VIEWS: 214 PAGES: 87

             S T R U C T U R AL P R O P E R TI E S

                Mikael Jonassohn
                        LUND 1996

                   LUND UNIVERSITY
A doctoral thesis at a university in Sweden is produced either as a monograph or as
a collection of papers. In the latter case, the introductory part constitutes the
formal thesis, which summarises the accompanying papers. These have either
already been published or are manuscripts at various stages (in press, submitted, or
in manuscript).

© 1996 Mikael Jonassohn
Department of Organic Chemistry 2
Lund University
P.O.Box 124, S-221 00 Lund (Sweden)

ISBN 91-628-2215-2
            List of papers
This thesis summarises the following papers which are referred to by roman
numerals I-VII.

I     Mikael Jonassohn, Heidrun Anke, Olov Sterner and Christer Svensson
      New Compounds Isolated from the Culture Filtrate
      of the Fungus Merulius tremellosus
      Tetrahedron Letters, 1994, 35(10), 1593-1596.

II    Mikael Jonassohn, Heidrun Anke, Paloma Morales and Olov Sterner
      Structure-Activity Relationships for Unsaturated Dialdehydes 10. The
      Generation of Bioactive Products by Autoxidation of Isovelleral and Merulidial
      Acta Chemica Scandinavica, 1995, 49(7), 530-535.

III   Mikael Jonassohn, Olov Sterner and Heidrun Anke
      Structure-Activity Relationships for Unsaturated Dialdehydes 12.
      The Reactivity of Triacetic Acid Lactone Towards Unsaturated Dialdehydes
      Tetrahedron, 1996, 52(4), 1473-1478.

IV    Jörgen Gustavsson, Mikael Jonassohn, Pia Kahnberg, Heidrun Anke
      and Olov Sterner
      The Reactivity of the Antibiotic Sesquiterpene Isovelleral
      Towards Primary Amines
      Accepted for publication in Natural Product Letters, 1997.

V     Mikael Jonassohn, Ralf Davidsson, Pia Kahnberg and Olov Sterner
      The Reactivity of Drimane Unsaturated Dialdehydes Towards Nucleophiles
      Tetrahedron, 1997, 53(1), 237-244.

VI    Mikael Jonassohn, Rikard Hjertberg, Heidrun Anke, Kim Dekermendjian,
      Arpad Szallasi, Eckhard Thines, Robin Witt and Olov Sterner
      The Preparation and Bioactivities of (-)-Isovelleral
      Accepted for publication in Bioorg. Med. Chem., 1997.

VII Mikael Jonassohn, Pia Kahnberg and Olov Sterner
    The reactivity of Isovelleral Towards Amino Acids
    In manuscript.
The natural unsaturated 1,4-dialdehydes warburganal (6), polygodial (14),
isovelleral (55) and merulidial (56a) and their corresponding unconjugated
aldehyde isomers (121), (15), (108) and (109a) were prepared, and their
bioactivity and reactivity towards various nucleophiles were investigated.
   New synthetic routes to prepare (-)-6 and its isomer (+)-121 from (-)-14, and
(+)-109a from (-)-56a, have been developed. The dialdehydes (-)-55 and (+)-108
were prepared by total synthesis. A natural adduct between (-)-56a and triacetic
acid lactone (155) was isolated from the culture filtrate of the fungus Merulius
tremellosus, and the same adduct was shown to be formed from the reaction
between 56a and 155 in aqueous solutions as well as in organic solvents.
   The reactivity of the eight unsaturated 1,4-dialdehydes towards lactone 155, L-
lysine, L -alanine and L -cysteine was investigated. The naturally occuring
dialdehydes 6, 14, 55, 56a all react faster than their corresponding isomer 121, 15,
108 and 109a, which correlates with for instance their antibiotic activities. The
drimane dialdehydes 6 and 14, are generally reactive towards various nucleophiles,
while the marasmane 55 and its isomer 108 are specifically reactive towards thiols.
The dialdehydes 6, 55 and 56a are easily autoxidised, while less bioactive
derivatives are not, and the biological activities of the autoxidation products of 55
and 56a are of the same order as their parent compounds. Furthermore, 55 was
found to oxidise cysteine to cystine in buffer at pH 7.4. The biological activities of
the dialdehydes can only in part be correlated with their reactivity towards
nucleophiles, and other molecular mechanisms, for example oxidative, for their
action must be considered.
   A correlation between the hot taste of the dialdehydes to the human tongue
and their affinity for the vanilloid receptor, which is the target for the pungent
compounds capsaicin and resiniferatoxin, was found. The hot taste of the
dialdehydes to the human tongue, as well as their interactions with the dopamine
D1 receptor and their mutagenicity in Ames' test, were at least to some extent
shown to be enantiospecific.

Key words: Unsaturated dialdehydes, isovelleral, polygodial, warburganal, merulidial,
reactivity, amino acids, triacetic acid lactone, structure-activity relationships, vanilloid
receptor, dopamine D1 receptor, autoxidation, mutagenic, antimicrobial, cytotoxic
          OR E T H AN O NE         hundred years ago, Emil Fischer announced his

M          elucidation of the structure of glucose (1). In comparison, the structures
           of the complex alkaloids vinblastine (2) and vincristine (3) were
elucidated seventy years later. The three compounds (see Figure 1) are clearly
natural products, yet glucose is ubiquitous and essential for life, while the two
alkaloids are produced by only a few plant species and have no apparent functions
in the plants. It is customary to distinguish between the primary and the secondary
metabolism. The former refers to the processes essential for life producing simple
and widely distributed low molecular weight carboxylic acids, proteins, lipids, α -
amino acids and carbohydrates. A characteristic feature of secondary metabolites,
e.g. vinblastine (2) and vincristine (3), is that their function is principally unknown.
It is hard to believe that they are just waste products, and that an organism should
allocate such a large proportion of metabolic resources for purposes voids of sense.

                                                H              COOMe

                                                       N        N
                               O                   R
                 HO                                        H
                                      OH          HO
                               OH               MeOOC

                    β-D-glucose (1)            R=Me Vinblastine (2)
                                               R=CHO Vincristine (3)

                       Figure 1 Primary and secondary metabolites

   What we do know about secondary metabolites is that their production is
connected with several external factors. They could be defined as non-nutritional
compounds, which influence the biology of the producer (e.g.hormones) itself as

well as other species in the environment. Interestingly, secondary metabolites also
play a prominent role in the coexistence and co-evolution of species. Torsell [1]
exemplified their function in several areas of chemical control:
   • sexual attractants
   • development, e.g. sterilisation by queen substance (bees), metamorphosis,
        egg laying, growth suppressors (animal-plant, animals, plants)
   • social behaviour, e.g. building behaviour (termites), territorial claims (bees),
        track indicators (ants), accumulation (locusts), etc.
   • feedants, antifeedants, repellents and toxins (animal-plant, animals)
   • defence and alarm
Notably, every class of secondary plant constituent is represented as a particular
attractant, repellent, etc. [1]. For instance, the cabbage butterfly Pieris brassicae1 has
a taste for glycosinoates, e.g. sinigrin (4), which is a feeding attractant for this
particular insect. Simultaneously, sinigrin (4) acts as a repellent, or is even
poisonous, for most other insects. However, insects often adapt and become
dependant to these attractants. Since the general nutritional value of most plants
varies comparatively little, the insect population is actually controlled by the host

                                               O         CO2Me
              OH                                   O O                                  HO   CHO
                                                                O                              CHO
 HO                      S               AcO                          O
                                                                OH         O
                             N                                        HO
                                          MeO2C        O

       Sinigrin (4)                                Azadirachtin (5)                Warburganal (6)

                       Figure 2 Some insect attractants/repellents produced by plants

    Plants and other organisms also want to protect themselves against predators.
This is often done by producing repellents, antifeedants and toxins. An antifeedant
has been defined [2] as a compound that inhibits feeding but does not kill the
predator directly, although it may die by starvation. For instance, the African
neem tree Melia azadirachta is never attacked by desert locusts, because it contains
the powerful antifeedant terpene azadirachtin (5) [3]. The sesquiterpene
warburganal (6), which is produced by the East African tree Warburgia stuhlmannii,
is a specific antifeedant against larvae of the African army worm. However, it does
not have any repellent effect on locusts. Warburganal (6) interferes with the
stimulus transduction process in the chemoreceptor cells of the army worms [4].
    This sesquiterpene contains an α ,β -unsaturated 1,4-dialdehyde moiety (see
Figure 3) in conformity with several other repellent or antifeedant terpenoids.
Their repellency implies a role in the natural defence systems of their host
organisms [5, 6]. In some cases, the predator do not even have to chew the plant.
For instance, aphids2 have been shown to detect polygodial (14), another plant
metabolite containing an unsaturated dialdehyde functionality, with sensilla
located on their antennal tips [7]. Until now, 81 terpenoid unsaturated

1 kålfjäril
2 bladlöss

dialdehydes have been isolated from most kingdoms in the nature (see Figure 4
and Appendix A). Most of them are sesqui-, di-, or sesterterpenes, while a few are


                   Figure 3 The α,β-unsaturated 1,4-dialdehyde moiety

    Several unsaturated dialdehydes with a drimane skeleton have been isolated
from Cannellaceae, a small family of plants consisting of nine species, grouped into
four genera. Of these, Winterana and Cinnamodendron are endemic to South
America, Warburgia to East Africa and Cinnamosma to Madagascar. Five
unsaturated dialdehydes have been isolated [8-11] from the stem bark of Cannella
winterana, a tree that grows in the subtropical areas of the Florida Keys and
throughout the Caribbean. The East African genus Warburgia (Canellaceae)
consists of two species, W. stuhlmannii and W. ugandensis. The barks of these are
employed widely in folk-medicine and as spices in food. The bark extract,
containing some unsaturated dialdehydes, possesses antifeedant activity against
African army worms Spodoptera littoralis and S. exempta, widely occurring African
crop pests [12].
    The plant Polygonum hydropiper (Polygonaceae) is also known as marsh pepper
because of its habitat and hot taste experienced on chewing the leaves. The plant
contains the unsaturated dialdehydes polygodial (14), warburganal (6) and
isotadeonal (15). The two former are responsible for the hot taste. The presence of
polygodial (14) (remarkably ca 10% wt/fresh wt in the flowerhead of Polygonum
hydropiper [13]) is probably a key factor, which makes this plant species resistant to
attack by most Lepidopterous larvae [14].
    A systematic investigation of the active compounds of the Hepaticae, made by
Asakawa and co-workers, has also revealed some unsaturated dialdehydes,
especially in the liverwort genus Porella [15-21]. Besides the sesquiterpene
polygodial (14), two groups of diterpene dialdehydes occur in liverworts: the
sacculatals and the perrottetianals (see Figure 4 and Appendix A).
    However, the occurrence of unsaturated dialdehydes is not restricted to
terrestrial plants. Several dialdehydes, e.g. dictyodial A (33), have been isolated
from brown algae, the principal seaweed of temperate and polar regions. The
trialdehyde halimedatrial (40) has been found in numerous species of the tropical
marine green algae Halimeda spp.[22,23], which are among the most common
seaweed on tropical reefs [24]. Green algae represent the largest and most diverse
group of all algae, comprising at least 7000 species. In some species, 25% of the
organic extracts consist of halimedatrial (40). This is not unexpected, since this
trialdehyde is an effective feeding deterrent for herbivorous fishes, and hence can
be considered as a key substance in the chemical defence of these species.
Consequently, some populations of Halimeda growing in reef slope habitats, where
herbivory is intense, have high concentrations of halimedatrial (40) [25]. When
fishes are biting and chewing the plant, the less-deterrent halimedatetraacetate (86)
is immediately converted to the deterrent halimedatrial (40) and epi-halimedatrial
(41) by the plant (see Scheme 1) [24].

Figure 4 Natural unsaturated dialdehydes (the compounds 25-32, 51-53 and 57 equilibry
         between the shown hydroxy lactones and the corresponding dialdehydes).
         *) Stereo chemistry not known

            HO      CHO              HO       CHO                  HO     CHO               HO   CHO
                      CHO                       CHO                         CHO                    CHO

                                     H                             H
                                          OAc                          OH
       Warburganal              Cinnamodial                 Mukaadial                 Muzigadial
          (6)                       (7)                       (8)                        (9)

            HO     CHO                    CHO                          CHO                      CHO
                     CHO                        CHO                          CHO                      CHO

        Epoxy-                   9-Deoxy-                  9-Epi-deoxy-                 9-Deoxy-
       muzigadial                muzigadial                 muzigadial               isomuzigadial
         (10)                      (11)                        (12)                       (13)

               CHO                        CHO                          CHO                      CHO
                       CHO                      CHO                          CHO                      CHO

                                                  AcO                          AcO

       Polygodial               Isotadeonal        3α-Acetoxypolygodial 3β-Acetoxypolygodial
          (14)                      (15)                   (16)                 (17)

                   CHO                     CHO
                                                               O       CHO              O       CHO
                       CHO                       CHO                         CHO                      CHO

      Ancistrodial       -2-cyclohexenyl)-3-pentenal           Panudial            Kuehneromycin B
         (18)                       (19)                         (20)                   (21)

           HO      CHO
                     CHO                                       O       CHO                  HO   CHO
                                                                             CHO                   CHO
                   O                                                                                        4
       RES-1149-1                                       oxypolygodial                Capsicodendrin
          (22)                                               (23)                         (24)

                                O         O      OH                                       O          O    OH
                       OAc    HO                                             OH         HO
                                                H CHO                                                     H CHO
                             O                                                      O

                        O                                                      O

                                 Mniopetal A                                                Mniopetal B
                                    (25)                                                       (26)

                                          O      OH                                         O O       O    OH
                       OH     HO                                                    OH
                                                H CHO                                                     H CHO
                             O                                                     HO

                                 Mniopetal C                                                Mniopetal D
                                    (27)                                                       (28)

          O      O     OH             O    O     OH               O    O     OH               O      O     OH
        HO                       HO                               O                      HO
   HO                H CHO                      H CHO                       H CHO                         H CHO


        Mniopetal E              Mniopetal F                 Kuehneromycin A                 Marasmal
           (29)                     (30)                          (31)                        (32)

         CHO                                    CHO                                  CHO
  CHO       H                             CHO     H                            CHO     H
                 H                                       H                                       H

                                                             OH                                      OH OAc

        Dictyodial A                      4β-Hydroxydictyodial A                 xenicatriene-18,19-dial
            (33)                                  (34)                                     (35)

          CHO                                  CHO                                   CHO
   CHO                                    CHO            OH                    CHO               OAc
                 H                                       H                                       H

7-Hydroxy-2,6-cyclo-9,13-                 9-Hydroxy-2,14-dichoto-                  9-Acetoxy-2,14-dichoto-
 xenicadiene-18,19-dial *                    madiene-19,20-dial                      madiene-19,20-dial
           (36)                                    (37)                                     (38)

          CHO                                                 CHO                                        CHO
                                                     H                                       H                 CHO
   CHO           H                                                    CHO

                                                                       CHO                                      CHO


 xenicadiene-18,19-dial *                       Halimedatrial                         Epi-halimedatrial
           (39)                                     (40)                                    (41)

                                                                                      OH    O

            AcO     CHO                               HO        CHO                             O    CHO
                            CHO                                       CHO                                   CHO

            H                                     H                                         H

         Scalaradial                     12-Deacetylscalaradial            12-(3-Hydroxy-1-oxybutoxy)-scalaradial
            (42)                                 (43)                                      (44)

            AcO     CHO                                         CHO                          HO      CHO
                            CHO                                       CHO                                   CHO

            H                                     H                                         H

      12-Epi-scalaradial                 12-Deacetoxyscalaradial                 12-Deacetyl-12-epi-scalaradial
            (45)                                  (46)                                       (47)

            AcO     CHO                                 O       CHO                          HO      CHO
                            CHO                                       CHO                                   CHO

            H                                       H                                       H

 12,18-Di-epi-scalaradial                    12-oxo-scalaradial              12-Deacetyl-12,18-di-epi-scalaradial
          (48)                                     (49)                                     (50)
                                   O                                O                           O
                                            O                              O                          O
                                                 OH                              OH                         OH

                                              CHO                            CHO                          CHO
                            O     OH                                                            OH

                        O         Pilatin                       Marasmic acid         9-Hydroxymarasmic acid
                                   (51)                             (52)                       (53)
                                                                      CHO                           CHO    OH
  OHC                                                                    CHO
OHC                                                                                                         O
                                                                      OH                                   O

 Hanegoketrial                  Isovelleral                      Merulidial                 Hyphodontal
     (54)                          (55)                            (56)                        (57)

                  CHO                           CHO                             CHO                         CHO

                 CHO                        CHO                              CHO                          CHO
                                    OH                                OH
      Velleral                  Piperdial                        Epi-piperdial             Chrysorrhedial
       (58)                       (59)                               (60)                      (61)

             CHO                         CHO                      OH    CHO                   OH    CHO
                     CHO                         CHO                          CHO                         CHO

                                                             1β-Hydroxy-                     1β-Hydroxy-
      Saccalutal                 Isosaccalutal                sacculatal                     isosacculatal
        (62)                         (63)                        (64)                            (65)

             CHO                         CHO                            CHO                         CHO
                     CHO                         CHO                          CHO                         CHO
     3β-Hydroxy-                                             18-Hydroxy-                     19-Hydroxy-
     isosacculatal               Durbinal C                   sacculatal                      sacculatal
         (66)                       (67)                         (68)                            (69)
             CHO                                                        CHO                                     CHO
       OHC                         OHC CHO
                                                       HO                     CHO
                           HO                           Cl

 Perrottetianal A           Perrottetianal B                 Hamiltonin C                  Ent-isocopalendial
       (70)                      (71)                           (72)                              (73)

                      CHO                                                        CHO
                                                       CHO                                                      CHO
                       CHO                        CHO                                 CHO
                                                                                 O                              CHO

  (E)-Labda-8(17),               (Z)-Labda-8(17),        8β,17-Epoxy-12E-
 12-diene-15,16-dial            12-diene-15,16-dial      labdene-15,16-dial             Iso-Ent-isocopalendial
        (74)                           (75)                    (76)                              (77)

                      CHO                         CHO                                 CHO                        CHO
                       CHO                             CHO                       CHO                         CHO

       Linaridial                 ∆3-(E)-Isolinaridial           ∆3-(Z)-Isolinaridial       (Z)-Isolinaridial
         (78)                            (79)                           (80)                      (81)
                                  CHO            OHC                                                            CHO
                                                                       OHC                                O


     Rhipocephenal                     Oxytoxin 2                  β-Acaridial              labdene-15,16-dial
         (82)                            (83)                         (84)                         (85)

This was the first example of an activated defence in a marine plant to be reported,
but similar defence systems have been found in mushrooms.

    AcO      CHO                                        CHO                          CHO
                                             H                               H
                 AcO                                          CHO                             CHO

                                                               CHO                             CHO


     Halimedatetraacetat (86)                 Halimedatrial (40)          Epi-halimedatrial (41)

                                                                    CHO                       CHO
             O        O
     Stearoylvelutinal (87)                      Isovelleral (55)             Velleral (58)

              Scheme 1 Examples on activated chemical defence in algae and fungi

   Fruit bodies of pungent Russulaceae species (genera Lactarius and Russula)
initially contains a biologically inactive fatty acid ester of velutinal, e.g.
stearoylvelutinal (87). Upon injury (by insects or snails, biting and chewing the
fruitbody) compound 87 is enzymatically transformed within seconds to pungent
unsaturated dialdehydes and related sesquiterpenes (see Scheme 1) [26-28].
Lactarius rufus3 and L. quietus 4 produces only isovelleral (55), L. bertillonii
produces only velleral (58), L. vellereus5 produces isovelleral (55) and velleral (58),
Russula queletii6 , L. piperatus 7 and L. torminosus8 all form velleral (58) and
piperdial (59), while L. necator9 produces velleral (58) and epi-piperdial (60) [27,
29-33]. This variation probably reflects enzymatic differences in the different
species, and has been used as a chemotaxonomic evidence for the division of L.
vellereus and L. bertillonii as different species [33]. The mechanisms of these
conversions have been thoroughly investigated by Sterner and Hansson [27, 31-
35], and have been suggested to constitute a chemical defence system that protects
the fruit bodies of these species against parasites and fungivors. Indeed, isovelleral
(55) has been shown to be an effective feeding deterrent toward opossum [36].
This is how Camazine and co-workers describe the reaction of opossums tasting
fungi (Lentinellus ursinus) containing isovelleral (55):
    The fungi were often spat out within seconds, and the animals frequently salivated
    profusely, frothed at the mouth, shook their head from side to side, and wiped

3   pepparriska
4   ekriska
5   luden vitriska
6   krusbärskremla
7   slät vitriska
8   skäggriska
9   svartriska

  their muzzles in their fur or on the bottom of the cage. When an unpalatable item
  was rejected, it was nearly always approached beforehand and apparently sniffed.

   Unsaturated dialdehydes are also found in the animal kingdom, for example in
molluscs belonging to the subclass Opisthobranchia. The molluscs have lost their
shell during the evolution, and so they lack the usual mechanical defensive system,
and yet few predators are known since a chemical defence system is employed [37-
39]. The chemical defence of Mediterranean nudibranchs have been thoroughly
investigated by Cimino and co-workers [5, 6, 40-44]. While nudibranchs in
general draw defensive chemicals from their preys, dorid nudibranchs elaborates
their own chemical defence by producing polygodial. For instance, the nudibranch
species Dendrodoris limbata and D. grandiflora store olepupuane (88) [45] in the
border of the mantle of these nudibranchs10 [42, 46]. The precursor olepupuane
(88) is transformed to polygodial (14), most probably by the contact with
predators (see Scheme 2) [46].

                          O                                   CHO


              Olepupuane (88)                        Polygodial (14)

              Scheme 2 Example on activated chemical defence in molluscs

   Sponges 11 are common on ocean floors, in particular along coast in shallow
water, throughout much of the world. Unsaturated dialdehydes, e.g. the scalarane
sesterterpenes 42-50, are sometimes found in sponges (see Appendix A). Sponge
chemical defences do not appear to be effective against specialist predators, since
nudibranchs specialise on certain sponges and may accumulate noxious sponge
compounds and store them in glands to be used for their own defence. The
chemical relationship between nudibranchs and sponges has been reviewed by
several investigators (see Rogers & Paul [39] and ref. therein). Indeed, the same
unsaturated dialdehydes have been isolated from nudibranchs as well as from the
sponges on which the nudibranchs feed [40].
   The occurrence of unsaturated dialdehydes is not limited to molluscs and
sponges in the kingdom of animals. The defence secretion of minor soldiers of the
termite Ancistrotermes cavithorax contains the dialdehyde ancistrodial (18), which
has been shown to be repellent to the ant Megaponera foetens, a major predator of
this species [47]. Field studies indicate that predation upon A. cavithorax is
markedly lower than that for other common termite species, most probably due to
this chemical defence system. Notably, the defence secretion of each soldier
contains tiny 2 µg ancistrodial (18).
   Another unsaturated dialdehyde, β -acaridial (84), has been isolated from
mites 12 [48, 49]. Interestingly, the dialdehyde 84 has a dual role, being both an

1 0 nakensnäckor
1 1 svampdjur
1 2 kvalster

antifungal defence substance and the sex pheromone for one of these mite species.
Since mites and fungi share stored-product habitats, the presence of β -acaridial
(84) should enhance the competitive strength of the mites. Notably, in cultures of
mites which secretes β-acaridial (84), moulds13 very rarely grow [50].
    In summary, unsaturated dialdehydes are widespread defence substances in
nature, used by many different organisms. While some dialdehydes, e.g.
merulidial (56), only occur in one species, others are widely distributed.
Polygodial (14), which is the most widespread unsaturated dialdehyde, has been
isolated from higher plants as well as liverworts and molluscs (see Appendix A). It
is interesting to note that the isomeric dialdehydes 18 and 19 (see Figure 5) have
completely different origins; ancistrodial (18) is produced by termites (vide supra)
[47], while 19 occurs in tropical green algae [51].

                                   CHO                     CHO

                                       CHO                     CHO

                              18                      19

        Figure 5 Compound 18 is produced by termites, and 19 by tropical green algae.

   The evolution of chemical defence substances has obviously converged into a
similar group of terpenoid secondary metabolites, and one might wonder how
"secondary" they actually are. Most probably the presence of unsaturated
dialdehydes in the organisms enhance their competitive strength.

How are the deterrent dialdehydes stored?
Two obvious questions may be posed: can the organisms protect themselves against
the dialdehydes, and how do the predators handle these deterrent terpenoids?
   There are at least two strategies for the storage of the dialdehydes. The first and
most straightforward is to store the dialdehyde in specially cavities or cells.
Hagendoorn et al. found unsaturated dialdehyde(s) in cavities in leaves and tepals
of the plant Polygonum hydropiper [52]. When a herbivore bites and chews the
plant, the cavity walls are crushed and the pungent dialdehyde is distributed.
Notably, comparable cavities containing polygodial-like compounds were absent in
the closely related species Polygonum persicaria. The need for specialised organs,
cells or cell compartments is likely, since polygodial (14) is reported to cause
damages to membranes [53].
   Other organisms have elaborated more sophisticated systems for the storage
and distribution of unsaturated dialdehydes. As discussed above, velutinal esters
are enzymatically transformed within seconds to unsaturated dialdehydes, upon
injury of the fruit bodies of several Lactarius and Russula species [26-28].
Consequently, the velutinal esters are the precursors of the dialdehydes. Anything
that attacks a fruitbody of a pungent Russulaceae species violently enough to
disrupt the laticiferous hyphae (where the velutinal esters are stored) will initiate

1 3 mögel

the production of the unsaturated dialdehydes, and will also be exposed to them.
In the same manner, halimedatetraacetate (86) (see Scheme 1) is the precursor of
the trialdehyde halimedatrial (40) [54], which displays a herbivore deterrent role.
   Protected forms of unsaturated dialdehydes are also contained in opisthobranch
molluscs. The formation of polygodial (14) from olepupuane (88) [45], which is
stored in the border of the mantle, was discussed above (see Scheme 2) [42, 46].
                                                           CHO                       CHO
                        CHO                                      CHO                       CHO

        Puulenal (89)                            Polygodial (14)         Isotadeonal (15)


                           CHO                                      CHO

        Onchidal (90)                           Ancistrodial (18)

                     OHC                                       OHC

          Oxytoxin 1 (91)                            Oxytoxin 2 (83)


                 92                             (Z)-Isolinaridial (81)

                 O                                        CHO                        CH2OH
                        O                                        CHO

                      93                         Polygodial (14)          Drimenol (94)

             Figure 6 The conversion of precursors to unsaturated dialdehydes

   Puulenal (89) is produced by the Hawaiian nudibranch Chromodoris albonotato,
and yields the dialdehydes polygodial (14) and isotadeonal (15) on hydrolysis (see
Figure 6) [37]. Similarly, the dialdehyde ancistrodial (18) is produced on
hydrolysis of onchidal (90), which occurs in the defence secretion of the
opisthobranch mollusc Onchidella binneyi. [55]. Interestingly, ancistrodial (18) also
occurs in the defence secretion of termites (vide supra) [47]. Oxytoxin-1 (91),
occurring in the opisthobranch mollusc Oxynoe olivacea, was found to be the
semiprotected form of the dialdehyde oxytoxin-2 (83) [43]. It is reasonable to
believe that the true chemical weapons used by these marine invertebrates are the
corresponding deprotected compounds, which often are too unstable to be
observed. For instance, onchidal (90) possesses biological activities which are
suggested to be due to its corresponding dialdehyde [56].
   Moreover, extracts of plants have been reported to contain metabolites, which
are transformed into unsaturated dialdehydes upon hydrolysis. The plant Linaria
saxatilis 14 contained diterpene 92 (isomeric mixture), which degraded to the
dialdehyde (Z)-isolinaridial (81) during chromatographic purification [57].
Another plant, Drimys winteri, produces the terpenoid 93, which on hydrolysis
gave polygodial (14) and drimenol (94) [58].

                                        O                      CHO OMe
                                            OMe                            OMe


                                  95                      96

          Figure 7 Dimethylacetals that yields unsaturated dialdehydes upon hydrolysis

   In addition, there are several reports (see Chapter 3) on the isolation of
dimethyl acetals, e.g. 95 and 96 (see Figure 7). These compounds can be regarded
as protected dialdehydes, since they can be transformed to their corresponding
dialdehydes by hydrolysis. However, they are most probably artefacts, formed
from the corresponding dialdehydes during extraction with methanol (see further
discussion in Chapter 3).

                              H                                O
                     S            NH2                                            S
               Lacrymator                                Lacrymator factors
             precursor (97)                                (98) and (99)

                     O                                             O
                     S                                             S
                                  NH2                                  S
               Alliin (100)                                 Allicin (101)

         Scheme 4 The formation of new compounds in onion and garlic upon injury

1 4 sporre fam. lejongapsväxter

    The conversion of one compound to another as a response to injury is naturally
not only limited to unsaturated dialdehydes. For instance, when an onion is sliced,
an enzyme immediately converts the lacrimatory precursor 97 to the lacrymators
98 and 99 (see Scheme 4) [59]. Similarly, alliin 100 is converted to the odoric
allicin 101 when garlic is injured. Both of these "tertiary metabolites" has a survival
value. The lacrymatory factors 98 and 99 are irritating and repugnant to certain
animals, while allicin 101 is antibiotic.

Antifeedant properties
Many unsaturated dialdehydes are antifeedants, which means that they inhibit
feeding. For instance, polygodial (14 ), warburganal (6 ), muzigadial (9) ,
cinnamodial (7) and 9-deoxymuzidadial (11), all isolated from plants, possess
antifeedant activity against larvae of African army worms Spodoptera littoralis and
S. exempta. [4, 9, 12, 60-62]. Analogously, 9-deoxymuzidadial (11) and
polygodial (14) possess antifeedant activity against the Australian carpet beetle
(Anthrenocerus australis) and the webbing clothes moth (Tineola bisselliella) [63].
Polygodial (14) also possesses potent antifeedant activity against a number of
Lepidopterous species [64-67], and inhibits aphid feeding and colonisation and
transmission of some plant viruses [65, 68-71]. However, polygodial (14) is not a
universal antifeedant, since it shows insignificant activity against tobacco
hornworm (Manducta sexta) the locust Schistocerca vaga (Scudder) [12], as well as
against wheat bulb fly Delia coarctata when used as a seed treatment [72].
   Polygodial (14), scalaradial (42) and 12-deacetoxyscalaradial (46), all occurring
in Mediterranean nudibranchs, have been shown to be distasteful to fishes [40, 73].
Rhipocephenal (82) and halimedatrial (40), occurring in a tropical marine algae,
have been shown to be toxic feeding deterrents against a herbivorous fish [22, 74],
while the fungal metabolite isovelleral (55) has been shown to be a feeding
deterrent towards the fungivor opossum (vide supra) [36]. Interestingly, the
combination of isovelleral (55) and stearic acid in the fungi enhanced the pungent
taste compared to isovelleral (55) alone [36]. Possibly, the long-chain fatty acids
facilitate the solubilisation of the hydrophobic isovelleral (5 5 ) in saliva.
Stearoylvelutinal (87) is the fungal precursor of several pungent dialdehydes, e.g.
isovelleral (55) and velleral (58) (vide supra). Remarkably, stearoylvelutinal 87 is
converted to velleral (58), not only by the enzymes in the fruit body, but also by
human saliva [75]. Even if the reaction in saliva is much slower than in fruit bodies,
this indicates that the predator may induce the formation of the deterrent by its
own saliva. However, stearoylvelutinal do not produce any hot taste to human
tongue, not even after several minutes [76].
   Several insects are adapted to toxins in their food. They can even accumulate
them, thereby making themselves distasteful for predators, e.g. birds. But can any
organism detoxify or metabolise unsaturated dialdehydes? As discussed above,
nudibranchs feed on certain sponges and may even accumulate noxious sponge
compounds and store them in glands to be used for their own defence. They can
also convert the sponge metabolites to new compounds. Cimino et al. suggested
that the marine mollusc Hypselodoris orsini converts the dialdehyde scalaradial (42)
to deoxoscalarin (102), and further to 6-keto-deoxoscalarin (103), which is
specifically accumulated in the glands of the mantle [44].

                                                      HO                                       HO
        AcO       CHO                         AcO           O                          AcO          O

        H                                     H                                        H

   Scalaradial (42)                   Deoxoscalarin (102)                    6-Keto-deoxoscalarin (103)

   Scheme 5 The proposed metabolisation of 42 by the marine mollusc Hypselodoris orsini

   Most likely, the mollusc is well protected against predators by the sponge
metabolites and the chemical transformation of the main sponge products could
be a detoxification process. On Guam, some populations of the sponge Hyrtios
erecta were found to produce the dialdehyde scalaradial (42) (see Scheme 5), while
other populations produced heteronemin (104) instead (see Figure 8) [39].
Glossodoris nudibranchs predating on the former population contained scalaradial
(42), as well as deoxyscalarin (105) and scalarin (106), while the nudibranchs
predating on the latter population contained 12-epi-scalaradial (45), as well as
heteronemin (104) (see Figure 6). Heteronemin (104) is very similar to olepupuane
(88), the precursor of polygodial (14) in Dendrodoria nudibranchs.

                                                      AcO                        AcO
                    AcO       CHO                   HO          O                          O

                                                                    OAc                    OAc
                    H                             H

            12-Epi-scalaradial (45)      Heteronemin (104)                 Olepupuane (88)

                                   HO                                 HO
                             AcO          O                     AcO          O

                             H                                  H

                        Deoxyscalarin (105)                 Scalarin (106)

                 Figure 8 Compound 45 and 104-106 are metabolites of marine
                          Glossodoris molluscs feeding on Hyrtios sponges.

   Also other organisms seem to be able to handle the dialdehydes. Jurgens et al.
reported that muzigadial (9) was metabolised to its hemiacetal by microorganisms
[77]. Similarly, a fungal parasite in fruit bodies of species of Russulaceae
transforms isovelleral (55) to its hemiacetal, as reported by Anke & Sterner [78].
This fungus, Calcarisporium arbuscula, resists high concentrations of isovelleral (55).
Clearly, the development of metabolic transformations of repellents is important
in the Darwinian struggle for survival.

Occurrence in spices and medicinal plants
There is no doubt that plants are good sources of biologically active natural
products. In addition, these phytochemicals are all biodegradable and, also
important, they are renewable. Naturally, medicine-men possess a wealth of
empirical knowledge on local plants and their use in folk-medicine. Drug
discovery have therefore often been based on ethnobotanical information, provided
by medicine men [79, 80]. Hence, it is not surprising that several pungent
unsaturated dialdehydes have been isolated from plants used in folk medicine and
as spices in food.
   The East African genus Warburgia (Canellaceae) consists of two species, W .
stuhlmannii and W. ugandensis. The bark extracts of these contains five different
dialdehydes: mukaadial (8), muzigadial (9), polygodial (14), ugandensidial (7)
and warburganal (6) [12, 81, 82]. Several of these dialdehydes possess broad
antimicrobial activity, as well as antifeedant activity against African army worms
Spodoptera littoralis and S. exempta, widely occurring African crop pests [9, 12, 62,
80, 83]. Interestingly, the barks of the two species are widely used in folk medicine
to alleviate toothache, rheumatism, general body pains, diarrhoea, malaria, and
also as spices in food [84, 85]. The bark of W. ugandensis is commonly known by
several different names in Kenya, depending on the local tribe, such as “apacha”
(Kakamega), “muthiga” (Kikuyu), “muziga” (Swahili), “olosogoni” (Masai),
“soget” (Kipsigis) “soke” (Tugen) and “sogomaitha” (Luo) [9, 86]. W. stuhlmannii,
which only occurs in the coastal areas of Kenya and Tanzania, is locally known as
“mkaa” or “mukaa” (Swahili) [81, 86]. Notably, four of the dialdehydes:
mukaadial (8), muzigadial (9), warburganal (6) and ugandensidial (7) have obtained
their names from the swahilian or latin names of these species.
   The folk-medicinal plant waterpepper15 , Polygonum hydropiper, contains the
intense pungent substances polygodial (14) and warburganal (6) in leaf and seed.
Waterpepper has sometimes been used as a substitute for pepper in Europe, and as
hot-tasting spice in China and Japan. The sprout of water pepper, called “mejiso”
or “benitade” in Japanese, is actually a well known relish for “sashimi” (raw fish)
[87, 88]. The expressed juice of the freshly gathered green leaves of water pepper is
used as a dip sauce together with Japanese traditional river fish, called ”ayu” [88].
Water pepper is also used as a folk medicine against tumours (scirrhous, hydropic,
and edematous tumours), uterine fibromas, and malignant ulcers [89, 90]. On the
other hand, it is known that this plant is toxic to fish [91, 92], pigs and sheep [92].
   The plant Drimys lanceolata contains polygodial (14) [93]. Its fruits are said to
have been used as a substitute for pepper in Tasmania [94]. The plant Linaria
japonica (Scrophulariaceae), once used as a Japanese folk medicine, has been
reported to contain the dialdehyde linaridial (78) [95]. The rhizomes of
Zingiberaceae-species16 has a world wide use in food flavouring. It shows that the
pungent dialdehydes aframodial (76) and its congener (+)-(E)-labda-8(17),12-
diene-15,16-dial (74) often occurs in these species. For instance, the plant
Aframomum daniellii grows in many regions of Cameroon and is locally known as
“achoh”. Its seeds, which contain aframodial (76), produce a hot taste on chewing,
and are widely used medicinally and as a food spice in Cameroon. The roots are
used as purgative [96]. Also three other Aframomum species contains aframodial

1 5 bitterpilört
1 6 ingefärsväxter

(76) [97]. The rhizomes of ordinary small type Japanese gingers (Zingiber
officinale) 17 [98], generally called “kintoki” and “yanaka”, also contain large
amounts aframodial (76), as well as the pungent compounds (6)-gingerol (107),
(8)-gingerol, (10)-gingerol and (6)-shogaol [99].

                            CHO                      CHO                     CHO

                             CHO                         CHO                    CHO

          Linaridial (78)         12-diene-15,16-dial (74)       Aframodial (76)

                                              O     OH

                                   OMe      [6]-Gingerol (107)

         Figure 9 Plant metabolites from Scrophulariaceae and Zingiberaceae-species

   The rhizomes of Alpinia galanga contain the dialdehyde 74, and have been used
for flavouring foods in the preparation of meat dishes and curries [100]. Actually,
the dialdehyde 74 occurs in the rhizomes of 13 different Zingiberaceae plants, in
particular in the genera Alpinia and Zingiber, of Malaysia [101].

Commercial applications
Unsaturated dialdehydes have been patented for drug use in a few cases. For
instance, aframodial (76) reduced cholesterol in blood and showed low toxicity
[102]. The same compound also inhibits leukotriene formation and possesses anti-
inflammatory activity [103]. Further, trichophyton-inhibiting pharmaceuticals
containing warburganal (6) and/or polygodial (14) have been used for treatment of
tinea pedis (athlete’s foot) [104]. Warburganal (6) has also been patented as virus
genome inactivator [105].
   The antifeedant properties of unsaturated dialdehydes have been shown to be
effective for reducing pest and disease damage of crop and woollen cloth [63, 69,
106]. This is of course of large commercially interest, since army worms
(Spodoptera species) cause >30% crop losses each year in India [107]. These
compounds are of natural origin and therefore more easily degradable than
commercial synthetic insecticides. Furthermore, it is unlikely that they will
accumulate and pollute the environment. Even if polygodial (14) is volatile and
unstable in field applications (half-life ca 3 weeks), field tests showed that it can be
used against aphids [108]. In an attempt to decrease plant virus transmission from
aphids, winter barley was treated with polygodial (14) (25 g/ha). Indeed, the
harvest increased from 3.83 to 5.22 tonne/ha. However, repeated applications were
needed to protect plants from aphids entering the plots over a period of several
weeks, why unsaturated dialdehydes were considered to be insufficiently persistent

1 7 ingefära

for commercial use [69, 71]. In another investigation, polygodial (14) and 9-
deoxymuzigadial (1 1 ) limited insect feeding damage to woollen cloth
considerably at rates comparable to that of synthetic insect proofing agents [63].
They might therefore be used to protect for instance a wool fabric from insect
attack. However, more work is needed to establish the feasibility of using
antifeedants such as these for the long term protection of animal fibre products
from insect damage.
    Different routes for large scale production of these compounds have been
investigated. For instance, racemic polygodial (14) is readily synthesised (see e.g.
Asakawa et al. [65]) and cell suspensions of Polygonum hydropiper have shown to
accumulate useful quantities of polygodial (14) [13]. Also work has commenced
towards isolating genes responsible for polygodial (14) production for use in the
genetic modification of crop plants for increased insect resistence [109].
   The use of unsaturated dialdehydes for mould-prevention in packaged foods is
not unexpected [110], since they possess significant antifungal activities. [83, 111].
The antifungal and antibacterial effects of polygodial (14) can be synergised by
anethole, perillaldehyde, safrole, methyleugenol and indole [112-115]. The
combination of polygodial (14) with these compounds may be a very effective
means of controlling troublesome human pathogenic fungi. These combination
effects have been discussed, but not fully explained [115]. Inversely, polygodial
(14) synergise the antifungal activity of maesanin and several antibiotics such as
actinomycin D and rifampicin, most probably by facilitating the transmembrane
transport of these chemicals into the cells [116, 117].
   Unsaturated dialdehydes have also found applications in other fields. Their
reactivity towards amines has made them useful as derivatisation agents for GLC-
analysis of chiral amines [118-120]. Furthermore, addition of polygodial (14) has
shown to improve mint flavouring for chewing gums, beverages, pharmaceuticals
and toiletries by suppressing bitterness and providing a cooling sensation with “a
longer lasting, fresher taste” [121].

Biological activity
Most of the unsaturated dialdehydes possess a wide variety of biological activity:
  • algaecidal [111,122]
  • allergen [123,124]
  • antibacterial [8,16,22,97,111,122,125-132]
  • antifungal [8,9,22, 53, 83,97,111-116,122,125,126,128-131,133-137]
  • anticomplemental [87]
  • antifeedant/deterrent [4,7,9,12,22,36,60,62-65,67,71,73,106,138-141]
  • anti-hypercholesteromic [102,142]
  • anti-inflammatoric [103]
  • antitumour-promoting [88]
  • cytotoxic [22, 73, 97,111,122,125,128-131,135,137,143-147]
  • electron transport inhibiting (in mitochondria) [86]
  • enzyme-inhibiting [83,128,129,131-133,137,142,146, 148-152]
  • fertilisation-inhibiting [22,153]

   •   helicocidal (snail-killing) - see molluscicidal
   •   haemolytic [154]
   •   membrane-toxicity [134,155-158]
   •   molluscicidal [60,81,138, 159]
   •   mutagenic [111,122,129,160-163]
   •   phytotoxic [65,111,122,164]
   •   piscicidal (fish-killing)[22,40,87,116,165,166]
   •   plant-growth regulatory [65,167-169]
   •   platelet aggregation inhibiting [128,146]
   •   pungent in mammals, including man [62,67,71,73,139,166,170-173]
   •   receptor affinity [7,71,132,171,174]
   •   skin irritant [15,175,176]
   •   synergistic [53,116,117]
   •   tumour-promoting [169]

   The biological activities have in many cases been shown to be strongly linked to
the unsaturated dialdehyde functionality. For instance, both the enal moiety and
the other aldehyde group were essential for the inhibition of tumour promotion on
mouse skin [88], as well as for the mutagenicity of isovelleral (55) in Ames
Salmonella assay [160]. However, the biological activities are not simply the result
of the reactivity of the aldehyde groups. Instead, large qualitative and quantitative
activity differences have been reported and small structural changes, like
stereoisomerisation, may affect the biological activity dramatically [65, 67, 111,
160, 161, 172, 174]. While polygodial (14) is one of the most potent natural
unsaturated dialdehydes, its epimer isotadeonal (15) is more or less inactive [12,
62, 67, 83, 156]. The isovelleral isomer 108 (see Figure 10) is less antibiotic,
cytotoxic and phytotoxic compared to isovelleral itself [111], and the mutagenic
activity of merulidial (56) is lost after isomerisation to 109, or by acetylation of its
secondary alcohol [122].

                      CHO                                         CHO

             Polygodial (14)        Isovelleral (55)       Merulidial (56)

                      CHO                                         CHO

            Isotadeonal (15)             108                    109

                                      Figure 10

Reactivity towards bionucleophiles
Warburganal (6) was observed to reduce the excitability of the chemoreceptors
and thereby strongly suppress the feeding response of Spodoptera exempta larvae to
sucrose-flavoured agar [4]. However, when the experiment was repeated with a
mixture of warburganal (6) and cysteine or dithiothreitol, no decrease in
excitability was observed. Also the mercaptide forming organomercurial, p-
(chloromercuri)-bensoate, gave a qualitatively similar reaction. Therefore, Ma
suggested that warburganal (6) might act as an -SH acceptor and thereby affect
energy transduction in the chemoreceptors. On the other hand it is known that the
administration of thiol-containing substances decreases the taste in man and
animals [177].
   It has been shown that all antifeedant dialdehydes taste hot and spicy to the
human tongue, whereas nonantifeedant derivatives are devoid of hot taste [60, 62].
Both the enal moiety and the other aldehyde showed to be essential for these, as
well as many other bioactivities. Hence, it is not possible to draw simple parallels
between these compounds and other α ,β -unsaturated aldehydes, e.g. acrolein or
croton aldehyde.
    Taniguchi et al. reported that several physiological effects due to polygodial
(14), e.g. inhibition of growth, alcohol fermentation, and papain activity appeared
to result from its irreversible reaction with sulfhydryl groups [83, 133]. However,
based on kinetic data, Sodano and co-workers proposed that the biological activity
of the unsaturated dialdehydes is primarily related to their ability to form adducts
with amino groups, rather than sulfhydryl groups on the receptors [141]. While
isotadeonal (15) was less reactive than polygodial (14) towards cysteine, 15 did not
react at all with methyl amine or lysine or β-alanine in buffer (pH 9). Polygodial
(14) was however very reactive towards the amines (see Scheme 6), and it later
found an application as a derivatisation reagent for GLC-analysis of enantiomeric
primary amines, e.g. (-)-amphetamine (1 1 0 ) [118-120]. In subsequent
investigations, several dialdehydes were reacted with methyl amine, and the
reactions were monitored by NMR spectroscopy [172, 178]. The reactivity, as well
as the antifeedant activity and the hot taste, showed to be dependant on the
distance between the aldehyde groups, at least for the drimane dialdehydes.
Sodano and co-workers suggested that the biological mechanism of hot tasting
and antifeedant activity of unsaturated dialdehydes may result from covalent
binding to primary amino groups of the chemoreceptive sites, rather than from
Michael addition of membrane sulfhydryl groups, even though both groups might
be available at the receptor site [172].

             CHO                                                              N
                         +   H2N


     Polygodial (14)     (-)-Amphetamin (110)

     Scheme 6 Polygodial (14) react readily with amines, e.g. (-)-amphetamine (110)[118]

   The high reactivity of unsaturated dialdehydes towards primary amino groups
(and sulfhydryl groups) is often proposed to explain their antibiotic activity, e.g.
their ability to cause membrane leakage [116, 134, 158], and their enzyme
inhibiting properties [151, 152]. In fact, their reactivity towards primary amines in
vivo may produce some of the natural pyrrols (molliorins), which are co-isolated
with the corresponding dialdehyde scalaradial (4 2 ) [179-182]. Feeding
experiments established that labelled ornithine (111) was incorporated in the
corresponding pyrrole molliorin-b (112) (see Figure 11) [183].

                                  AcO        CHO

                              Scalaradial (42)                         Ornithine (111)

                          H                 OAc

                                                                   N       AcO

                                             Molliorin-b (112)

    Figure 11 Molliorin-b (112) is formed in vivo in the sponge Cacospongia mollior,
              most probably in a reaction involving scalaradial (42) and ornitin (111)

   Nevertheless, other mechanisms have been considered. The first example of a
reaction between a bi-functional nucleophile and an unsaturated dialdehyde was a
reaction between cysteine methyl ester (113) and muzigadial (9) (see Scheme 7)
         HO    CHO                                                                       HO   N
                 CHO                                                                                  S
                                MeO2C                          Et 3 N                             H
                                      H2N         SH          CDCl 3

   Muzigadial (9)     L-Cysteine methyl ester (113)

            CHO                                                                               N
                                  H2N       Lys
                                                                                                  S   Cys
                                    HS      Cys
   Polygodial (14)

      Scheme 7 Muzigadial (9) gives a tetracyclic adduct in its reaction with the
               cysteine ester 113. Fritz et al.[187] suggests other possible adducts in
               reactions between dialdehydes [e.g. polygodial (14)] and peptides

   In earlier investigations, cysteine was supposed to react in Michael reaction type
additions with unsaturated dialdehydes, even if no adduct had been isolated [83,
133] . This expectation was reasonable, since it is known that cysteine reacts
preferably in 1,4-additions with α , β -unsaturated aldehydes [185]. Saturated
aldehydes however, readily give thiazolidine adducts with cysteine [186]. Another
mechanism for reactions between unsaturated dialdehydes and peptides was
suggested by Fritz et al. two years later [187]. This mechanism included both a
pyrrole-formation and a Michael-addition (see Scheme 7).
   The molecular mechanisms by which these type of compounds exert their
bioactivities is still not known, and many questions concerning the unsaturated
dialdehydes and their activity remain. For instance, their hot taste to the human
tongue has been suggested to be correlated to their reactivity to amines [172].
However, this does not convincingly explain the hot taste of isovelleral (55) and its
isomer 108 (see Figure 10), which was observed to react very slow with amines, but
taste as hot as polygodial (14) [172]. Furthermore, there is no theory that explain
why isovelleral (55), its isomer 108 and merulidial (56) are mutagenic, in contrast
to most other unsaturated dialdehydes, including the merulidial isomer 109 (see
Figure 10). Furthermore, why may small structural changes, like
stereoisomerization, affect the biological activities so dramatically? Different
configurations may of course affect the shape of the molecules and hence the
reactivity as well as their interactions with receptors. If the dialdehydes bind
specific to macromolecules, e.g. DNA and receptors, then their molecular shape
will be as important as their reactivity. Other characteristics than their reactivity
against amines and sulfhydryl groups might also influence their biological activity.
This will be further discussed in the thesis.

       H E AI M OF T H I S     investigation was to study how structural properties

T       affect the reactivity and bioactivity of unsaturated dialdehydes. We
        decided to compare warburganal (6), polygodial (14), isovelleral (55) and
merulidial (56a), with their isomers 121, isotadeonal (15), 108 and 109a (see
Figure 12). Compound 14, 55 and 56 were chosen since they possess different
skeletons and were naturally available. Compound 6 was chosen since it is an
interesting derivative of 14. The isomers 15, 108 and 109a have all been prepared
before and are known to possess less potent bioactivities compared to their
congeners. Compound 121 has never been assayed before, and has only been
prepared as a racemate.

        HO    CHO               CHO                                      CHO
                CHO                   CHO

  Warburganal (6)     Polygodial (14)          Isovelleral (55)   Merulidial (56a)

         HO   CHO               CHO                                      CHO
                CHO                   CHO

        121           Isotadeonal (15)               108               109a

                                         Figure 12

    The general intention was to compare the reactivity towards biological
nucleophiles (e.g. amino acids) of the 8 compounds in Figure 12, and to assay
them for various biological activities. In addition we planned to investigate the
influence of the absolute configuration on their affinity for receptor activities and
other bioactivities. Since isovelleral (55) is one of the most bioactive dialdehydes,

we wanted to prepare and compare (+)-55, (-)-55, (+)-108 and (-)-108 (see Figure

       H                     H                       H                 H
                  CHO                     CHO                 CHO                    CHO

                  CHO                     CHO                 CHO                CHO
       H                     H                       H                 H
         (+)-55                  (-)-55                  (+)-108           (-)-108

                                      Figure 13

   None of these compounds are commercially available. Hence they have to be
prepared by isolation, by transformation of other natural products, or by total
synthesis. If the natural source for the dialdehyde is readily available, isolation is
normally to prefer, and for some dialdehydes there are actually several sources (see
Appendix A). Otherwise one can choose from the numerous synthetic pathways
which have appeared in the literature during the last twenty years. The subject has
recently been exhaustively reviewed by Gustafsson [188]. In particular, there have
been extensive development of general synthetic routes to the drimane
dialdehydes, previously reviewed by Jansen & de Groot [189]. For instance, there
are about 20 synthetic routes to warburganal (6) published to date. The main
purpose of the preparations has been to get access to the chosen dialdehydes (vide
supra). For this purpose we have used the quickest and most convenient routes
published. Furthermore, some new synthetic routes have been developed.

Isolation of isovelleral (55), merulidial (56a) and polygodial (14)
The marasmane dialdehyde (+)-isovelleral (55) has been reported to occur in fruit
bodies of basidiomycetes [29, 36, 190, 191] and from submerged cultures of
another basidiomycete [154]. We isolated 55 from fruit bodies of Lactarius
vellereus FR ., collected in October in birch woods in the vicinity of Lund, and
followed the procedure published by Hansson et al. [192]. 10 kg of fresh fruit
bodies afforded 7 g isovelleral (55).
   (-)-Merulidial (56a) has only been isolated from one source, the basidiomycete
Merulius tremellosus FR . [193, 194]. We isolated 56a from submerged cultures of
Merulius tremellosus as described previously [193]. Recrystallisation from ether
gave light yellow rods (about 6 g) with melting point 115-117˚C. Notably,
Steglich and co-workers [194] reported a lower melting point (99˚C) for needles
of (-)-merulidial (56a), obtained by crystallisation from chloroform/ethanol 95/5.
   (-)-Polygodial (14) was first isolated in 1962 from the plant Polygonum
hydropiper L. [175], and has since then been isolated from a number of different
organisms, spanning from plants, through liverworts to molluscs (see Appendix A).
We collected 19.3 kg of Polygonum hydropiper in August in the vicinity of Lund,
and ground the fresh specimens in an ordinary meat mincer. The resulting mush
was extracted with EtOAc. The concentrated extract was subjected to trituration
with dichlormethane, followed by three chromatographical steps, including
passage through an ODS column, eluating with acetonitrile (to remove
chlorophyll) [195], and eventually crystallisation. 19.3 kg of fresh plant afforded
about 10 g (-)-polygodial (14).

Semisynthetic preparation of (-)-warburganal (6) (paper V)
(-)-Warburganal (6) has been isolated from the bark of the trees Warburgia
ugandensis and W. stuhlmanni [12, 60, 196], Drimys granadensis [197], Canella
winterana [10], and from the plant Polygonum hydropiper [87]. Since the former
species were unavailable, the latter should be the species to choose for extraction of
6. However, this plant only contains minor amounts of warburganal (6). Hence, an
efficient semisynthetic route from natural (-)-polygodial (14) to (-)-warburganal
(6) would be desirable, since 14 is readily available from natural sources, while 6 is
not. This particular semisynthetic route has actually never been reported.
    Attempts to oxidise the enolate of polygodial have been reported to fail, as well
as the oxidation of the TMS enol hemiacetal 114 (see Figure 14) via mCPBA
oxidation and treatment with QF [198]. We considered different semisynthetic
strategies such as allylic oxidation of natural (-)-polygodial (1 4 ). In our
investigation, attempts to oxidise polygodial (14) or compound 115 were
unsuccessful. We therefore prepared the diacetate 116b (R=OAc) and oxidised it
according to Hollinshead et al. [199].

           CHO                      O                            O                   OR
                  CHO                    OTMS                        OMe

  Polygodial (14)             114                          115                 116

                                         Figure 14

    (-)-Polygodial (14) was reduced [200, 201] to the diol 116a (R=H) [202],
which was converted to (-)-warburganal (6) in an excellent overall yield [60% from
(-)-polygodial (14), 5 steps] via acetylation, oxidation, deacetylation and oxidation
by the use of Urones protocol [203], originally developed by Ley and co-workers
[198, 199] (see Scheme 8). The spectroscopic data of compound 6 were identical
in all respects with those of natural (-)-warburganal (6).

        CHO                         OR                       HO       OR             HO   CHO
              CHO                                                                           CHO
                                         OR                             OR

                  a                           c                         e

Polygodial (14)                116a R=H                          117a R=Ac   Warburganal (6)
                        b                              d
                               116b R=Ac                         117b R=H

Scheme 8. a) LiAlH4 , ether; b)Ac2 O, pyridine; c) SeO2 , dioxan; d) K2 CO3 , MeOH;
          e) (COCl)2 , DMSO, CH2 Cl2 , Et3 N.

   Compound 116a could also be used as substrate in the route to warburganal (6)
reported by Tanis & Nakanishi [204]. However, the overall yield is lower [43%
from (-)-polygodial (14)] and the number of steps (ten) is higher. Notably, the
diol 116a can be used as a substrate towards the natural compounds cinnamolide
(117), isodrimeninol (118) and drimenin (119) (see Figure 15) [199]

                                   HO                     O
              O                         O                      O                 CHO OR

   Cinnamolide (117)    Isodrimeninol (118)         Drimenin (119)              120

                                            Figure 15

   Warburganal (6) has been reported to be prepared from compound 120 (see
Figure 15) [204-206] in two single steps (96% and 80% yields, respectively)
according to Okawara et al. [205]. Polygodial (14) may be transferred to
compound 120 by ketalisation [207, 208]. This route was not investigated, but
seems promising and convenient if the selective ketalisation can be optimised.

Preparation of (-)-isotadeonal (15) (paper V)
Polygodial (14) and its more stable epimer isotadeonal (15) can be transformed to
each other by treatment with base, e.g. alumina in refluxing dichloromethane [58,
209]. We obtained the same result by treatment with Cs2CO 3 at 55˚C in THF.
Although the epimers are diastereomers, toluene/MTBE (99/1) showed to be the
only mobile phase that afforded an acceptable separation on silica columns,
yielding (-)-isotadeonal (15) (54 %, higher Rf), and unchanged (-)-polygodial (14)
(41 %, lower Rf), which could be recycled.

Semisynthetic preparation of the warburganal isomer 121 (paper V)
We intended to prepare and assay the pure enantiomers of polygodial (14),
warburganal (6), and their epimers isotadeonal (15) and 121 (respectively), in
order to evaluate structure-activity relationships for these drimanes. (+)-121 has
never been prepared. However Kende & Blacklock [210] obtained the (±)-epimer
121 as a by-product in their total synthesis of (±)-warburganal (6) (see Scheme 9).

                                                                          HO   CHO

                   O       O            MeO
                               O                                     Warburganal (6)

                           a,b                           c,d              HO   CHO
                  122                                                            CHO

                  HO    CHO                  HO     CHO
                          CHO                         CHO

          Muzigadial (9)                      123

         Scheme 9. a) TMS(MeO)CHLi; b) KH, THF; c) m-CPBA; d) H3 O+

    On treatment with TMS(MeO)CHLi, ketone 122 was transferred to a
diasteromeric mixture of alcohols, which underwent elimination upon treatment
with KH in THF to afford a 1:3 mixture of (E)- and (Z)-enol ethers. Epoxidation
of the (Z)-enol ether gave a 1:4 mixture of α - and β -epoxides, which led after
hydrolysis to a corresponding 1:4 mixture of warburganal (6), and the 9β-epimer
121. Similarily, the (±)-muzigadial 9β-epimer 123 (see Scheme 9) was obtained as
a side product in total synthesis of (±)-muzigadial (9) [211].
Attempts to prepare compound 121 from (-)-isotadeonal (15), by using the same
route as for the preparation of warburganal (6) (see Scheme 8), were unsuccessful.
Allylic oxidation of the epi-diacetate 124 did not yield any 9β-hydroxydiacetate
125. Instead the aldehyde 126 was slowly formed as sole product in 67 % yield
(see Scheme 10). During the corresponding oxidation of the 9β -diacetate 116b,
9α-hydroxydiacetate 117a was the major product and the aldehyde 127 [62, 204]
was formed in minor amounts as a side product (see Scheme 10).

                                           OAc                 HO      OAc

                                                   OAc                    OAc

                                        124                      125

                                                    a               OAc


                OAc                  HO                             OAc
                      OAc                        OAc
             116b                      117a                      127

                         Scheme 10. a) SeO2 , dioxan, reflux

    The failure to obtain 9β-hydroxydiacetate 125 was probably due to steric
crowding around the 9β-proton, and to the ungoverned boat conformation of the
terpenoid ring in the transition state (see Figure 16). In comparison, the terpenoid
ring of compound 116b adopt the more governed chair conformation in the
corresponding transition state. Several investigators [205, 206] reported about
difficulties with LDA mediated deprotonation of the α -aldehyde 128, probably
due to steric hindrance around the 9β-proton (see Figure 16), while the β-aldehyde
129 was easily deprotonated with LDA, and subsequently oxidised with
MoO 5 ·HMPA·Pyr. to compound 130 [204-206]. At this point, we considered
other synthetic strategies towards epimer 121. Homologation of ketones to α -
hydroxyaldehydes is a well-known procedure, which has been reviewed by several
authors [212-214]. Hence, a substrate containing an ketofunction in C-9 should be
a suitable substrate for the introduction of masked acylanions from the less
hindered side, producing 9β-hydroxycompounds.

                                                          O                                                  H
                                                                   H                               H             H
                   9                                          Se
                                  OAc                                                                        R
                                                                         OAc                                         H
                   H                                                          OAc        R                   OAc
      116b             Se    O
               O                                   124                                                 124

                            CHO O                  CHO O                       HO       O
                                    O                         O

                            128                    129                          130

                                             Figure 16

   By oxidative cleavage with sodium periodate in THF/water, we managed to
transfer the triol 117b to the ketone 131a in a nearly quantitative yield (see Scheme
11). With this semisynthetic key substrate available, we now considered several
ways to introduce an aldehyde function.

                            HO      OH                                   O

                                        OH     NaIO4                                OH


                        117b                                           131a
                                             Scheme 11

    Kende and Blacklocks route [210] from ketone 122 to the 9β-epimer 121 was
discussed in Scheme 9 (vide supra). However, this route is unselective, affording
both 6 and 121. CN- is a useless formylsynthon in this approach, since it gives
preferentially 1,4-additions (see ref. 23 in [215]). Peterse & de Groot used
lithium-1,3-dithiane as a masked acylanion in a model synthesis towards α -
hydroxy drimane dialdehydes [216]. The adduct could be hydrolysed to the α -
hydroxyaldehyde by treatment with CdCl2 in acetonitrile and hydrochloric acid.
This procedure would be expected to yield epimer 121 when applied to ketone
122 (see Scheme 9), as suggested by Goldsmidt and Kezar (see ref. 7 in [217]).

                                    O                                               HO

          HO                                                  HO
                        OH                                                    OH

                                             Figure 17

   Another promising, and the most direct approach, includes the use of α-alkoxy
organolithiums, which are carbinyl carbanion equivalents. The adduct is easily
hydrolysed to afford a carbinylalcohol, which can be oxidised to the aldehyde.
This synthetic tool was used by Corey et al. [218] to introduce a carbinyl moeity
in a total synthesis of (±)-aphidicolin (132) (see Figure 17).
   Addition of 1-ethoxyethoxymethyllithium 135 to O-protected 131a would
after hydrolysis give the triol 137 (see Scheme 12). For this purpose the
ketoalcohol 131a was treated with t-butyldimethylsilyl chloride to give the
protected species (131b). The α-alkoxy organolithium reagent was prepared with
the Still protocol [219]. Thus addition of tributylstannyllithium to
paraformaldehyde followed by protection with α-chloroethylether (133) (prepared
from ethanol, paraldehyde and dry hydrogen chloride immediately before use) in
the presence of N,N-dimethylaniline gave the O-ethoxyethyl compound 134 (see
Scheme 12). Addition of n-butyllithium to 134 in THF at -78ºC caused rapid
tin/lithium exchange to yield the corresponding α -alkoxy organolithium reagent
135. Reaction with ketone 131b gave the expected product in 58% yield, (83% in
respect to consumed 131b) after chromatography.

                                      Cl   O
           O       O
                                           133                 O                           O

                                      HO                   O                          O
                       b,c                  d                            e
         Bu3SnH                  Bu3Sn             Bu3Sn                      Li
                                                               134                         135

                O                          HO                                 HO               OH
                                                   O       O
                             OR                        OTBDMS                                       OH
                                  g                                  h

     f         131a R=CH2OH                  136                                      137
               131b R=CH2OTBDMS
               131c R=CHO                                                         i

                                                                                  HO 11CHO


Scheme 12 a) HCl (g), EtOH; b) LDA, THF; c) (HCHO)n, THF; d) 133, CH2 Cl2 ,
            N,N-dimethylanilin; e) n-BuLi, THF; f) TBDMSCl, imidazole, DMF;
            g) 135, THF; h) HOAc/MeOH/water 2:2:1; i) (COCl)2 , DMSO, CH2 Cl2 , Et3 N.

   The uncomplete consumption of enone 131b, most probably caused by
formation of the enolate, was not unexpected owing to the strong basicity of the
reagent. One way to circumvent the enolisation might be by using an

organocerium reagent, prepared from the organolithium and anhydrous
cerium(III)chloride [220, 221]. Hence, compound 135 was added to a suspension
of dry CeCl3 in THF at -78˚C. After 50 min the ketone 131b in THF was added
and the mixture was stirred for several hours at -78˚C. However, no adduct (136)
at all was formed under these conditions, not even after several hours stirring at
room temperature. TLC analysis showed no traces of compound 134. Hence, the
tin/lithium exchange was successful. Whether the corresponding α -alkoxy
organocerium reagent was formed or not remains unclear. Assuming that is
actually was formed, it might have been to bulky to accomplish an addition to the
C-9 carbonyl. We did not investigate this route further.
   Hydrolysis of 136 with 2:2:1 HOAc/MeOH/water proceeded smoothly,
affording the triol 137 in 82 % yield (see Scheme 12). The overall yield from
ketone 131b (67% in respect to consumed 131b, two steps) should be possible to
increase by repeated experiments in larger scale, and by excluding the
chromatographic separation between the two steps. However, these steps were not
optimised because of the small amounts of material available.
   Finally, the triol 137 was oxidised (see Scheme 12) to obtain the dialdehyde
121 in 82 % yield. The overall yield from (-)-polygodial (14) to (+)-121 was 26%
(nine steps). The spectral data of 121 agreed with those reported by Kende et al.
[210], and the stereochemistry of 121 was confirmed by a NOESY experiment.
Correlations observed between 5-H and 11-H (but not between 5-H and 9-OH),
as well as between 15-H3 and 9-OH (but not between 15-H3 and 11-H,
established that 9-OH has β-orientation (see Scheme 12 for atom numbers).
   We were still curious on the lithium-1,3-dithiane approach (vide supra). On
treatment with lithium-1,3-dithiane in THF at -78 ˚C, raising to -20 ˚C for 3 h,
the ketone 131b gave the corresponding 9α-dithian adduct 138 as the sole adduct
in 86% yield (see Figure 18).




                                   Figure 18

    Unfortunately the dethioketalisation caused problems. Peterse & de Groot
[216] used CdCl2 in acetonitrile and hydrochloric acid for this purpose, but did
not report any experimental details. In our hands this treatment hydrolysed the
silylether, but did not affect the thioketal. A more widespread method to
hydrolyse thioketals includes treatment with Hg2+ (HgCl2, CaCO3, acetonitrile,
water), but a 1,3-dithiane is quite stable under these conditions [222], and de
Groot and co-workers failed using this method for their substrate [216]. Our
efforts to hydrolyse the thioketal 138 by treatment with CAN in aqueous acetone
solution (see e.g. [223]) failed. Because of lack of material we were not able to
investigate any other cleavage procedures [224]. Paulsen et al. [225, 226] reported
about a more convenient desulphurisation under milder conditions by using 4,5-
dihydro-2-lithio-5-methyl-1,3,5-dithiazine (instead of dithiane), which can be

deprotected with Hg2+ . This reagent is also slightly more reactive than lithium-
1,3-dithiane, and seems reasonable to use in future investigations.
   In summary, we have developed a new and more efficient route to (+)-121 than
the one previous published by Kende et al. [210]. With our procedure, the
dialdehyde 121 is obtained in four steps and 46% overall yield from the
ketoalcohol 131a. Following the Kende protocol, the dialdehyde 121 is obtained
in four steps and 35% overall yield from the aldehyde 131c (R=CHO, see Scheme
12), which in addition has to be prepared from the ketoalcohol 131a. I f
warburganal (6), rather than polygodial (14), is available, then 121 can be obtained
via an initial reduction of 6 to the corresponding triol 117b, followed by the
procedure above (see Scheme 11 & 12). Notably, according to several investigators
[199, 207], reductions of α -hydroxy dialdehydes or the corresponding diesters
may be very problematic.

Preparation of (-)-and (+)-isovelleral (55) (paper VI)
Two synthetic pathways to isovelleral (55) have been reported. One leading to the
natural enantiomer (+)-55 [227], and another yielding racemic (±)-(55) [228, 229].

          OH                      O                                   b       Cl                      c,d

                             e                                f                                       g
                                              CHO                                        CO2Me
             OH                                                                    140
         H                                    H                                    H
                            h                                 i,j                        O             k
                  CO2Me                             CO2H                                 CO2H
         H                                    H                                    H

                  O         l,m                                   n                           CO2Me
                      O                                 O                                    O

                      CO2Me q                           CO2Me r                                  OH

                      OTf                              CO2Me

                                          Isovelleral (55)

Scheme 13 a) PCC; b) HCl ; c) Mg, THF; d) 139; e) KH, THF, 18-crown-6; f) Trimethyl-
             phosphonoacetate, n-BuLi, DME; g) 235˚C; h) KOH, water; i) O3 ; j) DMS;
             k) (COCl)2 , PhH; l) LiCH2 CO2 Me; m) MsOH, PhH; n) dimethylsulfoxonium -
             methylide, THF; o) LDA; p) PhNTf2 ; q) Pd(OAc)2 , CO, PPh3 , MeOH, Et3 N,
             DMF; r) DIBAL, PhMe, THF s) (COCl)2 , DMSO, CH2 Cl2 , Et3 N.

The latter is outlined in Scheme 13 and leads to racemic isovelleral 55 in 12 steps
in a total yield of 15% from 140, which is obtained in three steps from 3-methyl-
but-2-enal (139).
   Hence, unnatural (-)-55 could be prepared either by adopting the former
synthesis to produce the (-)-enantiomer, or by the resolution of the racemic
mixture obtained in the latter. Since we were interested in some of the
intermediates formed in the racemic synthesis, we employed the latter to obtain
the racemic diol (±)-141. To our knowledge, there are only two reports dealing
with resolution of enantiomeric mixtures in synthesis of enantiomeric pure
dialdehydes. In the first one, (-)- and (+)-polygodial (14) were prepared by
treating the corresponding racemic diol (±)-116 (see Figure 19) with (-)-
menthoxyacetyl chloride in pyridine [67]. The diastereomeric ester mixture was
separated by chromatography, saponificated and oxidised to yield pure (-)- and
(+)-(14). In the second report, the similar problem was solved in a different way by
Asakawa and co-workers [65]. They separated the enantiomers by preparing and
crystallising the diastereomeric salts from chiral amines and the monoester (±)-142
(see Figure 19).

                            CH2OH                  CO2Me
                                  CH2OH                 COOH

                            116                   142

                                    Figure 19

    We applied the former strategy [67] on the racemic diol (±)-141, and
transferred them to their diastereomeric esters 143 and 144 by treatment with (-)-
menthoxyacetyl chloride in pyridine (see Scheme 14). The diastereomers 143 and
144 were separated by preparative, centrifugally accelerated, radial TLC (using a
Chromatotron) [230, 231] , and by MPLC. The only mobile phase that gave
acceptable separation showed to be CH2Cl2:n-BuOAc (300:1). 143 and 144 were
then compared with the diester prepared from the diol of the natural (+)-isovelleral
(55). The latter was identical in all respects with compound 143. Since no traces of
its diastereomer 144 could be detected by 1H NMR spectroscopy we established
that (+)-isovelleral (5 5 ) isolated from Lactarius vellereus is at least 99%
enantiomerically pure. The recovered diastereomers 143 and 144 were >99%
diastereomeric pure, as judged by 1H NMR. Compound 143 and 144 were then
saponificated in 5% KOH-MeOH solution, and Swern oxidised [229] to yield
(+)- and (-)-isovelleral (55) (respectively) in 85% yield over two steps (see Scheme
14). Their spectral data were identical in all respects to those of an authentic
sample of (+)-isovelleral (55).

Preparation of the isovelleral isomers (+)- and (-)-108 (paper VI)
In the course of a total synthesis of (+)-isovelleral (55), an unexpected thermal
isomerisation to its isomer (108) was discovered [227] The rearrangement
proceeds via an intramolecular ene reaction (see Scheme 15), and has previously
been investigated by Hansson et al. [192, 232].Notably, Magnusson and Froborg



                          O                                              O
          H                                              H
                                  O                                              O
                      O                                              O

          H                                              H
                  O                                              O
          143                 O                          144                 O
                      O                                              O
              b                                              b

          H                                              H
                      OH                                             OH

          H                                              H
                  OH                                             OH
              c                                              c

          H                                              H
                   CHO                                           CHO

                  CHO                                            CHO
          H                                              H
         (+)-55                                      (-)-55

         Scheme 14 a) (-)-Menthoxyacetyl chloride, pyridine; b) KOH, water;
                     c) (COCl)2 , DMSO, CH2 Cl2 , Et3 N.

attempted to perform this rearrangement in 1978, by heating neat isovelleral (55)
at 210˚C for 30 min, but obtained pyrovellerofuran (145) as the major product
    Using this technique, (+)- and (-)-108 were readily prepared by heating a
solution of (-)- and (+)-55 (respectively) in toluene for 1 h at 180˚C under N2
atmosphere in a high-pressure glass tube. The concentrated residue was purified by
silica chromatography, using toluene: MTBE (200:1) as the most suitable eluent,
yielding a 1:1-mixture of 55 (higher R f) and 108 (lower Rf). Isovelleral (55) could
of course be recycled to enhance the yield of 108.

Preparation of the (-)-merulidial isomer 109a (paper II)
On the analogy of isovelleral (55) (vide supra), merulidial (56a) and its acetate 56b
[194] have been shown to isomerise to isomer 109a and its corresponding acetate
109b, when heated in toluene [122, 192]. However, more by-products were
formed in these reactions than in the very clean isomerisation of isovelleral (55).

                 H                                                                      H
           H2C           O                                H                   H2C               O
                                  ∆                        O       ∆
                             H                                                                      H

                             H                                                                   H
                     O                                                                      O
            55                                                                    108




                 Scheme 15            Thermal isomerisation of isovelleral (55)

Notably, the ratio between the merulidial acetate 56b and its isomer 109b was ca
2:8, which differed considerably from the ratio (ca 6:4) between (-)-merulidial
(56) and its isomer 109a [192]. Apparently 109b is the favoured, more stable
isomer in the pair 56b/109b. Hence, in a preparative route to the isomer, acetate
56b is the preferred substrate. Sterner used this strategy in his preparation of
compound 109a, but failed with the final deacetylation, which proceeded in
discouraging 1~2 % yield [122].

                                       CHO                    CHO
                                          CHO                    CHO

                                       OR                     OR
                                 (56a) R=H           (109a) R=H
                                 (56b) R=Ac          (109b) R=Ac
                                 (56c) R=TBDMS       (109c) R=TBDMS

                                            Figure 20

   In our efforts to prepare larger amounts of the (-)-merulidial isomer 109a, we
considered different strategies. Since the diastereomers 56a and 109a are hardly
separable with normal flash chromatography and MPLC techniques, and since the
equilibrium for their acetates is forced towards the isomer 109b, the isomerisation
and deprotection of O -protected derivatives of 56a was investigated. The
deacetylation with ethanolysis in NaOEt-EtOH was reported to be unsuccessful
[122]. Therefore, the acetate 56b was treated with dry K2 CO 3 in anhydrous
MeOH. The substrate was consumed to yield a mixture of products, presumably
acetals, but no merulidial (56a). However, attempts to hydrolyse these acetals by
treatment of the crude product with HCl in ether failed, and no traces of 56a
could be detected by TLC or 1 H NMR. The deacetylation is of course
complicated by the presence of aldehydes, and to avoid these problems the
aldehydes should be protected before the deacetylation.

    In view of these facts, silylated derivatives of 56a should be preferred, since
desilylation with aqueous HF has been reported to proceed smoothly, without
affecting the aldehydes [234]. Attempts to transfer merulidial (56a) to its tert-
butyldimethylsilyl derivative, by treatment with tert-butyldimethylsilyl chloride in
DMF in the presence of imidazole [235], were unsuccessful. Another, more
powerful method yielding tert-butyldimethylsilyl ethers is to treat the alcohol with
tert-butyldimethylsilyl chloride in acetonitrile in the presence of lithium sulphide
[236]. Applying these conditions on merulidial (56a) afforded unfortunately not
the expected product, but a great number of by-products. In contrast, one single
product was obtained when four equivalents of tert-butyldimethylsilyl triflate was
added to a solution of merulidial (56a) in dichlormethane in the presence of 2,6-
lutidine. Interestingly, 56a was not completely consumed if less than four
equivalents of the silylating agent was used. In addition, the reaction mixture
became more sluggish if the silylating agent was added to the solution of
merulidial (56a), instead of the opposite. During workup and on contact with silica
during the purification, the product degraded spontaneously to a more polar
compound. Analysis by 1 H NMR established the latter to be the silylated
dialdehyde 56c. The former was not pure enough to be completely structure
elucidated by NMR. Mass spectroscopic analysis, however, revealed a molecular
ion from the main component to have the molecular weight 476.3142, suggesting
a disilylated merulidial derivative with the composition C27H48Si2O3. In the light
of this composition, its acid lability, the presence of several signals between 4.9-5.9
ppm as well as the lack of aldehyde signals in its NMR spectrum, it is reasonable to
presume this compound to be the disilylated cyclised hemiacetal 146 (see Figure

                                   O                 OHC

                             OTBDMS                     OTBDMS
                             146                       147

                                       Figure 21

   When a dichloromethane solution of the alkoxysilane 146 was stirred with silica
gel/10% aqueous solution of oxalic acid, it was smoothly hydrolysed to 56c in 20
minutes. The same treatment of the crude product, afforded compound 56c in
91% chomatographic yield. The merulidial silyl ether 56c was transferred to its
isomer 109c by thermal isomerisation in mesitylene for 1 h at 180˚C. The
concentrated residue was purified by silica chromatography, using
CH 2 Cl 2 /EtOAc (50/1) as the most suitable eluent, to afford the hydroazulenic
dialdehyde 147 (see Figure 21, 4%, higher R f), the isomer 109c (54%, lower R f)
and recovering 56c (17%, middle R f). Compound 109c was desilylated by
treatment with aqueous HF in CH3CN, affording the (-)-merulidial isomer 109a
in 81% yield after chomatographic purification.

       HE     T O XI COL O G I CAL      effects of α , β -unsaturated aldehydes are

T      undoubtedly connected to their ability to function as direct-acting
       alkylating agents (see Witz [237] for a review). They appear to react mainly
with thiol groups in 1,4-additions, and give 1:2 aldehyde-thiol adducts containing
a thiazolidine ring. In contrast, they react with amines only under special
circumstances, for instance alkylation of nucleosides and DNA [238-242]. Since
Schiff-base formation takes place much slower compared with attack by
sulphydryls, the latter will predominate if a choice is given.
   Correspondingly, the reactivity of unsaturated dialdehydes towards biological
nucleophiles have been suggested to be essential for their biological activity [83,
172]. Since they are tri-functional compounds, the potential exists for a more
complicated reaction pattern, including intermolecular crosslinking of proteins. As
it was shown that the biological effects of unsaturated dialdehydes could be
inhibited in the presence of cysteine or dithiothreitol [4, 60, 83, 133], they were
suggested to act mainly as SH-acceptors in Michael reaction type additions with
unsaturated dialdehydes, even if no such adduct ever has been isolated. This theory
was later revised by Sodano and co-workers, who proposed that the ability of
unsaturated dialdehydes to form adducts with amino groups, rather than
sulphydryl groups, is responsible for some of their bioactivities [67, 141, 172].

                                                 CH3                           CH3
            CHO                              N                             N


         14                            149                           150

                                  Scheme 15

   Their suggestion was based on reactivity studies between unsaturated
dialdehydes and methyl amine at pH 9.0 (see Scheme 15). The antifeedant
activity and the hot taste were correlated with the distance between the aldehyde

groups for the drimane dialdehydes; a short distance was found to be a
requirement for the formation of a charged azomethine derivative, e.g. 149 (see
Scheme 15), which in turn can be transformed to a pyrrole, e.g. 150. The capability
of the unsaturated dialdehyde in forming the charged azomethine derivative was
proposed to be the minimum requirement for exhibiting a hot taste.
   The reactivity towards amines has been suggested to be responsible also for
other biological interactions of the unsaturated dialdehydes. For instance,
scalaradial (42) is believed to inactivate bee venom phospholipase A2 by reacting
with lysine residues at or near the substrate binding site [151, 152]. Furthermore,
the ability of polygodial (14) to increase the cell membrane permeability has been
suggested to depend on its reactions with cell membrane proteins (e.g.
neurotransmitter receptors and G-proteins) [116, 134, 158, 243]. However, since
both the enal moiety and the other aldehyde are essential for, for instance, their
mutagenic and tumour promotion inhibiting activity [62, 88, 160, 161], there are
probably also other molecular mechanisms by which this group of terpenes exert
their bioactivities. For example, polygodial (14) is approximately as pungent,
antibiotic and cytotoxic as isovelleral (55) [111, 172], but it completely lacks
mutagenic activity in bacteria as well as in mammalian cells [160]. Obvious, the
two compounds exert their toxic effects in partly different ways on the molecular
level, and this may be due to differences in chemical reactivity towards
nucleophiles. We have therefore investigated the reactivity and instability of
unsaturated dialdehydes under biomimetic conditions.

Isolation of a natural adduct (paper I)
During our isolation procedure of merulidial (56a) from submerged cultures of the
fungus Merulius tremellosus (see Chapter 2), we obtained two new compounds: the
pyranone 151 and meruliolactone (152) (see Figure 23), both with new skeletons.
The new merulane skeleton of meruliolactone (152) resembles the lactarane and the
tremulane skeletons, and analogously, meruliolactone (152) is similar to the fungal
metabolites vellerolactone (153) and tremulenolide A (154) [244].
   Meruliolactone (152) has previously been shown to be formed during thermal
isomerisation of acetylmerulidial (56b) at 180°C in toluene/triethylamine [192]. In
this report, Hansson et al. suggested that meruliolactone (152) was formed by
elimination of the acetoxy group followed by prototropic shifts. In view of the
enzymatic conversion of marasmane sesquiterpenoids to marasmane and lactarane
dialdehydes and lactones [e.g. isovelleral (55) and vellerolactone (153)] in fruit
bodies of species belonging to Russulaceae [34] the co-formation of merulidial
(56a) and meruliolactone (152) is reasonable. Thus, meruliolactone (152) should be
considered to be a true natural compound.
   The structure of the second new compound, 151, was not easily elucidated. MS
and 13C NMR data suggested its composition to be C21H 24O 5, and by the use of
HMBC experiments, the terpenoid portion of 151 could be established. The
remaining part (C6H 4O 3) included one acetylic methyl group (δ 20.2/2.22 ppm),
one protonated olefinic carbon (δ 100.0/5.83 ppm), and four unprotonated carbons
(162.8, 162.8, 162.0, 101.6). Because of the few hydrogens, HMBC experiments
did not reveal the structure. Its IR spectra included one absorption at 1560 cm-1,
which indicated a furan or an pyranone. The absence of absorption around 3100

cm -1 excluded any furan, why we suggested the structure of 151 to be as outlined
in Figure 23.


                       CHO                                          O            CHO
                                            O     O

                    OR                    HO                                OH
              56a R=H            Triacetic acid lactone (155)        151
              56b R=Ac

                O    O                                                  O    O



          Meruliolactone (152)    Vellerolactone (153)          Tremulenolide A (154)

           Merulane (new)               Lactarane                   Tremulane

                                         Figure 23

    We considered different strategies to prove the structure. The compound could
for example be transferred to other derivatives by hydrogenation or by addition of
dimethyl acetylenedicarboxylate in a Diels-Alder reaction [245, 246]. A total
synthesis of compound 151 would have been more elegant, but probably also more
time consuming. An additional solution to the problem would be structure
determination by X-ray diffraction. This turned out to be feasily, and the structure
of pyranone 151 was shown to be identical to the one we had suggested.
Compound 151 was shown to crystallise in the monoclinic space group P21 (No.
4) with the unit cell a = 10.009, b = 9.225, c =10,381 Å, ß = 107.98° with the
volume 911.7 Å3. Figure 24 shows a stereo picture of the unit cell (for the crystal
structure see Figure 3 in Paper I). However, the absolute configuration could not
be established.
    We realised that compound 151 could be formed by the condensation reaction
between merulidial (56a) and triacetic acid lactone (155) (see Figure 23), which is a
natural product of polyketide origin produced by Penicillium and Pseudomonas
strains [247-251]. Indeed, 151 was formed slowly when merulidial (56a) was
stirred at room temperature with lactone 155 in buffer at pH 5.6 (the same
conditions as during the fermentation [193, 252]). When the experiment was
performed in refluxing ethyl acetate, compound 151 was formed quantitatively in
24 hours. Since this adduct was identical in all respects (including CD-spectra)

with previously isolated material, the absolute configuration of 151 was established
to be the same as for merulidial (56a).

             (Figure 24 is unfortunately not avilable in this www version)

                                     Figure 24

  Most probably pyranone 151 is formed chemically from lactone 155 and
merulidial (56a) during the fermentations, but 155 has not been reported from
Merulius tremellosus. Hence, we performed TLC and NMR analyses of daily
samples of continuously fermentations of Merulius tremellosus, and in fact detected
lactone 155 as a fungal metabolite. Consequently, Merulius tremellosus is the first
reported organism to produce significant amounts of triacetic acid lactone (155)
under normal fermentation conditions [193]. Since the fungus produce
considerable amounts of merulidial (56a), which is antifungal and possibly a
defence compound, it is intriguing that it also produces a metabolite that
inactivates merulidial (56a). Could 155 be protecting the fungus from its own

Reactivity towards
triacetic acid lactone (paper III & V)
The chemistry of triacetic acid lactone (155) has recently been reviewed by
Moreno-Mañas and Pleixats [253]. The preparation of lactone 155 was first
described in the pioneering work by Collie in 1891 [254], and it was a delight to
use this convenient procedure. Compound 155 is highly nucleophilic at C-2 [253],
and has previously been reported to react with aliphatic α,β-unsaturated aldehydes
[255]. However, the major product is formed by Michael addition of C-2 to the β-
carbon and the product corresponding to the pyranone 151 was only obtained as a
minor product when lactone 1 5 5 was reacted with crotonaldehyde.
Mechanistically, the reaction between merulidial (56a) and 155 may proceed by
two principal routes, as shown in Scheme 17.
   Since electrophiles always appear to alkylate C-2 in 155 [255-257], the lower
route in Scheme 17 seems most probable. Furthermore, this type of electrocyclic
ring closure is known from the literature [258].


     O         O                                           O          CHO
           2                                                                                                            O
                                                                   OH                                        O
               CHO                                                                                                          CHO
                  CHO                                          O

               OH                                          O               CHO
     56a                                                                                                         151


    Scheme 17 Possible intermediates in the formation of adduct 151 from 56a and 155

   Triacetic acid lactone (155) is a bi-functional nucleophile that to some extent
resembles for instance cytosine (156) and guanine (157) (see figure 25). Their
corresponding deoxynucleosides have, as well as DNA, been shown to react with
several mutagenic α ,β -unsaturated carbonyl compounds, for instance acrolein
(158), crotonaldehyde (159), malondialdehyde (160) and methylvinyl ketone
(161), to form cyclic adducts, e.g. 162, under biomimetic conditions (see Figure
25) [238-242, 259-262].

                               O                       O                               O

                           O                      HN                               N       NH

                                        OH                     H2N                 N   N        H2N
                               151                Cytosine (156)                   Guanine (157)

                          O                 O          O                       O           O

                     H                  H          H                                   H

                                                                   OH                                       OH
                         158                159        160                 161                        56a

                               O        OH                         O       OHC
                     N              N                      O
                     N         N        N                                  O
                     H                  H
                              162                              151

                                                       Figure 25

   These aldehydes have, in correspondence with some unsaturated dialdehydes,
e.g., merulidial (56a), been shown to be mutagenic without activation in Ames

Salmonella assay [160, 263]. It has been suggested that their mutagenicity or
possible tumourigenicity may relate to their reactions with nucleotides and DNA
[240]. In fact, in a reactivity screening study it was shown that all α,β-unsaturated
carbonyl compounds forming adducts were also mutagenic in Ames Salmonella
assay towards strain TA100 [242]. These reactions actually occur in vivo since the
corresponding adducts have been isolated from both rat and human urine. As
shown in Figure 25, the adducts between the nucleotides and α ,β -unsaturated
carbonyl compounds, e.g. 162, are structurally similar to pyranone 151, and
consequently triacetic acid lactone (155) reacts as a bi-functional nucleophile in a
similar manner as some nucleotides. This resemblance motivated us to use the
reaction between lactone 155 and some unsaturated dialdehydes as a model
system, in order to investigate how small changes in structure might influence the
chemical reactivity of these substances towards bi-functional nucleophiles. If an
addition of this kind of nucleophile is stereo selective, two diastereomeric
dialdehydes may react at different rates. This may then explain the differences in
the biological activities of a given pair of stereoisomers.
   We have compared the reactivity of four dialdehyde pairs (see Figure 26)
towards lactone 155, by measuring their half-lives in solutions buffered at pH 4.0,
5.6, 7.4 and 9.0. A large excess (20 equivalents) triacetic acid lactone (155) was
used to obtain pseudo-first order kinetics, and 5 % acetonitrile was added as co-
solvent for the dialdehydes.

         HO   CHO               CHO                                       CHO
                CHO                   CHO

  Warburganal (6)     Polygodial (14)          Isovelleral (55)    Merulidial (56a)

         HO   CHO               CHO                                       CHO
                CHO                   CHO

        121           Isotadeonal (15)               108                109a

                                         Figure 26

   The reactions are slow at neutral pH, but are catalysed by acid (see Table 1). As
some of the dialdehydes are autoxidised more rapidly at alkaline pH than they
reacted with compound 155, the half-life at pH 9.0 has not been calculated. For
practical reasons the reactivity of the dialdehydes can only been compared at pH
4.0 and pH 5.6. Merulidial (56a) and polygodial (14) were the most reactive
dialdehydes in this series. In addition, the difference in reactivity within the pairs
of stereoisomers turned out to be dramatic. Merulidial (56a) was at least 20 times
more reactive than its isomer 109a, and polygodial (14) at least 10 times more
reactive than its isomer 1 5 . Isovelleral (5 5 ) and warburganal (6 ) reacted
considerably slower, and were only slightly more reactive than their corresponding
isomers 108 and 121.

Table 1. Reactivity of unsaturated dialdehydes with lactone 155 in buffer at 37°Ca
                           pH 4.0:               pH 5.6:                pH 7.4:
           no.            T1/2 a (h)             T1/2 a (h)            T1/2 a (h)
            6                770                   1200                 > 2000
           121            > 2000                 > 2000                 > 2000
            14                 72                       180                480
            15                820                    > 2000             > 2000
            55                630                     1100              > 2000
           108               1200                     1700              > 2000
            56a                30                       140             > 2000
           109a               660                    > 2000             > 2000
a The spontaneous degradation has been subtracted.
Solutions were made 0.2 mM in dialdehyde and 4 mM in nucleophile. The reactions were monitored
by disappearance of dialdehyde, as analysed by HPLC.

    The result is somewhat unexpected, since isovelleral (55) is a potent mutagen
towards mammalian cells [162], and has been shown to cause base-pair
substitutions and frame-shift mutations in Salmonella typhimurium strains [160].
As mentioned above, there is a correlation between the mutagenicity of α , β -
unsaturated carbonyl compounds in Ames Salmonella assay towards strain TA100
(sensitive to base-pair substitutions), and their ability to form adducts with
deoxyguanosine [242]. Hence, isovelleral (55) was expected to be at least as
reactive towards this type of bi-functional nucleophiles as the non-mutagen
polygodial (14). Furthermore, if merulidial (56) is compared with isovelleral (55),
the former is more reactive (approximately 10 times) in this investigation, while the
latter possesses higher biological activities. Similarly, polygodial (14) is many times
more reactive than warburganal (6) in this investigation, while the latter is as least as
bioactive as 14 . In fact, the reactivities of the potent isovelleral (55) and
warburganal (6) are comparable to those of the less bioactive isotadeonal (15) [83]
and the merulidial derivative 109a. The bioactivities of isovelleral (3) and
warburganal (6) probably depend on other chemical interactions, since isovelleral
(55) and warburganal (6) are at least as bioactive as merulidial (56a) and polygodial
(14) in several bioassays [111].
    Nevertheless, the natural dialdehydes 6, 14, 55 and 56a are each more reactive
than their less bioactive isomers 121, 15, 108 and 109a. In order to understand
these differences, one has to consider the mechanisms behind the adduct
formations. The adducts between the dialdehydes and 155 were synthesised in
refluxing ethyl acetate, as this procedure enhanced the reaction rates and the
yields. Comparison by TLC and HPLC showed that the same adducts were
formed in buffer. Each dialdehyde yielded a single product (see Figure 27), except
for the polygodials 14 and 15, which gave the same epimeric adduct mixtures. The
facial selectivity probably arise from the preferred trans-decalin like (rather than the
less favoured cis-decalin like) transition state (see Figure 28). In addition, it is
reasonable that steric hindrance by the cyclopentane ring in the marasmane and
isolactarane dialdehydes favours addition from the exo-face of the double bond.

                          CHO           O                         HO   CHO       O

                                             O                                       O

                                O                                       O

                          CHO           O                         HO   CHO       O

                                             O                                       O

                                O                                       O

          O                         O
               O                         O
                                                                       CHO                      CHO

      O            CHO          O            CHO
                                                                  O          O             O          O

                                                                        O                        O

              OH                        OH

    Figure 27 Adducts from the reactions between lactone 155 and unsaturated dialdehydes.

   As the stereo selective control of nucleophilic additions influence their chemical
reactivity, this may also affect the biological activity of a given isomeric pair of

                                                     ‡                                                    ‡
                                             O                    O                  OHC
                                                 O           O
                          O                                                      O               OH

          Figure 28        Hypothetical trans-decalin like transition states in the reactions
                           between lactone 155 and unsaturated dialdehydes

   In summary, the presence and orientation of the unconjugated aldehyde
influences the reactivity of the unsaturated dialdehyde moiety towards lactone
155, without participating in the reaction itself. Although several QSAR-studies
have established the importance of the orientation of the unconjugated aldehyde,
this has until now been related to the dipole moments and the formation of
pyrolles with amines [155, 156, 161]. Consequently, this complicates the picture
that at a first glance appeared to be quite simple. It is possible that the reactivity
studied here may be important for the biological activity of certain types of

dialdehydes, for instance isolactaranes like merulidial (56a), but it is evident that
other reactions of the unsaturated dialdehyde functionality have to be considered.

Reactivity towards amino acids
As discussed in the beginning of this chapter, the reactivity of unsaturated
dialdehydes towards amino acid side chains containing nucleophilic groups has
been suggested to be important for their biological activities. However, only a few
reports deal on the reactivity and adduct formation between amino acids and
unsaturated dialdehydes have appeared [62, 83, 133, 141, 184, 187]. Brooks and
co-workers reacted polygodial (14) with phenylalanine, tyrosine and tryptophan
esters [118-120], and although the adducts were not further characterised, EIMS
suggested that the corresponding pyrroles, e.g. 163, were formed. In another
investigation, muzigadial (9) was reacted with L- cysteine methyl ester in
triethylamine and chloroform to afford the sensitive and instable tetracyclic
adduct 164, but no Michael addition (see Figure 29).
                                    MeO2C                                CO2Me
                   CHO                    N                  HO      N
                                                 N                        S
                         CHO                     H

           Muzigadial (9)           163                        164

                                                CHO                      CHO

                                                CHO                      CHO

          Polygodial (14)         Isovelleral (55)            108

                                     Figure 29

   Polygodial (14) has been shown to react readily with methyl amine, but
isovelleral (55) reacted considerably slower, and its isomer 108 hardly reacted at all
[172]. Nonetheless, 55 and 108 possess mutagenic activities and are more
antibacterial, algaecidal and cytotoxic than the nonmutagenic 14. We have
therefore investigated whether the marasmane dialdehydes [e.g. isovelleral (55)]
and the drimane dialdehydes [e.g. polygodial (14)] differ in reactivity towards
amino acids.

The reactivity of marasmane and drimane dialdehydes (paper IV, VII)
In order to compare the reactivity of marasmane dialdehydes with drimane
dialdehydes, polygodial (14), isovelleral (55) and its isomer (108) (0.2 mM) were
reacted with either of L-cysteine, L-lysine, L-arginine-HCl, L-alanine, L-tryptophan
and propylamine. The isovelleral isomer 108 was included, since its bioactivities
normally differs somewhat from the one observed for isovelleral (55). Ten
equivalents of nucleophile was used to obtain pseudo first order reactions, and the
reactions were performed in phosphate buffer (pH 7.4) at 37°C. The amino acids

were selected to obtain a maximum of diversity in side chains, and propylamine
was chosen as an simplified analogue to lysine. Acetonitrile (5%) was used as a co-
solvent for the dialdehydes. The disappearance of the dialdehydes was monitored
by HPLC (see Table 2).

Table 2. Reactivity of unsaturated dialdehydes towards five amino acids and propylamine
         in buffer (pH 7.4a) at 37°C
                                _________Unsaturated dialdehyde no._________
                                   14              55                108
                                    b               b
    Nucleophile                 T1/2 (h)        T1/2 (h)           T1/2 b (h)
    L-Cysteine                     0.28                 0.33                  0.22
    L-Lysine                       0.40               120                   280
    L-Alanine                      7.4              >1800                 >1800
    L-Arginine-HCl                 1.6               1300                   760
    L-Tryptophan                   n.t.               520                   880
    Propylamine                    0.60                71                     n.t.
  The pH has not been adjusted after the addition of amino acids and may therefore be somewhat
  The spontaneous degradation has been subtracted.
Solutions were made 0.2 mM in dialdehyde and 2 mM in nucleophile. The reactions were monitored
by disappearance of dialdehyde, as analysed by HPLC. n.t. = not tested

    Polygodial (14) is clearly reactive towards amines, but also towards thiols, since
alanine, lacking the sulphydryl group in cysteine, reacts considerably slower than
the latter. In fact, polygodial (14) react faster with cysteine than with any other
nucleophile in this assay. Propylamine reacts approximately like lysine, and
arginine somewhat slower, while alanine is nearly 20 times less reactive than lysine.
This indicates that polygodial (14) mainly reacts with unhindered side chain amino
groups, and only to some extent with α-amino groups.
    Isovelleral (55) and its isomer 108 show the same reactivity pattern, both being
very reactive towards cysteine, but more or less inert towards most other amino
acids. The more than 200 fold greater reactivity of isovelleral (55) and its isomer
108 towards cysteine compared with that towards lysine and propylamine
illustrates the specific reactivity of the isovelleraloids. In contrast, polygodial (14)
possess a more general reactivity towards amino acids. The differences between the
three dialdehydes in reaction times with propylamine and lysine are in accordance
with the observations made by Sodano and co-workers [172]. Our results suggest
that interactions with thiols are more important than interactions with amines for
unsaturated dialdehydes in general. As isovelleral (55) and its isomer 108 are as
pungent as polygodial (14) (see Table 8 in Chapter 4), the pungency can not be
linked to their ability to react with amines. Thus, the hot taste must be correlated
to some other mechanism, for example their reactivity towards thiols, or their
interactions at the vanilloid receptor (see Chapter 4).
    Interestingly, the reactions between cysteine and isovelleral (55) or compound
108 came to a standstill after approximately 45 minutes. In contrast, the reaction
between polygodial (14) and cysteine proceeded until all dialdehyde was
consumed. The experiment with isovelleral (55) and cysteine was repeated with
increased amounts of reagents (1.1 mM isovelleral and 10 equivalents of cysteine)
and co-solvent (acetonitrile, 30%). Again the reaction came to a standstill, and a

precipitation was formed. This was identified to be cystine, the oxidised form of
cysteine. In a similar experiment without isovelleral (55) no cystine was formed
during the same time. When the experiment with isovelleral (55) and cysteine was
repeated under inert gas the reaction proceeded without any standstill until all
dialdehyde was consumed. Isovelleral (55), or the adducts formed in the reaction,
are most probably responsible for the oxidation of cysteine to cystine, but
remarkably no reduced form of isovelleral (55) could be detected. As polygodial
(14), or its adducts with cysteine, did not oxidise cysteine, this is clearly a
difference between these two dialdehydes. Compounds that can generate oxygen
radicals are known to cause toxicity due to lipid peroxidation, and the cells use
cysteine and glutathione (both sensitive to oxidation) to protect themselves from
reactive oxygen species. Isovelleral (55) is known to damage cell membranes, and
thereby exert cytotoxic activity and its ability to oxidise cysteine indicates that it is
capable of forming reactive oxygen radicals, probably as a result of its autoxidation
(vide infra). This property may therefore be important for the biological activity of
the isovelleraloids.

Reactivity of drimane dialdehydes towards amino groups (paper V)
Although warburganal (6) is at least as bioactive as its congener polygodial (14) [9,
83, 88, 156], the presence of the α -hydroxyl group has been suggested to be less
important for the bioactivities of unsaturated drimane dialdehydes [62, 83, 264].
Cimino et al. proposed that the ability of unsaturated dialdehydes to form
adducts with amino groups is related to at least some of their bioactivities [67,
172], and we therefore reacted the drimane dialdehydes with L- lysine and L -
alanine to see whether they react at the same rate. For comparison, their less
bioactive C-9 isomers 15 and 121 were also investigated. To obtain biomimetic
conditions and to prevent base mediated epimerisation and degradation of the
dialdehydes, the reactions were performed in phosphate buffer (pH 7.4) at 37°C.
In order to prevent a high pH (by exceeding the buffer capacity) only two
equivalents of each amino acid was used, and acetonitrile (5%) was used as a co-
solvent for the dialdehydes. The disappearance of the dialdehydes was monitored
by HPLC (see Table 3).

Table 3. Reactivity of unsaturated dialdehydes towards lysine and alanine
         in buffer (pH 7.4) at 37°C
                      ___________Unsaturated dialdehyde no._____________
                           14          15              6          121
   Nucleophile          T1/2 a (h)  T1/2 a (h)     T1/2 a(h)    T1/2 a (h)
   L-lysine                   2.6            580               35           280
   L-alanine                 37              830              210           360
a The spontaneous degradation has been subtracted.
Solutions were made 0.2 mM in dialdehyde and 0.4 mM in nucleophile. The reactions were
monitored by disappearance of dialdehyde, as analysed by HPLC.

   The dramatic difference in reaction rate between polygodial (14) and its epimer
isotadeonal (15) was expected [141], since polygodial (14) easily forms pyrrols, in
contrast to isotadeonal (15) [172]. While 15 hardly react at all, 14 reacts rapidly
and more than 200 times faster than 15 in the presence of lysine and reasonably
fast also in the presence of alanine. Similarly, warburganal (6) reacts faster than

121, although the difference is less pronounced within this isomeric pair. While
polygodial (14) and warburganal (6) react much faster in the presence of alanine,
their isomers do not show the same specificity, probably due to their inability to
form pyrroles. The reaction rates for polygodial (14) are several times lower at
these conditions, compared to the previous experiment with 10 equivalents of the
lysine or alanine (see Table 2). This is probably due to differences in reaction order
or in pH. Interestingly, polygodial (14) is approximately one order of magnitude
more reactive than warburganal (6) towards lysine and alanine. Inversely, 121 is
twice as reactive as 15.
    The reactivity showed to be strongly affected by the presence of the α-hydroxyl
group, presumably due to steric interactions, or by induction effects. It is
interesting to note that these diladehydes showed the same reactivity pattern in
their reactions with triacetic acid lactone (155) (vide supra). This result is not in
complete agreement with their biological activities, since warburganal (6) is
approximately as bioactive as polygodial (14) [9, 83, 88, 156]. The reactivity of
unsaturated dialdehydes towards amines is therefore only one of several
interactions at the biomolecular level. For instance, Fritz et al. [187] suggested a
slightly different mechanism including a bi-functional nucleophile, e.g. a peptide
or an protein (as shown in Scheme 7 in Chapter 1). If the α -C-9 hydroxide is
displaced by an SN 2’ attack of the SH moiety on the β -carbon of the enal, the
resultant alkene would force the C-8 and C-9 substituents into an even closer
proximity, and thereby enhancing the pyrrole formation. Although Lam & Frazier
[184] showed that cysteine methyl ester gives 1,2-addition and pyrrole formation
rather than Michael addition, a cysteine side chain in a peptide does not include
any α-amino group, and thus would be expected to react differently. However, all
bioactivities of these compounds does not necessarily have to depend on the
formation of new covalent bonds. Some activities could certainly be caused by
other interactions, and in such cases the presence of the C-9 hydroxyl group be

Isolation of amino acid adducts (paper IV & VII)
Some of the drimane aldehydes have been shown to form pyrrols with amines and
thiazolidine derivatives with cysteine methyl ester (vide supra, Figure 29 and
Scheme 15) [118-120, 172, 184]. As isovelleral (55) showed a completely different
reactivity pattern in a previous experiment (vide supra), we have investigated its
reaction with L-cysteine and L-lysine to find out if it reacts differently compared to
the drimane dialdehydes.
   Our initial experiments were performed in phosphate buffer (pH 7.4) at 37°C
under inert gas, with ethanol (10 %) as co-solvent for isovelleral (55) (0.25 mM)
and 10 equivalents of cysteine. The disappearance of isovelleral (55) was monitored
by HPLC, and when all 55 was consumed, the reaction mixture was extracted
several times with ethyl acetate. The organic phase was dried and concentrated.
TLC analysis of the extract showed >20 spots, and all attempts to separate the
mixture by chromatography on silica gel were unsuccessful, partly due to the
decomposition of the material on silica gel. We obtained a similar mixture of
adducts when polygodial (14) was reacted with L-glycine in another experiment.
The reversibility of the adduct forming reaction, i.e. the dissociation of adducts to
the aldehyde and thiol/amine, is well known for α,β-unsaturated aldehydes [237].

    We therefore modified the experiment and reacted 200 mg isovelleral (0.43 M)
with one equivalent L-cysteine or L-lysine under inert gas in ethanol. The solvent
was chosen in accordance to the results of Sodano and co-workers [172], who
proposed that one molecule of solvent (water) is incorporated in the adduct
between methyl amine and isovelleral (55). Ethanol should be expected to
equivalent to water in this matter, besides being a better solvent for the
dialdehydes. The reactions were stirred at 55°C to increase the reaction rates. TLC
analysis showed a number (>20) of products, including one or two major adducts
(less polar) in both the reactions. The reaction was stopped, and the concentrated
reaction mixtures were partitioned between heptane and water. The organic phases
were dried, concentrated and purified by chromatography, to afford one major
adduct (165, see Figure 30) for the cysteine reaction, and two major adducts (166
and 167a, n=3) for the lysine reaction.

                                       OEt                                          OEt

                                       O                                            O
                              HN                                                    NH
                 H2N                                                     166

                                                                                                 O       O
                                           N                                                     N
                      N            n
                                                    EtO                                 165
                          167a n=3
                          167b n=1
                                                                                         HO      N
HO                            O

             HO                        OH                 H
                                                                       OAc                 164

                               N               OH

                          O                                    OH
                                                                                N                    H
     NH2                                                                            AcO

   Putrescine (169)                 Spirodihydrobenzo-                         Molliorin-b (112)
                                   furanlactam VI (168)

                                                   Figure 30

   The minor lysine adduct 166 is an imine, probably formed via an electrocyclic
reaction as outlined in Figure 30. Its structure was unexpected, but indicates a

diverse reaction pattern, far from existing theories, which might also explain the
large mixture of products. The major lysine adduct 167a (n=3) is a dimeric
pyrrole, congenerous to adduct 150 (vide supra, Figure 15) obtained in a reaction
between isovelleral (55) and methyl amine, but also an analogue to the natural
compound molliorin-b (112). Feeding experiments have shown that labelled
ornithine (111) (see Figure 11 in Chapter 1) was incorporated in the corresponding
pyrrole molliorin-b (112), (Iengo -79) most probably in a reaction between
scalaradial (42) and ornithine (111) (vide supra). Similarly, the fungal metabolite
spirodihydrobenzofuranlactam VI (168) is probably formed from L-lysine and two
terpenoid fractions [265, 266]. Molliorin-b (112) might also be formed in a
reaction between scalaradial (42) and putrescine (169) (see Figure 30), which can
be formed via decarboxylation of ornithine (111) in vivo. Indeed, diamines give the
same type of products as lysine with unsaturated dialdehydes. Isovelleral (55) was
reacted with 1,3-diaminopropane according to the same conditions as for lysine
(vide supra), and the major product was isolated and characterised to be 167b (see
Figure 30).
The major cysteine adduct 165 is analogous to compound 164, as shown in Figure
30. It is interesting to note that compound 165 and 166 contain an intact
cyclopropane ring. The cyclopropane opening reaction involves an attack from a
solvent molecule, and if this is a rate determining step, then adducts with an intact
cyclopropane ring, e.g. 165 and 166, would be expected to form easier than ring
opened adducts, such as 167a (see Figure 30). It should be noted that the
cyclopropane ring did not open up in the extremely slow reaction between the
isovelleral isomer 108 and methyl amine, as reported by Cimino et al. [172]. The
product was observed to be the charged azomethine derivative 170 (see Figure 31).
Isovelleral (55) reacted somewhat faster under similar conditions, and a pyrrole was
formed as the sole product. It is reasonable that this property of the isovelleraloids
is responsible for their specific reactivity, and it may also be important for their
bioactivities. The extremely toxic illudins are for example known to only react with
thiol nucleophiles [267].


                                               N    CH3


                                     Figure 31

   The adducts between amino acids and dialdehydes may very well possess
interesting bioactivities. Spirodihydrobenzofuranlactam VI (168) (see Figure 30) is
for instance a potent inhibitor of both HIV-1 protease and endothelin-1 (ET-1)
[265], while the 1:2 crotonaldehyde-cysteine adduct and the 1:1 trans-4-
hydroxypentenal:cysteine adduct exhibit carcinostatic activities [268, 269]. In
contrast to the corresponding free aldehydes, which are also carcinostatic but also
highly toxic, the adducts were practically nontoxic [269]. The reactive free
aldehyde is probably regenerated at the target site by a slow dissociation of the
adduct, providing a prolonged action [269]. It has also been suggested that the
aldehyde-thiol adducts may have bioactivities which are not based on dissociation
to the reactive aldehyde [270]. This proposal was based on studies which showed

inhibition of a cardiac glutathione-S-transferase by the GSH adduct of trans-4-
hydroxynonenal. The latter adduct also appears to have inhibitory activity on a
hepatic glutathione-S-transferase [271]. The bioactivities and biological importance
of the amino acid adducts of unsaturated dialdehydes could therefore be
interesting, but have not yet been investigated.

The results show a general correlance between the reactivity and for instance the
antibiotic effect of the drimane dialdehydes. The presence of a C-9 hydroxyl
group affects the reactivity considerably, and the C-9 configuration is obviously of
vital importance for their reactivity towards amines. While the drimane
dialdehydes show a more general reactivity towards amino acids, the
isovelleraloids react preferentially with cysteine, probably since the opening of the
cyclopropane ring is a rate determining step. Isovelleral (55) react with lysine and
cysteine in water or ethanol to form a number of products. Although the yields
were poor, the major adducts from the ethanol solution were isolated and
characterised to be heterocyclic compounds, formed by the reaction of both
aldehyde functions with hetero atoms of the amino acid. Consequently, there is a
general correlation between the reactivity of unsaturated dialdehydes towards sole
amines or sulphydryls, and for instance their antibiotic effect. However, their
antifeedant effect and their pungency can not be correlated to their reactivities
towards e.g., lysine and alanine. This indicates a more complex mix of interactions
at the molecular level. It should for example be noted that isovelleral (55) appear
to catalyse the oxidation of cysteine in presence of oxygen, while polygodial (14)
does not.

Instability and degradation
Unsaturated dialdehydes are sensitive and reactive compounds that may degrade
and oxidise during chemical and biological tests [272]. They may for instance
form acetals/hemiacetals with alcohols, epimerise, autoxidise, or react with the
biological assay medium. New compounds are formed which may be equally,
more, or less potent than the parent dialdehyde, and it is important to be aware of
what is happening. The degradation products may even be solely responsible for
the observed effect in some bioassay, while the dialdehydes are inactive. Since the
instability of unsaturated dialdehydes could be expected to affect their biological
activities, it was investigated.

Acetal formation
Unsaturated dialdehydes reacts readily with alcohols, e.g. methanol, to form
acetals [195]. Methanol is often used for extraction during isolation procedures,
and consequently there are several reports about dimethyl acetals of dialdehydes,
isolated from plants [57, 273-275], nudibranch [45], sponges [276, 277] and alga
[278]. Although the acetals are reported as natural compounds, the authors have
often considered that they might be artefacts. In methanol extracts, bioactive
dialdehydes may be inactivated by acetal formation. This has been discussed by
Taniguchi & Kubo, who avoided methanol extraction in the isolation procedures in
a screening project of East African plants [80]. There are other occasions when

methanol may occur as solvent, for instance during chromatographic separations,
and during bioassays.
   Isotadeonal (15) is not as soluble in water as polygodial (14), as reported by
Taniguchi et al. [83]. Polygodial (14) is proposed to react with water to form a
dihemiacetal, which would increase the water-solubility of 14. If 14 would undergo
this reaction more readily than 15, the difference should affect their movement
across the cell membranes, and could therefore explain their different bioactivities.
We have investigated if any of the dialdehydes warburganal (6), polygodial (14),
isovelleral (55), merulidial (56a), or their isomers 121, 15, 108 and 109a form
dihemiacetals in deuterated phosphate buffer solutions (pH 7.4), containing
CD 3CN (20%) as co-solvent. Interestingly, no dihemiacetals could be detected by
1H NMR, not even after 24 hours, in contradiction with Taniguchi’s propose. The
same phenomenon was observed for dialdehydes solved in deuterated
acetone/water mixtures [76].

Polygodial (14) was isolated from the plant, Polygonum hydropipe§r L. by two
independent groups in the early sixties [175, 200]. Interestingly, one of these
groups also isolated isotadeonal (15), the epimer of polygodial (14), from the same
plant [279]. Since the extract was treated with alkali during the isolation
procedure, one might speculate whether 15 is an artefact, formed by base-catalysed
epimerisation of 14 [58], or if different populations of the same species produce
different metabolites. Cortes et al. isolated 14 and 15 from the bark of the tree
Drimys winteri, but regarded 15 to be an artefact, produced during the isolation
process [176].
   We have investigated if isotadeonal (15) occurs in specimens of Polygonum
hydropiper collected in the vicinity of Lund, and extracted 3.3 g leaves with
hexane/ethyl acetate 4:1 [280]. 1H NMR analysis (500 MHz) of the concentrated
extract (0.043 g) show polygodial (14) and warburganal (6) (100:1), but no
detectable amounts of isotadeonal (15). Hence, the possibility that 15 is an artefact
rather than a true metabolite in Polygonum hydropiper cannot be excluded.
   Polygodial (14) epimerise readily under basic conditions. Schulte & Scheuer
obtained epimerisation of 14 to 15 when a sample of 14 was passed through a
QAE-Sephadex A-25 column, washed and prepared in the hydroxide form with
methanol/dichloromethane 1/1 [37]. Treatment of polygodial (14) and ent-
isocopal - 12-ene-15,16-dial (73) with alumina, a widely used solid phase in
chromatography, is known to procedure their epimers isotadeonal (15) and iso-ent-
isocopal -12-ene-15,16-dial (77) [58, 209]. Both 14 and 15 may consequently
epimerise during bioassays at alkaline pH. According to our experiences (see paper
V), the epimerisation is slow at pH 7.4, but appears to be substantial under more
basic conditions, for instance at pH 9. If for instance 14 or 15 is assayed for
biological activity during several days at alkaline pH, the activity may be
unexpected high or low due to epimerisation. Since 14 and 15 often differ
considerably in biological activity, this may cause serious consequences, in
particular if the results are evaluated in QSAR studies.

Autoxidation (paper II & V)
Some unsaturated dialdehydes autoxidise readily to form allylic oxidation
products. Fungal extracts, containing isovelleral (55) and velleral (58), were

observed to produce the autoxidation products 9-hydroxyisovelleral (171) and 9-
hydroxyvelleral (172) during storage at -25°C, and 171 was formed when 55 was
chromatographed on alumina gel [281]. Notably, autoxidation is catalysed on
alumina gel [282], but also occurs in solutions of the dialdehydes at room
temperature, according to our experience. This process is naturally a problem
during isolation procedures and during biological tests. There are indeed reports
concerning isolation of natural compounds which may have been formed by
spontaneous oxidation, for instance 173 from 74, or 53 from 52 (see Figure 32)
[154, 283, 284]. In one of these investigations, the fungi were extracted by soaking
the fruit bodies in ethanol for 45 days [284]. The possibility that oxidised
compounds are artefacts should therefore not be excluded. Another compound of
this type, 9-hydroxymarasmic acid (53), was however proved to be a true natural
product [154].

                       CHO                                                        OH
            HO                    HO           CHO                             CHO

              171                   172                          Marasmic acid (52)
                     CHO                      CHO

                                          O                                  O
                       CHO                    O

                                              OH                                 CHO
       (+)-(E)-Labda-8(17),         173                      9-Hydroxymarasmic acid (53)
      12-diene-15,16-dial (74)

                                    Figure 32

   In an autoxidation experiment of merulidial (56a), isovelleral (55) and their
isomers 56c, 108, 109a and 109c (see Scheme 18) on alumina, we observed
considerable differences in autoxidation rate (see Paper II). Merulidial (56a)
degraded within minutes, while its less bioactive isomers 56c, 109a, 109c were
considerable more stable. In contrast, isovelleral (55) and its isomer 108 showed to
oxidise at approximately equal rates. The same pattern was observed when the
degradations of dialdehydes in buffers (pH 5.6, 7.4 and 9.0) were monitored by
HPLC (see Table 4).
   Although all the compounds are reasonably stable at pH 5.6, they degrade at
alkaline pH. Obviously, basic conditions catalyse the autoxidations of the
dialdehydes. Merulidial (56a) has been shown to autoxidise to compound 174 and
175 (see Scheme 18) on alumina gel [282], and we found that these products are
also formed in the alkaline buffer solutions. Similarily, isovelleral (55) is
autoxidised mainly to compound 171, on alumina gel as well as in the buffers. In
addition to 171, minor amounts of the norvellerane 176 were also obtained during
a preparative oxidation of isovelleral under basic conditions. The oxidation
products are most probably formed via allylic oxidations of the parent
dialdehydes. The norvellerane 176 might be formed via the base-sensitive

hypothetical intermediate 177, a congener to the merulidial oxidation product

Table 4. Spontaneous degradation of unsaturated dialdehydesa in buffer at 37°C
                          pH 5.6:                pH 7.4:                 pH 9.0:
           no.            T1/2 a (h)             T1/2 a (h)              T1/2 a (h)
           56a              170                     23                      10
          109a             >2000                  1000                      63
            55              >2000                      390                 180
           108              >2000                      420                 280
           14b               980                       310                  71
           15b               700                       210                  47
            6                570                       100                  10
           121               360                        67                  34
a Solutions were made 0.2 mM in dialdehyde.
b The spontaneous epimerisation has been subtracted
The reactions were monitored by disappearance of dialdehyde, as analysed by HPLC.

                   CHO                                  CHO       EtO-
                  CHO      O2                           O
           55                                    177                                  176 OH

                   CHO                                  CHO

                  CHO                                   CHO
           108                                   171

           CHO                                    CHO                        O
              CHO                                    CHO                          CHO
           OR                                     OH                         OH
     56a R=H                                     174                        175
     56b R=Ac
     56c R=TBDMS


     109a R=H
     109c R=TBDMS

                Scheme 18 Autoxidation of isovelleral (55) and merulidial (56a).

   It is reasonable to suggest that the autoxidation of the unsaturated dialdehydes
proceeds via the formation of an allylic carbanion, which either react directly with
oxygen to the corresponding peroxide anion, or is oxidised to the corresponding
radical, that can react with oxygen to the peroxy radical [285]. We have compared
the autoxidation rates with the calculated enthalpy change for abstracting the C-9
proton of compounds 55, 108, 56a and 109a to form their corresponding C-9
anions (see Paper II). The enthalpy change for this is 6.6 kcal/mol less for
merulidial (56a) compared to its isomer 109a, while the difference between
isovelleral (55) and its isomer 108 is only 1.2 kcal/mol (less for the latter), as
predicted by AM1 calculations. Consequently, the C-9 anion of 56a should be
formed considerably easier than that of its isomer 109a, while one can not expect a
big difference between 55 and its isomer 108. This is in agreement with the
observed rates of the autoxidation of the compounds, and may be an explanation
to the observed differences in degradation rate.
   Since bioassays are normally conducted in aqueous media at pH~7, the
dialdehydes may be autoxidised to new compounds during the bioassays.
Interestingly, we observed that isovelleral (55) and merulidial (56a), both being
mutagenic, were autoxidised to form compound 174 and 171 (respectively) as
major products after six hours incubation in the medium used in the Ames'
mutagenicity test. Compound 174 and 171 are both more mutagenic than their
parent dialdehydes [122, 160], while the deformylated merulidial derivative 175 is
as mutagenic as merulidial (56a). Notably, the nonmutagenic 8-O-protected
derivatives 56b and 56c (see Scheme 18), autoxidised considerably slower than
merulidial (56a). Exactly the same pattern was observed for the isomer 108a. We
therefore propose that the compounds formed by the autoxidation of natural
unsaturated dialdehydes in bioassay media are partially responsible for at least the
mutagenic activities of the dialdehydes.
   As shown in Table 4, warburganal (6) and its isomer 121 were shown to degrade
considerably faster than polygodial (14) and isotadeonal (15) in buffer solution at
alkaline pH. As for isovelleral (55) and merulidial (56a) the degradation may be
caused by base-mediated allylic autoxidation. Attempts were therefore made to
investigate the autoxidation of 6 (but not 121, due to lack of material). As shown
above, alumina gel may catalyse the autoxidation of unsaturated dialdehydes.
Warburganal (6) was therefore dissolved in ether and stirred in the presence of
alumina gel. The dialdehyde degraded almost completely in 1 hour, forming one
major (178) (see Scheme 19) and one minor compound.
   Notably the latter degraded to compound 178 on the silica column during the
purification. Most probably, compound 178 is formed in an α-aldol rearrangement
[286], via the hypothetical intermediates 179 and 180, as outlined in Scheme 19.
The minor product, which degraded to ketoaldehyde 178 during the purification,
may be 179 or 180. This rearrangement is in agreement with the ring expansions to
D-homosteroids, reviewed by the Fiesers [287]. This ring expansion was first
observed when 17-hydroxy-20-oxo-steroids were treated with alkali [288, 289].
Interestingly, compound, 182, which is similar to 180, has been reported [211] to
form when a mixture of stereomeric alcohols, 1 8 1 , was treated with
HgCl2/HCl/CH3CN (see Scheme 19).

            HO    CHO               HO    CHO                   CHO                       CHO
                    CHO                     CHO                        CHO                      CHO

        6        Alumina            121                         14                        15

             O                                    O                                   O

                         CHO                                  CHO                               CHO

                 179                               180                                178

            HO                                            O

                            SBun HgCl2
             181                                          182

                                           Scheme 19

   Since isovelleral (55) and merulidial (56a) are autoxidised readily in a solution
of water/pyridine 1:1 (pH ~8), warburganal (6) was stirred in the same solvent
mixture for 24 hours at 37ºC. Two major products were formed (see Figure 34),
and were identified to be the carboxylic acid 183, and the natural product
mukaadial (8).

                   HO       CHO                   HO      CHO                HO   CHO
                              CHO                           COOH                    CHO
                   H                              H                          H
                                                      O                          OH
                    6                             183                        8

                                            Figure 34

   These compounds are most probably formed by allylic autoxidation, and were
shown to be identical with the major degradation products in aqueous buffer
solution at pH 7.4. It should be noted that compound 178, that was formed on
alumina (vide supra), was not detected in buffer or in water/pyridine. However, the
possibility that 178 is formed, and immediately degraded to several other minor
products (due to degradation in water) should not be excluded. The isolated
degradation products of warburganal (6) have not yet been assayed for bioactivity.
Since 6 often is at least as bioactive as polygodial (14), but less reactive towards
nucleophiles (vide supra), the formation of autoxidation products may modulate
the bioactivities of 6.

Polygodial (14) and isotadeonal (15) also degrade at alkaline pH (see Table 4).
However, in contrast to all other dialdehydes no major autoxidation product could
be detected in buffer (pH 7.4) or in water/pyridine. According to TLC analysis of
a water/pyridine solution of polygodial (14) stirred at 37ºC over three days,
polygodial (14), isotadeonal (15) (formed by epimerisation) and some extremely
polar or polymeric material in the starting spot on the TLC plate. The difference
between polygodial (14) and warburganal (6) in this respect is striking, and should
be investigated further.
    In summary, unsaturated dialdehydes are instable in solutions at alkaline pH,
and degrade by autoxidation to bioactive compounds. Small structural changes
have been shown to affect the degradation of the compounds considerably. We
propose that the degradation products are responsible for some of the bioactivities
of the dialdehydes. Furthermore, as mukaadial (8) has been reported to be isolated
together with warburganal (6) after five days extraction [10], the question if
mukaadial 8 is a true metabolite, or an autoxidation product, formed from 6 is

Instability in bioassay
As discussed above, the unsaturated dialdehydes can be instable during bioassays,
either because of autoxidation, or by reaction with the solvent or the assay media.
The latter possibility was recently discussed in a QSAR investigation of these
compounds [272]. In this study, several dialdehydes that are reasonably stable in
buffer are considerably less stable in media used to assay cytotoxicity (Eagle’s
minimum essential medium with L-glutamine, and Ham’s F-12, 1:1), in particular
in presence of fetal calf serum (10%). While the degradation in buffer probably
can be related to autoxidation of the dialdehydes (vide supra), their accelerated
degradation in assay media may be related to reactions with amino acids (and
proteins in media containing fetal calf serum).
   In order to investigate this, we stirred warburganal (6), polygodial (14),
isovelleral (55) and merulidial (56a) with either Ham’s F-12 medium, or artificial
Ham’s F-12 medium without amino acids, at 37°C The degradations of
dialdehydes were monitored by HPLC (see Table 5).

Table 5.    Spontaneous degradation of dialdehydesa in Ham’s F-12 medium (pH 7.9) at 37°C
                           without amino acids:         with amino acids:
                 no.            T1/2 a (h)                  T1/2 a (h)
                   6               n.t.                       0.67
                  14                43                        0.15
                  55                18                         3.5
                 56a               6.6                        0.83
 Solutions were made 0.2 mM in dialdehyde. The reactions were monitored by disappearance of
dialdehyde, as analysed by HPLC. n.t. = not tested

   All compounds degraded rapidly in presence of amino acids, and polygodial
(14) showed to be the most sensitive. Isovelleral (55) degraded considerably slower
than the other compounds, and warburganal (6) was more stable than polygodial
(14), which is in accordance with their reactivity towards lysine (vide supra). The
dialdehydes also degrades in absence of amino acids, most probably due to their

autoxidation. Merulidial (56a) was the most instable compound in this sense,
which is in accordance with its ability to autoxidise (see Table 4).
    As shown above, the dialdehydes degrade rapidly in Ham’s F-12 medium. This
medium is mixed with other constituents, e.g. glutathione and fetal calf serum,
during an assay. In order to gain a better understanding of the influence of
different constituents in assay medium on the stability of unsaturated dialdehydes,
we investigated the degradation of 55 in presence of different combinations of
serum and glutathione with the Ham’s F-12 medium, or PBS buffer at 37°C.
Isovelleral (55) was chosen, since it was the most stable compound in the latter
investigation (see Table 5), and its disappearance was monitored by HPLC (see
Table 6).

Table 6. Spontaneous degradation of isovellerala in different mediums at 37°C.
          medium                                                         T1/2 a (min)
          PBS buffer                                                  21000
          PBS buffer + glutathioneb                                      60
          Ham’s F-12                                                    210
          Ham’s F-12 + calf serumc                                       30
          Ham’s F-12 + glutathioneb                                      <1
          Ham’s F-12 + calf serumc + glutathioneb                        <1
  Solutions were made 0.27 mM in isovelleral, with acetone (1%) as co-solvent.
  Solutions were made 2.0 mM in glutathione.
  Solutions were made 10% in fetal calf serum.
The reactions were monitored by disappearance of isovelleral, as analysed by HPLC.

    The presence of fetal calf serum accelerated the degradation of isovelleral (55),
while the combination of glutathione and Ham’s F-12 medium (with, or without
serum) caused totally degradation of 55 within minutes (see Table 6). This is
remarkable, since isovelleral (55) degrades considerably slower in presence of
glutathione in PBS buffer at the same pH. The serum contains a complex mixture
of proteins, lipids, lipoproteins, etc., and it is reasonable that isovelleral (55) react
with sulphydryl groups in proteins, according to our previous results (vide supra).
Glutathione contains a sulphydryl group, and is known to react extremely fast with
α ,β -unsaturated aldehydes, e.g. acrolein and crotonaldehyde [237]. Nevertheless
isovelleral (55) degrades much faster in F-12 than in PBS buffer, in the presence of
glutathione. The reason for this synergetic effect remains to be investigated.

The thermal instability of merulidial in DMSO
When merulidial (56a) is heated in toluene, it rearranges via an intramolecular ene
reaction to its isomer 109a [192]. This reaction is slow at 100°C, and compete with
autoxidation in presence of oxygen. However, when 56a is stirred in DMSO at
temperatures at 100°C in presence of oxygen, meruliolactone (152) is formed as
the major product (see Scheme 20). The reaction was inhibited by inert conditions.
Notably, acetylmerulidial (56b) was reported to form meruliolactone (152) in the
presence of triethylamine in toluene at 180°C [192]. Most probably 152 was
formed by initial pyrolysis of the acetate group in 56b followed by an elecrocyclic
ring opening and a Cannizzaro reaction, or vice versa. It is well known that several
unsaturated dialdehydes give the Canizzaro reaction under alkaline conditions [81,
82, 170, 175, 290, 291]. Sulfonate esters are known to undergo elimination at
100°C in DMSO [292]. Is it possible that the secondary alcohol coordinates a

DMSO-molecule, and decompose (see Scheme 20) to the olefin in a electrocyclic
reaction? The olefin could then form 152 as described above. However, since the
reaction was inhibited under inert conditions, the Canizzaro reaction may have
been the initial step, followed by a series of electrocyclic shifts (see Scheme 20).

                                             H            S
                           OHC                        H

                CHO                       CHO                          O     O
                     CHO                         CHO

         56a    OH

               O                         O
                     O                        O                        O     O


    Scheme 20. Hypothetical mechanisms for the formation of 152 from 56a in DMSO.

Unsaturated dialdehydes are instable and reactive compounds. They are instable at
alkaline pH due to epimerisation and autoxidation, and they react readily with for
instance amino acids, serum and glutathione in assay medium. The dialdehydes
are also reactive towards nuclophilic solvents such as methanol. This chemical
instability must therefore be considered when results from biological assays of
unsaturated dialdehydes are compared.

       H E U NSAT U R AT ED         dialdehydes investigated in this thesis have been

T      assayed for a number of different biological activities. Some of the test
       results are included in paper II, IV, V and VI. Other results have been
published elsewhere, or are still unpublished, why they have been included in this
chapter. All the biological tests (except for the taste tests) have been performed by
our colaborators: Department of Biotechnology, University of Kaiserslautern in
Germany; Department of Biochemistry, Sct. Hans Hospital, Roskilde in
Denmark; Department of Physiology and Pharmacology, Karolinska Institute,
Stockholm in Sweden; and National Cancer Institute, Bethesda in USA.

Interactions with the vanilloid receptor
Hot peppers (Capsicum spp.) contain capsaicin (184) (see Figure 35), which is
responsible for the burning sensation in the human tongue [293]. Interestingly,
capsaicin has recently been shown to interact at a specific receptor [294]. This was
demonstrated by the specific binding of [3 H]-resiniferatoxin (RTX) (185), a
naturally occurring ultrapotent capsaicin analogue, and by the development of a
synthetic, competitive capsaicin antagonist, capsazepine (186) [295]. However, as
RTX (185) and capsaicin (184) only have the (homo)vanilloid fraction in common,
and the rest of these molecules differ dramatically, the receptor has been described
as the vanilloid receptor [296]. Notably, vanilloid receptors contain thiol groups
essential for RTX binding, and sulphydryl-rective agents, e.g. N-ethylmaleimide,
have been shown to activate capsaicin-sensitive neurones in cross-tachyphylaxis with
capsaicin [297].
   As several unsaturated dialdehydes are known to be pungent [62, 67], and
likewise react readily with sulphydryls (see Chapter 3), we have investigated
whether they interact with the vanilloid receptor [171]. In addition, we have
assayed the dialdehydes for pungency (minimum amount to cause a pungent
sensation) on human tongue (see ref. [171] for experimental details).
Phtaldialdehyde (187) was included in the test for comparison, since it has been
reported to be hot-tasting [172].


       H                                 O                              S        NH
       N                          O                   O

    OH                          HO       OCH3                                         OH

   Capsaicin (184)                    Resiniferatoxin (185)                 Capsazepine (186)

         HO    CHO                CHO                                                       CHO
                 CHO                      CHO

 Warburganal (6)        Polygodial (14)             Isovelleral (55)            Merulidial (56a R=H)
                                                                                           (56b R=Ac)
         HO    CHO                CHO                                                       CHO
                 CHO                     CHO

         121           Isotadeonal (15)                    108                        109a R=H
                                                                                      109b R=Ac
                          CHO                       CHO

                          CHO                      CHO
               HO                        HO
                                          188                               OH
                171                                                 174
                                        CHO                        CHO

                                      CHO           HO            CHO                 187
                      Velleral (58)                  172

                                             Figure 35

   The results (see Table 1) should be regarded as relative rather than absolute
numbers, since the definition of minimum pungent amount is subjective and require
calibration. However, this is this first investigation in which the relative pungency
has been investigated. In previous reports, a specific amount of the dialdehydes
have been reported to be pungent or non pungent [62, 67]. In this investigation all
the tested dialdehydes showed to be pungent, depending on the amount that was

Table 7.     Pungency (min. pungent amount) of 17 unsaturated dialdehydes, phtaldialdehyde and
             capsaicin on the human tongue, and their affinity for specific [3H]-resiniferatoxin
             binding sites in rat spinal cord preparations (see Figure 35).
                            Pungency            Pungency                 Affinity
Compound                   (µg/tongue)        (nmol/tonguea )          (IC50; µM)
6                               0.5                2.0                     6.8±2.4
14                              0.1                0.43                    7.6±0.9
15                              5                 21                      43.2±17.1
55                              0.5                2.2                     5.2±0.9
56a                             0.4                1.6                     1.2±0.4
56b                             0.4                1.4                     3.8±0.7
58                              0.1                0.43                    1.7±0.9
108                             0.1                0.43                    1.3±0.4
109a                          20                  81                    >100
109b                          20                  69                    >100
121                             8                 32                      51±12
171                             2                  8.1                     6.6±2.0
172                             5                 20                    >100
174                             5                 19                    >100
184                             0.02               0.066
187                             5                 37                        n.t.
188                             1                  4.0                      4.0±1.2
 Calculated from the measured [µg/tongue]-value, which only have one number of accuracy.
n.t. = not teated

   A good, although not perfect, correlation was found between the pungency on
the human tongue and affinity for vanilloid receptors in the rat spinal cord. These
results suggests that unsaturated dialdehydes possess their pungency by interacting
with vanilloid receptors on capsaicin-sensitive sensory neurones. Results from
additional investigations showed that isovelleral (55) interacts with vanilloid
receptors on capsaicin-sensitive sensory neurones. For instance, isovelleral (55)
induced a dose-dependent calcium uptake by rat dorsal root ganglion neurones
cultured in vitro, which was fully inhibited by the competitive vanilloid receptor
antagonist capsazepine (186). The affinity of isovelleral (55) for inducing calcium
uptake or inhibiting RTX binding was also in very good agreement with the
threshold dose (2.2 nmol) at which it provoked pungency on the human tongue.
Repeated applications of capsaicin (184) is known to cause desensitisation on
human tongue [298]. During several hours after a desensitisation, even a 5000-fold
increase above the minimum pungent amount did not elicit any burning sensation.
We have investigated this phenomenon on the pair isovelleral (55)/capsaicin (184),
and found that the minimum pungent amount of isovelleral (55) was 10-fold
increased after desensitisation with 4x200 µg capsaicin (184). Similarly, isovelleral
(55) showed cross-tachyphylaxis with capsaicin (184) when pungency of these
compounds were assayed in rat eye.
   The mechanism by which the pungent dialdehydes bind to vanilloid receptors
remains unclear, since they are structurally distinct from the known classes of
vanilloids. Even if sulphydryl-rective agents activate capsaicin-sensitive neurones in
cross-tachyphylaxis with capsaicin (184), it is by no means certain that such agents
excite capsaicin-sensitive nerves by activating vanilloid receptors. It is therefore not
clear whether the reactivity of unsaturated dialdehydes play any role in their

interaction at vanilloid receptors. In isomeric pairs (e.g., 56a/109a; 14/15, 6/121)
the more reactive isomer was also the more pungent, while polar dialdehydes are
less pungent than unpolar. The isovelleral isomer 108 was one of the most active
dialdehydes, and at least as active as for instance polygodial (14). In contrast, it
reacts very slow with amines (see Chapter 3). For the isovelleraloids, it therefore
seems more reasonable to correlate pungency with reactivity to sulphydryls (see
Chapter 3). Since most tricyclic and tetracyclic dialdehydes are “non pungent“, it
has been suggested that only smaller molecules, for instance the sesquiterpenoid
dialdehydes, can enter the receptor sites. However, de Rosa et al. reported that 12-
deacetoxyscalaradial (46) was pungent, while scalaradial (42) was not [73]. This
difference may very well facilitate its interaction to a taste receptor, but may also
affect the reactivity.
    It has been generally accepted that vanilloid-like activity requires the presence of
a homovanillyl substituent. The present study has provided new clues to the
understanding how vanilloid receptors recognise ligands, and how unsaturated
dialdehydes exert their pungent activity.

Bioactivities of the drimane dialdehydes (paper V)
The drimane dialdehydes (-)-warburganal (6), (-)-polygodial (14), and their
isomers (+)-121 and (-)-15 have been assayed for a number of biological activities.

Table 8. Antibacterial, antifungal, cytotoxic and phytotoxic activities of unsaturated
                                  _______Unsaturated dialdehyde no._______
                                   (-)-6    (+)-121      (-)-14     (-)-15
  BHK 21                         10-20         20-50             10-20        20
  L 1210                         10            20                10           20
  B16-F1                         10            20-50             20           20-50
  S. italica                     10            10                10           10
  L. sativum                     10            50                10           50
  B. brevis                      10            20                100          20
  B. subtilis                    10            20                100          20
  E. dissolvens                  100           >100              >100         100
  M. luteus                      50            50                10           20
  M. miehei                      5             50                10           20
  P. variotii                    50            >100              10           50
  P. notatum                     5             50                1            50
  N. coryli                      5             50                10           100
*   (LD100, µM), the tests were performed according to [299]
**   (IC50, µg/plate), the tests were performed according to [300]

   In general, warburganal (6) is more antibiotic than its isomer 121. Polygodial
(14) is more cytotoxic and antifungal than its isomer 15, but less antimicrobial.
Warburganal (6) and polygodial (14) possess approximately the same activities,
except for the antimicrobial activity. Similarly, 121 and 15 possess almost identical
activities. The drimane dialdehydes are in particular antifungal, while their other

activities differs between different investigations. Warburganal (6), polygodial (14)
and isotadeonal (15) have been reported to lack antibacterial activity in one
investigation [83], while 14 and 15 possessed potent antibacterial activities in
another report [111]. Furthermore, they possess lower activity in comparison to the
isovelleraloids in all these bioassays (see Table 9), which indicates an important
difference between these two classes of dialdehydes.
    The results are not in complete agreement with the reactivities observed in this
investigation (see Chapter 3), as polygodial (14) is approximately one order of
magnitude more reactive than warburganal (6) towards lysine, alanine and the
lactone 155. However, within the two couples of epimers the agreement is better.

Enantiospecific bioactivities of the isovelleraloids (paper VI)
We attempted to compare the influence of the absolute configuration on activities
that possibly are caused by a specific influence on molecular targets such as
receptors. The enantiomers of isovelleral (55) and its diastereomer 108 were
therefore prepared (see Chapter 2) and have been assayed in a number of biological
   Interestingly, (-)-55 is as pungent as its diastereomer (-)-108, while a fifth-fold
higher dose is needed of their enantiomers (+)-55 and (+)-108 to give the same
effect. However, no significant difference in the affinity to the vanilloid receptor
between the enatiomeric pairs (+)-55 / (-)-55 or (+)-108 / (-)-108 could be
observed. Nevertheless, there is a significant difference in pungency between these
enantiomers. In comparison, both enantiomers of polygodial (14) have been
reported to be pungent [65, 67].
   The unsaturated dialdehydes also have an specific affinity for the dopamine
D1 receptor in the CNS, but not to other CNS receptors [174]. Compound (+)-
55 and (-)-108 showed comparable affinity for the dopamine D1 receptor, while
(-)-55 was a slightly less potent inhibitor. However, (+)-108 was shown to be ten
times less potent, and also differs from the other isovelleraloids by not fully
inhibiting the binding of 3H-SCH 23390. These results indicate that (+)-108 has
different affinities for the various dopamine D1 receptor subtypes.
   The antibiotic, phytotoxic and cytotoxic activities of these enantiomers are
comparable, except for the remarkable low antifungal activity of (+)-isovelleral
(55) against the deuteromycete Mucor miehi. This exception is significant and
may be due to enantiospecific interactions.
   The results from Ames' mutagenicity assay with the Salmonella typhimurium
strain TA98 were remarkable. The mutagenic response of (+)-55 was 244
revertants per µ g, while that of (-)-55 was only 23 revertants per µ g (same test
procedure). (-)-108 has previously [160] been shown to possess ca 10% of the
mutagenicity of (+)-55, and (+)-108 was found to be completely devoid of
mutagenic activity in this investigation towards TA98. Natural (+)-isovelleral (55)
appears to interact enantiospecifically with either the genetic material, or some
active site of importance for the DNA replication in the cells. (-)-108 has been
reported [162] to be devoid of mutagenic effect towards mammalian cells, while
(+)-isovelleral (55) showed to be a potent mutagen, also in comparison to well-
known mutagens such as ethyl methanesulfonate. The mutagenic activity of the
isovelleraloids is certainly one of the most fascinating properties among all of the

Table 9. Receptor, antibacterial, antifungal, cytotoxic and phytotoxic activities of unsaturated
                                 _______Unsaturated dialdehyde no._______
                              (+)-55     (-)-55         (-)-108      (+)-108
Receptor affinity
  Vanilloid*                2.7 ± 0.3    2.4 ± 0.6    1.3 ± 0.4    0.9 ± 0.2
  Dopamine D1 **            0.29 ± 0.011 0.48 ± 0.012 0.26 ± 0.018 2.26 ± 0.56

  Pungency***               0.5                0.1               0.1               0.5
  BHK 21                    2-5                1-2               2                 1-2
  L 1210                    5                  1-2               2                 1-2
  B16-F1                    2-5                2-5               5                 1-2
  S. italica                1                  1                 5                 5
  L. sativum                5                  5                 5                 5
  B. brevis                 1                  2                 5                 2
  B. subtilis               1                  2                 5                 2
  E. dissolvens             10                 10                >100              50
  M. luteus                 5                  5                 10                5
  M. miehei                 10                 0.1               0.5               0.5
  P. variotii               10                 5                 20                5
  P. notatum                1                  1                 1                 1
  N. coryli                 1                  1                 1                 1
  TA98                      244                23                #                 0
*        affinity (IC50, µM) for specific [3H]-RTX binding sites in rat spinal cord preparations, the
         tests were performed according to [171]
**       inhibition (IC50, µM) of the specific binding of 3H-SCH 23390 to the dopamine D1
***      Pungency (µg/tongue), the tests were performed according to [171]
****     (LD100, µM), the tests were performed according to [299]
*****    (IC50, µg/plate), the tests were performed according to [300]
******   Ames' mutagenicity assay towards the S. typhimurium strain TA98 (revertants per µg)
#        (-)-108 has been shown to possess only ~ 10 % of the mutagenicity of (+)-55 [160]

   In summary, the drimane dialdehydes appear to behave differently from the
isovelleraloids in biological test systems as well as in reactivity studies. The
isovelleraloids appear to exert their biological activities by both being generally
toxic and being selective towards receptors. The general toxicity does not differ
between the enantiomers, but their mutagenicity, pungency and affinity for the
dopamine D1 receptor are highly enantiospecific. The cell toxicity most probably
depends on the general reactivity of the dialdehyde functionality towards
bionucleophiles. In contrast, their selective mechanism is strongly influenced by
their configuration, but probably not by reactivity. The findings presented in this
thesis certainly warrant further investigations of the isovelleraloids. The fact that
the mutagenicity of (-)-108 was eliminated by inversion of the absolute
configuration, while the toxicity of 108 remained unchanged, is astonishing. By

identifying the factors that determine such differences, it may be possible to
distinguish between the different kinds of bioactivities more efficiently.

       ERPENOIDS          CONT AI NI NG           an unsaturated 1,4-dialdehyde

T       functionality are products of an evolution that has been going on from the
        beginning of times. They are produced by a number of evolutionary
completely separated organisms, from termites to plants. It appears as if evolution
in the past has made the crucial experience that secondary metabolites containing
an unsaturated 1,4-dialdehyde functionality possess properties that gives the
producing organism some kind of advantage over a “less gifted” organism. This
possibility alone strongly motivates the detailed studies of the unsaturated
dialdehydes that have been carried out in this investigation.
   Unsaturated dialdehydes possess a number of bioactivities, and as they also are
electrophilic and reactive towards thiols and amines, they may react as
electrophiles in vivo. While the drimane dialdehydes, for instance polygodial (14)
and warburganal (6), are generally reactive against various nucleophiles, isovelleral
(55) and isovelleraloids show specifically reactivity towards thiols. The dialdehydes
also react with bi-functional nucleophiles, such as triacetic acid lactone (155).
Remarkably, the unconjugated aldehyde influence the reaction rate significantly,
without participating in the reaction with 155. Generally, the α-substituent to the
unconjugated aldehyde, and its configuration, has been found to be of major
importance for the reactivity.
   The isovelleraloids are differently reactive, probably due to the cyclopropane
ring. Compared to polygodial (14), they react quite slowly with amino groups to
form pyrrole derivatives. This reaction involves an opening of the cyclopropane
ring, which may be a rate determining step, since the isovelleraloids react
considerably faster with thiols to form adducts, without opening of the
cyclopropane ring.
   The dialdehydes are sensitive and may degrade in solution, for instance by
reacting with methanol, which may deactivate them in bioassays. Polygodial (14)
and isotadeonal (1 5 ) are epimerised to each other at alkaline pH. This
epimerisation may affect their reactivity and bioactivity. Merulidial (56a),
isovelleral (55) and some of their derivatives, are sensitive to autoxidation. The
oxidation products formed possess potent biological activities, and may in some
cases be more active than their parent dialdehydes. It is therefore reasonable that

such autoxidation products are at least partly responsible for the activities of
unsaturated dialdehydes. It should be noted that merulidial derivatives, which are
less sensitive to autoxidation, also possess less bioactivities. Polygodial (14) and its
congener warburganal (6) both degrade in solution in the presence of oxygen, but
while the latter is transformed to allylic autoxidation products, polygodial (14)
most probably oxidise to polymeric material.

                       CHO                  CHO
                         CHO                  CHO

     -                              HO

                       CHO                  CHO                  CHO               CHO
                         CHO                  CHO                  CHO                CHO

     +                              HO                                                     189


    ++                     (+)-55


    Figure 36 Relative mutagenic effect of natural and synthetic dialdehydes [160, 163, 301, 302]

   Consequently, there are striking differences in the reactivity of the dialdehydes,
and it is difficult to correlate the observed bioactivities with a certain type of
reactivity. Nevertheless, the natural dialdehydes (14, 55, 56a and 6) are in this
study more reactive and more bioactive than their unnatural isomers 15, 108, 109a
and 121. Some general molecular features (molecular shape and lipophilicity) may
therefore be responsible for some differences between the drimane dialdehydes
and the isovelleraloids, while the orientation of the unconjugated aldehyde and the
diheder angle can be correlated with the difference in activity between the isomers.
In addition, their polarity is also of importance for their activities. Unpolar
dialdehydes are more antibiotic and pungent than their corresponding oxidised
congeners, although the opposite relationship has been observed in mutagenicity
assays. Their polarity probably affect their ability to penetrate cell membranes, and
their interactions to active sites of receptors.
Even if the dialdehydes have been proposed to possess cytotoxic and antifungal
activity by reacting with proteins in the cell membrane [116, 134, 158, 243], there
are bioactivities which may be caused by other molecular mechanisms. For
instance, we have observed that the enantiomers of isovelleral (55) and its isomer
108 possess the same antibiotic activity, but differ significantly in pungency,
mutagenic activity and affinity for the dopamine D1 receptor. The mutagenic

effect of (-)-108 can be cancelled by stereo inversion, while the antibiotic effects
are constant. As there is a resemblance in structural properties between inactive and
active mutagens (as shown in Figure 36), it would be possible to evaluate the
structural demands for the mutagenicity for these dialdehydes. Hence, it would be
possible to optimise a desirable (or undesirable) biological effect, for instance
   The synthetic compound 189 was less antibiotic than isovelleral (55), while 190
possessed considerably higher antibiotic activities (see Paper IV). Apparently, the
presence of the cyclopentane ring in isovelleral (55) enhance its activities, either by
increasing the lipophilicity, or by affecting the conformation (and reactivity) of the
molecule. The methyl group adjacent to the cyclopropane ring influence the
bioactivity, since 190 is more bioactive than 189. It would therefore be interesting
to synthesise and assay compound 191, since contains a cyclopentane ring and
lacks the methyl group, it would be expected to possess higher activities than 55.



                                     Figure 37

   The unsaturated dialdehydes have been shown to interact specifically with some
receptors. Since there are a number of synthetic routes developed towards
unsaturated dialdehydes, preparations of labelled dialdehydes appear to be
feasible. They would therefore be suitable to use for investigation of receptors.
   During millions of years, Nature has developed a group of bioactive terpenoids
which are used by a number of different organisms to enhance their competitive
strength. These compounds are often produced in their most active isomeric form,
and possess several general and selective biological effects. Obviously, they are
optimised for each organism and its habitat. During the last two decades, scientists
have collected these terpenoids to one class of compounds - unsaturated 1,4-
dialdehydes - since they all look alike. This investigation show that there are
considerable reasons to divide them in subclasses, for instance drimanoids,
isovelleraloids and merulidialoids, since they possess different reactivity and
bioactivity. The findings presented in this thesis motivates further investigations of
these subclasses and their mechanisms of action.

                          Appendix A
Natural unsaturated dialdehydes, their CAS-numbers and sources.
#) CAS-number not available
Unsaturated dialdehyde                                          CAS-No    Source
β-Acaridial (84)                                            121362-23-0   Mite [48, 49, 303,
7-Acetoxy-2,6-cyclo-9,13-xenicadiene-18,19-dial (39)         81574-91-6   Alga [305]
9-Acetoxy-2,14-dichotomadiene-19,20-dial (38)                         #   Alga [305]
17-Acetoxy-4-hydroxy-1(9),6,13-xenicatriene-18,19-dial (35) 133585-91-8   Alga [306]
3α-Acetoxypolygodial (16)                                   160668-33-7   Plant [275]
3β-Acetoxypolygodial (17)                                   128718-16-1   Plant [11, 275]
Aframodial [see 8β,17-Epoxy-12E-labdene-15,16-dial (76)]
Ancistrodial (18)                                            68398-28-7   Termite [47]
Canellal [see Muzigadial (9)]
Capsicodendrin (24)                                          74749-35-2   Plant [144]
Chrysorrhedial (61)                                         147396-20-1   Fungus [307]
Cinnamodial / Ugandensidial (7)                              23599-45-3   Plant
1β-p-Coumaroyloxypolygodial (23)                            104006-82-8   Plant [309]
12-Deacetoxyscalaradial (46)                                154554-90-2   Sponge [73]
12-Deacetyl-12,18-di-epi-scalaradial (50)                    75266-25-0   Sponge [310]
12-Deacetyl-18-epi-12-oxoscalaradial (49)                   104900-66-5   Mollusc [173]
12-Deacetyl-12-epi-scalaradial (47)                         104900-65-4   Sponge [276, 277],
                                                                          Mollusc [173]
12-Deacetylscalaradial (43)                                  77282-60-1   Sponge [173, 311],
                                                                          Mollusc [173]
9-Deoxyisomuzigadial (13)                                   128718-15-0   Plant [11]
9-Deoxymuzigadial (11)                                      128718-14-9   Plant [11, 63]
Dictyodial A (33)                                            70552-61-3   Alga [126, 305, 312,
12,18-Di-epi-scalaradial (48)                                72300-73-3   Sponge [276, 277,
Durbinal C (67)                                             165337-80-4   Sponge [145]
Ent-isocopal-12-en-15,16-dial (73)                           84807-61-4   Sponge [139, 315]
Epi-deoxymuzigadial (12)                                    160668-37-1   Plant [275]
Epi-halimedatrial (41)                                      116836-91-0   Alga [25]
Epi-piperdial (60)                                                    #   Fungus [32]
Epi-polygodial [see Isotadeonal (15)]
12-Epi-scalaradial (45)                                      72300-72-2   Sponge [277, 314,
8α,12-Epoxy-12Z -labdene-15,16-dial (85)                     74513-41-0   Plant [317]
8β,17-Epoxy-12E-labdene-15,16-dial / Aframodial (76)         71641-23-1   Plant [96, 97, 99, 125]
4,13α-Epoxymuzigadial (10)                                  120166-31-6   Plant [318]
3-Formyl-(2,6,6-trimethyl-2-cyclohexenyl)-3-pentenal (19)   109326-06-9   Alga [51]
Halimedatrial (40)                                           87425-38-5   Alga [22, 23]
Hamiltonin C (72)                                           166774-52-3   Alga [319]
Hanegoketrial (54)                                           73616-89-4   Liverwort [320]
7-Hydroxy-2,6-cyclo-9,13-xenicadiene-18,19-dial (36)         81574-90-5   Alga [305]
9-Hydroxy-2,14-dichotomadiene-19,20-dial (37)                81574-88-1   Alga [305]
4β-Hydroxydictyodial A (34)                                  89482-11-1   Alga [140, 321]
1β-Hydroxyisosacculatal (65)                                171599-66-9   Liverwort [165]
3β-Hydroxyisosacculatal (66)                                 76475-27-9   Liverwort [322]

9-Hydroxymarasmic acid (53)                              109883-99-0   Fungus [154]
12-(3-Hydroxy-1-oxybutoxy)-scalaradial (44)               74054-35-6   [323]
1β-Hydroxysacculatal (64)                                171527-46-1   Liverwort [165]
18-Hydroxysacculatal (68)                                 73039-14-2   Liverwort [322]
19-Hydroxysacculatal (69)                                 73039-15-3   Liverwort [322]
Hyphodontal (57)                                         159736-40-0   Fungus [137]
Iso-ent-isocopal-12-en-15,16-dial (77)                    84807-62-5   Sponge [139, 315]
(Z)-Isolinaridial (81)                                    85120-60-1   Plant [324]
∆3 -(Z)-Isolinaridial (80)                               150134-18-2   Plant [325]
∆3 -(E)-Isolinaridial (79)                               150134-17-1   Plant [325]
Isopolygodial [see Isotadeonal (49)]
Isosacculatal (63)                                        64282-29-7   Liverwort [166, 170,
Isotadeonal / Isopolygodial / Epi-polygodial (15)          5956-39-8   Plant [127, 275, 279]
Isovelleral (55)                                          37841-91-1   Fungus [29, 36, 154,
Kuehneromycin A (31)                                     162666-36-6   Fungus [128]
Kuehneromycin B (21)                                     162810-05-1   Fungus [128]
(Z)-Labda-8(17),12-diene-15,16-dial (75)                  84413-88-7   Plant [327]
(-)-(E)-Labda-8(17),12-diene-15,16-dial (-)-(74)          77346-42-0   Plant [328]
(+)-(E)-Labda-8(17),12-diene-15,16-dial (+)-(74)         104263-85-6   Plant [100, 101, 147,
Linaridial (78)                                           55890-21-6   Plant [273]
Marasmal (32)                                            124869-10-9   Fungus [164]
Marasmic acid (52)                                         2212-99-9   Fungus [129, 130, 154,
                                                                       290, 334]
Merulidial (56)                                           68053-32-7   Fungus [193, 194]
Mniopetal A (25)                                         158760-98-6   Fungus [131, 335]
Mniopetal B (26)                                         158760-99-7   Fungus [131, 335]
Mniopetal C (27)                                         158761-00-3   Fungus [131, 335]
Mniopetal D (28)                                         158761-01-4   Fungus [131, 335]
Mniopetal E (29)                                         158761-02-5   Fungus [131, 335]
Mniopetal F (30)                                         158761-03-6   Fungus [131, 335]
Mukaadial (8)                                             87420-14-2   Plant [10, 81]
Muzigadial / Canellal (9)                                 66550-09-2   Plant [8, 9]
Oxytoxin 2 (83)                                          129932-68-9   Mollusc [43]
Panudial (20)                                            154648-88-1   Fungus [146]
Perrottetianal A (70)                                     73483-87-1   Liverwort [16-21, 336,
Perrottetianal B (71)                                     73483-86-0   Liverwort [16]
Pilatin (51)                                             119525-97-2   Fungus [338]
Piperdial (59)                                           100288-36-6   Fungus [31]
Polygodial / Tadeonal (14)                                 6754-20-7   Plant [12, 93, 127, 175,
                                                                       176, 200,279,339, 340]
                                                                       Liverwort [15, 169,
                                                                       341, 342]
                                                                       Mollusc [6, 40, 45]
RES-1149-1 (22)                                          160219-86-3   Fungus [132, 208]
Rhipocephenal (82)                                        71135-77-8   Alga [74]
Sacculatal (62)                                           64242-90-6   Liverwort [166, 170,
Scalaradial (42)                                          53527-28-9   Sponge [343]
                                                                       Mollusc [40]
Tadeonal [see Polygodial (14)]
Ugandensidial [see Cinnamodial (7)]
Velleral (58)                                             50656-61-6   Fungus [30-33, 190,
Warburganal (6)                                           62994-47-2   Plant [12, 197]

1.    Torsell, K.B.G. (1983) Natural Product Chemistry - A Mechanistic and Biosynthetic Approach
      to Secondary Metabolism. John Wiley & Sons Limited, Chichester.
2.    Munakata, K. (1975) Pure & Appl. Chem. 42, 57-66.
3.    Nakanishi, K. (1975) Rec. Adv. Phytochem. 9, 283
4.    Ma, W.-C. (1977) Physiol. Entomol. 2, 199-207.
5.    Cimino, G., De Rosa, S., De Stefano, S., Sodano, G., Villani, G. (1983) Science 219 (4589),
6.    Cimino, G., De Rosa, S., De Stefano, S., Morrone, R., Sodano, G. (1985) Tetrahedron 41,
7.    Powell, G., Hardie, J., Pickett, J.A. (1995) Physiol. Entomol. 20, 141-6.
8.    El-Feraly, F.S., McPhail, A.T., Onan, K.D. (1978) J. Chem. Soc., Chem. Commun. , 75-6.
9.    Kubo, I., Miura, I., Pettei, M.J., Lee, Y.W., Pilkiewicz, F., Nakanishi, K. (1977) Tetrahedron
      Lett. , 4553-6.
10.   Kioy, D., Gray, A.I., Waterman, P.G. (1989) J. Nat. Prod. 52, 174-7.
11.   Al-Said, M.S., El-Khawaja, S.M., El-Feraly, F.S., Hufford, C.D. (1990) Phytochemistry 29, 975
12.   Kubo, I., Lee, Y.-W., Pettei, M., Pilkiewicz, F., Nakanishi, K. (1976) J. Chem. Soc., Chem.
      Commun. , 1013-14.
13.   Banthorpe, D.V., Brooks, C.J.W., Brown, J.T., Lappin, G.J., Morris, G.S. (1989)
      Phytochemistry 28, 1631-3.
14.   Yan, F., Schoonhoven, L.M. (1993) Kunchong Xuebao 36, 1-7.
15.   Asakawa, Y., Aratani, T. (1976) Bull. Soc. Chim. Fr. , 1469-70.
16.   Asakawa, Y., Toyota, M., Takemoto, T. (1979) Phytochemistry 18, 1681-5.
17.   Asakawa, Y., Yamamura, A., Waki, T., Takemoto, T. (1980) Phytochemistry 19, 603-7.
18.   Asakawa, Y., Campbell, E.O. (1982) Phytochemistry 21, 2663-7.
19.   Toyota, M., Nagashima, F., Asakawa, Y. (1989) Phytochemistry 28, 1661-5.
20.   Toyota, M., Nagashima, F., Asakawa, Y. (1989) Phytochemistry 28, 3383-7.
21.   Toyota, M., Nagashima, F., Shima, K., Asakawa, Y. (1992) Phytochemistry 31, 183-9.
22.   Paul, V.J., Fenical, W. (1983) Science 221(4612), 747-9.
23.   Nagaoka, H., Miyaoka, H., Yamada, Y. (1990) Tetrahedron Lett. 31, 1573-6.
24.   Paul, V.J., Van Alstyne, K.L. (1992) J. Exp. Mar. Biol. Ecol. 160, 191-203.
25.   Paul, V.J., Van Alstyne, K.L. (1988) Coral Reefs 6, 263-9.
26.   Sterner, O., Bergman, R., Kesler, E., Nilsson, L., Oluwadiya, J., Wickberg, B. (1983)
      Tetrahedron Lett. 24, 1415-18.
27.   Sterner, O., Bergman, R., Kihlberg, J., Wickberg, B. (1985) J. Nat. Prod. 48, 279-88.
28.   Camazine, S., Lupo, A.T., Jr. (1984) Mycologia 76, 355-8.
29.   Magnusson, G., Thorén, S., Wickberg, B. (1972) Tetrahedron Lett. , 1105-8.
30.   Magnusson, G., Thorén, S., Drakenberg, T. (1973) Tetrahedron 29, 1621-4.
31.   Sterner, O., Bergman, R., Franzen, C., Wickberg, B. (1985) Tetrahedron Lett. 26, 3163-6.
32.   Sterner, O. (1989) Acta Chem. Scand. 43, 694-7.
33.   Hansson, T., Sterner, O., Strid, A. (1995) Phytochemistry 39, 363-5.
34.   Hansson, T., Sterner, O. (1991) Tetrahedron Lett. 32, 2541-4.
35.   Hansson, T., Pang, Z., Sterner, O. (1993) Acta Chem. Scand. 47, 403-5.
36.   Camazine, S.M., Resch, J.F., Eisner, T., Meinwald, J. (1983) J. Chem. Ecol. 9, 1439-47.
37.   Schulte, G.R., Scheuer, P.J. (1982) Tetrahedron 38, 1857-63.
38.   Faulkner, D.J., Molinski, T.F., Andersen, R.J., Dumdei, E.J., Dilip de Silva, E. (1990) Comp.
      Biochem. Physiol., C: Comp. Pharmacol. Toxicol. 97C, 233-40.
39.   Rogers, S.D., Paul, V.J. (1991) Mar. Ecol.: Prog. Ser. 77, 221-32.
40.   Cimino, G., De Rosa, S., De Stefano, S., Sodano, G. (1982) Comp. Biochem. Physiol. B 73B,
41.   Cimino, G., De Rosa, S., De Stefano, S., Sodano, G. (1986) Pure & Appl. Chem. 58, 375-86.
42.   Cimino, G., Sodano, G., Spinella, A. (1988) J. Nat. Prod. 51, 1010-11.
43.   Cimino, G., Crispino, A., Di Marzo, V., Gavagnin, M., Ros, J.D. (1990) Experientia 46, 767-
44.   Cimino, G., Fontana, A., Gimenez, F., Marin, A., Mollo, E., Trivellone, E., Zubia, E. (1993)
      Experientia 49, 582-6.
45.   Okuda, R.K., Scheuer, P.J., Hochlowski, J.E., Walker, R.P., Faulkner, D.J. (1983) J. Org.
      Chem. 48, 1866-9.

46.   Avila, C., Cimino, G., Crispino, A., Spinella, A. (1991) Experientia 47, 306-10.
47.   Baker, R., Briner, P.H., Evans, D.A. (1978) J. Chem. Soc., Chem. Commun. , 410-11.
48.   Leal, W.S., Kuwahara, Y., Suzuki, T. (1989) Agric. Biol. Chem. 53, 875-8.
49.   Leal, W.S., Kuwahara, Y., Suzuki, T., Kurosa, K. (1989) Naturwissenschaften 76, 332-3.
50.   Kuwahara, Y., Leal, W.S., Suzuki, T., Maeda, M., Masutani, T. (1989) Naturwissenschaften 76,
51.   Paul, V.J., Littler, M.M., Littler, D.S., Fenical, W. (1987) J. Chem. Ecol. 13, 1171-85.
52.   Hagendoorn, M.J.M., Geelen, T.A.M., van Beck, T.A., Jamar, D.C.L., Tetteroo, F.A.A., van der
      Plas, L.H.W. (1994) Physiol. Plant. 92, 595-600.
53.   Taniguchi, M., Yano, Y., Motoba, K., Tanaka, T., Oi, S., Haraguchi, H., Hashimoto, K., Kubo,
      I. (1988) Agric. Biol. Chem. 52, 1881-3.
54.   Paul, V.J., Fenical, W. (1982) Tetrahedron Lett. 23, 5017-20.
55.   Ireland, C., Faulkner, D.J. (1978) Bioorg. Chem. 7, 125-31.
56.   Abramson, S.N., Radic, Z., Manker, D., Faulkner, D.J., Taylor, P. (1989) Mol. Pharmacol. 36,
57.   San Feliciano, A., Barrero, A.F., del Corral, J.M.M., Gordaliza, M., Medarde, M. (1985)
      Tetrahedron 41, 671-80.
58.   Aasen, A.N., Nishida, T., Enzell, C.R., Appel, H.H. (1977) Acta Chemica Scandinavica B 31,
59.   Block, E. (1982) Scientific American , 94-99.
60.   Nakanishi, K., Kubo, I. (1977) Isr. J. Chem. 16, 28-31.
61.   Nakanishi, K. (1976) Pontif. Acad. Sci. Scr. Varia 41, 185-210.
62.   Kubo, I., Ganjian, I. (1981) Experientia 37, 1063-4.
63.   Gerard, P.J., Perry, N.B., Ruf, L.D., Foster, L.M. (1993) Bull. Entomol. Res. 83, 547-52.
64.   Blaney, W.M., Simmonds, M.S.J., Ley, S.V., Katz, R.B. (1987) Physiol. Entomol. 12, 281-91.
65.   Asakawa, Y., Dawson, G.W., Griffiths, D.C., Lallemand, J.Y., Ley, S.V., Mori, K., Mudd, A.,
      Pezechk-Leclaire, M., Pickett, J.A., et al. (1988) J. Chem. Ecol. 14, 1845-55.
66.   Schoonhoven, L.M., Yan, F.S. (1989) J. Insect Physiol. 35, 725-8.
67.   Caprioli, V., Cimino, G., Colle, R., Gavagnin, M., Sodano, G., Spinella, A. (1987) J. Nat. Prod.
      50, 146-51.
68.   Gibson, R.W., Rice, A.D., Pickett, J.A., Smith, M.C., Sawicki, R.M. (1982) Ann. Appl. Biol.
      100, 55-9.
69.   Dawson, G.W., Griffiths, D.C., Hassanali, A., Pickett, J.A., Plumb, R.T., Pye, B.J., Smart, L.E.,
      Woodcock, C.M. (1986) Proc. - Br. Crop Prot. Conf.--Pests Dis. , 1001-8.
70.   Pickett, J.A., Dawson, G.W., Griffiths, D.C., Hassanali, A., Merritt, L.A., Mudd, A., Smith,
      M.C., Wadhams, L.J., Woodcock, C.M., Zhang, Z. (1987) Development of plant-derived
      antifeedants for crop protection. In Pestic. Sci. Biotechnol. (R. R. Greenhalgh, Terence Robert,
      ed) Blackwell Scientific, Oxford, UK 125-8.
71.   Griffiths, D.C., Pickett, J.A., Smart, L.E., Woodcock, C.M. (1989) Pestic. Sci. 27, 269-76.
72.   Scott, G.C. (1981) Proceedings 1981 British Crop Protection Conferance - Pests and Deseases ,
73.   De Rosa, S., Puliti, R., Crispino, A., De Giulio, A., Mattia, C.A., Mazzarella, L. (1994) J. Nat.
      Prod. 57, 256-62.
74.   Sun, H.H., Fenical, W. (1979) Tetrahedron Lett. , 685-8.
75.   Hashimoto, T., Tanaka, H., Asakawa, Y. (1994) Chem. Pharm. Bull. 42, 1542-4.
76.   Sterner, O., pers. commun. .
77.   Jurgens, T.M., Hufford, C.D., Clark, A.M. (1992) Xenobiotica 22, 569-77.
78.   Anke, H., Sterner, O. (1988) Phytochemistry 27, 2765-7.
79.   Taniguchi, M., Chapya, A., Kubo, I., Nakanishi, K. (1978) Chem. Pharm. Bull. 26, 2910-13.
80.   Taniguchi, M., Kubo, I. (1993) J. Nat. Prod. 56, 1539-46.
81.   Kubo, I., Matsumoto, T., Kakooko, A.B., Mubiru, N.K. (1983) Chem. Lett. , 979-80.
82.   Canonica, L., Corbella, A., Gariboldi, P., Jommi, G., Krepinsky, J., Ferrati, G., Casagrande, C.
      (1969) Tetrahedron 25, 3895-902.
83.   Taniguchi, M., Adachi, T., Oi, S., Kimura, A., Katsumura, S., Isoe, S., Kubo, I. (1984) Agric.
      Biol. Chem. 48, 73-8.
84.   Kokwaro, J.O. (1976) Medicinal Plants of East Africa. East African Literature Bureau,
85.   Watt, J.M., Breyer-Brandwijk, M.G. (1962) Medicinal and Poisonous Plants of Southern and
      Eastern Africa. E. S. Livingstone Ltd., Edinburgh and London.
86.   Rajab, M.S., Olowookere, J.O. (1993) Bull. Chem. Soc. Ethiop. 7, 17-21.
87.   Fukuyama, Y., Sato, T., Asakawa, Y., Takemoto, T. (1982) Phytochemistry 21, 2895-8.
88.   Matsumoto, T., Tokuda, H. (1990) Basic Life Sci. 52, 423-7.
89.   Hartwell, J.L. (1970) Lloydia 33, 288-89.

90.    Hartwell, J.L. (1982) Plant used against cancer. Quarterman Publications, Inc., Massachusetts.
91.    Cribb, J.W., Cribb, A.B. (1981) Useful Wild Plants in Australia. Collins, Sydney.
92.    Akamatsu, K. (1970) Wakanyaku. Ishiyakushuppan, Tokyo.
93.    Loder, J.W. (1962) Austr. J. Chem. 15, 389-90.
94.    Maiden, J.H. (1889) Useful Native Plants of Australia. Turner & Henderson, Sydney.
95.    Kitagawa, I., Yoshihara, M., Tani, T., Yosioka, I. (1975) Tetrahedron Lett. , 23-6.
96.    Kimbu, S.F., Njimi, T.K., Sondengam, B.L., Akinniyi, J.A., Connolly, J.D. (1979) J. Chem. Soc.,
       Perkin Trans. 1 , 1303-4.
97.    Ayafor, J.F., Tchuendem, M.H.K., Nyasse, B., Tillequin, F., Anke, H. (1994) Pure Appl. Chem.
       66, 2327-30.
98.    Burkill, I.H. (1966) , Volume 2Ministry of Agiculture and Co-operatives, Kuala-Lumpur,
       Malaysia 1323.
99.    Kano, Y., Tanabe, M., Yasuda, M. (1990) Shoyakugaku Zasshi 44, 55-7.
100.   Morita, H., Itokawa, H. (1986) Chem. Lett. , 1205-8.
101.   Sirat, H.M., Masri, D., Rahman, A.A. (1994) Phytochemistry 36, 699-701.
102.   Kano, Y., Miura, O., Tatsumi, Y. In Chem. Abstr.Kanebo, Ltd., Japan, Jpn. Kokai Tokkyo
       Koho 118:66855.
103.   Matsumoto, T., Noguchi, M., Matsuzaki, T., Ninomya, M., Kawagishi, S. In Chem.
       Abstr.Nippon Tobacco Sangyo, Japan, Jpn. Kokai Tokkyo Koho 119:86036.
104.   Matsumoto, A., Kageyama, S. In Chem. Abstr.Daicel Chemical Industries, Ltd., Japan, Jpn.
       Kokai Tokkyo Koho 109:176376.
105.   Tokuda, H., Matsumoto, T. In Chem. Abstr.Daicel Chemical Industries, Ltd., Japan, Jpn.
       Kokai Tokkyo Koho 111:167386.
106.   van Beek, T.A., de Groot, A. (1986) Recl. Trav. Chim. Pays-Bas 105, 513-27.
107.   Singhal, S., Mathur, S.C. (1993) Chemistry and Industry , 112-14.
108.   Zhang, Z., Liu, X., Lou, Z., Li, H., Zhu, S., Zou, F. (1993) Kunchong Xuebao 36, 172-6.
109.   Dawson, G.W., Hallahan, D.L., Mudd, A., Patel, M.M., Pickett, J.A., Wadhams, L.J.,
       Wallsgrove, R.M. (1989) Pestic. Sci. 27, 191-201.
110.   Shono, A., Sakaguchi, T., Adachi, H. In Chem. Abstr.Sanyoo Fuain Kk, Japan, Jpn. Kokai
       Tokkyo Koho 123:79473.
111.   Anke, H., Sterner, O. (1991) Planta Med. 57, 344-6.
112.   Kubo, I., Himejima, M. (1991) J. Agric. Food Chem. 39, 2290-2.
113.   Kang, R., Helms, R., Stout, M.J., Jaber, H., Chen, Z., Nakatsu, T. (1992) J. Agric. Food Chem.
       40, 2328-30.
114.   Himejima, M., Kubo, I. (1992) J. Nat. Prod. 55, 620-5.
115.   Himejima, M., Kubo, I. (1993) J. Agric. Food Chem. 41, 1776-9.
116.   Kubo, I., Taniguchi, M. (1988) J. Nat. Prod. 51, 22-9.
117.   Yano, Y., Taniguchi, M., Tada, E., Tanaka, T., Oi, S., Haraguchi, H., Hashimoto, K., Kubo, I.
       (1989) Agric. Biol. Chem. 53, 1525-30.
118.   Brooks, C.J.W., Watson, D.G., Cole, W.J. (1985) J. Chromatogr. 347, 455-7.
119.   Brooks, C.J.W., Brindle, P.A., Cole, W.J., Watson, D.G. (1988) J. Chromatogr. 438, 108-10.
120.   Brooks, C.J.W., Cole, W.J., Brindle, P.A. (1989) J. Chromatogr 468, 201-13.
121.   Ishikawa, H., Suzuki, Y., Sakai, A., Ishizuka, S. In Chem. Abstr.Lotte Co Ltd, Japan; Toyotama
       Perfumery, Jpn. Kokai Tokkyo Koho 123:168244.
122.   Anke, H., Sterner, O., Steglich, W. (1989) J. Antibiot. 42, 738-44.
123.   Stampf, J.L., Benezra, C., Asakawa, Y. (1982) Arch. Dermatol. Res. 274, 277-81.
124.   Benezra, C., Stampf, J.-L., Barbier, P., Ducombs, G. (1985) Contact Dermatitis 13, 110-114.
125.   Ayafor, J.F., Tchuendem, M.H.K., Nyasse, B., Tillequin, F., Anke, H. (1994) J. Nat. Prod. 57,
126.   Finer, J., Clardy, J., Fenical, W., Minale, L., Riccio, R., Battaile, J., Kirkup, M., Moore, R.E.
       (1979) J. Org. Chem. 44, 2044-7.
127.   McCallion, R.F., Cole, A.L.J., Walker, J.R.L., Blunt, J.W., Munro, M.H.G. (1982) Planta
       Med. 44, 134-8.
128.   Erkel, G., Lorenzen, K., Anke, T., Velten, R., Gimenez, A., Steglich, W. (1995) Z. Naturforsch.
       50c, 1-10.
129.   Kupka, J., Anke, T., Mizumoto, K., Giannetti, B.M., Steglich, W. (1983) J. Antibiot. 36, 155-
130.   Kavanagh, F., Hervey, A., Robbins, W.J. (1949) Proc. Nat. Acad. Sci. 35, 343-49.
131.   Kuschel, A., Anke, T., Velten, R., Klostermeyer, D., Steglich, W., Köning, B. (1994) J.
       Antibiotics 47, 733-740.
132.   Ogawa, T., Ando, K., Tanaka, T., Uosaki, Y., Matsuda, Y. (1996) J. Antibiotics 49, 1-5.
133.   Taniguchi, M., Adachi, T., Haraguchi, H., Oi, S., Kubo, I. (1983) J. Biochem. 94(1), 149-54.

134.   Taniguchi, M., Yano, Y., Tada, E., Ikenishi, K., Oi, S., Haraguchi, H., Hashimoto, K., Kubo, I.
       (1988) Agric. Biol. Chem. 52, 1409-14.
135.   Morita, H., Itokawa, H. (1988) Planta Med. 54, 117-20.
136.   Kim, T.H., Isoe, S. (1983) J.C.S., Chem Commun. , 730-31.
137.   Erkel, G., Anke, T. (1994) Z. Naturforsch. 49c, 561-570.
138.   Cimino, G., De Rosa, S., De Stefano, S., Sodano, G. (1985) Experientia 41, 1335-6.
139.   Cimino, G., Morrone, R., Sodano, G. (1982) Tetrahedron Lett. 23, 4139-42.
140.   Tanaka, J., Higa, T. (1984) Chem. Lett. , 231-2.
141.   D'Ischia, M., Prota, G., Sodano, G. (1982) Tetrahedron Lett. 23, 3295-8.
142.   Tanabe, M., Chen, Y.D., Saito, K., Kano, Y. (1993) Chem. Pharm. Bull. 41, 710-13.
143.   Forsby, A., Andersson, M., Lewan, L., Sterner, O. (1991) Toxicol. in Vitro 5, 9-14.
144.   Mahmoud, I.I., Kinghorn, A.D., Cordell, G.A., Farnsworth, N.R. (1980) J. Nat. Prod. 43, 365-
145.   Rudi, A., Kashman, Y., Benayahu, Y., Schleyer, M. (1995) Tetrahedron Lett. 36, 4853-56.
146.   Lorenzen, K., Anke, T., Anders, U., Hindermayr, H., Hansske, F. (1994) Z. Naturforsch., C:
       Biosci. 49, 132-8.
147.   Itokawa, H., Morita, H., Katou, I., Takeya, K., Cavalheiro, A.J., De Oliveira, R.C.B., Ishige,
       M., Motidome, M. (1988) Planta Med. 54, 311-15.
148.   Marshall, L.A., Winkler, J.D., Griswold, D.E., Bolognese, B., Roshak, A., Sung, C.M., Webb,
       E.F., Jacobs, R. (1994) J. Pharmacol. Exp. Ther. 268, 709-17.
149.   Marshall, L.A., Bolognese, B., Raymond, H. (1994) Pharmacol. Commun. 5, 27-38.
150.   Barnette, M.S., Rush, J., Marshall, L.A., Foley, J.J., Schmidt, D.B., Sarau, H.M. (1994)
       Biochem. Pharmacol. 47, 1661-1668.
151.   De Carvalho, M.S., Jacobs, R.S. (1991) Biochem. Pharmacol. 42, 1621-6.
152.   Potts, B.C.M., Faulkner, D.J., De Carvalho, M.S., Jacobs, R.S. (1992) J. Am. Chem. Soc. 114,
153.   Bergquist, J., Strandberg, C., Andersson, M., Sterner, O., Pesando, D., Girard, J.P. (1993)
       Toxicol. in Vitro 7, 205-12.
154.   Anke, H., Hillen-Maske, E., Steglich, W. (1989) Z. Naturforsch., C: Biosci. 44, 1-6.
155.   Andersson, M., Bocchio, F., Sterner, O., Forsby, A., Lewan, L. (1993) Toxicol. in Vitro 7, 1-6.
156.   Forsby, A., Walum, E., Sterner, O. (1992) Chem.-Biol. Interact. 84, 85-95.
157.   Forsby, A., Witt, R., Walum, E. (1994) Nat. Toxins 2, 89-95.
158.   Andres, M.-I., Forsby, A., Walum, E. (1996) Polygodial induces H-noradrenaline release in
       human neuroblastoma SH-SY5Y cells in INVITOX 96 workshop.
159.   Bastos, J.K., Gottlieb, O.R., Kaplan, M.A.C., dos Santos Filho, D., Sarti, S.J., Rodrigues, C.P.S.
       (1991) Rev. Cienc. Farm. 13, 83-9.
160.   Sterner, O., Carter, R.E., Nilsson, L.M. (1987) Mutat. Res. 188, 169-74.
161.   Nilsson, L.M., Carter, R.E., Sterner, O., Liljefors, T. (1988) Quant. Struct.-Act. Relat. 7, 84-91.
162.   Morales, P., Andersson, M., Lewan, L., Sterner, O. (1992) Mutat. Res. 268, 315-21.
163.   Jonassohn, M., Anke, H., Morales, P., Sterner, O. (1995) Acta Chem. Scand. 49, 530-5.
164.   Ayer, W.A., Craw, P.A. (1989) Can. J. Chem. 67, 1371-1380.
165.   Hashimoto, T., Okumura, Y., Suzuki, K., Takaoka, S., Kan, Y., Tori, M., Asakawa, Y. (1995)
       Chem. Pharm. Bull. 43, 2030-2032.
166.   Asakawa, Y., Harrison, L.J., Toyota, M. (1985) Phytochemistry 24, 261-2.
167.   Asakawa, Y., Takemoto, T. (1979) Experientia 35, 1420-1.
168.   Asakawa, Y., Huneck, S., Toyota, M., Takemoto, T., Suire, C. (1979) J. Hattori Bot. Lab 46,
169.   Asakawa, Y. (1984) Rev. Latinoam. Quim. 14, 109-14.
170.   Asakawa, Y., Takemoto, T., Toyota, M., Aratani, T. (1977) Tetrahedron Lett. 16, 1407-10.
171.   Szallasi, A., Jonassohn, M., Acs, G., Biro, T., Acs, P., Blumberg, P.M., Sterner, O. (1996) British
       Journal of Pharmacology, 119, 283-290.
172.   Cimino, G., Sodano, G., Spinella, A. (1987) Tetrahedron 43, 5401-10.
173.   Terem, B., Scheuer, P.J. (1986) Tetrahedron 42, 4409-12.
174.   Bocchio, F., Kalf-Hansen, S., Dekermendjian, K., Sterner, O., Witt, R. (1992)
         Tetrahedron Lett. 33, 6867-70.
175.   Barnes, C.S., Loder, J.W. (1962) Austral. J. Chem. 15, 322-327.
176.   Cortes, M.J., Oyarzun, M.L. (1981) Fitoterapia 52, 33-5.
177.   Henkin, R.I., Bradley, D.F. (1969) Proc. N. A. S. 62, 30.
178.   Cimino, G., Spinella, A., Sodano, G. (1984) Tetrahedron Lett. 25, 4151-2.
179.   Cafieri, F., De Napoli, L., Fattorusso, E., Santacroce, C. (1977) Experientia 33, 994-5.
180.   Cafieri, F., De Napoli, L., Fattorusso, E., Santacroce, C., Sica, D. (1977) Tetrahedron Lett. ,
181.   Cafieri, F., De Napoli, L., Iengo, A., Santacroce, C. (1978) Experientia 34, 300-301.

182.   Cafieri, F., De Napoli, L., Iengo, A., Santacroce, C. (1979) Experientia 35, 157-158.
183.   Iengo, A., Pecoraro, C., Santacroce, C., Sodano, G. (1979) Gazz. Chim. Ital. 109, 701-2.
184.   Lam, P.Y.S., Frazier, J.L. (1987) Tetrahedron Lett. 28, 5477-80.
185.   Schmolka, I.R., Spoerri, P.E. (1957) J. Org. Chem. 22, 943-46.
186.   Friedman, M. (1973) The Chemistry and Biochemistry of the Sulfhydryl Group in Amino Acids,
       Peptides and Proteins. Pergamon Press, Oxford.
187.   Fritz, G.L., Mills, G.D., Jr., Warthen, J.D., Jr., Waters, R.M. (1989) J. Chem. Ecol. 15, 2607-
188.   Gustafsson, J. (1994) In Organic Chemistry 2Lund University, Lund 109.
189.   Jansen, B.J.M., De Groot, A. (1991) Natural Product Reports , 319-337.
190.   List, P.H., Hackenberg, H. (1969) Arch. Pharm. Weinheim 302, 125-43.
191.   Resch, J.F. (1982) , 2214.
192.   Hansson, T., Sterner, O., Wickberg, B., Bergman, R. (1992) J. Org. Chem. 57, 3822-8.
193.   Quack, W., Anke, T., Oberwinkler, F., Giannetti, B.M., Steglich, W. (1978) J. Antibiot. 31,
194.   Giannetti, B.M., Steffan, B., Steglich, W., Quack, W., Anke, T. (1986) Tetrahedron 42, 3579-
195.   Tozyo, T., Yasuda, F., Nakai, H., Tada, H. (1992) J. Chem. Soc., Perkin Trans. 1 , 1859-66.
196.   Kioy, D., Gray, A.I., Waterman, P.G. (1990) Phytochemistry 29, 3535-8.
197.   Ferreto, L., Ciccio, J.F., Castro, V., Andrade, R. (1988) Spectroscopy 6(3-4), 133-6.
198.   Ley, S.V., Mahon, M. (1981) Tetrahedron Lett. 22, 3909-3912.
199.   Hollinshead, D.M., Howell, S.C., Ley, S.V., Mahon, M., Ratcliffe, N.M., Worthington, P.A.
       (1983) J. Chem. Soc., Perkin Trans. 1 , 1579-89.
200.   Ohsuka, A. (1962) Nippon Kagaku Zasshi 83, 757-60.
201.   Peña, W., López, J.T., Cortés, M. (1989) Synth. Commun. 19, 2841-50.
202.   Appel, H.H., Connolly, J.D., Overton, K.H., Bond, R.P.M. (1960) J. Chem. Soc. , 4685-4692.
203.   Urones, J.G., Marcos, I.S., Perez, B.G., Diez, D., Lithgow, A.M., Gomez, P.M., Basabe, P.,
       Garrido, N.M. (1994) Tetrahedron 50, 10995-1012.
204.   Tanis, S.P., Nakanishi, K. (1979) J. Am. Chem. Soc. 101, 4398-4400.
205.   Okawara, H., Nakai, H., Ohno, M. (1982) Tetrahedron Lett. 23, 1087-90.
206.   Jansen, B.J.M., Sengers, H.H.W.J.M., De Groot, A., Bos, H.J.T. (1988) J. Org. Chem. 53, 855-
207.   Garlaschelli, L., Mellerio, G., Vidari, G. (1989) Tetrahedron 45, 7379-86.
208.   Uosaki, Y., Yoshida, M., Ogawa, T., Saitoh, Y. (1996) J. Antibiotics 49, 6-12.
209.   Mischne, M.P., Sierra, M.G., Ruveda, E.A. (1984) J. Org. Chem. 49, 2035-7.
210.   Kende, A.S., Blacklock, T.J. (1980) Tetrahedron Lett. 21, 3119-22.
211.   Bosch, M.P., Camps, F., Coll, J., Guerrero, A., Tatsuoka, T., Meinwald, J. (1986) J. Org. Chem.
       51, 773-84.
212.   Martin, S.F. (1979) Synthesis , 633-665.
213.   Gröbel, B.-T., Seebach, D. (1977) Synthesis , 357-402.
214.   Lever, O.W.J. (1976) Tetrahedron 32, 1943-71.
215.   De Groot, A., Van Beek, T.A. (1987) Recl. Trav. Chim. Pays-Bas 106, 1-18.
216.   Peterse, A.J.G.M., Roskam, J.H., De Groot, A. (1978) Recl. Trav. Chim. Pays-Bas 97, 277-8.
217.   Goldsmith, D.J., Kezar, H.S., III (1980) Tetrahedron Lett. 21, 3543-6.
218.   Corey, E.J., Tius, M.A., Das, J. (1980) Journal of the American Chemical Society 102, 1742-
219.   Still, W.C. (1978) Journal of the American Chemical Society 100, 1481-1487.
220.   Imamoto, T., Kusumoto, T., Tawarayama, Y., Sugiura, Y., Mita, T., Hatanaka, Y., Yokohama,
       M. (1984) J. Org. Chem. 49, 3902-3912.
221.   Imamoto, T., Sugiura, Y., Takiyama, N. (1984) Tetrahedron Lett. 25, 4233-4236.
222.   Corey, E.J., Bock, M.G. (1975) Tetrahedron Lett. , 2643.
223.   Lo, T.-L., Ho, H.C., Wong, C.M. (1972) J. Chem. Soc., Chem. Commun. , 791.
224.   Greene, T.W., Wuts, P.G.M. (1991) Protective Groups in Organic Chemistry. John Wiley &
       Sons, Inc., New York.
225.   Paulsen, H., Stubbe, M., Heiker, F.R. (1980) Liebigs Ann. Chem. 1980, 825-837.
226.   Paulsen, H., Mielke, B., von Deyn, W. (1987) Liebigs Ann. Chem. 1987, 439-445.
227.   Bergman, R., Hansson, T., Sterner, O., Wickberg, B. (1990) J. Chem. Soc., Chem. Commun. ,
228.   Thompson, S.K., Heathcock, C.H. (1990) J. Org. Chem. 55, 3004-5.
229.   Thompson, S.K., Heathcock, C.H. (1992) J. Org. Chem. 57, 5979-89.
230.   Hostettmann, K., Hostettmann-Kaldas, M., Sticher, O. (1980) Journal of Chromatography 202,
231.   Stahl, E., Müller, J. (1982) Chromatographia 15, 493-497.

232.   Hansson, T., Bergman, R., Sterner, O., Wickberg, B. (1990) J. Chem. Soc., Chem. Commun. ,
233.   Froborg, J., Magnusson, G. (1978) Tetrahedron 34, 2027-8.
234.   Trost, B.M., Hipskind, P.A. (1992) Tetrahedron Lett. 33, 4541-4.
235.   Corey, E.J., Venkateswarlu, A. (1972) J. Am. Chem. Soc. 94, 6190-6191.
236.   Olah, G.A., Gupta, B.G.B., Narang, S.C., Malhotra, R. (1979) Journal of Organic Chemistry 44,
237.   Witz, G. (1989) Free Radical Biology & Medicine 7, 333-349.
238.   Chung, F.-L., Hecht, S.S. (1983) Cancer Research 43, 1230-1235.
239.   Chung, F.-L., Young, R., Hecht, S.S. (1984) Cancer Research 44, 990-995.
240.   Chung, F.-L., Roy, K.R., Hecht, S.S. (1988) J. Org. Chem. 53, 14-17.
241.   Eder, E., Hoffman, C. (1992) Chem. Res. Toxicol. 5, 802-808.
242.   Eder, E., Hoffman, C., Deininger, C., Scheckenbach, S. (1994) Toxic. in Vitro 8, 707-710.
243.   Forsby, A., Andres, M.-I., Walum, E. (1993) In Human cells in in vitro pharmaco-toxicology.
       Present status within Europe. (V. Rogiers, W. Sonck, E. Shephard and A. Vercruysse, eds),
       VUBPRESS, Brussels 169-184.
244.   Ayer, W.A., Cruz, E.R. (1993) J. Org. Chem. 58, 7529-7534.
245.   Gingrich, H.L., Roush, D.M., Van Saun, W.A. (1983) J. Org. Chem. 48, 4869-73.
246.   Seyferth, D., Blank, D.R., Evnin, A.B. (1967) J. Amer. Chem. Soc. 89, 4793-95.
247.   Bentley, R., Zwitkowits, P.M. (1967) Journal of the American Chemical Society 89, 676-680.
248.   Bentley, R., Zwitkowits, P.M. (1967) Journal of the American Chemical Society 89, 681-685.
249.   Light, R.J., Harris, T.M., Harris, C.M. (1966) Federation Proc., Abstr. 25, 768.
250.   Harris, T.M., Harris, C.M., Light, R.J. (1966) Biochim. Biophys. Acta 121, 420-423.
251.   Kotani, T., Nonomura, S., Tatsumi, C. (1964) Nippon Nogeikagaku Kaishi 38, 585.
252.   Jonassohn, M., Anke, H., Sterner, O., Svensson, C. (1994) Tetrahedron Lett. 35, 1593-1596.
253.   Moreno-Mañas, M., Pleixats, R. (1992) Adv. Heterocycl. Chem. 53, 1-84.
254.   Collie, J.N. (1891) J. Chem. Soc. 59, 607-617.
255.   De March, P., Moreno-Mañas, M., Casado, J., Pleixats, R., Roca, J.L., Trius, A. (1984) J.
       Heterocyclic Chem. 21, 85-89.
256.   Ikawa, M., Stahmann, M.A., Link, K.P. (1944) Journal of the American Chemical Society 66,
257.   Appendino, G., Cravatto, G., Nano, G.M., Palmisano, G., Annunziata, R. (1993) Helvetica
       Chimica Acta 76, 1194-1202.
258.   Hutchinson, D.W., Tomlinson, J.A. (1968) Tetrahedron Lett. 9, 5027-5028.
259.   Galliani, G., Pantarotto, C. (1983) Tetrahedron Lett. 24, 4491-92.
260.   Hadley, M., Draper, H.H. (1990) Lipids 25, 82-5.
261.   Stone, K., Ksebati, M.B., Marnett, L.J. (1990) Chem. Res. Toxicol. 3, 33-38.
262.   Stone, K., Uzieblo, A., Marnett, L.J. (1990) Chem. Res. Toxicol. 3, 467-72.
263.   Lutz, D., Eder, E., Neudecker, T., Henschler, D. (1982) Mutation Research 93, 305-315.
264.   Ying, B.-P., Peiser, G.D., Ji, Y.-Y., Mathias, K.M., Karasina, F., Hwang, Y.-S. (1995) J. Agric.
       Food Chem. 43, 826-9.
265.   Roggo, B.E., Petersen, F., Sills, M., Roesel, J.L., Moerker, T., Peter, H.H. (1996) J. Antibiot.
       49, 13-19.
266.   Roggo, B.E., Hug, P., Moss, S., Stämpfli, A., Kriemler, H.-P., Peter, H.H. (1996) J. Antibiot.
       49, 374-79.
267.   McMorris, T.C., Kelner, M.J., Wang, W., Moon, S., Taetle, R. (1990) Chem. Res. Toxicol. 3,
268.   Tillian, H.M., Schauenstein, E., Ertl, A., Esterbauer, H. (1976) Europ. J. Cancer 12, 989-93.
269.   Tillian, H.M., Schauenstein, E., Esterbauer, H. (1978) Europ. J. Cancer 14, 533-36.
270.   Ishikawa, T., Esterbauer, H., Sies, H. (1986) J. Biol. Chem. 261, 1576-81.
271.   Åhlin, P., Danielson, U.H., Mannervik, B. (1985) FEBS Letters 179, 267-70.
272.   Sterner, O., Andersson, M., Forsby, A., Morales, P. (1991) ATLA 19, 171-7.
273.   Kitagawa, I., Yoshihara, M., Tani, T., Yosioka, I. (1976) Chem. Pharm. Bull. 24, 294-302.
274.   Fukuyama, Y., Sato, T., Miura, I., Asakawa, Y. (1985) Phytochemistry 24, 1521-4.
275.   Ying, B.-P., Peiser, G., Ji, Y.-Y., Mathias, K., Tutko, D., Hwang, Y.-S. (1995) Phytochemistry
       38, 909-15.
276.   Crews, P., Bescansa, P. (1986) J. Nat. Prod. 49, 1041-52.
277.   Bergquist, P.R., Cambie, R.C., Kernan, M.R. (1990) Biochem. Syst. Ecol. 18, 349-57.
278.   Enoki, N., Ishida, R., Matsumoto, T. (1982) Chem. Lett. , 1749-52.
279.   Ohsuka, A. (1963) Nippon Kagaku Zasshi 84, 748-752.
280.   van Beek, T.A., van Dam, N., de Groot, A., Geelen, T.A.M., van der Plas, L.H.W. (1994)
       Phytochem. Anal. 5, 19-23.
281.   Sterner, O. (1985) Thesis, In Organic Chemistry 2Lund University, Lund 123.

282.   Sterner, O., Steglich, W. (1988) Liebigs Ann. Chem. , 823-4.
283.   Nakatani, N., Kikuzaki, H., Yamaji, H., Yoshio, K., Kitora, C., Okada, K., Padolina, W.G.
       (1994) Phytochemistry 37, 1383-8.
284.   Daniewski, W., Kroszczynski, W., Skibicki, P., De Bernardi, M., Fronza, G., Vidari, G., Vita-
       Finzi, P. (1988) Phytochemistry 27, 187.
285.   Gersmann, H.R., Nieuwenhuis, H.J.W., Bickel, A.F. (1963) Tetrahedron Lett. , 1383.
286.   Collins, C.J., Eastham, J.F. (1966) Chapter 15. Rearrangements Involving the Carbonyl
       Group. In The Chemistry of the Carbonyl Group (S. Patai, ed) Interscience Publishers, John
       Wiley & Sons, London, New York, Sydney 761-821.
287.   Fieser, L.F., Fieser, M. (1959) Steroids. Reinhold Publishing Corp., New York.
288.   Ruzicka, L., Meldahl, H.F. (1938) Helv. Chem. Acta 21, 1760-70.
289.   Ruzicka, L., Meldahl, H.F. (1939) Helv. Chem. Acta 22, 421-24.
290.   Dugan, J.J., de Mayo, P., Nisbet, M., Robinson, J.R., Anchel, M. (1966) J. Am. Chem. Soc. 88,
291.   Canonica, L., Corbella, A., Jommi, G., Krepinsky, J., Ferrari, G., Casagrande, C. (1967)
       Tetrahedron Lett. , 2137-41.
292.   Corey, E.J., Posner, G.H., Atkinson, R.F., Wingard, A.K., Halloran, D.J., Radzik, D.M., Nash,
       J.J. (1989) J. Org. Chem. 54, 389-93.
293.   Borssén, J. (1996) In Vin & Sprit-Journalen 21-26.
294.   Szallasi, A., Blumberg, P.M. (1990) Brains Res. 524, 106-111.
295.   Bevan, S., Hothi, S., Hughes, G., James, I.F., Rang, H.P., Shah, K., Walpole, C.S.J., Yeats, J.C.
       (1992) Br. J. Pharmacol. 107, 544-552.
296.   Szallasi, A. (1994) Gen. Pharmacol. 25, 223-43.
297.   Evangelista, S., Renzi, D., Guzzi, P., Surrenti, C., Santicioli, P., Maggi, C.A. (1992) Gen.
       Pharmacol. 23, 39-41.
298.   Karrer, T., Bartoshuk, L. (1991) Physiology & Behavior 49, 757-764.
299.   Zapf, S., Hossfeld, M., Anke, H., Velten, R., Steglich, W. (1995) J. Antibiotics 48, 36-41.
300.   Anke, H., Bergendorff, O., Sterner, O. (1989) Food Chem. Toxicol. 27, 393-98.
301.   Gustafsson, J., Sterner, O. (1995) Tetrahedron 51, 3865-72.
302.   Gustafsson, J., Sandström, J., Sterner, O. (1995) Tetrahedron Asymmetry 6, 595-602.
303.   Leal, W.S., Kuwahara, Y., Suzuki, T., Nakao, H. (1989) Agric. Biol. Chem. 53, 3279-84.
304.   Suzuki, T., Haga, K., Leal, W.S., Kodama, S., Kuwahara, Y. (1989) Appl. Entomol. Zool. 24, 222
305.   Blount, J.F., Dunlop, R.W., Erickson, K.L., Wells, R.J. (1982) Aust. J. Chem. 35, 145-63.
306.   König, G.M., Wright, A.D., Sticher, O. (1991) Tetrahedron 47, 1399-1410.
307.   De Bernardi, M., Garlaschelli, L., Toma, L., Vidari, G., Vita-Finzi, P. (1993) Tetrahedron 49,
308.   Brooks, C.J.W., Draffan, G.H. (1969) Tetrahedron 25, 2887-98.
309.   Vichnewski, W., Kulanthaivel, P., Herz, W. (1986) Phytochemistry 25, 1476-78.
310.   Walker, R.P., Thompson, J.E., Faulkner, D.J. (1980) J. Org. Chem. 45, 4976-4979.
311.   Yasuda, F., Tada, H. (1981) Experientia 37, 110-11.
312.   Nagaoka, H., Kobayashi, K., Yamada, Y. (1988) Tetrahedron Lett. 29, 5945-6.
313.   De Nys, R., Wright, A.D., Koenig, G.M., Sticher, O. (1993) Phytochemistry 32, 463-5.
314.   Cimino, G., De Stefano, S., Di Luccia, A. (1979) Experientia 35, 1277-8.
315.   Zubia, E., Gavagnin, M., Scognamiglio, G., Cimino, G. (1994) J. Nat. Prod. 57, 725-31.
316.   Crews, P., Bescansa, P., Bakus, G.J. (1985) Experientia 41, 690-1.
317.   Bohlmann, F., Jakupovic, J. (1979) Phytochemistry 18, 1987-92.
318.   Al-Said, M.S., Khalifa, S.I., El-Feraly, F.S. (1989) Phytochemistry 28, 297-8.
319.   Pika, J., Faulkner, D.J. (1995) Tetrahedron 51, 8189-98.
320.   Matsuo, A., Atsumi, K., Nakayama, M., Hayashi, S. (1981) J. Chem. Soc. Perkin Trans. I , 2816-
321.   Kirkup, M.P., Moore, R.E. (1983) Phytochemistry 22, 2539-41.
322.   Asakawa, Y., Toyota, M., Takemoto, T. (1980) Phytochemistry 19, 1799-1803.
323.   Jamieson, D.D., De Rome, P.J., Taylor, K.M. (1980) J. Pharm. Sci. 69, 462.
324.   De Pascual Teresa, J., San Felicano, A., Barrero, A.F., Gordaliza, M., Miguel del Corral, J.M.,
       Medarde, M. (1982) An. Quim., Ser. C 78, 425-6.
325.   San Feliciano, A., Gordaliza, M., Miguel del Corral, J.M., de la Puente, M.L. (1993)
       Phytochemistry 33, 631-33.
326.   Asakawa, Y., Takemoto, T. (1978) Phytochemistry 17, 153-4.
327.   Lin, J.H. (1982) Kuo Li Chung-kuo I Yao Yen Chiu So Yen Chiu Pao Kao , 147-62.
328.   Itokawa, H., Morita, H., Mihashi, S. (1980) Chem. Pharm. Bull. 28, 3452-54.
329.   Ramiandrasoa, F., Chuilon, S., Moretti, C., Kunesch, G. (1986) Plant. Med. Phytother. 20,

330.   Kimbu, S.F., Ngadjui, B., Sondengam, L.B., Njimi, T.K., Connolly, J.D., Fakunle, C.O. (1987)
       J. Nat. Prod. 50, 230-1.
331.   Firman, K., Kinoshita, T., Itai, A., Sankawa, U. (1988) Phytochemistry 27, 3887-91.
332.   Lognay, G., Marlier, M., Severin, M., Haubruge, E., Gibon, V., Trevejo, E. (1991) Flavour
       Fragrance J. 6, 87-91.
333.   Sirat, H.M. (1994) Planta Med. 60, 497.
334.   Cradwick, P.D., Sim, G.A. (1971) J. Chem. Soc., Chem. Commun. , 431-432.
335.   Velten, R., Klostermeyer, D., Steffan, B., Steglich, W., Kuschel, A., Anke, T. (1994) J. Antibiotics
       47, 1017-1024.
336.   Asakawa, Y., Toyota, M., Taira, Z., Takemoto, T., Kido, M., Ichikawa, Y. (1980) J. Chem. Soc.,
       Chem. Commun. , 1232-3.
337.   Spörle, J., Becker, H., Allen, N.S., Gupta, M.P. (1991) Z. Naturforsch., C: Biosci. 46, 183-8.
338.   Heim, J., Anke, T., Mocek, U., Steffan, B., Steglich, W. (1988) J. Antibiot. 41, 1752-7.
339.   Ciccio, J.F. (1984) Ing. Cienc. Quim. 8, 45-6.
340.   Nagashima, M., Nakatani, N. (1991) Chem. Express 6, 993-6.
341.   Asakawa, Y., Toyota, M., Uemoto, M., Aratani, T. (1976) Sentai Chii Zappo 7, 124-8.
342.   Asakawa, Y., Toyota, M., Takemoto, T. (1978) Phytochemistry 17, 457-60.
343.   Cimino, G., De Stefano, S., Minale, L. (1974) Experientia 30, 846-7.
344.   Froborg, J., Magnusson, G. (1978) Journal of the American Chemical Society 100, 6728-6733.

Finally I would like to express my sincere gratitude to all the people who are and
have been working at Department of Organic Chemistry 2, for contributing to
stimulating discussions in the coffee room as well as in the pub, and for creating a
pleasant and very stimulating atmosphere.
In particular, I wish to express my deep gratitude to:
• my supervisor docent Olov Sterner. I am very grateful for his encouraging and
   pragmatic spirit, as well as his humour, always having the glint in the eye. Olle’s
   support and never failing enthusiasm made my work very stimulating.
• professor Börje Wickberg for sharing his broad experience in chemistry with me.
• docent Peter Somfai, docent Ulf Berg, professor Göran Magnusson and docent
   Jan Kihlberg for stimulating discussions.
• doctors Ola Bergendorff and Jörgen Gustavsson, my dear colleagues, with
   whom I shared the same office for four memorable years. Ola and Jörgen
   introduced me to many things in the early days, and generously shared their
   great experience in chemistry as well as in beer with me.
• doctor Magnus Polla, for fruitful discussions about various matters. I spent a
   great time together with Magnus, Ola and Jörgen, especially during the
   unforgettable beer exploration trips to Czechoslovakia and Bornholm.
• my friend and colleague Rudong Shan, with whom I share the same office, for
   stimulating discussions and good friendship.
• Pia Kahnberg, who has contributed with hard labour in the reactivity studies
   with amino acids, and for her positive spirit, always spreading sunshine in the
• Rikard Hjertberg for great efforts and skilful work in the total synthesis of
• Ralf Davidsson for the isolation of polygodial, and for his efforts towards the
   drimane dialdehydes.
• the co-workers from all over the world in our group during the last years -
   Marco Clericuzio, Eremias Dagne, Mona Ezzelarab, Jean-Yves Fouron, Tomas
   Jarevång, Paloma Morales, Zijie Pang, Marc Stadler and Pierre Tane - for good
   friendship and stimulating discussions.
• the professors Heidrun & Timm Anke, and their biological team in
   Kaiserslautern for their co-operation, as well as their supervision and kind
   assistance during my fermentations at their laboratory. In particular, I am most
   grateful to Anke Mayer, Gudrun Schneider, Eckhard Thines and ”Mick” Gehrt
   for their exceptionally hospitality during my visits. I hope we continue to see
   each other in future.
• professor Mogens Nielsen and his co-workers at Sct Hans Hospital in Roskilde,
   for their co-operation, and for stimulating discussions at Gimle.
• Fritiof Pontén, my near friend and colleague, for many good laughs during all
   these years.
• Ulf Ellervik, for good friendship, fruitful co-operation in developing the courses
   and the demonstrations of our department. Ulf and ”Fifi” share my interest in
   trying new ideas in the progress and development of teaching and laboratory

• Johan Broddefalk, Tomas Jarevång, Ulf Nilsson, Michael Wilstermann and Jens
  Åhman for fruitful discussions in synthetic organic chemistry.
• doctor Mikael Elofsson and Katarina Flemmer, my everlasting colleagues, with
  whom I have shared my undergraduate and graduate studies since 1986,
  including many gastronomic events with ”matlaget”.
• Jens Åhman and Fritiof Pontén for constant co-operation in arranging
  department parties and pub-evenings. These guys are always keen on new
  parties, and Jens is a superior chef!
• Fritiof Pontén, Tatjana Vuljanic, doctor Ola Bergendorff and doctor Gerhard
  Erkel for CAS-online search assistance.
• Ulf Ellervik, Jörgen Gustavsson and Maria Levin for reading and commenting
  on the manuscript for this thesis (at a time when at least Ulf had häcken full
• Anders Sundin, Niclas Falk and Anders Wirdheim for assistance in computer
• Herman Lindborg for expert technical support.
• doctor Karl-Erik Bergquist for NMR assistance.
• Maria Levin for invaluable help with various practical matters, and for
  contributing to a rough, open and very comfortable atmosphere.
• all colleagues who are and have been working at Department of Organic
  Chemistry 1, in particular Fredrik Almquist, Håkan & Nina Bladh, Anna-Lena
  Gustavsson, Catarina Ludwig, Roger Olsson, Tina Persson, Johan Wennerberg,
  Ulf Wellmar and David Wensbo for good friendship, nice pub-evenings and
  parties, and for assistance with various things.
• Elsa Bohus-Jensen and Pelle Enestubbe, my high school teachers at Teknikum in
  Växjö, for creating a very stimulating atmosphere, and for introducing me and
  my classmates to the fascinating world of chemistry and bioscience. Elsa and
  Pelle were superb mentors, and without their enthusiasm and encouragement I
  would never even thought of university studies. Elsa even gave us personal grants
  so that we could go to the universities to meet scientists. Thank you very much
• Fredrik Svensson, my colleague and fellow twitcher, for 15 unforgettable years
  as brothers-in-arms. I am looking forward to the following 15!
• Anders Ekstrand, my near friend, for pioneering chemical experiments in my
  parents clothes closet 1978-1981.
• Lundell, Springsteen, Dylan, Young, Ebba Grön, Nationalteatern and Torson
  for creating creative conditions on lab.
• Martin & Karin, Fredrik Åsa & Viktor, Mikael & Cecilia and Kattis & Even for
  creating my spare time cheerful, for instance by offering me good beer and
  superb whisky.
• My parents Britta and Gunnar for their constant love and support, whatever I
  decided to do.
Finally, I thank my wife Pernilla for her love, support and patience throughout my
research, and during the time-consuming work the last year of completing this
thesis. She has all too many times been told ”We’ll do that when the thesis is
                                                            Nevertheless, now it is!


To top