
   peptides. Annu. Reu. Pharm. Tox. 29,23-54.
Nichols, D. J . (1987).Fluid volumes in rainbow trout, Salmo gairdneri: Application of
   compartmental analysis. Comp. Biochem. Physiol. [A] 87, 703-709.
Nichols, D. J., Weisbart, M., and Quinn, J. (1985).Cortisol kinetics and fluid distribution
   in brook trout (Saloelinusfontinalis). J . Endocrinol. 107,57-69.
Nielsen, K. E., and Gesser, H. (1984).Energy metabolism and intracellular pH in trout
    heart muscle under anoxia and different [Ca’’]. J . Comp. Physiol. B 154,523-527.
Nikinmaa, M. (1982).Effects of adrenalin on red cell volume and concentration gradient
   of protons across the red cell membrane in the rainbow trout, Salmo gairdneri. M o l .
    Physiol. 2,287-297.
Nikinmaa, M., Soivio, A., and Railo, E. (1981).Blood volume of Salmo gairdneri: Influ-
   ence of ambient temperature. Comp. Biochem. Physiol. [ A ] 69,767-769.
Nishimura, H. (1980a). Evolution of the renin-angiotensin system. In “Evolution of
   Vertebrate Endocrine Systems” (P. K. T. Pang and A. Epple, eds.), pp. 373-403.
    Texas Tech Univ. Press, Lubbock.
Nishimura, H. (1980b). Comparative endocrinology of renin and angiotensin. In “The
    Renin-Angiotensin System” ( J . A. Johnson and R. R. Anderson, eds.), pp. 29-77.
    Plenum, New York.
Nishimura, H. (1985a). Endocrine control of renal handling of solutes and water in
    vertebrates. Renal Physiol. Busel 8,279-300.
Nishimura, H. (1985b). Evolution ofthe renin-angiotensin system and its role in control
    of cardiovascular function in fishes. Euol. Biol. Primitioe Fishes 103,275-293.
Nishimura, H. (1987). Role of the renin-angiotensin system on osmoregulation. Vert.
    Endocrinol. 2,157-187.
Nishimura, H., and Bailey, J. R. (1982). Intrarenal renin-angiotensin system in primitive
    vertebrates. Kidney Int. 22, S185-Sl92.
Nishimura, H., and Madey, M. A. (1989).Signals controlling renin release in aglomerular
    toadfish. Fish Physiol. Biochem. 7,323-329.
Nishimura, H., and Ogawa, M. (1973). The renin-angiotensin system in fishes. Amer.
    Zool. 13,823-838.
Nishimura, H., and Sawyer, W. H. (1976). Vasopressor, diuretic, and natriuretic re-
    sponses to angiotensins by the American eel, Anguilla rostruta. Gen. Comp. Endo-
    crinol. 29,337-348.
Nishimura, H., Lunde, L. G., and Zucker, A. (1979).Renin response to hemorrhage and
    hypotension in the aglomerular toadfish Opsunus tau. Am. /. Physiol. 237, H105-
Nishimura, H., Norton, V. M., and Bumpus, F. M. (1978).Lack of specific inhibition of
    angiotensin I1 in eels by angiotensin antagonists. A m . / . Physiol. 235, H95-Hl03.
Nishimura, H., Ogawa, M., and Sawyer, W. H. (1973). Renin-angiotensin system in
    primitive bony fishes and a holocephalian. Am. J . Physiol. 224,950-956.
Nishimura, H., Oguri, M., Ogawa, M . , Sokabe, H., and Imai, M. (1970).Absence of renin
    in kidneys of elasmobranchs and cyclostomes. Am. J . Physiol. 218,911-915.
Nishimura, H., Sawyer, W. H., and Nigrelli, R. F. (1976). Renin, cortisol and plasma
    volume in marine teleost fishes adapted to dilute media./. Endocrinol. 70,47-59.
Nolly, H. L., and Fasciolo, J. C. (1972). The renin-angiotensin system through the
    phylogenetic scale. Comp. Biochem. Physiol. 41A, 249-254.
Nustad, K., Johansen, L., Orstavik, T. B., and Pierce, J. V. (1980). Submandibular and
    kidney kallikreins: Biochemistry, origin and function. In “Enzymatic Release of
    Vasoactive Peptides” (F. Gross and H. G. Vogel, eds.), pp. 89-100. Raven Press,
    New York.
Ogawa, M . , and Sokabe, H. (1982).Hypocalcemic effect of homologous angiotensin-like
3. BLOOD AND EXTRACELLULAR FLUID VOLUME REGULATION                                      247

    substances produced by the renin-like enzyme in the corpuscles of Stannius in the
   Japanese eel Anguilla japonica. Gen. Comp. Endocrinol. 47,36-41.
Ogilvie, R. I. (1985).Comparative effects of vasodilator drugs on flow distribution and
   venous return. Can.J. Physiol. Pharmacol. 63,1345-1355.
Ogilvy, C. S., and Du Bois, .4. B. (1982).Effect of tilting on blood pressure and interstitial
    fluid pressures of bluefish and smooth dogfish. Am. J. Physiol. 242, R70-R76.
Ogilvy, C. S., Tremml, P. G., and DuBois, A. B. (1988). Pressor and hemodilution
    responses compensate for acute hemorrhage in bluefish. Comp. Biochem. Physiol.
   [A] 91,807-813.
O’Grady, S. M . (1989). Cyclic nucleotide-mediated effects of ANF and VIP on flounder
    intestinal ion transport. A m . ] . Physiol. 256, C142-C146.
O’Grady, S. M., Field, M., Nash, N . T., and Rao, M. C. (1985). Atrial natriuretic factor
   inhibits Na-K-CI cotransport in teleost intestine. Am. J . Physiol. 249, C531-C534.
Oguri, M. (1978). Presence of juxtaglomerular cells in the holocephalian kidney. Cen.
   Comp. Endocrinol. 36, 170-173.
Oguri, M. (1986).Absence ofjuxtaglomerular cells in the kidneys of freshwater stingrays
   and primitive bony fishes. Bull. Jpn. Soc. Sci. Fish. 52,1735-1737.
Okawara, Y., and Kobayashi, H. (1988). Enhancement of water intake by captopril
    (SQ14225),an angiotensin I-converting enzyme inhibitor, in the goldfish, Carassius
    auratus. Gen. Comp. Endocrinol. 69, 114-118.
Okawara, Y., Karakida, T., Aihara, M . , Yamaguchi, K., and Kobayashi, H. (1987).Involve-
    ment of angiotensin I1 in water intake in the Japanese eel, Anguillajaponica. Zool.
    Sci. 4,523-528.
Olson, K. R. (1984). Distribution of flow and plasma skimming in isolated perfused gills
    of three te1eosts.j. E x p . Biol. 109,97-108.
Olson, K. R., and Duff, D. W. (1992).Cardiovascular and renal effects of eel and rat atrial
    natriuretic peptide in rainbow trout. J. Comp. Physiol. (in press).
Olson, K. R., and Meisheri, K. D. (1989). Effects of atrial natriuretic factor on isolated
    arteries and perfused organs of trout. Am. J . Physiol. 256, R10-R18.
Olson, K. R., and Villa, J. (1991).Evidence against non-prostanoid endothelial-derived
    relaxing factor(s) in trout vessels. Am. J. Physiol. 260, R925-R933.
Olson, K. R., Duff, D. W., Farrell, A. P., Keen, J., Kellogg, M. D., Kullman, D., and Villa,
    J. (1991).Cardiovascular effects of endothelin in trout. A m . J . Physiol. 260, H1214-
Olson, K. R., Kullman, D., Narkates, A. J., and Oparil, S. (1986).Angiotensin extraction
    by trout tissues in uiuo and metabolism by the perfused gill. Am. J. Physiol. 250,
Olson, K. R., Lipke, D. W., Kullman, D., Evan, A. P., and Ryan, J. W. (1989).Localization
    of angiotensin-converting enzyme in the trout gill. J. E x p . Zool. 250, 109-115.
Olson, K. R., Lipke, D. W., Munshi,J. S. D., Moitra,A.,Ghosh,T. K., Kunwar, G.,Ahmad,
    M., Roy, P. K., Singh, 0.N., Nasar, S. S. T., Pandey, A., Oduleye, S. O., and Kullman,
    D. (1987). Angiotensin-converting enzyme in organs of air-breathing fish. Cen.
    Coinp. Endocrinol. 68,486-491.
Opdyke, D. F., and Holcombe, R. (1976). Response to angiotensins I and I1 and to
    AI-converting-enzyme inhibitor in a shark. Am. J. Physiol. 231, 1750-1753.
Opdyke, D. F., and Holcombe, R. F. (1978). Effect of angiotensins and epinephrine on
    vascular resistance of isolated dogfish gut. Am. J. Physiol. 234, R196-R200.
Opdyke, D. F., Carroll, R. G., Keller, N. E., and Taylor, A. A. (1981).Angiotensin I1
    releases catecholamines in dogfish. Comp. Biochem. Physiol. [ C l 7 0 , 131-134.
Opdyke, D. F., Cook, J. E., and Rausch, A. (1975).The vascular volume-pressure rela-
    tionship in the dogfish, Squalus acanthias. Comp. Biochem. Physiol. 51A, 431-437.
Opdyke, D. F., Wilde, D. W., and Holcombe, R. F. (1982).Effect of angiotensin I1 on
248                                                               KENNETH R. OLSON

    vascular resistance in whole-body perfused dogfish. Comp. Biochem. Physiol. [C]
Orice, G. G., and Castillo, G. A. (1986).A role for the kallikrein-kinin system in the
    regulation of osmotic permeability in the toad bladder. Comp. Biochem. Physiol. [C]
O’Toole, L. B., Armour, K. J., Decourt, C., Hazon, N., Lahlou, B., and Henderson, I. W.
    (1990). Secretory patterns of la-hydroxycorticosterone in the isolated perifused
    interrenal gland of the dogfish, Scyliorhinus canicula. J . Mol. Endocrinol. 5,55-60.
Pandey, B. N. (1977).Haematological studies in relation to environmental temperature
    and different periods of breeding cycle in an air breathing fish, Heteropneustes
   fossilis. Folia Haematol., Leipzig 104,69-74.
Pandey, B. N., Pandey, P. K., Choubey, B. J., and Munshi, J. S. D. (1976). Quantitative
    measurement of blood of an air-breathing siluroid fish, Heteropneustes fossilis
    (Bloch), in relation to body weight and respiratory surface areas. Matsya 2, 1-12.
Pandey, P. K., Moitra, A., Munshi, J . S. D., and Choubey, B. J. (1978). Role of L-
    thyroxine, hydrocortisone and progesterone in regulation of blood volume in a
    fresh-water air-breathing fish, Clarias batrachus (Linn.). Folia Haematol., Leipzig
Pang, P. K. T., Pang, R. K., Liu, V. K. Y., and Sokabe, H. (1981a).Effect of fish angioten-
    sins and angiotensin-like substances on killifish calcium regulation. Gen. Comp.
    Endocrinol. 43,292-298.
Pang, P. K. T., Yang, M. C. M., and Sokabe, H. (1981b).Effects of a converting enzyme
    inhibitor and angiotensin antagonists on the rat pressor effect of angiotensin-like
    substance formed by goosefish Stannius corpuscles and homologous plasma. Gen.
    Comp. Endocrinol. 45,402-405.
Parry, G . (1961).Osmotic and ionic change in blood and muscle of migrating salmonids.
    J. E x p . Biol. 38,411-427.
Pearson, M. P., and Stevens, E. D. (1991).Size and hematological impact of the splenic
    erythrocyte reservoir in rainbow trout, Oncorhynchus mykiss. Fish Physiol. Bio-
    chem. 9,39-50.
Perrier, H., and Perrier, C. (1978). L’analyse immunoelectrophoretique des proteines
    plasmatiques de la truite arc-en-ciel (Salmo gairdneri) et d’autres salmonides
    (Salmo trutta, Oncorhynchus kisutch). Ann. Znst. Michel Pacha 11, 15-20.
Perrier, H., Perrier, C., Peres, G., and Gras, J. (1977).The perchlorosoluble proteins of
    the serum of the rainbow trout (Salmo gairdneri Richardson): Albumin-like and
    hemoglobin binding fraction. Comp. Biochem. Physiol. [Bl57,325-327.
Perrott, M. N., and Balment, R. J. (1990).The renin-angiotensin system and the regula-
    tion of plasma cortisol in the flounder, Platichthys pesus. Gen. Comp. Endocrinol.
Perry, S. F., Laurent, D. J., and Booth, C. E. (1984).Absence of branchial edema in
    perfused heads of rainbow trout (Salmo gairdneri).J. E x p . Zool. 231,441-445.
Plaza-Yglesias, M., Laufer, M., and Herrera, F. C. (1988).Ionic and osmotic regulation in
    blood, aqueous humor, gills and retina in the euryhaline fish, Eugerres plumieri.
    Comp. Biochem. Physiol. [A] 89,377-382.
Polanco, M. J., Mata, M. I., Agapito, M. T., and Recio, J. M. (1990). Angiotensin-
    converting enzyme distribution and hypoxia response in mammal, bird, and fish.
    Gen. Comp. Endocrinol. 79,240-245.
Power, G. G., and Brace, R. A. (1983). Does the autonomic nervous system regulate
    whole-body lymph flow? Am. J . Physiol. 245, R695-R700.
Price, D. A,, Doble, K. E., Lee, T. D., Galli, S. M., Dunn, B. M., Parten, B., and Evans,
    D. H. (1990). The sequencing, synthesis, and biological actions of an ANP-like
3.   BLOOD A N D EXTRACELLUL.4R FLUID VOLUME REGULATION                               249

    peptide isolated from the brain of the killifish Fundulus heteroclitus. Biol. Bull. 178,
Primmett, D. R. N., Randall, D. J . , Mazeaud, M., and Boutilier, R. C . (1986).The role of
    catecholamines in erythrocyte pH regulation and oxygen transport in rainbow trout
    (Salmo gairdneri) during exercise. J . E x p . B i d . 122, 139-148.
Rabito, S. F., Binia, A,, and Segovia, R. (1972). Plasma kininogen content oftoads, fowl
    and reptiles. Comp. Biochem. Physiol. [A] 41,281-284.
Randall, D. J., and Daxboeck, C. (1982).Cardiovascular changes in the rainbow trout
    (Salmo gairdneri Richardson) during exercise. Can.J . Zool. 60,1135-1140.
Rasio, E. A., and Coresky, C. A. (1985).Passage of ions and dextran molecules across the
    rete mirabile of the eel. The effects of charge. Circ. Res. 57,74-83.
Reed, R. K., Bowen, B. D., and Bert, J. L. (1989). Microvascular exchange and interstitial
    volume regulation in the rat: Implication of the model. Am. J . Physiol. 257, H384-
Reinecke, M., Betzler, D., and Forssmann, W. G. (1987a). Immunocytochemistry of
    cardiac polypeptide hormones (cardiodilatin/atrial natriuretic polypeptide) in brain
    and hearts of Myxine ghtinosa (Cyclostomata). Histochemistry 86,233-239.
Reinecke, M., Betzler, D., Forssmann, W. G., Thorndyke, M., Askensten, U., and Falk-
    mer, S. (1987b).Electromicroscopical, immunohistochemical, immunocytochemical
    and biological evidence for the occurrence of cardiac hormones (ANP/CDD) in
    chondrichthyes. Histochemistry 87,531-538.
Reinecke, M., Nehls, M., and Forssmann, W. G. (1985).Phylogenetic aspects of cardiac
    hormones as revealed by immunocytochemistry, electronmicroscopy, and bioassay.
    Peptides 6,321-331.
Reinking, L. N. (1983).Aldosterone response to renin, angiotensin, ACTH, hemorrhage
    and sodium depletion in a freshwater teleost, Catostomus macrocheilus. Comp.
    Biochem. Physiol. [A] 74,873-880.
Rhodin, J . A. (1972). Fine structures of elasmobranch arteries, capillaries and veins in
    the spiny dogfish, Squalus acanthias. Comp. Biochem. Physiol. [A] 42,59-64.
Riegel, J . A. (1978).Factors affecting glomerular function in the pacific hagfish Eptatre-
    tus stouti (Lockington).J. E x p . Biol. 73,261-277.
Riegel, J . A. (1986). Hydrostatic pressures in glomeruli and renal vasculature of the
    hagfish, Eptatretus stouti. J. E x p . Biol. 123,359-371.
Robertson, J. D. (1974). Osmotic and ionic regulation in cyclostomes. In “Chemical
    Zoology” (M. Florkin and B. T. Scheer, eds.), pp. 149-191. Academic Press, New
Robertson, J. D. (1975). Osmotic constituents of the blood plasma and parietal muscle of
    Squalus acanthias L. Biol. Bull. 148,303-319.
Robertson, J . D. (1976). Chemical composition of the body fluids and muscle of the
    hagfish Myxine glutinosa and the rabbit-fish Chimaera monstrosa. J. Zool. Lond.
Robertson, J . D. (1984).The composition of blood plasma and parietal muscle of Oslo
    fjord eels [Anguilla anguilla (L.)] and the river lamprey [Lampetrafluuiatilis (L.)].
    Comp. Biochem. Physiol. [A] 77,431-439.
Robertson, J. D. (1986). The osmotic concentration of parietal muscle of the hagfish
    Myxine glutinosa L., including estimates of free and bound constituents. Comp.
    Biochem. Physiol. [A] 84,751-757.
Robertson, J . D. (1989).Osmotic constituents of the blood plasma and parietal muscle of
    Scyliorhinus canicula (L.). Comp. Biochem. Physiol. [A] 93,799-805.
Roman, R. J. (1986).Pressure diuresis mechanism in the control of renal function and
    arterial pressure. Fed. Proc. 45,2878-2884.
250                                                                 KENNETH R. OLSON

                    Venous system: Physiology of the capacitance vessels. In “Hand-
Rothe, C. F. (1983a).
   book of Physiology” ( J . T. Shepherd and F. M. Abboud, eds.), pp. 397-452.Ameri-
   can Physiological Society, Bethesda, Maryland.
Rothe, C. F. (1983b).Reflex control of veins and vascular capacitance. Physiol. Rec. 63,
Rothe, C. F. (1986).Physiology of venous return. An unappreciated boost to the heart.
   Arch. Intern. Med. 146,977-982.
Ryan, J. W. (1983).Assay of peptidase and protease enzymes in oivo. Biochem. Pharma-
   col. 32,2127-2137.
Samson, W. K., and Quirion, R. (1990).  “Atrial Natriuretic Peptides.” CRC Press, Boca
   Raton, Florida.
Sander, G. E., and Huggins, C. G. (1972).Vasoactive peptides. Annu. Rev. Pharmacoll2,
Satchell, G . H. (1986).     Cardiac function in the hagfish, Myxine (Myxinoidea : Cyclo-
    stomata). Acta Zool. 67, 115-122.
Satchell, G.H.(1991).      “Physiology and Form of Fish Circulation.” Cambridge Univer-
    sity Press, Cambridge.
Sawyer, W. H.(1970).        Vasopressor, diuretic, and natriuretic responses by lungfish to
    arginine vasotocin. Am. J. Physiol. 218, 1789-1794.
Sawyer, W. H., Blair-West, J. R., Simpson, P. A., and Sawyer, M. K. (1976).              Renal
    responses of Australian lungfish to vasotocin, angiotensin 11, and NaCl infusion. Am.
    J . Physiol. 231,593-602.
Schachter, M. (1980).Kallikreins (kininogenases)-A group of serine proteases with
    bioregulatory actions. Pharmacol. Rev. 31, 1-17.
Scheide, J. I., and Zadunaisky, J. A. (1988).  Effect of atriopeptin I1 on isolated opercular
    epithelium of Fundulus heteroclitus. Am. J . Physiol. 254, R27-R32.
Schiffman, R. H., and Fromm, P. 0.(1959).      Measurement of some physiological parame-
    ters in rainbow trout (Salmo gairdneri). Can.J . Zool. 37,25-32.
Schmidt-Nielsen, B. (1975).      Comparative physiology of cellular ion and volume regula-
    tion. J . Exp. Biol. 194,207-220.
Schmidt-Nielsen, B. (1977).      Volume regulation of muscle fibers in the killifish, Fundulus
    heteroclitus. J . E x p . Zool. 199,411-418.
Schmidt-Nielsen, B., Renfro, J. L., and Benos, D. (1972).        Estimation of extracellular
     space and intracellular ion concentrations in osmoconformers, hypo- and hyperos-
     moregulators. Bull. Mt. Desert Island Biological Laboratory 12,99-104.
Scicli, A. G., and Carretero, 0. A. (1986).  Renal kallikrein-kinin system. Kidney Int. 29,
    120- 130.
Seki, T., Miwa, I., Nakajima, T., and Erdos, E. G. (1973).    Plasma kallikrein-kinin system
    in non-mammalian blood: Evolutionary aspects. Am. J . Physiol. 224,1425-1430.
Shearer, K. D.(1984).      Changes in elemental composition of hatchery-reared rainbow
    trout, Salmo gairdneri, associated with growth and reproduction. Can. J. Fish.
    Aquat. Sci. 41,1592-1600.
Silva, P., Stoff, J. S., Solomon, R. J., Lear, S., Kniaz, D., Greger, R., and Epstein, F. H.
    (1987).  Atrial natriuretic peptide stimulates salt secretion by shark rectal gland by
    releasing VIP. Am. J. Physiol. 252, F99-F103.
Sinha, N. D. P., and Munshi, J. S. D. (1979).    Muscle fluid compartments and electrolyte
    distribution in relation to gonadal cycle in the fresh water teleost Channa punctatus.
    Indian J . Exp. Biol. 17,885-887.
Sinha, N. D.P., and Munshi, J. S. D. (1980).         Dimensions of body fluid spaces and
    electrolyte distribution in an amphibious fresh water teleost Channu punctatus
    Bloch. Indian J . E x p . Biol. 18,980-982.
3.   BLOOD A N D EXTRACELLULAR FLUID VOLUME REGULATION                                  25 1

Sinha, I . P., and Munshi, J. S. D. (1981).Intravascular components ofan air breathing
         i D.
    teleost Channa punctatus (B1och)-Their dimensions and responses to osmotic
    stress. lndian J . E x p . B i d . 19, 763-765.
Siret, J. R., Carmena, A. O., and Callejas, J. (1976). Erythrokinetic study in the fish
    manjuari (Atractosteus tristoechus). Comp. Biochem. Physiol. [A] 55, 127-128.
Sleet, R. B., and Weber, L. J . (19834.Water and electrolyte imbalances associated with
    laboratory manipulation of a marine teleost involve the gut. Can. J . Zool.61,1202-
Sleet, R. B., and Weber, L. J. (1983b). Blood volume of a marine teleost before and after
    arterial cannulation. Comp. Biochem. Physiol. [A] 76,791-794.
Smart, G. R. (1978). Investigations of the toxic mechanisms of ammonia to fish-gas
    exchange in rainbow trout (Salmo gairdneri) exposed to acutely lethal concentra-
    tions. J . Fish Biol. 12,93-104.
Smit, G. L., and Hattingh, J. (1979).Haematological studies on some freshwater teleosts.
    S. Afr. J . Anim. Sci. 9 , 6 5 6 8 .
Smith, L. S. (1966). Blood volumes of three salmonids. J. Fish. Res. Board Can. 23,
Sokabe, H. (1968). Comparative physiology of the renin-angiotensin system.J.l a p . Med.
    Assn. 59,502-512.
Sokabe, H. (1974). Phylogeny ofthe renal effects ofangiotensin. Kidney Znt. 6,263-271.
Sokabe, H., and Ogawa, M. (1974).Comparative studies of the juxtaglomerular appara-
    tus. Znt. Rev. Cytol. 37,271-327.
Sokabe, H . , Mizogami, S., and Sato, A. (1968).Role of renin in adaptation to sea water in
    euryhaline fishes. l a p . J. Pharmacol. 18,332-343.
Sokabe, H., Nishimura, H., Ogawa, M., and Oguri, M. (1970). Determination of renin in
    the corpuscles of Stannius of the teleost. Gen. Comp. Endocrinol. 14,510-516.
Sokabe, H., Ogawa, M., Oguri, M., and Nishimura, H. (1969). Evolution of the juxta-
    glomerular apparatus in the vertebrate kidneys. Texas Rep. Biol. Med. 27,867-885.
Sokabe, H., Oide, H., Ogawa, M., and Utida, S . (1973). Plasma renin activity in Japanese
    eels (Anguillajaponica) adapted to seawater or in dehydration. Gen. Comp. Endo-
    crinol. 12, 160-167.
Solomon, R. J., Dubey, A., Silva, P., and Epstein, F. H. (1988).Effect ofatrial natriuretic
    peptide on renal function in Squalus acanthias. Bull. Mt. Desert Island Biological
    Laboratory 27,18-21.
Solomon, R. J . , Protter, A., Epstein, F. H., and Silva, P. (1992). C-type natriuretic
    peptides stimulate chloride secretion in the rectal gland of Squalus acanthias. Am.J .
    Physiol. 262, R707-R711.
Solomon, R. J., Taylor, M . , Dorsey, D., Silva, P., and Epstein, F. H. (198Sa). Atriopeptin
    stimulation of rectal gland function in Squalus acanthias. Am. J . Physiol. 18, R348-
Solomon, R. J., Taylor, M., Sheth, S., Silva, P., and Epstein, F. H. (1985b).  Primary role of
    volume expansion in stirnulation of rectal gland function. Am.J. Physiol. 248, R638-
Spence, B., Meier, A. H., and Dietz, T. H. (1977).Circadian chloride rhythms in the gulf
    killifish, Fundulus grandis. Copeia 3,557-560.
Stevens, E. D. (1968).The effect ofexercise on the distribution ofblood to various organs
    in rainbow trout. Comp. Biochem. Physiol. 25,615-625.
Sulya, L. L., Box, B. E., and Gunter, G. (1961). Plasma proteins in the blood offishes from
    the Gulf of Mexico. Am. J . Physiol. 200, 152-154.
Suzuki, Y., and Hibiya, T. (1981a).Morphological changes and increased permeability of
    blood vessels in acute inflammation of fish. Bull. / p n . SOC. Sci. Fish. 47,323-327.
252                                                               KENNETH R. OLSON

Suzuki, Y., and Hibiya, T. (1981b). Effect of increased vascular permeability on tissue
   fluid protein composition in inflammation of fish. Bull. Jpn. Soc. Sci. Fish. 47,
Suzuki, R., Takahashi, A,, Hazon, N., and Takei, Y. (1991). Isolation of high-molecular-
   weight C-type natriuretic peptide from the heart of a cartilaginous fish (European
   dogfish, Scyliorhinus canicula). FEBS 282,321-325.
Takahashi, A., Kibota, J., Kawauchi, H., and Hirano, T. (1985). Effects of N-terminal
   peptide of salmon proopiocortin on interrenal function of the rainbow trout. Gen.
   Comp. Endocrinol. 58,328-335.
Takei, Y. (1988). Changes in blood volume after alteration of hydromineral balance in
   conscious eels, Anguilla japonica. Comp. Biochem. Physiol. 91A, 293-297.
Takei, Y., Hirano, T., and Kobayashi, H. (1979). Angiotensin and water intake in the
   Japanese eel, Anguilla japonica. Gen. Comp. Endocrinol.38,466-475.
Takei, Y., Okubo, J., and Yamaguchi, K. (1988). Effects of cellular dehydration on
   drinking and plasma angiotensin I1 level in the eel, Anguilla japonica. Zool. Sci. 5,
Takei, Y., Takahashi, A,, Watanabe, T. X., Nakajima, K., and Sakakibara, S. (1989).Amino
   acid sequence and relative biological activity of eel atrial natriuretic peptide. Bio-
   chem. Biophys. Res. Commun. 164,537-543.
Takei, Y., Takahashi, A,, Watanabe, T. X., Nakajima, K., Sakakibara, S., Takao, T., and
    Shimonishi, Y. (1990a). Amino acid sequence and relative biological activity of a
   natriuretic peptide isolated from eel brain. Biochem. Biophys. Res. Commun. 170,
Takei, Y., Tamaki, H., and Ando, K. (1990b).Identification and partial characterization of
    immunoreactive and bioactive atrial natriuretic peptide from eel heart. J . Comp.
    Physiol. [B] 160, 119-126.
Takemoto, Y., Nakajima, T., Hasegawa, Y., Watanabe, T. X., Sokabe, H., Kumangae, S.,
    and Sakakibara, S. (1983). Chemical structures of angiotensins formed by incubating
    plasma with the kidney and the corpuscles of Stannius in the salmon, Oncorhynchus
    Keta. Gen Comp. Endocrinol. 51,219-227.
Talbot, C., and Potts, W. T. W. (1989). Osmoregulation in immature Atlantic salmon
    (Salmo salar L.) following transfer from seawater to freshwater. Comp. Biochem.
    Physiol. [A], 92,235-239.
Talbot, C., Preston, T., and East, B. W. (1986). Body composition of Atlantic salmon
    (Salmo salar L.) studied by neutron activation analysis. Comp. Biochem. Physiol.
    85A, 445-450.
Taylor, A. A. (1977).Comparative physiology of the renin-angiotensin system. Fed. Proc.
Taylor, A. E. (1981). Capillary fluid filtration, Starling forces and lymph flow. Circ. Res.
Taylor, A. E., and Granger, D. N. (1984). Exchange ofmacromolecules across the micro-
    circulation. In “Handbook of Physiology. The Cardiovascular System” (E. M. Ren-
    kin and S. R. Michel, eds.), Vol. IV, pp. 467-520. Williams and Wilkins, Baltimore,
Thorson, T. (1958). Measurement of the fluid compartments of four species of marine
    Chondrichthyes. Physiol. Zool. 31, 16-23.
Thorson, T. (1959). Partitioning of the body water in sea lamprey. Science 130,99-100.
Thorson, T. (1961). The partitioning of body water in Osteichthyes: Phylogentic and
    ecological implications in aquatic vertebrates. Biol. Bull. 120,238-254.
Trippodo, N. C., Cole, F. E., Frohlich, E. D., and MacPhee, A. A. (1986). Atrial natri-
3. BLOOD AND EXTRACELLULAR FLUID VOLUME REGULATION                                       253

    uretic peptide decreases circulatory capacitance in areflexic rats. Circulation Re-
    search 59,291-296.
Turner, A. H. (1937). Serum protein measurements in the lower vertebrates. 11. In
    marine teleosts and elasmobranchs. Biol. Bull. 73,511.
Uemura, H., Naruse, M., Hirohama, T., Nakamura, S., Kasuya, Y.,and Aoto, T. (1990).
    Immunoreactive atrial natriuretic peptide in the fish heart and blood plasma exam-
    ined by electron microscopy, immunohistochemistry and radioimmunoassay. Cell
    Tissue Res. 260,235-247.
Vetter, S., and Hanke, W. (1985). A study comparing in vitro adrenal cortex responses of
    fish and amphibia. Acta Endocrinol. (Copenhagen) 108,157-161.
Vislie, T., and Fugelli, K. (1975). Cell volume regulation in flounder (Platichthyspesus)
    heart muscle accompanying an alteration in plasma osmolality. Comp. Biochem.
    Physiol. [A] 52,415-418.
Vogel, W. 0. (1985a). Systemic vascular anastomoses, primary and secondary vessels in
    fish, and the phylogeny of lymphatics. In “Cardiovascular shunts, Alfred Benzon
     Symposium 21” (K. Johansen and W. W. Burggren, eds.), pp 143-159. Munksgaard,
Vogel, W. 0. (1985b). The caudal heart of fishes: Not a lymph heart. Acta Anat. 121,
Vogel, W. O., and Claviez, M. (1981).Vascular specialization in fish, but no evidence for
    lymphatics. Z. Naturforsch. 36c, 490-492.
Walsh, P. J., and Moon, T. W. (1982). The influence of temperature on extracellular and
    intracellular pH in the American eel, Anguilla rostrata (LeSueur). Respir. Physiol.
Wardle, C. S. (1971). New observations on the lymph system of the plaice Pleuronectes
    platessa and other teleosts. J. Mar. Biol. Assn. U . K . 51,977-990.
Warner, M. C., and Williams, R. W. (1977). Comparison between serum values of pond
    and intensive raceway cultured channel catfish lctalurus punctatus (Rafinesque).
    J . Fish Biol. 11,385-391.
Weld, M. M., and Fryer, J. N. (1987). Stimulation by angiotensins I and I1 of ACTH
    release from goldfish pituitary cell columns. Gen. Comp. Endocrinol. 68, 19-27.
Weld, M. M., and Fryer, J. N. (1988). Angiotensin I1 stimulation ofteleost adrenocortico-
    tropic hormone release: Interactions with urotensin I and corticotropin-releasing
    factor. Gen. Comp. Endocrinol. 69,335-340.
Wells, R. M. G., and Forster, M. E. (1989). Dependence ofblood viscosity on haematocrit
    and shear rate in a primitive vertebrate. J . E r p . Biol. 145,483-487.
Westenfelder, C., Baranowski, R. L., Shiozawa, D. K., and Kablitz, C. (1988a). Atrial
    natriuretic factor (ANF) in two teleost fish Gila atraria (GA) and Salmo gairdneri
    (SG). Proc. Am. Soc. Neph. 142A. [Abstract]
Westenfelder, C., Birch, F. M., Baranowski, R. L., Rosenfeld, M. J., Shiozawa, D. K., and
    Kablitz, C. (198813).Atrial natriuretic factor and salt adaptation in the teleost fish Gila
    atraria. Am. J . Physiol. 255, Fl281-F1286.
White, F. C. (1988). Organ blood flow haemodynamics and metabolism of‘the albacore
    tuna Thunnus alalunga (Bonnaterre).E x p . Biol. 47,161-169.
Wiig, H., and Reed, R. K. (1987). Volume-pressure relationship (compliance)of intersti-
    tium in dog skin and muscle. Am. J . Physiol. 253, H291-H298.
Wilkes, P. R. H., and Mchlahon, B. R. (1986). Responses of a stenohaline freshwater
    teleost (Catostomus commersoni) to hypersaline exposure 11. Transepithelial flux of
    sodium, chloride and ‘acidic equivalents.’]. E z p . Biol. 121,95-113.
Wilkes, P. R. H., Walker, R. L., McDonald, D. G., and Wood, C. M. (1981). Respiratory,
254                                                               KENNETH R. OLSON

   ventilatory, acid-base and ionoregulatory physiology of the white sucker Catosto-
   mus commersoni: The influence of hyperoxia. J . E x p . Biol. 91,239-254.
Wilson, J. X. (1984a). Coevolution of the renin-angiotensin system and the nervous
   control of blood circulation. Can.J . Zool. 62, 137-147.
Wilson, J. X. (1984b). The renin-angiotensin system in nonmammalian vertebrates.
   Endocr. Rev. 5,45-61.
Wood, C. M., and Milligan, C. L. (1987).Adrenergic analysis of extracellular and intra-
    cellular lactate and H+ dynamics after strenuous exercise in the starry flounder
    Platichthys stellatus. Physiol. Zool. 60,69-81.
Wood, C. M., and Randall, D. J. (1971). The effect of anaemia on ion exchange in the
    southern flounder (Paralichthys lethostigma). Comp. Biochem. Physiol. [A] 39,391-
Wood, C. M., Turner, J. D., Munger, R. S., and Graham, M. S. (1990). Control of
    ventilation in the hypercapnic skate Raja ocellata. 11. Cerebrospinal fluid and intra-
    cellular pH in the brain and other tissues. Respir. Physiol. 80,279-298.
Wright, P. A., Randall, D. J., and Wood, C. M. (1988). The distribution of ammonia and
    H + between tissue compartments in lemon sole (Parophrys vetulus) at rest, during
    hypercapnia and following exercise. J . E x p . Biol. 136, 149-175.
Yamada, C., and Kobayashi, H. (1987).Immunoreactive angiotensin I1 in the corpuscles
    of Stannius of the rainbow trout, Salmo gairdneri. Zool. Sci. 4,387-390.
Yamamoto, K. (1987). Contraction of spleen in exercised cyprinid. Comp. Biochem.
    Physiol. [A] 87,1083-1087.
Yamamoto, K. (1988). Contraction of spleen in exercised freshwater teleost. Comp.
    Biochem. Physiol. [A] 89,65-66.
Yamamoto, K., and Itazawa, Y. (1989).Erythrocyte supply from the spleen of exercised
    carp. Comp. Biochem. Physiol. [A] 92, 139-144.
Yamamoto, K., Itazawa, T., and Kobayashi, H. (1980). Supply of erythrocytes into the
    circulating blood from the spleen of exercised fish. Comp. Biochem. Physiol. [A] 65,
Yamamoto, K., Itazawa, T., and Kobayashi, H. (1983).Erythrocyte supply from the spleen
    and hernoconcentration in hypoxic yellowtail. Mar. Biol. 73,221-226.
Yamamoto, K., Itazawa, Y., and Kobayashi, H. (1985).Direct observation offish spleen by
    an abdominal window method and its application to exercised and hypoxic yellow-
    tail. Jap./. Zchthyol. 31,427-433.
Yanagisawa, T., and Hashimoto, K. (1984).Properties and functions of yellowtail plasma
    albumins. B u 1 l . J p . SOC.Sci. Fish. 50,627-633.
Yokota, S. D., and Benyajati, S. (1986).Regulation ofglomerular filtration rate in a marine
    elasmobranch, the dogfish (Squalus acanthias). Bull. M t . Desert Island Biological
    Laboratory 26,87-90.
Yokota, S . D., and Benyajati, S. (1988).Hormonal control of renal function in the spiny
    dogfish (Squalus acanthias). FASEB]. 2, A949. [Abstract]
Youson, J . H., and Butler, D. G. (1988). Morphology of the kidney of adult bowfin, Amia
    calva, with emphasis on “renal chloride cells” in the tubule. J . Morphol. 196,
Zucker, A., and Nishimura, H. (1981). Renal responses to vasoactive hormones in the
    aglomerular toadfish, Opsanus tau. Gen. Comp. Endocrinol. 43, 1-9.

Department of Zoology
University of British Columbia
Vancouver, British Columbia, Canada

Department of Biology
University of Ottawa
Ottawa, Ontario, Canada

  I. Catecholamine Metabolism
     A. Catecholamine Biosynthesis
     B. Catecholamine Degradation
 11. Control of Blood Catecholamine Levels
     A. Circulating Levels during “Rest” and “Stress”
     B. Origin of Plasma Catecholamines
     C. Control of Catecholamine Release
     D. Fate of Plasma Catecholamines
111. Actions of Circulating Catecholamines
     A. Introduction
     B. Catecholamines and Gill Diffusing Capacity
     C. Catecholamines and Blood Oxygen Capacity
     D. Catecholamines and Gill Ventilation
     E. Catecholamines and Ion Movements
     F. Catecholamines and Blood Flow and Distribution
     G. Carbon Dioxide Transport
     H. General Conclusions Concerning the Action of Circulating Catecholamines
IV. Factors Influencing Actions of Catecholamines


    The biosynthesis and metabolic degradation of catecholamines in
fish, with few exceptions, appears to be essentially identical to the
metabolism of catecholamines in other vertebrate groups. There is,
FlSH PHYSIOLOGY, VOL. XIIB                           Copyright D 1992 by Academic Press, Inc.
                                                All rights of reproduction in any form reserved.
256                                       D. J. RANDALL A N D S. F. PEKKY

however, relatively little information on the metabolic pathways in
fish, and the data available are based on studies of very few species.
The topic of catecholamine metabolism in fish was reviewed by
Nilsson (1983).

A. Catecholamine Biosynthesis
    Catecholamines are synthesized in both nonneural chromaffin cells
and adrenergic neurons by identical processes. A single biosynthetic
pathway is responsible for the production of the three catecholamines,
dopamine, noradrenaline, and adrenaline. This series of enzymatic
reactions, known as the “Blaschko pathway” (Blaschko, 1939; Holtz,
1939), is summarized in Fig. 1. The first step is the hydroxylation of
tyrosine to dihydroxphenylalanine (DOPA) by tyrosine hydroxylase
(TH).This conversion, which occurs in the cytoplasm, is believed to be
the rate-limiting process in the synthesis of dopamine and noradrena-
line. The activity of T H is rapidly controlled, in part, via negative
feedback from the end products dopamine and noradrenaline (e.g.,
Spector et al., 1967). Thus, increases in catecholamine release will
cause an immediate increase in the formation of DOPA. In addition,
T H may be subject to long-term modulation in which additional en-
zyme molecules are induced slowly during chronic hypersecretion of
catecholamines (e.g., Lewander et al., 1977). For these reasons, cate-
cholamine stores are rarely depleted in chromaffin tissue even during
periods of intense release into the circulation. Molecular oxygen is a
co-factor in the hydroxylation reaction catalyzed by TH, thus the avail-
ability of oxygen may also limit catecholamine synthesis. Indeed, it
was demonstrated that long-term exposure of crucian carp (Carassius)
to anoxia markedly reduced the catecholamine content of head kidney
chromaffin tissue (Nilsson, 1989, 1990). Upon return to normoxia after
 17 days of anoxia (Nilsson, 1990), the noradrenaline levels recovered
only very slowly indicating that chronic lack of oxygen may damage the
catecholamine synthesizing system within the chromaffin tissue. The
pH and temperature optima of T H (see Table I) in chromaffin tissue
of Atlantic cod (Gadus morhua) are 6.0 and 3Oo-35”C, respectively
(Jonsson and Nilsson, 1983).
    The decarboxylation of DOPA to yield dopamine (DA) is catalyzed
b y the cytoplasmic enzyme aromatic L-amino acid decarboxylase
(AADC). Although the activity of this enzyme has not been directly
measured in fish, its presence in Atlantic cod is clearly indicated by the
increased levels of DA in a variety of peripheral tissues following
injections of DOPA (Johansson and Henning, 1981). In mammals,
4.CATECHOLAMINES                                                        257

   Fig. 1. The biosynthesis of catecholamines within a chromaffin cell by the
“Blaschko pathway.” Enzyme abbreviations: TH, tyrosine hydroxylase; AADC,
aromatic L-amino acid decarboxylase; DBH, dopamine-P-hydroxylase; P N M T ,
phenylethanolamine-N-methyl transferase.

AADC is distributed ubiquitously in both neural and nonneural tissue
and has a broad substrate specificity. Aromatic L-amino acid decar-
boxylase is available in sufficient quantities so that DOPA does not
accumulate to any great extent, and AADC has never been shown to be
rate-limiting in the production of catecholamines. The dopamine thus
formed is taken up into storage vesicles and is either stored as such for
258                                                D. J. RANDALL A N D S. F. PERRY

                                     Table I
     pH and Temperature Optima ofthe Blaschko Pathway Enzymes TH, DBH, and
       PNMT in Chromaffin Tissue of Atlantic Cod (Gadas rnorhua) and Dogfish
                               (Squalus acanthius)

                 PH           Temperature
Enzyme        optimum          optimum              Species                      Reference

TH            6.0             30-35°C             C. morhzra       Jonsson and Nilsson
DBH           5.4          27°C                   G. morhua        Jonsson and Nilsson
DBH           5.4-6.2         24.5-32°C           S. acanthius     Jonsson (1982)

PNMT          7.9             37°C                S. ucanthias         Abrahamsson (1979)
PNMT          -               30°C                C. morhua            Abrahamsson (1979)
                    ~     ~     _    _    _   _     _    _     _   _     _   ~     ~         ~

   Abbreviations: DBH, dopamine-p-hydroxylase; PNMT, phenylethano1amine-N-
methyl transferase; TH, tyrosine hydroxylase.

later release (e.g., dopaminergic neurons) or further metabolized (e.g.,
adrenergic neurons, chromaffin cells).
    The conversion of dopamine to L-noradrenaline occurs within stor-
age vesicles by a stereospecific hydroxylation by the enzyme do-
pamine-P-hydroxylase (DBH),which is a copper containing tetrameric
glycoprotein that requires ascorbate and molecular oxygen as co-
factors. This requirement of oxygen for noradrenaline synthesis may
also partially explain the drastic reductions in chromaffin tissue cate-
cholamine content in chronically anoxic carp (Nilsson, 1990; see ear-
lier). A portion of the vesicular DBH is bound to the vesicular matrix
while another portion is soluble and released with other vesicular
contents (catecholamines, adenosine triphosphate (ATP), chromo-
granin A; see Nilsson, 1983)on neural or other appropriate stimulation.
The pH and temperature optima for DBH are shown in Table I. The
pH optima of 5.4 and 5.4-6.2 for DBH from chromaffin tissue of Atlan-
tic cod (Jonsson and Nilsson, 1979) and dogfish (Jonsson, 1982), re-
spectively, may be physiologically significant because the internal pH
of isolated bovine chromaffin cell storage vesicles has been estimated
to be 5.5 (Johnson and Scarpa, 1976). The temperature optima for cod
and dogfish DBH appear somewhat lower than values reported for the
mammalian enzyme (Nilsson, 1983).
    The final step in the biosynthesis of adrenaline is the N -
methylation of noradrenaline catalyzed by the enzyme phenyletha-
nolamine-N-methyl transferase (PNMT) in the cytoplasm. It is unclear
how the noradrenaline formed within storage vesicles is made avail-
4. CATECHOLAMINES                                                   259

able to the cytoplasmic enzyme or how the adrenaline synthesized is
repackaged into vesicles. In mammals, the enzyme is specific for phe-
nylethylamines with the highest affinity for L-noradrenaline sug-
gesting that this catecholamine is the natural substrate. In fish, unlike
mammals, adrenaline is synthesized in both adrenergic neurons and
chromaffin cells, and thus both catecholamines can act as neuro-
transmitters. Phenylethanolarnine-N-methyltransferase activity has
been reported in a variety of fish species including rainbow trout
(Mazeaud, 1972), Atlantic cod (Abrahamsson and Nilsson, 1976; Abra-
hamsson, 1980), and dogfish (Abrahamsson, 1979).
    A comparison of the optimal catalytic activities of TH, DBH, and
PNMT from cod chromaffin tissue (cardinal vein) or coeliac ganglia
(PNMT, Abrahamsson and Nilsson, 1976; DBH, Jonsson and Nilsson,
1979; TH, Jonsson and Nilsson, 1983) indicates that the rate-limiting
step in the biosynthesis of adrenaline is the PNMT-catalyzed methyl-
ation of noradrenaline because the activity of PNMT is so much lower
than either T H or DBH. This view is further supported by the observa-
tion of accumulated [3H]noradrenaline in Atlantic cod after injection
of [3H]tyrosine (Jonsson and Nilsson, 1983). For these reasons, it is
likely that the ratio of adrenaline/noradrenaline in any catecholamine-
synthesizing tissue will be determined, in part, by the activity of
PNMT. In Atlantic cod, adrenaline is the predominant catecholamine
in chromaffin tissue and most adrenergically innervated organs with
the exception of the swim bladder and urinary bladder (Abrahamsson
and Nilsson, 1976). This is perhaps surprising given the relatively low
activities of PNMT and presumably indicates a low turnover rate of
    Although adrenaline may be formed from noradrenaline produced
locally in both neural and nonneural tissue, an additional route may
involve uptake of circulating noradrenaline and subsequent methyl-
ation. The importance of this pathway for adrenaline synthesis will
depend on the extent of blood flow to the particular organ.
    The synthesis of catecholamines in some vertebrate groups is under
hormonal control specifically by the stimulatory actions of adrenocor-
tical steroids on DBH and PNMT (Nilsson, 1983). In fish, little is
known about the endocrine control of catecholamine biosynthesis al-
though it would appear that in rainbow trout, PNMT is not modified by
corticosteroids (Mazeaud, 1972; Nilsson, 1983; Jonsson et al., 1983).
On the other hand, administration of cortisol was shown to signifi-
cantly increase the activity of chromaffin tissue DBH in rainbow trout
(Jonsson et al., 1983), presumably by reducing the extent of enzyme
260                                      D. J. RANDALL AND S. F. PERRY

    The biosynthesis of catecholamines can be blocked at several steps
by specific enzyme inhibitors (see Nilsson, 1983). The most widely
used compounds are similar to L-tyrosine in structure and are used to
inhibit TH. Enzyme inhibitors have been used only rarely and with
limited success in fish. Metcalfe and Butler (1989) administered
a-methyl-p-tyrosine, an inhibitor of T H (Spector et al., 1965), to dog-
fish (Scyliorhinus canicula) for 5 days in an attempt to prevent the
increase in circulating levels of catecholamines on exposure to envi-
ronmental hypoxia. This protocol was effective in lowering both the
“resting” and hypoxic levels of circulating catecholamines. However,
the hypoxic fish still released physiologically significant quantities of
catecholamines (see Table I in Metcalfe and Butler, 1989) into the
circulation so it is unclear whether or not this protocol can adequately
assess the importance of elevated catecholamines during such envi-
ronmental stresses. Furthermore, as discussed by Metcalfe and Butler
(1989),chronic treatment of dogfish with a-methyl-p-tyrosine substan-
tially altered oxygen consumption and blood oxygen status because of
specific effects on oxidative metabolism at the cellular level. Similar
usage of a-methyl-p-tyrosine in rainbow trout does not impair the
release of catecholamines during acute external hypoxia or physical
disturbance (S. F. Perry and R. Kinkead, unpublished observations).
    The methylation of noradrenaline to adrenaline can be blocked by
SK&F 64139, a specific inhibitor of PNMT (Pendleton et al., 1976).
This compound has been used in Atlantic cod to effectively deplete
sympathetic neurons of adrenaline (Abrahamsson, 1980).Owing to the
apparent low rate of adrenaline turnover, several days of treatment are
required to significantly deplete stores of adrenaline. This protocol,
although effective in reducing the adrenaline content of cate-
cholamine-synthesizing tissues, is not surprisingly associated with sig-
nificant increases in noradrenaline content (Abrahamsson, 1980).
    An alternate technique to deplete chromaffin tissue or adrenergic
neurons of endogenous catecholamines is to present a-methylated
precursors (e.g., a-methyltyrosine, a-methyldopa, a-methyldopamine)
that are recognized as substrates by the Blaschko pathway enzymes
and metabolized to inactive (or less active) “false transmitters” (see
Nilsson, 1983).

B. Catecholamine Degradation
   The two predominant enzymes involved in the catabolism of cate-
cholamines are monoamine oxidase (MAO) and catechol-O-methyl
transferase (COMT), and these produce inactive deaminated and 0-
4.   CATECHOLAMINES                                                  26 1

methylated catabolites, respectively. The mitochondrial enzyme,
MAO, is particularly important in neuronal degradation whereas
COMT may play a more important role in the catabolism of circulating
catecholamines because it is largely extraneural. The relative impor-
tance of deamination versus O-methylation for the degradation of
circulating catecholamines in fish is probably related to species
(Mazeaud and Mazeaud, 1973; Ungell and Nilsson, 1979, 1983; Nek-
vasil and Olson, 1986a,b; Colletti and Olson, 1988). In Atlantic cod
(Ungell and Nilsson, 1979), the major urinary catabolite of adrenaline
is metanephrine (MN) suggesting that O-methylation is the predomi-
nant route of degradation. In dogfish (Mazeaud and Mazeaud, 1973;
Ungell and Nilsson, 1983), the urinary catabolites of adrenaline are
MN, vanillylmandelic acid (VMA), and 3-methoxy-4-hydroxyphenyl-
glycol (MOPEG),indicating that both O-methylation and deamination
are involved. In rainbow trout (Nekvasil and Olson, 1986a; Colletti and
Olson, 1988),deamination by MA0 appears to be the more important
route of catecholamine inactivation. The relative importance of the two
enzymes cannot be determined with certainty because at least two
catabolites (VMA and MOPEG; see Fig. 2) are produced by the com-
bined effects of O-methylation/deamination (see Nekvasil and Olson,
    In trout (Nekvasil and Olson, 1986a) and cod (Ungell and Nilsson,
1979), O-methylated derivatives are the first to appear following injec-
tions of exogenous catecholamines. These results are consistent with
the established view (Kopin, 1960) that COMT acts before deamina-
tion. The temporal differences in the appearance of O-methylated
and deaminated catabolites may reflect the mitochondrial and cy-
toplasmic locations of MA0 and COMT, respectively.
    A variety of tissues exhibit MA0 and COMT activities (Edwards et
al., 1986; Nekvasil and Olson, 1986b; Colletti and Olson, 1988)includ-
ing gill, liver, and kidney. The relative importance of any particular
organ to the overall metabolic degradation of circulating catechol-
amines depends on the enzymatic activities, percentage of blood flow,
relative mass, and the ability to extract catecholamines from the
plasma. Based on the few available data, the gill is probably the domi-
nant site of plasma catecholamine degradation (Nekvasil and Olson,
 1986a,b; Colletti and Olson, 1988). The gill receives 100% of the
cardiac output, possesses both deaminating and O-methylating activi-
ties, and forms an extensive cell-plasma interface through which cate-
cholamines can be extracted. There are more catabolites released into
the arteriovenous (AV) circulation of the gill compared to the arterio-
arterial circulation (Colletti and Olson, 1988)indicating that cells per-

                                                        HOO 3

                                               P -       HOO                   " q9 - - -   -   h   -   Z

         Fig. 2. Metabolic dcgradittion of' adrenaline arid Iioratireiialiue by dearnination (MAO. monoamine oxidase) and
      O-methylation (COMT, catechol-0-methyl transfel-ase). Abbreviations: AD, adrenalirte: DOMA, 3,4-dihydroxy-
      mandelic acid; IIOPEC, 3,4-dihydroxyphenylglycol; MN, metanephrine; NA, n o r a d r e d i n e ; N M , noi-mctanephrine;
      MOPEG, 3 - 1 ~ 1 ~ ~ l i ~ ~ x y - 4 - 1 i ~glycol;xVMA,1vanill ylmandelic acid.
                                                   ~~~ y~l ~11v~
4. CATECHOLAMINES                                                     263

fused by the AV circulation are the major sites of catecholamine degra-
dation. Since chloride cells are most abundant on the gill filamental
epithelia (bathed by the AV circulation), these cells may be especially
important in this regard. Noradrenaline is the preferred substrate for
catabolism in both gill tissue homogenates (Nekvasil and Olson,
l986b) and perfused gill preparations (Colletti and Olson, 1988).
    The oxidative deamination of catecholamines requires molecular
oxygen. Thus, the oxygen status of the blood, in vivo or in vitro, will
affect the rate of catecholamine catabolism. Exposure of blood in vitro
to hyperoxia is a convenient technique to rapidly deplete endogenous
catecholamines. It is possible that hypoxic conditions may attenuate
the rate of degradation and in this way prolong the physiological ef-
fects of circulating catecholamines.
    The deamination of catecholamines by M A 0 can be blocked by a
class of compounds termed M A 0 inhibitors; these include pargyline
and nialamide (see Nilsson, 1983).Monoamine oxidase inhibitors have
been used in vitro to impede catecholamine degradation in whole
blood suspensions (Milligan and Wood, 1987), although the benefit is
questionable owing to the low activity of M A 0 in blood.


A. Circulating Levels during “Rest” and “Stress”
    The levels of plasma catecholamines under resting conditions and
after a variety of imposed stresses have been summarized previously
(Nilsson, 1983; Tetens et al., 1988; Hart et al., 1989; Perry et al., 1989;
Thomas and Perry, 1991; McDonald and Milligan, 1992; see also Fig.
3).In unanesthetized cannulated fish, the concentrations of adrenaline
or noradrenaline are usually between 1and 5 nmol.liter-l. The absolute
values and the ratios of adrenalinelnoradrenaline at rest vary among
and between species with no obvious pattern emerging. There is a
slight predominance of adrenaline in teleosts whereas noradrenaline
appears to be the dominant catecholamine in elasmobranchs. In gen-
eral, however, any one species does not appear markedly different
from any other. The eel may be an exception among all fish in that it has
unusually low levels of plasma catecholamines (Le Bras, 1982; Epple
et al., 1982; Epple and Nibbio, 1985; Hyde and Perry, 1990).
    The circulating levels of catecholamines rise during or immedi-
ately following a variety of physical and environmental disturbances
264                                                      D. J. RANDALL AND S. F. PERRY



         "   --  (1)   (2)
                                    =      -
                                 (4) (5) (6) (7) ( 8 ) (9) (10) (11) (12) (13)
     Fig. 3. Changes in the plasma levels of (A) adrenaline and ( B ) noradrenaline in fish
during environmental hypoxia. The data were obtained from the following studies
corresponding to the numbers in parentheses in the lower panel. 1.Thomas et (11. (1991);
2 . Metcalfe and Butler (1989); 3. Wright et a / . (1989); 4.Aota et a / . (1990); 5 . Boutilier et
al. (1988);6. Ristori and Laurent (1989); 7. Kinkead et al. (1991); 8. Thomas et a / . (1991);
9. Perry et al. (1991b); 10. Fievet et al. (1987); 11. Tetens and Christensen (1987);
12. Fritsche and Nilsson (1990); 13. Fievet et a1. (1990). See text for further details.
4.CATECHOLAMINES                                                     265

including external hypoxia (Fritsche and Nilsson, 1990; Perry et al.,
1991b; Thomas et al., 1991; Metcalfe and Butler, 1989; Aota et al., 1990;
Boutilier et al., 1988; Fievet et al., 1990; Ristori and Laurent, 1989;
Wright et al., 1989; Butler et al., 1979; Fievet et al., 1987; Kinkead et
al., 1991; Tetens and Christensen, 1987), hypercapnia (Hyde and
Perry, 1990; Perry et al., 1989; Perry et al., 1987), air exposure (Fuchs
and Albers, 1988; Nilsson et al., 1976),exhaustive or “violent” exercise
(Wood et al., 1990; Ristori and Laurent, 1985; Primmet et al., 1986;
Axelsson and Nilsson, 1986; Butler et al., 1986; Milligan and Wood,
1987; Tang and Boutilier, 1988; Wood, 1991; Opdyke et al., 1982;
Nakano and Tomlinson, 1967), metabolic acidosis (Boutilier et al.,
1986; Aota et al., 1990), anemia (Iwama et al., 1987; Perry et al., 1989),
and exposure to “softwater” (Perry et al., 1988a).There is considerable
disparity among the various studies that have evaluated plasma cate-
cholamine levels during stress owing, in part, to species and method-
ological differences. Adrenaline usually is the predominant catechol-
amine during stress. Although it is expected that the relative quantities
of adrenaline and noradrenaline in the plasma should reflect, at least
partially, the storage levels in the chromaffin cells of the head kidney
and elsewhere, this is only partly supported by the available data. In
Atlantic cod, the ratio of stored catecholamine within the head kidney
is 86% adrenaline to 14% noradrenaline (Abrahamsson and Nilsson,
1976), while in rainbow trout the levels of each catecholamine in the
chromaffin cells of the anterior kidney are approximately equal (Na-
kano and Tomlinson, 1967). The ratio of circulating adrenaline/
noradrenaline in trout and cod during hypoxia (e.g., cod, Fritsche and
Nilsson, 1990; Perry et al., 1991; trout, Fievet et al., 1990; Thomas et
al., 1991) or exhaustive exercise (e.g., trout, Primmett et al., 1986;
Wood et al., 1991; cod, Butler et al., 1989)do not appear to be different.
On the other hand, the chromaffin tissue of the axillary bodies of
dogfish (Squalus acanthias) contains predominantly noradrenaline
(83% noradrenaline/l7% adrenaline) (Abrahamsson, 1979). This may
explain the obvious predominance of noradrenaline in the circulation
of dogfish during hypoxia (Metcalfe and Butler, 1989; Butler et al.,
1978; Butler et al., 1979) or after exhaustive exercise (Opdyke et al.,
 1982) or intravascular K+ injection (Opdyke et al., 1983).
    Several other factors may potentially influence the levels of circu-
lating catecholamines during stress including the rate of metabolic
degradation (Nekvasil and Olson, 1986a,b; Colletti and Olson, 1988),
accumulation by neural and extraneural tissues (Nekvasil and Olson,
 1986a; Busacker and Chavin, 1977; Nekvasil and Olson, 1985; Ungell,
 1985a,b), and overflow from adrenergic neurons. Furthermore, dis-
266                                       D. J. RANDALL A N D S. F. PERRY

crepancies among the various studies may be related to the site of
blood withdrawal. Hathaway and Epple (1989) demonstrated regional
differences in the levels of plasma catecholamines with the highest
levels occurring near the presumed major site of release, the head
kidney. Since the gill is believed to be an important location of cate-
cholamine removal from the blood (Nekvasil and Olson, 1986a,b), pre-
versus postbranchial blood samples might be expected to vary consid-
erably. However, the only two studies to measure catecholamine lev-
els in pre- and postbranchial blood (eel, Hathaway and Epple, 1989;
cod, Kinkead et al., 1991)demonstrated essentially equivalent values.
Finally, the measured levels in the blood may be affected by the speed
with which the stress is imposed and the nature of the sampling pro-
tocol. Perry et d. (1991b) demonstrated two different patterns of cate-
cholamine release in Atlantic cod during hypoxia depending on how
rapidly the final water Po, (Pwo,) was attained. A gradual reduction of
PWO,   caused significant elevations of plasma adrenaline levels without
affecting noradrenaline levels while a more rapid induction of hypoxia
caused similar increases in the levels of both catecholamines. During
progressive stress, one catecholamine may be more important initially
while the other appears later. For example, during progressive hy-
poxia, noradrenaline appearance in the plasma may precede adrena-
line appearance by several minutes (Fritsche and Nilsson, 1990). In
contrast, dogfish preferentially release adrenaline into the blood dur-
ing brief stress, but if the stress persists, noradrenaline becomes the
more important plasma catecholamine (Abrahamsson, 1979). In other
instances, adrenaline and noradrenaline levels may initially increase
in the plasma at equal rates but noradrenaline may subsequently sta-
bilize while adrenaline levels continue to rise (Thomas et al., 1991). It
would appear that the chromaffin cells release variable quantities of
adrenaline and noradrenaline depending on the nature and duration of
the particular stimulus. Because of the rapid clearance of catechol-
amine from the blood by the combined effects of tissue accumulation
and metabolism (see following discussion), peak levels normally occur
within several minutes after cessation of the stress (e.g., Tang and
Boutilier, 1988; Wood et al., 1990). Less severe or prolonged stresses
may cause delayed increases in plasma catecholamine levels (Thomas
et al., 1991).Catecholamine levels may return to control values despite
the continuation of the stress that originally initiated the release (e.g.,
Perry et al., 1987).
    It is informative that plasma catecholamine levels remain essen-
tially unchanged during a variety of physical and/or environmental
disturbances that are usually classified as “stressful” including moder-
4.   CATECHOLAMINES                                                  267

ate or mild hypoxia (Boutilier et al., 1988; Ristori and Laurent, 1989;
Kinkead and Perry, 1990) and sustained aerobic exercise (Ristori and
Laurent, 1985; Butler et al., 1986; Hughes et al., 1988; see also review
by Wood, 1991). It is becoming increasingly clear that plasma catechol-
amine levels do not rise substantially until the degree of stress be-
comes very severe. Only at such times can circulating catecholamines
play a significant role in the initiation of adaptive responses.

B. Origin of Plasma Catecholamines
    The two potential sources of circulating catecholamines are “over-
flow” from adrenergic nerve terminals and release from chromaffin
cells. Few studies have directly assessed the importance of adrenergic
neural overflow. It has been argued (Wahlqvist and Nilsson, 1980;
Butler et al., 1989) that the inability to totally prevent the rise in
circulating catecholamines during stress after bilateral sectioning of
the nerves to the chromaffin tissue may reflect a minor component of
adrenergic neural overflow. On the other hand, Perry et al. (1991b)
concluded that “overflow” from peripheral adrenergic nerve terminals
did not significantly contribute to the elevation of circulating catechol-
amines in hypoxic cod since pretreatment with bretylium (an inhibitor
of catecholamine release from adrenergic nerve terminals) did not alter
the pattern of catecholamine release. Although it is unlikely that “over-
flow” of catecholamines significantly contributes to systemic changes
in the circulating levels, it is conceivable that local levels may change
in the region of densely adrenergically innervated organs or tissues.
Pennec and Le Bras (1984) showed that the perfused eel (Anguilla)
heart spontaneously releases catecholamines stored within adrenergic
neurons into the perfusion fluid. This “overflow” of neuronal catechol-
amines was enhanced by stimulation of the vagus nerve and signifi-
cantly affected cardiac function. The relative proportions of catechol-
amines released b y the eel heart closely matches the storage levels
(noradrenaline > adrenaline > dopamine). Significant quantities of
neuronal catecholamines also “overflow” from the perfused cod
spleen after electrical stimulation (Nilsson and Holmgren, 1976; Eh-
renstrom and Ungell, 1990). The predominant catecholamine released
by the cod spleen is adrenaline in accordance with the higher concen-
tration of this catecholamine stored within this tissue (Abrahamsson
and Nilsson, 1976). The physiological significance of catecholamine
“overflow” by the spleen has yet to be elucidated. Overflow from other
adrenergically innervated organs may also contribute slightly to circu-
lating levels or cause regional increases. In teleosts, adrenaline is the
268                                              D. J. RANDALL AND S. F. PERRY

dominant catecholamine in most adrenergic neurons (Abrahamsson
and Nilsson, 1976; Nilsson, 1983)while in elasmobranchs noradrena-
line appears to be more important (Abrahamsson, 1979).
    By far, the most important source of circulating catecholamines is
the chromaffin tissue (Nakano and Tomlinson, 1967; Abrahamsson and
Nilsson, 1976; Abrahamsson, 1979; Nilsson, 1983; Hathaway and Ep-
ple, 1989). I n teleosts, the chromaffin tissue is contained primarily
within the anterior or head kidney often in association with the walls of
the posterior cardinal veins (Nilsson, 1983). Table 11 summarizes the
levels of adrenaline and noradrenaline in the chromaffin tissue of
various teleost and elasmobranch species. It would appear that the
absolute levels of the catecholamine stored within the chromaffin
tissue-containing organs vary considerably among species as does the
ratio adrenaline/noradrenaline. It is unclear to what extent method-
ological differences have contributed to these variable patterns. Fur-
thermore, the values presented in Table I1 are expressed in terms of
tissue weight. Since the proportion of chromaffin cells varies in the
particular tissues, little can be inferred about the absolute values in the
chromaffin tissue itself. I n elasmobranchs, the axillary bodies are the
major sites of chromaffin tissue and unlike in teleosts, noradrenaline
                                   Table 11
  Catecholamine Levels in the Chromaffin Tissue of Teleosts and Elasmobranchs*

         Reference              Adrenaline      Noradrenaline         Site

  S. gairdneri                      4.7            4.5          Head kidney
    Nakano and
    Tomlinson (1967)
 A . rostrata                      84.8          42.4           PCB
     Hathaway and
     Epple (1989)
  G. morhua                        38.2           14.3          PCV
    Abrahamsson and
     Nilsson (1976)
  C . carassius                     0.05           0.16         Kidney
     Nilsson (1990)
  C . carpio                        0.05           0.84         Head kidney
     Stabrovskii (1968)
  S. acanthias                    445           2139            Axillary body
     Abrahamsson (1979)
  C . monstrosa                  3780           9390            Axillary body
     Pettersson and
     Nilson (1979)

      * All values are given as pg g-' tissue
4.   CATECHOLAMINES                                                               269

always is the prevalent catecholamine (Abrahamsson, 1979; Nilsson,
1983; see also Table 11). In cyclostomes and dipnoans, the heart is an
important site of chromaffin tissue (Nilsson, 1983).Depletion of car-
diac catecholamine stores reduces heart rate in hagfish but the exact
role of these catecholamine stores in cardiac function is not clear (see
Part A, Chapter 1).

C. Control of Catecholamine Release
   The chromaffin tissue of teleosts and elasmobranchs is innervated
by sympathetic preganglionic nerve fibers (Gannon and Campbell,
1972; Nilsson, 1976; Nilsson et al., 1976; Hathaway et al., 1989). These
fibers are evidently cholinergic because electrical stimulation (Nilsson
et al., 1976; Abrahamsson, 1979; Wahlqvist, 1981) or application of
acetylcholine (Ach) to simulate nerve activity (Nilsson et al., 1976;
Perry et al., 1991b) causes the release of both adrenaline and noradren-
aline from in situ perfused preparations; the cholinergic ganglionic
blocker hexamethonium abolishes the release of catecholamines after
nerve stimulation. The ratio of adrenaline/noradrenaline released
from perfused chromaffin tissue may deviate markedly from the pre-
dicted ratio based on the chromaffin tissue catecholamine content
(Abrahamsson, 1979; Perry et al., 1991b;see also Fig. 4). Moreover, the

                    0   1   2   3   4   5   6   7   8   9   1011121314
                                        Time (min)
     Fig. 4. Catecholamine overflow in an in situ, saline-perfused head kidney prepara-
tion of Atlantic cod (Gadus rnorhua) during a control period and immediately after
administration of 10-6M acetylcholine. Noradrenaline overflow is represented by the
open boxes, and adrenaline overflow is represented by the shaded boxes. * indicates a
statistically significant difference compared to the corresponding catecholamine over-
flow value immediately before addition of acetylcholine (6 min); t indicates a statisti-
cally significant difference compared to the overflow value of other catecholamines.
[From Perry et al. (1991b).]
270                                       D. J. RANDALL AND S. F. PERRY

proportions of catecholamines released may vary in relation to the
nature and duration of the stimulus. For example, Nilsson et al. (1976)
demonstrated variable adrenaline/noradrenaline overflow ratios in a
perfused head kidney preparation of Atlantic cod depending on
whether the preparation was stimulated electrically or with Ach. Simi-
larly, Abrahamsson (1979)reported that brief (1min) electrical stimu-
lation caused preferential adrenaline overflow in a perfused dogfish
preparation although noradrenaline clearly is the dominant circulating
catecholamine in elasmobranch plasma during sustained stress. These
results suggest that adrenaline and noradrenaline may be stored in
different chromaffin cell types, as demonstrated in amphibians
(Coupland, 1971). Furthermore, these cell types may respond differ-
ently to potential catecholamine-releasing stimuli.
    The importance of neural stimulation ofthe chromaffin tissue in the
mobilization of circulating catecholamines during stress has been eval-
uated by bilateral sectioning of spinal nerves 1-4 innervating the head
kidney. This procedure was shown to significantly, although not en-
tirely, decrease the increase of plasma catecholamines during air expo-
sure (Wahlqvist and Nilsson, 1980) or after exhaustive exercise (Butler
et al., 1989). Although it is obvious that neural stimulation of chromaf-
fin tissue and consequent release of Ach contributes to the elevation of
plasma catecholamines during periods of stress, the intermediary path-
ways are unknown. A variety of chemoreceptors exist in fish that moni-
tor the chemical or physical properties of the internal, external, or both
environments. These receptors have been implicated primarily in the
control of cardiovascular and ventilatory functions, although it is also
conceivable that they (or other similar receptors) may play a role in the
control of catecholamine release. If so, the sensory thresholds to elicit
release must b e considerably different than those eliciting cardiovas-
cular and ventilatory effects because the latter responses commence
with slight changes in external or internal chemistry, or both whereas
catecholamine release only begins under conditions of extreme stress.
    The inability of bilateral denervation to completely prevent cate-
cholamine release into the circulation or to substantially alter adrena-
line mobilization indicates that other factors are involved in the control
of catecholamine release from chromaffin tissue. This contention is
further supported by several experimental observations. First, the
catecholamine-induced (catecholaminotropic) release of catechola-
mines from chromaffin tissue of American eels (Anguilla rostrata) is
unaffected by removal of the preganglionic innervation (Hathaway
et al., 1989; Hathaway and Epple, 1989). Second, dogfish (Squalus
acanthias) liberate catecholamines into the circulation in response to
4. CATECHOLAMINES                                                     27 1

intravascular injections of potassium even after ganglionic blockade
using hexamethonium (Opdyke et al., 1983). This observation supports
the idea that increases in plasma K+ arising from skeletal muscle may
be an important secondary stimulus contributing to the release and
maintenance of plasma catecholamine after exhaustive exercise in
elasmobranchs (Opdyke et al., 1982).Third, the noninnervated cardio-
vascular chromaffin cells of the sea lamprey release catecholamines in
response to carbon dioxide (Dashow and Epple, 1985). Fourth, the
demonstration by Perry et al. (199lb) of a local direct stimulatory effect
of blood hypoxemia on adrenaline release from the chromaffin tissue in
an in situ, blood perfused head kidney preparation of Atlantic cod.
This observation of a specific modulatory effect of blood oxygen on the
release of catecholamine from the chromaffin tissue is consistent with
the general consensus that lowering of blood oxygen content is the
dominant factor initiating the release of catecholamines and also con-
sistent with the view that the principal effects of elevated circulating
catecholamine levels are related to enhancement of blood oxygen
transport (see following discussion).
     It was suggested (Perry et al., 1991b) that the release of either
stored catecholamine (adrenaline or noradrenaline) from the chromaf-
fin tissue could be controlled independently as a function of the in-
flowing (local) plasma levels of that particular catecholamine (Perry et
al., l99lb). The basis of this “negative feedback” phenomenon is that cate-
cholamine “overflow” from the chromaffin tissue into the circulation is
the net result of two opposing processes; clearance of inflowing cate-
cholamines and release of sequestered catecholamines. Clearance, in
turn, is the summated effect of reuptake, metabolism, and tissue bind-
ing. Thus, in the event of a prolonged catecholamine-releasing stimu-
lus, catecholamine net “overflow” from chromaffin tissue into the cir-
culation will stop when the inflowing plasma levels rise to a point
when clearance exceeds release. This mechanism is probably one of
several means (see later) by which plasma catecholamines are pre-
vented from rising to unnecessarily high levels and, in addition, allows
independent control of the plasma levels of each catecholamine.
     A common feature shared by the various stresses in which catechol-
amine levels are elevated is a requirement for enhanced oxygen trans-
port (see Table I in Thomas and Perry, 1991). This additional require-
ment may arise from the increased metabolic demands associated with
exercise or reductions in blood oxygen content caused by environmen-
tal hypoxia or internal acidosis. Since blood acid-base status (specifi-
cally red blood cell acid-base status) and oxygen content are interre-
lated, it is difficult to distinguish the potential independent effects of
272                                       D. J. RANDALL AND S. F. PERRY

acidosis and hypoxemia on promoting the release of catecholamines
from chromaffin tissue. There is no evidence, however, that acidosis
per se is a direct stimulus for catecholamine release or that it is a
prerequisite for catecholamine release during hypoxia. The release of
catecholamines during periods of internal acidosis, at least in rainbow
trout, is related to the lowering of blood oxygen content by the effects
of H + on hemoglobin-oxygen binding (Perry et al., 1989) and can be
prevented by exposing fish to hyperoxic water (Perry et al., 1989; Aota
et al., 1990).Acidotic conditions may augment catecholamine release
during hypoxemia because the extent of the reduction in oxygen con-
tent required to elicit catecholamine release is much larger during
anemia than during an acidosis.
    It is likely that the lowering of the blood oxygen content (or a
closely related variable), rather than a reduction of PO, per se, is the
proximate stimulus for catecholamine mobilization since blood PO, is
often unchanged or even elevated when catecholamines are released
during several stresses including anemia (Iwama et al., 1987; Perry et
al., 1989), hypercapnia (Perry et al., 1987), and exhaustive exercise
(Wood et al., 1990).Moreover, results of studies suggest that the prin-
cipal “zone” of catecholamine release corresponds to the area of maxi-
mal capacitance on the oxygen dissociation curve (Thomas et al., 1992;
S. F. Perry and S. D. Reid, unpublished observations). It would appear
that both inter- and intraspecific differences in the pattern of catechol-
amine release during hypoxia simply reflect intrinsic differences in the
properties of hemoglobin-oxygen binding. For example, rainbow
trout and American eel both release catecholamines during exposure
to environmental hypoxia but with widely different arterial P o , thresh-
olds (Fig. 5 ) corresponding to nearly identical arterial oxygen content
thresholds. In each case, the release of catecholamines is initiated at
arterial P o , values roughly equivalent to the hemoglobin P50 value as
determined from in vivo oxygen dissociation curves (S. F. Perry and
 S. D. Reid, unpublished observations; Fig. 5).

D. Fate of Plasma Catecholamines
   Catecholamines released into the circulation are rapidly cleared
from the plasma by the combined effects of tissue accumulation/
binding and metabolic degradation (see previous discussion). In rain-
bow trout, the biological half-time of an injected dose of adrenaline or
noradrenaline is less than 10 min (Nekvasil and Olson, 1986a). Indeed,
by 10 min after a bolus injection of catecholamine in trout, only 10% of
the injected dose is physiologically active (Nekvasil and Olson, 1986a).
4. CATECHOLAMINES                                                                      273



                  1 .o
                                                                           750   ?

                  0.8                                                      500

            z                                                                     s

                  0.4                                                      250

                  0.0                                                      0
                         0       20       40   60     80     100   120   140
                                               PaO, (torr)



             'M    1.o
                                 P    =   1l.lton




                  0.0             ..
                         0       20       40   60     80     100   120   146
                                               PaO, (torr)
    Fig. 5. The relationships between arterial Po, (Pa0,)  and plasma adrenaline levels
during acute external hypoxia in (A) rainbow trout (Salmo gairdneri) and (B) American
eel (Anguilla rostrata).In each case, in oioo hemoglobin oxygen dissociation curves are
shown. The shaded areas represent the zones of adrenaline release. For each species,
the release of adrenaline commences at 50-60% Hb-02 saturation. Note the expanded
scale ofthe adrenaline axis for A. rostrata.[From S. F. Perryand S. Reid (unpublished).]

Only a few studies have attempted to quantify the half-time of cate-
cholamines in the circulation of fish (Mazeaud, 1972; Ungell and
Nilsson, 1979), but they appear to be in general agreement with the
detailed study of Nekvasil and Olson (1986a). The relatively short
residence time of biologically active catecholamines in the circulation
274                                       D. J. RANDALL AND S . F. PERRY

corresponds with the brief physiological effects of catecholamines
following single injections of exogenous catecholamines or sudden
releases of endogenous catecholamines.
    The accumulation of catecholamines into tissues occurs by two
processes termed neuronal (type 1)or extraneuronal (type 2 ) uptake
(see Nilsson, 1983).Neuronal uptake refers to the absorption of cate-
cholamines by adrenergic nerve terminals after which the amines are
either metabolized or repackaged within the storage vesicles for sub-
sequent rerelease. The neuronal uptake mechanism can be blocked by
cocaine or desmethylimipramine (Nilsson, 1983).Extraneuronal up-
take refers to accumulation by nonneural tissues for subsequent meta-
bolic degradation. Extraneuronal uptake can be blocked by either
metanephrine or corticosterone (Nilsson, 1983).
    The importance of any particular tissue in the accumulation of
circulating catecholamines is determined by several factors including
its relative mass, blood flow, complement of catabolic enzymes, uptake
affinity, endogenous catecholamine levels, and density of adrenergic
neurons (neuronal uptake only). Thus, skeletal muscle, although rela-
tively inefficient at accumulating catecholamines (Ungell, 1985; Nek-
vasil and Olson, 1986a), may nevertheless play an important role in
inactivating circulating catecholamines owing to its enormous mass
(Nekvasil and Olson, 1986a). Conversely, the highly efficient chromaf-
fin tissue may be relatively unimportant because of its small size. With
the exception of the brain, all tissues that have been examined accumu-
late both catecholamines to varying degrees (Busacker and Chavin,
1977; Ungell, 1985a,b; Nekvasil and Olson, 1986a). The brain appar-
ently does not accumulate adrenaline (Busacker and Chavin, 1977;
Nekvasil and Olson, 1986a) and has an extremely low affinity for nor-
adrenaline accumulation (Nekvasil and Olson, 1986a). In general, tis-
sues preferentially accumulate noradrenaline regardless of the nature
of the uptake mechanism (neuronal versus extraneuronal; Busacker
and Chavin, 1977; Ungell, 1985a,b; Nekvasil and Olson, 1985,
     The gill tissue is presented with enormous levels of inflowing
catecholamines owing to its anatomical location. It is probably not
surprising, therefore, that the gill possesses highly efficient “neuronal-
like” and extraneuronal uptake mechanisms (Colletti and Olson,
 1988).Although the gill is innervated by adrenergic nerves (Donald,
 1984), these apparently are not the sites of neuronal-like accumulation
since autoradiographic studies (Nekvasil and Olson, 1985) have re-
vealed preferential accumulation/binding of noradrenaline by the
noninnervated pillar cells. A considerably greater quantity of catechol-
4.CATECHOLAMINES                                                       275

amines are extracted and metabolized as blood flows through the AV,
relative to the arterio-arterial circulation. Since the relative proportion
of the cardiac output flowing through the AV circulation of the gill is, in
part, controlled by circulating catecholamines, a model has been pro-
posed (Nekvasil and Olson, 1986b) in which inactivation of plasma
catecholamines is partially controlled by the resistance of the AV path-
way. In this scheme, the release of catecholamines will reduce blood
flow through the AV circuit (see Nilsson, 1984) and the rate of catechol-
amine inactivation, and hence it will assist in maintaining high circu-
lating systemic levels during periods of stress.


A. Introduction
    Changing levels of circulating catecholamines have numerous
physiological effects, both direct and indirect, all of which lead to
either increases in, or maintenance of, energy turnover and oxygen
supply under adverse conditions such as extreme hypoxia or acidosis.
The metabolic effects have been studied extensively in fish and in-
jected catecholamines, acting via p-adrenoreceptors, have been shown
to increase plasma glucose levels in trout (Wright et al., 1989) activat-
ing liver glycogenolysis, gluconeogenesis, or both and inhibiting gly-
colysis. Catecholamine injections had no effect on either muscle or
liver glycogen levels, and glycogen depletion seen during hypercap-
nia is related to factors other than adrenergic activation of liver glyco-
gen phosphorylase (Perry et al., 1988).Hypercapnia (Perry et al., 1988)
and hypoxia (Wright et al., 1989) depress glycogen phosphorylase
activity but increase pyruvate kinase activity leading to a reduction in
glucose levels. The effect of catecholamines, in both cases, was to
reverse this trend and maintain glucose availability. Catecholamines
also help to sustain oxidative metabolism in trout red blood cells
during acidotic states resulting from exhaustive exercise. This is
achieved in part by enhanced erythrocyte lactate oxidation (Wood et
al., 1990), an effect only seen under acidic conditions. Catecholamines
were without effect on this aspect of erythrocyte metabolism at normal
blood pH (Wood et al., 1990).
    Treatment of dogfish with a-methyl-tyrosine reduced circulating
levels of catecholamines and oxygen uptake in resting fish (Metcalfe
and Butler, 1989). They concluded that the reduction in oxygen con-
276                                         D. J . RANDALL A N D S. F. PERRY

sumption was due to a direct effect of the drug on cellular metabolism,
having observed that a-methyl-tyrosine depresses oxygen uptake in
isolated hepatocytes. It is also reported, however, that heart function in
vitro is dependent on low levels of catecholamines in the perfusate
(see Part A, Chapter l ) , so it is possible that the reduction in circulating
catecholamines following a-methyl-tyrosine treatment also contrib-
uted to the decrease in oxygen uptake.
    Catecholamines have a marked effect on oxygen delivery to the
tissues in teleosts, modulating changes in gill diffusing capacity, in-
creases in erythrocyte number, volume and intracellular pH, and
changes in blood flow (see Part A, Chapter 2) and breathing. In elasmo-
branchs infusion of catecholamines caused an increase in breathing in
quiet dogfish (Randall and Taylor, 1991) but inhibition of the rise in
circulating catecholamines associated with hypoxia, by treatment with
a-methyl-tyrosine, did not have any marked effect oxygen transfer or
breathing rate in dogfish exposed to moderate hypoxia (Metcalfe and
Butler, 1989).

B. Catecholamines and Gill Diffusing Capacity
    Increases in gill diffusing capacity probably result from elevations
in circulating catecholamines (Randall and Daxboeck, 1984) that occur
during hypoxia or following exhaustive exercise. An increase in circu-
lating catecholamines raises dorsal aortic blood pressure, which, in
turn, increases the width of the gill blood sheet and results in a more
even distribution of blood throughout the lamellae (Farrell et al.,
1980). In addition, there is probably an increase in the number of
lamellae perfused (Randall and Daxboeck, 1984), and it has been sug-
gested that adrenaline increases the permeability of the gill epithe-
lium to oxygen and other nonelectrolytes (Isaia et al., 1978). All of
these factors will increase the gill diffusing capacity. There is no direct
evidence for a role of catecholamines in augmentation of gill diffusing
capacity in intact animals, but Pettersson (1983)and Perry et al. (1985)
demonstrated that adrenaline enhanced gill oxygen diffusing capacity
in saline perfused cod and trout heads, respectively.

C. Catecholamines and Blood Oxygen Capacity
    Catecholamines have a marked effect on blood oxygen capacity via
a stimulation of Na+/H+ exchange across the erythrocyte membrane,
and the subsequent elevation of intracellular p H (Nikinmaa, 1990;
Motais et al., 1990) and via increases in hematocrit due to the splenic
release of red blood cells into the circulation (Nilsson, 1983).
4. CATECHOLAMINES                                                  277

    Many teleosts have a blood with a marked Root shift: That is, a
reduction in blood p H results in a decrease in hemoglobin oxygen
binding capacity. This is important for oxygen transfer into the
swimbladder as acidosis in the gas gland causes the release of oxygen
from hemoglobin for diffusion into the bladder. If the acidosis was
general, however, one might expect blood oxygen capacity to be re-
duced and oxygen delivery to other tissues impaired. This does not
occur as the release of catecholamines into the blood maintains eryth-
rocytic p H in the face of an acidosis, and no Root shift is observed
(Tufts and Randall, 1989).
    This increase in erythrocytic pH is caused by a P-adrenergic activa-
tion of Na+/H+transfer across the red blood cell membrane that results
in a disequilibrium of the proton gradient across the erythrocyte mem-
brane, raising intracellular p H (Nikinmaa, 1990).Adrenergic stimula-
tion of trout erythrocytes increases nucleotide triphosphate (NTP)use,
probably as a result of increased adenosine triphosphatase (ATPase)
activity and production of cyclic adenosine monophosphate (CAMP)
(Ferguson and Boutilier, 1989). Nucleotide triphosphate levels do not
fall under aerobic conditions because oxygen uptake and NTP produc-
tion is increased as well, to match use. Under anoxic conditions,
however, NTP levels fall following adrenergic stimulation, presum-
ably production no longer keeps pace with use (Ferguson and Bou-
tilier, 1989). Decreases in erythrocytic levels of NTP during hypoxia,
however, are not simply a consequence of adrenergic stimulation be-
cause ATP levels can fall during hypoxia in the absence of any change
in circulating catecholamines (Val, 1991). The erythrocytic membrane
is relatively impermeable to protons and so acid is transferred between
the plasma and the red blood cell via the Jacobs-Stewart cycle. The
increase in red blood cell pH is not immediately short-circuited by the
Jacobs-Stewart cycle, because the plasma carbon dioxide hydration/
dehydration reaction is uncatalyzed and the rate of proton transfer is
much faster than the plasma bicarbonate dehydration reaction velocity
(Forster and Steen, 1969; Motais et al., 1990; Nikinmaa et al., 1990).
This means that, unlike the situation in mammals, the capillary endo-
thelium in fish cannot contain carbonic anhydrase activity because if it
did the catecholamine-stimulated rise in erythrocytic p H would be
short-circuited and blood oxygen-carrying capacity would be reduced.
Thus, all carbon dioxide transfer must occur through the red blood cell,
as shown by Perry et al. (1982), and there must be a tight coupling of
oxygen uptake and carbon dioxide excretion. Hemoglobin oxygenation
supplies the protons for bicarbonate dehydration and, therefore, car-
bon dioxide excretion. Hemoglobin oxygenation not only supplies
protons for bicarbonate dehydration but also for Na+/H+ exchange
278                                                  D. J. RANDALL AND S. F. PERRY




     1.00~-    *
                           *                  1.oo
                               I-            0.90


                         iii          CI -

                     c 2            0 2

    Fig. 6. (A) The regulation of erythrocytic pH during a metabolic acidosis, following
adrenergic stimulation. The increased Na+/H+ transfer transports protons out of the
erythrocyte at a greater rate than the Jacobs-Stewart cycle replaces them, owing to the
absence of carbonic anhydrase (CA) in the plasma. Hb, hemoglobin.
4.   CATECHOLAMINES                                                                     279

(Motais et al., 1990) (Fig. 6). Thus adrenergic stimulation of the
sodium/proton antiporter may limit proton availability for bicarbonate
dehydration. Carbon dioxide excretion is maintained, however, even
though erythrocytic sodium/proton exchange is stimulated by cate-
cholamines (Steffensen et al., 1987).
    Catecholamines cause the release of erythrocytes from the spleen
into the blood stream. The splenic contraction causing this release is
mediated by stimulation of a-adrenoreceptors (Nilsson and Grove,
1974), first appears at low levels of catecholamines, and is dose-
dependent in hypercapnic trout (Perry and Kinkead, 1989).Catechol-
amines are probably involved in the recruitment of red blood cells
from the spleen during hypoxia (Yamamoto et al., 1985) and following
exhaustive exercise (Nikinmaa et al., 1984; Primmett et al., 1986).
Perry and Kinkead (1989) showed that splenic contraction was the
dominant response causing arterial blood oxygen content to increase
during hypercapnic induced elevations in circulating catecholamines.
During exhaustive exercise, however, reductions in plasma volume,
due to fluid shifts between blood and muscle, probably also contribute
to the increase in blood oxygen content.
    Thus, oxygen content of arterial blood is maintained or even ele-
vated under a variety of adverse conditions due to a rise in both the
number of circulating red blood cells and erythrocytic p H caused by
the rise in circulating catecholamines.

D. Catecholamines and Gill Ventilation
   Catecholamines infused into the dorsal aorta of eels caused a hyper-
ventilation in the summer animals but a hypoventilation in winter eels
(Peyreaud-Waitzeneggar, 1979). Hyperventilation was blocked by the
P-adrenergic antagonist propranolol whereas the hypoventilatory re-
sponses were inhibited by a-adrenergic blockers. Infusion into trout
resulted in a reduction in the breathing rate but an increase in
breathing amplitude and gill water flow (S. Aota, unpublished observa-
tions). In Amia, there was an increase in both rate and amplitude of
breathing following catecholamine infusion (McKenzie et al., 1991a).

      (B) Respiratory exchange ratio (R. E.) (i) prior to and following a burst swim and
(ii) prior to and following adrenaline infusion in fish at rest. [Modified from Steffenson et
al. (1987).] Bars show SD ( n = 6). Asterisks indicate values significantly different from
resting value ( P 5 0.025).(iii) Maintenance of COz excretion during adrenergic stimula-
tion of the erythrocyte. Protons essential for HCOS- dehydration are released from the
hemoglobin on oxygenation. [From Randall and Brauner (1991).]
280                                       D. J. RANDALL A N D S. F. PERRY

In dogfish catecholamine infusion resulted in a marked increase in
breathing rate but only in those fish with a low initial rate (Randall and
Taylor, 1991). If the dogfish were agitated and had a high initial
breathing rate, there was no effect of catecholamines infusion on
breathing rate. Catecholamine infusion into trout has been reported to
depress gill ventilation in normoxic (Kinkead and Perry, 1990; Playle
et al., 1990.), hypoxic, and hypercapnic trout (Kinkead and Perry,
1991). Kinkead et al. (1991) found that neither a-nor P-adrenoreceptor
agonists significantly impaired the hyperventilatory responses of cod
to hypoxia, despite an increase in circulating catechokmine levels.
Thus, there is evidence that catecholamine infusion has an effect on
breathing in fish, causing either an increase or a decrease in rate
depending on the species, the time of year, and the physiological state
of the animal. In some instances, for example in hypoxic cod (Kinkead
et al., 1991), elevated catecholamines may have little or no effect on
gill ventilation.
    The question remains: Is there a physiological role for catechol-
amines in the control of ventilation (see Randall and Taylor, 1992;
Perry et al., 1992)?Catecholamine levels in the blood increase follow-
ing anemia, hypoxia, and exhaustive exercise, and all these conditions
are associated with increases in ventilation. Changes in ventilation in
intact fish during moderate hypoxia (trout, Kinkead and Perry, 1990;
cod, Kinkead et al., 1991) and hyperoxia (trout, Kinkead and Perry,
1990),however, were not due to changes in circulating catecholamines
but rather to gill oxygen chemoreceptor activity (Burleson and Smat-
resk, 1990; Burleson, 1991). These chemoreceptors can be stimulated
by NaCN to increase breathing (Burleson and Smatresk, 1990). Dener-
vation of the gills of Amia resulted in a reduced ventilatory response
to hypoxia and the effects of NaCN are obliterated. The reduced
response in the denervated fish is similar to that seen following cate-
cholamine infusion in the normoxic fish. Hypoxia in Amia causes an
increase in circulating catecholamines and could account for the re-
sponse to hypoxia following denervation of the peripheral oxygen
chemoreceptors. Thus, although the hyperventilatory response to hy-
poxia in fish is largely driven by chemoreceptor stimulation due to
oxygen lack, there may be some additional or even supplementary
action of catecholamines in extreme hypoxia, when chemoreceptor
output has been reported to decline. It has been shown that proprano-
lo1 does not effect the initial increase in breathing but does impair the
sustained hyperventilatory response associated with hypercapnia in
trout (R. Kinkead and S. F. Perry, unpublished observations), hence
catecholamines may play some role in prolonged ventilatory responses
4.CATECHOLAMINES                                                     281

to hypercapnia. Catecholamines may stimulate breathing during ex-
haustive exercise and during anemia, but this has yet to be demon-
    If there is a role for circulating catecholamines in the control of
breathing how could they have an effect? In mammals, P-adrenergic
stimulation of the carotid body of rabbits (Milsom and Sadig, 1983),
lambs (Jansen et al., 1986),and cats (Mulligan et al., 1986) results in an
increase in ventilation. Denervation of the carotid body of lambs inhib-
its the ventilatory response to catecholamines. Catecholamines have
no effect on the carotid body activity but stimulate ventilation in goats
(Hudgel et al., 1986). Denervation of the carotid and aortic bodies of
cats does not inhibit the increase in ventilation following catechol-
amine infusion (Eldridge and Gill-Kumar, 1980). The increase in ven-
tilation could not be explained by the associated brain acidosis caused
by catecholamines (Eldridge et al., 1985).Thus it was concluded that
there was a central stimulation of breathing by catecholamines in some
mammals. I n others, peripheral stimulation of chemoreceptors by cate-
cholamines was of greater importance. In the lamb the blood brain
barrier appears to be complete and there is no central effect of circu-
lating catecholamines, hence only peripheral mechanisms can re-
spond to changes in circulating catecholamines. In cats perhaps circu-
lating catecholamines can cross the blood brain barrier and have a
central, as well as peripheral, action.
    The situation in fish is much less clear and is based on only a few
studies. Catecholamines have no direct effect on activity from trout gill
oxygen chemoreceptors (Burleson, 1991). Amia with denervated gill
chemoreceptors show no response to NaCN either peripherally (Mc-
Kenzie et al., 1991b) or centrally (Hedrick et d.,1991)but still retain a
ventilatory response to catecholamines. This rather fragmentary and
 sparse evidence indicates that catecholamines in fish do not stimulate
ventilation via action on peripheral chemoreceptors.
     Catecholamines have a marked effect on the circulation and
 changes in breathing could be secondary to changes in blood flow,
 pressure, or both. S. Aota (unpublished observations), however, found
that infusion of low doses of catecholamines (3-5 nM) resulted in a rise
 in blood pressure but no change in breathing, indicating that the two
responses were separate. At moderate (physiological) doses of cate-
cholamines there was a hyperventilation, as well as an increase in
blood pressure, whereas at high levels there was no effect on ventila-
tion but a very marked increase in blood pressure. Kinkead et al. (1991)
 observed no change in breathing associated with modulations of blood
 pressure during hypoxia and concluded that adrenergically medi-
282                                                 D. J. RANDALL AND S. F. PERRY

ated changes in blood pressure had no effect on breathing in the hy-
poxic cod.
    Hedrick et al. (1991)were unable to stimulate breathing by altering
the pH, oxygen, or carbon dioxide content of the extradural fluid of
Amia brains. They could find no evidence for central pH/C02 or oxy-
gen receptors. This is despite the fact that many investigators have
observed correlations between blood pH and ventilation (Janssen and
Randall, 1975). In many fish, however, acid infusion is associated with
a rise in circulating catecholamines. If hyperoxic fish are made acido-
tic, catecholamines are not released and ventilation is in fact reduced,
presumably due to the rise in blood oxygen content. Aota et al. (1990)
concluded that the increase in ventilation following acid infusion was
due to the action of catecholamines but changes in blood oxygen
content could also be involved. These experiments did not differenti-
ate between the effects of catecholamines and hypoxemia.
    Circulating noradrenaline can cross the blood brain barrier in fish
(Peyreaud-Waitzeneggar et al., 1980; Nekvasil and Olsen, 1986a) and
stimulate fictive ventilation when injected into the caudal vein or
directly into the fourth ventricle of the dogfish brain (Fig. 7) (Randall
and Taylor, 1991). Injections of noradrenaline close to the cell body of
a respiratory motorneurone in the dogfish medulla caused an increase
in its firing rate, which occurs within 3-5 sec of the injection
(Randall and Taylor, 1991). This was in contrast to much longer laten-


 &&+                                                           I I
                                                                         , , I
                                                                         I J I
      Adr Prop

                                                       1 .HI'  I         1 'M
                                        I S
      Adr Prop                                                           15 s
    Fig. 7. Effect of injection of adrenaline into the fourth ventricle of a dogfish (male
Squalus ucunthias, 1260) upon bursting efferent activity recorded from the central cut
end of the third branchial branch ofthe vagus. Injection of20 pl of lo-* mol . I-' adrena-
line (Adr), after a delay of about 100 sec, induced slow bursts of increased amplitude,
which progressively increased in rate. Repeated injections of adrenaline together with
propranolol (AdrProp) abolished the stimulatory response, and on the second injection
may have unmasked an a-adrenergic inhibition of efferent respiratory activity. [From
Randall and Taylor (1991).]
4.   CATECHOLAMINES                                                    283

cies following infusion into the blood or the fourth ventricle of the
brain, presumably because noradrenaline must first cross the blood
brain barrier before it can act on these central respiratory sites. The
short latency also indicates that the site of action is in the region of the
respiratory motoneurons. This central effect of catecholamines can be
blocked by infusion of propranolol into the blood, indicating that not
only is the effect via P-adrenoreceptors but that the site is accessible
from the blood.
   Thus, it seems possible that there is a central respiratory site sensi-
tive to catecholamines from the blood that may be involved in regulat-
ing gill ventilation in fish. The importance of this mechanism has been
debated, with Randall and Taylor (1992) arguing for and Perry et al.
arguing (1992) against physiological relevance.

E. Catecholamines and Ion Movements
    McDonald and Rogano (1986) showed that during a 2-h period of
epinephrine infusion, NaCl efflux was initially elevated by about 6-
fold above resting levels, declined to resting levels about 15 min after
the onset of infusion, and remained at that level for the rest of the
infusion period. Vermette and Perry (1987), also based on infusion
studies, concluded that catecholamines were involved in anion and
cation transfer across fish gills. Despite the fact that catecholamines are
involved in ion transfer across the gills, Vermette and Perry (1988)
could find no positive evidence that catecholamines played a role in
pH regulation during hypercapnic exposure in trout nor could Perry et
al. (1988a) find any evidence for a role for catecholamines in calcium
balance in fish exposed to low calcium water.
    Catecholamines may be involved in the retention of lactate in mus-
cle following exhaustive exercise in plaice (Wardle, 1978), but this has
not been confirmed in other fish (Wood and Milligan, 1987; Tang et al.,
    Catecholamines have a marked effect on erythrocytic pH due to
P-adrenoreceptor activation of NaIH exchange, as discussed earlier
(see Nikinmaa, 1990; Motais et al., 1990; for further discussion and

F. Catecholamines and Blood Flow
   and Distribution
   The role of circulating catecholamines on the cardiovascular sys-
tem has been reviewed by Nilsson (1983, 1984), and some aspects are
284                                      D . J. RANDALL AND S. F. PERKY

discussed in Part A, Chapter 1by Farrell and Jones. Adrenergic control
of the heart is largely neural, and it is only during stresses such as
exhaustive exercise (Primmett et al., 1986)or severe hypoxia (Boutilier
et al., 1988) that circulating levels may increase enough to have an
effect on the heart (Axelsson and Nilsson, 1986; Butler et al., 1989). A
neural innervation of the heart is present in most teleosts but absent in
elasmobranchs (Nilsson, 1983; Nilsson and Holmgren, Chapter 5).
Adrenergic stimulation generally elicits positive inotropic and chrono-
tropic effects on the heart, but under some conditions, in some species,
negative chronotropic effects have been observed (see Farrell and
Jones, Part A, for further discussion). There is some adrenergic tone to
the teleost heart, but this appears to be less important than cholinergic
tone during both rest and exercise (Axelsson and Nilsson, 1986;
Axelsson, 1988).
    The vasculature of teleosts, unlike that of elasmobranchs, is usually
well innervated by adrenergic nerves. Catecholamine infusion usually
evokes a rise in blood pressure and a vasoconstriction mediated by
a-adrenoreceptors. In addition, there is a pool of P-adrenoreceptors
that cause a vasodilation when stimulated (Nilsson, 1983). Vascular
resistance in the systemic circuit is modulated by adrenergic tone;
circulating catecholamines have an effect only under very adverse
conditions (Wood and Shelton, 1975; Smith et al., 1985; Axelsson,
1988). Exercise results in an increase in adrenergic tone to the heart
and the rise in blood pressure is modulated by neurally mediated
a-adrenergic vasoconstriction in the systemic circuit but a vasodilation
in the branchial circuit (Jones and Randall, 1978). Gut blood flow
increases following feeding but decreases during exercise (Axelsson et
al., 1989).These changes in gut blood flow are modulated, at least in
part, by adrenergic fibers innervating a-adrenoreceptors that cause

G. Carbon Dioxide Transport
    The pattern of carbon dioxide transport in blood and the mecha-
nisms of its excretion across the gill epithelium have been reviewed
(Perry, 1986; Perry and Wood, 1989; Perry and Laurent, 1990; Randall,
1990). As blood arrives at the gills, plasma HC03- is rapidly dehy-
drated by carbonic anhydrase within red blood cells to form molecular
COZ. The rate-limiting step in the overall conversion of plasma HC03-
to COZ is probably the entry of HC03- into the red blood cell (Perry et
al., 1982) in exchange for C1- via the band 3 anionic exchanger (Ro-
mano and Passow, 1984). The COe thus formed diffuses from the red
4.   CATECHOLAMINES                                                 285

cell into the plasma and across the gill lamellar epithelium into the
water according to existing COz partial pressure gradients. Relatively
few studies have addressed the potential impact of elevated circulating
catecholamines on COZ excretion or blood COZ transport. According to
theory, however, activation of the red blood cell Na+/H+exchanger by
catecholamines is expected to impair branchial carbon dioxide excre-
tion and markedly affect blood COz transport. A transient inhibition of
COz excretion is predicted for the following reasons. First, the passive
entry of HCO3- into the red cell must be reduced owing to a rise in the
intracellular levels of HC03-- associated with the abrupt alkalinization
of the cell interior (see previous discussion). This, in turn, would slow
the rate of formation of excretory COz. Second, the PC02 gradient
between the plasma and red blood cell must be temporarily reversed
owing to a reduction of intracellular PCOZ (see Thomas and Perry,
1991).At such times, the red blood cell is excluded from any role in net
C02 excretion. Third, the dehydration of HC03- to COZ within the red
blood cell must be temporarily “rate-limited” by the availability of H +

owing to the enormous consumption of H + by the Na+/H+ exchanger.
The extent of this reduction of bicarbonate dehydration will depend on
the relative rates of Nat/H+ exchange and the catalyzed bicarbonate
dehydration reaction during the early phases of oxygenation and sub-
sequently on the relative rates of Na+/H+transfer and Cl-/HC03- ex-
change. There would appear to be some adrenergically mediated ex-
pulsion of the protons liberated from the hemoglobin into the plasma
because there is a large, but slow, increase in blood PaCOz after
exhaustive exercise (Perry and Wood, 1989)that is reduced by infusion
of carbonic anhydrase into the blood.
    The notion of adrenergic inhibition of COz excretion remains con-
troversial with considerable empirical evidence both for and against
the hypothesis as originally proposed by Wood and Perry (1985). Di-
rect measurements of COz excretion in vivo (Steffensen et al., 1987;
Playle et al., 1990) or in vitro (Tufts et al., 1988) failed to provide
evidence in favor of adrenergic impairment. On the other hand, in
vitro studies (Perry et al., 1991a; Wood and Perry, 1991)clearly demon-
strate an inhibitory effect of catecholamines on red blood cell COz
excretion and establish that the underlying mechanism is related to
adrenergic activation of the Na+/H+ exchanger. A difference between
the in vitro and in vivo studies is the absence of proton production as a
result of hemoglobin oxygenation that occurs in vivo simultaneously
with bicarbonate dehydration. This probably explains the differences
between the in vitro and in vivo results.
    Elevation of circulating catecholamines markedly affects blood car-
286                                                         D. J. RANDALL A N D S. F. PERRY

bon dioxide transport by the concerted actions of ( a )inhibition of COZ
excretion and ( b )titration of plasma HC03- by Hf extruded by the red
blood cell Na+/H+ exchanger. Both of these effects contribute to an
elevation in blood Pco, during intravascular infusion of exogenous, or
release of endogenous, catecholamines (Perry and Vermette, 1987;
Vermette and Perry, 1988; Perry and Thomas, 1991) that cannot be
otherwise explained by ventilatory adjustments (see Fig. 8). It has

                     2.6   -A
                     2.4   -
                     2.2   -
             E       2.0   -
                     1.8   -
             0-      1.6 -PaCO, = -0.016(Pa02)           + 3.564
                     1.4 - r = 0.83 (PeO.01)
                           n = 3061 observations
                     1.2 -
                     1.0   '

                                 20    40
                                           I         I







                                (15   2o       25   30

                                                     I      I           I        I        I

                           0     20    40           60     80      100      120      140

                                                    PaO, (tom)
    Fig. 8. P , O , - P , ~ o , diagrams obtained from continuous recordings of arterial blood
respiratory status of rainbow trout using an extracorporeal loop. Injections of exogenous
adrenaline during normoxia and hypoxia (inset) or the release of endogenous catechol-
amines during hypoxia cause pronounced deviations from the predicted P,~,-P,~o, rela-
tionship shown in (A) and reproduced in (B). In each case, the elevation of circulating
catecholamines is associated with a rise of arterial PC02 without an appropriate accom-
panying change in arterial Po,. These data indicate that the origin of the adrenergic
respiratory acidosis is unrelated to ventilatory or other branchial adjustments but likely
reflects activation of red blood cell Na+/H+exchange. See text for further details. [Data
from S. F. Perry and S. Thomas (unpublished observations).]
4. CATECHOLAMINES                                                   287

been suggested (e.g. Wood and Perry, 1985; Perry and Wood, 1989)
that the rise in arterial blood PCO, associated with the recovery from
exhaustive exercise in fish may partially result from the effects of
catecholamines on blood COZ transport and excretion. The functional
significance of the catecholamine-induced effects on arterial blood
COZ transport is unknown although it has been proposed (Perry and
Wood, 1989) that the associated respiratory acidosis may assist in the
stimulation of ventilation during recovery from exhaustive exercise.

H. General Conclusions Concerning the Action of
   Circulating Catecholamines
    Catecholamines may be involved in regulating gill water flow, gill
diffusing capacity, blood oxygen content, and blood flow and distribu-
tion, as well as energy turnover. The metabolic effects appear to be to
maintain function in the face of adverse conditions rather than to
increase metabolism. The same can be said to be true for both cardiac
function (see Part A, Chapter 1)and oxygen transfer. The evidence is
implied rather than direct, increasing catecholamine levels have a
marked effect on each ofthese factors. In a study on dogfish in which an
increase in circulating catecholamines during hypoxia was inhibited,
no marked effect on oxygen transfer was observed (Metcalfe and But-
ler, 1989). Randall et al. (1987), however, observed that fish swim as
well after exhaustive exercise as before, even though they are acidotic,
whereas fish in acidic water do not swim as well as in neutral water (Ye
and Randall, 1991). Fish in acid water become acidotic but do not
release catecholamines, whereas catecholamines are released after
exhaustive exercise and the fish can maintain oxygen delivery to the


    Fish exhibit large intra- and interspecific differences in both the
nature of catecholamine mobilization in response to stress and the
responsiveness of target tissues to the catecholamines. These differ-
ences are related to diurnal and seasonal cycles, interactive effects of
other hormones, acclimation history, and the prevailing internal respi-
ratory/acid-base status.
    The quantities of catecholamines stored within tissues and circu-
lating in the plasma vary diurnally and seasonally both under resting
288                                      D. J. RANDALL AND S. F. PERRY

conditions and in response to stress. At rest, diurnal or daily fluctua-
tions (Boehkle et al., 1967; Le Bras, 1984; Ehrenstrom and Johansson,
1987)are probably more important than seasonal variations (Van Dijk
and Wood, 1988; Milligan et al., 1989; Temma et al., 1990). Indeed,
there is little agreement on the influence of season on resting plasma
catecholamine levels; for example, the unusually high values in rain-
bow trout plasma reported by Van Dijk and Wood (1988) were attrib-
uted to “winter” acclimation yet Milligan et al. (1989)were unable to
demonstrate any effects of seasonality on resting levels in trout. Simi-
larly, the effects of season on stress-induced release of catecholamines
are poorly understood. Milligan et al. (1989) observed an obvious reduc-
tion in the circulating levels of noradrenaline yet an increase in the
levels of adrenaline associated with exhaustive exercise in winter
acclimated rainbow trout. On the other hand, Van Dijk and Wood
(1988) measured extremely high values of both catecholamines after
exhaustive exercise that, again, were attributed to winter acclimation.
    Seasonality may influence the pattern of catecholamine release in
response to environmental disturbances owing to temperature related
effects on metabolism and blood oxygen transport because the internal
oxygen status (or a closely related variable) is a key factor controlling
catecholamine release (see previous discussion). Thus, the water PO,
threshold for catecholamine release during hypoxia is expected to be
directly proportional to temperature from the combined effects of in-
creased metabolic rate and decreased affinity of hemoglobin-oxygen
binding. This idea has not been tested directly, although a comparison
of two studies on hypoxic trout (Pwo, = 60 torr) performed at 12°C
(Thomas et al., 1991) or 4°C (S. D. Reid and S. F. Perry, unpublished
observations) reveals that significantly greater quantities of catechol-
amines were released at the warmer water temperature (see Fig. 9).
    Diurnal, seasonal, or long-term elevations of plasma catechol-
amines resulting from chronic stress may significantly affect the
adrenergic responsiveness of target tissues by desensitization or
  down regulation” of adrenoreceptors (Lefkowitz et al., 1990) al-
though this has yet to be clearly demonstrated in fish. An often cited
example of seasonal effects on adrenergic function is the variable
responsiveness of the teleost red blood cell Na+/H+ exchanger to
catecholamines (Nikinmaa and Jensen, 1986; Van Dijk and Wood,
 1988;Cossins and Kilbey, 1989).Although not always obvious (Tetens
et al., 1988; Milligan et al., 1989),it would appear that the red blood
cells of winter-acclimated fish are considerably less responsive to
adrenergic stimulation than summer-acclimated fish. The reasons for
these seasonal differences remain uncertain but may involve changes
4. CATECHOLAMINES                                                                            289

                            A       "Winter" fish

                                                                   0    Normoxia
                200                                                o    Hypoxia


                   0    l       5     10   15   20   25       30       35    40    45

                            B       "Summer" fish
                 400 -                                   T                  Normoxia

                            0   5     10   15   20       25   30       35     40   45   50

                                                 Time (h)
    Fig. 9. The effects of long-term (48 h) exposure to moderate environmental hypoxia
(Pwo2 = 60 torr, open symbols) or normoxia (closed symbols) on plasma catecholamine
(adrenaline and noradrenaline) levels in rainbow trout acclimated to (A) winter condi-
tions (temperature = 12°C) or (B) summer conditions (temperature = 4°C). All values
shown are means 2 1 SEM. [Data from Thomas et al. (1991)and S. G . Reid and S. F. Perry
(unpublished observations).]

in cell surface P-adrenoceptor numbers, affinities, or both (Marttila and
Nikinmaa, 1988).
    Changes in the surface population of red blood cell P-adrenocep-
tors may also partially explain the enhanced adrenergic responsive-
ness during acute hypoxia (Motais et al., 1987; Marttila and Nikinmaa,
290                                             1 . J. RANDALL A N D S. F. PERRY

1988; Fuchs and Albers, 1988; Salama and Nikinmaa, 1990; Reid and
Perry, 1991). In both trout (Reid and Perry, 1991) and carp (Marttila
and Nikinmaa, 1988),short-term exposure of red blood cells to severe
hypoxia facilitates the recruitment of internal cytomplasmic p-adre-
noceptors (Reid et al., 1991) to the cell surface where they become
functionally coupled to adenylate cyclase. Interestingly, chronic expo-
sure of fish to hypoxia appears to desensitize red blood cell adrenergic
responsiveness (Thomas et al., 1991) and is perhaps related to down
regulation of P-adrenoceptors.
    The sensitivity of teleost red blood cells to catecholamines is also
related to the extracellular acid-base status with increasing sensitivity
at lower p H values (Nikinmaa et al., 1987; Borgese et al., 1987; Cossins
and Kilbey, 1989).
    There is also a marked influence of season on the cardiovascular
(e.g. Part et al., 1982) and ventilatory responses (Peyreaud-Waitzeneg-
ger et al., 1980) to catecholamines. Specifically, in “summer”-
acclimated fish, P-adrenoceptor mediated responses are dominant
while in winter-acclimated fish, a-adrenoceptor mediated responses
appear to be relatively more important. The reason(s) for such seasonal
switches in adrenoceptor dominance is unknown but may involve
alterations in the proportions of adrenoceptor subtypes present in
particular target tissues. The heart from a cold-acclimated rainbow
trout shows a greater sensitivity to adrenaline than that from a warm-
acclimated fish. This increase in sensitivity is probably due to the
twofold increase in P-receptors in the sarcolemma with no change in
the total P-receptor population in hearts from cold-acclimated fish
(Keen, 1992).
    Little is known about the interactive effects of noncatecholamine
hormones on adrenergic function. The glucocorticoid, cortisol,
however, has been shown to enhance the responsiveness of trout red
blood cells (Reid and Perry, 1991) and hepatocytes (Reid et al., 1991)to
catecholamines specifically by increasing the abundance of cell sur-
face P-adrenoceptors. It is possible that chronic stress and the associ-
ated elevation of plasma cortisol levels may increase the ability of fish
to physiologically adapt to any subsequent acute stresses, at least with
respect to the adrenergic stress responses.


Abrahamsson, T. (1979). Phenylethanolamine-N-methyl    transferase (PNMT) activity
   and catecholamine storage and release from chromaffin tissue of the spiny dogfish,
   Squalus acanthias. Comp. Biochem. Physiol. C 64, 169-172.
4. CATECHOLAMINES                                                                       29 1

Abrahamsson, T. (1980). The effect of SK and F 64139, an inhibitor of' phenyletha-
    nolamine-N-methyl transferase (PNMT), on adrenaline and noradrenaline content
    in sympathetic neurons of the cod Gadus morhua. Comp. Biochem. Physiol. C 67,
Abrahamsson, T., and Nilsson, S. (1976). Phenylethanolamine-N-methyltransferase
    (PNMT) activity and catecholamine content in chromaffin tissue and sympathetic
    neurons in the cod, Gadus morhua. Acta Physiol. Scand. 96,94-99.
Aota, S., Holmgren, K. D., Gallaugher, P., and Randall, D. J . (1990).A possible role for
    catecholamines in the ventilatory responses associated with internal acidosis or
    external hypoxia in rainbow trout Oncorhynchus mykiss. J . E x p . Biol. 151,57-70.
Axelsson, M . (1988).The importance of nervous and humoral mechanisms in the control
    of cardiac performance in the Atlantic cod Gadus morhua at rest and during nonex-
    haustive exercise. J . E x p . Biol. 137,287-303.
Axelsson, M., and Nilsson, S. (1986). Blood pressure control during exercise in the
    Atlantic cod, Gadus morhua. J. E x p . Biol. 126,225-236.
Axelsson, M., Driedzic, W. R., Farrell, A. P., and Nilsson, S. (1989).Regulation ofcardiac
    output and gut blood flow in the sea raven, Hemitripterus americanus. Fish Physiol.
    Biochem. 6,315-326.
Blaschko, H. (1939). The specific action ofL-DOPA decarboxylase.]. Physiol. (Lond.)        96,
Boehkle, K. W., Tiemeier, 0. W., and Elephteriou, B. E. (1967). Diurnal rhythm in
    plasma epinephrine and norepinephrine in the channel catfish (Ictalurus puncta-
     tus). Gen. Comp. Endocrinol. 8, 189-192.
Borgese, F., Garcia-Romeu, F., and Motais, R. (1987). Ion movements and volume
    changes induced by catecholamines in erythrocytes of rainbow trout: Effect of pH.
    J. Physiol. (Lond.)382, 145-157.
Boutilier, R. G., Iwama, G. K., and Randall, D. J . (1986).The promotion ofcatecholamine
    release in rainbow trout, Salmo gairdneri, by acute acidosis: Interaction between
    red cell pH and haemoglobin oxygen carrying-capacity.]. E x p . Biol. 123,145-157.
Boutilier, R. G., Dobson, G. P., Hoeger, U., and Randall, D. J . (1988).Acute exposure to
    graded levels of hypoxia in rainbow trout (Salmo gairdneri): Metabolic and respira-
    tory adaptations. Respir. Physiol. 71,69-82.
Burleson, M. L., and Smatresk, N. J. (1990). Effects of sectioning cranial nerves IX and X
     on cardiovascular and ventilatory reflex responses to hypoxia and NaCN in channel
    catfish. J. E x p . Biol. 154,407-420.
Burleson, M. L. (1991). Oxygen-sensitive chemoreceptors and cardiovascular and venti-
     latory control in rainbow trout. Ph.D. Thesis, University of British Columbia.
Busacker, G. P., and Chavin, W. (1977). Uptake, distribution, and turnover of catechol-
    amine radiolabel in the goldfish, Carassius auratus L. Can. J . Zool. 55, 1656-
Butler, P. J., Axelsson, XI., Ehrenstrom, F., Metcalfe, J. D., and Nilsson, S. (1989).
    Circulating catecholamines and swimming performance in the Atlantic cod, Gadus
    morhua. J . E x p . Biol. 141,377-387.
Butler, P. J . , Metcalfe, J. D.,and Ginley, S . A. (1986).Plasma catecholamines in the lesser
    spotted dogfish and rainbow trout at rest and during different levels of exercise.
   /. E x p . Biol. 123,409-421.
Butler, P. J . , Taylor, E. W., Capra, M. F., and Davison, W. (1978).The effect of hypoxia on
    the levels of circulating catecholamines in the dogfish Scyliohinus canicula. J .
    Comp. Physiol. 127,325-330.
Butler, P. J., Taylor, E. W., and Davison, W. (1979). The effect of long term, moderate
    hypoxia on acid-base balance, plasma catecholamines and possible anaerobic end
292                                                  D. J. RANDALL AND S. F. PERRY

    products in the unrestrained dogfish Scyliorhinus canicula. J. Comp. Physiol. 132,
Colletti, A. E., and Olson, K. R. (1988). Catecholamine metabolism by the perfused
    rainbow trout gil1.J. Exp. Zool. 248, 177-184.
Cossins, A. R., and Kilbey, V. K. (1989). The seasonal modulation of Na+/H+exchanger
    activity in trout erythrocytes. J. E x p . Biol. 144,463-478.
Coupland, R. E. (1971). Observations on the form and size distribution of chromaffin
    granules and on the identity of adrenaline- and noradrenaline-storing chromaffin
    cells in vertebrates and man. Mem. Soc. Endocr. 19,611-635.
Dashow, L., and Epple, A. (1985). Plasma catecholamines in the lamprey: Intrinsic
    cardiovascular messengers? Comp. Biochem. Physiol. C 82,119-122.
Donald, J . (1984). Adrenergic innervation of the gills ofbrown and rainbow trout, Salmo
    trutta and S. gairdneri. J. Morphol. 182,307-316.
Edwards, D., Hall, T. R., and Brown, J. A. (1986). The characteristics and distribution of
    monoamine oxidase (MAO) activity in different tissues of the rainbow trout, Salmo
    gairdneri. Comp. Biochem. Physiol. C 84,73-77.
Ehrenstrom, F., and Johansson, P. (1987).Circadian rhythms and contents ofcatechols in
    different brain structures, peripheral organs and plasma of the Atlantic cod, Gadus
    morhua. Comp. Biochem. Physiol. C 87(1), 193-202.
Ehrenstrom, F., and Ungell, A. L. (1990).Nerve impulse-induced release of endogenous
    noradrenaline and adrenaline from the perfused cod spleen. J. Comp. Physiol. B 160,
Eldridge, F. L., and Gill-Kumar, P. (1980).Mechanisms ofhyperpnea induced by isopro-
    terenol. Resp. Physiol. 40,349-363.
Eldridge, F. L., Kiley, J. P., and Milhorn, D. E. (1985). Mechanisms of respiratory
    response to isoproterenol in glomectomized cats. J. Appl. Physiol. 58,83-88.
Epple, A., and Nibbio, B. (1985). Catecholaminotropic effects of catecholamines in a
    teleost fish, Anguilla rostrata. J. Comp. Physiol. B 155,285-290.
Epple, A., Vogel, W. H., and Nibbio, B. (1982).Catecholamines in head and body of eels
    and rats. Comp. Biochem. Physiol. C 71, 115-118.
Farrell, A. P., Sobin, S. S., Randall, D. J., and Crosby, S. (1980).Intralamellar blood flow
    patterns in fish gills. Am. J. Physiol. 239, R429-R436.
Ferguson, R. A., and Boutilier, R. G. (1989). Metabolic-membrane coupling in red blood
    cells of trout: The effects of anoxia and adrenergic stimulation. J. E x p . Biol. 143,
     149- 164.
Fievet, B., Caroff, J., and Motais, R. (1990). Catecholamine release controlled by blood
    oxygen tension during deep hypoxia in trout: Effect on red blood cell Na/H ex-
    changer activity. Respir. Physiol. 79,81-90.
Fievet, B., Motais, R., and Thomas, S. (1987). Role of adrenergic-dependent H+ release
    from red cells in acidosis induced by hypoxia in trout. Am. J. Physiol. 252, R269-
Forster, R. E., and Steen, J. B. (1969). The rate ofthe “Root shift” in eel red cells and eel
    haemoglobin solutions. J. Physiol. Lond. 204,259-282.
Fritsche, R., and Nilsson, S. (1990). Autonomic nervous control of blood pressure and
    heart rate during hypoxia in the cod, Gadus morhua. J. Comp. Physiol. B 160,
Fuchs, D. A., and Albers, C. (1988).Effect of adrenaline and blood gas conditions on red
    cell volume and intraerythrocytic electrolytes in the carp, Cyprinus carpio. J. E r p .
    B i d . 137,457-477.
Cannon, B. J., and Campbell, G. D. (1972).Monoamine storage in relation to cardiac
4. CATECHOLAMINES                                                                     293

    regulation in port jackson shark Heterodontus portusjacksoni. Z . Zellforsch Mik-
    rosk. Anat. 134,437-450.
Hart, B. B., Stanford, G. G . , Ziegler, M. G., Lake, C. R., and Chernow, B. (1989).
    Catecholamines: Study of interspecies variation. Crit. Care Med. 17, 1203-1222.
Hathaway, C. B., and Epple, A. (1989).The sources of plasma catecholamines in the
    American eel, Anguilla rostrata. Gen. Comp. Endocrinol. 74,418-430.
Hathaway, C. B., Brinn, J . E., and Epple, A. (1989).Catecholamine release by catechol-
    amines in the eel does not require the presence of brain or anterior spinal cord.
    J. E x p . Zool. 249,338-342.
Hedrick, M. S., Burleson, M. L., Jones, D. R., and Milsom, W. K. (1991).An examination
    of central chemosensitivity in an air-breathing fish (Amia calua).J . E x p . Biol. 155,
Holtz, P. (1939).Dopadecarboxylase. Naturwissenschaften 43,724-725.
Hudgel, D. W., Kressin, N. A., Nielsen, A. M., and Bisgard, G . E. (1986). Role of
    beta-adrenergic receptors in carotid body function of the goat. Respir. Physiol. 64,
    203-2 11.
Hughes, G . M., Le Bras-Pennec, Y., and Pennec, J.-P, (1988). Relationships between
    swimming speed, oxygen consumption, plasma catecholamines and heart perfor-
    mance in rainbow trout ( S . gairdneri R.). J. E x p . Biol. 48,45-49.
Hyde, D. A., and Perry, S. F. (1990). Absence of red cell pH and oxygen content
    regulation in American eel (Anguilla rostrata) during hypercapnic acidosis in ciao
    and in uitro. J. Comp. Physiol. B 159,687-693.
Isaia, J . , Maetz, J., and Haywood, G. P. (1978). Effects of epinephrine on branchial
    nonelectrolyte permeability in trout. J . E x p . Biol. 74, 227-237.
Iwama, G . K., Boutilier, R. G., Heming, T. A., Randall, D. J., and Mazeaud, M. (1987).
    The effects of altering gill water flow on gas transfer in rainbow trout. Can.J. Zool.
Jansen, A. H., Ioffe, S., and Cherniak, V. (1986).Stimulation of foetal breathing activity
    by P-adrenergic mechanisms. J . Appl. Physiol. 60,1938-1945.
Janssen, R. G., and Randall, D. J. (1975). The effect of changes in pH and PCO, in blood
    and water on breathing in rainbow trout, Salmo gairdneri. Respir. Physiol. 25,
Johansson, P., and Henning, M. (1981). The effect of L-dopa on blood pressure and
    catecholamine content in the brain and peripheral organs in the Atlantic cod, Gadus
    morhua. Comp. Biochem. Physiol. C 70,249-253.
Jonsson, A. C. (1982). Dopamine-/3-hydroxylase activity in the axillary bodies, the heart
    and the splanchnic nerve in two elasmobranchs, Squalus acanthias and Etmmop-
    t e r m spinax. Comp. Biochem. Physiol. C 71,191-194.
Jonsson, A. C., and Nilsson, S. (1979). Effects of pH, temperature and Cu2+ on the
    activity of dopamine-P-hydroxylase from the chromaffin tissue of the cod, Gadus
    morhua. Comp. Biochem. Physiol. C 62,5-8.
Jonsson, A. C., and Nilsson, S. (1983).Catecholamine biosynthesis in uitro and in ciao in
    the chromaffin tissue ofthe Atlantic cod, Gadus morhua. Comp. Biochem. Physiol. C
Jonsson, A. C., Wahlqvist. I., and Hansson, T. (1983). Effects of hypophysectomy and
    cortisol on the catecholamine biosynthesis and catecholamine content in chromaf-
    fin tissue from rainbow trout, Salmo gairdneri. Gen. Comp. Endocrinol. 51, 278-
Johnson, R. G., and Scarpa, A. (1976).Internal pH of isolated chromaffin vesicles.] Biol
    Chem 251,2189-2191.
294                                                D. J. RANDALL A N D S. F. PERRY

Jones, D. R., and Randall, D. J. (1978). The respiratory and circulatory systems during
   exercise. Fish Physiol. 7,425-501.
Keen, K. (1992).Thermal acclimation, cardiac performance and adrenergic sensitivity in
    rainbow trout (Oncorhynchus mykiss). Ph.D. thesis, Simon Fraser University.
Kinkead, R., and Perry, S. F. (1990).An investigation of the role of circulating catechol-
   amines in the control of ventilation during acute moderate hypoxia in rainbow trout
    (Oncorhynchus mykiss).J. Comp. Physiol. B 160,441-448.
Kinkead, R., and Perry, S. F. (1991). The effects of intraarterial infusion of catechol-
   amines on ventilation in rainbow trout (Oncorhynchus mykiss) exposed to external
    hypoxia or hypercapnia. Respir. Physiol. 84,77-92.
Kinkead, R., Fritsche, R., Perry, S. F., and Nilsson, S. (1991). The role of circulating
    catecholamines in the ventilatory and hypertensive responses to hypoxia in the
    Atlantic cod (Gadus morhua).Physiol. 2001. 1087-1109.
Kopin, I. J. (1960). Technique for the study of alternate pathways: Epinephrine metabo-
    lism in man. Sciences NY 131, 1372 (Abstr.).
Le Bras, Y. M. (1982). Effects of anaesthesia and surgery on levels of adrenaline and
    noradrenaline in blood plasma of the eel (Anguilla L.). Comp. Biochem. Physiol. C
Le Bras, Y. M. (1984). Circadian variations of catecholamine levels in brain, heart, and
    plasma in the eel, Anguilla L., at three different times of year. Gen. Comp. Endocri-
    nol. 55,472-479.
Lefiowitz, R. J., Hausdorff, W. P., and Caron, M. G. (1990). Turning off the signal-
    Desensitization of beta-adrenergic receptor function. FASEB /. 4(11), 2881-2889.
Lewander, T., Joh, T. H., and Reis. D. J. (1977).Tyrosine hydroxylase: Delayed activa-
    tion in central noradrenergic neurons and induction in adrenal medulla elicited by
    stimulation of central cholinergic receptors.J. Pharmacol. E x p . Ther. 200,523-526.
Marttila, 0. N. T., and Nikinmaa, M. (1988). Binding of P-adrenergic antagonists 3H-
    DHA and 3H-CGP 12177 to intact rainbow trout (Salmo gairdneri) and carp (Cypri-
    nus carpio) red blood cells. Gen. Comp. Endocrinol. 70,429-435.
Mazeaud, M. M. (1972). Epinephrine biosynthesis in Petromyzon marinus (cyslostoma)
    and Salmo gairdneri (teleost). Comp. Gen. Pharmacol. 3,457-468.
Mazeaud, M. M. (1979).Catecholamine turnover rate in a salmonid, the rainbow trout
    (Salmo gairdneri R . ) . Zchthyophysiol. Acta. 3,70-82.
Mazeaud, M. M., and Mazeaud, F. (1973).Excretion and catabolism ofcatecholamines in
    fish. Part 2. Catabolites. Comp. Gen. Pharmacol. 4,209-217.
McDonald, D. G., and Milligan, C. L. (1992). Chemical and physical properties ofthe
    blood. In “Fish Physiology” (W. S. Hoar and D. J. Randall, eds.), Vol. 12, pp.
    55-133. Academic Press, New York.
McDonald, D. G., and Rogano, M. S. (1986).Ion regulation by the rainbow trout Salrno
    gairdneri in ion poor water. Physiol. 2001.   59,318-331.
McKenzie, D. J., Aota, S., and Randall, D. J . (1991a). Ventilatory and cardiovascular
    responses to blood pH, plasma PC02, blood O2 content and catecholamines in an
    air-breathing fish, the bowfin (Amia caloa). Physiol. Zool. 64,432-450.
McKenzie, D. J., Burleson, M. L., and Randall, D. J . (1991b).The effects of branchial
    denervation and pseudobranch ablation on cardioventilatory control in an air-
    breathing fish. /. E x p . Biol. 161,347-365.
Metcalfe, J. D., and Butler, P. J. (1989). The use of alpha-methyl-p-tyrosine to control
    circulating catecholamines in the dogfish Scyliorhinus canicula: The effects on gas
    exchange in normoxia and hypoxia. /. E x p . Biol. 141,21-32.
Milligan, C. L., and Wood, C. M. (1987).Regulation of blood oxygen transport and red
4. CATECHOLAMINES                                                                      295

    cell pH, after exhaustive exercise in rainbow trout (Salmo gairdneri) and starry
    flounder (Platichthys stellatus).J. E x p . Biol. 133,263-282.
Milligan, C. L., Graham, M. S., and Farrell, A. P. (1989).The response oftrout red cells to
    adrenaline during seasonal acclimation and changes in temperature. J. Fish Biol. 35,
Milsom, W. K., and Sadig, T. (1983). Interaction between norepinephrine and hypoxia on
    carotid body chemoreception in rabbits. J. Appl. Physiol.: Respirat. Environ. Exer-
    cise Physiol. 55, 1893-1898.
Motais, R., Garcia-Romeu, F., and Borgese, F. (1987). The control of Na+/H+ exchange
    by molecular oxygen in trout erythrocytes. J. Gen. Physiol. 90,197-207.
Motais, R., Scheuring, F., Borgese, F., and Garcia-Romeu, F. (1990). Characteristics of
    P-adrenergic-activated Na-proton transport in red blood cells. In “Progress in Cell
    Research” ( J . M. Ritchie, P. J. Magistretti, and L. Bolis, eds.), Vol. 1,pp. 179-193.
Mulligan, E., Lahiri, S., Mokashi, A., Matsumoto, S., and McGregor, K. H. (1986).
    Adrenergic mechanisms in oxygen chemoreception in the cat aortic body. Respir.
    Physiol. 63,375-382.
Nakano, T., and Tomlinson, N. (1967). Catecholamine and carbohydrate concentrations
    in rainbow trout (Salmo gairdneri) in relation to physical disturbance. Fish Res. Bd.
    Canada 24,1701-1715.
Nekvasil, N. P., and Olson, K. R. (1985). Localization of3H-norepinephrine binding sites
    in the trout gill. J. E x p . Zool. 235,309-313.
Nekvasil, N. P., and Olson, K. R. (1986a). Plasma clearance, metabolism, and tissue
    accumulation of 3H-labelled catecholamines in trout. Am. J. Physiol. 259, R519-
Nekvasil, N. P., and Olson, K. R. (1986b). Extraction and metabolism of circulating
    catecholamines by the trout gill. Am. J. Physiol. 250, R526-R531.
Nikinmaa, M. (1990). “Vertebrate Red Blood Cells. Zoophysiology,” Vol. 28, 262 pp.
    Springer-Verlag, Heidelberg.
Nikinmaa, M., and Jenson, F. B. (1986). Blood oxygen transport and acid-base status of
    stressed trout (Salmo gairdneri):Pre- and postbranchial values in winter fish. Comp.
    Biochem. Physiol. A 84(2),391-396.
Nikinmaa, M., Cech, J. J., and McEnroe, M. (1984). Blood oxygen transport in stressed
     striped bass (Morone saxatilis): Role of beta-adrenergic responses.J. Comp. Physiol.
     154B, 365-369.
Nikinmaa, M., Steffensen, J. F., Tufts, B. L., and Randall, D. J. (1987). Control of red cell
    volume and pH in trout: Effects of isoproterenol, transport inhibitors, and extracellu-
     lar pH in bicarbonate/carbon dioxide-buffered media. J. Exp. Zool. 242,273-281.
Nilsson, G. E. (1989). Effects of anoxia on catecholamine levels in brain and kidney of
     the crucian carp. Am. J. Physiol. 257, R10-Rl4.
Nilsson, G. E. (1990). Long-term anoxia in crucian carp: Changes in the levels of amino
     acid and monoamine neurotransmitters in the brain, catecholamines in chromaffin
     tissue, and liver glycogen. J . E x p . Biol. 150,295-320.
Nilsson, S. (1976). Fluorescent histochemistry and cholinesterase staining of sympa-
     thetic ganglia in a teleost, Gadus morhua. Acta Zool. (Stockh.)57,69-77.
Nilsson, S . (1983). “Autonomic Nerve Function in the Vertebrates. Zoophysiology,” Vol.
     13, Springer-Verlag, Berlin, Heidelberg, New York.
Nilsson, S. (1984). Innervation and pharmacology of the gills. In “Fish Physiology”
    (W. S. Hoar and D. J. Randall, eds.), Vol. XA, pp. 185-227. Academic Press, New York.
Nilsson, S., and Grove, D. J . (1974).Adrenergic and cholinergic innervation ofthe spleen
    of the cod: Gadus morhua. Eur. J . Pharmacol. 28,135-143.
296                                               D. J. RANDALL AND S. F. PERRY

Nilsson, S., and Holmgren, S. (1976). Uptake and release of catecholamines in sympa-
     thetic nerve fibres in the spleen of the cod, Gadus morhua. Eur. 1.Pharmacol. 39,
Nilsson, S., Abrahamsson, T., and Grove, D. J. (1976).Sympathetic nervous control of
     adrenaline release from the head kidney ofthe cod, Gadus morhua. Comp. Biochem.
     Physiol. C 55,123-127.
Opdyke, D. F., Bullock, J., Keller, N. E., and Holmes, K. (1983). Effect of ganglionic
     blockade on catecholamine secretion in exercised dogfish. Am. 1. Physiol. 245,
     R9 15- R9 19.
Opdyke, D. F., Carroll, R. G., and Keller, N. E. (1982). Catecholamine release and
     blood pressure changes induced by exercise in dogfish. A m . ] . Physiol. 242, R306-
Part, P., Kiessling, A., and Ring, 0. (1982). Adrenalin increases vascular resistance in
    perfused rainbow trout (Salmo gairdneri Rich.) gills. Comp. Biochem. Physiol. C 72,
Pendleton, R. G., Kaiser, C., and Gessner, G. (1976). Studies on adrenal phenyletha-
    nolamine-N-methyl transferase (PNMT) with SK&F 64139, a selective inhibitor.
    J. Pharmacol. E x p . Ther. 197,623-632.
Pennec, J. P., and Le Bras, Y. M. (1984). Storage and release of catecholamines by
    nervous endings in the isolated heart of the eel (Anguilla anguilla L.). Comp.
    Biochem. Physiol. C 77, 167-171.
Perry, S. F. (1986). Carbon dioxide excretion in fishes. Can.]. Zool. 64,565-572.
Perry, S. F., Booth, C. E., and McDonald, D. G. (1985). Isolated perfused head of
    rainbow trout. I. Gas transfer, acid-base balance and hemodynaniics. Amer. J.
    Physiol. 249, R246-R254.
Perry, S. F., and Kinkead, R. (1989). The role of catecholamines in regulating arterial
    oxygen content during hypercapnic acidosis in rainbow trout (Salmo gairdneri).
    Respir. Physiol. 77,365-378.
Perry, S. F., and Laurent, P. (1990). The role of carbonic anhydrase in carbon dioxide
     excretion, acid-base balance and ionic regulation in aquatic gill breathers. I n
     “Animal Nutrition and Transport Processes” ( J . P. Truchot and B. Lahlou, eds.),
     Vol. 6, Part 2, pp. 39-57. S. Karger, Basel.
Perry, S . F., and Thomas, S. (1991). The effect of endogenous or exogenous catechol-
     amines on blood respiratory status during acute hypoxia in rainbow trout (Oncor-
     hynchus mykiss).].Comp. Physiol. B . 161.489-497.
Perry, S . F., and Vermette, M. G. (1987). The effects of prolonged epinephrine infusion
     on the physiology of the rainbow trout, Salmo gairdneri. I. Blood respiratory, acid-
     base and ionic states.]. E x p . Biol. 128,235-253.
Perry, S. F., and Wood, C. M. (1989).Control and coordination of gas transfer in fishes.
     Can.]. Zool. 67,2961-2970.
Perry, S. F., Wood, C. M., Thomas, S., and Walsh, P. J. (1991a).Adrenergic inhibition of
     bicarbonate dehydration through trout erythrocytes is mediated by the activation of
     N a + / H + exchange.J. E r p . Biol. 157,367-380.
Perry, S. F., Fritsche, R., Kinkead, R., and Nilsson, S. (1991b).Control ofcatecholamine
     release in uiuo and in situ in the Atlantic cod (Gadus morhua)during hypoxia.]. Erp.
     Biol. 155,549-566.
Perry, S. F., Davie, P. S., Daxboeck, C., and Randall, D. J. (1982). A comparison of COz
     excretion in a spontaneously ventilating blood-perfused trout preparation and
     saline-perfused gill preparations: Contribution of the branchial epithelium and red
     blood cell.]. Erp. B i d . 101,47-60.
4.   CATECHOLAMINES                                                                     297

Perry. S. F., Kinkead, R., and Fritsche, R. (1992). Are circulating catecholamines in-
    volved in the control of breathing by fishes? Reu. Fish Biol. Fish. 2,65-83.
Perry. S . F., Kinkead, R., Gallaugher, P., and Randall, D. J. (1989). Evidence that
    hypoxemia promotes catecholamine release during hypercapnic acidosis in rainbow
    trout (Salmo gairdneri).Respir. Physiol. 77,351-364.
Perry, S. F., Malone, S., and Ewing, D. (1987).Hypercapnic acidosis in the rainbow trout
    ( S a l n o gairdneri). I. Branchial ionic fluxes and blood acid-base status. Can.J. Zool.
Perry, S . F., Verbost, P. M., Vermette, M. G., and Flik, G. (1988a).Effects ofepinephrine
    on branchial and renal calcium handling in the rainbow tr0ut.J. E x p . Zool. 246,l-9.
Perry, S. F., Walsh, P. J., Mommsen, T. P., and Moon, T. W. (1988b). Metabolic conse-
    quences of hypercapnia in the rainbow trout, Salmo gairdneri: P-adrenergic effects.
    Gen. Comp. Endocrinol. 69,439-447.
Petterson, K. (1983). Adrenergic control of oxygen transfer in perfused gills of cod,
    Gadus morhua. J. E x p . Biol. 102,327-335.
Pettersson, K., and Nilsson, S. (1979). Catecholamine stores in the holocephalan fish
    Chimaera monstrosa L. Mar. Biol. Lett. 1,41-46.
Peyreaud-Waitzenegger, M. (1979). Simultaneous modifications of ventilation and ar-
    terial PO2 by catecholamines in the eel, Anguilla anguilla L.: Participation o f a and P
    effects. J. Comp. Physiol. B 129,343-354.
Peyreaud-Waitzenegger, M., Barthelemy, L., and Peyreaud, C. (1980). Cardiovascular
    and ventilatory effects of catecholamines in unrestrained eels (Anguilla anguilla L.).
    J. Comp. Physiol. B 138,367-375.
Playle, R. C., Munger, R. S., and Wood, C. M. (1990). Catecholamine effects on gas
    exchange and ventilation in rainbow trout (Salmo gairdneri). J. E x p . Biol. 152,
Primmett, D. R. N., Randall, D. J . , Mazeaud, M., and Boutilier, R. G. (1986).The role of
    catecholamines in erythrocyte pH regulation and oxygen transport in rainbow trout
    (Salmo gairdneri) during exercise. J. E x p . Biol. 122, 139-148.
Randall, D. J. (1990). Control and co-ordination of gas exchange in water breathers. In
    “Advances in Comparative and Environmental Physiology,” Vol. 6, pp. 253-278.
    Springer-Verlag, Berlin, Heidelberg.
Randall, D. J., and Brauner, C. (1991). Effects of environmental factors on exercise in
    fish. J. E x p . Biol. 160, 113-126.
Randall, D. J., and Daxboeck, C. (1984). Oxygen and carbon dioxide transfer across fish
    gills. In “Fish Physiology” (W. S. Hoar, and D. J. Randall, eds.), pp. 263-314.
    Academic Press, New York.
Randall, D. J., and Taylor, E. W. (1992).Control of breathing in fish: Evidence of a role
    for catecholamines. Reu. Fish B i d . Fish. 1, 139-157.
Randall, D. J., Mense, D., and Boutilier, R. G. (1987). The effects of burst swimming on
    aerobic swimming in chinook salmon (Oncorhynchus tshawytscha). M a r . Behau.
    Physiol. 13, 77-88.
Reid, S., and Perry, S. F. (1991).The effects and physiological consequences ofelevated
    cortisol on rainbow trout (Oncorhynchus mykiss) beta-adrenoceptors. J. E x p . Biol.
Reid, S., Moon, T. W., and Perry, S. F. (1991).Characterization ofbeta-adrenoceptors of
    rainbow trout (Oncorhynchus mykiss) erythrocytes. J. E x p . Biol. 158, 199-216.
Reid, S., Moon, T. W., and Perry, S. F. (1992). Rainbow trout hepatocyte beta-
    adrenoreceptors, catecholamine responsiveness, and effects of cortisol. Amer. J.
    Physiol., in press.
298                                                   D. J. RANDALL AND S . F. PERRY

Ristori, M. T., and Laurent, P. (1985). Plasma catecholamines and glucose during moder-
    ate exercise in the trout: Comparisons with bursts of violent activity. J. E x p . Biol. 44,
Ristori, M. T., and Laurent, P. (1989).Plasma catecholamines in rainbow trout (Salmo
    gairdneri) during hypoxia. J. E x p . Biol. 48,285-290.
Romano, L., and Passow, H. (1984). Characterization of anion transport system in trout
    red blood cell. A m . ] . Physiol. 246, C330-C338.
Salama, A., and Nikinmaa, M. (1990). Effect of oxygen tension on catecholamine-
    induced formation of CAMP and on swelling of carp red blood cells. Am. J. Physiol.
    259(5), C723-C726.
Smith, D. G., Wahlqvist, I., Nilsson, S., and Eriksson, B.-M. (1985). Nervous control of
    the blood pressure in the Atlantic cod, Gadus morhua.]. E r p . Biol. 117,335-347.
Spector, S., Gordon, R., Sjoerdsma, A., and Udenfriend, S. (1967). Endproduct inhibition
    of tyrosine hydroxylase as a possible mechanism for regulation of norepinephrine
    synthesis. Mol. Pharmacol. 3,549-555.
Spector, S., Sjoerdsma, A,, and Udenfriend, S. (1965). Blockade of endogenous norepi-
    nephrine synthesis with a-methyltyrosine, an inhibitor of tyrosinehydroxylase.
    J. Pharmacol. E x p . Ther. 147,86-95.
Stabrovskii, E. M. (1968). Adrenaline and noradrenaline in the organs of carp, Cyprinus
    carpio, at rest and under functional stresses. J. Evol. Biochem. Physiol. 4,337-341.
Steffensen, J. F., Tufts, B. L., and Randall, D. J . (1987). Effect of burst swimming and
    adrenaline infusion on O2 consumption and COz excretion in rainbow trout, Salmo
    gairdneri.]. E r p . Biol. 131,427-434.
Tang, Y., and Boutilier, R. G. (1988). Correlation between catecholamine release and
    degree of acidotic stress in rainbow trout, Salmo gairdneri. Am. J. Physiol. 255,
Tang, Y., McDonald, D. G . , and Boutilier, R. G. (1989). Adrenergic regulation of blood
     acid-base status following exhaustive exercise in seawater-adapted rainbow trout,
     Salmo gairdneri. Physiol. 2001.62,950-963.
Temma, K., Iwata, M., Kondo, H., and Ohta, T. (1990). Seasonal variations in the content
     of catecholamines in carp heart (Cyprinus carpio). Comp. Biochem. Physiol. C 97,
Tetens, V., and Christensen, N. J. (1987). Beta-adrenergic control of blood oxygen
     affinity in acutely hypoxia exposed rainbow trout. J. Comp. Physiol. B 157,667-675.
Tetens, V., Lykkeboe, G., and Christensen, N . J . (1988). Potency of adrenaline and
     noradrenaline for beta-adrenergic proton extrusion from red cells of rainbow trout,
     Salmo gairdneri. J. E x p . Biol. 134,267-280.
Thomas, S., and Perry, S. F. (1992). Control and consequences ofadrenergic activation of
     red blood cell Na+/H+ exchange on blood oxygen and carbon dioxide transport.
    J. E x p . Zool.,in press.
Thomas, S., Kinkead, R., Wood, C. M., Walsh, P. J., and Perry, S. F. (1991). Desensitiza-
     tion of adrenaline-induced red blood cell H + extrusion i n uitro after chronic expo-
     sure of rainbow trout (Salmo gairdneri) to moderate environmental hypoxia. I . E x p .
     Biol. 156,233-248.
Thomas, S., Perry, S . F., Pennec, Y., and Maxime, V. (1992). Metabolic alkalosis and the
     response of the trout (Salmo fario) to acute severe hypoxia. Respir. Physiol. 87,
Tufts, B. L., and Randall, D. J. (1989). The functional significance of adrenergic pH
     regulation in fish erythrocytes. C a n . ] . 2001.67,235-238.
4.   CATECHOLAM INE S                                                                  299

Tufts. B. L., Ferguson, R. A., and Boutilier, R. G. (1988).In tjiuo and in uitro effects of
    adrenergic stimulation on chloride/bicarbonate exchange in rainbow trout erythro-
    cytes./. E x p . Biol. 140,301-312.
Ungell, A. L. (1985a).Accumulation of intra-arterially administered [3Hladrenaline and
    [3H]noradrenaline in various tissue of the Atlantic cod, Gadus morhua. C o m p .
    Biochem. Physiol. C 81,25-28.
Ungell, A. L. (198513).Uptake of 3H-adrenaline and 14C-noradrenaline into neuronal and
    extraneuronal tissue compartments in the perfused gas gland of the swimbladder of
    the Atlantic cod, Gadus m0rhua.J.Comp. Physiol. B 155,479-485.
Ungell, A. L., and Nilsson, S. (1979). Metabolic degradation of I3H]-adrenaline in the
    Atlantic cod, Gadus morhua. Comp. Biochem. Physiol. C 64, 137-141.
Ungell, A. L., and Nilsson, S. (1983). Catabolism and excretion of [3H]adrenaline in the
    spiny dogfish, Squalus acanthias. C o m p . Biochem. Physiol. C 74,319-322.
Val, A. L. (1991). Adaptations of fishes to extreme conditions in freshwaters. In “The
    Vertebrate Gas Transport Cascade: Adaptations to Environment and Mode of Life          ”

    (E. Bicudo, ed.). (In press.)
Van Dijk, P. L. M., and Wood, C. M. (1988). The effect ofp-adrenergic blockade on the
    recovery process after strenuous exercise in the rainbow trout, Salmo gairdneri
    Richardson. J . Fish Biol. 32,557-570.
Vermette, M. G., and Perry, S. F. (1987). The effects of prolonged epinephrine infusion
    on the physiology of the rainbow trout, Salmo Gairdneri. J. E x p . Biol. 128,255-267.
Vermette, M. G., and Perry, S. F. (1988). Effects of prolonged epinephrine infusion on
    blood respiratory and acid-base states in the rainbow trout: Alpha and beta effects.
    Fish. Physiol. Biochem. 4(4), 189-202.
Wahlqvist, I. (1981). Branchial vascular effects of catecholamines released from the head
    kidney of the Atlantic cod, Gadus morhua. Molec. Physiol. 1,235-241.
Wahlqvist, I., and Nilsson. S. (1980).Adrenergic control of the cardio-vascular system of
    the Atlantic cod, Gadus morhua, during “stress.”/. C o m p . Physiol. B 137, 145-150.
Wardle, C. S. (1978). Non-release of lactic acid from anaerobic swimming muscle of
    plaice, Pleuronectes platessa L.: A stress reaction./. E x p . Biol. 77, 141-155.
Wood, C. M. (1991). Acid-base and ion balance, metabolism, and their interactions after
     exhaustive exercise in fish./. E x p . Biol. 160,285-308.
Wood, C. M., and Milligan, C. L. (1987).Adrenergic analysis of extracellular and intra-
    cellular lactate and H + dynamics after strenuous exercise in the starry flounder,
     Platichthys stellatus. Physiol. 2001.60,69-81.
Wood, C. M., and Perry, S. F. (1985).Respiratory, circulatory and metabolic adjustments
     to exercise in fish. I n “Circulation, Respiration and Metabolism” (R. Gilles, ed.), pp.
     1-22. Springer-Verlag, Berlin, Heidelberg.
Wood, C. M., and Perry, S. F. (1991). A new in uitro assay for C 0 2 excretion by trout red
     blood cells: Effect of catecho1amines.J. E x p . Biol. 157, 349-366.
Wood, C. M., and Shelton, G. (1975). Physical and adrenergic factors affecting systemic
     vascular resistance in the rainbow trout: A comparison with branchial vascular
    resistance. J . E x p . Biol. 63,505-523.
Wood, C. M., Walsh, P. J., Thomas, S., and Perry, S. F. (1990). Control of red blood cell
    metabolism in rainbow trout (Oncorhynchus mykiss) after exhaustive exercise.
    /. E x p . Biol. 154,491-507.
Wright, P. A., Perry, S. F., and Moon, T. W. (1989).Regulation ofhepatic gluconeogene-
     sis and glycogenolysis by catecholamines in rainbow trout during environmental
     hypoxia./. E x p . Biol. 147, 169-188.
300                                               D. J. RANDALL AND S. F. PERRY

Yamamoto, K., Itazawa, Y., and Kobayashi, H. (1985). Direct observation offish spleen by
    an abdominal window method and its application to exercised and hypoxic yellow-
    tail. J a p .J . Icth. 31,427-433.
Ye, X., and Randall, D. J. (1991). The effect of water pH on swimming performance in
    rainbow trout (Salmo gairdneri, Richardson). Fish. Physiol. Biochem. 9, 15-21.

Department of Zoophysiology
University of Goteborg
Giitehorg, Sweden

  I. Introduction
 11. Origin of Vasomotor and Cardiac Nerves
       A. Vasomotor Nerves
       B. Cardiac Innervation
111. Purines
       A. Purine Derivatives
       B. Purinergic Nerves
       C. Purine Actions on the Heart
       D. Purine Actions on the Vasculature
IV. 5-Hydroxytryptamine (Serotonin)
       A. 5-HT in Cyclostomes
       B. 5-HT in Elasmohranchs
       C. 5-HT in Teleosts
  V. Neuropeptides
       A. Vasoactive Intestinal Polypeptide
       13. Bombesin
       C. Neuropeptide Y
       D. Somatostatin
       E. Substance P
       F. Galanin
       G. Gastrin/Cholecystokinin
 V l . Endothelial Factors


   Although cardiovascular ph rsiology in nonmammalian rertebrates
and its relation to respiration is a major area of comparative physiology,
the systems that control the cardiovascular functions have been given
FISH PHYSIOLOGY, VOL XIIB                          Copylight D 1992 by Academic Preaa, Inc
                                              All rights of reproduction in any form reserved
302                          STEFAN NILSSON A N D SUSANNE H O L M G R E N

modest attention. Knowledge of vasomotor control in fish is still frag-
mentary: Apart from studies on certain peptides with major functions
in cardiovascular control in mammals (notably angiotensin and atrial
natriuretic peptide; see Chapter 3), research on cardiovascular control
systems has been largely restricted to adrenergic vasomotor control
(via adrenergic neurons and circulating catecholamines; see also other
chapters in this volume) and the double antagonistic cholinergic (in-
hibitory) and adrenergic (excitatory) innervation of the teleost heart.
Only recently has data started to accumulate, which demonstrate the
involvement in cardiovascular control of a number of neuropeptides,
and also amines (5-hydroxytryptamine, serotonin) and certain purine
derivatives (adenosine and its nucleotides).
    With the exception of the branchial innervation in some teleosts,
vasomotor nerves in fish appear to be derived solely from the spinal
autonomic (“sympathetic”) division of the autonomic nervous system.
These fibers were previously regarded as solely adrenergic (i.e., re-
leasing adrenaline, and/or noradrenaline as the neurotransmitter). In-
creasing evidence does, however, suggest that vasomotor control in
many organs, especially the gut, may also depend on several types of
spinal autonomic neurons that release nonadrenergic, noncholinergic
(NANC) transmitters (alone or co-released with catecholamines or
acetylcholine; see later).
    The concept of neurons releasing a transmitter substance other than
the classical (adrenaline/noradrenaline or acetylcholine) arose in the
1960s through work on the guinea pig gut by Greeff et al. (1962) and
Burnstock et al. (1963,1964). The nature of these NANC neurons is, in
many cases, still not at all clear although a number of candidates,
notably the neuropeptides (sometimes also known as “brain-gut pep-
tides” or “regulatory peptides”), have been proposed as putative neu-
rotransmitters, also in the cardiovascular system (Burnstock and Grif-
fith, 1988).
    General descriptions of the autonomic innervation patterns in fish
and other vertebrates are given by Nicol(1952), Burnstock (1969), and
Nilsson (1983). In this chapter, we give a brief description of the origin
of vasomotor and cardiac nerves in fish outlining the “classical”
adrenergic and cholinergic pathways. The status of knowledge regard-
ing purines, 5-hydroxytryptamine (serotonin), and neuropeptides in
the cardiovascular system of fish is discussed. However, the available
information is still fragmentary, and parallels need to be drawn with
the situation in mammals where the functions of NANC neurotransmit-
ters and their relations to the adrenergic and cholinergic systems are
better understood.
5. PURINES, 5-HT,   A N D NEUROPEPTIDES                              303


    The autonomic nervous system has a major function in the control
of the cardiovascular system in all vertebrates, with the possible excep-
tion of some cyclostomes (myxinoids) that have a poorly developed
autonomic nervous system. Nerve fibers from the cranial autonomic
division (“parasympathetic nerves”) reach the heart in the paired va-
gus nerve, and fibers from the sympathetic chains (spinal autonomic
nervous system or sympathetic nervous system) either run in separate
nerves to the heart (for instance, the nervi acceleruntes of mammals) or
join the vagi and run as a “vago-sympathetic trunk” to the heart. Spinal
autonomic (sympathetic) fibers run to the viscera in the splanchnic
nerves, which consist of postganglionic neurons in fish, or join the
 spinal nerves and run to somatic vascular beds. A detailed account of
the structure and function of the autonomic nervous system of fishes
can be found in Nilsson (1983). A brief summary of the origins of the
vasomotor and cardiac nerves is offered in Fig. 1.

A. Vasomotor Nerves
    Knowledge of the structure of the autonomic nervous system of
cyclostomes is fragmentary, and differentiation between autonomic
and sensory neurons is hard to make. Visceral branches from the ven-
tral spinal nerves in the hagfish, Myxine glutinosu, probably carry
vasomotor fibers (Nicol, 1952; Fange et al., 1963; Campbell, 1970). In
the lamprey, Lampetra sp., vasomotor fibers that may be regarded as
sympathetic leave the central nervous system (CNS) in both the dorsal
and ventral spinal nerves (Tretjakoff, 1927; Nicol, 1952). Histochemi-
cal evidence suggests that these fibers are adrenergic (Leont’eva, 1966;
Govyrin, 1977). There is no evidence for vasomotor innervation of the
branchial vasculature of cyclostomes.

    The paravertebral autonomic ganglia in elasmobranchs are ar-
ranged segmentally, but longitudinal connections are irregular, which
means that true sympathetic chains of the type found in teleosts and
tetrapods are absent. The most anterior sympathetic ganglia on each
side form, together with masses of chromaffin cells, the axillary bodies.
These contain large quantities of catecholamines, mainly in the chro-
304                                 STEFAN NILSSON AND SUSANNE HOLMGREN

   Gyclostome                                  Elasmobranch

         I                                           I
   @              vein                          KI9
      artery                  heart               artery                  heart
   Dipnoan                                     Teleost

      artery                  heart                artery                 heart
    Fig. 1. Simplified diagrammatic representation of the origins of'the vasomotor and
cardiac nerves in fish from four different groups. Preganglionic neurons are shown as
solid lines, while the postganglionic neurons are shown as broken lines. Chromaffin
tissue occurs in the (large) veins of all four groups and in cyclostomes and dipnoans also
within the heart and possibly in some arteries (such as the intercostal arteries of Pro-
topterus). The vagal innervation of the heart is inhibitory in all vertebrates except
cyclostomes (lampetroids) and may be lacking (myxinoids). Note that the sympathetic
chains in teleosts continue into the head, bearing ganglia that send gray rami commu-
nicantes into the cranial nerves. cc, chromaffin cells; grckpl, gray rami communicantes
or splanchnic nerves; sc, sympathetic chains; sg, sympathetic ganglion; wrc, white rami
communicantes, X, vagus nerve.

maffin tissue, which can be released to the blood within the posterior
cardinal sinuses (Nicol, 1952; Young, 1933; Nilsson, 1983; Nilsson and
Holmgren, 1988).
   The elasmobranch paravertebral ganglia are connected to the spi-
nal nerves via white rami communicantes. Recurrent gray rami are
absent in elasmobranchs, and vasomotor fibers to the somatic vascula-
ture, if such an innervation does indeed exist, must take other path-
ways. Splanchnic nerves, carrying vasomotor (and other) fibers to the
5.   PURINES, 5-HT, AND NEUROPEPTIDES                                305

viscera, are plentiful in the elasmobranchs, and there is thus a rich
autonomic innervation of spinal autonomic (sympathetic) origin.
    Early anatomical reports suggested an innervation of the branchial
vasculature of elasmobranchs via the cranial nerves (Nicol, 1952).
However, later examination has shown that effects on gill blood flow
observed during stimulation of the branchial nerves can be attributed
to contractions of skeletal muscle of the gill arch (Metcalfe and Butler,
1984). Spinal autonomic pathways do not enter the head in elasmo-
branchs, and there is, therefore, no evidence for an autonomic inner-
vation of the branchial vasculature in these fish (Nilsson, 1983).
              AND          ACTINOPTERYGIANS
    In teleosts and other actinopterygians, there are well-developed
sympathetic chains and, at least in teleosts and some ganoids, these
chains continue into the head bearing ganglia in connection with the
cranial nerves. Both white and gray rami communicantes are present,
and fibers of spinal autonomic (sympathetic) origin thus run in the
spinal nerves to the systemic blood vessels. Although there are no
white rami connecting the sympathetic chains to the cranial nerves,
autonomic fibers from the spinal segments of the sympathetic chains
can run forward in the chains and enter the cranial nerves via gray rami
communicantes. A spinal autonomic (sympathetic) vasomotor inner-
vation in those parts of the body that are innervated by the cranial
nerves (or rather cranio-sympathetic nerve trunks) is thus possible
(Nicol, 1952; Nilsson, 1983).
    In addition to the vasomotor control exerted by fibers that originate
in the sympathetic chains, there is good evidence for a cranial (para-
sympathetic) innervation of the branchial vasculature (see extensive
reviews by Laurent (1984)) and Nilsson (1984)). Histochemical and
electron microscopical studies of the innervation of the branchial vas-
culature of teleosts has revealed the presence of cholinergic-type
nerve profiles, that also show strong acetylcholinesterase activity, in
the sphincter at the base of the efferent filamental arteries (Bailly and
Dunel-Erb, 1986) (Fig. 2). These findings are in concert with physio-
logical studies that demonstrate a cholinergic constriction of the ar-
terioarterial vascular pathway of the gills (Smith, 1977, 1978; Pet-
tersson and Nilsson, 1979; Nilsson and Pettersson, 1981).
    Formaldehyde-induced fluorescence histochemistry (Falck-
Hillarp technique) and electron microscopy have been used to show
the distribution of adrenergic nerves in the gills of several teleost
species (Donald, 1984, 1987; Dunel-Erb and Bailly, 1986; Nilsson,
306                                 STEFAN NILSSON AND SUSANNE HOLMGREN



    Fig. 2. Simplified and generalized diagram showing the distribution of cholinergic
(ACh: solid line), serotonergic (5-HT-immunoreactive:dotldash lines), and adrenergic
(Adr, formaldehyde induced fluorescence: dashed lines) fibers in the teleost gill. Seroto-
nergic fibers running in the branchial nerve innervate the sphincter at the base of the
efferent filamental artery, the bases of the efferent filamental arterioles and the central
venous sinus. The occurrence ofadrenergic nerve terminals varies with species. In many
species, an innervation of the afferent filamental arteries, afferent lamellar arterioles,
and the central venous sinus and nutritive vasculature has been observed, and in some
species there are also occasional fibers to the efferent lamellar arterioles. ABA, EBA,
afferent and efferent branchial artery; AFA, EFA, afferent and efferent filamental artery;
ALa, ELa, afferent and efferent lamellar arteriole; BV, branchial vein; CVS, central
venous sinus; FC, filamental cartilage; Sph, sphincter at the base of the efferent fila-
mental artery. [Figure based primarily on histochemical data from Bailly et al. (1989)and
Donald (1984, 1987).]

1986; Dunel-Erb et al., 1989). An innervation of the branchial arteries
b y fluorescent nerve fibers was observed in the carp (Cyprinus carpio)
only, while an innervation of the afferent filamental artery and lamellar
arterioles appears to be a general feature of the teleost gill. In addition,
there are generally dense adrenergic plexuses in the nutritive vascula-
ture and central venous sinus of the gill filament.

B. Cardiac Innervation
   With the probable exception of the myxinoids, all vertebrates pos-
sess a vagal innervation of the heart. This innervation is excitatory in
lampetroids, possibly due to release of local stores of catecholamines
5.   PURINES, 5-HT, AND NEUROPEPTIDES                                307

(adrenaline and noradrenaline), although evidence for an extrinsic
innervation of the endogenous catecholamine storing cells is wanting
(Caravita and Coscia, 1966; Beringer and Hadek, 1973). In all other
vertebrates studied, the vagal innervation is inhibitory. For an in-depth
account of the vagal innervation and control of the fish heart, see
Chapter 6 in this volume. Other reviews on fish cardiac innervation
and control are those by Nilsson (1983), Laurent et al. (1983), and
Farrell (1984).
   The presence of spinal autonomic (sympathetic) pathways to the
heart has not been confirmed either in cyclostomes or in elasmo-
branchs, and anatomical evidence for the existence of such an inner-
vation in dipnoans is ambiguous. In teleosts (with the exception of
pleuronectids where they seem to be absent) and the holostean (Lepi-
sosteus platyrhincus),however, there is a well-established cardiac
innervation by spinal autonomic fibers that reach the heart chiefly in
the vagosympathetic trunks, but that may also run along the first pair of
spinal nerves or enter the heart along the coronary arteries (Gannon
and Burnstock, 1969; Holmgren, 1977; Nilsson, 1983; Donald and
Campbell, 1982). These fibers are excitatory and were originally dem-
onstrated as adrenergic neurons using histochemical and ultrastruc-
tural techniques (Govyrin and Leont'eva, 1965; Yamauchi and Burn-
stock, 1968).


A. Purine Derivatives
    Adenosine 5'-triphosphate (ATP) and its derivatives occur ubiqui-
tously in living cells and serve major functions in the life of the cells.
Adenosine 5'-triphosphate is catabolized by the action of various
adenosine triphosphatase (ATPases), and the formed adenosine di-
phosphate (ADP) and adenosine monophosphate (AMP) can be further
degraded by 5'-nucleotidase to adenosine and finally deaminated to
inosine (Burnstock, 1972).
    From mammalian systems it is known that ATP from cells of the
vascular endothelium may be released in response to disturbances of
the endothelium (such as hypoxia) and affect the vascular smooth
muscle. This occurs either directly, with ATP activating Pz-purinocep-
tors that cause contraction of the vessel, or via release of endothelium
derived relaxing factor(s) (EDRF) that inhibit the musculature of the
media (Fig. 3) (Mione et al., 1990).
308                                  STEFAN NILSSON AND SUSANNE HOLMGREN

                    Adventitia         Media        Endothelium Vascular
                             I             I             I      lumen


     Fig. 3. Hypothetical diagram, based mainly on observations from mammals, show-
ing possible relationships in the mechanisms involving various transmitters and factors
that control the tension ofthe blood vessel wall. Nerve fibers (vasomotor nerves) running
as a plexus at the adventitio-medial border release transmitters, neuromodulators, or
both that may affect the smooth muscle of the media directly or control (via autoregula-
tion) the release of neurotransmitters.
     Catecholamines (CA) from adrenergic nerves or in the form of circulating (humoral)
catecholamines normally constrict arteries by acting on a-adrenoceptors, although in-
hibitory P-adrenoceptors may be dominating in certain vascular beds (e.g., the branchial
vasculature of teleosts).
     Whether released as a cotransmitter from adrenergic nerves, as a transmitter in its
own right from purinergic nerves (if such exist in fish), or as ATP or its metabolites from
endothelial cells, ATP may act directly on vascular purinoceptors (PI or P2-receptors) of
the media, or by stimulating release of endothelium-derived relaxing factor (EDRF)
from endothelial cells. It should be emphasized, however, that the presence of an EDRF
in fish blood vessels has yet to be confirmed.
     5-Hydroxytryptamine (serotonin) is released either as a neurotransmitter from sero-
tonergic vasomotor nerves, from enterochromaffin cells in the gut, or from endothelial
cells of the type demonstrated in fish (see Fig. 4). As in the case of ATP, 5 - H T may also
exert some of its (inhibitory) effects via the release of EDRF.
     A number of neuropeptides, exemplified by substance P (SP) and vasoactive intesti-
nal polypeptide (VIP), may also act in a vasoregulatory system directly from nerves or
carried in the blood stream, and may also act via factors from the endothelium (EDRF).
5.   PURINES, 5-HT, AND NEUROPEPTIDES                                309

   The effects of purine compounds show certain patterns, which have
made a classification of the receptors into PI- and Pe-purinoceptors
useful. The PI-purinoceptors are more sensitive to adenosine than
to ATP, can be blocked by methylxanthines such as caffeine and
theophylline, and affect the levels of cyclic adenosine 3’,5‘-
monophosphate (CAMP). The Pz-purinoceptors are more sensitive to
ATP than to adenosine. Quinidine and 2,2’-pyridylisatogen act as an-
tagonists, and the effect is not mediated via the adenylate cyclase
system (Burnstock, 1976, 1978).

B. Purinergic Nerves
    One of the first hypotheses regarding the nature of the NANC
transmitters was the “purinergic nerve hypothesis,” originally con-
ceived by Burnstock and collaborators (1972, 1975). This hypothesis
suggests that ATP or a related purine nucleotide acts as a transmitter of
autonomic nerves. Most of the evidence in favor of this hypothesis
derives from studies of mammalian systems, notably the gut inner-
vation, although purinergic transmission in other systems, including
the cardiovascular system, has also been suggested (Burnstock and
Griffith, 1988). It is well known and accepted that ATP is stored and
released with other neurotransmitters (e.g., catecholamines), and one
function of nervously released ATP (or ATP-like substances) may be
that of a neuromodulator that affects synaptic transmission of the
“true” transmitter (Burnstock, 1990; Mione et al., 1990).
    Practically nothing is known about the possible existence of pu-
rinergic nerves in the fish cardiovascular system. Thus, the control
exerted by purine compounds will be summarized disregarding their
origin (nervous, endothelial, metabolic, etc.).

C. Purine Actions on the Heart
   During periods of hypoxia, ATP resynthesis may be inhibited in
some tissues and the concentration of adenosine can increase. Al-
though several studies of the fish heart suggest a close control of the
ATP levels [which can be sustained by linkage to creatine phosphate
(JGrgensen and Mustafa, 1980; Nielsen and Gesser, 1984; Koke and
Anderson, 1986)1, levels of the dephosphorylated adenosine com-
pounds may increase, and the physiological effects of these can be of
significance in cardiac control during hypoxia. The pattern of effects of
adenosine compounds on the fish heart is, however, not uniform.
   A positive inotropic response to adenosine occurs in the heart ofthe
310                          STEFAN NILSSON AND SUSANNE HOLMGREN

flounder, Platichthys flesus. The effect was antagonized by the PI-
purinoceptor antagonist caffeine (Lennard and Huddart, 1989). In the
carp (Cyprinus carpio), adenosine and adenine nucleotides produced
negative inotropic and chronotropic effects that could be blocked by
theophylline, again demonstrating purinoceptors of the PI-variety (Co-
hen et al., 1981; Rotmensch et al., 1981).
    Studies of the rainbow trout (Salmo gairdneri = Oncorhynchus
mykiss) heart by Meghji and Burnstock (1984a) showed negative ino-
tropic (and positive chronotropic) effects of adenosine and ATP on the
atrium: The effects were, however, insensitive to a PI-antagonist, and
it was concluded that the rainbow trout heart purinoceptor differs from
the type found in other vertebrates (Meghji and Burnstock, 1984a).
Studies by the same authors on the heart of the dogfish, Scyliorhinus
canicula, show the presence of PI-purinoceptors, mediating negative
inotropic and chronotropic effects in the atrium, while the ventricle
was largely insensitive to the adenosine compounds (Meghji and
Burnstock, 198413).

D. Purine Actions on the Vasculature
    Theophylline-sensitive contraction of coronary artery ring prepara-
tions caused by adenosine, ATP, and ADP have been demonstrated in
rainbow and steelhead trout (Small et al., 1990; Small and Farrell,
1990), and also in the skate, Raja nasutu, at low concentrations while
higher concentrations produced relaxation (Farrell and Davie, 1991b).
In contrast, vascular rings from coronary arteries of the maco shark,
Zsurus oxyrinchus, relaxed in response to adenosine and ADP (in high
concentration). This effect was also inhibited by theophylline (Farrell
and Davie, 1991a).
    The effect of adenosine compounds on other systemic vessels is
similarly variable. Adenosine infusion was without effect on the iso-
lated perfused trunk preparation of the rainbow trout (Colin et al.,
1979), while Wood (1977) demonstrated systemic vasoconstriction in
response to ATP injected in vivo.
    Adenosine dilates the branchial vasculature in the hagfish, Myxine
glutinosa (Axelsson et al., 1990), and several studies have demon-
strated effects of adenosine compounds on the branchial vasculature in
teleosts. A marked arterioarterial branchial vasoconstriction occurs in
the isolated-perfused head preparation of the rainbow trout, and
adenosine infusion in vivo also markedly increased the branchial vas-
cular resistance. Isolated gill arches with the filaments removed
showed a higher sensitivity in the efferent than in the afferent vessels,
5. PURINES,   5-HT, AND NEUROPEPTIDES                                31 1

and the authors implicated the sphincter at the base of the efferent
filamental artery (see Fig. 2) in the contractile response (Ristori and
Laurent, 1977; Colin et al., 1979). Similarly, adenosine produced theo-
phylline-sensitive vasoconstriction in the isolated perfused gills of
Oreochromis niloticus, and in these experiments a vasodilator effect of
ATP was also observed (Okafor and Oduleye, 1986).
    From a thorough pharmacological analysis of the relative activities
of a number of adenosine analogues, Colin and Leray (1981) concluded
the presence of specific vascular purinoceptors of the gill vasculature.
It would thus seem that the branchial vasculature of teleost fish pos-
sesses PI-purinoceptors responsible for arterioarterial vasoconstriction
in response to adenosine.


    5-Hydroxytryptamine (5-HT, serotonin, enteramin), originally de-
scribed as a vasoactive factor present in blood serum (hence the name
serotonin) (Rapport et al., 1948; Erspamer, 1954), emanates from sev-
eral sources, notably enterochromaffin cells of the gut, but also from
  serotonergic” nerves. 5-HT fulfills the criteria for a neurotransmitter
in mammals (Griffith, 1988) and it is reasonable to believe that the
histochemical findings of 5-HT nerves indicate a role for 5-HT as a
transmitter also in nonmammalian species. The species variations in
the source of 5-HT, and whether or not the neuronally located 5-HT is
involved in cardiovascular control in fish, are discussed later.
    In addition to the possible effects of 5-HT as a neurotransmitter or
hormone from enterochromaffin cells, 5-HT stored in and released
from specialized endothelial cells may affect blood vessels directly or
via stimulation of the release of EDRF (Figs. 2 and 4).
    Tables I, 11, and I11 summarize the known distribution and effects
of 5-HT in fish.

A. 5-HT in Cyclostomes
    In cyclostomes (Atlantic hagfish, Myxine glutinosa, and Pacific hag-
fish, Polistotrema [Eptatretus]stouti) the effects of 5-HT on the car-
diovascular system may be indirect, due to release of catecholamines.
Injections of 5-HT caused a small increase in dorsal aortic blood pres-
sure and heart rate and perfusion of 5-HT with constant flow rate
through the branchial system or systemic vessels to the gut, in situ,
produced both increases and decreases in vascular resistance. The
312                                STEFAN NILSSON AND SUSANNE HOLMGREN

    Fig. 4. Electron micrograph of endothelial cells (E)lining the vascular lumen (L) of
a swimbladder artery of the eel (Anguilla anguilla). Smooth muscle cells of the media
(M) are also seen. Note the darker 5-HT immunoreactive cell. [Courtesy of K. Lundin,

responses mimicked, but were much weaker than, the responses to the
same concentrations of adrenaline or noradrenaline and were blocked
by adrenergic antagonists but not by the serotonergic antagonist methy-
sergide (Reite, 1969).
    The formaldehyde-induced fluorescence histochemical technique
(Falck-Hillarp technique), autoradiography studies of the uptake of
tritiated 5-HT, and immunohistochemistry performed on the gut of
different species of cyclostomes have demonstrated the presence of
5-HT-containing nerve fibers in the gut (Honma, 1970; Baumgarten et
al., 1973; Sakharov and Salimova, 1980; S. Nilsson and S. Holmgren,
unpublished results). However, these fibers show no particular rela-
tion to the vessels of the gut, and no 5-HT-containing (enterochro-
maffin) cells of the mucosa (which could possibly provide a “hormonal
source” of 5-HT) have been observed in either species (El-Salhy et al.,
1985; S. Nilsson and S. Holmgren, unpublished studies). It is, there-
fore, possible that the effects of 5-HT observed by Reite (1969) are
merely pharmacological and of little physiological significance.

B. 5-HT in Elasmobranchs
   In Squalus acanthias there are numerous endocrine cells showing
5-HT-like immunoreactivity (IR) in the gut mucosa, but there is only a
sparse innervation of the muscular layers of the gut, and no 5-HT fibers
                                                               Table I
                    Summary of Anatomical Data Concerning the Innervation of the Branchial Vasculature in Teleost Fish

         Species                BA      AFA      ALA       Lam      ELA       EFA      Sph      CVS      Nut               Reference

Adrenergic neurons
 C yprinus carpio                +                 +                                                              Donald (1'387)
 Platycephalus bassensis         -                 +                                                              Donald (1987)
 P . caeuruleopunctatus          -                 +                                                              Donald (1987)
 Tetractenos glaber              ~
                                                                                                                  Donald (1987)
 Anguilla australis              -                 +                                                              Donald (1987)
 Gadopsis marmoratus             -                 +                                                              Donald (1987)
 Salmo trutta                    +                 +                                                              Donald (1984)
 Salmo gairdneri                 -                 +                                                              Donald (1984); Dunel-Erh
                                                                                                                  and Bailly (1986)
  Gadus morhua                   -                 +                                             +        +       Nilsson (1986)
Cholinergic neurons
  Perca fluviatilis                                                                     +                         Bailly and Dunel-Erb
  Gadus morhua                                                                          +                         (1986)
Serotonergic neurons/cells                                                                                        Bailly and Dunel-Erb
  Salmo gairdneri                -                                    t        +        +        +         +      (1986)

   Abbreviations: AFA, EFA, afferent and efferent filamental artery; ALA, ELA, afferent and efferent filamental arterioles; BA, branchial artery;
CVS, central venous sinus; Lam, (secondary) lamellae; Nut, nutritive vasculature; Sph, sphincter at the base of the efferent filamental artery.
    314                               STEFAN NILSSON A N D SUSANNE HOLMGREN

                                           Table I1
Putative Neurotransmitters in Perivascular Nerves of Fish, Revealed by Immunohistochemistry

    Tissue         Transmitter             Species                       Reference

Gut/c.a,ni.a        BM            Lampetrapuviatilis           S. Holmgren, unpublished
                                  Raja erinacea                Bjenning e t al. (1991)
                                  Scyliorhinus canicula        Tagliafierro e t al. (1988)
                                  Squatina aculeata            Tagliafierro e t al. (1988)
                                  Squalus acanthias            Bjenning e t al. (1990)
                    CGRP          Lampetra fluviutilis         S . Holmgren, unpublished
                    GAL           G a d u s morlzua            S. Holmgren, unpublished
                    N PY          Raja erinucea, R. radiata    Bjenning e t al. (1989)
                    SOM           Squalus acanthias            Holmgren and Nilsson
                    SP            Gadus morhua                 Jensen and Holmgren (1991)
                    VIP           Squalus acanthias            Holmgren and Nilsson
                                  Lepisosteus platyrhincus     Holmgren and Nilsson
                                  Gadus morhua                 (198313)
                                                               Lundin and Holmgren (1984);
                                                               Jensen and Holmgren (1985);
                                                               S. Holmgren, unpublished
                                  Salmo gairdneri              Holmgren et al. (1982)
Gills               GiCCK         Gadus morhua                 S. Holmgren, unpublished
                    5-HT          Salmo gairdneri              Bailly e t al. (1989)
                    NPY           Raja erinacea, R . rudiata   Bjenning et al. (1989)
                    VIP           Gadus morhua                 S. Holmgren, unpublished
Coronary artery     BM            Raja erinacea                Bjenning et al. (1991)
                                  Salmo guirdneri              Bjenning and Holmgren
                     N PY         Raja erinacea, R. radiata    Bjenning et ul. (1989)
Swimbladder         5-HT          Anguilla anguilla            Lundin and Holmgren (1989);
                                                               K. Lundin, unpublished
                    VIP           Gudus niorhua                Lundin and Holmgren (1984)
Gallbladder         VIP           Gudus morhua                 Aldman and Holmgren (1987)
Gonads              VIP           Gudus morhua                 Uematsu e t al. (1989)
Urinary bladder      VIP          Gadus morhua                 Lundin and Holmgren (1986)
Brain               BM            Scyliorhinus                 Vallarino e t a / . (1990)
Ducts of            VIP           G a d u s morhua             S. Holmgren, unpublished

    Abbreviations: 5-HT, 5-hydroxytryptamine (serotonin); BM, bombesin; c.a, coeliac artery;
CGRP, calcitonin gene related peptide; GAL, galanin; GICCK, gastrin/cholecystokinin; m a . ,
mesenteric artery; NPY, neuropeptide Y; SOM, somatostatin; SP, substance P; VIP, vasoactive
intestinal polypeptide.
    5.   PURINES, 5-HT, AND NEUROPEPTIDES                                              315

                                           Table 111
                     Effects of 5-HT on the Cardiovascular System in Fish

         Effect         Organ!tissue             Species                     Reference

Increased blood       Dorsal aorta        Myxine glutinosa         Reite (1969)
  pressure                                Polistotrema stouti      Reite (1969)
                      Ventral aorta       Squalus suckley          Reite (1969)
                                          Hydrolagus collei        Reite (1969)
                                          Gadus morhua             Reite (1969)
                                                                   R. Fritsche, unpublished
                                          Anguilla                 Reite (1969)
Decreased blood       Dorsal aorta        Five teleost species     Reite (1969)
Tachycardia           Heart               Myxine glutinosa         Reite (1969)
                                          Polistotrema stouti      Reite (1969)
Increased pressure    Ventricle           Salmo gairdneri          Reite (1969)
Increased vascular    Gills               Myxine, Polistotrema     Reite (1969)
  resistance                              Squalus acanthias        Ostlund and Fange (1962)
                                          Anguilla anguilla        Ostlund and Fange (1962)
                                          Zoarces uiuiparus        Ostlund and Fange (1962)
                                          Labrus berggyl tu        Ostlund and Fagne (1962)
                                          Gadus morhua             Ostlund and Fagne (1962)
                                          Salmo gairdneri          Katchen et a / . (1976)
                       Systemic vascul.   Myxine, Polistotrema     Reite (1969)
Relaxation             Coronaiy vascul.   Salmo gairdneri           Small et a / . (1990)

    innervating visceral vessels have been observed (Holmgren and
    Nilsson, 1983a; El-Salhy et al., 1985). However, in Scyliorhinus
    canicula and Squatina aculeata fibers surround gut vessels (Taglia-
    fierro et al., 1988). This may indicate true species differences but may
    also depend on different sensitivity of the antisera used. Using the
    formaldehyde-induced fluorescence method, Bailly (1983) failed to
    show a serotonergic innervation of the gill filaments of the dogfish,
    Scyliorhinus canicula. However, this method is not as sensitive as
    immunochemical methods and does not, therefore, completely rule
    out the possible existence of branchial serotonergic nerves.
        I n uivo experiments in dogfish (Squalus suckley) and ratfish ( H y -
    drolagus collei) showed a small increase in ventral aortic pressure after
    5-HT injection, while the pressure in the dorsal aorta was unaffected.
    In skates, 5-HT produced no response (Reite, 1969).The indications of
    an effect on the gill vasculature is in agreement with the findings in
    teleosts (see later) and could point to a similar control function of 5-HT
316                          STEFAN NILSSON AND SUSANNE HOLMGREN

on the gills, although histochemical evidence for serotonergic nerves
in elasmobranch gills is still wanting.

C. 5-HT in Teleosts
    The most conspicuous action of 5-HT on the cardiovascular system
in teleosts is a marked constriction of the branchial vasculature. Injec-
tions of 5-HT caused a decrease in dorsal aortic pressure in five teleost
species studied, while the ventral aortic pressure, measured in cod and
eel (anesthetized), and the intraventricular systolic pressure in rain-
bow trout (anesthetized)increased rapidly (Reite, 1969). Direct effects
on systemic vessels were small, while perfused branchial vessels from
cod and eel showed a marked constriction that could be blocked by the
specific serotonergic antagonist methysergide (Ostlund and Fange,
1962; Reite, 1969; Katchen et aZ., 1976). Similar results were obtained
in a preliminary study of unrestrained cod in our laboratory. Injection
of 5-HT caused an increase in ventral aortic pressure and heart rate,
while dorsal aortic pressure was reduced and lost its pulsatile nature,
indicating a constriction of the branchial vessels (R. Fritsche, personal
    In the rainbow trout, injection of 5-HT produced the same response
as exposure to acidified water or infusion of HCl (i.e., an increased
frequency and magnitude of opercular movements, a drop in arterial
oxygen tension (P,o,), a rise in arterial COZ tension (PLco2),      and a
decrease in arterial pH) while the heart rate remained unchanged. The
responses to both acidification and 5-HT were antagonized by methy-
sergide. The results were interpreted as a mechanism for redistribut-
ing blood resulting in a decrease in gas exchange during acidification
(Thomas et al., 1979).
    Careful anatomical and histological studies of the gills from rain-
bow trout made by Dunel-Erb and co-workers have shown the
presence of 5-HT in at least three types of cells: neurons (and nerve
fibers), polymorphous granular cells (PGCs), and neuroepithelial cells
(NECs) (Dunel-Erb et aZ., 1982; 1989; Bailly et al., 1989). The neurons
are of vagal origin, they innervate the proximal part of the efferent
filament artery (including the sphincter) and extend to the efferent
lamellar arterioles and central venous sinus (see Fig. 2). The fibers
were observed impinging on the vascular smooth muscle, which sup-
ports the theory that 5-HT is involved in branchial vasomotor control.
 In view of the constrictor effects on gills obtained in physiological
experiments (see earlier), it may be hypothesized that 5-HT, released
 from nerves or other cells within the gills, cause a constriction of the
5.   PUKINES, 5-HT, AND NEUROPEPTIDES                               317

efferent arteries, thereby increasing lamellar recruitment during con-
ditions such as hypoxia, exercise, or stress.
     In mammals, 5-HT has been demonstrated in endothelial cells of
certain blood vessels (Lincoln et al., 1990), and 5-HT has been shown
to produce vasodilatation by stimulating a release of EDRF from the
epithelium of coronary vessels (Cocks and Angus, 1983)(see also Fig.
3 ) .In the eel, Anguilla anguilla, 5-HT-immunoreactive cells with long
varicose processes, possibly paracrine in nature, are present in the
mucosa of the pneumatic duct of the swimbladder (Lundin and
Holmgren, 1989). At the ultrastructural level, it was observed that
these cells were confined to the vascular endothelium of small vessels
supplying the pneumatic duct (Lundin, 1991; Fig. 4). It is possible that
these 5-HT-immunoreactive cells are involved in the control of the
blood flow through this resorptive part of the swimbladder, thereby
affecting the rate of resorption of gas from the swimbladder. Whether
or not this mechanism involves an EDRF in fish, and whether or not
5-HT has a dilatory effect on the swimbladder vessels remains to be
elucidated. 5-HT has weak inhibitory effects on isolated coronary ar-
teries from the rainbow trout, but in this case, it is unlikely that the
response is mediated by EDRF, because most of the endothelium was
removed during preparation of the artery rings (Small et al., 1990).


    Neuropeptides are bioactive peptides of about 4-40 amino acids
present in autonomic neurons as well as in neurons of the CNS. Identi-
cal or closely related peptides (“gut hormones”) occur in endocrine
cells of the gut and pancreas. Variations in the amino acid sequences
between animal species are common. The number of amino acids
substituted, their position, and the types of substitution indicate two
things: (a) the evolutionary relationship between different animal
groups, and (b) the importance of different parts of the molecule in the
bioactivity of the peptide.
    Related peptides form families such as the tachykinins, the vasoac-
tive intestinal polypeptide/peptide histidine isoleucin-like (VIP/PHI)
peptides, and the gastrin/cholecystokinin-like  (gastrin/CCK-like)pep-
tides. Several members of a peptide family often occur in one animal
species, and may have either similar or clearly separate functions.
Closely related peptides from different species may be given the same
name with a species prefix (e.g., porcine VIP, human VIP, and cod
    318                                   STEFAN NILSSON AND SUSANNE HOLMGREN

       Research on neuropeptides and gut hormones has evolved dramati-
    cally over the last two decades, and accumulating evidence shows the
    involvement of these compounds in cardiovascular control (Burnstock
    and Griffith, 1988). In the following account, only those peptides have
    been included that have been implicated in the control of the fish
    cardiovascular system (Tables 11, IV-VII).

    A. Vasoactive Intestinal Polypeptide
        The 28-amino-acid peptide vasoactive intestinal polypeptide (VIP)
    holds a prominent position in mammalian cardiovascular physiology as
    a neuropeptide early recognized to possess vasoactive properties. In
    mammals, cerebral arteries are densely innervated by VIP-containing
    fibers, while the density of the innervation of peripheral systemic
    vessels and vascular beds varies; blood vessels of the gastrointestinal
    tract, the respiratory tract, and the urogenital tract are densely inner-
    vated, while systemic, coronary, and blood vessels of the liver, the

                                           Table I V
                     Effects of VIP on the Cardiovascular System in Fish

      Effect             Orgadtissue               Species                 Reference

Decreased vascular   Perfused intestine       lctalurus melas     Holder et al. (1983)
                     Perf. swimbladder        Gadus morhua        Lundin and Holmgren
                     Perf. rectal gland       Squalus acanthias   Solomon et al. (1984)
                                                                  Thorndyke et al. (1989)
                     Perfused gills           Salmo trutta        Bolis et al. (1984)
                     A.mes. in uiuo           Gadus morhua        Jensen et al. (1991)
                     A.coe1 in uioo           Cadus morhua        Jensen et al. (1991)
                     Ventral aorta in oioo    Gadus morhua        Jensen et al. (1991)
Increased vascular   Overall systemic         Squalus acanthias   S. Holmgren, M. Axelsson,
  resistance                                                        and A. P. Farrell,
                     Coeliac vasculature      Squalus acanthias   S. Holmgren, M. Axelsson,
                                                                    and A. P. Farrell,
Increased stroke     Heart                    Gadus morhua        Jensen et al. (1991)
Tachycardia          Heart                    Squalus acanthias   S. Holmgren, M. Axelsson,
                                                                    and A. P. Farrell,
5.   PURINES, 5-HT, AND NEUROPEPTIDES                               319

spleen, the pancreas, and the kidney generally show a sparse inner-
vation (Edvinsson and Uddman, 1988).
    Also in fish, an increasing number of studies point to the involve-
ment of one or several VIP-like peptides in the cardiovascular control
(Tables I1 and IV). Immunohistochemical studies suggest that major
systemic arteries are more densely innervated than the peripheral
vascular beds. In the spiny dogfish, Squalus acanthias, a moderately
dense plexus of fibers was found in the walls of the coeliac and mes-
enteric arteries (Holmgren and Nilsson, 1983a). In the cod, Gadus
morhua, there is a dense perivascular nerve plexus in the adven-
titiomedial border of the mesenteric artery and its branch to the swim-
bladder (swimbladder artery) (Lundin and Holmgren, 1984). Well-
developed perivascular plexuses were observed along the small
mesenterial branches of the coeliac and mesenteric arteries, and in
the branches on the surface of the gut wall (S. Holmgren unpublished;
Fig. 5 ) .
    The VIP innervation of small arteries and veins running within the
walls of the swimbladder, the urinary bladder, the gonads, and the
gallbladder appears to be sparse (Lundin and Holmgren, 1984, 1986,
1989; Aldman and Holmgren, 1987; Uematsu et al., 1989).Similarly,
the density of perivascular VIP fibers innervating vessels intrinsic to
the gut wall was low in investigated fish species, such as the holostean,
Lepisosteus platyrhiricus (Holmgren and Nilsson, 1983b), the cod
(Jensen and Holmgren, 1985), and the rainbow trout (Holmgren et
al., 1982). In a study of the myenteric plexus and muscular layers
of 18 elasmobranch and teleost species, it was concluded that VIP-
immunoreactive fibers follow and surround vessels of these gut layers
to some extent only (Bjenning and Holmgren, 1988).
    VIP extracts from rainbow trout and catfish gut, like porcine VIP,
induced a vasodilation in a perfused intestinal loop of the catfish,
Zctalurus melas (Holder et aZ., 1983).This is to our knowledge the only
published study using native fish VIP on the fish cardiovascular sys-
tem. Instead, most studies of the presence and function of VIP in fish
have been performed using mammalian VIP or antibodies raised
against mammalian VIP. Sequence analyses of VIP from the dogfish,
Scyliorhinus canicula, and the cod, Gadus morhua, showed that the 28
amino acid sequence of dogfish and cod VIP varies in five positions
only from porcine VIP (Dimaline and Thorndyke, 1986; Dimaline et
al., 1987; Thwaites et aZ., 1989). Furthermore, the elasmobranch VIP
has full affinity for mammalian pancreatic VIP receptors (Dimaline et
al., 1987). It is, therefore, reason to believe that available immunohis-
tochemical data give a good indication of the distribution of nerve
320                                STEFAN NILSSON AND SUSANNE HOLMGREN

    Fig. 5. Immunohistochemistry of blood vessels from cod (Gadus rnorhua) showing
VIP-immunoreactive nerve fibers in the wall of the ducts of Cuvier (A), and transverse
section (B) and whole mount preparation (C) of small arteries on the surface of the gut.
(D)and (E)show NPY and bombesin immunoreactive fibers, respectively, in the wall of
small arteries entering the stomach wall.

fibers containing VIP-like material in fish. Most studies in fish using
mammalian VIP give results that agree with the effects of VIP on the
mammalian cardiovascular system, but it must be kept in mind that the
differences in amino acid sequence between mammalian and fish VIP
could give the mammalian VIP antagonistic, rather than agonistic,
properties when tested in fish.
   Generally, porcine VIP appears to be vasodilator in teleosts, as in
mammals. In the cod, VIP caused a long-lasting decrease in vascular
resistance during perfusion of the isolated gas gland and swimbladder,
probably due to vasodilation of the vascular beds fed by the mesenteric
and swimbladder arteries, which possess VIP-immunoreactive peri-
vascular nerves (Lundin and Holmgren, 1984).
   The arterioarterial flow through isolated gill arches from the brown
5.   PURINES, 5-HT, AND NEUROPEPTIDES                                  321

trout, Salmo trutta, was dose dependently increased by VIP (Bolis et
al., 1984). However, in the cod only a few, usually single, varicose
fibers were observed running along the gill vessels (S. Holmgren,
unpublished), and injection of VIP in vivo produced no effects on
blood pressure, flow, or vascular resistance that may be attributed to an
effect on the gills (Jensen et al., 1991). It is premature at this stage to
speculate whether or not these differences between the rainbow trout
and the cod studies are due to the different experimental approaches or
to true species differences.
    Injections of porcine VIP into the cardiovascular system of
unanesthetized cod in vivo caused an increase in the gut blood flow,
due to an increase in flow in the coeliac and mesenteric arteries and in
the ventral aorta. The increased cardiac output was caused by an
increase in stroke volume, while the heart rate was largely unaffected.
Surprisingly, the increase in cardiac output appeared to be the only
reason for the increase in flow in the mesenteric artery. In the coeliac
artery, on the other hand, a decrease in vascular resistance caused by
vasodilation further increased the flow (Jensen et al., 1991). The rea-
sons for this discrepancy between the two vascular beds are uncertain,
but preliminary immunohistochemical studies show a more dense VIP
innervation of the coeliac artery than the mesenteric artery, indicating
that this vessel normally is more influenced by a VIP-like peptide than
the mesenteric artery.
    In the elasmobranch Squalus acanthias, a different effect of por-
cine VIP has been obtained. Injections into unrestrained fish in vivo
caused increased total vascular resistance including the coeliac artery
vascular bed. The blood flow to the gut was consequently reduced.
The dorsal aortic blood pressure was slightly increased (S. Holmgren,
M. Axelsson, and A. P. Farrell, unpublished). Whether this reflects true
differences between elasmobranchs and teleosts, differences between
individual species, or depends on the difference between porcine VIP
and the native Squalus VIP remains to be elucidated. It is notable,
however, that in the rectal gland of Squalus, VIP caused the same
effect as that described in mammalian exocrine glands: a vasodilation,
combined with an increase in glandular secretion (Solomon et al.,
 1984; Thorndyke et al., 1989).

B. Bombesin
    It is not yet clear exactly which of the approximately 20 bombesin-
like peptides known today are present in fish. Most studies indicate
the presence of several related peptides, and the presence of both
longer, gastrin-releasing peptide-like (GRP) forms and shorter,
322                          S T E F A N NILSSON A N D SUSANNE H O L M G R E N

bombesin-like forms, possibly with species differences, have been
concluded. Thus, Conlon et al. (1987) found that bombesin-like mate-
rial extracted from the intestine of the common dogfish, Scyliorhinus
canicula, occurs in large and small forms; the long, partly sequenced
25 amino acid form shows a clear homology with mammalian and
chicken GRP, and it has been argued that authentic bombesin is not
present in this species. On the other hand, one of the two bornbesin/
GRP-related peptides present in the gut of Scyliorhinus stellaris ap-
pears closely related to bombesin (Cimini et al., 1985). Several forms of
bombesin-like material were obtained from the cod gut, one of which
shows similar properties to bombesin (Holmgren and Jonsson, 1988),
and bombesin-like material isolated from the intestine of the ratfish,
Hydrolagus colliei, was equipotent with synthetic amphibian bom-
besin in stimulating amylase secretion; the effect being blocked by a
specific bombesin antagonist (Thorndyke et al., 1990).
    Binding sites for bombesin have been demonstrated in the stomach
of the teleost Scorphaeichthys marmoratus (Vigna and Thorndyke,
1989), and there are several studies in elasmobranchs and teleosts that
show effects of exogenous bombesin on gut smooth muscle (Lundin et
d., Holmgren, 1983;Jensen and Holmgren, 1985; Holmgren and
Jonsson, 1988; Thorndyke and Holmgren, 1990), on gastric acid secre-
tion (Holstein and Humphrey, 1980), and on behavior (Kavaliers and
Hawkins, 1981; Beach et al., 1988).
    Less attention has been paid to the possible bombesin innervation
of the cardiovascular system in fish (Table 11), but a moderately dense
to dense plexus of perivascular nerves showing bombesin-like immu-
noreactivity has been demonstrated in systemic vessels to the gut in
the elasmobranchs, Squalus acanthias and Raja erinacea; the distribu-
tion of the fibers to the adventitiomedial border implies an involve-
ment in the control of the vascular smooth muscle (Bjenning et d.,
1990, 1991).
     A sparse innervation has been found in the heart and coronary
vessels of the little skate, Raja erinacea (Bjenning et al., 1991). A
colocalization of bombesin-like peptide(s) with 5-HT in perivascular
 nerves of the gut was suggested after immunohistochemical studies in
the elasmobranchs Scyliorhinus canicula and Squatina aculeata (Tag-
liafierro et al., 1988), but the physiological significance of this has not
been looked into further. In the brain of Scyliorhinus, bombesin im-
munoreactive fibers innervate vascular structures of the median emi-
 nence and may be involved in hypophysiotropic actions similar to the
 situation in mammals (Vallarino et al., 1990). In the rainbow trout,
 immunoreactive fibers surround ganglion cells in the sinoatrial region
    5.   PURINES, 5-HT, AND NEUROPEPTIDES                                           323

   and innervate the myocardium (Bjenning and Holmgren, 1989).This is
   compatible with the view of a regulatory function on the heart activity.
       Physiological experiments (Table V) further suggest a role of
   bombesin-like peptides in the cardiovascular control of fish. The flow
   through the vascularly perfused stomach of the spiny dogfish shows
   several phases of increase and decrease in resistance, possibly re-
   flecting activation of several mechanisms (Bjenning et al., 1990). In
   uuiuo, bombesin causes an major increase in somatic vascular resis-
   tance, which causes a shunting of blood into the coeliac artery, although
   the resistance in the coeliac artery vascular bed is slightly increased (S.
   Holmgren, M. Axelsson, and A. P. Farrell, unpublished).
       A small, but significant bradycardia is obtained in the spiny dogfish
   in uiuo; this occurs after a significant increase in dorsal aortic pressure
   and may possibly demonstrate the presence of a reflex bradycardia (S.
   Holmgren, M. Axelsson, and A. P. Farrell, unpublished). However, a
   negative chronotropic effect (and a negative inotropic effect) of bom-
   besin is obtained in the isolated perfused heart of the rainbow trout (C.
   Bjenning and S. Holmgren, unpublished), suggesting a direct effect on
   the heart tissues at least in this species. Coronary vessels from the
   longnose skate, Raja rhina, contract in response to bombesin
   (Bjenning et al., 1991).
       Injections of bombesin in the cod in viuo failed to produce an effect
   on blood pressure, heart performance, or flow to the gut (M. Axelsson

                                           Table V
                   Effects of Bombesin on the Cardiovascular System in Fish

         Effect             Organ/tissue             Species                  Reference
ncreasedldecreased     Perfused stomach       Squalus acanthias       Bjenning et (11. (1990)
 vascular resistance
mcreased vascular      Overall systemic and   Squalus acanthias       S. Holmgren, M.
 resistance             perfused tail                                   Axelsson, and A. P.
                                                                        Farrell, unpublishe
Bradycardia            Perfused heart         Oncorhynchus mykiss     Bjenning and
                                                                         Holmgren (1989)
                       I n aiuo               Squalus acanthias       S. Holmgren, hl.
                                                                        Axelsson, and A. P.
                                                                        Farrell, unpublishe
Decreased stroke       Perfused heart         Oncorhynchus mykiss     Bjenning and
 force                                                                  Holmgren (1989)
Zontraction            Coronary vasculature   Raja rhina              Bjenning et al. (1991)
324                          STEFAN NILSSON AND SUSANNE H O L M G R E N

and S. Holmgren, unpublished). Blood vessels of the cod appear to be
sparsely innervated by bombesin fibers: bundles of bombesin-
immunoreactive fibers run along arteries on the surface of the gut wall,
but very few fibers seem to innervate the vessels themselves (S.
Holmgren, unpublished).
   T h e results available give little support for a general theory on the
effect of bombesin-like peptides on fish circulation. The immunohisto-
chemical data suggest a variation in innervation between species and
between cardiovascular beds. It is possible that bombesin, in some
species such as Squalus acanthias, is involved in the redistribution of
blood between different vascular beds. This has been observed in the
crocodiles, Caiman crocodylus and Crocodylus porosus, where the
flow to the gut increased and the flow to the lung decreased due to
changes in vascular resistance after bombesin administration
(Holmgren et al., 1989).In mammals, nerve fibers showing bornbesin/
GRP-like immunoreactivity supply the brain vasculature and lung
vessels, but no vasomotor effects could be demonstrated in isolated
pial arteries (Uddman et al., 1983,1984), and it has been suggested that
the nerves are sensory or that the major function of bombesin/GRP in
the perivascular nerves is that of a modulator. Indeed, in the rainbow
trout and cod, bombesin potentiates the effect of acetylcholine on gut
wall smooth muscle (Thorndyke and Holmgren, 1990).

C. Neuropeptide Y
    Immunohistochemical neuropeptide Y (NPY) studies in fish sug-
gest a striking species variation in the cardiovascular innervation. Thus
a comparative study of three species of elasmobranchs, Squalus
acanthias, Raja erinacea, and Raja radiata, and of two teleosts, the cod
and the rainbow trout, revealed NPY-like immunoreactivity in the two
Raja species only, although in all these species, as well as in several
other teleost and Raja species, the same antisera reveal immunoreac-
tive fibers innervating the gut smooth muscle (Bjenning and
Holmgren, 1988; Burkhardt-Holm and Holmgren, 1989; Bjenning et
al., 1989). It is, however, clearly premature to conclude that skates are
unique among fish in the possession of cardiovascular NPY-containing
    The amino acid sequences of NPY isolated from the fish pancreas
show good homology with mammalian NPY, especially in the C-
terminal region. Interestingly, the primary structure has been strongly
conserved among skate (Raja rhina), ganoids (Lepisosteus spatula,
Amia calua), and salmon and eel NPY, and (in the C-terminal region)
5.   PURINES, 5-HT, AND NEUROPEPTIDES                                325

between these species and mammals, while the variation among tele-
ost appears much larger (Andrews et al., 1985; Conlon et al., 1986a,
1991;Kimmel et al., 1986; Pollock et al., 1987). No neuronally derived
NPY from fish has been sequenced to date, but immunoreactive mate-
rial from the brain of the anglerfish, Lophius americanus, shows even
closer similarity to mammalian NPY than fish pancreatic NPY (aPY) in
elution and radioimmunoassay studies (Andrews et a/., 1985; Noe et
al., 1986, 1989). There is as yet no information on the characteristics of
NPY-like peptides in cardiovascular nerves from fish, or whether or not
different forms are expressed depending on which organ is innervated.
    In skates, the perivascular fibers form a plexus in the adventitiome-
dial border. This plexus is especially dense in systemic vessels to the
gut; less dense in the conus arteriosus, the coronary vessels, the gill
arteries, the dorsal aorta, vessels within the gut wall, and the portal
vein; and sparse in the ducts of Cuvier. In the heart, the sinus venosus
and the atrium receive a sparse innervation, while the ventricle is
moderately innervated in Raja erinacea but devoid of fibers in Raja
radiata (Bjenning et al., 1989).
     Neuropeptide Y often (but not always) coexists with adrenaline/
noradrenaline in perivascular nerves in mammals, amphibians, and
reptiles (Gibbins et al., 1988; Morris, 1989). In Raja radiata, adrener-
gic nerves innervate gut arteries and arterioles and to some extent
coronary vessels, while the heart, larger arteries, and veins are devoid
of such nerves (Bjenning et al., 1989). The distribution thus agrees
with part of the NPY-immunoreactive nerves, making coexistence a
     In Squalus acanthias in vivo, cardiac output and coeliac artery Bow
increased in response to NPY; both the overall systemic vascular resis-
tance and the vascular resistance of the coeliac vascular bed decrease,
which suggests an inhibitory effect on the vascular smooth muscle
(Table VI). The response is probably independent of adrenergic mech-
anisms, since injected adrenaline or noradrenaline increase total sys-
temic and coeliac vascular resistance (S. Holmgren, M. Axelsson, and
A. P. Farrell, unpublished).
     himunohistochemical studies have failed to demonstrate cardio-
vascular nerves containing NPY in Squalus acanthias (see earlier).
However, NPY-like peptides are common in the fish pancreas (see
earlier), and a humoral action of these peptides on the cardiovascular
system is possible.
     Experiments performed on isolated coronary arteries from the long-
nose skate, Raja rhina, show that NPY only occasionally produces an
effect of its own on these vessels, while the amplitude of the response
326                                  STEFAN NILSSON AND SUSANNE HOLMGREN

                                     Table VI
        Effects of Neuropeptide Y (NPY) on the Cardiovascular System in Fish

      Effect          Organ/tissue            Species              Reference

Decreased         Overall systemic       Squalus acanthias S . Holmgren, M.
 vascular                                                     Axelsson, and A. P.
 resistance                                                   Farrell, unpublished
                  Coeliac vasculature    Squalus acanthias S. Holmgren, M.
                                                              Axelsson, and A. P.
                                                               Farrell, unpublished
Tachycardia       Heart                  Squalus acanthias S . Holmgren, M.
                                                              Axelsson and A. P.
                                                              Farrell, unpublished
Potentiates     Coronary vasculature     Raja rhina         C. Bjenning and A. P.
  noradrenaline                                               Farrell, unpublished

to noradrenaline was significantly enhanced in the presence of NPY
(C. Bjenning and A. P. Farrell, unpublished). This has also been seen
in isolated canine coronary arteries, while vascular resistance in the
coronary vascular bed of the whole perfused heart increased by NPY
alone, suggesting a different effect in the more peripheral parts of the
vessels (Macho et al., 1989).The potentiating effect of NPY on adrener-
gic vasoconstriction has been reported from several studies in mam-
mals (see Polak and Bloom, 1988), and NPY coexists with catechol-
amines in sympathetic perivascular neurons of most vertebrate species
examined, including the toad, Bufo marinus (Morris et a,?.,1986; Mor-
ris, 1989).
    Taken together, these results point to some interesting features:
   1. It is evident that the interaction between NPY and catechol-
      amines occurs early in the vertebrate lineage.
   2. The interaction of NPY with catecholamines appears indepen-
      dent of the effect of NPY alone, which may vary between tissues
      and species.
   3. Regional differences in NPY-mechanisms may occur along the
      vascular tree.

D. Somatostatin
   Somatostatins of various lengths are widely distributed in verte-
brate tissues: as a neuropeptide in the CNS and in endocrine cells in
the gut and pancreas. Somatostatin also occurs in peripheral nerves,
5.   PURINES, 5-HT, AND NEUROPEPTIDES                                 327
but the presence in cardiovascular nerves is not widespread
(Dahlstrom et al., 1988; Morris, 1989; Plisetskaya, 1989). In general,
somatostatin has inhibitory effects on secretory events. In the toad,
Bufo marinus, somatostatin is present in vagal postganglionic neurons
and has inhibitory effects on the heart, supporting the cholinergic
effects (Campbell et al., 1982).
    Dense plexuses of nerve fibers that show somatostatin-like immu-
noreactivity are present in the walls of the coeliac and mesenteric
arteries of the spiny dogfish, Squalus acanthias (Holmgren and
Nilsson, 1983a). In the cod in vivo, somatostatin had inconsistent and
usually weak effects on heart rate and on flow in the ventral aorta, the
coeliac artery and the mesenteric artery, and on dorsal aortic pressure.
The ventral aortic pressure increased in 50% of the tested fish (S.
Holmgren and M. Axelsson, unpublished).

E. Substance P
    Substance P was the first peptide to be recognized of the large
tachykinin family with the common C-terminal sequence Phe-X-Gly-
Leu-Met-NHZ (Euler and Gaddum, 1931). Tachykinins were demon-
strated in fish at a relatively early stage (Euler and Ostlund, 1956), and
a number of fish tachykinins have been sequenced since (Conlon et al.,
1986b, 1990; Conlon and Thim, 1988).
    The teleost gut is densely innervated by substance P-immunoreac-
tive fibers (Jensen, 1989), but in the cod the innervation of the cardio-
vascular system is sparse with only few blood vessels in the gut wall
innervated by substance P-immunoreactive fibers. However, endo-
crine cells containing tachykinin-like peptides are common in most
fish species investigated, and, in addition, transmitter overflow during
nerve activity may affect the vasculature of the stomach ( Jensen, 1989;
Jensen and Holmgren, 1991).
    Numerous mammalian studies have demonstrated a vasodilator
effect of substance P, and studies in the cod and the spiny dogfish
demonstrate a similar situation (Jensen et al., 1991; S. Holmgren, M.
Axelsson, and A. P. Farrell, unpublished; Table VII). However, in
contrast to the effects of substance P in mammals, heart rate remains
unaffected in these two fish species.
    In Squalus acanthias, substance P reduced the overall systemic
vascular resistance; the effect was particularly evident in the vascular
bed of the coeliac artery. Both cardiac output and coeliac artery blood
flow increased, while blood pressure decreased slightly (S. Holmgren
and M. Axelsson, unpublished).
    In vivo injections of substance P in the cod decreased the vascular
328                                S T E F A N NILSSON AND SUSANNE H O L M G R E N

                                      Table VII
             Effects of Substance P on the Cardiovascular System in Fish

  Effect           Orgadtissue               Species                Reference

Decreased    Overall systemic           Squalus ucanthius    S. Holmgren, M.
vascular                                                       Axelsson, and A. P.
resistance                                                     Farrell, unpublished
             Coeliac vasculature        Squalus acanthias    S. Holmgren, M.
                                                               Axelsson, and A. P.
                                                                Farrell, unpublished
              Coeliac vasculature       Gudus morhua         Jensen et al. (1991)
              Mesenteric vasculature    Gadus morhua         Jensen et al. (1991)

resistance of the coeliac and mesenteric arteries with very small effects
on dorsal or ventral aortic blood pressures (Fig. 6). The flow in the
coeliac artery reached a single peak, and then declined to its basic
value, while the flow in the mesenteric artery described a three-phasic
pattern. After an initial increase in flow, the flow decreased to its initial
value, or below, and then increased again. Blockade with atropine
abolished the phase of flow decrease, leaving an increase in flow
roughly agreeing in time with the three phases prior to blockade. The
transient flow decrease occurred even during continuous exposure to
substance P and after vagotomy and was not caused by mechanical
obstruction of the vessel walls due to contractile activity in the stom-
ach. It thus appears that substance P somehow triggers a local choliner-
gically mediated vasoconstrictor reflex in the vascular bed perfused by
the mesenteric artery, but the physiological significance of this re-
mains to be elucidated (Jensen et al., 1991).
    Ingestion of food leads to an increase in blood flow to the gut
(mammals, Fara, 1984; crocodiles, M. Axelsson, unpublished; teleost
fish, Axelsson et al., 1989), and in the dog there is a postprandial
increase in plasma levels of a substance P-like peptide, suggesting an
involvement of a tachykinin in the postprandial hyperemia. Whether
or not this is also the case in fish remains to be elucidated.

F. Galanin
   Galanin, a 29-amino acid neuropeptide, was first isolated and de-
scribed by Tatemoto et al. (1983), and its actions in mammals have
mostly been related to pancreatic functions (Plisetskaya, 1989).
5.    PURINES, 5-HT, AND NEUROPEPTIDES                                              329

     k Pa

             ‘] A


     k Pa

     k h
             ‘1                                               --

                   A                            A         A
                   Substance P                 Atropine   Substance P           2 rnin

    Fig. 6. Effects of substance P injection in wioo in the cod (Gadus rnorhua) on the
intraluminal pressure in the stomach (PSTOM), mesenteric artery blood flow ( F i t * ) ,
coeliac artery blood flow (FCA),dorsal and ventral aortic blood pressure ( P D A and PvA,
respectively), and heart rate (HR). Note triphasic response in FMA,
                                                                  which is changed to a
simple flow increase after atropine treatment.

However, studies in mammals (e.g., Kummer, 1987; Morris et al.,
1992),toads (Morris et al., 1989),and lizards (Gibbins et al., 1989)show
the presence of galanin in perivascular, sympathetic nerves, often in
co-existence with NPY. In the cod, Gadus morhua, galanin-like immu-
noreactivity is present in perivascular nerves innervating arterial
branches on the gut surface (Fig. 7), and galanin stimulates strip prepa-
rations of the coeliac and mesenteric arteries by a direct action on the
smooth muscle (P. Karila and S. Holmgren, unpublished results; Fig.
7). Little is known of the effects of galanin in the cardiovascular system
of vertebrates other than mammals, but the results in the cod agree
with the direct effects of galanin on smooth muscle of the gut reported
in the rat (Ekblad et al., 1985).
330                                 STEFAN NILSSON AND SUSANNE HOLXIGREN

 A *

      mN0   4
            1       t
      Galanin (nM) 10
                              100      300 10 min

    Fig. 7. Contractile effect of galanin on isolated mesenteric artery strip preparation
from cod (Gadus morhua) (A), and galanin-immunoreactive nerve fibers running along a
small artery at the surface of the stomach (B) and forming a perivascular plexus around a
branching visceral artery (C).

G. Gastrin/Cholecystokinin
   Although there has been much discussion of the exact identity of
the gastrin/CCK-like peptide(s) present in fish, there is to date no
sequence analyses made of a native fish gastrin or CCK. With the
support of circumstantial evidence, evolutionary theories have been
put forward suggesting that caerulein is the most primitive variant, first
appearing in fish (Larsson and Rehfeld, 1977)or that ancestral gastrin/
CCK resembles CCK rather than gastrin (Crim and Vigna, 1983; Vigna,
1985). Whichever the case, it is clear from radioimmunoassay studies
combined with Sephadex gel filtration and ion-exchange fractionation
that multiple forms of gastrin/CCK exist in extracts from the fish gut
(Aldman et al., 1989; Jonsson, 1989).
   Gastrin/CCK has mainly been associated with digestive events,
and the few investigations dealing with their effects on the cardiovas-
cular system in mammals have mainly focused in the control of the
gastrointestinal blood flow. In our preliminary experiments with the
cod (Gadus morhua) in vivo, there appears to be a reduction of gut
blood flow after injection of sulfated CCK8 and caerulein ( J . Gun-
narsson, unpublished, Fig. 8). There was also a dramatic increase in
ventral aortic blood pressure in response to these peptides, without a
5.   PURINES, 5-HT, A N D NEUROPEPTIDES                                            331


     Fig. 8. Effect of sulfated caerulein (CAER-S, 1 pmollkg) on heart rate (fH, beats/
min), ventral aortic blood pressure (Pv,, kPa), mesenteric and coeliac artery blood flow
(F,,, and Fcoel, respectively, both shown as uncorrected AkHz doppler shift) in the cod
(Gadus rnorhua). Note marked increase in ventral aortic blood pressure, concomitant
with the (probably barostatic reflexogenic) decrease in heart rate and reduction in gut
blood flow. [Courtesy of J . Gunnarsson, unpublished.]

significant effect on the dorsal aortic pressure ( J . Jensen, M. Axelsson,
and S. Holmgren, unpublished). In addition, CCK8-S caused a marked
vasoconstriction of the branchial vasculature in the isolated-perfused
head preparation from the cod (Fig. 9; L. Sundin, unpublished). This
suggests an effect of the gastrin/CCK-like peptides on the gill vascula-
ture, a conclusion that was supported by immunohistochemical find-
ings of CCK-immunoreactive nerve fibers along gill blood vessels in
the cod (S. Holmgren, unpublished).


   There is an increasing number of studies in mammals that indicate
that neurotransmitters may act both directly on the vascular smooth
muscle cells or indirectly by the release of endothelial factors. It is
332                                 STEFAN NILSSON A N D SUSANNE H O L M G R E N

                            8 7

      Fig. 9. Effect of a bolus injection of sulfated CCK-8 (0.5 nmol) on inflow pressure
   kPa), efferent arterial outflow (Fa;dropsimin), and inferior jugular vein outflow (F”;
dropslmin) in the gill apparatus from the cod (Gadus rnorhua) perfused at constant flow
from a pulsatile pump. Note marked vasoconstriction, reflected as both an increase in
inflow counter-pressure and a reduction in both flows. [Courtesy of L. Sundin, unpub-

important to bear in mind that the effect produced by a certain amine or
peptide may be quite different depending on the route of administra-
tion of the substance during an experiment (Cocks and Angus, 1983;
D’Orleans-Juste et al., 1985; Daly and Hieble, 1987; Burnstock, 1988).
    In mammals, the endothelium-related compounds include EDRF,
later identified as nitric oxide (NO) (Ignarro et al., 1986; Palmer et al.,
1987; Mione et al., 1990; Burnstock, 1990).Whether or not EDRF also
occurs in fish is not clear, but nitroglycerine (which is broken down to
nitrous oxide) causes relaxation of coronary artery rings (Small et al.,
    A second substance that has major effects on fish blood vessels is
endothelin. Endothelin-1, a 21-amino acid peptide that was originally
isolated from porcine endothelial cells, contracts blood vessels from
several vertebrates including the catfish (Arniurus rnelas) and rainbow
trout (Oncorhynchus mykiss) (Poder et al., 1991; Olson et al., 1991).
Olson et al. (1991) also observed a transient decrease in dorsal aortic
blood pressure after injection of 500 ngekg-’ endothlin-1 into rainbow
trout, while higher doses (1500ngskg-’) produced a triphasic (pressor/
depressor/pressor) response. They concluded that the rainbow trout
vasculature “is exquisitely sensitive” to endothelin-1 and suggested
that the physiological expression of the peptide has been highly con-
served during the course of vertebrate evolution.
5. PURINES,    5-HT, AND NEUROPEPTIDES                                               333


    Our own work on cardiovascular control in fish is currently supported by the
Swedish Natural Science Research Council and the Swedish Forestry and Agriculture
Research Council. We thank Kersti Lundin, Michael Axelsson, Jorgen Jensen, Regina
Fritsche, Jonas Gunnarsson, and Lena Sundin for letting us use and quote their, as yet,
unpublished material.


Aldman, G., and Holmgren, S. (1987). Control of gallbladder motility in the rainbow
    trout, Salmo gairdneri. Fish Physiol. Biochem. 4, 143-155.
Aldman, G . , Jonsson, A. C., Jensen, J., and Holmgren, S. (1989). GastrinICCK-like
    peptides in the spiny dogfish, Squalus acanthias: Concentrations and actions in the
    gut. Comp. Biochem. Physiol. 92C, 103-109.
Andrews, P. C., Hawke, D., Shively, J. E., and Dixon, J. E. (1985). A nonamidated
    peptide homologous to porcine peptide YY and neuropeptide YY. Endocrinology
Axelsson, M., Farrell, A. P., and Nilsson, S. (1990). Effect of hypoxia and drugs on the
    cardiovascular dynamics of the Atlantic hagfish, Myxine glutinosa. J . Exp. Biol. 151,
Axelsson, M., Driedzic, W. R., Farrell, A. P., and Nilsson, S. (1989).Regulation ofcardiac
    output and gut blood flow in the sea raven, Hemitripterus americanus. Fish Physiol.
    Biochem. 6,315-326.
Bailly, Y. (1983). “Recherches sur Quelques Innervations d e la Branchie des Poissons.”
    Th&sede 3eme cycle de cytologie, UniversitC Pierre et Marie Curie, Paris.
Bailly, Y., and Dunel-Erb, S. (1986). The sphincter of the efferent filament artery in
    teleost gills: I. Structure and parasympathetic innervati0n.J. Morphol. 187,219-237.
Bailly, Y., Dunel-Erb, S., Geffard, M., and Laurent, P. (1989).The vascular and epithelial
    serotonergic innervation of the actinopterygian gill filament with special reference
    ttr the trout, Salmo gairdneri. Cell Tissue Res. 258,349-363.
Baumgarten, H. G., Bjorklund, A., Lachenmayer, L., Nobin, A., and Rosengren, E.
    (1973). Evidence for the existence of serotonin-, dopamine- and noradrenaline-
    containing neurons in the gut of Lampetra puuiatilis. 2. Zellforsch. 141,33-54.
Beach, M. A,, McVean, A,, Roberts, M. G., and Thorndyke, M. C. (1988). The effects of
    bombesin on the feeding of fish. Neurosci. Lett. 32,46.
Beringer, T., and Hadek, H.(1973).Ultrastructure of sinus venosus innervation in Petro-
    myzon marinus. J . Ultrastruct. Res. 42,312-323.
Bjenning, C., and Holmgren, S. (1988). Neuropeptides in the fish gut. A study of evolu-
    tionary trends. Histochemistry 88, 155-163.
Bjenning, C., and Holmgren, S. (1989).The distribution and effect of bombesin in the
    rainbow trout heart. “Strategies of Physiological Adaptations” The Kjell Johansen
    Commemorative Symposium, Copenhagen, Denmark.
Bjenning, C., Driedzic, W., and Holmgren, S. (1989).Neuropeptide Y-like immunoreac-
    tivity in the cardiovascular nerve plexus of the elasmobranchs Raja erinacea and
    Raja radiata. Cell Tissue Res. 255,481-486.
334                                 STEFAN NILSSON AND SUSANNE HOLkfGREN

Bjenning, C., Farrell, A. P., and Holmgren, S. (1991). Bombesin-like immunoreactivity
    in skates, and the in uitro effect of bombesin in the coronary vascular system of the
    longnose skate, Raja rhina. Regul. Pept. 35,207-219.
Bjenning, C., Jonsson, A.-C., and Holmgren, S. (1990). Bombesin-like immunoreactive
    material in the gut, and the effect of bombesin on the stomach circulatory system of
    an elasmobranch fish, Squalus acanthias. Regul. Pept. 28,57-69.
Bolis, L., Mandolfino, M., Marino, D., and Rankin, J. C. (1984). Vascular actions of
    vasoactive intestinal poly-peptide and beta endorphin in isolated perfused gills of
    the brown trout, Salmo trutta. Mol. Physiol. 5,221-226.
Burkhardt-Holm, P., and Holmgren, S. (1989). A comparative study of neuropeptides in
    the intestine of two stomachless teleosts (Poecilia reticulata, Leuciscus idus mela-
    notus) under conditions of feeding and starvation. Cell Tissue Res. 255,245-254.
Burnstock, G. (1969). Evolution of the autonomic innervation of visceral and cardiovas-
    cular systems in vertebrates. Pharmacol. Reu. 21,247-324.
Burnstock, G. (1972). Purinergic nerves. Pharmacol. Rev. 24,509-581.
Burnstock, G. (1975). Comparative studies of purinergic nerves. ]. E r p . Zool. 194, 103-
Burnstock, G. (1976).Purinergic receptors.]. Theoret. Biol. 62,491-503.
Burnstock, G. (1978). A basis for distinguishing two types of purinergic receptor. I n
    “Cell Membrane Receptors for Drugs and Hormones: A Multidisciplinary Ap-
    proach” (L. Bolis and R. W. Straub, eds.), pp. 107-118. Raven Press, New York.
Burnstock, G. (1988). Regulation of local blood flow by neurohumoral substances re-
    leased from perivascular nerves and endothelial cells. Acta Physiol. Scand. 133,
Burnstock, G. (1990). Noradrenaline and ATP as cotransmitters in sympathetic nerves.
    Neurochem. Znt. 17,357-368.
Burnstock, G. (1990).Local mechanisms ofblood flow control by perivascular nerves and
    endothelium. I. Hypertension 8, S95-Sl06.
Burnstock, G., and Griffith, S. G. (1988).“Non-adrenergic Innervation of Blood Vessels.”
    CRC Press, Boca Raton, Florida.
Burnstock, G., Campbell, G., Bennett, M., and Holman, M. (1963). Inhibition of the
    smooth muscle of the taenia coli. Nature (London)200,581-582.
Burnstock, G., Campbell, G., Bennett, M., and Holman, M. (1964). Innervation of the
    guinea-pig taenia coli: Are there intrinsic inhibitory nerves which are distinct from
    sympathetic nerves? Znt. J . Neuropharmacol. 3, 163-166.
Campbell, G. (1970).Autonomic nervous system. I n “Fish Physiology” (W. S. Hoar and
    D. J. Randall, eds.), Vol. IV, pp. 109-132. Academic Press, New York, London.
Campbell, G., Gibbins, I. L., Morris, J. L., Furness, J. B., Costa, M., Oliver, J. R.,
    Beardsley, A. M., and Murphy, R. (1982).Somatostatin is contained in and released
    from cholinergic nerves in the heart of the toad Bufo marinus. Neuroscience 7,
Caravita, S., and Coscia, L. (1966). Les cellules chromaffines du coeur de la Iamproie
    (Lampetra zanandreai). Etude au microscope 6lktronique avant et apres un
    traitement 31 la reserpine. Arch. B i d . 77,723-753.
Cimini, V., Van Noorden, S., Giordano-Lanza, G., Nardini, V., McGregor, G. P., Bloom,
    S. R., and Polak, J. M. (1985). Neuropeptides and 5-HT immunoreactivity in the
    gastric nerves of the dogfish (Scyliorhinus stellaris). Peptides 6(Suppl3), 373-377.
Cocks, T. M . , and Angus, J . A. (1983). Endothelium-dependent relaxation of coronary
    arteries by noradrenaline and serotonin. Nature 305,627-630.
Cohen, S., Rotmensch, H. H., Rubinstein, R., and Lass, Y . (1981). Lack of uptake or
5.   PURINES, 5-HT, AND NEUROPEPTIDES                                                335

    degradation of adenosine in the termination of its action in the beating carp Cypri-
    nus carpio atrium. Biochem. Pharmacol. 30,890-893.
Colin, D. A., and Leray, C. (1981).Vasoactivities ofadenosine analogs in trout gill Salmo
   gairdneri. Biochem. Pharmacol. 30,2971-2978.
Colin, D. A,, Kirsch, R., and Leray, C. (1979).Hemodynamic effects of adenosine on gills
    of the trout Salmo gairdneri.J. Comp. Physiol. 130,325-330.
Conlon, J . M., and Thim, L. (1988).Isolation of the tachykinin, des[Ser’ Pro2]scylior-
    hinin I1 from the intestine of the ray, Torpedo marmorata. Gen. Comp. Endocrinol.
Conlon, J . M., Bjenning, C., Moon, T. W., Youson, J. H., and Thim, L. (1991).Neropep-
    tide Y-related peptides from the pancreas of teleostean (eel), holostean (bowfin) and
    elasmobranch (skate) fish. Peptides (in press).
Conlon, J. M., Deacon, C. F., O’Toole, L., and Thim, L. (1986b). Scyliorhinin I and 11:
    Two novel tachykinins from the dogfish gut. FEBS Lett. 200,111-116.
Conlon, J . M., Henderson, I. W., and Thim, L. (1987).Gastrin-releasing peptide from the
    intestine of the elasmobranch fish, Scyliorhinus canicula (common dogfish). Gen.
    Comp. Endocrinol. 68,415-420.
Conlon, J. M., O’Harte, F., and Kah, 0. (1990). Carassin: A tachykinin from the brain of
    the goldfish that is structurally related to neuropeptide-gamma. Abstract from the
    meeting: Substance P and related peptides: Cellular and molecular physiology.
    Worchester Mass., July 18-21.
Conlon, J. M., Schmidt, W. E., Gallwitz, B., Falkmer, S., and Thim, L. (1986a). Charac-
    terization of an amidated form of pancreatic polypeptide from the daddy sculpin
    (Cottus scorpius).Regul. Pept. 16,261-268.
Crim, J. W., and Vigna, S. R. (1983). Brain, gut and skin peptide hormones in lower
    vertebrates. Am. Zool. 23,621-638.
Dahlstrom, A., Nilsson, O., Lundgren, O., and Ahlman, H. (1988).Nonadrenergic, non-
    cholinergic innervation of gastrointestinal vessels, morphological and physiological
    aspects. In “Nonadrenergic Innervation of Blood Vessels” (G. Burnstock and S. G.
    Griffith, eds.), Vol. 11. CRC Press, Boca Raton, Florida.
Daly, R. N., and Hieble, J. P. (1987). Neuropeptide Y modulates adrenergic neuro-
    transmission by an endothelium dependent mechanism. Eur. J. Pharmacol. 138,
Dimaline, R., and Thorndyke, M. C. (1986).Purification and characterisation ofVIP from
    two species of dogfish. Peptides 7,21-25.
Dimaline, R., Young, J., Thwaites, D. T., Lee, C. M., Shuttleworth, T. J., andThorndyke,
    M. C. (1987).A novel vasoactive intestinal peptide (VIP) from elasmobranch intes-
    tine has full affinity for mammalian pancreatic VIP receptors. Biochem. Biophys.
    Acta 930,97-100.
Donald, J. (1984).Adrenergic innervation of the gills of brown and rainbow trout, Salmo
    trutta and S . gairdneri.J . Morphol. 182,307-316.
Donald, J . A. (1987). Comparative study of the adrenergic innervation of the teleost gill.
   J . Morphol. 193,63-73.
Donald, J . , and Campbell. G. (1982). A comparative study of the adrenergic innervation
    of the teleost heart.J. Comp. Physiol. 147,85-91.
D’Orleans-Juste, P., Dion, S., Mizrahi, J., and Regoli, D. (1985). Effects ofpeptides and
    non-peptides on isolated arterial smooth muscle: Role of endothelium. Eur. J . Phar-
    macol. 114,9-21.
Dunel-Erb, S., and Bailly, Y. (1986). The sphincter of the efferent filament artery in
    teleost gills. 11. Sympathetic innervation. J. Morphol. 187,239-246.
336                                 STEFAN NILSSON AND SUSANNE HOLMGREN

Dunel-Erb, S., Bailly, Y., and Laurent, P. (1982). Neuroepithelial cells in fish gill pri-
    mary lamellae. J . Appl. Physiol. 53, 1342-1353.
Dunel-Erb, S., Bailly, Y., and Laurent, P. (1989).Neurons controlling the gill vasculature
    in five species of teleosts. Cell Tissue Res. 255,567-573.
Edvinsson, L., and Uddman, R. (1988).Vasoactive intestinal polypeptide (VIP): A pu-
    tative neurotransmitter in the cardiovascular system. In “Nonadrenergic Inner-
    vation of Blood Vessels” (G. Burnstock and S. G. Griffith, eds.), Vol. I, pp. 101-125.
    CRC Press, Inc., Boca Raton, Florida.
Ekblad, E., Hikanson, R., Sundler, F., and Wahlestedt, C. (1985).Galanin neuromodula-
    tory and direct contractile effects on smooth muscle preparations. Br. J . Pharmacol.
El-Salhy, M., Wilander, E., and Lundqvist, M. (1985).Comparative studies of serotonin-
    like immunoreactive cells in the digestive tract of vertebrates. Biomed. Res. 6,
Erspamer, V. (1954). Pharmacology of indolealkylamines. Pharmacol. Reu. 6,425-487.
Euler, von U. S., and Gaddum, J. H. (1931).An unidentified depressor substance in
    certain tissue extracts. J . Physiol. 7 2 , 7 4 4 7 .
Euler, von U. S., and Ostlund, E. (1956).Occurrence of a substance P-like polypeptide in
    fish intestine and brain. Br. J . Pharmacol. 11,323-325.
Fange, R., Johnels, A. G., and Enger, P. S. (1963). The autonomic nervous system. I n
    “The Biology of Myxine” (A. Brodal and R. Fange, eds.), pp. 124-136. Univ. For-
    laget, Oslo.
Fara, J . W. (1984).Postprandial mesenteric hyperemia. In “Physiology of the Intestinal
    Circulation” (A. Shepherd and D. N. Granger, eds.), pp. 99-106. Raven Press, New
Farrell, A. P. (1984).A review of cardiac performance in the teleost heart: Intrinsic and
    humoral regulation. Can.J . 2001.62,523-536.
Farrell, A. P., and Davie, P. S. (1991a). Coronary artery reactivity in the maco shark,
    Isurus oryrinchus. Can.J . Zool. 69,375-379.
Farrell, A. P., and Davie, P. S. (1991b). Coronary vascular reactivity in the skate, Raja
    nasuta. Comp. Biochem. Physiol. 99c, 555-560.
Gannon, B. J., and Burnstock, G. (1969). Excitatory adrenergic innervation of the fish
    heart. Comp. Biochem. Physiol. 29,765-773.
Gibbins, I. L., Morris, J. L., Furness, J. B., and Costa, M. (1988).Innervation of systemic
    blood vessels. In “Non-adrenergic Innervation of Blood Vessels” (G. Burnstock and
    S. Grifith, eds.), Vol. 11, pp. 1-36. CRC Press, Boca Raton, Florida.
Gibbins, I. L., Morris, J. L., and Holmgren, S. (1989).Distribution ofneuropeptide Y and
    galanin in cardiovascular sympathetic neurons of Australian lizards. In “Proc. XXXI
     ICPS Meeting,” Helsinki.
Govyrin, V. A. (1977). Development of vasomotor adrenergic innervation in onto- and
     phylogenesis. J . Euol. Biochem. Physiol. 13, 614-620.
Govyrin, V. A., and Leont’eva, G. R. (1965). Distribution of catecholamines in myocar-
     dium of vertebrates./. Eool. Biochem. Physiol. 1, 38-44.
Greeff, K., Kasperat, H., and Oswald, W. (1962). Paradoxe Wirkungen der elektrischen
    Vagusreizung am isolierten Magen ond Herzvorhofpraparat des Meerschweinchens
     sowie deren Beieinflussung durch Ganglienblocker, Sympathicolytica, Reserpin
     und Cocain. Naunyn-Schmiedebergs Arch. E x p . Pathol. Pharmakol. 243,528-545.
Griffith, S. G. (1988).Serotonin (5-HT) as a neurotransmitter in blood vessels. I n “Non-
     adrenergic Innervation ofBlood Vessels” (G. Burnstock and S. G. Griffith, eds.), Vol.
     I, pp. 27-40. CRC Press, Inc., Boca Raton, Florida.
5.   PURINES, 5-HT, A N D NEUROPEPTIDES                                              337

Holder, F. C., Vincent, B., Ristori, M. T., and Laurent, P. (1983).Vascular perfusion ofan
   intestinal loop in the catfish lctalurus melas. Demonstration of the vasoactive effects
   of the mammalian vasoactive intestinal peptide and gastro-intestinal extracts from
   teleost fish. C . R . Seances Acad. Sci. Ser. Ill Sci. Vie 296,783-788.
Holmgren, S. (1977).Regulation of the heart of a teleost, Gadus morhua, by autonomic
   nerves and circulating catecholamines. Acta Physiol. Scand. 99,62-74.
Holmgren, S. (1983). The effect of putative non-adrenergic, non-cholinergic autonomic
   transmitters on isolated strips from the stomach of the rainbow trout, Salmo
   gairdneri. Comp. Biochem. Physiol. 74C, 229-238.
Holmgren, S., and Jonsson, A. C. (1988).Occurrence and effects on motility ofbombesin-
   related peptides in the gastrointestinal tract of the Atlantic cod, Gadus morhua.
   Comp. Biochem. Physiol. 89C, 249-256.
Holmgren, S., and Nilsson, S. (1983a). Bombesin-, gastrin/CCK-, 5-hydoxytryptamine-,
   neurotensin-, somatostatin-, and VIP-like immunoreactivity and catecholamine flu-
   orescence in the gut of the elasmobranch, Squalus acanthias. Cell Tissue Res. 234,
Holmgren, S., and Nilsson, S. (1983b).VIP-, bombesin-, and neurotensin-like immuno-
   reactivity in neurons of the gut of the holostean fish, Lepisosteus platyrhincus. Acta
    ZOO^. (Stockh.)64,25-32.
Holmgren, S., Axelsson, M., Jensen, J., Aldman, G., Sundell, K., and Jonsson, A. C.
    (1989). Bombesin-like immunoreactivity and the effect of bombesin in the gut,
    circulatory system and lung of the caiman, Caiman crocodylus crocodylus, and the
    crocodile, Crocodylus porosus. J . E x p . Biol. 48,261-271.
Holnigren, S., Vaillant, C., and Dimaline, R. (1982). VIP-, substance P-, gastrin/CCK-,
    bombesin-, somatostatin-, and glucagon-like immunoreactivities in the gut of the
    rainbow trout, Salmo gairdneri. Cell Tissue Res. 223, 141-153.
Holstein, B., and Humphrey, C. S. (1980). Stimulation of gastric acid secretion and
    suppression of VIP-like immunoreactivity by bombesin in the atlantic codfish,
    Gadus morhua. Acta Physiol. Scand. 109,217-223.
Honma, S. (1970).Presence ofmonoaminergic neurons in the spinal cord and intestine of
    the lamprey. Arch. Histol. Jap. 32,383-393.
Ignarro, L. J., Byrns, R. E., and Wood, K. S. (1986).Pharmacological and biochemical
    properties of endothelium-derived relaxing factor (EDRF): Evidence that it is
    closely related to nitric oxide (NO) radical. Circulation 74 (Suppl. II), 287.
Jensen, J. (1989). Substance P and other tachykinins. In “The Comparative Physiology of
    Regulatory Peptides” (S. Holmgren, ed.), pp. 130-149. Chapman and Hall, London,
    New York.
Jensen, J., and Holmgren, S . (1985).Neurotransmitters in the intestine of the Atlantic
    cod, Cadus morhua. Comp. Biochem. Physiol. 82C, 81-89.
Jensen, J., and Holmgren, S. (1991).Tachykinins and intestinal motility in different fish
    groups. Gen. Comp. Endocrinol. (in press.)
Jensen, J., Axelsson, M., and Holmgren, S. (1991). Effects of substance P and vasoactive
    intestinal polypeptide on gastrointestinal blood flow in the Atlantic cod Gadus
    morhua. J. E x p . Biol. 156,361-373.
Jonsson, A.-C. (1989).Gastrin/cholecystokinin-related peptides-Comparative aspects.
    In “The Comparative Physiology of Regulatory Peptides” (S. Holmgren, ed.), pp.
    61-86. Chapman and Hall, London, New York.
JZrgensen, J. B., and Mustafa, T. (1980).Effect of hypoxia on carbohydrate metabolism in
    flounder Platichthys Jesus 2. High energy phosphate compounds and the role of
    glycolytic and gluconeogenetic enzymes. Comp. Biochem. Physiol. 67,249-256.
338                                 STEFAN NILSSON AND SUSANNE HOLMGREN

Katchen, M. S., Olson, K. R., and Wayne, C. (1976).Effects ofhistamine and serotonin on
    isolated perfused gill of rainbow trout (Salmo gairdneri). Fed. Proc. 35,528.
Kavaliers, M., and Hawkins, M. F. (1981).Bombesin alters behavioral thermoregulation
    in fish. Life Sci. 28, 1361-1364.
Kimmel, J. R., Plisetskaya, E. M., Pollock, H. G., Hamilton, J . W., Rouse, J. B., Ebner,
    K. E., and Rawitch, A. B. (1986). Structure of a peptide from coho salmon endocrine
    pancreas with homology to neuropeptide Y. Biochem. Biophys. Res. Commun. 141,
    1084- 1091.
Koke, J. R., and Anderson, D. R. (1986).Changes in metabolite levels and morphology of
    teleost ventricular myocytes due to hypoxia ischemia and metabolic inhibitors.
    Cytobios 45,97-108.
Kummer, W. (1987). Galanin- and neuropeptide Y-like immunoreactivities coexist in
    paravertebral sympathetic neurons of the cat. Neurosci. Lett. 78,127-131.
Larsson, L.-I., and Rehfeld, J. F. (1977).Evidence for a common evolutionary origin of
    gastrin and cholecystokinin. Nature 269,335-338.
Laurent, P. (1984).Gill internal morphology. In “Fish Physiology” (W. S. Hoar and D. J.
    Randall, eds.), Vol. X, pp. 73-183. Academic Press, Orlando.
Laurent, P., Holmgren, S., and Nilsson, S. (1983). Nervous and humoral control of the
    fish heart: Structure and function. Comp. Biochem. Physiol. 76A, 525-542.
Lennard, R., and Huddart, H. (1989). Purinergic modulation of cardiac activity in the
    flounder during hypoxic stress. J . Comp. Physiol. 159, 105-114.
Leont’eva, G. R. (1966). Distribution of catecholamines in blood vessel walls of cyclo-
    stomes, fishes, amphibians and reptiles. J . Evol. Biochem. Physiol. 2,31-36.
Lincoln, J., Loesch, A., and Burnstock, G. (1990). Localization of vasopressin, serotonin
    and angiotensin-I1 in endothelial cells ofthe renal and mesenteric arteries ofthe rat.
    Cell Tissue Res. 259,341-344.
Lundin, K. (1991). “The Teleost Swimbladder: A Study of the Non-adrenergic, Non-
    cholinergic Innervation.” Ph.D. Thesis, Department of Zoophysiology, University
    of Goteborg.
Lundin, K., and Holmgren, S. (1984).Vasointestinal polypeptide-like immunoreactivity
    and effects of VIP in the swimbladder of the cod, Gadus morhua. J . Comp. Physiol.
    154B, 627-633.
Lundin, K., and Holmgren, S. (1986). Non-adrenergic, non-cholinergic innervation of
    the urinary bladder of the Atlantic cod, Gadus morhua. Comp. Biochem. Physiol.
    84C, 315-323.
Lundin, K., and Holmgren, S. (1989). The occurrence and distribution of peptide- or
    5-HT-containing nerves in the swimbladder of four different species of teleosts,
    (Gadus morhua, Ctenolabrus rupestris, Anguilla anguilla, Salmo gairdneri). Cell
    Tissue Res. 257,641-647.
Lundin, K., Holmgren, S., and Nilsson, S. (1984). Peptidergic functions in the dogfish
    rectum. Acta Physiol. Scand. 121,46A.
Macho, P., Perez, R., Huidobro-Toro, J. P., and Domenech, R. J. (1989).Neuropeptide Y
    (NPY): A coronary vasoconstrictor and potentiator of catecholamine-induced coro-
    nary constriction. Eur. J . Pharmacol. 167,67-74.
Meghji, P., and Burnstock, G. (1984a).Actions of some autonomic agents on the heart of
    the trout (Salmo gairdneri) with emphasis on the effects of adenyl compounds.
     Comp. Biochem. Physiol. 78C, 69-75.
Meghji, P., and Burnstock, G. (1984b). The effect of adenyl compounds on the heart of
     the dogfish, Scyliorhinus canicula. Comp. Biochem. Physiol. 77C, 295-300.
Metcalfe, J. D., and Butler, P. J. (1984).On the nervous regulation ofgill blood flow in the
     dogfish Scyliorhinus canicula. J . E x p . Biol. 113,253-267.
5.   PURINES, 5-HT, AND NEUROPEPTIDES                                                   339

Mione, M. C., Ralevic, V., and Burnstock, G. (1990). Peptides and vasomotor mecha-
    nisms. Pharmacol. Ther. 46,429-468.
Morris, J. (1989).The cardiovascular system. In “The Comparative Physiology of Regu-
    latory Peptides” (S. Holmgren, ed.), pp, 273-307. Chapman and Hall, London.
Morris, J. L., Gibbins, I. L., Campbell, G., Murphy, R., Furness, J. B., and Costa, M.
    (1986). Innervation of the large arteries and heart of the toad (Bufo marinus) by
    adrenergic and peptide-containing neurons. Cell Tissue Res. 243, 171-184.
Morris, J. L., Gibbins, I. G., and Holmgren, S. (1992).Galanin is more common than NPY
    in vascular sympathetic neurons ofAustralian marsupials. Regul. Pept. 37, 101-109.
Morris, J. L., Gibbins, I. L., and Osborne, P. B. (1989). Galanin-like immunoreactivity in
    sympathetic and parasympathetic neurons of the toad Bufo marinus. Neurosci. Lett.
    102, 142-148.
Nicol, J. A. C. (1952). Autonomic nervous systems in lower chordates. Biol. Reo. 27,
Nielsen, K. E., and Gesser, H. (1984).Energy metabolism and intracellular pH in trout
    heart muscle under anoxia and different calcium ii concentrations.]. Comp. Physiol.
Nilsson, S. (1983). “Autonomic Nerve Function in the Vertebrates.” Springer-Verlag,
    Berlin, Heidelberg, New York.
Nilsson, S. (1984). Innervation and pharmacology of the gills. I n “Fish Physiology”
    (W. S. Hoar and D. J. Randall, eds.), Vol. XA, pp. 185-227. Academic Press, Orlando.
Nilsson, S. (1986).Control of gill blood flow. In “Fish Physiology: Recent Advances” (S.
    Nilsson and S. Holmgren, eds.), pp. 86-101. Croom Helm, Dover, New Hampshire.
Nilsson, S., and Holmgren, S. (1988).The autonomic nervous system of elasmobranchs:
    Structure and function. In “Physiology of Elasmobranch Fishes.” (T. J. Shuttle-
    worth, ed.), pp. 143-169. Springer-Verlag, Berlin.
Nilsson, S., and Pettersson, K. (1981). Sympathetic nervous control of blood flow in the
    gills of the Atlantic cod, Gadus morhua.]. Comp. Physiol. 144, 157-163.
Noe, B. D., McDonald, J. K., Greiner, F., Wood, J. G. and Andrews, P. C. (1986).
    Anglerfish islets contain NPY immunoreactive nerves and produce the NPY analog
    aPY. Peptides 7,147-154.
Noe, B. D., Milgram, S. L., Balasubramaniam, A,, Andrews, P. C., Calka, J., and
    McDonald, J. K. (1989). Localization and characterization of neuropeptide Y-like
    peptides in the brain and islet organ of the anglerfish (Lophius americanus). Cell
    Tissue Res. 257,303-:311.
Okafor, M . C. J., and Oduleye, S. 0.(1986). Hemodynamic effects ofpurinergic receptor
     stimulation in isolated fish Oreochromis niloticus gills. Actu Physiol. Hung. 67,
Olson, K. R., Duff, D. W., Farrell, A. P., Keen, J , , Kellogg, M. D., Kullman, D., and Villa,
    J . (1991). Cardiovascular effects of endothelin in trout. A m . J .Physiol. 260, H1214-
Ostlund, E., and Fange, R. (1962).Vasodilation by adrenaline and noradrenaline, and the
     effects of some other substances on perfused fish gills. Comp. Biochem. Physiol. 5 ,
Palmer, R. M. J., Ferrige, A. G., and Moncada, S. (1987).Nitric oxide release accounts for
    the biological activity of endothelium-derived relaxing factor. Nature 327,524-526.
Pettersson, K., and Nilsson, S. (1979).Nervous control of the branchial vascular resis-
     tance of the Atlantic cod, Gadus morhua.]. Comp. Physiol. 129, 179-183.
Plisetskaya, E. M. (1989). Pancreatic peptides. In “The Comparative Physiology
     of Regulatory Peptides” (S. Holmgren, ed.), pp. 174-202. Chapman and Hall,
340                                  STEFAN NILSSON AND SUSANNE HOLMGREN

Poder, T. C., Silberberg, S. D., and Rampe, D. (1991).Contraction of reptile, amphibian
    and fish blood vessels by endothelin-1. Can.J . Physiol. Pharmacol. 69,215-217.
Polak, J. M., and Bloom, S. R. (1988).Atrial natriuretic peptide (ANP), neuropeptide Y
    (NPY) and calcitonin gene-related peptide (CGRP) in the cardiovascular system of
    man and animals. In “Nonadrenergic Innervation of Blood Vessels” (G. Burnstock
    and S. G. Griffith, eds.), Vol. I, pp. 127-144. CRC Press, Inc., Boca Raton, Florida.
Pollock, H. G., Kimmel, J. R., Hamilton, J. W., Rouse, J. B., Ebner, K. E., Lance, V., and
    Ravitch, A. B. (1987). Isolation and structures of alligator gar (Lepisosteus spatulu)
    insulin and pancreatic polypeptide. Gen. Comp. Endocrinol. 67,375-382.
Rapport, M. M., Green, A. A., and Page, I. H. (1948).Serum vasoconstrictor (serotonin).
    IV. Isolation and characterization. J . Biol. Chem. 176, 1243-1251.
Reite, 0. B. (1969).The evolution of vascular smooth muscle responses to histamine and
    5-hydroxytryptamine. I. Occurrence and stimulatory actions in fish. Acta Physiol.
    Scand. 75,221-239.
Ristori, M.-T., and Laurent, P. (1977).Action ofhypoxia on the gill vascular system ofthe
    perfused isolated trout head. C . R . Seances Soc. Biol. 171,809-813.
Rotmensch, H. H., Cohen, S., Rubinstein, R., and Lass, Y. (1981).Effects ofadenosine in
    the isolated carp atrium. Isr. J . Med. Sci. 17,393.
Sakharov, D. A., and Salimova, N. B. (1980).Serotonin neurons in the peripheral nervous
    system of the larval lamprey, Lampetra planeri. A histochemical, microspectro-
    fluorimetric and ultrastructural study. Zool. Jahrb. Physiol. 84,231-239.
Small, S. A., Macdonald, C., and Farrell, A. P. (1990).Vascular reactivity of the coronary
    artery in rainbow trout Oncorhynchus mykiss. Am. J . Physiol. 258, R1402-R1410.
Small, S. A., and Farrell, A. P. (1990). Vascular reactivity of the coronary artery in
    steelhead trout Oncorhynchus mykiss. Comp. Biochem. Physiol. 97C, 59-63.
Smith, D. G. (1977). Sites of cholinergic vasoconstriction in trout gills. Am. J . Physiol.
    233, R222-R229.
Smith, D. G. (1978). Neural regulation of blood pressure in rainbow trout (Salnzo
    gairdneri). Can.J . 2001.56, 1678-1683.
Solomon, R., Taylor, M., Stoff, J . S., Silva, P., and Epstein, F. H. (1984).In uivo effect of
    volume expansion on rectal gland function. I. Humoral factors. Am. J . Physiol. 246,
Tagliafierro, G., Zaccone, G., Bonini, E., Faraldi, G., Farina, L., Fasulo, S., and Rossi,
    G. G. (1988).Bombesin-like immunoreactivity in the gastrointestinal tract of some
    lower vertebrates. Ann. N.Y. Acad. Sci. 547,458-460.
Tatemoto, K., Rokaeus, A. J., Jornvall, H., McDonald, T. J., and Mutt, V. (1983).
    Galanin-A novel biologically active peptide from porcine intestine. F E B S Lett.
Thomas, S., Belaud, A,, and Peyraud, C. (1979).Arguments for serotonergic adjustments
    in gill blood circulation in fish. ZRCS Medical Science; Cardiovascular System;
    E rioironmental Biology and Medicine; Experimental Animals; Physiology; Respiru-
    tory System 7,543.
Thorndyke, M., and Holmgren, S. (1990). Bombesin potentiates the effect of acetyl-
    choline on isolated strips of fish stomach. Regul. Pept. 30, 125-135.
Thorndyke, M. C., Reeve, J . R., and Vigna, S. R. (1990).Biological activity ofa bombesin-
    like peptide extracted from the intestine of the ratfish, Hydrolagus colliei. Comp.
    Biochem. Physiol. 96C, 135-140.
Thorndyke, M. C., Riddell, J. H., Thwaites, D. T., and Dimaline, R. (1989).Vasoactive
     intestinal polypeptide and its relatives-Biochemistry, distribution, and functions.
    Biol. Bull. 177, 183-186.
5.   PURINES, 5-HT, AND NEUROPEPTIDES                                             341

Thwaites, D. T., Young, J., Thorndyke, M. C., and Dimaline, R. (1989).The isolation and
    chemical characterization of a novel vasoactive intestinal peptide-related peptide
    from a teleost fish, the cod, Gadus morhua. Biochem. Biophys. Actu 999,217-220.
Tretjakoff, D. (1927). Das peripherische Nervensystem des Flupneunauges. 2. Wiss.
    Bull. 129,359-452.
Uddman, R., Edvinsson, L., Owman, C., and Sundler, F. (1983).Nerve fibres containing
    gastrin releasing peptide around pial vessels. J . Cereb. Blood Flow Metub.,S, 386-
Uddman, R., Moghimzade, E., and Sundler, F. (1984). Occurrence and distribution of
    GRP-immunoreactive nerve fibres in the respiratory tract. Arch. Otorhinolaryngol.
    239, 145.
Uematsu, K., Holmgren, S., and Nilsson, S. (1989).Autonomic innervation ofthe ovary in
    the Atlantic cod, Gadus morhuu. Fish Physiol. Biochem. 6,213-291.
Vallarino, M., Deste, L., Negri, L., Ottonello, I., and Renda, T. (1990). Occurrence of
    bombesin-like immunoreactivity in the brain of the cartilaginous fish, Scyliorhinus
    canicula. Cell Tissue Res. 259, 177-181.
Vigna, S. R. (1985). Evolution of hormone and receptor diversity: Cholecystokinin and
    gastrin. Am. Zool. 26, 1033-1040.
Vigna, S. R., and Thorndyke, M. C. (1989).Bombesin. I n “The Comparative Physiology
    of Regulatory Peptides” (S. Holmgren, ed.), pp. 34-60. Chapman and Hall, London.
Wood, C. M. (1977). Cholinergic mechanisms and the response to ATP in the systemic
    vasculature of the rainbow trout. J . Comp. Physiol. 122,325-346.
Yamauchi, A., and Burnstock, G. (1968). An electronmicroscope study of the innervation
    of the trout heart. J . Comp. Neurol. 132,567-588.
Young, J. Z. (1933).The autonomic nervous system of selachians. Quart.]. Micr. Sci. 75,
    57 1-624.
This Page Intentionally Left Blank

School of' Biological Sciences
The University of Birmingham
Birmingham, United Kingdom

  I. Introduction
 11. Innervation of the Heart
     A. Cyclostomes
     B. Elasmobranchs
     C. Teleosts
111. The Central Location of Vagal Preganglionic Neurons
     A. Cyclostomes
     B. Elasmobranchs
     C. Teleosts
IV. Control of the Heart and Branchial Circulation
     A. Vagal Tone on the Heart
     B. Efferent Activity Recorded from Cardiac Vagi
     C. Central Origin of Efferent Activity in Cardiac Vagi
 V. Cardiorespiratory Interactions
     A. Reflex Modulation of Heart Rate
     B. Central Interactions Modulating Heart Rate
VI. Cardiorespiratory Synchrony


    The fish heart is composed of typical vertebrate cardiac muscle
fibers. Contraction is initiated by a propagated muscle action potential
that originates from a myogenic pacemaker and generates a character-
istic electrocardiogram (ECG) wave form (Randall, 1968; Satchell,
1991).Its functioning is influenced by intrinsic mechanisms, such as
FISH PHYSIOLOGY, VOL. XIIB                              Copyright 0 1992 by Academic Press, Inc.
                                                   All rights of reproduction in any form reserved.
344                                                         E. W. TAYLOR

the relationship between the force of contraction and stretch applied to
the muscle fibers, which is identical to the Frank-Starling relationship
described in mammals (Johansen, 1971). Thus, the increase in dia-
stolic filling time that accompanies cardiac slowing, because it results
in increased atrial volume, causes an increase in cardiac stroke vol-
ume. Short et al. (1977) concluded that maintenance of cardiac output
in the dogfish during a hypoxia-induced bradycardia was wholly attrib-
utable to the Frank-Starling relationship.
    Heart rate operates under the influence of nervous and hormonal
controls in order that it may respond to changes in supp:y or demand
with respect to oxygen or metabolites. This chapter considers the
efferent arm of the reflex nervous control of the fish heart. The afferent,
sensory arm is reviewed by Burleson et al. in Chapter 7 of this volume.
    Anatomical and pharmacological evidence suggests that efferent
nervous control of the heart in elasmobranchs is due solely to an
inhibitory parasympathetic input supplied by the cardiac vagus. Al-
though the teleost heart receives an excitatory sympathetic supply,
parasympathetic control predominates, and the heart in both groups
operates under varying levels of inhibitory vagal tone. Consequently,
this chapter concentrates on efferent vagal innervation of the heart and
considers the central projections of the cardiac vagi together with
associated branchial and visceral branches of the vagus. A detailed
description of the neuranatomy of the vagal motor column is related to
recordings of efferent activity in branches of the vagus. Changes in
heart rate with temperature and oxygen partial pressure, together with
cardiorespiratory interactions, are described with emphasis on their
neurophysiological bases, including the roles of central interactions
and peripheral chemoreceptors and mechanoreceptors in determining
the efferent output from cardiac vagal motoneurons in the brainstem.


A. Cyclostomes
   This group of vertebrates is composed of the myxinoids (e.g., My-
xine, the hagfish) and the petromyzonts (e.g., Lampetra, the lamprey).
The heart of myxinoids is aneural, that is, it is not innervated by the
vagus or the sympathetic nervous system (Green, 1902; Carlson, 1904);
whereas the heart of the lamprey (although similarly devoid of a sym-
pathetic supply) is innervated by the vagus (Ransom and Thompson,
1886; Augustinsson et al., 1956). The cardiac fibers leave the thin,

nonmyelinized epibranchial trunk of the vagus and run to the median
jugular vein (Fange, 1972).
    The main effect of vagal stimulation in petromyzonts is an acceler-
ation of the heart with an accompanying decrease in the force of
contraction (Falck et al., 1966).Acetycholine induces an acceleration
of the heart, a response unique among vetebrates. Nicotinic cho-
linoceptor agonists, such as nicotine, have the same effect (Au-
gustinsson et al., 1956; Falck et al., 1966).The excitatory effect ofvagal
stimulation or nicotinic agonists can be blocked by nicotinic cho-
linoceptor antagonists such as tubocurarine and hexamethonium (Au-
gustinsson et al., 1956; Falck et al., 1966; Lukomskayo and Michelson,
 1972).The heart in cyclostomes contains large quantities of adrenaline
and noradrenaline, stored in chromaffin-type cells, that may be re-
leased to maintain normal cardiac function (Nilsson and Axelsson,
 1987). Adrenaline, noradrenaline, isoprenaline, and tyramine stimu-
late the petromyzont heart, although the effects are less pronounced
than that of acetylcholine. These effects are blocked by propranolol,
suggesting involvement of P-adrenoceptors on the heart as in the
higher vertebrates (Augustinsson et al., 1956; Falck et al., 1966; Nayler
and Howells, 1965).

B. Elasmobranchs
    The elasmobranchs are phylogenetically the earliest group of verte-
brates in which a well-developed autonomic nervous system with
clearly differentiated parasympathetic and sympathetic components
has been described (Nicol, 1952). They are also the earliest group
known to have an inhibitory vagal innervation of the heart. In the
dogfish, Scyliorhinus canicula, the vagus nerve divides to form, at its
proximal end, branchial branches 1, 2, 3, and 4 that contain skeleto-
motor fibers innervating the intrinsic respiratory muscles of gill arches
2, 3, 4, and 5, respectively, as well as sensory fibers to the gill arches
and walls of the pharynx (Fig. 1).The first gill arch is innervated by the
glossopharyngeal (IXth cranial) nerve. Other respiratory muscles oper-
ating around the jaws and pharyngeal skeleton are innervated by
branches of cranial nerves V and VII (Fig. 1).The vagus also sends, on
each side of the fish, two branches to the heart: the branchial cardiac
branch, which arises from the fourth branchial branch (Norris and
Hughes, 1920), and the visceral cardiac branch, which arises from the
visceral branch of the vagus (Marshall and Hurst, 1905). These
branches were redescribed by Lutz (1930c), Taylor et al., (1977), Bar-
rett and Taylor (1985a), and Withington-Wray et al. (1986) and are
346                                                                       E. W. TAYLOR


                        gill cleft
                                                           post. lat.
                          X br. 1                                lin.
                          X br. 2                          ant. card.
                          X br.
                                  4                        heart
                         X br. c.
                       duct. Cuv.                          pericard.
                        Xvisc. c.                          visc. X

    Fig. 1. Schematic diagram of a dorsal view of the left side of the head of the dogfish
to display the course of the cranial nerves innervating respiratory muscles and the heart.
These constitute the mandibular branch of the trigeminal, Vth nerve (V mand.); the
spiracular branches of the facial, VIIth nerve; the glossopharyngeal, IXth; and the vagus
Xth nerves. The vagus divides to form four branchial branches (X br.1-4), which inner-
vate the intrinsic respiratory muscles in the gill arches on either side of gill clefts 2-5.
The first gill cleft is innervated by IX. Both IX and X are located on the wall of the
anterior cardinal sinus (ant. card. sin.). The vagus, X also supplies two branches to the
heart: the branchial cardiac (X br. c), which branches from the fourth branchial branch,
and visceral cardiac (X visc. c), which branches from the visceral branch (visc. X),
supplying the esophagus and foregut. Both cardiac branches enter the ductus Cuveri
(duct. Cuv.) and run toward the heart where they form a dense plexus on the sinus
venosus and atria. (pericard., pericardium; post. lat. lin., posterior lateral line nerve).

illustrated in Fig. 1. The two pairs of cardiac nerves pass down the
ductus Cuveri and then break up into an interwoven plexus on the
sinus venosus, terminating at the junction with the atrium (Young,
1933).The sinoatrial node is thought to be the site of the pacemaker in
elasmobranch fishes (Rybak and Cortok, 1956; Satchell, 1971). The
remainder of the vagus is termed the visceral branch, and this inner-
vates the anterior part of the gut down to the pylorus and the anterior
part of the spiral intestine (Young, 1933). Stimulation of the vagus
nerve, as well as application ofacetycholine, has an inhibitory effect on
heart rate (Short et al., 1977). The effects are antagonized b y atropine,

implying that the effect is mediated by muscarinic cholinoceptors as in
the higher vertebrates (Lutz, 1930a-c; Johansen et al., 1966; Butler
and Taylor, 1971; Capra and Satchell, 1977; Taylor et al., 1977). Injec-
tion of atropine into intact fish abolished a reflex bradycardia in re-
sponse to hypoxia (Butler and Taylor, 1971; Taylor et al., 1977) and
hyperoxia (Barrett and Taylor, 1984a; and see Fig. 9).
    The sympathetic system of elasmobranchs consists of an irregular
series of ganglia, lying dorsal to the posterior cardinal sinus and ex-
tending back above the kidneys (Young, 1950). These paravertebral
ganglia are arranged approximately segmentally, except in the most
anterior part where there is a concentration of associated neurosecre-
tory tissue forming the auxillary bodies (see later). The segmentally
arranged ganglia are irregularly connected longitudinally and with the
contralateral paravertebral ganglia, but there are no distinct sympa-
thetic chains of the type found in higher vertebrate groups. A peculiar-
ity of the sympathetic system of elasmobranchs is that it does not
extend into the head. This condition is unique among vertebrates, but
it is not clear whether it is primary or the result of a secondary loss
(Young, 1950).
    Gannon et al. (1972) described a sparse adrenergic innervation of
the sinus venosus in Heterodontus, and it is conceivable that sympa-
thetic fibers reach the elasmobranch heart and may influence heart
rate, but the pathway is disputed. With rare exceptions (e.g., Mustelus,
Pick, l970), contributions to the vagi or direct cardiac nerves from
paravertebral ganglia have not been traced anatomically (Izquierdo,
1930; Lutz, 1930a-c; Young, 1933; Pick, 1970; Short et al., 1977).
However, Izquierdo (1930) described a branch from the suprarenal
bodies (see following discussion) that joined the vagi at the ductus
Cuveri and found that electrical stimulation of this region increased
sinoatrial conduction velocity in atropinized preparations. Short et al.
(1977) found that vagal stimulation in atropinized dogfish had no effect
on heart rate. However, our stimulation sites may have excluded a
sympathetic nerve joining the vagi close to the heart. In turn, Izquier-
do’s methods did not exclude an effect on the heart due to release of
catecholamines from the auxillary body (see later). Thus, the existence
of a cardiac sympathetic innervation in elasmobranchs is not proven,
and there is no direct evidence for it having a functional role (Randall,
1970; Nilsson and Axelsson, 1987).
    Aggregates of chromaffin tissue, the “suprarenal bodies,” are juxta-
posed to the paravertebral ganglia in elasmobranchs. They represent
the homologue of the more discrete adrenal medulla of mammals,
birds, and reptiles and contain high concentrations of catecholamines,
348                                                          E . W. TAYLOR

predominantly noradrenaline, that are released into the circulation by
activity in sympathetic preganglionic fibers. The most anterior and
largest suprarenal body, the auxillary body, lies in the dorsal wall of the
posterior cardinal sinus from where its products will b e aspirated
directly into the heart (Johansen, 1971).The reported effects of adren-
aline and noradrenaline on the elasmobranch heart are variable
(Nilsson, 1983) but a positive chronotropic and large inotropic effect,
mediated by a P-adrenoreceptor mechanism, has been described for
the isolated heart (Capra and Satchell, 1977), and some degree of
cardioregulation may be exercised by catecholamines released from
the auxillary bodies. Circulating levels of catecholamines are rela-
tively high in elasmobranchs and increase during hypoxia (Butler et
al., 1978) so that it seems possible that they may exert tonic control
over the cardiovascular system (Short et al., 1977); compensating for
lack of sympathetic innervation of the heart and branchial circulation.
An additional adrenergic influence on the heart may be exerted by
specialized catecholamine storing endothelial cells in the sinus veno-
sus and atrium. These cells are innervated by cholinergic vagal fibers
(Saetersdal et al., 1975; Pettersson and Nilsson, 1979).
     There is no evidence for any vasomotor innervation of elasmo-
branch gills (Nilsson, 1983; Metcalfe and Butler, 1984b), but an intrin-
 sic vasoconstriction during deep hypoxia (Satchell, 1962) may be re-
leased by a rise in circulating catecholamines (Butler et al., 1978), and
it is possible that these vasomotor effects on the branchial vasculature
involve variations in the relative proportion of total blood flow directed
through the parallel arterioarterial and arteriovenous routes or changes
in the patterns of perfusion of the gill lamellae (Nilsson, 1983).

C. Teleosts
   In teleost fish the vagus innervates the gills, the heart, and the
viscera (pharynx, esophagus, stomach, and swimbladder) (Campbell,
1970). The teleost heart is innervated via a branch of the visceral vagus
(Nicol, 1952; Randall, 1970; Johansen, 1971). The cardiac branches of
the vagi follow the ductus Cuveri to the sinus venoms and atrium but
vagal fibers may not reach the ventricle. Vagal ganglia lie close to the
sinoatrial border and appear to consist solely of nonadrenergic cell
bodies (Laurent, 1962; Gannon and Burnstock, 1969; Santer and Cobb,
1972; Santer, 1972; Holmgren, 1977,1981). The vagus is cardioinhibi-
tory as in all vertebrates, with the exception of the cyclostomes. As in
elasmobranchs, this inhibitory affect is due to the release of acety-
choline affecting muscarinic cholinoceptors associated with the car-

diac pacemaker and atrial musculature (Young, 1936; Randall, 1966;
Randall and Stevens, 1967; Gannon and Burnstock, 1969; Holmgren,
1977,1981; Cameron, 1979).Atropine or hyoscine block a reflex brady-
cardia in intact fish (Stevens et al., 1972; Priede, 1974; Nilsson and
Axelsson, 1987), the inhibition of rate following vagal stimulation
(Randall, 1966; Holmgren, 1977) and the negative inotropic effects of
acetylcholine on isolated perfused atria (e.g., Donald and Cambell,
1982).Acetylcholine was without effect on the contractility of isolated
ventricles from the cod (Holmgren, 1977) and seven other teleost
species (Donald and Cambell, 1982).Although the negative inotropic
influence of the vagi does not reach the ventricle, cardiac output is
greatly affected by the inotropic control of the atrium, which directly
regulates the filling of the ventricle (Jones and Randall, 1978; Johan-
son and Burggren, 1980).
    Historically, sympathetic cardioaccelatory innervation was gener-
ally assumed to be lacking in teleosts (Randall, 1968). However, an
adrenergic innervation of the heart via a mixed vagosympathetic trunk
as well as separate sympathetic postganglionic fibers has since been
described in several teleost species (Gannon and Burnstock, 1969;
Holmgren, 1977; Cameron, 1979; Donald and Campbell, 1982). As a
whole the teleosts may be considered phylogenetically at the first
group of vertebrates in which there is both sympathetic and parasym-
pathetic control of the heart, with innervation similar to that found in
tetrapods (Laurent et al., 1983; Nilsson and Axelsson, 1987).
    An adrenergic tonus has been demonstrated on the heart of the cod
(Gadus), but the relative importance of the neuronal and humoral
adrenergic control of the heart remains uncertain (Nilsson, 1983). The
positive chronotropic and inotropic effects on the teleost heart, pro-
duced by adrenergic agonists and adrenergic nerves, are mediated via
P-adrenoceptor mechanisms associated with the pacemaker and the
myocardial cells (Randall and Stevens, 1967; Gannon and Burnstock,
1969; Holmgren, 1977; Wahlqvist and Nilsson, 1977; Cameron and
Brown, 1981).Adrenergic control may be important during exercise as
a rapid cardioacceleration induced by enforced swimming in the gold-
fish was abolished by propranolol (Cameron, 1979), and Priede (1974)
found that increases in heart rate associated with swimming in rainbow
trout continued after bilateral vagotomy.
    In contrast to elasmobranchs where the branchial branches are
solely skeletomotor (Metcalfe and Butler, 1984b), the branchial
branches of the vagus (going to the gills) have both a vasomotor and
skeletomotor function (Pettersson and Nilsson, 1979). There are sym-
pathetic ganglia associated with cranial nerves IX and X in teleosts and
350                                                          E. W. TAYLOR

the branchial nerves are mixed vago/glossopharyngeosympathetic
trunks (Nilsson, 1983). Stimulation of these nerves may produce a
cholinergically mediated constriction of the arterioarterial pathway in
the gills whereas stimulation of the adrenergic fibers favors blood
flow through this respiratory route rather than the arteriovenous route.
Despite the clear demonstration of mixed autonomic innervation of the
heart and branchial vasculature in teleosts, it remains probable that
much of the functional control of gill perfusion is exercised via circu-
lating catecholamines (Nilsson, 1983; Part A, Chapter 1).


A. Cyclostomes
    The central nervous system of the cyclostomes may represent a
prototype of the vertebrate brain (Ariens Kappers, 1929, 1947). The
hindbrain is identical in superficial appearance to that of the rest of the
vertebrates, with vagal rootlets leaving on either side to innervate the
viscera. No study has been made of the topographical representation of
vagally innervated structures within the vagal motor column of cyclo-
stomes. In Lampetra two separate divisions of the vagal motor column
have been identified using normal staining techniques: a rostral and a
caudal motor nucleus of X (Niewenhuys, 1972). The caudal motor
nucleus of X, which cannot be delineated from the spinal visceromotor
cells, is thought to represent a splanchnic center, and the rostral nu-
cleus is considered to be branchiomotor in nature (i.e., to innervate the
branchial pouches) (Addens, 1933). The location of the caudal motor
nucleus in cyclostomes, which centers around obex, is similar to the
region of the dorsal vagal motor nucleus (DVN) in the cat (Bennett et
al., 1981)and to the nucleus motorius nervi vagi medialis (Xmm) in the
dogfish (Barrett and Taylor, l985b) in which the cell bodies contribut-
ing axons to the cardiac vagi are found.

B. Elasmobranchs
   T h e gross location of the vagal motor column in the hindbrain has
been described in a number of elasmobranchs, although almost noth-
ing was known of the topographical origin of vagal preganglionic fibers
in elasmobranchs (Smeets et al., 1983). Classic neuranatomical tech-
niques were used to describe a continuous column of large cell bodies

of preganglionic neurons constituting motor nuclei of the IXth and Xth
cranial nerves in a number of elasmobranchs, namely Selache maxima
(Black, 1917), Squalus acanthias and Scyliorhinus canicula (Smeets
and Niewenhuys, 1976), and Hydrolagus collei and Raja clavata
(Smeets et al., 1983).
    In the shark Cetorhinus and in the Holocephali, Addens (1933)
divided the vagal motor nucleus into separate rostral and caudal parts
and suggested that the rostral portion subserves either a visceromotor
or branchiomotor function, whereas the caudal portion represents a
general visceromotor or splanchnic center. In Squalus this vagal part of
the visceromotor column was designated the nucleus motorius nervi
wagi medialis (Xmm) (Smeets and Niewenhuys, 1976). An area lateral
to the caudal part of the visceromotor column contained a distinct
aggregation of large bipolar and triangular cells and Smeets and
Niewenhuys (1976) considered this to represent part of the motor
nucleus of X and accordingly named it the nucleus motorius nervi vagi
lateralis (Xml). The Xmm and Xml, by virtue of their locations, may be
the homologues of the mammalian DVN and the nucleus ambiguus
(NA), respectively (Smeets and Niewenhuys, 1976; Barrett et al.,
1983). In this account they will be referred to as the DVN and LVN
signifying dorsal and lateral vagal motonuclei.
    Retrograde intraaxonal transport of horseradish peroxidase (HRP)
to identify vagal preganglionic neurons (Withington-Wray et al., 1986;
Levings, 1990) showed that the vagal motor column in the dogfish,
Scyliorhinus canicula, extends over 5 mm in the hindbrain (2.1 mm
caudal to 2.9 mm rostral to obex, Figs. 2, 3, and 4 ,which agrees well
with the extent described by Smeets and Niewenhuys (1976) for fish of
similar size. Caudal to obex there appeared at first to be two distinct
groups of vagal motoneurons, the majority found dorsomedially, and a
smaller ventromedial group, both close to the lateral edge of the fourth
ventricle (Withington-Wray et al., 1986).The ventromedial group were
continuous with cells in the spino-occipital motor nucleus (Black,
1917), and almost certainly constituted a forward extension of this
nucleus-contributing axons to the hypobranchial nerve that inner-
vates the ventral muscles of the gill region (Levings, 1990; Levings and
Taylor, 1988).The majority of vagal motoneurons caudal of obex con-
tribute axons to the visceral branch of the vagus. The other branch of
the vagus whose motoneurons were found caudal of obex was the
visceral cardiac branch (Fig. 3). Visceral cardiac motoneurons were
found in the dorsomedial division of the vagal motor column. Rostra1 of
obex the medial vagal motoneurons were found clustered close to the
ventrolateral edge of the fourth ventricle (Fig. 5) in the visceromotor
352                                                                     E. W. TAYLOR


    Fig. 2. Schematic diagram of a dorsal view of the hindbrain and anterior spinal cord
of the dogfish showing the distribution of the neuron cell bodies supplying efferent
axons to cranial nerves innervating respiratory muscles and the heart; namely, the vagus,
Xth; glossopharyngeal, IXth; spiracular branches of the facial, VIIth; and the adductor
mandibulae branches of the trigeminal, Vth nerves. The individual motor nuclei are
represented as hatched areas; the area of overlap between adjacent motor nuclei is
indicated in black. 1, Adductor mandibulae motor nucleus (V); 2, Facial motor nucleus
(VII); 3, Glossopharyngeal motor nucleus (IX); 4, Dorsal motor nucleus of the vagus (X);
5, Lateral motor nucleus of the vagus (X); 6, Occipital nerve (XI); 7, Vagus nerve (X); 8,
Glossopharyngeal nerve (IX); 9, Octavus nerve (VIII); 10, Branches of the facial nerve
(VII); 11, Branches of the facial and trigeminal nerves (VII and V); aur, cerebellar
auricle; smi, sulcus medianus inferior; siv, sulcus intermedius ventralis; slH, sulcus
limitans of His; oli, inferior olive; R, rhomboid fossa; 0, obex. [Redrawn from Levings
6.   CONTROL OF HEART AND CARDIORESPIRATORY INTERACTIONS                                                                  353

      Mand Vth
      Facial '411th
      Vagus Xth
      Glossa. lXth
        X Br. 1
        X Br. 2
       X Br. 3
        X Br. C.                -
        X visc. C              -
                       I           I         I      I        I     I        I          I    I       I         I      I
                      -2     -1          OBEX      +1        *2    +3     +4          +5    +6      +7     +8        +9
                    caudal                                                                                 rostral
                                         Distribution with respect to obex (mm)
    Fig. 3. The topographical organization of the vagal motor column and respiratory
motor nuclei of the Vth, VIIth, IXth, and Xth cranial nerves in the dogfish to show their
sequential rostro-caudal distribution on either side of obex. The lines indicate, from the
top down, the rostro-caudal extent of pools of motoneurons supplying the mandibular
branch of the Vth cranial nerve; the facial branch of the VIIth cranial nerve; the whole
vagus (X), the glossopharyngeal (IXth) cranial nerve; and the vagus (Xth) cranial nerve,
separated into its constituent branches (the first three branchial branches, X Br 1-3; the
branchial cardiac, X Br C; and visceral cardiac, X visc. C, branches, and the visceral
branch, visc. X). [Redrawn from Taylor (1989).]

            IJl   Medial     42:
                  Lateral 20
                                     E                             A
                                                                           n         : 77
                                         I               I         I             I              I         I
                                       -2               -1        OBEX           +1         +2           +3
                                   caudal                                                                rostral
                                                 Distribution with respect to obex (mm)
    Fig. 4. Rostro-caudal distribution of vagal preganglionic motoneurons with respect
to obex in the medulla of the dogfish, Scyliorhinus canicula. The majority of labeled
motoneurons are located medially in the DVN. A small number (8%)are located ventro-
laterally and supply axons solely to the branchial cardiac branch of the vagus innervating
the heart. Medial cells supplying this nerve are indicated by the unshaded portion of the
upper histogram. Motoneurons supplying axons to branchial branches 1, 2, and 3 (Br
1-Brlll) of the vagus,innervating gill arches 2, 3, and 4,occupy the rostral part of the
vagal motor column, while the more caudal motoneurons supply axons sequentially to
the heart, esophagus, and stomach. [Redrawn from Taylor and Elliott (1989).]
354                                                                       E. W. TAYLOR

    Fig. 5. The location of preganglionic motoneurons identified in the brainstem of
elasmobranch fish, following retrograde transport of HRP along identified branches of
the Xth, vagus, and IXth glossopharyngeal cranial nerves. Cell bodies of preganglionic
motoneurons, their axons and dendrites, as well as afferent sensory projections are
stained with dark reaction product. (A) T.S. of medulla from dogfish, Scyliorhinus
canicula, taken at obex to show preganglionic vagal motoneurons in the dorsal moto-
nucleus (DVN) close to the 4th ventricle and in a scattered ventrolateral location (the
LVN) (scale bar 200 pm). (B) A schematic diagram of a T.S. at obex to show the relation-
ship between vagal motoneurons and other structures in the brainstem. (C) T.S. of
dogfish medulla taken 0.4 mm rostral of obex to show the location of cardiac vagal
motoneurons that are distributed both in the DVN close to the 4th ventricle and are
scattered ventrolaterally right to the lateral edge of the brainstem in the LVN (scale bar
100 pm). (D) T.S. of medulla from the ray Raia claoata, 1.2 mm rostral of obex following
application of HRP to the right glossopharyngeal nerve. Sensory neurons are labeled in
the visceral sensory nucleus of IX (IX sn) together with their axons that course through
the solitary tract (ts). The motor nucleus of IX is located medially in the visceral motor
column surrounding Steida’s fasiculus (fst) (scale bar 200 pm). (ax, axons; ax. X, axons of
vagus nerve; DVN, dorsal vagal motonucleus; F1, nucleus funiculi lateralis; flm, fasi-
culus longitudinalis medialis; fst, fasiculus medianus of Steida; IXsn, glossopharyngeal
sensory nucleus; LVN, lateral vagal motonucleus; Oh, Oliva inferior; rdv, radix de-
scendens nervi trigemini; siv, sulcus intermedius ventralis; slH, sulcus limitans of His;
smi, sulcus medianus inferior; ts, nucleus tractus solitarius; 4 Vent, fourth ventricle; Xr,
vagal rootlet; Xsn, vagal sensory nucleus.)

column. The majority of vagal motoneurons were found in this ros-
tromedial division of the vagal motor column (i.e., the DVN), which
contributed axons to the branchial cardiac branch and to the visceral
branch in its most caudal one-third, and to the branchial branches of
the vagus in its rostra1 two-thirds (Fig. 3 ) .
     A clearly distinguishable ventrolaterally scattered group of cells
was identified that had a rostrocaudal extent of approximately 1 mm,
rostrally from obex (Figs. 2,4, and 5). This population of motoneurons
contributed axons solely to the branchial cardiac branch of the vagus
(Barrett et al., 1983;Barrett and Taylor, 1985b) and composed 8% ofthe
total population of vagal motoneurons. The cells in this lateral division
supply 60% of the efferent axons running in the branchial cardiac
nerve, with the other 40% supplied by cells in the rostromedial divi-
sion (Fig. 4).  When the medial cells contributing efferent axons to the
heart via the visceral cardiac branches are taken into account, then
the lateral cells supply 45% of vagal efferent output to the heart
(Withington-Wray et al., 1987). Thus branchial cardiac motoneurons
are found rostromedially and solely compose the lateral division
ofthe vagal motor column (Fig. 4). It is thought that these two locations
of cardiac vagal preganglionic neurons give rise to the two different
types of efferent activity recorded from the cardiac vagi of the dogfish
(Taylor and Butler, 1982; Barrett and Taylor, 1985a,c); this point is
examined in the following discussion.
     An as yet unpublished study using retrograde transport of HRP
along identified branches of the vagus to identify the detailed topogra-
phy of the vagal motor column in two species of rays Raja clavata and
R . microocellata (Levings, 1990) supported our results for the dogfish.
T h e majority of vagal preganglionic neuron cell bodies were located as
a “classic DVN” forming a continuous longitudinal column in the
ipsilateral hindbrain. In the midportion of the vagal motor nucleus,
 labeled neurons were observed to be located both in the DVN and
 scattered in a separate ventrolateral location, called the LVN that
 extended rostro-caudally for approximately 2 mm (Fig. 6). Barry (1987)
described the vagal motor nucleus of Raja eglanteria as being located
as separate dorsal and ventral nuclei. Examination of his data revealed
 a typical DVN (which he termed Xmd) and a separate group of neurons,
 which were scattered ventrolaterally, almost to the edge of the hind-
brain (termed Xmv). The location of these scattered neurons is very
 similar to that of neurons sited in the LVN of the Raja species exam-
 ined by Levings (1990). However, the location of neurons sited cau-
 dally in his Xmv closely resembles that of hypobranchial motoneurons,
 and it is probable that they arise from confusion between vagal and
356                                                                     E. W. TAYLOR

         Whole vagus
         Branchial 1
         Branchial 2
         Branchial 3
         Branchial 4
                          I     I      I      I      I       1      I       I
                          -4   -3      -2     -1    OBEX    *1     +2      +3
                               Distribution with respect to obex (mm)
    Fig. 6 . The rostro-caudal distribution, with respect to obex, of preganglionic
motoneurons in the vagal motor nucleus of the ray Raja clauata. (A) The distribution of
vagal motoneurons found following application of HRP to the whole vagal trunk. Dorsal
vagal motonucleus, DVN; lateral vagal motonucleus, LVN. (B) The topography of the
vagal motor nucleus. The distribution of vagal motoneurons supplying efferent axons to
the four branchial branches, the cardiac branch (supplied by the DVN and LVN) and the
visceral branches of the vagus nerve. [Redrawn from Levings (1990).]

hypobranchial branches, similar to that experienced by Withington-
Wray et al. (1986)working on the dogfish.
    Application of HRP to the individual branches of the vagus re-
vealed that the cells of origin of the branchial nerves were serially
represented in the DVN in the rostral and midportion of the vagal
motor nucleus (Figs. 6 and 7). A similar sequential topographical rep-
resentation of the cells of origin of the individual branchial branches of
the vagus in the DVN was described in the dogfish (Fig. 3 ) and can be
identified in Fig. 4.Additional neuranatomical study of branches ofthe
Vth and VIIth cranial nerves to respiratory muscles in dogfish and rays
(Levings, 1990) revealed that their efferent cell bodies were located
ipsilaterally in the brainstem rostral of the vagal motor column, con-
tributing to the sequential rostro-caudal distribution of discrete motor
nuclei innervating the respiratory apparatus (Figs. 2 and 3 ) .
    The rays possess only one pair of cardiac vagi. Cardiac vaga