UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Studies with tripterygium wilfordii : synthesis of diterpene analogues with potential pharmacological… Zetina-Rocha, Carlos B. 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1998-34651X.pdf [ 7.97MB ]
Metadata
JSON: 831-1.0061609.json
JSON-LD: 831-1.0061609-ld.json
RDF/XML (Pretty): 831-1.0061609-rdf.xml
RDF/JSON: 831-1.0061609-rdf.json
Turtle: 831-1.0061609-turtle.txt
N-Triples: 831-1.0061609-rdf-ntriples.txt
Original Record: 831-1.0061609-source.json
Full Text
831-1.0061609-fulltext.txt
Citation
831-1.0061609.ris

Full Text

STUDIES WITH TRIPTERYGIUM WILFORDIL SYNTHESIS OF DITERPENE ANALOGUES WITH POTENTIAL PHARMACOLOGICAL ACTIVITY by CARLOS B. ZETINA-ROCHA B.Sc, Universidad Nacional Autonoma de Mexico, 1982 M.Sc, Universidad Nacional Autonoma de Mexico, 1986  A THESIS SUBMITTED IN PARTIAL FULFILMENT 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 June, 1998 © Carlos B. Zetina-Rocha, 1998  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 of The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  \2  JQU^t^l^g  Abstract  This thesis is concerned with the development of synthetic routes leading to the preparation of a family of novel diterpenoid analogues structurally related to triptolide (1), an active constituent of the Chinese herbal plant Tripterygium wilfordii.  The  objective was to design a methodology which could allow access to a variety of potentially active analogues, which may contribute to a better understanding of the structure-activity relationship of this family of compounds by the evaluation of their pharmacological activity. In the first part of this research two alternative short syntheses to the phenolic diterpene isotriptophenolide  (71) were developed, using the readily  dehydroabietic acid (60) as starting material.  available  Subsequently, this ring C "activated"  compound (71) was used as a key intermediate from which several synthetic pathways were derived, leading to the preparation of a number of new analogues possessing several unsaturated and oxygen functionalities in rings B and C of the molecule. These derivatives comprised quinoid-type compounds such as C8-methoxylated dienones (153) and the (7,8)a-epoxy-dienone, 72. Epoxidation of the latter compound produced two novel di-epoxidized analogues. A different synthetic sequence was also developed to the corresponding (7,8)P-epoxy-isomer (73), which upon epoxidation produced a di-epoxy compound. Another series of mono-epoxidized derivatives were obtained from the 8hydroxy-dienones 157 and 158, prepared by changing the oxidation conditions of  Ill  isotriptophenolide (71). Several alternative pathways were developed to some of the key synthetic intermediates and to the epoxy-dienones 72 and 73. In the later stages of this research, a new "unified" synthetic scheme was achieved which allowed the synthesis of the analogues 72, 73,157 and 158 (and consequently their corresponding epoxidized derivatives), in a remarkably short 1 to 3-step sequence. Also, a short series of biotransformation experiments was performed by incubating the epoxydienones  72  and  73  with  TRP4a  cell  cultures,  giving  several  ring-B  oxygenated/unsaturated derivatives of isotriptophenolide (71). A total of 15 new quinoid/epoxy analogues, belonging to three main "series" of derivatives (i.e., 8-methoxy-, (7,8)epoxy-, and 8-hydroxy-dienones) were prepared in the present research.  60  153  72 = a-epoxide 73 = P-epoxide  71  157 = a-OH 158 = p-OH  IV  Notes  The numbering system used throughout this thesis is that used by contemporary natural products chemists and is illustrated below.  abietane  The abietane skeleton is designed as 18-nor if CI 8 is absent or if a double bond is present at C3 or C4. The rearranged abietane skeleton is designated as 18(4—>-3) abeo.  18-nor  18(4->3)a6eo  V  Table of Contents Abstract  ii  Notes  iv  Table of Contents  v  List of Figures  xi  List of Schemes.  xiii  List of Tables  xv  List of Abbreviations  xvi  Acknowledgements CHAPTER 1 INTRODUCTION  xx 1  1.1  Historical background of the plant Tripterygium wilfordii  1  1.2  Pharmacological properties and clinical uses of refined extracts of Tripterygium wilfordii and triptolide  3  1.2.1 Immunosuppressive and anti-inflammatory properties  4  1.2.2 Anticancer properties  6  1.2.3 Male fertility control  7  1.2.4 Side-effects of GTW and triptolide  9  1.3  1.4  Approaches utilized to produce triptolide and its derivatives  10  1.3.1 Isolation of diterpenoids from Tripterygium species  12  1.3.2 Use of plant cell cultures of T. wilfordii  20  1.3.3 Total syntheses of triptolide and related compounds  26  Attempts to establish a structure-activity relationship for triptolide-related epoxides  36  VI  1.5.  Current research interests  40  1.6  Objectives of the present research  42  CHAPTER 2 RESULTS AND DISCUSSION 2.1  Synthesis of isotriptophenolide (71) 2.1.1  Modified synthesis of isotriptophenolide (71)  43 44 50  2.2  Preliminary experiments involving oxidation of isotriptophenolide (71)..55  2.3  Synthesis of the 8-methoxy-dienones, 153 2.3.1  2.4  2.5  2.6  Attempted epoxidation of the 8-methoxy-dienones, 153  Synthesis of the (7,8)a-epoxy-dienone series  68 73 77  2.4.1  Synthesis ofthe (7,8)a-epoxy-dienone, 72  77  2.4.2  Epoxidation ofthe (7,8)a-epoxy-dienone, 72  88  Synthesis ofthe (7,8)P-epoxy-dienone series 2.5.1  Synthesis ofthe (7,8)p-epoxy-dienone, 73  2.5.2  Epoxidation ofthe (7,8)P-epoxy-dienone, 73  Synthesis ofthe 8-hydroxy series  95 95 100 Ill  2.6.1  Synthesis of 8-hydroxy-dienones  Ill  2.6.2  Epoxidation ofthe 8a-hydroxy-dienone, 157  119  2.6.3  Epoxidation ofthe 8p-hydroxy-dienone, 158  136  2.7  Unified synthesis ofthe (7,8)a-epoxy-, (7,8)P-epoxy- and 8-hydroxy.140 series.  2.8  Attempted epoxidation of quinoid-type analogues by alternative methods 2.8.1  143  Experiments with DMD and enzymes as epoxidizing agents.. 143  Vll  2.8.2  2.9  Biotransformation experiments with TRP4a cell cultures  145  Biotransformation of 72 with TRP4a cell cultures of different ages  147  Biotransformation of (7,8)P-epoxy-dienone, 73  150  Conclusions  154  CHAPTERS EXPERIMENTAL  155  3.1  General  155  3.2  Synthesis of isotriptophenolide (71)  157  3.2.1  Synthesis of isodehydroabietenolide (65)  157  Purification of crude dehydroabietic acid (DHA, 60)  158  18-Norabieta-4(19),8,ll,13-tetraene(134)  159  Acid chloride 129  159  Isocyanate 131  160  Monomethylamine 132  161  Dimethylamine 133  162  Exo-olefin 134  162  18,19-Dinorabieta-8,ll,13-trien-4-one(135)  164  3-Dimethylthiomethylene-l 8,19-dinorabieta-8,11,13trien-4-one (136)  166  19-Hydroxy-18(4-^3)aieo-abieta-3,8,11,12-tetraen-18-oic acid lactone (65) 168 Alternative simplified syntheses of the exo-olefin (134)... 170 Method A  170  Method B  171  VUI  Ozonolysis of the mixture of olefins prepared by the alternative method A 3.2.2 Synthesis of isotriptophenolide (71) 12-Acetyl-19-hydroxy-18(4^3)a6eo-abieta-3,8,11,13tetraen-18-oic acid lactone (137)  171 172  172  12-Acetoxy-19-hydroxy-18(4^3)a6eo-abieta-3,8,11,13tetraen-18-oic acid lactone (138) 173 12,19-Dihydroxy-18(4-^3)a!6eo-3,8,ll,13-tetraen-18-oic acid lactone (71) 175 3.3  3.4  Preliminary experiments for oxidation of isotriptophenolide (ITP, 71)..176 I.  Oxidation with Fremy's salt  176  II.  Oxidation with ferric chloride  179  III.  Oxidation of ITP with DDQ  180  Synthesis of 8-methoxy-dienones Diastereomeric mixture of 8-methoxy-l 9-hydroxy-12-oxo18(4^3)a6eo-abieta-3,9(ll),13-trien-18-oic acid lactones (153)  182  182  8a-Methoxy-19-hydroxy-12-oxo-18(4-^3)a&eo-abieta-3,9(l 1),13trien-18-oic acid lactone (153a) 184 3.5  Synthesis of the (7,8)a-epoxy series 3.5.1  3.5.2  Syntheses of 7a-hydroxy-isotriptophenolide (146a)  187 187  7a,12,19-Trihydroxy-18(4^3)a6eo-abieta-3,8,ll,13tetraen-18-oic acid lactone (146a)  187  Method A (from 7a-methoxy-isotriptophenolide, (150a)  187  Method B (from isotriptophenolide, 71)  189  Synthesis of (7,8)a-epoxy-dienone, 72  192  (7,8)a-epoxy-19-hydroxy-12-oxo-18(4^^3)abeo-abieta3,9( 11),13-trien-18-oic acid lactone (72) 192  IX  Method A (from 7a-hydroxy-isotriptophenolide, 146a)... 192 Method B (from 7a-hydroxy-isotriptophenolide, 146a)... 193 Method C (from isotriptophenolide, 71) 3.5.3 3.6  Epoxidation of (7,8)a-epoxy-dienone, 72  Synthesis of the (7,8)P-epoxy series 3.6.1  Synthesis of 7p-hydroxy-isotriptophenolide (146b) 12-Acetoxy-19-hydroxy-7-oxo-18(4—>3)a6eo-abieta3,8,1 l,13-tetraen-18-oic acid lactone (162)  194 195 198 198  198  12,19-Dihydroxy-7-oxo-18(4->3)aZ)eo-abieta-3,8,l l,13tetraen-18-oic acid lactone (151) 199 7p,12,19-Trihydroxy-18(4^3)a6eo-abieta-3,8,ll,13tetraen-18-oic acid lactone (146b) 3.6.2  Synthesis of the (7,8)(3-epoxy-dienone, 73  200 201  (7,8)P-Epoxy-19-hydroxy-12-oxo-18(4^'3)a6^o-abieta3,9(1 l),13-trien-18-oic acid lactone (73) 201 3.6.3 3.7  3.8  Epoxidation of (7,8)P-epoxy-dienone 73  Synthesis of the 8-hydroxy series  203 205  3.7.1  Epoxidation of the 8a-hydroxy-dienone 157  205  3.7.2  Epoxidation of 8p-hydroxy-dienone 158  208  Attempted bio-epoxidation of (7,8)-epoxy-dienone precursors with TRP4a cell cultures 3.8.1  General  209 210  Growth conditions of TRP4a cell cultures  210  Harvesting and extraction procedures  210  X  3.8.2  3.8.3  General procedure for the biotransformation of the synthetic precursors  211  Measurement of peroxidase activity (pyrogallolpurpurogallin assay)  212  Biotransformation of the (7,8)a-epoxy-dienone (72) with TRP4a cell culture  213  Biotransformation of (7,8)P-epoxy-dienone (73)  214  REFERENCES  215  APPENDIX  223  XI  List of Figures  Figure 1.1  Diterpenoids isolated from Tripterygium wilfordii cell cultures  21  Figure 2.1  The three dimensional structure of quinone 145  59  Figure 2.2  The stereochemical structure of diastereoisomers 153a and 153b  72  Figure 2.3  Major NOE's observed for compound 154  76  Figure 2.4  Major correlations observed in the COSY spectrum of compound 154  76  Figure 2.5  Major NOE's observed for compound 72  83  Figure 2.6  Major proton correlations observed in the COSY spectrum of compound 72  Figure 2.7  Comparison between the two possible orientations of the CI 3-C 14 epoxide  Figure 2.8  83  90  Comparison of the ' H NMR spectra of diepoxides 160 and 161 (400 MHz, in CDCI3)  92  Figure 2.9  Major NOE's observed for compound 161  94  Figure 2.10  Comparison of the ' H NMR spectra of epoxy-dienones 72 and 73 (400 MHz, in CDCI3)  99  Figure 2.11  Expanded COSY spectrum of compound 164  102  Figure 2.12  Major NOE's observed for compound 164  103  Figure 2.13  Expanded COSY spectrum of compound 163  105  Figure 2.14  NOE difference spectra of compound 163  107  Figure 2.15  Conformation of the C13-C14 epoxy group of 163  108  XII  Figure 2.16  Expanded COSY spectrum of compound 165  122  Figure 2.17  Stereochemical view of compound 164  123  Figure 2.18  Expanded COSY spectrum of compound 166  126  Figure 2.19  Expanded HMQC spectrum of compound 166  128  Figure 2.20  The orientation of the C13-C14 epoxide ring in compound 166  131  Figure 2.21  Expanded COSY spectrum of compound 167  132  Figure 2.22  Major NOE's observed for compound 167  134  Figure 2.23  NOE difference spectra of compound 172  138  Figure 2.24  Orientation of the C13-C14 epoxide of 172  140  Figure 2.25  Required transition state for the epoxidation of the dienone substrates with dimethyl dioxirane  144  Figure 2.26  General extraction procedure of TRP4a cell cultures  147  Figure 2.27  Consumption of starting material during the biotransformation of 72 with TRP4a cell cultures  Figure 2.28  Changes in the amounts of products formed during the biotransformation of 72 with TRP4a cell cultures (cell ages: A, 14 days; B, 25 days)  Figure 2.29  149  150  Changes in the amounts of products formed during the biotransformation of 73 with TRP4a cell cultures (cell ages: A, 14 days; B, 25 days)  152  Xlll  List of Schemes  Scheme 1.1  Proposed biosynthetic pathway to tripdioUde (2)  Scheme 1.2  Synthesis of racemic triptoHde (1) via BC^-ABC abietane  23  construction  27  Scheme 1.3  Synthesis of triptoUde (1) from dehydroabietic acid (60)  29  Scheme 1.4  Synthesis of triptolide (1) via AB^'ABC abietane construction  32  Scheme 1.5  Biogenetic-type synthesis of racemic triptoUde (1)  33  Scheme 1.6  Construction of the epimeric triepoxide systems from levopimaric acid (115)  Scheme 1.7  35  Alkylation of thiols by the diterpene triepoxides via hydroxylassisted epoxide ring opening  37  Scheme 2.1  Synthesis of isotriptophenolide (71) from dehydroabietic acid (60)  45  Scheme 2.2  The mechanism of decarboxylation of dehydroabietic acid (60) with lead tetraacetate  53  Scheme 2.3  Possible quinoid-type derivatives of isotriptophenolide (71)  56  Scheme 2.4  Reaction of isotriptophenolide (71) with Freriiy's salt  57  Scheme 2.5  Proposed mechanism for the formation of compounds 145 and 146  62  Scheme 2.6  Proposed mechanism for the formation of compounds 150 and 151  66  Scheme 2.7  Proposed trapping of the quinone methide 142 with a peroxide anion... .67  Scheme 2.8  Oxidation of/7ara-substituent phenols with PIDA  69  Scheme 2.9  Reaction of isotriptophenolide (71) with PIDA in methanol  70  Scheme 2.10 Attempted epoxidation of the 8-methoxy-dienone 153  75  XIV  Scheme 2.11 Proposed synthetic plan for the (7,8)a-epoxy-dienone 72  79  Scheme 2.12 Demethylation of 150a with boron trichloride  80  Scheme 2.13 Preparation of the (7,8)a-epoxy-dienone 72 from 146a  81  Scheme 2.14 A possible mechanism to compound 151 from 146a via oxirane ring opening  84  Scheme 2.15 An alternative mechanism for the oxidation of 146a to 151  85  Scheme 2.16 Proposed mechanism for the formation of compound 72 and 151  86  Scheme 2.17 Proposed retrosynthesis of the (7,8)P-epoxy-dienone 73  95  Scheme 2.18 Synthesis of 7-oxo-isotriptophenolide (151) from 138  96  Scheme 2.19 Synthesis of 7-oxo-isotriptophenolide (151) from 146b  97  Scheme 2.20 The mechanism of formation of compound 163 and 164  109  Scheme 2.21 Steric interactions in the possible reaction intermediate 164-11  Ill  Scheme 2.22 Synthesis of the 8-hydroxy-dienones 157 and 158 from ITP (71)  113  Scheme 2.23 The mechanism of oxidation of isotriptophenolide (71) with PID A/water Scheme 2.24 The epoxidation of hydroxy-dienone 157  115 119  Scheme 2.25 A feasible mechanism for the formation of compounds 165, 166 and 167  135  Scheme 2.26 The epoxidation of hydroxy-dienone 158  136  Scheme 2.27 The original synthetic sequence to diol 146a and epoxide 72  141  Scheme 2.28 The "unified" synthesis of the epoxy-dienones 72 and 73; and the hydroxy-dienones 157 and 158  142  Scheme 2.29 Products of biotransformation of 72 and 73 with TRP4a cell cultures... 151  XV  List of Tables  Table 1.1  18(4^'3) ^6eo-abietane type diterpenoids isolated from Tripterygium species  14  Table 1.2  Abietane-type diterpenoids isolated from Tripterygium species  18  Table 2.1  ' H NMR data of hydroxy-dienones 157 and 158 (400 MHz, in CDCI3; 5 in ppm, J in Hz in parentheses)  118  XVI  List of Abbreviations  APT  attached proton test  [oiVo  specific rotation recorded at t °C using sodium D-line ligh  ABq  AB quartet  Ac  acetyl  B-5  standard tissue culture medium developed by Gamborg  br  Eveleigh broad  brine  saturated sodium chloride solution  Bu  butyl  c  concentration (g/100 ml)  °C  degree Celsius  cm"  wave number  COSY  IH-IH bidimensional Correlated NMR Spectroscopy  d  doublet  dd  doublet of doublets  ddd  doublet of doublet of doublets  dec  decomposition  DHA  dehydroabietic acid  Av  chemical shift difference  DDQ  2,3-dichloro-5,6-dicyano-l,4-benzoquinone  DMD  dimethyl dioxirane  dt  doublet of triplet  ED50  median effective dose  Et  ethyl  EIMS  electron impact mass spectrum  g GC  gram  GTW  a multi-glycoside extract from the plant Tripterygium wilfordii  71  gas chromatography  xvu  h  hour  [H]  reduction  hexanes  generally a mixture of several isomers of hexane (CeHn), predominantly n-hexane and methylcyclopentane (CeHn)  HMQC  Heteronuclear Multiple-Quantum Coherence  HPLC  high pressure (performance) liquid chromatography  HRMS  high resolution mass spectrum  hv  light radiation  Hz  hertz  IR  infrared  ITP  isotriptophenolide  J  coupling constant  KB  a tissue culture cell line derived from human carcinoma of the nasopharynx  X  wavelength  L  liter  L-1210  a tissue culture cell line derived from mouse leukemia  LAH  lithium aluminum hydride  LD50  median lethal dose  LDA  lithium diisopropyl amide  logs  the log of extinction coefficient  LTA  lead tetraacetate  LRMS (or MS)  low resolution mass spectrum  m  multiplet  M  molar  M"  molecular ion  m-CPBA  OTeto-chloroperbenzoic acid  Me  methyl  mg  milligram  MHz  megahertz  min  minute  xvm  mL  milliliter  mmol  millimol  mp  melting point  MSNA0.5K0.5  MS medium of Murashige and Skoog^'' supplement with naphthaleneacetic acid (NA, 0.5 mg/L) and kinetin (K, 0.5 mg/L)  1^  micro (10"^)  m/z  mass to charge ratio  V  frequency  NBS  N-bromosuccinimide  r\D  refractive index  nm  nanometre  NMR  nuclear magnetic resonance  NOE  nuclear Overhauser effect  [0]  oxidation  P-388  a tissue culture cell line derived from mouse leukemia  PCC  pyridinium chlorochromate  Ph  phenyl  PIDA  phenyl iodoso diacetate  ppm  parts per million  Pr  propyl  PRDaCoioo  PRL-4 medium of Gamborg and Eveleigh^' supplement with 2,4dichlorophenoxyacetic acid (D, 2 mg/L) and coconut milk (Co, lOOmL/L)  PRL-4  standard tissue culture medium developed by Gamborg and Eveleigh'''  r.t.  room temperature  s  singlet  sept  septet  sh  shoulder  t  triplet  td  triplet of doublets  XIX  Td  tripdiolide  TFA  trifluoroacetic acid  TFAA  trifluoroacetic anhydride  THF  tetrahydrofuran  Tl  triptolide  TLC  thin layer chromatography  TMS  tetramethylsilane  Tosyl (Ts)  j9ara-toluenesulfonyl  TRP4a  a cell line of plant cell culture developed from Tripterygium wilfordii  UV  ultraviolet  XX  Acknowledgements  I would like to thank my supervisor. Professor James P. Kutney, for his guidance throughout this work. I would also like to thank to those members of our group, past and present, that in someway helped and contributed to this research. I wish to express my deepest gratitude to Dr. Rajina Naidu for her invaluable suggestions, her advise, and helpful discussions throughout the development of my synthetic work and the preparation of this thesis. I am grateful to Gary Hewitt, Elena Polishchuk and David Chen of the Biological Services Facility for their assistance in the preparation of the cell cultures and for showing me how to work with them. The expertise and help from the staff of the NMR service, mass spectrometry, and Mr. Peter Borda of microanalytical services are highly appreciated. A fellowship from Syntex, S.A. and financial assistance from the University of British Columbia are gratefully acknowledged. Last, but never least, I would like to express my deep gratitude to my beloved parents, for their support, understanding and encouragement throughout the course of my education.  CHAPTER 1  INTRODUCTION  1.1  Historical background of the plant Tripterygium wilfordii  Kupchan and co-workers in 1972 isolated two novel diterpenoid triepoxides, triptolide (Tl, 1) and tripdiolide (Td, 2) from the root extract of Tripterygium wilfordii.' These compounds were the first reported natural products containing the 18 (4—>3) abeoabietane skeleton and the first recognized diterpenoid triepoxides from a natural source.  Triptolide (1)  Tripdiolide (2)  Tripterygium wilfordii Hook.f (Hooker filius) is a perennial twining vine, typically about 2-3 meter in height with reddish brown twigs and oval-shaped leaves. It belongs to the family Celastraceae. The genus Tripterygium is indigenous to China, where three species are found namely: Tripterygium wilfordii Hook, f., Tripterygium hypoglaucum Level and Tripterygium regelii Sprague et Takeda.^'^ These species are  distributed in different geographical areas of China.  T. wilfordii grows in the  mountainous areas of south eastern and southern China; T. hypoglaucum is found in south western and southern China and T. regelii occurs in the north eastern part of China and Japan.'*'^ The most common species of Tripterygium is T. wilfordii, and as a result this species has been most extensively utilized in traditional Chinese medicine and also analyzed in detail in terms of its constituents. This plant is commonly known in China as Lei Gong Teng (Thunder God vine) or Mang Cao (rank grass) and its use in Chinese folk medicine can be traced back almost two thousand years. "Rank grass" is first mentioned in the Saint Peasant's Scripture of Materia Medica,*^ written between the first and second centuries of our era, as being used for the treatment of fevers, chills, carbuncle and oedema. The powdered root of this plant was used by Chinese farmers as a pesticide to protect their crops from chewing insects, application of which is still being studied nowadays.^ Recently , extracts and preparations from T. wilfordii have been used increasingly for the treatment of rheumatoid arthritis, chronic hepatitis, ankylosing spondititis and a variety of skin disorders.*'''" Since the isolation of triptolide (1) and tripdiolide (2) from this plant in the early seventies, and since these compounds showed significant antileukemic and antitumor activities', much interest was generated within this family of compounds. Over the past 25 years, a great deal of research has been performed on this plant and as a result a "refined extract" (a so-called total multi-glycosides extract, or GTW) has been found to have anti-inflammatory and immunosuppressive activity.' Later, a reversible antifertility activity was observed as well when GTW was administered orally in male rats and in men, without showing significant side effects."''^  Such a diversity of pharmacological activities exhibited by constituents of this plant presents a promising future for some of the compounds responsible for those activities, among them being triptolide (1) and tripdiolide (2).  1.2  Pharmacological properties and clinical uses of refined extracts of T. wilfordii and triptolide (1)  Although the plant Tripterygium wilfordii has been used since ancient times in China as a herbal medicine {vide supra), and it has even been used in agriculture as an insecticide (later it was found that five alkaloids present in the roots were responsible for such an activity),'^ it was not used in medical practice until some 25 years ago, when Chinese researchers noted the efficacy of extracts of the plant in relieving the symptoms of rheumatoid arthritis.''' Since then physicians have been using it increasingly to treat several other immune-related diseases with very promising results {vide infra). In the early clinical trials in China, many crude preparations of Tripterygium wilfordii were used, but their side-effects and quality couldn't be easily controlled. Due to this, a refined extract, with definite quality control criteria and supplied in the form of tablets, referred to as the "multi-glycosides" extract of Tripterygium wilfordii (or GTW),'^ was developed and it is now used widely in clinics throughout China. The root xylem is extracted with water and then with chloroform, the resulting solution is concentrated and finally column chromatographed to obtain the so-called GTW extract.  Twenty-five  grams of xylem yields 1 mg of GTW which is then made commercially available as tablets, each containing 10 mg of the extract. GTW seems not to be an appropriate name  for the preparation as no chemical evidence of the presence of glycosides has ever been found. This extract contains the large majority of the pharmacologically active compounds of the plant, including triptolide (1) and tripdiolide (2) as two of its main components.  It also has only minute amounts of alkaloids, which are its main toxic  constituents, hence minimizing the side effects as compared to the crude decoctions. Interest in this plant was renewed and prompted a greater deal of research when in addition to their strong antileukemic activity, yet another remarkable property of these compounds, a fully reversible male antifertility action, was later found." A more detailed discussion of each one of those activities is presented below.  1.2.1  Immunosuppressive and anti-inflammatory properties  The most extensive clinical use of GTW preparations in traditional Chinese medicine for the past 20 years has been for the treatment of auto-immune related diseases, such as rheumatoid arthritis and a series of skin disorders ranging from systemic lupus erythematosus, allergic angitis, anaphylactoid purpura nephritis and Behcet's disease.'* Clinical trials of the GTW extract started in the early 1970's, and numerous reports of its effectiveness against the aforementioned diseases have since been published.'''•'^•'^ In a study with 144 patients,^ GTW extract proved to be effective in the treatment of acute and active rheumatoid arthritis. After only 3 to 5 days of medication, arthroses pain, oedema and joint morning stiffness showed remarkable improvements. In a similar trial on patients with systemic lupus erythematosus in Shanghai, the preparation  was effective in 88% of cases." In the treatment of several types of nephritis inconsistent results were observed.^" Good therapeutic results were obtained in the treatment of other skin disorders such as Behcet's disease and some allergic diseases.'^  The anti-  inflammatory and analgesic efficacy of the drug is superior to any of the currently available non-steroidal anti-inflammatory drugs, such as salicylates, phenylbutazone, etc.. Its potency is similar to that of corticosteroids and it can be used as a substitute in patients with steroids-dependency or steroid-allergy problems.'"•^' Laboratory studies in rats showed that GTW has a superior effect in prolonging the mean survival time of cardiac allografts, as compared to cyclosporine A, a wellknown anti-rejection agent.^^  In another in vivo experiment carried out in order to  determine the immunosuppressive efficacy of an extract of T. wilfordii against rheumatoid arthritis, the preparation showed a potent immunosuppressive effect on type II collagen induced arthritis in mice, a widely use experimental animal model of human rheumatoid arthritis.^'' Much research has been done to elucidate the biological mechanism for the immunosuppressive activity of T. wilfordii preparations. Some early reports showed the ability of those extracts to suppress antibody synthesis, decrease E-rosette formation of lymphocytes and reduce lymphocyte response in vitro to both mitogens and alloantigens.^'*'^^ Subsequent in vitro studies revealed that GTW decreases prostaglandin E2 (PGEj) synthesis by monocytes^'' and inhibits antigen and mitogen-stimulated proliferation by T (thymus derived) and B (bone marrow derived) cells. It also inhibits interleukin-2 (IL-2) production by T cells and suppresses immunoglobulin production by B cells."  Most of these studies have been conducted on extracts of T. wilfordii, such as GTW, rather than with triptolide (1) or tripdiolide (2) alone. Therefore it could be possible that some of the other constituents present, such as triterpenes or alkaloids could be responsible for the observed pharmacological activity. Nevertheless, two recent reports^^'^' on the isolation and identification of the immunosuppressive components of different extracts of T. wilfordii have established that triptolide (1) and tripdiolide (2) are indeed the major components and are the main agents responsible for the anti-rheumatic properties of the T. wilfordii extracts. Furthermore, triptolide (1) was also reported to inhibit Coenzyme A-induced proliferation of T-Iymphocytes, mixed lymphocyte reaction and cytotoxic T-lymphocytes activity and to prolong skin allograf survival.^"'^'  1.2.2  Anticancer properties  The initial findings of antileukemic activity from an alcoholic extract of the roots of T. wilfordii prompted investigation on this plant, leading to the isolation and identification of triptolide (1) and tripdiolide (2) as the compounds responsible for such an activity.'  Biological evaluation of these novel diterpene triepoxides revealed  significant activity in vivo against the L-1210 and P-388 leukemias in mice.  They  showed a considerable life-prolonging effect, observed at the 0.1 mg/kg level for both compounds and in vitro cytotoxicity (ED50) against cells derived from human carcinoma of the nasopharynx (KB) at concentration as low as 10"^ to 10"'' \iglm\.?^  The efficacy of triptolide (1) against L-615 leukemia in mice was also demonstrated when the mean survival time was prolonged, in some cases for more than a month.^'' In those cases, the rechallenge of the surviving mice, with leukemic cells did not cause recurrence of the disease.  More recently, triptolide (1) has also showed  remarkable growth inhibition activity on colony formation of several human breast and stomach cancer cell lines (BT-20, MCF-7, MKN-45, MKN-7 and Kato-III)." Several studies have been carried out to establish the mechanism of action of Tl (1) and Td (2)."^^'^' Although no conclusive evidence has been found, it is postulated that its tumor-inhibitory activity is due to selective alkylation, which proceeds via nucleophilic attack of the thiol group of growth regulation enzyme to an epoxide group of Tl or Td '^ {vide infra. Sect. 1.4).  1.2.3  Male fertility control  The incidental discovery of a reversible male antifertility effect of GTW in male rats and in men in 1986, during a study of the immunosuppressive activity of this drug, arose worldwide interest.^*'" It was noticed that rats fed with a dose of 10 mg/kg/day of GTW, 6 times a week, became infertile after 8 weeks of dosing. Observation of the sperm showed a sharp decrease of motility to almost zero and a moderate reduction in the sperm density. No apparent toxicity was noticed at those dose levels, the histology of important somatic organs were unaltered, the levels of testosterone in serum and the sex behavior were unaffected.  Fertility was fully restored within 4-5 weeks after the  treatment was discontinued. Female rats did not suffer any effect on fertility.  Men  treated for rheumatoid arthritis or psoriasis with GTW, presented similar effects in fertility at dose levels one third of that for treatment of arthritis or skin disorder. These dose levels further decrease the risk of side effects.^^ In a later more detailed clinical trial,^' 26 men suffering from psoriasis were treated with oral doses of 20 mg GTW/day for 4-6 months and the changes in the reproductive function were systematically observed for a period of one year. It was noted that after one month, the sperm motility and the concentration of spermatozoa decreased to about 25% of that before treatment. After two months, the motility dropped to 12% and the density was lowered further.  No significant side effects were observed, no  discernible change occurred in the morphology of the testis and potency and libido were unaffected throughout the treatment. One month after cessation of treatment, both the sperm density and motility increased to fertile levels and after one more month were fully restored to normal. The data gathered up to now indicate that GTW mode of action is achieved by causing damage to the epidydimal spermatozoa, and to a lesser extent disturbs the spermatogenic dynamics."" The inhibition of spermatogenesis seem to be caused by partial blockage to the synthesis of basic nuclear proteins in late spermatids, an effect which takes place at a faster rate than that caused by gossypol, another known oral antifertility agent in men and male animals.  This postulate serves as a possible  explanation to the observed reduction of the sperm density and to the fact that the remaining spermatozoa present deformations, with head swelling, head-tail separation and marked curling of the broken mid-piece.  It is worthy to mention that the related diterpene epoxide, tripchloride (7), an artifact formed reversibly from triptolide (1) during the isolation process,'" was recently found to have a nearly 200 times higher antifertility activity than GTW. An additional advantage of this compound is that, unlike GTW or triptolide (1), which cause relatively broad damage to the sperm, especially to their head (a potential source of mutation), tripchloride (7) only damages the spermatozoa in the epididymis without causing significant anomalies to the sperm head.''^ Further investigation on this compound is presently been carried out.'"  1.2.4  Side-effects of GTW and triptolide (1)  At regular clinical dose levels of GTW for the treatment of rheumatoid arthritis or dermatological disorders, i.e. 60-90 mg/day (1-1.5 mg/kg/day), the main observed side effects are gastrointestinal disturbances, including nausea, vomiting, anorexia, epigastric burning, xerotomia and diarrhea or constipation.''^ In most cases, discontinuation of the medication is not necessary as those effects are not severe and generally will subside during the course of the treatment. Leucopenia or thrombocytopenia were developed in a few patients, but rapid recovery occurred after cessation of the medication. Menstrual disturbances in women, oligospermia or azospermia and decrease in the size of the testis were other side effects in men observed infrequently.^ It is important to note that the dosage of GTW necessary to induce reversible infertility in men is much lower than that for the treatment of rheumatoid arthritis and dermatitis, this being only 20 mg/day, administered orally. At these dose levels the above  10  mentioned disturbances were not observed.'^ Ahhough the use of GTW (or triptoUde) as a male antifertihty agent might seem to be limited by the fact that the antifertihty activity appear to be inseparable from the immunosuppressant activity, animal tests showed that the antifertihty dosage levels were 5-12 times lower than those for immunosuppressive activity. This indicates that an effective antifertihty activity can be achieved without immunosuppressive side effects, such as increased susceptibility to secondary infections.'''' The antileukemic activity of triptolide (1) has also been tested in vivo}'^^ These experiments showed that the survival time of mice injected with leukemic cells was significantly extended when treated with doses of triptolide of 0.1 mg/kg/day. The LD50 of triptolide (1) administrated by intravenous injection was found to be 0.8 mg/kg for a single dose or 0.16 mg/kg/day for seven days, causing fatal degeneration of the heart tissue and bone marrow.'^ Those results made evident the dangerous proximity between the therapeutic and the lethal dosage, for that reason this compound has not been used clinically for the treatment of leukemia.  1.3  Approaches utilized to produce triptolide and its derivatives  The broad diversity of pharmacological properties exhibited by T. wilfordii have motivated much research in the last half of this century. Three different approaches have been adopted in order to make available new compounds and to produce them in sufficient quantities for biological evaluations and future clinical trials, namely: (1) Direct  11  isolation from the plant; (2) Use of plant cell cultures of T. wilfordii, and (3) Chemical synthesis. Each one of these methods has its own advantages and draw backs. The isolation of biologically active agents from the intact plant presents various problems for large scale production of individual constituents to be used as pharmaceuticals. Generally, isolations from such sources yield only minute amounts of the desired compounds, and because of the complexity of the crude extracts, separation from co-occurring materials is often difficult, costly and time consuming. Furthermore, the concentration of the desired compounds may vary according to the time of year the plant is harvested. Additionally, some other factors, such as slow growth of the plant or geographical, political and environmental constraints may make the required plant unavailable. The search for a better alternative which may allow to overcome most of those problems lead to the development of plant cell cultures from Tripterygium wilfordii in the early 1980's by Dr. Kutney's research team. The tissue culture technology presents the advantages of having controlled growth conditions which allow reproducibility of results and year-round availability of the natural products.  Growth parameters can also be  manipulated in order to develop cell lines capable of producing higher yields of the desired compounds. Additionally, cell cultures provide an excellent media for the study of the biosynthetic processes of the plant or even isolation of useful enzymes.  A  significant limitation to this technology, though, is the dependency on the metabolic capabilities of the cell culture to produce one specific compound or a series of compounds of particular interest in quantities large enough for one's requirements. Likewise, it can be mentioned that, despite the fast development of this area in the last  12  two or three decades, this methodology still offers to the present day, limited predictability on the ability of the cultures to metabolize or biotransform a "foreign" precursor and to predict the possible outcomes of such biotransformations. Another more versatile approach has been the development of various synthetic strategies in order to produce not only the natural products of interest, but interesting structurally-related "un-natural" compounds as well. The most significant advantage of this method is perhaps, that it allows for a better control of the direction to follow towards the preparation of a desired compound. However, a significant disadvantage of a total synthetic approach is that the multi-step sequences involved may in some instances be lengthy and/or have overall low yields. Lastly, the structural complexity of the targeted molecules, particularly important in some natural products, may represent another challenge to overcome. Thus, each method has in turn, contributed to some extent to supply novel analogs, in the search for compounds with improved pharmacological activity and lower side-effects. The choice of approach has ultimately been left to the specific objectives of the different research groups involved.  1.3.1  Isolation of diterpenoids from Triptetygium species  The research in this area has focussed on the isolation and identification of the different components of the herbal extracts and their subsequent biological evaluation, in the search for the compounds responsible for such activities.  13  As a result, up to the present over 170 compounds have been isolated from Tripterygium plants, including mostly alkaloids, sesquiterpenes, diterpenes and triterpenes. Several of the most active components have been found to be diterpenes. Since the interest of our investigation focussed only on derivatives of the diterpene triepoxide triptolide (1), only the compounds belonging to the diterpene family will be discussed in detail here.  Aside from the diterpene triepoxides triptolide (1) and  tripdiolide (2), found in the plant in concentrations as low as 0.001%', another 46 diterpenoids have been isolated from Tripterygium species to the present day. All of them exhibit either the abietane or the 18(4—>3) abeo-ahietane skeleton, with various degrees of oxygenation throughout their structures. It has been observed that the compounds presenting the rearranged 18(4—>-3) a^eo-abietane skeleton show the highest degree of oxygenation present in at least two rings and most commonly in rings A-C of the structure. The other type of diterpenes identified in Tripterygium species is the abietane-type, which do not possess a butyrolactone in ring A and exhibit oxygenated functionality only in ring A and/or C, but not in ring B (with only two very recently reported exceptions, compounds 40 and 41''^'^^). It is worthy to note that none of the latter show any epoxide functionality in their structures (see Tables I.l, 1.2 and the corresponding structures). Another interesting feature, common to all, is the presence of a carbonyl group, either as a ketone, aldehyde, carboxylate or lactone (with the only four exceptions of 30, 31, 42 and 43), feature that seems to be important for their biological activity (vide infra, sect. 1.4).  14  Table 1.1 Code 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29  18(4->3) Abeo-ahiotane type diterpenoids isolated from Tripterygium species Compound Triptolide Tripdiolide Tripterolide Triptonide 16-hydroxytriptolide Triptolidenol Tripchloride Triptriolide 12-epitriptriolide Triptetraolide Isotriptetraolide Tripdioltonide 13,14-epoxi-9,ll,12trihydroxytriptolide Triptonolide Triptophenolide (hypolide) Triptophenolide methyl ether Triptobenzene I Triptobenzene E Triptobenzene F Triptobenzene G Neotriptophenolide Isoneotriptophenolide Triptobenzene D Triptinin-B Triptoditerpenic acid Hypoglic acid Triptonoditerpenic acid Triptoquinonoic acid A Triptoquinonoic acid C  m.p. (OC) 226-227 210-211 225-228 210-211 232-234 193-194 256-258 260-262 268.5±1.0 258-260 250-252 222-224 268-270  Species Tw Tw Th Tw Tw Tw Tw Tw Tw Tw Tw Tw Tw  Reference 1 1 3 1 46 47 48 49 50 51 52 53 53  214-215 232-234 152-154  Tw Tw Tw Tw Tw V. reg Tw V. reg Tw V. reg Tw Tw Tw V. reg Tw Th Th Th Tr Tr  54 55 55 56 45 45 45 55 57 45 58 59 60 61 62 63  -  189-191 185-187 174-177 -  219-224 189-191 182-183 202-203  Tw = T. wilfordii; Th = T. hypoglaucum; Tr = T. regelii; Tw v. reg = T. wilfordii var. regelii  15  19  18(4—>3) abeo-ah'iQtane skeleton  1 R=H 2 R = P -OH 3 R = a -OH  16  8 R = P -OH 9 R = a -OH  10 R ' = OH; R^ = H 11 R ' = H ; R ^ = P - O H  13  12  14  17  15  R'  = H; R^ = H  16  R ' = C H 3 ; R^  17  R'  = H;  R^  18 R ' = OH; R^ = H 19* R ' = H ; R^ = OH 20* R ' = H ; R^ = OH  =H  = OH  HOOC  21 R ' = OH; R^ = H 22 R^ = H; R^ = OH  23 R - H 24 R = O H 25 R=0CH3  OCH,  HOOC  HOOC  26  HOOC  27  Epimers at C I 5 . Their absolute configuration was not determined.  28 R = H 29 R = OH  Table 1.2 Code 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48  Abietane-type diterpenoids isolated from Tripterygium species Compound 11-hydroxy-14-methoxydehydroabietane Triptobenzene B Triptonoterpene Triptonoterpene methyl ether Neotriptonoterpene Methylneotriptonoterpene 12-methoxytriptonoterpene Triptonoterpenol wilforol F Triptobenzene A Triptobenzene C 8p, 19-dihydroxy-3-oxopiran15-ene Triptoquinonol Triptoquinondiol 3 -oxo-triptoquinonol Triptoquinonal Tritoquinonoic acid B Triptoquinone G Tripterifordin*  m.p. (OC) 194-195  153-155 209-211 205-207 175-176 196-197 197-199  165-166 183-184 135-136 127-128 212-213 255-256  Species Tr  Reference 63  Tw V. reg Tw Tw Tw Tw Tw Tw Tw Tw V. reg Tw V. reg Tw V. reg  45 47 47 64 65 65 66 67 45 45 68  Tr Tr Tr Tr Tr Tw V. reg Tw  62 62 62 62 62 69 70  Tw = T. wilfordii; Th = T. hypoglaucum; Tr = T. regelii; Tw v. reg = T. wilfordii var. regelii * = Kaurane skeleton  abietane skeleton  30  OCHj  31 = P-OH; a-H 32 = 0  33 R ' = OH; R^ = H 34 R^ = H; R^ = OH  19  OCH-,  35 R = 0CH3 36 R = OH  OCHi  37  39 R = 2 H 40 R = 0  45 R ' = CHO; R^ = H 46 R ' = COOH; R^ = H 47 R ' = COOH; R^ = OH  38  42 R = 2H 43 R = P-OH; a-H 44 R = 0  41  48  20  1.3.2  Use of plant cell cultures of T. wilfordii  With the primary purpose of developing a plant cell culture capable of producing larger quantities of triptolide (1) and tripdiolide (2) than the whole plant, an intensive program aimed towards this objective was undertaken by Kutney et al in 1980. As a result, a cell culture of T. wilfordii grown in modified B-5 and PRL-4 suspension media (standard tissue culture medium developed by Gamborg and Eveleigh^') was established. This cell culture produced a yield of 0.003% tripdiolide (2), based on the dry cell weight, and TLC evidence of the presence of triptolide was also reported.'^ Further optimization of the media and growth conditions, as well as a careful selection of calli based on growth vigor and on TLC and KB cytotoxicity activity analyses, led to the development of a stable cell line designated as TRP4a." Stock suspension cultures of TRP4a are presently maintained  in  FRDjCojoo  broth  (PRL-4  medium  supplemented  with  2,4-  dichlorophenoxyacetic acid (D) (2mg/ml) and coconut milk (100 ml/L) as a growth medium, and 10% inoculum into MSNAojKgj broth (MS medium of Murashige and Skoog^" supplemented with napthaleneacetic acid (NA) (0.5 mg/ml) and kinetin (K) (0.5 mg/L) and incubating for 4-5 weeks. The levels of tripdiolide (2) produced this way are 4 mg/L or 36 times greater than that isolated from the whole plant. A thorough investigation on the isolation and characterization of the metabolites produced by TRP4a cell culture has been carried out by several members of Dr. Kutney's research group and a number of new compounds have been isolated."""  These  compounds include triterpenoids, phytosterols and diterpenoids. Figure 1.1 shows only  21 the new diterpenoids isolated from T. wilfordii cell cultures and which have not been previously isolated from Tripterygium plants.  Ha.N^^^' 50  49  54 R = H 55 R = 0CH3  51 R ' = R^ = H 52 R ' = H ; R^ = OCH3 53 R^ = OCH3; R^ = H  56  59  57 R = H 58 R = OCH3  60  Figure 1.1: Diterpenoids isolated from Tripterygium wilfordii cell cultures  22  ROOC  61 R ' = R^ = H 62 R ' = H ; R^ = OH 63 R ' = CH3; R^ = OH  65  64  66  69  67 R = H 68 R = CH,  70  Figure 1.1 (cont.): Diterpenoids isolated from Tripterygium wilfordii cell cultures  In addition to these compounds, triptolide (1), tripdiolide (2), and compounds 4, 14,15,16, 21, 32, 33 and 34 were also isolated. This series of metabolites provided some  23  clues as to the order of events leading to tripdiolide (2). Based on those findings, a biosynthetic pathway to tripdiolide (2) from DHA (60) was proposed'^ (Scheme 1.1).  OPP  Geranylgeranyl pyrophosphate  Dehydroabietic acid (60)  65  61  15  67 1) Oxidation 2) Epoxidation  Tripdiolide (2)  Scheme 1.1:  70  Proposed biosynthetic pathway to tripdiolide (2)  24  Having completed the isolation and characterization of metabolites in the cultures of T. wilfordii, the study moved into the next stage. The new objective was to make use of the TRP4a cell culture as a tool to biotransform synthetic substrates possessing the abietane skeleton at low oxidation levels, with the purpose of obtaining related novel diterpene epoxide analogs to perform biological activity evaluations.  Thus, three  precursors for biotransformation studies, butenolide (65), isotriptophenolide (ITP, 71) and triptophenolide (TP, 15) were prepared and incubated with cell cultures of TRP4a under a large variety of conditions."'^"^'^^  OH  65  71  15  It was found that the T. wilfordii cell cultures metabolized only partly these substrates (particularly low conversions, c.a. 35% and 41%, were observed for the ring C "activated" precursors 71 and 15 respectively), in spite of many changes in the incubation conditions. The products isolated from those experiments were largely constituted by rings A or B hydroxylated compounds (mostly at the benzylic carbon, C7), some products of further oxidation of the alcohol group to the corresponding ketone and a small amount of methylation of the phenolic alcohol groups in substrates 15 and 71. Nevertheless, a  25  few important exceptions were observed. When isotriptophenolide (71) was incubated with a 21 day-old TRP4a cell culture for a period of 4 days, the (7,8)-epoxy derivatives 72 and 73 were isolated. A larger scale experiment (1.5g) carried out under similar conditions provided the epoxides 72 and 73 in 13% yield (or 27% based on recovered starting material) as a mixture c.a. 1:1 of the a and p isomers. In this experiment, another interesting product, 74, which was not detected in the smaller scale experiments was isolated as well, although in a small quantity (0.2%).  72 a-epoxide 73 p-epoxide  74  Although the compounds obtained constituted a modest success, the overall limited ability of the TRP4a cell cultures to biotransform the synthetic precursors in the "right" or desired direction in order to produce a richer variety of epoxidized analogues, represent an important limitation to the application of this technology. The achievement of the above mentioned objectives, is a challenge yet to be overcome in future research in this area.  26  1.3.3  Total syntheses of triptolide and related compounds  Another major alternative approach to prepare sufficient quantities of triptolide (1) and analogues for pharmacological evaluation has been via organic synthesis. The unique abietane-type triepoxide system present in this compound represents an interesting challenge that has captured the attention of researchers since its isolation was first reported in 1972. The total syntheses reported to date can be classified into two main categories, the synthesis of racemic Tl (1) from simple, achiral raw materials and the synthesis of the optically pure /-triptolide, with the same stereochemistry as the natural compound. The two major challenges for the construction of the framework of the molecule have been the triepoxide system in ring C and the butenolide moiety in ring A. This has led to a number of approaches, differing mainly in the sequence of construction of the various structural units of the target molecule. The strategy  adopted  by Berchtold  et al.^^'^" was to synthesize  the  dihydronaphthalenone 75 as a starting material with the B/C ring fragment of the abietane skeleton (Scheme 1.2). Construction of ring A was done via alkylation and annulation of the naphthalenone, which provided a suitable functionalized tricyclic intermediate 78 for the construction of the ring C triepoxide system and the butenolide in ring A. Annulation was achieved by alkylation of 75 with the lodobutyrolactone 76, followed by opening of the lactone to give 77. Aldol condensation to give 78, reduction of the aldehyde, acid hydrolysis and rearrangement of the double bond completed the synthesis of the  27  1^ ^ 79 R ' = C H 3 ; R% R^ = O *»- 14 R ' = H ; R ^ R ^ = 0  '  (^ 67 R ' = R^ = H; R^ = OH  1 R ' = OH; R^ = H 81 R ' = H ; R^ = OH Scheme 1.2:  Synthesis of racemic triptohde (1) via BC -*- ABC abietane construction  a) NaOH, DMF; b) M^NH; c) CrOj, pyridine, CI^Cl2; d) neutral A^Oj, EtOAc; e) NaBH,, EtOH, 2N HCl; f) m-CPBA, CH2CI2; g) EtjN, CH2CI2; h) 2,4,6-timethylpyridine, MeSQCI, DMF; i) n,, Pd-C, EtOAc; j) CrOj, AcOH; k) BBrj, CH2CI2, 0°C; 1) NaBH,, EtOH; m) NaI04, MeOH; n) m-CPBA, CI^Clj; o) NaBH4, EtOH  28  butenolide ring. Construction of the ring C system was accomplished by hydroxylation at C7 and subsequent conversion to the epoxy dienone 80, by means of periodate oxidation, which upon further epoxidation gave racemic triptonide (4). This compound (4), on reduction with sodium borohydride yielded a 3:1 mixture of racemic 14-epitriptolide (81) and triptolide (1), respectively. A number of alternative routes to the total synthesis of triptolide (1) and triptonide (4) have been developed by van Tamelen and coworkers. A total chiral synthesis of 1triptonide (4) ( which could provide 1-triptolide 1 by reduction at the C14 carbonyl) from dehydroabietic acid (60) carried out by van Tamelen et al. consisted of appropriate functionalization of the aromatic ring at C14 first, followed by construction of the butenolide moiety in ring A and finally further elaboration on ring C to build up the epoxide system (Scheme 1.3)*'. The aromatic ring was first functionalized by nitration of dehydroabietic acid (60) to give the 12,14-dinitro derivative 82, which was then subjected to a series of functional group manipulations in order to introduce the required oxygenated group at CI4. Ring A butenolide construction was then undertaken. Curtius degradation of 86 gave the isocyanate 87, which was then reduced to the corresponding secondary amine and methylated with formaldehyde to provide the tertiary amine 88. Cope elimination to give the exo-olefin 89 was followed by oxidative cleavage to yield ketone 90. After a sequence of several steps, 90 was transformed to the aldehyde 94, which was then cyclized to form the butenolide 95. Finally, introduction of a P-hydroxyl group at C7 gave Berchtold's intermediate, o-hydroxymethylphenol 67 (Scheme 1.2), which was then further elaborated to obtain /-triptonide (4) in an overall yield of 0.27% from dehydroabietic acid (60).  29  60  jj / - 82 R ' = N 0 2 , R^ = N 0 2 ^, ^ ,  83 R ' = N H 2 , R^ = N 0 2  !> 84  R'=I,R^=N02  ^ 85 R ' = H , R^ =NH2  OCOCF3  f,g  86  OCOCF3  h,i  87  j,k  89  88 Scheme 1.3:  Synthesis of triptoHde (1) from dehydroabietic acid (60)  a) HNO3-H2SO4, AcOH; b) Hj, Pd/C, AcOH, CF3CO2H; c) NaN02, CF3CO2H; KI; d) Zn, AcOH, 65-7(?C; e) NaNOi, CF3CO2H; f) SOC^, PhH, DMF; g) NaN3, acetone, HjO; PhH, lOQPC; h) LAH, THF, refluxing; i) HCOjH-aq HCHO; j) m-CPBA, CHCI^, -20°C; k) CHCI3, refluxing; 1) OsQ,NaI04, AcOH-dioxane-HjO  30  t,u  m 2vo^^ r\'>  r  91  R'=H;R^=CH20H;  R^ R'*= 0  R ^ = OH; R ' ' = CHjOCHjPh q-s (^ 93 R '•= Ac; ' R^= - CHjOH; ^ ' R^= - OH;' R'*= CHjOCHzPh >  92 R ' = H ; R ^ = C H 2 0 H ;  V.W  aa  OHC  -c  R^=H  95  R'=AC;R^-  96  R'=AC;R^R^=0  67  R'=H;R^=0H;R^=H  80 Scheme 1.3 (cont.): Synthesis of triptolide (1) from dehydroabietic acid (60) m) LDA, HCHO, THF, -78PC; n) MeOC(CH3)=CH2, AcOH; o) PhCH20CH2Li, THF, -78°C; p) HCI, THF; q) MeOC(CH3)=CH2, AcOH; r) AC2O, pyridine; s) HCI, MeOH; t) PCC, Cp|ci2; u) o-C6H4(NH2)2, PhC02H, EtOH, HCI; v) NaCl62, HOSO2NH2, dioxane-H20; w) Hj, Pd/C, EtOH; x) CrOj, ACOH-H2O, 40°C; y) KOH, MeOH-H20; z) NaBH4, EtOH; aa) NaI04, EtOH  31  A shorter route to racemic triptolide (1), also devised by van Tamelen (Scheme 1.4), involved construction of ring C, possessing a hydroxyl group at C14 onto an appropriate A/B fragment derived from decalone 97.^^ In this synthesis, an efficient Diels Alder addition to the furan derivative 99 was used to form ring C with the desired functionality (100). The butenolide moiety in ring A was constructed from the alkene intermediate 101 via introduction of a hydroxyl group at C3 to give 102, followed by rearrangement with thionyl chloride and conversion of the CI9 allylic chloride to the allylic alcohol 103. Addition of dimethylformamide to this alcohol, followed by a carbene [2,3]-sigmatropic rearrangement gave 104.  Further elaboration yielded the  important intermediate 16, from which triptolide (1) can be derived via a known route, (see Scheme 1.2). Another original approach^^ presented a biogenetic-type synthesis of the racemic key intermediate 16. This is one of the most efficient routes yet designed, which results in a 15% yield after 12 steps with only four purifications required (Scheme 1.5). The key step is the cyclization of the geranylgeraniol-type intermediate 111 to the tricyclic skeleton 112. This intermediate contains both A and B rings with the correct trans junction and suitable functionality in ring A to subsequently construct the butenolide moiety and yield intermediate 16.  32  97  98  CO,Me 0-Si—/-Bu  e,f  g-J  99  OMe  k,l  OMe Hd^'  101  102  OMe  Scheme 1.4: Synthesis of triptolide (1) via AB—^ ABC abietane construction a) CS2, 2,6-di-;-Bu-4-Me-C6H20Li, THF, Mel; b) (^3)38*!", NaH, DMSO, -10°C; c) HCl (aq)-MeOH; d) LDA, HMPA, THF, TBDMSCl; e) CH2=CC02Me, PhH, 65-70°C; f) 5:1 MeOH-6M HCl; g) Mel, NaH, THF; h) MeLi, THF, -IS^C; i) MeSOjCl, EtjN, CHjClj; j) Li, NH3, THF, -78''C; k) m-CPBA, CH2CI2; 1) LDA, THF; m) SOCI2, EtjO, pyridine, CPC; n) KOAc, DMSO, 75°C; o) NaOMe, NaOH; p) (MeO)2CHNMe2, xylene. A, molecular sieve; q) m-CPBA, CliCl2; r) [(CH3)3Si]2NLi, THF, 0°C; s) IM HCl  33  O  OMe  105 107 R ' = COzMe, R ^ R^ = O 108 R ' = H , R ^ R ^ = 0 109 R ' = R ^ = H , R ^ = O H  MeOjC  110  111  OMe MeOzC  MeOiCf^  113  112  OMe  OMe MeO,C*  114  Scheme 1.5:  16  Biogenetic-type synthesis of racemic triptolide (1)  a) NaH, THF, 0°C; b) BaCOH)^, H j O - E t p , 90°C; c) LAH, E t p , 0°C; d) LiBr, PBrj, collidine, E t p , 40°C; e) ZnBr^, Et^O, 0°C; f) LiH, CHjCOCHjCOjMe, DMF, 75°C; g) SnCl4, CH^CIj, 0°C; h) MeSO^Cl, EtjN, CH^Cl^, 0°C; i) m-CPBA, CHjCl^; j) LDA, -78°C  34  Tokoroyama et al.*'' devised an alternative to the construction of the epoxide system in rings B/C, starting from levopimaric acid (115) (Scheme 1.6).  This route  involved initial formation of an endoperoxide which was then rearranged to the diepoxide 116.  A series of manipulations of the double bonds, and several epoxidations and  reductions led to the desired epoxide system 125 along with two other epimeric triepoxides, 123 and 124.  35  tOjMe 115  116 R = H  118  117 R = Me  g,h  119  120  121  122  COjMe 123  124  125  Scheme 1.6: Construction of the epimeric triepoxide systems from levopimaric acid (115) a) Rose Bengal, O2, hv; b) FeSO^; c) CHjNj; d) HCl (cat), EtjO; e) CrOj, H2SO4, acetone; f) NBS, CCI4; g) Zn, THF; h) /n-CPBA, Na2C03, CH2CI2; i) NaBH^, MeOH; j) m-CPBA, MeCN; k) (MeC0)2, PhH, hv, O2  36  1.4  Attempts to establish a structure-activity relationship for triptolide-related epoxides  Since the antileukemic properties of triptolide (1) and tripdiolide (2) were first reported by Kupchan,' some investigation has been made in order to elucidate what functionality in the molecule is responsible for their pharmacological activities. Those compounds exhibited significant activity both in vitro and in vivo against several types of leukemia (vide supra) and showed impressive life prolonging effects in mice at levels as low as 0.1 mg/kg. In contrast, triptonide (4) a variant of triptolide (1) in which the C14 P-hydroxyl group has been replaced by a carbonyl group, showed no antileukemic activity at doses up to 0.4 mg/Kg. Another similar compound, 14-epitriptolide (81), which differs from Tl (1) only in the orientation of the C14 hydroxyl group, also showed no significant activity against leukemia.  1 R=H 2 R = p-OH  81  37  These observations led to the postulate that the 9,1 l-epoxy-14p-hydroxy system is necessary for the antileukemic activity of triptolide (1) and tripdiolide (2), and that an intramolecular hydrogen bond formed between these two functional groups may facilitate the opening of the 9,11-epoxide during selective alkylation of thiol groups of biological molecules (Scheme l.?).''^ This postulate was supported by the observation that other plant-derived tumor inhibitors are also believed to act by mechanisms that involve intramolecular catalysis of this type, resulting in selective alkylation of the thiol groups of growth regulation enzymes*^ (compounds of this kind are classified as alkylating antineoplastic agents).^^ This rationale seemed to explain the fact that triptonide (4) and 14epitriptolide (81) were inactive against leukemia.  1 R=H 2 R = P-OH  Scheme 1.7  126  Alkylation of thiols by the diterpene triepoxides via hydroxyl-assisted epoxide ring opening  Nevertheless, in more recent years this hypothesis was proven to be incorrect.^'' An X-ray analysis of the product obtained from the reaction of triptolide (1) with propanethiol under identical conditions as those reported by Kupchan,' revealed that the  38  epoxide ring opening by attack of the thiol took place exclusively at CI2 from the convex side of the molecule, thus opening the 12,13-epoxide instead of the 9,11-epoxide, as previously believed. The series of analogues of Tl (1), 121 to 125, synthesized by Tokoroyama and coworkers (vide supra) were screened for antineoplastic activity against cells derived from human carcinoma of the nasopharynx (KB) and L-1210 leukemia in mice^*" and were found inactive. These results suggested that the a,P-unsaturated butenolide moiety may also be important for the tumor-inhibitory activity.  Derivative 127, latter  synthesized by Berchtold et al. also failed to show such activity against P-388 lymphocytic leukemia, which further supported this proposal.'^  Additional evidence  indicates that nucleophilic addition of thiols to the a,p-unsaturated system could also occur and therefore it may play a role in the antitumor activity of some plant-derived compounds.^^  /v5  COjMe  121  t02Me  123  124  39  ^  tOjMe 125  127  More recently, Takaishi et al. isolated the diterpenic quinones triptoquinone A (28) and B (44) from a bark extract of T. wilfordii var. regelii.''^^ These compounds do not possess any epoxide functionality in rings B/C, nor the butenolide in ring A and yet, they were found to have extremely potent inhibitory activity (in )a,molar concentrations) against the release of the inflammatory mediators interleukin l a and ip (IL-1 a and IL1(3) from lipopolysaccharide-stimulated human peripheral monocytes.*^'''^' It is reported that overproduction of IL-1 by synovium is strongly related to the degree of inflammation of arthricular synovial membrane.^^ The strong activity of this quinoid-abietane type of compounds is not unprecedented, though, since the structurally related taxodione (128), for instance, is another well known tumor-inhibitor diterpene.^^  HOOC  28  44  40  The latter seems to indicate that even the presence of the butenolide moiety of triptolide-related compounds may also not be an indispensable requisite for their biological activity. It also suggests that it is feasible that other related quinoid-type derivatives of triptolide (1) may present similar pharmacological activities, although that postulate will only be evaluated as more related derivatives become available for biological evaluation.  1.5  Current research interests.  From the above it becomes clear that there is importance and a high potential today for the development of new, more active compounds of this kind, with improved antineoplastic, immunosuppressive and even male antifertility activities with lower side effects.  The relatively small number of derivatives available to date and the small  quantities in which some of them have been isolated, have been important limiting factors in achieving that goal, as larger amounts are necessary to carry out more complete biological screening and eventually clinical trials of the most promising compounds. From the pharmacological point of view, some of the problems yet to be overcome are the overlapping immunosuppressive and antifertility activities found in several of the naturally occurring compounds. Another problem is the relatively high toxicity when used in large doses or for long periods of time. These factors greatly limit the therapeutic applications of these compounds as a safe method of male fertility control and in prolonged treatments of rheumatoid arthritis and some skin disorders.  41  As for the approaches currently utilized to prepare analogues of triptolide (1), it can be said that although several novel diterpene epoxides have been found in recent years (mentioned in section 1.3.1), the number of new compounds is gradually decreasing since the majority of the active diterpene components of the Tripterygium plants have already been isolated and identified. A few more derivatives have been produced by treatment of triptolide (1) with several reagents, such as HCl, HBr, PrS", etc. which attach themselves to the molecule by opening the 12,13-epoxide (in some kind of "reverse synthesis" approach)."^'' From those studies, tripchloride (7) is important since, as previously mentioned, it shows significantly higher activity and lesser toxicity than triptolide (1). On the other hand, plant cell culture methodology has provided larger amounts of Tl (1) and Td (2) than those found in the plant. However, it was also found that the cell culture of TRP4a, at the present stage of development, have a very limited capability to biotransform  foreign  synthetic  substrates  into the  desired  highly  epoxidized  intermediates. Consequently, this narrows down considerably the variety and amount of products structurally related to Tl (1) that can be obtained by this method. Thus, at present the trends of investigation in this area are gradually moving toward the use of synthetic chemistry. In our laboratory it also became clear that given the conditions mentioned above, biotransformations would not be the most practical way to prepare an ample variety of mono-, di-, or even triepoxide analogues of Tl (1). Therefore, at that point the use of synthetic chemistry seemed to be a better option to achieve that objective, since this approach would hopefully provide a reliable way to  42  prepare larger amounts of any given analogue which might turn out to have important pharmacological activity. In addition, if the synthetic route was appropriately designed, it could potentially provide enough versatility to produce a wider variety of analogues than what could be achieved by means of biotransformations with cell cultures.  1.6  Objectives of the present research  In view of the importance of having a method to prepare a variety of analogues of the diterpene triepoxides in diversity and quantities large enough to allow future pharmacological evaluations, and with the aim of contributing to a better understanding of the structure-activity relationship of this class of compounds, the present investigation is aimed to the development of new and versatile synthetic pathways leading to the preparation of a variety of novel synthetic analogues, structurally related to triptolide (1). These synthetic routes should ideally be able to give access to a series of compounds with a diverse degree of oxygen functionality in rings B and C of the molecule (i.e. a range of mono-, di-, and tri-epoxide derivatives). They should also provide enough flexibility to prepare derivatives with several different stereochemical orientations of the oxygen functionality in the molecule, and desirably render pure diastereomeric compounds. Additionally, since it has been seen that quinoid-type analogues are also very likely candidates to present important pharmacological activities, the development of synthetic pathways to that type of derivatives will also be explored in our investigation.  43  CHAPTER 2  RESULTS AND DISCUSSION  The triptolide project has been under continuous investigation in Dr. Kutney's laboratories for several years, as mentioned in the introduction. The initial studies, once a stable cell line of T. wilfordii was developed (coded TRP4a), focused mainly on the isolation and characterization of the metabolites present, in order to understand the different biosynthetic pathways involved. Next, the interest turned to biotransformation studies on different precursors. The objective was to investigate the capabilities of the TRP4a cell culture to biotransform various intermediates structurally related to the isolated metabolites. In order to carry out those studies, several synthetic precursors with abietane and a6eo-abietane type structures were prepared by some members of our group. A synthesis to isodehydroabietenolide (65) from dehydroabietic acid (DHA, 60) was developed by M. Roberts.^^ Part of that synthesis was then modified by F. Kuri-Brena^^ by adapting a reported procedure for 14-trifluoroacetyl-DHA^' (Scheme 1.4), in order to increase the overall yield.  This synthetic sequence was later extended to produce  isotriptophenolide (71).^^ In the final stages of this project, the interest turned to develop diverse synthetic sequences leading to a variety of epoxidized compounds structurally related to triptolide (1), with different degrees of oxygen functionality, and to new quinoid-type derivatives possessing the aieo-abietane skeleton. The present investigation is concerned with the  44  development of those synthetic routes to structural variations of the triptolide molecule. Initially, the previously developed  conditions to synthesize isotriptophenolide  (71) from dehydroabietic acid (60) were utilized, with a few modifications to the original sequence in order to facilitate its repetition. Significant shortening of this sequence and optimization of the yield of most of the steps involved in the synthesis was later achieved by the present author (vide infra). Dehydroabietic acid (60) was chosen as the starting material for this synthetic route because it is available in large quantities at a very low cost, possesses the required stereocenters at C5 and CIO, and incorporates most of the carbon skeleton of triptolide (1) . Nevertheless, the drawback with DHA is that "the range of synthetic approaches is limited by the chemistry of the starting material rather than by one's imagination".^^ This become more apparent when ring C functionalization was attempted {vide infra). As a consequence, the number of strategies available for the subsequent oxidation/epoxidation of isotriptophenolide (71) was significantly limited.  2.1  Synthesis of Isotriptophenolide (71)  Treatment of dehydroabietic acid (60), in benzene with excess thionyl chloride and a catalytic amount of DMF gave the acid chloride 129 (Scheme 2.1). This reaction (and all the following transformations until the exo-olefm 134) was monitored by IR spectroscopy, which showed only one carbonyl stretch at 1780 cm" (acid chloride) after 2 hrs of reaction. The crude product was treated with sodium azide in acetone and the  In fact, it was later found that dehydroabietic acid (60) was indeed an actual metabolite produced by the TRP4a cell cultures (see Figure 1.1).  45  . r 129 R = C1 ° ^ 130 R = N->  60  C 132 133  R' = H; R = Me R ' = R^ = Me  134  131  135  MeS SMe  o 136  65 m  Scheme 2.1:  c  137 R = CO-Me 138 R = OOC-Me  Synthesis of isotriptophenolide (71) from dehydroabietic acid (60).  a) SOCI2, benzene, DMF; b) NaNj, acetone; c) A, toluene; d) LiAIH4, THF; e) HCHO, HCO2H; f) m-CPBA; g) A; h) O3, -78°C; MezS; i) lithium 4-methyl-2,6-di-t-butyl phenolate, THF, CS2; Mel; j) Me2S=CH2, THF, -20^; k) HCI, MeCN; 1) acetyl chloride, anh. AICI3, CS2, reflux; m) m-CPBA, CH2CI2; n) HCI, MeOH  46  resulting acyl azide 130 was subsequently heated in toluene to effect the Curtis rearrangement to the isocyanate 131 (IR: 2250 cm"').  Reduction of 131 was  accomplished by lithium aluminum hydride (LAH), yielding the secondary methyl amine 132, as evidenced by the disappearance of the isocyanate band in the IR spectrum. An Eschweiler-Clarke methylation on this amine was performed by refluxing with aqueous formaldehyde and formic acid to give the tertiary amine 133. The latter was converted to the corresponding amine oxide by treatment with meto-chloroperbenzoic acid at -20°C. This mixture was then refluxed to eliminate dimethyl hydroxylamine, giving the exoolefin 134, which after purification by column chromatography was obtained as an isomerically pure, colorless oil, in 52% yield from dehydroabietic acid (60). Next, the alkene 134 was converted to the ketone 135 by ozonolysis in a mixture of methanol-methylene chloride at -78°C, followed by treatment with dimethyl sulphide (83% yield). A by-product, result of benzylic oxidation at C7 to give the diketone 135a was obtained (24%) yield) in one experiment in which the time of ozonolysis was prolonged, consequently decreasing the yield of 135 (64%)). This result was in agreement with the observation of previous workers in our laboratory,^^'^^ who were unable to maintain consistent high yields (80-90%)) due to uncontrolled over oxidation of the ketone 135, especially when multi-gram scale experiments were performed.  135 a  47  In the present study it was possible to greatly control this side reaction by modifying certain experimental conditions. It was found that several factors contributed to the over oxidation of 135, one of them was the presence of a very localized site of high concentration of ozone within the reaction solution, when a one-end bubbling tube was used. This problem was easily prevented by utilizing a gas-dispersion tube instead, along with vigorous stirring of the mixture. Other important factors were the temperature and the reaction time. Maintaining a careful control of the temperature at -78°C and a close monitoring of the reaction time (typically 30-60 min., depending on the scale of the experiment) by TLC, proved that it was possible to obtain a consistently high yield for large scale experiments (e.g. 30 g), with only minimal amounts of diketone 135a (<5%) being produced. The next step in the synthetic sequence was the attachment of a one carbon unit at the C3 position (which later would become CI8). This was achieved by deprotonation at C3 with the sterically hindered phenoxide, 4-methyl-2,6-di-^butylphenolate as a base (in order to prevent deprotonation-epimerization at C5) followed by alkylation with carbon disulphide; and lastly treatment with iodine to obtain the ketene dithioketal 136 in nearly quantitative yield after purification by column chromatography. The final transformation to obtain isodehydroabietenolide (65) was achieved in a one-pot conversion comprising two steps. The dithioketal 136 was first treated with dimethyl sulphonium methylide in THF to give a mixture of epoxides at the C4 position, which was hydrolyzed without isolation to give directly the butenolide 65. Purification of this crude mixture  by crystallization  with hexanes-ethyl  acetate  yielded  isodehydroabietenolide 65 as white needles in 55% yield (22% overall yield from  48  dehydroabietic acid, (60). The IR spectrum of the butenolide 65 presented absorption bands at 1750 cm"' for the carbonyl group of the lactone ring, and for the double bonds at 1675 and 1600 cm'' respectively. Its ' H N M R spectrum showed the presence of three aromatic protons at 66.96 (d, J - 1.5 Hz, H14), 7.03 (dd, J = 8.0, 1.5 Hz, H12) and 7.25 (d, J = 8.0 Hz, HI 1). A broad AB quartet was observed at 64.76, characteristic of the methylene protons at CI9. The C20 methyl appeared at 61.01 and the two isopropyl methyl groups were at 51.22 (d, J = 6.9 Hz), respectively.  Some changes of these  characteristic ' H NMR signals were of great usefulness later during the course of this research. It was noted, for example, that changes in the chemical shift or a wider splitting between the AB quartet of the HI9 protons usually reflected a change in the stereochemistry at C5.  Also, when the A/B ring junction is cis in the series of  compounds relative to the corresponding A/B trans compounds, a dowoifield shift of the QQ  C20 methyl signal by 0.2 to 0.3 ppm is commonly observed . Thus, this shifting usually places the C20 methyl resonance at a lower field than the isopropyl signals if the A/B ring junction is cis, which otherwise appears at higher field than the isopropyl methyl signals for the compounds with an A/B ^ra«5 junction. The butenolide 65 was our first intermediate containing all of the basic framework found in triptolide (1), but lacking oxygen fiinctionality on rings B and C. Previous attempts to directly oxidize the aromatic ring of 65, to a quinone-type derivative for example, were unsuccessful and only products of benzylic oxidation were observed.^^'^ These results suggested that further "activation" of the aromatic ring may be required. This led to the preparation of the corresponding CI2 hydroxylated derivative.  49  isotriptophenolide (71). Earlier studies showed that biotransformation of 71 to produce oxidized/epoxidized derivatives at ring C of this compound proceeded only with very limited success (see introduction section). For the purpose of this research, it was thought that this ring C hydroxylated compound would be an interesting intermediate to test chemical oxidation on it, in order to obtain intermediates leading to a new "family" of quinone-type derivatives and diterpene epoxides. The desired quinoid derivatives could also lead to a variety of new oxygenated derivatives at rings B and C of the molecule. Therefore, our attention was directed to prepare the hydroxy butenolide 71. The synthesis of isotriptophenolide (71) involved Friedel-Crafts acylation of dehydroisoabietenolide (65) followed by Baeyer-Villiger oxidation and hydrolysis of the resulting acetate 138 to the desired phenol (Scheme 2.1). The acylation of 65 with acetyl chloride and aluminum chloride in carbon disulphide occurred exclusively at the CI2 position, as seen by ' H NMR, to give 137. The H12 signal disappeared and only C14 and C l l were observed as singlets; the latter being shifted dovrafield to 57.44, as compared with 57.25 in the starting material, as a consequence of the presence of the acetyl group at the neighboring carbon. A nearly quantitative yield of the crude product was obtained. The latter was of sufficient purity to be used without further purification in the next step. The Baeyer-Villiger oxidation of 137 utilizing m-CPBA in methylene chloride provided the acetate 138 in almost quantitative yield. The IR spectrum of this compound displayed the carbonyl band of the acetate at 1740 cm"' and the ' H NMR spectrum showed an upfield shift of the HI 1 and H14 singlets to 56.92 and 7.02 respectively.  50  Finally, hydrolysis of the crude 12-acetoxy-isodehydroabietenolide (138) with concentrated hydrochloric acid in methanol provided, after purification by column chromatography, isotriptophenoiide (71) in 85% overall yield from butenolide 65, as a white crystalline solid (m.p. = 196-198°C). The IR spectrum of this phenol showed a hydroxyl band at 3300 cm"'. The ' H NMR spectrum displayed the hydroxyl proton at 54.62 (IH, s, exchangeable with D2O) and singlets for the protons at C14 and C l l at 66.72 and 6.90, respectively. This step completed the synthesis of isotriptophenoiide (71), which was then used in oxidation experiments involving ring C of the molecule (vide infra).  2.1.1  Modified synthesis of isotriptophenoiide (71)  Although the sequence described above provided us with a route to isotriptophenoiide (71) for preliminary experiments, the requirement for a larger quantity of this intermediate in order to continue our studies motivated the development of a shorter, less time-consuming route to the phenol 71. The existing synthetic route, although efficient, was quite lengthy and required the manipulation of large quantities of reagents and solvents during the course of many steps, making this sequence rather tedious and impractical for the purpose of large-scale preparations. A closer inspection of the synthetic route to 71 showed that the latter steps of the synthesis, i.e., the functionalization at the aromatic ring of isodehydroabietenolide (65) to obtain isotriptophenoiide (71) was short arid efficient (three steps, 85% overall yield;  51  Scheme 2.1), therefore needing no further improvements. In contrast, the preparation of the butenolide 65 was a very lengthy process, in particular at its initial stages, where a six-step sequence was required simply to eliminate the carboxylic group of dehydroabietic acid (60) in order to form the exo-cyclic olefin 134; a synthetic transformation which usually can be achieved in a much simpler way. Therefore, our attention was focussed on improving this part of the synthesis. Oxidative decarboxylation of dehydroabietic acid (60) with lead tetraacetate (LTA) should, in principle, provide the required olefin 134 in one simple step.^*' This decarboxylation reaction is not regiospecific towards formation of the desired exo-olefin 134, and consequently a mixture of the three possible products of elimination was possible (Scheme 2.2). Nevertheless, it would be reasonable to expect a predominance of the product of elimination towards the less sterically hindered and more hydrogensubstituted C19 methyl, favoring the formation of the exo-cyclic olefin 134.  LTA  19/vn tOjH  134 a  134  60  + 134 b  52  This idea was in fact tested by M. Roberts at the starting of the triptolide project in our group. ^^ It was found that a mixture of the alkenes 134, 134a and 134b, in a ratio of 4:2.5:1 was obtained after treating DHA (60) with lead tetraacetate in benzene at reflux for four hours, in a rather low 50% yield. Initial experiments carried out under the same conditions as reported by Roberts confirmed these observations.  Another  additionally discouraging problem in this transformation was the great difficulty encountered in separating the required product from the mixture of olefins. A special chromatographic technique utilizing silver nitrate impregnated silica gel was required to purify 134.^^ These problems made this method inefficient and impractical, and as a consequence this approach was abandoned, and the longer although higher yielding route showed in Scheme 2.1 was adopted at that time. We decided to investigate further this transformation in order to improve the yield of the required product and to find a better alternative for its purification. A survey of the literature revealed that copper (II) acetate has been used as a co-oxidant in conjunction with lead tetraacetate to obtain better yields of alkenes for the oxidative decarboxylation of primary and secondary carboxylic acids.^' The function of the Cu^"^ ions is to oxidize the corresponding primary or secondary radical intermediates to alkenes, thus giving good yields of olefins from this type of substrate. For tertiary carboxylic acids, little or no effect is commonly observed because the tertiary radical formed is efficiently oxidized to olefins by lead tetraacetate.  Nevertheless, the low yield observed (50%) for the  decarboxylation of dehydroabietic acid (60) suggested that a low rate of oxidation for the radical intermediate 140 to the corresponding carbocation 141 might be taking place.  * The ratio of isomeric olefins 134,134a, and 134b was evaluated from integration of their clearly resolved ' H N M R signals for the C20 methyl at 51.02, 1.04 and 1.38, respectively.  53  since this would give the opportunity for side-reactions, such as rearrangements, dimerizations, esterifications, etc. to occur, consequently reflecting in a low yield of the olefins 134-134b (Scheme 2.2).  +  • Pb(OAc)3  CO,  t00-Pb(0A4 139  60  139 -AcO"  140  +  1 R^ tOO-Pb(OAc:b  "slow" 141  Elimination  134,13' 4 a, 134 b  Scheme 2.2: The mechanism of decarboxylation of dehydroabietic acid (60) with lead tetraacetate.  If the above was true, then the addition of Cu(0Ac)2 should, in principle, lead to a faster oxidation of the free radical 140 to 141, therefore leading to higher yields of the corresponding olefins. In order to test this hypothesis, the decarboxylation of DHA (60) was carried out  54  using lead tetraacetate, copper (II) acetate and pyridine in benzene at reflux. After four hours, a mixture of olefins 134,134a and 134b, (similar to that reported previously) was obtained in a ratio of 47:35:18, respectively as evidenced by H NMR analysis, but this time in a 90% yield, after column chromatography.  This result represented a 40%  increase in the yield for this transformation, which could provide a better route to 134. Nevertheless, the problem of separation of this mixture of alkenes remained yet to be solved. The three alkenes present in the mixture had identical Rf values and therefore the required olefin 134 could be separated only by means of special column chromatography using silver nitrate impregnated silica gel.'^ It was clear to us that a purification under those conditions would be a setback for a large scale preparation. Alternatively, a more practical approach was devised. Since the next synthetic step was the cleavage of the double bond of 134 by ozone in order to obtain the ketone 135 (Scheme 2.1), it was thought that the corresponding products of ozonolysis of the other two olefins present in this mixture (134a and 134b) should be of different polarity to 135, which could allow an easier purification of the products in the mixture at this stage. The alkene mixture, 134-134b was then subjected to ozonolysis under the same conditions utilized earlier {vide supra) to provide an easily separable mixture of ozonized products. Purification of this crude mixture by flash column chromatography (hexanesEtOAc 9:1) gave 90%) yield of the ketone 135 (considering the contents of olefin 134 in the alkene mixture), as compared to the 83% yield obtained in our previous sequence (Scheme 2.1).' The overall yield from DHA for this new synthetic approach was 38%), which very closely compared to the 42% yield obtained when following the route to 135, showed in scheme 2.1, despite the low regio-selectivity observed during the  55  decarboxylation step. Further modification of the reaction conditions and amounts of reagents did not have a significant effect on the ratio of alkenes in the mixture. The significant improvement in the yield of this transformation, and the successful elimination of the difficult purification of the exo-olefin 134 from the alkene mixture, provided an efficient and very simplified route to the preparation of isotriptophenolide (71). This route provided only a slightly lower overall yield (2.2%) for the total synthesis and shortening the sequence by 6 steps. This faster and simpler route to the phenol 71 greatly facilitated the preparation of large quantities of this compound for further oxidation experiments. Later, during the course of this research, another alternative method to carry out the oxidative decarboxylation of 60 was also developed (see Appendix).  2.2  Preliminary experiments involving oxidation of isotriptophenolide (71)  As mentioned above in the introduction (Section 1.4), the key to the pharmacological activity of the triptolide-related diterpenes seems to lie in one or more of the epoxide rings present in the abeo-abietanQ skeleton of the molecule; and in some instances in the conjugated unsaturated system of some quinoid-abietane type compounds of this family. Therefore, our first  objective was to transform  the phenolic ring of  isotriptophenolide (71) into a quinoid-type or a dienone system, leading to ring Cunsaturated derivatives. This could give rise to biologically active compounds (these derivatives, even at this stage could already show some pharmacological activity) and  56  should also open several synthetic avenues to functionalize the molecule with various degrees of epoxidation on rings C and B of the skeleton.  71  142  143  or  144  Scheme 2.3: Possible quinoid-type derivatives of isotriptophenolide (71)  The first attempts to carry out the oxidation of the aromatic ring involved treatment of 71 at -20°C with sodium nitrite and oxygen gas in methylene chloride with a catalytic amount of hydrochloric acid. ' TLC monitoring showed a complicated mixture with some unreacted starting material. Use of nitrogen oxide (NO2) and oxygen^^ under the same conditions led to similar results. Decreasing the reaction temperature to -70°C did not showed any significant improvement. On the other hand, use of a dilute aqueous solution of chromium trioxide and acetic acid to 5^C resulted in a similarly complicated crude mixture. In view of these results, an alternative method was attempted, involving a "one-  57  electron oxidation" procedure.  Potassium nitrosodisulfonate (Fremy's salt) has been  reported in the literature^"* to easily carry out oxidation of phenols to ortho- or paraquinones (depending on whichever position is unsubstituted). An experiment with 71 using Fremy's salt (2.6 equiv.) and KH2PO4 in an acetone/water mixture at room temperature for 3h, gave a crude mixture showing two main spots on TLC, along with a small amount (10%) of starting material remaining. The less polar spot on TLC showed a yellow color (before spraying), which suggested that perhaps this could be the orthoquinone 143 or the quinone methide 142. After purification by column chromatography, the less polar compound was isolated as yellow flakes (m.p. = 181-183°C), in 34% yield. This was not any of the expected compounds, but instead, a product of an unexpected bond breakage between C7 and C8 which yielded the para-quinone 145 (Scheme 2.4). The other product formed in this reaction was 7-hydroxy-isotriptophenolide (146), isolated as a mixture of the 7a and 7p isomers (1.3:1 respectively by ' H N M R ) in 34% yield.  OH  K(S03)2NO • , KH2PO4 H2O / acetone  o  71  Scheme 2.4: Reaction of isotriptophenolide (71) with Fremy's salt  For the sake of simplicity, the same numbering used for the 18(4->3)a6eo-abietanes was used for 145.  58 Compound 145 had a molecular formula of C20H22O5 by high resolution mass spectroscopy, indicative of the addition of two extra oxygen atoms and loss of two hydrogen atoms, with respect to the starting material 71. The IR spectrum of 145 showed the butenolide carbonyl at 1745 cm'', characteristic aldehyde bands at 2875, 2740, and another carbonyl band at 1720 cm'', along with typical quinone bands at 1675 and 1600 cm''.  The ' H N M R spectrum  exhibited a signal at 59.72 (IH, s), not exchangeable with D2O, confirming the presence of an aldehyde group in the molecule. The characteristic broad singlet of the H5 signal shifted to 54.17, an extremely low field, as compared to its position at 62.69 in ITP (71), which suggested the presence of an electron-rich or a heteroatom-carrying group in the nearby environment. The H5 proton showed only one cross peak in the COSY spectrum with a multiplet at 62.50, indicating that this signal must correspond to the H6 protons. Irridation of the H5 proton at 64.17 caused NOE enhancement of the H6 signal at 52.50, the aldehyde proton at 69.75, a signal at 56.63 (IH, s) and very weakly to one of the H19 protons at 54.57. Similarly, irradiation of the aldehyde signal (69.75) enhanced the H5 and H6 signals, therefore indicating that the aldehyde proton must be located at C7. The proton at 66.63 showed a weak cross peak with H20 at 51.10 (3H, s) in the COSY spectrum and its irradiation resulted in enhancement of the H5 and H20 signals, thus identifying this signal as Hll (free rotation of the quinone ring around the C9-C10 bond allowed the Hll to be close in space to both of these protons). This suggested that the configuration of this compound was that showed in figure 2.1.  59  145 Figure 2.1: The three dimensional structure of quinone 145.  At the low field there was observed another singlet at 56.47, which had cross peaks with the H15 septet at 52.98 (IH, J = 6.0 Hz). Irradiation of this signal at 56.47 showed enhancement to the HI5 proton and the two isopropyl methyl groups, HI6 and H17, at 51.11 and 1.13 (3H each, both d, J = 6.0 Hz); confirming that the irradiated signal corresponded to H14. The H19 protons were located at 54.57, presenting a wide splitting (Av = 0.26 ppm) between the AB quartet due to the nearby presence of the aldehyde group. This signal showed cross peaks with the signals at 52.23 (IH, m) and 52.36 (IH, br d, J = 18 Hz); which, in turn had cross peaks with each other and therefore were assigned as the two H2 protons. Those H2 signals also presented cross peaks with the protons at 61.58 (IH, ddd, J = 13.6, 5.7, 2.4 Hz) and an overlapped multiplet at 52.50, which were coupled with each other and thus must correspond to the HI protons. These signals accounted for all the protons observed in the ' H N M R spectrum. The presence of the quinone's carbonyl groups was evidenced by  C NMR  spectrum. Two signals, at 5187.1 and 187.9 were clearly due to the C8 and C12 carbonyl groups of the quinone ring. The C7 aldehyde carbon was found at 5199.2. The signal at  60  §173.3 was assigned to the C18 lactone carbonyl and the two quaternary signals at 5125.7 and 160.8 were assigned to the vinylic carbons C3 and C4 respectively, since they appeared at the same chemical shifts as in all the other related compounds carrying the butenolide moiety. The three remaining quaternary carbons found at §41.1, 150.8 and 153.7 corresponded to the CIO, CI3, and C9, respectively. The vinylic carbons, C14 and C l l , carrying single protons were found at § 132.7 and 135.3, respectively.  The  remaining aliphatic carbons of the molecule were similarly assigned on the basis of its HMQC (heteronuclear multiple quantum coherence) spectrum.  Thus, the signals at  §17.8, 21.1, 21.3, 26.4, 32.7, 34.1, and 42.8 were assigned as C2, C16, C17, C15, CI, C5 and C6, respectively.  146 146 a = a-OH 146 b = p-OH  The second product of the oxidation of ITP with Fremy's salt was the diol 146. The IR spectrum of 146 showed an hydroxyl band at 3500 cm"' as well as the butenolide bands at 1745 (C=0) and 1680 (C=C). This compound presented a molecular ion peak at m/z 328 in its mass spectrum, consistent with the addition of one oxygen atom to the  61  molecule. Its ^H NMR spectrum showed two sets of signals, clearly indicating that this compound was a diastereomeric mixture of the 7a and 7P-hydroxy phenol. The H7 protons appeared as a broad singlet at 64.89. The H14 and HI 1 signals of the a-hydroxy compound (146a) were found at 57.14 (IH, s) and 6.73 (IH, s); while the same signals for the (3 isomer (146b) were at 57.36 (IH, s) and 6.69 (IH, s), respectively.  The  significant difference in chemical shift of the H14 signal (0.22 ppm) between these two isomers was due to the different stereochemistry of the hydroxyl group at C7. The Phydroxyl group in 146b is in a quasi-equatorial orientation and in close proximity to proton H14 in space. On the other hand, in 146a the a-hydroxyl group is in an axial orientation, placing it further apart from proton HI4, and consequently affecting to a lesser extent the chemical shift of the H14. These distinct ^H NMR features allowed us to evaluate the ratio of each diastereoisomer in this inseparable mixture as a:P 1.3:1. Since the pure diastereoisomers were later prepared, a more comprehensive discussion of their spectral data is presented below (vide infra). The formation of compounds 145 and 146 in this reaction probably proceeded via the radical intermediate 147, which then underwent a nitroso radical addition at C8 to give 148, followed by fragmentation to give the quinone-aldehyde 145 (Scheme 2.5). Alternatively, the intermediate 147 could also suffer elimination of an hydrogen radical at C7 to generate the quinone methylide 142 followed by a Michael-addition-type reaction by water to produce the diol 146.  62  ITP  71  ON(KS03)2 -H ^  147  H  148 • ON(KS03)2  H O  142  H2O  O—N(KS03)2  • HN(KS03)2  OH  146  Scheme 2.5: Proposed mechanism for the formation of compounds 145 and 146.  The latter suggested that the desired quinone methylide was in fact obtained under the reaction conditions, but it may be too labile to be isolated and consequently, underwent further reaction very rapidly to generate 146. In order to test this postulate, an  63  experiment was performed using ferric chloride, and an oxidant which acts by a similar mechanism, i.e., by "one-electron" oxidation. Isotriptophenolide (ITP, 71) was treated with a catalytic amount of ferric chloride (0.05 equiv.) in methanol-water solution at 60''C for 9h. Purification of the crude product by column chromatography gave the yellow compound 149 as the only product (58%), along with some unreacted starting material (12%).  OH  Compound 149 was obtained as bright yellow needles (m.p. 287-289°C), having a molecular formula of C20H20O4. Its IR spectrum showed a new carbonyl signal at 1645 cm'' along with the phenolic hydroxyl group at 3495 cm'' and the butenolide carbonyl at 1745 cm''. The ' H NMR spectrum showed several significant changes with respect to that of ITP (71). The aromatic protons H14 and HI 1 shifted downfield to 58.07 and 6.90, as compared to 56.90 and 6.72, respectively in isotriptophenolide (71) due to the presence of a carbonyl group at C7. The characteristic broad signal of H5 at 52.69 and the H6 multiplet at 51.88, seen in the ' H NMR spectrum of the starting material, were not observed in the spectrum of 149. Instead, a new signal at 56.31 (IH, s), corresponding to a vinylic H6 proton was observed.  The identity of this signal as H6 was further  evidenced by an NOE enhancement of the H19 signal at 55.04 when H6 (56.31) was  64  irradiated, and by the wider spHtting (Av = 0.17 ppm) of the HI9 AB quartet due to change at the C5-C6 positions. The strong absorptions observed in the UV spectrum of 149 at X (log s) 213.2 (4.31), 251.3 (4.16), 284.5 (4.02) and 344.3 (3.66), were in support of the proposed structure. The isolation of this compound also suggested that the reactions may have proceeded via the same quinone methide intermediate 142 but, once again, it was too reactive and it was oxidized further to form the insaturated ketone 149. Variations in the reaction temperature showed that at lower temperature (40-5 O^C) the reaction progressed much slower (to give the same product) and at room temperature no reaction at all was observed. In order to investigate the possibility of forming this quinone methide 142 under conditions other than through free radicals, and also to establish further evidence of its presence as a reaction intermediate, an experiment of oxidation with ITP (71) utilizing DDQ (1.2 equiv) in methanol (Ih, room temperature) was performed in order to trap this intermediate. Two main products, comprising of a diastereomeric mixture(2:l) of 7a and 7pmethoxy-isotriptophenolide (150, 62%) and 7-oxo-isotriptophenolide (151, 15%), were obtained (Scheme 2.6).  150  65  The H NMR spectrum of compound 150 was similar to that of the diol 146. The assignment of the H14 singlet of the two isomers, the Vp-methoxy ITP (150b) at 57.28 and 7a-methoxy-ITP (150a) at 57.13; and the corresponding H7 protons at 54.64 (dd, J = 8.8 Hz) and 4.30 (br s) respectively, was based on their NOE results.  The methoxy  group for both the isomers appeared as overlapping singlets at 53.44. The presence of these methoxylated products provided further supporting evidence of the in situ generation of the quinone methide intermediate 142, which due to its highly unstable nature, suffers immediate subsequent attack by nucleophiles present in the reaction medium (such as methanol) to give the observed products (Scheme 2.6).  151  The keto-phenol 151 was obtained as a white solid (m.p. 286-288°C, dec.) with a molecular formula of C20H22O4 (i.e. one oxygen atom more and two hydrogen atoms less than isotriptophenolide, 71). The appearance of an extra carbonyl band at 1690 cm'' was observed in its IR spectrum, corresponding to a ketone group. The ' H NMR spectrum showed a large downfield shift for the H14 (67.97), H6 (52.68) and H5 (53.17) signals, as compared to the 66.90, 1.88 and 2.69, respectively for the same protons in the starting material. The H7 protons, appearing at 52.95 in isotriptophenolide (71) were not present  the pure 7a-methoxy-diastereomer was later obtained in another reaction, so a more detailed discussion of the corresponding spectroscopy will be presented in the next section below.  66  in the spectrum of the keto-phenol 151. All of this evidence indicated the presence of a carbonyl group at the C6 position.  MeOH 1,6-addition o  71  142  150  DDQ  151  Scheme 2.6: Proposed mechanism for the formation of compounds 150 and 151.  In view of these results, it was clear that the quinone methide 142 could not be isolated, but perhaps it could be possible to "trap" this reactive intermediate with an epoxidizing agent, in order to obtain the epoxy dienone 72 or 73 (Scheme 2.7). An experiment of oxidation of isotriptophenolide (71) with DDQ (1.2 equiv.) was carried out under the same conditions as before, but replacing methanol by THF (to  67  R-OQ9  71  142  72 = a-epoxide 73 = P-epoxide  Scheme 2.7: Proposed trapping of the quinone methide 142 with a peroxide anion.  prevent nucleophilic attack by the solvent). After 20 minutes of reaction time, a freshly prepared solution of hydrogen peroxide and dilute sodium hydroxide (1.5 equiv.) was added and stirring continued for Ih.  The yellow crude was a complex mixture of  compounds, difficult to purify, so only a partial purification was achieved by column chromatography, giving two fractions. The ^H NMR analysis of these fractions indicated the presence of recovered starting material, 7-oxo-ITP (151), ketone 149, a small amount of the mixture of diols 146 and some products of dehydrogenation of ring B. The latter was evidenced by the series of vinylic signals (d and dd) between 65.5-6.7. Unfortunately there was no indication of epoxidation at ring B or C in neither of these fractions since no signal between 53.6 to 3.8, corresponding to a H7 or HI 1 epoxy proton was observed, as expected if an epoxide ring was located at C7-C8 or C9-Cll'''*^'''^ Epoxidation at CI 3-CI4 did not occur either, as shown by the absence of a signal around 53.0 for an HI4 epoxy proton in the spectrum.  Another experiment under similar  conditions, but using a solution of tert-hxxtyl hydroperoxide and triton-B in pentane (1.8 equiv.) as the epoxidating agent instead of hydrogen peroxide-sodium hydroxide, gave similar results. Variation in the reaction time allowed before addition of the peroxide did  68  not enhance the reaction profile. These results indicated that the quinone methide 142 was difficult to epoxidize under these reaction conditions. It was apparent from these observations that the side reactions leading to by-products were much faster than the nucleophilic attack of the peroxide anion, or perhaps steric hindrance prevented the peroxide reagent from approaching the intermediate 142. Because of these findings, further investigation of this reaction was not pursued.  2.3  Synthesis of the 8-methoxy-dienones, 153  Since neither isolation of the quinone methide 142 nor further in situ epoxidation seemed feasible, another strategy was explored. Among the wide variety of oxidation methods utilized for oxidation of phenols, the one electron processes to give phenoxy radicals are some of the most commonly used. Such oxidants include reagents containing Fe (III),^^-^^ Ag (I),^^ Pb (IV),^^ Mn (111),'°° Cu (11),'°' among others. Two electron phenolic oxidations have also been frequently used.' "  Para- and or^/zo-quinones are  among the most frequently obtained products, although the reactive intermediate species generated may also undergo further reactions such as coupling, quinone methide formation, etc. leading to unexpected by-products.  Unfortunately, as seen from our  earlier experiments, the latter proved to be the case for the oxidation of isotriptophenolide (71) with some of those reagents. As mentioned previously, another very useful product of oxidation of isotriptophenolide (71) for the present study, would be the dienone derivative such as 144  69  (Scheme 2.3). Such dienones could be further epoxidized in order to obtain several potentially bioactive compounds. Several methods have been reported to produce that type of derivatives.  Our attention focused on the two electron phenolic oxidation  processes which give aryloxonium ions (Ar-O"^), since these seemed to offer the advantage of reacting with nucleophiles in a slower and more controllable manner (in either, inter- or intra-molecular fashion) than the corresponding radical counterparts (Ar-0-). ^ An increasingly used method to generate such species is by utilizing hypervalent iodine reagents,'^^''"^ such as phenyl iodoso diacetate (PIDA). This reagent is most frequently used since it is readily available and gives rise to intermediates which decompose under mild conditions to efficiently generate Ar-0* species. It has been reported that PIDA in nucleophilic solvents, such as methanol, interacts smoothly with some ;7(3ra-substituted phenols to yield the corresponding 4-alkoxy-dienone such as 152 (Scheme 2.8) 108, 109  Y  (x  /)  OH +  PhI(0Ac)2  -^  Y  C\^  Phi *-  /)—-0—IPh-(OAc)  Y = i-Bu- , PhCt^-, CH3- .CHjOY/^OH  or  MeO  \  MeOH  / 152  Scheme 2.8: Oxidation of/?ara-substituted phenols with PIDA  Thus, isotriptophenolide (71) was treated with PIDA (1.2 equiv.) in methanol under argon. After chromatographic purification, a diastereomeric mixture of the 8-  70  methoxy-dienones 153 (a-methoxy:P-methoxy = 29:71 by ' H NMR; 45% yield), pure 7a-methoxy-isotriptophenolide (150a, 26%) and 7-oxo-isotriptophenolide, (151, 17%) were obtained (Scheme 2.9).  OH  +  71  153  151  150 a  Scheme 2.9: Reaction of isotriptophenolide (71) with PIDA in methanol  The diastereomeric product, 8a- and 8p-methoxy dienone 153 was isolated as an inseparable mixture showing one single spot on TLC (s.s. CHCla-MeOH 98:2, hexanesEtOAc 1:1 or CH2Cl2-acetone 94:6). Attempted recrystallization of the solid mixture in several solvent systems was unsuccessful in achieving any significant separation. The ' H NMR spectrum of 153 showed two sets of signals. Four singlets were observed at low field, corresponding to the vinylic protons of ring C of the dienones. From these, the two highest peaks appeared at 66.34 and 6.24, while the lowest were located at 56.39 and 6.12 respectively. Irradiation of the signal at 56.24 and, in turn, the signal at 6.12. both showed NOE enhancement of the isopropyl methyl protons located between 51.03 and 51.10, as well as the corresponding C8 methoxide singlets at 53.00 and 2.98 respectively. This suggested that the irradiated signals were the H14 protons of the two diastereoisomers and consequently the signals at 56.34 (major isomer) and 66.39  71  (minor) must be the corresponding HI 1 protons, although at this point it was not clear to which isomer each of those signals corresponded. Nevertheless, irradiation of the H20 protons of the major isomer, located at 61.24 (3H, s), resulted in enhancement of the corresponding 8-methoxide signal located at 53.00. This strongly suggested that these signals (major isomer) must correspond to the diastereoisomer with the (B-oriented methoxide, since for this isomer the C20 methyl and the methoxy group at C8 are in a 1,3-diaxial position with respect to each other, and therefore should present an NOE enhancement. Further irradiation of the methoxide signals at 63.00 (major isomer) and 2.98 (minor isomer) showed enhancement of only the H20 proton at 61.24 (major), confirming this assignment. Therefore, the ratio of diastereoisomers 8P-methoxy to 8amethoxy-dienone was 71:29 respectively.  This ratio seemed to be in apparent  disagreement with the expected result considering the steric interactions in the ring B of the molecule. Since a methoxy group at position C8 with a P orientation would place this group in a 1,3-diaxial relationship with respect to the C20 methyl group, it could be expected, taking into consideration the stronger steric interactions for this isomer than the corresponding 8a-oriented methoxy isomer, that this should reflect in a larger percentage of the 8a-isomer being formed over the 8p-compound; which was opposite to what it was observed.  Nevertheless, in a more detailed analysis it could be seen that the  conformation of the B ring in the 8p-methoxy-dienone (153b) was chair, while in the aisomer (153a) it was a twisted boat. Consequently, the former (153b) would be expected to be more stable than 153a; and for this reason it was formed in a larger amount than the a-isomer (Figure 2.2). Furthermore, the minimized tridimensional model of compound  See note 1 in appendix.  72  153b showed that both the SP-methoxy group and the C20 methyl group had a semi-axial orientation; and the methyl group of the methoxyl could be oriented away from the C20 methyl due to the free rotation along the C8-oxygen bond. Both of these effects reduce considerably the unfavorable steric interactions.  153 b Figure 2.2: The sterochemical structure of diastereoisomers 153a and 153b.  Another product of this reaction, the methoxy phenol 150a, was obtained in a diastereomerically pure form, as a white powder (m.p. 109-111°C) with a molecular formula of C21H26O4.  150 a  It was optically active ([a] p : +54.5°) and its IR spectrum indicated the presence of an hydroxyl group (3580 cm''), a carbonyl corresponding to the lactone (1740 cm"'),  73  and a benzene ring (3010, 1610 cm"'). The ' H N M R spectrum showed clear signals, with better defined coupling patterns than the spectrum of the previously obtained diastereomeric mixture of 7a- and 7p-methoxy-ITP (150) (see Section 2.3). The two aromatic proton signals at 57.10 (IH, s) and 6.71 (IH, s) were assigned to H14 and Hll respectively, based on their NOE results. Irradiation of Hll enhanced the hydroxyl signal at 55.21 (IH, s, D2O exchangeable) and the multiplet at 52.36 corresponding to Hip. Irradiation of the H14 (57.10) proton resulted in enhancement of the signals at 54.27 (IH, d, J = 3.9 Hz, H7), 53.44 (3H, s, C7-methoxy), 53.15 (sept, J = 6.9 Hz, H15), and the isopropyl methyl groups at 51.25 and 1.23, respectively. In the COSY spectrum, H7 (54.27) presented cross peaks with signals at 61.96 and 2.07. These two signals were coupled with each other and showed cross peaks with a proton overlapped with HI5 at 53.15. Thus, the signals at 51.96 and 2.07 must be the H6 protons and the one at 53.15 was assigned to H5. When one of the H6 protons, located at 51.96, was irradiated, NOE enhancements were observed for the other H6 (52.07), the H7 (54.27), and the H20 singlet (50.94); with no enhancement observed for the H5 signal (53.15). Since H20 is Poriented, then the irradiated H6 and H7 must also have the same P configuration, and consequently the methoxy group at C7 (53.44) must be a-oriented.  2.3.1  Attempted epoxidation of the 8-methoxy-dienones (153)  Due to the difficulty in achieving separation of the mixture of 7a- and 7pmethoxy-dienones (153), direct epoxidation of this diastereomeric mixture was attempted. In order to prevent possible hydrolysis of the lactone moiety under the basic  74  conditions required for the epoxidation reaction, relatively "weak" bases were used for our first experiments. Treatment of the mixture of methoxy-dienones 153 with hydrogen peroxide and sodium bicarbonate (2 equiv.) in a mixture of THF and water (room temperature) gave no reaction after 48 hrs. The use of potassium bicarbonate under similar conditions or in a methanol-water mixture, produced only trace amounts of a more polar compound. Finally, reaction with hydrogen peroxide (10 equiv.) and 5% aqueous sodium hydroxide (1.5 equiv.) in methanol (room temperature for 32hrs), or treatment with an excess of tert-huty\ hydroperoxide and Triton B (1 equiv.) in benzene at 40°C for 21 hrs (little reaction progress was observed when this reaction was carried out at temperatures lower than 40°C), led to partial consumption of the starting material 153. After the above mentioned times, a slightly more than half of the starting material was transformed to a slightly more polar product (TLC: hexanes-EtOAc, 4:6) and no further change was observed by extending the reaction time or adding more amounts of reagents. Both of these sets of reaction conditions produced the same results, except that the treatment with tert-huty\ hydroperoxide gave somewhat better (5%) total recovery yield. As a result, a larger scale experiment (200 mg of 153) was performed using those conditions. After purification by preparative TLC, 54 mg of diastereomerically pure 8a-methoxy-dienone 153a (a 93% recovered yield, based on a 29 % contents of this isomer in the starting material) and 30 mg of dienone 154, with the epimerized A/B-ring junction (21% yield, considering that the starting material contained 71% of the Sp-methoxy-isomer) were obtained, instead of the expected epoxidized compounds 155 and 156.  75  ^BuOOH / Triton-B  -X— Benzene, 40°C  153  155  156  153 a  154  Scheme 2.10: Attempted epoxidation of the 8-methoxy-dienones 153  Compound 153a, a white solid (mp 203-205°C) with a molecular formula of C21H26O4, was easily identified since its ' H NMR spectrum showed the same set of signals observed for the minor isomer present in the starting material, previously assigned as the 8a-methoxy-dienone, 153a (see Section 2.4). A clear NOE enhancement of the methoxy singlet at 62.98, observed when the H5a proton at 53.72 (IH, br s) was irradiated, confirmed the a orientation of the methoxy group. Compound 154 was isolated as an optically active ([aj^ : -16.7°) white solid (m.p. 162-164''C). Its IR spectrum showed a band for a dienone carbonyl group at 1668 cm"' and a lactone carbonyl at 1757 cm"'. In the ' H NMR spectrum of 154 the H20 protons (51.39) appeared at lower field than in 153a (51.04). This downfield shift has usually been observed when the A/B ring junction is cis.  Irradiation of this H20 signal  enhanced both the H5 proton at 62.71 (IH, s) and the 8-methoxy group singlet at 52.96.  76  These NOE effects established the (3 stereochemistry of both the C8 methoxy and the H5, therefore indicating that epimerization took place at C5. The vinylic protons on ring C also shifted slightly, appearing as singlets at 56.14 (H14) and 6.36 (Hll), as compared to 56.12 and 6.39 respectively for 153a. The HI9 signals appeared as a broad singlet at 54.64. The rest of the NOE and COSY results were in agreement with the structure assigned (Figures 2.3 and 2.4).  Figure 2.3 : Major NOE's observed for compound 154  Figure 2.4 : Major correlations observed in the COSY spectrum of compound 154  The recovery of pure 8a-methoxy-dienone 153a, (in practically the same amount contained in the original diastereomeric mixture 153, i.e., 93% recovered yield), together  77  with the fact that the other isomer (the 8|3-methoxy-dienone, 153b) only suffered epimerization at C5, but no epoxidation of any of these derivatives took place, was probably due to steric factors within rings B and C of the molecule. Apparently, the large volume of the methoxy group at C8 in both diastereomers 153 blocked the accessibility of the peroxide anion to the neighbouring C14, thus preventing epoxidation at the CISCM double bond. The double bond between C9 and Cll was blocked as well by the angular C20 methyl group. On the other hand, because of the boat conformation of ring B of the 8a-isomer 153a, the C8 methoxy group was in close proximity to the H5 (Figure 2.2), thus, the steric hindrance of the methoxide effectively prevented epimerization at C5 of this diastereomer.  In contrast, in the SP-methoxy-isomer (153b) the methoxy group was  oriented opposite to the H5, so it did not exert any blockage to this proton. Consequently, this allowed the base (HO") to have free access to abstract the proton at C5, resulting in the diastereoselectively epimerization of 153b at this center to give the corresponding AfQ-cis fused compound, 154.  The low recovery (21%) of this C5  isomerized 8p-methoxy-dienone (154) indicated that some other side reactions (such as hydrolysis of the lactone ring) may have taken place as well.  2.4 Synthesis of the (7,8)a-epoxy- series  2.4.1  Synthesis of the (7,8)a-epoxy-dienone, 72  From the above results it was obvious that in order to achieve epoxidation of the  78  dienone system in ring C, it would be necessary to replace the bulky methoxy group at C8 for a smaller group, such as an hydroxyl group, or better yet, an epoxy group between C7 and C8. Attempts to demethylate the 8-methoxy-dienones 153 were unsuccessful (BCI3 in CH2CI2; BF3-OEt2 / «-Bu4NI in CHCI3; at different temperatures)"°''" giving complicated mixtures and low recovery yields.  This indicated that the required 8-  hydroxy-dienones, 157 and 158, needed to be prepared using a different strategy.  Demethylation  X-  Epoxidation  •  153  EPOXIDIZED DERIVATIVES  157 = a-OH 158 = p-OH  Nevertheless, a possible alternative route was devised to obtain the (7,8)a-epoxydienone (72), perhaps a better candidate than the 8-hydroxy-dienone for further epoxidation; since the epoxy group between C7 and C8 could exert less steric hindrance than an hydroxyl group at C8. Furthermore, if epoxy-dienone 72 could be obtained, this compound itself would be of great interest since it already incorporates an epoxide functionality into the molecule's framework. Therefore, our attention turned into obtaining this compound. Previous attempts to obtain the (7,8)a-epoxy-dienone (72), or its (7,8)P-isomer by epoxidizing the intermediate quinone methide 142 were unsuccessfhl (Scheme 2.7). The new approach to this problem was based on the possibility of carrying out an intramolecular "trapping" of  79  a carbocation intermediate formed at C8 during the oxidation of an appropriately substituted isotriptiphenolide derivative 146a, with PIDA (Scheme 2.11).  72  146 a  150 a  PIDA  Scheme 2.11: Proposed synthetic plan for the (7,8)a-epoxy-dienone 72  In order to prepare the diol 146a required to test this concept, we decided to take advantage of the availability of the pure 7-a-methoxy-isotriptophenolide (150a), obtained previously from oxidation of ITP (71) with PIDA (Scheme 2.9). It was found that the best conditions to carry out the de-methylation of 150a were by slow addition of a dilute solution of BCI3 (0.9 equiv.) in methylene chloride to a solution of 150a in the same solvent, kept at -75°C under argon; and subsequently carefiil neutralization (sodium bicarbonate solution) and work up to give the desired 7a-hydroxy-isotriptophenolide (146a) in 62% yield (Scheme 2.12).  80  OH  BCI3 / CH2CI2 -75 °C  146 a  150 a Scheme 2.12: Demethylation of 150a with boron trichloride  Compound 146a was obtained as an optically active ([a] • : +27°) white powder (mp 131-132.5°C) with a molecular formula of C20H24O4. Its ' H NMR spectrum showed disappearance of the methoxy group signal and a significant downfield shift of the H7p proton (from 54.27 in 150a to 54.85 in the diol 146a). The AB quartet of the H19 protons at 54.78 showed a wider splitting pattern (Av = 0.12 ppm) due to the stronger deshielding effect of the hydroxyl group on HI 9a. The rest of the signals presented little change with respect to the starting material. A small amount (13%) of the corresponding dehydrated product (the C6-C7 olefin 159) was isolated as a by-product.  The amount of this compound was larger in  experiments using more than 0.9 equivalents of BCI3 or when the reaction was carried out at higher temperatures.  159  With the diol 146a on hand, the next step was to test its cyclization reaction with PIDA.  An experiment was carried out by adding a solution of 7a-hydroxy-  isotriptophenolide (146a) in acetonitrile to a cooled solution (0''C) of PIDA (1.2 equiv.) in the same solvent. The crude mixture was purified by preparative TLC to yield the desired (7,8)a-epoxy-dienone 72, in 22% yield and the oxidize product 7-oxoisotriptophenolide (151, 54% yield) (Scheme 2.13). A trace amount (3% yield) of the quinone-aldehyde 145 was also isolated.  146 a  72  151  Scheme 2.13: Preparation of the (7,8)a-epoxy-dienone, 72 from 146a.  The formation of the epoxy-dienone 72, even in a low yield, was a very significant result. This compound had been isolated in our group as a diastereomeric mixture with the corresponding (7,8)P-epoxy-isomer in small quantities (c.a. 6% yield of  82  each isomer, based on its contents in the mixture) from biotransformation experiments using TRP4a cell culture.^^ However, this was the first time compound 72 was obtained synthetically and with the added advantage of being able to prepare it diastereomerically pure. Compound 72 was obtained as white needles (m.p 187-188°C) with a molecular formula of C20H22O4. It was optically active ([a]^ : +169.3°) and its IR spectrum showed absorptions indicative of, a dienone carbonyl at 1662 cm'' and for the carbonyl of the lactone at 1752 cm' , as well as disappearance of the hydroxyl signals at 3375 cm''. Its ' H N M R spectrum showed some significant changes with respect to the starting material. The ring C proton signals shifted upfield to 66.48 (IH, s) and 5.99 (IH, d, J = 0.9 Hz), respectively; and a new sharp doublet was observed at 63.81 (IH, d,J= 2.2 Hz). Irradiation of the proton at 66.48 showed NOE's to the signals of the Hip at 52.19 (IH, br dd, J = 6.2 Hz) and of the C20 methyl at 50.95, suggesting that the irradiated signal corresponded to Hll (Figure 2.5). Irradiation of signal at 55.99 enhanced the isopropyl group signals at 52.95 (IH, sept., J - 6.9 Hz, H15); 51.05 and 1.08 (3H each, both d, J = 6.9 Hz, H16, H17), and the new signal at 63.81. This indicated that the signal at 65.99 was HI4 and the doublet at 63.81 corresponded to the H7 proton. The H7 signal was coupled with two protons at 62.02 (IH, br dd, J = 14.4, 12.8 Hz) and 2.30 (IH, br ddd, J = 14.4, 3.2, 2.2 Hz), suggesting that those were the H6 protons (Figure 2.6). The H6 signals were in turn coupled with a signal at 63.23 (IH, br d, J = 12.7 Hz). This proton presented a small coupling with the H19 signal at 54.70 (2H, br ABq, Av = 0.08 ppm, J = 17.2 Hz). This indicated that the signal at 63.23 was H5.  83  The H19 proton also showed cross peaks to the two H2 signals at 52.31 (IH, m) and 2.47 (lH,brd,J=17.9Hz). Finally, the low chemical shift at which H5 appeared, together with the fact that the a-stereochemistry of the hydroxy group at C7 in the staring material was well established and it should not change during the course of the transformation, established the a-orientation of the epoxide group between C6 and CI.  A comparison of these  spectral data with those of the corresponding compound isolated from TRP4a cell cultures^^ confirmed the structural assignment.  Figure 2.5: Major NOEs observed for compound 72.  Figure 2.6: Major proton correlations observed in the COSY spectrum of compound 72.  84  On the other hand, the fact that significant oxidation of the C7 alcohol group of 146a, to the corresponding ketone took place was rather unexpected. Although PIDA is known to be able to oxidize both primary and secondary alcohols into carbonyl compounds, this usually takes place at temperature above 80°C."^ Alcohols normally do not appreciably react with PIDA at ambient temperature.'"^ Thus, an hypothesis could be that the PIDA rapidly reacts with the epoxide functionality of the epoxy dienone 72 as soon as this compound is formed, effectively promoting opening of the oxirane ring to give 7-oxo-ITP (151) (Scheme 2.14).  OAc  Scheme 2.14: A possible mechanism to compound ISlfrom 146a via oxirane ring opening  85  In order to test this hypothesis, an experiment was carried out by subjecting the epoxy dienone 72 to the same reaction condition utilized for the cyclization reaction of 146a (1.2 equiv. PIDA in acetonitrile, 0°C). After Ih no change was observed (the original cyclization/oxidation reaction of 146a took place in 30 min) so the reaction mixture was warmed up to room temperature and stirred for another hour; but still no reaction was observed and 72 was recovered unchanged after work-up. The results of this experiment showed that the formation of 151 did not proceed by that pathway. An alternative postulate could be that the keto phenol 151, was formed by direct oxidation of 146a by PIDA (Scheme 2.15).  OH  I  Ph  OAc  146 a  L  "OAc  151  Scheme 2.15: An alternative mechanism for the oxidation of 146a to 151.  This unusually high reactivity of the C7 hydroxy group over the CI2 phenolic hydroxy group may have been due to a combination of several factors. The benzylic nature of the C7 hydroxy group and its easy accessibility could have made it more labile than a "regular" secondary alcohol towards oxidation by PIDA. On the other hand, the bulkiness of the isopropyl group next to the phenolic hydroxyl, probably made it more difficult for the PIDA to approach to the C12 hydroxy group, consequently decreasing its "reactivity".  86  Nevertheless, a more likely explanation contemplates the formation of the two products of this reaction via a common intermediate, which follows two different pathways. The first reaction which takes place during the oxidation of phenols with I no  PIDA is the formation of a phenolic oxonium ion,  which is in resonance with the  corresponding carbocation species at the para position, 146-1 (Scheme 2.16).  This  reaction intermediate could then have reacted in two possible ways: an intramolecular nucleophilic attack of the C7 hydroxyl group to the C8 carbocation to form the oxirane ring, leading to 72 (path A); or suffer elimination of the benzylic proton at C7 to generate a quinone methide, leaving an enol function at C7-C8, which quickly tautomerize to give 7-oxo-isotriptophenolide (151) (path B).  Scheme 2.16: Proposed mechanism for the formation of compounds 72 and 151  87  Given the acidic nature of the benzylic proton H7 and the strong driving force for stabilization of the molecule by forming a conjugated system (the quinone methide), the path B (elimination of the H7 proton) seemingly was favored over path A (the cyclization to generate an strained oxirane ring at position C7-C8).  This explained better the  predominance of 7-oxo-isotriptophenolide (151) over the (7,8)a-epoxy-dienone, 72 (c.a. 2.5:1, respectively). Numerous attempts to increase the yield of the desired epoxy dienone 72 were carried out by changing several reaction conditions such as; the solvent used (methylene chloride, benzene, acetone, trifluoroethanol, THF), the addition of bases (potassium carbonate, pyridine, potassium hydroxide), the reaction temperature (from room temperature to -75°C), reaction time, amount of PIDA (0.5 to 2 equiv.), and the order in which the reagents were added. Only a small increase in yield was achieved when this reaction was carried out in trifluoroethanol at 0°C in the presence of KOH (2 equiv.) and 1.2 equivalents of PIDA (added to the substrate, 146a). Under those reaction conditions, a 29% yield of the dienone 72 was obtained, as compared to a 22% yield under the previous conditions, along with the formation of 7-oxo-ITP (151) in 56% yield. In view of these results, it was decided to investigate the use of other reagents which act in similar fashion to PIDA, in the pursuit of a better way to carry out this transformation. A number of experiments of oxidation of 7a-hydroxy-ITP (146a) with sodium metaperiodate^^'^" (methanol/water or THF/water; room temperature to 65°C), lead tetraacetate"'''"'* (benzene, trifluoroethanol or dioxane; 0°C to room temperature) and manganese (III) tris (acetylacetonate)  (acetonitrile; room temperature to 65°C)  were performed, but they all failed to effect the cyclization to give the dienone 72 (except  one experiment using lead tetraacetate, in which a trace amount (c.a. 1%) of 72 was obtained). Thus, despite the similarity between those reagents and PIDA when effecting several other types of oxidations,"^"'^^ only PIDA was successful in carrying out the required cyclization of 146a to give the (7,8)a-epoxy-dienone 72.  2.4.2  Epoxidation of the (7,8)a-epoxy-dienone, 72  The synthesis of this important epoxy-compound 72, provided us with the opportunity to investigate fiirther the possibility of epoxidation of the dienone system, in order to obtain new di- or possibly tri-epoxide analogs of triptolide (1). Since the epoxy group between C7 and C8 in 72, must show less steric hindrance than the methoxy group of the 8-methoxy-dienones (153), this new diterpene epoxide should allow for easier epoxidation of the ring C of the molecule. Slow addition of a freshly prepared mixture of ^butyl hydroperoxide and TritonB in toluene (1.6 equiv.) to 72 in toluene at 0°C, were found to be the best condition to carry out the epoxidation.  An HPLC method was developed to monitor the reaction  progress, since one of the diepoxides formed had a very similar Rf value to the starting material on TLC, making it difficult to visualize reaction completion by that means. After 4 hrs of reaction time all the starting material was consumed to give after chromatographic purification, the diepoxides 160 and 161 in 29% and 31% yield, respectively.  Several other reaction conditions were tested (different bases, solvents, reaction temperatures and use of hydrogen peroxide), all giving similar or lower yields of the corresponding diepoxides.  89  160  The diepoxide 160 was obtained as an optically active ([aJn : -241.0°) white crystalline solid (m.p 211-213°C). It had a molecular formula of C20H22O5 and its IR spectrum displayed similar absorption to those of the starting material (72), with the band of the unsaturated ketone at 1681 cm''.  The presence of a second epoxide in the  molecule was evidenced from its ' H N M R spectrum, which showed only one olefinic signal at 66.16 (IH, s) and two epoxide protons at 63.80 (IH, s) and 3.06 (IH, s), respectively (Figure 2.8, vide infra). The characteristic signals of the isopropyl group were shifted upfield as compared to those signals in 72, showing the HI 5 proton at 62.55 (IH, sept, J = 6.9 Hz) and the C16, C17 methyl groups at 50.94 and 0.99 (3H each, both d, J = 6.9 Hz), respectively. These changes suggested that the second epoxide was located between CI3 and CI4. This was confirmed by the NOE enhancement of the isopropyl signals and the H7 proton at 63.80, observed when the new epoxide proton signal at 63.06 was irradiated. The determination of the stereochemistry of this new epoxide was difficult to achieve by NOE experiments. Nevertheless, it was noted that the H14 proton in 160 appeared at unusually higher field (63.06) as compared to the chemical shifts of all the epoxide protons of triptolide (1) (63.30 to 3.90). A closer examination of the spatial relationship between the proton at C14 and the neighboring  90  C7-C8 epoxide revealed that, if the C13-C14 epoxide were P oriented, the H14 proton would be directly above the C7-C8 epoxide ring (Figure 2.7-1) and this would make it appear at higher field than expected, due to the shielding effect."^  I  II  Figure 2.7: Comparison between the two possible orientations of the C13-C14 epoxide  In contrast, if the CI 3-C14 epoxy function were a oriented, the HI 4 proton would be situated away from the C7-C8 epoxide ring (Figure 2.7-II), and consequently subject to no shielding or perhaps even being deshielded by the epoxide; this should cause the HI4 signal to appear in the normal region of epoxy protons or at a lower field. Furthermore, from the steric point of view, the a face of the dienone ring is subjected to certain steric blockage due to the a epoxy ring at C7-C8, while the p face is more accessible. This situation favors an attack of the reagent from the p face of the molecule. All of the above, indicated that the C13-C14 epoxide group had a P configuration. The results of COSY and NOE spectra, as well as comparison with other epoxy diterpenes previously reported,''^^'^^ were all in support of the structure assigned.  91  161  The other product of the epoxidation of 72 was the diepoxide 161.  This  compound was isolated as fine white needles (m.p 210-212°C) with the same molecular formula as 160, C20H22O5. It was optically active ([a]p : -27.3°) and its IR spectrum closely resembled that of diepoxide 160, showing two carbonyl bands at 1748 cm'' (lactone C=0) and 1676 cm"' (enone C=0). The ' H N M R spectrum displayed similar signals to those of 160, presenting also one olefinic proton at 86.22 (IH, s) and two epoxy signals at 63.71 (IH, d, J = 1.3 Hz) and 62.96 (IH, s), respectively. However, some significant changes were observed (Figure 2.8), such as: the HI9 quartet at 64.73 had a much wider splitting pattern (2H, Av = 0.24 ppm, J = 16.3 Hz); the H5 signal had disappeared or shifted to higher field and was overlapping with other aliphatic protons; and lastly the C20 methyl group was found at lower field, at 51.24 (3H, s). The COSY spectrum showed cross peaks between the epoxy proton at 63.71 and two other signals contained in the multiplets at 62.28 (2H) and 2.48 (3H) which were correlated to each other. This suggested that the proton at 63.71 was H7 and the other two contained the H6 protons. Another proton partly overlapped in the multiplet at 62.48  92  Compound 160  —r——'—'—•—I—^—I—•—'—r-—I—'—'—:  6.S  S.i  '———i  •i.'i  -j.a  '———r  %,.s  !.o  ^.i  ;.!  1.J  1.0  Compound 161  f  JLJL I  J.-i  !  6 e  '  5 3  '  '  '  !  5.11  j  J.5  J.3  I  iUv_A/UL  '  '  1.5  :  5.6  '  '  •  '  I  2.5  '  •  •—•  i.l  1  1.5  '  L  I.I!  Figure 2.8: Conqjarison of the ^H NMR spectra of diepoxides 160 and 161 (400MHz, in CDCI3)  93  was HI5, showing cross peaks with the HI6 and HI7 at 60.90 and 0.99, respectively. The multiplets at 52.48 also displayed weak cross peaks with the HI9 protons (54.73), therefore suggesting that the signal at 62.48 must also contain the H5 proton; thus accounting for the three protons under that signal. The considerable shift to higher field of H5 as compared to the same signal in the epoxide 160 (52.87), together with the wide splitting of the C19 protons (due to the closeness in space of the HI 9a to the C7-C8 epoxide), strongly suggested that the stereochemistry at C5 had changed. This was also supported by the change in chemical shift of its '^C signal, showing C5 at 637.7, as compared to 532.6 for the C5 signal in 160. On the other hand, irradiation of the H20 methyl group at 51.24 showed NOE enhancement of two partly overlapping signals at 62.48 (H5, H6P), thus confirming the (3 orientation of the H5 proton (Figure 2.9). Irradiation of the H20 signal also enhanced the vinylic proton at 56.22, which was then assigned to H l l . Therefore the second epoxide function was between CI3 and CI4, which was confirmed by the NOE enhancement of the isopropyl group signals and the H7 proton (63.71) when the H14 singlet at 52.96 was irradiated. The orientation of the C13-C14 epoxide apparently remained the same as in 160 (i.e. P), as only a minimum change of chemical shift of the H14 was noted (52.96 as compared with 63.06 in 160) in its ^H NMR spectrum.  The '^C NMR spectrum  supported this assignment, since the signals of C13 at 663.7 and C14 at 559.0 were at practically the same chemical shifts as the corresponding signals in 160, observed at 663.6 and 58.0, respectively.  94  Figure 2.9: Major NOEs observed for compound 161  Compound 161 was apparently formed by isomerization of diepoxide 160 under the reaction conditions, although it could also be formed via partial epimerization of the starting material 72, followed by epoxidation from the most accessible p face of the molecule. Given that the HPLC trace of this reaction showed formation of a larger amount of the A/B trans diepoxide 160 than that of A/B cis 161 since the beginning, and throughout the first two hours of reaction (i.e. 160 apparently formed first); the first theory seems more likely. Although formation of compound 161 could not be prevented despite changes in the reaction conditions (i.e., weaker bases or shorter reaction times produced lower amounts of 161, but lead to incomplete reaction, leaving inseparable mixtures of starting material 72 and diepoxide 160, in addition to compound 161), this provided an unexpected opportunity by making available a new type of structural variation of the 18(4->3) aZ?eo-abietane diterpene skeleton, possessing several oxygenated functionalities, and which effect in the pharmacological activity of this type of epoxidized diterpenes has never before been explored.  95  2.5 Synthesis of the (7,8)P-epoxy series  2.5.1  Synthesis of the (7,8)P-epoxy-dienone, 73  After the successful preparation of the (7,8)a-epoxy-dienone 72, it was thought that perhaps the isomeric (7,8)P-analog, 73 could be synthesized applying similar cyclization conditions with the corresponding 7P-hydroxy-phenol 146b (Scheme 2.17).  73  146 b  151  Scheme 2.17: Proposed retrosynthesis of the (7,8)P-epoxy-dienone, 73.  Previous reports showed that sodium borohydride reduction (in ethanol) of ketophenols structurally related to 151, or to its corresponding  14-hydroxy-isomer,  triptonolide (14), occurred diastereoselectively to give preferentially the corresponding 7p-alcohol (96-98% P: 4-2% a).^^''^ Therefore 7-oxo-isotriptophenolide (151) would be a good starting material to prepare 73. A small quantity of keto phenol 151 was already available, since this compound was obtained as a side-product from the cyclization reaction to obtain 72 and from other oxidation reactions {vide supra). Nevertheless, a  96  two-step sequence from the acetoxy derivative 138 (an intermediate compound in the synthesis of isotriptophenolide, 71; see Scheme 2.1) could provide a simple synthetic route to 151, making it accessible in large quantities, since it would be the main product, rather than a side-product (Scheme 2.18).  OAc  146 b  138  162  151  Scheme 2.18: Synthesis of 7-oxo-isotriptophenolide (151) from 138  Earlier experiments^^ on this route provided low yields (13%) of the hydroxyphenol 146b. An experiment of oxidation of 12-acetoxy-isodehydroabietenolide (138) using chromium trioxide in acetic acid^^ gave the corresponding benzylic ketone 162 in 54% yield.  This compound was then hydrolyzed with a dilute solution of  hydrochloric acid in methanol, to give 7-oxo-isotriptophenolide (151) in 89% yield. Reduction of 151 with sodium borohydride in a mixture of ethanol-methylene chloride diastereoselectively afforded the desired 7P-hydroxy-isotriptophenolide, 146b as a colorless solid (m.p 136-138°C, dec.) in a yield of 98% based on recovered starting material, with only a trace amount of the corresponding 7a-alcohol ( « 1 % ) . This yield was considerably higher than the 26% previously reported for this transformation,^^ and it  Direct oxidation of isotriptophenolide (71) to 151 (Cr03/AcOH or PCC) was ineffective since it provided very poor yields of 151, along with complex mixtures of several over oxidized compounds.  97  was due to changes in the reaction conditions and the isolation procedure.  OH  146b  The diol 146b was optically active ([a] ^ : +94.8°) and exhibited a band in its IR spectrum at 3580 cm"' for the hydroxy groups. Its ' H N M R spectrum showed the H7a proton at 54.95 (IH, br dd, J= 8.6, 8.6 Hz) and the NOE enhancement of the H5 signal at §2.75 (IH, br d, J = 13.8 Hz) observed when the proton at C7 was irradiated confirmed the a orientation of H7. Cyclization of 146b was carried out under similar conditions as in the synthesis of the (7,8)a-epoxy-dienone 72 (previous section) (Scheme 2.19).  The corresponding  (7,8)P-epoxy-dienone 73 and 7-oxo-isotriptophenolide (151) were obtained in 30% and 59% yield respectively; which were practically the same yields as in the cyclization step to obtain 72.  PIDA 'OH  146 b  K2CO3, CF3CH2OH  o:  73  Scheme 2.19: Synthesis of (7,8)P-epoxy-dienone 73 from 146b  151  98  Compound 73 was obtained as optically active ([a] ^ : -360.9°) colorless plates with a molecular formula of C20H22O4. Its IR spectrum displayed the same absorption bands as 72, with the dienone carbonyl absorption at 1655 cm"'. The ' H N M R spectrum also showed a similar pattern to that of the (7,8)a-isomer, 72, except that some signals were shifted and the HI9 protons at 54.68 appeared as a broad singlet rather than the AB quartet observed for the a-epimer (Figure 2.10). The H5 proton was shifted at higher field (52.58, br d, J = 13.4 Hz) and the H20 singlet at 61.17 was at a lower field position than the same signal in 72. These changes in the chemical shifts indicated a change in the stereochemistry (to P) of the epoxide between the C7 and C8 carbon. The rest of the signals appeared at very similar chemical shifts to those of the corresponding protons in compound 72. The H l l and H14 signals were located at 66.42 (IH, s) and 5.99 (IH, d, J = 1.0 Hz), respectively. Irradiation of the HI 1 signal enhanced the Hip proton at 62.10 (IH, br dd, J = 12.9, 5.6 Hz) and the H l a at 51.61 (IH, br ddd, J = 12.9, 12.6, 6.1 Hz). Those two protons were in turn coupled with protons at 62.30 (IH, m) and 2.47 (IH, br d, J = 18.5 Hz), which corresponded to the two H2 protons. Saturation of the H20 resonance at 61.17 (3H, s) caused NOEs to one H2 proton at 62.30 and the H6 proton at 62.13 (IH, br dd, J = 14.5, 13.5 Hz), indicating that both of these protons were p oriented. Irradiation of the H7 signal at 53.84 (IH, d, J = 5.8 Hz) resulted in enhancement of the HI4 proton (65.99) and the other H6 proton at 62.26 (IH, ddd, J = 14.5, 5.8, 5.7 Hz); which hence must be the H6a, therefore confirming the a orientation of H7 proton, and consequently the p stereochemistry of the epoxide group.  99  Compound 72  Li/LV  _A_A  w^  JU •  I  6-J  •  ••  I  •  •  5.S  •  I  i.l  •  •  •  I  4.5  •  •  •  I  '  '  4.a  '  '  I  3.5  '  '  '  '  I  i.B  '  '  '  '  i  J.5  •  '  •  '  I  2.1)  •  1 5  •  ;  1.  PPM  Figure 2.10: Comparison of the ' H NMR spectra of epoxy-dienones 72 and 73 (400 MHz, inCDCls)  100  2.5.2  Epoxidation of the (7,8)P-epoxy-dienone, 73  The epoxy-dienone 73 (40 mg) in benzene was treated with ^butyl hydroperoxide and Triton B for 4hrs, to give 30 mg of a pure mixture of the diepoxide 163 and compound 164 (38:62, respectively by ' H NMR) after purification by column chromatography. Several other epoxidation experiments were performed using either t-h\xty\ hydroperoxide or hydrogen peroxide with a variety of bases (KHCO3, K2CO3, KH, NaOH, KF/AI2O3, lithium-2,6-di-^butyl-4-methyl phenolate, etc.) and solvents (such as THF, acetonitrile, ethanol/water, toluene) under several different reaction conditions. These experiments gave either poor conversion or mixtures of 163 and 164 with similar or lower total yields, but with higher contents of the isomerized starting material, 164. Repeated recrystallization of the mixture with ethanol provided pure samples of compounds 163 and 164.  164 The epoxide 164 was obtained as white crystals (m.p. 189-191 °C, dec.) with the same molecular formula as the starting material, C22H22O4. Its IR spectrum closely resembled that of epoxide 73 with the carbonyl band of the dienone at a slightly higher frequency of 1681 cm"'. The ' H NMR spectrum showed a few important changes with  101  respect to that of 73. The HI9 protons at 54.71 appeared as a widely splitted AB quartet, instead of a singlet (2H, ABq, Av = 0.19 ppm, J= 16.8 Hz); and the characteristic signal of H5 shifted downfield to 52.87 (IH, br d, J - 12.8 Hz) as compared to 52.50 in the starting material. Some changes in the aliphatic protons between 51.60 and 2.50 were also observed. The vinylic protons at 56.52 (IH, s, H l l ) and 6.03 (IH, s, H14) and the H7a signal at 53.80 (IH, d, J = 3.6 Hz) had similar chemical shift to those of the corresponding signals in 73. In the COSY spectrum (Figure 2.11) the H7a proton displayed crossed peaks to a signal at 62.39 (IH, dt, J = 14.8, 3.6 Hz), which in turn was correlated to the H5 proton (52.87) and another proton at 51.89 (IH, dd, J = 14.8, 12.8 Hz). Therefore, from these correlations the signals at 52.39 and 1.89 must be due to the two H6 protons. The H19 signals (54.71) were coupled with the two H2 protons within the multiplets at 52.33 and 2.26, and they were assigned to H2a and H2p respectively, in comparison with the spectrum of 73. Both H2 signals were coupled with each other and with a signal at 61.70 (IH, m). The signal at 51.7 and H2p were coupled with another proton at 61.80 (IH, dd, J= 14.2, 5.8 Hz). Thus, the signals at 51.70 and 1.80 were assigned to the two HI protons. Irradiation of the HI 1 proton (66.52) enhanced the signals of the H20 at 61.17 and the HI proton at 51.80, therefore the latter was assigned to Hip (Figure 2.12). Saturation of the H20 singlet at 61.17 showed strong NOEs to the H l l , H5 (52.87) and the H2p (52.26). The enhancement of the H5 signal indicated that the stereochemistry at this center had changed, to leave the H5 proton P oriented. This also explained the downfield  102  i.a  i.s  i.e  Z.5  t.O  Correlated protons: a) 19/2P; b) 19/2a; c) 19/5;d) 7/6P; e) 5/6a; f) 5/6P; g) 2a/2p; h) 6a/6P; i) la/2a; j) lcx/2P; k) lp2P; 1) la/lp; m) 15/16,17  Figure-2.11: Expanded COSY spectrura of compound 164.'  103  downfield shift of this signal as compared to the H5 in 73, since now this proton was placed in close proximity to the oxygen atom of the C6-C7 P-epoxy fiinction, causing thus a deshielding effect on it. Irradiation of the H5 proton caused enhancement of the H20 signal, as expected, and also showed NOE to the H6 at 52.39, thus assigned to H6p.  Figure 2.12: Major NOEs observed for compound 164  The diepoxide 163 was obtained as a colorless solid (m.p. 198-200 "C) with a molecular formula of C20H22O5. Its IR spectrum was similar to that of the epoxides 164 and 73, showing the appearance of the unsaturated ketone frequency at 1676 cm''.  163  Its H NMR spectrum exhibited some similarities to that of the epoxide 164, but showed only one olefinic proton at 56.26 (IH, s) and two epoxy protons at 53.77 (IH, d, J= l).l Hz) and 3.06 (IH, s), clearly indicative of a new epoxide in ring C. The upfield shift of the H15 signal to 52.53 (IH, sept., J = 6.9 Hz) as compared to the corresponding  104  signal for the epoxides 73 (52.98) and 164 (52.97) suggested that the epoxidation took place at the double bond between C13 and CI4. This was further confirmed by the NOE enhancement of the HI 5 proton and the isopropyl methyl groups at 50.92 and 1.00 (3H each, both d, J = 6.9 Hz) when the new epoxy proton at 63.06 was irradiated, thus assigned to H14. The enhancement of the qther epoxy proton at 53.77 by irradiation of the HI4 signal, confirmed its identity as the H7 proton. The analysis of the COSY spectrum (Figure 2.13) for the proton signal of H7, showed strong cross peaks to a signal at 62.42 (IH, dt, J = 12.2, 3.8 Hz) and a weak correlation to a proton at 51.92 (IH, dd, J = 12.2, 11.4 Hz). Both these protons were correlated with each other and with the H5 signal at 62.62 (IH, dd, J - 11.4, 3.8 Hz). This indicated that the resonances at 51.92 and 2.42 were due to the H6 protons. The AB quartet of the HI9 protons at 54.70 ((2H, br ABq, Av = 0.19 ppm, J = 16.9 Hz), like in the spectrum of the epoxide 164 (and different from the spectrum of 73), was widely splitted suggesting that the stereochemistry at C5 was cis, i.e. the H5 proton was p oriented as in 164. Also, the HI9 signals showed cross peaks to the two H2 protons at 52.25 (IH, m) and 2.37 (IH, m). The two remaining unassigned protons in the spectrum of 163, appearing at 61.66 (IH, m) and 1.79 (IH, m) had cross peaks to each other and to the H2 signals; and they corresponded to the two HI protons. Irradiation of the C20 angular methyl group at 51.12 showed significant enhancement of the H l l vinylic proton (56.26) and the H5 signal at 62.62, therefore confirming the p orientation of the H5 proton (similarly, irradiation of H5 enhanced the  105,  4.S  1.0  OOlf  Correlated protons : a) 19/2a; b) 19/2P; c) 116a; d) 7/6p; e) 19ayi9P; f) 15/16,17; g) 5/6a; h) 5/6p; i) 6a/6p; j) ip/2a; k) ip/2p; 1) la/2a; m) lciy2p; n) la/lp  Figure 2.13: Expanded COSY spectrum of compound 163.  106  H20, as expected) (Figure 2.14). The H2 proton at 62.25 was also enhanced, and was therefore assigned to H2p. Saturation of the H7a resonance at 53.77 increased the intensity of the HI 4 signal (53.06) and the two H6 protons at 51.92 and 2.42 respectively. Irradiation of the signal at 51.92 caused NOEs to the geminal H6 and to the H7a (53.77) but it did not increase the signal of H5p, therefore the signal at 61.92 must be due to H6a and consequently the signal at 52.42 was H6p. The orientation of the epoxide between CI3 and C14 was difficult to determine by the results of the NOE difference experiments. However, the HI 4 proton had a very similar chemical shift (53.06) to that of the corresponding signal in the formerly obtained diepoxides 160 and 161 (53.06 and 2.96, respectively). This unusually high chemical shift observed in these compounds (as compared to all the epoxy protons of Tl(l), which appear between 53.30 and 3.90) was thought as a result of a shielding effect of the H14 protons by the epoxy function between C7 and C8 (see Section 2.5.1). This shielding could only take place if the proton at C14 was located directly above the C7, C8 epoxy ring, requiring an anti orientation of the two neighboring epoxides on ring B and C of the molecule {vide supra, Figure 2.7-1). For the diepoxide 163, the high field chemical shift of the proton HI4 closely resembles that found in the related compounds 160 and 161, which indicated that an equivalent situation as mentioned above must be taking place in this case. In other words, the two epoxy function in 163 must be anti to each other, placing the new epoxide  107 Irradiation at H7  yi,iiH^i.^lii^^<ij^  kradiationatH14  ^«ii<i^w»»>ix*miii,i(»»n>i» <i*»niini* ••w^rii-iUfMi'.''I  i^V'i"»***  InsKiadQaatHS  H16,17  6.?  6.C  S.5  S»  i.7  1.B  3.5  J.8  Figure 2.14:" NOE" difference spectra of compound 163'  2.S  i.  • 5  it  108  between CI3 and C14 a oriented, thus leaving the H14 in the shielding zone of the C7C8 p epoxy ring (Figure 2.15-1). Furthermore, if the C13-C14 epoxide was (3 oriented this would lead to a more unstable situation in the molecule, given the stronger steric and electronic repulsions between the two neighboring epoxides in this conformation (Figure 2.15-11).  O  CH.  ^ H  H  II Figure 2.15: Conformation of the CI 3-C14 epoxy group of 163  The epoxidation of the (7,8)P-epoxy-dienone 73 yielded only one diepoxide, 163, in which the AJB-trans ring junction was isomerized to a cis configuration. But, unlike the epoxidation of the isomeric (7,8)a-epoxy-dienone 72, no diepoxide having the original AJB-trans junction was formed.  During the HPLC monitoring of several  epoxidation experiments, it was noted that isomerization of the starting material 73 was the first reaction to take place, starting to occur after 20-30 minutes of reaction, and continuing to produce the A/B cis diastereomer 164 for several hours (typically 2-4 hrs, depending upon the reaction conditions). During the progi-ess of the reaction, the rate of production of the diepoxide 163 was much slower than the isomerization of epoxide 73 to 164 (Scheme 2.20).  109  164-1  Scheme 2.20: The mechanism of formation of compounds 163 and 164.  Complete consumption of the starting material was achieved in 4hrs to give a mixture of products 163 and 164 (4:6, respectively by ' H NMR spectroscopy). Excess of epoxidating agent (^BuOOH) or longer reaction times did not increase significantly the amount of diepoxide 163 and it resulted in a detriment of the total yield of the crude mixture. This could probably be caused by side reactions (hydrolysis of the lactone moiety of both compounds 163 and 164, for example) due to the basic reaction conditions. Therefore, to minimize losses in the yield the reaction mixture was quenched as soon as all the starting material was consumed. On the basis of these results, epoxidation of (7,8)P-epoxide 73 seems to follow a  no different reaction mechanism, i.e., a different sequence of events to that of the epoxidation of its epimer, the (7,8)a-epoxy-dienone 72. During the epoxidation of the aisomer 72, the epoxidation of the 13,14-double bond took place at a faster rate than the epimerization at the C5 position {vide supra. Section 2.4.2). That probably was due to the difficult accessibility of the base to the H5a in that compound, because of the steric blockage by the nearby C7-C8 a-epoxy function; as a result, diepoxides of both the A/B rings ^ra77^-fused compound and the epimeric A/B cw-fused isomer were obtained. In the case of the epoxidation reaction of 73, the H5a is much more accessible for the base since this time the epoxide between C7 and C8 is P-oriented and therefore it does not block the access to the proton at C5. Consequently, the epimerization of C5 is much easier and it took place faster than the epoxidation of the dienone in ring C. Attempts to prevent epimerization at this center by using weaker bases, such as KHCO3 or K2CO3 lead to incomplete and slower reaction, and still significant epimerization occurred. The use of large, sterically hindered bases, such as the lithium salt of 2,6-di-rbutyl-4-methyl-phenol, in the presence of ^butyl hydroperoxide did not gave any reaction, and the starting material was recovered unchanged after 12hrs (probably the base was too hindered even to abstract the proton of the ^butyl hydroperoxide). The epoxidation of the dienone took place from the less hindered a-face of the molecule since the P-face of the dienone ring was inaccessible to the /-butyl hydroperoxide anion due to the presence of the C20 methyl group and the C7-C8 epoxy function, both p-oriented, blocking the top face of the molecule. Like in the case of the epoxidation of 72, the double bond between C13 and C14 was selectively epoxidized.  Ill  This result was probably due to two factors; firstly, examination of the molecular model of epoxide 164 showed that the C14 position was slightly more accessible than the C9 position. But, perhaps more importantly was the unstability of the possible reaction intermediate that would be formed by attack of the ^butyl peroxide anion at C9 (compound 164-11, Scheme 2.21), caused by strong 1,3-steric interaction between the bulky ^butyl group and the chain fragments of ring A at CI and C5. Equivalent steric interactions are much less severe in the reaction intermediate 164-1 (Scheme 2.20). The above factors lead to the regioselective epoxidation of the double bond between C13 and CI 4, in a diastereoselective fashion by the less hindered a-face of the molecule.  t-BuOO  0  ~ \  164  164-11  Scheme 2.21: Steric interactions in the possible reaction intermediate 164-11  2.6  Synthesis of the 8-hydroxy series  2.6.1  Synthesis of 8-hydroxy-dienones  In the (7,8)-epoxy-dienone series further epoxidation at the C13-C14 position to  112  give the corresponding diepoxides could in fact be accomplished, while the diastereomeric mixture of 8-methoxy-dienones (153) did not suffer epoxidation even under harsh conditions. This indicated that the steric hindrance of the substituents close to the double bonds of the dienone system played a determinant factor to the successful introduction of a second epoxide function. Two substrates initially considered as good candidates for the introduction of an epoxide function to the ring C of the molecule were the 8-hydroxy-dienones 157 and 158, taking into account the less steric effect imposed on the CI3-CI4 double bond by the hydroxyl functionality.  157  158  Due to the initial unsuccessful attempts to obtain these compounds (157 and 158), and because a route to the related substrates, the epoxy-dienones 72 and 73 was developed, the preparation of the 8-hydroxy derivatives was not pursued further (see Section 2.4.1). Nevertheless, the positive results obtained in the epoxidation of 72 and 73 suggested strongly that epoxidation of the dienone ring of 157 and 158 could be achievable, since the hydroxyl group at C8 would exert a similar steric hindrance as an epoxy group between C7 and C8. Thus, the efforts to prepare these hydroxy-dienones were resumed. A different approach, based on the same strategy used to prepare the methoxy  113  derivatives 153 (vide supra, Scheme 2.9), was followed. The method utilized to obtain 153 was based on the fact that the oxidation of/)ara-substituded phenols with PIDA in the presence of nucleophiles (generally used as the solvent in the reaction) gives rise to dienones in which the nucleophile attaches to the para position of the ring (vide supra, Scheme 2.8). Therefore, if the nucleophile was water for this reaction, the desired 8hydroxy-dienones (157 and 158) should be obtained, along with 7a-hydroxyisotriptophenolide, 146a (which could be used to prepare the (7,8)a-epoxy-dienone, 72). To prove this concept, isotriptophenolide (71) was treated with PIDA (1.1 equiv.) in a mixture of acetonitrile-water at room temperature.  After chromatographic  purification of the crude mixture, 8a-hydroxy-dienone (157, 9%), 8p-hydroxy-dienone (158, 18%), 7a-hydroxy-isotriptophenolide (146a, 60%), and 7-oxo-isotriptophenolide (151, 9%) were obtained (Scheme 2.22).  +  71  157=a-OH  151  146 a  158=(3-OH  Scheme 2.22: Synthesis of the 8-hydroxy-dienones 157 and 158 from ITP (71)  Although the desired 8-hydroxy-dienones were obtained, the predominant transformation was the diastereospecific hydroxylation of isotriptophenolide (71) at the C7 position to give 146a. The extent in which this diol was formed with respect to  114  compounds 157 and 158 was rather unexpected considering that the corresponding methoxy derivatives, the 7a-methoxy-ITP (150a) and the 8-methoxy-dienones 153 (obtained when ITP, 71, was treated with PIDA in methanol instead of PIDA/water), were formed in 26%) and 45%) respectively {vide supra. Scheme 2.9). Therefore a similar ratio was anticipated for the corresponding hydroxylated derivatives.  These results  suggested that, similar to the mechanism of oxidation of 7a-hydroxy-ITP (146a) with PIDA {vide supra. Scheme 2.16), the carbocation reaction intermediate 71-1 is in fact highly unstable and is quickly "stabilized" by losing a benzylic proton, to form the corresponding quinone methide, 142 (Scheme 2.23). This intermediate then suffers a nucleophilic attack by water to give the diol 146a. Therefore, if the carbocation intermediate 71-1 has a shorter mean life than the quinone methide 142, this would lessen the opportunity of 71-1 to react with the nucleophile (water) to form the 8-hydroxy-dienones (157 and 158), consequently increasing the opportunities for a 1,6-addition of water to the quinone methide 142 to give 7a-hydroxy-isotriptophenolide (146a). In contrast, during the oxidation of isotriptophenolide (71) with PIDA in methanol {vide supra. Section 2.3), the carbocation 71-1 most likely was stabilized by solvatation by the solvent, thus extending the mean life time of this reaction intermediate. This favored the formation of the diastereomeric 8-methoxy-dienones (153) over the 7amethoxy-isotriptophenolide (150a) because of two reasons; firstly, the solvent (methanol) would have more time to attack 71-1, and secondly the solvating methanol molecules would be in very close proximity to the reactive site of the molecule (the carbocation at C8), thus leading to a faster formation of the 8-methoxylated product (153).  115  On the other hand, these observations were also in support of the related mechanism proposed for the formation of compounds 72 and 151 (vide supra. Scheme 2.16).  Furthermore, this indicated that the strain generated by the formation of the  oxirane ring to give epoxide 72 was not a significant factor contributing to its low yield of formation. Instead, the facile formation of the quinone methide by elimination of the H7 proton (path B) was the dominant factor in that reaction.  146 a  157 = a-OH 158 = P-OH  Scheme 2.23: The mechanism of oxidation of isotriptophenolide (71) with PIDA/water  157  Compound 157 was obtained as an optically active ([a]o : -164.8°) white solid (m.p. 196-198°C), with a strong absorption band in the UV at 229.2 nm (log s: 4.23). It had a molecular formula of C21H24O4 and its IR spectrum showed the presence of an hydroxyl band at 3397 cm"', the dienone carbonyl group at 1670 cm"' and a lactone carbonyl at 1742 cm"'. The ' H N M R spectrum of 157 closely resembled that of the 8a-methoxy-dienone, 153a. The ring C vinyHc protons appeared as singlets at 56.38 (H14) and 6.17 (Hll), as compared to 66.39 and 6.12 respectively for 153a. The chemical shift of the H19 protons was exactly the same as the corresponding 8a-methoxy analogue, appearing at 54.71 as a widely splitted AB quartet (2H, ABq, Av = 0.09 ppm, J = 17.2 Hz). The characteristic signal of H5a shifted downfield to 53.96 (IH, br s) as compared to 53.72 in 153a, due to the stronger deshielding effect of the 8a-hydroxy group as compared to a methoxy group. This a orientation of the H5 proton was further confirmed by NOE experiments. A singlet at 51.85 (IH, D2O exchangeable) was indicative of the C8 hydroxyl group. The chemical shift of the other aliphatic protons remained practically unchanged. Irradiation of the H14 at 56.38 enhanced the signals of the isopropyl group at 52.89, 1.03, 1.06 (H15, H16 and H17, respectively) and the H7a proton at 52.18.  117  Saturation of the H5a proton (53.96) caused NOEs to the H l a at 51.77, H6a at 61.63, and H7a at 52.18. The rest of the NOE and the COSY results were in good agreement with the structure assigned for compound 157.  158  The hydroxy-dienone 158 was a white crystalline solid (106-108°C) with the same molecular formula as that of compound 157. It was optically active ([a] 24 179.2°) and its UV and IR spectra revealed the same bands as the a-hydroxy isomer. Its ' H N M R spectrum showed some important changes in the chemical shift of several signals. The C20 methyl group shifted to lower field, appearing at 61.29 (3H, s) as compared to 51.03 in the spectrum of 157, because of the closeness in space to the deshielding C8 hydroxyl group (1,3-diaxial relationship). On the other hand, the H5 proton had a drastic upfield shift to 52.35 in comparison to 63.96 for the same signal in the 8a-isomer. This reflected the change in the orientation of the hydroxyl group to Pposition, since in the 8a-hydroxy-isomer the boat conformation of ring B placed the H5 and the hydroxyl group very close to each other (1,4-diaxial, see conformations of the corresponding 8-methoxy analogues in Figure 2.2), which strongly deshielded the proton at C5. This effect is lost when the hydroxyl group is in a p orientation, and consequently the H5 proton appears at much higher field. A similar effect was observed for the signal  118  of the H19 protons which appeared as a broad singlet at 64.71, while the same signal for the isomer 157 was a widely splitted AB quartet (54.71, Av = 0.09 ppm, J = 17.2 Hz). The assignment of the rest of the signals in the ' H NMR spectrum of compound 158 was accomplished on the basis of the NOE enhancements and COSY correlations.  A  comprehensive comparison of those signals for the diastereomers 157 and 158 is presented in Table 2.1.  Table 2.1  ' H NMR data of hydroxy-dienones 157 and 158 (400 MHz, in CDCI3; 6 in ppm, J in Hz in parentheses) Proton  157  158  la  1.77, m  1.64, m  IP  2.09, m  2.06, m  2a  2.43, d (15.4)  2.47, d (16)  2P  2.33, m  2.35, m  3.96, brs  2.35 brd (11.4)  6a  1.63, m  1.64, m  6P  2.09, m  2.06, m  7a  2.18, ddd (13.0, 10.8,2.2)  2.21, dt (13.6, 2.5)  7P  1.77, m  1.53, td (13.6, 4.5)  11  6.17, s  6.09, s  14  6.38, s  6.41, s  15  2.89, sept (6.8)  2.90, sept (6.9)  16,17  1.03, 1.06, d (6.8)  1.04, 1.06, d (6.9)  19  4.71, ABq (17.2)  4.71, brs  20  1.03,s  1.29, s  5  119  Having the hydroxy-dienones 157 and 158 available, the next step in this study was to investigate the epoxidation of the dienone moiety of these compounds.  2.6.2  Epoxidation of the 8a-hydroxy-dienone, 157  Given the structural resemblance between the 8a-hydroxy-dienone (157) and the (7,8)a-epoxy-dienone (72), similar epoxidation conditions to those used on the latter were utilized to epoxidize compound 157. Thus, the reaction was performed by addition of a freshly made solution of /^-butyl hydroperoxide and Triton-B in toluene to a cold solution of dienone 157 also in toluene. The solution was then brought to room temperature and stirred for 1.5 hrs. Workup and purification by preparative TLC gave the epoxides 165 (39%)), 166 (11%) and 167 (14%), with a combined total yield of 64% (Scheme 2.24).  t-BuOOH triton-B / toluene  157  165  166  Scheme 2.24: The epoxidation of hydroxy-dienone 157  167  120  The main product, epoxide 165, was obtained as an optically active ([a] ^ : + 219.5") white solid (mp 239-24rC) with a molecular formula of C20H24O5. Its IR spectrum showed an hydroxyl band at 3400 cm'', the lactone carbonyl at 1752 cm"' and a carbonyl of enone at 1673 cm"'. The UV spectrum of 165 showed two absorption bands at 216.4 (log s 4.17) and 239.5 nm (log s 4.02), corresponding to the unsaturated lactone and the enone, respectively. o  A first inspection of the H NMR spectrum revealed that a monoepoxidation had taken place since only one olefinic proton at 56.00 (IH, s) was observed and one epoxide proton appeared at 53.31 (IH, s). The HI9 signal at 64.71 had a wider splitting pattern (br ABq, Av = 0.24 ppm, J = 16.9 Hz); and the H5 proton, appearing at 53.96 in the starting material (157), was not found at this low field, and was probably shifted to higher field and was overlapped with the other signals of the aliphatic protons between 62.8 and 1.4.  The signals of the isopropyl group also shifted upfield as compared with the  corresponding signals for 157. The H15 proton appeared at 62.53 (IH, sept, J - 7.0 Hz) and the C16 and C17 methyl groups, at 60.88 and 0.97 (3H each, both d, J = 7.0 Hz), respectively. This suggested that the epoxide function was between CI3 and CI4, which was later confirmed by the NOE enhancement of all the isopropyl signals and the H7 proton (52.11) when the epoxy proton at 63.31 was irradiated.  121  The COSY spectrum of 165 (Figure 2.16) showed cross peaks between the H19 protons (54.71) and a signal at 52.39 (IH, dd, J = 18.0, 6.1 Hz) and a proton in the multiplet at 62.11, which must correspond to the two H2 protons. These signals were coupled with each other and had cross peaks to the signals at 62.79 (IH, ddd, J = 12.8, 12.6, 6.1 Hz) and 1.39 (IH, dd, J = 13.8, 6.1 Hz), assigned to the two HI protons. The remaining protons of the molecule were assigned based on the '^C APT (Attached Proton Test) and HMQC spectra, since the coupling patterns in the COSY spectrum gave insufficient information to make a clear identification. In the group of signals centered at 62.11 there were four protons, two of which were already assigned as one H7 and one H2. The HMQC spectrum correlated one of the protons in that group to a proton at 61.62 (IH, m) (both attached to a carbon at 635.9), and another proton in the same group to a signal at 61.84 (attached to a carbon at 6 24.6). It also showed the correlation between the two H2 already assigned, and finally one cross peak to a carbon at 643.9. The latter was attached to only one proton (also confirmed by the negative peak of this carbon in the APT spectrum), therefore this must be the H5 proton. Consequently, the two pairs of signals mentioned above corresponded to the two H6 and the two H7 protons, although a precise assignment was not possible at this stage of the analysis. Turning to the COSY spectrum, the last proton signal observed was a very fine doublet at 62.45 (IH, J = 2.1 Hz, D2O exchangeable), corresponding to the C8-hydroxyl, which showed a cross peak to a signal at 51.62 (long-range coupling). Therefore the pair of signals at 61.62 and 2.11 must be the two H7, and the H6 protons were at 61.84 and  122  .6  l.D  2.e 2.2  :.' :.t  :.s  3.2 i.i  Correlated protons: a) 15/16,17; b) la/lp; c) la/2p; d) la/2a; e) 2a/2P; f) lp/2P; g) 7p/7a; 7p/6p; h) 6(x/6P; 6a/7a; 6a/5; i) 0H/7p; j) 6a/7p  Figure 2.16: Expanded COSY spectrum of conqiound 165  123  2.11 respectively. The unexpected appearance of the hydroxyl group as a sharp doublet (given that the C8 holds no protons), with a small coupling constant, together with the observation that the chemical shift of this signal had very little dependency on the variation of the concentration of the NMR sample, suggested some degree of "rigidity" of this group (i.e., restricted rotation of the hydroxyl along the C-0 bond). Examination of the molecular model clearly indicated that if the epoxy group between CI 3 and C14 was a-oriented, the 8a-hydroxyl would be close enough to form an intramolecular hydrogen bond to the epoxide, therefore imposing a "rigid" position to the hydroxyl group (Figure 2.17).  Furthermore, this would explain why the hydroxyl presented a long-range  coupling to an H7 and even to which one of these H7 protons was coupled. From the model it could be seen that the hydroxyl group would be "locked" in such an orientation that coincidentally adopts a "W" arrangement with H7p (51.62), which favors an effective a-bond orbital overlap, thus allowing a four-bond coupling between these two protons.'•^° That coupling and the rigidity of the hydroxyl group would not be possible if the 13,14-epoxide was p-oriented, therefore indicating that this epoxide was a.  o "W coupling -^  \ .  •^  Figure 2.17: Stereochemical view of compound 165  ^-—• hydrogen bond  124  On the other hand, the dramatic shift to higher field of the H5 proton to 52.11 as compared to 53.96 in the starting material 157, strongly suggested that the stereochemistry at C5 had changed. The H5 (a) proton appeared at such a low field in 157 due to the deshielding effect of the close in space 8a-hydroxyl group, which should not change much if an epoxy group was introduced at C13-C14. Nevertheless, a large change in chemical shift of this signal was indeed observed, therefore indicating a very different magnetic environment around H5, one in which this proton was no longer close to the hydroxyl group. Thus, the H5 proton must be |3-oriented (Figure 2.17). This was also supported by the change in the '^C chemical shift of C5, appearing at 643.9 compared with 535.1 in dienone 157, and by the result of the irradiation of H l l , which caused NOE enhancement of the C20 methyl only, with no enhancement of the Hip, as normally observed when the ring junction is A/B trans. All this evidence therefore indicated that the stereochemistry at the C5 position was changed and H5 was p.  Another product of the epoxidation reaction, epoxy-dienone 166, was a colorless solid (80-82°C) with the same molecular formula as compound 165. It was optically active ([a] ^ : +230.0°) and its UV spectrum displayed only one absorption band at 244 nm (log s 3.94). Its IR spectrum showed a carbonyl band of unsaturated ketone at 1673  125  cm"' and the carbonyl of lactone at 1784 cm'' (a higher frequency than the usual 1752 cm"' for the lactone carbonyl) and no hydroxyl bands were observed. The ' H N M R spectrum of 166 showed one olefinic proton at 55.77 (IH, s) and one epoxy proton at 53.34 (IH, s). The HI9 signal was again a widely split ABq (Av = 0.22 ppm, J = 9.2 Hz) at 54.33. The isopropyl group signals were located at 52.53 (IH, sept, J = 7 Hz, HI5); 0.85 and 0.93 (3H each, both d, J = 7.0 Hz, HI6, HI7); and the C20 methyl was at 51.23 (3H, s). All the rest of the protons appeared as a cluster between 61.55 and 2.15, making the signal assignment difficult. In the COSY spectrum (Figure 2.18) the singlet at 55.77 had a weak cross peak to the C20 methyl signal (homoallylic coupling), suggesting that this was due to H l l .  This was confirmed by the NOE  enhancement of the H20 protons when the HI 1 signal was irradiated. Consequently, the singlet at 53.34 was the H14 thus indicating that the epoxide was between CI3 and C14. Surprisingly, the HI9 protons did not show any cross peaks in the COSY spectrum, unlike the starting material 157 and all the other similar derivatives prepared so far. This observation, and the evidence from the IR spectrum showing the carbonyl of the lactone shifted some 32 cm"' to higher frequency than its normal position, indicated that the lactone moiety had suffered a change. Examination of the HMQC (Figure 2.19) and '^C spectra provided more information about these changes and helped to identify the signals of the protons in rings A and B of the molecule in the ' H NMR spectrum. The carbonyl carbons C12 and CI8 were easily recognized by their characteristic low field chemical shifts at 6193 and 174.9, respectively. The C l l , C14 to C18 and C20 signals were correlated with their  126  —A  .A  AAAJI/ i.c  0  1.5  la  Correlated protons: a) 19a/19p; b) 11/20; c) 15/16,17; d) 2/3p; e) 7a/7p; f) 7ay6p; g) 6p/5p; h) 6P/7P; i) 6a/6p; j) 2/lp, lo/lp; k) 5p/6a  Figiire 2.18: Expanded COSY spectrum of conqjound 166  127  corresponding proton signals identified above from the ' H N M R spectrum. The vinylic C9 signal appeared at 5165.4, the same chemical shift as in the starting material (5165.6), but the characteristic signals about 6123 and 163 corresponding to the olefmic carbons of the lactone ring (C3 and C4 respectively) disappeared from the spectrum. This indicated that the double bond between those carbons was no longer present, therefore leaving a saturated lactone moiety instead. This was consistent with the shift to higher frequency observed for the carbonyl group of the lactone in the IR spectrum (vide supra) and with the appearance of only one absorption band in the UV spectrum (corresponding to the enone in ring C) of 166. The APT spectrum showed the presence of four more quaternary carbons at 639.5, 62.8, 70.4 and 79.7. From these, the signals appearing at 539.5 and 70.4 were assigned to CIO and C8, respectively in comparison with the corresponding signals in the starting material at 638.5 and 68.3; and also due to the fact that no significant change apparently took place at these carbon centers. Another quaternary signal not yet assigned was the CI 3 carbon. Observation of the chemical shift of this carbon in the C spectra of epoxy derivatives with related structures, such as compound 165 (565.8), and diepoxides 160 (663.6) and 161 (663.7) allowed assignment of the signal at 662.8 of 166 to its C13. Therefore, the remaining quaternary signal (579.7) must correspond to either C3 or C4, the other one being a methine carbon, in accordance with its APT spectrum. Since the molecular weight of 166 indicated a gain of 16 units with respect to the starting material (the oxygen atom of the epoxide at C13-C14), and considering the above evidence indicating the absence of both a double bond in the lactone ring and a hydroxyl group in the molecule; this strongly suggested that an intramolecular attack of the  128  JLUI\J S"  g^  1  •  f  d 1  O"  50  c  2 ^ u >»  b -  100  a u ... ppm  I  I  . ppm  1 1 1 1 1 1 I 1 1 1 1 1 1 1 1 1 1  ^  I  Correlations: a) Cll/Hll; b) C19/H19; c) C14/H14; d) C3/H3; e) C7/H7; f) C15/H15; g) C6/H6; h) C2/H2; i) C16/H16; j) C17/H17; k) C20/H20; I) C5/H5; m) Cl/Hl  Figure 2.19: Expanded HMQC spectrum of compound 166  129  original 8a-hydroxyl group to the C4 vinylic carbon had taken place (intramolecular 1,4addition), forming a new ether bond between these two centers and leaving the C3 carbon as a methine {vide infra. Scheme 2.25). The latter could only be accomplished if the A/B ring junction was cis, since only in this configuration the lactone ring could be in close proximity to the hydroxyl group so as to allow such a cyclization to occur. In the HMQC spectrum there were two methine signals left, at 547.9 and 40.7 with their corresponding protons at 52.15 (IH, dd, J = 11.8, 4.0 Hz) and 61.67 (IH, t, J = 2.9 Hz). These signals were due to C3 and C5, although at this stage it was not known which signal corresponded to which carbon. On the other hand, the signals for the C6 and C7 were easily assigned at 616.4 and 29.8 respectively, since the change in chemical shift with respect to the same signals in the starting material, 166 (appearing at 617.4 and 28.6, respectively) was minimal. The only two signals left to identify in the  C spectrum were located at 640.6 and 17.7, and  corresponded to CI and C2 respectively, in comparison with the spectrum of similar compounds. The assignments of CI, C2, C6 and C7 were confirmed by the correlations of their corresponding attached protons observed in the COSY spectrum (vide supra. Figure 2.18). This spectrum also showed cross peaks between the proton at 61.67 and the two H6 signals at 51.78 and 1.94, thus indicating that this proton was H5. Therefore, the other methine proton, appearing at 52.15 must be H3, which was confirmed by its cross peaks with the H2 protons. Thus the corresponding C5 carbon was at 540.7 and C3 signal at 547.9. This completed the assignment of all the carbon and proton signals of compound 166.  130  The change in chemical shift of C5 with respect to the same signal in compound 157 at 535.1 was consistent with a change of stereochemistry at this center. Irradiation of the C20 methyl group at 51.23 caused NOEs to the Hll (55.77), H6p (51.94), Hip (51.56), and the H5 (51.67). The latter confirmed the P orientation of the H5. Saturation of the H5P proton enhanced the signals of the H20, the Hip, H19p (54.20), the two H6 (51.78 and 1.94), and the H3 signal at 52.15, thus revealing the P stereochemistry of the H3 proton. Irradiation of the epoxy proton H14 (53.34) caused NOEs to the isopropyl group signals and to the H7a proton at 52.09 only, but not to the H7p. This observation indicated that the orientation of the HI4 proton was a, and consequently the C13-C14 epoxide had a P stereochemistry. By examination of the molecular model it could be seen that if the epoxy function was a-oriented, given the very rigid structure of the molecule, the H14 proton would be at almost the same distance to both the H7 protons (Figure 2.20-1).  Therefore, irradiation of the H14 signal should result in equal  enhancement of the two H7 protons. On the other hand, if the epoxide ring was p the HI4 proton would be parallel and in close proximity to the H7a, and further apart from H7p (Figure 2.20-II). Therefore, it would be expected an NOE enhancement of H7a, but not of H7p, which is precisely what it was observed. All these arguments were in complete agreement with the structure assignment of 166.  131  I  II  Figure 2.20: The orientation of the C13-C14 epoxide ring in compound 166  Compound 167 was obtained as an optically active ([a]p : + 273.6°) colorless solid (mp 168-170°C). It had the same molecular formula as 166 (C20H24O5), and showed also practically the same absorption bands in its IR (1780 and 1677 cm"' for a saturated ketone and an enone, respectively) and UV spectra (243 nm, log s 3.95). Its ' H NMR spectrum was very similar to that of epoxide 166, showing only a few changes in the shape of the group of signals of the aliphatic protons between 51.70 and 1.82; and the appearance of a new proton signal at 52.66 (IH, br d, J = 7.1 Hz). The rest of the signals showed minimal change of chemical shift.  167  All of the above suggested that this compound had a very similar structure to epoxide 166. The COSY spectrum of 167 (Figure 2.21) also displayed a similar cross-  132.  JA_JUJU  J-_ 1.3  1.3  2. a  2.S  3.3  3.5  Correlated protons: a) 3o(/2a; b) 3a/2P; c) 15/16,17; d) ip/2p, 7a/7P, 7p/6, 5p/6; e) 2a/2p, la/2P; f) 6a/6p, 6/7a; g) Ip/lcx, lp/2a; h) \aJ2a; i) 19a/19p  Figure 2.21: Expanded COSY spectrun: of compound 167  133  peak pattern to that compound. The HI 9 protons were not correlated to any other signal, and the vinylic proton at 55.82 (IH, s) showed cross peaks to the singlet of the C20 methyl group at 51.21, suggesting that h was H l l . This was confirmed later by the enhancement of the H20 protons and the HI a (IH, dd, J = 11.9, 3.6 Hz) at 61.70, consequently placing the epoxide group at the C13-C14 position. The new doublet at 52.66 was correlated to a signal at 51.77 (IH, m), and all the other cross peaks were in a cluster between 61.50 and 2.10. The H5 signal was located by irradiation of the H20 protons, which caused NOEs to the H l l (55.82), H6p (62.02) and H5 at 51.52; therefore confirming the p-orientation of the H5 proton. Irradiation of the unidentified doublet at 52.66 enhanced the HI9a doublet at 63.93 and the H2 signals at 51.77 and 62.02 respectively, thus indicating that this signal was H3, and it was in an a orientation. The change in stereochemistry of this proton was supported by its shift to lower field (52.66) as compared to the H3p signal at 62.15 in the spectrum of 166, caused by its closeness to the oxygen atom linking C4 and 1T  C8 due to its a orientation. A change in the corresponding C3 signal in the C spectrum, at 45.6 as compared with 47.9 for 166 was observed as well. In order to assign the rest of the protons, a similar sequence utilizing the information from the  C and HMQC spectra  as described above for compound 166 was followed. All the assignments were confirmed by their correlations in the COSY and HMBC spectra and by their consistent NOE results (Figure 2.22). A possible rationale to the interesting formation of epoxides 165-167 from the epoxidation of 8a-hydroxy-dienone (157) involved an initial epimerization at C5 to give  134  the A/B-c/^ fused intermediate 168 (Scheme 2.25). The a-hydroxyl function of this isomerized compound could then have formed an hydrogen bond interaction with the ^butoxide anion, thus controlling the direction of the epoxidation from the a face of ring C to give the major product of this reaction, compound 165 (path A). The formation of an hydrogen bond in this product, this time between the hydroxyl group and the neighboring C13-C14 epoxide ring (as evidenced by its ' H NMR spectrum, vide supra) provided additional "stabilization" of this compound.  Figure 2.22: Major NOEs observed for compound 167  Alternatively, the alcohol group of the intermediate compound could suffer deprotonation by a base in the reaction mixture to give the alkoxide 169 (path B). Due to the conjunction of rings A/B, the lactone moiety was placed in close proximity to the C8 alkoxide, which apparently favored an intramolecular Michael-type addition of this anion to the unsaturated lactone, thus forming an ether bond between C8 and C4. Subsequent protonation of the enolate from either the a or p face of the lactone ring would yield compounds 170 and 171 (in an approximately 1:1 ratio), respectively.  135  168  157  165  path B; deprotonation of OH  169  171  C13-C14 epoxidation (from top face)  C13-C14 epoxidation (from top face)  166  167  Scheme 2.25: A feasible mechanism for the formation of compounds 165, 166 and 167  Finally, the formation of this new ether bridge gave a concave shape to the a face of the molecule, increasing therefore the steric blockage of this side and consequently  136  making difficult the approach of the bulky peroxide anion from the a side. In contrast, the p face (the convex side of the molecule) was more accessible, allowing a better approach of the peroxide anion from the p face of the dienone ring, resulting thus in formation of a P-epoxide at the C13-C14 position. On the other hand, the utilization of weaker bases for the epoxidation reaction (KHCO3 or K2CO3 / H2O2) in order to prevent epimerization at C5, gave much lower yields of epoxidation even with extended reaction times (up to 48 hrs). Lower recovery yields and isomerization of H5 was still observed in the latter experiments.  2.6.3  Epoxidation of the SP-hydroxy-dienone, 158  The hydroxy-dienone 158 was epoxidized using the same reaction conditions as described above for the dienone 157.  The reaction was completed in 1.5 hrs and  purification of the crude mixture by preparative TLC yielded 41% of a diastereomeric mixture of the epoxides 172 and 173 (85:15 respectively by ' H N M R , Scheme 2.26).  t-BuOOH triton-B / toluene  172  Scheme 2.26: The epoxidation of hydroxy-dienone 158  173  137  This diastereomeric mixture, appearing as a single spot on TLC (s.s. hexanesEtOAc 6:4, CHiCb-acetone 96:4 or CH2Cl2-isopropyl ether 1:1), was inseparable by chromatographic methods, but pure diastereomer 172 was obtained by repeated recrystallization with ethyl acetate-hexanes mixtures.  The fact that some of the trans  isomer was obtained in this reaction probably reflected a slightly lower acidity of the H5 proton of the hydroxy-dienone 158 with respect to the same proton of the 8a-hydroxy diastereoisomer, 157, or simply a slower reaction of the base with this substrate.  Compound 172 was obtained as colorless fine needles (190-192°C) with a molecular formula of C20H24O5. It was optically active ([a] p : - 227.0°) and its IR spectrum showed two carbonyl bands at 1752 and 1677 cm"', corresponding to the unsaturated lactone and enone respectively, and an hydroxyl absorption at 3449 cm"'. Its ' H N M R spectrum indicated epoxidation of one of the double bonds of ring C, showing only one olefmic signal at 55.96 (IH, s) and one epoxy proton at 63.35 (IH, s). Irradiation of the signal at 63.35 caused NOE enhancement of the characteristic isopropyl group signals and two multiplets at 61.67 and 2.14, corresponding to the two H7 protons.  * Due to the small content of diastereomer 173 in the mixture, a pure sample of this compound could not be obtained. Nevertheless, ' H NMR of an enriched sample showed a very similar signal pattern to that of 172, being the main difference the position of H5 at 52.42, indicating that the junction of rings A/B was trans. This was later confirmed by the isomerization of an enriched sample of 173 to 172 upon basic treatment.  138  InadiationatHll  ,|*>MI*wwf'WvHW«fm(*^^-Vrtk^V^  Y  Irradiation at H14  **%|V**'V*^1 Irradiation atH5  U  M Kl  UNf^,4M*HrfWl(V'«t#M(fV^i  Irradiation at H20  mfff,jt^Mm*^^f*^^^ H20  Hll  H16,17  H14 H15, OH H6P, H7P H2p/  H19  H6a|l  H5  1 I  _^ -T—'—'—•—•—r  — ' ;—  '  I ' I.H  '—r-" 5.U  -1—[—'  in  Figure 2.23: NOE DiEFerence spectra of compound 172  7.',  .'.«  '—1— 2.0  I l.O  139  thus indicating that the epoxide was at the C13-C14 position (Figure 2.23). The H19 protons appeared as a split AB quartet (2H, Av = 0.12 ppm, J = 16.8 Hz) at 54.70 in contrast to the singlet signal (64.71) for these protons of the starting material. Two other major changes in the spectrum were the significant down field shift of the H5 signal, appearing at 53.05 (IH, br d, J = 9.3 Hz) in comparison with the same signal at 52.35 in compound 158; and the appearance of the C8 hydroxyl signal as a sharp doublet at 62.49 (IH, d, J = 1.9 Hz, D2O exchangeable). Irradiation of the H5 proton enhanced the signals of the C20 methyl group at 51.29 (3H, s), one of the HI 9 doublets at 54.66 and a signal in the multiplet at 62.14, corresponding to one of the H6 protons. Saturation of the H20 signal confirmed the enhancement of H5 (and HI 1 at 65.96, Figure 2.23). This clearly indicated that the H5 was p-oriented, which explained its large down field shift since in this orientation, this proton is placed close in space to the C8 hydroxyl group, therefore being deshielded. On the other hand, the appearance of the hydroxyl signal as a sharp doublet, suggested that this group was "locked" in a rigid position by an hydrogen bonding effect with the neighboring C13-C14 epoxide, similar to that observed for epoxide 165 {vide supra Figure 2.17). Such hydrogen bonding would only be possible if the epoxide ring was P-oriented, therefore indicating that this was the stereochemistry of the epoxy ring (Figure 2.24-1). Additional support for this assignment came from the fact that the H14 appeared at 53.35, which was within the expected chemical shift region as compared to all the epoxy protons of triptolide (1) (53.30 to 3.90); and particularly similar to that of the same proton of the related epoxide 165 (63.31). If the C13-C14 epoxide ring was a-oriented.  140  the HI4 proton would be close in space to the C8 hydroxyl group, and this would make it appear at lower field than expected due to the deshielding effect of the oxygen atom (Figure 2.24-11). The rest of the protons in the molecule were assigned on the basis of the COSY and NOE spectra.  "W"  coupling  II  Figure 2.24: Orientation of the C13-C14 epoxide of 172  2.7 Unified synthesis of the (7,8)a-epoxy-, (7,8)P-epoxy- and 8-hydroxy- series  The fact that during the preparation of the 8-hydroxy-dienones, 157 and 158 from isotriptophenolide (ITP, 71), a high yield of 7a-hydroxy-isotriptophenolide (146a) was obtained (60% yield), suggested that this method could indeed also be utilized as a short and efficient synthesis of the diol 146a (vide supra. Scheme 2.22). 7a-hydroxy-ITP (146a), which cyclization lead to the important analogue (7,8)aepoxy-dienone 72, was originally prepared by demethylation of 7a-methoxy-ITP (150a) in a fair yield (61%)). Nevertheless, the methoxy derivative 150a was only obtained in a low 26%) yield from the oxidation of isotriptophenolide (71) with PIDA in methanol; thus bringing the corresponding overall yield to 146a to only 16%) (Scheme 2.27).  141  Considering that the next step in the synthesis of the epoxy-dienone 72 (the intramolecular cyclization of diol 146a) was also a rather low yielded step (29%), it was clear that this three-step synthetic sequence was inefficient in terms of yield.  +  71  151  153  150 a  BCI3/CH2CI2  146 a  Scheme 2.27: The original synthetic sequence to diol 146a and epoxide 72  Therefore, the new method of oxidation of ITP (71) with PIDA/water evidently represented a major improvement in the preparation of 7a-hydroxy-ITP (146a) as it provided an alternative method to prepare this intermediate in only one step and with a much better yield (60%, as compared to 16% by the method described above). Furthermore, a possibility was visualized for applying a slightly modified procedure in order to extend the scope of this procedure even further. Since the cyclization of diol 146a to epoxy-dienone 72 was performed also by treatment with PIDA (although under  142  different reaction conditions and in the absence of water), in principle it would be feasible to carry out this transformation in situ, i.e., without having to isolate the diol 146a, by simply adding two equivalents of PIDA instead of one. If this was achievable, then the preparation of the (7,8)a-epoxy-dienone (72) from isotriptophenohde (71) could be accomplished in one single synthetic step. In order to test this hypothesis, an experiment of oxidation of isotriptophenohde (71) was carried out utilizing two equivalents of PIDA in a mixture of acetonitrile/water. After stirring for 1 h at room temperature all the starting material was consumed to give, after chromatographic purification: (7,8)a-epoxy-dienone (72, 20%), 8a-hydroxydienone (157, 8%), 8p-hydroxy-dienone (158, 18%), and 7-oxo-isotriptophenolide (151, 32%) (Scheme 2.28).  72  71  157 = a-OH 158 = (3-OH  2 steps  73  151  Scheme 2.28: The "unified" synthesis of the epoxy-dienones 72 and 73; and the hydroxydienones 157 and 158  143  Thus, this transformation provided the epoxy-dienone 72 in a yield about four times higher than previously obtained (20%, as compared with a 4.6% overall yield by the route portrayed in Scheme 2.27), without affecting the yields of the expected 8-hydroxydienones 157 and 158. In addition, since the keto-phenol 151, obtained among the products of this reaction, could be easily transformed to the (7,8)P-epoxy-dienone (72) by a two-step sequence (diastereospecific reduction of the C7 ketone to the corresponding alcohol with NaBH4, followed by cyclization using PIDA), this new reaction in fact opened a remarkably short and easy alternative access to the (7,8)-epoxy- and 8-hydroxyseries of analogues.  2,8 Attempted epoxidation of quinoid-type analogues by alternative methods.  2.8.1  Experiments with DMD and enzymes as epoxidizing agents  Some alternative methods of epoxidation were evaluated in order to explore the possibility of obtaining different stereochemical variations of the epoxide functionality of the analogues prepared in the present investigation, and perhaps achieve epoxidation at the C9-C11 position in ring C. The three-membered cyclic peroxide, dimethyl dioxirane (DMD) has been 1 '5 1 1 00  employed to epoxidize a variety of alkenes  '  }')'X  and even some enone systems.  The  neutral reaction conditions under which this reaction is carried out should also prevent epimerization at the C5 position, as observed in some of our previous experiments. A freshly distilled acetone solution of dimethyl dioxirane (1.5 equiv) prepared as  144  reported in the literature,124 (concentration: 0.03M) was added to cold acetone-methylene chloride solutions (2:1, at 2-5°C) of the following substrates: the epoxy-dienones 72 and 73, the hydroxy-dienones 157 and 158, and the diastereomeric mixture of 8-methoxydienones (153), each containing 10 mg of the appropriate compound. After stirring for 8 hrs at 2-5°C, no reaction was observed as revealed by TLC and the starting materials were recovered unchanged.  In a similar set of experiments carried out at room  temperature for 10 hrs, once again no change took place. Although it was not completely clear why no epoxidation occurred, a likely reason could be a low nucleophilicity of the reagent, or because of the hindrance of the various substituents near the double bonds in ring C which did not allow the required transition state to form (it is generally believed that this epoxidation process involves a concerted spiro "butterfly" type transition state,'^^ i.e. 174, Figure 2.25).  Further  investigation in this area may be required to bring some light to the understanding of the factors involved.  .^  Epoxidize Products  174  Figure 2.25: Required transition state for the epoxidation of the dienone substrates with dimethyl dioxirane  Since the epoxidation of the respective dienone substrates with DMD was  145  unsuccessfiil, another methodology of epoxidation was tested, and this involved the use of enzymes as "biological" epoxidizing agents. Enzymes potentially allows both regio and stereo specific oxygenation reactions, which are very difficult to carry out chemically. Chloroperoxidase (CPO) and horseradish peroxidase (HRP) are two relatively stable and commercially available enzymes, which have been found to catalyze the epoxidation of some poorly activated double bonds and even a few conjugated systems in the presence of hydrogen peroxide (for HRP) or ^butyl hydroperoxide (for CPO).''^^"'^^ Ethanolic stock solutions of compounds 72, 73 and the diastereomeric mixture 153 (4 mg compound / 0.5 ml ethanol) were added to a solution containing 0.13 mg of HRP (Sigma, code EC 1.11.1.7, 1100 units/mg) in 11 mM phosphate buffer (4.5 ml, pH 6.5), at 8-10°C, followed by addition of hydrogen peroxide solution (0.5%, 0.2 ml). TLC examination of the reaction mixture after 3hrs of stirring at this temperature showed only unchanged starting material. To a second set of experiments using the same substrates and a solution of CPO (0.7 mg, Sigma, code EC 1.11.1.10, 500 units/mg) in a 10 mM sodium citrate buffer (4 ml, pH 5.5) at 18-20°C, was added /-BuOOH (4|aL). This also showed no reaction by TLC after 6 hrs. The failure of these enzymes to accomplish epoxidation of the tested compounds was most likely due to the lack of recognition of these substrates by the active site of the enzymes, which unfortunately is not unusual when effecting enzymatic transformations.  2.8.2  Biotransformation experiments with TRP4a cell cultures  From previous studies'^'^^'^^ it was known that the developed cell line of TRP4a  146  was capable of producing the triepoxides, triptolide (1) and tripdiolide (2). It was also found that these cell cultures had certain capacity to biotransform some "foreign" synthetic intermediates, such as isotriptophenoHde (71) to produce small amounts of epoxidized compounds (i.e., 72-74, vide supra). It was clear therefore, that "appropriate" abietane-type diterpenes could act as substrates in enzyme-catalyzed oxidation reactions to the corresponding oxidized products. The question as to whether the enzymes present in the TRP4a cell cultures could act as "reagents" in catalyzing the epoxidation of the dienone system of the compounds prepared in this investigation, to the di- or triepoxide system was of interest and required evaluation. The selection and evaluation of optimal incubation conditions for cultures of the TRP4a cell line has been extensively studied in Dr. Kutney's laboratory, and parameters such as type of growth medium, pH, units of peroxidase enzyme per mmol of substrate to be biotransformed, etc. are well established.^^'^^ Thus, the present study focussed on the evaluation of the appropriate age of the cell culture (which dictates what type of enzymes are present in the medium) and the determination of the optimal reaction times to biotransform the substrates, in the hope that epoxidation on ring C of the molecule could be accomplished. For the biotransformation studies, a set of experiments of incubation of the epoxydienones 72 and 73 with TRP4a cell cultures of three different ages: 7, 14 and 25 days old was performed. MSNA0.5K0.5 media was utilized to grow the cell culture. In general, the substrate was dissolved in an appropriate amount of ethanol, added to the culture and incubation carried out under standard conditions (see Experimental). Samples were taken  147  and extracted at certain time intervals and analyzed by TLC and high performance liquid chromatography (HPLC). A general extraction procedure, as outlined in Figure 2.26 was followed, in which the cells and the spent medium were extracted separately for convenience, and later combined for the analysis. Extraction of the spent medium was carried out using ethyl acetate. The cells were freeze-dried and extracted also with ethyl acetate.  Cell suspension culture TM  Filtration through Miracloth  Freeze dry  ^' Broth  y  ^ Cells Saturation with NaCI Extraction with EtOAc Solvent removal  Homogenization in EtOAc Filtration, Separation T'  1  Broth extract  Solvent 1f  Cell extract  V  Cell pellets  Figure 2.26: General extraction procedure of TRP4a cell cultures  Biotransformation of 72 with TRP4a cell cultures of different ages  A solution of (7,8)a-epoxy-dienone in ethanol (72, 15 mg in 1 ml) was added to  148  TRP4a cell cultures (7, 14 and 25 days old, respectively) and the culture was incubated under standard conditions. Samples were withdrawn every 24 hrs and the culture was harvested after incubation for 4 days. A control experiment and a blank experiment were also conducted simultaneously. The samples were extracted and then analyzed by TLC and HPLC, initially. For TLC, two different solvent systems were developed to analyze the compounds, namely A (toluene-chloroform-ethyl acetate-formic acid, 105:48:45:3) and B (methylene chloride-methanol, 97:3). The results showed that there were no significant changes in the 7-day-old culture samples in comparison with control and blank samples. produced  In the 14-day-old culture, the Rf values of the major compounds  corresponded  isotriptophenolide  to  the  7a-hydroxy-isotriptophenolide  (151), 7p-ethoxy-isotriptophenolide  (146a),  (175b) and  7-oxo-  (6,7)-dehydro-  isotriptophenolide (159) {vide infra. Scheme 2.29). These crude samples were then further analyzed by HPLC. The HPLC retention times from the crude samples matched exactly to those of the above mentioned products, in comparison to authentic samples of these compounds.  The final confirmation came from the MS analysis of the crude  mixture isolated at the end of the 4-day incubation period, which showed the molecular ion peaks appearing at m/z 328, 326, 356 and 310, matching the molecular weights of 7a-hydroxy-isotriptophenolide  (146a),  7-oxo-isotriptophenolide  (151),  7p-ethoxy-  isotriptophenolide (175b) and (6,7)-dehydro-isotriptophenolide (159), respectively. This 14-day-old culture, also showed almost complete consumption of the starting material as evidenced by both TLC and HPLC results. Figure 2.27 shows the consumption of the starting material with respect to time for the different ages of the TRP4a cell cultures.  149  20 U)  15  g  10  c  3 O  E <  5 0  -7-day-old -14-day-old  " 5 ^ 1  2  3  .25-day-old  4  Incubation Time (days)  Figure 2.27: Consumption of starting material during the biotransformation of 72 with TRP4a cell cultures  Addition of (7,8)a-epoxy-dienone (72) in ethanol to a 25-day-old TRP4a cell culture, showed a very similar observation as to the 14-day-old culture, except that only about 50% consumption of the starting material was achieved.  The samples were  extracted, and analyzed by TLC after developing twice in solvent systems A and B; and by HPLC. The results showed that the starting material (Rf 0.61, system B) remained in substantial quantities, although two more polar spots (Rf 0.13 and 0.24; 146a and 151, respectively) appeared after 1 day, and two additional spots (Rf 0.31 and 0.69; 175b and 159, respectively) were noted after 2 days of incubation. The culture was harvested after 4 days since no significant changes were observed by tic or hplc after the third day (Figure 2.28).  150  0.06 0)  3 U  1  0.05 O04  0.03  O)  E  0.02.  4^  C 3  o E n  Q01  0-  1  2  3  4  Incubation Time (days)  1  2  3  4  Incubation Time (days)  Figure 2.28: Changes in the amounts of products formed during the biotransformation of 72 with TRP4a cell cultures (cell ages: A, 14 days; B, 25 days)  Biotransformation of (7,8)|3-epoxy-dienone, 73  It was shown by the results from the biotransformation of (7,8)a-epoxy-dienone (72), that the ages of the cell culture played an important role in the biotransformation.  151  Therefore, a series of similar experiments as to compound 72 were performed with (7,8)P-epoxy-dienone. The starting material was incubated with TRP4a cell cultures 7, 14 and 25 day-old respectively. Samples were taken at certain intervals (1,2 and 3 days) and the cultures were harvested after 4 days of incubation. TLC analysis of these samples showed very similar results to those with the (7,8)a-isomer as substrate, showing the same spots, with no appearance of any other new product. The 7-day-old experiment gave total recovery of the starting material. The hplc results confirmed the formation of the same products as in the biotransformation of 72, except that the orientation of the C7 hydroxyl of the diol was (3 (146b) and the C7 ethoxy group of 175a was a, by comparison with the retention times of authentic samples (Scheme 2.29).  TRP4a  146a = a-OH 146b = P-OH  14 or 25-day old  151  +  72 = a-epoxide 73 = P-epoxide  175a = a-OEt 175b = P-OEt  159  Scheme 2.29: Products of the biotransformation of 72 and 73 with TRP4a cell cultures  152  Figure 2.29 illustrates the conversion of the starting material (73) into the different products formed during these experiments.  B 0.05 if 0.04 E o) 0.03 E XT 0.02 c  o 0.01 w 0 1  B  c  2 3 Incubation Time (days)  4  0.035  1146a 1151 1175a ,159  0.01  3  1  2  3  4  Incubation Time (days)  Figure 2.29: Changes in the amounts of products formed during the biotransformation of 73 with TRP4a cell cultures (cell ages: A, 14 days; B, 25 days)  The products formed in both sets of experiments with compounds 72 and 73 were  153  most likely derived from opening of the epoxide at the 7,8 position in the substrates under the acidic medium (pH 5.0) or, most likely, by enzymes present in the TRP4a cell cultures, giving the corresponding 7-hydroxy-isotriptophenolide as the major product. This compound then could have been transformed further to yield the other products observed in these experiments. Enzymatic oxidation of the C7 hydroxyl group would yield ketone 151, while dehydration of the same functional group would produce compound 159. Apparently the C7 ethoxy derivatives (175a and 175b) were formed by nucleophilic substitution of the protonated hydroxyl group (formed under the acidic medium) of 146a and 146b, respectively by the ethanol present in the culture media; as suggested by the observation of opposite stereochemistry at C7 between the corresponding hydroxyl and the ethoxy groups. This result was consistent with some previous observations on the biotransformation of 7-hydroxy-ITP (146) with TRP4a cell cultures and with the expected lability of this diol under acidic conditions. In summary, the biotransformation of (7,8)a-epoxy-dienone (72) and (7,8)Pepoxy-dienone (73) with TRP4a cell cultures of different ages gave very similar results, with higher consumption of starting material for the 14-day-old cultures. From these experiments, it can be shown that the TRP4a cell line had the ability to biotransform compounds 72 and 73, but they did not seem to contain the appropriate enzymes to carry out epoxidation of these type of substrates to give the di- or triepoxide system as in triptolide (1) or tripdiolide (2).  154  2.9  Conclusions The development of two short alternative syntheses to isotriptophenolide (71)  greatly facilitated the preparation of large quantities of this key intermediate. A variety of new diterpene analogues of the abeo-abietane family were synthesized, comprising a number of mono- and di-epoxy derivatives possessing a quinoid-type structure in ring C of the skeleton (15 derivatives were prepared in total). From these compounds, the epoxy-dienones 72 and 73, previously isolated in small quantities from biotransformation experiments with TRP4a cell cultures, were synthesized here for the first time. Given the structural similarities between some of the active analogues previously reported in the literature and the quinoid/epoxy skeleton of the group of derivatives prepared in the present research, it is feasible that some of the latter compounds may possess  interesting  biological  activity, which will  be determined  by  future  pharmacological studies. In addition, it will be interesting to evaluate the activity of the new structural variation to the basic 18(4—>3) abeo-ahietane framework, present in those epoxidized derivatives having a ring A/B cw-j unction, since this is a new class of compounds which consequently has never been studied before. On the other hand, the biotransformation of dienones 72 and 73 with TRP4a cell cultures showed that the enzymes present in this cultures were unable to perform the required epoxidation of the ring C dienone system of this kind of substrates.  This  indicates that these enzymes are highly specific to the epoxidation of the biosynthetic precursors of triptolide (1) and tripdiolide (2), and do not recognize compounds with the present structural variations in the molecule.  See note 2 in Appendix.  155  CHAPTER 3  EXPERIMENTAL  3.1  General  Chemicals and solvents used in all synthetic reactions were of reagent grade. Solvents used for HPLC analyses were of HPLC grade. Technical grade solvents were used for extractions without purification.  Anhydrous magnesium or sodium sulphate  were used as drying agents for organic extracts. Prior to use, anhydrous THF, ether and benzene were freshly distilled in the presence of sodium and benzophenone under argon. Moisture-sensitive reactions were carried out in oven-dried glassware and under a positive pressure of argon or nitrogen. Analytical TLC was performed on aluminum-backed, silica gel plates (Merck silica gel 60 F254). Preparative TLC was carried out on glass-backed, pre-coated plates (Merck silica gel 60 F254, 0.25, 0.5 or 1 mm). Developed TLC plates were initially visualized with UV illumination, and then by heat treatment of plates sprayed either with a 5% solution of ammonium molybdate in \0% sulfuric acid, or with a 30% solution of concentrated sulfuric acid in acetic acid followed by a 5% solution of anisaldehyde in isopropanol. Flash column chromatography on silica gel (Merck silica gel 60, 230-400 mesh) was run under moderate air pressure to maintain a proper eluent flow rate. Reagent grade  156  solvents or distilled technical grade solvents were employed for column chromatography. For some chromatographic purifications a chromatotron model 8924 was used. Gas Chromatography (GC) analyses were performed on a Hewlett Packard 5890, equipped with a flame ionization detector. High Performance Liquid Chromatography (HPLC) analyses were carried out on a Waters 71 OB, using a UV detector. Melting points were measured on a Reichert melting point apparatus and are uncorrected. Optical rotations were recorded on a Perkin-Elmer 141 or a 241 MC polarimeter using a quartz cell with a 10 cm path length. The concentration of the samples in units of grams per 100 mL is given in parentheses. UV  spectra  were  obtained  with  a Perkin-Elmer  Lambda  2 UVA'^is  spectrophotometer using quartz cells of 1 cm path length. IR  spectra  spectrophotometer  were recorded or  on  a  either  Perkin  on a Perkin  Elmer  1710  Elmer  infrared  71 OB infrared  Fourier  transform  spectrophotometer. IH NMR spectra were recorded on Bruker WH-400, AE-200 or AMX-500 spectrometers.  Chemical shifts (5) are reported in ppm relative to tetramethylsilane  (TMS), and in most instances the residual 'H or '^C signal of CDCI3 were used as indirect references to TMS. Those values of 6 are reported to be at 7.24 ppm for 'H and the carbon signal of CDCI3 at 77.0 ppm. All COSY experiments and NOE experiments were performed on the Bruker WH-400 spectrometer.  157  13c NMR spectra were obtained on Varian XL-300 (at 75.3 MHz) or Bruker AMX-500 (at 125.8 MHz) spectrometers. Chemical shifts (5) are cited in ppm relative to TMS. HMQC and HMBC spectra were performed on the AMX-500 spectrometer. Low resolution mass spectra (LRMS) or electron impact mass spectra (EIMS) were determined on Kratos MS 50 or MS 80 mass spectrometers. High resolution mass spectra (HRMS) were recorded on the Kratos MS 50 spectrometer. Elemental analyses were performed by Mr. P. Borda, in the Microanalytical Laboratory of the University of British Columbia.  3.2  Synthesis of isotriptophenolide (71)  For those synthetic intermediates (129-138, 65, 71) who had also been obtained previously in Dr. Kutney's laboratory, additional spectroscopic data, with more detailed analyses were obtained in the present study (including some corrections for previous assignments). Also, several modifications to the original procedures to optimize the yields, as well as some entirely new synthetic routes to some of these intermediates were made by the present author. All of these data and changes are included in the appropriate sections below.  3.2.1  Synthesis of isodehydroabietenolide (65)  158  Purification of crude dehydroabietic acid (DHA, 60)  Technical grade dehydroabietic acid (DHA) (Pfaltz and Bauer, Inc. 300 g) was dissolved in warm ethanol (500 mL, 40-50 °C) in a 4.0 L capacity liquid-liquid extractor. After complete dissolution of DHA, ethanolamine (50 mL) and water (500 mL) were added with stirring. The resulting brown-colored mixture was extracted continuously with petroleum ether for 20 h while stirring at 55-60 °C. The resulting aqueous solution, thus freed of neutral material (organic phase) by separatory ftinnel, was heated at 65 °C for 15-20 min to evaporate any dissolved petroleum ether.  The aqueous solution was left to cool down at room  temperature for 10-15 min. Cooling this solution to 0-4 °C overnight, resulted in a semisolid mass, which was broken up manually. Filtration provided the ethanolamine salt of dehydroabietic acid, which was resuspended in 50% aqueous ethanol (380 mL) at 4 ^C and vacuum filtered. The resulting moist salt was then dissolved in hot ethanol (410 mL) followed by the addition of acetic acid (50 mL). To this refluxing solution, water (200 mL) was added gradually until a cloudy solution was obtained. The hot solution was filtered and cooled to room temperature. The crude crystals of DHA was isolated by vacuum filtration and washing with 50% aqueous ethanol (100 mL). Recrystallization of the crude acid was carried out by dissolving the product in boiling ethanol (300 mL), followed by the addition of water (164 mL) just to reach the solution's cloudy point. The solution was cooled to room temperature and kept at 4 °C. The resulting crystals were  159  filtered and dried to give 79.8 g of purified dehydroabietic acid (60), 89% pure by GC analysis (see appendix) with mp of 160-163 °C.  18-Norabieta-4(19),8,ll,13-tetraene(134)*  Acid chloride 129  Dehydroabietic acid (DHA) 60 (74.0 g, 0.246 mol, purity, 89%) dissolved in dry benzene (575 mL), was placed in a flask equipped with a thermometer, a magnetic stirrer, a Dean-Stark trap and a condenser with a moisture trap (anhydrous calcium chloride). The solution was brought to reflux for 1 h and then allowed to cool to room temperature. The Dean-Stark trap was replaced by a condenser and thionyl chloride (99%), 21.4 mL, 0.295 mol) was added to the reaction mixture. Dry dimethyl formamide (0.82 mL) was added dropwise and the solution was stirred for 1 h at room temperature. After this time, the mixture was warmed up to 68-73 °C and the temperature maintained for 30 min. The solution was then refluxed for 15 min and cooled to room temperature. Removal of the solvent by rotary evaporator in vacuo provided the crude acid chloride 129 (80.49 g) as a thick brown oil [IR (neat) v,^^^: 1780 cm"!].  In this part of the synthesis, the intermediates 129-133 were not purified but converted directly in sequence to the exo-olefin 134. The latter was then purified and characterized.  160  Isocyanate 131  The crude acid chloride 129 (80.49 g, 0.253 mol) was dissolved in reagent grade acetone (494 mL).  The  solution was cooled to -5-0 °C using an ice-salt bath. A solution of sodium azide (19.80 g, 0.304 mol) in water (65 mL) was added dropwise to the reaction mixture with vigorous stirring. After the addition was completed the suspension was stirred for 5-10 min at -5-0 ^C and then allowed to warm up to room temperature. Toluene (164 mL) was then added at room temperature and the mixture was stirred vigorously for an additional 10 min. The organic layer was decanted, dried over anhydrous sodium sulphate, filtered and the remaining acetone was removed under vacuum to a volume of 197 mL. The solution of acyl azide 130 [IR (toluene) v^^^: 2120 (azide), 1695 (carbonyl) cm''] was brought to a volume of 494 mL with additional amounts of toluene and was heated slowly to reflux for 1 h. The reaction mixture was monitored continually by IR. The solvent was then removed in vacuo and the crude isocyanate 131 (78.46 g) was obtained as a dark thick oil [IR (neat) v,„,,: 2250 cm"!].  161  Monomethylamine 132  A solution of crude 131 (78.46 g, 0.264 mol) in anhydrous THF (348 mL) was slowly added through an addition funnel to a stirred suspension of lithium aluminum hydride (LAH) (12.18 g, 0.321 mol) in anhydrous THF (608 NHCH  mL) at -5 to 0 ^C (ice-MeOH bath) under argon. After the  addition, the ice-methanol bath was removed and the stirred reaction mixture was allowed to reach room temperature and was stirred for 1 h. The reaction mixture was then heated to 66 <^C and was refluxed with stirring under argon for 22 h. After the elapsed time, the mixture was allowed to cool to room temperature. Sequential dropwise additions of reagent grade acetone (17.4 mL), water (12.2 mL), 15% aqueous sodium hydroxide (12.2 mL), and more water (34.8 mL) produced a thick white suspension which was filtered through a sintered glass fimnel (packed with Celite-545) and washed with hot THF (350 mL). The filtrate was concentrated and the resulting residue was redissolved in diethyl ether (348 mL), dried over anhydrous sodium sulphate and evaporated in vacuo to form the crude monomethyl amine 132 (74.7 g) as a thick yellow syrup [IR (neat) v^^^: 3300 cm'l].  162  Dimethylamine 133  The crude monomethyl amine 132 (74.7 g, 0.262 mol) was dissolved in reagent grade formic acid (200.6 mL) followed by the addition of 37% formaldehyde solution (100.3 mL) dropwise. The mixture was then gently refluxed *N(CH3)2  (115-120 °C) for 3 h, cooled to room temperature, and the solvents were removed in vacuo at 70 °C. The resulting tarry mass was dissolved in diethyl ether (616 mL) and a solution of 4N NaOH (401 mL) was added dropwise to complete dissolution. The organic layer was separated, dried, filtered and evaporated to dryness yielding a golden syrup of crude dimethylamine 133 (74.51 g) [IR (neat) v^^^:  3350 cm-1].  Exo-olejin 134 To a stirred solution of the crude dimethylamine 133 (74.51 g, 0.249 mol) in reagent grade chloroform (1490 mL) at -40 °C (dry ice-CCl4-acetone bath), /w-chloroperbenzoic acid (m-CPBA) (60%, 93.8 g, 0.393 mol) was added by portions over a 20 min period. After the addition had been completed, the mixture was stirred for 30 min at -40 °C. Triethylamine (14.9 mL, 0.107  163  mol) was added dropwise and the mixture was warmed to room temperature and then refluxed at 60 ^C for 1 h. The reaction was examined by thin layer chromatography (TLC) (mobile phase: hexanes). The reaction mixture was cooled to room temperature and the solvent removed in vacuo. The residue was taken up in diethyl ether (976 mL) and washed consecutively with 10% sulfuric acid (976 mL), 10% potassium carbonate (2 X 976 mL) and brine (976 mL). The organic layer was dried, filtered, and the solvent was evaporated to afford the crude exo-olefm (94.48 g). The crude product was purified by column chromatography using silica gel (230-400 mesh), eluting with hexanes followed by hexanes-EtOAc (96:4) to give the pure exo-olefm 134 as a colorless oil (28.91 g, 45.7%; or 51.8% overall yield from dehydroabietic acid, 60 considering the DHA purity as 89%).  Physical data of 18-norabieta-4(19),8,l 1,13-tetraen (134):  A colorless oil; [a]^^: +215.2° (c = 2.20, CHCI3); UV ^tj^^f" (log s): 218.2 (3.55), 267.8 (2.90), 276.3 (2.92); IR (neat) v„,, cm"': 3055 (aromatic and olefinic CH), 2965 (CH), 1640 (C-C), 1610 (aromatic C-C); 'H NMR (400 MHz, CDCI3) 6: 1.01 (3H, s, H20), 1.24 (6H, d, J = 6.9 Hz, H16, H17), 1.58 (IH, ddd, J = 12.8, 12.8, 4.6 Hz, Hla), 1.65-1.89 (4H, m, H2, H6), 2.06 (IH, ddd, J = 13.0, 12.8, 5.8 Hz, H3a), 2.23 (IH, dd, J = 11.9, 1.0 Hz, H5), 2.28 (IH, br d, J= 12.8 Hz, Hip), 2.38 (IH, br d, 7 = 12.9 Hz, H3P), 2.85 (IH, sept, J = 6.9 Hz, H15), 2.90 (2H, m, H7), 4.61 (IH, d, J = 1.6 Hz,  H19A),  (IH, d, J = 1.6 Hz, H19B), 6.94 (IH, br s, H14), 7.01 (IH, dd, J= 8.2, 1.6 Hz, H12),  4.86 7.22  164  (IH, d, J = 8.2 Hz, H l l ) ; ''C NMR (75.3 MHz, CDCI3) 5: 21.4, 22.8, 23.7, 24.0 (2C), 30.1, 33.5, 36.4, 38.5, 39.2, 47.9, 106.3, 124.0, 125.4, 127.1, 134.9, 144.7, 145.7, 150.7; EIMS m/z (rel. intensity): 254 (M^ 35.8), 239 (90.5), 211 (5.1), 197 (base peak), 169 (10.8), 155 (12.1), 141 (25.5); HRMS calcd. for C.gH^s: 254.2034; found: 254.2031.  18,19-Dinorabieta-8,ll,13-trien-4-one(135)  The purified exo-olefin 134 (28.91 g, 0.114 mol) was dissolved in methanol-methylene chloride (5:1, 1126 mL), and was divided into 2 equal portions of 563 mL each in two flasks and ozonized.  Ozone was generated by a  laboratory ozonator (Welsbach, model T-23) set at 2.2 psi of oxygen, 90 volts and a flow of ozone 0.014-0.015 L/min. These solutions were cooled to -78 °C in a dry ice-acetone bath. Ozone was passed into the vigorously stirred solutions at -78 °C via a gas-dispersion tube, until the solution turned a pale blue color (about 60 min for each portion). hexanes:EtOAc 9:1).  The reaction was monitored every 15 min by TLC (eluent: The reaction mixtures were continuously stirred at the bath  temperature for an additional 20 min and dimethyl sulphide (2.45 mL) was added to each flask. The mixtures were allowed to warm to room temperature and stirred for 20 h. Analytical TLC using hexanes:EtOAc (9:1) showed that both reactions were identical, and they were combined and the solvent removed in vacuo. The residue was dissolved in hexanes-diethyl ether 2:1 (1000 mL) and washed with water (3 x 200 mL) and brine (200  165  mL).  The aqueous layer was extracted again with diethyl ether (400 mL) and the  combined organic layers were dried, filtered and concentrated to afford the crude ketone (29.18 g) as a light yellow oil.  This crude oil was purified by flash column  chromatography using hexanes-EtOAc (9:1, 8:2), yielding pure ketone 135 (24.08 g, 83%) as a colorless oil which slowly crystallized to a bright white solid at about 5 ^C.  Physical data of 18,19-dinorabieta-8,l l,13-trien-4-one (135):  Colorless prisms; mp: 37-39 «€ (lit: 40-42 oc''); [a]^^: +173.50 (^ = o.92, CHCI3); UV X^^^°" (log s): 205.0 (4.24), 267.1 (3.07), 275.6 (3.10); IR (neat) v,„,, cm"': 3050 (aromatic CH), 2960 (CH), 1695 (C=0), 1605 and 1500 (aromatic C=C); iH NMR (400 MHz, CDCI3) 5: 1.05 (3H, s, H20), 1.22 (6H, d, J = 6.9 Hz, H16, H17), 1.74-1.94 (2H, m, HI a, H6p), 1.95-2.16 (3H, m, H2, H6a), 2.40 (3H, m. Hip, H3), 2.58 (IH, dd,J = 12.4, 2.4 Hz, H5), 2.82 (3H, m, H7, H15), 6.92 (IH, br s, H14), 7.02 (IH, dd, J = 8.2, 1.6 Hz, H12), 7.20 (IH, d,J= 8.2 Hz, Hll); ''C NMR (75.3 MHz, CDCl,,) 5: 17.5, 22.6, 23.7, 23.9 (2C), 28.7, 33.5, 36.9., 40.9, 42.3, 55.3, 124.1, 124.9, 127.3, 134.6, 143.0, 146.4, 212.3; EIMS m/z (rel. intensity): 256 (M\ 29.8), 241 (base peak), 223 (15.8), 213 (18.0), 199 (5.6), 181 (33.1), 171 (19.9); HRMS calcd. for C,gH240: 256.1827; found: 256.1822.  Physical data of 18,19-dinorabieta-8,l l,13-trien-4,7-dione (135a):  166  Colorless crystals; mp: 105-107 ^C (lit.: 107-108 oQ"); IR (CHCI3) v„^ cm"': 3035 (aromatic CH), 2960 (CH), 1710 (C=0), 1680 (C=0), 1600 and 1495 (aromatic -O  C-C); 'H NMR (400 MHz, CDCI3) 6: 1.15 (3H, s, H20), 1.25 (6H, d, J = 6.9 Hz, H16, H17), 1.94-2.25 (3H, m, Hla,  H2), 2.33-2.53 (3H, m, Hip, H3), 2.73 (IH, dd, J = 18.6, 4.3 Hz, H5), 2.86 (IH, dd, J 18.8, 13.2 Hz, H6p), 2.93 (IH, sept, J = 6.9 Hz, H15), 3.09 (IH, dd, J - 13.2, 4.2 Hz, H6a), 7.35 (IH, d, J= 8.1 Hz, HI 1), 7.42 (IH, dd, J = 8.1, 2.0 Hz, H12), 7.92 (IH, d, J = 2.0 Hz, H14); EIMS m/z (rel. intensity): 270 (M^ 48.7), 255 (base peak), 227 (8.3), 213 (27.4), 199 (10.1), 185 (32.2), 173 (5.0).  3-Dimethylthiomethylene-18,19-dinorabieta-8,ll,13-trien-4-one (136)  To a stirred solution of 4-methyl-2,6-di-rbutylphenol (54.2 g, 0.246 mol) in anhydrous THF (1278 mL) at -5 to 0 «€ (ice-MeOH bath) was MeS  added a solution of «-butyl lithium (1.6 M in SMe  o  hexanes, 154 mL, 0.246 mol) dropwise under  argon, followed by carbon disulphide (54.8 mL, 0.813 mol) addition. The resulting red solution was allowed to warm to room temperature. A solution of the 4-ketone 135 (24.08 g, 0.094 mol) in anhydrous THF (186 mL) was added dropwise to the reaction mixture. The flask containing the ketone solution was washed with an additional amount  167  of anhydrous THF (68 mL) and was added to the reaction flask. The reaction mixture was stirred at room temperature for 48 h. The completion of the reaction was monitored by TLC (hexanes-EtOAc, 9:1) and then developed with anisaldehyde spray reagent. Methyl iodide (33 mL, 0.53 mol) was then added dropwise and the resulting reaction mixture was stirred at room temperature for a further 20 h, with the reaction flask wrapped in aluminum foil to minimize photo-oxidation. The solvent was evaporated and the residue dissolved in diethyl ether (1865 mL), washed with water (3 x 1000 mL) and brine (1000 mL).  The organic layer was dried, filtered and concentrated in vacuo  yielding the crude red oil of the ketene thioketal 136 (101.99 g). This crude product was purified repeatedly by flash column chromatography with hexanes-EtOAc (9:1) as the eluant. The pure ketene thioketal 136 was obtained as a light orange oil which rapidly solidified upon cooling (31.82 g, 94.0%).  Physical data of 3-dimethylthiomethylene-18,10-dinorabieta-8,l l,13-trien-4-one (136):  A yellow solid; mp: 67-69 ^C UV X^^  (log e): 206.2 (4.31), 318.8 (3.80); IR  (neat) v„„, cm"': 3040 (aromatic CH), 2960 (CH), 1675 (C=0); iH NMR (400 MHz, CDCI3) 5: 1.09 (3H, s, H20), 1.22 (6H, d, J = 7.0 Hz, H16, H17), 1.81 (IH, m, H6p), 1.93 (IH, ddd, J = 12.6, 12.5, 6.2 Hz, Hla), 2.24 (IH, m, H6a), 2.35 (3H, s, -SCH3), 2.37 (3H, s, -SCH3), 2.47 (IH, m, Hip), 2.60 (IH, dd, 7 = 12.2, 2.5 Hz, H5), 2.68-2.94 (4H, m, H2p, H7, H15), 3.38 (IH, ddd, J = 16.5, 6.4, 2.5 Hz, H2a), 6.93 (IH, br s, H14), 7.02 (IH, dd, J = 8.0, 1.6 Hz, H12), 7.20 (IH, d, J = 8.0 Hz, Hll); ''C NMR (75.3 MHz,  168  CDCl,) 5: 17.8, 17.9, 18.1, 23.8, 23.9 (2C), 29.1, 29.7, 33.4, 37.0, 39.9, 56.3, 124.2, 125.0, 127.2, 134.8, 139.5, 142.7, 143.8, 146.5, 201.1; EIMS m/z (rel. intensity): 360 (M\ 45.2), 345 (base peak), 313 (16.4), 297 (8.9), 199 (35.1), 127 (51.4); HRMS calcd. for C2,H280S2: 360.1582; found: 360.1579.  19-Hydroxy-18(4->3)aZ>eo-abieta-3,8,ll,12-tetraen-18-oic acid lactone (65)  A solution of n-butyl lithium (1.6M in hexanes, 71.7 mL, 0.114 mol) was added dropwise to a stirred suspension of trimethylsulphonium iodide (26.99 g, 0.133 mol) in anhydrous THF (552 mL) at -70 ^C (dry ice-acetone bath) under argon. The reaction mixture was allowed to warm to -10 ^C and stirred at this temperature for 30 min. The reaction mixture was then cooled to -70 °C. The ketene thioketal 136 (31.80 g, 0.088 mol) in anhydrous THF (166 mL) was added dropwise to the reaction mixture while the temperature was maintained between -65 to -70 ^C. The resulting yellow-brown mixture was stirred at -70 <^C for another 30 min and then allowed to reach room temperature (in approximately 2 h).  After this time the reaction was completed according to TLC  monitoring (hexanes-EtOAc, 9:1; developer: anisaldehyde spray and sulphuric acid). The solvent was evaporated and the residue was taken up in diethyl ether (175 mL) and again evaporated to dryness. The residue was then redissolved in diethyl ether (2210 mL) and washed with water (2 x 300 mL). The organic phase was evaporated to give a residue  169  which was then dissolved in acetonitrile (205 mL) and methanol (565 mL). This solution was cooled in an ice-salt bath to -5 to 5 °C and concentrated HCl (70 mL) was added while stirring. After the addition was completed the cooling bath was removed and the heterogeneous mixture was stirred at room temperature for 40 h. The solvents were then removed and the residue was extracted with diethyl ether (2210 mL). The organic solution was washed with saturated sodium bicarbonate (3 x 300 mL) and water (100 mL). The water phase was extracted with ethyl acetate (2 x 500 mL) and all the organic layers were combined, dried, filtered and concentrated to yield an orange-brown crude butenolide (22.43 g). This crude was then purified by recrystallization with hexanesEtOAc and the mother liquor by column chromatography using hexanes-EtOAc (4:1), to give pure butenolide 65 (14.30 g, 54.9%; 42.5% yield from exo-olefin 134 or 22.1% overall yield from dehydroabietic acid) as a colorless oil which slowly solidified to give white needles.  Physical data of 19-hydroxy-18(4-^3)flZ>eo-abieta-3,8,l l,13-tetraen-18-oic acid (65):  White needles; mp: 97-99 ^C; [a]^^ : +41.5° (c = 0.81, MeOH); UV X^l^^  (log  s): 220.2 (4.28), 267.9 (2.83), 276.0 (2.84); IR (CHCI3) v„,, cm"': 2960 (CH), 1750 (C=0), 1675 (C=C), 1600 (aromatic C=C), 1340, 1020 (C-O-C); iR NMR (400 MHz, CDCl3)5: 1.01 (3H, s,H20), 1.22 (6H, d, J = 6.9 Hz, HI6, HI7), 1.70 (IH, ddd, J = 12.4, 12.4, 6.4 Hz, Hla), 1.90 (2H, m, H6), 2.37 (IH, m, H2P), 2.50 (2H, m, Hip, H2a), 2.71 (IH, m, H5), 2.82 (IH, sept, J = 6.9 Hz, HI5), 3.01 (2H, m, H7), 4.76 (2H, br AB^ Av -  170  0.06 ppm, J = 17.2 Hz, H19), 6.96 (IH, d, J = 1.5 Hz, H14), 7.03 (IH, dd, J = 8.0, 1.5 Hz, H12), 7.25 (IH, d, J = 8.0 Hz, HI 1); "'C NMR (75.3 MHz,CDCl3) 5: 18.1, 20.2, 22.3, 24.1 (C16, C17), 28.4, 32.6, 33.6, 36.3, 41.4, 70.5 (C19), 124.1 (Cll, C12), 125.0 (C3), 127.5, 134.3, 142.5, 146.7, 163.1 (C4), 174.2 (C18); EIMS m/z (rel. intensity): 296 (M", 42.1), 281 (base peak), 239 (24.6), 221 (10.2), 193 (26.7), 186 (48.1); HRMS calcd. for C20H24O2: 296.1776; found: 296.1770; Anal, calcd. for  CJQHJA:  C,  81.05; H, 8.16;  found:C, 80.97; H, 8.15.  Alternative simplified syntheses of the exo-olefin (134)  Method A  To a stirred solution of DHA (28 g, 93.2 mmol, purity: 86%) in 196 mL of benzene under argon was added sequentially copper (II) acetate monohydrate (5.58 g, 28.0 mmol) and pyridine (8.55 mL, 106.3 mmol). This solution was refluxed and lead tetraacetate (45.5 g, 102.6 mmol) was added gradually over a period of 2 h with vigorous stirring. The reaction mixture was stirred for another 45 min under these conditions and was monitored by TLC. Since a small amount of DHA was still present as visualized by TLC, an extra amount of lead tetraacetate (6.2 g) was added to the reaction mixture and let it to stir for an additional Ih, until all the starting material was consumed. Then the reaction mixture was cooled down to about 20 ^C, filtered through celite, washed with ether (300 mL). This solution was then washed with 10% HCl (2 x 100 mL) followed by  171  water (3 x 100 mL) and brine (1 x 60 mL), dried and evaporated under vacuum to obtain a thick yellow residue (24.75 g). This crude mixture was then purified by column chromatography using silica gel and eluting with hexanes-EtOAc (96:4) to give a pure mixture of the olefins 134, 134a and 134b in the ratio of 47:35:18 respectively, in accord with its 'H NMR analysis (18.2 g, 89.2%, starting from 86% pure DHA).  Method B  Dehydroabietic acid, 60 (1.0 g, 3.3 mmol, purity: 86%) was dissolved in dry benzene (105 mL) containing copper (II) acetate monohydrate (0.23 g, 1.15 mmol) and pyridine (0.33 mL, 3.9 mmol) and stirred for 15 min under argon. To this solution was added PIDA (6.35 g, 19.8 mmol) in portions of 0.53 g every hour and the reaction mixture was refluxed for another 12 h (total time, 24 h) After this time, the reaction mixture was cooled down to room temperature and filtered through a short pad of silica gel and washed with hexanes. The residue after evaporating the solvent was purified by flash chromatography using silica gel and eluting with hexanes-EtOAc (96:4) affording a pure mixture of olefins 134 and 134a, 82:18 by 'H NMR (565 mg, 113% considering the purity of86% for DHA).  Ozonolysis of the mixture of olefins prepared by the alternative method A: The purified mixture of olefins (18.2 g, 71.54 mmol) was ozonized following the same procedure as described above for the pure exo-olefin 134. Purification of the desired ketone 135 from the oxidized products of the other two olefins present in the  172  crude mixture was easily accomplished under the same conditions as above. The ketone 135 (7.75 g) was obtained in 89.9 % yield considering the content of exo-olefin 134 in the mixture, or 37.7% overall yield from DHA.  3.2.2  Synthesis of isotriptophenolide (71)  12-Acetyl-19-hydroxy-18(4—>3)a6eo-abieta-3,8,ll»13-tetraen-18-oic  acid  lactone  (137)  °%^  A solution of 65 (9.10 g, 30.7 mmol) in carbon disulfide (190 mL) and acetyl chloride (10.92 mL, 153.5 mmol) was added to a suspension of anhydrous AICI3 (13.56 g, 101.7 mmol) in carbon disulfide (692 mL) with vigorous stirring under argon. The reaction mixture was  refluxed overnight (about 16 h). The carbon disulfide was removed under vacuum and cold aqueous HCl (514 mL, concentrated HCl-water, 3:10) was added to the residue. The resulting suspension was stirred in an ice-water bath for 15 min and then at room temperature for 30 min. After all the complex was destroyed, the mixture was stirred vigorously with diethyl ether (1020 mL) for another 30 min. The aqueous layer was further extracted with ether (200 mL) and all the combined ethereal extracts, were washed sequentially with water (1 x 340 mL), saturated NaHCOj (1 x 340 mL) and brine (1 x 340 mL), dried and concentrated. The crude product (10.35 g, 99.6%) was obtained as a pale  173  white solid, which showed only one spot on TLC (hexanes-EtOAc, 6:4) and was thus used directly in the next step. An analytical sample of 137 was obtained by purifying on preparative TLC, eluting with hexanes-EtOAc (6:4).  Physical data of 12-acetyl-19-hydroxy-18(4—>'3)a6eo-abieta-3,8,ll,13-tetraen-18-oic acid lactone (137):  MeOH  White crystals; mp: 159-161 °C; UV Xj^l]^ (log s): 217.0 (4.43), 253.2(3.98); IR (CHCI3) Vmax cm-1: 2950 (CH), 1745 (lactone C=0), 1680 (C=0); iH NMR (400 MHz, CDCI3) 6: 1.03 (3H, s, H20), 1.19, 1.21 (3H each, both d, J = 6.9 Hz, H16, H17), 1.71 (IH, ddd, J = 12.5, 12.5, 6.2 Hz, Hla), 1.94 (2H, m, H6), 2.32-2.62 (3H, m. Hip, H2), 2.55 (3H, s, -COCH3), 2.70 (IH, m, H5). 3.03 (2H, m, H7), 3.45 (IH, sept, J = 6.9 Hz, H15), 4.77 (2H, br AB,, Av = 0.06 ppm, J = 17.1 Hz, H19), 7.12 (IH, s, H14), 7.44 (IH, s, H l l ) ; "C NMR (75.3 MHz, CDCI3) 5: 18.1, 20.0, 22.2, 24.1, 24.2, 28.3, 28.8, 30.6, 32.4, 36.3, 41.3, 70.5, 123.9, 125.0, 127.7, 136.7, 138.1, 142.2, 145.7, 162.5, 174.0, 203.1; EIMS m/z (rel. intensity): 338 (M", 28.7), 323 (91.2), 281 (8.5), 193 (6.6), 43 (base peak); HRMS calcd. for C22H26O3: 338.1882; found: 338.1879.  12-Acetoxy-19-hydroxy-18(4->3)fl6eo-abieta-3,8,ll?13-tetraen-18-oic (138)  acid  lactone  174  OAc  Xo a stirred solution of the acetyl lactone 137 (9.70 g) in CH2CI2 (50 mL) was added w-CPBA (80%, 12.8 g) in one portion and the mixture was cooled to -5 to 0 *^C. A catalytic amount of trifluroacetic acid (2.31 mL) was added. The mixture was then allowed to warm to room  temperature and stirred overnight (wrapped in aluminum foil to minimize decomposition of m-CPBA), until all the starting material was consumed (18 h, TLC: isopropyl ether, developed twice).  The reaction mixture was diluted with EtOAc (413 mL) and the  solution was washed with 10% Na2S03 (1 x 62 mL), saturated KHCO3 (1 x 62 mL), brine (1 x 62 mL) and dried, filtered and concentrated. The crude product (a whitish solid, 11.72 g) showed one major spot on TLC (isopropyl ether, developed twice) and was used without purification in the next step.  A pure sample for spectroscopic analyses was  purified by column chromatography, eluting with isopropyl ether.  Physical data of  12-acetoxy-19-hydroxy-18(4—>3)a6eo-abieta-3,8,ll,13-tetraen-18-oic  acid lactone (138):  A white powder; mp: 69-71 oC; UV X'^^'' (log s): 218.6 (4.22), 268.0 (3.23), 276.2 (3.22); IR (CHCI3) Vmaxcm-l; 2965 (CH), 1740 (C=0), 1680 (C=C), 1510 (C=C); IH NMR (400 MHz, CDCI3) 5: 1.01 (3H, s, H20), 1.17, 1.18 (3H each, both d, J = 6.9 Hz, H16, H17), 1.72 (IH, m, H l a ) , 1.91 (2H, m, H6), 2.30 (3H, s, -OCOCH3), 2.26-2.55 (3H, m. H i p , H2), 2.69 (IH, m, H5), 2.92 (IH, sept, J = 6.9 Hz, HI5), 2.99 (2H, m, H7),  175  4.76 (2H, br AB^, Av = 0.07 ppm, J - 17.1 Hz, H19), 6.92 (IH, s, Hll), 7.02 (IH, s, H14); ''C NMR (75.3 MHz, CDCI3) 6: 18.1, 20.1, 20.9, 22.2, 23.0 (2C), 27.1, 27.9, 32.5, 36.4, 41.0, 70.5 (C19), 118.1, 125.0, 127.7, 132.4, 137.9, 143.6, 146.2, 162.8, 169.9, 174.2 (C18); EIMS m/z (rel. intensity): 354 (M\ 4.3), 312 (45.8), 297 (12.5), 255 (5.6), 149(29.7), 110 (base peak); HRMScalcd. for C22H26O4: 354.1836; found: 354.1831.  12,19-Dihydroxy-18(4-»3)a6e<?-3,8,ll,13-tetraen-18-oic acid lactone (71)  Concentrated HCl (28.7 mL) was added dropwise at room temperature to a stirred solution of 138 (10.00 g) in MeOH (311 mL). The mixture was stirred in the dark o-—'  at room temperature until reaction was completed  according to TLC (about 6 h, TLC: isopropyl ether, developed twice). The mixture was then evaporated to remove MeOH. The aqueous suspension was extracted with EtOAc (400 mL), and the EtOAc layer was washed with saturated NaHCOj (3 x 75 mL), brine (2 X 75 mL), dried, filtered and concentrated. The resulting crude product was purified by column  chromatography  using  isopropyl  ether-hexanes  (9:1)  to  yield  pure  isotriptophenolide (71, 6.53 g, 85.1% overall yield from butenolide 65) as white crystals.  Physical data of 12,19-dihydroxy-18(4—>3)aZ?eo-3,8,ll,13-tetraen-18-oic acid lactone (71):  176  A white crystalline solid; mp: 196-198 «€ (dec); [a]^^: +45.6o (c = 0.411, MeOH); UV X^^^^ (log s): 224.0 (4.10), 285.0 (3.56); IR (thin film) v^ax cm-l; 3300 (OH), 2950 (CH), 1740 (C=0), 1660,1610,1500 (aromatic C=C); ^H NMR (400 MHz, CDCI3) 6: 1.00 (3H, s, H20), 1.22, 1.23 (3H, each, both d, J = 6.9 Hz, HI6, HI7), 1.69 (IH, m, Hla), 1.88 (2H, m, H6), 2.30-2.54 (3H, m, Hip, H2), 2.69 (IH, m, H5), 2.94 (2H, m, H7), 3.13 (IH, sept, J = 6.9 Hz, H15), 4.62 (IH, s, C12-0H, D^O exchangeable), 4.76 (2H, br AB^, Av = 0.07 ppm, J = 17.2 Hz, H19), 6.72 (IH, s, H l l ) , 6.90 (IH, s, H14); "C NMR (125.8 MHz, CDCI3) 5: 18.2 (C2), 20.3 (C6), 22.2 (C20), 22.5 (C16), 22.6 (C17), 26.8 (C15), 27.6 (C7), 32.6 (CI), 36.3 (CIO), 41.5 (C5), 70.5 (C19), 111.1 (Cll), 124.9 (C3), 126.3 (C13), 127.4 (C14), 132.7 (C8), 143.4 (C9), 151.0 (C12), 163.3 (C4), 174.4 (C18); ElMS m/z (rel. intensity): 312 (M", 86.8), 297 (base peak), 293 (33.0), 277 (8.5), 267 (18.7), 253 (33.7), 237 (12.3), 149 (58.1); HRMS calcd. for C20H24O3: 312.1725; found: 312.1720; Anal, calcd. for C20H24O3: C, 76.90; H, 7.73; found: C, 76.89; H, 7.78.  3.3  Preliminary experiments for oxidation of isotriptophenolide (ITP, 71)  I.  Oxidation with Fremy's salt  A solution of potasium nitrosodisulfonate (115.6 mg, 0.43 mmol) and KH2PO4 (35 mg, 0.26 mmol) in distilled water (10.5 mL) was added to a solution of 71 (50 mg.  177  0.16 mmol) in acetone (5 mL) under argon atmosphere. The mixture was stirred at room temperature for 3 h. The acetone was removed in vacuo and the residue was extracted with CH2CI2 (3x10 mL). The organic layer was washed with water and brine, dried, filtered and evaporated. The crude product was then purified by column chromatography, using gradient elution with hexanes-EtOAc (65:35 to 65:55) to give recovered starting material (4.9 mg, 9.8%), the quinone-aldehyde 145 (18.4 mg, 33.6%), 7-hydroxyisotriptophenolide (146, 17.9 mg, a:p, 1.3:1 by 'H NMR, 34% yield) and 5.3 mg of a mixture of unidentified compounds.  Physical data of the quinone-aldehyde 145*:  MeOH  Yellow flakes, mp: 181-183 °C; UV Cmax ^\;  (log  s): 219.0 (sh, 4.13), 257.2 (3.97); IR (CHCI3) v ,„,, cm"': 3050 (olefinic CH), 2875, 2740, 1720 (aldehyde C=0), 1745 (lactone C=0), 1645 (quinone C=0), 1600 (C=C); 'H NMR (400 MHz, CDCI3) 5: 1.10 (3H, s, H20),l.ll, 1.13 (3H each, both d, J = 6.0 Hz, H16, H17), 1.58 (IH, ddd, J=13.6, 5.7, 2.4 Hz, Hla), 2.23 (IH, m, H2p), 2.36 (IH, br d, J = 18 Hz, H2a), 2.50 (3H, m. Hip, overlapped with H6), 2.98 (IH, sept, J = 6.0 Hz, H15), 4.17 (IH, br d, J = 5.3 Hz, H5), 4.57 (2H, br ABq, Av = 0.26 ppm,J= 17.3 Hz, H19), 6.47 (IH, s, H14), 6.63 (IH, s, Hll), 9.75 (IH, s.  For the sake of simplicity the same numbering sequence as for the 18(4->3)fl6eo-abietane was used for this compound.  178  -CHO); "C NMR (75.3 MHz, CDC13) 6: 17.8 (C2), 18.5 (C20), 21.1 (C16), 21.3 (C17),26.4 (C15), 32.7 (CI), 34.1 (C5), 41.1 (CIO), 42.8 (C6), 71.2 (C19), 125.7 (C3), 132.7 (C14), 135.3 ( C l l ) , 150.8 (C13), 153.7 (C9), 160.8 (C4), 173.3 (C18), 187.1 (C12*), 187.9 (C8*), 199.2 (C7), (note: * interchangeable); EIMS m/z (rel. intensity): 342 (M^ 46.0), 313 (15.1), 298 (base peak), 283 (60.3), 254 (24.7), 239 (19.8), 225 (12.7), 211 (25.9), 192 (49.4), 163 (21.3), 152 (41.3), 128 (22.3), 115 (24.9), 91 (41.9), 29 (9.1), 28 (16.9); HRMS calcd. for C20H22O5: 342.1467; found: 342.1467.  Physical data of 7-hydroxy-isotriptophenolide (146)*:  OH  IR (thin film) Vmax cm"!: 3500 (OH), 2990 (CH), 1745 ( C - 0 ) , 1680 (C=C); 'H NMR (400 MHz, CDCl,) 6: (a-isomer, partial): 0.95 (3H, s, H20), 6.73 (IH, s, H l l ) , 7.17 (IH, s, H14); 'H NMR (400 MHz, CDCI3) 6: (P-isomer, partial): 1.10 (3H, s, H20), 2.75 (IH, br d, J =  13.8 Hz, H5), 6.69 (IH, s, H l l ) , 7.36 (IH, s, H14); EIMS m/z (rel. intensity): 328 (M^ 15.6), 310 (base peak), 295 (43.2), 267 (34.5), 253 (76.1).  * See below, Sects. 3.5.1 and 3.6.1 for the full spectral data of the pure diastereomers.  179  II.  Oxidation with ferric chloride  To a stirred solution of isotriptophenolide (ITP, 71) (40 mg, 0.13 mmol) in methanol-water (2:1, 6 mL) was added a solution of ferric chloride (1.2 mg, 0.007 mmol) in water (2 mL) and the mixture was heated to 60 °C for 9 h. The methanol was evaporated under vacuum and the product was precipitated by adding water (30 mL). The crude was then vacuum filtered, washed with water (20 mL) and purified by preparative TLC eluting twice with hexanes-EtOAc (1:1) to give the product 149 (24.1 mg, 58.1%) as a yellow solid, along with recovered starting material (4.9 mg, 12.2%).  Physical data of 12,19-dihydroxy-18(4—>3)a6eo-abieta-3,5,8,ll,13-pentaen-7-oxo-18-oic acid lactone (149): Yellow needles; mp: 287-289 ^C; UV X'' J  (log  s): 213.2 (4.31), 251.3 (4.16), 284.5 (4.02), 344.3 (3.66); IR (CHCI3) Vmax cm-l: 3495 (OH), 2980 (CH), 1745 (lactone C-0), 1645 (unsaturated C=0), 1600 (C=C); 'H NMR (400 MHz, CDCI3) 6: 1.29, 1.31 (3H each, both d, J = 7.0.Hz, H16, H17), 1.36 (3H, s, H20), 1.88 (IH, ddd, J = 13.3, 11.2, 6.2 Hz, Hla), 2.50 (IH, m. Hip), 2.61-2.76 (2H, m, H2), 3.18 (IH, sept, J = 7.0 Hz, H15), 5.04 (2H, br ABq, Av = 0. 17 ppm, J = 16.4 Hz, H19), 5.36 (IH, br s, OH), 6.31 (IH, s, H6), 6.90 (IH, s, H l l ) , 8.07 (IH, s, H14); LRMS m/z (rel. intensity): 324 (M^ 35.6), 309 (base  180  peak), 294 (6.2), 281 (16.3), 267 (12.9); HRMS calcd. for C20H20O4: 342.1362; found: 324.1359.  III.  Oxidation of ITP with DDQ  To a stirred solution of 71 (50 mg, 0.16 mmol) in dry methanol (5 mL) under argon was added DDQ (45 mg, 0.20 mmol). On completion of the reaction after 1 h, as evidenced by TLC (hexanes-EtOAc, 1:1), the reaction mixture was poured on ice (13 g) and stirred until the ice melted completely. The mixture was then extracted with EtOAc (3x15 mL), washed successively with 10% solution of sodium hydrosulphite, water, saturated sodium bicarbonate and brine.  This solution was dried over MgS04 and  evaporated in vacuo. The residue was then dissolved in hexanes-EtOAc (1:1) and filtered through a short pad of silica gel, eluting with the same solvent mixture. The solvent was removed and the crude product was purified in the chromatotron, eluting with hexanesEtOAc 6:4 to yield a diastereomeric mixture (2:1) of 7a and 7P-methoxyisotriptophenolide (150, 33.9 mg, 61.8%), and 7-keto-isotriptophenolide (151, 7.9 mg, 15.1%).  Physical data of 7-methoxy-isotriptophenolide (150)*  Full spectroscopic data for the pure 7a-methoxy-isomer are presented in the following section below.  181  IR (thin film) Vmax cm-l; 3550 (OH), 2960 (CH), 1744 (C=0); 'H NMR (400 MHz, CDCI3) 6: (a-isomer, partial): 3.44 (3H, s, -OCH3), 4.30 (IH, br s, H7p), 6.75 (IH, s, Hll), 7.13 (IH, s, H14); "H NMR (400 MHz, CDCI3) 5: (p-isomer, partial): 3.43 (3H, s, -OCH3), 4.64 (IH, dd, J = 8.8 Hz, H7a), 6.72 (IH, s, Hll), 7.28 (IH, s, H14); EIMS m/z (rel. intensity): 342 (M\ 12.1), 310 (base peak), 295 (59.3), 267 (32.2), 253 (87.5).  Physical data of  12,19-dihydroxy-7-oxo-18(4—>3)a6eo-abieta-3,8,ll,13-tetraen-18-oic  acid lactone (151):  OH  MeOH  White needles; mp: 286-288 °C (dec); UV X.max  (log s): 208.2 (4.40), 231.6 (4.38), 287.4 (4.20); IR (KBr) v„,, cm-': 3125 (br, OH), 2980 (CH), 1750 (lactone C=0), 1690 (ketone C=0), 1650, 1585 (aromatic C=C); 'H NMR (400 MHz, CDC13) 5: 1.13 (3H, s, H20), 1.25, 1.27 (3H each, both d, J = 7.0 Hz, HI6, H17), 1.80 (IH, m, Hla), 2.43 (2H, m. Hip, H2p), 2.57 (IH, m, H2a), 2.68 (2H, d, J = 11.7 Hz, H6P; overlapped d, J = 7.4 Hz, H6a), 3.14 (IH, sept, J = 7.0 Hz, H15), 3.17 (IH, m, H5), 4.75 (2H, m, H19), 5.37 (IH, s, C12-0H, D20 exchangeable), 6.78 (IH, s, Hll), 7.97 (IH, s, H14); EIMS m/z (rel. intensity): 326 {M\ 1.8), 310 (54.3), 295 (base peak), 253 (23.7), 225 (8.3), 213 (38.9); HRMS calcd. for C20H22O4: 326.1518; found: 326.1511.  182  3,4  Synthesis of 8-methoxy-dienones  Diastereomeric  mixture  of  8-methoxy-19-hydroxy-12-oxo-18(4^^3)flZ»eo-abieta-  3,9(ll),13-trien-18-oic acid lactones (153)  To a stirred solution of isotriptophenolide (71) (600 mg, 1.92 mmol) in MeOH (40 mL) under argon, PIDA (740.4 mg, 2.3 mmol) was added. The mixture was stirred at room temperature for 1.5 h. After all the starting material was consumed NaHCOj (160 mg) was added and the suspension was stirred for 10 min; and the mixture was evaporated in vacuo to remove MeOH. The residue was then resuspended in EtOAc and filtered thorough a cotton plug to remove NaHCOj. The solution was evaporated to dryness. The crude product (893 mg) was partly purified by the chromatotron, eluting with a mixture of chloroform-methanol (99:1) to obtain three semi-pure fractions containing 8-methoxydienones, 7a-methoxy-ITP and 7-keto-ITP respectively. Each of these fractions was then purified separately by column chromatography eluting with hexanes-EtOAc (65:35) to give: a pure mixture of 8-methoxydienones (153, 298.8 mg, a:p = 29:71 by ^H NMR , 45.4% yield), 7a-methoxy-isotriptophenohde (150a, 172.2 mg, 26.4 % yield) and 7-keto-isotriptophenohde (151, 89.8 mg, 14.3 % yield).  Physical data of the 8-methoxy-dienones 153*  183  IR (thin film) Vmax cm-1: 2960 (CH), 1745 (lactone C=0), 1665 (dienone C=0), 1640 (C=C), 1240 f^OMe  (C-O-C);  'H NMR (400 MHz, CDCI3) 5: (a-isomer,  partial): 1.04 (3H, s, H20), 2.98 (3H, s, -OCH3), 6.12 (IH, s, H14), 6.39 (IH, s, Hll); 'H NMR (400 MHz, CDCI3) 5: (p-isomer, partial): 1.24 (3H, s, H20), 3.0 (3H, s, -OCH3), 4.70 (2H, br s, H19), 6.34 (IH, s, H14), 6.24 (IH, s, Hll); EIMS m/z (rel. intensity): 342 (M^ 19.3), 310 (81.2), 300 (23.5), 295 (base peak), 285 (11.4), 267 (21.4), 179 (38.5), 147 (60.0), 43 (66.2).  Physical data of  7a-methoxy-12,19-dihydroxy-18(4->3)a6eo-abieta-3,8,ll,13-tetraen-  18-oic acid lactone (150a):  A white powder; mp: 109-111 ^C; [a] ^^ : +54.5^ MPOH  (c = 0.110, CHCI3); UV X^^^  (log s): 204.2 (sh, 4.73),  220.5 (4.31); 279.7 (3.45); IR (CHCI3) Vmax cm-^: 3580 (OH), 2950 (CH), 1740 (C=0), 3010, 1610 (aromatic C=C); 1200, 1058, 1010; iH NMR (400 MHz, CDCI3) 5: 0.94 (3H, s, H20), 1.23, 1.25 (3H each, both d, J = 6.9 Hz, H16, H17), 1.71 (IH, m, Hla), 1.96 (IH, td, J = 13.5, 4.1 Hz, H6p), 2.07 (IH, dt, J= 13.5, 2 Hz, H6a), 2.36 (2H, m. Hip, H2p), 2.48 (IH, br d, J  Complete spectroscopic data of the pure 8a-diastereomer are provided below in this section.  184  = 13.1 Hz, H2a), 3.15 (2H, sept, J = 6.9 Hz, H15, and overlapped m, H5), 3.44 (3H, s, OCH3), 4.27 (IH, d, J = 3.9 Hz, H7P), 4.78 (2H, br AB^, Av = 0.09 ppm, J = 17.0 Hz, H19), 5.21 (IH, s, OH, D p exchangeable), 6.71 (IH, s, Hll), 7.10 (IH, s, H14); ''C NMR (75.3 MHz, CDCI3) 6: 18.1, 21.9, 22.4, 22.5, 22.8, 25.3, 27.8, 32.5, 36.5, 37.0, 56.5 (-OMe), 70.6, 76.0 (C7), 110.9, 125.0, 129.2, 133.0, 143.5, 153.6, 163.9, 174.6 (C18); EIMS m/z (rel. intensity): 342 {M\ 23.8), 310 (base peak, M-MeOH ), 295 (68.6), 279 (6.3), 267 (46.9), 253 (96.0), 213 (12.5), 207 (13.4), 179 (13.2), 43 (53.5); HRMS calcd. for C2,H2604: 342.1831; found: 342.1826; Anal, calcd. for C2,H2604 : C, 73.66; H 7.65; found:C, 73.61, H, 7.85.  8a-Methoxy-19-hydroxy-12-oxo-18(4—>3)«Z»eo-abieta-3,9(ll),13-trien-18-oic  acid  lactone (153a)  To a stirred solution of diastereomeric mixture of 8-methoxy-dienones 153 (200 mg, 0.58 mmol) in benzene (60 ml), 1 ml of 90% ^butyl hydroperoxide (9.6 mmol) was added, followed by 0.24 ml of triton-B (40% in methanol, 0.58 mmol). The solution was stirred to 40 "C for 21 h until no more starting material was consumed as seen by TLC (hexanes-ethyl acetate, 4:6). The reaction mixture was diluted with 20 ml of benzene and washed with water and brine. The combined aqueous layers were re-extracted with ether and this extract combined to the main organic phase. The combined extracts were dried, filtered and evaporated. The crude product was purified by preparative TLC eluting twice with hexanes-EtOAc (55:45) to give 54 mg (93% yield, considering that the  185  Starting material contained 29% of the a-isomer) of diastereomerically pure 8a-methoxydienone 153a and 30.4 mg (21.4% yield, based on a 71% contents of the P-isomer in the starting material) of the isomerized 5P-8P-methoxy-dienone 154.  Physical data of 8a-methoxy-19-hydroxy-12-oxo-18(4^3)aZ)eo-abieta-3,9(ll),13-trien18-oic acid lactone (153a):  A white solid, mp: 203-205 ^C; UV X^^^^ (log max  s): 202.8 (4.54), 221.2 (4.31); IR (CHCI3) v^ax cm-l; 2964 (CH), 1749 (lactone C=0), 1668 (dienone C=0), 1635 (C=C); 1059 (C-O-C); iH NMR (400 MHz, CDCI3) 6: 1.04 (3H, s, H20), 1.05, 1.09 (3H each, both d, J = 6.9 Hz, H16, H17), 1.59 (IH, dd, J = 11.0, 12.8 Hz, H6P), 1.75 (2H, br ddd, J = 14.9, 6.8, 5.8 Hz, H7p, and overlapped m, Hla), 2.05 (2H, dd, J = 12.8, 12.5 Hz, H6a, and overlapped m, Hip), 2.18 (IH, t, J = 12.5 Hz, H7a), 2.36 (IH, m, H2P), 2.45 (IH, br d, J = 16.7 Hz, H2a), 2.98 (4H, br s, -OCH3 , H15), 3.72 (IH, br s, H5), 4.71 (2H, br ABq, Av = 0.10 ppm, J = 18.9 Hz, H19), 6.12 (IH, s, H14), 6.39 (IH, s, Hll); '^C NMR (75.3 MHz, CDCI3) 6: 17.4 (C6), 17.9 (C2), 21.6 (C16), 22.0 (C17), 22.6 (C20), 26.0 (C15), 29.0 (C7), 33.1 (CI), 35.1 (C5), 38.5 (CIO), 50.9 (-OCH3), 70.1 (C19), 73.4 (C8), 123.9 (C3), 128.6 (Cll), 140.4 (C14), 146.6 (C13), 162.2 (C9*), 163.0 (C4*), 173.7 (C18), 186.2 (C12) (note: * interchangeable); EIMS m/z (rel. intensity): 342 (M^ 19.3), 310 (81.2, M-MeOH ), 295 (base peak), 285 (11.4), 267 (21.4), 191 (11.4), 179 (38.5),  186  165 (21.2), 147 (60.0), 137 (31.7), 119 (29.1), 115 (20.2), 43 (38.0); HRMS calcd. for Cj.HjiOA- 342.1831; found: 342.1840; Anal, calcd. for C2,H2604 : C, 73.66; H 7.65; found: C, 73.76, H, 7.68.  Physical data of 5P-8|3-Methoxy-19-hydroxy-12-oxo-18(4->3)a6eo-abieta-3,9(ll),13trien-18-oic acid lactone (154):  A white solid, mp: 162-164 «€; [a]^^ : -16.7° (c MeOH  0.119, CHCI3); UV X^^^  (log s): 220.5 (4.23), 240.6  (sh, 4.06); IR (CHCI3) v^ax cm-l; 2937 (CH), 1757 (lactone C=0), 1668 (dienone C=0), 1637 (C=C); 1064 (C-O-C); IH NMR (400 MHz, CDCI3) 5: 1.04, 1.09 (3H each, both d, J = 6.9 Hz, H16, H17), 1.32 (IH, td, J = 13.4, 4.7 Hz, H7a), 1.39 (3H, s, H20), 1.59 (2H, m, Hla, H6p), 1.98 (IH, dt, J = 13.4, 3.8 Hz, H7P), 2.20 (IH, dd,J= 14.4, 3.4 Hz, Hip), 2.27 (2H, m, H2), 2.55 (IH, m, H6a), 2.71 (IH, br s, H5), 2.96 (4H, br s, -OCH3, H15), 4.64 (2H, br s, H19), 6.14 (IH, s, H14), 6.36 (IH, s, Hll); ''C NMR (75.3 MHz, CDCI3) 5: 17.4 (C2), 20.0 (C6), 21.7 (C16), 21.8 (C17), 25.9 (C15), 27.4 (C20), 33.3 (CI), 33.8 (C7), 39.1 (CIO), 41.9 (C5), 51.1 (-OCH3), 70.4 (C19), 72.5 (C8), 127.2 (C3), 128.9 (Cll), 142.4 (C14), 146.8 (C13), 157.1 (C9), 161.5 (C4), 173.6 (C18), 186.1 (C12); EIMS m/z (rel. intensity): 342 (M^ 41.1), 311 (62.7), 300 (60.4), 295 (8.0), 271 (4.3), 255 (5.7), 239 (4.0), 211 (6.0), 191 (26.1), 178 (base peak), 163 (32.3), 147 (43.0), 137 (23.3); HRMS  187 calcd. for C2,H2604: 342.1831; found: 342.1842; Anal, calcd. for C2,H2604 : C, 73.66; H 7.65;found:C, 73.39, H, 7.51.  3.5  Synthesis of the (7,8)a-epoxy series  3.5.1  Syntheses of 7a-hydroxy-isotriptophenoIide (146a)  7a,12,19-trihydroxy-18(4—>3)a6eo-abieta-3,8,ll?13-tetraen-18-oic acid lactone (146a) OH  Method A (from 7a-niethoxy-isotriptophenolide, 150a):  A solution of BCI3 (0.59 mL of IM solution, 0.59 mmol, diluted with 23 mL of anhydrous CH2CI2) was added dropwise over a period of 30 min. to a solution of 7amethoxy-isotriptophenolide (150a) (234.6 mg, 0.68 mmol) in methylene chloride (47 mL) at -75 "C under argon. The reaction mixture was stirred at this temperature for 1.5 h and then quenched by pouring it quickly over a rapidly stirring saturated sodium bicarbonate solution (70 mL) at 0-5 °C.  The mixture was stirred for 10 min then  extracted with CH2CI2, washed with water and brine; dried, filtered and evaporated. The  crude product was purified by preparative TLC eluting with hexanes-EtOAc (30:70) to give 7a-hydroxy-isotriptophenolide (146a, 100.8 mg, 61.7% yield) along with some of the corresponding dehydrated product, (6,7)-dehydro-ITP, 159 (28.5 mg, 13.4%).  Physical data of 7a,12,19-trihydroxy-18(4-»3)aZ7eo-abieta-3,8,ll,13-tetraen-18-oic acid lactone (146a):  A white powder, mp: 131-132.5 ^C; [a]^": +27° (c = 0.056, CHCl,); UV ^J^l^ (log s): 204.7 (4.66), 221.0 (4.29) 280.1 (3.45); IR (CHCI3) Vmax cm-^: 3375 (OH), 2963 (CH), 1734 (C=0), 1672, 1619, 1510 (aromatic C-C) 1022 (C-0); 'H NMR (400 MHz, CDCI3) 6: 0.95 (3H, s, H20), 1.24, 1.26 (3H each, both d, J = 6.8 Hz, H16, H17), 1.72 (IH, m, Hla), 1.98 (IH, dt, J= 13.7, 2.1 Hz, H6a), 2.10 (IH, td, J = 13.7, 4.2 Hz, H6p), 2.32-2.56 (3H, m. Hip, H2), 3.15 (2H, m, H5, H15), 4.78 (2H, br ABq, Av = 0.12 ppm, J = 17.2 Hz, H19), 4.85 (IH, m, H7p), 5.14 (IH, s, C14-0H, D^O exchangeble), 6.73 (IH, s, Hll), 7.17 (IH, s, H14); ••''C NMR (75.3 MHz, CDCI3) 6: 18.3 (C2), 22.1 (C20), 22.4 (C16), 22.6 (C17), 26.9 (C15), 29.4 (C6), 32.6 (CI), 36.5 (C5), 66.9 (C7), 70.5 (C19), 111.2 (Cll), 125.4 (C3), 128.5 (C13), 128.7 (C14), 133.5 (C8), 143.8 (C9), 153.2 (C12), 163.2 (C4), 174.1 (CI8); LRMS m/z (rel. intensity): 328 (M\ 23.3), 310 (base peak), 295 (56.2), 285 (14.8), 267 (36.1), 253 (74.3), 239 (3.5); HRMS calcd. for C20H24O4: 328.1674; found: 328.1673.  189  Physical data of 12,19-dihydroxy-18(4->3)a6eo-abieta-3,6,8,ll,13-pentaen-18-oic acid lactone (159):  A reddish powder; IR (CHCI3) Vmax cm-l; 3525 (OH), 2985 (CH), 1745 (C=0), 1680, 1625, 1610, 1590 (aromatic C=C); ^H NMR (400 MHz, DMSO-d^) 5: 0.95 (3H, s, H20), 1.21, 1.23 (3H each, both d, J = 6.9 Hz, H16, H17), 1.86 (IH, m, Hla), 2.25-2.49 (3H, m. Hip, H2), 3.29 (IH, sept, J= 6.9 Hz, H15), 3.54 (IH, br s, H5), 5.01 (2H, br ABq, Av = 0.10 ppm, J = 17.2 Hz, H19), 5.85 (IH, dd, J - 9.4, 2.5, H6), 6.62 (IH, dd, J = 9.4, 3.3, H7), 6.88 (IH, s, H l l ) , 7.02 (IH, s, H14). EIMS m/z (rel. intensity): 310 (M^ 41.6), 295 (27.8), 293 (9.2), 267 (33.6), 253 (base peak), 167 (11.5). HRMS calcd. for C20H22O3: 310.1569 found: 310.1560.  Method B (from isotriptophenolide, 71):  To a solution of 71 (100 mg, 0.32 mmol) in acetonitrile (30 ml) and water (10 mL) was slowly added (c.a. 10 min.) a solution of PIDA (113.5 mg, 0.35 mmol) in acetonitrile (10 ml). The mixture was stirred at room temperature for 30 min. and then water (25 ml) was added. The reaction mixture was then extracted with methylene chloride, washed with water, saturated sodium bicarbonate and brine. The solution was dried, filtered and evaporated in vacuo. The crude mixture was then purified by the  190  chromatotron using ethyl acetate-hexanes (30:70) to give two fractions. These were then re-purified by preparative TLC in methylene chloride-acetone 92:8 to yield four products, namely 7a-hydroxy-ITP (146a, 63.2 mg, 60.1%), 7-oxo-ITP (151, 9.3 mg, 8.9%), 8ahydroxy-dienone 157 (8.9 mg, 8.5%) and 8p-hydroxy-dienone 158 (18.7 mg, 17.8%).  Physical data of  8a,19-dihydroxy-12-oxo-18(4->3)a6eo-abieta-3,9(ll),13-trien-18-oic  acid lactone (157):  White crystals, mp: 196-198 «€; [a]^^ : -164.8° (c = 0.108, CHCI3); UV X^l^^  (log s): 229.2 (4.23); IR  (CHCI3) Vmax cm-1: 3397 (OH), 2961 (CH), 1742 (lactone C=0), 1670 (dienone C=0), 1635 (olefmic C=C); IH NMR (400 MHz, CDCI3) 5: 1.03, 1.06 (3H each, both d, J = 6.8 Hz, HI6, H17), 1.03 (3H, s, H20), 1.63 (IH, m, H6a), 1.77 (2H, m, Hla, H7p), 1.85 (IH, s, C8OH, D2O exchangeable), 2.09 (2H, m. Hip, H6p), 2.18 (IH, ddd, J = 13.0, 10.8, 2.2 Hz, H7a), 2.33 (IH, m, H2p), 2.43 (IH, br d, J = 15.4 Hz, H2a), 2.89 (IH, sept, J = 6.8 Hz, H15), 3.96 (IH, br s, H5), 4.71 (2H, br ABq, Av = 0.09 ppm, J = 17.2 Hz, H19), 6.17 (IH, s, Hll), 6.38 (IH, s, H14); ''C NMR (75.3 MHz, CDCI3) 6: 17.4 (C6), 18.0 (C2), 21.4 (C16), 21.7 (C17), 22.1 (C20), 25.8 (C15), 28.6 (C7), 33.9 (CI), 35.1 (C5), 38.5 (CIO), 68.3 .(C8), 70.1 (C19), 123.8 (C3), 124.8 (Cll), 142.7 (C14), 142.8 (C13), 163.2 (C4*), 165.6 (C9*), 173.9 (C18), 186.2 (C12) (note: * interchangeable); DCI m/z (rel. intensity): 346 (M + NH/, 18), 330 (47), 313 (base peak), 297 (32), 269 (8), 165 (8), 149  191 (3); HRMS calcd. for C20H25O4: 329.1753; found: 329.1750. Anal, calcd. for C20H24O4 : C, 73.13; H 7.37; found: C, 72.87, H, 7.31.  Physical data of  8p,19-dihydroxy-12-oxo-18(4^3)aZ>eo-abieta-3,9(ll),13-trien-18-oic  acid lactone (158):  White crystals, mp: 106-108 «€; [a]^^ : -179.2° MeOH  (c = 0.159, CHCI3); UV X j ^ " (log s): 225.5 (4.19); IR (CHCI3) Vmax cm-1: 3416 (OH), 2963 (CH), 1747 O—  (lactone C=0), 1668 (dienone C=0), 1635 (olefinic  C=C); IH NMR (400 MHz, CDCI3) 5: 1.04, 1.06 (3H each, both d, J = 6.9 Hz, HI6, HI 7), 1.29 (3H, s, H20), 1.53 (IH, td, J = 13.6, 4.5 Hz, H7p), 1.64 (2H, m, HI a, H6a), 1.90 (IH, s, C8-0H, D2O exchangeable), 2.06 (2H, m, Hip, H6p), 2.21 (IH, dt, J = 13.6, 2.5 Hz, H7a), 2.35 (2H, br d, J = 11.4 Hz, H2p, H5), 2.47 (IH, br d, J - 16 Hz, H2a), 2.90 (IH, sept, J = 6.9 Hz, H15), 4.71 (2H, br s, H19), 6.09 (IH, s, Hll), 6.41 (IH, s, H14);"CNMR(75.3MHz, C D C y S : 16.2 (C20), 17.9 (C2), 19.2 (C6), 21.4 (CI6), 21.6 (C17), 25.7 (C15), 32.5 (CI), 38.8 (C7), 39.9 (CIO), 45.4 (C5), 69.1 (C8), 70.2 (C19), 123.9 (Cll), 124.8 (C3), 142.5 (C13), 144.7 (C14), 161.3 (C9*), 163.6 (C4*), 173.6 (C18), 186.3 (C12) (note: * interchangeable); DCI m/z (rel. intensity): 346 (M + NH4\ base peak), 330 (84), 313 (52), 297 (8), 269 (6), 165 (4); HRMS calcd. for C20H25O4: 329.1753; found: 329.1750. Anal, calcd. for C20H24O4: C, 73.13; H 7.37; found: C, 73.15, H, 7.43.  192  3.5.2  Syntheses of the (7,8)a-epoxy-dienone, 72  (7,8)cx-epoxy-19-hydroxy-12-oxo-18(4^3)aZieo-abieta-3,9(ll),13-trien-18-oic  acid  lactone (72)  Method A (from 7a-hydroxy-isotriptophenolide, 146a):  A solution of 7a-hydroxy-isotriptophenolide (50 mg, 0.15 mmol) in acetonitrile (12.5 ml) was slowly added (c.a. 10 min.) to a stirred solution of PIDA (57.5 mg, 0.18 mmol) in acetonitrile (10 ml) kept at 0-5 ^C and under argon atmosphere. Stirring was continued for 20 min and then water (25 ml) was added. The reaction mixture was extracted with ether, washed with water, saturated sodium bicarbonate and brine. The solution was dried, filtered and evaporated.  The crude mixture was purified by  preparative TLC eluting twice with mixture methylene chloride-methanol (97:3) to give 10.7 mg (21.6%) of (7,8)a-epoxy-dienone 72, 26.7 mg (53.7%) of 7-oxo-ITP (151), and 1.5 mg of compound 145.  193  Method B (from 7a-hydroxy-isotriptophenolide, 146a):  7a-hydroxy-ITP (146a, 60 mg, 0.18 mmol) was dissolved in trifluoroethanol (12 mL) and cooled to 0-5 °C under argon. KOH (21.3 mg, 0.38 mmol) was added and the mixture was stirred for 5 min followed by addition of a solution of PIDA (70 mg, 0.22 mmol) in trifluoroethanol (6 mL). The reaction mixture was stirred to 0-5 *^C for 20 min, until all the starting material was consumed as visualized by TLC (CHjClj-acetone, 96:4). Then the reaction mixture was evaporated to dryness under vacuum. The residue was redissolved in ethyl acetate, filtered through a short pad of silica gel and eluted with additional amounts of ethyl acetate. The combined filtrate was then evaporated under vacuum and the crude was purified by preparative TLC eluting twice with methylene chloride-methanol (97:3) to give (7,8)a-epoxy-dienone (72, 17 mg, 28.6 %) and 7-oxoisotriptophenolide (151, 33.4 mg, 56 %).  Physical data of (7,8)a-epoxy-19-hydroxy-12-oxo-18(4^'3)fl6eo-abieta-3,9(l l),13-trien18-oic acid lactone (72):  White needles; mp: 187-188 ^C (dec); [a]^^: +169.3° (c = 0.127, CHCI3); UV ^mlx^ (log e): 219.1 (4.13), 258.1 (4.19); IR (CHCI3) Vmax cm-1: 3020 (olefinic CH), 2969 (CH), 1752 (lactone C=0), 1662 (quinone C-0), 1636 (C=C), 1057, 1028, 906; 'H NMR (400 MHz, CDCI3) S: 0.95 (3H, s, H20),1.05, 1.08 (3H each, both d, J = 6.9 Hz, H16, H17), 1.69 (IH, br ddd, J = 13.3, 12.5, 6.4 Hz, Hla), 2.02 (IH, br dd, J = 14.4,  194  12.8 Hz, H6p), 2.19 (IH, br dd, J = 13.3, 6.2 Hz, Hip), 2.31 (IH, m, H2p, overlapped with H6a), 2.30 (IH, br ddd, J = 14.4, 3.2, 2.2 Hz, H6a), 2.47 (IH, br d, J = 17.9 Hz, H2a), 2.95 (IH, sept, J = 6.9 Hz, H15), 3.23 (IH, br d, J = 12.7 Hz, H5), 3.81 (IH, d, J = 2.4 Hz, H7P), 4.70 (2H, br ABq, Av = 0.08 ppm, J = 17.2 Hz, H19), 5.99 (IH, d, J = 0.9 Hz, H14), 6.48 (IH, s, Hll); '^C NMR (75.3 MHz, CDCI3) 5: 17.6 (C2), 21.4 (C16), 21.5 (C17), 22.2(C20), 23.6 (C6), 26.4 (C15), 30.9 (CI), 32.4 (C5), 37.3 (CIO), 54.3 (C8), 60.9 (C7), 69.9 (C19), 125.3 (C3), 130.4 (Cll), 137.7 (C14), 150.0 (C13), 157.3 (C9), 160.8 (C4), 173.2 (C18), 186.0 (C12); EIMS m/z (rel. intensity): 326 (M^ 68.6), 311 (base peak), 310 (35.5), 295 (25.7), 279 (9.5), 267 (14.6), 253 (30.0), 229 (11.6), 203 (26.8), 191 (6.5), 169 (14.5), 163 (23.4), 141 (20.0); HRMS calcd. for C20H22O4: 326.1518; found: 326.1510; Anal, calcd. for C20H22O4: C, 73.59; H, 6.80; found: C, 73.46; H, 6.62.  Method C (from isotriptophenolide, 71):  A solution of PIDA (1.045 g, 3.1 mmol) in acetonitrile (68 mL) was added in one portion to a solution of 71 (500 mg, 1.5 mmol) in acetonitrile (150 mL) and water (50 mL) under argon. The reaction mixture was stirred for 30 min. and then a 2.5% solution of NaHCOj (150 mL) was added. The reaction mixture was extracted with CH2CI2, washed with brine, dried, filtered and evaporated. The residue was suspended in EtOAc and filtered through a short pad of silica gel, washing with EtOAc. The filtrate was evaporated and the resulting crude product was purified by the chromatotron, eluting with methylene chloride-acetone (98:2) to give two fractions, the less polar one containing the  195  semi-pure (7,8)a-epoxy-dienone 72 and the other was a mixture of three compounds. These fractions were re-purified separately in the chromatotron. The dienone fraction was eluted with methylene chloride-acetone (99:1) to yield pure (7,8)a-epoxy-dienone 72 (104.7 mg, 20.0%).  The second fraction was eluted with a mixture of methylene  chloride-acetone-ether (95: 2: 3), yielding 151 (170 mg, 32.5%)), 8a-hydroxy-dienone 157 (42.5mg, 8.1%), and 8p-hydroxy-dienone 158 (94.1 mg, 17.9%). Also, a very small amount (3.1 mg) of the quinone-aldehyde 145 was isolated.  3.5.3  Epoxidation of (7,8)a-epoxy-dienone, 72  The (7,8)a-epoxy-dienone (72, 30 mg, 0.09 mmol) was dissolved in toluene (5 mL) and cooled down to 0 *^C under argon. A freshly prepared solution of trimethyl benzyl ammonium ^butylperoxide (0.12 ml, 1.2M in toluene) was added dropwise. The reaction mixture was stined for 30 min to 0 °C and then allowed to slowly reach room temperature.  The progress of the reaction was monitored by HPLC (see appendix,  method 1). The reaction was completed after 4 h , then the mixture was diluted with EtOAc and filtered through a short pad of silica gel, washing with ethyl acetate (40 mL). The solvent was removed and the residue purified by preparative TLC eluting with hexanes-EtOAc, (45:55) to yield the diepoxides 160 (9.1 mg, 29%) and 161 (9.7 mg, 30.9%).  196  Physical data of  (7,8)a,(13,14)p-bis(epoxy)-19-hydroxy-12-oxo-18(4^3)flZ7eo-abieta-  3,9(1 l)-dien-18-oic acid lactone (160): A white crystalline solid; mp: 211-213 ^C; [a] Q : MeOH  -241.00 (c = 0.105, CHCI3); UV X^l^  (log e): 222.8  (4.28), 256.0 (sh, 3.87); IR (CHCI3) Vmax cm-': 3020 (olefinic CH), 2970 (CH), 1752 (lactone C=0), 1681 (enone C=0), 1628 (C=C), 1172, 1083, 1027; 'H NMR (400 MHz, CDCI3) 6: 0.94, 0.99 (3H each, both d,J= 6.9 Hz, H16, H17), 0.94 (3H, s, H20), 1.56 (IH, m, Hla), 1.97 (IH, ddd, J = 14.5, 13.1, 1.4 Hz, H6p), 2.12 (IH, br dd, J = 13.2, 5.9 Hz, Hip), 2.25 (IH, m, H2p), 2.31 (IH, ddd, J = 14.5, 4.1, 2.3 Hz, H6a), 2.47 (IH, br d, J = 18.4 Hz, H2a), 2.55 (IH, sept, J = 6.9 Hz, H15), 2.87 (IH br d, J = 13.1 Hz, H5a), 3.06 (IH, s, H14), 3.80 (IH, s, H7p), 4.69 (2H, m, H19), 6.16 (IH, s, Hll); ''C NMR (125.8 MHz, CDCI3) 5: 16.6 (C16), 17.4 (C2), 18.3 (C17), 20.0 (C20), 23.7 (C6), 24.8 (C15), 30.0 (CI), 32.6 (C5), 36.6 (CIO), 54.2 (C8), 57.6 (C7), 58.0 (C14), 63.6 (C13), 69.9 (C19), 125.5 (C3), 126.1 (Cll), 153.7 (C9), 159.9 (C4), 173.1 (C18), 194.2 (C12); EIMS m/z (rel. intensity): 342 {M\ 3.3), 326 (5.0), 314 (15.9), 299 (94.7), 272 (13.8), 256 (11.4), 128 (8.9), 115 (10.7), 91 (20.6), 71 (36.2), 65 (13.3), 43 (base peak); HRMS calcd. for C20H22O5: 342.1467; found: 342.1461; Anal, calcd. for C20H22O5: C, 70.16; H, 6.48; found: C, 69.97; H, 6.43.  197  Physical data of 5p-(7,8)a,(13,14)p-bis(epoxy)-19-hydroxy-12-oxo-18(4^3)<36eo-abieta3,9(1 l)-dien-18-oic acid lactone (161):  Fine white needles; mp: 210-212 ^C; [ a ] ^ : -27.30 (c = 0.110, CHCI3); UV X^l^^  (log s): 222.8  (4.23), 256.5 (sh, 3.85); IR (thin film) Vmax cm-l; 2969 (CH), 1748 (lactone C=0), 1676 (enone C=0), 1057, 1023; 'H NMR (400 MHz, CDCI3) 6: 0.90, 0.99 (3H each, both d,J= 6.8 Hz, H16, H17), 1.24 (3H, s, H20), 1.70 (IH, ddd, J = 14.1, 11.5, 6.5 Hz, Hla), 2.17 (2H, m, H2), 2.28 (2H, m. Hip, H6a), 2.48 (3H, sept, J = 6.8 Hz, H15, and overlaped m, H5p, H6P), 2.96 (IH, s, H14), 3.71 (IH, d, J = 1.3 Hz, H7p), 4.73 (2H, br ABq, Av = 0.24 ppm, J = 16.3 Hz, H19), 6.22 (IH, s, Hll); ''C NMR (125.8 MHz, CDCI3) 6: 16.3 (C16), 17.3 (C2), 18.4 (C17), 23.0 (C6), 25.0 (C15), 29.4 (C20), 33.5 (CI), 37.2 (CIO), 37.7 (C5), 52.9 (C8), 57.7 (C7), 59.0 (C14), 63.7 (C13), 70.4 (C19), 123.6 (C3), 129.0 (Cll), 152.3 (C9), 163.3 (C4), 173.5 (C18), 193.1 (C12); EIMS m/z (rel.intensity): 342 (M\ 1.6), 326 (2.6), 314 (63.2), 299 (16.5), 286 (6.7), 272 (15.3), 257 (5.2), 243 (9.6), 229 (8.3), 215 (5.3), 201 (4.3), 187 (49.9), 173 (6.9), 159 (29.6), 115 (8.7), 91 (18.3), 43 (base peak); HRMS calcd. for C20H22O5: 342.1467; found: 342.1470; Anal, calcd. for C20H22O5: C, 70.16; H, 6.48; found: C, 70.36; H, 6.52.  198  3.6  Synthesis of the (7,8)P-epoxy series  3,6.1  Synthesis of 7p-hydroxy-isotriptophenolide (146b)  12-Acetoxy-19-hydroxy-7-oxo-18(4—>3)a6eo-abieta-3,8,ll5l3-tetraen-18-oic  acid  lactone (162) To a stirred solution of 12-acetoxy butenolide, OAc  138 (3.70 g, 10.44 mmol) in glacial HOAc (185 mL), a freshly prepared solution of CrOj (7.78 g, 77.8 mmol) in aqueous HOAc (90%. 500 mL) was added slowly in 3 O-—'  portions (166.7 mL each portion). The reaction mixture  was allowed to stirrer for 2 h after each addition (TLC, hexanes-EtOAc, 1:1) and then isopropanol (75 mL) was added followed by stirring for another 15 min. The reaction was poured into water (1480 mL) and the aqueous solution was extracted with CHCI3 (4 x 740 mL). The combined CHCI3 extract was washed with saturated NaHCO, (2 x 200 mL), brine (200 mL), dried and concentrated. The crude product was purified by column chromatography using hexanes-EtOAc (7:3) to give 162 as a yellowish powder (2.08 g, 54.2 %).  Physical data of  12-acetoxy-19-hydroxy-7-oxo-18(4->3)a6eo-abieta-3,8,ll,13-tetraen-  18-oic acid lactone (162):  199  A yellowish powder; mp: 181-183 ^C; 'H NMR (400 MHz, CDCI3) 5: 1.16 (3H, s, H20), 1.21, 1.23 (3H each, both d, J - 7.0 Hz, H16, H17), 1.83 (IH, m, Hla), 2.34 (3H, s, -CO2-CH3), 2.43 (IH, m, H2p), 2.47 (IH, dd, J - 12.6, 6.1 Hz, Hip), 2.57 (IH, br d, J = 17.4 Hz, H2a), 2.73 (IH, d, J = 11.9 Hz, H6P), 2.74 (IH, d, J = 7.2 Hz, H6a), 3.01 (IH, sept, J = 7.0 Hz, H15), 3.23 (IH, m, H5), 4.76 (2H, m, H19), 7.10 (IH, s, Hll), 8.06 (IH, s, H14); ''C NMR (75.3 MHz, CDC 13) 5: 17.7, 21.0, 21.7, 22.7 (2C), 27.3, 31.6, 36.3, 36.5, 40.8, 70.0, 118.1, 125.9, 127.8, 129.2, 139.5, 150.2, 152.8, 160.0, 169.0, 173.3, 194.6; EIMS m/z (rel. intensity): 368 (M^ 15.4), 326 (86.3), 311 (base peak), 298 (10.2), 229 (35.1), 203 (70.4), 163 (80.0); HRMS calcd. for C22H24O5: 368.1624; found: 368.1620.  12,19-dihydroxy-7-oxo-18(4->3)a6eo-abieta-3,8,ll,13-tetraen-18-oic  acid  lactone  (151) Concentrated HCl (6.5 mL) was added to a stirred solution of 162 (1.98 g, 5.37 mmol) in MeOH (66 mL). The reaction mixture was stirred for 24 h at room temperature and reaction completion was monitored by TLC eluting twice with isopropyl ether. Water (135 mL) was added slowly with proper stirring, cooling the suspension to 4 °C and the stirring continued for 1 h. The suspension was then filtered under vacuum and washed with cold water (96 mL), until the pH was neutral, yielding practically pure 151 as a whitish powder (1.55 g, 88.6% yield).  200  7p,12,19-Trihydroxy-18(4->3)aZ>eo-abieta-3,8,ll5l3-tetraen-18-oic  acid  lactone  (146b)  The above compound 151 (1.50 g, 4.60 mmol) was dissolved in EtOH (90 mL) with CH2CI2 (22.5 mL) to increase solubility. NaBH4 (330 mg, 8.70 mmol) was added in small portions and the reaction mixture was stirred for 8 h, until all the starting material disappeared (TLC, CH2Cl2-acetone, 96:4). The reaction was quenched with water (120 mL) followed by addition of CH2CI2 (270 mL). The organic phase was separated and the aqueous phase was re-extracted with more CHjClj (100 ml). The combined organic extract was washed with brine, dried, filtered and evaporated. The crude product was purified by the chromatotron using gradient elution with CH2Cl2-acetone (96:4 to 80:20) to give the desired 7P-hydroxy compound 146b, as a colorless crystalline solid (1.18 g, 98.3 % yield based on recovered starting material) and unreacted 7-oxo-ITP (151, 301 mg, 20.0 %).  Physical data of 7P,12,19-trihydroxy-18(4->-3)aZ)eo-abieta-3,8,ll,13-tetraen-18-oic acid lactone (146b):  A colorless crystalline solid; mp: 136-138 ^C (dec); [a]'^: +94.8° (c = 0.048, CHCI3); UV X^l^^  (log s): 208.0 (4.49), 220 (4.28), 283.0 (3.45); IR (CHCI3) Vmax  cm-i: 3580 (OH), 2960 (CH), 1750 (C=0), 1670, 1620, 1518 (aromatic C=C); 'H NMR  201  (400 MHz, CDCI3) 6: 1.10 (3H, s, H20), 1.24, 1.26 (3H each, both d, J = 6.8 Hz, H16, H17), 1.66 (IH, m, Hla), 1.87 (IH, ddd, J= 13.7, 12.9, 9.0 Hz, H6p), 2.30 (IH, ddd, J 12.9, 7.7, 2.7 Hz, H6a), 2.34-2.55 (3H, m, Hip, H2), 2.75 (IH, br d, J = 13.8 Hz, H5), 3.14 (IH, sept, J = 6.8 Hz, H15), 4.76 (2H, br ABq, Av = 0.07 ppm, J = 17.1 Hz, H19), 4.93 (IH, s, C12-0H), 4.95 (IH, br dd, J = 8.6, 8.6 Hz, H7a), 6.69 (IH, s, H l l ) , 7.35 (IH, s, H14); ''C NMR (75.3 MHz, CDCI3) 6: 18.1 (C2), 22.5 (C16,17), 22.8 (C20), 27.0 (C15), 31.0 (C6), 32.7 (CI), 36.7 (CIO), 40.8 (C5), 69.1 (C7), 70.3 (C19), 110.9 (Cll), 125.4 (C3), 126.6 (C14), 130.0 (C13), 137.3 (C8), 144.0 (C9), 156.9 (C12), 161.9 (C4), 174.9 (CI8); LRMS m/z (rel. intensity): 328 (M^ 14.4), 310 (90.3), 295 (77.4), 285 (25.5), 267 (49.5), 253 (base peak), 178 (28.3), 165 (21.7), 43 (38.1); HRMS calcd. for C20H24O4: 328.1674; found: 328.1671; Anal.calcd.for C20H24O4: C, 73.13; H, 7.37; found: C, 73.08; H, 7.22.  3.6.2  Synthesis of the (7,8)P-epoxy-dienone, 73  (7,8)p-epoxy-19-hydroxy-12-oxo-18(4->3)a6eo-abieta-3,9(ll),13-trien-18-oic  acid  lactone (73)  Potassium carbonate (630 mg, 4.56 mmol) was partly dissolved in trifluroethanol (60 mL) by sonicating in a warm water bath for 5 min. To this solution 7Phydroxy-isotriptophenolide (146b, 500 mg, 1.52 mmol)  202  was added and the mixture was stirred for 15 min followed by addition of PIDA (567.5 mg, 1.76 mmol). Stirring was continued at room temperature for 30 min, until complete consumption of the starting material as evidenced by TLC. The reaction mixture was then evaporated to dryness under vacuum and the residue was redissolved in ethyl acetate, and filtered through a short pad of silica gel eluting with additional amounts of ethyl acetate. The combined filtrate was evaporated under vacuum and purified by the chromatotron using methylene chloride-acetone (94:6) to give (7,8)P-epoxy-dienone 73 (149.8 mg, 30.2 %) and 7-oxo-isotriptophenolide (151, 292 mg, 58.8 %).  Physical data of (7,8)p-epoxy-19-hydroxy-12-oxo-18(4^3)aZ)eo-abieta-3,9(ll),13-trien18-oic acid lactone (73):  Colorless long plates; mp: 164-165 ^C (dec); [a]^^ : -360.9° (c = 0.133, CHCI3); UV X^l^^  (log s): 218.0 (4.13), 258.6 (4.23); IR (Thin film) Vmax cm"': 2963 (CH),  1752 (lactone C=0), 1655 (dienone C=0), 1631 (C=C), 1231, 1058,1021, 907 (C-O-C); 'H NMR (400 Mhz, CDCI3) 5: 1.06, 1.08 (3H each, both d,J= 6.9 Hz, H16, H17), 1.17 (3H, s, H20), 1.61 (IH, br ddd, J = 12.9, 12.6, 6.1 Hz, Hla), 2.10 (IH, br dd, J = 12.9, 5.6 Hz, Hip), 2.13 (IH, br dd, J = 14.5, 13.4 Hz, H6p), 2.26 (IH, ddd, J = 14.5, 5.8, 5.7 Hz, H6a), 2.30 (IH, m, H2p), 2.47 (IH, br d, J = 18.5 Hz, H2a), 2.58 (IH, br d, J = 13.4 Hz, H5), 2.98 (IH, sept d, J = 6.9 Hz, H15), 3.84 (IH, d,J= 5.8 Hz, H7a), 4.68 (2H, br s, H19), 5.99 (IH, d, y = 1.0 Hz, H14), 6.42 (IH, s, Hll); ''C NMR (75.3 MHz, CDCI3) S: 16 (C20), 17.4 (C2), 21.3 (C16), 21.6 (C17), 23.9 (C6), 26.3 (C15), 32.7 (CI), 38.1  203  (CIO), 42.0 (C5), 56.0 (C8), 64.0 (C7), 69.9 (C19), 125.6 (C3), 128.5 (Cll) 139.0 (C14), 149.8 (C13), 157.9 (C9), 159.4 (C4), 173.1 (CIS), 186.1 (C12); EIMS m/z (rel. intensity): 326 (M\ 71.4), 311 (base peak), 295 (27.1), 267 (15.8), 253 (29.3), 203 (27.1), 178 (16.1), 165 (18.8), 163 (30.7), 141 (25.4), 115 (17.9), 43 (18.7); HRMS calcd. for C20H22O4: 326.1518; found: 326.1522; Anal, calcd. for C20H22O4: C, 73.59; H, 6.80; found: C, 73.28; H, 6.87.  3.6.3  Epoxidation of (7,8)P-epoxy-dienone 73  A solution of (7,8)p-epoxy-dienone (73, 40 mg, 0.12 mmol) in benzene (6 mL) was epoxidized by the addition of Nbutyl hydroperoxide (90%, 0.08 mL) and Triton B (40%, 0.04 mL) followed by stirring at room temperature for 4 h. The progress of the reaction was monitored by TLC using hexanes-EtOAc, (1:1), or ether as eluerit. The reaction mixture was worked up by filtering through a short pad of silica gel and washing with ethyl acetate (50 mL). The crude ethyl acetate residue was purified by preparative TLC using ethyl acetate-hexanes (1:1) to give 30.1 mg of a pure mixture of the diepoxide 163 and the isomerized starting material, 164 (62:38 of 164:163 as shown by 'H NMR). Repeated recrystallization of this mixture using ethanol gave pure samples of both products.  Physical data of 5(3-(7,8)p-epoxy-19-hydroxy-12-oxo-18(4->3)aZ7eo-abieta-3,9(ll),13trien-18-oic acid lactone (164):  204  White crystals; mp: 189-191 ^C (dec); IR (Thin film) Vmax cm-l; 2969 (CH), 1752 (lactone C-0), 1681 (dienone C=0), 1632 (C=C), 1217, 1034, 907 (C-O-C); IH NMR (400 MHz, CDCI3) 5: 1.07, 1.08 (3H each, both d, J = 6.9 Hz, H16, H17), 1.17 (3H, s, H20), 1.70 (IH, m, Hla), 1.80 (IH, dd, J = 14.2, 5.8 Hz, Hip), 1.89 (IH, dd, J = 14.8, 12.8 Hz, H6a), ), 2.26 (IH, m, H2p), 2.33 (IH, m, H2a), 2.39 (IH, dt , J = 14.8, 3.6 Hz, H6P), 2.87 (IH, br d, J = 12.8 Hz, H5p), 2.97 (IH, sept, J = 6.9 Hz, H15), 3.80 (IH, d, J = 3.6 Hz, H7a), 4.71 (2H, br AB,,, Av = 0.19 ppm, J = 16.8 Hz, H19), 6.03 (IH, s, H14), 6.52 (IH, s, Hll); EIMS m/z (rel. intensity): 326 (M\ base peak), 310 (76.8), 295 (50.1), 284 (15.3), 267 (40.4), 253 (93.0), 203 (17.9), 165 (21.4), 152 (17.0), 141 (23.4), 43 (47.1); HRMS calcd. for C20H22O4: 326.1518; found: 326.1520.  Physical data of 5p-(7,8)p,( 13,14)a-bis(epoxy)-19-hydroxy-12-oxo-18(4->3)a6eo-abieta3,9(1 l)-dien-18-oic acid lactone (163):  Colorless solid; mp: 198-200 ^C (dec); IR (Thin film) Vmax cm-1: 2964 (CH), 1752 (lactone C=0), 1676 (dienone C-0), 1628 (C=C), 1218, 1035 (C-O-C); IH NMR (400 MHz, CDCI3) 6: 0.92, 1.00 (3H each, both d,J= 6.9 Hz, H16, H17), 1.12 (3H, s, H20), 1.66(1H, m, Hla), 1.79 (IH, m. Hip), 1.92 (IH, dd, J = 12.2, 11.4 Hz, H6a), 2.25  205  (IH, m, H2p), 2.37 (2H, m, Hip, H2a), 2.42 (IH, dt, J = 12.2, 3.8 Hz, H6P), 2.53 (IH, sept, J = 6.9 Hz, H15), 2.62 (IH, dd, J = 11.4, 3.8 Hz, H5p), 3.06 (IH, s, H14p), 3.77 (IH, d, J = 3.7 Hz, H7a), 4.70 (2H, br AB,, Av = 0.19 ppm, J= 16.9 Hz, H19), 6.26 (IH, s, H l l ) ; EIMS m/z (rel. intensity): 342 (M^ 7.7), 328 (8.1), 326 (14.9), 310 (12.1), 299 (13.2), 285 (9.2), 271 (12.7), 253 (18.1), 243 (16.4), 231 (10.9), 192 (20.7), 164 (19.9), 149 (13.3), 43 (base peak); HRMS calcd. for C20H22O5: 342.1467; found: 342.1464.  3.7  Synthesis of the 8-hydroxy series  The 8a-hydroxy-dienone 157 and the SP-hydroxy-dienone 158 were prepared as previously described in Section 3.5.1. The present section illustrates the epoxidation of those compounds to prepare a series of monoepoxidized derivatives.  3.7.1  Epoxidation of the 8a-hydroxy-dienone 157  A solution of 8a-hydroxy-dienone 157 (49 mg, 0.15 mmol) in toluene (9.2 mL) was cooled down to 0 ^C under argon atmosphere. A freshly prepared solution of trimethylbenzylammonium /-butylperoxide (0.26 mL, 1.2M in toluene) was added dropwise and the mixture was stirred at 0 ^C for 15 min. Then the ice bath was removed and the solution was warmed up to room temperature in 15 min. The mixture was then stirred to 28-30 ^C, monitoring the progress of the reaction by HPLC (see Appendix, method 2). The reaction was completed after 1.5 h and the mixture was diluted with  206  EtOAc and filtered through a short pad of silica gel washing with ethyl acetate. The solvent was removed under vacuum and the residue was purified by preparative TLC eluting with ether to yield the epoxides 165 (19.8 mg, 38.6%), 166 (5.9 mg, 11.5%) and 167 (7.4 mg, 14.4%), (combined yield = 64.5%).  Physical  data  of  5p-(13,14)a-epoxy-8a,19-dihydroxy-12-oxo-18(4->3)aZ)eo-abieta-  3,9(1 l)-dien-18-oic acid lactone (165):  A white sohd, mp: 239-241 ^C (dec); [aV^: +  O  •y 'OH  219.50 (c = 0.143, CHCI3); UV X^l^^  (log s): 216.4  (4.17), 239.5 (4.02); IR (thin film) v^ax cm-l; 3400 (OH), 2938 (CH), 1752 (lactone C=0), 1673 (enone  C=0), 1089, 1049, 1026 (C-O-C); iH NMR (400 MHz, CDCI3) 5: 0.88, 0.97 (3H each, both d, J= 7.0 Hz, H16, H17), 1.08 (3H, s, H20), 1.39 (IH, dd, J = 13.8, 6.1 MHz, Hip), 1.62 (IH, m, H7p), 1.84 (IH, m, H6a), 2.11 (4H, m, H2p, H5, H6p, H7a), 2.39 (IH, dd, J= 18.0, 6.1 Hz, H2a), 2.45 (IH, d, J = 2.1 Hz, C8-0H, D^O exchangeable), 2.53 (IH, sept, J = 7 Hz, H15), 2.79 (IH, ddd, J = 12.8, 12.6, 6.1 Hz, Hla), 3.31 (IH, s, H14), 4.71 (2H, br ABq, Av = 0.24 ppm, 7 = 16.9 Hz, H19), 6.00 (IH, s, Hll); '^C NMR (125.8 MHz, CDCI3) 5: 15.6 (C16), 18.2 (C2), 18.7 (C17), 23.5 (C20), 24.6 (C6), 25.1 (C15), 29.2 (CI), 35.9 (C7), 40.5 (CIO), 43.9 (C5), 64.1 (C14), 65.8 (C8), 70.8 (C19), 124.0 (Cll), 124.2 (C3), 162.6 (C9*), 163.5 (C4*), 173.6 (C18), 192.3 (C12) (note: * interchangeable); DCI m/z (rel. intensity): 362 (M + NH/, base peak), 345 (M+ H, 70.6),  207  329 (18.0), 316 (13.6), 299 (16.7); HRMS calcd. for C20H25O5 (M + H): 345.1702; found: 345.1706; Anal, calcd. for C20H24O5: C, 69.75; H, 7.02; found: C, 69.53; H, 6.96.  Physical data of epoxy-derivative 166:  MeOH  A colorless solid; mp: 80-82 ^C; [a] '^ : + 230.0° (c = 0.050, CHCI3); UV X.max  (log £): 244.1 (3.94); IR (thin film) Vmax cm-l; 2966 (CH), 1784 (sat. lactone C=0), 1673 (enone C=0), 1130, 1048, 998, 982 (C-O-C); ^H NMR (400 MHz, CDCI3) 5: 0.85, 0.93 (3H each, both d, J = 7.0 Hz, HI6, HI7), 1.23 (3H, s, H20), 1.56 (2H, m, Hip, H7p), 1.67 (IH, t, J = 2.9 Hz, H5p), 1.78 (4H, m, Hla, H2, H6a), 1.94 (IH, m, H6p), 2.09 (IH, m, H7a), 2.15 (IH, dd, J = 11.8, 4.0 Hz, H3P), 2.53 (IH, sept, J = 7 Hz, H15), 3.34 (IH, s, H14), 4.32 (2H, br ABq, Av = 0.22 ppm, J = 9.2 Hz, H19), 5.77 (IH, s, HI 1); "C NMR (125.8 MHz, CDCI3) 5: 15.9 (C16), 16.4 (C6), 17.7 (C2), 18.9 (C17), 24.6 (C15), 25.4 (C20), 29.8 (C7), 39.5 (CIO), 40.6 (CI), 40.7 (C5), 47.9 (C3), 58.0 (C14), 62.8 (C13), 70.4 (C8), 7.1 (CI9), 79.7 (C4), 119.3 (Cll), 165.4 (C9), 174.9 (CI8), 193.3 (CI2); DCI m/z (rel. intensity): 362 (M + N H / , 42.5), 345 (M + H, 76.1), 344 (M^ 54.5), 329 (72.6), 315 (base peak), 282 (99.7); HRMS calcd. for C20H24O5: 344.1624; found: 344.1626.  Physical data of epoxy-derivative 167:  208  Short colorless needles; mp: 168-170 oC; [a] ^ : + 273.60 (c = 0.1027, CHCI3); UV X^^^"" (log s): 243.4 (3.95); IR (thin film) Vmax cm-l; 2967 (CH), 1780 (sat. lactone C=0), 1677 (enone C=0), 1118, 1080, 1013 (C-O-C); ^H NMR (400 MHz, CDCI3) 5: 0.86, 0.94 (3H each, both d, J = 7.0 Hz, H16, H17), 1.21 (3H, s, H20), 1.52 (3H, m. Hip, H5P, H7p), 1.70 (IH, dd, J = 11.9, 3.6 Hz, Hla), 1.77 (IH, m, H2a), 2.02 (4H, m, H2p, H6, H7a), 2.57 (IH, sept, J = 7 Hz, H15), 2.66 (lH,br d, J = 7.1 Hz, H3a), 3.30 (IH, s, H14), 4.13 (2H, br ABq, Av = 0.41 ppm, J = 8.9 Hz, H19), 5.82 (IH, s, Hll); ''C NMR (125.8 MHz, CDCI3) 6: 15.4 (C6), 15.7 (C16), 16.5 (C2), 18.9 (C17), 24.7 (C15), 28.5 (C20), 29.0 (C7), 35.5 (CI), 37.4 (C5), 37.7 (CIO), 45.6 (C3), 58.8 (C14), 63.4 (C13), 70.3 (C19), 71.6 (C8), 78.9 (C4), 121.0 (Cll), 165.3 (C9), 174.3 (C18), 192.5 (C12); DCIMS m/z (rel. intensity): 362 (M + NH/, 0.1), 345 (M + H, 73.7), 315 (base peak); HRMS calcd. for C20H25O5: (M + H): 345.1702; found: 345.1704.  3.7.2  Epoxidation of 8p-hydroxy-dienone 158  Compound 158 (95 mg) was epoxidized (1.5 h) and the product was purified using the same conditions described above for 157. A pure mixture (41.0 mg, 41.2%) of epoxides 172 and 173 (85:15, respectively, as shown by 'HNMR) was obtained. A sample of the pure P-epoxy-diastereoisomer was obtained by recrystallizing the mixture twice with hexanes-EtOAc to give 172 as very fine needle-shaped crystals.  209  Physical  data  of  5|3-(13,14)(3-epoxy-8p,19-dihydroxy-12-oxo-18(4^3)a6eo-abieta-  3,9(1 l)-dien-18-oic acid lactone (172):  Colorless fine needles; mp: 190-192 »€; [a]^^ : -227.0° (c = 0.1238, CHCI3); UV \^l^  (log s): 215.8  (4.28), 328.0 (sh, 2.35); IR (CHCI3) Vmax cm-1: 3449 (OH), 2966 (CH), 1752 (lactone C=0), 1677 (enone C=0); IH NMR (400 MHz, CDCI3) 5: 0.88, 0.96 (3H each, both d, J = 6.9 Hz, HI6, HI7), 1.29 (3H, s, H20), 1.42 (IH, m, H6a), 1.67 (IH, br t, J =7.6 Hz, H7a), 1.77 (2H, m, HI), 2.14 (2H, m, H6p, H7P), 2.26 (IH, m, H2P), 2.34 (IH, br d, J = 19.1 Hz, H2a), 2.49 (IH, d,J= 1.9 Hz, C8-0H, D^O exchangeable), 2.52 (IH, sept, J = 6.9 Hz, H15), 3.05 (IH, br d, J = 9.3 Hz, H5), 3.35 (IH, s, H14), 4.70 (2H, br ABq, Av = 0.12 ppm, J = 16.8 Hz, H19), 5.96 (IH, s, Hll); ''C NMR (125.8 MHz, CDCI3) 6: 15.8 (C16), 17.2 (C2), 18.8 (C17), 23.4 (C6), 25.1 (C15), 27.5 (C20), 29.7 (CI), 33.2 (C7), 39.2 (C5), 39.5 (CIO), 63.0 (C14), 65.5 (C13), 69.4 (C9), 70.8 (C19), 123.3 (Cll), 125.3 (C3), 161.8 (C4 & C9), 173.5 (C18), 192.2 (C12); DCl m/z (rel. intensity): 362 (M + N H / , base peak), 345 (15.5), 316 (12.4), 299 (21.0); HRMS calcd. for C20H25O5: (M+ H): 345.1702; found: 345.1697; Anal, calcd. for  CJQH^A: C,  69.75; H,  7.02; found: C, 69.87; H, 7.02.  3.8  Attempted bio-epoxidation of (7,8)-epoxy-dienone precursors with TRP4a cell cultures  210  3.8.1  General Growth conditions of TRP4a cell cultures  Production and propagation of plant cell cultures and the TRP4a cell line were conducted in the Biological Services Facility, Department of Chemistry, University of British Columbia and essentially in accord with the earlier published procedure.^^.^S Cultures were grown in conical flasks and incubated without illumination at 27 ± 1 ^C on a rotary shaker with a 7/8" throw and run at 140 rpm. The cultures were grown in PRD2C0100 growth medium for 14 days and then resuspended as a 10% inoculum in MSNA0.5K0.5 production medium. The growth and viability of the cultures were assayed by measuring refractive indices (r\Q, Galileo refractometer, 25.0 ^C), and monitoring microscopic purity and pH.  Harvesting and extraction procedures  T. wilfordii cell cultures grown in MSNA0.5K0.5 media were used for biotransformation studies. The age of the cell culture at the time of biotransformation was calculated from the day when the inoculation from PRD2C0100 media to MSNA0.5K0.5 media was made. Cells were harvested after growth in MSNA0.5K0.5 medium for a certain period of time (between 7 and 25 days) by filtration through Miracloth^'^, a coarse fibrous cloth which separates cellular material from liquid broth. The cells were fi^ozen until the time of extraction.  The broth (spent medium) was  211  saturated with NaCl and extracted with EtOAc (3 x broth volume). The extracts were combined, dried over anhydrous Na2S04 or MgS04, and filtered through paper. The solvent was removed by rotary evaporation and the residual extract was dried in vacuo. Frozen cells were thawed and homogenized with a homogenizer (Ultra Turrax, TK-25, 28,000 rpm) in EtOAc (3-5 x cells volume). The homogenized cell suspension in EtOAc was filtered through a pad of Celite, and separated into an organic and an aqueous fraction. The aqueous layer of the filtrate was saturated with NaCl and further extracted with EtOAc (3 x aqueous layer volume). All the EtOAc extracts of the cells were combined, washed with saturated NaCl solution (equal volume) and dried. The EtOAc extract was filtered, the solvent removed by rotary evaporation and the residual extract dried in vacuo.  General procedure for the biotransformation of the synthetic precursors  The precursors were dissolved in a certain amount of ethanol and then added to the cell culture of a specific age (7, 14 or 25-day-old) under sterile conditions. Incubation was carried out under the same conditions as for the intact cell cultures (standard conditions).  Samples were taken at specified intervals under sterile conditions and  filtered to separate the broth and cells. Refractive indices (r\^) and pH measurement as well as microbial contamination determination were carried out on the broth samples. The cells were kept frozen in the freezer until the time of extraction.  Subsequent  harvesting, homogenization and extraction of the cells were carried out as outlined above.  212  The extracts from biotransformation experiments (broth and cells) were analyzed by TLC, HPLC and MS. For TLC analysis, the cell culture samples (cells with broth) were sonicated with ethyl acetate (same volume as the sample) for 5 min, and the resulting extracts were used for analysis. HPLC was performed using a Waters reverse phase "Radial-pack" 8C18 cartridge (5|j.), and a UV absorbance detector set at 254 nm (recording wavelength) and 280 nm. The eluent was a mixture of methanol, water, acetonitrile, acetic acid, in a ratio: 30:55:15:0.1, at a flow rate of 1.5 ml/min. Samples which had been filtered through a MiUipore HV filter (0.45 |j.m) were quantitatively dissolved in MeOH prior to HPLC analysis. Control experiments were performed as references and they were similar to the corresponding biotransformation, except that an equal volume of pure solvent, same as that used to dissolve the precursor, was added instead of the starting material. The control experiments were always carried out simultaneously and under identical conditions. To examine whether any changes was brought about by the solvent, medium or atmospheric oxygen, blank experiments (i.e., no solvent nor precursor added to it) were usually carried out under identical conditions as those for the biotransformation experiments.  Measurement of peroxidase activity (pyrogallol-purpurogallin assay)  A sample of the broth for each different age (7, 14 or 25-day-old) was analyzed for the peroxidase activity by established procedures.'^°  213  The broth (1 mL) was added to a 50 mL Erlenmeyer flask containing 5% aqueous pyrogallol solution (2 mL), 0.1 M phosphate buffer (2 mL, pH 6.6), freshly prepared 0.5% hydrogen peroxide (H2O2) solution (1 mL) and distilled water (14 mL) at 20 °C. This mixture was allowed to stand for 20 seconds at 20 °C and then 2M H2SO4 (1 mL) was added to quench the reaction and the solution was then extracted with ether (2 x 25 mL).  The absorbance reading of the extract at 420 nm (Bausch & Lomb, model  Spectronic 20 UV spectrometer) was measured against a blank solution containing an ethereal extract (2 x 25 mL) of a mixture comprised of 5% pyrogallol solution (2 mL), 0.1 M phosphate buffer (3 mL, pH 6.6), freshly prepared 0.5% H2O2 solution (1 mL) and distilled water (14 mL). The peroxidase activity was calculated from a standard curve, which was prepared by measuring the absorbance at 420 nm of a set of standard solutions of purpurgallin (0.5 to 3.5 mg) in ether (50 mL).  3.8.2  Biotransformation of the (7,8)a-epoxy-dienone (72) with TRP4a cell culture  A stock solution of the (7,8)a-epoxy-dienone (72) was prepared in EtOH (concentration, 15 mg/mL). One mL of this solution was added sequentially to a flask containing 125 mL of different ages of TRP4a cell cultures: 7, 14 and 25-day-old and incubated under normal conditions.  A control (1 mL EtOH in 125 mL culture)  experiment and a blank experiment (125 mL culture only) were carried out under the same conditions for each of the different ages. Samples were taken at every 24 h, with the culture harvested after 4 days of incubation. The samples (5 mL) were extracted with  214  EtOAc (sonicated in EtOAc) and analyzed by TLC (two different solvent systems were used; A: toluene-chloroform-ethyl acetate-formic acid, 105:48:45:3, developed twice; and B: CH2Cl2-methanol, 97:3) and HPLC. There was no significant change in the 7-day-old culture samples in comparison with control and blank samples. In the 14-day-old culture, almost all the staring material was consumed and the following products were formed: 7-oxo-ITP (151), 7a-hydroxyITP (146a), 7P-ethoxy-ITP (175b) and (6,7)-dehydro-ITP (159); in comparison to the authentic samples. For the 25-day-old culture, similar observations were noted as to the 14-day-old culture, except that less starting material was consumed (c.a.50%). These results are presented in Figure 2.28, in Section 2.8.2 of this thesis.  3.8.3  Biotransformation of (7,8)P-epoxy-dienone (73)  One mL of the stock solution of (7,8)P-epoxy-dienone 73 (15 mg/mL, in EtOH) was added to TRP4a cell cultures of different ages (7, 14 and 25-day-old, 125 mL each) and incubated under normal conditions as used above for the (7,8)a-epoxy-dienone 72. Samples (5 mL) were taken after 1, 2, 3, and 4 days of incubation for analysis. The samples were extracted with EtOAc and analyzed by TLC (vide supra) and HPLC. The cultures were harvested after 4 days, and the cells and broth were extracted with EtOAc and all the samples analyzed by HPLC and MS. The HPLC results were very similar to those of (7,8)a-epoxy-dienone experiments. Detailed results are shown in Figure 2.29 in Section 2.8.2.  215  REFERENCES (1)  S. M. Kupchan, W. A. Court, R. G. Dailey Jr., C. J. Gilmore, and R. F. Bryan. J. Am. Chem. Soc. 1972, 94, 7194-7195.  (2)  X.-Y. Lu, C.-P. Zhang, and Y. Chen. Jiang Su Yi Yao 1987, 640-643.  (3)  D.-G. Wu, X.-C. Sun, and F. Li. Acta Botanica Yunnanica 1979, 1, 29-35.  (4)  An Encyclopedia of Traditional Chinese Medicinal Substances; 1 st ed.; Jiangsu College of New Medicine, Ed.; Shanghai Publishing House of Science and Technology: Shanghai, 1977; Vol. 2.  (5)  Iconographia Cormophytorum Sinicorum, Supplementum II; I st ed.; Botany Institute-Chinese Academy of Science, Ed.; Science Publishing House: Beijing, 1983,p 266.  (6)  S. Huang, Ed. "The Saint Peasant's Scripture of Materia Medica," Publishing House for Chinese Medical Classics, Beijing, republ. 1982, pp 309-310.  (7)  S.-F. Chiu and Y.-T. Qiu. Journal of Applied Entomology 1993, 116, 479-486.  (8)  D. Y. Yu, J Traditional Chinese Med 1983, 3, 125-129.  (9)  J.-L. Guo, S.-X. Yuan, X.-C. Wang, S.-X. Xu, and D.-D. Li. Chinese Medical Journal (English Edition) 1981, 94, 405-412. '  (10)  W.-Y. Xu, J.-R. Zheng, and X.-Y. Lu. Intl. J Dermatol. 1985, 24, 152-157.  (11)  J. Zheng, J. Fang, J. Gao, H. Guo, L. Xu, Y. Yeng, Y. Yu, H. Sun. Zhongguo Yixue Kexueyuan Xuebao 1986, 8, 19-23.  (12)  S.-Z. Qian. Contraception 1987, 36, 335-345.  (13)  (a) M. Beroza. J Amer. Chem. Soc. 1953, 75, 44-49. (b) M. Beroza. Ibid 1953, 75, 2136-2138.  (14)  R.-L Li and D.-F. Su. Experimental and Clinical Application of Tripterygium wilfordii. Chinese Academic Press, Beijing, 1989, pp. 47-100.  (15)  One of the main manufacturers of commercially available tablets of GTW is Taizhou Pharmaceutical Factory, in the province of Jiangsu, China.  (16)  J. Li, YaoXue TongBao 1985, 20, 101-105.  216  (17)  X.-L. Tao, Y. Sun, Y. Dong, Y. Shi, Y.-L. Xiao, H. Di and N.-C. Chang. Chin. J. Med. 1987, 26, 399-402.  (18)  X.-Y. Li. Int. J. Immunotherapy 1993, IX, 181-187.  (19)  W. Quin, C. Liu and S. Yang. Chin. Med J. 1981, 94, 827.  (20)  L. Li, X. Zhang and H. Chen. Chin. Med J. 1982, 62, 581.  (21)  X.-L. Tao, Y.-P. Shi, X.-H. Chen and N.-Z. Zhang. Acta Acad Med Sinica 1988, 10,361-364.  (22)  X.-F. Yang, L.-T. Xu, H. Ren, Z.-J. Li, C.-F. Sun and Z.-Y. Zhang. Chin. Med Sci. J. 1992, 7, 232-234.  (23)  W.-Z. Gu, S. R. Brandwein, S. Banerjee. J. Rheumatol. 1992, 19, 682-688.  (24)  T.-M. Zhang, Z.-Y. Chen and C. Lin. Acta Pharmacologica Sinica 1981, 2, 128131.  (25)  J.-R. Zheng, L.-F. Xu, L. Ma, D.-H. Wang and J.-W. Gao. Acta Academiae Medicinae Sinicae 1983, 5, 1-8.  (26)  X. Tao, H. Dai, W. Ye and N. Zhang. Arthritis and Rheumatism 1989, 32 (suppl. 4):S 131.  (27)  X.-L. Tao, L. S. Davis and P. E. Lipsky. Arthritis and Rheumatism 1991, 34, 1274-1281.  (28)  W.-Z. Yu, R. Chen, S. Brandwein, J. McAlpine, N. Burres. Int. J. Immunopharmac. 1995, 17, 351-356.  (29)  X. Tao, J.-J. Cai, P.-E. Lipsky. J. Pharmacol Exp. Ther. 1995, 272, 1305-1312.  (30)  L.-X. Zhang, K.-X. Gu, Z.-F. Cao, D.-J. Pan and W.-J. Zhang. Shanghai J. Immun. 1986, 6, 277-279.  (31)  S.-X. Yang, H.-L. Gao, S.-S. Xie, W.-R. Zhang, Z.-Z. Long. Int. J. Immunopharmac. 1992, 14, 963-969.  (32)  S. M. Kupchan and R. M. Schubert. Science 1974, 185, 791-792.  (33)  Y.-S. Wei and L Adachi. Acta Pharmacologica Sinica 1991, 12, 406-410.  (34)  D.-Q. Yu, D.-M. Zhang, H.-B. Wang and X.-T. Liang. Acta Pharmaceutica Sinica 1992, 27, 830-836.  217  (35)  Goodman and Gilman's The Pharmacological Basis of Therapeutics; 8"^ Ed.; A. G. Gilman, T. W. Rail, A. S. Nies and P. Taylor, Eds.; Pergamon Press: Elmsford, NY, 1990, Section XII.  (36)  S.-Z. Qian, C.-Q. Zhong, Y. Xu. Contraception 1986, 33, 105-110.  (37)  S.-Z. Qian, C.-Q. Zhong, N. Xu, Y. Xu. Adv. Contracep. 1986, 2, 253-254.  (38)  S.-Z. Qian, C.-Q. Zhong, Y. Xu. Adv. Contracep. 1986, 2, 298.  (39)  S.-Z. Qian, Y.-Z. Hu and J. S. Tong. Chinese J. Androl. 1989, 3, 129-132.  (40)  S.-Z. Qian, Y. Xu, J.-W. Zhang. Contraception 1995, 51, 121-129.  (41)  S. A. Matlin, A. Belenguer, V. E. Stacey, S.-Z. Qian, Y. Xu, J.-W. Zhang, J. K. M. Sanders, S. R. Amor and C. M. Pearce. Contraception 1993, 47, 387-400.  (42)  W.-S. Ye, Y.-C. Deng, Y.-L. Huang, Y.-H. Han and S.-P. Xue. Acta Anatomica Sinica 1992, 23, 290-293.  (43)  J. Guo, S. Yuan, X. Wang, S. Xu and D. Li. Chinese Med J. 1981, 7, 405-412.  (44)  J.-R. Zheng, K.-X. Gu, J.-W. Gao, L.-F. Xu, Y.-H. Yu and M.-Y. Tang. Acta Academiae Medicinae Sinicae 1991, 13, 398-403.  (45)  Y. Takaishi, N. Wariishi, H. Tateishi, K. Kawazoe, K. Miyagi, K. Li and H. Duan. Phytochemistry 1997, 45, 979-984.  (46)  P.-C. Ma, X.-Y. Lu, J.-J. Yang, and Q.-T. Zheng. Acta Pharmaceutica Sinica 1991, 26, 759-763.  (47)  F.-X. Deng, S.-Q. Huang, J.-H. Cao, Z.-L. Xia, S. Lin, D.-Y. Zhu, S.-H. Jiang, and Y.-L. Zhu. Acta Botanica Sinica 1985, 27, 516-519.  (48)  X.-Y. Lu, P.-C. Ma, Y. Chen, C.-P. Zhang, Y.-G. Zhang, Z.-X. Zhang, L.-S. Sheng, S.-Z, Li, D.-K. An, C.-H. He, and Q.-T. Zheng. Acta Academiae Medicinae Sinicae 1990, 12, 157-161.  (49)  P.-C. Ma, X.-Y. Lu, C.-H. He and Q.-T. Zheng. Acta Botanica Sinica 1991, 33, 370-377.  (50)  P.-C. Ma and C.-L. Yang. Acta Botanic Sinica 1993, 35, 637-643.  (51)  F.-X. Deng, Z.-L. Xia, R.-Q. Xu and J.-Y. Chen. Acta Bontanica Sinica 1992, 34, 618-621.  218  (52)  S. Lin, F.-X. Deng, N. Sakurai, D. Hong, and Y.-Z. Chen. Acta Botanica Sinica 1993, 35, 385-389.  (53)  C.-P. Zhang, X.-Y. Lu, P.-C. Ma, Y. Chen, Y.-G. Zhang, Z. Yan, G.-F. Chen, Q.-T. Zheng, C.-H. He, and D.-Q. Yu. Acta Pharmaceutica Sinica 1993, 28, 110115.  (54)  F.-X. Deng, S.-Q. Huang, Z.-D. Wang, G.-G. Ma, G.-Q. Song, and Z.-X. Chen. Acta Pharmaceutica Sinica \9%1, 16, 155-157.  (55)  F.-X. Deng, B.-N. Zhou, G.-Q. Song and C.-Q. Hu. Acta Pharmaceutica Sinica 1982, 17, 146-150.  (56)  K. Li, H. Duan, K. Kawazoe and Y. Takaishi. Phytochemistry 1997, 45, 791-796.  (57)  K.-C. Chen, R.-Z. Yang and C.-B.Wang. Zhong Cao Yao 1986, 17, 242-245.  (58)  J. Xu, T. Ikekawa, M. Ohkawa, I. Yokota, N. Hara and Y. Fujimoto. Phytochemistry 1997, 44, 1511-1514.  (59)  L. Zhang, Z.-X. Zhang, L.-S. Sheng, D.-K. An, Y. Lu, Q.-T. Zheng, and S.-C. Wang. Acta Pharmaceutica Sinica 1992, 28, 32-34.  (60)  X.-M. Zhang, C.-F. Wang and D.-G. Wang. Acta Botanica Yunnanica 1992, 14, 319-322.  (61)  L. Zhang, Z.-X. Zhang, D.-K. An and C. Kong. Acta Phamaceutica Sinica 1991, 26,515-518.  (62)  (a) J.-H. Shen and B.-N. Zhou. Chin. Chem. Lett. 1992, 5, 113-116. (b) Y. Takaishi, K. Shishida, N. Wariishi, M. Shibuya, K. Goto, M. Kido, M. Takai and Y. Ono. Tet. Lett. 1992, 33, 7177-7180.  (63)  J.-H. Shen. Ph.D. Thesis, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, 1990.  (64)  B.-N. Zhou, D.-Y. Zhu, F.-X. Deng, C.-G. Huang, J.P. Kutney, and M. Roberts. Planta Medica 1988, 54, 330-332.  (65)  X.-Y. Wu, G.-W. Qin, R.-S. Xu, D.J. Fan. Planta Medica 1994, 60, 189.  (66)  F.-X. Deng, J.-H. Cao, Z.-L. Xia, S. Lin, D.-Y. Zhu, and S.-H. Jiang. Acta Pharmaceutica Sinica 1987, 22, 377-379.  219  (67)  T. Morota, W.-Z. Qin, K. Takagi, L.-H. Xu, M. Maruno and B.-H. Yang. Phytochemistry 1995, 40, 865-870.  (68)  Y. Takaishi, K. Miyagi, K. Kawazoe, K. Nakano, K. Li and H. Duan. Phytochemistry 1997, 45, 975-978.  (69)  K. Shishido, K. Nakano, N. Wariishi, H. Tateishi, T. Omodani, M. Shibuya, K. Goto, Y. Ono and Y. Takaishi. Phytochemistry 1994, 35, 731-737.  (70)  K. Chen, Q. Shi, T. Fujioka, D.-C. Zhang, C.-Q. Hu, J.-Q. Jin, R.E. Kilkuskie and K.-H. Lee. J. Nat Prod 1992, 55, 88-92.  (71)  O. L. Gamborg and D. E. Eveleigh. Can. J. Biochem. 1968, 46, 417-421.  (72)  J. P. Kutney, G. M. Hewitt, T. Kurihara, P. J. Salisbury, R. D. Sindelar, K. L. Stuart, P. M. Townsley, W. T. Chalmers and G. G. Jacoli. Can. J. Chem. 1981, 59, 2677-2683.  (73)  J. P. Kutney, L. S. L. Choi, R. Duffin, G. M. Hewitt, N. Kawamura, T. Kurihara, P. Salisbury, R. Sindelar, K. L. Stuart, P. M. Townsley, W. T. Chalmers, F. Webster, and G. G. Jacoli. Planta Medica 1983, 48, 158-163.  (74)  T. Murashige and F. Skoog. Physiol. Plant 1962, 15, 473-497.  (75)  M. Roberts. Ph. D. Thesis, University of British Columbia, 1989.  (76)  K. Han. Ph. D. Thesis, University of British Columbia, 1994.  (77)  J. P. Kutney, K. Han. references cited therein.  (78)  F. Kuri-Brena. Ph. D. Thesis, University of British Columbia, 1992.  (79)  C. K. Lai, R. S. Buckanin, S. J. Chen, D. F. Zimmerman, F. T. Sher and G. A. Berchtold. J. Org. Chem. 1982, 47, 2364-2369.  (80)  F. T. Sher and G. A. Berchtold. J. Org. Chem. 1977, 42, 2569-2574.  (81)  E. E. van Tamelen, J. P. Demers, E. G. Taylor and K. KoUer. J. Am. Chem. Soc. 1980,102,5424-5425.  (82)  L. C. Garver and E. E. van Tamelen. J. Amer. Chem. Soc. 1982, 104, 867-869.  (83)  E. E. van Tamelen and T. M. Leiden. J. Amer. Chem. Soc. 1982, 104, 1785-1786.  (84)  H. Koike, T. Tokoroyama. Tetrahedron Lett. 1978, 46, 4531-4534.  Reel. Trav. Chim. Pays-Bas 1996, 115, 77-93 and  220  (85  B. H. Barton and Y. Houminer. J. Chem. Soc, Chem. Commun. 1973, 839-840.  (86  T. Tokoroyama, H. Koike, A. Kondo, K. Hirotsu. Y. Ezaki, Tennen Yuki Kagobutsu ToronkaiKoen Yoshishu, 24'\ 1981, 330-339.  (87  N. Miyazaki, K. Sato, M. Goto, M. Sasano, M. Natsuyama, K. Inoue and K. Nishioka. Arthritis and Rheumatism 1988, 31, 480-486.  (88  S. M. Kupchan, A. Karim and C. Marcks in: J. Amer. Chem. Soc. 1968, 90, 5923 and J. Org. Chem. 1969, 34, 3912.  (89  J. P. Demers. Ph. D. Thesis, Stanford University, 1980.  (90  For a review on decarboxylation of carboxylic acids see: R. A. Sheldon and J. K. Kochi. Org React 1972, 19, 279-421.  (91  Y. N. Ogibin, M. I. Katzin and G. I. Nikishin. Synthesis, 1974, 889-890.  (92  J. Bacha and J. K. Kochi. J. Org Chem., 1968, 33, 83-93. J. Bacha and J. K. Kochi. ibid. 1968, 33, 2746-2754.  (93  E. Bosch, R. Rathore and J. K. Kochi. J. Org. Chem. 1994, 59, 2529-2536.  (94  H. Zimmer, D. C. Lankin and S. W. Horgan. Chem. Rev. 1971, 71, 229-246.  (95  T. Kametani and K. Fukumoto. Synthesis 1972, 657-674.  (96  W. J. Taylor and A. R. Battersby, Eds. Arnold Hall, London, 1967.  (97)  H. Musso. Angew. Chem. Int Ed Engl. 1963, 2, 723-734.  (98)  V. Balogh, M. Fenzon and M. Golfier. J. Org Chem. 1971, 36, 1339-1341.  (99)  D. Taub, C. H. Kuo, H. L. Slates and N. L. Wendler. Tetrahedron 1963, 19, 1-17.  "Oxidative Coupling of Phenols", E.  (10,0) M. J. S. Dewar and T. Nakaya. J. Am. Chem. Soc. 1968, 90, 7134-7135. (101) J. Brussee and A. C. Jansen. Tetrahedron Lett. 1983, 3261-3262. (102) D. J. Hart, P. A. Cain and D. A. Evans. J. Am. Chem. Soc. 1978, 100, 1548-1557. (103) D. G. Hewitt. J. Chem. Soc, C 1971, 2967-2973. (104) H.-D. Becker and K. Gustafsson. J. Org Chem. 1979, 44, 428-432.  221  ^105) A. Pelter and S. M. A. Elgendy. J. Chem. Soc, Perkin Trans. 71993, 1891-1896. ^106) A. Varvoglis. Chem. Soc. Rev. 1981, 10, 377-407. ^107) A. Varvoglis. Synthesis 1984,709-726. ;i08) R. Raju and G. S. K. Rao. Ind J. Chem. 1987, 26B, 469-473. :i09) A. Pelter, R. S. Ward and A. Abb-El-Ghani. J. Chem. Soc. Perkin Trans. 11992, 2249-2251. ^110) M. V. Bhatt and S. U. Kuikarni. Synthesis 1983, 249-282. ;i 11) A. K. Mandal, N. R. Soni and K. R. Ratnam. Synthesis 1985, 274-275. :i 12) S. Spyroudis and A. Varvoglis. Synthesis 1975, 445-447; 1976, 837-838. :i 13) M. J. Harrison and R. O. C. Norman. J. Chem. Soc. (C) 1970, 728-730. ^114) R. B. Gammill. Tetrahedron Lett. 1985, 26, 1385-1388. ^115) A. R. Forrester, R. H. Thomson and S.-O. Woo. J. Chem. Soc, Perkin 11975, 2340-2353. 116) R. Criegee. Oxidation in Organic Chemistry, Academic Press, New York, 1965, 365-367. ;i 17) D. F. Banks. Chem. Rev. 1966, 66, 243-266. ;i 18) S. J. Angyal and R. J. Young. J. Am. Chem. Soc. 1959, 81, 5251-5255. ^119) L. M. Jackman and S. Stemhell. Applications of Nuclear Magnetic Resonance Spectroscopy in Organic Chemistry, 2 "'' ed., Pergamon Press, Oxford, 1969. ^120) D. H. Williams and I. Fleming. Spectroscopic Methods in Organic Chemistry, 4'^ ed., McGraw-Hill Books Co., London, 1989. ;i21) R. Curci, M. Fiorentino and L. Trousi. J. Org Chem. 1980, 45, 4758-4760. ^122) R. W. Murray. Chem. Rev. 1989, 1187. :i23) P. Bovicelli and P. Lupattelli. J. Org Chem. 1994, 59, 4304-4307. ^124) W. Adam, Y.-Y. Chan, D. Cremer, J. Gauss, D. Scheutzow and M. Schindler. J. Org Chem. 1987, 52, 2800-2803.  222  (125) Y. Angelis, X. Zhang and M. Orfanopoulos. Tet Lett. 1996, 37, 5991-5994. (126) F. L. Lakner and L. P. Hager. J. Org. Chem. 1996, 61, 3923-3925. (127) A. F. Dexter, F. J. Lakner, R. A. Campbell and L. P. Hager. J. Am. Chem. Soc. 1995, 117,6412-6413. (128) K. L. Stuart, J. P. Kutney and B. R. Worth. Heterocycles 1978, 9, 1015. (129) H. L. Holland. Organic Synthesis with Oxidative Enzymes, VCH Publishers, Cambridge, U.K. 1992. (130) B. Chance and A. C. Maehly. Methods in Enzymology, Academic press. New York, 1955, Vol. 2, p. 773 (131) J. I. Consepcion, C. G. Francisco, R. Freire, R. Hernandez, J. A. Salazar and E. Suarez. J. Org Chem. 1986, 51, 402-407.  223  APPENDIX  Notes:  Note 1: A simple MMX minimization software program, included in the CS Chem 3D Pro V.3.5, program for molecular modeling and analysis was used to obtain the minimized structures of the compounds indicated. Supplier: Cambridge Soft Corporation, Cambridge, MA (USA). Note 2: Samples of compounds 153, 154; and the epoxy-dienones 72 and 73 were sent to the British Columbia Cancer Agency for pharmacological in vitro testing. The results of these studies were not yet available at the time of printing of this thesis.  Analytical methods:  Gas chromatographic analysis of DHA (60): The purity of dehydroabietic acid was analyzed as its methyl ester using either one of the following methods for sample preparation. Method A: Dissolve 3-5 mg of sample in 5 mL of ether. Add dropwise a solution of diazomethane in ether until a very light yellow color remains for at least 15 seconds. Carefully heat the solution in a water bath to 28-30 °C for a few seconds till the color disappears completely. Evaporate the solvent under vacuum and then re-dissolve the sample in 1 mL of methanol.  224  Method B: Dissolve 3-5 mg of sample in 1 mL of methanol. Add one drop of 1% w/v phenolpthalein in methanol solution, followed by dropwise addition of 10% w/v tetramethyl ammonium hydroxide solution in methanol and mixing well until the pink color remains unchanged for about 5 minutes. Instrumental conditions: Initial temperature, 220 °C, final temperature, 245 °C; run time, 20 min; rate, 0.7 °C/min; attenuation, 1; detector temperature, 300 °C; injector temperature, 300 °C.  Under these conditions, the observed retention time for the  methyldehydroabietate was 15 minutes.  HPLC Method 1: Sample preparation:  5 |J,L of reaction mixture were dissolved in 3 mL of  methanol and evaporated to dryness under vacuum. The residue was redissolved in 0.5 mL of methanol. Mobile phase: solvent A, methanol; solvent B, water. A gradient elution was used as shown in the following table: Time  Flow  %A  %B  0  1.5  80  20  5  1.5  80  20  6  1.5  60  40  24  1.5  60  40  25  2.0  80  20  35  2.0  80  20  36  1.5  80  20  225  Instrumental conditions:  column: Waters "Radial-pack" 8C18 reverse phase  cartridge, 5|j,; detector: UV, X= 254 nm; sensitivity: 0.02. Using the above conditions the observed retention times for the compounds analyzed were: (7,8)a-epoxy-dienone, 72: 17.43 min; diepoxide 161: 10.96 min; and diepoxide 160: 16.16 min.  HPLC Method 2: Sample preparation:  5 \iL of reaction mixture were dissolved in 3 mL of  methanol and evaporated to dryness under vacuum. The residue was redissolved in 0.5 mL of methanol. Mobile phase: solvent A: a mixture of water, MeOH, acetonitrile, and acetic acid, 55.3:29.7:15:0.1; solvent B: a mixture of water and acetonitrile, 8:2. Solvents A and B were then mixed isocratically in proportion 8:2 respectively at a flow rate of 1.5 ml/min. Instrumental conditions:  column: Waters "Radial-pack" 8C18 reverse phase  cartridge, 5|a; detector: UV, X= 254 nm; sensitivity: 0.02. The observed retention times under these conditions for the compounds analyzed were; 8a-hydroxy-dienone, 157: 10.50 min and 8p-hydroxy-dienone, 158: 9.00 min.  An alternative synthesis of the exo-olefin 134:  Decarboxylation of dehydroabietic acid (DHA, 60) under conditions based on a method reported in the literature,''^' by adding slowly PIDA (6 equiv), to a mixture of the  226  substrate, copper (II) acetate and pyridine in benzene at reflux; gave after 24 h, a mixture of olefins 134 and 134a (82:18, respectively by ' H N M R analysis) in 77% yield. This alternative method of decarboxylation of DHA had the advantage of preventing the formation of one of the undesired isomeric alkenes (134b) and consequently a higher yield of the required olefin 134 was obtained. Nevertheless, a very large excess of reagent and longer reaction times were required in order to accomplish reaction completion (shorter reaction times or lesser amounts of PIDA lead to lower yields and incomplete consumption of starting material). Due to this, the present method is less convenient than the procedure using lead tetraacetate for multigram-scale preparations.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0061609/manifest

Comment

Related Items