Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Total synthesis of (±)-kelsoene and synthetic approaches to (±)-isomulinic acid Garcia, Josué Arturo Orellana 2003

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

Item Metadata

Download

Media
831-ubc_2003-86071X.pdf [ 11.76MB ]
Metadata
JSON: 831-1.0061225.json
JSON-LD: 831-1.0061225-ld.json
RDF/XML (Pretty): 831-1.0061225-rdf.xml
RDF/JSON: 831-1.0061225-rdf.json
Turtle: 831-1.0061225-turtle.txt
N-Triples: 831-1.0061225-rdf-ntriples.txt
Original Record: 831-1.0061225-source.json
Full Text
831-1.0061225-fulltext.txt
Citation
831-1.0061225.ris

Full Text

TOTAL SYNTHESIS OF ( ± ) - K E L S O E N E AND SYNTHETIC APPROACHES TO (±) - ISOMULINIC ACID by Josue Arturo Orellana Garcia B. Sc. (with honours), University of Ottawa, 1997 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Chemistry) We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A Fall 2003 © Josue Arturo Orellana Garcia In presenting this thesis in partial fulfilment of the requirements f o r 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 f o r extensive copying of this thesis f o r scholarly purposes may be granted by the head o f my department or by his or her representatives. It is understood that copying or publication o f this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada nat_ lQ^oto4>t^- ti %t>*3 DE-6 (2/88) II ABSTRACT The total synthesis of the unique tricyclic terpene natural product (±)-kelsoene was achieved in 15 steps from commercially available cyclopent-2-en-1-one (418) and the Afunctional reagent 4-chloro-2-trimethylstannylbut-1-ene (32). Bicycle rac-34 was constructed in two steps from enone 418 and served as the diquinane core of the natural product. The relative configuration of the carbon chirality center at C-8 was established through a highly diastereoselective homogeneous hydrogenation of rac-34 using Wilkinson's catalyst. Compound rac-59 was converted to enone rac-61 in four steps using established methods. A highly diastereoselective [2+2]-photocycloaddition of ethylene to enone rac-61 served to construct the four-membered ring of ketone rac-64 and establish the carbon chirality centers at C-2 and C-5. Ketone rac-64 was converted to alkene 143 using Lombardo's reagent. Hydroboration of alkene 143 and subsequent oxidation of the resulting material furnished alcohol 144. This material was transformed into (±)-kelsoene in five steps using established methods. 32 418 rac-34 rac-59 rac-61 H rac-64 143 144 (±)-kelsoene Ill The total synthesis of the tricyclic sesquiterpenoid natural product (±)-isomulinic acid (39) was attempted using four distinct strategies. Bicyclic enone 193 has the correct relative configuration at C-3 and C-5 and served as the platform for all the synthetic approaches attempted. Keto ester 189 was constructed from enone 193 in 8 steps. The alkylation of keto ester 189 with bifunctional reagent 36 furnished compound 226, which has the incorrect relative configuration at C-10, as the major product. Alkene 228 was viewed as a key intermediate in the second synthetic approach to (±)-isomulinic acid. It was hoped that a highly diastereoselective [2+2]-photocycloaddition of 228 to a suitable partner would furnish a tricyclic such as 229. Ketone 249 was prepared from enone 193 in 5 steps but did no yield access to alkene 228. Similarly, alkene 273 was constructed from enone 193 in 6 steps but did not yield access tb alkene 228. Enone 293 was efficiently prepared from enone 193 and was transformed into ketone 299 through a heterogeneous hydrogenation. Although ketone 299 possesses the correct relative configuration at C-9 and C-10, it could no be prepared in large scale due to its instability under the conditions used for its preparation. Ketone 309 was prepared from enone 193 using a sequence of reactions similar to those used to prepare ketone 299. Ketone 309 was elaborated to a bicycle 377, using a 12-step sequence of reactions. Due to an unforeseen side reaction, the ettempted homologation of the aldehyde moiety of 377 resulted in the formation of aldehyde 398. It could be shown that tricycle 402 could be prepared from 398. 309 377 398 402 V TABLE OF CONTENTS Abstract' ii Table of Contents v List of Tables vii List of Figures ix List of Schemes xii List of Symbols and Abreviations xix Acknowledgements xxiv Dedication xxvi 1 INTRODUCTION 1 1.1 General 1 1.1.1 Origins of Organic Chemistry 1 1.1.2 Organic Chemistry in the 2CT Century 2 1.2 Creativity in Total Synthesis 4 1.3 Background 8 1.4 Proposals 10 2 TOTAL SYNTHESIS OF ( ± ) - K E L S O E N E 12 2.1 Introduction 12 2.2 Biogenesis 13 2.3 Spectroscopic Determination of the Absolute Configuration of 15 (+)-Kelsoene 2.4 Reported Syntheses of Kelsoene 17 2.4.1 Total Synthesis of (±)-Kelsoene by Mehta et al. 17 2.4.2 Enantioselective Total Synthesis of (+)-Kelsoene and (-)-Kelsoene By Mehta et al. Determination of the Absolute Configuration of (+)-Kelsoene 20 2.4.3 Enantiospecific Total Synthesis of (-)-Kelsoene by Schulz and Coworkers 22 2.4.4 Total Synthesis of (±)-Kelsoene by Zhang et al. 25 2.4.5 Total Synthesis of (±)-Kelsoene by Bach et al. 29 vi TABLE OF CONTENTS - Continued 2.5 Total Synthesis of (±) -Kelsoene 32 2.5.1 Retrosynthetic Analysis 32 2.5.2 Total Synthesis of (±)-Kelsoene 33 2.5.3 Summary 52 3 SYNTHETIC APPROACHES TO (±) - ISOMULINIC ACID 53 3.1 Introduction 53 3.2 Isolation, Structure Determination and Biological Activity 54 3.3 Biogenesis 58 3.4 Synthetic Approaches to (±)- lsomulinic Acid 60 3.4.1 Introduction 60 3.4.2 First Synthetic Approach to (±)-lsomulinic Acid 61 3.4.2.1 Retrosynthetic Analysis 61 3.4.2.2 First Synthetic Approach to (±)-lsomulinic Acid 69 3.4.3 Second Synthetic Approach to (+)-lsomulinic Acid 87 3.4.3.1 Retrosynthetic Analysis 87 3.4.3.2 Second Synthetic Approach to (±)-lsomulinic Acid 93 3.4.4 Third Synthetic Approach to (±)-lsomulinic Acid 106 3.4.4.1 Retrosynthetic Analysis 106 3.4.4.2 Third Synthetic Approach to (±)-lsomulinic Acid 111 3.4.5 Fourth Synthetic Approach to (±)-lsomulinic Acid 122 3.4.5.1 Retrosynthetic Analysis 122 3.4.5.2 Fourth Synthetic Approach to (±)-lsomulinic Acid 125 3.5 Summary 167 4 EXPERIMENTAL 177 4.1 General 177 4.1.1 Data Acquisition, Presentation and Experimental Techniques 177 4.1.2 Solvents and Reagents 179 4.2 Experimental - Total Synthesis of (±) -Kelsoene 181 4.3 Experimental - Synthetic Approaches to (±)- lsomulinic Acid 196 5 REFERENCES 295 VII LIST OF TABLES 2.1 Comparison of the HREIMS and IR data reported for naturally occurring 50 (+)-kelsoene with the HRMS and IR data obtained for synthetic (±)-kelsoene 2.2 Comparison of the 1 3C NMR data reported for naturally occurring 50 (+)-kelsoene with the 1 3C NMR data obtained for synthetic (±)-kelsoene 2.3 Comparison of the 1H NMR data reported for naturally occurring 51 (+)-kelsoene and the 1H NMR data obtained for synthetic (±)-kelsoene 4.1 NMR data for (1 r7p5S*,6^)-6-methylbicyclo[3.3.0]octan;2-one (rac-59) 183 4.2 NMR data for (1 f?*,5S*,6S*)-6-methylbicyclo[3.3.0]octan-2 one (rac-59) 184 4.3 NMR data for (1 R*,2S*,5/?*,6S*,7S*,8S*)-2,8-dimethyl-6- 190 (hydroxymethyl)tricyclo[5.3.0.02,5]decane (144) 4.4 NMR data for (1 /?,>2S*>5rT,,6S*>7S*I8S*)-2l8-dimethyl-6- 191 (hydroxymethyl)tricyclo[5.3.0.02'5]decane (144) 4.5 NMR data for (1 S*,5rT,6fl*,9rT)-5-hydroxy-9-lsopropylbicyclo 208 [4.3.0]nonane-1 -carbaldehyde-2',2,-dimethylpropylene acetal (194) 4.6 NMR data for (1S*,5AT,6rT,9rT)-5-hydroxy-9-lsopropylbicyclo 209 [4.3.0]nonane-1 -carbaldehyde-2',2,-dimethylpropylene acetal (194) 4.7 NMR data for (1 S*,6fT,I9rT,)-9-isopropyl-5-oxobicyclo[4.3.0]nonane-1 - 216 carbaldehyde-2',2'-dimethylpropylene acetal (193) 4.8 NMR data for (1 ^6rT,9rT)-9-isopropyl-5-oxobicyclo[4.3.0]nonane-1 - 217 carbaldehyde-2',2'-dimethylpropylene acetal (193) 4.9 NMR data for (1 S*,6S*,9rT)-9-isopropyl-5-oxobicyclo[4.3.0]nonane-1 - 219 carbaldehyde-2',2'-dimethylpropylene acetal (220) 4.10 NMR data for methyl (1 rT\3S*,6S*7rT)-6-formyl-7-isopropyl-3-methyl-2- 222 oxobicyclo[4.3.0]nonane-3-carboxylate 2',2'-dimethylpropylene acetal (221) VIII LIST OF TABLES - Continued 4.11 NMR data for methyl (1 rT,3S*,6S*,7rT)-6-formyl-7-isopropyl-3-methyl-2- 223 oxobicyclo[4.3.0]nonane-3-carboxylate 2',2'-dimethylpropylene acetal (221) 4.12 NMR data for methyl (1 /?\6/^,9/?>9-isopropyl-4-xobicyclo[4.3.0]nonane- 233 1 -carboxylate (256) 4.13 NMR data for methyl (1rT,6fT,9rT)-9-isopropyl-4-xobicyclo[4.3.0]nonane- 234 1 -carboxylate (256) 4.14 NMR data for (1 fr*,6S*,9R*)-9-isopropyl-4-oxobicyclo[4.3.0]nonane- 237 1 -carbaldehyde (258) 4.15 NMR data for (1 fT,6^9/?>9-isopropyl-4-oxobicyclo[4.3.0]nonane- 238 1-carbaldehyde (258) 4.16 NMR data for methyl (1 fi*,6S*,9fl*)-9-isopropyl-4-oxobicyclo[4.3.0] 241 nonane-1-carboxylate (249) 4.17 NMR data for methyl (1 rT,6S*,9F?*)-9-isopropyl-4-oxobicyclo[4.3.0] 242 nonane-1-carboxylate (249) 4.18 NMR data for Methyl (1S*I4fl*,5S*>6S*,9rT,11/T)-4>il-epoxy-4-ethyl-9- 255 isopropyl-5[2-(tert-butyldimethylsiloxy)ethyl]bicyclo[4.3.0]nonane-1-carboxylate (302) 4.19 NMR data for Methyl (IS'^/T.S^.e^.g/T.llfTH.II-epoxy^-ethyl-g- 256 isopropyl-5[2-(tert-butyldimethylsiloxy)ethyl]bicyclo[4.3.0]nonane-1-carboxylate (302) 4.20 NMR data for methyl (1 ^2^5/^,6^9/r)-5-isopropyl-9-methyl-10- 293 oxotricyclo[7.5.0.02,6]tetradec-12-ene-6-carboxylate (402) 4.21 NMR data for methyl (1 S*,2S*,5rT,6S*,9rT)-5-isopropyl-9-methyl-10- 294 oxotricyclo[7.5.0.02,6]tetradec-12-ene-6-carboxylate (402) ix LIST OF FIGURES 1.1 Couper's formulas for acetic acid (1) and ethyl ether (2) 2 1.2 Impressive structures of calicheamicin yi' (3) and dodecahedrane (4) 4 1.3 Structure of the CP molecules 5 1.4 Methyl vinyl ketone as two different but-1-ene synthons 8 1.5 Methyl vinyl ketone as two different but-1-ene synthons 9 1.6 Rationale for a total synthesis of kelsoene (35) 11 1.7 Rationale for a total synthesis of isomulinic acid (39) 11 2.1 Kelsoene and related natural products 12 2.2 Differentially labeled a//o-aromadendranyl cations 15 2.3 Chiral alkenes derivatized with (aS)-MBCC (49) 16 2.4 Steric interactions between the C-8 methyl group and the C-6 aldehyde in 42 139 and rac-80 2.5 Approach of a hydroborating reagent to olefin 143 46 2.6 NOE correlations in alcohol 144 47 3.1 Different numbering systems for the mulinane carbocyclic skeleton 53 (150 and 151) and structure of (-)-isomulinic acid (39) 3.2 Structures of isomulinic acid (39) and mulinic acid (152) 54 3.3 Structures of 17-acetoxymulinic acid (153), mulinenic acid (154) and 55 mulinolic acid (155) 3.4 Structures of compounds 156 and 157 55 3.5 Structures of mulinol (158) and 11,12-epoxymuli-13-en-20-oic acid (159) 56 3.6 Structures of mulin-12,14-dien-11-on-20-oic acid (160), mulin-12-ene- 57 11,14-dion-20-oic acid (161) and 13-epimulinolic acid (162) 3.7 Structures of azorellanol (163) and compounds 164 and 165 57 3.8 Structures of compounds 199, 204, 205 and 206 71 3.9 Diagnostic 1H NMR data for the acetal moiety of 195 75 3.10 Spin system present in alkene 212 as determined using a selective 1D 75 TOCSY experiment X LIST OF FIGURES - Continued 3.11 Selected NOE difference data observed for alcohol 194 79 3.12 Selected NOE difference data observed for ketone 193 80 3.13 Selected NOE difference data observed for ketone 217 80 3.14 Selected NOE difference data for ketoester 221 83 3.15 Proposed intermediates susceptible to epimerization 84 3.16 Key proposed intermediates in the second retrosynthesis of (±)-isomulinic 87 acid (39) 3.17 Selected NOE difference data for keto aldehyde 258 96 3.18 Selected NOE difference data for enol ether 261 99 3.19 Structures of alkene 228, ketones 193 and 249, and alcohol 252 100 3.20 Selected NMR data for keto lactone 272 104 3.21 Structure of ketones 249 and 275 and of enolates 259 and 276 106 3.22 Potential preparation of ketone 279 from enone 278 107 3.23 Comparison of 1H NMR data for ketone 249 with 1H NMR data 115 for ketone 299 3.24 Selected 1H NMR data for alkenes 300 and 301 117 3.25 Two possible approaches to the double bond of alkene 300 by an oxidant 118 and expected shape of the transition state for the epoxidation of methylenecyclohexane by DMDO under steric control 3.26 Selected NOE difference data for epoxide 302 119 3.27 Structures of mulin-11,13-dien-20-oic acid (157),enone 293 and ketones 122 294 and 299 3.28 Comparison of 1H NMR data for ketones 249 and 299 with 127 1H NMR data for ketone 309 3.29 Selected structural data expected for ketone 322 and alkene 321 128 3.30 Selected spectral data for aldehydes 310 and 327 131 3.31 Selected NMR data for the major cyclopropyl enol ether obtained from the 138 Simmons-Smith reaction of enol ethers 342 and 343 LIST OF FIGURES - Continued xi 3.32 NOE difference data for acetal 385 152 3.33 Selected 1H NMR data for alkene 398 159 3.34 Selected NOE difference data for alcohols 416 and 417 165 3.35 Structures of isomulinic acid (39), mulin-11,13-dien-20-oic acid (157) and 167 tricyclic enone 402 3.36 Structures of tricycle 229 and alkene 228 170 xi LIST OF SCHEMES 1.1 Construction of diketone 11 from keto ester 7 6 1.2 Construction of diketone 11 from keto ester 7 7 1.3 The use of methyl vinyl ketone as a but-1 -ene a1 ,a3-synthon 9 1.4 Complex carbocycle construction using a trifunctional reagent 10 2.1 Konig and Wright's proposed biosynthesis of kelsoene 13 2.2 Nabeta's proposed biosynthesis of kelsoene 14 2.3 Generation of an axially-chiral nitrile oxide 15 2.4 Derivatization of (-)-B-pinene with (aS)-MBCC (49) 16 2.5 Derivatization of (+)-kelsoene with (aS)-MBCC 17 2.6 Construction of ketone rac-64 by Mehta and Srinivas 19 2.7 Total synthesis of (±)-kelsoene by Mehta and Srinivas 20 2.8 Enantioselective total synthesis of (+)-kelsoene and (-)-kelsoene by 21 Mehta and Srinivas 2.9 Chemical degradation of (+)-kelsoene 22 2.10 Construction of enf-64 from (R)-(+)-pulegone 24 2.11 Enantiospecific total synthesis of (-)-kelsoene from (R)-(+)-pulegone 25 2.12 Homo-Favorskii rearrangement of y-keto-p-toluenesulfonate 82 26 2.13 Acid catalyzed isomerization of bicyclo[3.1.1]heptanone to 26 bicyclo[3.2.0]heptanone 2.14 Preparation of enyne 93 27 2.15 Preparation of y-keto-p-toluenesulfonate 98 28 2.16 Completion of the synthesis of (±)-kelsoene by Zhang and Koreeda 29 2.17 Synthesis of 1,6-diene 106 30 2.18 Completion of the synthesis of (±)-kelsoene by Bach and Spiegel 31 2.19 Retrosynthetic analysis of (±)-kelsoene 33 2.20 Franck-Neumann's synthesis of alkenone rac-34 34 2.21 Synthesis of alkenone rac-34 34 xii LIST OF SCHEMES - Continued 2.22 Diastereoselective hydrogenation of alkenone rac-34 35 2.23 Synthesis of enone rac-61 37 2.24 PCC induced oxidative rearrangement of tertiary allylic alcohol 121 38 2.25 [2+2]-photocycloaddition of ethylene to 3-methylcyclopent-2-en-1 -one 39 (125) 2.26 [2+2]-photocycloaddition of ethylene to a,(3-unsaturated bicyclic lactam 39 127 2.27 [2+2]-photocycloaddition of ethylene to bicyclic enone 130 40 2.28 synthesis of tricyclic ketone rac-64 41 2.29 Homologation of ketone 136 to aldehyde 138 in a study towards the 42 synthesis of hirsutene 2.30 Taguchi's synthesis of an a-chloroaldehyde from cyclohexanone 43 2.31 Modified synthetic plan to aldehyde rac-80 44 2.32 Wittig olefination of ketone 131 44 2.33 Olefination of fenchone with Tebbe's reagent 45 2.34 Lombardo olefination of ketone rac-64 45 2.35 Synthesis of alcohol 144 47 2.36 Synthesis of ketone rac-81 48 2.37 Completion of the synthesis of (±)-kelsoene 49 2.38 Overview of the total synthesis of (±)-kelsoene 52 3.1 Proposed biosynthesis of the mulinane diterpenoids from a labdane 59 precursor 3.2 Retrosynthesis of isomulinic acid to mulin-11,13-dien-20-oic acid 61 3.3 Preparation of enone 38 using bifunctional reagent 36 62 3.4 Late stages of the synthetic problem 63 3.5 Retrosynthesis of mulin-11,13-dien-20-oic acid to enone 183 64 3.6 Retrosynthetic analysis of mulinenic (154) acid to ketone 182 65 3.7 Retrosynthesis of enone 183 to ketoester 189 66 xiv LIST OF SCHEMES - Continued 3.8 Dynamic equilibrium between ketones 190 and 191, structure of ketone 67 193, and keto and enol forms of keto ester 189. 3.9 Retrosynthesis of ketone 183 to ketoester 198 68 3.10 Synthesis of dione 199 69 3.11 Pathway for the Fe(lll) mediated alkylation of keto ester 198 70 3.12 Synthesis of enone 197 72 3.13 Synthesis of alkene 196 73 3.14 Preparation of aldehyde 211 74 3.15 Preparation of acetal 195 74 3.16 Hydroboration and oxidation of alkene 195 75 3.17 Oxidation of diol 214 to keto aldehyde 215 77 3.18 Proposed mechanism for the formation of diol 214 78 3.19 Oxidation of alcohols 194 and 213 to ketones 193 and 220 79 3.20 Optimized conditions for the hydroboration of alkene 195 to alcohol 194 81 3.21 Preparation of keto ester 189 81 3.22 Stereoselective alkylation of keto ester 189 82 3.23 Epimerization of ketoester 218 to ketoester 221 83 3.24 Alkylation of ketoester 189 with the bifunctional reagent 84 c/'s-5-iodo-1 -tri-n-butylstannylpent-1 -ene (36) 3.25 Preparation of alkenyl iodide 227 85 3.26 Overview of the de Mayo reaction 88 3.27 Possible construction of 237 through a de Mayo reaction 89 3.28 Overview of the photocycloaddition-reduction-cyclization sequence 90 developed by Baldwin 3.29 Possible construction of tricycle 246 using the 91 reaction sequence developed by Baldwin. 3.30 Retrosynthetic analysis of alkene 228 92 3.31 Preparation of alcohol 252 and lactone 253 93 XV LIST OF SCHEMES - Continued 3.32 Inversion of configuration at C-3 of allylic alcohol 252 94 3.33 Attempted homogeneous hydrogenation of alcohol 251 using Wilkinson's 95 catalyst and Crabtree's catalyst 3.34 Preparation of keto aldehyde 258 96 3.35 Directed homogeneous hydrogenation of allylic alcohol 251 using Brown's 97 catalyst 3.36 Preparation of ketone 249 97 3.37 Planned deprotonation of ketone 249 98 3.38 Formation of enol ether 261 98 3.39 Myers' method for reductive 1,3-transposition of allylic alcohols 100 3.40 Planned preparation of alkene 266 from allylic alcohol 252 101 3.41 Formation of hydroxylactone 268 from epoxyester 267 as demonstrated 101 by Barrero etal. 3.42 Implementation of Barrero's and Myers' methods in a planned synthesis of 102 alkene 225 3.43 Preparation of epoxide 270 103 3.44 Preparation of keto lactone 272 104 3.45 Attempted preparation of allylic alcohol 274 105 3.46 Application of McKenzie's rationale for the heterogeneous hydrogenation 108 of enones in a proposed synthesis of ketone 275 3.47 Preparation of a tertiary (284) aldehyde from a trisubstituted epoxide (283) 108 as demonstrated by Yamamoto and coworkers. 3.48 Application of Yamamoto's epoxide rearrangement to the 109 planned preparation of aldehyde 286 3.49 Potential conversion of the (2-terf-butyldimethylsiloxy)-ethyl group of 287 110 to an allyl group in 288 or a vinyl group in 289 3.50 Synthesis of enone 293 111 LIST OF SCHEMES - Continued xvi 3.51 Attempted hydrogenation of enone 293 to ketone 294 under 112 heterogenenous conditions 3.52 Hydrogenation of enone 293, formation of ketones 294, 298 and 299 114 3.53 Equilibration of ketones 294, 298 and 299 115 3.54 Preparation of alkenes 300 and 301 116 3.55 Epoxidation of alkenes 300 and 301 118 3.60 Lewis acid mediated rearrangement of epoxide 302 120 3.61 Pathway for the MABR catalyzed rearrangement of epoxide 302 to ketone 304 3.62 Retrosynthesis of aldehyde 311 to enone 308 123 3.63 Retrosynthesis of mulin-11,13-dien-20-oic acid (157) 124 to aldehyde 311 3.64 Synthesis of enone 308 125 3.65 Heterogeneous hydrogenation of enone 308 126 3.66 Equilibration of a mixture of compounds 320, 309 and 321 126 3.67 Proposed mechanism for the generation of alkene 321 129 from enone 308 under heterogeneous hydrogenation conditions 3.68 Preparation of aldehydes 310 and 327 from ketone 309 130 3.69 equatorial alkylation of aldehyde 328 as shown by Ireland and Mander in a 132 total synthesis of (±)-rimuene 3.70 Axial alkylation of aldehyde 331 as shown by Snider et al. in a total 133 synthesis of (+)-erinacine 3.71 Expected alkylation of aldehyde 310 134 3.72 Attempted alkylation of aldehydes 310 and 327 under the reaction 135 conditions reported by Ireland and Mander 3.73 Bransted acid catalyzed rearrangement of cycropropanols 136 to a-methyl aldehydes xvii LIST OF SCHEMES - Continued 3.74 Preparation of silyl enol ethers 342 and 343 136 3.75 Simmons-Smith cyclopropanation of silyl enol ethers 342 and 343 138 3.76 Preparation of aldehyde 311 139 3.77 Preparation of aldehyde 318 140 3.78 Preparation of alkene 317 141 3.79 Formation of homoallylic bromide 351 during the attempted 142 deprotection of methyl ether 350 using BBr3 3.80 Formation of lactone 358 and bromide 356 during the BBr3 mediated 143 deprotection of methyl ether 360 3.81 Potential side reaction during the BBr3 mediated 144 deprotection of methyl ether 317 3.82 Mechanism for the conversion of alkyl methyl ethers to alcohols and 145 alkyl iodides as proposed by Jung 3.83 Attempted deprotection of methyl ether 317 with an excess of TMSI 146 3.84 Proposed conversion of methyl ether 311 to alcohol 374 or alkyl iodide 147 375 3.85 Two possible preparations of alkene 377 148 3.86 Revised synthetic route to (±)-mulin-11,13-dien-20-oic acid 157 149 3.87 Preparation of iodide 375 150 3.88 Attempted E2 elimination of HI from alkyl iodide 375, preparation of 151 mixed acetal 385 3.89 Preparation of selenide 378 and alkene 377 153 3.90 Oxidation of Selenide 378 with Nal0 4 154 3.91 Oxidation of selenide 378 with ozone, formation of lactone 390 155 3.92 Preparation of alkenes 393 and 377 156 3.93 Proposed homologation of aldehyde 377 to aldehyde 379 158 3.94 Preparation of aldehyde 398 158 xvii LIST OF SCHEMES - Continued 3.95 Proposed mechanism for the Lewis acid catalyzed formation of alkene 160 398 from epoxides 394 3.96 Proposed construction of tricycle 402 161 3.97 Major catalytic pathway for the ring-closing metathesis of diene 405 163 using catalyst 404 3.98 Preparation of tricyclic alcohols 416 and 417 from aldehyde 398 164 3.99 Preparation of p,y unsaturated ketone 402 166 3.100 Preparation of alkene 195 168 3.101 Preparation of keto ester 189 169 3.102 Alkylation of keto ester 189 with bifunctional reagent 36 169 3.103 Preparation of silyl enol ether 258 171 3.104 Preparation of lactone 273 172 3.105 Preparation of ketone 304 173 3.106 Final synthetic approach to the (±)-isomulinic acid 175 xix LIST OF SYMBOLS AND ABBREVIATIONS 1D - one-dimensional 1H - proton 2H - deuterium 1 3C - carbon-13 9-BBN - 9-Borabicyclo[3.3.1]nonane a - below the plane of a ring or 1,2 relative position (3 - above the plane of a ring or 1,3 relative position 8 - chemical shift in parts per million or 1,5 relative position A - reflux y - 1,4 relative position X - wavelength pm - micrometer AcOH - acetic acid APT - Attached Proton Test (see J-mod) aq - aqueous ax - axial br - broad BF3«Et20 - boron trifluoride-dimethyl etherate complex BH3»SMe - borane-dimethyl sulfide complex BH3»THF - borane-tetrahydrofuran complex BuLi - butyllithium °C - degrees Celsius calcd. - calculated cm - centimeter COSY - (1H-1H)-homonuclear correlation spectroscopy CuBr»DMS - copper(l) bromide-dimethyl sulfide complex CuCN - copper(l) cyanide Cul - copper(l) iodide CuOTf - copper(l) trifluoromethanesulfonate LIST OF SYMBOLS AND ABBREVIATIONS - Continued C-X - carbon number X d - doublet DBU - 1,8-diazabicyclo[5.4.0]undec-7-ene DCI - deuterium chloride DCM - dichloromethane DIBAL-H - diisobutylaluminum hydride DMAP - 4-dimethylaminopyridine DMDO - dimethyl dioxirane DME - dimethoxyethane DMF - A/,A/-dimethylformamide DMSO - dimethyl sulfoxide E - entgegen (configuration) ee - enantiomeric excess ent - enantiomeric eq - equatorial equiv. - equivalents EtOAc - ethyl acetate EtOH - ethanol Et20 - diethyl ether g - grams(s) GC - gas chromatography GC-LRMS - coupled gas chromatography-low resolution mass spectrometry GC :MS - coupled gas chromatography-mass spectrometry GLC-MS - coupled gas-liquid chromatography-mass spectrometry h - hours(s) HCI - hydrochloric acid HMBC - 1H detected multiple bond heteronuclear multiple quantum Coherence HMPA - hexamethylphosphoramide xxi LIST OF SYMBOLS AND ABBREVIATIONS - Continued HMQC - 1H detected heteronuclear multiple guantum coherence HOAc - acetic acid HREIMS - high resolution electron impact mass spectrometry HRMS - high resolution mass spectrometry hu - light energy Hz - hertz (s"1) IR - infrared J - coupling constant in hertz J-mod - J-modulated spin echo (see APT) n J S n -H - n bond coupling constant between tin and proton nuclei (in hertz) KH - potassium hydride KHMDS - potassium bis(trimethylsilyl)amide KOH - potassium hydroxide LDA - lithium diisopropylamide LHMDS - lithium bis(trimethylsilyl)amide LiBr - lithium bromide LiCI - lithium chloride m - meter(s) M - Molar M + - molecular ion MABR - methylaluminum bis(4-bromo-2,6-di-tert-butylphenoxide) MCPBA - mefa-chloroperbenzoic acid Me - methyl MeCN - acetonitrile Mel - iodomethane, methyl iodide MeLi - methyllithium MeOH - methanol Mg - milligram(s) MHz - megahertz LIST OF SYMBOLS AND ABBREVIATIONS - Continued min minute(s) mL milliliter(s) mm millimeter(s) mmol millimole(s) moi mole(s) mp melting point NaH sodium hydride NaOAc sodium acetate NaOH sodium hydroxide NaOMe sodium methoxide NMO A/-methylmorpholine /V-oxide NMR nuclear magnetic resonance NOE nuclear Overhauser effect PCC pyridinium chlorochromate PDC pyridinium dichromate PhH benzene PhSeCI phenylselenenyl chloride PPL porcine pancreatic lipase ppm parts per million psi pounds per square inch p-TSA para-toluenesulfonic acid q quartet R rectus (configuration) rac racemic s singlet S sinister (configuration) r.t. room temperature sat. saturated TBSCI tenf-butyldimethylsilyl chloride LIST OF SYMBOLS AND ABBREVIATIONS - Continued t - tripet t - tertiary TBAF - tetrabutylammonium fluoride TBDMSCI - tert-butyldimethylsilyl chloride f-BuOH - terf-butyl alcohol f-BuOK - potassium terf-butoxide f-BuOMe - tert-butyl methyl ether tert - tertiary TFA - trifluoromethanesulfonic acid THF - tetrahydrofuran TLC - thin layer chromatography TMSCI - trimethylsilyl chloride, chlorotrimethylsilane TMSI - trimethylsilyl iodide, iodotrimethylsilane TMSOTf - trimethylsilyl trifluoromethanesulfonate TOCSY - total correlation spectroscopy TPAP - tetrapropylammonium perruthenate TsCI - para-toluenesulfonyl chloride UBC - University of British Columbia UV - ultraviolet -ve - negative +ve - positive W - watts Z - zusammen (configuration) ± - racemic ACKNOWLEDGEMENTS xxiv The completion of this thesis brings to a close a remarkable six-year period in my life. A great many people have influenced me in ways that neither they nor I imagined. Professor Edward Piers is gratefully acknowledged for his supervision and mentorship. I was granted more academic freedom than any graduate student could ever wish and have learned a great deal during my time in his research laboratory. I regard him as a great role model and feel fortunate to have been his pupil. My time in Professor Piers' research laboratory witnessed a number of changes. I am grateful to all the members who contributed in creating the vibrant atmosphere of the early years and the cohesiveness of my final year. Dr. Krystyna Skupinska, Dr. Michael Gilbert, Dr. Sebastien Caille, Dr. Shawn Walker, Dr. Robert Britton, Dr. Stephen Lau, Cristian Harrison and Diana Wallhorn deserve special mention. My research benefited greatly from numerous discussions with members of the department of chemistry. The insights and ideas provided by Professor David Perrin and Dr. Robert Britton are specially acknowledged. The technical support provided by the UBC NMR and MS facilities was invaluable. Mrs. Liane Darge of the NMR facility is specially thanked for sharing her expert advice and providing technical assistance on countless occasions. This thesis was proofread in part by Dr. Stephen Lau, Cristian Harrison and Carrie Gillon, and in whole by Professor Michael Wolf. Their help in this regard is gratefully noted. Any and all errors of a scientific nature or misuse of the English language are my own. XXV I have been fortunate to receive generous financial support in the form of a scholarship from the Natural Sciences and Engineering Research Council of Canada (PGS A and B) and a continued Gladys Estella Laird Research Fellowship from the University of British Columbia. For their own very special way of making me feel immensely fortunate while in Vancouver I thank Jennifer Yip, Victoria Yip, Krystyna Skupinska and David Vocadlo, Shawn Walker, David Fenwick, Michael Fenster, Carrie Gillon, Cristian Harrison, Harry Brummer III, Sarah Everts, Manuela Scharph, Spencer Williams, Michael Gilbert and Sara Falkner, Robert Britton and Tiffany Wright-Britton, Mark Klinger, Roger Linington, Todd Barsby and Ginger Warden, Robert Gerl, Carl Rye, Ladan Mehranvar and Megan McCusker. All the members of The Elm Street Society, The Flipper Babies Ultimate Club and The British Columbia Brew Crew are thanked for filling my idle time in creative and often unpredictable ways. Bianca, Jenny, Mom and Dad; I could not have persevered without your love, support and inspiration. xxvi Este tesis es dedicado, con mucho carino, a mi familia 1 1. Introduction 1.1 General 1.1.1 Origins of Organic Chemistry 1 2 Until the late 1700s organic chemistry was a primitive science. Indeed, no one could claim to know how organic and inorganic chemicals differed. Some even believed that organic chemicals possessed a vital force and could not be prepared in the laboratory. In 1786 Lavoisier defined organic chemicals as combinations of oxygen with radicals that contained carbon and hydrogen. If the compounds were of animal origin, then they could also contain nitrogen and phosphorous. He did not, however, specify what a radical was. It was not until 1828, with Friedrich Wohler's preparation of urea from ammonium cyanate, that the notion of a vital force began to lose credence. Conclusive evidence against this concept was provided in 1845 by Kolbe and Berthelot, who were able to synthesize organic compounds directly from the elements. Some argue that Wohler's preparation of urea marks the birth of synthetic organic chemistry. However, the lack of an understanding of the relationship between carbon, hydrogen and other elements suggests that the emergence of synthetic organic chemistry as a rigorous science occurred much later. New theories addressing the nature of organic compounds began to emerge in the second part of the 19th century. The first breakthrough came in a report entitled "A New Chemical Theory" authored in 1858 by Archibald Scott Couper. In it, the 27 year old stated: I propose to consider the single element carbon. This body is found to have highly distinguished characteristics: 1. It combines with equal numbers of equivalents of hydrogen, chlorine, oxygen, sulfur, etc. 2. It enters into chemical union with itself. These two properties in my opinion explain all that is characteristic of organic chemistry. 1. Introduction 2 Although obscure, Couper's report was revolutionary. He defined structure as the connection between atoms to form molecules (Figure 1.1) at a time when Kekule considered the concept of atoms a "convenient fiction". He even speculated on the composite nature of the atom 40 years before the work of J. J. Thompson. Unfortunately, Couper's work remained neglected until 1900 and, in the meantime, Kekule's ideas formed the basis for work on structural covalent chemistry. Figure 1.1: Couper's formulas for acetic acid (1) and ethyl ether (2) Ideas regarding the tetravalent nature of the carbon atom and its implications to the structure of organic materials were advanced by Pasteur in 1860, Butlerow in 1862, and (surprisingly) Kekule in 1867. They did not gain credibility until 1874 when Van't Hoff and Le Bel, in an attempt to explain optical activity, independently proposed that a tetrahedral carbon atom with four different substituents could exist in two forms. In spite of skeptics like Kolbe, who proclaimed that these ideas were "not far removed from belief in witchery and ghost-rapping", Van't Hoffs views were widely accepted by 1894, except by those who did not embrace the atomic hypothesis. This situation would change upon the revolution in physics which demonstrated the existence of an atom and its composite nature. 1.1.2 Organic Chemistry in the 20th Century Organic chemistry flourished in the last century and is now, by most measures, a mature science. The Nobel Prize for Chemistry3 serves as a convenient chronicle of major achievements in the field. A comment on the state of the art in organic chemistry is provided in the 1912 lectures. In his presentation speech Soderbaum4 claimed that "no method of organic synthesis superior to that of Grignard's is known", and queried if we would "ever learn how to produce artificially alkaloids or vegetable organic bases?" Grignard5 had a more positive outlook and hinted at advances to come; he noted C - O - O H I o2 C - H 3 | H2 H 2 | C—H 3 H 3—C 1 2 1. Introduction 3 ...a new group of very interesting syntheses, namely asymmetrical synthesis....(McKenzie) esterified benzoylformic, pyruvic and laevulic acids by /-menthol, and he caused the organic magnesium compound to react with the CO of each of these active esters. Then by saponifying the products obtained in order to eliminate the menthol he obtained slightly levorotatory alcohol acids. It would have been difficult for Grignard and Soderbaum to imagine the breakthroughs that would occur in organic chemistry during the remainder of the century. The number of methods available to the synthetic chemist has grown at an intimidating rate and was estimated to be greater than 35,000 in 1990.6 The Swedish Royal Academy of Sciences has recognized contributions in this area on a number of occasions. Notable recipients of the prize include Alder and Diels7 for their "discovery and development of the synthesis of dienes," Brown8 for his work on borohydrides and organoboranes, and Wittig8 for the development of phosphorous ylides. Methods for asymmetric synthesis abound, and for their work in this field Knowles,9 Noyori10 and Sharpless11 received the Nobel Prize in 2001. The synthesis of organic materials is no longer viewed with the pessimism of Soderbaum. To conceive molecules of such astonishing complexity as calicheamicin y-i1 (3)12 and awesome symmetry as dodecahedrane (4) (Figure 1.2), would have been difficult in 1912, and their chemical synthesis13"15 would have been unimaginable. Yet, by the 1960's the art of organic synthesis had reached new heights, and in 1965 Woodward's sophisticated contributions were acknowledged with the Nobel Prize. Thirty five years later Corey was credited with "turning the art of chemical synthesis into a science".6 1. Introduction 4 o MeSSS^ / = > NHC0 2 Me Me Me O OMe OH ™ O J ^ O n H M e \ n H 4 -HO MeO calicheamicin yi (3) dodecahedrane (4) Figure 1.2: Impressive structures of calicheamicin y/ (3) and dodecahedrane (4) These accomplishments by no means indicate that organic chemistry is a dying science.16 The need to develop new methods and strategies for the construction of organic compounds is continually fueled by the elucidation of novel structures found in nature. Moreover, Wender's notion of the "ideal synthesis"17 and Trost's concept of "atom economy"18 remain lofty goals, and the most celebrated synthetic achievements often exceed the 25-step "rational limit" proposed by Hudlicky.19 1.2 Creativity in Total Synthesis The total synthesis of natural products has occupied a prominent position in the field of organic chemistry over the last half century, and many accounts outlining the value of this exercise have been published. Experience teaches that there is arguably no better training ground for young synthetic organic chemists than the total synthesis of a challenging natural product. Furthermore, the complexity of the target molecule may be such that the development of new synthetic methods or the formulation of an innovative synthetic strategy may be required. Often, a total synthesis serves as a rigorous test for a newly developed synthetic method. There are even some, including this author, who argue that the total synthesis of natural products has intrinsic value as a form of art. Synthetic chemists express their creative spirit in various forms. Some choose to construct a target in the most ingenious ways, often assembling intricate structures 1. Introduction 5 through a series of transformations occurring in a single synthetic operation. Others approach the challenge of a daunting synthesis with a sense of adventure, and cherish the opportunity to develop new technology during the journey. Natural products which possess both elaborate carbocyclic frameworks and unusual functionality, and which display potentially useful biological activity, are the targets of choice for total synthesis. An instructive example follows. In 1995 the two novel structures known as CP-225,917 (5) and CP-263,114 (6) were disclosed20,21 and immediately attracted the attention of synthetic chemists (Figure 1.3). The key structural features embedded in these molecules include a bicyclo[4.3.1]decane system, a bridgehead double bond, a maleic anhydride, a y-hydroxylactone, an cc-hydroxyketone and a quaternary center. The y-hydroxylactone and the a-hydroxyketone moieties found in CP-225,917 combine to form a tetrahydropyran system in CP-263,114. ^ C 0 2 H ^ C 0 2 H (+)-CP-225,917 (5) (-)-CP-263,114 (6) Figure 1.3: Structure of the CP molecules Both CP-225,917 and CP-263,114 inhibit squalene synthase and farnesyl protein transferase. Squalene synthase catalyzes the condensation of two molecules of farnesyl pyrophosphate to squalene, an intermediate in the biosynthesis of cholesterol. Thus, it is possible that the CP molecules may lead to new cholesterol lowering drugs. Farnesyl protein transferase mediates farnesylation of the protein p21. Mutated p21 is rendered permanently active and thus leads to unregulated cell growth and division. As farnesyl protein transferase inhibitors, the CP molecules may interfere with a step crucial to p21 activity and the carcinogenic process. Clearly, these molecules are ideal targets for total synthesis. 1. Introduction 6 A significant body of literature related to the synthesis of the CP molecules exists, however the work of Shair2 2 , 2 3 is conspicuous for its creativity. Shair's approach to (+)-CP-263,114 (enf-6) is ingenious on two counts: 1. His incisive retrosynthetic analysis allows the construction of a carbocycle reminiscent of the CP molecules from a relatively simple material in a single synthetic operation. 2. His conditions for establishing the quaternary carbon chirality center constitute a new method for chemical synthesis. The enantiomerically enriched keto ester 7 is obtained in 3 steps from 2-iodocyclopent-2-en-1-one and the adduct resulting from its reaction with the Grignard reagent 8 undergoes a remarkable series of transformations (Scheme 1.1). The 1,5-diene embedded in bromomagnesium alkoxide 9 undergoes the well known anion accelerated oxy-Cope rearrangement, and yields a bromomagnesium enolate (10). This material collapses to diketone 11 upon a transannular Dieckmann-like cyclization. anion-accelerated oxy-Cope rearrangement Scheme 1.1: Construction of diketone 11 from keto ester 7 The synthesis continues with the elaboration of diketone 11 to enol carbonate 12 in 6 steps. A likely mechanism for the number of transformations which result in the conversion of 12 into acid 15 is depicted in Scheme 1.2. Compound 12, upon TMSOTf 1. Introduction 7 promoted ionization of the enol carbonate moiety, releases a silylketene acetal (13) and a carbomethoxenium ion fragment. Recombination of these two units results in the formation of a new carbon-carbon bond in a manner reminiscent of the Fries rearrangement, and establishes the quaternary carbon chirality center. The resulting intermediate (14) undergoes a number of deprotection and cyclization reactions under the influence of TMSOTf, which culminate in the formation of polycyclic acid 15. This material leads to the target compound in five synthetic operations. Fries-like rearrangement Me-O, Scheme 1.2: Conversion of enol carbonate 12 to acid 15 This relatively short synthesis of such a complex natural product was facilitated by the preparation of carbocycle 11, which resembles the natural product, at an early point in the sequence. This, in turn, resulted from a penetrating retrosynthetic analysis which called for the use of three reactions, namely a Grignard addition, an anion accelerated oxy-Cope rearrangement and a Dieckmann-like cyclization, in a single synthetic operation. In addition, the challenge presented by the quaternary carbon chirality center resulted in the development of a new synthetic method. 1. Introduction 8 1.3 Background Many small molecules with two or more functional groups are useful in the construction of carbocycles. Methyl vinyl ketone (16, Figure 1.4) is perhaps the best known example of such compounds and its use in the Robinson annulation is a fundamental method for the construction of six-membered rings. The polar nature of the carbon-oxygen double bond imposes a reactivity pattern upon the carbon skeleton of methyl vinyl ketone. As a result, C-1 and C-3 have donor (d) properties, as defined by Seebach,24 and are reactive towards electrophiles. Similarly, C-2 and C-4 have acceptor (a) properties and are susceptible to attack by nucleophiles. In the Robinson annulation, methyl vinyl ketone may be regarded as the synthetic equivalent of the but-1-ene a1,d4-synthon25 (17). O a 16 17 18 Figure 1.4: Methyl vinyl ketone as two different but-1-ene synthons An annulation sequence developed by Corey and used in a synthesis of the natural product helminthosporal26 demonstrates the (less common) use of this versatile reagent as a but-1-ene a1,a3-synthon (18, Scheme 1.3). Triethylamine promoted reaction of the (-)-carvomenthone derived keto aldehyde (19) with methyl vinyl ketone results in the axially alkylated material 20, which is deformylated under mildly basic conditions to provide diketone 21. Subjection of this diketone to Robinson annulation conditions would result in the fused bicycloalkenone product. In this example, treatment of 21 with BF3»Et20 results in the formation of the bridged bicycle 22. This mode of cyclization is observed as a side reaction in the Robinson annulation. Although the desired bridged bicycle was obtained and led to the synthesis of the target molecule, the low yield for this transformation (21 to 22) detracts from the utility of this method. 1. Introduction 9 H CHO 19 Et 3N, 16 r.t:, 3 d 71% 20 2% K 2 C 0 3 , EtOH, A, 16 h 70% .0 CH 2 CI 2 l r.t., 16 h BF3« Et 2 0 32% 21 ° 22 Scheme 1.3: The use of methyl vinyl ketone as a but-1-ene a1,a3-synthon2 6 The development of more sophisticated small molecules for the construction of carbocycles has been a research theme in this group for some time. Indeed, a significant number of bifunctional reagents have been prepared, and many have been featured prominently in the synthesis of natural products, including crinipellin27 (23) and variecolin28 (24, Figure 1.5). In some cases trifunotional reagents have served to construct relatively complex polycyclic materials using a short sequence of reactions (Scheme 1.4). Alkylation of vinyligous ester 26 with the trifunctional reagent 25 proceeds in good yield under standard conditions to generate compound 27. Reduction of the carbonyl moiety in 27, and treatment of the resulting material with a catalytic amount of acid reveals the Figure 1.5: Structures of crinipellin (23) and varicolin (24) 1. Introduction 10 latent conjugated enone to provide 28. Copper cyanide mediated intramolecular addition of the alkenyltrimethylstannane function to the ct,p-unsaturated ketone results in the formation of the c/'s-fused bicycle 29, albeit in modest yield. Finally, intramolecular alkylation of the potassium enolate derived from 29 provides the tricyclic compound 30 as a single diastereomer. Thus, the use of trifunctional reagent 25, which served as the frans-hex-2-ene a\d3,a6-synthon (31), facilitated the preparation of carbocycle 30 from vinyligous ester 26 in only 5 synthetic operations. Scheme 1.4: Complex carbocycle construction using a trifunctional reagent 1.4 Proposals As part of a continuing effort to highlight the use of bifunctional reagents developed in this laboratory, two total synthesis projects were undertaken. The structure of the natural product (+)-kelsoene (35, Figure 1.6) was elucidated30 in 1996 and its unique tricarbocyclic skeleton attracted our attention. The bifunctional reagent 4-chloro-2-trimethylstannylbut-1-ene31 (32) has been shown to serve as the synthetic 1. Introduction 1 1 equivalent to the but-1-ene d2,a4-synthon (33) in a cyclization sequence leading to bicycle rac-34.3 2 The structural similarity between kelsoene (35) and bicycle 34 suggested that an expedient total synthesis of the natural product could be achieved, thus showcasing the utility of bifunctional reagent 32. Figure 1.6: Rationale for a total synthesis of kelsoene (35) The mulinane class of natural products comprises 12 members all having the carbocyclic framework embedded in isomulinic acid3 3 (39, Figure 1.7). A recently developed cycloheptane annulation sequence34 based on the use of the bifunctional reagent c/'s-5-iodo-1-tributylstannylpent-1-ene (36) provides expedient access to bicyclic systems such as compound 38. In this context reagent 36 may be viewed as the synthetic equivalent of the pent-1-ene d1,a5-synthon (37). It was hoped that this method would be part of a successful strategy leading to the synthesis of a number of mulinane natural products. OH 36 37 38 (-)-isomulinic acid (39) Figure 1.7: Rationale for a total synthesis of isomulinic acid (39) 12 2. Total Synthesis of (±)-Kelsoene 2.1 Introduction In 1996 Wright et al. reported the isolation of the novel sesquiterpenoid (+)-kelsoene (35) from the dichloromethane extracts of Cymbastella hooperi, a marine sponge collected from the Kelso reef, Great Barrier Reef, Australia.30 In this initial report the authors noted that, at the time of its isolation, the 6-isopropenyl-2,8-dimethyltricyclo[5.3.0.02,5]decane35 kelsoane carbon skeleton had no precedent in the natural products literature, and provided only a partial assignment of its relative configuration. Indeed, only a single natural product, sulcatine G 3 6 (40), was known to possess the tricyclo[5.3.0.02,5]decane carbocyclic system. The following year the same group showed that kelsoene possesses the relative configuration depicted in 35. 3 7 In 1997 it was reported that this "most unusual carbocyclic skeleton" is shared by the tetraterpenoid poduran (41 ),38 which was isolated from the springtail Podura aquatica. Kelsoene has been subsequently isolated from the liverworts Ptychanthus striatus,39'40 Tritomaria quinquedentata,^ and Calipogeia muelleriana42 (Figure 2.1). .-15 (+)-kelsoene (35) sulcatine G (40) poduran (41) Figure 2.1: kelsoene and related natural products Due to its unique and unprecedented structure, and in order to determine its absolute configuration, kelsoene has been the subject of biosynthetic,39,43 spectroscopic,44 and synthetic studies.45"50 2., Total synthesis of (±)-kelsoene 13_ 2.2 Biogenesis Shortly after their initial report Konig and Wright proposed a biosynthetic path to kelsoene.37 They reasoned that kelsoene might be derived from a guaiane skeleton (42), which must first undergo a ring closure between C-6 and C-11 to generate an allo-aromadendrane species (43). The bond between C-7 and C-11 is then broken, and a new bond between C-7 and C-10 is formed to generate the kelsoane skeleton (44, Scheme 2.1). 42 43 44 Scheme 2.1: Konig and Wright's proposed biosynthesis of kelsoene A year later Nabeta et a/. 3 9 , 4 3 outlined a more explicit biosynthetic path to kelsoene from farnesyl diphosphate (46, Scheme 2.2). In their study, cells of the liverwort Ptychanthus striatus were incubated with [2-2H2]-, [5-2H2]-, and [2-13C]-potassium mevalonate (45), and the labeling patterns of the biosynthetically labeled 2 H-or 13C-kelsoene were determined by GLC-MS, 2 H- and 1 3C-NMR analyses. Thus, initial cyclization of farnesyl diphosphate (46) from the re-face forms the (-)-germacradienyl cation (47). This cation then cyclizes stereospecifically between C-1 and C-5, and between C-6 and C-11 to furnish the (+)-a//o-aromadendranyl cation (48). Finally, the cyclopropane ring is cleaved and a ring closing reaction between C-7 and C-10 ensues, with concomitant loss of a proton, to afford kelsoene. 2. Total synthesis of (±)-kelsoene 14 Curiously, Nabeta et al. observed randomization of the 1 3C label between C-12 and C-13 (Scheme 2.2). They proposed two possible explanations for the observed randomization. The first involves an "equivalent loss of a proton from the gem-dimethyls". Thus, loss of a proton through path a would yield kelsoene labeled at the terminal sp2 carbon of the isopropenyl moiety. Alternatively, loss of a proton through path b would place the 1 3C label at the sp3 carbon of the isopropenyl moiety of kelsoene. A second and more likely explanation invokes free rotation about the C-7 and C-11 bond in 47. Thus a stereospecific ring closure between C-11 and C-6 may occur when the 1 3C label is on the (3-face of germacradienyl cation to afford 48a. Alternatively, the same stereospecific ring closure may occur when the 1 3C label is on the a-face of the germacradienyl cation to afford 48b (Figure 2.2). A stereospecific elimination-cyclization sequence would lead from cations 48a and 48b to 35a and 35b. 2. Total synthesis of (±)-kelsoene 15 2.3 Spectroscopic determination of the absolute configuration of (+)-kelsoene. In 1998 Fukui et al. disclosed a new method for the recognition of chiral organic compounds by NMR spectroscopy.51 Their research in this field was prompted by the lack of chiral derivitazing agents for chiral alkenes. This new method relies on the well known dipolar cycloaddition of nitrile oxides to electron rich alkenes. Thus, when the axially chiral reagent 2'-methoxy-1,1'-binaphthalen-2-carbohydroximoyl chloride ((aS)-MBBC, 49) is treated with a tertiary amine, a reactive chiral nitrile oxide (50) is generated (Scheme 2.3). (aS)-MBCC (49) 50 Scheme 2.3: Generation of an axially-chiral nitrile oxide5 1 In principle, this chiral nitrile oxide can form a covalent adduct with a chiral olefin of undetermined absolute configuration (Scheme 2.4). The relative stereochemistry of the adduct can then be assigned based on NMR spectroscopic data. Thus, since the 2. Total synthesis of (±)-kelsoene 16 absolute stereochemistry of the chiral reagent is known, the absolute configuration of the alkene in question can be assigned. This hypothesis was substantiated by a series of experiments carried out on adducts formed between nitrile oxide 50 and (-)-B-pinene (51), (-)-a-pinene (53) and an olefin derived from (-)-menthone (54) (Figure 2.3). 51 52 Scheme 2.4: Derivatization of (-)-B-pinene with (aS)-MBCC (49) 51 53 54 enf-35 Figure 2.3: Chiral alkenes derivatized with (aS)-MBCC (49)bl In view of the fact that kelsoene possesses an olefin as its single functional group, Nabeta et al. decided to apply Fukui's method to determine its absolute configuration.44 Thus, treatment of a mixture of (+)-kelsoene and the axially chiral reagent (aS)-MBBC (49) with triethylamine for 14 days provided a mixture of covalent adducts 55 and 56 in a 62:38 ratio and 10% overall yield (Scheme 2.5). Careful analysis of the spectroscopic data derived from a substance believed to be 55 led these authors to (incorrectly) assign the absolute configuration of (+)-kelsoene as that depicted in enf-35 (Figure 2.3). 2. Total synthesis of (±)-kelsoene 17 (+)-kelsoene 55 56 Scheme 2.5: Derivatization of (-t-)-kelsoene with (aS)-MBCC44 In retrospect, the fact that two covalent adducts were formed in a nearly 3:2 ratio should have cast doubts on the conclusions reached by Nabeta et al. Indeed, careful inspection of the three substrates used in Fukui's study (51, 53 and 54) reveals that in all cases at least one of the sp2 centers resides in the carbocyclic core of the molecule. Furthermore, based on the structure of these substances, addition of the nitrile oxide is expected to occur with a high degree of facial selectivity. Clearly, neither of these conditions is met in the spectroscopic study of (+)-kelsoene. 2.4 Reported Syntheses of kelsoene 2.4.1 Total synthesis of (±)-kelsoene by Mehta and Sr inivas. 4 5 4 6 It is evident that the unusual nature of the kelsoane skeleton, the presence of a single functional group and its six contiguous carbon chirality centers combine to make kelsoene a significant synthetic challenge. The first foray into a total synthesis of this natural product was reported by Mehta and Srinivas in 1999. 4 5 Their synthetic approach was clearly influenced by the recognition that quick access to the diquinane core of kelsoene would facilitate the construction of the natural product in an expedient fashion. 2. Total synthesis of (±)-kelsoerte 18 The sequence begins with the preparation of the C2-symmetric diketone (rac-58) from commercially available 1,5-cyclooctadiene (57) using the method described by Henry et al.52 (Scheme 2.6). Wittig mono-methylenation of this diketone provided alkenone rac-34 in good yield. Hydrogenation of the exocyclic olefin under heterogeneous conditions proceeded, as expected, from the convex surface of the bicycle with a good degree of facial selectivity (60% diastereomeric excess) and in excellent yield. The bicyclic ketone (rac-59) was dehydrogenated using the method established by Saegusa,53 and the resulting enone (rac-60) was alkylated with methyllithium to generate a tertiary allylic alcohol. This unstable intermediate was treated with pyridinium chlorochromate to furnish the desired enone (rac-61) in good yield. At this stage the authors elected to construct the four membered ring of kelsoene using a [2+2]-photocycloaddition between enone rac-61 and trans-l ,2-dichloroethylene. Protection of the carbonyl function of the resultant tricycle (62) was necessary for the reductive dehalogenation of the four membered ring to proceed efficiently. Finally, hydrogenation of the cyclobutene moiety in 63 followed by deprotection of the carbonyl group provided the desired tricyclic ketone rac-64. Having achieved a synthesis of the tricarbocyclic core of kelsoene, Mehta and Srinivas attempted the installation of the isopropenyl moiety in the natural product. Unfortunately, ketone rac-64 proved to be "extremely refractory" towards various reagents, a behaviour which was attributed to the sterically encumbered environment around the carbonyl group. 2. Total synthesis of (±)-kelsoene 19 57 i) PdCI2, Pb(OAc)4, AcOH, 70% ii) KOH, MeOH, 95% iii) PCC, 85% O rac-58 o MePPh 3 l, f-BuOK, PhH, 65% H H O rac-34 H2(g), 10% Pd/C, EtOAc, 72% i) MeLi, Et 20 ii) PCC, 80% transA ,2-dichloroethylene, C 6 H 1 2 , hu, Pyrex, 85% rac-60 i) LHMDS, TMSCI ii) Pd(OAc)2, MeCN 80% (2 steps) H rac-59 O i) 1,2-ethanediol, p-TSA, PhH, 85% ii) sodium-Cl naphthalenide, 62 DME, 95% i) H2(g), 10% Pd/C, EtOAc, 95% ii) Amberlyst-15, acetone, 85% Scheme 2.6: Construction of ketone rac-64 by Mehta and Srinivas 45 Undeterred by these results, the authors revised their strategy towards the synthesis of kelsoene.4 6 In order to enhance the reactivity of ketone rac-64 it became necessary to alleviate the steric congestion around its carbonyl group. It was postulated that simply transforming ketal 63 into ketone rac-65 would afford a substrate that could be elaborated to the natural product (Scheme 2.7). Indeed, ketone rac-65 could be converted into methyl enol ether 66 in excellent yield using a Wittig olefination protocol. Acid hydrolysis of this substrate provided a single aldehyde (67) as the major product. This aldehyde was easily alkylated with methyllithium to provide a mixture of epimeric alcohols (68) which could be oxidized to the corresponding ketone (69) in excellent yield. This material was readily hydrogenated, and the resulting ketone (rac-70) was converted to (±)-kelsoene through a Wittig methylenation. 2. Total synthesis of (±)-kelsoene 20 Amberlyst-15 acetone, 85% PCC, DCM 90% 69 H2(g), 10% Pd/C, EtOAc, 98% MePPh 3 l f-BuOK, Et 2 0, 80% (±)-kelsoene MeOCH 2PPh 3CI NaOf-Am, Et 2 0 85% MeLi, Et 2 0 97% H jOMe H 66 35% HCI0 4, Et 2 0, 90% Scheme 2.7: Total synthesis of (±)-kelsoene by Mehta and Srinivas 2.4.2 Enantioselective total synthesis of (+)-kelsoene and (-)-kelsoene by Mehta and Srinivas. Determination of the absolute configuration of (+)-kelsoene. 4 7 Having achieved a total synthesis of (±)-kelsoene, Mehta and Srinivas4 7 set out to determine the absolute configuration of this natural product using the synthetic sequence previously developed. To accomplish this, diketones 58 and enf-58 were required with a high degree of enantiomeric purity. Thus the bicyclic diol rac-71 was 2. Total synthesis of (±)-kelsoene 21 constructed in racemic form using Henry's procedure52 (Scheme 2.8). A lipase catalyzed kinetic resolution was employed to provide diol 71 and diacetate 72 in good yield and excellent enantiomeric excess. Diol 71 was readily oxidized to diketone 58, while diacetate 72 was first hydrolyzed under basic conditions to the corresponding diol and subsequently oxidized to diketone enf-58. The absolute configuration of the chiral diones 58 and enf-58 had been previously determined. HQ H i) PdCI 2, Pb(OAc)4, AcOH, 70% ii) KOH, MeOH, 95% H OH rac-71 57 vinyl acetate, Amano lipase PS-30 f-BuOMe PCC, DCM 94% i) KOH, MeOH ii) PCC 90% O H H 58 O enf-58 (+)-kelsoene (35) (-)-kelsoene (enf-35) Scheme 2.8: Enantioselective total synthesis of (+)-kelsoene and (-)-kelsoene by Mehta and Srinivas 4 7 2. Total synthesis of (±)-kelsoene 22 Diketone 58 was subjected to the 15-step synthetic sequence used for the synthesis (±)-kelsoene and provided (+)-kelsoene (35) as determined by a measurement of its optical rotation. In a similar fashion, diketone enf-58 provided (-)-kelsoene (enf-35). Thus the absolute configuration of naturally occurring (+)-kelsoene could be unambiguously assigned as that depicted in 35. 2.4.3 Enantiospecific total synthesis of (-)-kelsoene by Schulz and coworkers. Determination of the absolute configuration of (+)-kelsoene.48 A second synthetic study aimed at determining the absolute configuration of (+)-kelsoene was carried out by Schulz and coworkers.48 Their approach relied on the chemical degradation of naturally occurring (+)-kelsoene and on the enantiospecific synthesis of (-)-kelsoene from {R)-(+)-pu\egor\e. It was reasoned that naturally occurring (+)-kelsoene could be degraded to a tricyclic intermediate which could be constructed in enantiomerically pure form from (f?)-(+)-pulegone. Chiral GC analysis of these materials would then allow the absolute configuration of (+)-kelsoene to be determined. Scheme 2.9: Chemical degradation of (+)-kelsoene The chemical degradation of (+)-kelsoene was carried out in a straightforward manner (Scheme 2.9). Thus, treatment of the natural product with p-toluenesulfonic acid isomerized the olefin to its more stable isomer (73). The double bond was then cleaved with ozone to provide the expected tricyclic ketone (64). 2. Total synthesis of (±)-kelsoene 23 At this stage it was necessary to construct tricyclic ketone 64 (or its enantiomer) in enantiomerically rich form. Thus (R)-(+)-pulegone (74) was transformed into a mixture of cis- and trans-pulegonic acids using a bromination reaction followed by a Favorskii rearrangement (Scheme 2.10). These diastereomers were chromato-graphically separated and the c/s-isomer (75) was converted to the bicyclic lactone 76 under acidic conditions. This material was subjected to a base mediated ring opening reaction, which served to establish the isopropenyl moiety. Carboxylic acid 77 was transformed to the corresponding acid chloride, and this material was converted into the necessary bicyclic ketone (enf-61) in the presence of a strong Lewis acid. The four-membered ring was installed in a manner similar to that used by Mehta and Srinivas. Thus irradiation of a mixture of ethylene gas and bicyclic enone enf-61 provided the requisite tricyclic ketone (enf-64). Chiral GC analysis of compounds 64 and enf-64 revealed that these two ketones exhibited different retention times and, therefore, have an enantiomeric relationship. Thus, being confident of the absolute configuration of the tricyclic ketone derived from (R)-(+)-pulegone, it could be deduced that the absolute configuration of naturally occurring (+)-kelsoene is that depicted in 35. Further support for this assignment was provided by an X-ray analysis on p-toluenesulfonate 78. This material was prepared by hydride reduction of ketone enf-64, followed by a tosylation reaction (Scheme 2.10). 2. Total synthesis of (±)-kelsoene 24 i) Br 2, Et 2 0, 0 °C ii) 25% aq. KOH, A 74 75 ethylene (g), hu, DCM, 0 °C i) LiAIH4, Et 2 0 ii) p-TsCI, pyridine aq. HCI, MeOH, A i) f-BuOK, DMF 140 °C ii) H® i) (COCI)2, DCM ii) AICI3, DCM C 0 2 H Scheme 2.10: Construction of enf-64 from (R)-(+)-pulegone 48 Finally, (-)-kelsoene (enf-35) was prepared from bicyclic ketone enf-61. The latter substance was elaborated to tricyclic alkenone 65 using the protocol devised by Mehta and Srinivas45,46 (Scheme 2.11). Alkenone 65 was treated with the ylide derived from trimethylsulfoxonium iodide to generate epoxide 79. Hydrogenation of the cyclobutene moiety in 79 was accompanied by the rearrangement of the epoxide to the corresponding aldehyde to furnish compound 80. This aldehyde was alkylated with methyllithium, and the resulting mixture of secondary alcohols was oxidized to the corresponding ketone (81). Finally, this ketone was equilibrated to its more thermodynamically stable epimer and subsequently converted to (-^kelsoene (enf-35) using the methylenation method described by Petasis.54 An optical rotation 2. Total synthesis of (±)-kelsoene 25 measurement verified that this material is indeed enantiomeric to naturally occurring (+)-kelsoene (35). as reported by Mehta et al. i) NaOMe, MeOH ii) Cp 2TiMe 2 , THF, A (-)-kelsoene (enf-35) Me 3SOI, base DMSO H2(g), Pd/C, E t 2 0 i) MeLi, THF, -78 °C ii) PDC, Et 2 0 CHO 80 Scheme 2.11: Enantiospecific total synthesis of (-)-kelsoene from (R)-(+)-pulegone48 2.4.4 Total Synthesis of (±)-kelsoene by Zhang and Koreeda 50 This synthesis relies on two key transformations to construct the tricyclic framework of kelsoene.50 It is known that y-keto-p-toluenesulfonates undergo a stereospecific homoallyl rearrangement, commonly referred to as the homo-Favorskii rearrangement. For example, treatment of 2-methyl-2-tosyloxymethylcyclohexane (82) with hot NaOH provides bicyclo[3.2.0]heptanone 83 through the homo-Favorskii rearrangement (Scheme 2.12). The isomeric bicyclo[3.1.1]heptanone 84 is obtained as the major side product in this reaction. The occurrence of these two isomers indicates that the reaction likely proceeds through the intermediacy of the non-classical zwitterion 85. 2 . Total synthesis of (±)-kelsoene 26 OTs NaOH (1.33 moi equiv.) dioxane / water 90 °C, 3 h 82 + 83 56:44 84 Scheme 2.12: Homo-Favorskii rearrangement of y-keto-p-toluenesulfonate 82 Interestingly, bicyclo[3.1.1]heptanones are readily rearranged to the isomeric bicyclo[3.2.0]heptanones by treatment with mild acid (Scheme 2.13). It is thought that this rearrangement is driven by the relatively higher strain energy of bicyclo[3.1.1]heptanones.187,188 Thus, the two constitutional isomers arising from the homo-Favorskii rearrangement of y-keto-p-toluenesulfonates converge to the more stable bicyclo[3.2.0]heptanone products upon treatment with mild acid. 84 86 calculated strain calculated strain energy = 35.85 kcal/mol energy = 30.48 kcal/mol Scheme 2.13: Acid catalyzed isomerization of bicyclo[3.1.1]heptanone to bicyclo[3.2.0]heptanone50 The synthesis begins with the conversion of the enol ether moiety of 2,5-dihydroanisole (87) to the corresponding ethylene ketal (Scheme 2.14). The remaining double bond was then oxidized with MCPBA to provide a mixture of diastereomeric epoxides (88). The oxirane was opened with sodium p-chlorophenylselenide and the resulting adduct (89) was oxidized to the corresponding selenoxide. Thermolysis of this material afforded the desired allylic alcohol (90). The cyanocuprate derived from copper(l) cyanide and Grignard reagent 91 reacted smoothly with the pivaloate derived from compound 90, and desilylation of the resulting adduct furnished enyne 92. A palladium catalyzed enyne cyclization afforded the desired bicyclic carbocycle 93. 2. Total synthesis of (±)-kelsoene 27 OCH, 87 i) (HOCH 2) 2, p-TSA (cat.), DCM * ii) MCPBA (1.5 moi equiv.), DCM, r.t., 10 h 99% di(p-chlorophenyl)diselenide NaH.THF, r.t., 10 h 94% HO"" 88 Pd(OAc)2 (5 mol%), AsPh 3 (15mol%) AcOH, PhH, 60 °C H " " 18 min 81% 92 SeAr 89 i) 30% H 2 0 2 , EtOH, r.t., 0.5 h ii) NaHC0 3 > EtOH, A, 12 h 92% i) pivaloyl chloride, DMAP, DCM, r.t., 4 h ii) B r M g — v 91 CuCN.THF, rt, 12 h iii) K 2 C 0 3 , MeOH, r.t., 4 h -TMS 0 ^ HO 90 94% Scheme 2.14: Preparation of enyne 93 The reduction of the exocyclic double bond in 93 was accomplished through the use of CoCI2/LiBH4 with a high degree of diastereoselectivity (Scheme 2.15). At this stage, the ketal moiety was converted to the corresponding ketone to furnish enone 94. Methylation of compound 94, followed by hydroxymethylation of the resulting material provided enone 96. As expected, both alkylation steps occurred from the convex face of the bicyclic system. The hydroxyl group of enone 96 was then converted to the corresponding tosylate in high yield. Conjugate addition of an isopropenyl group to the convex face of enone 97 generated the y-keto-p-toluenesulfonate (98) required for the crucial homo-Favorskii rearrangement. 2. Total synthesis of (±)-kelsoene 28 i) CoCI2»2H20, LiBH 4 2:1 THF/MeOH -78 °C, 2 h ''''H ii) TFA, hexanes r.t., 14 h 75% BrMg. 98 Cul, LiCI, TMSCI, THF, 0°C, 12 h 68% i) LDA, THF, -78 °C ii) Mel, HMPA, -78 °C to -20 °C 73% i) LDA, THF, -78 °C ii) C H 2 0 (g) -78 °C to 0 °C 85% H O — TsCI, DMAP, H -< • — H ' " > DCM, 0 °C 93% 97 96 Scheme 2.15: Preparation of y-keto-p-toluenesulfonate 98 50 Treatment of keto-p-toluenesulfonate 98 with an excess of f-BuOK provided, as expected, a mixture of cyclobutanone products 99 and 100 (Scheme 2.16). Although these materials could be easily separated, the mixture was treated with mild acid in order to isomerize cyclobutanone 99 to its more stable constitutional isomer (101). Cyclobutanones 100 and 101 were converted to the corresponding tosylhydrazones, and these were subsequently reduced to furnish (±)-kelsoene in 76% yield over the two 2. Total synthesis of (±)-kelsoene 29 TsO f-BuOK, f-BuOH r.t., 2 min 95% 98 99 + 5 : 4 100 (±)-kelsoene i) TsNHNH 2 , PhH, 60 °C, 12 h ii) NaBH 3CN, p-TSA sulfolane/DMF/ hexanes (1:1:2) 110 °C, 6 h 76% p-TSA, C F 3 C H 2 O H 0 °C, 4 h 90% 101 + 1 :1 100 Scheme 2.16: Completion of the synthesis of (±)-kelsoene by Zhang and Koreeda 50 2.4.5 Total Synthesis of (±)-kelsoene by Bach and Spiegel The most recently reported synthesis of (±)-kelsoene differs significantly in approach from those reported earlier, including the work described in this thesis. The synthesis begins with the preparation of the 1,6-diene 103 through a Knoevenagel condensation of dimethyl malonate with aldehyde 102 (Scheme 2.17). The highly diastereoselective intramolecular ene reaction of 103 leading to the cyclopentane 104 is thought to occur through a chair-like transition state as depicted in 107.55 Reduction of diester 104, followed by a porcine pancreatic lipase catalyzed acetylation of the resulting diol provides compound 105 as a mixture of diastereomers. The free alcohol is oxidized to the aldehyde and subsequently converted to the corresponding olefin (106) using a Wittig protocol. 2. Total synthesis of (±)-kelsoene 30 102 Knoevenagel Condensation diastereoselective C 0 2 C H 3 ene reaction C 0 2 C H 3 103 106 OAc i) (COCI)2, DMSO, Et 3N, t DCM, -78 °C, 2 h ii) Ph 3P=CH 2, THF -78 °C to r.t., 12 h 50% C 0 2 C H 3 104 C 0 2 C H 3 i) LiAIH4, Et 2 0 A, 2 h ii) vinyl acetate, PPL, r.t., 11 h 75% 105 OAc Scheme 2.17: Synthesis of 1,6-diene 106 The diastereomeric mixture of substances 106 was subjected to photochemical [2+2]-cycloaddition conditions56 and furnished the tricarbocyclic compounds 108 (Scheme 2.18). As expected, the rings embedded in this key intermediate exist in a trans-anti-cis arrangement. Methanolysis of ester 108 to alcohol 109 was followed by Swern oxidation to the corresponding aldehyde. This material provided a-bromoaldehyde 110 as a single diastereomer upon treatment with py»HBr 3 5 7 but the relative configuration of the carbon chirality center associated with the aldehyde was not ascertained. Addition of MeLi to a-bromoaldehyde 110 and oxidation of the resulting secondary alcohol furnished a-bromoketone 111. Elimination of hydrogen bromide from 111 provided the expected conjugated enone 112. As anticipated, hydrogenation of its double bond provided the requisite cis-anti-cis carbocycle with the incorrect relative configuration at the carbon chirality center associated with the acetyl group. Epimerization of this intermediate under acidic conditions to ketone rac-70 and olefination using a Wittig protocol resulted in the synthesis of (±)-kelsoene. 2. Total synthesis of (±)-kelsoene 31 H H 106 v OAc hu (254 nm), CuOTf, Et 2 0, r.t., 15 h 89% LiBr, L i 2 C0 3 , DMF, 140°C, 1 h 77% 112 i) H 2 (g), Pd/C, EtOH, r.t., 12 h ii) aq. HCI0 4 (35%), CHCI 3, A, 20 h. 65% as reported by 'H Mehta et al. rac-70 111 K 2 CQ 3 , MeOH, r.t., 3 h 90% i) (COCI)2, DMSO, Et 3N, DCM, -78 °C, 2 h ii) pyHBr 3 , CHCI 3, r.t., 12 h 81% i) MeLi, Et 2 0, -78 °C, 1.5 h H ii) PCC, NaOAc, DCM, 0 0°Ctor . t . ,4h 73% (±)-kelsoene 110 Scheme 2.18: Completion of the synthesis of (±) -kelsoene by Bach and Spiegel 49 2. Total synthesis of (±)-kelsoene 2.5 Total synthesis of (±)-kelsoene 2.5.1 Retrosynthetic analysis 32 Notwithstanding its low molecular weight, kelsoene represents a significant synthetic challenge. In addition to its unusual tricarbocyclic skeleton, kelsoene possesses six contiguous carbon chirality centers, one of which is quaternary. The retrosynthetic analysis of kelsoene began with the recognition that a tricarbocyclic ketone such as rac-64, in which the correct relative configuration of five carbon chirality centers is set, could be elaborated to the natural product (Scheme 2.19). This process would entail homologation of the ketone to the corresponding aldehyde, followed by an alkylation to generate the requisite secondary alcohol (113). Finally, oxidation of the alcohol to the corresponding ketone, followed by a methylenation reaction would furnish racemic kelsoene. It was expected that the four membered ring in ketone rac-64 could be constructed using a [2+2]-photocycloaddition of a suitable enone such as rac-61 with ethylene, using the protocol devised by Caldwell and coworkers.58 It was expected that the requisite cis-anti-cis relationship in ketone rac-64 would be established through the preferential approach of ethylene from the convex surface of enone rac-61 during the photocycloaddition event.59 Enone rac-61 could, in principle, be obtained from ketone rac-59 by oxidation to the corresponding enone, followed by 1,2 addition of methyllithium and oxidative rearrangement of the resulting tertiary allylic alcohol. The relative configuration of the methyl group in ketone rac-59 was expected to be established through a preferential hydrogenation of an exocyclic olefin from the convex face of bicyclic ketone rac-34. 2. Total synthesis of (±)-kelsoene 33 rac-34 rac-59 rac-61 Scheme 2.19: Retrosynthetic analysis of (±)-kelsoene 2.5.2 Total synthesis of ( ± ) - k e l s o e n e The planned total synthesis of (±)-kelsoene required an efficient synthesis of the diquinane core structure rac-34. A review of the literature revealed that in spite of the large body of work related to cyclopentane annulation methods60,61 and polyquinane natural product synthesis,62 only a couple of methods exist which establish the desired relationship between the ketone and the alkene moieties present in rac-34. The preparation of alkenone rac-34 reported by Franck-Neumann and Kastler.63 (Scheme 2.20) relies on the trimethylamine oxide mediated decomplexation of a tricarbonyl [trimethylenemethane] iron complex bearing an enone (115). Unfortunately, this reaction and the synthesis of the necessary starting material both proceed in low yield. 2. Total synthesis of (±)-kelsoene 34 Scheme 2.20: Franck-Neumann's synthesis of alkenone rac-34 It was decided to synthesize ketone rac-34 using a procedure modified from that reported earlier by this research group.32 The bifunctional reagent 4-chloro-2-trimethylstannylbut-1-ene (32) was prepared from 3-butyn-1-ol (116) in two steps using the method devised by Piers and Chong31 (Scheme 2.21). Sequential treatment of this material with MeLi and CuCN-LiCI in THF afforded the "lower order" heterocuprate64 lithium cyano(4-chlorobut-1-en-2-yl)cuprate (117). Reaction of this reagent with commercially available cyclopent-2-en-1-one in the presence of BF3«Et20, followed by a suitable workup procedure, provided the conjugate addition product 118 in good yield. Treatment of 118 with KH in THF caused an intramolecular alkylation reaction that provided the desired alkenone. C l ' "CuLiCN 117 i) MeLi, THF, -78 °C ii) CuCN-LiCI, THF, -78 °C i) BF3« Et 2 0 ii) cyclopent-2-en-l-one, -78 °C, 1 h iii) sat. aq NH4CI 69% (from 32) KH, THF, r.t. 75% 118 H O rac-34 Me 3 Sn ' 2 steps VCI 32 OH 116 Scheme 2.21: Synthesis of alkenone rac-34 2. Total synthesis of (±)-kelsoene 35 With an efficient synthesis of alkenone rac-34 accomplished, the stage was set to establish the correct configuration at C-8 (kelsoene numbering) through a stereoselective hydrogenation of the exocyclic alkene. It appeared likely that this hydrogenation would, for steric reasons, occur predominantly from the convex face of the bicycle. However, treatment of rac-34 with hydrogen in the presence of palladium-on-carbon provided ketones rac-59 and 119 (see Scheme 2.22) in nearly equal amounts. Fortunately, the use of homogeneous hydrogenation catalysts gave better results. Thus, hydrogenation of rac-34 using Crabtree's catalyst ( [lr(cod)py(Pcy3)]PF6 )65 furnished ketones rac-59 and 119 in a ratio of 6:1, but the hydrogenation proceeded at a very slow rate. On the other hand, reduction of rac-34 with hydrogen in the presence of Wilkinson's catalyst66 afforded a 95% yield of a mixture of rac-59 and 119 in a ratio of 95:5 as determined by 1H-NMR spectroscopy (Scheme 2.22). At this stage no experiments were carried out to verify the relative configuration of the major product. These diastereomers proved difficult to separate and, therefore, were carried through the next synthetic operation as a mixture. Scheme 2.22: Diastereoselective hydrogenation of alkenone rac-34 The spectral data collected on ketone rac-59 was in full accord with the structural assignment of the compound. The IR spectrum of rac-59 displayed a strong C=0 stretching absorption (1736 cm"1) characteristic of a saturated five-membered ring ketone. In addition, the 1 3C NMR spectrum recorded for rac-59 included a carbonyl resonance at 5= 223.1. Furthermore, the 1H NMR spectrum displayed a 3-proton doublet at 8 = 0.97, with a coupling constant of 6.9 Hz, which could be attributed to the methyl group at C-8 (kelsoene numbering). Finally, an HRMS measurement on the parent ion revealed that the material obtained from the hydrogenation had a molecular formula consistent with the structure of ketone rac-59. 2. Total synthesis of (±)-kelsoene 36 Ketone rac-59 was converted into the desired photocycloaddition precursor using a three-step sequence (Scheme 2.23). Oxidation of the 95:5 mixture of ketones rac-59 and 119 to the corresponding enones was accomplished using the method established by Reich ef a/. 6 7 , 6 8 Thus the mixture of ketones was treated with an excess of LDA in order to form the corresponding kinetic enolates. These enolates were quenched with phenylselenenyl chloride to generate the requisite a-phenylselenides (120). Oxidation of the phenylselenides with hydrogen peroxide and chromatographic purification of the acquired material provided enone rac-60 in 83% yield. No attempt to identify or isolate the enone derived from the minor ketone (119) was made. The spectroscopic data collected for enone rac-60 was in full accord with the assigned structure. The IR spectrum of rac-60 displayed a strong C=0 stretching absorption at 1707 cm"1 that can be attributed to a conjugated carbonyl group. The 1 3C NMR revealed an enone carbonyl resonance at 8 = 213.7 and two olefinic carbon resonances at 8 = 135.8 (a) and 165.3 (p). In addition, the 1H NMR spectrum included two 1-proton resonances at 8 = 6.17 (a) and 7.57 (P) with a mutual coupling constant of 5.8 Hz. Enone rac-60 was treated with an excess of methyllithium in THF at 0 °C to generate the expected tertiary allylic alcohol (121). This alcohol proved to be very prone to elimination. Treatment of the reaction mixture with 10% aq. HCI or exposure of the crude alcohol to silica gel resulted in the formation of conjugated diene 122. Therefore, the isolation of the crude tertiary allylic alcohol required a work-up procedure which included a quench of the reaction mixture with a pH = 8 aqueous buffer. 2. Total synthesis of (±)-kelsoene 37 PCC-on-alumina DCM, 0 °C rac-6~\ Scheme 2.23: Synthesis of enone rac-61 Finally, the crude tertiary allylic alcohol was treated with PCC-on-alumina to generate the required conjugated enone (rac-61, Scheme 2.23). This transformation occurs through the initial formation of a chromate ester (123) followed by a [3,3]-sigmatropic rearrangement and subsequent oxidation of the rearranged chromate ester (124, Scheme 2.24). This PCC-induced oxidative rearrangement of tertiary allylic alcohols has played a key role in the development of new cyclopentenone,73 cyclohexenone74 and cycloheptenone34 annulation methods in this laboratory. rac-61 124 Scheme 2.24: PCC induced oxidative rearrangement of tertiary allylic alcohol 121 Analysis of the spectral data collected on enone rac-61 verified the assigned structure. The IR spectrum of this material included the diagnostic absorption at 1698 cm"1 attributed to a conjugated carbonyl group. The 1 3C NMR spectrum contained a carbonyl resonance at 8 = 210.2 and two olefinic resonances at 5 = 132.4 (a) and 179.6 (p). More importantly, the 1H NMR spectrum included a 3-proton singlet at 8= 2.02 attributed to the alkenyl methyl group. In addition, a single 1-proton broad singlet (8 = 5.82-5.84) was observed in the olefinic region of the spectrum. The plan for the synthesis of (±)-kelsoene called for a [2+2]-photocycloaddition reaction to construct the four membered ring. This strategy was based on two literature reports in which the efficient synthesis of four membered rings through similar [2+2]-bhotocycloadditions had been demonstrated. Caldwell and coworkers58 showed that when a solution of enone 125 in DCM saturated with ethylene gas at -70 °C is irradiated with UV-light through a Pyrex filter, bicyclic ketone 126 is obtained in 90% yield (Scheme 2.25). 2. Total synthesis of (±)-kelsoene 39 Scheme 2.25: [2+2]-photocycloaddition of ethylene to 3-methylcyclopent-2-en-1-one (125)58 Similarly, Meyers59 has demonstrated the synthesis of the tricyclic lactam 128 using similar conditions. In this case two tricyclic products, 128 and 129, are observed in a 12:1 ratio, respectively (Scheme 2.26). Presumably, the approach of ethylene from the sterically less encumbered convex side of the a,p-unsaturated bicyclic lactam (127) provides the major product. In this case, the diastereocontrol imposed by the convex nature of the bicycle is less than would be expected with an enone such as rac-61, due to the presence of the angular methyl group which, to some extent, serves to block the approach of ethylene from the convex face. In addition, the sp2 character of the nitrogen atom at the ring fusion reduces the effect of the convex nature of the bicyclic system. Scheme 2.26: [2+2]-photocycloaddition of ethylene to a,p-unsaturated bicyclic lactam 1275 9 At the time that the retrosynthetic analysis of kelsoene was carried out, a relevant report regarding [2+2]-photocycloadditions was overlooked. As it turns out, the synthesis of a tricyclic system very similar to that required for the synthesis of (±)-kelsoene has been carried out by Ohfune et al.75 (Scheme 2.27). Thus, irradiation of 2. Total synthesis of (±)-kelsoene 40 bicyclic enone 130 in the presence of ethylene provided a mixture of the cis-anti-cis-photoproduct 131 and its cis-syn-cis isomer 132 in 75% and 8% respectively. That the c/s-sy/7-c/s-photoproduct was obtained in a significant amount indicates that the convex nature of the bicyclic enone is not sufficient to provide complete selectivity in the photocycloaddition. Scheme 2.27: [2+2]-photocycloaddition of ethylene to bicyclic enone 130 With a sufficient amount of enone rac-61 and significant literature precedent for the desired photocycloaddition, the stage was set to attempt the construction of the requisite tricyclic ketone (rac-64). Irradiation of a DCM solution of enone rac-61 saturated with ethylene gas at -78 °C with UV-light through a Pyrex filter provided the expected cis-anti-cis tricyclic ketone in 90% yield (Scheme 2.28). The absence of the isomeric cis-syn-cis tricycle, and the high yield of the reaction indicates that the secondary methyl substituent in enone rac-61 exerts some diastereocontrol beyond that provided by the convex nature of the bicyclic system. This reaction is thought to proceed through a stepwise mechanism (Scheme 2.28).7 6'1 8 9 The first step involves an n^>n* transition of enone rac-61 to its triplet excited state (133), which behaves as a biradical and reacts with ethylene to generate a biradical adduct. This species can be represented as resonance forms 134 and 135. A radical recombination closes the four membered ring to furnish the desired cis-anti-cis tricycle. 2. Total synthesis of (±)-kelsoene 41 H H hu (pyrex filter, A>290 nm) DCM, -78 °C ,0* rac-61 133 ethylene H H H H rac-64 135 134 Scheme 2.28: synthesis of tricyclic ketone rac-64 The spectral data collected on ketone rac-64 supported the assigned structure. The IR spectrum showed a strong absorption at 1728 cm"1, which is diagnostic of a saturated 5-membered ring ketone. The 1 3C NMR spectrum contained the correct number of carbon resonances, including a carbonyl resonance at 5 = 223.4. The olefinic carbon resonances previously observed in enone rac-61 were absent. Analysis of a separate APT experiment revealed the presence of 2 methyl, 4 methylene and 4 methine carbon resonances. Moreover, an HRMS measurement on the parent ion was consistent with the molecular formula corresponding to ketone rac-64. Thus, the tricyclic core of (±)-kelsoene, with the correct relative configuration at each of the five carbon chirality centers, had been assembled from commercially available cyclopent-2-en-1-one and the . known bifunctional reagent 4-chloro-2-trimethylstannylbut-1-ene (32) in eight synthetic operations. All that remained to complete the synthesis of kelsoene was the installation of the isopropenyl group with the correct relative configuration at C-6 (kelsoene numbering). As originally planned, the total synthesis of (±)-kelsoene would proceed with the homologation of tricyclic ketone rac-64 to the corresponding aldehyde. This transformation was expected to occur through a Wittig olefination of ketone rac-64 with 2. Total synthesis of (±)-kelsoene 42 (methoxymethylene)triphenylphosphorane to generate the corresponding exocyclic methyl enol ether. This substance would, in turn, be hydrolyzed under acidic conditions to furnish the homologated aldehyde. This sequence has been successfully employed in triquinane sesquiterpenoid synthesis (Scheme 2.29).77 MeO OMe ^ = P P h q 100% hydrolysis 64% Scheme 2.29: Homologation of ketone 136 to aldehyde 138 in a study towards the synthesis of hirsutene77 In the present work, the configuration of the aldehyde was expected to be that required for the synthesis of kelsoene due to a significant non-bonded interaction between the C-8 methyl group and the a-oriented aldehyde (rac-80). This interaction is absent in the required (5-oriented aldehyde (139, Figure 2.4) H O' H // \ CH 3 H . CH-, 139 rac-80 Figure 2.4: Steric interactions between the C-8 methyl group and the C-6 aldehyde in 139 and rac-80 Surprisingly, subjection of ketone rac-64 to the homologation conditions previously described (vide supra) did not provide the desired methyl enol ether, and resulted only in recovery of the starting material. Initially, it was thought that the bulky nature of the phosphorane reagent was the major cause for the lack of reaction. 2. Total synthesis of (±)-kelsoene 43 Therefore, alternative methods involving sterically less hindered reagents were sought. In this connection, Taguchi etal.78 have shown that ketones can be homologated to the corresponding a-chloroaldehydes under the action of dichloromethyllithium at low temperatures, followed by heating of the initial adduct (Scheme 2.30). It is believed that the initial 1,2-addition product undergoes a ring closing reaction to the chloro epoxide (141), releasing a chloride ion as a leaving group. Heating of chloroepoxide 141 promotes its rearrangement to a-chloroaldehyde 142. 140 - 2 5 °C, 30 min 141 142 Scheme 2.30: Taguchi's synthesis of an a-chloroaldehyde from cyclohexanone Unfortunately, the attempted application of this method in the synthesis of (±)-kelsoene was unsuccessful, as the reaction of dichloromethyllithium with ketone rac-64 provided an intractable mixture of products. It became apparent that, in order to accomplish the desired homologation, the original synthetic plan required some strategic modification. The revised approach relied on the formation of an exocyclic olefin (143) from ketone rac-64, likely through the use of a relatively small organometallic reagent. This exocyclic olefin could then be hydroborated and oxidized to provide a primary alcohol (144), and elaborated to the target aldehyde (Scheme 2.31). 2. Total synthesis of (±)-kelsoene 44 Scheme 2.31: Modified synthetic plan to aldehyde rac-80 It was expected that the olefination of ketone rac-64 could be carried out through the use of a Wittig protocol. Indeed, precedent for this transformation on a very similar substrate (131) exists in the literature (Scheme 2.32).75 It was, therefore, somewhat surprising that reaction of ketone rac-64 with methylenetriphenyl-phosphorane resulted only in the recovery of starting material. This result, and the successful olefination of ketone rac-65 carried out by Mehta and Srinivas (Scheme 2.7) suggest that the lack of reactivity observed with ketone rac-64 is largely due to the steric congestion imposed by presence of the C-8 methyl group and the C-4 methylene group (kelsoene numbering, vide supra). 2. Total synthesis of (±)-kelsoene 4 5 The difficulties associated with methylenation of sterically hindered ketones have been well documented. Titanium based methylenating reagents have surfaced as superior alternatives to the Wittig phosphoranes.79 A comparative study80 revealed that the Tebbe reagent (146)81 is comparable to or more effective than the Wittig reagent in most cases. However, highly hindered ketones such as fenchone (147) still fail to give satisfactory results (Scheme 2.33). Thus, it was not surprising that the attempted methylenation of tricyclic ketone rac-64 with Tebbe's reagent did not provide the desired olefin. Scheme 2.33: Olefination of fenchone with Tebbe's reagent Takai et al. developed a similar titanium based reagent which is more reactive than the Tebbe reagent and which is still not well characterized.82"84 An alternative method for the preparation of the active species has been developed by Lombardo,85,86 and both methods have proven successful in the methylenation of hindered and easily enolizable ketones. Gratifyingly, subjection of ketone rac-64 to the olefination conditions described by Lombardo provided the desired alkene in 63% yield after a reaction time of 16 hours87 (Scheme 2.34). Scheme 2.34: Lombardo olefination of ketone rac-64 2. Total synthesis of (±)-kelsoene 46 The spectral data collected on olefin 143 was in full support of the assigned structure. The 1 3C NMR spectrum showed the correct number of carbon resonances, including two alkene resonances at 8= 108.8 (exocyclic) and 159.9 (endocyclic). Analysis of a separate APT experiment revealed 2 methyl, 5 methylene and 4 methine resonances. Moreover, the newly introduced exocyclic methylene group was observed as a 2-proton multiplet at 8 = 4.90-4.95 of the 1H NMR spectrum. Having synthesized the requisite exocyclic olefin, the stage was set to attempt the hydroboration of this material. Based on the lack of reactivity observed with ketone rac-64, it seemed likely that bulky hydroboration reagents such as 9-BBN would fail at achieving the desired transformation. Therefore, the hydroboration of compound 143 was attempted using borane-dimethyl sulfide complex. It was expected that the hydroboration would proceed with a high degree of regioselectivity and facial selectivity. With respect to facial selectivity, analysis of molecular models indicates that the hydroborating reagent should approach the alkene moiety from the less sterically hindered [3-face to furnish alcohol 144, which possesses the incorrect relative configuration at the newly formed carbon chirality center (Figure 2.5). The hydroboration of alkene 143 proceeded with a high degree of regio- and facial-selectivity and in good yield (Scheme 2.35). At this stage it was deemed prudent to verify the structure of this substance, and to ascertain the relative configuration of its six carbon chirality centers. To this end, the IR, HRMS, 1H NMR, 1 3C NMR, HMQC, HMBC and COSY spectra of this substance were recorded. As expected, the IR spectrum of this material contained an absorption at 3325 cm"1 due to the primary H H H Figure 2.5: Approach of a hydroborating reagent to olefin 143 2. Total synthesis of (±)-kelsoene 47 alcohol moiety. The 1 3C NMR spectrum exhibited a resonance at 8 = 64.7 attributed to the carbinol carbon. The 1H NMR spectrum showed two doublets-of-doublets at 8 = 3.33 and 3.93, with a mutual coupling constant of 10.2 Hz, which can be attributed to the carbinol protons. Finally, an HRMS measurement on the parent ion was consistent with the molecular formula of the expected alcohol. i) BH3«DMS THF, 0 °C ii) H 2 0 2 , NaOH THF, H 2 0 , r.t. 80% Scheme 2.35: Synthesis of alcohol 144 Due to a fortuitous dispersion of signals, the 1H NMR spectrum of alcohol 144 could be fully assigned (see experimental section for details). A set of one dimensional NOE difference experiments verified that the configuration at the newly generated carbon chirality center was as expected (vide supra). As shown below, the secondary methyl group (MeA) in alcohol 144 gives rise to a doublet at 8 = 1.02, the carbinol protons H A and H B produce doublets of doublets at 8 = 3.33 and 8 = 3.93, and the signals due to the tertiary and secondary hydrogens He and H D appear as multiplets at 8 = 2.30 and 8 = 1.34 respectively (Figure 2.6). Me A |_| C H A H B O H 144 Me A : 5 = 1.02 (d); signal enhanced by irradiation at 5 = 3.33 or 3.93 H A , H B : 5 = 3.33 (dd), 3.93 (dd) H c : 5 = 2.30 (m); signal enhanced by irradiation at 8 = 3.33 or 3.93 H D : 5 = 1.34 (m); signal enhanced by irradiation at 8 = 3.33 Figure 2.6: NOE correlations in alcohol 144 2. Total synthesis of (±)-kelsoene 48 In nuclear Overhauser enhancement difference (NOED) experiments, irradiation at either 8 = 3.33 or 8 = 3.93 (HA, HB) enhances the signal intensities at both 8 = 1.02 (MeA) and 8 = 2.30 (He). Also, irradiation at 8 = 3.33 (HA or HB) causes an increase in the intensity of the resonance at 8 = 1.34 (HD). These experiments show clearly that the CH2OH function is on the a-face of the molecule as drawn above (144, figure 2.6). The next step in the sequence involved the oxidation of alcohol 144 to the corresponding aldehyde. This was accomplished smoothly and in high yield using the method developed by Ley et al.88 (Scheme 2.36). The aldehyde obtained (rac-80) proved quite unstable and was therefore used shortly after its preparation. However, the IR spectrum of rac-80 showed an absorption at 1716 cm"1, indicative of the aldehyde moiety. Furthermore, an HRMS measurement of the parent ion was consistent with the molecular formula of the expected aldehyde. Addition of methyllithium to aldehyde rac-80, and oxidation88 of the resultant diastereomeric mixture of secondary alcohols provided ketone rac-81 in 86% yield from alcohol 144. DCM, r.t. TPAP H I JH H : H : 144 rac-80 i) MeLi, E t 2 0, 0 °C ii) HOAc, H 2 0 H : H : rac-61 149 86% Scheme 2.36: Synthesis of ketone rac-81 2. Total synthesis of (±)-kelsoene 4 9 The structure of ketone rac-81 was verified with the usual spectroscopic techniques. Thus, the IR spectrum showed the expected absorption for the ketone carbonyl stretch at 1707 cm"1. The 1 3C NMR spectrum contained the expected number of carbon resonances, including a carbonyl resonance at 8= 210.0. Finally, the 1H NMR spectrum included a three-proton singlet at 8= 2.19 attributed to the newly introduced methyl group. At this stage it was necessary to invert the configuration of the chirality center associated with the acetyl group. This operation was accomplished by refluxing a biphasic mixture consisting of a deuteriochloroform solution of ketone rac-81 and a solution of perchloric acid in water (Scheme 2.37). The required epimeric ketone rac-70 was obtained in 80% yield and showed a carbonyl stretch absorption at 1709 cm"1 in the IR spectrum. The expected carbonyl resonance appeared at 8= 208.7 in the 1 3C NMR spectrum. Furthermore, the methyl group associated with the acetyl moiety appeared as a three-proton singlet at 8 = 2.02. All that remained to complete the synthesis of kelsoene was the transformation of the ketone moiety in rac-70 into the corresponding olefin. This was accomplished in high yield through the use of Lombardo's reagent (Scheme 2.37). The 1 3C NMR, 1H NMR, IR and HRMS spectral data collected on the synthetic sample of (±)-kelsoene are in full accordance with that reported for (+)-kelsoene by Konig and Wright37 (see Tables 2.1,2.2, and 2.3). H rac-81 rac-70 (±)-kelsoene Scheme 2.37: Completion of the synthesis of (±)-kelsoene 2. Total synthesis of (±)-kelsoene 50 Table 2.1. Comparison of the HREIMS and IR data reported for naturally occurring (+)-kelsoene with the HRMS and IR data obtained for synthetic (+)-kelsoene Data naturally occurring (+)-kelsoene synthetic (±)-kelsoene 3085 3083 IR 1650 1647 (cm"1) 1450 1452 1375 1374 885 886 HREIMS (m/z) 204.1869 204.1870 Table 2.2. Comparison of the 1 3C NMR data reported for naturally occurring (+)-kelsoene with the 1 3C NMR data obtained for synthetic (±)-kelsoene 015 carbon number 5 (ppm) naturally occurring (+)-kelsoenea'b synthetic (±)-kelsoenec 1 57.8 57.8 2 45.7 45.8 3 33.0 33.1 4 14.6 14.6 5 47.4 47.5 6 48.1 48.1 7 49.9 50.0 8 36.3 36.3 9 33.2 33.3 10 26.0 26.0 11 145.6 145.6 12 109.8 109.8 13 24.1 24.1 14 23.5 23.5 15 17.7 17.7 Assignments are as reported by Konig and Wright.37 Recorded at 75 MHz in CDCI3. C Recorded at 100 MHz in CDCI3. 2. Total synthesis of (±)-kelsoene 51 Table 2.3. Comparison of the 1H NMR data reported for naturally occurring (+)-kelsoene and the 1H NMR data obtained for synthetic (±)-kelsoene proton number3'13 6 (ppm)(mult; J(Hz)) naturally occurring (+)-kelsoene0 synthetic (±)-kelsoened H-1 2.09 (ddd, 9.1, 8.9, 6.7) 2.06 (ddd, 9, 9, 6.8) H-3a 1.64 (m) part of the m at 1.61-1.69 H-3b (or H-4b 1.68 (m) see footnote e H-4a 1.52 (m) part of the m at 1.43-1.54 H-4b (or H-3b) 1.68 (m) see footnote e H-5 2.42 (m) part of the mat 2.31-2.41 H-6 2.37 (m) part of them at 2.31-2.41 H-7 2.87 (ddd, 10.5, 6.8, 6.7) 2.84 (ddd, 10.4, 7, 7) H-8 2.27 (m) 2.24 (m) H-9a 1.34 (m) part of the m at 1.24-1.40 H-9b 1.76 (m) part of the m at 1.69-1.81 H-10a 1.35 (m) part of the m at 1.24-1.40 H-1 Ob 1.47 (m) part of the m at 1.43-1.54 H-12a 4.80 (br s) 4.77 (s) H-12b 4.87 (br s) 4.84 (s) H-13 1.61 (brs) 1.59 (s) H-14 1.17 (s) 1.14 (s) H-15 0.91 (d, 6.8) 0.89 (d, 7.0) Assignments are as reported by Konig and Wright.37 Methylene protons are designated H-Xa and H-Xb arbitrarily. Recorded at 300 MHz in CDCI3. Recorded at 500 MHz in CDCI3. H-3b and H-4b account for a proton at 8 = 1.61 -1.69 and a proton at S = 1.69-1.81, but could not be unambiguously assigned. 2. Total synthesis of (±)-kelsoene 52 2.5.3 Summary The total synthesis of the unique natural product (±)-kelsoene (rac-35) was accomplished89 through a 15-step sequence from commercially available cyclopent-2-en-1-one and the bifunctional reagent 4-chloro-2-trimethylstannylbut-1ene31 (32, Scheme 2.38). The synthetic approach was guided by the realization that the diquinane core of the natural product (rac-34) could be assembled quickly and efficiently through an annulation method previously reported by this group.32 Enone rac-61 was prepared form ketone rac-34 in a straightforward manner. A highly diastereoselective [2+2]-photocycloaddition of ethylene to enone rac-61 served to install the cyclobutane ring of (±)4<elsoene. Tricyclic ketone rac-64 could not be elaborated to the natural product using the original synthetic plan due to the severe steric congestion around its carbonyl group. This problem was circumvented through the use of Lombardo's reagent, which is reactive towards highly hindered and easily enolizable ketones, to generate the corresponding olefin (143). The natural product was obtained from olefin 143 through an efficient 6-step reaction sequence. [2+2]-photocycloaddition to ethylene (±)-kelsoene 143 rac-64 Scheme 2.38: Overview of the total synthesis of (±)-kelsoene 53 3. Synthetic Approaches to (±)-lsomulinic Acid 3.1 Introduction The mulinane family of natural products comprises twelve diterpenoid substances, all of which possess the 5-isopropyl-6,9,12-trimethyltricyclo[7.5.0.02,6]-tetradecane framework depicted in structure 150 (Figure 3.1). Note that the carbon atoms of the parent tricycle are numbered according to the IUPAC nomenclature rules90'91 for bridged hydrocarbons in 150. The community of natural products chemists has adopted the numbering system depicted in 151 based on biosynthetic considerations.33 For the sake of clarity, the numbering system depicted in 151 has also been adopted for sections 3.2 and 3.3 and 3.5 of this chapter. 150 151 (-)-isomulinic acid (39) Figure 3.1: Different numbering systems for the mulinane carbocyclic skeleton (150 and 151) and structure of (-)-isomulinic acid (39) (-)-lsomulinic acid3 3 (39, Figure 3.1) is arguably the most structurally complex member of the mulinane family of diterpenoids. Notable structural features embedded in this natural product include nine contiguous carbon chirality centers (two of which are quaternary), a tertiary carboxylic acid moiety, two epoxide rings, and a frans-fused hydrindane system. Clearly, this target compound presents a significant synthetic challenge. 3. Synthetic Approaches to (±)-lsomulinic Acid 54 3.2 Isolation, Structure Determination and Biological Activity Mulinic acid (152) and isomulinic acid (39, Figure 3.2), the first reported members of the mulinane family of natural products, were isolated from the petrol extracts of Mulinum crassifolium Phil. (Umbelliferae) by Loyola et al.33 This shrub is found in the north of Chile and its bitter infusions are used in folk medicine to treat diabetes and bronchial and intestinal disorders. The structure of mulinic acid (152) was elucidated using a combination of NMR spectroscopic and single-crystal X-ray diffraction analyses. The structure of isomulinic acid (39) was deduced using NMR spectroscopic analysis and by comparison of its spectral data with that obtained for mulinic acid (152). In addition, it was shown that isomulinic acid (39) can be obtained by heating a sample of mulinic acid (152) to 195 °C for 5 minutes. This thermal rearrangement proceeds with retention of configuration at C-11 and C-14. Figure 3.2: Structures of isomulinic acid (39) and mulinic acid (152) Since the first report by Loyola et al., ten other mulinane natural products have been isolated. 17-Acetoxymulinic acid (153),92 mulinenic acid (154),93 and mulinolic acid (155)94 were also isolated from Mulinum crassifolium Phil. (Umbelliferae) (Figure 3.3). The structures of 17-acetoxymulinic acid and mulinolic acid were elucidated by analysis of their NMR spectral data. No determination of their absolute configuration was made, although it is postulated to be as depicted in 153 and 155 based on biogenetic considerations. The structure of mulinenic acid was determined to be as shown in structure 154 by analysis of NMR and X-ray crystallographic data. 152 39 G""0 153 154 155 Figure 3.3: Structures of 17-acetoxymulinic acid (153), mulinenic acid (154) and mulinolic acid (155) Nicoletti et al. disclosed the structures of two new mulinane diterpenoids from the extracts of Mulinum spinosum (Cav.) Pers.,95 a shrub found in the Patagonia steppe of Chile and Argentina. Known as neneo, this plant is used in folk medicine to treat dental neuralgias, hepatic diseases and altitude sickness. The structures of compounds 156 and 157 were determined by analysis of their NMR spectral data (Figure 3.4). 156 157 Figure 3.4: Structures of compounds 156 and 157 In 1997 Loyola era/, showed that compounds 155 (Figure 3.3) and 157 (Figure 3.4) also occur in Azorella compacta Phil. (Umbelliferae),96 a resinous shrub which grows in the high Andes mountains of southern Peru, Bolivia, northwestern Argentina and northeastern Chile. Bitter tasting infusions of this plant are used in folk medicine for the treatment of diabetes, asthma, colds, bronchitis, and kidney and "womb 3. Synthetic Approaches to (±)-lsomulinic Acid 56 complaints." In this report, compound 157 is referred to as mulin-11,13-dien-20-oic acid. Further investigation of the extracts of Azorella compacta Phil. (Umbelliferae) resulted in the isolation of two new mulinane natural products. The structures of mulinol (158) and 11,12-epoxymulin-13-en-20-oic acid (159)97 were determined by analysis of their NMR spectral data (Figure 3.5). Figure 3.5: Structures of mulinol (158) and 11,12-epoxymuli-13-en-20-oic acid (159) In 1999, Timmermann and coworkers isolated two new mulinane diterpenoid natural products which, unlike all previously described mulinanes, displayed antibacterial activity. Mulin-12,14-dien-11-on-20-oic acid (160) and mulin-12-ene-11,14-dion-20-oic acid (161) were isolated from Azorella compacta Phil. (Apiaceae) (Figure 3.6). Finally, 13-epimulinolic acid (162) was isolated from the extracts of Laretia acaulis (Cav)99 (Figure 3.6). 3. Synthetic Approaches to (±)- lsomulinic Acid 57 Figure 3.6: Structures of mulin-12,14-dien-11-on-20-oic acid (160), mulin-12-ene-11,14-dion-20-oic acid (161) and 13-epimulinolic acid (162) In 1998 Loyola et al. disclosed the structure of azorellanol (163, Figure 3.7), a diterpenoid substance with a new carbon skeleton isolated from Azorella compacta Phil. (Umbelliferae).100 Recall that petrol extracts of this plant yielded a number of mulinane natural products. The same year Timmermann and coworkers disclosed the structure of a related natural product from Azorella madreporica Clos1 0 1 (164, Figure 3.7). Compound 164 decomposes in CDCI3 at room temperature. Interestingly, the major component in the mixture of decomposition products of compound 164 was shown to possess the mulinane carbocyclic skeleton (165, Figure 3.7). azorellanol (163) 164 165 Figure 3.7: Structures of azorellanol (163) and compounds 164 and 165 3. Synthetic Approaches to (±)-lsomulinic Acid 58_ 3.3 Biogenesis No biosynthetic studies have been carried out on the mulinane family of natural products. A possible biosynthetic pathway to the mulinane skeleton (172) from a labdane derivative (167) has been proposed by Loyola etal.33 (Scheme 3.1). The suggested biogenetic route to the mulinane family of natural products begins with alcohol 167, which presumably is derived from labdadienyl pyrophosphate102 (166) in a number of steps. The ring contraction that leads from alcohol 167 to compound 168 involves simultaneous cleavage of the bond between C-4 and C-5, bond formation between C-5 and C-3 and displacement of the hydroxyl group on C-3. This process leaves a positive charge localized at C-4 as shown in 168. Loss of the proton from C-3 results in the formation of a double bond between C-3 and C-4 (see 169). Migration of the methyl group from C-10 to C-5 occurs as shown in structure 169. This sequence of steps establishes the relative configuration at C-3 and C-5, and leaves a positive charge localized at C-10 (see 170). Two consecutive hydride shifts, as shown in 170, establish the relative configuration at C-9 and C-10 and place the positive charge at C-8 (170 to 171). Carbon-carbon bond formation between C-8 and C-15, double bond migration and loss of a proton from C-11 as shown in 171 result in the formation of the mulinane tricyclic system (172). A number of oxidative reactions at C-17, C-20 and on the 7-membered ring of 172 give rise to the mulinane family of natural products. It must be noted that Loyola et al. presented the biosynthetic pathway described above in their first report33 of the mulinane diterpenoids. At that time, only two mulinanes (39 and 152) were known. In addition, the carbocyclic framework of the azorellane family of natural products (see 163 and 164 in Figure 3.7 for example) was not reported until 1998. The structural similarity between the azorellane and mulinane families of natural products and their occurrence in the same organism suggests that they share a common biosynthetic pathway to an advanced intermediate (such as 171) in which the relative configuration at C-3, C-5 and C-10 has been established. 3. Synthetic Approaches to (±)-lsomuiirtic Acid 59 Scheme 3.1: Proposed biosynthesis of the mulinane diterpenoids from a labdane precursor33 3. Synthetic Approaches to (±)-lsomulinic Acid 60 3.4 Synthetic Approaches to (±)-lsomulinic Acid 3.4.1 Introduction The structural complexity of (-)-isomulinic acid (39) renders it a significant synthetic challenge, and this led us to attempt its total synthesis in racemic form. Our efforts in this regard evolved through four distinct strategies, each drawing upon the lessons learned from the previous failed attempt. The final strategy provided access to a tricyclic system reminiscent of the mulinane skeleton, with the correct relative configuration at the five carbon chirality centers present in all the members of the mulinane family of diterpenoids. Unfortunately, due to time and material constraints, completion of the total synthesis of (39) was not achieved despite our best efforts. Although none of the synthetic plans presented in this chapter provided access to (±)-isomulinic acid, the retrosynthetic analysis leading to each of the four strategies is presented in order to provide context to the work that was carried out. The reader should note that, for the sake of clarity, the numbering system adopted by the community of natural products chemists (see 151 in Figure 3.1) is used in the discussion dealing with the retrosynthetic analyses of isomulinic acid. All synthetic intermediates presented are numbered according to the IUPAC protocol for bridged carbocycles,90,91 and any discussion related to them makes use of this system unless specified otherwise. 3. Synthetic Approaches to (±)-lsomulinic Acid 61 3.4.2 First approach to ( ± ) - i s o m u l i n i c acid 3.4.2.1 Retrosynthetic analysis The retrosynthetic analysis of isomulinic acid (39) began with the realization that the two epoxides on its seven-membered ring can be obtained through the rearrangement of the endoperoxide moiety of mulinic acid (152, Scheme 3.2). Indeed, Loyola et al.33 showed that this process occurs thermally. However, the yield for this transformation under these conditions was low. The same type of transformation is known to occur in high yield under the influence of transition metal catalysts.103,104 Mulinic acid (152) can, in turn, be obtained through a hetero-Diels-Alder reaction of mulin-11,13-dien-20-oic acid (157) with singlet oxygen. Analysis of molecular models indicates that the methyl group on C-8 blocks the p-face of the diene on the seven-membered ring. Therefore, the hetero-Diels-Alder reaction of 157 with singlet oxygen should occur preferentially from the a-face of the molecule thereby establishing the correct relative configuration at C-11 and C-14. Scheme 3.2: Retrosynthesis of isomulinic acid to mulin-11,13-dien-20-oic acid It is clear that part of the mulinane carbocyclic framework could potentially be prepared through the use of the seven-membered ring annulation procedure recently developed in this laboratory. Piers et al.34 have shown that compounds such as bicycle 38 can be prepared from the bifunctional reagent c/'s-5-iodo-1-tri-n-butylstannylpent-1-ene (36) and a suitable keto ester, such as 173, in four steps. Indeed, the bicyclic structure of 38 maps well onto mulin-11,13-dien-20-oic acid (157, Scheme 3.3). Alkylation of keto ester 173 with bifunctional reagent 36 yields compound 174. The 3. Synthetic Approaches to (±)-lsomulinic Acid 62 alkenylstannane group is transformed into the corresponding alkenyl iodide upon treatment with iodine. A BuLi promoted lithium-iodine exchange105 results in the formation of alkenyllithium 176, which undergoes an intramolecular cyclization to furnish tertiary-allylic alcohol 177 after a suitable work-up procedure. Treatment of 177 with PCC-on-alumina causes an oxidative rearrangement69"72 which yields enone 38 (see Scheme 2.24). 173 SnBu 3 alkylation 36 C0 2 Et 174: R = SnBih I I ~ i / f . n  O M D U 3175: R = l n-BuLi 176: R = Li C0 2 Et PCC-on-alumina I oxidative rearrangement C0 2 Et 38 177 Scheme 3.3: Preparation of enone 38 using bifunctional reagent 36 3. Synthetic Approaches to (±)-lsomulinic Acid 63 Having decided to make use of the bifunctional reagent c/'s-5-iodo-1-tri-n-butylstannylpent-1-ene (36) to construct the seven-membered ring of the mulinane diterpenoids, the task then becomes the elaboration of a suitable material such as 178 to mulin-11,13-dien-20-oic acid (157, Scheme 3.4). Note that the presence of a methoxycarbonyl group on C-8 of 178 precludes the use of this functional group as a surrogate for the carboxylic acid function on C-5 of the natural product. Three synthetic issues must then be resolved: 1) adjustment of the oxidation state of C-17, 2) use of a suitable functional group at C-5 and its transformation to a carboxylic acid moiety, and 3) preparation of the seven-membered ring diene from the seven-membered ring enone. 178 157 Scheme 3.4: Late stages of the synthetic problem It appeared reasonable to use an acetal moiety at C-5 as a surrogate for the carboxylic acid moiety (Scheme 3.5). Thus, preparation of the carboxylic acid would likely involve an acid catalyzed deprotection of the aldehyde, followed by oxidation to the corresponding acid, and would occur at a late stage in the synthesis (179 to 157). 3. Synthetic Approaches to (±)- lsomulinic Acid 64 Scheme 3.5: Retrosynthesis of mulin-11,13-dien-20-oic acid to enone 183 Preparation of the diene on the seven-membered ring of 179 should be possible from enone 180. This material, in turn, should be accessible from ketone 181 using standard methods. In principle, the methyl group on C-8 (181) can be derived from the corresponding methyl carboxylate (182). It was recognized that this transformation (182 to 181) would potentially require protection of the ketone function at C-12. Finally, it was hoped that the c/s-fusion between the six- and seven-membered rings would be established through the direct hydrogenation of enone 183. Alternatively, the allylic alcohol obtained from the reduction of enone 183 could be used to direct the hydrogenation of the double bond under homogenous conditions.106,107 3. Synthetic Approaches to (±)-lsomulinic Acid 65 The presence of the methyl carboxylate moiety on C-8 may at first seem problematic. However, it must be noted that it imparts flexibility to this synthetic strategy. Indeed, a retrosynthetic analysis of mulinenic acid (154), which is oxygenated at C-17, reveals that it could potentially be derived from ketone 182 (Scheme 3.6). As mentioned earlier, the acetal moiety on C-5 should provide access to the carboxylic acid function of 154. The double bond between C-11 and C-12 could be established through oxidation of the arylselenide moiety at C-12 of compound 185, followed by thermolysis of the resulting selenoxide.108 The etherification reaction leading to 185 from alcohol 186 would occur under the action of phenylselenenyl chloride as shown by Nicolaou et a/.1 0 9 or /V-phenylselenophthalimide as demonstrated by Kuhnert and Maier.110 Preparation of alcohol 186 would require the reduction of the methoxycarbonyl group at C-8. Finally, alkene 187 should be accessible from ketone 182 using standard methods. ArSe Scheme 3.6: Retrosynthetic analysis of mulinenic (154) acid to ketone 182. 3. Synthetic Approaches to (±)- lsomulinic Acid 66 It was expected that the annulation method described in Scheme 3.3 would yield access to the pivotal enone (183). This process would require alkylation of keto ester 189 with c/'s-5-iodo-1-tri-n-butylstannylpent-1-ene (36) as shown in Scheme 3.7. Scheme 3.7: Retrosynthesis of enone 183 to ketoester 189 The planned use of keto ester 189 requires some comment. It is well known that the relative energy of cis- and frans-fused hydrindane systems depends largely on their substitution pattern.111 Dana and Lo Cicero112 have shown that hydrindanone systems similar to ketone 193 (Scheme 3.8) exist in dynamic equilibrium as a mixture of epimers under equilibrating conditions. The example illustrated in Scheme 3.8 indicates that ketone 193 could be susceptible to epimerization at C-10 in basic or acidic media. In principle, keto ester 189 should enolize as depicted in 192 (Scheme 3.8) and, as a result, should prevent epimerization at C-10. 3. Synthetic Approaches to (±)- lsomulinic Acid 67 Scheme 3.8: Dynamic equilibrium between ketones 190 and 191, 1 1 2 structure of ketone 193, and keto and enol forms of keto ester 189. It was expected that keto ester 189 would be accessible from ketone 193 through the use of Mander's method113 (Scheme 3.9). Oxidation of alcohol 194 under neutral conditions should provide access to ketone 193 without epimerizing the carbon chirality center at C - 10 . It was envisioned that a diastereoselective hydroboration reaction of alkene 195 would furnish alcohol 194 as the major product. It was postulated that the bulky 2',2'-dimethylpropylene acetal moiety of alkene 195 would impart a large degree of facial selectivity during the hydroboration event. Reduction of the ester moiety of 196 to the corresponding aldehyde, and subsequent protection of the aldehyde as the dimethylpropylene acetal would provide alkene 195. The reduction of the carbonyl group at C-8 of 197 to a methylene group was expected to occur through formation of the corresponding, dithioethylene ketal, followed by desulfurization with Raney nickel. Bicyclic enone 197 should be accessible from keto ester 198 through the use of a Robinson annulation. The preparation of keto ester 198 from inexpensive materials has been described in the literature.114 3. Synthetic Approaches to (±)-Ssomulinic Acid 68 3. Synthetic Approaches to (±)- lsomulinic Acid 69 3.4.2.2 First synthetic approach to (±)- isomulinic acid The proposed synthesis of (±)-isomulinic acid (39) required an efficient preparation of bicyclic enone 197 from keto ester 198. It was expected that this task would be accomplished through a Robinson annulation. Unfortunately, attempts to carry out this transformation in one step were not successful and, as a result, a three step synthesis was devised. The alkylation of keto ester 198 with methyl vinyl ketone could be carried out in small scale using thf lie acid as a catalyst as shown by Kotsuki et a/.115 Unfortunately, formation of a number of side products was observed when more than 3 mmol of keto ester 198 were used. The desired alkylation could also be carried out under the influence of an Fe(lll) catalyst as demonstrated by Christoffers116 (Scheme 3.10). Gratifyingly, this reaction could be carried out in large scale (162 mmol) and did not suffer from extensive formation of undesired products. Indeed, dione 199 was obtained in 63% yield and the major side-product (200) was isolated in only 5% yield. Some of the starting material (18%) was also recovered. This alkylation reaction is thought to occur through the initial formation of a stable octahedral complex (201) in which the enolate of keto ester 198 behaves as a resonance stabilized ligand (Scheme 3.11). Methyl vinyl ketone then coordinates to complex 201 to form intermediate 202. Note that although methyl vinyl ketone may approach the metal center from either side of the chelating ligand, the alkylation at C-1 occurs preferentially anti to the isopropyl group at C-5 for steric reasons. As a result the correct relative configuration is established at C-1 and C-5 of dione 199. Note also Scheme 3.10: Synthesis of dione 199 3. Synthetic Approaches to (±)-lsomuiinic Acid 70_ that methyl vinyl ketone must adopt an S-c/'s configuration in order for the alkylation to occur. Complex 203 is destabilized relative to complex 202 and, as a result, the ligand is released. Scheme 3.11: Pathway for the Fe(lll) mediated alkylation of keto ester 198 The spectral data collected on dione 199 was in full accord with the assigned structure. The IR spectrum clearly showed three strong absorptions at 1749 cm"1, 1731 cm"1 and 1718 cm"1, which account for the carbonyl functions in the molecule. In addition, the 1 3C NMR spectrum contained an ester carbonyl resonance at 8= 171.0 and two ketone carbonyl resonances at 8= 207.9 and 216.5. The 1H NMR spectrum contained the correct number of signals, including a three proton singlet at 8 = 2.08 corresponding to the methyl ketone function. Finally, an HRMS measurement of the parent ion was consistent with the molecular formula of the desired dione. The relative configuration at C-1 and C-5 could not be verified spectroscopically. However, the assigned relative configuration was supported by literature precedent. Indeed, Coates has shown that the analogous diketone 205 could be obtained from keto ester 204. Diketone 205 was converted to mesylate 206, the structure of which was determined by X-ray crystallography (see Figure 3.8).1 1 7 3. Synthetic Approaches to (±)-lsomulinic Acid 71 O S 0 2 Me 199 204 205 206 Figure 3.8: Structures of compounds 199, 204, 205 and 206 117 The major side product formed during the alkylation of keto ester 198 was determined to be trione 200 based on the spectral data collected. Indeed, both the 1H NMR and 1 3C NMR spectra contained the correct number of signals. The 1 3C NMR spectrum contained signals at 5 = 171.4, 207.8, 210.7 and 216.6, which account for the four carbonyl groups in the molecule. The 1H NMR spectrum showed two methyl singlets at 8= 2.13 and 2.17, which account for the two methyl ketone carbonyl functions. The IR spectrum displayed three carbonyl resonances at 1747, 1730 and 1716 cm"1. Moreover, the molecular mass of 200 was confirmed by an HRMS measurement of its parent ion. The relative configuration of the carbon chirality center at C-3 of trione 200 was not determined. The next step in the sequence involved a classical aldol condensation-dehydration sequence to form bicyclic enone 197. A review of the literature suggested the use of Scanio and Hill's method.118 Thus, reaction of dione 199 with pyrrolidine led to the formation dienamine 207 upon removal of water (Scheme 3.12). This material could not be fully characterized since it decomposed readily upon contact with silica gel. However, a GC-LRMS analysis indicated that the molecular mass of the major component of the crude mixture did indeed correspond to the expected dienamine. Compound 207 was hydrolyzed using the method devised by Johnson et a/.119 to furnish the desired enone (197). 3. Synthetic Approaches to (±)-lsomulinic Acid 72 H o .0 6 N a O A c , H O A c H 2 0 , P h H , A .0 P h H , A D e a n - S t a r k 199 207 197 7 6 % , 2 s t e p s Scheme 3.12: Synthesis of enone 197 The spectral data collected on enone 197 was in full accord with the assigned structure. The IR spectrum showed a resonance at 1664 cm"1, which was attributed to the oc,p-unsaturated ketone moiety. The 1H NMR spectrum displayed a one-proton singlet at 6 = 5.81 corresponding to the alkenyl proton. The 1 3C NMR spectrum showed an alkenyl signal at 8 = 123.4 corresponding to the unsaturated carbon at the position a to the ketone. Two signals at 8= 171.2 and 172.3 accounted for the ester carbonyl carbon and the unsaturated carbon at the fj position of the ketone. A signal observed at 8= 198.8 was assigned to the enone carbonyl carbon. An HRMS measurement of the parent ion was consistent with the molecular formula of the desired product. With an efficient synthesis of a bicyclic compound which possesses the correct relative configuration at C-3 and C-5 (mulinane numbering), the stage was set to prepare alkene 195. The reduction of the ketone carbonyl of enone 197 to a methylene group was achieved using a two-step sequence. Formation of dithioketal 208 was accomplished efficiently upon treatment of enone 197 with an excess of 1,2-ethanedithiol and a catalytic amount of BF3»Et20 in DCM (Scheme 3.13). The structure of dithiane 208 was verified by inspection of its spectral data. The IR spectrum showed a single carbonyl resonance at 1718 cm"1 corresponding to the ester carbonyl group. Both the 1 3C NMR and 1H NMR spectra displayed the correct number of signals. The carbonyl resonance observed at 8 = 198.8 in enone 197 was notably absent in the 1 3C NMR spectrum of dithiane 208. An HRMS measurement of the parent ion confirmed the molecular formula of 208. 3. Synthetic Approaches to (±)-lsomulinic Acid 73 The reduction of the dithiane function of 208 to a methylene group occurred under the action of Raney nickel (Scheme 3.13). Unfortunately, this desulfurization took place at a slow rate and in modest yield. A GC-LRMS trace indicated that the major product was the desired alkene and that some over-reduction to the saturated bicycle (209) had also occurred. Luckily, the over-reduction process took place at a much slower rate than the desired desulfurization. 197 208 196 66% 209 trace Scheme 3.13: Synthesis of alkene 196 Alkene 196 displayed spectral data in accordance with the assigned structure. The 1H NMR spectrum contained the correct number of signals, and the 1 3C NMR spectrum showed 14 carbon resonances. A J-mod 1 3C NMR experiment demonstrated that compound 196 contained five methylene groups, and a total of six methine and methyl groups. The IR spectrum was not particularly informative, but an HRMS measurement of the parent ion confirmed the molecular formula of alkene 196. Conversion of the ester moiety of compound 196 into an aldehyde was efficiently accomplished in two steps. Thus, treatment of alkene 196 with two equivalents of DIBAL-H resulted in the formation of the corresponding alcohol (210) in good yield (Scheme 3.14). The strong absorption at 3369 cm"1 was attributed to a hydroxyl OH stretching vibration. The three-proton singlet previously observed at 8 = 3.63 in the 1H NMR spectrum of methyl ester 196 was notably absent. A two-proton multiplet that appeared at 8= 3.54-3.59 in the 1H-NMR spectrum was assigned to the carbinol protons of alcohol 210. The signal occurring at 8 = 64.9 in the 1 3C NMR spectrum was attributed to the carbinol carbon. Finally, a J-mod 1 3C NMR experiment verified that the product contained six methylene groups, and an HRMS measurement of the parent ion verified the molecular formula of alcohol 210. 3. Synthetic Approaches to (±)- lsomulinic Acid 74 DIBAL-H (1.0 M in hexanes) Et 2 0, 0 °C 85% TPAP, NMO DCM, r.t. 91% 210 211 Scheme 3.14: Preparation of aldehyde 211 Alcohol 210 was oxidized to the corresponding aldehyde in excellent yield using Ley's method88 (Scheme 3.14). The appearance of a singlet at 5 = 9.65 in the 1H NMR spectrum served as evidence for the desired aldehyde. The carbonyl carbon was observed at 5 = 203.2 in the 1 3C NMR spectrum. The IR spectrum displayed a strong carbonyl stretch resonance at 1718 cm"1. Moreover, an HRMS measurement of the parent ion confirmed the molecular formula of aldehyde 211. The next task involved protection of the tertiary aldehyde moiety of 211 as a 2,2-dimethylpropylene acetal. This transformation was expected to occur through the use of an excess of 2,2-dimethylpropanediol and a catalytic amount of a suitable Bransted acid (Scheme 3.15). It was recognized that migration of the double bond from the six-membered ring to the five-membered ring could occur under these conditions. In the event, the desired acetal (195) was obtained in good yield and was accompanied by small amounts of a side product in which the double bond had migrated to the five membered ring (as in 212). 95% trace Scheme 3.15: Preparation of acetal 195 3. Synthetic Approaches to (±)-lsomulinic Acid 75 Inspection of the 1H NMR and 1 3C NMR data collected on the major product made it clear that the aldehyde moiety had been converted to the corresponding acetal. The 1 3C NMR spectrum showed a signal at 8= 103.4 that was assigned to the acetal carbon atom, and two signals at 8 = 77.4 and 78.2 that were assigned to the C-1' and C-3' of the 2',2'-dimethylpropylene acetal moiety. The 1H NMR spectrum displayed the diagnostic signals outlined in Figure 3.9. The long-range coupling (W-coupling) constant of 2.9 Hz allowed for the distinction between the equatorial and the axial protons. Figure 3.9: Diagnostic 1H NMR data for the acetal moiety of 195 As indicated in Scheme 3.15, a second acetal could be isolated in very low yield from the reaction mixture. The spectral data collected on this material was very similar to that collected for acetal 195. However, a 1D-TOCSY experiment demonstrated the presence of the spin system depicted in Figure 3.10 and thus verified that this material was the isomerized alkene (212). equatorial protons 8 = 3.57-3.62, dd, J = 10.7, 2.9 Hz 5 = 3.60-3.65, dd, J = 10.7, 2.9 Hz acetal proton 8 = 4.33, s H axial protons 5 = 3.28, d, J=10.7 Hz 5 = 3.36, d, J=10.7 Hz methyl groups 5 = 0.67, s 5 = 1.16, s HA:,5 = 5.44 (m) H B , H c , H D and H 5 = 2.22-2.30 (m, 1H) 8= 1.99-2.12 (m, 2 H) . 8 = 1.43-1.55 (m 1 H) irradiation of H A enhances H B , H c , H D , H E , Me A and Me B Figure 3.10: Spin system present in alkene 212 as determined using a selective 1D TOCSY experiment 3. Synthetic Approaches to (±)-lsomulinic Acid 76 With alkene 195 in hand, the stage was set for the hydroboration experiment that was expected to introduce the carbon chirality center at C-10 (mulinane numbering) with the correct relative configuration. A number of hydroborating agents were screened including BH3»THF, BH3«SMe2, 9-BBN, catechol borane and catechol borane-LiBH4. These preliminary experiments demonstrated that the best results were obtained using BH3«SMe2 as the hydroborating reagent in DCM. A mixture of alcohols 194 and 213 (see Scheme 3.16) was invariably obtained. However, the composition of the mixtures and the total yield of the two alcohols (194 and 213) varied. An examination of the more polar material derived from the hydroboration-oxidation reaction revealed that diol 214 was the major side product. OH OH OH Scheme 3.16: Hydroboration and oxidation of alkene 195 The mixture of alcohols 194 and 213 was easily separated from diol 214 using flash column chromatography on silica gel. Diol 214 displayed two three-proton singlets at 8 = 2.11 and 8 = 2.20 in the 1H NMR spectrum. The diagnostic signals corresponding to the acetal moiety (see Figure 3.9) were notably absent. The 1 3C NMR spectrum displayed signals at 8 = 71.0, 71.5, 75.4 and 80.4, which account for all four carbon atoms bonded to oxygen. The IR spectrum clearly showed the expected absorbance at 3370 cm"1. Finally, an HRMS measurement of the parent ion confirmed its molecular formula. The structure of diol 214 was further confirmed by its oxidation88 to the corresponding keto aldehyde (215, Scheme 3.17). This material exhibited two distinct 3. Synthetic Approaches to (±)-lsomulinic Acid 77 AB spin systems in its 1H NMR spectrum, which can be assigned to the two methylene groups flanking the ether oxygen atom. In addition, a singlet at 6 = 9.51 indicated the presence of the expected tertiary' aldehyde. The 1 3C NMR spectrum showed two carbonyl signals at 5 = 213.8 and 8 = 204.9 which can be assigned to the ketone and aldehyde carbonyl carbons, respectively. The IR spectrum of this material displayed two strong absorptions at 1707 cm"1 and 1729 cm"1 which also indicated the presence of a ketone and an aldehyde carbonyl. Finally, an HRMS measurement of the parent ion was consistent with the molecular formula of keto aldehyde 215. Scheme 3.17: Oxidation of diol 214 to keto aldehyde 215 The fact that no isomers of diol 214 were isolated indicates that the reductive opening of the acetal moiety occurred in an intramolecular fashion. A postulated mechanism for this side reaction is presented in Scheme 3.18. Thus, after initial hydroboration of alkene 195 to provide the c/'s-fused borane derivative (216), one of the oxygen atoms on the acetal group can coordinate to the boron atom to form ylide 217. The juxtaposition of the activated acetal carbon and the borohydride moiety allows for the internal reduction of C-20 (mulinane numbering) to form intermediate 218. Oxidation of the borane (219) results in formation of the observed diol (214). 3. Synthetic Approaches to (±)- lsomulinic Acid 78 214 219 218 Scheme 3.18: Proposed mechanism for the formation of diol 214 Although alcohols 194 and 213 were difficult to separate by flash column chromatography, a small amount of the trans-fused alcohol (194) was purified and its structure was fully assigned based on extensive NMR experiments. The relative configuration of the two newly introduced carbon chirality centers at C-5 and C-6 was confirmed through a number of selective NOE difference experiments as shown in Figure 3.11. 0 3. Synthetic Approaches to (±)- lsomulinic Acid 79 partial structure of alcohol 194 H A : 5 = 4.25 (s) signal is enhanced by irradiation of H B and H E H B : 5 = 4.10-4.17 (ddd, J= 10, 10, 5 Hz) signal is enhanced by irradiation of H A and H c H c : 5 = 1.96-2.04 (m) signal is enhanced by irradiation of H B H D : part of a multiplet at 5 = 1.74-1.91 signal is enhanced by irradiation of H A and H B H E : 8 = 2.41-2.50 (ddd, J= 13.0, 3.0, 3.0 Hz) signal is enhanced by irradiation of H A H F : part of a multiplet at 8 = 1.23-1.48 signal is not enhanced by irradiation at H A or H B Figure 3.11: Selected NOE difference data observed for alcohol 194 In view of the tedious separation of alcohols 194 and 213, it was decided to carry out an oxidation88 of the mixture (Scheme 3.19). It was gratifying to find that the mixture of the corresponding ketones (193 and 220) could be easily separated by flash column chromatography on silica gel. TPAP, NMO DCM 194 213 193 57% 220 38% Scheme 3.19: Oxidation of alcohols 194 and 213 to ketones 193 and 220 Ketone 193 exhibited a strong carbonyl absorption at 1708 cm"1. In addition, its molecular formula was confirmed through an HRMS measurement of its parent ion. A resonance at 5 = 210.7 in the 1 3C NMR spectrum confirmed the presence of a ketone carbonyl. Although, the structure of ketone 193 could be fully assigned, no conclusive evidence regarding the relative configuration of the carbon chirality center at C-6 could 3. Synthetic Approaches to (±)-lsomulinic Acid 80 be obtained. However, in a selective NOE difference experiment, irradiation of the acetal proton did not enhance the signal corresponding to the proton at C-6 (see Figure 3.12). This result indicated that ketone 193 was likely the desired frans-fused product. 193 H B Figure 3.12: Selected NOE difference data observed for ketone 193 The c/s-fused ketone (220) displayed a strong carbonyl resonance at 1708 cm"1. An HRMS measurement of the parent ion verified the molecular formula. A signal at 5 = 214.4 in the 1 3C NMR spectrum demonstrated the presence of a ketone carbonyl. In order to confirm the conclusions regarding the newly introduced carbon chirality center, a full suite of NMR experiments were performed on ketone 220 and its structure was fully assigned. A number of selective NOE difference experiments performed on ketone 220 confirmed the relative configuration at C-6 (Figure 3.13). irradiation of H A does not enhance H B HA:8 = 4.15(s) H B ; 5 = 2.11-2.18 (dd, J= 12.0, 7.3 Hz) H A : 5 = 4.27 (s) signal is enhanced by irradiation of H B H B : 5 = 2.86-2.93 (dd, J = 7.5, 7.5 Hz) signal is enhanced by irradiation of H A H c : 8 = 2.14-2.24 (ddd, J= 14.7, 8.3, 3.1 Hz) signal is enhanced by irradiation of H B 220 Figure 3.13: Selected NOE difference data observed for ketone 217 As indicated in Scheme 3.16, the hydroboration-oxidation sequence leading to alcohols 194 and 213 was not very selective and required optimization. To this end, a series of experiments were carried out in which the reaction solvent (Et20, pentane, 3. Synthetic Approaches to (±)-lsomulinic Acid 81 DCM and THF), reaction temperature (0 °C, 10 °C and r.t.) and the amount of hydroborating agent (1.3 and 10 equivalents of BH3»SMe2) were varied. Scheme 3.20 shows the best conditions found in small "scale reactions (0.23 mmol of alkene 195). Similar results were observed when the reaction was scaled up (see experimental section for details). OH OH 195 194 213 96 % combined yield based on recovered starting material (2:1) Scheme 3.20: Optimized conditions for the hydroboration of alkene 195 to alcohol 194 With access to sufficient amounts of ketone 193, it became possible to explore the preparation of keto ester 189. The installation of the methoxycarbonyl moiety was efficiently accomplished using the method developed by Mander113 (Scheme 3.21). Unfortunately, characterization of the keto ester proved difficult. TLC analysis of the reaction mixture indicated the presence of two products. Inspection of the 1H NMR spectrum indicated that the compound existed as a mixture of isomers 1) LDA, THF,-78 °C 2) O u NC OMe -78 °C to r.t. 80% 193 C 0 2 M e Scheme 3.21: Preparation of keto ester 189 3. Synthetic Approaches to (±)-lsomuiiriic Acid 82 According to the synthetic plan, keto ester 189 was to be alkylated with the bifunctional reagent c/'s-5-iodo-1-tri-n-butylstannylpent-1-ene (36, see Scheme 3.3 and Scheme 3.7). It was decided to probe the stereochemical outcome of the alkylation process using methyl iodide as the electrophile (Scheme 3.22). Recall that this reaction was expected to establish the new carbon chirality center at C-8 (mulinane numbering) with the correct relative configuration. Thus, the alkylation of enolate 222 should proceed to give the axial methyl group based on stereoelectronic120 grounds. In the event, alkylation of keto ester 189 furnished keto ester 221 as a single product. Scheme 3.22: Stereoselective alkylation of keto ester 189 Compound 221 exhibited a three-proton singlet at 8= 1.40 in the 1H NMR spectrum corresponding to the newly introduced methyl group. The methyl ester moiety was observed as a three-proton singlet at 8 = 3.69 in the 1H NMR spectrum. In addition, the ester carbonyl group gave rise to a signal at 8= 174.3 in the 1 3C NMR spectrum. Due to a fortuitous dispersion of signals in the 1H NMR spectrum, the structure of keto ester 221 could be fully assigned. A set of one dimensional NOE difference experiments verified the relative configuration at C-4 and C-6 (Figure 3.14). Note that these experiments also serve to indirectly prove the structure of keto ester 189. 3. Synthetic Approaches to (±)-lsomulinic Acid 83 H A : 8 = 1.28-1.36 (ddd, J = 9.4, 9.4, 9.4 Hz) signal is enhanced by irradiation of H B H B : 8 = 2.38-2.43 (dd, J = 11.8, 7.2 Hz) H c : 8 = 4.20 (s) signal is not enhanced by irradiation of H B Me A : 8 = 1.40 (s) signal is enhanced by irradiation of H B Figure 3.14: Selected NOE difference data for ketoester 221 Although keto ester 221 could be synthesized with the correct relative configuration at C-3, C-5, C-8 and C-10 (mulinane numbering), it was recognized that this material was susceptible to epimerization at C-10 (mulinane numbering). Indeed, the sample used to obtain the NMR data for compound 221 had already begun to epimerize to keto ester 224 in the NMR tube, presumably through the enolization of the ketone group under the action of trace amounts of DCI to form enol 223 (Scheme 3.23). Scheme 3.23: Epimerization of ketoester 218 to ketoester 221 This undesired epimerization indicated that the proposed intermediates 225 and 188 (Figure 3.15) would also be susceptible to epimerization, and would have to be used shortly after their preparation. 3. Synthetic Approaches to (±)- lsomulinic Acid 84 Figure 3.15: Proposed intermediates susceptible to epimerization Surprisingly, the proposed alkylation of keto ester 189 with the bifunctional reagent c/s-5-iodo-1-tri-n-butylstannylpent-1-ene (36) did not take place under the conditions outlined by Walker121 or under the conditions used to prepare keto ester 221. The use of KHMDS as the base led to an inseparable mixture of the desired ketoester (225) and its C-10 (mulinane numbering) epimer 226. Initially it was thought that the observed epimerization resulted from deprotonation of keto ester 225 with excess base. However, a mixture of 225 and 226 was observed even when a deficiency of base was used. The reason(s) for the epimerization of 225 to 226 remain unclear. Scheme 3.24: Alkylation of ketoester 189 with the bifunctional reagent c/s-5-iodo-1 -tri-n-butylstannylpent-1 -ene (36) The problems associated with separating keto ester 225 from its C-10 epimer (226) rendered it difficult to characterize. However, it could be shown that the mixture 3. Synthetic Approaches to (±)- lsomulinic Acid 85_ of keto esters 225 and 226 converges to 226 exclusively upon treatment with base (Scheme 3.25). It could also be shown that keto ester 226 can be prepared as a single compound from ketone 220. Alkenylstannane 226 was converted to alkenyl iodide 227 upon treatment with iodine. Bu 3 SrK B u 3 S n . B u 3 S n 227 Scheme 3.25: Preparation of alkenyl iodide 227 The spectral data collected on alkenyl iodide 227 verified its structure. The 1H NMR spectrum displayed a two-proton multiplet at 5= 6.10-6.20, which was assigned to the alkenyl protons. The 1 3C NMR spectrum displayed two signals at 8= 82.8 and 140.7 which were attributed to the vinyl iodide moiety. A J-mod 1 3C NMR experiment demonstrated the presence of nine methylene groups and a total of eleven methyl and methine groups. The IR spectrum showed two carbonyl stretching absorptions at 1702 and 1736 cm"1. The molecular mass of 227 was verified through an HRMS measurement of its parent ion. 3. Synthetic Approaches to (±)-lsomulinic Acid 86 The fact that the relative stereochemistry at C-10 (mulinane numbering) could not be controlled, combined with the difficulties in obtaining alcohol 194 exclusively from the hydroboration of alkene 195 led us to abandon this synthetic approach. At this point it was apparent that the seven-membered annulation method developed by Piers et al.34 would not yield access to (±)-isomulinic acid (39). However, the intriguing synthetic challenge posed by this natural product prompted the formulation of a new synthetic plan. 3. Synthetic Approaches to (±)-lsomulinic Acid 87 3.4.3 Second synthetic approach to (±)-isomulinic acid 3.4.3.1 Retrosynthetic analysis The work described in section 3.4.2.2 made it clear that the presence of a carbonyl group adjacent to C-10 jeopardized the integrity of the carbon chirality center at that position. Therefore, the planned use of any intermediates in which the methine group at C-10 was activated, and thus prone to epimerization, was avoided. The second synthetic approach to the mulinane family of natural products focused on trying to quickly establish a tricyclic intermediate, exemplified by the general structure 229 (Figure 3.16), which could potentially be elaborated to the final product. Note that structure 229 has in place all the carbon chirality centers present in 157 with the correct relative configuration. Recall that compound 157 can, in principle, give rise to isomulinic acid (39, see Scheme 3.2). At this stage, the size and the functionalization of the third ring, which is to become the seven-membered ring in the final product, was left undefined. It was envisioned that the relative stereochemistry at C-8 and C-9 of 229 would be established during the photocycloaddition of alkene 228 to a suitable partner. Note that bicycle 228 has in place three of the five carbon chirality centers present in the proposed tricyclic structure (229). Note also the planned use of a methoxycarbonyl group at C-5 in 228 as a surrogate for the carboxylic acid moiety of 157. 228 229 157 Figure 3.15: Key proposed intermediates in the second retrosynthesis of (±)- isomulinic acid (39) 3. Synthetic Approaches to (±)-lsomulinic Acid 88 Two photochemical processes for the construction of carbocycles were considered. De Mayo and coworkers122 have demonstrated that the photocycloaddition of an alkene (230) to a vinylogous anhydride such as 231 can be used to construct seven-membered rings. Note that the double bond of alkene 230 may be embedded within a ring. An overview of this useful reaction sequence is shown in Scheme 3.26. The photocycloaddition can, in theory, give rise to a total of eight possible products. These can be divided into two subsets of regioisomers, which are exemplified by 232 and 233. Each subset may contain up to four isomers which arise from photocycloaddition at either face of alkene 230 and from the endo and exo modes of photocycloadditon. The number of isomers is reduced to four during the ring expansion step (234, 235 and their corresponding diastereomers arising from photocycloaddition on the other face of alkene 230). Scheme 3.26: Overview of the de Mayo reaction1 1 2 Analysis of molecular models of alkene 228 indicates that it should exist in the conformation depicted in Scheme 3.27, and that the p-face of the molecule is blocked by the methoxycarbonyl group. As a result, the vinylogous anhydride 231 should approach the double bond from the less sterically hindered a-face during the photocycloaddition. Therefore the major photocycloaddition product should be 3. Synthetic Approaches to (±)-lsomulinic Acid 89 compound 236. Treatment of compound 236 with NaOMe should promote the strain-driven ring expansion and furnish tricycle 237. Note that 237 possesses all five carbon chirality centers present in 157 (Figure 3.15) with the correct relative configuration. At this stage no specific plan was drawn for the elaboration of diketone 237 to compound 157. However, the functional groups on the seven-membered ring should allow for the preparation of the desired diene. o Scheme 3.27: Possible construction of 237 through a de Mayo reaction The second and lesser known photochemical process that could be potentially used for the preparation of a tricyclic structure such as 229 was developed by Baldwin and Wilkinson.123 An overview of the reaction sequence is presented in Scheme 3.28. Photochemical cycloaddition of 230 to the vinylogous carbonate124 238 can yield up to eight possible products. As before, these can be subdivided into two subsets of regioisomers (239 and 240). Each subset may contain up to four isomers which arise from photochemical cycloaddition at either face of alkene 230 and from the endo and exo modes of photochemical cycloaddition. The number of isomers is reduced to four during the reduction of the ester carbonyl and the ensuing, strain-driven, carbon bond 3. Synthetic Approaches to (±)-lsomulinic Acid 90 cleavage (241, 242 and their corresponding diastereomers arising from photocycloaddition on the other face of alkene 230). Keto aldehydes 241 and 242 readily undergo an aldol condensation to form enones 243 and 244. An intramolecular version of this photocycloaddition reaction has been developed by Winkler190 and has been employed in the total synthesis of challenging natural products.191,192 ;"R,TR2 + H T 0 ••'R'' M e R 2 M e M e 230 238 M e r - ' R i ' UK 239 M e R 3 M e M e 240 1) reduction 2) work-up (H+) cyclization R 3 241 O 242 Scheme 3.28: Overview of the photocycloaddition-reduction-cyclization sequence developed by Baldwin1 2 3 Inspection of molecular models of alkene 228 and the vinylogous carbonate 238 indicates that their major photocycloaddition product should be 245 (Scheme 3.29). Treatment of 245 with DIBAL-H, followed by a suitable work-up procedure should furnish enone 246. Note that 246 has in place the five carbon chirality centers present in 157 with the correct relative configuration. At this stage no specific plan was drawn for the elaboration of enone 246 to compound 157. However, the functional groups on the six-membered ring should allow for the preparation of the desired diene. 3. Synthetic Approaches to (±)-lsomulinic Acid 91 1 238 Scheme 3.29: Possible construction of tricycle 246 using the reaction sequence developed by Baldwin. At this stage the synthetic problem is reduced to the preparation of alkene 228. The initial strategy involved the preparation of 228 from enol triflate 247 and a methyl cuprate (Scheme 3.30). It was anticipated that a mixture of enol triflates 247 and 248 would be obtained from ketone 249, therefore, a separation and recycle process was part of the plan. Ketone 249 was expected to be accessible from alcohol 250. A directed homogeneous hydrogenation106,107 of allylic alcohol 251 was expected to furnish alcohol 250, which has the correct relative configuration at the newly formed carbon chirality center (C-10). It was anticipated that the hydroxyl group of alcohol 251 would exert a stronger directing effect than the methoxycarbonyl group.125 The preparation of alcohol 251 from enone 197 was expected to be a straightforward process. 3. Synthetic Approaches to (±)-lsomulinic Acid 92 Scheme 3.30: Retrosynthetic analysis of alkene 228 3. Synthetic Approaches to (±)-lsomulinic Acid 93 3.4.3.2 Second synthetic approach to (±)- isomulinic acid The reduction of enone 197 to alcohol 252 was easily accomplished under Luche conditions126,127 (Scheme 3.31). A small amount (5%) of the C-3 epimer of alcohol 252 was also generated, but could be easily separated from the desired material using flash column chromatography on silica gel. The IR spectrum of 252 contained a strong absorption at 3383 cm"1, which indicated the presence of the hydroxyl function. The carbinol proton was observed as a broad signal at 8 = 4.22 and the alkenyl proton was observed as a broad singlet at 8 = 5.46 in the 1H NMR spectrum. An HRMS measurement of the parent ion was consistent with the molecular formula. H 197 252 253 Scheme 3.31: Preparation of alcohol 252 and lactone 253 The relative configuration at C-3 was easily confirmed by formation of the corresponding bridged 8-lactone (253) when alcohol 252 was treated with a stoichiometric amount of KH. Lactone 253 exhibited a strong carbonyl stretching absorption at 1742 cm"1. The 1H NMR spectrum showed that the C-3 proton (observed as a ddd at 8 = 5.03-5.11) and the alkenyl proton at C-2 (observed as a ddd at 8 = 6.06-6.12) shared a mutual coupling constant of 5.0 Hz. The configuration at C-3 of allylic alcohol 252 was inverted using the method developed by Mitsunobu128 (Scheme 3.32). As expected, the IR spectrum of alcohol 251 showed an absorption at 3428 cm'1, which accounted for the hydroxyl group, and a strong absorption at 1725 cm"1 due to the ester carbonyl function. The carbinol proton was observed as a broad signal at 8 = 4.06 in the 1H NMR spectrum. 251 Scheme 3.32: Inversion of configuration at C-3 of allylic alcohol 252 With alcohol 251 in hand, the stage was set for the crucial directed homogeneous hydrogenation106,107 which would establish the carbon chirality center at C-10 (mulinane numbering). No reaction was observed when Wilkinson's catalyst66 was used in an attempted hydrogenation (Scheme 3.33). Treatment of alcohol 251 with a catalytic amount of Crabtree's catalyst6 5 , 1 2 9 , 1 3 0 under hydrogen at one atmosphere provided a mixture of products (Scheme 3.33). GC-LRMS analysis of the mixture indicated that the major component had the same molecular mass as the starting material. In addition, TLC analysis of the mixture showed that the major product was much less polar than alcohol 251. These observations suggested that allylic alcohol 251 had isomerized to enol 255, which tautomerized to the corresponding ketone (256). 3. Synthetic Approaches to (±)- lsomulinic Acid 95 ..OH isomerization H .OH No Reaction H 2 (g, 1 atm) DCM Crabtree's catalyst H 2 (g, 1 atm) DCM Wilkinson's catalyst C0 2Me 255 tautomerization + RhCI(PPh3)3 H Wilkinson's catalyst Crabtree's catalyst P F 6 -C0 2Me 256 80% .0 Scheme 3.33: Attempted homogeneous hydrogenation of alcohol 251 using Wilkinson's catalyst and Crabtree's catalyst Ketone 256 exhibited a carbonyl absorption at 1724 cm'1 in the IR spectrum. The 1H NMR spectrum did not show evidence of a carbinol proton. A resonance at 5 = 212.6 in the 1 3C NMR spectrum was attributed to the ketone carbonyl carbon. Although the structure of ketone 256 could be fully assigned, it was not possible to determine the relative configuration of the newly formed carbon chirality center by spectroscopic means. The relative configuration at C-6 of ketone 256 was determined indirectly. Keto ester 256 was efficiently reduced to diol 257 with DIBAL-H. This material was oxidized to the corresponding keto aldehyde (258) using Ley's method88 (Scheme 3.34). Keto aldehyde 258 exhibited carbonyl stretching absorptions at 1721 cm"1 and 1709 cm'1 in the IR spectrum. The 1 3C NMR spectrum displayed two carbonyl resonances at 5 = 203.5 and 8= 212.0, which were attributed to the aldehyde and ketone carbonyl carbons respectively. A singlet at 5= 9.71 in the 1H NMR spectrum provided further evidence for the aldehyde moiety. 3. Synthetic Approaches to (±)-lsomulinic Acid 96 H H H ,0 .OH .0 1) DIBAL-H TPAP, NMO Et 2 0 2) MgS0 4»7H 20 DCM C 0 2 M e 256 257 258 73%, 2 steps Scheme 3.34: Preparation of keto aldehyde 258 A full suite of NMR experiments allowed the complete structural assignment of keto aldehyde 258. The relative configuration at C-6 was determined using a selective 1D NOE difference experiment (Figure 3.15). Enhancement of the angular proton (HB) upon irradiation of the aldehyde proton (HA) provided strong evidence for the c/s-fusion of keto aldehyde 258. Figure 3.15: Selected NOE difference data for keto aldehyde 258 In view of these results, it became necessary to explore other methods for directed hydrogenation. Evans131 has demonstrated that Brown's catalyst132 can be used to effect directed hydrogenations of allylic alcohols similar to 251. This process is rather sensitive to substrate concentration, hydrogen concentration and catalyst load. After some experimentation, the desired frans-fused alcohol (250) could be obtained as the sole product from the directed hydrogenation of 251. P H A :5 = 9.71 (s) HQ. 6 = 2.60-2.71 (dddd, J= 10.4, 6.8, 6.4, 5.9 Hz) irradiation of H A enhances H B 3. Synthetic Approaches to (±)-lsomulinic Acid 97 Scheme 3.35: Directed homogeneous hydrogenation of allylic alcohol 251 using Brown's catalyst The carbinol proton of alcohol 250 was observed as a multiplet at 8= 4.02-4.10 in the 1H NMR spectrum. The previously observed alkenyl proton signal was notably absent. A J-mod 1 3C NMR experiment demonstrated the presence of five methylene groups and a total of seven methyl and methine groups. An HRMS measurement of the parent ion was consistent with the molecular formula of alcohol 250. Alcohol 250 was cleanly oxidized to the corresponding ketone using Ley's method88 (Scheme 3.36). A resonance at 8= 210.9 in the 1 3C NMR spectrum of ketone 249 was attributed to the ketone carbonyl carbon. The molecular formula of 249 was confirmed by an HRMS measurement of the parent ion. A fortuitous dispersion of signals in the 1H NMR spectrum allowed the complete structural assignment of ketone 249. The relative configuration of the carbon chirality center at C-6 could not be verified by spectroscopic means, but was assigned to be as depicted in Scheme 3.36 by comparison to ketone 256. Scheme 3.36: Preparation of ketone 249 3. Synthetic Approaches to (±)- lsomulinic Acid 98 The synthetic plan called for the preparation of alkene 228 through the intermediacy of the related enolate (259, see Scheme 3.37). As mentioned previously, it was expected that a mixture of enolates (259 and 260) would be obtained upon deprotonation of ketone 249 with strong base. Scheme 3.37: Planned deprotonation of ketone 249 Treatment of ketone 249 with strong base under conditions which allow the formation of the most thermodynamically stable product (e.g. a slight deficiency of base with respect to the amount of ketone used), followed by trapping of the enolate with a sililating agent provided a single enol ether (261) as the major product (Scheme 3.38). The position of the double bond was established using a series of selective 1D NOE difference experiments as shown in Figure 3.18. Scheme 3.38: Formation of enol ether 261 3. Synthetic Approaches to (±)- lsomulinic Acid 99 Me M e / .OMe OTBS H A ; 5 = 4.80-4.89 (m) H B ; 5 = 2.88-2.98 (dd, J= 16.2, 6.4 Hz) H c ; 8 = 1.80-1.91 (m) Me A ;5 = 0.96 (d, .7=6.4 Hz) 261 irradiation of H A enhances H B and H c irradiation of H B enhances H A , H c and Me A Figure 3.18: Selected NOE difference data for enol ether 261 Note that the C-5 methylene group of ketone 249 (see Scheme 3.38) is sterically hindered relative to the methylene group at C-3 by virtue of the adjacent five-membered ring. Consequently, it was expected that the deprotonation of ketone 249 under kinetic conditions, followed by trapping of the enolate with a silylating agent would favour the formation of enol ether 261. Therefore it appeared that a feasible preparation of alkene 228 would not be achieved using the synthetic plan outlined in Scheme 3.30. It is clear that the preparation of alkene 228 from enone 197 (Scheme 3.31) required the installation of the carbon chirality center at C-6 with the correct relative configuration, and the selective activation of C-4 and C-5 (Figure 3.19). The work outlined in section 3.4.2.2 suggested that a substrate similar to ketone 193, in which C-4 and C-5 are activated by the carbonyl group, would not be a suitable intermediate in a synthesis of alkene 228. Moreover, it has been established that although the carbonyl group at C-4 of ketone 249 activated C-5 and did not jeopardize the integrity of the carbon chirality center at C-6, this substrate was not useful in the preparation of alkene 228. After some thought, it became apparent that the allylic alcohol moiety embedded in compound 252 activated the desired carbon atoms. It seemed that it should be possible to use this feature of alcohol to 252 to prepare alkene 228. 3. Synthetic Approaches to (±)-lsomulinic Acid 1 0 0 C 0 2 M e 228 C 0 2 M e 249 C 0 2 M e 252 Figure 3.19: Structures of alkene 228, ketones 193 and 249, and alcohol 252 At this time two literature reports inspired a second possible strategy for the preparation of alkene 228. The first of these was Myers' highly regio- and stereoselective method for the reductive 1,3-transposition of allylic alcohols133"135 (Scheme 3.39). Treatment of an allylic alcohol (262) with NBSH 1 3 6 , 1 3 7 under Mitsunobu128 conditions at -30 °C results in the formation of the corresponding /V-allylic sulfonyl-hydrazine (263). This material decomposes at higher temperatures to form an allylic diazene intermediate (264). Diazene 264 undergoes elimination of molecular nitrogen which results in the reductive transposition of the double bond to form alkene 265. OH H 262 NBSH, DEAD Ph 3P, NMM -30 °C H 2N \ ^SOoAr N ,<H R 2 263 r.t. H—N 4^ Pu *N H R 3 ,H R 2 264 j -N 2 (g) R 2 265 Scheme 3.39: Myers' method for reductive 1,3-transposition of allylic alcohols It appeared likely that this method could be utilized to prepare alkene 266 from allylic alcohol 252 (Scheme 3.40). Note that alkene 266 possesses the correct relative 3. Synthetic Approaches to (±)-lsomutinic Acid 101 configuration at C-6 and has C-4 and C-5 activated by virtue of the double bond. In order to prepare alkene 228 it was necessary to install a carbon atom on C-4, and this would require the discrimination between C-4 and C-5. Scheme 3.40: Planned preparation of alkene 266 from allylic alcohol 252 Barrero et a/.138 have shown that substrates with appropriately positioned epoxide and methoxycarbonyl groups, such as 267, can be efficiently converted to hydroxylactones (268) upon treatment with a Lewis acid (Scheme 3.41). The reaction is thought to occur in a concerted fashion as shown in 269. Scheme 3.41: Formation of hydroxylactone 268 from epoxyester 267 as demonstrated by Barrero etal.™ In the context of a planned synthesis of alkene 228, alkene 266 would have to be oxidized to oxirane 270 (Scheme 3.42). For steric reasons, the relative configuration of the carbon chirality centers at C-4 and C-5 of epoxide 270 was expected to be as shown. Treatment of 270 with a Lewis acid should furnish the expected bridged hydroxylactone (271). Alcohol 271 should be readily oxidized to 3. Synthetic Approaches to (±)-lsomulinic Acid 102 ketone 272, which should furnish alkene 273 upon selective olefination. It was hoped that lactone 273 would provide allylic alcohol 274 upon treatment with NaOMe in MeOH. In principle, Myers' method for reductive transposition of allylic alcohols133 would allow access to alkene 228 from alcohol 274. Scheme 3.42: Implementation of Barrero's1 3 8 and Myers' 1 3 4 methods in a planned synthesis of alkene 225 Subjection of alcohol 252 to Myers' conditions for reductive transposition of the double bond furnished the desired alkene, albeit in modest yield (Scheme 3.43). The alkene function of 266 was evidenced by two signals at 5 = 5.40-5.49 and 6 = 5.68-5.75 with a mutual coupling constant of 9.9 Hz in the 1H NMR spectrum. The 1 3C NMR spectrum displayed two resonances at 5= 126.6 and 8= 129.4 which provided further evidence for the presence of the double bond. No indication of a hydroxyl group was found in the IR spectrum. Finally, an HRMS measurement of the parent ion was consistent with the molecular formula of the requisite alkene. No effort to establish the relative configuration of the carbon chirality center at C-6 of alkene 266 was made at this stage. However, based on the mechanism for the reductive transposition of the double bond, the configuration at C-6 was assumed to be as shown in Scheme 3.43. 3. Synthetic Approaches to (±)-lsomulinic Acid 103 The next step involved the diastereoselective epoxidation of alkene 266 to epoxide 270. Analysis of molecular models of alkene 266 indicates that it should adopt a conformation in which the (3-face of the double bond is blocked by the methoxycarbonyl group. As a result, the oxidizing agent should approach the double bond from the less sterically hindered a-face. Therefore, the relative configuration at C-4 and C-5 of epoxide 270 was anticipated to be as depicted in Scheme 3.43. The epoxidation of alkene 266 was accomplished smoothly under the action of dimethyl dioxirane (DMDO). The 1H NMR spectrum of a crude mixture of products indicated that two epoxides had been obtained in approximately 10:1 ratio. Fortunately, the major product could be isolated using flash column chromatography on silica gel. The 1H NMR spectrum and the 1 3C NMR spectrum of epoxide 270 both showed the correct number of signals. The absence of any signals in the olefinic region of the 1H NMR spectrum of 270, and the appearance of two signals at 8 = 2.97-3.02 and 8 = 3.16-3.20 with a mutual coupling constant of 4.1 Hz clearly showed that the alkene moiety had been oxidized to the desired epoxide. Finally, the molecular formula of 270 was confirmed by an HRMS measurement of its parent ion. At this stage no efforts were made to establish the relative configuration at C-4 and C-5 of compound 270. Scheme 3.43: Preparation of epoxide 270 With access to sufficient amounts of epoxide 270, the stage was set for the proposed lactonization-epoxide opening reaction under the conditions reported by Barrero et a/.1 3 8 Curiously, treatment of epoxide 270 with BF3»Et20 in DCM at 0 °C resulted in the formation of a complex mixture of products. However, treatment of epoxide 270 with BF3»Et20 in the presence of lithium dimethyl cuprate in Et20 at 0 °C 3. Synthetic Approaches to (±)- lsomulinic Acid 104 led to the formation of hydroxylactone 271 (Scheme 3.44). This material did not lend itself to chromatographic purification on silica gel and was, therefore, directly oxidized to the corresponding keto lactone (272). Two absorptions at 1798 cm"1 and at 1742 cm"1 in the IR spectrum of 272 clearly indicated the presence of a y-lactone and a ketone moiety, respectively. The molecular formula of 272 was confirmed by an HRMS measurement of its parent ion. The ketone carbonyl carbon was observed at 8 = 204.5 and the lactone carbonyl was observed at 8 = 177.1 in the 1 3C NMR spectrum. 270 271 272 Scheme 3.44: Preparation of keto lactone 272 Gratifyingly, the disperse nature of the 1 H NMR spectrum of keto lactone 272 facilitated the complete assignment of all the protons on the six-membered ring with the aid of a COSY experiment (Figure 3.20). This allowed the corroboration of the relative configuration of the carbon chirality center at C-6 through the use of two selective NOE difference experiments. It is interesting to note that the lack of coupling between HA and H B , and between H C and H F , indicates that the six-membered ring of 272 resides in a twist-boat conformation. H H 5 0 272 partial structure of 272 H A : 8 = 2.32-2.38 (dd, J= 9, 9 Hz) H B : 8 = 4.35 (s) H c : 8 = 2.47-2.55 (dd, J = 12.7, 9.1 Hz) H D : 8 = 1.66-1.76 (ddd, J= 12.7,11.5, 6.9 Hz) H E : 8 = 2.59-2.69 (ddd, J= 16.5, 11.5, 9.1 Hz) Hp: 8 = 2.36-2.45 (dd, J = 16.5, 6.9, Hz) irradiation of H D and H B both enhance H A Figure 3.20: Selected NMR data for keto lactone 272 3. Synthetic Approaches to (±)-lsomulinic Acid 105 At this stage it was necessary to install the carbon atom on C-4 of keto lactone 272. Subjection of 272 to Takai8 2"8 4 , 8 7 olefination conditions did not prove fruitful. The desired transformation was smoothly accomplished with the use of a Wittig phosphorane reagent (Scheme 3.45). The two exocyclic methylene protons of alkene 273 appeared as singlets at 5 = 4.86 and 6 = 4.79, and the carbinol proton appeared as a singlet at 6= 4.69 in the 1H NMR spectrum. The 1 3C NMR spectrum exhibited a carbonyl resonance at 8= 178.6, two alkenyl resonances at 8= 144.5 and 8= 108.7, and a carbinol resonance at 8 = 83.7. The absorption at 1770 cm"1 in the IR spectrum of 273 was attributed to the y-lactone carbonyl group. An HRMS measurement of the parent ion confirmed the molecular formula for 273. Scheme 3.45: Attempted preparation of allylic alcohol 274 With the structure of 273 firmly established, the stage was set for the proposed methanolysis of the lactone moiety under basic conditions. Compound 273 turned out to be remarkably stable under these conditions. This result was rather surprising since it was expected that the strain in the six-membered ring of alkene 273 would be similar to that of keto lactone 272 (see Figure 3.18), and would therefore drive the methanolysis of the lactone moiety. The failure to convert compound 273 to allylic alcohol 274 led us to abandon the approach to alkene 228 outlined in Scheme 3.40 and Scheme 3.42. At this stage, a different and apparently more expedient approach to the mulinane diterpenoids was devised. 3. Synthetic Approaches to (±)-lsomulir»ic Acid 106 3.4.4 Third synthetic approach to (±)-isomulinic acid 3.4.4.1 Retrosynthetic analysis The third synthetic approach to the mulinane diterpenoids was influenced by the results described in section 2.4.3.2 and by two reports in the literature which suggested a new and expedient strategy to establish the carbon chirality centers at C-8, C-9 and C-10 of the mulinane skeleton. It has been shown that although ketone 249 can be prepared efficiently, enolate 259 could not be formed selectively (Figure 3.19). It appeared likely that if a carbon substituent was present at C-9 (as in 275); it would be possible to selectively form an enolate such as 276 under thermodynamic conditions. 249 259 275 276 Figure 3.21: Structure of ketones 249 and 275 and of enolates 259 and 276 It was expected that a ketone such as 275 would be accessible through a directed hydrogenation of alcohol 277 (see Figure 3.22), followed by oxidation of the product. Alcohol 277, in turn, would be prepared from enone 278 as before (see Schemes 3.31 and 3.32). It was realized that this idea would not prove useful in preparing alkene 228 as outlined in Scheme 3.30 since the substituent at C-9 of 228 is a hydrogen atom. However, note that the carbonyl group of 275 activates C-9 and should allow for the inversion of configuration at that carbon to yield a system such as 279, in which the substituent at C-9 is equatorial. Bicycle 279 possesses four of the five carbon chirality centers found in all the mulinane natural products (see 157) with the correct relative configuration. 3. Synthetic Approaches to (±)-lsomulinic Acid 107 R R R 2) oxidation 1) directed hydrogenation 275 .0 epimerization 1 228 157 .0 279 Figure 3.22: Potential preparation of ketone 279 from enone 278 A report from the literature indicated that hydrogenation of an enone such as 278 under heterogeneous conditions would likely yield the corresponding frans-fused system (279) after equilibration at C-9 of ketone 275 (Scheme 3.46). According to the classical Horiuti-Polanyi mechanism for hydrogenation of olefins over metal catalysts, the two hydrogen atoms add to the same side of the double bond. McKenzie139 has pointed out that there is a stereoelectronic requirement for the two metal-carbon bonds in intermediates 280 and 281 to be eclipsed, which forces the six-membered ring into a boat conformation. Note that in intermediate 281, which leads to the c/'s-fused product (282), the boat forces the R group to clash with one of the methylene protons on the five-membered ring. If the R group is sufficiently large, this non-bonded interaction will destabilize the complex. As a result, intermediate 280, in which the boat forces the R group away from the five-membered ring, is favoured. The fact that a correlation exists between the size of the side chain (R) and the amount of frans-fused product observed in the heterogeneous hydrogenation of substrates similar to enone 278 supports McKenzie's proposal. 3. Synthetic Approaches to (±)- lsomulinic Acid 108 Pd, H2(g) 278 M = surface metal atom R = carbon M M 280 Scheme 3.46: Application of McKenzie's1 3 9 rationale for the heterogeneous hydrogenation of enones in a proposed synthesis of ketone 275 If the hydrogenation depicted above proved successful in establishing the correct relative configuration at C-9 and C-10, it would still be necessary to establish the quaternary center at C-8. The efficient and stereoselective Lewis acid-catalyzed rearrangement of epoxides to tertiary aldehydes has been demonstrated by Yamamoto and coworkers140 among others.141 The example depicted in Scheme 3.47 was particularly interesting since it indicated that this protocol might prove useful in establishing the carbon chirality center at C-8. Treatment of epoxide 283 with a stoichiometric amount of a bulky Lewis acid results in the formation of a tertiary aldehyde (284) in excellent yield. , 0 MABR OHC H 98% TBSO 283 TBSO' 284 MABR Scheme 3.47: Preparation of a tertiary (284) aldehyde from a trisubstituted epoxide (283) as demonstrated by Yamamoto and coworkers.1 4 0 3. Synthetic Approaches to (±)- lsomulinic Acid 109 It seemed likely that an epoxide such as 285 could be readily synthesized from ketone 279 using standard methods (Scheme 3.48). The relative configuration at C-8 of 285 was expected to be as shown based on steric arguments. The relative configuration at the other epoxide carbon (which corresponds to C-17 in the natural product) does not need to be defined at any stage. It has been established that, during the epoxide rearrangement, the migration of the alkyl group occurs anti to the oxygen atom (see inset, Scheme 3.48). Therefore, it was expected that if an aldehyde was obtained from the rearrangement of 285, the configuration at C-8 would be retained (as in 286). M = metal centered Lewis acid Scheme 3.48: Application of Yamamoto's epoxide rearrangement140 to the planned preparation of aldehyde 286 It was expected that if the proposed rearrangement of epoxide 285 to aldehyde 286 proved successful, this approach could be extended to the construction of the mulinane carbocyclic skeleton. At this stage, no specific plan had been devised for the construction of the seven-membered ring. However, it was postulated that a substrate such as 286 with appropriately functionalized groups at C-8 and C-9 would allow the use of a ring closing metathesis strategy. With this idea in mind, it was decided to install a (2-tert-butyldimethylsiloxy)ethyl group at C-9 since it would potentially allow access to two olefins (288 or 289) that could be used to construct the seven membered ring through a ring closing metathesis (Scheme 3.49). The nature of the group at C-8 of 287 was left undefined at this stage. This issue would be revisited if the 3. Synthetic Approaches to (±)- lsomulinic Acid 110 installation of the carbon chirality centers at C-8, C-9 and C-10 with the correct relative configuration could be accomplished as planned. Thus, for the time being, epoxide 290 became the synthetic target (see Scheme 3.49). OTBS 288 Scheme 3.49: Potential conversion of the (2-ferf-butyldimethylsiloxy)-ethyl group of 287 to an allyl group in 288 or a vinyl group in 289 3. Synthetic Approaches to (±)- lsomulinic Acid 111 3.4.4.2 Third synthetic approach to (+)-isomulinic acid The first step in this new synthetic strategy involved the preparation of enone 293. This operation was accomplished by forming the thermodynamic extended sodium enolate of alkene 197 in DME, and quenching it with an excess of iodide 292 1 4 2 (Scheme 3.50). Scheme 3.50: Synthesis of enone 293 The spectral data collected on enone 293 verified its structure. The incorporation of the side chain was evidenced by the appearance of two three-proton singlets at 5 = -0.02 and -0.04 and a nine-proton singlet at 5 = 0.83 in the 1H NMR spectrum, which correspond to the te/t-butyldimethylsiloxy moiety. In addition, a two-proton multiplet at 5 = 3.50-3.64 was assigned to the methylene group attached to oxygen. The 1 3C NMR spectrum showed the correct number of signals. A J-mod 1 3C NMR experiment demonstrated the presence of six methylene groups and a total of ten methyl and methine groups. Furthermore, the molecular formula of enone 293 was verified by an HRMS measurement of its parent ion. Access to enone 293 set the stage for the heterogenenous hydrogenation that was expected to provide the frans-fused ketone (294) with the correct relative configuration at C-10 (mulinane numbering, see Scheme 3.51). Overman143 has reported heterogeneous hydrogenation conditions which allow the in situ epimerization of the carbon chirality center a to the carbonyl group. Treatment of enone 293 with Pd(OH)2/C in the presence of DBU under hydrogen at one atmosphere provided a polar, UV-active material. This material was determined to be enone 295 on the basis 3. Synthetic Approaches to (±)-lsomulinic Acid 1 1 2 of 1H NMR and GC-LRMS data. Enone 293 could be obtained from 295 under standard protection conditions, thereby confirming the structure of the latter compound. Alternatively, treatment of enone 293 with a catalytic amount of Pd/C under hydrogen at one atmosphere provided two main products (Scheme 3.51). TLC analysis of the reaction mixture showed that these materials did not absorb ultraviolet light (X = 2 5 4 nm). Their 1H NMR spectra revealed that the terf-butyldimethylsilyl group had been removed under the hydrogenation conditions. Analysis of the products by GC-LRMS showed that their molecular mass is consistent with the structure of hydroxy ketones 296 or their corresponding hemiketals 297. Attempts to re-protect the hydroxyl group (as had been done with enone 295) failed, and thus indicated that hemiketals 297 had been formed. This type of hemiketal formation has been previously reported by Molander.144 The failure to protect the alcohol also indicated that no equilibrium exists between hydroxyketone 296 and hemiketal 297 under the conditions used for protection of the primary alcohol. 294 297 Scheme 3.51: Attempted hydrogenation of enone 293 to ketone 294 under heterogenenous conditions 3. Synthetic Approaches to (±)-lsomulinic Acid 113 It was hypothesized that the formation of hemiketals 297 occurred through the initial hydrogenation of the double bond, followed by deprotection of the alcohol and subsequent hemiketal formation. This suggested that the hydrogenation of 293 should be carried out using a large amount of palladium-on-carbon in order to increase the rate of the reaction. Indeed, this hydrogenation was complete in less than one hour and furnished three products (Scheme 3.52). GC-LRMS analysis of the mixture of products showed that their molecular masses were consistent with the structures of compounds 294, 298 and 299. According to the mechanistic rationale presented in Scheme 3.46, the hydrogenation of enone 293 should provide, initially, two diastereomeric ketones in which the newly incorporated hydrogen atoms have a cis relationship to each other (294 and 298 in Scheme 3.52). The c/'s-fused product (298) may adopt a conformation in which the alkyl group at C-5 resides in the equatorial position. It is conceivable that ketone 298 would isomerize to its C-5 epimer (300) under thermodynamic conditions. Ketone 300 can also adopt a conformation in which the C-5 alkyl group lies in the equatorial position. 3. Synthetic Approaches to (±)-lsomulinic Acid 114 H2(g),l0% Pd/C EtOH, < 1 h V 300 299 Scheme 3.52: Hydrogenation of enone 293, formation of ketones 294, 298 and 299 The initially formed frans-fused product (294) is conformationally locked, and therefore, the C-5 alkyl group resides in the axial position. The severe 1,3-diaxial steric interaction between the alkyl group and the methoxycarbonyl present in ketone 294 makes it less stable relative to its C-5 epimer (299). In theory, this should allow for the complete conversion of ketone 294 to ketone 299 under equilibrating conditions. Subjection of a mixture of ketones 294, 298 and 299 to epimerization conditions resulted in the complete conversion of ketone 294 to ketone 299. Ketone 298 remained unchanged. 3. Synthetic Approaches to (±)-lsomuliriic Acid 115 294 + 298 + 299 NaOMe MeOH ,0 OTBS Scheme 3.53: Equilibration of ketones 294, 298 and 299 The IR spectrum of ketone 299 displayed two strong carbonyl stretching absorptions at 1712 and 1724 cm"1. The mass spectrum did not display an M + ion, however, an HRMS measurement of the [M-f-Bu]+ ion verified the molecular formula. The 1H NMR and 1 3 C NMR spectra displayed the correct number of signals. A J-mod 1 3 C NMR spectrum verified the presence of six methylene groups and a total of twelve methyl and methine groups. The results obtained from the hydrogenation and epimerization reactions shown in Schemes 3.52 and 3.53 demonstrated that ketone 299 was indeed the desired material. No further efforts to ascertain the relative configuration at C-5 were made. However, comparison of the 1H NMR data obtained for ketone 299 with that obtained for ketone 249, particularly the signal corresponding to the equatorial proton at C-2, indicated that the configuration at C-6 was as desired (see Figure 3.23). H A : 8 = 2.75-2.83 (ddd, J= 13.1, 6.8, 1.8 Hz) 249 OTBS h A : 8 = 2.73-2.83 (ddd, J= 13.1, 6.7, 2.1 Hz) 299 Figure 3.23: Comparison of H NMR data for ketone 249 with 1H NMR data for ketone 299 3. Synthetic Approaches to (±)-lsomulinic Acid 116 Unfortunately, attempts to carry out the hydrogenation of enone 293 in large scale always resulted in the formation of undesired sideproducts (297, see Scheme 3.51). Rather than trying to optimize the hydrogenation process at this stage, it was decided to proceed with the synthetic plan using the amount of ketone 299 acquired thus far. The next step in the sequence involved the preparation of alkenes 300 and 301 from ketone 299 through the use of a Wittig protocol. In the event, an 11:1 mixture of alkenes 300 and 301 were obtained in 78% overall yield. The geometry about the double bond in the major product was determined to be as shown using a number of 1H NMR experiments on a derivative of 300 (see Tables 4.18 and 4.19 in the experimental section). OTBS OTBS OTBS Scheme 3.54: Preparation of alkenes 300 and 301 Unfortunately, the acquired mixture of alkenes proved inseparable by flash column chromatography on silica gel, and this prevented their full characterization. However, the 1H NMR spectrum obtained on the mixture did show the expected signals for the two products (Figure 3.24). 3. Synthetic Approaches to (±)-lsomulinic Acid 117 OTBS HA:8.= 5.17(q; J = .6,8Hz) Me A : 8 = 1.46 (d; J=6.8 Hz) Me B : 8 = 3.53 (s) H A:S = 5.11 (q; J = 6.7Hz) Me A :8 = 1.45 (d; J=6.7 Hz) Me B : 5 = 3.60 (s) Figure 3.24: Selected 1H NMR data for alkenes 300 and 301 The synthetic plan called for the treatment of alkenes 300 and 301 with a suitable oxidant in order to establish the epoxide moiety. Based on the calculated structure of the transition state for the oxidation of alkenes by DMDO, 1 4 5 , 1 4 6 the epoxidation was expected to occur through the approach of the oxidant from the p-face of the alkene, as shown in Figure 3.25. Indeed, Johnson et al.U7 have stated that during the oxidation of methylenecyclohexanes "the effect of steric hydrance favours, obviously, the equatorial attack." At the time, it was not conceived that the planned epoxidation of alkene 300 would be subject to any stereoelectronic effects.147 3. Synthetic Approaches to (±)- lsomulinic Acid 118 Q ^ X > TBSO ^ (3-approach (equatorial) a-approach (axial) 300 Figure 3 . 2 5 : Two possible approaches to the double bond of alkene 300 by an oxidant and expected shape of the transition state for the epoxidation of methylenecyclohexane by DMDO under steric control The dimethyldioxirane (DMDO) mediated epoxidation of alkenes 300 and 301 furnished a mixture of products which were easily separated by flash column chromatography on silica gel (Scheme 3.55). The structure of the major product was determined to be as shown in 3 0 2 (see Tables 4 .18 and 4.19 in the experimental section for details). No effort was made to isolate or characterize the minor product. OTBS OTBS QTBS Scheme 3 . 5 5 : Epoxidation of alkenes 300 and 301 The 1H NMR spectrum of epoxide 302 exhibited a three-proton doublet at 8 = 1.08 and a one-proton quartet at 8 = 2.67-2.72 with a mutual coupling constant of 5.4 Hz. These signals could be assigned to the proton and the methyl group on the epoxide ring. The 1 3C NMR spectrum displayed a quaternary carbon signal at 8 = 63.4 and a methine carbon signal at 8 = 61.8, which correspond to the two epoxide carbons. The molecular formula of epoxide 302 was verified through an HRMS measurement of its parent ion. 3. Synthetic Approaches to (±)- lsomulinic Acid 119 A full suite of NMR experiments were performed on a sample of epoxide 302 and allowed the complete assignment of its structure (see Tables 4.18 and 4.19 in the experimental section). A number of 1D selective NOE difference experiments established the configuration of the carbon chirality centers at C-4 and C-11 (Figure 3.26). These experiments also served to indirectly determine the geometry of the double bond in alkene 300. OTBS 302 H A: 5 = 1.45-1.52 (dd; 13.7,10.3 Hz) H B: 5 = 2.70-2.79 (ddd; 13.7, 9.3, 9.3 Hz) H c : 5 = 0.90-0.98 (dd; 13.4, 9.3 Hz) H D : 8 = 2.67-2.72 (ddd; 13.4, 10.3, 9.3 Hz) H E : 5 = 2.14-2.22 (ddd; 5.4, 5.4, 5.4 Hz) MeA: 8 = 1.08 (d; 5.4 Hz) MeA: 8 = 3.65 (s) Irradiation of H B enhances H A and H c Irradiation of H D enhances H c and H A Irradiation of H E enhances H c Irradiation of Me A enhances Me B and H E Figure 3.26: Selected NOE difference data for epoxide 302 Note that epoxide 302 does not possess the desired relative configuration at C-4. This meant that the synthetic plan outlined in section 3.4.4.1 would not yield access to the mulinane family of diterpenoids. However, if treatment of epoxide 302 with MABR promoted the migration of the methyl group to C-4, aldehyde 303 would be obtained (Scheme 3.60). Note that aldehyde 303 has the correct relative configuration at all the carbon chirality centers present in 157. Furthermore, the aldehyde moiety could potentially be used to install an alkenyl chain which would participate in a ring-closing metathesis reaction to establish the seven-membered ring. 3. Synthetic Approaches to (±)- lsomulinic Acid 120 OTBS OTBS 302 MABR DCM -78 °C alkyl migration TBSO ^ - O H O r > O M e 303 CrOMe 304 Scheme 3.60: Lewis acid mediated rearrangement of epoxide 302 Unfortunately, treatment of epoxide 302 with a stoichiometric amount of MABR 1 4 0 resulted in the formation of methyl ketone 304 in excellent yield (Scheme 3.60). The 1H NMR spectrum of 304 displayed a three-proton singlet at 5= 2.07 that was attributed to the methyl ketone group. In addition, a signal at 5 = 2.24-2.34 (ddd, 1 H, J= 12.6, 11.0, 4.3 Hz) was assigned to the axial proton at C-4. The ketone carbonyl carbon appeared at 8= 212.5 in the 1 3C NMR spectrum. The IR spectrum clearly displayed two carbonyl stretch absorptions at 1724 and 1712 cm"1, which correspond to the ester and the ketone groups, respectively. Finally, the molecular formula for ketone 304 was verified through an HRMS measurement of its parent ion. Based on significant literature precedent, the rearrangement of epoxide 302 with migration of the hydride was expected to provide ketone 305. However, ketone 304 was observed as the single product of this reaction. This indicated that the Lewis acid 3. Synthetic Approaches to (±)-lsomulinic Acid 121 also catalyzed the epimerization of the initially formed ketone (see Scheme 3.61). In order to confirm that the acetyl group resides in the equatorial position, a solution of 304 in NaOMe/MeOH was allowed to stir for two days at room temperature. The product isolated from this reaction was identical to the starting material. Scheme 3.61: Pathway for the MABR catalyzed rearrangement of epoxide 302 to ketone 304 Clearly, this synthetic plan would not provide access to the mulinane family of diterpenoids. However, it seemed likely that, after some modifications, the heterogeneous hydrogenation of enone 293 or a similar substance would provide efficient and expedient access to a bicyclic system with the correct relative configuration at four of the five carbon chirality centers found in the mulinane carbocyclic system. Thus, a new synthetic strategy that incorporated the heterogeneous hydrogenation of an a-alkyl enone was devised. 3. Synthetic Approaches to (±)- lsomulinic Acid 122 3.4.5 Fourth synthetic approach to (+)-isomulinic acid 3.4.5.1 Retrosynthetic analysis The work described in section 3.4.4.2 demonstrated that the hydrindane system embedded in the mulinane diterpenoids (see 157, Figure 3.27), with the correct relative configuration at C-3, C-5, C-9 and C-10, could be obtained in an expedient fashion from enone 293. This process involved the hydrogenation of enone 293 under heterogeneous conditions to provide ketone 294. This material, in turn, allowed access to ketone 299 after epimerization of the carbon chirality center at C-9. 157 293 294 299 Figure 3.27: Structures of mulin-11,13-dien-20-oic acid (157), enone 293 and ketones 294 and 299 Unfortunately, the hydrogenation process could not be scaled up without the generation of undesired sideproducts. It appeared likely that replacing the tert-butyldimethylsilyl group of enone 293 with a more stable protecting group would allow for the efficient generation of a bicyclic system analogous to 299 in large scale. Therefore, it was decided to protect the primary alcohol as a methyl ether (as in 308, Scheme 3.62) since this group is stable under a variety of conditions. This would allow access to a system such as 309 using the same hydrogenation and epimerization sequence. The installation of the methyl group at C-8 was envisioned to occur through the alkylation of aldehyde 310 or a derivative thereof. The relative configuration of the 3. Synthetic Approaches to (±)-lsomulinic Acid 123 carbon chirality center at C-8 was expected to be as shown in 311 based on significant literature precedent. Aldehyde 310 would be prepared from ketone 309 using standard methods. Scheme 3.62: Retrosynthesis of aldehyde 311 to enone 308 It appeared likely that the aldehyde group at C-8 and the 2-methoxyethyl group at C-9 of aldehyde 311 could be transformed into two alkenyl chains that would participate in a ring-closing metathesis reaction. Therefore, the retrosynthetic analysis for the construction of compound 157 from aldehyde 311 was based on the use of a ring-closing metathesis as a key step (Scheme 3.63). Compound 157 should be accessible from diene 312 upon conversion of the ester group to the corresponding acid. Dehydration of allylic alcohol 313 would furnish diene 312. Allylic alcohol 313 was expected to be the major product of the crucial ring-closing metathesis reaction utilizing diene 314 as the substrate. Compound 314 would be accessed upon alkylation of aldehyde 315 with a suitable Grignard reagent. Note that the relative configuration of the carbon chirality center at C-12 of allylic alcohol 314 does not need to be defined at any stage since both isomers should eventually lead to diene 312. The oxidation of alcohol 316 to the corresponding aldehyde was expected to occur under 3. Synthetic Approaches to (±)-lsomulinic Acid 124 standard conditions. It was envisioned that the conversion of methyl ether 317 to primary alcohol 316 would take place under the action of B B r 3 1 4 8 , 1 4 9 or TMSI . 1 5 0 The homologation of aldehyde 311 to aldehyde 318, and the subsequent conversion of aldehyde 318 to alkene 317 was expected to occur under standard conditions. 317 318 311 Scheme 3.63: Retrosynthesis of mulin-11,13-dien-20-oic acid (157) to aldehyde 311 3, Synthetic Approaches to (±)- lsomulinic Acid 1 2 5 3.4.5.2 Fourth synthetic approach to (±)- isomulinic acid The first step in the present strategy involved the installation of the 2-methoxyethyl group at C-2 of enone 197. This operation was accomplished using conditions similar to those used to prepare enone 293 (see Scheme 3.64 and Scheme 3.50). Thus, treatment of enone 197 with NaH in DME provided the thermodynamic sodium enolate. The alkylation of this enolate with 2-bromoethyl methyl ether (319) occurred at a slow rate and in good yield. Scheme 3.64: Synthesis of enone 308 The IR spectrum of enone 308 displayed two carbonyl stretching absorptions at 1665 and 1725 cm"1, which were attributed to the a,p-unsaturated ketone and the methyl ester group, respectively. The 1H NMR spectrum displayed the correct number of signals, including a three-proton singlet at 5 = 3.24, which was attributed to the newly introduced methyl ether. In addition, a two-proton multiplet at 5= 3.25-3.32 was attributed to the methylene group adjacent to the ether oxygen atom. The 1 3C NMR spectrum displayed the correct number of signals, including two alkenyl carbon signals at 8 = 129.9 and 8 = 167.9. A J-mod 1 3C NMR experiment demonstrated the presence of six methylene groups and a total of six methyl and methine groups. The molecular formula of enone 308 was verified by an HRMS measurement of its parent ion. Access to enone 308, set the stage for the proposed heterogeneous hydrogenation reaction that was expected to establish the correct relative configuration of the carbon chirality center at C-6 of ketone 308. Treatment of enone 308 with a catalytic amount of 10% palladium-on-carbon under hydrogen at one atmosphere 3. Synthetic Approaches to (±)-lsomulinic Acid 126 provided a mixture of three products. At this stage the nature of the products was not determined, but was hypothesized to be as shown in Scheme 3.65 based on the mechanistic rationale outlined in Scheme 3.46 and the results obtained for the homogeneous hydrogenation of enone 293 (see Scheme 3.52). 308 320 309 Scheme 3.65: Heterogeneous hydrogenation of enone 308. Treatment of the mixture of compounds 320, 309 and 321 with HCI in MeOH resulted in the convergence of compounds 320 and 309 into 309. Compound 321 remained unchanged (Scheme 3.66). At this stage compounds 309 and 321 were easily separated using flash column chromatography on silica gel. 320 + 309 + 321 78% (2 steps) 17% (2 steps) Scheme 3.66: Equilibration of a mixture of compounds 320, 309 and 321 The spectroscopic data collected on compound 309 was consistent with the assigned structure. The IR spectrum showed that the two carbonyl stretching absorptions coincided at 1723 cm"1. The 1H NMR spectrum displayed the correct number of signals. A J-mod 1 3C NMR spectrum demonstrated the presence of six 3. Synthetic Approaches to (±)-lsomulinic Acid 127 methylene groups and a total of eight methyl and methine groups. In addition, its molecular formula was verified through the HRMS measurement of its parent ion. The relative configuration of the newly formed carbon chirality centers at C-5 and C-6 of ketone 309 was not determined spectroscopically at this stage. However, comparison of the 1H NMR data collected on ketone 309 with that collected for ketones 249 and 299 indicated that the relative configuration at C-5 was as desired (Figure 3.28). H A : 5 = 2.75-2.83 (ddd, J = 13.1, 6.8, 1.8 Hz) 249 OTBS h A : 8 = 2.73-2.83 (ddd, J = 13.1, 6.7, 2.1 Hz) 299 H A : 6 = 2.74-2.83 (ddd, J= 13.0, 6.7, 2.2 Hz) 309 Figure 3.28: Comparison of 1H NMR data for ketones 249 and 299 with 1H NMR data for ketone 309 3. Synthetic Approaches to (±)-lsomulinic Acid 128 The structure of compound 321 was assigned to be as shown in Scheme 3.66 based on the spectroscopic data collected. The IR spectrum showed a carbonyl stretching absorption at 1723 cm"1. The 1H NMR spectrum displayed the correct number of signals. Taken together, these data cannot be used to differentiate between compound 321 or the c/s-fused ketone 322 (see Figure 3.29). However, an HRMS measurement of the parent ion of compound 321 indicated its molecular formula to be as shown in Figure 3.29. The 1 3C NMR spectrum displayed two resonances at 5 = 139.0 and 5 = 126.7, which indicated the presence of a carbon-carbon double bond in the molecule. No resonances corresponding to a ketone carbonyl were observed. Furthermore, a J-mod 1 3C NMR experiment showed that compound 321 possesses seven methylene groups and a total of six methyl and methine groups. Figure 3.29: Selected structural data expected for ketone 322 and alkene 321 3. Synthetic Approaches to (±)- lsomulinic Acid 129 The generation of alkene 321 as the major side-product from the heterogeneous hydrogenation of enone 308 requires some comment. A proposed pathway for this transformation is presented in Scheme 3.67. The reduction of cyclic ketones to secondary alcohols is known to proceed under heterogeneous conditions using dispersed metals, including palladium, as catalysts.151 Therefore, it is possible that some of enone 308 would provide allylic alcohol 323 during the hydrogenation reaction. Alcohol 323 is prone to dehydration to provide diene 324. Finally, the hydrogenation of the carbon-carbon double bond between C-3 and C-4 of diene 324 would furnish alkene 321. Although conceivable, it is unlikely that alkene 321 would arise through the direct hydrogenolysis of alcohol 323 since forcing conditions (high pressure and temperature) are usually required for this process.152 321 Scheme 3.67: Proposed mechanism for the generation of alkene 321 from enone 308 under heterogeneous hydrogenation conditions. 3. Synthetic Approaches to (±)-lsomulinic Acid 130 The synthetic plan called for the conversion of ketone 309 to aldehyde 310. This transformation was efficiently accomplished in two steps (Scheme 3.68). Ketone 309 was converted to a mixture of spiroepoxides 324 and 325 using the method developed by Corey and Chaykovsky153"155 (see inset, Scheme 3.68) and the conditions reported by Danishefsky.156 These epoxides proved inseparable by flash column chromatography on silica gel. The mixture of epoxides was treated with a Lewis acid (MABR, see Scheme 3.47), which promoted the rearrangement of the epoxide moiety to the corresponding aldehyde.140 The mechanism of this rearrangement is similar to that previously discussed in Scheme 3.48. Aldehydes 309 and 327 were also inseparable by flash column chromatography on silica gel, and were thus characterized as a mixture. 310 327 Scheme 3.68: Preparation of aldehydes 310 and 327 from ketone 309 3. Synthetic Approaches to (±)-lsomulinic Acid 131 Aldehydes 310 and 327 exist as a 3:1 mixture of C-4 epimers in HCI/CDCI3. Although no efforts were made to ascertain the relative configuration of the carbon chirality center at C-4 of the two aldehydes, it is reasonable to expect the aldehyde group of the major isomer to reside in the equatorial position (see Figure 3.30). The IR spectrum of the mixture exhibited two carbonyl stretching absorptions at 1721 and 1704 cm"1. The molecular formula was confirmed through an HRMS measurement of the parent ion. Both the 1H NMR and 1 3C NMR spectra showed resonances diagnostic of the aldehyde group (see Figure 3.30). CHO 310 H 327 major (equatorial) epimer minor (axial) epimer 13, 'C NMR: 5 = 205.6 H NMR: 5 = 9.85 (s) 13, 'C NMR: 5 = 204.5 H NMR: 8 = 9.47 (d, J=3.9 Hz) Figure 3.30: Selected spectral data for aldehydes 310 and 327 3. Synthetic Approaches to (±)-lsomulinic Acid 132 With aldehydes 309 and 327 in hand, the stage was set for the installation of the methyl group at C-4. Based on significant literature precedent,120 the relative configuration of the new carbon chirality center was expected to be as desired. Ireland and Mander157 have shown that the methylation of the potassium enolate of aldehyde 328 provides the axial aldehyde 329 as the sole product (Scheme 3.69). This high degree of diastereoselectivity has been attributed to the fact that the approach of the electrophile (Mel) to the a-face of the enolate (axial approach) is blocked by the two axial protons H A and H B (see 330, Scheme 3.69). On the other hand, approach of the electrophile to the (3-face of the enolate (equatorial approach) is unhindered.158 a-approach (axial) Scheme 3.69: equatorial alkylation of aldehyde 328 as shown by Ireland and Mander in a total synthesis of (± ) - r imuene 1 5 7 3. Synthetic Approaches to (±)- lsomulinic Acid 133 It must be noted that equatorial alkylation of exocyclic enolates is not always observed. Snider et a/.159 have shown that the alkylation of the potassium enolate of aldehyde 331, generated using the same conditions used by Ireland and Mander, provided the equatorial aldehyde (332) as the major product (Scheme 3.70). In this example, the approach of the electrophile from the a-face of the enolate (axial approach) is only hindered by one axial proton (HA) on the six-membered ring (see 334). Furthermore, the approach of the electrophile from the p-face of the enolate (equatorial approach) is likely hindered by the axial isopentenyl chain. 334 a-approach (axial) Scheme 3.70: Axial alkylation of aldehyde 331 as shown by Snider etal. in a total synthesis of (+)-erinacine159 3. Synthetic Approaches to (±)-lsomulinic Acid 134 Based on these and other results from the literature, it was expected that the alkylation of aldehydes 310 and 327 would provide the axial aldehyde, thereby establishing the correct relative configuration at the newly generated carbon chirality center. Indeed, examination of molecular models of the enolates derived from aldehydes 310 and 327 indicated that the two axial protons on the a-face of the molecule (HA and HB) should hinder the axial approach of the electrophile (see 335, Scheme 3.71). The side chain on C-5, which resides in the equatorial position, was not expected to exert any influence on the approach of the electrophile. Thus, the axial aldehyde (311) was expected to be the major product from this alkylation reaction. base deprotonation 1 p-approach (equatorial) OMe Mel OMe alkylation H CHO O K 335 311 a-approach (axial) Scheme 3.71: Expected alkylation of aldehyde 310 3. Synthetic Approaches to (±)- lsomulinic Acid 135 Subjection of aldehydes 310 and 327 to the alkylation conditions reported by Ireland and Mander provided a mixture of products. 1H NMR analysis of the crude product mixture revealed the presence of two methyl enol ethers and two new aldehydes (Scheme 3.72). OMe OMe equatorial axial alkylation alkylation Scheme 3.72: Attempted alkylation of aldehydes 310 and 327 under the reaction conditions reported by Ireland and Mander The mixture of products obtained from the alkylation reaction of aldehyde 310 proved inseparable by flash column chromatography on silica gel. The methyl enol ethers produced were evidenced by two singlets at 8 = 5.68 and 8 = 5.63 (60%, ratio 3.6:1) in the 1H NMR spectrum of the mixture. The two new aldehydes gave rise to singlets at 8= 9.58 and 8= 9.24 (32%, ratio 7.4:1). The signal at 8= 9.58 was attributed to the desired product (311) based on the mechanistic arguments presented in the preceding discussion. The spectroscopic verification of the relative configuration of the newly generated carbon chirality center was carried out at a later stage on a material derived from aldehyde 311. 3. Synthetic Approaches to (±)- lsomulinic Acid 136 Unfortunately, all attempts to alkylate aldehydes 310 and 327 under a variety of conditions were plagued by the generation of methyl enol ethers as the major products. Similarly, attempts to alkylate hydrazone and imine derivatives of aldehydes 310 and 327 were not successful. In view of these results, a four step sequence was used to install the methyl group at the C-4 position. This method relied on the known Bronsted acid catalyzed rearrangement of cyclopropanols to methyl aldehydes160"165 (Scheme 3.73). + Scheme 3.73: Bronsted acid catalyzed rearrangement of cycropropanols to a-methyl aldehydes The first step in this alkylation sequence was the conversion of aldehydes 310 and 327 to a mixture of silyl enol ethers 342 and 343 (Scheme 3.40). These silyl enol ethers were easily separable by flash column chromatography on silica gel. Although these materials could be characterized independently of each other, no effort was made to ascertain the geometry about the double bond of the major product. 310: = H, R 2 = CHO 327: R 1 = CHO, R 2 = H 1) KHMDS, THF -78 to 0 °C 2) TBSCI THF, 0 °C 75% OTBS Scheme 3.74: Preparation of silyl enol ethers 342 and 343 3. Synthetic Approaches to (±)-lsomuHnic Acid 137 The 1H NMR spectrum of the major silyl enol ether displayed a one-proton singlet at 8 = 6.00 which was attributed to the alkenyl proton. The terf-butyldimethyl silyl group was evidenced by the appearance of a nine-proton singlet at 8 = 0.88 and two three proton singlets which coincided at 8= 0.07. The 1 3C NMR spectrum displayed two alkenyl carbon signals at 8= 121.6 and 8= 132.7. The IR spectrum of this material showed a carbonyl stretching absorption at 1724 cm"1 and an absorption at 1666 cm"1 which was attributed to the stretching of the carbon-carbon double bond. The molecular mass of the major silyl enol ether was confirmed by an HRMS measurement of its parent ion. The 1H NMR spectrum of the minor silyl enol ether showed a one-proton singlet at 8 = 5.97 which was attributed to the alkenyl proton. The te/t-butyldimethyl silyl group was evidenced by the appearance of a nine-proton singlet at 8 = 0.90 and two three proton singlets which coincided at 8 = 0.08. The 1 3C NMR spectrum displayed two alkenyl carbon signals at 8= 119.2 and 8= 134.0. No IR or mass spectral data was recorded on this material. At this point a Simmons-Smith166"168 reaction was utilized to prepare cyclopropanes 344 and 345 from the mixture of silyl enol ethers (342 and 343, Scheme 3.75). The resulting mixture of cyclopropanes proved inseparable by flash column chromatography on silica gel. Analysis of molecular models of compounds 342 and 343 indicated that the cyclopropanation reaction should occur predominantly from the p-face of the double bond (see inset, Scheme 3.75). Therefore, it was expected that the relative configuration of the carbon chirality center at C-4 of compounds 344 and 345 would be as shown. 3. Synthetic Approaches to (±)-lsoroulinic Acid 138 Et2Zn, CH 2 I 2 Et 20, r. t. 63% (brsm) 342: R, = H, R 2 = OTBS 343: RT = OTBS, R 2 = H 344: R, = H, R 2 = OTBS 345: R, = OTBS, R 2 = H p-approach cyclopropanation OMe Scheme 3.75: Simmons-Smith cyclopropanation of silyl enol ethers 342 and 343 Although the mixture of cyclopropyl silyl ethers 344 and 345 was inseparable, the major product could be independently prepared form the major enol ether. The 1H NMR spectrum of the major product of the cyclopropanation reaction showed clear evidence for the presence of the cyclopropane ring (see Figure 3.31). The 1 3C NMR spectrum displayed the correct number of signals. A J-mod 1 3C NMR experiment demonstrated the presence of seven methylene groups and a total of fourteen methyl and methine groups. The molecular mass of the major cyclopropyl silyl ether was verified through an HRMS measurement of its parent ion. Figure 3.31: Selected NMR data for the major cyclopropyl enol ether obtained from the Simmons-Smith reaction of enol ethers 342 and 343 H A : 5 = 0.11-0.15 (dd, J = 6.2, 3.4 Hz) H B : 5 = 0.63-0.80 (dd, J= 6.7, 6.2 Hz) H c : 5 = 3.21-3.25 (dd, J = 6.7, 3.4 Hz) 3. Synthetic Approaches to (±)-lsomulinic Acid 139 The mixture of cyclopropyl enol ethers could be efficiently converted to the corresponding cyclopropanols (346 and 347) upon deprotection with TBAF (Scheme 3.76). The cyclopropanols were not characterized, but were immediately converted to aldehyde 311 by treatment with mineral acid in hot THF. OMe 344: = H, R 2 = OTBS 346: R, = H, R 2 = OH 311 345: RT = OTBS, R 2 = H 347: R, = OH, R 2 = H Scheme 3.76: Preparation of aldehyde 311 The spectral data collected on aldehyde 311 was consistent with the assigned structure. The 1H NMR spectrum contained the correct number of signals, including a one-proton singlet at 8 = 9.58 which was attributed to the aldehyde group and a three-proton singlet at 5 = 1.04 which accounted for the methyl group a to the aldehyde. Note that the 1H NMR spectrum of the major a-methyl aldehyde obtained using the reaction conditions described by Ireland and Mander also displayed a singlet at 8 = 9.58. The 1 3C NMR spectrum displayed the correct number of signals, including a resonance at 8 = 206.7 which was attributed to the aldehyde carbonyl carbon. A J-mod 1 3C NMR spectrum demonstrated the presence of six methylene groups and a total of ten methyl and methine groups. The IR spectrum showed a carbonyl stretching absorption at 1721 cm"1. Finally, the molecular mass was confirmed through an HRMS of the parent ion. 3. Synthetic Approaches to (±)-lsomulinic Acid 1 4 0 The homologation of aldehyde 311 to aldehyde 318 was efficiently accomplished using the same sequence of reactions used to prepare aldehyde 310 (Scheme 3.77). Thus, a mixture of epoxides 348 and 349 was prepared from aldehyde 311 using the Corey-Chaikovsky epoxidation protocol153"156 (see Scheme 3.68). This mixture of epoxides converged to aldehyde 318 upon treatment with MABR 1 4 0 (see Scheme 3.47). O Me + / JJ Me—s' 1 1 \ CH, THF, 0 °C 318 Scheme 3.77: Preparation of aldehyde 318 The IR spectrum of aldehyde 317 exhibited a carbonyl stretching absorption at 1720 cm"1. The 1H NMR spectrum contained the correct number of signals. The aldehyde proton resonated at 8= 9.81 (dd, J= 3.9, 2.7 Hz). The 1 3C NMR spectrum displayed the correct number of signals, including a resonance at 8 = 204.0 which was attributed to the aldehyde carbonyl carbon. A J-mod 1 3C NMR spectrum demonstrated the presence of seven methylene groups and a total of ten methyl and methine groups. 3. Synthetic Approaches to (±)-lsomulinic Acid 141 The synthetic plan called for the conversion of aldehyde 318 to alkene 317. The desired transformation could be achieved using a Wittig or Takai8 2"8 4'8 7 olefination protocol. However, the best yield of alkene 317 was obtained using Tebbe's methylenation reagent81 (146, Scheme 3.78). Scheme 3.78: Preparation of alkene 317 The 1H NMR spectrum of alkene 317 displayed a one-proton multiplet at 8 = 5.66-5.79 (dddd, J = 17.0, 10.2, 8.0, 6.9 Hz) and a two-proton multiplet at 8 = 4.92-5.02. These signals were attributed to the three olefinic protons in the molecule. The 1 3C NMR spectrum displayed the correct number of signals, including two olefinic carbon resonances at 8 = 116.9 and 8 = 135.5. A J-mod 1 3C NMR experiment showed clearly the presence of eight methylene groups and a total of ten methyl and methine groups. An HRMS of the parent ion confirmed the molecular mass of alkene 317. Having efficiently installed an allyl group at C-4, it became necessary to transform the 2-methoxy ethyl group at C-5 into an alkenyl chain that would participate in a ring-closing metathesis reaction to construct the seven-membered ring. This required the conversion of methyl ether 317 to the corresponding primary alcohol. This transformation is typically carried out using a Lewis acid, BBr3 being the most widely used. 3. Synthetic Approaches to (±)-lsomulinic Acid 142 Grieco148 has shown that the conditions used for the deprotection of methyl ethers of highly functionalized substrates can often result in the promotion of side reactions. For example, treatment of compound 350 with BBr3 in DCM at low temperature provides homoallylic bromide 351 as the major product (Scheme 3.79). This reaction is thought to occur through the initial activation of the ether group by coordination of the ether oxygen atom to the Lewis acid, as shown in 352. The vinyl group attacks the activated methylene group and forms a carbocationic species (see 353 and 354). Nucleophilic attack of carbocation 354 by bromide results in the formation of the observed product. Scheme 3.79: Formation of homoallylic bromide 351 during the attempted deprotection of methyl ether 350 using BBr 3 1 4 8 3. Synthetic Approaches to (±)-lsomulinic Acid 143 The attempted deprotection of a related substrate used in the same study led to the formation of a different side product149 (Scheme 3.80). Treatment of 355 with BBr3 in DCM resulted in the formation of two main products (358 and 360). Deprotection of the methyl ether proceeds as expected to generate a primary alkoxide bound to the Lewis acid (357). This material cyclizes onto the adjacent methyl ester to form lactone 358, which was the desired product. Unfortunately, the Lewis acid also activates one of the acetate groups to form the stabilized oxonium ion 359. Attack of 359 by bromide results in the formation of 360. Scheme 3.80: Formation of lactone 358 and bromide 356 during the BBr3 mediated deprotection of methyl ether 360 1 4 9 3. Synthetic Approaches to (±)- lsomulinic Acid 144 The relative scarcity of functional groups in methyl ether 317 indicated that the only possible side reaction would involve the attack of the alkene double bond onto the methylene group of the activated ether as shown in 361 (Scheme 3.81). Thus, in addition to the desired primary alcohol, side products arising from cation 362 were also expected. Scheme 3.81: Potential side reaction during the BBr3 mediated deprotection of methyl ether 317 Remarkably, methyl ether 317 proved stable to the deprotection conditions described by Grieco, even after an extended period of time. As a result, the transformation of the methyl ether to a primary alcohol was attempted using TMSI as the Lewis acid. According to Jung,150 the deprotection of methyl ethers with TMSI occurs as shown in Scheme 3.82. The methyl ether (363) and TMSI exist in a fast equilibrium with the activated ether (364). Presumably the conversion of the activated ether (364) to a trimethyl silyl ether (365) occurs through an SN2 mechanism. The trimethyl silyl ether is converted to the corresponding alcohol (366) during an aqueous work-up. However, in the presence of excess TMSI the trimethyl silyl ether provides the corresponding alkyl iodide (368). 3. Synthetic Approaches to (±)- lsomulinic Acid 145 Me fast \ | SiMe 3 l+ 3 Me SiMe 3 I JO .OH Me 3Sil slow S N 2 H ,0 Mel 363 364 Me 3Sil 365 366 M e 3 S i \ ^ S i M e 3 + SiMe 3 l+ 3 .0^ ^ S i M e 3 368 367 Scheme 3.82: Mechanism for the conversion of alkyl methyl ethers to alcohols and alkyl iodides as proposed by Jung 1 5 0 Unfortunately, treatment of alkene 317 with a large excess of TMSI (~ 10 equiv.) did not provide either the primary alcohol or the corresponding alkyl iodide (Scheme 3.83). Analysis of the 1H NMR spectrum of the crude product mixture indicated that the two major products were cyclic ethers 372 and 373, which are epimeric at C-14 (mulinane numbering). 3. Synthetic Approaches to (±)-lsomulinic Acid 146 Scheme 3.83: Attempted deprotection of methyl ether 317 with an excess of TMSI It is likely that the formation of ethers 372 and 373 occurred through the intermediacy of the desired trimethyl silyl ether 369. The double bond of 369 may react with adventitious HI to generate a secondary carbocation (370). The juxtaposition of the trimethyl silyl ether and the secondary carbocation facilitates the formation of the oxonium ion 371. This material provides the observed ethers upon elimination of TMSI. The 1H NMR spectrum of the major ether showed three one-proton multiplets at 5= 3.44-3.48, 5= 3.63-3.68 and 8 = 4.20-2.26, which account for all the protons adjacent to the ether oxygen atom. A three-proton doublet at 8 = 1.95 (J= 6.8 Hz) was attributed to the methyl group a to the ether oxygen atom. The 1 3C NMR spectrum displayed the correct number of signals. A J-mod 1 3C NMR experiment demonstrated the presence of seven methylene groups and a total of ten methyl and methine groups. The relative configuration of the carbon chirality center at C-14 (mulinane numbering) was not determined. 3. Synthetic Approaches to (±)-lsomulinic Acid 147 Due to the failure to deprotect methyl ether 317, it was decided to revise the synthetic strategy. According to the mechanistic rationale presented by Jung,1 5 0 the TMSI mediated deprotection of methyl ether 311 should provide either alcohol 374 or alkyl iodide 375 (Scheme 3.84). It was recognized that primary alcohol 374 could potentially be converted to the corresponding ortfro-nitrophenyl selenide (376) using the procedure developed by Grieco169 (Scheme 3.85). This material, in turn, should furnish alkene 377 upon oxidation of the selenide and subsequent syn-elimination.170 Alternatively, alkyl iodide 375 could be converted to the corresponding phenyl selenide (378) as described by Sharpless.1 0 8 , 1 7 1 Again, oxidation of the selenide and subsequent syn-elimination of the corresponding selenoxide should provide alkene 377. 311 374 : R = OH 375: R = I Scheme 3.84: Proposed conversion of methyl ether 311 to alcohol 374 or alkyl iodide 375 3. Synthetic Approaches to (±)-isomulinic Acid 148 375 3 7 8 377 Scheme 3.85: Two possible preparations of alkene 377 If the preparation of alkene 377 proved successful, this material could potentially yield access to the mulinane family of natural products using the sequence of reactions outlined in Scheme 3.86. Thus, homologation of aldehyde 377 could be carried out as before (see Scheme 3.77). Allylic alcohol 380 should be accessible by treatment of aldehyde 379 with allyl magnesium bromide. The relative configuration of the carbon chirality center at C-14 (mulinane numbering) does not need to be defined at any stage. Subjection of diene 380 to ring-closing metathesis conditions should provide the desired tricyclic system (381). Oxidation of the secondary alcohol under neutral conditions should provide the corresponding (3,y-unsaturated enone (382). Alkylation of enone 377 with methyl iodide would install the requisite methyl group at C-13 (mulinane numbering). Note that compound 383 possesses the complete carbocyclic framework present in the mulinane family of natural products. Formation of alkenyl triflate 384 was expected to occur by treatment of the thermodynamic enolate of enone 383 with a suitable triflating agent. Reduction of the triflate group was expected to provide the 3. Synthetic Approaches to (±)-lsomulinic Acid 149 desired seven-membered ring diene. Finally, hydrolysis of the methyl ester on 312 would provide (±)-mulin-11,13-dien-20-oic acid 157. alkenyl triflate formation 384 312 157 Scheme 3.86: Revised synthetic route to (±)-mulin-11,13-dien-20-oic acid 157 3. Synthetic Approaches to (±)- lsomulinic Acid 1 5 0 Treatment of aldehyde 311 with TMSI provided alkyl iodide 375 as the major product (Scheme 3.87). The methylene protons at C-12 (mulinane numbering) of iodide 375 resonated at 5 = 3.04-3.15 (ddd; J= 9.5, 9.5, 9.5 Hz) and 5 = 3.29-3.37 (ddd; J= 9.5, 9.5, 4.9 Hz) in the 1H NMR spectrum. The aldehyde proton gave rise to a signal at 5 = 9.48. Scheme 3.87: Preparation of iodide 375 At this time it was postulated that treatment of iodide 375 with a bulky base would promote the E2 elimination of HI to establish a double bond between C-11 and C-12 (mulinane numbering). However, treatment of iodide 375 with f-BuOK resulted in the formation of mixed acetal 385. Presumably, the approach of f-butoxide anion to the hydrogen atoms (3 to the iodine atom is too sterically encumbered to allow for the E2 elimination process to occur. The formation of acetal 385 is thought to occur as shown in Scheme 3.88. Thus, 1,2 addition of the f-butoxide anion to the aldehyde generates anion 386. This material may revert back to aldehyde 375 and the f-butoxide anion. Alternatively, anion 386 may undergo an irreversible internal SN2 displacement reaction to generate the observed mixed acetal. 3. Synthetic Approaches to (±)-lsomulinic Acid 151 + H E 12 CHO E2 elimination •CHO 375 377 O M e Scheme 3.88: Attempted E2 elimination of HI from alkyl iodide 375, preparation of mixed acetal 385 The 1H NMR spectrum of acetal 385 displayed a nine-proton singlet at 8= 1.19 which was attributed to the f-butyl group. The acetal proton gave rise to a singlet at 8 = 4.80. The protons on the methylene group adjacent to the acetal oxygen atom were observed as one-proton multiplets at 8= 3.70-3.77 and 8= 3.52-3.65. The 1 3 C NMR spectrum displayed the correct number of signals, including a resonance at 8 = 94.9 which was attributed to the acetal carbon. The f-butyl group gave rise to a three-carbon resonance at 8= 28.8. A J-mod 1 3 C NMR experiment demonstrated the presence of six methylene groups and a total of twelve methyl and methine groups. The molecular mass of this compound was verified through an HRMS measurement of its parent ion. The relative configuration of the acetal carbon chirality center was determined using a number of 1D NOE difference experiments (Figure 3.32). Note that although acetal 385 is not useful in a synthetic sense, it served to confirm the conclusions regarding the relative configuration of the carbon chirality center at C-8 (mulinane numbering). 3. Synthetic Approaches to (±)-fsomulinic Acid 152 Me B _C02Me H A : 5 = 2.38-2.45 (ddd; J= 13.0, 3.6, 3.6 Hz) H B : 5 = part of a multiplet at 1.07-1.34 H c : 8 = 3.52-3.65 (m) H D : 8 = 4.80 (s) Me A : 8 = 0.94 (s) Me B : 8= 1.00 (d; J=6.4 Hz)) f-Bu:8 = 1.19(s) irradiation of H D enhances H B , H c and f-Bu irradiation of H A enhances H B and Me B Figure 3.32: Selected NOE difference data for acetal 385 In order to establish the double bond between C-11 and C-12 (mulinane numbering), iodide 375 was transformed into selenide 378 using the procedure outlined by Sharpless and Lauer.108 (Scheme 3.89). The 1H NMR spectrum of selenide 378 contained the correct number of signals. The phenyl group was evidenced by a three-proton multiplet at 5= 7.18-7.26 and a two-proton multiplet at 5= 7.42-7.48. The aldehyde proton gave rise to a singlet at 8 = 9.51. The methylene group adjacent to the selenium atom was observed as a pair of multiplets at 5 = 2.75-2.85 and 5 = 3.03-3.12 with a mutual coupling constant of 11.9 Hz. The 1 3C NMR spectrum displayed a two-carbon resonance at 5= 129.0, a two-carbon resonance at 132.6 and two one-carbon resonances at 8 = 126.8 and 5 = 130.1, all of which were assigned to the phenyl group. A J-mod 1 3C NMR experiment demonstrated the presence of six methylene groups and a total of fourteen methyl and methine groups. The molecular mass selenide 378 was confirmed through an HRMS measurement of its parent ion. 3. Synthetic Approaches to (±)-lsomulinic Acid 153 Scheme 3.89: Preparation of selenide 378 and alkene 377 The oxidation of selenide 378 to the corresponding selenoxide, and the subsequent syn-elimination reaction (see 387, Scheme 3.89) that was to establish the double bond between C-11 and C-12 (mulinane numbering) of alkene 377 was not a straightforward process. Treatment of selenide 378 with hydrogen peroxide provided a mixture of products. Alkene 377 could be isolated from this mixture in low yield. The three alkenyl protons of compound 377 gave rise to a one-proton multiplet at 8= 5.72-5.84 and a two-proton multiplet at 8= 5.05-5.12 in the 1H NMR spectrum. The aldehyde proton was observed as a singlet at 8 = 9.56. The 1 3C NMR spectrum of alkene 377 displayed a resonance at 8 = 118.0 and a resonance at 8= 136.2, which were both attributed to the alkenyl group. 3. Synthetic Approaches to (±)-lsomulinic Acid 154 The low yield of alkene 377 prompted the exploration of alternative conditions for the oxidation of selenide 378. The attempted oxidation of the phenyl selenide moiety of compound 378 with Nal0 4 in basic MeOH did provide the desired alkene, but in low yield. Alternatively, the oxidation of 378 with Nal0 4 under acidic conditions did not provide the desired alkene. Although the products from this reaction were not isolated, analysis of the 1H NMR spectrum of the crude reaction mixture revealed that none of the products in the mixture possessed an aldehyde group. 378 377 Scheme 3.90: Oxidation of Selenide 378 with Nal0 4 Clearly, the standard reagents for oxidation of selenides to selenoxides (H202, Nal04) were not sufficiently selective in the preparation of alkene 377. At this stage, the identity of the side-products arising from the oxidation of 378 with H 2 0 2 and Nal04 was unclear. However, it was hoped that a more selective oxidation protocol would provide clean access to alkene 377. It has been shown that selenides are cleanly and quantitatively oxidized to selenoxides with ozone.68 Indeed, this is the best method for the formation of selenoxides when special conditions are required for the elimination step. Treatment of selenide 378 with ozone at -78 °C in DCM, followed by the addition of benzene and heating of the reaction mixture to reflux resulted in the clean formation of lactone 390 (Scheme 3.91). The 1H NMR spectrum of lactone 390 exhibited the correct number of signals, including a two-proton multiplet at 8 = 4.32-4.43 which was attributed to the methylene group adjacent to the lactone oxygen. The 1 3C NMR spectrum displayed two carbonyl carbon resonances at 8= 176.2 and 8 = 3. Synthetic Approaches to (±)-lsomulinic Acid 155 174.5. A resonance at 8 = 66.9 was attributed to the methylene group adjacent to the lactone oxygen. A J-mod 1 3C NMR spectrum demonstrated the presence of six methylene groups and a total of eight methyl and methine groups. The generation of lactone 390 likely occurs through the oxidation of the phenyl selenide moiety to the corresponding selenoxide and the oxidation of the aldehyde group to the corresponding acid. Reich et ai}72 have shown that the syn-elimination of selenoxides is suppressed in the presence of protic solvents as a result of the strong hydrogen bonding properties of selenoxides. It is possible that the carboxylic acid and the selenoxide functions of compound 388 form an intramolecular hydrogen bond (as in 391), which helps to suppress the syn-elimination pathway to alkene 389. As a result, the selenoxide decomposes through an alternate pathway. Scheme 3.91: Oxidation of selenide 378 with ozone, formation of lactone 390 3. Synthetic Approaches to (±)-lsomulinic Acid 156 Due to the unexpected oxidation of the aldehyde group of compound 378 the oxidation of the selenide moiety was carried out on the corresponding alcohol. The reduction of aldehyde 378 to alcohol 392 was accomplished with the use of NaBH4 in MeOH. It seemed likely that the hydroxyl group could also be involved in some form of internal hydrogen bonding thereby suppressing the syn-elimination of the selenoxide. Thus, it was postulated that the inclusion of an amine base in the reaction mixture during the elimination step would allow the clean syn-elimination of the selenoxide.172 SePh SePh 378 392 i) 0 3 , DCM, -78 °C ii) PhH, Et 3 N, DCM, A Scheme 3.92: Preparation of alkenes 393 and 377 The oxidation of the phenyl selenide moiety of compound 392 was carried out with ozone at -78 °C. Benzene and triethyl amine were added to the solution of the selenoxide and the resulting mixture was heated to reflux. Gratifyingly, these reaction conditions provided efficient access to alkene 393. Oxidation of the alcohol moiety of compound 393 using Ley's method provided aldehyde 377, which had been previously characterized. ) 3. Synthetic Approaches to (±)-lsomulinic Acid 157 The 1H NMR spectrum of alkene 393 exhibited a one-proton resonance at 8 = 5.42-5.56 (ddd, J = 17.0, 10.0, 10.0 Hz) which was attributed to the alkenyl proton on C-11 (mulinane numbering). The alkenyl protons on C-12 (mulinane numbering) were observed as a one-proton multiplet at 8 = 5.00-5.05 and a one-proton multiplet at 8 = 4.97-5.00. The two carbinol protons at C-15 (mulinane numbering) were observed as a pair of doublets at 8= 3.68 and 8= 3.49 with a mutual coupling constant of 10.8 Hz. The 1 3C NMR spectrum displayed the correct number of signals, including two alkenyl carbon resonances at 8= 137.4 and 8= 116.7. The carbinol carbon was observed at 8= 58.1. A J-mod 1 3C NMR spectrum demonstrated the presence of six methylene groups and a total of nine methyl and methine groups. The synthetic plan called for the homologation of aldehyde 377 to aldehyde 379. This process was expected to occur through the formation of epoxides 394 from aldehyde 377 using the method developed by Corey and Chaykovsky (Scheme 3.93). Treatment of these epoxides with a Lewis acid should promote their rearrangement to the desired aldehyde. This rearrangement was expected to occur through the formation of secondary carbocation 396 followed by the migration of a hydride to form carbocation 397. Intermediate 397 should give rise to the desired aldehyde. Note that this two-step aldehyde homologation sequence provided efficient access to aldehyde 318 from aldehyde 311 (see Scheme 3.77). 3. Synthetic Approaches to (±)- lsomulinic Acid 158 Scheme 3.93: Proposed homologation of aldehyde 377 to aldehyde 379 Subjection of aldehyde 377 to epoxidation conditions furnished a mixture of oxiranes (394), which are epimeric at C-12 (mulinane numbering, Scheme 3.94). This mixture of epoxides proved inseparable by flash column chromatography on silica gel. Treatment of epoxides 394 with a bulky Lewis acid (MABR, see Scheme 3.47) resulted in the formation of an aldehyde as the main product. Unfortunately, the spectral data collected on the acquired material indicated that it was not the desired product. Scheme 3.94: Preparation of aldehyde 398. 3. Synthetic Approaches to (±)-lsomulinic Acid 159 The IR spectrum of aldehyde 398 displayed a carbonyl stretching absorption at 1719 cm'1. The aldehyde proton was observed as a singlet at 8 = 9.70 in the 1H NMR spectrum. Furthermore, the alkenyl protons at C-13 (mulinane numbering) were observed as a pair of signals at 8 = 4.91-4.96 and 8 = 4.96-5.02. The alkenyl proton at C-12 (mulinane numbering) resonated at 8= 5.74-5.87. (see Figure 3.33). The 1 3C NMR spectrum of alkene 398 displayed the correct number of signals, including a carbonyl carbon resonance at 8 = 207.0. In addition, the alkenyl carbons gave rise to signals at 8= 139.1 and 8= 115.4. Hc: 5 = 4.91-4.96 (ddd) A c = 10.1 Hz JBC = 3.0 Hz J C D or JCE = 1.5 Hz HF: 5 = 9.70 (s) Figure 3.33: Selected 1H NMR data for alkene 398 A proposed mechanism for the unexpected formation of alkene 398 is presented in Scheme 3.95. The first step involves the coordination of the oxirane oxygen of compound 394 to the Lewis acid to form complex 395. It was expected that the Lewis acid would promote the formation of secondary carbocation 396, which should give rise to the desired aldehyde (see Scheme 3.93). Instead, the activated epoxide is attacked by the vinyl group at C-9 (mulinane numbering). This results in the simultaneous opening of the epoxide ring and the formation of a carbon-carbon bond to provide secondary carbocation 399. A 1,2-hydride shift leads from secondary carbocation 399 to secondary carbocation 400. This material collapses to the Lewis acid bound aldehyde as shown (400 to 401). Finally, release of the catalyst provides the observed product (398). H A : 5 = 5.74-5.87 (dddd) J A B = 17.0 Hz J A C = 10.1 Hz JAD = JAE = 7.0 Hz H B : 5 = 4.96-5.02 (ddd) J A B = 17.0 Hz J B c = 3.0 Hz J B D or JBE = 1.7 Hz 3. Synthetic Approaches to (±)- lsomulinic Acid 1 6 0 C O 2 C H 3 396 M A B R -formation of secondary carbocation \ CO2CH3 3 9 4 MABR DCM, -78 °C C 0 2 C H 3 9 395 MABR C 0 2 C H 3 J 0 C l H I / C0 2 Me 398 c o 2 C H 3 1,2-hydride migration \ C02CH 3 |_| 400 M A B R -Scheme 3.95: Proposed mechanism for the Lewis acid catalyzed formation of alkene 398 from epoxides 394 3. Synthetic Approaches to (±)-lsomulinic Acid 161 Recall that the synthetic plan called for the preparation of tricyclic enone 382 from aldehyde 379 using a three-step sequence (see Scheme 3.86 and Scheme 3.96). Tricycle 382 has in place one of the double bonds present in mulin-11,13-dien-20-oic acid (157). In addition, the ketone carbonyl positioned at C-14 of compound 382 was expected to facilitate the installation of the methyl group at C-13 and the formation of a double bond between C-13 and C-14. Unfortunately, all of bicyclic aldehyde 377 had been committed to forming aldehyde 398. The implementation of the planned three-step sequence for the construction of the seven-membered ring using aldehyde 398 would furnish enone 402. Despite the functional groups present on the seven-membered ring of enone 402, its elaboration to compound 157 was not expected to be a straightforward process. 377 398 402 Scheme 3.96: Proposed construction of tricycle 402 3. Synthetic Approaches to (±)-lsomulinic Acid 162 The ring-closing diene metathesis reaction was viewed as a key step in the construction of the seven-membered ring of the mulinane diterpenoids. In recent years, this transformation has surfaced as one of the most powerful methods for ring construction.173"176 A variety of catalysts for ring-closing metathesis exist. In this synthetic study, the ruthenium carbene catalyst 403 was utilized177 (see Scheme 3.97). A kinetic study has revealed that the ring-closing metathesis reaction catalyzed by compound 404, which is similar to catalyst 403, occurs through two distinct pathways.178 In the case studied, the major catalytic cycle (>95%) is as depicted in Scheme 3.97. The first step involves coordination of one of the double bonds of diene 405 to the metal center of catalyst 404 to form complex 406. Dissociation of a tricyclohexyl phosphine ligand results in the formation of complex 407. An intramolecular metallacyclobutane formation reaction occurs to yield 408. This complex can undergo a cycloreversion reaction to provide complex 409, in which ethylene is coordinated to the metal center. The loss of ethylene gas leaves a vacant coordination site on the metal center. This site may be occupied by the second double bond of diene 405 to form complex 410. A second intramolecular metallacyclobutane formation reaction ensues to provide 411. A second cycloreversion reaction serves to close the carbocycle and yields complex 412. Dissociation of the carbocycle from the metal center and the incorporation of a tricyclohexyl phosphine ligand regenerates the catalyst. Note that when catalyst 403 is used, the catalytic species after the first cycle is compound 404. Therefore it was expected that the reaction mechanism depicted in Scheme 3.97 would operate during the planned ring-closing metathesis. 3. Synthetic Approaches to (±)-tsomulinic Acid 163 metallacyclobutane cycloreversion formation Scheme 3.97: Major catalytic pathway for the ring-closing metathesis of diene 405 using catalyst 404 1 7 8 The minor catalytic pathway is very similar to that depicted in Scheme 3.97. It differs in that all the ligands in the catalyst (404) remain bound to the metal center throughout the catalytic cycle. 3. Synthetic Approaches to (±)- lsomulinic Acid 1 6 4 Treatment of aldehyde 398 with allyl magnesium bromide provided a nearly 1:1 mixture of homoallylic alcohols epimeric at C-15 (mulinane numbering, 414 and 415, Scheme 3.98). Although these alcohols could be separated on TLC, they were carried on to the next step as a mixture. Thus, subjection of alcohols 414 and 415 to ring-closing metathesis reaction conditions provided a mixture of tricyclic alcohols 416 and 417. 416 417 Scheme 3.98: Preparation of tricyclic alcohols 416 and 417 from aldehyde 398 Alcohols 416 and 417 were easily separated by flash column chromatography on silica gel. The 1H NMR spectrum of alcohol 416 exhibited a two proton multiplet at 8 = 5.62-5.76 which accounted for the alkenyl protons. The carbinol proton was observed as a one-proton broad doublet at 8 = 3.84-3.92 (J = 11.4 Hz). The 1 3C NMR displayed two alkenyl carbon resonances at 8= 127.7 and 8= 131.2. A J-mod 1 3C NMR experiment demonstrated the presence of six methylene groups and a total of eleven methyl and methine groups. The relative configuration of the carbon chirality center at C-15 (mulinane numbering) was determined to be as shown based on an NOE difference experiment (see Figure 3.34). 3. Synthetic Approaches to (±)-lsomulinic Acid 165 The 1H NMR spectrum of alcohol 417 displayed two one-proton resonances at 5= 5.63-5.71 and 8 = 5.82-5.91 which were attributed to the alkenyl protons. The carbinol proton resonated at 8= 3.46-3.53 (dd; J= 7.0, 7.0 Hz). The 1 3C NMR spectrum contained two alkenyl carbon resonances at 8= 127.7 and 8= 132.7. A J-mod 1 3C NMR experiment demonstrated the presence of six methylene groups and a total of eleven methyl and methine groups. The relative configuration of the carbon chirality center at C-15 (mulinane numbering) was assigned to be as shown based on an NOE difference experiment (see Figure 3.34). H A : 8 = 1.00-1.10 (ddd, J= 14.6, 14.6, 3.7 Hz) H A : 5 = 3.46-3.53 (dd J= 7.0, 7.0 Hz) H B : 5 = 3.84-3.92 (br. d, J= 11.4 Hz) Me A : 5 = 0.90 (s) Me A : 5 = 0.98 (s) irradiation of H A enhances Me A irradiation of H B enhances H A , does not enhance Me A Figure 3.34: Selected NOE difference data for alcohols 416 and 417 3. Synthetic Approaches to (±)-lsomulinic Acid 166 Tricyclic alcohols 416 and 417 converged to enone 402 upon oxidation using Ley's method (see Scheme 3.99). Due to a fortuitous dispersion of signals, the 1H NMR spectrum of enone 402 could be fully assigned. The alkenyl protons gave rise to two one-proton multiplets at 8= 5.67-5.76 and 8= 5.79-5.88. The C-11 methylene protons were observed as a one-proton multiplet at 8= 3.67-3.74 and a one-proton doublet of doublets at 8= 2.62-2.71 (J = 12.7, 8.7 Hz). The 1 3C NMR spectrum of contained two alkenyl carbon resonances at 8= 123.8 and 8= 131.4. The ketone carbonyl carbon resonated at 8= 208.4. A J-mod 1 3C NMR experiment verified the presence of six methylene groups and a total of ten methyl and methine groups. 416 417 402 Scheme 3.99: Preparation of (3,y unsaturated ketone 402 At this point, due to time constraints and in consideration of the expected difficulties in accessing the mulinane family of diterpenoids from compound 402, the proposed synthesis of (±)-isomulinic acid was abandoned. 3. Synthetic Approaches to (±)-lsomulinic Acid 167 3.5 Summary The total synthesis of isomulinic acid (39, Figure 3.35) was attempted in racemic form. The natural product mulin-11,13-dien-20-oic acid (157) was identified as a key intermediate en route to isomulinic acid. Four distinct synthetic approaches were explored but none provided access to the mulinane carbocyclic framework. Each approach provided insights which influenced the design of the next. The final synthetic approach allowed access to tricyclic enone 402 in a total of 23 linear steps from a known compound. Note that tricycle 402 possesses all the carbon chirality centers present in mulin-11,13-dien-20-oic acid (157) with the correct relative configuration. Figure 3.35: Structures of isomulinic acid (39), mulin-11,13-dien-20-oic acid (157) and tricyclic enone 402 The first synthetic plan incorporated the seven-membered ring annulation method developed in this laboratory as a key step.34 The known keto ester 198 1 1 4 was prepared in three steps from commercially available materials (Scheme 3.100). Bicyclic enone 197 was constructed from keto ester 198 in three steps and served as a platform for all four synthetic approaches. Note that the carbon chirality centers at C-3 and C-5 of enone 197 possess the correct relative configuration. Alkene 195 was prepared from enone 197 in five steps using standard methods. 3. Synthetic Approaches to (±)-lsomulinic Acid 168 .0 P 48% 45% v»C0 2Me 3 steps 5 steps 198 197 195 Scheme 3.100: Preparation of alkene 195 A diastereoselective hydroboration reaction of alkene 195 was used to establish the carbon chirality center at C-10 of alcohol 194 (Scheme 3.101). Unfortunately, the diastereoselectivity of this reaction was poor. Furthermore, alcohols 194 and 213 proved difficult to separate and were, therefore, oxidized to the corresponding ketones (193 and 220) as a mixture. Note that the carbon chirality center at C-10 of ketone 193 is configurationally stable under the oxidation conditions. Ketones 193 and 220 were easily separated and ketone 193 was elaborated to keto ester 189. It could be shown that the alkylation of keto ester 189 proceeds with a high degree of diastereoselectivity to establish the carbon chirality center at C-8 (Scheme 3.102). Unfortunately, the alkylation of keto ester 189 with the bifunctional reagent cis-5-iodo-1-tri-n-butylstannylpent-1-ene (36) provided a mixture of keto esters 225 and 226, which are epimeric at C-10. Evidently the desired product is configurationally unstable under the alkylation conditions. The failure to prepare alcohol 194 in high yield from alkene 195, and the configurational instability of keto ester 225 caused us to abort this synthetic plan. 3. Synthetic Approaches to (±)- lsomulinic Acid 169 96% combined yield based on recovered starting material (2:1) oxidation 80% 57% 38% Scheme 3.101: Preparation of keto ester 189 Bu 3 Sn . Bu 3Sn Scheme 3.102: Alkylation of keto ester 189 with bifunctional reagent 36 3. Synthetic Approaches to (±)-lsomulinic Acid 170 The second synthetic approach focused on trying to gain expedient access to a tricyclic substance, exemplified by the general structure 229, which possesses the requisite carbon chirality centers with the correct relative configuration (Figure 3.36). It was envisioned that the carbon chirality centers at C-8 and C-9 of tricycle 229 would be established through a highly diastereo- and regioselective [2+2]-photocycloaddtion reaction between alkene 228 and a suitable partner. Thus, the synthetic plan focused on gaining access to bicycle 228, which possesses the correct relative configuration at C-10. Bicyclic enone 197 was efficiently converted to allylic alcohol 251 using standard methods (Scheme 3.103). The hydroxyl group at C-8 of alcohol 251 was utilized to direct the hydrogenation of the double bond and thus provide access to alcohol 250. This material was oxidized to ketone 249. The formation of the thermodynamic enolate of ketone 249, followed by treatment with a sililating agent, provided silyl enol ether 261, in which the carbon-carbon double bond resides between C-7 and C-8, as the major product. This indicated that it would be difficult to establish the desired double bond between C-8 and C-9 of alkene 228. 229 228 Figure 3.36: Structures of tricycle 229 and alkene 228 3. Synthetic Approaches to (±)-lsomulinic Acid 171 Scheme 3.103: Preparation of silyl enol ether 258 An alternate route to alkene 228 was devised. This strategy allowed for the selective activation of C-9 versus C-7. Thus, enone 197 was reduced to alcohol 252 (Scheme 3.104). This material was transformed to alkene 266 using the method developed by Myers and Zheng.133 The diasteroselective epoxidation of alkene 266 provided oxirane 270. A three step sequence inspired by the work of Barrero138 provided access to bridged lactone 273. It was hoped that methanolysis of lactone 273 would provide access to allylic alcohol 274 which could, in principle, yield alkene 228 through the use of Myers' method.133 Unfortunately, compound 273 proved unreactive under methanolysis conditions. The stability of lactone 273 coupled with an interest in exploring a more expedient route to the mulinane natural products led us to abandon this synthetic approach. 3. Synthetic Approaches to (±)-lsomulinic Acid 172 274 273 270 Scheme 3.104: Preparation of lactone 273 The results obtained from the second synthetic approach demonstrated that it was necessary to differentiate between C-7 and C-9 prior to establishing the carbon chirality center at C-10. A report by McKenzie139 indicated that frans-fused hydrindane systems bearing an alkyl substituent at C-9 could be obtained through the heterogeneous hydrogenation of the corresponding enones. It was realized that if Mackenzie's mechanistic rationale held true when applied to an enone such as 293 (see Scheme 3.105), it would provide expedient access to a hydrindane system which possessed four of the five requisite carbon chirality centers with the correct relative configuration (299). The fifth carbon chirality center was expected to be installed using Yamamoto's procedure for the rearrangement of trisubstituted oxiranes to tertiary aldehydes.140 The thermodynamic enolate of ketone 197 was alkylated with iodide 292 to provide enone 293 (Scheme 3.105). The hydrogenation of 293 under heterogeneous conditions led to undesired sideproducts when carried out in large scale. However, small scale reactions provided sufficient material to further explore the synthetic plan. Subjection of the material acquired from the hydrogenation reaction to equilibrating conditions furnished ketone 299. Note that this compound possesses four of the 3. Synthetic Approaches to (±)- lsomulinic Acid 173 requisite carbon chirality centers with the correct relative configuration. Wittig olefination of ketone 299, followed by epoxidation of the acquired alkene furnished oxirane 302. Unfortunately, treatment of this epoxide with Yamamoto's Lewis acid provided ketone 304 as the major product. OTBS OTBS 304 302 Scheme 3.105: Preparation of ketone 304 The exploration of this synthetic approach demonstrated that a frans-fused hydrindanone with four of the requisite carbon chirality centers, such as 299, could be prepared in an expedient fashion. Clearly, the hydrogenation process required some modifications. It was evident that the terf-butyldimethyl silyl protecting group was too labile under the hydrogenation conditions. Therefore, it was decided to replace it with a methyl group in the fourth and final approach explored. 3. Synthetic Approaches to (±)-lsomulinic Acid 174 As expected, the alkylation of enone 197 with 2-bromoethyl methyl ether provided the required C-9 alkylated enone (not shown). Hydrogenation of this enone, followed by equilibration of the acquired material, provided ketone 309 (Scheme 3.106). Note that bicycle 309 possesses the correct relative configuration at C-3, C-5, C-9 and C-10. Ketone 309 was converted to a mixture of spiroepoxides (not shown) using the Corey-Chaykovsky protocol.154,155 Treatment of this mixture of epoxides with Yamamoto's Lewis acid1 4 0 provided an inseparable mixture of aldehydes 310 and 327, which are epimeric at C-8. The direct alkylation of aldehydes 310 and 327 to establish the carbon chirality center at C-8 failed under a variety of conditions. Therefore a reaction sequence which relied on the known rearrangement of cyclopropanols to a -methyl aldehydes160"165 was used to establish the C-8 quaternary center. Thus, aldehydes 310 and 327 were converted to a mixture of spirocyclopropanols 346 and 347 in three steps. Alcohols 346 and 347 converged to aldehyde 311 upon treatment with a Bronsted acid in hot THF. Note that aldehyde 311 possesses all five carbon chirality centers present in all the members of the mulinane family of diterpenoids, with the correct relative configuration: 3. Synthetic Approaches to (±)- lsomulinic Acid 175 O M e O M e 4 o C H O 57% 378 4 steps U C H O 310: = H, R 2 = C H O 327: R, = C H O , R 2 = H 3 steps 0 X H O 311 346: RT = H, R 2 = O H 347: RT = O H , R 2 = H % o C H O 75% 3 steps H H C 0 2 M e 402 Scheme 3.106: Final synthetic approach to the (±)- isomulinic acid Having obtained access to aldehyde 311, all efforts were focused on establishing the third ring of the mulinane carbocyclic skeleton. Treatment of aldehyde 311 with TMSI1 5 0 provided the corresponding C-12 iodide (not shown). This material was converted to phenylselenide 378 using Sharpless' protocol. The preparation of alkene 377 from phenylselenide 378 required a four-step sequence. An unexpected rearrangement occurred when the homologation of aldehyde 377 was attempted. 3. Synthetic Approaches to (±)-lsomulinic Acid ; •. 176 Conversion of aldehyde 377 to a mixture of spiroepoxides, followed by treatment of this mixture with a Lewis acid resulted in the conversion of the C-9 vinyl group to an allyl group. Finally, it could be shown that the seven-membered ring could be constructed using a sequence of reactions which included a ring-closing diene metathesis as a key step. Unfortunately, due primarily to time constraints, the total synthesis of (±)-isomulinic acid (39) or any other member of the mulinane family of diterpenoids was not achieved. However, it could be shown that the hydrindane system embedded in the mulinane carbocyclic skeleton, with the correct relative configuration at all five carbon chirality centers, could be accessed in an efficient and expedient manner (see 311, Scheme 3.106). Furthermore, the seven-membered ring could be constructed using a strategy which included a ring-closing diene metathesis as a key step. Unfortunately, due to an unexpected rearrangement reaction that occurred during the attempted homologation of aldehyde 377, the position of the functional groups on the seven-membered ring of tricyclic enone 402 was not conducive to an expedient construction of (±)-isomulinic acid (39). 177 4. Experimental 4.1 General 4.1.1 Data acquisition, presentation and experimental techniques All reported compounds have been given systematic names according to the IUPAC protocol for bridged carbocycles.90,91 Melting points were recorded on a Fischer-Johns melting point apparatus. Infrared (IR) spectra were recorded as thin films between sodium chloride plates employing a Perkin-Elmer 1710 FT-IR spectrophotometer with internal calibration. Proton nuclear magnetic resonance (1H NMR) spectra were recorded on Bruker spectrometer models AMX-500 (500 MHz), WH-400 (400 MHz), AV-400 (400 MHz) or AV 300 (300 MHz) using deuteriochloroform (CDCI3) as the solvent.179'180 Signal positions (5 values) are reported in parts per million (ppm) from tetramethylsilane (5 = 0) and were measured relative to the signal for chloroform (8 = 7.24). Whenever possible, the 1H coupling constants (J values) were determined using the guidelines to first-order multiplet analysis reported by Hoye et a/. 1 8 1 ' 1 8 2 Coupling constants are reported in Hertz (Hz). The tin-proton coupling constants (Jsn-H) are reported as an average of the 1 1 7Sn and the 1 1 9Sn values. The spectral data are reported in the following format: chemical shift (ppm), multiplicity, number of protons, coupling constants(s), and assignments (when known). The abbreviations used for multiplicities are: s (singlet), d (doublet), t (triplet), q (quartet), m (multiplet), and br (broad). In some cases, the proton assignments were supported by COSY (1H-1H homonuclear correlation spectroscopy) and/or 1D-NOED (one dimensional nuclear Overhauser enhancement difference) experiments. Carbon nuclear magnetic resonance (13C NMR) spectra were recorded on Bruker spectrometer models AMX-500 (125.8 MHz), AV-400 (100.4 MHz) or AV 300 (75.5 MHz) using deuteriochloroform as the solvent. Signal positions (8 values) are 4. Experimental 178 reported in parts per million (ppm) from tetramethylsilane (8 = 0) and were measured relative to the signal for chloroform (8 = 77.0). In some cases, J-modulated spin-echo experiments (also called attached proton test (APT) experiments) were used to differentiate the signals due to methyl and methine groups from the signals due to methylene groups and quaternary carbons. The ATP spectra were phased such that the signals due to methyl and methine groups are negative (-ve) and the signals due to methylene groups and quaternary carbons are positive (+ve). In some cases, the proton and carbon assignments were supported by (1H, 13C) heteronuclear multiple quantum coherence (HMQC) experiments and heteronuclear multiple bond correlation (HMBC) experiments, which were carried out on a Bruker AMX-500 spectrometer or a Bruker AV-400 spectrometer. Coupled gas chromatography-low resolution mass spectrometry analyses (GC-MS) were carried out on an Agilent 6890 series gas chromatograph equipped with an HP-5 column (cross-linked 5% diphenyl, 95% dimethylpolysiloxane, length: 30 m, internal diameter: 0.25 mm, film thickness: 0.25 pm) and coupled to an Agilent 5973Network mass selective detector. High resolution electron impact mass spectra (HREIMS) were recorded on a Kratos MS80 or on a Kratos Concept II HQ mass spectrometer. The molecular ion [M+] masses are reported unless otherwise noted. All compounds subjected to high resolution mass measurements were homogeneous by TLC and/or GC-MS analysis. Thin layer chromatography (TLC) analyses were carried out on commercially available aluminium-backed silica gel 60 F254 plates (E. Merck or Macherey-Nagel brand, thickness: 0.2 mm). Visualization was accomplished by using uv light (254 nm) and/or staining the plates with one of the following reagents: a) phosphomolybdic acid (PMA) in EtOH (20% w/v, Aldrich Chemical Co.), b) ammonium molybdate (5% w/v) and cerium(IV) sulfate (0.1% w/v) in 10% aqueous sulfuric acid, c) vanillin (6% w/v) in sulfuric acid (4% v/v)-EtOH (10% water v/v in EtOH). 4. Experimental 179 Purifications by flash column chromatography were performed using 230-400 mesh silica gel (SiliCycle) following the technique described by Still.1 8 3 All reactions which are sensitive to moisture were carried out using flame- or oven-dried glassware under an atmosphere of dry argon. Glass syringes, Teflon® cannulae and stainless steel needles used to handle dry solvents and reagents were oven dried, cooled in a dessicator, and flushed with argon prior to use. Sub-milliliter volumes of reagents were measured using gas-tight™ microliter syringes which had been dried under reduced pressure (vacuum pump) and stored in a dessicator prior to use. Concentration, evaporation, or removal of solvent under reduced pressure (water aspirator) refers to solvent removal using a Buchi rotary evaporator at 20-25 Torr. Removal of solvents using fractional distillation at atmospheric pressure was carried out using a Vigreux column (length: 100 mm, internal diameter: 10 mm). Cold temperatures were maintained using one of the following: 10 °C; Julabo cryobath (model FP80), 0 °C, ice-water bath or Julabo cryobath (model FP80); -30 °C, aqueous CaCI2-dry ice (35 g CaCI2/100 mL H 20); 1 8 4 -78 °C, acetone-dry ice or Julabo cryobath (model FP80). 4.1.2 Solvents and reagents All solvents and reagents were purified using established procedures.185 Benzene, DCM, pentane, and DME were distilled from calcium hydride. Et20 and THF were distilled from sodium benzophenone ketyl. All of these solvents were distilled under an atmosphere of dry argon and used immediately. Methanol was distilled from Mg with a catalytic amount of l 2 under an atmosphere of nitrogen. Methyl vinyl ketone was distilled prior to use. lodomethane and diiodomethane were passed through a short column of oven dried basic alumina prior to use. Diisopropylamine and BF3»Et20 were distilled from calcium hydride under an atmosphere of argon. 4. Experimental 180 MABR was prepared using the procedure developed by Yamamoto and coworkers.140 NBSH was prepared using the method of Myers.136 DMDO was prepared using the protocol outlined by Adam et a/.186 Solutions of LDA were prepared by adding a solution of butyllithium (1.0 equiv.) in hexanes to a solution of freshly distilled diisopropylamine in THF at -78 °C, followed by warming to 0 °C for 20 min prior to use. Petroleum ether refers to a hydrocarbon mixture with a boiling point range of 30-60 °C Argon and nitrogen were dried by bubbling the gas through concentrated sulfuric acid and then passing it through a column of Drierite® and potassium hydroxide. All other solvents and reagents were commercially available and were used without further purification. 4 . Experimental 181 4.2 Experimental - Total Synthesis of (±)-Kelsoene 3-(4-Chlorobut-1-en-2-yl)cyclopentanone (118) and (1fl*, 5S*)-6-methylidenebicyclo[3.3.0]octan-2-one (rac-34). ^ S n M e 3 Cl Cl O 32 118 rac-34 To a cold (-78 °C) stirred solution of 4-chloro-2-trimethylstannylbut-1-ene (32) (7.08 g, 27.9 mmol) in dry THF was added MeLi (1.6 M solution in Et20, 19 mL, 30.5 mmol) and the mixture was stirred at -78 °C for 5 min. A solution of CuCN (2.71 g, 30.3 mmol) and LiCI (1.29 g, 30.3 mmol) in dry THF (35 mL) was added and the solution was stirred at -78 °C for 5 min. BF3»Et20 (3.79 mL, 29.9 mmol) was added dropwise over 10 min and the solution was stirred at -78 °C for 15 min. Freshly distilled cyclopent-2-en-1-one (2.79 mL, 33.4 mmol) was added and the mixture was stirred at -78 °C for 1 h. Sat. aq NH4CI (30 mL) was added and the stirred mixture was allowed to warm to r.t. The phases were separated and the aqueous layer was extracted with Et20 (2 x 30 mL). The combined organic extracts were washed (brine, 30 mL), dried (MgS04), and concentrated. Flash chromatography (400 g of silica gel, 5:1 petroleum ether-Et20) of the crude oil afforded compound 118 as a colourless oil that displayed 1H NMR spectral data in accordance with that previously reported in the literature; 3 2 yield: 3.84 g (79%). Conversion of the chloroketone 118 into the bicyclic ketone rac-34 was accomplished by treatment of the former substance with KH in THF as previously described.32 4. Experimental 182 (1 /?*,5S*,6S*)-6-Methylbicyclo[3.3.0]octan-2-one (rac-59) and (1 /7*,5S*,6/?*)-6-methylbicyclo[3.3.0]octan-2-one (119) rac-34 rac-59 119 To a solution of ketone rac-34 (2.235 g, 16.4 mmol) in DCM (120 mL) was added chlorotris(triphenylphosphine)rhodium(l) (Wilkinson's catalyst, 106 mg, 0.11 mmol). The solution was stirred at r.t. under an atmosphere of hydrogen (-760 torr) for 48 h. The solution was filtered through a pad of silica gel (25 g, elution with DCM) and the filtrate was concentrated via fractional distillation. Flash chromatography (200 g of silica gel, 7:1 penf.ane-Et.2O) of the remaining material provided a colourless oil that consisted of a mixture of ketones rac-59 and its C-6 epimer 119 (ratio = 95:5 by 1H NMR spectroscopy); yield: 2.15 g (95%). IR (film): v = 2953, 2874, 1736, 1459, 1161, 1107 cm"1. 1H NMR (500 MHz): 5 = 0.97 (d, 3 H, J = 6.9 Hz), 1.12 (m, 1 H), 1.48 (dddd, 1 H, J = 13.0, 11.4, 9, 8.3 Hz), 1.62 (dddd, 1 H, J=12.7, 6, 6, 3 Hz), 1.77-1.83 (m, 2 H), 1.84 (dddd, 1 H, J -13 .0 , 9.4, 3.4,1.0 Hz), 2.03-2.13 (m, 1 H), 2.17 (ddd, 1 H, J -17 .6 , 8.3, 3.4 Hz), 2.25 (dddd, 1 H, J = 17.6, 11.4, 9.4, 1.4 Hz), 2.58 (ddd, 1 H, J = 9, 9, 5.0 Hz), 2.66 (m, 1 H). 1 3C NMR (125 MHz): 5 = 14.2, 21.4, 28.8, 32.3, 38.3, 39.6, 45.0, 51.7, 223.1. Exact mass calculated for C 9 H 1 4 0: 138.1045; found 138.1044. 4. Experimental 183 Table 4.1. NMR data for (1 fT,5S*,6S*)-6-methylbicyclo[3.3.0]octan-2-one (rac-59) Carbon 1 3 C 1H HMBC No. 5 (ppm)a 5 (ppm)(mult; J(Hz))b c'd Correlations'3'6 1 51.7 H-1: 2.58 (ddd; 9,9,5.0) H-5, H-8a, H-8b 2 223.1 H-1, H-3a, H-3b, H-4b 3 39.6 H-3a: 2.17 (ddd; 17.6, 8.3, 3.4) H-3b: 2.25 (dddd; 17.6, 11.4, 9.4, 1.4) 4 21.4 H-4a: 1.48 (dddd; 13.0, 11.4, 9, 8.3) H-1, H-3a, H-3b, H-5, H-6 H-4b: 1.84 (dddd; 13.0, 9.4, 3.4, 1.0) 5 45.0 H-5: 2.66 (m) H-3a, H-3b, H-4a, H-4b, H-6, H-7, H-9 6 38.3 H-6: 2.03-2.13 (m) H-4a, H-7b, H-8a, H-8b, H-9 7 32.3 H-7a: 1.12 (m) H-5, H-6, H-8a, H-8b, H-9 H-7b: 1.62 (dddd; 12.7, 6, 6, 3) 8 28.8 H-8a: part of the m at 1.77-1.83 (m) H-1, H-5, H-6, H-7b H-8b: part of the m at 1.77-1.83 (m) 9 14.2 0.97 (d; 6.9) H-6, H-7a Recorded at 125 MHz. Recorded at 500 MHz. Assignments based on HMQC data recorded at 500 MHz. Methylene protons are designated H-Xa and H-Xb arbitrarily. Only those correlations which could be unambiguously assigned are recorded. 4. Experimental 184 Table 4.2. NMR data for (1 R*,5S*,6S*)-6-methylbicyclo[3.3.0]octan-2-one (rac-59) Proton No. 1H 5 (ppm)(mult; J(Hz))a'b'° COSY Correlations3^ H-1 2.58 (ddd; 9,9,5.0) H-8a, H-8b H-3a 2.17 (ddd; 17.6, 8.3, 3.4) H-3b, H-4a, H-4b H-3b 2.25 (dddd; 17.6, 11.4, 9.4, 1.4) H-3a, H-4a, H-4b H-4a 1.48 (dddd; 13.0, 11.4, 9 ,8.3) H-3a, H-3b, H-4b, H-5 H-4b 1.84 (dddd; 13.0, 9.4, 3.4, 1.0) H-3a, H-3b, H-4a, H-5 5 2.66 (m) H-4a, H-4b, H-6 6 2.03-2.13 (m) H-5, H-9 H-7a 1.12 (m) H-7b, H-8a, H-8b H-7b 1.62 (dddd; 12.7, 6, 6, 3) H-7a, H-8a, H-8b H-8a part of the m at 1.77-1.83 (m) H-1, H-7a, H-7b H-8b part of the m at1.77-1.83 (m) H-1, H-7a, H-7b 9 0.97 (d; 6.9) H-6 Recorded at 500 MHz. Assignments are based on HMQC data recorded at 500 MHz. ' Methylene protons are designated H-Xa and H-Xb arbitrarily. Only those correlations which could be unambiguously assigned are recorded. 4. Experimental 185 (1 /7*,5A?*,6S*)-6-Methylbicyclo[3.3.0]oct-3-en-2-one (rac-60) H H rac-59 119 rac-60 To a cold (-78 °C) solution of LDA (0.321 M in THF, 25 mL, 8.03 mmol) was added a cold (-78 °C) solution of ketones rac-59 and 119 (0.923 g, 6.69 mmol) in THF (20 mL). The solution was warmed to 0 °C for 30 min and then was cooled to -78 °C. A cold (-78 °C) solution of phenylselenenyl chloride (1.60 g, 8.35 mmol) in THF (10 mL) was added, the mixture was stirred at -78 °C for 1 h, and then was treated with 10% aq HCI (10 mL). The phases were separated and the organic phase was washed (brine, 10 mL), dried (MgSC^) and concentrated. The crude product was dissolved in DCM (20 mL) and the solution was cooled to 0 °C. Aq H 2 0 2 (8.8 M, 1.5 mL, 13 mmol) was added, the mixture was stirred for 1 h, and then was treated with 10% aq HCI (10 mL). The phases were separated and the organic phase was washed (sat. aq NaHC03, 10 mL), dried (MgS04) and concentrated by fractional distillation. Flash chromatography (80 g of silica gel, 5:1 pentane-Et20) of the crude oil provided enone rac-60 as a colourless oil; yield: 0.755 g (83%). No attempt to identify or isolate the product derived from ketone 119 was made. IR (film): v = 2954, 2873, 1707, 1585, 1459, 1353, 1187, 732 cm"1. 1H NMR (400 MHz): 5 = 0.83-0.91 (m, 1 H), 1.01 (d, 3 H, J =7.0 Hz), 1.49-1.57 (m, 1 H), 1.63-1.75 (m, 1 H), 1.84 (dd, 1 H, J=12.6, 6.4 Hz), 1.97-2.01 (m, 1 H), 2.63 (dd, 1 H, J = 10.4, 5.5 Hz), 3.23 (dddd, 1 H, J = 8.5, 5.5, 2.7, 2.7 Hz), 6.17 (dd, 1 H, J = 5.8, 2.7 Hz), 7.57 (dd, 1 H, J = 5.8, 2.7 Hz). 1 3C NMR (100 MHz): 5 = 15.7, 28.9, 31.3, 36.8, 49.9, 50.2, 135.8, 165.3, 213.7. Exact mass calculated for C 9 H 1 2 0: 136.0888; found 136.0893. 4. Experimental 186 (1 S*,5/?*,8S*)-4,8-Dimethylbicyclo[3.3.0]oct-3-en-2-one (rac-61) H H .0 rac-60 rac-61 To a cold (-78 °C) solution of enone rac-60 (0.872 g, 6.41 mmol) in Et20. (30 mL) was added MeLi (1.6 M solution in Et20, 4.5 mL, 7.2 mmol) and the mixture was allowed to warm to r.t. over a period of 1 h. The solution was treated with tris(hydroxy-methyl)aminomethane buffer (pH = 8, 0.5 M, 20 mL) and the phases were separated. The organic phase was washed with brine (20 mL), dried (MgS04) and concentrated. The remaining crude oil was dissolved in DCM (40 mL) and the solution was cooled to 0 °C. Pyridinium chlorochromate adsorbed on basic alumina (0.79 mmol/g, 12 g, 9.5 mmol) was added to the solution in one portion and the resulting slurry was stirred for 4 h. The mixture was filtered through a pad of Florisil® (60 g, elution with 100 mL of Et20). The filtrate was concentrated by fractional distillation and the residual oil was purified by flash chromatography (50 g of silica gel, 3:1 pentane-Et20) to provide enone rac-61 as a colourless oil; yield: 0.557 g (60 %). IR (film): v = 2953, 1699, 1622, 1439, 1379, 1273, 1190, 882 cm"1. 1H NMR (400 MHz): 5 = 0.95-1.01 (m, 1 H), 1.03 (d, 3 H, J = 7.0 Hz), 1.57-1.72 (m, 3 H), 2.01-2.10 (m, 1 H), 2.02 (s, 3 H), 2.60 (dd, 1 H, J = 9.9, 5,6 Hz), 3.08-3.15 (m, 1 H), 5.82-5.84 (m, 1 H). 1 3C NMR (100 MHz): 6 = 15.6, 17.5, 27.7, 32.3, 37.0, 50.9, 53.7, 132.4, 179.6, 210.2. Exact mass calculated for Ci 0H 1 4O: 150.1045; found 150.1048. 4. Experimental 187 (1 R*,2S*,5S*,7S*,8S*)-2,8-Dimethyltricyclo[5.3.0.02'5]decan-6-one (rac-64) rac-61 rac-64 Enone rac-61 (37 mg, 0.25 mmol) was dissolved in DCM (25 mL) that had been presaturated with ethylene (C. P. grade) at -78 °C. The resulting solution was irradiated (-78 °C, 12 h) through a Pyrex filter (X > 290 nm) using a 450-W Hanovia medium pressure mercury arc lamp. Ethene was bubbled through the stirred solution during the irradiation. The solution was concentrated by fractional distillation and the residual oil was purified by flash chromatography (2 g of silica gel, 4:1 pentane-Et20) to provide ketone rac-64 as a colourless oil; yield: 40 mg (90%). IR (film): v = 2952, 1728, 1455, 1378, 1268 cm'1. 1H NMR (400 MHz): 5 = 0.99 (d, 3 H, J=7.3 Hz), 1.04-1.14 (m, 1 H), 1.23 (s, 3 H), 1.47-1.63 (m, 2 H), 1.64-1.79 (m, 3 H), 1.82-1.91 (m, 1 H), 2.09-2.15 (m, 1 H), 2.23-2.37 (m, 3 H), 2.93 (dd, 1 H, J =9, 9 Hz). 1 3C NMR (100 MHz, APT): 5 = 16.4 (-ve), 20.9 (+ve), 21.0 (-ve), 27.8 (+ve), 34.1 (+ve), 35.0 (+ve), 37.9 (-ve), 44.0 (+ve), 51.8 (-ve), 52.5 (-ve), 57.9 (-ve), 225.4. Exact mass calculated forC 1 2H 1 80: 178.1358; found 178.1361. 4. Experimental 188 (1 fr^^.S/r.y^^^^.S-Dimethyl-e-methylidenetricycloIS.S.O.O^ldecane (143) H : H : rac-64 143 To a cold (0 °C) stirred suspension of Lombardo's reagent ' (~ 1.4 mmol) in THF (4 mL) was added, sequentially, DCM (10 mL) and, via a cannula, a solution of ketone rac-64 (128 mg, 0.72 mmol) in DCM (2 mL). The mixture was allowed to warm to r.t. and then was stirred for 16 h. Sat. aq. NaHC03 (10 mL) was added and the resulting biphasic mixture was filtered through a pad of neutral alumina (10 g). The alumina was eluted with Et20 (20 mL). The phases of the combined eluate were separated and the organic phase was washed (brine, 10 mL), dried (MgS04), and concentrated by fractional distillation. Purification of the crude oil by flash chromatography (10 g of silica gel, pentane) provided alkene 143 as a colourless oil; yield: 81 mg (63%). 1H NMR (400 MHz): 5 = 0.89 (d, 3 H, J = 6.4 Hz), 1.14 (s, 3 H), 1.20-1.30 (m, 1 H), 1.32-1.43 (m, 1 H), 1.45-1.52 (m, 1 H), 1.53-1.63 (m, 3 H), 1.84 (dd, 1 H, J=19,6, 9.1 Hz), 2.00-2.12 (m, 1 H), 2.18 (ddd, 1 H, J = 9, 9, 4.1 Hz), 2.23-2.36 (m, 2 H), 3.21-3.28 (m, 1H), 4.90-4.95 (m, 2 H). 1 3C NMR (75 MHz, APT): 8= 16.7 (-ve), 20.6 (-ve), 25.6 (+ve), 27.9 (+ve), 29.7 (+ve), 34.6 (+ve), 34.7 (+ve), 38.2 (-ve), 51.0 (-ve), 54.1 (-ve), 55.7 (-ve), 108.8 (+ve), 115.4. Exact mass calculated for C 1 3H 2n: 176.1565; found 176.1568. 4. Experimental 189 (1 fr^^^/r.e^.y^.S^^.S-Dimethyl-e^hydroxymethyOtricycloIS.S.O.O2,5] decane(144) 143 144 To a cold (0 °C) stirred solution of BH3»THF (0.18 M in THF, 11 mL, 2 mmol) was added, via a cannula, a cold (0 °C) solution of alkene 143 (78 mg, 0.4 mmol) in pentane (3 mL). The solution was stirred at 0 °C for 3 h and then was treated with H 20 (110 uL, 6 mmol), aq NaOH (3 M, 3.3 mL, 10 mmol) and aq H 2 0 2 (8.8 M, 1.2 mL, 10 mmol). The mixture was stirred overnight. The phases were separated and the organic phase was washed (brine, 10 mL), dried (MgS04) and concentrated. Chromatographic purification of the residual oil (8 g of silica gel, 9:1 pentane-Et20) provided alcohol 144 as a colourless oil; yield: 68 mg (80%). IR (film): v = 3324, 2952, 1456, 1375, 1029 cm"1. 1H NMR (500 MHz): 8= 1.02 (d, 3 H, J = 7.2 Hz), 1.10-1.20 (m, 1 H), 1.16 (s, 3 H), 1.34 (dddd, 1 H, J = 14.0, 10.7, 9.1, 6.7 Hz), 1.38-1.46 (dddd, 1 H, J= 11.7, 8, 8, 3.8), 1.59 (dddd, 1 H, J=14.0, 9.7, 9.7, 4.4 Hz), 1.70-1.77 (m, 1 H), 1.78 (dd, 2 H, J = 8, 8 Hz), 1.98-2.04 (dddd, 1 H, J= 10.2, 7.3, 4.6, 3.1), 2.07-2.16 (m, 2 H), 2.21-2.32 (m, 2 H), 2.88 (dd, 1H, J = 7.3, 7.3, 7.3 Hz), 3.33 (dd, 1 H, J=10.2, 10.2 Hz), 3.93 (dd, 1 H, J = 10.2, 4.6 Hz). The hydroxyl proton was not observed in this spectrum. 1 3C NMR (125 MHz): 8= 16.3, 20.4, 23.2, 26.5, 34.3, 35.0, 36.5, 46.8, 50.2, 52.3, 53.5, 57.0, 64.7. Exact mass calculated for d 3 H 2 2 0 : 194.1671; found 194.1664. 4. Experimental 190 Table 4 . 3 . NMR data for (1 fl*,2S*,5/?')6S*)7S*,8S*)-2)8-dimethyl-6-(hydroxymethyl)tricyclo[5.3.0.02,5]decane (144) 11 Carbon 13 C 1H HMBC No. 5 (ppm)a 6 (ppm)(mult; J (Hz))bc'd Correlations6 1 57.0 H-1: part of the m at 2.07-2.16 (m) H-7, H-3a, H-33, H-11 2 46.8 H-3a, H-33, H-11 3 35.0 H-3a: 1.78 (dd; 8, 8) H-11 H-3p: 1.78 (dd; 8, 8) 4 20.4 H-4a: part of the m at 2.07-2.16 (m) H-3a, H-33, H-4P; 1.38-1.46 (dddd; 11.7, 8, 8, 3.8) 5 50.2 H-5: part of the m at 2.21-2.32 (m) H-3a, H-33, H-12a, H-12b 6 52.3 H-6: 1.98-2.04 (dddd; 10.2, 7.3, 4.6, 3.1) H-7, H-8, H-12a, H-12b 7 53.5 H-7: 2.88 (ddd; 7.3, 7.3, 7.3) H-13 8 36.5 H-8: part of the m at 2.21-2.32 (m) H-7, H-13 9 34.2 H-9a; 1.10-1.20 (m) H-7, H-13 H-9b: 1.70-1.77 (m) 10 26.5 H-10o: 1.34 (dddd; 14, 10.7, 9.1, 6.7) H-7 H-10P; 1.59 (dddd; 14, 9.7, 9.7, 4.4) 11 23.2 H-11: 1.16 (s) H-3ct, H-33, 12 64.7 H-12a: 3.33 (dd; 10.2, 10.2) H-7 H-12b: 3.93 (dd; 10.2, 4.6) 13 16.3 H-13: 1.02 (d; 7.2) H-7 Recorded at 125 MHz. Recorded at 500 MHz. Assignments based on HMQC data recorded at 500 MHz. Methylene protons are designated H-Xa if they are known to reside below the plane of the paper in the structure depicted above. They are designated H-Xp if they are known to reside above the plane of the paper in the structure depicted above. If no information regarding their position relative to the plane of the paper is available then they are arbitrarily designated H-Xa and H-Xb. Only those correlations which could be unambiguously assigned are recorded. 4. Experimental 191 Table 4.4. NMR data for (1 Fr,2S*,5Fr;6S*,7S*,8S*)-2,8-dimethyl-6-(hydroxymethyl)tricyclo[5.3.0.02,5]decane (144) Proton 1H COSY NOE No. 5 (ppm)(mult; J (Hz))a'b' Correlations3^ Correlations'1,6 H-1 H-1: part of the m at 2.07-2.16 (m) H-7, H-10a, H-1 Op H-3ot H-3a: 1.78 (dd; 8, 8) H-4a, H-4p H-3P H-3P: 1.78 (dd; 8, 8) H-4a, H-4p H-4a H-4a: part of the m at 2.07-2.16 (m) H-3a, H-3p, H-4p, H-5 H-4P H-4P; 1.38-1.46 (dddd; 11.7, 8, 8, 3.8) H-3a, H-3p, H-4a H-5 H-5: part of the m at 2.21-2.32 (m) H-4a H-6 H-6: 1.98-2.04 (dddd; 10.2, 7.3, 4.6, 3.1) H-7, H-12a, H-12b H-3P, H-4p, H-7, H-8, H-12a, H-12b H-7 H-7: 2.88 (ddd; 7.3, 7.3, 7.3) H-1, H-6, H-8 H-1, H-4p, H-6, H-8, H-10P H-8 H-8: part of the m at 2.21-2.32 (m) H-7, H-13 H-9a H-9a; 1.10-1.20 (m) H-9b, H-10a, H-10P H-9b H-9b: 1.70-1.77 (m) H-9a, H-10a, H-10P H-10a H-10a: 1.34 (dddd; 14, 10.7, 9.1,6.7) H-1, H-9a, H-9b, H-10P H-1 OP H-10P; 1.59 (dddd; 14, 9.7, 9.7, 4.4) H-1, H-9a, H-9b, H-10a H-11 H-11: 1.16 (s) H-12a H-12a: 3.33 (dd; 10.2, 10.2) H-6, H-12b H-5, H-6, H-10a, H-11, H-12b, H-13 H-12b H-12b: 3.93 (dd; 10.2,4.6) H-6, H-12a H-6, H-11, H-13, H-12a H-13 H-13: 1.02 (d; 7.2) H-8 Recorded at 500 MHz. Assignments based on H M Q C data recorded at 500 MHz. Methylene protons are designated H -Xa if they are known to reside below the plane of the paper in the structure depicted above. They are designated H-Xp if they are known to reside above the plane of the paper in the structure depicted above. If no information regarding their position relative to the plane of the paper is available then they are arbitrarily designated H-Xa and H-Xb. Only those correlations which could be unambiguously assigned are recorded. Recorded as N O E difference at 500 MHz using 1D selective N O E difference experiments. 4. Experimental ; 192 (1 /r^^.S/T.e^^^.S^-e-Acetyl^.S-dimethyltricycloIS.S.O.O^ldecane (rac-81) 144 rac-80 149 rac-81 To a solution of alcohol 144 (20 mg, 0.10 mmol) in DCM (10 mL) was added TPAP (54 mg, 0.155 mmol) in one portion. The solution was stirred at r.t. for 30 min and then was filtered through a pad of silica gel (1 g) and the silica gel was eluted with Et20 (20 mL). The filtrate was concentrated and the crude aldehyde (rac-80) was dissolved in Et20 (10 mL). The solution was cooled (0 °C), treated with MeLi (1.6 M solution in Et20, 0.128 mL, 0.21 mmol), and allowed to warm up to r.t. over a period of 30 min. The mixture was treated with HOAc (11 u.L, 0.21 mmol) and H 20 (5 mL). The phases were separated and the organic phase was washed (brine, 5 mL), dried (MgS04), and concentrated. The acquired crude mixture of diastereomeric secondary alcohols (149) was dissolved in DCM (10 mL). The solution was treated with TPAP (54 mg, 0.155 mmol) and then was stirred at r.t. for 30 min. The mixture was filtered through a pad of silica gel (1 g) and the silica gel was eluted with Et20 (20 mL). The filtrate was concentrated and the crude oil was purified by flash column chromatography (1 g silica gel, 20:1 pentane-Et20) to provide ketone rac-81 as a colourless oil; yield: 18 mg (86%). IR (film): v = 2946, 1707, 1459, 1353, 1234, 1184 cm"1. 1H NMR (500 MHz): 5 = 0.56 (d, 3 H, J = 7.3 Hz), 1.16 (s, 3 H), 1.27-1.38 (m, 2 H), 1.48-1.63 (m, 3 H), 1.73-1.80 (m, 1 H), 1.85-1.92 (m, 1 H), 2.19 (s, 3 H), 2.14-2.24 (m, 2 H), 2.31-2.41 (m, 2 H), 3.04-3.13 (m, 2 H). 4. Experimental 193 1 3C NMR (100 MHz, APT): 5 = 16.1 (+ve), 21.1 (+ve), 22.3 (-ve), 25.4 (-ve), 29.7 (+ve), 35.7 (+ve), 35.8 (-ve), 36.5 (-ve), 47.7 (-ve), 48.4 (+ve), 55.0 (+ve), 55.1 (+ve), 62.8 (+ve), 210.0 (-ve). Exact mass calculated forCuH 220: 206.1671; found 206.1666. 4. Experimental 194 (Ifr^^^/T.efrj^.S^-e -Acetyl-a .S-dimethyltricycloIS.S.O.O^decane (rac-70) H ; H : rac-81 rac-70 To a solution of ketone rac-81 (18 mg, 0.08 mmol) in CDCI3 (3 mL) was added 35% aq HCI04 (1 mL) and the resulting biphasic mixture was heated to reflux for 20 h. The reaction mixture was cooled to r.t. and the phases were separated. The organic phase was washed (brine, 3 mL), dried (MgS04), and concentrated. Chromatography (1 g of silica gel, 20:1 pentane-Et20) of the crude oil provided ketone rac-70 as a colourless oil; yield: 17 mg (95%). IR (film): v = 2948, 1709, 1452, 1373, 1352, 1155 cm"1. 1H NMR (500 MHz): 5 = 0.75 (d, 3 H, J = 6.9 Hz), 1.15 (s, 3 H), 1.18-1.27 (m, 1 H), 1.27-1.35 (m, 1 H), 1.35-1.42 (m, 1 H), 1.46-1.57 (m, 1H), 1.63-1.81 (m, 3 H), 1.81 -1.95 (m, 1 H), 2.02 (s, 3 H), 2.05-2.14 (m, 1 H), 2.16-2.29 (m, 1 H), 2.61 (ddd, 1 H, J = 10.7, 6, 6 Hz), 2.89 (dd, 1 H, J = 10.7, 8.1 Hz), 3.21 (ddd, 1 H, J = 8, 7, 3.8 Hz). 1 3C NMR (100 MHz): 5 = 15.3 (-ve), 17.4 (+ve), 23.2 (+ve), 26.0 (-ve), 30.0 (+ve), 32.4 (-ve), 33.0 (-ve), 35.8 (+ve), 47.1 (+ve), 47.6 (-ve), 49.8 (+ve), 55.3 (+ve), 56.8 (+ve), 208.7. Exact mass calculated for C14H22O: 206.1671; found 206.1663. 4. Experimental : 195 (1 /7*,2S*,5/T,6/T,7S*3S*)-6-lsopropen^ [(±)-kelsoene] (rac-35) H H I H H To a cold (0 °C), stirred suspension of Lombardo's reagent85,86 in THF (1 mL, -0.35 mmol) was added, sequentially, DCM (5 mL) and, via a cannula, a solution of ketone rac-70 (11 mg, 0.05 mmol) in DCM (2 mL). The mixture was allowed to warm up to r.t. and then was stirred for 6 h. Sat. aq NaHC03 (3 mL) was added and the resulting biphasic mixture was filtered through a pad of neutral alumina (1 g). The alumina was eluted with Et20 (10 mL). The phases of the combined eluate were separated and the organic phase was washed (brine, 5 mL), dried (MgS04), and concentrated. Purification of the crude oil by chromatography (1 g of silica gel, pentane) provided (±)-kelsoene (rac-35) as a colourless oil; yield: 9 mg (85%). IR (film): v = 3083, 2948, 2869, 1647, 1452, 1374, 886 cm"1. 1H NMR (500 MHz): 8 = 0.89 (d, 3 H, J = 7.0 Hz), 1.14 (s, 3 H), 1.24-1.40 (m, 2 H), 1.43-1.54 (m, 2 H), 1.59 (s, 3 H), 1.61-1.69 (m, 2 H), 1.69-1.81 (m, 2 H), 2.06 (ddd, 1 H, J = 9, 9, 6.8 Hz), 2.24 (m, 1 H), 2.31-2.41 (m, 2 H), 2.84 (ddd, 1 H, J=10.4, 7. 7 Hz), 4.77 (s, 1 H), 4.84 (s, 1 H). 1 3C NMR (100 MHz, APT): 8= 14.6 (-ve), 17.7 (+ve), 23.5 (+ve), 24.1 (+ve), 26.0 (-ve), 33.1 (-ve), 33.3 (-ve), 36.3 (+ve), 45.8, 47.5 (+ve), 48.1 (+ve), 50.0 (+ve), 57.8 (+ve), 109.8 (-ve), 145.6. Exact mass calculated for Ci 5 H 2 4 : 204.1878; found 204.1870. 4 . Experimental 196 4.3 Experimental - Synthetic Approaches to (±)-lsomulinic Acid Methyl (1 S*,5fl*)-5-isopropyl-2-oxo-1 -(3-oxobutyl)cyclopentane-1 -carboxylate (199) and Methyl (1 S*,3S*,5/?*)-5-isopropyl-2-oxo-1,3-bis(3-oxobutyl)cyclopentane-1 -carboxylate (200) 198 199 Compound 199 was prepared using the method developed by Christoffers.116 Thus, to a stirred mixture of neat keto ester 198 (30.0 g, 162 mmol) and freshly distilled methyl vinyl ketone (12.5 g, 178 mmol) was added a catalytic amount of FeCla^HaO (440 mg, 1.6 mmol). Upon addition of FeCl3»6H20 the mixture turned deep purple in colour. The mixture was allowed to stir under an atmosphere of air. After 2 days, the mixture was filtered through a pad of silica gel and the silica gel was rinsed with Et20. The eluate was concentrated and the crude material was purified by flash column chromatography (700 g of silica gel, 3:1 petroleum ether-Et20) to yield compound 199 as a clear oil (25.8 g, 63%). Keto ester 198 (5.5 g, 18%) and trione 200 (2.6 g, 5%) were also isolated as clear oils from the crude mixture. Compound 199 exhibited: IR (film): v = 2958, 1749, 1731, 1718, 1439, 1360, 1229, 1162 cm"1. 4 . Experimental 197 1H NMR (400 MHz): 5 = 0.82 (d, 3 H, J = 6.6 Hz, Me-12 or Me-13), 0.91 (d, 3 H, J = 6.6 Hz, Me-12 or Me-13), 1.45-1.58 (m, 1 H, H-11), 1.67-1.86 (m, 2 H), 1.89-1.99 (ddd, 1 H, J= 14.4, 9.5, 4.7 Hz), 2.01-2.22 (m, 2 H), 2.08. (s, 3 H, Me-10), 2.26-2.50 (m, 3 H), 2.61-2.72 (ddd, 1 H, J = 17.3, 10.0, 5.9 Hz), 3.63 (s, 3 H, Me-1 V 1 3C NMR (100 MHz): 8 = 20.9, 22.2, 24.9, 28.0, 29.8, 30.5, 38.6, 38.6, 51.8, 53.4, 61.5 (C-1), 171.0 (C-6), 207.9 (C-9), 216.5 (C-2). Exact mass calculated for C14H22O4: 254.1518; found 254.1518. Compound 200 exhibited: 16 200 IR (film): v = 2956, 1747, 1730, 1716, 1225, 1161 cm"1. 1H NMR (400 MHz): 8 = 0.86 (d, 3 H, J= 6.6 Hz, Me-16 or Me-17), 0.94 (d, 3 H, J= 6.6 Hz, Me-16 or Me-17), 1.30-1.40 (m, 1 H), 1.44-1.54 (m, 1 H), 1.61 -1.73 (dddd, 1 H, J-11.8, 11.8, 11.8, 7.6 Hz), 1.75-1.91 (m, 2 H), 2.00-2.06 (dd, 1 H, J= 14.4, 1.6 Hz), 1.99-2.10 (m, 1 H), 2.11-2.26 (m, 2 H), 2.13 (s, 3 H, Me-10 or Me-14), 2.17 (s, 3 H, Me-10 or Me-14), 2.32-2.49 (m, 2 H), 2.56-2.68 (ddd, 1 H, J= 18.2, 7.5, 7.5 Hz), 2.67-2.74 (m, 1 H), 3.62 (s, 3 H, Me-1'). 1 3C NMR (100 MHz): 8 = 20.7, 22.2, 25.1, 25.9, 28.7, 29.9, 30.2, 34.7, 38.8, 39.8, 47.2, 51.7, 55.2, 61.4 (C-1), 171.4 (C-6), 207.8 (C-9 or C-13), 210.7 (C-9 or C-13), 216.6 (C-2). Exact mass calculated for Ci8H2805: 324.1937; found 324.1939. 4 . Experimental 198 Methyl (6S*,7fl*)-7-isopropyl-3-oxobicyclo[4.3.0]non-1 -ene-6-carboxylate (197) 199 207 197 Enone 197 was prepared using the method described by Scanio and Hill.1 1 8 Thus, a 500 mL round-bottom flask equipped with a Dean-Stark water separator and a condenser was charged with a solution of diketone 199 (25.8 g, 102 mmol) and pyrrolidine (14.5 g, 204 mmol) in benzene (250 mL). The mixture was stirred and heated to reflux for 3 h under an atmosphere of argon. After that time the mixture was concentrated to provide the crude dienamine 207 as a brown oil. A 250 mL round-bottom flask equipped with a condenser was charged with a benzene (70 mL) solution of crude dienamine 207. A solution of sodium acetate (1.9 g, 23 mmol) and glacial acetic acid (3.8 mL) in water (3.8 mL) was then added and the resulting biphasic mixture was heated to reflux for 4 h. After this time the mixture was washed successively with water (100 mL), 10% aq. HCI (100 mL), sat. aq. NaHC03 (100 mL) and brine (100 mL). The organic phase was dried (MgS04) and concentrated to provide crude enone 197 as a yellow oil which crystallized on standing. The crude material was recrystallized from hot DCM to furnish 18.1 g (76%) of pure enone 197 which exhibited: m.p. = 83-84 °C IR (film): v = 2967, 1721, 1664, 1333, 1229, 1168 cm"1. 4. Experimental 199 1H NMR (400 MHz): 5 = 0.88 (d, 3 H, J = 6.4 Hz, Me-12or Me-13), 1.03 (d, 3 H, J = 6.4 Hz, Me-12 or Me-13), 1.44-1.62 (m, 2 H), 1.63-1.79 (m, 2 H), 1.95-2.05 (m, 1 H), 2.28-2.37 (ddd, 1 H, J = 16.1, 4.3, 1.8 Hz), 2.37-2.43 (dd, 1 H, J=13.6, 4.6 Hz), 2.43-2.55 (m, 1 H), 2.78-2.89 (dddd, 1 H, 19.6, 11.0, 2.2, 2.2 Hz), 2.86-2.93 (ddd, 1 H, J = 13.1, 4.6, 2.5 Hz), 3.68 (s, 3 H, Me-1'), 5.81 (s, 1 H, H-2). 1 3C NMR (100 MHz): 8 = 22.1, 22.4, 27.6, 30.6, 30.9, 34.1, 34.7, 52.0, 56.8, 58.5, 123.4 (C-2), 171.2 (C-1 orC10), 172.3 (C-1 or C-10), 198.8 (C-3). Exact mass calculated for C14H20O3: 236.1412; found 236.1412. 4. Experimental 200 Methyl (6S*,7R*)-7-isopropyl-3-oxobicyclo[4.3.0]non-1-ene-6-carboxylate-3,3-ethylenedithioketal (208) 197 1 3 208 To a stirred solution of enone 197 (20.4 g, 86.4 mmol) in DCM (280 mL) under an atmosphere of argon were added, sequentially, 1,2-ethanedithiol (24.4 g, 21.7 mL, 105 mmol) and neat BF3«Et20 (1.23 g, 1.10 mL, 8.64 mmol). The resulting mixture was allowed to stir for 36 h. The mixture was then diluted with Et20 (250 mL), washed successively with tris(hydroxy-methyl)aminomethane buffer (pH = 8, 0.5 M, 250 mL) and brine (250 mL), dried (MgSCv) and concentrated. The residual oil was purified by flash chromatography (500 g of silica gel, 19:1 petroleum ether-Et20) to yield 25.0 g (93%) of dithioketal 208 as a clear oil. IR (film): v = 2948, 1718, 1436, 1169, 739 cm"1. 1H NMR (400 MHz): 5 = 0.81 (d, 3 H, J = 6.2 Hz, Me-12 or Me-13), 0.97 (d, 3 H, J = 6.2 Hz, Me-12 or Me-13), 1.32-1.59 (m, 4 H), 1.81-1.91 (m, 1 H), 1.98-2.08 (ddd, 1 H, J = 13.7, 13.7, 2.5 Hz), 2.18-2.29 (m, 2 H), 2.57-2.68 (m, 1 H), 2.69-2.78 (ddd, 1 H, 13.2, 4.2, 2.6 Hz), 3.12-3.20 (m, 1 H), 3.25-3.40 (m, 3 H), 3.62 (s, 3 H, Me-1"), 5.51-5.55 (ddd, 1 H, J = 1.5, 1.5, 1.5 Hz, H-2). 1 3C NMR (100 MHz): 5 = 22.2 (-ve, C-12 or C-13), 22.3 (-ve, C-12 or C-13), 27.6 (+ve), 29.5 (+ve), 31.3 (-ve), 34.1 (+ve), 39.4 (+ve), 40.1 (+ve), 40.3 (+ve), 51.4 (-ve), 55.4 (+ve), 58.6 (-ve), 65.3 (+ve), 125.0 (-ve, C-2), 145.5 (+ve, C-1), 173.6 (+ve, C-10). Exact mass calculated for C i 6 H 2 4 0 2 S 2 : 312.1218; found: 312.1216. 4. Experimental 2 0 1 Methyl (6S*,7/?*)-7-isopropylbicyclo[4.3.0]non-1 -ene-6-carboxylate (196) A 500 mL round bottom flask equipped with an overhead mechanical stirrer was charged with a solution of dithioketal 208 (25.0 g, 80.1 mmol) in EtOH (200 mL). Excess Raney nickel (W-2, 50% in H20, 150 g) was added in portions over a span of 2 days until no starting material could be detected by TLC analysis. 10% aq. HCI (50 mL) was added and the reaction mixture was stirred overnight. The mixture was filtered ® through a pad of Celite and the remaining solid was rinsed with petroleum ether until no product was detected in the eluate by TLC analysis. The phases of the eluate were allowed to separate and the petroleum ether phase was washed with brine (2 x 200 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (300 g of silica gel, 19:1 petroleum ether-Et20) to yield 10.3 g (66%) of alkene 196 as a clear oil. IR (film): v = 2948, 1724, 1433, 1164, 1008, 795 cm"1. 1H NMR (400 MHz): 5 = 0.84 (d, 3 H, J = 6.4 Hz, Me-12 or Me-13), 0.98-1.06 (m, 1 H, H-5a), 1.01 (d, 3 H, J=6.3 Hz, Me-12 or Me-13), 1.30-1.59 (m, 4 H), 1.64-1.75 (m, 1 H), 1.79-1.89 (m, 1 H), 1.91-2.04 (m, 2 H), 2.17-2.29 (m, 1 H), 2.56-2.68 (m, 1 H), 2.68-2.74 (ddd, 1 H, J = 12.6, 3.6, 3.1 Hz, H-5p), 3.63 (s, 3 H, Me-1'), 5.44 (br. s, 1 H, H-2). 1 3C NMR (100 MHz): 5 = 20.1 (+ve), 22.2 (-ve, C-12 orC-13), 22.6 (-ve, C-12 orC-13), 24.7 (+ve), 27.3 (+ve), 29.7 (+ve), 31.3 (-ve, C-7 or C-11), 33.4 (+ve), 51.2 (-ve, C-7 or C-11), 55.4 (+ve, C-6), 59.1 (-ve, C-1'), 120.5 (-ve, C-2), 143.8 (+ve, C-1), 174.9 (+ve, C-10). Exact mass calculated for C14H22O2: 222.1620; found: 222.1622. 13 208 196 4. Experimental 202 (6S*,7R*)-6-hydroxymethyl-7-isopropylbicyclo[4.3.0]non-1 -ene (210) C0 2 Me 196 12 13 210 To a cold (0 °C), stirred solution of ester 196 (3.16 g, 14.2 mmol) in Et20 (100 mL) was added a solution of DIBAL-H in hexanes (1.0 M, 31.3 mL, 31.3 mmol). The reaction mixture was stirred at this temperature until no starting material could be detected by TLC analysis. Excess MgS04»7H20 (8.0 g, 32.5 mmol) was added and the resulting heterogeneous mixture was stirred vigorously for 1 h. The mixture was then filtered through a sintered glass funnel and the funnel was rinsed with Et20 until no product could be detected in the eluate by TLC analysis. The collected eluate was concentrated and the residual oil was chromatographed (150 g of silica gel, 9:1 petroleum ether-Et20) to furnish alcohol 210 as a clear and colourless oil; yield: 2.35 g (85%). IR (film): v = 3369, 2932, 1473, 1034 cm"1. 1H NMR (400 MHz): 5 = 0.86 (d, 3 H, J = 6.6 Hz, Me-12 or Me-13), 0.99 (d, 3 H, J = 6.6 Hz, Me-12 or Me-13), 1.14-1.28 (m, 3 H), 1.34-1.47 (dddd, 1 H, J = 11.6, 11.6, 9.9, 6.7 Hz), 1.56-1.67 (m, 2 H), 1.77-1.91 (m, 2 H), 1.93-2.00 (m, 2 H), 2.11-2.20 (dddd, 1 H, 16.3, 8.3, 3.3, 1.6 Hz), 2.18-2.25 (ddd, 1 H, J = 13.1, 8, 8 Hz), 2.27-2.38 (m, 1 H), 3.54-3.59 (br. s, 2 H, H-10), 5.54-5.59 (m, 1 H, H-2). 1 3C NMR (100 MHz): 8= 19.2 (+ve), 23.1 (-ve, C-12), 23.1 (-ve, C-13), 24.5 (+ve), 27.3 (+ve), 28.3 (+ve), 30.2 (-ve), 34.7 (+ve), 48.0 (+ve, C-6), 58.3 (-ve), 64.9 (+ve, C-10), 121.9 (-ve, C-2), 144.0 (+ve, C-1). Exact mass calculated forCi 3H 2 20: 194.1671; found: 194.1676. 4. Experimental 203 (6S*,7R*)-7-lsopropylbicyclo[4.3.0]non-1 -ene-6-carbaldehyde (211) 12 13 210 211 Alcohol 210 was oxidized to aldehyde 211 using the method developed by Ley. Thus, to a solution of alcohol 210 (6.82 g, 35.2 mmol) in DCM (250 mL) were added, sequentially, NMO (4.94 g, 42.2 mmol) in one portion and TPAP (475 mg, 1.4 mmol) in one portion. The resulting mixture was stirred overnight under an atmosphere of argon. TLC analysis of the mixture indicated that most of the starting material had been consumed. The reaction mixture was filtered through a pad of silica gel and the silica gel was rinsed with Et20. The eluate was concentrated and the residual oil was purified by flash chromatography (350 g of silica gel, 19:1 petroleum ether-Et20) to furnish aldehyde 211 as a colourless oil; yield: 6.12 g (91%). IR (film): v = 2942, 1718, 1472, 1368, 880 cm"1. 1H NMR (400 MHz): 5 = 0.86 (d, 3H,J= 6.6 Hz, Me-12 or Me-13), 0.96 (d, 3 H, J= 6.6 Hz, Me-12 or Me-13), 1.02-1.11 (ddd, 1 H, J= 13.7, 13.7, 3.2 Hz, H-5cc), 1.32-1.46 (m, 2 H), 1.49-1.46 (dddd, 1 H, J= 13.0, 11.5, 11.5, 7.0 Hz), 1.58-1.72 (m, 2 H), 1.89-1.96 (m, 2 H), 1.94-2.04 (dddd, 1 H, J= 13.3, 9.8, 8.2, 3.8 Hz), 2.29-2.40 (m, 1 H), 2.52-2.63 (m, 1 H), 2.58-2.66 (ddd, 1 H, J = 13.0, 3.5, 3.5 Hz, H-5p), 5.56-5.62 (m, 1 H, H-2), 9.65 (s, 1 H, H-10). 1 3C NMR (100 MHz): 5=19.9 (+ve), 22.5 (-ve, C-12 or C-13), 22.6 (-ve, C-12 or C-13), 24.5 (+ve), 27.5 (+ve), 28.8 (+ve), 30.3 (+ve), 30.7 (-ve, C-7 or C-11), 59.0 (-ve, C-7 or C-11), 60.0 (+ve, C-6), 122.7 (-ve, C-2), 141.1 (+ve, C-1), 203.2 (-ve, C-10). Exact mass calculated for Ci 3H 2 0O: 192.1514; found 192.1515. 4. Experimental 204 (6S*,7/7*)-7-lsopropylbicyclo[4.3.0]non-1-ene-6-carbaldehyde 2',2'-dimethyl-propylene acetal (195) To a solution of aldehyde 211 (6.12 g, 22.0 mmol) in DCM (250 mL) at r.t. was added 2,2-dimethyl-1,3-propanediol (22.9 g, 220 mmol). A catalytic amount of p-TSA»H20 (210 mg, 1.1 mmol) was added and the reaction mixture was allowed to stir overnight. TLC analysis of the reaction mixture indicated that the starting material had been consumed. The mixture was diluted with Et20 (30 mL), washed successively with water (2 x 30 mL) and brine (30 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (450 g of silica gel, 49:1 petroleum ether-Et20) to furnish acetal 195 as a colourless oil; yield: 8.47 g (95%). IR (film): v = 2952, 1471, 1393, 1116, 1019 cm"1. 1H NMR (400 MHz): 8 = 0.67 (s, 3 H, Me-4' or Me-5'), 0.83 (d, 3 H, J = 6.7 Hz, Me-12 or Me-13), 0.82-0.98 (m, 1 H, H-5ot), 0.98 (d, 3 H, J = 6.4 Hz, Me-12 or Me-13), 1.05-1.14 (m, 1 H), 1.16 (s, 3 H, Me-4' or Me-5'), 1.38-1.51 (ddd, 1 H, J= 12, 12, 6.0 Hz), 1.52-1.60 (m, 1 H), 1.63-1.74 (dddd, 1 H, J= 13.1, 9,9, 6.5, 6.5 Hz), 1.76-1.87 (dddd, 1 H, 12.9, 9.9, 8.8, 4.3 Hz), 1.87-2.03 (m, 3 H), 2.08-2.17 (dddd, 1 H, J=15.9, 8.8, 6.2, 1.2 Hz), 2.26-2.38 (m, 1 H), 2.57-2.64 (ddd, 1 H, J= 12.9, 3, 3 Hz, H-5(3), 3.28 (d, 1 H, J = 10.7 Hz, H-1'ax or H-3'ax), 3.36 (d, 1 H, J = 10.7 Hz, H-1'ax or H-3'ax), 3.57-3.62 (dd, 1 H, J=10.7, 2.9 Hz, H-1'eq or H-3'eq), 3.60-3.65 (dd, 1 H, J=10.7, 2.9 Hz, H-1'eq, or H-3'eq), 4.33 (s, 1 H, H-10), 5.47 (d, 1 H, 1.5 Hz, H-2). 4. Experimental 205 1 dC NMR (100 MHz): 5= 19.8 (+ve), 21.8 (-ve), 23.0 (-ve), 23.2 (-ve), 23.2 (-ve), 24.8 (+ve), 27.6 (+ve), 28.8 (+ve), 30.0 (+ve, C-2'), 30.5 (-ve), 31.6 (+ve), 49.9 (+ve), 59.2 (-ve), 77.4 (+ve, C-1' or C-3'), 78.2 (+ve, C-1' or C-3'), 103.4 (-ve, C-10), 119.5 (-ve, C-2), 144.6 (+ve, C-1). Exact mass calculated for Ci8H3o02: 278.2246; found: 278.2247. 4. Experimental 206 Hydroboration of alkene 195, preparation of (1S*,5R*,6R*,9R*)-5-Hydroxy-9-isopropylbicyclo[4.3.0]nonane-1-carbaldehyde-2',2'-dimethylpropylene acetal (194), (1S*,5S*,6S*,9R*)-5-Hydroxy-9-isopropylbicyclo[4.3.0]nonane-1-carbaldehyde-2',2'-dimethylpropylene acetal (213) and (1S*,5S*,6S*,9r7*)-1-(3'-hydroxy-2',2'-dimethylpropoxymethyl)-5-hydroxy-9-isopropylbicyclo[4.3.0]nonane (214) OH OH OH To a stirred solution of alkene 195 (262 mg, 0.94 mmol) in DCM (54 mL) was added, via a syringe, neat BH3«SMe2 (134 u,L, 1.41 mmol, 1.5 equiv.). The resulting solution was stirred at room temperature for 24 h. TLC analysis of the solution indicated that all the starting material had been consumed. The reaction mixture was then treated with H 20 (1 mL, 55.6 mmol), aq NaOH (3 M, 2.83 mL, 8.5 mmol) and aq H 2 0 2 (8.8 M, 0.96 mL, 8.5 mmol) and was allowed to stir overnight. Neat AcOH (0.49 mL, 8.5 mmol) was added. The phases were separated and the organic phase was washed with brine (30 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (25 g of silica gel, 1:1 petroleum ether-Et20) to yield a mixture of alcohols 194 and 213 (ratio 1.5:1, 219 mg, 0.74 mmol) as a clear oil and in 79% combined yield. This mixture of alcohols 194 and 213 proved difficult to separate. Elution of the baseline material with Et20 provided diol 214 (53 mg, 0.18 mmol) as a clear oil in 19% yield. 4. Experimental 207 (1 S^^/^je/^.S/TJ-S-Hydroxy-S-isopropylbicyclo^.S.Olnonane-l-carbaldehyde-2',2'-dimethylpropylene acetal (194) A small amount (30 mg) of alcohol 194 could be separated from the previously obtained mixture of alcohols 194 and 213 by flash column chromatography (25 g of silica gel, DCM). Alcohol 194 displayed: 1H NMR (400 MHz): 5 = 0.62 (s, 3 H), 0.74-0.83 (m, 1 H), 0.78 (d, 3 H, J= 6.7 Hz), 0.92 (d, 3 H, J=6.7 Hz), 0.96-1.03 (m, 1 H), 1.02-1.10 (ddd, 1 H, J= 10, 10, 10 Hz), 1.12 (s, 3 H), 1.23-1.48 (m, 4 H), 1.51-1.62 (m, 2 H), 1.74-1.91 (m, 3 H), 1.96-2.04 (m, 1 H), 2.41-2.50 (ddd, 1 H, J= 13.0, 3.0, 3.0 Hz), 3.20-3.26 (ddd, 1 H, J= 10.2, 10.2, 5.0 Hz), 3.26-3.32 (d, 1 H, J= 11.2 Hz), 3.52-3.58 (d, 1 H, J= 11.2 Hz), 3.55-3.60 (dd, 1 H, J = 11.2, 2.7 Hz), 4.10-4.17 (dd, 1 H, J= 11.2, 2.7 Hz), 4.25 (s, 1 H). 1 3C NMR (100 MHz): 5 = 21.9 (-ve), 22.8 (-ve), 22.8 (+ve), 23.0 (+ve), 23.3 (-ve), 23.5 (-ve), 28.7 (+ve), 29.9 (+ve), 31.2 (-ve), 34.4 (+ve), 35.9 (+ve), 51.6 (+ve), 58.2 (-ve), 59.7 (-ve), 70.1 (-ve), 77.3 (+ve), 77.9 (+ve), 104.1 (-ve). OH 4. Experimental 208 Table 4.5: NMR data for (1 S*,5Ff\6Fr,9Fr)-5-hydroxy-9-isopropylbicyclo[4.3.0]nonane-1-carbaldehyde-2',2,-dimethylpropylene acetal (194) 4' Carbon No. 13Q 5 (ppm)a 1H 5 (ppm) (mult; J(Hz))bc'd HMBC Correlations8 1 51.6 H-2ax, H-2eq, H-9 2 34.4 H-2ax: 0.74-0.83 (m) H-4ax, H-4eq, H-9, H-10 H-2eq: 2.41-2.50 (ddd, J= 13.0, 3.0, 3.0) 3 22.8 H-3eq: part of the m at 1.51 -1.62 H-2ax, H-2eq, H-4eq H-3ax: part of the m at 1.74-1.91 4 35.9 H-4ax: 0.96-1.03 (m) H-5 H-4eq: 1.96-2.04 (m) 5 70.1 H-5: 4.10-4.17 (ddd, J= 10, 10, 5.0) H-4ax, H-4eq 6 58.2 H-6: part of the m at 1.23-1.48 H-2ax, H-2eq, H-4ax, H-4eq, H-5 7 23.0 H-7a: see footnote f H-2ax, H-2eq H-7b: see footnote f 8 28.7 H-8a: part of the m at 1.23-1.48 H-9 H-8b: part of the m at 1.74-1.91 9 59.7 H-9: 1.02-1.10 (ddd, J= 10, 10, 10) H-10, H-12, H-13 10 104.1 H-10: 4.25 (s) H-2ax, H-2eq, H-1'ax, H-1'eq, H-3'ax, H-3'eq 11 31.2 H-11: part of the m at 1.51-1.62 H-9, H-12, H-13 12 23.3 H-12: 0.78 (d, J=6.7) H-13 13 22.8 H-13: 0.92 (d, J=6.7) H-12 V 77.3 H-1'ax: 3.26-3.32 (d, J= 11.2) H-10, H-3'ax, H-3'eq, H-4', H-5' H-1 'eq: 3.51 -3.56 (dd, J = 11.2, 2.7) 2' 29.9 H-10, H-1'ax, H-1'eq, H-3'ax, H-3'eq, H-4', H-5' 3' 77.9 H-3'ax: 3.20-3.26 (d, J= 11.2) H-10, H-1'ax, H-1'eq, H-4', H-5' H-3'eq: 3.55-3.60 (dd, J= 11.2, 2.7) 4' 21.9 H-4': 0.62 (s) H-1'ax, H-3'ax, H-5' 5' 23.5 H-5': 1.12 (s) H-1'ax, H-1'eq, H-3'ax, H-3'eq, H-4' Recorded at 100 MHz. Recorded at 400 MHz. Assignments based on HMQC and JMOD data. Methylene protons are designated H-Xax and H-Xeq if they are known to occupy axial and equatorial positions, respectively, in the conformation depicted above. If no information regarding their relative position is known then they are designated H-Xa and H-Xb arbitrarily. Only those correlations which could be unambiguously assigned are recorded. One of the proton signals which occur at 8 = 1.74-1.91 and two of the proton signals which occur at 8 = 1.23-1.48 remain unassigned. They account for the methylene protons on C-7 and the hydroxyl group proton. 4. Experimental 209 Table 4.6: NMR data for (1 S*,5rT,6rT,9rT)-5-hydroxy-9-isopropylbicyclo[4.3.0]nonane-1 -carbaldehyde-2',2,-dimethylpropylene acetal (194) 4' Proton No 1H 5 (ppm) (mult; J(Hz))abc COSY Correlations'3 NOE Correlations^8 H-2ax 0.74-0.83 (m) H-2eq, H-3ax, H-3eq H-2eq 2.41-2.50 (ddd, J= 13.0, 3.0, 3.0) H-2ax, H-3ax, H-3eq H-2ax, H-3ax, H-3eq, H-10, H-13 H-3eq part of the m at 1.51-1.62 H-2ax, H-2eq, H-4ax, H-4eq H-3ax part of the m at 1.74-1.91 H-2ax, H-2eq, H-4ax, H-4eq H-4ax 0.96-1.03 (m) H-3ax, H-3eq, H-4eq, H-5 H-4eq 1.96-2.04 (m) H-3ax, H-3eq, H-4ax, H-5 H-4ax, H-5 H-5 4.10-4.17 (ddd, J= 10, 10, 5.0) H-4ax, H-4eq, H-6 H-3ax, H-4eq, H-10 H-6 part of the m at 1.23-1.48 H-6 H-7a see footnote f H-7b see footnote f H-8a part of the m at 1.23-1.48 H-9 H-8b part of the m at 1.74-1.91 H-9 H-9 1.02-1.10 (ddd, J= 10, 10, 10) H-8a, H-8b, H-11 H-10 4.25 (s) H-2eq, H-3ax, H-5, H-1'ax, H-3'ax H-11 part of the m at 1.51-1.62 H-9, H-12, H-13 H-12 0.78 (d, J=6.7) H-11 H-13 0.92 (d, J=6.7) H-11 H-1'ax 3.26-3.32 (d, J= 11.2) H-1'eq H-1'eq 3.51-3.56 (dd, J= 11.2, 2.7) H-1'ax, H-3'eq H-3'ax 3.20-3.26 (d, J= 11.2) H-3'eq H-3'eq 3.55-3.60 (dd, J= 11.2, 2.7) H-3'ax, H-1'eq* H-4' 0.62 (s) H-5' 1.12 (s) Recorded at 400 MHz. Assignments based on HMQC and JMOD data. Methylene protons are designated H-Xax and H-Xeq if they are known to occupy axial and equatorial positions, respectively, in the conformation depicted above. If no information regarding their relative position is known then they are designated H-Xa and H-Xb arbitrarily. Only those correlations which could be unambiguously assigned are recorded. Recorded as NOE difference at 400 MHz using 1D selective NOE experiments. One of the proton signals which occur at 8 = 1.74-1.91 and two of the proton signals which occur at 8 = 1.23-1.48 remain unassigned. They account for the methylene protons on C-7 and the hydroxyl group proton. 4. Experimental 210 (1 S*,5S*,6S*,9/^)-1-(3'-Hydroxy-2',2'-dimethylpropoxymethyl)-5-hydroxy-9-isopropylbicyclo[4.3.0]nonane (214) IR (film): v = 3370, 2949, 1460, 1111, 1042, 739 cm"1. 1H NMR (400 MHz): 5 = 0.80 (d, 3 H, J=6.6 Hz, Me-12 or Me-13), 0.88 (d, 3 H, J=6.5 Hz, Me-12 or Me-13), 1.20-1.37 (m, 1 H), 1.37-1.48 (m, 1 H), 1.55-1.68 (dddd, 1 H, J = 11.6, 11.6, 11.6, 7.6 Hz), 1.68-1.82 (m, 2 H), 1.92-2.20 (m, 7 H), 2.05 (s, 3 H, Me-4' or Me-5'), 2.09 (s, 3 H, Me-4' or Me-5'), 2.24-2.42 (m, 3 H), 2.49-2.60 (ddd, 1 H, J= 18.4, 7.5, 7.5 Hz), 2.59-2.66 (dddd, 1 H, J= 8.9, 8.9, 4.0, 1.9 Hz), 3.54-3.60 (m, 3 H). The hydroxyl protons were not observed in this spectrum. 1 3C NMR (100 MHz): 5 = 19.0, 21.9, 21.9, 22.6, 23.5, 24.9, 27.0, 29.2, 29.8, 33.5, 36.3, 49.1, 49.4, 50.9, 71.0, 71.3, 75.4, 80.4. Exact mass calculated for Ci8H3503[M-i-H]+: 299.2587; found 299.2587. OH 4. Experimental 211 (1 S*,6S*,9R*)-1-(2^2'-dimethyl-3-oxopropoxymethyl)-9-isopropyl-5-oxobicyclo-[4.3.0]nonane (215) 13 4 . 5 ' 214 215 Diol 214 was oxidized to the keto aldehyde 215 using the method developed by Ley.88 Thus, to a stirred solution of diol 214 (53 mg, 0.18 mmol) in DCM (10 mL) at r.t. was added, sequentially, NMO (31 mg, 0.27 mmol) in one portion and TPAP (10 mg, 0.03 mmol) in one portion. The resulting mixture was stirred for 30 min. TLC analysis of the mixture indicated that the starting material had been consumed. The reaction mixture was filtered through a pad of silica gel (1 g) and the silica gel was rinsed with Et20. The collected eluate was concentrated and the residual oil was purified by flash column chromatography (5 g of silica gel, 1:1 petroleum ether-Et20) to afford keto aldehyde 215 as a colourless oil. Yield: 47 mg (90%). IR (film): v = 2956, 2872, 1729, 1707,1471, 1114, 738 cm"1. 1H NMR (400-MHz): 8 = 0.78 (d, 3 H, J= 6.7 Hz, Me-12 or Me-13), 0.89 (d, 3 H, J= 6.3 Hz, Me-12 or Me-13), 1.03 (s, 6 H, Me-4' and Me-5'), 1.22-1.34 (m, 2 H), 1.46-1.62 (m, 3 H), 1.68-1.94 (m, 4 H), 1.98-2.09 (m, 1 H), 2.14-2.30 (m, 2 H), 2.61-2.65 (dd, 1 H, J = 8.3, 4.9 Hz), 3.23 (d, 1 H, J= 8.8 Hz), 3.29 (d, 1 H, J= 8.8 Hz), 3.24 (d, 1 H, J= 8.9 Hz), 3.40 (d, 1 H, J= 8.9 Hz), 9.51 (s, 1 H, H-10). 1 3C NMR (100 MHz): 8= 19.0 (2 C, C-4' and C-5'), 20.9, 22.1, 22.5, 23.2, 28.1, 28.7, 31.7, 39.4, 47.2 (C-1), 52.0 (C-2'), 52.6, 55.9, 75.1 (C-10 or C-1'), 76.4 (C-10 or C-1'), 204.9 (C-3'), 213.8 (C-5). Exact mass calculated for Ci8H3i03[M+H]+: 295.2274; found 295.2278. r 4. Experimental 212 Optimized conditions for the hydroboration of alkene 195 OH OH OH 195 r ^ X ^ O H Small scale: To a cold (0 °C), stirred solution of alkene 195 (65 mg, 0.23 mmol) in DCM (5 mL) was added, via a syringe, neat BH 3»SMe 2 (23 mg, 0.30 mmol). The reaction mixture was allowed to stir at this temperature for 6 h. After this time, H 2 0 (250 u.L, 13.9 mmol) was added and the mixture was allowed to stir for 10 min. The reaction mixture was then treated with aq NaOH (3 M, 700 uL, 2.1 mmol) and aq H 2 0 2 (8.8 M, 250 uL, 2.2 mmol). The mixture was allowed to stir overnight. Neat AcOH (120 uL, 2.1 mmol) was added. The phases were separated and the organic phase was washed with brine (10 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (5 g of silica gel, 1:1 petroleum ether-Et20) to yield an inseparable mixture of alcohols 194 and 213 (36 mg, ratio 2:1, 52%, 96% based on recovered starting material) as a clear oil. Alkene 195 (30 mg, 46%) was also recovered as a clear oil. Large scale: To a cold (0 °C), stirred solution of alkene 195 (2.16 g, 7.77 mmol) in DCM (167 mL) was added, via a syringe, neat BH 3»SMe 2 (1.18 mg, 15.5 mmol). The reaction mixture was allowed to stir at this temperature for 2 days. After this time, H 2 0 (7 mL, 390 mmol) was added and the reaction mixture was allowed to stir for 1 h. The mixture was then treated with aq NaOH (3 M, 20.7 mL, 62 mmol) and aq H 2 0 2 (8.8 M, 7.1 mL, 62 mmol). The mixture was allowed to stir overnight. Neat AcOH (3.9 mL, 62 mmol) was added. The phases were separated and the organic phase was washed with brine (100 mL), 4. Experimental 213 dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (100 g of silica gel, 1:1 petroleum ether-Et20) to yield an inseparable mixture of alcohols 194 and 213 (1.02 g, ratio 2:1, 44%, 69% based on recovered starting material) as a clear oil. Alkene 195 (777 mg, 2.80 mmol, 36%) was also recovered as a clear oil. In addition, elution of the column with Et20 provided diol 214 (349 mg, 1.17 mmol, 15%) as a clear oil. 4. Experimental 214 Oxidation of alcohols 194 and 213, preparation of (1 S*,6/?*,9/?*)-9-lsopropyl-5-oxobicyclo[4.3.0]nonane-1-carbaldehyde-2',2'-dimethylpropylene acetal (193) and (1S*,6S*,9/?*)-9-lsopropyl-5-oxobicyclo[4.3.0]nonane-1-carbaldehyde-2',2'-dimethylpropylene acetal (220) 194 213 193 220 Alcohols 194 and 213 were oxidized to ketones 193 and 220, respectively, using the method developed by Ley.88 To a stirred solution of a mixture of alcohols 194 and 213 (ratio 1.5:1, 219 mg, 0.74 mmol) in DCM (15 mL) at r.t. was added, sequentially, NMO (130 mg, 1.11 mmol) in one portion and TPAP (26 mg, 0.07 mmol) in one portion. The reaction mixture was stirred overnight. The mixture was then filtered through a pad of silica gel (2 g) and the silica gel was rinsed with Et20. The collected eluate was concentrated. Purification of the residual oil by flash column chromatography (15 g of silica gel, 3:1 petroleum ether-Et20) provided ketone 193 (124 mg, 0.42 mmol, 57%) and ketone 220 (82 mg, 0.28 mmol, 38%) as clear oils. 4. Experimental 2 1 5 (1S*,6fl*,9R*)-9-lsopropyl-5-oxobicyclo[4.3.0]nonane-1-carbaldehyde-2',2'-dimethylpropylene acetal (193) IR (film): v = 2878, 1708, 1472, 1127, 1020, 914, 732 cm"1. 1H NMR (400 MHz): 5 = 0.63 (s, 3 H), 0.81 (d, 3 H, J = 6.7 Hz), 0.97 (d, 3 H, J = 6.3 Hz), I. 11 (s, 3 H), 1.20-1.33 (m, 2 H), 1.33-1.50 (m, 2 H), 1.54-1.66 (m, 1 H), 1.66-1.76 (m, 1 H), 1.82-1.94 (m, 2 H), 2.01-2.11 (m, 1 H), 2.11-2.18 (dd, 1 H, J = 12.0, 7.3 Hz), 2.25-2.33 (dd, 1 H, J=15.9, 5.9 Hz), 2.30-2.44 (m, 1 H), 2.60-2.69 (br d, 1 H, J = 14.5 Hz), 3.16-3.22 (d, 1 H, J = 11.0 Hz), 3.27-3.33 (d, 1 H, J = 11.0 Hz), 3.48-3.54 (dd, 1 H, J = II. 0, 2.7 Hz), 3.54-3.60 (dd, 1 H, J = 11.0, 2.7 Hz), 4.15 (s, 1 H). 1 3C NMR (100 MHz): 5 = 19.7 (+ve), 21.9, (-ve), 22.7 (-ve), 23.0 (-ve), 23.7 (-ve), 24.0 (+ve), 27.6 (+ve), 30.0 (+ve), 31.0 (-ve), 33.0 (+ve), 39.7 (+ve), 54.8 (+ve), 59.0 (-ve), 60.0 (-ve), 77.3 (+ve), 78.1 (+ve), 102.8 (-ve), 210.7 (+ve). O Exact mass calculated for Ci8H 3o0 3: 294.2195; found 294.2194. 4. Experimental 216 Table 4.7: NMR data for (1 S*,6rT,9fT)-9-isopropyl-5-oxobicyclo[4.3.0]nonane-1 -carbaldehyde-2',2'-dimethylpropylene acetal (193) 4' Carbon No. 1 3 C 5 (ppm)a 1H 5 (ppm) (mult; J (Hz))b'c'd HMBC Correlations6 1 54.8 H-2ax, H-2eq, H-3eq, H-6, H -7a, H -8a, H-12, H-13 2 33.0 H-2eq: 2.60-2.69 (m) H-2ax, H-4ax, H-6 H-2ax: part of the m at 1.20-1.33 3 24.0 H-3eq: part of the m at 1.82-1.94 H-2eq, H-4ax, H-4eq H-3ax: 2.30-2.44 (m) 4 39.7 H-4eq: 2.01-2.11 (m) H-2ax, H-2eq, H-3ax, H-3eq, H-6 H-4ax: 2.25-2.33 (dd, J = 15.9, 5.9) 5 210.7 H-3eq, H-4ax, H-4eq, H-6, H-7p 6 60.0 H-6: 2.11-2.18 (dd, J = 12.0, 7.3) H-2eq, H-4ax, H-7P 7 19.7 H-7p: 1.54-1.66 (dddd, J = 12.4, 12.4, 12.0, 6.4) H-6 H -7a: 1.66-1.76 (dddd, J = 13.1, 12.4, 7.3, 3.2) 8 27.6 H-8P: part of the m at 1.33-1.50 H -7a, H-7p H -8a: part of the m at 1.82-1.94 9 59.0 H-9: 1.20-1.33 (m) H-6, H -8a, H-10, H-12, H-13 10 102.8 H-10: 4.15 (s) H-2eq, H-6, H-9, H-1'ax, H-1'eq H-3'ax, H-3'eq 11 31.0 H -11: part of the m at 1.33-1.50 H-12, H-13 12 23.0 H-12: 0.97 (d, J = 6.3) H -11, H-9 13 22.7 H-13: 0.81 (d, J = 6.7) H-11, H-9 1' 77.3 H-1'ax: 3.27-3.33 (d, J = 11.0) H-10, H-3'ax, H-3'eq, H-4', H-5' H-1 'eq: 3.54-3.60 (dd, J = 11.0, 2.7) 2' 30.0 H-1'eq, H-3'eq, H-4', H-5' 3' 78.1 H-3'ax: 3.16-3.22 (d, J = 11.0) H-10, H-1'ax, H-1'eq, H-4', H-5' H-3'eq: 3.48-3.54 (d, J = 11.0, 2.7) 4' 21.9 H-4': 0.63 (s) H-1'ax, H-3'ax, H-5' 5' 23.7 H-5': 1.11 (s) H-1'ax, H-1'eq, H-3'ax, H-3'eq, H-4' Recorded at 100 MHz. Recorded at 400 MHz. Assignments based on HMQC and JMOD data. Methylene protons are designated H-Xax and H-Xeq if they are known to occupy axial and equatorial positions, respectively, in the conformation depicted above. Methylene protons are designated H-Xa if they are known to reside below the plane of the paper in the structure depicted above. They are designated H-Xp if they are known to reside above the plane of the paper in the structure depicted above. Only those correlations which could be unambiguously assigned are recorded. 4. Experimental 217 Table 4.8: NMR data for (1 ^6^9/TO-9-isopropyl-5-oxobicyclo[4.3.0]nonane-1 -carbaldehyde-2',2'-dimethylpropylene acetal (193) Proton 1H COSY NOE No 5 (ppm) (mult; J(Hz)) a b c Correlations01 Correlations6 H-2eq 2.60-2.69 (m) H-2ax, H-3ax, H-3eq H-3ax, H-3eq, H-2ax, H-12 H-2ax part of the m at 1.20-1.33 H-2eq, H-3ax, H-3eq H-3eq part of the m at 1.82-1.94 H-2ax, H-2eq, H-3ax, H-4ax, H-4eq H-3ax 2.30-2.44 (m) H-2ax, H-2eq, H-3eq, H-4ax, H-4eq H-4eq 2.01-2.11 (m) H-4ax, H-3ax, H-3eq H-4ax 2.25-2.33 (dd, J = 15.9, 5.9) H-4eq, H-3ax, H-3eq H-6 2.11-2.18 (dd, J = 12.0, 7.3) H-7a, H-7p H-7p .1.54-1.66 (dddd, J = 12.4, 12.4, 12.0,6.4) H-6, H-7a, H-8a, H8p H-7a 1.66-1.76 (dddd, J = 13.1, 12.4, 7.3, 3.2) H-6, H-7p, H-8a, H-8p H-8P part of the m at 1.33-1.50 H-7a, H-7P H-8a part of the m at 1.82-1.94 H-7a, H-7p H-9 1.20-1.33 (m) H-10 4.15 (s) H-7B, H-8B, H-11, H-1'ax, H-3'ax H-11 part of the m at 1.33-1.50 H-12, H-13 H-12 0.97 (d, J = 6.3) H-11 H-13 0.81 (d, .7 = 6.7) H-11 H-1'ax 3.27-3.33 (d, J = 11.0) H-1'eq H-1'eq 3.54-3.60 (dd, J = 11.0, 2.7) H-1'ax H-3'ax 3.16-3.22 (d, J = 11.0) H-3'eq H-3'eq 3.48-3.54 (d, J = 11.0, 2.7) H-3'ax H-4' 0.63 (s) H-5' 1.11 (s) Recorded at 400 MHz. Assignments based on HMQC and JMOD data. Methylene protons are designated H-Xax and H-Xeq if they are known to occupy axial and equatorial positions, respectively, in the conformation depicted above. Methylene protons are designated H-Xa if they are known to reside below the plane of the paper in the structure depicted above. They are designated H-X(3 if they are known to reside above the plane of the paper in the structure depicted above. Only those correlations which could be unambiguously assigned are recorded. Recorded as NOE difference at 400 MHz using 1D selective NOE experiments. 4. Experimental 218 (1S*,6S*,9ff*)-9-lsopropyl-5-oxobicyclo[4.3.0]nonane-1-carbaldehyde-2',2'-dimethylpropylene acetal (220) IR (film): v = 2937, 1708, 1471, 1121, 994, 734 cm"1. 1H NMR (400 MHz): 5 = 0.59 (s, 3 H), 0.78 (d, 3 H, J = 6.7 Hz), 0.89 (d, 3 H, J = 6.7 Hz), 1.11 (s, 3 H), 1.33-1.45 (m, 2 H), 1.58-1.71 (m, 2 H), 1.66-1.81 (m, 3 H), 1.81-1.92 (m, 1 H), 1.92-2.07 (m, 2 H), 2.14-2.24 (m, 1 H), 2.25-2.34 (m, 1 H), 2.86-2.92 (dd, 1 H, J = 7.5, 7.5 Hz), 3.25-3.32 (d, 1 H, J = 11.1 Hz), 3.29-3.35 (d, 1 H, J = 11.1), 3.49-3.58 (br d, 2 H, J = 11.1 Hz), 4.27 (s, 1 H). 1 3C NMR (100 MHz): 6 = 21.0 (+ve), 21.7 (-ve), 22.1 (-ve), 23.1 (-ve), 23.6 (-ve), 24.7 (+ve), 27.9 (+ve), 28.4 (-ve), 29.1 (+ve), 30.0 (+ve), 38.9 (+ve), 54.0 (-ve), 54.4 (+ve), 54.5 (-ve), 77.0 (+ve), 77.2 (+ve), 104.3 (-ve), 214.4 (+ve). O Exact mass calculated for Ci8H 3o0 3: 294.2195; found 294.2200. 4 . Experimental 219 Table 4.9: NMR data for (1 S*,6S*,9rT)-9-isopropyl-5-oxobicyclo[4.3.0]nonane-1 -carbaldehyde-2',2'-dimethylpropylene acetal (220) 4' Carbon No. 13Q 5 (ppm)a 1H 6 (ppm) (mult; J (Hz))b'cd HMBC Correlations 1 54.4 H-2a, H-2b, H-9, H-10 2 29.1 H-2a: part of the m at 1.58-1.71 H-10, H-4ax, H-4eq H-2b: part of the m at 1.92-2.07 3 21.0 H-3a: part of the m at 1.66-1.81 H-2a, H-2b, H-4ax, H-4eq H-3b: part of the m at 1.92-2.07 4 38.9 H-4ax: 2.14-2.24 (ddd, J= 14.7, 8.3, 6.1) H-3a, H-3b H-4eq: 2.25-2.34 (ddd, J= 14.7, 5.9, 5.9) 5 214.4 H-3b, H-4ax, H-4eq, H-6, H-7b 6 54.5 H-6: 2.86-2.93 (dd, J=7.5 , 7.5) H-3a, H-3b, H-4eq 7 24.7 H-7a: part of the m at 1.58-1.71 H-6, H-8a, H-8b H-7b: part of the m at 1.81 -1.92 8 27.9 H-8a: part of the m at 1.33-1.45 H-6, H-9 H-8b: part of the m at 1.66-1.81 9 54.0 H-9: part of the m at 1.33-1.45 H-8a, H-8b, H-11, H-12, H-13 10 104.3 H-10: 4.27 (s) H-2a, H-2b, H-6, H-9, H-1'ax, H-1'eq, H-3'ax, H-3'eq 11 28.4 H-11: part of the m at 1.66-1.81 H-12, H-13 12 22.1 H-12: 0.78 (d, J=6.7) H-9, H-11 13 23.6 H-13: 0.89 (d, J=6.7) H-9, H-11 1' 77.0 H-1'ax: 3.29-3.35 (d, J= 11.1) H-10, H-3'ax, H-3'eq, H-4', H-5' H-1'eq: part of the br. d at 3.49-3.58 2' 30.0 H-1'eq, H-3'eq, H-4', H-5' 3' 77.2 H-3'ax: 3.25-3.32 (d, J= 11.1) H-10, H-1'ax, H-1'eq, H-4', H-5' H-3'eq: part of the br. d at 3.49-3.58 4' 21.7 H-4': 0.59 (s) H-1'ax, H-3'ax, H-5' 5' 23.1 H-5': 1.11 (s) H-1'ax, H-1'eq, H-3'ax, H-3'eq, H-4' Recorded at 100 MHz. Recorded at 400 MHz. Assignments based on H M Q C and J M O D data. Methylene protons are designated H-Xax and H-Xeq if they are known to occupy axial and equatorial positions, respectively, in the conformation depicted above. If no information regarding their position is available then they are arbitrarily designated H-Xa and H-Xb. 4. Experimental 220 Methyl (1 f?*,3S*J6S*J7r7*)-6-formyl-7-isopropyl-2-oxobicyclo-[4.3.0inonane-3-carboxylate 2',2'-dimethylpropylene acetal (189) and Methyl (1/?*,3/?*,6S*,7/?*)-6-formyl-7-isopropyl-2-oxobicyclo-[4.3.0]nonane-3-carboxylate 2',2'-dimethylpropylene acetal (189) C 0 2 M e » " H 193 189 To a cold (-78 °C) solution of LDA (0.823 mmol) in THF (5 mL) was added a cold (-78 °C) solution of ketone 193 (121 mg, 0.411 mmol) in THF (2 mL). The resulting solution is allowed to stir at this temperature for 1.5 h. Neat methyl cyanoformate (Manders reagent,113 38 mg, 0.45 mmol) was added drop-wise, via a syringe, and the solution was allowed to warm to r.t. TLC analysis of the mixture revealed that ketone 193 had been consumed and that two new materials had been formed. Sat. aq NH4CI (10 mL) and Et20 (10 mL) were added and the phases were allowed to separate. The organic phase was washed with brine (10 mL), dried (MgSO-4) and concentrated. The 1H NMR spectrum of the acquired material revealed the presence of more than two products. Attempts to separate the two materials by flash column chromatography on silica gel failed. Attempts to characterize the mixture of products by GC-LRMS were not successful. 4. Experimental 221 Methyl (1R*,3S*,6S*,7f?*)-6-formyl-7-isopropyl-3-methyl-2-oxobicyclo-[4.3.0]nonane-3-carboxylate 2',2'-dimethylpropylene acetal (221) o o O Q OCH3 O C H 3 189 221 To a stirred solution of keto ester 189 (70 mg, 0.20 mmol) in THF (5 mL) at r.t was added NaH (16 mg of a 60% dispersion in mineral oil, 0.40 mmol). The resulting reaction mixture was stirred for 10 min. Neat Mel (280 mg, 3.29 mmol) was added drop-wise, via a syringe, and the reaction mixture was allowed to stir overnight. H 20 (5 mL) and Et20 (5 mL) were added and the layers were separated. The organic layer was washed with brine (5 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (4 g of silica gel, 3:2 petroleum ether-Et20) to provide keto ester 221 (51 mg, 0.14 mmol, 70%) as a clear oil. 1H NMR (400 MHz): 5 = 0.67 (s, 3 H), 0.83 (d, 3 H, J= 6.6 Hz), 0.99 (d, 3 H, J= 6.3 Hz), 1.20 (s, 3 H), 1.28-1.36 (ddd, 1 H, J= 9.4, 9.4, 9.4 Hz), 1.36-1.52 (m, 2 H),1.40 (s, 3 H), 1.53-1.64 (m, 2 H), 1.64-1.74 (m, 2 H), 1.86-1.97 (dddd, 1 H, J= 13.2, 9.5, 9.4, 6.7 Hz), 2.38-2.43 (dd, 1 H, J= 11.8, 7.2 Hz), 2.62-2.68 (ddd, 1 H, J= 13.6, 5.0, 2.5 Hz), 3.09-3.18 (ddd, 1 H, J= 13.8, 13.8, 5.0 Hz), 3.20-3.25 (d, 1 H, J= 11.1 Hz), 3.32-3.36 (d, 1 H, J= 11.1 Hz), 3.51-3.56 (dd, 1 H, J= 11.1, 2.8 Hz), 3.58-3.62 (dd, 1 H, J= 11.1, 2.8 Hz), 3.69 (s, 3 H), 4.20 (s, 1 H). 1 3C NMR (100 MHz): 5 = 19.9, 22.1, 22.7, 23.0, 23.4, 23.7, 27.7, 29.4, 30.0, 30.0, 31.0, 34.5, 51.9, 55.5, 57.0, 58.0, 77.3, 77.9, 101.8, 174.3, 208.5. 4. Experimental 222 Table 4.10: NMR data for methyl (1 Fr,3S*,6S*,7F?*)-6-formyl-7-isopropyl-3-methyl-2-oxobicyclo[4.3.0]nonane-3-carboxylate 2',2'-dimethylpropylene acetal (221) 11 Carbon 1 3 C 1 H HMBC No. 5 (ppm)a 8 (ppm) (mult; J (Hz))b'c'd Correlations 1 55.5 H-1: 2.38-2.43 (dd, J= 11.8, 7.2) H-5b, H-9a 2 208.5 H-1, H-11 3 57.0 H-4b, H-11 4 34.5 H-4a: part of the m at 1.64-1.74 (ddd, J = 13.8, 4.6, 2.5) H-11 H-4b: 3.09-3.18 (ddd, J= 13.8, 13.8, 5.0) 5 30.0 H-5a: part of the m at 1.36-1.52 (ddd, J = 13.8, 13.6, 4.6) H-1, H-4b, H-12 H-5b: 2.62-2.68 (ddd, J = 13.6, 5.0, 2.5) 6 29.4 7 58.8 H-7: 1.28-1.36 (ddd, J= 9.4, 9.4, 9.4) H-12, H-14, H-15 8 27.7 H-8a: part of the m at 1.53-1.64 H-7 H-8b: 1.86-1.97 (dddd, J= 13.2, 9.5, 9.4, 6.7) 9 19.9 H-9a: part of the m at 1.53-1.64 H-1, H-8a H-9b: part of the m at 1.64-1.74 10 174.3 H-4b, H-1', 11 243 H-11: 1.40 (s) H-4a, H-4b 12 101.8 H-12: 4.20 (s) H-1, H-5a, H-1 "a, H-1"b, H-3"b 13 31.0 H-13: part of the m at 1.36-1.52 H-14, H-15 14 22.7 H-14: 0.99 (d, J=6.3) H-15 15 23.0 H-15: 0.83 (d, J=6.6) H-14 1' 51.9 H-1': 3.69 (s) 1" 77.3 H-1 "a: 3.32-3.36 (d, 11.0) H-12, H-3"a, H-3"b, H4", H-5" H-1"b: 3.58-3.62 (dd, J= 11.0, 2.8) 2" 30.0 H-4", H-5" 3" 77.9 H-3"a: 3.20-3.25 (d, J = 11.1) H-12, H-1 "a, H-1"b, H-4", H-5" H-3"b: 3.51-3.56 (d, J = 11.1, 2.8) 4" 22.1 0.67 (s) H-1 "a, H-3"a, H-5" 5" 23.7 1.20 (s) H-1 "a, H-3"a, H-4" Recorded at 100 MHz. Recorded at 400 MHz. Assignments based on HMQC data. Methylene protons are designated H-Xa and H-Xb arbitrarily. 4. Experimental 223 Table 4.11: NMR data for methyl (1 Rk,3S*,6Sk,7Rk)-6-formyl-7-isopropyl-3-methyl-2-oxobicyclo[4.3.0]nonane-3-carboxylate 2',2'-dimethylpropylene acetal (221) 11 Proton 1H COSY NOE No 8 (ppm) (mult; J(Hz))abc Correlations01 Correlations6 H-1 2.38-2.43 (dd, J= 11.8, 7.2) H-9a, H-9b H-7, H-11, H-5a H-4a part of the m at 1.64-1.74 (ddd, J = 13.8, 4.6, 2.5) H-4b, H-5a, H-5b H-4b 3.09-3.18 (ddd, J= 13.8, 13.8, 5.0) H-4a, H-5a, H-5b H-4a, H-5", H-5b H-5a part of the m at 1.36-1.52 (ddd, J= 13.8, 13.6, H-4a, H-4b, H-5b 4.6) H-5b 2.62-2.68 (ddd, J= 13.6, 5.0, 2.5) H-4a, H-4b, H-5a H-4a, H-4b, H-5a, H-14 H-7 1.28-1.36 (ddd, J=9.4, 9.4, 9.4) H-8b H-8a part of the m at 1.53-1.64 H-8b H-8b 1.86-1.97 (dddd, J= 13.2, 9.5, 9.4, 6.7) H-7, H-8a H-9a part of the m at 1.53-1.64 H-1 H-9b part of the m at 1.64-1.74 H-1 H-11 1.40 (s) H-12 4.20 (s) H-13 part of the m at 1.36-1.52 H-14, H-15 H-14 0.99 (d, J=6.3) H-13 H-15 0.83 (d, J=6.6) H-13 H-1' 3.69 (s) H-1 "a 3.32-3.36 (d, 11.0) H-1"b H-1"b 3.58-3.62 (dd,J= 11.0, 2.8) H-1 "a, H-3"b H-3"a 3.20-3.25 (d, J= 11.1) H-3"b H-3"b 3.51-3.56 (d, J= 11.1, 2.8) H-1"b, H-3"a H-4" 0.67 (s) H-5" 1.20 (s) Recorded at 400 MHz. Assignments based on HMQC data. Methylene protons are designated H-Xa and H-Xb arbitrarily. Only those correlations which could be unambiguously assigned are recorded. Recorded as NOE difference at 400 MHz using 1D selective NOE experiments. 4. Experimental 224 Methyl (1R*,3S*,6S*,7/?*)-6-formyl-7-isopropyl-3-[(2)-5-tri-n-butylstannylpent-4-enyl]-2-oxobicyclo[4.3.0]nonane-3-carboxylate 2',2'-dimethylpropylene acetal (226) H II C 0 2 C H 3 H || C 0 2 C H 3 189 ^ ^ S n B u 3 O 10 v H || C 0 2 C H 3 225 i 4 ^ ^ S n B u 3 5" 15 To a stirred solution of keto ester 189 (116 mg, 0.33 mmol) in THF (5 mL) at r.t. was added, via a syringe, KHMDS (0.66 mL of a 0.5 M solution in toluene, 0.33 mmol). The resulting mixture was stirred for 30 min. A solution of c/'s-5-iodo-1-tributylstannylpent-1-ene (160 mg, 0.33 mmol) in THF (1 mL) was transferred to the reaction vessel via a cannula. The resulting reaction mixture was heated to reflux overnight. Sat. aq NaHC03 (5 mL) and Et20 (5 mL) were added and the layers were separated. The organic layer was washed with brine (5 mL), dried (MgS04) and concentrated. The acquired oil was purified by flash column chromatography (10 g of silica gel, 9:1 petroleum ether-Et20) to provide an inseparable mixture of keto esters 225 and 226 (ratio 4:5, 161 mg, 0.23 mmol, 69%) as a clear oil. To a stirred solution of keto esters 225 and 226 (161 mg, 0.23 mmol) in MeOH (2 mL) was added NaOMe (95%, 20 mg, 0.37 mmol). The resulting reaction mixture was heated to reflux for 18 h. Sat. aq NH4CI (5 mL) and Et20 (5 mL) were added and the layers were separated. The organic layer was washed with brine (5 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (10 g of silica gel, 9:1 petroleum ether-Et20) to provide keto ester 226 (145 mg, 0.20 mmol, 90%) as a clear oil. 4. Experimental 225 IR (film): v = 2954, 1736, 1709, 1459, 1111 cm"1. 1H NMR (400 MHz): 5 = 0.66 (s, 3 H, Me-4" or Me-5"), 0.77 (d, 3 H, J= 6.6 Hz, Me-18 or me-19), 0.79-0.89 (m, 12 H), 0.91 (d, 3 H, J= 6.6 Hz, Me-18 or Me-19), 1.13 (s, 3 H, Me-4" or Me-5"), 1.15-1.17 (m, 23 H), 1.71-1.85 (m, 2 H), 1.85-1.94 (ddd, 1 H, J= 13.2, 13.2, 4.5 Hz), 1.94-2.05 (m, 2 H), 2.05-2.20 (m, 2 H), 2.41-2.51 (ddd, 1 H, J= 14.0, 5.3, 4.0 Hz), 3.10-3.15 (dd, 1 H, J = 7.7, 2.3 Hz), 3.30 (d, 1 H, J= 10.0 Hz, H-1 "ax or H-3" ax), 3.33 (d, 1 H, J= 10.0 Hz, H-1"ax or H-3"ax), 3.52-3.57 (dd, 1 H, J= 10.0, 2.9 Hz, H-1"eq or H-3"eq), 3.54-3.59 (dd, 1 H, J= 10.0, 2.9 Hz, H-1"eq or H-3"eq), 3.66 (s, 3 H, Me-1'), 4.18 (s, 1 H, H-16), 5.76 (d, 1 H, J= 12:5 Hz; 2JSn.H = 36 Hz, H-15), 6.41-6.50 (ddd, 1 H, J= 13.9, 12.5, 7.0 Hz; 3JSn.H = 71 Hz, H-14). 1 3C NMR (100 MHz): 5 = 10.2 (+ve, 3 C), 13.7 (-ve, 3 C), 21.0 (+ve), 21.7 (-ve), 22.8 (-ve), 23.2 (-ve), 23.3 (-ve), 24.2 (+ve), 24.8 (+ve), 27.3 (+ve, 3 C), 27.9 (+ve), 29.2 (+ve, 3 C), 29.3 (-ve), 30.0 (+ve), 30.2 (+ve), 34.8 (+ve), 37.3 (+ve), 50.9 (-ve), 52.1 (-ve), 53.9 (-ve), 56.2 (+ve), 59.3 (+ve), 77.2 (+ve, C-1"), 77.3 (+ve, C-3"), 103.1 (-ve, C-16), 128.4 (-ve), 148.5 (-ve), 172.4 (+ve, C-10), 209 (+ve, C-2). Exact mass calculated for CasHsyCV^Sn [M -Bu] +: 653.3228; found: 653.3226. C33H 5 70 5 1 1 8 Sn [M -Bu] +: 651.3222; found: 651.3215. C 3 3 H 5 7 0 5 1 1 6 S n [M -Bu] +: 649.3223; found: 649.3226. 4. Experimental 226 Methyl (1R*,3S*,6S*,7ff*)-6-formyl-3-[(2)-5-iodopent-4-enyl]-7-isopropyl-2-oxobicyclo[4.3.0]nonane-3-carboxylate 2',2'-dimethylpropylene acetal (227) To a stirred solution of alkenylstannane 226 (318 mg, 0.45 mmol) in dry DCM (10 mL) at r.t. was added a solution of iodine (136 mg, 0.54 mmol) in dry DCM (10 mL). The reaction mixture was stirred at r.t. for 15 min. Sat. aq Na 2 S 2 0 3 (20. mL) was added and the layers were separated. The aqueous layer was extracted with Et20 (10 mL). The organic layers were combined and washed with brine (20 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (15 g of silica gel, 19:1 petroleum ether-Et20) to provide alkenyl iodide 227 (220 mg, 90%) as a clear oil. IR (film): v = 2954, 1736, 1709, 1459, 1111 cm"1. 1H NMR (400 MHz): 5 = 0.68 (s, 3 H, Me-4" or Me-5"), 0.80 (d, 3 H, J= 6.6 Hz, Me-18 or Me-19), 0.92 (d, 3 H, J= 6.6 Hz, Me-18 or Me-19), 1.14 (s, 3 H, Me-4" or Me-5"), 1.24-1.48 (m, 4 H), 1.62-1.99 (m, 7 H), 2.00-2.21 (m, 5 H), 3.07-3.13 (dd, 1 H, J= 8.6, 6.5 Hz), 3.32 (d, 1 H, J= 10.8 Hz, H-1 "ax or H-3"ax), 3.35 (d, 1 H, J= 10.8 Hz, H-1 "ax or H-3"eq), 3.52-3.59 (dd, 1 H, J= 10.8, 2.8 Hz, H-1"eq or H-3"eq), 3.55-3.62 (dd, 1 H, J= 10.8, 2.8 Hz, H-1"eq or H-3"eq), 3.67 (s, 3 H, Me-1'), 4.30 (s, 1 H, H-16), 6.10-6.20 (m, 2 H, H-14and H-15). 4. Experimental 227 1 J C NMR (100 MHz): 5 = 21.7 (-ve), 22.2 (-ve), 22.9 (+ve), 23.3 (-ve), 23.7 (-ve), 24.6 (+ve), 26.7 (+ve), 28.2 (+ve), 28.5 (-ve), 28.9 (+ve), 30.0 (+ve), 33.7 (+ve), 34.8 (+ve), 52.1 (-ve), 52.5 (-ve), 54.3 (+ve), 55.6 (-ve), 59.1 (+ve), 77.1 (+ve, C-1" or C-3"), 77.2 (+ve, C-1" or C-3"), 82.8 (-ve, C-15), 104.5 (-ve, C-16), 140.7 (-ve, C-14), 173.1 (+ve, C-10), 211.0 (+ve, C-2). Exact mass calculated for C25H39IO5: 546.1842; found: 546.1842. 4. Experimental 228 Methyl (3S*,6S*,7/7*)-3-hydroxy-7-isopropylbicyclo[4.3.0]non-1-ene-6-carboxylate (252) ,0 ,0H 12 13 252 To a stirred solution of enone 197 (543 mg, 2.3 mmol) and CeCI3»7H20 (1.028 g, 2.8 mmol) in MeOH (65 mL) at r.t. was added NaBH4 (104 mg, 2.7 mmol) in four portions over 10 min. TLC analysis of the reaction mixture indicated that all of the starting material had been consumed shortly after all of the reducing agent had been added. Aq HCI (10% HCI, 5.5 mL) was added and the reaction mixture was stirred for 10 min. H 20 (50 mL) and Et20 (50 mL) were added and the phases were separated. The organic phase was washed with sat. aq NaHC03 (25 mL) and brine (25 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (25 g of silica gel, 3:2 Et20-petroleum ether) to yield alcohol 252 (471 mg, 2.0 mmol, 86%) as a clear oil. IR (film): v = 3382, 2954, 1726, 1434, 1166, 1006 cm"1. 1H NMR (300 MHz): 5 = 0.83 (d, 3 H, J = 6.2 Hz, Me-12 or Me-13), 0.99 (d, 3 H, J = 6.1 Hz, Me-12 or Me-13), 1.51-1.61 (m, 6 H), 1.81-1.94 (m, 1 H), 2.04-2.13 (m, 1 H), 2.16-2.32 (m, 1 H), 2.59-2.74 (m, 2 H), 3.66 (s, 3 H, Me-1'), 4.22 (br. s, 1 H, H-3), 5.46 (s, 1 H, H-2). 1 3C NMR (100 MHz): 5 = 22.3 (C-12 or C-13), 22.4 (C-12 or C-13), 27.4, 29.3, 31.3 (2 C), 33.6, 51.6, 56.0, 58.7 (C-1'), 68.1, 123.9,(C-2), 148.2 (C-1), 174.1 (C-10) Exact mass calculated for C u H 2 2 0 3 : 238.1569; found: 238.1573. 4. Experimental 229 (3S*,6S*,7R*)-7-lsopropylbicyclo[4.3.0]non-1-ene-6,3-carbolactone (253) To a stirred solution of alcohol 252 (84 mg, 0.35 mmol) in THF (5 mL) at r.t. was added dry KH (16 mg, 0.4 mmol) in one portion. After 30 min, TLC analysis of the mixture indicated that all the starting material had been consumed. H 20 (5 mL) and Et20 (5 mL) were added and the phases were separated. The organic phase was washed with brine (5 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (5 g of silica gel, 1:1 petroleum ether-Et20) to yield lactone 253 (66 mg, 0.32 mmol, 91 %) as a clear oil. IR (film): v = 2958, 1742, 1473, 1363, 1064 cm"1. 1H NMR (300 MHz): 5 = 0.95 (d, 3 H, J = 6.7 Hz, Me-12 or Me-13), 1.00 (d, 3 H, J = 6.5 Hz, Me-12 or Me-13), 1.20-1.28 (dd, 1 H, J - 11.2, 7.2 Hz), 1.40-1.61 (m, 2 H), 1.65-1.75 (m, 1 H), 2.05-2.31 (m, 4 H), 2.34-2.45 (dd, 1 H, J = 16.6, 6.9 Hz), 2.63-2.82 (m, 1 H), 5.03-3.11 (ddd, 1 H, J = 5.0, 3.7, 1.1 Hz, H-3), 6.06-6.12 (ddd, 1 H, J = 5.0, 2.7, 1.5 Hz, H-2). 1 3C NMR (100 MHz): 5 = 23.1 (-ve, C-12 or C-13), 23.5 (-ve, C-12 or C-13), 27.6 (+ve), 28.1 (+ve), 28.1 (+ve), 28.4 (-ve), 32.7 (+ve), 54.6 (+ve, C-6), 55.4, (-ve), 74.3 (-ve, C-3), 120.8 (-ve, C-2), 154.1 (+ve, C-1), 174.2 (C-10). Exact mass calculated for C i 3 H 1 8 0 2 : 206.1307; found: 2061310. H 252 253 4. Experimental 230 Methyl (3/?*,6S*,7fl*)-3-hydroxy-7-isopropylbicyclo[4.3.0]non-1-ene-6-carboxylate (251) 252 254 1 3 251 Alcohol 252 was converted to alcohol 251 using the method developed by Mitsunobu.128 Thus, to a cold (0 °C), stirred, solution of alcohol 252 (617 mg, 2.59 mmol), p-nitrobenzoic acid (866 mg, 5.18 mmol), and Ph3P (1.36 g, 5.18 mmol) in THF (55 mL) was added, drop-wise, via a syringe, neat DEAD (990 mg, 5.68 mmol). The reaction mixture was allowed to warm up to r.t. overnight. The reaction mixture was concentrated and the residual material was purified by flash column chromatography (50 g of silica gel, 3:1 petroleum ether-Et20) to provide ester (254). This, material was dissolved in methanol (20 mL) and the resulting solution was treated with NaOMe (95%, 270 mg, 4.7 mmol). After 1 h, H 20 (30 mL) and Et20 (30 mL) were added. The phases were separated and the organic phase was washed with brine (30 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (35 g of silica gel, 3:2 Et20-petroleum ether) to yield alcohol 251 (553 mg, 2.32 mmol, 89%) as a clear oil. IR (film): v = 3428, 2953, 1725, 1436, 1169, 1011 cm"1. 1H NMR (400 MHz): 5 = 0.83 (d, 3 H, J = 5.9 Hz, Me-12 or Me-13), 1.01 (d, 3 H, J = 5.8 Hz, Me-12 or Me-13), 1.26-1.36 (ddd, 1 H, J=13.3, 13.3, 3.1 Hz), 1.36-1.57 (m, 3 H), 1.59-1.81 (m, 3 H), 1.82-1.91 (m, 1 H), 2.21-2.32 (m, 1 H), 2.49-2.57 (ddd, 1 H, J = 16.3, 3.1, 3.1 Hz), 2.58-2.69 (m, 1 H), 3.60 (s, 3 H, Me-1'), 4.06 (br. s, 1 H, H-3), 5.53 (d, 1 H, J = 1.5 Hz, H-2). 4. Experimental 231 1 J C NMR (100 MHz): 5 = 22.2 (-ve, C-12 or C-13), 22.5 (-ve, C-12 or C-13), 27.2 (+ve), 28.2 (+ve), 29.6 (2 C, +ve), 31.1 (-ve), 51.4 (-ve), 55.9 (+ve, C-6), 58.7 (-ve), 63.3 (-ve), 121.8 (-ve, C-2), 148.6 (+ve, C-1), 173.7 (+ve, C-10). Exact mass calculated for C14H22O3: 238.1569; found 238.1570. 4. Experimental 232 Methyl (1 /?*,6ff*,9ff*)-9-isopropyl-4-oxobicyclo[4.3.0]nonane-1-carboxylate (256) 251 catalyst 256 A stirred solution of alcohol 251 (469 mg, 1.97 mmol) and Crabtree's catalyst (10 mg, 0.12 mmol) in DCM (10 mL) at r.t. and the atmosphere in the reaction vessel was purged with H2 (g). During the purge, the solution turned from orange to pale yellow in colour. The mixture was allowed to stir for 24 h under H2 (g) at 1 atmosphere. TLC analysis of the mixture indicated that all the starting material had been consumed. The mixture was filtered through a pad of silica gel (10 g) and the silica gel was rinsed with Et20. The collected eluate was concentrated and the residual oil was purified by flash column chromatography (25 g of silica gel, 1:1 petroleum ether-Et20) to provide ketone 256 (366 mg, 1.53 mmol, 78%) as a clear oil which crystallized on standing. m.p. =38 °C IR (film): v = 2955, 2872, 1724, 1457, 1433, 1204 cm'1. 1H NMR (400 MHz): 5 = 0.82 (d, 3 H, J = 6.5 Hz), 0.89 (d, 3 H, J = 6.5 Hz), 1.14-1.29 (dddd, 1 H, J = 11.0, 11.0, 10.6, 7.4 Hz), 1.48-1.71 (m, 3 H), 1.75-1.85 (m, 2 H), 1.92-2.02 (dddd, 1 H, J = 7.4, 7.4, 7.4, 6.7 Hz), 2.19-2.27 (m, 3 H), 2.42-2.57 (m, 2 H), 2.80-2.90 (dddd, 1 H, 10.6, 6.7, 6.7, 6.7 Hz), 3.70 (s, 3 H). 1 3C NMR (100 MHz): 5 = 21.1 (-ve), 22.7 (-ve), 28.2 (+ve), 29.8 (-ve), 30.7 (+ve), 32.1 (+ve), 36.7 (+ve), 42.3 (-ve), 43.3 (+ve), 51.7 (-ve), 54.2 (+ve), 57.5 (-ve), 176.6 (+ve), 212.6 (+ve). Exact mass calculated for C 1 4 H 2 2 0 3 : 238.1569; found 238.1568. 4. Experimental 233 Table 4.12: NMR data for methyl (1 R*,6R*,9Fr)-9-isopropyl-4-oxobicyclo[4.3.0]nonane-1 -carboxylate (256) H 13 Carbon 1 3 Q 1H HMBC No. 5 (ppm)a 5 (ppm) (mult; J (Hz))b'c'd Correlations6 1 54.2 H-2a, H-2b, H-6, H-7b 2 36.7 H-2a: part of the m at 2.19-2.27 H-3a, H-3b H-2b: part of the m at 2.19-2.27 3 30.7 H-3a: part of the m at 1.75-1.85 H-3b: part of the m at 2.42-2.57 4. 212.6 H-3a, H-5a, H-5b, H-6 5 43.3 H-5a: part of the m at 2.19-2.27 H-6 H-5b: part of the m at 2.42-2.57 6 42.3 H-6: 2.80-2.90 (dddd, J = 10.6, 6.7, 6.7, 6.7) H-5a, H-5b, H-7a, H-7b, H-8b 7 32.1 H-7a: 1.14-1.29 (dddd, J = 11.0, 11.0, 10.6, 7.4) H-5b, H-6, H-8b H-7b: 1.92-2.02 (dddd, J = 7.4, 7.4, 7.4, 6.7) 8 28.2 H-8a: part of the m at 1.48-1.71 H-7b, H-9, H-11 H-8b: part of the m at 1.75-1.85 9 57.5 H-9: part of the m at 1.48-1.71 H-7b, H-8b, H-12, H-13 10 176.6 H-2a, H-2b, H-3a, H-3b, H-5b, H-6, H-8b 11 29.8 H-11: part of the m at 1.48-1.71 H-8a, H-9, H-12, H-13 12 21.1 H-12: 0.82 (d, J=6.5) H-9, H-11, H-13 13 22.7 H-13: 0.89 (d, J=6.5) H-9, H-11, H-12 1' 51.7 H-1': 3.70 (s) Recorded at 100 MHz. Recorded at 400 MHz. Assignments based on HMQC and JMOD data. Methylene protons are designated H-Xa and H-Xb arbitrarily. Only those correlations which could be unambiguously assigned are recorded. 4. Experimental 234 Table 4.13: NMR data for methyl (1 rT,6rT,9fi*)-9-isopropyl-4-oxobicyclo[4.3.0]nonane-1 -carboxylate (256) Proton 1H COSY No. 5 (ppm) (mult; J(Hz))a'bc Correlations01 H-2a part of the mat 2.19-2.27 H-3a, H-3b H-2b part of the mat 2.19-2.27 H-3a, H-3b H-3a part of the m at 1.75-1.85 H-2a, H-2b, H-3b H-3b part of the m at 2.42-2.57 H-2a, H-2b, H-3a H-5a part of the m at 2.19-2.27 H-5b, H-6 H-5b part of the m at 2.42-2.57 H-5a, H-6 H-6 2.80-2.90 (dddd, J = 10.6, 6.7, 6.7, 6.7) H-5a, H-5b, H-7a, H-7b H-7a 1.14-1.29 (dddd, J = 11.0, 11.0, 10.6, 7.4) H-6, H-7b, H-8a H-7b 1.92-2.02 (dddd, J = 7.4, 7.4, 7.4, 6.7) H-6, H-7a, H-8a, H-8b H-8a part of the m at 1.48-1.71 H-7a, H-7b, H-8b H-8b part of the m at 1.75-1.85 H-7a, H-8b, H-9 H-9 part of the m at 1.48-1.71 H-11 part of the m at 1.48-1.71 H-12, H-13 H-12 0.82 (d, J=6.5) H-11 H-13 0.89 (d, J=6.5) H-11 H-1' 3.70 (s) Recorded at 400 MHz. Assignments based on HMQC and JMOD data. Methylene protons are designated H-Xa and H-Xb arbitrarily. Only those correlations which could be unambiguously assigned are recorded. 4. Experimental 235 To a cold (0 °C), stirred solution of keto ester 256 (40 mg, 0.17 mmol) in Et20 (5 mL) was added, via a syringe, DIBAL-H (1.0 M in hexanes, 0.59 mL, 0.59 mmol). The reaction mixture was stirred at this temperature for 1 h. MgS04»7H20 (200 mg, 0.81 mmol) was added and the resulting heterogeneous mixture was stirred vigorously for 1 h. The mixture was filtered through a sintered glass funnel and the funnel was rinsed with Et20. The eluate was concentrated and the residual oil was purified by flash column chromatography (5 g of silica gel, 2:1 Et20-petroleum ether) to provide a mixture of alcohols 257. The acquired mixture of alcohols 257 was dissolved in DCM (5 mL). To the resulting solution were added, sequentially, NMO (30 mg, 2.5 mmol) in one portion and TPAP (10 mg, 0.03 mmol) in one portion. The reaction mixture was allowed to stir at r.t. for 2 h. The mixture was filtered through a pad of silica gel (1 g) and the silica gel was rinsed with Et20. The collected eluate was concentrated and the residual oil was purified by flash column chromatography (15 g silica gel, 1:1 petroleum ether-Et20) to yield keto aldehyde 258 (25 mg, 0.12 mmol, 72%) as a clear oil. 4. Experimental 236 Keto aldehyde 258 exhibited: IR (film): v = 2962, 1721, 1709, 1474, 737 cm"1. 1H NMR (400 MHz): 5 = 0.90 (d, 3 H, J= 6.3 Hz), 0.93 (d, 3 H. J= 6.3 Hz), 1.21-1.33 (dddd, 1 H, J= 12.2, 12.2, 10.2, 6.4 Hz), 1.36-1.49 (dddd, 1 H, J= 12.2, 12.2, 12.2, 7.1 Hz), 1.66-1.82 (m, 3 H), 1.92-2.01 (m, 1 H), 2.01-1.11 (m, 1 H), 2.22-2.30 (m, 2 H), 2.26-2.33 (dd, 1 H, J= 15.9, 6.8 Hz), 2.37-2.44 (dd, 1 H, J= 15.9, 5.9 Hz), 2.41-2.49 (ddd, 1 H, J= 14.0, 7.7, 5.5 Hz), 2.60-2.71 (dddd, 1 H, J= 10.4, 6.8, 6.4, 5.9 Hz), 9.71 (s, 1 H). 1 3C NMR (100 MHz): 5 = 22.4 (-ve), 22.4 (-ve), 27.2 (+ve), 29.0 (-ve), 30.3 (+ve), 32.7 (+ve), 35.5 (+ve), 39.1 (-ve), 42.7 (+ve), 56.3 (+ve), 57.5 (-ve), 203.5 (-ve), 212.0 (+ve). Exact mass calculated for Ci 3H 2o0 2: 208.1463; found 208.1461. 4. Experimental 237 Table 4.14: NMR data for (1 fl*,6S*,9f?*)-9-isopropyl-4-oxobicyclo[4.3.0]nonane-1 -carbaldehyde (258) H 13 Carbon 1 3 C 1H HMBC No. 5 (ppm)a 5 (ppm) (mult; J(Hz))b'c'd Correlations6 1 56.3 H-2b, H-3a, H-3b, H-7b, H-8b, H-10 2 27.2 H-2a: part of the m at 1.66-1.82 H-3a, H-3b, H-10 H-2b: 2.41-2.49 (ddd, J= 14.0, 7.7, 5.5) 3 35.5 H-3a: part of the m at 2.22-2.30 H-2a, H-2b H-3b: part of the m at 2.22-2.30 4 212.0 H-2a, H-2b, H-5a, H-5b, H6 5 42.7 H-5a: 2.26-2.33 (dd, J= 15.9, 6.8) H-5b: 2.37-2.44 (dd, J= 15.9, 5.9) 6 39.1 H-6: 2.60-2.71 (dddd, J= 10.4, 6.8, 6.4, 5.9) H-5a, H-5b, H-8b 7 32.7 H-7a: 1.21-1.33 (dddd, J= 12.2, 12.2, 10.2, 6.4) H-5a, H-5b, H-8a, H-8b H-7b: 2.01-2.11 (m) 8 30.3 H-8a: 1.36-1.49 (dddd, J= 12.2, 12.2, 12.2,7.1) H-7a, H-7b H-8b: 1.92-2.01 (m) 9 57.5 H-9: part of the m at 1.66-1.82 H-7b, H-8a, H-12, H-13 10 203.7 H-10: 9.71 (s) H-2a, H-2b, H-6, H-9 11 29.0 H-11: part of the m at 1.66-1.82 H-8a, H-12, H-13 12 22.4 H-12: 0.90 (d, J=6.3) H-13 13 22.4 H-13: 0.93 (d, J=6.3) H-13 Recorded at 100 MHz. Recorded at 400 MHz. Assignments based on HMQC and JMOD data. Methylene protons are designated H-Xa and H-Xb arbitrarily. Only those correlations which could be unambiguously assigned are recorded. 4 . Experimental 238 Table 4.15: NMR data for (1 R*,6S*,9rT)-9-isopropyl-4-oxobicyclo[4.3.0]nonane-1-carbaldehyde (258) Proton 1H COSY No 5 (ppm) (mult; J(Hz))a'b'c Correlations'3 H-2a part of the m at 1.66-1.82 H-2b, H-3a, H-3b H-2b 2.41-2.49 (ddd, J= 14.0, 7.7, 5.5) H-2a, H-3a, H-3b H-3a part of the m at 2.22-2.30 H-2a, H-2b, H-3b H-3b part of the m at 2.22-2.30 H-2a, H-2b, H-3a H-5a 2.26-2.33 (dd, J= 15.9, 6.8) H-5b, H-6 H-5b 2.37-2.44 (dd, J= 15.9, 5.9) H-5a, H-6 H-6 2.60-2.71 (dddd, J= 10.4, 6.8, 6.4, 5.9) H-5a, H-5b, H-7a, H-7b H-7a 1.21-1.33 (dddd, J= 12.2, 12.2, 10.2, 6.4) H-6, H-7b H-7b 2.01-2.11 (m) H-6, H-7a H-8a 1.36-1.49 (dddd, J= 12.2, 12.2, 12.2, 7.1) H-8b, H-9 H-8b 1.92-2.01 (m) H-8a, H-9 H-9 part of the m at 1.66-1.82 H-8a, H-8b H-10 9.71 (s) H-11 part of the m at 1.66-1.82 H-12, H-13 H-12 0.90 (d, .7=6.3) H-11 H-13 0.93 (d, J= 6.3) H-11 Recorded at 400 MHz. Assignments based on HMQC and JMOD data. Methylene protons are designated H-Xa and H-Xb arbitrarily. Only those correlations which could be unambiguously assigned are recorded. 4. Experimental 239 Methyl (1 /T^r^je^jgfTH-hydroxy-g-isopropylbicyclo^.S.OJnonane-l -carboxylate (250) To a solution of alcohol 251 (336 mg, 1.4 mmol) in DCM (35 mL) at r.t. was added Brown's catalyst (100 mg, 0.14 mmol). The mixture was placed in a high pressure hydrogenation apparatus and purged with H2 (g) three times at high pressure (800 psi). The hydrogenation vessel was pressurized to 800 psi and the reaction mixture was stirred for 24 h. After this time the hydrogenation chamber was vented to the atmosphere. The reaction mixture was filtered through a pad of silica gel (5 g) and the silica gel was rinsed with Et20. The collected eluate was concentrated and the residual oil was purified by flash column chromatography (10 g of silica gel, 1:1 petroleum ether-Et20) to afford alcohol 250 (315 mg, 1.3 mmol, 94%) as a clear oil. IR (film): v = 3404, 1723, 1445, 1192, 1006 cm"1. 1H NMR (400 MHz): 5 = 0.82 (d, 3 H, J= 6.5 Hz, Me-12 or Me-13), 0.99 (d, 3 H, J= 6.4 Hz, Me-12 or Me-13), 1.20-1.94 (series of m, 12 H), 2.09-2.20 (dddd, 1 H, J= 12.4, 12.4, 9.0, 3.0 Hz), 2.40-2.47 (ddd, 1 H, J= 12.1, 3.1, 3.1 Hz), 3.63 (s, 3 H, Me-1'), 4.02-4.10 (m, 1 H, H-4). 1 3C NMR (100 MHz): 5 = 22.1 (-ve, C-12 orC-13), 22.8 (-ve, C-12 orC-13), 26.8 (+ve), 28.6 (+ve), 31.1 (+ve), 31.6 (+ve), 32.0 (-ve), 34.9 (+ve), 43.5 (-ve), 50.7 (-ve), 57.2 (+ve, C-1), 58.1 (-ve), 66.3 (-ve), 174.8 (+ve, C-10). Exact mass calculated for C i 4 H 2 4 0 3 : 240.1725; found 240.1722. 4. Experimental 240 Methyl (1 f?*,6S*,9/7*)-9-isopropyl-4-oxobicyclo[4.3.0]nonane-1-carboxylate (249) 250 249 Alcohol 250 was oxidized to ketone 249 using the method developed by Ley.88 Thus, to a stirred solution of alcohol 250 (208 mg, 0.87 mmol) in DCM (10 mL) at r.t. was added, sequentially, TPAP (15 mg, 0.4 mmol) in one portion and NMO (112 mg, 0.95 mmol) in one portion. The mixture was allowed to stir overnight. The reaction mixture was filtered through a pad of silica gel (5 g) and the silica gel was rinsed with Et20. The collected eluate was concentrated and the residual oil was purified by flash column chromatography (10 g of silica gel, 2:1 petroleum ether-Et20) to yield ketone 249 (184 mg, 77 mmol, 88%) as a clear oil. IR (film): v = 2951, 1720, 1456, 1179, 1000,738 cm"1. 1H NMR (400 MHz): 5 = 0.85 (d, 3 H, J= 6.1 Hz), 0.98 (d, 3 H, J= 6.1 Hz), 1.36-1.50 (m, 2 H), 1.45-1.55 (ddd, 1 H, J= 13.1, 12.6, 5.8 Hz), 1.61-1.72 (m, 2 H), 1.79-1.92 (dddd, 1 H, J= 11.9, 11.9, 11.9, 4.4 Hz), 1.98-2.16 (m, 2 H), 2.28-2.40 (m, 3 H), 2.40-2.51 (ddd, 1 H, J= 17.0, 12.6, 6.8 Hz), 2.75-2.83 (ddd, 1 H, J= 13.1, 6.8, 1.8 Hz), 3.72 (s, 3 H). 1 3C NMR (100 MHz): 8 = 22.2 (-ve), 22.6 (-ve), 26.8 (+ve), 30.1 (+ve), 31.9 (-ve), 34.3 (+ve), 38.6 (+ve), 43.7 (+ve), 50.4 (-ve), 51.2 (-ve), 56.1 (+ve), 57.3 (-ve), 173.9 (+ve), 210.9 (+ve). Exact mass calculated for C i 4 H 2 2 0 3 : 238.1569; found 238.1567. 4. Experimental 241 Table 4.16: NMR data for methyl (1 fl*,6S*,9f?*)-9-isopropyl-4-oxobicyclo[4.3.0]nonane-1 -carboxylate (249) H 13 Carbon 13Q 1H HMBC No. 5 (ppm)a 8 (ppm) (mult; J (Hz))b'c'd Correlations6 1 56.1 H-1", H-2a, H-2b, H-3a, H-3b 2 34.3 H-2a: 1.45-1.55 (dddd, J= 13.1, 12.6, 5.8) H-3a, H-3b, H-6, H-9 H-2b: 2.75-2.83 (ddd, J= 13.1, 6.8, 1.8) 3 38.6 H-3a: 2.28-2.40 (ddd, J= 17.0, 5.8, 1.8) H-2a, H-2b H-3b: 2.40-2.51 (ddd, J= 17.0, 12.6, 6.8) 4 210.9 H-2a, H-2b, H-3a, H-3b, H-6 5 43.7 H-5a: part of the m at 2.28-2.40 H-7 H-5b: part of the m at 2.28-2.40 6 50.4 H-6: part of the m at 1.98-2.16 H-2b, H-5a, H-5b, H-7b 7 26.8 H-7a: part of the m at 1.61 -1.72 H-5b, H-6, H-8b H-7b: 1.79-1.92 (dddd, J= 11.9, 11.9, 11.9,4.4) . 8 30.1 H-8a: part of the m at 1.61-1.72 H-7a, H-7b H-8b: part of the m at 1.98-2.16 9 57.3 H-9: part of the m at 1.36-1.50 H-12, H-13 10 173.9 H-1', H-3a, H^3b, H-6, H-11, H-12 11 31.9 H-11: part of the m at 1.36-1.50 H-8a, H-12; H-13 12 22.2 H-12: 0.98 (d, J=6.1) H-9, H-11, H-13 13 22.6 H-13: 0.85 (d, J=6.1) H-9, H-11, H-12 V 51.2 H-1': 3.72 (s) Recorded at 100 MHz. Recorded at 400 MHz. Assignments based on HMQC and JMOD data. Methylene protons are designated H-Xa and H-Xb arbitrarily. Only those correlations which could be unambiguously assigned are recorded. 4. Experimental 242 Table 4.17: NMR data for methyl (1 fl*,6S*,9f?*)-9-isopropyl-4-oxobicyclo[4.3.0]nonane-1 -carboxylate (249) 13 Proton 1H COSY No 5 (ppm) (mult; J (Hz))a'bc Correlations01 H-2a 1.45-1.55 (dddd, J= 13.1, 12.6, 5.8) H-2b, H-3a, H-3b H-2b 2.75-2.83 (ddd, J= 13.1, 6.8, 1.8) H-2a, H-3a, H-3b H-3a 2.28-2.40 (ddd, J= 17.0, 5.8, 1.8) H-2a, H-2b, H-3b H-3b 2.40-2.51 (ddd, J= 17.0, 12.6, 6.8) H-2a, H-2b, H-3a H-5a part of the m at 2.28-2.40 H-5b, H-6 H-5b part of the m at 2.28-2.40 H-5a, H-6 H-6 part of the mat 1.98-2.16 H-5a, H-5b, H-7a, H-7a H-7a part of the m at 1.61-1.72 H-6, H-7b, H-8b, H-7b 1.79-1.92 (dddd, J= 11.9, 11.9, 11.9, 4.4) H-6, H-7a, H-8a, H-8b H-8a part of the mat 1.61-1.72 H-7b, H-9 H-8b part of the m at 1.98-2.16 H-7a, H-7b, H-9 H-9 part of the m at 1.36-1.50 H-8a, H-8b H-11 part of the mat 1.36-1.50 H-12, H-13 H-12 0.98 (d, .7=6.1) H-11 H-13 0.85 (d, .7=6.1) H-11 H-1' 3.72 (s) Recorded at 400 MHz. Assignments based on HMQC and JMOD data. Methylene protons are designated H-Xa and H-Xb arbitrarily. Only those correlations which could be unambiguously assigned are recorded. 4. Experimental 243 Methyl (1 S*,6S*,9R*)-9-isopropylbicyclo[4.3.0]non-4-ene-1-carboxylate (266) Alkene 266 was prepared from alcohol 252 using the method developed by Myers.133 Thus, to a cold (-30 °C), stirred solution of Ph3P (8.18 g, 31.2 mmol) in NMM (92 mL) was added drop-wise, via a syringe, DEAD (5.58 mL, 31.2 mmol). The resulting solution was allowed to stir for 5 min. A cold (-30 °C) solution of alcohol 252 (6.19 g, 26 mmol) in NMM (15 mL) was added via a cannula and the resulting reaction mixture was allowed to stir for 10 min. Solid NBSH (6.27 g, 31.2 mmol) was added in one portion and the reaction mixture was stirred for 1 h. TLC analysis of the mixture indicated that the starting material had been consumed. The reaction mixture was allowed to warm up to r.t. over 1 h. Et20 (500 mL) and H 20 (500 mL) were added and the layers were allowed to separate. The organic layer was washed with H 20 (500 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (300 g of silica gel, 9:1 petroleum ether-Et20) to provide alkene 266 (3.65 g, 16.4 mmol, 63%) as a clear oil. IR (film): v = 2948, 1724, 1433, 1163, 685 cm"1. 1H NMR (400 MHz): 8 = 0.81 (d, 3 H, J = 6.4 Hz, Me-12 or Me-13), 0.99 (d, 3 H, J = 6.4 Hz, Me-12 or Me-13), 1.27-1.60 (m, 4 H), 1.66-1.78 (m, 1 H), 1.87-2.02 (m, 2 H), 2.04-2.17 (m, 1 H), 2.20-2.34 (m, 1 H), 2.41-2.51 (m, 1 H), 2.60-2.69 (dd, 1 H, J = 12.9, 7.8 Hz), 3.58 (s, 3 H, Me-1'), 5.40-5.49 (dddd, 1 H, J = 9.8, 3.3, 3.3, 3.3 Hz, H-4 or H-5), 5.68-5.75 (dddd, 1 H, J = 9.8, 2.2, 2.2, 2.2 Hz, H-4 or H-5). .OH H 252 13 266 4. Experimental 244 1 J C NMR (100 MHz): 5 = 22.3 (C-12 or C-13), 22.7 (C-12 or C-13), 25.0, 26.0, 29.5, 32.1, 33.9, 48.6, 50.6, 56.6 (C-1), 56.8, 126.6 (C-4 or C-5), 129.4 (C-4 or C-5), 174.8 (C-10). Exact mass calculated for C14H22O2: 222.1620; found: 222.1627. 4. Experimental 245 Methyl (1 S*,4R*,5S*,6R*,9R*)-9-isopropyl-4,5-epoxybicyclo[4.3.0]nonane-1 -carboxylate (270) 266 270 To a cold (-78 ° C ) , stirred solution of freshly distilled DMDO (-0.09-0.11 M in acetone, 300 mL, -30 mmol) was added a solution of alkene 266 (3.20 g, 14.4 mmol) in acetone (40 mL). The solution was allowed to warm to r.t. overnight. Et20 (300 mL) and H 20 (300 mL) were added and the layers were allowed to separate. The organic layer was washed with sat. aq NaHC03 (300 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (200 g of silica gel, 9:1 petroleum ether-Et20) to provide epoxide 270 (1.96 g, 8.2 mmol, 57%) as a clear oil. IR (film): v = 2954, 1725, 1455, 1174, 834 cm"1. 1H NMR (400 MHz): 5 = 0.80 (d, 3 H, J=6.4 Hz, Me-12 or Me-13), 1.01 (d, 3 H, J=6.3 Hz, Me-12 or Me-13), 1.08-1.18 (ddd, 1 H, J= 12.9, 11.7, 6.6 Hz), 1.23-1.44 (m, 3 H), 1.57-1.66 (ddd, 1 H, J= 11.2, 9.1, 1.6 Hz), 1.72-1.84 (dddd, 1 H, J= 15.0, 11.0, 7.3, 3.4 Hz), 1.85-2.12 (m, 4 H), 2.29-2.38 (dd, 1 H, J= 13.0, 7.3 Hz), 2.97-3.02 (dd, 1 H, J = 4.1, 4 Hz. H-4 or H-5), 3.16-3.20 (dd, 1 H, J= 4.1, 2.1 Hz, H-4 or H-5), 5.35 (s, 3 H, Me-1'). 1 3 C NMR (100 MHz): 8 = 22.4, 22.6, 22.8, 24.6, 28.8, 31.1, 31.6, 51.0 (C-1), 51.1 (2 C ) , 55.6 (2 C ) , 55.8, 174.4 (C-10). Exact mass calculated for C H H 2 2 0 3 : 238.1569; found: 238.1574. 4. Experimental 246 (1 S*,5/?*,6R*,9fl*)-9-lsopropyl-4-oxobicyclo[4.3.0]nonane-1,5-carbolactone (272) 270 271 272 To a cold (0 °C), stirred suspension of CuBr»DMS (1.79 g, 8.72 mmol) in Et20 (100 mL) was added, via a syringe, MeLi (12.5 mL, 1.4 M in Et20, 17.4 mmol). The resulting mixture was allowed to stir for 20 min. A solution of epoxide 270 (1.73 g, 7.27 mmol) in Et20 (25 mL) at r.t. was transferred to the reaction vessel via a cannula. The resulting mixture was allowed to cool to 0 °C over 15 min. Neat BF3»Et20 (1.14 g, 7.99 mmol) was added drop-wise via a syringe. The reaction mixture was allowed to stir for 10 min. TLC analysis of the reaction mixture indicated that the starting material had been consumed. Sat. aq NH4CI (50 mL) was added and the layers were separated. The organic layer was washed with brine (50 mL), dried (MgS04) and concentrated to provide crude hydroxy lactone 271. Hydroxy lactone 271 was converted to keto lactone 272 using the method developed by Ley.88 Thus, to a stirred solution of crude hydroxy lactone 271 in DCM (100 mL) at r.t. was added, sequentially, TPAP (255 mg, 0.73 mmol) in one portion and NMO (1.02 g, 8.73 mmol) in one portion. The resulting reaction mixture was allowed to stir overnight. The reaction mixture was filtered through a pad of silica gel (5 g) and the silica gel was rinsed with Et20. The collected eluate was concentrated and the residual oil was purified by flash column chromatography (100 g of silica gel, 1:1 petroleum ether-Et20) to provide keto lactone 272 (1.15 g, 5.18 mmol, 71?/o) as a clear oil which crystallized on standing. m.p. = 64-66 °C IR (film): v = 2959, 2872, 1790, 1742, 1456, 1201, 1088, 975 cm"1. 4. Experimental 247 1H NMR (400 MHz): 8 = 0.86 (d, 3 H, J = 6.7 Hz, Me-12 or Me-13), 1.05 (d, 3 H, J = 6.6 Hz, Me-12 or Me-13), 1.45-1.66 (m, 3 H), 1.66-1.76 (ddd, 1 H, J=12.7, 11.5, 6.9 Hz, H-2a), 1.90-2.00 (ddd, 1 H, 14.5, 9.6, 5.0 Hz), 2.00-2.08 (m, 1 H), 2.17-2.28 (m, 1 H)2.32-2.38 (dd, 1 H, J=9, 9 Hz, H-6), 2.36-2.45 (dd, 1 H, J- 16.5, 6.9 Hz, H-3a), 2.47-2.55 (dd, 1 H, J = 12.7, 9.1 Hz, H-2(3), 2.59-2.69 (ddd, 1 H, J = 16.5, 11.5, 9.1 Hz, H-3p), 4.35 (s, 1 H, H-5). 1 3C NMR (100 MHz): 8 = 22.9, 23.1, 23.8, 29.4, 30.0, 31.0, 33.8, 52.7, 55.3 (C-1), 55.9, 83.8 (C-5), 177.1 (C-10), 204.5 (C-4). Exact mass calculated for Ci 3 Hi 8 0 3 : 222.1256; found: 222.1255. 4. Experimental 248 (1S*,5R*,6R*,9R*)-9-lsopropyl-4-methylenebicyclo[4.3.0]nonane-1,5-carbolactone (273) H H H To a cold (0 °C), stirred suspension of methyltriphenylphosphonium bromide (926 mg, 2.59 mmol) in THF (10 mL) was added drop-wise, via a syringe, n-BuLi (1.62 mL, 1.6 M in hexanes, 2.59 mmol) and the resulting mixture was allowed to stir for 30 minutes. The reaction mixture was then cooled to (-78 °C). A solution of keto lactone 272 (192 mg, 0.87 mmol) in THF (1 mL) was transferred to the reaction vessel via a cannula and the reaction mixture was allowed to warm to r.t. overnight. TLC analysis of the reaction mixture indicated that the starting material had been consumed. Et20 (10 mL) and H 20 (10 mL) were added and the layers were separated. The organic layer was washed with brine (10 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (10 g of silica gel, 9:1 petroleum ether-Et20) to afford alkene 273 (160 mg, 0.72 mmol, 83%) as a clear oil. IR (film): v = 2960, 1770, 1475, 1456, 1203, 1094, 956, cm"1. 1H NMR (400 MHz): 8 = 0.87 (d, 3 H, J = 6.9 Hz, Me-13 or Me-14), 1.07 (d, 3 H, J = 6.9 Hz, Me-13 or Me-14), 1.39-1.65 (m, 4 H), 1.82-2.01 (m, 2 H), 2.05-2.12 (dd, 1 H, J = 9.6 Hz), 2.17-2.29 (m, 1 H), 2.32-2.47 (m, 3 H), 4.69 (s, 1 H, H-5), 4.79 (s, 1 H, H-11), 4.86 (s, 1 H, H-11). 1 3C NMR (100 MHz): 8 = 23.1, 23.4, 24.3, 27.3, 29.8, 30.4, 32.3, 53.6, 56.3 (C-1), 57.2, 83.7 (C-5), 109.7 (C-11), 144.5 (C-4), 178.6 (C-10). Exact mass calculated for d 4 H 1 2 0 2 : 220.1463; found: 220.1465. 4. Experimental 249 Methyl (6S*,7R*)-7-isopropyl-3-oxo-2-[2-(terf-butyldimethylsiloxy)ethyl]bicyclo [4.3.0]non-1 -ene-6-carboxylate (293) To a stirred suspension of NaH (70 mg of a 60% suspension in mineral oil, 1.75 mmol) in DME (10 mL) at r.t. was added, via a cannula, a solution of enone 197 (472 mg, 2.00 mmol) in DME (5 mL). The resulting reaction mixture was allowed to stir for 1 h. During this time the mixture became dark purple in colour. A solution of iodide 292 (858 mg, 3.00 mmol) in DME (5 mL) was transferred to the reaction vessel via a cannula. The reaction mixture was allowed to stir overnight. Sat. aq NH4CI (10 mL) was added and the resulting mixture was extracted with Et20 (3 x 20 mL). The combined organic extracts were washed with brine (20 mL), dried (MgS04) and concentrated. The residual oil was purified using flash column chromatography (25 g of silica gel, 9:1 petroleum ether-Et20) to afford enone 293 (498 mg, 78% based on recovered starting material) as a clear oil. In addition, enone 197 was recovered (91 mg, 19%). Enone 293 exhibited: IR (film): v = 2952, 1729, 1668, 1357, 1100, 837 cm"1. 1H NMR (400 MHz): 5 = -0.04 (s, 3 H, Si-CH3), -0.02 (s, 3 H, Si-CH3), 0.83 (s, 9 H, Si-C(CH3)3), 0.86 (d, 3 H, J= 6.4 Hz, Me-14 or Me-15), 1.02 (d, 3 H, J= 6.4 Hz, Me-14 or Me-15), 1.42-1.58 (m, 2 H), 1.62-1.78 (m, 2 H), 1.91-2.01 (m, 1 H), 2.98-2.37 (ddd, 1 H, J= 18.2, 5.1, 2.3 Hz), 2.37-2.49 (m, 3 H), 2.60-2.75 (m, 1 H), 2.70-2.81 (ddd, 1 H, J = 19.6, 9.1, 1.9 Hz), 2.81-2.88 (ddd, 1 H, J= 12.7, 4.5, 2.5 Hz), 3.50-3.64 (m, 2 H, H-11), 3.65 (s, 3 H, Me-1'). 4. Experimental 250 1 3C NMR (100 MHz): 5 = -5.5 (-ve, Si-CH3), -5.4 (-ve, Si-CH3), 18.2 (+ve, Si-C(CH3)3), 22.0 (-ve, C-14.or C-15), 22.5 (-ve, C-14 or C-15), 25.9 (3 C, -ve, Si-C(CH3)3), 27.7 (+ve), 29.8 (+ve), 30.0 (+ve), 30.6 (-ve), 33.6 (+ve), 34.8 (+ve), 51.8 (-ve), 57.1 (+ve, C-6), 58.8 (-ve, C-1'), 61.6 (+ve, C-11), 130.2 (+ve, C-2), 167.9 (+ve, C-1 or C-12), 172.1 (+ve, C-1 or C-12), 198.0 (+ve, C-3). Exact mass calculated for C 2 2 H 3 8 0 4 S i : 394.2539; found: 394.2534. 4. Experimental 251 Methyl (1S*,5S*,6S*,9R*)-9-isopropyl-4-oxo-5-[2-(terf-butyldimethylsiloxy)ethyl] bicyclo[4.3.0]nonane-1 -carboxylate (299) The hydrogenation of enone 293 was not a reproducible process. Best results were obtained in small scale reactions with a high load of palladium-on-carbon. To a solution of enone 293 (115 mg, 0.16 mmol) in EtOH (5 mL) was added palladium on carbon (60 mg, 10% by weight) and the reaction was allowed to stir under hydrogen at one atmosphere. After 1 h TLC analysis of the mixture indicated that the starting material had been consumed. The reaction mixture was filtered through a pad of silica (1 g) and the silica gel was rinsed with Et20. The combined filtrated was washed with H 20 (2x10 mL) and brine (5 mL), dried (MgS04) and concentrated. The residual oil was dissolved in MeOH (5 mL) and NaOMe (20 mg, 0.37 mmol) was added in one portion. The resulting reaction mixture was stirred overnight. H 20 (10 mL) and Et20 were added and the layers were separated. The organic layer was washed with brine (5 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography to provide ketone 299 (89 mg, 77%) as a clear oil. 15 293 299 4. Experimental 252 Ketone 299 exhibited: IR (film): v = 2954, 1724, 1712, 1092, 836 cm"1. 1H NMR (400 MHz): 5 = -0.02 (s, 3 H, Si-CH3), -0.01 (s, 3 H, Si-CH3), 0.83 (d, 3 H, J = 6.2 Hz, Me-14 or Me-15), 0.85 (s, 9 H, Si-C(CH3)3), 0.96 (d, 3 H, J= 6.2 Hz, Me-14 or Me-15), 1.34-1.84 (m, 7 H), 1.84-1.95 (m, 2 H), 1.95-2.05 (m, 1 H), 2.28-2.37 (ddd, 1 H, J= 15.6, 5.3, 2.1 Hz), 2.41-2.54 (m, 2 H), 2.73-2.83 (ddd, 1 H, J= 13.1, 6.7, 2.1 Hz, H-2p), 3.51-3.59 (ddd, 1 H. J= 9.9, 8.0, 6.3 Hz, H-12), 3.64-3.72 (ddd, 1 H, J= 10.1, 7.2, 4.7 Hz, H-12), 3.70 (s, 3 H, Me-1'). 1 3C NMR (100 MHz): 8= -5.4 (2 C, Si-CH3), 18.2 (+ve, Si-C(CH3)3), 22.1 (-ve, C-14 or C-15), 22.5 (-ve, C-14 or C-15), 25.8 (+ve), 25.9 (3 C, -ve, Si-C(CH3)3), 30.0 (+ve), 30.3 (+ve), 31.9 (-ve), 34.9 (+ve), 39.2 (+ve), 48.1 (-ve), 51.2 (-ve), 55.3 (-ve), 57.0 (+ve, C-1), 57.4 (-ve), 61.2 (+ve, C-12), 174.1 (+ve, C-10), 211.9 (+ve, C-4). Exact mass calculated for Ci 8 H 3 i0 4 Si [M - t-Bu]+: 339.1992; found: 339.1991. 4. Experimental 253 Methyl (1 S*,4R*,5S*,6S*,9/7*,11 fT)-4,11 -epoxy-4-ethyl-9-isopropyl-5[2-(tert-butyldimethylsiloxy)ethyl]bicyclo[4.3.0]nonane-1-carboxylate (302) To a cold (0 °C), stirred suspension of ethyltriphenylphosphonium bromide (112 mg, 0.30 mmol) in THF (3 mL) was added BuLi (113 u.L, 1.6 M in hexanes, 0.18 mmol) and the resulting reaction mixture was allowed to stir for 30 min. A solution of ketone 299 (60 mg, 0.15 mmol) in THF (2 mL) was transferred to the reaction vessel via a cannula and the reaction mixture was allowed to warm up to r.t. overnight. H 20 (3 mL) and Et20 (5 mL) were added and the layers were separated. The organic layer was washed with brine (5 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (5 g of silica gel, 9:1 petroleum ether-Et20) to afford an inseparable mixture of alkenes 300 and 301 (48 mg, ratio 11:1, 78%) as a clear oil. To a cold (-78 °C), stirred solution of the acquired mixture of alkenes 300 and 301 in DCM (6 mL) was added, dropwise, via a pipette, a solution of freshly distilled DMDO (1.4 mL, -0.09-0.11 M, -0.13-0.15 mmol). The resulting mixture was allowed to stir for 1 h. TLC analysis of the mixture indicated that all the starting material had been consumed. H 20 (5 mL) and Et20 (10 mL) were added and the layers were separated. The organic layer was washed with brine (5 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (10 g of silica gel, 9:1 petroleum ether-Et20) to afford oxirane 302 (45 mg, 89%) as a clear oil. No effort to isolate or characterize the oxirane derived from alkene 301 was made. 299 300: RT = Me, R 2 = H 301: RT = H, R 2 = Me 302 4. Experimental 254 Oxirane 302 exhibited: IR (film): v = 2955, 1723, 1255, 1164, 1095, 836 cm"1. 1H NMR (500 MHz): 5 = 0.00 (s, 3 H, Si-CH3), 0.01 (s, 3 H, Si-CH3), 0.80 (d, 3 H, J = 6.5 Hz), 0.86 (s, 9 H, Si-C(CH3)3), 0.89 (d, 3 H, J= 6.5 Hz), 0.90-0.98 (m, 1 H), 1.08 (d, 3 H, J=5.4 Hz), 1.14-1.28 (m, 1 H), 1.37-1.75 (m, 6 H), 1.76-1.84 (m, 1 H), 1.84-1.92 (m, 1 H), 1.92-2.01 (m, 1 H), 2.05-2.14 (ddd, 1 H, J= 19.7, 10.7, 10.7 Hz), 2.14-2.22 (ddd, 1 H, J= 13.4, 10.3, 10.3 Hz), 2.67-2.72 (q, 1 H, J= 5.4 Hz), 2.70-2.79 (ddd, 1 H, J= 13.7, 9.3, 9.3 Hz), 3.56-3.71 (m, 2 H), 3.65 (s, 3 H). 1 3C NMR (125 MHz): 5 = -5.4 (-ve, Si-CH3), -5.3(-ve, Si-CH3), 14.0 (-ve), 18.2 (+ve, Si-C(CH3)3), 21.9 (-ve), 22.6 (-ve), 25.9 (3 C, -ve, Si-C(CH3)3), 27.3 (+ve), 29.0 (+ve), 30.7 (+ve), 31.1 (+ve), 32.0 (-ve), 34.2 (+ve), 34.3 (-ve), 50.8 (-ve), 52.3 (-ve), 55.5 (+ve), 60.0 (-ve), 61.6 (+ve), 61.8 (-ve), 63.4 (+ve), 175.3 (+ve). Exact mass calculated for C 24H 440 4Si: 424.3009; found: 424.3007. 4. Experimental 255 Table 4.18: NMR data for Methyl (1 S*>4rri5S*l6S*,9r?M 1 ff)-4,11-epoxy-4-ethyl-9-isopropyl-5[2-(tert-butyldimethylsiloxy)ethyl]bicyclo[4.3.0]nonane-1-carboxylate (302) 17 Carbon 1 3 C 1H No. 5 (ppm)a 5 (ppm) (mult; J(Hz))bcd 1 55.5 2 31.1 H-2a: part of the m at 1.37-1.75 (dd; 13.7, 10.3) H-2b: 2.70-2.79 (ddd; 13.7, 9.3, 9.3) 3 29.0 H-3a: 0.90-0.98 (ddd; 13.4, 9.3) H-3b: 2.14-2.22 (ddd; 13.4, 10.3, 9.3) 4 63.4 5 34.3 H-5: part of the m at 1.37-1.75 6 52.3 H-6: 2.05-2.14 (ddd; 19.7, 10.7, 10.7) 7 27.3 H-7a: 1.84-1.92 (m) H-7b: part of the m at 1.37-1.75 8 30.7 H-8a: 1.92-2.01 (m) H-8b: part of the m at 1.37-1.75 9 60.0 H-9: part of the m at 1.37-1.75 10 175.3 11 61.8 H-11: 2.67-2.72 (q; 5.4) 12 14.0 H-12: 1.08 (d; 5.4) 13 34.2 H-13a: 1.76-1.84 (m) H-13b: part of the m at 1.37-1.75 14 61.6 H-14a: part of the m at 3.56-3.71 H-14b: part of the m at 3.56-3.71 15 32.0 H-15: 1.14-1.28 (m) 16 21.9 H-16: 0.89 (d; 6.5) 17 22.6 H-17: 0.80 (d; 6.5) V 50.8 H-1': 3.65 (s) Recorded at 125 MHz. Recorded at 500 MHz. c Assignments based on HMQC and JMOD data. Methylene protons are designated H-Xa and H-Xb arbitrarily. 4. Experimental 256 Table 4.19: NMR data for Methyl (1 S*,4f?*)5S*,6S*,9fl*,11 fl*)-4,11-epoxy-4-ethyl-9-isopropyl-5[2-(tert-butyldimethylsiloxy)ethyl]bicyclo[4.3.0]nonane-1-carboxylate (302) 17 Proton 1H COSY NOE No. 5 (ppm) (mult; J (Hz))a'b'c Correlations01 Correlations01,8 H-2a part of the m at 1.37-1.75 (dd; 13.7, 10.3) H-2b, H-3b H-2b 2.70-2.79 (ddd; 13.7, 9.3, 9.3) H-2a, H-3b, H-3a H-2a, H-3a H-3a 0.90-0.98 (dd; 13.4, 9.3) H-2b, H-3b H-3b 2.14-2.22 (ddd; 13.4, 10.3, 9.3) H-2a, H-2b, H-3a H-3a H-5 part of the m at 1.37-1.75 H-6 2.05-2.14 (ddd; 19.7, 10.7, 10.7) H-7a 1.84-1.92 (m) H-7b part of the m at 1.37-1.75 H-8a 1.92-2.01 (m) H-8b part of the m at 1.37-1.75 H-9 part of the m at 1.37-1.75 H-11 2.67-2.72 (q; 5.4) H-12 H-3a H-12 1.08 (d;5.4) H-11 H-11, H-1' H-13a 1.76-1.84 (m) H-13b, H-14a, H-14b H-13b part of the mat 1.37-1.75 H-13a, H-14a, H-14b H-14a part of the m at 3.56-3.71 H-13a, H-13b, H-14b H-14b part of the m at 3.56-3.71 H-13a, H-13b, H-14a H-15 1.14-1.28 (m) H-16, H-17 H-16 0.89 (d; 6.5) H-15 H-17 0.80 (d; 6.5) H-15 H-1' 3.65 (s) Recorded at 500 MHz. Assignments based on HMQC and JMOD data. Methylene protons are designated H-Xa and H-Xb arbitrarily. Only those correlations which could be unambiguously assigned are recorded. Recorded as 1D selective NOE difference at 400 MHz. 4. Experimental 257 Methyl (1S*,4S*,5S*,6S*,9/?*)-4-acetyl-9-isopropyl-5-[2-(ferf-butyldimethyl siloxy)ethyl]bicyclo[4.3.0]nonane-1 -carboxylate (304) A cold (-78 °C) solution of MABR (0.18 M, 1.7 mL, 0.31 mmol) in DCM was prepared according to the procedure described by Yamamoto and coworkers.140 The above mentioned solution was added to a cold (-78 °C) solution of oxirane 302 (66 mg, 0.16 mmol) in DCM (3 mL). The resulting reaction mixture was allowed to stir for 30 min. TLC analysis of the mixture indicated that the starting material had been consumed. 10% aq. HCI (5 mL) and DCM (5 mL) were added and the layers were separated. The organic layer was washed with sat. aq NaHC03, dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (5 g of silica gel, 9:1 petroleum ether-Et20) to provide ketone 304 (61 mg, 92%) as a clear oil which exhibited: IR (film): v = 2952, 1724, 1722, 1255, 1091,836 cm"1. 1H NMR (400 MHz): 6 = -0.01 (s, 3 H, Si-CH3), 0.00 (s, 3 H, Si-CH3),), 0.79 (d, 3 H, J = 6.2 Hz, Me-16 or Me-17), 0.84 (s, 9 H, Si-C(CH3)3), 0.95 (d, 3 H, J= 6.2 Hz, Me-16 or Me-17), 1.05-1.15 (ddd, 1 H, J= 13.1, 13.1, 3.8 Hz, H-2a), 1.20-1.52 (m, 7 H), 1.63-1.94 (m, 5 H), 2.07 (s, 3 H, Me-11), 2.24-2.34 (ddd, 1 H, J= 12.6, 11.0, 4.3 Hz, H-4), 2.63-2.71 (ddd, 1 H, J= 13.1, 3.4, 3.4 Hz, H-2p), 3.48-3.62 (m, 2 H;.H-14), 3.63 (s, 3 H, Me-1'). 302 MABR 1 7 304 4. Experimental 258 1 J C NMR (100 MHz): 5 = -5.4 (2 C, -ve, Si-CH3), 18.3 (+ve, Si-C(CH3)3), 22.3 (-ve, C-16 or C-17), 22.6 (-ve, C-16 or C-17), 25.5 (+ve), 25.9 (3 C, -ve, Si-C(CH3)3), 27.8 (+ve), 28.2 (-ve), 28.4 (+ve), 31.8 (-ve), 35.4 (+ve), 35.9 (-ve), 36.0 (+ve), 50.8 (-ve), 54.5 (-ve), 57.2 (+ve, C-1), 57.4 (-ve), 57.9 (-ve), 60.8 (+ve, C-14), 174.4 (+ve, C-10), 212.5 (C-11). Exact mass calculated for C 24H 440 4Si: 424.3009; found: 424.3007. 4. Experimental 259 Methyl (6S*,7/T*)-7-isopropyl-2-(2-methoxyethyl)-3-oxobicyclo[4.3.0]non-1-ene-6-carboxylate (308) r 11 197 1 5 308 To a stirred suspension of NaH (2.821 g of a 60% suspension in mineral oil, 70.5 mmol) in dry DME (115 mL) at r.t. was added a solution of enone 197 (15.14 g, 64.1 mmol) in dry DME (50 mL) and the resulting mixture was allowed to stir for 1 h. During this time the reaction mixture became dark purple in colour. Neat 2-bromoethyl methyl ether (319, 9.08 mL, 13.43 g, 96.6 mmol) was added via a syringe. The resulting mixture was allowed to stir at r.t. for 2 days. Sat. aq NH4CI (100 mL) was added and the resulting mixture was extracted with Et20 (3 x 100 mL). The combined organic extracts were washed with brine (100 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (700 g of silica gel, 3:1 petroleum ether-Et20) to afford enone 308 (12.84 g, 68%) as a clear oil. IR (film): v = 2952, 1725, 1665, 1356, 1230, 1170, 1112 cm"1. 1H NMR (400 MHz): 8 = 0.84 (d, 3 H, J = 6.5 Hz, Me-14 or Me-15), 1.00 (d, 3 H, J = 6.5 Hz, Me-14 or Me-15), 1.41-1.75 (m, 4 H), 1.93-2.02 (m, 1 H), 2.28-2.36 (ddd, 1 H, J = 18.2, 5.2, 2.4 Hz), 2.35-2.42 (dd, 1 H, J = 13.4-4.7 Hz), 2.41-2.48 (br. dd, 2 H, J = 7, 7 Hz), 2.50-2.62 (ddd, 1 H, J = 17.6, 8.9, 8.9 Hz), 2.68-2.79 (dd, 1 H, J=19.9, 10.1 Hz), 2.78-2.85 (ddd, 1 H, J= 13.0, 4.7, 2.5 Hz), 3.24 (s, 3 H, Me-1'), 3.25-3.32 (m, 2 H, H-11), 3.64 (s, 3 H, Me-1"). 4. Experimental 260 1 dC NMR (100 MHz): 5 = 22.0 (-ve, C-14 or C-15), 22.4, (-ve, C-14 or C-15), 26.8 (+ve), 27.7 (+ve), 29.6 (+ve), 30.6 (-ve), 33.5 (+ve), 34.7 (+ve), 51.8 (-ve), 57.1 (+ve, C-6), 58.4 (-ve), 58.5 (-ve), 70.8 (+ve, C-11), 129.9 (+ve, C-2), 167.9 (+ve, C-1), 172.0 (+ve, C-12), 197.8 (+ve, C-3). Exact mass calculated for C 1 7 H 2 6 0 4 : 294.1831; found: 294.1831. 4. Experimental 261 Methyl (1S*,5S*,6S*,9r?*)-9-isopropyl-5-(2-methoxyethyl)-4-oxobicyclo[4.3.0] nonane-1-carboxylate (309) and Methyl (6S*,7/?*)-7-isopropyl-2-(2-methoxyethyl)bicyclo[4.3.0]non-1-ene-6-carboxylate (321) To a solution of enone 308 (12.84 g, 43.6 mmol) in MeOH (200 mL) at r.t. was added palladium-on-carbon (10%, 1 g, 0.94 mmol). The resulting heterogeneous mixture was purged with H2 (g) three times. The mixture was allowed to stir overnight at r.t. under H2 (g) at one atmosphere. TLC analysis of the mixture indicated that the starting material had been consumed. The mixture was filtered through a pad of Celite® (10 g) and the Celite® was rinsed with MeOH (50 mL). The collected eluate was transferred to a 500 mL round-bottomed flask and concentrated HCI (~10M, 5 mL) was added. The solution was allowed to stir overnight. H 20 (200 mL) and Et20 (200 mL) were added and the layers were separated. The aqueous layer was extracted with Et20 (200 mL). The combined organic extracts were washed with brine (200 mL), dried (MgS04) and concentrated. The residual material was purified by flash column chromatography (700 g of silica gel, 7:3 petroleum ether-Et20) to afford ketone 309 (10.08 g, 78%) and alkene 321 (2.08 g, 17%) as clear oils. 308 309 321 4. Experimental 262 Ketone 309 exhibited: 1" 12 IR (film): v = 2945, 1724, 1712, 1458, 1215, 1117 cm"1. 1H NMR (400 MHz): 5 = 0.83 (d, 3 H, = 6.2 Hz, Me-14 or Me-15), 0.96 (d, 3 H, J = 6.2 Hz, Me-14 or Me-15), 1.32-1.47 (m, 2 H), 1.47-1.52 (ddd, 1 H, J=12.9, 12.9, 5.5 Hz), 1.56-1.77 (m, 3 H), 1.79-2.07 (m, 4 H), 2.29-2.38 (ddd, 1 H, J = 15.6, 5.5, 2.5 Hz), 2.39-2.53 (m, 2 H), 2.74-2.83 (ddd, 1 H, J= 13.0, 6.7, 2.2 Hz, H-2B), 3.24 (s, 3 H, Me-1"), 3.30-3.38 (ddd, 1 H, J= 14.4, 7.5, 7.2 Hz, H-12), 3.39-3.46 (ddd, 1 H, J = 9.4, 7.5, 5.2 Hz, H-12), 3.72 (s, 3 H, Me-1'). 1 3C NMR (100 MHz): 5 = 22.2 (-ve, C-14 or C-15), 22.6 (-ve, C-14 or C-15), 25.7 (+ve), 27.2 (+ve), 29.9 (+ve), 31.8 (-ve), 34.8 (+ve), 39.2 (+ve), 48.7 (-ve), 51.3 (-ve), 55.3 (-ve), 57.0 (+ve, C-1), 57.4 (-ve), 58.4 (-ve), 70.8 (+ve, C-12), 174.1 (+ve, C-10), 211.8 (+ve, C-4). Exact mass calculated for C17H28O4: 296.1988; found: 296.1987 4. Experimental 263 Alkene 321 exhibited: IR (film): v = 2951, 1723, 1459, 1365, 1164, 1117 cm"1. 1H NMR (400 MHz): 5 = 0.85 (d, 3 H, J= 6.5 Hz, Me-14 or Me-15), 0.95-1.04 (m, 1 H), 1.01 (d, 3 H, J= 6.5 Hz, Me-14 or Me-15), 1.25-1.35 (ddd, 1 H, J= 11.9, 7.4, 7.4 Hz), 1.35-1.57 (m, 3 H), 1.68-1.76 (m, 1 H), 1.79-1.97 (m, 3 H), 2.17-2.37 (m, 3 H), 2.41-2.52 (m, 1 H), 2.64-2.71 (ddd, 1 H, J= 12.6, 3.4, 3.4 Hz, H-5B), 3.30 (s, 3 H, Me-1'), 3.33-3.38 (m, 2 H, H-11), 3.60 (s, 3 H, Me-1"). 1 3C NMR (100 MHz): 5 = 20.7 (+ve), 22.2 (-ve, C-14 or C-15), 22.7 (-ve, C-14 or C-15), 27.3 (+ve), 27.4 (+ve), 28.5 (+ve), 30.9 (-ve), 33.0 (+ve), 34.1 (+ve), 51.1 (-ve), 56.0 (+ve, C-6), 58.5 (-ve), 58.7 (-ve), 71.0 (+ve, C-11), 126.7 (+ve, C-2), 139.0 (+ve, C-1), 175.2 (+ve, C-12). Exact mass calculated for Ci 7H 280 3: 280.2038; found: 280.2034. 4. Experimental 264 Methyl (1S*, AR*, 5S*, 6S*, 9/?*)-4-formyl-9-isopropyl-5-(2-methoxyethyl)bicyclo [4.3.0]nonane-1-carboxylate (310) and Methyl (1S*, AS*, 5S*, 6S*, 9/7*)-4-formyl-9-isopropyl-5-(2-methoxyethyl)bicyclo [4.3.0]nonane-1 -carboxylate (327) To a cold (0 °C) solution of Me 3 ST (885 mg, 4.33 mmol) in THF (20 mL) was added dry KHMDS (865 mg, 4.33 mmol) in portions over 30 min. The resulting reaction mixture was stirred for 1 h. A solution of ketone 309 (988 mg, 3.33 mmol) in THF (15 mL) was transferred to the reaction vessel via a cannula. The reaction mixture was stirred for 30 min. H 20 (10 mL) was added and the mixture was concentrated. The residue was partitioned between Et20 (20 mL) and H 20 (20 mL). The layers were separated and the aqueous layer was extracted with Et20 (20 mL). The combined organic extracts were washed with H 20 (10 mL), brine (20 mL), dried (MgS04) and concentrated. The residual oil was purified using flash column chromatography (50 g of silica gel, 7:3 petroleum ether-Et20) to afford an inseparable mixture of epoxides 325 and 326. The acquired mixture of epoxides 325 and 326 was dissolved in DCM (10 mL) and the resulting solution was cooled to -78 °C. A cold (-78 °C) solution of MABR 1 4 0 (22 mL of a 0.18 M solution in DCM, 3.67 mmol) was transferred to the reaction vessel via a cannula. The resulting mixture was stirred for 1 h. 10% aq HCI (20 mL) and DCM (10 mL) were added and the layers were separated. The organic layer was washed with sat. aq NaHC03 (20 mL), dried (MgS04), and concentrated. The residual oil was purified by flash column chromatography (50 g of silica gel, 7:3 petroleum ether-Et20) 309 325 326 310: RT = H, R 2 = C H O 327: RT = C H O , R 2 = H 4. Experimental 265 to yield an inseparable mixture of aldehydes 310 and 327 (ratio 3:1, 0.981 g, 95%) as a clear oil. The mixture of aldehydes 310 and 327 exhibited: IR (film): v = 2951, 2872, 1720, 1705, 1456, 1169 cm"1. Exact mass calculated for Ci 8H 3o0 4: 310.2144; found: 310.2143. Aldehyde 310 exhibited 1H NMR (400 MHz): 6 = 0.79 (d, 3 H, J= 6.5 Hz, Me-15 or Me-16), 0.95 (d, 3 H, J= 6.4 Hz, Me-15 or Me-16), 1.10-1.95 (series of m, 12 H), 1.99-2.07 (m, 1 H), 2.51-2.57 (ddd, 1 H, J= 13.3, 3.3, 3.3 Hz), 2.53-2.63 (m, 1 H), 3.27 (s, 3 H, Me-1"), 3.38-3.49 (m, 2 H), 3.64 (s, 3 H, Me-1'), 9.85 (s, 1 H, H-11). 1 3C NMR (100 MHz): 8 = 22.3 (-ve, C-15 or C-16), 22.7 (-ve, C-15 or C-16), 24.1 (+ve), 25.3 (+ve), 28.3 (+ve), 30.2 (+ve), 31.8 (-ve), 33.4 (+ve), 37.1 (-ve), 48.4 (-ve), 50.8 (-ve), 52.3 (-ve), 57.6 (+ve, C-1), 58.1 (-ve), 58.6 (-ve), 71.1 (+ve, C-13), 174.5 (+ve, C-10), 205.6 (-ve, C-11). 4. Experimental 266 Compound 327 exhibited: 13 1H NMR (400 MHz): 5 = 0.81 (d, 3 H, J= 6.9 Hz, Me-15 or Me-16), 0.98 (d, 3 H, J= 6.3 Hz, Me-15 or Me-16), 1.10-1.95 (series of m, 13 H), 2.06-2.16 (m, 1 H), 2.68-2.75 (ddd, 1 H, J= 12.9, 3.3, 3.3 Hz), 3.22 (s, 3 H, Me-1"), 3.25-3.38 (m, 2 H, H-13), 3.66 (s, 3 H, Me-1'), 9.47 (d, 1 H, J= 3.9 Hz, H-11). 1 3C NMR (100 MHz): 8 = 22.3 (-ve, C-15 or C-16), 22.7 (-ve, C-15 or C-16), 24.7 (+ve), 25.3 (+ve), 28.3 (+ve), 31.8 (-ve), 32.2 (+ve), 34.8 (-ve), 35.3 (+ve), 50.9 (-ve), 54.1 (-ve), 55.2 (-ve), 57.0 (+ve, C-1), 58.0 (-ve), 58.4 (-ve), 69.9 (+ve, C-13), 174.3 (+ve, C-10), 204.5 (-ve, C-11). 4. Experimental 267 Methyl (1 S*,4E ,5S*,6S*,9/?*)-9-isopropyl-5-(2-methoxyethyl)-4-tert-butyldimethylsiloxymethylenebicyclo[4.3.0]nonane-1 -carboxylate (342) and Methyl (1 S*,4Z ^ ^.e^^fTJ-g-isopropyl-S^-methoxyethylH-ferf-butyldimethylsiloxymethylenebicyclo[4.3.0]nonane-1-carboxylate (343) 310: RT = H, R 2 = C H O 342: RT = H, R 2 = O T B S 327: RT = C H O , R 2 = H 343: RT = O T B S , R 2 = H To a cold (-78 °C), stirred solution of KHMDS (1.93 g, 9.68 mmol) in THF (20 mL) was added a cold (-78 °C) solution of aldehydes 310 and 327 (2.00 g, 6.45 mmol) in THF (10 mL). The resulting mixture was allowed to warm to 0 °C over 1 h. A cold (0 °C) solution of TBSCI (1.46 g, 9.68 mmol) in THF (20 mL) was transferred to the reaction vessel via a cannula. The resulting mixture was allowed to warm to r.t. overnight. H 20 (50 mL) and Et20 (50 mL) were added and the layers were separated. The aqueous layer was washed with Et20 (50 mL). The combined organic extracts were washed with brine (50 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (100 g of silica gel, 9:1 petroleum ether-Et20) to yield silyl enol ethers 342 and 343 (2.06 g, 75%). Although the two silyl enol ethers (ratio 0.8:1) could be separated and characterized independently, no effort was made to establish the geometry about the double bond of the major product. The major isomer exhibited: IR (film): v = 2954, 2860, 1724, 1666, 1462, 1255, 1161,839 cm"1. 4. Experimental 268 1H NMR (400 MHz): 6 = 0.07 (s, 6 H, Si-CH3), 0.78 (d, 3 H, J= 6.4 Hz, Me-15 or Me-16), 0.88 (s, 9 H, Si-C(CH3)3), 0.95 (d, 3 H, J= 6.3 Hz, Me-15 or Me-16), 1.08-1.17 (ddd, 1 H, J= 13.0, 10.3, 7.0 Hz), 1.16-1.64 (m, 5 H), 1.65-1.98 (m, 5 H), 2.05-2.13 (m, 1 H), 2.49-2.59 (ddd, 1 H, J= 10.3, 6.3, 4.3 Hz), 2.61-2.70 (ddd, 1 H, J= 12.6, 6.2, 4.0 Hz), 3.28 (s, 3 H, Me-1"), 3.30-3.40 (m, 2 H, H-13), 3.60 (s, 3 H, Me-1'), 6.00 (s, 1 H, H-11). 1 3C NMR (100 MHz): 8= -5.3 (2 C, -ve, Si-CH3), 18.2 (+ve, Si-C(CH3)3), 21.4 (+ve), 22.3 (-ve, C-15 or C-16), 22.6 (-ve, C-15 or C-16), 25.7 (3 C, -ve, Si-C(CH3)3), 25.9 (+ve), 29.1 (+ve), 30.4 (+ve), 31.7 (-ve), 35.7 (+ve), 38.3 (-ve), 50.7 (-ve), 55.9 (-ve), 57.5 (+ve, C-1), 58.1 (-ve), 58.5 (-ve), 71.1 (+ve, C-13), 121.6 (+ve, C-4), 132.7 (-ve, C-11), 174.9 (+ve, C-10). Exact mass calculated for C 2 4H 4 4 0 4 S i : 424.3008; found: 424.3009. The minor isomer exhibited: 1H NMR (400 MHz): 5 = 0.08 (s, 6 H, Si-CH3), 0.79 (d, 3 H, J= 6.5 Hz, Me-15 or Me-16), 0.90 (s, 9 H, Si-C(CH3)3), 0.91 (d, 3 H, J= 6.5 Hz, Me-15 or Me-16), 1.11-1.32 (m, 2 H), I. 32-1.46 (m, 2 H), 1.46-1.60 (m, 2 H), 1.64-1.98 (m, 5 H), 2.04-2.17 (m, 1 H), 2.62-2.72 (m, 1 H), 2.67-2.76 (ddd, 1 H, J= 13.1, 10.0, 7.7 Hz), 3.28 (s, 3 H, Me-1"), 3.29-3.43 (m, 2 H, H-13), 3.57 (s, 3 H, Me-16), 5.97 (s, 1 H, H-11). 1 3C NMR (100 MHz): 5 = -5.2 (-ve, Si-CH3), -5.1 (-ve, Si-CH3), 18.0 (+ve, Si-C(CH3)3), 22.1 (-ve, C-15 or C-16), 22.6 (-ve, C-15 or C-16), 22.9 (+ve), 25.6 (3 C, -ve, Si-C(CH3)3), 27.1 (+ve), 30.2 (+ve), 31.8 (-ve), 33.6 (+ve), 35.2 (+ve), 36.2 (-ve), 50.7 (-ve), 52.5 (-ve), 56.2 (+ve, C-1), 58.5 (-ve), 58.9 (-ve), 72.4 (+ve, C-13), 119.2 (+ve, C-4), 134.0 (-ve, C-11), 175.3 (+ve, C-10). No IR or mass spectral data were collected on this material. 4. Experimental 269 Methyl (l^^'AT^^.S^.e^.g/rj-g-isopropyl-S^-methoxyethyl^'-terf-butyl dimethylsiloxybicyclo[4.3.0]nonane-4-spiro-1 'cyclopropane-1 -carboxylate (344) and Methyl (1S*,2'S*,4S*,5S*,6S*,9/?*)-9-isopropyl-5-(2-methoxyethyl)-2'-fert-butyl dimethylsiloxybicyclo[4.3.0]nonane-4-spiro-1 'cyclopropane-1 -carboxylate (345) 342: Pi! = H , R 2 = O T B S 344: R, = H , R 2 = O T B S 343: R , = O T B S , R 2 = H 345: R , = O T B S , R 2 = H To a solution of silyl enol ethers 342 and 343 (2.737 g, 6.45 mmol) and diethylzinc (16.1 mL of a 1.0 M solution in hexanes, 16.1 mmol) in Et20 (25 mL) at r.t. was added, drop-wise, via a syringe, diiodomethane (6.91 g, 2.08 mL, 25.8 mmol). The resulting reaction mixture was allowed to stir for 4 h. Et20 (100 mL) and 1 M aq HCI (25 mL) were added and the layers were separated. The organic layer was washed successively with sat. aq NaHC03 (25 mL) and brine (25 mL). The organic extracts were dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (125 g of silica gel, 9:1 petroleum ether-Et20) to afford an inseparable mixture of spirocyclopropyl silyl ethers 344 and 345 (ratio 0.8:1, 1.634 g, 60%) as a clear oil. A mixture of silyl enol ethers 342 and 343 (245 mg, 9%) was also recovered from the reaction mixture. A clean sample of the major spirocyclopropyl silyl enol ether was synthesized from the major silyl enol ether following the procedure described above. This material exhibited: IR (film): v = 2953, 2860, 1725, 1462, 1197, 1157, 1120, 838 cm 4. Experimental 270 1H NMR (400 MHz): 8 = 0.02 (s, 3 H, Si-CH3), 0.04 (s, 3 H, Si-CH3), 0.11-0.15 (dd, 1 H, J= 6.2, 3.4 Hz, H-3'), 0.63-0.80 (dd, 1 H, J= 6.7, 6.2 Hz, H-3'), 0.80 (d, 3 H, J= 6.5 Hz, Me-14 or Me-15), 0.84 (s, 9 H, Si-C(CH3)3), 0.98-1.08 (m, 1 H), 0.99 (d, 3 H, J= 6.5 Hz, Me-14 or Me-15), 1.10-1.48 (m, 5 H), 1.48-1.78 (m, 5 H), 1.80-2.00 (m 2 H), 2.48-2.56 (ddd, 1 H, J= 12.7, 3.6, 3.6 Hz, H-2p), 3.13-3.21 (ddd, 1 H, J= 9.3, 9.3, 6.5 Hz, H-12), 3.21-3.25 (dd, 1 H, J= 6.7, 3.4 Hz, H-2'), 3.26 (s, 3 H, Me-1'"), 3.32-3.40 (ddd, 1 H, J = 9.3, 9.3, 5.4 Hz, H-14), 3.64 (s, 3 H, Me-1"). 1 3C NMR (100 MHz): 8 = -5.0 (-ve, Si-CH3), -4.9 (-ve, Si-CH3), 17.8 (+ve, Si-C(CH3)3> 22.4 (-ve, C-14 or C-15), 22.6 (-ve, C-14 or C-15), 25.6 (+ve), 25.7 (-ve), 25.8 (-ve, 3 C, Si-C(CH3)3), 26.0 (+ve), 28.4 (+ve), 28.6 (+ve), 28.6 (+ve), 31.9 (-ve), 35.6 (+ve), 37.9 (-ve), 50.6 (-ve), 52.7 (-ve), 56.4 (-ve), 57.9 (-ve), 58.0 (+ve, C-1), 58.5 (-ve), 72.8 (+ve, C-12), 175.1 (+ve, C-10). Exact mass calculated for C25H4604Si: 438.3165; found: 438.3168. 4. Experimental 271 Methyl (1 S*,4S*,5S*,6S*,9ff*)-4-formyl-9-isopropyl-5-(2-methoxyethyl)-4-methylbicyclo[4.3.0]nonane-1 -carboxylate (311) H H ,»>CHO 344: R 1 = H, R 2 = OTBS 345: Ri = OTBS, R 2 = H 346: R, = H, R 2 = OH 347: R, = OH, R 2 = H 311 To a solution of spirocyclopropyl silyl ethers 344 and 345 (1.63 g, 3.73 mmol) in THF (25 mL) was added a solution of TBAF (5 mL of a 1.0 M solution in THF, 5 mmol) and the resulting reaction mixture was stirred for 7 h. H 20 (25 mL) and Et20 (25 mL) were added and the layers were separated. The aqueous layer was extracted with Et20 (25 mL). The combined organic extracts were washed with brine (25 mL) and concentrated to furnish a crude mixture of spirocyclopropanols 346 and 347. The acquired crude mixture of spirocyclopropanols 346 and 347 was dissolved in THF (20 mL) and concentrated aq HCI (-10 M, 1 mL) was added. The resulting reaction mixture was heated to 70 °C for 1 h. Et20 (25 mL) and H 20 (25 mL) were added and the layers were separated. The aqueous layer was extracted with Et20 (25 mL). The combined organic extracts were washed with brine (25 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (100 g of silica gel, 7:3 petroleum ether-Et20) to provide aldehyde 311 (665 mg, 55% over two steps) as a colourless oil. 4. Experimental 272 1" 14 311 Aldehyde 311 exhibited: IR (film): v = 2950, 2873, 1721, 1460, 1365, 1194, 1118 cm"1. 1H NMR (400 MHz): 5 = 0.78 (d, 3 H, J=6.4 Hz, Me-16 or Me-17), 0.94 (d, 3 H, J=6.3 Hz, Me-16 or Me-17), 1.04 (s, 3 H, Me-12), 1.14-1.46 (m, 6 H), 1.61-1.74 (m, 2 H), 1.77-1.98 (m, 5 H), 2.44-2.52 (ddd, 1 H, J= 12.9, 3.2, 3.2 Hz, H-28), 3.22-3.31 (m, 1 H, H-14), 3.28 (s, 3 H, Me-1"), 3.37-3.45 (ddd, 1 H, J= 9.1, 9.1, 5.1 Hz, H-14), 3.64 (s, 3 H, Me-1'), 9.58 (s, 1 H, H-11). 1 3C NMR (100 MHz): 5 = 22.2 (-ve), 22.3 (-ve), 22.6 (-ve), 25.7 (+ve), 28.4 (+ve), 29.5 (+ve), 31.8 (-ve), 33.5 (+ve), 33.7 (+ve), 42.1 (-ve), 49.9 (+ve, C-4), 50.7 (-ve), 53.3 (-ve), 57.6 (-ve), 57.7 (+ve, C-1), 58.5 (-ve), 73.0 (+ve, C-14), 174.5 (+ve, C-10), 206.7 (-ve, C-11). Exact mass calculated for Ci9H3204: 324.2301; found: 324.2300. 4. Experimental 273 Methyl (1S*,4S*,5S*,6S*,9/?*)-4-formylmethyl-9-isopropyl-5-(2-methoxyethyl)-4-methylbicyclo[4.3.0]nonane-1 -carboxylate (318) 311 348 349 318 To a cold (0 °C) solution of Me 3 ST (1.05 g, 5.12 mmol) in THF (40 mL) was added dry KHMDS (1.02 g, 5.12 mmol) in portions over 30 min. The resulting reaction mixture was stirred for 1 h. A solution of aldehyde 311 (1.11 g, 3.42 mmol) in THF (15 mL) was transferred to the reaction vessel via a cannula. The reaction mixture was stirred for 30 min. H 20 (10 mL) was added and the mixture was concentrated. The residue was partitioned between Et20 (20 mL) and H 20 (20 mL). The layers were separated and the aqueous layer was extracted with Et20 (20 mL). The combined organic extracts were washed with H 20 (10 mL), brine (20 mL), dried (MgS04) and concentrated. The residual oil was purified using flash column chromatography (50 g of silica gel, 7:3 petroleum ether-Et20) to afford an inseparable mixture of epoxides 348 and 349. The acquired mixture of epoxides 348 and 349 was dissolved in DCM (10 mL) and the resulting solution was cooled to -78 °C. A cold (-78 °C) solution of MABR 1 4 0 (27 mL of a 0.15 M solution in DCM, 4.1 mmol) was transferred to the reaction vessel via a cannula. The resulting mixture was stirred for 1 h. 10% aq HCI (20 mL) and DCM (10 mL) were added and the layers were separated. The organic layer was washed with sat. aq NaHC03 (20 mL), dried (MgS04), and concentrated. The residual oil was purified by flash column chromatography (50 g of silica gel, 7:3 petroleum ether-Et20) to yield aldehyde 318 (0.923 g, 89% over two steps) as a clear oil. 4. Experimental 274 1" 15 318 Aldehyde 318 exhibited: IR (film): v = 2951, 1720, 1457, 1364, 1192 cm"1. 1H NMR (400 MHz): 5 = 0.80 (d, 3 H, J= 6.5 Hz, Me-17 or Me-18), 0.98 (d, 3 H, J= 6.4 Hz, Me-17 or Me-18), 1.10 (s, 3 H, Me-11), 1.19-1.45 (m, 7 H), 1.56-1.72 (m, 4 H), 1.82-2.03 (m, 2 H), 2.14-2.21 (dd, 1 H, J= 14.0, 1.2 Hz), 2.42-2.52 (m, 2 H), 3.23-3.32 (m, 1 H, H-15), 3.30 (s, 3 H, Me-1"), 3.41-3.49 (ddd, 1 H, J= 9.3, 9.3, 4.6 Hz, H-15), 3.65 (s, 3 H, Me-1'), 9.81 (dd, 1 H, J= 3.9, 2.7 Hz, H-13). 1 3C NMR (100 MHz): 8 = 22.3 (-ve, C-17 or C-18), 22.7 (-ve, C-17 or C-18), 25.5 (+ve), 27.0 (-ve), 28.4 (+ve), 29.6 (+ve), 31.8 (-ve), 32.4 (+ve), 36.4 (+ve), 37.5 (+ve, C-4), 45.0 (-ve), 45.4 (+ve, C-12), 50.8 (-ve), 51.4 (-ve), 57.6 (-ve), 57.9 (+ve, C-1), 58.6 (-ve, C-1'), 73.0 (+ve, C-15), 174.6 (+ve, C-10), 204.0 (-ve, C-13). Exact mass calculated for C20H34O4: 338.2457; found: 338.2460. 4. Experimental 275 Methyl (1S*,4S*,5S*,6S*,9ff*)-4-allyl-9-isopropyl-5-(2-methoxyethyl)-4-methyl bicyclo[4.3.0]nonane-1 -carboxylate (317) To a cold (0 °C) solution of aldehyde 318 (88 mg, 0.26 mmol) in THF (2 mL) was added Tebbe's reagent (146, 573 uL of a 0.5 M solution in toluene, 0.29 mmol). The resulting reaction mixture was allowed to warm up to r.t. over 1 h. The reaction mixture was diluted with Et20 (5 mL) and aq NaOH (500 u.L of a 0.1 M solution) was added in a dropwise fashion via a syringe. H 20 (5 mL) was added and the layers were separated. The organic layer was dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (5 g of silica gel, 19:1 petroleum ether-Et20) to afford alkene 317 (70 mg, 80%) as a clear oil which exhibited: IR (film): v = 2951, 2877, 1722, 1637, 1458, 1191, 1157, 1119 cm"1. 1H NMR (400 MHz): 8 = 0.79 (d, 3 H, J= 6.6 Hz, Me-18 or Me-19), 0.82 (s, 3 H, Me-11), 0.95-1.05 (m, 1 H), 0.98 (d, 3 H, J=6.4 Hz, Me-18 or Me-19), 1.14-1.42 (m, 6 H), 1.49-1.56 (ddd, 1 H, J= 14.4, 3.1, 3.1 Hz), 1.56-1.77 (m, 3 H), 1.77-1.85 (m, 1 H), 1.86-1.99 (m, 2 H), 2.15-2.25 (dd, 1 H, J= 13.6, 8.1 Hz), 2.33-2.41 (ddd, 1 H, J= 13.1, 3.4, 3.4 Hz, H-2p), 3.21-3.29 (ddd, 1 H, J= 9.1, 9.1, 6.6 Hz, H-16), 3.29 (s, 3 H, Me-1"), 3.41-3.49 (ddd, 1 H, J= 10.5, 9.1, 5.0 Hz, H-16), 3.63 (s, 3 H, Me-1'), 4.92-5.02 (m, 2 H, H-14), 5.66-5.79 (dddd, 1 H, J= 17.0, 10.2, 8.0, 6.9 Hz, H-13). 19 318 146 317 4. Experimental 276 1 dC NMR (100 MHz): 5 = 22.3 (-ve, C-18 or C-19), 22.7 (-ve, C-18 or C-19), 25.6 (+ve), 26.5 (-ve, C-11), 28.5 (+ve), 29.3 (+ve), 31.8 (-ve), 32.4 (+ve), 34.9 (+ve), 35.4 (+ve), 36.7 (+ve, C-4), 44.6 (-ve), 50.6 (-ve), 51.2 (-ve), 57.7 (-ve), 58.1 (+ve, C-1), 58.5 (-ve), 73.6 (+ve, C-16), 116.9 (+ve, C-14), 135.5 (-ve, C-13), 174.9 (+ve, C-10). Exact mass calculated for C21H36O3: 336.2665; found: 336.2664. 4. Experimental 277 Methyl (1 S*,2S*,5ff*,6S*,9S*,11 R*)-5-isopropyl-9,11 -dimethyl-12-oxabicyclo [7,5,0,02'6]tetradecane-6-carboxylate (372) and Methyl (1 S*,2S*,5/7*,6S*,9S*,11 S*)-5-isopropyl-9,11 -dimethyl-12-oxabicyclo [7,5,0,02,6]tetradecane-6-carboxylate (373) 317 372 373 To a stirred solution of alkene 317 (20 mg, 0.059 mmol) in DCM (5 mL) was added, via a syringe, neat TMSI (100 u,L, 0.70 mmol). The reaction mixture was allowed to stir for 3 h. TLC analysis of the reaction mixture mixture indicated that the starting material had been consumed. Et20 (5 mL) was added and the reaction mixture was allowed to stir for 30 min. H 20 was added and the layers were separated. The organic layer was washed with brine (5 mL) dried (MgS04) and concentrated. The residual material was purified by flash column chromatography (1 g of silica gel, 9:1 petroleum ether-Et20) to yield the major tricyclic ether (372 or 373, 12 mg, 63%) as a clear and colourless oil. The major isomer exhibited 1H NMR (400 MHz): 5 = 0.80 (d, 3 H, J= 6.5 Hz, Me-16 or Me-17), 0.91 (s, 3 H, Me-19), 0.96-1.10 (m, 1 H), 1.00 (d, 3 H, J=6.4 Hz, Me-16 or Me-17), 1.10-1.46 (m, 6 H), 1.48-1.74 (m, 4 H), 1.84-1.98 (m, 3 H), 1.95 (d, 3 H, J= 6.8 Hz, Me-20), 2.41-2.50 (m, 2 H), 3.44-3.48 (m, 1 H, H-13), 3.62 (s, 3 H, Me-1'), 3.63-3.68 (ddd, 1 H, J= 9.8, 9.8, 5.9 Hz, H-13), 4.20-4.26 (ddq, 1 H, J=6.8, 6.8, 6.8 Hz, H-11). 4. Experimental 278 l d C NMR (100 MHz): 5 = 22.3 (-ve, C-16 or C-17), 22.7 (-ve, C-16 or C-17), 23.9 (-ve), 25.6 (+ve), 26.6 (-ve), 28.4 (+ve), 31.9 (-ve), 32.3 (+ve), 32.5 (-ve), 33.1 (+ve), 34.0 (+ve), 38.5 (+ve, C-9), 44.6 (+ve), 45.6 (-ve), 50.7 (-ve), 51.1 (-ve), 57.6 (-ve), 58.0 (+ve, C-6), 63.2 (+ve, C-13), 174.9 (+ve, C-18) 4. Experimental 279 Methyl (1 S*,4S*,5S*,6S*,9f?*)-4-formyl-5-(2-iodoethyl)-9-isopropyl -4-methylbicyclo[4.3.0]nonane-1 -carboxylate (375) 14 To a cold (-78 °C), stirred solution of ether 311 (3.60 g, 11.1 mmol) in DCM (100 mL) was added, via a syringe, a cold solution of TMSI (5 g, 25.0 mmol). The resulting reaction mixture was allowed to warm up to r.t. overnight. Et20 (50 mL) was added and the reaction mixture was allowed to stir for 1 h. H 20 (50 mL) was added and the layers were separated. The organic layer was washed with brine (50 mL), dried (MgSG*4) and concentrated. Attempts to purify the material by flash column chromatography on silica gel (200 g of silica gel, 9:1 petroleum ether-Et20) invariably resulted in the isolation of an orange oil (3.46 g, ~ 74%) which appeared clean by 1H NMR spectroscopy. The acquired material was stored at low temperature (-5 °C) and used in the next reaction without further purification. 1H NMR (400 MHz): 5 = 0.79 (d, 3 H, J= 6.3 Hz, Me-16 or Me-17), 0.95 (d, 3 H, J= 6.4 Hz, Me-16 or Me-17), 1.08 (s, 3 H, Me-12), 1.11-1.21 (ddd, 1 H, J= 13.6, 13.6, 3.6 Hz, H-2a), 1.23-1.50 (m, 5 H), 1.64-1.75 (m, 1 H), 1.83-2.08 (m, 5 H), 2.09-2.19 (m, 1 H), 2.47-2.55 (ddd, 1 H, J= 13.6, 3.6, 3.6 Hz, H-2p), 3.06-3.15 (ddd, 1 H, J= 8.6, 8.6, 8.6 Hz, H-14), 3.29-3.37 (ddd, 1 H, J= 8.6, 8.6, 4.9 Hz, H-14), 3.68 (s, 3 H, Me-1'), 9.48 (s, 1 H, H-11). 4 . Experimental 280 Methyl (1 S*,2S*,5R*,6S*,9ff*)-5-isopropyl-9-methyl-10-tert-butoxy-11-oxabicyclo [7.4.0.02'6]tridecane-6-carboxylate (385) 12 375 385 To a stirred solution of iodide 375 (43 mg, 0.10 mmol) in THF (5 mL) at r.t was added t-BuOK (17 mg, 0.15 mmol) in one portion. The resulting heterogeneous reaction mixture was allowed to stir for 30 min. H 20 (5 mL) and Et20 (5 mL) were added and the phases were separated. The organic phase was washed with brine (5 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (4 g of silica gel, 93:7 petroleum ether-Et20) to furnish mixed acetal 385 (32 mg, 87%) as a clear oil which exhibited: IR (film): v = 2972, 1721, 1508, 1364, 1165, 1088 cm"1. 1H NMR (400 MHz): 6 = 0.82 (d, 3 H, J= 6.5 Hz, Me-14 or Me-15), 0.94 (s, 3 H, Me-17), 0.94-1.06 (m, 1 H), 1.00 (d, 3 H, J=6.4 Hz, Me-14 or Me-15), 1.07-1.34 (m, 3 H), 1.19 (s, 9 H, 0-C(CH3)3), 1.39-1.50 (m, 2 H), 1.54-1.82 (m, 4 H), 1.87-1.96 (m, 1 H), 1.94-2.05 (m, 1 H), 2.12-2.23 (ddd, 1 H, J= 12.2, 12.2, 9.0 Hz), 2.38-2.45 (ddd, 1 H, J= 13.0, 3.6, 3.6 Hz, H-7p), 3.52-3.65 (m, 1 H, H-12), 3.65 (s, 3 H, Me-1'), 3.70-3.77 (m, 1 H, H-12), 4.80 (s, 1 H, H-10). 1 3C NMR (100 MHz): 5 = 21.3 (-ve), 22.3 (-ve), 22.7 (-ve), 24.1 (+ve), 25.6 (+ve), 28.7 (+ve), 28.8 (3 C, -ve, 0-C(CH3)3), 31.9, (-ve), 32.8 (+ve), 33.4 (+ve), 37.1 (+ve, C-9), 42.2 (-ve), 47.9 (-ve), 50.7 (-ve), 57.6 (+ve, C-6), 58.5 (-ve, C-1'), 61.7 (+ve, C-12), 74.3 (+ve, 0-C(CH3)3), 94.9 (-ve, C-10), 174.8 (+ve, C-16). Exact mass calculated for C 2 2 H 3 8 0 4 : 366.2770; found: 366.2769. 4. Experimental 281 Methyl (1S*,4r7*,5S*,6S*,9r?*)4-formyl-9-isopropyl-4-methyl-5-(2-phenylselanyl-ethyl)bicyclo[4.3.0]nonane-1 -carboxylate (378) To a stirred solution of diphenyl diselenide (1.54 g, 4.93 mmol) in EtOH (25 mL) at r.t. was added NaBH4 (373 mg, 9.87 mmol) in three portions over 10 minutes. The reaction mixture was stirred until the yellow solution became colourless. A solution of iodide 375 (3.456 g, 8.23 mmol) in EtOH (10 mL) was transferred to the reaction vessel via a cannula and the reaction mixture was heated to reflux for 2 h. H 20 (30 mL) and Et20 (30 mL) were added and the layers were separated. The aqueous layer was washed with Et20 (30 mL). The combined organic extracts were washed with brine (30 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (175 g of silica gel, 9:1 petroleum ether-Et20) to furnish aldehyde 378 (3.46 g, 94%) as a clear oil which exhibited: IR (film): v = 2948, 1720, 1476, 1364, 1148 cm"1. 1H NMR (400 MHz): 5 = 0.78 (d, 3 H, J= 6.2 Hz, Me-16 or Me-17), 0.95 (d, 3 H, J= 6.2 Hz, Me-16 or Me-17), 0.97 (s, 3 H, Me-11), 1.13-1.22 (ddd, 1 H, J= 13.4, 13.4, 3.4 Hz), 1.22-1.46 (m, 4 H), 1.47-1.55 (m, 1 H), 1.55-1.67 (m, 1 H), 1.76-2.02 (m, 6 H), 2.45-2.53 (ddd, 1 H, J= 13.1, 3.2, 3.2 Hz, H-2(3), 2.75-2.85 (ddd, 1 H, J= 11.9, 9.8, 7.2 Hz, H-14), 3.03-3.12 (ddd, 1 H, J= 11.9, 10.2, 5.1 Hz, H-14), 3.66 (s, 3 H, Me-1'), 7.18-7.26 (m, 3 H), 7.42-7.48 (m, 2 H), 9.51 (s, 1 H, H-12). 17 375 378 4. Experimental 282 1 J C NMR (100 MHz): 8 = 22.2 (-ve), 22.3 (-ve), 22.6 (-ve), 25.8 (+ve), 28.5 (+ve), 28.6 (+ve), 30.3 (+ve), 31.8 (-ve), 33.6 (+ve), 33.6 (+ve), 46.3 (-ve), 50.2 (+ve, C-4), 50.8 (-ve), 53.2 (-ve), 57.5 (-ve), 57.7 (+ve, C-1), 126.8 (-ve, C-1"), 129.0 (2 C, -ve, C-2" and C-6"), 130.1 (+ve, C-1'), 132.6 (2 C, -ve, C-3" and C-5"), 174.5 (+ve, C-10), 206.4 (-ve, C-12). Exact mass calculated for C24H34O3 8 0Se: 450.1673; found: 450.1673. 4. Experimental 283 Methyl (1 S*,2S*,5R*,6S*,9R*)-5-isopropyl-9-methyl-10-oxo-11 -oxatricyclo[7.4.0.02'6]tridecane-6-carboxylate (390) 378 390 A stream of O2 and O 3 was passed thorugh a cold (-78 °C), stirred solution of selenide 378 (55 mg, 0.12 mmol) in DCM (5 mL) until a grey-blue colour persisted. Benzene (5 mL) was added and the resulting solution was heated to reflux for 1 h. The resulting solution was allowed to cool to r.t. Et20 (10 mL) and H 20 (10%, 10 mL) were added and the layers were separated. The organic layer was washed with sat. aq. NaHC03 (10 mL) and brine (10 mL), dried (MgSCv) and concentrated. The residual oil was purified by flash column chromatography (5 g of silica gel, 19:1 petroleum ether-Et20) to provide lactone 390 (27 mg, 72%) as a clear and colourless oil. IR (film): v = 2952, 1724, 1448, 1166, 1013 cm"1. 1H NMR (400 MHz): 6 = 0.80 (d, 3 H, J=6.5 Hz), 0.99 (d, 3 H, J=6.4 Hz), 1.12-1.33 (m, 3 H), 1.26 (s, 3 H), 1.35-1.50 (m, 2 H), 1.54-1.71 (m, 3 H), 1.79-2.01 (m, 3 H), 2.26-2.39 (m, 1 H), 2.39-2.56 (m, 2 H), 3.66 (s, 3 H), 4.32-4.43 (m, 2 H). 1 3C NMR (100 MHz): 5 = 21.7 (+ve), 22.2 (-ve), 22.6 (-ve), 24.7 (+ve), 27.1 (-ve), 28.6 (+ve), 31.7 (-ve), 34.2 (+ve), 34.9 (+ve), 40.2 (-ve), 44.4 (+ve), 50.9 (-ve), 51.0 (-ve), 57.2 (+ve), 58.1 (-ve), 66.9 (+ve), 174.5 (+ve), 176.2 (+ve). 4. Experimental 284 To a stirred solution of aldehyde 378 (3.46 g, 7.69 mmol) in MeOH (20 mL) at r.t. was added NaBH4 (238 mg, 11.6 mmol). The reaction mixture was allowed to stir for 1 h. The reaction mixture was diluted with Et20 (20 mL) and was transferred to a separatory funnel. 10% aq HCI (20 mL) was added and the layers were separated. The aqueous layer was washed with Et20 (20 mL). The combined organic extracts were washed with brine (20 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (200 g of silica gel, 6:4 petroleum ether-Et20) to afford alcohol 393 (3.26 g, 93%) as a clear oil. A stream of 0 2 and 0 3 was passed thorugh a cold (-78 °C), stirred solution of alcohol 392 (3.26 g, 7.22 mmol) in DCM (25 mL) until a grey-blue colour persisted. Benzene (25 mL) and triethylamine (5 mL) were added and the resulting solution was heated to reflux for 1 h. The resulting solution was allowed to cool to r.t. Diethyl ether (25 mL) and aq. HCL (10%, 25 mL) were added. The layers were separated and the organic layer was washed successively with sat. aq. NaHC03 (25 mL) and brine (25 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography to provide alcohol 393 (1.68 g, 79% yield) as a clear and colourless oil. 4. Experimental 285 H NMR (400 MHz): 5 = 0.80 (d, 3 H, J= 6.5 Hz, Me-16 or Me-17), 0.86 (s, 3 H, Me-11), 1.00 (d, 3 H, J= 6.4 Hz, Me-17), 1.06-1.16 (ddd, 1 H, J= 14.1, 14.1, 3.5 Hz, H-2a), 1.20-1.34 (m, 3 H), 1.34-1.50 (m, 3 H), 1.64-1.75 (m, 1 H), 1.76-1.83 (ddd, 1 H, J = 14.7, 3.4, 3.4 Hz), 1.83-1.91 (m, 1 H), 2.01-2.09 (dd, 1 H, J= 12.1, 10.0 Hz), 2.42-2.50 (ddd, 1 H, J= 14.1, 3.4, 3.4 Hz, H-2p), 3.49 (d, 1 H, J= 10.8 Hz, H-12), 3.66 (s, 3 H, H-1'), 3.68 (d, 1 H, J= 10.8 Hz, H-12), 4.97-5.00 (m, 1 H, H-14), 5.00-5.05 (m, 1 H, H-14), 5.42-5.56 (ddd, 1 H, J= 17.0, 10.0, 10.0 Hz, H-13). The hydroxyl proton was not observed in this spectrum. 1 3C NMR (100 MHz): 5 = 22.3 (-ve, C-16 or C-17), 22.7 (-ve, C-16 or C-17), 25.0 (-ve, C-11), 25.8 (+ve), 28.6 (+ve), 31.9 (-ve), 32.9 (+ve), 33.2 (+ve), 38.3 (+ve, C-4), 49.5 (-ve), 50.7 (-ve), 52.8 (-ve), 57.6 (+ve, C-1), 58.1 (-ve), 65.5 (+ve, C-12), 116.7 (+ve, C-14), 137.4 (-ve, C-13), 174.7 (+ve, C-10). 4. Experimental 286 Methyl (l^^rr.S^.e^.SfrH-formyl-S-isopropyl^-methyl-S-vinylbicyclo^.S.O] nonane-1-carboxylate (377) H H 393 377 Alcohol 393 was oxidized to aldehyde 377 using the method developed by Ley. Thus, to a stirred solution of alcohol 377 (1.68g , 5.7 mmol) in DCM (25 mL) at r.t. was added, sequentially, TPAP (200 mg, 0.57 mmol) in one portion and NMO (995 mg, 8.5 mmol) in one portion. The mixture was allowed to stir overnight. The reaction mixture was filtered through a pad of silica gel (10 g) and the silica gel was rinsed with Et20. The collected eluate was concentrated and the residual oil was purified by flash column chromatography (100 g of silica gel, 19:1 petroleum ether-Et20) to yield aldehyde 377 (1.31 g, 4.5 mmol, 78%) as a clear oil which exhibited. IR (film): v = 2953, 1720, 1638, 1459, 1365, 1167 cm"1. 1H NMR (400 MHz): 5 = 0.79 (d, 3 H, J= 6.5 Hz, Me-16 or Me-17), 0.96 (d, 3 H, J= 6.4 Hz, Me-16 or Me-17), 0.99 (s, 3 H, Me-11), 1.21-1.44 (m, 5 H), 1.44-1.55 (m, 1 H), 1.64-1.77 (m, 1 H), 1.80-1.98 (m, 3 H), 2.05-2.15 (dd, 1 H, J= 11.6, 11.6 Hz), 2.47-2.56 (m, 1 H), 3.66 (s, 3 H, Me-1'), 5.05-5.12 (m, 2 H, H-14), 5.72-5.84 (m, 1 H, H-13), 9.56 (s, 1 H, H-12). 1 3C NMR (100 MHz): 5 = 22.2 (-ve), 22.3 (-ve), 22.7 (-ve), 25.7 (+ve), 28.6 (+ve), 31.8 (-ve), 32.7 (+ve), 33.5 (+ve), 49.5 (+ve, C-4), 50.8 (-ve), 50.9 (-ve), 52.2 (-ve), 57.2 (+ve, C-1), 58.0 (-ve), 118.0 (+ve, C-14), 136 (-ve, C-13), 174.5 (+ve, C-10), 206.5 (-ve,C-12). Exact mass calculated for C i 8 H 2 8 0 3 : 292.2038; found: 292.2038. 4. Experimental 287 Methyl (1 S*,4fl*,5S*,6S*,9ff*)-5-allyl-4-formyl-9-isopropyl-4-methylbicyclo[4.3.0] nonane-1-carboxylate (398) 18 377 394 398 To a cold (0 °C) solution of Me 3 ST (1.37 g, 6.72 mmol) in THF (50 mL) was added dry KHMDS (1.34 g, 6.72 mmol) in portions over 30 min. The resulting reaction mixture was stirred for 1 h. A solution of aldehyde 377 (1.31 g, 4.46 mmol) in THF (20 mL) was transferred to the reaction vessel via a cannula. The reaction mixture was stirred for 30 min. H 20 (10 mL) was added and the mixture was concentrated. The residue was partitioned between Et20 (25 mL) and H 20 (25 mL). The layers were separated and the aqueous layer was extracted with Et20 (25 mL). The combined organic extracts were washed with H 20 (15 mL), brine (25 mL), dried (MgS04) and concentrated. The residual oil was purified using flash column chromatography (70 g of silica gel, 9:1 petroleum ether-Et20) to afford an inseparable mixture of epoxides 394 (1.34 g, 98%). The acquired mixture of epoxides 394 was dissolved in DCM (10 mL) and the resulting solution was cooled to -78 °C. A cold (-78 °C) solution of MABR 1 4 0 (53 mL of a 0.10 M solution in DCM, 5.3 mmol) was transferred to the reaction vessel via a cannula. The resulting mixture was allowed to warm up to r.t. overnight. 10% aq HCI (30 mL) and DCM (20 mL) were added and the layers were separated. The organic layer was washed with sat. aq NaHC03 (30 mL), dried (MgS04), and concentrated. The residual oil was purified by flash column chromatography (70 g of silica gel, 9:1 petroleum ether-Et20) to yield aldehyde 398 (636 mg, 47%) as a clear oil. 4. Experimental 288 Aldehyde 398 exhibited: IR (film): v = 2952, 2873, 1719, 1459, 1366, 1165,910 cm"1. 1H NMR (400 MHz): 8 = 0.80 (d, 3 H, J= 6.4 Hz, Me-17 or Me-18), 0.96 (d, 3 H, J= 6.4 Hz, Me-17 or Me-18), 1.08 (s, 3 H, Me-11), 1.21-1.44 (m, 6 H), 1.59-1.73 (m, 2 H), 1.81-1.97 (m, 4 H), 2.21-2.31 (m, 1 H), 2.34-2.43 (dddd, 1 H, J= 15.2, 8.1, 3.7, 1.7 Hz), 2.45-2.56 (m, 1 H), 3.66 (s, 3 H, Me-1'), 4.91-4.96 (ddd, 1 H, J= 10:1, 3, 1.5 Hz, H-15a), 4.96-5.02 (ddd, 1 H, J = 17.0, 3.0, 1.7 Hz, H-15b), 5.74-5.87 (dddd, 1 H, J= 17.0, 10.1, 7.0, 7.0 Hz, H-14). 1 3C NMR (100 MHz): 8 = 22.3 (-ve), 22.6 (-ve), 22.7 (-ve), 25.9 (+ve), 28.6 (+ve), 31.8 (-ve), 33.6 (+ve), 33.8 (+ve), 34.1 (+ve), 46.0 (-ve), 49.9 (+ve, C-4), 50.8 (-ve), 52.9 (-ve), 57.7 (-ve), 57.9 (+ve, C-1), 115.4 (+ve, C15), 139.1 (-ve, C-14), 174.6 (+ve, C-10), 207.0 (-ve, C-12). Exact mass calculated for C19H30O3: 306.2195; found: 306.2194. 4 . Experimental 289 Methyl (1 S*,2S*,5/7*,6S*,9/7*510/7*)-10-hydroxy-5-isopropyl-9-methyltricyclo [7.5.0.026]tetradec-12-ene-6-carboxylate (416) a n d Methyl (1 S*,2S*,5R*,6S*,9R*,10S*)-10-hydroxy-5-isopropyl-9-methyltricyclo [7.5.0.02,6]tetradec-12-ene-6-carboxylate (417) 398 414: R, = H, R 2 = O H 403 416: = H, R 2 = O H 415: RT = O H , R 2 = H 417: RT = O H , R 2 = H To a cold (0 °C), stirred solution of aldehyde 398 (636 mg, 2.07 mmol) in THF (25 mL) was added, via a syringe, allyl magnesium bromide (2.1 mL of a 1.0 M solution in diethyl ether, 2.1 mmol). After 10 min, TLC analysis of the mixture indicated that all the starting material had been consumed. Sat. aq. NH4CI (15 mL) was added and the layers were separated. The organic layer was washed with brine (15 mL), dried (MgS04) and concentrated. The residual oil was purified by flash column chromatography (30 g of silica gel, 3:1 petroleum ether-Et20) to furnish a nearly 1:1 mixture of alcohols 414 and 415 (687 mg, 95%) as a clear oil. To a solution of alcohols 414 and 415 (111 mg, 0.32 mmol) in degassed DCM was added, via a cannula, a solution of catalyst 403 (36 mg, 0.044 mmol) in degassed DCM. The resulting solution was heated to reflux for 24 h. The reaction mixture was allowed to cool to r.t. and filtered through a pad of silica gel (2 g). The silica gel was rinsed with Et20. The combined eluate was concentrated and the residual oil was purified by flash column chromatography (7 g of silica gel, 7:3 petroleum ether, Et20) silica to provide a nearly 1:1 mixture of tricyclic alcohols 416 and 417 (90 mg, 88%) as a clear oil. 4. Experimental 290 Clean samples of alcohols 416 and 417 were obtained by flash column chromatography (silica gel, 7:3 petroleum ether-Et20). Compound 416 exhibited C O 2 C H 3 15 r 1H NMR (400 MHz): 5 = 0.79 (d, 3 H, J= 6.5 Hz, Me-18 or Me-19), 0.99 (d, 3 H, J= 6.3 Hz, Me-18 or Me-19), 0.98 (s, 3 H, Me-16), 1.00-1.10 (ddd, 1 H, J= 14.6, 13.3, 3.7 Hz, H-7a), 1.10-1.20 (m, 1 H), 1.20-1.31 (m, 1 H), 1.31-1.49 (m, 4 H), 1.51-1.60 (m, 1 H), 1.79-1.98 (m, 4 H), 1.98-2.05 (ddd, 1 H, J - 4 . 6 , 3.2, 3.2 Hz), 2.02-2.10 (m, 1 H), 2.39-2.46 (ddd, 1 H, J = 13.3, 3.4, 3.4 Hz, H-7p), 2.44-2.52 (m, 1 H), 2.55-2.65 (m, 1 H), 3.65 (s, 3 H, Me-1'), 3.84-3.92 (br. d, 1 H, J= 11.4 Hz, H-10), 5.62-5.76 (m, 2 H, H-12 and H-13). 1 3C NMR (100 MHz): 8 = 20.5 (-ve, C-16), 22.3 (-ve, C-18 or C-19), 22.7 (-ve, C-18 or C-19), 24.7 (+ve), 25.5 (+ve), 28.7 (+ve), 31.9 (-ve), 32.1 (+ve), 33.2 (+ve), 35.9 (+ve), 42.4 (+ve, C-9), 44.9 (-ve), 48.0 (-ve), 50.8 (-ve), 58.1 (-ve), 58.4 (+ve, C-6), 67.7 (-ve, C-10), 127.7 (-ve, C-12 or C-13), 131.2 (-ve, C-12 or C-13), 175.3 (+ve, C-15). 4. Experimental 291 Compound 417 exhibited 13 12 1H NMR (400 MHz): 5 = 0.78 (d, 3 H, J=6.5 Hz, Me-18 or Me-19), 0.90 (s, 3 H, Me-16), 0.96 (d, 3 H, J=6.4 Hz, Me-18 or Me-19), 1.17-1.31 (m, 1 H), 1.32-1.48 (m, 4 H), 1.48-1.59 (m, 2 H), 1.70-1.93 (m, 4 H), 2.03-2.13 (m, 2 H), 2.28-2.38 (ddd, 1 H, J= 14.6, 7.1, 7.1 Hz), 2.42-2.59 (m, 3 H), 3.46-3.53 (dd, 1 H, J = 7.0, 7.0 Hz, H-10), 3.63 (s, 3 H, Me-1'), 5.63-5.71 (m, 1 H, H-12 or H-13), 5.82-5.91 (m, 1 H, H-12 or H-13). 1 3C NMR (100 MHz): 5 = 22.2 (-ve, C-18 or C-19), 22.7 (-ve, C-18 or C-19), 25.5 (+ve), 26.3 (+ve), 28.7 (+ve), 29.4 (-ve), 31.9 (-ve), 32.5 (+ve), 35.8 (+ve), 38.7 (+ve), 41.1 (+ve, C-9), 43.1 (-ve), 49.4 (-ve), 50.5 (-ve), 57.6 (+ve, C-6), 58.4 (-ve), 78.7 (-ve, C-10), 127.7 (-ve, C-12 or C-13), 132.7 (-ve, C-12 or C-13), 175.4 (+ve, C-15). 4. Experimental Methyl (1 S*,2S*,5/?*,6S*,9/7*)-5-isopropyl-9-methyl-10-oxotricyclo[7.5.0.02'6] tetradec-12-ene-6-carboxylate (402) 292 416: = H, R 2 = OH 402 417: Ri = OH, R 2 = H A mixture of alcohols 416 and 417 was oxidized to ketone 402 using the method developed by Ley.88 Thus, to a stirred solution of alcohols 416 and 417 (50 mg, 0.16 mmol) in DCM (5 mL) at r.t. was added, sequentially, TPAP (5 mg, 0.015 mmol) in one portion and NMO (35 mg, 0.30 mmol) in one portion. The mixture was allowed to stir overnight. The reaction mixture was filtered through a pad of silica gel (1 g) and the silica gel was rinsed with Et20. The collected eluate was concentrated and the residual oil was purified by flash column chromatography (5 g of silica gel, 17:3 petroleum ether-Et20) to yield ketone 402 (45 mg, 90%) as a clear oil which exhibited. IR (film): v = 2951, 1720, 1698, 1365, 1155, 679 cm"1. 1H NMR (400 MHz): 8= 0.77 (d, 3 H, J= 6.5 Hz), 0.92-1.02 (ddd, 1 H, J= 14.3, 13.8, 4.1 Hz), 0.96 (d,'3H,J=6.4 Hz), 1.19-1.29 (m, 1 H), 1.27 (s, 3 H), 1.31-1.41 (m, 2 H), 1.43-1.58 (m, 3 H), 1.76-1.94 (m, 3 H), 2.04-2.11 (ddd, 1 H, J= 14.3, 3.5, 3.3 Hz), 2.12-2.22 (ddd, 1 H, J= 14.4, 8.2, 5.6 Hz), 2.41-2.49 (ddd, 1 H, J = 13.2, 4.1, 3.5 Hz), 2.62-2.71 (dd, 1 H, J= 12.7, 8.7 Hz), 2.73-2.82 (m, 1 H), 3.64 (s, 3 H), 3.67-3.74 (m, 1 H), 5.67-3.74 (m, 1H), 5.79-5.88 (m, 1H). 1 3C NMR (100 MHz): 8 = 22.2 (-ve), 22.6 (-ve), 25.4 (+ve), 25.8 (+ve), 26.0 (-ve), 28.5 (+ve), 31.8 (-ve), 34.1 (+ve), 35.6 (+ve), 40.8 (+ve), 46.8 (-ve), 50.1 (-ve), 50.7 (-ve), 53.8 (+ve), 57.8 (+ve), 58.0 (-ve), 123.8 (-ve), 131.4 (-ve), 175.2 (+ve), 208.4 (+ve). 4. Experimental 293 Table 4.20: NMR data for methyl (1 S*,2S*,5/?*,6S*,9rT)-5-isopropyl-9-methyl-10-oxotricyclo[7.5.0.02,6]tetradec-12-ene-6-carboxylate (402) 13 12 Carbon 13Q 1H HMBC No. 8 (ppm)a 8 (ppm) (mult; J (Hz))b'c'd Correlations6 1 46.8 H-1: part of the m at 1.74-1.98 H-2, H-8eq, H-14a, H-14b, H-19 2 50.1 H-2: part of the m at 1.43-1.58 H-3a, H-3b, H-7eq, H-14b 3 25.4 H-3a: part of the m at 1.43-1.58 H-3b: part of the m at 1.76-1.94 4 28.5 H-4a: part of the m at 1.31 -1.41 H-5, H-15 H-4b: part of the m at 1.76-1.94 5 58.0 H-5: part of the m at 1.31-1.41 H-16, H-17 6 57.8 H-7ax, H-8eq, H-15 7 34.1 H-7ax: part of the m at 1.43-1.58 H-7eq: 2.41-2.49 (ddd; 13.2, 4.1, 3.5) 8 35.6 H-8ax: 0.92-1.02 (ddd; 14.3, 13.8, 4.1) H-19 H-8eq: 2.04-2.11 (ddd; 14.3, 3.5, 3.3) 9 53.8 H-7eq, H-11a, H-14a, H-14b 10 208.4 H-1, H-8ax, H-11a, H-11b, H-19 11 40.8 H-11a: 2.62-2.71 (dd; 12.7, 8.7) H-12, H-13 H-11b: 3.67-3.74 (m) 12 123.8 H-12: 5.67-5.76 (m) H-11a, H-11b, H-14a, H-14b 13 131.4 H-13: 5.79-5.88 (m) H-1, H-11a, H-11b, H-14a, H-14b 14 25.8 H-14a: 2.12-2.22 (ddd; 14.4, 8.2, 5.6) H-12, H-13 H-14b: 2.73-2.82 (m) 15 31.8 H-15: 1.19-1.29 (m) H-16, H-17 16 22.6 H-16: 0.77 (d; 6.4) H-15, H-17 17 22.2 H-17: 0.96 (d; 6.5) H-15, H-16 . 18 175.2 H-5, H-7ax, H-1' 19 26.0 H-19: 1.27 (s) 1" 50.7 H-1': 3.64 (s) Recorded at 100 MHz. Recorded at 400 MHz. Assignments based on HMQC and JMOD data. Methylene protons are designated H-Xax and H-Xeq if they are known to occupy axial and equatorial positions, respectively, in the conformation depicted above. If no information regarding their position is available then they are arbitrarily designated H-Xa and H-Xb. Only those correlations which could be unambiguously assigned are recorded. 4. Experimental 294 Table 4.21: NMR data for methyl (1 S*,2S*,5R*,6S*,9f?*)-5-isopropyl-9-methyl-10-oxotricyclo[7.5.0.02,6]tetradec-12-ene-6-carboxylate (402) Proton No 1H 5 (ppm) (mult; J (Hz))a'b'c COSY Correlations01 H-1 part of the m at 1.74-1.98 H-14a H-2 part of the m at 1.43-1.58 H-3a part of the m at 1.43-1.58 H-3b part of the m at 1.76-1.94 H-4a part of the m at 1.31-1.41 H-4b part of the m at 1.76-1.94 H-5 part of the m at 1.31-1.41 H-7ax part of the m at 1.43-1.58 H-7eq, H-8ax, H-8eq H-7eq 2.41-2.49 (ddd 13.2, 4.1, 3.5) H-7ax, H-8ax, H-8eq H-8ax 0.92-1.02 (ddd 14.3, 13.8, 4.1) H-7ax, H7eq, H-8eq H-8eq 2.04-2.11 (ddd 14.3, 3.5, 3.3) H-7ax, H-7eq, H-8ax H-11a 2.62-2.71 (dd; 12.7, 8.7) H-11b, H-12 H-11b 3.67-3.74 (m) H-11a, H-12, H-13 H-12 5.67-5.76 (m) H-11a, H-11b, H-13 H-13 5.79-5.88 (m) H-11b, H-12, H-14a, H-14b H-14a 2.12-2.22 (ddd; 14.4, 8.2, 5.6) H-1, H-13, H-14b H-14b 2.73-2.82 (m) H-13, H-14a H-15 1.19-1.29 (m) H-16, H-17 H-16 0.77 (d; 6.4) H-15 H-17 0.96 (d; 6.5) H-15 H-19 1.27 (s) H-1' 3.64 (s) Recorded at 400 MHz. Assignments based on HMQC and JMOD data. Methylene protons are designated H-Xax and H-Xeq if they are known to occupy axial and equatorial positions, respectively, in the conformation depicted above. If a proton occupies an a relative position it is designated H-Xa, if it occupies a p relative position it is designated H-Xp. Only those correlations which could be unambiguously assigned are recorded. 295 5. References (1) Salzberg, H. W. From Caveman to Chemist: Circumstances and Achievements; American Chemical Society: Washington, DC, 1990. (2) Jaffe, B. Crucibles: The Story of Chemistry, Fawcett World Library: New York, 1957. (3) A comprehensive collection of speeches and lectures can be found at the Nobel Prize Foundation website: www.nobel.se. (4) Presentation speech, Professor H. G. Soderbaum, 1912 (see reference 3). (5) V. Grignard, Nobel Lecture, 1912 (see reference 3). (6) Presentation speech, Professor S. Gronowitz, 1990 (see reference 3). (7) Nobel Prize, 1950 (see reference 3). (8) Nobel Prize, 1979 (see reference 3). (9) Knowles, W. S. Angew. Chem., Int. Ed. Engl. 2002, 41, 1998. (10) Noyori, R. Angew. Chem., Int. Ed. Engl. 2002, 41, 2008. (11) Sharpless, K. B. Angew. Chem., Int. Ed. Engl. 2002, 41, 2024. (12) Lee, M. D.; Dunne, T. S.; Siegel, M. M.; Chang, C. C ; Morton, G. O.; Borders, D. B. J. Am. Chem. Soc. 1987, 109, 3464. (13) Nicolaou, K. C ; Hummel, C. W.; Pitsinos, E. N.; Nakada, M.; Smith, A. L; Shibayama, K.; Saimoto, H. J. Am. Chem. Soc. 1992, 114, 10082. (14) Ternansky, R. J.; Balogh, D. W.; Paquette, L. A. J. Am. Chem. Soc. 1982, 104, 4503. (15) Fessner, W.-D.; Murty, B. A. R. C ; Worth, J.; Hunkler, D.; Fritz, H.; Prinzbach, H.; Roth, W. D.; Schleyer, P. R.; McEwen, A. B.; Maier, W. F. Angew. Chem., Int. Ed. Engl. 1987, 26, 452. (16) Seebach, D. Angew. Chem., Int. Ed. Engl. 1990, 29, 1320. (17) Wender, P. A.; Miller, B. L. In Organic Synthesis: Theory and Applications; Hudlicky, T., Ed.; JAI Press: Greewich, CT, 1993; Vol. 2, p 27. (18) Trost, B. M. Science 1991, 254, 1472. (19) Hudlicky, T. Chem. Rev. 1996, 96, 3. (20) Stinson, S. Chem. Eng. News 1995, 73, 29. 5. References 296 (21) Dabrah, T. T.; Kaneko, T.; Massefski, W. J.; Whipple, E. B. J. Am. Chem. Soc. 1997, 119, 1594. (22) Chen, C ; Layton, M. E.; Shair, M. D. J. Am. Chem. Soc. 1998, 120, 10784. (23) Chen, C ; Layton, M. E.; Sheehan, S. M.; Shair, M. D. J. Am. Chem. Soc. 2000, 722,7424. (24) Seebach, D. Angew. Chem., Int. Ed. Engl. 1979, 18, 239. (25) Corey, E. J.; Cheng, X.-M. The Logic of Chemical Synthesis; John Wiley & Sons: New York, 1989. (26) Corey, E. J.; Nozoe, S. J. Am. Chem. Soc. 1963, 85, 3528. (27) Piers, E.; Renaud, J.; Rettig, S. J. Synthesis 1998, 590. (28) Piers, E.; Walker, S. D.; Hong-Ping, Y. unpublished results. (29) Piers, E.; McEachern, E. J. Syn/eff 1996, 1087. (30) Wright, A. D.; Konig, G. M.; Angerhofer, C. K.; Greenidge, P.; Linden, A.; Desqueyroux-Faundez, R. J. Nat. Prod. 1996, 59, 710. (31) Piers, E.; Chong, J. M. Can. J. Chem. 1988, 66, 1425. (32) Piers, E.; Karunuratne, V. Tetrahedron 1989, 45, 1089. (33) Loyola, L. A.; Morales, G.; Rodriguez, B.; Jimenez-Barbero, J.; de la Torre, M. C ; Perales, A.; Torres, M. R. Tetrahedron 1990, 46, 5413. (34) Piers, E.; Walker, S. D.; Armbrust, R. J. Chem. Soc, Perkin Trans. 1 2000, 635. (35) Previous literature reports have used different structural numbering systems for kelsoene. In this thesis the kelsoane skeleton is numbered according to IUPAC protocol for tricyclic ring systems. (36) Arnone, A.; Nasini, G.; Vajna de Pava, O. J. Chem. Soc, Perkin Trans. 1 1993, 2723. (37) Konig, G. M.; Wright, A. D. J. Org. Chem. 1997, 62, 3837-3840. (38) Schulz, S.; Messer, C ; Dettner, K. Tetrahedron Lett. 1997, 38, 2077. (39) Nabeta, K.; Yamamoto, K.; Hashimoto, M.; Koshino, H.; Funatsuki, K.; Katoh, K. J. Chem. Soc, Chem. Commun. 1998, 1485. (40) Hashimoto, T.; Ikeda, H.; Takaoka, S.; Tanaka, M.; Asakawa, Y. Phytochemistry 1999, 52, 501. (41) Konig, W. A.; Warmers, U. Phytochemistry 1999, 52, 1519. (42) Warmers, U.; Wihstutz, K.; Bulow, N.; Fricke, C ; Konig, W. A. Phytochemistry 1998, 49, 1723. 5. References 297 Nabeta, K.; Yamamoto, M.; Fukushima, K.; Katoh, K. J. Chem. Soc, Perkin Trans. 1 2000, 2703. Nabeta, K.; Yamamoto, M.; Koshino, H.; Fukui, H.; Fukushi, Y.; Tahara, S. Biosci., Biotechnol., Biochem. 1999, 63, 1772. Mehta, G.; Srinivas, K. Syn/eff 1999, 5, 555-556. Mehta, G.; Srinivas, K. Tetrahedron Lett. 1999, 40, 4877-4880. Mehta, G.; Srinivas, K. Tetrahedron Lett. 2001, 42, 2855. Fietz-Razavian, S.; Schulz, S.; Dix, I.; Jones, P. J. Chem. Soc, Chem. Commun. 2001, 2154. Bach, T.; Spiegel, A. Synlett2002, 1305. Zhang, L; Koreeda, M. Org. Lett. 2002, 4, 3755. Fukui, H.; Fukushi, Y.; Tahara, S. Tetrahedron Lett. 1999, 40, 325. Henry, P. M.; Davies, M.; Ferguson, G.; Philips, S.; Restivo, R. J. Chem. Soc, Chem. Commun. 1974, 112. Ito, Y.; Hirao, T.; Saegusa, T. J. Org. Chem. 1978, 43, 1011. Petasis, N. A.; Bzowej, E. I. J. Am. Chem. Soc. 1990, 112, 6392. Tietze, L. F.; Beifuss, U.; Ruther, M.; Ruhlmann, A.; Antel, J.; Sheldrick, G. M. Angew. Chem., Int. Ed. Engl. 1988, 27, 1186. Bach, T.; Spiegel, A. Eur. J. Org. Chem. 2002, 645. Corey, E. J.; Guzman-Perez, A.; Loh, T. P. J. Am. Chem. Soc. 1994, 116, 3611. Cargill, R. L; Dalton, J. R.; Morton, G. H.; Caldwell, W. E. Org. Synth. 1984, 1984, 118. Meyers, A. I.; Fleming, S. A. J. Am. Chem. Soc. 1986, 108, 306. Ramaiah, M. Synthesis 1984, 529. Hudlicky, T.; Price, J. D. Chem. Rev. 1989, 89, 1467. Mehta, G.; Srikrishna, A. Chem. Rev. 1997, 671. Franck-Neumann, M.; Kastler, A. Syn/eff 1995, 61. Lipshutz, B. H.; Sengupta, S. Organic Reactions 1992, 41, 135. Suggs, J. W.; Cox, S. D.; Crabtree, R. H.; Quirk, J. M. Tetrahedron Lett. 1981, 22, 303. Young, J. F.; Osborn, J. A.; Jardine, F. H.; Wilkinson, G. Chem. Commun 1965, 131. Reich, H. J.; Renga, J. M.; Reich, I. L. J. Am. Chem. Soc. 1975, 97, 5434. 5. References 298 (68) Reich, H. J.; Wollowitz, S. Organic Reactions 1993, 44, 1. (69) Dauben, W. G.; Michno, D. K. J. Org. Chem. 1977, 42, 682. (70) Corey, E. J.; Suggs, J. W. Tetrahedron Lett. 1975, 2467. (71) Piancatelli, G.; Scettri, A.; DAuria Synthesis 1982, 245. (72) Cheng, Y.-S.; Liu, W.-L; Chen, S. Synthesis 1980, 223. (73) Piers, E.; Cook, K. L; Rogers, C. Tetrahedron Lett. 1994, 35, 8573. (74) Piers, E.; Boulet, S. L. Syn/erf 1998, 516. (75) Ohfune, Y.; Shirahama, H.; Matsumoto, T. Tetrahedron Lett. 1975, 49, 4377. (76) Crimmins, M. T.; Reinhold, T. L. Organic Reactions 1993, 44, 297. (77) Majetich, G.; Defauw, J. Tetrahedron 1988, 44, 3833. (78) Taguchi, H.; Tanaka, S.; Yamamoto, H.; Nozaki, H. Tetrahedron Lett. 1973, 27, 2465. (79) Pine, S. H. Organic reactions 1993, 43, 1. (80) Pine, S. H.; Shen, G. S.; Hoang, H. Synthesis 1991, 165. (81) Tebbe, F. N.; Parshall, G. W.; Reddy, G. S. J. Am. Chem. Soc. 1978, 100, 3611. (82) Takai, K.; Hotta, Y.; Oshima, K.; Nozaki, H. Tetrahedron Lett. 1978, 2417. (83) Hibino, J.; Okazoe, T.; Takai, K.; Nozaki, H. Tetrahedron Lett. 1985, 26, 5579. (84) Okazoe, T.; Hibino, J.; Takai, K.; Nozaki, H. Tetrahedron Lett. 1985, 26, 5581. (85) Lombardo, L. Tetrahedron Lett. 1982, 23, 4293. (86) Lombardo, L. Org. Synth. 1987, 65, 81. (87) The addition of a Pb(ll) source has been shown to dramatically increase the rate of olefination in the Takai and Lombardo reactions. Takai, K.; Kakiuchi, T.; Kataoka, Y.; Utimoto, K. J. Org. Chem. 1994, 59, 2668. (88) Ley, S. V.; Norman, J.; Griffith, W. P.; Marsden, S. P. Synthesis 1994, 639. (89) Piers, E.; Orellana, A. Synthesis 2001, 2138. (90) Thompson, H. W.; Weedon, B. C. L; Cullis, C. F.; Gujral, P. D., Eds. Nomenclature of Organic Chemistry, Butterworth & Co.: London, 1971. (91) Panico, R.; Powell, W. H.; Richer, J. C. A Guide to IUPAC Nomenclature of Organic Compounds-Recommendations 1993; Blackwell Scientific Publications: Oxford, 1993. (92) Loyola, L. A.; Morales, G.; de la Torre, M. C ; Pedreros, S.; Rodriguez, B. Phytochemistry\990, 29, 3950. 5. References 299 (93) Loyola, L. A.; Morales, G.; Rodriguez, B.; Jimenez-Barbero, J.; Pedreros, S.; de la Torre, M. C.; Perales, A. J. Nat. Prod. 1991, 54, 1404. (94) Loyola, L. A.; Borquez, J.; Morales, G.; San Martin, A. Phytochemistry1996, 43, 165. (95) Nicoletti, M.; Di Fabio, A.; D'Andrea, A.; Salvatore, G.; van Baren, C.; Coussio, J. D. Phytochemistry 1996, 43, 1065. (96) Loyola, L. A.; Borquez, J.; Morales, G.; San Martin, A. Phytochemistry1997', 44, 649. (97) Loyola, L. A.; Borquez, J.; Morales, G.; San Martin, A. Phytochemistry 1998, 49, 1091. (98) Wachter, G. A.; Matooq, G.; Hoffmann, J. J.; Maiese, W. M.; Singh, M. P.; Montenegro, G.; Timmermann, B. N. J. Nat. Prod. 1999, 62, 1319. (99) Loyola, L. A.; Borquez, J.; Morales, G.; San Martin, A. Phytochemistry 2000, 53, 961. (100) Loyola, L. A.; Borquez, J.; Morales, G.; San Martin, A.; Manriquez, V.; Wittke, O. Tetrahedron 1998, 54, 15533. (101) Wachter, G. A.; Franzblau, S. G.; Montenegro, G.; Suarez, E.; Fortunato, R. H.; Saavedra, E.; Timmermann, B. N. J. Nat. Prod. 1998, 61, 965. (102) Dewick, P. M. Medicinal Natural Products: A Biosynthetic Approach; 2nd ed.; John Wiley & Sons Ltd.: Chichester, 2002. (103) Boyd, J. D.; Foote, C. S.; Imagawa, D. K. J. Am. Chem. Soc. 1980, 102, 3641. (104) Sutbeyaz, Y.; Secen, H.; Balci, M. J. Org. Chem. 1988, 53, 2312. (105) Cahiez, G.; Bernard, D.; Normant, J. F. Synthesis 1976, 245. (106) Hoveyda, A. H.; Evans, D. A.; Fu, G. C. Chem. Rev. 1993, 93, 1307. (107) Brown, J. M. Angew. Chem., Int. Ed. Engl. 1987, 26, 190. (108) Sharpless, K. B.; Lauer, R. F. J. Am. Chem. Soc. 1973, 95, 2697. (109) Nicolaou, K. C ; Magolda, R. L.; Sipio, W. J.; Barnette, W. E.; Lysenko, Z.; Jouliie, M. M. J. Am. Chem. Soc. 1980, 102, 3784. (110) Kuhnert, S. M.; Maier, M. E. Org. Lett. 2002, 4, 643. (111) Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic Compounds; John Wiley & Sons, Inc: New York, 1994. (112) Lo Cicero, B.; Weisbuch, F.; Dana, G. J. Org. Chem. 1981, 46, 914. (113) Crabtree, S. R.; Mander, L. N.; Sethi, P. Org. Synth. 1991, 70, 256. 5. References 300 (114) Taber, D. F.; Petty, E. H. J. Org. Chem. 1982, 47, 4808. (115) Kotsuki, H.; Arimura, K.; Ohishi, T.; Maruzasa, R. J. Org. Chem. 1999, 64, 3770. (116) Christoffers, J. Chem. Commun 1997, 943. (117) Vettel, P. R.; Coates, R. M. J. Org. Chem. 1980, 45, 5430. (118) Scanio, C. J. V.; Hill, L. P. Synthesis 1970, 651. (119) Johnson, J. L; Herr, M. E.; Babcock, J. C ; Fonken, A. E.; Stafford, J. E.; Heyl, F. W. J. Am. Chem. Soc. 1956, 78, 430. (120) Caine, D. In Comprehensive Organic Synthesis; First ed.; Trost, B. M., Ed.; Pergamon Press: New York, 1991; Vol. 3, pp 1-63. (121) Walker, S. D. In Ph. D. Thesis; University of British Columbia: Vancouver, 2002. (122) Challand, B. D.; Hikino, H.; Kornis, G.; Lange, G.; de Mayo, P. J. Org. Chem. 1969, 34, 794. (123) Baldwin, S. W.; Wilkinson, J. M. J. Am. Chem. Soc. 1980, 102, 3634. (124) Bader, A. R.; Carroll, M. F. J. Am. Chem. Soc. 1953, 75, 5400. (125) Thompson, H. W. J. Org. Chem. 1971, 36, 2577. (126) Luche, J.-L. J. Am. Chem. Soc. 1978, 100, 2226. (127) Gemal, A. L; Luche, J.-L. J. Am. Chem. Soc. 1981, 103, 5454. (128) Hughes, D. L. Organic Reactions 1992, 42, 335. (129) Corey, E. J.; Engler, T. A. Tetrahedron Lett. 1984, 25, 149. (130) Stork, G.; Kahne, D. H. J. Am. Chem. Soc. 1983, 105, 1072. (131) Evans, D. A.; Morrisey, M. M. J. Am. Chem. Soc. 1984, 106, 3866. (132) Brown, J. M.; Naik, R. G. J. Chem. Soc, Chem. Commun. 1982, 348. (133) Myers, A. G.; Zheng, B. Tetrahedron Lett. 1996, 37, 4841. (134) This protocol can also be used to prepare allenes from propargilic alcohols. Myers, A. G.; Zheng, B. J. Am. Chem. Soc 1996, 118, 4492. (135) A different protocol for this transformation has been reported by Tsuji. Mandai, T.; Matsumoto, T.; Kawada, M.; Tsuji, J. Tetrahedron 1993, 49, 5483. (136) Myers, A. G.; Zheng, B.; Movassaghi, M. J. Org. Chem. 1997, 62, 7507. (137) NBSH can also be used to effect the deoxygenation of unhindered alcohols Myers, A. G.; Movassaghi, M.; Zheng, B. J. Am. Chem. Soc. 1997, 119, 8572. (138) Barrero, A. F.; Oltra, J. E.; Alvarez, M.; Rosales, A. J. Org. Chem. 2002, 67, 5461. (139) McKenzie, T. C. J. Org. Chem. 1974, 39, 629. 5. References 301 (140) Maruoka, K.; Ooi, T.; Nagahara, S.; Yamamoto, H. Tetrahedron 1991, 47, 6983. (141) Jung, M. E.; D'Amico, D. C. J. Am. Chem. Soc. 1995, 117, 7379. (142) Heslin, J. C ; Moody, C. J. J. Chem. Soc, Perkin Trans. 1 1988, 1417. (143) Overman, L. E.; Tomasi, A. L. J. Am. Chem. Soc. 1998, 120, 4039. (144) Molander, G. A.; Quirmbach, M. S.; Silva, L. F.; Spencer, K. C ; Balsells, J. Org. Lett. 2001, 3, 2257. (145) Miaskiewicz, K.; Smith, D. A. J. Am. Chem. Soc. 1998, 120, 1872. (146) Bach, R. D.; Dmitrenko, O.; Waldemar, A.; Schambony, S. J. Am. Chem. Soc. 2003, 125, 924. (147) Johnson, C. R.; Tait, B. D.; Cieplak, A. S. J. Am. Chem. Soc. 1987, 109, 5875. (148) Grieco, P. A.; Noguez, J. A.; Masaki, Y. J. Org. Chem. 1977, 42, 495. (149) Grieco, P. A.; Nishizawa, M.; Oguri, T.; Burke, S. D.; Marinovic, N. J. Am. Chem. Soc 1977, 99, 5773. (150) Jung, M. E.; Lyster, M. A. J. Org. Chem. 1977, 42, 3761. (151) Harada, K.; Munegumi, T. In Comprehensive Organic Synthesis; First ed.; Trost, B. M., Ed.; Pergamon Press: New York, 1991; Vol. 8, pp 139-158. (152) McCombie, S. W. In Comprehensive Organic Synthesis; Trost, B. M., Ed.; Pergamon Press: New York, 1991; Vol. 8, pp 811-833. (153) Corey, E. J.; Chaykovski, M. J. Am. Chem. Soc. 1962, 84, 866. (154) Corey, E. J.; Chaykovski, M. J. Am. Chem. Soc. 1962, 84, 3782. (155) Corey, E. J.; Chaykovski, M. J. Am. Chem. Soc. 1965, 87, 1353. (156) Danishefsky, S. J.; Masters, J. J.; Young, W. B.; Link, J. T.; Snyder, L. B.; Magee, T. V.; Jung, D. K.; Isaacs, R. C. A.; Bornmann, W. G.; Alaimo, C. A.; Coburn, C. A.; Di Grandi, M. J. J. Am. Chem. Soc. 1996, 118, 2843. (157) Ireland, R. E.; Mander, L. N. J. Org. Chem. 1967, 32, 689. (158) Malhotra, S. K.; Johnson, F. J. Am. Chem. Soc. 1965, 87, 5493. (159) Snider, B. B.; Vo, N. H.; O'neil, S. V.; Foxman, B. M. J. Am. Chem. Soc. 1996, 118, 7644. (160) DePuy, C. H.; Breitbeil, F. W. J. Am. Chem. Soc. 1963, 85, 2176. (161) DePuy, C. H.; Gibson, D. H. Chem. Rev. 1974, 74, 605. (162) Wenkert, E.; Berges, D. A. J. Am. Chem. Soc. 1967, 89, 2508. (163) Wenkert, E.; Mueller, R. A.; Reardon, E. J.; Sathe, S. S.; Scharf, D. J.; Tosi, G. J. Am. Chem. Soc. 1970, 92, 7428. 5. References 302 (164) Reusch, W.; Grimm, K.; Karoglan, J. E.; Martin, J.; Subrahamanian, K. P.; Venkataramani, P. S.; Yordy, J. D. J. Am. Chem. Soc. 1977, 1958. (165) Reusch, W.; Yordy, J. D. J. Am. Chem. Soc. 1977, 99, 1965. (166) Simmons, H. E.; Smith, R. D. J. Am. Chem. Soc. 1958, 80, 5323. (167) Charette, A. B.; Beauchemin, A. Organic reactions 2001, 58, 1. (168) Bertinato, P.; Sorensen, E. J.; Meng, D.; Danishefsky, S. J. J. Org. Chem. 1996, 61, 8000. (169) Grieco, P. A.; Gilman, S.; Nishizawa, M. J. Org. Chem. 1976, 41, 1483. (170) Sharpless, K. B.; Young, M. W. J. Org. Chem. 1975, 40, 947. (171) Bieber, L. W.; de Sa, A. C. P. F.; Menezes, P. H.; Goncalves, S. M. C. Tetrahedron Lett. 2001, 42, 4597. (172) Reich, H. J.; Wollowitz, S.; Trend, J. E.; Chow, F.; Wendelborn, D. F. J. Org. Chem. 1978, 43, 1697. (173) Armstrong, S. J. Chem. Soc, Perkin Trans. 11998, 371. (174) Grubbs, R. H.; Chang, S. Tetrahedron 1998, 54, 4413. (175) Furstner, A. Angew. Chem., Int. Ed. Engl. 2000, 39, 3012. (176) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18. (177) Wilhelm, T. E.; Belderrain, T. R.; Brown, S. N.; Grubbs, R. H. Organometallics 1997, 16, 3867. (178) Dias, E. L; Nguyen, S. T.; Grubbs, R. H. J. Am. Chem. Soc 1997, 119, 3887. (179) Horst Friebolin's book is an excellent introduction to NMR techniques. Friebolin, H. Basic One- and Two-Dimensional NMR Spectroscopy, 2nd ed.; VCH Publishers: New York, 1993. (180) The signals due to trace amounts of common solvents and impurities could be easily identified. Gottlieb, H. E.; Kotlyar, V.; Nudelman, A. J. Org. Chem. 1997, 62, 7512. (181) Hoye, T. R.; Hanson, P. R.; Vyvyan, J. R. J. Org. Chem. 1994, 59, 4096. (182) Hoye, T. R.; Zhao, H. J. Org. Chem. 2002, 67, 4014. (183) Still, W. C ; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923. (184) Bryan, W. P.; Byrne, R. H. J. Chem. Educ 1970, 47, 2923. (185) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory Chemicals; Pergamon Press: New York, 1988. (186) Adam, W.; Bialas, J.; Hadjiarapoglou, L. Chem. Ber. 1991, 124, 2377. 5. References 303 (187) Engler, E. M.; Andose, J. D.; Schleyer, P. V. R. J. Am. Chem. Soc. 1973, 95, 8005. (188) Liebman, J. F.; Greenberg, A. Chem. Rev. 1976, 76, 311-365. (189) Schuster, D. I.; Lem, G.; Kaprinidis, N. A. Chem. Rev. 1993, 93, 3. (190) Winkler, J. D.; Bowen, C. M.; Liotta, F. Chem. Rev. 1995, 95, 2003-2020. (191) Winkler, J. D.; Rouse, M. B.; Greaney, M. F.; Harrison, S. J.; Jeon, Y. T. J. Am. Chem. Soc. 2002, 124, 9726. (192) Winkler, J. D.; Doherty, E. J. Am. Chem. Soc. 1999, 121, 7425. 

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.831.1-0061225/manifest

Comment

Related Items