NOVEL SECONDARY METABOLITES FROM SELECTED
COLD WATER MARINE INVERTEBRATES
By
DAVID ELLIS WILLIAMS
B.Sc.Hons, King's College, University of London, 1983
A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
in
THE FACULTY OF GRADUATE STUDIES
(Department of Chemistry)
We accept this thesis as conforming
to the required standard
THE UNIVERSITY OF BRITISH COLUMBIA
October 1987
® David Ellis Williams, 1987 In presenting this thesis in partial fulfilment of the requirements for an advanced
degree at the University of British Columbia, I agree that the Library shall make it
freely available for reference and study. I further agree that permission for extensive
copying of this thesis for scholarly purposes may be granted by the head of my
department or by his or her representatives. It is understood that copying or
publication of this thesis for financial gain shall not be allowed without my written
permission.
Department
The University of British Columbia 1956 Main Mall Vancouver, Canada
DE-6(3/81) ABSTRACT.
A study of the secondary metabolism of two nudibranchs and one soft coral has led to the isolation of eighteen new and two known secondary metabolites. The structures of all compounds were determined by a combination of the interpretation of spectral data, chemical degradations and interconversions, and single crystal x-ray diffraction analysis.
The British Columbian dorid nudibranch Diaulula sandiegensis yielded two new steroidal metabolites, diaulusterols A (41) and B
(42). The 25-(3-hydroxybutanoate) residue of diaulusterol A (41)
and the 2a,3a-diol array of both 41 and 42 are not commonly
encountered in naturally occurring steroids. Both metabolites
exhibited considerable antibacterial and antifungal activity.
Steroid 41 exhibited fish antifeedant activity. The relative
concentration of 4.1 and 42 in the skin extracts of D_.
sandiegensis appears to be related to the animals' seasonal
abundance.
Extracts of the British Columbian soft coral Gersemia
rubiformis yielded a series of ten diterpenes possessing cembrane
f170-175), pseudopterane (167-169) and gersolane (176) carbon
skeletons. The structure of an eleventh diterpene remains
unresolved. In addition, the structure of a degraded diterpene
possessing a 13-membered ring (122) is tentatively proposed.
G. rubiformis represents the first example of a soft coral to
- ii - yield pseudopterane diterpenes. The organism is the first to contain cembrane, pseudopterane and gersolane metabolites, a fact which has biogenetic implications. Two new sesquiterpenes were also isolated. Tochuinyl acetate (165) and dihydrotochuinyl acetate (166) represent the first examples of cuparane sesquiterpenes to be isolated from a soft coral. A biogenesis is proposed. Metabolite 166 exhibited fish antifeedant activity.
Investigations of Gersemia rubiformis collected in
Newfoundland waters revealed that the secondary metabolism differed from west coast specimens. The isolation of the new unstable sesquiterpene (+)-/?-cubebene-3-acetate (178) resulted.
Skin extracts of the dendronotoid nudibranch Toquina tetraquetra were examined in an attempt to correlate its feeding dependency and lack of predation to the presence of allomones.
Metabolites 165. 166. 170. 179 and the new butanoate diterpene
180 could be traced to the coelenterates which make up the animal's diet. Tochuinyl acetate (165). dihydrotochuinyl acetate
(166) and rubifolide (170) were previously found in extracts of
Gersemia rubiformis. Ptilosarcenone (179) has been reported as one of the major metabolites of the sea pen Ptilosarcus 213 gurneyi . The exact origin of a sixth metabolite, pukalide
(£1) , remains unknown. It is proposed that Tochuina tetraquetra selectively sequesters dietary metabolites for defensive purposes.
- iii - Table OJL Contents.
Page #
Abstract ii
List of Figures vi
List of Schemes ix
List of Tables x
Acknowledgements xii
Abbreviations xiii
I. Introduction 1
A. Overview 1
B. Natural Product Chemistry 1
C. Marine Natural Products 4
II. Nudibranchs 10
A. introduction: Nudibranch Zoology 10
B. Nudibranch Defenses: Evolutionary and Ecological Influences 13
C. The Chemistry of Nudibranchs 19
III. Secondary Metabolites from the Dorid Nudibranch,
Diaulula sandiegensis (Cooper, 1862) 23
A. Introduction 23
B. Isolation and Structural Elucidation 25
C. Biological Activities of Diaulusterol A (11) and Diaulusterol B (12) 46 D. Further Observations and Discussion 47
- iv - IV. Soft Corals 52
A. Introduction: Alcyonacean Zoology 52
B. Symbiotic Associations Between Alcyonarians and Algae (Zooxanthellae): Implications to the Origin of Isolated Metabolites 55
C. Why The Interest: The Biological and Ecological Significance of Some Coelenterate Metabolites 58
D. Chemistry of Soft Corals and Alcyonarians In General ..... 61
E. Review of the Chemistry of the Family Nephtheidae 70
V. Secondary Metabolites from the Soft Coral, Gersemia rubiformis (Ehrenberg, 1834), and the Dendronotoid Nudibranch, Toouina tetraquetra (Pallas, 1788) 78
A. Introduction 78
B. Isolation of the Metabolites 83
C. Structural Elucidation and Related Studies . . 87
(i) Sesquiterpenes 165 and 166 87 (ii) Pseudopterane diterpenoids, 167-169 . . . 105 (iii) Cembrane diterpenoids, 170-175 125 (iv) New carbon skeletal metabolites, 176-177 . 175 (v) Sesquiterpene 123. 192 (vi) Pukalide (£3J 206 (vii) Briarein diterpenoids, 179 and 180 .... 206 D. Biological Activities 210
E. Biosynthetic Postulates 212
F. Further Observations and Discussion 216
VI. Experimental 224
VII. Bibliography 258
- v - LAPt Of Figures.
Figure Page #
1 Phylogenetic classification of two nudibranchs . 11
2 Typical cryptobranch dor id nudibranch ...... 12
3 13C NMR spectrum of diaulusterol A (H) 26
4 NMR spectrum of diaulusterol A (11) 27 13 5 C NMR spectrum of diaulusterol A triacetate (13J 29 6 *H NMR spectrum of diaulusterol A triacetate (12) 31 7 Contour plot of 2D-HETC0R NMR spectrum of diaulusterol A triacetate (12.) 33
8 Comparison of the coupling constants of 41
and 11 with models 15. and 49-52 38
9 *H NMR spectrum of diaulusterol B (12) 44
10 "^H NMR spectrum of diaulusterol B diacetate (52.) 46 11 Structure of a typical alcyonarian 53 12 Phylogenetic classification of the soft
coral, Gersemia rubiformis 54
13 *H NMR spectrum of tochuinyl acetate (165) ... 88
14 13C NMR spectrum of tochuinyl acetate (165) ... 89
15 Contour plot of 2D-HETCOR NMR spectrum of tochuinyl acetate (lfL5J 90 16 Gated decoupled JC NMR spectrum of tochuinyl acetate (1£5J 94
17 SINEPT NMR spectra of tochuinyl acetate (165) . . 96
- vi - 18 H NMR spectrum of dihydrotochuinyl acetate (1£6J 101 19 C NMR spectrum of dihydrotochuinyl acetate
(1££) 102
20 13C NMR spectrum of gersemolide (167) 107
21 DEPT NMR spectra of gersemolide (167) 108
22 *H NMR spectrum of gersemolide (167) 109 23 Computer generated perspective drawings of gersemolide (167) 113
24 1H NMR spectra of A4,5Z, A7,17-isogersemolide 119
25 1H NMR spectrum of A4'5E, A7,17-isogersemolide (lfi&) 121
26 13C NMR spectrum of A4,5E, A7,17-isogersemolide
(169) 122
27 ^ NMR spectrum of rubifolide (170) ...... 126
28 13C NMR spectrum of rubifolide (170) 127
29 400MHz ^-H NMR spectrum of rubifolide (170) . . . 136
30 400MHz lanthanide shifted 1H NMR spectra
of rubifolide (12H) 137
31 *H NMR spectrum of epilophodione (171) 146
32 13C NMR spectrum of epilophodione (171) 147
33 ADEPT NMR spectra of epilophodione (171) .... 148
34 1H NMR spectrum of isoepilophodione A (122) . • . 151
35 13C NMR spectrum of isoepilophodione A (172) . . 152
36 *H NMR spectrum of isoepilophodione B (173) . . . 155
37 Low temperature 1H NMR spectra of isoepilophodione B (122.) 157 38 NMR spectrum of isoepilophodione C (174) . . . 160
- vii - 39 C NMR spectrum of isoepilophodione C (174) . . 161
40 1H NMR spectrum of rubifol (175) 165
41 13C NMR spectrum of rubifol (125.) 166
42 APT NMR spectrum of rubifol (1751 ...... 167
1 43 H NMR spectrum of rubifol (175) in CgDg .... 171
44 "*"H NMR spectrum of unknown diterpene 172
45 Proposed structures for unknown diterpene .... 174
46 XH NMR spectrum of gersolide (176) 176
47 Computer generated perspective drawing of
gersolide (176) 178
48 1H NMR spectrum of rubiformate (177) 182
49 13C NMR spectrum of rubiformate (177) 183
1 50 H NMR spectrum of rubiformate (177) in CgD6 . . 191
51 C NMR spectrum of ( + ) -y?-cubebene-3-acetate (1781 194 52 *H NMR spectrum of (+)-y?-cubebene-3-acetate (178) 196
53 1H NMR spectrum of the mixture of dihydro epimers 192 and 193 198
54 1H NMR spectrum of the mixture of dihydro
epimers 192 and 193 in CgDg 201
55 1H NMR spectrum of the butanoate 1M 207
56 Biogenetic speculation for (+)-/?-cubebene-3-acetate (12&) 216
- viii - ULs_t QL Schemes.
Scheme Page #
1 Biogenetic proposal for tochuinyl acetate (165) and dyhydrotochuinyl acetate (166) .... 213
2 Biogenetic proposals for the cembrane, pseudopterane, gersolane and degraded 13-membered ring diterpenoids, 167-177 214
3 Biogenetic proposal for (+) -fj-cubebene-3-acetate (12SJ 217
- ix - List ££ Tables.
Table Page #
1 13C NMR data comparison of 41 and 44 32
2 1H NMR data comparison of 41 and 45. 32
3 *H NMR data comparison of the steroidal methyl resonances of Air 45_, A3 and 42 35 13 4 C NMR data comparison of the steroidal methyl resonances of 41f 4£ and A3 36 5 Partial *H NMR assignments for diaulusterol A (11), diaulusterol B (42), triacetate 41 and diacetate 5_3_ 40
6 Partial C NMR assignments for diaulusterol A (41) and triacetate 42 42 7 Results of in vitro antibiotic assays for the steroidal metabolites, diaulusterol A (41) and diaulusterol B (42) 47
8 *H NMR data comparison of 165 and 181 91
9 13C NMR data comparison of 165 and 1£1 92
10 1H NMR assignments for tochuinyl acetate (165) and dihydrotochuinyl acetate (166) 99
11 13C NMR assignments for tochuinyl acetate (1£5_)
and dihydrotochuinyl acetate 166 100
12 1H NMR data comparison of 166. 183 and 184 ... 104
13 13C NMR data comparison of 166. 132. and 184 . . . 105 14 13C NMR assignments for gersemolide (167) and 169. and spectral comparison with 133. and 133 . • HI
15 1H NMR assignments and spectral comparison for gersemolide (1£2), 133 and 1£9_ 116
- x - 16 "^C NMR data comparison of 170 and 186 128
17 13C NMR assignments for rubifolide (HQ.) , epilophodione (171), isoepilophodiones A (172) and C (174) and rubifol (175), and spectral comparison with JJL8. and 189 129
18 *H NMR assignments for rubifolide (170), epilophodione f171), isoepilophodiones A (122), B (122.) and C (121) and rubifol (125.), and
spectral comparison with 188 and 189 131
19 Coupling constant data for rubifolide (170) . . . 141
20 Correlations observed in a long range 2D-HETCOR NMR spectrum of rubifolide (170) . . . 144 21 Subsequent NOE results observed for
epilophodione (171) 149
22 NOE results observed for rubifol (125.) 168
23 Spin-spin decoupling results for the eight-spin system found on Cll to C2 in rubifol (175) . . . 170 24 NOE results observed for gersolide (176) .... 179
25 *H NMR assignments for gersolide (176) and rubiformate (122) 180 1 3 26 C NMR assignments for rubiformate (177) and spectral comparison with epilophodione (171) . . 185 27 13C NMR data comparison of 177 and 1M 186 1 3 28 C NMR assignments for (+)-/?-cubebene-3-acetate (1781 195 29 NOE results observed for the dihydro epimers 1£2 and 122 200 30 C NMR data comparison of the methyl resonances of 12£, Ulr 123. and 1££ 203 31 ''"H NMR assignments for ( + ) -/?-cubebene-3-acetate (1781 and the dihydro epimer 1E2 205 32 1H NMR assignments for butanoate 180 and spectral comparison with 179 208
- xi - Results of in vitro cytotoxicity assays for rubifolide f 170) and epilophodione (HI) Acknowledgements.
I wish to express my appreciation to Professor Raymond J.
Andersen for his patience, ceaseless encouragement and guidance
throughout the course of this work, and for his assistance during
the preparation of the thesis. It has been a pleasure and a
learning experience to work with him.
Assistance given by Mike LeBlanc for performing bioassays
and collecting specimens is greatfully acknowleged. I thank the
staff of the Bamfield Marine Station and Drs. G. K. Eigendorf and
S. 0. Chan and their staff for efficient assistance throughout.
I must express my deepest appreciation and apologies to the
nudibranchs, Diaulula sandieaensis and Tpqvina tetraquetrfl, and
the soft coral, Gersemia rubiformis. I thank Mark Tischler for
his assistance in drawing structures. Finally and not least,
thanks to Peter Northcote and Peter Clifford for continually
bearing the cold waters of the Pacific, and Judy Needham for
unfailing support.
- xiii - Abbreviations.
A Angstrom units
ADEPT Automatic DEPT
APT Attached proton test
c concentration, gram per lOOmL solvent
CaSO, Calcium sulphate
C D Benzene-dg 6 6 CDC1. Chloroform-d^
CH2C12 Methylene chloride
DEPT Distortionless enhancement by polarisation transfer
e electronic glitch
EIMS Electron impact mass spectrum or spectroscopy
EtOAc Ethyl acetate
EU(F0D)3 Europium-1,1,1,2,2,3,3-heptafluoro-7,7- dimethyl-4,6-octanedione
FAB Fast atom bombardment
g grease resonance
GC Gas chromatography .
GCMS Gas chromatographic mass spectrum
2D-HETCOR Two dimensional heteronuclear correlation
HPLC High performance liquid chromatography
HRMS High resolution mass spectrum
IR Infra red
LRMS Low resolution mass spectrum
- xiv - M Parent Ion
MeOH Methanol mins minutes mp melting point
MS Mass spectrum or spectroscopy m/z mass to charge ratio
1H NMR Proton nuclear magnetic resonance 13 Carbon-13 nuclear magnetic resonance "LJC NMR Nuclear Overhauser enhancement NOE Octadecylsilane ODS relative rel., Rates of flow Rf solvent resonance s
SCUBA Self contained underwater breathing apparatus Selective insensitive nuclei enhancement SINEPT by polarisation transfer
TLC Thin layer chromatography
U unknown impurity resonance
UV Ultra violet w water resonance
W width (Hz) of NMR resonance signal at half h/2 height wt weight
- xv - Abbreviations for multiplicities of NMR signals:
s = singlet
d = doublet
t = triplet ("apparent" triplets are also referred as a triplet)
q = quartet
dd = doublet of doublets, etc..
b = broad signal
m = multiplet
Structures in this thesis are drawn, where appropriate, as depicted in reference source.
When relative stereochemistry ( a (down) and ft (up)) is assigned to geminal protons the assignment is made with respect to centres with a "known" relative stereochemistry in the corresponding 2D structural drawing.
- xvi - X. INTRODUCTION a., overview.
The purpose of the research undertaken and described in this
thesis was to isolate and elucidate the structures of previously
unknown natural products from selected cold water marine
invertebrates.
The majority of interesting secondary metabolites* exhibit 1 2
biological activities which are of pharmacological , ecological
or physiological interest. With this in mind, organisms for
study were chosen on the basis of in. vitro screening for
antifungal and antibacterial activities (it has been found that
such activities can generally be correlated to wider biological
activities) and ecological indicators such as a lack of
predation, or an unfouled surface. These determinants may prove
to be of little importance in terms of the roles that the
metabolites actually play in nature. However, they have proven to be an effective guide to the discovery of interesting secondary 4 metabolism .
£. Natural Product Chemistry.
The main driving force behind current natural product
research has been the desire to discover new metabolites of
potential use in the pharmaceutical or agricultural industries. * natural products and secondary metabolites are in this thesis regarded as synomynous.
1 Encouragment for these efforts stems from the fact that many
drugs in use today, such as morphine, atropine, penicillin and
streptomycin, are natural products or semisynthetic derivatives
thereof.
Natural products chemistry (the chemistry of secondary metabolites) has traditionally focused on terrestrial plants and
microorganisms. Historically only the major metabolites were
characterised. Structures were solved by chemical degradations
and interconversions and were confirmed by synthesis, since the
1960's, technological advances have enabled researchers to
elucidate the structures of more and more complex metabolites
present in ever smaller amounts. The recent isolation and 5
structural elucidation of the metabolites palytoxin (1) and
brevetoxin B (2.)^ would have been unimaginable 20 years ago.
As the process of structure elucidation becomes more
routine, there is an increased effort by natural products
chemists to understand why organisms produce such compounds. The
historical definition of secondary metabolites as compounds that
are not "essential to the basic protoplasmic metabolism of the
organism", or as "end products (waste or storage) of metabolism",
is becoming obsolete. It is now apparent that many secondary
metabolites are involved in the physiology of the producing
organism. For example, bonellin (3J , isolated from the marine
Echurian worm, Bonellia viridis. acts as a sex determinant and
produces marked sexual dimorphism33. Many other "secondary
metabolites" are used to mediate ecological interactions by
acting as chemical communicants. Some allomones (interspecies
chemical communicants) are used to ward off potential predators
2 3 C02H C02H 3
and pheromones (intraspecies chemical communicants) can be utilised to warn members of the same species of a potential OH "7 danger nearby ' .
c. Marine Natural Products.
Since its birth in the late 1960*s, marine natural products research has undergone an explosive growth. The technological advances in structural analysis and the advent of "cheap" safe
SCUBA have greatly facilitated the growth. New compounds are now discovered and reported at a rate that is fast enough to rapidly outdate any compilation of structures. Research in the field has spread geographically, and the active investigators have begun to explore in earnest some fascinating phenomena at the interface between biology and chemistry. Most metabolites isolated from marine organisms are exclusive to the marine environment, and many of the metabolites are restricted to genera or even single species.
The potential of the marine environment as a vast source of bioactive compounds is widely recognised. In the early 1970's,
1665 diverse marine species were screened for antitumor activity^ and it was found that nine percent contained compounds with activity compared with two or three percent of terrestrial organisms. Since there are more than a million species of marine invertebrates and more than 25000 known species of fish, such results are particularly exciting.
Marine natural product chemists seeking to isolate interesting compounds have followed three basic search strategies. One is to screen for biological activity the extracts of large numbers of organisms collected more or less randomly. A second strategy attempts to narrow the search by following promising lines of evidence. For example, tumors are known to occur in fin fish and molluscs, but they are conspicuously absent from other marine animals such as sharks, sponges and tunicates.
Extracts from many of the latter organisms have been examined^".
The third search strategy capitalizes on clues about the presence of bioactive metabolites by studying the ecology and behaviour of marine organisms. Selection pressures, in densely populated habitats such as coral reefs, have favoured the evolution of chemical defenses. Many organisms living in these habitats biosynthesise toxic metabolites or concentrate them from food sources in order to deter predation, to keep competitors from approaching too closely, and to keep themselves unfouled.
To date, literally thousands of compounds isolated from marine invertebrates have shown bioactivities which give them potential for pragmatic application, particularly in the areas of
5 human and veterinary medicine. The bryostatins isolated by Petit g et al. from the sessile plant like bryozoan Bugula neritina are strongly cytotoxic and anti-leukemic. An example of this
interesting family of metabolites is provided by bryostatin 1
(A) • Extracts of the soft bodied sessile organisms Euplexaura
flara. a gorgonian, and a Haliclona species, a sponge, yielded g the anti-mf lammatory butenolides, 5_-fL , and the anti-leukemic alkaloid manzamine A (j9_)*°, respectively. A unique metabolite
imbricatine (1H) isolated from the starfish Dermasterias
imbricata^ is capable of eliciting the "swimming" response of
the northeastern Pacific anemone, Stomphia coccinea. Imbricatine
(10) displays significant activity in the L1210 (ICgQ P388 (T/C 139 at 0.5mg/Kg) antineoplastic assays. Unfortunately, many marine natural products that are active in vitro are inactive in vivo. In addition, many are so toxic that they kill the host organism. However, the marine environment is gradually yielding a number of compounds (or synthetic analogues), that have found, or show a potential for, practical application. Nereistoxin f11), which is responsible for the insecticidal activity of the marine polychaete worm Lumbricpnerejs heteropoda12. led to the development of a 13 synthetic analogue padan (12) . which is now in common use in Japan. The semisynthetic manipulation of spongouridine (Ara-U) 14 (12) isolated from the sponge Cryptotethia crypta , and more 15 recently from the gorgonian Eunicella cavolini^ , has led to 1-p- D-arabinofuranosylcytosine (Ara-C) (14)16 an effective antitumor drug that has been used for over a decade in the treatment of leukemia. A number of other compounds are currently undergoing 17 18 clinical trials An example is provided by didemnin B (15.) isolated from extracts of the tunicate Trididemnum la 19 cyanophorum ' . Didemnin B (15.) is a potent antitumor agent 20 (IC500.00lMg/mL) and inhibitor of the Herpes simplex virus , 7 CH3CHOHCO—•-N-CH-CO —^MeLeu—*Thr —^Sta—•Hip—^Leu—. „_ I ()-•—Me2Tyr-—Pro J 15 *" and is undergoing clinical trials as an anti-leukemic druglb. Finally, a most satisfying aspect of marine natural products is the high level of interest exhibited by scientists outside the field. There is a genuine desire to determine the functions of new compounds in the marine environment. Interdisciplinary research involving chemists, marine biologists and marine ecologists can only serve to strengthen all these fields. 21 An excellent series of books edited by Scheuer and a 22 series of review articles by Faulkner provide timely, and exhaustive reviews of important aspects of marine natural 23 products research. Other authoritative reviews by Baker , 8 Christophersen , Faulkner et al. , Fenical , Moore , 28 29 Scheuer and Shields et al. , provide additional highlights. 9 XI. NUDTBRANCHS. A. introduction: Nudibranch Zoology- Molluscs of the class Gastropoda, subclass Opisthobranchia (Figure 1.) are slow moving animals whose synonyms include sea slug and naked snail. Of the 3000 described species of opisthobranchs30'3^" most are shell-less, though some have internal or greatly reduced shells. Nudibranchs comprise the largest of seven orders in the subclass Opisthobranchia, and are characterised by a complete absence of a shell. Adult nudibranchs range in size from 3mm to 300mm and are found throughout the world. Over 130 species have 32 been described on the west coast of North America . Colouration ranges from cryptic to very striking, conspicuous colours. The outer covering of a nudibranch is termed the mantle or dorsum (Figure 2.). A pair of rhinophores located near the nudibranch*s head are used as a chemosensory apparatus. The branchial plume is a respiratory structure (gill) usually located near the anus (some species of nudibranchs possess cerata which are finger like or club-shaped structures arranged in uniform 3 0 groups or clusters along each side of the dorsum , these function as respiratory or digestive organs) and the foot anchors the animal and is involved in locomotion. When nudibranchs are disturbed, most species retract their external structures leaving only the dorsum exposed. A few respond by swimming. The swimming response seems to be a transient means of escape, usually lasting for several seconds, swimming is evoked only when the animals are 10 MOLLUSCA CLASS MONOPLACOPHORA APLACOPHORA POLYPLACOPHORA BIVALVIA SCAPHOPODA GASTROPODA CEPHALOPODA SUBCLASS PROSOBRANCHIA OPISTHOBRANCHIA GYMNOMORPHA PULMONATA ORDER PYRAMIDELLIDA ACHOCHLIDIOIDA PLATYHEDYLOIDA CEPHALASPIDA NOTASPIDEA ANASPIDA THECOSOMATA GYMNOSOMATA SACOGLOSSA NUDIBRANCHIA ENTOCHONCHIDA SUBORDER AEOLIDACEA DENDRONOTOIDA ARMINOIDEA DORIDOIDA STTPFRFAMTI.Y EUDORIDACEA TRITONIIDAE DISCODORIDIDAE TOCHUINA DISCODORIS (DIAULULA) SPECTER TETRAQUETRA SANDIEGENSIS Figure 1. Phylogenetic classification of two nudibranchs^ N.B. Organisms classified according to Austin . 11 alarmed and it is not used as an active form of locomotion under normal circumstances. All nudibranchs are predators on other invertebrates. Dorids (suborder Doridacea) predominantly eat sponges, bryozoans (ectoprocts) and tunicates, while members of the other three suborders (Dendronotacea, Arminacea and Aeolidacea) primarily feed on coelenterates (sea anemones, soft corals, etc..)30. Nudibranchs are hermaphroditic. One advantage of hermaphrodism is the increased probability of finding a mate, since every individual of the same species is eligible. Little is known about how nudibranchs locate and recognise each other. They may secrete sex pheromones (a possible role for some of the 33 secondary metabolites isolated from nudibranchs) and the 3 4 rhinophores may act as the pheromone receptors . Nudibranch egg masses vary in size, shape and colour from species to species. Dorid egg masses are ribbon shaped, while aeolids and dendronotids lay egg "strings". Egg masses are very distinct and many can be identified to species. Defenses: Evolutionary M Ecological Influences. Even though nudibranchs have exposed soft tissue, limited mobility, and are often brightly coloured, they have few known predators35. Only a very few selected members of four groups of animals, opisthobranchs, crustaceans, asteroids and fishes, prey on nudibranchs34'36. This relative impunity to predation has led to speculation that nudibranchs utilize a variety of sophisticated defensive strategies . The primary level of defense employs colour which cryptically mimics the preferred dietary substrate of a species. Some nudibranchs are known to concentrate pigments from their •57 diet as a means of achieving cryptic colouration . The nudibranch AJLsLLSjg. sanouineae cooperi lives exclusively on the orange sponge Anthoarcuata graceae and apparently incorporates 38 3 5 carotenoid pigments from the sponge . However, Thompson demonstrated that cryptic nudibranchs may still not be acceptable to fish when placed in a setting where colour offered no concealment. It also appears that some well defended nudibranchs 37 blatently advertise themselves with conspicuous colouration . Thus, some nudibranchs apparently rely on a second level of defense. The second level of nudibranch defense may be some 3 6 combination of behavioral, morphological, or chemical effects . The most basic behavioral defense mechanism is the swimming response. On this coast, accomplished swimmers are species of the genera Tritonia. Melibe, Cumanotus. Flabellinopsis and 30 Dendronotus . Certain nudibranchs of the suborders 13 Dendronotacea, Aeolidacea and Arminacea possess large cerata which can be sacrificed to predators at the slightest provocation, hence enabling captured nudibranchs to escape (the cerata are rapidly regenerated). Some dorid (suborder Doridacea) nudibranchs possess spicules acquired from their dietary sponges. The spicules give the animal a rigid shape. A voracious nudibranch predator, the opisthobranch Navoanax inermis. does not eat spicule-containing dorids, while readily consuming non-spiculose types . Spicules may provide some protection against predators, however, the relative importance of this morphological defense is believed to be rather low34. Certain aeolid nudibranchs selectively sequester and utilize intact nematocyst stinging cell organelles35'40 obtained from dietary coelenterates. The unfired nematocysts are stored in a functional state in specialised cynidocyst cells located in the skin, typically at the tips of the cerata34. When an aeolid is disturbed, clouds of stinging nematocysts are released causing some tropical aeolids to be hazardous to man40. It is now widely accepted that opisthobranchs utilize some form of chemical defense at the secondary level, and it is the nature of these secretions that has led to an intense interest by marine natural product chemists in recent years. Several dorids secrete sulphuric acid (pH=l) along their skin surface as a chemical deterrent4*. Nudibranchs, particularly dorids, possess abundant numbers of non-mucous secretory skin glands in the 35 dorsum . The position and function of the glands is most 14 consistent with a defensive role . Their placement explains the observation that most secondary metabolites from nudibranchs can be extracted by a relatively short immersion of the intact animal in a suitable solvent. These inferences are corroborated by studies that have found feeding inhibitors in the dorsum but not in the gut, foot, or head parts of nudibranchs42. It has been found that all four nudibranch suborders secrete non-acidic 43 noxious substances , some of which taste bitter, or bitter sour35. Nudibranch secondary metabolites can have several origins: (i) derived directly from a dietary source; (ii) derived indirectly by chemical modification of a metabolite obtained from a dietary source; (ii) or £e_ novo biosynthesis. Observations by 43c Johannes in 1963 , suggested that the nudibranch Phyllidia varicosa secreted a toxic mucus. This led to the isolation in 1975 of a structurally novel metabolite 9-isocyanopupukeanane (IfL)44 which had fish antifeedant activity (one of the first examples of a marine metabolite with a known biological role). The nudibranch obtained this metabolite directly by concentrating 45 it from a dietary Hymeniacidon species of sponge . In the nudibranch, Chromodoris marislae. a metabolite is derived 46 indirectly by chemical modification . The major metabolite marislin (12) is related to the metabolite pleraplysillin-2 (12.), isolated from the sponge Pleraplysilla spinifera47. by a simple rearrangement. Only a few species of nudibranchs appear to be capable of de novo biosynthesis of defensive or other secondary metabolites. Dendrodoris limbata biosynthesises polygodial (12)48, a 15 metabolite endowed with antifeedant properties. The incorporation 3 4 of [ C] mevalonic acid into the terpene portions of 2SL and 21 suggests that species of the genus Archidoris can also elaborate 49 terpenes . OH CHO 20 21 16 Since nudibranch metabolites are mainly of dietary origin 50 (the origin of a number remains unclear ) the skin chemistry of a species can differ greatly from one collecting location to another. A number of clear examples are documented. Specimens of Cadlina luteomarginata from Vancouver, British Columbia contained 2a furodysin (22) i furodysinin (21) and microcionin-2 (24.) • Specimens collected in the Queen Charlotte Islands, British 51 Columbia contained the furanoditerpene, mariginatafuran (25.) • Dihydropallescensin-2 (2fL) was found in the skin extracts of specimens of £. luteomarginata collected off of La Jolla, 42c California . It was simultaneously shown that the majority of metabolites present in the skin extracts of £.. luteomarginata were obtained from the sponges in the nudibranch's diet, and that the observed variation in skin chemistry was a consequence of changing diet with changing collection site. H H 22 23 24 25 26 17 The secondary metabolites isolated from nudibranchs for which toxic or antifeedant properties have been established, fall A *5 A A very roughly into three groups: (i) isocyanosesquiterpenes ' ; 52 (i) furanoterpenes ; and (iii) sesquiterpenes with a drimane 53 skeleton . The first two groups appear to be of dietary origin, while the drimane sesquiterpenes are biosynthesised d_e_ novo by the nudibranchs. It appears that compounds of these three groups may display antifeedant properties towards organisms other than fish54 and that similar defense strategies are operating among 55 opisthobranch molluscs and insects . The development of a "dietary-derived" chemical defense mechanism has probably had a profound influence on the evolutionary development of nudibranchs. It is believed that the loss of the typical molluscan shell is related to the gain of this chemical defense. Circumstantial evidence supports the notion that the chemical defense was preadaptive to the loss of 37 the shell , since a number of interesting secondary metabolites have been isolated from shelled Gastropod cousins56. Also the chemical defense may have played an important role in the adaptive radiation of nudibranchs through their dependency on 37 specialist diets . Further to this, it is asserted that sponge- feeding nudibranchs tend to predominate in tropical faunas where toxicity among sponges as a general rule is at its greatest371,57'58, while those that feed upon bryozoans and other 37 58 animals are more common in temperate waters ' Little is known about the relative importance of any specific 18 nudibranch defense mechanism, but it seems apparent that nudibranchs utilize a combination of strategies. It has not been unequivocally proven, but circumstantial evidence and evolutionary hypothesis suggest that chemical defensive secretions (allomones) play an important role. C. The Chemistry of. Nudibranchs. Many novel secondary metabolites have been isolated from gastropods, particularly nudibranchs. The scope of this work is too extensive to be reviewed in full in this thesis. Reviews of nudibranch chemistry up to 1982 were provided by Schulte 59 42c et al. and Thompson et al. . An update can be found in 22 Faulkner's review papers . Additional in depth discussions are to be found in the Ph.D., dissertations of K. Gustafson53<^ and S. W. Ayer60. It is appropriate, however, that a brief overview of the literature since Faulkner's last review paper covering the published work up to and including July 1985 be included in this thesis. The following review covers the period from July 1985 up to the citations reported in CA Selects, issue 9, May 4tn, 1987, and/or Chemical Abstracts, No. 20, May 18tn, 1987. Although chemical studies of shelled molluscs are of increasing interest61, shell-less members of the subclass 32 Opisthobranchia, aplysiomorphs (lately known as anaspids ) (sea 62 hares) and nudibranchs, continue to yield interesting natural products. A partially classified nudibranch belonging to the genus Phyllidia from Sri Lanka, yielded the previously unreported 63 metabolite 3-isocyanotheonellin (22) • Extraction of the nudibranch, Chromadoris macfarlandi19 . has resulted in the isolation of the two aromatic norditerpenes macfarlandins A (28) and B (22)64* and the three diterpenes M, 21 and 2265. All of these compounds display antimicrobial activities, and all appear to be derived from sponge metabolites in the nudibranch's diet. Similarly, extracts of £.. funera yielded the metabolites 22 and 34f which appear to be oxidation products of compounds derived from the sponge Dysidea herbacea66. Both metabolites cause food rejection in the "spotted kelpfish" Gibbonsia elegans. The nudibranch D_oxi£. verrucosu yielded the first naturally occurring analogue of methylthioadenosine 25., which possesses "diverse 67 regulatory activities" . 0 20 OAc Attention has also turned toward nudibranch eggmasses which are yielding extraordinary macrolides. The brilliantly coloured nudibranch, Hexabranchus sanguineus, deposits its striking red eggmasses in underwater caves. Though exposed and vulnerable these eggs have only one predator, the aeolid nudibranch, Favorinus japonicus. This virtual immunity to predation led . 6 8 Scheuer et al. to investigate the organic constituents of the eggs resulting in the isolation of two extraordinary macrolides, ulapaulide A (3JL) and B (3_1) , which inhibit L1210 leukemia cell proliferation (IC5Q0.01-0.03 g/mL). Egg masses of an unidentified nudibranch collected at Kabira Bay, located in Japanese waters, yielded a series of novel macrolides, for example kabiramide C (3JL) . This compound possesses antifungal activity and inhibits cell division of fertilized starfish eggs at low 21 XIX. SECONDARY METABOLITES FROM TJH£ DORID NUDIBRANCH, Diaulula sandieoensis (Cooper. 3 862). h. introduction. The intertidal and subtidal regions of the Pacific Coast of North America are inhabited by sizeable populations of dorid 32 nudibranchs . Many of the species have distributional ranges that extend along the entire coastline from Alaska to Mexico. PiauJ-ulfl sandjegepsis (recently Discodoris sandieoensis32) . suborder Doridoida, family Discodorididae (Figure 1.), provides an example of a dorid which has a documented distribution range that extends along the Pacific Coast of North America from 32 southern Alaska to Mexico, and off the coasts of Japan . Walker and Faulkner isolated a series of nine chlorinated acetylenes related to 3_9_ from specimens of D_. sandieoensis collected at Point Loma, California70. Fuhrman et al.7* isolated isoguanosine (40) from the digestive glands of specimens collected at Monterey Bay and the Channel Islands off Santa Barbara, California. Consideration of the reported findings, the relative abundance of the nudibranchs and their apparent lack of predation led us to examine specimens of Diaulula sandieqensis collected in Barkley Sound, British Columbia. The investigation of the secondary metabolites from the skin extracts of D_. sandiegnesis was first initiated by S. W. Ayer in September 1980, and the studies reported herein were begun in earnest in April 1985. Diaulula sandieoensis was probably first identified by 3 2 Cooper in 1862 . The animal typically inhabits exposed rocky 23 channels at a depth of 1-5 meters, but can be found intertidally and to a depth of 30 meters. Body length can reach 8cm, but most fall into the range of 4-5cm. The dorsum has a velvety appearance produced by minute tubercles. The ground colour of this dorid is whitish-yellow to very pale brown broken by various numbers (typically 2-8) of brown to black markings shaped either as blotches or as distinctive doughnut-shaped rings. Specimens having no spots have been reported. The rhinophores possess 20-30 lamellae and the six branchial plumes are white tripinnate and are retractable. The animal supposedly feeds on intertidal and subtidal sponges of the genera Halichondria and Haliclona30'72. Diaulula sandiegensis is well camouflaged in its natural environment, but with a little effort and a sense of "tuning in" on the objective they were readily collected by hand using SCUBA. D_. sandiegensis were most abundant on exposed rocky reefs, particularly in surge channels. The animal was collected throughout the year at depths of 1 to 15 meters. 24 B_. isolation and Structural Elucidation. Freshly collected animals (152 specimens) were immediately immersed in methanol and allowed to extract at 2°C for two days. The extraction solvent was decanted from the nudibranchs, evaporated in vacuof and the residue was partitioned between brine and "ethyl acetate. Fractionation of the ethyl acetate soluble material by flash, Sephadex LH-20 and reverse-phase HPLC chromatographies produced pure samples of diaulusterol A (41) and diaulusterol B (42). 2 43 R' = Ac.R =COCH2CH(OAc)Me 53 R^Ac. r2 = h Diaulusterol A (41) was isolated as a UV absorbing clear oil that showed IR bands at 3380, 1722, 1664 and 1626cm"1, appropriate for hydroxyl, saturated carbonyl, unsaturated carbonyl and alkene functionalities. The EIMS of 41 provided little information since no parent ion, or recognisable fragment ions were observed. Diaulusterol A (41) showed resonances for 13 thirty one carbon atoms in its C NMR spectrum (Figure 3.) and its 1H NMR spectrum (Figure 4.) possessed a series of five methyl resonances at 0.65(s, 3H) , 0.96(d, J=6.8Hz, 3H), 1.17(s, 3H), 25 ww I I I wI I I I I I I I I I I M I I I I I J I M I I I I I I I I I I 60 40 20 PPM 0 27 1.21(d, J=7.0Hz, 1H) and 1.46(sf 6H)ppm. Other notable features in the *H NMR spectrum of 41 were three very broad (W^y2=15-27Hz) one proton "singlets" at 2.66, 2.87, 3.25ppm, three possible carbinol methine proton resonances at 3.88(dt, J=10.7,4.9Hz, 1H), 4.15(m, 1H), 4.28(t, 4.9Hz, lH)ppm, and two olefinic proton resonances at 5.87(t, J=0.9Hz, 1H) and 6.54(d, J=4.9Hz, lH)ppm. Diaulusterol A (41) steadily decomposed to give a complex mixture of very polar materials when it was exposed to NMR solvents or other manipulations. Metabolite 41 possessed an IR 3 3 * band appropriate for hydroxyl and the C and ADEPT (optimised 73 for polarisation transfer through 140Hz coupling) NMR experiments indicated the presence of three secondary alcohol carbons resonating at 64.4(CH), 65.2(CH) and 66.7(CH)ppm. Confirmation for this was provided by the three carbinol methine proton resonances and the three very broad one proton "singlets". Hence, 41 was expected to acetylate readily. Acetylation of 41 with acetic anhydride and pyridine at room temperature generated the triacetate 42. which proved to be a very stable substance. Triacetate 42. was shown by EIMS to have a molecular C formula of 37H540g (observed m/z 642.3779, requires 642.3768), which represents the addition of six carbon atoms to the thirty 13 one that were observed in the C NMR spectrum of 41. The thirty seven resonances observed in the C NMR spectrum of 42. (Figure 5.) and the three new methyl singlet resonances at 2.03, 2.06 and 2.10ppm in addition to the observed downfield shifts of the An experiment which allows for the full proton count to be assigned to each carbon atom, for an example see Figure 33., page 148. For a detailed discussion see reference 73. 28 carbinol methine resonances to 5.11(dt, J=13.8f4.8Hz, 1H) , 5.24(m, 1H) and 5.60(t, J=5.4Hz, lH)ppm and the loss of the three very broad "singlets" in the *H NMR spectrum of 42 (Figure 6.) confirmed the formation of a triacetate. Subtraction of the atoms present in the three acetyl residues from the molecular formula of 42 resulted in a molecular formula C3iH48°s ^or diaulusterol A (41)t which corresponds to a very low intensity M+ peak (rel., intensity <0.2) at a m/z of 516 daltons in the FAB MS of 41. Three of the oxygen atoms in this formula could be accounted for by the three secondary alcohols which underwent acetylation. The IR bands at 1722 and 1664cm-1 13 and the C NMR resonances at 172.4(C) and 188.7(C)ppm indicated that the remaining oxygen atoms were possibly present as ester and cross-conjugated ketone funtionalities. Compare for example, the partial 13C NMR resonances of the cross conjugated keto "7 A 1 steroid 44 and the H NMR resonances of the ecdysone-like 75 metabolite, acetylpinnasterol 45. » with those of 41 (Tables 1. and 2.). The molecular formula C31H4gOg requires 8 units of unsaturation. Two of these could be accounted for by the two carbonyls. The *H NMR spectrum of diaulusterol A 41 possessed two olefinic proton resonances at 5.87 and 6.54ppm. The short range 2D-HETC0R NMR spectrum76 of triacetate 42 (see Figure 7.) by analogy correlated the olefinic proton resonances with the carbon resonances at 123.6(CH) and 128.1(CH)ppm, respectively, and the resonances at 146.2(C) and 166.9(C)ppm were assigned to the remaining carbons of two trisubstituted olefins. Comparison of 13 74 the C NMR chemical shift values with those for metabolite 44 30 44 45 R=Ac 57 R=H 13 Table 1. C NMR data comparison of Al and 44. Chemical shift, ppm. Chemical shift, ppm. Carbon # Al Carbon # AAa 6 188.7 3 186.3 7 123.6 4 123.9 8 166.9 5 168.9 reference 74, 25.20MHz, CDCl-j. Table 2. H NMR data comparison of Al and 45. Chemical shift, ppm. Carbon # Al AST 4 6.54(d, J=4.9Hz, 1H) 6.82(d, J=2.0Hz, 1H) 7 5.87(t, J=0.9HZ, 1H) 6.03(t, J=2.0Hz, 1H) 19 1.17(s, 3H) 1.14(s, 3H) reference 75, 400MHz, C5D5N. 32 (Table 1.) provided additional credence to the assignments. The remaining four units of unsaturation could be accounted for by four rings. The presence of four rings, the methyl resonances in the *H NMR spectrum at 0.65, 0.96, 1.17 and 1.46ppm, and the existence 13 of methyl carbons in the C NMR spectrum of AX at 12.5(CH3), 18.7(CH3), 22.3(CH3), 26.1(CH3) and 26.2(CH3)ppm, implied that diaulusterol A (11) consisted of a steroidal nucleus. Supporting 1 13 this hypothesis was the close agreement of the H and C NMR methyl chemical shifts of Al and the steroidal metabolites 4575, AL, 4277 and 1£78 (Tables 3. and 4.). A series of decoupling experiments resulted in the elaboration of the ester containing fragment. Irradiation of the carbinol methine resonance at 4.15ppm in the *H NMR spectrum of 41 collapsed the methyl doublet at 1.21ppm to a singlet, and simplified the two resonances at 2.34(dd, J=16.4,9.0Hz, 1H) and 2.24(dd, J=16.4,3.8Hz, lH)ppm to an AB quartet (J=16.4Hz). This isolated six proton spin system was assigned to a 3- hydroxybutanoate moiety. Subtraction of the four carbons of this fragment from the molecular formula of Al leaves the twenty seven carbons of a basic steroid skeleton, supporting the hypothesis of a steroidal structure for 41. The short range 2D-HETCOR NMR spectrum of the triacetate 42 correlated the1 H NMR resonance at 13 1.46(s, 6H)ppm in H with the C NMR resonances at 26.1(CH3) and 26.2(CH3)ppm (C26 and C27). The values are all somewhat deshielded from typical steroidal chemical shifts at the C26 and C27 positions (see Table 3. and 4.). By inference, the presence of a 13C NMR resonance at 83.6(C)ppm indicated that the 34 OS03Na OAc Table 3. H NMR data comparison of the steroidal methyl resonances of Al, AJLr A£ and Al Chemical shift, ppm. Carbon # Al A5.a AL A2b 18 10.65s 1.05s 10.67s 0.78s 19 21.17s 21.14s 1.00s 1.02s 21 30.96d 1.51s 30.92d 1.28d 26 41.46s 0.97dz 0.87d 41.42s 27 41.46s 0.98dz 0.87d 41.42s areference 75, 400MHz, CgDgN. breference 77, 500MHz, CgDgN. zassignments within a column may be interchanged. Compare the values of the chemical shifts with the same superscripts 1, 2, 3, or 4. 35 Chemical shift, ppm. Carbon #41 M 4£a 18 112.5 112.0 15.5 19 222.3 19.3 224.0 21 318.7 318.6 13.3 26 426.1 22.5 429.7Z 27 426.2 22.7 429.4Z reference 78, p. 238, C5D5N, spectrometer frequency not reported. Assignments within a column may be interchanged. Compare the values of the chemical shifts with the same superscripts 1, 2, 3, or 4. butanoate fragment was attached at C25 of the steroid sidechain. The existence of a butanoate moiety was confirmed by the observation of a fragment ion at a m/z of 496 daltons in the MS of the triacetate 41r which corresponds to the loss of 3- acetoxybutanoic acid via a McLafferty rearrangement. A second series of *H NMR decoupling experiments allowed the placement of much of the remaining functionality on the steroid nucleus. Irradiation of the methine proton at 3.88ppm in the spectrum of 41 collapsed the methine triplet at 4.28ppm to a doublet (J=4.9Hz) and simplified a pair of resonances at 1.75- 1.85ppm to an apparent AB quartet. Irradiation of the methine resonance at 4.28ppm simplified the doublet of triplets at 3.88ppm to a doublet of doublets (J=10.7,4.9Hz) and collapsed the olefinic doublet at 6.54(J=4.9Hz)ppm to a singlet, while irradiation at 6.54ppm simplified the methine resonance at 4.28ppm to a doublet (J=4.9Hz). The only way to situate this five proton spin system on a steroid nucleus was to assign the reson• ance at 1.75-1.85ppm to a pair of geminal protons at CI, the resonance at 3.88 and 4.28ppm to carbinol methine protons at C2 and C3, respectively, and the resonance at 6.54ppm to an olefinic proton at C4. The demonstration of a NOE between the methyl protons at 1.23ppm (C19) and a methine proton at 5.10ppm (H2) in the triacetate 41 required that H2 be /?. An additional NOE between the methine proton at 5.10ppm (H2) and its vicinal neigh• bour, which appears at 5.60ppm (H3) in the *H NMR spectrum of 41, showed that H3 was also p. Comparison of the observed coupling constants for this spin system in 41 and 41 to the corresponding 75 798081 coupling constants in the model compounds 45. and 45_-52 ' confirmed the configurational assignment (see Figure 8.). 37 Ac H°^2^vl/f AcO^^sf^j Ov^J/i H0JkA; Aco-'-^i HO-'^AJ = 41 J12=107Hz 43 Jl>2=13-8Hz 45 J1(2 10 5.3 5Hz J2'3 = 49Hz J23 = 4-8Hz J2 3=7 0Hz J 9Hz 3,V* J3#4 = 5-*Hz J3; = 20Hz 6=3 90.dt 49 J, 2--100,4-5Hz 50 J12M2,A5Hz J2 3 = 8 0Hz J23=A 5Hz 81 51 R = H, J23=10Hz 52 J3i< = 5Hz R=Ac,J23 = 9Hz J3>A=<1Hz Figure 8. Comparison of the coupling constants of 41, 42. and the models 45. and 49-52. Since the nature of the A ring in 41 and 42. had been established, the remaining functionality indicated by the 38 spectral data and required by the molecular formula, the cross conjugated ketone and the associated trisubstituted olefin, had to be placed with the ketone functionality at C6 and the olefin 1 13 at C7-C8. The resulting H and C NMR assignments for diaulusterol A (11) and triacetate 12 (facilitated by the short range 2D-HETCOR NMR spectrum of 12) are given in Tables 5. and 6., respectively, and are in agreement with the proposed structure. Consistent with the structural assignment is the splitting of the H7 olefinic proton resonance (5.87ppm) in H into a triplet (J=0.9Hz) via allylic coupling to H9 and H14, which resonate at 2.08(ddm, J=12.0,7.2Hz, 1H) and 2.41(m, lH)ppm. The allylic coupling was verified by irradiating the olefinic proton at 5.87ppm (H7) which resulted in a sharpening of both the H9 and H14 resonances. The observation of intense fragment ions at m/z Of 582(M+ - HOAc), 522(M+ - 2xHOAc), 376(M+ - (2xHOAc + 3- acetoxybutanoic acid)), 361(M+ - (2xHOAc + 3-acetoxybutanoic acid + + + CH3)), 265(M - (sidechain + 2xHOAc)) and 496(M - (3- acetoxybutanoic acid)) daltons in the MS of !2r provides further confirmation of the proposed structure. The relative stereochemistry for ring A was established by NMR as discussed in the preceeding paragraphs. The observed chemical shifts for the C18 (^ 0.65ppm, 13C 12.5ppm) and C21 (1H 13 0.96ppm, C 18.7ppm) methyl groups in H were virtually identical to those for cholesterol (Ifi.) (C18 0.67, 12.0ppm: C21 0.92, 18.6ppm), suggesting that the relative configurations at C13, C14, C17 and C20 were identical in the two molecules. The 39 Table 5. Partial NMR assignments for diaulusterol A (11) , diaulusterol B (12) t triacetate 12 and diacetate 5_3_. Chemical shift , ppm. Carbon # 11 12 1 1.75-1.85 2 3.88(dt, J=10.7f4.9Hz, IH) 3.88(m, IH) 3 4.28(t, J=4.9Hz, IH) 4.27(t, J=5. 2Hz, IH) 4 6.54(df J=4.9Hz, IH) 6.54(d, J=5. 2Hz, IH) 7 5.87(t, J=0.9Hz, IH) 5.87(bs r IH) Z Z 9 2.08(ddm, J=12.0f7.2Hzf 1H) 2.07(m, 1H) 2 1 Z 14 2.41(m/ 1H) ' 2.42(m, 1H) 18 0.65(s, 3H) 0.65(s, 3H) 19 1.17(s, 3H) 1.17(s, 3H) 21 0.96(df J=6.8Hz, 3H) 0.97(d, J=7. 2Hz, 3H) 26 1.46(s) 1.23(s) 27 1.46(s) 1.23(s) 29 2.34(dd, J=16.4,9.0Hz, IH) / 2.42(dd, J=16.4,3.8Hz, IH) / 30 4.15(m, IH) / 31 1.21(d, J=7.0Hz, 3H) / -OH 2.66(bs, W, /9=27Hz, IH) / 2.87(bs, W?^=27Hz, IH) / 3.25(bs, W^/2=15Hz' 1R) / cont'd, absolute stereochemistry at C20 in cholesterol (M) is 20R. It 82 has been demonstrated by Vanderah and Djerassi that in 20S 40 Table 5. continued. Chemical shift, ppm. Carbon # 4JL 52. 1 ? ? 2 5.10(dt, J=13.8,4.8Hz, 1H) 5.11(dt, J=13.4,3.9Hz, 1H) 3 5.60(m, 1H) 5.60(m, 1H) 4 6.43(d, J=5.4Hz, 1H) 6.43(d, J=6.9Hz, 1H) 7 5.89(bs, 1H) 5.89(bs, 1H) 9 ? ?Z 14 ? 2.42(tm, J=9.5Hz, 1H)Z 18 0.65(s, 3H) 0.65(s, 3H) 19 1.23(s, 3H) 1.24(s) 21 0.96(d, J=7.2Hz, 3H) 0.98(d, J=7.lHz, 3H) 26 1.42(s) 1.24(s) 27 1.42(s) 1.24(s) 29 2.42(dd, J=16.2,6.3Hz, 1H) / 2.57(dd, J=16.2,7.7HZ, 1H) / 30 5.23(m, 1H) / 31 1.29(d, J=7.6Hz, 3H) / -OC(0)CH^ 2.03(s, 3H), 2.06(s, 3H), 2.04(s, 3H), 2.08(S,3H) J 2.10(s, 3H) Assignments within a column may be interchanged. Aits underneath doublet of doublets at 2.42ppm. "?nunable to assign. steroids, for comparable molecules, the C21 methyl group is shifted approximately O.lppm upfield in the *H NMR spectrum. This 41 Table 6. Partial C NMR assignments for diaulusterol A (H) and triacetate 43. Chemical shift, ppm. Carbon # ll3 12 2 65.2(CH)Z 67.5(CH)Z 3 64.4(CH) 64.7(CH) 4 128.KCH) 124.2(CH) 5 146.2(C) 147.6(C) 6 188.7(C) 187.6(C) 7 123.6(CH) 123.7(CH) 8 166.9(C) 166.5(C) 18 12.5(CH3) 12.5(CH3) y 19 22.3(CH3) 20.4(CH3) 21 18.7(CH3) 18.6(CH3) 25 83.6(C) 83.1(C) X X 26 26.1(CH3) 26.0(CH3) X X 27 26.2(CH3) 25.9(CH3) 28 172.4(C) 170.1(C)W 29 ? 47.7(CH2) 30 66.7(CH)Z 67.6(CH)Z y 31 21.1{CH3) 19.8(CH3) -oc CO- / 169.4W, 170 / 170.1W. "?"unable to assign C29 by comparison with 12. "Assignments within a column may be interchanged. 42 further supports the hypothesis that the relative configuration at C20 in Al is identical to that in cholesterol (AfL) • Irradiation of the C18 methyl protons in A3, at 0.65ppm failed to induce an observable NOE into the C14 proton and irradiation of the C19 methyl protons at 1.23ppm likewise failed to induce an NOE into the C9 proton. This negative evidence, while not proving the stereochemical assignment, is at least consistent with that shown. It is postulated from biosynthetic reasoning that diaulusterol A (Al) possesses the absolute stereochemistry typically associated with naturally occurring steroids. The relative stereochemistry at C30, the carbinol chiral carbon in the butanoate moiety, remains undetermined. Attempts to make the 2,3-acetonide of Al/ with intention of subsequent application of 83 the Horeau method to determine the absolute configuration at C30, failed. Diaulusterol B (A2) was also isolated as a clear oil that was shown by EIMS to have a molecular formula of C27H42°4 (observed m/z 430.3055, requires 430.3083), requiring 7 units of unsaturation. The 1H NMR spectrum of A2 (Figure 9.) was identical to that of Al in all respects except for the absence of the resonances due to the 3-hydroxybutanoate residue. Similarly, the IR was essentially identical apart from the absence of an ester absorption band (1722cm"1 in 141). The observation of intense + + fragment ions at m/z of 412(M - H20), 397(M - (H20 + CH3)), + + + 379(M - (2xH20 + CH3)), 301(M -(sidechain)) and 283(M - (K20 + sidechain)) daltons in the MS of A2r further confirmed that diaulusterol B should be assigned structure A2. Treatment of the triol A2 with acetic anhydride and pyridine at room 43 temperature generated the diacetate 52. as expected. The H NMR spectrum (Figure 10.) was perfectly consistent with the proposed structure 52., as was the IR and mass spectrum. Partial *H NMR assignments for 42 and 52. are included in Table 5. Diaulusterol B (42) was isolated from rapidly processed extracts of Diaulula sandieqensis and in all the isolations undertaken the ratio of 41 to 42 was virtually identical (7 1/2:1), thereby eliminating the possibility that 42 is an artifact formed by hydrolysis of diaulusterol A (41). £. Biological Activities of. Diaulusterol A. (41) aM Diaulusterol £ (42). Diaulusterol A (41) and diaulusterol B (42) were present in relatively high concentrations (0.16mg (0.06% dry wt.) and 0.02mg (0.007% dry wt.) per animal, respectively). It seems reasonable, therefore, to expect that these two steroidal metabolites might have a biological role. The metabolites 41 and 42 exhibit considerable antimicrobial activity as demonstrated by the results of the in vitro antibacterial and antifungal assays, Table 7. A fish antifeedant bioassay was undertaken using commercial goldfish tCarassius auratus) as the test species. The results showed that 41 inhibits feeding at a concentration of 0.20mg/pellet (0.0074rag/mg of food pellet), and there was a marked preference for the control at concentrations of less than 0.05mg/pellet (0.0019mg/mg of food pellet). Metabolite 42 showed no activity up to the maximum concentration tested, 0.20mg/pellet. Since the average concentration of metabolite 41 45 Table 7. Results of jji vitro antibiotic assays for the steroidal metabolites, diaulusterol A (41) and diaulusterol B (42). (Minimum inhibitory concentration reported in /ig/disc.) Micro-organism. Compound ScA BS PythD RhyzS diaulusterol A (41) 1270 63 1270 63 diaulusterol B (42) 48 10 387 48 Bacteria: Fungi: ScA: Staphylococcus aureus. PythU: Pvthium ultiftium.. BS: Bacillus subtilis. RhyzS: Rhizoctonia solani. per animal was 0.16mg the experimental inhibitory concentration is quite comparable. The comparability is even more apparent when you consider that loses will have been incurred during the isolation and that the metabolites were not released directly 84 into the mouth of an inquisitive predator . D_. Further Observations and Discussion. The 25-(3-hydroxybutanoate) residue and the 2a,3c»-diol array of the diaulusterols A (41) and B (42) are not commonly encountered in naturally occuring steroids. Three of the small number of steroids containing the 2a, 3e*-diol moieties include compound 54., which was recently isolated from the marine hydroid F.udendrium qlomeratum85. the potent anticancer reagent 2a- 8 6 hydroxyhippuristanol (5_5J from the gorgonian Xs_is_ hippuris and 47 azedarachol (5JL) isolated from the root bark of Melia 87 azedarach , which also possesses an unsaturated sec-butanoate moiety at the C20 position of the pregnane skeleton. Diaulusterol A (11) and diaulusterol B (12) are related to the two phytosterols, acetylpinnosterol (15.) and pinnosterol (.52), which were isolated from the red alga Laurencia pinnata75. All four steroids share structural features with the ecdysones, for example ot-ecdysone (12.). Metabolites 15. and 5_2 showed biological activity as insect moulting hormones, however, metabolites H and 12 exhibited no such activity. The relative concentrations of diaulusterols A (H) and B (12)f in the skin extracts of Diaulula sandiegnesis. varies greatly throughout the year. Minimum concentrations, sometimes so low that it was impossible to isolate the metabolites (they were 48 apparently not present at all on occasions), were observed from February to April. During the summer the concentrations gradually increased reaching a maximum in September/October. During the fall the concentrations decreased to the minimum values observed in late winter. During the winter months the nudibranchs generally hid themselves deeply in cracks and fissures and became very difficult to collect. Presumably this provides protection from the battering that exposed rocky reefs experience from winter storms. This behavior must also provide protection from predation at a time when the chemical defense reserves are at a minimum. The retreat to physically inaccessible habitats appears to coincide with the depletion in the nudibranch*s chemical arsenal and not with the approach of winter storms which start in November (the animals can be found in reasonable numbers from May through to early January). The seasonal variation in the relative concentrations of diaulusterols A (11) and B (12) is probably the result of one of two scenarios: i) The nudibranchs take shelter from the winter storms and stop feeding, and thus the chemical arsenal becomes depleted; Or ii) they take shelter as a result of a spent chemical defense? Further study is required to distinguish which of the two cause and effects operates. In addition to Diaulula sandiegensis. Aldisa sanguinea cooperi provides another example of a dorid nudibranch that sequesters a steroidal fish antifeedant, 3-oxo-chol-4-ene-24-oic acid (JLBJ . Further steroidal metabolites have been isolated from nudibranchs for which no defensive role is ascribed. For example, the highly oxygenated steroid 5_9_ from Hervie peregrina. Flabellina affinis and r.oryphella lineata89. and the steroidal 49 peroxide 5a,8a-epidioxysteroid (££.) isolated from a species of 90 Adalaria . The use of steroids for chemical defense is not limited to nudibranchs. Throughout nature steroids exhibit a wide variety of defensive biological activities, depending on oft- 91 times subtle molecular changes . In conclusion, these results, in combination with the 7 0 71 previous findings of Walker et al. and Fuhrman et al. , have shown that Diaulula sandiegensis represents another example of a dorid nudibranch possessing different skin chemistry at different collecting sites (Fuhrman et al.71 only investigated extracts isolated from digestive glands). It seems reasonable to assume that the steroids found in the British Columbian specimens, and 50 the chlorinated acetylenes found in the Californian specimens are being produced by the different dietary organisms found at the two locations. In fact, it has been demonstrated that the chlorinated acetylenes are derived directly from a dietary 92 sponge . A dietary source for the steroidal metabolites has not yet been found. 51 IY. SOFT CORALS. A.. Introduction: Alcyonacean Zoology. Soft corals are members of the phylum Coelenterata (or Cnidaria), and except for the hydras and a few other freshwater hydrozoans, all coelenterates are marine. Coelenterates are rather primitive animals that exhibit two different body forms, the medusa which is adapted to a pelagic existence, and the polyp which is adapted to a sessile benthic existence. The phylum is composed of approximately 9000 living species, and has a rich fossil record. The coelenterates show a number of distinguishing 93—97 features . They all lack organs or differentiated muscle cells, many species are brilliantly coloured, all have a radial symmetry and the gastrovascular cavity (Figure 11.) opens to the outside at just one end to form the mouth. There are three classes of coelenterates: The Hydrozoa (hydroids and hydras); the Scyphozoa (mainly jellyfish); and the Anthozoa (sea-anemones and most corals) (Figure 12.). The class Anthozoa is the largest of the three coelenterate classes and consists of over 6000 species. In total, 123 species of anthozoans 3 2 are known from Central California to Southern Alaska . The class Anthozoa can be divided into the two subclasses Zoantharia (sea anemones, stony corals) and Alcyonaria (soft corals, sea pens, sea pansies, sea fans, whip corals, pipe corals). The 98 Alcyonarians are very abundant in tropical reef environments . The order Alcyonacea (soft corals52 ) dominates almost exclusively I tube Figure 11. Structure of a typical alcyonarian . in Indo Pacific reefs, while Western Atlantic reefs are populated primarily by species of the order Gorgonacea. In the order Alcyonacea (soft corals) the internal skeleton supports the colony and consists of separate calcareous spicules integrated into the tissue (the coenenchyme). Hence, these 53 PHYLUM CNIDARIA (or COELENTERATA) CLASS. HYDROZOA ANTHOZOA SCYPHOZOA (hydroids, hydra) (sea anemones, (large jellyfish) corals) SUBCLASS ALCYONARIA ZOANTHARIA (sea anemones, stony corals) ORDER GORGONACEA STOLONIFERA HELIOPORACEA PENNATULACEA (horny corals, (organ (sea pens or sea whips and pipe corals) sea pansies) sea fans) PROTOALCYONARIA GASTRAXONACEA ALCYONACEA EAUILi ALCYONIIDAE ASTEROSPICULARIIDAE NIDALIIDAE XENIIDAE PARALCYONIIDAE NEPHTHEIDAE CEffllS GERSEMIA SPECIES RUBIFORMIS Figure 12. Phylogenetic classification of the soft coral, Gersemia rubiformis. N.B. Organisms classified according to Austin . 54 animals consist typically of a rubbery mass with a massive mushroom, or a variously lobate growth form. Alcyonacean colonies can be reproductively hermaphroditic, dioceous (one colony male, another female), or asexual. However, in general the Alcyonaceans are asexual and the single polyps that are produced and released by sexual reproduction, attaches and by asexual budding becomes the parent of all other members of the colony. Alcyonaceans feed by catching prey in the tentacles surrounding the oral cavity. The presence of a mouth and a digestive cavity permits the use of a wide range of food sizes, which includes zooplankton and fine particulate matter. Upon capture, prey is paralyzed by stinging structures called nematocysts and carried to the mouth. Nematocysts are a unique structure of all coelenterates. They are contained in specialised cells known as cnidoblasts which are located throughout the epidermis, particularly in the tentacles. £. Symbiotic Associations Between Alcyonarians and Algae f zooxantheiiae): implications For Tiie. Origin oL Isolated Metabolites. Many coelenterates, particularly those living in shallow tropical seas (eg., stony and soft corals, gorgonians and sea 99 anemones), contain symbiotic algae known as zooxanthellae . The algae are classified as dinoflagellates100 and were assigned to the single species, Zooxanthella microadriatica101. This single 102 species designation is now in question . It has been proposed 55 from studies of coelenterate zooxanthellae associations that the photosynthetic algae provide substantive levels of nutrition for 103 the host . The host, in turn, provides essential elements for autotrophic metabolism with the zooxanthellae augmenting the recycling of nutrients ("waste products") 93,103Cf<^'104. It would appear that the very existence of coral reefs rests on the photosynthetic activity of the unicellular algae and their 1 05 association with coelenterates . In order to maintain the symbiotic association, most coelenterates reside in the upper photic zone and coral reefs are restricted to shallow, clear waters. Soft corals (order Alcyonacea), invariably tropical species containing algal symbionts, have been a rich source of interesting terpenoid natural products (other Alcyonarians have 22 been a rich source of prostanoids as well as terpen.es) . The very existence of the symbiotic association obscures the origin of the substances isolated from zooxanthellae bearing- coelenterates106. It has been established that the sterol composition of such animals reflects both a dietary accumulation and a contribution from their symbiotic zooxanthellae. Zooxanthellae-bearing coelenterates show the presence of several unique, highly alkylated sterols such as the cyclopropane containing sterols gorgosterol (£1)107 and 23-demethylgorgosterol 108 (£2) . However, the position and degree of alkylation appear dependent on whether the algae live symbiotically or alone10 2a_(^. Direct evidence in the literature indicates that both gorgonians and soft corals are capable of independen56 t terpene synthesis. 61 62 rC02Me 64 63 Pukalide (£2.) was isolated from Lophoaoraia rigida109'11 0 and L_. 111 alba. . Pacifiaoroia pulchra e^ilis. yielded furanodiene (£1)112. These species are tropical gorgonians which lack algal symbionts. Other examples are known109'111'113. Pukalide (£2.) and furnodiene (£A) have also been isolated from the soft corals, Sinularia abrupta114. and an Efflatounaria species115, respectively, which possess symbionts. Further to this, measurements of the 13C/12C ratios of individual metabolites has determined that terpenes can only have been produced by the host coral or gorgonian102^, and these conclusions have been confirmed by a series of incorporation studies with labelled precursors116. 57 £. Why The Interest: The. Biological and. Ecological Significance OL Some Coelenterate Metabolites. The members of the class Scyphozoa (jelly fish, etc.) which utilize nematocysts to deter predators have, to date, provided no interesting natural products. Similarly, the physically protected hard corals have yielded little of interest to the marine natural products chemist. The majority of interesting coelenterate metabolites have been isolated from soft corals and gorgonians, 117 with fewer examples from sea pens (order Pennatulacea) , 118 zoanthids (sea anemones, etc..) and species of the order 119 Stolonifera . The tissues of many of these exposed coral reef invertebrates are toxic or show antifeedant properties to 120 121 fish ' . Soft corals and gorgonians in particular show very 120 122 little evidence of predation . Only a few gastropods feed on 123 these animals. An example of a compound that is toxic to fish is provided by crassolide (65). isolated from the soft coral, LPbpphyturo crassum124. OAc 65 In addition to predatory defensives it appears that soft coral metabolites can be released into the water column 125 surrounding a colony to inhibit the growth of competitors . 58 67 126 Flexibilide (££), from Sinularia flexibilis . and the diterpene 127 67. from Lobophytum pauciflorum-1"*' . can kill or retard the growth 128 of the hard coral Acropora formosa . Sessile reef coelenterates need to ensure that they are not invaded by microorganisms, larvae and/or algae. A chemical defense strategy is once again apparently employed to handle the potential problem. Activity against marine unicellular algae is 123 widespread . Further to this, a sea pen (order Pennatulacea), a Stylatula species, yielded stylatulide (£8J which is toxic to 129 copepod larvae . The growth of diatoms is inhibited by the muricins 1-4 (69-72). These saponins appear to contribute to reduced fouling of the gorgonian Muricea fructicosa by epiphytes and epizoa130. 59 Me C9H19 1 2 NH(CH2)3N(CH2)4NMe2 69 R = R =Ac 70 R1 = C0Prn, R2=Ac 71 R1 = Ac, R2 =C0Prn 73 72 R' = R2 = C0Prn Terpenoids from soft corals exhibit other diverse biodynamic 131 132 properties. Physiological , ichthyotoxic and 133 antineoplastic activities have all been attributed to soft coral metabolites. The in vivo antimicrobial activity of the spermidine derivative 22, isolated from a Sinularia species, is of note134. Alcyonarians in general appear to be capable of £e_ novo terpene synthesis. Many species of this subclass possess no skeletal protection, and hence their survival appears dependent on alternative strategies. These strategies are based certainly, in part, on the production of noxious and toxic chemicals. The wide distribution of biologically and ecologically active metabolites associated with these animals make them a prime target for marine natural products chemists. 60 EL. chemistry pJL soft corals and Alcyonarians In. General. Soft corals have been a rich source of novel terpenoid metabolites. A full review of the literature far exceeds the scope of this thesis, and is quite unnecessary. An excellent 9 8 review by Tursch et al. covers the literature on coelenterates up to 1978, and diterpenes, in particular cembranoid metabolites, which are widely distributed in soft corals, were reviewed by Fenical135 in 1978, and Weinheimer et al.136 in 1979. The literature is further reviewed and updated in Faulkner's review 22 papers . However, it is appropiate that a brief overview of the literature since Faulkner's last review paper covering work published up to and including July 1985, excluding steroidal 137 metabolites , be included in this thesis. The following review covers the period from July 1985 up to the citations reported in CA Selects, issue 9, 1987, and/or Chemical Abstracts, No. 20, May 18th, 1987. Many sesquiterpenes have been isolated from soft corals. The majority of sesquiterpenes isolated from marine coelenterates exist as the optical antipode of the form found, when known, in terrestrial or marine plants. A number of sesquiterpenes have been reported from soft corals over the past two years. Two new 138 139 capnellane sesquiterpenes, 74 and 75 , were isolated from the Chinese soft coral, Capnella imbricata. Extracts of the Australian soft coral, Lemnalia cervicornis yielded eleven calamenane sesquiterpenes, including compound 7JL and two non- 140 aromatic minor metabolites 22 an6d1 78 . Studies on several Australian Xenia species resulted in the isolation of the new sesquiterpenes 79-81 . The Mediterranean alcyonacean, Alcyonium corallcides. yielded the two metabolites (+)-corlloidin-A (82) and (-)-coralloidin-B (JL2.) , where £2 is the first naturally 142 occurring 5,6-dehydroeudesmane 74 75 62 H -ChUOAc 82 83 Diterpenes are the most frequently encountered metabolites of soft corals, and of these cembranoids are probably the most common. Denticulatolide (M) is an ichthyotoxic peroxide- containing cembranolide from the soft coral, Lobophytum 143 denticulatum . Studies of Australian soft corals of the genus Efflatounaria revealed the presence of two isomeric 7,8- 144 epoxycembra-3,ll,15-trien-16,2-olides, £5_ and 86 Coralloidolides A (£2) and B (M) r the first and rather unusual cembranoids from a Mediterranean organism, were isolated from the 145 soft coral Acyonium coralloides . Extraction of the soft coral, Sinularia mayi. resulted in the isolation of a neocembranoid metabolite, 13-hydroxyneocembrene (&9J146. A Pacific soft coral of the genus Sinularia. collected in Palau, Western Caroline Islands, gave £Q. and three other new norcembranoid metabolites147. Ten new diterpenoids, including minabein-10 (£1) , 148 possessing the known briarein skeleton have been isolated from a species of Minabea collected in the Eastern Caroline 149 Islands . This is the first report of the occurrence of this class of compounds in soft corals. Briareins are typically 22 associated with gorgonians and sea pens (order Pennatulacea) 63 Coll et al. have looked at a series of soft corals of the genus Lobophvtum and noted the co-occurrence of different diterpene skeletons in the same colony. The two diterpenes 22. and 93 which might be considered as a prenylated sesquiterpene cubebol and a prenylated germacrene, respectively, were found to co-occur with known cembranoid diterpenes in two seperate species of Lobophytum. The new bicyclic diterpene 2A was isolated from another species of Lobophytum along with a known cembranoid. 87 88 64 90 91 94 Species of the family Xeniidae had until recently yielded only xenicane and xeniaphyllane diterpenes such as xenicin f 95) 150 from Xenia elongata and xeniaphyllenol (2A) from macraspiculata151. Recently, a species of Cespitularia of the family Xeniidae yielded 4,5-deoxyneodolabelline (JLD / which can be related to the dolabellane skeleton by migration of a methyl from CI to Cll152. The two cyclized cembranes 2JL and 22. were also isolated. A xenia species from Okinawan waters yielded two 65 pairs of diasteriomeric diterpenes. One pair consisted of the perhydroazulene derivatives hydratoxeniolone (ULQ.) and hydratoisoxeniolone (101). The putative biogenetic precursor 153 germacrexeniolone (1£2) was also noted. 96 97 95 99 100 101 66 HO 103 104 A number of steroidal metabolites have been isolated from 137 soft corals , but only two unique metabolites will be mentioned here. A Capnella species contained the norpregane C20 154 derivatives, 103 and 104 . Naturally occuring pregnane-derived steroids, particularly from the marine environment, are rare. 105 106 Finally, two very unusual metabolites, methyl sartortuoate 155 156 f105) and methyl isosartortuoate (106) f have been isolated from the soft coral Sarcophvton tartuosum collected in the South China Sea. They both represent unprecedented tetracyclic tetraterpenoids with unknown biogenetic origins. However, since diterpenes of the cembrene class are commonly found in soft 67 corals, a plausible biogenesis would involve generation of the cyclohexene ring and hence the required carbon skeleton by a Diels-Alder coupling of two cembrenes. Other orders of the subclass Alcyonaria, particularly gorgonians, also continue to yield interesting metabolites. The anti-inflammatory diterpenoid kallolide A (107) possessing the rare pseudopterane skeleton, was isolated from Pseudopterogorgia 157 kallos . A new hydroquinone glycoside, moritoside (108)f which inhibits the cell division of fertilized starfish eggs, was 158 isolated from a gorgonian of the genus Euplexaura . A marine sea whip, Pseudopterogorgia elisabethae. yielded pseudopterosin C (109), which represents a new class of anti-inflammatory and 159 analgesic diterpene pentosides . Species of the order Stolonifera continue to yield prostanoids160, such as the antitumor agents bromovulone (110) and iodovulone (111) isolated from Clavularia viridis161. Tubipora musica yielded tubipofuran (112)r which is ichthyotoxic toward a killifish, Orizias latipes . Clavularia koellikeri yielded five kericembrenolides related to metabolite 113. which all inhibited the growth of B-16 3 63 melanoma cells . The stolonifer, Sarcodictyon roseum, contained 164 sarcodictyenone (114) , which represents the first example of a chiral, optically active prenyl derivative of a reduced benzoquinone. Finally, the sea pansy (order Pennatulacea), Renilla reniformis contained the metabolite 115 which inhibits the settlement of larvae of the barnacle Balanus amphitrite165 68 69 R. Review Of. the. Chemistry Ol. the. Family Nephtheidae. The chemistry of the soft coral, Gersemia rubiformisP a species of the family Nephtheidae, is to be discussed in this thesis. Hence, a review of the chemistry of this family is in order. The family Nephtheidea comprises many genera of which Lemnflliflf Paralemnalia, Capnella. Litophvton and Nephthea have received considerable attention from organic chemists. In the first three genera, only sesquiterpenes have been reported while the latter two genera have afforded mainly cembranoid 22 98 diterpenes ' . The sesquiterpenes observed include the simple terrestrial sesquiterpene germacrene-C (116) isolated from Nephthea chabrolii together with the guaiane-derived alcohol HI166. Two new metabolites, eudesma-4,7(ll)-dien-8B-ol (1181 and eudesma-4,7(11)-diene-8-one (119)f were isolated from an 45a unidentified species of Nephthea . Numerous sesquiterpenes bearing the capnellane carbon skeleton, but differing in the level of oxygenation and unsaturation, have been isolated from 22 98 soft corals of the genus Capnella ' . Examples include metabolites H138, 25139, 12£ and 121167 which were all isolated from Capnella imbricata. 116 117 118 70 119 ! H OH 121 120 Soft corals of the genera Lemnalia and Paralemnalia have been characterised by the exclusive production of sesquiterpenes. The majority of sesquiterpenes isolated from a number of different species have been based on a nardosinane skeleton. An example is provided by lemnacarnol (122) isolated from L. carnosa168. L. africana has yielded a wide variety of sesquiterpenes which includes the germacrene alcohols 123 and 124169, the racemic diol, 125_166, africanol (126J170 (which was also noted in extracts of L. nitida171), a 17 2 norsesquiterpene 127 and the two sesquiterpenes 128 and 172 129 . Extracts of h.. africana have also yielded the two unusual compounds the monoacetate 130 and the diacetate 131. that appear to be derived from the nardosinane skeleton (lemnalane skeleton) by ring expansion. These two sesquiterpenes were found 173 to co-occur with the eremophilane derivative 132 71 122 R=OH 123 124 138 R=H ^OCHO 125 126 127 72 Lemnacarnol (122) was found to co-occur with 7-epi-lemnalactone (ill)171. Lemnalol (134) was isolated from L. 174 tenuis and L.. cervicornis yielded metabolites 7JL, 22 and 140 78 . The discovery of two aristolane derivatives 135 and 136 in L. humesi could indicate that the nardosinane derivatives are derived from precursors having a cyclopropane ring175. 2-Deoxy-12-oxolemnacarnol (137) was found to co- occur with 2-deoxylemnacarnol (US.) in Paralemnalia 171 thyroides , the presence of 138 was also noted in Lemnalia af rieana and L.. iaexis.171. Extracts of Paralemnalia diaitiformis resulted in the isolation of both 133 and 176 lemnalactone (139) . A specimen of p_. thyroides from the Townsville area was the source of two sesquiterpenes, 140 and 177 141. together with 2-deoxylemncarnol (138) . Specimens of E« thyroides from Palau contained, along with 2- deoxylemnacarnol (138), the two new norsesquiterpenes 142 178 and 143 . It is interesting to note that the carbon skeleton of lemnacarnol, lemnalactone and their derivatives is antipodal to that of known nardosinane sesquiterpenes isolated 98 from terrestrial sources Y OH 133 134 135 73 136 137 139 142 143 Diterpenes are the most frequently encountered Nephtheidae metabolites and again they are dominated by cembranoid derived compounds. For example, Schmitz and co-workers described nephthenol f144) and epoxynephthenol acetate (145), as 179 metabolites of a Nephthea species from Enewetrak . Metabolite 144 was later found in extracts from Litophyton flrbpreum and L. 74 180 viridis . 2-Hydroxynephthenol (146) was also noted from L. 181 viridis . A large series of "simple" cembranoids (144 and 147- 182 166 155) have been isolated from H. brassica . H. chabrolii and 183 184 J LO, unknown species of Nephthea '- *t Further structurally novel diterpenes include the two xeniaphyllanes 156 and 157. which were 185 isolated from H. chabrolii 144 145 146 166,182,184 147 166,182,183 148 75 182 153 156 R="V 157 R=X' Steroidal related metabolites isolated from the family, Nephtheidea, include the two pregnane derivatives, 158 and 159. which together with the two steroidal metabolites, 160 and 161. were obtained from the cold water species, Gersemia 186 rubiformis . An unidentified species, reportedly a Nephthea species, contained the diol 162 187 158 159 76 OH 0 160 R = H HO 161 R=OAc 162 Very few aromatic compounds have been isolated from soft corals. However, the familiar 5-methyl-2-tetraprenyl-l,4- benzoquinone and the corresponding benzopyran, 164. were 1 88 isolated from a species of the genus Nephthea . where they may play a role as symbiont attractants. 163 164 The chemistry of just a few tropical and one cold water species of the family Nephtheidea have been studied. However, numerous secondary metabolites have been uncovered possessing diverse structural forms. This family by itself provides a prime target for marine natural products chemists. 77 Y. SECONDARY METABOLITES ££0J1 THE. SOFT CQBAL., Gersemia rubjfprmjs (Ehrenbera. 1834) , AHE THE. DKNnRONOTOTD NUDIBRANCHi Toauina tetraauetra (Pallas. 1788). A.. Introduction. (i). Gersemia rubiformis (Ehrenbera. 18JL4) . Soft corals and gorgonians, two groups of coelenterates that belong to the orders Alcyonacea and Gorgonacea respectively, are op very abundant in tropical reef environments . Indo Pacific reefs tend to be dominated almost exclusively by soft corals, while western Atlantic reefs are populated primarily by gorgonians. The abundance and ease of collection of these invertebrates has facilitated an extensive investigation of their secondary 22 98 135 136 metabolism Sesquiterpenes and diterpenes, having a wide variety of both well known and rare carbon skeletons, are the most common soft coral and gorgonian metabolites. In contrast to tropical waters the cold temperate waters of British Columbia harbour only one reported species of intertidal 32 and subtidal soft coral, Gersemia rubiformis . G_. rubiformis. a soft coral of the family Nephtheidae (Figure 12.), has a documented distribution range in the Pacific that extends from Japan north to the Bering Sea and down along the west coast of North America from Alaska to Northern California (north of Point Arena). £. rubiformis is also reported from the northern 32 Atlantic . The species was probably first identified by 3 2 Ehrenberg in 1834 and is known locally as the "sea strawberry". 78 The animal can obtain a height of up to 15cm and consists of a branching mass with polyps having eight tentacles. Colour varies from pale orange to red. When the tentacles are expanded, the tips of the polyps and the whole colony takes on a whitish hue. £. rubiformis is found attached to rocks at the very lowest tides and subtidally to a depth of 15 meters in current swept 72 189 surge channels and other exposed areas ' 1 86 Kingston et al. previously reported the isolation of the 186a 186t3 rare C2^ pregnane steroids I58 and 159 , and the two unusual C2g steroids 12/j-hydroxy-24-norcholesta-l,4,22-triene- 3-one (JL£Q.) and its acetate 161186c,d from specimens of Gersemia rubiformis collected off the Newfoundland coastline. Until now, this work probably represented the first and only study on a cold temperate or cold water species of soft coral. No in vitro antimicrobial or other biological activities were reported for any of these steroidal metabolites. 190 Our routine bioactivity testing program revealed that extracts of Gersemia rubiformis. collected off the British Columbian coastline, exhibited potent in vitro antimicrobial activity. The organism competes extremely well in a very crowded environment and is very clean and unfouled. G_. rubiformis has, as far we know, just two predators, the dendronotoid nudibranch, 122a Toquina tetraquetra in the Pacific and the prosobranch gastropod, Calliostoma occidentale (Mighels and Adams, 1842) in the Atlantic12213. Considering just how blatently G_. rubiformis' advertises itself and that X. tetraquetra grazes on this soft coral with such apparent culinary delight (we have observed up to a dozen on a single wall (approximat79 e area 3x7 meters) covered with G. rubiformis, each rasping a large groove into the colony) it is extremely surprising that no other organism, not even another nudibranch, is observed to feed on £. rubiformis on the west coast. Prompted by these observations, we undertook an investigation of the secondary metabolites of £. rubiformis collected off the east and west coasts of Canada. The investigation was initiated in August 1985. Gersemia rubiformis was collected on several exposed rocky reefs located within the Gordon Group of Islands, Port Hardy, British Columbia, and at Admiral's Cove, Cape Broyle and Bay Bulls, Newfoundland (south of St. Johns). Once a colony of £. rubiformis had been located it was readily collected by hand using SCUBA. The British Columbian animals were collected on four occasions in the spring, late summer and early fall. The Newfoundland animals were collected on a number of occasions between October 1986 and January 1987. Although found in an assortment of habitats, G_. rubiformis was most abundant on exposed cliff faces at a depth of 10-15 meters and also in exposed surge channels at depths of 1-3 meters. (ii). Toauina tetraquetra (Pallas, 1788). The cold temperate waters of North America are inhabited by more than one hundred and thirty species of subtidal and 32 intertidal nudibranchs . The skin extracts of nudibranchs have 22 been found to contain interesting secondary metabolites . The majority of the compounds isolated to date have been sesquiterpenoids or diterpenoids. It is found that many of the 80 1 QT metabolites are of a dietary origin . A variation in skin chemistry from one collecting site to the next, reflecting a change in the nudibranch's diet, provides strong evidence for a dietary origin23'420'191. It has been demonstrated that nudibranchs can sequester secondary metabolites, sometimes 2a 42c 19 2 selectively, from dietary sponges ' , bryozoans , soft corals453 and hydroids193. Toguina tetraquetra. suborder Dendronotoida, family Tritoniidae (Figure 1.) feeds primarily on two coelenterates, the cold water soft coral, Gersemia rubiformis (order 122a 72 189 Alcyonacea) ' ' and the sea pen, PtUPSarcus aurneyi 1 pq (Gray, 1860) (order Pennatulacea) . Tochuina tetraquetra is found in the north Pacific from central Japan north to the Bering Sea and down along the west coast of North America from Alaska to Southern California (south 32 of Point Conception to the Mexican border) . The animal 3 2 was probably first identified by Pallas in 1788 , and is known locally as the "orange peel nudibranch". This striking dendronotoid is found at depths from 3-100 meters in rocky areas when feeding on Gersemia rubiformis and flat sandy-mud bottoms when feeding on the sea pen Ptilosarcus aurneyi. I. tetraquetra is one of the worlds largest nudibranchs, reaching 30cm in length. The body is a deep orange-yellow covered with tubercles tipped with white and the foot is salmon pink to yellow. A white band marks the margin of the dorsum and branchial plumes form an irregular series of low, white tufts along the undulating body margins. A point of interest is that these molluscs were eaten by Aleutian indians raw or cooked in the Kuril Islands, U.S.S.R., 81 where it is known as "Tochni". It may also have been consumed by 3 0 189 North American west coast indians in times of hardship Specimens of Toquina tetraquetra were collected throughout the year by hand using SCUBA at depths of 3-15 meters from two locations. The first from several exposed rocky surge channels located within the Gordon Group of Islands, Port Hardy, British Columbia, where they were found feeding on or sitting nearby the soft coral, Gersemia rubiformis. The second collections were made on sandy bottoms located within the Deer Group of Islands, Bamfield, British Columbia, where the sea pen Ptilosarcus aurneyi is commonly found. The investigation of the skin extracts of Toquina tetraquetra was prompted by a series of observations. Firstly, the nudibranch grazes with apparent "delight" on the soft coral, Gersemia rubiformis. for which no other predator is known on the west coast of North America. Observations indicate that this nudibranch is so specialised that the animal would preferably starve to death in the absence of £. rubiformis even when offered a variety of other coelenterates (the sea pen, Ptilosarcus 122a gurneyi was not offered) . Hence, the soft coral may possess a chemical defense which this one specialised nudibranch has learned to ignore and quite probably modify for its own defensive purposes. In addition, I. tetraquetra has no known natural predators. Also the animal's size, relative abundance and "totally excessive" mucus covering makes the isolation of secondary metabolites a relatively undemanding task. Encouraged by these observations an investigation of the skin extracts of I. 82 tetraquetra was initiated in November 1986. £. Isolation Q_f Th_e_ Metabolites. (i). Gersemia rubiformis (Ehrenbera. 1834). (a). British Columbian specimens: Freshly collected specimens were immediately immersed in methanol, homogenised and allowed to extract at room temperature. The extraction solvent was filtered off, concentrated in vacuo and partitioned between brine and organics. Fractionation of the organic soluble material through a complex combination of flash, Sephadex LH-20, preparative TLC, radial TLC, reverse and normal phase HPLC chromatographies produced pure samples of two new 194 sesquiterpenes 165 and 166 , and eleven new diterpenes, possessing pseudopterane f167-169)l90d. cembrane f170- 175)190d'194 and the new gersolane (176)195 carbon skeletons (the structure of the eleventh remains unresolved). In addition, a degraded diterpene (177) was isolated. 2 165 166 83 84 175 176 177 (b). Newfoundland specimens: Freshly collected specimens were immediately frozen and stored at -5°C. After defrosting, the animals were homogenised in methanol and allowed to extract at room temperature. The extraction solvent was filtered off, concentrated in vacuo and the residue was partitioned between brine and ethyl acetate. Fractionation of the organic extract by a combination of flash, radial TLC, normal and reverse phase HPLC chromatographies produced a pure sample of sesquiterpene 178. The presence of the steroidal metabolites 15B-161186 was noted by 300MHz XH NMR analysis on crude fractions. These metabolites were never fully purified. 15 178 85 (ii). Toauina tetraquetra fPallas, 3788). Freshly collected animals were immediately immersed in methanol and allowed to extract at room temperature. The methanol was decanted, concentrated in vacuo and partitioned between brine and ethyl acetate. Fractionation of the organic soluble material by a combination of preparative TLC and reverse phase HPLC chromatographies yielded two unrelated sets of pure secondary metabolites. (a) . Port Hardy animals: From Port Hardy animals pure samples of the two 194 sesquiterpenes 165 and 166 and two cembranes, the diterpene 17Q190d,194 an(3 tjie previously reported diterpene r, 109, 111, 114,194 Li-j 63 i were obtained. (b) . Bamfield animals: From Bamfield animals pure samples of two diterpenes 148 possessing the previously reported briarein carbon skeleton , the known metabolite 179 and the previously unreported butanoate 194 analogue 180 , were obtained. 86 C Structural Elucidation ajid. Related Studies. (i). sesquiterpenes 165 and 166; Tochuinyl acetate (165) was obtained as a clear oil that gave a parent ion at a m/z of 260.1777 daltons in the EIMS appropriate for a molecular formula of ci7H24°2 (rec3u^re^ m/z *s 260.1777 daltons) which required 6 units of unsaturation. A carbonyl band at 1734cm"1 in the IR, in combination with a methyl singlet resonance at 1.94ppm in the "^H NMR spectrum (Figure 13.) 13 and a carbonyl resonance at 171.2ppm in the C NMR spectrum (Figure 14.), identified an acetate residue. The carbinol 7 6 methylene (a short range 2D-HETCOR NMR experiment gave a proton 13 count, see Figure 15.) resonance at 70.6ppm in the C NMR spectrum of 1£5_ suggested that the acetate was an ester of a primary alcohol. The additional singlets at 1.13, 1.33 and 2.30ppm in the NMR spectrum suggested that the remaining fifteen carbon atoms of tochuinyl acetate (165) constituted a sesquiterpene fragment. One degree of unsaturation was accounted for by the acetate carbonyl. Four more were readily assigned to the 4-methylphenyl moiety indicated by the NMR methyl resonance at 2.30(s, 3H)ppm and the two aromatic resonances at 7.08(d, J=8.1Hz, 2H) and 7.22(d, J=8.lHz, 2H)ppm. These resonances were correlated in the 13 short range 2D-HETCOR NMR spectrum (Figure 15.) with the C NMR resonances at 20.8(CH3), 128.6(2xCH) and 126.7(2xCH)ppm, respectively. The resonances at 135.3(C) and 142.9(C)ppm were assigned to the remaining two aromatic carbons. 87 o co- Details: 187mg 5mm tube 7hours 42-1mins Figure 15. Contour plot of 2D-HETC0R NMR spectrum of tochuinyl acetate (1£5_). I I II | t (1 I] I I I I | I I ! I | 1 II I | I I I I |I I I Ij 1 ii I | i I ( I | I I I I | I I 1 I j I ! I I | I I I I j I I I I | I i I PPM F1 (PPM) Comparison of the partial H and C NMR data of the model metabolite 181196 with those for 165 (see Tables 8. and 9.), supports the assignment. The 4-methylphenyl moiety was further confirmed by a MS fragment ion at an m/z of 91 daltons (C^H7) which is indicative of a tropylium ion derived from an alkyl substituted 197 benzene ring 181 Table 8. 1H NMR data comparison of 165 and 181• Chemical shift, ppm. Carbon # 1£5_ l£la 1,5 7.22(d, J=8.1Hz, 2H) 7.26(d, J=8.3Hz, 2H)Z 2,4 7.08(d, J=8.1HZ, 2H) 7.20(d, J=8.3Hz, 2H)Z 12 2.30(s, 3H) 2.33(s, 3H) 13 1.33(s, 3H) 1.42(s, 3H) reference 196, 360MHz, solvent not reported. Assignments within a column may be interchanged. The short range 2D-HETC0R NMR spectrum correlated the NMR resonances at 3.36(d, J=ll.lHz, IH) and 3.59(d, J=ll.lHz, lH)ppm with the 13C NMR carbinol methylene resonance at 70.6ppm. The 91 13 Table 9. C NMR data comparison of 165 and 181. Chemical shift, ppm. Carbon # 165 1,5 126.7 2,4 128.6 3 135.3 6 142.9 144.44 7 49.8 8 37.6 40.46 12 20.8 20.97 13 25.0 29.71 reference 196, 90MHz, solvent not given. No chemical shifts reported for "?" (C3 and C7). results of two decoupling experiments were consistent with the correlation. Irradiation of the signal at 3.36ppm collapsed the doublet at 3.59ppm to a singlet and irradiation at 3.59ppm similarly collapsed the doublet at 3.36ppm to a singlet. Hence, the resonances at 3.36 and 3.59ppm were assigned to the geminal carbinol protons of a tertiary acetoxymethyl group. Since the molecular formula of tochuinyl acetate f165) required six units of unsaturation and only five could be accounted for by the structural fragments so far ascribed (acetate and 4-methylphenyl moieties), it was assumed that metabolite 165 possessed an additional carbocyclic ring. The short range 2D-HETCOR NMR spectrum indicated that the structure had to further accomodate: two additional quaternary tetra-alkyl substituted carbons (47.4(C) and 49.8(C)ppm); three additional 92 methylene carbons (20.2(CH2)f 34.8(CH2) and 37.6(CH2)ppm); and two additional methyl carbons (19.SfCH^) and 25.0(CHj)ppm). The three methylene carbons were correlated to the NMR multiplets resonating at 1.58(m, IE), 1.75-1.88(m, 4H) and 2.48(m, lH)ppm. A series of decoupling experiments established that these three sets of multiplets were highly spin-coupled. Irradiation of the multiplet at 2.48ppm sharpened the multiplets at 1.58 and 1.75-1.88ppm. Similarly irradiation at 1.58ppm sharpened the multiplets at 1.75-1.88 and 2.48ppm. 182 Two biogenetically sound sesquiterpenes, 165 and 182. accomodated all the structural fragments thus far ascribed to 1 13 tochuinyl acetate. Comparison of the partial H and C NMR chemical shift assignments for 181196 with 165 (Tables 8. and 9.), suggested that the correct gross structure was represented by 165. However, more definitive proof was sought. A gated decoupled 13C NMR spectrum (Figure 16.) (which allows long range couplings to be observed) distinguished between the two structures, 165 and 182. In Figure 16., the carbinol carbon at 70.6ppm appears as a triplet (large one bond C-H coupling) of quartets (small three bond C-H coupling to the methyl protons resonating at either 1.33 or 1.13ppm). The methyl resonating at 93 s 13 J. 15 OAc 165 • 1 I ' IB 71 IS 1« » 11 160 1 40 1 '2 0 10-'0- 80 60 40 20 PPM 0 Figure 16. Gated decoupled 13C NMR spectrum of tochuinyl acetate (liLS.) . 25.0ppm (correlated to the methyl singlet at 1.33ppm in H NMR spectrum) appears as a quartet (large one bond coupling) of triplets (small three bond coupling). The results were consistent with the gross structure 165 and not 182. Further consistencies were noted. The C12 resonance at 20.8ppm appeared as a quartet of triplets, the C17 methyl at 20.9ppra appeared as a straight quartet. With the additional knowledge that aromatic two (and J - four) bond C-H coupling is small relative to three bond ( CH=1 3 198 2Hz, JCH=7-12Hz) the experiment also allowed for the chemical shift assignment of the two sets of aromatic methine carbons. The CI and C5 resonance at 126.7ppm appeared as a doublet (one bond coupling) of doublets (three bond coupling to H5 and Hi respectively, as they are not magnetically equivalent), and the C2 and C4 resonance at 128.6ppm appeared as a doublet (one bond coupling) of quintets (three bond coupling to C12 methyl protons and the H4 and H2 protons, respectively). Structure 165. based on a cuparane sesquiterpene skeleton, effectively accounted for all the spectral features of tochuinyl 199 acetate. Two SINEPT experiments, optimised for polarisation transfer through 7Hz coupling, connected the structural fragments of 165 together and in so doing unambiguously confirmed the proposed structural constitution. Irradiation of the methyl resonance at 1.13ppm (Mel5) showed polarisation transfer to carbon resonances at 34.8(C10), 47.4(C7 or Cll), 49.8(C7 or Cll) and 70.6(C14)ppm (Figure 17a.). Thus, the five membered ring and the acetate moiety were connected through Cll. Irradiation of the methyl resonance at 1.33ppm (Mel3) showed transfer to carbon 95 12^3 OAc Details for each: 187mg llhours 21 2mins 165 (a). Irradiated at M3ppm (Me15) (b). Irradiated at 133ppm (Me13) - 1 1 1 1 I i i i i ]• i i iI i i i i | i i i i | i i i r j i i r r j • i \ r i | i t1" i | r"i • ii rI| i i i• i i i• | i i i i | i i i i [ i i i i | i i i i | i i i i || ii i i i i i I i i i i | i i i i I 200 180 160 140 120 100 80 60 40 20 PPM Figure 17. SINEPT NMR spectra of tochuinyl acetate (UL5.) resonances at 37.6(C8), 47.4(C7 or Cll), 49.8(C7 or Cll) and 142.9(C6)ppm (Figure 17b.). Thus, the five membered ring and the 4-methylphenyl moiety were connected through the C7-C6 bond. The experiment also allowed for further chemical shift assignments. The 13C NMR resonances at 19.5, 25.0, 142.9ppm and hence 135.3ppm could be assigned to the C15, C13, C6 and C3 carbons, respectively, and the resonances at 34.8, 37.6 and hence 20.2ppm to the CIO, C8 and C9 carbons, respectively. A series of difference NOE experiments established that the relative stereochemistry about the cyclopentane ring was as shown in 165. Irradiation of the methyl resonance at 1.13ppm (Mel5) induced NOE's into the methyl resonance at 1.33ppm (Mel3), the geminal carbinol protons at 3.36 and 3.59ppm (H14 protons) and into the aromatic protons at 7.22ppm (Hi and H5). Irradiation of the methyl resonance at 1.33ppm (Mel3) induced NOE's into the methyl resonance at 1.13ppm (Mel5) and the aromatic resonance at 7.22ppm (HI and H5). The reverse irradiations resulted in the expected NOE's. Inducement of an NOE into the aromatic protons at 7.08ppm (H2 and H4), on irradiation of the methyl resonating at 2.30ppm (Mel2), provided a further consistency. One small ambiguity needs to be explained. Decoupling experiments indicated that the H8b proton, resonating at 2.48ppm in the *H NMR spectrum of 165. was coupled to the HlOa proton resonating at 1.58ppm. W-coupling can readily account for this observation. Molecular models indicate that the conformation required for W-coupling is quite strained. Since an NOE was observed between the H8b (2.48ppm) and the aromatic HI and H5 protons (7.22ppm), the H8b proton has to possess an ct- 97 configuration and hence, the HlOa must also be a to allow for the planar configuration required for W-coupling. The deshielded chemical shift of the H8b proton (2.58ppm) is consistent with an a-configuration, for it would be positioned within the deshielding region established by the ring current of the aromatic ring. Similarly, the Mel3 protons, resonating at 1.33ppm, would to some extent be situated within the same deshielding region. These protons show a much stronger NOE with the aromatic protons at 7.22ppm (HI and H5) than do the methyl protons at 1.13ppm (Mel5), which are situated outside the deshielding region. Full ^H and 13 C NMR assignments for 165 are given in Tables 10. and 11., respectively. Dihydrotochuinyl acetate (166) was also isolated as a colourless oil. It gave a parent ion at a m/z of 262.1926 daltons in the EIMS appropriate for a molecular formula of ci7H26°2 (required m/z is 262.1934 daltons) that required 5 units of unsaturation. The similarity of the MS of 165 and 1££ and a facile loss of two mass units from the parent ion of 166 suggested that it was the dihydro derivative of The *H and 13C NMR spectra of 166 (Figures 18. and 19.) were fully consistent with the proposal. Resonances assigned to the 4- methylphenyl fragment in the H and -LJC NMR spectra of 165 (Tables 10. and 11.), were replaced in the NMR spectra of dihydrotochuinyl acetate f166) with ^H resonances at 1.65(s, 3H), 2.61(m, 2H), 2.70(m, 2H), 5.39(m, IH) and 5.53(bs, lH)ppm, and 13 C resonances at 28.4(CH2), 31.9(CH2), 119.0(CH), 119.KCH), 130.5(C) and 138.5(C)ppm, which could be assigned to a l-alkyl-4- 98 Table 10. H NMR assignments for tochuinyl acetate (165) and dihydrotochuinyl acetate (166). Chemical shift, ppm. Carbon # 1£5_ 1M 1 7.22(d, J=8.1Hz, IH) 2.70(m, 2H)zl 2 7.08(d, J=8.lHz, IH) 5.39(m, lH)yl 4 7.08(d, J=8.1Hz, IH) 2.61(m, 2H)z2 5 7.22(d, J=8.1Hz, IH) 5.53(bs, lH)y2 8a 1.75-1. 88(m) 1.65-1.80(m)x 8b 2.48(m, IH) 2.20(m, IH) 9 1.75-1. 88(m) 1.42-1.55(m)x 1.65-1.80(m)x 10a 1.58(m, IH) 1.42-1.55(m) 10b 1.75-1. 88(m) 1.65-1.80(m)x 12 2.30(s, 3H) 1.65(s, 3H) 13 1.33(s, 3H) 1.07(s, 3H)W 14 3.36(d, J=ll.lHz , IH) 3.74(d, J=11.0Hz, IH) 3.59(d, J=ll.lHz , IH) 3.81(d, J=11.0Hz, IH) 15 1.13(s, 3H) 1.04(s, 3H)W 17 1.94(s, 3H) 2.02(s, 3H) Assignments within a column may be interchanged. Were appropriate interchange as complete pairs according to the second superscript 1 or 2. 99 Table 11. C NMR assignments for tochuinyl acetate (165) and dihydrotochuinyl acetate (166). Chemical shift, ppm. Carbon # 165 166 1 126.7(CH) 28.4Z 2 128.6(CH) 119.0y 3 135.3(C) 130.5X 4 128.6(CH) 31.92 5 126.7(CH) 119.ly 6 142.9(C) 138.5X 7 49.8(C)2 50.3W V 8 37.6(CH2) 37.0 9 20.2(CH2) 20.1 V 10 34.8(CH2) 35.3 11 47.4(C)2 46.9W U 12 20.8(CH3) 22.7 U 13 25.0(CH3) 22.8 14 70.6(CH2) 70.1 15 19.5(CH3) 19.5 16 171.2(C) 171.4 17 20.9(CH3) 21.0 assignments within a column may be interchanged. 100 101 rP OAc 166 o to I i i I I | I I I I | I I I i | i I I i | i I I I [ i I i i | I I I i | I l M | M M | l l l I | I I I I | I I I I | I I i I | I I I i | n i i | i m i | i i i i | 160 140 120 100 80 60 40 20 PPM 0 Figure 19. 13C NMR spectrum of dihydrotochuinyl acetate (!££). methyl-1,4-cyclohexadiene residue. Apart from these differences the two sets of spectra were virtually identical. Comparison of the partial ^-H and 13C NMR data of 1£6_ with the two models 1£3_196 and 184200 (Tables 12. and 13.) provided further confirmation and allowed for a number of chemical shift assignments. Assignment of the NMR resonances attributed to the l-alkyl-4-methyl-lf4- cyclohexadiene moiety was partially achieved by a series of decoupling experiments. Irradiation of the multiplet at 2.61ppm markedly sharpened the olefinic resonance at 5.53ppm and weakly sharpened the multiplet at 5.39ppm. Irradiation at 2.70ppm markedly sharpened the resonance at 5.39ppm and weakly sharpened the broad singlet at 5.53ppm. The proposed assignments (see Table 10.) are based on the knowledge that vicinyl coupling is larger 197 than allylic coupling in such spin systems. However, as indicated, the appropriate interchanges are also possible. Consistent with the proposed structure for dihydrotochuinyl acetate f166) were the NMR resonances (Table 10. and 11.) and an IR band (1731cm-1) appropriate for the acetate functionality. Decoupling the three sets of multiplets at 1.42-1.55, 1.65-1.80 and 2.20ppm established that they were all once again highly spin-coupled. Finally, dihydrotochuinyl acetate (1£&) was quantitatively converted to tochuinyl acetate (1£5_) with 1 13 palladium on charcoal in refluxing ethanol. Proposed H and C NMR assignments for 166 are included in Tables 10. and 11., respectively. 103 183 184 Table 12. H NMR data comparison of 166. 183 and 184. Chemical shift, ppm. Carbon # 166 184b 1 2.70(m, 2H)Z 2.57(bs) 2.50(m) 2 5.39(m, lH)y 5.40(bs, 1H)Z 5.34(m) 4 2.61(m, 2H)Z 2.57(bs) 2.50(m) 5 5.53(bs, 1H)Y 5.56(bs, 1H)Z 5.34(m) 8b 2.20(m, 1H) ? 2.19(bq, 1H) 12 1.65(s, 3H) 1.64(s, 3H) 1.63(bs, 3H) 13 1.07(s, 3H) 1.15(s, 3H) 0.98(s, 3H) reference 196, 360MHz, solvent not given. breference 200, 100MHz, CC1.. No chemical shift reported forB ?n (H8b). 4 z yassignments within a column may be interchanged. 104 Table 13. C NMR data comparison of 166. 1£2 and ISA.. Chemical shift, ppm. Carbon # 166 l£la 184b 1 28.4Z 26.45 28.6 2 119.0y 119.19 117.8 3 130.5X 130.92 130.9 4 31.9Z 32.22 31.8 5 119.lv 119.51 118.8 6 138.5X 138.47 139.2 7 50.3 52.13 49.5 8 37.0 36.61 ? 11 46.9 159.08 48.8( d) 12 22.7W 22.85 22.8 13 22. 8W 26.71 25.8 areference 196, 90MHz, solvent not given. ^reference 200, 25MHz, CDC1,. Resonances were not assigned. Not possible to distinguish between 27.4(t) or 32.8(t)ppm (C8/C9) for n z Assignments within a column may be interchanged. (ii). Pseudopterane diterpenoids. 167~169: Gersemolide (167). obtained as colourless needles, was shown by EIMS to possess a molecular formula of C20H24°4 (°bserved m/z 328.1686, requires 328.1675) having 9 units of unsaturation. The IR bands at 1699, 1648 and 1762cm"1 and resonances at 208.3, 105 197.6 and 173.7ppm in the 13C NMR spectrum of 167 (Figure 20.) could be assigned to the carbonyls of two ketones and one ester functionality, respectively, thereby accounting for all four 201 202 oxygen atoms. Eight olefinic carbon resonances in the DEPT ' (optimised for polarisation transfer through 140Hz coupling) 13 (Figure 21.) and C NMR spectra at 112.3(CH2), 116.1(CH2), 123.8(CH), 134.2(C), 138.1(C), 146.4(C), 150.0(CH) and 156.2(C)ppm indicated four carbon-carbon double bonds. Subtracting the seven sites of unsaturation required by the carbonyl and olefinic functionalities from the total of nine in the molecule, indicated that gersemolide f167) was bicyclic. Proton resonances at 7.12(bs, 1H) and 5.38(m, lH)ppm in the 1H NMR spectrum (Figure 22.), and 13C NMR resonances at 173.7(C), 150.0(CH), 134.2(C) and 79.0(CH)ppm could be assigned to an a,r- disubstituted a, ^-unsaturated y-lactone by comparison with the reported spectral data for the model compounds pseudopterolide (1£5J203, kallolides A-C (lfil, 1££ and 1£2)157 and lophodione f188)which all contain such a fragment (see Table 14.). The lactone accounts for one of the required rings in gersemolide f167). The assignment of the two proton resonances at 5.38 and 7.12ppm to the H8 and H9 lactone protons, respectively, was consistent with the results of a decoupling experiment. The H8 resonance (5.38ppm) was markedly sharpened on irradiation of the olefinic H9 resonance at 7.12ppm. Two isopropenyl groups were 13 identified from the C NMR resonances at 112.3(CH2) and 1 116.1(CH2)ppm and the H NMR resonances at 1.64(bs, 3H), 1.73(bs, 3H), 4.76(m, 2H), 5.20(bs, 1H) and 5.47(bs, lH)ppm. Again, 106 107 Details for each: 3-5mg 3hours 67mins 1 i i i i i i i i i i i i i i i i i i i i i i i i i i ' i i i i ' i 1 1 1 i 60 40 20 PPM 0 Figure 21. DEPT NMR spectra of gersemolide (167) H-- 0 0 185 186 HQ 0 187 188 compare the partial C NMR resonances of the respective carbons for metabolites 186157 and 188111 with those of 1£I in Table 14. A decoupling experiment indicated that the two proton multiplet at 4.76ppm was coupled to the broad methyl singlet at 1.64ppm, since the methyl was markedly sharpened on irradiation at 4.76ppm. Also, the methyl doublet resonating at 1.83(J=1.2Hz)ppm collapsed to a singlet on irradiation of the olefinic quartet (J=1.2Hz) at 6.33ppm. Presumably the methyl resonance at 1.83ppm and the olefinic resonance at 6.33ppm constituted part of a trisubstituted double bond. The two olefinic protons resonating at 5.20 and 5.47ppm were by inference, therefore, associated with the allylic methyl protons, resonating at 1.73ppm, of the second isopropenyl group. 110 Table 14. C NMR assignments for gersemolide (167) and 169. and spectral comparison with 186 and 188. Chemical shift, ppm. Chemical shift, ppm. Carbon # 1£2 1££ 186a Carbon # lB8b 1 41.2(CH) 43.5Z 42.1 1 41.3 2 2 44.9(CH2) 44.3 / 2 45.7 3 208.3(C)2 203.3y / 3 205.4 4 156.2(C) 150.0X / 4 144.82 5 123.8(CH) 130.8X / 5 133.4 6 197.6(C)2 193.8y / 6 190.7 7 62.4(CH) 145.3X 48.8 / 8 79.0(CH) 77.6 81.2 10 80.1 9 150.0(CH) 148.lx 147.1 11 148.4 10 134.2(C)Y 135.lx 136.8 12 134.1 V 11 22.0(CH2) 23.1 23.1 13 25.7 12 31.4(CH2) 28.4 34.5 14 30.3 13 146.4(C)Y 147.3X 148.22 15 145.5Z X y 14 112.3(CH2) 112.7 110.7 16 115.9 W V X 15 19.8(CH3) 21. 9 19.5 17 17.0 W V 16 21.9(CH3) 23.8 / 18 21.5 17 138.1(C)y 131.4X 142.7Z / X V y 18 116.1(CH2) 14.2 114.5 / W V X 19 23.1(CH3) 18.4 21.8 / 20 173.7(C) 172.4 175.6 20 173.1 cont'd. Ill Table 14. continued. a reference 157, 50MHz, CDC13, "/" means no advantage in comparing, ^reference 111, 20MHz, CDCl^, "/" means no advantage in comparing, z~vassignments within a column may be interchanged. The preceeding arguments suggested that gersemolide (167) possessed a pseudopterane carbon skeleton with the 12-membered ring accounting for the remaining site of unsaturation. At this stage of the project only a small quantity of gersemolide (167) had been isolated (<4.0mg) and the spectral information obtained thus far did not allow for an unambiguous structural proof, or for the determination of relative stereochemistry. With some effort reasonable crystals of gersemolide (167) were obtained by slow evaporation from 1:1 diethyl ether/methanol at 0°C. The complete structure of gersemolide (167) was secured via single crystal x-ray diffraction analysis, performed by G. D. Van Duyne and J. Clardy of Cornell University. A computer generated perspective drawing of the final x-ray model of gersemolide (167) is represented in Figure 23a. It should be noted that the x-ray experiment did not define the absolute configuration, so the enantiomer shown is arbitrary. While the structure in Figure 23a., was completely compatible with the spectral data, Van Duyne and Clardy indicated that there * were some troubling aspects in the molecular geometry . The C10- C11-C12-C1 region of the x-ray model was unacceptable. The Cll- C12 distance was 1.417A, much too short for a sp -sp bond, and 112 C10-C11-C12-C1 C10-C11-C12-C1 TORSION OF -37* TORSION OF +43* CM. (c). Figure 23. Computer generated perspective drawings of gersemolide f167). 113 the C10-C11-C12-C1 torsional angle was -8°, an essentially eclipsed conformation. This type of situation is not unprecedented and is usually explained by assuming a disordered structure. There were presumably at least two low energy conformations which occur in the crystal, and the resulting x-ray structure was their superposition. Van Duyne and Clardy carefully inspected various electron density syntheses, this did not reveal discrete peaks for the disordered mates, but did reveal that the electron densities of CIO and Cll were rather broad. Since the x- ray analysis did not resolve the problem, they used molecular mechanics to find the suspected minima. Two local minima on the three dimensional surface were found by varying the C10-C11-C12- Cl torsional angle. The first had a torsional angle of -37° (Figure 23b.) and the second, +43° (Figure 23c). The region around 0° represented a broad maximum. The calculated heats of formation for the minima differed by only 0.5kcal/mole, and hence, the x-ray geometry (Figure 23a.) was, as expected, the superposition of the two minima. Neither conformation allows for both the C3 and C6 ketones to be conjugated with the C4-C5 trisubstituted olefin. Figures 23b. and c, indicate that the C6 ketone is almost co-planar with the C4-C5 bond, which is in 1 3 agreement with the C NMR resonances at 208.3 (non-conjugated) and 197.6ppm (somewhat conjugated) for C3 and C6, respectively. All x-ray analysis and associated work kindly performed and interpreted by G. D. Van Duyne and J. Clardy, without whom I would be at a complete loss here. Hence, only interpreted results are reported. For supplementary material concerning this analysis see reference 190d. 114 Comparison of the *H NMR chemical shifts of 167 with those of the model compounds lfi^-lM111'157'203 and the diterpenes 3 68- 177 (to be discussed in the following pages), combined with a series of decoupling experiments allowed for a complete assignment of *H NMR resonances. Irradiation of the multiplet at 2.26ppm (HI) simplified the resonances at 2.38(dd, J=15.1,9.5Hzf H2a) and 2.53(dd, J=15.1,4.5Hz, H2b)ppm to two doublets (J=15.lHz). The HI resonance was also observed to couple with the rather shielded two proton multiplet resonating at 1.28ppm. One or both of the H12 protons (or equivalent protons (H14) in cembrane metabolites) generally lie in the upfield region from 0.4 to 1.3ppmin'157,190d'203. This presumably results from anisotropic shielding by either the nearby isopropenyl moiety or the lactone carbonyl, or by a combination of the two effects. Molecular models readily demonstrate that either effect is feasible. These arguments are consistent with the decoupling results and hence the H12 protons were assigned to the resonance at 1.28ppm. The H12 protons were subsequently shown to couple to a multiplet resonating at 2.01ppm (Hlla) and one in the region 2.32-2.45ppm (Hllb) lying underneath the H2 doublet of doublets at 2.38ppm. Irradiation at 2.0lppm (Hlla) markedly affected the geminal partner resonating in the region 2.32-2.45ppm (Hllb). These assignments were confirmed by the observation that irradiation of the lactone olefinic resonance at 7.12ppm (H9) weakly sharpened the region 2.32-2.45ppm, which is consistent with allylic coupling between the H9 and the Hllb protons. Proposed *H NMR assignments for 167 are included in Table 15. 1 3 Proposed C NMR assignments, based on comparison to model 115 Table 15. H NMR assignments and spectral comparison for gersemolide f167). 168 and 169. Chemical shift, ppm. Carbon 1£1 1LS. 1 2.26(m, 1H) 1.95-2.06(m ,1H)Z 2a 2.38(dd, J=15.1,9.5Hz, 1H) 2.31-2.49(m)z b 2.53(ddf J=15.1,4.5Hz, 1H) 2.72(dd, J=13.6,3.8Hzr 1H) 5 6.33(q, J=1.2Hz, 1H) 6.11(q, 1.6Hz, 1H) 7 3.29(d, J=0.6Hz, 1H) / 8 5.38(m, 1H) 5.93(bsf 1H) 9 7.12(bsf 1H) 7.10(bsf 1H) 11a 2.01(m, 1H) 2.31-2.49(m)Z b 2.32-2.45(m, 1H)1 12 1.28(m, 2H) 1.18(mr 1H) 1.26(mf 1H) zl 14a 5.20(bs, 1H) 4.94(m, 1H) zl b 5.47(bsf 1H) 5.07(bsf 1H) 15 1.73(bs, 3H)y i 1.73(bs, 3H) 16 1.83(d, J=1.2Hz, 3H) 2.14(d, J=1.6Hz, 3H) z2 18 4.76(mf 2H) 1.85(sr 3H) 19 1.64(bs, 3H) y2 1.93(s, 3H) cont'd. compounds, particularly kallolide B (186) , lophodione (JJL2.) and the diterpenes 168-177. to be discussed in the following pages, are included in Table 14. A4'5Z,2'17-Isogersemolide (168). a very minor metabolite 116 Table 15. continued. Chemical shift, ppm. Carbon # 169 1 2.33(m, IH) 2a 2.49-2.57(m) b 2.71(dd, J=14.0,5.7Hz, IH) 5 6.21(q, J=1.2Hzr IH) 7 / 8 5.92(d, J=0.6Hz, IH) 9 7.02(tr J=0.6Hz, IH) 11a 2.17(tdm, J=12.2,1.5Hz, IH) b 2.49-2.57(m) 12 1.26-1.41(m, 2H) 14a 4.76(bs, IH) b 4.86(q, J=0.9Hz, IH) 15 1.72(bs, 3H) 16 2.08(d, J=1.2Hz, 3H) 18 2.01(s, 3H) 19 2.06(sf 3H) 'yassignments within a column may be interchanged. Were appropriate interchange as complete groups according to the second superscript 1 or 2. 1lies underneath the H2 resonance at 2.38ppm. (<1.5mg) obtained as clear crystalline needles, was shown by EIMS to possess a molecular formula of C2oH24°4 (observed m/z 328.1684, requires 328.1675), identical to that of gersemolide f167) . A number of fragments were tentatively identified by the 117 combination of the results of a series of NOE experiments and spectral comparison of the *H NMR resonances of 168 (Figure 24.) and gersemolide (167) (see Table 15.). Induction of an NOE into the olefinic proton at 6.11(q, J=1.6Hz, lH)ppm upon irradiation of the methyl doublet (J=1.6Hz) at 2.14ppmf identified a methyl substituted ene-dione, identical with the corresponding substructure of gersemolide (167). Irradiation of the methyl resonance at 1.73(bs, 3H)ppm induced NOE's into the olefinic resonance at 4.94(m, lH)ppm and part of the multiplet region centred at 2.44ppm. Comparison with gersemolide (167) allowed for the tentative identification of an isopropenyl residue. Two methyl singlets resonating at 1.85 and 1.93ppm and the two upfield one proton multiplets at 1.18 and 1.26ppmf typical of H12 protons situated in a pseudopterane skeleton, were also observed. The pseudopterane structure 168 effectively accomodates all the identified structural fragments of A4'5Z,A7'17- isogersemolide. A series of NOE experiments verified that 168 represented the correct structure. Simultaneous irradiation of the Mel9 resonance (1.93ppm) and the HI multiplet (1.95-2.06ppm) induced NOE's into the doubly allylic lactone carbinol resonance at 5.93(bsf lH)ppm (H8, from Mel9), the H9 olefinic resonance at 7.10ppm (from Mel9), the methyl resonance at 1.85ppm (Mel8, from Mel9), the H12 multiplet at 1.18ppm (from Hi) and the Hl4b resonance at 5.07(bs, lH)ppm (from HI). Irradiation of the methyl (Mel8) resonance at 1.85ppm induced a small NOE into the olefinic proton at 6.12ppm (H5). Thus, the Mel8 and Mel9 methyls had to be located in the vicinity of both the methyl substituted 118 ene-dione and lactone moieties, i.e., at the C17 position. Irradiation at 2.14ppm (Mel6) not only induced an NOE into the H5 proton (6.11ppm), but also into the resonances at 2.72ppm (H2b) and 5.07ppm (H14b). This set of experiments demonstrated the expected connectivities about the 12-membered ring. Finally, gersemolide (167) was partially isomerised to 168 in the presence of formic acid. The gentle refluxing of reactants yielded a 30:3:20 mixture of 167. 168 and one other unidentified compound 4 5 (presumably a A ' E isomer of gersemolide), respectively, as evidenced by 300MHz *H NMR analysis. Proposed ^H NMR assignments for A4'5Z,A7,17-isogersemolide (168) are included in Table 15. A4,5E, A7,17-lsogersemolide (169). obtained as colourless needles, was shown by EIMS to possess a molecular formula of C20H24°4 (°bserved m/zr 328.1669, requires 328.1675), identical to that of both gersemolide (167) and A4'5Z, A7,17-isogersemolide (168). A number of fragments could readily be identified from the XH and 13C NMR spectra of A4,5E, A7'17-isogersemolide (1£9_) (Figures 25. and 26.). These included: a methyl substituted ene- dione (13C NMR: 203.3, 150.0, 130.8, 193.8 and 23.8ppm; XH NMR: 2.08(d, J=1.2Hz, 3H) and 6.21(q, J=1.2Hz, lH)ppm); o,y- 13 disubstituted os,^-unsaturated r-lactone ( C NMR: 77.6, 148.1, 135.1 and 172.4ppm; XH NMR: 7.02(t, J=0.6Hz, 1H) and 5.92(d, J=0.6Hz, lH)ppm); and isopropenyl (13C NMR: 147.3, 112.7 and 21.9ppm; XH NMR: 1.72(bs, 3H), 4.76(bs, 1H) and 4.86(q, J=0.9Hz, lH)ppm) residues. The NMR resonances for each of these residues were virtually identical with those assigned to the corresponding substructures in either gersemolide (1£1) and/or A4'5Z,A7'17- isogersemolide (168) (see Tables 14. and 15.). A marked 120 121 122 similarity in the H NMR spectra of 169 and 16J1, particularly in the methyl singlets resonating at 2.01 and 2.06ppm, was noted. These methyl resonances were comparable to those assigned to the Mel8 and Mel9 protons of 168. which resonated at 1.85 and 1.93ppm, respectively. The above arguments, hence suggested that 168 and 169 had the same constitution but differed in some stereochemical sense. The geometrical isomer of A4f^Z,A7,l7-isogersemolide (16J1) possessing the pseudopteranoid structure 169 effectively accomodates all the identified structural fragments. A series of NOE and spin-spin decoupling experiments verified that 169 represented the correct structure. Irradiation of the methine triplet (J=0.6Hz) at 7.02ppm (H9) collapsed the doublet (J=0.6Hz) at 5.92ppm (H8) to a singlet and sharpened the methylene Hll resonance at 2.17(tdm, J=12.2,1.5Hz, lH)ppm. Irradiation of the H8 carbinol doublet at 5.92ppm simplified the H9 triplet at 7.02ppm to a doublet (J=0.6Hz). Hence, H8 through to Hll were connected via scalar coupling. Irradiation of the olefinic H5 resonance at 6.21(q, J=1.2Hz, lH)ppm collapsed the methyl doublet (J=1.2Hz) at 2.08ppm (Mel6) to a singlet and irradiation of both the H14 protons, resonating at 4.76 and 4.86ppm, sharpened the methyl singlet at 1.72ppm (Mel5). Also an NOE was observed between Mel5 and the Hl4b resonance at 4.86ppm. The stereochemical arrangement about the C4-C5 double bond was proposed on the basis of spectral evidence. Unlike A4f 5Z ,^7'"17- isogersemolide (168) no NOE could be demonstrated between the H5 and Mel6 protons, and although this does not prove the stereochemical assignment it was at least consistent with 123 proposed structure. Secondly, the H5, Mel6 proton coupling constant in 168 was 1.6Hz while in 169 it was 1.2Hz. This is consistent with the general observation that allylic cis coupling 197 is larger than allylic trans coupling . To confirm the proposed structure A4,5E, A7'"^-isogersemolide (169) was treated with a catalytic amount of iodine in benzene. *H NMR analysis showed that a 1:3 mixture of 168 and 169. respectively, had resulted. To allow for a complete assignment of the NMR resonances exhibited by 169. a second series of decoupling experiments were performed. Irradiation of the Hlla methylene resonance at 2.17ppm markedly sharpened the multiplet regions resonating between 2.49- 2.57 (consisting of H2a and Hllb) and 1.26-1.41ppm (Hl2a and Hl2b). Irradiation of the Hl2a, Hl2b two proton multiplet at 1.26-1.41ppm sharpened the HI multiplet at 2.33ppm, simplified the Hlla at 2.17(tdm, J=12.2,1.5Hz)ppm to a doublet of multiplets (J=12.2Hz) and sharpened the Hllb multiplet at 2.49-2.57(m)ppm. Irradiation of the Hi multiplet at 2.33ppm sharpened the Hl2a, H12b two proton multiplet (1.26-1.41ppm) and the multiplet region 2.49-2.57ppm (H2a and Hllb), and reduced the H2b doublet of doublets (J=14.0,5.7Hz) at 2.71ppm to a doublet (J=14.0Hz). Finally, irradiation of the H2b resonance at 2.71ppm affected the geminal H2a resonating in the multiplet region 2.49-2.57ppm and simplified the HI multiplet at 2.33ppm to an apparent broad triplet (J-8.2HZ). Hence, all the proton resonances were assigned (see Table 15.). 124 (iii). Cembrane diterpenoids. 120.-125. Rubifolide f170). obtained as a white crystalline solid, was shown by EIMS to possess a molecular formula of C2oH24°3 (observed m/z 312.1275, requires 312.1276), requiring 9 units of unsaturation. A number of fragments could readily be identified from the H and XJC NMR spectra (Figures 27, and 28.). The first 13 fragment, a 2,5-dialkyl-3-methylfuran, was indicated by the C NMR resonances at 9.5(q) 149.4(s), 117.l(s), 113.8(d) and 149.9(s)ppm (carbon multiplicities deduced from a fully coupled 13 C NMR spectrum) which were almost identical in chemical shift with the resonances assigned to the corresponding fragment in kallolide B (1££)157 (Table 16.). In the model compound 386157 the /3-proton and the methyl substituent were assigned *H NMR resonances at 5.85(s, IH) and 1.96(s, 3H)ppm, respectively. Irradiation of the olefinic resonance at 5.99(bs, lH)ppm sharpened the methyl singlet at 1.92(bs, 3H)ppm. Thus these resonances were assigned to the equivalent protons in the methylfuran fragment of 170. The fragment accounted for three units of unsaturation. Carbon resonances at 78.7(d), 152.1(d), 132.8(s) and 174.5(s)ppm, and a NMR resonance at 6.86(bs, lH)ppm, identified the second fragment as an «i?'-disubstituted a,f3- unsaturated y-lactone identical with the corresponding 13 substructure of gersemolide f167). The appropriate C NMR comparisons are given in Tables 14. and 17. The *H NMR resonance at 6.86ppm and the IR band at 1753cm-1 were, as expected, typical of such a system111'1^7'197'203. The lactone moiety accounted for 125 126 Table 16. C NMR data comparison of 170 and 186. Chemical Shift, ppm. Chemical shift, ppm. Carbon # Carbon # 3 86a 3 149.4 3 149.6Z 4 117.1 4 116.4 5 113.8 5 112.4 6 149.9 6 150.0Z 18 9.5 16 9.7 reference 157, 50MHz, CDC13. Assignments within a column may be interchanged. a further 3 units of unsaturation. The third structural fragment, an isoproprenyl residue, was readily apparent from the 13C NMR resonances at 145.4(s), 112.9(t) and 19.2(q), and ''"H NMR resonances, typical of such a system111'157'197'203, at 4.88(bs, 1H), 4.90(sharp m, 1H) and 1.74(bs, 3H)ppm (see Tables 14. and 17., 15. and 18., for spectral comparisons). The fragment was confirmed by two observations. First, on irradiating the 1H NMR resonance at 4.90ppm the methyl signal at 1.74ppm sharpened and second, an NOE was demonstrated between this olefinic proton and the methyl. Finally, a fourth structural fragment, a trisubstituted 13 olefin bearing a methyl, was evidenced by the C NMR resonances at 117.4(d), 127.0(s) and 25.7(q)ppm, and 1H NMR resonances at 6.07(bs, 1H) and 1.98(bs, 3H)ppm. The preceeding arguments indicated that rubifolide f170) 128 Table 17. C NMR assignments for rubifolide (170) , epilophodione (171) , isoepilophodiones A (122) and C (121) and rubifol (125.) , and spectral comparison with 188 and 189. Chemical shift, ppm. Carbon # 120. 121 122 121 1 43.3(d) 41.8(CH) 41.52 35.72 Z Z 2 31.2(t) 45.5(CH2) 42.8 41.7 3 149.4(s) 205.5(C) 202.6 202.0 4 117.l(s) 143.9(C)2 149.8y 154.3y 5 113.8(d) 133.5(CH) 132.3 128.3 6 149.9(s) 192.1(C) 194.6 196.9 7 117.4(d) 127.KCH) 130.2 48.32 8 127.0(s) 151.8(C) 140.0y 143.7y 2 Z 9 39.5(t) 43.6(CH2) 38.9 41.6 10 78.7(d) 78.4(CH) 78.1 80.0 11 152.1(d) 148.1(CB) 147.8 148.0 12 132.8(s) 136.2(C) 134.1 132.9 X X 13 20.0(t) 21.0(CH2) 22.1 21.2 14 30.5(t) 33.5(CH2) 29.0 27.6 15 145.4(s) 145.4(C)2 144.9y 145.8y 16 112.9(t) 115.6(CH2) 113.2 113.0 17 19.2(q) 17.2(CH3) 14.3 18.1 y X X 18 9.5(q) 21.1(CH3) 18.3 27.3 y X 19 25.7(g) 22.3(CH3) 23.4 125.3 20 174.5(s) 173.6(C) 172.2 173.9 cont'd. 129 Table 17. continued. Chemical shift, ppm. Carbon # 125. 1883 189a 1 40.7(CH) 41.3 40.5 Z 2 39.3(CH2) 45.7 44.8 3 217.3(C)y 205.4 208.3 4 155.2(C)X 144.8Z 154.1 5 126.3(CH) 133.4 126.5 6 202.9(C)Y 190.7 189.0 Z 7 50.2(CH2) 125.9 128.0 8 80.7(C) 156.3 146.4 Z 9 37.6(CH2) 43.5 37.3 10 78.3(CH) 80.1 77.9 11 147.2(CH) 148.4 148.6 12 132.3(C) 134.1 133.1 13 21.3(CH2) 25.7 22.2 14 26.6(CH2) 30.3 31.2 15 145.2(C)X 145.5Z 146.4 16 112.9(CH2) 115.9 114.0 17 18.2(CH3) 17.0 19.1 W 18 26.7(CH3) 21.5 22.2 W 19 24.9(CH3) 22.6 26.9 20 173.6 173.1 173.1 a reference 111, 20MHz, CDC13. Hydrogen attachments3were determined using a single frequency off resonance C NMR spectrum. z_wassignments within a column may be interchanged. 130 Table 18. H NMR assignments for rubifolide (170) . epilophodione (121), isoepilophodiones A (122), B (173) and C (174) and rubifol (175), and spectral comparison with 188 and 189. Chemical shift, ppm. Carbon # 120. 121 1 2.36(td, J=12.6f4.2Hz, 1H) 2.20-2.47(m) 2a 2.55(m, 2H) 2.51(m, 2H) b 5 5.99(bs, 1H) 6.38(q, J=1.6Hz, 1H) 7 6.07(bs, 1H) 6.12(bs, 1H) 9a 2.68(dd, J=11.4,4.2Hz, 1H) 2.67(dd, J=13.3,4.5Hz, 1H) b 3.22(t, J=11.4Hz, 1H) 2.82(dd, J=13.3,4.8Hz, 1H) 10 4.95(dm, J=11.4Hz, 1H) 5.27(m, 1H) 11 6.86(bs, 1H) 7.15(bd, J=0.9Hz, 1H) 13a 2.08(dm, J=14.0Hz, 1H) 2.20-2.47(m) b 2.43(td, J=14.0,2.7Hz, 1H) 14a 1.18(tq, J=14.0,2.7Hz, 1H) 1.38(m, 1H) b 1.64(m, 1H) 1.86(m, 1H) 16a 4.88(bs, 1H) 4.92(m, 2H) b 4.90(m, 1H) 17 1.74(bs, 3H) 1.61(bs, 3H) 18 1.92(bs, 3H) 1.91(d, J=1.6Hz, 3H) 19 1.98(br, 3H) 2.19(d, J=1.2Hz, 3H) cont'd. possessed five carbon-carbon double bonds, a lactone carbonyl and the two rings, the furan and lactone. Thus, 170 had to possess one further carbocyclic ring to account for the total of nine 131 Table 18. continued. Chemical shift, ppm. Carbon # 112 173' 1 2.33-2.45(m)z 2.20-2.62(b)2 2a 2.50(m, 2H)Z 2.20-2.62(b)2 b 2.68-2.92(b) 5 6.68(q, J=1.2Hz, 1H) 6.21(bm, 1H) 7 6.07(bs, 1H) 6.25(bs, 1H) 9a 2.98(ddm, J=13.5 ,5.1Hz, 1H) 2.68-2.92(b) b 3.04(ddm, J=13.5 ,4.5Hz, 1H) 4.03(b, 1H) 10 5.23(bm, 1H) 5.27(b, 1H) 11 7.19(bs, 1H) 7.16(b, 1H) 13a 2.33-2.45(m)z 2.2g-2.62(b)z b 14a 2.06(m, 1H)Z 1.41(b, 1H) b ? 16a 4.54(bs, 1H) 4.64(b, 1H) b 4.76(q, J=0.6Hz, 1H) 4.83(bm, 1H) 17 1.66(bs, 3H) 1.70(bs, 3H) 18 2.16(d, J=1.2Hz, 3H) 1.88(bs, 3H) 19 1.99(bs, 3H) 2.06(ra, 3H) cont'd. units of unsaturation. Further to this, the ring had to accomodate the remaining unassigned aliphatic methine and four aliphatic methylene carbons (13C NMR: 43.3(d), 20.0(t), 30.5(t), 31.2(t) and 39.5(t)ppm). The cembranoid structure 170 effectively accomodates all the identified structural fragments of rubifolide. 132 Table 18. continued. Chemical shift, ppm. Carbon # 174 175 1 1.85-2.06(m)z 3.34(bd, J=10.7Hz, IH) 2a 2.20(m, 2H)Z 2.09-2.19(m) b 2.21-2.37(m) upfield part 5 6.29(q, J=1.2Hz, IH) 6.31(bs, IH) 7a 3.08(d, J=15.0Hz, IH) 2.47(d, J=17.3Hz, IH) b 3.73(d, J=15.0Hz, IH) 3.37(d, J=17.3Hz, IH) 9a 2.14(m, IH) 2.59(dd, J=10.8,3.9Hz, IH) b 3.05(dd, J=ll.l,1.8Hz, IH) 3.91(bt, J=10.8Hz, IH) 10 5.06(bm, IH) 4.83(m, IH) 11 6.79(d, J=1.8Hz, IH) 6.64(d, J=0.6Hz, IH) 13a 1.85-2.06(m)z 2.09-2.19(m) b 2.21-2.37(m) downfield part 14a 1.70-1.82(m, 2H)Z 1.67-1.77(m, IH) b 1.82(bt, J=10.7Hz, IH) 16a 4.69(bs, IH) 4.82(bs, IH) b 4.81(m, IH) 4.84(bs, IH) 17 1.61(bs, 3H) 1.70(bs, 3H) 18 1.95(d, J=1.2Hz, 3H) 2.06(bs, 3H) 19 5.90(d, J=0.9Hz, IH) 1.20(s, 3H) 6.08(s, IH) cont'd. NOE and spin-spin decoupling experiments verified that 170 represented the correct structure. Irradiation of the methyl resonance at 1.98ppm (Mel9) in the 1H NMR spectrum of rubifolide 1110) induced positive NOE's into an olefinic proton at 6.07ppm (H7), a methylene proton at 2.68ppm (H9a), the H10 carbinol 133 Table 18. continued. Chemical shift, ppm. Carbon # l££b 189b 1 / / 2 2.40(m, 1H) / 2.64(bd, J=14.0Hz, 1H) 5 6.42(bs, 1H) 6.15(bs, 1H) 7 6.12(bs, 1H) 6.26(bs, 1H) 9a 2.61(bd, J=13.3Hz, 1H) 2.41(m, 1H) b 3.04(dd, J=13.3,4.6Hz, 1H) 2.67(m, 1H) 10 5.31(m, 1H) 4.99(m, 1H) 11 6.97(bs, 1H) 7.00(bs, 1H) 13a / / b / / 14a / / b / / 16a 4.72(bs, 1H) 4.70(bs, 1H) b 4.97(bs, 1H) 4.90(bs, 1H) 17 1.60(bs, 3H) 1.68(bs, 3H) 18 1.84(bs, 3H) 1.89(bs, 3H) 19 2.19(bs, 3H) 2.04(bs, 3H) awithin this column "b" means very broad signals with no fine structure. breference 111, 220 and 360MHz, CDC1,. No chemical shifts reported for "/" (HI, H13a and b, and H14a and b). zassignments within a column may be interchanged. "?"could not assign from data available. 134 proton at 4.95ppm, and induced a weak negative NOE into a methylene proton at 3.22ppm (H9b). The lactone methine proton at 4.95ppm (H10) showed vicinal coupling to the methylene protons H9a and H9b (2.68 and 3.22ppm, respectively) and to the olefinic Hll (6.86ppm) and it showed homoallylic coupling to the H13a methylene proton at 2.08ppm. Irradiation of the furan methyl resonance at 1.92ppm (Mel8) induced NOE's into an olefinic proton at 5.99ppm (H5) and into the H2 methylene protons at 2.55ppm. Similarly, irradiation of the isopropenyl methyl resonance at 1.74ppm (Mel7) also induced an NOE into the H2 protons (2.55ppm), and with this the Mel7 and Mel8 were connected through space. Hence, connectivity was established from the isopropenyl group to H5 and from H7 to H13a. The chemical shift assignments, thus far, are summarised in Table 18. A second set of decoupling experiments extended the connectivity to include all protons from H7 to HI. The results are summarised in Table 18. However, due to the overlap of resonance signals in the region from 2.32 to 2.60ppm the chemical shift assignments, coupling constants and connectivities for Hi, H2a, H2b and Hl3b were confused. A lanthanide shift experiment with decoupling resolved the problem and allowed all protons, but the H2 methylene protons, to be viewed as first order in the 400MHz •'"H NMR spectrum (Figure 29.). The results of the addition of Eu(FOD)3 are shown in Figures 30a-c. It is seen that the region from 2.32 to 2.46ppm, in Figure 29., can be viewed as two pairs of triplet of doublets resonating at 2.36(td> J=12.6,4.2Hz, IH) and 2.43(td, J=14.0,2.7Hz, lH)ppm, and the resonance at 135 136 Figure 30. 400MHz lanthanide shifted XH NMR spectra of rubifolide (170). 2.55ppm as a second order two proton multiplet. A final series of decoupling experiments allowed for the unambiguous assignment of resonances and extended the scalar coupling connectivity to include the H2 protons. Irradiation of the broad triplet, in Figure 30c, resonating at 3.93ppm (H13b) simplified the broad triplet at 1.69ppm (H14a) to a doublet, affected the resonance lying underneath the two methyl resonances between 2.07 to 2.13ppm (Hl4b) and dramatically sharpened the H13a resonance presumably centered at 2.76ppm. Irradiation of the Hi multiplet (3.32ppm) affected the H14b resonance (2.07-2.13ppm), markedly sharpened the H2a, H2b multiplet at 2.78ppm and weakly sharpened 3+ the H14a resonance (1.69ppm). Plots of Eu volume (ul) against observed chemical shift resulted in a series of straight line plots. The plots that corresponded to the resonances at 2.76ppm (Hl3a), 3.32ppm (Hi), 3.93ppm (Hl3b) and 2.78ppm (H2a, H2b), in Figure 30c, crossed the chemical shift axis at 2.075, 2.340, 2.390 and 2.545ppm, respectively. Although each plot was based on only three data points the results allowed for the assignments given in Table 18. A summary of the coupling constant data for 170 is given in Table 19. The spin-spin decoupling experiments demonstrated that H14a and Hi zero coupled (or had a very small coupling in the shifted spectra). A dihedral angle of 90° (required for zero coupling) was readily obtainable in molecular models for either of the H14 protons in a manner that allowed for all the structural restraints imposed by the NOE results. The model and NOE results demonstrated that the resonances at 3.22(t, J=11.4Hz, H9b) and 140 Table 19. Coupling constant data for rubifolide (170). Observed Proton # multiplicity Magnitude, Hz Jx,y HI td J l,2a 12.6 Jl,2b 4.2 J 0.0 lr14a Jl,14b 12.6 H2 m J 4.2 and 12.6 l,2 ? J2,2 H5 bs J sharpened 5,18 H7 bs J sharpened 7,9b J 0.8 7,19 H9a dd J 11.4 9a,9b J 4.2 9a,10 H9b t J sharpened 7,9b J 11.4 9af9b J 11.4 9b,10 H10 dm J 4.2 9a/10 J 11.4 9b,10 J 1.4 10fH J 2.0 10r13a Hll bs J 1.4 10,ll J 1.0 ll,13a H13a dm J 2.0 10,13a J 1.0 ll,13a cont'd. 141 Table 19. continued. Observed Proton # multiplicity Jv Magnitude, Hz J 14.0 13a,13b J 2.7 13a,14a J 2.7 13a,14b H13b td J 14.0 13a,13b J 14.0 13b,14a J 2.7 13b,14b H14a td J 2.7 13a,14a J 14.0 13b,14a J 14.0 14a,14b J 0.0 l,14a J 2.7 H14b m 13a,14b J 2.7 13b,14b J 14.0 14a,14b J 12.6 l,14b -means not observed and/or unable to deduce. sharpened-on irradiating either of the appropriate protons the other proton was sharpened, but an actual J value could not be assigned. The J's are probably, therefore <0.5Hz. 2.68(dd, J=11.4,4.2Hz, H9a)ppm were to be assigned to the H9/? and H9a protons, respectively. The H10 and H9a (H9a) would effectively display an axial-equatorial array (with an approximate dihedral angle of 50°), with the observed coupling constant of 4.2Hz, and the H10 and H9/3 (H9b) protons an axial- 142 axial array (with an approximate dihedral angle of 175°) with coupling of 11.4Hz. The magnitude of the two coupling constants 197 are consistent with the proposed configuration . 13 Since there was some ambiguity over the C NMR assignments for the C3 and C6 carbons (at 149.4(s) or 149.9(s)ppm), and as connectivity had only been established from H7 through to H5f a long range 2D-HETC0R NMR spectrum76'204 of 170 was obtained. As 7 6 with the short range experiment (see pages 33. and 90.) connectivities between two different nuclei are established. However, by utilising the appropriate delay times the 2 3 experiment can be optimised to detect long range JCH and JCH couplings while suppressing the much larger ^JQ^ coupling. The experiment was optimised for polarisation transfer through 8Hz coupling, since a long range coupling of approximately 8Hz had been observed into the carbon resonating at 149.9ppm in the fully 1 3 coupled C NMR spectrum of 170. Both the H5 and H7 protons transferred magnetisation to the 13C NMR resonance at 149.9(s)ppm identifying it as C6, while only H5 transferred magnetisation to 149.4(s)ppm identifying it as C3. Hence, the experiment allowed for the assignment of C3 and C6, and completed the connectivity about the 14-membered ring of 170. Further connectivities were apparent, these were fully consistent with structure 170. and 13 allowed for the unambiguous assignment of C NMR resonances. The correlation results are summarised in Table 20., and the subsequent chemical shift assignments are summarised in Table 17. The occurrence of an NOE between the C19 methyl protons (1.98ppm) and the H7 olefinic proto143 n (6.07ppm) requires that the C7-C8 olefinic bond of rubifolide (170) has a Z configuration. Table 20. Correlations observed in a long range 2D-HETCOR NMR spectrum of rubifolide (170). Proton Coupling Carbon coupled to H2af H2b 2-bond C3 H5 3-bond C3 2-bond C6 1-bond C5 H7 3-bond C9 2-bond C6 1-bond C7 Hll 3-bond C20 1-bond Cll Mel7 3-bond C16 2-bond C15 Mel8 3-bond C5 3-bond C3 2-bond C4 1-bond C18 Mel9 3-bond C9 3-bond C7 2-bond C8 1-bond C19 Finally, it was assumed that the relative configurations at the CI and CIO carbons in rubifolide (170) were the same as those assigned to the corresponding carbons in gersemolide (167). The enantiomer shown is arbitrary. Epilophodione f171), obtained as colourless needles, was shown by EIMS to possess a molecular formula of C20H24°4 (observed m/z 328.1680, requires 328.1675) which required nine 144 1 13 units of unsaturation. The H and C NMR spectra (Figures 31. 73 and 32.), in combination with the ADEPT spectrum (Figure 33), readily identified methyl substituted ene-dione, a,r- disubstituted a,/?-unsaturated r-lactone, isopropenyl and trisubstituted olefin residues, identical with the corresponding substructures in gersemolide (167) and rubifolide (170). The NMR resonances are compared in Tables 14, 15, 17 and 18. The 1H and 13C NMR data (Tables 18. and 17.) observed for epilophodione (171) showed a striking resemblence to 11J the data reported by Bandurraga et al. for lophodione (188,) "r suggesting that the two metabolites had the same constitution and differed only in some stereochemical sense. The structure 171 was proposed for epilophodione and was confirmed by a series of NOE experiments. Irradiation of the Mel8 methyl protons (1.91ppm) in 171 induced an NOE into the H5 olefinic proton (6.38ppra) indicating that the C4-C5 olefinic bond in epilophodione (171) had the Z configuration in common with the corresponding bond in lophodione (188)111. No NOE could be demonstrated between the Mel9 protons (2.19ppm) and the H7 olefinic proton (6.12ppm) in epilophodione (171), a result that was consistent with the C7-C8 olefinic bond having the E configuration found in lophodione (188)11]". The results of subsequent NOE irradiations were consistent with the proposed structure and are summarised in Table 21. The gross constitution was thus confirmed, and proposed 1H and 13C NMR assignments are given in Table 18. and 17., respectively. Since the olefinic geometries in epilophodione (171) were identical to those of lophodione (lflS.)111/ the stereochemical 145 147 Details for each: 16-0mg 2hours 20 Omins CH3 Peaks s 00 CH2 Peaks CH Peaks CH,CH2,CH3 Peaks *i>iUi)Mi»«Mi|i|4iWMi'M»NiWi»W Figure 33. ADEPT NMR spectra of epilophodione (111). Table 21. Subsequent NOE results observed for epilophodione (111). Proton irradiated Protons with observed NOE H7 H5 H10 H9a H9b Hll Mel7 H16(a and/or b) H2ar H2b Mel8 H16(a or b) H2a and/or H2b Mel9 Hll H10 H9a difference in the two structures had to involve the relative configurations at the CI and CIO chiral centers. The reported relative configurations, derived from an x-ray crystallographic analysis, for the CI and CIO chiral centers in lophodione are * * S ,S . If the two chiral centers in epilophodione (121) have the same configurations as the corresponding chiral centers in gersemolide (167), the relative configurations at CI and CIO must be R ,S , respectively, making 171 epimeric with 188. This is perhaps supported by the observation that the two metabolites rotate the plane of polarised light in opposite directions; in epilophodione (171) the optical rotation was positive, for lophodione (188) it was negative. The absolute configuration of lophodione (188)111 was not reported and the absolute 149 configuration of epilophodione (121) has not been determined, so it is not possible to say which of the two chiral centers has the opposite configuration in the two metabolites. Isoepilophodione A (122) i isolated as a pale yellow oil, was shown by EIMS to possess a molecular formula of C2oH24°4 (observed m/z 328.1686, requires 328.1675), identical to that of 1 13 epilophodione (171). Examination of the IR, DV and the H and X"X NMR spectra (Figures 34. and 35.) of isoepilophodione A (172) - identified a number of structural fragments; afr disubstituted a» ^-unsaturated r-lactone, methyl substituted ene-dione, trisubstituted olefin and isopropenyl moieties, identical with the corresponding substructures of epilophodione (171) (allbeit with very slightly different chemical shifts, see Tables 17. and 18.). The strong correlation of all spectral data with 171 suggested that the two compounds could be geometrical isomers of one another at either or both of the trisubstituted olefins. This type of situation has a precedent. The C4-C5 double bond in the pseudopterolides, 167-169r can possess either the Z or E configuration, and lophodione (188) was known to co-exist with the C4-C5 E and C7-C8 Z geometric isomer, isolophodione (189)111. 0 189 150 The AH and •LJC NMR data for 189 are included in Tables 17. and 18. Hence, isomerization of epilophodione (121) to isoepilophodione A (172), and visa-versaf was accomplished using iodine in benzene, providing confirmation of the proposal. However, the geometry of the two trisubstituted olefins in 172 was still in question: the C4-C5 Z and C7-C8 E olefins in epilophodione (171) could potentially isomerize to either the corresponding E,E, Z,Z, or E,Z olefins. Irradiation of the methyl protons resonating at 1.99ppm induced NOE's into the olefinic proton resonating at 6.07ppm and the lactone carbinol methine proton at 5.23ppm (H10), identifying them as the Mel9 and H7 protons, respectively. The NOE experiment demonstrated that the C7-C8 carbon-carbon double bond possessed a Z configuration. Consistent with this assignment was the observation that irradiation of H7 (6.07ppm) only induced an NOE into Mel9 (1.99ppm). Irradiation of the methyl protons resonating at 2.16ppm only induced a small positive NOE into the H16a proton resonating at 4.54ppm, identifying this as the Mel8. Assignment of the E configuration to the C4-C5 double bond of 172 was consistent with this NOE result. Construction of a molecular model for 172. possessing the structural restraints imposed by the above NOE results, readily demonstrated that the H5 olefinic proton (6.68ppm) pointed into the interior of the ring. For a number of conformations the H2, H9 and H13 (or H14, but not both H13 and H14 at the same time) protons, similarly, all pointed into the interior, so that one or both protons in each of the methylene pairs were situated close to the H5 proton. Irradiation of H5 (6.68ppm) induced NOE's into the H9a doublet of doublets resonating at 2.98ppm and into a 153 number of protons resonating in the two multiplet regions, 2.33- 2.45ppm and the one centred at 2.50ppm (H2a, H2b). The preceeding 1 13 arguments all support the proposed structure, 372. and C NMR assignments are included in Tables 18. and 17. Interestingly for both'isoepilophodione A (172) and A4,5E, 7 17 A ' -isogersemolide (1£9_) the chemical shift of the C3 carbon was more shielded (more conjugated) compared to epilophodione (171) and gersemolide (167) by 2.9 and 5.0ppm, respectively. The C6 ketone resonance of 172 was more deshielded (less conjugated), compared to 121, by 2.5ppm. It appears that planarity between the C6 ketone and C4-C5 double bond is partially lost while planarity between the C3 ketone and C4-C5 double bond is partially gained in going from a Z to E configuration. Isoepilophodione B (173). obtained as a pale yellow oil, was shown by EIMS to possess a molecular formular of C2oH24°4 (observed m/z 328.1670, requires 328.1675), identical to that of epilophodione (171) and isoepilophodione A (172). The *H NMR spectrum of 173 (Figure 36.) consisted of very broad signals so that very little structural information could be gained. Fortunately, however, isomerization of both epilophodione (171) and isoepilophodione A (172) had also generated isoepilophodione B (122.). Isomer ization of 173 to 171 and 172f as expected, was readily accomplished using iodine in benzene. Thus, 173 had to exist as one of the two remaining geometrical isomers of epilophodione (171). ie., with C4-C5 and C7-C8 carbon-carbon double bonds in either the E,E or Z,Z configurations. Irradiation of the methyl protons resonating at 1.88ppm induced an NOE into 154 the olefinic proton at 6.21ppm. Similarly, irradiation of the methyl protons resonating at 2.06ppm induced an NOE into the proton at 6.25ppm. Hence, both double bonds had to possess the Z configuration and structure 173 was proposed for isoepilophodione B. Isoepilophodione B (173) must exist in at least two slowly interconverting low energy conformations in solution, and hence, those protons that sit in different chemical environments due to the exchange are excessively broad. Initial attempts to resolve the problem involved high temperature *H NMR experiments. At 50°C the problem was still not removed and resulted in the partial decomposition of 173. A series of low temperature NMR experiments resulted in the freezing out of two conformations (in a ratio of 4:1) (see Figure 37., -40°C appeared to be the optimum temperature, going to lower temperatures had no apparent effect). Spectral comparison of the low temperature ^H NMR data of 173 to the room temperature data given in Table 18., for 170. 171. 122* 188 and 189 indicated that at -40°C the H10 lactone carbinol proton resonated at 5.42ppm (see Figure 37b.). With this a series of spin-spin decoupling experiments at -40°C allowed for proton NMR assignments. Irradiation of the broad doublet, of the major conformer, resonating at 4.07ppm (J=11.2Hz, H9b) simplified the broad doublet of doublets at 2.79(J=11.2,8.0Hz, H9a)ppm to a doublet (J=8.0Hz) and sharpened the broad olefinic singlet at 6.30ppm (H7). Irradiation of the doublet of doublets at 2.79ppm (H9a) collapsed the methylene doublet at 4.07ppm (H9b) to a broad singlet and sharpened the methine proton at 5.42ppm (H10) to a broad singlet. These assignments were correlated to the room temperature H NMR assignments given in Table 18. H5 could be assigned to 6.21ppm since H7 was assigned the resonance at 6.25ppm, and hence the Mel8 protons were assigned to the resonance at 1.88ppm and Mel9 to 2.06ppm. Finally, at -40°C irradiation of the H9b broad triplet (3.68ppm), of the minor conformer, sharpened H9a (resonating underneath the H9a broad doublet of doublets (2.79ppm) of the major conformer). Previously irradiation at 2.79ppm had also simplified the broad triplet at 3.68ppm (H9b) to a broad doublet and collapsed the H10 doublet of multiplets (5.01ppm) to a broad singlet. Thus, the above partial 1H NMR assignments for the minor conformer were also made. Isoepilophodione C (174)f isolated as a clear oil, was shown by EIMS to possess a molecular formula of C2oH24°4 (°Dserved m/z 328.1671, requires 328.1675) which required nine units of 113 unsaturation. The H and C NMR spectra (Figures 38. and 39.) and IR spectra readily identified methyl substituted ene-dione, a»r-disubstituted a,^-unsaturated r-lactone and isopropenyl residues, identical with the corresponding substructures in gersemolide (167). rubifolide (170) and epilophodione (171) (see 13 Tables 14, 15, 17 and 18). Further to these fragments, the C NMR resonances at 125.3 and 143.7ppm and the 1H NMR resonances at 5.90(d, J=0.9Hz, IH) and 6.08(s, lH)ppm tentatively identified an exocyclic methylene moiety. This fragment was consistent with the observation that on irradiation of the olefinic proton resonating at 5.90ppm an NOE was induced into the proton at 6.08ppm. Isoepilophodione A (172) partially isomerized to isoepilophodione C (174) on treatment with formic acid under reflux conditions, suggesting that the two metabolites were 159 related by a simple double bond migration. The cembranoid structure 174 effectively accomodated all the identified structural fragments of isoepilophodione C. A series of NOE, 199 spin-spin decoupling and SINEPT NMR experiments verified that 174 represented the correct structure. Irradiation of the Mel8 protons resonating at 1.95ppm induced NOE's into the H5 proton at 6.29ppm and the H16a at 4.69ppm, and an across ring NOE was also observed into the olefinic lactone proton (Hll) (6.79ppm). Irradiation of the methyl resonance at 1.61ppm (Mel7) induced an NOE into the H16b proton at 4.81(m, lH)ppm, and under decoupling conditions the same irradiation sharpened the H16b resonance. Thus, assignment of the Z configuration to the C4-C5 double bond was confirmed and the Mel8 protons and the isopropenyl residue were connected through space and could be assigned to their "usual" positions at C4 and CI, respectively. Irradiation of the H7b broad doublet (J=15.0Hz) resonating at 3.73ppm collapsed the H7a doublet (J=15.0Hz) at 3.08ppm to a singlet and sharpened the Hl9a doublet at 5.90ppm to a singlet. As expected, irradiation at 3.08ppm (H7a) collapsed H7b (3.73ppm) to a singlet. Irradiation of H7b (3.73ppm) under NOE conditions induced NOE's into the H7a doublet (3.08ppm) and H5 (6.29ppm). Simultaneous irradiation of H7a (3.08ppm) and H9b (3.05ppm) induced NOE's into H7b (3.73ppm), Hl9b (6.08ppm) and the H9a multiplet at 2.14ppm and small NOE's were observed into H5 (6.29ppm) and H19a (5.90ppm). Irradiation of the exocyclic methylene H19a proton at 5.90ppm not only induced an NOE into Hl9b, but also H7a, and irradiation of H19b induced an NOE into the H9b proton resonating at 3.05ppm. This series of experiments connected the H5 to the H9 protons through 162 spatial and scalar coupling connectivities of the two H19 protons with the H7 and H9 protons. Finally, irradiation of the lactone H10 proton resonating at 5.06ppm sharpened both the H9a multiplet centered at 2.14ppm and the lactone olefinic proton (Hll) at 6.79ppm, and irradiation at 2.14ppm (H9a) reduced H9b (3.05ppm) to a broad singlet. Hence, the isopropenyl moiety was connected spatially to the Mel8 protons which were subsequently connected around the ring to the lactone Hll proton through both dipolar and scalar coupling correlations. A molecular model demonstrated that the H7a, H7b, H9a and H9b protons were to be assigned to the Elft, H7a, H9a and H9/3 protons, respectively. 199 Three SINEPT experiments, optimised for polarisation transfer through 7Hz coupling, connected the exocyclic methylene and ene-dione fragments, and allowed for 13C NMR assignments to be made. First, irradiation of the methylene proton at 3.73ppm (H7b) showed polaristion transfer to the carbon resonances at 128.3(C5), 143.7(C8) and 196.9(C6)ppm. Second, irradiation of the exocyclic methylene proton resonating at 5.90ppm (H19a) showed polaristion transfer to carbon resonances at 48.3(C7 or C9, more likely C7) and 143.7(C8)ppm. Thus, the structural constitution was further supported. The third experiment involved irradiation of the H10 lactone methine proton (5.06ppm) and it confirmed the assignments made for C12 and C20, since transfer of polarisation was observed into the carbon resonances at 132.9(C12) and 1 13 173.9(C20)ppm. Proposed H and C NMR assignments are included in Tables 18. and 17. Rubifol f175). isolated as colourless needles, was shown by 163 EIMS to possess a molecular formula of C20H26°5 (°DServe(^ m/z 346.1785, requires 346.1781) which required 8 units of unsaturation. The IR, 1H, 13C and APT205 NMR spectra (for latter three, see Figures 40., 41., and 42.) readily identified methyl substituted ene-dione, a,y-disubstituted a, ^-unsaturated r-lactone and isopropenyl residues, identical with the corresponding substructures in gersemolide (167), rubifolide (170) and epilophodione (171) (see Tables 14., 15., 17. and 18.). — 1 13 The IR band at 3509cm and a C NMR resonance at 80.7(C)ppm identified a fourth fragment, a tertiary alcohol. The presence of the alcohol was confirmed by the loss of water from the parent ion (m/z 346 daltons) in the MS to give a fragment ion at an m/z 197 of 328 daltons . Hence, seven degrees of unsaturation and all five oxygens were accounted for. The preceeding arguments indicated that rubifol (175) had to possess a second ring to account for the total of eight units of unsaturation. The cembranoid structure 175 effectively accomodated all the identified structural fragments. An extensive series of NOE and spin-spin decoupling experiments verified that 175 represented the correct structure. The NOE results are summarized in Table 22. Decoupling experiments, in combination with the NOE results, demonstrated that the *H NMR resonances at 2.47 and 3.37ppm were to be assigned to a geminal pair, the H7a and H7b protons, respectively, and similarly the C9 methylene protons were assigned the resonances at 2.59 and 3.91ppm. The H9a, H9b protons also showed vicinal coupling to the lactone methine proton at 4.83ppm (H10). The decoupling and NOE results, discussed thus far, were consistent with the proposed structure 164 s Details: 76mg lOhours 50-1mins C's,CH2's up, CH's,CH3's down J>|»>i||l0l^wL»llNi>>l^l>l*^W>*W^i(l 11 1 1 1 1 11 11 1 111 111 i 111 i | 11 i 11 i 11111 111 111 1111 i 11 i i 111 11 i i i 111 i 11 11 i 1111 i 111 11 i 11 i i i | i i i i 11 i 11 | i i i i i i 11 i | i i i 11 i 1 ' i 220 200 180 160 140 120 100 80 60 40 20 PPM 0 Figure 42. APT NMR spectrum of rubifol f175). Table 22. NOE results observed for rubifol (175). Proton irradiated Protons with observed NOE HI H2b H14b Mel9 H2a and H13aa H9a Hll Hl4a H7a H7b Mel9 H7b H7a H5 Mel9 H9a H9b H10 H9b H9a Hll Hll H2a, H2b H9b Mel7 and H14aa H16b H2a H2b Hl3b Mel8 H5 H16a, H16b Mel9 HI H7af H7b result of a simultaneous irradiation. and connected through from the isopropenyl moiety to the olefinic lactone proton (Hll). Assuming that the CI and CIO carbons in 168 rubifol (175) possess the same relative configuration as the appropriate carbons in gersemolide f167), a molecular model possessing the configurational restraints imposed by the NOE results demonstrated that the relative configuration at C8 had to be S . A second series of spin-spin decoupling experiments in combination with the NOE results and a molecular model clearly demonstrated the connectivity in the eight-spin system found on Cll to C2 and thus completed the connectivity about the 14- membered ring. The results of the decoupling experiments are summarised in Table 23., and the subsequent *H NMR assignments 13 are included in Table 18. Proposed C NMR assignments are included in Table 17. Finally, to remove ambiguities associated with the assignment of the lactone methine proton (H10) to the resonance at 4.83(m, lH)ppm (lying underneath the Hl6a (4.82ppm) and H16b (4.84ppm) resonances) a *H NMR spectrum of rubifol (175) was obtained in deuterated benzene (CgDg). The chemical shift assignment and the multiplet coupling pattern had been based on the observation that irradiation of the H9a resonance at 2.59ppm induced an NOE into the resonance at 4.83ppm (H10). The *H NMR spectrum run in CgDg is shown in Figure 43. As expected most resonances are shifted. Irradiation of the H10 multiplet resonance (4.18ppm) reduced the H9b triplet (J=10.8Hz) at 3.78ppm to a doublet (J=10.8Hz) and sharpened the lactone Hll resonance at 6.40ppm. Thus, the carbinol lactone proton (H10) decoupled as expected and the previously discussed decoupling and NOE results in CDClj were consistent with the proposed assignment. 169 Table 23. Spin-spin decoupling results for the eight-spin system found on Cll to C2 in rubifol (US.) Proton Coupling Proton Multiplicity prior Result3 irradiated coupled to to irradiation H2a vicinal HI bd sh H2b vicinal HI bd sh Hll allylic H13a m sh Hl3a allylic Hll d sh vicinal Hl4a m sh vicinal Hl4b bt sh H13b vicinal Hl4a m sh vicinal Hl4b bt dm H14b vicinal HI bd bs vicinal Hl3a m sh vicinal Hl3b m sh multiplicity after irradiation, sh-means sharpened. A seventh metabolite*, possibly a cembranoid, was isolated only once. It was obtained as a clear oil that was shown by EIMS to possess a molecular formular of C21H28°5 (°bserved m/z 360.1949, requires 360.1937), which required eight units of unsaturation. The *H NMR spectrum (Figure 44.) consisted of rather broad peaks, suggesting that the metabolite existed in at least two low energy conformations in solution. A *H NMR methyl resonance at 3.18(s, 3H)ppm suggested the existence of a methyl ether. The low intensity mass spectral fragment at a m/z of 345 daltons, corresponding to the loss of a methyl, supported the * Referred to as the/an unknown diterpene in this thesis. 170 notion. Due to the lack of material (2.2mg) and the conformational 13 problem, it was possible to obtain only a partial C NMR spectrum of the unknown diterpene. However, the *H and the 13 available C NMR data did allow for the identification of the following structural fragments: an a.,j^-disubstituted a,f1- unsaturated r-lactone (13C NMR: 80.9, 132.0, 148,6 and 174.1ppm; 1H NMR: 5.17(very bs, IH) and 6.98(bs, lH)ppm); a isopropenyl moiety (13C NMR: 17.8, 112.8 and 143.9ppm; 1H NMR: 1.62(bs, 3H), 4.58(bs, IH) and 4.79(bs, lH)ppm); and a trisubstituted olefin (13C NMR: 126.1ppm; XH NMR: 1.97(d, J=1.5Hz, 3H) and 6.19(bs, lH)ppm). Compare these chemical shifts with those given in Tables 17. and 18. The presence of isopropenyl and trisubstituted olefin moieties was confirmed by two NOE experiments. Irradiation of the methyl resonance at 1.62ppm induced an NOE into the isopropenyl olefinic proton at 4.79(bs, lH)ppm, and irradiation of the methyl resonance at 1.97ppm induced NOE's into the olefinic proton at 6.19ppm and into the second olefinic isopropenyl proton resonating at 4.58(bs, lH)ppm. The broad "bump" on the baseline centered at 3.83(IH)ppm in the NMR spectrum of the unknown diterpene indicated that an 13 alcohol residue may have been present. Also a C NMR resonance at 50.6ppm may have corresponded to the methyl ether carbon, but this resonance would also be appropriate for the carbon at C7 in Figure 45 (a). Not only was the metabolite isolated in small quantity it was also rather unstable and IR, UV and low or high temperature *H NMR spectra were not obtained prior to decomposition. Hence, further evidenc173 e for the presence of the (a). (b). (c). Figure 44. Proposed structures for the unknown diterpene. alcohol etc., was not obtained. Three possible structures for the metabolite are proposed in Figure 45. Extensive NOE and spin-spin decoupling experiments suggested that all three were reasonable candidates. The NOE's previously discussed allowed the isopropenyl and trisubstituted olefin moieties to be placed in the "usual" positions, at CI and C4-C5, respectively, with the C4-C5 double bond possessing a Z configuration. NOE's were observed between the methyl ether protons at 3.18ppm and the methyl resonance at 1.43(s, 3H)ppm. The doublet at 3.15(J=11.7Hz, lH)ppm, as evidenced by both NOE and decoupling results, was the geminal or vicinal neighbour of the resonance at 1.80(td, J=11.7, 0.3Hz, lH)ppm. Irradiation of the olefinic proton at 6.19ppm induced NOE's into the isolated methylene AB quartet resonating at 2.84(d, J=17.9Hz, 1H) and 3.00(d, J=17.9Hz, lH)ppm. The three proposed structures fit this data accordingly. However, each structure exhibits discrepencies. For example, in Figure 45 (a)., the arguments for the presence of a secondary alcohol at C3 are weak and for the structures (b). 174 and (c)., there was no apparent coupling to, or protons that could be readily associated with the C9 position. It was thought that the unknown diterpene might be an artifact of isolation. However, attempts to generate methyl ethers of the cembranoids 171-174. through treatment with methanol in the presence of a catalytic amount of formic acid, have failed to generate this compound. Obviously the structural solution of this unknown diterpene needs further work. (iv) . Hew. CarbOP Skeletal Metabolites. 17_£ and 122: Gersolide (176). isolated as a very minor metabolite (1.5mg) and obtained as colourless needles, was shown by EIMS to possess a molecular formula of C2oH24°4 (°Dserved m/z 328.1674, requires 328.1675) which required nine units of unsaturation. A number of fragments were identified from the IR and *H NMR spectra (Figure 46.). The first fragment, a ai^-disubstituted a»fl-unsaturated Y- lactone, was indicated by the IR band at 1758cm"1 and the "^H NMR resonances at 5.14(d, J=0.6Hz, 1H) and 6.76(bs, lH)ppm. 1H NMR resonances at 2.06(d, J=0.6Hz, 3H) and 5.96(q, J=0.6Hz, lH)ppm, and the IR bands at 1683, 1679 and 1609cm-1 identified a second fragment as a methyl substituted ene-dione. The observation of an NOE between the olefinic proton and the methyl resonance provided further support for the proposed fragment. Finally, the third structural fragment, a isopropenyl group, was evidenced by the 1H NMR resonances at 1.72(d, J=0.6Hz, 3H) , 4.97(bs, 1H) and 5.22(bs, lH)ppm. On irradiating the methyl resonance at 1.72ppm an NOE was observed into the olefinic 175 isopropenyl resonance at 4.97ppm. The three structural fragments thus far proposed were identical with the corresponding substructures in gersemolide (167), rubifolide f170) and epilophodione (171). The spectral data reported is typical of such systems111,157'197,2°3 (see Tables 15. and 18.). The three identified structural fragments ascribed to gersolide (176) accounted for all the oxygen atoms and seven units of unsaturation. Thus, the two degrees of unsaturation not accounted for had to be carbocyclic rings. A 1H NMR resonance at 1.30(s, 3H)ppm required that a methyl substituent be situated at one of the ring junctions. No known diterpenoid carbon skeleton could account for the above features and further spectral data was lacking since only a small quantity of gersolide (176) had been isolated (1.5mg). Crystals of gersolide f176) were obtained by slow evaporation from methanol at 0°C and the complete structure of gersolide (176) was secured via single crystal x-ray diffraction analysis, kindly performed by L. Parkanyi and J. Clardy of Cornell University. A computer generated perspective drawing of the final x-ray model of gersolide (176) is represented in Figure 47. Gersolide (176) represents a diterpenoid with a new rearranged carbon skeleton possessing a 13-membered carbocyclic ring . It should be noted that the x-ray experiment did not define the absolute configuration, so the enantiomer drawn is arbitrary. The a,ft- unsaturated ^-lactone fragment was planar within experimental error. The C4-C5 double bond was not co-planar with either the C6 *We propose the name "gersolane" for the new diterpeniod carbon skeleton of gersolide (176). 177 Figure 47. Computer generated perspective drawing of gersolide (1Z£>. or the C3 carbonyl groups. These two structural observations are consistent with those observed for gersemolide (167) and are also 1 13 consistent with the H and JC NMR chemical shift assignments made for the diterpenes, 167-175. An intensive series of NOE and spin-spin decoupling experiments with reference to a molecular model allowed for the assignment of the majority of the NMR resonances, and provided confirmation of the structure, 176. The NOE results are summarized in Table 24. The H9a cyclopropyl proton (0.98(dd, J=8.3, 5.8Hz, lH)ppm) showed geminal coupling to H9b (1.83(dd, 178 Table 24. NOE results observed for gersolide f176). Proton irradiated Protons with observed NOE H9a H7 H9b Hll H9b H9a Mel7 H2b Hl6a Mel8 H2b H5 H16b Mel9 H5 H9b H10 J=8.3, 5.8Hzf lH)ppm) and vicinal coupling to H7 (1.43(tf J=5.8Hz, H7)ppm), and irradiation of the H2b doublet (J=12.3Hz) at 2.81ppm simplified the H2a triplet at 2.18ppm to a broad doublet (J=12.3Hz). The subsequent "'"H NMR assignments are given in Table 25. Finally, it could be readily demonstrated, with a molecular model, that the methylene H9a and H9b protons (resonating at 0.98 and 1.83ppm, respectively) had to be assigned to the H9a and H9/3 protons, respectively. Rubiformate (177). isolated as a minor metabolite (<4.0mg) and obtained as a clear oil, was shown by EIMS to possess a molecular formula of C20H24°5 (°bserved m/z 344.1623, requires 344.1624) which required nine units of unsaturation. A number of 179 Table 25. H NMR assignments for gersolide f176) and rubiformate (122). Chemical shift, ppm. Chemical shift, ppm. Carbon \ 176 Carbon # 122 1 2.31(tdm, J=12.3,5.2Hz, 1H) 2 1 2.68(m, 1H) 2a 2.18(t, J=12.3Hz, 1H) 2 2.70-2.78(m, 2H) b 2.81(bd, J=12.3Hz, 1H) 5 5.96(g, J=0.6Hz, 1H) 7.03(s, 1H) 7 1.43(t, J=5.8Hz, 1H) / 9a 0.98(dd, J=8.3,5.8Hz, 1H) 8a 2.98(ddm, J=6.6,2.7Hz, 1H) b 1.83(dd, J=8.3,5.8Hz, 1H) b 3.04(ddm, J=6.6,2.7Hz, 1H) 10 5.14(d, J=0.6Hz, 1H) 9 5.29(m, 1H) 11 6.76(bs, 1H) 10 7.11(m, 1H) 13a 2.33-2.48(m, 2H)Z 12a 1.63(m, 1H) b 12b 2.20(m, 1H) 14a 1.25-1.36(m, 1H)Z 13a 2.55-2.65(m, 2H) b 1.98-2.13(m, 1H)Z b 16a 4.97(bs, 1H) 15a 4.69(bs, 1H) b 5.22(bs, 1H) b 4.78(q, J=0.9Hz, 1H) 17 1.72(d, J=0.6Hz, 3H) 16 1.67(bs, 3H) 18 2.06(d, J=0.6Hz, 1H) 17 2.01(s, 3H) 19 1.30(s, 3H) 18 2.23("dn, 3H) 20 9.46(s, 1H) 'assignments within a column may be interchanged. 180 1 13 fragments could be identified from the H and C NMR (Figures 48. and 49.) and other spectral evidence. The first fragment, a Oi^-disubstituted ot,^-unsaturated r-lactone, was identified from 13C NMR resonances at 76.8, 134.2, 147.7 and 173.1ppm and XH NMR resonances at 5.29(m, IH) and 7.11(m, lH)ppm. An IR band at 1758cm"1, typical of such a system111,157,197f203, and the observation that irradiating either proton resonance sharpened the other, were consistent with the proposal. Carbon resonances at 18.4, 113.3 and 145.0ppm and NMR resonances at 1.67(bs, IH) , 4.69(bs, IH) and 4.78(q, J=0.9Hz, lH)ppm identified a second fragment, an isopropenyl residue. Confirmation was provided by decoupling and NOE experiments. Irradiation at 1.67ppm collapsed the downfield quartet at 4.78ppm to a singlet and sharpened the upfield resonance at 4.69ppm. The same irradiation under NOE conditions induced an NOE into the resonance at 4.78ppm. IR absorptions at 1676 and 1606cm"1 identified a third structural fragment as a a,^-unsaturated ketone with the 206 possibility of further conjugation . The proposal was supported by the presence of a carbonyl resonance at 204.5ppm and olefinic resonances at 125.5, 157.8, 119.2 and 150.7ppm in the 13C NMR spectrum of rubiformate (177). Evidence for an enone with extended conjugation was provided by the DV absorption at 297nm (s 16780). Irradiation of the methyl resonance at 2.01(s, 3H)ppm sharpened the olefinic singlet at 7.03ppm. The same irradiation under NOE conditions induced an NOE into this olefinic singlet and into part of the multiplet region resonating between 2.70- 2.78ppm; the induced part appeared as a triplet (7.3Hz) centered 181 at 2.73ppm. On irradiating the isopropenyl methyl resonance at 1.67ppm an NOE was also observed into the triplet at 2.73ppm. This is typical for metabolites 167-175 and is the result of both an isopropenyl and a methyl substituted ene-dione lying in the vicinity of the methylene H2 protons. Hence, the enone was to be treated as a methyl dialkylsubstituted enone. The three structural fragments thus far identified were identical to the corresponding substructures in gersemolide (167). rubifolide f170), epilophodione f171) and gersolide (176). NMR comparisons are made in Tables 15. 16. 17. 18. 25. and 26. Finally, a fourth structural fragment, a methyl substituted enol formate, was apparent from 13C NMR resonances at 9.8, 119.2, 150.7 and 176.8ppm, ^-H NMR resonances at 2.23("d", J=2.lHz, 3H) and 9.46(s, lH)ppm and IR bands at 1720 and 1650cm-1. Compare, 13 for example, the appropriate C NMR resonances of the model 207 compound, norpectinatone (190) . with those of rubiformate (122), Table 27. The IR bands also compare favourably. For example, the carbon-carbon double bond stretch of the enol 206 -1 and tne formate 191 occurs at 1650cm (run in CgH12) carbonyl absorption band of formates is observed in the region 1730- — 1 1 97 1715cm . The presence of the formate moiety was consistent with a MS fragment ion at a m/z of 315 daltons (M+ -29) corresponding to acyl cleavage and the loss of CHO. The preceeding arguments indicated that rubiformate (122) possessed five carbon-carbon double bonds, a ketone, a formate, a lactone carbonyl and a lactone ring. Since no additional unsaturated functionalities were evident, 177 had to possess a second ring to account for the total of nine units of 184 Table 26. C NMR assignments for rubiformate f177) . and spectral comparison with epilophodione f171). Chemical shift, ppm. Chemical shift, ppm. Carbon # 121a Carbon # 121° 1 45.5Z 1 41.8(CH) Z 2 46.5 2 45.5(CH2) 3 204.5 3 205.5(C) 4 157.8 4 143.9(C)2 5 125.5(br) 5 133.5(CH) 6 150.7 ./ 7 119.2 / Z 8 31.2 9 43.6(CH2) 9 76.8 10 78.4(CH) 10 147.7 11 148.1(CH) 11 134.2 12 136.2(C) Z y 12 29.8 ' 13 21.0(CH2) Z 13 30.5 14 33.5(CH2) 14 145.0 15 145.4(C)2 15 113.3 16 115.6(CH2) 16 18.4 17 17.2(CH3) y y 17 23. l 18 21.1(CH3) y 18 9.8 19 22.3(CH3) 19 173.1 20 173.6(C) 20 176.8 / in this column "br" at C5 (125.5(br)) means broad signal. cont'd. 185 Table 21. continued. bin this column "/" means no advantage in comparing, z,vassignments within a column may be interchanged. 0 191 190 i 3 Table 27. C NMR data comparison of 177 and 190. Chemical shift, ppm. Chemical shift, ppm. Carbon # 177 Carbon # 190a 5 125.5 9 126.1 6 150.7 10 159.2 7 119.2 11 106.5 18 9.8 19 8.7 areference 207, spectrometer frequency not reported, CDCl-j. unsaturation. This carbocyclic ring had to accomodate five 13 remaining unassigned aliphatic carbons ( C NMR: 29.8, 30.5, 31.2, 45.4 and 46.5ppm). The new rearranged degraded cembrane diterpenoid structure, 177. possessing a 13-membered ring, 186 effectively accomodated all the identified structural fragments. log A series of NOE, spin-spin decoupling and SINEPT NMR experiments verified that 177 represented the correct structure. As discussed above, the isopropenyl and methyl substituted enone moieties could be placed at their respective positions through the observance of NOE's into one of the H2 protons (2.70- 2.78ppm). Irradiation of the methyl protons resonating at 2.01ppm (Mel7) also induced a small NOE into the two proton multiplet y region resonating between 2.55-2.65ppm, which presumably should be assigned to the H21, HI, or the H12 or H13 methylene protons. Decoupling experiments (to be discussed) demonstrated that the multiplet should be assigned to H13a, H13b. Irradiation of the olefinic proton resonating at 7.03ppm (H5) not only induced an NOE into Mel7 (2.01ppm), confirming the assignment of the Z configuration given to the C4-C5 double bond, but also the formate proton (H20) at 9.46ppm. The reverse result also held true. Hence, the placement of the formate at the C6 position was confirmed. Irradiation of the multiplet resonance at 2.20ppm (Hl2b) markedly sharpened the H12a multiplet (1.63(m, lH)ppm) situated between the water (1.56ppm) and- Mel7 resonances and as a result of allylic coupling sharpened the olefinic lactone resonance at 7.11ppm (H10), and via homoallylic coupling sharpened the lactone carbinol (H9) resonance at 5.29ppm. Irradiation of the Hl2a multiplet (1.63ppm) markedly sharpened its geminal partner, H12b (2.20(m, lH)ppm) and sharpened the multiplet resonating between 2.55-2.65ppm (H13a, Hl3b). Irradiation of the lactone carbinol resonance at 5.29ppm (H9) sharpened the olefinic H10 resonance at 7.11ppm and the two 187 methylene protons at 2.98(dd"m", J=6.6,2.7Hz, H8a) and 3.04(ddm, J=6.6,2.7Hz, H8b)ppm to two broad multiplets. Irradiation of either of the H8 protons sharpened both the lactone H9 resonance (5.29ppm) and Mel8 (2.23ppm). The observance of coupling between the Mel8 and H8 protons was consistent with the couplings 145 reported by M. D'Ambrosio et al. between the Mel9 and H9 protons in the cembranoid metabolites £2 and ££. This presumably can be accounted for by W-coupling between the Mel8 and H8 protons. The proton connectivity about the ring was consistent with the proposed structure, 177. The assignment of an E configuration to the C6-C7 double bond was consistent with the result that NOE's were observed between the formate and olefinic H5 resonance but not the Mel8 protons, ie., the formate proton (H20) points toward the H5 proton and away from the methyl protons (Mel8). A molecular model possessing both; the same relative configuration at the CI and C9 carbons as the appropriate carbons in gersemolide (167) and gersolide (176? • and the structural constraints imposed by the NOE results, demonstrated that when the C6-C7 bond has an E configuration the Mel8 protons need not nessessarily lie in the vicinty of any other protons. However, if the C6-C7 bond had the opposite, Z configuration the Mel8 protons would have to lie in close proximity to at least the lactone carbinol proton (H9). On irradiating the methyl resonance at 2.23ppm (Mel8) no NOE's were observed. These results though only providing negative evidence are consistent with the proposal of an E configuration. The proposed arrangement also minimises the 188 steric hinderence between the "bent" formate chain and the Mel8 protons. 199 Four SINEPT experiments, optimised for polaristion transfer through 7Hz coupling, provided further evidence for the proposed constitution of rubiformate (122) and confirmed several 13 C NMR assignments. Irradiating the formate proton at 9.46ppm (H20) resulted in three bond polarisation transfer to the carbon resonance at 150.7(C6)ppm and four bond transfer into the resonance at 119.2(C7)ppm. A second irradiation at 7.03ppm (H5) showed two bond transfer to the carbon resonances at 157.8(C4) and 150.7(C6)ppm. Irradiation of the methyl resonance at 2.23ppm (Mel8) showed two bond polarisation transfer to the carbon resonance at 119.2(C7)ppm. Hence, the SINEPT results confirmed the structural constitution by connecting the enol formate and the enone moieties through the C5-C6 bond and unambiguously established the presence of a methyl group at the C7 position. A final experiment involved irradiation of the H10 olefinic lactone proton (7.11ppm) and resulted in polarisation transfer into the carbon resonances at 77.8(C9), 134.2(C11) and 173.1(C19)ppm. Proposed *H and 13C NMR assignments are included in Tables 25. and 26. Several observations in the NMR spectra indicated that rubiformate (122) exists in two low energy conformations in solution. The methyl resonance at 2.23ppm (Mel8) appeared as an apparent "doublet". However, irradiation of this resonance did not sharpen a signal with an equivalent coupling. Similarly, sequential irradiation of all resonances in the spectrum did not reduce the methyl signal to a singlet. At 400MHz the methyl 189 "doublet" (Me-18) had a "coupling constant" of 2.7Hz as compared to 2.1Hz at 300MHz and at 80MHz the coupling was not observed at all. The olefinic lactone resonance at 7.11ppm (H10) appeared as a multiplet possessing a large coupling of approximately 5.8Hz, the resonance was more complex and broader than that observed for metabolites 167-176. In addition to these ambiguities the H8 methylene protons at 2.98 and 3.04ppm did not decouple to simple doublets on irradiating the lactone carbinol resonance at 5.29ppm (H9). In the *H NMR spectrum obtained in deuterated benzene (CgDg) (see Figure 50.) the Hl6a isopropenyl olefinic resonance at 4.62ppm appeared as an apparent "doublet" of multiplets with j=5.7Hz. No coupling partner for J=5.7Hz could be located by decoupling experiments. The H8a and H8b protons, both appearing as two sets of two pairs of doublet of doublets, resonating between 1.69-1.80ppm (for both pairs J*s=7.2,5.1Hz) and between 2.11-2.22ppm (downfield dd, J=10.8,6.3Hz, and upfield dd, J=ll.l,6.3Hz), respectively, were coupled only to their geminal H8 partner and the H9 lactone proton (4.78ppm). Ambiguities, in the methyl signals resonating between 1.46-1.58ppm and the olefinic lactone resonance at 6.52ppm (H10) were also noted. In both CgDg and CDCI2 the ratio of the two conformers was approximately 5:4 (evidenced by the NMR integrations). In the 1 3 C NMR spectrum the C5 resonance at 125.5ppm appears as a broad signal and many other resonances appear as two signals lying virtually on top of one another. It is unlikely that the problem results from the presence of two closely related metabolites, 190 such as a pair of diasteriomers, for one would expect the 13 chemical shifts of at least some of the C NMR resonances to be significantly different as was noted for the two metabolites epilophodione (171) and lophodione (188)Also during the isolation there was no evidence that two compounds were present in the rubiformate (177) fraction. The existence of two conformations has a precedent; it was noted that both gersemolide (167) and isoepilophodione B (173) existed in two low energy conformations. Variable temperature ^"H NMR experiments would be informative to the problem, but were precluded by the decomposition of the metabolite. Chemical studies aimed at interconverting rubiformate (177) to a crystalline ene-dione or methyl enol ether, for the purpose of x-ray analysis, were unsuccessful. Rubiformate (177) was treated, under both reflux and non-reflux conditions, with potassium carbonate/"wet" methanol or "wet" acetone followed by the subsequent addition of the methylating agent, methyl iodide. Under the various conditions employed decomposition of rubiformate resulted, to yield a complex mixture of very polar compounds which were not characterised further. The structure of rubiformate (177) can only be viewed as a tentative proposal; an unambiguous structural proof, through x-ray analysis of a crystalline derivative or synthesis, is required (v). Sesquiterpene 178; (+)-7?-Cubebene-3-acetate (178). obtained as a clear oil, did not give a parent ion in the mass spectrum. However, a carbon 192 s 15 r 160 140 120 100 80 60 40 20 PPM 0 Pigure 51. C NMR spectrum of {+) -yj-cubebene-3-acetate (122.). 13 count in the C NMR spectrum (Figure 51.) of 178 indicated the 205 presence of seventeen carbons. An APT experiment demonstrated that metabolite 178 consisted of three quaternary, six methine, four methylene and four methyl carbons (see Table 28.). The IR —l 13 absorption at 1732cm , the C NMR resonances at 21.6(CH3), 76.1(CH) and 170.4(C)ppm and the resonance at 5.39(d, J=8.0Hz, lH)ppm in the NMR spectrum of 178 (Figure 52.) indicated the presence of a secondary acetate. In the MS of 178. a fragment ion at a m/z of 202 daltons (C15H22, observed m/z 202.1713, requires 202.1722), corresponding to the loss of acetic acid via a McLafferty rearrangement, confirmed the presence of the acetate. It is known that loss of acetic acid is particularly facile for 197 allylic acetates on six membered rings . The preceeding arguments indicated that (+)-/9-cubebene-3-acetate (178) had a molecular formula of ci7H26°2* Subtraction °f the acetate moiety from this molecular formula left fifteen carbons, thus 178 was probably a sesquiterpene acetate containing five units of unsaturation. A number of other structural fragments were apparent from 13 spectral evidence. C NMR resonances at 109.0(CH2) and 152.8(C)ppm and the two broad one proton singlet resonances at 5.01 and 5.05ppm identified an exocyclic methylene. A shielded one proton multiplet resonating at 0.58ppm tentatively suggested the presence of a cyclopropyl group. In the MS, a base peak was observed at a m/z of 159 daltons corresponding to the loss of a C3Hg fragment (m/z 43) from the C15H22 fragment ion. The three methyl signals resonating between 0.90-0.96ppm in the *H NMR spectrum all appeared as doublets193, and hence the C^Hg fragment Table 28. C NMR assignments for ( + )-/?-cubebene-3-acetate (178). Chemical shift, ppm. Chemical shift ppm. Carbon # Carbon # 1 37.4(C) 10 30.4(CH)Y Z 2 38.9(CH2) 11 109.0(CH2) 3 76.KCH) 12 26.5(CH)Y X 4 152.8(C) 13 18.8(CH3) Y X 5 44.2(CH) 14 19.8(CH3) Y X 6 34.6(CH) 15 20.0(CH3) 7 33.5(CH)Y 16 170.4(C) Z X 8 31.1(CH2) 17 21.6(CH3) Z 9 31.2(CH2) Assignments may be interchanged. probably resulted from the loss of an isopropyl residue. Two units of unsaturation were accounted for by the acetate carbonyl and the olefinic exocyclic methylene. Hence, (+)-/?- cubebene-3-acetate (178) had to possess three carbocyclic rings, one of which was assigned to a cyclopropyl ring. The sesquiterpenoid cubebene structure 178 effectively accomodated all the identified structural fragments. The connectivity from the H2 methylene protons through to the Hll protons was established by a series of decoupling experiments. Irradiation of the "^H NMR carbinol resonance at 5.39(d, J=8.0Hz, H3)ppm sharpened 195 the exocyclic methylene singlet resonances (5.01 (Hlla) and 5.05ppm (Hllb)) and reduced the one proton doublet of doublets (J=14.4f8.0Hz) at 2.38ppm (H2b) to a doublet (J=14.4Hz). These results demonstrated an allylic coupling between H3 and the exocyclic methylene protons (Hlla and Hllb). Irradiation of H2b (2.38ppm) reduced the carbinol resonance at 5.39ppm (H3) to a singlet and collapsed the part of the multiplet resonating between 1.55-1.67ppm possessing the large doublet coupling of 14.4Hz. The magnitude of the coupling suggested that H2b possessed a geminal partner. The geminal partner, H2a, had only a small coupling (<0.5Hz) to the vicinyl H3. On irradiating at 1.62ppm (H2a) the carbinol doublet at 5.39ppm (H3) sharpened and the H2b doublet of doublets (2.38ppm) simplified to a doublet (J=8.0Hz). Thus, (+)-/?-cubebene-3-acetate (178) possessed the above described five-spin system and the exocyclic methylene and acetate moieties were placed at their appropriate positions in structure 178. (+)-^-Cubebene-3-acetate (178) rapidly decomposed to give a complex mixture of very polar material when it was exposed to NMR solvents, other manipulations, or when left standing. Hydrogenation of 178 resulted in a mixture of two dihydro epimers that proved to be very stable compounds. The epimers, which were not readily seperable, were obtained in a ratio of 4:1 as evidenced by the 1H NMR spectrum (Figure 53.). The fragmentation patterns of the two epimers were virtually identical on GCMS analysis, however, neither epimer gave a parent ion. In the EIMS and FAB MS of the mixture, very low intensity M+ peaks (rel., intensity <0.6) at m/z of 264 daltons were observed. Fragment 197 peaks at m/z of 204 and 161 (100%) daltons corresponded to the loss of acetic acid followed by the subsequent loss of the isopropyl residue. AcO. AcOv 192 193 A series of NOE and spin-spin decoupling experiments verified that 178 represented the correct structure for the natural product (+)-f?-cubebene-3-acetate and that the dihydro epimers 192 and 193 represented the major and minor products, respectively. Irradiation of the H3 triplet resonance (J=7.2Hz), of the major epimer 192. at 5.12ppm reduced H2b at 2.33(dd, J=14.8f 7.2Hz, lH)ppm to a doublet (J=14.8Hz) and sharpened the H4 multiplet at 2.47ppm. Equivalent decouplings were apparent for the minor epimer, 193. upon irradiating at 4.67ppm (H3). Irradiation of the cyclopropyl methine resonance at 0.54ppm (H6) markedly sharpened the obscured H5 "multiplet" at 0.83ppm and sharpened the H7 multiplet at 1.38ppm. Irradiation at 0.83ppm (H5) sharpened both the H6 multiplet at 0.54ppm and the methine H4 multiplet resonance at 2.47ppm. As expected, irradiation of H4 (2.47ppm) markedly sharpened H5 (0.83ppm), simplified the carbinol H3 triplet (5.12ppm) to a doublet and collapsed the Mell doublet (J=6.0Hz) at 0.97ppm to a singlet. Hence, connectivity was extended from the methylene H2 protons to the methine H7 199 proton for both epimers. Subsequent NOE experiments (the results are summarized in Table 29.) allowed the relative stereochemistry from the H2b proton through to the H5 proton to be defined for both the major and minor dihydro epimers 192 and 1£3_, respectively, and hence that of (+)-/?-cubebene-3-acetate (178) was also established. Table 29. NOE results observed for the dihydro epimers, 192 and 122.. Protons with observed NOE in: Proton irradiated 192 193 H2b H3 H3 Mel5 ? H3 H2b H2b H4 X H4 H3 X H5 ? Mell ? ?-not possible to say whether NOE resulted. X-no equivalent NOE resulted for 193. A H NMR spectrum of the epimer mixture, 192 and 193. was obtained in deuterated benzene (CgDg) (see Figure 54.). Many resonances were shifted, particularly the H5 and H6 cyclopropyl methine resonances at 0.64(m, 1H) and 0.50(m, lH)ppm, respectively. Irradiation of the methyl resonance at 0.86(d, J=7.2Hz)ppm (Mel5) induced NOE's into the H5 resonance at 200 s Figure 54. H NMR spectrum of the mixture of dihydro epimers 192 and 121 in CCDC. 0.64ppm, the methylene H2b resonance at 2.18ppm and into part of the two multiplet regions resonating between 1.35-1.40 and 1.48- 1.66ppm. The H2b proton possessed a /9-configuration as did H5 and hence, the relative configuration at CIO was defined, i.e., Mel5 also possessed a ft-configuration. Irradiation of the methine H6 resonance at 0.50ppm induced an NOE, after careful inspection, into the methine H4 resonance of the minor epimer 193 at 2.20ppm. The result was consistent with the observation that no NOE existed between the two cyclopropyl protons, H5 and H6, ie., H6 possessed an a-configuration and hence, the two protons were disposed in a trans arrangement. The preceeding arguments were totally consistent with the gross structural constitution of 178. NOE experiments and coupling constant arguments established the relative configuration at all of the chiral centers except C7. The relative configuration given to the C7 chiral carbon and represented in structure 178 was consistent with the result that no NOE was observed between the methine H6 resonance, centered at 0.54ppm, and the H7 methine multiplet at 1.38ppm (assignment based on a decoupling result, see above). The configurational assignment at C7 was also consistent with the observation that the three 13C NMR methyl resonances of C13, C14 and C15 were virtually identical with the appropriate carbon resonances reported for (-)-epicubebol (121)208, (+)-cubebol (125J209 and 210 (±)-/?-cubebene-nor-ketone (196) (see Table 30.). Since 13 . C NMR is an excellent method for comparing the relative stereochemistries of similar molecules, one would expect that if the relative stereochemistry at either the C7 or CIO carbons were 202 194 195 196 3 3 Table 30. C NMR data comparison of the methyl resonances of Hfir ISA, 1£5_ and 19_£. Chemical shift, ppm. Carbon # 12£ 194a 195b 196c 13 18.8Z 19.lz 18.7Z 18.9Z 14 19.8Z 19.8Z 19.6Z 19.5Z 15 20.0Z 20.0Z 20.0Z 20.0Z a reference 208, 100MHz, CDC13. ^reference 209, spectrometer frequency not reported, CDCl^. cauthentic sample kindly provided by Dr. E. Piers of the University of British Columbia. Assignments within a column may be interchanged. different the chemical shifts of the C13, C14 and C15 methyl resonances would be shifted by a ppm or two. Sesquiterpenes isolated from soft corals and in fact from coelenterates in general tend to exist as the optical antipod of the form found in terrestrial and marine plants. This is exemplified by the two metabolites 194 and 195. isolated from the 203 197 198 199 brown alga Dictvopteris divaricatazuo and a soft coral, a 209 Cespitularia species , respectively. They exist as the optical antipodes of one another (ignoring the stereochemistry at C-4, 211 (-)-cubebol is known from terrestrial plants ) and are spectroscopically identical. Similarly, a- and /?-cubebene (197 and 198. respectively) isolated from the oil of Piper cubeba L. 212 both showed negative optical rotations and (+)-a-cubebene (199) isolated from gorgonians of the genus Pseudoplexaura 98 possesses a positive optical rotation . Thus, since (+)-/?- cubebene-3-acetate f178) has a positive optical rotation and is derived from a soft coral it is suggested that its absolute configuration is as represented in structure 178. There were a few anomalies in the NMR spectra (Figures 52-54., and Table 31.). A number of the coupling patterns could not readily be rationalised. An example is provided by the 21 cyclopropyl H6 multiplet (0.58ppm). An authentic sample of 196 had an identical coupling pattern for H6. The complexity was probably the result of observed long range (3 and 4 bond) couplings. These couplings may be attributed to the extreme rigidity of (+)-/?-cubebene-3-acetate (178). easily demonstrable 204 by a molecular model, which results in a variety of possible W- Table 31. H NMR assignments for (+)-/?-cubebene-3-acetate (178) and the dihydro epimer 192. Chemical shift, ppm. Carbon # 122. 122a 2a 1.55-1.67(m) 1.52-1.65(m) b 2.38(dd, J=14.4, 8.0Hz, IH) 2.33(dd, J=14.8,7.2Hz, IH) 3 5.39(d, J=8.0Hz, IH) 5.12(t, J=7.2Hz, IH) 4 / 2.47(m, IH) 5 0.80-0.90(obscured) 0.83(m, IH) 6 0.58(m, IH) 0.54(m, IH) 7 1.55-1.67(m)z,y 1.38(m, IH) 8 1.37-1.48(m, 2H)y ? 9 1.05-1.15(m, 2H)y ? 10 1.27(m, lH)z'y ? 11a 5.01(bs, IH) 0.97(d, J=6.0Hz, 3H) b 5.05(bs, IH) / 12 1.74(m, IH) 1.69(m, IH) 13 0.90-0.96(3 x d's) 0.88-1.03(m) 14 0.90-0.96(3 x d's) 0.88-1.03(m) 15 0.90-0.96(3 x d's) 0.94(d, J=6.7Hz, 3H) 17 2.01(s, 3H) 2.01(s, 3H) awithin this column could not assign "?". z,yassignments within a column may be interchanged. coupling arrangements. In a rigid system W-coupling can be as 197 1 13 large as 7HzA . Proposed H and x C NMR assignments for (+)-/?- 205 cubebene-3-acetate (178) are included in Tables 28. and 31.. The proposed NMR assignments for the major dihydro epimer 192. are also included in Table 31. (vi) . Pukalide (£1) : Pukalide (£3.) , obtained as a white crystalline solid, was identified by comparison of its physical and spectral data with literature values114 and authentic IR and 1H NMR spectra kindly provided by Dr. P. J. Scheuer of the University of Hawaii. (vii) . Briarein diterpenoids148, U3. and. 1M: Ptilosarcenone (179). obtained as an amorphous solid, was identified by comparison of its physical and spectral data with 213 1 literature values and an authentic H NMR spectrum kindly provided by Dr. D. J. Faulkner of the Scripps Institution of Oceanography. Butanoate 178 was obtained as a glass that showed a parent ion in the EIMS at a m/z of 508.1873 daltons corresponding to a molecular formula of C2gH33ClOg (requires a m/z of 508.1865 daltons) which required ten units of unsaturation. Spectral data, particularly the 1H NMR spectrum (Figure 55.), indicated that 180 was closely related to ptilosarcenone (12SJ (see Table 32). However, one of the acetate resonances in the 1H NMR spectrum of 179 was replaced in the spectrum of 180 by resonances at 0.99(t, J=7.2Hz, 3H), 1.71(m, 2H) and 2.38(t, J=7.2Hz, 2H)ppm which were assigned to a butanoate ester. A series of NOE experiments demonstrated that the butanoate 206 207 Table 32. H NMR assignments for butanoate 180 and spectral comparison with 179. Chemical shift, ppm. Carbon # lfifl. 179a 2 5.79(df J=8.7Hzf IH) 5.72(dr J=9Hz, IH) 3 5.60(dd, J=12.4,8.7Hz, IH) 5.63(dd, J=12,9Hz, IH) 1 4 S^-S^fm) 5.99(d, J=12Hzr IH) 6 5.23(dm, J=3.9Hz,lH) 5.26(ddd, J=4,2,lHz, IH) 7 4.98(dr J=3.9Hz, IH) 5.01(d, J=4Hz, IH) 9 5.46(d, J=7.5Hzf IH) 5.49(d, J=7Hzf IH) 10 2.73(m, IH) 2.82(m, IH) 11 2.79(mf IH) 2.82(m, IH) 13 5.87(d, J=10.4,0.9Hz, IH) 5.90(d, J=llHz, IH) 14 6.58(d,J=10.4Hz, IH) 6.61(d, J=llHzf IH) 15 1.17(s, 3H) 1.18(s, 3H) 1 16a S^-S^dii) 5.96(dd, J=2f lHzflH) b 6.12(dm, J=0.9Hz, IH) 6.12(ddf J=l,lHz, IH) 17 2.36(q, J=7.2Hzf IH) 2.41(qf J=7Hz, IH) 18 1.20(d, J=7.2Hzf 3H) 1.21(d, J=7Hz, 3H) 20 1.30(df J=7.2Hz, 3H) 1.30(df J=7Hzf 3H) 22 2.38(t, J=7.2Hz, 2H) 2.15(s, 3H)j 23 1.71(m, 2H) / 24 0.99(t, J=7.2Hzf 3H) / 26 2.20(s, 3H) 2.20(s, 3H)^ -OH 3.42(s, IH) 3.43(s, IH) cont'd. 208 Table 32. continued. areference 213a, 220MHz, solvent not reported. 1signals lie partially on top of one another. •'chemical shifts obtained from my spectra, and may be interchanged. The assignment of the resonance at 2.20ppm should be to the Me24 protons in metabolite 179. ester in 180 was attached to C2, while the acetate ester was attached to C9. Irradiation of the acetate methyl resonance at 2.20ppm (Me26) induced NOE's into the Mel5, H3 and H7 protons resonating at 1.17, 5.60 and 4.98ppm, respectively. Simultaneous irradiation of the butanoate a-methylene (H22's) proton resonance at 2.38ppm and the H17 resonance at 2.36ppm induced NOE's into the resonances at 4.98 (H7, from H17), 5.46 (H9, from H17) and 1.20ppm (Mel8, from H17). The irradiation also induced NOE's into the resonances at 6.58 (H14, from the butanoate a-CH2), 1.71 (methylene H23, from the butanoate a-CH2) and 0.99ppm (Me24, from the butanoate e*-CH2). Hence, the butanoate and the acetate had to be placed at the C2 and C9 positions, respectively. Irradiation of the hydroxyl -OH proton (3.42ppm) attached to C8, demonstrated structural consistencies by inducing NOE's into the methyl resonance at 1.20ppm (Mel8) and the H10 and Hll methine multiplets at 2.73 and 2.79ppm, respectively. Proposed 1H NMR assignments are included in Table 32. 209 D_. Biological Activities. Initial interest in the chemistry of Gersemia rubiformis was sparked by the potent in vitro antimicrobial activities shown by crude methanolic extracts. None of the pure sesquiterpenes and diterpenes, which were abundant enough to be tested, showed appreciable antimicrobial activity. It was not possible to test the very minor or unstable and quite possibly the most interesting metabolites in any of the bioassays ( A4,5Z, A7'17- isogersemolide (168). A4,5E, A7'17-isogersemolide (169) . the unknown diterpene, gersolide (176) and rubiformate (177) were not tested). Rubifolide (170) and epilophodione (171) showed no insecticidal activities in a confused beetle bioassay (Tribolium confusum), and although both 170 and 171 showed mild herbicidal activity in a Lemna paucicostata (duck weed) bioassay, when tested on higher plants no activity was observed. The metabolites 165-177 isolated from Gersemia rubiformis were present in relatively high concentrations (see Experimental for comprehensive details). The total percentage yield of —2 terpenoid metabolites was 2.8 x 10 %. Hence, it seems reasonable to expect that a number, if not all of these metabolites would exhibit a biological role/activity. Of the metabolites tested, dihydrotochuinyl acetate (166) exhibits fish antifeedant activity. Metabolite 166 completely inhibited feeding at a concentration of 0.15mg/pellet (0.0056mg/mg of food pellet) though there was a marked preference for the control at concentrations as low as 0.005mg/pellet (0.0002mg/mg of food pellet). Since the concentration of metabolite 166 per mg of £. 210 rubiformis was O.Ollmg and the average concentration obtained from Toauina tetraquetra per animal was 0.35mg (C.00004mg/mg of nudibranch), the experimental inhibitory concentration is very comparable. Thus, it would appear that £. rubiformis utilises a defensive allomone and the highly specialised feeder, the nudibranch, T_. tetraquetra has learned to ingest and sequester metabolite 166 for the purpose of its own defense. This observation possibly provides an explanation for the marked 122a dietary dependency of the nudibranch 203 Diterpenes, such as pseudopterolide (185) , kallolide A (1122)157 and lophotoxin (2M)109, that are structurally related to metabolites 167-177 exhibit diverse bioactivities. 203 Pseudopterolide (185) shows potent cytotoxic activity, 157 kallolide A (107) exhibits potent antiinflammatory activity and lophotoxin (200) f co-occurring with pukalide (£3.), is a i no neuromuscular toxin . The diterpenes 167-177 will likely yield similar results. The preliminary in vitro cytotoxic activities of rubifolide (170) and epilophodione (171) are reported in Table 33. Epilophodione (171) is currently undergoing in vivo screens. CH0 211 Table 33. Results of in. vitro cytotoxicity assays for rubifolide (HQ.) and epilophodione (171) . (IC-50, ng/mL of active dilution). Cell lines: Metabolite A549-human lung murine melanoma human colon 170 na na na 171 27 24 5.8 Moser-human lung SWl271-human lung 170 na na 171 6.8 6.6 na-no significant activity at maximum concentrated tested. Biological activities of the novel allylic acetate, (+)-/?- cubebene-3-acetate (178) were precluded due to its instability, and for similar reasons pukalide (£3_), ptilosarcenone (179) and the butanoate 180 were not tested. £.. Biosvnthetic Proposals. A biogenetic pathway for tochuinyl acetate (167) and dihydrotochuinyl acetate (166), proceeding through the epoxide containing intermediate 201. is proposed in Scheme 1. The pathway involves the conversion of farnesylpyrophosphate (2112) to the bisabolene intermediate 203. Generation of the parent alcohol 204 followed by acetylation, leads to lfLfc. Dehydrogenation followed by acetylation readily gives 1£5_. It may be that the 212 aromatic metabolite 165 is an artifact. The cembrane skeleton f205) can obviously be generated from geranylgeranylpyrophosphate f206), as illustrated in Scheme 2. 213 Scheme 2. Biogenetic proposals for cembrane, pseudopterane, gersolane and degraded 13-membered ring diterpenoids, 167-177. 214 Biological oxidations and other secondary transformations would eventually give metabolites 171-175. and subsequent reduction of an ene-dione moiety followed by condensation would lead to a furan-containing skeleton as required by rubifolide f170). The co-occurrence of both ene-dione and furano- functionalities was previously noted in the gorgonian Lophogorgia alba from which the ene-dione metabolites lophodione (188) and isolophodione f189) and the furan-containing metabolites pukalide (£3_) and lophotoxin (200) were isolated1^. It has been suggested that the metabolites 188 and 189 may represent possible precursors to the furan-containing cembrenolides. 203 In their report of pseudopterolide (185). Fenical et al. suggested that the pseudopterane carbon skeleton, which superficially looks like a simple dimerization of two geranyl residues, might instead arise from ring contraction of a cembranoid precursor. The gersolane skeleton of 176 and the proposed skeleton of rubiformate (180), like the pseudopterane skeleton of 185, also appear superficially to have been formed by coupling of two geranyl units with subsequent secondary transformations. Since Gersemia rubiformis contains similarly fuctionalized diterpenoids having cembrane, pseudopterane and gersolane skeletons etc., it seems more likely that the latter two (or three) skeletons arise from rearrangements of a cembrane precursor*90*3'203, as illustrated in Scheme 2. Epilophodione (171) and gersemolide (167), for example, are in principle directly related by such a contraction (through an allylic rearrangement), although it is likely that the contraction would take place before complete functionalization of the hydrocarbon 215 skeleton (see Scheme 2.). The presence of both ene-dione and furano- functionalities in pseudopterolides was noted in the 157 kallolides A-C (107. 1B6P 1£2) , and hence the interconversion of pseudopterane metabolites to rubifolide (170) or visa versa also appears feasible. (+)-/?-Cubebene-3-acetate (17g) could be generated by folding of a farnesylpyrophosphate precursor with subsequent oxidation at the appropriate carbon in the manner represented in Figure 56. However, the precursor from which cubebanes are considered to be derived is the cadinane skeleton (207), which is itself derived 214 from either the bisabolene or germacrene carbonium ions (2M and 209. respectively) shown in Scheme 3.. Figure 56. Biogenetic speculation for (+)-/?-cubebene-3-acetate (12fi>. £. Further Observations and Discussion. Tochuinyl acetate (165) and dihydrotochuinyl acetate (166) represent, to the best of our knowledge, the first examples of cuparane sesquiterpenes isolated from a soft coral. The cuparane sesquiterpene skeleton, however, is well documented in the 216 Scheme 3. Biogenetic proposal for (+) -f?-cubebene-3-acetate 210 marine environment, particularly from red algae of the genus 215 Laurencia . In nearly all cases the metabolites are 216 halogenated, most often with bromine . a-Bromocuparene 217 (210) . for example was isolated from L. glandulifera. Cuparanes isolated from molluscs, for example the sea hare Aplysia dactylomela62b. are known, but these are probably derived from algal dietary sources. The isolation of diterpenes 167-177 from Gersemia rubiformis 217 collected off the west coast of Canada is of interest for a number of reasons. The pseudopterane diterpenoids, only recently 157 203 discoverd by Fenical's group ' , represent a family of Alcyonarian metabolites that have a previously unknown carbon skeleton. To date, the pseudopteranes have been taxonomically restricted to the order Gorgonacea. Hence, the pseudopteranes 167-169 are the first examples isolated from any source other than a gorgonian. Secondly, G_. rubiformis represents the first organism to contain both pseudopterane and cembrane diterpenes, a fact that has biosynthetic implications (see preceeding section). Finally, fi. rubiformis contains gersolide (176) and rubiformate (177). the first examples of diterpenes (though 177 would appear to be a degraded diterpene) possessing new rearranged carbon skeletons with 13-membered rings. The British Columbian specimens of Gersemia rubiformis were collected on four occasions. The collections were made in May 1986, April 1987 (spring), August 1985 (summer) and October 1986 (fall). On each occasion, all the metabolites 165-177 were present. The unknown diterpene was isolated only from the May collection. The percentage yields of the metabolites were at a maximum for the collections made in August and October (quoted in the Experimental). All percentage yields were at least halved in the two spring collections. The yields of the two pseudopterolides, 168 and 169. isoepilophodione A (122)t rubifol (175), gersolide (176) and rubiformate (177) were reduced to trace quantities (<0.2mg). Such observations imply that there is a "slow down" in metabolic prossesses during the winter months, 218 and that either the metabolites are simply degraded and excreted, or are used for a biological purpose. Rubifolide f170), the major secondary metabolite isolated from G_. rubiformis. is inactive in all the biological screens undertaken thus far (see preceeding section). The position of 170 at the end of possible converging biosynthetic pathways implies that it is the biosynthetic end or "waste" product of the diterpenoid biosyntheses. (see preceeding 1 57 section). However, Fenical et al. have demonstrated that the furan-containing metabolite kallolide A (107) can be converted to kallolide C (187) through singlet photooxidation of 107. This may mean therefore that 187 is not a true secondary metabolite, or on the other hand it may mean that 107 and thus rubifolide (170) are the initial biosynthetic precursors for all the other diterpenoids present in the respective organisms (see Scheme 2.). The two pseudopteranes 168 and 169. the four isomers 171- 174. rubifol (175) and the unknown diterpene are all probably true secondary metabolites. The metabolites 168. 169. 171-174 and 175 were always present from rapidly processed extracts. Also gersemolide (167) only partially isomerised to A4,5Z,A7'17- isogersemolide (168) after extensive reflux in the presence of a catalytic quantity of formic acid, no reaction was observed at room temperature with 48 hours of stirring. Rubifol (175) and the unknown diterpene could not be prepared by treatment of eplilophodione (171), or isoepilophodiones A (172), B (173) and C (174) with water or methanol in the presence of catalytic amounts of formic acid. None of the metabolites, 165-177. isolated from west coast Gersemia rubiformis were found in specimens collected on the east 219 coast of Canada. Instead, the steroidal metabolites JL5£.-1£1 186 previously reported by Kingston et al. were noted, along with the rather unstable sesquiterpene acetate, (+) -/?-cubebene-3- acetate f178), possessing the cubebane skeleton. Cubebane sesquiterpenoids have been previously reported in other 209 Alcyonarian species, for example, (+)-cubebol (195) and (+)-a- 9 8 cubebene (199) were isolated from soft coral and gorgonian species, respectively. The skeleton is also well represented in both terrestrial and marine plants, though generally in these sources they exist as the optical antipod of the metabolites 218 noted in coelenterates . Oxygenation at C3 has, as far as we know, not previously been reported. The role of metabolite 178 in the host organism remains unknown. Its instability precluded bioassay studies. Since the west coast specimens of £. rubiformis possess a chemical defense it is likely that the east coast animals utilize a similar defensive strategy. Hence, sesquiterpene 178 may have a defensive role. Gersemia rubiformis is a species of the family Naphtheidae. The chemistry of this family was reviewed on page 70.. The chemistry of £. rubiformis bears little resemblance to that previously noted for the family. The discovery of interesting metabolites in a cold water soft coral, some of which are virtually identical to those previously reported from tropical soft corals and gorgonians, suggests that although not many species of Gorgonacea and Alcyonacea grow in cold temperate waters the species that do might reasonably be expected to have chemistry similar to their tropica220 l relatives. All of the metabolites present in the skin extracts of Toouina tetraquetra. with the exception of pukalide (£3J , can be directly traced to the coelenterates that make up its diet. Tochuinyl acetate f165), dihydrotochuinyl acetate f166) and rubifolide (170) are all metabolites of the soft coral Gersemia rubiformis which is the major dietary organism of the Port Hardy 122a animals . Pukalide has been reported as a metabolite of the tropical soft coral, Sinularia flbrupta114r and the tropical gorgonians, Lophooorgia rigida109'110. and L. alba111. However, £. rubiformis is the only soft coral reported in British 32 Columbian waters and we have not been able to find pukalide (63) in fi. rubiformis extracts. Thus, the exact origin of pukalide (£3.) remains unknown. It could be an extremely minor metabolite that has eluded us thus far, it could be a metabolite of an as yet unreported soft coral, or it could result from a metabolic transformation carried out by the nudibranch on some £. rubiformis metabolite. Of interest here is the fact that Coll has demonstrated convincingly that in a Sinularia species of soft coral, pukalide (£1) can be found only during one or two days of 219 the year when animals are spawning . Perhaps, therefore, it is not surprising that we failed to isolate pukalide (£1) from extracts of G_. rubiformis. Also, with this precedent it is not so surprising to find that the unknown diterpene was isolated only from the May collection. Further studies into these two observations should be undertaken. Ptilosarcenone (179) has been reported as one of the major metabolites of the sea pen (order Pennatulacea), Ptilosarcus 213 221 gurneyi . which is a component of the diet of Bamfield 18x 9 specimens of Toquina tetraquetra . An exhaustive study of the metabolites of £. gurneyi collected at Seattle, Washington and Sidney, British Columbia failed to uncover the butanoate analogue 180 or any metabolite which could be viewed as a precursor to it . Probably the chemistry of the Bamfield population of £. gurneyi differs slightly from Strait of Georgia and Puget Sound populations. Ptilosarcenone (179) was reported to be generated t 213a facile elimination of butanoic acid from ptilosarcone (211) • To discount the possibility that 179 and 180 were isolation artifacts, a rapid workup of a chloroform extract of freshly collected T_. tetraquetra specimens was carried out. Only diterpenes 179 and 180 were found with no trace of the putative parent compounds, for example ptilosarcone (211). It appears that ptilosarcenone (179) and its butanoate analogue, 180. are the metabolites actually present in the skin extracts of I. 34 tetraquetra. Nudibranchs possess acidic dietary tracts , just the conditions required for the facile elimination of butanoic acid from ptilosarcone (211). Thus, it may be that the 222 metabolites originally ingested by the nudibranch were ptilosarcone (211) and the butanoate 180. Specialist feeders, it was suggested in chapter II, can ingest large quantities of otherwise toxic or "distastful" metabolites in order to gain the benefits of an exclusive food source, etc.. The findings reported here are consistent with this and other suggestions that some dentronotoid nudibranchs feed exclusively on particular coelenterates. Findings indicate that T_. tetraquetra ingests and sequesters the sesquiterpene fish antifeedant, dihydrotochuinyl acetate (166). Since J_. tetraquetra collected at Port Hardy possess a chemical defense it is likely that animals at Bamfield possess a similar defensive strategy. Ptilosarcenone (179) is a known insecticide and an inhibitor 213a of enzyme esterases . Perhaps it is utilised as a defensive allomone by the nudibranch. Ptilosarcenone (179) needs to be tested for fish antifeedant activity. The instability of butanoate 180 precluded biological assays, but one would suspect due to its structural similarity with 179 that it might possess similar activities. 223 YI. EXPERIMENTAL. The H NMR spectra were recorded on Bruker WP-80, Varian XL- 300 and Bruker WH-400 spectrometers. C NMR spectra were recorded on Varian XL-300 and Bruker WH-400 spectrometers. Tetramethylsilane was used as an internal standard for the *H NMR spectra. Deutero-chloroform was used as the NMR solvent unless otherwise stated, and as an internal standard (6=77.Oppm) in the 13 13 C NMR spectra. The hydrogen attachment in the C NMR spectra is given only when it has been determined by an NMR experiment. Low resolution mass spectra were recorded on an AEI MS902 spectrometer and high resolution mass spectra were recorded on an AEI MS50 instrument. Low resolution gas chromatograph mass spectra were run on a Kratos MS80RFA spectrometer and CarloErba 4160 gas chromatograph. Infrared spectra were recorded on a Perkin-Elmer 1710 Fourier Transform spectrometer. Ultra violet- visible spectra were recorded on a Bausch and Lomb Spectronic- 2000 spectrophotometer. Optical rotation measurements were recorded on a Perkin-Elmer 141 polarimeter using a 10cm microcell. A Fisher-Johns apparatus was used to determine melting points and these values are uncorrected. Gas chromatography was performed on a Hewlett Packard 5890A instrument and a flame ionisation detector was used. High performance liquid chromatography (HPLC) was performed on Perkin- Elmer series 2 or Waters 501 liquid chromatograph instruments, and Perkin-Elmer LC55 spectrophotometer or Waters 440 absorbance detector and/or Perkin-Elmer LC-25 refractive index detectors were employed for peak detection. Whatman Magnum-9 Partisil 10 or 224 Magnum-9 ODS-3 Partisil 10 columns were used for the preparative HPLC. The HPLC solvents were BDH OmniSolv grade or Fisher HPLC grade; water was glass distilled; all other solvents were reagent grade. sneets Merck Silica Gel 60 PF-254 and pre-coated 60 F254 were used for preparative thin layer and thin layer chromatography (TLC), respectively. Merck Silica Gel 230-400 Mesh was used for flash chromatography and Merck Silica Gel 60 PF-254 with CaS04.l/2H20 was employed in radial thin layer chromatrography (Harrison Research chromatotron model 7924, FMI lab pump model R PG-150). Sephadex LH-20 resin was used for molecular exclusion chromatography. To prevent potentially interesting metabolites from being discarded all partially purified fractions were broadly characterised using TLC and NMR spectroscopic analysis. When a metabolite is stated as being pure it consists of a 1 13 single peak on HPLC, and its spectral data (MS, H and C NMR etc..) is consistent with this. In the fish antifeedant bioassays goldfish (Carassius auratus) were fed both treated (metabolites H, 12, 165. 166. 167. 170. 171. 173 and 175 were dissolved in dichloromethane and the required volume transferred onto the surface of a pellet (Wardleys shrimp pellets), average weight of a pellet, 27mg) and control pellets (treated with IOJJL of dichloromethane) at the same time. The fish were then observed until all the control had been consumed and for 15-20 minutes thereafter. If the treated pellet had only been nosed or spat out during this period it was 225 considered to be unpalatable. The experiments were repeated three times and the reported results were duplicated. 226 Diaulula sandieaensis. Collection Data. Specimens of D_. sandiegensis (152 animals, 42.6g dry weight, after extraction) were collected by hand using SCUBA (-1 to -15m) on several exposed rocky reefs located in the Deer Group of Islands, Bamfield, British Columbia. Freshly collected animals were immediately immersed in methanol and allowed to extract at 2°C. Fxtraction and Chromatographic Separation. A number of collections of D_. sandieoensis were made. The following represents typical yields and isolation precedures used for a September collection (when the concentration of metabolites 41 and AZ was at a maximum). After storage at 2°C for two days the methanol was decanted and replaced with fresh solvent and the sample was left to extract at 2°C for a further two days. The first and second methanol extracts were combined and evaporated in vacuo. The resulting residue was partitioned between brine (200mL) and ethyl acetate (5 x 125mL). The combined ethyl acetate extracts were dried over sodium sulphate for 1 hour and evaporated in vacuo to give 1.4g of brown oil. Flash chromatography of the brown oil (1:1 acetone/methylene chloride) gave a fraction (90mg) which contained diaulusterols A (11) i B (12) and fats. Sephadex LH-20 chromatography (80% methanol/chloroform, 2.5 x 100cm column) removed most of the 227 fats. The resulting enriched mixture of compounds Al and A2 (47mg) was purified via reverse phase HPLC (70% methanol/water) to yield pure steroids as clear oils (41: 24.0mg; 42: 3.2mg. 0.16mg (0.06% dry weight) and 0.02mg (0.007% dry weight) per animal, respectively). When spotted on TLC both steroids could be visualised by short wave UV light, iodine absorptions, and on exposure to sulphuric acid spray (40% sulphuric acid/methanol) followed by heat green chars turning bright red were observed. Physical. Spectral and Chemical Conversion D_a±a.. 25 Diaulusterol A (Al) : clear oil; [a] D+29.0°(c 1.15, CH2C12); Rf0.18 (EtOAc); UV(MeOH) Xmax 265nm (c 9830); IR(CH2C12) 3380, 2957, 2875, 1722, 1664, 1626, 1459, 1377, 878, and 753cm~1; AH NMR (300MHz) 0.65(s, 3H), 0.96(d, J=6.8Hz, 3H), 1.06(tm, J=9.2Hz, IH), 1.17(s, 3H) , 1.21(d, J=7.0Hz, 3H), 1.30-1.42(m region), 1.46(s, 6H), 1.51-2.0(m region), 2.08(ddm, J=12.0,7.2Hz, IH), 2.19(dm, J=13.6Hz, IH), 2.34(dd, J=16.4,9.0Hz, IH), 2.41(m, IH), 2.42(dd, J=16.4,3.8Hz, IH), 2.66(very bs, Why2=27Hz, IH), 2.87(very bs, Wh^2=27Hz, IH), 3.25(very bs, Wh^2=15Hz, IH), 3.88(dt, J=10.7,4.9Hz, IH), 4.15(m, IH), 4.28(t, J=4.9Hz, IH), 5.87(t, J=0.9Hz, IH), 6.54(d, J=4.9Hz, lH)ppm; 13C NMR (75MHz) 12.5(CH3), 18.7(CH3), 20.4(CH2), 21.1(CH3), 21.8(CH2), 22.3(CH3), 22.6(CH2), 26.1(CH3), 26.2(CH3), 27.6(CH2), 35.8(CH2), 35.9(CH), 38.5(CH2), 38.6(CH2), 40.6(C), 41.1(CH2), 43.7(CH2), 44.5(C), 47.8(CH), 56.0(CH), 56.2(CH), 64.4(CH), 65.2(CH), 66.7(CH), 83.6(C), 123.6(CH), 128.KCH), 146.2(C), 166.9(C), 172.4(C), 188.7(C)ppm. 228 Preparation o£ Diaulusterol A, triacetate (41): 41 (24.0mg) was stirred at room temperature for 13 hours under nitrogen in a solution of pyridine/acetic anhydride (3:1; 8mL). Evaporation of the reagents under high vacuum generated a residue that was purified via preparative TLC (ethyl acetate) to give pure 25 triacetate 41 (25.0mg). Compound 41: clear oil; [ct] D+70.0° (c 1.35, CH2C12); Rf0.56 (EtOAc); UV(MeOH) > IR(CH2C12) 2960, 2880, 1740, 1667, 1628, 1460, 1371, 1249, 1061, 740cm-1; XH NMR (300MHz) 0.65(s, 3H), 0.96(d, J=7.2Hz, 3H), 1.04(tm, J=9.0Hz), 1.23(s, 3H), 1.29(d, J=7.6Hz, 3H), 1.42(s, 6H), 1.46-1.98(m, region), 2.03(s, 3H), 2.06(s, 3H), 2.10(s, 3H), 2.19(dm, J=13.5Hz, 1H) 2.42(dd, J=16.2,6.3Hz, 1H), 2.57(dd, J=16.2,7.7Hz, 1H), 5.10(dt, J=13.8,4.8Hz, 1H), 5.23(m, 1H), 5.60(m, 1H), 5.89(bs, 1H) , 6.43(d, J=5.4Hz, lH)ppm; 13C NMR (75MHz) 12.5, 18.6, 19.8, 20.4, 20.7, 20.9, 21.0, 21.2, 21.8, 22.5, 25.9, 26.0, 27.5, 35.8, 35.8, 35.9, 38.5, 40.5, 41.1, 42.1, 44.4, 47.7, 55.9, 56.2, 64.7, 67.5, 67.6, 83.1, 123.7, 124.2, 147.6, 166.5, 169.4, 170.0, 170.1, 170.1, 187.6ppm; HRMS, observed m/z 642.3779, C37H54°g requires 642.3768; LRMS, m/z (rel., intensity) 642 (M+, 3), 582 (4), 522 (10), 496 (9), 494 (10), 376 (100), 361 (55), 265 (33), 249 (23), 171 (37), 147 (41), 129 (38), 119 (75). Attempts Ifi. Make Ths. 2/l-Acetonjd,e of Diaulusterol A. (41) : To 41 (5.0mg) was added 3mL 2,2-dimethoxypropane or acetone, lmL dimethylformamide and one small crystal of p-toluene sulphonic acid monohydrate. The reaction mixture was refluxed for 2 hours, 20ml of CH0C1- was added and the reaction quenched and extracted 229 with 5% sodium bicarbonate solution (2 x 15raL) and water (15mL). The organic extract was dried over. anhydrous sodium sulphate for 1 hour filtered and concentrated in vacuo to yield 3.5mg of yellowy oil. The oil was fractionated by preparative TLC (100% ethyl acetate, collected two short wave UV active bands with Rf's of 0.80 and 0.70). The two bands collected (1.4mg and 1.6rag, respectively) were not readily identifiable. 300MHz XH NMR analysis indicated that they were not the expected acetonides. The Mel8 resonance observed at 0.65(s, 3H)ppm in 41 was not apparent and would not be expected to shift. The characterisation of the two products was not pursued further. 25 Diaulusterol fi (42): clear oil; [a] D-84.7°(c 0.32, CH2C12); Rf0.15 (EtOAc); DV(MeOH) Xmax 265nm (*4319); IR(CH2C12) 3410, 2966, 2877, 1664, 1626, 1458, 1384, 1064, 1029, 879cm"1; XH NMR (300MHz) 0.65(s, 3H), 0.97(d, J=7.2Hz, 3H), 1.09(tm, J=9.5Hz, IH), 1.17(s, 3H), 1.23(s, 6H), 1.30-1.87(m region), 1.95(m, IH), 2.07(m, IH), 2.19(m, IH), 2.29(m, IH), 2.42(m, IH), 3.88(m, IH), 4.27(t, J=5.2Hz, IH), 5.87(bs, IH), 6.54(d, J=5.2Hz, lH)ppm; HRMS, observed m/z 430.3055, C27H4204 requires 430.3083; LRMS, m/z (rel., intensity) 430 (M+, 14), 412 (79), 397 (49), 394 (28), 379 (27), 339 (23), 301 (20), 283 (20), 275 (37), 173 (54), 105 (64), 95 (61), 91 (57), 81 (72), 69 (87), 59 (100). Preparation c_f_ Djaulvsterpl a diacetate (52): 42 (3.2mg) was stirred for 11 hours at room temperature under nitrogen in a solution of pyridine/acetic anhydride (3:1; 3mL). Evaporation of the reagents under high vacuum generated a residue that was purified via preparative TLC (ethyl acetate) to give pure diacetate 53_. Compound 52: clear oil; Rf0.55 (EtOAc); DV(MeOH) 230 x 265nm £ max ( 7121); IR(CH2C12) 2967, 2880, 1742, 1668, 1628, 1458, 1371, 1262, 1044cm"1; 1H NMR (300MHz) 0.65(s, 3H), 0.98(d, J=7.lHz, 3H), 1.09(m, 1H), 1.24(s, 9H), 1.30-2.00(m region), 2.04(s, 3H), 2.08(s, 3H), 2.20(m, 1H), 2.42(tm, J=9.5Hz, 1H), 5.11(dt, J=13.4,3.9Hz, 1H), 5.60(m, 1H), 5.89(bs, 1H)), 6.43 (d, J=6.9Hz, lH)ppm; HRMS, observed m/z 514.3257, C31H4gOg requires 514.32958; LRMS, m/z (rel., intensity) 514 (M+, 2), 496 (2), 454 (9), 412 (27), 394 (55), 379 (50), 376 (36), 361 (33), 265 (29), 171 (42), 119 (100), 105 (67), 95 (27), 91 (39). 231 Gersemia rubiformis. Collection Data. £. rubiformis was collected both on the West and East coasts of Canada. Specimens were collected by hand using SCUBA on and around several exposed rocky reefs (-1 to -15m) located within the Gordon Group of Islands, British Columbia, and at Admiral's Cove, Cape Broyle and Bay Bulls, Newfoundland (south of St. Johns) Extraction and Chromatographic Separation For British Columbian Specimens. Four collections of £. rubiformis were made. The following represents typical isolation procedures and the maximum yields observed are quoted (these varied greatly with season). Freshly collected animals (1.5Kg dry weight, after extraction) were immediately immersed in methanol and allowed to extract at room temperature for 2-3 days. They were homogenised in a Waring blender and vacuum filtered through Celite. The homogenised material was extracted in fresh methanol at room temperature for a further two days. The first and second methanol extracts were combined and concentrated in vacuo to lOOOmL (approximately 1/3 of the original volume). The yellow brown solution was sequentially extracted with hexanes (4 x 250mL), methylene chloride (2 x 250mL), ethyl acetate (2 x 250mL) and brine (500mL). The combined organic extract was concentrated to 232 500mL and dried over anhydrous sodium sulphate for 3 hours. Vacuum filtration followed by evaporation in vacuo gave a dark red oil (33g). The oil was fractionated by flash chromatography (50mm diameter columns, 20cm silica gel, step gradient of 5% ethyl acetate/hexanes to 10% methanol/ethyl acetate, run repetitively with 4g samples) to yield fractions containing fats, steroids and carotenoids. Fractions eluting with 5%, 25% ethyl acetate/hexanes and two fractions eluting with 100% ethyl acetate (collected as test tube fractions and combined according to TLC (5% methanol/methylene chloride) results) gave fractions A (2.1g), B (3.25g), C (2.3g) and D (2.2g), respectively. Subsequent fractionation of these fractions followed. (i). Fraction A; Purification of fraction A by radial TLC (step gradient consisting of two solvent systems, 100% hexanes followed by 100% methylene chloride) gave a crude fraction which contained fats, tochuinyl acetate (1£5J and dihydrotochuinyl acetate (!££.) (250mg). This fraction was further purified by reverse phase EPLC (80%/15%/5% methanol/water/methylene chloride) to yield crude samples of 165 (42mg) and 166 (31.8mg). Repetitive reverse phase HPLC (80%/15%/5% methanol/water/methylene chloride) resulted in _3 pure samples of 165 (18.7mg, 1.3 x 10 % dry weight) and 1££ _3 (16.7mg, 1.1 x 10 % dry weight), both as clear oils. When spotted on TLC both sesquiterpenes could be visualised by short wave UV light, iodine absorptions, and on exposure to sulphuric acid spray (40% sulphuric acid/methanol) followed by heat brown 233 chars turning to purple were observed. (ii) . FractJPP E: Impurities (fats, steroids, etc..) in fraction B were partially removed via preparative TLC (2% methanol/methylene chloride, collected short wave UV active band, R^O.64) and Sephadex LH-20 column chromatography (70% methanol/chloroform, 2.5 x 100cm column). The resulting enriched sample of rubifolide (120.) (350mg) was further purified by a combination of normal phase HPLC (3:3:1 diethyl ether/hexanes/methylene chloride) and reverse phase HPLC (7:2:1 methanol/water/methylene chloride) to yield pure 170 (220mg, 1.5 x 10~2% dry weight) as a white crystalline solid (by slow evaporation from 50% methanol/methylene chloride at 4°C). When spotted on TLC 170 could be visualised by short wave UV light, iodine absorption, and on exposure to sulphuric acid spray (40% sulphuric acid/methanol) followed by heat a yellow char turning bright yellow brown then to purple was observed. (iii) . Fraction £: Purification of fraction C commenced with preparative TLC (5% methanol/methylene chloride, collected a series of close running short wave UV active bands, Rf's0.71-0.83) to give a fraction which contained fats, carotenoids, gersemolide f167). A4'5z, A7' 17-isogersemolide (lfifi.), A4,5E, A7' 17-isogersemolide (1££) , isoepilophodiones A (122), B (123.) and C (124), rubifol (125.), an unknown diterpene and gersolide (176) (485mg). 234 Fractionation of this fraction by radial TLC (step gradient consisting of two solvent systems, 50% methylene chloride/hexanes followed by 1:1:1 diethyl ether/hexanes/methylene chloride) gave two fractions. The least polar one contained 167, 168. 169. 173. 174. 175. an unknown diterpene, 176 and fats (314.6mg). The other consisted of 172 and fats (23.8mg) which was combined with fraction D after the preparative TLC purification (see below). Normal phase HPLC (3:3:1 diethylether/hexanes/methylene choride) on the more complex mixture gave six fractions three of which contained just fats, the other three subfractions consisted of: (i) 167. 169. an unknown diterpene and fats (41.6mg); (ii) 168. 174 and fats (30.4mg); (iii) HI, 123., 12£ and fats (161.6mg). Reverse phase HPLC (60% methanol/water) on subfraction (i) —4 yielded pure 167 (14.2mg, 9.5 x 10 % dry weight), pure 1£9_ (3.0mg, 2.0 x 10~4% dry weight), both as clear crystalline needles, and a pure sample of an unknown diterpene (2.2mg, 1.5 x —4 10 % dry weight) as a clear oil. Reverse phase HPLC (60% methanol/water) on subfraction (ii) yielded pure 1£8_ (l.Omg, 6.7 —5 x 10 % dry weight) as clear crystalline needles and pure 174 (17.4mg, 1.2 xlO % dry weight) as a clear oil. Reverse phase HPLC (60% methanol/water) on subfraction (iii) yielded pure 173 (77.7mg, 5.1 xlO % dry weight) as a pale yellow oil, pure 175 (7.6mg, 5.1 x 10~4% dry weight) and pure 176 (1.5mg, 1.0 x 10~4% dry weight), both as clear crystalline needles. (iv) . Fraction D_: Purification of fraction D commenced with preparative TLC (5% methanol/methylene chloride, collected a series of close 235 running short wave DV active bands, *sO.50-0.63) to yield a fraction (179.3mg) containing epilophodione (171). isoepilophodiones A (122) and B (173), rubiformate (177), fats and carotenoids. Radial TLC on this fraction (step gradient consisting of two solvent systems, 50% methylene chloride/hexanes followed by 1:1:1 diethyl ether/hexanes/methylene chloride) gave two fractions. The least polar fraction (28.4mg) contained 173 and fats, and was combined with fraction C after the preparative TLC purification (see above). The other consisted of 171. 172. 177 and fats (66.9mg). Reverse phase HPLC (60% methanol/water) on the more polar fraction yielded pure 171 (16.0mg, 1.1 x 10-3% dry weight) as clear crystalline needles, pure 172 (8.9mg, 5.9 x 10~ 4 % dry weight) as a pale yellow oil and pure 177 (3.6mg, 2.4 x —4 10 % dry weight) as a clear oil. When spotted on TLC all the diterpenes isolated from fractions C and D could be visualised by short wave UV light, iodine absorptions, and on exposure to sulphuric acid spray (40% sulphuric acid/methanol) followed by heat yellow/browm chars turning to grey/black were observed. The intensity of the spots varied somewhat from one diterpene to another. Extraction, chromatographic separation Hydroqenation £££ Newfoundland Specimens. Specimens of £. rubiformis were collected on a number of occasions between October 1986 and January 1987. The following represents typical isolation procedures. 236 Freshly collected animals (250g dry weight, after extraction) were immediately frozen and stored at -5°C. In February 1987 the specimens were defrosted, homogenised in a Waring blender and extracted in methanol at room temperature for 3 days. After vacuum filtration through celite the homogenised material was extracted in fresh methanol at room temperature for a further 3 days, and then vacuum filtered. The two extracts were kept seperated and concentrated in vacuo to 300mL (approximately 1/10 of the original volumns). The yellow brown solutions were partitioned between ethyl acetate (6 x 175mL) and brine (200mL). The combined organic extracts in each case were concentrated to 200mL and dried over anhydrous sodium sulphate for 2 hours. Vacuum filtration followed by evaporation in vacuo gave two dark orangy brown oils (approximately 2g in each case). The two oils were fractionated by flash chromatography (40mm diameter colums, 20cm silica gel, step gradient of 100% hexanes to 5% ethyl acetate/hexanes) to yield a fraction, eluting with 5% ethyl acetate/hexanes solution, containing fats and (+)-/?- cubebene-3-acetate (178). Crude samples of steroids 158-161x . eluting with 25%-100% ethyl acetate/hexanes solutions, were evidenced by 300MHz 1H NMR spectral comparison to the literature 186 data . These were not purified further. Fractionation of the least polar fraction by radial TLC (step gradient consisting of two solvent systems, 100% hexanes followed by 2 1/2% ethyl acetate/hexanes) gave a fraction which in the case of the first methanol extract was purified by normal phase HPLC (2 1/2% ethyl acetate/hexanes) to yield 253.4mg of crude 178. 237 Purification of this fraction by reverse phase HPLC (80%/10%/10% methanol/water/methylene chloride) gave pure 178 (4.7mg, 1.88 x —3 10 % dry weight) as a clear oil, which on standing rapidly decomposed to a yellow/black oil consisting of a complex mixture of very polar materials. When spotted on 178 could be visualised by exposure to sulphuric acid spray (40% sulphuric acid/methanol) followed by heat to give a distinctive bright yellow char fading to faint brown (a faint iodine absorption was also observed). In the case of the second methanol extract, after the radial TLC purification, the crude (+)-/?-cubebene-3-acetate (178) was subjected to hydrogenation. lOmL of methanol and a catalytic amount of palladium (on activated charcoal) was added to the crude 178 (75.3mg, mixture of fats and 178). The solution was saturated with hydrogen for 15 mins and left under hydrogen at balloon pressure for 13hrs with stirring. The palladium catalyst was filtered off (on fine scintered glass) and the volume reduced to dryness to yield 61.3mg of white solid. This solid was purified on normal phase HPLC (2 1/2% ethyl acetate/hexanes) to yield a fraction containing the two dihydro epimers 192 and 123. (1.2mg) in a ratio of 4:1, respectively, as evidenced by NMR and GC analysis. The two epimers 192 and 193 were not seperable by by HPLC, but the mixture did give two distinguishable peaks on GC analysis using a 15 meter DB-1 column at a temperature of 140°C. When spotted on TLC the dihydro epimer mixture could be visualised by iodine absorption, and sulphuric acid spray (40% sulphuric acid/methanol). In both cases nothing was initially observed, after a period of 15 mins, a faint brown/yellow char 238 turning bright orange yellow then to purple. A strong iodine absorption was also observed. physical. spectral and Chemical Interconvecsion Data. 25 Tochuinyl acetate (1£5J: clear oil; [a] D-42.5°(c 1.09, CH2Cl2); Rf0.80 (5% MeOH/CH2Cl2); UV(MeOH) \max 219nm (« 5047); IR(CH2C12) 2963, 2882, 1734, 1516, 1480, 1466, 1379, 1245, and 1034cm"1; XH NMR (300MHz) 1.13(s, 3H), 1.33(s, 3H), 1.58(m, 1H), 1.75-1.88(m, 4H), 1.94(s, 3H) , 2.30(s, 3H) , 2.48(m, 1H), 3.36(d, J=ll.lHz, 1H), 3.59(d, J=ll.lHz, 1H), 7.08(d, J=8.1Hz, 2H), 13 7.22(d, J=8.1Hz, 2H)ppm; C NMR (75MHz) 19.5(CH3), 20.2(CH2), 20.8(CH3), 20.9(CH3), 25.0(CH3), 34.8(CH2), 37.6(CH2), 47.4(C), 49.8(C), 70.6(CH2), 126.7(CH), 128.6(CH), 135.3(C), 142.9(C), 171.2(C)ppm; HRMS, observed m/z 260.1777, C17H2402 requires 260.1777; LRMS, m/z (rel., intensity) 260 (M+, 23), 200(12), 185 (13), 171 (9), 158 (51), 143 (61), 132 (100), 119 (77), 105 (48), 91 (39), 77 (13), 65 (10), 55 (12). 25 o Dihydrotochuinyl acetate (16JL) : clear oil; [a] D~29.3 (c 1.11, CH2C12); Rf0.84 (5% MeOH/CH2Cl2); DV(MeOH) \max 224nm (* 371); IR(CH2C12) 2963, 2882, 2820, 1731, 1653(very weak), 1468, 1379, 1238, 1034, 985, and 952cm"1; 1H NMR (300MHz) 1.04(s, 3H), 1.07(s, 3H), 1.42-1.55(m, 2H), 1.65(s, 3H) , 1.65-1.80(m, 3H), 2.02(s, 3H), 2.20(m, 1H), 2.61(m, 2H), 2.70(m, 2H), 3.74(d, J=11.0Hz, 1H), 3.81(d, J=11.0Hz, 1H), 5.39(m, 1H), 5.53(bs, lH)ppm; 13C NMR (75MHz) 19.5, 20.1, 21.0, 22.7, 22.8, 28.4, 31.9, 35.3, 37.0, 46.9, 50.3, 70.1, 119.0, 119.1, 130.5, 138.5, 171.4ppm; HRMS, observed ra/z 262.1926, C,7H9,-0~ requires 239 + 262.1934; LRMS, m/z (rel.f intensity) 262(M f 24), 260 (7), 219 (2), 218 (1), 217 (1), 202 (21), 200 (10), 187 (26), 185 (11), 173 (17), 171 (6), 159 (31), 145 (55), 132 (85), 119 (100), 105 (78), 91 (70), 77(39), 67 (33), 61 (41), 55 (42). Oxidation oL Dihydrotochuinyl acetate (1&£): 3.8mg of 166 were refluxed for 18 hours on palladium (on activated charcoal) in 95% ethanol (3mL). The palladium catalyst was filtered off (on fine scintered glass) and the volume reduced to dryness to yield 3.5mg of 165 (in quantitive yield). Gersemolide (167); clear crystalline needles (50% 25 methanol/diethyl ether at 0°C); mp 138-140°; [a] D-31.8°(c 0.90, CH2C12); Rf0.77 (5% MeOH/CH2Cl2); UV(MeOH) \max 233nra (* 9008); IR(CH2C12) 3072, 2992, 2943, 2873, 1762, 1699, 1678, 1648, 1604, 1442, 1221, 1135, 1065, and 911cm"1; "'"H NMR (300MHz) 1.28(m, 2H) , 1.64(bs, 3H), 1.73(bs, 3H), 1.83(d, J=1.2Hz, 3H), 2.01(m, IH), 2.26(m, IH), 2.32-2.45(m, IH), 2.38(dd, J=15.1,9.5Hz, IH), 2.53(dd, J=15.1,4.5Hz, 1H) , 3.29(d, J=0.6Hz, IH), 4.76(m, 2H), 5.20(bs, IH), 5.38(m, IH), 5.47(bs,lH), 6.33(q, J=1.2Hz, IH), 13 7.12(bs, lH)ppm; C NMR (75MHz) 19.8(CH3), 21.9(CH3), 22.0(CH2), 23.1(CH3), 31.4(CH2), 41.2(CH), 44.9(CH2), 62.4(CH), 79.0(CH), 112.3(CH2), 116.1(CH2), 123.8(CH), 134.2(C), 138.1(C), 146.4(C), 150.0(CH), 156.2(C), 173.7(C), 197.6(C), 208.3(C)ppm; HRMS, observed m/z 328.1686, C2nH2404 requires 328.1675 ; LRMS, ra/z (rel., intensity) 328 (M+, 31), 283 (21), 268 (13), 246 (33), 232 (10), 229 (14), 213 (11), 201 (16), 189 (23), 178 (62), 164 (42), 159 (32), 151 (85), 135 (52), 121 (43), 119 (43), 105 (79), 91 (100), 82 (86), 68 (64), 53 (62). Tsomerisation o_f Gersemolide with Formic AC id, (167) : 167 240 (2.0mg) and 3 drops of formic acid were left stirring under gentle reflux for 16 hours in 4mL of ethyl acetate. After the addition of a further 6ml of ethyl acetate the reaction was quenched and extracted with 5% sodium bicarbonate (2 x lOraL) and water (lOmL). The organic extract was dried over anhydrous sodium sulphate for 20 mins, filtered through scintered glass and concentrated in vacuo to yield a 30:3:20 mixture (1.5mg) of 167. 168 and one other unidentified compound, respectively (evidenced by 300MHz *H NMR analysis). Distinguishable 1H 300MHz NMR data for unidentified compound (presumably a A^'^E isomer of 167): 1.15(m), 1.74(bs, 3H), 1.93 (bs, 3H), 2.18(d, J=1.5Hz, 3H), 2.78(dd, J=13.5,3.lHz, 1H), 3.69(s, 1H), 4.94(ra, 1H), 5.09(bs, 2H), 5.17(m, 1H), 5.33(m, 1H) , 6.05(q, J=1.5Hz, 1H), 7.25(bsr lH)ppm. A4F 5z.A7'17-isoaersemolide (168): clear crystalline needles (50% diethyl ether/methanol at room temperature); [a] D~27.3 (c 0.02, CH2C12); Rf0.75 (5% MeOH/CH2Cl2); UV(MeOH) XMAX 218 (« 3047), shoulder at 237nm; XH NMR (300 and 400MHz) 1.18(m, 1H) , 1.26(m, lH)r 1.73(bsf 3H), 1.85(sf 3H), 1.93(sr 3H), 1.95-2.06(m region), 2.14(df J=1.6Hz, 3H), 2.31-2.49(m region), 2.72(dd, J=13.6,3.8Hz, 1H), 4.94(m, 1H), 5.07(bs, 1H), 5.93(bs, 1H), 6.11(q, J=1.6Hz, 1H), 7.10(bs, 1H); HRMS, observed 328.1684, C20H24°4 re<3uires 328.1675; LRMS, m/z (rel., intensity) 328 (M+, 100), 310 (25), 300 (21), 283 (60), 267 (22), 255 (17), 239 (20), 201 (38), 189 (38), 173 (50), 164 (45), 150 (60), 136 (37), 121 (49), 105 (46), 95 (68), 91 (65), 82 (51), 79 (53), 67 (75), 55 (35), 41 (77). 241 A4Y5E.A7'17-lsogersemolide (l£i): clear crystalline needles 25 (methanol at 0°C); mp 121-124°C; [a] D-225.3°(c 0.37, CH2Cl2); R.0.77 (5% Me0H/CHoClo); UV(MeOH) \ „ 205 (* 4759), shoulder at 242nm; IR(CH2Cl2) 3017, 2960, 2930, 2855, 1761, 1695, 1672, 1645, 1607, 1458, 1379, 1175, 1045, and 871cm"1; 1H NMR (300MHz) 1.26- 1.41(m region, 2H), 1.72(bs, 3H), 2.01(s, 3H), 2.06(s, 3H), 2.08(d, J=1.2Hz, 3H), 2.17(tdm, J=12.2,1.5Hz, IH), 2.33(m, IH), 2.49-2.57(m region, 2H), 2.71(dd, J=14.0,5.7Hz, IH), 4.76(bs, IH) 4.86(q, J=0.9Hz, IH) , 5.92(d, J=0.6Hz,. IH) , 6.21(q, J=1.2Hz, IH) , 7.02(t, J=0.6Hz, lH)ppm; 13C NMR (75MHz) 14.2, 18.4, 21.9, 23.1, 23.8, 28.4, 43.5, 44.3, 77.6, 112.7, 130.8, 131.4, 135.1, 145.3, 147.3, 148.1, 150.0, 172.4, 193.8, 203.3ppm; HRMS, observed m/z 328.1669, C2oH24°4 re<3uires 328.1675; LRMS, m/z (rel., intensity) 328 (M+, 25), 310 (10), 300 (15), 283 (27), 267 (19), 255 (13), 241 (20), 201 (35), 189 (38), 173 (54), 159 (42), 149 (65), 135 (52), 121 (59), 105 (68), 91 (95), 79 (78), 67 (100). Isomerisation ojL A4'5E.A7,17-lsoaersemolide (1££) Kith. Iodine; To a stirring solution of 169 (1.9rag) in 2mL of ethyl acetate was added 5 drops of a solution of one small crystal of iodine in lOmL benzene. The reaction was quenched after 16 hours with 5mL of aqueous sodium bisulphite solution (stirred for 10 mins), and partitioned with ethyl acetate (3 x 5mL). The combined organic extracts were dried over anhydrous sodium sulphate for 30 mins, filtered through scintered glass and concentrated in vacuo to yield a 3:1 mixture (1.3rag) of 169 and 168, respectively, as evidenced by 3 00MHz 1H NMR analysis. Rubifolide (120.): white cristalline solid (50% methanol/ 25 methylene chloride at 4°C); mp 159-160°C; [a] D+31.7°(c 0.39, 242 CH2C12); Rf0.80 (5% MeOH/CH2Cl2); DV(MeOH) \max 279nm [e 16433); IR(CH2Cl2) 3076, 2998, 2931, 2867, 1753, 1644, 1606, 1447, 1199, 1069, 902, and 863cm"1; 2H NMR (300 and 400MHz) 1.18(td, J=14.0,2.7Hz, 1H), 1.64(m, 1H), 1.74(bs, 3H), 1.92(bs, 3H), 1.98(bs, 3H), 2.08(dm, J=14.0Hz, 1H), 2.36(td, J=12.6,4.2Hz, 1H), 2.43(td, J=14.0,2.7Hz, 1H) , 2.55(m, 2H), 2.68(dd, J=ll.4,4.2Hz, 1H), 3.22(t, J=11.4Hz, 1H), 4.88(bs, 1H), 4.90(m, 1H), 4.95(dm, J=11.4Hz, 1H), 5.99(bs, 1H), 6.07 (bs, 1H), 6.86(bs, lH)ppm; 13C MMR (75MHz) 9.5(q), 19.2(q), 20.0(t), 25.7(q), 30.5(t), 31.2(t), 39.5(t), 43.3(d), 78.7(d), 112.9(t), 113.8(d), 117.l(s), 117.4(d), 127.0(s), 132.8(s), 145.4(s), 149.4(s), 149.9(s), YL 152.1(d), 174.5(s)ppm; HRMS, observed m/z 312.1275, ^•20 2A°2 requires 312.1276; LRMS, m/z (rel., intensity) 312 (M+, 41), 216 (12), 201 (11), 173 (6), 148 (100), 133 (44), 120 (22), 105 (34), 91 (19), 79 (13), 77 (15), 69 (7), 68 (11), 67 (11), 65 (9), 55 (5), 53 (12). 1R Lanthanide Shift ME Experj.ment On RtifrjfPljde (Hfl) : 6.5 —4 x 10 moles (674.Omg) of Eu(FOD)3 were dissolved in CDC13 (1.0ml). 400MHz '''H NMR spectra were run with a 5mm sample tube —5 containing 6.5 x 10 moles (20.3mg) of 170. The subsequent addition of 10/JL (6.5 x 10~6moles), followed by two further 10^/L additions of shift reagent solution resulted in the spectra given 3 + in Figures 30a, b and c, corresponding to [Eu ]/[substrate (170)1 ratios of 1:10, 2:10 and 3:10, respectively. All manipulations were performed under nitrogen in a glove box with pre-dried (75°C) glassware. Epilophodione (171): clear crystalline needles (methanol at 243 25 0°C); mp 153-155°C; [a] D+136.2°(c 0.42, CH2Cl2);Rf0.67 (5% MeOH/CH-Cl-); UV(MeOH) X „ 264 (* 10893), 211nra (e 14133); IR(CH2C12) 3077, 2940, 2865, 1756, 1683, 1645, 1619, 1607, 1443, 1201, 1065, 948, and 900cm"1; 1H NMR (300MHz) 1.38(m, IH), 1.61(bs, 3H), 1.86(m, IH), 1.91(d, J=1.6Hz, 3H), 2.19(d, J=1.2Hz, 3H), 2.20-2.47(m, 2H), 2.51(m, 2H), 2.67(dd, J=13.3,4.5Hz, IH), 2.82(dd, J=13.3,4.8Hz, IH), 4.92(m, 2H), 5.27(m, IH), 6.12(bs, IH), 6.38(q, J=1.6Hz, IH), 7.15(bd, J=0.9Hz, lH)ppm; 13C NMR (75MHz) 17.2(CH3), 21.0(CH2), 21.1(CH3), 22.3(CH3), 33.5(CH2), 41.8(CH), 43.6(CH2), 45.5(CH2), 78.4(CH), 115.6(CH2), 127.1(CH), 133.5(CH), 136.2(C), 143.9(C), 145.4(C), 148.1(CH), 151.8(C), 173.6(C), 192.1(C), 205.5(C)ppm; HRMS, observed m/z 328.1680, C20H24°4 re<3uires 328.1675; LRMS, m/z (rel., intensity) 328 (9), 310 (3), 295 (1), 285 (2), 267 (3), 232 (7), 219 (9), 201 (6), 178 (50), 151 (100), 135 (55), 133 (32), 107 (31), 105 (31), 95 (31), 91 (40), 82 (95), 79 (45), 77 (36), 67 (48), 53 (39). Isomerization o_f. Epilophodione (HI) With Iodine: To a stirring solution of 171 (6.7mg) in 2mL ethyl acetate was added 5 drops of a solution of one small crystal of iodine dissolved in lOmL benzene. Partial conversion of 171 was noted after 12 hours by TLC (5% methanol/methylene chloride) examination. The reaction was quenched with 5mL of aqueous sodium bisulphite solution (left stirring for 35mins) and partitioned using ethyl acetate (3 x 5mL). The combined organic extracts were dried over sodium sulphate for 15 rains, filtered through scintered glass and concentrated in vacuo to yield a 2:1:4:1 mixture (5.5rag) of 121, 172. 173 and one other unidentified compound, respectively (evidenced by 300MHz 1H NMR analysis). 244 25 Isoepilophodione A (112): pale yellow oil; [ct] D+137.6°(c 0.55, CH2C12); Rf0.71 (5% MeOH/CH2Cl2); UV(MeOH) \max 246 (* 12818), 208nm (* 16167); IR(CH2Cl2) 3070, 2985, 2936, 2845, 1761, 1681, 1648, 1619, 1605, 1445, 1380, 1276, 1065, and 1026cm"1; 1H NMR (300MHz) 1.66(bs, 3H), 1.99(bs, 3H), 2.06(m, 1H), 2.16(d, J=1.2Hz, 3H), 2.33-2.45(m region, 3H), 2.50(m, 2H), 2.98(dd, J=13.5,5.1Hz, 1H), 3.04(dd, J=13.5,4.5Hz, 1H), 4.54(bs, 1H), 4.76(q, J=0.6Hz, 1H), 5.23(bm, 1H), 6.07(bs, 1H), 6.68(q, J=1.2Hz, 1H), 7.19(bs, lH)ppm; 13C NMR (75MHz) 14.3, 18.3, 22.1, 23.4, 29.0, 38.9, 41.5, 42.8, 78.1, 113.2, 130.2, 132.3, 134.1, 140.0, 144.9, 147.8, 149.8, 172.2, 194.6, 202.6ppm; HRMS, observed m/z 328.1686, C2QH2404 requires 328.1675; LRMS, m/z (rel., intensity) 328 (M+, 4), 310 (3), 295 (2), 282 (1), 267 (3), 232 (8), 219 (6), 178 (21), 151 (26), 135 (25), 121 (28), 107 (30), 95 (26), 91 (35), 82 (100), 67 (43). Tsomerisation Q£_ Isoepilophodione A. (112) With iMinfi.: To a stirring solution of 172 (2.5mg) in 2mL ethyl acetate was added 5 drops of a solution of one small crystal of iodine dissolved in lOmL benzene. The reaction was quenched after 8 hours with 5mL of aqueous sodium bisulphite solution (left stirring for 35 mins) and partitioned using ethyl acetate (3 x 5mL). The combined organic extracts were dried over sodium sulphate for 15 rains, filtered through scintered glass and concentrated in vacuo to yield a 1:3:2 mixture (1.9mg) of 171. 172 and 173. respectively (evidenced by 300MHz 1H NMR analysis). Isomerisation Q£ Isoepilophodione A, (172) W±til Formic Acid: 172 (1.5mg) and 3 drops of formic acid were left stirring under 245 gentle reflux for 5 hours in 3mL of methanol. After the addition of lOmL of ethyl acetate the reaction was quenched and extracted with 5% sodium bicarbonate (2 x lOmL) and water (lOmL). The organic extract was dried over anhydrous sodium sulphate for 20 mins, filtered through cintered glass and concentrated in vacuo to yield a 3:1:1 mixture (0.9mg) of 122/ 173 and 174. respectively (evidenced by 300MHz XH NMR analysis). 25 Tsoepilophodione £ (122): pale yellow oil; [a] D+298°(c 0.40, CH2C12); Rf0.77 (5% MeOH/CH2Cl2); UV(MeOH) Xmax 267 (* 14429), 212nm (* 13111); IR(CH2Cl2) 2982, 2935, 2875, 1758, 1683, 1643, 1620, 1610, 1443, 1381, 1260, 1120, 1105, 1064, 1026, and 882cm"1; 1H NMR (300MHz) 1.41(very broad, IH), 1.70(bs, 3H), 1.88(bs, 3H), 2.06(m, 3H), 2.20-2.62(broad m region, 3H), 2.68- 2.92(very broad, 2H), 4.03(very broad, IH), 4.64(bs, IH), 4.83(bm, IH), 5.27(very broad, IH), 6.21(bm, IH), 6.25(bs, IH), H 7.16(very broad, lH)ppm; HRMS, observed m/z 328.1670, C20 24°4 requires 328.1675; LRMS, m/z (rel., intensity) 328 (M+, 11), 312 (3), 232 (7), 219 (10), 201 (6), 178 (49), 151 (95 ), 135 (45), 122 (42), 107 (33), 95 (31), 91 (40), 82 (100), 67 (48). Isomerisation oj. Isoepilophodione B. (173) With Iodine: To a stirring solution of 173 (8.2mg) in 2mL of ethyl acetate was added 5 drops of a solution of one small crystal of iodine dissolved in lOmL benzene. Partial conversion of 173 was noted after 14 hours by TLC (5% methanol/methylene chloride) examination. The reaction was quenched with 5mL of aqueous sodium bisulphite solution (stirred for 30 mins) and partitioned using ethyl acetate (3 x 5mL). The combined organic extracts were dried over anhydrous sodium sulphate for 30 mins, filtered through 246 scintered glass and concentrated in vacuo to yield a 1:1:10:1 mixture (7.6mg) of 171f 172. 173 and one other unidentified compound, respectively (evidenced by 300MHz XH NMR analysis). 25 Isoepilophodione C. (Hi): clear oil; [«] D-170°(c 0.26, CH2C12); Rf0.75 (5% MeOH/CH2Cl2); DV(MeOH) Xmax 230nm (« 2426); IR(CH2C12) 2982, 2942, 2868, 1756, 1689, 1668, 1643, 1614, 1441, 1380, 1335, 1102, 1091, 1042, and 958cm""1; 1H NMR (300 and 400MHz) 1.61(bs, 3H), 1.70-1.82(m ,2H), 1.85-2.06(m, 3H), 1.95(d, J=1.2Hz, 3H), 2.14(m, IH), 2.20(m, 2H), 3.05(dd, J=ll.lrl.8Hz, IH), 3.08(d, J=15.0Hz, IH), 3.73(dr J=15.0Hz, IH), 4.69(bs, IH), 4.81(m, IH), 5.06(m, IH), 5.90(d, J=0.9Hz, IH), 6.08(s, IH), 6.29(q, J=1.2Hz, IH), 6.79(d, J=1.8Hz, lH)ppm; 13C NMR (75MHz) 18.1, 21.2, 27.3, 27.6, 35.7, 41.6, 41.7, 48.3, 80.0, 113.0, 125.3, 128.3, 132.9, 143.7, 145.8, 148.0, 154.3, 173.9, 196.9, H 202.0ppm; HRMS, observed m/z 328.1671, C20 24°4 requires 328.1675; LRMS, m/z (rel., intensity) 328 (M+, 8), 310 (6), 300 (3), 283 (6), 267 (4), 259 (4), 178 (19), 161 (17), 150 (100), 135 (14), 121 (27), 109 (29), 91 (28), 82 (60). Rubifol (175): clear crystalline needles (methanol at room 25 temperature); mp 127-130°C; [a] D -41.3°(c 0.38, CH2C12); Rf0.76 (5% MeOH/CH2Cl2); DV(MeOH) \max 230 (*6658), 207nm (« 10284); IR(CH2C12) 3509(broad), 2980, 2934, 2869, 1756, 1700, 1671, 1644, 1610, 1447, 1382, 1338, 1200, 1094, and 1042cm"1; XE NMR (300MHz) 1.20(s, 3H), 1.67-1.77(m, IH), 1.70(bs, 3H), 1.82(bt, J=10.7Hz, IH), 2.06(bs, 3H), 2.09-2.19(m region, 2H), 2.21-2.37(m region, 2H), 2.47(d, J=17.3Hz, IH), 2.59(dd, J=10.8,3.9Hz, IH), 3.34(bd, J=10.7Hz, IH), 3.37(d, J=17.3Hz, IH), 3.91(bt, J=10.8Hz, IH), 247 4.82(bsf 1H), 4.83(m, 1H), 4.84(bs, IE), 5.1(very broad W =46HZ 6 31 bs lH 13 h/2 ' ' * ( ' ) ' 6.64(df J=0.6Bz, lH)ppm; C NMR (75MHz) 18.2(CB3)V 21.3(CH2), 24.9(CH3), 26.6(CH2), 26.7(CH3), 37.6(CH2), 39.3(CH2), 40.7(CH), 50.2(CH2)f 78.3(CH), 80.7(C), 112.9(CH2), 126.3(CH), 132.3(C), 145.2(C), 147.2(CH), 155.2(C), 173.6(C), 202.9(C), 217.3(C)ppm; HRMS, observed m/z 346.1785, C20H26°5 rec3uires 346.1781; LRMS, m/z (rel., intensity) 346(M+, 2), 328 (5), 310 (3), 303 (10), 285 (5), 243 (6), 215 (14), 178 (17), 161 (16), 151 (89), 140 (45), 133 (33), 121 (33), 109 (54), 91 (62), 82 (100), 67 (67). 1 Unknown Diterpene: clear oil; Rf0.79 (5% MeOH/CH2Cl2); H NMR (300MHz) 1.43(bs, 3H), 1.62(bs, 3H), 1.64(m, 1H), 1.80(td, J=11.7,0.3Hz, 1H), 1.97(d, J=1.5Hz, 3H), 2.02-2.11(m, 3H), 2.19- 2.29(m, 2H), 2.84(bd, J=17.9Hz, 1H), 3.00(bd, J=17.9Hz, 1H), 3.15(d, J=11.7Hz, 1H), 3.18(s, 3H), 3.83(baseline bump, 1H), 4.58(bs, 1H), 4.79(bs, 1H) , 5.17(very bs, 1H), 6.19(bs, 1H), 6.98(bs, lH)ppm; Partial 13C NMR (75MHz) 17.8, 20.4, 27.2, 35.9, 39.9, 41.0, 50.6, 80.9, 112.8, 126.1, 132.0, 143.9, 148.6, 174.1ppm; HRMS, observed m/z 360.1949, C2iH28°5 re<3uires 360.1937; LRMS, m/z (rel., intensity) 360 (M+, 2), 345 (1), 332 (5), 328 (1), 317 (1), 301 (2), 288 (2), 242 (2), 205 (6), 178 (7), 165 (18), 161 (14), 154 (37), 137 (18), 123 (25), 109 (15), 99 (20), 93 (16), 82 (100), 67 (20). Attempts l£ Chemically Tnterconvert EpilPPhodjpne (171) and Tsoepilophodiones A. (122) , £ (121) £ (111) IP. Rubjfpl (125.) aM tiifi Unknown Diterpene: Metabolites 121, 122, 122 and 121 (1- 3mg), and 3 drops of formic acid were left stirring or under reflux for upto 16 hours (shorter reaction times were tried) in 248 3mL of methanol (for methanolysis to generate unknown diterpene) or 3mL of 5:1 acetone/water or 3mL 5:1 tetrahydrofuran/water (for hydration to generate 175). After the addition of lOraL of ethyl acetate the reaction was quenched and extracted with 5% sodium bicarbonate (2 x lOmL) and water (lOmL). The organic extract was dried over anhydrous sodium sulphate for 20 mins filtered through scintered glass and concentrated in vacuo to yield: with water and formic acid as reagents, 171 gave a 1:2 mixture of 171 and 173, respectively, 173 was apparently uneffected, and 174 gave a 5:1 mixture of 174 and an unidentified compound that had previously been noted as a decomposition product of 174 (might presumably be 4 5 the A ' E isomer of 174): with methanol and formic acid as reagents, 171 gave a 3:1 mixture of two unidentified compounds, possibly methyl ethers but not the unknown diterpene, for 172 the results are reported on p. 245., and 173 gave a 1:2:3:3 mixture of 173, the two unidentified compounds obtained in the reaction of 171 (the major product from previous, was know the minor product) and one further unidentified compound, respectively. All the above results were evidenced by 300MHz 1H NMR analysis. Distinguishable •'"H 300MHz NMR data for the unidentified decomposition and "isomerisation" product of 174: 1.70(bs, 3H), 1.77-2.26(m regions), 2.22(bs, 3H), 2.93(dd, J=12.0, 5.2Hz, 1H), 3.07(m, 1H), 4.66(bs, 1H), 4.84(bs, 1H) , 5.09(bm, 1H), 5.93(bs, 1H), 6.07(bs, 1H), 6.19(bs, 1H), 6.90(bs, lH)ppm. flersolide (176): clear crystalline needles (methanol at room 25 temperature); mp 176-178°C; [a] D-54.9°(c 0.31, CH2Cl2); Rf0.76 (5% MeOH/CH0Cl,); UV(MeOH) X_.Y 209 (*4213), shoulder at 237nra; 249 IR(CH2C12) 3091, 3052, 3037, 2995, 2937, 1758, 1683, 1679, 1609, 1480, 1446, 1394, 1333, 1242, 1198, 1084, and 1064cm"1; 1H NMR (300MHz) 0.98(dd, J=8.3,5.8Hz, 1H), 1.25-1.36(m, 1H), 1.30(s, 3H), 1.43 (t, J=5.8Hz, 1H), 1.72(d, J=0.6Hz, 3H), 1.83(dd, J=8.3,5.8Hz, 1H), 1.98-2.13(m, 1H), 2.06(d, J=0.6Hz, 3H), 2.18(t, J=12.3Hz, 1H), 2.31(tdm, J=12.3,5.2Hz, 1H), 2.33-2.48(m region, 2H), 2.81(bd, J=12.3Hz, 1H), 4.97(bs, 1H), 5.14(d, J=0.6Hz, 1H), 5.22(bs, 1H), 5.96(q, J=0.6Hz, 1H), 6.76(bs, lH)ppm; HRMS, observed m/z 328.1674, C2oH24°4 re(3uires 328.1675; LRMS, m/z (rel., intensity) 328 (M+, 5), 313 (6), 300 (3), 285 (5), 267 (4), 253 (4), 232 (6), 205 (16), 177 (66), 159 (29), 149 (100), 135 (41), 123 (75), 105 (50), 97 (60), 91 (79), 79 (69), 67 (61). 25 Rubiformate (122): clear oil; [a] D+194.5°(c 0.06, CH2C12); Rf 0.69 (5% MeOH/CH2Cl2); DV(MeOH) \max 297 (« 16780), 207nm (* 14295); IR(CH2C12) 3064, 2931, 2872, 2842, 1758, 1720, 1676, 1650, 1606, 1520, 1413, 1368, 1339, 1164, 1142, and 1065cm"1; 2H NMR (300MHz) 1.63(m, 1H), 1.67(bs, 3H), 2.01(s, 3H), 2.20(m, 1H), 2.23("d", 3H), 2.55-2.65(m region, 2H), 2.68(m, 1H), 2.70- 2.78(m region, 2H), 2.98(ddm, J=6.6,2.7Hz, 1H), 3.04(ddm, J=6.6,2.7Hz, 1H), 4.69(bs, 1H) , 4.78(q, J=0.9Hz, 1H), 5.29(ra, 1H), 7.03(S, 1H), 7.11(m, 1H), 9.46(s, lH)ppm; 13C NMR (75MHz) 9.8, 18.4, 23.1, 29.8, 30.5, 31.2, 45.5, 46.5, 76.8, 113.3, 119.2, 125.5, 134.2, 145.0, 147.7, 150.7, 157.8, 173.1, 176.8, 204.5ppm; HRMS, observed m/z 344.1623, C20H24°5 requires 344.1624; LRMS, m/z (rel., intensity) 344 (M+, 55), 326(3), 315 (6), 297 (3), 283 (3), 255 (2), 221 (6), 191 (34), 177 (54), 173 (36), 161 (36), 133 (47), 124 (89), 123 (100), 105 (26), 95 (43), 91 (29), 79 (20), 67 (36). 250 Attempts &t converting Rubiformate (Hi) T_C_ h Crystalline Derivative. An. Ene-Dione Qr_ L Methvl Enol Eikex: To a stirring solution of 177 (1.6mg), in 3mL of undried acetone or methanol, was added 5 drops of a 1:5 solution of potassium carbonate in the appropriate solvent. After 40 mins 170 L of methyl iodide was added. After stirring or refluxing for 2 hours the reaction was quenched with water (10mL), and extracted with methylene chloride (4 x 5mL) and two drops of acetic acid. The organic extract was dried over anhydrous sodium sulphate, filtered through cintered fine glass and evaporated in vacuo. In each case a complex mixture of polar compounds was obtained (evidenced by TLC) which were not identifiable by 300MHz AH NMR analysis and were not characterised further. 25 (+)-£-Cubefrene-2-acetate (lift): clear oil; [a] D+28.9°(c 0.24, CH2C12); Rf0.34 (5% EtOAc/hexanes); DV(MeOH) \max 207nm (* 6139); IR(CH2C12) 3057, 2962, 2934, 2875, 1732, 1609, 1464, 1374, 1246, and 1023cm"1; 1H NMR (300MHz) 0.58(m, IH), 0.80- 0.90(obscured by methyl doublets, IH), 0.90-0.96(3 x methyl doublets, 9H), 1.05-1.15(m region, 2H), 1.27(m, IH), 1.37-1.48(m region, 2H), 1.55-1.67(m region, 2H), 1.74(m, IH), 2.01(s, 3H), 2.38(dd, J=14.4, 8.0Hz, IH) , 5.01(bs, IH), 5.05(bs, IH), 5.39(d, 13 J=8.0Hz, lH)ppm; C NMR (75MHz) 18.8(CH3), 19.8(CB3), 20.0(CH3), 21.6(CH3), 26.5(CH), 30.4(CH), 31.1(CH2), 31.2(CH2), 33.5(CH), 34.6(CH), 37.4(C), 38.9(CH2), 44.2(CH), 76.1(CH), 109.0(CH2), 152.8(C), 170.4(C)ppm; HRMS, no parent ion observed -C2H402 m/z 202.1713, C15H22 requires 202.1722; LRMS m/z (rel., intensity) 202 (M+-60, 18), 159 (100), 157 (74), 142 (38), 131 (59), 117 251 (50) , 105 (56), 91 (69), 77 (38), 69 (29), 60 (26), 54 (41). Pihydrp Epjper Mixture Qf_ f +) -ft-Cubebene-3-acetate f 178) (122 SM 132.): For *H NMR (300MHz) see Figure 54., and Table 31; LRMS of mixture, m/z (rel., intensity) 264 (M+, 0.6), 204 (12), 161 (100), 133 (10), 119 (30), 105 (50), 91 (23), 81 (24), 69 (11), 55 (23), 43 (41). 252 Tochuina tetraquetra. Collection D_ai&. Specimens of X« tetraquetra (8.6g average dry weight per animal, after extraction) were collected by hand using SCOBA in several exposed rocky channels (-5 to -15m) located within the Gordon Group of Islands, Port Hardy, British Columbia (found feeding on, or near by, Gersemia rubiformis-two animals collected), and typically sandy bottoms located within the Deer Group of Islands, Bamfield, British Columbia ("Bamfield animals" three animals collected). Freshly collected animals were immediately immersed in methanol and allowed to extract at room temperature for 2-3 days before decanting the methanol to give a first skin extract. The animals were again immersed in fresh methanol and left for a further two days. The second methanol extract was decanted and the two skin extracts combined. Extraction and. chromatographic separation for Port Hardy Animals The methanol extract (lOOOmL) was concentrated in vacuo to 50mL and partitioned between brine (lOOmL) and ethyl acetate (4 150mL). After drying over sodium sulphate for 2 hours the organi extract was reduced to dryness to yield 0.5g of an orangy oil. The oil was fractionated by preparative TLC (2% methanol/methylene chloride) to yield two short wave UV active bands. The least polar band (R^0.66, consisting of a series of close running short wave UV active bands) contained 10.7mg of a 253 mixture of tochuinyl acetate (165), dihydrotochuinyl acetate (166). rubifolide (170) and fats. The other band contained pukalide (£2) and fats (Rf0.28) and was fractionated by reverse phase HPLC (70% methanol/water) to yield pure £2. (15.7mg (7.85mg (0.09% dry weight) per animal) as a white solid (powdery). When spotted on TLC fL3_ could be visualised by short wave UV light, iodine absorption, and on exposure to sulphuric acid spray (40% sulphuric acid/methanol) followed by heat a purple char turning a dirty green/brown was observed. The second, the least polar, UV active band was fractionated on reverse phase HPLC (80%/15%/5% methanol/water/methylene chloride) to yield crude samples of tochuinyl acetate (165), dihydrotochuinyl acetate (166) and rubifolide (170). Repetitive reverse phase HPLC purifications (80%/15%/5% methanol/water/methylene chloride) on these crude samples yielded both pure 165 (1.4mg, 0.7mg (0.008% dry weight) per animal) and 166 (0.7mg, 0.35mg (0.004% dry weight) per animal) as clear oils, and pure 170 as a white crystalline solid (4.2mg, 2.1mg (0.02% dry weight) per animal). Extraction and chromatographic separation for Bamfjeld Animals. The methanol extract (lOOOmL) was concentrated to 200mL and extracted with brine (200mL) and ethyl acetate (4 x lOOmL). After drying over sodium sulphate for 3 hours the organic extract was reduced to dryness to yield 0.6g of an orangy oil. The oil was fractionated by preparative TLC (2% methanol/methylene chloride) to yield a short wave UV active band 254 (RfO.25). This band was further purified on reverse phase HPLC (70% methanol/water) to yield ptilosarcenone (179) (7.2mg, 2.4rag (0.03% dry weight) per animal), as an amorphous solid, and its butanoate analogue, 1££, (5.1mg, 1.7mg (0.02% dry weight) per animal) as a glass. When spotted on TLC both 179 and 180 could be visualised by short wave UV light, faint iodine absorptions, and on exposure to sulphuric acid spray (40% sulphuric acid/methanol) followed by heat reddy chars turning a reddy/purple brown were observed. In the rapid workup freshly collected animals were immediately immersed in chloroform and left standing at room temperature for one day. The chloroform was decanted off and the volume (600mL) reduced to 200mL and extracted with brine (200mL) and ethyl acetate (4 x 200mL). After drying over sodium sulphate the organic extract was reduced to dryness. The orangy oil obtained was fractionated by non-activated preparative layer chromatography (8% ethyl acetate/methylene chloride). The short wave UV active band obtained (R^O.23) was further fractionated on normal phase HPLC (15% ethyl acetate/methylene chloride) to yield ptilosarcenone (179) and its butanoate analogue, 180. only. Total workup took less than three days. physical and Spectral Data. Pukalide (63): white powdery solid (methanol at room 25 temperature); mp 203-205°C; [a] D+40°(c 1.0, CH2C12); Rf0.39 (2 1/2% MeOH/CH2Cl2); UV(MeOH) Xmax 249nm {c 5175); IR(CH2C12) 3140, 3019, 1761, 1715, 1580, 1267, 1230, 1082, 906, 891, 869, 255 and 830cm ,• """H NMR (80MHz) 1.05(sf 3H) , 1.18(m, 1H) , 1.65(m, 1H), 1.79(bs, 3H), 2.00-2.75(m region, 4H), 2.95(m, 2H), 3.58(td, J=11.0,3.5Hz, 1H), 3.76(s, 3H) , 4.13(s, 1H), 4.90(bs, 1H), 5.18(bs, 2H), 6.38(bs, IE), 7.08(bs, lH)ppm; 13C NMR (100MHz) 18.7, 19.8, 22.8, 32.4, 32.5, 40.0, 40.7, 51.2, 55.0, 57.0, 77.8, 106.4, 112.9, 113.9, 137.3, 145.8, 148.2, 148.3, 160.0, 163.8, 173.7ppm; HRMS, observed m/z 372.1563, C2iH24°6 re(3uires 372.1573; LRMS, m/z (rel., intensity) 372 (M+, 39), 340 (29), 315 (13), 276 (17), 208 (100), 204 (47), 168 (65), 165 (89). Tochuinyl acetate (165), Dihydrotochuinyl acetate (166) and Rubifolide (170): Data reported above. Ptilosarcenone (179): amorphous solid (methanol at room 25 temperature); mp 150-152°C; [a] D~67.0°(c 0.57, CH2Cl2); Rf0.68 (5% MeOH/CH2Cl2); UV(MeOH) Xmax 224nm {e 12532); IR(CH2C12) 3538, 2978, 1789, 1741, 1689, 1379, 1371, 1218, and 1019cm"1; XH NMR (300MHz) 1.18(s, 3H) , 1.21(d, J=7.2Hz, 3H), 1.30(d, J=7.2Hz, 3H), 2.15(s, 3H), 2.20(s, 3H), 2.41(q, J=7.2Hz, 1H), 2.69-2.84(m region, 2H), 3.43(s, 1H), 5.01(d, J=4.0Hz, 1H), 5.26(dm, J=4.0Hz, 1H), 5.49(d, J=7.5Hz, 1H), 5.63(dd, J=12.4,8.7HZ, 1H), 5.72(d, J=8.7Hz, 1H), 5.90(dd, J=10.5,0.9Hz, 1H), 5.93-5.96(m region, 2H), 6.12(d, J=0.9Hz, 1H), 6.61(d, 13 J=10.5Hz, lH)ppm; C NMR (75MHz) 6.2(CH3), 14.7(CH3), 14.9(CH3) 21.0(CH3), 21.8(CH3), 38.9(CH), 43.4(C), 45.1(CH), 45.8(CH), 62.KCH), 68.9(CH), 76.6(CH), 77.7(CH), 84.0(C), 118.8(CH2), 124.KCH), 128.6(CH), 129.6(CH), 136.5(C), 154.1(CH), 169.7(C), 170.0(C), 174.3(C), 202.2(C)ppm; HRMS, no parent ion observed - C2H20 m/z 438.1423, C22H27C107 requires 438.1446; LRMS, m/z 256 (rel., intensity) 480 (M+, <1), 438 (5), 402 (3), 378 (3), 360 (4), 344 (5), 342 (5)f 325 (4), 324 (4), 314 (3), 236 (6), 226 (4), 219 (15), 191 (21), 175 (17), 147 (23), 135 (34), 123 (100), 107 (32), 95 (30), 91 (29), 79 (30), 69 (19). 25 Butanoate Analogue. 18JQ: clear glass; [a] D~49.2°(c 0.41, CH2C12); Rf0.71 (5% MeOH/CH2Cl2); UV(MeOH) \max 220nm {c 4857); IR(CH2C12) 3590, 2940, 2880, 1790, 1735, 1689, 1371, 1218, 1176, 1034, and 1020cm"1; AH NMR (300 and 400MHz) 0.99(t, J=7.2Hz, 3H) , 1.17(s, 3H), 1.20(d, J=7.2Hz, 3H), 1.30(d, J=7.2Hz, 3H), 1.71(m, 2H), 2.20(s, 3H), 2.36(q, J=7.2Hz, IH), 2.38(t, J=7.2Hz, 2H), 2.73(m, IH), 2.79(m, IH), 3.42(s, IH), 4.98(d, J=3.9Hz, IH), 5.23(dm, J=3.9Hz, IH), 5.46(d, J=7.5Hz, IH), 5.60(dd, J=12.4,8.7Hz, IH), 5.79(d, J=8.7Hz, IH), 5.87(dd, J=10.4, 0.9Hz, IH), 5.93-5.96(m region, 2H), 6.12(dm, J=0.9Hz, IH), 6.58(d, J=10.4Hz, IH), HRMS, observed m/z 508.1873, C2gH33Cl08 requires 508.1865; LRMS, m/z (rel., intensity) 508 (M+, 0.2), 438 (5), 422 (3) , 402 (3), 378 (4), 360 (4), 342 (4), 324 (4), 307 (4), 297 (4) , 287 (4), 279 (5), 271 (4), 269 (7), 255 (5), 253 (5), 251 (7), 231 (14), 219 (12), 203 (10), 191 (18), 175 (15), 147 (17), 135 (32), 123 (72), 107 (27), 95 (17), 81 (14), 71 (95), 55 (13), 43 (100). 257 211. BIBLIOGRAPHY. 1. For example see: (a). K. L. Rinehart, J. B. Gloer, J. C. Cook, S. A. Mizsak, T. A. Scahill. J. Amer. Chem. Soc., 103. 1857, 1981; (b). J. B. Tucker. Bioscience., 23, 541, 1985. 2. For example see: (a). J. Hellou, J. E. Thompson, R. J. Andersen. Tetrahedron., 3_8_, 1875, 1982; (b) . For examples and a summary concerning natural product literature up to 1981 see: J. B. Harborne. In "Introduction to Ecological Biochemistry", Academic Press, Toronto, 1982. 3. For example see: (a). J. A. Ballantine, A. F. Psaila, A. Pelter. J. C. S. Perkin I., 1080, 1980; (b). D. H. Griffin. In "Fungal Physiology", Wiley Interscience, New York, Chapter 12, 1981; (c). H. van den Ende. In "The Filamentous Fungi", Eds., J. E. Smith, D. R. Berg. J. Wiley & Sons, New York, volume 3, Chapter 13, 1978. 4. The screening of large numbers of marine organisms has lead to generalities about the distributions of toxins and other bioactive substances, see: A. J. Weinheimer, J. A. Matson. Lloydia., 23., 378, 1975. 5. (a). R. E. Moore, G. Bartolini, J. Barchi, A. A. Bothner-By, J. Dodak, J. Ford. J. Am. Chem. Soc, 104. 3776, 1982, and references cited therein. For correction see: ibid.. p. 5572; (b). J. K. Cha, W. J. Christ, J. M. Finan, H. Fujioka, Y. Kishi, L. L. Klein, S. S. Ko, J. Leder, W. W. McWhorter, Jr., K-P. Pfaff, M. Yonaga, D. Uemura, Y. Hirata. J. Am. Chem. Soc, 104. 7369, 1982, and references cited therein. 6. Y. Lin, M. Risk, S. M. Ray, D. Van Engen, J. Clardy, J. Golik, J. C. James, K. J. Nakanishi. J. Am. Chem. Soc, 103. 6773, 1981. 7. For detailed definitions of these words see: M. Barbier. In "Marine Natural Products, Chemical and Biological Perspectives", Ed., P. J. Scheuer, Academic Press, New York, volume 4, p. 147, 1981. 8. G. R. Petit, Y. Kamano, C. L. Herald, M. Tozawa. J. Am. Chem. Soc, 106. 6768, 1984, and references cited therein. 9. (a). H. Kikuchi, Y. Tsukitani, H. Nakanishi, I. Shimizu, S. Saitoh, K. Iguchi, Y. Yamada. Chem Lett., 233, 1982; (b). H. Kuchi, Y. Tsukitani, H. Nakanishi, I. Shimizu, S. Saitoh, K. Iguchi, Y. Yamada. Chem. Pharm. Bull., 21, 1172, 1983. 10. R. Sakai, T. Higa, C. W. Jefford, G. Bernardinelli. J. Am. Chem. Soc, HL8_, 6404, 1986. 11. C. Pathirana, R. J. Andersen. J. Amer. Chem. Soc, lMi 8288, 1986. 258 12. T. Okaichi, Y. Hashimoto. Agri. Biol. Chem., 2fL, 224, 1962, and refernces cited therein. 13. Y. Hashimoto, M. Sakai, K. Konishi. In "Proceedings of the Pood-Drugs from the Sea Conference 1972", Ed., L. R. Worthen. Marine Technology Society, Washington D.C., p. 129, 1972. 14. (a). W. Bergmann, R. J. Feeney. J. Org. Chem., 1£, 981, 1951; (b). W. Bergmann, D. C. Burkholder. ibjLd.., 2SL, 1501, 1955. 15. G. Cimino, S. de Rosa, S. de Stephano. Experientia., 10., 339, 1984. 16. S. S. Cohen. Perspect. Biol. Med., £, 215, 1963. 17. P. N. Kaul. Fed. Proc. Fed. Am. Soc. Exp. Biol., ML, 10, 1981. 18. P. N. Kaul, C. J. Sinderman, Eds.. In "Drugs and Food from the Sea, Myth or Reality". University of Oklahoma Press, Norman, 1978. 19. K. L. Rinehart. Jr., J. B. Gloer, R. G. Hughes. Jr., H. E. Renis, J. P. McGovren, E. B. Swynenberg, D. H. Stringfellow, S. L. Kuentzel, L. H. Li. Science., 212, 933, 1981. 20. For effectiveness against cutaneous Herpes simplex virus in mice see: S. D. Weed, D. A. Stringf ellow. Antiviral. Res., 1, 269, 1983. 21. (a). P. J. Scheuer, Ed.. "Marine Natural Products, Chemical and Biological Perspectives", Academic Press, New York, volume 6, 1986 ; (b). ib_id.., volume 5, 1983; (c). ihi&., volume 4, 1981; (d). ib_id., volume 3, 1980; (e). ibjjd.., volume 2, 1979; (f). ibid.. volume 1, 1978; (g). P. J. Scheuer. "Marine Natural Products", Academic Press, New York, 1973. 22. (a). D. J. Faulkner. Tetrahedron., 13., 1421, 1977; (b) . D. J. Faulkner. Natural Products Reports., 1, 251, 1984; (c) . i£id_., 1, 551, 1984; (d). ibid.. 1, 1, 1986. 23. J. T. Baker. Pure. Appl. Chem., Mr 35, 1976. 24. C. Christophersen, N. Jacobsen. Annu. Rep. Prog. Chem. Sect. B., 7_6_, 433, 1980. 25. (a). D. J. Faulkner, R. J. Andersen. In "In The Sea", Ed., E. D. Goldberg. Wiley, New York, volume 5, p. 679, 1974; (b). D. J. Faulkner, W. Fenical, Ed.. "Marine Natural Products Chemistry", Plenum Press, New York, 1977; (c). D. J. Faulkner. In "Topics In Antibiotic Chemistry", Ed., P. G. Sammes. Ellis Horwood, volume 2, p. 9, 1978; (d). D. J. Faulkner. In "Handbook of Environmental Chemistry", Ed., 0. Hutzinger. Springer-Verlag, Berlin, volume 1, p. 229, 1980. 259 26. (a). W. Fenical. In "Topics In The Biochemistry of Natural Chemistry", Ed., G. R. Waller. Plenum Press, New York, volume 13, p. 219, 1975; (b), W. Fenical, 0. J. McConnell. In "Marine Algae in Pharmaceutical Science", Ed., Y. Tanaka. Walter de Gruyter Company, Berlin, p. 403, 1979; (c). W. Fenical. In "Organic Chemistry Of Seawater", Ed., E. K. Duursma. Elsevier, Amsterdam, 1980. (d). W. Fenical. Science., 215., 923, 1982; (e). W. Fenical. In "Handbook of Biosolar Resources", Ed., 0. R. Zaborsky. CRC Press, volume 1, p. 467, 1982. 27. R. E. Moore. Aliphatic & Related Natural Product Chemistry, Ed., F. D. Gunstone (Specialist Periodical Reports). The Chemical Society, London, volume 1, p. 20, 1979. 28. P. J. Scheuer. Isr. J. Chem., 1£, 52, 1977. 29. L. S. Shields, K. L. Rinehart, Jr.. In "Antibiotics", Ed., G.H.Wagman.- Elsevier, Amsterdam, p. 309, 1978. 30. See for example: D. W. Behrens. In "Pacific Coast" Nudibranchs, A Guide To The Opisthobranchs Of The Northeastern Pacific", Sea Challengers, Los Osos, California, 1980. 31. T. E. Thompson, G. H. Brown. In "British Opisthobranch Molluscs", Academic Press, London 1976. 32. W. C. Austin. In "An Annotated Checklist Of Marine Invertebrates In The Cold Temperate Northeast Pacific", Khoyatan Marine Laboratory, Cowichan Bay, British Columbia, volumes 1, 2, and 3, 1985. 33. H. L. Sleeper, V. J. Paul, W. Fenical. J. Chem. Ecol., £, 57, 1980, and references cited therein. 34. L. G. Harris. In "Current Topics in Comparative Pathobiology", Academic Press, New York, volume 2, p. 213, 1973, and refences cited therein. 35. T. E. Thompson. J. Mar. Biol. Assoc. U.K., 3JL, 123, 1960. 36. C. D. Todd. Oceanogr. Mar. Biol. Ann. Rev., 19_, 141, 1981. 37. D. J. Faulkner, M. T. Ghiselnin. Mar. Ecol. Prog. Ser., 11, 295, 1983, and references cited therein. 38. M. Tischler. M.Sc, Dissertation, University Of British Columbia, 1987. 39. R. T. Paine. Veliger., £, 1, 1964. 40. T. E. Thompson, I. Bennett. Science., 166. 1532, 1969, and refences cited therein. 41. (a). T. E. Thompson. J. Mar. Biol. Assoc. U.K., 19_, 115, 260 1960. (b). M. Edmunds. Proc. Malac. Soc. Lond., 23, 121, 1968; (c). T. E. Thompson. Aust. J. Zool., Hi 755, 1969. 42. For example see: (a). M. R. Hagadone, 3. J. Burreson, P. J. Scheuer, J. S. Finer, J. Clardy. Helv. Chim. Acta., £2r 2484, 1979; (b). G. Cimino, S. de Rosa, S. de Stephano, G. Sodano. Comp. Biochem. Physiol (B)., 23_, 471, 1982; (c). J. E. Thompson, R. P. Walker, S. J. Wratten, D. J. Faulkner. Tetrahedron., 2_8_, 1865, 1982. 43. For example see: (a). H. P. K.Agersberg. Am. Natur., 5_5_, 222 1921; (b). M. Edmunds noted that most nudibranch skin secretions are non-acidic in reference 41b; (c). R. E. Johannes. Veliger., 5r 104, 1963. 44. B. J. Burreson, P. J. Scheuer, J. Finer, J. Clardy. J. Am. Chem. Soc, £2, 4763, 1975. 45. For further examples of metabolites obtained directly from dietary source see: (a). J. C. Coll, B. F. Bowden, D. M. Tapiolas, R. H. Willis, P. Djura, M. Streamer, L. Trott. Tetrahedron., 41, 1085, 1985; (b). D. Mebs. J. Chem. Ecol., H, 713, 1985. 46. J. H. Hochlowski, D. J. Faulkner. Tetrahedron Lett., 22, 271 1981. 47. G. Cimino, S. de Stefano, L. Minale. Experientia., 3J1, 846, 1974. 48. G. Cimino, S. de Rosa, S. de Stefano, G. Sodano, G. Villani. Science., 219. 1237, 1983. For isolation see: Reference 54b. 49. K. Gustafson, R. J. Andersen, M. H. M. Chen, J. Clardy, J. E Hochlowski. Tetrahedron Lett., 25., 11, 1984. 50. For example see: S. W. Ayer, R. J. Andersen, C. -H. He, J. Clardy. J. Org. Chem., A3., 2653, 1984. 51. K. Gustafson, R. J. Andersen, H. Cun-heng, J. Clardy. Tetrahedron. Lett., 23, 2521, 1985. 52. For examples see: (a). Reference 42b; (b). Reference 42c; (c). Reference 2a; (d). G. Schulte, P. J. Scheuer, 0. J. McConnell. Helv. Chim. Acta., £3_, 2159, 1980. 53. For examples see: (a). Reference 42b; (b). Reference 2a; (c) R. K. Okuda, P. J. Scheuer, J. E. Hochlowski, R. P. Walker, D. J Faulkner. J. Org. Chem., A3., 1866, 1983; (d). K. Gustafson. Ph.D Dissertation, University of British Columbia, 1984. 54. For example see: (a). D. Brewer, A. Taylor. J. Chem. Soc Chem. Commum., 1061, 1979; (b). G. Cimino, S. de Rosa, S. de Stefano, R. Morrone, G. Sodano. Tetrahedron., Al, 1093, 1985; (c). K. Nakanishi, I. Kubo. Isr. J. Chem., 1£, 28, 1977. 261 55. Dialdehydes are widespread in the defensive secretions of many insects, for example see: (a). R. Baker, P. H. Briner, D. A. Evans. J. Chem. Soc. Chem. Commum., 411, 1978; (b). J. Meinwald, T. H. Jones, T. Eisner, K. Hicks. Proc. Natl. Acad. U.S.A., 74. 2189, 1977. 56. See for example: (a). From Sacoglossans: (i). C. Ireland, D. J. Faulkner. Tetrahedron, 22, 233, 1981; (ii). C. Ireland, D. J. Faulkner, B. A. Solheim, J. Clardy. J. Am. Chem. Soc, 100. 1002, 1978; (iii). C. Ireland, D. J. Faulkner, J. S. Finer, J. Clardy. J. Am. Chem. Soc, 101. 1275, 1979; (b) . From Siphonariids: (i). J. E. Hochlowski, D. J. Faulkner. J. Org. Chem., Al, 3838, 1984; (ii). J. E. Hochlowski, J. C. Coll, D. J. Faulkner, J. E. Biskupiak, C. M. Ireland, Q. -T. Zheng, C. -H. He, J. Clardy. J. Am. Chem. Soc, 106. 6748, 1984. 57. (a). G. J. Bakus. Biotropica., £, 229, 1974; (b). G. J. Bakus, G. Green. Science. N.Y., 185. 951, 1974; (c). G. Green. Mar. Biol., AD., 207, 1977. 58. H. Bertsch, S. Johnson. Ciencias. Marinas (Mex)., £, 125,1982. 59. G. R. Schulte, P. J. Scheuer. Tetrahedron., 2£, 1857, 1982. 60. S. W. Ayer. Ph.D., Dissertation, University of British Columbia, 1985. 61. See for example: (a). From Sacoglossans: M. B. Ksebati, F. J. Schmitz. J. Org. Chem., £fi,, 5637, 1985; (b). From Siphonar iids: (i). D. C. Manker, D. J. Faulkner, C-F. Xe, J. Clardy. J. Org. Chem., 5JL, 814, 1986; (ii). J. E. Biskupiak, C. M. Ireland. Tetrahedron. Lett., 2£, 4307, 1985; (iii). D. M. Roll, J. E. Biskupiak, C. L. Mayne, C. M. Ireland. J. Am. Chem. Soc, 108. 6680, 1986. (c). From Prosobranch molluscs: (i). R. J. Andersen, D. J. Faulkner, C-H. He, G. D. Van Duyne, J. Clardy. J. Am. Chem. Soc, 107. 5492, 1985; (ii). K. F. Albizati, J. R. Pawlik, D. J. Faulkner. J. Org. Chem., J5_Q.r 3428, 1985; (d) . From Cephalaspidean molluscs: (i). S. J. Coval, P. J. Scheuer. J. Org. Chem., 50. 3024, 1985; (ii). S. J. Coval, G. R. Schulte, G. K. Matsumoto, D. M. Roll. P. J. scheuer. Tetrahedron. Lett., 2JL, 5359, 1985. 62. For example see: (a). T. Miyamoto, R. Higuchi, T. Komori, T. Fujioka, K. Mihashi. Tetrahedron. Lett, 22, 1153, 1986; (b). T. Ichiba, T. Higa. J. Org. Chem., 5_1, 3364, 1986; (c). R. Sakai, T. Higa, C. W. Jefford, G. Bernardinelli. Helv. Chim. Acta., £9_, 91, 1986. 63. K. N. Gulavita, E. D. DeSilva, M. R. Hagadone, P. Karuso, P. J. Scheuer, G. D. Van Duyne, J. Clardy. J. Org. Chem., 5J., 5136, 1986. 64. T. F. Molinski, D. J. Faulkner. J. Org. Chem., 5J., 2601, 1986. 262 65. T. F. Molinski, D. J. Faulkner, H. Cun-Heng, G. D. Van Duyne, J. Clardy. J. Org. Chem., 5±, 4564, 1986. 66. B. Carte, M. R. Kernan, E. B. Barrabee, D. J. Faulkner, G. K. Matsumoto, J. Clardy. J. Org. Chem., $1, 3528, 1986. 67. G. Cimino, A. Crispino, S. de Stefano, M. Gavagnin, G. Sodano. Experientia., 12, 1301, 1986. 68. J. A. Roesener, P. J. Scheuer. J. Am. Chem. Soc, 1D_8_, 846, 1986. 69. S. Matsunaga, N. Fusetani, K. Hashimoto, K. Koseki, M. Noma., J. Am. Chem. Soc, 108. 847, 1986. 70. R. P. Walker, D. J. Faulkner. J. Org. Chem., Mi, 1475, 1981. 71. F. A. Fuhrman, G. J. Fuhrman, R. J. Nachman, H. S. Mosher. Science., 212. 557, 1981. 72. R. M. Harbo. In "Tidepool and Reef Marine Life Guide to the Pacific Northwest Coast", Hancock House, Vancouver, 1980. 73. Recently developed by Varian Associates, Inc., see: XL- Series, NMR Spectrometer System, Advanced Operation, Publication No. 87-146056, Revision B1084, p. 2-53, 1984. 74. G. Cimino, S. de Rosa, S. de Stefano, G. Sodano. Experientia., M, 246, 1984. 75. A. Fukuzawa, Y. Kumagi, T. Masamune, A. Furusaki, C. Katayama, T. Matsumoto. Tetrahedron. Lett., 22, 4085, 1981. 76. Proposed and developed by: A. Bax, G. A. Morris. J. Magn. Reson., 42, 501, 1981, and references cited therein. 77. Z. Saatov, N. D. Abdullaev, M. B. Gorovits, N. K. Abubakirov. Khim. Prir. Soedin., 467, 1984 (English version, 441, 1984). 1 3 78. E. Breitmaier, W. Voelter. In ""c NMR Spectroscopy Methods and Applications In Organic Chemistry", Monographs in Modern Chemistry, series Ed., H. F. Ebel, 2nd Edition, Verlag Chemie, Weinheim, New York, volume 5, 1978. 79. A. Murai, M. Ono, A. Abiko, T. Masamune. Bull. Chem. Soc. Jpn., 5£r 1195, 1982. 80. (a). M. L. Mihailovic, J. Forsek, L. Lorenc, Z. Maksimovic, H. Fuhrer, J. Kalvoda. Helv. Chim. Acta., 52., 459, 1969; (b) . K. Shimada, T. Nambara, I. Uchida, S. M. Kupchan. Heterocycles., 12, 1445, 1979. 81. S. Honma, S. Iwamura, K. Iizuka, A. Kambegawa, K. Shida. Chem. Pharm. Bull., 25., 2019, 1977. 263 82. D. J. Vanderah, C. Djerassi. J. Org. Chem., 12, 1442, 1978. 83. A. Horeau, A. Nouaille. Tetrahedron. Lett., 1939, 1971, and references cited therein. 84. Evidence for the importance of the direct release of metabolites into the mouth of fish stems from the fact that fish have been observed to consume dead nudibranchs which when alive had been unpalatable. See reference 35. 85. E. Fattorusso, V. Lanzotti, S. Magno, E. Novellino. J. Org. Chem., 5JI, 2868, 1985. 86. T. Higa, J. Tanaka, Y. Tsukitani, H. Kikuchi. Chem. Lett., 1647, 1981. 87. M. Nakatani, H. Takao, I. Miura, T. Hase. Phytochemistry., 21, 1945, 1985. 88. (a). S. W. Ayer, R. J. Andersen. Tetrahedron. Lett., 22, 1039, 1982; (b). Reference 60. 89. (a). G. Cimino, S. de Rosa, S. de Stefano, G. Sodano. Tetrahedron. Lett., 21, 3303, 1980; (b). G. Cimino, S. de Stefano, S. de Rosa, G. Sodano, G. Illani. Bull. Soc. Chim. Belg., ££, 1069, 1980. 90. R. J. Stonard, J. C. Petrovitch, R. J. Andersen. Steroids., 2£, 81, 1980. 91. For examples see: (a). Reference 87; (b). R. W. Miller, J. Clardy, J. Kozlowski, K. L. Mikolajezak, R. D. Plattner, R. G. Powell, C. R. Smith, D. Weisleder, Z. Qi-Tai. Planta. Med., 40, 1985; (c). H. Schildknecht. Angew. Chem. Int. Ed. Engl., £, 1, 1970, and references cited therein; (d). Reference 2b, p. 198; (e). A. Sturararo, A. Guerriero, R. de Clauser, F. Pietra. Experientia., 2fL, 1184, 1982; (f). A. Guerriero, P. Traldi, F. Pietra. J. Chem. Soc. Chem. Commun., 40, 1986. 92. Private communication to Dr. R. J. Andersen from Dr. D. J. Faulkner of the Scripps Institution of Oceanography. 93. R. D. Barnes. In "Invertebrate Zoology", 2 Ed., W. B. Saunders Company, p. 77-112, 1968. 94. G. O. Mackie, Ed.. In "Coelenterate Ecology and Behaviour", Plenum Press, New York, 1976. 95. L. Muscatine, H. M. Lenhoff, Ed.. In "Coelenterate Biology, Reviews and New Perspectives", Academic Press, New York, 1974. 96. P. Tardent, R. Tardent, Ed.. In "Developmental and Cellular Biology of Coelenterates", Elsevier/North-Holland, Biomedical Press, Amsterdam, 1980. 264 97. For well illustrated examples of Alcyonarians see: F. M. Bayer, M. Grasshoff, J. Verseveldt, Ed.. In "Illustrated Trilingual Glossary of Morphological and Anatomical Terms Applied To Octacorallia", E. J. Brill, Leiden, The Netherlands, 1983. 98. B. Tursch, J. C. Braekman, D. Daloze, M. Kaisin. In "Marine Natural Products, Chemical and Biological Perspectives", Ed., P. J. Scheuer, Academic Press, New York, volume 2, p. 247, 1978. 99. For examples see: (a). D. L. Taylor. In "Symbiosis In The Sea", Ed., W. B. Vernberg, University of South Carolina Press, Columbia, South Carolina, p. 245, 1972; (b). D. L. Taylor. Annu. Rev. Microbiol., 22, 171, 1973; (c). D. L. Taylor. Symp. Soc. Exp. Biol., 22, 267, 1975; (d). R. K. Trench. Annu. Rev. Plant. Physiol., 22, 485, 1979. 100. M. R. Droop. Symp. Soc. gen. Microbiol., 12, 171, 1963. 101. A. R. Loeblich III., J. L. Sherley. J. Mar. Biol. Ass. U.K., 52, 195, 1979. 102. (a). W. C. M. C. Kokke, W. Fenical, L. Bohlin, C. Djerassi. Comp. Biochem. Physiol (B)., £J3_, 281, 1981; (b). N. W. Withers, W. C. M. C. Kokke, W. Fenical, C. Djerassi. Proc. Natl. Acad. Set. U.S.A., 22, 3764, 1982; (c) . ibid.. 22, 6390, 1982; (d). W. C. M. C. Kokke, S. Epstein, S. A. Look, G. H. Rau, W. Fenical, C. Djerassi. J. Biol. Chem., 259. 8168, 1984; (e). D. A. Schoenberg, R. K. Trench. Proc. R. Soc. Lond. (B)., 207. 405, 1980; (f). ibid., 222, 429, 1980; (g). ibid.. 202, 445, 1980; (h). R. K. Trench. Pure. Appl. Chem. , 52, 819, 1981; (i). D. A. Schoenberg, R. K. Trench. In reference 85: p. 423; (j). R. J. Blank, R. K. Trench. Science., 229. 656, 1985; (k). R. A. Kinzie III., G. S. Chee. Biol. Bull., 15£, 315, 1979. 103. (a). C. Von Holt, M. Von Holt. Comp. Biochem. Physiol., 24, 73, 1968; (b). R. K. Trench. Proc. R. Soc. Lond. (B)., 177. 225, 1971; (c). L. Muscatine in reference 86: Chapter IX, P. 359-395, and references cited therein; (d). D.;L. Taylor. Adv. Mar. Biol., 11, 1, 1973. 104. R. K. Trench. Proc. R. Soc. Lond. (B)., 177. 251, 1971. 105. C. M. Young. Adv. Mar. Biol. 1, 209, 1963. 106. For an early summary of these debates upto 1978 see: Reference 98. 107. L. S. Ciereszko, M. A. Johnson, R. W. Schmidt, C. B. Koons. Comp. Biochem. Physiol., 24, 899, 1968. 108. F. J. Schmitz, T. Pattabhiraman. J. Am. Chem. Soc, 92. 6073, 1970. 109. Isolated from samples collected in the Gulf of Mexico: W. 265 Fenical, R. K. Okuda, M. M. Bandurraga, P. Culver, R. S. Jacobs. Science., 212, 1512, 1981. 110. Isolated from samples collected in the Las Tres Marias Islands (Pacific coast of Mexico): M. M. Bandurraga, Ph.D., Dissertation, University of California at San Diego, private communication to Dr.R. J. Andersen. 111. M. M. Bandurraga, B. McKittrick, W. Fenical, E. Arnold, J. Clardy. Tetrahedron., 2£, 305, 1982. 112. R. R. Izac, M. M. Bandurraga, J. M. Wasylyk, F. W. Dunn, W. Fenical. Tetrahedron., 23., 301, 1982. 113. R. E. Schwartz, P. J. Scheuer, V. Zabel, W. H. Watson. Tetrahedron., 22, 2725, 1981. 114. L. M. G. Missakia, B. J. Burreson, P. J. Scheuer. Tetrahedron., 21, 2513, 1975. 115. B. F. Bowden, J. C. Braekman, J. C. Coll, S. J. Mitchell. Aust. J. Chem., 22, 927, 1980. 116. Reference 45a, and references cited therein. 117. For examples see: (a). B. N. Ravi, J. F. Marwood, R. J. Wells. Aust. J. Chem., 22, 2307, 1980; (b). A. Clastres, P. Laboute, A. Ahond, C. Poupat, P. Potier. J. Nat. Prod., 42, 162, 1984; (c). A. Clastres, P. Laboute, A. Ahond, C. Poupat, P. Potier, S. K. Kan. J. Nat. Prod., 42, 155, 1984; (d). M. N. Do, K. L. Erickson. J. Org. Chem., A3, 4410, 1983. 118. For examples see: (a). Reference 5; (b). C. B. Rao, A. S. R. Anjaneyulu, N. S. Sarma, Y. Venkateswarlu, R. M. Rosser, D. J. Faulkner, W. H. M. Chen, J. Clardy. J. Am. Chem. Soc, 106. 7983, 1984; (c). R. E. Schwartz, M. B. Yunker, P. J. Scheuer, T. Ottersen. Can. J. Chem., 31, 1707, 1979; (d). S. Takano, D. Uemura, Y. Hirata. Tetrahedron. Lett., 4909, 1978. 119. For examples see: (a). M. Kobayashi, T. Yasuzawa, M. Yoshihara, H. Akutsu, Y. Kyogoku, I. Kitagawa. Tetrahedron. Lett., 22, 5331, 1982; (b). M. Endo, M. Nakagawa, Y. Hamamoto, T. Nakanishi. J. Chem. Soc. Chem. Commun., 322, 1983; (c). B. J. Baker, R. K. Okuda, P. T. K. Yu, P. J. Scheuer. J. Am. Chem. Soc., 1£2, 2976, 1985. 120. G. J. Bakus. Science., 211, 497, 1981. 121. J. C. Coll, B. F. Bowden, G. M. Konig, R. Braslau. Bull. Soc. Chim. Belg., 23, 815, 1986, and references cited therein. 122. For example see: (a). M. K. Wickstein, J. D. DeMartini. Veliger., 13, 195, 1973; (b). F. E. Perron, R. D. Turner. J. Molluscan. Stud., 44, 100, 1978; (c). See reference 45a; (d). For a mollusc that probably feeds on an Alcyonarian see: G. S. Linz, 266 R. Sanduja, A. J. Weinheimer, M. Alam, G. E. Martin. Tetrahedron. Lett., 21, 4833, 1986. 123. B. Tursch. Pure. Appl. Chem., A3, 1, 1976. 124. Correct structure obtained by X-ray analysis-B. Tursch, J. C. Braekman, D. Daloze, H. Dedeurwaerder, R. Karlsson. Bull. Soc. Chim. Belg., 31, 75, 1978. 125. Release into the water column has been demonstrated for soft corals, see: J. C. Coll, B. F. Bowden, D. M. Tapiolas, W. C. Dunlap. J. Exp. Biol. Ecol., 33, 293, 1982. 126. R. Kazlauskas, P. T. Murphy, R. J. Wells, P. Schonholzer, J. C. Coll. Aust. J. Chem., 21, 1817, 1978. 127. B. F. Bowden, J. C. Coll, N. Liyanage, S. J. Mitchell, G. J. Stokie, I. A. Van Altena. Aust. J. Chem., 21, 163, 1978. 128. J. C. Coll, P. W. Sammario. Toxicon. Suppl. 3, 69, 1983. 129. S. J. Wratten, D. J. Faulkner, K. Hirotsu, J. Clardy. J. Am. Chem. Soc, 22, 2824, 1977. 130. M. M. Bandurraga, W. Fenical. Tetrahedron., 41, 1057, 1985. 131. For example see: (a). J. M. Frincke, D. E. Mclntyre, D. J. Faulkner. Tetrahedron. Lett., 21, 735, 1980; (b). J. Kobayashi, Y. Ohizumi, H. Nakamura, T. Yamakado, T. Matsuzaki, Y. Hirata. Experientia., 22., 67, 1983; (c). D. J. Estrella, R. S. Jacobs. Fed. Proc, 42, 67, 1983. 132. For example see: (a). M. Albericci, J. C. Braekman, D. Daloze, B. Tursch, J. P. Declercq, G. Germain, M. Van Meerssche. Bull. Soc. Chim. Belg., 31, 487, 1978; (b). Y. Kashman, A. Groweiss. Tetrahedron. Lett., 1159, 1977. 133. For example see: A. J. Weinheimer, J. A. Matson, M. B. Hossain, D. van der Belm. Tetrahedron. Lett., 2923, 1977. 134. R. Kazlauskas, P. T. Murphy, B. N. Ravi, R. L. Sanders, R. J. Wells. Aust. J. Chem., 25., 69, 1982. 135. W. Fenical. In "Marine Natural Products Chemistry, Chemical and Biological Perspectives", Ed., P. J. Scheuer, Academic Press, New York, volume 2, p. 173, 1978. 136. A. J. Weinheimer, C. W. J. Chang, J. A. Matson. Fortschr. Chem. Org. Naturst., 23, 285, 1979. 137. For examples see: (a). J. R. Proudfoot, C. Djerassi, J. Thompson. Steroids, 45., 263, 1985; (b). T. Gebreyesus, I. Stoilov, F-T. Luo, C. Djerassi. Steroids, 45., 447, 1985; (c). B. M. Jagodzinska, J. S. Trimmer, W. Fenical, C. Djerassi. J. Org. Chem., 5J1, 2988, 1985. 267 138. R. Li, K. Long, C. W. Zhao, Z. Fang, M. Zhang. Zhongshan Daxue Xuebao, Ziran Kexueban., 50, 1985. 139. R. Li, K. Long, Z. Fang, H. Zhao, M. Zhang. Zhongshan Daxue, Xuebao, Ziran Kexueban., 17, 1985. 140. B. F. Bowden, J. C. Coll, L. M. Engelhardt, D. M. Tapiolas, A. H. White. Aust. J. Chem., 21, 103, 1986. 141. B. F. Bowden, J. C. Coll, R. H. Willis. Aust. J. Chem., 21, 1717, 1986. 142. A. Guerriero, B. Dematte, M. D'Ambrosio, F. Pietro. J. Nat. Prod., Al, 608, 1986. 143. Y. Uchio, S. Eguchi, J. Kuramoto, M. Nakayama, T. Hase. Tetrahedron. Lett., 23, 4487, 1985. 144. B. F. Bowden, J. C. Coll, L. M. Engelhardt, G. V. Meehan, G. G. Pegg, D. M. Tapiolas, A. H. White, R. H. Willis. Aust. J. Chem., 21, 123, 1986. 145. M. D'Ambrosio, D. Fabbri, A. Guerriero, F. Pietra. Helv. Chim. Acta., 20., 63, 1987. 146. M. Aoki, T. Kato, Y. Uchio, M. Nakayama, M. Kodama. Bull. Chem. Soc. Jpn., 33, 779, 1985. 147. A. Sato, W. Fenical, Z. Qi-Tai, J. Clardy. Tetrahedron, Al, 4303, 1985. 148. Named after the metabolite briarein A reported by: J. E. Burks, D. van der Helm, C. Y. Chang, L. S. Ciereszko. Acta. Crystallogr (B)., 21, 704, 1977. 149. M. B. Ksebati, F. J. Schmitz. Bull. Soc. Chim. Belg., 23., 835, 1986. 150. D. J. Vanderah, P. A. Steudler, L. S. Ciereszko, F. J. Schmitz, J. D. Ekstrand, D. van der Helm. J. Am. Chem. Soc, 22., 5780, 1977. 151. A. Groweiss, Y. Kashman. Tetrahedron. Lett., 2205, 1978. 152. B. F. Bowden, J. C. Coll, J. M. Gulbis, M. F. Mackay, R. H. Willis. Aust. J. Chem., 21, 803, 1986. 153. I. Kitagawa, Z. Cui, Y. Cai, M. Kobayashi, Y. Kyogoku. Chem. Pharm. Bull., 2A, 4641, 1986. 154. A. J. Blackman, A. Heaton, B. W. Skelton, A. H. White. Aust. J. Chem., 23, 565, 1985. 155. J. Su, K. Long, T. Peng, Q. Zheng, X. Lin. Huaxue Xeudao 268 (Acta. Chimica. Sinica.)., 12, 796, 1985. 156. J. Su, K. Long, T. Pang, C-H. He, J. Clardy. J. Am. Chem. Soc, Mi 177, 1986. 157. S. A. Look, M. T. Burch, W. Fenical, Q. Zheng, J. Clardy. J. Org. Chem., 5£, 5741, 1985. 158. N. Fusetani, K. Yasukawa, S. Matsunaga, K. Hashimoto. Tetrahedron. Lett., 2fL, 6449, 1985. 159. S. A. Look, W. Fenical, G. K. Matsumoto, J. Clardy. J. Org. Chem., 51, 5140, 1986. x 160. For examples see: (a). K. Iguchi, S. Kaneta, K. Mori, Y. Yamada, A. Honda, Y. Mori. Tetrahedron. Lett., 2fL, 5787, 1985; (b). I. Kitagawa, M. Kobayashi, T. Yasuzawa, B. -W. Son, M. Yoshihara, K. Kyogoku. Tetrahedron., 11, 995, 1985. 161. K. Iguchi, S. Kaneta, K. Mori, Y. Yamada, A. Honda, Y. Mori. J. Chem. Soc. Chem. Commun., 981, 1986. 162. K. Ikuchi, K. Mori, M. Suzuki, H. Takashashi, Y. Yamada. Chem. Lett., 1789, 1986. 163. M. Kobayashi, B. -W. Son, Y. Kyogoku, I. Kitagawa. Chem. Pharm. Bull., 21, 2306, 1986. 164. M. D'Ambrosio, A. Guerriero, D. Fabbri, F. Pieta. Helv. Chim. Acta., £9_, 1581, 1986. 165. P. A. Keifer, K. L. Rinehart, Jr, I. R. Hooper. J. Org. Chem., 51, 4450, 1986. 166. B. F. Bowden, J. C. Coll, S. J. Mitchell. Aust. J. Chem., 22, 1833, 1980. 167. Y. M. Sheikh, G. Singy, M. Kaisin, H. Eggert, C. Djerassi, B. Tursch, R. Karlsson. Tetrahedron., 22, 2115, 1977. 168. B. Tursch, M. Colin, D. Daloze, D. Losman, R. Karlsson. Bull. Soc. Chim. Belg., £1, 81, 1975. 169. R. R. Izac, M. M. Bandurraga, J. M. Wasylyk, F. W. Dunn, W. Fenical. Tetrahedron., 28., 301, 1982. 170. B. Tursch, J. C. Braekman, D. Daloze, P. Fritz, A. Kelecom, R. Karlsson, D. Losman. Tetrahedron. Lett., 747, 1974. 171. D. Daloze, J. C. Braekman, P. Georget, B. Tursch. Bull. Soc. Chim. Belg., ££, 47, 1977. 172. B. F. Bowden, J. C. Coll, S. J. Mitchell, B. W. Skelton, A. H. White. Aust. J. Chem., 22, 2737, 1980. 269 173. R. R. Izac, W. Fenical, B. Tagle, J. Clardy. Tetrahedron., 22, 2569, 1981. 174. H. Kikuchi, Y. Tsukitani, Y. Yamada, K. Iguchi, S. A. Dexler, J. Clardy. Tetrahedron Lett., 21, 1063, 1982. 175. B. F. Bowden, J. C. Coll, S. J. Mitchell. Aust. J. Chem., 11, 681, 1980. 176. (a). A. Ahond, J. C. Coll, J. D. Fourneron. Tetrahedron. Lett., 1879, 1979; (b). A. Ahond, A. Chiaroni, J. C. Coll, J. D. Fourneron, C. Riche, J. C. Braekman, D. Daloze, B. Tursch, P. J. Dunstan. Bull. Soc. Chim. Belg., Mr 313, 1979. 177. B. F. Bowden, J. C. Coll, S. J. Mitchell. Aust. J. Chem., 21r 885, 1980. 178. R. R. Izac, P. Schneider, M. Swain, W. Fenical. Tetrahedron. Lett., 21r 817, 1982. 179. F. J. Schmitz, D. J. Vanderah, L. S. Ciereszko. Chem. Commun., 407, 1974. 180. Y. Kashman, M. Bodna, Y. Loya, Y. Bernayahu. Isr. J. Chem., 1£, 1, 1977. 181. B. Tursch, J. C. Braekman, D. Daloze. Bull. Soc. Chim. Belg., Mr 767, 1975. 182. A. J. Blackman, B. F. Bowden, J. C. Coll, B. Frick, M. Mahendran, S. J. Mitchell. Aust. J. Chem., Mr 1873, 1982. 183. D. J. Vanderah, N. Rutledge, F. J. Schmitz, L. S. Ciereszko. J. Org. Chem., H, 1614, 1978. 184. S. E. Poet, B. N. Ravi. Aust. J. Chem., Mr 77, 1982. 185. A. Ahond, B. F. Bowden, J. C. Coll, J. -D. Fourneron, S. J. Mitchell. Aust. J. Chem., 2A, 2657, 1981. 186. (a). J. K. Kingston, B. Gregory, A. G. Fallis. Tetrahedron. Lett., 4261, 1977; (b). J. K. Kingston, B. Gregory, A. G. Fallis. J. Chem. Soc. Perkin. Trans. 1., 2064, 1979; (c). J. K. Kingston. Tetrahedron. Lett., 21r 4295, 1980; (d). J. K. Kingston, A. G. Fallis. Can. J. Chem., £Q, 820, 1982. 187. A Naphthea species cited by original authors: J. P. Englebrecht, B. Tursch, C. Djerassi. Steroids., 2D_r 121, 1972. But cited as Sinularia mayi by: J. C. Braekman. in "Marine Natural Products Chemistry", Ed., D. J. Faulkner, W. Fenical, Plenum Press, New York, p. 5, 1977. 188. B. F. Bowden, J. C. Coll. Aust. J. Chem., 21, 2677, 1981. 189. D. W. Gotshall, L. L. Laurent. In "Pacific Coast Subtidal 270 Marine Invertebrates, A Fishwatcher1s Guide", Sea Challengers, Los Osos, California, 1979. 190. For examples see: (a). E. Fahy, R. J. Andersen, He. Cun- heng, J. Clardy. J. Org. Chem., 5_0_, 1149, 1985; (b) . E. Fahy, R. J. Andersen. J. Org. Chem., 5J., 57, 1986; (c) . E. Fahy, R. J. Andersen, C. Xu, J. Clardy. J. Org. Chem., 5JL, 5145, 1986.; (d) . D. E. Williams, R. J. Andersen, G. D. Van Duyne, J. Clardy. J. Org. Chem., £2, 332, 1987. 191. For example see: D. E. Williams, S. W. Ayer, R. J. Andersen. Can. J. Chem., £1, 1527, 1986. 192. For example see: B. Carte, D. J. Faulkner. J. Org. Chem., A3., 2314, 1983. 193. For example see: G. Cimino, S. de Rosa, S. de Stephano, G. Sodano. Tetrahedron. Lett., 21, 3303, 1980. 194. D. E. Williams, R. J. Andersen. Can. J. Chem., in press. 195. D. E. Williams, R. J. Andersen, L. Parkanyi, J. Clardy. Tetrahedron, in press. 196. A. G. Gonzalez, J. M. Arteaga, J. J. Fernandez, J. D. Martin, M. Norte, J. Z. Ruano. Tetrahedron., A3, 2751, 1984. 197. R. M. Silverstein, C. G. Bassler, T. C. Morril. In "Spectroscopic Identification of Organic Compounds", John Wiley and Sons, Chichester, 4th Edition, 1981. When referenced on p. 124 see p. 227. 198. G. C. Levy, G. L. Nelson. In "Carbon-13 Nuclear Magnetic Resonance for Organic Chemists", Wiley-Interscience, John Wiley & Sons, Inc., Toronto, 1972. 199. (a). Proposed and developed by: A. Bax. J. Magn. Reson., 57f 314, 1984; (b). For an example of where this ID NMR heteronuclear chemical shift correlation experiment has been employed see: A. Bax, J. A. Ferretti, N. Nashed, D. M. Jerina. J. Org. Chem., 50. 3031, 1985. 200. T. Suzuki, H. Kikuchi, E. Kurosawa. Bull. Chem. Soc. Jpn., 5_5_, 1561, 1982. 201. For a review on multipulse NMR techniques and includes see: R. Benn, H. Gunther. Angew. Chem. Int. Ed. Engl., 22, 350, 1983. 202. Proposed and developed by: (a). D. M. Droddrell, D. T. Pegg, M. R. Bendall. J. Magn. Reson., A3, 323, 1982; (b). D. T. Pegg, D. M. Droddrell, M. R. Bendall. J. Chem. Phys., 22., 2745, 1982. 203. M. M. Bandurraga, W. Fenical, S. F. Donovan, J. Clardy. J. Am. Chem. Soc, 104. 6463, 1982. 271 204. For an example of the use of the long range 2D-HETC0R NMR experiment see: J. N. Schoolery. J. Nat. Prod., 12, 226, 1984. 205. S. L. Patt, J. N. Schoolery. J. Magn. Reson., A3., 535, 1982, and references cited therein. 206. D. Dolphin, A. Wick. In "Tabulation Of Infrared Spectral Data", Wiley-Interscience, John Wiley and Sons, Toronto, 1977. 207. R. J. Capon, D. J. Faulkner. J. Org. Chem., 19_, 2506, 1984. 208. M. Suzuki, N. Kowata, E. Kurosawa. Bull. Chem. Soc. Jpn., 5A, 2366, 1981. 209. B. F. Bowden, J. C. Coll, D. M. Tapiolas. Aust. J. Chem., 23, 211, 1983. 210. Authentic sample kindly provided by Dr. E. Piers of the Chemistry Department, University of British Columbia. 211. (a). F. Vonasek, V. Herout, F. Sorm. Coll. Czech. Chem. Comm., 23, 919, 1960; (b). A. Tanaka, R. Tanaka, H. Uda, A. Yoshikoshi. J. Chem. Soc. Perkin. Trans. I., 1721, 1972. 212. Y. Ohta, T. Sakai, Y. Hirose. Tetrahedron. Lett., 6365, 1966. 213. (a). S. J. Wratten, W. Fenical. D. J. Faulkner, J. C. Wekell. Tetrahedron. Lett., 1559, 1977; (b). R. L. Hendrikson, J. H. Cardellina II. Tetrahedron., £2, 6565, 1986. 214. From the course notes of Dr R. J. Andersen, Chem 568, Marine Natural Products, taken at the University of British Colunbia in 1985. 215. For examples of closely related metabolites with laurane and cyclolaurane skeletons see: (a). M. Suzuki, E. Kurosawa. Bull. Chem. Soc. Jpn., 32., 3349, 1979; (b) . ihlsL., 32, 3352, 1979. 216. The presence of a facile leaving group, bromine, and an array of structurally related metabolites such as the lauranes (eg., 181 and 183) led to a proposed biogenetic pathway involving chamigrene intermediates. See: A. G. Gonzalez, J. Darias, A. Diaz, J. D. Fourneron, J. D. Martin, C. Perez. Tetrahedron. Lett., 3051, 1976. 217. T. Suzuki, M. Suzuki, E. Kurosawa. Tetrahedron. Lett., 3057, 1975. 218. For examples see refences 208, 211 and 212. 219. Private communication to Dr. R. J. Andersen from Dr. P. J. Scheuer of the University of Hawaii. 272