UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Biologically active secondary metabolites isolated from marine and terrestrial sources Cinel, Bruno 2001

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

Item Metadata

Download

Media
831-ubc_2001-714497.pdf [ 9.35MB ]
Metadata
JSON: 831-1.0061267.json
JSON-LD: 831-1.0061267-ld.json
RDF/XML (Pretty): 831-1.0061267-rdf.xml
RDF/JSON: 831-1.0061267-rdf.json
Turtle: 831-1.0061267-turtle.txt
N-Triples: 831-1.0061267-rdf-ntriples.txt
Original Record: 831-1.0061267-source.json
Full Text
831-1.0061267-fulltext.txt
Citation
831-1.0061267.ris

Full Text

BIOLOGICALLY ACTIVE SECONDARY METABOLITES ISOLATED FROM MARINE AND TERRESTRIAL SOURCES by B R U N O C I N E L B . S c , Simon Fraser University, 1994 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in T H E F A C U L T Y O F G R A D U A T E S T U D I E S 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 October 2001 © Bruno Cinel, 2001 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada Date OCT T /I / 2 . 0 ^ I DE-6 (2/88) ABSTRACT 11 A series of new and known secondary metabolites were isolated from marine and terrestrial sources guided by two newly developed, cell-based assays. Investigations into the antimitotic properties of a crude extract from the Caribbean octocoral Erythropodium caribaeorum resulted in the isolation and identification of the known antimitotic agent eleutherobin (61) and six novel structural analogues, 69-74. E. caribaeorum proved to be a new and abundant source of eleutherobin (61), whose pre-clinical development had been impeded by its scarcity, and the structural variations of the new diterpenoids offered key insights into proposed pharmacophore models for microtubule-stabilizing compounds. In addition, single crystal X-ray diffraction analysis and N O E difference experiments provided the first solid-state and solution conformations for eleutherobin (61) and may aid in the development of new models for microtubule stabilization. 61 R 1 = A c ; R 2 = H ; R 3 = M e ; A 2 ' ' 3 ' ( £ ) 69 Rj = A c ; R 2 = H ; R 3 = H ; A 2 , 3 (E) 70 R, = H ; R 2 = H ; R 3 = Me ; A 2 , 3 (E) 71 Rj = H ; R 2 = A c ; R 3 = M e ; A 2 ' 3 ' (E) ,2',3' OH 73 R = W ^ O H I OAc 74 R = Ac 72 R{ = A c ; R 2 = H ; R 3 = M e ; A ^ (Z) The crude extract from a marine sponge, Stylissa flabelliformis, exhibited potent activity in a new bioassay for G2 cell cycle checkpoint inhibitors. Bioassay-guided fractionation of this active extract resulted in the isolation and identification of the natural product i i i debromohymenialdisine (91) and three related alkaloids (85, 94, 103). These compounds were the first G 2 checkpoint inhibitors to be found by a rational screen and were structurally distinct from previously reported G 2 checkpoint inhibitors. In addition, debromohymenialdisine (91) was found to specifically act on the protein kinases C h k l and Chk2, thus providing a new biochemical tool to probe the molecular basis of G2 checkpoint inhibition. 91 85 94 103 As a result of a large-scale screen of natural extracts for G 2 checkpoint inhibitors, a series of a-pyrones from the Taiwanese tree Cryptocarya concinna were found to exhibit potent inhibitory activity. Bioassay-guided purification resulted in the isolation and identification of the natural product cryptofolione (130) and three related polyketide lactones (148, 131, 138). Synthetic modifications on these metabolites yielded four additional analogues and led to mode of action and structure-activity relationship studies. The G2 checkpoint inhibitors isolated from C. concinna were structurally different from previously known inhibitors and appear to act by a novel mechanism of action. 131 138 iv TABLE OF CONTENTS Abstract i i Table of Contents iv List of Tables v i i i List of Figures ix List of Schemes xvi List of Abbreviations xvii Acknowledgements xxi CHAPTER 1: General Introduction 1.1. Natural Products Chemistry 1 1.2. Marine Natural Products Chemistry 4 1.3. Novel Anticancer Screens 11 1.3.1. The Cell Cycle 11 1.3.2. Antimitotic Agents 14 1.3.3. The Antimitotic Bioassay 16 1.3.4. Cell Cycle Checkpoints 19 1.3.5. The G2 Checkpoint Inhibition Bioassay 21 1.3.6. Biochemical Tools and Chemical Genetics 24 1.4. Research Summary 27 References 29 CHAPTER 2: Antimitotic Diterpenoids from the Gorgonian Erythropodium caribaeorum 2.1. Introduction 36 2.1.1. Introduction to Coelenterates and Gorgonians 36 2.1.2. Review of Diterpenoids from Octocorals 38 V 2.1.3. Review of Known Metabolites from Erythr op odium caribaeorum 43 2.1.4. Review of Microtubule-Stabilizing Antimitotic Agents 46 2.2. Results and Discussion 50 2.2.1. Isolation of Antimitotic Compounds 50 2.2.2. Structure Elucidation of Antimitotic Diterpenoids 54 2.2.2.1. Desmethyleleutherobin 55 2.2.2.2. Eleutherobin 67 2.2.2.3. Desacetyleleutherobin 7 2 2.2.2.4. Isoeleutherobin A 77 2.2.2.5. Z-Eleutherobin 81 2.2.2.6. Caribaeoside 85 2.2.2.7. Caribaeolin 91 2.2.2.8. Diacetyleleutherobin 96 2.2.2.9. Known Compounds 102 2.2.3. Biological Activity 104 2.2.4. Solid-State and Solution Conformations 110 2.2.5. Additional Results 123 2.2.6. Future Studies 127 2.3. Conclusions 128 References 129 CHAPTER 3: G2 Cell Cycle Checkpoint Inhibitors from Stylissa flabelliformis 3.1. Introduction 135 3.1.1. Introduction to Marine Sponges 135 3.1.2. Review of Oroidin Alkaloids from Marine Sponges 138 vi 3.1.3. Review of G2 Checkpoint Inhibitors 143 3.2. Results and Discussion 145 3.2.1. Isolation of G2 Checkpoint Inhibitors 145 3.2.2. Structure Elucidation of G2 Checkpoint Inhibitors 148 3.2.2.1. Debromohymenialdisine 148 3.2.2.2. Hymenialdisine 152 3.2.2.3. Z-Debromoaxinohydantoin 155 3.2.2.4. Aldisin 158 3.2.3. Biological Activity 161 3.3. Conclusions 165 References 166 CHAPTER 4: G2 Cell Cycle Checkpoint Inhibitors from Cryptocarya concinna 4.1. Introduction 170 4.1.1. Review of Plant-Derived Compounds 170 4.1.2. Review of Cryptocarya Phytochemistry 173 4.2. Results and Discussion 177 4.2.1. Isolation of G2 Checkpoint Inhibitors 177 4.2.2. Structure Elucidation of G2 Checkpoint Inhibitors 179 4.2.2.1. -^Cryptofolione 180 4.2.2.2. Z-Cryptofolione 185 4.2.2.3. Cryptofolione ketone 189 4.2.2.4. Compound 138 193 4.2.3. Synthetic Derivatives of ZJ-Cryptofolione 197 4.2.3.1. Cryptofolione acetonide 197 vii 4.2.3.2. Diacetylcryptofolione ' 201 4.2.3.3. p-mercaptoethanol adduct 205 4.2.3.4. Compound 152 209 4.2.4. Biological Activity 213 4.3. Conclusions 215 References 216 EXPERIMENTAL 218 APPENDIX 225 vii i LIST OF TABLES Page Table 2.1. N M R data for desmethyleleutherobin (69) recorded in D M S O - J 6 . 61 Table 2.2. N M R data for eleutherobin (61) recorded in DMSO-c/6. 69 Table 2.3. N M R data for desacetyleleutherobin (70) recorded in DMSO-c/6. 74 Table 2.4. N M R data for isoeleutherobin A (71) recorded in D M S O - J 6 . 79 Table 2.5. N M R data for Z-eleutherobin (72) recorded in DMSO-t /6 . 83 Table 2.6. N M R data for caribaeoside (73) recorded in D M S O - J 6 . ' 88 Table 2.7. N M R data for caribaeolin (74) recorded in D M S O - J 6 . 93 Table 2.8. N M R data for diacetyleleutherobin (75) recorded in DMSO-c/6. 98 Table 2.9. Summary of antimitotic activity of the compounds isolated from E. caribaeorum. 107 Table 2.10. Difference N O E and R O E S Y data for eleutherobin (61) in C D C 1 3 andDMSO-e?6. 112 Table 3.1. G 2 checkpoint inhibition by debromohymenialdisine (91) and related compounds. 162 Table 4.1. Summary of a-pyrones isolated from various Cryptocarya species. 175 Table 4.2. N M R data for £-cryptofolione (130) recorded in C D C 1 3 . 181 Table 4.3. N M R data for Z-cryptofolione (148) recorded in C D C 1 3 . 186 Table 4.4. N M R data for cryptofolione ketone (131) recorded in C D C 1 3 . 190 Table 4.5. N M R data for compound 138 recorded in C D C 1 3 . 194 Table 4.6. N M R data for cryptofolione acetonide (149) recorded in C D C 1 3 . 198 Table 4.7. N M R data for diacetylcryptofolione (150) recorded in C D C 1 3 . 202 Table 4.8. N M R data for P-mercaptoethanol adduct (151) recorded in C D C 1 3 . 206 Table 4.9. N M R data for compound 152 recorded in C D C 1 3 . 210 ix LIST OF FIGURES Page Figure 1.1. Distribution of marine natural products reported in 1999 according to the phylum of the source organism. 10 Figure 1.2. A n overview of the eukaryotic cell cycle. 12 Figure 1.3. Fluorescent micrographs depicting cells in various stages of mitosis. 13 Figure 1.4. Microtubule structure. 15 Figure 1.5. The antimitotic bioassay. 18 Figure 1.6. The series of signal transduction pathways in mammalian cell cycle checkpoints. 19 Figure 1.7. Rationale for the use of G2 checkpoint inhibitors in cancer therapy. 22 Figure 1.8. The G2 cell cycle checkpoint inhibitor bioassay. 24 Figure 1.9. The relationship between genetics and chemical genetics in exploring protein function. 26 Figure 1.10. Small molecule-based studies of the cell cycle and cell cycle checkpoints. 27 Figure 2.1. Structural classes of gorgonian di terpenoids. 41 Figure 2.2. Erythropodium caribaeorum (photo by P. Humann). 43 Figure 2.3. Antimitotic activity (reported as Absorbance4o5nm) of each fraction collected from liquid chromatography. 53 Figure 2.4. Selected H M B C and C O S Y correlations for substructure A of compound 69. 56 Figure 2.5. Selected H M B C and C O S Y correlations for substructure B of compound 69. 57 Figure 2.6. Selected H M B C and C O S Y correlations for substructure C of compound 69. 59 Figure 2.7. Selected H M B C and C O S Y correlations for substructure D of compound 69. 60 Figure 2.8. ' H N M R spectrum of desmethyleleutherobin (69) recorded in DMSO-c/6 at 500 M H z . 62 Figure 2.9. I 3 C N M R spectrum of desmethyleleutherobin (69) recorded in DMSO-J6 at 100 M H z . ' • 63 X Figure 2.10. C O S Y spectrum of desmethyleleutherobin (69) recorded in D M S O - J 6 at 500 M H z . 64 Figure 2.11. H M Q C spectrum of desmethyleleutherobin (69) recorded in D M S O - J 6 at 500 M H z . 65 Figure 2.12. H M B C spectrum of desmethyleleutherobin (69) recorded in D M S O - J 6 at 500 M H z . 66 Figure 2.13. Selected H M B C and C O S Y correlations for the cyclic ketal subfragment of compound 61. 68 Figure 2.14. ' H N M R spectrum of eleutherobin (61) recorded in D M S O - J 6 at 500 M H z . 70 Figure 2.15. l 3 C N M R spectrum of eleutherobin (61) recorded in D M S O - J 6 at 100 M H z . 71 Figure 2.16. Selected C O S Y correlations for the arabinose subunit of compound 70. 73 Figure 2.17. ' H N M R spectrum of desacetyleleutherobin (70) recorded in DMSO-<i6 75 at 500 M H z . Figure 2.18. 1 3 C N M R spectrum of desacetyleleutherobin (70) recorded in D M S O - J 6 76 at 100 M H z . Figure 2.19. Selected C O S Y correlations for the arabinose subunit of compound 71. 78 Figure 2.20. ' H N M R spectrum of isoeleutherobin A (71) recorded in D M S O - J 6 at 500 M H z . 80 Figure 2.21. Selected H M B C and C O S Y correlations for the yV-methylurocanic acid ester subunit of compound 72. 82 Figure 2.22. ' H N M R spectrum of Z-eleutherobin (72) recorded in D M S O - d 6 at 500 M H z . 84 Figure 2.23. Selected H M B C and C O S Y correlations for the cyclohexene ring subunit of compound 73. 86 Figure 2.24. Selected R O E S Y correlations for the diterpene core of compound 73. 87 Figure 2.25. ' H N M R spectrum of caribaeoside (73) recorded in DMSO-c/6 at 500 M H z . 89 Figure 2.26. Expanded R O E S Y spectrum of caribaeoside (73) recorded in D M S O - J 6 at 500 M H z . 90 Figure 2.27. Selected H M B C correlations around C-15 of compound 74. 92 Figure 2.28. Selected R O E S Y correlations for the diterpene core of compound 74. 92 xi Figure 2.29. ' H N M R spectrum of caribaeolin (74) recorded in D M S O - J 6 at 500 M H z . 94 Figure 2.30. Expanded R O E S Y spectrum of caribaeolin (74) recorded in D M S O - J 6 at 500 M H z . 95 Figure 2.31. Selected C O S Y and H M B C correlations for the arabinose subunit of compound 75. 97 Figure 2.32. ' H N M R spectrum of diacetyleleutherobin (75) from the acetylation of eleutherobin (61). Recorded in DMSO-( i6 at 500 M H z . 99 Figure 2.33. ' H N M R spectrum of diacetyleleutherobin (75) from the acetylation of desacetyleleutherobin (70). Recorded in D M S O - J 6 at 500 M H z . 100 Figure 2.34. *H N M R spectrum of diacetyleleutherobin (75) from the acetylation of isoeleutherobin A (71). Recorded in D M S O - J 6 at 500 M H z , 101 Figure 2.35. Immunofluorescence photograph of A549 cells after exposure to eleutherobin (61) at a concentration of 60 ng/ml. 105 Figure 2.36. Antimitotic activity of the compounds isolated from E. caribaeorum. 106 Figure 2.37. Overlay of nonataxel (cyan) with a) paclitaxel b) epothilone B c) eleutherobin and d) discodermolide, all in yellow. 109 Figure 2.38. Areas of common overlap among microtubule-stabilizing compounds. 109 Figure 2.39. O R T E P drawing of eleutherobin (61). 111 Figure 2.40. Torsional angles around the cyclohexene ring of eleutherobin (61). 113 Figure 2.41. N O E difference experiments of eleutherobin (61) recorded in C D C 1 3 at 400 M H z . 114 Figure 2.42. N O E difference experiments of eleutherobin (61) recorded in C D C 1 3 at 400 M H z . 115 Figure 2.43. R O E S Y spectrum of eleutherobin (61) recorded in C D C 1 3 at 500 M H z . 116 Figure 2.44. R O E S Y spectrum of eleutherobin (61) recorded in DMSO-rf6 at 500 M H z . 117 Figure 2.45. Comparison of the conformations of eleutherobin (61). 119 Figure 2.46. Proposed common overlap for paclitaxel (4), epothilone B (53), and eleutherobin (61). 120 Figure 2.47. Proposed common overlap and pharmacophore for paclitaxel (4), epothilone B (53), and sarcodictyin A (55). 121 xi i Figure 2.48. Superposition of nine solution conformations of eleutherobin (61) in DMSO-c/6. 123 Figure 2.49. Taxonomy of the antimitotic extracts from the N C I Open Repository relative to E. caribaeorum. 124 Figure 2.50. Map of the geographic collection locations of sarcodictyin/eleutherobin-producing organisms. 125 Figure 3.1. Stylissa flabelliformis (photo by P.L. Colin). 145 Figure 3.2. ' H N M R spectrum of debromohymenialdisine (91)-HC1 recorded in D M S O - ^ 6 at 400 M H z . 150 Figure 3.3. 1 3 C N M R spectrum of debromohymenialdisine (91)-HC1 recorded in D M S O - J 6 at 100 M H z . 151 Figure 3.4. *H N M R spectrum of hymenialdisine (85)-HCl recorded in D M S O - d 6 at 500 M H z . 153 Figure 3.5. 1 3 C N M R spectrum of hymenialdisine (85)-HCl recorded in D M S O - J 6 at 100 M H z . 154 Figure 3.6. ' H N M R spectrum of Z-debromoaxinohydantoin (94) recorded in D M S O - J 6 at 400 M H z . 156 Figure 3.7. 1 3 C N M R spectrum of Z-debromoaxinohydantoin (94) recorded in D M S O - J 6 at 100 M H z . 157 Figure 3.8. ' H N M R spectrum of aldisin (103) recorded in D M S O - ^ 6 at 400 M H z . 159 Figure 3.9. I 3 C N M R spectrum of aldisin (103) recorded in DMSO-t /6 at 100 M H z . 160 Figure 3.10. Inhibition of the G2 checkpoint by debromohymenialdisine (91). 162 Figure 3.11. Inhibition of protein kinase C h k l by debromohymenialdisine (91). 164 Figure 3.12. Inhibition of protein kinase Chk2 by debromohymenialdisine (91). 164 Figure 4.1. Characteristic conformations and l 3 C N M R chemical shift values for syn-and £M /M,3-dio ls . 182 Figure 4.2. ' H N M R spectrum of £-cryptofolione (130) recorded in C D C 1 3 at 400 M H z . 183 Figure 4.3. 1 3 C N M R spectrum of ^-cryptofolione (130) recorded in C D C 1 3 at 100 M H z . 184 Figure 4.4. ' H N M R spectrum of Z-cryptofolione (148) recorded in C D C 1 3 at 500 M H z . 187 Figure 4.5. 1 3 C N M R spectrum of Z-cryptofolione (148) recorded in C D C 1 3 at 100 M H z . 188 xiii Figure 4.6. ' H N M R spectrum of cryptofolione ketone (131) recorded in C D C 1 3 at 500 M H z . 191 Figure 4.7. L , C N M R spectrum of cryptofolione ketone (131) recorded in C D C 1 3 at 100 M H z . 192 Figure 4.8. ' H N M R spectrum of compound 138 recorded in C D C 1 3 at 500 M H z . 195 Figure 4.9. 1 3 C N M R spectrum of compound 138 recorded in C D C 1 3 at 100 M H z . 19.6 Figure 4.10. ' H N M R spectrum of cryptofolione acetonide (149) recorded in C D C 1 3 at 400 M H z . 199 Figure 4.11. 1 3 C N M R spectrum of cryptofolione acetonide (149) recorded in C D C 1 3 at 100 M H z . 200 Figure 4.12. ' H N M R spectrum of diacetylcryptofolione (150) recorded in C D C 1 3 at 500 M H z . 203 Figure 4.13. 1 3 C N M R spectrum of diacetylcryptofolione (150) recorded in CDCI3 at 100 M H z . 204 Figure 4.14. Selected C O S Y correlations in the (3-mercaptoethanol adduct (151). 206 Figure 4.15. ' H N M R spectrum of p-mercaptoethanol adduct (151) recorded in C D C 1 3 at 500 M H z . 207 Figure 4.16. I 3 C N M R spectrum of (3-mercaptoethanol adduct (151) recorded in C D C 1 3 at 100 M H z . " 208 Figure 4.17. ' H N M R spectrum of compound 152 recorded in C D C 1 3 at 500 M H z . 211 Figure 4.18. 1 3 C N M R spectrum of compound 152 recorded in C D C 1 3 at 100 M H z . 212 Figure 4.19. Inhibition of the'G2 checkpoint by natural and semi-synthetic a-pyrones. 214 Figure A.l. C O S Y spectrum of eleutherobin (61) recorded in D M S O - J 6 at 500 M H z . 225 Figure A.2. H M Q C spectrum of eleutherobin (61) recorded in D M S O - d 6 at 500 M H z . 226 Figure A.3. H M B C spectrum of eleutherobin (61) recorded in DMSO-t /6 at 500 M H z . 227 Figure A.4. C O S Y spectrum of desacetyleleutherobin (70) recorded in D M S O - J 6 at 500 M H z . 228 Figure A.5. H M Q C spectrum of desacetyleleutherobin (70) recorded in D M S O - ^ 6 at 500 M H z . 229 xiv Figure A.6. H M B C spectrum of desacetyleleutherobin (70) recorded in D M S O - d 6 at 500 M H z . 230 Figure A.7. C O S Y spectrum of isoeleutherobin A (71) recorded in D M S O - d 6 at 500 M H z . 231 Figure A.8. H M Q C spectrum of isoeleutherobin A (71) recorded in D M S O - J 6 at 500 M H z . 232 Figure A.9. H M B C spectrum of isoeleutherobin A (71) recorded in DMSO-rf6 at 500 M H z . 233 Figure A.10. C O S Y spectrum of Z-eleutherobin (72) recorded in D M S O - J 6 ! at 500 M H z . 234 Figure A.11. H M Q C spectrum of Z-eleutherobin (72) recorded in D M S O - ^ 6 at 500 M H z . 235 Figure A.12. H M B C spectrum of Z-eleutherobin (72) recorded in DMSO-<i6 at 500 M H z . 236 Figure A.13. C O S Y spectrum of caribaeoside (73) recorded in D M S O - ^ 6 at 500 M H z . 237 Figure A.14. H M Q C spectrum of caribaeoside (73) recorded in D M S O - J 6 at 500 M H z . 238 Figure A.15. H M B C spectrum of caribaeoside (73) recorded in DMSO-c/6 at 500 M H z . 239 Figure A.16. R O E S Y spectrum of caribaeoside (73) recorded in D M S O - J 6 at 500 M H z . 240 Figure A.17-. C O S Y spectrum of caribaeolin (74) recorded in D M S O - d 6 at 500 M H z . 241 Figure A.18. H M Q C spectrum of caribaeolin (74) recorded in D M S O - J 6 at 500 M H z . 242 Figure A.19. H M B C spectrum of caribaeolin (74) recorded in D M S O - J 6 at 500 M H z . 243 Figure A.20. R O E S Y spectrum of caribaeolin (74) recorded in D M S O - J 6 at 500 M H z . 244 Figure A.21. C O S Y spectrum of debromohymenialdisine (91)-HCl recorded in DMSO-ci6 at 500 M H z . 245 Figure A.22. H M Q C spectrum of debromohymenialdisine (91)-HC1 recorded in D M S O - J 6 at 500 M H z . 246 Figure A.23. H M B C spectrum of debromohymenialdisine (91)-HC1 recorded in DMSO-</6 at 500 M H z . 247 Figure A.24. C O S Y spectrum of hymenialdisine (85)-HCl recorded in DMSO-c/6 at 500 M H z . 248 X V Figure A.25. H M Q C spectrum of hymenialdisine (85)-HCl recorded in DMSO-c/6 at 500 M H z . 249 Figure A.26. H M B C spectrum of hymenialdisine (85)-HCl recorded in DMSO-c/6 at 500 M H z . 250 Figure A.27. C O S Y spectrum of £-cryptofolione (130) recorded in C D C 1 3 at 500 M H z . 251 Figure A.28. H M Q C spectrum of £-cryptofolione (130) recorded in C D C 1 3 at 500 M H z . 252 Figure A.29. H M B C spectrum of ^-cryptofolione (130) recorded in C D C 1 3 at 500 M H z . 253 Figure A.30. C O S Y spectrum of Z-cryptofolione (148) recorded in C D C 1 3 at 500 M H z . 254 Figure A.31. H M Q C spectrum of Z-cryptofolione (148) recorded in C D C 1 3 at 500 M H z . 255 Figure A.32. H M B C spectrum of Z-cryptofolione (148) recorded in C D C 1 3 at 500 M H z . 256 Figure A.33. C O S Y spectrum of cryptofolione ketone (131) recorded in C D C 1 3 at 500 M H z . 257 Figure A.34. H M Q C spectrum of cryptofolione ketone (131) recorded in C D C 1 3 at 500 M H z . 258 Figure A.35. H M B C spectrum of cryptofolione ketone (131) recorded in C D C 1 3 at 500 M H z . 259 Figure A.36. C O S Y spectrum of compound 138 recorded in C D C 1 3 at 500 M H z . 260 Figure A.37. H M Q C spectrum of compound 138 recorded in C D C 1 3 at 500 M H z . 261 Figure A.38. H M B C spectrum of compound 138 recorded in C D C 1 3 at 500 M H z . 262 xvi LIST OF SCHEMES Page Scheme 2.1. Formation of geranylgeranyl pyrophosphate. 39 Scheme 2.2. Proposed biosynthetic relationship between cembranoid diterpenes. 40 Scheme 2.3. Isolation procedure for antimitotic diterpenoids from E. caribaeorum. 52 Scheme 2.4. Taxonomic scheme showing the related chemistry isolated from octocorals. 125 Scheme 3.1. Biogenetic precursors of odiline/stevensine from the sponge T. mprchella. 139 Scheme 3.2. Representative structural groups of oroidin alkaloids. 140 Scheme 3.3. The six cyclization modes of the oroidin skeleton. 141 Scheme 3.4. Isolation procedure for G 2 checkpoint inhibitors from S. flabelliformis. 146 Scheme 4.1. Isolation procedure for G 2 checkpoint inhibitors from C. concinna. 178 xvii LIST OF ABBREVIATIONS A angstrom [a]d2° specific rotation at wavelength of sodium D line at 20 °C Ac acetyl A T M ataxia telangiectasia mutated kinase ATR ataxia telangiectasia mutated-related kinase ATP adenosine triphosphate ax axial B.C. British Columbia br broad c concentration (g/100 ml) CD circular dichroism CDCI3 deuterated chloroform Cdk cyclin-dependent kinase CHCI3 chloroform CH2C12 dichloromethane CI chemical ionization COSY correlation spectroscopy ° degrees 5 chemical shift in parts per million d doublet ID one-dimensional 2D two-dimensional D B H debromohymenialdisine xvni D C M dichloromethane D M S O dimethyl sulfoxide D M S O - J 6 deuterated dimethyl sulfoxide D N A - P K DNA-activated protein kinase e molar absorptivity coefficient EI electron impact E L I C A Enzyme Linked Immuno-Cytochemical Assay E L I S A Enzyme Linked Immuno-Sorbent Assay Et ethyl E t O A c ethyl acetate eq equatorial g gram G T P guanosine triphosphate Gy grays H M B C heteronuclear multiple bond multiple quantum coherence H M Q C heteronuclear multiple quantum coherence H P L C high performance liquid chromatography H R C I M S high resolution chemical ionization mass spectrometry H R E L M S high resolution electron impact mass spectrometry H R F A B M S high resolution fast atom bombardment mass spectrometry H R P horseradish peroxidase H z hertz i signal due to impurity IC50 Inhibitory Concentration resulting in 50% maximal response ( 'Pr) 2 NH diisopropyl amine X I X J scalar coupling constant L R F A B M S low resolution fast atom bombardment mass spectrometry m multiplet M molar M + molecular ion Me methyl M e O H methanol M H z megahertz ml milliliter m/z mass to charge ratio u M micromolar N normal N C I National Cancer Institute nm nanometers N M R Nuclear Magnetic Resonance N O E Nuclear Overhauser Effect O R T E P Oak Ridge Thermal Ellipsoid Plot Ph phenyl ppm parts per million q quartet R O E S Y rotating frame Overhauser effect spectroscopy s singlet or signal due to solvent S C U B A self-contained underwater breathing apparatus SD standard deviation sp. species t triplet T F A trifluoroacetic acid T H F tetrahydrofuran U B C University of British Columbia US United States U V ultraviolet w signal due to water x times xxi ACKNOWLEDGEMENTS I wish to express my deepest appreciation to my research supervisor, Professor Raymond J. Andersen for the many years of endless enthusiasm, brilliant insight, constant encouragement and support. I wi l l be forever grateful for the opportunity to learn and work under his esteemed guidance. Many heartfelt thanks are extended to Dr. David E . Wil l iams and Dr. Michel Roberge for all their invaluable instruction and patient assistance over the past years. I have learned much and benefited enormously from their efforts. M y appreciation to all the members of the Andersen and Roberge research groups, especially John Coleman, Lynette L i m , Darko Curman, Cristina Biggs, and Rebecca Osborne who assisted on the biological testing and other aspects of this thesis research. Thanks also to Michael LeBlanc and Robert Britton for collection of the marine specimens used in this research, and the staff of the N M R , M S , and X-ray facilities at U B C for their assistance in collecting data. To my friends for their kindness and encouragement throughout the years: Jason, Erica, Patti, and especially Kyle , Susie, and above all, Carine with all my love. Finally, I wish to acknowledge the incalculable and unwavering support of my parents, Ivo and Bianca Cinel , for everything. Words fail to describe my profound gratitude for all you have done for me... CHAPTER 1 1. General Introduction Cancer remains one of the leading causes of death worldwide, and the limitations of conventional therapies coupled with an ageing population wi l l continue to impact the pervasiveness of this disease well into the new millennium. This threat, along with the emergence of resistance mechanisms to current therapeutic agents, has continued to drive the search for novel therapies. Recent breakthroughs in molecular biology have identified important genetic differences between normal and cancerous cells, yielding new insights into the molecular basis of this disease. B y identifying new chemical entities with specific antimitotic or cell cycle checkpoint inhibitory activities, the intricate machinery of a cancerous cell may be better understood and new treatments or cures developed. The goal of this thesis was to use two recently developed cell-based assays to identify antimitotic compounds and G2 checkpoint inhibitors from marine and terrestrial sources. These screens facilitated the isolation and structure determination of novel antimitotic diterpenoids from a Caribbean octocoral, alkaloid G2 checkpoint inhibitors from a marine sponge, and polyketide G2 checkpoint inhibitors from a Taiwanese tree. Further investigations on these biologically active secondary metabolites and their semi-synthetic analogues provided key insights into the structural features required for activity and helped to identify molecular targets and possibly novel mechanisms of action. 1.1. Natural Products Chemistry From the ancient Mesopotamian and Chinese civilizations to the present day, plant extracts have been widely used as traditional medicinal agents.1 A t the dawn of the new millennium, natural sources continue to provide important compounds as possible new 2 therapeutic agents or biological tools in the fight against life-threatening diseases such as cancer. Cancer is currently the leading cause of death in Canada and is set to emerge as the number one killer in the United States. 2 , 3 This ominous news comes nearly three decades after a "war on cancer" was declared in the U S and it highlights the urgent need for improved medical treatments. The chemotherapeutic approach represents one well-established and still attractive approach for treating this serious disease.4'5 New chemical entities offer the promise of future drugs with increased specificity, fewer side effects, novel modes of action to counter resistance, and potential to act as essential biological tools to probe the cellular mechanisms of this disease. Historically, natural products have been proven to be the foremost source for new leads in drug discovery and development. Indeed, some of the earliest active compounds isolated as pure entities are still in widespread use today. Morphine (1) was isolated in 1806 from an extract of Papaver somniferum and became the first commercially-pure medicinal product some twenty years later.6 At present, there are still no alternatives to morphine for patients suffering from the severe pain that is common in the late stages of cancer. The discovery that the analgesic and antipyretic properties of the bark of the willow tree were due to salicylic acid led to the development of the natural product derivative, aspirin (2). Over one hundred years after its initial introduction, aspirin remains the most successful drug ever developed. 6 The serendipitous discovery of penicillin by Fleming in 1928 (as penicillin G , 3)7 revolutionized drug discovery by revealing the potential of soil microorganisms to be a rich new source of biopharmaceuticals. Cultures of microbes have since provided a seemingly unlimited array of unique and complex structures with a broad variety of biological activities. Production of drugs via microbial fermentation removes the barriers presented by large-scale harvesting of wild specimens and difficult or costly syntheses that plague the development of some terrestrial plant agents. 3 Despite the obstacles involved in large-scale production of phytogenic drugs, plant metabolites have continued to be invaluable as therapeutics. Paclitaxel, or Taxol® (4), was originally isolated in the late 1960's from the Pacific Yew tree, Taxus brevifolia, after an extensive screen of over 110,000 plant extracts for cytotoxic agents.8 Its novel mechanism of action and the urgent need for improved anticancer chemotherapeutics led to its present day use in the treatment of ovarian and metastatic breast cancers. The traditional Chinese medicinal plant qinghao, or Artemisia annua, has been used for centuries in the treatment of fever and malaria. 9 Isolation of the active agent, artemisinin (5), provided a therapeutic treatment for quinine-resistant forms of malaria. 1 0 Both paclitaxel (4) and artemisinin (5) have spawned clinically useful semi-synthetic analogues, including docetaxel (6) and artemether (7), which have improved medicinal properties and decreased side effects. 4 5 4 O H Q O O H O' N H O H O O H ; o :o H O M e H 6 7 1.2. Marine Natural Products Chemistry In addition to terrestrial plants and soil microorganisms, a third resource of natural product chemical diversity is the marine environment. The total inhabitable space in the world's oceans greatly exceeds the narrow land surface zone and provides a vast array of species biodiversity throughout its wide range of habitats. Of the 28 principle animal phyla, 26 are represented in the marine biota and eight of these are exclusively aquatic in nature. In particular, many marine invertebrates exhibit soft bodied, sessile, slow moving, slow growing, or brightly coloured characteristics that leave them seemingly unprotected from predation. However, these organisms may chemically defend themselves by use of a highly complex arsenal of toxic secondary metabolites. Greater access to the marine environment came with the advent of S C U B A in 1943, opening the door to nearly 40 years of investigation into marine natural products. During this time, research in the secondary metabolites from marine organisms flowed along three parallel tracks: marine chemical ecology, marine toxins, and marine biopharmaceuticals." In contrast to primary metabolites such as amino acids and nucleotides, secondary metabolites are not ubiquitous in nature, are nonessential to the daily existence of an organism, and are often species specific. However, these compounds are thought to play key ecological roles as antifeedant,12 antifouling, 1 3 antiovergrowth, or antimicrobial 1 4 agents and as chemical cues to mating, settling, or metamorphosis. 1 3 ' 1 5 The biological roles of most marine invertebrate natural products remain unknown, but such biodiversity is an attractive lure for modern drug discovery. When marine plants and microorganisms are also included, marine organisms present themselves as a vast and largely unexplored source of species and chemical diversity. Some of the largest and most impressive marine metabolites have emerged from the study of marine toxins related to algal blooms and the associated seafood poisonings. The polyether "ladder-like" structure common to a majority of these toxins was first revealed by the study of the dinoflagellate metabolite brevetoxin B (8). The brevetoxins, produced by the "red tide" dinoflagellate Gymnodinium breve, comprise a series of related neurotoxins that interfere with sodium voltage channels and are responsible for massive marine life fatalities. 1 6 ' 1 7 Maitotoxin (9), isolated from the dinoflagellate Gambierdiscus toxicus and involved in ciguatera seafood poisoning, represents the largest and possibly most lethal non-biopolymer identified to 18 date. These very complex compounds exhibiting such potently harmful biological activities stand in contrast to the many marine metabolites showing promising therapeutic effects. Much of the research in marine natural products has focused on finding new chemotherapeutics or pharmaceutical agents. The first success in this area came early in the 1950's when Bergmann et al. isolated two ara-nucleosides, spongouridine (10) and spongothymidine (11.) from the Caribbean sponge Cryptotethya crypta}9 These compounds were the first nucleosides found to possess an arabinose sugar moiety and they served as "lead structures" for the development of many modified nucleosides with antiviral and antitumour properties.1 The synthetic compound Ara -A, or arabinosyladenine (12), has been used since the 1970's as an antiviral agent to treat Herpes encephalitis and was later found to occur naturally 20 in the Mediterranean gorgonian Eunicella cavolini. Fifteen years after Bergmann's discovery of 10 and 11, the cytosine arabinoside Ara-C (13) was developed into a clinical anticancer 21 agent and it currently remains one of the most active agents available for treatment of acute non-lymphocytic leukemia. 5 OH OH OH 10 R = H 11 R = C H 3 12 13 7 Many compounds more recently isolated from marine organisms have progressed to clinical trials for evaluation as new chemotherapeutics. Bryostatin 1 (14), one of a series of potently cytotoxic macrolides isolated from the bryozoan Bugula neritina, is currently in phase I and phase U clinical trials for the treatment of melanoma. 2 2 A n insufficient supply of 14 from harvesting wi ld bryozoan specimens led to the successful implementation of an aquaculture program capable of producing enough material for testing. ' With recent studies suggesting a microbial symbiont as the true source of bryostatin 1, advances in the genetic manipulation of biosynthetic genes may forever eliminate future supply problems. 2 4 Investigations of the shallow water tunicate Ecteinascidia turbinata by two independent research groups led to the simultaneous discovery of a series of tetrahydroisoquinoline alkaloids, including ecteinascidin 743 (15).25 2 6 This compound, which acts as a guanine-specific D N A alkylating agent, is currently in clinical trials as an antitumour agent.2 7 The similarity in structure between the ecteinascidins and the safracins isolated from Streptomyces lavendulae27"28 suggest these metabolites may all be microbial in origin and thus may eliminate the need to harvest wild specimens or undertake expensive chemical syntheses to obtain sufficient quantities for future use. 14 15 8 Chemical syntheses were required to confirm the structure of dolastatin 10 (16), a potent antimitotic agent originally isolated from approximately two tonnes of the Indian Ocean sea hare Dolabella auricularia29 This compound contains a series of novel amino acids and exhibits a complex mechanism of action involving inhibition of tubulin assembly during cell d iv i s ion . 2 2 5 Promising antineoplastic activity against all experimental cancers tested so far has led to the clinical trials of dolastatin 10 and related analogues. In contrast to dolastatin 10's inhibition of tubulin polymerization, the Caribbean sponge metabolite discodermolide (17)30 interferes with cell division by stabilizing microtubules against depolymerization. 3 1 Although this mechanism of action is similar to that of paclitaxel (4), discodermolide was also found to inhibit the growth of paclitaxel-resistant ovarian and colon cancer cel ls . 3 2 Another marine sponge natural product, the unusual sesterterpenoid dysidiolide (18), was found to be a possible inhibitor of the protein phosphatase Cdc25A. 3 ; > Inhibition of Cdc25A phosphatase, an enzyme known to be overexpressed in a number of tumour cell lines, should arrest the cell cycle at the G 2 / M transition just prior to cell division. Compounds such as dysidiolide (18) could potentially exploit the genetic differences between normal and cancerous cells and lead to new cancer chemotherapeutics. 17 18 9 While most of the efforts in finding new biopharmaceuticals from marine natural products have focused on anticancer compounds, there have been key discoveries in other areas as well. In particular, the pseudopterosins isolated from the Caribbean gorgonian, Pseudopterogorgia elisabethae, exhibit potent anti-inflammatory and analgesic activities. 3 4 ' 3 3 The diterpene pentose glycoside pseudopterosin C (19) mediates inflammation by acting on both lipooxygenase and phospholipase A? and has found commercial use as an active ingredient in a cosmetic, anti-ageing skin cream. A compound isolated from the Pap'ua New Guinean sponge Petrosia contignata, has been found to inhibit the release of histamine from anti-Ige stimulated cells , 3 6 thus showing promise as an anti-asthmatic treatment. This discovery of contignasterol (20) and its biological activity has led to the clinical development of an orally-active synthetic analogue that appears to lack the toxic side-effects associated with current steroidal anti-asthma drugs. OH Investigations of the secondary metabolites produced by marine organisms have revealed thousands of unique terpenoid, alkaloid, and polyketide chemical structures that have no terrestrial counterparts.3 7 As shown in Figure 1.1, the two most prolific marine sources of new compounds reported in the recent literature are the sponges (phylum Porifera) and the corals (phylum Coelenterata). Organisms from these two phyla remain a rich source of 10 structurally diverse organic molecules and form the basis of two of the chemical investigations presented in this thesis. Figure 1.1. Distribution of marine natural products reported in 1999 according to the phylum of the source organism. 3 7 Although interest in natural products as a source of chemical structural diversity occasionally becomes overshadowed by other approaches to drug discovery such as combinatorial chemistry and computer-based molecular modeling, statistics show natural products or synthetic analogues account for over 60% of recently approved drugs or late stage clinical trial drug candidates for cancer and infections diseases. Synthetic combinatorial libraries often contain closely related structural analogues, whereas a collection of biological extracts from widely divergent taxonomic groups, ecological niches, and geographic regions possess a much higher degree of molecular diversity. In addition, only a small percentage of the world's plant and animal species have been systematically evaluated for their medicinal value 3 9 and traditional screens for biological activity have focused on cytotoxicity or antimicrobial activity. While current drugs are estimated to target only 417 different enzymes or receptors, it has been suggested that the study of human genomics wil l improve drug therapy by providing 11 an estimated 10,000 potential new drug targets.6 The recent development of novel mechanism-based bioassays searching for specific inhibitors of molecular targets or biological processes demands a reassessment of previously examined natural extracts along with investigations into new collections. 1.3. Novel Anticancer Screens The term "cancer" refers to a broad group of more than 100 different forms of disease that are characterized by uncontrolled proliferative growth and the spread of aberrant cells from their site of origin. Most of the major cancers, or solid tumours, occur in localized tissues and arise from a common ancestral cell that has lost its ability to control cellular growth and reproduction as a result of mutations. Advances in genetics and molecular biology have identified many genetic differences between cancerous and normal cel ls , 4 0 " 4 3 differences that can be exploited to selectively target and control malignant cells. 1.3.1. The Cell Cycle4 4 4 5 The growth and reproduction of eukaryotic cells is determined by a precisely timed sequence of events known as the cell cycle. A normal functioning cell progresses through various stages of development and ultimately divides to form two genetically identical daughter cells. Any alteration or mutation in the complex cellular machinery responsible for these functions may result in uncontrolled cell proliferation, altered phenotype, or cell death. These genetic changes and cell cycle pathways may be exploited by anticancer agents that act by interfering with some aspect of the cell cycle. The cell cycle can be divided into two main stages, interphase and M phase (see Figure 1.2). Interphase represents the stage of cellular growth and replication of D N A and consists of three separate phases beginning with the G l or Gap l phase. During G l , the cell grows and produces enzymes necessary for the duplication of its genetic material. The S phase that 12 follows is responsible for synthesizing an exact copy of the cell 's D N A . After completion of the S phase, the cell enters the second gap phase, or G2. At this time, there is an increase in protein synthesis in preparation for cell division. Interphase Figure 1.2. A n overview of the eukaryotic cell cycle. The M phase of the cell cycle consists of mitosis, the distribution of a complete set of chromosomes to each daughter nucleus, and cytokinesis, the formation of two genetically identical cells. Mitosis itself can be divided into four sequential stages: prophase, metaphase, anaphase and telophase. Although these divisions have distinct morphological characteristics, the transition from one stage to the next is smooth and continuous (see Figure 1.3, A - F ) . The first stage of mitosis, prophase, is characterized by the condensation of the sister chromatids of each chromosome, the migration of the centrioles to opposite ends of the cell, and the early assembly of microtubules into the mitotic spindle. A t the onset of metaphase, the nuclear envelope disperses and the spindle microtubules begin to attach to the center 13 (kinetochore) of each chromosome. Towards the end of metaphase, the mitotic spindle is completely formed and each chromosome is aligned along the spindle equator. Anaphase marks the separation of the two sister chromatids of each chromosome as the microtubules slowly pull the now independent daughter chromosomes towards opposite poles. The last stage of mitosis, telophase, concludes with the decondensation of the chromosomes and the formation of a new nuclear envelope to surround each complete set of D N A . Cytokinesis results in the final division of the cytoplasm and creation of two daughter cells. B / C D E F ••••• • Figure 1.3. Fluorescent micrographs depicting cells in various stages of mitosis. In these examples, the microtubules (red) were stained using indirect immunofluorescent methods, while the chromosomes (green) were stained with the DNA probe Hoechst 33342. (A) A late prophase cell in which the two centrioles and their associated microtubules have separated to opposite sides of the nucleus. (B) A early metaphase cell that contains one monooriented chromosome (white arrow). (C) A metaphase cell in which all the chromosomes are aligned on the spindle equator. (D) A cell just entering anaphase in which the chromatids are disjoining. (E) A late anaphase cell in which the two groups of chromosomes are at the spindle poles. (F) A telophase cell in which the two groups of chromosomes are reforming nuclei and in which cytokinesis (between the white arrowheads) is almost complete. 14 1.3.2. Antimitotic Agents The uncontrolled cellular proliferation characteristic of cancerous tumours may be inhibited by compounds acting as antimitotic agents. Indeed, several natural products that act by arresting cells in mitosis have become clinically important anticancer drugs. The Vinca alkaloids vincristine (21) and vinblastine (22) were originally isolated from the periwinkle plant and have been used for over three decades in the chemotherapy of neoplastic diseases, such as childhood leukemia and Hodgkin's disease.47 They, along with the taxanes paclitaxel (4) and docetaxel (6), cause mitotic arrest by interfering with the assembly or disassembly of a- and P-48 tubulin into microtubules. R C 0 2 M e 21 R = CHO 22 R = Me As shown earlier, microtubules play a fundamental role throughout the process of cellular division, most importantly in attaching to and segregating the chromosomes. Microtubules are largely composed of the protein tubulin, a dimer comprised of a and (3 subunits, that joins in a head-to-tail fashion in the presence of GTP to form a long protein fiber called a protofilament. Typically 12 or 13 of these protofilaments group together to ultimately form the microtubule's hollow, pipe-like structure (see Figure 1.4). Rapid assembly and disassembly of tubulin dimers occurs at the ends of the microtubules resulting in a continuous, dynamic state of elongation and contraction. It is this dynamic polymerization and depolymerization of the microtubules that is disrupted by antimitotic agents. 15 There are two main mechanisms by which most antimitotic compounds act on tubulin and microtubules. A t high concentrations most antimitotic agents, including the Vinca alkaloids and dolastatin 10 (16),48'47 inhibit microtubule assembly and cause complete depolymerization. A second group of antimitotic compounds, such as the taxanes and discodermolide (17),49'47 stabilizes the microtubules against depolymerization and leads to abnormal microtubule arrays or bundles. Neither depolymerization nor bundling is observed at low concentrations but the dynamics of tubulin addition and removal from the mitotic spindle microtubules are sufficiently affected so as to inhibit normal function. Thus, the microtubules are prevented from attaching to and segregating the chromosomes during mitosis and the cells arrest at this stage of cell d iv i s ion . 5 0 , 5 1 Prolonged arrest eventually leads to cell death, either in mitosis or after an eventual escape from mitotic arrest. The current antimitotic chemotherapeutics are far from ideal. They display numerous unwanted side effects, including myelosuppression and neurotoxicity, and many cancers now demonstrate multi-drug resistance to these therapies. 4 8 ' 4 9 Resistance mechanisms can arise from overexpression of the P-glycoprotein drug efflux pump, from tubulin isotypes exhibiting weaker drug binding, and from mutational or post-translational alterations in the structure of tubulin. 16 Screening natural extracts may provide new chemical entities with increased specificity, fewer unwanted side effects, and antineoplastic activity against drug-resistant cancers. The antimitotic agents currently available, including those in pre-clinical studies or clinical trials, were discovered either by screening for cytotoxicity, by exhibiting cytotoxicity patterns against multiple cell lines similar to other antimitotics, or by serendipity. " The determination of their mode of action was secondary to their isolation and characterization as cancer-killing compounds. In response to the lack of a rational drug discovery screen for antimitotic agents, a rapid and reliable cell-based assay was recently developed in the laboratory of Michel Roberge. 3 3 1.3.3. The Antimitotic Bioassay53 Roberge's antimitotic bioassay was found to be suitable for drug discovery and for quantitative determination of antimitotic activity. In addition, this cell-based system provides many advantages over receptor or enzyme-based screens. These advantages include the elimination of many non-specific false positives associated with cell-free screens due to the need for agents to cross the cell membrane to demonstrate activity. 5 4 In a recent study, over 90% of compounds identified from in vitro target-based assays either failed to cross the plasma membrane or were degraded rapidly, resulting in no biological activity. 5 5 In Roberge's assay, exposure to antimitotic compounds, either in pure form or as crude natural extracts, causes rapidly cycling human breast carcinoma cells to arrest in mitosis. The arrested cells are detected by an Enzyme-Linked Immuno-Sorbent Assay (ELISA) using the monoclonal antibody T G - 3 . 5 6 The TG-3 monoclonal antibody is highly specific for mitotic cells and flow cytometry studies show a >50-fold more intense TG-3 immunofluorescence in mitotic cells than in interphase cel ls . 5 7 The antibody specifically recognizes^ a phosphoepitope of nucleolin, 5 8 found abundantly in cells arrested in mitosis but only present in low levels in 17 cycling cells. This method detects phosphorylated nucleolin and thus mitotic cells by an E L I S A 5 9 using 96-well microtitre plates and the addition of an HRP-conjugated secondary antibody. The colorimetric determination of H R P activity provides a quantitative determination of antimitotic activity. The antimitotic bioassay is shown schematically in Figure 1.5. Human breast carcinoma M C F - 7 cells were cultured as monolayers, seeded on 96-well microtitre plates, and allowed to grow overnight. Pure compounds or crude natural extracts were then added to the cells and incubated for 16-20 hours. A t this stage, the presence of an antimitotic agent would cause the cells to arrest in mitosis and the phosphorylation of nucleolin to occur. After incubation with the extracts, the cell culture medium was carefully withdrawn from the plate without removing any cells. The remaining attached cells were lysed by buffer and transferred to 96-well protein binding E L I S A plates. Addition of the TG-3 monoclonal antibody was followed by incubation for 16-20 hours, during which time any phosphorylated nucleolin would bind to the antibody. Introduction of the secondary antibody, HRP-conjugated goat antimouse IgM, and subsequent overnight incubation would attach the secondary antibody to any TG-3/protein complexes present. After a final rinse and addition of hydrogen peroxide, the H R P activity was determined by measuring the absorbance at 405 nm. A positive colorimetric result would indicate the cells had been arrested in mitosis due to the addition of an antimitotic sample. Improvements to the bioassay resulted from the substitution of the lengthy and labour-intensive E L I S A procedure with a simpler, shorter protocol. The E L I C A , or Enzyme-Linked Immuno-Cytochemical Assay, involves fixing the cells on the 96-well microtitre plates with formaldehyde and permeabilizing the cells with methanol and detergents. These modifications, along with the subsequent simultaneous addition of the TG-3 primary antibody and the H R P -conjugated secondary antibody, reduce the total time and steps by half and eliminate the transfer to new plates. The suitability of both the E L I S A and E L I C A procedures for quantifying 18 antimitotic activity was tested by using the known antimitotic paclitaxel (4) and confirmed by microscopy. Human breast carcinoma MCF-7 cells seeded on 96-well plates grown overnight incubate 16-20 hours Antimitotic agent present cells arrested in mitosis nucleolin phosphorylated No antimitotic agent present cells continue cycling nucleolin un-phosphorylated cell culture medium removed cells lysed and transferred to ELISA plates TG-3 antibody added incubate for 16-20 hours TG-3 antibody binds nucleolin-P No binding of TG-3 antibody HRP-conjugated antibody added Complex of two antibodies and nucleolin-P • No complex incubate overnight rinse and add hydrogen peroxide POSITIVE SIGNAL ANTIMITOTIC AGENT PRESENT IN EXTRACT Figure 1.5. The antimitotic bioassay. NEGATIVE SIGNAL NO ANTIMITOTIC PRESENT IN EXTRACT The antimitotic screen developed by Roberge et al. provided a new rational assay for natural products capable of arresting cells in mitosis. In total, over 24,000 extracts of marine microorganisms, marine invertebrates, and terrestrial plants were screened for antimitotic activity, with 119 extracts testing positive. The bioassay-guided investigation of one of the 19 active extracts, originating from a Caribbean octocoral, is the subject of Chapter Two of this thesis. 1.3.4. Cell Cycle Checkpoints As a cell replicates and divides, it is vital that the genetic information encoded in its DNA is duplicated and passed on accurately to the two new daughter cells. Any errors or deficiencies in these processes could lead to mutations and the manifestation of proliferative disorders such as cancer or even cell death. In response to D N A damage, a series of signal transduction pathways called checkpoints are activated which delay cell cycle progression and allow for D N A repair.60"62 Checkpoints arrest cells in G l phase to prevent the replication of damaged D N A and in G2 phase to prevent the segregation of damaged chromosomes during mitosis. While many investigations have focused on genetic studies in yeast, the complex mechanisms in mammalian cell cycle checkpoints have only recently become better understood (see Figure 1.6). ft .V //•• //• ?'? Apoptosis Figure 1.6. The series of signal transduction pathways in mammalian cell cycle checkpoints. (Biochemically well-defined responses are shown in red). 20 The first D N A damage checkpoint occurs prior to the S phase of the cell cycle and exists to prevent replication of defective genetic material. The G l cell cycle checkpoint is regulated by three important gene products: A T M , p53 and p21, whose levels and activities strictly control entry into S phase. 6 4" 6 6 In response to D N A damage, the ataxia telangiectasia mutated ( A T M ) kinase is thought to contribute to p53 activation by phosphorylation at the serine-15 res idue. 6 7 ' 6 8 Recently, A T M was also found to activate the kinase Chk2, which in turn phosphorylates p53 at serine-20 (see inset Figure 1.6).69 In undamaged cells that are dividing normally, p53 is highly unstable and rapidly degraded. After D N A damage induced by ionizing radiation, the two phosphorylation events by A T M and Chk2 lead to significant stabilization of p53 and subsequent increase in its transcription of p21. 6 5 ' - 7 0 Cells deficient in p53 function are unable to arrest in G l in response to y-irradiation and this transcriptional activation of the cyclin-dependent kinase (Cdk) inhibitor p21 is linked to the ability of p53 to induce G l arrest. The elevated levels of p21 lead to the binding to and inactivation of a variety of cyclin/Cdk complexes, such as cyclin E / C d k 2 , 7 1 thus preventing entry into S phase and the replication of genetic material. In addition to activating the G l cell cycle checkpoint, D N A damage also initiates D N A repair processes, G2 checkpoint pathways and, depending on the severity of the genetic damage, apoptosis. According to the current understanding of the G2 checkpoint signaling pathways, D N A damage activates members of the phosphoinositide-3 kinase family including A T M , ATM-related kinase (ATR) , and DNA-activated protein kinase ( D N A - P K ) . 7 2 The A T M -dependent signaling pathway involves activation of the downstream protein kinases C h k l and C h k 2 . 6 4 ' 7 3 " 7 7 These kinases directly mediate the phosphorylation of serine-216 on Cdc25c phosphatase, resulting in an increased binding affinity for the 14-3-3 family of proteins (see Figure 1.6). 7 6 ' 7 8" 8 1 The 14-3-3 proteins, present in larger amounts due to p53 transcription, inactivate the Cdc25c phosphatase and promote its export from the nucleus and sequestration in 21 the cytoplasm. The inactivation of Cdc25c results in the inhibition of the cyclin B/Cdc2 kinase complex responsible for the G 2 / M cell cycle transition, leading to G2 arrest.8 2 C h k l and Chk2 also phosphorylate and activate W e e l , a kinase that catalyses Cdc2 inhibitory phosphorylation. 8 3 , 8 4 Since it is necessary for Cdc2 to be dephosphorylated for entry into mitosis, this provides a second mechanism by which the ATM-dependent Chk proteins arrest cells in G2 as a result of D N A damage. 1.3.5. The G 2 Checkpoint Inhibition Bioassay85,86 In response to D N A damage, cells can activate biochemical control pathways called checkpoints to halt progression of the cell cycle or to trigger cell death. When these tightly regulated responses break down, D N A replication and cell division may occur regardless of the cellular damage and the resulting genetic changes may lead to malignancy. However, this failure of cell cycle arrest responses in malignant or cancerous cells can also be exploited therapeutically. 8 7 Over half of all human cancers completely lack a G l checkpoint as a result of a mutation in the p53 tumour suppressor gene. This fact has important clinical implications since agents that disrupt the G2 cell cycle checkpoint are found to selectively potentiate the cytotoxic effects of DNA-damaging agents on cancer cells that lack p53_ 4 2 - 8 5 ' 8 9 - 9 0 ' 9 1 j n combination with D N A -damaging treatments, inhibition of the G2 checkpoint in tumour cells lacking p53 function would facilitate mitotic entry and lead to cell death (see Figure 1.7a). Normal cells exposed to D N A damage and a G2 checkpoint inhibitor would still arrest at the functioning G l checkpoint to repair damaged D N A (Figure 1.7b), leading to an increased survival over mutant p53- tumour cells. For this reason, components of the G2 checkpoint are particularly attractive targets for novel antineoplastics, and their inhibition may increase the sensitivity of tumour cells to standard chemotherapy and radiation. " 22 A. p53- tumour cells Gl checkpoint mutated y G l G2 checkpoint inhibited, G2 M B. p53+ normal cells Gl checkpoint Cell Death G2 checkpoint inhibited p \ | G l s G2 M Cells Arrested Cell Death Figure 1.7. Rationale for the use of G2 checkpoint inhibitors in cancer therapy. A. In response to D N A damaging agents, tumour cells will not arrest at cell checkpoints, leading to cell death. B. Under the same conditions, normal cells will arrest in G l and repair their DNA. The few G2 checkpoint inhibitors identified to date include caffeine (23) and 1-substituted caffeine analogues, 2-aminopurine and 6-dimethylaminopurine, staurosporine and 7-hydroxystaurosporine (UCN-01 , 24), SB-218078, and isogranulatimide (25).86 Several of these compounds, including caffeine (23), U C N - O l (24), and isogranulatimide (25), have shown the ability to enhance the cytotoxicity of DNA-damaging agents. 8 5 ' 9 3 Caffeine and its analogues demonstrate in vitro inhibition of the A T M and A T R kinases, 6 2 ' 9 4 however their numerous pharmacological effects prevent clinical use. 9 5 ' 9 6 Although UCN-01 displays broad protein kinase activity, it is currently in phase I clinical trials and has been shown to increase the cytotoxic effect of DNA-damaging agents in breast cancer ce l l s . 8 9 ' 9 7 Because this is a new 23 approach to anticancer treatment, there is a need for a rational screen to identify better G2 checkpoint inhibitors that possess greater specificity and reduced unwanted cellular effects. 23 24 25 In response to the need for an assay for G2 checkpoint inhibition, a cell-based screen was developed in the laboratory of Michel Roberge and used to screen thousands of natural extracts from marine and terrestrial sources (see Figure 1.8). 8 5 8 6 In this assay, human breast cancer cells lacking p53 function ( M C F - 7 p53-) were cultured as monolayers, seeded on 96-well microtitre plates, and allowed to grow for 24 hours. The cells were then exposed to D N A damage in the form of 6.5 G y of y-irradiation. Pure compounds or crude natural extracts were then added to the cells 16 hours after irradiation along with the antimitotic agent, nocodazole. At this point, cells in the presence of a G2 checkpoint inhibitor w i l l not arrest in G2 as a result of the induced D N A damage but wi l l enter and arrest in mitosis. The cells, extracts, and nocodazole were incubated together for 8 hours after which time the cell culture medium was carefully withdrawn from the plate without removing any cells. The remaining steps of the bioassay involve detection of the presence of phosphorylated nucleolin using the TG-3 antibody and HRP-conjugated secondary antibody (as was seen earlier in section 1.3.3). A positive signal from this E L I S A protocol would indicate a G2 cell cycle checkpoint inhibitor was present in the extract, thus allowing the cells to overcome the barrier to entering mitosis. 24 Human breast carcinoma M C F - 7 p53- cells seeded on 96-well plates incubate for 24 hours DNA damage 6.5 Gy of y-irradiation incubate for 16 hours Extracts and Nocodazole added G2 checkpoint inhibitor present cells enter mitosis and arrested nucleolin phosphorylated incubate for 8 hours No G2 checkpoint inhibitor present cells arrested in G2 nucleolin un-phosphorylated cell culture medium removed cells lysed and transferred to ELISA plates TG-3 antibody added incubate for 16-20 hours HRP-conjugated antibody added Complex of two antibodies and nucleolin-P t No complex incubate overnight rinse and add hydrogen peroxide POSITIVE S I G N A L G2 C H E C K P O I N T INHIBITOR P R E S E N T IN E X T R A C T N E G A T I V E S I G N A L N O G2 C H E C K P O I N T INHIBITOR P R E S E N T IN E X T R A C T Figure 1.8. The G2 cell cycle checkpoint inhibitor bioassay. 1.3.6. Biochemical Tools and Chemical Genetics Not only can G2 checkpoint inhibitors provide potential chemotherapeutic benefits, they are also invaluable as biochemical tools to investigate the machinery of cell cycle checkpoints. 25 Since most of our understanding of cell cycle checkpoint pathways has come from genetic studies on yeast signaling pathways, finding agents that prevent the G2 arrest in mammalian cells would provide useful information on their components and gene targets. The use of natural products and natural product-like compounds to decipher and control the cellular and 98 physiological roles of proteins is referred to as "chemical genetics". The traditional genetic approach for determining the cellular function of a protein requires indirectly altering its function by an inactivating or activating genetic mutation. This method is extremely powerful for identifying the roles of genes that encode proteins of interest. However, limitations arise in using the classical genetic approach in mammals because of their slow reproductive rates, large physical size, and more complex genome." In addition, indirect genetic mutations do not allow for the instantaneous or conditional alteration of protein function and investigations into essential genes render organisms nonviable. 1 0 0 A complementary approach to classical genetics involves the use of exogenous ligands to alter the function of a single gene product or protein. The chemical genetics approach involves small molecules capable of either inactivating or activating proteins directly. For instance, the first antimitotic agent to be characterized, colchicine, inactivates the protein function of tubulin whereas steroid hormones, such as dexamethasone, activate transcription in nuclear hormone receptors (see Figure 1.9).98 Thus, this approach emulates the logic of a classical genetic screen but differs in that it relies upon small molecules, rather than mutations, to conditionally modulate biological processes.' 0 1 This direct method allows for "genetic-like" screens in cells and tissues of higher organisms and complements other indirect genetic screening approaches. 1 0 0 26 inactivating mutation in gene C activating mutation in gene C \ genetic / cell division, differentiation, chemical . ^ cell death, etc. OH genetic X f Y W o o , a P P ™ a c h A= :0 colchicine Q x ^ - ^ dexamethasone Figure 1.9. The relationship between genetics and chemical genetics in exploring protein function. 9 8 One of the fundamental requirements of chemical genetics is the existence of a diverse library of ligands to bind very specific molecular targets within an organism. The chemical diversity necessary for this approach relies, at present, on compounds primarily from natural sources since most of the large combinatorial libraries synthesized so far contain simple 102 compounds with few or no stereocenters. Indeed, compounds rivaling the structural complexity, high potency, and target specificity of natural products are not available using existing combinatorial synthetic methods. 1 0 0 To date, biological screens of natural and synthetic libraries have identified small molecule ligands for many proteins (including kinases, 1 0 3 phosphatases, 1 0 4 proteases, 1 0 5 and various receptors 1 0 - l 0 1 ' m ) and agents that act on cellular processes (such as blocking mitotic progression, 1 0 1 ' 1 0 9 inducing or suppressing cell-cycle arrest, 1 1 0 ' 8 5 and inducing apoptosis 1"). Small molecule-based studies of the cell cycle and cell cycle checkpoints have been used to gain new insights into the circuitry of cell cycle progression, including the effects of discodermolide," 2 lactacystin," 3 and rapamycin" 4 (see Figure 1.10). Biologically active small molecules identified by rational screens may thus be 27 used as molecular probes to elucidate cell cycle pathways and protein functions, such as for the D N A damage checkpoint and its mediator c h k l . 1 1 5 ' 7 3 Figure 1.10. Small molecule-based studies of the cell cycle and cell cycle checkpoints. The checkpoint inhibitor screen developed by Roberge et al. provided the first rational assay for natural products capable of overcoming the G2 checkpoint. These natural agents can be investigated as potentiators of D N A damage or used as molecular probes to further elucidate the G2 cell cycle checkpoint. In total, over 30,000 extracts of marine microorganisms, marine invertebrates, and terrestrial plants were screened for checkpoint inhibitors. 1 1 6 The bioassay-guided investigation of two of these active extracts, originating from a marine sponge and a Taiwanese tree, is the subject of Chapters Three and Four of this thesis, respectively. 1.4. Research Summary A series of new and known secondary metabolites were isolated from marine and terrestrial sources guided by two newly developed, cell-based assays. The crude extract from a Caribbean octocoral Erythropodium caribaeorum caused cells to arrest in mitosis and to display morphological characteristics similar to the microtubule-stabilizing compound, paclitaxel (4). 28 Bioassay-guided fractionation resulted in the isolation and identification of the known antimitotic agent eleutherobin and six novel structural analogues. E. caribaeorum proved to be a new and abundant source of eleutherobin, whose pre-clinical development had been impeded by its scarcity, and the structural variations of the new diterpenoids offered key insights into proposed pharmacophore models for microtubule-stabilizing compounds. In addition, single crystal X-ray diffraction analysis and N O E difference experiments provided the first solid state and solution conformations for eleutherobin. These results advance the current understanding of this structural class of marine diterpenoids and may facilitate the development of new models for microtubule stabilization. The crude extract from a marine sponge, Stylissa flabelliformis, exhibited potent activity in a new bioassay searching for G 2 cell cycle checkpoint inhibitors. Bioassay-guided fractionation of this active extract resulted in the isolation and identification of the natural product debromohymenialdisine and three related alkaloids. These compounds were the first G2 checkpoint inhibitors to be found by a rational screen and were structurally distinct from previously reported G2 checkpoint inhibitors. In addition, debromohymenialdisine was found to specifically act on the protein kinases C h k l and Chk2, thus blocking two branches of the checkpoint pathway downstream of A T M . The discovery of these biologically active natural products validated the power of the new bioassay and contributed to the understanding of the G2 checkpoint pathway by providing new biochemical tools to probe the molecular basis of inhibition. As a result of a large-scale screen of natural extracts for G2 checkpoint inhibitors, a series of a-pyrones from the Taiwanese tree Cryptocarya concinna were found to exhibit potent inhibitory activity. Bioassay-guided purification resulted in the isolation and identification of the natural product cryptofolione and three related polyketide lactones. Synthetic modifications on these metabolites yielded four additional analogues and led to mode of action and structure-29 activity relationship studies. The G2 checkpoint inhibitors isolated from C. concinna were structurally distinct from previously known inhibitors and appear to act by a novel mechanism of action. Continuing studies are underway to determine the cellular target and structural requirements of inhibition. REFERENCES 1 Newman, D.J . ; Cragg, G . M . ; Snader, K . M . Nat. Prod. Rep. 2000, 77, 215. 2 Greenlee, R.T.; Murray, T.; Bolden, S.; Wingo, P .A . CA Cancer J. Clin. 2000, 50, 7. 3 Schreiber, S.L. Science 2000, 287, 1964. 4 Andersen, R.J. ; Wil l iams, D . E . Pharmaceuticals from the Sea. In Chemistry in the Marine Environment; Hester, R .E . ; Harrison, R . M . , Eds.; Issues in Environmental Science and Technology, The Royal Society of Chemistry: Cambridge, 2000; Volume 13, pp. 55-79. 5 Avramis, V . I . In A Clinician's Guide to Chemotherapy, Pharmacokinetics and Pharmacodynamics; Growchow, L . B . ; Aines, M . M . , Eds,; Will iams and Wilkins: Baltimore, 1998; p. 209. 6 Grabley, S.; Thiericke, R. In Drug Discoveij from Nature; Grabley, S.; Thiericke, R., Eds.; Springer-Verlag: New York, 1999; pp. 3-39. 7 Fleming, A . Brit. J. Exp. Pathol. 1929,10, 226. 8 a) Wani, M . C . ; Taylor, H . L . ; Wal l , M . E . ; Coggin, P.; McPhai l , A . T . J. Am. Chem. Soc. 1971, 93, 2325. b) Cragg, G . M . ; Schepartz, S.A.; Suffness, M . ; Grever, M . R . J. Nat. Prod. 1993, 56, 1657. 9 Dewick, P . M . In Medicinal Natural Products: A Biosynthetic Approach; John Wiley & Sons: Toronto, 1997; pp. 178-179. 1 0 a) Schmid, G . ; Hofheinz, W . J. Am. Chem. Soc. 1983,105, 624. b) Klayman, D . L . Science, 1985, 228, 1049. c) Hien, T.; Shen, C C . Lancet 1993, 341, 603. 1 1 Faulkner, D.J. Nat. Prod. Rep. 2000, 77, 1. 1 2 a) Maia, L .F . ; Epifanio, R . A . ; Eve, T.; Fenical, W . J. Nat. Prod. 1999, 62, 1322. b) Vervoort, H .C . ; Richards-Gross, S.E.; Fenical, W. ; Lee, A . Y . ; Clardy, J. J. Org. Chem. 1997, 62, 1486. 30 l i Tsukamoto, S.; Kato, H . ; Hirota, H . ; Fusetani, N . J. Nat. Prod. 1997, 60, 126. 1 4 Osterhage, C ; Kaminsky, R.; Konig, G . M . ; Wright, A . D . J. Org. Chem. 2000, 65, 6412. 1 3 Pawlik, J.R. Induction of Marine Invertebrate Larval Settlement: Evidence for Chemical Cues. In Ecological Roles of Marine Natural Products; Paul, V . J . , Ed. ; Cornell University Press: New York, 1992; pp. 189-236. 1 6 L in , Y . - Y . ; Risk, M . ; Ray, S . M . ; VanEngen, D . ; Clardy, J.; Golik, J.; James, J . C ; Nakanishi, K . J. Am. Chem. Soc. 1981, 103, 6113. 1 7 Pol i , M . A . ; Mende, T.J.; Baden, D . G . Mol. Pharmacol. 1986, 30, 129. 1 8 a) Murata, M . ; Iwashita, T.; Yokoyama, A . ; Sasaki, M . ; Yasumoto, T. / . Am. Chem. Soc. 1992, 114, 1975. b) Murata, M . ; Nakoi, PL; Iwashita, T.; Matsunaga, S.; Sasaki, M . ; Yokoyama, A . ; Yasumoto, T. J. Am. Chem. Soc. 1993, 775, 2060. c) Nonomura, T.; Sasaki, M . ; Matsumori, N . ; Miata, M . ; Tachibana, K . ; Yasumoto, T. Angew. Chem., Int. Ed. Engl. 1996, 35, 1675. 1 9 a) Bergmann, W. ; Feeney, R.J . / . Am. Chem. Soc. 1950, 72, 2809. b) Bergmann, W. ; Feeney, R.J. J. Org. Chem. 1951, 16, 981. c) Bergmann, W. ; Burke, D . C J. Org. Chem. 1951, 16, 981. 2 0 Cimino, G . ; De Rosa, S.; De Stefano, S. Experientia 1984, 40, 339. 2 1 McConnel l , O.J.; Longley, R .E . ; Koehn, F . E . In The Discovery of Natural Products with Therapeutic Potential; Gullo, V . P . , Ed. ; Butterworth-Heinemann: Boston, 1994; p. 109. 2 2 a) Pettit, G.R.; Herald, C . L . ; Doubek, D . L . ; Herald, D . L . ; Arnold, E . ; Clardy, J. / . Am. Chem. Soc. 1982, 104, 6846. b) Pettit, G.R. 1. Nat Prod. 1996, 59, 812; and references cited therein. 2 3 Schaufelberger, D .E . ; Koleck, M . P . ; Beutler, J .A. ; Vatakis, A : M . ; Alvarado, A . B . ; Andrews, P.; Marzo. L . V . ; Muschik, G . M . I. Nat. Prod. 1991, 54, 1265. 2 4 a) Haywood, M . G . ; Davidson, S.K. Appl. Env. Microbiol. 1997, 63, 4612. b) Haywood, M . G . ; Schmidt, E .W. ; Davidson, S.K.; Faulkner, D.J . J. Mol. Microbiol. Biotechnol. 1999, 7,33. 2 5 Wright, A . E . ; Forleo, D . A . ; Gunawardana, G.P.; Gunasekera, S.P.; Koehn, F .E. ; McConnell , O.J. 1. Org. Chem. 1990, 55, 4508. 2 6 Rinehart, K . L . ; Holt, T .G . ; Fregeau, N . L . ; Stroh, J.G.; Kiefer, P .A. ; Sun, F.; L i , L . H . ; Martin, D . G . J. Org. Chem. 1990, 55, 4512. 2 7 Shu, Y . - Z . J. Nat. Prod. 1998, 61, 1053; and references cited therein. 2 8 a) Arai , T.; Takahashi, K . ; Nakahara, S.; Kubo, A . Experientia 1980, 36, 1025. b) Arai , T.; Takahashi, K . ; Kubo, A . ; Nakahara, S.; Sato, S.; Aiba, K . ; Tamura, C. Tetrahedron Lett. 1979, 2355. 2 9 Pettit, G.R.; Kamano, Y . ; Herald, C . L . ; Tuinman, A . A . ; Boettner, F .E . ; K i z u , H . ; Schmidt, J . M . ; Baczynskyj, L . ; Tomer, K . B . ; Bontems, R .J . J. Am. Chem. Soc. 1987,109, 6883. 3 0 Gunasekera, S.P.; Gunasekera, M . ; Longley, R .E . ; Schulte, G . K . J. Org. Chem. 1990, 55, 4912 (correction: 1991, 56, 1346). 3 1 a) ter Haar, E . ; Kowalski , R.J . ; Hamel, E . ; L i n , C . M . ; Longley, R . E . ; Gunasekera, S.P.; Rosenkranz, H.S. ; Day, B . W . Biochemistry 1996, 35, 243. b) Hung, G.T. ; Chen, J.; Schreiber, S.L. Chem. Biol. 1996, 3, 287. 3 2 Kowalski , R.J. ; Giannakakou, P.; Gunasekera, S.P.; Longley, R .E . ; Day, B . W . ; Hamel, E. Mol. Pharm. 1997, 52, 613. 3 3 a) Gunasekera, S.P.; McCarthy, P.J.; Kelly-Borges, M . ; Lobkovsky, E. ; Clardy, J. J. Am. Chem. Soc. 1996, 118, 8759. b) Blanchard, J .L.; Epstein, D . M . ; Boisclair, M . D . ; Rudolph, J.; Pal, K . Bioorg. Med. Chem. Lett. 1999, 9, 2537. c) Takahashi, M . ; Dodo, K . ; Sugimoto, Y . ; Aoyagi, Y . ; Yamada, Y . ; Hashimoto, Y . ; Shirai, R. Bioorg. Med. Chem. Lett. 2000, 10, 2571. 3 4 Look, S.A.; Fenical, W. ; Matsumoto, G . K . ; Clardy, J. J. Org. Chem. 1986, 51, 5141. 3 5 Roussis, V . ; W u , Z. ; Fenical, W. ; Strobel, S.A.; Van Duyne, G .D. ; Clardy, J. J. Org. Chem. 1990, 55, 4916. 3 6 a) Burgoyne, D . L . ; Andersen, R.J. ; Al len , T . M . / . Org. Chem. 1992, 57, 525. b) Takei, M . ; Burgoyne, D . L . ; Andersen, R.J . J. Pharm. Sci. 1994, 83, 1234. 3 7 Faulkner, D .J . Nat. Prod. Rep. 2001, 18, 1; and previous reports in this series. 3 8 Cragg, G . M . ; Newman, D.J . ; Snader, K . M . J. Nat. Prod. 1997, 60, 52. 3 9 Artuso, A . In Drugs of Natural Origin: Economic and Policy Aspects of Discovery, Development, and Marketing; The Haworth Press: Binghamton, 1997; p. 10. 4 0 Gibbs, J.B.; Oliff, A . Cell 1994, 79, 193. 4 1 Levitski, A . ; Gazit, A . Science 1995, 267, 1782. 4 2 Nurse, P. Cell 1997, 91, 865. 4 3 Omer, C . A . ; Koh l , N . E . TIPS 1997,18, 437. 4 4 Starr, C ; Taggart, R. In Biology: The unity and diversity of life; Fourth edition; Wadsworth Publishing: Belmont, 1987; pp. 140-146. 4 5 Karp, G. In Cell and Molecular Biology: Concepts and Experiments; Second edition; John 32 Wiley & Sons: Toronto, 1999; pp 608-623. 4 6 In The molecular basis of cell cycle and growth control; Stein, G.S.; Baserga, R.; Giordano, A . ; Denhardt, D.T. , Eds.; Wiley-Liss: Toronto, 1999. 4 7 Hamel, E . Med. Res. Rev. 1996,16, 207; and references cited therein. AO Rowinsky, E . K . ; Donehower, R . C . Antimicrotubule agents. In Cancer chemotherapy and biotherapy; Chabner, B . A . ; Longo, D . L . , Eds.; Lippincott-Raven Publishers: 49 Philadelphia, 1996; pp. 263-296. Rowinsky, E . K . Anna. Rev. Med. 1997, 48, 353. 5 0 Jordan, M . A . ; Thrower, D . ; Wilson, L . J. Cell Sci. 1992,102, 401. :, 5 1 Jordan, M . A . ; Toso, R.J . ; Thrower, D . ; Wilson, L . Proc. Natl. Acad. Sci. USA 1993, 90, 9552. 5 2 Jordan, A . ; Hadfield, J .A. ; Lawrence, N . J . ; McGowan, A T . Med. Res. Rev. 1998,18, 259. 53 Roberge, M . ; Cinel , B . ; Anderson, H.J . ; L i m , L . ; Jiang, X . ; X u , L . ; Bigg, C . M . ; Kel ly, M.T . ; Andersen, R.J . Cancer Res. 2000, 60, 5052. 5 4 Hertzberg, R.P. Curr. Opin. Biotechnol. 1993, 4, 80. 5 5 Corbett, T.; Valeriote, F.; LoRusso, P.; Polin, L . ; Panchapor, C ; Pugh, S.; White, K . ; Knight, J.; Demchik, L . ; Jones, J.; Jones, L . ; Lis low, L . In vivo methods for screening and preclinical testing. In Drug Development Guide; Teicher, B . , Ed.; Humana Press: Totowa, NJ , 1997; pp. 75-99. 5 6 Vincent, I.; Rosado, M . ; Davies, P. J. Cell Biol. 1996, 132, 413. 5 7 Anderson, H.J . ; deJong, G. ; Vincent, I.; Roberge, M . Exp. Cell Res. 1996, 238, 498. 5 8 Dranovsky, A . ; Gregori, L . ; Schwartzman, A . ; Vincent, I.; Davies, P.; Goldgaber, D. Soc. Neurosci. Abstracts 1997, 27, 2219. 5 9 Perlmann, H . ; Perlmann, P. Enzyme-linked immunosorbent assay. In Cell biology-a practical approach; Celis, J.E., Ed. ; Academic Press: San Diego, 1994; Volume 2, pp. 322-328. 60 Lowndes, N .F . ; Murguia, J.R. Curr. Opin. Genet. Dev. 2000, 10, 17. 6 1 Clarke, D.J . ; Gimenez-Albian, J.F. BioEssays 2000, 22, 351. 6 2 Zhou, B . -B .S . ; Elledge, S.J. Nature 2000, 408, 433. 6 3 Carr, A . M . Science 2000, 287, 1765. 6 4 Savitsky, K . ; Bar-Shira, A . ; Gilad, S.; Rotman, G . ; Z iv , Y . ; Vanagaite, L . ; Tagle, D . A . ; 33 Smith, S.; Uzie l , T.; Sfez, S.; Ashkenazi, M . ; Pecker, I.; Frydman, M . ; Harnik, R.; Patanjali, S.R.; Simmons, A . ; Clines, G . A . ; Sartiel, A . ; Gatti, R . A . ; Chessa, L . ; Sanal, O.; Lavin, M . F . ; Jaspers, N . G . J . ; Malcolm, A . ; Taylor, R.; Arlett, C.F. ; M i k i , T.; Weissman, S . M . ; Lovett, M . ; Collins, F.S. ; Shiloh, Y . Science, 1995, 268, 1749. 6 5 K o , L . J . ; Prives, C. Genes Dev. 1996, 10, 1054. 6 6 a) Brugarolas, J.; Chandrasekaran, C ; Gordon, J.I.; Beach, D . ; Jacks, T.; Hannon, G.J . Nature 1995, 377, 552. b) Deng, C ; Zhang, P.; Harper, J.W.; Elledge, S.J.; Leder, P. Cell 1995, 82, 675. 6 7 Shieh, S. -Y. ; Dceda, M . ; Taya, Y . ; Prives, C. Cell 1997, 91, 325. 6 8 X u , Y . ; Baltimore, D . Genes Dev. 1996, 10, 2401. 6 9 Hirao, A . ; Kong, Y . - Y . ; Matsuoka, S.; Wakeham, A . ; Ruland, J.; Yoshida, H . ; L i u , D . ; Elledge, S.J.; Mak, T .W. Science 2000, 257, 1824. 7 0 Sherr, C.J , Science 1996, 274, 1672. 7 1 Levine, A . J . Cell, 1997, 88, 323; and references cited therein. 7 2 Canman, C . E . ; L i m , D.S. Oncogene 1998, 17, 3301. 7 3 Cl iby, W . A . ; Roberts, C.J . ; Cimprich, K . A . ; Stringer, C M . ; Lamb, J.R.; Schreiber, S.L.; Friend, S.H. EMBO J. 1998,17, 159. 7 4 Wright, J .A. ; Keegan, K . S . ; Herendeen, D.R. ; Bentley, N.J . ; Carr, A . M . ; Hoekstra, M . F . ; Cancannon, P. Proc. Natl. Acad. Sci. USA 1998, 95, 7445. 7 5 Matsuoka, S.; Huang, M . ; Elledge, S.J. Science 1998, 282, 1893. 7 6 Sanchez, Y . ; Wong, C ; Thoma, R.S.; Richman, R.; W u , Z. ; Piwnica-Worms, H . ; Elledge, S.J. Science, 1997, 277, 1497. 7 7 Chaturverdi, P.; Eng, W . K . ; Zhu, Y . ; Mattern, M . R . ; Mishra, R.; Hurle, M . R . ; Zhang, X . ; Annan, R.S. ; L u , Q.; Faucette, L .F . ; Scott, G.F. ; L i , X . ; Carr, S.A.; Johnson, R . K . ; Winkler, J.D.; Zhou, B . B . Oncogene 1999,18, 4047. 7 8 Peng, C . - Y ; Graves, P.R.; Thoma, R.S.; W u , Z. ; Shaw, A.S . ; Piwnica-Worms, H . Science, 1997,277,1501. 7 9 Yang, J.; Winkler, K . ; Yoshida, M . ; Kornbluth, S. EMBO J. 1999, 75, 2174. 8 0 Dalai, S.N.; Schweitzer, C M . ; Gan, J.; DeCaprio, J. Mol. Cell. Biol. 1999,19, 4465. 8 1 Zeng, Y . ; Piwnica-Worms, H . Mol. Cell. Biol. 1999,19, 7410. 8 2 Lopez-Girona, A . ; Furnari, B . ; Mondesert, O.; Russell, P. Nature 1999, 397, 172. 34 O'Connell , M . J . ; Raleigh, J . M . ; Verkade, H .D. ; Nurse, P. EMBO J. 1997,16, 545. 8 4 Raleigh, J . M . ; O'Connell , M . J . J. Cell Sci. 2000, 113, 1727. 8 3 Roberge, M . ; Berlinck, R .G.S . ; X u , L . ; Anderson, H.J . ; L i m , L . Y . ; Curman, D. ; Stringer, C M . ; Friend, S.H.; Davies, P.; Haggarty, S.J.; Kel ly , M . T . ; Britton, R.; Piers, E . ; Andersen, R.J . Cancer Res. 1998, 58, 5701. 8 6 Curman, D . ; Cinel , B . ; Will iams, D .E . ; Rundle, N . ; Hutchins, J.; Clarke, P.; Goodarzi, A . A . ; Block, W . D . ; Lees-Miller, S.P.; Andersen, R.J . ; Roberge, M . J. Biol. Chem. 2001, 276, 17914; and references cited therein. 8 7 Shapiro, G.I.; Harper, J .W. J. Clin. Invest. 1999,104, 1645. 8 8 Kastan, M . B . ; Onyekwere, O.; Sidransky, D. ; Vogelstein, B . ; Craig, R . W . FASEB I. 1996, 10, 238. 8 9 Wang, Q.; Fan, S.; Eastman, A . ; Worland, P.J.; Sausville, E . A . ; O'Connor, P . M . 1. Natl. Cancer Inst. 1996, 88, 956. 9 0 Weinert, T.; Lydall , D . Seminars Cancer Biol. 1993, 4, 129. 9 1 Kao, G.D. ; McKenna, W . G . ; Maity, A . ; Blank, K . ; Muschel, R.J . Cancer Res. 1997, 57, 753. 9 2 Hartwell, L . H . ; Kastan, M . B . Science 1994, 266, 1821. 9 3 Jackson, J.R.; Gilmartin, A . ; Imburgia, C ; Winkler, J.D.; Marshall, L . A . ; Roshak, A . Cancer Res. 2000, 60, 566. 9 4 a) Blasina, A . ; Van de Weyer, I.; Laus, M . C . ; Luyten, W . H . M . L . ; Parker, A . E . ; McGovvan, C. Curr. Biol. 1999, 9,1. b) Sarkaria, J .N. ; Busby, E . C ; Tibbetts, R.S.; Roos, P.; Taya, Y . ; Karnitz, L . M . ; Abraham, R.T. Cancer Res. 1999, 59, 4375. c) Hall-Jackson, C . A . ; Cross, D. A . E . ; Morrice, N . ; Smythe, C. Oncogene, 1999, 75, 6707. 9 5 Arnaud, M . J . Progress in Drug Research 1987, 31, 273. 9 6 Jiang, X . ; L i m , L . Y . ; Daly, J.W.; L i . A . H . ; Jacobson, K . A . ; Roberge, M . Int. J. Oncol. 2000, 7f5, 971. 9 7 Sausville, E . A . ; Arbuck, S.G.; Messmann, R.; Headlee, D. ; Lush, R .D . ; Bauer, K . ; Murgo, A . ; Figg, W . D . ; Lahusen, T.; Jaken, S.; Roberge, M . ; Fuse, E . ; Kuwabara, T.; Senderowicz, A . M . J. Clin. Oncol. 2001, 79, 2319. 3 8 Schreiber, S.L. Bioorg. Med. Chem. 1998, 6, 1127. 5 9 Justice, M . J . ; Noveroske, J .K. ; Weber, J.S.; Zheng, B . ; Bradley, A . Hum. Mol. Genet. 1999, 8, 1955. 1 0 0 Stockwell, B .R . Nature Genetics 2000, 7, 116. 35 1 0 1 Haggarty, S.J.; Mayer, T . U . ; Miyamoto, D.T. ; Fathi, R.; K ing , R .W. ; Mitchison, T.J.; Schreiber, S.L. Chem. Biol. 2000, 7, 275. 1 0 2 Dolle, R . E . J. Comb. Chem. 1999, 7, 235. 1 0 3 Colas, P.; Cohen, B . ; Jessen, T.; Grishina, I.; M c C o y , J.; Brent, R. Nature 1996, 380, 548. 1 0 4 Aramburu, J.; Yaffe, M . B . ; Lopez-Rodriguez, C ; Cantley, L . C . ; Ffogan, P .G. ; Rao, A . Science 1999, 255,2129. 1 0 5 Apletalina, E . ; Appel , J.; Lamango, N.S . ; Houghten, R . A . ; Lindberg, I. J. Biol. Chem. 1998, 273, 26589. 1 0 6 Tian, S.-S.; Lamb, P.; King , A . G . ; Mi l le r , S.G.; Kessler, L . ; Luengo, J.I.; Aver i l l , L . ; Johnson, R . K . ; Gleason, J .G. ; Pelus, L . M . ; Di l lon , S.B.; Rosen, J. Science 1998, 281,251. 1 0 7 Zhang, B . ; Salituro, G . ; Szalkowski, D. ; L i , Z . ; Zhang, Y . ; Royo, I.; Vi le l la , D . ; Diez, M.T . ; Pelaez, F.; Ruby, C ; Kendall, R . L . ; Mao, X . ; Griffin, P.; Calaycay, J.; Zierath, J.R.; Heck, J .V.; Smith, R . G . ; Moller , D E . Science 1999, 284, 91A. 1 0 8 Norris, J .D.; Paige, L . A . ; Christensen, D.J . ; Chang, C . - Y . ; Huacani, M . R . ; Fan, D. ; Hamilton, P.T.; Fowlkes, D . M . ; McDonnel l , D.P. Science 1998, 255, 744. 1 0 9 a) Norman, T .C . ; Smith, D . L . ; Sorger, P . K . ; Drees, B . L . ; O'Rourke, S . M . ; Hughes, T.R.; Roberts, C.J . ; Friend, S.H.; Fields, S.; Murray, A . W . Science 1999, 255, 591. b) Mayer, T . U . ; Kapoor, T . M . ; Haggarty, S.J.; King , R .W. ; Schreiber, S.L.; Mitchison, T.J. Science 1999,256, 97'1. 1 1 0 a) Stockwell, B .R . ; Haggarty, S.J.; Schreiber, S.L. Chem, Biol. 1999, 6, 71. b) Stockwell, B .R . ; Hardwick, J.S.; Tong, J .K. ; Schreiber, S.L. J. Am. Chem. Soc. 1999,121, 10662. 1 1 1 Lu , X . P . Nature Med. 1997, 3, 686. 1 1 2 a) Nerenberg, J .B.; Hung, D.T. ; Somers, P .K . ; Schreiber, S.L. J. Am. Chem. Soc. 1993,115, 12621. b) Hung, D.T.; Nerenberg, J .B.; Schreiber, S.L. Chem. Biol. 1994,1, 67. c) Hung, D.T. ; Chen, J.; Schreiber, S.L. Chem. Biol. 1996, 3, 287. d) Hung, D.T.; Nerenberg, J.B.; Schreiber, S.L. J. Am. Chem. Soc. 1996, 775, 11054. 1 1 3 a) Fenteany, G . ; Standaert, R.F. ; Lane, W.S. ; Choi , S.; Corey, E.J . ; Schreiber, S.L.; Science 1995, 2r55, 726. b) Fenteany, G. ; Schreiber, S.L. J. Bio. Chem. 1998, 273, 8545. 1 1 4 Brown, E.J . ; Schreiber, S.L. Cell 1996, 86, 517. 1 1 5 a) Cimprich, K . A . ; Shin, T .B . ; Keith, C.T. ; Schreiber, S.L. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 2850. b) Keith, C.T. ; Schreiber, S.L. Science 1995, 270, 50. 1 1 6 Roberge, M . , personal communication. 36 CHAPTER 2 2.1. Antimitotic Diterpenoids from the Gorgonian Erythropodium caribaeorum In searching for biologically active natural products from marine sources, the crude extract from a Caribbean octocoral, Erythropodium caribaeorum, was found to contain a series of new diterpenoids. Specimens of E. caribaeorum were collected off the coast of Dominica and the methanolic crude extract exhibited promising activity in the antimitotic bioassay. In addition, cells arrested in mitosis by this extract displayed morphological characteristics similar to the microtubule bundling seen with paclitaxel. Bioassay-guided fractionation of the active extracts yielded six new antimitotic diterpenoids and the known antimitotic compound, eleutherobin. As well as identifying an important new source for eleutherobin, this discovery led to structure-activity relationships offering key insights into pharmacophore models for microtubule-stabilization and the first solid-state and solution conformational studies on this class of promising anticancer agents. 2.1.1. Introduction to Coelenterates and Gorgonians The animal phylum Coelenterata, also known as Cnidaria, represents a diverse group of marine invertebrates that includes hydras, jellyfish, sea anemones, and corals. With more than 10,000 known species, the coelenterates are arguably the most common and conspicuous invertebrates present in shallow tropical waters, reaching levels of diversity and importance unequalled by any other phylum. 1 Coral reefs are responsible for some of our earth's most fascinating, complex, and varied ecosystems by hosting an incredible multitude of animal and plant life. Consisting of four main classes, the coelenterates all exhibit radial symmetry, a single opening (coelenteron) for ingesting and expelling food, and tentacles armed with stinging structures (nematocysts) for capturing prey and defense. Two basic body types are found in 37 coelenterates, a sessile polyp form that lives attached to a substrate and an unattached, free-swimming medusa form. While most members of the Hydrozoa, Scyphozoa, and Cubozoa classes display alternate polyp and medusa stages, the remaining Anthozoa class exists only as solitary polyps or colonial polyps. Chemically, the Coelenterate phylum is second only to the sponges as the most prolific marine source for natural products reported in the recent literature (as seen in Figure 1.1), with compounds arising almost exclusively from the class Anthozoa (corals, sea fans, sea anemones, and zoanthids). 2 The anthozoans represent the largest and most diverse class within the Coelenterata and can conveniently be divided into two subclasses depending on their polyp morphology. The Hexacorallia usually display six unbranched tentacles or multiples thereof, whereas the Octocorallia always exhibit eight tentacles that bear tiny pinnate (feather-like) projections or side branches. The colonial octocorals, in turn, can be divided into four primary groups: the stoloniferans, alcyonaceans (soft corals), penatulaceans (sea pens), and gorgonians. Easily recognized as "sea fans" or "sea whips", the gorgonians are prominent and eye-catching members of the reef fauna of the West Indies and represent an estimated 38% of the known fauna. However, they are often mistaken for plants because of their permanent attachment to a substrate, bushy colony form, and passive mode of existence (their branches move only because of currents or wave action). 4 Many shallow water species of gorgonians and octocorals also play host to scores of symbiotic unicellular algae, or zooxanthellae, present in their surface tissues and responsible for the relatively dull brownish appearance of the host organism. 5 Like most octocorals, gorgonians are filter feeders that inhabit areas where currents flow and are represented by about 1 7 0 species in the Caribbean region. While most of these are restricted to depths greater than 5 0 meters or are not found in reef habitats, the remaining shallow water gorgonians are among the most common and abundant species.4 The morphological differences in the arrangement and shape of the polyps along with the branching 38 of the stem anchoring the polyp to the substrate can often be used to distinguish the genus and occasionally species of gorgonian. However, positive identification requires the microscopic examination of the location, pattern, shape, and size of the skeletal spicules embedded in the polyp's and colony's common tissue.6 The calcareous spicules are very small structures (0.05-1.00 mm) of various shapes that form the internal skeleton of gorgonians and may also function to deter predators. In addition, the rich chemistry often associated with these organisms is 5 7 thought to contain toxic compounds that play a key role as feeding deterrents. ' 2.1.2. Review of Diterpenoids from Octocorals The natural products isolated from coelenterates and, specifically, octocorals is dominated by the diterpenoid class of compounds. 2 Biosynthetically, the large and structurally diverse family of diterpenoids arises from the linking of four isoprene (C5) units in a head-to-tail fashion to form a 20-carbon structural framework. The assembly of the diterpene skeleton begins with the creation of two biochemical isoprene units, dimethylallyl pyrophosphate ( D M A P P ) and isopentenyl pyrophosphate (IPP), from acetate precursors through a mevalonic acid intermediate. The enzymatic formation of geranyl pyrophosphate from D M A P P and IPP is thought to involve the generation of an allylic cation, electrophilic addition yielding a tertiary cation, and the subsequent stereospecific loss of a proton, HR (see Scheme 2.1). 8 This process is repeated two more times, ultimately forming the C20 linear geranylgeranyl pyrophosphate (GGPP). a) formation ol allylic cation b) electrophilic addition giving tertiary cation "OPP OPP DMAPP IPP a) allylic cation b) electrophilic addition OPP geranyl pyrophosphate (GPP) OPP IPP a) allylic cation b) electrophilic addition OPP farnesyl pyrophosphate (FPP) OPP IPP c) stereospecilic loss of proton C\ OPP H R H S c) stereospecilic loss of proton c) stereospecilic loss of proton OPP geranylgeranyl pyrophosphate (GGPP) Scheme 2.1. Formation of geranylgeranyl pyrophosphate. 39 The linear G G P P can undergo a variety of oxidation, reduction, and cyclization reactions to produce the multitude of diterpenoids isolated from octocorals. In a recent review of the secondary metabolites reported from Caribbean gorgonians, diterpenoids were found to account for close to 65% of the compounds isolated over a 37 year period, with one third of these belonging to the cembranoid structural class. 9 The formation of the cembranoid diterpene framework involves a cyclization between carbons 14 and 1 of G G P P (as shown in Scheme 2.2). One of the simplest members of this class is cembrene A (26), isolated from marine soft corals and a variety of terrestrial sources. 1 0" 1 2 The carbon skeletons of two related classes of bicyclic diterpenes, the briaranes and the eunicellanes (or cladiellanes), are thought to arise from additional cyclizations of the cembrane ring system, as shown in Scheme 2.2. 1 3 Briarein A (27) 40 and eunicellin (28) were the first known members of these diterpene structural classes and were isolated from the gorgonians Briareum abestinum14 and Eunicella stricta,^ respectively. 28 Scheme 2.2. Proposed biosynthetic relationship between cembranoid diterpenes. In total, over twelve different structural classes of diterpenes have been reported from gorgonians (see Figure 2.1), with each class displaying an impressive number of members possessing unique substitution patterns and functionalities. In addition to the cembranes, briaranes, and eunicellanes, the most common carbon skeletons include the asbestinins (related to the eunicellins by a 1,2-methyl migration), 1 3 the pseudopteranes and cubitanes (both rearranged, 12-membered, monocyclic rings containing two isopropyl groups), the related dolabellane/dolastanes, and dilophol/elemene classes.9 Gorgonians of the genus 41 Pseudopterogorgia have proven to be especially prolific in terms of diterpene structural classes, producing secondary metabolites from the amphilectane, 1 6 serrulatane,1 7 elisabethane,18 elisapterane,1 9 and gersolane 2 0 structural classes, just to name a few. asbestinin pseudopterane cubitane dolabellane dolastane dilophol elemene (fuscol) amphilectane serrulatane elisabethane elisapterane gersolane Figure 2.1. Structural classes of gorgonian diterpenoids. There has been considerable interest in the secondary metabolites of gorgonians because of the many biological and pharmaceutical activities associated with these compounds. Investigations of Briareum excavatum have led to the isolation of 40 briarane diterpenoids, the 21 22 excavatolides A - Z and briaexcavatolides A - N , ' some of which exhibit significant cytotoxic activity against a variety of cancer cell lines (i.e., excavatolide M (29)). Pinnatin A (30), a polycyclic diterpenoid lactone belonging to the rare gersolane class," was recently found to show significant differential antitumour activity in the National Cancer Institute's 60 cell-line panel. 2 0 In addition to cytotoxicity, anti-inflammatory properties have been reported from 42 numerous briarane diterpenes and diterpene glycosides isolated from gorgonians. 2 4 These diterpene glycosides include the amphilectane-containing pseudopterosins,2 5 such as 26 pseudopterosin A (31), the elemene-containing fuscosides, and the dilophol-containing calyculaglycoside B (32). 2 7 Furthermore, gorgonian secondary metabolites continue to attract 28 29 30 ^ I interest because of the antiviral, antibacterial, insecticidal, immunomodulatory, and antifouling 3 2 properties associated with some of these compounds. Numerous ecological 33 studies indicate the diterpenoids produced by gorgonians can act as potent feeding deterrents to reduce predation by fish (i.e., furanocembranolide 33). 3 4 O 32 33 While diterpenoids comprise the majority of secondary metabolites reported from gorgonians and octocorals, there are also frequent reports of novel sesquiterpenoids, acetogenins, prostanoids, and highly functionalized steroids isolated from these organisms.2 Although over forty years have passed since Burkholder and Burkholder highlighted the occurrence of antibiotic substances from a Caribbean gorgonian, 3 5 only a small percentage of the many species of gorgonians from this region have been systematically scrutinized to date.9 Even species that have been extensively studied in the past continue to excite researchers with 4 3 novel compounds and structures, often with new biological activities, that were previously overlooked. 2.1.3. Review of Known Metabolites from Erythropodium caribaeorum The gorgonian Erythropodium caribaeorum (Duchassaing & Michelotti, 1860) is endemic to the Caribbean region and is classified as "uncommon", typically found from the southern tip of Florida, east to the lesser Antilles, and into the western Caribbean. 3 Its genus is taxonomically monospecific and is described morphologically as an encrusting octocoral, with a tan upper surface colouration and reddish underside. The colonies are quite bushy but grow in flat, vertical planes with distinctive, contrasting light yellow-brown polyps and dark brown rind. 6 Colony sizes vary from 10 centimeters to 1 meter and usually inhabit depths easily accessible by S C U B A , from 1 to 30 meters. Figure 2.2 shows the typical appearance of this gorgonian with its fine, hair-like polyps extended and retracted. Figure 2.2. Erythropodium caribaeorum (photo by P. Humann). 6 44 The chemistry of E. caribaeorum has been extensively investigated by many research groups over the past 17 years. The first report dates back to 1968 when Schmidt and Ciereszko 36 isolated choline sulphate (34) and scyllitol (35) from this gorgonian coral. A further 16 years passed before Fenical et al. reported the structures of two new chlorinated diterpenoids. erythrolides A (36) and B (37), from a sample collected in the waters off Bel ize . 7 The authors suggest erythrolide A arises from a photochemical di-7r-methane rearrangement of erythrolide B , the first observation of such a rearrangement in a natural product. Later that same year, the Fenical and Djerassi groups published a study investigating whether the terpenes isolated from gorgonians are produced by the organisms themselves or by their associated symbiotic 37 zooxanthellae. Most octocorals inhabiting the Caribbean Sea host symbiotic algae known as zooxanthellae which provide a large portion of the nutrient requirements of the octocoral through photosynthesis and carbon transfer.7 Since the true origin of secondary metabolites isolated from gorgonians was in doubt, the authors used 1 3 C / 1 2 C isotope ratio mass spectrometry to compare compounds known to be of algal origin, such as the sterol gorgosterol (38),38 and the newly identified diterpenoids, the erythrolides. Their data suggests the sterols isolated from E. caribaeorum are produced by the zooxanthellae, a result consistent with data from other gorgonians, whereas the diterpenes are true gorgonian metabolites. OH 36 37 38 45 A few years later, erythrolides A and B along with seven additional erythrolides, C-I (39-45), were isolated from E. caribaeorum collected in the U .S . Virgin Islands and Jamaica/ 9 A l l of these erythrolides, with the exception of erythrolide H (44), are chlorinated at the C-6 position. While erythrolides C (39), D (40), and H (44) all possess epoxide functionalities, the remaining diterpenes (41-43, 45) comprised a new group of ether-cyclized briaranes featuring an unprecedented C-2 to C-8 ether link. While in the process of conducting feeding-deterrent studies on the compounds from this gorgonian, Fenical and Pawlik reported the presence of a new spirobicyclic sesquiterpene, named erythrodiene (46).40 Results of the feeding studies showed, however, that the erythrolides, particularly erythrolides B (37) and D (40), were responsible for the defensive properties of the gorgonian towards a collection of predatory reef fishes. O 44 46 46 Subsequent studies of E. caribaeorum collected off the waters of Tobago and Jamaica led to the isolation of erythrolide J (47),41 erythrolide K (48),42 and the three acetate analogues of the erythrolides E (49), H (50), and I (51), respectively. 4 3 Unt i l our present work, the secondary metabolites isolated from the gorgonian Erythropodium caribaeorum had offered insight into their ecological roles as feeding deterrents and provided challenges to synthetic chemistry but were devoid of any pharmaceutical interest or potential. 4 4 49 R = A c 51 O 50 R= ^ ^ ^ ° Y ^ O 2.1.4. Review of Microtubule-Stabilizing Antimitotic Agents Compounds capable of binding to tubulin and stabilizing the microtubule assembly during cell division provide one important mechanism to halt the uncontrolled cellular proliferation associated with cancerous tumours. Arresting these cells in mitosis can eventually lead to apoptosis, or cell death, and presents a clinically important approach to the treatment of 47 slow-growing solid tumours usually occurring in adults, such as those targeting the lungs, colon, prostrate, and brain. Paclitaxel (4), originally isolated from the bark of the Pacific Yew tree, was the first chemical entity found to exhibit this unique mechanism of action. The subsequent clinical success of paclitaxel (4) and the related docetaxel (6) has generated considerable interest in identifying additional chemotypes that stabilize microtubules. Four additional structural classes of microtubule-stabilizing compounds that have been identified include the epothilones A and B , discodermolide, laulimalide, and the sarcodictyin/eleutherobin classes. 4 6 Epothilones A (52) and B (53) were initially isolated from the soil bacteria Sorangium cellulosum and were the first compounds identified after paclitaxel (4) to exhibit microtubule-stabilizing properties. 4 5 In contrast to paclitaxel, the epothilones can be easily produced in large quantities through bacterial fermentations and are equally active against multi-drug resistant human cancer cells overexpressing the P-glycoprotein efflux pump 4 6 The recent isolation of the biosynthetic gene cluster responsible for production of the epothilones has revealed another promising avenue of investigation and possible new source of analogues via combinatorial biosynthesis. 4 7 The next two classes of microtubule-stabilizing agents were discovered as secondary metabolites from marine sponges. Discodermolide (17) was originally isolated from the Caribbean sponge Discodermia dissoluta as an immunosuppressant and was later screened for antimitotic activity on the basis of a computer-assisted structure analysis that compared it to 48 other tubulin-interacting drugs 4 8 Laulimalide (54), however, was identified from the sponge Cacospongia mycofijiensis in a mechanism-based screening program specifically aimed at the discovery of new antimicrotubule agents.4 9 54 The fourth structural class of antimitotic compounds with microtubule-stabilizing properties is represented by the soft coral metabolites eleutherobin and the sarcodictyins. Pietra et al. originally isolated sarcodictyins A (55) and B (56) from a Mediterranean stolonifer Sarcodictyon roseum and later reported four other sarcodictyins, C to F (57-60) from the same source. 5 0 ' 5 1 There was, however, no mention of any biological activity associated with these new eunicellin-type diterpenoids. It was eight years after the discovery of the sarcodictyins that Fenical et al. filed a Patent application describing the isolation of eleutherobin (61) and its cytotoxic properties. 5 2 The authors reported the isolation of eleutherobin from a rare alcyonacean, identified as an Eleutherobia species, collected in the waters of Western Australia and showed eleutherobin (61) to be a potent cancer cell growth inhibitor against a diverse panel of tumour tissue cell lines. The mechanism of action was still undetermined at this point and 49 shortly thereafter, Kashman et al. isolated sarcodictyin A (55) and eleuthosides A (62) and B (63) from the South African soft coral Eleutherobia aurea, again with no mention of any biological activity. 5 3 This flurry of activity spanning 1995 and 1997 included a Patent application by Pietra et al. describing the microtubule-stabilizing properties of the sarcodictyins, 5 4 followed soon after by a journal article by Fenical et al. reporting the isolation and paclitaxel-like activity of eleutherobin (61),55 and the subsequent total synthesis of eleutherobin (61) by Nicolaou's group. 5 6 Despite all this research interest and further total syntheses of eleutherobin (61) and the sarcodictyins, 5 7" 5 9 the pre-clinical development of eleutherobin (61) as an antimitotic agent had been impeded by its scarcity. In fact, the licensing agreement to advance the development of eleutherobin (61) by Bristol-Myers Squibb was recently terminated because "we couldn't get any of it" [Fenical]. 6 0 Me 55 R 56 R 57 R 58 R M e R 2 = H Et R 2 = Ff M e R 2 = O H M e R 2 = O A c 59 Me i N 60 In addition to the five main structural classes listed above (taxane, epothilone, discodermolide, laulimalide and sarcodictyin/eleutherobin), there have been recent reports of other compounds that demonstrate the ability to stabilize microtubules. These include the hexacyclic WS9885B (64) from a Streptomyces species, 6 1 some steroidal derivatives (65, 66) of estradiol, 6 2 and certain polyisoprenylated benzophenones (67, 68) from a Malaysian plant. 6 3 It is clear from the success of paclitaxel (4) that further investigations into microtubule-stabilizing compounds wi l l yield promising new anticancer therapeutics. 66 R = A c gg A36,37 2.2. Results and Discussion 2.2.1. Isolation of Antimitotic Compounds A s part of a general collection of marine invertebrates from the tropical waters off Dominica, specimens of an encrusting gorgonian were collected near Prince Rupert Bay using 51 S C U B A . Samples were harvested by hand from the shallow water reefs, frozen on site, and transported to Vancouver over dry ice. The organism was identified as Erythropodium caribaeorum (Duchassaing & Michelotti , 1860) by Dr. Leen van Ofwegan of the Nationaal Natuurhistorisch Museum in Leiden, The Netherlands. Voucher samples are stored both at the University of British Columbia in Vancouver and in Leiden. A n initial, smaller collection of E. caribaeorum in the summer of 1997 resulted in the isolation of nearly equivalent amounts of two diterpene glycosides, one new and one previously reported. What is described below is the equivalent workup on a second, larger collection later that same year, resulting in the isolation of a total of six novel compounds. Frozen samples of E. caribaeorum (5.3 kg wet weight) were thawed and extracted exhaustively with MeOFf over a period of several days. The combined methanolic extracts were concentrated in vacuo to give a dark brown gum (280 g). A sample of this concentrated gum showed potent activity in Roberge's cell-based antimitotic assay. 6 4 In addition, microscopic examination of the cells arrested in mitosis exhibited evidence of tubulin bundling, similar to the characteristic morphological effects of paclitaxel (4). The combined crude extract was suspended in two liters of water and partitioned against E t O A c (2 liters x 5). The EtOAc fraction was concentrated in vacuo, suspended in two liters of 9:1 M e O H / F L O , and then partitioned against hexanes (2 liters x 5). Next, water was added to the remaining methanolic fraction in order to achieve a 6:4 M e O H / F L O ratio. This fraction was finally partitioned against two liters of C H C I 3 (2 liters x 5). Bioassay-guided fractionation showed the antimitotic activity was entirely present in the E t O A c , the 9:1 M e O H / H 2 0 , and the C H C I 3 fractions, respectively (see Scheme 2.3). Crude Methanol Extract Octocoral Erythropodium caribaeorum H 2 0 E t O A c 9:1 M e O H : H 2 0 Hexane 6:4 M e O H : H 2 0 C H C I 3 Reversed-Phase chromatography 80% H 20/MeOH 70% H20/MeOHj 60% H 20/MeOH 5 0 % H,0/MeOHi 4 0 % H 20/MeOH| 30% H 20/MeOH| 2 % MeOH/EtOAc 4 % MeOH/EtOAc 6 % MeOH/EtOAc High Performance liquid chromatography 8 % MeOH/EtOAc 74 ( lmg) 20 % H 20/MeOH| 10% H 20/MeOH| 100% M e O H Normal Phase chromatography 10% MeOH/EtOAc 12 % MeOH/EtOAc 14 % MeOH/EtOAc 71 (3mg) 72 (2mg) 61 (50mg) 69 (7mg) 70(6mg) 73 ( lmg) Scheme 2.3. Isolation procedure for antimitotic diterpenoids from E. caribaeorum. 53 Fractionation of the active chloroform fraction (59 g) was accomplished with vacuum, reversed-phase, flash liquid chromatography by first adsorbing the active extract onto C-18 silica gel. The dried extract and powder were then subjected to liquid chromatography using step-gradient elution (80:20 Ff 2 0/MeOH to MeOFf in 10% increments). The antimitotic activity was only observed in the 30:70 FEO/MeOH fraction. This active fraction was further subjected to normal phase, flash liquid chromatography using a step-gradient elution scheme (EtOAc to 20:80 M e O H / E t O A c in 2% increments). Instead of using the traditional thin layer chromatographic analysis and combination of like-fractions, a sample of each fraction was individually placed on a 96-well plate and screened for antimitotic activity. This was the first time this technique was used and was made possible by the development of the new antimitotic bioassay. The biological activity associated with each fraction is presented in Figure 2.3. At this stage, the antimitotic activity was centered in the three fractions corresponding to 6, 8 and 10% M e O H / E t O A c . ' H N M R spectral screening of these and surrounding fractions revealed signals for a family of related diterpene glycosides. 0.8 r-< 0 10 20 30 40 50 Fraction # Figure 2.3. Antimitotic activity (reported as Absorbance^nm) of each fraction collected from liquid chromatography. 54 A final purification step involving normal phase, high performance liquid chromatography (HPLC) using isocratic elution conditions (7:93 M e O H / C H 2 C l 2 ) yielded pure samples of 61 (50 mg), 69 (7 mg), 70 (6 mg), 71 (3 mg), and 72 (2 mg). The disappearance o f certain signals in the *H N M R spectra appeared to suggest the presence of compounds that were partially decomposing when subjected to the H P L C conditions. Thus, these fractions were ultimately purified on cyano-bonded-phase H P L C using a 56:42:2 EtOAc/hexane/( 'Pr) 2NH solvent mixture system as the eluent and gave small amounts of pure 73 (1 mg) and 74 (1 mg). O H 71 Rj = H ; R 2 = A c ; R 3 = Me; A (E) 72 R , = A c ; R 2 = H ; R 3 = M e ; A 2 ' 3 ' (Z) 2.2.2. Structure Elucidation of Antimitotic Diterpenoids The structures of eleutherobin (61), desmethyleleutherobin (69), desacetyleleutherobin (70), isoeleutherobin A (71), Z-eleutherobin (72), caribaeoside (73), and caribaeolin (74) were solved by extensive analysis of 1-dimensional and 2-dimensional N M R spectroscopic data recorded in DMSO-dg at 500 M H z . Low and high resolution F A B mass spectrometry provided important information on the molecular formula of the compounds. Proton spin systems were identified from C O S Y data and proton-carbon attachments were determined by H M Q C experiments. The H M B C data proved invaluable for determining the connectivity within the 55 compounds and also in the assignment of any quaternary carbons present. The relative stereochemistry and conformational information was obtained by analysis of N O E difference, R O E S Y , and X-ray diffraction experiments presented later in this chapter. A s mentioned earlier, an initial, smaller scale workup of the crude extract from E. caribaeorum resulted in the isolation of nearly equivalent amounts of the two diterpene glycosides, desmethyleleutherobin (69) and eleutherobin (61). Although compound 61 was present in much larger quantities in the second, larger scale isolation, the structural determination of 69 preceded that of the previously reported eleutherobin (61) and thus wil l be presented in detail first. 2.2.2.1. Desmethyleleutherobin OH Desmethyleleutherobin (69) was isolated as a white amorphous solid that gave an [ M + H ] + ion in the H R F A B M S at m/z 643.32230 appropriate for a molecular formula of C34H46N2O10 ( A M -1.21 ppm) and thirteen degrees of unsaturation. A quick analysis of the I D and 2D N M R spectra indicated the presence of two esters, eight olefinic protons, one methyl singlet attached to a nitrogen heteroatom, and one acetal suggesting a glycoside moiety. A summary of all N M R assignments, C O S Y , and H M B C correlations for desmethyleleutherobin (69) can be found in Table 2.1. 56 Beginning with the signals furthest downfield in the ' H N M R spectrum, there are two singlets (5 7.68 and 5 7.57 ppm) integrating for one proton each and showing H M Q C correlations to carbon signals in the aromatic region of the C N M R spectrum. H M B C correlations were observed from these two carbons into a singlet proton resonance occurring at 5 3.66 ppm, appropriate for a methyl group attached to a nitrogen atom (see Figure 2.4). The two proton resonances at 5 7.68 and 5 7.57 ppm showed H M B C correlations into a quaternary carbon at 5 136.9 ppm, thus suggesting the presence of an Af(l)-methylimidazole functional group. The H M B C experiment revealed further connectivity to this quaternary carbon from two isolated olefinic protons appearing at 5 7.51 and 8 6.34 ppm. The coupling constant for these protons was determined to be 15.5 Hz , corresponding to an £-olefinic configuration. Finally, both of these proton resonances displayed H M B C correlations to an ester carbon at 5 166.0 ppm, forming an .E-o^P-unsaturated ester. The resonance effects of this conjugated system result in the observed downfield shift of the P-proton. Together with the attached methyl-substituted imidazole, the a,P-unsaturated ester forms an extended conjugated system displaying a strong U V absorbance at 290 nm. Substructure A of 69 was thus identified as a /V-methylurocanic acid ester functionality. H M B C C O S Y Figure 2.4. Selected H M B C and C O S Y correlations for substructure A of compound 69. 57 Investigation of the ' H N M R spectrum indicated several proton resonances between 5 3.3 and 8 5.0 ppm, characteristic of a sugar moiety. A detailed analysis of the various coupling constants, C O S Y correlations, and chemical shift considerations indicated the presence of an a-2"-0-acetyl-arabinoside. The 1 3 C N M R and H M Q C spectra supported this by exhibiting a diagnostic anomeric carbon resonance at 8 92.9 ppm linked to its anomeric proton at 8 4.71 ppm. This proton appears as a doublet, with a coupling constant value of 3.2 Hz, and C O S Y correlations were observed between this proton and a proton further downfield at 8 4.82 ppm. The downfield proton showed additional C O S Y correlations to a third proton resonance at 8 3.73 ppm and appeared as a doublet of doublets with coupling constants of 3.2 and 10.0 Hz, respectively. The large J value locks the downfield proton into an axial position together with the third proton, and thus establishes an equatorial configuration for the anomeric proton (see Figure 2.5). The presence of an acetate ester linked to the glycoside was suggested by the downfield proton and carbon chemical shifts observed at the 2" position ( 'H : 8 4.82; 1 3 C : 8 70.8 ppm). This was further supported by L R F A B M S evidence of a mass spectral fragment at m/z 451 corresponding to the loss of a C 7 H i i 0 6 moiety (cleavage of the - O — C I " - bond). C O S Y H M B C C O S Y Figure 2.5. Selected H M B C and C O S Y correlations for substructure B of compound 69. A key heteronuclear correlation was observed from an up field proton at 8 3.44 ppm into the anomeric carbon resonance. This upfield proton and its geminal partner (8 3.62 ppm) 58 established the 5" position of the glycoside and exhibited typical geminal coupling (/ = 11.9 Hz) and C O S Y correlations. The downfield partner was assigned the equatorial position characteristic for a deshielded equatorial substituent. The multiplicity of the axial H-5" resonance in the ' H N M R , coupling constant value of 2.6 Hz, and C O S Y correlation into § 3.73 ppm revealed a neighbouring equatorial proton at the H-4" position. Both the H-4" and H-3" proton resonances lie in the 8 3.70-3.75 ppm region and their coupling constant values and C O S Y correlations are obscured by signal overlap. However, the full complement of ID and 2D N M R spectral data fully confirms substructure B of compound 69 as an a-2"-6>-acetyl-arabinoside. Analysis of the remaining atoms and degrees of unsaturation required by the molecular formula indicated substructures A and B of compound 69 account for all of the nitrogen atoms, eight of the ten oxygen atoms, and seven of the thirteen sites of unsaturation. What remains is a C20 unit and, based on the related chemistry of E. caribaeorum and other gorgonians, this suggested a diterpenoid structure. Inspection of the *H and 1 3 C N M R spectra in addition to the C O S Y and H M Q C experiments revealed an isolated methylene group (8 4.24 and 5 3.77 ppm) attached to an oxygen heteroatom and a quaternary carbon center. H M B C correlations were observed between these methylene protons and the anomeric carbon of substructure B , thus establishing the glycosidic linkage (see Figure 2.6). These methylene protons also showed H M B C correlations to an olefinic carbon signal at 8 133.0 ppm whose attached proton (8 5.29 ppm) only exhibited homonuclear coupling to a neighbouring allylic methine ( 'H: 8 3.95, 1 3 C : 8 33.5 ppm). H M B C correlations were also present from the methylene protons to a carbon at 8 111.6 ppm that perhaps suggested a tetrahedral carbon deshielded by two oxygen atoms of a hemiketal/ketal. Heteronuclear correlations of a hydroxyl proton to the carbons at 8 111.6 and 8 135.0 ppm supported the hemiketal structure and led to the assignment of the quaternary allylic carbon. 59 / " \ C O S Y Figure 2.6. Selected H M B C and C O S Y correlations for substructure C of compound 69. The ] H N M R spectrum of compound 69 showed evidence of an isolated cw-olefin from the multiplicity and the characteristic value of the coupling constant (7 = 5.6 Hz , typical for a five-membered ring). These olefinic proton resonances at 5 6.07 and 8 6.11 ppm displayed H M B C correlations into the carbon resonance at 8 111.6 ppm and an additional signal at 8 88.6 ppm. Due to the cw-configuration of the olefin and the limit of two oxygen atoms unaccounted for in substructures A and B , the five-membered cyclic hemiketal system was identified. A methyl singlet at 8 1.35 ppm in the ' H N M R spectrum exhibited H M B C correlations into the downfield carbon of the c/s-olefin, the deshielded carbon resonance at 8 88.6 ppm, and an additional carbon signal at 8 80.6 ppm. The proton (8 4.61 ppm) attached to this carbon atom provided the link to substructure A in the form of a heteronuclear correlation to the ester carbon of the /V-methylurocanic acid moiety. With all of the heteroatoms accounted for and the connectivities to substructures A and B established, all that remained was a Cio hydrocarbon portion providing three additional degrees of unsaturation. Investigation of the C O S Y spectrum revealed that the proton at 8 4.61 ppm was correlated into a proton at 8 1.32 ppm. This proton is one of the two methylene protons attached to the carbon at 8 31.1 ppm and is also coupled into a neighbouring methine proton at 8 2.45 60 ppm (see Figure 2.7). The final methyl singlet present in the ' H N M R spectrum (8 1.47 ppm) exhibited H M B C correlations into the downfield shifted carbon (8 38.4 ppm) attached to the neighbouring methine proton, a quaternary carbon resonance at 8 133.8 ppm, and the remaining olefinic carbon signal at 8 120.8 ppm. The proton attached to this olefinic carbon was the starting point of a series of C O S Y correlations that flowed to a vicinal methylene group (5 2.28 and 8 1.95 ppm), onto a methine proton (8 1.14 ppm), and ultimately to the single proton resonance at 8 3.95 ppm identified earlier in substructure C. The remaining upfield signals present in the ' H N M R spectrum (8 0.92 and 8 0.93 ppm) were readily identified as belonging to an isopropyl group attached to the carbon at 8 42.1 ppm. A final H M B C correlation from a methylene proton at 8 1.32 ppm into the carbon resonance at 8 33.5 ppm completed the cyclohexene ring and accounted for all the degrees of unsaturation expected from the molecular formula. Figure 2.7. Selected H M B C and C O S Y correlations for substructure D of compound 69. The I D and 2 D N M R data obtained for compound 69, shown in Figures 2.8 to 2.12, was in complete agreement with the structure of desmethyleleutherobin. The relative stereochemistry of desmethyleleutherobin (69) was found to be identical to that of eleutherobin (61), presented later in this chapter. A summary of all N M R assignments and correlations for desmethyleleutherobin (69) can be found in Table 2.1. Table 2.1. N M R data for desmethyleleutherobin (69) recorded in DMSO-J6. position 5 ' H a (7 value in Hz) 6 1 3 C b C O S Y Correlations H M B C Correlations 1 3.95 (m) 33.5 H2 H9b 2 5.29 (d, 7=9.1) 133.0 HI H15a/15b 3 ' 135.0 H15a, 4-OH 4 111.6 H2, 5,6, 15b, 4-OH 4-OH 6.40 (s) 5 6.07 (d, 7=5.6) 131.0 H6* H6 6 6.11 (d, 7=5.6) 133.5 H5* H5, 8, 16 7 88.6 H5, 6, 8, 16 8 4.61 (d, 7=7.2) 80.8 H9b H9a*, 16* 9a 1.46 (m) 31.1 H9b b 1.32 (m) H9a, 8, 10 10 2.45 (m) 38.4 H9b H8, 17 11 133.8 H17 12 5.26 (m) 120.8 H13b H17 13a 2.28 (m) 23.9 H13b, 14 b 1.95 (m) H13a, 12, 14 14 1.14 (m) 42.1 H13a/b, 1 H2, 19, 20 15a 4.24 (d, 7=12.5) 68.2 H15b H2 b 3.77 (d, 7=12.5) H15a 16 1.35 (s) 25.1 H8 17 1.47 (s) 21.7 18 1.46 (m) 28.8 H19, 20 H19, 20 19 0.93 (d, 7=6.7) 20.3 H18 H20 20 0.92 (d, 7=6.7) 22.0 H18 H19 r 166.0 H8, 2', 3' 2' 6.34 (d, 7=15.5) 113.7 H3' 3' 7.51 (d, 7=15.5) 137.6 H2' 4' 136.9 H 3 \ 5', 7' 5' 7.57 (s) 124.8 6'-NMe 6'-NMe 3.66 (s) 33.2 7' 7.68 (s) 140.0 6'-NMe 1" 4.71 (d, 7=3.2) 92.9 H2" H15a, 5"ax 2" 4.82 (dd, 7=3.2,10.0) 70.8 HI" , 3" 3" • 3.73 (m) 68.6 H2", 3"-OH, 4"* H5"ax 3"-OH 4.78 (d, 7=3.6) H3" 4" 3.76 (m) 66.2 H3"*, 4"-OH, 5"ax H2", 5"ax 4"-OH 4.88 (d, 7=6.4) H4" 5"eq 3.62 (d, 7=11.9) 63.1 H5"ax ax 3.44 (dd, 7=11.9,2.6) H5"eq, 4" 1"' 170.0 H2'" 2"' 2.01 (s) 21.0 a recorded at 500 MHz. recorded at 100 M H z ; exact assignments based on H M Q C and H M B C data. * unambiguous assignments not possible due to signal overlap. 62 64 O H 8.0 6.0 4.0 2.0 0.0 (ppm) Figure 2.10. C O S Y spectrum of desmethyleleutherobin (69) recorded in D M S O - d 6 at 500 M H z . (ppm) 7.00 6.00 5.00 4.00 3.00 2.00 Figure 2.11. H M Q C spectrum of desmethyleleutherobin (69) recorded in D M S C M 6 at 500 M H z . 66 Figure 2.12. H M B C spectrum of desmethyleleutherobin (69) recorded in DMSO-J6 at 500 M H z . 67 2.2.2.2. Eleutherobin O H Me N N i O H 61 The structural elucidation of 61 was solved by extensive analysis of I D and 2D N M R spectral data and analysis of the mass spectrometric data. The final structure was found to have Eleutherobin (61) was isolated as a white amorphous solid that gave an [ M + H ] + ion in the H R F A B M S at m/z 657.33719 appropriate for a molecular formula of C 3 5 H 4 8 N 2 O 1 0 ( A M -2.32 ppm). This molecular formula differed from the value obtained for desmethyleleutherobin (69) by the gain of fourteen mass units, corresponding to the addition of C H 2 . The *H N M R spectrum of 61 differed from the ' H N M R spectrum of desmethyleleutherobin (69) only by the presence of the singlet methyl resonance at 5 3.09 ppm, consistent with a methoxyl substituent. The mass spectral and ' H N M R data also suggested that 61 was identical to desmethyleleutherobin (69) except for the presence of a methoxyl group instead of a hydroxyl group at C-4 on the diterpene core. Detailed analysis of the H M Q C and H M B C experimental data confirmed the presence of this group, showing a three-bond correlation from the methoxyl proton into the quaternary ketal carbon at 8 115.6 ppm (see Figure 2.13). Also , the replacement of a hydrogen atom by a methyl group results in the observed downfield shift of the ketal carbon from 8 111.6 ppm in desmethyleleutherobin (69) to 8 115.6 ppm in eleutherobin (61). been previously published months earlier by other authors. 55 68 x ( / ^ ~ ^ \ H M B C X C O S Y Figure 2.13. Selected H M B C and C O S Y correlations for the cyclic ketal subfragment of The I D and 2D N M R data obtained for 61 was in complete agreement with this assignment and all spectra reacquired in CDCI3 matched those previously reported in that solvent. 5 5 A detailed analysis of the relative stereochemistry by R O E S Y and N O E difference experiments can be found in Section 2.2.4. A summary of all N M R assignments and correlations for eleutherobin (61) in D M S O - J 6 can be found in Table 2.2 and additional 2D spectra can be found in the Appendix. compound 61. Table 2.2. N M R data for eleutherobin (61) recorded in D M S O - J 6 . position 5 'H a 5 1 3 C b C O S Y H M B C (7 value in Hz) Correlations Correlatior 1 3.88 (m) 33.5 H2, 14 H9a 2 5.39 (d, 7=9.4) 134.9 HI H15a 3 133.3 H15a 4 115.6 H2, 5, 6, 15 4-OMe 3.09 (s) 49.1 5 6.08 (d, 7=5.9) 130.6 H6 H6 6 6.28 (d, 7=5.9) 133.6 H5 H5, 8 7 89.2 H5, 6, 8, 16 8 4.65 (d, 7=7.3) 80.7 H9a, 16 9a 1.49 (m) 30.9 H9b b 1.32 (m) H9a 10 2.45 (m) 38.4 H8 11 133.7 H17 12 5.27 (m) 120.9 H17 13a 2.27 (m) 23.9 b 1.95 (m) 14 1.14 (m) 42.0 HI H2, 19, 20 15a 4.16 (d, 7=12.6) 67.7 H15b H2 b 3.78 (d, 7=12.6) H15a 16 1.37 (s) 23.9 H8 17 1.47 (s) 21.5 18 1.45 (m) 28.8 H19, 20 H19, 20 19 0.93 (d, 7=6.6) 20.2 H18 20 r 0.91 (d, 7=6.6) 21.9 166.0 H18 H8, 2', 3' 2' 6.35 (d, 7=15.6) 113.5 H3' 3' 7.53 (d, 7=15.6) 137.8 H2' 4' 136.8 H 2 \ 3', 5', 5' 7.57 (s) 124.8 6'-NMe 6'-NMe 3.66 (s) 33.2 7' 7.69 (s) 140.0 6'-NMe 1" 4.70 (d, 7=3.0) 92.9 H2" H15a, 5"ax 2" 4.82 (dd, 7=3.0,9.4) 70.8 HI" , 3" 3" 3.73 (m) 68.5 H2", 3"-OH, 4"* 3"-OH 4.83 (d, 7=3.2) H3" 4" 3.76 (m) 66.1 H3"*, 4"-OH, 5"ax H2", 5"ax 4"-OH 4.92 (d, 7=6.0) H4" 5"eq 3.59 (d, 7=11.8) 63.0 H5"ax ax 3.44 (dd, 7=11.8,2.2) H5"eq, 4" H I " 1"' 170.1 H2", 2"' T" 2.01 (s) 20.9 a recorded at 500 MHz. b recorded at 100 MHz. * unambiguous assignments not possible due to signal overlap. 71 72 2.2.2.3. Desacetyleleutherobin O H Me N N i O H 70 Desacetyleleutherobin (70) was isolated as a white amorphous solid that gave an [M + H ] + ion in the H R F A B M S at m/z 615.32813 appropriate for a molecular formula of C 3 3 H 4 6 N 2 O 9 ( A M -0.05 ppm). This molecular formula differed from the value obtained for eleutherobin (61) by the loss of 42 mass units, corresponding to the loss of C 2 H 2 0 . The ' H N M R spectrum of 70 displayed a strong resemblance to the ! H N M R spectrum of eleutherobin (61) except for the absence of a methyl singlet resonance at 5 2.01 ppm that had been assigned to the 2" acetyl substituent and differences in the chemical shifts of the signals attributed to the arabinose protons. The mass spectral and ' H N M R data thus suggested that 70 was simply the desacetyl analogue of eleutherobin (61). Detailed analysis of the C O S Y and H M Q C experimental data confirmed the absence of this group by showing a homonuclear correlation from the anomeric proton at 8 4.57 ppm to the neighbouring 2" proton at 8 3.55 ppm (see Figure 2.16). The upfield shift in both the H-2" and C-2" resonances demonstrated they no longer experienced the deshielding effects of an attached, electron-withdrawing acetyl group. Relative to the H-2" and H-3" axial proton resonances, the ! H N M R chemical shift assignments for H-4" and H-5"eq displayed the expected downfield shifts of equatorial substituents. 73 C O S Y Figure 2.16. Selected C O S Y correlations for the arabinose subunit of compound 70. In order to confirm the identity of the arabinose sugar moiety, compound 70 was acetylated with acetic anhydride in pyridine. The resulting product was identical to diacetyleleutherobin (75) prepared by acetylation of eleutherobin (61) using the same reaction conditions (see Section 2.2.2.8), thus confirming the proposed structure of desacetyleleutherobin (70). The 2D N M R data obtained for 70 was in complete agreement with this assignment. A summary of all N M R assignments and correlations for desacetyleleutherobin (70) in DMSO-c/6 can be found in Table 2.3 and additional 2D spectra can be found in the Appendix. Table 2.3. N M R data for desacetyleleutherobin (70) recorded in D M S O - J 6 . position 5 'H a 5 1 3 C b C O S Y H M B C (7 value in Hz) Correlations Correlations 1 3.89 (m) 33.3 H2, 14 2 5.51 (d, 7=9.2) 132.3 HI H15a/15b 3 133.9 H15a 4 115.6 H2, 5, 6, 15b, 4-OMe 4-OMe 3.09 (s) 49.0 5 6.23 (d, 7=6.0) 130.9 H6 H6 6 6.26 (d, 7=6.0) 134.0 H5 H5, 8, 16 7 89.1 H5, 8, 16 8 4.66 (d, 7=6.6) 80.7 H9a/9b H16 9a 1.51 (m) 31.1 H9b, 8 b 1.41 (m) H9a, 8 10 2.46 (m) 38.4 H8 11 133.5 H17 12 5.26 (m) 121.2 H13b H17 13a 2.35 (m) 23.9 b 1.93 (m) H12 14 1.14 (m) 42.2 HI H2, 19, 20 15a 4.16 (d, 7=12.9) 67.9 H15b H2 b 3.83 (d, 7=12.9) H15a 16 1.37 (s) 23.9 17 1.47 (s) 21.7 18 1.49 (m) 28.7 H19, 20 H19, 20 19 0.94 (d, 7=6.9) 20.2 H18 20 r 0.92 (d, 7=6.9) 22.0 166.0 H18 H8, 3' 2' 6.35 (d, 7=15.7) 113.5 H3' 3' 7.53 (d, 7=15.7) 137.8 H2' 4' 136.8 H 2 ' , 3 \ 5 ' , 7 ' 5' 7.56 (s) 124.8 6'-NMe 6'-NMe 3.66 (s) 33.1 7' 7.68 (s) 140.0 6'-NMe 1" 4.57 (br.d) 97.1 H2" H15a 2" 3.55 (m) 69.2* HI" , 3"* 2"-OH * 3" 3.55 (m) 68.5* H2"* 3"-OH * 4" 3.66 (m) 68.1* 4"-OH * 5"eq 3.54 (d, 7=12.3) 63.1 H5"ax ax 3.37 (dd, 7=12.3,3.3) H5"eq a recorded at 500 MHz. b recorded at 100 MHz; exact assignments based on H M Q C and H M B C data. * unambiguous assignments not possible due to signal overlap or poor signal to noise. 77 2.2.2.4. Isoeleutherobin A OH M e N-N i OAc 71 Isoeleutherobin A (71) was isolated as a white amorphous solid that gave an [ M + H] + ion in the H R F A B M S at m/z 657.33834 appropriate for a molecular formula of C35H48N2O10 ( A M -0.58 ppm). This molecular formula was identical to that of eleutherobin (61). The ' H N M R spectrum of 71 displayed a strong resemblance to the *H N M R spectrum of eleutherobin (61) except for differences in the chemical shifts of the signals attributed to the arabinose protons. The mass spectral and ' H N M R data also suggested that 71 differed only in the acetylation pattern on the arabinose fragment. Detailed analysis of the C O S Y and H M Q C experimental data confirmed the absence of the acetyl group from the C-2" position by showing a homonuclear correlation from the anomeric proton at 5 4.64 ppm to the neighbouring 2" proton at § 3.81 ppm (see Figure 2.19). The upfield shift in both the H-2" and C-2" resonances relative to those observed in eleutherobin (61) showed they no longer experienced the deshielding effects of an attached, electron-withdrawing acetyl group. Likewise, C O S Y correlations observed between the axial C - 5 " methylene proton at 5 3.38 ppm and a methine resonance at 8 3.83 ppm revealed the acetate was not at C-4". This H-4" methine resonance also displayed a C O S Y correlation to a downfield resonance at 5 4.80 ppm, which was thus assigned to the remaining H-3" position. The significant deshielding of both the H-3" and C-3" 78 resonances of compound 71 relative to the corresponding chemical shifts in eleutherobin (61) indicated the electron-withdrawing acetate group was present at C-3". Figure 2.19. Selected C O S Y correlations for the arabinose subunit of compound 71. The splitting pattern observed in the ' H N M R spectrum for H-3" was consistent with axial-axial coupling to H-2" (/ = 10.1 Hz) and axial-equatorial coupling to H-4" (7 = 2.6 Hz). However, in order to confirm the identity of the arabinose sugar moiety, compound 71 was acetylated with acetic anhydride in pyridine. The resulting product was identical to diacetyleleutherobin (75) prepared by acetylation of eleutherobin (61) using the same reaction conditions (see Section 2.2.2.8), thus confirming the proposed structure of isoeleutherobin A (71). The 2D N M R data obtained for 71 was in complete agreement with this assignment. A summary of all N M R assignments and correlations for isoeleutherobin A (71) in DMSO-d6 can be found in Table 2.4 and additional 2D spectra can be found in the Appendix. Table 2.4. N M R data for isoeleutherobin A (71) recorded in D M S O - J 6 . position 5 ' H a (7 value in Hz) 8 l 3 C b C O S Y Correlations H M B C Correlations 1 3.90 (m) 33.1 H2 2 5.54 (d, .7=10.1) 132.3 HI H15a/15b 3 133.7 H15a 4 115.7 H2, 5,6, 15b, 4-OMe 4-OMe 3.09 (s) 49.0 5 6.23 (d, 7=5.7) 130.7 H6 H6 6 6.26 (d, 7=5.7) 133.9 H5 H5, 8 7 89.2 H5, 6, 8, 16 8 4.67 (d, 7=7.1) 80.8 H9b H16 9a 1.52 (m) 31.1 H9b b 1.43 (m) H9a, 8, 10 10 2.47 (m) 38.4 H9b H8, 17 11 133.3 H17 12 5.28 (m) 121.4 H13b H17 13a 2.37 (m) 23.9 HI 3b, 14 b 1.93 (m) H13a, 12 14 1.14 (m) 42.1 H13a H2, 19, 20 15a 4.17 (d, 7=12.7) 67.6 H15b H2 b 3.89 (d, 7=12.7) H15a 16 1.37 (s) 23.9 17 1.47 (s) 21.7 18 1.48 (m) 28.8 H19, 20 H19, 20 19 0.94 (d, 7=6.6) 20.2 H18 20 0.92 (d, 7=6.6) 22.0 H18 r 166.0 H8, 3' 2' 6.35 (d, 7=15.4) 113.6 H3' 3' 7.53 (d, 7=15.4) 137.6 H2' 4' 136.8 H 2 \ 5' 5' 7.57 (s) 124.8 6'-NMe 6'-NMe 3.66 (s) 33.2 7' 7.70 (s) 140.0 6'-NMe 1" 4.64 (d, 7=3.2) 97.0 H2" H15a 2" 3.81 (m)* 65.9* HI" , 2"-OH, 3"* 2"-OH 4.90 (d, 7=5.8)* H2"* 3" 4.80 (dd, 7=2.6,10.1) 72.7 H2", 4"* 4" 3.83 (m)* 64.9* H3", 4"-OH, 5"ax 4"-OH 4.96 (br.d)* H4"* 5"eq 3.62 (d, 7=12.2) 63.1 H5"ax ax 3.38 (m) H5"eq, 4"* 1"' 170.2 H2'" 2" 1 2.01 (s) 21.0 recorded at 500 MHz. b recorded at 125 MHz; exact assignments based on H M Q C and H M B C data. * unambiguous assignments not possible due to signal overlap or poor signal to noise. 80 81 2.2.2.5. Z-eleutherobin OH O - V 72 Z-eleutherobin (72) was isolated as a white amorphous solid that gave an [ M + H ] + ion in the H R F A B M S at m/z 657.33830 appropriate for a molecular formula of C 3 5 H 4 8 N 2 O 1 0 ( A M -0.65 ppm). This molecular formula was again identical to that of eleutherobin (61). The l H N M R spectrum of 72 displayed a strong resemblance to the ' H N M R spectrum of eleutherobin (61) except for differences in the chemical shifts of the signals attributed to the N-methylurocanic acid ester residue. Detailed analysis of the ' H N M R spectrum suggested that the only structural change was in the configuration of the A 2 3 olefin, changing from an E-configuration with a coupling constant of 15.6 H z in eleutherobin (61) to a Z-configuration with a coupling constant of 12.6 H z in compound 72. Further inspection of the 2D N M R data confirmed this change in configuration and allowed the assignment of all the nuclear resonances. In particular, the three-bond H M B C correlation from the olefinic proton at 8 6.94 ppm into an imidazole ring carbon (8 126.0 ppm) revealed this proton to be H-3 ' (see Figure 2.21). The methine proton attached to the imidazole carbon was found at 8 8.25 ppm, most likely a result of its spatial proximity to the anisotropic deshielding effects of the carbonyl functionality. Finally, H M B C correlations from the methyl singlet at 8 3.71 ppm and from the highly deshielded methine proton at 8 8.25 ppm established connectivity to the remaining imidazole carbons. 82 3.71 Figure 2.21. Selected H M B C and C O S Y correlations for the /Y-methylurocani.c acid ester subunit of compound 72. The N M R sample of compound 72 partially isomerized over time to eleutherobin (61), thus confirming the assigned structure of 72 as Z-eleutherobin. The 2D N M R data obtained for 72 was in complete agreement with this assignment. A summary of all N M R assignments and correlations for Z-eleutherobin (72) in DMSO-<i6 can be found in Table 2.5 and additional 2D spectra can be found in the Appendix. Table 2,5. N M R data for Z-eleutherobin (72) recorded in D M S O - J 6 . position 5 ' H a 5 I 3 C b C O S Y H M B C (J value in Hz) Correlations Correlations 1 3.92(m) 33.6 H2 H9a 2 5.40 (d, 7=9.4) 134.9 H15a/15b 3 133.3 H15a 4 115.6 H2, 5,6, 15b, 4-OMe 4-OMe 3.09 (s) 49.1 5 6.07 (d, 7-5.6) 130.7 H6 H6 6 6.18 (d, 7=5.6) 133.4 H5 H5, 8, 16 7 89.2 H5, 6, 8, 16 8 4.66 (d, 7=7.2) 80.8 H9b H16 9a 1.51 (m) 30.8 H9b b 1.33 (m) H9a, 8 10 2.47 (m) 38.4 H8, 17 11 133.8 H17 12 5.29 (m) 121.1 H13b H17 13a 2.28 (m) 23.9 H13b H18 • b 1.95 (m) H13a, 12 14 1.14 (m) 42.4 H2, 19, 20 15a 4.15 (d, 7=12.6) 67.8 H15b H2 b 3.78 (d, 7=12.6) H15a 16 1.36 (s) 24.0 17 1.49 (s) 21.3 18 1.47 (m) 28.8 H19, 20 H19, 20 19 0.95 (d, 7=6.6) 20.4 H18 H20 20 0.93 (d, 7=6.6) 21.8 H18 H19 1' 165.4 H8 2' 5.75 (d, 7=12.6) 112.3 H3' 3' 6.94 (d, 7=12.6) 138.9 H2' 4' 136.2 H5' 5' 8.25 (s) 126.0 6'-NMe 6'-NMe 3.71 (s) 33.2 7' 7.69 (s) 138.6 6'-NMe 1" 4.70 (d, 7=3.0) 93.2 H2" HI5a, 5"ax 2" 4.82 (dd, 7=3.0,9.6) 70.9 HI" , 3" 3" 3.76 (m) 68.5 H2", 3"-OH*, 4"* 3"-OH 4.81 (d, 7=2.8)* H3"* 4" 3.78 (m) 66.3 H3"*, 4"-OH* H5"ax 4"-OH 4.90 (d, 7=6.0)* H4"* 5"eq 3.59 (d, 7=12.0) 63.2 H5"ax ax 3.44 (dd, 7=12.0,2.1) H5"eq H I " 1"' 170.0 H2'" T" 2.00 (s) 20.8 a recorded at 500 MHz. b recorded at 125 MHz; exact assignments based on H M Q C and H M B C data. * unambiguous assignments not possible due to signal overlap. 84 85 2.2.2.6. Caribaeoside OH Me N N i O H 13 73 Caribaeoside (73) was isolated as a colourless glass that gave an [ M + H ] + ion in the H R F A B M S at m/z 673.33474 appropriate for a molecular formula of C35H48N2O11 ( A M 1.64 ppm). This molecular formula differed from the value obtained for eleutherobin (61) by the gain of sixteen mass units, indicating the presence of an additional oxygen atom. Detailed analysis of the N M R data revealed that caribaeoside (73) was a diterpene glycoside possessing the same A^-(6')-methylurocanic acid ester and 2"-0-acetylarabinose subunits present on the diterpene core of eleutherobin (61). However, it was evident that the central core of caribaeoside (73) differed from eleutherobin (61) in the C - l 1 to C-13 region. In comparing the ' H N M R spectra of 73 and eleutherobin (61), two characteristic features were missing: the C - l 7 olefinic methyl resonance at 6 1.47 ppm and the H-12 olefinic methine signal at 8 5.27 ppm. The 2D N M R data for caribaeoside (73) revealed these signals were replaced by a "methyl singlet resonance at 8 0.82 ppm and a pair of overlapping olefinic methine resonances around 8 5.52 ppm. Several key H M B C correlations from the methyl singlet at 8 0.82 ppm defined the changes observed in caribaeoside (73) (see Figure 2.23). This Me-17 singlet showed correlations into an olefinic carbon resonance at 8 137.4 ppm, a quaternary carbon resonance at 8 68.3 ppm, and a methine resonance at 8 45.7 ppm. The olefinic carbon resonance was linked to one of the protons at 8 5.52 ppm by an H M Q C 86 correlation, the quaternary carbon resonance suggested a hydroxyl substituent (consistent with the molecular formula), and the methine carbon bore a proton resonating at 8 2.07 ppm in the ' H N M R spectrum. This methine proton showed C O S Y correlations into a neighbouring methine proton at 8 4.00 ppm, corresponding to H - l of eleutherobin (61), and into two methylene protons, corresponding to H-9ab. The second olefinic proton at 8 5.53 ppm led to a series of C O S Y correlations that ultimately defined an isopropyl group from two overlapping methyl doublets at 8 0.94-0.95 ppm. Thus, the resulting structure for compound 73 maintained the C - l , C-2, CIO, and C14 centers of eleutherobin (61), shifted the cw-olefin to C-12 and C-13, and added a hydroxyl substituent to C - l 1. 5.52 5.53 5.52 5.53 H M B C C O S Y Figure 2.23. Selected H M B C and C O S Y correlations for the cyclohexene ring subunit of compound 73. The relative stereochemistry about the cyclohexene ring of caribaeoside was determined by analysis of R O E S Y and scalar coupling constant data. R O E S Y correlations were observed between the isopropyl methyl proton resonances (Me-19 and Me-20) and the proton resonances at 8 4.00 and 8 2.07 ppm ( H - l and H-10, respectively; see Figure 2.24). This demonstrated that the isopropyl group, H - l , and H-10 were on the same face of the caribaeoside molecule, as in eleutherobin (61). Additional R O E S Y correlations between H-8 and both H-10 and Me-16, and between Me-16 and OMe-21 confirmed that caribaeoside (73) and eleutherobin (61) had 87 identical relative stereochemistries at these centers. The Me-17 resonance at 8 0.82 ppm showed a strong R O E S Y correlation to the H-2 proton resonance, indicating the Me-17 and H-2 lie cis to each other. Typically, the methyl resonance would be expected at ~ 8 1.2 ppm, however, models of caribaeoside (73) show that the methyl group can sit in the shielding region of the A 2 , 3 olefin, resulting in a significant upfield shift for the methyl proton resonance (8 0.82 ppm). Finally, the chemical shifts and vicinal coupling constant of H - l and H-2 in caribaeoside (73) were nearly identical to those seen in eleutherobin (61), indicating the dihedral angle between these protons was essentially the same for both compounds. A summary of all N M R assignments and correlations for caribaeoside (73) can be found in Table 2.6. Figure 2.24. Selected R O E S Y correlations for the diterpene core of compound 73. Table 2.6. N M R data for caribaeoside (73) recorded in DMSO-c/6. position 5 'H a (7 value in Hz) 6 1 3 C b C O S Y Correlations H M B C Correlations 1 4.00 (m) 33.8 H2, 10 2 5.38 (d, 7=9.7) 136.3 HI 3 131.3* H15a/15b 4 115.4 H4-OMe 4-OMe 3.08 (s) 49.0 5 6.13 (d, 7=5.5) 130.9 H6 6 6.28 (d, 7=5.5) 133.8 H5 H16 7 90.0 H16 8 4.85 (d, 7=7.5) 78.7 H9b H16 9a 1.56 (rn) 29.1 H9b, 10 b 1.38 (m) H9a, 8, 10 10 2.07 (m) 45.7 H9a/9b, 2 H17 11 68.3 H17 11-OH 4.32 (s) 12 5.52 (s) 137.4 H17 13 5.53 (d, 7=6.3) 125.6 H14 14 1.54 (m) 46.5 H13, 18 H19, 20 15a 4.14 (d, 7=12.4) 68.4 H15b b 3.71 (d, 7=12.4) H15a 16 1.33 (s) 24.1 17 0.82 (s) 25.0 18 1.68 (m) 32.1 H19, 20, 14 19 0.94 (d, 7=6.8) 21.1 H18 20 0.95 (d, 7=6.8) 21.8 H18 1' 165.6* H2' 2' 6.36 (d, 7=15.5) 113.5 H3' 3' 7.52 (d, 7=15.5) 137.7 H2' 4' 136.9* H5' 5' 7.57 (s) 124.6 6'-NMe 3.66 (s) 33.2 7' 7.69 (s) 140.0 6'-NMe 1" 4.76 (d, 7=3.3) 94.4 H2" 2" 4.82 (m) 71.1 HI" , 3"* 3" 3.74 (m) 68.6 H2"*, 3"-OH*, 4"* 3"-OH 4.84 (br.d)* H3"* 4" 3.76 (m) 66.3 H3"*, 4"-OH* 4"-OH 4.92 (d, 7=5.8)* H4"* 5"eq 3.59 (d, 7=12.4) 63.1 H5"ax ax 3.47 (m) H5"eq 1"' 170.1 H2'" 2"' 2.03 (s) 20.9 a recorded at 500 MHz. b recorded at 125 MHz; exact assignments based on H M Q C and H M B C data. * unambiguous assignments not possible due to signal overlap or poor signal to noise. 90 Figure 2.26. Expanded R O E S Y spectrum of caribaeoside (73) recorded in DMSO-J6 at 500 M H z . 91 2.2.2.7. Caribaeolin Me N 20 ,OAc 19 13 74 Caribaeolin (74) was isolated as a clear oil that gave an [M + H ] + ion in the H R F A B M S at m/z 541.29111 appropriate for a molecular formula of C 3 0 H 4 0 N 2 O 7 ( A M -0.49 ppm). This molecular formula differed from the value obtained for caribaeoside (73) by the loss of 132 mass units, indicating a loss of C 5 H 8 0 4 . Detailed analysis of the N M R data revealed that caribaeolin (74) was a diterpenoid possessing a Af-(6')-methylurocanic acid ester subunit attached to the diterpene core of caribaeoside (73) but was missing the 2"-0-acetylarabinose. Several key H M B C correlations from the protons of a methylene group at 5 4.46 ppm defined the changes observed in caribaeolin (74) (see Figure 2.27). This methylene group showed correlations into an olefinic carbon resonance at 5 135.7 ppm, a quaternary olefinic carbon resonance at 8 131.4 ppm, and a carbon at 8 115.6 ppm that suggested a tetrahedral carbon deshielded by two oxygen atoms of a ketal. Further inspection of the H M Q C experimental data revealed this C H 2 group was attached to an oxygen atom, as expected by the carbon chemical shift of 65.0 ppm. Finally, H M B C correlations were observed between a carbonyl resonance at 8 169.9 ppm and both the methylene protons at 8 4.46 ppm and a methyl singlet resonance at 8 1.97 ppm, thus establishing the presence of a C - l 5 acetyl substituent for caribaeolin (74) in place of the C-l5 arabinose sugar residue found in caribaeoside (73). Figure 2.27. Selected H M B C correlations around C-15 of compound 74. The relative stereochemistry of caribaeolin (74) was determined by analysis of R O E S Y data and found to be identical to caribaeoside (73). The Me-17 resonance at 5",0.77 ppm showed a strong R O E S Y correlation to the H-2 proton resonance, indicating the Me-17 and H-2 lie cis to each other as in caribaeoside (73) (see Figure 2.28). R O E S Y correlations were also observed between the isopropyl methyl proton resonances (Me-19 and Me-20) and the proton resonances at 5 4.01 and 5 2.07 ppm ( H - l and H-10, respectively). This demonstrated that the isopropyl group, H - l , and H-10 were on the same face of the caribaeolin molecule, as in eleutherobin (61). Additional R O E S Y correlations between H-8 and both H-10 and Me-16, and between Me-16 and OMe-21 confirmed that caribaeolin (74) and caribaeoside (73) share the same relative stereochemistry. A l l the 2D N M R data was in complete agreement with the structure proposed for 74 and a summary of all N M R assignments and correlations for caribaeolin (74) can be found in Table 2.7. R O E S Y Figure 2.28. Selected R O E S Y correlations for the diterpene core of compound 74. Table 2.7. N M R data for caribaeolin (74) recorded in DMSO-r/6. position 5 ! H a 5 1 3 C b COSY H M B C (J value in Hz) Correlations Correlations 1 4.01 (m) 33.4 H2, 10 H9a*, 14* 2 5.38 (d, 7=10.0) 135.7 HI H15a/15b 3 131.4 H15a/15b 4 115.6 H2, 5,6, 15a/15b, 4-OMe 4-OMe 3.08 (s) 49.1 5 6.16 (d, 7=5.9) 130.4 H6 H6 6 6.34 (d, 7=5.9) 134.6 H5 H5, 8, 16 7 90.2 H5, 6, 8, 161 8 4.85 (d, 7=7.7) 78.4 H9b H16 9a 1.55 (m) 29.2 H9b, 10 b 1.41 (m) H9a, 8, 10 10 2.07 (m) 45.5 H9a/9b, 2 H8, 17 11 68.3 H17 11-OH 4.31 (s) 12 5.53 (m) 137.3 H17 13 5.53 (m) 125.7 H14 14 1.54 (m) 46.2 H13 H12*, 13*, 19, 20 15a 4.46 (s) 65.0 H15b* H2 b 4.46 (s) H15a* 16 1.34 (s) 23.8 17 0.77 (s) 24.8 11-OH 18 1.69 (m) 32.0 H19, 20 H19, 20 19 0.95 (d, 7=6.6) 20.9 H18 H20 20 0.96 (d, 7=6.6) 21.8 H18 H19 r 165.6 H8, 3' 2' 6.35 (d,.7=15.7) 113.6 H3' 3' 7.53 (d, 7=15.7) 137.8 H2' 4' 136.9 H5 ' , 7 ' 5' 7.56 (s) 124.6 6'-NMe 6'-NMe 3.66 (s) 33.1 7' 7.68 (s) 140.0 6'-NMe 1"' 169.9 2"' 1.97 (s) 20.6 a recorded at 500 MHz. b recorded at 125 MHz; exact assignments based on HMQC and HMBC data. * unambiguous assignments not possible due to signal overlap or poor signal to noise. 9 4 Figure 2.30. Expanded R O E S Y spectrum of caribaeolin (74) recorded in D M S O - J 6 at 500 M H z . 96 2.2.2.8. Diacetyleleutherobin OAc Me N i OAc 75 Diacetyleleutherobin (75) was formed from three separate acetylation reactions performed on eleutherobin (61), desacetyleleutherobin (70), and isoeleutherobin A (71). These compounds were acetylated with acetic anhydride in pyridine in order to confirm the identity of the arabinose sugar moiety. The discussion presented below specifically refers to data from acetylated eleutherobin (61), however, a comparison of the I D and 2 D N M R data revealed all the acetylation products were identical. Diacetyleleutherobin A (75) was isolated as a white amorphous solid that gave an [ M + H ] + ion in the H R F A B M S at m/z 741.33834 appropriate for a molecular formula of C39H52N2O12 ( A M -1.58 ppm). This molecular formula differed from the value obtained for eleutherobin (61) by the gain of 84 mass units, corresponding to the gain of two C 2 H 2 O units.. The ' H N M R spectrum of 75 differed from the spectrum of eleutherobin (61) by the presence of two additional singlet methyl resonances at 8 2.08 and 8 1.95 ppm and the absence of two hydroxyl signals. In addition, the signals attributed to H-3" and H-4" on eleutherobin (8 3.73-3.76 ppm) were replaced by two overlapping signals further downfield at 8 5.22 ppm. The M S and N M R data suggested that 75 was the fully acetylated analogue of eleutherobin (61). Detailed analysis of the N M R experimental data confirmed the presence of two additional acetate groups on the arabinose sugar at the C-3" and C-4" positions. The C O S Y 97 spectrum showed homonuclear correlations from the anomeric proton at 8 4.87 ppm to the neighbouring 2" proton at 8 4.96 ppm (see Figure 2.31). Likewise C O S Y correlations were observed between this axial 2" proton and the deshielded H-3" resonance at 8 5.22 ppm. The splitting pattern observed in the ! H N M R spectrum for H-2" was consistent with axial-equatorial coupling to H - l " (7= 3.4 Hz) and axial-axial coupling to H-3" (J= 10.7 Hz) . o C O S Y H M B C Figure 2.31. Selected C O S Y and H M B C correlations for the arabinose subunit of compound 75. The H M B C data showed correlations from the axial 5" methylene proton at 8 3.60 ppm to both the anomeric C - l " (8 92.9 ppm) and the neighbouring C-4" (8 66.5 ppm), whereas the equatorial H - 5 " at 5 3.85 ppm displayed correlations into C - 3 " (8 68.6 ppm). The ' H N M R chemical shifts of the arabinose protons (H-2", H-3" , and H-4") showed they all experienced the significant deshielding effects of an electron-withdrawing acetyl group. A summary of all the N M R assignments and correlations for diacetyleleutherobin (75) in DMSO-rf6 can be found in Table 2.8 and additional 2D spectra can be found in the Appendix. Table 2.8. N M R data for diacetyleleutherobin (75) recorded in D M S O - d 6 . Position 5 ' H a (7 value in Hz) 5 1 3 C b C O S Y Correlations H M B C Correlations 1 3.90 (m) 33.4 H2 2 5.43 (d, 7=9.5) 135.4 HI H15a 3 133.1 H15a 4 115.5 H5,6, 15b, 4-OMe 4-OMe 3.10 (s) 49.1 5 6.08 (d, 7=5.9) 130.5 H6 H6 6 6.30 (d, 7=5.9) 133.9 H5 H5, 16 7 89.3 H5, 6, 8, 16 8 4.66 (d, 7=7.6) 80.7 H9b H16 9a 1.51 (m) 30.9 H9b b 1.30 (m) H9a, 8, 10 10 2.47 (m) 38.3 H9b H8, 17 11 133.7 H17 12 5.30 (m) 121.0 H13b H17 13a .2.29 (m) 24.0 H13b b 1.96 (m) H13a, 12 14 1.15 (m) 42.0 H19, 20 15a 4.24 (d, 7=12.7) 67.9 H15b H2 b 3.85 (d, 7=12.7) H15a 16 1.38 (s) 24.0 17 1.48 (s) 21.4 18 1.49 (m) 28.8 H19, 20 H19, 20 19 0.93 (d, 7=7.0) 20.2 H18 H20 20 0.92 (d, 7=7.0) 21.9 H18 H19 r 166.0 H3' 2' 6.35 (d, 7=15.6) 113.5 H3' 3' 7.53 (d, 7=15.6) 137.9 H2' 4' 136.9 H 2 ' , 3 ' , 5 \ 7 ' 5' 7.56 (s) 124.8 6'-NMe 6'-NMe 3.66 (s) 33.2 7' 7.68 (s) 140.1 6'-NMe 1" 4.87 (d, 7=3.4) 92.9 H2" H15a/15b, 5"ax 2" 4.96 (dd, 7=3.4,10.7) 67.6 HI" , 3" 2"-OAc 2.02 (s)* 169.8*, 20.6 H2'" 3" 5.22 (m) 68.6 H2" H5"eq 3"-OAc 2.08 (s)* 169.8*, 20.6 H3" ' 4" 5.22 (m) 66.5 H5"ax 4"-OAc 1.95 (s)* 169.6*, 20.4 H5"ax H4'" 5"eq 3.85 (d, 7=12.8) 60.2 H5"eq ax 3.60 (dd, 7=12.8,1.5) a recorded at 500 MHz. b recorded at 125 MHz; exact assignments based on H M Q C and H M B C data. * unambiguous assignments not possible due to signal overlap or poor signal to noise. 99 100 102 2.2.2.9. Known compounds Sarcodictyin A Me N "4 ,OMe N O 2(1 1 9 55 Sarcodictyin A (55) was isolated as a clear oi l that gave an [ M + H ] + ion in the H R F A B M S at m/z 497.26514 appropriate for a molecular formula of C 2 8H36N 2 0 6 ( A M -0.04 ppm). Detailed analysis of the I D and 2D N M R data recorded in D M S O - J 6 identified the N-methylurocanic acid ester functionality and diterpene core common to the eleutherobins previously described. However, instead of a sugar substituent attached to C-15, there was evidence of a methyl ester. In addition, compound 55 displayed the cyclic hemiketal functionality also present in desmethyleleutherobin (69). The final structure of sarcodictyin A (55) was ultimately confirmed by comparison of spectra obtained in pyridine-t/5 with previously published data. 5 0 A complete assignment of the ' H and 1 3 C resonances can be found in the Experimental Section. 103 Erythrolide A and E Erythrolide A (36) was isolated as a white, amorphous solid that gave an [ M ] + ion in the H R E P M S at m/z 538.15613 appropriate for a molecular formula of C 2 6 H 3 i O 1 0 C l ( A M -0.34 ppm). Detailed analysis of the I D and 2D N M R data recorded in C D C 1 3 identified three acetates, one ketone, a y-lactone, two double bonds, and a single chlorine atom. The characteristic signals of an exocyclic olefinic methylene in the ! H N M R spectra and a shielded quaternary carbon in the C N M R spectra suggested one of the previously isolated erythrolides from E. caribaeorum. The final structure of erythrolide A (36) was ultimately confirmed by comparison of spectra with previously published data. 7 ' 4 1 A complete assignment of the ' H and 1 3 C resonances can be found in the Experimental Section. Erythrolide E (41) was isolated as a white, amorphous solid that gave an [ M ] + ion in the H R E I M S at m/z 496.15080 appropriate for a molecular formula of C24H29O9CI ( A M -0.26 ppm). Detailed analysis of the ID and 2D N M R data recorded in C D C I 3 indicated the presence of two acetates, one ketone, a y-lactone, one double bond, and a chlorine atom.. The final structure of erythrolide E (41) was ultimately confirmed by comparison of spectra with previously published data. 3 9 A complete assignment of the ' H and l 3 C resonances can be found in the Experimental Section. 104 2.2.3. Biological Activity In the search for biologically active secondary metabolites from marine and terrestrial sources, over 24,000 crude extracts were screened for antimitotic activity using the new cell-based assay developed in the laboratory of Michel Roberge. 6 4 The majority of antimitotic agents known at this time had been discovered by serendipity, by cytotoxicity screening, or by displaying a similar activity profile in a 60 cell-line panel to other known antimitotic agents. Roberge's newly developed, rational drug screen identified 119 extracts that demonstrated significant antimitotic activity. These extracts were then rescreened using tubulin immunofluorescence microscopy 6 5 to examine their effects on microtubule structure. The crude methanolic extract of a Caribbean octocoral, Erythropodium caribaeorum, was selected for further investigation based on the morphological characteristics displayed by cells treated with this extract. The E. caribaeorum extract produced paclitaxel-like bundling of microtubules that suggested a similar mechanism of action. At this point, the only antimitotic agents known to exhibit this mechanism of action were paclitaxel (4), the epothilones (52, 53), and discodermolide (17). A normal cell that has not been exposed to an antimitotic agent morphologically displays a complete, fibrous-like network of microtubules that permeates throughout the cell. Cells treated with an agent such as paclitaxel (4) that stabilizes microtubules against depolymerization exhibit distinctive bundling and stacking of the microtubules. 6 6 Figure 2.35 shows A549 cells treated with eleutherobin (61) and visualized using tubulin immunofluorescence microscopy. The appearance of these microtubules is consistent with the morphology of microtubules treated with paclitaxel (4). On the contrary, cells treated with an antimitotic agent that promotes destabilization of microtubules, such as nocodazole, result in a diffuse, ubiquitous white staining of tubulin throughout the cell as a result of the impaired polymerization into microtubule fibers. 105 Figure 2.35. hnmunofluorescence photograph of A549 cells after exposure to eleutherobin (61) at a concentration of 60 ng/ml. 6 7 The antimitotic assay was found to be well-suited for bioassay-guided fractionation of the crude E. caribaeorum extract and for quantitative evaluation of antimitotic activity of the pure compounds isolated. This is a useful measure of a compound's pharmacological potential since it not only evaluates the interaction of a compound and its target, as an in vitro target-based assay would do, but also its ability to permeate the cell and interact with its target inside the environment of a cell. For the bioassay, human breast cancer MCF-7 cells were treated at various concentrations of the compounds for 20 hours and the extent of mitotic arrest was determined by ELICA. The activity profiles of eleutherobin (61) and the six related diterpenes (69-74), as well as diacetyleleutherobin (75) and sarcodictyin A (55), were determined and appear in Figure 2.36. 106 100 80 60 ' 40 20 o t ^ ^ - f f . , r 0.01 0.1 1 10 100 Concentration ((iM) 0.01 0.1 1 Concentration (uM) 10 #— Eleutherobin (61) *?— Desmethyleleutherobin (69) •0— Desacetyleleutherobin (70) Isoeleutherobin A (71) Z-Eleutherobin (72) -•—Caribaeoside (73) — Caribaeolin (74) •—Diacetyleleutherobin (75) 6— Sarcodictyin A (55) Figure 2.36. Antimitotic activity of the compounds isolated from E. caribaeorum. Absorbance values were converted to percentage of mitotic cells using a standard curve. 6 8 Bars, S D . The results of the antimitotic activity of compounds 61, 69-75, and 55 revealed important information on the structural features required for biological activity. Analysis of the I C 5 0 values obtained from the graphs (see Table 2.9) showed eleutherobin (61), desmethyleleutherobin (69), diacetyleleutherobin (75), and isoeleutherobin A (72) all displayed very similar levels of biological activity, with values ranging from 20-50 n M . This suggested the hydroxyl substitution on the cyclic ketal of eleutherobin (61) and the acetyl group position on the arabinose sugar do not result in significant changes to the antimitotic activity of these compounds. A slight decrease in activity was observed, however, when the configuration of the C-2 ' ,3 ' olefin group was changed from E to Z as in Z-eleutherobin (72), perhaps reflecting the preference for a specific configuration to access the active site on the tubulin target. Similarly, removal of the acetate group found at the 2" position on the sugar moiety of eleutherobin (61) results in a slight loss of antimitotic activity, as shown by desacetyleleutherobin (71). This may 107 indicate an unfavourable interaction that occurs because of the presence of an additional hydroxyl group on the arabinose subunit. Table 2.9. Summary of antimitotic activity of the compounds isolated from E. caribaeorum. Compound i c 5 0 Structural changes from eleutherobin (61) (in nM) desmethyleleutherobin (69) 20 C-4 hydroxyl instead of methoxyl group eleutherobin (61) 30 diacetyleleutherobin (75)s 30 C-3" and C-4" acetyl groups on arabinose isoeleutherobin A (71) 50 C-3" acetyl group instead of C-2?' Z-eleutherobin (72) 250 Z-configuration of C-2 ' ,3 ' olefin desacetyleleutherobin (70) 400 C-2" hydroxyl instead of acetyl group sarcodictyin A(55) 2000 C-15 methyl ester instead of arabinose, C-4 hydroxyl instead of methoxyl group caribaeoside (73) 20000 C - l l hydroxyl and C-12,13 olefin caribaeolin (74) 20000 C - l l hydroxyl and C-12,13 olefin, C-15 acetate group instead of arabinose s synthetically acetylated More significant decreases in antimitotic activity were seen with the remaining compounds. In sarcodictyin A (55), the sugar residue is replaced with a methyl ester and a hydroxyl group replaces the C-4 methoxyl group, resulting in a 100-fold decrease in biological activity. Since desmethyleleutherobin (69) was fully active, it is the C-15 linked 2"-0-acetyl-arabinopyranose side chain and not the C-4 methoxyl group that was required for activity. The most significant loss in antimitotic activity in this series of compounds was associated with the caribaeoside (73) and caribaeolin (74) structures. The introduction of a hydroxyl group at C - l 1 and migration of the olefin group to the A 1 2 ' 1 3 position resulted in nearly a 1000-fold decrease in antimitotic potency relative to eleutherobin (61). This change altered both the shape and polarity of the diterpene core and revealed for the first time the critical importance of the C - l l to C - l 3 region for microtubule-stabilizing activity. The additional replacement of the arabinose 108 fragment in caribaeoside (73) with a simple acetate residue, as in caribaeolin (75), resulted in no further loss of activity. Experiments on MCF-7 cells showed the cytotoxicity I C 5 0 values for all these compounds corresponded to their antimitotic I C 5 0 values, suggesting cell death was ultimately a result of the overstabilization of microtubules. Indeed, cells treated with eleutherobin (61) for one hour and then washed clean of the drug would eventually all arrest in mitosis and die, suggesting irreversible damage to its microtubule targets. Results of the antimitotic activity of eleutherobin (61) and desmethyleleutherobin (69) against a multi-drug resistant ( M D R ) cell line indicated a greater than 10-fold drop in antimitotic activity, consistent with previous results reported for eleutherobin (61).57 This M D R phenotype is associated with the overexpression of P-glycoprotein, an energy-dependent drug efflux pump that maintains a low intracellular drug concentration. 6 9 Since the discovery of eleutherobin (61) by Lindel et al. in 1997,55 there have been many synthetic analogues that have been prepared as part of structure-activity relationship (SAR) studies. 7 0 ' 7 1 However, to date they have all been based on the eleutherobin/sarcodictyin diterpenoid core. Recently, Ojima et al. proposed a common pharmacophore for microtubule-stabilizing compounds that accommodates the paclitaxel (4), eleutherobin (61), epothilone (52, 53), and discodermolide (17) structural classes. 7 2 Their model used molecular dynamics to predict three areas of common overlap that are important domains for binding to tubulin (labeled A, B, and C in Figures 2.37 and 2.38). Figure 2.37. Overlay of nonataxel (cyan) with a) paclitaxel b) epothilone B c) eleutherobin and d) discodermolide, all in yel low. 7 2 110 The antimitotic compounds isolated from E. caribaeorum (61, 69-74, 55) appeared to agree with the proposed pharmacophore model. Changes in the A region identified by Ojima, such as altering the A 2 , 3 ' configuration in Z-eleutherobin (72), resulted in a small decrease in activity. Changes in the arabinose fragment, representing alterations in the C region of the pharmacophore, resulted in a little change (i.e., isoeleutherobin A (71)) all the way to a significant decrease in activity (i.e., sarcodictyin A (55)). Replacement of the methoxyl group by a hydroxyl group (i.e., desmethyleleutherobin (69)) had no real effect on the antimitotic potency and correspondingly, this alteration formally exists outside the proposed tubulin binding regions. The pharmacophore presented by Ojima suggested changes in the C - l l to C-13 region of eleutherobin should have an impact on the ability of analogues to act as microtubulin stabilizers. Indeed, caribaeoside (73) is the first such analogue ever tested in an antimitotic assay and its significant loss of activity associated with the introduction of a hydroxyl group at C - l l and migration of the olefin to A 1 2 1 3 further strengthens the pharmacophore model. Although these structural analogues were found to support Ojima's proposal, the creation of such a predictive pharmacophore model depends on comprehensive knowledge about the conformation of all the known microtubule-stabilizing compounds. A t the time this model was proposed, there had been no published solid-state or solution conformational analyses of eleutherobin (61) and the molecular dynamics calculations used to define its lowest energy conformation were carried out without any N O E constraints. 2.2.4. Solid-State and Solution Conformations During the course of characterizing the antimitotic diterpenes from E. caribaeorum, several crystals of eleutherobin (61) were obtained by the fortuitous slow evaporation of a concentrated N M R sample dissolved in D M S O - J 6 . The crystals proved to be suitable for X-ray I l l 73 diffraction analysis and provided the first solid-state conformation for eleutherobin (61). ' The computer-generated O R T E P for eleutherobin (61) is shown in Figure 2.39. 08 C20 Figure 2.39. O R T E P drawing of eleutherobin (61). Hydrogen atoms have been omitted for clarity. In order to facilitate a comparison of the solid-state conformation of eleutherobin (61) with its solution conformation, a series of R O E S Y and difference N O E experiments were performed in both a polar and nonpolar solvent. N M R samples were prepared in both D M S O -d6 and CDCI3 and then degassed, purged of oxygen, and subsequently flushed with argon and the tubes sealed. The difference N O E results obtained in C D C 1 3 and the R O E S Y results collected in CDCI3 and DMSO-c/6 are found in Table 2.10 along with the solid-state internuclear distances between protons measured from the X-ray diffraction analysis. 1 12 Table 2,10. Difference N O E a and R O E S Y b data for eleutherobin (61) in C D C h and D M S O - J 6 . H- Position H- Position Internuclear Difference ROESY ROESY (5 in CDCI.0 (5 in CDCI.0 Distance from NOE ( % correlations correlations X-ray data (A) enhancement)0 (CDCh) H2 (5.54) "Hl7a'(2.29) a~ 2.12 7.49 y y H2 H14 (1.21) 2.61 3.51 y y H8 (4.80) HI (3.94) 2.44. 4.28 y y H8 H10 (2.60) 2.30 6.16 y y H8 H16 (1.43) 2.52 0.91 • y y H8 H19(0.96) 2.64 0.52 y n H10(2.60) HI (3.94) 2.33 4.79 y y H10 H8 (4.80) 2.30 2.16 y y H10 HI9 (0.96) 2.51 0.77 . n n H13cc(2.29)d H2 (5.54) 2.12 9.69 y y H14 (1.21) HI (3.94) 2.37 6.40 y y H19 (0.96) HI 2.02 0.97 y V H19 H8 (4.80) 2.64 0.32 y n H19 H10 (2.60) 2.51 0.42 n n H-20 (0.94) H13f3(1.97)d 2.03 0.96 y y OMe (3.20) H16 (.1.43) 2.28 1.01 y y a Acquired at 400 MHz . b Acquired at 500 MHz. c ' H in first column was irradiated. d p and a are defined as being above and below the plane of diterpenoid core as in the structural representations of 61. y = ROESY correlation clearly observed, n = ROESY correlation not clearly observed. Examination of the solid-state conformation represented in Figure 2.40 revealed the presence of a twist in the cyclohexene ring relative to the remaining diterpene core of eleutherobin (61). Analysis of a number of key NOEs indicated that the solution conformation of the diterpenoid core in both the polar and nonpolar solvent was extremely similar, if not identical, to the solid-state conformation. In particular, irradiation of the H - l 3 a resonance resulted in a significant enhancement of the H-2 resonance, highlighting the close proximity of these protons that was also evident in the solid-state conformation. Detailed analysis of this X -ray diffraction experiment revealed a torsional angle of -61° for the C2-C1-C14-C13 bonds, thus o placing H-13a a distance of only 2.118 A from H-2 and consistent with the strong difference N O E observed (see Figure 2.40). Additionally, irradiation of the H-8 resonance resulted in 113 noticeable enhancements in the H-10, H - l , and Me-19 resonances, in full accord with the spatial arrangement of the isopropyl group relative to H-8 seen in the solid-state conformation. The X -ray diffraction analysis gave a torsional angle of -66° for C8-C9-C10-C1 which placed H-8 a distance of 2.304 A from H-10 and 2.438 A from H - l , in good agreement with the solution difference N O E s observed. Furthermore, it was possible to explain the weak difference N O E present between the Me-19 proton resonances and the H-8 resonance by correlation to a solid-state C9-C10-C1-C14 torsional angle of 177.7°. Figure 2.40. Torsional angles around the cyclohexene ring of eleutherobin (61). o Distances are in A . 114 115 Figure 2.43. ROESY spectrum of eleutherobin (61) recorded in CDC13 at 500 MHz. 117 Figure 2.44. R O E S Y spectrum of eleutherobin (61) recorded in D M S O - J 6 at 500 M H z . 118 Taken together, the combination of the H-2 to H - l 3 a and the.H-8 to H-10, H - l , and Me-19 difference N O E s represent an extremely restrictive set of conformational restraints. The interproton distances necessary for the observed N O E s between these protons are extremely difficult to attain with any significant changes from the solid-state conformation. In addition, the 2D R O E S Y experiments in C D C 1 3 and D M S O - d 6 revealed the corresponding correlations between H-2 to H - l 3 a and the H-8 to H-10, and H - l , providing further evidence of a major contributing solution conformation for the diterpenoid core that is essentially identical to the solid-state conformation. In light of details provided by the solid-state and solution conformational analyses of eleutherobin (61), the pharmacophore proposed by Ojima et al. for microtubule-stabilizing agents was re-evaluated. As mentioned earlier, the molecular dynamics calculations used to obtain a conformation for eleutherobin (61) were performed without using any torsional angle or 72 N O E constraints. Analysis of the eleutherobin conformation presented in their paper indicated a torsional angle approaching 180° for the C2-C1-C14-C13 bonds and a torsional angle greater than -90° for the C8-C9-C10-C1 substructure (see Figure 2.45). Such a conformation would not be expected to show the difference N O E ' s or R O E S Y correlations that were described earlier. Indeed, a significant difference exists in the spatial orientation of the C11-C12-C13 tubulin binding region identified and illustrated by Ojima compared to the solid-state and solution conformation present here. This key difference in the diterpene core conformation relative to the urocanic acid side chain and arabinose moiety of eleutherobin (61) suggests the solid-state and solution conformation would display dramatically different overlay fits with the other microtubule-stabilizing structures used to construct the Ojima pharmacophore. 119 Figure 2.45. Comparison of the conformations of eleutherobin (61). Left: the Ojima pharmacophore conformation (yellow overlayed with nonataxel in cyan); Right: the solid-state conformation obtained by X-ray diffraction analysis. Recently, two papers have been published proposing alternate pharmacophore models for the microtubule-stabilizing compounds. 7 4 ' 7 5 Both reports are based on studies involving the taxane and/or epothilone classes of antimitotic agents and extend their findings to the eleutherobin/sarcodictyin group. B y evaluating the biological activity of a taxane analogue and then using molecular modeling studies to investigate its binding to tubulin, He et al. observed that the C - l 3 side chain of paclitaxel (4) is not an absolute requirement for biological activity and that the C-2 side chain can play a critical role in the binding to tubulin. The authors proposed that the 14-member tricyclic ring system of eleutherobin (61), like the macrolide ring of the epothilones (i.e., 53), overlays with the taxane ring of paclitaxel (4) and that the urocanic acid side chain, like the thiazole side chain of the epothilones, mimics the C-2 side chain in paclitaxel (see Figure 2.46). 7 4 120 53 O H 61 Figure 2.46. Proposed common overlap (blue and red) for paclitaxel (4), epothilone B (53), and eleutherobin (61). A n alternate pharmacophore model was presented in the work of Giannakakou et al who investigated the binding of taxane and epothilone analogues on (3-tubulin from an epothilone-resistant cell l ine . 7 5 Molecular modeling studies of different conformations and docking studies on tubulin identified one preferred conformation for the epothilones. The pharmacophore model proposed was consistent' with structure-activity studies of various analogues and predicted a single common theme for the taxanes, epothilones, and sarcodictyins (see Figure 2.47). The authors suggested the isopropyl group at C - l 4 of sarcodictyin A (55) corresponds to the gem-dimethyl groups of the epothilones (i.e., 53) and taxanes (i.e., 4) while the ether oxygen attached to C-4 and C-l correlates to the oxetane and epoxide oxygen atoms of the taxanes and epothilones, respectively. It is important to note that the sarcodictyin A 121 conformation presented in this model is consistent with the X-ray crystal structure and solution conformation presented earlier for eleutherobin (61). This model also shows the methylimidazole functionality of the sarcodictyins lies in the vicinity of the C - l 3 side chain of the taxanes and the aromatic thiazole side chain of the epothilones. 4 Me i 55 Figure 2.47. Proposed common overlap (blue) and pharmacophore (red) for paclitaxel (4), epothilone B (53), and sarcodictyin A (55). Since both these models were generated prior to the publication of this thesis research, it is anticipated that the availability of solid-state and solution conformation data for eleutherobin (61) wil l facilitate further refinements of the microtubule-stabilizing pharmacophore models and, consequently, aid the eventual design of new antimitotic compounds. 122 The sample of eleutherobin (61) isolated and purified from Erythropodium caribaeorum was subjected to further solution conformational analysis in D M S O - J 6 and C D C 1 3 by researchers at Emory University in Atlanta. 7 6 Extensive ID and 2D N M R experiments were performed including five different R O E S Y spectra acquired with varying mixing times to determine the linearity of the cross-relaxation buildup. From these and additional experiments, a total of 40 intramolecular R O E distances and five side-chain three bond coupling constants ( 37('H, 1 3C)) were obtained. Although these measured quantities are averages derived from a rapidly equilibrating ensemble of conformations, the authors were able to deconvolute the average N M R spectra into constituent conformers using a combination of multiple force field conformational searching and the N A M F I S ( N M R analysis of molecular flexibility in solution) procedure. 7 7 ' 7 8 These results provided nine distinct and preferred conformations for eleutherobin (61) ranging in populations from 1-19%. The conformational diversity arose from the urocanic acid and arabinose side chains, with the nine-membered B-ring of the diterpene core existing as an invariant twist-boat-chair as seen in the solid-state (see Figure 2.48). The second most populated conformer obtained by this detailed analysis was the conformation observed by X-ray diffraction analysis (16%), again in full agreement with the solid-state and solution conformation determined in this thesis research. 123 Figure 2.48. Superposition of nine conformations of eleutherobin (61) as determined by N A M F I S analysis of R O E S Y and 3 7 ( ' H , I 3 C ) data. 7 6 The X-ray conformation is shown in red. a and b denote the two most populated arabinose locations. 2.2.5. Additional Results Since previous studies on octocorals have revealed significant geographic variation in the diterpenoid content, 7 9 evidence for the presence of eleutherobin (61) was investigated by H P L C and biological activity from two additional collections of Erythropodium caribaeorum. Crude methanolic extracts of E. caribaeorum from Florida and the Bahamas showed potent activity in the antimitotic assay and were subjected to bioassay-guided fractionation similar to the original isolation scheme. The antimitotic activity corresponded to the fractions expected for eleutherobin (61). The final normal phase H P L C purification step resulted in a small peak eluting at the time anticipated for standard eleutherobin and exhibited the same U V chromophore ( X m a x 290 nm). After collecting the entire H P L C run in a deep-well, 96-well plate, 124 the fractions were tested for antimitotic activity. The results confirmed activity associated with the fractions corresponding to eleutherobin (61). During the antimitotic screening of over 60,000 natural extracts obtained from the Open Repository Program of the N C I Developmental Therapeutics Program, it was found that eleven additional octocoral extracts exhibited activity in the assay. Since these extracts originated from organisms that were taxonomically related to Erythropodium caribaeorum, they might also contain related chemistry (see Figure 2.49). Sarcodictyin A , for instance, was first described from the Mediterranean stolonifer Sarcodictyon roseum by Pietra et al.,50 later isolated from a 53 South African alcyonacean along with the eleuthosides A and B by Kashman et al., and now reported from the Caribbean gorgonian E. caribaeorum (see Scheme 2.4 and Figure 2.50). Thus, the eleven additional active extracts may possess sarcodictyin or eleutherobin-type compounds. Phylum Class Order Family Genus Species Cnidaria Anthozoa Gorgonacea Anthothelidae Erythropodium caribaeorum Gorgoniidae Rumphella ? Rumphella sp. 8 Rumphella sp. e Isididae Mopsea whiteleggei Muricellisis sp. a Subergorgiidae Subergorgia sp. 1 cf. Mol l i s Subergorgia cf. Mol l i s Ellisellidae Junceella sp. d Verrucella sp. b Alcyonacea Alcyoniidae Sinularia sp. c Sinularia sp. h Figure 2.49. Taxonomy of the antimitotic extracts from the N C I Open Repository relative to E. caribaeorum. PHYLUM Porifera (sponges) Cnidaria Mollusca (nudibranchs) Chordata (tunicates) CLASS O R D E R Hydrozoa Scyphozoa Cubozoa (hydras) (jellyfish) I Stolonifera Sarcodictyins Anthozoa (sea anemones, corals, sea fans) Subclass Octocorallia Alcyonacea (soft corals) Sarcodictyins Eleuthosides Eleutherobin Gorgonacea (sea fans) (sea whips) Pennatulacea (sea pens) Sarcodictyin A Eleutherobins Caribaeoside/Caribaeolin Scheme 2.4. Taxonomic scheme showing the related chemistry isolated from octocorals. J ) IB Figure 2.50. Map of the geographic collection locations of sarcodictyin/eleutherobin-producing organisms: A : sarcodictyins A - F from a Mediterranean stolonifer; 5 0 ' 5 1 B: sarcodictyin A and eleuthosides A and B from a South African alcyonacean; C : eleutherobin from a Western Australian alcyonacean; 5 5 D: sarcodictyin A , eleutherobin, and analogues from a Caribbean gorgonian. 8 0 126 In order to determine i f these extracts possessed compounds similar to eleutherobin or the sarcodictyins, a similar bioassay-guided fractionation was performed on the 20 mg samples provided by the N C I . Each sample was subjected to reversed-phase liquid chromatography using the same stepwise gradient elution scheme as with the large E. caribaeorum purification (80:20 H 2 0 / M e O H to M e O H in 10 % increments). As expected for eleutherobin or sarcodictyin-type compounds, the antimitotic activity was found to almost exclusively reside in the 30:70 H 2 0 / M e O H fraction for each of the extracts. Thus, the polarity of the unknown active compounds seemed to agree with the observed polarity range of eleutherobin (61). The active fractions were then subjected to normal phase H P L C using the solvent system developed for eleutherobin (7:93 M e O H / C H 2 C l 2 ) and the entire run was collected in deep-well, 96-well plates and submitted for bioassay. The results showed antimitotic activity co-eluting with the solvent front for each of the extracts except one. This lack of retention on the column would be consistent with the behavior of the sarcodictyins, however, the solvent system was optimized for eleutherobin (61) and thus did not indicate the presence of that compound. One extract, from the gorgonian Subergorgia cf. Mol l i s , did show activity in the fractions corresponding to the elution of eleutherobin (61) under these standardized conditions, but the scarcity of material prevented further characterization and identification of this active agent. Although subsequent requests for additional crude extract material from the N C I failed to provide sufficient material to establish the identity of the active agents, it appears as though eleutherobin (61) is also produced by the Subergorgia octocoral and sarcodictyin-type chemistry may be present in the other extracts. In addition to the in vitro antimitotic studies on the M C F - 7 human breast cancer cell line, there have been preliminary results from the in vivo studies of desmethyleleutherobin (71). This work by Dr. Peggy Olive of the B . C . Cancer Research Centre is currently investigating the effect of desmethyleleutherobin (69) on S C C V H tumours in C 3 H mice. These mice were 127 injected, either intraperitoneal or intravenous, with a single dose of desmethyleleutherobin (69) ranging from 10-50 mg/kg and 24 hours later, the tumour was excised and the cells were analysed. The preliminary results show that desmethyleleutherobin (69) kills 30% of the tumour cells regardless of whether they were close to or distant from the blood supply. While this work is currently ongoing, it does strengthen the anticancer therapeutic potential of desmethyleleutherobin (69) by showing in vivo efficacy. 2.2.6. Future Studies Examination of the antimitotic properties of the Erythropodium caribaeorum extract resulted in the discovery of eleutherobin (61) and a series of related antimitotic diterpenes. Their interesting biological properties along with the conformational studies have spawned numerous avenues for continued investigation. While caribaeoside (73) and caribaeolin (74) were the first compounds featuring key structural changes to the diterpene core to be evaluated for microtubule-stabilizing biological activity, there are many changes to the framework of eleutherobin (61) that remain to be explored. Many synthetic studies have created analogues of the urocanic acid and arabinose side chains but modifications to the C-4 to C-l ether bridge, the three olefinic groups, and the tricyclic nature of the diterpene core have yet to be investigated. In light of the recent pharmacophore model suggesting the importance of the isopropyl group, 7 5 alterations to this area of the eleutherobin/sarcodictyins may provide further insights into defining a common model for microtubule-stabilization. In order to better define the solution conformation of eleutherobin (61), additional studies using R O E S Y and three-bond coupling constants are continuing by the researchers at Emory University. Application of these solution and solid-state conformations to docking studies of eleutherobin (61) to its f3-tubulin target would provide an even greater understanding of the features required for activity. Studies are also in progress using isotopically-labeled 128 eleutherobin to investigate the nature and dynamics of the binding to its cellular target. These results wi l l prove to be invaluable in the refinement of pharmacophore models for microtubule-stabilization and the eventual design of new compounds with antimitotic activity. Adding to the discovery of a new and relatively abundant source of eleutherobin (61) from Erythropodium caribaeorum, obtaining sufficient crude extract material from the eleven related marine octocorals may confirm the presence of eleutherobin (61) in at least one additional source and may provide a series of new structural analogues that are beyond the imagination of synthetic chemistry. The isolation of eleutherobin from E. caribaeorum alone has revived the pharmaceutical interest in this exciting molecule and can provide sufficient material for clinical investigations. Successive collecting trips in the waters off Dominica have revealed the potential for the aquaculture of this organism, as was successfully implemented 81 with the bryostatin-producing Bugula neritina, due to the rapid regrowth of harvested 82 material. Comparisons of the amounts and types of compounds isolated from the new versus old growth may reveal new analogues, biosynthetic precursors, or greater quantities of the active agents. Indeed, continuing studies in our lab on E. caribaeorum have led to the isolation of possible biosynthetic intermediates in the pathway to eleutherobin (61) and have shown that the C-4 methyl ketal in eleutherobin arises from the isolation procedure. 8 3 2.3. Conclusions In searching for new antimitotic compounds that act by stabilizing microtubules against depolymerization, a recently developed, rational cell-based screen was used to isolate the known antimitotic agent eleutherobin (61) and six novel structural analogues (69-74) from a marine organism. The Caribbean octocoral Erythropodium caribaeorum proved to be a new and abundant source of eleutherobin, whose pre-clinical development had been impeded by its scarcity, and led to structure-activity studies on the various analogues that were also identified. 129 The structural variations in this group included caribaeoside (73), the first analogue to demonstrate the critical importance of the C - l l to C - l 3 region of the di terpenoid core, and offered key insights into proposed pharmacophore models for microtubule-stabilizing compounds. In addition, single crystal X-ray diffraction analysis and difference N O E experiments provided the first solid-state and solution conformations for eleutherobin (61), revealing a twist-boat-chair orientation for the diterpene core that has been further supported by molecular modeling and conformational analyses. Promising investigations on the in vitro and in vivo biological activities of these compounds are currently underway and studies on antimitotic natural extracts from marine organisms that are related taxonomically to E. caribaeorum have revealed strong evidence for the presence of eleutherobin and the related sarcodictyins in these organisms. Finally, the discoveries arising from investigation of the antimitotic activity of E. caribaeorum have renewed the pharmacological interest in eleutherobin (61) and opened the door to many new and important avenues of scientific investigation. REFERENCES 1 Col in , P .L . ; Arneson, A . C . In Pacific Invertebrates; Coral Reef Press: Beverly Hi l ls , 1995; pp. 63-79. 2 Faulkner, D.J . Nat. Prod. Rep. 2001,18, 1; and previous reports in this series. 3 Bayer, F . M . In The Shallow-Water Octocorallia of the West Indian Region; Martinus Nijhoff: The Hague, 1961; pp. 1-373. 4 Cairns, S. In Guide to the commoner shallow-water gorgonians (sea whips, sea feathers, and sea fans) of Florida, the Gulf of Mexico, and the Caribbean region; Voss, G . , Ed. ; University of Miami Sea Grant Program: M i a m i , 1976; pp. 8-10. 3 Will iams, G .C . In Coral reef octocorals; Durban Natural Science Museum: Durban, 1993; pp. 8-22. 130 6 Humann, P. In Reef Coral Identification; DeLoach, N . , Ed. ; New Wor ld Publications: Jacksonville, 1993; pp. 8-29. 7 Look, S.A.; Fenical, W. ; Van Engen, D . ; Clardy, J. / . Am. Chem. Soc. 1984,106, 5026. 8 Dewick, P . M . In Medicinal Natural Products: A Biosynthetic Approach; John Wiley & Sons: Toronto, 1997; pp. 178-179. 9 Rodriguez, A . D . Tetrahedron 1995, 51, 4571. 1 0 a) Herin, M . ; Tursch, B . Bull. Soc. Chim. Belg. 1976, 55, 707. b) Vanderah, D.J . ; Rutledge, N . ; Schmitz, F.J. ; Ciereszko, L .S . J. Org. Chem. 1978, 43, 1614. 1 1 a) Anjaneyulu, A .S .R . ; Rao, G . V . ; Rao, N . S . K . Ind. J. Chem., Sect. B. 1996, 35, 815. b) Anjaneyulu, A .S .R . ; Sagar, K . S . ; Rao, G . V . J. Nat. Prod. 1997, 60, 9. c) Duh, C - Y . ; Wang, S.-K.; Weng, Y . - L . ; Chiang, M . Y . ; Dai , C.-F. J. Nat. Prod. 1999, o~2, 1518. 1 2 a) Birch, A . J . ; Brown, W . V . ; Corrie, J.E.T.; Morre, B .P . J. Chem. Soc, Perkin Trans. 1 1972, 2653. b) Patil, V . D . ; Nayak, U .R . ; Dev, S. Tetrahedron 1973, 29, 341. c) Mattern, D . L . ; Scott, W . D . ; McDaniel , C . A . ; Weldon, P.J.; Graves, D . E . / . Nat. Prod. 1997, 60, 828. 1 3 Stierle, D . B . ; Carte, B . ; Faulkner, D.J . ; Tagle, B . ; Clardy, J. J. Am. Chem. Soc. 1980, 102, 5088. 1 4 a) Hyde, R . W . Thesis, University of Oklahoma, 1966. b) Bartholome, C. Thesis, Universite Libre de Bruxelles, 1974. c) Burk, J.E.; Van der Helm, D. ; Chang, C . Y . ; Ciereszko, L.S. Acta Crystallogr., Sect. B 1977, 33, 704. 1 5 Kennard, O.; Watson, D . G . ; R iva di Sanseverino, L . ; Tursch, B . ; Bosmans, R.; Djerassi, C. Tetrahedron Lett. 1968, 24, 2879. 1 6 Rodriguez, A . D . ; Ramirez, C ; Rodriguez, I.I. J. Nat. Prod. 1999, 62, 997. 1 7 Rodriguez, A . D . ; Ramirez, C. J. Nat. Prod. 2001, 64, 100. 1 8 Rodriguez, A . D . ; Gonzalez, E . ; Huang, S.D. J. Org. Chem. 1998, 63, 7083. 1 9 Rodriguez, A . D . ; Ramirez, C ; Rodriguez, I.I.; Barnes, C . L . J. Org. Chem. 2000, 65, 1390. 2 0 Rodriguez, A . D . ; Shi, J .G.; Huang, S.D. J. Org. Chem. 1998, 63, 4425. 2 1 a) Sheu, J . -H.; Sung, P.-J.; Su, J . -H.; Wang, G . - H . ; Duh, C . - Y . ; Shen, Y . - C ; Chiang, M . Y . ; Chen, I.-T. J. Nat. Prod. 1999, 62, 1415. b) Sheu, J . -H. ; Sung, P.-J.; Cheng, M . - C ; L iu , H . - Y . ; Fang, L . -S . ; Duh, C . - Y . ; Chiang, M . Y . J. Nat. Prod. 1998, 61, 602. c) Sung, P.-J.; Su, J . -H.; Wang, G . - H . ; L i n , S.-F.; Duh, C . - Y . ; Sheu, J . -H. J. Nat. Prod. 1999, 62, 457. d) Neve, J.E.; M c C o o l , B . J . ; Bowden, B . F . Aust. J. Chem. 1999, 52, 359. 131 " a) Sheu, J . -H.; Sung, P.-J.; Su, J . -H. ; L i u , H . - Y . ; Duh, C . - Y . ; Chiang, M . Y . Tetrahedron 1999, 55, 14555. b) Sung, P.-J.; Su, J . -H.; Duh, C . - Y . ; Chiang, M . Y . ; Sheu, J . -H. J. Nat. Prod. 1999,64,318. 2 3 Williams, D .E . ; Andersen, R.J . ; Parkanyi, L . ; Clardy, J. Tetrahedron Lett. 1987, 25, 5079. 2 4 a) Pordesimo, E .O. ; Schmitz, F.J. ; Ciereszko, L .S . ; Hossain, M . B . ; Van der Helm, D. 7. Org. Chem. 1991, 56, 2344. b) Kobaysashi, J.; Cheng, J.-F.; Nakamura, H . ; Ohizumi, Y . ; Tomotake, Y . ; Matsuzaki, T.; Grace, K.J .S . ; Jacobs, R.S.; Kato, Y . ; Brinen, L .S . ; Clardy, J. Experientia 1991, 47, 501. c) Shin, J.; Park, M . ; Fenical, W . Tetrahedron 1989, 45, 1633. 2 5 a) Look, S.A.; Fenical, W. ; Matsumoto, G . K . ; Clardy, J. 7. Org. Chem. 1986, 51, 5140. b) Harvis, C . A . ; Burch, M . T . ; Fenical, W . Tetrahedron Lett. 1988, 29, 4361. c) Roussis, V . ; W u , Z. ; Fenical, W. ; Strobel, S.A.; Van Duyne, G.D. ; Clardy, J. J. Org. Chem. 1990, 55, 4916. d) Mayer, A . M . S . ; Jacobson, P .B. ; Fenical, W. ; Jacobs, R.S.; Glaser, K . B . Life Sci. 1998,62, PL401. 2 6 a) Shin, J.; Fenical, W . J. Org. Chem. 1991, 56, 3153. b) Jacobson, P .B. ; Jacobs, R.S.J. Pharmacol. Exp. Ther. 1992, 262, 866. 2 7 Cobar, O . M . ; Rodriguez, A . D . ; Padilla, O.L . ; Sanchez, J .A . J. Org. Chem. 1997, 62, 7183. 2 8 a) Coval, S.J.; Cross, S.; Bernardinelli, G . ; Jefford, C . W . J. Nat. Prod. 1988, 51, 918. b) Shin, J.; Park, M . ; Fenical, W . Tetrahedron 1989, 45, 1633. 2 9 a) Subrahmanyam, C ; Kulatheeswaran, R.; Ward, R.S. 7. Nat. Prod. 1998, 61, 1120. b) Rodriguez, A . D . ; Ramirez, C . J. Nat. Prod. 2001, 64, 100; and references cited therein. 3 0 Grode, S.H.; James, T.R.; Cardellina, J .H. , U. J. Org. Chem. 1983, 48, 5203. 3 1 Hamann, M . T . ; Harrison, K . N . ; Carroll, A . R . ; Scheuer, P.J. Heterocycles 1996, 42, 325. 3 2 Keifer, P . A . ; Rinehart, K . L . ; Hooper, I.R. J. Org. Chem. 1986, 57, 4450. 3 3 a) Epifanio, R .D . ; Maia , L . F . ; Fenical, W . J. Brazil Chem. Soc. 2000, 77, 584. b) Fenical, W. ; Pawlik, J.R. Mar. Ecol.-Prog. Ser. 1991, 75, 1. c) Wylie , C.R. ; Paul, V . J . / . Exp. Mar. Biol. Ecol. 1989, 729, 141. d) Van Alstyne, K . L . ; Wylie , C.R. ; Paul, V . J . 7. Exp. Mar. Biol. Ecol. 1994, 775, 17. 3 4 Epifanio, R .D . ; Martins, D . L . ; Vil laca, R.; Gabriel, R. J. Chem. Ecol. 1999, 25, 2255. 3 5 Burkholder, P.R.; Burkholder, L . M . Science 1958, 727, 1174. 3 6 Schmidt, R .W. ; Ciereszko, L .S . Proc. Okla. Acad. Sci. 1968, 48, 258. 3 7 Kokke, W . C . M . C . ; Epstein, S.; Look, S.A.; Rau, G . H . ; Fenical, W. , Djerassi, C. J. Biol. Chem. 1984, 259, 8168. 132 3 S a) Ciereszko, L .S . ; Johnson, M . A . ; Schmidt, R .W. ; Koons, C . B . Comp. Biochem. Physiol. 1968, 24, 899. b) Kokke, W . C . M . C . ; Bohlin, L . ; Fenical, W. , Djerassi, C . Comp. Biochem. Physiol. 1981, B68, 281. c) Withers, N . W . ; Kokke, W . C . M . C . ; Fenical, W. , Djerassi, C. Proc. Natl. Acad. Sci. U.S.A. 1982, 79, 3764. 3 9 a) Pordesimo, E .O. ; Schmitz, F.J . ; Ciereszko, L .S . ; Hossain, M . B . ; van der Helm, D. J. Org. Chem. 1991, 56, 2344. b) Pordesimo, E .O. Diss. Abstr. Int. B 1990, 50, 4542. 4 0 a) Fenical, W. ; Pawlik, J.R. Mar. Ecol. Prog. Ser. 1991, 75, 1. b) Pathirana, C ; Fenical, W. ; Corcoran, E . ; Clardy, J. Tetrahedron Lett. 1993, 34, 3371. 4 1 Dookran, R.; Maharaj, D. ; Mootoo, B.S . ; Ramsewak, R.; McLean, S.; Reynolds, W.F . ; Tinto, W.F . J. Nat. Prod. 1993, 5f5, 1051. 4 2 Banjoo, D. ; Maxwel l , A . R . ; Mootoo, B .S . ; Lough, A . J . ; McLean, S.; Reynolds, W.F . Tetrahedron Lett. 1998, 39, 1469. 4 3 Maharaj, D. ; Pascoe, K . O . ; Tinto, W . F . J.Nat. Prod. 1999, 62, 313. 4 4 a) Huang, H . ; Forsyth, C F . Tetrahedron Lett. 1993, 34, 7889. b) Huang, H . ; Forsyth, C F . J. Org. Chem. 1995, 60, 2773. c) Tokunaga, Y . ; Yagihashi, M . ; Ihara, M . ; Fukumoto, K . J. Chem. Soc. Chem. Commun. 1995, 955. d) Tokunaga, Y . ; Yagihashi, M . ; Ihara, M . ; Fukumoto, K . J . Chem. Soc. Perkin Trans. 1 1997, 189. 4 5 Bollag, D . M . ; McQueney, P .A. ; Zhu, J.; Hensens, O.; Koupal, L . ; Liesch, J.; Goetz, M . ; Lazarides, E . ; Woods, C A . Cancer Res. 1995, 55, 2325. 4 6 a) Kowalski , R.J . ; Giannakakou, P.; Hamel, E . / . Biol. Chem. 1997, 272, 2534. b) Chou, T . - C ; Zhang, X . - G . ; Balog, A . ; Su, D.-S.; Meng, D. ; Savin, K . ; Bertino, J.R.; Danishefsky, S.J. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 9642. 4 7 Molnar, I.; Schupp, T.; Ono, M . ; Zirkle, R .E . ; Milnamow, M . ; Nowak-Thompson, B . ; Engel, N . ; Toupet, C ; Stratmann, A . ; Cyr, D . D . ; Gorlach, J.; Mayo, J . M . ; Hu, A . ; Goff, S.; Schmid, J.; Ligon, J . M . Chem. Biol. 2000, 7, 97. 4 8 a) ter Haar, E . ; Kowalski , R.J . ; Hamel, E . ; L i n , C M . ; Longley, R .E . ; Gunasekera, S.P.; Rosenkranz, H.S. ; Day, B . W . Biochemistry 1996, 35, 243. b) Hung, G.T.; Chen, J.; Schreiber, S.L. Chem. Biol. 1996, 3, 287. 4 9 Mooberry, S.L.; Tien, G. ; Hernandez, A . H . ; Plubrukarn, A . ; Davidson, B .S . Cancer Res. 1999,59,653. 5 0 D 'Ambrosio , M . ; Guerriero, A . ; Pietra, F. Helv. Chim. Acta 1987, 70, 2019. 5 1 D 'Ambrosio , M . ; Guerriero, A . ; Pietra, F. Helv. Chim. Acta 1988, 71, 964. 5 2 Fenical, W. ; Jensen, P.R.; Lindel, T. "Eleutherobin and Analogues Thereof." U.S. Patent No. 5,473,057 A, December 5,1995. 133 3 3 Ketzinel, S.; Rudi, A . ; Schleyer, M . ; Benayahu, Y . ; Kashman, Y . J. Nat. Prod. 1996, 59, 873. 5 4 Battistini, C ; Ciomei, M . ; Pietra, F.; D'Ambrosio, M . ; Guerriero, A . Patent application WO 96/36335, November 21,1996. 5 5 Lindel , T.; Jensen, P.R.; Fenical, W. ; Long, B . H . ; Casazza, A . M . ; Carboni, J.; Fairchild, C R . 7. Am. Chem. Soc. 1997,119, 8744. 5 6 Nicolaou, K . C . ; van Delft, F . L . ; Ohshima, T.; Vourloumis, D . ; X u , J .Y. ; Hosokawa, S.; Pfefferkorn, J.; K i m , S.; L i , T. Angew. Chem. Int. Ed. Engl. 1997, 36, 2520. 5 / Long, B . H . ; Carboni, J . M . ; Wasserman, A . J . ; Cornell, L . A . ; Casazza, A . M . ; Jensen, P.R.; Lindel, T.; Fenical, W. ; Fairchild, C R . Cancer Res. 1998, 55, l l l l . 5 8 a) Lindel, T. Angew. Chem. Int. Ed. 1998, 37, 774. b) Chen, X . - T . ; Zhou, B! ; Bhattacharya, S.K.; Gutteridge, C .E . ; Pettus, T.R.R. ; Danishefsky, S.J. Angew. Chem. Int. Ed. 1998, 37, 789. c) Nicolaou, K . C . ; Ohshima, T.; Hosokawa, S.; van Delft, F .L . ; Vourloumis, D . ; X u , J .Y. ; Pfefferkorn, J.; K i m , S. J. Am. Chem. Soc. 1998, 120, 8674. 5 9 a) Nicolaou, K . C ; K i m , S.; Pfefferkorn, J.; X u , J .Y. ; Ohshima, T.; Hosokawa, S.; Vourloumis, D . , L i , T. Angew. Chem. Int. Ed. 1998, 37, 1418. b) Nicolaou, K . C ; X u , J .Y. ; K i m , S.; Pfefferkorn, J.; Ohshima, T.; Vourloumis, D . ; Hosokawa, S. 7. Am. Chem. Soc. 1998, 120, 8661. c) Nicolaou, K . C ; Winssinger, N ; Vourloumis, D. ; Ohshima, T.; K i m , S.; Pfefferkorn, J.; X u , J .Y. ; L i , T. J. Am. Chem. Soc. 1998, 120, 8661. 6 0 Rayl , A.J .S . The Scientist 1999,13,1. 6 1 Muramatsu, H . ; Miyauchi , M . ; Sato, B . ; Yoshimura, S. 40 t h Symposium on The Chemistry of Natural Products, Fukuoka, Japan, 1998; Paper 83, p. 487. See also: Vanderwal, C D . ; Vosburg, D . A . ; Weiler, S.; Sorensen, E.J . Org. Lett. 1999,1, 645. 6 2 a) Verdier-Pinard, P.; Wang, Z . ; Mohanakrishnan, A . K . ; Cushman, M . ; Hamel, E . Mol. Pharmacol. 2000, 57, 568. b) Wang, Z . ; Yang, D . ; Mohanakrishnan, A . K . ; Fanwick, P.E.; Nampoothiri, P.; Hamel, E . ; Cushman, M . J. Med. Chem. 2000, 43, 2419. 6 3 Roux, D. ; Hadi, H . A . ; Thoret, S.; Guenard, D. ; Thoison, O.; Pais, M . ; Sevenet, T. J. Nat. Prod. 2000, 63, 1070. 6 4 Roberge, M . ; Cinel , B . ; Anderson, H.J . ; L i m , L . ; Jiang, X . ; X u , L . ; Bigg, C M . ; Kelly, M.T. ; Andersen, R.J . Cancer Res. 2000, 60, 5052. 6 5 Anderson, H.J . ; Coleman, J.E.; Andersen, R.J . ; Roberge, M . Cancer Chemother. Pharmacol. 1997, 39, 223. 6 6 Schiff, P .B. , Horwitz, S.B. Proc. Natl. Acad. Sci. U.S.A. 1980, 77, 1561. 6 7 Meneghetti, V . B.Sc. Thesis, University of British Columbia, 1999. 134 6 8 Roberge, M . ; Berlinck, R.G.S. ; X u , L . ; Anderson, H.J . ; L i m , L . Y . ; Curman, D. ; Stringer, C M ; Friend, S.H. ; Davies, P.; Haggarty, S.J.; Kel ly, M . T . ; Britton, R.; Piers, E . ; Andersen, R.J. Cancer Res. 1998, 55, 5701. 6 9 Gottesman, M . ; Pastan, I. Annu. Rev. Biochem. 1993, 62, 385. 7 0 McDaid , H . M . ; Bhattacharya, S.K.; Chen, X . - T . ; He, L . ; Shen, H.-J . ; Gutteridge, C .E . ; Horwitz, S.B.; Danishefsky, S.J. Cancer Chemother. Pharmacol. 1999, 44, 223. 7 1 Nicolaou, K . C ; X u , J .Y. ; K i m , S.; Pfefferkorn, J.; Ohshima, T.; Vourloumis, D . ; Hosokawa, S. 7. Am. Chem. Soc. 1998, 120, 8661. 7 2 Ojima, I.; Chakravarty, S.; Inoue, T.; L i n , S.; He, L . ; Horwitz, S.B.; Kuduk, S.D.; Danishefsky, S.J. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 4256. 7 3 Cinel , B . ; Patrick, B . O . ; Roberge, M . ; Andersen, R.J . Tetrahedron Lett. 2000, 41, 2811. 7 4 He, L . ; Jagtap, P .G. ; Kingston, D.G.I . ; Shen, H.-J. ; Orr, G . A . ; Horwitz, S.B. Biochemistry 2000, 39, 3972. 7 5 Giannakakou, P.; Gussio, R.; Nogales, E . ; Downing, K . H . ; Zaharevitz, D . ; Bollbuck, B . ; Poy, G. ; Sackett, D. ; Nicolaou, K . C ; Fojo, T. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 2904. 7 6 Cornett, B . ; Monteagudo, E . ; Cicero, D.O. ; Cinel, B . ; Andersen, R.J . ; Roberge, M . ; Milanesio, M . ; Liotta, D . C ; Snyder, J.P. J. Am. Chem. Soc. submitted. 7 7 Cicero, D .O. ; Barbato, G . ; Bazzo, R. J. Am. Chem. Soc. 1995, 117, 1027. 7 8 Nevins, N . ; Cicero, D . ; Snyder, J.P. J. Org. Chem. 1999, 64, 3979. 7 9 X u , L . ; Patrick, B . O . ; Roberge, M . ; Al len , T.; van Ofwegen, L . ; Andersen, R.J . Tetrahedron Lett. 2000, 56, 9031; and references cited therein. on Cinel , B . ; Roberge, M . ; Behrisch, H . ; van Ofwegen, L . ; Andersen, R .J . Organic Lett. 2000, 2, 257. 8 1 Schaufelberger, D .E . ; Koleck, M . P . ; Beutler, J .A. ; Vatakis, A . M . ; Alvarado, A . B . ; Andrews, P.; Marzo, L . V . ; Muschik, G . M . J. Nat. Prod. 1991, 54, 1265. " Will iams, D .E . , personal communication. 8 3 Britton, R.; Roberge, M . ; Berisch, H . ; Andersen, R.J . Tetrahedron Lett. 2001, 42, 2953. 135 CHAPTER 3 3.1. G2 Cell Cycle Checkpoint Inhibitors from Stylissa flabelliformis A recently proposed strategy for cancer therapy involving compounds that inhibit the G2 cell cycle checkpoint led to the development of a new high-throughput assay to screen natural extracts for G2 checkpoint inhibitors. The crude extract from a marine sponge, Stylissa flabelliformis, collected in the waters off Motupore Island in Papua New Guinea exhibited potent activity in the new assay. Bioassay-guided fractionation of the active extract resulted in the isolation and identification of the natural product debromohymenialdisine and three related alkaloid inhibitors of the G2 checkpoint. These compounds were the first G2 checkpoint inhibitors to be found by a rational screen and were structurally distinct from previously known G2 checkpoint inhibitors. The discovery of these biologically active natural products has contributed to the understanding of the G2 checkpoint signaling pathway and provided essential new tools to probe the molecular basis of inhibition. 3.1.1. Introduction to Marine Sponges Sponges, which constitute the phylum Porifera, are the oldest living group of multicellular organisms and, although commonly thought of as primitive, are very successful and highly evolved organisms.1 In total, there are an estimated 5000-9000 species of marine sponges, in addition to 150 or so freshwater species, that have adapted to all the diverse ecological habitats of the world's oceans. This ecological success may arise from the simple architecture and sedentary lifestyle characteristic of sponges. Morphologically, sponges vary greatly in size and shape, from thin encrusting sheets scarcely a millimeter in thickness to large barrels or vases that may grow to over a meter in height and diameter.1 Most species are often 136 brightly coloured and while some members exhibit radial symmetry, the majority are irregular and display massive, erect, encrusting, or branching growth patterns.2 Sponges are sessile, filter-feeding organisms that lack internal organs but possess a well-developed system of cells arranged to form a network of canals.1 Water is pumped into the body of the sponge through small pores, called oscula, by a single layer of flagellated cells (choanocytes). The water is filtered for organic food particles and oxygen and then its unidirectional journey ends with expulsion though exit pores called oscules. The skeletal support in sponges is composed of either calcareous or siliceous spicules and protein spongin fibers and often represents the primary method of species identification. 2 While the phylum Porifera is divided into four main classes, the largest and most diverse class is the Demospongiae consisting of over 90% of all sponge species. Studies on the Demosponges have revealed a great range in habitat, spectacular variability in morphology, and an exquisitely rich chemistry. A s seen previously in Figure 1.1, sponges are the most prolific source of marine natural products described in the recent literature, accounting for nearly 44% of the compounds reported.' In addition to the thousands of unique compounds isolated from these organisms, many of the marine sponge metabolites have biosynthetic origins without known terrestrial counterparts.4 For example, the direct incorporation of isocyanide and isothiocyanide into compounds from the marine sponge Acanthella cavernosa (76, 77) 5 and the unique biosynthetic pathway to the 3-alkylpiperidine alkaloids such as manzamine A (78) 6 are exclusive to the marine world. 137 Marine sponges have provided remarkably complex molecules possessing a variety of potent biological activities making them an extremely valuable resource in finding lead structures or new molecular targets for drug discovery programs. The development of the potent anticancer agent Ara-C (13) from compounds originally isolated by Bergmann from a Caribbean sponge was presented in Chapter One. The diterpenoid manoalide (79) was originally reported by deSilva and Scheuer from the Indo-Pacific sponge Laffareilla variabilis1 and subsequently found to exhibit exciting anti-inflammatory properties by inhibiting the key enzyme phospholipase A 2 . While manoalide (79) has not achieved therapeutic use, it is widely used as a commercial reagent and biochemical tool by the pharmaceutical industry and may yet lead to new synthetic analogues for the treatment of inflammation. 4 Many marine sponge-derived natural products are currently in pre-clinical evaluation as promising antimitotic agents, including the microtubule-stabilizing compounds discodermolide (17) and laulimalide (54) discussed earlier in Chapter Two. Halichondrin B (80), originally isolated from the Japanese sponge Halichondria okadai,9 also acts as an antimitotic agent but inhibits the polymerization of tubulin instead of stabilizing the microtubules against depolymerization. 1 0 These examples and many others illustrate how marine sponge metabolites have played important roles in defining new molecular targets against disease and providing structurally novel lead compounds for drug discovery. In fact, the use of natural products, such as sponge-derived compounds, as essential 3.1.2. Review of the Oroidin Alkaloids from Marine Sponges12 Marine sponges have become well known as a rich source of structurally diverse and pharmacologically interesting C 1 1 N 5 pyrrole-imidazole alkaloids. Collectively, this large structural class of marine natural products is known as the oroidin alkaloids, named after the first member of this fascinating group of secondary metabolites. Oroidin (81) was first isolated from the marine sponge Agelas oroides in 1971 1 3 and subsequently has been found in various other sponges throughout the world . 1 4 " 2 1 In the thirty years since the initial discovery of oroidin (81), over 60 pyrrole-imidazole alkaloids have been isolated from marine sponges of the orders Agelasida, Axinell ida, and Halichondria. These compounds all appear to be biogenetically-derived from a single common precursor. The simplest structural entity related to oroidin (81) 139 is 3-amino-l-(2-aminoimidazol-4-yl)prop-l-ene (82) which lacks the pyrrole moiety. This 22 derivative was isolated from the Axinellidae sponge Teichaxinella morchella and is thought to be a possible biosynthetic precursor to the entire oroidin family of compounds. Recently, a pioneering biosynthetic feeding study using a cell culture of T. morchella by Kerr et al. demonstrated the incorporation of the labeled amino acids histidine and proline into odiline/stevensine (83) via 82 and 4,5-dibromopyrrole-2-carboxylic acid (84) (see Scheme 3.1). 2 3 This research finding led to the proposal of a universal chemical pathway for the biosynthesis of all the known oroidin alkaloids, represented as distinct structural groups in Scheme 3.2, with mechanisms outlining their formation from precursors 82 and 84.12 H 83 proline 84 Scheme 3.1. Biogenetic precursors of odiline/stevensine from the sponge T. morchella. 12 Scheme 3.2. Representative structural groups of oroidin alkaloids. Thus, Nature's combinatorial chemistry seems to have elaborated an extensive group of diverse secondary metabolites from the economical use of these two simple precursors. Structurally, the oroidin alkaloids can also be conceived as derivatives of the Q1N5 skeleton of oroidin (81) through (i) isomerization of the double bond and/or oxidation/reduction; (ii) dimerization; and (iii) cyclization.2 4 To date, there have been six modes of oroidin cyclization discovered in nature and Fattorusso et al. have proposed these modes be classified based on the atoms involved in the linkage formation.24 Consequently, the authors have categorized the pyrrole-imidazole alkaloids into six groups to form the hymenialdisine (85)," 141 dibromoagelaspongine (86),26 dibromophakellin (87),27 dibromoisophakellin (88),28 cyclooroidin (89),24 and agelastatin (90)29 structural types (see Scheme 3.3). 88 89 90 Scheme 3.3. The six cyclization modes of the oroidin skeleton. One of the oroidin cyclization modes, the C-4/C-10 linkage, gives rise to a unique class of tricyclic marine sponge alkaloids that includes the hymenialdisines (85, 91, 92), the axinohydantoins (93, 94), odiline/stevensine (83), and hymenin (95). These compounds share a common bicyclic pyrrolo[2,3-c]azepin-8-one ring system connected to a cyclic guanidine or hydantoin ring system. Debromohymenialdisine (91) was the first member of this group reported and was originally isolated in 1980 from the Caribbean sponge Phakellia flabellata.30 It has since been isolated from marine sponges of the genera Hymeniacidon,15'31 Axinella,32 and 33 25 Pseudoaxinyssa" along with its related analogue hymenialdisine (85). Recently, a number of reports have appeared describing the modulation of protein kinase C and the proinflammatory 142 transcription factor, N F K B , by both debromohymenialdisine (91) and hymenialdisine (85).34 The studies indicate that these compounds may well have beneficial effects in the treatment of inflammatory diseases and are promising new drug candidates against rheumatoid and osteoarthritis. The X-ray crystal structure of hymenialdisine (85) has been reported 1 5 along with numerous synthetic studies on both hymenialdisines and the related 3-bromohymenialdisine (92), previously isolated from the marine sponge Axinella carteri?6 91 R; = R 2 = H 93 94 83 A 9 ' 1 0 85 Rj =Br , R 2 = H 9 5 A 1 0 ' 1 92 R[ = H , R 2 = Br The structurally related axinohydantoin (93)32 and debromoaxinohydantoin (94)37 have been reported from various marine sponges along with the hymenialdisines. Axinohydantoin (93) exhibits moderate activity against a murine P-388 lymphocytic leukemia and its crystal structure features a reversal of configuration at the A 1 0 ' " olefinic group connecting the pyrroloazepinone ring to the hydantoin group. This suggested axinohydantoin was not simply a hydrolysis product of the hymenialdisines, but may arise from a C-4/C-10 cyclization of an oxidized linear precursor similar to the oroidin-like midpacamide (96) isolated by Scheuer et 38 al. Although debromoaxinohydantoin (94) was originally suggested to possess the same E-A 1 0 1 1 olefinic configuration, 3 7 a careful examination of the spectral data for axinohydantoin (93) and debromoaxinohydantoin (94) reveals the compound reported in the literature should be reassigned the Z - A 1 0 ' 1 1 olefinic configuration, consistent with a later report of Z-143 debromoaxinohydantoin. 3 9 Thus, evidence of f-debromoaxinohydantoin has yet to be reported. Odiline/stevensine (83)33'40 was isolated independently by two groups in 1985 and, along with hymenin (95),16 features the dibrominated pyrrole functionality found in oroidin (81). In terms of biological activity, hymenin (95) was originally described as exhibiting a-adrenoceptor blocking activity while ecologically, the family of C n N 5 pyrrole-imidazole alkaloids including odiline/stevensine (83) have been implicated in deterring the feeding of predatory reef fishes. 4 1 H O 96 93 3.1.3. Review of G2 Checkpoint Inhibitors As seen earlier in Chapter One, cells can activate biochemical control pathways in response to D N A damage, thus delaying cell cycle progression and allowing time for D N A repair. 4 2 These signal transduction pathways are called checkpoints and they can prevent replication of damaged D N A by causing cells to arrest in the G l phase of the cell cycle or prevent the separation of damaged chromosomes during mitosis by arresting cells in G2. Considering the critical role checkpoints play in the normal life cycle of every cell, a more detailed understanding of checkpoints in mammalian systems is needed, especially since most of the current understanding of checkpoints stems primarily from genetic studies in yeast and mice. Finding small molecule inhibitors of checkpoint pathways would provide valuable chemical tools to probe the cellular machinery of these control mechanisms. In addition, over half of all human cancers lack a functioning G l checkpoint due to mutations in the p53 tumour suppressor 144 gene. 4 j Thus, compounds that inhibit the G2 checkpoint may be valuable in cancer therapy to enhance the actions of DNA-damaging agents in tumours with a defective G l checkpoint. 4 4" 4 6 A t the time this research was beginning, there were few G2 checkpoint inhibitors known and all had been found serendipitously. Studies revealed caffeine (23) and the related purine analogues pentoxifylline (97),47 2-aminopurine (98), and 6-dimethylaminopurine (99)48 can act. as G2 checkpoint inhibitors but are not considered drug candidates due to their multiple pharmacological effects. 4 9 Staurosporine (100) and its related analogue UCN-01 (24) also display this biological activity but are found to be nonspecific protein kinase inhibitors, 4 9 , 5 0 thus limiting their usefulness. In addition, the protein phosphatase inhibitors okadaic acid (101) and fostriecin (102) can act as G2 checkpoint inhibitors but also lead to premature mitosis in the absence of D N A damage. 5 1 Thus, in order to identify new compounds as biochemical tools and potential lead compounds for cancer therapy, a rational cell-based assay was developed by Michel Roberge to screen natural extracts for G2 checkpoint inhibitors. O O O M e x - M e 23 97 98 99 145 3.2. Results and Discussion 3.2.1. Isolation of G2 Checkpoint Inhibitors As part of a general collection of marine invertebrates from the tropical Pacific Ocean reefs off Motupore and Madang in Papua New Guinea, specimens of a yellow marine sponge were collected using S C U B A . Samples were harvested by hand from the shallow water reefs, frozen on site, and transported to Vancouver over dry ice. The marine sponge was identified as Stylissa flabelliformis (Hentschel, 1912) by Dr. Rob van Soest of the Zoologisch Museum in Amsterdam, The Netherlands. Voucher samples are stored both at the University of British Columbia in Vancouver and in Amsterdam. Figure 3.1. Stylissa flabelliformis (photo by P .L . Colin) . 1 Specimens (87 g, wet weight) of Stylissa flabelliformis were thawed and extracted exhaustively with M e O H over a period of several days. The combined methanolic extracts were filtered and concentrated in vacuo to give a dark brown solid (-5 g). A sample of this 146 concentrated extract exhibited potent activity in Roberge's G2 checkpoint inhibition assay being developed at the time. Approximately 3 g of this crude extract was fractionated on a Sephadex LH-20 column using methanol as the eluent, yielding nine fractions, A through I (see Scheme 3.4). Sponge (wet weight 87 g) PNG sponge Stylissaflabelliformis Crude Methanol Extract Fraction A Fraction B Fraction C basify with 5% K 2 C 0 3 concentrate in vacuo Fraction DD Reversed Phase HPLC 80:20 H 2 0 / M e O H 103 aldisin Sephadex LH-20 MeOH Fraction D Fraction E Fraction F Fraction G Fraction H Normal Phase chromatography, stepwise gradient 100% CH 2 C1 2 to 50% C H 2 C l 2 : M e O H sat. with N H 3 Fraction A A Fraction BB Fraction CC Fraction I Reversed Phase high performance liquid chromatography 80:20:0.05 H 2 0 / M e O H / C F 3 C O O H 91-TFA 85-TFA 94 Acidify with 3N HC1 concentrate in vacuo Z-debromoaxinohydantoin 91-HC1 85-HC1 debromohymenialdisine-HCl hymenialdisine-HCl Scheme 3.4. Isolation procedure for G2 checkpoint inhibitors from S.flabelliformis. 147 Fractions D, E , and F were found to contain mainly debromohymenialdisine (91) and hymenialdisine (85) present as their taurine salts, whereas fractions F , G , and H also contained Z-debromoaxinohydantoin (94). These fractions all exhibited inhibitory activity in the G2 checkpoint assay and were further subjected to silica-gel flash chromatography using stepwise gradient elution ( C H 2 C 1 2 to 1:1 C H 2 C l 2 / M e O H saturated with NH 3 (g)) . Further purification was achieved by repeated fractionation on reversed-phase H P L C using 80:20:0.05 H 2 0 / M e O H / T F A as the eluent. The pure fractions of debromohymenialdisine (91) and hymenialdisine (85), now as their trifluoroacetic acid salts, were then converted to the hydrochloride salts by addition of 3N HC1 followed by concentration under reduced pressure. The cleavage product 103 was obtained by subjecting a portion of the original fraction D , containing mainly debromo-hymenialdisine-taurine, to a 5 % K 2 C 0 3 solution followed by concentration in vacuo. Finally, reversed-phase H P L C using 80:20 H 2 0 / M e O H as the eluent yielded a pure debromopyrrololactam, aldisin (103). Samples of the four purified marine alkaloids were tested for G2 checkpoint inhibitory activity and their structures were confirmed by N M R and mass spectrometric analyses. H Q H o H Q 91 85 94 103 148 3.2.2. Structure Elucidation of G2 Checkpoint Inhibitors The structures of debromohymenialdisine (91), hymenialdisine (85), Z-debromoaxinohydantoin (94) and aldisin (103) were solved by extensive analysis of I D and 2D N M R spectroscopic data recorded in D M S C M 6 at 400 and 500 M H z . Low and high resolution EI mass spectrometry provided important additional information on the molecular formula. Proton spin systems were identified from C O S Y data and proton-carbon attachments were determined by H M Q C experiments. The H M B C data proved invaluable for determining the connectivity within the compounds and also in the assignment of any quaternary carbons present. 3.2.2.1. Debromohymenialdisine 91 Debromohymenialdisine (91) was isolated as a yellow amorphous solid that gave an [ M ] + ion in the HRETMS at m/z 245.09128 appropriate for a molecular formula of C u H n N 5 0 2 ( A M -1.6 ppm). As the free base, compound 91 was poorly soluble in the common N M R solvents and gave poor spectra. Sharper N M R signals and a significant improvement in solubility were observed when 91 was converted to the hydrochloride salt. A quick examination of the I D and 2D N M R spectra indicated the presence of two amide carbons, four quaternary sp2 carbons, and two isolated proton spin systems, one of which displayed chemical shifts 149 consistent with the a and (3 protons on a pyrrole ring. Analysis of the molecular formula and the N M R spectra obtained for compound 91 identified it as debromohymenialdisine, originally isolated from Phakellia flabellata by Sharma et al. in 1980. 3 0 The authors recognized the possibility of geometrical isomerization at the C-10/C-11 olefin, however the large downfield shift of the C-8/C-9 methylene protons (8 3.30 ppm) due to the anisotropic effects of the neighbouring C - l 2 amide carbonyl indicated a Z-olefin configuration. This was confirmed a few years later by the X-ray crystallographic analysis of the related hymenialdisine (85)13 and comparison to the spectra reported from the isolation of /^-debromohymenialdisine in 1996. 5 2 The N M R assignments are presented in the Experimental section and a complete set of spectral data can be found in the Appendix. 150 151 152 3.2.2.2. Hymenialdisine H2N 14 6 'NH Br H O 85 Hymenialdisine (85) was isolated as a yellow amorphous solid that gave, an [ M ] + ion in the H R E I M S at m/z 325.08734 appropriate for a molecular formula of C n H i o N 5 0 2 8 1 B r ( A M -1.0 ppm). As with debromohymenialdisine (91), the free base of compound 85 was poorly soluble in the common N M R solvents and gave poor spectra. Sharper N M R signals and a significant improvement in solubility were observed when 85 was converted to the hydrochloride salt. The absence of an H-2 proton signal along with the upfield shift in the C-2 resonance from 5 123.2 to 5 104.5 ppm suggested that 85 was the 2-bromo derivative of debromohymenialdisine (91). The pronounced downfield shift of the pyrrole N H resonance in the ' H N M R spectra (5 12.92 ppm) can be attributed to the electron-withdrawing inductive effect of the neighbouring bromine atom and possible intramolecular hydrogen bonding to the adjacent C-6 carbonyl. Analysis of the molecular formula and the N M R spectra obtained for compound 85 confirmed the structure as that of hymenialdisine, originally isolated from Axinella verrucosa and Acanthella aurantiaca by Cimino et al. in 1982. 1 5 The A 1 0 ' 1 1 geometric isomer, Tf-hymenialdisine has also been recently reported. 5 2 The N M R assignments are presented in the Experimental section and a complete set of spectral data can be found in the Appendix. 153 154 155 3.2.2.3. Z-Debromoaxinohydantoin Z-Debromoaxinohydantoin (94) was isolated as a yellow amorphous solid that gave an [ M ] + ion in the H R E I M S at m/z 246.07530 appropriate for a molecular formula of C 1 1 H 1 0 N 4 O 3 ( A M -0.2 ppm). Examination of the ID N M R spectra indicated compound 94 possessed the same pyrroloazepinone ring system present in debromohymenialdisine (91) but featured a hydantoin group in place of the aminoimidazolidinone. A reversal of configuration at the A 1 0 ' 1 1 olefinic group relative to debromohymenialdisine (91) was proposed in the original isolation of debromoaxinohydantoin (94) by Pettit et al. in 1995. 3 7 However, the authors had failed to compare the ' H N M R chemical shift values of the H-8 and H-9 methylene protons (both at 5 3.25 ppm) to the values in their previous isolation of £"-axinohydantoin. 3 2 The large downfield shift of the H-9 methylene protons due to the anisotropic effects of the neighbouring C - l 2 amide carbonyl, almost identical to the effects observed with the hymenialdisines (85, 91), indicated a Z-olefinic configuration. The ' H chemical shift of H-9 protons that do not experience the anisotropic effects, as in E-axinohydantoin, was reported as 5 2.67 ppm. Comparison of N M R spectra with the recently reported isolation of Z-debromoaxinohydantoin j 9 confirmed the proposed structure for 94. A complete set of N M R assignments for Z-debromoaxinohydantoin (94) is presented in the Experimental section. 156 157 158 3.2.2.4. Aldisin O 103 Aldis in (103) was isolated as a pale yellow amorphous solid that gave an [ M ] + ion in the HREEVIS at m/z 164.05605 appropriate for a molecular formula of C g H s ^ C h ( A M 0.5 ppm). Examination of the ID N M R spectra indicated compound 103 possessed the same pyrroloazepinone ring system present in debromohymenialdisine (91) but signals at 8 194.5 and 13 8 162.2 ppm in the C N M R spectrum suggested the presence of one ketone and one amide group. Analysis of the molecular formula and the N M R spectra obtained for compound 103 confirmed the structure as that of aldisin, originally isolated in 1985 by Schmitz et al. along with debromohymenialdisine (91) from Hymeniacidon aldis and an unidentified Fijian sponge.3 1 The authors described aldisin (103) as a likely degradation product of 91 formed either by oxidation or by conjugative addition of H 2 O at C-10 followed by a reverse aldol reaction resulting in the loss of the guanidine moiety. A complete set of N M R assignments.for aldisin (103) can be found in the Experimental section. 161 3.2.3. Biological Activity Searching for new G2 cell cycle checkpoint inhibitors using the recently developed cell based assay led to the isolation of debromohymenialdisine (91), hymenialdisine (85), Z-debromoaxinohydantoin (94), and aldisin (103). The bioassay-guided isolation procedure and ultimate purification of active compounds provided the first example of G2 checkpoint inhibitors isolated as a result of a rational screen; all previous inhibitors had been identified by chance. This validated the potential of the new bioassay to identify novel active compounds that were structurally distinct from previously known G2 checkpoint inhibitors. The G2 checkpoint inhibitory activity of debromohymenialdisine (91) and the related analogues was determined along with its effects on cell cycle checkpoint and signal transduction protein kinases. 5 3 Debromohymenialdisine (91) was found to exhibit dose dependent G2 checkpoint inhibition in human breast carcinoma cells ( M C F - 7 mp53) that were arrested in G2 phase as a result of exposure to ionizing radiation (see Figure 3.10). A n I C 5 0 value of 8 u M was determined, with higher concentrations leading to maximal activity at 40 u M followed by a decrease in activity, a phenomenon also seen with other checkpoint inhibitors. 5 4 ' 5 ' 1 Hymenialdisine (85) exhibited similar potent G2 checkpoint inhibitory activity with an ICso value of 6 u M (see Table 3.1), indicating the presence of a bulky bromine substituent at C-2 in 85 has little effect on the activity. Z-debromoaxinohydantoin (94) and aldisin (103) both show a dramatic decrease in biological activity indicating that the presence of the aminoimidazolidinone subunit is required for G2 checkpoint inhibition. The two commercial products, 2-aminoimidazole and 2-amino-4,5-imidazole-dicarbonitrile, lack the pyrrololactam and were found to be essentially inactive in the bioassay. 162 1 10 100 DBH (pM) Figure 3.10. Inhibition of the G2 checkpoint by debromohymenialdisine (91). Table 3.1. G2 checkpoint inhibition by debromohymenialdisine (91) and related compounds. Compound G2 checkpoint inhibition I C 5 0 (uM) Debromohymenialdisine (91) 8 ± 4 Hymenialdisine (85) 6 ± 3 Z-debromoaxinohydantoin (94) » 2 0 0 Aldis in (103) » 2 0 0 2-aminoimidazole* » 1 0 000 2-amino-4,5,-imidazole- » 1 0 000 dicarbonitrile* obtained from Sigma-Aldrich. Hymenialdisine (85) was recently reported to be a potent inhibitor of several protein kinases in vitro, including cyclin-dependent kinases, glycogen synthase kinase-3[3, and casein kinase l . 5 6 The authors also present a crystal structure of a protein kinase Cdk2-hymenialdisine 163 complex that indicates key hydrogen bonding and van der Waals interactions between hymenialdisine (85) and the amino acid residues in the ATP-binding region of the protein. These interactions are consistent with the structure-activity results obtained for G2 checkpoint inhibition with compounds 91, 85, 94, and 103, suggesting the mechanism of action involves inhibition of a protein kinase. In particular, the deleterious effect of Z-debromoaxinohydantoin (94) or the negligible influence of a C-2 bromine substituent can both be rationalized in terms of respective nonfavourable or nonaltered drug-kinase interactions. Although hymenialdisine (85) was previously reported to inhibit a number of kinases at nanomolar concentrations and several more at micromolar concentrations in vitro,56 studies performed on the hymenialdisines (91, 85) isolated from Stylissa flabelliformis reveal a much narrower spectrum of inhibitory activity in vivo. In addition, the previous investigations did not test kinases involved in the G2 checkpoint. Debromohymenialdisine (91) isolated from S. flabelliformis was found to inhibit the checkpoint kinases C h k l and Chk2 in vitro with I C 5 0 values of 3 u M and 3.5 u M , concentrations close to those required for in vivo checkpoint inhibition ( I C 5 0 8 pM) (see Figures 3.11 and 3.12). However, there was no significant in vitro inhibition of the phosphoinositide-3 kinases A T M , A T R , and D N A - P K , thus indicating that debromohymenialdisine (91) acts as a checkpoint inhibitor by blocking two major branches of the checkpoint pathway downstream of A T M (see Figure 1.6). These results suggested the molecular targets for checkpoint inhibition are C h k l and Chk2, although the untested kinases Wee l and M y t l cannot be excluded. 164 D B H (pm Figure 3.11. Inhibition of protein kinase C h k l by debromohymenialdisine (91). o-i • i I 0.1 1 1 10 D B H (uM) Figure 3.12. Inhibition of protein kinase Chk2 by debromohymenialdisine (91). Additional studies performed on the cellular localization of a variety of cell cycle regulatory proteins implicated in the control of the G 2 / M transition (Cdc2, Cdc25C, cyclin A , 165 cyclin B , Cdk2, Cdc25B, W e e l , and 14-3-3) ruled out the possible effect of cellular compartmentalization in G2 checkpoint arrest by debromohymenialdisine (91). Because protein kinases play an essential role in the control of signal transduction pathways, the effects of debromohymenialdisine (91) on the activation or inhibition of 24 other kinases were determined by analysis of their mobility shifts in Western blots. None of the protein kinases tested ( C a M K 4 , Cdk2, C K l a , C K 2 a , E r k l , Erk2, p38 Hog M A P K , M e k l , Mek2, Mek3 , Mek4, Mek5, Mekk3, P i m l , P K B a , PKC(3, P K C s , PKCC, Rskl, p70 S 6 k , p46 S A P K , p54 S A P K , and Takl ) revealed changes in abundance or mobility upon treatment with debromohymenialdisine (91), again supporting a narrower in vivo inhibitory spectrum. A complementary approach to classical genetics to probe the function of a single gene product or protein involves the use of exogenous ligands such as the marine alkaloids isolated from the marine sponge Stylissa flabelliformis. Natural products research is embracing this "chemical genetics" approach outlined earlier in Chapter One. The power of exploiting the vast structural diversity found in nature to identify small molecules capable of either inactivating or activating biochemical pathways or individual proteins is being recognized in searching for novel G2 cell cycle checkpoint inhibitors. Since this discovery of the hymenialdisines (91, 85) as the first G2 checkpoint inhibitors identified from a rational search, additional studies have reported the isolation of staurosporine (100) and the new inhibitor isogranulatimide (25).54 These studies validate the new bioassay by identifying a known inhibitor and providing an additional structural inhibitor chemotype. 3.3. Conclusions The crude extract from a marine sponge, Stylissa flabelliformis, collected in the waters off Motupore Island in Papua New Guinea exhibited potent activity in a new bioassay searching for G2 cell checkpoint inhibitors. Bioassay-guided fractionation of this active extract resulted in 166 the isolation and identification of the natural products debromohymenialdisine (91), hymenialdisine (85), Z-debromoaxinohydantoin (94), and aldisin (103). These compounds were the first G2 checkpoint inhibitors to be found by a rational screen and were structurally distinct from previously reported G2 checkpoint inhibitors. The discovery of these biologically active natural products validated the power of the new bioassay and contributed to the understanding of the G2 checkpoint pathway by providing a new biochemical tool to probe the molecular basis of inhibition. The G2 checkpoint inhibitory activity of debromohymenialdisine (91) and the related analogues was determined along with its effects on cell checkpoint and signal transduction protein kinases. Debromohymenialdisine (91) was found to inhibit the G2 checkpoint pathway with an IC50 of 8 • M and specifically inhibited the protein kinases C h k l (IC50 3 • M ) and Chk2 ( I C 5 0 3.5 • M ) while not affecting A T M , A T R , or D N A - P K . Thus, debromohymenialdisine (91) was seen to block two branches of the checkpoint pathway downstream of A T M and was found to inhibit a narrow range of signal transduction proteins in vivo. These studies on the marine sponge alkaloids have provided small molecule ligands for a chemical genetics approach to probe the G2 checkpoint pathway. REFERENCES 1 Col in , P .L . ; Arneson, A . C . In Pacific Invertebrates; Coral Reef Press: Beverly Hil ls , 1995; pp. 17-21. 2 Rupert, E .E . ; Barnes, R .D . In Invertebrate Zoology; Sixth edition; Saunders College Publishing: Toronto, 1994; pp 73-93. Faulkner, D.J . Nat. Prod. Rep. 2001,18, 1; and previous reports in this series. 4 Andersen, R.J . ; Wil l iams, D . E . Pharmaceuticals from the Sea. In Chemistry in the Marine Environment; Hester, R .E . ; Harrison, R . M . , Eds.; Issues in Environmental Science and Technology, The Royal Society of Chemistry: Cambridge, 2000; Volume 13, pp. 55-79. 167 5 Simpson, J.S.; Garson, M . J . Tetrahedron Lett. 1990, 40, 3909. 6 Andersen, R.J. ; Van Soest, R . M . W . ; Kong, F. In Alkaloids: Chemical & Biological Perspectives; Pelletier, S.W., Ed. ; Pergamon: New York, 1996; Volume 10, pp. 301. 7 de Silva, E .D. ; Scheuer, P.J. Tetrahedron Lett. 1980, 21, 1611. 8 Potts, B . C . ; Faulkner, D.J . ; Jacobs, R.S. J. Nat. Prod. 1992, 55, 1701. 9 Hirata, Y . ; Uemura, D . Pure Appl. Chem. 1986, 58, 701. 1 0 Ba i , R.; Paull, K . D . ; Herald, C . L . ; Malspeis, L . ; Pettit, G.R. ; Hamel, E . J. Biol. Chem. 1991, 266, 15882. 1 1 Schreiber, S.L. Bioorg. Med. Chem. 1998, 6, 1127. 1 2 For an excellent review of the molecular diversity of the oroidin alkaloids see A l Mourabit, A . ; Potier, P. Eur. J. Org. Chem. 2001, 237. 1 3 a) Forenza, S.; Minale, L . ; Riccio, R.; Fattorusso, E. J. Chem. Soc, Chem. Commun. 1971, 1129. b) Garcia, E .E . ; Benjamin, L . E . ; Fryer, R.I. J. Chem. Soc, Chem. Commun. 1973, 78. 1 4 Cimino, G . ; De Stefano, S.; Minale, L . ; Sodano, G . Comp. Biochem. Physiol. 1975, SOB, 279. l i Cimino, G . ; De Rosa, S.; De Stefano, S.; Mazzarella, L . ; Puliti , R.; Sodano, G . Tetrahedron Lett. 1982, 23, 767. 1 6 Kobayashi, J.; Ohizumi, Y . ; Nakamura, H . ; Hirata, K . ; Wakamatsu, K . ; Miyazawa, T. Experientia, 1986, 42, 1064. 1 7 Keifer, P .A. ; Schwartz, R .E . ; Koker, M . E . S . ; Hughes, Jr., R . G . ; Rittschof, D. ; Rinehart, K . L . J. Org. Chem. 1991, 56, 2965. 1 8 Rosa, R ; Silva, W. ; Escalona de Motta, G . ; Rodriguez, A . D . ; Morales, J.J.; Ortiz, M . Experientia 1992, 48, 885. 1 9 Kinnel , R . B . ; Gehrken, H.-P.; Scheuer, P.J. J. Am. Chem. Soc. 1993, 115, 3376. 2 0 Cafieri, F.; Fattorusso, E . ; Mangoni, A . ; Tagliatela-Scafati, O.; Carnuccio, R. Bioorg. Med. Chem. Lett. 1995, 5, 799. 2 1 Tsukamoto, S.; Kato, H . ; Hirota, H . ; Fusetani, N . / . Nat. Prod. 1996, 59, 501. 2 2 Wright, A . E . ; Chiles, S.A.; Cross, S.S. J. Nat. Prod. 1991, 54, 1684. 2 3 Andrade, P.; Willoughby, R.; Pomponi, S.A.; Kerr, R . G . Tetrahedron Lett. 1999, 40, 4775. 2 4 Fattorusso, E. ; Tagliatela-Scafati, O. Tetrahedron Lett. 2000, 41, 799. 168 2 3 Kitagawa, I.; Kobayashi, M . ; Kitanaka, K . ; Kido, M . ; Kyogoku, Y . Chem. Pharm. Bull. 1983,57,2321. 2 6 Fedoreyev, S.A.; Iliyin, S.G.; Utkina, N . K . ; Maxinov, O.B.; Reshetnyak, M . V . Tetrahedron 1989, 45, 3487. 2 7 Sharma, G . ; Magdoff-Fairchild, B . J. Org. Chem. 1977, 42, 4118. 2 8 Fedoreyev, S.A.; Utkina, N . K . ; Ilyin, S.G.; Reshetnyak, M . V . ; Maximov, O.B. Tetrahedron Lett. 1986,27, 3177. 2 9 D 'Ambrosio, M . ; Guerriero, A . ; Debitus, C. ; Ribes, O.; Pusset, J.; Leroy, S.; Pietra, F . 7. Chem. Soc, Chem. Commun. 1993, 1305. 3 0 Sharma, G . ; Buyer, J.S.; Pomerantz, M . W . J. Chem. Soc, Chem. Commun. 1980, 435. 3 1 Schmitz, F.J. ; Gunasekera, S.P.; Lakshmi, V . ; Tillekeratne, L . M . V . J. Nat. Prod. 1985, 48, 47. 3 2 Pettit, G.R.; Herald, C . L . ; Leet, J .E.; Gupta, R.; Schaufelberger, D . E . ; Bate, R . B . ; Clewlow, P.J.; Doubek, D . L . ; Manfredi, K . P . ; Rutzler, K . ; Schmidt, J . M . ; Tackett, L .P . ; Ward, F .B. ; Bruck, M . ; Camou, F. Can. J. Chem. 1990, 68, 1621. 3 3 De Naanteuil, G . ; Ahond, A . ; Guilhem, J.; Poupat, C. ; Tran Huu Dau, E . ; Potier, P.; Pusset, M . ; Pusset, J.; Laboute, P. Tetrahedron 1985, 41, 6019. 3 4 a) Breton, J.J.; Chabot-Fletcher, M . J. Pharmacol. Exp. Ther. 1997, 282, 459. b) Roshak, A . ; Jackson, J.R.; Chabot-Fletcher, M . ; Marshall, L . A . J. Pharmacol. Exp. Ther. 1997, 283, 955. 3 5 a) Annoura, H . ; Tatsuoka, T. Tetrahedron Lett. 1995, 36, 413. b) X u , Y . - Z . ; Yakushijin, K . ; Home, D . A . 7. Org. Chem. 1997, 62, 456. c) Sosa, A . C . B . ; Yakushijin, K . ; Home, D . A . J. Org. Chem. 2000, f55, 610. 36 Supriyono, A . ; Schwarz, B . ; Wray, V . ; Witte, L . ; Muller, W . E . G . ; van Soest, R.; Sumaryono, W. ; Proksch, P .Z. Z. Naturforsch. 1995, 50c, 669. 3 7 Groszek, G . ; Kantoci, D. ; Pettit, G.R. Liebigs Ann. Chem. 1995, 715. 3 8 Chevolot, L . ; Padua, S.; Ravi , B . N . ; Blyth, P.C.; Scheuer, P.J. Heterocycles 1977, 7, 891. 3 9 Patil, A . D . ; Freyer, A . J . ; Kil lmer, L . ; Hofmann, G. ; Randall, K . Nat. Prod. Lett. 1997, 9, 201. 4 0 Albizati , K . F . ; Faulkner, D.J . J. Org. Chem. 1985, 50, 4163. 4 1 a) Wilson, D . M . ; Puyana, M . ; Fenical, W. ; Pawlik, J.R. J. Chem. Ecol. 1999, 25, 2811. b) Assmann, M . ; Lichte, E ; Pawlik, J.R.; Kock, M . Mar. Ecol. Prog. Ser. 2000, 207, 255. c) Lindel, T.; Hoffmann, H . ; Hochgurtel, M . ; Pawlik, J.R. 7. Chem. Ecol. 2000, 26, 1477. 4 2 Zhou, B . -B .S . ; Elledge, S.J. Nature 2000, 408, 433. 169 4 3 Kastan, M . B . ; Onyekwere, O.; Sidransky, D. ; Vogelstein, B . ; Craig, R . W . FASEBJ. 1996, 10, 238. 4 4 Nurse, P. Cell 1997, 97, 865. 4 5 Kao, G .D. ; McKenna, W . G . ; Maity, A . ; Blank, K . ; Muschel, R.J. Cancer Res. 1997, 57, 753. 4 6 Weinert, T.; Lydal l , D . Seminars Cancer Biol. 1993, 4, 129. 4 7 a) Walters, R . A . ; Gurley, L .R . ; Tobey, R . A . Biophys. J. 1974, 14, 99. b) Busse, P . M . ; Bose, S.K.; Jones, R .W. ; Tolmach, L . J . Radiat. Res. 1978, 76, 292. c) Schlegel, R.; Pardee, A . B . Science 1986, 232, 1264. d) Dowries, C.S.; Musk, C.S.S.; Watson, J .V.; Johnson, R . V . 7. Cell Biol. 1990, 770, 1855. 4 8 Andreassen, P.R.; Margolis, R . L . Proc. Nad. Acad. Sci. USA 1992, 89, 2272. 4 9 a) Mizuno, K . ; Noda, K . ; Ueda, Y . ; Hanaki, H . ; Saido, T.C. ; TJcuta, T.; Kuroki , T.; Tamaoki, T.; Hirai , S.; Osada, S.; Ohno, S. FEBS Lett. 1995, 359, 259. b) Kawakami, K . ; Futami, H . ; Takahara, J.; Yamaguchi, K . Biochem. Biophys. Res. Comm. 1996, 279, 778. 5 0 Akinaga, S.; Gomi , K.; Morimoto, M . ; Tamaoki, T.; Okabe, M . Cancer Res. 1991, 57, 4888. M a) Roberge, M . ; Tudan, C ; Hung, S . M . ; Harder, K . W . ; Jirik, F.R.; Anderson, H .J . Cancer Res. 1994, 54, 6115. b) Yamashita, K . ; Yasuda, H . ; Pines, J.; Yasumoto, K . ; Nishitani, H . ; Ohtsubo, M . ; Hunter, T.; Sugimura, T.; Nishimoto, T. EMBO J. 1990, 9, 4331. 5 2 Will iams, D . H . ; Faulkner, D.J . Nat. Prod. Lett. 1996, 9, 57. 5 3 Curman, D. ; Cinel , B . ; Will iams, D .E . ; Rundle, N . ; Block, W . D . ; Goodarzi, A . A . ; Hutchins, J.; Clarke, P.; Zhou, B . - B . ; Lees-Miller, S.P.; Andersen, R.J. ; Roberge, M . 7. Biol. Chem. 2001,276,17914. 5 4 Roberge, M . ; Berlinck, R.G.S . ; X u , L . ; Anderson, H.J . ; L i m , L . Y . ; Curman, D . ; Stringer, C M . ; Friend, S.H. ; Davies, P.; Haggarty, S.J.; Kel ly, M . T . ; Britton, R.; Piers, E . ; Andersen, R.J . Cancer Res. 1998, 58, 5701. 5 5 Jiang, X . ; L i m , L . Y . ; Daly, J.W.; L i . A . H . ; Jacobson, K . A . ; Roberge, M . Int. J. Oncol. 2000, 16, 971. 5 6 Meijer, L . ; Thunissen, A . - M . W . H . ; White, A . W . ; Gamier, M . ; Nikol ic , M . ; Tsai, L . - H . ; Walter, J.; Cleverley, K . E . ; Salinas, P.C. ; Wu , Y . - Z . ; Biernat, J.; Mandelkow, E . - M . ; Pettit, G.R. Chem. Biol. 2000, 7, 51. 170 CHAPTER 4 4.1. G2 Cell Cycle Checkpoint Inhibitors from Cryptocarya concinna The successful application of the G2 cell cycle checkpoint inhibitor assay in identifying new G2 checkpoint inhibitors led to a large screening of marine and terrestrial natural extracts obtained from the N C I Open Repository. A crude extract from the Taiwanese tree Cryptocarya concinna was found to exhibit potent activity in the G2 checkpoint inhibition assay. Bioassay-guided fractionation of this extract resulted in the isolation and identification of the natural product cryptofolione and a series of related polyketide lactones. Synthetic modifications on these natural products yielded four additional analogues and led to mode of action and structure-activity relationship studies. The polyketide G2 checkpoint inhibitors isolated from C. concinna were structurally distinct from previously known inhibitors and do not appear to act on the known G2 checkpoint molecular targets. 4.1.1. Review of Plant-Derived Compounds For thousands of years, plants have formed the basis of sophisticated medicinal systems throughout the world. These plant-based systems continue to play an essential role in health care, with estimates indicating approximately 80% of the world's population rely mainly on traditional medicines for their primary health care.1 Recently, there has been a resurgence in the use of traditional herbal medicines and the importance of plant-derived pharmaceuticals continues to grow. There are currently 119 chemical substances, originally derived from 90 plant species, which are in use as important drugs and three quarters of these compounds were discovered from plants used in traditional medicine. Plant-derived compounds have especially had a significant impact on the treatment of proliferative diseases such as cancers. 171 Perhaps the best-known anticancer drugs in clinical use are the Vinca alkaloids, vincristine (21) and vinblastine (22), originally isolated from the Madagascar periwinkle Catharanthus roseus.3 Their discovery and success in the treatment of cancer has led to many semi-synthetic analogues currently in clinical use and pre-clinical trials, including vinorelbine (104) and vindesine (105). The Vinca alkaloids are plant-derived antimitotic agents like the previously described taxanes, but act by inhibiting the polymerization of tubulin into microtubules as opposed to stabilizing microtubules against depolymerization. 21 R ^ C H O R 2 = C 0 2 M e R 3 = O A c 105 22 R t = M e R 2 = C 0 2 M e R 3 = O A c 104 Rj = Me R 2 = C O N H 2 R 3 = O H In addition to the Vinca alkaloids and the taxanes, two other important classes of anticancer compounds are also based on plant-derived compounds. Currently in general clinical use, the semi-synthetic derivatives etoposide (106) and teniposide (107) are based on podophyllotoxin (108), an antitumour agent from the roots of various Podophyllum plants.4 The Chinese ornamental tree Camptotheca acuminata was the source of camptothecin (109), which led to the development of the clinical therapeutics topotecan (110) and irinotecan (111).5 172 109 R 1 = H R 2 = H 111 110 R i = O H R 2 = C H 2 N M e 2 The natural product rohitukine (112), reported from the plant Dysoxylum binectariferum, has served as the basis for the totally synthetic flavone, flavopiridol (113).6 This synthetic agent is currently scheduled to advance to a phase U clinical trial against a broad range of tumours and acts as an inhibitor of cyclin dependent kinases that regulate the transition from G2 to M phase.7 The plant natural product has also served as a lead compound for a series of compounds, such as the paullones (114).1 These discoveries highlight the importance of plant natural product "lead" compounds that give rise to semi-synthetic and synthetic compounds with a variety of potencies and specificities. 173 4.1.2. Review of Cryptocarya Phytochemistry The genus Cryptocarya, which includes about 350 species, belongs to the Lauraceae family of plants and is wide-spread throughout the world especially in tropical regions. For years, traditional healers of Southern Africa have used the bark of Cryptocarya and the related Ocotea genus for medicinal as well as mythical purposes.9 Recently, the use and importance of Cryptocarya species has increased due to accessibility and supply issues concerning the Ocotea plants. 1 0 This interest has led to over four decades of investigations on the chemical constituents of Cryptocarya species, with numerous reports detailing the isolation of various alkaloids and a-pyrones. The alkaloids isolated from the genus Cryptocarya have been reported to display a range of interesting biological activities. Cryptopleurine (115), originally isolated from the bark of Cryptocarya pleurosperma in 1954," demonstrates cytotoxicity towards human epidermoid 12 13 carcinoma, antiviral activity, and has been shown to be a highly-active inhibitor of protein synthesis. 1 4 Chemical investigations into Cryptocarya chinensis, a perennial woody plant widely distributed in the forests of Taiwan and China, have revealed it to be a rich source of pavine alkaloids (compounds 116 to 118). 1 5" 1 8 In addition to reports of these alkaloids possessing behavioural and antitumour effects,1 9 (-)-caryachine (118) was recently shown to display exciting antiarrhythmic activity and represents a very promising new drug for the treatment of cardiac arrhythmias. 2 0 Most of the remaining alkaloids are structurally related to 174 reticuline (119) from the New Caledonian Cryptocarya longifolia11 via different cyclization modes, as evident in cryptowolline (120) or cryptodorine (121). O C H 3 119 120 121 In addition to the alkaloid constituents of the various Cryptocarya species, these laurel plants also contain a collection of 6-substituted 5,6-dihydro-a-pyrones as secondary metabolites. These metabolites are biosynthesized from either acetate or mixed acetate-shikimate pathways 2 1 and are widely distributed in both plants and fungi. 2 2 While many of these compounds display a diverse range of biological activities, such as plant growth inhibitors, insect antifeedants, antifungal and antitumour agents, the reports from Cryptocarya species are noticeably lacking in biological information. The 6-arylalkyl-5,6-dihydro-a-pyrones from lauraceous plants are characteristic of Aniba and Cryptocarya species, however, Aniba a-pyrones commonly possess a C-4 methoxyl group while the pyrones from Cryptocarya are not substituted on C-4 but often feature a styryl group attached to the C-6 side chain. A summary of the a-pyrones isolated from various Cryptocarya species are presented in Table 4.1 along with additional cyclized derivatives. Table 4.1. Summary of a-pyrones isolated from various Cryptocarya species. 175 Compounds Organisms n • 24 C . massoia 122 C. latifolia 25,26 123 124 125 26 C. caloneura C latifolia C. wylei26 27 C. bourdilloni 28 C. moschata C. wylei26 30 C. kurzii31 C. myrtifolia9 C. latifolia26 C. liebertiana32 176 Compounds Organisms 135 R = H 136 R = Ac 137 Recently, investigations into the chemistry of Cryptofolia moschata resulted in the identification of thirteen 6-[co-arylalkenyl]-5,6-dihydro-a-pyrones (126, 130, 138-148) including the known goniothalamin (126) and cryptofolione (130).23 Many of the compounds reported were found to exist as diastereomers and all contain a styryl group and 5,6-dihydro-a-pyrone connected by either a linear and hydroxylated carbonic chain or a carbonic chain with a dihydroxy-tetrahydropyrane ring. O 138 C2. = R 130 C 4 =/? 139 C 2 = 5 140 C 4 = 5 177 O H O 143 C 4 . , C 5 . = /?,5 144 C 4 . , C 5 . = 5,5 145 C 4 . , C 5 . = R,R O H O 146 C 6 . , C 7 , C 8 . = /?,5,/? 147 C 6 . , C T , C 8 . = 5,5,/? 148 C 6 . , C T , C 8 . = /?,/?>5 4.2. Results and Discussion 4.2.1. Isolation of G2 Checkpoint Inhibitors As part of a general screening of marine and terrestrial extracts obtained from the N C I Open Repository, a methanolic crude extract of the Taiwanese tree Cryptocarya concinna (Lauraceae) exhibited potent activity in the G2 checkpoint inhibition assay. A 500 mg sample of this crude extract was suspended in one liter of water and partitioned against E tOAc (1 liter x 5). The E t O A c fraction was concentrated in vacuo, suspended in one liter of 9:1 M e O H / I U O , and then partitioned against hexanes (1 liter x 5). Next, water was added to the remaining methanolic fraction in order to achieve a 6:4 M e O H / H ^ O ratio. This fraction was finally partitioned against two liters of C H 2 C I 2 (1 liter x 5). Bioassay-guided fractionation showed the G2 checkpoint inhibitory activity was entirely present in the E tOAc , the 9:1 M e O H / F L O , and the C H 2 C I 2 fractions, respectively (see Scheme 4.1). 178 Crude Methanol Extract (approx. 500 mg) Taiwanese Tree Cryptocarya concinna H7O EtOAc 9:1 M e O H : H 2 0 Hexane 6:4 MeOH:H 2 Q| CH 2 C1 2 Fraction A Fraction B Sephadex LH-20 80:20 M e O H / C H 2 C l 2 Fraction C Fraction D Fraction E Fraction F Fraction G Normal Phase chromatography 75:25 EtOAc/Hexanes Fraction A Fraction B Fraction C Fraction D High Performance liquid chromatography Fraction E Fraction F Fraction G Fraction H Fraction I 131 (lmg) 148 (15mg) 130 138 (2 lmg) (lmg) Scheme 4.1. Isolation procedure for G2 checkpoint inhibitors from C. concinna. Approximately 125 mg of this dichloromethane fraction was fractionated on a Sephadex LH-20 column using 80:20 M e O H / C H 2 C l 2 as the eluent, yielding seven fractions, A through G. Only fraction D exhibited significant inhibitory activity in the G 2 checkpoint assay and was thus 179 subjected to normal phase flash chromatography using an isocratic 75:25 EtOAc/hexanes elution. The resulting nine fractions were screened for inhibitory activity and interesting ' H N M R signals. A final purification of four fractions was achieved by reversed-phase H P L C using 48:52 HaO/MeOH as the eluent yielding pure cryptofolione (130), Z-cryptofolione (148), cryptofolione ketone (131), and compound 138. The four purified a-pyrones were tested for G2 checkpoint inhibitory activity and their structures confirmed by detailed N M R and mass spectrometric analyses. 131 138 4.2.2. Structure Elucidation of G2 Checkpoint Inhibitors The structures of cryptofolione (130), Z-cryptofolione (148), cryptofolione ketone (131), and compound 138 were solved by extensive analysis of ID and 2D N M R spectroscopic data recorded in DMSO-<r/6 at 400 and 500 M H z . Low and high resolution EI and CI mass spectrometry provided important information on the molecular formula. Proton spin systems were identified from C O S Y data and proton-carbon attachments were determined by H M Q C experiments. The H M B C data proved invaluable for determining the connectivity within the compounds and also in the assignment of any quaternary carbons present. The relative stereochemistry was obtained by formation of acetonide derivatives, analysis of circular dichroism experiments, and comparison with previously reported spectral data. 130 f-Cryptofolione (130) was isolated as a white amorphous solid that gave an [ M ] + ion in the HREDvIS at m/z 314.15178 appropriate for a molecular formula of C19H22O4 ( A M -0.09 ppm). A quick analysis of the I D and 2D N M R spectra indicated the presence of a monosubstituted benzene ring, two rrans-disubstituted olefins, a c/j-a,P-unsaturated ester functionality, three oxymethine protons, and three methylene groups. The H M B C and C O S Y experimental data revealed a trans-styryl double bond (coupling constant J of 15.9 Hz) connected to a 1,3-diol system and ultimately linked to an a-pyrone through a second trans-olefin (/ = 15.5 Hz). Thus, the spectral data from compound 130 matched the N M R data for the 6-[co-arylalkenyl]-5,6-dihydro-a-pyrone structures characteristic of Cryptocarya secondary metabolites. The final structure of £-cryptofolione (130) was confirmed by comparison of spectra with previously published data. 9 ' 2 3 It should be noted that the most recent published N M R data 2 3 misidentifies the structure of £-cryptofolione (130) and incorrectly assigns the , 3 C resonances for C-7 ' and C-8 ' , which should be interchanged. A summary of all N M R assignments, C O S Y , and H M B C correlations for ZT-cryptofolione (130) can be found in Table 4.2 and 2D N M R spectra can be found in the Appendix. Table 4.2. N M R data for £-cryptofolione (130) recorded in CDCI3. 181 position 8 ' H a 5 1 3 C b C O S Y H M B C (7 value in Hz) Correlations 3 Correlations 3 l ( O ) 2 164.1 H3 3 5.99 (d, 7=9.8) 121.4 H4 H5 4 6.83 (dt, 7=9.7, 4.5) 144.9 H3, H5 H 5 , H 6 5 2.38 (m) 29.6 H4, H6 6 4.84 (m) 77.8 H5, H I ' H I ' , H 2 ' r 5.65 (dd, 7=15.5, 6.5) 129.7 H6, H 2 ' H6, H 3 ' 2' 5.84 (dd, 7=15.5,7.4) 131.2 H 1 ' , H 3 ' H6, H 3 ' 3' 2.27 (t, 7=6.7) 40.3 H 2 ' , H 4 ' H 1 ' , H 2 ' 4' 4.03 (m) 68.1 H 3 ' , H 5 ' H3 ' ,H5? 4 ' - O H 3.14 (br) 5' 1.75 (m) 42.2 H 4 \ H 6 ' H 3 ' 6' 4.61 (m) 70.2 H 5 ' , H 7 ' H 5 ' , H 7 ' , H 8 ' 6 ' - O H 3.14 (br) 7' 6.24 (dd, 7=15.9, 6.0) 131.8 H6 . ' ,H8 ' H 5 ' 8' 6.60 (d, 7=15.9) 129.9 H 7 ' H 6 ' , H 2 ' 7 6 " 1" 136.6 H 7 ' , H 3 ' 7 5 " 2'76" 7.34 (d, 7=7.9) 126.4 H3'V5" H 8 ' , H 4 " 3'75" 7.28 (t, 7=7.6) 128.5 H2'76", H 4 " 4" 7.20 (t, 7=7.3) 127.6 H3'75" H2'76" a recorded at 400 M H z . b recorded at 100 M H z . The relative stereochemistry of the 1,3-diol in the linear unsaturated carbonic chain was determined by examination of the acetonide derivative 149. In papers presented by Rychnovsky et a/ . 3 3 and Evans et al.,34 acetonides formed from syn-1,3-diols are shown to exist in well-defined chair conformations with the two bulky substituents occupying equatorial positions (Ri and R2 in Figure 4.1). B y contrast, acetonides formed from ann'-l,3-diols adopt a twist-boat conformation to minimize unfavourable 1,3-diaxial interactions. The 1 3 C N M R chemical shifts of the acetonide methyl groups then become diagnostic for the relative stereochemistry of the 1,3-diols, indicating an ann'-configuration for £-cryptofolione (130). A detailed analysis of the acetonide derivative (149) follows in Section 4.2.3.1. anti Figure 4.1. Characteristic conformations and 1 3 C N M R chemical shift values for syn- and anti-\,3-6\o\s. The C-6 configuration in Zs-cryptofolione (130) was determined from the positive Cotton effect in the C D curve at 254-272 nm due to n—>n* transitions. Using the Snatzke rule 3 5 and considering a pseudo-equatorial substituent, this result defines a 6R configuration. Analysis of the complete set of N M R spectral data along with optical rotation values and C D curves suggest is-cryptofolione (130) isolated from Cryptocarya concinna corresponds to Cavalheiro and Yoshida's compound 5, which has a 6R,4'S,6'R configuration. 183 184 Z-Cryptofolione (148) was isolated as a white amorphous solid that gave an [ M ] + ion in the HRCEVIS at m/z 314.15226 appropriate for a molecular formula of C19H22O4 ( A M -1.5 ppm). Analysis of the I D and 2D N M R spectra indicated considerable similarities to compound 130 including the presence of a monosubstituted benzene ring, two disubstituted olefins, a cis-a,p-unsaturated ester functionality, three oxymethine protons, and three methylene groups. The H M B C and C O S Y experimental data revealed the only difference from 130 was the presence of a c/s-styryl double bond with a coupling constant J of 11.7 Hz . This group was connected to a similar 1,3-diol system and linked to an a-pyrone through a second trans-olefm (J = 15.5 Hz). The relative stereochemistry of the 1,3-diol of Z-cryptofolione (148) was determined by examination of its acetonide derivative. The l 3 C N M R chemical shifts of the acetonide methyl groups again indicated an Gtnfz'-configuration for Z-cryptofolione (148). The C-6 configuration in Z-cryptofolione (130) was determined from the positive Cotton effect in the C D curve at 254-272 nm due to n—*n* transitions, defining a 6R configuration. A summary of all N M R assignments, C O S Y , and H M B C correlations for Z-cryptofolione (148) can be found in Table 4.3 and 2D N M R spectra can be found in the Appendix. Table 4.3. N M R data for Z-cryptofolione (148) recorded in CDCI3. position 5 ' H a 5 1 3 C b C O S Y H M B C (7 value in Hz) Correlations 3 Correlations 3 1 (0) 2 164.0 H3 3 6.00 (d, 7=9.8) 121.5 H4 H5 4 6.83 (m) 144.7 H3, H5 H5, H6 5 2.37 (m) 29.7 H4, H6 H 3 , H 1 ' 6 4.84 (m) 77.8 H5, H I ' H5, H 1 ' , H 2 ' r 5.63 (dd, 7=15.5, 6.5) 129.8 H6, H 2 ' H6, H 3 ' 2' 5.83 (dt, 7=15.5, 7.4) 131.2 H 1 ' , H 3 ' H6, H 3 ' 3' 2.25 (t, 7=6.7) 40.2 H 2 ' , H 4 ' Hl ' ,H2*,H5 'ab 4' 4.04 (m) 68.2 H 3 ' , H 5 ' b H2 ' ,H3 ' ,H5 'ab 4 ' - 0 H 2.84 (br)* 5'a 1.79 (ddd, 7=14.5, 7.8, 3.2) 42.4 H5 'b , H 6 ' , H 4 ' H 3 ' 5'b 1.73 (ddd, 7=14.4, 8.5 3.6) H5'a , H 4 ' , H 6 ' 6' 4.92 (m) 65.6 H5'a , H7 ' H5'ab, H8 ' 6 ' - O H 2.66 (br)* T 5.75 (dd, 7=11.7, 9.2) 133.8 H 6 ' , H8 ' H5'ab 8' 6.50 (d, 7=11.7) 130.5 H 7 ' H2'76" 1" 136.4 H 7 ' , H 3 " / 5 " 2"/6" 7.24 (d, 7=7.2) 128.7 H3'75" H 8 ' , H 4 " 3"/5" 7.32 (t, 7=7.7) 128.3 H2'76", H 4 " 4" 7.23 (t, 7=7.3) 127.3 H3'75" H2'76", H3'75" a recorded at 500 M H z . b recorded at 100 M H z . * assignments interchangeable. 187 188 189 Cryptofolione ketone (131) was isolated as a white amorphous solid that gave an [M+H] + ion in the HRCJJVIS at m/z 313.14358 appropriate for a molecular formula of C19H20O4 ( A M 1.3 ppm). Analysis of the I D and 2D N M R spectra indicated considerable similarities to compound 130 including the presence of a monosubstituted benzene ring, two trans-disubstituted olefins, a c/5-a,(3-unsaturated ester functionality, two oxymethine protons, three methylene groups, and a ketone functionality. The H M B C and C O S Y experimental data revealed the only difference from 130 was the presence of a ketone with a ° C N M R chemical shift of 5 200.6 ppm in place of the C-6 ' alcohol group. Since cryptofolione ketone (131) can serve as a biosynthetic precursor to cryptofolione (130), it is most likely an authentic natural product and not an isolation artifact. The relative stereochemistry at the two chiral centers of cryptofolione ketone (131) was not determined due to the lack of sufficient material. However, it should be noted that while N M R data for cryptofolione ketone (131) has been reported first as a P C C oxidation product 9 and later as a natural product, 2 6 the stereochemistry suggested in the 1 ^  literature was not supported by experimental data and the C resonances for C-7 ' and C-8' should be interchanged. A significant difference also exists in the reported values for the H-5 ' methylene protons and those determined from the present isolation from C. concinna (8 2.61 ppm versus 8 2.74 and 8 2.90 ppm.) The C-6 configuration in 131 was determined from the positive Cotton effect in the C D curve at 254-272 nm due to n—>7i* transitions, defining a 6R configuration. A summary of all N M R assignments, C O S Y , and H M B C correlations for cryptofolione ketone (131) can be found in Table 4.4 and 2D N M R spectra can be found in the Appendix. Table 4.4. N M R data for cryptofolione ketone (131) recorded in C D C 1 3 . position 5 ' H a 5 1 3 C b C O S Y H M B C (7 value in Hz) Correlations 3 Correlations 3 1(0) 2 163.9 3 6.03 (d, 7=9.6) 121.6 H4 4 6.86 (dt, 7=9.9, 4.2) 144.6 H3, H5 H5 5 2.44 (m) 29.7 H4, H6 6 4.90 (m) 77.9 H 5 , H 1 ' H 2 ' r 5,72 (dd, 7=15.8, 6.6) 130.0 H6, H 2 ' 2' 5.89 (dt, 7=15.0, 7.2) 130.9 H l ' , H 3 ' a b H3'ab 3'a 2.37 (m) 39.3 H 2 ' , H3 'b 3'b 2.30 (m) H 2 ' , H 3 ' a , H 4 ' 4' 4.21 (m) 67.2 H3 'b , H5'ab H3'ab, H5 'b 4 ' - O H 3.31 (br) 5'a 2.90 (dd, 7=17.3, 2.7) 46.0 H 4 ' , H5 'b 5'b 2.74 (dd, 7=17.5, 9.0) H5 ' a 6' 200.6 H5 'b , H 7 ' , H 8 ' 7' 7.57 (d, 7=16.2) 143.9 H 8 ' 8' 6.72 (d, 7=16.1) 126.1 H 7 ' 1" 134.1 H8 ' 2"/6" 7.55 (d, 7=7.5) 128.5 H 3 7 5 " H 7 ' , H 3 7 5 " , H4" 3"/5" 7.39 (m) 129.0 H2"/6", H 4 " 4" 7.39 (m) 130.9 H3" /5" H 2 7 6 " a recorded at 500 M H z . b recorded at 100 M H z . 191 192 193 Compound 138 was isolated as a white amorphous solid that gave an [ M ] + ion in the HRCEVIS at m/z 288.13514 appropriate for a molecular formula of C 1 7 H 2 0 O 4 ( A M 3.6 ppm). Analysis of the I D and 2D N M R spectra indicated considerable similarities to Zi-cryptofolione (130) including the presence of a monosubstituted benzene ring, one fra/u-disubstituted olefin, a cw-a,P-unsaturated ester functionality, two oxymethine protons, and three methylene groups. The H M B C and C O S Y experimental data revealed the only difference from 130 was the absence of the second trans-olef'm l inking the 1,3-diol to the 5,6-dihydro-a-pyrone. The relative stereochemistry at the three chiral centers of compound 138 was not determined due to the lack of sufficient material. However, the final structure of compound 138 was confirmed by comparison of spectra with previously published data. 2 3 Analysis of the complete set of N M R spectral data along with optical rotation values and C D curves suggest compound 138 isolated from Cryptocarya concinna corresponds to Cavalheiro and Yoshida's compound 3, which exists in the 6R,2'S,4'R configuration. A similar misassignment of the l 3 C resonances for C-5 ' and C-6' exists in this report, as with £-cryptofolione (130), and thus the values should be interchanged. A summary of all N M R assignments, C O S Y , and H M B C correlations for compound 138 can be found in Table 4.5 and 2D N M R spectra can be found in the Appendix. Table 4.5. N M R data for compound 138 recorded in C D C 1 3 . position 6 ' H a (7 value in Hz) 5 1 3 C b C O S Y Correlations 3 H M B C Correlations 3 H O ) 2 167.2 3 6.01 (d, 7=9.9) 121.4 H4 H5 4 6.88 (dt, 7=9.7, 4.2) 145.2 H3, H5 H3, H5 5 2.37 (m) 30.0 H4, H6 6 4.74 (m) 75.0 H 5 , H l ' a H l ' a l ' a 1.90 (m) 42.4 H6 l ' b 1.78 (m) H 2 ' T 4.37 (m) 65.0 H l ' b , H 3 ' H l ' b , H 3 ' 2 ' - O H 2.25 (br)* 3' 1.82 (m) 43.1 H 2 ' , H 4 ' 4' 4.65 (m) 70.9 H 3 ' , H 5 ' H 3 ' , H 5 ' , H 6 ' 4 ' - 0 H 2.83 (br)* 5' 6.29 (dd, 7=15.9, 6.3) 131.4 H4 ' , H6 ' H4 ' 6' 6.62 (d, 7=16.1) 130.6 H 5 ' H 2 7 6 " 1" 136.4 H 5 ' , H 3 " / 5 " 2"/6" 7.37 (d, 7=7.0) 126.6 H 3 7 5 " H6 ' , H 4 " 375" 7.31 (t, 7=7.6) 128.6 H 2 7 6 " , H 4 " 4" 7.24 (under solvent) 127.9 H 3 7 5 " H2"/6" 3 recorded at 500 M H z . recorded at 100 M H z . * assignments interchangeable. 195 196 4.2.3. Synthetic Derivatives of Zs-Cryptofolione 4.2.3.1. Cryptofolione acetonide 5" 149 In order to establish the relative configuration of the 1,3-diol system in compound 130, cryptofolione acetonide (149) was prepared by adding 2 ml of 2,2-dimethoxy-propane and a catalytic amount of pyridinium p-toluene sulfonate to approximately 2 mg of /f-cryptofolione (130). After workup, cryptofolione acetonide (149) was isolated as a white amorphous solid that gave an [ M + H ] + ion in the HRCJJVIS at m/z 355.19102 appropriate for a molecular formula of C22H26O4 ( A M -0.2 ppm). A quick analysis of the I D and 2D N M R spectra indicated the presence of an acetonide group bridging the two alcohol oxygen atoms. The addition of two methyl singlets (8 1.39 and 8 1.40 ppm) in the ' H N M R spectrum and an acetal carbon resonance at 8 100.5 ppm was diagnostic for acetonide formation. As mentioned earlier in Section 4.2.2.1, acetonides formed from syn- and anti- 1,3-diols are found to exist in distinct and well-defined conformations 3 3 , 3 4 (see Figure 4.1). Thus, the 1 3 C N M R chemical shifts of the acetonide methyl groups appearing at 8 24.9 and 8 25.5 ppm were diagnostic for an anti-configuration for the 1,3-diols in E-cryptofolione (130). The spectral data obtained for cryptofolione acetonide (149) was in full agreement with previously published data.9 A summary of all N M R assignments, C O S Y , and H M B C correlations for cryptofolione acetonide (149) can be found in Table 4.6. Table 4.6. N M R data for cryptofolione acetonide (149) recorded in CDCI3. position 5 ' H a 5 1 3 C b C O S Y H M B C (7 value in Hz) Correlations 3 Correlations 3 1 (0) 2 164.0 H4 3 6.03 (d, 7=9.9) 121.7 H4 H5 4 6.86 (dt, 7=9.7, 4.5) 144.5 H3.-H5 H5, H6 5 2.42 (m) 29.8 H4, H6 H3 6 4.88 (m) 77.9 H 5 . H 1 ' H4, H5, H 1 ' , H 2 ' r 5.68 (dd, 7=15.5, 6.5) 129.4 H6, H 2 ' H6, H2 , H3'ab 2' 5.83 (dd, 7=15.6, 7.3) 130.7 H l ' , H 3 ' a b H6, H l ' , H 3 ' a b 3'a 2.34 (m) 38.5 H 2 ' , H 4 ' H l ' , H 2 ' , H 5 ' b 3'b 2.25 (m) H 2 ' 4' 3.93 (m) 65.8 H3 'a , H5 'b H3'ab, H5 'b 5'a 1.84 (m) 37.5 H 6 ' H3'ab, H7 5'b 1.75 (m) H 4 ' 6' 4.49 (m) 67.7 H5 'a , H7 ' H5 'a , H 7 ' , H8 ' . 7' 6.20 (dd, 7=16.0, 6.3) 129.7 H 6 \ H 8 ' H5 'a , H6 ' 8' 6.54 (d, 7=15.9) 130.5 H7 ' H6 ' , H2"/6" 1" 136.7 H 7 ' , H 8 ' , H 3 7 5 " 276" 7.36 (d, 7=7.5) 126.5 H3" /5" H4" 375" 7.28 (t, 7=7.5) 128.5 H2'76", H 4 " 4" 7.22 (t, 7=7.4) 127.7 H3"/5" H2"/6" 1"' 100.5 H 2 " ' , H 3 ' " 2"' 1.40 (s)* 24.9* H 3 ' " 3"' 1.39 (s)* 25.5* H 2 ' " 3 recorded at 400 M H z . recorded at 100 M H z . * assignments interchangeable. 199 200 201 Diacetylcryptofolione (150) was prepared by adding 0.5 ml of acetic anhydride and 1.5 ml of pyridine to approximately 2 mg of £-cryptofolione (130). After workup, diacetylcryptofolione (150) was isolated as a white amorphous solid that gave an [ M ] + ion in the H R E I M S at m/z 398.17274 appropriate for a molecular formula of C 2 3 h 2 6 0 6 ( A M -0.51 ppm). A quick analysis of the I D and 2D N M R spectra indicated the presence of two acetate groups in place of the two alcohol functionalities in 130. The addition of two methyl singlets (8 1.99 and 5 2.04 ppm) in the 'ff N M R spectrum and two ester carbon resonances at 8 170.2 and 5 170.5 ppm were diagnostic for acetate subunits. A summary of all N M R assignments, C O S Y , and H M B C correlations for diacetylcryptofolione (150) can be found in Table 4.7. Table 4.7. N M R data for diacetylcryptofolione (150) recorded in CDCI3. position 5 ' H a 5 1 3 C b COSY H M B C (J value in Hz) Correlations3 Correlations3 1 (0) 2 163.8 H3 3 6.03 (dt, 7=9.9, 1.8) 121.7 H4, H5 H5 4 6.85 (dt, 7=9.7, 4.2) 144.5 H3,H5 H5 5 2.39 (m) 29.6 H3, H4, H6 H3 6 4.87 (dt, 7=8.0, 6.6) 77.6 H5, H I ' H5, H1 ' ,H2 ' r 5.65 (dd, 7=15.5, 6.3) 130.5 H6, H2' H6, H2, H3' 2' 5.76 (dd, 7=15.3, 7.4) 129.3 H 1 ' , H 3 ' H6, H 1 ' , H 3 ' 3' 2.35 (m) 37.6' H 2 \ H4' H1 ' ,H2 ' 4' 5.03 (m) 68.7 H3' ,H5 'b H3',H5fb 5'a 1.94 (m) 38.5 H6' H3 ' ,H7 5'b 1.90 (m) H4' 6' 5.46 (m) 70.7 H5'a, H7' H5'a, H7', H8' T 6.09 (dd, 7=15.9, 7.4) 127.0 H6 ' ,H8 ' 8' 6.60 (d, 7=15.9) 132.9 H7' H6 ' ,H276" 1" 136.1 H7 ' ,H8 ' ,H3"/5" 276" 7.35 (d, 7=7.1) 126.6 H3"/5" H8 ' ,H4" 375" 7.29 (t, 7=7.5) 128.6 H2"/6", H4" H276" 4" 7.23 (t, 7=7.1) 128.1 H3"/5" H2"/6" 1"' 170.2* H2'" 2"' 2.04 (s)* 21.2* 3"' 170.5* H3'" 4"' 1.99 (s)* 21.0* 3 recorded at 500 MHz. recorded at 100 MHz. * assignments interchangeable. 203 204 205 4.2.3.3. p-mercaptoethanol adduct O 4" 3 ^ 'S 2"' 5" O H 151 The P-mercaptoethanol adduct of E-cryptofolione (151) was prepared by adding 10 ul of P-mercaptoethanol in a phosphate buffer to approximately 2 mg of £-cryptofolione (130) dissolved in THF. After workup, the P-mercaptoethanol adduct (151) was isolated as a pale yellow oil that gave an [M] + ion in the HRCHVIS at m/z 392.16585 appropriate for a molecular formula of C 2 iH 2 80 5 S (AM -0.3 ppm). A quick analysis of the ID and 2D N M R spectra indicated the absence of proton signals for the cw-olefin of the a,P-unsaturated lactone (5 5.99 and 5 6.83 ppm in £-cryptofolione (130)). These proton resonances were replaced by three new methylene signals, two of which were attributed to the new p-mercaptoethanol substituent. H M B C and COSY correlations confirm the presence of P-mercaptoethanol at C-4, consistent with a Michael-type addition. The relative stereochemistry of the P-mercaptoethanol substituent at C-4 was assigned based on the coupling constants measured between H-4 and the two H-3 methylene protons. The deshielded equatorial H-3 (5 2.91 ppm) shows a coupling constant 7 of 5.6 Hz to H-4 whereas the axial H-3 (8 2.55 ppm) shows a larger coupling to H-4 (7 = 7.7 Hz). This larger coupling constant is close to the typical 7 value of 8-10 Hz expected between two axial protons, thus supporting an equatorial P-mercaptoethanol substituent (see Figure 4.14). A summary of all N M R assignments, COSY, and H M B C correlations for the P-mercaptoethanol adduct (151) can be found in Table 4.8. C O S Y Figure 4.14. Selected C O S Y correlations in the p-mercaptoethanol adduct (151). Table 4.8. N M R data for p-mercaptoethanol adduct (151) recorded in C D C 1 3 . position 5 ' H a 5 1 3 C b C O S Y H M B C (7 value in Hz) Correlations 3 Correlations 3 1 (O) 2 169.2 H3ab 3a 2.91 (dd, 7=17.4, 5.6) 36.8 H3b 3b 2.55 (dd, 7=17.4, 7.7) H3a 4 3.39 (m) 34.3 H3ab, H5ab, H6 5a 2.09 (m) 34.6 H5b, H6 H6 5b 2.01 (m) H5a 6 5.08 (m) 76.8 H5a, H I ' H5a, H 1 ' , H 2 ' r 5.59 (dd, 7=15.6, 5.8) 130.6 H6, H 2 ' H5ab, H6, H 3 ' 2' 5.81 (dt, 7=15.4, 7.5) 130.3 H 1 \ H 3 ' H 3 ' 3' 2.29 (t, 7=6.7) 40.4 H 2 ' , H 4 ' H 1 ' , H 2 ' 4' 4.05 (m) 68.2 H 3 ' , H5'ab H 2 ' , H 3 ' , H 5 ' a b 5'a 1.81 (m) 42.2 H 4 ' , H 5 ' b , H 6 ' H 3 ' , H 7 ' 5'b 1.75 (m) H 4 ' , H 5 ' a , H 6 ' 6' 4.64 (m) 70.6 H5'ab, H 7 ' , H 8 ' 7' 6.27 (dd, 7=15.9, 6.2) 131.6 H 8 ' H5'ab 8' 6.62 (d, 7=15.8) 130.3 H 7 ' 1" 136.5 H 7 ' , H 8 ' , H 3 7 5 2'76" 7.37 (d, 7=7.0) 126.5 H 3 7 5 " H 8 ' , H 4 " 3 7 5 " 7.30 (t, 7=7.4) 128.6 H 2 7 6 " , H4" 4" 7.22 (t, 7=7.2) 127.8 H3'75" H2'76" 1"' 2.75 (t, 7=5.9) 33.8 2"' 3.77 (t, 7=5.9) 61.4 3 recorded at 400 M H z . recorded at 100 M H z . * assignments interchangeable. 4 207 208 209 4.2.3.4. Compound 152 O 152 £-cryptofolione (130) was subjected to hydrogenation conditions using H>(g) in the presence of catalytic Pd-C to yield compound 152. After workup, the hydrogenation product 152 was isolated as a white amorphous solid that gave an [ M + H ] + ion in the H R C T M S at m/z 323.22162 appropriate for a molecular formula of C 1 9 H 3 0 O 4 ( A M 1.9 ppm). A quick analysis of the I D and 2D N M R spectra indicated the absence of proton and carbon resonances for the three olefins groups that were present in ZT-cryptofolione (130). The absence of a third oxymethine proton resonance in the region around 5 4 to 5 ppm and a downfield shift in the carbonyl resonance at 5 177.6 ppm supported the opening of the lactone ring as first indicated by the molecular formula. The olefin and oxymethine resonances were replaced by seven new methylene group signals, with H M B C and C O S Y correlations consistent with the proposed structure of compound 152. A summary of all N M R assignments, C O S Y , and H M B C correlations for compound 152 can be found in Table 4.9. Table 4.9. N M R data for compound 152 recorded in C D C 1 3 . position S ' H a 6 1 3 C b C O S Y H M B C (7 value in Hz) Correlations 3 Correlations 3 1 (OH) 2 177.6 H3 3 2.33 (m) 33.8 H4 4 1.62 (m) 24.6 H 3 , H 5 H3, H5 5 1.30 (m) 28.8 H4 H3, H I ' * 6 r 1.39 (m) 1.30-1.39 (m)* 25.5 29.0* H3, H I ' * 2' 1.30-1.39 (m)* 29.2* 3'a 1.49 (m) 37.4 H3 'b , H4 ' H 5 ' .3'b 1.42 (m) H3 'a , H4 ' 4' ' 3.94 (br.s) 69.4 H3'ab, H 5 ' 5' 1.62 (m) 42.5 H 4 ' , H 6 ' 6' 3.96 (bs) 69.0 H 5 \ H 7 ' a b H7'ab, H8'ab 7'a 1.84 (m) 39.1 H 6 \ H 7 ' b , H 8 ' H8'ab 7'b 1.77 (m) H 6 ' , H 7 ' a , H8 8'a 2.78 (m) 32.2 H7'ab, H8 'b 8'b 1.75 (m) H7'ab, H8 ' a 1" 142.0 H7 'a , H8'ab, H 3 7 5 " 276" 7.18 (d, 7=8.0) 128.4 H 3 7 5 " H8'ab, H3"/5 ' H4" 375" 7.26 (t, 7=7.4) 128.5 H2"/6", H4" H2"/6" 4" 7.17 (t, 7=7.6) 125.9 H3" /5" H2"/6" 3 recorded at 500 M H z . recorded at 100 M H z . * assignments interchangeable. 212 213 4.2.4. Biological Activity Using Roberge's cell-based assay, a large-scale screen for G2 cell cycle checkpoint inhibitors from crude extracts obtained from the N C I Open Repository led to the isolation of E-cryptofolione (130), Z-cryptofolione (148), cryptofolione ketone (131), and compound 138. The bioassay-guided isolation and purification identified these four 6-[co-arylalkenyl]-5,6-dihydro-a-pyrones from the Taiwanese tree Cryptocarya concinna, consistent with the phytochemistry previously reported for this genus. Relatively simple synthetic transformations on the major metabolite, f-cryptofolione (130), resulted in four additional analogues for structure-activity relationship (SAR) studies, namely cryptofolione acetonide (149), diacetylcryptofolione (150), a p-mercaptoethanol adduct (151), and a hydrogenated cryptofolione (152). Preliminary results show the natural products £-cryptofolione (130) and its shorter homologue, compound 138, exhibited similar G2 checkpoint inhibitory potency as the previously described debromohymenialdisine (91) (see Figure 4.19). Altering the configuration of the A 7 ' 8 olefin (i.e., Z-cryptofolione (148)) or the oxidation state of the C-5 ' alcohol (i.e., cryptofolione ketone (131)) resulted in almost complete loss of inhibitory activity. Among the synthetic analogues, the cryptofolione acetonide (149) and the P-mercaptoethanol adduct (151) appear to retain biological activity while diacetylcryptofolione (150) and hydrogenated cryptofolione (152) are essentially inactive. It is clear that further studies are required to completely determine the structural features necessary for G2 checkpoint inhibition. 214 Concentration (ug/ml) Figure 4.19. Inhibition of the G2 checkpoint by natural and semi-synthetic a-pyrones. To date, all the known G 2 checkpoint inhibitors have been shown to act either as protein phosphatase inhibitors, such as okadaic acid (101) and fostriecin (102), or as protein kinase inhibitors, such as caffeine (23), staurosporine (100), U C N - O l (24), debromohymenialdisine (91), and isogranulatimide (25). However, E-cryptofolione (130) was found to be inactive against the known phosphatase and kinase G 2 checkpoint targets, suggesting it acts by a novel mechanism on an undetermined cellular target. Structurally, Zs-cryptofolione (130) possesses an a,P~unsaturated 8-lactone similar to leptomycin B (154), a compound recently reported to inhibit the nuclear export protein C r m l . 3 5 C r m l normally acts to exclude the protein kinase regulatory subunit cyclin B from the nucleus until the onset of mitosis. The inhibition of C r m l by leptomycin B thus results in the accumulation of cyclin B in the nucleus and premature entry into mitosis, overriding the G2 checkpoint. However, leptomycin B (153) is inactive in yeast C r m l mutants possessing a serine substitution at cysteine-529, suggesting it binds covalently through a Michael-type addition of the sulfhydryl group on Cys-529. When E-cryptofolione (130) was tested against this yeast 215 mutant, it retains G2 checkpoint inhibitory activity, suggesting its mode of action does not involve a Michael-type addition to the a,P-unsaturated 5-lactone. This result is further supported by the significant inhibitory activity of the p-mercaptoethanol adduct (151) of E-cryptofolione. It is still possible that £-cryptofolione (130) inhibits the action of C r m l via an alternate mechanism. OH O 154 In contrast, there are several nuclear export proteins that do not possess a cysteine residue in their active site, such as Nip 1.36 The effects of is-cryptofolione (130) on these possible cellular targets are currently under investigation in the laboratory of Michel Roberge at the University of British Columbia. 4.3. Conclusions As a result of a large-scale screen of natural extracts for G2 checkpoint inhibitors, a series of a-pyrones from the Taiwanese tree Cryptocarya concinna were found to exhibit potent inhibitory activity. Bioassay-guided purification resulted in the isolation and identification of E-cryptofolione (130), Z-cryptofolione (148), cryptofolione ketone (131), and compound 138. Synthetic modifications on the major metabolite, /f-cryptofolione (130), yielded four additional analogues: cryptofolione acetonide (149), diacetylcryptofolione (150), a P-mercaptoethanol adduct (151), and a hydrogenated cryptofolione (152). The G 2 checkpoint inhibitors isolated from C. concinna were structurally distinct from previously known inhibitors and appear to act by a novel mechanism of action. Continuing studies are underway to determine the cellular target and structural requirements of inhibition. 216 REFERENCES 1 Newman, D.J . ; Cragg, G . M . ; Snader, K . M . Nat. Prod. Rep. 2000,17, 215. 2 Arvigo, R.; Balick, M . In Rainforest Remedies; Lotus Press: Twin Lakes, 1993. 3 Hamel, E . Med. Res. Rev. 1996,16, 207; and references cited therein. 4 Cragg, G . M . ; Boyd, M . R . ; Cardellina, J .H. , U ; Newman, D.J . ; Snader, K . M . ; McCloud , T .G. In Ethnobotany and the Search for New Drugs; Ciba Symposium No. 185; 1994, p. 178. 3 Potmeisel, M . ; Pinedo, H . In Camptothecins: New Anticancer Agents; C R C Press: Boca Raton, 1990. | 6 Naik, R . G . ; Kattige, S.L.; Bhat, S.V.; Alreja, B . ; de Sousa, N . J . ; Rupp, R . H . Tetrahedron 1988,44,2081. 7 Christian, M . C . ; Pluda, J . M . ; Ho, T .C. ; Arbuck, S.G.; Murgo, A . J . ; Sausville, E . A . Semin. Oncol. 1997, 24, 219. 8 Cave, A . ; Leboeuf, M . ; Moskowitz, H . ; Ranaivo, A . ; Bick , I.R.C.; Sinchai, W. ; Nieto, M . ; Sevenet, T.; Cabalion, P. Aust. J. Chem. 1989, 42, 2243. 9 Sehlapello, B . M . ; Drewes, S.E.; Scott-Shaw, R. Phytochemistry 1994, 37, 847; and references cited therein. 1 0 Drewes, S.E.; Sehlapello, B . M . ; Horn, M . M . ; Scott-Shaw, R.; Sandor, P. Phytochemistry 1995, 38, 1427. 1 1 Fridrichson, J.; Mathieson, A . M c L . Nature 1954,173, 132; Acta Crystallogr. 1955, 8, 761. 1 2 Hoffmann, J.J.; Luzbetak, D.J . ; Torrance, S.J.; Cole, J.R. Phytochemistry 1978,17, 1448. 1 3 Krmpotic, E . ; Farnsworth, N . R . ; Messmer, W . M . J. Pharm. Sci. 1972, 61, 1508. 1 4 Donaldson, G.R.; Atkinson, M . R . ; Murray, A . W . Biochem. Biophys. Res. Commun. 1968, 31, 104. 1 5 Chang, W.-T. ; Lee, S.-S.; Chueh, F.-S.; L i u , K . C . S . Phytochemistry 1998, 48, 119. 1 6 Lee, S.-S.; L i u , Y . C . ; Chen, C H . J. Nat. Prod. 1990, 53, 1267. 1 7 Bick, I.R.C.; Sevenet, T.; Sinchai, W. ; Skelton, B . W . ; White, A . H . Aust. J. Chem. 1981, 34, 195. 1 8 Chen, C . - H . ; Lee, S.-S.; L a i , C. -F. ; W u , J.; Beal, J.L. J. Nat. Prod. 1979, 42, 163. 217 1 9 a) Meisenberg, G . ; Simmons, W . H . ; Collins, M . A . Pharmacol. Biochem. Behaviour 1984, 20, 355. b) Wu , Y . C . ; L iou , Y . F . ; L u , S T . ; Chen, C . H . ; Chang, J.J.; Lee, K . H . Planta Medica 1989, 55, 163. 2 0 a) W u , M . - H . ; Su, M . - J . ; Lee, S.-S.; L i n , L . -T . ; Young, M . - L . Br. J. Pharmacol. 1995,116, 3211. b) Chen, L . ; Su, M . - J . ; W u , M . - H . ; Lee, S.-S. J. Cardiovasc. Pharmacol. 1996, 27, 740. 2 1 Leete, E . ; Muir , A . ; Towers, G . H . N . Tetrahedron Lett. 1982, 23, 2635. 2 2 Davies-Coleman, M . T . ; Rivett, D . E . A . Prog. Chem. Org. Nat. Prod. 1989, 55, 1; and references cited therein. 2 3 Cavalheiro, A . J . ; Yoshida, M . Phytochemistry 2000, 53, 811. 2 4 Cavi l l , G . W . K . ; Clark, D . V . ; Whitfield, F . B . Aust. J. Chem. 1968, 21, 2819. 2 3 a) Drewes, S.E.; Sehlapello, B . M . ; Horn, M . M . ; Scott-Shaw, R.; Sandor, P. Phytochemistry 1995, 38, 1427. b) Drewes, S.E.; Horn, M . M . ; Wijewardene, C S . Phytochemistry 1996, 41, 333. c) Collett, L . A . ; Davies-Coleman, M . T . ; Rivett, D . E . A . ; Drewes, S.E.; Horn, M . M . Phytochemistry 1997, 44, 935. 2 6 Drewes, S.E.; Horn, M . M . ; Shaw, R.S . Phytochemistry 1995, 40, 321. 2 7 Hlubucek, J.R.; Robertson, A . V . Aust. J. Chem. 1967, 20, 2199; stereochemistry of 126 later redefined as R in: Meyer, H . H . Liebigs Ann. Chem. 1984, 484. 2 8 Govindachari, T.R.; Parthasarathy, P . C Tetrahedron Lett. 1971, 37, 3401. 2 9 Govindachari, T.R.; Parthasarathy, P .C . Tetrahedron Lett. 1972, 35, 3419. 3 0 Spencer, G.F. ; England, R .E . ; Wolf , R . B . Phytochemistry 1984, 25, 2499. 3 1 a) Fu, X . ; Sevenet, T.; Remy, F.; Pais, M . ; Hadi, A . ; Hamid, A . ; Zeng, L . M . 7. Nat. Prod. 1993, 5f5, 1153. b) Fu, X . ; Sevenet, T.; Hamid, A . ; Remy, F.; Pais, M . Phytochemistry 1993, 55,1272. 3 2 Drewes, S.E.; Horn, M . M . ; M a v i , S. Phytochemistry 1997, 44, 437. 3 3 a) Rychnovsky, S.D.; Skalitzky, D.J . Tetrahedron Lett. 1990, 31, 945. b) Rychnovsky, S.D.; Rogers, B . N . ; Yang, G.J. Org. Chem. 1993, 58, 3511. c) Rychnovsky, S.D.; Richardson, T .L; Rogers, B . N . J. Org. Chem. 1997, 62, 2925. 3 4 Evans, D . A . ; Rieger, D . L . ; Gage, J.R. Tetrahedron Lett. 1990, 31, 7099. 3 : 5 Kudo, N . ; Matsumori, N . ; Taoka, H . ; Fujiwara, D . ; Schreiner, E.P. ; Wolff, B . ; Yoshida, M . ; Horinouchi, S. Proc. Natl. Acad. Sci. USA 1999, 96, 9112. 3 6 Farjot, G . ; Sergeant, A . ; Mikaelian, I. 7. Biol. Chem. 1999, 274, 17309. EXPERIMENTAL 218 The extractions and column chromatography purifications were performed using reagent grade solvents. H P L C purifications utilized H P L C grade solvents that had been filtered and degassed prior to use. A l l other solvents, reagents, and standards were reagent or commercial grade and were used without further purification. Biichi rotary evaporators facilitated the removal of solvents in vacuo. Normal phase flash column chromatography was carried out using 230-400 mesh Merck silica gel G60. Reversed-phase flash column chromatography was performed using Whatman brand reversed-phase C - l 8 silica and size-exclusion chromatography was performed using Sephadex L H - 2 0 beads (25-100 u.m). Normal and reversed-phase thin layer chromatography (TLC) was carried out on commercial Kieselgel 60 F254 and Whatman M K C 1 8 F plates, respectively. Compounds were visualized by U V detection (k 254 nm) and by charring using a vanillin/sulphuric acid spray reagent. High performance liquid chromatography (HPLC) separations were performed on one of two systems: a Waters 600E H P L C pump with a Waters 486 U V tunable absorbance detector, or a Waters 600E H P L C pump with a Waters 996 U V photodiode array detector. Each system directed the data to a personal computer using Mil lenium™ 2010 chromatography software. Normal phase H P L C separations involve using either a Whatman Partisil 10 Magnum column or an Econosil cyano-bonded Si 5(0. column. Reversed-phase H P L C purifications were performed on either a Whatman Partisil 10 ODS-3 Magnum column or an Econosil C18 5u column. N M R data was collected on three different spectrometers: a Bruker A M X 5 0 0 , a Bruker WH400, or a Bruker A M 4 0 0 . Each was fitted with a 5mm probe and the spectra were obtained in either D M S O - d 6 or CDCI3 using the solvent residual peaks as references (DMSO-J6 : 8 2.49 219 ( 'H) ; 5 39.5 ( 1 3 C) ; C D C 1 3 : 8 7.24 ( 'H), 5 77.0 ( 1 3 C) ppm). Spectra were processed using Bruker Windows™ compatible W I N N M R software and coupling constants were reported in Hz . L o w and high resolution EI mass spectra were collected on a Kratos M S 5 0 / D S 5 5 S M mass spectrometer while the low and high resolution F A B mass spectra were recorded on a Kratos Concept II H Q spectrometer. A l l M S data acquisition was performed by the staff at the U B C Department of Chemistry Mass Spectrometry Center. Optical rotations were measured on a Jasco P-1010 polarimeter or a Perkin-Elmer 241 M L polarimeter (1 cm quartz cell), and the [OC]D values are given in 10"'degcm2g"1. U V measurements were performed on a Waters 2487 Dual Absorbance Detector with maximum absorptions recorded as ^ m a x in nanometers and molar absorptivity coefficients £ in units of Lmof ' cm" 1 . Circular dichroism (CD) spectra were obtained using a Jasco J-710 spectropolarimeter. A l l antimitotic and G2 checkpoint inhibition assays were performed by members of Dr. Michel Roberge's research group at U B C . X-ray diffraction analysis was performed by Brian Patrick at U B C . Eleutherobin (61) crystallized in space group ?2{2\2X with a=\2.8291(8), 6=13.6209(6), and c= 19.168(1) A . A crystal with dimensions 0.30 x 0.20 x 0.15 mm, was mounted on a glass fiber. Data were collected at - 1 0 0 ° C on a Rigaku /ADSC C C D area detector in two sets of scans (O = 0.0 to 190.0°, x = 0°; and co =-18.0 to 23.0°, x =- 90°) using 0.50° oscillations with 58.0 s exposures. The crystal-to-detector distance was 40.55 mm with a detector swing angle of -5.52°. O f the 7035 unique reflections measured ( M o - K a radiation, 29max=55.8°, i?m r=0.071, Friedels not merged), 4520 were considered observed (I>3o(I)). The final refinement residuals were R=0.046 (on F, l>3o(l)) and wR2=0.141 (on F2, all data). The data was processed using the d * T R E K program and corrected for both Lorentz and polarization effects. The structure was solved by direct methods and all non-hydrogen atoms were refined anisotropically, while all hydrogens involved in hydrogen-bonding were refined isotropically. A l l other hydrogens were 220 included in calculated positions. A l l calculations were performed using the teXsan crystallographic software package of the Molecular Structure Corporation. Desmethyleleutherobin (69): White amorphous solid; H R F A B M S [ M + H ] + m/z 643.32230 ( C 3 4 H 4 6 N 2 O 1 0 , A M -1.21 ppm); [ a ] D 2 ° = -31.3°, (c = 0.2, M e O H ) ; U V (MeOH) X m a x 2 9 1 nm (log 8 3.929); ' H N M R ( D M S O - J 6 , 500 M H z ) , 1 3 C N M R (DMSO-c/6, 100 M H z ) , and 2D N M R data listed in Table 2.1. Eleutherobin (61): White amorphous solid; H R F A B M S [ M + H ] + m/z 657.33719 C 3 5H4 8 N 2 Oio ( A M -2.32 ppm); [ a ] D 2 0 = -35.7°, (c = 0.4, M e O H ) ; U V (MeOH) Xmax 292 nm (log e 4.053); *H N M R ( D M S O - J 6 , 500 M H z ) , l 3 C N M R ( D M S O - J 6 , 100 M H z ) , and 2D N M R data listed in Table 2.2. Desacetyleleutherobin (70): White amorphous solid; H R F A B M S [ M + H ] + m/z 615.32813 C33H46N2O9 ( A M -0.05 ppm); [ a ] D 2 0 = -40.0°, (c = 0.3, M e O H ) ; U V (MeOH) X m a x 290 nm (log e 4.076); ' H N M R (DMSO-</6, 500 M H z ) , 1 3 C N M R (DMSO-^6 , 100 M H z ) , and 2D N M R data listed in Table 2.3. Isoeleutherobin A (71): White amorphous solid; H R F A B M S [ M + H ] + m/z 657.33834 C 3 5 H 4 8 N 2 O 1 0 ( A M -0.58 ppm); [cc]D 2° = -42.8°, (c = 0.1, MeOH) ; U V (MeOH) A™* 290 nm (log e 4.132); ' H N M R ( D M S O - J 6 , 500 M H z ) and 2D N M R data listed in Table 2.4. Z-eleutherobin (72): White amorphous solid; H R F A B M S [ M + H ] + m/z 657.33830 C 3 5H48N2O 1 0 ( A M -0.65 ppm); ' H N M R ( D M S O - J 6 , 500 M H z ) and 2D N M R data listed in Table 2.5. Caribaeoside (73): Colourless glass; H R F A B M S [ M + H ] + m/z 673.33474 C35H48N2O11 ( A M 1.64 ppm); [ a ] D 2 0 = -10.0°, (c = 0.08, M e O H ) ; U V (MeOH) X m a x 288 nm (log £ 3.792); ' H N M R (DMSO-46, 500 M H z ) and 2D N M R data listed in Table 2.6. 221 Caribaeolin (74): Clear o i l ; H R F A B M S [ M + H ] + m/z 541.29111 C35H48N2O11 ( A M -0.49 ppm); [ a ] D 2 0 = -12.0°, (c = 0.08, M e O H ) ; U V (MeOH) Xmax 289 nm (log e 4.071); ' H N M R (DMSO-^6 , 500 M H z ) and 2D N M R data listed in Table 2.7. Diacetyleleutherobin (75): Formed from three separate acetylation reactions on eleutherobin (61), desacetyleleutherobin (70), and isoeleutherobin A (71). Approximately 1 mg of compound was dissolved in 1.5 ml of pyridine and then 0.5 ml of acetic anhydride was added and mixture was stirred at room temperature overnight. Isolated ~ 1 mg of white amorphous solid; H R F A B M S [ M + H ] + m/z 741.33834 C 3 9 H 5 2 N 2 O i 2 ( A M -1.58 ppm); ! H N M R (DMSO-J6 , 500 M H z ) and 2D N M R data listed in Table 2.8. Sarcodictyin A (55): Clear oi l ; H R F A B M S [ M + H ] + m/z 497.26514 C 2 8H36N 2 0 6 ( A M -0.04 ppm); ' H N M R (pyridine-^5, 400 M H z ) 5 8.05 (d, 7=15.5, 1H), 7.68 (s, 1H), 7.35 (s), 7.14 (d, 7=15.5, 1H), 7.12 (d, 7=5.6, 1H), 6.97 (d, 7=9.5, 1H), 6.28 (d, 7=5.6, 1H), 5.26 (bs, 1H), 5.18 (d, 7=7.0, 1H), 4.58 (m, 1H), 3.65 (s, 3H), 3.39 (s, 3H), 2.95 (m, 1H), 2.38 (m, 1H), 1.98 (m, 1H), 1.94 (m, 1H), 1.58 (s, 3H), 1.56 (s, 3H), 1.53 (s, 3H), 1.43 (m, 1H), 1.17 (m, 1H), 0.91 (d, 7=6.6, 3H), 0.81 (d, 7=6.6, 3H); 1 3 C N M R (pyridine-^, 100 M H z ) 5 168.0, 167.2, 144.0, 140.4, 138.3, 138.0, 135.6, 134.7, 134.3, 133.0, 124.5, 121.8, 115.4, 112.3, 89.6, 81.8, 51.8, 42.1, 39.3, 34.9, 33.3, 32.2, 29.0, 25.8, 24.5, 22.2, 22.1, 20.4. Erythrolide A (36): White, amorphous solid; HREUvIS [ M ] + m/z 538.15613 C 2 6 H 3 i O i o C l ( A M -0.34 ppm); ' H N M R (CDC1 3 , 500 M H z ) 5 6.36 (dd, 7=16.1, 7.1, 1H), 5.90 (s, 1H), 5.79 (d, 7=16.2, 1H), 5.51 (s, 1H), 5.39 (s, 1H), 5.35 (s, 1H), 5.05 (d, 7=9.3, 1H), 4.46 (d, 7=9.3, 1H), 3.42 (s, 1H), 3.08 (m, 1H), 2.86 (m, 1H), 2.25 (dd, 7=10.5, 7.9, 1H), 2.13 (s, 3H), 2.03 (s, 3H), 1.90 (s, 3H), 1.87 (d, 7=10.1, 1H), 1.48 (s, 3H), 1.27 (s, 3H), 1.10 (s, 3H); 1 3 C N M R (CDC1 3 , 100 M H z ) 5 204.7, 175.0, 171.4, 169.5, 167.5, 138.6, 132.8, 127.2, 126.8, 87.2, 83.0, 80.6, 79.9, 75.4, 59.2, 43.6, 43.4, 39.6, 37.2, 28.6, 22.3, 21.3, 21.3, 20.8, 20.5, 9.2. 222 Erythrolide E (41): White, amorphous solid; HREUvIS [ M ] + m/z 496.15080 C24H29O9CI ( A M -0.26 ppm); ' H N M R (CDCI3, 500 M H z ) 8 6.63 (d, 7=10.0, 1H), 6.00 (d, 7=10.0, 1H), 5.73 (dd, 7=5.2, 2.5, 1H), 5.67 (s, 1H), 5.39 (s, 1H), 4.78 (s, 1H), 4.36 (d, 7=1.2, 1H), 4.18 (d, 7=1.6, 1H), 3.95 (m, 1H), 3.21 (d, 7=2.2, 1H), 2.73 (m, 2H), 2.66 (m, 1H), 2.24 (s, 3H), 2.11 (s, 3H), 1.62 (s, 3H), 1.45 (s, 3H), 1.03 (s, 3H); 1 3 C N M R (CDC1 3 , 100 M H z ) 8 194.1, 174.1, 169.9, 169.9, 152.3, 138.2, 124.5, 123.0, 86.1, 85.7, 83.1, 80.5, 69.8, 68.7, 59.3, 48.9, 41.6, 41.3, 36.3, 22.3, 21.2,21.1,21.1,6.5. Debromohymenialdisine (91)HC1: Yel low, amorphous solid; HREEVIS [ M ] + m/z 245.09128 C i H u N g O z ( A M -1.6 ppm); ' H N M R ( D M S O - J 6 , 400 M H z ) 8 12.10 (s, 1H), 9.5-8.5 (br, 3H), 8.02 (br.s, 1H), 7.10 (s, 1H), 6.49 (s, 1H), 3.29 (m, 4H); 1 3 C N M R (DMSO-^6 , 100 M H z ) 8 163.5, 162.9, 154.5, 130.2, 126.7, 122.7, 120.2, 120.0, 109.5, 39.2,31.6. Hymenialdisine (85)-HCl: Yel low, amorphous solid; HREEVIS [ M ] + m/z 325.08734 C n H 1 0 N 5 O 2 8 1 B r ( A M -1.0 ppm); ' H N M R (DMSO-J6 , 500 M H z ) 8 12.88 (s, 1H), 11.02 (br, 1H), 9.33 (br, 1H), 8.85 (br, 1H), 8.14 (br.s, 1H), 6.63 (s, 1H), 3.25 (m, 4H); 1 3 C N M R ( D M S O -d6, 100 M H z ) 8 163.0, 162.2, 154.5, 128.5, 128.3, 121.4, 120.5, 111.3, 105.2,39.7,31.9. Z-Debromoaxinohydantoin (94): Yel low, amorphous solid; HREEVIS [ M ] + m/z 246.07530 C 1 1 H 1 0 N 4 O 3 ( A M -0.2 ppm); *H N M R ( D M S O - J 6 , 400 M H z ) 8 11.79 (br, 1H), 11.04 (br, 1H), 9.44 (br, 1H), 7.89 (br, 1H), 6.96 (br, 1H), 6.51 (s, 1H), 3.2 (m, 4H); 1 3 C N M R (DMSO-^6 , 100 M H z ) 8 165.3, 162.9, 154.3, 125.4, 122.7, 122.4, 121.9, 121.3, 109.7, 40.4,30.6. Aldisin (103): Pale yellow, amorphous solid; HREEVIS [ M ] + m/z 164.05605 C 8 H 8 N 2 0 2 ( A M -0.5 ppm); ' H N M R ( D M S O - J 6 , 400 M H z ) 8 12.20 (br, 1H), 8.34 (br, 1H), 6.97 (s, 1H), 6.53 (s, 1H), 3.36 (m, 2H), 2.68 (m, 2H); 1 3 C N M R (DMSO-c/6, 100 M H z ) 8 194.3, 162.2, 128.0, 123.5, 122.4, 109.4,43.5, 36.5. 223 E-Cryptofol ione (130): White, amorphous solid; HRETMS [ M ] + m/z 314.15178 C19H22O4 ( A M -0.09 ppm); [ a ] D 2 0 = 57.5°, (c = 0.5, M e O H ) ; U V (MeOH) k m a x 240 nm (log e 3.867); ' H N M R (CDCI3, 400 M H z ) , 1 3 C N M R (CDC1 3 , 100 M H z ) , and 2D N M R data listed in Table 4.2. Z-Cryptofolione (148): White, amorphous solid; H R C I M S [ M ] + m/z 314.15226 C i 9 H 2 2 0 4 ( A M -1.5 ppm); [ a ] D 2 0 = 120.5°, (c = 1.2, M e O H ) ; U V (MeOH) X m a x 239 nm (log e 3.462); ! H N M R (CDCI3, 4 0 0 M H z ) , 1 3 C N M R (CDC1 3 , 100 M H z ) , and 2D N M R data listed in Table 4.3. Cryptofolione ketone (131): White, amorphous solid; H R C I M S [ M + H ] + m/z 313.14358 C 1 9 H 2 o 0 4 ( A M 1.3 ppm); [ a ] D 2 0 = 105.0°, (c = 0.08, M e O H ) ; U V (MeOH) X m a x 286 nm (log e 3.823); ' H N M R (CDCI3, 500 M H z ) , 1 3 C N M R (CDC1 3 , 100 M H z ) , and 2D N M R data listed in Table 4.4. "-Compound 138: White, amorphous solid; H R C I M S [ M ] + m/z 288.13514 C 1 7 H 2 o 0 4 ( A M 3.6 ppm); [ a ] D 2 0 = -13.8°, (c = 0.08, MeOH) ; U V (MeOH) lmaK 277 nm (log e 2.343); lU N M R (CDCI3, 500 M H z ) , 1 3 C N M R (CDC1 3 , 100 M H z ) , and 2D N M R data listed in Table 4.5. Cryptofolione acetonide (149): Prepared by adding 2 ml of 2,2-dimethoxy-propane and a catalytic amount of pyridinium p-toluene sulfonate to approximately 2 mg of Zs-cryptofolione (130), stirred at 70°C overnight. Y i e l d 1.8 mg. White, amorphous solid; H R C I M S [M + H ] + m/z 355.19102 C 2 2 H 2 6 0 4 ( A M -0.2 ppm); [ a ] D 2 0 = 49.4°, (c = 0.7, MeOH) ; U V (MeOH) Xmax 250 nm (log s 4.211); ! H N M R (CDC1 3 , 400 M H z ) , 1 3 C N M R (CDC1 3 , 100 M H z ) , and 2D N M R data listed in Table 4.6. Diacetylcryptofolione (150): Prepared by adding 0.5 ml of acetic anhydride and 1.5 ml of pyridine to approximately 2 mg of E-cryptofolione (130), stirred at room temperature overnight. Yie ld 1.9 mg. White, amorphous solid; H R E P M S [ M ] + m/z 398.17274 C 2 3 H 2 6 0 6 ( A M -0.51 ppm); [ a ] D 2 0 = 46.9°, (c =• 0.8, M e O H ) ; U V (MeOH) X m a x 250 nm (log e 4.151); *H N M R (CDCI3, 400 M H z ) , 1 3 C N M R (CDCI3, 100 M H z ) , and 2D N M R data listed in Table 4.7. 224 P-mercaptoethanol adduct (151): Prepared by adding 10 pi of P-mercaptoethanol in a phosphate buffer (pH 7.4) to approximately 2 mg of is-cryptofolione (130) dissolved in T H F , stirred at room temperature for 1 hour. Y ie ld 2.1 mg. Pale yellow o i l ; H R C I M S [ M ] + m/z 392.16585 C 2 i H 2 8 0 5 S ( A M -0.3 ppm); [ a ] D 2 0 = -10.2°, (c = 1.6, M e O H ) ; U V (MeOH) Xmax 250 nm (log s 3.881); *H N M R (CDC1 3 , 400 M H z ) , 1 3 C N M R (CDC1 3 , 100 M H z ) , and 2D N M R data listed in Table 4.8. Hydrogenated ^ -cryptofolione (152): Prepared by using H 2 (g) in the presence of catalytic Pd-C in ethanol, stirred at room temperature overnight to yield 0.8 mg of compound 152. White, amorphous solid; H R C I M S [ M + H ] + m/z 323.22162 C i 9 H 3 0 O 4 ( A M 1.9 ppm); C 2 3 H 2 6 0 6 ( A M -0.51 ppm); [ a ] D 2 0 = -7.2°, (c = 1.2, M e O H ) ; U V (MeOH) X m a x 259 nm (log £ 2.039); ' H N M R (CDC1 3 , 400 M H z ) , 1 3 C N M R (CDC1 3 , 100 M H z ) , and 2D N M R data listed in Table 4.9. 225 APPENDIX 3 JUL JUL J U din i i ^ ^ j u i i . as ^ K a© ©.© g • • to a 2899 :: B-i if S -o JO 0 O p • 0 o B o • (ppm) Figure A . l . C O S Y spectrum of eleutherobin (61) recorded in D M S O - ^ 6 at 500 M H z . 226 Figure A .2 . H M Q C spectrum of eleutherobin (61) recorded in D M S O - ^ 6 at 500 M H z . Figure A . 3 . H M B C spectrum of eleutherobin (61) recorded in D M S O - ^ 6 at 500 M H z . 228 Figure A.4 . C O S Y spectrum of desacetyleleutherobin (70) recorded in DMSO-rf6 at 500 M H z . 229 OH 7 00 6.00 5.00 4.00 3.00 2.0C 1.00 (ppm) Figure A . 5 . H M Q C spectrum of desacetyleleutherobin (70) recorded in D M S O - J 6 at 500 M H z . 230 Figure A.6 . H M B C spectrum of desacetyleleutherobin (70) recorded in DMSO-c/6 at 500 M H z . Figure A.7 . C O S Y spectrum of isoeleutherobin A (71) recorded in D M S O - J 6 at 500 M H z . 7.00 6.00 5.00 4.00 3.00 2.00 (ppm) Figure A.8 . H M Q C spectrum of isoeleutherobin A (71) recorded in D M S O - ^ 6 at 500 M H z . 233 Figure A.9 . H M B C spectrum of isoeleutherobin A (71) recorded in D M S O - J 6 at 500 M H z . 234 (ppm) Figure A . 10. C O S Y spectrum of Z-eleutherobin (72) recorded in D M S O - d 6 at 500 M H z . 235 Figure A . 11. H M Q C spectrum of Z-eleutherobin (72) recorded in DMSO-J6 at 500 M H z . 236 (ppm) OH 0- -<>4" OMei 2 OAc ,0 OH 72 JUL_J_JU 7.00 6.00 5.00 (ppm) e a 3.00 2.00 1.00 Figure A . 12. H M B C spectrum of Z-eleutherobin (72) recorded in D M S O - J 6 at 500 M H z . Figure A . 13. C O S Y spectrum of caribaeoside (73) recorded in D M S O - J 6 at 500 M H z . Figure A . 14. H M Q C spectrum of caribaeoside (73) recorded in D M S O - J 6 at 500 M H z . Figure A . 15. H M B C spectrum of caribaeoside (73) recorded in D M S C M 6 at 500 M H z . Figure A . 16. R O E S Y spectrum of caribaeoside (73) recorded in D M S O - J 6 at 500 M H z . Figure A . 17. C O S Y spectrum of caribaeolin (74) recorded in D M S O - J 6 at 500 M H z . Figure A . 18. H M Q C spectrum of caribaeolin (74) recorded in DMSO-J6 at 500 M H z . Figure A . 19. H M B C spectrum of caribaeolin (74) recorded in DMSO-<i6 at 500 M H z . 244 Me 7.00 6.00 5.00 4.00 3.00 2.00 1.00 (ppm) Figure A.20. R O E S Y spectrum of caribaeolin (74) recorded in D M S O - ^ 6 at 500 M H z . (ppm) Figure A.21. COSY spectrum of debromohymenialdisine (91)HC1 recorded in DMSO-d6 at 500 MHz. Figure A.22. H M Q C spectrum of debromohymenialdisine (91)HC1 recorded in D M S O - d 6 at 500 M H z . 1 ( p p m ) (ppm) Figure A.23. H M B C spectrum of debromohymenialdisine (9l)-HCl recorded in DMS0-J6 at 500 M H z . 248 2 ' , \ l 4 H 0 85 L - J L Figure A.24. COSY spectrum of hymenialdisine (85)HC1 recorded in DMSO-d6 at 500 MHz. 249 85 (ppm) JL_JL Figure A.25. HMQC spectrum of hymenialdisine (85)-HCl recorded in DMSO-J6 at 500 MHz. 250 85 (ppm) (ppm) Figure A.26. H M B C spectrum of hymenialdisine (85)-HCl recorded in D M S O - J 6 at 500 M H z . Figure A.27. C O S Y spectrum of £-cryptofolione (130) recorded in C D C 1 3 at 500 M H z . Figure A.28. HMQC spectrum of ^ -cryptofolione (130) recorded in CDC13 at 500 MHz. Figure A.29. H M B C spectrum of ^-cryptofolione (130) recorded in C D C 1 3 at 500 M H z . T 5' 3' r 5 148 \ 4 .A. 6.4 5.6 4.0 3.2 2.4 Figure A.30. C O S Y spectrum of Z-cryptofolione (148) recorded in C D C 1 3 at 500 M H z . 148 J L J L J L JU JU Figure A.31. HMQC spectrum of Z-cryptofolione (148) recorded in CDC13 at 500 MHz. Figure A.32. H M B C spectrum of Z-cryptofolione (148) recorded in C D C 1 3 at 500 M H z . 257 O (pom) 6.0 4.0 2.0 (BP") Figure A.33. C O S Y spectrum of cryptofolione ketone (131) recorded in C D C 1 3 at 500 M H z . 258 O Figure A.34. H M Q C spectrum of cryptofolione ketone (131) recorded in C D C 1 3 at 500 M H z . 259 Figure A.35. H M B C spectrum of cryptofolione ketone (131) recorded in CDC1 3 at 500 MHz. ;ure A.36. COSY spectrum of compound 138 recorded in CDC13 at 500 MHz. Figure A.37. HMQC spectrum of compound 138 recorded in CDC13 at 500 MHz. 262 Figure A.38. H M B C spectrum of compound 138 recorded in C D C 1 3 at 500 M H z . 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0061267/manifest

Comment

Related Items