Biomimetic Organic Synthesis Edited by Erwan Poupon and Bastien Nay
Related Titles Nicolaou, K. C., Chen, J. S.
Breslow, R. (ed.)
Classics in Total Synthesis III
Artificial Enzymes
Further Targets, Strategies, Methods 2011
2005 ISBN: 978-3-527-31165-1
ISBN: 978-3-527-32958-8
Berkessel, A., Gr¨oger, H. Dewick, P. M.
Medicinal Natural Products A Biosynthetic Approach Third Edition 2009
Asymmetric Organocatalysis From Biomimetic Concepts to Applications in Asymmetric Synthesis 2005 ISBN: 978-3-527-30517-9
ISBN: 978-0-470-74168-9
Nicolaou, K. C., Snyder, S. A. Dalko, P. I. (ed.)
Classics in Total Synthesis II
Enantioselective Organocatalysis
More Targets, Strategies, Methods
Reactions and Experimental Procedures
ISBN: 978-3-527-30684-8
2007 ISBN: 978-3-527-31522-2
2003
Biomimetic Organic Synthesis
Volume 1 Alkaloids
Edited by Erwan Poupon and Bastien Nay
Biomimetic Organic Synthesis
Volume 2 Terpenoids, Polyketides, Polyphenols, Frontiers in Biomimetic Chemistry
Edited by Erwan Poupon and Bastien Nay
The Editors Prof. Dr. Erwan Poupon Universit´e Paris-Sud Facult´e du Pharmacie 5, rue Jean-Baptiste Cl´ement 92260 Chˆatenay-Malabry France Dr. Bastien Nay Museum National d’Histoire Naturelle, CNRS 57, rue Cuvier 75005 Paris France
All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate. Library of Congress Card No.: applied for British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at
. 2011 Wiley-VCH Verlag & Co. KGaA, Boschstr. 12, 69469 Weinheim, Germany
All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law. Composition Laserwords Private Ltd., Chennai Printing and Binding Cover Design Schulz Grafik-Design, Fußg¨onheim Printed in the Federal Republic of Germany Printed on acid-free paper ISBN: 978-3-527-32580-1 ePDF ISBN: 978-3-527-63477-4 ePub ISBN: 978-3-527-63476-7 Mobi ISBN: 978-3-527-63478-1
V
Foreword The beauty and diversity of the biochemical pathways developed by Nature to produce complex molecules is a good source of inspiration for chemists who want to guided in their synthetic approach by biomimetic strategies. The first biomimetic syntheses were reported at the beginning of the 20th century, with the famous examples of Collie’s and Robinson’s related to the synthesis of phenolics (orcinol) and alkaloids (tropinone). Since then, the number of reported biomimetic syntheses, especially in the last twenty years, has increased, demonstrating the power of these approaches in contemporary organic and bioorganic chemistry. Biomimetic strategies allow the construction of complex natural products in a minimum of steps which is in accordance with the ‘‘atom economy’’ principle of green chemistry and, in addition, simple reagents can be used to access the targets. Furthermore, the bioorganic consequences of such successful syntheses allow the comprehension of the biosynthetic origin of natural compounds and these processes can produce sufficient quantities of pure products to achieve biological investigations. The biomimetic synthesis field came to maturity thanks to interconnexions between biosynthetic studies and organic synthesis, especially in the total synthesis of complex molecules. Biomimetic syntheses could even be considered as the latest stage of biosynthetic studies, confirming or invalidating the intimate steps leading to natural product skeletons. For example, the Johnson’s polycyclization of squalene precursors is one of the most impressive achievements in this field. This is still organic synthesis as the reactions are taking place in the chemist’s flask under chemically controlled experimental conditions, while biosynthetic steps can involve enzymatic catalysis, at least to a certain extent. However, concerning complex biochemical transformations, the exact role of enzymes has not always been clear, and has even been questionned by synthetic chemists. The two book volumes ‘‘Biomimetic Organic Synthesis’’ fill the gap in the organic chemistry literature on complex natural products. These books gather 25 chapters from outstanding authors, not only dealing with the most important families of natural products (alkaloids, terpenoids, polyketides, polyphenols. . .), but also with biologically inspired reactions and concepts which are truly taking part in biomimetic processes. By assembling these books, the editors E. Poupon and B. Nay succeeded in gathering specialists in complex natural product chemistry
VI
Foreword
for the benefit of the synthetic chemist community. With an educational effort in discussions and schemes, and in comparing both the biosynthetic routes and the biomimetic achievements, the demonstration of the power of the biomimetic strategies will become obvious to the readers in both research and teaching areas. These books will be a great source of inspiration for organic chemists and will ensure the continued development in this exciting field. ESPCI-ParisTech Paris, France
Janine Cossy
VII
Contents to Volume 1 Preface XVII List of Contributors XIX Biomimetic Organic Synthesis: an Introduction XXIII Bastien Nay and Erwan Poupon Part I 1
1.1 1.1.1 1.1.2 1.1.2.1 1.1.2.2 1.1.3 1.1.4 1.1.5 1.2 1.2.1 1.2.1.1 1.2.1.2 1.2.2 1.2.2.1 1.2.2.2 1.2.2.3 1.2.3 1.2.3.1 1.2.3.2
Biomimetic Total Synthesis of Alkaloids 1
Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples 3 Erwan Poupon, Rim Salame, and Lok-Hang Yan Ornithine/Arginine and Lysine: Metabolism Overview 3 Introduction: Three Important Basic Amino Acids 3 From Primary Metabolism to Alkaloid Biosynthesis 5 l-Ornithine Entry into Secondary Metabolism 5 l-Lysine Entry into Secondary Metabolism 5 Closely Related Amino Acids 6 The Case of Polyamine Alkaloids 7 Biomimetic Synthesis of Alkaloids 8 Biomimetically Related Chemistry of Ornithine- and Lysine-Derived Reactive Units 9 Ornithine-Derived Reactive Units 9 Biomimetic Behavior of 4-Aminobutyraldehyde 9 Dimerization 10 Lysine-Derived Reactive Units 11 Oxidative Degradation of Free l-Lysine 11 Clemens Sch¨opf ’s Heritage: 50 Years of Endocyclic Enamines and Tetrahydroanabasine Chemistry 12 Spontaneous Formation of Alkaloid Skeletons from Glutaraldehyde 13 Biomimetic Access to Pipecolic Acids 15 Pipecolic Acids: Biosynthesis and Importance 15 Biomimetic Access to Pipecolic Acids 16
VIII
Contents
1.3 1.3.1 1.3.2 1.3.3 1.3.4 1.3.5 1.3.5.1 1.3.5.2 1.3.6 1.4 1.4.1 1.4.2 1.4.2.1 1.4.2.2 1.4.2.3 1.4.3 1.4.3.1 1.4.3.2 1.4.3.3 1.4.4 1.5 1.5.1 1.5.2 1.5.2.1 1.5.2.2 1.5.3 1.5.4 1.5.4.1 1.5.4.2 1.5.4.3 1.5.4.4 1.5.4.5 1.5.4.6 1.5.4.7 1.5.4.8
Biomimetic Synthesis of Alkaloids Derived from Ornithine and Arginine 18 Biomimetic Access to the Pyrrolizidine Ring 18 Biomimetic Syntheses of Elaeocarpus Alkaloids 19 Biomimetic Synthesis of Fissoldhimine 22 Biomimetic Synthesis of Ficuseptine, Juliprosine, and Juliprosopine 25 Biomimetic Synthesis of Arginine-Containing Alkaloids: Anchinopeptolides and Eusynstyelamide A 26 Natural Products Overview 26 Biomimetic Synthesis 26 A Century of Tropinone Chemistry 29 Biomimetic Synthesis of Alkaloids Derived from Lysine 30 Alkaloids Derived from Lysine: To What Extent? 30 Lupine Alkaloids 31 Overview and Biosynthesis Key Steps 31 Biomimetic Synthesis of Lupine Alkaloids 32 A Biomimetic Conversion of N-Methylcytisine into Kuraramine 33 Biomimetic Synthesis of Nitraria and Myrioneuron Alkaloids 34 Biomimetic Syntheses of Nitraramine 35 Biomimetic Syntheses of Tangutorine 37 Endocyclic Enamines Overview: Biomimetic Observations 39 Biomimetic Synthesis of Stenusine, the Spreading Agent of Stenus comma 39 Pelletierine-Based Metabolism 42 Pelletierine: A Small Alkaloid with a Long History 42 Biomimetic Synthesis of Pelletierine and Pseudopelletierine 43 Pelletierine (129) 43 Pseudopelletierine 44 Lobelia and Sedum Alkaloids 44 Lycopodium Alkaloids 44 Overview, Classification, and Biosynthesis 44 Biomimetic Rearrangement of Serratinine into Serratezomine A 47 Biomimetic Conversion of Serratinine into Lycoposerramine B 47 Biomimetic Interrelations within the Lycoposerramine and Phlegmariurine Series 49 When Chemical Predisposition Does Not Follow Biosynthetic Hypotheses: Unnatural ‘‘Lycopodium-Like’’ Alkaloids 50 Total Synthesis of Cermizine C and Senepodine G 51 Biomimetic Steps in the Total Synthesis of Fastigiatine 52 Biomimetic Steps in the Total Synthesis of Complanadine A 53 References 54
Contents
2
2.1 2.2 2.2.1 2.2.2 2.3 2.4 2.4.1 2.4.2 2.4.3
3
3.1 3.1.1 3.1.2 3.2 3.3 3.3.1 3.3.2 3.3.3 3.3.4 3.3.5 3.3.6 3.4 3.5 3.5.1 3.5.2 3.6
Biomimetic Synthesis of Alkaloids Derived from Tyrosine: The Case of FR-901483 and TAN-1251 Compounds 61 Huan Liang and Marco A. Ciufolini Introduction 61 Biomimetic Total Syntheses of FR-901483 and TAN-1251 Compounds 63 Snider Synthesis of FR-901483 64 Snider Synthesis of TAN-1251 Substances 67 Oxidative Amidation of Phenols 71 Biomimetic Syntheses of FR-901483 and TAN-1251 Compounds via Oxidative Amidation Chemistry and Related Processes 77 Sorensen Synthesis of FR-901483 78 Honda Synthesis of TAN-1251 Substances 79 Ciufolini Synthesis of FR-901483 and TAN-1251C 80 References 86 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Indolemonoterpene Alkaloids 91 Sylvie Michel and Fran¸cois Tillequin Introduction 91 Indolemonoterpene Alkaloids 91 Classification and Botanical Distribution 91 Biomimetic Synthesis of Indolomonoterpene Alkaloids with a Non-rearranged Monoterpene Unit: Aristotelia Alkaloids 93 Biomimetic Synthesis of Secologanin-Derived Indolomonoterpene Alkaloids 96 Strictosidine, Vincoside, and Simple Corynanthe Alkaloids: Heteroyohimbines and Yohimbines 96 Antirhine Derivatives 99 Conversion of the Corynanthe Skeleton into the Strychnos Skeleton 99 Fragmentation and Rearrangements of Corynanthe Alkaloids: Ervitsine-, Ervatamine-, Olivacine-, and Ellipticine-Type Alkaloids 102 Iboga and Aspidosperma Alkaloids 106 Fragmentation and Rearrangements of Aspidosperma Alkaloids: Vinca Alkaloids and Rhazinilam 106 Biomimetic Synthesis of Secologanin-Derived Quinoline Alkaloids 109 Biomimetic Synthesis of Dimeric Indolomonoterpene Alkaloids 110 Anhydrovinblastine and the Anticancer Vinblastine Series 110 Strellidimine 113 Conclusion 113 References 114
IX
X
Contents
4
4.1 4.2 4.2.1 4.2.2 4.2.3 4.3 4.3.1 4.4
5 5.1 5.2 5.2.1 5.2.2 5.2.3 5.2.4 5.2.5 5.2.6 5.2.7 5.2.8 5.2.9 5.2.10 5.3
6 6.1 6.2 6.2.1 6.2.2 6.2.3 6.3
Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids 117 Timothy R. Welch and Robert M. Williams Introduction 117 Prenylated Indole Alkaloids 117 Dioxopiperazines Derived from Tryptophan and Proline 119 Dioxopiperazine Derived from Tryptophan and Amino Acids other than Proline 122 Bicyclo[2.2.2]diazaoctanes 126 Non-prenylated Indole Alkaloids 141 Epidithiodioxopiperazines 141 Conclusion 146 Acknowledgment 147 References 147 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus 149 Tanja Gaich and Johann Mulzer Introduction 149 Individual Examples 150 (±)-Camptothecin 150 (±)-Discorhabdins C and E 154 (±)-Brevianamides, Paraherquamides, VM55599, and Marcfortines 155 (+)-Stephacidin A and (−)-Stephacidin B 158 (±)-Chartelline C 160 (+)-Welwitindolinone A and (−)-Fischerindole I 164 (−)-Gelselegine 166 Communesin, Calycanthines, and Chimonanthines 168 (+)-11,11 -Dideoxyverticillin A 171 (±)-Borreverine and (±)-Isoborreverine 173 Conclusion 175 References 175 Biomimetic Synthesis of Manzamine Alkaloids 181 Romain Duval and Erwan Poupon Introduction 181 Two Complementary Hypotheses: An ‘‘Acrolein Scenario’’ and a ‘‘Malondialdehyde Scenario’’ 182 From Fatty Aldehydes Precursors to Simple 3-Alkyl-Pyridine Alkaloids 182 Biomimetic Synthesis of Dihydropyridines and Dihydropyridinium Salts 188 A Tool Box of Biomimetic C5 Reactive Units from the ‘‘Old’’ Zincke Reaction 189 Biomimetic Synthesis of Pyridinium Marine Sponge Alkaloids 191
Contents
6.3.1 6.3.2 6.3.3 6.3.4 6.3.4.1 6.3.4.2 6.4 6.4.1 6.4.2 6.4.2.1 6.4.2.2 6.4.3 6.4.3.1 6.4.3.2 6.4.3.3 6.5
6.5.1 6.5.2 6.6 6.6.1 6.6.2 6.7 6.7.1 6.7.2 6.7.3 6.8 6.9 6.9.1 6.9.2 6.9.3 6.9.4 6.9.5 6.9.6 6.9.7
Biomimetic Total Synthesis of Cyclostellettamine B and Related 3-Alkylpyridiniums 191 Biomimetic Synthesis of Xestospongins and Related Structures 191 Is the Zincke-Type Pyridine Ring-Opening Biomimetic? 193 Alkylpyridines with Unusual Linking Patterns 194 Biomimetic Synthesis of Pyrinodemin A 194 Biomimetic Synthesis of Pyrinadine A 195 Development of Baldwin’s Hypothesis: From Cyclostellettamines to Keramaphidin-Type Alkaloids 195 Linking Pyridinium Alkaloids and Manzamine A-Type Alkaloids 195 Biomimetic Total Synthesis of Keramaphidin B 197 Model Studies (1994) 197 Total Synthesis of Keramaphidin B (1998) 197 Drawbacks of the ‘‘Acrolein’’ Scenario 198 Very Low Yield of the Endo-Intramolecular Diels–Alder Reaction 198 Undesirable Transannular Hydride Transfers 199 Conversion of a ‘‘Keramaphidin’’ Skeleton into an ‘‘Ircinal/Manzamine’’ Skeleton Was Not Experimentally Possible 200 ‘‘Malondialdehyde Scenario:’’ A Modified Hypothesis Placing Aminopentadienals as Possible Precursors of Manzamine Alkaloids 200 Keramaphidin/Ircinal Connection 200 Halicyclamine Connection 201 Testing the Modified Hypothesis in the Laboratory 203 Biomimetic Models toward Manzamine A 203 Biomimetic Models toward Halicyclamines 205 Biomimetic Approaches toward Other Manzamine Alkaloids 208 Biomimetic Models of Madangamine Alkaloids 208 Biomimetic Model of Nakadomarine A 210 Biomimetic Models of Sarains: A Side Branch of the Manzamine Tree 211 A Biomimetic Tool-Box for the Synthesis of Manzamine Alkaloids: Glutaconaldehydes and Aminopentadienals 213 Biosynthesis of Manzamine Alkaloids: Towards a Universal Scenario 215 From Fatty Acids to Long-Chain Aminoaldehydes and Sarain Alkaloids 215 Pyridine Alkaloids: Theonelladine, Cyclostellettamine, and Xestospongin-Type Alkaloids 215 From Cyclostellettamines to Keramaphidin and Halicyclamine/Haliclonamine Alkaloids 218 Spinal Cord of Manzamine Metabolism: The Ircinal Pathway 218 From Ircinal and Pro-ircinals to Manzamine A Alkaloids 218 From Pro-ircinals to Madangamine Alkaloids 218 From Pro-ircinals to Manadomanzamine Alkaloids 219
XI
XII
Contents
6.9.8 6.10 6.11
From Ircinals and Pro-ircinals to Nakadomarine Alkaloids 219 Total Syntheses of Manzamine-Type Alkaloids 219 Conclusion 220 References 221
7
Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids 225 J´erˆome Appenzeller and Ali Al-Mourabit Introduction 225 Introduction to Pyrrole-2-Aminoimidazole (P-2-AI) Marine Alkaloids 226 Proposed Biogenetic Hypothesis for Clathrodin (1) and Related Monomers Starting from α-Amino Acids 229 Ground Work of George B¨uchi: Dibromophakellin (7) Synthesis from Dihydrooroidin (31) 233 Biomimetic Synthesis of P-2-AI Linear and Polycyclic Monomers 234 Biomimetic Synthesis of Linear Monomers 237 Debromodispacamides B (18) and D (39) and Dispacamide A (4) 237 Clathrodin (1) and Its Brominated Derivative Oroidin (3) 237 Biomimetic Synthesis of Cyclized Monomers 238 Cyclooroidin (48) 238 Dibromoagelaspongin (6) 238 Dibromophakellin (7) and Dibromophakellstatin (69) 243 Hymenialdisines (91) 247 Agelastatins 250 Biomimetic Synthesis of P-2-AIs Simple Dimers 253 Mauritiamine 253 Sceptrins, Ageliferins, and Oxysceptrins 254 Biomimetic Synthesis of Complex Dimers: Palau’amine and Related Congeners 255 Common Chemical Pathway for P-2-AI Biosynthesis 256 First Proposal Based on a Diels–Alder Key Step 257 Universal Chemical Pathway 257 Intramolecular Aziridinium Mediated Mechanism for the Formation of Massadine (141) from Massadine Chloride (155) 259 Aziridinium Mechanism for the Formation of the Tetramer Stylissadine A 259 Synthetic Achievements 261 Axinellamines A/B 262 Massadine Chloride (149) and Massadine (135) 263 Palau’amine (11) 265 New Challenging P-2-AI Synthetic Targets and Perspectives 266 References 267
7.1 7.1.1 7.1.2 7.2 7.3 7.3.1 7.3.1.1 7.3.1.2 7.3.2 7.3.2.1 7.3.2.2 7.3.2.3 7.3.2.4 7.3.2.5 7.4 7.4.1 7.4.2 7.5 7.5.1 7.5.2 7.5.3 7.5.4 7.5.5 7.5.6 7.5.6.1 7.5.6.2 7.5.6.3 7.6
Contents
8 8.1 8.2 8.2.1 8.2.2 8.3 8.3.1 8.3.2 8.3.3 8.4 8.4.1 8.4.2 8.4.3 8.4.4 8.5 8.5.1 8.5.2 8.6
9
9.1 9.1.1 9.1.2 9.1.3 9.2 9.2.1 9.2.2 9.2.3 9.2.4 9.2.5 9.2.6 9.3 9.3.1 9.3.2 9.3.3 9.3.4
Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin Edmond Gravel Introduction 271 Galbulimima Alkaloids 271 Alkaloids of Class I 272 Alkaloids of Class II and Class III 273 Cyclic Imine Marine Alkaloids 275 Symbioimine and Neosymbioimine 276 Pinnatoxins and Pteriatoxins 279 Gymnodimine and Derivatives 282 Other Polyketide Derived Alkaloids 284 Cassiarins A and B 284 Decahydroquinoline Alkaloids 285 Zoanthamine Alkaloids 288 Azaspiracids 291 Alkaloids Derived from Terpene Precursors 293 Cephalostatins and Ritterazines 294 Daphniphyllum Alkaloids 298 Conclusion 305 References 307 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids 317 Hans-Dieter Arndt, Roman Lichtenecker, Patrick Loos, and Lech-Gustav Milroy Introduction 317 Peptide Alkaloids: An Overview 317 Sources of Peptide Alkaloids 318 Key Features of Biosynthesis 319 Azole-Containing Peptide Alkaloids 321 Structural Features 321 Biomimetic Elements in Azole-Containing Peptide Alkaloids 323 Thiangazole 324 Lissoclinamide 7 326 Thiostrepton 328 GE2270A 334 Peptide Alkaloids Cyclized by Oxidation of Aryl Side Chains 336 Cyclic Peptides Containing Aryl-Alkyl Ethers 336 Cyclic Peptides Containing Biaryl Ethers 339 Cyclopeptides Containing Biaryls 344 Vancomycin 345 References 350
271
XIII
XIV
Contents
10
10.1 10.1.1 10.1.2 10.1.2.1 10.1.2.2 10.1.2.3 10.1.3 10.1.3.1 10.1.3.2 10.1.3.3 10.1.4 10.1.4.1 10.1.4.2 10.1.4.3 10.1.5 10.1.5.1 10.1.5.2 10.1.5.3 10.1.5.4 10.1.5.5 10.1.5.6 10.2 10.2.1 10.2.2 10.2.3 10.2.4 10.3
Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids 357 Hans-Dieter Arndt, Lech-Gustav Milroy, and Stefano Rizzo Indole-Oxidized Cyclopeptides 357 Introduction 357 TMC-95A–D 358 Formation of the Trp-Tyr Biaryl Bond by Metal-Catalyzed Cross Coupling 361 Stereocontrolled Oxidation of the Oxindole Fragment 361 Late-Stage Stereoselective (Z)-Enamide Formation 362 Celogentin C 363 Intramolecular Knoevenagel Condensation/Radical Conjugate Addition 366 C–H Activation–Indolylation 367 NCS-Mediated Oxidative Coupling 368 Himastatin and Chloptosin 369 Synthesis of the Himastatin Pyrroloindole Core 372 Synthesis of the Chloptosin Pyrroloindole Core 373 Macrolactamization 373 Diazonamide 375 Late-Stage Aromatic Chlorination 378 Bisoxazole Ring System via Oxidative Dehydrative Cyclization 379 Oxidative Annulation 379 Sequential Nucleophilic 1,2-Addition, Electrophilic Aromatic Substitution 380 Reductive Aminal Formation 380 Indole–Indole Coupling 381 A Complex Peptide Alkaloid: Ecteinascidine 743 (ET 743) 382 Biosynthesis and Biomimetic Strategy 383 Pentacycle Formation 385 Bridge Formation 389 Endgame 390 Outlook 391 References 392
Contents to Volume 2 Part II Biomimetic Synthesis of Terpenoids and Polyprenylated Natural Compounds 395 11
Biomimetic Rearrangements of Complex Terpenoids 397 Bastien Nay and Laurent Evanno
Contents
12
Polyprenylated Phloroglucinols and Xanthones 433 Marianna Dakanali and Emmanuel A. Theodorakis Part III
Biomimetic Synthesis of Polyketides
469
13
Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings 471 Gr´egory Genta-Jouve, Sylvain Antoniotti, and Olivier P. Thomas
14
Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides 503 Bastien Nay and Nassima Riache
15
Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening 537 Ivan Vilotijevic and Timothy F. Jamison
16
Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products 591 James Burnley, Michael Ralph, Pallavi Sharma, and John E. Moses Part IV
Biomimetic Synthesis of Polyphenols 637
17
Biomimetic Synthesis and Related Reactions of Ellagitannins 639 Takashi Tanaka, Isao Kouno, and Gen-ichiro Nonaka
18
Biomimetic Synthesis of Lignans 677 Craig W. Lindsley, Corey R. Hopkins, and Gary A. Sulikowski
19
Synthetic Approaches to the Resveratrol-Based Family of Oligomeric Natural Products 695 Scott A. Snyder
20
Sequential Reactions Initiated by Oxidative Dearomatization. Biomimicry or Artifact? 723 Stephen K. Jackson, Kun-Liang Wu, and Thomas R.R. Pettus Part V Frontiers in Biomimetic Chemistry: From Biological to Bio-inspired Processes 751
21
The Diels–Alderase Never Ending Story Atsushi Minami and Hideaki Oikawa
753
22
Bio-Inspired Transfer Hydrogenations 787 Magnus Rueping, Fenja R. Schoepke, Iuliana Atodiresei, and Erli Sugiono
XV
XVI
Contents
23
Life’s Single Chirality: Origin of Symmetry Breaking in Biomolecules 823 Michael Mauksch and Svetlana B. Tsogoeva Part VI
24
Conclusion: From Natural Facts to Chemical Fictions 847
Artifacts and Natural Substances Formed Spontaneously 849 Pierre Champy Index 935
XVII
Preface When we decided to start this project, at the end of 2008, we were perfectly aware that the amount of work to provide on it, the Biomimetic Organic Synthesis saga, would be very important. In fact, we were far from reality since the field not only concerns the huge universe of natural product chemistry, but also tends to embrace many fields beyond. We tried to design this book according to natural product chemistry principles, mainly by compound classes, and hope that few of them slipped our notice. Hopefully, the contributors who were asked to write a chapter in their respective field have welcomed this project with a great enthusiasm and worked hard to finish their chapter on time. Our editing adventure is now ending and we want now to warmly thank all of them for their outstanding contribution to this lengthy book. We also want to pay tribute to Professor Franc¸ois Tillequin, so happy with natural product chemistry, who recently passed away. Special thanks are also due to the staff of Wiley-VCH especially to Dr Gudrun Walter and Lesley Belfit for excellent collaboration. Biomimetic synthesis is the construction of natural products by chemical means using Nature’s hypothetical or established strategies, i.e. starting from synthetic mimicry of Nature’s biosynthetic precursors, ideally by way of biologically compatible reactions. In theory, this principle can be applied to all natural product classes, from the simplest to the most complex compounds. Yet the activation methods in the laboratory can be far from Nature’s enzymatic environment, and the biomimetic step can then be more difficult than expected at first glance. The way may therefore be tricky, even for a skilled chemist. We hope this book will delight readers by materializing most of organic synthesis concepts built from biochemical (biosynthetic) inspirations. Fortunately, readers may find solutions to synthetic problems or, at least, find a new way to improve their knowledge, as we did. Enjoy reading. March 2011
Erwan Poupon Universit´e Paris-Sud, Chˆatenay-Malabry, France Bastien Nay Mus´eum National d’Histoire Naturelle, Paris, France
XIX
List of Contributors Ali Al-Mourabit Centre de Recherche de Gif-sur-Yvette Institut de Chimie des Substances Naturelles UPR 2301 CNRS Avenue de la Terrasse 91198 Gif-sur-Yvette France J´erˆome Appenzeller Centre de Recherche de Gif-sur-Yvette Institut de Chimie des Substances Naturelles UPR 2301 CNRS Avenue de la Terrasse 91198 Gif-sur-Yvette France Hans-Dieter Arndt Technische Universit¨at Dortmund Fakult¨at Chemie Otto-Hahn-Strasse 6 44221 Dortmund Germany and
Max-Planck Institut f¨ur Molekulare Physiologie Otto-Hahn-Strasse 11 44227 Dortmund Germany Marco A. Ciufolini The University of British Columbia Department of Chemistry 2036 Main Mall Vancouver British Columbia V6T 1Z1 Canada Romain Duval Institut de Recherche pour le D´evelopement UMR 152 Facult´e des Sciences Pharmaceutiques 118 Route de Narbonne 31062 Toulouse France Tanja Gaich Leibniz Universit¨at Hannover Institute of Organic Chemistry Schneiderberg 1 30167 Hannover Germany
XX
List of Contributors
Edmond Gravel CEA, iBiTecS Service de Chimie Bioorganique et de Marquage 91191 Gif-sur-Yvette France Huan Liang The University of British Columbia Department of Chemistry 2036 Main Mall Vancouver British Columbia V6T 1Z1 Canada Roman Lichtenecker Technische Universit¨at Dortmund Fakult¨at Chemie Otto-Hahn-Strasse 6 44221 Dortmund Germany
Max-Planck Institut f¨ur Molekulare Physiologie Otto-Hahn-Strasse 11 44227 Dortmund Germany Sylvie Michel Universit´e Paris Descartes Facult´e de Pharmacie Laboratoire de Pharmacognosie U.M.R.-C.N.R.S. n◦ 8638 4 Avenue de l’Observatoire 75006 Paris France Lech-Gustav Milroy Technische Universit¨at Dortmund Fakult¨at Chemie Otto-Hahn-Strasse 6 44221 Dortmund Germany and
and Max-Planck Institut f¨ur Molekulare Physiologie Otto-Hahn-Strasse 11 44227 Dortmund Germany Patrick Loos Technische Universit¨at Dortmund Fakult¨at Chemie Otto-Hahn-Strasse 6 44221 Dortmund Germany and
Max-Planck Institut f¨ur Molekulare Physiologie Otto-Hahn-Strasse 11 44227 Dortmund Germany Johann Mulzer University of Vienna Institute of Organic Chemistry W¨ahringer Strasse 38 1090 Vienna Austria Erwan Poupon Universit´e Paris-Sud 11 Facult´e de Pharmacie 5 rue Jean-Baptiste Cl´ement 92260 Chˆatenay-Malabry France
List of Contributors
Stefano Rizzo Technische Universit¨at Dortmund Fakult¨at Chemie Otto-Hahn-Strasse 6 44221 Dortmund Germany and Max-Planck Institut f¨ur Molekulare Physiologie Otto-Hahn-Strasse 11 44227 Dortmund Germany Rim Salame Universit´e Paris-Sud 11 Facult´e de Pharmacie 5 rue Jean-Baptiste Cl´ement 92260 Chˆatenay-Malabry France Franc¸ois Tillequin Universit´e Paris Descartes Facult´e de Pharmacie Laboratoire de Pharmacognosie U.M.R.-C.N.R.S. n◦ 8638 4 Avenue de l’Observatoire 75006 Paris France
Timothy R. Welch Colorado State University Department of Chemistry Fort Collins, CO 80523-1872 USA Robert M. Williams Colorado State University Department of Chemistry Fort Collins, CO 80523-1872 USA Lok-Hang Yan Universit´e Paris-Sud 11 Facult´e de Pharmacie 5 rue Jean-Baptiste Cl´ement 92260 Chˆatenay-Malabry France
XXI
XIX
List of Contributors Sylvain Antoniotti Universit´e de Nice-Sophia Antipolis Facult´e des Sciences D´epartment de Chimie 28 AvenueValrose 06108 Nice Cedex 2 France Iuliana Atodiresei RWTH Aachen University Institute of Organic Chemistry Landoltweg 1 52074 Aachen Germany James Burnley University of Nottingham Faculty of Science School of Chemistry University Park Nottingham NG7 2RD United Kingdom
Pierre Champy Universit´e Paris-Sud 11 Chimie des Substances Naturelles CNRS UMR 8076 BioCIS Facult´e de Pharmacie 5 rue Jean-Baptiste Cl´ement 92296 Chˆatenay-Malabry France Marianna Dakanali University of California San Diego Department of Chemistry and Biochemistry 9500 Gilman Drive La Jolla San Diego, CA 92093-0358 USA Laurent Evanno Mus´eum National d’Histoire Naturelle Unit´e Mol´ecules de Communication et Adaptation des Micro-organismes associ´ee au CNRS (UMR 7245) 57 rue Cuvier 75005 Paris France
XX
List of Contributors
Gr´egory Genta-Jouve Universit´e de Nice-Sophia Antipolis Facult´e des Sciences D´epartment de Chimie 28 AvenueValrose 06108 Nice Cedex 2 France Corey R. Hopkins Vanderbilt University Medical Center Department of Chemistry Department of Pharmacology Vanderbilt Program in Drug Discovery Nashville, TN 37272-6600 USA Stephen K. Jackson University of California Department of Chemistry and Biochemistry Santa Barbara, CA 93106-9510 USA Timothy F. Jamison Massachusetts Institute of Technology Department of Chemistry 77 Massachusetts Avenue Cambridge, MA 02139 USA
Isao Kouno Nagasaki University Graduate School of Biomedical Sciences Department of Molecular Medicinal Sciences 1-14 Bunkyo-machi Nagasaki 852-8521 Japan Craig W. Lindsley Vanderbilt University Medical Center Department of Chemistry Department of Pharmacology Vanderbilt Program in Drug Discovery Nashville, TN 37272-6600 USA Michael Mauksch University of Erlangen- Nuremberg Department of Chemistry and Pharmacy Henkestrasse 42 91054 Erlangen Germany Atsushi Minami Hokkaido University Graduate School of Science Division of Chemistry Sapporo 060-0810 Japan
List of Contributors
John E. Moses University of Nottingham Faculty of Science School of Chemistry University Park Nottingham NG7 2RD United Kingdom
Michael Ralph University of Nottingham Faculty of Science School of Chemistry University Park Nottingham NG7 2RD United Kingdom
Bastien Nay Mus´eum National d’Histoire Naturelle Unit´e Mol´ecules de Communication et Adaptation des Micro-organismes associ´ee au CNRS (UMR 7245) 57 rue Cuvier 75005 Paris France
Nassima Riache Mus´eum National d’Histoire Naturelle Unit´e Mol´ecules de Communication et Adaptation des Micro-organismes associ´ee au CNRS (UMR 7245) 57 rue Cuvier 75005 Paris France
Gen-ichiro Nonaka Usaien Pharmaceutical Company Ltd. 1-4-6 Zaimoku Saga 840-0055 Japan
Magnus Rueping RWTH Aachen University Institute of Organic Chemistry Landoltweg 1 52074 Aachen Germany
Hideaki Oikawa Hokkaido University Graduate School of Science Division of Chemistry Sapporo 060-0810 Japan Thomas R.R. Pettus University of California Department of Chemistry and Biochemistry Santa Barbara, CA 93106-9510 USA
Fenja R. Schoepke RWTH Aachen University Institute of Organic Chemistry Landoltweg 1 52074 Aachen Germany Pallavi Sharma University of Nottingham Faculty of Science School of Chemistry University Park Nottingham NG7 2RD United Kingdom
XXI
XXII
List of Contributors
Scott A. Snyder Columbia University Department of Chemistry Havemeyer Hall 3000 Broadway New York, NY 10027 USA Erli Sugiono RWTH Aachen University Institute of Organic Chemistry Landoltweg 1 52074 Aachen Germany Gary A. Sulikowski Vanderbilt University Medical Center Department of Chemistry Department of Pharmacology Vanderbilt Program in Drug Discovery Nashville, TN 37272-6600 USA Takashi Tanaka Nagasaki University Graduate School of Biomedical Sciences Department of Molecular Medicinal Sciences 1-14 Bunkyo-machi Nagasaki 852-8521 Japan Emmanuel A. Theodorakis University of California San Diego Department of Chemistry and Biochemistry 9500 Gilman Drive La Jolla San Diego, CA 92093-0358 USA
Olivier P. Thomas Universit´e de Nice-Sophia Antipolis Facult´e des Sciences D´epartment de Chimie 28 AvenueValrose 06108 Nice Cedex 2 France Svetlana B. Tsogoeva University of Erlangen-Nuremberg Department of Chemistry and Pharmacy Henkestrasse 42 91054 Erlangen Germany Ivan Vilotijevic Massachusetts Institute of Technology Department of Chemistry 77 Massachusetts Avenue Cambridge, MA 02139 USA Kun-Liang Wu University of California Department of Chemistry and Biochemistry Santa Barbara, CA 93106-9510 USA
XXIII
Biomimetic Organic Synthesis: an Introduction Bastien Nay and Erwan Poupon
Nature always makes the best of possible things Aristotle 1 General remarks
‘‘Biomimetic’’, ‘‘biomimicry’’ and ‘‘biologically inspired’’ are terms that can be used whenever Nature symphonic processes inspire human creation. This will encompass science, arts, architecture and so on. In this book, we will focus our attention on organic chemistry. To assemble these two volumes, we were spoiled for choice. A selection of topics was made to give a wide perspective on biomimetic synthesis, especially when dedicated to natural product chemistry and total synthesis which will constitute the major part and a guiding principle all along this book. We are of course conscious that entire fields are left aside such as material chemistry or supramolecular chemistry. Yet we wish that this book will convince most readers that applying biomimetic strategies to organic synthesis can provide a shortcut toward efficiency, beauty and originality, in a wide scope of fields (Figure 1).
2 Natural products as a vital lead
For many decades, the question why living organisms of all kingdoms produce secondary metabolites (‘‘natural products’’) has been the subject of many debates. As soon as the first structures were determined, chemists also started thinking about the possible origin of the molecules [1]. Natural products are at the center of chemical ecology and have been forged in the crucible of Darwinian evolution. Many theories have tried to explain the incredible diversity of natural substances, including appealing views concluding that the living organisms that may be selected by evolution are the ones that favor chemical diversity [2], which may be the product of biochemical combinatorial processes. It is needless to remind here the importance of secondary metabolites
XXIV
Biomimetic Organic Synthesis: an Introduction
state of the art tools for bioorganic chemistry e.g.:
natural product chemistry, e.g.: - state of the art strategies for total synthesis - biosynthetic pathways understanding - chemical interrelations state of the art tools for organic chemistry e.g.: - cascade reactions, multicomponent reactions - organocatalysis - Diversity Oriented Synthesis - green chemistry
- Diversity Oriented Synthesis - chemical biology applications - prebiotic chemistry domains of organic synthesis that can benefit from biomimetic strategies
Figure 1
Tentacular influence of biomimetic strategies.
for Humanity notably as a source of drug candidates, pharmaceuticals, flavours, fragrances, food supplements. This aspect has been widely covered over the years. Back to the biological functions, activities of natural substances may be explain because they interact with and modulate almost all type of biological targets including proteins (enzymes, receptors, and cytoskeleton), membranes, or nucleic acids. Here again, important notions such as the conservation of protein domains in living organisms or the selection of privileged scaffolds have been discussed and should not be ignored by chemists interested in natural substances [3].
3 Biomimetic synthesis
Biomimetic synthesis is the construction of natural products by chemical means using Nature’s hypothetical or established strategy. It therefore stands in close relation with biosynthetic studies. Engaged in biomimetic strategies, the chemists 30 25 20 15 10 5
10
08
20
06
20
04
20
20
02
00
Figure 2 Analysis of bibliographical search in SciFinder with the terms ‘‘biomimetic total synthesis’’ from 1960 to 2010, leading to 339 references (30/10/2010).
20
98
20
96
19
94
19
92
19
90
19
88
19
86
19
84
19
82
19
80
19
78
19
19
19
76
0
Biomimetic Organic Synthesis: an Introduction
will face pragmatic issues for planning their synthesis but will undoubtedly wonder about the exact role of enzymes in nature’s way to construct sometimes highly complex structures. From highly evolved biosynthetic pathways involving enzymes with very high selectivity to less evolved routes or less specific enzymic catalysis, secondary metabolism pathways embrace a wide range of chemical efficiency. Biomimetic strategies will often, usually unintentionally, point out this aspect. An increasing number of total syntheses have been termed ‘‘biomimetic’’ or ‘‘biosynthetically inspired’’ and so on, especially during the last decade. Basically a quick search in SciFinder, using the term ‘‘biomimetic total synthesis’’ over the period 1960–2010, afforded 339 occurrences, beginning in 1976. As illustrated in Figure 2, the last ten years have shown an increasing number of publications in this field. Different situations and ‘‘degree of biomimicry’’ can then be instinctively distinguished when closely analyzing the final total synthesis including: • a total synthesis featuring a biomimetic crucial step after a multistep total synthesis of the natural product precursor; • a total synthesis featuring a biomimetic cascade reaction from more or less simple precursors. Many examples in both situations will be found in this book. Simple parameters can at first sight help defining the relevance of the biomimetic step especially in terms of complexity generation. These include among others: the number of new carbon-carbon bonds and cycles formed, the number of changes in hybridization state of carbon atoms, the global oxidation state and also the stereochemical changes. The chemical reactions borrowed from Nature tool box for building carbon-carbon bonds that will emerge will particularly stand out century-old reactions such as aldolization, Claisen condensation, Mannich reaction and Diels-Alder and other cycloadditions. Situations where self-assembly relies merely on inherent reactivity of the precursors are probably situations that will be most likely mimicked successfully in the laboratory [4]. Beautiful examples will be presented in this book. Since simplicity should be the hallmark of total syntheses approaching the perfect or ideal total synthesis [5], the use of biomimetic strategies can advantageously bring solutions to intricate synthetic problems [6]. Let us finally add that by many aspects we will not debate here, biomimetic strategies may fulfill the criteria of ‘‘green chemistry’’ and ‘‘atom economy’’ when exploiting for example multicomponent strategies [7].
4 On the organization of the book, in close relation with secondary metabolism biochemistry
The chief purpose of this book is not to give a full coverage of the main biosynthetic pathways of secondary metabolites. Yet a particular care has been brought by
XXV
XXVI
Biomimetic Organic Synthesis: an Introduction
authors in providing basic key elements of biosynthesis in the different chapters, to make them as comprehensive as possible to readers. If more has to be known about biosynthetic elements, we suggest referring to excellent books that have already covered the subject [8]. 4.1 Alkaloids
An alkaloid is a cyclic organic compound containing nitrogen in a negative oxidation state which is of limited distribution among living organisms. This is a modern definition for a heterogeneous class of natural substances given by S. W. Pelletier in the first volume of the series of famous periodical books Alkaloids [9]. In our Biomimetic Organic Synthesis, all chapters related to alkaloids have been gathered in the first volume of this edition. Many classifications were proposed for this class of compounds. They could be based on the biogenesis, structure, biological origin, spectroscopic properties or also biological properties. The great lack of general principles towards a unified classification is obvious and the borderline between alkaloids sensu stricto and other natural nitrogen-containing secondary metabolites (such as peptides or nucleosidic compounds) is often unclear. A classification based on the nitrogen source of the alkaloid will guide
Chapter 2.1 by R.Salame, L.-Y. Yan and E. Poupon
Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples O Me H H HO H H2 N N N H H N CO2H HO H2 N H L-ornithine NH N O NH2 "C4N" units H N H2N NH O H H NH O CO2H NH (+)-cycloanchinopeptolide C H2 N N L-arginine H NH2 O OH O O N L-lysine H2N CO2H "C5N" units CH3 N H N NH2 OH nitraramine serratezomine A Chapter 2.2 by H. Liang and M. A. Ciufolini
Biomimetic synthesis of alkaloids derived from tyrosine: the case of FR-901483 and TAN 1251 compounds OMe NH2
HOOC
NMe
2×
HO
HO
N
H
N
O
H
NHMe O
L-tyrosine
OP
Scheme 1
FR-901483
TAN-1251A
Alkaloids derived from ornithine/arginine, lysine and tyrosine.
OH
Biomimetic Organic Synthesis: an Introduction
our choice of topics tackled in this book. This has the advantage of linking the biosynthetic origin (and thereby the biomimetic approach) and the chemical structure of the secondary metabolites. Accordingly, chapters will be devoted to alkaloids primarily deriving from ornithine, arginine, lysine, tyrosine (Scheme 1) and of course tryptophan (which highly diverse chemistry will be envisaged in three chapters, Scheme 2). Particularly, we thought that the important class of indolomonoterpenic alkaloids, despite largely discussed along the years, deserved an overview chapter putting forward crucial ideas and challenges when approaching their chemistry. A large array of natural substances isolated from microorganisms displays a diketopiperazine ring system in more or less rearranged form. Because of constant efforts towards the comprehension of their biosynthesis and their total synthesis, a chapter is dedicated to these alkaloids. In fact, they are probably among the secondary metabolites that have largely benefited from biomimetic strategies
Chapter 2.3 by S. Michel and F. Tillequin
Biomimetic synthesis of alkaloids derived from tryptophan: indolemonoterpene alkaloids H NH
N H H
NH H
H H3COOC
O
H
H N
Glc
N
N
O
H
O
H strychnine
aristoteline
strictosidine Chapter 2.4 by T. R. Welch and R. M. Williams
Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids
O
H N
N N H
O N
O
O
N
Me
HN O
N H
H
HN
Me Me H N
O
N
Me
OH O
brevianamide A
H
okaramine N
Chapter 2.5 by T. Gaich and J. Mulzer
Biomimetic synthesis of alkaloids with a modified indole nucleus O N
O N N
Br O OH O
camptothecin
Scheme 2
N N Br N H chartelline C
Alkaloids derived from tryptophan.
Cl Me
Cl
CN
H Me Me O
N H welwitindolinne A
H O
XXVII
Biomimetic Organic Synthesis: an Introduction
XXVIII
Chapter 2.6 by R. Duval and E. Poupon
Biomimetic synthesis of manzamine alkaloids
N H
N H
O
O
OH
N
O
N
N
O
N
NH2
HO N
HO
H
manzamine A
sarain A
Chapter 2.7 by J. Appenzeller and A. Al Mourabit
Biomimetic synthesis of marine pyrrole-2-aminoimidazole and guanidium alkaloids Cl
H2 N
H H N O
NH2 N
O
NH
pyrrole-2-amino-imidazole (P-2-AI)
H OH NH2 N N NH
Br NH
Br Br N H
H N O oroidine
N N H
NH2
Br
palau'amine
Scheme 3
H N N H
O
NH2
NH HN
NH NH2 sceptrine
Focus on two classes of complex marine alkaloids.
with undeniable success. Also of special interest in biomimetic chemistry, several alkaloids are derived in nature by profound modifications of the indole nucleus itself giving rise to secondary metabolites for which the biosynthetic origin is not obvious at first glance. The examples of quinine and camptothecin were among the first structures where such phenomena were suspected. Up to now, such biomimetic syntheses imply an initial oxidation step of the indole nucleus, which selected examples are disclosed in a proper chapter. Two chapters will also cover the biomimetic synthesis of two classes of important marine alkaloids: the manzamine type alkaloids and the pyrrole-2-aminoimidazole alkaloids (Scheme 3). Alkaloids encompass also secondary metabolites obviously deriving from terpenes/steroids or polyketides, they will be considered as well in an individualized chapter (Scheme 4). Despite more related to polyketides in terms of biosynthetic machinery (see below), peptides alkaloids will be covered by two chapters in this section (Scheme 5). 4.2 Terpenes and terpenoids
Terpenes and terpenoids will be covered by chapters of the second volume of Biomimetic Organic Synthesis. They are made by terpene cyclases which catalyze
Biomimetic Organic Synthesis: an Introduction
XXIX
Chapter 2.8 by E. Gravel
Biomimetic synthesis of alkaloids with a non amino-acid origin polyketide alkaloids Me O
O
MeMe
HH
O
O
O
Me AcO
Me
Me
Me
Galbulimima alkaloid 13
zoanthamine
Scheme 4
daphniphylline
NH Me H H O H
Me O
O
HO
N Me
Me O
terpenoid alkaloids
Me
N
Polyketide and terpenoid alkaloids.
Chapter 2.9 by H.-D. Arndt, R. Lichtenecker, P. Loos and L.-G. Milroy
Biomimetic synthesis of azole- and aryl-peptide alkaloids HO H
HO H N
H2 N
OH OH H N
O
N H
NH2
NH
N
S
H O
N
HN
NH2
HS OH Me
HO thiostrepton
O
H
Me
N Me S O
S HN N
H N H Me
S O
OH N
O
O
H N
N H
O
H N
N H
N
N
O
OH O NH
HN biphenomycin C
O NH
Me
OH
O
N H H OH O
O
Me
O
O
HN
H N
Me O
H Me Me
Me
OH
Chapter 2.10 by H.-D. Arndt, L.-G. Milroy, S. Rizzo
Biomimetic synthesis of indole-oxidized and complex peptide alkaloids Me
O HO H
Me HO NH O H
Me H N O O
O
Me N HO
N N
O
N H
diazonamide A
Scheme 5
NH
HO
Cl
O
Cl
O
H
NH
NH O N H
HO NH
MeO O CONH2 Me
O AcO
Me
OMe
HO O
S
Me
H N
Me
Me
N O
O
TMC-95A-D
N
Me
O
OH
ecteinascidin 743
H
Complex peptide alkaloids.
highly efficient reactions at the origin of such a rich chemistry. The cationic cascade aspect of terpene biosynthesis from oligomers of activated forms of isoprene is very appealing for many biomimetic endeavors. Current aspects of terpene biosynthesis include the interesting notion of accuracy of terpene cyclases, an issue that is closely
O NH2
XXX
Biomimetic Organic Synthesis: an Introduction
n
two possible biosynthetic outcomes of prenyl cation terpene synthases = terpenes cyclases
aromatic prenyl transferases
terpenes sensu stricto Chapter 3.2 by M. Dakanali and E. Theodorakis
Chapter 3.1 by B. Nay and L. Evanno
Biomimetic rearrangements of complex terpenoids HO O
H
H
O
hyperforine HO
H O CO2H
antheridic acid
Scheme 6
O
O
CO HO
O
Polyprenylated xanthones and phloroglucinol O
O
O intricarene
O
O O
O O
O
O
OH
lateriflorone
Biomimetic synthesis of terpenes and terpenoids.
related to the quest for selectivity in biomimetic synthesis of such compounds. Leading review articles have already been published elsewhere about cationic cascade cyclizations [10]. A chapter will focus on the post-polycyclization events, dealing with biomimetic rearrangements of already complex terpene structures. Polyprenylated secondary metabolites resulting primarily from the transfer of prenyl units to aromatic rings by aromatic prenyl transferases, and sometimes followed by rearrangements, will be covered by another chapter in the second volume (Scheme 6). Other aspects of terpene alkaloids have been developed in the first volume of Biomimetic Organic Synthesis (especially, the reader can refer to chapters 2.3 and 2.8). 4.3 Polyketides
Manipulations of polyketide gene clusters have contributed to a revolution in the comprehension of polyketides (PK), and also of non-ribosomal peptides (NRP), biosynthesis. The genome sequencing of numerous PK and NRP producing microorganisms has revealed a large number of cryptic metabolites mostly unknown. Current challenges include the discovery of such natural compounds by allowing the expression of the corresponding genes (‘‘turn them on’’) and the programming/reprogramming of fungal PKS. Not to be forgotten is the implication of the PK pathways in aromagenesis in nature via the biosynthesis of phenols. We therefore
Biomimetic Organic Synthesis: an Introduction O H3C
O SCoA
O
O
O
H3C
poly-β-ketoester
SCoA
n
assembly mimic aliphatic polyketides aromatic polyketides
Chapter 4.2 by B. Nay and N. Riache
Chapter 4.1 by G. Genta-Jouve, S. Antoniotti, O. P. Thomas
Biomimetic synthesis of non-aromatic polycyclic polyketides
Polyketide assembly mimics and biomimetic access to aromatic rings
Me
KS AT ACP
Me
O H3C
CH3
SCoA SCoA O O
O
O
O O
O
O O
O
O O
O
O O
O Me
SCoA
OH abyssomycine C
O H
Chapter 4.3 by I. Vilotijevica and T. F. Jamison
CO2H phomoidride B
HO H
Polyketide assembly mimics and biomimetic access to aromatic rings Me Me O O O O O O O
O O
O
O O O O
O
Me Chapter 4.4 by J. Burnley, M. Ralph, P. Sharma and J. E. Moses
Biomimetic Electrocyclisation Reactions Towards Polyketide-derived Natural Products O O
H Me O
Me Me
H
elysiapyrone B
Scheme 7
O
HO2C O O
O
HO2C
O
O H O O
O
H
O
H H
torreyanic acid
Biomimetic synthesis of polyketides.
decided to ask for a contribution on biomimetic mimics of the fundamental steps of PK assembly and phenol ring formation. Turning our attention to more complex structures, beautiful examples of biomimetic synthesis of complex non aromatic polycyclic-PKs will be presented. The two following chapters will deal with two specific classes of natural substances characterized by their seminal mechanism of formation: i.e. polyepoxide ring opening and electrocyclization (Scheme 7). 4.4 Polyphenolic compounds
Another important biosynthetic route to aromatic rings in nature is provided by the shikimate/chorismate pathway. Simple phenolic acids enter the biosynthesis
XXXI
Biomimetic Organic Synthesis: an Introduction
XXXII
Chapter 5.1 by T. Tanaka, I. Kouno, G.-i. Nonaka
Biomimetic synthesis and related reactions of ellagitannins OH HO OH OH CO2H HO HO HO
OH
OH OH O
O O O O
OH HO
OH gallic acid
O
HO
OH OH OH
HO accutissimine B
OH
O
O
HO
OH
O
O O
HO OH
Chapter 5.2 by C. W. Lindsley, C. R. Hopkins and G. A. Sulikowski
Biomimetic synthesis of lignans
OH
MeO HO
Me
OMe
O
Me
H
OH
H
H
coniferylic alcohol
O
H O O
O pinoresinol
HO
O
O
OMe
O
carpanone
Chapter 5.3 by S. A. Snyder
Synthetic approaches to the resveratrol-based family of oligomeric natural products HO
OH
HO
HO
OH
OH
resveratrol OH
Scheme 8
HO
O OH OMe hopeanol
OH
H
O
HO
OH
H H HO
O O HO
H
H O
OH
OH
vaticanol C
HO
Biomimetic synthesis of polyphenolic natural substances.
of sometimes highly complex ellagitannins, a class of hydrolysable tannins widely studied for their health benefits (Scheme 8). Among phenylpropanoids natural substances directly derived from chorismate are the lignans that are discussed in the following chapter. Typical extended phenylpropanoids include compounds such as flavonoids and stilbenes. The chemistry of flavonoids has been widely studied and reviewed over the years [11]. This is not the case for natural substances deriving from resveratrol which hold center stage in the last few years because of the growing importance of resveratrol itself in human health, and because of new developments
Biomimetic Organic Synthesis: an Introduction
in the total synthesis of this very interesting class of polycyclic molecules. For these last three classes of molecules (ellagitannins, lignans, resveratrol derived), radical phenolic couplings plays a center role as the main source of carbon-carbon bonds. 4.5 Frontiers in biomimetic synthesis
At the cross-roads of methodology and total synthesis, a few topics will show how nature observation, especially enzymic mechanisms, can lead to new discoveries in organic chemistry (Scheme 9). A discussion on the engaging issue of occurrence of the Diels-Alder reaction in nature will be conducted in a chapter. The exponential impact of organocatalysis in organic chemistry will be illustrated by the challenging problem of transfer hydrogenations in a bio-inspired manner. Once again a plethora of review articles and books deals with the other aspects of organocatalysis [12]. Finally, by many aspects, biomimetic organic chemistry may be closely linked Chapter 6.1 by A. Minami and H. Oikawa
The Diels-Alderase never ending story Mg2+
−
O
O
−
Tyr169
O
NH
H2N
R
1
Arg101 O
NH2
O R2
O
macrophomate synthase mechanism Chapter 6.2 by M. Rueping, F. R. Schoepke, I. Atodiresei and E. Sugiono
Bio-inspired Transfer Hydrogenations O H H O
2 N R
R
EtO
OEt
R1 N H
conditions O
2 HN R
R
H H O
EtO
R1
OEt N H
Chapter 6.3 by M. Mauksch and S. B. Tsogoeva
Life's single chirality: symmetry-breaking reactions OMe
OMe autocatalysis
N EtO2C
Scheme 9
O HN H
O
CO2Et
Frontiers in biomimetic organic synthesis.
XXXIII
XXXIV
Biomimetic Organic Synthesis: an Introduction Chapter 7.1 by P. Champy
Artifacts and natural substances formed spontaneously artifactual ? alkaline media air heat light acidic media silica gel solvent reactivity
O O
artifactual ? O
H
artifactual? H O
O
O
cis-Anethole
Desmosine
O H
OGlc
N N
O
O
air oxidation?
O H3CO H3CO
O
O OH
Korolkoside
artifactual light-induced isomerization?
O
O O HO
O H
H
NH2
N
Vasicolinone
from NH4OH?
Scheme 10
Artifacts as a matter of debate for the conclusion.
to prebiotic chemistry. Key-words such as spontaneous evolution, molecular and supramolecular self-organization of organic molecules can indeed refer to both domains. A chapter will be devoted to the emergence of life single chirality on earth in a manner, once again, understandable to a broad readership. Eventually, we thought that a chapter about artifacts in natural product chemistry might provide the matter of debate for an open conclusion, just to spin out the discussion (Scheme 10). May the readers enjoy their trip in the fascinating science of Biomimetic Organic Synthesis.
References 1. See, this article of great interest:
Thomas, R. (2004) Nat. Prod. Rep., 21, 224–248. 2. See among others: (a) Firn, R.D. and Jones, C.G. (2009) J. Exp. Bot., 60, 719–726 and references cited therein; (b) Jenke-Kodama, H. and Dittmann, E. (2009) Phytochemistry, 70, 1858–1866. 3. See among others: (a) Breinbauer, R., Vetter, I.R., and Waldmann, H. (2002) Angew. Chem. Int. Ed., 41, 2878–2890; (b) Bon, R.S. and Waldmann H. (2010) Acc. Chem. Res., 43, 1103–1114 and references cited therein; (c) Dobson, C.M. (2004) Nature, 432, 824–828 and references cited therein; (d) Welsch, M.E., Snyder, S.A., and Stockwell, B.R. (2010) Curr. Opin. Chem. Biol., 14, 347–361. 4. (a) Gravel, E. and Poupon, E. (2008) Eur. J. Org. Chem., 27–42; (b) E.J.
Sorensen (2003) Bioorg. Med. Chem., 11, 3225–3228. 5. (a) Wender, P.A., Handy, S.T., and Wright, D.L. (1997) Chemistry & Industry, 765; (b) Wender, P.A. and Miller, B.L. (2009) Nature, 460, 197–20; (c) Gaich, T. and Baran, P.S. (2010) J. Org. Chem., 75, 4657–4673. 6. Among other review articles, interesting thoughts and historical perspectives are discussed in: (a) Scholz, U. and Winterfeldt, E. (2000) Nat. Prod. Rep., 17, 349–366; (b) de la Torre, M.C. and Sierra, M.A. (2004) Angew. Chem. Int. Ed., 43, 160–181; (c) Heathcock, C.H. (1996) Proc. Natl. Acad. Sci. USA, 93, 14323–14327. 7. Tour´e, B.B. and Hall, D.G. (2009) Chem. Rev., 109, 4439–4486.
Biomimetic Organic Synthesis: an Introduction 8. (a) Dewick, P.M. (2009) Medicinal nat-
Bioorg. Chem., 5, 51–98; (b) Yoder, ural products: a biosynthetic approach, R.A. and Johnston, J.N. (2005) Chem. 3rd Edition, Wiley, Chichester (UK); Rev., 105, 4730–4756. 11. Andersen, Ø. M. and Markham, K.R. (b) Bruneton, J. (2009) Pharmacognosie, (Eds) (2006) Flavonoids: chemistry, biophytochimie et plantes m´edicinales, 4th chemistry, and applications, CRC Taylor Edition, Tec et Doc, Paris; (c) see also and Francis, Boca Raton. the book series: Barton, D., Nakanishi, 12. (a) Berkessel, A. and Groger, H. K., Meth-Cohn, O. (Eds) (1999) Com(2005) Asymmetric organocatalyprehensive Natural Products Chemistry, sis: from Biomimetic Concepts To 1–9, Elsevier Science Ltd, Oxford; (d) Applications In Asymmetric SyntheMander, L. and Liu, H.-W. (Eds) (2010) sis, Wiley-VCH, Weinheim; (b) Comprehensive Natural Products ChemReetz, M.T., List, B., Jaroch, S., istry II, 1–10, Elsevier Science Ltd, and Weinmann, H. (eds) (2008) Oxford; (d) See also the monthly issues Organocatalysis, Springer Verlag, Berlin; of Nat. Prod. Rep. 9. Pelletier, S.W. (1983) The nature (c) with specific applications in total synand definition of an alkaloid in thesis, see for example: Marqu´ez-L´opez, Alkaloids: Chemical and Biological PerE., Herrera, R.P., and Christmann, spectives, Vol. 1 (ed. Pelletier, S.W.) M. (2010) Nat. Prod. Rep., 27, Wiley-Interscience, New York, pp. 1138–1167. 1–32. 10. For example, see the following early and late reviews: (a) Johnson, W.S. (1976)
XXXV
1
Part I Biomimetic Total Synthesis of Alkaloids
Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
3
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples Erwan Poupon, Rim Salame, and Lok-Hang Yan
1.1 Ornithine/Arginine and Lysine: Metabolism Overview1) 1.1.1 Introduction: Three Important Basic Amino Acids
• l-Ornithine (l-1) (l-Orn, Figure 1.1) is a non-proteinogenic amino acid produced from l-glutamic acid (4) in plants and from l-arginine (2) in animals. l-Ornithine plays a central role in the urea cycle in terrestrial vertebrates [1]. • l-Arginine: With its guanidine residue, l-Arginine (l-2) (l-Arg, R) is a highly basic amino acid. It is encoded by DNA and is the direct precursor of l-ornithine (l-1), urea, and also nitric oxide. It is also be encountered in some natural products (see below) [1, 2]. • l-Lysine (l-3) (l-Lys, K): is the only amino acid to have two different biosynthetic pathways. One is the aspartate (5) pathway present in bacteria, plants, and algae. The other starts from α-ketoglutarate (6) and is present in fungi [3, 4]. Lysine is an essential amino acid for humans. Scheme 1.1 reflects some of the biochemical relations between l-ornithine (l-1)/l-arginine (l-2) and l-lysine (l-3). It is of course not the aim of this chapter to provide further details concerning their respective biosynthesis.2) Only important metabolic intermediates, helpful for a better comprehension of the following sections, have been stressed. 1) These three amino acids (Figure 1.1) share
common chemical reactivity and are implicated in more or less similar biosynthetic pathways. We thus decided, in an effort to establish useful comparisons, to garner
lysine and arginine/ornithine derived natural substances in a single chapter. 2) The interested reader is referred to classical biochemistry textbooks.
Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
4
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
pKa 13.2
NH
H2N
CO2H
N H
L-arginine
"C4N" units
pKa 1.8
NH2 pKa 9
2 (R)
(encoded by DNA)
a CO2H pKa 1.7
pKa 10.7 H N d 2 L-ornithine
NH2
1
pKa 8.7
(not encoded by DNA)
nitrogen of the derived alkaloids
pKa 10.5 H N e 2
"C5N" units
L-lysine
a CO H pKa 2.2 2
NH2 pK 8.9 a
3 (K)
(encoded by DNA)
Figure 1.1
Structure of the amino acids.
CO2H
HO2C oxaloacetic acid
CO2H
HO2C
O
CO2H
NH2 NH2
Asp 5
NH2 L-Lys L-3
citric acid cycle (Krebs cycle)
HO2C
L-pipecolic acid L-9
NH HO2C
CO2H
H2N
O 2-ketoglutarate 6
CO2H L-Arg L-2
N H
NH2 H 2N
urea cycle (animals) HO2C L-Glutamic acid
CO2H NH2
NH2 L-Ornithine 1
(Glu) 4
HO2C
Scheme 1.1
NH H2N urea CO2H
H2N
H N
H 2N L-proline L-8
Place of the three amino acids in primary metabolism.
H N
NH2 putrescine 7
1.1 Ornithine/Arginine and Lysine: Metabolism Overview
5
1.1.2 From Primary Metabolism to Alkaloid Biosynthesis
The parallel between l-ornithine (l-1) and l-lysine (l-3) metabolism concerning their catabolism/biotransformation and subsequent chemical reactivity to form alkaloids is obvious (even if the incorporated nitrogen atom is different, that is, incorporation of the α-amino group for ornithine and ε-amino group for lysine). Mainly, both amino acids will be able to undergo decarboxylation to the corresponding diamine [putrescine 7 (C4 ) and cadaverine 10 (C5 ), respectively] and then oxidative deamination into aminoaldehydes (see below). Thereby, rather stable amino acids are turned into highly reactive units suitable for natural organic chemistry. 1.1.2.1 L-Ornithine Entry into Secondary Metabolism3) The diamine putrescine (7) can be formed directly from the decarboxylation of l-ornithine (1) (Scheme 1.2); it can also be derived from l-arginine (2) [6] after decarboxylation and transformation of the guanidine functional group. Putrescine (7) is then mono-N-methylated4) by putrescine N-methyltransferase (PMT). This reaction is the first purely ‘‘secondary metabolite’’ step and 11 is the first specific metabolite towards alkaloids. N-Methylputrescine (11) may then be oxidatively deaminated by diamine oxidase to 4-methylaminobutanal (12), which generates the N-methyl-1 -pyrrolinium cation 13, a cornerstone electrophilic intermediate and a central precursor of numerous alkaloids belonging to the pyrrolidine or tropane groups when the reaction with an appropriate nucleophile occurs. CO2H
H2N
decarboxylation H2N
NH2 L-ornithine L-1
pyrrolidine alkaloids
Me N
13 Nu Me
Scheme 1.2
putrescine 7
–CO2 CH3 +N
NH2 N-methylation
− Nu
Me NH
H2N 11
–H2O
Me NH
O
oxidative deamination
12 N
tropane alkaloids
Key elements of ornithine metabolism towards alkaloids.
1.1.2.2 L-Lysine Entry into Secondary Metabolism5) Decarboxylation of l-lysine (l-3) into the diamine cadaverine (10) (Scheme 1.3) followed by oxidative deamination leads to aminopentanal 14, which can cyclize into tetrahydropyridine 15.6) This latter is most likely the universal intermediate 3) The fundamental first steps of ornithine (or
lysine) catabolism towards alkaloids now constitute a classic in biosynthesis textbooks. See, among others, Reference [5]. 4) This important step has been widely studied. PMT is closely related to spermidine synthase. See Reference [7] for a valuable review.
5) The interested reader is referred to classical
biochemistry textbooks. 6) Synonymous terms: 2,3,4,5-tetrahydropyri-
dine = 1 -piperideine (imine form); 1,2,3, 4-tetrahydropyridine = 2 -piperideine (enamine form).
6
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples O HO
NH2
–CO2
oxidative deamination
NH2 NH2
O
cadaverine 10
5-aminopentanal 14
NH2 L-lysine
3
H N
H N
Nu
−
H N H lupine alkaloids
NH2
N
–H2O
15
tetrahydroanabasine 16 Nu
H N piperidine alkaloids
Scheme 1.3
Key elements of lysine metabolism toward alkaloids.
to lysine-derived piperidine alkaloids. This imine is too unstable and reactive to be isolated as such from plant material. Nucleophilic addition reactions at the imine function with suitable nucleophiles help stabilize 15 and are at the origin of various piperidine alkaloids. Dimerization into tetrahydroanabasine (16) is an important alternative in the lysine metabolism (the enamine form of 15 being the nucleophile). This latter reaction is spontaneous at physiological pH (vide infra) though stereospecific coupling involves an appropriate enzymic intervention in plants [8]. Tetrahydroanabasine (16) (which is probably also quite unstable as such in vivo) can then undergo various transformations and is at the origin of a class of alkaloids known as ‘‘lupine alkaloids’’ (Section 1.4.2). 1.1.3 Closely Related Amino Acids
• l-Proline: (l-8) (l-Pro, P, Scheme 1.1) is one of the 20 amino acids of the genetic code but the only one with a secondary amine function. It is biosynthesized from l-glutamic acid [9]. l-Proline recently increased in importance with the successful development of organocatalysis. • l-Pipecolic acid: Unlike l-proline (l-8), l-pipecolic acid (l-9) (Scheme 1.1) is a non-proteogenic amino acid. It derives from l-lysine and is at the origin of several classes of secondary metabolites. Biosynthetic and biomimetic aspects of pipecolic acid are discussed below.7) 7) Other close amino acids exist; of particular
interest are piperazic acids (structure shown here). Their biosynthesis from lysine is discussed in Chapter 10 (Indole-Oxidized and Complex Peptide
Alkaloids); see H HO2C N NH
also
Reference
[10].
1.1 Ornithine/Arginine and Lysine: Metabolism Overview
7
1.1.4 The Case of Polyamine Alkaloids
Cases where a C4 N2 building block is incorporated [i.e., the polyamine putrescine (7)] include ‘‘polyamine alkaloids’’ (Scheme 1.4). Comprehensive review articles, especially by M. Hesse and colleagues, have appeared detailing the massive amount of work done in the field of these secondary metabolites during the last 20 years [11, 12]. Essentially, six basic backbone components, namely, putrescine (7), spermidines (17, 18), homospermidines (19, 20), spermine (21), and homospermine (22), participate in the skeleton of polyamine alkaloids. Figure 1.2 gives examples of cyclic polyamine alkaloids [piriferine (23), celacinnine (24), aphelandrine (25), and lipogrammistin A (26)], organized according to the classification of M. Hesse and colleagues (see Reference [11]). Interestingly, cadaverine units are very rarely present in such cyclic molecules. Despite interesting biomimetic syntheses [13], polyamine alkaloids will not be covered in this chapter (except for the biosynthesis of pyrrolidine alkaloids; see Section 1.3.1).
CO2H H2 N L-ornithine
CH3
H N
H2N
R
S
H2N N
Adenosyl H3C
or L-arginine
H2N C3 putrescine 7
H2N N H
H2N
H2N H2N
H2N Scheme 1.4
H2N H N
C3
O
HO CO2H
N
OH
H N
NH2
H N
NH2
H N
NH2
sym -Nor - spermidine 18
H N
NH2
homospermidine 19
H N H N
N
S -adenosylmethionine
NH2
H2N
spermidine 17 spermine 21
C4
S
N
NH2 NH2
sym -homospermidine 20 sym -homospermine 22
Main polyamine backbones encountered in polyamine alkaloids.
8
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples cyclic putrescine alkaloids: O N H
N
O piriferine 23
Ph
cyclic spermidine alkaloids: O H Ph
[Aglaia pirifera]
cyclic spermine alkaloids:
O
N H N H
N Ph
celacinnine 24
[Maytenus heterophylla...]
cyclic sym-homospermine alkaloids: (rare)
aphelandrine 25
O
O
[Aphelandra aurantiaca...]
O
H O
N H N
N H
H O N H
H
N N
O lipogrammistin A 26 NH
[Diploprion bifasciatum]
HN
Figure 1.2
Representative polyamine alkaloids.
1.1.5 Biomimetic Synthesis of Alkaloids
Ionic iminium/enamine reactions are central to the construction of the title alkaloids, especially the Mannich reaction. This is probably one reason why biomimetic strategies have been particularly efficient in this class of secondary metabolites. We will, of course, approach in this chapter only some of the multifaceted aspects of the biomimetic chemistry of l-lysine or l-ornithine/arginine derived secondary metabolites. We will select topics and examples that are not covered (or not with the scrutiny we think opportune) in other review articles. This is especially the case when dealing with the manipulation of small reactive C4 and C5 units derived from the three amino acids. Some selected examples are presented in the following sections and organized as follows:
• biomimetic syntheses from l-lysine and l-ornithine/l-arginine or C4 , C5 reactive units presumably derived from the amino acids; • selected examples of biomimetic syntheses of more complex structures.8)
8) For the implementation of highly complex
metabolisms within marine sponges and involving, according to biosynthetic proposals, l-lysine, l-arginine, and l-proline
see Chapter 7 (Biomimetic Synthesis of Marine Pyrrole-2-aminoimidazole and Guanidinium Alkaloids).
1.2 Biomimetically Related Chemistry of Ornithine- and Lysine-Derived Reactive Units
1.2 Biomimetically Related Chemistry of Ornithine- and Lysine-Derived Reactive Units 1.2.1 Ornithine-Derived Reactive Units 1.2.1.1 Biomimetic Behavior of 4-Aminobutyraldehyde The Christophersen group studied the evolution of aqueous solutions of 4-aminobutyraldehyde (27) [prepared from aminobutyraldehyde dimethyl acetal (28), Scheme 1.5] by 1 H NMR over a wide pH range (1–12) [14]. Entropic factors explain the rapid formation of cyclic imine 29, which can trimerize into 30. When in aqueous solution, different species are in equilibrium and are depicted in Scheme 1.5. Along with aminobutanal 27, two neutral (pyrrolidine 29 and trimer 30), and four protonated entities (31–34) were detected and tracked as a function of pH. Around physiological pH, the four protonated species predominate. Owing to the rapid emergence of acid-catalyzed aldol condensation products, the authors cautiously avoided concentrated solution. As we will detail in a coming section, no dimeric structure such as 35 has been characterized from the mixtures.
H3N
+
33
10% at pH 1-4 O +
OMe H2N
2 M HCl
H3N
H2N
OMe
28
HO OH
O 27
N N
50% at pH 1-6
29 100% at pH 9
20% at pH 13
H N
+
OH
32 H + N
N 30
H
31 ± H2O
N
20% at pH 1-7
N
34 20% at pH 1-7
H N 35 not found
% indicate the highest yield of the entities and the corresponding pH +
6
H N
34 (20%)
Scheme 1.5
H
H
+
N
+
OH
32 (50%)
H3N
HO
+
OH H3N
31 (20%)
O 33 (10%)
Biomimetic reactions from 4-aminobutyraldehyde (27).
With substituted nitrogen atom, for example, with a biosynthetically relevant aminobutyl side chain (Scheme 1.6), the formation of the pyrrolidinium ring is nearly exclusive, with 36 predominant among several other entities [15].
9
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
10
+N
NH3 +
+N
NH 37
36
pH 1.1-8
pH 7.2-8
N
HO
Scheme 1.6
N
NH2
NH2
pH 8.0-12.7
± H2 O
pH 9
Biomimetic behavior of a substituted pyrrolidinium ion.
Notably, aminal 37 is a bicyclic compound that is formed spontaneously at basic pH.9) 1.2.1.2 Dimerization Whereas dimerization of six-membered ring enamines is a major outcome (as we will discuss in detail when dealing with lysine-derived units), the dimerization of the corresponding five-membered ring systems such as 29 is by far less described in the literature. The chemistry of these molecules is, correspondingly, simpler than that of lysine, probably because of a lesser propensity to react as an enamine (Scheme 1.7).
+
H N
H N
29
N
? H N
N
35
Me N
Me N
Me
Me N
Me N
Me N+ 38
13
39
[Nicotiana tabacum]
Scheme 1.7
Dimerization in the pyrrolidine series.
The dimer type 35 has never been described, although it has been postulated in some biosynthesis (Section 1.3.3). In the case of N-methyl substituted 13,10) dimerization has been observed and dimer 38 characterized. Notably, the corresponding reduced dimer 39 has been detected as a mixture of diastereomers from the root system of Nicotiana tabacum, as well as monomeric 13 [18]. In fact, at physiological pH (around 7.2 in a growing tobacco plant), the coexistence of imine and enamine forms of 13 should provide the opportunity for more or less spontaneous condensations. In the laboratory (Scheme 1.8), starting from N-methyl-4-aminobutanal diethyl acetal (40) in acidic conditions, deprotection affords 12, which cyclizes into biosynthetic intermediate 13, which in turn can dimerize into 38. The product of a retro-Michael reaction, 41, has also been fully characterized; it was usually observed as an impurity in the course of the synthesis of monomers 13 [19]. Alternative 9) Diazabicyclononane 37 was isolated earlier
from in vitro chemical or enzymic conversion of spermidine with pea seedling diamine oxidase (PSDO); see Reference [16]. 10) Is monomeric 29 too reactive and unstable? N-Monomethylputrescine (11) is
in fact an early and central precursor of many alkaloids as the first specific metabolite en route to alkaloids such as nicotine or tropanes via N-methylpyrrolinium (13) (which could therefore be more stable than 15); see References [5, 17].
1.2 Biomimetically Related Chemistry of Ornithine- and Lysine-Derived Reactive Units
Me
OEt
H N
H+ Me
OEt 40
H N
Me N
42
43
Me N 38
13 Me + Cl N
−
retroMichael
HCl
Me HN
H4LiAl
41 Scheme 1.8
Me N
Me + Cl − N
Hg(OAc)2
Me N
O
O 12
11
Biomimetic dimerization in the pyrrolidine series.
procedures toward 13 consist of the oxidation (e.g., with mercuric diacetate [20]) of N-methylpyrrolidine (42) or the reduction of lactam 43 with aluminum hydrides [21]. 1.2.2 Lysine-Derived Reactive Units 1.2.2.1 Oxidative Degradation of Free L-Lysine Despite a seemingly simple pathway, mimicking the fundamental l-lysine (and l-ornithine) catabolism pathways in the laboratory is far from trivial. Only a few publications report on the direct oxidation of l-lysine. In 1966, B. Franck and colleagues oxidized l-lysine with alkaline NaOCl in water (Scheme 1.9) [22]. Cyclic oxidation products were identified and compared with authentic samples. Scheme 1.9 highlights the possible pathways toward the isolated compounds; it may be assumed that the cyclic oxidation products arise from l-lysine by one (intermediates 46, 47), two (intermediate 48) or three oxidation steps (15, 16, 44, 45). Despite an incomplete conversion of l-lysine and the complex mixtures obtained, this simple reaction is of great interest as a totally biomimetic reaction that mimics (i) the decarboxylation/oxidative deamination steps and (ii) the spontaneous evolution of the resulting reactive species (15, dimer 16, lactam 44).
oxidation
NH2
1
H2N
CO2H L-lysine 3
NH2
–NH3
H N
oxidation 2
H N
N 16
Cyclic oxidative products of L-lysine.
15 (4%)
NH 48 Cl + oxidation 3
H2O
N
(4%)
Scheme 1.9
NH2 47
46
–CO2
conditions: alkaline NaOCl (2 equiv.), 3 hrs, 20 °C
H N
NH
H N
O
44 (9%)
Cl N
NH (3%)
45
12
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
1.2.2.2 Clemens Sch¨opf ’s Heritage: 50 Years of Endocyclic Enamines and Tetrahydroanabasine Chemistry Numerous compounds resulting from the self-condensations of endocyclic enamines, which are closely related to metabolic pathways, have been described, especially in the pioneering work of Sch¨opf, starting from the late 1930s.
N N
H N
N
N N
N H
conformations a-49 based on 13C [24]
H
15
pH<2 slow
N
H N
N
fast pH8
pH>12 15 H+
−
N
H+ H N
H
15 H N
aldo-49 [Haloxylon salicornicum]
Scheme 1.10
−
isotripiperideine 50 OH
H N
N 16
N HH
16
H b-49
H
H N
OH
H N
N
9
N
N
N
N NH
conformation based on 13C [24]
NH N
Evolution of tetrahydropyridine.
The simplest compound, that is, 1 -piperideine (15), does not exist in monomeric forms but, instead, trimeric assemblies of types α- and β-49 and 50 have been isolated (Scheme 1.10).11) Structures 16, 49, and 50 were characterized by Sch¨opf in 1948 [23] and the configurations and conformations studied by the Kessler group in 1977 by 13 C NMR [24]. At neutral or slightly basic pH (∼8), and as in nature, monomers 15 dimerize into tetrahydroanabasine (16) (which can be isolated as a dihydrobromide crystalline salt in the laboratory [25]). As early as 1956, Sch¨opf and colleagues clearly demonstrated the importance of pH on the kinetic and yield of conversion of 15 into 16. Studies were conducted with pioneering and clear-sighted biosynthetic considerations (zellm¨oglichen Bedingungen, see Scheme 1.11) [26]. Tetrahydroanabasine also reacts in solution at pH 9 with imine 15 and gives in almost quantitative yield trimer 50, also called isotripiperideine [23]. Consequently, trimer 50 is, in turn, in equilibrium with tetrahydroanabasine 16 and free 1 -piperideine 15 [27] and can, therefore, be considered as a stable, protected form of tetrahydroanabasine with interesting synthetic potential. Aldotripiperideine (aldo-49) is another trimer that results from a rearrangement of α-tripiperideine in acidic conditions or in basic conditions at pH 9.2 at 100 ◦ C. Interestingly, aldo-49 was isolated from Haloxylon salicornicum as a natural substance [28]. 11) Besides, they constitute a suitable way to
store piperideine as a crystalline solid on a 100 g scale.
1.2 Biomimetically Related Chemistry of Ornithine- and Lysine-Derived Reactive Units 261. Clemens Schöpf, Franz Braun und Alfred Komzak1) : Der Übergang von ∆1-Piperidein in Tetrahydro-anabasin under zellmöglichen Bedingungen
H N
(mitbearbeitet ven Hermann Koop)
H N
[Aus dem Institut für organische Chemie der Technischen Hochschule Darmstadt] (Eingegangen am 11. April 1956)
H N
H N
% Tetrahydro-anabasin
15
90
pH7.8
80
pH7.2
70 60
pH5.9–6.0 pH9.7–10.0
50
pH4.9–5.0
40 30 20
pH11.9–12.0
10
pH2.9
pH4.1
16
5
10
15
20
25
Stdn. Ausb. an Tetrahydro-anabasin in % d.Th. beim Aufbewahren einer 0.1m Lösung von ∆1-Piperidein in verdünnten Pufferlösungen bei 25°
Scheme 1.11
Sch¨opf’s pioneering works.
1.2.2.3 Spontaneous Formation of Alkaloid Skeletons from Glutaraldehyde Glutaraldehyde (51) is a well-known crosslinking agent in biochemistry or histology and is used as a biocide.12) It can also be advantageously seen as a convenient surrogate of lysine by considering a hypothetical oxidative deamination on aminopentanal 14.13) Simple reactions (Scheme 1.12) were recently disclosed, highlighting an impressive propensity of 51 to mimic several lysine metabolism elements [29]. Whereas 51 was known to polymerize14) according to different mechanisms and kinetics (Scheme 1.12) depending, for example, on the pH, very few studies previously described compounds resulting from self-condensations into small molecules. Products formed during the treatment of 51 in an aqueous solution at pH 8.5 and 60 ◦ C were investigated. A double homoaldolization followed by crotonization of one of the aldol adducts can easily explain the formation of bicyclic 52. This compound was previously described but no information was available concerning its stereochemistry [30]. Oxidation of 52 with Dess–Martin periodinane permitted crystallization of the major diastereomer 53. Compound 52 is, interestingly, related to a biosynthetic intermediate postulated in the course of the biosynthesis of 12) The numerous applications of glutaralde- 14) Glutaraldehyde is quite stable as an aque-
hyde have been extensively reviewed: see Reference [8] in Reference [29]. 13) Of course, one needs to keep in mind that in plants reactive functional groups such as aldehydes will most likely be masked in a transitory way; they are, in this chapter, presented in their reactive form for simplicity.
ous solution, where it exists as several hydrates with the slow formation of oligomers and polymers (Scheme 1.12). It undergoes rapid polymerization in neat conditions with a catalytic amount of water.
13
14
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples O
O
O
O
n O polymers and oligomers
O
O
O
O
O
O
O
O
O
O
O
O O
biomimetic intermediate in the Nitraria genus
51 O
O
HO
O
H HO
O
O O
52 H 4 diastereomers (46/38/9/7)
O
O
O
O
Dess-Martin periodinane, CH2Cl2
dimeric molecules
O
O O
H
(80%) O H
OH H O O O
HO HN
O O
O
O
OEt
isonitramine
O
NaOH, 2 M, 100 °C
EtO
OEt EtO
OEt
55 OEt
O
N
56 (97%) O
O
OEt OEt
TFA , H2O, THF, RT
H O
N 16
OH
OH
54 [X-ray] EtOH, H +
53 major compound [X-ray]
57 (75%) oxidized analog of 16
Scheme 1.12
Various condensations of glutaraldehyde.
Nitraria alkaloids (Section 1.4.3). From the same mixture, a crystalline compound 54 that displayed a tricyclic structure with a spiranic quaternary carbon and contiguous acetal and hemiacetal functions was isolated. This intriguing molecule has a striking analogy with known simple spiroalkaloids such as nitramine also isolated from different species of the Nitraria genus and its plausible mechanism of formation totally parallels the postulated biosynthesis of such natural substances [31]. Diethoxypentanal 55, which is easily available from monoprotection of 51, has
1.2 Biomimetically Related Chemistry of Ornithine- and Lysine-Derived Reactive Units
chirality inducer
iminium stabilizer Ph
(R)-(−)-60
nitrogen source O
OH NH2
NC
−
Scheme 1.13
hydrogenolysis Ph +
O
Ph
−
N
NC
N
O
58
59
Cyanophenyloxazolopiperidine: a convenient building block.
been treated in a boiling sodium hydroxide solution to furnish quantitatively compound 56 by aldolization/crotonization as a single (E)-stereoisomer. Deprotection in acidic conditions gave 57, an interesting oxidized analog of tetrahydroanabasine. Glutaraldehyde (51) has been widely used for the synthesis of various heterocycles through the formation of dihydropyridine-type intermediates. The most powerful applications are probably the so-called ‘‘CN(R,S)’’15) method (Scheme 1.13) with the use of the chiral non-racemic N-cyanomethyloxazolidine ring system integrated in a piperidine structure as an ideal way to stabilize dihydropyridine 58 into compound 59 [prepared in a single step from glutaraldehyde 51, (R)-(−)-phenylglycinol (60) and a source of cyanide ions in water]. This strategy, developed by the Husson and Royer groups, permitted the total synthesis of many alkaloids in a diastereoselective manner. Some of them are closely related to biomimetic strategies. This strategy has been extensively reviewed over the years [32]. As, interestingly, naturally occurring piperidine alkaloids bearing α-side chains are either in the (R) or (S) configuration, the possibility of using building blocks such as 59 to modulate the stereochemistry at α or α positions constituted a real breakthrough in piperidine total synthesis. 1.2.3 Biomimetic Access to Pipecolic Acids 1.2.3.1 Pipecolic Acids: Biosynthesis and Importance l-Pipecolic acid (l-9) was first identified in 1952 as a constituent of leguminous plants [33]. It is now recognized as a universal lysine-derived entity present in plants, animals, and microorganisms [34]. In natural substances, l-9 is a key element in molecules as diverse as swainsonine (61) or castanospermine (62) (small indolizidine alkaloids known for their glycosidase-inhibiting properties, Scheme 1.14) or FK 506/tacrolimus (63) (a polyketide/non-ribosomal peptide hybrid clinically approved as an immunosuppressant). Over the years, many studies have sought to establish the biosynthetic routes to l-9. Different metabolic pathways (with different proposed mechanisms [35]) are involved in the formation of l-9 (Scheme 1.15). Chemically speaking, these basic routes are distinguishable at the loss of the amino group of lysine 3. The reality of immediate precursors, that is, piperideine carboxylic acids 64 (known as P2C) and 65 (P6C), has been 15) Named after the French Centre National
de la Recherche Scientifique (CNRS).
± −
O
51
15
∗ N −
+ −
16
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples indolizidine alkaloids
incorporation in PK/NRP natural products O
HO swainsonine 61 HO [Swainsona H canescens] HO -glycosidase inhibitorcastanospermine 62 [Castanospermum australe] -glycosidase inhibitor-
HO
N
L-pipecolic
HO
OH
H N
MeO
acid 9
Me O
N
H HO
N OH OH
O HO Me
FK 506 = tacrolimus 63 (Prograf) [Streptomyces tsukubaensis] -immunosuppressive-
OH O
O O
Me OMe Me
Me
O H OMe
Scheme 1.14 Selected example of pipecolic acid derived alkaloids and pipecolic acid containing secondary metabolites. HO2C a-oxidation H2N
∆1-piperideine-2-carboxylic acid P2C 64
a CO2H
e
NH2 L-3
N 2
HO2C
N
reduction
HO2C
H N
HO2C
H N
L-pipecolic acid L-9
6
e-oxidation ∆1-piperideine-6-carboxylic acid P6C 65
Scheme 1.15
D-pipecolic acid D-9
Biosynthesis of pipecolic acid.
demonstrated by many feeding experiments and will not be further developed in the present chapter as many research and review articles are available [33]. The reverse pathways converting l-pipecolic acid (l-9) into (P6C 65) [36] or lysine (3) [37] are also known. d-Pipecolic acid (d-9) was also reported and derives from d-lysine (whereas l-9 can be biosynthesized from both l- and d-lysine (3) [38]) and was found as a constituent of a few natural substances. 1.2.3.2 Biomimetic Access to Pipecolic Acids The chemical synthesis of pipecolic acid has been a subject of great interest. Powerful methods of asymmetric piperidine synthesis have been developed toward this aim [39] and have been reviewed [40]. We only outline, in this section, reactions directly involving l-lysine (3) (or protected l-3) to access 9 in a somewhat biomimetic way. In the 1970s, a first conversion of l-lysine (3) into optically active pipecolic acids was disclosed by Yamada and colleagues (Scheme 1.16) [41]. Sodium nitrite–hydrochloric acid was used as a deaminating agent of l-lysine, followed by barium or sodium hydroxide treatment to afford d-pipecolic acid (d-9) with more than 90% optical purity and satisfactory overall yield. Net retention of configuration was explained by the formation of lactonic intermediate 66 followed by halogeno-acid 67. On the other hand, l-lysine could be converted directly into natural l-pipecolic acid starting from ε-tosyl-l-lysine (68) but in very low yield (∼1%)
1.2 Biomimetically Related Chemistry of Ornithine- and Lysine-Derived Reactive Units NH2 CO2H H L-lysine L-3
NH2
O
66
D-pipecolic acid 9
[~30%]
O HO
NHTos
H
H N
67
O
NHTos CO2H
HO
CO2H
Cl conditions: NaNO2, HCl then Ba(OH)2 or NaOH
NH2
68
O
NH2
O
17
R N
O
R = Tos R = H:L-pipecolic acid L-9 [1% from 68]
conditions: CF3CO2H, CF3CO2Na
Tos = tosyl
Scheme 1.16
Yamada’s biomimetic access to pipecolic acids.
in strong acidic conditions. This work by Yamada and colleagues is not ‘‘truly’’ biomimetic, but is worth mentioning as it converts in a single step the acyclic skeleton of l-3 into the piperidine ring of 9. In 2004, the fully protected l-lysine 69 was converted into aminal 70 by selective oxidation of the side chain using a Mn(OAc)2 /peracetic acid system by the Rossen group (Scheme 1.17) [42]. No epimerization occurred under the buffered oxidation conditions. Treatment under mild acidic conditions gave enamide 71, which could be reduced to l-pipecolate 72. This study constitutes one of the rare examples of biomimetic conversion of a side chain amino group into an aldehyde oxidation state (obtained as the cyclic N,N-acetal 70). Boc BocHN
CO2Me 69
MeO2C
NHBoc
conditions: Mn(OAc)2 (7 mol%), AcOOH (4.2 eq.), AcONa, AcOH
Scheme 1.17
N
Boc
Boc NHBoc
MeO2C
N
AcOH 70 (77%)
MeO2C
N
H2 (Pd/C) 71 (79%)
Rossen’s biomimetic synthesis of pipecolate derivatives.
But one of the most interesting examples was probably the work disclosed by the Ohtani group (Scheme 1.18). The photocatalytic redox synthesis of pipecolic acid was achieved in a one-step procedure directly from unprotected l-lysine [43]. Several catalytic systems [various TiO2 /co-catalyst (Pt, Rh, Pd)] were investigated to define the best conditions in terms of selectivity (oxidation of ε-amino versus α-amino – which influences the final optical purity of pipecolic acid), yields, and rates. The mechanism was proved to proceed via (i) oxidation of l-lysine with positive holes, leading to P2C (64) and P6C (65), depending on the oxidized nitrogen and (ii) reduction of the imines with electron, with both steps taking place at the surface of the catalyst. Titanium oxides were shown to predominantly oxidize the ε-amino group, permitting enantiomeric excess up to 90%. The ins and outs of the multiple combinations of catalysts/co-catalysts were studied; the interested reader is referred to the original article for details. With the recourse to catalysis
72 >99% ee
18
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples photocatalytic process
oxidation predominantly oxidized with TiO2 catalysts
reduction
h+
HO2C
2e
N
−
2H
+
HO2C
H N
DL-pipecolic
acid 9 H 2N
a CO2H
e L-3
NH2
a e
+ HO2C
NH3
P2C 64
N
HO2C
H N
P6C 65
conditions: various TiO2 catalysts, co-catalysts (Pt, Pd, Rh), l > 300 nm, 25°C, Argon h+:positive holes
Scheme 1.18
L-pipecolic
acid 9 (yield up to 70% ee up to 90%)
Totally biomimetic synthesis of pipecolic acid by photocatalysis.
and the release of ammonia as the only by-product of the reaction, this synthesis is indubitably a green chemistry process and a beautiful achievement.16)
1.3 Biomimetic Synthesis of Alkaloids Derived from Ornithine and Arginine 1.3.1 Biomimetic Access to the Pyrrolizidine Ring
The pyrrolizidine nucleus consists of two fused pyrrolidine cycles; similarities between both biosyntheses may therefore be expected. These alkaloids are biosynthesized from homospermidine, which comes from putrescine (7) (Scheme 1.19). Oxidative deamination and subsequent formation of a first five-membered cycle (74) through dehydration is followed by an intramolecular Mannich reaction exploiting the enolization capacity of the remaining aldehyde of 75. Biosyntheses of such alkaloids have been reviewed, as well as their structure elucidation, chemistry, and pharmacology [44]. The direct conversion of an acyclic precursor into a pyrrolizidine-type bicyclic structure was accomplished by the Marson group in 2000 (Scheme 1.20) [45]. Treatment of compound 76 in acidic conditions permitted the deprotection of the masked aldehyde, formation of the 1 -pyrrolidine, and subsequent formation of bicyclic 77 by an aza-Prins type cyclization. Such a cationic cyclization (radical cyclizations to pyrrolizidine are known) is closely related to biosynthetic pathways and was the first example of a non-enzymic synthesis of the pyrrolizidine ring from an acyclic precursor. The relative configuration of 77, which was the sole diastereomer isolated, was ascertained by X-ray crystallography. 16) The term ‘‘biomimetic’’ does not appear in
this article and no references to biosynthesis are made.
1.3 Biomimetic Synthesis of Alkaloids Derived from Ornithine and Arginine H 2N
NH2
L-Arg 2
NH2
NH2 oxidative deamination
O H N
H N
putrescine 7
73
homospermidine 20
–H2O NH2
CHO
H
intramolecular Mannich reaction
CHO
N
Scheme 1.19
N +
75
74
Pyrrolizidine ring biosynthesis.
OEt H H N
76
Scheme 1.20
oxidative deamination
N +
pyrrolizidine ring
EtO
NH2
−
+
N
H
+
N
+
O
OCHO
H
HCO2
O
N
O 77 (49%) (X-ray)
O (12%)
Biomimetic access to the pyrrolizidine ring.
1.3.2 Biomimetic Syntheses of Elaeocarpus Alkaloids
Homologous to intermediate 75 derived from homospermidine and implicated in the biosynthesis of indolizidine alkaloids, intermediate 78 (Scheme 1.21) has been postulated to be at the origin of interesting natural substances. Biomimetically speaking, aldehydic intermediate 78 was prepared in the laboratory by the Gribble group in the 1980s (Scheme 1.22) [46]. Bis-acetal 79 was prepared in three steps from chloroacetal 80 in 47% overall yield. It constitutes a stable protected equivalent of intermediate 81, which under acidic conditions can be deprotected to give in situ 78. In fact, non-isolated 81 when buffered at pH 5.5 presumably generates pyrrolinium aldehyde 78. This latter was reduced to pyrrolidine 82 or more interestingly trapped by various nucleophiles in a biomimetic manner, thus giving a very interesting unified access to a group of alkaloids isolated from different species of Elaeocarpus and Peripentadiena (commonly known as Elaeocarpus alkaloids [47], see examples on Scheme 1.21, 83–86). Most of them share a common indolizidine backbone functionalized at positions 7 and 8. Onaka, in the early 1970s, suggested intermediate 78 as the universal precursor of these natural substances [48]. The isolation some years after of alkaloids such as peripentadenine (86) [49] ascertained a spermidine (17) metabolism pathway. Although numerous syntheses of Elaeocarpus alkaloids have been reported [47], we deliberately delineate herein syntheses based on the in situ generation of intermediate 78.
19
20
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples Me
O
H H
N
NH2
O
H elaeocarpine 83 [Elaeocarpus polydactylus]
NH
O
N+ 78
spermidine 17
H2N OH
8 7
N
N O
H H
Me
N
HO
elaeokanine C 84 [Elaeocarpus kaniensis]
Scheme 1.21
N H H
N
H N
Me
NHCOC5H11
O
peripentadenine 86 [Peripentadenia mearsii ]
elaeocarpidine 85 [Elaeocarpus dolichostylis]
Elaeocarpus alkaloids.
Trapping of intermediate 78 with tryptamine gave a straightforward isohypsic synthesis of (±)-elaeocarpidine (85) in a stereoselective cascade of reaction from acetal 79 by just adjusting the pH of the solutions (Scheme 1.22) [46]. The elaeocarpidine aminal function was then reduced with sodium cyanoborohydride to give tarennine (87). In turn, trapping of intermediate 78 with β-keto ester 88, thus engaged in a tandem Mannich/aldol condensation, gave 89 as a mixture of two major diastereomers (axial and equatorial hydroxyl at C7) in 62% overall yield. Compound 89 appeared to be a common intermediate for the synthesis of both (±)-elaeokanine A (90) and C (84). Decarboalkoxylation in strong acidic conditions was accompanied with dehydration at position 7 and gave rise to (±)-elaeokanine A (90) in 91% yield. Milder conditions were necessary to carry out the synthesis of (±)-elaeokanine C (84). The conditions were carefully studied by the authors, who finally chose catalytic transfer hydrogenation with ammonium formate and palladium in methanol to afford the desired reaction. Although some degree of selectivity was expected, the outcome of the reaction clearly showed a preference for the wrong isomer, with a predominance of (±)-7-epi-elaeokanine C (91) over (±)-elaeokanine C (84) despite a good overall yield of 60% from bis-acetal 79. This experiment and its outcome in terms of stereochemistry can be rationalized when considering transition state 92 in which an equatorial hydroxyl is to be expected. This supposition was compared to a similar case studied by Tufariello and Ali [50] a few years before with the intramolecular kinetic aldol reaction of a ketone instead of a β-ketoester. In the latter, a stereocontrol, totally in favor of the axial configuration, could have been governed by transition state 93. Returning to a biosynthesis hypothesis and taking into account these findings, one can suggest that should such a pathway occur in Nature the decarboxylation step has to be prior to the aldol reaction. Starting from readily accessible acetal 79, it is worth noting how simple starting materials (tryptamine, ketoesters) and reaction conditions (aqueous solutions at
1.3 Biomimetic Synthesis of Alkaloids Derived from Ornithine and Arginine
80
3 steps (47%) MeO
OMe
N H
N 82
(36%)
79
OMe
tarennine 87
N H H
O
78
+
N H
N
O
N+
NaBH3CN
NH
(63%)
N
N
N H H
1.H2O,HCl M 2.pH 5.5
biosynthetic intermediate 78 (see Scheme 1.21)
NH2
pH 1.5
N
OMe
OH
NaBH3CN
O +
81
OMe
OMe
Cl
O
N H
O
21
H N H H
N
NH N +
elaeocarpidine 85
pH 5.5 O
O 88
O R'
O H
OR H
O +
N
8M HCl,
N
elaeokanine A 90
(91%)
HO 89 R'=CO2CH2Ph
N
R=CH2Ph O
H
7-epi-elaeokanine C 91 7
H
HO
N
HCO2NH4, Pd/C, MeOH, RT= CO2CH2Ph then H3O + = −CO2 O H N elaeokanine C 84 + H HO
CO2H
(3:1 ratio, 60% from bisacetal 79) Gribble and coll. [46] RO N 92
Scheme 1.22
O
H O O
N
Tufariello and coll. [50]
N
N
OH
O 91
O 93
O
O
OH
H
Elaeocarpus alkaloids: biomimetic synthesis.
various pH) enabled the design of a divergent pathway to Elaeocarpus alkaloids following, and thereby reinforcing, previously proposed biosynthetic hypotheses. Concomitant with Gribble’s work, the Hortmann group used α-cyanopyrrolidine 94 as a surrogate of 79 (Scheme 1.23) [51]. It was prepared by oxidative cyanation of the corresponding pyrrolidine 95 with chlorine dioxide (as an alternative to classical mercury acetate oxidation or modified Polonovski reaction). Compound 94 was used in a total synthesis of (±)-elaeocarpidine (85). Lactam 96 was prepared by L´evy and colleagues for the synthesis of 85 [52]. It was engaged in a reductive
84
22 EtO
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples EtO
OEt
OEt
N
CN
AgOTs, then tryptamine (38%)
ClO2, NaCN (~90%)
N H H
N CN 95
94
Scheme 1.23
N
1-tryptamine, H2 Pd/C (58%) 2-LiAlH4, THF (64%)
H N
N
reductive process
O
(±)-elaeocarpidine 85
96
Elaeocarpus alkaloids: alternative biomimetic synthesis.
Me grandisine A 97 [Elaeocarpus O grandis] Me
Figure 1.3
O
H H
N
grandisine B 98 O [Elaeocarpus Me grandis]
OH
N H N
grandisine D 99 [Elaeocarpus grandis]
Me O H H N O
Other Elaeocarpus alkaloids.
Pictet–Spengler reaction under catalytic hydrogenation conditions. The recourse to trivalent functional groups (lactam of 96) where divalent ones (imine) are needed places this former approach at the borderline of biomimetic synthesis. The Elaeocarpus alkaloid group gained constant interest with the discovery of new structures (see examples 97–99 in Figure 1.3) along with interesting biological properties (such as selective γ -opioid receptor affinity for grandisine-type alkaloids) [53]. These complex structures also stimulated state of the art total syntheses [54]. 1.3.3 Biomimetic Synthesis of Fissoldhimine
Fissoldhimine (100) was isolated in 1994 by Sankawa et al. [55] from fresh stems of Fissistigma oldhamii (Annonaceae), a shrub mainly found in Southern China and Taiwan (Scheme 1.24). Its structure was unambiguously confirmed by X-ray analysis. Since n-butanol was used to extract this basic molecule from an alkaline solution, it was suggested that fissoldhimine was an artifact resulting from the aminoacetalization of compound 101 (which may therefore be the ‘‘true’’ natural product) with a molecule of n-butanal presumably present in n-butanol.17) The authors proposed a biosynthetic pathway (Scheme 1.25) to fissoldhimine (100/101) from two molecules of cyclic enamine 29 that came from l-ornithine via a dimerization (see above). R. A. Batey and colleagues revisited the hypotheses in 2007 when disclosing the first biomimetic investigations toward fissoldhimine [56]. 17) Defined as spontaneously formed natu-
ral products, artifacts are closely related
to biomimetic chemistry (see Chapter 24 of this book).
1.3 Biomimetic Synthesis of Alkaloids Derived from Ornithine and Arginine traces in extraction solvent? (n-butanol)
O biosynthesis
N
O HN H H 2N N
N
Scheme 1.24
O O
O HN
artifact?
H H true natural product? 101
29
H N
N 1' 4' 3'
N 2'
H
4 3 1 2
H
fissoldhimine [Fissistigma oldhamii] 100
Fissoldhimine: possible structures. O HN
N H
N
29
N
H
H2PO4CONH2 H
H2N
N
35 N
early carbamoyl transfer
H
H
biogenetic hypotheses by Sankawa and coll. (1994) [55] O O H
HN
O N
N
H2N
H
H2PO4CONH2
H
HN H HN
N
H
H
101
aza-Mannich
N 102 O
NH2 H + dimerization
O
N
H2N N
H2N
+
H
+N
O
H
NH2
+
NH2
O
two alternative biogenetic hypotheses by Batey and coll. (2007) [56]
rearrangement
O
NH O H N H2N
O
N
H
H
N
O N H2N
Scheme 1.25
23
cycloaddition [4+2]
Biosynthetic hypotheses.
Benzyl- or para-methoxybenzyl-protected urea 103 (Scheme 1.26) was used as biomimetic equivalent of biosynthetic intermediate 102 postulated in Scheme 1.25. The best conditions, the authors found, for the formation of the desired exo-dimer 104b were the use of trifluoroacetic anhydride in THF (endo/exo: 2 : 1 ratio) but separation of the two diastereomers was impossible. In addition, final deprotection of the urea nitrogens was also problematic (removal of the benzyl group was unsuccessful and DDQ (2,3-dichloro-5,6-dicyano-1,4-benzoquinone)-mediated deprotection of PMB groups resulted in the deprotection of only one of the nitrogens). Despite obtaining an endo relative stereochemistry that contrasts with the exo stereochemistry of natural fissoldhimine 100/101, this example is interesting as it opens the way to discussions concerning the biosynthesis and the implication of enzymes, particularly since fissoldhimine was shown, by X-ray, to be a racemate in nature! Another point is that an early carbamoyl transfer is likely, in direct
NH2
Scheme 1.26
R N H
N
R N H N
endo (major)
O
O
R
H 104a
H
N H
N
biomimetic equivalents
HO R=Bn or PMB O
103
O
O
10% Dy(OTf)3, CH3CN (R=Bn) 10% Dy(OTf)3, CH3CN (R=PMB) BF3 Et2O, THF (R=PMB) TFAA (1.1 eq.), THF (R=PMB)
R N H N H N R N H H exo 104b (minor)
+
O
selected conditions (RT)
2-DIBAL-H, CH2Cl2, –78°C (R=Bn: 74%, R=PMB: 83%, 2 steps)
1-RNCO, tol, reflux
Biogenetically inspired heterodimerization toward fissoldhimine.
Bn=benzyl PMB=para-methoxybenzyl
biogenetic intermediate
O
N
102
>95/5 (73%) >95/5 (84%) >95/5 65/35
endo/exo
H
N
HN H N H N H2N H natural skeleton 100/101 O
O
24
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
1.3 Biomimetic Synthesis of Alkaloids Derived from Ornithine and Arginine
25
correlation with the critical difficulty of dimerization of imine 29 as already evoked (see above, Scheme 1.5). 1.3.4 Biomimetic Synthesis of Ficuseptine, Juliprosine, and Juliprosopine
As we will now discuss, imine 29 was successfully engaged by the Snider group [57] in a biomimetic synthesis of two dihydroindolizinium alkaloids: ficuseptine (105) [58] and juliprosine (106) [59] (Scheme 1.27). Biosynthetic considerations led to the postulation of a Chichibabin pyridine synthesis type reaction as a cornerstone convergent step for both 105 and 106. Mixing aldehyde 107 (which is presumably a tyrosine metabolite in Nature) with imine 29, generated in situ from 108, in acetic acid provided ficuseptine (105) in a single step in 52% yield (Scheme 1.28). OMe R
R ficuseptine 105 [Ficus septica]
N Cl
N OH MeO
Scheme 1.27
O
OH NH Me
NH Me
juliprosine 106 [Ficus septica]
R
N
R
N 29
OMe
+
MeO 107
OMe MeO H2N
AcOH 95 °C, via imine 29
ficuseptine 105
(52%)
N
108
Cl MeO
Scheme 1.28
1
O biosynthetic Chichibabin synthesis of pyridines
Ficuseptine (105) and juliprosine structures (106), and plausible biogenesis.
O
L-ornithine
Ficuseptine: biomimetic synthesis.
This strategy appeared to be as effective (Scheme 1.29) when dealing with more complex juliprosine (106) in terms of side chains engaged in the pyridine formation (probably of polyketide origin in Nature). Aldehyde 109 was prepared in six steps from bromopyridinol 110 and reacted with preformed imine 29 at room temperature in acetic acid to give Troc-106, from which juliprosine 106 was liberated by protecting group removal. Juliprosopine (111) [60], a reduced tetrahydropyridine counterpart of juliprosine was then targeted by the authors as a final destination. Exposure of Troc-106 to sodium borohydride gave a separable 1 : 1 mixture of diastereomers. Prior to this last step, model studies concluded that the natural stereochemistry of 111 should be trans. Therefore, the trans isomer was deprotected to afford juliprosopine (111), for which the stereochemistry was in consequence clarified 25 years after its isolation. It is striking in terms of
26
Br
N
110
Troc-106
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
Me
6 steps (32%)
9
OH
NTroc+ Me 109
OH
1-NaBH4, EtOH, 2-KOH, H2O, i-PrOH,
H
N
(36%)
NH Me
N 9
OH H
+
Troc=2,2,2-trichloroethoxycarbonyl
5
OH
(35%)
29
H N
Scheme 1.29
1-AcOH, 25°C, 2-Zn, HCl, MeOH, 65°C
O
NH juliprosine 106 Me
H N
juliprosopine 111
(36%)
Juliprosine (106) and juliprosopine (111) biomimetic synthesis.
evolutionary convergence that a pivotal biosynthetic (and synthetic!) reaction can give rise to different natural substances in different plants. These examples once again raise questions about the intervention of enzymes in such biosyntheses. 1.3.5 Biomimetic Synthesis of Arginine-Containing Alkaloids: Anchinopeptolides and Eusynstyelamide A 1.3.5.1 Natural Products Overview Original dihydroxybutyrolactams were isolated from marine ascidian Eusynstyela latericius collected in the Australian Great Reef and named eusynstyelamides A–C (112–114, Figure 1.4) by the Tapiolas group in 2009 [61]. They are in fact closely related to other natural substances, namely, anchinopeptolides A–D (115–118) and surprisingly cycloanchinopeptolide C (119) (with its central core consisting in a tricyclic 5-12-4 ring system) isolated earlier from a marine sponge [62]. Biosynthetically, they might all arise from dimerization of monomeric modified α-ketoacid-containing peptides (dipeptides or tripeptides in the case of eusynstyelamides and anchinopeptolides, respectively) by an aldol reaction (formation of the carbon–carbon bond) and an addition to the amide (Figure 1.4, box). Biomimetic syntheses of these two classes of alkaloids have been performed by the Snider group. 1.3.5.2 Biomimetic Synthesis (±)-Anchinopeptolide D (118) The aldol dimerization of protected modified tripeptide 120 (synthesis not described herein) was undertaken (Scheme 1.30) [63]. Treatment of 120 with potassium hydroxide in MeOH afforded a mixture of three isomeric compounds (Boc-118, Boc-121, 122), two of which were deprotected to afford 118 (anchinopeptolide D) and 121 (epi-anchinopeptolide D). Equilibration studies clearly demonstrated epimerization at the hemiaminal center: heating pure Boc-118 or Boc-121 for 1 h resulted in a 2 : 1 mixture of Boc-118 and Boc-121. Equilibration of the aldol stereocenter occurred more slowly (from pure Boc-118
NH
OH
H
HO
O N OH NH
NH2
NH
NH2 NH
HN
O
OH NH
H N
[Eusynstyela latericius]
N
N
NH O
O
Br
H
OH
O
H N
N H HO
N
O
O OH NH
N
O
O
N OH NH
OH NH
O not found
HO H
O
H 2N
H2N
(–)-eusynstyelamide A 112
Eusynstyelamides and anchinopeptolides.
O
O
O
HO H
O O eusynstyelamide B 113 (+)-eusynstyelamide C 114
H
HO
N H
Figure 1.4
Br
H N
HO O
O
O
N
H HO N H R1 O
R2
HO
N
O
O
N H
H N
H
H N H H O N H H
Me
relative stereochemistry
O
H HO
[Anchinoe tenacior]
(+)-cycloanchinopeptolide C 119
NH
NH
NH
NH
NH
NH
H N
OH
OH
OH (–)-anchinopeptolide A 115: R1=R2=Me (–)-anchinopeptolide B 116: R1=Me, R2=H (–)-anchinopeptolide C 117: R1=H, R2=Me (+)-anchinopeptolide D 118: R1=R2=H
H 2N
H 2N
OH
1.3 Biomimetic Synthesis of Alkaloids Derived from Ornithine and Arginine 27
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
28
Boc H
N
O
N N H Boc
O
O H HO HO
O 120
O H HO N H
N
H N
N H
HO
N
N H
O
O
121 (17%)
anchinopeptolide D 118 (52%)
H2N
O H HO
H N NH
H2N
1-KOH, MeOH, THF, 0 °C 2-TFA, CH2Cl2, 25 °C
OAc
HO
NH
N O
H HO
H N
N H
HO
N
N H
O O
O
OH
122 (Boc-protected) step 2 was not conducted (<5%)
NH OH
Scheme 1.30
O
Biomimetic synthesis of anchinopeptolides.
or Boc-121, 17 h of heating in KOH THF–MeOH provided a 6 : 3 : 1 mixture of Boc-118, Boc-121, and 122). ‘‘(±)-Cycloanchinopeptolide D’’ (123) A head to head [2 + 2] cycloaddition between the two double bonds of the plausibly tyrosine-derived part of anchinopeptolide C (117) may explain the formation of the cyclobutane ring of cycloanchinopeptolide C (119). To confirm this hypothesis, access to analog 123 (unnatural ‘‘cycloanchinopeptolide D’’) was targeted (Scheme 1.31). Careful choice of the photochemical conditions, especially the solvent, permitted the obtaining of 123 avoiding, thereby, the isomerization of the double bonds. Indeed, in water, irradiation of a 0.005 M solution of 118 at 350 nm afforded cycloanchinopeptolide D (123) in 48% yield. In water, a hydrophobic effect may cause the nonpolar chain to pack closely and favor the [2 + 2] cycloaddition versus the trans/cis isomerization. H N
H2 N
NH H2N
H HO HO
NH NH
N O
O N H
H N
hυ, 350 nm D2O, 28°C, 5h (48%)
O
NH
O HN
OH
H2N
H HO HO
NH
N O
O N H O
H NH H O N H H
OH H
NH
118 OH
Scheme 1.31
H N
H2N
"cycloanchinopeptolide D" 123
OH
Biomimetic synthesis of cycloanchinopeptolide D.
(±)-Eusynstyelamide A (112) Adopting a similar philosophy, access to 112 was possible (Scheme 1.32) [64] from unstable dipeptide 124 (prepared in four steps), which could undergo butyrolactam core formation under basic conditions. The obtaining of Boc-protected-112 as a major compound (37%) was accompanied by small amounts of Boc-113 (<4%). Deprotection of Boc-112 afforded eusynstyelamide A (112). At this stage, chemical and reactivity observations stimulated interesting thoughts. As previously mentioned, anchinopeptolide D (118) was shown to equilibrate into a mixture of epimeric compounds. The question of
1.3 Biomimetic Synthesis of Alkaloids Derived from Ornithine and Arginine
HO
O N
H O
H N
Br N H
O
N H
NHBoc
OH (±)-eusynstyelamide A NH 112 (95%)
unstable (readily
NaOH, 0.25M THF, 25°C, 2 d
N
H
OH Boc-(±)-eusynstyelamide B NH Boc-113 (<4%) O
Scheme 1.32
Biomimetic synthesis of eusynstyelamide A.
whether eusynstyelamides B (113) and C (114) are natural products or artifacts spontaneously generated during extraction was logically posed. In fact, when the isolation of eusynstyelamides was conducted, some samples, carefully processed, of the sponge contained only eusynstyelamide A (112) whereas some others contained the two isomeric eusynstyelamides A (112) and B (113) (1 : 1 ratio) or the three isomeric compounds [eusynstyelamides A–C (112–114), 2 : 6 : 1 ratio]. Additionally, no evidence of equilibration was observed by the authors as confirmed by Snider and colleagues with synthetic eusynstyelamide A (112). Indeed, synthetic 112 did not equilibrate under various conditions (acidic, storage in CD3 OD) and decomposed in basic conditions. The formation of Boc-eusynstyelamide A (Boc-112) as the major isomer from monomeric 124 is consistent with the isolation of eusynstyelamide A (112) as a unique compound from some sample of raw material. In the laboratory, the aldol reaction gave rise to a major isomer with high selectivity according to the proposed mechanism (Scheme 1.33) and may be correlated with the fact that it is the main relative stereochemistry observed in the two series of alkaloids. The influence of the side chains when comparing the eusynstyelamide and anchinopeptolide series is probably prominent in controlling both the equilibration propensity and the stereochemical outcome of the aminal formation. R1 R Z-enolate 1 O O H O M R2HN R2HN O
Scheme 1.33
HO O
R1 R1 hemiaminal
aldol
H R2HN R2HN
O formation O O M O
CF3CO2H/ CH2Cl2
no equilibration observed
+ O O HO
NBoc
124 oxidizes with air)
Boc-(±)-eusynstyelamide A Boc-112 (35%)
29
R1 H R1
NHR2 OH CONHR2
major relative stereochemistry in Nature and laboratory
Dimerization: a plausible mechanism.
1.3.6 A Century of Tropinone Chemistry
The biomimetic synthesis of tropinone (125) by Robinson in 1917 [65] is probably among the most renowned synthesis of organic chemistry. More than many other,
30
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
it has been analyzed, cited and is usually presented as the first biomimetic total synthesis and as a pioneering multicomponent reaction (Scheme 1.34). We will therefore not develop further this textbook classic. The reaction was revisited by Sch¨opf in the 1930s with an in-depth study of the reaction conditions [66] and has been adapted to other target molecules many times since then. Questions interfacing biosynthesis and chemistry arose continuously concerning tropane alkaloids; they were compiled and analyzed by O’Hagan and Humphrey in a review article [67]. The now well-established biosynthesis of the tropinone (125) skeleton is depicted in Scheme 1.34. Biosynthesis of tropinone:
13
Me N
oxidation 2 × AcSCoA
H2O
Me N
Me
SCoA O
N CO2 O
O
− CO2
Biomimetic synthesis (Robinson, 1917): CHO + NH2Me + O
CO2
CHO
CO2
Ca2
Me
N
CO2
HCl,
Me
CO2 O − 2 CO2
Scheme 1.34
N
O tropinone 125
Tropinone biosynthesis and landmark biomimetic synthesis.
Nearly one century later, this reaction still stimulates interesting discoveries. By way of an example, a solid-phase version was developed as an interesting means of generating tropane analogs in a combinatorial way [68]. The use of siloxy(allyl)silane was also reported in place of acetonedicarboxylates (avoiding thereby the double thermal decarboxylation step) and afforded tropinone in good yield [69]. Asymmetric syntheses of substituted tropinones, some using biomimetically related intramolecular Mannich reactions have also been disclosed [70]. The Mannich reaction and the Robinson–Sch¨opf condensation are absolute musts in the chemical toolbox for biomimetic synthesis of alkaloids [71].18) 1.4 Biomimetic Synthesis of Alkaloids Derived from Lysine 1.4.1 Alkaloids Derived from Lysine: To What Extent?
A list (if not a classification sensu stricto) according to the number of carbons that have been incorporated via the C5 N units originating from l-Lys can be drawn 18) ‘‘Time can never diminish the exquisite
beauty of Robinson’s total synthesis of tropinone. The simplicity of this remarkable synthesis, the splendor of the cascade and the stunning analysis that led to its
conception will always stand out, marking this endeavor as a true classic in total synthesis, and inspiring new generations of chemists’’ [72].
1.4 Biomimetic Synthesis of Alkaloids Derived from Lysine 1 Alkaloids totally derived from L-Lys: H
N
N
sparteine 126 [Cytisus scoparius] (3 ×Lys)
H H
N H
H N
H
H
N
HO2C
H
OH
N H
H N
H myrobotinol 127
N
HO2C
H N N
N H [Punica granatum] (1× Lys)
3 Alkaloids containing L-Lys:
[Oryza sativa] (4×Lys)
O
O
O pelletierine 129
HN oryzamutaic acid A 128 CO2H
O
[Myrioneuron nutans] (4× Lys)
N H
NH2
HO2C
2 Alkaloids mainly derived from L-Lys:
N H
N
H H O
NH2
HO2C
N
HN
N H
[Withania somnifera] (2 ×Lys)
O
CO2H
O
anaferine 130 N H O
N H N
piperine 131 [Piper nigrum]
O
Figure 1.5
Lysine-derived alkaloids: selected examples.
up – some examples are given in Figure 1.5. One may then distinguish, more or less artificially, between three categories, that is, alkaloids: (i) exclusively derived from l-Lys19) [e.g., sparteine (126), myrobotinol (127),20) oryzamutaic acid A (128)21) , (ii) mainly derived from l-Lys [e.g., pelletierine (129) and anaferine (130)], and (iii) containing l-Lys [e.g., piperine (131)]. For each of the two first groups, examples of biomimetic syntheses will be discussed in this section. 1.4.2 Lupine Alkaloids 1.4.2.1 Overview and Biosynthesis Key Steps The ‘‘lupine alkaloids’’ are an important and biosynthetically homogeneous group of natural substances found especially, but not only, in the Lupinus genus (Fabaceae). The vast majority of them are structurally characterized by a quinolizidine rings and they are classified in ten or so main subgroups defined by the central skeleton. Lupinine (132) (the simplest alkaloid of this group), sparteine (126), 19) Every single carbon atom of the alkaloids
comes from l-Lys. It is, at this point, striking to see how complex structures can arise from a single amino acid. 20) See below for comments on this intriguing new family of alkaloids. 21) Oryzamutaic acid A (128) is one of the rare alkaloids that contain four molecules
of l-lysine [22]. It is a novel yellow pigment isolated from the endosperm of an Oryza sativa (rice) mutant. Its rather elaborate architecture presumably incorporates several types of units derived from l-lysine, including lysine itself. It undoubtedly constitutes an appealing target for biomimetic synthesis! [73].
31
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
32
hydrolytic cleavage H2 N H N
N
O H H N
H 137 tetrahydroanabasine 16
oxidative deamination
O
O H H N
H
136
lupine alkaloids
H
OH H
NH
N
N
O
lupinine 132 oxosparteine 133 [Lupinus luteus] [Cytisus scoparius]
Scheme 1.35
O
N
N
NH N N
cytisine 134 [Laburnum sp.]
varenicline 135 [synthetic drug]
Lupine alkaloids: origin and examples.
oxosparteine (133), and cytisine (134) are probably among the most well-known structures of lupine alkaloids (Scheme 1.35). The existence of multiple higher oxidized derivatives of lupine alkaloids and transformed skeletons thereof explains part of the diversity within this class of natural substances. They are of great interest for their biological properties such as nicotine-like properties. Varenicline (135) (Chantix, Champix), a synthetic drug that acts as a partial agonist of α4 β2 -subtype of the nicotinic acetylcholine receptors, is directly inspired by the structure of natural cytisine. This drug is prescribed to treat smoking addiction in many countries in the USA and Europe. Many comprehensive reviews of interest have been published over the years covering many aspects of lupine alkaloids [74]. The suggestion that the quinolizidine is built up from l-lysine has now been widely confirmed by incorporation studies and chiral non-racemic intermediate 136 arising from selective hydrolysis of 16 (via intermediate 137) appears to be a cornerstone in lupine alkaloid metabolism. The biosyntheses of alkaloids such as lupinine (132) or sparteine (126) are now textbook examples of l-lysine metabolism in plants. 1.4.2.2 Biomimetic Synthesis of Lupine Alkaloids A unified biomimetic strategy starting from tetrahydroanabasine (16) was beautifully designed by Koomen and colleagues for the total synthesis of representative lupine alkaloids [75, 76]. Four natural substances, that is, lupinine (132), epilupinine (138), aminolupinane (139), and sparteine (126), as well as anabasine (140), were retrosynthetically traced back to the common precursor tetrahydroanabasine. Mimic of the Fundamental Oxidative Deamination Step Topaquinone (2,4, 5-trihydroxyphenylalanine) was identified as the covalently bound active site cofactor of copper-containing amine oxidases in the early 1990s [77]. These enzymes catalyze the fundamental two-electron oxidative deamination of primary amines into the corresponding aldehyde along with ammonia. The formation of a Schiff base that undergoes proton abstraction with aromatization as a driving force is implicated in the intimate mechanism of the enzyme. Several model studies of topaquinone or more generally quinoprotein enzymes have been
1.4 Biomimetic Synthesis of Alkaloids Derived from Lysine
H2N
COOH t-Bu
R' O
t-Bu
H 2N
N
R
O 141
O
t-Bu equivalent of topaquinone mimic of the active site of topaquinonecontaining amine oxidase O HO topaquinone
Scheme 1.36
O
MeOH t-Bu
aromatization = driving force
R' R
N
t-Bu
R OH
H
R' O R
+
t-Bu t-Bu
N R
if R'=H
made but are beyond the scope of this chapter [77, 78]. Commercially available and stable di-tert-butyl-o-quinone 141 was successfully used as a topaquinone surrogate to achieve the desired oxidative deamination without any overoxidation to benzoxazoles of type 142 (Scheme 1.36). Quinone 141 was used to oxidize compound 143, which resulted from the opening of tetrahydroanabasine (16) with methoxyamine (compound in which the stereochemistry was preserved), to give iminium or enamine 144 depending on the pH (Scheme 1.37). Reduction of 144 gave a stable quinolizidine (Scheme 1.37), which was further reduced to aminolupinane (139, isolated as its acetate). Hydrolysis of the oxime of 145, necessary to access lupinine, had to be performed under mild conditions to avoid epimerization of the intermediate aldehyde into the more stable epilupinine skeleton. Ozonolysis of 145 at low temperature followed by sodium borohydride reduction yielded exclusively lupinine (132). Under more vigorous conditions, a mixture of lupinine (132) and epilupinine (138) was obtained. The pathway to sparteine required participation of another tetrahydropyridine molecule (15). Reaction of 144 with 15 gave 146, the oxime group of which was split under mild conditions to furnish aldehyde 147, which is a direct precursor of the sparteine skeleton in which both the piperidine ring and the aldehyde occupy an axial position that can cyclize to 148. Reduction of 148 furnished sparteine 126 but, as already observed with lupinine, oxime cleavage under more drastic conditions was accompanied with a loss of stereoselectivity, giving isomeric β-isosparteine along with sparteine. This synthesis clearly demonstrated that the whole pathway from tetrahydroanabasine (16) to polycyclic sparteine (126) can be efficiently mimicked in the laboratory. Finally, simple hydrolytic conditions from 144 surprisingly gave rise to anabasine (140) in a single step.22) 1.4.2.3 A Biomimetic Conversion of N-Methylcytisine into Kuraramine Many examples of semi-synthetic conversions of lupine alkaloids have been disclosed over the years [74]; only one recent example will be given. Kuraramine (149) was isolated in 1981 from Sophora flavescens [79]. An obvious biosynthetic speaking, anabasine (140), for example, found in Nicotiana sp., is derived from tetrahydropyridine and nicotinic acid and not from
R'
H2O
Topaquinone and a topaquinone-mimic.
22) Biosynthetically
33
tetrahydroanabasine, nor does tetrahydroanabasine (16) derive from the reduction of anabasine (140).
O t-Bu
142
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
34
O
t-Bu
H NH
N
NH2OCH3 MeOH, rt (98%)
N
NH
H
N H
N
H
MeOH (45%)
N
OCH3
N
OCH3
OCH3
H
−H 144
H3O
,
N
H
(56%)
(±)-anabasine 140
AcHN H
1-NaBH4 (89%) 2-LAH 3-Ac2O (85%)
N
NH
O
NH
H
141 t-Bu
143 NH2
(±)-tetrahydroanabasine 16
N
N
O
OCH3
H
(±)-N -acetyllupinamine 139 N
OCH3 NaBH4 (89%)
144
N H N
OH
OCH3 1- 1 or 2
15
1 : TiCl3, H2O,
H +
N
145
N
OH
H
(see box) 2-NaBH4
N
(±)-lupinine 132 (±)-epilupinine 138 , 132/138: 5:2 (30%)
2 : O3 ,TFA, H2O, −50°C 132/138 1:0 (54%)
MeOH, H (80%) N
O
H
OCH3
N
H N
1 or 2 (see box)
H
H H N
N
H
H
147
H
148
NaBH3CN
1 : TiCl3, HCl, sparteine/b-isosparteine 1:1 (20%) 2 : O3, HCl, H2O, −50 °C, sparteine (21%)
H
H N
N N
N (±)-sparteine 126
Scheme 1.37
N
N
H
146
H
H
H
Biomimetic unified access to lupine alkaloids.
relationship with N-methylcytisine (150) which could involve a N(1)–C(10) oxidative cleavage has been postulated and strengthened by a biomimetic conversion (Scheme 1.38). Indeed, in 2010, Gallagher et al. [80] first silylated 150 into 151, which in turn was submitted to a Fleming–Tamao oxidation (153) followed by sodium borohydride reduction to give (−)-149 in 18% overall yield. 1.4.3 Biomimetic Synthesis of Nitraria and Myrioneuron Alkaloids
Mainly from the steppes of Uzbekistan, intriguing alkaloids [such as sibirine (154), nitraramine (155), and nitrarine (156)] have been isolated from Nitraria species
1.4 Biomimetic Synthesis of Alkaloids Derived from Lysine
Me N 10
1
biosynthetic link?
O
N
Me N OH 10 H N
Me N O =
1
(−)-N-methylcytisine 150
1-PhMe2SiCl, THF 2-LDA(48%, 2 steps)
Me N
35
HO
H N
O
1
10
(−)-kuraramine 149 Me N
biomimetic conversion O Hg(OAc)2 (152), AcO2H, AcOH
N
NaBH4, MeOH (70%)
O N
(26%, 3 steps)
PhMe2Si
Scheme 1.38
HO
151
153
Biomimetic synthesis of kuraramine from N-methylcytisine.
(Nitrariaceae). In the 1990s, an unusual lysine-derived metabolism was postulated by Koomen and colleagues to account for the biosynthesis of these often-complex molecules (Scheme 1.39) [81]. Often isolated as racemic mixtures, they may arise from an alternative opening/rearrangement of tetrahydroanabasine as compared to the ‘‘classical’’ Lupinus pathway (Scheme 1.39, pathway 1 and Scheme 1.35 above). Central to this hypothesis is the formation of compound 157, referred to as the ‘‘key intermediate’’ in Nitraria metabolism, by a retro-Michael reaction followed by an oxidative deamination (pathway ). 2 O N
"Nitraria alkaloids"
OH sibirine 154 [N. sibirica]
Me
N
N NH
NH
nitrarine 156 (±)-nitraramine 155 [N. schoberi ] [Nitraria schoberi ]
N H
1
O
O HH N 136 H
chiral precursor
1
H N
N
2
H N
1- H 2O H 2- oxidative deamination tetrahydroanabasine
2
O
achiral precursor
157
1- retro-Michael 2- oxidative deamination
16 HO O
"Lupinus alkaloids"
"Myrioneuron alkaloids"
HO N
O
N N
myrioneurinol 158 myrobotinol 127 [Myrioneuron nutans] [Myrioneuron nutans]
Scheme 1.39
Nitraria and Myrioneuron alkaloids: origin and selected examples.
1.4.3.1 Biomimetic Syntheses of Nitraramine Nitraramine (155) is probably one of the most original and intricate structures among lysine-derived alkaloids (Scheme 1.40). It has been isolated from Nitraria
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
36
H 15
N
N
H N
H N 15
a
HO
b
N
N
N
− H2O
b O
159
161
157
a N
N
H N
spirocyclization
spirocyclization
N O
OH 160
N
OH NH
158 ring inversion
biomimetic synthesis [E. Gravel et al.]
N
H N
H
OH
N N
15
N H H N
Si face
nitraramine
15 O
155
O EtOH, reflux, 3 h (2-3%)
51
O
O
N N H
Boc O N O
biomimetic synthesis N [G.-J. Koomen et al.]
O
H2O, pH 7 (0.5% global yield from 162)
O 55
Scheme 1.40
N
achiral precursor 161
Biosynthetic hypotheses of nitraramine and total biomimetic synthesis.
as a racemate. With several heterocycles (three peripheral cycles in a chair-like conformation and three central cycles in a boat-like conformation) and six contiguous stereogenic centers (one of which is a quaternary spiro center) this molecule, despite its rather modest molecular weight, can be seen as a real challenge for organic chemists. The Koomen group proposed a biogenetic hypothesis according to which nitraramine might result from the assembly of lysine-derived simple precursors (Scheme 1.40), through a series of simple reactions. From key intermediate 157 as a pivotal achiral precursor, the addition of a piperideine molecule (15) could
1.4 Biomimetic Synthesis of Alkaloids Derived from Lysine
37
then afford compound 158 (or 159). An intramolecular spirocyclization reaction could then give rise to the spiro quaternary center (intermediate 160) found in the natural product. A ring inversion on the newly created cyclohexane of 160 is then needed to explain two ring-closures by imine trappings that end the cascade toward nitraramine 155. The Koomen group targeted compound 161 in their beautiful pioneering total synthesis of nitraramine in 1995 (Scheme 1.40). Achiral synthetic 161 was then treated in neutral conditions in water and afforded the natural compound in a global sequence of a dozen steps from 55 and a piperidone (0.5% yield) [82]. In 2005, a second straightforward synthesis strongly reinforced the Koomen hypotheses [83]. In view of the fact that 157 can be obtained from the reaction of enamine 15 with glutaraldehyde (51) followed by dehydration, we set up a biomimetic total synthesis of nitraramine by a one-pot sequence. In fact, from simple starting materials 15 and 51 the reaction cascade quickly takes place. By treating one equivalent of 15 with two equivalents of 51 in boiling ethanol, nitraramine (155) was obtained in a low yield that competed with the Koomen total synthesis and also with quantities available from natural sources. Interestingly, this second total synthesis of nitraramine showed that the natural substance previously described as epinitraramine (162) was actually an artifact resulting from the protonation of nitraramine in the NMR tube (Scheme 1.41). In fact, 1 H NMR spectra of protonated nitraramine (with traces of hydrochloric acid liberated from CDCl3 ) differ slightly from those of 155 as a free base. Thereby, the absence of 162 reflects the high stereoselectivity of the cascade reaction. Finally, the success of the total synthesis of 155, despite low yields, raises interesting questions concerning the implication of enzymes in the biosynthesis of such alkaloids.
O
1
N N
N H 1 H N
O
(b) attack on Re face
H epinitraramine 162 = artifact
Scheme 1.41
N
OH
(b)
(a)
N
(a) attack on Si face
N H 1 H N
O
O
N 1
N H nitraramine 155
Epinitraramine is an artifact.
1.4.3.2 Biomimetic Syntheses of Tangutorine As seen above (Scheme 1.12), the self-condensation of glutaraldehyde (51) furnishes bicyclic compound 52. A simple condensation of 52 with tryptamine under acidic conditions allowed the synthesis of 163, which is a direct precursor of tangutorine (164), another alkaloid isolated from a Nitraria species (Scheme 1.42) [84]. Indeed, reduction of 163 with sodium borohydride gave tangutorine (164) after recrystallization to isolate the major diastereomer of the (75 : 25) mixture [85]. The reaction course with tryptamine could be rationalized by assuming the in situ formation of intermediate 165. Interestingly, compound 166, resulting from a direct Pictet–Spengler reaction of the aldehyde function of 52 with tryptamine, was also isolated from the reaction. It is reminiscent of the structure of many alkaloids from
38
O
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
HO
O
O
O HO
O
O
O
51
+ tryptamine − H 2O AcOH / CH2Cl2
H NH 166 (5%)
H
N H
H O N
O
165 O biomimetic equivalent
OAc
2- recrystallization: CH2Cl2 / MeOH [95:5] (~50%)
2 diastereomers
NH
Scheme 1.42
N N H
N H N
komarovinine 167
OH
H N H H (±)-tangutorine 164
N
N H
[Nitraria komarovii]
H N
1- NaBH4 / CeCl3, MeOH (>95%)
163 (25-30%)
tetrahydro-komarovinine 168
O
RT, 48 h (25-30%)
+
N H
O
52
N
[Nitraria komarovii]
komaroine 169 [Nitraria komarovii ]
Biomimetic synthesis of tangutorine.
the Nitraria genus in terms of carbon skeleton [see examples such as komarovinine (167), tetrahydro-komarovinine (168), and komaroine (169)], demonstrating thereby the impressive uniqueness of lysine metabolism in the Nitraria genus. It was therefore possible to postulate intermediate 165 as another achiral cornerstone metabolic intermediate along with already mentioned 157. Both intermediates may, at least in part, contribute to explain the occurrence of many Nitraria alkaloids as racemates in Nature (this is especially the case for indolic Nitraria alkaloids). This straightforward strategy also cast doubt on the previously proposed biosynthetic pathway to tangutorine (164), involving a complex rearrangement of a yohimbine-type indolomonoterpenic precursor, proposed by Jokela and colleagues (Scheme 1.43) [86]. These hypotheses appeared unlikely both in terms of chemical reactivity and chemotaxonomy. In 2010, one of us published a comprehensive review covering the state of knowledge and detailing the isolation, structure determination (and revision), and the biomimetic syntheses of Nitraria alkaloids [87]; the interested reader is directed to this article as well as to older reviews [86]. Many achievements in biomimetic synthesis of Nitraria alkaloids have been disclosed by the Koomen group [88] and the Husson group [89]. Some of them beautifully addressed the issues of efficiency and selectivity in total synthesis.
1.4 Biomimetic Synthesis of Alkaloids Derived from Lysine Initial biosynthetic proposal:
Alternative biosynthetic pathway according to Nitraria metabolism:
R 9 10
6
5
2
3
N
7 8 13
11 12
9
14 15
N H
yohimbine skeleton
10
21 20
11
13
18 16 17
6
5
2
3
N H
12
16 17
N
7
8 19
21
9
6
14 15
tangutorine skeleton
11
13 12
Scheme 1.43
B
5
N H
2
3
N H
19
D 20
E 16
L-lysine
165 O
1- Pictet-Spengler Michael 1,6
18 17
N
C
7
8
A
O
NH2
18
20 19
R 10
a plausible new achiral precursor
2- reduction
OH
R
H N
21
14 15
H N H H tangutorine 164
Comparison of two biosynthetic proposals for tangutorine.
More recently, a phytochemical study of Myrioneuron nutans (a species collected in Vietnam from a small genus within the Rubiaceae family) by Bodo and colleagues at the Museum National d’Histoire Naturelle in Paris revealed the presence of a new class of alkaloids that was named ‘‘Myrioneuron alkaloids.’’ Interestingly, these new alkaloids are closely related to the Nitraria alkaloids despite the taxonomic distance of the two families. Their biosynthesis is also discussed in the aforementioned review article. Representative structures [myrioneurinol (158) and myrobotinol (127)] are presented in Scheme 1.39 [90]; they constitute ideal targets for total synthesis inspired by biosynthetic pathways. 1.4.3.3 Endocyclic Enamines Overview: Biomimetic Observations From five- to seven-membered rings, the propensity of enamines to dimerize is schematically represented in Figure 1.6. Only two cases with biosynthetic consequences seem to be favorable for dimerization (16 and 38). Especially at neutral pH (e.g., ‘‘physiological’’ conditions), dimerization is spontaneous and therefore likely. This intrinsic reactivity pattern has probably partly governed Darwinian selection and the subsequent development of two metabolic pathways from tetrahydroanabasine (Scheme 1.39) [91, 92]. 1.4.4 Biomimetic Synthesis of Stenusine, the Spreading Agent of Stenus comma
Rove beetles Stenus comma (Coleoptera, Staphylinidae) synthesize the alkaloid stenusine (170) (Figure 1.7) in their pygidial glands as an escape mechanism on H N
R N
R N
n
R=H or Me n = 0 or 1
Figure 1.6
n
H N 35
Me N
39
H N Me N 38
Dimerization issue and evolutive consequences.
Me N H N
16
Me N
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
40
8
7
N
N
N
N
N
1
3 10
(3S, 10S ) 43%
stenusine 170 [Stenus comma]
Figure 1.7
(3R, 10S ) 40%
(3R, 10R ) 13%
(3S, 10R ) 4%
Stenusine: structure and stereochemical particularities.
water surfaces. The beetle propels itself over the water by expelling an oily substance with a high spreading capacity. Since the isolation and structure determination of stenusine (170) as the main component of the propulsion fluid, and a toxic chemical substance by Schildknecht et al. [93], numerous syntheses have been developed based on racemic and asymmetric strategies. Importantly, the enantioselective syntheses of isomeric (2S,3S)-stenusine and (2S,3R)-stenusine by Enders et al. [94] accompanied by chiral GC analysis for synthetic and natural samples revealed a strange fact. Indeed, stenusine is in fact a mixture of the four possible enantiomers in a ratio of 43 : 40 : 13 : 4 = (S, S) : (S, R) : (R, R) : (R, S) with a predominance (83 : 17) of the two epimers with a (S)-configuration on the side chain. It is unusual that an organism produces a natural compound as a mixture of the two pairs of enantiomers in a particular ratio. Husson, Kunesch, and colleagues proposed a very appealing biogenetic scheme for stenusine (170) that explained both the origin and the stereochemical particularities (Scheme 1.44) [95]. The essence of the biogenetic scenario is that stenusine derives most likely from l-lysine and l-isoleucine (171) and could be considered as the biological condensation of two of their metabolites, namely piperideine (15) and methylbutyraldehyde (172). Intermediates resulting from the iminium/enamine equilibrium could explain the formation of enantiomers at the C3 center and the partial racemization on the side chain leading after reduction and N-ethylation to stenusine. L-isoleucine 171 H N L-lysine
H
H N H
3 15
O
H 172
dehydration
N
OH
N
−reduction −N-ethylation
H N
stenusine 170
partial racemization
Scheme 1.44
Putative biosynthetic pathway to stenusine.
Phenyloxazolopiperidine 173, a stable equivalent of piperideine (15), was used to test the chemical basis of the hypothesis (Scheme 1.45). Reaction of 173 with (S)-2-methylbutyraldehyde (172) in ethanol at room temperature was conducted.
1.4 Biomimetic Synthesis of Alkaloids Derived from Lysine
Ph
Ph N
O
N
O
OH
(3R,10S ): ~40% (3S,10S ): ~40% (3R,10R ): ~10% (3S,10R ): ~10%
Raney Ni
H
N
N
Ph N
OH
Raney Ni, reflux
N
3
*
O
172
*
51% from 173
10
174
O
solvent
N
Ph
O
EtOH, RT
173 synthetic "biomimetic" stenusine 170
Ph
Ph
OH
41
Raney Ni
N
H
OH N
N Raney Ni
Scheme 1.45
Biomimetic synthesis of stenusine.
In the absence of air and after consumption of 173, addition of Raney nickel to the mixture directly led to the formation of stenusine in an impressive succession of reactions (oxazolidine ring opening, condensation with aldehyde, dehydration, reduction, debenzylation, oxidation of ethanol to ethanal, Schiff base formation, and reduction to give the N-ethyl group). Perhaps the most significant feature of the reaction was the total lack of stereoselectivity when starting from chiral non-racemic 173 and (S)-172! The total synthesis was not shadowed for all that: in fact, comparison of the optical rotations of natural stenusine (170) and synthetic stenusine brought striking support to the biosynthetic hypothesis (Figure 1.8), which was confirmed when a 13 C NMR quantitative evaluation of the stereomeric composition of the Mosher acid salt of synthetic stenusine gave quite similar proportions. The isolation of small amounts of aldehyde 175 or amounts up to 60% in boiling ethanol in the presence of air and molecular sieves provided strong support for the enticing sequence depicted in Scheme 1.45, and particularly for the occurrence enamine 174. A mechanism involving formation of an oxetane (176) and double β-cleavage was proposed that could account for the formation of 175 (Scheme 1.46). The story ended a few years later when Lusebrink and colleagues disclosed feeding experiments that clearly confirmed the biosynthetic scheme proposed earlier (Scheme 1.47) [96]. Deuterium-labeled amino acids and acetate combined
natural stenusine: [a]
20°C = 365
+5.8 (c =0.115, EtOH) [90]
synthetic "biomimetic" stenusine: (3R,10S ): ~43% (3S,10S ): ~40% (3R,10R ): ~13% (3S,10R ): ~4% chiral GC [90]
Figure 1.8
N
(3R,10S ): ~40% (3S,10S ): ~40% (3R,10R ): ~10% (3S,10R ): ~10% 13
[a]
C NMR estimation [91]
Natural and ‘‘biomimetic’’ stenusine: data comparisons.
20°C = 365
+5.0 (c =0.115, EtOH) [91]
42
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
Ph
OH N
Ph O2 (air)
OH
Ph
N
Ph
OH
OH N
N
H 176 O O
174
O
O
175
O
yield up to 60% Scheme 1.46
Biomimetic synthesis of stenusine: side reaction.
O [4,4,5,5-D4]-lysine H2N
CO2H
N
NH2
Scheme 1.47
[D3]-acetate ONa HO2C
NH2 L-[1′,1′,1′,2,3,4,4,5,5,5-D10]isoleucine
Stenusine biosynthesis.
with GC/MS analysis showed that lysine forms the piperidine ring of stenusine, the side chain originates from isoleucine, and the N-ethyl group from acetate.23)
1.5 Pelletierine-Based Metabolism 1.5.1 Pelletierine: A Small Alkaloid with a Long History
Pelletierine (129)24) (Scheme 1.48) was isolated by Tanret in 1878 [97] from pomegranate (Punica granatum L.) (stem and root traditionally used as an anthelmintic against tapeworms) as a volatile, optically active compound along with three other alkaloids: methylpelletierine, pseudopelletierine (177), and isopelletierine (178). Pelletierine (129) was described as a colorless oil and its structure was regarded by Hess and Eichel in 1917 as an aldehyde: that is, 3-(2-piperidyl)propionaldehyde (179) [98]. However, its exact structure has long been debated. Among many studies worldwide, a Japanese team isolated pelletierine from pomegranate root bark following Hess’ procedure, made a comparison with data provided by Hess and Eichel and concluded that the so-called pelletierine (129) was in fact isopelletierine (178) or 1-(2-piperidyl)-2-propanone [99]. NMR and IR studies by Gilman and Marion in 1961 were conducted on a amino acid origin are discussed in Chapter 8. usually have a polyketide and/or fatty acid origin; the case of stenusine is 24) Named in honor of Joseph Pelletier, French pharmacist and chemist (1788–1842). thereby remarkable. Alkaloids with a non-
23) Piperidinic alkaloids isolated from insects
1.5 Pelletierine-Based Metabolism 1878:"Sur la pelletiérine, alcaloïde de l'écorce du grenadier" by C. Tanret
1899: Piccinini suspects a piperidinic nucleus
Biomimetic synthesis of isopelletierine by Schöpf (1948) and by Ritchie (1949) 1969: elucidation of biosynthesis by Spenser
1900 H N
O 179
H N
Me O
Scheme 1.48
1950
1917: first structure proposed by Hess and Eichel
Me
H2N O
(−)-pelletierine 129 [Punica granatum]
43
180 racemization
1965: absolute configuration
1961: final clear conclusions on the structure of the "pelletierine of Tanret" by Gilman and Marion
H N
Me
H N
Me O
Me N
pseudopelletierine 177
O isopelletierine 178 = (±)-pelletierine
O
Pelletierine story from its discovery.
sample of pelletierine sulfate isolated by Tanret himself and showed the presence of a ketone instead of an aldehyde, confirming that the pelletierine of Tanret was 1-(2-piperidyl)-2-propanone [100]. From this point, (−)-pelletierine (129) is referred to as (−)-1-(2-piperidyl)-2-propanone and isopelletierine (177) (a term that should be avoided) to racemic pelletierine. The curious history of this small molecule has been brilliantly analyzed with an historical perspective in the 1960s [101]. Later, both the absolute configuration (D) [102] and biosynthesis, via l-lysine (3), were disclosed [103]. Part of the questioning concerning the exact structure of pelletierine (129) found its roots in the epimerization that is now a well-known phenomenon with 129 and is believed to be a base-catalyzed reaction operating via retro-Michael intermediate 180. Also found in pomegranate, pseudopelletierine (177), the existence of which unlike pelletierine has never been questioned, is the higher homolog of tropinone (125). 1.5.2 Biomimetic Synthesis of Pelletierine and Pseudopelletierine 1.5.2.1 Pelletierine (129)25) A Mannich reaction of acetoacetic acid followed by decarboxylation was early-on assumed to be at the origin of the propanone side chain of pelletierine (129) (Scheme 1.49). This was verified independently by Ritchie and colleagues and Sch¨opf in 1949 [104] and was proved to be biosynthetically totally correct 30 years 25) Compared to coniine, a similar structure in
terms of complexity, pelletierine has been
by far less studied in terms of asymmetric syntheses.
44
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
O O
H N
various conditions (see bibliography)
Me N
OH Scheme 1.49
(±)-pelletierine 129
O
[up to 80%]
−CO2
Pelletierine biomimetic conditions.
later by incorporation experiments in vivo [102]. Many improvements of these pioneering syntheses were subsequently disclosed (including the study of competitive dimerization of piperideine (15) into tetrahydroanabasine (16) [105]) as well as other ‘‘non-biomimetic’’ syntheses, including asymmetric strategies [106]. Pelletierine was also used as a starting material for biomimetic Claisen–Schmidt reactions known as the ‘‘pelletierine condensation’’ by different groups [107]. 1.5.2.2 Pseudopelletierine Pseudopelletierine (177) was prepared by the same biomimetic Mannich-type reaction as for tropinone but starting from glutaraldehyde (51) by Robinson in 1924, Sch¨opf in 1937, and reinvestigated by Cope in 1951 [108]. 1.5.3 Lobelia and Sedum Alkaloids
Lobelia and Sedum alkaloids are closely related to pelletierine in terms of structures and biosynthesis. Examples are given in Scheme 1.50 (181–183); the central piperidine ring of these alkaloids is known to derive from l-lysine. Interesting reviews have been published by Bates [109] and Felpin and Lebreton [110] to which the reader can refer advantageously. Numerous total syntheses have also been disclosed [111]. Me N
H N OH
OH
H
Me N O
O
(+)-norallosedamine 181
(−)-lobeline 182
[Sedum acre]
[Lobelia inflata]
N H
SCoA N
O
L-lysine
Ph (+)-lobinaline 183 [Lobelia cardinalis]
Scheme 1.50
Sedum and Lobelia alkaloids: selected structures and biosynthesis.
1.5.4 Lycopodium Alkaloids 1.5.4.1 Overview, Classification, and Biosynthesis A significant number (>250 in 2010) of alkaloids with original and intricate structures are found in the Lycopodium genus. This genus has over 500 species
1.5 Pelletierine-Based Metabolism
Me
Me
Me
N H N
O
lycopodine 184 [Lycopodium clavatum]
Figure 1.9
N H
N
lycodine 185 [Lycopodium annotinum]
N
OH
O
fawcettimine 186 [Lycopodium fawcetti ]
N H
Me
phlegmarine 187 [Lycopodium phlegmaria]
Examples of Lycopodium alkaloids.
and only around 10% of the species have been studied for their alkaloid content. Biological activities of such compounds can also be promising, as exemplified by well known huperzine A, which exhibits potent anticholinesterase activity. These alkaloids have attracted a great deal of interest from biogenetic and biological points of view as well as for providing challenging targets for total synthesis, especially because of their scarcity. W. A. Ayer has divided the Lycopodium alkaloids into four classes: lycopodine, lycodine, fawcettimine, and miscellaneous, with lycopodine (184), lycodine (185), fawcettimine (186), and phlegmarine (187) as representative compounds, respectively, for each class (Figure 1.9). Owing to the number of alkaloids now described, the classification has become tricky, especially with the isolation of highly complex structures (especially in the last ten years). Many comprehensive reviews have been published covering extraction, structure determination, biosynthetic hypotheses, total syntheses, and biological properties [112]. Based on biosynthetic considerations, another classification of Lycopodium alkaloids may be considered. In fact, even if the biosynthesis of the alkaloids is still not completely understood with very few feeding experiments conducted up to now, research groups have provided evidence for a lysine origin of these alkaloids. Furthermore, such as in the case of the biosynthesis of lycopodine (184) demonstrated by Spenser and colleagues (Scheme 1.51) [113], pelletierine units or analogs thereof are integrated in the structure. Therefore, Lycopodium alkaloids can be formally seen as pelletierine-type (propylpiperidine units26) – C5 N–C3 ) derived alkaloids and could then be classified according to the number of such units in the final natural substance, as proposed as early as 1960 by Conroy27) [114]. As a consequence, some alkaloids may result from the condensation of two [e.g., nankakurine A (188), huperzine A (189), lyconadin A (190)], three [e.g., himeradine A (191), lycoperine A (192)], or even four [complanadine D (193)] pelletierine-type units as determined by the number of piperidine cycles in the molecules for the less rearranged skeletons (Scheme 1.52). At least chemically speaking, Lycopodium alkaloids can be 26) Whether this unit comes from pelletierine 27) Conroy anticipated a C5 –C3 common pat-
itself or from the de novo condensation of a piperideine of type 15 with C3 -acetone-type equivalent such as acetonedicarboxylic acid.
tern in Lycopodium alkaloids but proposed a totally polyketide origin for it.
45
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
46
*Biosynthesis of Lycopodium alkaloids: COOH
CO2
COOH
O
COOH O O
NH CO2
15
N H
NH
NH
N
N H pelletierine 129
O
lycopodine 184
*Lycopodium alkaloids formalism: d
H d N
d d
N
N
N
H
H
N
O d pelletierine 129
fundamental synthons "C5NC3" pelletierine-type
Scheme 1.51
N
nankakurine A 188 [Lycopodium hamiltonii ]
Biosynthetic origin and analogy with pelletierine.
considered as being biosynthetically derived from simple more or less oxidized pelletierine-type units [see the example of nankakurine (188) in Scheme 1.51]. H N
N ×4
×2 O "C5 NC3" formal pelletierine unit
N H
H N H N
×3
Me
N H N
O
O
N
complanadine D 193 [Lycopodium complanatum]
N N
H
O
himeradine A 191 [Lycopodium chinense]
NH2 huperzine A 189 [Huperzia serrata]
NH
O
H H HN H N H
Scheme 1.52
N H
O
lycoperine A 192 [Lycopodium hamiltonii ]
lyconadin A 190 H
[Lycopodium complanatum]
A classification based on the formal analogy with pelletierine.
From the 1960s to the 1980s, the emergence of new strategies capable of addressing the Lycopodium alkaloid challenges permitted the total syntheses of complex structures by pioneering groups. This includes, among many others, work by Wenkert et al. [115], Evans [116], Heathcock et al. [117], and Schumann et al. [118], which often display interesting biomimetically related steps. Recent total syntheses of Lycopodium alkaloids were reviewed in 2009 (covering up to 2008) [112c, 119]. Most of them constitute the state of the art in total synthesis. Concerning biomimetic chemistry, we focus in this chapter on:
1.5 Pelletierine-Based Metabolism
• Recent relevant ‘‘biomechanistic’’ conversions of alkaloids that permitted structure elucidation and the establishment of clear biosynthetic links. They constitute ideal examples of chemical predisposition in the world of natural substances. These studies exemplify magnificently the interconnection of numbers of Lycopodium alkaloids and highlight how simple reactions can explain part of both the diversity and complexity within this large family of alkaloids, which all derive from simple building blocks. • Recent total synthesis that may feature at least one biomimetic step (usually a cascade reaction to form polycyclic skeletons). 1.5.4.2 Biomimetic Rearrangement of Serratinine into Serratezomine A Serratezomine A (194) (Scheme 1.53) has been isolated from Lycopodium serratum var. serratum by the Kobayashi group [120], which proposed a biogenetic origin from serratinine (195), also found in this plant, through its N-oxide form 196 followed by a Polonovski-type fragmentation (Scheme 1.53). O
O
OH
OH
CH3 N
O
OH CH3
CH3 O
N
OH
N OR
OH
OH
196
serratinine 195
[Lycopodium serratum var. serratum ]
Polonovski-type fragmentation
O O
O CH3
N
OH
HO
O
reduction
O CH3
CH3 N
N OR OH
OH
serratezomine A 194
Scheme 1.53
Plausible biosynthetic correlation between serratinine and serratezomine A.
To verify this proposal (Scheme 1.54), the conditions of a modified Polonovski reaction (also known as the Polonovski–Potier reaction) were applied to serratinine (195): treatment with m-chloroperbenzoic (m-CPBA) acid followed by addition of trifluoroacetic anhydride and then sodium cyanoborohydride gave two compounds [121]. One of them was identified as serratezomine A (194) and was predominant at lower temperature (−50 and –20 ◦ C) while the amount of the other (197) increased with temperature (0–20 ◦ C). Formation of this latter indicates an acid-catalyzed-cleavage of the lactone ring. 1.5.4.3 Biomimetic Conversion of Serratinine into Lycoposerramine B More recently, an alkaloid possessing an oxime function, lycoposerramine B (198, Scheme 1.55), was isolated from Lycopodium serratum. To confirm its structure, which was inferred by classical spectroscopic analysis, its semi-synthesis from serratinine (195) (for which the absolute configuration was known) was successfully attempted [122]. The retrosynthetic analysis (see box in Scheme 1.55) consisted of
47
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
48
O O
OH CH3 N
N
serratinine 195
CH3
+ O
OH O
serratezomine A 194 p -TsOH
O H
OH
CH3
3) NaBH3CN
OH
N
O
1) m - CPBA 2) TFAA
O
197
OH HO
CH3 N
O CH3
OH N
OH
stereoselective reduction
translactonization
Scheme 1.54
Biomimetic conversion of serratinine into serratezomine A.
regioselective oximation
selective oxidation
OH
O 5
13
4
CH3 N
OH
serratinine 195 [Lycopodium serratum]
O
CH3
8
N
OH N
removal
lycoposerramine B 198 [Lycopodium serratum ]
H3C
ring-opening /methylation
O
OH
1- C- 8 /C-13 acetylation: Ac2O, py, 100 °C (98%)
CH3 2- C- 8 hydrolysis: 10% HCl, N
O
(84%)
N
3-C-8 dehydroxylation: NaH, HMPA, CS2, MeI, rt,THF (47% + 49% sm) (77%) then Bu3SnH, AIBN, PhMe,
OH
serratinine 195
(quant.)
1-KOH, MeOH,
OAc CH3
2-Jones reagent, acetone, rt (98%)
199 O
O CH3
tert -BuOK, tert-BuOH, rt, (202: 32%; 201: 30%)
N 200
O H
O H
O CH3 N
H3C
+
O CH3 N
H3C
202
reductive ring-opening 1-MeOTf, MeCN, rt
2-Zn, AcOH, rt 201 (X-ray) (202: 30%, 201: 34%) Et2NH, EtOH
OH HO N
O
N
CH3
Scheme 1.55
NH2OH·HCl
+
Et2N
O
CH3
N
(46%) H3C lycoposerramine B 198
O
N H3C
(19%)
CH3
N H3 C
203
Biomimetic conversion of serratinine into lycoposerramine B.
1.5 Pelletierine-Based Metabolism
a removal of the hydroxyl group at C8, oxidation of the secondary hydroxyl at C13, ring-opening at the C4-N bond, and regioselective oximation at C5. A monoacetate was first prepared starting from serratinine (195) in a two-step sequence followed by a Barton–McCombie dehydroxylation. After acetate removal and oxidation to 200, the reductive ring-opening reaction was then conducted and yielded two C4 epimeric compounds (201 and 202; the latter could be epimerized in basic conditions). To selectively convert the carbonyl at C5 into an oxime, 201 was first treated with diethylamine (steric hindrance at C13 probably explains the selectivity). On the putative intermediate 203, hydroxylamine reacted preferentially with the more reactive iminium function at C5. The major isomer was identical with natural lycoposerramine B (198) (including CD spectra). 1.5.4.4 Biomimetic Interrelations within the Lycoposerramine and Phlegmariurine Series Several alkaloids belonging to the fawcettimine group were isolated from Lycopodium serratum by the Takayama group, including several lycoposerramines [B (198) and C (204)] and related phlegmariurines A (205) and B (206) (Scheme 1.56) [123]. Lycoposerramine C could be converted into phlegmariurines A and B by ring opening in the presence of a base (whereas potassium tert-butylate furnished exclusively 205, a mixture of 205 and 206 was obtained when treating with sodium methoxide). CD curves of semi-synthetic and natural phlegmariurine A were compared and shown to be identical, demonstrating thereby the same absolute configuration.
H3C
O
HO
O
HO
O
H3C N
N
fawcettimine 186
O H3C
H3C
O N
O N
lycoposerramine C 204
phlegmariurine A 205 phlegmariurine B 206
O NaOMe, MeOH
H3C
+
O N O HO
O
H3C
H3C N
O N
O phlegmariurine A 205 (45%)
N
phlegmariurine B 206 (5%)
O
lycoposerramine C 204
Scheme 1.56
O H3C
ter -BuOK THF, 0 °C
phlegmariurine A 205
H3C O N
(95%)
Biomimetic conversions in the lycoposerramine/phlegmariurine series.
The structure of lycoposerramine D (207) (Scheme 1.57) revealed the presence of an isolated methylene carbon that corresponds to an extra carbon compared to common C16 -type Lycopodium alkaloids. A Mannich reaction involving iminium
49
50
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples HO
O
O
O
H3C
H3C
N
4
H3C
HN
OH lycoposerramine P 209
Scheme 1.57
O
O
H3C
OH
O
OH
OH HCHO,
O
N H
208
O
H3C
N
OH
O
HBr cat, MeOH / H2O
N
lycoposerramine D 207 (12%)
Biomimetic conversion of lycoposerramine P into lycoposerramine D.
208 could easily explain the biosynthesis of lycoposerramine D (207) from lycoposerramine P (209), another alkaloid isolated and characterized concomitantly. This hypothesis was clearly ascertained as, when treated with formaldehyde, lycoposerramine P (209) yielded a semi-synthetic lycoposerramine D (207) that was totally identical to its natural counterpart [123a]. 1.5.4.5 When Chemical Predisposition Does Not Follow Biosynthetic Hypotheses: Unnatural ‘‘Lycopodium-Like’’ Alkaloids Yang and coworkers described an original work featuring a total synthetic dead end that unexpectedly gave interesting results (Scheme 1.58) [124]. The total synthesis of advanced intermediate 210 towards the total synthesis of lycojapodine A (211) was achieved but failed to give, through a possible biomimetic cascade, the desired carbinolamine lactone of the target molecule. Two competitive cyclizations can occur on such a diketone. But in fact, whatever the conditions tested, the initial 6/9 bicycle of 212 underwent the ‘‘wrong’’ pathway, leading after an impressive imine/enamine cascade to unnatural compound 213. This unexpected O HO H3C O
H
O H3C N
O RO O
O
H
H3C BocN
1
HN
O
O
H3C
13 steps
RO O H3C
O
O
N
O 1 expected biomimetic final stage
lycojapodine A 211 [Lycopodium `japonicum]
2
212 advanced biomimetic intermediate 210
2 observed unexpected chemical predisposition
O RO O H3C N
O
O OH
N
N
RO
RO
CH3
conditions: - R =OH: AcOH / H2O,100 °C (213: 45%) - R =OMe: ZnBr2, CH2Cl2 or TFA, CH2Cl2 (213: 65%)
CH3 O
O O unnatural "Lycopodium like" alkaloid 213
Scheme 1.58 Unexpected results: an interesting dead end in the course of the total synthesis of lycojapodine A.
N O
CH3
1.5 Pelletierine-Based Metabolism
51
chemoselectivity may facilitate a better understanding of the biosynthesis and rearrangements of complex Lycopodium alkaloids. 1.5.4.6 Total Synthesis of Cermizine C and Senepodine G Among others, two small quinolizidine alkaloids, namely cermizine C (214) and senepodine G (215) (Scheme 1.59), were isolated from the club moss Lycopodium cernuum and Lycopodium chinense, respectively, as minor compounds (respectively 0.00008 and 0.00005%) by the Kobayashi group [125]. Simple biosynthetic hypotheses may be put forward to explain the formation of senepodine G, a possible direct precursor of cermizine C through iminium reduction. N
H3 C
N
H3C
H
HN
(+)-cermizine C 214 (–)-senepodine G 215 [Lycopodium cernuum] [Lycopodium chinense]
Scheme 1.59
O
RO
CH3
H3C
O
O
HN
pelletierine
N
O
–CO2
Cermizine C and senepodine G: structure and plausible biosynthesis. O
O O
H
O
(2.4 eq)
(±)-pelletierine 129
O
H
H 3C O
O HN
O
AcOH EtOH 60 °C
O HN
H
H3C
N O
O 216
H
O
O
H
H3C
K2CO3, MeOH
N O
219
H
H 3C
O
O
THF, 60 °C, then 3M HCl in MeOH (100%)
H 3C
H 7
NaBH4, MeOH H3C
5
N
9
1
5 7
N 9
H
O 217 H 5
7
(82%)
CH3 (±)-7-epi-senepodine G 7-epi-215 H3C
9
N
1
CH3 (±)-5-epi-cermizine C 5-epi-214 5
1
H 3C
Scheme 1.60
N O
OH
MeMgBr N
O
218 (68%)
H
H
H 3C
N
N
H2 (50 psi) PtO2, EtOH (96%) H 3C
H
H3C
H3C
7
H 3C
9
N
Biomimetic synthesis of epi-senepodines from pelletierine.
Snider and colleagues set out to synthesize these two alkaloids [126]. Despite no mention of biosynthesis, their first approach started from pelletierine (129) and is worth noting as an interesting example of the conversion of one simple natural product into another (Scheme 1.60). A Knoevenagel reaction between the
1
52
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
ketone of (±)-pelletierine and Meldrum’s acid (a synthetic surrogate of biosynthetic acetate-derived building blocks) afforded intermediate 216, which would cyclize to give directly the corresponding quinolizidine skeleton in a beautiful cascade of reactions. Indeed, 217 was not isolable but furnished β, γ -isomer 218, which equilibrated in basic conditions to a mixture of α, β-isomer 219 and 218 (3 : 1). Hydrogenation of the double bonds of 218 and 219 followed by alkylation furnished (±)-7-epi-senepodine G (7-epi-215), which provided (±)-5-epi-cermizine C (5-epi-214) after stereospecific reduction using sodium borohydride (axial attack from the less hindered top face). An alternative synthetic scheme starting from (S)-220 via unsaturated lactam 221 enabled the synthesis of (−)-senepodine G (215) and (−)-cermizine C (214) (Scheme 1.61). Discrepancies in the comparison of optical rotation signs and magnitudes between natural and synthetic samples were also discussed by the authors. Very small amounts of natural compounds and the small value of the optical rotation clearly make the establishment of the absolute configuration of senepodine G and cermizine C difficult (although probably closely biosynthetically related, they were isolated from two different species). HO
H
H
4 steps (51%)
N
HN (S )-220
O 221
Scheme 1.61
2 steps (78%) H3C
H
(99%) H3C
N
N CH3
H
CH3 ·HCl
Cl
(–)-senepodine G 215
(–)-cermizine C 214
synthetic: [a]23D–77 (c 1.0, MeOH)
synthetic: [a]23D−2 (c 0.4, MeOH)
natural: [a]26D –35 (c 0.3, MeOH)
natural: [a]26D +4 (c 0.8, MeOH)
Biomimetic synthesis of senepodine G and cermizine C.
1.5.4.7 Biomimetic Steps in the Total Synthesis of Fastigiatine Fastigiatine (222, Scheme 1.62) was isolated as a minor alkaloid from Lycopodium fastigiatum in 1985 by the MacLean group [127]. The first total synthesis of this complex alkaloid was only accomplished in 2010 by the Shair group (Scheme 1.62) [128]. Despite being non-biomimetic in essence, the strategy beautifully demonstrated that such a total synthesis not only implied the development of highly efficient chemical transformations but also permitted interesting biosynthetic proposals, that is, ring formation implying an intermediate of type 223. The approach relies on the readily available cyclopropane 224, which was converted in ten steps into intermediate 225. The latter underwent an impressive sequence of reactions including a transannular Mannich reaction in acidic conditions to give 226 featuring the core of the alkaloid. Two steps consisting of functional manipulations of the skeleton afforded (+)-fastigiatine (222) in 15 steps from 224 with an average 30% yield. The suitability of intermediate 227 to easily undergo an amazing cascade to build in one step the challenging skeleton of fastigiatine encouraged the author to propose an intermediate such as 223 as the plausible biosynthetic intermediate toward these molecules.
1.5 Pelletierine-Based Metabolism Me
Me O H N
Me N
?
Me (+)-fastigiatine 222 [Lycopodium fastigiatum] (X-ray)
N N
Me
223
tBuO O
O O
4 steps
Cl
O
EtO
10 steps
Me
NH
O
O 224
H
NH
NHNs
225
1
formal [3+3] Me CO2t Bu H N
Me CO2t Bu TFE,
N OH
O
13
O
Me CO2t Bu H N
4 2
NHNs
OH NHMe
MeHN
Me CO2t Bu 6 H 7 N
NHR R= Ns (92%) 227: R= Me (87%)
Me CO2t Bu H N Me N
Me CO2t Bu H N
transannular Mannich Me
N
Me 1-p -TsOH 2-Ac2O
O N
(+)-fastigiatine 222 (X-ray)
Me N 226 (85%)
(81%, 2 steps)
Ns = nosyl
Scheme 1.62
A biomimetic cascade as a crucial step in the total synthesis of fastigiatine.
1.5.4.8 Biomimetic Steps in the Total Synthesis of Complanadine A Earlier in 2010, a similar cascade reaction with evident biomimetic relevance was highlighted in the beautiful total synthesis of complanadine A (228) by Sarpong and colleagues (Scheme 1.63). [119c,d] who exploited a synthesis previously described by Schumann and Naumann [129]. N-Desmethyl α-obscurine (229) was prepared in a one-pot procedure from enamide 230 and enantiomerically pure 231. The next step consisted of an oxidation of lactam 232 to the corresponding pyridone 233 in a somewhat biomimetic way using lead tetraacetate. Such a pyridone nucleus is shared by many Lycopodium alkaloids such as huperzine A (189) and lyconadin A (190) (Scheme 1.52). These few highlights of selected total syntheses conclude this section dedicated to Lycopodium alkaloids and also conclude this chapter devoted to the rich chemistry of alkaloids derived from ornithine and lysine. The exploration of the chemistry of these natural products is still ongoing. With the discovery of new complex structures – but in minute amounts – along with promising biological activities (e.g., neurotrophic properties for certain Lycopodium alkaloids), new synthetic strategies will undoubtedly be needed to secure enough material for biological purposes. Toward this end, no doubt biomimetic considerations will be able to
53
54
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples
O
Me
NH2
H
O
H N
Me
231 O Me
Me
Me
O
H N
NH2
O
NH2
O
O
230 70% HClO4, Me H O N N Boc
N H Me H N O
Boc2O, NEt3 (65% over 2 steps)
232 Pb(OAc)4,CHCl3, rt (84%)
N H N - desmethyl a-obscurine 229 (multigram scale) Me
Suzuki cross coupling N
Me 5 steps (20%)
H N
O
N
N Boc 233
H
H N
N H
(+)-complanadine A H 228 Me
Scheme 1.63
Total synthesis of complanadine A by the Sarpong group.
answer synthetic problems. As mentioned above, organic synthesis is endowed with highly efficient biomimetic reactions such as the Mannich reaction and the Robinson–Sch¨opf condensation to tackle the access to more and more complex structures. References 1. Wu, G. (2009) Amino Acids, 37, 1–17. 2. Wu, G., Bazer, F.W., Davis, T.A., Kim,
S.W., Li, P., Rhoads, J.M., Satterfield, M.C., Smith, S.B., Spencer, T.E., and Yin, Y. (2009) Amino Acids, 37, 153–168. 3. Torruella, G., Suga, H., Riutort, M., Peret´o, J., and Ruiz-Trillo, I. (2009) J. Mol. Evol., 69, 240–248 and references cited therein. 4. Azevedo, R.A. and Lea, P.J. (2001) Amino Acids, 20, 261–279. 5. Dewick, P.M. (2009) Medicinal Natural Product: A Biosynthetic Approach, 3rd edn. John Wiley & Sons, Ltd, Chichester.
6. Hashimoto, T., Yukimune, Y., and
7. 8.
9.
10.
Yamada, Y. (1989) Planta, 178, 131–137. Biastoff, S., Brandt, W., and Dr¨ager, B. (2009) Phytochemistry, 70, 1708–1718. See among many others: Brown, A.M., Robins, D.J., Witte, L., and Wink, M. (1991) Plant Physiol. (Life Sci. Adv.), 10, 179–185. (a) Tanner, J.J. (2008) Amino Acids, 35, 719–730; (b) Verbruggen, N. and Hermans, C. (2008) Amino Acids, 35, 753–759 (a) Vidal, J. (2009) in Amino Acids, Peptides and Proteins in Organic Chemistry, vol. 2 (ed. A.B. Hughes), Wiley-VCH
References
11.
12.
13.
14. 15.
16.
17. 18.
19.
20.
Verlag GmbH, Weinheim, pp. 35–92; (b) Ciufolini, M.A. and Xi, N. (1998) Chem. Soc. Rev., 27, 437–446. See these four extensive review articles: (a) Guggisberg, A. and Hesse, M. (1983) in The Alkaloids, Chemistry and Pharmacology, vol. 22 (ed. A. Brossi), Academic Press Inc., New York, pp. 85–188; (b) Sch¨afer, A., Benz, H., Fiedler, W., Guggisberg, A., Bienz, S., and Hesse, M. (1994) in The Alkaloids, Chemistry and Pharmacology, vol. 45 (eds A. Brossi and G.A. Cordell), Academic Press Inc., San Diego, pp. 1–125; (c) Bienz, S., Detterbeck, R., Ensch, C., Guggisberg, A., H¨ausermann, U., Meisterhans, C., Wendt, B., Werner, C., and Hesse, M. (2002) in The Alkaloids, Chemistry and Pharmacology, vol. 58 (ed. G. Cordell), Academic Press Inc., San Diego, pp. 83–338; (d) Bienz, S., Bisseger, P., Guggisberg, A., and Hesse, M. (2005) Nat. Prod. Rep., 22, 647–648. The following article is more focused on polyamine catabolism (especially in humans) but is worth reading: Seiler, N. (2004) Amino Acids, 26, 217–233. Dimitrov, V., Geneste, H., Guggisberg, A., and Hesse, M. (2001) Helv. Chim. Acta, 84, 2108–2118. Struve, C. and Christophersen, C. (2003) Heterocycles, 60, 1907–1914. Houen, G., Struve, C., Søndergaard, R., Friis, T., Anthoni, U., Nielsen, P.H., Christophersen, C., Petersen, B.O., and Duus, J.Ø. (2005) Bioorg. Med. Chem., 13, 3783–3796. Crocker, S.J., Loeffler, R.S.T., Smith, T.A. and Sessions, R.B. (1983) Tetrahedron Lett., 24, 1559–1560 and references cited therein. Feth, F., Wray, F., and Wagner, K.G. (1985) Phytochemistry, 24, 1653–1655. Wei, X., Sumithran, S.P., Deaciuc, A.G., Burton, H.R., Bush, L.P., Dwoskin, L.P., and Crooks, P.A. (2005) Life Sci., 78, 495–505. (a) Leete, E., Kim, S.H., and Rana, J. (1988) Phytochemistry, 27, 401–406; (b) see ref 17. Leonard, N.J. and Cook, A.G. (1959) J. Am. Chem. Soc., 81, 5627–5631.
21. Swan, G.A. and Wilcock, J.D. (1974) J.
Chem. Soc. Perkin Trans. 1, 885–891. 22. Franck, B. and Randau, D. (1966)
Angew. Chem. Int. Ed. Engl., 5, 131. 23. Sch¨ opf, C., Komzak, A., Braun,
24.
25.
26. 27.
28. 29.
30.
31.
32.
33.
34.
35.
36.
F., Jacobi, E., Bormuth, M.-L., Bullnheimer, M., and Hagel, I. (1948) Justus Liebigs Ann. Chem., 559, 1–42. Kessler, H., M¨ohrle, H., and Zimmermann, G. (1977) J. Org. Chem., 42, 66–72. Sch¨opf, C., Braun, F., Koop, H., Werner, G., Bressler, H., Neisius, K., and Schmadel, E. (1962) Justus Liebigs Ann. Chem., 658, 156–168. Sch¨opf, C., Braun, F., and Komzak, A. (1956) Chem. Ber., 89, 1821–1833. See among others: (a) Sch¨opf, C., Arm, H., and Krimm, H. (1951) Chem. Ber., 84, 690–699; (b) Sch¨opf, C. and Otte, K. (1956) Chem. Ber., 89, 335–340. El-Shazly, A.M., Dora, G., and Wink, M. (2005) Pharmazie, 60, 949–952. Salame, R., Gravel, E., Retailleau, P., and Poupon, E. (2010) Org. Biomol. Chem., 8, 2522–2528. Tashima, T., Imai, M., Kuroda, Y., Yagi, S., and Nakagawa, T. (1991) J. Org. Chem., 56, 694–697. See for example: Franc¸ois, D., Lallemand, M.-C., Selkti, M., Tomas, A., Kunesch, N., and Husson, H.-P. (1998) Angew. Chem. Int. Ed., 37, 104–105. (a) See for example: Roulland, E., Cecchin, F., and Husson, H.-P. (2005) J. Org. Chem., 70, 4474–4477 and references cited therein; (b) Review article: Husson, H.-P. and Royer, J. (1999) Chem. Soc. Rev., 28, 383–394. Zacharius, R.M., Thompson, J.F., and Steward, F.C. (1952) J. Am. Chem. Soc., 74, 2949. Review articles: (a) He, M. (2006) J. Ind. Microbiol. Biotechnol., 33, 401–407; (b) Broquist, H.P. (1991) Annu. Rev. Nutr., 11, 435–448. Gatto, G.J., Boyne, M.T., Kelleher, N.L., and Walsh, C.T. (2006) J. Am. Chem. Soc., 128, 3838–3847 and references cited therein. Zabriskie, T.M., Kelly, W.L., and Liang, X. (1997) J. Am. Chem. Soc., 119, 6446–6447.
55
56
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples 37. See for example: Naranjo, L., Martin
38.
39.
40. 41.
42.
43.
44.
45.
46.
47.
48. 49.
50.
de Valmaseda, E., Ba˜ nuelos, O., Lopez, P., Ria˜ no, J., Casqueiro, J., and Martin, J.F. (2001) J. Bacteriol, 183, 7165–7172. Matsumoto, S., Yamamoto, S., Sai, K., Maruo, K., Adachi, M., Saitoh, M., Nishizaki, T. (2003) Brain Res., 980, 179–184 and references cited therein. Recent synthesis of l-pipecolic acid and analogs: Lemire, A. and Charette, A.B. (2010) J. Org. Chem., 75, 2077–2080 and references cited therein for previous work. Kadouri-Puchot, C. and Comesse, S. (2005) Amino Acids, 29, 101–130. Aketa, K.-I., Terashima, S., and Yamada, S.-I. (1976) Chem. Pharm. Bull., 24, 621–631. Rossen, K., Kolaroviˇc, A., Baskakov, D., and Kiesel, M. (2004) Tetrahedron Lett., 45, 3023–3025. (a) Pal, B., Ikeda, S., Kominami, H., Kera, Y., and Ohtani, B. (2003) J. Catal., 217, 152–159; (b) previous related work: Ohtani, B., Tsuru, S., Nishimoto, S., Kagiya, T., and Izawa, K. (1990) J. Org. Chem., 55, 5551–5553. (a) Ober, D. and Kaltenegger, E. (2009) Phytochemistry, 70, 1687–1695; (b) Robins, D.J. (1995) in The Alkaloids, Chemistry and Pharmacology vol. 46 (ed. G.A. Cordell), Academic Press Inc., San Diego, pp. 1–61. Marson, C.M., Pink, J.H., Smith, C., Hursthouse, M.B., and Abdul Malik, K.M. (2000) Tetrahedron Lett., 41, 127–129. (a) Gribble, G.W. and Soll, R.M. (1981) J. Org. Chem., 46, 2433–2434; (b) Gribble, G.W., Switzer, F.L., and Soll, R.M. (1988) J. Org. Chem. 53, 3164–3170. Lamberton, J.A. (1973) in The Alkaloids, vol. 14 (ed. R.H.F. Manske), Academic Press Inc., New York, pp. 325–346 and references cited in Reference [54]. Onaka, T. (1971) Tetrahedron Lett., 12, 4395–4398. Bick, I.R.C., Gunawardana, Y.A.G.P., and Lamberton, J.A. (1985) Tetrahedron, 41, 5627–5631. Tufariello, J.J. and Ali, Sk.A. (1979) Tetrahedron Lett., 20, 4445–4448.
51. Chen, C.-K., Hortmann, A.G., and
52.
53.
54.
55.
56.
57. 58.
59.
60.
61.
Marzabadi, M.R. (1988) J. Am. Chem. Soc., 110, 4829–4831. Diker, K., El Biach, K., D¨oe´ de Maindreville, M., and L´evy, J. (1997) J. Nat. Prod., 60, 791–793. (a) See for examples: grandisines A and B: Carroll, A.R., Arumugan, G., Quinn, R.J., Redburn, J., Guymer, G., and Grimshaw, P. (2005) J. Org. Chem., 70, 1889–1892; (b) grandisines C–G: Katavic, P.L., Venables, D.A., Forster, P.I., Guymer, G., and Carroll, A.R. (2006) J. Nat. Prod., 69, 1295–1299; (c) habbemines A and B: Katavic, P.L., Venables, D.A., Rali, T., and Carroll, A.R. (2007) J. Nat. Prod., 70, 866–868. See for example, the total synthesis of grandisine D: Kurasaki, H., Okamoto, I., Morita, N., and Tamura, O. (2009) Org. Lett., 11, 1179–1181 and references cited therein. Wu, J.-B., Cheng, Y.-D., Kuo, S.-C., Wu, T.-S., Iitaka, Y., Ebizuka, Y., and Sankawa, U. (1994) Chem. Pharm. Bull., 42, 2202–2204. Twin, H., Wen, W.W.-H., Powell, D.A., Lough, A.J., and Batey, R.A. (2007) Tetrahedron Lett., 48, 1841–1844. Snider, B.B. and Neubert, B.J. (2005) Org. Lett., 7, 2715–2718. Baumgartner, B., Erdelmeier, C.A.J., Wright, A.D., Rali, T., and Sticher, O. (1990) Phytochemistry, 29, 3327–3330. D¨atwyler, P., Ott-Longoni, R., Sch¨opp, E., and Hesse, M. (1981) Helv. Chim. Acta, 64, 1959–1963. Ott-Longoni, R., Viswanathan, N., and Hesse, M. (1980) Helv. Chim. Acta, 63, 2119–2129. (a) Tapiolas, D.M., Bowden, B.F., Abou-Mansour, E., Willis, R.H., Doyle, J.R., Muirhead, A.N., Liptrot, C., Llewellyn, L.E., Wolff, C.W.W., Wright, A.D., and Motti, C.A. (2009) J. Nat. Prod., 72, 1115–1120; (b) This latter publication includes the revision of the structure of previously described ‘‘eusynstyelamide’’ isolated from Eusynstyela misakiensis, see: Swersey, J.C., Ireland, C.M., Cornell, L.M., and Peterson, R.W. (1994) J. Nat. Prod., 57, 842–845.
References 62. (a) Casapullo, A., Finamore, E., Minale,
63.
64. 65. 66. 67. 68.
69. 70.
71.
72.
73.
74.
75. 76.
L., and Zollo, F. (1993) Tetrahedron Lett., 34, 6297–6300; (b) Casapullo, A., Minale, L., Zollo, F., and Lavayre, J. (1994) J. Nat. Prod., 57, 1227–1233. Snider, B.B., Song, F., and Foxman, B.M. (2000) J. Org. Chem., 65, 793–800. Barykina, O.V. and Snider, B.B. (2010) Org. Lett., 12, 2664–2667. Robinson, R.J. (1917) J. Chem. Soc., 111, 762–768. Sch¨opf, C. (1937) Angew. Chem., 50, 779–787. Humphrey, A.J. and O’Hagan, D. (2001) Nat. Prod. Rep., 18, 494–502. J¨onsson, D., Molin, H., and Und´en, A. (1998) Tetrahedron Lett., 39, 1059–1062. Mikami, K. and Ohmura, H. (2002) Chem. Commun., 2626–2627. See for example: Davis, F.A., Theddu, N., and Gaspari, P.M. (2009) Org. Lett., 11, 1647–1650. Arend, M., Westermann, B., and Risch, N. (1998) Angew. Chem. Int. Ed., 37, 1044–1070. Nicolaou, K.C. and Montagnon, T. (2008) Molecules That Changed the World, Wiley-VCH Verlag GmbH, Weinheim. Nakano, H., Kosemura, S., Suzuki, T., Hirose, K., Kaji, R., and Sakai, M. (2009) Tetrahedron Lett., 50, 2003–2005. (a) Saito, K. and Murakoshi, I. (1995) in Studies in Natural Products Chemistry, vol. 15 (ed. A.-U. Rahman), Elsevier Science, Amsterdam, pp. 519–549; (b) Ohmiya, S., Saito, K., and Murakoshi, I. (1995) in The Alkaloids, Chemistry and Pharmacology, vol. 47 (ed. G.A. Cordell), Academic Press Inc., San Diego, pp. 1–114; (c) Sekine, T., Saito, K., Minami, K., Arai, N., Suzuki, H., Koike, Y., and Murakoshi, I. (1993) Yakugaku Zasshi, 113, 53–62. Wanner, M.J. and Koomen, G.-J. (1996) J. Org. Chem., 61, 5581–5586. Numerous total syntheses of lupine alkaloids have been published; refer to general review articles [74]. For the single example of sparteine, see
77.
78.
79.
80. 81.
82.
recent ‘‘conventional’’ total syntheses: (a) Norcross, N.R., Melbardis, J.P., Ferris Solera, M., Sephton, M.A., Kilner, C., Zakharov, L.N., Astles, P.C., Warriner, S.L., and Blakemore, P.R. (2008) J. Org. Chem., 73, 7939–7951; (b) Hermet, J.-P.R., McGrath, M.J., O’Brien, P., Porter, D.W., and Gilday, J. (2004) Chem. Commun., 1830–1831; (c) Smith, B.T., Wendt, J.A., and Aub´e, J. (2002) Org. Lett., 4, 2577–2579; (d) See also, the beautiful biomimetic synthesis of sparteine by van Tamelen: van Tamelen, E.E. and Foltz, R.L. (1969) J. Am. Chem. Soc., 91, 7372–7377. Selected publications: (a) Mure, M. and Klinman, J.P. (1995) J. Am. Chem. Soc., 117, 8698–9702; (b) Mure, M. and Klinman, J.P. (1995) J. Am. Chem. Soc., 117, 8707–8718; (c) Lee, Y. and Sayre, L.M. (1995) J. Am. Chem. Soc., 117, 11823–11828. Largeron, M., Neudorffer, A., and Fleury, M.-B. (2003) Angew. Chem. Int. Ed., 42, 1026–1029. Murakoshi, I., Kidoguchi, E., Haginiwa, J., Ohmiya, S., Higashiyama, K., and Otomasu, H. (1981) Phytochemistry, 20, 1407–1409. Frigerio, F., Haseler, C.A., and Gallagher, T. (2010) Synlett, 729–730. (a) Wanner, M.J. and Koomen, G.-J. (1994) in Studies in Natural Products Chemistry, vol. 14 (ed. A.-U. Rahman), Elsevier, Amsterdam, pp. 731–768; (b) Wanner, M.J. and Koomen, G.J. (1994) Pure Appl. Chem., 66, 2239–2242; (c) Wanner, M.J. and Koomen, G.J. (1996) Pure Appl. Chem., 68, 2051–2056; (d) Wanner, M.J. and Koomen, G.J. (1997) J. Indian Chem. Soc., 74, 891–895; (e) Wanner, M.J. and Koomen, G.J. (1997) Invited lecture presented at the International Conference on Biodiversity and Bioresources: Conservation and Utilization, Phuket, Thailand, November 23–27, 1997 (see: http://old.iupac.org/symposia/proceedings/ phuket97/wanner.html, accessed 26 August 2010). Wanner, M.J. and Koomen, G.-J. (1995) J. Org. Chem., 60, 5634–5637.
57
58
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples 83. Gravel, E., Poupon, E., and
84.
85. 86.
87. 88.
89.
90.
91.
92. 93.
Hocquemiller, R. (2005) Org. Lett., 7, 2497–2499. Duan, J.-A., Williams, I.D., Che, C.-T., Zhou, R.-H., and Zhao, S.-X. (1999) Tetrahedron Lett., 40, 2593–2596. Salame, R., Gravel, E., and Poupon, E. (2009) Org. Lett., 11, 1891–1894. Putkonen, T., Tolvanen, A., Jokela, R., Caccamese, S., and Parrinello, N. (2003) Tetrahedron, 59, 8589–8595. Gravel, E. and Poupon, E. (2010) Nat. Prod. Rep., 27, 32–56. See for examples: (a) nitraramine: Wanner, M.J. and Koomen, G.-J. (1995) J. Org. Chem., 60, 5634–5637; (b) nitrarine: Wanner, M.J. and Koomen, G.-J. (1994) J. Org. Chem., 59, 7479–7484. See for examples: (a) biomimetic approach to nitrarine: Franc¸ois, D., Lallemand, M.-C., Selkti, M., Tomas, A., Kunesch, N., and Husson, H.-P. (1997) J. Org. Chem., 62, 8914–8916; (b) biomimetic synthesis of isonitramine and sibirine: Franc¸ois, D., Lallemand, M.C., Selkti, M., Tomas, A., Kunesch, N., and Husson, H.-P. (1998) Angew. Chem. Int. Ed., 37, 104–105. See for examples: (a) myrioneurinol: Pham, V.C., Jossang, A., S´evenet, T., Nguyen, V.H., and Bodo, B. (2007) Tetrahedron, 63, 11244–11249; (b) myrobotinol: Pham, V.C., Jossang, A., S´evenet, T., Nguyen, V.H., and Bodo, B. (2007) J. Org. Chem., 72, 9826–9829. See, among others, pioneering work: (a) Leonard, N.J. and Cook, A.G. (1959) J. Am. Chem. Soc., 81, 5627–5631; (b) Leonard, N.J. and Hauck, F.P. Jr. (1957) J. Am. Chem. Soc., 79, 5279–5292. Leonard, N.J. and Musker, K. (1959) J. Am. Chem. Soc., 81, 5631–5633. (a) Schildknecht, H., Krauss, D., Connert, J., Essenbreis, H., and Orfanides, N. (1975) Angew. Chem. Int. Ed. Engl., 14, 427; (b) Schildknecht, H., Berger, D., Krauss, D., Connert, J., Gehlhaus, H., and Essenbreis, H. (1976) J. Chem. Ecol., 2, 1–11.
94. Enders, D., Tiebes, J., De Kimpe, N.,
95.
96. 97.
98. 99.
100.
101. 102.
103.
104.
105.
106.
Keppens, M., Stevens, C., Smagghe, G., and Betz, O. (1993) J. Org. Chem., 58, 4881–4884. Poupon, E., Kunesch, N., and Husson, H.-P. (2000) Angew. Chem. Int. Ed., 39, 1493–1495. Lusebrink, I., Dettner, K., and Seifert, K. (2008) J. Nat. Prod., 71, 743–745. (a) Tanret, C. (1878) C. R. Acad. Sci., 86, 1270–1272; (b) Tanret, C. (1880) C. R. Acad. Sci., 90, 695–698. Hess, K. and Eichel, A. (1917) Ber. Dtsch. Chem. Ges., 50, 1192–1199. (a) Kuwata, S. (1960) Bull. Chem. Soc. Jpn., 33, 1668–1672; (b) Kuwata, S. (1960) Bull. Chem. Soc. Jpn., 33, 1672–1678. Gilman, R.E. and Marion, L. (1961) Bull. Soc. Chim. Fr., 1993–1995 and references cited therein. Drillien, G. and Viel, C. (1963) Bull. Soc. Chim. Fr., 2393–2340. (a) Determination of absolute configuration: Beyerman, H.C., Maat, L., van Veen, A., Zweistra, A., and von Philipsborn, W. (1965) Rec. Trav. Chim. Pays-Bas, 84, 1367–1379; (b) full chiroptical properties: Craig, J.C., Lee, S.Y.C., and Roy, S.K. (1978) J. Org. Chem., 43, 347–349. (a) Gupta, R.N. and Spenser, I.D. (1969) Can. J. Chem., 47, 445–447; (b) Hemscheidt, T. and Spenser, I.D. (1990) J. Am. Chem. Soc., 112, 6360–6363 and references cited therein. (a) Anet, E.F.L., Hughes, G.K., and Ritchie, E. (1949) Nature, 164, 501; (b) Anet, E.F.L., Hughes, G.K., and Ritchie, E. (1950) Aust. J. Sci. Res. A, 3, 336–341; (c) Sch¨opf, C. (1949) Angew. Chem., 61, 31 See for examples: (a) Wisse, J.H., De Klonia, H., and Visser, B.J. (1964) Rec. Trav. Chim. Pays-Bas, 83, 1265–1272; (b) Quick, J. and Oterson, R. (1976) Synthesis, 745–756. For recent asymmetric syntheses of pelletierine, see: (a) Takahata, H., Kubota, M., Takahashi, S., and Momose, T. (1996) Tetrahedron: Asymmetry, 7, 3047–3054; (b) Turcaud, S., Martens, T., Sierecki, E., P´erard-Viret,
References
107.
108.
109. 110. 111.
112.
113.
114. 115.
116. 117.
118.
119.
J., and Royer, J. (2005) Tetrahedron Lett., 46, 5131–5134; (c) Carlson, E.C., Rathbone, L.K., Yang, H., Collett, N.D., and Carter, R.G. (2008) J. Org. Chem., 73, 5155–5158. Quick, J. and Meltz, C. (1979) J. Org. Chem., 44, 573–578 and references cited therein. Cope, A.C., Dryden, H.L., Overberger, C.G., and D’Addieco, A.A. (1951) J. Am. Chem. Soc., 73, 3416–3418 and references cited therein. Bates, R.W. and Sa-Ei, K. (2002) Tetrahedron, 58, 5957–5978. Felpin, F.-X. and Lebreton, J. (2004) Tetrahedron, 60, 10127–10153. See for example: Krishnan, S., Bagdanoff, J.T., Ebner, D.C., Ramtohul, Y.K., Tambar, U.K., and Stoltz, B.M. (2008) J. Am. Chem. Soc, 130, 13745–13754 and references cited therein for recent syntheses of Lobelia and Sedum alkaloids. Review articles: (a) Ma, X. and Gang, D.R. (2004) Nat. Prod. Rep., 21, 752–772; (b) Kobayashi, J. and Morita, H. (2005) in The Alkaloids, Chemistry and Biology, vol. 61 (ed. G.A. Cordell), Elsevier, San Diego, pp. 1–57; (c) Hirasawa, Y., Kobayashi, J., and Morita, H. (2009) Heterocycles, 77, 679–729. Hemscheidt, T. and Spenser, I.D. (1996) J. Am. Chem. Soc., 118, 1799–1800. See also the impressive number of discoveries from the Spenser group in the field of biosynthesis of alkaloids. Conroy, H. (1960) Tetrahedron Lett., 1, 34–37. Wenkert, E., Chauncy, B., Dave, K.G., Jeffcoat, A.R., Schell, F.M., and Schenk, H.P. (1973) J. Am. Chem. Soc., 95, 8427–8436. Scott, W.L. and Evans, D.A. (1972) J. Am. Chem. Soc., 94, 4779–4780. Heathcock, C.H., Kleinman, E.F., and Binkley, E.S. (1982) J. Am. Chem. Soc., 104, 1054–1068. Schumann, D., M¨uller, H.-J., and Naumann, A. (1982) Liebigs. Ann. Chem., 2057–2061. For recent highly elegant first total synthesis of some complex Lycopodium alkaloids published from 2009, see:
120.
121. 122.
123.
124.
125.
(a) (−)-deoxyserratinine: Yang, Y.-R., Lai, Z.-W., Shen, L., Huang, J.-Z., Wu, X.-D., Yin, J.-L., and Wei, K. (2010) Org. Lett., 12, 3430–3433; (b) (+)-sieboldine A: Canham, S.M., France, D.J., and Overman, L.E. (2010) J. Am. Chem. Soc., 132, 7876–7877; (c) (+)-complanadine A: Yuan, C., Chang, C.-T., Axelrod, A., and Siegel, D. (2010) J. Am. Chem. Soc., 132, 5924–5925; (d) (+)-complanadine A: Fischer, D.F. and Sarpong, R. (2010) J. Am. Chem. Soc., 132, 5926–5927; (e) (±)-lycoposerramine R: Bisai, V. and Sarpong, R. (2010) Org. Lett., 12, 2551–2553; (f) (+)-serratezomine: Chandra, A., Pigza, J.A., Han, J.-S., Mutnick, D., and Johnston, J.N. (2009) J. Am. Chem. Soc., 131, 3470–3471; (g) (+)-lyconadin A: Nishimura, T., Unni, A.K., Yokoshima, S. and Fukuyama, T. (2011) J. Am. Chem. Soc., 133, 418–419. Morita, H., Arisaka, M., Yoshida, N., and Kobayashi, J. (2000) J. Org. Chem., 65, 6241–6245. Morita, H. and Kobayashi, J. (2002) J. Org. Chem., 67, 5378–5381. Katakawa, K., Kitajima, M., Aimi, N., Seki, H., Yamaguchi, K., Furihata, K., Harayama, T., and Takayama, H. (2005) J. Org. Chem., 70, 658–663. (a) Takayama, H., Katakawa, K., Kitajima, M., Yamaguchi, K., and Aimi, N. (2002) Tetrahedron Lett., 43, 8307–8311; (b) Nakayama, A., Kogure, N., Kitajima, M., and Takayama, H. (2009) Org. Lett., 11, 5554–5557; (c) see also an interesting cascade of reaction in the first total synthesis of (+)-lycoflexine: Ramharter, J., Weinstabl, H., Mulzer, J. (2010) J. Am. Chem. Soc., 132, 14338–14339. Yang, Y.-R., Shen, L., Wei, K., and Zhao, Q.-S. (2010) J. Org. Chem., 75, 1317–1320. Morita, H., Hirasawa, Y., Shinzato, T., and Kobayashi, J. (2004) Tetrahedron, 60, 7015–7023. From the same group, see also the chemical correlation between lyconadins B and F: Ishiuchi, K., Kubota, T., Ishiyama, H., Hayashi, S., Shibata, T., Kobayashi, J. (2011) Tetrahedron Lett., 52, 289–292.
59
60
1 Biomimetic Synthesis of Ornithine/Arginine and Lysine-Derived Alkaloids: Selected Examples 126. Snider, B.B. and Grabowski, J.F. (2007)
128. Liau, B.B. and Shair, M.D.
J. Org. Chem., 72, 1039–1042. 127. Gerard, R.V., MacLean, D.B., Fagianni, R., and Lock, C.J. (1986) Can. J. Chem., 64, 943–949.
(2010) J. Am. Chem. Soc., 132, 9594–9595. 129. Schumann, D. and Naumann, A. (1983) Liebigs Ann. Chem., 220–225.
61
2 Biomimetic Synthesis of Alkaloids Derived from Tyrosine: The Case of FR-901483 and TAN-1251 Compounds Huan Liang and Marco A. Ciufolini
2.1 Introduction
Tyrosine is the biosynthetic precursor of such a large number of nitrogenous substances [1] that a general review of biomimetic syntheses of natural products derived from that amino acid would occupy considerable space. On the other hand, most of the compounds in question are plant alkaloids that have already been the subject of numerous reviews [2]. This chapter eschews any discussion of chemical syntheses of plant-derived agents, focusing instead on a small group of fungal metabolites discovered during the 1990s. Much of the impetus for the work described herein derived from a noteworthy paper that appeared in 1996 in the Journal of Antibiotics. In this publication, scientists at the Fujisawa (now Astellas) pharmaceutical company described a structurally novel natural product, FR-901483 (1, Figure 2.1) that displayed considerable immunosuppressive activity [3, 4]. The new substance was isolated from a culture of a Cladobotryum species and its structure and relative configuration were ascertained by X-ray diffractometry. Formula 1 depicts the molecule with its actual absolute configuration, which, however, was determined only later through chemical synthesis. The bioactivity of FR-901483 is intimately associated with the presence of the phosphate ester, which unfortunately is labile. Rapid dephosphorylation of 1 ensues even upon introduction into cellular cultures, resulting in formation of the corresponding alcohol, 2, which is devoid of activity. The new immunosuppressant appeared to be structurally related to the so-called TAN-1251 family of compounds (3–6, Figure 2.2), a group of substances with noteworthy antimuscarinic activity produced by a Penicillium species [5]. One can easily recognize that FR and TAN substances are formal dimers of tyrosine (Scheme 2.1). Moreover, the two types of compounds may share a common biosynthetic precursor in the form of aldehyde 7. Indeed, TAN compounds could emerge upon cyclization of 7 to iminium ion 9. By contrast, the framework of FR-901483 would result upon aldol cyclization to 8. An interesting question pertains to the biosynthetic origin of 7 (Scheme 2.2). This aldehyde could arise through oxidative cyclization of a tyrosinyl tyrosine, 11, Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
2 Biomimetic Synthesis of Alkaloids Derived from Tyrosine
62
Figure 2.1 Structure of FR-901483 (1) and of the corresponding diol (2, inactive).
OMe
HO
N
H NHMe
Z O 1 Z = P(O)(OH)2 FR-901483 2Z=H
NMe N
O
NMe
H
N
O
H
HO O
O TAN-1251A (3)
TAN-1251B (4) H NMe
N
O
NMe
H
N
O
O
O
TAN-1251C (5) Figure 2.2
H
TAN-1251D (6)
Structures of TAN-1251 compounds. OR
OH
H NMe N
RO
HO
N
HOOC
NHMe
NH2
O
NHMe
H Z O
FR-901483 architecture
H
COOH TAN1251 architecture
HO
OR H CHO HO
O
N
NHMe 8
Scheme 2.1
cyclization
H NHMe
aldol
N
RO O
H
NMe
iminium ion formation
7
N
RO O
H 9
FR-901483 and TAN-1251 as dimers of tyrosine.
to spirolactam 10, followed by appropriate redox modifications. The conversion of 11 into 10 is envisioned to involve an initial oxidative activation of the phenol to furnish a reactive electrophilic intermediate, naively represented in Scheme 2.3 as 13. The nitrogen atom of the amide subsequently intercepts 13, thereby expressing nucleophilic reactivity. The overall process may be described as the
2.2 Biomimetic Total Syntheses of FR-901483 and TAN-1251 Compounds OR
OR H CHO N
RO
O
O
NHMe HOOC
H
HOOC
N
NHMe
N H
NH-R O
7
10
O
11
HO
2 x HOOC
NH2
HO (L)-tyrosine
Scheme 2.2
Presumed biosynthetic origin of aldehyde 8.
O
O R-HN
NH–P
R-HN
NH–P
−2e−
HO Scheme 2.3
12
N
−H+
−H+
O
13
O
R
O
NH–P 14
Oxidative amidation of phenols.
‘‘oxidative amidation’’ of a phenol. One may anticipate that if this transformation could be duplicated in the laboratory, the synthesis of 1–6 would be greatly simplified.
2.2 Biomimetic Total Syntheses of FR-901483 and TAN-1251 Compounds
The architectural novelty of 1–6 and their interesting biological activity have inspired a great deal of synthetic research, and various strategies, some biomimetic, some not, have been explored in order to construct the framework of these natural products. In the context of this chapter, a synthesis is regarded as biomimetic if it proceeds through an early intermediate, that is, a tyrosinyl tyrosine or an analog thereof (Scheme 2.4). Of course, in no way does this detract from the brilliancy and significance of alternative approaches explored in connection with total syntheses [6, 7] and synthetic studies [8, 9] of the foregoing natural products. The criterion just stated qualifies as biomimetic the Snider, Sorensen, and Ciufolini syntheses of 1, as well as the Snider, Ciufolini, and Honda syntheses of TAN-1251 substances. We shall now review these efforts in detail.
63
64
2 Biomimetic Synthesis of Alkaloids Derived from Tyrosine Y Building Blocks HO
P
NMe
N H
1 and / or 2-6
Y = O; H,H Z = COOH, COOR, CH2OH
tyrosinyl tyrosine (or analog thereof)
Scheme 2.4
OR
Z
Biomimetic synthesis: one that proceeds via a tyrosinyl tyrosine intermediate.
2.2.1 Snider Synthesis of FR-901483
The first total synthesis of FR-901483 was achieved in 1999 by Snider [10]. Key strategic aspects of this endeavor were the early construction of tyrosinyl tyrosine derivative 15 and regio- and stereoselective aldol cyclization thereof as an avenue to the tricyclic core of 1 (Scheme 2.5). PAN O OHC
PAN O
BOC
N
[base]
N
HO
Me O
Scheme 2.5
N
FR-901483 (1) Me
O
15
BOC
N
H 16
PAN = para -anisyl
Key steps in Snider’s synthesis of FR-901483.
The assembly of 21 avoided the oxidative cyclization of a phenolic amide, because at that time no technology was available to effect such a transformation. Accordingly, its preparation relied instead on an ingenious 1,3-dipolar cycloaddition of tyrosine-derived nitrone 19 [11] to ethyl acrylate (Scheme 2.6). This step proceeded OMe OH MeOOC
NH2 1. anisaldehyde,
MeOOC
O
NH O
O
2. m CPBA
17
MeO
COOMe
3. NH2OH•HCl 73 %
EtOH
MeO
N 18
O
O 19
O MeO PAN
PAN
O CO2Et
O
toluene 77 %
H H N
COOMe O
MeOOC
COOEt
MeOOC
O O
21
EtOOC
N O
O (6 : 1)
H
20
Scheme 2.6
N
O
22 O
PAN = para -anisyl
Synthesis of intermediate 21 by 1,3-dipolar cycloaddition chemistry.
COOEt
2.2 Biomimetic Total Syntheses of FR-901483 and TAN-1251 Compounds
65
diastereoselectively to furnish a 6 : 1 mixture of adducts 21 (major) and 22 (minor). The stereochemical outcome of this step was rationalized by assuming that nitrone 19 reacts preferentially from conformation 20, which minimizes allylic-type interactions and promotes the approach of the methyl acrylate molecule from the Re face of the imino linkage. Furthermore, the reaction proceeds with exo topology due to the absence of secondary orbital interactions. Adducts 21 and 22 were not separated at this stage. Rather, hydrogenolysis of the N–O bond triggered formation of spirolactam 23, which was advanced to 15 in a conventional fashion (Scheme 2.7). The crucial aldol cyclization of the latter presented several complex issues. The success of this step depends on: the sequential occurrence of a chemoselective enolization of the cyclohexanone without enolization of the aldehyde; regioselective formation of enolate 26 (Z = H, H, or O; P = protecting group); and diastereoselective addition thereof to the Si-face of the aldehyde to yield 28 (Scheme 2.8). If enolization of the cyclohexanone segment 25 were indeed kinetically faster than that of the aldehyde, then the action of a gentle base could promote reversible, PAN O
PAN NO
MeOOC O
COOEt
45 psi H2, Pd/C MeOOC AcOH, 86 %
3. Pd/C, H2, Boc2O 4. NaH, MeI, 83 %
O
[?]
BOC
N
HO
N Me
Me O
O
15
Ar Z
O
16
(N) 29
O
axial H Ar N
Z
Ar
N
base
N
(N) base
(N) OHC
N
25 Ar
30
Ar equatorial N H CHO
(N)
O
Z (N)
Z Z
HO
26
O
Ar
Z
N
(N)
HO
O
31 undesired
Aspects of the aldol cyclization of 25.
N
(N) 28
desired
Z (N)
N
O
O
27
Z
O
Scheme 2.8
BOC
N
N
Synthesis of aldol substrate 15.
Ar
O
24
PAN O
PAN O OHC
BOC N Me
N
O
23 (85 % ee) rac. before nitrone cycloaddn.
2. Boc2O, Et3N, 95 % 3. Dess-Martin
OHC
MeOOC
O
1. LiBH4 then HCl, HOAc/H2O, 97 %
Scheme 2.7
OH
O
21 (+ 22)
O
N
PAN O
1. TsCl, Et3N 2. NaN3, 78 %
Ar H OH
28 O
66
2 Biomimetic Synthesis of Alkaloids Derived from Tyrosine
non-regioselective formation of the ketone enolate.1) For a fixed (S)-configuration at the α-position of the aldehyde, aldol cyclization from the incorrect regioisomer of the cyclohexanone enolate, 29, forces the 4-methoxybenzyl substituent into an axial orientation in transition state 30. This generates severe compression against a methylene group of the enolate ring (dashed semicircles). No such problems exist in regioisomeric transition state 27, where the substituent in question is pseudo-equatorial in the developing ring. This should favor selective formation of the desired 28. A potential complication is that for Z = O, the newly formed six-membered ring is an N-acylpiperidine. Substituents at the α-position of the nitrogen atom in such structures strongly prefer an axial orientation [12]. The incorrect aldol regioisomer might then become favored under thermodynamic conditions, and possibly even under kinetic conditions, if the transition state for the aldol cyclization were product-like. No such complication would affect aldol substrates in which Z is a pair of H atoms, and, indeed, the later Sorensen synthesis implemented precisely this second principle. Conversely, the question of diastereoselectivity in the aldol step is significantly more complex. Some aspects of this issue will be addressed later. Snider found that the aldol cyclization of 15 and related substrates occurred regioselectively without epimerization of the aldehyde α-stereocenter and with useful levels of stereocontrol (Scheme 2.9). Model substrate 32 reacted with NaOMe in MeOH at room temperature to afford a mixture of three diastereomers in which compound 33, which possesses the correct relative configuration of all ring stereocenters, was the major component. An interesting solvent effect on stereoselectivity was also unveiled, in that the action of t-BuOK in toluene afforded 34 as the major isomer and 36 as the minor one. The predominant formation PAN O
PAN O OHC
HO
N 32
HO
N
base
O
33
O
NaOMe, MeOH, rt
not detected
PAN O
O
N
N
BOC
t-BuOK
Me
t-BuOH rt
HO
35
36
O
13 %
not detected
not detected
22 %
PAN O HO
N 16
36 %
Scheme 2.9
N
HO
O
PAN O BOC
N
N
Me O
15
N
23 %
BOC
N
PAN O
70 % PAN O
N
PAN O HO
34
O
51 %
t-BuOK, toluene, rt
OHC
PAN O
HO
BOC
N
N
Me O
37 16 %
Me O
38 5%
Crucial aldol step in the Snider synthesis of FR-901483.
1) The lactam unit in 25 is much less C–H
acidic than either the ketone or the aldehyde, and it was anticipated not to interfere
with the operations leading to the tricyclic aldol intermediate.
2.2 Biomimetic Total Syntheses of FR-901483 and TAN-1251 Compounds
67
of 34 and 36 in toluene was attributed to stabilization of a transient alkoxide by chelation with the ketone, while the selective formation of 33 in MeOH was ascribed to strong solvation of the alkoxide, presumably by hydrogen bonding to the protic solvent. In like manner, cyclization of FR-901483 precursor 15 with t-BuOK in t-BuOH afforded a mixture of 16 (36%), 37 (16%), and 38 (5%). Improved diastereoselectivity was observed in the aldol cyclization of 15 with t-BuOK/t-BuOH relative to MeONa/MeOH. Unfortunately, 16 and 38 were inseparable and were employed in subsequent steps as a mixture. The resistance of the aldehyde to epimerization is consistent with earlier observations by Garner [13], and especially subsequent ones by Myers [14], who exploited this property in several brilliant total syntheses [15]. Release of the Boc group and LAH reduction afforded 39 (plus the corresponding product from 38, Scheme 2.10). The virtually complete diastereoselectivity achieved during the reduction of the keto carbonyl is attributable to preferential approach of the reducing agent from the convex face of the molecule. Protection of the amino group as a Cbz derivative permitted the removal of the isomeric material derived from 39. The reactivity of the OH groups in 39 differs significantly, permitting a highly selective conversion into 40. Treatment of the latter with CsOAc/18-crown-6 resulted primarily in SN 2 displacement of the nosyl group by acetate ion (70% yield). An olefinic by-product (20% yield) arising through a competitive E2 reaction was also obtained. Release of the acetate and formation of dibenzyl phosphate derivative 42 set the stage for a final hydrogenolysis of all benzyl groups. The target 1 was best isolated and characterized as the mono-hydrochloride, which was most conveniently prepared by hydrogenolysis of the hydrochloride salt of O-deprotected 42. The Snider synthesis of (−)-FR-901483 thus proceeded in 2% overall yield from O-methyltyrosine methyl ester (17) over 26 steps. PAN
PAN O HO
HO
BOC
N
N
1. LAH
N Me 2. CbzCl, 16 (+ 38)
O
HO
Et3N, 52 %
H
H OH
Scheme 2.10
41
TESO
1. CsOAc,
CBZ 18-crown-6, 70 % N Me 2. K CO , 2 3
N
NsO
MeOH
40
H
PAN
PAN Cbz tetrazole N (BnO)3PN(i -Pr)2 Me
N
Et3N, 99 %
39
PAN TESO
PAN Cbz 1. NsCl, Et3N N Me 2. TESOTf,
t-BuOOH, 93 %
TESO
N
H (BnO)2 P O O
Cbz 1. TBAF, 96 % N Me 2. HCl, then H2/Pd, MeOH 92 %
42
HO
N
H
(HO)2 P O O
NHMe • HCl
FR-901483 (1)
Completion of the synthesis of FR-901483.
2.2.2 Snider Synthesis of TAN-1251 Substances
Snider also takes the credit for achieving the first total synthesis of TAN-1251 compounds, thereby ascertaining their absolute configuration [16]. The opening
2 Biomimetic Synthesis of Alkaloids Derived from Tyrosine
68
moves of this effort retrace substantially the route to the key intermediate for FR-901483 (Scheme 2.6). Thus, tyrosine benzyl ether 43 was converted into α-hydroxyl amino acid 44 (Scheme 2.11). Condensation with 1,4-cyclohexanedione mono 1,4-butyleneketal and 1,3-dipolar cycloaddition of the ensuing nitrone to ethyl acrylate afforded 45. In accord with previous observations, this material was also formed as a 6 : 1 mixture of diastereomers. Without separation, these were advanced to 47. Nucleophilic substitution of tosylate with azide ion in 47 delivered a 6 : 1 mixture of diastereomeric azides, which contrary to early intermediates were readily separated. The synthesis thus continued with the major azide diastereomer, which possesses the correct configuration. Straightforward manipulations ultimately produced amino-alcohol 49.
MeOOC
NH2 1. anisaldehyde,
MeOOC
H N
2. m CPBA 3. NH2OH • HCl 53 % BnO
43
BnO
OBn (6 : 1)
O
OH 1.
O EtOH
O
2. ethyl acrylate, toluene 68 %
44
NO
MeOOC O
45
O O
HO 1. prenyl Br 86 %
O
2. 5 % HOAc/ CH2Cl2, 68 %
N
MeOOC
OH
O O
Scheme 2.11
Ar
(6 : 1)
(6 : 1) 1. 45 psi H2, Pd/C, HOAc
O MeOOC
N
OTs
O
2. TsCl, Et3N
COOEt
1. NaN3, 79 % 2. LAH
HO
N
NH-R
O
3. HCOOAc
O O
46
47
LAH, 58 %
48 R = CHO 49 R = Me
Snider intermediate 49 for the synthesis of TAN-1251 compounds.
Exposure of 49 to DMSO/TFAA/Et3 N induced oxidation of the primary alcohol to the aldehyde and concomitant trifluoroacetylation of the amino group (Scheme 2.12). The free amino-aldehyde obtained upon release of the N-COCF3 unit (methanolic K2 CO3 ) instantly cyclized to 51, which upon acidic hydrolysis of the ketal furnished TAN-1251C (5). Ar HO
Ar
N
NHMe
O
TFAA DMSO Et3N
O
49
OHC
COCF3 NMe
N
N K2CO3
O aq. MeOH
O O
X
50 0.1 M HCl aq acetone 75 %
Scheme 2.12
N
51 X = O(CH2)4O, 60 % 5
X = O, TAN-1251C
Snider synthesis of TAN-1251C.
It is likely that TAN-1251C is the biosynthetic forerunner of the other members of the family. For instance, TAN-1251D (6) probably ensues upon reduction of an iminium ion formed by protonation of TAN-1251C (Scheme 2.13). Snider utilized
2.2 Biomimetic Total Syntheses of FR-901483 and TAN-1251 Compounds
X
X
H+
H
N
Me N
N N
O O
H
H+ H
N H
Me N
Ar O 51
55 X = O(CH2)4 O kinetic product more energetic
Ar 52 X = O(CH2)4O thermodynamic product less energetic
[H]
[H] H
H N
N N
O
Scheme 2.13
N
O
X 0.1 M HCl aq acetone 79 %
69
X 56 X = O(CH2)4O 6 X = O, TAN-1251D
0.1 M HCl aq acetone 83 %
53 X = O(CH2)4O 54 X = O, epi -TAN-1251D
Solvent effects in the reduction of TAN-1251C to TAN-1251D.
precisely such a transformation to convert TAN-1251C into TAN-1251D, and in the process he uncovered yet another remarkable solvent effect. Thus, reduction of 51 with NaBH(OAc)3 in HOAc afforded 90% of a 1 : 9 mixture of ketals 56 and 53. Acidic hydrolysis of the major product 53 surrendered epi-TAN-1251D (54) (83%). A reversal of selectivity occurred upon reduction of 51 with the more active reductant, NaBH3 CN, in the more polar solvent, MeOH (dropwise addition of HOAc), leading to a 2 : 1 mixture of 56 and 53. An amelioration of selectivity in the desired sense was achieved by operating in even more polar solvents. Thus, the ratio of 56 : 53 increased to 6 : 1 when the reaction was carried out with NaBH3 CN in CF3 CH2 OH, and it peaked at >25 : 1 by switching to the even more polar (CF3 )2 CHOH. Acidic hydrolysis of ketal 56 afforded TAN-1251D (6). Snider offered the following rationale for the above phenomena. Stereochemical control in the reduction of 51 occurs during the protonation step, which determines the configuration of the benzyl-type substituent. Protonation from the convex face of the molecule is kinetically favored on grounds of reduced steric interaction with the approaching Brønsted acid. Protonation in this sense would form 55, wherein the benzyl substituent occupies an equatorial position. However, MM2 calculations revealed that 55 contains 3 kcal mol−1 more steric energy than the axial diastereomer 52, probably because of a serious non-bonded interaction between the equatorial benzyl group and the spiro cyclohexane ring. Evidently, the energy penalty incurred in repositioning the benzyl substituent at an axial position is much smaller than the extent of the foregoing steric interaction.
70
2 Biomimetic Synthesis of Alkaloids Derived from Tyrosine
Therefore, in media where the reduction of the iminium ion is slow and protonation is reversible, equilibration of iminium diastereomers can occur and the major iminium species present in the medium will be 52. Reduction of the latter will then lead to an epimeric series of products. This is what happens with the weak reductant, NaBH(OAc)3 in MeOH, a polar solvent of modest protonic acidity. In MeOH, the stronger reducing agent NaBH3 CN evidently intercepts the transient iminium species at a rate comparable to that of equilibration; therefore, reduction of the kinetic iminium ion 55 becomes the major process. As the polarity and Brønsted acidity of the solvent increase, the ionic iminium species is stabilized relative to the electrostatically neutral enamine, and the equilibration of iminium diastereomers is retarded. Therefore, the ratio of diastereomers 56 : 53 steadily improves upon progressing from MeOH to CF3 CH2 OH and to (CF3 )2 CHOH. TAN-1251A was reached starting with DDQ (2,3-dichloro-5,6-dicyano-1,4benzoquinone) oxidation of 51 in dry CH2 Cl2 (Scheme 2.14). This furnished eniminium ion 57, which upon reduction with NaBH3 CN in acidic MeOH afforded 58. Hydrolysis of the ketal delivered fully synthetic 3.
N N
O
N DDQ CH2Cl2
O O 51
Scheme 2.14
O
N O O
57
NaBH3CN MeOH HOAc
0.1 M HCl aq acetone 89 %
N O
N
X 58 X = O(CH2)4O, 79 % 3 X = O, TAN-1251A
Snider synthesis of TAN-1251A.
The synthesis of TAN-1251B required a solution to the challenging problem of regio- and diastereoselective hydroxylation of 3 in the presence of other readily oxidizable functionality (amines, double bonds). The action of mCPBA on the trimethylsilyl enol ether derivative 59 produced the N-oxide of the methylamino unit. The hydroxylation of the Na enolate of 3 (NaHMDS) with the Davis 10-(camphorsulfonyl)oxaziridine [17] produced inseparable mixtures of 60 and 61 in 40–50% yield. While no oxidation of the double bonds or the amines was detected, the reaction was taking place from the incorrect face of the enolate. Finally, it transpired that osmylation of enol ether derivatives 59 was moderately selective in the desired sense (Scheme 2.15). Thus, 3 was converted into a mixture of regioisomeric silyl enol ethers. Reaction of these with OsO4 , and NMO afforded a 2 : 1 : 8 : 4 mixture of 60, 61, TAN-1251B (4), and 62, in 39% yield, plus a mixture of triols resulting from hydroxylation of both the enol ether and the prenyl group (23% yield), plus recovered 3 (9%). HPLC separation of this complex mixture provided pure 4. Attempts to achieve a more regioselective enolization of 3 with chiral bases were fruitless.
2.3 Oxidative Amidation of Phenols
N
N N
O
OsO4
N
O
NMO
TMSO
O
59
TAN-1251A (3) N
Ar
N
Ar
N
N
Ar
N
H
N
Ar
N
N
H
HO
HO
H
H
60 O :
O 2
61
O
O
TAN-1251B (4)
HO
62
OH 1
:
8
:
4
+ products of prenyl dihydroxylation + recovered 3
Scheme 2.15
Snider synthesis of TAN-1251B.
2.3 Oxidative Amidation of Phenols
The promise of a fully biomimetic synthesis of 1–6 awaited the development of methodology for the oxidative amidations of phenols according to the format of Scheme 2.3. This precise transformation was unknown until the late-1990s, and, indeed, it was believed not to be feasible. This lore was rooted in an important observation recorded in 1987 by Kita, who studied the oxidation of amides 63 with PhI(OAc)2 (DIB), or PhI(OCOCF3 )2 (‘‘PIFA’’), intending perhaps to create spirolactams 64 [18]. However, the reaction produced only lactone 67, arguably through capture of electrophilic intermediate 65 by the carbonyl oxygen of the amide, followed by hydrolysis of the transient 66 upon aqueous workup (Scheme 2.16). Z R
R-HN
N
O
O
O DIB 64
Scheme 2.16
O
DIB HO
63
O
65
R NH
O O 66 Z = N-R 67 Z = O
H2O
Kita oxidation of phenolic amides leading to spirolactones.
Technology for oxidative amidation of phenols finally emerged in 1998 thanks to research carried out in our own laboratories [19]. First of all, we perceived a mechanistic analogy between the formation of lactone 67 upon oxidative activation of the phenol and the results obtained by Knapp during a study of the iodoamidation of olefins [20]. Knapp determined that the reaction of 68 with I2 results in formation of 69 instead of the desired 71 (Scheme 2.17). This suggests that resonance
71
72
2 Biomimetic Synthesis of Alkaloids Derived from Tyrosine R-HN
O
R-N
"I+"
O I
68
69 N–R
R NH
O
O O
O
65
66
Scheme 2.17 Analogy between the Knapp iodocyclization of olefinic amides and the Kita lactone formation.
interactions between the N atom and the carbonyl system in an amide induce accumulation of electronic density on the oxygen atom, which therefore becomes nucleophilic at the expense of the N atom. The same effect accounts for the formation of 66 upon cyclization of presumed intermediate 65. But when Knapp engaged imino ethers 70 in such iodocyclization reactions, the desired 71 was obtained in good yield (Scheme 2.18). Evidently, resonance interactions in 70 now promote accumulation of electronic density on the nitrogen atom, which therefore can express nucleophilic reactivity.
R'3SiO
"I+"
N–R
O
R N I
70
Scheme 2.18
71
Knapp iodolactamization reaction of olefinic imino ethers.
We surmised that an iminoether analog of the amide functionality should enable the desired spirolactamization reaction. In the end, oxazolines emerged as effective substrates for the target transformation [21]. The starting 74 were best prepared by Vorbr¨uggen condensation of a phenolic carboxylic acid with a suitable 1,2-aminoalcohol [22]. An advantage of this technique is that it required no protection of the phenol. Alternative procedures were less satisfactory in the present case [23]. Exposure of 74 to the action of DIB in fluoroalcohol solvents [CF3 CH2 OH (TFE) or (CF3 )2 CHOH (HFIP)] according to Kita [24] afforded the long sought 77 [25] (Scheme 2.19). Some oxazoline-derived products of oxidative amidation were observed to undergo spontaneous Michael cyclization to morpholine derivatives more or R
HOOC
O
R H2N
OH 73
N
PPh3, CCl4 Et3N, pyr.
HO
HO
72 Scheme 2.19
R PhI(OAc)2
R
O N
O
HO
N
O
R N
CF3CH2OH
74
O
75
O
Oxidative spirocyclization of phenolic oxazolines.
76
O
77
2.3 Oxidative Amidation of Phenols
73
less rapidly, depending on structural details. In all cases, cyclization was highly diastereoselective. For instance, compound 78 gave 79 exclusively (Scheme 2.20; structure ascertained by X-ray crystallography). The stereoselective formation of 79 is attributable to the strong preference for the axial orientation on the part of alkyl substituents flanking the N atom in N-acyl piperidines and related six-membered heterocycles [12]. O O HO
Bn O
N
N H H O
78
Scheme 2.20
O
79
Stereoselective cyclization of products of oxidative amidation.
Morpholine formation is useful in particular circumstances because it induces a stereocontrolled desymmetrization of the ‘‘locally symmetrical’’ dienone segment of the primary products, securing a specific configuration of the now stereogenic spiro center. In other instances, the proclivity to cyclize is problematic and must be suppressed. It is then convenient to acetylate the OH group prior to purification (cf. 83→84, Scheme 2.21). Overall yields of acetates 84 from 82 are typically around 40–45%. Such moderate yields must be weighed against the fact that the reaction rapidly converts inexpensive amino acid-derived substances into valuable enantiopure intermediates. OH
COOH
+ NH2 HO 80
NH-Ts
O
PPh3
N
NH-Ts
CCl4 73 %
81
DIB
R O
CF3CH2OH
HO
82
R N-Ts
O Ac2O, py, 41 %
Scheme 2.21
O N
83 R = H 84 R = Ac
Formation of spirolactam 84.
A comment is in order regarding the choice of an N-sulfonyl protecting group for 82. Experiments carried out with similar substrates that incorporated carbonyl-type N-protection furnished complex mixtures containing only some of the desired spirolactams. At least some of the by-products obtained from these reactions appeared to arise through interception of reactive intermediate 85 by the carbonyl unit (Scheme 2.22, pathway b). Iminocarbamate 86 thus formed could subsequently yield various secondary products. A sulfonyl protecting group is insufficiently nucleophilic to interact with the electrophilic moiety of 85, maximizing formation of the desired spirolactam. The chemistry of Scheme 2.19 constitutes what may be termed a ‘‘first generation’’ solution for the oxidative amidation of phenols. The methodology was later extended
74
2 Biomimetic Synthesis of Alkaloids Derived from Tyrosine
Ar
spirolactam
O N a +
a pathway
b
OR
O
b
O
N byproducts
N
pathway O
O Scheme 2.22
OR
H N
O
85
Ar
86
By-products arising from substrates incorporating a carbamoyl group.
to the oxidative cyclization of phenolic sulfonamides [26] and to a bimolecular variant of the reaction, wherein a nitrile captures the presumed intermediate 90 in a Ritter mode [27]. These processes are outlined in Scheme 2.23; however, they are outside the scope of the present chapter. R-SO2
N H
R-SO2
N
DIB
[ref 24]
60-97 %
HO
O
87 R'
R' DIB
[ref 25]
O
O 90
89 Scheme 2.23
NHAc R'
66-86 %
TFA MeCN
HO
88
91
Alternative modes of oxidative amidation of phenols.
A noteworthy variant of the above methodology achieves the oxidative cyclization of phenolic secondary amines. This chemistry was developed by Sorensen as the centerpiece of his biomimetic synthesis of FR-901483 [28]. A key step in this endeavor was the conversion of 92 into 93 upon exposure to DIB (Scheme 2.24). Notice that a sulfonamido protecting group was present on the spectator amino functionality in 92. We presume that this choice was dictated by the difficulties adumbrated in Scheme 2.22. OMe Me MeOOC
N H
N Ns
OMe DIB (CF3)2CHOH 51 % at 70 % conversion
Ns = 4-NO2C6H4SO2
HO
MeOOC
N
Me N Ns
O
92 Scheme 2.24 Oxidative cyclization of a phenolic amine in the Sorensen synthesis of FR-901483.
93
2.3 Oxidative Amidation of Phenols
A similar reaction described by Honda [29] accomplishes the DIB-mediated oxidative cyclization of 94→95 (Scheme 2.25). This transformation was central to a biomimetic synthesis of TAN-1251 substances. O
O NMe
HN
DIB (CF3)2CHOH
H
NMe N
H
69 %
94
HO Scheme 2.25
95
O
Honda oxidative cyclization of phenolic amines.
We had an opportunity to examine the behavior of primary amines under Sorensen–Honda conditions, but we found them to be exceedingly poor substrates for the reaction [30]. Thus, cyclization of 96 furnished 97, as a component of a complex mixture of products, in less than 10% yield (Scheme 2.26). H2N
DIB (CF3)2CHOH
HN
<10 %
HO
O 96
97
Scheme 2.26
Inefficient oxidative cyclization of phenolic primary amines.
We stress that the terminology ‘‘oxidative amidation of phenols’’ as used herein implies a mechanism that involves an initial oxidation of the phenol to an electrophilic species, which is then captured by the nitrogen atom of an appropriate amide equivalent. On the other hand, electronically complementary spirolactam syntheses involving the capture of an electrophilic N atom by an electron-rich aromatic ring have been known since the early 1980s, having been independently developed by Glover [31] and Kikugawa [32]. Notable examples of this chemistry evolve from N-alkoxy-acylnitrenium ions (Scheme 2.27). O MeO
N H
O −2e
−
MeO
O
MeO N
N
− H+
RO
O
RO 98
Scheme 2.27
99
100
Glover–Kikugawa reaction.
The synthetic potential of these processes has become apparent in recent times through the work of Wardrop and collaborators [33]. In particular, these workers disclosed a synthesis of TAN-1251A (3) based on an initial oxidative cyclization of tyrosine derivative 102→103 (Scheme 2.28). Whether this approach qualifies as ‘‘biomimetic’’ is a matter of debate. By the criterion enounced at the beginning of this chapter, the judgment would be in the negative (it does not involve a tyrosinyl
75
76
2 Biomimetic Synthesis of Alkaloids Derived from Tyrosine OMe
(L)-tyrosine
1. MeOCOCl, NaOH then NaOH, H2O
PhI(OCOCF3)2, CH2Cl2, MeOH
1. NaOH, H2O
OMe 2. MeONH2HCl,
2. (MeO)2SO2, K2CO3 HN 76 %
MeO
O
HN
DCC, HOBT, 73 %
O
NHOMe then H2O, 69 %
MeO
O
O
N OMe HN
MeO
O O 103
102
101
Scheme 2.28
O
OMe
Wardrop oxidative cyclization of 102→103.
tyrosine equivalent). Regardless, the strategy is sufficiently interesting as to merit a brief parenthesis. Thus, spirolactam 103 was elaborated to intermediate 107, which underwent aldol condensation with 108 to provide 109 and subsequent alane reduction yielded (−)-TAN-1251A (3) (Scheme 2.29) [34]. Overall, 3 was prepared in only 13 steps from (l)-tyrosine. This is one of the more efficient syntheses of TAN compounds reported to date. O
103
1. H2, 10 % Pd/C 2. (CH2OH)2, PPTS, 86 %
N
O N H
MeO
O
O OMe
LAH
N
65%
prenyl-O
57 %
N
O O
N
Me
O 107
OMe
1. LDA then 108 2. MsCl, Et3N then t-BuOK, THF, 60 %
OBn
N O
N Cbz
105 CHO 108
O
1. CbzCl, Et3N 2. Zn, HOAc 3. BrCH2CO2Bn 66 % Me
Me N H
O 104
1. H2, Pd/C 2. DPPA
O
O
106
Me N
O
Me
N O 2. 1 M HCl,
O
N
1. AlH3, Et2O
N
O
61 %
O
109
O TAN-1251A (3)
Scheme 2.29
Wardrop synthesis of TAN-1251A.
A similar strategy is apparent in a synthetic study toward FR-901483 [8c]. Oxidative cyclization of 99 (R = H) and reduction of the emerging 100 yielded 110, which upon N-propargylation and enol silyl ether formation afforded compound 112 (Scheme 2.30). Reaction of the latter with tributyltin hydride under radical conditions, followed by acidic treatment, delivered a mixture of 115 (40%), 116 (2%), and 117 (45%). Tricyclic lactam 115 is structurally related to 1. Its formation may be rationalized in terms of addition of tributyltin radical to the alkyne and radical cyclization of intermediate 113, followed by protiodestannylation of the resultant vinyltin species upon acidic workup. By-product 116 is likely to form through reduction of 113 and destannylation, while pyrrolizidinone 117 presumably arises via intramolecular H-atom transfer of 113 leading to 114 and radical cyclization of the latter.
2.4 Biomimetic Syntheses of FR-901483 and TAN-1251 Compounds O
O 100
O
O
1. H2, Pd/C 2. (CH2OH)2, PPTS
1. NaH then BrCH2CCH
3. Na, NH3 then NH4Cl, 81 %
N
then TMSI
Me3SiO
86 %
O 111
O 110
112
O
O
O
N
N
N
+
HO
Me3SiO
113
Bu3SnH, AIBN
116 (2 %)
O Bu3Sn
O
Bu3Sn
O
115 (40 %)
H atom transfer
then MeOH, HCl
N
N
Me3SiO
Scheme 2.30
N
(TMS)2NH
NH 2. 0.5 M HCl,
Bu3Sn
77
114
117 (45 %)
O
Wardrop approach to FR-901483.
The straightforward sequence delineated in Scheme 2.31 then converted intermediate 115 into desmethylamino FR-901483 (123). O
O N
2. OSO4 then NaIO4, 85 %
HO
O
1. NaH, BnBr
O
N
KHMDS then Et3SiCl
BnO
115
Et3SiO
ZnCl2·Et2O then
N
p -MeOBnBr 120, 68 %
BnO
118
119
OMe
OMe OMe
O O
N
BnO
1. LAH
SmI2,THF 85 %
O HO
N
2. Pd(OH)2/C, H2, 66 % 3. ref 8b, 38 %
HO
N
BnO 120
121 OP(OH)2
desmethylamino FR-901483 (123)
O
Scheme 2.31
Wardrop synthesis of desmethylamino FR-901483 (123).
2.4 Biomimetic Syntheses of FR-901483 and TAN-1251 Compounds via Oxidative Amidation Chemistry and Related Processes
The advent of methodology for the oxidative amidation of phenols enabled biomimetic syntheses of 1 and of TAN-1251 compounds through the actual
78
2 Biomimetic Synthesis of Alkaloids Derived from Tyrosine
union of two molecules of tyrosine. The Sorensen synthesis of FR-901483 was the first one to document this approach, and it was followed shortly thereafter by the Ciufolini synthesis of FR-901483 and of TAN-1251C. The 2005 Honda synthesis of TAN-1251C and TAN-1251D concludes the cycle of biomimetic syntheses relying on oxidative amidation chemistry. 2.4.1 Sorensen Synthesis of FR-901483
In 2000, Sorensen disclosed a synthesis of FR-901483 that relied on the oxidative cyclization of amine 92 (Scheme 2.24) as the key step [28]. The route to 92 started with the elaboration of tyrosine methyl ester hydrochloride into building blocks 125 and 129 (Scheme 2.32). COOH 1. MeI, K CO 2 3 NHBoc
HO
2. TFA
124
COOMe 1. MeI, K2CO3 NH-G
MeO NsCl, 88 %
2. AlCl3, EtSH 84 %
Z HO
125 G = H
Me
Ns
127 Z = COOMe
126 G = Ns
BH3 • THF SO3 • Py, DMSO 94 %
Scheme 2.32
N
128 Z = CH2OH 129 Z = CHO
Synthesis of building blocks 125 and 129.
The sensitive aldehyde 129 was employed in crude form in a subsequent reductive amination with 125 (Scheme 2.33). This provided tyrosinyl tyrosine derivative 92, which upon reaction with PhI(OAc)2 in HFIP underwent oxidative cyclization to 93. Exchange of the N-protection group, hydrogenation of the dienone 93 and redox manipulation of the resultant 130 delivered aldehyde 131. OMe OHC
Ns N Me
MeOOC
NH2 NaBH(OAc)3
+
129
125
51 % brsm
PAN
PAN Ns N Me
N
92
HO
PAN
O
N H
(CF3)2CHOH
MeO
HO
MeOOC
MeOOC
80 %
PhI(OAc)2
Ns N Me
93
Scheme 2.33
1. NaSPh then (Boc)2O, py 2. H2, Raney Ni, 66 %
MeOOC
HO
Boc N Me
N
130
1. LAH 2.(COCl)2, DMSO 90 %
OHC
Boc N Me
N
O 131
Sorensen route to aldehyde 131.
In contrast to the Snider substrate, 15, the stereoselective aldol cyclization of 131 occurred best in methanol containing MeONa (Scheme 2.34). Notice that 131 differs from 15 in that it lacks a carbonyl group on the pyrrolidine segment.
2.4 Biomimetic Syntheses of FR-901483 and TAN-1251 Compounds
PAN
PAN HO 131
Boc
N
NaOMe
Raney Ni
Boc (4-ClPh)3P, N Me DIAD, Et3N, 37 %
N
HO
92 %
132
(PhO)2P(O)OH,
HO
H2
N Me
34 %
O
133
H
PAN
PAN HO
Boc N Me
N
H (BnO)2P O O Scheme 2.34
HO
1. H2, Pd/C, MeOH 2. 4 N HCl, dioxane, 80 %
N
(HO)2P O O
134
NHMe•HCl
H FR-901483 (1)
Sorensen synthesis of FR-901483.
Tricyclic intermediate 132 was thus isolated in 34% yield from the ensuing mixture of products. Stereoselective reduction of the ketone from the convex face was accomplished by hydrogenation, and the resultant 133 underwent a Mitsunobu reaction with dibenzyl hydrogen phosphate to afford 134. In accord with Snider, the considerably less reactive OH group adjacent to the benzyl-type substituent did not interfere. Deblocking of 134 produced the target 1, which was thus reached in 18 steps from N-Boc tyrosine 124. 2.4.2 Honda Synthesis of TAN-1251 Substances
A key aspect of the Honda synthesis of TAN-1251 compounds is the reduction of dienones of the type 95 to the corresponding cyclohexanones (Scheme 2.35). This step was problematic and it necessitated the implementation of a variant of a procedure developed earlier by Buchwald [35]. Thus, treatment of 95 O
O NMe
N
H
O NMe
Et3SiH, CuCl
N
H
dppf
O
O
95
NMe N
+
135 (9 %)
O
NMe N
PPTS
H
[ref 32]
TAN-1251A (3)
66 %
O O Scheme 2.35
107
Honda formal synthesis of TAN-1251A (3).
H
136 (60 %)
O HO(CH2)2OH
79
80
2 Biomimetic Synthesis of Alkaloids Derived from Tyrosine
with 3 equiv. of Et3 SiH in the presence of 20 mol% of CuCl and 20 mol% of diphenyl(phosphino)ferrocene (dppf) accomplished reduction to 136 in 60% yield. The reaction also afforded some enone 135 (9% yield). Ketalization of 136 produced a substance identical to the Wardrop TAN-1251A precursor, 107, thereby realizing a formal synthesis of that natural product [29, 34]. In 2005, Honda reported actual total syntheses of TAN-1251C and TAN-1251D based on his oxidative cyclization reaction [36]. The route to 5→6 commenced with the assembly of intermediate 143 by a method that retraced the Sorensen avenue to 92 (Scheme 2.32). Thus, (l)-tyrosine methyl ester was transformed into aldehyde 140, which upon reductive amination with 141 afforded 142 (Scheme 2.36). Release of the Boc group and cyclization of the resultant compound surrendered piperazinone 144. The crucial oxidative cyclization of 145 with DIB in HFIP delivered 146 in 49% yield. This material was reduced to the corresponding cyclohexanone as seen earlier (Scheme 2.35). A sequence involving ketalization, hydrogenolysis of the benzyl group and O-prenylation of the free phenol led to intermediate 148, which was the precursor of both TAN-1251C and TAN-1251D (Scheme 2.36). The action of DIBAL upon 148 produced enamine 149, presumably through elimination from a carbinolamine-type intermediate formed upon delivery of a hydride to the carbonyl system (Scheme 2.37). Aqueous hydrolysis of the ketal liberated totally synthetic TAN-1251C. By contrast, LAH in the presence of AlCl3 converted 148 into 150, which upon ketal hydrolysis furnished TAN-1251D. 2.4.3 Ciufolini Synthesis of FR-901483 and TAN-1251C
The 2001 Ciufolini syntheses of FR-901483 and TAN-1251C [37] relied on the oxidative cyclization of phenolic oxazolines obtained through the union of a molecule of N-tosyltyrosine, 152, with one of tyrosinol methyl ether, 151. Quantities of 152 of excellent chemical and optical purity were prepared from tyrosine ethyl ester according to a 1915 procedure by Emil Fischer (Scheme 2.38) [38]. Notably, while tosylation of tyrosine ester produced the anticipated sulfonamide very selectively, that of free tyrosine gave only the N,O-ditosyl derivative [39]. The choice of a tosylamide, instead of a more readily cleavable Fukuyama nitrosulfonamide [40], was motivated by a desire to retain the original N-protecting group during later transformation requiring reductants or nucleophiles. The N-tosyl group would be cleaved during hydride reduction of the pyrrolidinone to the corresponding pyrrolidine. By contrast, the elaboration of tyrosinol 151 by literature methods [41] incurred significant erosion of optical integrity. Fortunately, the route shown in Scheme 2.38 afforded a product of excellent optical and chemical quality. Merger of 151 and 152 in the manner of Vorbr¨uggen afforded oxazoline 153, which underwent oxidative cyclization and acetylation to provide 155 (Scheme 2.39). Notice that the latter step results also in N-acylation of the tosylamide. However, this is inconsequential, because the acetyl groups are released simultaneously at a
O
Scheme 2.36
BnO
PMBO
N
N
O
146
H
NMe
Boc
CHO 141 OBn
PMBO 137
2. ethylene glycol, PPTS, 99 %
BnO
ZnBr2 99 %
PMBO
G N
O O
N
147
O
143 G = H
85 %
NaOMe
N G
OH
TFA 92 %
O
2. prenyl bromide, NaH, 73 %
O
145 R = H
O O
H
NMe DIB
N
H
NMe
148
O
(CF3)2CHOH 49 %
99 %
(COCl)2 DMSO
144 R = PMB
R-O
HN
139 G = Boc
138 G = H
Me
BnO
(Boc)2O 99 %
PMBO
NMe 1. H2, Pd(OH)2, 99 % H
Me
OBn
82 %
LAH
142 G = Boc
N H
NHCOOEt
MeOOC
1. CuCl, dppf, Et3SiH, t -BuONa 65 %
NaBH3CN, 90 %
H2N
MeOOC
2. PMBCl, 96 %
COOMe
Honda route to 148, precursor of TAN-1251 substances.
140
Me
(L)-tyrosine methyl ester
1. ClCOOEt, 88 %
2.4 Biomimetic Syntheses of FR-901483 and TAN-1251 Compounds 81
82
2 Biomimetic Synthesis of Alkaloids Derived from Tyrosine
O NMe N
O
H
O
N
O
83 %
DIBAL
H
O
150
O
NMe
NMe
LAH AlCl3
N
O
70 % 148
O
H
O 149
O
1 M HCl 67 %
1 M HCl 70 % NMe N
O O
COOH NHTs 152
N
O
TAN-1251D (6)
Scheme 2.37
HO
NMe
H
TAN-1251C (5)
H
O
Honda synthesis of TAN-1251C (5) and D (6).
COOH
1. EtOH, HCl(g), 100 % 2. TsCl, 98 % 3. aq NaOH, 98 %
NH2
HO
1. MeOOCCl, NaHCO3, 99 %
OH
K2CO3, 99 %
(L)-tyrosine
NH2
MeO
2. Me2SO4,
151
3. LAH, 85 % 4. aq KOH, 86 %
Scheme 2.38
Preparation of building blocks 151 and 152.
MeO
151 + 152
PPh3, CCl4,
DIB,
NH-Ts
N
PAN O
R O
O
N
CF3CH2OH
73 %
Ac2O, py, 41 %
O
153 HO PAN O PtO2,
R N-Ts
G-O
N
G NTs
PAN O MeI, K2CO3
N
97 %
96 %
O K2CO3 MeOH 79 %
Scheme 2.39
HO
156 G = Ac 157 G = H
Ciufolini synthesis of alcohol 158.
O 158
Me N Ts
154 R = H 155 R = Ac
2.4 Biomimetic Syntheses of FR-901483 and TAN-1251 Compounds
83
later stage. Saturation of the dienone was carried out by hydrogenation over PtO2 (Adams catalyst). Palladium or rhodium catalysts were unsatisfactory, because they promoted various degrees of reductive aromatization of 155 through C–N bond cleavage. Hydrogenation in the presence of supported platinum, for example, Pt(C), resulted in a variable extent of ketone reduction to the corresponding cyclohexanol. No such problem was observed with the Adams catalyst. Simultaneous N- and O-deacetylation of 156 prepared the molecule for selective methylation of the tosylamide, which was readily accomplished with MeI in the presence of K2 CO3 . The emerging 158 is the forerunner of both TAN-1251C and of FR-901483. The assembly of 5 (Scheme 2.40) commenced with O-demethylation of 158 and prenylation of the intermediate phenol. The resultant 159 underwent stereoselective LAH reduction to aminodiol 160. The latter is recognized as the Snider TAN-1251 intermediate. Whereas elaboration of 160 as prescribed by Snider (oxidation to keto aldehyde 162 with DMSO/TFAA followed by treatment with methanolic K2 CO3 ) did effect conversion into TAN-1251C, we found the overall yield of the final sequence disappointing. We suspected that the basic treatment required to release the N-trifluoroacetyl group may have diverted a portion of 162 into aldol manifolds. Considerable improvement in overall efficiency was observed upon protection of the amino group as a 2,2,2-trichloroethyl (Troc) carbamate, followed by Ley oxidation [42] of the resulting 160–162 and final Troc release with Cd/Pb couple [43]. Because Snider had shown that all other TAN-1251 compounds are accessible from TAN-1251C (5), a synthesis of 5 is tantamount to a formal synthesis of 3, 4, and 6. OMe 1. BBr3, 87 %
O HO
N
Ts 158
O
2. prenyl bromide, 98 %
LAH
O HO
Me N
O
O
N
Me
NMO 63 %
Scheme 2.40
O
N
Me N
Me N R
O
HO Troc-Cl 62 %
Troc O
N
Ts 159
O
TPAP
HO
88 %
N
162
161 R = COOCH2CCl3
NMe
Cd/Pb NH4OAc buffer 79 %
160 R = H
N
O
H
O TAN-1251C (5)
Ciufolini synthesis of TAN-1251C (5).
The synthesis of FR-901483 commenced with the oxidation of 158 to keto aldehyde 163 (Scheme 2.41), which is destined to undergo a crucial aldol cyclization. The Snider synthesis of 1 had already appeared in print at this juncture of our own work: the important observations recorded by the Brandeis team greatly assisted us in our endeavor. An aspect of our work that merits some space here pertains to the results of earlier computational (MM+) and experimental results. First, an MM+
84
2 Biomimetic Synthesis of Alkaloids Derived from Tyrosine MeO
OMe
TPAP
O HO
N
N
Me
O
NMO 77 %
Ts
O N
N
Me Ts
O
O
158
Scheme 2.41
163
Oxidation of 158 to the aldol substrate 163.
simulation carried out with simplified model substrate 164 failed to detect a distinct preference for any one regio- or stereoisomer of the aldol product. An estimate of the relative energy demand of transition states 165 (leading to the desired regioisomer) and 168 (producing the incorrect regioisomer) indicated that 165a,b are less energetic than 168 by about 0.8 kcal mol−1 . However, such a small energy difference was below the confidence level of the calculation (±1–2 kcal mol−1 ). Furthermore, 165a and 165b were calculated to be isoenergetic, as were the two diastereomers of either regioisomer of the aldol product. This implied that no substrate-directed diastereocontrol could be anticipated, and that the aldol step was likely to become non-stereoselective if carried out under conditions of thermodynamic reversibility (Scheme 2.42). O
O axial H Ar N O
MeO 168
O
base Ar
HO
N O
O
more energetic TS
N
O
O N
O
base less energetic TS
Ar equatorial H H
165a O N H
164
Scheme 2.42
Ar equatorial H O O
165b
169 O isoenergetic diastereomers
O
Ar HO
166
O
Ar HO
O
N
O
N
O
167
isoenergetic diastereomers
Pathways of aldol cyclization.
Turning to experiment, we found that cyclization of 163 using DBU in CH2 Cl2 formed 170 stereoselectively (Scheme 2.43, stereoisomeric products were also formed). This result reaffirmed the fact that the enolization of the cyclohexanone is kinetically faster than that of the aldehyde. More importantly, it indicated that the formation of the major product was taking place in accord with the so-called Seebach rule [44]. This induced us to suspect that Seebach forces, rather than the chelation metal alkoxide intermediate invoked by Snider, may determine stereoselectivity in nonprotic solvents. Such a reactivity model is valid only in nonprotic solvents: polar ones may induce erosion, or even reversal, of Seebach diastereoselectivity, presumably due to solvation of reactive species through H-bonding. Faced with the choice of advancing 170 into the synthesis through an inversion of configuration of
2.4 Biomimetic Syntheses of FR-901483 and TAN-1251 Compounds
MeO Ar O
O N
DBU
HO
NMeTs CH2Cl2
O
N
NMeTs
O
O 163 Scheme 2.43
H
170
Stereoselective aldol cyclization of 163 promoted by DBU.
the newly formed carbinol, or relying on the then-newly published work of Snider and endeavor to create the correct aldol diastereomer by operating in protic solvents, we opted for the second solution. One of the reasons that determined our choice was the anticipated (and in retrospect unjustified) difficulty of inducing an SN 2 reaction at the level of a secondary center flanked on either side by branched carbon atoms. In fact, Funk subsequently demonstrated that this reaction is quite feasible [6a]. The Snider aldol conditions (t-BuOH, t-BuOK), so successful with substrate 15, proved to be unsatisfactory with ketoaldehyde 163. First, the compound suffered from modest solubility in t-BuOH, necessitating the use of a 3 : 1 mixture of t-BuOH/THF to effect aldol cyclization. The resulting decrease in polarity and proticity of the medium was deleterious to yields and selectivity: 171 emerged as component of a mixture of products in a poor 21% yield after chromatography (Scheme 2.44). A switch to EtOH/EtONa and then to MeOH/MeONa progressively improved yields and stereoselectivities. This again contrasted with the behavior of the Snider substrate, but it was consonant with the Sorensen findings. Evidently, optimal conditions for this step are intimately dependent on the structure of the substrate. Ar
MeO O O
O
Scheme 2.44
N
HO solvent NMeTs base
163
O
H
O
N
NMeTs 171
• 3 : 1t-BuOH/THF; t-BuOK: 21 % • 9 : 1MeOH/H2O; MeONa: 44 %
Improved stereoselectivity of the aldol cyclization of 163 in protic media.
The fact that media of greater polarity2) and hydrogen bonding ability3) had a favorable influence on the outcome of the reaction induced us to examine the effect of added water. A methanolic solution of 163 was found to remain homogeneous upon dilution with up to 10 vol.% of water. Addition of solid NaOMe to such an aqueous/methanolic solution triggered rapid and diastereoselective aldol 2) Dielectric constants at 25 ◦ C: t-BuOH =
10.9; EtOH = 24.3; MeOH = 32.6 [45]. 3) The OH group is more accessible in the less sterically hindered MeOH and EtOH
relative to the more sterically encumbered t-BuOH. This should result in stronger H-bonds.
85
86
2 Biomimetic Synthesis of Alkaloids Derived from Tyrosine
cyclization leading to 171 in 44% chromatographed yield. We speculate that if the biosynthesis of FR-901483 (perhaps via intermediate of type 171) indeed involves aldol cyclization of a species related to 163, then the occurrence of such an event within an enzyme possessing a hydrophilic active site hosting numerous water molecules may well assist in the formation of the correct aldol diastereomer. The synthesis was completed rapidly and uneventfully from 171 (Scheme 2.45). Vigorous LAH reduction produced 172 (82%) plus a small amount of the epimeric alcohol. Mitsunobu reaction of 172 with dibenzyl phosphate according to Sorensen yielded extremely polar 173, which was best converted into the N-Cbz derivative 174 to effect purification. Hydrogenolysis in the presence of aqueous HCl provided the bis-hydrochloride salt of 1, which was identical in all respects to material prepared from an authentic sample of FR-901483, kindly provided by the Fujisawa Pharmaceutical Company. In summary, the synthesis of FR-901483 required 17 steps from (l)-tyrosine and resulted in a 1.3% overall yield of 1, while that of TAN-1251C proceeded in 16 linear steps from (l)-tyrosine and in 4% overall yield. Ar
Ar 171
LAH 91%
HO
HO
H N
NHMe
HO H
172
H
DIAD, (BnO)2P(O)OH, (4-ClPh)3P
O (BnO)2P O CbzCl, 26 %
Scheme 2.45
N
N
R
H2
Me 94%
FR-901483 (1)
173 R = H 174 R = Cbz
Ciufolini synthesis of FR-901483 (1).
The oxidative amidation of phenols remains the centerpiece of several ongoing synthetic endeavors in our laboratory. Present efforts retain very little of the original ‘‘biomimetic’’ flavor that stimulated the development of the technology. Still, our initial investigations spawned a great deal of new chemistry that has proved, and continues to prove, quite valuable in the synthesis of diverse nitrogenous substances. Undeniably, biosynthetic hypotheses remain a major source of inspiration for the development of new reactions and for the progress of organic chemical technology. References 1. Stanforth, S.P. (2006) Natural Prod-
uct Chemistry at a Glance, Blackwell Publishers, Malden. 2. Leading reviews: (a) Misra, N., Luthra, R., Singh, K.L., and Kumar, S. (1999) Comprehensive Natural Product Chemistry, vol. 4, Elsevier, Amsterdam, pp. 25–59; (b) Tillequin, F. (2007) Phytochem. Rev., 6, 65–79; (c) Sato, F., Inui, T., and Takemura, T. (2007) Curr. Pharm.
Biotechnol., 8, 211–218; (d) Liscombe, D.K. and Facchini, P.J. (2008) Curr. Opin. Biotechnol., 19, 173–180. 3. Sakamoto, K., Tsujii, E., Abe, F., Nakanishi, T., Yamashita, M., Shigematsu, N., Izumi, S., and Okuhara, M. (1996) J. Antibiot., 49, 37–44. 4. Tsujii, E., Nakanishi, T., Takase, S., Yamashita, M., Izumi, S., and
References
5.
6.
7.
8.
Okuhara, M. (1993) PCT Int. Appl. WO 9, 312,125. Shirafuji, H., Tsubotani, S., Ishimaru, T., and Harada, S. (1991) PCT Int. Appl. WO 9, 113,887. Total syntheses of FR-901483: (a) Maeng, J.-H. and Funk, R.L. (2001) Org. Lett., 3, 1125–1128; (b) Kan, T., Fujimoto, T., Ieda, S., Asoh, Y., Kitaoka, H., and Fukuyama, T. (2004) Org. Lett., 6, 2729–2731; (c) Ieda, S., Asoh, Y., Fujimoto, T., Kitaoka, H., Kan, T., and Fukuyama, T. (2009) Heterocycles, 79, 721–738 Total syntheses of TAN-1251 compounds: (a) Carson, C.A. and Kerr, M.A. (2009) Org. Lett., 11, 777–779; (b) Nagumo, S., Nishida, A., Yamazaki, C., Murashige, K., and Nawahara, N. (1998) Tetrahedron Lett., 39, 4493–4496; (c) Nagumo, S., Nishida, A., Yamazaki, C., Matoba, A., Murashige, K., and Kawahara, N. (2002) Tetrahedron, 58, 4917–4924; (d) Nagumo, S., Matoba, A., Ishii, Y., Yamaguchi, S., Akutsu, N., Nishijima, H., Nishida, A., and Kawahara, N. (2002) Tetrahedron, 58, 9871–9877 Synthetic studies toward FR-901483: (a) Bonjoch, J., Diaba, F., Puigbo, G., Sole, D., Segarra, V., Santamaria, L., Beleta, J., Ryder, H., and Palacios, J.-M. (1999) Bioorg. Med. Chem., 7, 2891–2897; (b) Brummond, K.M. and Lu, J. (2001) Org. Lett., 3, 1347–1349; (c) Wardrop, D.J. and Zhang, W. (2001) Org. Lett., 3, 2353–2356; (d) Suzuki, H., Yamazaki, N., and Kibayashi, C. (2001) Tetrahedron Lett., 42, 3013–3015; (e) Bonjoch, J., Diaba, F., Puigbo, G., Peidro, E., and Sole, D. (2003) Tetrahedron Lett., 44, 8387–8390; (f) Panchaud, P., Ollivier, C., Renaud, P., and Zigmantas, S. (2004) J. Org. Chem., 69, 2755–2759; (g) Brummond, K.M. and Hong, S.-P. (2005) J. Org. Chem., 70, 907–916; (h) Gotchev, D.B. and Comins, D.L. (2006) J. Org. Chem., 71, 9393–9402; (i) Kaden, S. and Reissig, H.-U. (2006) Org. Lett., 8, 4763–4766; (j) Diaba, F., Ricou, E., and Bonjoch, J. (2006) Tetrahedron: Asymmetry, 17, 1437–1443; (k) Kropf, J.E., Meigh, I.C., Bebbington, M.W.P., and Weinreb, S.M. (2006)
9.
10.
11.
12.
13.
14.
15.
J. Org. Chem., 71, 2046–2055; (l) Simila, S.T.M. and Martin, S.F. (2007) J. Org. Chem., 72, 5342–5349; (m) Asari, A., Angelov, P., Auty, J.M., and Hayes, C.J. (2007) Tetrahedron Lett., 48, 2631–2634; (n) Diaba, F., Ricou, E., Sole, D., Teixido, E., Valls, N., and Bonjoch, J. (2007) Arkivoc, iv, 320–330 Synthetic studies toward TAN-1251 compounds: (a) Seike, H. and Sorensen, E.J. (2008) Synlett, 695–701; (b) Nagumo, S., Akutsu, N., Yamazaki, C., Kawahara, N., and Nishida, A. (1998) Tennen Yuki Kagobutsu Toronkai Koen Yoshishu, 40, 595–600; (c) Auty, J.M.A., Churcher, I., and Hayes, C.J. (2004) Synlett, 1443–1445 (a) Snider, B.B. and Lin, H. (1999) J. Am. Chem. Soc., 121, 7778–7786; (b) Snider, B.B., Lin, H., and Foxman, B.M. (1998) J. Org. Chem., 63, 6442–6443. (a) Available in three steps from O-methyl tyrosine methyl ester (17) according to the procedure of: Grundke, G., Keese, W., and Rimpler, M. (1987) Synthesis, 1115–1116. (a) Chow, Y.L., Col´on, C.J., and Tan, J.N.S. (1968) Can. J. Chem., 46, 2821–2825; (b) Lunazzi, L., Cerioni, G., Foresti, E., and Macciantelli, D. (1980) J. Chem. Soc. Perkin Trans. 2, 717–723; (c) Johnson, F. (1968) Chem. Rev., 68, 375–413 (a) Garner, P. (1984) Tetrahedron Lett., 25, 5855–5858; (b) review: Jurczak, J. and Golebiowski, A. (1989) Chem. Rev., 89, 149–164 (a) Myers, A.G. and Kung, D.W. (1999) J. Am. Chem. Soc., 121, 8401–8402; (b) Myers, A.G., Zhong, B., Movassaghi, M., Kung, D.W., Lanman, B.A., and Kwon, S. (2000) Tetrahedron Lett., 41, 1359–1362; (c) Myers, A.G., Kung, D.W., and Zhong, B. (2000) J. Am. Chem. Soc., 122, 3236–3237; (d) Myers, A.G., Zhong, B., Kung, D.W., Movassaghi, M., Lanman, B.A., and Kwon, S. (2000) Org. Lett., 2, 3337–3340 (a) Myers, A.G. and Kung, D.W. (1999) J. Am. Chem. Soc., 121, 10828–10829; (b) Myers, A.G. and Kung, D.W. (2000) Org. Lett., 2, 3019–3022; (c) Myers, A.G.
87
88
2 Biomimetic Synthesis of Alkaloids Derived from Tyrosine
16. 17.
18.
19.
20.
21.
22.
23.
24.
25.
and Lanman, B.A. (2002) J. Am. Chem. Soc., 44, 12969–12971; (d) Charest, M.G., Lerner, C.D., Brubaker, J.D., Siegel, D.R., and Myers, A.G. (2005) Science, 308, 395–398; (e) Charest, M.G., Siegel, D.R., and Myers, A.G. (2005) J. Am. Chem. Soc., 127, 8292–8293; (f) Kwon, S. and Myers, A.G. (2005) J. Am. Chem. Soc., 127, 16796–16797. Snider, B.B. and Lin, H. (2000) Org. Lett., 2, 643–646. (a) Davis, F.A. and Haque, M.S. (1986) J. Org. Chem., 51, 4083–4085; (b) Davis, F.A. and Chen, B.-C. (1993) Chem. Rev., 92, 919–934. (a) Tamura, Y., Yakura, T., Haruta, J.-I., and Kita, Y. (1987) J. Org. Chem., 52, 3927–3930; See also: (b) Kita, Y., Tohma, H., Kikuchi, K., Inagaki, M., and Yakura, T. (1991) J. Org. Chem., 56, 435–438. (a) Review: Ciufolini, M.A., Braun, N.A., Canesi, S., Ousmer, M., Chang, J., and Chai, D. (2007) Synthesis, 3759–3772. Knapp, S. (1996) in Advances in Heterocyclic Natural Product Synthesis, vol. 3 (ed. W.H. Pearson), JAI Press, Greenwich, pp. 57–98. Oxazolines also participate effectively in iodolactamization reactions: Kurth, M.J. and Bloom, S.H. (1989) J. Org. Chem., 54, 411–414. (a) Vorbr¨uggen, H. and Krolikiewicz, K. (1981) Tetrahedron Lett., 22, 4471–4474; (b) Vorbr¨uggen, H. and Krolikiewicz, K. (1993) Tetrahedron, 49, 9353–9372. For example, the Wipf method: (a) Wipf, P. and Miller, C.P. (1992) Tetrahedron Lett., 33, 907–910; (b) Wipf, P., Kim, Y., and Goldstein, D.M. (1995) J. Am. Chem. Soc., 117, 11106–11112; (c) Wipf, P. and Li, W. (1999) J. Org. Chem., 64, 4576–4577. Kita, Y., Takada, T., Gyoten, M., Tohma, H., Zenk, M.H., and Eichhorn, J. (1996) J. Org. Chem., 61, 5857–5864, see also Reference [17]. (a) Braun, N.A., Ciufolini, M.A., Peters, K., and Peters, E.-M. (1998) Tetrahedron Lett., 39, 4667–4670; (b) Braun, N.A., Ousmer, M., Bray, J.D., Bouchu, D., Peters, K., Peters, E.-M., and Ciufolini, M.A. (2000) J. Org. Chem., 65, 4397–4408.
26. (a) Canesi, S., Belmont, P., Bouchu,
27.
28.
29.
30.
31.
32.
33.
D., Rousset, L., and Ciufolini, M.A. (2002) Tetrahedron Lett., 43, 5193–5195; (b) Canesi, S., Bouchu, D., and Ciufolini, M.A. (2004) Angew. Chem. Int. Ed., 43, 4336–4338. (a) Canesi, S., Bouchu, D., and Ciufolini, M.A. (2005) Org. Lett., 7, 175–177; (b) Liang, H. and Ciufolini, M.A. (2008) J. Org. Chem., 73, 4299–4301; (c) Mendelsohn, B., Lee, S., Kim, S., Teyssier, F., Aulakh, V.S., and Ciufolini, M.A. (2009) Org. Lett., 11, 1539–1542; (d) Mendelsohn, B.A. and Ciufolini, M.A. (2009) Org. Lett., 11, 4736–4739. (a) Scheffler, G., Seike, H., and Sorensen, E.J. (2000) Angew. Chem. Int. Ed., 39, 4593–4596. Mizutani, H., Takayama, J., Soeda, Y., and Honda, T. (2002) Tetrahedron Lett., 43, 2411–2414. Canesi, S. (2004) La Cylindricine C: Synth`ese et M´ethodologie, Dissertation, University Claude Bernard Lyon 1. (a) Glover, S.A., Goosen, A., McCleland, C.W., and Schoonraad, J.L. (1984) J. Chem. Soc. Perkin Trans. 1, 2255–2260; (b) Glover, S.A. and Scott, A.P. (1989) Tetrahedron, 45, 1763–1776. (a) Kikugawa, Y. and Kawase, M. (1984) J. Am. Chem. Soc., 106, 5728–5729; (b) Kawase, M., Kitamura, T., and Kikugawa, Y. (1989) J. Org. Chem., 54, 3394–3403; (c) Kikugawa, Y. and Kawase, M. (1990) Chem. Lett., 581–584; (d) Kikugawa, Y., Shimada, M., and Matsumoto, K. (1994) Heterocycles, 37, 293–301. (a) Wardrop, D.J., Burge, M.S., Zhang, W., and Ortiz, J.A. (2003) Tetrahedron Lett., 44, 2587–2591; (b) Wardrop, D.J., Landrie, C.L., and Ortiz, J.A. (2003) Synlett, 1352–1354; (c) Wardrop, D.J. and Burge, M.S. (2004) Chem. Commun., 1230–1231; (d) Wardrop, D.J., Zhang, W., and Landrie, C.L. (2004) Tetrahedron Lett., 45, 4229–4231; (e) Wardrop, D.J. and Burge, M.S. (2005) J. Org. Chem., 70, 10271–10284; (f) Wardrop, D.J., Bowen, E.G., Forslund, R.E., Sussman, A.D., and Weerasekera, S.L. (2010) J. Am. Chem. Soc., 132, 1188–1189.
References 34. (a) Wardrop, D.J. and Basak, A. (2001) 35.
36.
37.
38.
Org. Lett., 3, 1053–1056. Moritani, Y., Appella, D.H., Jurkauska, V., and Buchwald, S.L. (2000) J. Am. Chem. Soc., 122, 6797–6798. (a) Mizutani, H., Takayama, J., and Honda, T. (2005) Synlett, 328–330; (b) Mizutani, H., Takayama, J., Soeda, Y., and Honda, T. (2004) Heterocycles, 62, 343–355. (a) Ousmer, M., Braun, N.A., and Ciufolini, M.A. (2001) Org. Lett., 3, 765–767; (b) Ousmer, M., Braun, N.A., Bavoux, C., Perrin, M., and Ciufolini, M.A. (2001) J. Am. Chem. Soc., 123, 7534–7538; (c) Ciufolini, M.A. (2005) Il Farmaco, 60, 627–641; (d) Ciufolini, M.A., Canesi, S., Ousmer, M., and Braun, N.A. (2006) Tetrahedron, 62, 5318–5337. Fischer, E. and Lipschitz, W. (1915) Ber. Dtsch. Chem. Ges., 48, 360–378.
39. McChesney, E.V. and Swann, W.K.
40. 41.
42.
43.
44. 45.
Jr. (1937) J. Am. Chem. Soc., 59, 1116–1118. Kan, T. and Fukuyama, T. (2004) Chem. Commun., 353–359. (a) Abarbri, M., Guignard, A., and Lamant, M. (1995) Helv. Chim. Acta, 78, 109–121; (b) Jung, M.E., Jachiet, D., and Rohloff, J.R. (1989) Tetrahedron Lett., 30, 4211–4214, the material thus produced was essentially racemic. Ley, S.V., Normand, J., Griffith, W.P., and Marsden, S.P. (1994) Synthesis, 639–666. Dong, Q., Anderson, C.E., and Ciufolini, M.A. (1995) Tetrahedron Lett., 36, 5681–5682. Seebach, D. and Golinski, J. (1981) Helv. Chim. Acta, 64, 1413–1423. CRC Handbook of Chemistry and Physics 59th edn (ed. R. C. Weast) (1978) CRC Press, Boca Raton, Florida.
89
91
3 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Indolemonoterpene Alkaloids Sylvie Michel and Franc¸ ois Tillequin
3.1 Introduction 3.1.1 Indolemonoterpene Alkaloids
Tryptophan-derived indolemonoterpene and related quinoline alkaloids represent a group of circa 3000 different compounds isolated to date from higher plants [1]. Many are active principles used in therapeutics that were introduced in clinics during the nineteenth century, following their isolation from plants that were used in medicine at that time, such as quinine (1) from Cinchona bark or strychnine (2) from Strychnos nux-vomica seeds (Figure 3.1). The isolation, in the 1950s of the anti-psychotic and anti-hypertensive reserpine (3) from Rauwolfia serpentina, followed by that of the anticancer vinblastine (4) from the aerial parts of Catharanthus roseus stimulated a new expansion in research on indolemonoterpene alkaloids, including exploration of the biosynthetic pathways giving rise to these natural products [2] and, since the early 1970s, the biomimetic syntheses of these compounds [3], that is, synthetic attempts toward their assembly inspired by the biosynthetic lines, without recourse to the enzymatic machinery present in Nature. Following a brief overview of the structures and botanical distribution of indolemonoterpene alkaloids, this chapter will describe a few selected highlights concerning their biomimetic synthesis, chosen to illustrate useful methods and significant achievements toward the obtainment of the main skeletons encountered in this series of natural products. 3.1.2 Classification and Botanical Distribution
Only a few indolemonoterpene alkaloids, which have been almost exclusively obtained from species belonging to the genus Aristotelia, arise from the condensation of tryptamine with a non-rearranged (geranyl) monoterpene unit [4]. Most of them derive, from a biosynthetic point of view, from the initial condensation of a Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
92
3 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Indolemonoterpene Alkaloids H H HO
H3CO
H
N
H
N N O
N 1 Quinine
H
H H
2 Strychnine OH
N N H H
H3CO
N H O
H H3COOC
OCH3
O
N
N H H3COOC
H N
OCOCH3 R HO COOCH3
H3CO
OCH3
OCH3 OCH3
3 Reserpine
Figure 3.1
O
4 R = CH3 Vinblastine 11 R = CHO Vincristine
Selected examples of indolemonoterpene alkaloids.
tryptamine (5) unit with the iridoid glycoside secologanin (6), under the influence of the enzyme strictosidine synthase, to give stereospecifically strictosidine (7) (Scheme 3.1) [5]. 9 10
CHO N H
NH2
5 Tryptamine
H
+
H H3COOC
11
O
Glc
O
6 Secologanin
12
8 13
6
7 1
5 2
N H H
4
NH
3 19
14 15
H H3COOC
16
20
18
H
O
21
Glc
O 17
7 Strictosidine Scheme 3.1
Biosynthesis of strictosidine (7).
This glycoside is the common precursor to most indolemonoterpene alkaloids, despite the impressive array of their final structures. These alkaloids are essentially encountered in plant families belonging to the order Gentianales (Apocynaceae, Loganiaceae, and Rubiaceae) and in some species belonging to the related order Cornales (Cornaceae, Nyssaceae, Alangiaceae. . .). Compounds of this group with a terpene moiety remaining in an unrearranged form constitute the Corynanthe and Strychnos alkaloids (corynane-strychnane or type I series), whereas rearrangement of the terpenoid part gives rise to the Iboga alkaloids (ibogane or type II series, with fragmentation of the C15–C16 bond and creation of a C14–C17 bond) and the Aspidosperma alkaloids (aspidospermane or type III series, with fragmentation of the C15–C16 bond and creation of a C17–C20 bond) (Scheme 3.2) [6]. Rearrangements permitting ring expansion lead to quinoline alkaloids, exemplified by camptothecin (8) and quinine (1), whereas ring cleavage leads to ring-opened alkaloids such as rhazinilam (9) (Figure 3.2).
3.2 Aristotelia Alkaloids 21
3 19 21
15
18
20 15
3 16
14
20
19 18
H3COOC
14 17
16
16
17
H3COOC
Iboga-type alkaloids
3 15
17 19
Aspidosperma-type alkaloids
Corynanthe and Strychnos-type alkaloids Main skeletons of secologanin-derived indolemonoterpene alkaloids.
N
O
N
N O OH O 8 Camptothecin Figure 3.2
21 20
18
COOCH3
Scheme 3.2
93
O N H 9 Rhazinilam
N H H H H3C
N H
N
H H N
10 Usambarine
Structures of camptothecin (8), rhazinilam (9), and usambarine (10).
Subsequent condensation of an indolemonoterpene with a second tryptamine unit gives rise to quasi-dimeric alkaloids, exemplified by usambarine (10). Dimerization involving two indolomonoterpene units leads to dimeric alkaloids, illustrated by vinblastine (4) and vincristine (11), both isolated from the aerial parts of Catharanthus roseus and currently widely used in cancer chemotherapy.
3.2 Biomimetic Synthesis of Indolomonoterpene Alkaloids with a Non-rearranged Monoterpene Unit: Aristotelia Alkaloids
The Aristotelia alkaloids constitute a small group of some 50 alkaloids. All of them have been isolated in minute amounts from plants belonging to the eleocarpaceous genus Aristotelia indigenous in the Southern hemisphere: Australia, New Zealand, and Chile. A biogenetic scheme leading to Aristotelia alkaloids, through a linkage between tryptamine and an unarranged monoterpene unit such as geraniol, was first hypothesized by Bick [7] and extended later on by Bick and Hesse [8]. Accordingly, nucleophilic attack of tryptamine (5) on the α-terpinyl cation arising from the cyclization of linalyl or neryl diphosphate should give rise to an α-terpinyltryptamine (12) unit possessing the same skeleton as the natural alkaloid (+)-fruticosimine (13), which only differs in its oxidation level. Dehydrogenation of the putative α-terpinyltryptamine unit 12 should generate an aldimine intermediate (14), which should cyclize upon protonation to makomakine (15), aristoteline (16), or hobartine (17), which are the major alkaloids isolated from Aristotelia species (Scheme 3.3).
14
94
3 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Indolemonoterpene Alkaloids O
NH2
N H
N H 5 Tryptamine
12
13 Fruticosimine
N
14
H N
15 Makomakine
N H
H
N H
N H
H N
H N
O-P-P
H N
H N
N H
N H
17 Hobartine
NH H N 16 Aristoteline
Scheme 3.3
Biosynthesis of Aristotelia alkaloids.
Following a pioneering approach by L´evy and coworkers involving isatin as source of indole nucleus [9], the first fully biomimetic synthesis of the Aristotelia alkaloids makomakine (15), aristoteline (16), and hobartine (17) was performed by Stevens and Kenney [10]. Thus, mercury(II) nitrate-initiated Ritter reaction [11] of indol-3-ylacetonitrile (18), bringing the carbon and nitrogen atoms of tryptamine, with (−)-β-pinene (19) gave the corresponding imine, which was reduced with sodium borohydride in alkaline medium to afford (+)-makomakine (15). Further treatment of 15 with concentrated hydrochloric acid gave (+)-aristoteline (16) (Scheme 3.4). Similarly, Ritter condensation of (+)-α-pinene (20) with 18 gave (±)-hobartine (17) (Scheme 3.5). Notably, the enantiospecificity observed in the synthesis of 15 and 16 was not observed in that of 17, since the allylic mercurial intermediate can cyclize at either of the two enantiomeric sites in the latter case (Scheme 3.6). Following the work of Stevens, several syntheses of Aristotelia alkaloids were performed, through an initial condensation of (S)-(−)-α-terpinylamine (21)
3.2 Aristotelia Alkaloids
C
N
Hg(NO3)2 CH2Cl2, −30°C NaOH
NaBH4 CH3OH 3 M NaOH
N
+
N H 18
95
H N
(17%)
N H
19 (−)-b-Pinene
N H 15 (+)-Makomakine cc. HCl
NH H N 16 (+)-Aristoteline
Scheme 3.4
Biomimetic synthesis of (+)-makomakine (15) and (+)-aristoteline (16).
C
N H 18
Scheme 3.5
N
+
H N
N
Hg(NO3)2 CH2Cl2
NaBH4, CH3OH
−30°C NaOH
3 M NaOH (11%)
N H
20 (+)-a-Pinene
N H 17 (±)-Hobartine
Biomimetic synthesis of (±)-hobartine (17).
From (−)-b -pinene
Hg(NO3)2
N
N
O3NHg HgNO3
HgNO3 N H
From (+)-a-Pinene
N
δ− δ−
NO3Hg
N H
Scheme 3.6 Comparison of the mechanisms involved in the syntheses of (+)-makomakine (15) and (+)-aristoteline (16) from (−)-β-pinene, and that of (±)-hobartine (17) from (+)-α-pinene.
N H
N
N H
96
3 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Indolemonoterpene Alkaloids
with various indole substrates, including tryptophyl bromide (22) [12] and indol-3-ylacetaldehyde (23) (Figure 3.3) [13]. This latter approach proved particularly successful, since most of the Aristotelia alkaloids isolated to date could be prepared in the group of Borschberg [14], by condensation of unprotected or protected indol-3-ylacetaldehyde (23) with α-terpinylamine (21), terpinyl-derived amino-alcohols, or synthetic equivalents such as phenylthioterpinylamines. The use of these latter reagents permitted the obtainment of Aristotelia alkaloids at various oxidation levels, exemplified by aristoserratine (24) [15], aristofruticosine (25) [16], and serratenone (26) (Figure 3.4) [17, 18]. Br
CHO
NH2 N H 23
N H 22
21
Figure 3.3 Structures of (S)-(−)-α-terpinlamine (21), trptophl bromide (22), and indol-3-ylacetaldehyde (23).
H N
N
NH H N O 24 Aristoserratine Figure 3.4
N H 25 Aristofruticosine
N H 26 Serratenone
Structures of aristoserratine (24), aristofruticosine (25), and serratenone (26).
3.3 Biomimetic Synthesis of Secologanin-Derived Indolomonoterpene Alkaloids 3.3.1 Strictosidine, Vincoside, and Simple Corynanthe Alkaloids: Heteroyohimbines and Yohimbines
The isolation of the iridoid secologanin (6) in large amounts from Lonicera species permitted the development of the pioneering work performed in the group of R.T. Brown towards the biomimetic synthesis of indolomonoterpene alkaloids [19]. The first achievements, taking place in the early 1970s, implied the condensation of tryptamine (5) with secologanin (6), to give strictosidine (7) and/or its C3 epimer vincoside (27), followed by the hydrolysis of the sugar moiety with β-glucosidase and subsequent rearrangement and/or reduction to give monoterpenoid indole alkaloids (Figure 3.5). However, this initial approach suffered several drawbacks, such as the facile reaction of N4 of 7 and 27 with the carbomethoxy group at C16 to give the corresponding lactams.
3.3 Biomimetic Synthesis of Secologanin-Derived Indolomonoterpene Alkaloids
4
N HH
NH
3
H
H H3COOC 16
O
Glc
O
N H
HN 28
27 Vincoside Figure 3.5
Structures of vincoside (27) and Nb -benzyltrptamine (28).
N H H H H3COOC
N
N H H
H O
H
29 3-Iso-19-epiajmalicine
H H3COOC
N
N H CH3
H O
H
30 19-Epiajmalicine
H H3COOC
N H O
H
31 1-Methyltetrahydroalstonine
Figure 3.6 Structures of 3-iso-19-epiajmalicine (29), 19-epiajmalicine (30), and 1-methyltetrahydroalstonine (31).
A benzyl protecting group at N4, introduced by using Nb -benzyltryptamine (28) as starting material, circumvented this problem and afforded the first biomimetic syntheses of several heteroyohimbine alkaloids, exemplified by 3-iso-19-epiajmalicine (29), 19-epiajmalicine (30) [20], and 1-methyltetrahydroalstonine (31) (Figure 3.6) [21]. Finally, a more efficient and elegant approach, developed more recently in the same group, implied an alteration of the initial sequence, which permitted the selective generation of a single C3 epimer and also avoided the use of the N-benzyl protecting group. In those biogenetically inspired syntheses, secologanin was first selectively converted into variously protected aglucone aldehydes. Pictet–Spengler condensation with tryptamine was carried out afterwards, as one of the last steps of the envisioned synthesis. This strategy proved particularly efficient, since it permitted the group of R.T. Brown to successfully prepare several series of non-rearranged indolemonoterpene alkaloids, including heteroyohimbines and yohimbines, but also anthirine derivatives, rearranged Aspidosperma alkaloids, and ring-expanded compounds in the camptothecin series [22]. The synthesis of ajmalicine (32) illustrates the approach towards heteroyohimbine alkaloids by this modified strategy. Secologanin (6) was first acetylated (Scheme 3.7). The free aldehyde group on the aglucone part was then protected as ` an ethylene acetal, by treatment with ethylene glycol in acidic medium. Zemplen deacetylation of the sugar moiety gave the dihydropyran aldehyde 33. When performed at pH 5, hydrolysis of the sugar part of 33 with β-glucosidase afforded 34. Condensation of 34 with tryptamine (5) followed by reduction of the imine intermediate gave 35, which was deprotected and stereospecifically cyclized to 36 in acidic medium. Finally, epimerization at C3 afforded the desired ajmalicine (32) (Scheme 3.7) [23].
97
OH
O CHO H
Scheme 3.7
O
H
+
O
H O
H H3COOC
OH
CHO
N
O
N H
H H3COOC
Glc
H
(iii) NaOMe, MeOH
O
75%
33
O
H
H H3COOC
N H H
34
O
O
36
N
H
CHO
O
O
H H3COOC
O
H H3COOC
(i) Ac2O, Pyridine (ii) (CH2OH)2, TFA, THF
O
H H
(ii) NaBH4
(i) Pb(OAc)2, AcOH
85%
O
N
H
OH
H
32 Ajmalicine
H H3COOC
N H H
35
HN
O H H3COOC
N H O
50-70%
b-Glucosidase, H2O, pH = 5, 37°C
5, MeOH, NaBH3CN
Glc
Biomimetic synthesis of the heteroyohimbine alkaloid ajmalicine (32) from secologanin (6).
1M HCl in H2O / Me2CO
H H3COOC
O
O
H
6 Secologanin
H3COOC
H
CHO
98
3 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Indolemonoterpene Alkaloids
3.3 Biomimetic Synthesis of Secologanin-Derived Indolomonoterpene Alkaloids
When hydrolysis of the sugar part of the dihydropyran aldehyde 33 was carried out at pH 7 in the presence of β-glucosidase, the cyclohexene carbaldehyde 37 was obtained as the major reaction product, providing an entry toward yohimbine alkaloids. Thus, reaction of 37 with tryptamine (5), followed by reduction, afforded 38, which was cyclized into (−)-3-iso-19,20-dehydro-β-yohimbine (39) by treatment under acidic conditions [24]. This latter compound could be further converted into deserpidine analog 40 and reserpine analog precursor 41 (Scheme 3.8) [25]. 3.3.2 Antirhine Derivatives
A straightforward biogenetically inspired entry to the antirhine group of alkaloids followed the observation that the treatment of secologanin ethylene acetal (33) with baker’s yeast (Saccharomyces cerevisiae) at pH 6.4 gave in high yield the lactol 42, in equilibrium with the open form 43, through hydrolysis of the glucose unit followed by reduction of the aldehyde group at C1. Pictet–Spengler condensation with tryptamine (5), carried out by heating in pH 3.5 aqueous buffer and acetone, afforded stereospecifically 16-methoxycarbonyl-16,17-dehydroantirhine (44) as major reaction product (Scheme 3.9) [26]. For the synthesis of antirhine itself, the lactol 42 was first heated in alkaline medium to afford the decarbomethoxy lactone 45. Condensation of 45 with tryptamine (5) gave the amide 46, which was reduced to the corresponding amine with lithium aluminum hydride. Acid-catalyzed hydrolysis of the acetal protecting group occurred with simultaneous Pictet–Spengler cyclization to afford the desired antirhine (47) (Scheme 3.10). 3.3.3 Conversion of the Corynanthe Skeleton into the Strychnos Skeleton
The group of Stephen Martin described an elegant conversion of the Corynanthe into the Strychnos skeleton, in the continuation of an enantioselective total synthesis of geissoschizine (48) (Figure 3.7) [27]. In this approach, the treatment of the racemic synthetic Corynanthe alkaloid 49 with tert-butyl hypochlorite afforded a mixture of the corresponding epimeric α-and β-chloroindolenines 50 (Scheme 3.11). At this step, the addition of a Lewis acid, such as SnCl4 , prior to the conversion into chloroindolenines played a crucial role in obtaining correct control of the stereochemistry. Indeed, without addition of Lewis acid, the attack of the chloronium ion proceeded essentially by the less hindered convex α-face of the molecule, giving the α-chloroindolenine as major reaction product, which proved inert in the following step. Initial addition of a Lewis acid permitted complexation of the electron lone pair on the basic nitrogen atom, which is projected on the αface, ensuring an increased β attack of the oxidative reagent. More specifically, the use of tin tetrachloride permitted the formation of the β-chloroindolenine as predominant reaction product, accompanied by only trace amounts of the α-isomer. Further deprotonation in strong alkaline medium
99
O
H
O
Glc
75%
O
H H3COOC
10% HCl in H2O/Me2CO
70%
b -Glucosidase H2O, pH = 7, 37°C
39
H H3COOC
N H H
O
O CHO H
Biomimetic synthesis of yohimbine alkaloids.
33
O
Scheme 3.8
H H3COOC
O
N
OH
O
37
H H3COOC
O
OH
CHO
41
H H3COOC
N HH
N H
N
OH
H
40 OH
H H3COOC
N HH
85%
5, MeOH, NaBH3CN
HN
OH
O H H3COOC 38
N H O
OH
100
3 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Indolemonoterpene Alkaloids
3.3 Biomimetic Synthesis of Secologanin-Derived Indolomonoterpene Alkaloids
O
O H H
O
Glc
O
H3COOC
O
Saccharomyces cerevisiae
O
O
O
O
CHO
CHO
COOCH3
CHO
H3COOC
O
H3COOC
33
OH H
OH 42
30-50%
43
N
N H H
5, Me2CO, H2O pH = 3.5
O
H
pH = 6.4, 25°C 60-80%
101
COOCH3
H
OH H
44
Scheme 3.9
O
Biomimetic synthesis of antirhine derivatives.
O
(i) 4 M aq. NaOH (ii) HCl
O
O O 5, EtOH
O
H3COOC
OH 42
Scheme 3.10
N HH
O
HN
N H O
H
O
H
O
(i) AlLiH4, THF (ii) H+
H
OH
46
47 Antirhine (69% overall yield from 42)
Biomimetic synthesis of antirhine (47).
N
H H3COOC OH 48 Geissoschizine
Figure 3.7
OH H
H
45
N
N HH
Structure of geissoschizine (48).
of the crude β-chloroindolenine at C16, to give an enolate that could cyclize onto C2, ensured the skeletal reorganization, affording the Strychnos alkaloid (±)-akuammicine (51) in a good 52% overall yield from 49 (Scheme 3.11) [28]. The same reaction sequence, starting from 52, a protected hydroxylated analog of 49, gave 53, which was further deprotected to 18-hydroxyakuammicine (54), providing an entry to the total biomimetic synthesis of strychnine (2). Indeed, the conversion of 54 into strychnine (2) can be achieved through the intermediacies of the dihydro-derivative 55 and Wieland–G¨umlich aldehyde (56), according to standard procedures (Scheme 3.12) [29]. More recently, the same type of intermediates also permitted a biomimetic entry to the sarpagan alkaloids [30].
3 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Indolemonoterpene Alkaloids
102
Cl
N H H
(i) SnCl4 (ii) t-BuOCl
N
Cl 2
N
2
N
N H
H
H H3COOC
H H3COOC
49
50
16
Cl
H
N
H
N
COOCH3
N
N N H
N
LHMDS, THF
16
H H
COOCH3
N
H COOCH3
H COOCH3
51 Akuammicine (52% overall yield from 49)
Cl+
b-attack
COOCH3 H H
R
N
N H H a-attack
Scheme 3.11
Cl+
Biomimetic synthesis of (±)-akuammicine (51).
3.3.4 Fragmentation and Rearrangements of Corynanthe Alkaloids: Ervitsine-, Ervatamine-, Olivacine-, and Ellipticine-Type Alkaloids
The use of simple or conjugated iminium ions derived from a tetrahydro- or dihydro-pyridine units as intermediates proved an efficient method to construct the backbone of several alkaloids that arise from a classical skeleton through a fragmentation reaction followed by a rearrangement, such as ervitsine, ervatamine, olivacine, ellipticine, and related alkaloids. The groups of Husson [31] and Bosch [32] have in particular studied this methodology, which reproduces the key step invoked in the biosynthesis of a large number of indolemonoterpene alkaloids [33]. For instance, the key step of the pioneering biomimetic synthesis of ellipticine (57) by Langlois et al. [34] (Scheme 3.13) involves cyclization of the imminium 58, followed by oxidation, according to the biogenetic hypothesis previously published by Potier and Janot, which involves fragmentation of the C5–C6 bond of the Corynanthe alkaloid stemmadenine (59) followed by rearrangement [35] (Scheme 3.14).
3.3 Biomimetic Synthesis of Secologanin-Derived Indolomonoterpene Alkaloids
N H H H H3COOC
H
(i) SnCl4, t-BuOCl
N
(ii) LHDMS, THF
H
N
N
HCl, MeOH N H
N H
H COOCH3 OTBS
OTBS 52
53 H
(i) Zn, 10% H2SO4 MeOH
H COOCH3 OH
54 (22% yield from 52) H
N DIBALH, CH2Cl2, −90°C
(ii) NaOMe, MeOH N H
H
NaOAc, CH2(COOH)2,
N
H
H COOCH3 OH 55
Ac2O, AcOH, 110°C
56
N HH HO
H H
O
N
H N
H
O
H O
H 2 Strychnine
Scheme 3.12
Biomimetic synthesis of strychnine (2).
N N H
H
CH3 92%
N N H
CH3
N
CH3
N H
58 35%
Pd-C decaline
N N H 57 Ellipticine
Scheme 3.13
103
Key step of the biomimetic synthesis of ellipticine (57).
The enantioselective total synthesis of N(a) -methylervitsine (60) more recently developed by Bennasar et al. [36] is a typical example of an asymmetric version of this approach. Addition of the enolate of 2-acetyl-1-methylindole (61) onto the N-methylpyridinium salt 62, bearing a chiral auxiliary derived from (S)-O-methylprolinol, afforded the intermediate 1,4-dihydropyridine 63. Conversion into the imminium cation 64, upon treatment with Eschenmoser’s salt, permitted a direct cyclization into the tetracycle 65, obtained as a mixture of diastereoisomers, in which the 15β isomer predominated. Cope elimination performed on the corresponding N-oxides gave methylene derivatives at C16, from which the major diastereoisomer 66 could be separated by crystallization. Removal of the chiral auxiliary required prior reduction to the secondary alcohol 67, which afforded 68 upon reaction with methyllithium. Re-oxidation to the corresponding 2-acylindole followed by final stereoselective elaboration of the (20E )-ethylidene side chain gave the desired (−)-N(a) -methylervitsine (60) (Scheme 3.15).
H CH2OH
4 N 21 3 18 14 20 16 15 19
5
Scheme 3.14
−HCHO
CH3 N N H
N H H3COOC CH2OH
CH2 N
N
CH3
N H
N H
Hypothesis of ellipticine Biosynthesis according to Potier and Janot.
N H
59 Stemmadenine
H3COOC
N H
6
CH2 N
N
CH3 N H
18
17
16
19
15
14
20 21 3
N4
CH3 N
57 Ellipticine
N H
OH
104
3 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Indolemonoterpene Alkaloids
Scheme 3.15
80%
LiBH4, THF
64
O
H
N
N
H
H
O H3CO
54%
68
N CH3OH
H
N
65
H O H3CO
N
N(CH3)2
CH3 MeLi, THF
CH3 O
5 15
N
CH3
N N
40%
N
H
63
N CH3 O
CH3
67
CH3 OH
N
H3CO
H
CH3
62
O H3CO
N
LDA
Biomimetic synthesis of (−)-N(a) -methylervitsine (60).
N CH3 O
(H3C)2N
61
N CH3 O
+
CH3 N
O N
N
CH3
O
25%
(iii) NaBH4, MeOH
N
H
N
H
CH3
H O H3CO
N CH3 O
66
16
N
CH3
20
60 (−)-N(a)-Methylervitsine
CH3 O
N
H
(CH3)2N+ =CH2I−
(ii) Me3OBF4, CH2Cl2
(i) MnO2, CH2Cl2
45% overall yield
(iii) Crystallization
(ii) C6H5CH3, Reflux
(i) m-CPBA, CH2Cl2
H3CO
H
3.3 Biomimetic Synthesis of Secologanin-Derived Indolomonoterpene Alkaloids 105
106
3 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Indolemonoterpene Alkaloids
3.3.5 Iboga and Aspidosperma Alkaloids
From a biogenetic viewpoint, there is general agreement to consider that both Iboga and Aspidosperma alkaloids derive from the Corynanthe precursor stemmadenine (59), through a retro-aldol fragmentation sequence, leading to the intermediate diene dehydrosecodine (69), followed by a Diels–Alder type cycloaddition, which should give rise to either the ibogane skeleton, for example, catharanthine (70), or the aspidospermane system, exemplified by tabersonine (71) (Scheme 3.16) [37]. Based on this biogenetic sequence, Martin Kuehne and his group developed a highly elegant and flexible approach to Iboga and Aspidosperma alkaloids, involving dehydrosecodine synthetic equivalents as key intermediates [38]. The syntheses of catharanthine (70) and tabersonine (71), from the common 15-oxosecodine intermediate 72, can be considered as typical examples of this synthetic strategy [39]. Indeed, keto compound 72 appears as a direct precursor of Aspidosperma alkaloids, whereas a stabilized derivative of its tautomeric enol form can be considered as a precursor of the Iboga skeleton. Thus, the oxosecodine 72 was prepared by condensation of indoloazepine 73 with 1-chloro-2-ethylpenta-1,4-dien-3-one (74), which was obtained by addition of vinylmagnesium bromide followed by Swern oxidation of the Vilsmeier reaction product of butyraldehyde and N,N-dimethylformamide (Scheme 3.17). To construct the Iboga skeleton, the oxosecodine precursor 72 was treated with t-butyldimethylsilyl chloride or triflate in alkaline medium to give the corresponding t-butyldimethylsilyl enol ether, which could not be isolated but underwent spontaneous Diels–Alder cyclization to afford the catharanthine derivative 75. Further conversion of 75 into catharanthine (70) was ensured by transformation into the corresponding oxocoronaridines (76a, 76b) with fluoride ion, conversion into thiones 77a and 77b by reaction with phosphorus pentasulfide, S-methylation to S-methylcatharanthine (78), and final desulfurization with Raney nickel (Scheme 3.18). To access the Aspidosperma series, the oxosecodine precursor 72 was heated in refluxing toluene to give 15-oxovincadifformine (79) as unique Diels–Alder reaction product, obtained in almost quantitative yield. Bromination of ketone 79, followed by sodium borohydride reduction, afforded the bromohydrin 80, which was converted into tabersonine (71) through a radical elimination process upon treatment with TiCl4 -AlLiH4 (Scheme 3.19). 3.3.6 Fragmentation and Rearrangements of Aspidosperma Alkaloids: Vinca Alkaloids and Rhazinilam
Based on a biogenetic hypothesis of formation of the vincamine skeleton through oxidation of an Aspidosperma precursor such as vincadifformine (81) [40], Jean L´evy described an industrial process for the partial synthesis of the clinically interesting alkaloid vincamine from tabersonine, which can be extracted in high yield from
16
OH
Scheme 3.16
H3COOC
N H OH
N
Biosynthesis of Iboga and Aspidosperma alkaloids.
59 Stemmadenine
H3COOC
N H
15
N
N H
N H
N H
69
15
N
COOCH3
N
COOCH3
16
3
COOCH3
N
21
21
3
17
15
14
71 Tabersonine
N 16 H COOCH 3
H
21 20
N
20
14
3
COOCH3
16
70 Catharanthine
N H
17
N 15
3.3 Biomimetic Synthesis of Secologanin-Derived Indolomonoterpene Alkaloids 107
3 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Indolemonoterpene Alkaloids
108
H O
(CH3)2NCHO POCl3
(i) CH2=CHMgBr
H
(ii) Oxdn. (Swern)
O
O
Cl
+
Cl 74
NH N H 73
COOCH3 N
N
O
N
N H
N H
Scheme 3.17
N
N H
O 72 COOCH3
COOCH3
COOCH3
Synthesis of the of Iboga and Aspidosperma precursor 72.
t-BuMe2SiCl DBU
N
N Ot BDMS 98%
N H
N H
O 72 COOCH3
COOCH3 N
F−
O
N H H3COOC 76a, 76b
O
Ot BDMS
N H
N
P4S10 C6H6
N H H3COOC
83%
COOCH3
75 N S
CH3I
77a, 77b
SCH3
N H H3COOC 78 Raney Ni N N H H3COOC
70 Catharanthine (70% from 77a, 77b)
Scheme 3.18
Biomimetic synthesis of the Iboga alkaloid catharanthine (70). N
N O
N N H 72
COOCH3
(i) Br2
N
(ii) NaBH4, MeOH
H
61% N H
Scheme 3.19
99%
N H
O COOCH3
C6H5CH3, 110°C H N H
O
COOCH3
79 15-Oxovincadifformine
Br
N TiCl4, AlLiH4
OH
COOCH3 80
H
71% N H
COOCH3
71 Tabersonine
Biomimetic synthesis of the Aspidosperma alkaloid tabersonine (71).
3.4 Biomimetic Synthesis of Secologanin-Derived Quinoline Alkaloids
109
the seeds of the African trees Voacanga africana and Voacanga thouarsii [41]. Vincadifformine (81), obtained by catalytic hydrogenation of tabersonine C14 = C15 double bond, was first converted into the corresponding hydroxyindolenine N-oxide 82, upon prolonged treatment with a suitable peracid. Reduction of the N-oxide with triphenylphosphine and rearrangement in acidic medium, performed in a one-pot procedure, smoothly afforded the desired vincamine (83), accompanied by its epimer at C16 (Scheme 3.20). O N
14 15
H N H
p - Nitroperbenzoic acid C6H6 80-90%
COOCH3
81 Vincadifformine
N
N H N HO COOCH3
PPh3 AcOH
H N HO COOCH3
82 1,2-Shift
H N HO
16
H3COOC 83 Vincamine
Scheme 3.20
H
N 50-65%
N OH
H+
H
N
H3COOC
N H
O H3COOC
Biomimetic conversion of vincadifformine (81) into vincamine (83).
A similar reaction sequence, performed later on starting from tabersonine (71), gave a mixture of 14,15-dehydrovincamine (84) and 16-epi-14,15-dehydrovincamine (85) in good yields, accompanied by smaller amounts (circa 5%) of the ring-opened N-oxide 86, resulting from the oxidative cleavage of the C2–C3 bond of the indole moiety, that possessed the same chromophore as rhazinilam (9). Catalytic reduction of 86 to 87, followed by saponification, decarboxylation, and lithium aluminum hydride reduction of the intermediate alcohol 88 afforded a biomimetic entry to rhazinilam (9) (Scheme 3.21) [42].
3.4 Biomimetic Synthesis of Secologanin-Derived Quinoline Alkaloids
The intriguing structure of the anticancer quinoline alkaloid camptothecin (8), isolated from the bark of Camptotheca acuminata (Nyssaceae) and the seeds of Nothapodytes foetida (Icacinaceae), was early-on shown to derive biogenetically from strictosidine (7), on the basis of both chemical considerations [43] and biosynthetic studies [44]. The major breakthrough towards the synthesis of the camptothecin chromophore was the observation of the facile oxidative rearrangement of indoloquinolizidine-derived lactams (e.g., 89) to the corresponding quinoloindolizidinones (e.g., 90), through a ring-opened intermediate (e.g., 91) [45]. A biomimetic total synthesis of camptothecin, using tryptamine (5) and secologanin
N
110
3 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Indolemonoterpene Alkaloids O N
(i) m-CPBA, C6H6 (ii) PPh3, AcOH
H N H
H N
H
N
+
HO
COOCH3
H3COOC
71 Tabersonine
N
N N
+
H3COOC HO
84
85
O N H COOCH3 86 HO (5% from 71)
H2, Pt, MeOH N
N
O
AlLiH4,THF
N H 9 Rhazinilam
Scheme 3.21
O
(i) KOH, MeOH, H2O (ii) HCl, H2O, Reflux
N H OH 88
N O N H HO COOCH
3
87
Biomimetic conversion of tabersonine (71) into rhazinilam (9).
(6) as starting materials, was recently developed on the basis on this rearrangement [46]. When performed at pH 4, the condensation of 5 and 6 gave a 3 : 2 mixture of vincoside (27) and strictosidine (7). Upon heating in alkaline medium, vincoside (27) was converted into the corresponding lactam (92), which was also obtained, together with its epimer at C3, starting from the crude vincoside–strictosidine mixture. Sodium periodide oxidation, performed after protection of the sugar moiety by acetylation, gave the expected ring-opened intermediate 91. Rearrangement of 91 in alkaline medium afforded the protected glycoside 90, possessing the same carbon skeleton as camptothecin. Conversion of the quinolone into the corresponding quinoline was ensured by successive chlorination with thionyl chloride, catalytic hydrogenolysis and hydrogenation, and re-aromatization by use of DDQ (2,3-dichloro-5,6-dicyano-1,4-benzoquinone) to afford 93. Deprotection of the sugar moiety, followed by treatment with β-glucosidase, gave the lactol 94, which was oxidized to the corresponding lactone 95. Finally, oxidation at C20 by oxygen in the presence of cupric chloride afforded the desired alkaloid camptothecin (8) (Scheme 3.22).
3.5 Biomimetic Synthesis of Dimeric Indolomonoterpene Alkaloids 3.5.1 Anhydrovinblastine and the Anticancer Vinblastine Series
The presence in only minute amounts (circa 5–50 g per ton of dried Catharanthus roseus aerial parts) and the clinical importance of the dimeric alkaloids vinblastine (4) and vincristine (11), currently used in cancer chemotherapy, strongly stimulated synthetic research toward their biomimetic synthesis, starting from the two monomeric units catharanthine (70) and vindoline (96), which can be isolated in relatively high yield (circa 1%) from the plant material. This dimerization sequence
H
H
Scheme 3.22
50%
O
H
O
H
H
H
N
H
N
94
N
89%
PCC CH2Cl2
H
O
H
O
60%
O2,CuCl2, DMF
73%
DDQ, Dioxane
N
N
93
N
N
70%
O
O
H OGlc(OAc)4
O
SOCl2, DMF 0°C
OGlc(OAc)4
O
O
N
O H OGlc(OAc)4
O
H
N
OH O 8 Camptothecin 95
H
H OGlc(OAc)4
O
H
N
H
89
N H H
O
N
H
N
90
N H H
O
Ac2O, Pyridine
H OH
N H H
78%
OGlc
O
O
65%
92
H
N
Et3N, MeOH
N H H
O
O
O
OGlc(OAc)4
Raney Ni, H2, EtOH
91 (64% from 92)
H OGlc(OAc)4
O
O
Glc
N O H
O
NEt3 MeOH
Biomimetic synthesis of camptothecin (8).
(i) MeONa, MeOH (ii) b -Glucosidase H2O, pH = 5
N
N
NaIO4, MeOH-H2O
Cl
NH
27 Vincoside
H H3COOC
N H H
3.5 Biomimetic Synthesis of Dimeric Indolomonoterpene Alkaloids 111
3 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Indolemonoterpene Alkaloids
112
was successfully conducted by the group of Pierre Potier, using a modification of the classical Polonovski fragmentation reaction. Indeed, treating a N-oxide with trifluoroacetic anhydride, instead of acetic anhydride in the initial sequence, permitted blocking of the reaction at the iminium stage [47, 48]. Thus, catharanthine N-oxide, prepared by treatment of catharanthine (70) with p-nitroperbenzoic acid, was acylated by trifluoroacetic anhydride to generate the strongly electrophilic iminium 97. Subsequent nucleophilic attack by the electron-rich aromatic ring of vindoline (96), followed by reduction of the conjugated iminium, led to the desired anhydrovinblastine (98), having the same (16 S) configuration as the natural biologically active compounds, when the whole reaction sequence was performed at −50 ◦ C (Scheme 3.23) [49]. (i) p -Nitroperbenzoic acid (ii) (CF3CO)2O
N N 16 21 H H3COOC
OCOCF3 N N H COOCH3
70 Catharanthine N 97
(i) 96 (ii) Reduction − 50°C
N H H3COOC
N H3CO
N 16' N H H3COOC H3CO
H OCOCH3 N CH3OH COOCH3
N H OCOCH3 N CH3 COOCH3 HO
98 Anhydrovinblastine
96 Vindoline
Scheme 3.23
Biomimetic synthesis of anhydrovinblastine (98).
In contrast, when the reaction temperature was allowed to rise to 0 ◦ C, only the biologically inactive (16 R)-epimer isoanhydrovinblastine (99) was obtained, due to conversion of the kinetic iminium 97 into the thermodynamic conformer 100 (Scheme 3.24). N
N 0°C
N H 97
COOCH3
N H 100
(i) 96 (ii) Reduction 0°C
COOCH3
N N H
COOCH3 N
99 H3CO
Scheme 3.24
H N OCOCH3 CH3 COOCH3 HO
Formation of isoanhydrovinblastine (99).
More recently, an efficient alternative method for coupling the two monomeric units present in anhydrovinblastine (98) and vinblastine (4) with an excellent stereocontrol of the C16 chiral center has been developed in the group of Boger,
3.6 Conclusion
113
based on the use of a FeCl3 oxidation of the catharanthine unit in the presence of a non-nucleophilic cosolvent, such as 2,2,2-trifluoroethanol [50].
3.5.2 Strellidimine
Another type of dimerization mechanism is involved in the formation of strellidimine (101), a dimeric ellipticine alkaloid, isolated from the bark of the African Loganiaceae Strychnos dinklagei, together with several monomeric ellipticine derivatives, including 3,14-dihydroellipticine (102) and 10-hydroxyellipticine (103) [51]. The biomimetic synthesis of strellidimine, performed to confirm the structure of the natural alkaloid, involved first the oxidation of 10-hydroxyellipticine (103) to the corresponding highly electrophilic quinone-imine 104, upon treatment with hydrogen peroxide in the presence of horseradish peroxidase (HRP). Nucleophilic attack by the basic nitrogen atom of 3,14-dihydroellipticine (102) followed by rearrangement afforded the desired bisindole alkaloid strellidimine (101) in almost quantitative yield (Scheme 3.25) [52]. H N H N
102 3, 14-Dihydroellipticine N HO
H2O2 HRP
N H
Scheme 3.25
N H
N
O
O N H
103 10-Hydroxyellipticine
H N
N
104
N
N H
N O N H
N H
101 Strellidimine
Biomimetic synthesis of strellidimine (101).
3.6 Conclusion
In conclusion, the complex structures of natural compounds, particularly well illustrated by the monoterpene indole alkaloids, in connection with the versatile reactivity of both the indole nucleus and the iridoid moiety, which includes two carbon–carbon double bonds and two masked aldehyde groups, have stimulated the imagination of synthetic chemists and biochemists. The biomimetic syntheses of these compounds permit us to better understand and rationalize the mechanisms involved in their formation and remain a complex challenge for some of their representatives. In that way, Nature is a major source of inspiration for organic chemists, both in terms of reaction mechanisms and structural diversity.
114
3 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Indolemonoterpene Alkaloids
References 1. (a) Saxton, J.E. (ed.) (1983) Indoles: The
2.
3.
4.
5.
6. 7.
8.
9. 10.
11. 12. 13.
14.
Monoterpenoid Indole Alkaloids, The Chemistry of Heterocyclic Compounds, vol. 25, Part 4, John Wiley & Sons, Ltd, Chichester; (b) Tillequin, F., Michel, S., and Seguin, E. (1993) in Alkaloids and Sulphur Compounds (eds P.M. Dey, J.B. Harborne, and P.G. Waterman), Methods in Plant Biochemistry, vol. 8, Academic Press, London, pp. 309–371. Atta-ur-Rahman and Basha, A. (1983) Biosynthesis of Indole Alkaloids, Clarendon Press, Oxford. Takayama, H. and Sakai, S.-I. (1998) in The Alkaloids, Chemistry and Pharmacology, vol. 50 (ed. G.A. Cordell), Academic Press, New York, pp. 415–452; (b) Scholz, U. and Winterfeldt, E. (2000) Nat. Prod. Rep., 17, 349–366. Bick, I.R.C. and Hai, M.A. (1985) in The Alkaloids, Chemistry and Pharmacolog, vol. 24 (ed. A. Brossi), Academic Press, New York, pp. 113–151. (a) Nagakura, N., R¨uffer, M., and Zenk, M.H. (1979) J. Chem. Soc. Perkin Trans. 1, 2308–2312; (b) Loris, E.A., Pahjikar, S., Ruppert, M., Barleben, L., Unger, M., Sch¨ubel, H., and St¨ockigt, J. (2007) Chem. Biol., 14, 979–985. Le Men, J. and Taylor, W.I. (1965) Experientia, 21, 508–510. Bick, R.I.C., Hai, M.A., and Preston, N.W. (1979) Heterocycles, 12, 1563–1565. Kburz, R., Sch¨opp, E., Bick, R.I.C., and Hesse, M. (1981) Helv. Chim. Acta, 64, 2555–2561. Mirand, C., Massiot, G., and L´evy, J. (1982) J. Org. Chem., 47, 4710–4711. Stevens, R.V. and Kenney, P.M. (1983) J. Chem. Soc., Chem. Commun., 384–386. Delpech, B. and Khuong-Huu, Q. (1978) J. Org. Chem., 43, 4898–4900. Gribble, G.W. and Barden, T.C. (1985) J. Org. Chem., 50, 5900–5902. Darbre, T., Nussbaumer, H.-J., and Borschberg, H.-J. (1984) Helv. Chim. Acta, 67, 1040–1052. (a) Borschberg, H.-J. (1991) Chimia, 45, 329–341; (b) Borschberg, H.-J. (1992) in Studies in Natural Products
15. 16. 17. 18.
19.
20.
21.
22. 23.
24.
25.
26.
27. 28.
Chemistry, vol. 11 (ed. Atta-ur-Rahman Elsevier, Amsterdam, pp. 277–334; (c) Borschberg, H.-J. (1994) in Indoles: The Monoterpenoid Indole Alkaloids, The Chemistry of Heterocyclic Compounds, vol. 25, Part 4, Supplementary Volume (ed. J.E. Saxton John Wiley & Sons, Ltd, Chichester, pp. 15–56; (d) Borschberg, H.-J. (1996) in The Alkaloids, Chemistry and Pharmacology, vol. 48 (ed. G.A. Cordell Academic Press, New York, pp. 192–248; (e) Borschberg, H.-J. (2005) Curr. Org. Chem., 9, 1465–1491. Burkard, S. and Borschberg, H.-J. (1989) Helv. Chim. Acta, 72, 254–263. Beerli, R. and Borschberg, H.-J. (1991) Helv. Chim. Acta, 72, 110–116. Burkard, S. and Borschberg, H.-J. (1991) Helv. Chim. Acta, 74, 275–289. Galli, R., Dobler, M., G¨uller, R., Stahl, R., and Borschberg, H.-J. (2002) Helv. Chim. Acta, 85, 3400–3413. Brown, R.T. (1980) in Indole and Biogenetically Related Alkaloids (eds J.D. Phillipson and M.H. Zenk), Academic Press, London, pp. 171–184. Brown, R.T. and Chaple, C.L. (1974) J. Chem. Soc., Chem. Commun., 740–742. Brown, R.T., Chaple, C.L., Platt, R., and Spencer, H. (1974) J. Chem. Soc., Chem. Commun., 929–930. Brown, R.T. (2000) J. Indian Chem. Soc., 77, 609–616. Brown, R.T., Dauda, B.E.N., Pratt, S.B., and Richards, P. (2002) Heterocycles, 56, 51–58. Brown, R.T., Pratt, S.B., and Richards, P. (2000) Tetrahedron Lett., 41, 5627–5630. Binns, F., Brown, R.T., and Dauda, B.E.N. (2000) Tetrahedron Lett., 41, 5631–5635. Brown, R.T., Dauda, B.E.N., Jameson, S.B., and Santos, C.A.M. (2000) Tetrahedron Lett., 41, 8861–8865. Martin, S.F., Chen, K.X., and Eary, C.T. (1999) Org. Lett., 1, 79–81. Ito, M., Clark, C.W., Mortimore, M., Goh, J.B., and Martin, S.F. (2001) J. Am. Chem. Soc., 123, 8003–8010.
References 29. (a) Knight, S.D., Overman, L.E.,
30.
31.
32.
33.
34.
35. 36.
37.
38.
and Pairaudeau, G. (1993) J. Am. Chem. Soc., 115, 9293–9294; (b) Knight, S.D., Overman, L.E., and Pairaudeau, G. (1995) J. Am. Chem. Soc., 117, 5776–5788; (c) Anet, F.A.L. and Robinson, R. (1953) Chem. Ind. (London), 245. Deiters, A., Chen, K., Eary, C.T., and Martin, S.F. (2003) J. Am. Chem. Soc., 125, 4541–4550. (a) Husson, A., Langlois, Y., Riche, C., Husson, H.-P., and Potier, P. (1973) Tetrahedron, 29, 3095–3098; (b) Besseli´evre, R. and Husson, H.-P. (1976) Tetrahedron Lett., 17, 1873–1876; (c) Husson, H.-P. (1980) in Indole and Biogeneticallty Related Alkaloids (eds J.D. Phillipson and M.H. Zenk), Academic Press, London, pp. 185–200. (a) Bennasar, M.-L., Vidal, B. and Bosch, J. (1993) J. Am. Chem. Soc., 115, 5340–5341; (b) Bennasar, M.-L., Vidal, B., and Bosch, J. (1996) J. Org. Chem., 61, 1916–1917; (c) Bennasar, M.-L., Vidal, B., and Bosch, J. (1997) J. Org. Chem., 62, 3597–3609. Andriantsiferana, M., Besseli´evre, R., and Husson, H.-P. (1977) Tetrahedron Lett., 18, 2587–2590. Langlois, Y., Langloisy, N., and Potier, P. (1975) Tetrahedron Lett., 16, 955–958. Potier, P. and Janot, M.M. (1973) C. R. Acad. Sci. (Paris), 276, 1727–1728. Bennasar, M.-L., Zulaica, E., Alonso, Y., Mata, I., Molins, E., and Bosch, J. (2001) Chem. Commun., 1166–1167. Scott, A.I., Lee, S.L., Wan, W., Hirata, T., Gu´eritte, F., Baxter, R.L., Nordlov, H., Dorschel, C.A., Mizukawi, H., and Mackenzie, N.E. (1981) Heterocycles, 20, 1257–1274. (a) Kuehne, M.E., Kirkemo, C.L., Matsko, T.H., and Bohnert, J.C. (1980) J. Org. Chem., 45, 3259–3265; (b) Kuehne, M.E., Okuniewicz, F.J., Kirkemo, C.L., and Bohnert, J.C. (1982) J. Org. Chem., 47, 1335–1343; (c) Kuehne, M.E., Bohnert, J.C., Bornmann, W.G., Kirkemo, C.L., Kuehne, S.E., Seaton, P.J., and Zebovitz, T.C. (1985) J. Org. Chem., 50, 919–923; (d) Kuehne, M.E. and
39.
40. 41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
Podhorez, D.E. (1985) J. Org. Chem., 50, 924–929; (e) Kuehne, M.E. and Zebovitz, T.C. (1987) J. Org. Chem., 52, 4331–4339; (f ) Kuehne, M.E. and Pitner, J.B. (1989) J. Org. Chem., 54, 4553–4569. Kuehne, M.E., Bornmann, W.G., Earle, W.G., and Marko, I. (1986) J. Org. Chem., 51, 2913–2927. Wenkert, E. and Wickberg, B. (1965) J. Am. Chem. Soc., 87, 1580–1589. L´evy, J. (1971) Proc´ed´e de pr´eparation de la vincamine naturelle a` partir de la tabersonine et deriv´es indoliques nouveaux, Patent BE 7, 616,28 [to Omnium Chimique, SA]. Hugel, G., Gourdier, B., L´evy, J., and Le Men, J. (1980) Tetrahedron, 36, 511–513. (a) Hutchinson, C.R., O’Loughlin, G.J., Fraser, S.B., and Brown, R.T. (1974) Chem. Commun., 928; (b) Winterfeldt, E. and Radunz, H. (1971) J. Chem. Soc., Chem. Commun., 374–375. (a) Hutchinson, C.R., Heckendorf, A.H., Sraughn, J.L., Daddona, P.E., and Cane, D.E. (1979) J. Am. Chem. Soc., 101, 3358–3369; (b) Hutchinson, C.R. (1981) Tetrahedron, 37, 1047–1065. (a) Warneke, J. and Winterfeldt, E. (1972) Chem. Ber., 105, 2120–2125; (b) Boch, M., Korth, T., Nelke, J.M., Pike, D., Radunz, H., and Winterfeldt, E. (1972) Chem. Ber., 105, 2126–2142; (c) Krohn, K. and Winterfeldt, E. (1975) Chem. Ber., 108, 3030–3042; (d) Hutchinson, C.R., Hsia, M.-T.S., Heckendorf, A.H., and O’Loughlin, G.J. (1976) J. Org. Chem., 41, 3493–3494. Brown, R.T., Jianli, L., and Santos, C.A.M. (2000) Tetrahedron Lett., 41, 859–862. Ahond, A., Cav´e, Ad., Kan-Fan, C., and Potier, P. (1970) J. Chem. Soc., Chem. Commun., 517. Potier, P. (1980) in Indole and Biogenetically Related Alkaloids (eds J.D. Phillipson and M.H. Zenk Academic Press, London, pp. 159–169. Langlois, N., Langlois, Y., Gu´eritte, F., and Potier, P. (1976) J. Am. Chem. Soc., 98, 7017–7024. (a) Ishikawa, H., Colb, D.A., Seto, S., Va, P., Tam, A., Kakei, H.,
115
116
3 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Indolemonoterpene Alkaloids Rayl, T.J., Hwang, I., and Boger, D.L. (2009) J. Am. Chem. Soc., 131, 4904–4916; (b) Va, P., Campbell, E.L., Robertson, W.M., and Boger, D.L. (2010) J. Am. Chem. Soc., 132, 8489–8495. 51. (a) Michel, S., Tillequin, F., and Koch, M. (1980) Tetrahedron Lett., 21,
4027–4030; (b) Michel, S., Tillequin, F., Koch, M., and Ak´e-Assi, L. (1982) J. Nat. Prod., 45, 489–494. 52. Michel, S., Tillequin, F., and Koch, M. (1987) J. Chem. Soc., Chem. Commun., 229–230.
117
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids Timothy R. Welch and Robert M. Williams
4.1 Introduction
Countless secondary metabolic indole alkaloids produced in both marine and terrestrial fungi are derived from tryptophan. Our crude attempts to synthesize some of the vast array of structurally diverse natural alkaloids only serves to showcase the efficiency and elegance with which Nature is able assemble the same molecules. Still, we strive to mimic and in turn better understand the mechanisms inside the cell that are able to produce molecular architecture of such synthetic complexity. Moreover, it has often been found advantageous to exploit Nature’s evolutionary creative design of alkaloids in search of new compounds of therapeutic potential. The rich subclass of biologically active, tryptophan-derived dioxopiperazines found in nature has inspired medicinal chemists to use the dioxopiperazine core in drug design efforts to mimic the interactions of natural peptides while reducing susceptibility to metabolic amide bond cleavage. Furthermore, a dioxopiperazine should pay a small entropic penalty upon binding to a target in comparison to an analogous peptide, as a direct consequence of the reduced conformational mobility inherent in the dioxopiperazine ring. In this chapter, we present a brief review of a select group of (partly) biomimetic syntheses of tryptophan-derived dioxopiperazine alkaloids (Figure 4.1). In most of these syntheses, a single step or key transformation has been deemed to constitute the ‘‘biomimetic’’ aspect of that particular work. As the actual biosynthetic pathways to most, if not all, of the alkaloid natural products covered here are either unknown or known only in part we have attempted, where appropriate, to point out the particular biomimetic step or transformation.
4.2 Prenylated Indole Alkaloids
Birch, Wright, and Russell first isolated brevianamide A from Penicillium brevicompactum in 1969 [1–3]. Several years later, Birch and coworkers determined that brevianamide A was biosynthetically derived from tryptophan, proline, and Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids
118
O H
N H O
N H
O
Me Me O
Me Me N OH H N O H H N O
N N
O
HN
N H brevianamide F
HO
N Me H Me okaramine N
spirotryprostatin B Section 4.2.1
H
Me H O Me N N
N N Me H O H Me
OH
gypsetin Section 4.2.2
H Me Me N H OH N N
Cl Cl
O Me Me H
NH
NH
OH N N
Me O Me
O (+)-versicolamide B
malbrancheamide
O
N O ent-alantrypinone
Section 4.2.3
Figure 4.1
O Me N
HN
O N HO H S S N Me O OH (+)-gliotoxin Section 4.3.1
Section 4.3
Representative molecules discussed in this chapter.
mevalonic acid through feeding experiments (Scheme 4.1) [4]. Furthermore, Birch showed that radiolabeled brevianamide F was incorporated into 5. It was postulated at the time and later supported with experimental evidence by Williams and coworkers that deoxybrevianamide E was also a biosynthetic precursor [5]. 14
CO2H
CH2
NH2
N H
3
H
[2-14C]-D,L-tryptophan (1)
3
Me OH N H
H CO2H
[5-3H]-L-proline (2)
14
CH2
O
O
[2-14C]-D,L-mevalonic acid lactone (3)
H
O N H 14 CH2 N H
OMe H O
N H brevianamide F (4)
Scheme 4.1
N H
Me(14C) H
HO N N O
(+)-brevianamide A (5)
O HH
3 3
H
N H N O MeH Me
N H deoxybrevianamide E (6)
Proposed biosynthesis of the brevianamides.
Since these early studies, Williams has developed a proposal for the biosynthesis of the brevianamides, with that of brevianamide E shown in Scheme 4.2. Derived from tryptophan and proline, 6 was proposed to undergo oxidation to hydroxyindolenine 7. Irreversible nucleophilic ring closure was supposed to lead to brevianamide E (8), supported by incorporation of [8-3 H2 ]6 into 8 in significant radiochemical yield [5].
4.2 Prenylated Indole Alkaloids
3 3
O H
H
H
H
O
N N H Me Me
O
N
H H
H [ox]
N H Me Me
HO
N H
Scheme 4.2
O
HO
N
nucleophilic
O
addition
N
deoxybrevianamide E (6)
119
N H
N Me O
H
Me
brevianamide E (8)
7
Proposed biosynthesis of brevianamide E.
4.2.1 Dioxopiperazines Derived from Tryptophan and Proline
Brevianamide F (12) was first isolated in 1972 and is one of the simplest tryptophan-derived dioxopiperazine natural products. It is readily synthesized through amino acid coupling of N-Boc-tryptophan (9) to proline ethyl ester (10), Boc-deprotection, and ring closure, in modest overall yield (Scheme 4.3) (R.M. Williams, unpublished results). H
CO2H
EtO2C
NHBoc + EtO2C N H
9
HATU, i Pr2NEt N H 10
O
MeCN
1. TFA, CH2Cl2
O H
56%
62% 11
1:1 = cis :trans
N N H
NHBoc 2. 2-OH-pyridine PhCH3 N H
Scheme 4.3
N
H O
N H brevianamide F (12)
Biomimetic synthesis of brevianamide F.
Brevianamide F lacks only the reverse prenyl group found in deoxybrevianamide E, which has been synthesized by Kametani and coworkers en route to brevianamide E (Scheme 4.4) [6]. N-Benzyloxycarbonyl-l-proline (13) was subjected to Schotten–Baumann conditions with dimethyl aminomalonate to give amide 14. Debenzyloxycarbonylation of 14 followed by heating with catalytic 2-hydroxypyridine effected cyclization to dioxopiperazine 15 in 93% yield. Condensation with indole 16 gave a separable mixture of diastereomers, individually hydrolyzed to the corresponding free acids (17). Heating of the desired diastereomer in dioxane gave deoxybrevianamide E (18) and its epimer (19) in 29 and 55% yield, respectively. Irradiation of methanolic 18, containing Rose Bengal in the presence of oxygen, followed by addition of dimethyl sulfide, resulted in the biomimetic hydroxylation of deoxybrevianamide E, furnishing brevianamide E (8) and 20 as a separable mixture of diastereomers. Nineteen years later, a more efficient synthesis of brevianamide E was completed by Danishefsky and coworkers [7]. The synthesis commenced with C3 chlorination of the known phthaloylated tryptophan derivative 21, followed by addition of fresh prenyl-9-borabicyclo[3.3.1]nonane (prenyl-9-BBN) to the resultant 3-chloroindolenine (Scheme 4.5). Hydrazinolysis in ethanol provided amino ester
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids
120
CO2Me HO2C
H CO2Me CO2Me CO2Bz 1. Pd/C, H2 N O 2. 70 °C, 2-OHH pyridine (cat) 14 93%
H2N
CO2Bz N
69% 13 HO2C O 1. NaH, ∆ 2. NaOH 22%
HN Me H O Me
H HO Rose Bengal
N HN
HN Me H O Me
hn, O2; Me2S
Scheme 4.4
H
1. t BuOCl, Et3N THF, −78 °C
CO2Me
2. Me
9-BBN
H
CO2Me 1. Boc-proline,
NH2 Me Me
HN
95%
BOPCl
HN
HN 2. TFA; NH3, MeOH
23
H
H
Me
21
H
19
HO
O
N N Me H O Me
+ N H
brevianamide E (8)
N HN Me H O Me
HN
O
N N Me H O Me
O
bis(epi)brevianamide E (20)
Kametani’s total synthesis of deoxybrevianamide E and brevianamide E.
NPhth
N H
N H
63% 8:20 = 2:1
deoxybrevianamide E (18)
+
HN Me H O Me deoxybrevianamide E (18)
18:19 = 1:1.9
O
H
H
HN
17
N MeMe H 16
O N
84%
NMe2
15
H
dioxane, ∆
N HN
H CO2Me O HN + N O H
HN
CO2Me
52%
CO2Me
N2H4, EtOH, rt
65%
NPhth Me Me 22
H HO N H
O
N HN Me H O Me
N N Me H O Me
brevianamide E (8)
DMDO CH2Cl2, acetone
76%
O
H HO
deoxybrevianamide E (18) N H
O
N N Me H O Me
bis(epi)brevianamide E (20)
Scheme 4.5
Danishefsky’s total synthesis of brevianamide E.
23 in 65% yield, which was coupled to N-Boc-l-proline, deprotected, and cyclized to afford deoxybrevianamide E (18) in 52% yield. Compound 18 was elaborated to brevianamide E (8) and bis(epi)brevianamide E (20) in a ratio of ∼1 : 5 upon treatment with dimethyldioxirane (DMDO) in a biomimetic oxidative cyclization sequence reminiscent of the original Kametani work discussed above. The structural similarities between deoxybrevianamide E and the then newly isolated natural products tryprostatin A and B did not escape notice of Danishefsky
4.2 Prenylated Indole Alkaloids
121
and coworkers. While prenylation failed in attempts to use a reverse prenylborane nucleophile directly as in the method used to synthesize 22 above, a solution was found in treating chloroindolenine 24 with tri(n-butyl)prenylstannane and BCl3 to afford the desired prenyl functionality at C2 in excellent yield (Scheme 4.6). Phthalimide deprotection, peptide coupling, Boc-deprotection, and cyclization were achieved to afford tryprostatin B (30) in 43% overall yield [7]. H HN
CO2Me
NPhth
t BuOCl, CCl4 Et3N
21 Me CH2Cl2 + BCl3
n-Bu3Sn
Me 25
Cl N
24
H HN
CO2Me
O Me
27
N Boc
H 2. NH3, MeOH
Me 29
Scheme 4.6
Me Me
82% O N
1. TMSI, MeCN
67%
28
Me
HN
HN O Me
CO2Me
NH2
HN
MeOH, CH2Cl2
Me
CO2Me
H
CO2Me
NPhth N2H4 • H2O
HN 83%
HN
94%
H
NPhth
Me Me BCl2 26
N-Boc-L-Pro-F CH2Cl2, NaHCO3
H
H
Me tryprostatin B (30)
Biomimetic total synthesis of tryprostatin B.
Danishefsky and coworkers also completed the total synthesis of the spirooxindole spirotryprostatin B [8]. l-Tryptophan methyl ester was converted into the oxindole derivative 32, followed by addition of prenyl aldehyde under basic conditions to afford an inseparable four-component mixture of spirooxindoles (33–36, Scheme 4.7). Peptide coupling and subsequent treatment of the mixture with lithium bis(trimethylsilyl)amide (LHMDS) followed by selenylation presumably gave phenyl selenide mixture 38. Oxidative elimination produced a mixture from which 39 was separated and elaborated to spirotryprostatin B (40) via Boc-deprotection and base-induced cyclization. Danishefsky took a markedly different approach in the synthesis of spirotryprostatin A [9]. A potentially biomimetic Pictet–Spengler reaction of tryptophan derivative 42 with thioaldehyde 41 as a masked isoprene equivalent gave the desired cis-tetrahydrocarboline (43) with marginal selectivity (Scheme 4.8). N-Bromosuccinimide (NBS)-mediated oxidative rearrangement proceeded via intermediate 44 to the oxindole was followed by deprotection of the carbamate to give amine 45. The modest yield of the sequence (57%) reflects the susceptibility of the oxindole to electrophilic aromatic bromination under the spiro-rearrangement conditions. Peptide coupling and Troc-deprotection resulted in cyclization to the dioxopiperazine, after which oxidation and sulfoxide elimination revealed the prenyl group to afford selectively spirotryprostatin A (48). The synthesis of notoamide J was completed by Williams and coworkers, starting with the Boc-protection of 7-hydroxyindole (Scheme 4.9) [10]. Chlorination at C3 was followed by reverse prenylation of the resultant 3-chloroindolenine 51 to afford
122
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids H CO2Me
H CO2Me
NH2•HCl
NH2•HCl
DMSO, 12N HCl
O
AcOH, PhOH
N H
N H
31 H CO Me 2
O HN
O HN
+
NH Me
NH Me
Me
+
HN
O CO2H
HN
NBoc
Me
BOPCl
HN
NH Me Me
35
LHMDS
N O Boc Me
36 O PhSe CO2Me
H CO Me 2 N
+
NH
H CO Me 2
O
Me
34
H
Et3N, py.
H CO Me 2
Me
33
Me
73% (2 steps) 2:3:3:3
32
O
H CO Me 2
CHO
Me
HN N
PhSeCl
Me 37
DMDO
N O Boc Me
Me 38 O
O
CO2Me
HN N
Me 39
Scheme 4.7
N O Boc Me
N
Me 1. TFA, CH2Cl2
N Me
2. Et3N
O
O
HN
7% (5 steps) spirotryprostatin B (40)
Danishefsky’s synthesis of spirotryprostatin B.
52. The corresponding gramine was prepared by treating 52 with formaldehyde and dimethylamine, and subsequent Somei–Kametani coupling and imine hydrolysis gave tryptophan derivative 54 in good yield. Protection of the free amine as the Boc-carbamate and ester hydrolysis gave 55, which was coupled to proline ethyl ester in the presence of O-(7-azabenzotriazol-1-yl)-N,N,N ,N -tetramethyluronium hexafluorophosphate (HATU) to afford amide 56. Cyclization to dioxopiperazine 57 was followed by a biomimetic oxidation sequence accompanied by pinacol-type rearrangement to the oxindoles notoamide J (58) and 3-epi-notoamide J (59) in a 2 : 1 separable mixture. 4.2.2 Dioxopiperazine Derived from Tryptophan and Amino Acids other than Proline
Corey and coworkers designed a succinct synthesis of okaramine N, featuring a Pd-promoted dihydroindoloazocine formation [11]. Readily available tryptophan derivative 60 was reduced to indoline 61 and subsequently prenylated via copper(I)-catalyzed alkylation with butyne 62 (Scheme 4.10). Treatment with
O
N 42 H
NH2
H CO2Me
H
Scheme 4.8
74%
57% (4 steps)
NBoc SPh
CO2Me
MeO
43
HN
O
47
N Me H Me 2:1, cis:trans
H
Danishefsky’s synthesis of spirotryprostatin A.
45
3.NaIO4, MeOH H2O 4. PhCH3, ∆
2. Zno, THF, MeOH
MeO
1. CH2Cl2, Et3N
MeCN, Et3N
2. (Boc)2O
1. TFA, 4Å CH2Cl2
TrocN + NH H SPh ClOC H 46 Me Me
H CO2Me
41 +
MeO
HN
MeO
PhS
Me Me O
N H
H
O Me
O
H
N
AcOH, H2O
NBS, THF
45%
EtOH, ∆
RhCl3.3H2O
MeO
Br
CO2Me TFA, CH Cl
O
O Me
N H
H
H
N
Me spirotryprostatin A (48)
MeO
HN
O
2 2 NBoc H SPh 57% (2 steps) N OH Me H Me 44
H
4.2 Prenylated Indole Alkaloids 123
N H
O
EtO2C
N H
47% 58:59 = 2:1
O
O N
HO
59%
Me Me
BocO
O H
BocO
N H
N
76%
O
H
N H 56
O
N H
+ HO
N H
NH2 Me Me
CO2Et
48%
Et3N, THF
Me
Me Me
N H 59
O H O
N H
N
53% cis:trans = 1:1
2. 2-OH-pyridine PhCH3
N H
N H 57
83%
N N Me H Me
O H
2. LiOH, THF/H2O
H O
52
1. Boc2O, 1M NaOH dioxane
BocO
HO
9-BBN
1. TFA, CH2Cl2
54
Cl Me
NHBoc Me Me
N
BocO
51
EtO2C
PBu3, MeCN, 2. 1 M HCl, THF
CO2Et Ph N 1. Ph
83%
DMF
O N H notoamide J (58)
HATU, i Pr2NEt MeCN
S
n Bu
2.
N H
Me Me
NMe2
50
1. LiOH, THF/H2O
53
N H
BocO
CH2Cl2
NHBoc Me Me
CO2H
BocO
70%
MeCN
NCS
Williams’ biomimetic synthesis of notoamide J.
N H 55
84%
AcOH
HNMe2, CH2O
49
Scheme 4.9
BocO
HO
Boc2O, DMAP
O
H
Me Me
124
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids
4.2 Prenylated Indole Alkaloids
CO2Me
NaBH3CN, AcOH
NHBoc
NH
1. CuCl, i Pr2NEt Me OAc 62 Me
CO2Me
60%
2. DDQ; H2, quinoline, Pd/C 87%
NHBoc
NH
60
61
CO2H
CO2Me 1. SOCl2 NHBoc
N Me Me
N
2. LiOH; Fmoc-Cl
Me Me
81%
63
Scheme 4.10
125
NHFmoc 64
Synthesis of the N-reverse-prenylated tryptophan derivative.
2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) effected the dehydrogenation of the indoline, and the resulting tryptophan derivative 63 was deprotected, saponified, and reprotected as the N-Fmoc derivative (64). The synthesis of okaramine N was completed through reductive amination of 3-methyl-buten-2-al onto l-tryptophan methyl ester, followed by coupling of the resultant product (65) with acid 64 to form the desired tetracycle (Scheme 4.11). Treatment of 66 with Pd(OAc)2 provided the eight-membered ring (67) in modest yield (38%). The free amine obtained upon Fmoc cleavage underwent cyclization to form dioxopiperazine 68 in 95% yield. A potentially biomimetic oxidative cyclization was effected by treating 68 with N-methyltriazolinedione (MTAD), which reacted selectively with the N-unsubstituted indole subunit, and after photooxidation and Me Me N
Me Me CO2Me
NaBH4, MeOH
NH2
NH
CHO DCM;
31
CO2Me NH
Me
HN
BOPCl, 64
FmocHN MeO2C N
70%
NH Me 66 N
N
N Me Me H 67
Scheme 4.11
Me
Me Me
Me Me
Et2NH O
O
Me
65
FmocHN MeO2C N
Pd(OAc)2, O2
95%
O H
H N N
MTAD; O2, sunlamp, methylene blue; H O
Me2S; ∆
N Me H Me 68
Completion of the total synthesis of okaramine N.
70%
Me Me N OH H N O H H O N
N Me H Me okaramine N (69)
38%
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids
126
NaOH, Boc2O,
CO2Me NH
(n-Bu)4NHSO4
91%
NH2
N-PSP, PPTS
CO2Me
93%
NHBoc
BocN
31
70
SePh
SePh BocN
+ CO2Me BocN
CO2Me N H Boc 72 (minor) endo
N Boc 71 (major) exo
Me Me MeOTf, 2,6-di(tbutyl)pyridine
Me Me
Me Me TMSI
Sn(n-Bu)3 BocN 60%
N Boc
CO2R
83%
18:1 (a:b) NaOH
THF/MeOH/H2O 98%
Scheme 4.12
NH
1. 74, BOPCl, Et3N
2. TMSI N CO2Me 34% H 75
73 (a) R =Me
NH
H O N O H
N
NH
Me Me amauromine (76)
74 (a) R=H
Danishefsky’s total synthesis of amauromine.
then reduction formed a hydroxylated octacycle that was directly converted into okaramine N (69) via thermolysis in 70% yield. Amauromine was synthesized as shown in Scheme 4.12, starting with the bis(Boc) protection of l-tryptophan [12]. Conversion into the selenide by reaction with N-phenylselenophthalimide (N-PSP) in the presence of pyridinium p-toluenesulfonate (PPTS) afforded a mixture of 71 and 72 (9 : 1). Photolysis of the mixture in the presence of prenyltri(n-butyl)tin produced a mixture of reverse prenylated pyrrolodinoindolines, the desired major diastereomer of which was separated by crystallization from hexanes. The Boc groups were globally cleaved using iodotrimethylsilane (TMSI) to afford the ester (75). Coupling of 75 with acid 74 gave the amide, which readily cyclized to the desired dioxopiperazine (76, amauromine) upon treatment with TMSI. Danishefsky and coworkers completed a total synthesis of gypsetin in the same fashion as their effort on brevianamide E [7]. The reverse prenylated amine, 23, was synthesized as shown above (Scheme 4.5). Boc protection and cleavage of the methyl ester afforded acid 77, which was coupled to amine 23, deprotected, and cyclized to dioxopiperazine 78 (Scheme 4.13). Treatment with DMDO effected the biomimetic oxidative conversion into the natural product gypsetin (79). 4.2.3 Bicyclo[2.2.2]diazaoctanes
In 1970, Sammes proposed a hetero-Diels–Alder cycloaddition to be the biosynthetic origin of the bicyclo[2.2.2]diazaoctane core found in brevianamides A and B (Scheme 4.14) [13]. Support for this proposal was observed upon treating dihydroxypyrazine 82 with dimethyl acetylenedicarboxylate (83) or with norbornadiene (84) to give cycloadducts 85 or 86, respectively (Scheme 4.15) [13].
4.2 Prenylated Indole Alkaloids
N H
H CO Me 2
1. t BuOCl, Et3N THF, −78 °C
NPhth
2. Me
H HN
Me
9-BBN 95%
H CO Me 2
CO2Me N2H4, EtOH, rt 65%
NPhth Me Me 22
HN
NH2 Me 23 Me
21 1. (Boc)2O, Et3N THF 94% 2. LiOH, MeOH, THF
H
O
Me Me NH
HN Me H Me O 78
HN
NH
DMDO
HO
H
40%
1. BOPCl, CH2Cl2
2. TFA, CH2Cl2 3. NH3MeOH 87% H CO2H
Me H O Me N N
N Me H N H MeO
CH2Cl2, acetone
HN
OH
NHBoc Me 77 Me
gypsetin (79) (+ 18% all syn-isomer + 20% double anti -isomer)
Scheme 4.13
Synthesis of gypsetin.
Me Me N Me
O N
Me
HO N N
OH O 80
81
Scheme 4.14
Proposed hetero-Diels–Alder formation of bicyclo[2.2.2]diazaoctanes. MeO2C
83
MeO2C HN
DMF, r.t. PhH2C HO
N
OH
N
Me
82
Scheme 4.15
CO2Me
CO2Me
O
CH2Ph O
N Me H 85 O HN
CH2Ph O
Me
NH 86
84
127
Sammes’ model study of proposed cycloaddition.
Williams and coworkers expanded on the pioneering work of Sammes with the following biosynthetic proposal for the brevianamides [5]. Deoxybrevianamide E (18) was thought to undergo oxidation to hydroxyindolenine 7, which could undergo a pinacol-type rearrangement to indoxyl 87 (Scheme 4.16). Subsequent two-electron oxidation and enolization to azadiene 88, followed by an intramolecular hetero-Diels–Alder reaction was, following from the original proposal of Sammes, envisioned to give the natural products (+)-brevianamide A and (+)-brevianamide
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids
128
O H
O H
N N H Me Me
H O
[ox]
N H
H N O H Me Me
HO
O H
N rearrangement
Me Me O
N H
N NH H
O
2e− ox enolization
N
deoxybrevianamide E (18)
7
O Me Me O
N
NH
N+ H
87
HN O Fe3+
OH N
O Me Me NH+
HN O Me Me H+ − N O H N O
O− 88a
88b
88
N H
OMe Me H HO N N
O (+)-brevianamide A (89)
Scheme 4.16
Me H OH N
Me H N O
N
O (+)-brevianamide B (90)
Biosynthetic proposal for the brevianamides.
B. The pseudo-enantiomorphic relationship between the two natural products was envisaged to arise from the equilibrium between conformers 88a and 88b, which undergo cycloaddition to give 89 and 90, respectively. Ab initio studies of the two transition states demonstrated that 88a is the more stable of the two, which is consistent with the observed product ratios of 89 and 90. The theoretical insights published by Domingo et al. lend support to the proposal of a biosynthetic intramolecular Diels–Alder reaction of intermediate 88 [14]. Unfortunately, experimental support for this pathway has been elusive to secure and thus it remains a speculative biogenetic construction. The total synthesis of d,l-brevianamide B demonstrated the first congruent application of a biomimetic Diels–Alder reaction used to form the bicyclo[2.2.2]diazaoctane core common to the series of prenylated dioxopiperazines to be discussed in this section [15]. 9-Epi-deoxybrevianamide E (19) was converted into the lactim ether (91) and oxidized to give the Diels–Alder precursor 92 (Scheme 4.17). Treatment with aqueous methanolic KOH induced tautomerization to azadiene 93, which underwent a potentially biomimetic intramolecular Diels–Alder cycloaddition to give a mixture of diastereomers (94 and 95, 2 : 1). Oxidation, pinacol-type rearrangement, and lactim ether deprotection of the minor diastereomer (95) afforded brevianamide B (90) in 65% overall yield from 96. This study was one of the first to experimentally support the biogenetic origin of the core bicyclo[2.2.2] ring system as arising via a dioxopiperazine that undergoes a net two-electron oxidation to an azadiene moiety.
4.2 Prenylated Indole Alkaloids H N
NH
O
H
O
N Me3OBF4
O
OMe
N
N Me H 19
N
O
H 2O
40%
N Me H 91
OMe
N KOH MeOH
N
O
DDQ
69%
Me
OMe
N
129
Me
N Me H 92
Me
N Me Me H 93
H N Me Me H
+ 60% 94:95 = 2:1
Scheme 4.17
MeO N N O 94 H Me N Me H MeO N N O 95
1. mCPBA 2. HCl, 99%
N Me Me H MeO HO N N O
H Me Me N H
1. NaOMe 2. HCl, 65%
96
O
HO N N
O (±)-brevianamide B (90)
Application of the proposed biomimetic Diels–Alder reaction.
Williams and coworkers also applied the intramolecular Diels–Alder reaction to the racemic synthesis of VM55599 [16]. The reverse-prenylated tryptophan derivative 97 was coupled to β-methyl-β-hydroxyproline ethyl ester 98 to afford dipeptide 99 (Scheme 4.18). N-Boc deprotection and cyclization afforded dioxopiperazine 100, and elimination with thionyl chloride gave enamide 101. Formation of the lactim ether (102) and treatment with aqueous KOH gave azadiene 103, which spontaneously suffered intramolecular Diels–Alder reaction to give a separable mixture of all four possible racemic diastereomers (104–107, 2.6 : 3.7 : 1.0 : 1.6, respectively). Cleavage of the lactim ether and diisobutylaluminum hydride (DIBAL-H) reduction gave VM55599 in 73% yield from 105. The synthesis of VM55599 allowed for assignment of the absolute stereochemistry of the molecule, which places the methyl group at the β-position of the proline residue syn- to the bridging isoprene moiety [16]. In stark contrast, the analogous methyl group of paraherquamide A is anti to the bridging isoprene unit. Scheme 4.19 outlines a possible unified biosynthesis for VM55599 and paraherquamide A, arising from dimethylallyl pyrophosphate (DMAPP), l-isoleucine, and l-tryptophan. If a Diels–Alder cycloaddition is to be invoked, approach of the isoprene moiety must occur from the same face as the methyl group on the proline ring for synthesis of VM55599, and from the opposite face to the methyl group to give paraherquamide A. The diastereofacial selectivity of the Diels–Alder reaction gave a preponderance of the syn-relative stereochemistry alpha to the gem-dimethyl group in both molecules. As VM55599 is a very minor metabolite of Penicillium sp. IMI332995, it is plausible that cyclization of 112→114 is preferred and further metabolization gives paraherquamide A, whereas the minor cycloaddition via 113 produces VM55599 as a dead-end shunt metabolite. As shown in Scheme 4.18
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids
130
TFA⋅HN EtO2C
CO2H BocHN
O
Me OH CO2Et
N
HO Me 98 BocHN
BOPCl
1. TFA
83%
N Me Me H
N
N Me Me H
97
O
Me OMe
N
N
O
Me3OBF4
Me
N Me H 100
OMe
N
NH
SO2Cl, py. 75%
2. 2-OH-pyr. ∆, 95%
Me O
N
NH
O
99 Me
Me OH O
KOH
O
N
72%
N Me H 101 H N Me Me H MeO N 104 N O H N Me Me H MeO N N
106 O
Me
N Me H 102 H N Me Me H
H Me
N Me H 103 1. HCl 2. PhCH3, ∆
MeO N N
105
Me
Me H
H Me
107
MeO N N
H N Me Me H HO N N
3. DIBAL-H 73%
Me H
VM55599 (108)
O H N Me Me H
Me
Me H
O 104:105:106:107 = 2.6 : 3.7 : 1.0 : 1.6
Scheme 4.18
Williams’ biomimetic total synthesis of VM55599.
Me Me PPO DMAPP (109)
H N
Me
Me
H N Me Me
H Me N
Me
H
CO2H
H2 N HO2C
Me NH2 L-Ile (110)
HO N N
OH N N
Me
X 113
X 112
L-Trp (111)
H
syn-selective [4+2]
H Me
O Me Me H
[ox] SAM DMAPP
NH
OMe N N
O
H Me
O
Me Me paraherquamide A (115)
Scheme 4.19
Me H
H Me N
OH N N
X 114 major
Proposed biosynthesis of paraherquamide A and VM55599.
H N Me Me H
Me
HO N N
VM55599 (108) minor
H
4.2 Prenylated Indole Alkaloids
131
above, the intrinsic diastereofacial bias of the Diels–Alder reaction is modest at best, giving a slight excess (1.47 : 1) of cycloaddition from the same face as the methyl group, and favoring the syn-relative stereochemistry to the extent of 2.4 : 1. Such observations suggest that the biosynthesis may rely on protein organization of the precyclization conformers to stereoselectively produce the syn-isomers. A report from Liebscher and coworkers showed promise in terms of improving the diastereoselectivity of the Diels–Alder cycloaddition, using neutral conditions to prepare the azadiene in contrast to the basic conditions employed by Williams [17]. Compound 118 was prepared by a Horner–Wadsworth–Emmons reaction of aldehyde 116 with phosphonate 117 (Scheme 4.20). Treatment of 118 with neat acetyl chloride for 20 days gave the Diels–Alder product 119 as a single diastereomer in 48% yield. (MeO)2
O P
O N
O
HN
CHO
117
Me N Me MOM
O
N Me MOM Me 118
116
Scheme 4.20
H O
N H
KOt Bu
H N Me Me H HO N N O 119 one diastereomer
N
H
AcCl 20 d
48%
Liebscher’s Diels–Alder work.
The precedent set by Liebscher’s work was applied to an asymmetric total synthesis of VM55599 [18]. The loss of stereochemistry observed at the proline methyl group in 101 above led Williams to employ a dehydrotryptophan derivative as the Diels–Alder azadiene precursor, allowing for the preparation of VM55599 in an enantioselective fashion. Williams has demonstrated that the β-methylproline residue of paraherquamide A and VM55599 are biosynthetically derived from l-Ile. In a biomimetic construct, l-Ile was converted into optically pure β-methylproline 120 using Hoffman–L¨offler–Freytag conditions in 45% yield (Scheme 4.21). S Me S H2N CO2H
Me
45%
L-Ile (110)
S Me S CO2Et N Boc 120
Me LiOH 96%
CO2H
N Boc 121
ClH3N
Me
N N O 123
Scheme 4.21
NaH, MeSCH2Cl
Me
CO2Me
62%
N O 122
88% O
SMe O
O HN
2. TFA; Et3N 3. PhCH3, ∆
CHO N MeMe MOM 116 NaHMDS 87%
1. EDCI, HOBt Et3N
MeS HO
Me
N N
H
O
N Me MOM Me 124
Key dioxopiperazine synthesis.
Me
N 1. MeI, NaHCO3, 92% 2. 50% aq. HCO2H, 70%
O
O
N H N H 125
Me Me
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids
132 Me O
AcCl
Me
N H
N
Me Me
H Me N
O H S N N
H Me
AcO
O
O AcO
+
N
14 days N H 125
H
N
H
O
N H
Me H
Me
Me
N
N H 126 H N Me Me H
128 (10%)
Me H O S H N N
129 (35%)
O
O
H Me N Me H
H N Me Me H
O H S N N
130 (0%) O
Scheme 4.22
Me Me
N H 127
DIBAL-H
Me Me
H N Me Me H
Me H O H S N N
68%
(–)VM55599 (108)
Me H O H S N N
131 (15%) O
Completion of the asymmetric total synthesis of (−)-VM55599.
Cleavage of the ethyl ester, peptide coupling to glycine methyl ester hydrochloride, Boc deprotection, and cyclization gave dioxopiperazine 122 in good overall yield. Protection of the secondary amide and subsequent condensation with aldehyde 116 gave an epimeric mixture of dioxopiperazines, which gave selectively (Z)-isomer 125 following deprotection and dehydration. Following Liebscher’s protocol, 125 was treated with acetyl chloride for 14 days, yielding a mixture of three diastereomers (Scheme 4.22). The reaction is thought to proceed by initial acylation to the O-acyl lactim 126, tautomerization to azadiene 127, which suffers intramolecular Diels–Alder reaction from three of the four possible diastereomeric transition states, followed by loss of acetate to give compounds 128–131. The major diastereomer 129 was treated with excess DIBAL-H to effect reduction to (−)-VM55599 (108). Interestingly, cycloadduct 130 was not observed from the cycloaddition reaction. Penicillium sp. produces paraherquamide A in large excess over VM55599 (>600 : 1), so it is surprising that this provocative biogenetic precursor to paraherquamide A is not observed in laboratory cycloaddition reactions. Despite the structural differences between the laboratory and biological Diels–Alder precursors, the intrinsic facial selectivity of the cyclization does not appear to mimic the bias toward paraherquamide stereochemistry one would expect given the observed product ratios of the fungal metabolites. As many of the bicyclo[2.2.2]diazaoctane natural products differ only at the substitution of the indole ring, a convergent approach utilizing a Fischer indole synthesis was undertaken in an alternative racemic synthesis of brevianamide B [19]. This work had the objective of validating other possible biosynthetic pathways
4.2 Prenylated Indole Alkaloids
O
S
S
EtO2C
H
133
Me
Me
n BuLi, CuI
H
EtO2C S
LiOH
S
Me
S
S
Me
BOPCl, i PrNEt
Me
77%
135 NH2 O
Me Me
Me H
N S
Me Me
NCS, AgNO3
O
88% O
HO2C S
134
NH2 N
99%
Me
O H O H
CONH2 N 136 H
O
O
76%
132
137
O
138:139 = 45:43
H
138 +
AlCl3
O
N
48% Me
Me O
OH N
EtOAc
O NH
O 140
Me
N O
1. PhNHNH2 2. ZnCl2, 58%
H Me N Me H OH N N
139 O
1. m CPBA 2. NaOH, 16%
O 141
Scheme 4.23
133
Me H OH N
Me
H N O
N
O (±)-brevianamide B (90)
Concise synthesis of brevianamide B.
to reach the oxidation state of the azadiene. In the present instance, the α-ketoamide species (138) was posed to serve as a surrogate for the possible biosynthetic oxidative deamination of the tryptophan moiety and coupling to a proline amide species (Scheme 4.23). Conjugate addition of carboxylate 133 to ketone 132 gave ester 134 in 76% yield. Saponification, peptide coupling with l-proline amide, and dithiane deprotection gave a mixture of the uncyclized amide 138 and dioxopiperazine 139. Aluminum trichloride was added to the mixture to give the Diels–Alder cycloadduct (140) in exclusively the anti-configuration, whereas mixtures of both the syn- and anti-cycloadducts were observed in previous syntheses of VM55599 and brevianamide B discussed above. A Fischer indole synthesis was completed by treatment of 140 with phenyl hydrazine followed by ZnCl2 , affording 141 in good yield. Oxidation and pinacol-type rearrangement of this known intermediate gave brevianamide B. While an appealing convergent approach towards the synthesis of other related bicyclo[2.2.2]diazaoctanes, the utility of this strategy is limited to those natural products containing the anti-stereochemistry observed in formation of compound 140, the remainder of which must be able to withstand the harsh conditions of the Fischer indole synthesis. The same biomimetic Diels–Alder disconnection was exploited in the total synthesis of stephacidin A [20]. Starting with reverse prenylated tryptophan derivative 142 (prepared in eleven steps from 6-hydroxyproline), dipeptide 144 was prepared through bis(2-oxo-3-oxazolindinyl)phosphinic chloride (BOPCl) mediated coupling
O
O
O
148
O
MeO N N
H N Me Me H
Me
Me
146
+
N H
N
O
N H Me Me
O H
BOPCl, i PrNEt2 54%
N 143 H·HCl
EtO2C
Synthesis of stephacidin A.
91%
PBu3, DEAD
Me Me
N H Me Me
142
Scheme 4.24
Me
Me
Me Me
NHFmoc
CO2H
HO
O 81%
149
144
O
MeO N N
Me Me
O
96%
HCl
NHFmoc
N
N H Me Me
Me3OBF4 Cs2CO3
H N Me Me H
Me Me
O
O H
EtO2C
N H
O
Me Me
N
O
O
N
N H
N
H O N N
H N Me Me H
86% 148:149 = 1:2.4
KOH, MeOH
145
N H Me Me
O H
O stephacidin A (150)
OMe
Me Me
N H Me Me
O H
95%
147
OH
H O
OH
134
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids
4.2 Prenylated Indole Alkaloids
with cis-3-hydroxyproline ethyl ester (143, Scheme 4.24). Fmoc deprotection resulted in the cyclization to dioxopiperazine 145, which upon treatment with tributyl phosphine and diethyl azodicarboxylate (DEAD) underwent Mitsunobu dehydration to give enamide 146. Formation of the lactim ether (147) was followed by intramolecular Diels–Alder reaction to give a mixture of epimers enriched with the syn-isomer (149, 2.4 : 1). Deprotection of 149 gave stephacidin A (150) in excellent yield. Williams found that the bicyclo[2.2.2]diazaoctane core could be accessed directly from compound 145 by treating it with excess PBu3 and DEAD, effecting the dehydration, tautomerization, and Diels–Alder reaction in one pot to afford stephacidin A and its epimer (Scheme 4.25) [21]. O H N H O Me Me
N H Me Me
145 Scheme 4.25
Me
N H O
OH
PBu3 DEAD
Me
H N Me Me H
O
40 °C 64%
+ epi
HO N N O stephacidin A (150) 150:epi = 2.4:1
Improved biomimetic synthesis of stephacidin A.
Myers and coworkers, in the course of their total synthesis of avrainvillamide, discovered that synthetic (−)-152 spontaneously dimerized to stephacidin B under several mild conditions, including addition of triethylamine or exposure to silica gel [22, 23]. Baran and coworkers later determined that stephacidin A could be readily converted into avrainvillamide through reduction to indoline 151 followed by Somei oxidation (Scheme 4.26). In accord with Myers’ observations, dimerization to stephacidin B occurred readily upon exposure to silica gel, triethylamine, or on evaporation from dimethyl sulfoxide (DMSO) [24–26]. Myers has further demonstrated that the observed biological activity of stephacidin B may be due to the formation of 152 from 153 in vivo [22]. Stephacidin A is of particular biogenetic interest, as both enantiomers have been isolated in nature: (+)-stephacidin A from Aspergillus ochraceus [27] and from a marine-derived Aspergillus sp. [28] and (−)-stephacidin A from terrestrial Aspergillus versicolor [29–31]. Operating under the assumption that the biosynthesis of stephacidin A proceeds through a common, achiral intermediate, Williams and coworkers have proposed notoamide S (154) as the point of divergence in the two biosyntheses [32]. Oxidation of 154 could give achiral azadiene 155, which is postulated to undergo intramolecular Diels–Alder reaction to give either (+)or (−)-stephacidin A depending on the enantiofacial selectivity of the reaction (Scheme 4.27). Tryptophan derivative 142 used in the above synthesis of stephacidin A was also employed in the biomimetic total synthesis of marcfortine C (164) [33]. Pipecolic acid derivative 156 was coupled to acid 142 to form amide 157, which underwent cyclization to the dioxopiperazine following Fmoc-deprotection to give 158 as
135
136
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids
Me Me O
H N Me Me H
Me Me O
NaBH3CN
H N Me Me H
AcOH
HO N N
HO N N
O
O
stephacidin A (150)
Me Me O
SeO2, H2O2
151
O N Me Me H
Me Me O
OH Me Me N H H O
SiO2
O
N
N O H O N
HO N N O avrainvillamide (152)
N
Me Me
N O Me O Me H
stephacidin B (153) Scheme 4.26
Biomimetic conversion of stephacidin A into stephacidin B. Me Me Aspergillus versicolor
O H N H
H O
N H Me Me
HO
Me
O
N
Me
H O N N
N O N
[ox]
N H Me Me
HO
Me
notoamide S (154)
H N Me Me H
O
(−)-stephacidin A, (−)-150
OH IMDA [ox]
Me
Aspergillus sp.
Me 155 (achiral)
H Me N Me H
Me O
O H N
N
O (+)-stephacidin A, (+)-150
Scheme 4.27
Proposed biosynthesis of (+)- and (−)-stephacidin A through notoamide S.
an inconsequential mixture of diastereomers (Scheme 4.28). The biomimetic Diels–Alder reaction preferred the syn-diastereomer 160 in 2.4-fold excess, as expected. Excess DIBAL-H selectively reduced the tertiary amide of 160, and amine salt formation followed by a biomimetic oxidative rearrangement gave marcfortine C (164). Malbrancheamide and malbrancheamide B were synthesized via the biomimetic Diels–Alder reaction of enamides 165 and 166 to afford both syn-cycloadducts 167
Me
Scheme 4.28
Me
O
O
159
156
CO2Et
+
O H N N + H OTs
Me
O
162
H Me N Me H
O
Me
Me
Me
BOP, i Pr2NEt 77%
N H
OH
Total synthesis of marcfortine C.
PPTS
O H N N
H Me N Me H
142
N H Me Me
NHFmoc
CO2H
N H Me Me
O H N N 160
Me
Me O
O
H Me N Me H
157
O N
Me
77%
S O O 163
nBu
O
N H
O
Me
89%
N H
N
O
O H N N
Me
NH
O
161
Me
Me
H OH O
H Me N Me H
158
N H Me Me
O H
marcfortine C (164)
O H N
Me H
N
Me
O
DIBAL-H
94%
Me
OH
NHFmoc
N
EtO2C O
O
Me
Me
40 °C 60% 159:160 = 1:2.4
Bu3P, DEAD
4.2 Prenylated Indole Alkaloids 137
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids
138
Me H
X = Cl 89% 167:168 = 2:1
O H
Cl O
KOH
O
DIBAL-H 80%
Cl Cl
OH N
N
malbrancheamide (171)
+ epi (168)
MeOH
165: X = Cl 166: X = H
Me H
X=H 86% 169:170 = 1.8:1
H Me N
Cl DIBAL-H
Me H
74%
OH N O
H Me N
Cl
OH N
169
N
Scheme 4.29
H Me N
Me H
167
N
N Me H Me
Cl
Cl
OH N
N N H
X
H Me N
N
+ epi (170)
malbrancheamide B (172)
Completion of the total syntheses of the malbrancheamides.
and 169, as well as the anti-epimers 168 and 170 (Scheme 4.29) [34]. Treatment of the syn-cycloadducts with excess DIBAL-H gave either malbrancheamide or malbrancheamide B. As the malbrancheamides were the first of this family of prenylated indole alkaloids to possess a halogenated indole ring, Williams and coworkers probed the biosynthesis to establish the timing of the chlorination event [35]. Malbrancheamide is proposed to arise from tryptophan, proline, and dimethylallyl diphosphate, leading to deoxybrevianamide E (18, Scheme 4.30). Oxidation of 18 could give intermediate 174, which is expected to suffer intramolecular Diels–Alder reaction HN
CO2H
O H
HO2C 173
NH2
N H
H [ox] O
N N
IMDA
OH
Me
N H 111
Me
N Me Me H 18
OPP 109
Me H
X
N
H Me N red
OH N
N Me H Me 174, X = O 175, X = H2
H Me N
Me H
halogenase
OH N
N
halogenase
OH N N
H Me N
Cl Cl
malbrancheamide (171)
Scheme 4.30
Cl
OH N
177 Me H
H Me N
N
N
X 176, X = O
Me H
Proposed biosynthesis of the malbrancheamides.
malbrancheamide B (172)
O N
CH2Cl2 80% 178:179 = 3:1
S O O 163
Synthesis of key oxindoles.
OH
Scheme 4.31
H O
nBu
145
N H Me Me
N H
N
Me Me
O
O H
Me
Me
Me
Me
O
O
Me Me
HN O
O H
HN O
O H
N H 179
+
N H 178
Me Me
O
N
71%
CH2Cl2
OH DEAD, PBu3 H
CH2Cl2 78%
OH DEAD, PBu3 H
O
N
Me
Me
Me
Me
O
O
N H 181
Me Me
N H 180
Me Me
HN O
O H
HN O
O H
O
N
O
N
4.2 Prenylated Indole Alkaloids 139
140
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids
(IMDA) to cycloadduct 176. Reduction of the tertiary amide would provide premalbrancheamide (177). Alternatively, 18 could suffer reduction of the tertiary amide to 175, providing 177 directly upon cycloaddition. Premalbrancheamide is proposed to undergo subsequent halogenation events to give both malbrancheamide B and malbrancheamide. In feeding studies, labeled dioxopiperazine 176 and premalbrancheamide (177) were added to separate cultures of Malbranchea aurantiaca, but interestingly only premalbrancheamide 177 was incorporated into malbrancheamide B. This suggests that reduction of the tertiary amide must precede Diels–Alder construction of the bicyclo[2.2.2]diazaoctane core through an intermediate analogous to monooxopiperazine 175. Williams has reported that–like stephacidin A and notoamide B, which are produced in Nature as distinct enantiomers–the minor metabolite versicolamide B is likewise produced as distinct enantiomers in different strains of Aspergillus sp. The asymmetric syntheses of both (+)- and (−)-versicolamide B (182) have recently been accomplished by deploying compound 145, previously used in the synthesis of stephacidin A (Scheme 4.31). Oxaziridine oxidation of 145 and pinacol-type rearrangement consequent to oxidation of the indole gave a 3 : 1 separable mixture of 178 and 179 [36]. As previously reported, treatment with tributyl phosphine and DEAD induced Mitsunobu dehydration to give enamides 180 and 181. Treatment of 180 and 181 individually with potassium hydroxide in methanol induced the intramolecular Diels–Alder reaction to afford either (+)- or (−)-versicolamide B (182), along with the minor anti-diastereomers (+)-183 or (−)-183 (Scheme 4.32). As previous biomimetic Diels–Alder reactions containing indole-based azadienes display syn-selectivity (typically ∼2.5 : 1 syn : anti), the exclusive selectivity for the anti-products in the versicolamide B syntheses is of particular interest. The authors suggest that the anti-preference stems from a stable transition state leading to the anti-cycloadduct when an oxindolic azadiene is employed, whereas the syn- and anti-transition states arising from an indolic azadiene are of roughly equal stability [14, 37].
Me Me
Me
O
Me
N H 180
Me Me
Me Me
O
N H 181
O H
N
HN O
O
KOH MeOH 75% 182:183 = 1.4:1
Me H
MeO
O Me NH + Me Me O Me O Me
OH N N O
(+)-versicolamideB, (+)-182
O H
N
HN O
Scheme 4.32
O
HN
KOH MeOH Me 72% O 182:183 = 1.4:1 Me
O Me Me H
O
Me H H O N N
HN
HO N N
(−)-versicolamide B, (−)-182
O (+)-183
+
Me O Me H
NH
OH N
N
O (−)-183
Completion of the asymmetric syntheses of the versicolamides.
Me O Me
4.3 Non-prenylated Indole Alkaloids
While the above syntheses have differed in the approach to the azadiene, all share a common biomimetic intramolecular Diels–Alder reaction to give the bicyclo[2.2.2]diazaoctane core featured in the molecules presented in this section.
4.3 Non-prenylated Indole Alkaloids
Hart and coworkers have expressed interest in the biosynthesis of the fumiquinazoline family of alkaloid natural products, and reported a biomimetic synthesis of ent-alantrypinone as the initial effort at a synthetic program to access more complex, related substances [38]. l-Tryptophan methyl ester was coupled to isatoic anhydride to afford amide 184 in good yield (Scheme 4.33). Acylation of 184 with acyl chloride 186 (prepared in two steps from S-methyl-l-cysteine, 185) under Schotten–Baumann conditions furnished diamide 187, which underwent cyclodehydration to iminobenzoxazine 188. Treatment with excess Li[Me3 AlSPh] effected the rearrangement to quinazolinone 189, and Fmoc deprotection was accompanied by amide formation to give 190. Oxidation of 190 provided the sulfoxide, which was converted into enamide 191 upon heating in benzene with triphenylphosphine. Trifluoroacetic acid induced the conversion into bicycle 192, presumably through intramolecular electrophilic attack of an intermediate N-acyliminium ion onto indole. Conversion into the oxindole proceeded via oxidative rearrangement of 192 with NBS to give the polybrominated indolinone, which was hydrogenolyzed over platinum on carbon to give ent-alantrypinone (193), along with ent-17epi-alantrypinone (194). Three dimeric tryptophan-derived dioxopiperazines have succumbed to biomimetic total syntheses, all completed by Movassaghi and coworkers, namely, (+)-WIN 64821, (−)-ditryptophenaline, and (+)-11,11 -dideoxyverticillin A [39, 40]. Syntheses of the former two began with cleavage of the Boc carbamate to effect the cyclization to dioxopiperazine 196 (Scheme 4.34) [40]. Treatment with bromine then gave a separable mixture of two diastereomers, endo-(+)-197 and exo-(−)-198. endo-Bromide (+)-197 was carried on to (+)-WIN 64821 (201) by, first, treatment with tris(triphenylphosphine)cobalt chloride to afford the dimerized product 199, which was globally deprotected upon exposure to samarium diiodide. (+)-WIN 64821 was thus obtained in 75% yield. Similarly, (−)-ditryptophenaline (202) was synthesized following methylation of exo-(−)-198, dimerization, and deprotection to give the product in 79% yield. 4.3.1 Epidithiodioxopiperazines
(+)-11,11 -Dideoxyverticillin A (211) differs from dimers 201 and 202 in that it is likely derived from l-tryptophan and l-alanine, rather than from l-phenylalanine. Additionally, the dioxopiperazines are bridged by a disulfide, adding a difficult challenge to the synthetic construction of the molecule. Similar to their previous
141
Scheme 4.33
191
N
H N
N
N
89% O
HN
76%
Me
O 192
N
N H
N
N
SMe
96%
N
Me
NH
N
O
N H
O
N
H N
O
+
190
NH
O
187
O HN
O ent-alantrypinone (193)
O
HN
94%
NH
MeO2C
piperidine
74% 193:194 = 1:1.5
2. H2, Pt/C
1. NBS, THF, TFA, H2O
189
O NH
FmocHN MeO2C
DCM H2O, NaHCO3
NH2
NHFmoc
186
N
O
SMe
NH 184
O
NH
Cl
Li[Me3AlSPh]
TFA
SMe
89%
2. SOCl2
1. FmocCl
97%
∆
MeO2C
Synthesis of ent-alantrypinone.
188
O
NH
O
NH
N
O
FmocHN
NH2
SMe
NH2
O 185
31
MeO2C
HO
NH
CO2Me
isatoic anhydride
HN
O
HN
N
Me N
79%
2. Ph3P, ∆
1. m CPBA
80%
i Pr2NEt
Ph3P, I2, DCM
O ent-17-epialantrypinone (194)
O
N
SMe
SMe NHFmoc
142
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids
195
48%
MeN O
O
O
N
N
O
199
200
N H SO2Ph O
N
O
N H SO2Ph O
SO2Ph H N
N
SO2Ph H N
NMe
NH
PhSO2N O
NH
79%
SmI2, NMP
75%
SmI2, NMP
196
HN
O
N
H H N
N H O H
N
O
O
N
H H N
N H O H
N
O NMe
NH
198
NH
197
NH
O
N H O SO2Ph
N
(+)-WIN 64821(201)
O
O
Br
N N H SO2Ph O O
O
(−)-ditryptophenaline (202)
MeN
HN
86%
Br2
Concise total syntheses of (+)-WIN 64821 and (−)-ditryptophenaline.
2. CoCl(PPh3)3
1. MeI
48%
HN
O
80%
O NHBoc
CoCl(PPh3)3
Scheme 4.34
198
197
PhSO2N
TFA; morpholine
O
HN
MeO
Br
4.3 Non-prenylated Indole Alkaloids 143
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids
144
Me HN O NHBoc
PhSO2N
84%
N
O
MeN
Py2AgMnO4
NMe
O
N
O
H
SO2Ph N
N HO O
N H SO2Ph O
59%
OH N N H SO2Ph O 208
Me HS MeN
O
N HO O
O
HO O
87%
N O
KI3 NMe SH Me
pyridine 87%
O
OH N N H H 209
O
SH
210
N
N
O
N H H
OH
TBSCl
NMe OH Me
PPY 55%
H H N
Me HS
O
N H SO2Ph O
H H N
Scheme 4.35
O
Me HO MeN
207
NMe OTBS Me
N
O
205
63%
Me TBSO Na(Hg) MeN
O
Me
SO2Ph H N
Me
N H SO2Ph O 206
Me TBSO MeN
2. MeI
Me O
SO2Ph O H N
NMe N
204
Me
46%
Br
1. Br2
NH HN
PhSO2N
203
CoCl(PPh3)3
O
O
TFA; CO2Me morpholine
NMe 2. ethanolamine OTBS 56% Me
H H N
S N MeN S O
O
1. K2CS3, TFA
O S NMe N S Me N H O H
(+)-11,11′-dideoxyverticillin A (211)
Biomimetic total synthesis of (+)-11,11 -dideoxyverticillin A.
work, Movassaghi and coworkers showed that cleavage of the N-Boc carbamate was accompanied by cyclization to dioxopiperazine 204 (Scheme 4.35) [39]. Exposure of 204 to bromine produced the 3-bromopyrroloindoline, and the amides were subsequently methylated upon treatment with iodomethane. Reductive dimerization with the cobalt(I) complex as before gave the desired dimeric intermediate 206. The dimer was oxidized with bis(pyridine)-silver(I) permanganate to octacycle 207, and exposure to Fu’s (R)-(+)-4-pyrrolidinopyridinyl(pentamethylcyclopentadienyl)iron (PPY) catalyst with t-butyldimethylsilyl chloride (TBSCl) gave selectively the alanine-derived protected hemiaminals of 208. Removal of the benzenesulfonyl groups with sodium amalgam revealed diaminodiol 209. Treatment of 209 with K2 CS3 followed by ethanolamine gave diaminotetrathiol 210, which readily oxidized to (+)-11,11 -dideoxyverticillin A (211) when partitioned between aqueous hydrochloric acid and dichloromethane and treated with potassium triiodide.
4.3 Non-prenylated Indole Alkaloids
145
The total synthesis of sporidesmin A was completed by Kishi and coworkers in 1973. In a series of communications, Kishi described a novel strategy for the synthesis of epidithiodioxopiperazines using a dithioacetal moiety as a protecting group for the disulfide bridge [41–43]. Thus protected, the dithioacetal is stable to acidic, basic, and reducing conditions, allowing for the introduction of thiol groups at an early stage in a total synthesis. Synthesis of the sporidesmins began with the treatment of dioxopiperazine 212 with the dithiane derivative of p-anisaldehyde in the presence of acid to afford dithioacetal-protected dioxopiperazine 213 (Scheme 4.36) [43]. Condensation with acid chloride 214 and subsequent methoxymethyl deprotection gave compound 215. Treatment of ketone 215 with DIBAL-H at −78 ◦ C resulted in stereoselective reduction to the alcohol, which was then converted into acetate 216 in 80% yield. Cyclization to the diacetate (217) proceeded upon addition of iodosobenzene diacetate, and hydrolysis of the acetates gave the corresponding diol. Treatment of the diol with m-chloroperbenzoic acid (mCPBA) afforded an intermediate sulfoxide, which decomposed to the disulfide upon exposure to strong Lewis acid, revealing (±)-sporidesmin A (218). O MeO
N NMe
AcS
O
p -MeOC6H4CH2S2H2, H+
MeO
O 212
S NMe + N S Ar Me MeO O 213
O
MeO OMe
S NMe HN S Ar Me O N Me
1. DIBAL-H 2. Ac2O
Cl MeO
80%
Cl MeO
3. NaOH
N Me
50%
S NMe HN S Ar Me O N Me
O
S NMe N S Ar Me N H O Me
iodosobenzene diacetate 30%
OMe 216
1. NaOH, MeOH 2. mCPBA 3. BF3⋅Et2O
25%
OMe 217 Scheme 4.36
OMe 214
AcO
215 AcO AcO
2. HCl, THF
O
O Cl
COCl 1. BuLi
Cl
Cl MeO
HO HO
O
S NMe N S Me N H O Me
OMe (±)-sporidesmin A (218)
Total synthesis of (±)-sporidesmin A.
The biosynthesis of gliotoxin is believed to proceed through the intramolecular nucleophilic ring-opening of a phenylalanine-derived arene oxide and has been the subject of considerable speculation and interest. Kishi and coworkers drew inspiration from these biogenetic hypotheses in devising a brilliant total synthesis of gliotoxin. The total synthesis of (±)-gliotoxin was completed in 1976 utilizing the same disulfide protecting strategy as deployed above for the sporidesmins, and was re-engineered in 1981 by the same route starting from optically pure
146
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids O
O OtBu
OtBu O O HN S S N Ar + O Me 219
Triton B
OtBu O
220
98% 221:222 = 2:1
HO
OH
77%
AcO
O NS S N Ar Me O 223
O HO H N SS Ar N Me O OCH2OPh 225
Scheme 4.37
HO
3. NaOMe
90%
1. BCl3 2. m CPBA 3. HClO4
35%
O NS S N Ar Me O 222 PhLi, PhCH2OCH2Cl
1. MsCl 2. LiCl
1. Ac2O 2. TFA 3. ClCO2Et 4. NaBH2
O + NS S N Ar Me O 221 Cl
HO
O NS S N Ar Me O 224
53%
O N HO H S S N Me O OH (+)-gliotoxin (226)
Kishi’s total synthesis of (±)- and (+)-gliotoxin.
dithioacetal 219 obtained from resolution (Scheme 4.37) [44, 45]. Coupling of 219 with t-butoxy arene oxide 220 in the presence of triton B afforded 221 and 222 in a 2 : 1 ratio. Acylation, deprotection, mixed anhydride formation, and reduction gave alcohol 223 in 77% yield. Alcohol 223 was converted into the chloride following mesylation, and then deprotected to reveal alcohol 224. The key stereoselective cyclization–alkylation reaction was achieved upon addition of phenyllithium to 224 and phenoxymethyl chloride, affording cycloadduct 225 in modest yield (53%). The primary alcohol was revealed upon removal of the benzyl ether, and the thioacetal oxidatively removed to afford either (±)- or (+)-gliotoxin (226).
4.4 Conclusion
Dioxopiperazine alkaloids cover an astonishing array of molecular architecture and, with that, corresponding synthetic challenges to construct such substances. The biosynthesis of many of the natural substances touched on in this chapter has, in some instances, been studied going back several decades, and many workers have sought to exploit insights from Nature’s strategic bond constructions in a synthetic laboratory context. Advances in whole genome sequencing have brought new and invigorated interest in elucidating the biosynthesis of structurally intriguing and biomedically relevant secondary metabolites. The insights to be gained from educated guess work on what specific compounds might lie along a possible biosynthetic pathway, traditionally accomplished by isolation and structural
References
elucidation to map metabolite co-occurrence in conjunction with isotopically labeled precursor incorporation experiments, is in the process of giving way to a much higher resolution picture of secondary metabolism in microorganisms and plants. With the advent of powerful new genomics and proteomics tools to study and manipulate secondary metabolite production, advances in our understanding of Nature’s creative synthetic palette will surely explode in the coming years. The fruits of these insights will undoubtedly be extensively exploited by synthetic chemists working at the forefront of complex molecule synthesis. In addition, many new natural products–the biosynthetic intermediates themselves, often isolated in trace amounts if at all–will provide and constitute worthy new synthetic targets and substrates for various important applications.
Acknowledgment
R.M.W. is grateful to the National Institutes of Health (GM068011; CA70375; CA85419) for financial support.
References 1. Birch, A.J. and Russell, R.A. (1972) 2.
3. 4.
5.
6.
7.
8.
9.
10.
Tetrahedron, 28, 2999–3008. Birch, A.J. and Wright, J.J. (1969) J. Chem. Soc., Chem. Commun., 644–645. Birch, A.J. and Wright, J.J. (1970) Tetrahedron, 26, 2329–2344. Baldas, J., Birch, A.J., and Russell, R.A. (1974) J. Chem. Soc., Perkin Trans. 1, 50–52. Sanz-Cervera, J.F., Glinka, T., and Williams, R.M. (1993) Tetrahedron, 49, 8471–8482. Kametani, T., Kanaya, N., and Ihara, M. (1980) J. Am. Chem. Soc., 102, 3972–3975. Schkeryantz, J.M., Woo, J.C.G., Siliphaivanh, P., Depew, K.M., and Danishefsky, S.J. (1999) J. Am. Chem. Soc., 121, 11964–11975. von Nussbaum, F. and Danishefsky, S.J. (2000) Angew. Chem., Int. Ed., 39, 2175–2178. Edmondson, S., Danishefsky, S.J., Sepp-Lorenzino, L., and Rosen, N. (1999) J. Am. Chem. Soc., 121, 2147–2155. Finefield, J.M. and Williams, R.M. (2010) J. Org. Chem., 75, 2785–2789.
11. Baran, P.S., Guerrero, C.A., and Corey,
12.
13. 14.
15.
16.
17. 18.
19.
E.J. (2003) J. Am. Chem. Soc., 125, 5628–5629. Depew, K.M., Marsden, S.P., Zatorska, D., Zatorski, A., Bornmann, W.G., and Danishefsky, S.J. (1999) J. Am. Chem. Soc., 121, 11953–11963. Porter, A.E.A. and Sammes, P.G. (1970) J. Chem. Soc., Chem. Commun., 1103. Domingo, L.R., Sanz-Cervera, J.F., Williams, R.M., Picher, M.T., and Marco, J.A. (1997) J. Org. Chem., 62, 1662–1667. Williams, R.M., Sanz-Cervera, J.F., Sancenon, F., Marco, J.A., and Halligan, K. (1998) J. Am. Chem. Soc., 120, 1090–1091. Stocking, E.M., Sanz-Cervera, J.F., and Williams, R.M. (2000) J. Am. Chem. Soc., 122, 1675–1683. Jin, S.D., Wessig, P., and Liebscher, J. (2001) J. Org. Chem., 66, 3984–3997. Sanz-Cervera, J.F. and Williams, R.M. (2002) J. Am. Chem. Soc., 124, 2556–2559. Adams, L.A., Valente, M.W.N., and Williams, R.M. (2006) Tetrahedron, 62, 5195–5200.
147
148
4 Biomimetic Synthesis of Alkaloids Derived from Tryptophan: Dioxopiperazine Alkaloids 20. Greshock, T.J., Grubbs, A.W.,
21. 22. 23. 24.
25.
26.
27.
28.
29.
30.
31.
Tsukamoto, S., and Williams, R.M. (2007) Angew. Chem. Int. Ed., 46, 2262–2265. Greshock, T.J. and Williams, R.M. (2007) Org. Lett., 9, 4255–4258. Herzon, S.B. and Myers, A.G. (2005) J. Am. Chem. Soc., 127, 5342–5344. Myers, A.G. and Herzon, S.B. (2003) J. Am. Chem. Soc., 125, 12080–12081. Baran, P.S., Guerrero, C.A., Ambhaikar, N.B., and Hafensteiner, B.D. (2005) Angew. Chem. Int. Ed., 44, 606–609. Baran, P.S., Guerrero, C.A., Hafensteiner, B.D., and Ambhaikar, N.B. (2005) Angew. Chem. Int. Ed., 44, 3892–3895. Baran, P.S., Hafensteiner, B.D., Ambhaikar, N.B., Guerrero, C.A., and Gallagher, J.D. (2006) J. Am. Chem. Soc., 128, 8678–8693. Qian-Cutrone, J., Huang, S., Shu, Y.Z., Vyas, D., Fairchild, C., Menendez, A., Krampitz, K., Dalterio, R., Klohr, S.E., and Gao, Q. (2002) J. Am. Chem. Soc., 124, 14556–14557. Kato, H., Yoshida, T., Tokue, T., Nojiri, Y., Hirota, H., Ohta, T., Williams, R.M., and Tsukamoto, S. (2007) Angew. Chem. Int. Ed., 46, 2254–2256. Deyrup, S.T., Swenson, D.C., Gloer, J.B., and Wicklow, D.T. (2006) J. Nat. Prod., 69, 608–611. Mudur, S.V., Gloer, J.B., and Wicklow, D.T. (2006) J. Antibiot., 59, 500–506. Shim, S.H., Swenson, D.C., Gloer, J.B., Dowd, P.F., and Wicklow, D.T. (2006) Org. Lett., 8, 1225–1228.
32. McAfoos, T.J., Li, S., Tsukamoto, S.,
33.
34.
35.
36.
37.
38. 39.
40.
41.
42.
43.
44. 45.
Sherman, D.H., and Williams, R.M. (2010) Heterocycles, 82, 461–472. Greshock, T.J., Grubbs, A.W., and Williams, R.M. (2007) Tetrahedron, 63, 6124–6130. Miller, K.A., Welch, T.R., Greshock, T.J., Ding, Y.S., Sherman, D.H., and Williams, R.M. (2008) J. Org. Chem., 73, 3116–3119. Ding, Y.S., Greshock, T.J., Miller, K.A., Sherman, D.H., and Williams, R.M. (2008) Org. Lett., 10, 4863–4866. Miller, K.A., Tsukamoto, S., and Williams, R.M. (2009) Nat. Chem., 1, 63–68. Domingo, L.R., Zaragoza, R.J., and Williams, R.M. (2003) J. Org. Chem., 68, 2895–2902. Hart, D.J. and Magomedov, N.A. (2001) J. Am. Chem. Soc., 123, 5892–5899. Kim, J., Ashenhurst, J.A., and Movassaghi, M. (2009) Science, 324, 238–241. Movassaghi, M., Schmidt, M.A., and Ashenhurst, J.A. (2008) Angew. Chem. Int. Ed., 47, 1485–1487. Kishi, Y., Fukuyama, T., and Nakatsuk, S. (1973) J. Am. Chem. Soc., 95, 6492–6493. Kishi, Y., Fukuyama, T., and Nakatsuk, S. (1973) J. Am. Chem. Soc., 95, 6490–6492. Kishi, Y., Nakatsuk, S., Fukuyama, T., and Havel, M. (1973) J. Am. Chem. Soc., 95, 6493–6495. Fukuyama, T. and Kishi, Y. (1976) J. Am. Chem. Soc., 98, 6723–6724. Fukuyama, T., Nakatsuka, S., and Kishi, Y. (1981) Tetrahedron, 37, 2045–2078.
149
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus Tanja Gaich and Johann Mulzer
5.1 Introduction
Indole alkaloids make up the largest group of alkaloids, with more than 2000 members. They exhibit a huge structural diversity and an often mind dazzling molecular complexity. Owing to their architectural attractiveness and their pronounced pharmacological activities they have been under extensive investigations over recent decades, including the elucidation of their biosynthetic pathways, which have been pioneered by Battersby, Arigoni, Scott, and others. The landmark works were the unraveling of the biosynthetic pathway of monoterpenoid indole alkaloids (e.g., strictosidine 9 as the first biosynthetic node of the series), especially the monoterpenic part 7 derived from geraniol diphosphate 5 (highlighted in Scheme 5.1) [1]. These early biosynthetic studies relied on the administration of isotopically labeled common starter units, followed by isolation and structural characterization. These findings inspired synthetic chemists not only to ‘‘mimic’’ but also predict these biogenetic transformations and probe them in vitro.’’ The synthetic elegance, ease, and conciseness that can be achieved by applying a biomimetic route or even just one ‘‘biomimetic key step’’ has led to a myriad of awe-inspiring total syntheses. Some of these biomimetic syntheses lead to natural products in which the indole nucleus is lost and an indoline, indolenine, indoxyl, or indolinone ring system is obtained instead (Figure 5.1). This chapter is focused on such structures with a ‘‘broken indole nucleus.’’ We are only able to cover a small fraction of these syntheses, and therefore the selection was conducted to provide the reader with an overview of no more than 13 different structure types (Figure 5.2).
Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
150
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus O
OH CO2H CoAS 1 from 3x AcetylCoA
Mevalonate pathway
H
OP2O52−
OR
OH
3
OH OPO32−
Rohmer (MEP) pathway
4
OP2O5 5
OP2O52−
2−
O OH 2 from pyruvate and glyceraldehyde
All other monoterpenoid indolalkaloids
N H H
NH2 +
N H
O strictosidine (9)
Scheme 5.1
tryptamine (8)
O OGluc secologanin (7)
OGluc
Biosynthesis of monoterpenoid indole alkaloids.
R3 R1 R2 N
R1 R2
Biomimetic Synthesis
N H
indolenine R2 R1
R3 R1 R2 4 N R H indoline
O R2
O N H indolinone Figure 5.1
O O iridotrial (6)
CHO CO2Me
"PictetSpengler"
NHCO Me 2
H
N R1 H indoxyl
Conversion of the indole nucleus into first derivatives.
5.2 Individual Examples 5.2.1 (±)-Camptothecin
Camptothecin was discovered in 1966 by M. E. Wall and M. C. Wani in a systematic screening of natural products for anticancer drugs [2, 3]. It was isolated from the bark and stem of Camptotheca acuminata (Camptotheca, Happy tree), a tree native in China. Camptothecin showed remarkable anticancer activity in preliminary clinical trials, by inhibiting topoisomerase of type I enzyme, and thereby causing apoptosis. Its disadvantageous low solubility and (high) adverse drug reaction has initiated numerous syntheses of camptothecin derivatives, two of which have been approved, and are used as anticancer drugs (Figure 5.3) [4]. Camptothecin has a planar pentacyclic ring structure that includes a pyrrolo[3,4-β]-quinoline moiety (rings A, B, and C), a conjugated pyridone moiety (ring D), and one chiral center at position 20 within the α-hydroxy lactone ring
N N Br N H
N
O
Figure 5.2
S N Me N S Me N H O H
O
O
R HN
N R1
R
2
A
C N D
OH O
E O
N Me
H H H NHMe
NHMe
N H
N H
HN
N Me
N
H Me Me O
(+)-welwitindolinone A (22)
CN
Me
Cl
H
O
Me
NH
H
NHMe
NMe
H H
Me O O Me N H
N
N H
H H N
O Me Me O Me
Me
(−)-chimonanthine (19)
N N H H Me
MeN
(±)-brevianamide B (13)
O
19
O
N Me HO O Me (−)-stephacidin B (23)
N H H O H N O H
N
H O H N N
(±)-borreverine (18)
N H
communesin B (12)
O
(±)-isoborreverine (17)
N
calycanthine (21)
N H
H N
N H
Overview of structures discussed in this chapter.
OH
Me N
N
B
O
(±)-camptothecine (11)
R=H discorhabdin E (16) R=Br discorhabdin C (15)
N H
O
N N O H OMe
O
Cl
Br
gelselegine (20)
(±)-chartelline C (14)
Br
H H N
(+)-11,11'-dideoxyverticillin A (10)
O
S N Me N S
Me
O
5.2 Individual Examples 151
152
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus
24 and 25 - proposed biosynthetic precursors O N
N
N H H
N 20
O MeO
N
N H H
H MeO
H
OH O O (+)-camptothecin (11)
OMe
O O geissoschizine (25)
corynantheidine (24)
commercial camptothecin derivatives N HO
O
O OH O
N
O
N
N
topotecan (26)
O
N
N
O
N
irinotecan (27)
O OH O
Figure 5.3 Camptothecin (11) with its proposed biosynthetic precursors and derivatives used as anticancer drugs.
with (S)-configuration (ring E). Its planar structure is thought to be one of the most important factors in topoisomerase I inhibition. Biosynthetically, 11 is unique, because it is the first reported example of an alkaloid containing the pyrrolo[3,4-β]quinoline unit. In 1967 Wenkert [5], suggested that 11 might be formed in vivo from an indole alkaloid of the corynantheidine type (24), whereas Winterfeldt suggested a biosynthetic relationship between 11 and geissoschizine (25) [6]. It was Hutchinson et al., in brilliant work in 1974, who finally solved the biosynthetic maze by isotope labeling experiments [7]. They were able to show that strictosidine (9) is unequivocally the biosynthetic precursor via which Nature accesses structure 11 (Scheme 5.2). As camptothecin (11) appears to be more of a quinoline than of an indole alkaloid, the question arose how 9 is converted into 11. The first event after the formation of strictosidine (9) is the lactam formation of strictosamide (28). After this an oxidative cleavage of the indole nucleus with subsequent recyclization takes place to give the ABC-ring moiety of 11. The last events in the biosynthesis are the placements of the correct oxidation states at rings D and E. Soon after Hutchinson’s biosynthesis was published, Winterfeldt and coworkers completed the first biomimetic total synthesis of this molecule [8], following Hutchinson’s findings (Scheme 5.3). Their synthesis started from known compound 31, which was converted into 32 via diisopropyl-carbodiimide (DIC) mediated amide bond formation. Claisen ester condensation with potassium tert-butoxide furnished compound 33 (86% yield). This 1,3-dicarbonyl compound was then methylated with diazomethane, and a subsequent Michael addition/elimination reaction with di-tert-butyl malonate gave compound 34. Biomimetic oxidation and recyclization under mild basic conditions afforded quinolone 35 in impressive 75%
5.2 Individual Examples CHO CO2Me NH2 +
N H
"pictetspengler"
NH CO Me 2
N H H
O
N
N H H
O
O
OGluc tryptamine (8)
base
153
strictosidine (9)
secologanin (7)
O strictosamide (28)
OGluc
OGluc
[O] then base
−
O B
A
N
O
C N D
D-ring-[O] E-ring-[O]
B
A
C N D H
N E O OH O
camptothecin (11)
Scheme 5.2
O
O
N
N H H
O
E O
30
O
29
OGluc
OGluc
Biosynthesis of camptothecin (11) by Hutchinson et al.
b. KOt Bu
a. DIC
N H
NH
31
CO2Et
HO2C
CO2Et
N H
N
32
CO2Et
86%
O
N
N H
86%
CO2Et
O CO2Et
33 OH
O O
35
N H
68%
e. O2;NaH; DMF
N CO2Et
f. SOCl2; DMF g. Pd; BaSO4; H2 h. DME; DIBALH
t BuO2C
51%
O
36
CO2t Bu
A
i. EtI; NaH
N O
B N
j. CuCl2; O2 DMF 18%
C N D
O
E O OH O (±)-camptothecin (11)
O
The mechanism of the Winterfeldt-Witkop cyclization
O
O
O − O2; NaH; DMF
Scheme 5.3
CO2Et
CO2tBu
N
N H
O
34
t BuO2C
O
N
N H
75%
c. CH2N2 d. NaH; CH2(CO2t Bu)2
N
N H
O ON
N H O
N
Biomimetic synthesis of camptothecin (11) by Winterfeldt et al.
N N H
154
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus
yield. Conveniently, spontaneous oxidation of ring D to the final oxidation state occurred. Chlorination of the quinolone moiety, followed by dechlorination under palladium catalysis and reduction of the ester functionality, led to compound 36. This was alkylated and oxidized via copper(II) catalysis with oxygen to give the tertiary alcohol of (±)-camptothecin. This very elegant synthesis applies all the biosynthetic key reactions, namely, the formation of the quinoline from the indole moiety, the spontaneous oxidation of ring D, and the introduction of the tertiary alcohol in the final step of the total synthesis. The synthesis is straightforward and very short, even enabling facile access to camptothecin derivatives. However, owing to its biological importance, and although many synthetic approaches and syntheses have been achieved, camptothecin is still the subject of numerous synthetic investigations. Thereby, progress has been made with respect to oxidation conditions, which were developed for base-sensitive substrates [9]. 5.2.2 (±)-Discorhabdins C and E
Discorhabdins, like makaluvamines and damirones, belong to the group of pyrroloiminoquinone alkaloids. Over the last 20 years they have been regularly isolated from sponges generally confined to two families, Latrunculidae and Acarnidae [10]. The sponges producing these natural products have been collected from shallow and deep-water habitats, ranging from tropical regions of the Caribbean Sea to the cold water regions of New Zealand and the Antarctica. Discorhabdins, also known as prianosins, are the major pigments in these beautiful sponges. They exhibit extremely high activity against P-388 lymphotic leukemia cell lines in vitro (IC50 values 30–1500 ng ml−1 ). The mode of action of these natural products is the strong inhibition of the topoisomerase II enzyme, which plays a vital role in cell division and other fundamental processes in eukaryotic cells. Discorhabdins induce apoptosis and thereby serve as potent chemotherapeutic agents. The most active congener appears to be discorhabdin C (15). Munro et al. proposed a biosynthesis that includes all pyrroloiminoquinone alkaloids (makaluvamines, damirone, and discorhabdins), which shows how these alkaloids are biogenetically connected (Scheme 5.4) [11]. Tryptamine 7 is the starting point of the proposed biosynthesis. Oxidation of 7 gives pyrroloiminoquinone 37. 1,4-Addition of tyramine, and subsequent oxidation leads to makaluvamines (39). Intramolecular oxidative phenol coupling leads to the carboskeleton of discorhabdins (40). Sulfur-containing discorhabdins can, in principle, be accessed in two ways, either via makaluvamine F (42) or via discorhabdins (40) and a late sulfur introduction. Numerous approaches to discorhabdins have been reported [12]. Owing to lack of space we focus here on the biomimetic synthesis of racemic discorhabdin C by Heathcock et al. in 1999 (Scheme 5.5) [13]. For the pyrroloiminoquinone synthesis, Heathcock et al. started from nitrophenol 45. Benzylation of the phenolic alcohol and reduction of the nitro group to the aromatic amine was followed by iodination of the aromatic ring and alkylation of
5.2 Individual Examples +
NH3 +
Ra
HN [O]
[O]
N H 7
O 37
Path A
N H
+
N
N
Rb
Rb
R2 N R1
O
H2N
O 38 Damirones AC Makaluvamine O
ex Tryptophan
155
R2 N R1
O 39 Makaluvamines A-C,H,I,N isobatzellines A-E
HO Path B
O R5 R
3
R4
Ra
+
HN
ox. phenolcoupling HO
R6 R R7
NH2 then [O]
N H
HO
2
N H
40 O Discorhabdin C,E,F,G, K,O,P
S
O
N H O Makaluvamine F 42 N H
early-sulfurintroduction
N R1 41
Makaluvamines D,E,G,J-M,P O R5 3
R
O R4
R5
N
R
S H
R2 N H
O 43
N R1
Discorhabdin Q,T,W,S
Scheme 5.4
HN R2
N R1
late-sulfurintroduction
+
+
N
+
ox. phenolcoupling
R4 N
3
R2
S H
N H
O
N R1
44 Discorhabdin A,B,I,R
Biosynthesis of discorhabdins proposed by Munro et al.
the amine, giving 46. Palladium-catalyzed indole formation via a Heck coupling reaction and subsequent standard functional group manipulation gave Boc-protected indole 47. This was oxidized with Fr´emy’s salt [Na2 NO(SO3 )2 ] to yield quinone 48. Compound 48 was treated with trifluoroacetic acid (TFA) to deprotect the primary amine and subsequently cyclized to yield desired pyrroloiminoquinone 49 in eleven steps. Pyrroloiminoquinone 49 underwent a 1,4-addition elimination reaction with amines A and B, affording makaluvamines 50A and 50B in 54% yield. Application of biomimetic oxidative phenol coupling with copper(II) chloride gave the corresponding spiro-dienone compounds whose deprotection led to (±)-discorhabdins C (15) and E (16). 5.2.3 (±)-Brevianamides, Paraherquamides, VM55599, and Marcfortines
Prenylated indole alkaloids comprise a large group of secondary metabolites that have been isolated from various fungi (in particular Penicillium sp. and Aspergillus sp.) (Figure 5.4) [14, 15]. Owing to their molecular complexity and biogenesis these natural products have attracted considerable attention. Paraherquamides (e.g., 52) most notably display potent anti-parasitic activity.
2
NH2 N H
R N
MeO
N H
N H
64%
m. CuCl2 NEt3; O2 n. NaOMe
R1
N H
MeO
O 50
N
N Ts
N Ts OBn 47
BocHN
54%
l. NaHCO3 then A/B
82%
i. H2 /Pd j. Fremy's salt
Biomimetic synthesis of discorhabdins C (15) and E (16) by Heathcock et al.
R2
HO
53%
g. (Boc)2O h. Ts2O NaH
e. Pd(OAc)2 f. LAH
Biomimetic step
OBn 46
I
O R=Br (±)-Discorhabdin C 15 R=H (±)-Discorhabdin E 16
Br
O
d. BrCH2CH=CHCN NaHCO3 59%
NO2 c. ICl; K2CO3
A: R =Br R =H B: R1=Br R2=Br
1
R1
OH 45
Scheme 5.5
R2
HO
MeO
a. BnBr; K2CO3 b. Fe; HCl
CN
MeO
N Ts
NHBoc
O
N
49
N Ts
k. TFA DCM
O 48 54%
MeO
O
156
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus
5.2 Individual Examples O
O H N H
19
N H
O H N N
O
19
OMe N N
OH O O paraherquamide B (52)
O brevianamide B (13) O
O NH
H
19
O
19
OMe N N
OMe N N
OH O O sclerotamide (54)
O
NH
H
19
VM55599 (53)
Figure 5.4
NH
H 19
O brevianamide A (51) H N H O H N N
O
H
H O N N
157
OH
O
O marcfortine (55)
Selected members of the family of prenylated indole alkaloids.
The key feature of all these structures is their azabicyclo[2.2.2] ring system that has been proposed to arise from a biogenetic [4+2]-cycloaddition of the isoprene moiety across the diketopiperazine unit (Scheme 5.6) [16]. The relative configuration at C19 (brevianamide numbering) is different within this natural product family. For brevianamides A and B (51 and 13, respectively) it is anti-, but for all others it is a syn-relationship. This difference is explained by the facial divergence in the selectivity of the Diels–Alder reaction, which sets the relative stereochemistry of this carbon center. Although the precursor is an achiral molecule, the brevianamides exhibit optical activity; this example might be a strong O
OH
O
HN
HN
OPP
N
N N
4
O
N
[O]
O
N H
O
N H deoxybrevianamide E (57)
brevianamide F (56)
O
58 H N
N H
N H
H O N N
O brevianamide A (51)
[4+2] HO rearrangement
O
H O N N
62
O
[O]
H N
N
H N H
O brevianamide B (13)
Scheme 5.6
O H N N
H O N N
59
O
O H N N
achiral
N H
OH O
61
Biosynthesis of brevianamide B (13) by Williams et al.
O H N N 60
O
158
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus
hint and important case in the search for the Diels–Alderase enzyme [17]. Except for catalytic antibodies, to date only three reported examples of a Diels–Alderase exist, in addition to ribozymes that are known to catalyze Diels–Alder reactions [18]. The group of R.M. Williams, which is one of the leaders in biosynthetic investigations concerning this class of natural products, has completed numerous total syntheses of several congeners of this class [19], among them biomimetic total syntheses of VM55599 (53) [20], and brevianamide B (13), with the latter being discussed here (see also Chapter 4) [21]. This synthesis started from compound 57, which was converted into the lactim ether 62 with Meerwein’s salt (Scheme 5.7). Subsequent oxidation of 62 with DDQ (2,3-dichloro-5,6-dicyano-1,4-benzoquinone) gave diene 63, which was treated with methanolic potassium hydroxide to give in situ the desired azadiene 64, which smoothly underwent an intramolecular Diels–Alder reaction to give the central azabicyclo[2.2.2]ring system that is common to all members of this natural product family. In this case, brevianamide B precursors 65 and 66 were formed in a 2 : 1 ratio in favor of the undesired C19-diastereomer 66. Oxidation with mCPBA (m-chloroperbenzoic acid), and treatment with HCl gave indoxyls 67 and 68. These intermediates under basic conditions underwent ring contraction to form, after acid hydrolysis of the lactim ether, (±)-brevianamide B (13) and the 19-epi-compound 69. This elegant synthesis showcases how predictions about the biosynthetic origin of natural products influence synthetic routes, lead to new insights into chemical reactivity, and give evidence for possible enzymatic transformations of enzymes that remain to be discovered, encouraging those who are in constant search for them. 5.2.4 (+)-Stephacidin A and (−)-Stephacidin B
Prenylated indole alkaloids are an impressive example of how Nature assembles the most complex and diverse structures from a very limited number of rather simple building blocks [15, 22]. In the case of stephacidins these building blocks are tryptophan, proline, and ‘‘isoprene units.’’ Stephacidins are fungal metabolites that were isolated from Aspergillus ochraceus WC76466 by Bristol Myer Squibb in 2002 and independently in 2001 by Pfizer [23]. They are reported to inhibit the growth of cultured human cancer cells (IC50 values 50–100 nM), and are therefore biologically interesting structures. They belong to a large family of natural products together with brevenamides, notoamides, and paraherquamides. Their biosynthesis contains two key steps. The first is an intramolecular Diels–Alder reaction to form their characteristic azabicyclo[2.2.2] ring system [16], in the same fashion as discussed above in the case of brevenamides (Scheme 5.8). The second key step is an oxidation of stephacidin A (72) to avrainvillamide (73), which then cyclizes to give the most complex sibling of this natural product family, namely, stephacidin B (23). This last and most intriguing oxidation reaction was utilized first by Myers et al. [24] and the group of Baran [25] and then in 2007 stephacidin B (23) was synthesized by the group of Williams using this last
H
57
O
HN
N H
O
H
N
Scheme 5.7
69
N H
O
f. NaOMe g. HCl
79%
N
HO
N
HO
a. Me3OBF4
62
O 68
H O N N
H O N N O 67
HN
N
N
d. mCPBA e. HCl
H OMe
O
H N
H N
O
O
31%
b. DDQ
Biomimetic synthesis of brevianamide B (13) by Williams et al.
H O N N
O (±)-brevianamide B (13) O
O H N N
HN
O
MeO N N 66
MeO N N 65
HN 63
HN
OMe
N
66:65 = 2:1
N
O
64
N
N
OMe
O
60%
c. KOH; MeOH H2O
5.2 Individual Examples 159
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus
160
O
O
O
O
[O]
N
N HN
HN
O
70
N
N
Scheme 5.8
Me O Me
H N H
O Me
O H
O
dimerization
Me H
Me O N Me O Me H Me (+)-stephacidin A (72) [O]
Me O O Me N+ H
N H H N O
N OH
71 −
O
N
HN
H N H
O D.A.
N
Me HO O Me (−)-stephacidin B (23)
N
O Me Me
N +
−
O
Me Me
(+)-avrainvillamide (73)
Biosynthesis proposal for stephacidins.
oxidative dimerization reaction [26]. Baran’s synthesis is highlighted here. It started with the formation of the annelated pyran ring to compound 74 in three steps, affording 75 in 69% yield (Scheme 5.9). Protection of indole and ester hydrolysis gave carboxylic acid 76, which was reacted with amine 77 in a peptide coupling reaction to give compound 78. The next task was the formation of the azabicyclo[2.2.2] ring system, which was achieved via an oxidative intramolecular enolate coupling of diketopiperazine 79. This reaction, although very elegant, is not in accord with the proposed biosynthetic transformation for this step. Removal of the MOM protective group and introduction of the gem-dimethyl group was followed by dehydration to give isopropylidene compound 81, which upon heating at 200 ◦ C in sulfolane underwent an ene-reaction that was followed by a 1,2-shift to furnish (+)-stephacidin A (72). Oxidation of 72 to (+)-avrainvillamide (73) with hydrogen peroxide and selenium dioxide was followed by treatment of 73 with triethylamine, which resulted in a sequential double Michael addition, as outlined in Scheme 5.10, to give (−)-stephacidin B (23). Dimerization of 73 also occurred spontaneously in an NMR tube, albeit in lower yields. The biomimetic syntheses of stephacidins show how impressively biosynthetic considerations simplify synthetic routes of mind boggling molecular architectures. 5.2.5 (±)-Chartelline C
Chartelline marine alkaloids were isolated from the bryozoan Chartella papyracea by Christophersen and coworkers and characterized in the 1980s [27]. Chartelline C (14) is the scarcest member of the chartelline alkaloid family. It consists of an indolenine, an imidazole, and β-lactam heterocycle arrayed in a dense, π-stacking framework that poses numerous challenges for its synthesis [28]. Biosynthetically, chartellines are proposed to stem from one tryptamine, one histidine, and one prenyl unit (Scheme 5.11) [29].
O
80
N
O B Br O
N 79 Boc
l. MeMgBr m. Burgess 63%
k. MOM N O
O
69%
O
O
O
MeO2C
N
MOM H N
75
N H
O
CO2Me 77%
N Boc 81
H O
N
MOM O N
i. MOMCl, NaH 65%
O
O
H
N
N
H N
H CO Me 2
77
CO2Me
N Me O H Me
O
MeO2C
CO2H NHCbz
O
(−)-stephacidin A (ent-72)
N 78 Boc
O
HN MeO2C
CbzHN
g. HATU 81%
76
N Boc
(+)-stephacidin A (72) was synthesized through the same sequence after determination of the absolute configuration
28-45%
n. 200° C
e. Boc2O f. LiOH
h. Pd2dba3, CHCl3 NEt3, MeOH ∆
NHCbz
Total synthesis of stephacidin A (72) by Baran et al.
N CO2Me Boc O
H
61%
d. ∆
74a
a. Boc2O b. Mg/MeOH c. CuCl MeO2CO
j. LDA; Fe(acac)3
N 74 H
Scheme 5.9
TsO
NHCbz
CO2Me
5.2 Individual Examples 161
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus
162
H N
O
H
O
N O MeH Me
N
Me
H N
O
Me a. SeO , H O 2 2 2 27%
H
O Me
N+ O MeMe − O
N
(+)-avrainvillamide (73)
(+)-stephacidin A (72)
69% −
Me O O Me N + H
N O
Me O Me O
N H H N H H O
N
b. NEt3 dimerization
− Me O O Me + N
Me O Me
Me O
Me
N Me OH O Me
N
Me
N HH H N O N
Me
O
H
Me
+
N O Me O− Me
(−)-stephacidin B (23)
Scheme 5.10 O
Biomimetic synthesis of stephacidin B (23) by Baran et al.
H N
O
N H Cl securine B (82) N H
H
H+
N
Br
Br
H
H N N
Br
+
N H
N H
Cl
Br
N H
O N
N HN
Cl 83
securamine B (84) Br
+
[Cl ]
O O
H N
O
Cl Br
N +
N H
O
N H 85 Cl
Scheme 5.11
−2HCl
N
N
[1,5]
Cl H N N
Br N 86
N
N Br Br
N
HN Br
N N Br N H
Cl
chartelline C (14)
Biosynthesis proposal for chartelline C (14).
A related bryozoan, Securiflustra securifrons, also produces the securines and securamines, and they are plausible biogenetic precursors to the chartellines. It was reported that solutions of securine B (82) in DMSO-d6 converted into securamine B (84), presumably via the protonated intermediate 83 (Scheme 5.11). Redissolving 84 in CDCl3 converted 84 back into 82. One could imagine that if this isomerization was carried out in the presence of an electrophilic source of chlorine, 82 could be converted into chartelline C (14) through the intermediacy of 85 and 86. The transformation of 86 into 14 can be written as a [1,5]-shift and is a key step in the biomimetic strategy for the synthesis of chartelline C by Baran et al. (Scheme 5.12) [30].
N H
O
N N Boc N H 95
93%
MgBr
O
P(O)(OEt)2 CHO
CO2SEM
88
TBSO
CO2SEM 86%
94a P(O)(OEt)2
H2N
89
TBSO
m. BOPCl
84%
d. H2O2 e. TBSCl
b. KNCS
NH2 c. HCl
CHO
98%
l. LiOH
j. TBAF k. MnO2
P(O)(OEt)2 N2 90 63%
CO2H
MeOC
f. NaIO4, OsO4 TBSO g. K2CO3
N N Boc N H 94
N H
N
Biomimetic synthetic route towards chartelline C (14) by Baran et al.
CO2Me 87
NH2
Scheme 5.12
TBSO
a.
+ 92
N Boc
93
N N N H Boc
OTBS
CO2Me
Br
CO2Me
71% h. Pd(PPh3)4 i. H2; Lindlar
N H 91
N
5.2 Individual Examples 163
164
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus
Their synthesis started with the construction of prenylated imidazole 89, which was achieved in five standard transformations from serine derivative 87. For the introduction of the (Z)-double bond on the macrocycle and the coupling with indole fragment 92 the double bond in 89 was converted into alkyne 91. Sonogashira coupling of 91 and 92 gave compound 93, which was deprotected with TBAF, oxidized to give the aldehyde functionality in 94, and subjected to ester hydrolysis to finally furnish carboxylic acid 94. This acid was coupled with amine 94a to yield the substrate for the Horner–Wadsworth–Emmons reaction to close the macrocycle of chartelline C. Masamune–Roush conditions (Scheme 5.13) afforded the macrocycle in 56% yield, which was brominated to give securine-type compound 96. The endgame consisted of the same biosynthetic transformation as shown in Scheme 5.11 (86→14). Baran and coworkers used N-bromosuccinimide as an electrophilic bromine source. The reaction was carried out at 180 ◦ C to remove the Boc-protective group, followed by bromination of the C3-position at the indole to give intermediate securamine-type structure 97. This compound in situ underwent a [1,5]-shift to give the desired carboskeleton of chartelline C. Ester hydrolysis mediated by TFA was followed by decarboxylation at 200 ◦ C to yield (±)-chartelline C (14). The chartellines with their unique, densely functionalized, and very strained architecture are a very good example of how biosynthetic considerations can lead to the application of new transformations in organic synthesis, if the practitioners keenly follow the biosynthetic proposal, even if it is very difficult to execute in the laboratory.
95
O H N
n. LiCl, DIPEA o. Br2; NBA 31%
N Boc N
Br 96
O CO2SEM p. NBS 180°C Br
Br N
97
Br
N H
Br N H
93%
O
Cl N Br (±)-chartelline C (14)
Scheme 5.13
N
NaCl
O
N Br
Br
N HN
Br
CO2SEM
q. TFA then 200°C 64%
N H
N Br
N
CO2SEM Cl
N
Br N 98 H
Endgame of the biomimetic synthesis of chartelline C (14) by Baran et al.
5.2.6 (+)-Welwitindolinone A and (−)-Fischerindole I
The discovery of hapalindole, fischerindole, ambiguine, and welwitindolinone indole alkaloids isolated from marine blue-green algae by Moore and coworkers
5.2 Individual Examples
provided exciting new and unique structures for organic synthesis [31]. Welwitindolinones inspired organic chemists to pursue their synthesis also because of their array of promising biological properties such as insecticidal activities. Welwitindolinone A (22) is a densely functionalized oxindole harboring three all-carbon quaternary centers, a neopentyl chlorine atom, and a striking spiro-fused cyclobutane. 12-epi-Fischerindoles I (103) and G (104) represent the most complex congeners within the fischerindole alkaloid family. A proposal for their biogenetic origin has been put forward by Moore [31], in which a monoterpenoid geranyl-type unit 99 is attacked by an electrophilic chlorine, which again reacts with an indole moiety and thereby links the two different building blocks to give 12-epi-hapalindole E (101, Scheme 5.14). Two pathways are now conceivable: The first leads to the formation of oxindole 102, which cyclizes to form the four-membered ring of welwitindolinone A (22). Alternatively, 101 can cyclize under acid catalysis to give 12-epi-fischerindole G (104) with its five-membered ring. This in turn is converted into 12-epi-fischerindole I (103) via an additional oxidation reaction. Cl
Cl+
H
99 NC N H
Cl
H H Cl+
NC
[O]
CN O N H 102
N H 12-epi-hapalindole E (101)
100
+
+
[H ]
[H ]
Cl Cl
CN
CN [O]
H Me Me
N H 12-epi-fischerindole I (103)
Scheme 5.14
H
Cl H Me Me
N H 12-epi-fischerindole G (104)
Me CN
H Me Me O
N H (+)-welwitindolinone A (22)
Biosynthesis proposal for welwitindolinone A (22) by Moore et al.
To date, two total syntheses of (+)-welwitindolinone A (22) have been completed, by the groups of Baran and Wood [32, 33]. Additionally, numerous uncompleted approaches have been reported by other groups [34]. The synthesis by Baran et al. is discussed here because it closely follows the biosynthetic proposal. In accord with the biosynthesis their strategy was based on the coupling of the terpene moiety with the indole fragment. For the synthesis of the terpene fragment they started from (S)-carvone, which was epoxidized in a Scheffer–Weitz reaction, and subsequent vinyl-magnesium bromide addition gave compound 106 with the quaternary carbon center installed. The alcohol functionality of 106 was converted into the chloride via N-chlorosuccinimide (NCS). The coupling of the terpene fragment 107 with indole was conducted via a copper(II)-mediated oxidative radical coupling to yield compound 108. Montmorillonite K-10 (MK-10) enabled cyclization of 108 to give tetracycle 109, which contained the carboskeleton of fischerindoles G
165
166
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus
and I. Functional group transformations introduced isonitrile functionality and led to the synthesis of (−)-fischerindole I (110). Oxidation of 110 with xenon difluoride gave (+)-welwitindolinone A (22) in 44% yield. When comparing the synthetic route with Moore’s biosynthetic proposal, strong parallels can be observed. Compound 108 has the same carboskeleton as 12-epi-hapalindole E (101) in Scheme 5.14, which is the central precursor in the biosynthetic proposal. The subsequent cyclization reaction that yields tetracycle 109 is strongly related to the conversion of 101 into 104 in Scheme 5.15. In addition, the last step, the oxidative rearrangement, which yields welwitindolinone A (22) with its remarkable cyclobutane, can be found in slightly different version in Scheme 5.14 (102→22). Cl O
O
a. NaOH; H2O2 b. LHMDS;
O
H
d. LHMDS then Cu(II)
c. NCS, PPh3
O
H2C=CHMgBr HO H 30% (S )-carvone (105)
Cl
H 106
N H
H 107
N H
32% 57%
Cl Cl CN
Cl
CN
Me H Me Me O
CN j. XeF2
F N
N H (+)-welwitindolinone A (22)
H Me Me
111
44%
N H
H Me Me
f. NaCNBH3 NH4OAc g. HCO2H DMT-MM h. COCl2, NEt3 i. DDQ
38%
(−)-fischerindole I (110)
108
e. MK-10
Cl O H N H 109
H Me Me
DMT-MM: 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium chloride
Scheme 5.15
Biomimetic synthesis of welwitindolinone A (22) by Baran et al.
5.2.7 (−)-Gelselegine
In recent years, several new indole and oxindole alkaloids have been isolated from Gelsemium elegans a plant that is known in Chinese folk medicine as ‘‘Kou-Wen’’ or ‘‘Hu-Man-Teng’’ [35]. These alkaloids have highly strained polycyclic structures and can be classified into six groups, the sarpagine, humantenine, gelselegine, gelsedine, koumine, and gelsemine types, based on their skeletal types. Sakai and coworkers have considered a possible biogenetic pathway. Their biosynthetic proposal (Scheme 5.16) starts with koumidine (112) as central building block [36]. Quaternization of the tertiary amine triggers fragmentation into (19Z)-anhydrovobasinediol (113). Oxidation of 113 into oxindole rankinidine (114) provides access to humantenidine-type alkaloids. To get from humantenidine- to the gelsedine-type alkaloids a ring contraction from a six-membered ring (humantenidine) to a five-membered ring in gelsedine has to take place. The hydroxymethyl
5.2 Individual Examples
167
[sarpagine-type] O
HO A
B
C
N H
H
H N Me
N H
N D
21 20
20
(19Z )-Anhydrovobasinediol (113)
koumidine (112) [gelsedine-type]
[gelselegine-type]
O
[humantenine-type]
O
O 21
D 20
D
HN D N O 21 OMe
N O N 20 H OMe gelselegine (20)
N O N H OMe gelsedine (115)
Scheme 5.16
OH
20
rankinidine (114)
Biosynthesis proposal for gelselegine (20) by Sakai et al.
group at C20 in gelselegine (20) comes from such a rearrangement/ring contraction. Gelsedine, which lacks the hydroxymethyl group, is the biogenetic follow-up product of gelselegine. In this way gelselegine somewhat represents the ‘‘missing link’’ in the biosynthesis between gelsedine (lacking C21), rankinidine, and the humantenidine/sarpagine type alkaloids. Sakai and coworkers investigated their proposal with a series of chemical transformations, and were able to describe the first synthesis of these chemically and biogenetically unique oxindole alkaloids, in a manner consistent with this biogenetic sequence (Scheme 5.17). They started from koumidine (112), which was transformed into (19Z)-anhydrovobasinediol-type structure 116, by reaction HO
O
H a. TrocCl; N
N H
N
N H
94%
koumidine (112)
Troc
b. OsO4
N O H Troc 117
116
O O
NH N O OMe Troc
CH2N2 h. TMAD; PBu3
120
30%
O
Scheme 5.17
l. PtO2, H2
64%
c. CH(OMe)3 Ac2O ∆ KOH; 75%
O
OH NH N O OH Troc H 119
OH k. NaIO4 N N O H OMe (−)-gelselegine (20)
OH OH
d. TMSCl e. OsO4
50%
O
N
O
f. BH3 SMe2 g. H2O2, Na2WO4
i. Zn AcOH j. 5d rt
O
H
MgO
H N N O H OMe gelsedine (115)
Biomimetic synthesis of gelselegine (20) by Sakai et al.
89%
N N H
O 118
Troc
168
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus
with Troc-chloride and subsequent fragmentation reaction. Treatment of 116 with osmium tetroxide not only gave oxidized indolinone system but also afforded diol 117. This in turn was converted into the cyclic orthoester, which under thermal conditions underwent elimination to give olefin 118 with opposite olefin geometry to that of 116. Treatment of 118 with trimethylsilyl chloride (TMSCl) shifted the exocyclic double bond into the ring to give an enamide. This was dihydroxylated again with osmium tetroxide to afford diol 119. This instituted the ring contraction of the former six-membered ring in 118 to the desired five-membered ring of gelselegine (20). The following transformations were needed to achieve this goal: continuing from 119, the indolinone moiety was reduced to the indoline with borane dimethyl sulfide complex. Reoxidation with hydrogen peroxide and sodium tungstate gave the N-hydroxylated indolinone moiety. The diol of 119 was treated under Mitsunobu type conditions to afford epoxide 120. Reductive cleavage of the Troc-protective group was followed by epoxide opening, when left at room temperature for five days, to give the desired (−)-gelselegine (20). To gain evidence that gelselegine is the actual precursor or ‘‘missing link’’ in the biosynthesis of gelsedine (115), gelselegine was submitted to sodium periodate conditions, giving the glycol cleavage product, which upon reduction with platinum oxide and hydrogen was converted into gelsedine (115). 5.2.8 Communesin, Calycanthines, and Chimonanthines
The calycanthaceous alkaloids were first isolated from the plant genus Calycanthus [37]. Early isolations of individual members showed a closely related skeleton that differed only in the aminal connectivity for the various natural products. The subtle differences in possible structures made the establishment of the relative and absolute stereochemistry of these compounds very difficult. The first calycanthoid that was characterized was (+)-calycanthine (21). Its structure was chemically established by Robinson and Woodward, and crystallographically by Hamor and Robertson, using the dihydrobromide dehydrated salt [38–40]. Soon after the publication of the crystal structure, the absolute stereochemistry of calycanthine (21) was determined using circular dichroism analysis by Mason [41]. The structure of (−)-chimonanthine (19), isolated from Chimonanthus fragrans by Hodson et al., was elucidated by Hamor and Robertson through X-ray analysis of the dihydrobromide salt. Chimonanthine bears two indoline units, each with an annulated pyrrolidine, that are 3,3 − connected. All possible stereoisomers, both the C2-symmetric and meso-isomers, have been identified. The absolute stereochemistry of the vicinal quaternary carbon centers of (−)-chimonanthine is identical to that of (+)-calycanthine, and these natural products equilibrate under acidic conditions (Scheme 5.18). The total synthesis of meso- and racemic chimonanthine and calycanthine via this oxidative coupling was performed by Scott and coworkers (Scheme 5.19) [40c]. They used methyl-magnesium bromide as a base to deprotonate the indole ring, which made it more susceptible to oxidative conditions. Subsequent treatment with iron(III)-chloride furnished dimerization
5.2 Individual Examples
MeN MeHN
H H N
169
Me H H N N
N
+ MeHN N NHMe 121 (d,l )
Dimerase
N H
+
MeHN
N N H H Me
N N H H Me meso -chimonanthine (123)
(−)-chimonanthine (19) H+
N
(Me)-8
Me N N
H N
NHMe N N H Me (+)-calycanthine (21)
122 (meso)
Scheme 5.18 Biosynthesis proposal for calycantheous alkaloids by Woodward and Robinson. MeHN
MeHN
Me N
N
H N
MeN +
a. MeMgI
N H 120
Scheme 5.19
b. FeCl3
H N
H
N
NHMe 121 (d,l )
N H
N Me
(±)-calycanthine (21)
N N H Me (±)-chimonanthine (19)
Biomimetic synthesis of calycantheous alkaloids by Scott et al.
product 121, which underwent cyclization to (±)-chimonanthine (19). Equilibration indeed gave (±)-calycanthine (21). An alternative method of oxidative dimerization of indoles was developed by Takayama and coworkers, which uses hypervalent iodine reagents as oxidants, and provides a rapid synthetic entry into the chimonanthine and calycanthine alkaloids [42]. Woodward and Robinson proposed the biosynthesis of these natural products more than 50 years ago and this proposal was later refined by B. Stoltz. This proposal is shown in Scheme 5.20. It contains an oxidative dimerization of two tryptamine units. In 2006, Stoltz et al. developed a more detailed biosynthetic concept that not only attempted to explain the formation of the calycantheous alkaloids but also of communesins alkaloids from common precursors (Scheme 5.20) [43b]. In analogy to Scheme 5.18, dimer (R,R)-124 could be cyclized and recyclized to form, in a cascade process via intermediates 125 and 126, the hexacyclic structure 127, which is N-prenylated to 128 and then oxidatively converted into 12 (Scheme 5.20). An analogous sequence starting form meso-129 would end up with 130, which is a reasonable precursor to the known alkaloid perophoramidine (131). A few years earlier, in 2003, Stoltz had provided a totally different biosynthesis that contained another natural product–aurantioclavine–as an intermediate in the biosynthesis of communesins [44]. In a model study (Scheme 5.21) some evidence was provided for this concept by generating ortho-iminoquinonemethide
170
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus
H2N
H H H N N
N
N
NH2
NH2
N
H N
H N
126 O
H N
N
O
prenylation
NH
N H
NH
N H
N
H N
H N
NH2
N
N Cl
130
Scheme 5.20 Extended biosynthesis proposal for calycantheous alkaloids and communesins by Stoltz et al. H H N
H H N
NH2
+
N N
132 (aurantioclavine) H
N HO 133
NH O
N Me
134 O
H N
O
N H
Boc N
Stoltz's model study Cl Cs2CO3 CH2Cl2, −78°C
Scheme 5.21
N Me
6 N N Me H R
99%
137
N
Me
HN
NBoc
138
TsN
H
N N H Me communesin B(12)
135
HN Ts 136
NH2
O
N H
Me
HN
Me N Br
N N H Cl perphoramidine (131)
NH
N H
129 (meso -compound of 124)
H
N N Me H communesin B (12)
128
127
N
NH2
N
125
(R,R)-124
H2N
H2N
N
Mg, NH4Cl 80%
139 R = Ts 140 R = H
Biomimetic study on the communesin alkaloids by Stoltz et al.
5.2 Individual Examples MeO2C
Br
CO2Me
Br
NH2
141
N Me
H N
N
TeocHN
142
HN N O Me O 143 TMS
Cs2CO3
65%
N B F
E
TBAF
H C
N N H Me 144
H N
CO2Me
G
AuCl(PPh3)2 AgOTf
CO2Me H
D
N N H Me H
89%
146
Scheme 5.22
CO2Me
N N H Me 145
Related study on communesin alkaloids by Funk et al.
137 from precursor 136, and adding it to the aurantioclavine derivative 138. In fact a diastereomeric mixture of polycycle 139 was obtained containing communesin rings F, E, D, C, and G. The 13 C NMR shifts for C6 (84.8 and 83.9 ppm) were in agreement with those of communesin B (82.4 ppm). Unfortunately, they were never able to complete the total synthesis. The group of Funk also pursued for the synthesis of communesins, and thereby followed a path closely related to Stoltz’s older biosynthesis proposal (Scheme 5.22) [45]. Starting from substituted tryptamine 141 and dibromide 142, aziridine 143 was formed under base catalysis. ortho-Quinomethide 144 was formed via TBAF deprotection of the carbamide moiety in compound 143, which proceeded under formation of ethylene and carbon dioxide, and indeed gave Diels–Alder product 145. Treatment of 145 with a gold(I)-reagent closed ring G of communesins in 146, and was the closest that Funk and coworkers could get to the natural product. To date two non-biomimetic total syntheses of communesin F have been completed [46, 47]. The communesin story is one of the rare cases where following a non-biomimetic route has lead to better results. This may be due to the uncertainty in the biosynthesis of these natural products and points out that there is still a lot of research needed to solve this puzzle. Most probably, if the biosynthetic proposal is improved a biomimetic synthesis of communesins will be more efficient than the currently completed ones. 5.2.9 (+)-11,11 -Dideoxyverticillin A
The fungal metabolite (+)-11,11 -dideoxyverticillin A (10) is a cytotoxic alkaloid, isolated from a marine Penicillium sp., that shows a densely functionalized intricate dimeric epi-dithiodiketopiperazine structure [48]. A hypothesis for the biosynthesis of this molecule envisages the reductive dimerization of monomer 147 followed
171
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus
172
O Me Me
H H N
SO2Ph H N
O
S N N S
O
O N H
Me
S N Me N S Me H O
Me
N N
Br O
(+)-11,11′-dideoxyverticillin A (10)
Figure 5.5
147
Dimerization strategy for the synthesis of (10) by Movassaghi et al.
by tetrahydroxylation and tetrathiolation reactions (Figure 5.5). Most probably the biomimetic dimerization will not be reductive but oxidative in nature. Brominated compound 147 serves as a surrogate to achieve the same C–C-bond formation under reductive conditions. Therefore, with respect to bond formation it can be regarded as ‘‘biomimetic’’ but not with respect to the reaction conditions. In the synthesis performed by Movassaghi et al. (Scheme 5.23), dipeptide 150 was converted into the diketopiperazine and then the indole nucleus was destroyed by the introduction of bromine to form intermediate 152 which was N-methylated [49]. The crucial dimerization was performed by reduction with a Co(I)-complex to provide dimer 153, presumably via a radical mechanism. This step is remarkable for its high stereoselectivity, as the 5,5-cis-fused ring system is generated exclusively. CO2H
CO2H a. LiHMDS, PhSO2Cl
NHBoc
NHBoc
71%
N H
N SO2Ph
148
HN
b. (L)-Alanin methyl ester, EDC, HOBt, Et3N
N SO2Ph 150
94%
149
84%
SO2Ph H N
O Me
N
f. MeI, K2CO3
O
MeN NMe O 153
N N H SO2Ph O
Me
g. CoCl(PPh3)3, acetone 38%
Br
c. TFA, CH2Cl2 d. morpholine, tBuOH
O NH
N H N SO2Ph O 152
CO2Me
O NHBoc
O
e. Br2, acetonitrile 76%
NH HN N SO2Ph O
Me
151
Scheme 5.23 Biomimetic dimerization to form (+)-11,11 -dideoxyverticillin A by Movassaghi et al.
Dihydroxylation of 153 with bis(pyridine)-silver(I) permanganate led to the tetracyclic diol 154 as a single diastereomer. TBS-protection of the tertiary OH-groups and removal of the benzenesulfonyl groups with sodium amalgam gave 155, which on exposure to potassium trithiocarbonate was converted into bis-dithioepanethione 156. Addition of ethanolamine afforded the unstable proposed biosynthetic precursor, which on oxidation with potassium triiodide finally furnished 10 (Scheme 5.24).
5.2 Individual Examples SO2Ph H N
O
153
a. Py2AgMnO4, CH2Cl2
HO MeN
63%
b. TBSCl c. Na-Hg
O NMe
O
154
H H N
O
N
N
TBSO MeN
O
48%
NMe
O
N N H SO2Ph O
173
N N H H
OH 155
OTBS
O
56% d. TFA, K2CS3 CH2Cl2
10
f. KI3, pyridine 62%
S
H H N
O N HS HS MeN
e. HOCH2CH2NH2 acetone
O
MeN
NMe
O N H
H
O
S
N
O
S S
O
N SH 157
H N
O
S N
N H
SH 156
Scheme 5.24 Endgame of the synthesis of (+)-11,11 -dideoxyverticillin A by Movassaghi et al.
5.2.10 (±)-Borreverine and (±)-Isoborreverine
(±)-Borrerine (158), (±)-borreverine (18), and (±)-isoborreverine (17) (Scheme 5.25) were originally isolated in 1973 from Borreria verticillata, a plant used in traditional medicine of West Africa for the treatment of skin diseases [50]. Subsequently, in 1991, these natural products were re-isolated from Flindersia fournieri, a plant from New Caledonia (West Pacific) [51a]. NHMe
N H
NMe
H+
H N H
H H H
NH NMe borrerine (158)
NHMe
borreverine (18)
+
N H
H H
N NHMe
isoborreverine (17)
Scheme 5.25 Biosynthesis proposal for borreverine (18) and isoborreverine (17) by Koch et al.
The co-occurrence of these dimeric alkaloids 18 and 17 with (±)-borrerine (158) suggested a simple chemical relationship between them. The biosynthetic proposal (Scheme 5.25) contains an acid-catalyzed dimerization of (±)-borrerine (158), which leads after the loss of a proton either to (±)-borreverine or (±)-isoborreverine. In fact Koch and coworkers mimicked this reaction and where able to isolate both dimeric indole alkaloids (Scheme 5.26) [52]. The optimum conditions were trifluoroacetic acid in benzene at 65 ◦ C for 30 min. These conditions furnished (±)-borreverine and
S
O
NMe
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus
174
NHMe
+
H
NMe
N H
Me N H
N H
(±)-borrerine (158)
N H
159
NHMe
+
H
N H
H 161
160
160
NHMe
NHMe −H
H N H NH
NHMe
+
N H
H H NMe
NH
H H
N H
H H NHMe
NH NHMe 162
163 (±)-borreverine (18) NHMe M. Koch conditions: TFA, benzene, 65°C 30 minutes 80% yield of 18 and 17
H N H
NHMe −H
H H
+
H H
N H H N
N
NHMe
NHMe
164
(±)-isoborreverine (17)
Scheme 5.26
Detailed reaction mechanism of the dimerization reaction.
O OH H
CO2Me O
OH
NMe NMe2 neoselaginellic acid 165
N Me
minfiensine 166
OH
HO strychnochromine 168
vincorine 167
O
OMe N
H N
N
H O
N
N
O
NH N H
O HO HO
meleagrine 169
N OH
H O
OMe
macroxine 173
Figure 5.6
phalarine 171
N H
O
OH jasminiflorine 174
N OO H meloscandonine 172
OH
H
H
H
N H
N
OH
isatisine A 170
O H N
N
H N
O
O
N HN OMe
N
N H
N
O N H
N
N N N OH
O
O terengganesine B 175
O mersicarpine 176
Selected indole alkaloids that contain a broken indole nucleus.
References
(±)-isoborreverine in about equal amounts and a total yield of 80%. Longer reaction times changed the ratio of (±)-borreverine to (±)-isoborreverine in favor of the latter. (±)-Borreverine (18) can be quantitatively transformed into (±)-isoborreverine (17) after reacting for 12 h under the conditions mentioned above. Additional evidence that the dimerization pathway might definitely be operative in Nature is provided by the fact that all three natural products not only were isolated from the same plant but, so far, were only isolated as racemate, which means that no enzymatic process is involved and the process is spontaneous and thermodynamically favored.
5.3 Conclusion
Indole alkaloids provide a wide variety of different structure types with varying degrees of structural complexity. Some selected structures that were not mentioned in this chapter are given in Figure 5.6. As shown above, this huge architectural diversity is inevitably linked to the characteristic reactivity pattern of the indole nucleus, which can undergo a plethora of skeletal rearrangements that are in most cases initiated by an oxidation reaction. In Nature these oxidations are mostly non-enzymatic processes. This makes this family of natural products amenable to biomimetic synthesis. In this way, the intrinsic reactivity of the molecules allows an increase of molecular complexity in very few steps, if the biosynthetic proposal is followed. References 1. (a) Wenkert, E. (1962) J. Am. Chem.
Soc., 84, 98–102; (b) Thomas, R. (1961) Tetrahedron Lett., 16, 544–553; (c) Money, T., Wright, I.G., McCapra, F., and Scott, A.I. (1965) Proc. Natl. Acad. Sci. U.S.A., 53, 901–903; (d) Battersby, A.R., Burnett, A.R., and Parsons, P.G. (1968) J. Chem. Soc., Chem. Commun., 1280–1281; (e) Battersby, A.R., Burnett, A.R., and Parsons, P.G. (1969) J. Chem. Soc. C, 1187–1192; (f) Battersby, A.R., Byrne, J.C., Kapil, R.S., Martin, J.A., Payne, T.G., Arigoni, D., and Loew, P. (1968) J. Chem. Soc., Chem. Commun., 951–953; (g) Eisenreich, W., Bacher, A., Arigoni, D., and Rohdich, F. (2004) Cell. Mol. Life Sci., 61, 1401–1426; (h) Contin, A., van der Heijden, R., Lefeber, A.W.M., and Verpoorte, R. (1998) FEBS Lett., 434, 413–416. 2. For syntheses, see: (a) Chavan, S.P., Pathak, A.B., and Kalkote, U.R. (2007)
Synlett, 2635–2638; (b) Chavan, S.P., Pasupathy, K., Venkatraman, M.S., and Kale, R.R. (2004) Tetrahedron Lett., 45, 6879–6882; (c) Bennasar, M.-L., Zulaica, E., Juan, C., Alonso, Y., and Bosch, J. (2002) J. Org. Chem., 67, 7465–7474; (d) Chavan, S.P. and Venkatraman, M.S. (1998) Tetrahedron Lett., 39, 6745–6748; (e) Ciufolini, M.A. and Roschangar, F. (1996) Angew. Chem. Int. Ed., 35, 1692–1694; (f) Shen, W., Coburn, C.A., Bornmann, W.G., and Danishefsky, S.J. (1993) J. Org. Chem., 58, 611–617; (g) Curran, D.P. and Liu, H. (1992) J. Am. Chem. Soc., 114, 5863–5864; (h) Kametani, T., Ohsawa, T., and Ihara, M. (1980) Heterocycles, 14, 951–953; (i) Tang, C. and Rapoport, H. (1972) J. Am. Chem. Soc., 94, 8615–8616; (j) Stork, G. and Schultz, A.G. (1971) J. Am. Chem. Soc., 93, 4074–4075.
175
176
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus 3. Wall, M.E., Wani, M.C., Cook, C.E.,
4.
5.
6. 7.
8.
9.
45, 3113–3115; (g) Chavan, S.P. and Palmer, K.H., McPhail, A.I., and Sim, Venkatraman, M.S. (2005) Arkivoc, G.A. (1966) J. Am. Chem. Soc., 88, 165–169; (h) Kanazawa, A., Muniz, 3888–3890. M.N., Baumlova, B., Ljungdahl, N., and (a) Ulukan, H. and Swaan, P.W. (2002) Greene, A.E. (2008) Synlett, 2275–2278; Drugs, 62(2), 2039–2057; (b) Lu, A.J., (i) Liu, G.-S., Dong, Q.-L., Yao, Y.-S., Zheng, Z.S., Zou, H.J., Luo, X.M., and and Yao, Z.-J. (2008) Org. Lett., 10, Jiang, H.L. (2007) Eur. J. Med. Chem., 5393–5396; (j) Tagami, K., Nakazawa, 42, 307–314; (c) Adams, D.J., Wahl, N., Sano, S., and Nagao, Y. (2000) HetM.L., Flowers, J.L., Sen, B., Colvin, M., erocycles, 53, 771–776; (k) Tang, C.-J., Dewhirst, M.W., Manikumar, G., and Babjak, M., Anderson, R.J., Greene, Wani, M.C. (2005) Cancer Chemother. A.E., and Kanazawa, A. (2006) Org. Pharm., 57(2), 145–154. Biomol. Chem., 4, 3757–3759; (l) Wu, Wenkert, E., Dave, I.G., Lewis, R.G., and X.-J., Guo, Q.-C., Zhang, Y., Chen, Y.-Y., Sprague, P.W. (1967) J. Am. Chem. Soc., and Wang, X.-D. (2008) Anhui Yiyao, 89, 6741–6745. 12, 783–785; (m) Yu, S., Luo, Y., Liu, Winterfeldt, E. (1971) Justus Liebigs Ann. H., Liu, H., and Lu, W. (2010) Monatsh. Chem., 745, 23–30. Chem., 141, 245–249; (n) Zhang, L., (a) Hutchinson, C.R., Heckendorf, Bao, Y., and Chen, F. (2008) ZhongA.H., Daddona, P.E., Hagaman, E., guo Yiyao Gongye Zazhi, 39, 481–483; and Wenkert, E. (1974) J. Am. Chem. (o) Zhou, H.-B., Liu, G.-S., and Yao, Soc., 96, 5609–5611; (b) Hutchinson, Z.-J. (2007) Org. Lett., 9, 2003–2006; C.R. (1981) Tetrahedron, 37, 1047–1065; (p) Thomas, O.P., Zaparucha, A., and (c) Takayama, H., Kitajima, M., and Husson, H.-P. (2002) Eur. J. Org. Chem., Aimi, N. (1999) J. Synth. Org. Chem. 157–162; (q) Dumas, C., Kan-Fan, C., Jpn., 57, 181–193; (d) Carte, B.K., de Royer, J., and Husson, H.-P. (2000) Brosse, C., Eggleston, D., Hemling, M., Eur. J. Org. Chem., 3601–3606; (r) For Mentzer, M., Poehland, B., Troupe, N., an improved reagent, see: Jiang, W., Westley, J.M., and Hecht, S.M. (1990) Zhang, X., and Sui, Z. (2003) Org. Lett, Tetrahedron, 46, 2747–2760; (e) Aimi, 5, 43–46. 10. For comprehensive reviews see: (a) N., Hoshino, H., Nishimura, M., Sakai, Kita, Y. (2005) Curr. Org. Chem., 9, S., and Haginawa, J. (1990) Tetrahe1567–1588; (b) Makar’eva, T.N., dron Lett., 31, 5169–5170; (f) Kitajima, Krasokhin, V.B., Guzii, A.G., and M., Masumoto, S., Takayama, H., and Stonik, V.A. (2010) Chem. Nat. Comp., Aimi, N. (1997) Tetrahedron Lett., 38, 46, 152–153; (c) Jeon, J.-e., Na, Z., 4255–4258. (a) Boch, M., Korth, T., Nelke, J.M., Jung, M., Lee, H.-S., Sim, C.J., Nahm, Pike, D., Radunz, H., and Winterfeldt, K., Oh, K.-B., and Shin, J. (2010) J. Nat. E. (1972) Chem. Ber., 105, 2126–2142; Prod., 73, 258–262; (d) Na, M.K., (b) Krohn, K. and Winterfeldt, E. (1975) Ding, Y., Wang, B., Tekwani, B.L., Chem. Ber., 108, 3030–3042. Schinazi, R.F., Franzblau, S., Kelly, M., (a) Anderson, R.J., Raolji, G.B., Stone, R., Li, X.-C., Ferreira, D., and Kanazawa, A., and Greene, A.E. (2005) Hamann, M.T. (2010) J. Nat. Prod., Org. Lett., 7, 2989–2991; (b) Blagg, 73(3), 383–387; (e) Lang, G., Pinkert, B.S.J. and Boger, D.L. (2002) TetraA., Blunt, J.W., and Munro, M.H.G. hedron, 58, 6343–6349; (c) Chavan, (2005) J. Nat. Prod., 68, 1796–1798; S.P., Dhawane, A.N., and Kalkote, U.R. (f) Reyes, F., Martin, R., Rueda, A., (2010) Tetrahedron Lett., 51, 3099–3101; Fernandez, R., Montalvo, D., Gomez, C., (d) Chavan, S.P., Pathak, A.B., and and Sanchez-Puelles, J.M. (2004) J. Nat. Kalkote, U.R. (2007) Synlett, 2635–2638; Prod., 67, 463–465; (g) Gunasekera, (e) Chavan, S.P., Pathak, A.B., and S.P., Zuleta, I.A., Longley, R.E., Wright, Kalkote, U.R. (2007) Tetrahedron Lett., A.E., and Pomponi, S.A. (2003) J. Nat. 48, 6561–6563; (f) Chavan, S.P. and Prod., 66, 1615–1617; (h) Dijoux, M.-G., Rasapalli, S. (2004) Tetrahedron Lett., Gamble, W.R., Hallock, Y.F., Cardellina
References
11.
12.
13. 14.
II, J.H., Van Soest, R., and Boyd, M.R. (d) Kametani, T., Kanaya, N., and Ihara, (1999) J. Nat. Prod., 62, 636–637; (i) M. (1980) J. Am. Chem. Soc., 102(11), Gunasekera, S.P., McCarthy, P.J., 3974–3975; (e) Lee, B.H., Clothier, Longley, R.E., Pomponi, S.A., Wright, M.F., and Pickering, D.A. (1997) J. Org. E., Lobkovsky, E., and Clardy, J. (1999) Chem., 62(22), 7836–7840; (f) Lee, J. Nat. Prod., 62, 173–175; (j) Yang, B.H. and Clothier, M.F. (1997) J. Org. A., Baker, B.J., Grimwade, J., Leonard, Chem., 62(6), 1795–1798; (g) Trost, A., and McClintock, J.B. (1995) J. Nat. B.M., Cramer, N., and Bernsmann, Prod., 58, 1596–1599; (k) Perry, N.B., H. (2007) J. Am. Chem. Soc., 129(11), Blunt, J.W., Munro, M.H.G., Higa, 3086–3087. 15. (a) Birch, A.J. and Wright, J.J. (1969) T., and Sakai, R. (1988) J. Org. Chem., J. Chem. Soc., Chem. Commun., 53, 4127–4128; (l) Perry, N.B., Blunt, 644–645; (b) Birch, A.J. and Wright, J.W., and Munro, M.H.G. (1988) TetraJ.J. (1970) Tetrahedron, 26, 2329–2344; hedron, 44, 1727–1734; (m) Perry, (c) Birch, A.J. and Russell, R.A. (1972) N.B., Blunt, J.W., McCombs, J.D., and Tetrahedron, 28, 2999–3008; (d) Munro, M.H.G. (1986) J. Org. Chem., 51, Yamazaki, M., Okuyama, E., Kobayashi, 5476–5478. Lill, R.E., Major, D.A., Blunt, J.W., M., and Inoue, H. (1981) TetraheMunro, M.H.G., Battershill, C.N., dron Lett., 22, 135–136; (e) Ondeyka, McLean, M.G., and Baxter, R.L. (1995) J.G., Goegelman, R.T., Schaeffer, J.M., J. Nat. Prod., 58, 206–211. Kelemen, L., and Zitano, L. (1990) J. An(a) Nishiyama, S., Cheng, J.F., Tao, tibiot., 43, 1375–1379; (f) Liesch, J.M. X.L., and Yamamura, S. (1991) Tetraand Wichmann, C.F. (1990) J. Antibiot., hedron Lett., 32, 4151–4154; (b) Kita, 43, 1380–1386; (g) Blanchflower, S.E., Y., Tohma, H., Inagaki, M., Hatanaka, Banks, R.M., Everett, J.R., and Manger, K., and Yakura, T. (1992) J. Am. Chem. B.R. (1991) J. Antibiot., 44, 492–497; (h) Soc., 114(6), 2175–2180; (c) Roberts, Blanchflower, S.E., Banks, R.M., Everett, D., Joule, J.A., Bros, M.A., and Alvarez, J.R., and Reading, C. (1993) J. AnM. (1997) J. Org. Chem., 62(3), 568–577; tibiot., 46, 1355–1363; (i) Banks, R.M., (d) White, J.D., Yager, K.M., and Yakura, Blanchflower, S.E., Everett, J.R., Manger, T. (1994) J. Am. Chem. Soc., 116(5), B.R., and Reading, C. (1997) J. Antibiot., 1831–1838; (e) Knoelker, H.J. and 50, 840–846; (j) Whyte, A.C., Gloer, Hartmann, K. (1991) Synlett, 428–430; J.B., Wicklow, D.T., and Dowd, P.F. (f) Kubiak, G. and Confalone, P. (1990) (1996) J. Nat. Prod., 59, 1093–1095; (k) Tetrahedron Lett., 31, 3845–3848; (g) Polonsky, J., Merrien, M.-A., Prang´e, Wada, Y., Fujioka, H., and Kita, Y. T., and Pascard, C. (1980) J. Chem. Soc., (2010) Marine Drugs, 8, 1394–1416; (h) Chem. Commun., 601–602; (l) Prang´e, Tohma, H., Harayama, Y., Hashizume, T., Billion, M.-A., Vuilhorgne, M., M., Iwata, M., Kiyono, Y., Egi, M., Pascard, C., Polonsky, J., and Moreau, S. and Kita, Y. (2003) J. Am. Chem. Soc., (1981) Tetrahedron Lett., 22, 1977–1980; 125(37), 11235–11240; (i) Tohma, H., (m) Sugie, Y., Hirai, H., Inagaki, T., Harayama, Y., Hashizume, M., Iwata, Ishiguro, M., Kim, Y.-J., Kojima, Y., M., Egi, M., and Kita, Y. (2002) Angew. Sakakibara, T., Sakemi, S., Sugiura, A., Chem. Int. Ed., 41(2), 348–350. Suzuki, Y., Brennan, L., Duignan, J., Aubart, K.M. and Heathcock, C.H. Huang, L.H., Sutcliffe, J., and Kojima, (1999) J. Org. Chem., 64(1), 16–22. N. (2001) J. Antibiot., 54, 911–916. (a) Schkeryantz, J.M., Woo, J.C.G., 16. (a) Sanz-Cervera, J.F., Williams, R.M., Siliphaivanh, P., Depew, K.M., and Alberto, M.J., Maria Lopez-Sanchez, Danishefsky, S.J. (1999) J. Am. Chem. J., Gonzalez, F., Eugenia Martinez, Soc., 121(51), 11964–11975; (b) Ritchie, M., and Sancenon, F. (2000) TetraheR. and Saxton, J.E. (1981) Tetrahedron, 56(34), 6345–6358; (b) Williams, dron, 37(24), 4295–4303; (c) Kametani, R.M., Sanz-Cervera, J.F., Sancenon, F., T., Kanaya, N., and Ihara, M. (1981) Marco, J.A., and Halligan, K.M. (1998) J. Chem. Soc., Perkin Trans. 1, 959–963; Bioorg. Med. Chem., 6(8), 1233–1241;
177
178
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus
17.
18. 19.
20.
21.
22. 23.
(c) Williams, R.M., Sanz-Cervera, J.F., Sancenon, F., Marco, J.A., and Halligan, K. (1998) J. Am. Chem. Soc., 120(5), 1090–1091. (a) Hilvert, D., Hill, K.W., Nared, K.D., and Auditor, M.T.M. (1989) J. Am. Chem. Soc., 111, 9261–9262; (b) Heine, A., Stura, E., Yli-Kauhaluoma, J.T., Gao, C., Deng, Q., Beno, B.R., Houk, K.N., Janda, K.D., and Wilson, I.A. (1998) Science, 279, 1934–1940; (c) Kasahara, K., Miyamoto, T., Fujimoto, T., Oguri, H., Tokiwano, T., Oikawa, H., Ebizuka, Y., and Fujii, I. (2010) ChemBioChem, 11, 1245–1252; (d) Katayama, K., Kobayashi, T., Chijimatsu, M., Ichihara, A., and Oikawa, H. (2008) Biosci. Biotechnol. Biochem., 72, 604–607; (e) Oikawa, H. (2001) Kagaku Seibutsu, 39, 424–426; (f) Oikawa, H. (2005) Bull. Chem. Soc. Jpn., 78, 537–554; (g) Ose, T., Oikawa, H., and Tanaka, I. (2004) Seibutsu Butsuri, 44, 32–35; (h) Ose, T., Yao, M., Oikawa, H., and Tanaka, I. (2003) Nippon Kessho Gakkaishi, 45, 384–390. Tarasow, T.M. and Eaton, B.E. (1999) Cell Mol. Life Sci., 55, 1463–1472. (a) Williams, R.M., Glinka, T., Kwast, E., Coffman, H., and Stille, J.K. (1990) J. Am. Chem. Soc., 112(2), 808–821; (b) Williams, R.M., Cao, J., Tsujishima, H., and Cox, R.J. (2003) J. Am. Chem. Soc., 125(40), 12172–12178; (c) Williams, R.M. (2002) Chem. Pharm. Bull., 50(6), 711–740; (d) Williams, R.M. and Cox, R.J. (2003) Acc. Chem. Res., 36(2), 127–139; (e) Cushing, T.D., Sanz-Cervera, J.F., and Williams, R.M. (1996) J. Am. Chem. Soc., 118(3), 557–579; (f) Greshock, T.J., Grubbs, A.W., and Williams, R.M. (2007) Tetrahedron, 63(27), 6124–6130. Sanz-Cervera, J.F. and Williams, R.M. (2002) J. Am. Chem. Soc., 124(11), 2556–2559. Adams, L.A., Valente, M.W.N., and Williams, R.M. (2006) Tetrahedron, 62(22), 5195–5200. Nising, C.F. (2010) Chem. Soc. Rev., 39(2), 591–599. (a) Qian-Cutrone, J., Huang, S., Shu, Y.-Z., Vyas, D., Fairchild, C., Menendez, A., Krampitz, K., Dalterio,
24.
25.
26. 27.
28.
R., Klohr, S.E., and Gao, Q. (2002) J. Am. Chem. Soc., 124, 14556–14557; (b) Qian-Cutrone, J., Krampitz, K.D., Shu, Y.-Z., Chang, L.-P., Lowe, S.E. (2001) Patent US 6291461; (2001) Chem. Abstr., 135, 236411. (a) Wulff, J.E., Herzon, S.B., Siegrist, R., and Myers, A.G. (2007) J. Am. Chem. Soc., 129(16), 4898–4899; (b) Herzon, S.B. and Myers, A.G. (2005) J. Am. Chem. Soc., 127(15), 5342–5344. (a) Baran, P.S., Guerrero, C.A., Ambhaikar, N.B., and Hafensteiner, B.D. (2005) Angew. Chem. Int. Ed., 44(4), 606–609; (b) Baran, P.S., Hafensteiner, B.D., Ambhaikar, N.B., Guerrero, C.A., and Gallagher, J.D. (2006) J. Am. Chem. Soc., 128(26), 8678–8693. Greshock, T.J. and Williams, R.M. (2007) Org. Lett., 9(21), 4255–4258. (a) For the debut of a chartelline alkaloid, see: Chevolot, L., Chevolot, A.-M., Gajhede, M., Larsen, C., Anthoni, U., and Christophersen, C. (1985) J. Am. Chem. Soc., 107, 4542–4543; (b) For the isolation of chartelline C, see: Anthoni, U., Chevolot, L., Larsen, C., Nielsen, P.H., and Christophersen, C. (1987) J. Org. Chem., 52, 4709–4712. For studies towards the chartellines and related alkaloids, see: (a) Lin, X. and Weinreb, S.M. (2001) Tetrahedron Lett., 42, 2631–2633; (b) Pinder, J.L. and Weinreb, S.M. (2003) Tetrahedron Lett., 44, 4141–4143; (c) Nishikawa, T., Kajii, S., and Isobe, M. (2004) Chem. Lett., 33, 440–441; (d) Korakas, P., Chaffee, S., Shotwell, J.B., Duque, P., and Wood, J.L. (2004) Proc. Natl. Acad. Sci. U.S.A., 101, 12054–12057; (e) Nishikawa, T., Kajii, S., and Isobe, M. (2004) Synlett, 2025–2027; (f) Sun, C., Lin, X., and Weinreb, S.M. (2006) J. Org. Chem., 71, 3159–3166; (g) Sun, C., Camp, J.E., and Weinreb, S.M. (2006) Org. Lett., 8, 1779–1781; (h) Black, P.J., Hecker, E.A., and Magnus, P. (2007) Tetrahedron Lett., 48(36), 6364–6367; (i) Kajii, S., Nishikawa, T., and Isobe, M. (2008) Tetrahedron Lett., 49(4), 594–597; (j) Kajii, S., Nishikawa, T., and Isobe, M. (2008) Chem. Commun., 27, 3121–3123.
References 29. (a) Baran, P.S., Shenvi, R.A., and
30. 31.
32.
33.
34.
35.
Mitsos, C.A. (2005) Angew. Chem. Int. Ed., 44, 3714–3717; (b) Black, P.J., Hecker, E.A., and Magnus, P. (2007) Tetrahedron Lett., 48(36), 6364–6367. Baran, P.S. and Shenvi, R.A. (2006) J. Am. Chem. Soc., 128(43), 14028–14029. Stratmann, K., Moore, R.E., Bonjouklian, R., Deeter, J.B., Patterson, G.M.L., Shaffer, S., Smith, C.D., and Smitka, T.A. (1994) J. Am. Chem. Soc., 116, 9935–9942. (a) Baran, P.S., Maimone, T.J., and Richter, J.M. (2007) Nature, 446, 404–408; (b) Baran, P.S. and Richter, J.M. (2005) J. Am. Chem. Soc., 127(44), 15394–15396. Reisman, S.E., Ready, J.M., Weiss, M.M., Hasuoka, A., Hirata, M., Tamaki, K., Ovaska, T.V., Smith, C.J., and Wood, J.L. (2008) J. Am. Chem. Soc., 130(6), 2087–2100. Studies toward the welwitindolinones: (a) Wood, J.L., Holubec, A.A., Stoltz, B.M., Weiss, M.M., Dixon, J.A., Doan, B.D., Shamji, M.F., Chen, J.M., and Heffron, T.P. (1999) J. Am. Chem. Soc., 121, 6326–6327; (b) Deng, H. and Konopelski, J.P. (2001) Org. Lett., 3, 3001–3004; (c) Jung, M.E. and Slowinski, F. (2001) Tetrahedron Lett., 42, 6835–6838; (d) Lopez-Alvarado, P., Garcia-Granda, S., Alvarez-Rua, C., and Avendano, C. (2002) Eur. J. Org. Chem., 1702–1707; (e) Ready, J.M., Reisman, S.E., Hirata, M., Weiss, M.M., Tamaki, K., Ovaska, T.V., and Wood, J.L. (2004) Angew. Chem. Int. Ed., 43, 1270–1272; (f) MacKay, J.A., Bishop, R.L., and Rawal, V.H. (2005) Org. Lett., 7, 3421–3424; (g) Baudoux, J., Blake, A.J., and Simpkins, N.S. (2005) Org. Lett., 7, 4087–4089; (h) Brailsford, J.A., Lauchli, R., and Shea, K.J. (2009) Org. Lett., 11(22), 5330–5333; (i) Trost, B.M. and McDougall, P.J. (2009) Org. Lett., 11(16), 3782–3785; (j) Zheng, P. and Harmata, M. (2007) Chemtracts, 20(1), 20–31. (a) Lin, L., Cordell, G.A., Ni, C., and Clardy, J. (1990) Phytochemistry, 29(9), 3013–3017; (b) Yamada, Y., Kitajima, M., Kogure, N., Wongseripipatana, S.,
36.
37.
38. 39. 40.
41. 42.
43.
44. 45.
46.
47.
48.
49.
50.
and Takayama, H. (2009) Tetrahedron Lett., 50(26), 3341–3344. Takayama, H., Kitajima, M., Ogata, K., and Sakai, S. (1992) J. Org. Chem., 57(17), 4583–4584. Hendrickson, J.B., Rees, R., and Goschke, R. (1962) Proc. Chem. Soc., 383–384. Robinson, R. and Teuber, H. (1954) J. Chem. Ind., 783. Woodward, R.B., Yand, N., and Katz, T.J. (1960) Proc. Chem. Soc., 76–78. (a) Hamor, T.A., Robertson, J.M., Shrivastave, H.N., and Silverton, J.V. (1960) Proc. Chem. Soc., 78–80; (b) Hamor, T.A. and Robertson, J.M. (1962) J. Chem. Soc., 194–205. (c) Scott, A.I., McCapra, F., and Hall, E.S., (1964) J. Am. Chem. Soc., 86, 302–303. Mason, S.F. (1962) Proc. Chem. Soc., 362. Ishikawa, H., Aimi, N., and Takayama, H. (2002) Tetrahedron Lett., 43, 5637–5639. (a) Numata, A., Takahashi, C., Ito, Y., Takada, T., Kawai, K., Usami, Y., Matsamura, E., Imachi, M., Ito, T., and Hasegawa, T. (1993) Tetrahedron Lett., 34, 2355–2358; (b) May, J.A. and Stoltz, B. (2006) Tetrahedron, 62, 5262–5271. May, A., Zeidan, R.K., and Stoltz, B.M. (2003) Tetrahedron Lett., 44, 1203–1205. (a) Crawley, S.L. and Funk, R.L. (2003) Org. Lett., 5, 3169–3171; (b) Crawley, S.L. and Funk, R.L. (2006) Org. Lett., 8, 3995–3998. Yang, J., Wu, H., and Shen Qin, J. (2007) J. Am. Chem. Soc., 129, 13794–13795. Liu, P., Seo, J.H., and Weinreb, S.M. (2010) Angew. Chem. Int. Ed., 49(11), 2000–2003. (a) Son, B.W., Jensen, P.R., Kauffman, C.A., and Fenical, W. (1999) Nat. Prod. Res., 13, 213–222; (b) Gardiner, D.M., Waring, P., and Howlett, B.J. (2005) Microbiology, 151, 1021–1032. Kim, J., Ashenhurst, J.A., and Movassaghi, M. (2009) Science, 324(5924), 238–241. Pousset, J.L., Kerharo, J., Maynart, G., Monseur, X., Cav´e, A., and Goutarel, R. (1973) Phytochemistry, 12(9), 2308–2310.
179
180
5 Biomimetic Synthesis of Alkaloids with a Modified Indole Nucleus 51. (a) Balde, A.M., Pieters, L.A.,
Gergely, A., Wray, V., Claeys, M., and Vlietinck, A.J. (1991) Phytochemistry, 30(3), 997–1000; (b) Tillequin, F. and Koch, M. (1979) Phytochemistry, 18(9), 1559–1561; (c) Tillequin, F., Rousselet, R.,
Koch, M., Bert, M., and S´evenet, T. (1979) Ann. Pharm. Fr., 37(11–12), 543–548. 52. Tillequin, F., Koch, M., Pousset, J.L., and Cav´e, A. (1978) J. Chem. Soc., Chem. Commun., (19), 826–828.
181
6 Biomimetic Synthesis of Manzamine Alkaloids* Romain Duval and Erwan Poupon
6.1 Introduction
Since the isolation of manzamine A (1) in 1986 [1] the manzamine group of alkaloids1) has been enriched continuously by the discovery of novel marine compounds with unprecedented molecular architectures (Figures 6.1 and 6.2). Nearly 100 of alkaloids have been isolated to date from sponges of the order Haplosclerida and Dictyoceratida. This apparently heterogeneous family of alkaloids2) encompasses: 1) 3-Alkylpyridines and 3-alkylpyridinium salts (see the examples of monomeric theonelladin A (2) [2], oligomeric cyclostellettamine A (3) [3], and niphatoxin B (4) [4], and also polymeric structures such as halitoxins (5) [5]–Figure 6.1); 2) Elaborated and sometimes highly complex structures; see the examples of manzamine A (1) [1], sarain A (6) [6], keramaphidin B (7) [7], halicyclamine A (8) [8], manadomanzamine A (9) [9], nakadomarine A (10) [10], madangamine C (11) [11], misenine (12) [12], and upenamide A (13) [13] (Figure 6.2). Despite their high structural diversity and variable sponge origin, the manzamine alkaloids exhibit common structural features, such as polycyclic bis-nitrogenated cores and macrocyclic alkyl loops, suggesting a common biosynthetic origin. This led to the proposal of ‘‘universal’’ biogenetic hypotheses for these complex secondary compounds, and motivated their biomimetic synthesis by several research groups.3) ∗
In memory of the late Dr Christian Marazano whose creativity and humanity will always inspire us. 1) The term ‘‘manzamine alkaloids’’ will be used throughout the chapter, as recommended by Marazano and colleagues, instead of less specific ‘‘3-alkylpiperidine alkaloids.’’ 2) Although identical alkaloids were sometimes isolated from different sponges, only
the one organism from which the molecule was first characterized is given in this chapter. 3) Of particular interest and with important consequences among sponge-derived secondary metabolites is the real origin of the molecules. Whether they are produced by the sponge itself or by associated symbionts raises exciting questions (who possess the genes?).
Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
182
6 Biomimetic Synthesis of Manzamine Alkaloids
theonelladin A 2 [Theonella swinhoei ] NH2
monomers: N dimers and oligomers:
5
cyclostellettamine A 3 [Stelletta maxima]
N
polymers:
N Cl
N
niphatoxin B 4 [Niphates sp.]
N n
n = 2-5
Me
N
n
Me
n
N
N
6
("C-N" connection)
N
("C-C" connection)
Me
N
halitoxins 5 [Haliclona spp.]
Figure 6.1 Examples of simple ‘‘manzamine alkaloids’’: 3-alkylpyridines and pyridinium salts.
While numerous articles have already reviewed these fascinating alkaloids and their total syntheses [14], to the best of our knowledge none has yet concentrated on comparing their perceived biogenesis and biomimetic synthesis. This chapter reviews several possible biogenetic relationships (mapped in Schemes 6.1, 6.24, 6.34, 6.36, and 6.38 and, mainly, Scheme 6.42) between the most representative members of the manzamine alkaloids. This tree-like, ‘‘from simple to complex,’’ description integrates biomimetic chemistry studies to illustrate how, and to what extent, some hypotheses were validated or invalidated by experimental synthesis. Comprehensive reviews on the structure and sources of manzamine alkaloids as well as their total synthesis and biological activities will be found elsewhere [14]. From the discovery of manzamine A to the brilliant intuitions of Baldwin and Marazano and to the latest development in total synthesis we will embark on a journey that covers 25 years of discoveries of fascinating natural substances and biosynthetically driven chemical endeavors (see timescale on Figure 6.3).
6.2 Two Complementary Hypotheses: An ‘‘Acrolein Scenario’’ and a ‘‘Malondialdehyde Scenario’’ 6.2.1 From Fatty Aldehydes Precursors to Simple 3-Alkyl-Pyridine Alkaloids
The manzamine alkaloids, which feature large alkyl chains or loops, must at least partly originate from polyacetate metabolism. In a memorable paper entitled ‘‘On the biosynthesis of manzamines’’ published in 1992, this observation led Baldwin and Whitehead from the University of Oxford to propose fatty dialdehydes
N
H
N
O
Figure 6.2
N
OH
N H
N
N
N
keramaphidin B 7 [Amphimedon sp.]
madangamine C 11 [Xestospongia ingens]
manzamine A 1 [Haliclona sp.]
H
N
N H
N
N
H
HN
N
H
x
O y
halicyclamine A 8 [Haliclona sp.]
N
H
misenine 12 [Reniera sp.]
Examples of complex manzamines: macrocyclic polycyclic alkaloids.
nakadomarine A 10 [Amphimedon sp.]
N
H
sarain A 6 HO [Reniera sarai ]
HO
N
O
macrocyclic complex alkaloids:
N
HO
O
OH
HN
N
O
H
N
upenamide A 13 [Echinochalina sp.]
O
H
manadomanzamine A 9 [Acanthostrongylophora sp.]
NH
N
O
6.2 Two Complementary Hypotheses: An ‘‘Acrolein Scenario’’ and a ‘‘Malondialdehyde Scenario’’ 183
N
3
NH2
see scheme 6.2 for details
−2 H2O and oxidation if X=H
15
3
−2 H2O
NH2
O
18
N
O
15
N
O
X
oxidation
N
3
cyclostellettamine A type
for details
N
see scheme 6.11 for details
−2 H2O X=OH (17) see scheme 6.2
O
H2N
N
−2 H2O and reduction
xestospongin O type
X = H: acrolein 16 O X = OH: malondialdehyde 17
dimerization or oligomerization
NH2
X
see scheme 6.2 X=H (16) for details
X
O
15
O transamination
14
Scheme 6.1 Proposed biosynthesis of 3-alkylpyridine alkaloids: (i) monomers: need for an exogenous nitrogen source (ammonia equivalents) and (ii) dimers (and oligomers): self-aminating process, that is, one amino-aldehyde partner is the nitrogen source for the other. Lipophilic chains and macrocycles of the natural alkaloids are depicted by loops in Schemes 6.1, 6.11, 6.25 and 6.30, allowing a vision of the possible relationships between heterocyclic cores of representative alkaloids, irrespective of specific structural differences (such as chain length, unsaturations, etc.).
N
theonelladine A type NH 2
"NH3"
O
existence of a biosynthetic Zincke-type reaction?
X
O
O
O
monomer formation
fatty acid catabolism
184
6 Biomimetic Synthesis of Manzamine Alkaloids
6.2 Two Complementary Hypotheses: An ‘‘Acrolein Scenario’’ and a ‘‘Malondialdehyde Scenario’’
185
O N H
N H
N
H
N
O N N N N 1986: discovery of OH manzamine A by 1992: the "acrolein scenario" HO Higa and coll. by Baldwin and HO Whitehead 1998: biomimetic synthesis of 2006: total synthesis of keramaphidin B by Baldwin and coll. sarain A by Overman and coll.
1980
1990
2000
2010 H
O
Figure 6.3
1998: a unified alternative "malonaldehyde scenario" by Marazano and coll. 1998: first total synthesis of manzamine A by Winkler and coll.
HN
O
N
H
H
2010: total synthesis of haliclonacyclamine C by Sulikowski and coll.
Selected milestones in 25 years of manzamine alkaloid chemistry.
(possibly coming from fatty acid catabolism) as universal, bifunctional precursors of what would become the ‘‘manzamine family’’ [15]. According to this seminal hypothesis, dialdehydes in C8 –C16 such as 14 would be monoaminated with either pyridoxamine (via transamination) or ammonia (via reductive amination) to yield amino-aldehydes 15 (Scheme 6.1). Also plausibly produced by fatty acid degradation (see below), acrolein 16 (the original C3 species hypothesized by Baldwin) or malondialdehyde 17 (alternative C3 species proposed by Marazano in 1998 at the Institut de Chimie des Substances Naturelles in Gif-sur-Yvette [16]) would react with 15 and a source of ammonia, furnishing pyridine alkaloids of the theonelladin A type. On the other hand, dimerization of amino-aldehydes 15 in the presence of two equivalents of acrolein 16 or malondialdehyde 17 would furnish alkaloids of the cyclostellettamine type via dihydropyridiniums 18. Formation of alkaloids of the xestospongin type would occur in the case of alkyl chains β-hydroxylated relative to the nitrogen (Section 6.3.2). Scheme 6.2 presents detailed mechanisms for pyridine ring formation according to both scenarios. Importantly, late oxidation of dihydropyridine species would be required to yield the final pyridine/pyridinium skeletons when acrolein 16 is incorporated, contrary to malondialdehyde 17. Also important and central to the modified hypothesis involving 17 is the postulate of the existence of two types of C5 reactive units: glutaconaldehydes and aminopentadienals. In vivo, malondialdehyde (17) results mainly from the catabolism and peroxidation of polyunsaturated fatty acids such as arachidonic or linolenic acid (Scheme 6.3). Different mechanisms have been proposed, some involving radical reactions with reactive oxygen species (ROS) [17]. The existence of an acrolein radical (19) has been postulated, which could react with a hydroxyl radical to give malondialdehyde 17. From this hypothesis, a simple reduction of 19 would explain the formation of acrolein 16.
186
6 Biomimetic Synthesis of Manzamine Alkaloids
O
O
X=H: acrolein 16 O general scheme for the formation of heterocycles
O reactive C3 units X
OH aza-aldol
N
N
N
oxidation
NH2
Baldwin's hypothesis (1992): a dihydropyridine chemistry O N NH2
X
Marazano modified hypothesis (1998): a pyridinium chemistry O
X=OH: malonaldehyde 17 O
−H2O + HX NH2 O
O
or
O
HN
HN
NH2 O type
1
glutaconaldehyde aminopentadienals 2 types of C5 units Scheme 6.2
O
R H
O
OH lipidic peroxidation OH O O
Scheme 6.3
2
Detailed pyridine ring formation according to both scenarios.
O poly-unsaturated fatty acids (e.g. arachidonic acid)
O type
O
? 19
OH
acrolein 16
O O malonaldehyde 17
Plausible origin of C3 -reactive units from lipidic peroxidation.
Alternative biosynthetic pathways have been proposed. To date, they have not been corroborated by biomimetic chemical synthesis but they merit attention. Amade, Thomas, and colleagues founded their hypothesis for the biosynthesis of 3-alkylpyridiniums when they isolated pachychalines A (20) and B (21) from a Caribbean Pachychalina species [18] and pachychaline D (22) from a Callyspongia species [19]. Given the presence of a homospermidine fragment on pachychaline B (21) and D (22), the authors proposed this diamine as a possible C3 unit provider and put together a unified scenario for both C-N and C-C connection patterns necessitating oxidation steps of primary amines into the corresponding imines (Scheme 6.4). Imines/enamines cascades could be responsible for the formation of pyridiniums with a C3 diamine acting as a leaving group. Whatever the biosynthetic scheme devised by chemists, it is striking that it is impossible to avoid some kind of C3 unit.
O
23
Scheme 6.4
R2
R2
NH2
R1
O
N
NH2
oxidations
N
NH2
R1
N
N
N
R2
R2
R1 =
R1 =
N
N
8
N
R1
10
R2
N
R2
N
R1
NH2
N H
12
C-N connection
N
NH2
pachychaline D 22 [Callyspongia sp.]
12 NH2
pachychaline A 20 [Pachychalina sp.]
12 NH2
NH2
N H
N H
NH2
N
H
N
NH2
R1 H
NH2
R1
oxidation
NH2
R2
R1
R2
R1
"C-N connection" -see above-
N
"C-C connection" -see above-
N
H homospermidine
10 N
HN
pachychaline B 21 [Pachychalina sp.]
R2
R2
N
HN
12 NH2
Alternative biosynthetic hypotheses for 3-alkylpyridiniums based on the pachychaline series.
N N H C unit 3
NH2
reductive amination
N H
NH2
8
C-C connection
N
6.2 Two Complementary Hypotheses: An ‘‘Acrolein Scenario’’ and a ‘‘Malondialdehyde Scenario’’ 187
188
6 Biomimetic Synthesis of Manzamine Alkaloids
6.2.2 Biomimetic Synthesis of Dihydropyridines and Dihydropyridinium Salts
The order of events in this postulated biosynthesis of pyridine rings, involving imine/enamine formation, Michael reaction, and aldol/aza-aldol reaction, is unknown. To our knowledge, its closest synthetic equivalent is the Chichibabin synthesis of pyridines, where ammonia or primary amines and aliphatic aldehydes in excess react at elevated temperatures to yield 3,4,5-trisubstituted dihydropyridines/dihydropyridinium salts that spontaneously oxidize to pyridines/ pyridinium salts (Scheme 6.5) [20]. This reaction cannot be exploited when pyridines substituted with different groups, or dihydropyridine intermediates, are desirable. However, Marazano and colleagues developed a versatile strategy related to the Chichibabin synthesis to by-pass these limitations and access Baldwin’s intermediates, based on the coupling of Strecker (intermediate 24), Michael (intermediate 25 formed with 16), and aza-aldol reactions. This ‘‘one-pot’’ procedure capitalizes on the particular reactivity of zinc triflate and directly furnishes 1,3-disubstituted dihydropyridiniums masked under the form of α-aminonitriles 26 (Scheme 6.5) [21]. Those stable equivalents favorably compare to dihydropyridinium 27 obtained from pyridinium salts 28, following the classical reduction-modified Polonovski reaction sequence developed by Husson and colleagues (via tetrahydropyridine 29 and N-oxide 30) [22]. Dihydropyridinium 27 can in turn be formed from 25 upon Historical Chichibabin synthesis of pyridines (1906): R O
R
R
R
spontaneous oxidation
R
O N
NH2 R
N
R
Ph
Ph
O Marazano biomimetic synthesis of dihydropyridine: R
NH2 O
Zn(OTf)2
R
N
16
Ph CN 24
Ph
R
O
O
H N
KCN
R
Ph
Ph
N
R
Ph CN 26
CN 25
O
AgBF4
NH2 NaBH4
O
biosynthetic intermediates
KCN
N Ph
R 27
Husson's strategy (modified Polonovski reaction): TFAA NaBH4, EtOH
N Ph
R 28
m CPBA
N Ph
R 29
O
N
Ph
R 30
TFAA: trifluoroacetic anhydride m CPBA: meta-chloroperbenzoic acid
Scheme 6.5
Syntheses of biomimetic equivalents of dihydropyridines.
R
6.2 Two Complementary Hypotheses: An ‘‘Acrolein Scenario’’ and a ‘‘Malondialdehyde Scenario’’
treatment with a silver salt and be reduced to 29 with sodium borohydride, usually with high selectivity, or trapped with cyanide ions to yield 26. Conditions for deprotonating stable salts such as 31, prepared in situ by heating masked dihydropyridinium 32, to give the corresponding dihydropyridine 33 (that was unstable but could be trapped), were disclosed by the Marazano group (Scheme 6.6) [23]. The possibility of favoring one isomer over the other (dihydropyridinium 34) soon appeared to be a challenging problem in gaining chemoselectivity (see the following sections). H OMe X
HX
N Ph
N
Me 34
Ph
Me 32
dihydropyridinium salts
Scheme 6.6
MeO N Ph
N
Me 31
Me
33 Ph dihydropyridine
Isomerization of dihydropyridinium salts to dihydropyridine.
6.2.3 A Tool Box of Biomimetic C5 Reactive Units from the ‘‘Old’’ Zincke Reaction
Marazano et al. revisited the century-old Zincke reaction, a nucleophilic ring opening of electron-deficient pyridinium (‘‘Zincke salts,’’ easily prepared from pyridines and electrophiles such as cyanogen bromide or 2,4-dinitrochlorobenzene) [24]. Scheme 6.7 presents the mechanism generally admitted for the Zincke reaction [25]. Ring opening of pyridinium 35 occurs with the first equivalent of amine (usually accompanied by a dark red coloration of the reaction mixture). A second equivalent of amine then reacts with 36 to provide 37 with extrusion of a 2,4-dinitroaniline moiety. In solution, 37 is in equilibrium with two aminopentadienimines (38 and 39). Upon heating, intermediate 37 may cyclize into pyridinium salts 40 with elimination of one amine moiety [26]. Aminopentadienimines of type 38 and 39 are of particular interest for accessing biomimetic equivalents of postulated intermediates, namely aminopentadienals and glutacondialdehydes (Scheme 6.2) [27]. Convenient and scalable accesses to substituted glutacondialdehyde salts from the corresponding pyridinium Zincke salts were recently disclosed4) [28]. Specifically, in these cases, a secondary amine such as dimethylamine is employed for the ring-opening to afford firstly salts 42, then biomimetic equivalents 43 of biosynthetic aminopentadienals. These latter can be hydrolyzed into glutacondialdehyde salts 44 when treated with potassium hydroxide (Scheme 6.8). 4) Overcoming thereby some drawbacks of
the ‘‘classical’’ Zincke reaction, such as the use of two equivalents of amine and the
propensity of aminopentadienals to form pyridinium salts.
189
190
6 Biomimetic Synthesis of Manzamine Alkaloids
DNP
DNP N
Cl
N
H2N R
35
R1
H N
R
R1
R2
R2
DNP= 2,4-dinitrophenyl
R N
(or other deactivating group)
H2N
pyridinium salts 40
R2 HCl
HCl R N
R
39
Scheme 6.7
36
R2 HCl
DNP NH2 R
R N
R1
R
aminopentadieniminiums
N H
R1
R2 HCl 37
R
Mechanism of the Zincke reaction.
DNP N
HN
Me Me
"Zincke salts"
Me
EtOH, reflux
R 41 K O glutaconaldehyde salts
Me N Me
N Me 42 R
O R 44
Me
KOH THF/MeOH (65-93%)
NaOH
51-91% from 41
N Me
O 43 R
"Zincke aldehydes"
O NH2
O
O HN O NH2 O glutaconaldehyde aminopentadienal postulated biosynthetic intermediates
O
Scheme 6.8 Glutacondialdehydes and aminopentadienals as biosynthetic intermediates and biomimetic equivalents.
t-BuNH2
R
O
R
quant.
N 47
i-LDA
t-Bu ii- Et2N
O
2
KO 44
R
K
O
O
Et
46
iii- HCl biosynthetic hypothesis
O O
R
R1
N 38
R1
H N
R2
R
R
R1
R2 N
H N
H 2N R
H N
R1
R2 N H
N
R
R1
R2 R
DNP
DNP NH N
NEt2
N Et
N 48
R HCl KOH
O
50-80%
R OBn 45
LDA: lithium diisopropylamide
Scheme 6.9 Alternative synthetic pathway towards substituted glutacondialdehyde salts, using vinamidinium salts as biomimetic equivalents of malonodialdehyde.
t-Bu >90%
6.3 Biomimetic Synthesis of Pyridinium Marine Sponge Alkaloids
More elaborated glutaconaldehyde salts of type 44 (as exemplified by compound 45, Scheme 6.9) can also be prepared starting from various aldehydes and vinamidinium salts 46 (via 47 and 48) [29]. Interestingly, this strategy is reminiscent of the first fundamental step in Marazano’s hypothesis of pyridine ring formation, that is, the reaction of a fatty aldehyde with malondialdehyde 17 (see Scheme 6.2 for details).
6.3 Biomimetic Synthesis of Pyridinium Marine Sponge Alkaloids 6.3.1 Biomimetic Total Synthesis of Cyclostellettamine B and Related 3-Alkylpyridiniums
To test their ‘‘malondialdehyde’’ scenario and to demonstrate the suitability of Zincke chemistry toward this end, the access to cyclostellettamine B (49) [3] was studied by Marazano and colleagues [16a]. They performed a pseudo-dimerization of two 3-aminoalkylpyridines (50, 51) of different chain lengths,5) using sequential pyridinium N-activation (via 52 and 53). Cyclostellettamine B (49) was thus elegantly obtained in a biomimetic ‘‘domino-Zincke’’ reaction (Scheme 6.10). A similar philosophy permitted the total synthesis of haliclamine A (54) [30, 31] and niphatoxin B (4) [32], and also that of two original molecules isolated and synthesized by the K¨ock group, that is, viscosamine (55) (a trimeric 3-alkylpyridinium) [33] and a monomeric but cyclic 3-alkylpyridinium alkaloid (56) (Scheme 6.10) [34]. 6.3.2 Biomimetic Synthesis of Xestospongins and Related Structures
Xestospongins are macrocyclic bis-1-oxaquinolizidine alkaloids isolated from Xestospongia exigua (syn. Neopetrosia exigua) [35]. Many other structures are closely related to xestospongins such as araguspongins, and the interested reader is referred to general review articles [14]. We focus herein on the biomimetic synthesis of xestospongins A (57) (Scheme 6.11) and C (58) [35] as well as (+)-araguspongin B (59) (Scheme 6.12) [36] by the Baldwin group in 1998, which also permitted the establishment of their correct absolute configurations [37]. Biosynthetically, starting from bis-hydroxypyridinium dimer 60 or the corresponding dihydropyridinium salt 61, intramolecular trapping of the iminiums would explain the formation of the oxaquinolizidine ring systems and the natural substances after a reduction step on 62 (Scheme 6.11). Conformational and configurational differences and/or equilibria between natural substances in this series can be seen to occur via iminium/enamine epimerizing equilibrium, involving ring opening/reclosure to cyclic aminals. The biomimetic 5) The incorporation of chains of various
lengths is of course a critical point in the total synthesis of such molecules.
191
6 Biomimetic Synthesis of Manzamine Alkaloids
192
NHBoc 50 N
NHBoc n -BuOH
Cl
N
reflux
DNP
51
N
Cl
N
H2 N
52
Cl-DNP then HCl
13 carbons
N
N
NH2
n -BuOH
N
reflux
12 carbons
53 Cl N
Cl
DNP
cyclostellettamine B 49 [Stelletta maxima]
N
5
haliclamine A 54 [Haliclona sp.]
N
Cl
niphatoxin B 4
N
N 6
N N
9
N
viscosamine [Haliclona 55 9 viscosa] 9
4
cyclic monomeric alkaloid 56 [Haliclona viscosa]
N
N
Scheme 6.10 Biomimetic synthesis of cyclostellettamine B and the structure of alkaloids synthesized using a similar philosophy.
60
OH N
HO
H O
reduction oxidation
5
H
O
5
N
N
61
62
see scheme 6.1 O O or
O 17
16
N
O N
reduction
O
*absolute configuration as corrected by Baldwin and coll.
Scheme 6.11
N
O
N
O N (+)-xestospongin A 57* [Xestospongia exigua]
Biosynthetic proposal for xestospongin A and related structures.
synthesis of ent-xestospongins A (ent-57) and C (ent-58) and ent-araguspongin B (ent-59) from ent-60 and ent-62 as depicted in Scheme 6.12 probably proceeds via this pathway, and implicates intermediate ent-61. Two distinct reaction conditions permitting a reduction of the unsaturated piperidine without reduction of the masked iminium were studied (i.e., hydrogenation with catalytic rhodium or Raney nickel), and gave different ratios of the three natural substances ent-57 and ent-58
6.3 Biomimetic Synthesis of Pyridinium Marine Sponge Alkaloids
193 N
BN O
O
N B
10 steps
N
O
O
OH 5
A N ent- 62 conditions O
A
N
ent- 60
fast
DEAD
OH
OEt
N
N
N O
N H
O N O
O
(+)-xestospongin C 58 (−)-xestospongin A 57 (unnatural isomer) (unnatural isomer)
9.5% 77%
17% 7%
23% -
1 Rh on alumina, MeOH, H2; 2 Raney Ni, MeOH, H2
Biomimetic synthesis by the Baldwin group. H
O O
H
H
N Me
N
O upenamide 13 [Echinochalina sp.]
O
H
63
H
O
N Me
CHO
H
64
N I
HO
Figure 6.4
H
HO
N
N
DEAD: diethyl azodicarboxylate
Scheme 6.12
H O
slow
+
+
N
1 2
fast
O
O
(+)-araguspongine B 59 (unnatural isomer)
O
Selected synthetic approaches to upenamide.
and ent-59. Clear establishment of the absolute configurations of (+)-xestospongin A (57) and (−)-xestospongin C (58) and questions concerning those of araguspongin (59) alkaloids were also discussed by Baldwin and colleagues [37]. Similar stereoelectronic outcomes, which we will not discuss here, were studied during the synthesis of the octahydropyrano-pyridine ring system of upenamide (13) by the Marazano [38] and Sulikowski groups [39]. The fragments (63 and 64) prepared by each group are presented in Figure 6.4. To date, this fascinating alkaloid has resisted total synthesis. 6.3.3 Is the Zincke-Type Pyridine Ring-Opening Biomimetic?
To the best of our knowledge, Zincke-type pyridine or pyridinium ring formations are poorly documented in biochemistry. As an example (Scheme 6.13), the biosynthesis of quinolinic acid (65)–a direct precursor of nicotinamide adenine dinucleotide (NAD)–from tryptophan-derived 66 was proposed to take place via acyclic 67 followed by a 6π-electrocyclization (demonstrated with model systems)
6 Biomimetic Synthesis of Manzamine Alkaloids
194
oxidation,
CO2H
CO2H isomerization
L-tryptophan
NH2 OH 66 Scheme 6.13
O H2 N
CO2H HO
CO2H 67
CO2H
N H
[6p]
CO2H N CO2H 65
Biosynthesis of quinolinic acid.
(Scheme 6.13) [40]. If discovered in secondary metabolism, such a transformation would putatively connect theonelladine-type and cyclostellettamine-type alkaloids via a biosynthetic Zincke-type reaction (cf. Scheme 6.1). 6.3.4 Alkylpyridines with Unusual Linking Patterns 6.3.4.1 Biomimetic Synthesis of Pyrinodemin A With a cis-cyclopent[c]isoxazolidine ring system linking two alkylpyridine chains, pyrinodemin A (68) (Scheme 6.14), the first representative of a small group of four alkaloids [41], has been the subject of several publications [42]. In fact, only the total synthesis in combination with degradation experiments of 68 permitted establishment of the correct structure of this intriguing natural product as far as the position of the side-chain double bond is concerned. In 2005 [43], the Kobayashi group put forward clear conclusions establishing the position of the double bond and the racemic character of the central core–despite a (−) reported optical rotation in the original paper [41a]. We will, in this section, primarily deal with the biomimetic access to the central bicyclic system of 68, which was logically proposed to biosynthetically arise from a [3 + 2] cycloaddition between a nitrone and a (Z)-alkene, which in turn arise from two precursors (aldehyde 69 and amine 70 sharing the same number of carbons and a similarly positioned cis-double bond). The key cycloaddition was exploited in most total syntheses of 68 and resulted in a stereocontrolled formation of the bicyclic system in good yields.
H 15
3
14 H
N
H N
16
O
N
pyrinodemin A 68 [Amphimedon sp.]
3'
16' 15'
conditions for the biomimetic cycloaddition step (various substrates): Ph or PhCH3, reflux
N O
oxidation
15 16
O N
69
70
H2N 16'
Scheme 6.14
15'
Biosynthetic considerations for pyrinodemin A.
N
6.4 Development of Baldwin’s Hypothesis: From Cyclostellettamines to Keramaphidin-Type Alkaloids
195
6.3.4.2 Biomimetic Synthesis of Pyrinadine A Another intriguing linking pattern is the one observed in pyrinadine A (71) isolated from a Cribochalina sp. by Kobayashi and colleagues (Scheme 6.15) [44]. It consists of an uncommon in Nature diazoxy group, presumably resulting from the oxidative dimerization of a hydroxylamine such as 72 (obtained from the oxidation of the corresponding amine 73). N O
N
reduction (Zn, AcOH) [44(a)]
N
N
biomimetic conditions : CH2Cl2, air (76%) [45]
pyrinadine A 71 oxidative [Cribochalina sp.] dimerization NH2
73 N
O N HO
H N
NHOH OH N
N O
N OH
dehydro-72
N
72 N
OH
N Scheme 6.15
Pyrinadine A: plausible biosynthetic origin and biomimetic access.
In 2009, Lee and colleagues [45] successfully mimicked the process in the laboratory with a clean and spontaneous conversion of synthetic precursor 72 into 71 under simple aerial conditions according to the mechanism proposed in the box in Scheme 6.15. Keeping in mind that plausible precursor dehydro-72 is a known natural product [41b], the exact role of enzymes is clearly questioned in such cases and an artifactual origin cannot be ruled out.6) 6.4 Development of Baldwin’s Hypothesis: From Cyclostellettamines to Keramaphidin-Type Alkaloids 6.4.1 Linking Pyridinium Alkaloids and Manzamine A-Type Alkaloids
Competitive to the redox processes that may take place between cyclostellettamine-related alkaloids 74 and postulated dihydropyridinium salts counterparts such as 75, intramolecular Diels–Alder cycloaddition of bis-dihydropyridinium 75 might occur as represented in Scheme 6.16, yielding a bridged pentacyclic 6) The existence of unsymmetrical mole-
cules such as pyrinadines C–G, see
Reference [44], adds further credence to this statement.
196
6 Biomimetic Synthesis of Manzamine Alkaloids
+N
X
redox interconversion
−
pyridinium alkaloids 74 X
+N
X
−
N
+
−
keramaphidin-type alkaloids N
reduction
75 X
− +
N
N
N
N
76 keramaphidin B 7
N +
dismutation
N
O
H2O
N 77 X
manzamine A 1 N H
N H
N
N
+
−
78
H
Pictet-Spengler and oxidations
N H
CHO
H OH
N
CHO
oxidations
N
OH
N
NH2 N H
N
HN
H N H
80 ircina l A 79 [Ircinia sp.]
manzamine A type alkaloids
Scheme 6.16 Baldwin’s hypothesis: the missing link between pyridinium alkaloids and manzamine A.
intermediate (76). Dismutation7) reaction would give rise to a new iminium (77), which upon hydrolysis would provide 78 as a direct precursor of ircinal A (79), a natural substance isolated for the first time in 1992 from Ircinia sp. (notably, after Baldwin’s proposals) [46]. Formation of manzamine A (1) through a Pictet–Spengler reaction with tryptamine (80) followed by oxidation into the final β-carboline would then be easily explained. The pertinence of Baldwin’s proposal comes from the fact that key pentacyclic intermediate 76 was postulated in 1992 for the biogenesis of manzamine A (1), before its natural occurrence became apparent some time later with the isolation of keramaphidin B (7) (1994) and related analogs. In fact, simple iminium reduction of 76 can explain the biosynthesis of keramaphidin B-type alkaloids, placing thereby 7) This term will be used cautiously in the
present chapter, regarding the absence of
knowledge of the precise redox mechanisms involved biosynthetically.
6.4 Development of Baldwin’s Hypothesis: From Cyclostellettamines to Keramaphidin-Type Alkaloids
76 as a cornerstone in the general biosynthetic mapping of manzamine alkaloids. This example is probably one of the most brilliant demonstrations of the power of ‘‘retrobiosynthesis’’ as it beautifully paved the way to rich biomimetic endeavors. 6.4.2 Biomimetic Total Synthesis of Keramaphidin B 6.4.2.1 Model Studies (1994) To validate their hypothesis, Baldwin and colleagues successfully carried out model reactions that permitted the synthesis of the central core of keramaphidin B (7) by intermolecular Diels–Alder reaction between two molecules of dihydropyridinium salts 81, prepared from picoline via N-oxide 82 (Scheme 6.17) [47]. Incubation of 81 in an aqueous buffer at pH 8.3 for 18 h followed by treatment with sodium borohydride yielded mainly unsaturated piperidine 83 but also 84 (10% yield based on N-oxide 82), the awaited polycyclic core reminiscent of that of keramaphidin B (7), probably formed via iminium 85.8)
N +
N
O−
Me
(CF3CO)2O
Me
Me
N
+
Me
82
3 steps
N
81
+
Me
X− N+
N Me NaBH4
Me
N
Me
N
NaBH4
Me
83
+ Me
major compound Scheme 6.17
N 85
N
N Me 10% 84, keramaphidin central core
Baldwin’s hypothesis: model studies toward keramaphidin B.
6.4.2.2 Total Synthesis of Keramaphidin B (1998) Four years later, the Baldwin group made a significant contribution to the art of total (biomimetic) synthesis (Scheme 6.18). They first obtained cyclostellettamine-type 74 by dimerization of 3-tosyloxyalkenyl pyridine 86, which was converted into postulated biosynthetic intermediate 75, via intermediate 87, using the classical reduction/modified Polonovski reaction sequence (vide supra Schemes 6.5 and 6.17). Following equilibrium of bis-dihydropyridinium 75 in aqueous medium 8) The publication ended with:
‘‘investigations into an intramolecular variant of the cycloaddition [ . . . ] are
in progress with the aim of accomplishing the total biomimetic syntheses of manzamines and ingenamine via keramaphidin B.’’
197
6 Biomimetic Synthesis of Manzamine Alkaloids
198
+
O
N
3 steps
N
N
X
−
X
TsO
−
N
+
74
86
NaBH4
N+ X
N +
N X
H
−
−
N 1 - mCPBA 2 - (CF3CO)2O
+
X
−
+
N
98%
75 MeOH, H2O
N 87 56% from 86
then NaBH4 60-85%
NaBH4 +
N X
N −
N
(0.2-0.3%)
biosynthetic intermediate 76
Scheme 6.18
N
keramaphidin B 7
Validation of Baldwin’s hypothesis: total synthesis of keramaphidin B.
and reduction, the authors were able to isolate keramaphidin B (7) in 0.2–0.3% overall yield [48]. As already pointed out in the model studies (Scheme 6.17), the major compound resulting from this last reaction was recyclable bis-tetrahydropyridine 87. 6.4.3 Drawbacks of the ‘‘Acrolein’’ Scenario
While this total synthesis achievement clearly demonstrated the validity of the model, the extremely low yield of keramaphidin B (7) became the subject of puzzling investigations. This result was explained by important side-reactions (mainly dismutation of dihydropyridines via intermolecular hydride transfer), and by the high energy barrier of the macrocyclic cycloaddition. The experimental drawbacks observed with this pioneering model are detailed below and were part of the reason why a modified scenario was concomitantly proposed by the Marazano group. 6.4.3.1 Very Low Yield of the Endo-Intramolecular Diels–Alder Reaction This is obviously the major drawback of Baldwin’s total synthesis of keramaphidin B (Scheme 6.19). This key step could benefit in vivo from the intervention of a
6.4 Development of Baldwin’s Hypothesis: From Cyclostellettamines to Keramaphidin-Type Alkaloids
N endo intramolecular Diels-Alder
+
N X
−
Scheme 6.19
+
low yield
N
76 X
-transition state energetic reasons? -in vivo Diels-Alderase?
75
N −
Low yield of the Diels–Alder reaction.
‘‘Diels–Alderase’’ that could limit the conformational mobility of the transition state, thereby minimizing the entropic factor. However, molecular modeling studies conducted on intermediate 75 revealed the existence of conformations close to the required transition state [48b]. The kinetic preference of 75 to disproportionate was thereby put forward as the main reason for the low yield of the synthesis. 6.4.3.2 Undesirable Transannular Hydride Transfers One of the reasons why the above Diels–Alder reaction was not as efficient as expected is probably the existence of a favorable transannular dismutation, leading after reduction to 3 -piperidines. This disproportionation phenomenon, which was observed on model systems with dihydropyridinium salt 81 [47], occurred to a greater extent with 75 in the natural product synthesis (Scheme 6.20). Several experimental evidences suggested that the isolation of reduced 87 after treatment of the reaction mixture with sodium borohydride resulted from the reduction of 88 (arising from spontaneous dismutation), and not from the reduction of biosynthetic intermediate 75. The intervention of a putative ‘‘Diels–Alderase’’ could also exclude or limit this disproportionation reaction.
N+ X
−
spontaneous dismutation
N+ X
−
disproportionation
X
−
+
N
N 88
75
NaBH4
N
NaBH4
N 87 Scheme 6.20
The disproportionation issue.
199
200
6 Biomimetic Synthesis of Manzamine Alkaloids
6.4.3.3 Conversion of a ‘‘Keramaphidin’’ Skeleton into an ‘‘Ircinal/Manzamine’’ Skeleton Was Not Experimentally Possible According to Baldwin’s biogenetic hypothesis, alkaloids of the keramaphidin B type are the immediate precursors of ircinal type and manzamine A type alkaloids, following regioselective oxidation and iminium hydrolysis (cf. Scheme 6.16). To test this hypothesis, the Marazano group synthesized aminonitrile 89, which was submitted to decyanation–hydrolysis using tetrafluoroboric acid in aqueous medium. Although the N2 -iminium 90 was clearly observed by NMR, aldehyde 91 could never be obtained even under various forcing hydrolytic conditions (Scheme 6.21) [49]. This result disfavors the perception that manzamine aldehydes originate from the hydrolysis of keramaphidin-type iminium as proposed by J.E. Baldwin [15]. However, one should keep in mind that this hydrolysis equilibrium could be driven biosynthetically by a connected equilibrium (e.g., proton shift to αβ-unsaturated aldehyde), or any irreversible transformation (e.g., Pictet–Spengler reaction). Me
Me
Me
H Me
Me CN
O
N
HBF4
N
N
Me
H2O
Me 89 Me keramaphidin B skeleton
H2O
N H
N Me 2 BF4 90
N H
Me Me
Me
N H
CHO
Me Me
Me
HN
Me
91
ircinal/manzamine skeleton
Scheme 6.21 Failure to convert a keramaphidin skeleton into a ircinal/manzamine A skeleton on a model system.
6.5 ‘‘Malondialdehyde Scenario:’’ A Modified Hypothesis Placing Aminopentadienals as Possible Precursors of Manzamine Alkaloids 6.5.1 Keramaphidin/Ircinal Connection
The modified hypothesis, based on the intervention of malondialdehyde C3 -reactive units to explain the formation of pyridinium salts, sets the stage for a universal model of biosynthesis for the manzamine alkaloids. Armed with a long experience in the pyridinium chemistry, the Marazano group turned its attention to the manzamine alkaloids in the mid-1990s. In their 1998 and 1999 papers [16], for the first time the pyridinium/aminopentadienal chemistry was put forward to explain the divergent formation of ircinal/manzamine and halicyclamine alkaloids. Whereas polycyclic intermediate 76 (resulting from the intramolecular Diels– Alder reaction between dihydropyridinium salts, cf. Scheme 6.16) was central in Baldwin’s hypothesis, intermediate 92 featuring (i) a dihydropyridinium moiety
6.5 Aminopentadienals as Possible Precursors of Manzamine Alkaloids
O
201
aminopentadienal
O
O NH3 OO
HN
1
N
X
see scheme 6.2
NH
OO O 16 O
NH3
93
O 17
reduction
X
74
N
O aminopentadienal
pyridinium salts alkaloids CHO
NH3 O O
N
see scheme 6.2
2
OO
N
X NH
NH3
O
17 O
N
N
O
type
1
(see scheme 6.2)
reduction H
−H2O
CHO
O
reduction
keramaphidin B 7
HN
92
N
N H
N 78
HN
manzamine A 1 Scheme 6.22 Biosynthetic scheme towards manzamine A and keramaphidin B according to Marazano’s hypothesis.
(resulting from either the ‘‘acrolein’’ or ‘‘malondialdehyde’’ pathways 1 and , 2 respectively, on Scheme 6.22) and (ii) an open-chain aminopentadienal, is obviously the key element of the modified hypothesis (Scheme 6.22). It should be mentioned here that intermediate 78 closely related to ircinal and resulting formally from an intramolecular Diels–Alder reaction of 92 is therefore a precursor of the keramaphidin (7) skeleton according to this model. Additionally, an intermediate such as 93 could also explain the biosynthesis of cyclostellettamine-related pyridinium 74. 6.5.2 Halicyclamine Connection
The aim of this model was also to explain the formation of halicyclamine-type alkaloids (Scheme 6.23) from precursor 94, that is, similar to precursor 92 with the difference that it integrates a type-2 aminopentadienal moiety (according to Scheme 6.2). Halicyclamine-type alkaloids would this time be generated by an intramolecular 1,4-addition of the aminopentadienal moiety reacting not as a diene
202
6 Biomimetic Synthesis of Manzamine Alkaloids
O
N X
H N
94
−H2O reduction
H
N
O N
95 found to be a natural product in 2004! [Amphimedon sp.]
X
H
H N
halicyclamine A 8
N H X
N
HN
reductions
H
Scheme 6.23 Biosynthetic scheme towards halicyclamine A according to the modified scenario.
O 16 NH2
N
15
O
N type
O
HN
1
O
NH2
O
aminopentadienal of type 2
O H
O NH
type 2 O
NH
17 O
aminopentadienal of type 1
O
formal (4+ 2) N
N
N HN
HN
O H O
N H N
N H
halicyclamine alkaloids
N
NH
HN
ircinal alkaloids
Scheme 6.24 The aminopentadienal connection between representative manzamine alkaloids.
keramaphidine alkaloids
6.6 Testing the Modified Hypothesis in the Laboratory
203
(Scheme 6.22) but as an enamine. Illustrated in Scheme 6.23 with the biosynthetic proposal for halicyclamine A (8) the scenario was consolidated some years later with the isolation of postulated pyridinium 95 as a natural substance from Amphimedon sp. [50].9) When considering only the biosynthesis of the central cores of the different sub-classes of alkaloids evoked up to now, the homogeneity of the model is even more striking, as represented in Scheme 6.24. Starting from intermediates of type 15, access to the halicyclamine, keramaphidin, and ircinal series is explained via the reactivity of the aminopentadienal system (acting as either diene or enamine).
6.6 Testing the Modified Hypothesis in the Laboratory 6.6.1 Biomimetic Models toward Manzamine A
According to Marazano’s hypothesis, which involves malondialdehyde as the crucial C3 building-block, aminopentadienal-dihydropyridinium species like 92 would undergo intramolecular (4 + 2) cycloaddition to yield an iminium that already possesses the main structural features of ircinal alkaloids (Scheme 6.22). To probe this hypothesis, Marazano and colleagues reacted dihydropyridinium 34 and aminopentadienoate 96, which unexpectedly furnished amine 97 (Scheme 6.25). This behavior was explained by intramolecular hydride transfer to the formed iminium 98, followed by loss of butanal upon hydrolysis of 99 [16b]. O
CO2Et
OEt 96
Me NHn-Bu Me
N Ph
CH2Cl2 0 – 4°C
N Ph H 98
Me
Me NH
34
Scheme 6.25
CO2Et
CO2Et H2O
N
Ph intramolecular hydride shift
Me Me N
99
~50%
N Ph
Me Me 97
A (4 + 2) cycloaddition strategy towards an ircinal model.
While many strategies relied on the use of the Diels–Alder reaction in the numerous synthetic approaches to manzamine A [51], this one seems to be among the most efficient since all crucial functionalities are brought together in a single step. Indeed, the construction of the ABC-ring system of manzamine A was published soon after by the same team (Scheme 6.26) [52]. The choice of a correct substitution pattern on the starting dihydropyridinium salt 100 permitted 9) The
Marazano group extended the Baldwin hypotheses (based on the dihydropyridinium chemistry) to explain the biosynthesis of halicyclamine A (and
also sarain A) in a 1995 publication [53] before developing their own modified model.
NH2
204
6 Biomimetic Synthesis of Manzamine Alkaloids
the construction of the C-ring with a (4 + 2) cycloaddition from 101, furnishing ircinal analog 102 in good overall yield. O Ph
OEt
H
CO2Et
H
(4+2)
N
N
Me TBDMSO
HN 96
100
Me
Ph
Cbz
N
A
A
B
OH
C
B Me
C
N
N
101 OH
Scheme 6.26
N
CO2Et
H
102 8 steps, 17% from diene 96
manzamine A 1
Biomimetic synthesis of the ABC-ring system of manzamine A.
No further developments concerning the biomimetic approach to ircinal/manzamine A series were published in the following years by the group of Marazano, until 2008 and the publication of a general approach validated by unprecedented results [21]. The authors combined Zincke-type chemistry, involving biomimetic species such as aminopentadienals and glutacondialdehydes, and their ‘‘Chichibabin-like’’ synthesis of dihydropyridinium depicted in Scheme 6.5. Thus, Strecker–Michael adduct 103 was reacted with aminopentadienal 104 in Lewis acid conditions, yielding in moderate yield bicyclic iminopentadienal 105 (Scheme 6.27). This stable compound, when treated with acetic anhydride followed by reduction and final hydrolysis, furnished dienal 106 as a biomimetic model of ircinal alkaloids, following an impressive cascade sequence of rearrangements of the aminopentadienal system (Scheme 6.27). O O
Ph
Me
Ph
Bn N
Ac N H
Ph
Me OAc
Me
N Ph
Me OAc
NaBH(OAc)3 Ac Ph N H
N
Me
Me Ph
N Ph
Me OAc
N
Me
(45% from 105)
Scheme 6.27
Ac
N
Ph
A
A
Me AcO Me
OAc H
H
105
O
N
Me Ph
CHO
HCl
Me
Ph N
N
Me
Ac2O
Bn N
Ac
O
N N (22%)
Me
Ph
Ph
Ph
NHBn
N
104 HN
CN 103
Ph
Me
ZnBr2
+ Me
N
H
O
CHO B
OH
N
ircinal A 79
B Me 106 (55%)
Me
H
Aminopentadienal scenario towards the AB-ring system of manzamine A.
6.6 Testing the Modified Hypothesis in the Laboratory
205
6.6.2 Biomimetic Models toward Halicyclamines
The first clear link between keramaphidin B (7) and halicyclamine alkaloids was experimentally demonstrated in 1995 by the Marazano team [53]. When performing a similar reaction to the one described in 1994 by Baldwin and colleagues [47] (cf. Scheme 6.17), consisting of studying the behavior of dihydropyridinium salt 108 in solution (generated in situ from 107), a significant amount of 109 was isolated along with awaited 110 and 111 (Scheme 6.28). Me
n -hex
OMe H
n -hex N
Me n -hex
N
Me
+
n -hex N
N n-hex Me keramaphidin skeleton (25%)
dismutated piperideine (40%) 110 +
Me 108
107
N
Me
n-hex N
Me
n-hex
H
N
H
HN
H N
n-hex
halicyclamine A 8
N
Scheme 6.28
n-hex
H
111
H Me H Me halicyclamine skeleton (7%) 109 H
HN
First biomimetic synthesis of halicyclamine-related structures.
Similarly, keramaphidin model 112, when submitted to regioselective photooxidative α-cyanation, was turned into halicyclamine model 113 via retroaza-Mannich fragmentation of iminium 114 and cyanide retrapping (Scheme 6.29) [49]. This result suggests that a regioselective N1 -oxidation of keramaphidin B type alkaloids might be implicated in the biogenesis of the halicyclamines. Further, Marazano and colleagues observed that treating polycyclic aminal adduct 115 resulting from nucleophilic reaction of aminopentadienal 116 with dihydropyridinium salt 34 [16b] in acidic conditions gave pyridinium 117, which was directly reduced with sodium borohydride to give a 3 : 1 ratio of compounds
n-hex
N
N Me
Me
Me
Me
n-hex
n-hex N
N
112
N
Me
N
n-hex
Me
Me CN
selective oxidation H
n -hex
N
H N H
CN
Scheme 6.29
n-hex
Me 114
conditions: hn (l> −495 nm), TPP, TMSCN, O2
TPP: tetraphenylporphine TMSCN: trimethylsilylcyanide
n-hex
Me
CN H
n-hex
113 (15%)
Selective oxidation of a keramaphidin model.
n-hex
N
H Me CN
H
N
n-hex
6 Biomimetic Synthesis of Manzamine Alkaloids
206
118 and 119 (Scheme 6.30a). The main compound (118) was found to possess the halicyclamine A (8) central core whereas 119, the minor adduct, can be considered as a halicyclamine B (120) model [54] (despite the opposite stereochemistry of one stereocenter on the tetrahydropyridine ring). In its recent development of the halicyclamine chemistry [21], the Marazano group observed that reaction of Strecker adduct 103 and aminopentadienal 116 in the presence of zinc triflate directly furnished pyridinium 121 (Scheme 6.30b). Pyridinium 121 was further reduced to the above-described mixture of bis-piperidines 118 and 119 in good overall yield. Taken together, these results suggest that dehydrohalicyclamines such as 95 (i.e., pyridinium species) could be the actual biogenetic precursors of halicyclamine-type alkaloids after reduction, and not the opposite. see scheme 6.27 possible evolution towards ircinal /manzamine A series Me O CH2Cl2 H N Me N n-Bu N N Ph Ph 115 117 Me Me (55%) Me
H
O
116
Me
HN Me
N Ph 34
Me
(a)
Me
Me
N
H
H Ph
+ Me Ph CN 103
Me
(b)
Scheme 6.30
n-Bu Ph
H
+
N
N
NaBH4
Me
H
N
n-Bu N
H
halicyclamine B 120 [Xestospongia sp.]
Me
Zn(OTf)2
N
N
116 NH n-Bu
H Me
H
O
HN
H Me 3:1 ratio 118 119 (23% from salt 117) (40% from pyridinium 121)
N
halicyclamine A 8
O
H
H
HN
O
N H
N
dehydrohalicyclamine A 95
NaBH4
H
H
NHn-Bu Me
N Ph
CN
N Ph
n -Bu
Me 121 (25%) CN
First (a) and second (b) generation approaches to halicyclamines.
The latest developments to tackle the total synthesis of halicyclamine A (8) include the synthesis of a macrocyclic target molecule, demonstrating thereby the feasibility of intramolecular reaction (Scheme 6.31) [55]. Precursor 123 was prepared in eleven steps from tetradecandioic acid (122). Intermediate 124 was effectively reached with, now classical in this series, zinc triflate. Compound 124 then collapsed in acidic conditions to provide α-aminonitrile/pyridinium 125, which was finally converted into halicyclamine A model 126 during a reduction step that appeared to be both regio- and diastereoselective (to be compared to the obtaining of a mixture of regioisomers in the case of intermolecular reaction,
6.6 Testing the Modified Hypothesis in the Laboratory HO2C
10 CO2H
122
Me
11steps
Me O
O
H N Ph
207
N
CN Zn(OTf)2
CN 123
O
H N
N Bn N
N Ph
(33%)
124
Me H N
H
HN H
Ph
N
HN
KCN, AcOH
Me AcO
NaBH4, MeOH/H2O
H
N Ph
CN 125
126 (27%, 2 steps) halicyclamine A 8
Scheme 6.31
Me
Latest developments toward a biomimetic synthesis of halicyclamine A.
see Scheme 6.30). In view of this achievement, a biomimetic total synthesis of halicyclamine A (8) is reasonably within reach. An alternative pathway, involving a late introduction of the nitrogen atoms, was communicated in 2003 (Scheme 6.32) [56]. Compound 127 was prepared as a stabilized equivalent of biomimetic intermediate 128, itself reminiscent of postulated biosynthetic intermediate 129. The strategy towards 127 firmly exploited the biosynthetic proposals as it consisted of sequential condensation of aldehyde and malonaldehyde equivalents. Its reactivity towards primary amines was then studied and led to the formation–via (among others) pyridinium 130–of four diastereomers, including compound 131, which display the same relative stereochemistry as halicyclamine A. Notably, even if conceivable in principle, no manzamine A type compounds were formed during these investigations, probably because of the irreversibility of pyridinium formation. Overall, these short and convergent biomimetic syntheses, which rely on identical reactants brought to distinct reaction fates, are based on the fusion of Baldwin’s seminal hypothesis (dihydropyridinium-based) and Marazano’s modified theory (aminopentadienal-based) (Scheme 6.33). In one of the most exciting recent achievements in natural product biomimetic synthesis, application of the Baldwin–Marazano concepts delivered core analogs of ircinal/manzamine A and the halicyclamines, constituting a strong presumption of how those alkaloids relate biosynthetically. Biomimetically speaking, the most impressive fact is that the entire sequences of reactions are promoted in cascades depending on the Lewis acid; it is therefore difficult to imagine more straightforward ways to access these complex families of molecules.
N
208
6 Biomimetic Synthesis of Manzamine Alkaloids conception of a biomimetic equivalent: Me
Me
N Si(CH3)3
Me O
O
O
Me
O
O
O
O
O
O
127= stabilized analog of 128 O O
Me
O
Me
O
O
O
OEt O
OEt
O
biosynthetic intermediate 129
Na
reactivity towards primary amines: Me
Me
O
Me
Me
Me
Me Me
O
O
Me
O
O
1- H
NH2 N then H (66%)
O
Me 127
O Me
OEt
NH2
130
H
Ph
2- NaBH4, N 3- 2N HCl then NaBH4 Ph (55%) Me
H Me
HN
Me
131
Scheme 6.32 Alternative strategy towards halicyclamines with the late introduction of nitrogens.
H O O R1
N
+ Me CN 103
CHO
ZnBr2
106 R1
N
Zn(OTf)2
HN
Me
Me
Me
H
HN
R2
104 or 116
R
N
ircinal / manzamine A model
Me
H Me
halicyclamine model R2
118
Scheme 6.33 Marazano divergent route to either ircinal/manzamine A or halicyclamine-type alkaloids.
6.7 Biomimetic Approaches toward Other Manzamine Alkaloids 6.7.1 Biomimetic Models of Madangamine Alkaloids
Pro-ircinal-type alkaloids could undergo core fragmentation via intracyclic (path A) or pericyclic (path B) vinylogous retro-Mannich reactions (Scheme 6.34). Following redox transformations, a sequence of enamination and vinylogous aza-Mannich reaction would eventually produce madangamine type alkaloids. Path (B) was pioneered biomimetically by Marazano and colleagues (Scheme 6.35) in an oxidized version, based on (carboxyethyl)acetoacetate dianion
6.7 Biomimetic Approaches toward Other Manzamine Alkaloids
H
[O]
N
O H
O H
O
N
HN
N
H
path B
retro-Mannich vinylogous then H
path A
HN
HN
retro-Mannich vinylogous O H
O
then [O]
Ircinal type alkaloids N
209
N
aza -Mannich vinylogous
N
HN
retro-Mannich vinylogous
[H]
then [H] O H
O
N
N
Mannich vinylogous HN
HN
O HN N
O
imine formation
Mannich vinylogous imine formation
HN
N N
N
aza-Mannich vinylogous then [H]
N
madangamine type alkaloids N
Scheme 6.34
Possible biogenesis of madangamine C type alkaloids.
and quaternarized dihydropyridinium 133 as a biomimetic equivalent of postulated intermediate 132 (see Scheme 6.34) [57]. Following a double Mannich addition, tricycle 134 was obtained in close analogy with the core of madangamine C-type alkaloids. In 2011, the total synthesis of madangamine type alkaloids remains a mountain to climb [58].
6 Biomimetic Synthesis of Manzamine Alkaloids
210
O
O
O
O 133
EtO
HN
OEt
N
HN N
132
O
O
EtO 133
O
O
Na OEt
O
1-THF, RT
HN COCF3 Bn N
2- K2CO3, EtOH/H2O Bn N reflux (50%) 134
N
Me
Me
Scheme 6.35
N
N H
madangamine C 11 [Xestospongia ingens]
Biomimetic synthesis of the madangamine C core.
6.7.2 Biomimetic Model of Nakadomarine A
Alkaloids of the ircinal A type could also undergo intracyclic fragmentation via vinylogous retro-Mannich reaction to give 135 (Scheme 6.36). Subsequently, a vinylogous Mannich reaction would enable ring closure of 135, yielding fused tetracycle 136. Final cyclization to furan would produce alkaloids of the nakadomarine A type. Alternatively, furan formation could occur from diketonic 135 to 137, enabling a furan-Mannich intramolecular cyclization to nakadomarine alkaloids. This last hypothesis was validated in the laboratory by Nishida and colleagues on models [59], before their publication of the first total synthesis of (+)-nakadomarine O H
O H OH retro -Mannich N vinylogous
N
H O OH Mannich vinylogous
OH N N
N
O OH N
N
N 136
Ircinal A 79
135
furan −H2O formation
furan −H2O formation
H O N
Mannich vinylogous
O N
N 137
Scheme 6.36
N nakadomarine A 10
Possible biogenesis of nakadomarine A from ircinal A.
H
6.7 Biomimetic Approaches toward Other Manzamine Alkaloids
211
O 19 steps Bs
N
O CO2Me
Bs =PhSO2
HN
AcO
H
AcO
AcO H
H O OAc p -TsOH Bs N N Boc
O
H
138 THPO
CO2Me 139
(R)-(–)-140
H N
O
4 steps
N Boc 141
H O
HCl Bs
N
H (87%, 2 steps)
THPO
12 steps
N Boc
O N N H
142
HO
(+)-nakadomarine A 10
Scheme 6.37 Focus on the biomimetic key step in the first total synthesis of nakadomarine A.
A (10) (see also Section 6.9.8) in 2003 (Scheme 6.37) [60], featured by a key biomimetic step. Advanced intermediate 138 was prepared in 23 steps from racemic 139 (via optically active intermediate 140). The authors made successful an intramolecular furan-iminium cyclization of spiropyrrolidine 141 into 142, an elegant way of mimicking the presumed core biosynthesis of nakadomarine A type alkaloids. 6.7.3 Biomimetic Models of Sarains: A Side Branch of the Manzamine Tree
With its highly intricate diazatricyclic central core and two macrocyclic side chains sarain A (6), isolated from Reniera sarai in 1986 and fully characterized in 1989 (X-ray analysis of a diacetate derivative), is one of the most complex manzamine alkaloid (Scheme 6.38), featuring an unprecedented pentacyclic, box-like heterocyclic architecture. Inspection of this alkaloid reveals a polyenic 1,2,3-aminodiol, sphingolipid-like moiety, suggesting a distinct biogenetic origin relative to the manzamines previously described in this chapter. Indeed, retro-biosynthetic analysis of sarain A type alkaloids10) using iminium-based disconnections provides with key amino-aldehyde 15 and two C3 synthons [postulated as malondialdehyde (17) according to Marazano’s model], along with cyclic amino acid 144 (that may result from the catabolism of sphingolipid 143) (Scheme 6.38) [61]. From these biogenetic elements, but with a philosophy identical to what we have seen up to now, it thus appears that sarain A type alkaloids are not directly connected to the elaborated manzamines presented before, with which they only share simple amino-aldehyde precursors 15. Alkaloids related to sarain A should thus be regarded as branching 10) So far, sarain A is the only isolated man-
zamine alkaloid to possess this unprecedented polycyclic core.
212
6 Biomimetic Synthesis of Manzamine Alkaloids
O O
b O
NH2
15
NH2 O
O
a
a : enamine formation b : aldol-crotonization
H N
b
a
NH2
O
HOOC
O
144
HOOC
−4 H2O Sphingolipid type 143 O
pH sensitive O proximity N
N
N
N HOOC
145
sarain A 6 HO
reduction
HO
O sarain type alkaloids
N
N
O
Mannich
N
N
147
Scheme 6.38
O H2O
HN
N
O
146
Proposed biogenesis of sarain A-type alkaloids.
off early on in the manzamine metabolism, in a case of divergent biosynthesis, with postulated intermediates 145–147 represented on Scheme 6.38. Marazano et al. pioneered the biomimetic synthesis of the heterocyclic core of sarain A. In their 1999 paper, a first model, albeit incomplete, gave the first experimental evidence that manzamines and sarains are biosynthetically related (Scheme 6.39). A thermodynamic mixture of two aldehydes (148 and 149) was obtained when reacting dihydropyridinium salt 34 and glutaconaldehyde salt 150. The sequence occurred via the rearrangement of aminal 151 into iminium 152 on contact with alumina. Compound 148 contains a bicyclic system reminiscent of that of sarain A 6. A few years later [61b], using β-bromoacrylamide 153 as aminated malondialdehyde equivalent (Scheme 6.40), and benzylidene 154 as an amino acid surrogate, another model study was disclosed. Those reactants were coupled under basic conditions to yield glutarimide 155 after benzylidene hydrolysis. Next, 155 was coupled with malondialdehyde (17) sodium salt to furnish aminopropenal 156, which was N-alkylated to yield 157. After a series of tedious reductions of the glutarimide moiety, the authors were able to obtain aminal 158, which underwent Sakurai-type cyclization under the described conditions to afford tricyclic sarain A model 159.
6.8 A Biomimetic Tool-Box for the Synthesis of Manzamine Alkaloids H
O
N Ph 34
Me 150
Me
Ph CH2Cl2 Me ONa
N
(95%)
Me
N
Al2O3 or SiO2 O
HO
Me
Ph
Bn
N
N O
Me 148
149
Biomimetic synthesis of a first sarain A model. O O
NH Br
154
O
Me 2:3 ratio (33%) O thermodynamic equilibrium
O
N
153
EtO
Me
Ph
N
O
O
O
152
HO
Scheme 6.39
Me
O
151 de: 90%
sarain A 6 alternative representation
Me
Ph
O
213
O
LDA Bn
N
(70%)
O N
then MeI Bn N
N Me H 156 O
Bn
Me 155
O
H
O
H
O
NaH
NH2 O
Ph
N O Me
17
ONa
N O Me Me 157 10 steps
O
R
N
17
O
N
HO
R'' R'
TMS
biosynthetic analogy
H N
R NH2 O
R''
O R' N
N
H Ts
N
N Me
biosynthetic analogy
Me
HO sarain A 6
Scheme 6.40
159
FeCl3 (61%) Ts
N HO
N Me Me 158
HO
Biomimetic synthesis of a second sarain A model.
6.8 A Biomimetic Tool-Box for the Synthesis of Manzamine Alkaloids: Glutaconaldehydes and Aminopentadienals
As seen above, simple aminopentadienals constitute C5 biomimetic equivalents of long-chain aminopentadienals (types 1 and 2), regarded as key precursors of the manzamine alkaloids. Glutacondialdehydes are hydrolyzed counterparts of aminopentadienals and can also be seen as equivalents of postulated aldehyde intermediates (refer to Scheme 6.2). Such biomimetic C5 nucleophiles can be obtained in a few steps from simple starting materials, using Mannich addition of imine anions onto vinamidinium salts (cf. Scheme 6.9), or, in a more versatile manner, using the Zincke opening of N-activated pyridinium salts (cf. Scheme 6.8). In their first model reactions (Schemes 6.25 and 6.26), the Marazano group exploited the reactivity of more stable aminopentadienoates (but presenting inappropriate
6 Biomimetic Synthesis of Manzamine Alkaloids
214
first generations
biomimetic electrophiles OMe
N
R CN O
in situ generateddihydropyridinium salts
O
H N
R
Ph
N
R''' R''
intermolecular reactions
R O NH Ph
Ph
R
"increasing biomimetism"
R
masked dihydropyridinium salts
Ph
biomimetic nucleophiles
R'
OMe wrong oxidation state
N
"increasing biomimetism"
Ph
O
aminopentadienoates
R''' HN R'' aminopentadienals
glutaconaldehydes
O
O
O
R'''
N R''' intramolecular reactions
N
R'' R'' equivalents of type 1
latest developments
R''' N
OK
R'' R'' equivalents of type 2
(see scheme 6.2)
Figure 6.5 Evolution of the biomimetic chemical tool-box used for accessing manzamine alkaloids.
oxidation relative to aldehydes), though those species were rapidly abandoned in favor of more versatile aminopentadienals (Figure 6.5). On the borderline of the chemistry of manzamine alkaloids, several experimental studies were conducted to delimit the scope of reactivity of glutaconaldehydes. More or less biosynthetically related structures were obtained. One of the simplest reactions reported is the self-dimerization of glutaconaldehyde 150 formed in situ from the corresponding sodium salt under acidic conditions (Scheme 6.41) [62]. In fact, according to the mechanism depicted, cinnamaldehyde 160 could be obtained in very good yield. A similar outcome was observed when a monoprotected malonaldehyde unit 161 was engaged in the cascade [63]. Formation of the resulting adduct 162, which displays the same aromatic pattern as the one encountered in 160, was explained by the loss of a carbon in the form of a formic acid molecule. The chemistry of glutaconaldehydes was recently explored beyond biomimetic considerations, especially by the Vanderwal group. Intramolecular ring openings O
ONa O
p -TSA, CH2Cl2
Me
150 Me O
O
O 150 Me
Me Me
O O Me Me
O O
O 161
O
Bu4NCl, CH2Cl2
Me O
O
Me
Me O Me
Me
O
O
O
Me
O
p -TSA: p -toluenesulfonic acid
NaO
O
–HCO2H
162 (32%)
Me
Me
Me O 150
Scheme 6.41
O
Me 160 (80%)
Me O
O H
Me O
Me O
O Me
O
OH
Selected examples of reactivity of glutaconaldehydes.
O
Me Me HO OH
6.9 Biosynthesis of Manzamine Alkaloids: Towards a Universal Scenario
of pyridinium salts (including ‘‘Zincke salts’’) gave easy access to multiple heterocycles [64]. Aminopentadienals (‘‘Zincke aldehydes’’) were also shown to undergo pericyclic cascades to provide synthetically useful (Z)-α, β, γ , δ,-unsaturated amides [65] or served for the preparation of δ-tributylstannyl-α, β, γ , δ,-unsaturated aldehydes [66]. Applications to the synthesis of natural products or natural product-like analogs include the access to indolomonoterpenic alkaloid cores of the strychnane, aspidospermane, or ibogane types (with a total synthesis of norfluorocurarine), a formal synthesis of porothramycins or the total synthesis of nicotine and analogs [67]. Nuhant, Delpech, and colleagues disclosed a methodology driven study dealing with the activation of aminopentadienals towards nucleophiles which also enabled completion of the synthesis of protoemetinol [68].
6.9 Biosynthesis of Manzamine Alkaloids: Towards a Universal Scenario
In the following, we propose a general mapping of the biosynthesis of typical manzamine alkaloids, as emerging from presumed biogenetic relationships as well as biomimetic chemistry evidence. This mapping finds its introduction in Schemes 6.1, 6.24, and 6.38 and is extended in Scheme 6.42. The reader will find additional details in terms of intermediates and alternative biosynthetic connections in the corresponding isolated schemes. To provide the simplest yet broadest vision of the possible relationships between the manzamine alkaloids, we again depict macrocycles by large loops (see footnote 3). 6.9.1 From Fatty Acids to Long-Chain Aminoaldehydes and Sarain Alkaloids
As depicted in Scheme 6.1, fatty acid degradation would produce two kinds of biosynthetic reactants: (i) dialdehydes 14 (C8 –C16 ) that would become monoaminated to yield amino-aldehydes 15; (ii) acrolein (16) (the original C3 species hypothesized by Baldwin) or malondialdehyde (17) (alternative C3 species proposed by Marazano in 1998). Incorporation of a sphingolipid moiety to this metabolism would open the path towards sarain A-type alkaloids (Scheme 6.38), which appear to branch off very early from the biogenetic trunk of the manzamine family. 6.9.2 Pyridine Alkaloids: Theonelladine, Cyclostellettamine, and Xestospongin-Type Alkaloids
Condensation of amino-aldehydes 15 with a source of ammonia would furnish theonelladine-type alkaloids. Alternatively, dimerization of 15 in the presence of two equivalents of acrolein or malondialdehyde would give rise not only to alkaloids of the cyclostellettamine type but also of the xestospongin-type when the alkyl chains are β-hydroxylated (Schemes 6.1 and 6.11).
215
NH
N N H HH OH N
O Mannich
Scheme 6.42
175
N
N
H 181
N
H
aza-Mannich vinylogous
N
H 2O
H 2O
174
O N H
N
HN
HN H
H2O
H2O
178
N H O H 2O
H 2O
N
N
O
HN
HN
N
HN H
furan formation
O H2O N
O O
180
H
N
173
N
177
HN
H
O
[O]
[O/H]
[H]
N H tryptamine 80 O Pictet-Spengler
H
N
H
N
H
HN
H
H
RetroMannich vinylogous 172
O Mannich vinylogous
176
O
O
H
N
H
N
O
N
N
HN
H
[O] O
HN
170
171
OH
N
N H OH
Manzamine A type alkaloids
H
N
N H
tryptamine 80 Pictet-Spengler [O] allylic amination
[O] allylic amination O H
[O]
tryptamine 80 OH Pictet-Spengler
Ircinal A type alkaloids
retro-Mannich vinylogous
Mannich OH vinylogous
O
N
N
N
H
Mapping of the presumed biogenetic relationships between representative manzamine alkaloids.
madangamine C type alkaloids
N
N
N
Mannich O vinylogous
nakadomarine A type alkaloids
H
N
retro-AzaMannich vinylogous
NH 179
NH N H HH OH retro-Mannich N
manadomanzamine A type alkaloids
O
H
2- oxidation allylicamination
1- Retro-Mannich vinylogous
H
216
6 Biomimetic Synthesis of Manzamine Alkaloids
O
O
oxidation
17 O
16
Scheme 6.42
(Continued)
Marazano's hypothesis
fatty acids catabolism
Baldwin's hypothesis
reduction
N
[H]
N
N
N H
O
N
aza-Mannich vinylogous
N
N
aza-Mannich
aza-Mannich vinylogous
H2O [O]
N
163
N
(4 + 2) cycloaddition
H
[H] N
N
[O] N N
N
N
[O]
[H]
165
166
N
N
H
H
H
H
N
HN
H2O H2O
N
dehydrohalicy clamine-type lkaloids
HN
halicyclamine A / haliclonamine type alkaloids
keramaphidine B type alkaloids 169 retro-aza-Mannich
164
N
O
N 168
formal (4+2) cycloaddition
NH
O
N
N
H O
HN
H
167
O
[H]
HN
H
tryptamine 80 Pictet-Spengler allylic amination
6.9 Biosynthesis of Manzamine Alkaloids: Towards a Universal Scenario 217
218
6 Biomimetic Synthesis of Manzamine Alkaloids
6.9.3 From Cyclostellettamines to Keramaphidin and Halicyclamine/Haliclonamine Alkaloids
Incorporation of acrolein as C3 precursor according to Baldwin’s model implies late, possibly spontaneous, oxidation of dihydropyridine/dihydropyridinium intermediates to pyridine/pyridinium. Competitive with these oxidation processes, two kinds of reaction might occur: (i) intramolecular Diels–Alder cycloaddition of bis-dihydropyridiniums 163 to bridged intermediates 164, which could either be reduced to keramaphidin B type alkaloids or (ii) undergo retro-aza-Mannich fragmentation to afford bis-piperidine skeletons 165 typical of halicyclamine/haliclonamine alkaloids following redox transformation. These alkaloids could also arise directly from bis-dihydropyridinium 163 upon intramolecular vinylogous aza-Mannich cyclization, and also from aminopentadienal-dihydropyridinium 166 if malondialdehyde (17) is involved as key C3 unit (Marazano’s model). 6.9.4 Spinal Cord of Manzamine Metabolism: The Ircinal Pathway
Aminopentadienal-dihydropyridiniums 166 are the direct precursors of isoquinoline aldehydes 167 following intramolecular [4 + 2] cycloaddition. Such pro-ircinals represent the earliest entry point in the biosynthesis of ircinal alkaloids (168 to ircinal A type alkaloids), of which they possess most structural features, except for their pyrrolidine ring. Pro-ircinals 168 could also be accessed by regioselective oxidation of keramaphidin type alkaloids to 169 followed by iminium hydrolysis, although this theory has not been supported experimentally. All pro-ircinals and ircinal intermediates should be considered central to the ‘‘manzamine metabolism,’’ and seem to constitute branching points toward the majority of structurally representative alkaloids (Scheme 6.24). 6.9.5 From Ircinal and Pro-ircinals to Manzamine A Alkaloids
Although ircinal-type molecules are seen as immediate precursors for manzamine A type alkaloids, it should be kept in mind that β-carboline formation is not the obligate last step toward these alkaloids. Indeed, pro-ircinals 170 and 171 are candidates for oxidative allylic amination/β-carboline formation, making them all potentially direct precursors of manzamine A-type alkaloids. 6.9.6 From Pro-ircinals to Madangamine Alkaloids
Pro-ircinal alkaloids 170 could undergo a vinylogous retro-Mannich fragmentation, giving rise to spiro-piperidine 172. Following intramolecular redox transfer or sequential oxidoreduction, spiranic tetrahydropyridinium 173 would be
6.10 Total Syntheses of Manzamine-Type Alkaloids
produced. Madangamine-type alkaloids would eventually be yielded upon cyclizing enamination (174) and vinylogous aza-Mannich reaction with a final reduction of 175. 6.9.7 From Pro-ircinals to Manadomanzamine Alkaloids
Intramolecular epoxidation of pro-ircinals 171 would produce isoquinoline aldehydes 176, prone to Pictet–Spengler reaction with tryptamine (80). Following regioselective oxidation, iminium 177 would be hydrolyzed to propionaldehyde derivative 178, which is amenable to iminium formation with the neo-formed tetrahydroisoquinoline system. Mannich addition of an acetone equivalent onto the residual iminium of 179 would eventually produce manadomanzamine-type alkaloids. 6.9.8 From Ircinals and Pro-ircinals to Nakadomarine Alkaloids
Ircinal A-type alkaloids could undergo vinylogous retro-Mannich fragmentation, identical to the one undergone by 170 in the supposed biosynthesis of madangamine alkaloids (cf. Scheme 6.34). Subsequently, cyclization of diketone 180 to furan 181 would enable intramolecular vinylogous Mannich addition to effect ring closure, yielding the fused tetracyclic system typical of nakadomarine A-type alkaloids. The implication of furan nucleophiles in this biosynthesis is directly suggested from biomimetic experiments (cf. Scheme 6.37). Approaching the end of this marine story, let us finally emphasize how the chemist’s intuitions proved to be right and were corroborated (sometimes afterwards) by the isolation and characterization of new informative structures of natural products. Bear in mind that the structure of ircinal A (79) and keramaphidin (7) were not yet known when Whitehead and Baldwin proposed their pioneering biosynthetic model, and that pyridinium-piperidine intermediates such as 95 were postulated before being discovered in sponges (cf. Scheme 6.23). As the latest nod from Nature to the chemical community, the manzamine-type alkaloids zamamidines [69] [see the structure of zamamidine C (182), Scheme 6.43] recently gave striking presumptive evidence for the C3 (acrolein/malonaldehyde) scenario. In fact, a C3 link such as 16 clearly unifies a tetrahydromanzamine A molecule with a β-carboline moiety; a plausible biosynthesis from ircinal A (79) is therefore easily conceivable with the intervention of two molecules of tryptamine (80).
6.10 Total Syntheses of Manzamine-Type Alkaloids
The Baldwin–Whitehead and Marazano’s biosynthetic hypotheses have provided a useful framework to develop synthetic approaches to the manzamines alkaloids.
219
220
6 Biomimetic Synthesis of Manzamine Alkaloids
NH N H
N zamamidine C 182 [Amphimedon sp.]
NH O
N H OH
N N
80
16 H2 N O
80 N H Pictet-Spengler Pictet-Spengler ircinal A and oxidation to b-carboline 79 NH2
H
Scheme 6.43
Retrobiosynthesis of zamamidine C.
Nevertheless, stepwise strategies have been initiated worldwide over the last 20 years toward these remarkable alkaloids. To date, the total synthesis of manzamine A (1) [70], nakadomarine A (10) [60, 71], haliclonacyclamine C [72], and sarain A (6) [73] have been achieved and beautifully illustrate the state-of-the art in methods for highly complex molecule construction. These total syntheses as well as the numerous chemical approaches are not included in this chapter. Natural ircinals and manzamines have been subjected to semi-synthetic transformations, especially in the Hamann group, providing a wide range of derivatives for diverse biological screenings and studies [74]. In addition, total synthesis has also enabled the preparation of various simplified analogs that are unreachable from natural material [75].
6.11 Conclusion
From a biosynthetic standpoint, the most striking observation is that the great majority of manzamine alkaloids can be connected by means of reversible reactions (such as Mannich, aza-Mannich, Michael, and aldol), and thus be potentially interconvertible at a biochemical level. According to this hypothesis, only a limited number of alkaloids should be considered structurally ‘‘terminal,’’ that is, those few formed by enzymatically irreversible steps (such as Pictet–Spengler cyclization to β-carboline systems). In this general biogenetic proposal, reductions would have the role of freezing reactive intermediates (e.g., iminium species) into stable alkaloids. However, it must be realized that several alternative reoxidations (i.e., at other levels of the molecule) probably remain feasible biochemically, due to similar redox potentials of related iminium alkaloids. Moreover, conformation-induced intramolecular electron transfers and dismutations have the potential to occur spontaneously, as suggested by several observations in the laboratory (cf. Schemes 6.18 and 6.25). Overall, the conception of ‘‘manzamines in equilibrium’’ for this rich metabolism of marine alkaloids (that might be driven by subtle ecological changes) yields a particularly striking picture of dynamic chemical evolution and diversity-oriented biogenesis. These latter suggestions have been, at least partly, magnificently demonstrated by track records of
References O
O
H N
N
N
N
H O
O N
HN
y
NH
HO OHC N
OH
x
misenine 12
Figure 6.6
manadomanzamine A 9
haliclonine A 183 [Haliclona sp.]
Challenging manzamines as future synthetic targets.
successful applications at the biomimetic chemistry level (cf. Schemes 6.18, 6.28, and 6.33). Synthetic effort toward the manzamine alkaloids will certainly continue in the future as new exciting structures periodically appear in the literature. Molecules such as misenine (12) (for which no logical biosynthetic route can yet be proposed), complex, highly rearranged haliclonine A (183) [76], or indolic manadomanzamine (9), will surely keep stimulating chemists because of their intrinsic beauty and not only because of interesting biological properties (Figure 6.6). Future synthetic endeavors will most probably be guided by the body of growing biosynthetic studies currently performed worldwide with marine organisms.
References 1. Sakai, R., Higa, T., Jefford, C.W., and
2.
3.
4.
5.
6.
7.
8.
Benardinelli, G. (1986) J. Am. Chem. Soc., 108, 6404–6405. Kobayashi, J., Murayama, T., Ohizumi, Y., Sasaki, T., Ohta, T., and Nozoe, S. (1989) Tetrahedron Lett., 20, 4833–4835. Fusetani, N., Asai, N., Matsunaga, S., Honda, K., and Yasumuro, K. (1994) Tetrahedron Lett., 35, 3967–3970. Talpir, R., Rudi, A., Ilan, M., and Kashman, Y. (1992) Tetrahedron Lett., 33, 3033–3034, see Reference [16] for an important comment concerning the structure of niphatoxin A. Schmitz, F.J., Hollenbeak, K.H., and Campbell, D.C. (1978) J. Org. Chem., 43, 3916–3922. Cimino, G., Mattia, C.A., Mazarella, L., Puliti, R., Scognamiglio, G., Spinella, A., and Trivellone, E. (1989) Tetrahedron, 45, 3863–3872. Kobayashi, J., Tsuda, M., Kawasaki, N., Matsumoto, K., and Adachi, T. (1994) Tetrahedron Lett., 35, 4383–4386. Jaspers, M., Pasupathy, V., and Crews, P. (1994) J. Org. Chem., 59, 3253–3255.
9. Peng, J., Hu, J.-F., Kazi, A.B., Li, Z.,
10.
11.
12.
13.
14.
Avery, M., P´eraud, O., Hill, R.T., Franzblau, S.G., Zhang, F., Schinazi, R.F., Wirtz, S.S., Tharnish, P., Kelly, M., Wahyuono, S., and Hamann, M.T. (2003) J. Am. Chem. Soc., 125, 13382–13386. Isolation and structure determination: Kobayashi, J., Watanabe, D., Kawasaki, N., and Tsuda, M. (1997) J. Org. Chem., 62, 9236–9239. (a) Kong, F., Andersen, R.J., and Allen, T.M. (1994) J. Am. Chem. Soc., 116, 6007–6008; (b) Kong, F., Graziani, E.I., and Andersen, R.J. (1998) J. Nat. Prod., 61, 267–271. Guo, Y., Trivellone, E., Sconamiglio, G., and Cimino, G. (1998) Tetrahedron, 54, 541–550. Jimenez, J.I., Goetz, G., Mau, C.M.S., Yoshida, W.Y., Sheuer, P.J., Williamson, R.T., and Kelly, M. (2000) J. Org. Chem., 65, 8465–8469. See among others: (a) Peng, J., Rao, K.V., Choo, Y.-M., and Hamann, M.T. (2008) in Modern Alkaloids, Structure,
221
222
6 Biomimetic Synthesis of Manzamine Alkaloids
15. 16.
17.
18.
19.
20. 21.
22.
23.
24.
25.
Isolation, Synthesis and Biology (eds E. Fattorusso and O. Tagliatela-Scafati), Wiley-VCH Verlag GmbH, Weinheim, pp. 189–232; (b) Hu, J.-F., Hamann, M.T., Hill, R., and Kelly, M. (2003) in The Alkaloids, Chemistry and Biology, vol. 60 (ed. G.A. Cordell), Academic Press, San Diego, pp. 207–285. Baldwin, J.E. and Whitehead, R.C. (1992) Tetrahedron Lett., 33, 2059–2062. (a) Kaiser, A., Billot, X., Gateau-Olesker, A., Marazano, C., and Das, B.C. (1998) J. Am. Chem. Soc., 120, 8026–8034; (b) Jakubowicz, K., Ben Abdeljelil, K., Herdemann, M., Martin, M.-T., Gateau-Olesker, A., Al Mourabit, A., Marazano, C., and Das, B.C. (1999) J. Org. Chem., 64, 7381–7387. Many mechanisms have been proposed involving among others the formation of endoperoxides or enzymatic reactions in the course of the biosynthesis of inflammatory mediators such as prostaglandins and thromboxanes and so on. See for example: Esterbauer, H., Schaur, R.J., and Zollner, H. (1991) Free Radic. Biol. Med., 11, 81–128. Laville, R., Thomas, O.P., Berru´e, F., Reyes, F., and Amade, P. (2008) Eur. J. Org. Chem., 121–125. Laville, R., Thomas, O.P., and Amade, P. (2009) Pure Appl. Chem., 81, 1033–1040. Chichibabin, A.E. (1906) J. Russ. Phys. Chem. Soc., 37, 1229. Wypych, J.-C., Nguyen, T.M., Nuhant, P., B´en´echie, M., and Marazano, C. (2008) Angew. Chem. Int. Ed., 47, 5418–5421. See for example: Husson, H.-P., Grierson, D., and Harris, M. (1980) J. Am. Chem. Soc., 102, 1064–1082. Gil, L., Gateau-Olesker, A., Marazano, C., and Das, B.C. (1995) Tetrahedron Lett., 36, 707–710. (a) Zincke, T. (1903) Justus Liebigs Ann. Chem., 330, 361–374; (b) Zincke, T. (1904) Justus Liebigs Ann. Chem., 333, 296–345; (c) Zincke, T. and Wurker, W. (1905) Justus Liebigs Ann. Chem., 338, 107–141. Review article: Cheng, W.-C. and Kurth, M.J. (2002) Org. Prep. Proc. Int., 34, 585–608.
26. Especially useful for the synthesis of
27. 28.
29.
30.
31.
32. 33.
34.
35.
36.
37.
38.
chiral non-racemic pyridinium salts, see among many applications: (a) Comp`ere, D., Marazano, C., and Das, B.C. (1999) J. Org. Chem., 64, 4528–4532; (b) Guilloteau-Bertin, B., Comp`ere, D., Gil, L., Marazano, C., and Das, B.C. (2000) Eur. J. Org. Chem., 1391–1399. Review article: Becher, J. (1980) Synthesis, 589–612. Nguyen, T.M., Peixoto, S., Ouairy, C., Nguyen, T.D., B´en´echie, M., Marazano, C., and Michel, P. (2010) Synthesis, 103–109. Wypych, J.-C., Nguyen, T.M., B´en´echie, M., and Marazano, C. (2008) J. Org. Chem., 73, 1169–1172. Isolation: Fusetani, N., Yasumuro, K., and Matsunaga, S. (1989) Tetrahedron Lett., 30, 6891–6894. Michelliza, S., Al Mourabit, A., Gateau-Olesker, A., and Marazano, C. (2002) J. Org. Chem., 67, 6474–6478; (b) See also, the total synthesis of cyclostellettamines A–F by the Baldwin group exploiting a N-oxide strategy: Baldwin, J.E., Spring, D.R., Atkinson, C.E., and Lee, V. (1998) Tetrahedron, 54, 13655–13680. Kaiser, A., Marazano, C., and Maier, M. (1999) J. Org. Chem., 64, 3778–3782. (a) Isolation of viscosamine: Volk, C.A. and K¨ock, M. (2003) Org. Lett., 5, 3567–3569; (b) total synthesis of viscosamine: Timm, C. and K¨ock, M. (2006) Synthesis, 2580–2584. (a) Isolation and total synthesis: Timm, C., Volk, C.A., Sasse, F., and K¨ock, M. (2008) Org. Biomol. Chem., 6, 4036–4040; (b) See also: Timm, C., Mordhorst, T., and K¨ock, M. (2010) Mar. Drugs, 8, 483–497. Nakagawa, M., Endo, M., Tanaka, N., and Gen-Pei, L. (1984) Tetrahedron Lett., 25, 3227–3230. Kobayashi, M., Kawazoe, K., and Kitagawa, I. (1989) Chem. Pharm. Bull., 37, 1676–1678. Baldwin, J.E., Melman, A., Lee, V., Firkin, C.R., and Whitehead, R.C. (1998) J. Am. Chem. Soc., 120, 8559–8560. Maia, A.A., Mons, S., Pereira de Freitas Gil R., and Marazano, C. (2004) Eur. J. Org. Chem., 1057–1062.
References 39. Kiewel, K., Luo, Z., and Sulikowski, 40. 41.
42.
43.
44.
45. 46.
47.
48.
49.
G.A. (2007) Org. Lett., 9, 5051–5054. Colabroy, K.L. and Begley, T.P. (2005) J. Am. Chem. Soc., 127, 840–841. (a) Tsuda, M., Hirano, K., Kubota, T., and Kobayashi, J. (1999) Tetrahedron Lett., 40, 4819–4820; (b) Hirano, K., Kubota, T., Tsuda, M., Mikami, Y., and Kobayashi, J. (2000) Chem. Pharm. Bull., 48, 974–977 (a) Baldwin, J.E., Romeril, S.P., Lee, V., and Claridge, T.D.W. (2001) Org. Lett., 3, 1145–1148; (b) Snider, B.B. and Shi, B. (2001) Tetrahedron Lett., 42, 1639–1642; (c) Romeril, S.P., Lee, V., Claridge, T.D.W., and Baldwin, J.E. (2002) Tetrahedron Lett., 43, 327–329; (d) Romeril, S.P., Lee, V., Baldwin, J.E., Claridge, T.D.W., and Odell, B. (2003) Tetrahedron Lett., 44, 7757–7761; (e) Morimoto, Y., Kitao, S., Okita, T., and Shoji, T. (2003) Org. Lett., 5, 2611–2614; (f) Pouilh`es, A., Amado, A.F., Vidal, A., Langlois, Y., and Kouklovsky, C. (2008) Org. Biomol. Chem., 6, 1502–1510 (a) Ishiyama, H., Tsuda, M., Endo, T., and Kobayashi, J. (2005) Molecules, 10, 312–316. (a) Pyrinadine A: Kariya, Y., Kubota, T., Fromont, J., and Kobayashi, J. (2006) Tetrahedron Lett., 47, 997–998; (b) pyrinadines B–G: Kariya, Y., Kubota, T., Fromont, J., and Kobayashi, J. (2006) Bioorg. Med. Chem., 14, 8415–8419 Anwar, M. and Lee, V. (2009) Tetrahedron Lett., 65, 5834–5837. Kondo, K., Shigemori, H., Kikuchi, Y., Ishibashi, M., Sasaki, T., and Kobayashi, J. (1992) J. Org. Chem., 57, 2480–2483. Baldwin, J.E., Claridge, T.D.W., Heupel, F.A., and Whitehead, R.C. (1994) Tetrahedron Lett., 35, 7829–7832. (a) Baldwin, J.E., Claridge, T.D.W., Culshaw, A.J., Heupel, F.A., Lee, V., Spring, D.R., Whitehead, R.C., Boughtflower, R.J., Mutton, I.M., and Upton, R.J. (1998) Angew. Chem. Int. Ed., 37, 2661–2663; (b) Baldwin, J.E., Claridge, T.D.W., Culshaw, A.J., Heupel, F.A., Lee, V., Spring, D.R., and Whitehead, R.C. (1999) Chem. Eur. J., 5, 3154–3161. Gomez, J.-M., Gil, L., Ferroud, C., Gateau-Olesker, A., Martin, M.-T., and
50.
51.
52.
53.
54.
55.
56. 57.
58.
59.
60.
61.
62.
63.
64.
Marazano, C. (2001) J. Org. Chem., 66, 4898–4903. Matsunaga, S., Miyata, Y., van Soest, R.W.M., and Fusetani, N. (2004) J. Nat. Prod., 67, 1758–1760. Among many others, see for example: Kita, Y., Toma, T., Kan, T., and Fukuyama, T. (2008) Org. Lett., 10, 3251–3253, and references cited therein. Herdemann, M., Al Mourabit, A., Martin, M.-T., and Marazano, C. (2002) J. Org. Chem., 67, 1890–1897. Gil, L., Baucherel, X., Martin, M.-T., Marazano, C., and Das, B.C. (1995) Tetrahedron Lett., 36, 6231–6234. Isolation: Harrison, B., Talapatra, S., Lobkovsky, E., Clardy, J., and Crews, P. (1996) Tetrahedron Lett., 37, 9151–9154. Sinigaglia, I., Nguyen, T.M., Wypych, J.-C., Delpech, B., and Marazano, C. (2010) Chem. Eur. J., 16, 3594–3597. Sanchez-Salvatori, M.R. and Marazano, C. (2003) J. Org. Chem., 68, 8883–8889. Tong, H.M., Martin, M.-T., Chiaroni, A., B´en´echie, M., and Marazano, C. (2005) Org. Lett., 7, 2437–2440. For recent approaches, see: Amat, M., P´erez, M., Proto, S., Gatti, T., and Bosch, J. (2010) Chem. Eur. J., 16, 9438–9441, and references cited therein. Nagata, T., Nishida, A., and Nakagawa, M. (2001) Tetrahedron Lett., 42, 8345–8349. Nagata, T., Nakagawa, M., and Nishida, A. (2003) J. Am. Chem. Soc., 125, 7484–7485. (a) Hourcade, S., Ferdenzi, A., Retailleau, P., Mons, S., and Marazano, C. (2005) Eur. J. Org. Chem., 1302–1310; (b) Ge, C.S., Hourcade, S., Ferdenzi, A., Chiaroni, A., Mons, S., Delpech, B., and Marazano, C. (2006) Eur. J. Org. Chem., 4106–4114. Sanchez-Salvatori, M.R., Lopez-Giral, A., Ben Abdeljelil, K., and Marazano, C. (2006) Tetrahedron Lett., 47, 5503–5506. Lopez-Giral, A., Mahuteau-Betzer, F., Gateau-Olesker, A., and Marazano, C. (2003) Eur. J. Org. Chem., 1859–1867. Kearney, A.M. and Vanderwal, C.D. (2006) Angew. Chem. Int. Ed., 45, 7803–7806.
223
224
6 Biomimetic Synthesis of Manzamine Alkaloids 65. Steinhardt, S.E., Silverston, J.S., and
66.
67.
68.
69.
70.
71.
Vanderwal, C.D. (2008) J. Am. Chem. Soc., 130, 7560–7561. Michels, T.D., Rhee, J.U., and Vanderwal, C.D. (2006) Org. Lett., 10, 4787–4790. (a) Martin, D.B.C. and Vanderwal, C.D. (2009) J. Am. Chem. Soc., 131, 3472–3473; (b) Michels, T.D., Kier, Kearney, A.M., Vanderwal, C.D. (2010) Org. Lett., 12, 3093–3095; (c) Peixoto, S., Nguyen, T.M., Crich, D., Delpech, B., Marazano, C. (2010) Org. Lett., 12, 4760–4763. Nuhant, P., Raikar, S.B., Wypych, J.-C., Delpech, B., and Marazano, C. (2009) J. Org. Chem., 74, 9413–9421. (a) Takahashi, Y., Kubota, T., Fromont, J., and Kobayashi, J. (2009) Org. Lett., 11, 21–24; (b) Yamada, M., Takahashi, Y., Kubota, T., Fromont, J., Ishiyama, A., Otoguro, K., Yamada, H., Omura, S., and Kobayashi, J. (2009) Tetrahedron, 65, 2313–1317 (a) Winkler, J.D. and Axten, J.M. (1998) J. Am. Chem. Soc., 120, 6425–6426; (b) Martin, S.F., Humphrey, J.M., Ali, A., and Hillier, M.C. (1999) J. Am. Chem. Soc., 121, 866–867; (c) Humphrey, J.M., Liao, Y., Ali, A., Rein, T., Wong, Y.-L., Chen, H.-J., Courtney, A.K., and Martin, S.F. (2002) J. Am. Chem. Soc., 124, 8584–8592; (d) Toma, T., Kita, Y., and Fukuyama, T. (2010) J. Am. Chem. Soc., 132, 10233–10235. (a) Ono, K., Nakagawa, M., and Nishida, A. (2004) Angew. Chem. Int. Ed., 43, 2020–2023; (b) Young, I.S. and Kerr, M.A. (2007) J. Am. Chem. Soc., 129, 1465–1469; (c) Jakubec, P., Cockfield, D.M., and Dixon, D.J. (2009) J. Am. Chem. Soc., 131, 16632–16633; (d) C. D.
72.
73.
74.
75.
76.
Vanderwal recently wrote a review article with a fine analysis of the different strategies, and the interested reader can advantageously refer to this article: Vanderwal, C.D. (2010) Angew. Chem. Int. Ed., 49, 2830–2832 (a) Smith, B.J. and Sulikowski, G. (2010) Angew. Chem. Int. Ed., 49, 1599–1602; (b) Smith, J.B., Qu, T., Mulder, M., Noetzel, M.J., Lindsley, C.W., and Sulikowski, G.A. (2010) Tetrahedron, 66, 4805–4810 (a) Garg, N.K., Hiebert, S., and Overman, L.E. (2006) Angew. Chem. Int. Ed., 45, 2912–2915; (b) Becker, M.H., Chua, P., Downham, R., Douglas, C.J., Garg, N.K., Hiebert, S., Jaroch, S., Matsuoka, R.T., Middleton, J.A., Ng, F.W., and Overman, L.E. (2007) J. Am. Chem. Soc., 129, 11987–12002. See among other recent publications: (a) Peng, J., Kudrimoti, S., Prasanna, S., Odde, S., Doerksen, R.J., Pennaka, H.K., Choo, Y.-M., Rao, K.V., Tekwani, B.L., Madgula, V., Khan, S.I., Wang, B., Mayer, A.M.S., Jacob, M.R., Tu, L.C., Gertsch, J., and Hamann, M.T. (2010) J. Med. Chem., 53, 61–76; (b) Wahba, A.E., Peng, J., Kudrimoti, S., Tekwani, B.L., and Hamann, M.T. (2009) Bioorg. Med. Chem., 17, 7775–7782 See among others: (a) Winkler, J.D., Londregan, A.T., and Hamann, M.T. (2006) Org. Lett., 8, 2591–2594; (b) Winkler, J.D., Londregan, J.R., Ragains, A.T., and Hamann, M.T. (2006) Org. Lett., 8, 3407–3409 Jang, K.H., Kang, G.W., Jeon, J., Lim, C., Lee, H.-S., Sim, C.J., Oh, K.-B., and Shin, J. (2009) Org. Lett., 11, 1713–1716.
225
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids J´erˆome Appenzeller and Ali Al-Mourabit
‘‘la chimie est a` la biologie ce que le solf`ege est a` la musique’’ (i.e., ‘‘chemistry is to biology what musical notation is to music.’’) Pierre Potier 7.1 Introduction
Targeted and exhaustive metabolomic studies have become crucial for the comprehension of the origin and reactivity of natural substances. The latter investigations have become extremely important for any biogenetic proposal and even for conceiving efficient biosynthetic experiments using labeled early precursors. Prior to the association of structures to build families of biogenetically related molecules, the bioprospecting and collecting of natural sources for molecule isolation have to focus on phylogenetically related species. Although the fingerprints of secondary metabolites generated by biogenetic key-steps in living systems are essentially highly controlled by the genome, it is important to take into account that enzymatic key events could be followed by intrinsic spontaneous reactions toward each final molecule. Some highly chemo- and stereoselective transformations could be due exclusively to structural pre-organization of the molecules and to the highly chiral environment in cells. Isolation and reactivity analysis of classes of molecules might contribute to efficient synthetic plans, in particular for molecules that could be synthesized by cascade processes. The objective of this chapter is to compile some relevant biomimetic reactions and synthetic achievements based on biomechanistic analyses in the pyrrole-2-aminoimidazole and guanidinium alkaloid series. To distinguish biomimetic from non-biomimetic syntheses, we propose some (not absolute) criteria: • synthetic plan inspired by biosynthetic hypotheses; • the conversion of a natural product into another natural product following biomechanistic thoughts; • the use of a minimum number of protecting groups; Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
226
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
• the use of original biomimetic reactions (often one-pot reactions) with mild conditions in a biomimetic medium (water as solvent, use of acids or bases, photochemistry, . . .); • the design of straightforward and highly chemoselective reactions. 7.1.1 Introduction to Pyrrole-2-Aminoimidazole (P-2-AI) Marine Alkaloids
Pyrrole-2-aminoimidazole alkaloids (P-2-AIs) constitute a family of alkaloids that to date have been isolated exclusively from marine sponges, particularly from Agelasidae, Halichondridae, and Axinellidae families. The extraordinary molecular diversity can be generated from the central precursor clathrodin (1) and its brominated derivatives, that is, hymenidin (2) and oroidin (3). These C11 N5 molecules contain a pyrrole carboxamide moiety linked to a 2-aminoimidazole ring by a propenamine chain. The highly diverse family of P-2-AIs metabolites can be conceived of as derivatives of these C11 N5 monomers with regards to oxidation, reduction, hydration, or methylation of both 2-amino-4(5)-vinylimidazole and pyrrole-2-carboxamide units. Polycyclization events or dimerization could take place, involving tautomerism of the 2-amino-4(5)-vinylimidazole system or the nucleophilic properties of the pyrrole (Scheme 7.1, path a) or the tautomerically controlled C–C and C–N bond formation between two monomers (path b), respectively. Complex transformations of clathrodin into dimeric P-2-AIs involving tautomerism, C–C bonds formation, C–N bonds formation, oxidation, and cascade reactions could lead to interesting molecules like palau’amine (11) congeners (path c). Finally, further fragmentation events of the previous compounds would afford degraded members, clinching this entire class of marine natural products (path d). The present chapter completes our previous comprehensive review analyzing the biomechanistic aspects of P-2-AIs [1], and is dedicated to biomimetic achievements. Many of the recent synthetic studies based on the biogenetic proposals culminated in the total synthesis of several P-2-AIs members. Importantly, the most intricate steps and retrosynthetic strategies for P-2-AIs alkaloids were directly inspired by the biochemical pathways crossed back and forth by the numerous biogenetic proposals put forward for their biosynthesis after isolation. The P-2-AIs metabolites will be classified in terms of linear monomers, cyclic monomers, simple dimers, complex dimers, and tetramers. Linear monomers such as dispacamide A (4) are directly related to oroidin while cyclic monomers such as dibromoagelaspongin (6), dibromophakellin (7), and agelastatin A (8) contain new rings in addition to 2-aminoimidazole and pyrrole. In simple dimers (2 × C11 N5 ), the two subunits are linked by one or two bonds, such as in sceptrin (9), whereas in complex dimers more than two bonds appear between the two C11 N5 subunits, such as in benzosceptrin A (10). Palau’amine (11) and its close congeners require several reactions, including chlorination, cascade reaction, oxidations, and hydrolysis. Only a few tetramers such as stylissadine A (12) have been reported; their presence indicates the promising molecular diversity of P-2-AIs marine alkaloids. Not surprisingly, some degradation compounds such as 13–17 were also reported.
2
5
NH
1
7 9
H N O
11
15
X
N 12 H
O
NH
H N O
N H
NH2
N
14
N H 13
Scheme 7.1
H2N
H 2N
N
COR
NH
16 R = OH 17 R = NH2
N H
N H 15 O
O
Y
Y
d
polycyclic tetramers
c
polycyclic dimers
b
polycyclic monomers
a
O
9
N
5
OH
H
7
12
N H
1 3
NH
H OH N NH
NH Br 3'
5
7 7'
HN
O
Br
3'
H2N
HN
Br
15'
NH
1'
H N
H N
Br
Br
7'
OH
O
HN
NH
N H H
N
N
H
O
HN
N
N
NH2
N H
N
NH2H2N
O
HN N
7
NH
O
O
9 NH
7
5
H
N 3
NH
1 Me
N H
N
O
O
N
7'
HN
NH
O
NH2
O
NH
Br
N H H
N
H N OH
HN Br
NH
Br
NH2
O
H N
8: agelastatin A
11
N
12
HO
10 benzosceptrin A
H2N
H N
Br
12 stylissadine A
7
O
7: dibromophakellin
NH2 N 2'
O
Br
Br
Complex Structures
9 sceptrin
3
NH
11 palau'amine
15
9N 5
H
7
HN
1
9 NH
O
H2N 2 N
15
H
N 12 11
Cl 1' HN
Br
7'
NH2
O
15 11
N12
H
1N
3N
2
6: dibromoagelaspongin
Br
Br
H2N
Clathrodin (1) and some examples of related P-2-AIs.
N
NH2
Degradation Compounds
d
4: X = Y = Br, dispacamide A 5: X = Br, Y = H, dispacamide B
N
H2N
1: X = Y = H, clathrodin 2: X = Br; Y = H, hymenidin 3: X = Y = Br, oroidin
3N
H2N
Reactive Monomers X
Br
7.1 Introduction 227
228
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
The case of the misleading propenamine 13 in regards to biogenesis is interesting. While it was considered to be an early precursor of oroidin through acylation by 4,5-dibromopyrrole-2-carboxylic acid, its formation is very likely the result of the hydrolysis of oroidin or similar derivatives. Exhaustive reviews dealing with P-2-AIs structures, activities, and syntheses have been published recently [2]. We focus here on the biomimetic syntheses of P-2-AIs published since 2005 and also on recent insights in terms of biogenetic hypotheses. The biogenesis of P-2-AIs can be studied at two levels of complexity: (i) from the pool of amino acids to clathrodin (1) and dispacamide (4, 5) skeletons or (ii) from the latter to more complex structures via polycyclizations or dimerizations (Scheme 7.2). In 2001, our group postulated that the tautomeric equilibria of the aminopropenylimidazole portion of oroidin (3) were the key processes involved in the formation of both polycyclic monomers and cyclized dimers [1]. This is certainly a consequence of the corresponding ambivalent nucleophilic/electrophilic behavior of oroidin derivatives. The proposed universal chemical pathway for the formation of P-2-AIs members, known up to 2000, is still valid. All complex dimeric P-2-AIs members isolated during the last decade [2f, 3] fit in with the general chemical pathway that we have previously suggested for their formation and relative stereochemical relationships. The correction of palau’amine relative configuration (11), and consequently those of congeners [4], supported the hypothesis. The universal biochemical pathway is also in accordance with the conserved stereochemical relationships within the dimer subclass. Some additional and particular biogenetic hypotheses have been suggested by Baran and K¨ock starting from sceptrin (9) [2c, 5]. In fact, the most fundamental question regarding the biosynthesis of P-2-AI alkaloids concerns the first event connecting monomers like hymenidin (2) to dimers like sceptrin (9).
Early precursors proline ornithine arginine histidine lysine
Complex structures
Central intermediates NH2 NH N O NH pre-clathrodin events
post-clathrodin events
HN 1: clathrodin HN
N
HN O 18: debromodispacamide B
Scheme 7.2
H H N
NH2
NH O
NH2 Cl HN
O
NH
NH H OH N NH NH N
11: palau'amine
Pre- and post-clathrodin (1) events involved in the biogenesis of P-2-AIs.
The biogenetic hypothesis defined the basic C11 N5 structure of P-2-AIs alkaloids and introduced the concept of the dual reactivity originated by the 2-aminoimidazole section (Figure 7.1). The ambivalent effect that propagates along the vinylogous chain defines five possible tautomers (I–V) with various modes of cyclization and dimerization. Nucleophilic and electrophilic reacting positions can vary depending on the tautomer engaged in the dimerization step [1]. The investigations, including molecular
7.1 Introduction
12 NH
15
11 9
NH
16
O
HN H 5 7 17 1N H2 N N 4 δ−
HN 1N
H2N
I
HN 1N 5
7
δ+ δ+
H2N N IV
4
5
NH
O 1N 5
7
δ− δ− 3 N 4 H II
δ+
HN
O
7
H2N N 4
δ+
III
NH
NH
O
O
HN δ− 7 H 5 1N H2N δ− N 4 V
Tautomers of type I, II, III, IV and V will be extensively used in the following section
Figure 7.1 Tautomerism in the building blocks of the P-2-AI monomer clathrodin (1); ambivalent reactivity of vinylogous 2-aminoimidazole.
calculations, concluded that the tautomeric interconversion of I–V is more easily explained in acidic conditions [6]. The process leading to the selection of the reacting tautomers in Nature raises an interesting question. The dynamic hydrogen-bonding interaction of monomers like clathrodin (1) with the host enzyme is probably one of the most intriguing biochemical catalytic systems. The proposed isomerization mechanism is probably operative with precise modulation dictated by the sponge genome. The consequence appears in the composition of the P-2-AI targeted metabolome, which changes with the sponge species. From a synthetic point of view, the apparently obvious dimerization step proved difficult to realize in the laboratory. The biogenesis process of P-2-AI metabolites can be divided into two major events (Scheme 7.2): (i) pre-clathrodin events linking early precursors, presumably amino acids, to monomers like clathrodin (1) or debromodispacamide B (18) (Scheme 7.2) and their brominated derivatives (e.g., oroidin) and (ii) chemical transformations of the crucial precursors clathrodin or debromodispacamide B into higher order P-2-AI members. 7.1.2 Proposed Biogenetic Hypothesis for Clathrodin (1) and Related Monomers Starting from α-Amino Acids
Regarding the pre-clathrodin events several more or less intuitive biosynthetic pathways have been proposed by different groups starting from various amino acids. The postulated key steps rely on some presumed key precursors isolated from phylogenetically related sponges or co-isolated together with other P-2-AIs members. In some cases, synthetic results and/or radioactive labeling experiments provided
229
Scheme 7.3
16
H N
HN
NH
19 HN
NH OH
13
N
NH2 O
N NH
NH2
Br
Br
N H
HO2C Br
O
N H
COOH
Br
NH COOH
N
N H
N NH2
NH2
O OH proline
H N
Kerr-Pomponi
C radiolabelled experiments
14
Br
23
Andersen
NH
NH2
NH2
21: stevensine
O
N
N HOOC H histidine ornithine HOOC
NH2
Lindel H2N
NH
N N H 22: clathramide A O
Br
HN HN O Hydrolysis 1: X = Y = H, clathrodin HN 2: X = H, Y = Br, hymenidin 3: X = Y = Br, oroidin
X NH
NH NH
Y
H 2N
acylation
H 2N
Biosynthetic proposals for clathrodin and its brominated derivative oroidin (3).
O
N H
COOH
Br
H N
H2N
OH H2N
O
COOH Köck
20 isolated
NH
lysine
HN
H2N
H 2N NH2
oxidation
O
proline OH
H N
NH2
COOH
ornithine
Kitagawa-Braeckman
H2N
COOH
230
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
7.1 Introduction
additional support. Failing any biosynthetic demonstration for the real early precursors involved in the formation of the 2-aminoimidazole and 2-aminoimidazolone parts, hypotheses on the pre-clathrodin events remain purely speculative at present. The most commonly cited route as deduced from the isolated metabolites would start from proline, ornithine, histidine, arginine, or lysine amino acids. Scheme 7.3 gives an overview of the different proposals. Pyrrole-2-carboxylic acid (16) and 2-amino-5-(3-amino)propylimidazole (13) have been isolated from many sponges. This ‘‘misleading’’ result suggests that clathrodin (1) could be formed by the acylation of 13 by 16 (Scheme 7.3). If the latter suggestion is right, the biosynthesis of oroidin could be inferred from understanding the biogenesis of compounds 13 and 16 independently. Proline was assumed by all authors to be the precursor of the pyrrole-2-carboxamide part by double oxidation, while several possibilities have been suggested for the origin of the 2-amino-5-(3-amino)propylimidazole unit. Many hypotheses based on primary amino acids have been proposed for 13 while proline was exclusively considered for pyrrole-2-carboxylic acid (16). According to us, the origin of the 2-aminoimidazole moiety could actually be much more subtle. In fact, 2-amino-5-(3-amino)propylimidazole (13), supposed to be the precursor of oroidin, might rather be the product of its hydrolysis. For Kitagawa [7] and Braeckman [8], ornithine and its guanidic analog are the precursors of the vinyl-2-aminoimidazole (13) (see Scheme 7.3). More recently Andersen [9] and coworkers completed this proposal by an alternative route for the incorporation of ornithine into oroidin alkaloids via a guanidine-containing fragment (23) related to stevensine (21). Kerr et al. reported the incorporation of the 14 C-labeled proline, ornithine, and histidine into the cyclized oroidin derivative stevensine (21) in a cell culture of the sponge Teichaxinella morchella [10]. The mechanistic incorporation of histidine into oroidin remains unresolved. Unfortunately, only low levels of radioactivity and specific incorporation levels were observed with amino acids in the Kerr experiments ([U-14 C]His, 1460 dpm (0.026%); [U-14 C]ornithine, 1300 dpm (0.024%)), and these results along with the lack of in vivo biosynthetic studies leave the question unanswered [11]. Lindel [12] complemented the Kerr’s proposal regarding the incorporation of histidine. This could proceed via the clathramide A like precursor 22 (Scheme 7.3). The quaternary methyl group in 22 would provide the additional carbon, leading to the C11 N5 skeleton. The inclusion of this carbon was thought to proceed through a cyclopropane intermediate. However, this step is based on yet unobserved amination of histidine at position 2 of the imidazole nucleus. K¨ock isolated [13] the guanidine-containing bromopyrrole compound 20 from the sponge Agelas wiedenmayeri. This compound, which corresponds to the guanidylated lysine 4-bromopyrrole-2-carboxamide, was also highlighted as a precursor of P-2-AI early intermediates. Further hydroxylation, cyclization, and decarboxylation reactions could afford the oroidin skeleton from compound 19 (see Scheme 7.3). More recently Al-Mourabit and coworkers proposed a biogenetic pathway [14] for P-2-AI early precursors from proline and related arginine. This proposal (Scheme 7.4) was based on the isolation of verpacamides A (24) and C (25) from the
231
Br
O
N NH
NH2
O arginine
HO
3: oroidin
HN
NH
+
H2N
NH
Br Br
N
H N
NH
H O
O
HN
N
N
NH
H O
O
NH
NH2
N
N
NH
O
N
NH
NH2
O
N
1: clathrodin
HN
NH
HN
HN O OH N O
NH
NH2
critical transfer of the guanidine H H2N N
26: putative intermediate for dibromodispacamide B
oxidation O
HN
H2N
HN O 18: debromodispacamide B
NH
O
or
25: verpacamide C cyclo(Pro-Pro)
O
H2N
N
4: dispacamide A
HN
NH
24: verpacamide A cyclo(Pro-Arg)
O H
H2N
HN
Scheme 7.4 Al-Mourabit’s retro-biogenetic proposal for P-2-AIs via cyclo(Pro-Arg) and further cyclo(Pro-Pro) diketopiperazines as precursors.
Br
NH
OH
proline
O
H2N
NH
232
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
7.2 Ground Work of George B¨uchi: Dibromophakellin (7) Synthesis from Dihydrooroidin (31)
233
sponge Axinella vaceletti. The consideration of cyclo(Pro-Arg) and cyclo(Pro-Pro) diketopiperazines distinguishes this hypothesis from others. This family of proline and arginine diketopiperazines could be the precursor of dispacamide A via dioxygen-mediated oxidation. Notably, two prolines and one guanidine (or one proline and one arginine) yield the C11 N5 skeleton of P-2-AIs. Importantly, the co-isolated verpacamides A (24) and C (25) show an increasing degree of oxidation en route to P-2-AIs through arginine cyclization into a proline residue bearing the required guanidine. The biomimetic syntheses of dispacamides following a similar pathway have been achieved recently in the group of Al-Mourabit (Section 7.3) [15, 16].
7.2 Ground Work of George B¨uchi: Dibromophakellin (7) Synthesis from Dihydrooroidin (31)
Among the many contributions of George B¨uchi (1921–1998) in the field of organic chemistry, his ground work on the synthesis of natural products over 50 years reveals his elegant style. Several creative and original syntheses of natural products were indeed achieved in B¨uchi’s laboratory. His synthetic conceptions were often associated with biomimetic synthesis. The synthesis of parazoanthoxanthin (27) [17] and dibromophakellin (7) [18], following some biomimetic thoughts, can be seen as a landmark in the biomimetic chemistry of P-2-AIs and complex guanidinium alkaloids in general. The first total synthesis of dibromophakellin (7) will be depicted and discussed in details. B¨uchi first investigated the oxidation of 2-aminoimidazoles, which was applied to the syntheses of parazoanthoxanthin (27) and dibromophakellin (7) (Scheme 7.5). The key steps involved the acidic dimerization of 2-aminoimidazolylethanol (28) to give 27 via vinylimidazole 29, and the oxidative intramolecular cyclization of dihydrooroidin (30) to give 7. Dibromophakellin (7) [19] is a complex and compact tetracyclic structure associated with mild biological activities. Dibromophakellin exhibits surprising stability to hydrolysis despite the presence of two aminal groups. Its unusual architecture made it a very attractive target for total synthesis. Dibromophakellin has been isolated from several sponges of the families Agelasidae and Axinellidae in its two enantiomeric forms. Another interesting facet of this structure is that O HN
HN H2N
H2N
N
N
HO
28
29
N H 2N
N
N H
O
NH
N
N NH2
27: parazoanthoxanthin
Scheme 7.5 Buchi’s ¨ foundation of oxidative and ambivalent reactivity of 2-aminoimidazole (2-AI) toward marine metabolites.
Br
N Br
HHN
NH N NH2
7: dibromophakellin
Br
N N H
NH2
Br
30: dihydrooroidin
234
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
dibromophakellin is a subunit met in one of the most challenging complex marine metabolites: palau’amine (11). The construction of the dibromophakellin skeleton en route to palau’amine is still a hot topic. Dibromophakellin is biogenetically related to oroidin through formal N9–C5 and N12–C4 cyclizations. B¨uchi proposed dihydrooroidin (30) as a common precursor to both oroidin (3) and dibromophakellin (7) (Scheme 7.6). Dihydrooroidin could lead to oroidin through oxidation at C6–C7 or to dibromophakellin under the same oxidative conditions through N9–C5 and N12–C4 cyclizations. This strategy was remarkable at the time in the sense that very few P-2-AI members had been isolated from sponges in 1980. The synthesis of dihydrooroidin (30) was achieved in five steps and 35% overall yield starting from l-citrulline. It began with the preparation of 2-aminoimidazole-4-propanamine (31) from citrulline. The key step was the Lancini condensation [20] of the corresponding aminoaldehyde with cyanamide leading to the 2-aminoimidazole nucleus. The aminoaldehyde was prepared by Akabori reduction [21]. The pyrrole subunit was introduced at a late stage of the synthesis using the acylation of 31 in the presence of 4,5-dibromo-2-trichloroacetylpyrrole. The dihydrooroidin (30) was then oxidized by elemental bromine to yield racemic dibromophakellin (7) after treatment with tBuOK. The precise mechanism involved in the latter reaction cascade remains unclear. This ground-breaking synthesis was straightforward and easily scalable, and still provides a rapid access to racemic dibromophakellin. A similar strategy was used 25 years later by Harran and coworkers in a synthetic study of dibromophakellin [22]. The principles illustrated in this synthesis were developed by Horne and coworkers in the synthesis of oroidin, dispacamides, dibromoisophakellin, dibromophakellstatin, and even dimeric mauritiamine [23]. The 2-aminoimidazole part, which is thought to be very sensitive to oxidative reactions, could be installed using preparative synthetic chemistry. Although the polycyclization of the dihydrooroidin (30) intermediate was seemingly biomimetic, the biosynthetic pathway leading to the 2-aminoimidazole (2-AI) from citrulline remains obscure. The principal findings of B¨uchi’s work are certainly the high propensity of 2-aminoimidazole to be oxidized and its ambivalent reactivity. These properties are clearly reminiscent of those employed in the synthesis of parazoanthoxanthin (27) by B¨uchi himself, and also of many subsequent developments of P-2-AIs chemistry.
7.3 Biomimetic Synthesis of P-2-AI Linear and Polycyclic Monomers
The pre-clathrodin synthetic details linking amino acids toward monomers like clathrodin or debromodispacamide and their brominated derivatives are unknown. Only the biomechanistic analyses of isolated P-2-AIs and their synthetic achievements based on speculative biosynthesis have contributed to progress in the understanding of these early steps. The isolation of an increasing number of
COOH
NH2
L-citrulline
NH2
H HN
4
5
N
NH2
N
a, b, c, d
N NH2
NH
31: 2-aminoimidazole -4-propanamine
H2N e
30: dihydrooroidin
Br HN
NH
O
Br NH
O
N H
f
HN
N H
Br
NH2
N
5
N
9
N 4 Br H HN
12
O
N NH2 rac-7: dibromophakellin
biomimetic step oxidative polycyclization
NH2
N
3: oroidin
Br
30: dihydrooroidin
Br
biogenetic hypothesis O N H Br NH N Br HN NH2
Scheme 7.6 Buchi’s ¨ biogenetic proposal for oroidin (3) and dibromophakellin (7), and biomimetic synthesis of rac-7 from L-citrulline. Reagents and conditions: (a) EtOH, HCl; (b) Na/Hg; (c) NH2 CN, HCl; (d) NaOH, heat, 71% overall yield from L-citrulline; (e) 4,5-dibromo-2-trichloroacetylpyrrole, Na2 CO3 , DMF, 50%; and (f) (i) Br2 , AcOH, (ii) tert-BuOK, 2-butanol; quantitative.
HN
O
Br
12 N
9
7: dibromophakellin
Br
O
7.3 Biomimetic Synthesis of P-2-AI Linear and Polycyclic Monomers 235
HN
Scheme 7.7
R = H : 90% R = OH : 71%
EDCI, DMAP, DCM reflux, 2h
N O
NH
N
O
R = H, 33 R = OH, 34
H3CO2C
N H R
NH2.1/2 H2CO3
NH
R = H : 56% R = OH : 27%
m. s. 4 Å, DMF, O2 90 °C, 30 mn
H2N
N N
R
O
HO
N
O
HN
NH O
R
HN
O
N
NH2
R = H, 18: debromodispacamide B R = OH, 39: debromodispacamide D
NH2
NH
26
H2N
N
N
NH
R
O
R = H, 37 R = OH, 38
O
N
O
HN
H2N
nucleophilic attack after α ketooxidation
R = H, 35 R = OH, 36
O
- oxidative proline rearrangement
- oxidative guanidine addition
Al-Mourabit’s biomimetic synthesis of debromodispacamides B and D.
R
CO2H
R=H R = OH 32
H3CO2C
+
N H 16
O
NH2
biogenetic hypothesis
18: debromodispacamide B
HN
NH
236
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
7.3 Biomimetic Synthesis of P-2-AI Linear and Polycyclic Monomers
putative biosynthetic intermediates was extremely important in developing hypotheses regarding the pre-clathrodin or pre-dispacamide biochemical events. The logic constructed through intermediate isolation from natural sources, combined with synthetic achievements, is of high value for the formulation of biosynthetic scenarios. The next section is based mainly on this reasoning. 7.3.1 Biomimetic Synthesis of Linear Monomers
In contrast with complex P-2-AI members, very little has been made in terms of the biomimetic synthesis of simple monomers. Considering that these early transformations are very important for the elaboration of chemical diversity, our group has focused over the past years on biogenetically inspired syntheses of oroidin (3), dispacamides, and other linear monomers. 7.3.1.1 Debromodispacamides B (18) and D (39) and Dispacamide A (4) The suggested hypothesis starts from the putative key diketopiperazine of type 26, combining proline and arginine (Schemes 7.4 and 7.7). The synthesis of debromodispacamides B (18) and D (39) was completed using the same strategy [16] in one step from the pseudodipeptides 33 or 34 in the presence of guanidine and air dioxygen. When the reaction with guanidine was run in degassed solvent under argon, no rearrangement of proline into 2-aminoimidazolone was observed. The one-pot reaction of the pseudopeptide pyrrole-proline methyl esters 33 and 34 in the presence of guanidine carbonate and air oxygen, in DMF at 90 ◦ C, was found to afford stereoselectively the desired debromodispacamides B (18) and D (39). The natural products were isolated in 44% yield from proline and 19% from 4-hydroxyproline, respectively, in two steps only and without protecting groups. Notably, the synthesis of (R)-debromodispacamide D from l-trans-4-hydroxyproline methyl ester (32) was completely stereoselective and stereospecific whereas the natural product was isolated racemic. The bromine-containing dispacamide A (4) was prepared from 33 using the same strategy (Scheme 7.8) [15]. 7.3.1.2 Clathrodin (1) and Its Brominated Derivative Oroidin (3) The biomimetic synthesis of oroidin and related monomers [24] was inspired by the biogenetic proposals involving pyraxinine (14) [25] and dibromoagelaspongin (6) [26]. The oroidin motif can be identified in 6, an intricate tetracyclic P-2-AI member with two adjacent quaternary carbons. The pyridine containing pyraxinine 14 could be obtained from the vinylaminoimidazole (13) isolated from sponges. The proximity between oroidin (3) and dibromoagelaspongin (6) led to the formulation of a common chemical approach using the oxidative addition of guanidine derivatives to 1,2-dihydropyridine (44) (Scheme 7.9). The mechanistic rationalization for the formation of the pyridine derivative 14 from the 2-aminoimidazole (13) was based on the cyclization of tautomer 43 into 42 followed by the its ring opening when exposed to acidic or basic conditions. Based on this biogenetic proposal, a retrosynthetic scheme for oroidin was planed. The conversion of 43 into the
237
238
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids Br NH O N H
N
HN HN
N H
Boc
THF, O2 reflux, 3h 46%
H3CO2C 33
HN
NH
Br
O
Boc
HN R1 N
N R2
O
(i) AcOH, Br2, rt,
HN
(ii) CH2Cl2, TFA, rt,
NH2 N
NH
74%
OH O
OH O
40 R1, R2 = H, Boc
41
65%
CH3SO3H, 80 °C,
Br Br
NH O NH2 HN
N
NH O
4: dispacamide A
Scheme 7.8
Al-Mourabit’s synthesis of dispacamide A (4).
guanidic pyridine intermediate was assumed to be reversible through the putative intermediate 42. Mechanistic evidence for this process was reported by Al-Mourabit et al. [27]. The key step of this sequence was the oxidative addition of 2-aminopyrimidine (2-AP) mediated by bromine on the 1,2-dihydropyridine derivative (44) obtained by reductive acylation of pyridine. The masked 2-aminoimidazoline (45) was then deprotected to yield the expected oroidin derivative 46 and (Z)-oroidin (47). A final isomerization in acidic medium yielded oroidin (3). This four-step synthesis proved to be short and afforded a reasonable overall yield. Similar access to clathrodin (1) was proposed using non-brominated reagents [28]. 7.3.2 Biomimetic Synthesis of Cyclized Monomers 7.3.2.1 Cyclooroidin (48) An efficient conversion of oroidin into racemic cyclooroidin was reported by Lindel and coworkers by heating oroidin in protic medium [29]. The N12–C7 bond formation is believed to proceed via the azafulvene type IV tautomer (Scheme 7.10). 7.3.2.2 Dibromoagelaspongin (6) The polycyclized monomer dibromoagelaspongin (6) first isolated by Struchkov et al. [26] exhibits an intriguing tetracyclic structure with a quaternary carbon
N N
N
O
Br
Br
H N
44
N
O
b
or
N
N
N
Br 45
N
N
NH
N O
14 pyraxinine
NH2
Br
H2N
c
H
N
Br
H 2N
R 2N
Br
N
R1
N
NH 46
N
H N
42
R3 N
O Br
+ H2N
H2N
Br
N
H N
O
43
HN
4
NH 47
N
N
biogenetic hypothesis
d
H2N
NHR1
N
N
H N
O
O
H N
HN
NH
3: oroidin
3: R1 =
N H 13: R1 = H
H2N
Scheme 7.9 Al-Mourabit’s synthesis of oroidin (3) based on the biogenetic proposal for pyraxinine (14) and dibromoagelaspongine (6). Reagents and conditions: (a) 4,5-dibromo-2-carbonyl chloride, NaBH4 , −78 ◦ C, 20%; (b) 2-aminopyrimidine, Br2 , 42%; (c) NH2 OH, TEA, EtOH, heat, 47%; and (d) TFA-DCM, 50 ◦ C, 71%.
N
a
6 dibromoagelaspongin
Br
Br
H 2N
H OH N
Br
Br
Br
Br
NHR1
7.3 Biomimetic Synthesis of P-2-AI Linear and Polycyclic Monomers 239
240
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids H N
Br NH
H N
Br O
+
NH2 N HCOO− H
Br
a
N
12 NH
Br
H N
7
O
3: oroidin
NH2 N + HCOO− H
Tautomer IV
Br O
N 12
HN
7
Br H N +
NH2 − NH HCOO
48: cyclooroidin
Scheme 7.10 Lindel’s conversion of oroidin into rac-cyclooroidin (48). Reagents and conditions: (a) H2 O/ EtOH (4 : 1), 95 ◦ C, sealed tube, 45 h, 93%.
connected to three different nitrogens. Struchkov et al. proposed that the putative imidazolinone 49 would cyclize to afford the skeleton of 6 (Scheme 7.11). Al-Mourabit and Potier proposed a biomimetic pathway from oroidin (3) through the dual tautomer of type III. After cyclization into aminals 50 (path a) [1] or 51 (path b) [27] and tautomerization into more stable intermediates 52 and 53, respectively, similar N9–C4 oxidative cyclizations could yield the hemiaminal dibromoagelaspongin (6). The notable macrocycle 50 (path a) was proposed as the common intermediate for both dibromophakellin and dibromoagelaspongin [1]. The first N–C bond proceeded through tautomer III of the natural product oroidin itself (Scheme 7.11). Although this proposal has not been tested experimentally, the suggested transannular cyclization of the nine-membered intermediate 50 was used in the recently reported total synthesis of palau’amine [30]. An alternative avenue based on B¨uchi’s oxidation of dihydrooroidin (30) for the formation of 6 (path c) [18] was proposed by Feldman [31]. The reaction was thought to occur via 54 and 55, followed by an oxidative closure into the dibromoagelaspongin (6). Direct C4–N9 bond formation is also plausible (path d) to yield 56. Inspired by paths b–d, Feldman and coworkers achieved the first total synthesis [31] of dibromoagelaspongin (6) in 16 steps and 4.7% overall yield from an advanced imidazole sulfoxide (57, Scheme 7.12). After acylation with N-SEM protected dibromopyrrole derivative, the dihydrooroidin intermediate 58 was formed. A Pummerer oxidative cyclization yielded the N9–C4 bond in compound 59. The second biomimetic oxidative cyclization was realized in presence of NCS to yield 60 after N12–C4 bond formation. Regioselective oxidation and cyclization allowed for the formation of the tetracyclic core of agelaspongin with a bis-aminal at C4. Amination of 60 through the azide 61 and methoxy/hydroxy exchange led to rac-dibromoagelaspongin (6).
Br 3: oroidin
12
N H H N
N
Br
tautomerization
NH2
O
a
H N
N
N
Br
a
b
Br
12 NH
5 4
NH
N H
N
Scheme 7.11
oxidation
tautomerization
NH2
Br
O
54
NH
Br
NH
d
N
N
d
c
O
O
51
N
50
Br
O
Br
NH
N
Br
NH2
Br
N
N
H NH
9
O
O
tautomerization
NH2 Br
55
Br
H N
N
H N
12
N N
H N
Biomimetic hypotheses for dibromoagelaspongin (6).
30: dihydrooroidin
Br
O
c
NH
cyclization
NH2
Br Tautomer of type III
NH
b
Feldman's hypothesis from dihydrooroidin (paths c/d)
O
9 NH
4
4 9 NH
Al-Mourabit's biogenetic hypotheses from oroidin (paths a/b) 4
N
NH
H N
53 Br
O
N H N
52
12
N N
H N
N
4
Br
12
N H N
N
49 O
NH
O
Br
N N
OH H N
N OH N
H N
Br
oxidation cyclization NH2
6: dibromoagelaspongin
cyclization
N
Struchkov
oxidation
56 Br
5
NH2
Br
Br
NH2
Br
Br
NH2
Br
NH2
7.3 Biomimetic Synthesis of P-2-AI Linear and Polycyclic Monomers 241
N
57
Br
O
Br
N
58
9
Br
O
Si
SO2N(CH3)2 5 N S+ O− NH 4 N b
O
O
5 N
O
S
Br
Br 59
N
12
Br
Si c
O
Br
SO2N(CH3)2
Br 53
N H H N
N 4 N
9
N
NH2
55
O
Br
NH
N NH
60
N 12
Br
Br
O
NH
N H H N
N
Br
NH2
O
N
61
N
N
H3CO H N
Br
Br
N3
Br 31: dihydrooroidin
Cl SO2N(CH3)2 N S 4 d,e,f N N
N
H N
biogenetic hypothesis N
O
N
N N
NH2
Br
Br
6: dibromoagelaspongin
g,h
OH H N
Scheme 7.12 Feldman’s total synthesis of rac-dibromoagelaspongin (6). Reagents and conditions: (a) (i) (NH2 )2 , (ii) 2-trichloroacetyl-4,5-dibromo-N-SEM-pyrrole, Na2 CO3 , and (iii) HCl 1, 5M, 45%; (b) Tf2 O, 2,6-lutidine, CH2 Cl2 , −78 ◦ C, 45%; (c) (i) TFA then Bu4 NF and (ii) NCS, CH2 Cl2 , 65%; (d) HCl, CH3 OH; (e) mCPBA; (f) TMSN3 , ZnI2 , 36%; (g) H2 , Pd, TFA; and (h) H2 O, 90%.
O
SO2N(CH3)2 N S+ N O− a O
N H N
NH2
Br 6: dibromoagelaspongin
O
N
OH N
242
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
7.3 Biomimetic Synthesis of P-2-AI Linear and Polycyclic Monomers
Al-Mourabit et al. tested the biomimetic approach considering the intermediate 53 (Scheme 7.13) [32]. The strategy was based on a six-step sequence using 2-aminopyrimidine, which has the notable advantage of being less polar than guanidine. The tetrahydropyridine (62) was prepared from 5-aminopentanol and 2-trichloroacetylpyrrole on a multi-gram scale. It reacted with 2-AP in the presence of NIS to afford 63 in 52% yield. Bromination of the pyrrole ring by Br2 yielded 64. Compared to other oxidative reagents, the use of BaMnO4 to promote the dehydrogenation of intermediate 64 proved more successful, affording 65 in 34% yield. The protected cyclic oroidin 65 was used to investigate the biomimetic oxidative cyclization into the fused tetracyclic core of 6. The protected dibromoagelaspongin 66 was obtained using DMDO as oxidant. Although deprotection attempts of 66 to give 6 failed, validation of this biomimetic approach was achieved by the preparation of the intricate tetracyclic compound 66. 7.3.2.3 Dibromophakellin (7) and Dibromophakellstatin (69) Dibromophakellin is the representative member of the cyclized monomers, previously synthesized by B¨uchi in a pioneering work. In a biomimetic strategy, the N12–C4 and N9–C5 bonds should be formed in one step starting from open-chain precursors related to oroidin. Other non-biomimetic syntheses have proved to be interesting alternative routes, as for the related (−)-dibromophakellstatin (69) [33] reported by Lindel and coworkers. B¨uchi’s synthesis of phakellin was improved by Horne [34] and extended to phakellstatin by using a NBS-mediated oxidation. Feldman reported a biomimetic synthesis of phakellstatin [35] based on a Pummerer extended reaction towards the aminoimidazole ring (Scheme 7.14). The sulfide 67 was obtained in five steps and converted into the oroidin-like intermediate 68 after deprotection and coupling with 4,5-dibromo-2-trichloroacetylpyrrole. Oxidative tetracyclization starting from compound 68 was successful using the Stang reagent, PhI(CN)OTf, with formation of the key N12–C4 and N9–C5 bonds. Further oxidation mediated by CAN led straightforwardly to dibromophakellstatin (69). The conversion of 69 into dibromophakellin (7) was finally achieved by enol ether formation with Meerwein’s reagent and amination using ammonium propionate. While the following syntheses are not biomimetic, they are deliberately included in our discussion to illustrate biological inspiration from proline and cyclo(Pro-Pro) diketopiperazine. In 2008, Romo [36], and later Nagasawa [37], reported the enantioselective syntheses of (+)-phakellin starting from l-proline to construct the pyrrole-proline ketopiperazine moiety of dibromophakellin (7). The same strategy was employed for the first asymmetric synthesis of the close (+)-phakellin derivative (+)-phakellstatin by Romo in 2003 [38]. The synthesis confirmed the absolute configuration of the natural product, which is (−)-phakellstatin (hydrolyzed phakellin). The synthesis of the natural (−)-enantiomer by Lindel [39] confirmed the same configuration. The main features of Romo’s synthesis of the unnatural enantiomer (+)-dibromophakellstatin are outlined here. A pivotal step in the approach was the diastereoselective desymmetrization of the optically active cyclo (l-Pro-l-Pro) diketopiperazine 70 (Scheme 7.15). Stereocontrolled C-acylation of the corresponding monoenolate with benzyl chloroformate produced 71 as the
243
62
H N
a
O
N
63
Br
N H N
N N b
O
N
64
N H N
N
Br
Br
N c
O
53
65
N
Br
N H N
N
O
b NH N H N
N
Br
NH2
Br
Br
N d
O
66
N
N
Br
Br
N N HCl
OH N
Br oroidin Tautomer of type III
biogenetic hypothesis H N NH2 N N H N O Br
N H N
Br
NH2
Br 6: dibromoagelaspongin
O
N
OH N
Scheme 7.13 Al-Mourabit’s approach to rac-dibromoagelaspongin (6). Reagents and conditions: (a) NIS, 2-aminopyrimidine (2-AP), DMF/CH3 CN, 52%; (b) Br2 , CH2 Cl2 ; (c) BaMnO4 , 34% overall yield; and (d) DMDO, acetone then HCl, Et2 O, quantitative.
O
N
N H N
NH2
Br 6: dibromoagelaspongin
O
N
OH N
244
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
five steps
5
N N
SPh
a
Pht = phthalimide 67 P = SO2N(CH3)2
PN
N O
Br
12 5
68 HN
4
NH
Br
55
9 NH
Br
b
N
SPh
N
Br
NH
N
Br
NH2
N H
N
O
54
Br
Br H HN
Br
N
N H
c
NH
N
SPh
N
N
NH
biogenetic hypothesis O O
Br HN
NH
N H N
Br Br
69
H HN
N
O N
O
NH
Br
H HN
N
O N
NH2
N
rac-7 : dibromophakellin
d,e Br
NH2 30: dihydrooroidin
Br
O
Scheme 7.14 Feldman’s synthesis of dibromophakellstatin (69) and dibromophakellin (7) using a Pummerer extended reaction. Reagents and conditions: (a) (i)(NH2 )2 and (ii) 4,5-dibromo-2-trichloroacetylpyrrole, 64%; (b) PhI(CN)OTf, iPr2 NEt, CH2 Cl2 /CH3 OH, 60–73%; (c) CAN, CH3 CN/H2 O, 80–93%; (d) Et3 OBF4 , NaHCO3 , CH2 Cl2 , 45%; and (e) EtCO2 −, NH4 +, heat, 59%.
HN
N
PhtN
4
Br HHN
12N
9
NH2 7: dibromophakellin
Br
O
7.3 Biomimetic Synthesis of P-2-AI Linear and Polycyclic Monomers 245
246
O H
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
N
a
H N
O BnOOC
N H N
O
70
b,c,d
O BnOOC
N
N
5 steps
HOOC N
O
AcO
4 steps
71
N
O
72
CbzHN
H
H2NOC O
73
N N
O
74 e,f,g
Br
Br H H N N O HN
N
O
69: (+)-dibromophakellstatin
44% overall yield; (e) PhI(TFA)2 , pyridine, CH3 CN; (f) H2 , Pd/C, MeOH; and (g) NBS, THF, 35% overall yield. KHMDS = potassium 1,1,1,3,3,3-hexamethyldisilazide.
Scheme 7.15 Romo’s synthesis of (+)-dibromophakellstatin (69). Reagents and conditions: (a) KHMDS, ClCOOBn, THF, 70%; (b) PhSeBr, KHMDS, THF; (c) DMDO, CH2 Cl2 ; (d) SeO2 , dioxane,
major product. A finely tuned sequence of oxidation then yielded the pyrrole derivative 72 in three steps. Acylcarbinol (73) could be constructed using four additional steps of functional group transformations. Compound 74 with the two nitrogens in place required five more steps. The next key step was a Hoffmann rearrangement under oxidative conditions leading to the cyclic urea moiety of the targeted natural product. The synthesis of (+)-dibromophakellstatin (69) was achieved after Cbz-hydrogenolysis and pyrrole dibromination. Lindel and coworkers proposed a concise synthesis of the natural (−)-dibromophakellstatin (69) using the same methodology they employed for the racemic mixture [40]. Scheme 7.16 shows the key steps of the synthesis. The diastereoselective synthesis of compound 75 was achieved in four steps starting from 4,5-dibromo-2-carboxypyrrole and l-trans-4-hydroxyproline methyl ester. Debromination of the pyrrole followed by dehydration led to the bis-enamine 76. This compound was the substrate of the key diamination step using TsONHCOOEt. Hydroxyproline was the source of stereogenic information to control selectively
Br
HO HN
a-d
N
TBSO
75
N O
g
e,f
N TBSO
O
76
Scheme 7.16 Lindel’s synthesis of (−)-dibromophakellstatin (69). Reagents and conditions: (a) SOCl2 , MeCN, reflux; (b) HO-Pro-OMe, Na2 CO3 , room temp. (rt); (c) TBSCl, imidazole, DMF, rt; (d) DIBAL-H, CH2 Cl2 , −78 ◦ C, 49% overall
N
N
N
N
H N
TBSO
77
N
O
O TsO
Br
Br
OEt
O
N
O HO
Br
Br
Br
h-k
O
HN
N
O
69: (−)-dibromophakellstatin
yield; (e) H2 , Pd/C, NEt3 , MeOH, CH2 Cl2 , rt; (f) MsCl, DBU, 0 ◦ C, 81% overall yield; (g) TsONHCOOEt, CaO, CH2 Cl2 , 15 ◦ C, 50%; (h) NEt3 · 3HF, THF, rt, 97%; (i) CBr4 , PPh3 , rt, 73%; (j) SmI2 , THF–MeOH, rt, 77%; and (k) NBS, THF–MeCN, rt, 47%.
7.3 Biomimetic Synthesis of P-2-AI Linear and Polycyclic Monomers
the stereochemistry of the annulation during the formation of tetracycle 77. The reductive removal of the hydroxyl group was achieved via alkyl bromide formation and its reduction using SmI2 . Final dibromination of the pyrrole afforded the natural product (−)-dibromophakellstatin (69). The remarkable synthesis of (+)-phakellin (83) by Romo is depicted here (Scheme 7.17). The concise phakellin annulation approach is reminiscent of the approach developed for palau’amine by the same group. The hemiaminal 78, prepared from l-proline in three steps, was aminated into diastereoisomers 79 and 80 using diphenylphosphoryl azide followed by hydrogenolysis (Scheme 7.17). The diastereoisomer 79 was converted into the thermodynamically stable diastereoisomer 80 in basic medium. After the amidination of 80, the following step consisted in the crucial oxidative cyclization of compound 81, leading to the N-Tces phakellin (82) and performed using iodonium benzene-diacetate and magnesium oxide. Final deprotection yielded the (+)-phakellin (83). The mechanism proposed by Romo for the oxidative phakellin cyclization parallels the plausible biomimetic interconversion of the natural ugibohlin (84) and dibromoisophakellin (85) (box in Scheme 7.17) [41]. The enantiopure pyrrole-hydroxyproline-diketopiperazine 36 (Scheme 7.18) used in the synthesis of debromodispacamides (Section 7.3.1) was the starting material of the synthesis of (+)-phakellin hydrochloride (90) reported by Nagasawa and coworkers [37]. After selective oxidation and dehydration, the resultant α, β-unsaturated amide 86 was reduced under Luche conditions to yield allylic alcohol 87 after acetylation and desilylation of hydroxyl groups. The key step of the synthesis was the Overman trichloroacetimidate rearrangement [42], which provided the tertiary amide 88 with complete transfer of chirality on the quaternary carbon. The acetyl group in 88 was then replaced by a silyl group, through the thermodynamically favored isomerization at the carbinolamine position. N,N-Boc protected guanidine was then introduced to afford compound 89 after three additional steps. The final cyclization took place after silyl group removal, formation of a mesylate ester, followed by cyclization and N-Boc deprotection in acidic conditions. The synthesis of (+)-phakellin hydrochloride (90) was achieved in 18 steps. This synthesis provided the first example of the non-intuitive Overman [3,3] sigmatropic rearrangement applied to an enamide (87). 7.3.2.4 Hymenialdisines (91) Hymenialdisines are another example of cyclized oroidin monomers isolated from numerous sponges and containing a pyrroloazepinone system [2i]. The two (Z)-91 and (E)-91 diastereoisomers of hymenialdisine interconvert in a pH- and concentration-dependent equilibrium, with (Z)-91 being the most thermodynamically stable isomer. (Z)-Hymenialdisine [(Z)-91] shows a nanomolar kinase inhibitory activity against a wide panel of kinases [43], making synthetic analogs with this type of scaffold very attractive from a medicinal chemistry point of view. Nguyen and Tepe recently reviewed the preparation of analogs and their evaluation as kinase inhibitors [44].
247
78
N
N
O
a,b
H2N H
HN
+
N
c
H
H 2N
O
NH
Br
N
N
84: ugibohlin
Br
O
H N NH2
79
N
N
80
O
d,e,
HN
H N
Br
H N H NH2
NH2
TcesN
+
N
N
N
O
NH
Br
81
O
f
N
H N
HN
H N
Br
N
82
N
N
O
NH
Br
O
g H 2N
85: (+)-dibromoisophakellin
H2N
TcesN
N
N O
83: (+)- phakellin
N
H N
Scheme 7.17 Romo’s enantioselective synthesis of phakellin (83) and biomimetic interconversion of dibromoisophakellin (85) into ugibohlin (84). Reagents and conditions: (a) diphenylphosphoryl azide, DBU, THF, rt, 63%; (b) H2 , Pd/C, MeOH, rt, 70% overall yield for the two epimers; (c) K2 CO3 , CH3 OH, 60 ◦ C, 94%; (d) TcesN = C(Cl)SMe, NEt3 CH2 Cl2 , rt, 93%; (e) HgCl2 , HMDS, CH3 CN, rt, 77%; (f) PhI(OAc)2 , MgO, CH2 Cl2 , microwaves, 30–38%; and (g) Zn, AcOH, CH3 OH, 40 ◦ C, 80%. DBU = 1,8-diazabicycloundec-7-ene, HMDS = 1,1,1,3,3,3-hexamethyldisilazane, and Tces = 2,2,2-trichloroethoxysulfonyl.
H
HO
248
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
7.3 Biomimetic Synthesis of P-2-AI Linear and Polycyclic Monomers O
N N
TBSO
36
a,b
O
N N
O TBSO
86
c,d,e
AcO
N N
O HO
87
f
O
Cl3C
AcO H N
N N
O
g-k
O
TBSO H N
N
BocHN
N
O
BocN
88
89 l,m,n
H N
H2N
N
+
Cl− HN
N
O
90
Scheme 7.18 Nagasawa’s enantioselective synthesis of (+)-phakellin hydrochloride (90). Reagents and conditions: (a) IBX, trimethylamine N-oxide, DMSO, rt; (b) MsCl, NEt3 , CH2 Cl2 , rt, 52% overall yield; (c) NaBH4 , CeCl3 , EtOH/THF (1 : 1), 0 ◦ C; (d) Ac2 O, pyridine, CH2 Cl2 , rt, 75% overall yield; (e) HF · NEt3 , THF, rt, 92%; (f) CCl3 CN, DBU, CH2 Cl2 , 0 ◦ C to rt,
48%; (g) K2 CO3 , MeOH, rt, 96%; (h) TBSCl, NaH, THF, 0 ◦ C, 82%; (i) DIBAL-H, toluene, −80 ◦ C, 97%; (j) H2 , Raney Ni, EtOH, rt, 99%; (k) NBoc = C(SMe)NHBoc, AgOTf, NEt3 , MeCN, rt, 95%; (l) TBAF, THF, 0 ◦ C, 99%; (m) MsCl, NEt3 , CH2 Cl2 , reflux, 91%; and (n) aq. HCl, MeOH, rt, 99%. DBU = 1,8-diazabicycloundec-7-ene.
Biosynthetically, hymenialdisines could arise from the related hymenin (92) by an oxidative process. Hymenin could be in turn formed from the corresponding tautomer III of oroidin (3), allowing for the C15–C6 bond formation [1]. This strategy has been applied by Horne and coworkers [45] in the synthesis of hydantoin analogs axinohydanthoins (95). The linear precursor 93 was obtained in four steps from ornithine methyl ester, using B¨uchi’s and Horne’s methodologies outlined in the synthesis of dibromophakellin (Scheme 7.19). The biomimetic key cyclization was performed using TFA to yield the pyrroloazepinone 94. A final oxidation with three equivalents of bromine yielded axinohydantoins 95 as a 7/9 mixture of the (E/Z) isomers. While the biomimetic approaches are inspired from biosynthetic considerations, some synthetic strategies for the synthesis of (Z)-hymenialdisine rely on its degradation into bromoaldisine 96 [46]. The construction of the pyrroloazepinone system in 91 is intuitive. The introduction of the 2-aminoimidazolone could be performed using a large variety of reactants [2i]. A similar synthetic plan was used by Horne [47] for the synthesis of hymenin (92) (Scheme 7.20). By finely tuned oxidations, hymenin was then converted into (Z)-hymenialdisine (91). The pyrroloazepinone 98 was obtained from the alcohol 97 in two steps and the aminoimidazole part was then introduced in acidic medium to yield hymenin (92). Bromination of the aminoimidazole yielded 99, from which 100 was obtained after hydrolysis by heating in acetic acid. An additional oxidation in acidic medium yielded the α, β-unsaturated 2-aminoimidazolone ring of hymenialdisine with partial protodebromination. The natural products (Z)-hymenialdisine (91) and (Z)-bromohymenialdisine (101) were isolated.
249
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
250
biogenetic hypothesis HN
NH
NH
H N
2
O
2
1
HN
NH
15
Br
NH N H O (E )- 91: hymenialdisine
O
15
a
Br MeO COOMe
15
NH N H
O 93
O
Br
NH
NH
O Br
or
N H
O 94
HN
O
HN Br
b
6 15
NH N H O 3: oroidin (tautomer III)
H N
NH
NH
Br
H O 92: hymenin
6
Br
NH
ornithine methyl ester
NH
13 N
O
HN
4 steps
N H
Br
Br
O NH
HN
NH2
H2N
6
Br 14
13
NH N H O (Z )-91: hymenialdisine Br
N
NH
HN O
NH
1
HN
NH
Br
N H O (E )-95
O (Z )-95
Scheme 7.19 Biogenetic hypothesis for hymenialdisines and Horne’s synthesis of axinohydantoins. Reagents and conditions: (a) TFA, rt, 30% and (b) Br2 , 3 equiv, AcOH, AcONa, 80% for the two diastereoisomers. H N
HN
O
H biogenetic hypothesis N O Hydrolysis HN O H
HN
HN
Br
NH Br N H O (Z )-91: hymenialdisine Br
NH N H
Br
N H
NH O 97
N H
NH O 98
H N
HN
N H
NH O
92: hymenin
Scheme 7.20 Hydrolysis of hymenialdisine [(Z)-91] into bromoaldisine 96 could be foreseen in Horne’s syntheses of hymenin (92) and (Z)-hymenialdisine (91). Reagents and conditions: (a) Swern
H N
HN
Br
NH N H
O 99
f Br
N H
O
HN X
e Br
H N
HN
O
HN Br
d Br
O
HN Br
c Br
NH N H
96
HN Br
Br a,b
Br
O
non biomimetic synthesis
H N
HN HO
O
NH O 100
Br
NH N H
O 91: X = H, 101: X = Br
oxidation; (b) CH3 SO3 H, rt, 80%; (c) 2-aminoimidazole, CH3 SO3 H, rt, 65%; (d) Br2 , TFA, rt, 95%; (e) AcOH, H2 O, reflux, 72%; and (f) CH3 SO3 H, HBr cat., 90 ◦ C, sealed tube, 33% for 101 and 27% for 91.
7.3.2.5 Agelastatins Agelastatins A–F (8, 105–109) belong to the class of tetracyclic P-2-AI monomers that were isolated from Agelas dendromorpha [48] and Cymbastela sp. [49]. Agelastatins E and F were isolated recently [50] in our group from Agelas dendromorpha as well. These natural products exhibit a densely functionalized tetracyclic core (A–D) with a central cyclopentane C ring bearing four contiguous stereocenters (Scheme 7.21). The reported derivatives resulted from the variation of the bromination degree in addition to the N1 methylation and C4/C5 hydroxylation/methoxylation.
14
15
11
6
2
N3
Y
6
NH 7
12
NH
5
N
Br
N Br
HN
O
NH H
H N
X
Br
Br
Y
110: nagelamide J
NH
NH
NH
NH
N H MeO H N
O
NH
1
O 103: pre-agelastatins
H2N
X
NH
O 104
5
H 4 NH 7 N 8 H H NH
N NH
X
Br
biogenetic hypothesis
Biogenetic hypothesis for agelastatins and related derivatives.
102
4
5
8 9 NH
10
O
NH
12
Scheme 7.21
X
Y
N1
H
N
O
H NH
O
8: X = H ; R1 = H ; R2 = Me ; R3 = H
109: X = Br ; R1 = H ; R2 = H; R3 = H
108: X = H ; R1 = Me ; R2 = Me ; R3 = H
107: X = H ; R1 = H ; R2 = H ; R3 = H
106: X = H ; R1 = H ; R2 = Me ; R3 = OH
105: X = Br ; R1 = H ; R2 = Me ; R3 = H
Br
O
A N B
NH
C
HO
H
N D
8: agelastatin A
NH
O
agelastatins A-F 8 and 105–109
R3
5
H 4 NH
7 8
R R1O N2
7.3 Biomimetic Synthesis of P-2-AI Linear and Polycyclic Monomers 251
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
252
Agelastatin A (8) has been reported to be a powerful antitumoral agent [51]. These synthetically challenging bioactive molecules have been targeted by many groups in the recent past. The first synthesis of racemic agelastatin A has been reported by Weinreb and coworkers [52]. The biogenetic hypothesis formulated in our group is depicted in Scheme 7.21. Agelastatins could be formed starting from the precursor pre-agelastatin 103 by N12–C7 bond formation via a conjugated addition of the nitrogen from pyrrole on the conjugated imine N1–C5–C6–C7 leading to the tetracycle skeleton 104. A pre-agelastatin motif is clearly shown by the dimeric natural product nagelamide J (110) recently isolated by Kobayashi and coworkers [53]. The pre-agelastatin could be readily formed from oxidized oroidin (102) by C4–C8 cyclopentane formation. Importantly, the bromination pattern (on C13) in agelastatin A is very seldom found in the P-2-AIs family and could provide a particular reactivity favoring the agelastatin pathway. From a synthetic point of view, the stereocontrolled installation of four stereocenters on the cyclopentane remains the most challenging task. Non-biomimetic syntheses of the cyclopentane have been completed using various synthetic tools and strategies reviewed elsewhere [2d, i]: N-sulfinyl Diels–Alder cycloaddition (Weinreb), ring-closing metathesis from a linear precursor (Davis, Hale, Ichikawa, Chida), or functionalization of a cyclopentane mainly by allylic substitution (Trost, Tanaka, Wardrop) have been proposed. We focus now on the later stage of these syntheses. Most agelastatin syntheses reported to date [54] made use, for the late stages, of a similar biogenetically related aza-Michael conjugate addition to form the crucial N12–C7 bond. The subsequent addition of methylurea to the remaining ketone led to ring D closure with hemiaminal formation at C5. This sequence is illustrated in Scheme 7.22 for Wardrop’s synthesis of agelastatin A (8). Treatment of cyclopentene OAc
HO
NH O
CCl3
111
OAc
a,b
c,d
OAc NH
NH
PhtN O
112
N Bn
O
NH O
113
e,f
O NH
NH N Bn
O
NH O
O g
N
NH N Bn
O
114
NH O
NH N Bn
115 h,i
Br A
N
HO N O D C NH B
H NH
O 8: (+/−) agelastatin A
Scheme 7.22 Wardrop’s completion of the synthesis of (±)-agelastatin A (8). Reagents and conditions: (a) DEAD, phthalamide, PPh3 , THF, 0 ◦ C to rt; (b) Bn(Me)NH, NaHCO3 , DMF, 100 ◦ C, 62% for two steps; (c) NH2 NH2 , THF, rt;
(d) 2-pyrrolecarboxylic acid, EDC, CH2 Cl2 , rt, 85% for two steps; (e) K2 CO3 , MeOH, CH2 Cl2 , rt; (f) IBX, DMSO, rt, 91% for two steps; (g) K2 CO3 , DMSO, 100 ◦ C, 48%; and (h) H2 , Pd(OH)2 , THF, rt, 61%; (i) NBS, MeOH, THF, rt, 75%.
7.4 Biomimetic Synthesis of P-2-AIs Simple Dimers
253
111 under Mitsunobu conditions with phthalamide followed by the trichloromethyl substitution with N-methyl-N-benzylamine provided the derivative 112 with N9 and the urea moiety in position. Phthalamide removal with hydrazine followed by acylation with 2-pyrrolecarboxylic acid led to compound 113. After methanolysis of the acetate ester, oxidation of the resulting allylic alcohol with o-iodoxybenzoic acid (IBX) yielded the cyclopentenone 114. The next biogenetically inspired step was the key N12–C7 bond formation by conjugate aza-Michael addition of the nitrogen from pyrrole onto the cyclopentenone ring, performed with potassium carbonate in DMSO. Hydrogenolysis of the N-benzyl group in ketone 115 using Pearlman’s catalyst afforded spontaneously the annulation of the N,N -disubstituted urea. Further regioselective bromination by N-bromosuccinimide yields racemic agelastatin A (8). Notably, Feldman proposed a similar synthesis in 2002 [55]. 7.4 Biomimetic Synthesis of P-2-AIs Simple Dimers 7.4.1 Mauritiamine
Mauritiamine (116) isolated by Fusetani [56] from Agelas mauritiana is an oxidative dimerization product of oroidin. Fusetani and coworkers proposed an oxidation–dehydration process prior to the oroidin dimerization connecting C4–C5 . The formation of mauritiamine fits into the general chemical pathway involving tautomers I–V (Figure 7.1) suggested by Al-Mourabit and Potier [1]. The first total synthesis of mauritiamine reported by Horne seems in accordance with the biomimetic suggestions [23b]. The strategy involving heterodimerization of intermediates 117 and 118 was applied (Scheme 7.23). H2N N H N H2N
N
N
Br
H H N
4
NH
5
O
H 5 17 1N H 2N N 4
NH O
1N
Br
O
H2N
HN
116: mauritiamine
Br
biogenetic hypothesis Br
N
Br
O 7
N H
I 7
Br
N
N H N
Br
H N
N H
5 4
O
1 H
H2N
H N
H2N
Br
Br
H N
4
NH
6
NH O Br
O
N
HN
IV
120: nagelamide D
Br
Br
NH2
N H2N
N H
a
H 2N
F3CCOO H2N N
H 2N H2N
N H
Cl +
H2N
N H2N
b N H
OCOCF3
H N
117
118
NH2
Scheme 7.23 Horne’s synthesis of C4–C5 connected mauritiamine (116) and the structure of the C4–C6 connected nagelamide D (120). Reagents and conditions: (a) NCS, TFA, rt; (b) MeOH–xylene, 135 ◦ C, 23% for two steps; and (c) 4,5-dibromo-2-(trichloroacetylpyrrole), DMF, rt, 65%.
O N
NH Br
NH
HN
HN
119
NH
HN
N
c
N
H2N
13
O H 2N
H2N
N
O
O
HN
116: mauritiamine
Br
Br Br
254
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
The bis-imidazole core 119 of mauritiamine (116) was obtained in 23% yield from aminovinyl-2-aminoimidazole 13 and was acylated twice with 4,5-dibromo-2-(trichloroacetyl)pyrrole to obtain the natural product in 65% yield. Although several natural P-2-AI dimers with one C–C connection were isolated [57], their synthesis does not seem experimentally obvious. It is worth mentioning here the total synthesis of the C4–C6 connected nagelamide D (120) reported by Lovely [58]. The authors indicated that the NMR of the synthetic compound did not match that of the natural product. The question of the existence of nagelamide D as a natural product is thus still open. 7.4.2 Sceptrins, Ageliferins, and Oxysceptrins
Sceptrin (9) [59] and ageliferin (121) are formally made up of two hymenidin (2) subunits. The biogenetic hypothesis formulating the formation of sceptrin from hymenidin (2) via [2 + 2] cycloaddition was proposed by Faulkner. The authors mentioned, though, some unfruitful [2 + 2] cycloaddition attempts from oroidin. A non-concerted and unified mechanism was later proposed by Rinehart for sceptrin (9) and ageliferin (121) [60] instead of independent [2 + 2] or [4 + 2] cycloaddition, respectively. A general and common chemical pathway was put forward for sceptrin (9), ageliferin (121), and palau’amine (11) congeners in 2001 by Al-Mourabit and Potier [1]. The dual reactivity introduced above, involving tautomers I–V, was proposed to govern the corresponding nucleophilic/electrophilic behavior of the monomers leading to multifaceted intermediates (Scheme 7.24). The first step engaging tautomers II and IV would lead to the key connection between the nucleophilic C7 of II and the electrophilic analogous carbon C7 of IV. The multifaceted intermediate 122 displayed the appropriate structure for conversion into sceptrin (9) or ageliferin (121) through two modes of cyclizations. 2: hymenidin NH H N
Br
II Br
IV
O NH H N
5
3 2 NH2
7 6
4 N
7
RHN
tautomerism RHN
N NH2
N NH2 7
H2N H+ N N
H2N H+ N N
H N1
RHN
7
N
N H
122
O
O Br NH Br
NH HN
H N N H
O
9: sceptrin
H2N
RHN
N
NH2
N
RHN
H
N
a) Baran shunt N NH2 N H
NH2 N H 124: biradical intermediate
Br Br N H
O N
HN H N O
NH NH
Br
O Br
H N NH2 N
121: ageliferin
Scheme 7.24 Common biosynthetic pathway proposed for sceptrin/ageliferin and the alternative Baran’s radical shunt. Reagents and conditions: (a) H2 O, 200 ◦ C, microwave, 40%.
7
H2N
H2N NH
N
7
RHN
multifaceted intermediates 123 NH2
NH
N
N H
HN H N
N
NH H N NH2 N
O
125: nagelamide E
7.5 Biomimetic Synthesis of Complex Dimers: Palau’amine and Related Congeners
255
The analysis shows that the scheme is extremely dynamic and that the interconversions of sceptrin and ageliferin are conceivable. The conversion of sceptrin (9) into ageliferin (121) has been realized by Baran [5]. The rearrangement under microwave conditions was compatible with a radical or ionic mechanism. The authors proposed a radical shunt that transformed the vinyl cyclobutane of sceptrin into the cyclohexane ring of ageliferin. Biradical intermediate 124 [61] was put forward to explain the formation of nagelamide E (125) (epi-ageliferin). In contrast to the universal pathway starting from the same linear precursor via multifaceted intermediates, the [1,3]-sigmatropic rearrangement of sceptrin eludes the question of what is the real link between monomers like hymenidin (2) and dimers. In other words, is the first dimerization event dissociable from the ionic 122 or the radical 124 intermediates? The drastic conditions used by Baran to turn back from sceptrin to the putative intermediate 124 indicate a high activation barrier required for the reverse transformation. The direct access to the multifaceted intermediate 122 through the formation of the first C7–C7 bond indicates the likely central role of the first dimerization event for the subsequent transformations. A series of oxysceptrins (127) could also derive from sceptrin by simple oxidation [60]. Exposure of sceptrin (9) to peracetic acid followed by an acidic treatment provided 127 [62], while the over oxidized natural product nakamuric acid (128) was obtained by oxidative cleavage of the dihydroxy intermediate 126 (Scheme 7.25). H N
O NH
H N
H N
NH2 N
Br Br
a
N H
O 9: sceptrin
N H
NH
H N
H N
NH2 N
Br Br
N NH
O
NH2 N H
O
O
H N
H N
NH2
NH b
N HO NH NH2 HO N H
126
N
Br Br N H
O
N H
127: oxysceptrin
NH
H N
NH2 N
Br Br
N NH O
O
CO2H NH
NH2 N H
O
128: nakamuric acid
c
Scheme 7.25 Baran’s synthesis of oxysceptrin (127) and nakamuric acid (128). Reagents and conditions: (a) AcO3 H, H2 O, 50%; (b) AcOH, 140 ◦ C, 65%; and (c) NaIO4 , AcOH–NaOAc buffer, 60%.
7.5 Biomimetic Synthesis of Complex Dimers: Palau’amine and Related Congeners
Palau’amine (11) is the most attractive member of this family of compounds since it displays an extremely complex architecture with six fused heterocycles and eight contiguous stereocenters. Furthermore, this compound shows an impressive nanomolar immunosuppressive activity [63]. Its synthesis recently by the group of Baran [30], partially based on previous biosynthetic proposals, proved that biomechanistic speculations can provide valuable approaches not only for biosynthetic experiments but also for synthetic strategies.
256
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
7.5.1 Common Chemical Pathway for P-2-AI Biosynthesis
Unlike other classes of alkaloids, the P-2-AI alkaloids show an unprecedented use of multifaceted intermediates that provide a high degree of complexity and molecular diversity. Complex P-2-AI dimers like palau’amine (11) can be mapped back to two C11 N5 clathrodin (1) subunits linked by several C–C and C–N bonds. The initial proposal for the structure (11’ Figure 7.2) of palau’amine was made in 1993 by Scheuer and coworkers. Its structure was revised in 2007 by K¨ock [4a] and Quinn [64] by NMR spectral means. This revision deals with the relative configuration of C6 and C7 stereocenters and made uniform the relative configuration around the cyclopentane core of complex P-2-AI dimers. The homogenization of the stereochemistry extended the list of arguments to a universal biochemical pathway for the entire family [1]. The correction into the highly strained trans-azabicyclo[3.3.0]octane subunit was not obvious, as trans-fused bicyclo[3.3.0]octanes are energetically disfavored. The extreme chemical complexity of palau’amine (11) and congeners axinellamines A (133) and B (134) and massadine (135) makes biomimetic approaches very attractive to design a common synthetic plan for these related alkaloids and to minimize the number of steps. As shown below, the fully substituted cyclopentane core E of palau’amine (with C5 bearing the spiroaminoimidazole ring F) (Figure 7.3) is a common feature of this family of compounds. The A-B-C subunit corresponds to phakellin. Thus, the reactivity generating the phakellin subunit should proceed along the same lines. From an oroidin-like precursor, the key bonds between the two oroidin subunits are C7–C7 and C5–C6 . NH
Cl HN H2N
H
H
O
NH
6
H2N
7'
N
NH
Cl
NH H OH NH N
H N
NH N H
HN
O
Br
NH
H N
Br
NH N H
NH
N H H
7'
H
H OH NH N
NH
Cl
O
HN
6
O
H OH N NH NH N H Br X
11′: palau'amine initial structure
11: palau'amine revised structure NH
Cl
NH H H N O
Br
HN Br
131: styloguanidine
HN NH
N H H
Br H OH N NH NH H
NH
Cl O
HN H2N
129: X = H konbu'acidin A 130: X = Br konbu'acidin B
NH
H
H OH N NH
N
Br
NH H
O
Br
HN Br
132: tetrabromostyloguanidine
Figure 7.2 Palau’amine: previous (11 ) and revised (11) structures and congeners (129–132).
7.5 Biomimetic Synthesis of Complex Dimers: Palau’amine and Related Congeners O
spiroimidazole
NH
Cl
H2N
HN F E
H H
pentasubstituted cyclopentane
N
H N
D
Br
NH OH NH
NH2 Br
N H
O
11
133: axinellamine A
HO
O
NH Cl 7
Br
N
NH2 N
7'
H NH HO N
Br NH
H NH
134: axinellamine B
Figure 7.3
7
N H
Br
NH
N 5
7' 6'
O
N
NH
HO NH
Br
O
HO
Br
NH2 N H
NH
NH2
NH
Br
H
6'
O Br
Br
NH2 N
H NH HO N
Br
A
phakellin moiety
5 7'
NH N H
B
7
Br
C
O
HO N NH Cl NH
NH
O
NH2
135: massadine
Structure of palau’amine (11) and related dimers.
Interestingly, the C7–C7 bond is conserved in all complex dimers. We will now discuss different biogenetic hypotheses proposed for palau’amine and congeners. 7.5.2 First Proposal Based on a Diels–Alder Key Step
Scheuer and Kinnel’s biosynthetic proposal for palau’amine (Scheme 7.26) was based on the previous structure of palau’amine (11): the key step proposed by these authors was a Diels–Alder cycloaddition with aminovinylimidazole (13) as dienophile and a dehydrophakellin of type 137 as diene [65]. The cycloadduct (138) with H6 /H7 in cis relation would be obtained. An additional oxidative ring contraction mediated by chlorohydroxylation could then lead to palau’amine (11). This hypothesis was weakened by the structural revision since in the revised structure H6 /H7 are trans and ring contractions usually proceed with retention of configuration. The isolation of konbu’acidins (129) as precursors of palau’amine has suggested another route with clathrodin (1) as the dienophile. Palau’amine would then be obtained through amide hydrolysis of the konbu’acidin like compound 142. Although the dehydrophakellin 137 was not isolated from sponges, the isolation of the phakellin oxidized product 136 was reported quite recently [66]. 7.5.3 Universal Chemical Pathway
A universal chemical pathway for the formation of more complex, dimeric, polycyclic P-2-AI members known up to 2000, involving the dual reactivity of the 2-aminoimidazole was proposed by Al-Mourabit and Potier [1]. The presence of
257
Cl
HN
H N
N
N
N
H N
O
Br
O R=
NH
O
'cis-ring juncture'
4+2
R=H
NH
O
O
N
H NH
N H
NH
NH N H
H N
H N
7'
N
Cl
+
N
6'
N
H
H N
NH
+
+
H2O
Cl
Cl H O NH2 2
N H
NH
138 H-6'/H-7' cis
7
H
O
H
+
140 H-6'/H-7' cis
N H
H2N
Cl
O
H
NH
O
H2N
Scheuer and Kinnel’s biogenetic proposal for palau’amine.
NH
R = H : 13 R= O 1
N
N
137
136
R NH
HN
Scheme 7.26
H2N
HN
H N
'dehydrophakellin'
HO
HN
Br
N N
N
6' H
NH
OH2
NH2
NH
NH
7'
N N
N
6' H
NH
OH2 NH2
NH
NH
141 H-6'/H-7' cis
O
N H H
Cl HN
139 H-6'/H-7' trans
7'
Cl HN
H2O
H2O
7'
N
H
6
Cl
H OH NH N
NH
NH
NH N H
6'
HN
O
7'
H N
NH H OH NH N NH N H
6'
NH
142 H-6'/H-7' cis
O
N H H
Cl HN
amide hydrolysis
11 H-6'/H-7' trans
O
H
NH
H2N
7
258
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
7.5 Biomimetic Synthesis of Complex Dimers: Palau’amine and Related Congeners
similar structural motifs and functionalities in several P-2-AIs regardless of their monomeric or dimeric nature suggests common biosynthetic transformations leading to their formation. Variations of Al-Mourabit’s postulated biosynthetic intermediates were further deepened by Baran and K¨ock [2c]. The C7–C7 multifaceted intermediate A 143 could follow many plausible modes of cyclization, including the formation of the second multifaceted intermediate B 145 (Scheme 7.27). The putative common precursor 145 towards several higher order P-2-AIs would provide the observed trans relative stereochemistry in the [3.3.0] bicyclic core common to all higher order P-2-AIs, including palau’amine (11), axinellamines (148), massadine chloride (149), and styloguanidine (150). Interestingly, the macrophakellin (151) and macroisophakellin (152) subunits were found in the palau’amines and the regioisomeric styloguanidines that could arise from macrocyclic fleeting intermediates ‘‘macropalau’amine’’ 146 and ‘‘macrostyloguanidine’’ 147, respectively (see also Scheme 7.33 below). The macrocyclic intermediates could then collapse in a transannular fashion, generating the strained trans-azabicyclo[3.3.0]octane core of these alkaloids. Variations of Al-Mourabit’s postulated biosynthetic intermediate 145 were exploited by Baran for the synthesis of axinellamine A and B [67], massadine [68], and palau’amine [30]. 7.5.4 Intramolecular Aziridinium Mediated Mechanism for the Formation of Massadine (141) from Massadine Chloride (155)
Displacement of the chlorine by a hydroxyl group at C6 through an aziridinium intermediate was investigated by Romo and coworkers [69]. The aziridinium mechanism is an interesting way to explain the conversion of massadine chloride (149) into massadine (135) with retention of configuration (Scheme 7.28). During synthetic studies toward palau’amine and congeners, the Romo group indeed reported an unexpected conversion of cyclopentyl chloride 154 into its derivative 155 in the presence of azide, which surprisingly proceeded with retention of configuration (Scheme 7.28). This group concluded that a mechanism involving neighboring group participation of the spiroaminoimidazole might be in operation through the aziridinium species 153. Furthermore, it was proposed that a similar process could account for the biosynthesis of massadine 135 from an earlier chlorinated precursor (149). 7.5.5 Aziridinium Mechanism for the Formation of the Tetramer Stylissadine A
In 2007 Baran and K¨ock [2c] provided support for the pathway proposed by Romo, by isolating the unprecedented massadine chloride (149), and reported its rapid conversion into massadine in an aqueous medium at 60 ◦ C. This behavior was thought to be due to the favorable geometrical disposition between the aminoimidazole nitrogen and the chlorine atom. Such a mechanism was proposed for the dimerization of massadine chloride (149) into stylissadine A (2) (Scheme 7.29).
259
X
Y
N H
N H
N H
O
O
NH
7'
7
H N
143
N
N
HN
NH
NH3
Cl+
O
H
N
OH2
X
Y N H
NH
N H
O
O
NH2 O
N
NH
HN
N
H N
Cl
NH
H
N
N
OH
NH2
NH
NH
NH N H
NH
144
7'
7
N
HN
O
macroisophakellin subunit
X
152
Y
151
N H
N H
N
H2O
Cl
N
NH2
H N
H2N
HC
O
O
HO
Y
NH
H NH
7'
Cl 7
N
X
146
NH
NH2
NH
145
N
NH H N
N H
NH2
NH NH OH N
H N
HN
O
7'
H
HN
NH
7
OH NH
NH
NH
NH
147
H N
Cl HN
NH NH
NH N H
N N H HO N
6 7'
HO
RHN
RHN
7
Cl
N
H HO N
6 7'
H2N
H
NH2
NH
NH2
H NH
O
NH
axinellamines (148)
RHN
7
Cl
NH2
H OH NH N
HN
O
H
HN
N
H
7'
NH
NH
NH H
H OH N NH
Cl HN
styloguanidine (150)
H2N
7
massadine chloride (149)
O oxidative cyclization & hydrolysis
N oxidative cyclization
7'
Cl
H N
7
palau'amine (11)
O
H
RHN
H2N amide N & acylpyrrole N bis-cyclization
amide N & acylpyrrole C bis-cyclization
2nd multifaceted C7-C7' intermediate B
RHN
H2N
6
Cl
Scheme 7.27 Al-Mourabit’s unified biogenetic proposal for palau’amine and congeners [1] and further developments of Baran and K¨ock [2c].
1st multifaceted C7-C7' intermediate A
X
Y
HN
O
NH
N
macrophakellin subunit
260
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
7.5 Biomimetic Synthesis of Complex Dimers: Palau’amine and Related Congeners
261
:
biogenetic hypothesis NH2
NH2 O
Cl
NH
O
:
H2 O
NH
N
N H
Br
−Cl
NH
RHN
HO
Br NH
Br
O
Br
aziridinium intermediate
massadine chloride 149
TIPSO Ts N
O
O
NH2
O
massadine (135) Ts N
O
O
O R
a R
Bn X
NH
NH
N
TIPSO
Ts N
N
Hb Ha
O
154: X = Cl, R = OTs
R
17
BnN
17
N
NH
O
O BnN
O HO
Br
NH2
NH
N
N H
RHN
Br
TIPSO
HO
Br
NH
N
O
NH
Br
NH
N
_
O
NH2
Hb X Ha
N
Bn
153
17
BnN
Hb Ha
O N3
Bn
155: X = N3, R = N3
Scheme 7.28 Biosynthesis proposal for massadine: Romo’s aziridinium intermediate suggested for the displacement of chloride with retention of configuration. Reagents and conditions: (a) NaN3 , DMF, 105–120 ◦ C. Br
Br
NH
HN O
Br
Br
2 X 149
Br
O
H HN N N H
H2N
N H
Scheme 7.29
N O
N H
NH2
H2N
OH O N
N
N H
O
N H
N H NH2
Br
Br NH
Br
H HN N N H H2N
O HO N N H
O
N H
NH2 H2N
Br
NH H N
O N
O
Br
HN O
Br
Br
NH H N
H2O
O HO N
Br Br
Br
N
N H O stylissadine A (2)
OH O N N H
Baran and K¨ock’s proposed formation of stylissadine A from massadine.
7.5.6 Synthetic Achievements
Inspired by the reactivities showed in these biogenetic proposals, several groups have proposed synthetic approaches for complex P-2AI dimers. Obviously, the most accomplished synthetic work that took advantage from these hypotheses was reported by Baran and coworkers. Using very effective preparative organic chemistry, including new and chemoselective reaction discoveries, this group completed the first total synthesis of complex dimers using late biomimetic steps [30, 68]. Scheme 7.30 depicts the synthesis of the common precursor. The Diels–Alder cycloadduct 159 was converted into the diazide 160 in six steps; the azide groups played the role of the masked amines needed for the introduction of pyrrolecarboxamides. Five additional steps, involving a key intramolecular aldol
N H NH2
Br
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
262
+ 157
CO2Me
158
COOMe
Me
N3 N3
Br
160
N3 Cl
162
N3
NH
HO
N H
N
O
NH2
4 steps
Palau'amine path
N3 N3
3 steps
Axinellamines path
N3 N3
H2N
Cl HN
NHBoc Br
Br Br
H N O
Br
2 steps
HN
H OH NH N NH N H 11
N3 N3
H2N
OH
N
Br 4 steps
HO
N
H HO N O
NH Br
H NH NH
NH O
HN Br Br
NH 166: X = Cl 167: X = OH
NH2
N
H HN HO HN O NH H2N 147
N H
HN
NH H HN
Cl
O
NH2
X
NBoc2 OH
N H
HN
H O N CHO OHC
NH H
N
O
NBoc
N
Massadine path
NBoc
N H
4 steps
165 NH
NBoc
BocN
164 3 steps 1 pot
Cl
163
2 steps Cl
HN NH OH
Cl
H O
NH2
N3 Cl
NHBoc
N
5 steps
PMBO
COOMe OTIPS 159 Cl NBoc
5 steps
161 O
O
6 steps
Cl N3
N3
O
OTIPS Me
MeO2C
H NH NH2
135
Scheme 7.30 Baran’s syntheses of precursors for axinellamines A/B, massadine (135) and palau’amine (11).
reaction, were needed to build the cyclopentene 161, which was functionalized to obtain the Boc-protected spiroimidazole 162 in five steps via a key stereoselective oxidation by IBX. Compound 162 shows the fully substituted cyclopentane core of palau’amine (11) and congeners with the right relative configurations of the five stereocenters. This compound – obtained in multi-gram quantity via scalable procedures at a relative early stage of the synthesis – was used as a common intermediate toward the synthesis of axinellamines, massadine, and palau’amine. In the axinellamines path, the introduction of a second aminoimidazole yielded bis-aminoimidazole 163 in three steps while in the palau’amine and massadine paths the bis-formamide 164 was first prepared. The biomimetic reactions of these syntheses are located in the last steps, which are detailed in the next sections. 7.5.6.1 Axinellamines A/B The crucial step in the synthesis of axinellamines was inspired from biomechanistic analysis (Scheme 7.31). N-Boc removal starting from 163 yielded bis-aminoimidazole 166. The oxidation of the C4 =C5 imidazolic double bond by DMDO led to the N1–C4 linkage and to the introduction of the hydroxyl group on
7.5 Biomimetic Synthesis of Complex Dimers: Palau’amine and Related Congeners
2nd multifaceted C7-C7′intermediate B
biogenetic hypothesis
Cl HO H N 7
RHN
7′
H NH
Y
NH H N
NH2
X
N3
N
N3
H BocN
NBoc N 3 H
N3 NBoc2
163
N 1 NH2 b H 4′
a
N
Cl
− NH TFA
NBoc
5′
NH TFA
HN 166
NH2
−
4
N3
H
NH2
c
NH2
HO NH TFA− NH2 H
N H
NH
HN
168
NH
N3
H
N3 HO
H
Cl
NH TFA− N
NH2
N
H HN HO HN O NH 170
Br
Cl
Cl
NH Br Br
O
OH NH
N H
Br
N N H 145
Cl
O
NH
TFA
NH
N3 HO
−
263
HN
NH2
TFA−
NH2
169 d
Cl
O N H
HN Br
Br
NH
NH2
N
H HN HO HN O
Br
170
Br
Scheme 7.31 Biogenetic hypothesis and Baran’s final biomimetic steps for the synthesis of axinellamines (170). Reagents and conditions: (a) TFA–CH2 Cl2 (2 : 1), 100%; (b) DMDO, H2 O then TFA-CH2 Cl2 (2 : 1);
OH NH
(c) silver(II) picolinate, H2 O; 40% for the two diastereoisomers over two steps; and (d) 1,3-propanedithiol, triethylamine, MeOH then 4,5-dibromo-2-trichloroacetylpyrrole, diisopropylamine, DMF, 69%.
C5 of 168. Two diastereoisomers with cis fusion were obtained. The highly selective silver(II) picolinate-mediated oxidation at the C4 position of the unprotected spirocyclic core (168) yielded the hemiaminal 169. Azide reduction in presence of 1,3-propanedithiol followed by acylation with 4,5-dibromo-2-trichloroacetylpyrrole yielded the two diastereoisomers axinellamines A and B (170) as TFA salts. 7.5.6.2 Massadine Chloride (149) and Massadine (135) The total syntheses of massadine chloride (149) and massadine (135) were achieved by the Baran’s group using the same methodology as for the constitutional isomeric axinellamines. The bis-formamide 164 (Scheme 7.32) was hydroxylated on C4 by silver(II) picolinate in the presence of TFA to deliver deformylated ammonium 171. The installation of the second 2-aminoimidazole ring was made by the Lancini’s cyanamide methodology [20]. This proceeded with partial chlorine displacement by a hydroxyl group on C6 with retention of configuration, as
H NH NH2
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
264
HO Cl RHN
7
NH 7′
NH H N
H NH
Y
N3
Br
N3
H O OHC
N CHO
NH2 135
Br
6
a
O NH
145
NH2
HO HN
6 HN
NH 4
NH2
NH2
HO
HN
N3
H NH
H HO N
Br Br
NBoc N
O
HN
Cl
NHBoc
NH
N
N H
HN NH2
X
Cl
HO
O
N N O H 2nd multifaceted C7-C7′ intermediate B
H2N
biogenetic hypothesis
H N
OH
H O + NH3 2 CF3CO2−
NH
N3
4
b
N3
4
N3
OH c,d HO N3 HN
N3 HN NH H2N
164
171
NH
166
173
O
5′ 4′
H NH
NH2 e,f
H2N HO
O N H
HN Br
HN
O
Br NH Br
NH O
H HO
Br
Scheme 7.32 Biogenetic hypothesis and Baran’s final biomimetic steps for the synthesis of massadine (135). Reagents and conditions: (a) silver(II) picolinate, H2 O–TFA then TFA, 84%; (b) cyanamide, NaOH to
N
N
H NH NH2
135
pH 5, 24%; (c) DMDO, H2 O–TFA; (d) TFA, 65% over two steps; (e) PtO2 , H2 ; and (f) 4,5-dibromo-2-trichloroacetylpyrrole, diisopropylamine, DMF, 40% over two steps.
proposed by Romo [69]. The diol 166 was then transformed into compound 173 with the ether bond C4–O–C4 of massadine through oxidation of the C4 =C5 double bond by DMDO followed by addition of TFA. Control of the pH was required to favor the regioselective cyclization versus the axinellamine cyclization mode. Surprisingly, the configuration of the major diastereomer 173 at C4 –C5 corresponds to that of epi-massadine. The authors have postulated that epi-massadine should exist in nature, as for axinellamines A and B isomers. Starting from 173, final azide hydrogenations and acylation of the resulting amines with 4,5-dibromo-2-trichloroacetylpyrrole yielded massadine (135). The synthesis of massadine chloride was completed starting from the C6 chloro-analog of 166 by the same procedures.
7.5 Biomimetic Synthesis of Complex Dimers: Palau’amine and Related Congeners
265
7.5.6.3 Palau’amine (11) Based on the latter strategy, Baran and coworkers reported the first total synthesis of palau’amine (11) in 2010 [30]. Scheme 7.33 describes the final biomimetic steps. The strained trans-fused 5,5-ring system was obtained from the macrocyclic intermediate 146, following the transannular attack that was put forward by Al-Mourabit and Potier for the formation of the phakellin subunit. The strategy provided evidence for the chemical pathway in which the ‘‘phakellin’’ substructure was reached through a transannular cyclization of a ‘‘macropalau’amine’’ precursor 146. After the elegant preparation of the intermediate 175 from 164, the ambivalent reactivity of the 2-aminoimidazole part allowed the access to palau’amine (11).
HO Cl RHN
NH X
NH H NH
O
biogenetic hypothesis Cl HN Phakellin annulation
RHN
NH2
H
amide hydrolysis
O
144 X 2nd multifaceted C7-C7′ intermediate B
NH
7′
H N
NH
Y
NH
7
N
Y X, Y = H, Br
NH2
NH H N
7′
O R=
H N
7
H OH NH N
7
H2 N
H N
NH N H X
172
NH
7′
H
NH
Cl HN
O
H OH N NH NH N H
palau'amine (11)
Y
OMe
BuOtOC OMe OMe
maropalau'amine strategy
pyrrole surrogate: 176 Cl NBoc N3 N3
N
NHBoc a-c
H O
Cl
4
OHC 164
Br
4′
HN
OH NH NH
174
NH2
Cl HN
N3
NH
N3 N3
N CHO
HN
NH2
d,e
HO O
N3 N
7
H2N NH non OH biomimetic NH
N H
NH2
7′
146
7
H
Cl HN 7′
H N O
NH NH
H OH NH N NH N H
palau'amine (11)
Scheme 7.33 Baran’s palau’amine synthesis with the final likely biomimetic macropalau’amine formation. Reagents and conditions: (a) H2 O–TFA then silver(II) picolinate; (b) cyanamide, brine, pH 5, 64%
NH
NH OH H 5′ N NH 9′ NH NH O N
175
H2 N
Cl HN
for two steps; (c) TFAA–TFA then Br2 , 54%; (d) AcOH, 176, THF then TFA–CH2 Cl2 TFA, 44%; and (e) Pt(OAc)2 , H2 , TFA–H2 O then EDC–HOBt–DMF then TFA, 17% one pot.
266
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids
7.6 New Challenging P-2-AI Synthetic Targets and Perspectives
Many efforts were made to clarify the chemical pathways toward complex P-2-AI metabolites, but the factors controlling the fleeting reactivity of P-2-AI alkaloids are still rather obscure. Several interesting and challenging molecules are currently under chemical and biological studies. Recent isolation of novel P-2-AI dimeric members, benzosceptrin A (10) [3] and stylissazole C (177) [70], demonstrates the incredible potential of P-2-AI metabolites to generate original molecules and rare organic architectures. While benzosceptrin A shows a highly strained benzocyclobutane, stylissazole C is one of the first example of dimerization involving exclusively N–C bond formations (Scheme 7.34). These members add another dimension to the molecular diversity of P-2-AI metabolites and further highlight the unique dual reactivity of the vinylogous 2-aminoimidazole clathrodin (1), hymenidin (2), and oroidin (3) precursors. N H2 N N H
O HH HN
N
H N
N H O
X Br
NH2
12
NH
oxidation cyclization
Y
11
HN
N
1N
O 7
oxidation NH cyclization
HN
N H H N
4
O NH N H O stylissazole C (177)
Scheme 7.34
X, Y = H, Br: clathrodin (1) hymenidin (2) and oroidin (3)
NH2 HN
O
HN
HN
+
NH
+ NH
NH2
benzosceptrin A (10)
New challenging structures: benzosceptrin A and stylissazole C.
Biomimetic synthesis of P-2-AI metabolites by various groups led to better understanding of the reactivity and the potential biosynthetic early precursors. The lack of biosynthetic studies by labeled precursors underpins the crucial role of research using the combination of isolation and synthesis to construct the chemical pathway puzzle. The isolation of new intermediates helps to explore the ability of the corresponding phylogenetically related sponges to create new molecules. The utility of biosynthetic speculations inspired from the reactivity of the molecules for understanding both synthetic and living systems seems obvious. The molecules identified in living matter result from complex processes that became understandable through the discovery of the impressive molecular diversity and reasoned methods of organic chemistry. In future, essential insights into the actual P-2-AI biosynthetic machinery are expected to give decisive information on the enzymes involved. The generation of the presumed universal precursors like clathrodin (1) and their conversion into higher order P-2-AIs like palau’amine (11) or benzosceptrin (10) seems very subtle. The dynamic hydrogen-bonding interaction induced by the duality of the 2-aminoimidazole is probably an interesting biochemical catalytic system still to be discovered. Synthetic plans inspired by biosynthetic hypothesis need straightforward and highly chemoselective reactions. In several cases of natural product
References
families, the chemical correlations/transformations following biomechanistic inspiration have demonstrated their high value for effective biomimetic and total synthesis.
References 1. Al-Mourabit, A. and Potier, P. (2001) 2.
3.
4.
5.
6.
Eur. J. Org. Chem., 237–243. (a) Hoffmann, H. and Lindel, T. (2003) Synthesis, 1753–1783; (b) Jacquot, D.E.N. and Lindel, T. (2005) Curr. Org. Chem., 9, 1551–1565; (c) K¨ock, M., Grube, A., Seiple, I.B., and Baran, P.S. (2007) Angew. Chem. Int. Ed., 46, 6586–6594; (d) Weinreb, S.M. (2007) Nat. Prod. Rep., 24, 931–948; (e) Arndt, H.D. and Riedrich, M. (2008) Angew. Chem. Int. Ed., 47, 4785–4788; (f) Aiello, A., Fattorusso, E., Menna, M., and Taglialatela–Scafati, O. (2008) in Modern Alkaloids: Structure, Isolation, Synthesis and Biology (eds E. Fatorusso and O. Taglialatela–Scafati), Wiley-VCH Verlag GmbH, Weinheim, pp. 271–304; (g) Feldman, K.S. and Fodor, M.D. (2009) Synthesis, 3162–3173; (h) Heasley, B. (2009) Eur. J. Org. Chem., 1477–1489; (i) Forte, B., Malgesini, B., Piutti, C., Quartieri, F., Scolaro, A., and Papeo, G. (2009) Mar. Drugs, 7, 705–753. See Appenzeller, J., Tilvi, S., El Bitar, H., Martin, M.T., Gallard, J.F., Tran Huu Dau, E., Debitus, C., Laurent, D., Moriou, C., and Al-Mourabit, A. (2009) Org. Lett., 11, 4874–4877, and references therein. (a) Grube, A. and K¨ock, M. (2007) Angew. Chem. Int. Ed., 46, 2320–2324; (b) Kobayashi, H., Kitamura, K., Nagai, K., Nakao, Y., Fusetani, N., van Soest, R.W.M., and Matsunaga, S. (2007) Tetrahedron, 48, 2127–2129; (c) Buchanan, M.S., Carroll, A.R., and Quinn, R.J. (2007) Tetrahedron Lett., 48, 4573–4574. Baran, P.S., Zografos, A.L., and O’Malley, D.P. (2004) Angew. Chem. Int. Ed., 43, 2674–2677. Wei, Y. and Zipse, H. (2008) Eur. J. Org. Chem., 47, 3811–3816, and references cited therein.
7. Kitagawa, I., Kobayashi, M., Kitanaka,
8.
9.
10.
11.
12.
13.
14.
15. 16.
17.
18. 19.
20.
K., Kido, K., and Kyogoku, Y. (1983) Chem. Pharm. Bull., 31, 2321–2328. Braeckman, J.-C., Daloze, D., Stoller, C., and van Soest, R.W.M. (1992) Biochem. Syst. Ecol., 20, 417–431. Linington, R.G., Williams, D.E., Tahir, A., van Soest, R., and Andersen, R.J. (2003) Org. Lett., 5, 2735–2738. Andrade, P., Willoughby, R., Pomponi, S.A., and Kerr, R.G. (1999) Tetrahedron Lett., 40, 4775–4778. Wang, Y.G., Morinaka, B.I., Reyes, J.C.P., Wolff, J.J., and Molinski, T.F. (2010) J. Nat. Prod., 73, 428–735. Lindel, T., Hochg¨urtel, M., Assmann, M., and K¨ock, M. (2000) J. Nat. Prod., 63, 1566–1569. Assmann, M., Lichte, E., van Soest, R.W.M., and K¨ock, M. (1999) Org. Lett., 1, 455–457. Vergne, C., Boury-Esnault, N., Perez, T., Martin, M.-T., Adeline, M.-T., Tran Huu Dau, E., and Al-Mourabit, A. (2006) Org. Lett., 8, 2421–2424. Travert, N. and Al-Mourabit, A. (2004) J. Am. Chem. Soc., 126, 10252–10253. Vergne, C., Appenzeller, J., Ratinaud, C., Martin, M.T., Debitus, C., Zaparucha, A., and Al-Mourabit, A. (2008) Org. Lett., 10, 493–496. Braun, M., B¨uchi, G., and Bushey, D.F. (1978) J. Am. Chem. Soc., 100, 4208–4213. Foley, L.H. and B¨uchi, G.H. (1981) J. Am. Soc. Chem., 104, 1776–1777. (a) Sharma, G.M. and Burholder, P.R. (1971) J. Chem. Soc., Chem. Commun., 151–152; (b) Sharma, G.M. and Magdoff-Fairshild, B. (1977) J. Org. Chem., 42, 4118–4124. (a) Lancini, G.C. and Lazzari, E. (1966) J. Antibiot., 3, 152–154; (b) Lancini, G.C., Lazzari, E., Arioli, V., and Bellani, P. (1969) J. Med. Chem., 12, 775–780.
267
268
7 Biomimetic Synthesis of Marine Pyrrole-2-Aminoimidazole and Guanidinium Alkaloids 21. Akabori, S. (1933) Chem. Ber., 66, 22. 23.
24.
25.
26.
27.
28.
29. 30.
31. 32.
33. 34.
35.
36. 37.
38.
151–158. Nakadai, M. and Harran, P.G. (2006) Tetrahedron Lett., 47, 3933–3935. (a) Wiese, K.J., Yakushijin, K., and Horne, D.A. (2002) Tetrahedron Lett., 43, 5135–5136; (b) Olofson, A., Yakushijin, K., and Horne, D.A. (1997) J. Org. Chem., 62, 7918–7919. Shroif-Gr´egoire, C., Travert, N., Zaparucha, A., and Al Mourabit, A. (2006) Org. Lett., 8, 2961–2964. Al-Mourabit, A., Pusset, M., Chtourou, M., Gaigne, C., Ahond, A., Poupat, C., and Potier, P. (1997) J. Nat. Prod., 60, 290–291. Fedoreyev, S.A., Ilyin, S.G., Utkina, N.K., Maximov, O.B., Reshetnyak, M.V., Antipin, M.Y., and Struchkov, Y.T. (1989) Tetrahedron, 45, 3487–3492. Abou-Jneid, R., Ghoulami, S., Martin, M.-T., Tran Huu Dau, E., Travert, N., and Al-Mourabit, A. (2004) Org. Lett., 6, 3933–3936. Schroif-Gr´egoire, C., Travert, N., Zaparucha, A., and Al-Mourabit, A. (2006) Org. Lett., 8, 2961–2964. P¨overlein, C., Breckle, G., and Lindel, T. (2006) Org. Lett., 8, 819–821. Seiple, I.B., Su, S., Young, I.S., Lewis, C.A., Yamaguchi, J., and Baran, P.S. (2010) Angew. Chem. Int. Ed., 49, 1095–1098. Feldman, K.S. and Fodor, M. (2008) J. Am. Chem. Soc., 130, 14964–14965. Picon, S., Tran Huu Dau, E., Martin, M.-T., Retailleau, P., Zaparucha, A., and Al Mourabit, A. (2009) Org. Lett., 11, 2525–2528. Z¨ollinger, M., Mayer, P., and Lindel, T. (2006) J. Org. Chem., 71, 9431–9439. Wiese, K.J., Yakhushijin, K., and Horne, D.A. (2002) Tetrahedron Lett., 43, 5135–5136. Feldman, K.S., Skoumbourdis, A.P., and Fodor, M.D. (2007) J. Org. Chem., 72, 8076–8086. Wang, S. and Romo, D. (2008) Angew. Chem. Int. Ed., 46, 1304–1306. Imaoka, T., Iwamoto, O., Noguchi, K., and Nagasawa, K. (2009) Angew. Chem. Int. Ed., 47, 3799–3801. Poullennec, K.G. and Romo, D. (2003) J. Am. Chem. Soc., 125, 6344–6345.
39. Zoellinger, M., Mayer, P., and Lindel, T.
(2007) Synlett, 2756–2758. 40. Jacquot, D.E.N., Zoellinger, M., and
41.
42. 43.
44. 45.
46.
47. 48.
49.
50.
51.
52.
53.
54.
Lindel, T. (2005) Angew. Chem. Int. Ed., 44, 2295–2298. Travert, N., Martin, M.-T., Bourguet-Kondracki, M.-L., and Al-Mourabit, A. (2005) Tetrahedron Lett., 46, 249–252. Overman, L.E. (1980) Acc. Chem. Res., 13, 218–224. Meijer, L., Thunnissen, A.M., White, A.W., Garnier, M., Nikolic, M., Tsai, L.H., Walter, J., Cleverley, K.E., Salinas, P.C., Wu, Y.Z., Biernat, J., Mandelkow, E.M., Kim, S.H., and Pettit, G.R. (2000) Chem. Biol., 7, 51–63. Nguyen, T.N.T. and Tepe, J.J. (2009) Curr. Med. Chem., 16, 3122–3143. Barrios Sosa, A.C., Yaskushijin, K., and Horne, D.A. (2002) J. Org. Chem., 67, 4498–4500. Schmitz, F.J., Gunasekara, S.P., Lakshmi, V., and Tillekeratne, L.M.V. (1985) J. Nat. Prod., 48, 47–53. Xu, Y., Yakushijin, K., and Horne, D.A. (1997) J. Org. Chem., 62, 456–464. D’ambrosio, M., Guerriero, A., Debitus, C., Ribes, O., Pusset, J., Leroy, S., and Pietra, F. (1993) J. Chem. Soc., Chem. Commun., 1305–1306. Hong, T.W., Jimenez, D.R., and Molinski, T.F. (1998) J. Nat. Prod., 61, 158–161. Tilvi, S., Moriou, C., Martin, M.-T., Gallard, J.-F., Sorres, J., Patel, K., Petek, S., Debitus, C., Ermolenko, L., and Al-Mourabit, A. (2010) J. Nat. Prod., 73, 720–721. Hale, K.J., Domostoj, M.M., El-Tanani, M., Campbell, F.C., and Mason, C.K. (2005) in Strategies and Tactics in Organic Synthesis(ed. M. Harmata), Academic Press, London. (a) Anderson, G.T., Chase, C.E., Koh, Y.-H., Stien, D., and Weinreb, S.M. (1998) J. Org. Chem., 63, 7594–7595; (b) Weinreb, S.M. (1999) J. Am. Chem. Soc., 121, 9574–9579. Araki, A., Tsuda, M., Kubota, T., Mikami, Y., Fromont, J., and Kobayashi, J. (2007) Org. Lett., 9, 2369–2371. Dickson, D.P. and Wardrop, D.J. (2009) Org. Lett., 11, 1341–1344.
References 55. (a) Feldman, K.S. and Saunders,
56.
57.
58. 59.
60.
61.
62.
63.
J.C. (2002) J. Am. Chem. Soc., 124, 9060–9061; (b) Feldman, K.S., Saunders, J.C., and Wrobleski, M.L. (2002) J. Org. Chem., 67, 7096–7109. Tsukamoto, S., Kato, H., Hirota, H., and Fusetani, N. (1996) J. Nat. Prod., 59, 501–503. Endo, T., Tsuda, M., Okada, T., Mitsuhashi, S., Shima, H., Kikuchi, K., Mikami, Y., Fromont, J., and Kobayashi, J. (2004) J. Nat. Prod., 67, 1262–1267. Bhandari, M.R., Sivappa, R., and Lovely, C.J. (2009) Org. Lett., 11, 1535–1538. Walker, R.P., Faulkner, D.J., van Engen, D., and Clardy, J. (1981) J. Am. Chem. Soc., 103, 6772–6773. Keifer, P.A., Schwartz, R.E., Koker, M.E.S., Hugues, R.G. Jr., Rittschof, D., and Rinehart, K.L. (1991) J. Org. Chem., 56, 2965–2975. Northrop, B.H., O’Malley, D.P., Zographos, A.L., Baran, P.S., and Houk, K.N. (2006) Angew. Chem. Int. Ed., 45, 4126–4130. O’Malley, D.P., Li, K., Maue, M., Zographos, A.L., and Baran, P.S. (2007) J. Am. Chem. Soc., 129, 4762–4775. Kinnel, R.B., Gehrken, H.P., and Scheurer, P.J. (1993) J. Am. Chem. Soc., 115, 3376–3377.
64. Buchanan, M.S., Carroll, R., and
65.
66.
67.
68.
69.
70.
Quinn, R.J. (2007) Tetrahedron Lett., 48, 4573–4574. (a) Kinnel, R.B., Gehrken, H.P. and Scheuer, P.J. (1993) J. Am. Chem. Soc., 115, 3376–3377; (b) Kinnel, R.B., Gehrken, H.-P., Swali, R., Skoropowski, G., and Scheuer, P.J. (1998) J. Org. Chem., 63, 3281–3286. Tsukamoto, S., Tane, K., Ohta, T., Matsunaga, S., Fusetani, N., and van Soest, R.W.M. (2001) J. Nat. Prod., 64, 1576–1578. O’Malley, D.P., Yamaguchi, J., Young, I.S., Seiple, I.B., and Baran, P.S. (2008) Angew. Chem. Int. Ed., 47, 3581–3583. Su, S., Seiple, I.B., Young, I.S., and Baran, P.S. (2008) J. Am. Chem. Soc., 130, 16490–16491. Wang, S., Dilley, A.S., Poullennec, K.G., and Romo, D. (2006) Tetrahedron, 62, 7155–7161. Patel, K., Laville, R., Martin, M.-T., Tilvi, S., Moriou, C., Gallard, J.-F., Ermolenko, L., Debitus, C., and Al-Mourabit, A. (2010) Angew. Chem. Int. Ed., 48, 4775–4779.
269
271
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin Edmond Gravel
8.1 Introduction
As discussed in other chapters, a vast majority of the alkaloids known to date are biosynthetically derived from amino acids such as tryptophan, tyrosine, lysine, and many others. Nevertheless, a certain number of alkaloids stand out as exceptions and find their origin in polyketide or terpene precursors instead. The present chapter is intended to be a survey of some relevant biomimetic synthetic efforts toward these particular natural products. Alkaloids of polyketide origin will be treated first and will constitute the major part of this contribution, following which terpenoid alkaloids will be discussed.
8.2 Galbulimima Alkaloids
In the late 1950s, Ritchie et al. discovered a new family of alkaloids [1] that were isolated throughout the following decade from the bark of relic trees native to Papua New Guinea and northern Australia belonging to the Galbulimima (also known as Himantandra) genus (Himantandraceae) [2]. These natural products have been divided into three classes [3] based on structural features [4] (Figure 8.1) and the number of isolated molecules in this family is still increasing [5]. While all three classes share the same trans-decalin ring system, the lactone derivatives of Class I seem rather simple compared to complex alkaloids of Class II or Class III. In the late 1960s, the Ritchie group had already pointed out a possible biogenetic precursor common to all alkaloids of the family [2d] without disclosing much detail about the plausible required transformations. A more detailed biosynthetic hypothesis for alkaloids of Class I was later proposed by the Baldwin group [6], followed by a unified biosynthetic route to alkaloids of Class II and Class III suggested by Movassaghi et al. [7]. Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
272
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin Me N
Me
AcO
H
O
N Me MeO
H O
H H H (−)-himbacine (1) (Class I)
Figure 8.1
HO
HH
Me
HH NH Me H H O H
O Bz OAc
H O OAc
(−)-himbosine (Class II)
(−)-Galbulimima alkaloid 13 (21) (Class III)
Representative structures of the different classes of Galbulimima alkaloids.
8.2.1 Alkaloids of Class I
Himbacine (1), belonging to Class I, was the first alkaloid of the family to be fully described [2a] and was shown to be a selective muscarinic antagonist, hence potentially opening the way to a new class of drug molecules for the treatment of Alzheimer’s disease [8]. Since then, 1 has motivated many synthetic efforts [9], among which Baldwin et al. proposed the first biomimetic total synthesis [6a] and later extended their approach to other alkaloids of Class I such as himbeline (2) and himandravine (3) [6b]. Me R N
Me H
Me
N H Me
endo [4+2] reductions
reductive lactonization reductive amination N -methylation or protonation
R
O
Me
Me
H O
H HH Class I alkaloids (R = H or Me)
O O
O OH
O 5
O
4
O
Scheme 8.1 Baldwin’s biosynthetic hypothesis for the origin of Class I Galbulimima alkaloids.
Their strategy relied on their own biosynthetic hypothesis (summarized in Scheme 8.1), according to which polyketide derivative 4 would first undergo reductive lactonization followed by reductive amination and N-methylation or protonation to give rise to intermediate 5. The tetrahydropyridinium ring of the latter would serve as activator for a biological Diels–Alder reaction via an endo transition state that would lead, after hydride reduction of the iminium ion to form a cis- or trans-piperidine ring, to the general scaffold of Class I alkaloids. As a synthetic precursor of postulated intermediate 5, the authors decided to prepare dienone 6. They predicted that the latter, upon removal of the Boc group, would cyclize into iminium 5 which would then spontaneously undergo the cycloaddition process (Scheme 8.2) [6a]. Key biomimetic intermediate 6 was assembled from chiral phosphonate 7 (obtained in 9% yield over three simple steps) and aldehyde 8 (obtained in 29% yield over seven steps) by a Horner–Emmons reaction that proceeded in 50%
8.2 Galbulimima Alkaloids H N
O
Boc
Me acidic Boc cleavage condensation
Me
H N
H H H H
IMDA, ...
Me 5′ O
endo transition state
O
Scheme 8.2
Class I alkaloids
O
6 O
Me
Conversion of dienone 6 into intermediate 5 .
yield. The Boc group was cleaved by treatment of 6 with trifluoroacetic acid at 0 ◦ C (very likely leading to the formation of 5 after condensation); the mixture was then allowed to slowly warm up to room temperature to enable the cycloaddition process, and the reaction was finally quenched by addition of sodium cyanoborohydride followed by aqueous sodium bicarbonate. This key biomimetic sequence yielded 9 as a 1 : 1 mixture of isomers (epimers at C6) that could not be separated. Boc protection was thus performed on the crude reaction mixture followed by selective reduction of the trisubstituted double bond with Adam’s catalyst and, after purification by flash chromatography, compound 10 and compound 10 were obtained in 11% yield each from 6. TFA treatment to remove the Boc groups of 10 and 10 permitted the formation of himbeline (2) and himandravine (3), respectively, in high yields. N-methylation of 2 finally afforded himbacine (1) in 74% yield (Scheme 8.3) [6b]. Interestingly, by their approach, Baldwin et al. have accessed three of the four Class I Galbulimima alkaloids. The fourth Class I alkaloid, himgravine (11), was not synthesized but might have been obtained by N-methylation of the isomer of 9 with C6 in the (R)-configuration, had the latter been separable from its epimer. 8.2.2 Alkaloids of Class II and Class III
Different research groups have studied the synthesis of Class II [10, 11] and Class III [7, 12] alkaloids and the group of Movassaghi has proposed a unified biosynthetic pathway for these molecules (Scheme 8.4). They postulated that polyketide precursor 12 (analog of 4, oxidized at C16) can undergo reductive amination followed by condensation, giving rise to tetrahydropyridinium 13, which will then cyclize via an intramolecular Diels–Alder (IMDA) cycloaddition. On the obtained cycloadduct 14, the ketone at C20 is free to react (contrary to the Class I pathway in which it is engaged in the lactone cycle) in its enol form by conjugate addition to permit the formation of the C21–C8 bond and yield intermediate 15. After tautomerization of the resulting imine into the corresponding endocyclic enamine, attack of the latter on the C20 ketone would lead to the formation of the C5–C20 bond, and upon reduction of the residual imine give rise to intermediate 16. Enone 17, resulting from the oxidation of 16, can be seen as the
273
274
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin Key biomimetic intramolecular Diels – Alder reaction
Boc 7 MeO
O P
3 steps O ca. 9% Me
OMe
Me
N H
+
unseparable epimers
H
(50%)
H
Me
H
Me O 6
8
7 steps O ca. 29%
HCHO, NaBH3CN, MeCN (74%)
R = H: himbeline (2)
H
O 9
O
Me H
H
H
N
R TFA, DCM
N
Boc
10 H
H
Me
H
H
O
(95%)
R = Me: himbacine (1)
O H
H
?
H
H
Me
Me O
H O himgravine (11)
Me H
Me
N
O
Me O
H
NH
6
TFA, DCM then NaBH3CN
EDIPA, LiCl MeCN
Me
Me
O
Boc
O
NH
1) Boc2O, TEA, DCM 2) H2, PtO2, EtOH (22% over 3 steps)
O
O
separable epimers
Me H
Me H
NH 10′
himandravine (3)
H
H
Me
TFA, DCM
H
H
Boc
Me
(92%)
O H
Scheme 8.3
N
H
O
O H
H
O
Baldwin’s biomimetic approach to Class I Galbulimima alkaloids [6b].
pivotal intermediate that will lead either to 18 (precursor of Class III alkaloids) by conjugate addition (forming the N–C19 bond) and decarboxylation or to 19 and then 20 (precursor of Class II alkaloids) by oxidation [7]. Inspired by this putative biosynthetic pathway, the authors achieved the total synthesis of both alkaloids of Class II (himandrine) [10] and Class III (Galbulimima alkaloid 13 or GB 13) [7]. During these investigations, they obtained exciting results that support their hypothesis. In 2006, with the synthesis of GB 13 (21) [7], they demonstrated the introduction of the C5–C20 bond from synthetic intermediate 22 (obtained in 15% yield over eleven steps). Deprotection of the silyl enol ether of 22 gave transient compound 23 (biomimetic equivalent of 15), which readily cyclized by formation of the C5–C20 bond and upon treatment with sodium borohydride afforded 24 in circa 70% yield1) (Scheme 8.5). In this one-pot process, three contiguous stereocenters were formed with a high level of diastereoselectivity, and target alkaloid 21 was obtained after three additional synthetic steps. This synthesis also provided the first evidence that revision of the absolute stereochemistry of Class II/III alkaloids was needed. 1) The yield here is given based on (−)-22
even though the isomers were separated at a later stage in the synthesis.
8.3 Cyclic Imine Marine Alkaloids Me
O O Me O HO
reductive amination condensation
H
Me
H H H
HO O
16
O 12
Me
H N
O
OH
IMDA
HO
N H
21 8
H O
OH
O
13
H
H OH H 14 conjugate addition tautomerization
HO
HO H
carbonyl addition imine reduction
H
oxidation
19
NH Me HO O O H 17
H
N H H HO
H OH H 15
HO H
H 9
OH
Class III alkaloids
Scheme 8.4
20
O
HO
HO
18
O
oxidation
conjugate addition decarboxylation
N Me
Me
NH H OH Me HO O H 16
5
NH Me HO
formation of the spiro-fused framework
OH O
OH 19
H N Me HO
O
OH 20
Class II alkaloids
Movassaghi’s biosynthetic hypothesis for Class II and Class III alkaloids.
More recently [10], while completing the first total synthesis of a Class II alkaloid, the authors demonstrated that the formation of the N–C9 bond, leading to the spirofused polycyclic framework, could result from the oxidative spirocyclization of intermediate 25 (synthetic equivalent of 17, obtained in circa 1% yield over 20 steps). The latter was treated with N-chlorosuccinimide to supposedly afford α-chloroester 26 (biomimetic equivalent of 19) and permit the formation of 27, by intramolecular allylic displacement, in 89% yield (Scheme 8.5). Only two further synthetic steps (reduction of the enone into enol and benzoylation) were required to complete the first synthesis of himandrine (28).
8.3 Cyclic Imine Marine Alkaloids
In the past two decades, Uemura et al. have been implicated in many studies regarding the chemistry of marine alkaloids. In the course of their work, it occurred to them that several of these natural products bore a common structural feature: a cyclic imine either spiro- or side-fused to a cyclohexene. Most of the alkaloids concerned were suspected to originate from dinoflagellates and the authors proposed that a general scenario, based on an IMDA reaction involving an iminium-activated dienophile (vide infra), might account for their biosynthesis. Members of this family of natural compounds include symbioimine, pinnatoxins, pteriatoxins, and gymnodimines that will be discussed in the following sections,
275
276
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin Introduction of C5—C20 bond (ref. [7]): TBSO
5
H Me
Et3
N H
3
O
H NH O
HO
20
H
23
H NaBH4
N H H H
Me H
22
N . (HF)
H NH
(ca. 70%)
O
O
O
ca. 15% over 11 steps
NH Me H H N H 24 O O HO
3 steps ca. 46%
HH
NH Me H H O H Galbulimima alkaloid 13 (21) Formation of N—C9 bond and spirofused polycyclic framework (ref. [10]): HO
HO
HO
H NCS
NH Me MeO
O
OH
OMe
9
NH Me MeO
25
H
Cl OH O 26
H
(89%)
N Me MeO
OMe
O
OH
OMe
27
ca. 1% over 20 steps 2 steps ca. 78%
HO HH N Me MeO
H O OBz
OMe
himandrine (28)
Scheme 8.5 Biomimetic processes in Movassaghi’s synthetic efforts toward Class II and Class III alkaloids.
but also spirolides [13], prorocentrolides [14], and spiroprorocentrimine [15], which are presented in Figure 8.2 but will not be detailed any further. 8.3.1 Symbioimine and Neosymbioimine
In 2004 the research group of Uemura reported the isolation of symbioimine (29) from Symbiodinium sp. (Symbiodiniaceae), a dinoflagellate cultivated from its symbiotic partner, the marine flatworm Amphiscolops sp. (Convolutidae) [16]. The molecule is composed of an uncommon 6,6,6-tricyclic iminium ring system and an aryl sulfate moiety (Figure 8.3). In terms of biological activity, this singular tetracyclic iminium sulfate alkaloid inhibits the differentiation of RAW264 cells into osteoclasts (EC50 = 44 µg ml−1 ) and can thus be considered a potential candidate for the prevention and treatment of osteoporosis. Symbioimine also shows a significant anti-inflammatory potential as it inhibits cyclooxygenase-2 activity (32%) at 10 µM without affecting cyclooxygenase-1 activity. The structure and biological activity of this new natural
8.3 Cyclic Imine Marine Alkaloids OH
Me
OH H N
H
OH H HO
O
H O OH Me
O
O
Me
HO O
HN
OSO3
Me spiro-prorocentrimine (Prorocentrum sp.)
Me
Me OH spirolide C (Alexandrium ostenfeldii)
OH OH
Me
HN
OSO3−
Me
R HRH R = H: symbioimine (29) R = Me: neosymbioimine (30)
Figure 8.3
O
Examples of marine alkaloids featuring a cyclic imine fused to a cyclohexene.
O3SO HNH
O O
OH
Me prorocentrolide B (Prorocentrum lima)
H
O HO
O
Me Me
Me
OH
N
Me
OH
Figure 8.2
O
OSO3 OH
Me
OH
O
Me
O OH H
Me Me Me
OH
HO
R R cyclic iminium side-fused to cyclohexene
Structure of symbioimine and neosymbioimine.
product have stimulated several efforts toward its synthesis [17]. One year after the discovery of symbioimine, Uemura et al. reported the isolation of neosymbioimine (30), a compound closely related to the former [18]. In their first paper, the authors proposed that the biosynthesis of symbioimine involved the cyclization of polyketide precursor 31 by an IMDA cycloaddition proceeding via an exo transition state followed by condensation to form the iminium. This proposal appeared quite unusual considering that most literature precedents concerning IMDA on such substrates were in favor of an endo transition state [19]. An alternative biosynthetic pathway was proposed along with the neosymbioimine isolation report and suggested that the IMDA occurred through an endo transition state after formation of dihydropyridinium precursor 32. The obtained cycloadduct 33 could then undergo epimerization at C4 to yield symbioimine (Scheme 8.6). This latter hypothesis was supported, in the following years, by studies toward the biomimetic synthesis of symbioimine (Scheme 8.7) from the groups of Snider [17c,d] and Thomson [17e]. Snider et al. started with the synthesis of biomimetic intermediate 34 from simple starting material in nine synthetic steps with 55% overall yield. Key compound 34 was obtained as a mixture of isomers with a cis/trans ratio of 3 : 1 at C2–C3 and an (E/Z) ratio of 6 : 1 for the C9=C10 olefin. Upon treatment with BF3 ·Et2 O, tetrahydropyridinol 34 was converted into dihydropyridinium 35, which directly
277
278
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin OR Me
HN epimerization
4
OH HN
endo [4+2]
HN
2
H
2
Me
OR
OR
H 33
H
32
OR
Me 2
H 4
OSO3
H
H symbioimine (29) (Symbiodinium sp.)
−
NH2
condensation
OR
4
2
exo [4+2]
Me
OR
2
H
OR
H
O
33'
H
Scheme 8.6
H2 N
Me
O
31
OR
Proposed biosynthetic origin of symbioimine.
Thomson's approach
Me
N
OMe
38 OMe
Me
HN
endo [4+2]
TFA
OMe
OMe
Me
HN
H 4
OMe ca. 18% over 10 steps
H epimerization
Snider's approach
Troc
Me
N
1) Zn, MeOH/AcOH
N
Me H
2) K2CO3
H 36
H
R Me H 4
R
H
cis / trans = 3:1
Troc Me endo [4+2]
H
(100%)
(25−37%)
R = H: deoxysymbioimine (37) R = OMe: 39
H
(31%)
N
CF3CO2 HN
TFA
H
(75%)
H
Troc
OMe
H
OMe
N
Me
Troc
Me
N
2
BF3.Et2O
34 3
OH
9
H H
35
10 E / Z = 6:1
ca. 55% over 9 steps
Scheme 8.7
Biomimetic syntheses of the symbioimine skeleton.
cyclized into tricyclic 36 in 31% yield [based on the (E)-isomer]. The latter was deprotected in presence of zinc dust then neutralized with K2CO3 and treated with trifluoroacetic acid to afford (±)-deoxysymbioimine (37) [17c,d]. Thomson et al. chose to synthesize a cyclic imine as a biomimetic equivalent of 32. Key intermediate 38 was obtained in 18% overall yield over ten steps as a mixture of isomers at the C3 position but with control over the absolute configuration at C2. Treatment of 38 with TFA permitted the generation of a transient dihydropyridinium that readily cyclized and epimerized into tricyclic adduct 39 in 25–37% yield. Optically active symbioimine (29) was then obtained upon phenol deprotection and selective sulfation (22% yield for the two steps), thus completing the first enantioselective total synthesis of this alkaloid [17e].
8.3 Cyclic Imine Marine Alkaloids
Both routes share the same key biomimetic IMDA during which two new rings and four stereocenters are diastereoselectively formed in a single step (with comparable yields in the two teams), which is quite impressive in terms of molecular complexity. Notably, the presence of a single methyl substituent at C2 is responsible for the π-facial selectivity of the cycloaddition. Apart from the enantioselectivity, the two approaches mainly differ from the fact that Thomson’s synthesis proceeds via an iminium ion intermediate rather than an acyl iminium ion (which requires further deprotection after cycloaddition) as proposed in Snider’s work. Very recently, the group of Chruma disclosed a model study with further arguments in favor of the revised biosynthetic hypothesis involving the endo-IMDA [20]. In this report, the authors showed that when all-(E) intermediate 40 (closely related to 31) underwent an IMDA reaction it was via an endo transition state, as depicted in Scheme 8.8, and not via an exo transition state as originally proposed. The reaction yielded a single diastereomer (41, epimeric at C3 and C4 compared to the postulated biosynthetic intermediate 33 ), confirming that the methyl substituent at C2 is sufficient to induce the π-facial selectivity of the IMDA. Moreover, this study demonstrated that (Z)-enones, such as 42, isomerized into the all-(E) form before undergoing endo-IMDA, thus highlighting the importance of constraining these precursors in the form of dihydropyridinium salts to prevent such isomerization and increase the dienophile’s reactivity.
Me
BocHN
OMe
2
Me2AlCl −20 °C
O
OMe
MeO
NHBoc 2 Me H 4
OMe
H OMe
3
H
41
H
endo -boat
NH2
−78 °C, 30 min
NHBoc Me
O
2
BocHN
all-E precursor
O
H
Me
40
enone isomerization
Cl Me Al Me O
MeO
O
OMe Me2AlCl −78 °C → −20 °C
NHBoc Me O H H
OMe
OMe
2
H 4
H OMe
OR Me H OR
3
33′
biosynthetic intermediate
42 (Z )-enone
H
Scheme 8.8 Chruma’s contribution to our understanding of the biosynthesis of symbioimine.
8.3.2 Pinnatoxins and Pteriatoxins
Described by the group of Uemura in 1995 [21], pinnatoxin A (43) was the first representative isolated in this macrocyclic toxin family that was later enlarged with the report of pinnatoxins B, C [22], D [23], E, F, and G [24] as well as
279
280
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin 27-membered carbocycle
Me [6,5,6]-dispiroketal
O
HO Me O
cyclic iminium
Me
HN
[6,7]-spiro-ring system
H O
cyclohexene
O R
O
[5,6]-bicycloketal
OH Me R=
CO2H
: pinnatoxin A (43)
CO2H : pinnatoxin B
R=
CO2H : pinnatoxin C
CO2H
R=
R=
O
CO2H : pinnatoxin D
OH O
: pinnatoxin F
R=
NH2 S
R=
: pteriatoxin B
CO2H OH
: pinnatoxin G R=
O
Figure 8.4
: pteriatoxin A
OH
: pinnatoxin E
NH2 R=
CO2H
S
OH
NH2 R=
H2N R=
NH2 S
: pteriatoxin C
CO2H
General structure of pinnatoxins and pteriatoxins.
pteriatoxins A–C (Figure 8.4) [25]. Extracted from Pinna muricata (Pinnidae),2) and also present in other Okinawan bivalves such as P. attenuata, P. atropurpurea, or Atrina pectinata, pinnatoxins are suspected of being responsible for the frequent food poisoning symptoms (paralysis, diarrhea, convulsion, etc.) following the ingestion of this commonly eaten shellfish, and have been shown to be Ca2+ channel activators. Structurally, all these molecules share the same highly complex macrocyclic skeleton featuring a [6,5,6]-dispiroketal moiety, a [6,7]-spiro-ring system composed of cyclic iminium attached to a cyclohexene and a [5,6]-bicycloketal (tetrasubstituted dioxabicyclo[3.2.1]octane). The difference between the toxins lies in the nature of the side-chain attached to the cyclohexene’s olefin (Figure 8.4). The intriguing molecular architecture of these natural products combined with their pronounced biological activity motivated several chemists to work on their synthesis, especially on that of pinnatoxin A (43) [26]. The first reported total synthesis of 43 was achieved by the group of Kishi [26a] and was based on Uemura’s biosynthetic hypothesis [21a] that proposed an IMDA reaction followed by condensation to cyclic imine for the formation of the [6,7]-spiro-ring system (Scheme 8.9 depicts the possible biosynthetic origin of pinnatoxin A). In this work, the authors first synthesized precursor diketone 44, which, upon treatment with camphorsulfonic acid, yielded the tricyclic dispiroketal core (formation of three new cycles and two new stereocenters) bearing a free hydroxyl at C24, as a mixture of epimers at C19. Fortunately, under classical silylation conditions, 2) Pteriatoxins were extracted from Pteria
penguin (Pteriidae).
8.3 Cyclic Imine Marine Alkaloids
Me HO Me O
O
Me
HN
O
O
OH
intramolecular Diels −Alder
HO HO
Me
Me
O
CO2H
OH
(+)-pinnatoxin A (43) (Pinna muricata)
Me
O
CO2
O
Scheme 8.9
H2N OH
HO Me O
H O
condensation to cyclic imine
formation of dispiroketal
cyclizing Me ketalization
OH
Plausible biosynthetic origin of pinnatoxin A.
Me NHAlloc HO Me 12 steps ca. 24%
OTBS
TBSO
1) CSA, MeOH 2) TBSOTf, lutidine
O 19
O
O
15
HO Me O
OTBS
OH
45
46
OTBS
Me
1) TFA, H2O 2) MsCl, NEt3 3) TESOTf, lutidine
(ca. 48%)
Key biomimetic intramolecular Diels − Alder
AllocHN Me O
Me
O TESO Me O
TESO Me O
32
O
47
CO2tBu
OTBS
NHAlloc Me
O 6
Me
H
1) HF.py, pyridine 2) Pd(PPh3)4, AcOH 3) 200 °C, vacuum HO Me 4) TFA (ca. 51%)
O O
OMs
Me
Me TESO Me O
O O
Me
15 O
Me
15 O
CO2tBu
IMDA
49
O DABCO NEt3
OTBS
48
desired exo product (34%)
AllocHN
O O
O
CO2tBu 32
24
44
O
Me
Me Me O OO
O
19 O
Me Me
O
24
16 steps ca. 9%
Me O
(ca. 78%)
O O
O
CO2tBu
O
Me
HN
O
H O
O CO2
O
28
OH
OTBS
Me
Me
(−)-pinnatoxin A (50)
Scheme 8.10
Kishi’s biomimetic synthesis of (−)-pinnatoxin A [26a].
the undesired epimer isomerized to the natural series and afforded intermediate 45 in 78% yield for the two steps (Scheme 8.10). Acetonide 46 was obtained after 16 synthetic steps (circa 9% overall yield) and was converted into bicycloketal 47 in 48% yield over three steps (acetonide cleavage, mesylation of C32 secondary hydroxyl, protection of C15 tertiary hydroxyl). The SN 2 displacement of the C32
281
282
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin
mesylate with DABCO permitted the transformation of 47 into transient diene 48, which then readily cyclized via an IMDA reaction to give rise to exo cycloadduct 49 as the major product (34%). Deprotection of C15 and C28 hydroxyls as well as terminal amine were easily carried out under classical conditions, but the cyclization into imine by condensation of the amine and C6 ketone required drastic conditions (heating at 200 ◦ C under vacuum for 1 h). Finally, (−)-pinnatoxin A (50) was obtained (after cleavage of the tert-butyl ester) in 51% yield over the four last steps (Scheme 8.10) [26a]. Notably, this impressive synthesis permitted the establishment of the absolute stereochemistry of pinnatoxin A as 43, the antipode of 50 originally proposed by Uemura. 8.3.3 Gymnodimine and Derivatives
Isolated from oysters collected in New Zealand [27], gymnodimine (51) is another spirocyclic imine marine toxin that was shown to be a powerful ligand of muscle-type nicotinic cholinergic receptors [28]. It was first described in 1995 by Yasumoto et al., who extracted it from the shellfish Tiostrea chilensis (Ostreidae) and dinoflagellates of the Karenia3) genus (Gymnodiniaceae). They established its gross structure, which was later detailed (in terms of stereochemistry) by Blunt and Munro [29]. Gymnodimine is a macrocyclic molecule composed of a trisubstituted tetrahydrofuran, a cyclic imine spiro-fused to a cyclohexene ring and a butenolide moiety (Figure 8.5). Several years later, the group of Miles reported the isolation of gymnodimines B [30] and C [31], oxidized analogs of 51 that bear an exocyclic C17=C29 double bond and an allylic hydroxyl at C18. With its unique structure and intriguing biological activity, gymnodimine has attracted the attention of several synthetic chemists [32]. Among them, the group of Kishi has developed a biomimetic approach [32e] headed towards the formation of the spiro-ring system by an IMDA reaction as a means to close the macrocyclic framework of the molecule. Their strategy was based on the biosynthetic hypothesis developed by Uemura for pinnatoxin A [21a] that seems reasonably applicable to gymnodimine (Scheme 8.11) considering the obvious structural similarities that exist between the two toxins. To achieve the biomimetic construction of the macrocycle (Scheme 8.12), the authors first prepared triene 52 (obtained in 62% yield over four steps), protected amine 53 (obtained in 60% yield over two steps), and tetrahydrofuran 54 (obtained in 5% yield over 14 steps). These three precursors were then assembled to permit the construction of compound 55 in 8% yield over eight additional steps. Under classical deprotection conditions 55 was transformed into 56, which was then treated with (PPh3 )4 Pd to cleave the alloc carbamate and give α,β-unsaturated imine 57 (51% for the two steps). The key biomimetic IMDA was carried out under 3) Organisms of the Karenia genus where for-
merly included in the Gymnodinium genus.
8.3 Cyclic Imine Marine Alkaloids disubstituted butenolide
O
O
Me H
Me
trisubstituted tetrahydrofuran
OH
cyclohexene
Me 18
[6,7]-spiro-ring system
N
cyclic imine
OH O
O 17
Me
18
16-membered carbocycle
C18-(S ): gymnodimine B C18-(R ): gymnodimine C
gymnodimine (51)
Figure 8.5
Structure overview of gymnodimine and congeners.
intramolecular Diels − Alder
O O
OH
Me H
Me
Me
Me
O O
Me
Me
Me
O Me
Me
N
O Me NH2 condensation
Me
gymnodimine (51) (Karenia selliformis)
Scheme 8.11
OH
Me O
to cyclic imine
Plausible biosynthetic origin of gymnodimine.
ca. 62% over 4 steps
TIPSO
Me 52
O
OR1 Me
8 steps, ca. 8%
Br Me
OR2
O
Me
Me OH
I AllocHN
Me
54
Me O
ca. 5% over 14 steps
TBAF (78%)
53
ca. 60% over 2 steps
OH
OH
Me H
Me
Me
R1 = TIPS, R2 = TES: 55 R1 = H, R2 = H: 56
OH
H2O pH 6.5
O
Me
(PPh3)4Pd (66%)
Key biomimetic intramolecular Diels − Alder reaction
OH
O
AllocHN
Me
O
57
Me
Me N 58
Me
Me N
exo -cycloadduct (35%)
Scheme 8.12 Kishi’s biomimetic approach to the macrocyclic framework of gymnodimine [32e].
mild conditions (aqueous buffer pH 6.5 at 36 ◦ C) and afforded a mixture of three cycloadducts. One of these was the desired exo-cycloadduct 58 (35%) while the other two were endo-cycloadducts that were isolated as keto-amines, suggesting that their imine form is unstable in aqueous conditions [32e]. Notably, 58 is the only exo-cycloadduct that was isolated, demonstrating that the exo-IMDA proceeds with high facial selectivity. Also interesting is the fact that the
283
284
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin OH
Me
OH
OH
OH
Me
H2O pH 6.5
Me
O
O
Me
Me
Me O
TBAF
56′
O
Me TeocHN
TeocHN TBSO
Me
OTBS
Me
TBSO O
OTBS
Me H
PhH, 185 °C
Me
4 Å MS
TBSO O
55′
Scheme 8.13
O
Me TeocHN
TeocHN
OTBS
Me H
+
Me
O Me
Me
Me O
Me
Me
O 2:1
Me
TeocHN
Attempts at cycloaddition from keto-amine precursors [32e].
authors submitted deprotected keto-amine 56 to the same aqueous conditions and observed nothing but the decomposition of the triene moiety. In addition, when they attempted to achieve cyclization of protected keto-amine 55 they observed the formation of the two endo-cycloadducts with poor facial selectivity (2 : 1), but could never detect any exo-product (Scheme 8.13). The results obtained in this study suggest that an IMDA reaction could account for the formation of gymnodimine without enzyme assistance but that the formation of the α,β-unsaturated imine dienophile should occur prior to the cycloaddition process.
8.4 Other Polyketide Derived Alkaloids 8.4.1 Cassiarins A and B
In 2007, the group of Morita reported the isolation of two new alkaloids [33], cassiarins A (59) and B (60), extracted from the leaves of Cassia siamea (Fabaceae) collected in Indonesia. The aromatic molecules were shown to exhibit promising antimalarial activities against Plasmodium falciparum (59 in particular, with an IC50 of 0.005 µg ml−1 ), and a vasodilatator effect of cassiarin A was demonstrated in a recent study [34]. Structurally, both molecules share the same tricyclic aromatic core composed of a 3-methylisoquinoline fused to a 2-methylpyran ring and mainly differ in the fact that 60 is N-substituted whereas 59 is not (Figure 8.6). To date, only two groups have been working on the synthesis of cassiarins: the Morita group [35] and the Yao group [36]. The latter published a biomimetic total synthesis of both alkaloids based on Morita’s biosynthetic proposal (Scheme 8.14) [33], which suggests that chromone precursor 61 could react with ammonia or
8.4 Other Polyketide Derived Alkaloids HO
O
Me
Me
O
N
N
OMe
O
O
Me
Me cassiarin A (59)
Figure 8.6
cassiarin B (60)
Structure of cassiarins A and B.
HO
Me N
R=H −H2O
O RHN
Me cassiarin A (59)
HO
O
N
Me
O
O O
Me R = (CH2)3CO2Me −H2O
Me
HO
O
Me
O
O NH2R −H2O
(Cassia siamea)
O
Me
62
Me 61
OMe
cassiarin B (60)
Scheme 8.14
Biosynthetic origin of cassiarins proposed by Morita.
a primary amine to give intermediate enamine 62 that would then cyclize into isoquinoline to yield cassiarins. To conduct their biomimetic study, Yao and Yao prepared chromone precursor 63 under literature conditions [37] and obtained key intermediate 64 after three additional steps (phenol protection, triflate formation, and Negishi coupling, 65% yield overall). Treatment of 64 with trifluoroacetic acid in the presence of a catalytic amount of silver nitrate afforded 65 (synthetic equivalent of postulated biosynthetic precursor 61), which could not be isolated. Exposing the reaction mixture containing crude 65 to either ammonia or methyl 4-aminobutyrate, followed by cleavage of the MOM ether with hydrochloric acid, permitted the obtention of cassiarin A (60% yield from 64) and cassiarin B (52% from 64) respectively (Scheme 8.15) [36]. 8.4.2 Decahydroquinoline Alkaloids
Isolated from different terrestrial [38] and marine [39] organisms, decahydroquinoline alkaloids represent one of the major classes of the wide variety of toxic amphibian alkaloids [40] produced by tropical frogs of the Dendrobatidae [41] and Mantellidae [42] families. It is still undecided whether these molecules can be synthesized de novo by the amphibians and some observations suggest that they might be of dietary origin (especially since they have been found to occur in ants [38c,d]) [43].
285
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin Me HO
HO
OH
OH
3 steps ca. 65%
ref. 37
O
O
OH Me
MOMO O
O 63
O 64
Me
MOMO
Me
Me O O
O Key biomimetic condensation
286
65
1) Et3N, RNH3Cl 2) HCl
1) NH3 (g) 2) HCl
HO
Me
NH
O
Me
N
N
R
O Me
H
CF3
Me
O
Scheme 8.15
AgNO3, TFA
O
Me
cassiarin A (59)
cassiarin B (60)
(60%)
(52%) R = (CH2)3CO2Me
Yao’s biomimetic synthesis of cassiarins [36].
H H
NH
H
H
H N
H
H
cis -233C
2-epi-cis -275B
(Dendrobates pumilio)
Figure 8.7
H
H N
trans -219A
trans -243A
(Dendrobates histrionicus)
Representative examples of decahydroquinoline alkaloids.
Amphibian alkaloids have been the subject of numerous synthetic studies over the last three decades [44] and recently Amat and Bosch have been developing the biomimetic synthesis of decahydroquinoline alkaloids (Figure 8.7) [45]. These natural compounds possess a cis- or trans-fused decahydroquinoline substituted at both C2 and C5 positions. It has been suggested that they might find their biosynthetic origin in a linear polyketide precursor (such as 66) which would undergo an intramolecular aldolization–crotonization process to form a cyclohexenone intermediate (67). Upon reductive amination, condensation, and reduction, the latter would lead to the 2,5-disubstituted decahydroquinoline framework (68) (Scheme 8.16) [46]. Based on this biosynthetic scenario, Amat et al. decided to undertake the synthesis of cis-195A (69) [45c], formerly known as (−)-pumiliotoxin C [47], a representative alkaloid of the group. As a synthetic equivalent of 66, they prepared diketo ester
8.4 Other Polyketide Derived Alkaloids
R
H N
2
2
R
R N
R NH2
reductive amination reduction
R
2
2
OO
condensation
R
5 1
R
68
67
Scheme 8.16
O O
aldol O condensation
1
R
1
R
2
1
R
66
Proposed biosynthetic origin of decahydroquinoline alkaloids.
70 (in 65% yield from 5-oxohexanoyl chloride) and submitted it to treatment with aqueous lithium hydroxide followed by trimethylsilyl chloride in methanol and obtained cyclohexenone 71 in 82% yield. Reaction of the latter with (R)-phenylglycinol under acidic catalysis enabled its cyclocondensation to tricyclic lactam 72 in 70% yield (Scheme 8.17). This biomimetic sequence provides experimental support for the proposed biosynthetic scenario from which it was rationalized.
Biomimetic sequence
EtO
OO
1 step 65%
1) LiOH, EtOH MeO 2) TMSCl, MeOH
O Me
Ph
OO
(R )-phenylglycinol PhH, AcOH O
(82%)
Me
70
(70%)
71
72
H NH 69
H Me
O
N
H
Me
8 steps ca. 12%
Me cis -195A (−)-pumiliotoxin C
Scheme 8.17
Amat’s biomimetic synthesis of cis-195A [45c].
At this point, eight additional synthetic steps (mostly reductions and an Eschenmoser sulfide contraction to implement the propyl side-chain at C2) were needed to complete the synthesis of the target compound 69 in 12% yield (Scheme 8.17) [45c]. Interestingly, the authors noted that, depending on the order in which the reduction of the endocyclic double bond and the Eschenmoser contraction were carried out, different series of cis-decahydroquinolines could be accessed. When the tricyclic lactam intermediate 73 is first submitted to selective hydrogenation of the C5=C6 double bond, hydrogen uptake is directed to the β-face by the presence of an axial C–O bond to give 74. After conversion of the lactone into thioamide and treatment under Eschenmoser’s sulfide contraction conditions [BrCH2 R3 then P(OMe)3 , Et3 N] intermediate 75 is obtained. Reduction of the latter begins by hydrogenation of the exocyclic double bond to give 76 on which the hydrogen uptake is directed to the less hindered α-face of the iminium, giving access to the cis series of decahydroquinolines (77), as depicted in Scheme 8.18 (Path A) [45a].
287
288
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin Path B
Path A
Ph
Ph (R ) N
R3
sulfide contraction
O
O
Ph
Ph O
N
O
N
O
1
H
R
R1
73
R1
O
H
74
H2 uptake
HO
2
R3
O
N
H2 uptake
C4a inversion
H2 uptake
H
R
79 R
5
H N
2
5
HH N
R2
R1 75
76
H H N
R2
2 H NH
2 5
H 81 ent -2-epi-cis series
R1
R2
5
H
O
N
H
R3
H2 uptake
H H
H2 3 AcOH R
H N
H
80
Ph
3
2
2
R2
R1
Ph H HO
H
2
1
R1
5 4a
N
H
4a
Scheme 8.18
Ph
R1
sulfide contraction
H2, AcOH
Ph
O
N
6
5
H
78
H2
5
R1
H R1
R1
77 cis series
Enantioselective synthesis of 2,5-disubstituted decahydroquinolines [45a].
In contrast, when 73 is first subjected to sulfide contraction, compound 78 is obtained and can be totally reduced in one-pot. This process begins with the hydrogenation of the exocyclic double bond to give intermediate 79, on which the presence of a β-oriented side-chain at C2 directs the hydrogen uptake to the α-face of the C5=C6 endocyclic double bond. At that point, an inversion of the C4a center is required to put the R1 substituent at C5 in an equatorial position and yield intermediate 80. Hydrogen uptake on the latter is directed to the less hindered β-face of the iminium, giving access to the ent-2-epi-cis series of decahydroquinolines (81) (Scheme 8.18, Path B). Access to the natural 2-epi-cis series can be achieved in the same manner from a tricyclic lactam precursor obtained from (S)- instead of (R)-phenylglycinol [45a]. 8.4.3 Zoanthamine Alkaloids
Zoanthamines are highly complex molecules extracted from polyps of the Zoanthus genus (Zoanthidae), collected in different regions around the world (Figure 8.8). The first members of the family, including zoanthamine (82), were isolated by the group of Faulkner and Rao in the 1980s from zoanthids collected off the coasts of India [48]. Later, Uemura et al. isolated new zoanthamine alkaloids from Zoanthus species collected off the coasts of southern Japan [49], Norte et al. reported the
8.4 Other Polyketide Derived Alkaloids O
H
R
O O H Me
B A
Me
Me H
Me
C
Me
N
D
O
O
E
H
H
MeO
O OH H O
Me
H
Me
N
N
O
O
Me O O
O Me
F
O
O
O H Me
HO O
Me
H
G
H O
Me O Me H
O O
Me N O
Me
Me
R = Me: zoanthamine (82) R = H: norzoanthamine
Figure 8.8
H
Me
zoanthenamine
cyclozoanthamine
zoanthenamide
Representative examples of zoanthamine alkaloids.
isolation of additional members of the family from polyps collected in the Canary Islands [50] and other groups have reported related molecules as well [51]. Zoanthamine alkaloids display a wide range of biological activities as some of them have been shown to be antiosteoporotic [52], anti-inflammatory [48a,b], cytotoxic [49], antibacterial [53], or active against platelet aggregation [54]. The challenging structural features and interesting biological activities of zoanthamines have prompted many research groups to develop their own approach to the synthesis of these alkaloids [55]; to date, only the groups of Miyashita [56] and Kobayashi [57] have reported total syntheses. Among the different synthetic studies toward these alkaloids, a biomimetic approach to the heterocyclic CDEFG ring system of zoanthamine has been disclosed by Kobayashi et al. [58]. Their strategy was based on Uemura’s hypothesis [49b, 59], according to which zoanthamines could be derived from linear polyketide precursor 83 (Scheme 8.19). O
Me O
Me
MeMe N
HO O
O
Me Me
Me O
Me Me
O OH Me HO2C H2N
Me
zoanthamine (82)
Scheme 8.19
O
OH
Me 83
Biosynthetic origin of zoanthamine proposed by Uemura.
Upon electrocyclization and Diels–Alder cycloaddition, enabling construction of the ABC rings, the latter could give rise to intermediate 84, which would then be transformed into 85 after hemiaminal ether formation, providing the FG rings. The E ring would then be formed by attack of the secondary amine on the ketone of ring C followed by dehydration to yield iminium 86. Finally, the D ring would be constructed by reduction of the cyclic iminium upon attack by the carboxylic acid to yield zoanthamine (Scheme 8.20). To mimic this scenario, Kobayashi and coworkers decided to study the cyclization of advanced intermediate 87 under different conditions. To this end, aldehyde 88 [obtained in 24% yield over ten steps from (+)-Wieland–Miescher ketone] [58b]
289
290
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin Me
Me
Me Me
O
HO
O B
OH
O OH Me HO2C H2N
Me
83
O Me H O
Me Me
H
A
O O
Me O
OE N F
O
Me
HN Me O HO2C
Me
Me
MeMe
O
N OG
N Me Me
Me
Me
O
G
Me Me
O
O
O
Me
O F
Me
85
D
Me
Scheme 8.20
Me Me
O OH
O Me O HO2C H2N 84
D C
Me
Me
C
Me
H
B
A
Me
Me Me
O
E
O
O
HO
Me
Me
86
zoanthamine (82)
Proposed biosynthetic route for zoanthamine [55].
and sulfone 89 (obtained in 11% yield over 17 steps from d-glutamic acid) were coupled and, after further modification, compound 87 was obtained in 38% yield over six steps (Scheme 8.21) [58a,b]. Me
Me Me OTBS
O
Me O
O
PhO2S Me
88
89
10 steps ca. 24%
17 steps ca. 11%
Scheme 8.21
Me
6 steps ca. 38%
CO2H
Me
+ O
Me Me
Boc N
O
O
O
Boc
MeMe
N
O
O O
HO Me
Boc N O
O
Me Me
O 87
Me
Synthesis of cyclization substrate 87 [58a,b].
With precursor 87 in hand, the authors explored different deprotection strategies. They thus found that upon treatment with 2N hydrochloric acid in tetrahydrofuran at room temperature only partial deprotection took place, allowing the formation of 90, which bears the bicyclohemiaminal ether FG ring system of zoanthamine and can be considered a synthetic equivalent of plausible biosynthetic intermediate 85 (Scheme 8.22). Heating of 90 in wet acetic acid for several hours and subsequent addition of a dehydrating agent permitted the deprotection of the Boc group and the formation of the D and E rings to yield compound 91 in 85% yield from 87. When the latter conditions were applied directly to substrate 87, pentacycle 91 was formed in 89% yield, presumably through the formation of transient intermediate 92, synthetic equivalent of plausible biosynthetic intermediate 84 [58a]. During this simple one-pot process, the fully functionalized CDEFG ring system of zoanthamine is formed in high yield, demonstrating the efficacy of this biomimetic strategy. Kobayashi et al. recently applied the stepwise version of this methodology to the total synthesis of norzoanthamine [57b].
8.4 Other Polyketide Derived Alkaloids plausible biosynthetic intermediate
Me O
HN Me O HO2C
Me
plausible biosynthetic intermediate
MeMe
Me Me
O
85
O O
O HO O
MeMe Boc 1) AcOH-H2O N O 2) Na2SO4 O
O H
90
N
O N
O E
O
O
(89%)
HO O
O
OH
O H2 N Me
G
91
Me
D
F
Me
OH
MeMe
C
Me
O H O
84
Me
MeMe
(95%)
Me
Me
O Me O HO2C H2N
1) AcOH-H2O 2) Na2SO4
87
HCl (89%)
HO
O
Me
Me
Me Me
O
Me Me
O
O
Me
Boc N
Me
92
Scheme 8.22 Kobayashi’s biomimetic approach of to the CDEFG rings of zoanthamine [58a].
8.4.4 Azaspiracids
Phykotoxins of the azaspiracid family were discovered in the late 1990s during an investigation into the cause of food poisonings [60] due to the consumption of Irish blue mussels (Mytilus edulis, Mytilidae) [61]. Subsequently, a total of eleven azaspiracids have been isolated [62] and their occurrence in different shellfish species [63] suggests that the origin of these alkaloids might be a marine dinoflagellate [64]. Azaspiracids are highly complex nonacyclic molecules with particular structural features such as a tetracyclic ring system composed of a substituted tetrahydrofuran side-fused to a trioxadispiroketal, a central highly substituted tetrahydropyran, a [6,6]-bicycloketal (dioxabicyclo[3.3.1]nonane) and a piperidine ring spiro-fused to a tetrahydrofuran, resulting in a spiro-hemiaminal ether moiety (Figure 8.9). Most azaspiracids differ from one another by the methylation (at C8 and C22) and hydroxylation (at C3 and C23) patterns and all share the same C26–C40 region. Recently, 20 new azaspiracid derivatives were identified, including dihydroxy (at C3 and C23) and carboxy-derivatives (at C22) [65]. The exceptional structural complexity of azaspiracids along with their biological activities (mainly toxic effects [66]) have motivated numerous synthetic efforts [67] that culminated in total syntheses by the groups of Nicolaou [68] (which led to the unambiguous elucidation of the structure of azaspiracid-1 (93) [69]) and Evans [70]. Like other polyether compounds produced by dinoflagellates, azaspiracids are thought to be of polyketide origin [71]. Based on the assumption that a linear precursor (94) might organize into the polycyclic framework of azaspiracids (Scheme 8.23), Forsyth et al. have devised two biomimetic strategies for the synthesis of the spiro-hemiaminal ether moiety of these molecules [67m].
291
292
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin Me spirohemiaminal ether I
40
highly substituted tetrahydropyran variable region 1 R = H or OH 2 R = H or Me 3 R = H or Me or CO2H 4 R = H or OH
R
O HO
C
O
H
D
H O
Me
H O OH
[6,6]-bicycloketal
HO
OH
O
O H
H O
O 2
4
O
Me
OH
H
H
Me
H HO
O
H
O
H
Me
[6,5,6]-trioxadispiroketal
3
R =R =R =H, R =Me: Azaspiracid-1 (93)
1
Figure 8.9
H
O
Me
H
1
R
General structure of azaspiracids and 3D structure of azaspiracid-1.
Me
Me
22
Me
O H
F
Me
3
O
3
HH N
26
E
B
O H
4 23
R H O
A
GO
O
Me R
2 8
H
H
Me
Me constant region O H H
HN
H
O HN H O H Me
H
spiro-hemiaminal ether formation
Me ketalization
HO H O HN O HO H Me H H H
H
H
Me
H2N
H
H
OHO
HO O HO
H
Me
H
H Me
Me
Me
azaspiracids
94 Me
NH
O
OH
Me
Me
27
NH2 40
O
Scheme 8.23
Me
OH
OH
O
Me
OH
OH O
Plausible biosynthetic origin of the FGHI rings of azaspiracids.
The first of these routes is based on a Staudinger/aza-Wittig sequence that was applied to intermediate 95 (obtained in 7% yield over 13 steps), synthetic equivalent of the C27–C40 region of postulated biosynthetic intermediate 94 (Scheme 8.24). Treatment of 95 with triethylphosphine furnished spiro-hemiaminal ether 96 in 75% yield, presumably through formation of iminophosphorane 97, which would cyclize into 98 by an intramolecular aza-Wittig reaction and then into 96 by addition of the C33 hydroxyl to the newly formed imine. The fact that 96 is obtained as the major product along with its C36 isomer (4 : 1 ratio) reflects either a kinetic pseudo-axial attack of the C33 hydroxyl on the Si face of the imine or a post-addition thermodynamic equilibration (or both) [67m]. The second route is based on a Staudinger/hetero-Michael sequence that was applied to carbamate-ynone 99 (obtained in 6% yield over 14 steps), another synthetic equivalent of the C27–C40 end (94). Conversion of 99 into spiro-hemiaminal ether 100 was carried out either stepwise or in a one-pot process. Under dichlorodicyano-quinone (DDQ) deprotection conditions, p-methoxybenzyl ether was cleaved but the conjugate addition of the nitrogen on the ynone was also initiated
8.5 Alkaloids Derived from Terpene Precursors Me
OH
13 steps ca. 7%
Me
N2
TMS
Me
O O TBS
Me
Boc
99
Me O O TBS TES
DDQ
Me N
36
PEt3
O 97
TBS Me O
TMS
101 OH
+ OPEt3
Me O O TBS TES 98
N
Me N Boc
OH
1) AgTFA, CH2Cl2 2) KI, EtOH
1) AgTFA, CH2Cl2 2) KI, EtOH (56%)
Me
Me
Me
(75%) 36
O
O
N H
Boc
NH Me
H
O Me O TBS TES ref. [67k]
I
Me 100
I
Me 96
Me O
Me
33
TES
H N
Et3P, PhMe
OH
TES
Me
OH
14 steps ca. 6%
O O Me O TBS TES 95
TES
TES
Me N3
O
O TBS
Me OH O Me
O N H H H Me
H
ref. [67k]
H
FGHI rings of azaspiracids
Scheme 8.24
Forsyth’s biomimetic approach to the HI rings of azaspiracids [67m].
to give 101. Treatment with silver trifluoroacetate and subsequent addition of ethanolic potassium iodine surprisingly afforded 100 (instead of the expected iodo-alkyne) by addition of the C33 hydroxyl to the enone-enamine. When the conditions initially intended to convert the terminal alkynyl silyl group into the corresponding iodo-alkyne were directly applied to 99, the two hetero-Michael additions were initiated and the C27 and C33 protecting groups were cleaved in a one-pot process to yield 100 in 56% yield (Scheme 8.24) [67m]. Both spiro-hemiaminal ethers 96 and 100 can then be easily converted into the FGHI ring-system of azaspiracids under conventional conditions, previously described by the Forsyth group [67k].
8.5 Alkaloids Derived from Terpene Precursors
Various alkaloids derived from terpene precursors have been reported in the literature (Figure 8.10). These include several monoterpenoid alkaloids (such as gardenamide [72]), a large number of diterpenoid alkaloids [73] (e.g., barbaline),
293
294
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin O H
OMe
NH HO
H
CO2Me H
OAc AcO OH Me Ac O O O Ac N Me O
O
gardenamide (Rothmania urcelliformis)
N
barbaline (Delphinium barbeyi)
Me H
H
Me
H
daphniglaucine A (Daphniphyllum glaucescens)
Me
Me N O H H
HO H
OH
Me
H
solasodine (Solanum sp.)
Figure 8.10
Examples of terpenoid alkaloids.
numerous triterpenoid alkaloids, either steroidal [74] (such as solasodine) or non-steroidal (such as Daphniphyllum alkaloids, vide infra [75]), and others. The following sections are devoted to the bis-steroidal alkaloids of the cephalostatin/ritterazine family and to the efforts developed around the biomimetic synthesis of Daphniphyllum alkaloids. 8.5.1 Cephalostatins and Ritterazines
Even though they were not isolated from the same marine organisms, cephalostatins and ritterazines are two structurally related groups of bis-steroidal pyrazine alkaloids [76]. Cephalostatins were extracted from the marine tubeworm Cephalodiscus gilchristi (Cephalodiscidae) collected off the coasts of southeastern Africa by the group of Pettit in the early 1970s. It took 15 years before the authors were able to isolate and elucidate the structure of cephalostatin 1 (102, Figure 8.11) [77], and in the following decade 18 more cephalostatins were reported by the same group [78]. Ritterazines, on the other hand, were extracted from the marine tunicate Ritterella tokioka (Polyclinidae) collected by Fusetani et al. off the coast of Japan. The first member of the family, ritterazine A (103), was reported in 1994 [79] and, in subsequent years, 25 other ritterazines were described [80]. The two groups share the same general structure consisting of two hexacyclic steroidal spiroketal units (one of which is called South and the other North)4) attached by a central pyrazine ring (Figure 8.11). Most of the differences between cephalostatins are found in the outer CDEF rings and especially in the spiroketal 4) South and North hemispheres refer to
the Fuchs nomenclature. Pettit calls these regions Left and Right, whereas Fusetani speaks of West and East regions. The numbers or letters that follow these denominations correspond to the alkaloids
to which the regions belong (e.g., South 1 refers to the southern region of cephalostatin 1 and North A refers to the northern region of ritterazine A) as depicted on Figure 8.11.
8.5 Alkaloids Derived from Terpene Precursors Me North 1
Me
H H
B'
D'
C'
H E' O
A'
Me
A
N
E
H
C B
D
H
F'
O OH
N
HO
H Me
N
H
O
O O
Me Me
OH
Me
Figure 8.11
Me
O
Me H
Me
N
H
O
Me
H
ritterazine A (103)
Me OH Me
OH
H
OH
O
H
Me
N
Me
H
HO O
Me
Me
Me OH
OH
H
cephalostatin 5
O Me OH H Me H O
South A
OH
H Me
H
O
Me
N
H
HO
H
Me
H
HO
OH
Me Me O OH Me H
North A
F
cephalostatin 1 (102) O Me OH South 1
O Me Me
H
Me
N
OH
O
OH Me
295
Me OH
H
H N
Me
H
ritterazine M
OH
Structures representative of cephalostatins and ritterazines.
end regions. Ritterazines also show little variability in their AB rings but can have different CD junction patterns (either side-fused 6,5 or spiro-fused 5,5). The striking structural similarities along with the fact that the two groups of alkaloids were isolated from two animals of different phyla (i.e., Hemicordata for Cephalodiscus sp. and Chordata for Ritterella sp.) suggest that the biosynthesis of these compounds might actually find its origin in a marine microorganism ingested by both C. gilchristi and R. tokioka [78h, 80a]. The cytotoxic activities displayed by both cephalostatins [77, 78] and ritterazines [79, 80] are very impressive. Cephalostatin 1, for instance, is one of the most potent anticancer agents tested to this day, with femtomolar activity against the P388 cell line. Ritterazines globally seem to be a little less active than cephalostatins but some of them nonetheless show subnanomolar activities against the P388 cell line as well (e.g., ritterazine B). Since the structure of cephalostatin 1 was disclosed in the late 1980s, the cephalostatin/ritterazine family has been the subject of numerous synthetic efforts [81]. The attractive structures and biological activities, along with the fact that only minute amounts of these alkaloids can be isolated from their natural sources, have prompted chemists to develop efficient synthetic routes to complete their evaluation as drug candidates. The group of Fuchs has had a prominent role in the chemistry of the cephalostatin/ritterazine family and has developed biomimetic approaches to different regions of the molecules. In 2005, Lee and Fuchs proposed biosynthetic pathways
OH O O
Me
296
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin
to both the southern region of cephalostatin 7 and the northern region of cephalostatin 1 [82]. This hypothesis postulates that hecogenin (104) can be modified to give 105, which then undergoes electrophilic spiroketal ring opening followed by elimination to produce 106. Dihydroxylation of the C25=C26 olefin, [4+2]-cycloaddition of singlet oxygen on the cyclopentadiene moiety and endoperoxide cleavage will then convert 106 into 107. The latter can, at that point, undergo an acid-catalyzed cyclization cascade by attack of either the C26 or the C25 hydroxyl on the C22 ketone to yield South 7 or North 1, respectively (Scheme 8.25). E O Me
H
Me
O O
H
H HO
Me
OH Me
Me reduction oxidation
Me
H
R O
Me
O
H
Me
H HO
hecogenin (104)
R OE
Me OH H Me
Me
Me O
H
H
H
HO
105
H
26
R
Me OH Me Me
25
Me
Me
Me
OH Me
O
H
EO R
Me
H O
H
H HO
H 106
H
HO
H
O2
HO Me OH HO Me Me H HO
H
H
(a)
OH O O
Me OH HO Me 22
Me (a)
Me
R=H
South 7
R
HO
H H
Me
Me
OH O
OH HO Me
(b)
H
(b)
Me
R = OH
OH H
HO
26 25
107
HO
H
H
MeO O
OH
North 1
H
Scheme 8.25 Biosynthetic hypothesis for the formation of South 7 and North 1 (E = electrophile).
Inspired by this biosynthetic scenario, the authors designed a synthesis that would reproduce the major steps described above. They started from 108 [obtained in 75% yield from hecogenin (104)], which underwent reductive spiroketal opening followed by iodination and elimination to afford 109 in 81% yield. Dihydroxylation by the Sharpless protocol and subsequent selective protection of the primary C26 alcohol followed by benzoylation of the C25 hydroxyl gave 110 in 94% yield over the three steps. Tetrahydrofuran opening upon treatment with trifluoroacetyl triflate (TFAT), removal of the trifluoroacetate under mild basic conditions, oxidation of the obtained hydroxyl into the corresponding ketone, and protection of the latter with 1,3-propanediol afforded 111 in 81% yield. Cycloaddition of singlet oxygen to the cyclopentadiene moiety of the latter took place stereospecifically to give only the α-face adduct and the endoperoxide was reductively cleaved by treatment with activated zinc and acetic acid to give 112 in 83% yield. In situ
8.5 Alkaloids Derived from Terpene Precursors
297
generated hydrogen cyanide unexpectedly led to the formation of hydroxypropyl ether 114 in 86% yield, presumably via formation of transient oxonium 113. After oxidation of the primary alcohol of 114 to the corresponding aldehyde, reaction with tetraisobutylammonium fluoride (TBAF) permitted the concurrent cleavage of the silyl ether and acrolein to directly furnish tetrahydrofuran hemiacetal 115 in 84% yield. Treatment of the latter with pyridinium p-toluenesulfonate (PPTS) completed the cyclization sequence and afforded a protected version of the southern hemisphere of cephalostatin 7 in 66% yield5) (Scheme 8.26) [82]. TBDPSO
26
Me OBz
25
Me
O
4 steps BzO Me ca. 75%
Me
Me 1) Et3SiH, BF3. OEt2
H
Me
(81% over 3 steps)
H AcO
BzO
2) PPh3, I2, imidazole 3) DBU, 90 °C
O
H
O
H
BzO
1) AD-mix 2) TBDPSCl, NEt3 3) Bz2O, MgBr2.OEt2
Me
Me
(94% over 3 steps)
H
AcO
108
Me
Me
H
(81% over 4 steps)
TBDPSO H2O BzO Me
OBz
DDQ, MeCN/H2O
OH O O
H
Me
AcO
H
H
Me OBz
Me Me
BzO 1) O2, sunlamp, TPP 2) Zn, AcOH
Me
(83% over 2 steps)
OH
H
113
110 1) TFAT 2) Na2CO3 3) DMSO, TFAA 4) diol, Sc(OTf)3
OBz
OO
H
O
TBDPSO
Me
Me BzO HO Me
Me Me
H AcO
TBDPSO
Me
Me
H
H
AcO
109
H
Me
H
AcO
112
OO
H
H
111
(86%)
TBDPSO
HO
Me OBz
BzO
Me Me
Me
H
OH O O
Me
(84%)
H AcO
H
Scheme 8.26
O
H
114
AcO
H
OH
PPTS
Me
(66%)
H OH
BzO
Me BzO HO Me
1) TPAP, NMO, 4 Å MS 2) TBAF
Me OBz
H AcO
115
Me HO Me
O
H
South 7
H
Biomimetic synthesis of South 7 [82].
This biomimetically inspired synthesis (carried out in 16 steps and 24% overall yield from 108) is a great improvement on the previously published method [83] (25 steps, 2% overall yield) for the preparation of South 7 (which is found in 18 of the 45 members of the cephalostatin/ritterazine family). The same strategy could be applied to the synthesis of North 1 but to date no report of progress in this direction has been disclosed. The Fuchs group has also studied the biomimetic coupling of the northern and southern units [84], following the hypothesis that Nature might utilize the random 5) The reaction actually has a yield of 81%,
but 15% corresponds to the C25 epimer of
OBz O
South 7 carried onward from the Sharpless dihydroxylation reaction.
25
Me
298
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin
coupling of different steroidal α-aminoketones to generate various bis-steroidal pyrazines [78c,h]. In this work, the authors reduced α-azidoketones 116 and 117, to generate α-aminoketones 118 and 119 in situ, by treatment with ethanolic NaHTe for an hour. They then added silica gel and exposed the mixture to the air and, 18 h later, observed the formation of heterodimer 120 and homodimers 121 and 122 in 35%, 23%, and 14% yields respectively. Upon deprotection with an excess of TBAF in refluxed THF, the corresponding cephalostatin 7, ritterazine K, and cephalostatin 12 were each obtained in about 80% yield (Scheme 8.27) [84b]. This biomimetic one-pot process led to the simultaneous syntheses of cephalostatin 7, cephalostatin 12, and ritterazine K and provided the first synthetic samples of these molecules. Considering that the most biologically active bis-steroidal pyrazines are the unsymmetrical ones, different groups have developed methods specifically designed to avoid the formation of symmetrical dimers. The Heathcock group disclosed their approach based on the reaction of 2-acetoxy-3-ketones (123) with 2-amino-3-methoximes (124) in 1992 [85]. In the following years, Guo and Fuchs proposed a similar approach based on the reaction of 2-azido-3-ketones (125) with 2-amino-3-methoximes [86] and the group of Winterfeldt also published a different strategy involving vinyl azides and α-aminoketones [87a], which was more recently improved to a simpler approach relying on the reaction between 2-hydroxy-3-ketones (126) and 2-amino-3-ketones (127) in the presence of ammonium acetate [87b] (Figure 8.12). 8.5.2 Daphniphyllum Alkaloids
At the beginning of the twentieth century, Yagi reported the isolation of daphnimacrine, a complex alkaloid from a common Japanese tree, Daphniphyllum macropodum (Daphniphyllaceae) also known as Yuzuriha [88]. At that time, the available spectroscopic techniques did not permit elucidation of the molecule’s structure.6) Almost 60 years later, Hirata and Yamamura initiated research on Daphniphyllum alkaloids and, since then, different groups have reported the structures of more than a 100 representatives of this growing family of natural compounds 6) The structure of daphnimacrine was elu-
cidated by X-ray analysis and reported in 1968 [89]. OAc Me
O O
Me daphnimacrine
Me Me
N
Me
8.5 Alkaloids Derived from Terpene Precursors H
HO
H
O
N3
Me
O
H
Me 116
OH
O OTBDPS
Me
N3
Me OAc
OTBDMS
Me O OAc Me
H
TMSO
Me
Me
O
O
OH
H
117
H
NaHTe
H
HO TMSO H
O
O OTBDPS
H O
OTBDMS
O
Me
H2N
Me 118
OH
Me OAc Me
NH2
Me
Me OAc
O
Me
Me
O
H
OH
H
119
H
SiO2, O2
HO
Me
OTBDMS
OTBDMS Me
Me O Me AcOMe
O
O
Me
O
O
OTMS
H
AcOMe
HO
OH
Me
H
Me
O
TMSO
Me 121 (23%)
O
H
Me
Me O
Me
H
Me OAc
O
H N
H
H
Me OAc
H
N
N
H
H TMSO
Me
H H
H HO
H
H
N
N
OTBDPS
OH
OH
H
N H
O
AcO Me
H Me
Me
OTBDPS
120 (35%)
HO Me TBDPSO
OAc
O Me O
122 (14%)
Me Me
TBAF (ca. 80%)
OH
ritterazine K
Scheme 8.27
Me
OH
TBAF (ca. 80%)
cephalostatin 7
TBDMSO
TBAF (ca. 80%)
cephalostatin 12
Biomimetic pseudo-combinatorial approach to bis-steroidal pyrazines [84b].
isolated from plants of the Daphniphyllum genus found in Japan, China, Taiwan, and New Guinea [75]. Some of these alkaloids have been shown to possess biological activities, including platelet aggregation inhibition [90], antioxidant activity [91], modulation of the expression of nerve growth factor (NGF) [92], and cytotoxicity [93]. Upon discovery, Daphniphyllum alkaloids have been classified into different structural types with respect to their backbone skeleton (Figure 8.13 below presents
299
300
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin Guo's method (1996)
Heathcock's method (1992) H H
O
Me
H2N
H
H
+ OAc
Me
H
N OMe
H
H
H
+ N3
Me
124
Me
H2N
H
H
123
O
N OMe
H
H
125 Winterfeldt's method (2004) H H
O
H
+ OH
Me
O
H
H
NH4OAc
126
Figure 8.12
Me
H2N
H
127
Different approaches to unsymmetrical bis-steroidal pyrazines.
Me O
Me O O
O O
Me
O
Me
AcO Me Me
N
Me
Me HN O
O
O Me
Me Me
daphniphylline
secodaphniphylline
HO
O Me
Me
O
N
O
Me N
Me
Me
O daphnilactone A
O
O
N
H
Me
daphnicyclidin A
N
daphnilactone B bukittinggine CO2Me H OH OAc Me AcO
N
yuzurimine
Me
OMe
CO2Me H
O Me N
CO2Me HO
O
H OAc
N Me O daphnezomine F
yuzurine
Figure 8.13 Examples of Daphniphyllum alkaloids, representative of their structural diversity.
major alkaloids representative of different structural types). Despite the chemical diversity of these often strikingly complex molecules, it has been proposed that they all arise from a common terpene precursor derived from squalene [94]. Since the 1980s, the Heathcock group has had a prominent role in the synthesis of Daphniphyllum alkaloids [95]. They have reported several biomimetic approaches based on their own biosynthetic hypothesis for the formation of a postulated pentacyclic intermediate, named proto-daphniphylline, that is likely to be common
8.5 Alkaloids Derived from Terpene Precursors Me
Me
Me
Me
Me
Me Me
NH2 Me
O
128
129
R'
O
Me
N
Me Me
prototropic rearrangement
− H2O
oxidation
squalene
R
O
R'
R Me
Me
1-aza diene
Me
R R R O HN
H N
R
R OH N
133
R'
O
R
R 1,4-addition
132 H N
O HN
N
NH2
R
R'
R'
2-aza diene
Me Me Me
Me
R R N 134
Me Me Me
Me
Me
Me
Me
Diels− Alder
Me N
Me ene-like reaction
Me
Me
N 135
Scheme 8.28
130
R'
131 H N
Me
Me HN
proto -daphniphylline (136)
Heathcock’s hypothesis for the biosynthesis of proto-daphniphylline.
to many types of Daphniphyllum alkaloids (Scheme 8.28) [96]. According to this scenario, squalene is oxidized into dialdehyde 128, which, upon condensation with a primary amine (e.g., amino acid or pyridoxamine), gives rise to 1-azadiene (129) that then isomerizes by a prototropic rearrangement to the corresponding 2-azadiene (130). The enimine double bond of the later is not very nucleophilic and the authors propose that, at that point, another nucleophilic species reduces the imine double bond to yield nucleophilic enamine 131. An intramolecular Michael addition of the enamine on the conjugated aldehyde gives monocyclic 132, which is then converted into 133 by hemiaminal ether formation. Proton-mediated addition and elimination processes allow the conversion of the bicyclic dihydropyran intermediate into a dihydropyridine derivative 134, which can undergo an IMDA reaction and yield tetracyclic intermediate 135. Finally, an ene-like reaction enables the formation of the fifth ring to form proto-daphniphylline (136). This hypothesis has been supported by the successful biomimetic total syntheses of different alkaloids that Heathcock et al. have managed to complete over the years. In 1988, they reported the synthesis of methyl homosecodaphniphyllate (137) [97, 98] via a tetracyclization reaction cascade from monocyclic diol 138 (obtained in 60% yield over eight steps). Under Swern oxidation conditions, 138 was converted into dialdehyde 139 (synthetic equivalent of 132) and upon subsequent addition of gaseous ammonia, replacement of the solvent with acetic acid, and heating
301
302
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin
at 70 ◦ C for 1.5 h, pentacyclic product 140 was formed in 77% overall yield. Removal of the benzyl protecting group and reduction of the terminal double bond were conducted simultaneously by hydrogenation over Pd/C. Jones oxidation of the obtained primary alcohol into the corresponding carboxylic acid followed by Fischer esterification then afforded (±)-methyl homosecodaphniphyllate (137) in 85% yield over the three steps (Scheme 8.29). Key biomimetic tetracyclization reaction
OBn
OBn
H
HO 8 steps HO ca. 60%
138
BnO
H
O (COCl)2, DMSO NEt3, CH2Cl2
1) NH3 2) AcOH 70 °C
O
Me
Me
(77%)
Me
139
Me
HN
Me
140
Me Me
Me
MeO methyl homosecodaphniphyllate (137) Me
O
1) H2, Pd/C, HCl 2) CrO3, H2SO4, H2O 3) MeOH, H2SO4
Me (85%)
Me
Scheme 8.29
HN
Heathcock’s synthesis of methyl homosecodaphniphyllate [98].
The key tetracyclization reaction, during which six bonds and four rings are formed, proceeds with impressive efficacy. In subsequent years, this strategy allowed the authors to achieve the total syntheses of other Daphniphyllum alkaloids, including secodaphniphylline [99, 100], bukittinggine7) (141) [101, 102], methyl homodaphniphyllate (142) [99, 103], daphnilactone A (143) [103b–105], and codaphniphylline [106, 107]. In the case of bukittinggine [102], the tetracyclization process was carried out from precursor 144 (benzyloxy derivative of 138, obtained in 12% yield over twelve steps) and gave pentacyclic 145 in 74–78% yield (Scheme 8.30). A palladium-assisted oxidative cyclization then permitted the formation of hexacyclic intermediate 146 in 70% yield. The latter was submitted to hydroboration-oxidation 7) Bukittinggine was isolated from Sapium
baccatum (Euphorbiaceae) [101a] but is considered a member of the Daphniphyllum alkaloids because of obvious structural similarities with alkaloids such as methyl
homosecodaphniphyllate and daphnilactone B. Moreover, three bukittinggine-type alkaloids have been isolated recently [101b] from Daphniphyllum calycinum:
O
OMe
O
O Me N R
R = H: bukittiggine R = OH: caldaphnidine N
Me
Me N R
R = H: caldaphnidine O R = OH: caldaphnidine P
8.5 Alkaloids Derived from Terpene Precursors OBn 1) Swern 2) NH3 3) AcOH 75 °C
H
HO 12 steps HO ca. 12%
OBn
BnO Me HN
(74-78%)
144
BnO
(CF3CO2)2Pd benzoquinone PPh3 OBn
OBn N
(70%)
145
146
Me Me
1) BH3 then NaBO3 2) p -TsCl, pyridine 3) LiEt3BH 4) Na − NH3 5) Ag2CO3 −Celite
O bukittinggine (141)
O
Me (52%)
N
Scheme 8.30
Heathcock’s synthesis of bukittinggine [102].
followed by tosylation and reduction to allow the conversion of the exocyclic double bond into the β-methyl derivative (along with negligible amounts of its α-methyl isomer). Removal of the benzyl protecting groups with sodium in condensed ammonia followed by oxidative lactonization with silver carbonate on Celite (F´etizon’s reagent) completed the synthesis of (±)-bukittinggine (141) in 52% yield from 146. The syntheses of methyl homodaphniphyllate (142) and daphnilactone A (143) [103b] gave another example of biomimetic tetracyclization reaction, but also provided arguments in favor of the hypothesis according to which the secodaphnane skeleton might be the initial biosynthetic skeleton of Daphniphyllum alkaloids [96a, 108] (despite the scarce occurrence of the related alkaloids compared to those of the daphnane type). In a plausible scenario [109], the secodaphnane skeleton (147) can be converted into intermediate 148 by a fragmentation process initiated by oxidation either at C9 (149a, path A in Scheme 8.31) or on the nitrogen atom oxidation on C9
Me Me
Me
fragmentation path A
R
R Me
Me
HN
R Me 9 X
149a
HN
Me
147 secodaphnane skeleton oxidation on nitrogen
Me Me
X
N
149b
R
Me
R Me H
N
Me
reduction
148
Me Me
Me HN
150
fragmentation path B
R
Me Me
Me N
151 daphnane skeleton
Scheme 8.31 Plausible mechanism for the conversion of the secodaphnane skeleton into the daphnane skeleton [109].
303
304
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin
(149b, path B in Scheme 8.31). Reduction of the imine to secondary amine 150 and nucleophilic attack of the nitrogen on the double bond can then lead to the daphnane skeleton (151). For the syntheses of 142 and 143 [103b], cyclic ether 152 (obtained in 73% yield over six steps) was submitted to the tetracyclization conditions described above and gave hexacyclic intermediate 153 (synthetic equivalent of 149a) in 47% yield (Scheme 8.32). Selective reduction of the terminal double bond by hydrogenation over Adam’s catalyst followed by treatment with diisobutylaluminum hydride (DIBAL-H) provided fragmentation product 154 (synthetic equivalent of 150) in 70% yield for the two steps. In this process, DIBAL-H serves both as activator for the alkoxy leaving group and as reducing agent for the transient iminium ion. Intermediate 154 can then lead to two different skeletons: Jones oxidation of the primary hydroxyl into the corresponding carboxylic acid followed by treatment with aqueous formaldehyde at pH 7 yielded (±)-daphnilactone A (143) in 50% yield, whereas successive treatment of 154 with phenyl isothiocyanate, refluxing formic acid, and methanolic potassium hydroxide furnished the daphnane skeleton-bearing intermediate 155. Oxidation of the hydroxyl followed by Fischer esterification completed the synthesis of (±)-methyl homodaphniphyllate (142) in 70% yield over the last five steps. O Me
O Me
Me
N
daphnilactone A (143)
(50%)
HO
O
1) Swern 2) NH3 3) AcOH 45 °C Me
HO 6 steps ca. 73%
Me
Me
152
HN
(47%)
Me O
OH
1) H2, PtO2 2) DIBAL-H MePh, reflux Me (70%)
Me
Me
HN
153
154 1) PhNCO 2) HCO2H, reflux 3) KOH, MeOH
Me MeO
O 4) CrO3, H2SO4 5) MeOH, H2SO4
Me Me
N
1) CrO3, H2SO4 2) CH2O, pH7
Me
(70%, 5 steps)
HO Me Me
methyl homodaphniphyllate (142)
N
Me
155
Scheme 8.32 Heathcock’s syntheses of methyl homodaphniphyllate and daphnilactone A [103b].
These results are in good agreement with the postulated biosynthetic conversion of the secodaphnane skeleton into the daphnane skeleton through oxidation at C9 (path A on Scheme 8.31). The relevance of the N-oxidation-mediated fragmentation
8.6 Conclusion
scenario (path B on Scheme 8.31) was also studied by the authors [109] but will not be detailed here. In the early 1990s [110], the Heathcock group reported results concerning the biomimetic total synthesis of proto-daphniphylline (136) in which they took their key polycyclization reaction one step further. In this work they achieved the pentacyclization of dialdehyde 156 (partially reduced equivalent of biosynthetic intermediate 128, see Scheme 8.28), directly into 136, under conditions similar to those described above for the tetracyclization process (i.e., treatment with gaseous ammonia followed by warm acetic acid), but in rather low yield (15%). Surprisingly, a few years later [95a], the authors reported that under the same conditions, but with methylamine (which was used by accident) in place of ammonia, the pentacyclization took place and afforded compound 157 (reduced equivalent of 136) in 65% overall yield (Scheme 8.33). Me
Me
Me
Me
Me
Me Me
Me
1) NH3 2) AcOH, ∆
Me
Me
proto -daphniphylline (136)
Me
1) MeNH2 2) AcOH, ∆
O Me
(15%)
Me
HN
Scheme 8.33
Me
(65%)
O
156
Me Me
Me HN 157
Pentacyclization of 156 into either 136 or 157 [95a].
The difference in efficacy between the tetracyclization and pentacyclization processes, when carried out with ammonia, points out the formation of the first cyclopentane ring as the limiting reaction. When methylamine is used instead of ammonia, the secondary enamine 158b that is formed after condensation with 156 is certainly more nucleophilic than primary enamine 158a (Scheme 8.34), thus resulting in much higher overall yield. As for the spontaneous reduction of the isoprenyl group, it is probably due to the fact that after the Prins-like reaction converting 159 into 160, the reduction of the carbocation leading to 161 occurs via a 1,5-hydride shift instead of a 1,5-proton transfer as proposed for the conversion of 162 into 136-H+ (Scheme 8.34). The authors pointed out that, since no isoprenyl-bearing Daphniphyllum alkaloid has ever been reported, the source of the biosynthetic nitrogen is very likely to be an alkylamine such as pyridoxamine or an amino acid [95a]. During the conversion of 156 into 157, seven bonds and five rings are formed in a fully diastereoselective way and one out of three similar double bonds is regioselectively saturated. 8.6 Conclusion
Throughout this short survey we have seen that biomimetic strategies can be applied successfully to alkaloids of non-amino acid origin. In some cases, the
305
306
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin Me
Me
Me
Me
Me
Me
Me
Me Me
156 O
Me Me
−H2O
Me
Me R
Me Me
1,4-addition
Me
Me
O
O
NH2
N R H
O
Me
limiting step
Me −H2O
Me
Me
Me N R
Me
Me R
R = H: 158a R = Me: 158b Me
formal Diels−Alder reaction
Me
Me
Me
Me
N
R = Me
Me
Me
Me
Me R=H
Me Me
Me
1,5-hydride shift
Me
Me
Me H
N
H 161
Prins-like reaction Me
Me
N H
160
Me Me
N
Me Me Me
159 Me
H2O
Me
Me
Me
Me
Me
Me
Me
N
Me
Me
Me
HN
H2N
157
136-H
Scheme 8.34
Me H
Me
Me
Me
Me
Me
1,5-proton transfer
Me H
+
Me
Prins-like reaction
HN 162
Proposed pentacyclization mechanism.
strategies provided by biosynthetic hypotheses have been applied to total syntheses of the target molecules, sometimes drastically improving the yields of delicate transformations and, most of all, the ease of the global synthetic process. In other cases, biomimetic strategies have been shown to be useful for the formation of particular regions of interest, which often display structural complexity or originality. Interestingly, in nearly all studies presented above, the key biomimetic steps concern the nitrogen-containing moieties of the molecules (the only exception is found in the biomimetic transformations of the cephalostatin/ritterazine spiroketal end). In these examples, the nitrogen atom is either involved directly in the chemical transformations or serves as an activator for cycloaddition processes. Many of the efforts presented in this chapter are only models and still need to be confirmed or taken a few steps further by completing the syntheses or evaluating their relevance with appropriate substrates. Moreover, several families of alkaloids of non-amino acid origin have not (or very scarcely) been studied in terms of biomimetic synthesis yet and would definitely deserve to be. This is especially true for terpene-derived molecules, such as diterpenoid alkaloids, which can display
References
impressive complex structures (e.g., Delphinium alkaloids; see the structure of barbaline, Figure 8.10). Another issue that, in many cases, needs to be dealt with is the biosynthetic origin of the nitrogen atom(s) found in the molecules. The work of the Heathcock group on Daphniphyllum alkaloids regarding this question stands as an exception and the findings were, to some extent, the fruit of thorough analysis of accidental results. References 1. Brown, R.F.C., Drummond, R.,
2.
3.
4.
5.
6.
7.
Fogerty, A.C., Hughes, G.K., Pinhey, J.T., Ritchie, E., and Taylor, W.C. (1956) Aust. J. Chem., 9, 283–287. (a) Pinhey, J.T., Ritchie, E., and Taylor, W.C. (1961) Aust. J. Chem., 9, 106–134; (b) Binns, S.V., Dunstan, P.J., Guise, G.B., Holder, G.M., Hollis, A.F., McCredie, R.S., Pinhey, J.T., Prager, R.H., Rasmussen, M., Ritchie, E., and Taylor, W.C. (1965) Aust. J. Chem., 18, 569–573; (c) Mander, L.N., Prager, R.H., Rasmussen, M., Ritchie, E., and Taylor, W.C. (1967) Aust. J. Chem., 20, 1473–1491; (d) Mander, L.N., Prager, R.H., Rasmussen, M., and Ritchie, E. (1967) Aust. J. Chem., 20, 1705–1718. Ritchie, E. and Taylor, W.C. (1967) in The Alkaloids, vol. IX (eds R.H.F. Manske and H.L. Holmes), Academic Press, New York, pp. 529–544. The stereochemistry of Class II and Class II alkaloids has recently been revised. See: Willis, A.C., O’Connor, P.D., Taylor, W.C., and Mander, L.N. (2006) Aust. J. Chem., 59, 629–632. Mander, L.N., Willis, A.C., Hertl, A.J., and Taylor, W.C. (2009) Tetrahedron Lett., 50, 7089–7092. (a) Tchabanenko, K., Adlington, R.M., Cowley, A.R., and Baldwin, J.E. (2005) Org. Lett., 7, 585–588; (b) Tchabanenko, K., Chesworth, R., Parker, J.S., Anand, N.K., Russell, A.T., Adlington, R.M., and Baldwin, J.E. (2005) Tetrahedron, 61, 11649–11656. Movassaghi, M., Hunt, D.K., and Tjandra, M. (2006) J. Am. Chem. Soc., 128, 8126–8127 (see Supporting Information for full biosynthetic hypothesis).
8. Kozikowski, A.P., Fauq, A.H., Miller,
9.
10.
11.
12.
J.H., and McKinney, M. (1992) Bioorg. Med. Chem. Lett., 2, 797–802. (a) Hart, D.J., Wu, W.-L., and Kozikowski, A.P. (1995) J. Am. Chem. Soc., 117, 9369–9370; (b) Hart, D.J., Li, J., Wu, W., and Kozikowski, A.P. (1997) J. Org. Chem., 62, 5023–5033; (c) Chackalamannil, S., Davies, R.J., Asberom, T., Doller, D., and Leone, D. (1996) J. Am. Chem. Soc., 118, 9812–9813; (d) Chakalamannil, S., Davies, R.J., Wang, Y., Asberom, T., Doller, D., Wong, J., and Leone, D. (1999) J. Org. Chem., 64, 1932–1940; (e) Takadoi, M., Katoh, T., Ishiwata, A., and Terashima, S. (1999) Tetrahedron Lett., 40, 3399–3402; (f) Takadoi, M., Katoh, T., Ishiwata, A., and Terashima, S. (2002) Tetrahedron, 58, 9903–9923; (g) De Baecke, G., and De Clercq, P.J. (1995) Tetrahedron Lett., 36, 7515–7518; (h) Hofman, S., De Baecke, G., Kenda, B., and De Clercq, P.J. (1998) Synthesis, 479–489; (i) Hofman, S., Gao, L.-J., Van Dingenen, H., Hosten, N.G.C., Van Haver, D., De Clercq, P.J., Milanesio, M., and Viterbo, D. (2001) Eur. J. Org. Chem., 2851–2860; (j) Wong, L.S.-M. and Sherburn, M.S. (2003) Org. Lett., 5, 3603–3606. Movassaghi, M., Tjandra, M., and Qi, J. (2009) J. Am. Chem. Soc., 131, 9648–9650. O’Connor, P.D., Mander, L.N., and McLachlan, M.M.W. (2004) Org. Lett., 6, 703–706. (a) Zi, W., Yu, S., and Ma, D. (2010) Angew. Chem., Int. Ed. Engl., 49, 5887–5890; (b) Larson, K.K. and Sarpong, R. (2009) J. Am. Chem.
307
308
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin
13.
14.
15.
16.
17.
18.
Soc., 131, 13244–13245; (c) Evans, D.A. and Adams, D.J. (2007) J. Am. Chem. Soc., 129, 1048–1049; (d) Shah, U., Chackalamannil, S., Ganguly, A.K., Chelliah, M., Kolotuchin, S., Bulevich, A., and McPhail, A. (2006) J. Am. Chem. Soc., 128, 12654–12655; (e) Mander, L.N. and McLachlan, M.M. (2003) J. Am. Chem. Soc., 125, 2400–2401. For a recent review, see: (a) Gu´eret, S.M. and Brimble, M.A. (2010) Nat. Prod. Rep., 27, 1350–1366; (b) Hu, T.M., Burton, I.W., Cembella, A.D., Curtis, J.M., Quilliam, M.A., Walter, J.A., and Wright, J.L.C. (2001) J. Nat. Prod., 64, 308–312; (c) Falk, M., Burton, I.W., Hu, T., Walter, J.A., and Wright, J.L.C. (2001) Tetrahedron, 57, 8659–8665; (d) Hu, T.M., Curtis, J.M., Walter, J.A., and Wright, J.L.C. (1996) Tetrahedron Lett., 37, 7671–7674. (a) Torigoe, K., Murata, M., Yasumoto, T., and Iwashita, T. (1988) J. Am. Chem. Soc., 110, 7876–7877; (b) Hu, T.M., deFreitas, A.S.W., Curtis, J.M., Oshima, Y., Walter, J.A., and Wright, J.L.C. (1996) J. Nat. Prod., 59, 1010–1014. Lu, C.K., Lee, G.H., Huang, R., and Chou, H.N. (2001) Tetrahedron Lett., 42, 1713–1716. Kita, M., Kondo, M., Koyama, T., Yamada, K., Matsumoto, T., Lee, K.-H., Woo, J.-T., and Uemura, D. (2004) J. Am. Chem. Soc., 126, 4794–4795. (a) Maier, M. and Varseev, G. (2006) Angew. Chem., Int. Ed. Engl., 45, 4767–4771; (b) Sakai, E., Araki, K., Takamura, H., and Uemura, D. (2006) Tetrahedron Lett., 47, 6343–6345; (c) Snider, B.B. and Che, Q. (2006) Angew. Chem., Int. Ed. Engl., 45, 932–935; (d) Zou, Y., Che, Q., and Snider, B.B. (2006) Org. Lett., 8, 5605–5608; (e) Kim, J. and Thomson, R.J. (2007) Angew. Chem., Int. Ed. Engl., 46, 3104–3106; (f) Born, S., Bacani, G., Olson, E.E., and Kobayashi, Y. (2008) Synlett, 2877–2881; (g) Born, S. and Kobayashi, Y. (2008) Synlett, 2479–2482. Kita, M., Ohishi, N., Washida, K., Kondo, M., Koyama, T., Yamada, K.,
19.
20.
21.
22.
23.
24.
25.
26.
and Uemura, D. (2005) Bioorg. Med. Chem., 13, 5253–5258. Roush, W.R. (1992) in Comprehensive Organic Synthesis, vol. 5 (ed. B.M. Trost), Pergamon Press, Oxford, pp. 513–550. Burke, J.P., Sabat, M., Iovan, D.A., Myers, W.H., and Chruma, J.J. (2010) Org. Lett., 12, 3192–3195. (a) Uemura, D., Chou, T., Haino, T., Nagatsu, A., Fukuzawa, S., Zheng, S.-Z., and Chens, H.-S. (1995) J. Am. Chem. Soc., 117, 1155–1156; (b) Chou, T., Kamo, O., and Uemura, D. (1996) Tetrahedron Lett., 37, 4023–4026. (a) Takada, N., Umemura, N., Suenaga, K., Chou, T., Nagatsu, A., Haino, T., Yamada, K., and Uemura, D. (2001) Tetrahedron Lett., 42, 3491–3494; (b) Matsuura, F., Hao, J., Reents, R., and Kishi, Y. (2006) Org. Lett., 8, 3327–3330. Chou, T., Haino, T., Kuramoto, M., and Uemura, D. (1996) Tetrahedron Lett., 37, 4027–4030. Selwood, A.I., Miles, C.O., Wilkins, A.L., van Ginkel, R., Munday, R., Rise, F., and McNabb, P. (2010) J. Agric. Food Chem., 58, 6532–6542. (a) Takada, N., Umemura, N., Suenaga, K., and Uemura, D. (2001) Tetrahedron Lett., 42, 3495–3497; (b) Hao, J., Matsuura, F., Kishi, Y., Kita, M., Uemura, D., Asai, N., and Iwashita, T. (2006) J. Am. Chem. Soc., 128, 7742–7743. See, for example: (a) McCauley, J.A., Nagasawa, K., Lander, P.A., Mischke, S.G., Semones, M.A., and Kishi, Y. (1998) J. Am. Chem. Soc., 120, 7647–7648; (b) Matsuura, F., Peters, R., Anada, M., Harried, S.S., Hao, J., and Kishi, Y. (2006) J. Am. Chem. Soc., 128, 7463–7465; (c) Sakamoto, S., Sakazaki, H., Hagiwara, K., Kamada, K., Ishii, K., Noda, T., Inoue, M., and Hirama, M. (2004) Angew. Chem., Int. Ed., 43, 6505–6510; (d) Stivala, C.E., and Zakarian, A. (2008) J. Am. Chem. Soc., 130, 3774–3776; (e) Ishihara, J., Tojo, S., Kamikawa, A., and Murai, A. (2001) Chem. Commun., 1392–1393; (f) Suthers, B.D., Jacobs, M.F., and Kitching, W. (1998) Tetrahedron Lett.,
References
27.
28.
29.
30.
31.
32.
33.
34.
39, 2621–2624; (g) Nakamura, S., Kikuchi, F., and Hashimoto, S. (2008) Angew. Chem., Int. Ed., 47, 7091–7094. Seki, T., Satake, M., Mackenzie, L., Kaspar, H.F., and Yasumoto, T. (1995) Tetrahedron Lett., 36, 7093–7096. Kharrat, R., Servent, D., Girard, E., Ouanounou, G., Amar, M., Marrouchi, R., Benoit, E., and Molg´o, J. (2008) J. Neurochem., 107, 952–963. Stewart, M., Blunt, J.W., Munro, M.H.G., Robinson, W.T., and Hannah, D.J. (1997) Tetrahedron Lett., 38, 4889–4890. Miles, C.O., Wilkins, A.L., Stirling, D.J., and MacKenzie, A.L. (2000) J. Agric. Food Chem., 48, 1373–1376. Miles, C.O., Wilkins, A.L., Stirling, D.J., and MacKenzie, A.L. (2003) J. Agric. Food Chem., 51, 4838–4840. The efforts of the Romo group have led to the only total synthesis known to date: (a) Kong, K., Romo, D., and Lee, C. (2009) Angew. Chem., Int. Ed., 48, 7402–7405; (b) Kong, K., Moussa, Z., and Romo, D. (2005) Org. Lett., 7, 5127–5130; (c) Ahn, Y., Cardenas, G.I., Yang, J., and Romo, D. (2001) Org. Lett., 3, 751–754; (d) Yang, J., Cohn, S.T., and Romo, D. (2000) Org. Lett., 2, 763–766; For synthetic studies by other research groups, see: (e) Johannes, J.W., Wenglowsky, S., and Kishi, Y. (2005) Org. Lett., 7, 3997–4000; (f) Tsujimoto, T., Ishihara, J., Horie, M., and Murai, A. (2002) Synlett, 399–402; (g) Ishihara, J., Miyakawa, J., Tsujimoto, T., and Murai, A. (1997) Synlett, 1417–1419; (h) White, J.D., Quaranta, L., and Wang, G. (2007) J. Org. Chem., 72, 1717–1728; (i) White, J.D., Wang, G., and Quaranta, L. (2003) Org. Lett., 5, 4983–4986; (J) White, J.D., Wang, G., and Quaranta, L. (2003) Org. Lett., 5, 4109–4112. Morita, H., Oshimi, S., Hirasawa, Y., Koyama, K., Honda, T., Ekasari, W., Indrayanto, G., and Zaini, N.C. (2007) Org. Lett., 9, 3691–3693. Matsumoto, T., Kobayashi, T., Ishida, K., Hirasawa, Y., Morita, H., Honda, T., and Kamata, K. (2010) Biol. Pharm. Bull., 33, 844–848.
35. (a) Rudyanto, M., Tomizawa, Y.,
36. 37. 38.
39.
40.
Morita, H., and Honda, T. (2008) Org. Lett., 10, 1921–1922; (b) Morita, H., Tomizawa, Y., Deguchi, J., Ishikawa, T., Arai, H., Zaima, K., Hosoya, T., Hirasawa, Y., Matsumoto, T., Kamata, K., Ekasari, W., Widyawaruyanti, A., Wahyuni, T.S., Zaini, N.C., and Honda, T. (2009) Bioorg. Med. Chem., 17, 8234–8240. Yao, Y.S. and Yao, Z.J. (2008) J. Org. Chem., 73, 5221–5225. Murti, S. (1949) Proc. Indian Sci. Sect. A, 30, 107–112. (a) Garraffo, H.M., Spande, T.F., Daly, J.W., Baldessari, A., and Gros, E.G. (1993) J. Nat. Prod., 56, 357–373; (b) Spande, T.F., Jain, P., Garraffo, H.M., Pannell, L.K., Yeh, H.J.C., Daly, J.W., Fukumoto, S., Imamura, K., Tokuyama, T., Torres, J.A., Snelling, R.R., and Jones, T.H. (1999) J. Nat. Prod., 62, 5–21; (c) Jones, T.H., Gorman, J.S.T., Snelling, R.R., Delabie, J.H.C., Blum, M.S., Garraffo, H.M., Jain, P., Daly, J.W., and Spande, T.F. (1999) J. Chem. Ecol., 25, 1179–1193; (d) Daly, J.W., Garraffo, H.M., Jain, P., Spande, T.F., Snelling, R.R., Jaramillo, C., and Rand, A.S. (2000) J. Chem. Ecol., 26, 73–85. (a) Steffan, B. (1991) Tetrahedron, 47, 8729–8732; (b) Kubanek, J., Williams, D.E., Dilip de Silva, E., Allen, T., and Andersen, R.J. (1995) Tetrahedron Lett., 36, 6189–6192; (c) Davis, R.A., Carroll, A.R., and Quinn, R.J. (2002) J. Nat. Prod., 65, 454–457; (d) Wright, A.D., Goclik, E., K¨onig, G.M., and Kaminsky, R. (2002) J. Med. Chem., 45, 3067–3072. (a) Witkop, B. and Gossinger, E. (1983) in The Alkaloids, vol. 21 (ed. A. Brossi), Academic Press, New York, pp. 139–253; (b) Daly, J.W. and Spande, T.F. (1986) in Alkaloids: Chemical and Biological Perspectives, vol. 4 (ed. S.W. Pelletier), John Wiley & Sons, Inc., New York, pp. 1–274; (c) Daly, J.W., Garraffo, H.M., and Spande, T.F. (1993) in The Alkaloids, vol. 43 (ed. G.A. Cordell), Academic Press, San Diego, pp. 185–288; (d) Daly, J.W. (1998) in The Alkaloids,
309
310
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin
41.
42.
43.
44.
vol. 50 (ed. G.A. Cordell), Academic Press, New York, pp. 141–169; (e) Daly, J.W., Garraffo, H.M., and Spande, T.F. (1999) in Alkaloids: Chemical and Biological Perspectives, vol. 13 (ed. S.W. Pelletier), Pergamon, New York, pp. 1–161; (f) Daly, J.W., Spande, T.F., and Garraffo, H.M. (2005) J. Nat. Prod., 68, 1556–1575. (a) Daly, J.W., Ware, N., Saporito, R.A., Spande, T.F., and Garraffo, H.M. (2009) J. Nat. Prod., 72, 1110–1114; (b) Daly, J.W., Spande, T.F., and Garraffo, H.M. (2005) J. Nat. Prod., 68, 1556–1575; (c) Tokuyama, T., Tsujita, T., Shimada, A., Garraffo, H.M., Spande, T.F., and Daly, J.W. (1991) Tetrahedron, 47, 5401–5414; (d) Tokuyama, T., Nishimori, N., Karle, I.L., Edwards, M.W., and Daly, J.W. (1986) Tetrahedron, 42, 3453–3460. Garraffo, H.M., Caceres, J., Daly, J.W., Spande, T.F., Andriamaharavo, N.R., and Andriantsiferana, M. (1993) J. Nat. Prod., 56, 1016–1038. (a) Daly, J.W., Kaneko, T., Wilham, J., Garraffo, H.M., Spande, T.F., Espinosa, A., and Donnelly, M.A. (2002) Proc. Natl. Acad. Sci. U.S.A., 99, 13996–14001; (b) Daly, J.W. (1995) Proc. Natl. Acad. Sci. U.S.A., 92, 9–13. Some recent achievements: (a) Voituriez, A., Ferreira, F., P´erez-Luna, A., and Chemla, F. (2007) Org. Lett., 9, 4705–4708; (b) Lee, C.-L.K. and Loh, T.-P. (2005) Org. Lett., 7, 2965–2967; (c) Holub, N., Neidh¨ofer, J., and Blechert, S. (2005) Org. Lett., 7, 1227–1229; (d) Davis, F.A. and Yang, B. (2005) J. Am. Chem. Soc., 127, 8398–8407; (e) Wrobleski, A., Sahasrabudhe, K., and Aub´e, J. (2004) J. Am. Chem. Soc., 126, 5475–5481; (f) Wrobleski, A., Sahasrabudhe, K., and Aub´e, J. (2002) J. Am. Chem. Soc., 124, 9974–9975; (g) Toyooka, N., Fukutome, A., Nemoto, H., Daly, J.W., Spande, T.F., Garraffo, H.M., and Kaneko, T. (2002) Org. Lett., 4, 1715–1717; (h) for a review on older syntheses of dendrobatid alkaloids, see: Kibayashi, C., Aoyagi, S. (1997) in Studies in Natural Products Chemistry, vol. 19 (ed. A.U. Rahman), Elsevier
45.
46. 47.
48.
Science B.V., Amsterdam, pp. 3–88; for biomimetic approaches to dendrobatid alkaloids, see: (i) Glanzmann, M., Karalai, C., Ostersehlt, B., Sch¨on, U., Frese, C., and Winterfeldt, E. (1982) Tetrahedron, 38, 2805–2810; (J) Bonin, M., Royer, J., Grierson, D.S., and Husson, H.-P. (1986) Tetrahedron Lett., 27, 1569–1572; for reviews on biomimetic syntheses, see: (k) Scholz, U. and Winterfeldt, E. (2000) Nat. Prod. Rep., 17, 349–366; (l) Nicolaou, K.C., Montagnon, T., and Snyder, S.A. (2003) Chem. Commun., 551–564; (m) De la Torre, M.C. and Sierra, M.A. (2004) Angew. Chem., Int. Ed. Engl., 43, 160–181. (a) Amat, M., Fabregat, R., Griera, R., Florindo, P., Molins, E., and Bosch, J. (2010) J. Org. Chem., 75, 3797–3805; (b) Amat, M., Fabregat, R., Griera, R., and Bosch, J. (2009) J. Org. Chem., 74, 1794–1797; (c) Amat, M., Griera, R., Fabregat, R., Molins, E., and Bosch, J. (2008) Angew. Chem., Int. Ed., 47, 3348–3351; (d) Escolano, C., Amat, M., and Bosch, J. (2006) Chem. Eur. J., 12, 8198–8207. Winterfeldt, E. (1979) Heterocycles, 12, 1631–1650. (a) Numerous enantioselective syntheses of cis-195A have been reported so far, see for example: (a) Oppolzer, W. and Flaskamp, E. (1977) Helv. Chim. Acta, 60, 204–207; (b) Murahashi, S.-I., Sasao, S., Saito, E., and Naota, T. (1993) Tetrahedron, 49, 8805–8826; (c) Comins, D.L. and Dehghani, A. (1993) J. Chem. Soc., Chem. Commun., 1838–1839; (d) Naruse, M., Aoyagi, S., and Kibayashi, C. (1996) J. Chem. Soc., Perkin Trans. 1, 1113–1124; (e) Riechers, T., Krebs, H.C., Wartchow, R., and Habermehl, G. (1998) Eur. J. Org. Chem., 2641–2646; (f) Dijk, E.W., Panella, L., Pinho, P., Naasz, R., Meetsma, A., Minaard, A.J., and Feringa, B.L. (2004) Tetrahedron, 60, 9687–9693. (a) Rao, C.B., Anjaneyula, A.S.R., Sarma, N.S., Venkatateswarlu, Y., Rosser, R.M., Faulkner, D.J., Chen, M.H.M., and Clardy, J. (1984) J. Am. Chem. Soc., 106, 7983–7984; (b) Rao,
References
49.
50.
51.
52.
53.
54.
55.
56.
C.B., Anjaneyula, A.S.R., Sarma, N.S., Venkatateswarlu, Y., Rosser, R.M., and Faulkner, D.J. (1985) J. Org. Chem., 50, 3757–3760; (c) Rao, C.B., Roa, D.V., and Raju, V.S.N. (1989) Heterocycles, 28, 103–106. (a) Fukuzawa, Y., Hayashi, Y., Uemura, D., Nagatsu, A., Yamada, Y., and Ijuin, Y. (1995) Heterocycl. Commun., 1, 207–214; (b) Kuramoto, M., Hayashi, K., Fujitani, Y., Yamaguchi, K., Tsuji, T., Yamada, K., Ijuin, Y., and Uemura, D. (1997) Tetrahedron Lett., 38, 5683–5686. (a) Daranas, A.H., Fern´andez, J.J., Gav´ın, J.A., and Norte, M. (1998) Tetrahedron, 54, 7891–7896; (b) Daranas, A.H., Fern´andez, J.J., Gav´ın, J.A., and Norte, M. (1999) Tetrahedron, 55, 5539–5546. See, for example: (a) Rahman, A.U., Alvi, K.A., Abbas, S.A., Choudhary, M.I., and Clardy, J. (1989) Tetrahedron Lett., 30, 6825–6828; (b) Fattorusso, E., Romano, A., Taglialatela-Scafati, O., Achmad, M.J., Bavestrello, G., and Cerrano, C. (2008) Tetrahedron Lett., 49, 2189–2192. Yamaguchi, K., Yada, M., Tsuji, T., Kuramoto, M., and Uemura, D. (1999) Biol. Pharm. Bull., 22, 920–924. Venkateswarlu, Y., Reddy, N.S., Ramesh, P., Reddy, P.S., and Jamil, K. (1998) Heterocycl. Commun., 4, 575–580. Villar, R.M., Gil-Longo, J., Daranas, A.H., Souto, M.L., Fern´andez, J.J., Peixinho, S., Barral, M.A., Santaf´e, G., Rodr´ıguez, J., and Jim´enez, C. (2003) Bioorg. Med. Chem., 11, 2301–2306. For a recent review, see: Behenna, D.C., Stockdill, J.L., and Stoltz, B.M. (2008) Angew. Chem., Int. Ed., 47, 2365–2386. (a) Miyashita, M., Sasaki, M., Hattori, I., Sakai, M., and Tanino, K. (2004) Science, 305, 495–499; (b) Sakai, M., Sasaki, M., Tanino, K., and Miyashita, M. (2002) Tetrahedron Lett., 43, 1705–1708; (c) Miyashita, M. (2007) Pure Appl. Chem., 79, 651–655; (d) Yoshimura, F., Sasaki, M., Hattori, I., Komatsu, K., Sakai, M., Tanino, K.,
57.
58.
59.
60.
61.
62.
63.
64.
and Miyashita, M. (2009) Chem. Eur. J., 15, 6626–6644. (a) Murata, Y., Yamashita, D., Kitahara, K., Minasako, Y., Nakazaki, A., and Kobayashi, S. (2009) Angew. Chem., Int. Ed., 48, 1400–1403; (b) Yamashita, D., Murata, Y., Hikage, N., Takao, K., Nakazaki, A., and Kobayashi, S. (2009) Angew. Chem., Int. Ed., 48, 1404–1406. (a) Hikage, N., Furukawa, H., Takao, K., and Kobayashi, S. (2000) Chem. Pharm. Bull., 48, 1370–1372; (b) Hikage, N., Furukawa, H., Takao, K., and Kobayashi, S. (1998) Tetrahedron Lett., 39, 6237–6240; (c) Hikage, N., Furukawa, H., Takao, K., and Kobayashi, S. (1998) Tetrahedron Lett., 39, 6241–6244. (a) Kuramoto, M., Hayashi, K., Yamaguchi, K., Yamada, K., Tsuji, T., and Uemura, D. (1998) Bull. Chem. Soc. Jpn., 71, 771–779; (b) Uemura, D. (2006) Chem. Rec., 6, 235–248. (a) McMahon, T. and Silke, J. (1996) Harmful Algae News, 14, 2; (b) for a review on azaspiracid shellfish poisoning see: Twiner, M.J., Rehmann, N., Hess, P., and Doucette, G.J. (2008) Mar. Drugs, 6, 39–72. Satake, M., Ofuji, K., Naoki, H., James, K.J., Furey, A., McMahon, T., Silke, J., and Yasumoto, T. (1998) J. Am. Chem. Soc., 120, 9967–9968. (a) Ofuji, K., Satake, M., McMahon, T., Silke, J., James, K.J., Naoki, H., Oshima, Y., and Yasumoto, T. (1999) Nat. Toxins, 7, 99–102; (b) Ofuji, K., Satake, M., McMahon, T., James, K.J., Naoki, H., Oshima, Y., and Yasumoto, T. (2001) Biosci. Biotechnol. Biochem., 65, 740–742; (c) James, K.J., Sierra, M.D., Lehane, M., Magdalena, A.B., and Furey, A. (2003) Toxicon, 41, 277–283. Furey, A., Moroney, C., Magdalena, A.B., Saez, M.J.F., Lehane, M., and James, K.J. (2003) Environ. Sci. Technol., 37, 3078–3084. (a) First thought to be Protoperidinium crassipes (Protoperidiniaceae), see: James, K.J., Moroney, C., Roden, C., Satake, M., Yasumoto, T., Lehane, M., and Furey, A. (2003) Toxicon, 41, 145–151; (b) but later corrected to
311
312
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin Azadinium spinosum (Peridiniphycidae), see: Tillmann, U., Elbracchter, M., Krock, B., John, U., and Cembella, A. (2009) Eur. J. Phycol., 44, 63–79. 65. Rehmann, N., Hess, P., and Quilliam, M.A. (2008) Rapid Commun. Mass Spectrom., 22, 549–558. 66. (a) Twiner, M.J., Hess, P., Dechraoui, M.Y.B., McMahon, T., Samons, M.S., Satake, M., Yasumoto, T., Ramsdell, J.S., and Doucette, G.J. (2005) Toxicon, 45, 891–900; (b) Colman, J.R., Twiner, M.J., Hess, P., McMahon, T., Satake, M., Yasumoto, T., Doucette, G.J., and Ramsdell, J.S. (2005) Toxicon, 45, 881–890; (c) Ronzitti, G., Hess, P., Rehmann, N., and Rossini, G.P. (2007) Toxicol. Sci., 95, 427–435; (d) Vilarino, N., Nicolaou, K.C., Frederick, M.O., Cagide, E., Ares, I.R., Louzao, M.C., Vieytes, M.R., and Botana, L.M. (2006) Chem. Res. Toxicol., 19, 1459–1466; (e) Kulagina, N.V., Twiner, M.J., Hess, P., McMahon, T., Satake, M., Yasumoto, T., Ramsdell, J.S., Doucette, G.J., Ma, W., and O’Shaughnessy, T.J. (2006) Toxicon, 47, 766–773; (f) Alfonso, A., Vieytes, M.R., Ofuji, K., Satake, M., Nicolaou, K.C., Frederick, M.O., and Botana, L.M. (2006) Biochem. Biophys. Res. Commun., 346, 1091–1099; (g) Vale, C., Nicolaou, K.C., Frederick, M.O., Gomez-Limia, B., Alfonso, A., Vieytes, M.R., and Botana, L.M. (2007) J. Med. Chem., 50, 356–363; (h) Bellocci, M., Sala, G.L., Callegari, F., and Rossini, G.P. (2010) Toxicol. Sci., 117, 109–121; (i) Cao, Z., LePage, K.T., Frederick, M.O., Nicolaou, K.C., and Murray, T.F. (2010) Toxicol. Sci., 114, 323–334; (J) Vale, C., Nicolaou, K.C., Frederick, M.O., Vieytes, M.R., and Botana, L.M. (2010) Toxicol. Sci., 113, 158–168; (k) Kellmann, R., Schaffner, C.A., Grønset, T.A., Satake, M., Ziegler, M., and Fladmark, K.E. (2009) J. Proteomics, 72, 695–707. 67. (a) Nicolaou, K.C., Pihko, P.M., Diedrichs, N., Zou, N., and Bernal, F. (2001) Angew. Chem., Int. Ed., 40, 1262–1265; (b) Nicolaou, K.C., Qian, W.Y., Bernal, F., Uesaka, N., Pihko, P.M., and Hinrichs, J. (2001) Angew. Chem., Int. Ed., 40, 4068–4071; (c)
Carter, R.G. and Weldon, D.J. (2000) Org. Lett., 2, 3913–3916; (d) Carter, R.G., Bourland, T.C., and Graves, D.E. (2002) Org. Lett., 4, 2177–2179; (e) Carter, R.G., Graves, D.E., Gronemeyer, M.A., and Tschumper, G.S. (2002) Org. Lett., 4, 2181–2184; (f) Zhou, X.T. and Carter, R.G. (2004) Chem. Commun., 2138–2140; (g) Zhou, X.T., Lu, L., Furkert, D.P., Wells, C.E., and Carter, R.G. (2006) Angew. Chem., Int. Ed., 45, 7622–7626; (h) Zhou, X.T. and Carter, R.G. (2006) Angew. Chem., Int. Ed., 45, 1787–1790; (i) Dounay, A.B. and Forsyth, C.J. (2001) Org. Lett., 3, 975–978; (J) Aiguade, J., Hao, J.L., and Forsyth, C.J. (2001) Org. Lett., 3, 979–982; (k) Forsyth, C.J., Hao, J.L., and Aiguade, J. (2001) Angew. Chem., Int. Ed., 40, 3663–3667; (l) Geisler, L.K., Nguyen, S., and Forsyth, C.J. (2004) Org. Lett., 6, 4159–4162; (m) Nguyen, S., Xu, J.Y., and Forsyth, C.J. (2006) Tetrahedron, 62, 5338–5346; (n) Forsyth, C.J., Xu, J.Y., Nguyen, S.T., Samdal, I.A., Briggs, L.R., Rundberget, T., Sandvik, M., and Miles, C.O. (2006) J. Am. Chem. Soc., 128, 15114–15116; (o) Li, Y.F., Zhou, F., and Forsyth, C.J. (2007) Angew. Chem., Int. Ed., 46, 279–282; (p) Ishikawa, Y. and Nishiyama, S. (2004) Heterocycles, 63, 539–565; (q) Ishikawa, Y. and Nishiyama, S. (2004) Heterocycles, 63, 885–893; (r) Oikawa, M., Uehara, T., Iwayama, T., and Sasaki, M. (2006) Org. Lett., 8, 3943–3946; (s) Yadav, J.S., Joyasawal, S., Dutta, S.K., and Kunwar, A.C. (2007) Tetrahedron Lett., 48, 5335–5340; (t) Yadav, J.S. and Venugopal, C. (2007) Synlett, 2262–2266; (u) Li, X., Li, J., and Mootoo, D.R. (2007) Org. Lett., 9, 4303–4306. 68. (a) Nicolaou, K.C., Li, Y.W., Uesaka, N., Koftis, T.V., Vyskocil, S., Ling, T.T., Govindasamy, M., Qian, W., Bernal, F., and Chen, D.Y.K. (2003) Angew. Chem., Int. Ed. Engl., 42, 3643–3648; (b) Nicolaou, K.C., Chen, D.Y.K., Li, Y.W., Qian, W.Y., Ling, T.T., Vyskocil, S., Koftis, T.V., Govindasamy, M., and Uesaka, N. (2003) Angew. Chem., Int. Ed., 42,
References 3649–3653; (c) Nicolaou, K.C., Pihko, P.M., Bernal, F., Frederick, M.O., Qian, W.Y., Uesaka, N., Diedrichs, N., Hinrichs, J., Koftis, T.V., Loizidou, E., Petrovic, G., Rodriquez, M., Sarlah, D., and Zou, N. (2006) J. Am. Chem. Soc., 128, 2244–2257; (d) Nicolaou, K.C., Chen, D.Y.K., Li, Y.W., Uesaka, N., Petrovic, G., Koftis, T.V., Bernal, F., Frederick, M.O., Govindasamy, M., Ling, T.T., Pihko, P.M., Tang, W.J., and Vyskocil, S. (2006) J. Am. Chem. Soc., 128, 2258–2267. 69. (a) Nicolaou, K.C., Vyskocil, S., Koftis, T.V., Yamada, Y.M.A., Ling, T.T., Chen, D.Y.K., Tang, W.J., Petrovic, G., Frederick, M.O., Li, Y.W., and Satake, M. (2004) Angew. Chem., Int. Ed., 43, 4312–4318; (b) Nicolaou, K.C., Koftis, T.V., Vyskocil, S., Petrovic, G., Ling, T.T., Yamada, Y.M.A., Tang, W.J., and Frederick, M.O. (2004) Angew. Chem., Int. Ed., 43, 4318–4324; (c) Nicolaou, K.C., Koftis, T.V., Vyskocil, S., Petrovic, G., Tang, W.J., Frederick, M.O., Chen, D.Y.K., Li, Y.W., Ling, T.T., and Yamada, Y.M.A. (2006) J. Am. Chem. Soc., 128, 2859–2872; (d) Nicolaou, K.C., Frederick, M.O., Loizidou, E.Z., Petrovic, G., Cole, K.P., Koftis, T.V., and Yamada, Y.M.A. (2006) Chem. Asian J., 1, 245–263; (e) Nicolaou, K.C., Frederick, M.O., Petrovic, G., Cole, K.P., and Loizidou, E.Z. (2006) Angew. Chem., Int. Ed., 45, 2609–2615. 70. (a) Evans, D.A., Kværnø, L., Mulder, J.A., Raymer, B., Dunn, T.B., Beauchemin, A., Olhava, E.J., Juhl, M., and Kagechika, K. (2007) Angew. Chem., Int. Ed., 46, 4693–4697; (b) Evans, D.A., Dunn, T.B., Kværnø, L., Beauchemin, A., Raymer, B., Olhava, E.J., Mulder, J.A., Juhl, M., Kagechika, K., and Favor, D.A. (2007) Angew. Chem., Int. Ed., 46, 4698–4703; (c) Evans, D.A., Kværnø, L., Dunn, T.B., Beauchemin, A., Raymer, B., Mulder, J.A., Olhava, E.J., Juhl, M., Kagechika, K., and Favor, D.A. (2008) J. Am. Chem. Soc., 130, 16295–16309. 71. Kalaitzis, J.A., Chau, R., Kohli, G.S., Murray, S.A., and Neilan, B.A. (2010) Toxicon, 56, 244–258.
72. Bringmann, G., Hamm, A., Kraus,
73.
74.
75.
76. 77.
78.
79.
J., Ochse, M., Noureldeen, A., and Jumbam, D.N. (2001) Eur. J. Org. Chem., 1983–1987. For a recent review, see: Wang, F.-P., Chen, Q.-H., and Liu, X.-Y. (2010) Nat. Prod. Rep., 27, 529–570. See, for example: (a) Ata, A. and Andersh, B.J. (2008) Alkaloids Chem. Biol., 66, 191–213; (b) Li, H.J., Jiang, Y., and Li, P. (2006) Nat. Prod. Rep., 23, 735–752; (c) Rahman, A.U. and Choudhary, M.I. (1999) Nat. Prod. Rep., 16, 619–635, and references therein. For a recent review, see: Kobayashi, J. and Kubota, T. (2009) Nat. Prod. Rep., 26, 936–962. For a review, see: Moser, B.R. (2008) J. Nat. Prod., 71, 487–491. Pettit, G.R., Inoue, M., Kamano, Y., Herald, D.L., Arm, C., Dufresne, C., Christie, N.D., Schmidt, J.M., Doubek, D.L., and Krupa, T.S. (1988) J. Am. Chem. Soc., 110, 2006–2007. (a) Pettit, G.R., Inoue, M., Kamano, Y., Dufresne, C., Christie, N.D., Niven, M.L., and Herald, D.L. (1988) J. Chem. Soc., Chem. Commun., 865–867; (b) Pettit, G.R., Kamano, Y., Dufresne, C., Inoue, M., Christie, N.D., Schmidt, J.M., and Doubek, D.L. (1989) Can. J. Chem., 67, 1509–1513; (c) Pettit, G.R., Kamano, Y., Inoue, M., Dufresne, C., Boyd, M.R., Herald, C.L., Schmidt, J.M., Doubek, D.L., and Christie, N.D. (1992) J. Org. Chem., 57, 429–431; (d) Pettit, G.R., Xu, J.P., Williams, M.D., Christie, N.D., Doubek, D.L., and Schmidt, J.M. (1994) J. Nat. Prod., 57, 52–63; (e) Pettit, G.R., Ichihara, Y., Xu, J.P., Boyd, M.R., and Williams, M.D. (1994) Bioorg. Med. Chem. Lett., 4, 1507–1512; (f) Pettit, G.R., Xu, J.P., Ichihara, Y., and Williams, M.D. (1994) Can. J. Chem., 72, 2260–2267; (g) Pettit, G.R., Xu, J.P., and Schmidt, J.M. (1995) Bioorg. Med. Chem. Lett., 5, 2027–2032; (h) Pettit, G.R., Tan, R., Xu, J.P., Ichihara, Y., Williams, M.D., and Boyd, M.R. (1998) J. Nat. Prod., 61, 955–958. Fukuzawa, S., Matsunaga, S., and Fusetani, N. (1994) J. Org. Chem., 59, 6164–6166.
313
314
8 Biomimetic Syntheses of Alkaloids with a Non-Amino Acid Origin 80. (a) Fukuzawa, S., Matsunaga, S., and
81.
82. 83.
84.
85. 86.
87.
88. 89.
90.
91.
92.
Fusetani, N. (1995) J. Org. Chem., 60, 608–614; (b) Fukuzawa, S., Matsunaga, S., and Fusetani, N. (1995) Tetrahedron, 51, 6707–6716; (c) Fukuzawa, S., Matsunaga, S., and Fusetani, N. (1997) J. Org. Chem., 62, 4484–4491. (a) For a recent review, see: Lee, S., LaCour, T.G., and Fuchs, P.L. (2009) Chem. Rev., 109, 2275–2314; for recent achievements, see: (b) Fortner, K.C., Kato, D., Tanaka, Y., and Shair, M.D. (2010) J. Am. Chem. Soc., 132, 275–280; (c) Poza, J.J., Rodr´ıguez, J., and Jim´enez, C. (2010) Bioorg. Med. Chem., 18, 58–63. Lee, J.S. and Fuchs, P.L. (2005) J. Am. Chem. Soc., 127, 13122–13123. (a) Jeong, J.U. and Fuchs, P.L. (1994) Tetrahedron Lett., 35, 5385–5388; (b) Jeong, J.U. and Fuchs, P.L. (1995) Tetrahedron Lett., 36, 2431–2434. (a) Jeong, J.U., Sutton, S.C., Kim, S., and Fuchs, P.L. (1995) J. Am. Chem. Soc., 117, 10157–10158; (b) Jeong, J.U., Guo, C., and Fuchs, P.L. (1999) J. Am. Chem. Soc., 121, 2071–2084. Smith, S.C. and Heathcock, C.H. (1992) J. Org. Chem., 57, 6379–6380. Guo, C., Bhandaru, S., Fuchs, P.L., and Boyd, M.R. (1996) J. Am. Chem. Soc., 118, 10672–10673. (a) Drogemuller, M., Jautelat, R., and Winterfeldt, E. (1996) Angew. Chem., Int. Ed. Engl., 35, 1572–1574; (b) Haak, E. and Winterfeldt, E. (2004) Synlett, 1414–1418. Yagi, S. (1909) Kyoto Igaku Zasshi, 6, 208–222. Nakano, T., Saeki, Y., Gibbons, C.S., and Trotter, J. (1968) J. Chem. Soc., Chem. Commun., 600–601. Li, C.S., Di, Y.T., He, H.P., Gao, S., Wang, Y.H., Lu, Y., Zhong, J.L., and Hao, X.J. (2007) Org. Lett., 9, 2509–2512. Mu, S.Z., Yang, X.W., Di, Y.T., He, H.P., Wang, Y., Wang, Y.H., Li, L., and Hao, X.J. (2007) Chem. Biodivers., 4, 129–138. Saito, S., Yahata, H., Kubota, T., Obara, Y., Nakahata, N., and Kobayashi, J. (2008) Tetrahedron, 64, 1901–1908.
93. (a) Morita, H. and Kobayashi, J. (2002)
94.
95.
96.
97.
98.
99.
100.
101.
102.
Tetrahedron, 58, 6637–6641; (b) Kobayashi, J., Ueno, S., and Morita, H. (2002) J. Org. Chem., 67, 6546–6549; (c) Zhang, Y., He, H.P., Di, Y.T., Mu, S.Z., Wang, Y.H., Wang, J.S., Li, C.S., Kong, N.C., Gao, S., and Hao, X.J. (2007) Tetrahedron Lett., 48, 9104–9107. (a) Suzuki, K.T., Okuda, S., Niwa, H., Toda, M., Hirata, Y., and Yamamura, S. (1973) Tetrahedron Lett., 14, 799–802; (b) Niwa, H., Hirata, Y., Suzuki, K.T., and Yamamura, S. (1973) Tetrahedron Lett., 14, 2129–2132. (a) Heathcock, C.H. (1996) Proc. Natl. Acad. Sci. U.S.A., 93, 14323–14327; (b) Heathcock, C.H. (1992) Angew. Chem., Int. Ed. Engl., 31, 665–681. (a) Ruggeri, R.B. and Heathcock, C.H. (1989) Pure Appl. Chem., 61, 289–292; (b) Heathcock, C.H., Piettre, S., Ruggeri, R.B., Ragan, J.A., and Kath, J.C. (1992) J. Org. Chem., 57, 2554–2566. For the isolation of methyl homosecodaphniphyllate, see: Toda, M., Hirata, Y., and Yamamura, S. (1972) Tetrahedron, 14, 1477–1484. (a) Ruggeri, R.B., Hamen, M.M., and Heathcock, C.H. (1988) J. Am. Chem. Soc., 110, 8734–8736; (b) Heathcock, C.H., Hansen, M.M., Ruggeri, R.B., and Kath, J.C. (1992) J. Org. Chem., 57, 2544–2553. For the isolation of secodaphniphylline, see: Irikawa, H., Toda, M., Yamamura, S., and Hirata, Y. (1969) Tetrahedron Lett., 9, 1821–1824. (a) Stafford, J.A. and Heathcock, C.H. (1990) J. Org. Chem., 55, 5433–5434; (b) Heathcock, C.H. and Stafford, J.A. (1992) J. Org. Chem., 57, 2566–2574. (a) For the isolation of bukittinggine, see: Arbain, D., Byrne, L.T., Cannon, J.R., Patrick, V.A., and White, A.H. (1990) Aust. J. Chem., 43, 185–190; (b) for bukittinggine-related alkaloids, see: Zhang, C.R., Yang, S.P., and Yue, J.M. (2008) J. Nat. Prod., 71, 1663–1668. Heathcock, C.H., Stafford, J.A., and Clark, D.L. (1992) J. Org. Chem., 57, 2575–2584.
References 103. (a) Ruggeri, R.B. and Heathcock, C.H.
(1990) J. Org. Chem., 55, 3714–3715; (b) Heathcock, C.H., Ruggeri, R.B., and McClure, K.F. (1992) J. Org. Chem., 57, 2585–2594. 104. For the isolation of daphnilactone A, see: (a) Sasaki, K. and Hirata, Y. (1972) Tetrahedron Lett., 13, 1275–1278; (b) Sasaki, K. and Hirata, Y. (1972) J. Chem. Soc., Perkin Trans. 2, 1411–1415. 105. Ruggeri, R.B., McClure, K.F., and Heathcock, C.H. (1989) J. Am. Chem. Soc., 111, 1530–1531. 106. For the isolation of codaphniphylline, see: Irakawa, H., Sakabe, N.,
107.
108.
109. 110.
Yamamura, S., and Hirata, Y. (1968) Tetrahedron, 24, 5691–5700. Heathcock, C.H., Kath, J.C., and Ruggeri, R.B. (1996) J. Org. Chem., 60, 1120–1130. (a) Niwa, H., Toda, M., Ishimaru, S., Hirata, Y., and Yamamura, S. (1974) Tetrahedron, 30, 3031–3036; (b) Yamamura, S. and Hirata, Y. (1976) Chem. Lett., 1381–1382. Heathcock, C.H. and Joe, D. (1995) J. Org. Chem., 60, 1131–1142. Piettre, S. and Heathcock, C.H. (1990) Science, 248, 1532–1534.
315
317
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids Hans-Dieter Arndt, Roman Lichtenecker, Patrick Loos, and Lech-Gustav Milroy
Das Runde muss ins Eckige. Sepp Herberger 9.1 Introduction
In general, peptides are not traditionally considered alkaloids. Therefore, this and the following chapter will have to stretch the borders of the field to identify common principles among structurally related molecules of various origins. We illustrate similarities and differences between typical peptide-derived natural products, which either are ‘‘alkaloids’’ in the sense of the typically assumed definition – that is, a generally alkaline (=basic), bioactive compound, preferentially of green plant origin – or are so much related to them either by biosynthetic roots or by structure that a unifying discussion may be appropriate. Owing to the high number of molecules, synthetic works, and structure–function investigations – all triggered by highly intriguing molecular structures and many promising bioactivities of peptide alkaloids – it is impossible to give a comprehensive treatment here. We apologize for the unavoidable, arbitrary selection of topics, molecules, and syntheses, and for the omission of all the excellent studies we could not cover in detail or even reference. 9.1.1 Peptide Alkaloids: An Overview
Peptide alkaloids occupy an ill-defined subsection of the rich structural universe of alkaloid natural products. They show similarities to many other peptide natural products, for example, lantibiotics or cyclo(depsi)peptides. A somewhat arbitrary distinction has to be made between the many amino-acid-derived alkaloids, for example, indole, diketopiperazine, or the ergot alkaloids such as lysergic acid (1) on the one hand, and linear peptides with basic properties such as dolastatin 15 (2) on the other hand. Historically, only cyclopeptides like 3, which defied the ‘‘typical’’ alkaloid or peptide classification and were basic and from plant origin, were termed ‘‘peptide alkaloids.’’ However, substances with various degrees of Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
318
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids Me HOOC
N
Me H N
Me Me2N
H
Me O N H Me
O Me
Me N O
N H
Me H O N H N
O
Me
H N O O
HN
H
H Me
NH O
N
H
O HN H Me
NH
Me H N O O
Me N N
N
Me Me
Me 3: ziziphine A
Me HO
S
O
S Me2N
O
Me
Me
NH N
O
2: dolastatin 15
H O
Me
OMe N
OMe Me
N O H O O Me Me
1: lysergic acid
O O
H
H N O Me
N
Me O
H
O
N H
Me
4: patellamide A
Cl
O
Cl N H
5: diazonamide A
Figure 9.1 Common alkaloid structures (1 and 2) in comparison with typical peptide alkaloids (3–5).
alkaloid and peptide elements are more often found and their number is steadily growing (Figure 9.1). Nonetheless, a recent classification that includes microbial and marine alkaloids (such as 4 or 5) remains unavailable. Therefore, this chapter and the one following are devoted to molecules that: • are composed of at least three amino acids with peptide bonds; • are preferentially cyclic, but go beyond ordinary cyclo(depsi)peptide structure; • feature additional motifs such as non-trivial (i.e., non-amino acid side chain) heterocycles, biaryls, or condensation loci distinguishing them from regular peptides. 9.1.2 Sources of Peptide Alkaloids
Peptide alkaloids have been first isolated from plants, but can be found in many lower organisms as well, for example, in bacteria, cyanobacteria, fungi, and marine invertebrates such as ascidians. For many of the compounds the actual producing organism has not been finally elucidated. Some materials isolated from plants in fact resemble products from bacterial or cyanobacterial origin. Indeed, in some cases it was shown that biosynthetically complex, highly modified secondary metabolites were produced by symbiotic microorganisms within the higher organism host from which the compounds have been isolated [1].
9.1 Introduction
9.1.3 Key Features of Biosynthesis
The common feedstock for all peptide alkaloids is amino acids, which get connected to canonical peptide chains and are then further modified. The backbones are assembled either by ribosomal peptide synthesis (RPS) [2] from genetically encoded templates on ribosomes or by non-ribosomal peptide synthesis (NRPS) [3] on dedicated multiprotein clusters (Figure 9.2). With RPS, a homochiral L-peptide is generated at the ribosome as a synthesis template (‘‘structure peptide’’), which typically carries a ‘‘leader peptide’’ as a tag for molecular recognition by the maturation enzymes. After maturation of the structure peptide, the leader peptide is removed. This process has only recently become more apparent [4]. With NRPS, more building blocks than only the gene-encoded amino acids are available for chain growth, which ensues in a conveyor-belt fashion on multiprotein complexes. In this case chemical modification can occur already during the assembly process. Both types of products then can undergo final enzymatic tailoring after assembly, such as methylation, prenylation, glycosidation, or acylation. Sometimes integration with other biosynthesis pathways (RPS→NRPS, NRPS→polyketide, NRPS→alkaloid) is observed, leading to additional building blocks or mixed-biosynthesis products. Canonical peptide chains are rather flexible and tend to fold into stable secondary structures only if their length exceeds 50–70 residues – with some exceptions in highly disulfide-rich ‘‘miniproteins.’’ Modification by post-assembly maturation and tailoring enzymes let the peptide mature into a structurally more rigid peptide alkaloid (Scheme 9.1a). Typical rigidifying elements are macrocycle formation, heterocycles formed from side-chain condensation reactions, or oxidative connections of side-chains, primarily at electron-rich aromatic side chains from Phe, Tyr, or Trp (Scheme 9.1b). Conceptually, this is reminiscent of terpene biosynthesis, where a flexible oligo-isoprenoid template is set up by oligomerization from isopentenyl pyrophosphate, which is followed by a distinct cyclization and tailoring phase [5]. This generates polycycles and oxidizes the chains at reactive loci (mostly double bonds) to arrive at very unique, defined structures. Just as in the case of terpenes, it can be assumed that diverse target structures can be reached from a single peptide precursor material by such a ‘‘covalent folding’’ – only depending on the topology of the maturation pathway. This will certainly offer benefits during evolution processes and organism adaptation, because small changes in biosynthesis can then generate entirely different structures. Of course, peptide alkaloids rarely reach the close-knit structural complexity of carbocyclic terpenes, but their richer heteroatom chemistry and the many functional groups available can lead to very specific shapes and drug-like activity from only a few amino acids as precursor material. Finally, post-synthetic tailoring might further complete the biosynthesis by adding alkyl groups or sugar residues to the core scaffold. For many peptide alkaloids it is not yet clear how the peptide template becomes assembled. Both RPS and NRPS have been documented in several cases. Furthermore, the interconnection of assembly and maturation remains mostly speculative. Biosynthesis research will still have to show at which steps the maturation happens
319
- Maturation during growth and/or after release possible
- Post-assembly tailoring (alkylations, glycosidations)
- Leader-peptide-mediated enzymatic maturation - Protease-mediated liberation from leader peptide
- Post-release modifications (alkylations, building block addition)
Figure 9.2
Distinct biosynthetic aspects of RPS and NRPS pathway logics.
- Chain growth as tRNA-ester at the ribosome - No chain modifications during growth - Hydrolytic release as canonical L-peptide chain tagged with a leader peptide
- Dedicated, MDa-protein-cluster-architecture-encoded synthesis - Less generic protein-cluster template, variation difficult - Amino acyl-AMP-mediated delivery of >200 non-proteinogenic AA - Chain growth as phosphopantheinyl-thioester at the NRPS cluster - Chain modifications during growth typical (redox, epimerization) - Diverse release pathways (hydrolysis, reduction, macrocycle formation) generating an untagged synthesis product
Non-ribosomal peptide synthesis (NRPS)
- Gene-encoded, mRNA templated synthesis at the ribosome - Generic variation of oligonucleotide template possible - tRNA-mediated delivery of 20 proteinogenic amino acids (AA)
Ribosomal peptide synthesis (RPS)
320
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids
9.2 Azole-Containing Peptide Alkaloids Assembly phase
Maturation phase R1
R OH
H2N
H2N
(a)
m
R
O
O amino acid
O
H N
Rn N H O n -2
OH
HN
Rn
O
H N R
O
peptide chain precursor
m
N H n -2
HO
HX O
−
+ PO3
− H2PO4
R
N H HO
N H HO
HN
−
O
Rn
O
N H Rm n -2
Side-chain linkage
Backbone or side-chain modification R
X N
a H
O
− H2O
− H2
R
R
R
X
O
N H
N H
X N
O
O
+
R
+ [O]
N HH
(b)
R
HH N
N HH O
R b OH
O
final biosynthesis product
O R
O
H N
Oxidation
Condensation
R
R1 O
matured peptide alkaloid
Dehydration Elimination
Rtailor
Tailoring phase R1
321
− H2
N H HO
H H N
R
R
N HH
O
Scheme 9.1 (a) Covalent folding of peptide chains into peptide alkaloid scaffolds (schematically simplified); (b) typical maturation steps in peptide alkaloid biosyntheses.
and if it indeed follows a model with distinct phases. However, such a model will provide a guiding principle for biomimetic synthesis strategies. For the sake of chemical discussion and to highlight similar features, we will often boldly simplify the structures in question in a ‘‘retrobiosynthetic’’ simplification guided by the typically available biosynthetic transformations (Scheme 9.1b), and attempt to align key synthetic contributions with the characteristic features of this impressive natural product class. This chapter is devoted to azole- and aryl-peptide alkaloids, whereas the following chapter will illustrate indole peptide alkaloids and the ecteinascidines.
9.2 Azole-Containing Peptide Alkaloids 9.2.1 Structural Features
Azole-containing molecules comprise probably the most diverse family among the peptide alkaloids. Their unifying feature is the occurrence of thioazol(in)e and oxazol(in)e units embedded in a peptide chain (Figure 9.3). These modifications
O
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids
322
Me Me H2N
Me H N
O N H H
O
Me H N
O
O
O
H N
N H
O
O
H N
N H
O
O
H N
N H
O O
HO H N
O HN S
N
O O
N
O
O N H
S
NH2 O
O
N H
O
O
H N
N H
O
O
O N H
O
O
N H
O
N N
HO
N
S
O
H S
HO
N
Me
N Me S O
S HN N
H N H
S
N H
O
O
Me
OH Me
OH N
O
H N Me
Me Me
N
N
Me
O
Me O
Me O NH
2
N H
R1 R2
S
N
N
N S
O Me
O H R3
OH
Azole-containing peptide alkaloids.
S
N
O
HN
N
O O 8: telomestatin (Streptomyces annulatus)
S
H N
O
O N
9: thiostrepton (Streptomyces azureus): R1 = CH3, R2 = H, R3 = CH2CH3 10: siomycin A (Streptomyces sioyaensis): R1 = R2 = CH2 (dehydroalanine), R3 = CH3
Figure 9.3
N
O
NH2
N H
N
Me
O
N
HO
H
NH
O O
H O Me Me Me 4: patellamide A (Prochloron spp.)
N
N COOH H H
O
N
N
N
O
N
HN
H S
H N
NH O
Me
O
N
Me
NH
N H
O
OH H N
HN H Me
S
H
Me
H O
H N
O
N
Me
O
S
O
7: thiangazole (Polyangium sp.)
O
O
N H
O
N
NH
S
S
H N
Me
N H
N
Me
N N
H Me
S Ph
Me O
O Me
Me
O
O 6: microcin B17 (Escherichia coli)
H
N H
S
N H
O
H N
N
Me Me
N
N
O
H N
N H
O
NH2
O
O
H N
H N
N
NH
O
Me
H2N
N
S
H N
N H
O
O
H N
N H
NH
Me
O N S
H N H
S N
N OH
HN O
Me
Me NH Me
11: micrococcin P1 (Microccocus spp.)
O
OH
9.2 Azole-Containing Peptide Alkaloids HO O
Me
O
H
O
NH
Me NH
N
N
O O
S
O
N
H S Me
HO
H
N H
N
N H
S Me
S
O HN
S
S
S
N
S
N
H N
N HN
O O
OH
S N
MeHN
NH2
N N
Me
O
N N
H O
N
O
N
S
H Me N HN
S
O
S
H
NH2
H NH O
NH
O
O
O OH
O
Me 12: nosiheptide (Streptomy ces actuosus)
Figure 9.3
13: GE2270A (Planobispora rosea)
(Continued)
can occur with a low frequency, such as exemplified in the long-chain molecule microcin B17 (6), or with impressive density, as can be seen with telomestatin (8) or thiostrepton (9), for instance. However, in most cases the amino acid nature of their backbone composition is still readily discerned. They can be found as linear [microcin B17, thiangazole (7)] and monocyclic azol(in)e-containing peptide alkaloids [lissoclinamides, patellamides (4), telomestatin], and as the structurally more complex thiopeptide (also called thiazolyl peptide) antibiotics, such as thiostrepton, micrococcin P1 (11), nosiheptide (12), or GE2270A (13). Most of them were retrieved from microorganisms of terrestrial or marine origin, prominently from Actinomyces and cyanobacteria. 9.2.2 Biomimetic Elements in Azole-Containing Peptide Alkaloids
Tailoring by hydroxylation, epoxidation, glycosidation, or alkylation/prenylation has been found for some azole-containing peptide alkaloids, but the large majority of them are composed purely from naturally occurring amino acid building blocks, also with respect to the stereochemistry of individual residues. Their key feature is the occurrence of azole heterocycles in the running peptide chain – mostly oxazol(in)es and/or thiazol(in)es. Owing to their complex molecular structures and their mostly microbial origin, it was assumed for many of them that their assembly would arise from NRPS pathways. However, Walsh et al. showed that microcin B17 (6) is formed from genetically encoded peptides by RPS [6]. A multi-enzyme complex transforms the linear peptide precursor by heterocycle formation [7]. While it is certainly too early to generalize, this trend has gained momentum in recent years, when RPS was shown to be involved with the Lissoclinum peptides, the patellamides (4), and the thiopeptide antibiotics as well (9–13). In all cases a serine, threonine, or cysteine peptide is dehydrated (cyclodehydratase, McbB for 6) and then sometimes further oxidized (dehydrogenase, McbC for 6) to generate peptide-embedded heterocycles [8].
323
324
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids Proposed biosynthesis pathway in Mcb B: B+
B Zn O
HX
R1
O
R2
N H
B
− H X
Zn
Zn OH
X
R1
R2
N H
O
B
X
R1
N
O
H B+
X−
B
HR
2
O
X=O,S Biomimetic deprotection/cyclodehydration with OPPh3 / Tf2O: Ph3P
+
−
2 TfO
O + Ph3P O R1
S Tr N H
Scheme 9.2
R2 O
−OPPh3 −TfO− −Tr+
Ph3P R1
+
SH
−OPPh3
O R2
N
−H+
S R1
O
N
HR 2 O
Enzymatic heterocycle formation and biomimetic cyclodehydration.
Mechanistically, the dehydratase enables a nucleophilic side-chain attack at an amide carbon (Scheme 9.2), which is not easily induced with chemical reagents. These typically activate the side chain [9, 10]. However, for thiazolines such a condensation was achieved with oxophilic Lewis acids such as TiCl4 [11] or Tf2 O-activated triphenylphosphine oxide in a biomimetic fashion [12], leading directly to peptide heterocycles (Scheme 9.2). The access to oxazoles by this method is less biomimetic, as the corresponding hydroxy group has to be oxidized to the ketone prior to ring closure and then serves as the electrophile [13]. Furthermore, at sensitive functionalities and longer peptide chains the limits of mimicking enzymes operating on peptide chain backbones with bulk reagents are currently reached rather quickly. 9.2.3 Thiangazole
The oligo-thiazoline thiangazole (7) was isolated from Polyangium spp. (strain PI 3007) and showed HIV-inhibitory properties [14, 15]. Together with the similar mirabazoles and tantazoles, its biosynthesis remains unclear to date, but is considered to involve SAM-dependent α-methylation of cysteine to account for the rare 2-methylcysteine subunits. Interestingly, when assuming a Hofmann-type deamination on a terminal phenylalanine and cyclodehydratation and dehydrogenation steps, 7 can be formally simplified to the pentapeptide 15 (Scheme 9.3). In a RPS-grounded assembly, a single methyltransferase would suffice to install all methyl groups, either on the peptide (15→14) or on the azoline level after azole ring formation. The former order of events seems more
9.2 Azole-Containing Peptide Alkaloids O Me
N H
O
Me
Me
O
N
N H
N
S
SH O Me
HN
H N
S
methylation
H 2N
H N H
O
O N H H SH
OH NH2 O
3
O SH Me
O HS
7
Scheme 9.3
HN
Ph
Me
Me
Ph elimination
oxidation
Me
N
S
OH O Me
Me N
Ph
H Me
HN
cyclodehydration
325
14
15
Hypothetical biomimetic simplification of thiangazole (7) to pentapeptide 15.
plausible since catenated thiazolines tend to auto-oxidize easily. Alternatively, 2-methylcysteine might be biosynthetically generated and independently integrated into the growing chain by NRPS. Such a bold proposal has not yet been realized in synthetic practice, but elements of it inspired the synthesis design in various groups. In most successful syntheses, the C-terminal oxazole was not part of a multiple ring closure but had to be independently generated by cyclodehydratation reactions (Figure 9.4). To assemble the oligo-thiazoline, several alternatives were worked out. The group of Pattenden successfully established an iterative condensation of nitriles with 2-methylcysteine to extend the chain C-terminally [16]. In a potentially more biomimetic variant, Heathcock showed that a string of 2-methylcysteines can be condensed to an oligothiazole with considerable efficiency. This method was later used by Wipf (below) [17] and Akaji [18] in their syntheses of 7, which featured further optimizations. To furnish the uncommon 2-methylcysteine, Seebach’s self-regeneration of chirality was instrumental. Despite the tendency of cysteine enolates to eliminate the sulfur substituent, Pattenden could prepare 2-methylcysteine via the tert-butyl Robinson-Gabriel cyclodehydration Akaji et al., TH 1999, 55, 10685 Heathcock et al. JOC 1994, 59, 4733 Burgess cyclodehydration Cu(I)-mediated oxidation Pattenden et al. TH 1995, 51, 7321 PPh3/I2 cyclodehydration Wipf et al., JOC 1995, 60, 7224
O
Me
H3CHN
O
N
Me N
DDQ oxidation Akaji et al., TH 1999, 55, 10685 PhSeO2H oxidation Wipf et al., JOC 1995, 60, 7224
Figure 9.4 Strategic disconnections in thiangazole (7) total syntheses. To improve readability, the following abbreviations have been adopted throughout: ACIE = Angew. Chem. Int. Ed., BMCL = Bioorg. Med. Chem. Lett., CAJ = Chem. Asian. J.,
N S
S Me
S Me
N
Ph
7: thiangazole (Polyangium sp.)
Iterative condensation of nitriles Pattenden et al. TH 1995, 51, 7321 TiCl4 mediated triple-cyclodehydration Akaji et al.,TH 1999, 55, 10685 Wipf et al., JOC 1995, 60, 7224 Heathcock et al. JOC 1994, 59, 4733
CPB = Chem. Pharm. Bull., CEJ = Chem. Eur. J., JACS = J. Am. Chem. Soc., JOC = J. Org. Chem., OL = Org. Lett., PNAS = Proc. Natl. Acad. Sci. U.S.A., TH = Tetrahedron, TL = Tetrahedron Lett.
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids
326
Cl
H3N
2 steps 81%
HS
COOMe
H
Ses N H
OBn
H HN N
Ph
19
O Me
O
O Me
OBn 20
tBu
OBn
34%
O
O Me
N
H HN N
H3CHN SAc
O
Me
O Me O
SAc 21
Me O Me
N
vi, vii
HN
Ph
S
Me
H3CHN
O Me
COOMe Me
18
iii-v
OBn
N
d.r. > 99:1
O
HN
7 steps 21%
56%
S
Me N
O Me
H
17
H3CHN
Me N
tBu
ii
OMe
N
16 O
MeO
H
S
L-Cys x HCl
O
i
N
tBu
O
O LiO
O
COOMe R
− +
40%
N
N
SAc
S viii
S Me
N
Ph
S Me 79%
7
Scheme 9.4 2-Methylcysteine synthesis and oxazoline–thiazoline conversion by Wipf. Reagents and conditions: (i) LDA, DMPU, THF, −90 ◦ C; (ii) MeI; (iii) Pd(OH)2 , H2 , MeOH; (iv) Burgess reagent, THF; (v) AcSH; (vi) NH3 , MeOH; (vii) TiCl4 , CH2 Cl2 ; and (viii) PhSeOOH, benzene, 60 ◦ C; Ses: (trimethylsilyl)ethylsulfonyl.
thioaminal 16, which was alkylated to 18 via enolate 17 with a dr of >99 : 1 (Scheme 9.4) [19]. To circumvent the preparation of the cysteine building blocks and to generate variants for structure–activity relationship studies, Wipf developed an interesting oxazoline–thiazoline conversion. Trisubstituted oxazole 19 was extended with 2-methylserine building blocks to oligomer 20 (seven steps, 21%) [17]. After debenzylation, multiple oxazoline formation was cleanly induced with the Burgess reagent (60%), giving a structural analog of 7. Treatment of the labile oxazolines with thioacetic acid then delivered the 2-methylcysteine precursor 21, which was processed by biomimetic dehydration and oxidation of the terminal phenylethyl group (PhSeOOH) to give thiangazole (7). Overall, these biomimetic investigations were setting the stage for approaches to more complex peptide alkaloid syntheses. 9.2.4 Lissoclinamide 7
In contrast to the linear oligothiazolines, much more is known on the biosynthesis of cyclic azole-containing peptide alkaloids. The gene clusters for the monocyclic lissoclinamides, the ulithiacyclamides, and the patellamides have been identified and found to be quite similar [20]. These data suggested a RPS assembly with a genetically encoded peptide template guided by a leader peptide [21]. Typical enzymatic activities, dehydration, dehydrogenation, and macrolactam formation, could be identified. For lissoclinamide 4, the peptide ring closure site was located between
9.2 Azole-Containing Peptide Alkaloids
327
a Cys and a Phe residue. Notably, many Lissoclinum peptides contain d-amino acid residues, but a formal epimerase enzyme was not found. It was then noted that the chemically synthesized l-Val lissoclinamide 7 converted into the d-isomer (22) upon treatment with pyridine, showing a conformation-driven preference [22]. It can hence be assumed that the d-residues at the acidic thiazoline α-carbons are formed by epimerization of 23 following biosynthetic assembly [23]. An according simplification of lissoclinamide 7 (22) is shown in Scheme 9.5. Notably, during the biosynthesis, azole-formation (24→23) could precede macrocyclization (25→24), but the full peptide chain 25 will be assembled first. Me H O
Me
Me O
R
S
N H
Me H H O
N
N
HN R
H N
H N
O
N
O
S N H
side chain Ph epimerisation
N
S
H N
RO O
H O
N
HN S
O
O H
Me Me O H
S
N H
N
22: lissoclinamide 7
H2N
O
O
cycloPh dehydration
N
S
N H SH
Me H H O N N N H H O O OH Me 25 Ph
O N H
SR H O
HN
NH
O Ph
23 Ph H N
Me
HN
H
O H
Ph
Ph
Scheme 9.5
Me
Me O
H N
HN
Ph O
O H
SR
24 Me H N O
macrocyclization
O OH SH
Biosynthesis-guided simplification of lissoclinamide 7 (22) to heptapeptide 25.
Many members of this peptide alkaloid class have been synthesized from oxazole- or thiazole-amino-acid building blocks, but very few approaches have been reported where the azoles are installed after assembly of the complete peptide chain. Additionally, for the lissoclinamides their exceptional configurational lability had to be taken into account [24]. Wipf et al. used multiple cyclodehydrations, oxazoline–thiazoline interconversions, and an pentafluorophenyl diphenylphosphinate (FDPP)-mediated macrocyclization (Figure 9.5), giving access to the full configurational assignment of lissoclinamide 7 and to a library of variants for biological testing. The suitably protected heptapeptide 26 was assembled (Scheme 9.6) and converted into macrocycle 27 by deprotection, macrolactam formation, and TIPS (triisopropylsilyl)-protection of the secondary allo-Thr-OH while unraveling the primary Ser-OH groups. These were used to initiate a double cyclocondensation to 28 followed by thiolysis to install thioamides in the peptide backbone (29). At this stage, deprotection of the Thr side chain enabled a triple azoline ring closure to furnish lissoclinamide 7 (22) in a stereochemically pure form with exceptional efficiency (90% yield). Using this strategy with late-stage introduction of the thiazoline rings, the stereochemistry of 22 could be firmly assigned. Notably, the intermediate oxazoline
328
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids
Macrocyclization with FDPP Burgess-reagent mediated triple-cyclodehydration
Me H O H
Oxazoline thiolysis Me
Me O
S N H
Burgess reagent mediated triple-cyclodehydration
N
H O
N HN N
H N
O
Ph
N
22: lissoclinamide 7 (Prochloron spp.)
S OH
Ph Figure 9.5 Me
allo-Thr Me O
Strategic disconnections of lissoclinamide 7 (22) used by Wipf [22, 24].
Me
HO O H
HN
Ph N
OTBS i-v
O Ph
NH HN O
O O 26
Me
Me Me O
Me O O HN H Cbz HN NH
OTBS
61%
OH
N HH NH
RO O H
HN HN
N
H N
O
vi
X
RO O H
H
N HN
H N
O
OH
O Ph
N O
O H
Ph
27 (X = O, R = TIPS) 29 (X = S, R = H)
O
N HH NH N
HN O
Ph
O Ph
Me
Me Me O
X
28 (R = TIPS) vii-viii 41% over three steps
ix
90%
22: lissoclinamide 7
Scheme 9.6 Completion of the total synthesis of lissoclinamide 7 (22). Reagents and conditions: (i) H2 , Pd/C, MeOH; (ii) NaOH, THF–MeOH–H2 O; (iii) FDPP, NaHCO3 , DMF-CH2 Cl2 (3 : 1), 40 ◦ C; (iv) TIPS-OTf,
2,6-lutidine, CH2 Cl2 ; (v) TsOH, THF–H2 O; (vi) Burgess reagent, THF, 65 ◦ C; (vii) H2 S, MeOH–NEt3 , 22 ◦ C, 4 d; and (viii) TBAF, THF, 37 ◦ C; (ix) Burgess reagent, THF, 40–70 ◦ C.
formation located the introduction of thioamides in the backbone. Owing to activation of all the OH-groups for SN 2-type substitution by the Burgess reagent in the last step [25] allo-Thr had to be protected during chain assembly. But, overall, without the use of enzymes this synthesis could hardly come closer to a biomimetic strategy – before the biosynthesis was elucidated. 9.2.5 Thiostrepton
Thiopeptide antibiotics are highly modified, sulfur-rich macrocyclic peptide alkaloids that share another common motif beyond the azoles in the peptide chain: a tri- or tetrasubstituted pyridine core [26]. Early studies on their biosynthesis were made by feeding labeled amino acids and it was shown that the whole framework
9.2 Azole-Containing Peptide Alkaloids
of thiostrepton (9) and nosiheptide (12) is exclusively derived from proteinogenic amino acids [27–30]. Surprisingly, genetics studies conclusively established that all important thiopeptide antibiotics (thiostrepton, thiocillin, siomycin, thiomuracin, nosiheptide) [31–33] are derived from RPS [2,33–35]. One of the most complex members, thiostrepton (9), was isolated in the 1950s from Streptomyces azureus. It was identified as a very potent antibiotic with strong activity against Gram-positive bacteria (Figure 9.3) [7]. The X-ray structure of 9 revealed a globular structure reminiscent of folded protein domains [9]. Thiostrepton inhibits bacterial protein biosynthesis by binding tightly to the GTPase-associated region (GAR) on the 70S ribosome. Scheme 9.7 summarizes key features of its biosynthesis. The genetically encoded, Cys-rich structural peptide is N-terminally tagged with a leader peptide sequence (30) [4]. Dehydrations lead to the development of five thiazole rings and seven dehydroamino acids (31). In a highly unprecedented step – proposed early on by Bycroft and Gowland [36] – two dehydroamino acids engage in a formal 2-azadiene Diels–Alder cycloaddition (32) to generate a macrocycle and the central piperidine core. This core is reduced in the case of thiostrepton (33), but in other thiopeptides a pyridine ring results from elimination reactions. The second macrocycle is completed by insertion of an activated quinaldic acid (34), which is apparently incorporated in a more NRPS-like fashion. After protease-mediated cleavage of the leader peptide, nucleophilic attack of the N-terminus at the epoxidized aromatic ring is assumed to insert the unconventional final linkage between the N-terminus and a threonine OH-group in the A-ring. Notably, this series of events suggests that the A-ring closes first. It remains to be clarified if the crucial cycloaddition during the biosynthesis is stepwise or concerted (Scheme 9.8), but the anticipated hydroxypiperidine intermediate certainly becomes reduced by an NADPH-dependent enzyme. The unconventional quinaldic acid is biosynthesized from tryptophan by a sequence involving C2 methylation, hydrolysis of the indole subunit, oxidative deamination to a putative diketone, condensation with the more electrophilic α-keto acid, and enantioselective tautomerization to a pyridyl alcohol. P450-mediated oxidation could generate a quinoline epoxide, to serve as an electrophile for the terminal Ile residue’s amino group. Two total syntheses of thiostrepton and the very similar siomycin have been published, the former by the Nicolaou group [37–40] and the latter by Nakata et al. [41, 42]. Figure 9.6 summarizes the key disconnections. Not surprisingly, the strategies realized do not slavishly follow the biosynthetic track – which was only clarified five years after the total synthesis was completed! Nonetheless, several structural elements are accessed in ingenious anticipation of potential biosynthetic possibilities, and the final ring closure of the B-ring is achieved at the same site where it probably occurs in biosynthesis. In setting up the peptide chains, effective but mostly conventional technologies are used in both syntheses, which we will not discuss in detail here. However, to set up the crucial central dehydropiperidine ring two distinct approaches were developed. The Nakata team developed an interesting auxiliary-controlled imine addition to a dehydropyrrolidine-derived azomethine ylide followed by a smart,
329
330
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids
Me H N
O
Me
HN
H
O
OH H N
N H
Me
O
N H
Me
Me O
OH H N
O
N H H O SH Me
O
SH H H O N
OH H H O N
N H OH
LP
O Me
N H
SH H H O N
SH H N
O Me
O
N H OH
SH H N
O N H OH
OH
O N H OH
O
R2 O
Me
30
Me
HN
Me H N H
O
O
Me
N H
O
Me O
O
H N
Me
N
N H
OH H N
N H
S
S
O Me
O
HO H
HN HO Me
O
NH
Me
H O
Me
HO H
O
N
OH Me
N H
N
OH S N HN S O H N N H S N O OH Me HS
S Me
O
H N
32
NH
LP
S
H O
Me
HN
HS
HO
H N H Me
S HN N S
H N
S Me
O
N H
O
O
N
HN HO Me
HS OH Me
N H N Me S O
N
O
A S
Me NH Me NH
Me
Me
NH
Me
N
NH2
O
N
Me
O
H O
H N
LP
33
O
O
O
X
NH
R2 O
O
O
O
HO H
H N
N H
OH
OH Me
Me
NH
O
N
Me
O N
N
A
O
N H
O
NH NH O
N
O
O
Me
Me
NH2
O
H N
H N
N
O
OH
O
O
N H
S
H N
Me
Me
O
NH
H H S N N O Me
31
LP
N
H H S N N
N
H N H Me
S HN N S O
B N
O
N H
O
OH H N
H N
N H
O
H N
O
HN
Me
Me
O
OH 34
NH2
O H
Me Me
OH
Me
9: thiostrepton
Scheme 9.7 Thiostrepton biosynthesis as deduced mostly from genetic information (LP: leader peptide).
directed five-ring-six-ring interconversion. Nicolaou’s group on the other hand realized the piperidine ring formation in alignment with the biosynthetic hypothesis of Bycroft and Gowland (Scheme 9.9a) by a 2-azadiene hetero-Diels–Alder reaction. Using Ag+ salts, 2-azadiene 36 was liberated from a thiazolidine precursor 35. Homodimerization of 36 was found to take place in Diels–Alder type fashion (37) even at low temperatures. In accordance with earlier work [43–45] the initial dehydropiperidine product 38 was highly prone to imine–enamine tautomerization, followed by an intramolecular aza-Mannich reaction. This unintended sequence could be suppressed using a scavenging amine (BnNH2 ), which removed the side-chain imine and liberated the labile amino-dehydropiperidine product 39 as a 1 : 1 mixture of diastereomers (separable later in the synthesis).
9.2 Azole-Containing Peptide Alkaloids
R1 R3HN
N OH N
S
R2
R
stepwise or concerted
R1 R3HN
NAD (P )H reduction
S
N HO R 2
O
R +H N 3
H2N
N R2 H
H2N
HOOC
oxidation ring opening
−OOC
ring closure isomerization tailoring epoxidation
O Me
methylation
Me
S
O
N
R'-O
R
N
O
H N
R=H R = Me
R
N
R1 R3HN
331
Ile
R''
OH
Scheme 9.8 Biosynthesis of the dehydropiperidine core and the quinoline unit of thiostrepton. Hetero-Diels-Alder reaction Nicolaou et al., ACIE 2004, 43, 5087-5092
Macrolactam formation Nicolaou et al., ACIE 2004, 43, 5087-5092 siomycin A: Hashimoto, Nakata et al., CAJ 2008, 3, 984
HO H
O
O
NH
S
O
H O
siomycin A: Oxazoline thiolysis Hashimoto, Nakata et al., CAJ 2008, 3, 984
N HN
HO
Me Yamaguchi macrolactonization Nicolaou et al., ACIE 2004, 43, 5087-5092
H S OH Me
N H
N
N
O
H N Me S O
S HN N
H N
NH
Me
TES-deprotection and OH-group elimination Nicolaou et al., ACIE 2004, 43, 5087-5092
siomycin A: imine equilibrium Hashimoto, Nakata et al., CAJ 2008, 3, 984
Me
N
H N H
S
Me
N H
O
O
OH N
O
H N Me
Me
OH
Jacobsen-Katsuki epoxidation Nicolaou et al., ACIE 2004, 43, 5087-5092 siomycin A: Hashimoto, Nakata et al., CAJ 2008, 3, 984
H N
HN H O R3
O NH2
O
O R1 R2
Selenium oxidation and elimination Nicolaou et al., ACIE 2004, 43, 5087-5092 Regio- and stereoselective epoxide opening LiClO4: Nicolaou et al., ACIE 2004, 43, 5087-5092 siomycin A: Yb(OTf)3: Nakata et al., CAJ 2008, 3, 984
9: thiostrepton (Streptomyces azureus, R1 = Me, R2 = H, R3 = Et) 10: siomycin A (Streptomyces sioyaensis, R1 = R2 = CH2 (dehydroalanine), R3 = Me)
Figure 9.6
Key synthetic disconnections in thiostrepton (9) and siomycin A (10) syntheses.
This sequence features both a breaking of symmetry and recycling of the accessory aldehyde 40 and is biomimetically inspired. The high reactivity and tendency to homodimerize can be attributed to the lower oxidation state of the 2-azadiene carbon-1 (imine instead of amide as in a peptide chain): studies by Moody with amide-derived 2-azadienes indicated much lower reactivity [46]. Whether the putative biosynthesis enzymes really promote a similar reaction between two crossing peptide strands remains to be elucidated. Nonetheless, it was shown that such transformations are chemically possible, and could be directed and/or catalyzed by suitable enzymes during biosynthesis.
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids
332
N
H S
COOEt
S
NH N
Me
Me
N
i
H
Boc Me N
N
Me Me
O
N +
S H Me
O
40
MeOOC
N O
H
ii-iv
OTBS
Me Me
N
Me (b)
Me
OTBS
COOAll H Me
Me
O
N O Boc Me Me 37 aza-DielsAlder dimerization
COOEt N N
S
6S
S N N
HS
O
Me
Me Me
OTBS 42
OH H N
Me
38
Me
Me Me
N
H N Me N N 5R EtOOC
39
N
41
MeOOC
O
Boc
52% brsm Me
H S
60%
S
MeOOC
S S H Me
O imine hydrolysis
S
1:1 mixture of diastereomers (5R, 6S and 5S, 6R)
N
N
Me Boc Me S N
6S
Me
N
N N
endo transition state
COOEt
N N
S N
Boc Me Me
(a) was recycled 68%
H2N N 5R
EtOOC
N Boc
Me Me
COOEt
N
S
EtOOC
36
CHO
Me
S
HS Me
35
Boc Me N
COOEt
N
Boc
HS
O
N
N O N Mn V O O PhPh
Me
Me
v
N O N FeV N
N
Me
S
69%
COOH manganese(V)-oxo complex of (R,R)-MnSal
Cys
HOOC
iron(V)-oxo complex in cytochrome P450 oxidation
Scheme 9.9 (a) Biomimetic dehydropiperidine synthesis; (b) biomimetic quinoline epoxide opening. Reagents and conditions: (i) Ag2 CO3 , pyridine, DBU, BnNH2 , −12 ◦ C; (ii) NaOCl, 4-Ph-py-N-oxide, (R,R)-MnSal, phosphate buffer, CH2 Cl2 ; (iii) NBS, AIBN, CCl4 ; (iv) DBU, THF; and (v) H-L-Ile-OAll, LiClO4 , MeCN; bvsm = based on recovered starting material.
During biosynthesis of the quinoline unit, a heme-catalyzed epoxidation is anticipated, which was followed in building block synthesis as well. Stereoselective epoxidation of the dihydroquinoline 41 was successful with Katsuki’s catalyst [47] (Scheme 9.9b) in moderate dr (87 : 13). Radical bromination and elimination installed the double bond (42), and regioselective opening with isoleucine esters under Li+ (Nicolaou) or Yb3+ -catalysis (Nakata) completed a biomimetic synthesis using the intrinsic reactivity [48, 49] of the biosynthetic intermediates. Successful completion of the thiostrepton synthesis again featured a biomimetic strategy (Scheme 9.10). The B-ring chain 44 was attached to the A-ring scaffold
N
H
N
S
O
O
N
Me
O
O
HS OH Me
N
S
NH
H N H Me
NH
Me
O
Me
O
S
S HN N
N
9
O FmO
+
O
N
OH
OH
O
N
H N
N H
Me
Me
O
O
H N
Me
H N H
Me
O NH2
Me
O
N
H
Me
HN
H
Me
O
O
HS OH Me
N
S
O
Me
H N H
NH
Me
Me
O
S
S HN N
42% N
O
HN
H N
N H
H N
N H
Me
H N
OTBS
H N Me S O
45
N
TBSO
Me
O
N H PhSe
N H N Me S O
O
Me
O
O
SePh H N
H
Me
O
O
SePh H N
46
OTBS
Me Me
O N H O PhSe HN TBSO Me H N N HO
iii
O HO
OH
S
S HN N
N
O
H N H
Me HS OH Me
N
S
TBSO H O Me NH
TBSO
TBSO
35%
iv,v
N
H
TBSO Me H O NH Me N NH O
HN
H O
TESO
TESO
59%
i,ii
NH2 SePh
O
O
O
44
HN
H N
O
SePh H N
OTBS
HO
HN
O
Me
H N Me
N H
PhSe
O
N H
HO
TBSO
H N Me S O
N
OH
NH2
N H N Me S
O NH2
NH2 SePh
O
SePh
O
Scheme 9.10 Completion of the synthesis of thiostrepton (9). Reagents and conditions: (i) HATU, HOAt, i-Pr2 NEt, DMF; (ii) Et2 NH, CH2 Cl2 ; (iii) 2,4,6-trichlorobenzoyl chloride, Et3 N, DMAP; (iv) t-BuOOH; and (v) HF–pyridine. Fm: 9-fluorenylmethyl.
HO
HN
H O
Me
HO H O
N S HN N H N S H
NH
Me
Me HS TBSO OH Me Me 43
HN
H O
TESO
TBSO H O Me NH
9.2 Azole-Containing Peptide Alkaloids 333
334
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids
43 by a HATU-mediated coupling. After C-terminal deprotection, hydroxy-acid 45 was ring-closed to the bicyclic macrolactam 46 at the putative ring closure site in the biosynthesis. The unmasking of the final product 9 was initiated with the oxidation-triggered elimination of the phenylselenyl alanines to the dehydro amino acids, and then HF treatment. During this process the Thr-OH group adjacent to the thiazoline ring was eliminated to give the (Z)-configured enamide side chain. This indicates that elimination of this residue may also be extremely facile during biosynthesis. Overall, the synthetic works toward thiostrepton (9) and siomycin (10) define milestones in peptide alkaloid total synthesis, and show how biomimetic considerations can successfully guide a complex total synthesis project toward crucial and productive bond formation chemistry. 9.2.6 GE2270A
In comparison it is worth taking a look at the thiopeptide GE2270A (13). This compound, isolated from Planobispora rosea, has been identified as an antibiotic as well, but its mode of action entails an allosteric blockade of elongation factor TU, a ribosome-associated factor. Structurally, some different features are noteworthy. Only one macrocycle is present, but it is larger in size than the A-ring of thiostrepton. The connecting core is now an aromatized pyridine ring, and some post-assembly tailoring is apparent (methylations, hydroxylations). Its biosynthesis has only been investigated by labeling experiments [50] but is believed to follow a similar pathway to thiostrepton, closely related to its relative thiomuracin [51]. Synthetically, GE2270 allows us to compare biomimetic and non-biomimetic strategies. Two distinct syntheses for this molecule have been worked out: one with biomimetic key steps [52, 53] and the second making extensive use of non-biological Pd-mediated sp2 -couplings [54, 55], even for the macrocyclization (Figure 9.7). Cyclodehydration with DAST Bach et al., CEJ 2008,14, 2322 Nicolaou et al., CAJ 2008, 3, 413
H O
N N
Negishi-coupling Bach et al., CEJ 2008, 14, 2322
N
S N
S
S
HN O O
13: GE2270A (Planobispora rosea)
NH N
MeHN
Figure 9.7
Stille macrocyclization Bach et al., CEJ 2008, 14, 2322
N
Hetero-Diels-Alder reaction and oxidation Nicolaou et al., CAJ 2008, 3, 413
Macrocyclization with FDPP Nicolaou et al., CAJ 2008, 3, 413
NH2 H
N
N S
O
O
H
O
Me
S Me
Key disconnections for GE2270A (13).
N H
Me H N S
H
OH O NH O OMe
9.2 Azole-Containing Peptide Alkaloids
335
Both approaches installed the side chain at the end stage of the synthesis. In Nicolaou’s biomimetic synthesis [52, 53], GE2270A and some of its variants were prepared. The strategy used to obtain the pyridine core was similar to that employed in their earlier synthesis of thiostrepton. Starting from the dehydropiperidine cycloadduct, NH3 was eliminated using DBU (1,8-diazabicyclo[5.4.0]undec-7-ene), and oxidation of the dihydropyridine delivered the aromatized pyridine 47 (Scheme 9.11). Similar steps are anticipated for the biosynthesis of pyridinecontaining thiopeptides. However, the building block resulting from this sequence contained two methyl esters in highly similar chemical environments due to its homodimeric origin. The methyl esters in 47 (Scheme 9.11) were challenging to differentiate downstream. Furthermore, chain extension by peptide couplings worked, but the O MeHN MeHN O O H
H N
N H
S
HN
O Me
S N O
H
N
OAll
NHBoc Me
S
MeO S
Me BocHN
N
HN H O OH
O
H
N H N
Me
O Me
HN
O Me
N
S
S
N
S
N
O
O 30 steps
N
N
L-cysteine
H 0.01% yield
O
N
N H Me S N HN H NH O OH
S
H
O
22 steps
NH2
4.8% yield
N-Boc valine
O 13: GE2270A
MeHN
MeHN O O
O O H
N H
S N
S
N S
N
H
Br
N H
S
O
N
N
Me
OR
O
S
SnMe3
N
H Me O S N
Pd-catalyzed couplings amide bond formation
HN
N
I S Br N Br 48 O S
N O Me
OR H RO
Scheme 9.11 Building blocks and key intermediates used by Nicolaou (top) and Bach (bottom).
S N
NH HN O
SnMe3 N
H Me N
O Me
O
H2N
O OtBu
IZn Me
S
S
BrZn
N
O
Me
S
H OTBS
OH BocHN
H
HN
OH BocHN
OO S
Me
47
H
O
Me
OMe
MeHN
macrocyclization
Pd-catalyzed Stille coupling
Boc N Me O Me
COOMe
peptide coupling cyclodehydration oxidation
O OMe
N
S
N
Me
OMe
N
S MeO
O
N
N
S
NHBoc
S
N
N
S
N
Me
S
S
N
MeO
O
HN H RO
336
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids
macrocycle formation by macrolactamization turned out to be rather difficult. The macrocycle could be closed with FDPP, but in yields ranging from only 15 to 25%, depending on the site of ring closure. Overall, this led to a synthesis in 30 steps in the longest linear sequence in modest overall yield. In the second synthesis by Bach [54, 55], a less biomimetic strategy was followed. The plan was to assemble the scaffold by consecutive Pd-mediated couplings around the pyridine core (Scheme 9.11, bottom). In prior work, methods for the coupling chemistry were developed from the zinc reagent 48, which allowed access to all of the building blocks. Similar to Nicolaou, macrolactam formations were found to be less effective in ring closure (30%). However, when a Stille coupling was used to close the macrocycle, the ring closing efficiency strongly improved (75%). In this way, more complex building blocks could be connected, leading to a shorter linear sequence (22 steps). Macrocycle formation included, the efficiency of this synthesis was fairly high, with an overall yield of 4.8%. These data show that biomimetic synthesis is not necessarily superior to a non-biomimetic approach, especially when crucial synthesis tools are not available or cannot be easily transferred. It critically shows that further method developments for the biomimetic synthesis of azole-containing peptide alkaloids are needed, with respect to chain assembly, maturation, and interconnection chemistry. Such efforts should be guided by close interaction with biosynthesis research to provide maximal mutual benefits.
9.3 Peptide Alkaloids Cyclized by Oxidation of Aryl Side Chains 9.3.1 Cyclic Peptides Containing Aryl-Alkyl Ethers
Numerous cyclopeptide alkaloids containing aryl-alkyl ethers have been isolated from a host of different plant families, such as Rhamnaceae, Rubiaceae, Olacaceae, and others [56, 57]. The structure of these 13- to 15-membered macrocycles features a phenyl alkyl ether with an enamide in ortho-para or ortho-meta orientation (Figure 9.8). This styrylamine moiety bridges two or three α-amino acids. N-Methyl or N,N-dimethyl substitutions are often found in the side-chains of these cyclic alkaloids. Although various biological activities such as antibacterial, antifungal, sedative, and immunostimulant properties have been identified for these compounds, detailed understanding about their biological function and their biosynthesis still needs to be developed [58]. However, their likely precursors are oligopeptides. Feeding experiments with labeled amino acids in the plant Ceanothus americanus resulted in the formation of the linear precursors of the corresponding cyclopeptides, suggesting that the precursor molecules are more likely to be generated by a specific enzyme system than by cleavage from genetically encoded peptides [59]. Three possible biosynthetic pathways leading to aryl-alkyl linked cyclopeptides can
9.3 Peptide Alkaloids Cyclized by Oxidation of Aryl Side Chains
337
OH O
Me H
O
O O
Me HN
H
NH Me
HN O
Me
49: sanjoinine A (frangufoline) Ziziphus vulgaris
Figure 9.8
H
N H HN O H Me NH2
H O O O
O O
O O
NH N H N H Me O Me O HN Me Me Me2N Me
NH
NH
Me Me
NMe2
OMe
O
N H HN O H Me NMe2
Me
N H 51: hemsine A
50: lotusine G
Ziziphus lotus
3: ziziphine A
Paliurus hemleyanus
Ziziphus oenoplia
Selected examples of aryl-alkyl ether peptide alkaloids.
OH R
3
OH H N
O
OH N H R2 O hypothetical precursor peptide
R1HN
H
elimination Michael-type addition
O
(a)
O
R3 1
R HN
R
O
R3 N H
2
(c)
−2H+, −2e−
HO
O
H N
−e− (SET)
(b)
OH
OH H N
1
R HN
R3
O
H O NR2
O
O
OH
N H
OH H N
1
R HN
O
H
O 2
O
R
H N H
OH
O
Ox
O R3
H
O
O O NH
HN
H OH
H
R2
−2 H+, −2 e−, −CO2
SET
O O NH
H OH
HN R1
O
OH
H
R3 HN
O O HN
−CO2, −H2O
H O O NH
R1HN
H NH
R1
R2
O R3
O
HO
O
R3 HN
HN R1
H
HO
O
HN R2
Scheme 9.12 Possible ring formations in the biosynthesis of aryl-alkyl ether peptide alkaloids, SET = single electron transfer.
R2
338
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids
be proposed, which differ only in the oxidation state of the intermediate aryl-ring (Scheme 9.12). Isolation of the dehydroamino-acid containing linear Rhamnaceae alkaloid lasiodine A led to speculation about a possible ring closure via endo-attack of a phenolic group to this double bond via Michael addition (Scheme 9.12, path a). This mechanism, however, is not fully suited to explain the ortho-meta orientation found in some other cyclopeptide alkaloids [60]. Oxidative pathways could more generally explain these patterns. Enzymatic oxidation of a tyrosine side chain could produce a radical cation that may be intercepted by a β-hydroxylated amino acid (Scheme 9.12, path b) and then further oxidized. Alternatively, a direct two-electron oxidation could be envisioned, forming an electrophilic para-quinone intermediate, which re-aromatizes after loss of water (Scheme 9.12, path c). In both cases a concomitant loss of the carboxyl group could ensue (cf. TMC-95A–D described in Chapter 10), potentially explaining the prevalence of an endocyclic (Z)-configured enamide found frequently among these peptide alkaloids. Owing to the limited availability of these alkaloids, several strategies for their total syntheses have been developed. No genuine biomimetic approach to the macrocyclization according to those already proposed (Scheme 9.12) has been established thus far. Ring closure as the key step in each of these synthetic routes has been accomplished at different positions of the molecular scaffold (Figure 9.9) [61]. Schmidt et al. have introduced a macrolactamization approach based on the activation of a carboxyl-group as a pentafluorophenyl ester, which was further applied to the synthesis of several cyclopeptide alkaloids (Schmidt et al. [62–69], Joulli´e et al. [70–76], Han et al. [77–79]). More biomimetic ring closures at the aromatic core have less frequently been realized. In their total synthesis of sanjonine G1 (52), Zhu et al. used an intramolecular SN Ar reaction of a 4-fluoro-3-nitroaryl-derivative with a β-hydroxy amino acid residue (53) and subsequent reductive removal of the activating nitro-group from 54 via the diazonium intermediate to obtain the desired cyclopeptide scaffold 55 (Scheme 9.13) [80–83]. A different method for macrocyclization was established by Evano, using a copper-catalyzed Ullmann-type vinyl amidation. This was applied to the synthesis of paliurine F (56, Scheme 9.14) [84–86]. The hydroxypyrrolidine fragment 57 was SNAr cyclization J. Zhu et al. 1999, 2002 (sanjoinine G1) 2005 (mauritine A, B, C and F)
O R3
Copper-mediated macroenamidation G. Evano et al. 2007 (paliurine F), 2007 (abyssenine A), 2009 (paliurine E and F, ziziphines N and Q, abyssenine A, mucronine E)
O O NH HN HN R2 R1 Macrolactamization U. Schmidt et al. 1981 (ziziphine A), 1983 (mucronine B), 1991 (frangulanine); M. M. Joulliè et al. 1992 (nummularine F), 1998 (sanjoinine G1), 1998 (frangufoline); B. H. Han et al. 1995 (sanjoinine G1)
Figure 9.9
Ring-closing strategies in the synthesis of aryl-alkyl ether peptide alkaloids.
9.3 Peptide Alkaloids Cyclized by Oxidation of Aryl Side Chains O 2N
NO2 Me
NBn2
Me
N H Me
OH O
OAc
F
O
H N
339
Me
i
O
ii,iii
O O
45%
Me Bn2N
OH
NH Me
HN
Me
73%
Me
OAc
O O O
Me Bn2N
HN 55
Me 53
NH Me
54
3 steps
Me 85%
52: sanjoinine G1
Scheme 9.13 Synthesis of sanjoinine G1 (52) [79]. Reagents and conditions: (i) TBAF, DMSO, 85 ◦ C, then Ac2 O, Et3 N, DMAP, CH2 Cl2 ; (ii) SnCl2 , DMF, 60 ◦ C; and (iii) NaNO2 , H3 PO2 , Cu2 O, THF-H2 O. OMe O
O
N
OMe I O
i-v H OTBS
Boc 57
O N
vi
O
50%
OMe
NH2
H HN H Boc Me 58
Me
70%
H
O
3 steps
O O NH N H HN H Boc Me
56: paliurine F 57%
Me
59
Scheme 9.14 Synthesis of paliurine F (56) [85]. Reagents and conditions: (i) [Ph3 PCH2 I]+ I− , NaHMDS, THF-HMPA, −78 ◦ C; (ii) TBAF, THF, −10 ◦ C to room temp. (rt); (iii) (COCl)2 , DMSO, Et3 N, CH2 Cl2 , −78 ◦ C; (iv) NaClO2 , NaH2 PO4 , 2-methylbut-2-ene, tBuOH–THF–H2 O, rt; (v) L-isoleucinamide, AcOH, EDC, HOBt, NMM, DMF, rt; and (vi) CuI (10%), N, N -dimethylethylenediamine, Cs2 CO3 , THF, 60 ◦ C.
constructed from N-Boc-d-serine and a chelation-controlled borohydride reduction. The (Z)-vinyl iodide in 58 was introduced by Stork–Zhao olefination, while isoleucinamide was coupled to the C-terminus of the linear precursor. For the macrocyclization step to 59, CuI/N,N -dimethylethylenediamine gave the best results, which was followed by appendage of l-isoleucine and N,N-dimethyl-l-leucine to afford paliurine F in 6.5% overall yield over 16 steps. 9.3.2 Cyclic Peptides Containing Biaryl Ethers
The biaryl ether is a common motif found in several constrained, biologically active cyclic peptides. The simplest ones are cyclic tripeptides containing isodityrosine 60 as a common element (Figure 9.10). Biosynthetically, they are derived from oxidative cyclization of a terminal tyrosine, either at the C-terminus (61) or at the N-terminus (62–70). These different biphenyl-ether linkages lead to diverse biological activities of otherwise related structures. While K-13 (61), isolated from Micromonospora halophytica [87] acts as a potent angiotensin I converting enzyme (ACE) inhibitor, its inhibition of aminopeptidase B is very weak [88]. The OF4949
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids
340
OR1
OH O
OR3 O
O
CO2H H2N
CO2H
NH2
H2N
OH
60 O
O
R2
N CO2H H H
O
H NH
O O
H N Me
Me
62: renieramide, 63: OF4949-I, 64: OF4949-II, O 65: OF4949-III, H N 66: OF4949-IV, AcHN N CO2H 67: eurypamide A, H H H O 68: eurypamide B, 69: eurypamide C, 70: eurypamide D, 61: K-13 OH
Figure 9.10
H
R4
Me N
O
H N
1
R =H, R1=Me, 1 R =H, 1 R =Me, R1=H, 1 R =H, 1 R =H, R1=H, 1 R =H,
2
R =CH(CH3)2 2 R =CH(OH)CONH2 2 R =CH(OH)CONH2 HN 2 R =CH2CONH2 2 R =CH2CONH2 2 R =CH(OH)CH(OH)CH2 N 2 H R =CH(OH)CH3 2 R =CH2OH 2 R =CH(OH)CH3
NH2
HN
O Me HN O N
Me
O
Me
OMe R3=H, R3=H, 3
R4=OH R4=H 4
71: bouvardin, 72: RA-V, 73: RA-VII, R =Me, R =H
Peptides containing the cyclic isodityrosine subunit.
compound class on the other hand (63–66), produced by Penicillium rugulosum, shows potent inhibition of aminopeptidase B [89]. Bouvardin (71) was first isolated from Bouvardia ternifolia (Rubiaceae) and features an 18-membered peptide ring next to the cycloisodityrosine unit (Figure 9.10) [90]. Several members of this peptide family have been found (deoxybouvardin, 6-O-methylbouvardin, RA-I to RA-XVII) and were investigated as potent antitumor leads [91, 92]. Bouvardin and RA-VII (73) bind to the eukaryotic ribosome and inhibit protein biosynthesis. In principle, two different approaches have been used to construct the cycloisodityrosine ring. Either the biaryl ether linkage has been formed prior to ring closure by macrolactamization or biaryl-ether formation is used to cyclize a linear peptide precursor (Figure 9.11) [93–98]. Ullmann-coupling Schmidt et al. 1988, Evans et al. 1989, Boger et al. 1990 Rao et al. 1992, Lygo et al. 2004 SN Ar-cyclization Zhu et al. 2000
AcHN
H N H
O
Figure 9.11
TTN-oxidative coupling Yamamura et al. 1988, 1989 OR1 OH Nishiyama et al. 2003, 2004, 2006 Nucleophilic addition to quinone O O Rao et al. 1993 SN Ar-cyclization Zhu et al. 1994 O O H Ru-mediated SN Ar-cyclization N Rich et al. 1997 H2N N COOH N COOH H H H H H O R2 Negishi-cross coupling Jackson et al., 2000, 2009 OH Macrocyclization Schmidt et al. 1988, Evans et al. 1989 Boger et al. 1990, Rao et al. 1992 Lygo et al. 2004, Jackson et al. 2009
Strategies for cyclic isodityrosine synthesis.
9.3 Peptide Alkaloids Cyclized by Oxidation of Aryl Side Chains
341
Alternatively, Negishi couplings [99] or Horner–Wadsworth–Emmons type olefinations [100] were used to cyclize the open-chain peptide. Nishiyama and Yamamura applied thallium trinitrate (TTN) mediated cyclization of halogenated bis-phenols as a biomimetic way to access cycloisodityrosine structures [101, 102]. The synthesis of eurypamide B (68, first isolated from Microciona eurypa) [103] began with the formation of a linear tripeptide (74) containing one diiodo- and one dibromotyrosine derivative, which served to direct a regioselective oxidation to 75 with TTN (Scheme 9.15). Dehalogenation using hydrogenolysis and deprotection yielded the target compound 68 without any observable racemization [104]. OH I
OH
OH I O
Br Me
OH H N
N H H O NHBoc 74
Br
I
O
O ii, iii
i Br H H O N
72% BocHN COOMe
OH
Br
H
90%
N COOMe H H Me
O HO 75
H2N
H HO N
N COOH H H O Me HO 68: eurypamide B H
Scheme 9.15 Synthesis of eurypamide B (68) [103]. Reagents and conditions: (i) thallium trinitrate (TTN), THF–MeOH; (ii) H2 , NaOAc, 10% Pd/C, MeOH; and (iii) 1 M NaOH, MeOH, then TFA, CH2 Cl2 .
Biaryl ethers can also be formed by an Ullmann reaction, where a phenolic hydroxyl group is made to couple with an aryl iodide upon addition of CuBr · SMe2 [105–107]. Other approaches have used SN Ar coupling to an electron-deficient aromatic ring, which could be achieved using electron-withdrawing substituents [108] or through the intermediate formation of a metal π-arene complex. Rich et al. applied an intramolecular SN Ar reaction of a preformed cationic ruthenium complex as the key step in their synthesis of OF4949-III and K-13 (Scheme 9.16) [109]. 3-Chlorotyrosine (76) was converted into the cationic cyclopentadienylruthenium complex 77 and coupled to the protected Tyr-Tyr dipeptide 78. The resultant cyclization precursor 79 smoothly formed the biaryl ether upon exposure to weak, non-nucleophilic base. Liberation of the auxiliary Ru-fragment by UV-irradiation and demethylation gave K-13 (61) in eight steps. Figure 9.12 summarizes the main approaches to the bouvardin scaffold. Again, macrolactamization, SN Ar-[110] and Ullmann macrocyclization [107] have been used to furnish this peptide alkaloid. The first total synthesis of RA-VII was achieved via a potentially biomimetic route using TTN-mediated coupling of halogenated aryl-phenols. However, low yields hampered the critical cyclization step [111]. More recent reports describe the synthesis of the cycloisodityrosine by copper(II)-mediated O-arylation of phenylboronic acids [112]. A class of compounds that is biosynthesized by oxidation of bromotyrosine derivatives is the bastadins [113]. More than 20 members of this family have been isolated from marine Ianthella sponges. Bastadins occur in linear or cyclic form, possessing one biaryl or one or two biaryl-ether bonds (Scheme 9.17) as well as (E)-configured oximes. Several bastadins exhibit antimicrobial, cytotoxic,
342
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids OH
Cl
Cl OH
i-iv
RuCp COOMe
H2N
COOH
+
PF6
61% H2N
OMe
OMe
76
O N H
AcHN
77
v
65%
OH
Cl
OH
O
COOH
78
PF6
vi-viii AcHN
H N H
O
O
H N
50% N COOH H H
OMe
AcHN
RuCp
O N H
O
OH
COOMe
OMe
61: K-13
79
Scheme 9.16 Synthesis of K-13 (61) via transition-metal activated SN Ar [108]. Reagents and conditions: (i) Boc2 O, NaOH–dioxane; (ii) TMS-CHN2 (4.0 equiv), THF–MeOH; (iii) 4 N HCl, dioxane, then NaHCO3 ; (iv) RuCp(CH3 CN)3 PF6 , ClCH2 CH2 Cl, reflux; (v) EDCI, HOBt, DMF, 0 ◦ C; (vi) sodium 2,6-di-tBu-phenoxide, THF; (vii) hν, CH3 CN; and (viii) AlBr3 , EtSH. TTN-mediated oxidation T. Inoue et al. CPB 1995, 43, 8, 1325 Cu (I)-mediated boronic acid coupling K. Takeya et al. TL 2003, 44, 5901 H.-D. Arndt et al., Synlett 2009, 720 SNAr-cyclization D.L. Boger et al. BMCL 1995, 5, 1187 J. Zhu et al. JOC 1996, 61, 771
OR1 O H
R2
Me N
O
O O
NH
H N Me
Me
Macrolactamization T. Inoue et al. CPB 1995, 43, 8,1325 J. Zhu et al. JOC 1999, 64, 17, 6283 D.L. Boger et al. JACS 1991, 113, 1427
HN Me
SNAr-cyclization J. Zhu et al. JOC 1999, 64, 17, 6283 Ullmann macrocyclization D.L. Boger et al. JACS 1991, 113, 1427
O Me HN O N
Me
Bouvardin R1 = H, R2 = OH RA-V R1 = H, R2= H RA-VII R1 = Me, R2= H
O OMe
Figure 9.12 (72,73).
Different strategies to synthesize the bouvardin (71) and RA-family scaffolds
and anti-inflammatory activity, as well as interesting modulation of intracellular calcium channels [114–116]. The bastadin 6 (83) biosynthesis is presumed to proceed via condensation of two brominated tyrosines followed by double oxidation of the primary amine yielding an N,N -dihydroxylated intermediate, which forms the oxime upon dehydration (Scheme 9.17). It is then assumed that the resulting hemibastadin is dimerized by oxidation-catalyzed biaryl ether formation. A second oxidative coupling reaction completes the bastadin macrocycle [117]. Analogous to the biosynthetic pathway three main protocols have been established for bastadin synthesis that apply an oxidative coupling (Scheme 9.18). Yamamura et al. used TTN for aryl ether formation from 84 (Scheme 9.18, A)
9.3 Peptide Alkaloids Cyclized by Oxidation of Aryl Side Chains OH
Br HO
Br
Br Br
OH
HO
Br
Br
+ 2[O]
O
Br
x2
Br
−H2O
N H
O
H (E) N Br
N
O
Br
Br
−HBr
O
−
Ar
OH Br
N
OH
N H
N
OH
OH 82
O Br
Br
O
O
Br Br
NHBoc
NaH, DMF (D),
O
OR
Br
− − HBr
CHO
NHBoc 89
O
86
90: bastadin 12
OR
O
NHBoc
85 OR
Br
88
Br
HN NHBoc
84
I
O
Br
cerium ammonium nitrate (C)
NHBoc
Br
Br
+
Br 83: bastadin 6
CHO
O O
Br
Br NHBoc 87
Scheme 9.18
O
+H 2e– − −Br +
O N H
TTN (A) or horseradish peroxidase, H2O2 (B) or
Br
OHC
Br
Bastadin 6 (83) and its presumed biosynthesis.
OH Br
Br
Br
83: bastadin 6
Scheme 9.17
+ O
O
Ar
N
OH
+
OH Br
N (E) H
Br O
O
Ar
O Br Br
Br
Br
O
HO
N
H N
OH
OH
O
Ar
N
OH 81: hemibastadin
80
N
H N
−2 H+ − −2 e
N H
NH2
343
O N H
O
Biomimetic synthesis approaches to bastadin aryl ethers.
[118, 119]. The second protocol from Shi et al. (Scheme 9.18, B) employed a horseradish-peroxidase-mediated coupling, which enabled the synthesis of bastadins 2, 3, and 6 [120]. More recently, Kobayashi et al. reported the synthesis of bastadin 6 via a Ce(IV)-mediated oxidative coupling (Scheme 9.18, C) [121]. All these methods proceed through radical intermediate 85. Alternatively, different members of the bastadin family have been accessed via phenol coupling of aryl iodonium salts (84→89) [122].
344
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids
9.3.3 Cyclopeptides Containing Biaryls
Biphenomycins 91–93 are cyclopeptides containing a meta-meta biaryl unit which bridges a tripeptide by forming a 15-membered ring (Figure 9.13). The compounds have been isolated from broths of Streptomyces filipinensis and Streptomyces griseorubuginosus and display potent antibacterial activities against various β-lactam resistant bacteria in vivo [123, 124]. Although very little is known about the biosynthesis of these biaryl peptide alkaloids, one could assume that a cytochrome-mediated oxidation of one aryl group forms either a radical or cationic intermediate, which is intercepted by the second aryl group to forge the biaryl linkage (Scheme 9.19). However, the generation of the uncommon ortho-hydroxylation remains unclear and might involve more complex biosynthetic events. A biomimetic analogon of this reaction was realized by applying an intramolecular Suzuki–Miyaura coupling to cyclize a linear precursor tripeptide by forming the HO
OH O
H N
H2N
N H H OH O
O
HO
OH
R OH
H N
H2N
OH H N
O
N H H OH O
O
OH
O N H
OH O NH
HN NH2
NH2
91: biphenomycin A (R = OH) 92: biphenomycin B (R = H)
Figure 9.13
NH2
93: biphenomycin C
Structures of the biphenomycins A–C (91–93). FeO OH
HO H N
H2N
O N H
O
H N
H2N O
HO
OH
O N H H OH O
O
NH2
H N
H2N
NH2
NH2
H
OH OH H N
H2N O
O
O
O
OH H N
FeIV N H H O OH NH2
Scheme 9.19
N H H OH O
O
−e−
[O]
HO
O
−e− + −H
H2N O
O
FeIV
O
O N H H OH O
H N
H2N
NH2
Possible biosynthetic routes to biphenomycin biaryls.
O
H O N H H OH O NH2
9.3 Peptide Alkaloids Cyclized by Oxidation of Aryl Side Chains Oi Pr I
i PrO O
H N
BocHN O
94
i PrO
Oi Pr
i
Bpin OMe N H H O O Me Me N Cbz MeO
O
H N
50%
BocHN
N H H O O Me
O
96
Cy P
Cy OMe
PdCl2
ii
N Cbz
OMe
Me
95%
2
95
92
Scheme 9.20 Synthesis of biphenomycin B (92) by intramolecular Suzuki coupling [125]. Reagents and conditions: (i) 95 (0.06 equiv), K2 CO3 , T(µW); and (ii) BCl3 , CH2 Cl2 , 0 ◦ C, then 2 M LiOH, rt.
biaryl linkage (Scheme 9.20) [125]. Not unexpectedly, most synthetic approaches featured macrolactamization at the critical ring closing step [126, 127]. In this synthetic route, coupling of three non-proteinogenic amino acids yields the linear tripeptide 94 containing an arylboronate and aryl iodide at the C- and N-terminus, respectively. The aryl–aryl bond in 96 was formed by microwave-assisted Suzuki– Miyaura cross-coupling triggered by Buchwald’s palladium catalyst (95), which proceeded in 50% yield after extensive optimization. An intramolecular Suzuki–Miyaura cross coupling reaction was also employed for the key ring-forming step in the total synthesis of arylomycin A2 (97, Scheme 9.21). This compound is a biaryl-bridged lipohexapeptide isolated from Streptomyces T¨u 6075 [128], displaying antibacterial activity against various Gram-positive bacteria by inhibiting the bacterial type I signal peptidase [129]. Romesberg et al. compared a route with final macrolactamization to a more biomimetic route (Scheme 9.21) featuring a final cyclization by biaryl bond formation [130]. In fact, poor yields were observed on attempted macrolactam formation from 98 to 99, owing to the 14-membered ring in the final product as well as the strained biphenyl linkage topology. Biaryl coupling from 100 was more successful, likely assisted by peptide backbone pre-organization and the templating effect of the palladium catalyst. 9.3.4 Vancomycin
Probably the most complex and most well studied example of an aryl-oxidized peptide alkaloid is vancomycin (101, Figure 9.14). This prominent glycopeptide is one of the most important agents against multidrug-resistant Gram-positive bacterial pathogens [131]. It acts as an inhibitor of peptidoglycan crosslinking during bacterial cell wall biosynthesis.
345
346
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids
MeO MeO
COOMe NH2
BocHN O
I
CO2H
49%
<10%
Me
BocHN
H N H
O
98
OMe
ii
i
H N
Bpin
OMe MeO MeO
HN
CO2Me
O
H N
BocHN O
O
N H
Me
CO2Me
Me 99
100 HO HO
steps Me N
Me
O N H OH
8
Me
Me
O
O
H N
N H Me O
O
HN
H N
CO2H O
Me
97: arylomycin A2
Scheme 9.21 Comparative syntheses of the arylomycin A2 core [130]. Reagents and conditions: (i) EDC, HOBt, DMF; and (ii) PdCl2 (PPh3 )2 , K2 CO3 , CH3 CN.
HO HO
1. monooxygenase (oxyB) Me Me O
O
O D
H N H H
O
O Cl H N O
HN
E O
B
H N H
OH H N O
O
H O
A HO
OH OH
3. monooxygenase (oxyC) 101 : vancomycin
Figure 9.14
OH
HO
NH2 O D-3,5-dihydroxyphenylglycine
OH
NH
NH2
HOOC
HO
NH2 O D-4-hydroxyphenylglycine
Cl
O
C HO
HO
2. monooxygenase (oxyA)
O
NH2 O
OH
OH
OH
NHMe O
Me Me
HO
OH NH2 D-β-hydroxytyrosine
Structure of vancomycin and some of its uncommon amino acids.
The biosynthesis of vancomycin was studied in quite some detail. A multi-domain non-ribosomal peptide synthetase (NRPS) builds up the hexapeptide core of this complex molecule via a thioester-templated mechanism [132, 133]. This hexapeptide contains several non-proteinogenic amino acids: a 4-hydroxyphenylglycine, 3,5-dihydroxyphenylglycine, and β-hydroxytyrosine (Figure 9.15) [134–136]. After chlorination at the β-hydroxytyrosine aryl side-chains, the biaryl- and biaryl ether oxidative coupling steps are accomplished while the precursor peptide is still attached to the NRPS assembly line [137, 138]. This process is catalyzed by three homologous P450 oxygenases (oxyABC) [139]. Gene knockout experiments
9.3 Peptide Alkaloids Cyclized by Oxidation of Aryl Side Chains
Glycosidation (Nicolaou et al.)
CuI-mediated cyclization (Nicolaou et al.) SNAr-cyclization Cl (Evans et al.)
R
CuI-mediated cyclization (Nicolaou et al.) SNAr-cyclization (Evans et al.)
O O
O D
C HO O
Macrolactamization (Nicolaou et al.)
H N H H
O
Cl H N
HN
E OH
O
O
H N
N H H
NH O
H N
O
O
B
Me Me
NH2
HOOC OH
A
OH HO
Me
Oxidative cyclization (Evans et al.) Suzuki coupling (Nicolaou et al.)
Peptide fragment coupling (Evans et al., Nicolaou et al.)
Figure 9.15 Overview of the key disconnections in vancomycin (101) synthesis (R = disaccharide).
OH
N N
FeIII N
102
N
N N
O2 2 x e−
HO
O FeIV N
O
H2O
HO
H
O
N H H
H N R
O
O
iii)
O
OH
N H H
O
N H H
O
OH HO
O
H N R
O
N H H
O
Fe
N
N N
O
O N H H
H N R
OH
R
Fe
N N
OH
O HO
H N
N H H
107 N
N
O
O
106
105
N
N H H
ii)
HO
O
R
104 i)
O
H N
N H H
103
OH
OH
O
N
347
O
N H H
108
Scheme 9.22 Possible mechanisms of biaryl-ether formation during the biosynthesis of vancomycin.
N N
Fe
N N
O
N H H
348
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids
specified the order of the oxidative coupling steps [140, 141]. A heme-containing oxygenase (oxyB) couples ring C and D, followed by the oxidative coupling of ring D and E (oxyA), while the final biaryl linkage is furnished by oxyC. X-Ray structures of OxyB and C show a cytochrome P450-type fold featuring a cysteine-bound heme iron [142, 143]. Labeling experiments with OxyB-catalyzed reactions in the presence of 18 O2 showed no 18 O incorporation into the product [144, 145]. By analogy to other P450 enzymes, this indicates a reaction mechanism that proceeds via dioxygen addition to a high-spin Fe(III) species (102→103, Scheme 9.22). The remaining transformations (104→108) are still under investigation [145]. Possible pathways include the formation of an arene epoxide (105), which is opened by an intramolecular attack of a phenol (Scheme 9.22, path i). Alternatively, a radical could be formed at one of the aromatic rings (106), which adds to a second aryl ring and then abstracts a hydroxyl radical from the Fe-heme complex (path ii). Finally, enzyme-mediated formation of diradical 107 could ensue, followed by coupling (path iii).
Cl
O
Br
O
i
H N
H
NHBoc
34%
EtOOC
N3
B
BnO MeO
NHBoc O
B
BnO
OMe OMe
A
Cl H O H N N H H
TBSO
O
N3
Br D
C
D
TBSO
H
N
N
OH Br
C
EtOOC
N N
N N
OMe OMe
A
MeO 110
109 N N
N N
N O
HOOC
NHBoc
86%
HO O
O NH2 A
B
OMe OMe 111
Br D
C ii
Cl H O H N N H H
BnO MeO
O
D
C
HO
N Br
O Cl H N
H N H
H
NHBoc O
NH B
BnO A
MeO
OMe OMe
112
Scheme 9.23 Ring closures in Nicolaou’s synthesis of vancomycin [146–151]. Reagents and conditions: (i) CuBr · Me2 S, K2 CO3 , pyridine, MeCN; and (ii) FDPP, iPr2 NEt, DMF.
9.3 Peptide Alkaloids Cyclized by Oxidation of Aryl Side Chains
The complex structure of vancomycin (101) poses an enormous challenge for synthesis (Figure 9.15). The main demands of a successful synthetic route include the asymmetric synthesis of the amino acid building blocks, their assembly, and most importantly the control over the atropisomerism ensuing from hindered rotation about the aryl–aryl and aryl–O-aryl axes. Two main strategies have been developed for the synthesis of the vancomycin aglycon. The first, developed by the group of K. C. Nicolaou [146–148], is based on a triazene driven copper-mediated coupling reaction of 109 to form the C-O-D ring 110 (Scheme 9.23) [149–151]. This coupling was not atropisomer-selective, but the undesired isomer could be separated. The A-B ring (112) was then closed starting from 111 by macrolactamization in an excellent yield and with high atropselectivity,
Cl
Cl F
F
C
OH
O
H N H
C
OH O H O N NH H MeHNOC O2N
O2N
i
NHTfa
65% dr > 95:5
OBn
MeHNOC
NH
A
A
OMe
OMe
OMe OMe
MeO
113
114
Cl HO
O2N O
C
OH
O
H N H
NH
MeHN
NO2
OAll
F
A
OMs
H N H OH O
HO ii
NHBoc
O
NH
A
OMe OMe 116
OPiv O
C HO O
H O
O NH
MeHN
N H
Cl H O H N N H H
HO iii, iv
NHTfa O
54% dr > 95:5
O
NH
O
NHTfa O
B
x
B
H N H
OMs D
C
D
NHBoc
OPiv O
OMs
Cl
H N H OTf O
B
MeO
OMs D
Cl
O
H N H
O
79% dr = 5:1 MeHN
OMe OMe 115
MeO
OAll O
C
D
O
B
NHTfa OH
B
B
MeO
O
A MeO
OMe OMe 117
MeHN H HO
A
OH OH 118
Scheme 9.24 Ring closures in Evans’ synthesis of vancomycin [152, 153]. Reagents and conditions: (i) VOF3 , BF3 · Et2 O, AgBF4 , TFA-CH2 Cl2 , then NaHB(OAc)3 ; (ii) Na2 CO3 , DMSO, 1.5 h, then Tf2 NPh; (iii) AlBr3 , then EtSH; and (iv) MeOH, 60 ◦ C.
349
350
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids
probably benefiting from very favorable conformational preorganization. Subsequently, the preformed tripeptide containing a phenolic hydroxyl was coupled to the C-O-D ring to set the stage for the final D-O-E ring closure, which was realized using a copper-mediated reaction of a phenolic hydroxyl and the triazene-activated aromatic ring D. The second synthetic route to the vancomycin aglycon was established by the group of D. A. Evans (Scheme 9.24) [152, 153]. In this approach, the AB ring 114 was formed first by a biomimetic oxidative biaryl coupling of 113 using VOF3 . This interesting transformation was atropselective (19 : 1), but unfortunately in the wrong stereochemical sense! A building block containing the aryl-ring D was then coupled (115) and the electron-withdrawing nitro-group on ring C then allowed for a SN Ar cyclization, which preferentially formed the desired atropisomer of the C-O-D macrocycle 116. In the biosynthetic pathway, the global structure of the molecule directs the geometry of rings A and B in the last oxidation step catalyzed by OxyC. During synthesis, gentle warming gave the correct AB ring atropisomer 118 in a ratio of 95 : 5 – after the methyl ethers had been cleaved from 117, which prohibited atropisomer interconversion. The strong directing character of the entire molecular scaffold was illustrated by thermal equilibration of the isolated biaryl, which favors the ‘‘non-natural’’ atropisomer in a ratio of 2 : 1. Dedicated atropisomer equilibrations were then extensively used in Boger’s total synthesis of vancomycin [154].
References 1. (a) Piel, J. (2004) Nat. Prod. Rep., 21,
2. 3. 4. 5. 6.
7.
8.
9.
519–538; (b) Schmidt, E.W. (2007) Nat. Chem. Biol., 4, 466–473; (c) Piel, J. (2009) Nat. Prod. Rep., 26, 338–362. Nolan, E.M. and Walsh, C.T. (2009) ChemBioChem, 10, 34–53. Sieber, S.A. and Marahiel, M.A. (2005) Chem. Rev., 105, 715–738. Oman, T.J. and van der Donk, W.A. (2010) Nat. Chem. Biol., 6, 9–18. Yoder, R.A. and Johnston, J.N. (2005) Chem. Rev., 105, 4730–4756. Milne, J.C., Roy, R.S., Eliot, A.C., Kelleher, N.L., Wokhlu, A., Nickels, B., and Walsh, C.T. (1999) Biochemistry, 38, 4768–4781. Li, Y.-M., Milne, J.C., Madison, L.L., Kolter, R., and Walsh, C.T. (1996) Science, 274, 1188–1193. Roy, R.S., Gehring, A.M., Milne, J.C., Belshaw, P.J., and Walsh, C.T. (1999) Nat. Prod. Rep., 16, 249–263. Hughes, R.A. and Moody, C.J. (2007) Angew. Chem., 119, 8076–8101; Angew. Chem. Int. Ed., 46, 7930–7954.
10. Wipf, P. (1995) Chem. Rev., 95,
2115–2134. 11. Parsons, R.L. and Heathcock, C.H. Jr.
(1994) J. Org. Chem., 59, 4733–4734. 12. You, S.-L., Razavi, H., Kelly, and J.W.
13. 14.
15. 16.
17. 18. 19.
(2003) Angew. Chem., 115, 87–89; Angew. Chem. Int. Ed., 42, 83–85. You, S.-L. and Kelly, J.W. (2003) J. Org. Chem., 68, 9506–9509. Jansen, R., Kunze, B., Reichenbach, H., Jurkiewicz, E., Hunsmann, G., and H¨ofle, G. (1992) Liebigs Ann. Chem., 357–359. Jansen, R., Schomburg, D., and H¨ofle, G. (1993) Liebigs Ann. Chem., 701–704. Boyce, R.J., Mulqueen, G.C., and Pattenden, G. (1995) Tetrahedron, 51, 7321–7330. Wipf, P. and Venkatraman, S. (1995) J. Org. Chem., 60, 7224–7229. Akaji, K. and Kiso, Y. (1999) Tetrahedron, 55, 10685–10694. Pattenden, G., Thom, S.M., and Jones, M.F. (1993) Tetrahedron, 49, 2131–2138.
References 20. Donia, M.S., Ravel, J., and Schmidt,
21.
22.
23.
24. 25. 26.
27.
28.
29.
30.
31. 32.
33. 34.
35.
36.
37.
E.W. (2008) Nat. Chem. Biol., 4, 341–343. Donia, M.S., Hathaway, B.J., Sudek, S., Haygood, M.G., Rosovitz, M.J., Ravel, J., and Schmidt, E.W. (2006) Nat. Chem. Biol., 2, 729–735. Wipf, P., Fritch, P.C., Geib, S.J., and Sefler, A.M. (1998) J. Am. Chem. Soc., 120, 4105–4112. Milne, B.F., Long, P.F., Starcevic, A., Hranueli, D., and Jaspars, M. (2006) Org. Biomol. Chem., 4, 631–638. Wipf, P. and Fritch, P.C. (1996) J. Am. Chem. Soc., 118, 12358–12367. Wipf, P. and Miller, C.P. (1993) J. Org. Chem., 58, 3604–3606. Bagley, M.C., Dale, J.W., Merritt, E.A., and Xiong, X. (2005) Chem. Rev., 105, 685–714. Priestley, N.D., Smith, T.M., Shipley, P.R., and Floss, H.G. (1996) Bioorg. Med. Chem., 4, 1135–1147. Frenzel, T., Zhou, P., and Floss, H.G. (1990) Arch. Biochem. Biophys., 278, 35–40. Mocek, U., Zeng, Z., O’Hagan, D., Zhou, P., Fan, L.D.G., Beale, J.M., and Floss, H.G. (1993) J. Am. Chem. Soc., 115, 7992–8001. Smith, T.M., Priestley, N.D., Knaggs, A.R., Nguyen, T., and Floss, H.G. (1993) J. Chem. Soc., Chem. Commun., 1612–1614. Kelly, W.L., Pan, L., and Li, C. (2009) J. Am. Chem. Soc., 131, 4327–4334. Yu, Y., Duan, L., Zhang, Q., Liao, R., Ding, Y., Pan, H., Wendt-Pienkowski, E., Tang, G., Shen, B., and Liu, W. (2009) ACS Chem. Biol., 4, 855–864. Li, C. and Kelly, W.L. (2010) Nat. Prod. Rep., 27, 153–164. Arndt, H.-D., Schoof, S., and Lu, J.-Y. (2009) Angew. Chem., 121, 6900–6904; Angew. Chem. Int. Ed., 48, 6770–6773. McIntosh, J.A., Donia, M.S., and Schmidt, E.W. (2009) Nat. Prod. Rep., 26, 537–559. Bycroft, B.W. and Gowland, M.S. (1978) J. Chem. Soc., Chem. Commun., 256–258. Nicolaou, K.C., Safina, B.S., Zak, M., Estrada, A.A., and Lee, S.H.
38.
39.
40.
41.
42.
43. 44.
45.
46.
47.
48.
49.
50.
51.
(2004) Angew. Chem., 116, 5197–5202; Angew. Chem. Int. Ed., 43, 5087–5092. Nicolaou, K.C., Zak, M., Safina, B.S., Lee, S.H., and Estrada, A.A. (2004) Angew. Chem., 116, 5202–5207; Angew. Chem. Int. Ed., 43, 5092–5097. Nicolaou, K.C., Safina, B.S., Zak, M., Lee, S.H., Nevalainen, M., Bella, M., Estrada, A.A., Funke, C., Zecri, F.J., and Bulat, S. (2005) J. Am. Chem. Soc., 127, 11159–11175. Nicolaou, K.C., Zak, M., Safina, B.S., Estrada, A.A., Lee, S.H., and Nevalainen, M. (2005) J. Am. Chem. Soc., 127, 11176–11183. Mori, T., Higashibayashi, S., Goto, T., Kohno, M., Satouchi, Y., Shinko, K., Suzuki, K., Suzuki, S., Tohmiya, H., Hashimoto, K., and Nakata, M. (2008) Chem. Asian J., 3, 984–1012. Mori, T., Higashibayashi, S., Goto, T., Kohno, M., Satouchi, Y., Shinko, K., Suzuki, S., Tohmiya, H., Hashimoto, K., and Nakata, M. (2008) Chem. Asian J., 3, 1013–1025. Wulff, G. and Klinken, H.T. (1992) Tetrahedron, 48, 5985–5990. Wulff, G. and B¨ohnke, H. (1986) Angew. Chem., 98, 101–102; Angew. Chem. Int. Ed. Engl., 25, 90–92. Wulff, G., Lindner, H.-J., B¨ohnke, H., Steigel, A., and Klinken, H.-T. (1989) Liebigs Ann. Chem., 527–531. Moody, C.J., Hughes, R.A., Thompson, S.P., and Alcaraz, L. (2002) Chem. Commun., 1760–1761. Sasaki, H., Irie, R., Hamada, T., Suzuki, K., and Katsuki, T. (1994) Tetrahedron, 50, 11827–11838. Boyd, D.R., Davies, R.J.H., Hamilton, L., McCullough, J.J., Malone, J.F., Porter, H.P., Smith, A., Carl, J.M., Sayer, J.M., and Jerina, D.M. (1994) J. Org. Chem., 59, 984–990. Bushman, D.R., Sayer, J.M., Boyd, D.R., and Jerina, D.M. (1989) J. Am. Chem. Soc., 111, 2688–2691. De Pietro, M.T., Marazzi, A., Sosio, M., Donadio, S., and Lancini, G. (2001) J. Antibiot., 54, 1066–1071. Morris, R.P., Leeds, J.A., N¨ageli, H.-U., Oberer, L., Memmert, K., Weber, E., LaMarche, M.J., Parker, C.N., Burrer, N., Esterow, S., Hein, A.E.,
351
352
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids
52.
53.
54.
55.
56. 57.
58.
59.
60.
61. 62.
63.
64.
65.
66.
Schmitt, E.K., and Krastel, P. (2009) J. Am. Chem. Soc., 131, 5946–5955. Nicolaou, K.C., Zou, B., Dethe, D.H., Li, D.B., and Chen, D.Y.-K. (2006) Angew. Chem., 118, 7950–7956; (2006) Angew. Chem. Int. Ed., 45, 7786–7792. Nicolaou, K.C., Dethe, D.H., Leung, G.Y.C., Zou, B., and Chen, D.Y.-K. (2008) Chem. Asian J., 3, 413–429. M¨uller, H.-M., Delgado, O., and Bach, T. (2007) Angew. Chem., 119, 4855–4858; Angew. Chem. Int. Ed., 46, 4771–4774. Delgado, O., M¨uller, H.-M., and Bach, T. (2008) Chem. Eur. J., 14, 2322–2339. Tan, N.-H. and Zhou, J. (2006) Chem. Rev., 106, 840–895. El-Seedi, H.R., Zahra, M.H., Goransson, U., and Verpoorte, R. (2007) Phytochem. Rev., 6, 143–165. Gournelis, D.C., Laskaris, G.G., and Verpoorte, R. (1997) Nat. Prod. Rep., 14, 75–82. Baig, M.A., Banthorpe, D.V., Coleman, A.A., Tampion, M.D., Tampion, J., and White, J.J. (1993) Phytochemistry, 34, 171–174. Marchand, J., Pa¨ıs, M., Moseur, X., and Jarreau, F.-X. (1969) Tetrahedron, 25, 937–954. Joulli´e, M.M. and Richard, D.J. (2004) Chem. Commun., 2011–2015. Schmidt, U., Lieberknecht, A., B¨okens, H., and Griesser, H. (1981) Angew. Chem., 93, 1121–1122; Angew. Chem. Int. Ed. Engl., 20, 1026–1027. Schmidt, U., Griesser, H., Lieberknecht, A., and Talbiersky, J. (1981) Angew. Chem., 93, 271–272; Angew. Chem. Int. Ed Engl., 20, 280–281. Schmidt, U., Lieberknecht, A., Griesser, H., and H¨ausler, J. (1981) Angew. Chem., 93, 272–273; Angew. Chem. Int. Ed. Engl., 20, 281–282. Schmidt, U., Lieberknecht, A., Grisser, H., and Talbiersky, J. (1982) J. Org. Chem., 47, 3261–3264. Schmidt, U. and Schanbacher, U. (1983) Angew. Chem., 93, 150–151; Angew. Chem. Int. Ed. Engl., 22, 152–153.
67. Schmidt, U., Lieberknecht, A., B¨ okens,
68. 69.
70.
71. 72.
73.
74.
75. 76.
77.
78.
79.
80.
81. 82.
83.
84.
85.
H., and Griesser, H. (1983) J. Org. Chem., 48, 2680–2685. Schmidt, U. and Schanbacher, U. (1984) Liebigs Ann. Chem., 1205–1215. Schmidt, U., Z¨ah, M., and Lieberknecht, A. (1991) J. Chem. Soc., Chem. Commun., 1002–1004. Nutt, R.F., Chen, K.M., and Joulli´e, M.M. (1984) J. Org. Chem., 49, 1013–1021. Heffner, J. and Joulli´e, M.M. (1989) Tetrahedron Lett., 30, 7021–7024. Heffner, J., Jiang, J., and Joulli´e, M.M. (1992) J. Am. Chem. Soc., 114, 10181–10189. Jiang, J., Li, W.R., Przeslawski, R.M., and Joulli´e, M.M. (1993) Tetrahedron Lett., 34, 6705–6708. Williams, L., Zhang, Z., Shao, F., Carroll, P.J., and Joulli´e, M.M. (1996) Tetrahedron, 52, 11673–11694. East, S.P. and Joulli´e, M.M. (1998) Tetrahedron Lett., 39, 7211–7214. East, S.P., Shao, F., Williams, L., and Joulli´e, M.M. (1998) Tetrahedron, 54, 13371–13390. Han, B.H., Kim, Y.C., Park, M.K., Park, J.H., Go, H.J., Yang, H.O., Suh, D.Y., and Kang, Y.C. (1995) Heterocycles, 41, 1909–1914. Suh, D.Y., Kim, Y.C., Han, Y.M., and Han, B.H. (1996) Heterocycles, 43, 2347–2351. Kim, Y.-A., Shin, H.-N., Park, M.-S., Cho, S.-H., and Han, S.-Y. (2003) Tetrahedron Lett., 44, 2557–2560. Temal-La¨ıb, T., Chanstanet, J., and Zhu, J. (2002) J. Am. Chem. Soc., 124, 583–590. Temal-La¨ıb, T. and Zhu, J. (1999) Tetrahedron Lett., 40, 83–86. Temal-La¨ıb, T., Bois-Choussy, M., and Zhu, J. (2000) Tetrahedron Lett., 41, 7645–7648. Cristau, P., Temal-La¨ıb, T., Bois-Choussy, M., Martin, M.-T., Vors, J.-P., and Zhu, J. (2005) Chem. Eur. J., 11, 2668–2679. Toumi, M., Couty, F., and Evano, G. (2007) Angew. Chem., 119, 578–581; Angew. Chem. Int. Ed., 46, 572–575. Evano, G., Toumi, M., and Coste, A. (2009) Chem. Commun., 4166–4175.
References 86. Toumi, M., Rincheval, V., Young, A.,
87. 88. 89.
90.
91.
92.
93.
94.
95. 96. 97.
98.
99.
100.
101.
102.
Gergeres, D., Turos, E., Couty, F., Mignotte, B., and Evano, G. (2009) Eur. J. Org. Chem., 3368–3386. Kase, H., Kaneko, M., and Yamada, K. (1987) J. Antibiot., 40, 450–454. Yasuzawa, T., Shirahata, K., and Sano, H. (1987) J. Antibiot., 40, 455–458. Sano, S., Kuroda, H., Ueno, M., Yoshikawa, Y., Nakamura, T., and Obayashi, A. (1987) J. Antibiot., 40, 519–525. Jolad, S.D., Hoffman, J.J., Torrance, S.J., Wiedhopf, R.M., Cole, J.R., Arora, S.K., Bates, R.B., Gargiulo, R.L., and Kriek, G.R. (1977) J. Am. Chem. Soc., 99, 8040–8044. Itokawa, H., Takeya, K., Mihara, K., Mori, N., Hamanaka, T., and Sonobe, T. (1983) Chem. Pharm. Bull., 31, 1424–1427. Itokawa, H., Takeya, K., Mori, N., Hamanaka, T., Sonobe, T., and Mihara, K. (1984) Chem. Pharm. Bull., 32, 284–290. Burgess, K., Lim, D., Bois-Choussy, M., and Zhu, J. (1997) Tetrahedron Lett., 38, 3345–3348. Nolasco, L., P´erez-Gonz¨alez, M., Caggiano, L., and Jackson, R.F.W. (2009) J. Org. Chem., 74, 8280–8289. Boger, D.L. and Yohannes, D. (1990) J. Org. Chem., 55, 6000–6017. Evans, D.A. and Ellman, J.A. (1989) J. Am. Chem. Soc., 111, 1063–1072. Schmidt, U., Weller, D., Holder, A., and Lieberknecht, A. (1988) Tetrahedron Lett., 29, 3227–3230. Paerson, A.J., Zhang, P.L., and Lee, K.J. (1996) J. Org. Chem., 61, 6581–6586. P´erez-Gonz¨alez, M. and Jackson, R.F.W. (2000) Chem. Commun., 2423–2424. Chattopadhyay, S.K., Bandyopadhyay, A., and Pal, B.K. (2007) Tetrahedron Lett., 48, 3655–3659. Nishiyama, S., Suzuki, Y., and Yamamura, S. (1988) Tetrahedron Lett., 29, 559–562. Nishiyama, S., Suzuki, Y., and Yamamura, S. (1989) Tetrahedron Lett., 30, 379–382.
103. Reddy, M.V.R., Harper, R.M., and
104.
105.
106.
107. 108.
109. 110.
111.
112.
113.
114.
115. 116.
117.
118.
119.
Faulkner, D.J. (1998) Tetrahedron, 54, 10649–10656. Ito, M., Yamanaka, M., Kutsumura, N., and Nishiyama, S. (2004) Tetrahedron, 60, 5623–5634. Rama Rao, A.V., Chakraborty, T.K., Laxma Reddy, K., and Rao, A.S. (1992) Tetrahedron Lett., 33, 4799–4802. Rama Rao, A.V., Gurjar, M.K., Reddy, A.B., and Khare, V.B. (1994) Tetrahedron Lett., 34, 1657–1660. Boger, D.L. and Yohannes, D. (1991) J. Am. Chem. Soc., 113, 1427–1429. Bigot, A., Bois-Choussy, M., and Zhu, J. (2000) Tetrahedron Lett., 41, 4573–4577. Janetka, J.W. and Rich, D.H. (1997) J. Am. Chem. Soc., 119, 6488–6495. Bigot, A., Tran Huu Dau, M.E., and Zhu, J. (1999) J. Org. Chem., 64, 6283–6296. Inoue, T., Inaba, T., Umezawa, I., Yuasa, M., Itokawa, H., Ogura, K., Komatsu, K., Hara, H., and Hoshino, O. (1995) Chem. Pharm. Bull., 43, 1325–1335. (a) Hitotsuyanagi, Y., Ishikawa, H., Naito, S., and Takeya, K. (2003) Tetrahedron Lett., 44, 5901–5903; (b) Kilitoglu, B., Arndt, H.-D. (2009) Synlett, 720–723. Kazlauskas, R., Lidgard, R.O., Murphy, P.T., Wells, R.J., and Blount, J.F. (1998) Aust. J. Chem., 34, 765–786. Carney, J.R., Scheuer, P.J., and Kelly-Borges, M. (1993) J. Nat. Prod., 56, 153–157. Pordesimo, E.O. and Schmitz, F.J. (1990) J. Org. Chem., 55, 4704–4709. Mack, M.M., Molinski, T.F., Buck, E.D., and Pessah, I.N. (1994) J. Biol. Chem., 269, 23236–23249. Jaspars, M., Rali, T., Laney, M., Schatzman, R.C., Diaz, M.C., Schmitz, F.J., Pordesimo, E.O., and Crews, P. (1994) Tetrahedron, 50, 7367–7374. Nishiyama, S. and Yamamura, S. (1984) Tetrahedron Lett., 23, 1281–1284. Nishiyama, S., Suzuki, T., and Yamamura, S. (1982) Tetrahedron Lett., 23, 3699–3702.
353
354
9 Biomimetic Synthesis of Azole- and Aryl-Peptide Alkaloids 120. Guo, Z.-W., Machiya, K., Salamonczyk,
121.
122.
123.
124.
125. 126.
127.
128.
129.
130.
131. 132. 133. 134. 135.
136.
137.
G.M., and Sih, C.J. (1998) J. Org. Chem., 63, 4269–4276. Kotoku, N., Tsujita, H., Hiramatsu, A., Mori, C., Koizumi, N., and Kobayashi, M. (2005) Tetrahedron, 61, 7211–7218. Couladouros, E.A., Pitsinos, E.N., Moutsos, V.I., and Sarakinos, G. (2005) Chem. Eur. J., 11, 406–421. Chang, C.C., Morton, G.O., James, J.C., Siegel, M.M., Kuck, N.A., Testa, R.T., and Borders, D.B. (1991) J. Antibiot., 44, 674–677. Ezaki, M., Shigematsu, N., Yamashita, M., Komori, T., Umehara, K., and Imanaka, H. (1993) J. Antibiot., 46, 135–140. L´epine, R. and Zhu, J. (2005) Org. Lett., 7, 2981–2984. Schmidt, U., Meyer, R., Leitenberger, V., Lieberknecht, A., and Griesser, H. (1991) J. Chem. Soc., Chem. Commun., 275–277. Waldmann, H., He, Y.-P., Tan, H., Arve, L., and Arndt, H.-D. (2008) Chem. Commun., 5562–5564. H¨oltzel, A., Schmid, D.G., Nicholson, G.J., Stevanovic, S., Schimana, J., Gebhardt, K., Fiedler, H.-P., and Jung, G. (2002) J. Antibiot., 55, 571–577. Paetzel, M., Karla, A., Strynadka, N.C.J., and Dalbey, R.E. (2002) Chem. Rev., 102, 4549–4580. Roberts, T.C., Smith, P.A., Cirz, R.T., and Romesberg, F.E. (2007) J. Am. Chem. Soc., 129, 15830–15838. Williams, D.H. (1996) Nat. Prod. Rep., 13, 469–477. Fischbach, M.A. and Walsh, C.T. (2006) Chem. Rev., 106, 3468–3496. Finking, R. and Marahiel, M.A. (2004) Annu. Rev. Microbiol., 58, 453–488. Widboom, P.F. and Bruner, S.D. (2009) ChemBioChem, 10, 1757–1764. S¨ussmuth, R.D. and Wohlleben, W. (2004) Appl. Microbiol. Biotechnol., 63, 344–350. Hubbard, B.K. and Walsh, C.T. (2003) Angew. Chem., 115, 752–789; Angew. Chem. Int. Ed., 42, 730–765. Zerbe, K., Woithe, K., Li, D.B., Vitali, F., Bigler, L., and Robinson, J.A. (2004) Angew. Chem., 116, 6877–6881; Angew Chem. Int. Ed., 43, 6709–6713.
138. Woithe, K., Geib, N., Zerbe, K., Li,
139.
140.
141.
142.
143.
144.
145. 146.
147.
148.
D.B., Heck, M., Fournier-Rousset, S., Meyer, O., Vitali, F., Matoba, N., Abdou-Hadeed, K., and Robinson, J.A. (2007) J. Am. Chem. Soc., 129, 6887–6895. Bischoff, D., Bister, B., Bertazzo, M., Pfeifer, V., Stegmann, E., Nicholson, G.J., Keller, S., Pelzer, S., Wohlleben, W., and S¨ussmuth, R.D. (2005) ChemBioChem, 6, 267–272. Bischoff, D., Pelzer, S., H¨oltzel, A., Nicholson, G.J., Stockert, S., Wohlleben, W., Jung, G., and S¨ussmuth, R.D. (2001) Angew. Chem., 113, 1736–1739; Angew. Chem. Int. Ed., 40, 1693–1696. Bischoff, D., Pelzer, S., Bister, B., Nicholson, G.J., Stockert, S., Schirle, W., Wohlleben, W., Jung, G., and S¨ussmuth, R.D. (2001) Angew. Chem., 113, 4824–4827; Angew. Chem. Int. Ed., 40, 4688–4691. Zerbe, K., Pylypenko, O., Vitali, F., Zhang, W., Rouset, S., Heck, M., Vrijbloed, J.W., Bischoff, D., Bister, B., S¨ussmuth, R.D., Pelzer, S., Wohlleben, W., Robinson, J.A., and Schlichting, I. (2002) J. Biol. Chem., 277, 47476–47485. Pylypenko, O., Vitali, F., Zerbe, K., Robinson, J.A., and Schlichting, I. (2003) J. Biol. Chem., 278, 46727–46733. Geib, N., Woithe, K., Zerbe, K., Li, D.B., and Robinson, J.A. (2008) Bioorg. Med. Chem. Lett., 18, 3081–3084. Holding, A.N. and Spencer, J.B. (2008) ChemBioChem, 9, 2209–2214. Nicolaou, K.C., Natarajan, S., Li, H., Jain, N.F., Hughes, R., Solomon, M.E., Ramanjulu, J.M., Boddy, C.N.C., and Takayanagi, M. (1998) Angew. Chem., 110, 2872–2878; Angew. Chem. Int. Ed., 37, 2708–2714. Nicolaou, K.C., Jain, N.F., Natarajan, S., Hughes, R.R., Solomon, R., Li, H., Ramanjulu, J.M., Takayanagi, M., Koumbis, A.E., and Bando, T. (1998) Angew. Chem., 110, 2879–2881; Angew. Chem. Int. Ed., 37, 2714–2716. Nicolaou, K.C., Takayanagi, M., Jain, N.F., Natarajan, S., Koumbis, A.E., Bando, T., and Ramanjulu, J.M.,
References (1998) Angew. Chem., 110, 2881–2883; Angew. Chem. Int. Ed., 37, 2717–2719. 149. Nicolaou, K.C., Li, H., Boddy, C.N.C., Ramanjulu, J.M., Yue, T.-Y., Natarajan, S., Chu, X.-J., and Br¨ase, S. (1999) Chem. Eur. J., 5, 2584–2601. 150. Nicolaou, K.C., Boddy, C.N.C., Li, H., Koumbis, A.E., Hughes, R., Natarajan, S., Jain, N.F., Ramanjulu, J.M., Br¨ase, S., and Solomon, M.E. (1999) Chem. Eur. J., 5, 2602–2621. 151. Nicolaou, K.C., Koumbis, A.E., Takayanagi, M., Natarajan, S., Jain, N.F., Bando, T., Li, H., and Hughes, R. (1999) Chem. Eur. J., 5, 2622–2647.
152. Evans, D.A., Wood, M.R., Trotter, B.W.,
Richardson, T.I., Barrow, J.C., and Katz, J.L. (1998) Angew. Chem., 110, 2864–2868; Angew. Chem. Int. Ed., 37, 2700–2704. 153. Evans, D.A., Dinsmore, C.I., Watson, P.S., Wood, M.R., Richardson, T.I., Trotter, B.W., and Katz, J.L. (1998) Angew. Chem., 110, 2868–2872; Angew. Chem. Int. Ed., 37, 2704–2708. 154. Boger, D.L., Miyazaki, S., Kim, S.H., Wu, J.H., Loiseleur, O., and Castle, S.L. (1999) J. Am. Chem. Soc., 121, 3226–3227.
355
357
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids Hans-Dieter Arndt, Lech-Gustav Milroy, and Stefano Rizzo
10.1 Indole-Oxidized Cyclopeptides 10.1.1 Introduction
In the previous chapter, biomimetic aspects of the synthesis of azole- and arylcontaining peptide alkaloids were discussed. Macrocyclic peptide alkaloids in which the ring is formed through covalent attachment to a Trp-based indole are less common in Nature than phenol-oxidized cyclopeptides, but are even more intriguing (Figure 10.1). HO H
O NH
HO O
Me
O
NH
Me
O H NH CONH2 O O Me Me N H O 1a-d: TMC-95A-D N
HO
Me HN H HN N O H Me
O
Me N H O
Me
H Me HH N N
Me
HO
Me 3: himastatin
Figure 10.1
NH2
N H NH
NO
COOH
N
N H
H
2: celogentin C
Me
Me Me
N
HN O
O
Me HO H N MeH HN H H H N HO O N O O O N O OH N O H H NH Me O
O
H
Me
Me Me O
HN
Me Me H N
O
O N O O O O H H NH N H Me OH
N N H H Me
Me
Me Me
HO NH HN O
N
O O
N O
OH O
N H
4: diazonamide A
Selected indole-oxidized peptide alkaloids (1–4).
In the broader sense, electron-rich indoles are highly susceptible to oxidation reactions, much like phenol derivatives, and prone to undergo coupling reactions Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
Cl Cl N H
358
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids
with latent electrophiles. In the context of a densely functionalized peptide chain, this leads to a range of unusual substitution patterns, sometimes macrocycle forming, which is evidenced by the structural diversity of the molecules presented in the following section. 10.1.2 TMC-95A-D
TMC-95A-D (Figure 10.2) are selective proteasome inhibitors isolated from the fermentation broth of Apiospora montagnei [1]. They consist most notably of a highly oxidized Trp-Tyr biaryl linkage with an all l-configured tripeptide-derived monocycle appended by unusual cis-propenyl- and 3-methyl-2-oxopentanyl amide residues. Compounds 1a–d inhibit the chymotrypsin-like, trypsin-like, and caspase-like peptidase activities of the proteasome at nanomolar concentrations [1]. Little is known about the biosynthesis of 1a–d, although knowledge about the biosynthesis of biphenyl cyclopeptides (see Chapter 9) may be indicative. Macrocyclization is thought to proceed via an intramolecular biaryl coupling resulting from oxidative tailoring of side chains. Oxidation of the Trp indole ring at the β, C2, and C3 positions may be a pre-requisite in this case (Scheme 10.1). Some experimental evidence suggests that the oxindole more readily undergoes intra-molecular biaryl coupling when sp3 -hybridized at the C3 position (6 → 1) [2], even more than the corresponding indole (7 → 5) [3]. The oxindole moiety is likely to result from multiple rounds of oxidation of Trp by heme peroxidase or cytochrome P450-type enzymes at the β, C2, and C3 positions (Scheme 10.1). The indolyl–phenyl coupling (Scheme 10.2) might either occur via ring opening of an arene epoxide (Path A, 8 → 9) or via a cytochrome P450-mediated phenoxy radical mechanism (Path B, 8 → 10), analogous to biphenyl-couplings. The 3-methyl-2-oxopentanoic side chain (Scheme 10.3) is predicted to result from an oxidative transamination of isoleucine followed by hydrolysis (11 → 12) [4]. A cis-propenyl amide oxidized Trp HO
all-L-peptide
2 R1R H
6
N HO
Trp-Tyr biaryl bond
O
7
O
TMC-95A-D
O NH
Me
NH
23
O H NH CONH2 O R3 R4 Me N 34 35 H O 2-oxopentanyl amide
Figure 10.2 Key features of TMC-95A-D (1a–d). To improve scheme readability, the following abbreviations are used throughout: ACIE = Angew. Chem. Int. Ed.,
R1 1a: 1b: 1c: 1d:
H H OH OH
R2 OH OH H H
R3 Me H Me H
R4 TMC-95 A H B Me C H D Me
(Apiospora montagnei Sacc. TC 1093)
- proteasome inhibitor Koguchi et al., JAB 2000, 53, 105 Kohno et al., JOC 2000, 65, 990
CEJ = Chem. Eur. J., JAB = J. Antibiot. JACS = J. Am. Chem. Soc., JOC = J. Org. Chem., OL = Org. Lett., PNAS = Proc. Natl. Acad. Sci. U.S.A., TH = Tetrahedron, TL = Tetrahedron Lett.
10.1 Indole-Oxidized Cyclopeptides
359
OH Me O
O HO H
NH
HO
N H O
HO Me
N H
O
NH NH
O CONH2 NH O Me
N H O
OH
H
side - chain tailoring
NH
O HO
Me
NH O
1a-d:TMC-95A-D
Me Me
N H
O
O CONH2
NH2
5
oxidative side chain coupling
oxidative side chain coupling
Me O
HO
H2NOC
O
H H N
N H
N H b OH O HO O
Me
O
H N
HO
Me O
Me
3
side -chain tailoring
O
Me O H N H
H2NOC H N
N
N
H OH
6
Scheme 10.1
enzymatic arene epoxidation
O O
H
O
[o]
NH N HO
H
O
O
Me
Path B
N H
O radical C-C bond formation
HO NH
H
N
−H2O
O
H HO
N H
Me
O
rearomatization
O
NH O N H
H
O
HO
NH
O
Scheme 10.2
N
Me
HO H
H
9
O
tyrosyl radical formation −H·
O CONH2
HO N
HO
O
HO
H
NH O N H 8
HO H
Me
O
nucleophilic ring opening
H
HO NH
Path A
N H
HO O
OH
O HO H HO
N O
H
Simplified retrobiosynthetic analysis of TMC-95A-D.
HO
H
NH2
O
7
HO
H
N H
O
2
O
Me H N
rearomatization −H·
O CONH2
O
H
O N H H O
Me Me
O 1a-d:TMC-95A-D
Two alternative mechanisms for indolyl–phenyl coupling.
N H 10
360
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids
B O
Me
N H
PLP-dependent Me transaminase −NH3 +H2O
NH2
Me H Me epimerization
O N H
O
11
O N H
12
Me Me O 13
Scheme 10.3 Biosynthetic explanation for the origin of the 3-methyl-2-oxopentanoic side-chain [4]. PLP = pyridoxal phosphate
base-mediated epimerization of the labile β-stereogenic center (12 → 13) – in vivo or during isolation – would then offer a convenient explanation for the occurrence of two sets of TMC-95 diastereomers (Figure 10.2, TMC-95 A with C, and B with D). The biosynthesis of the rare terminal enamide (Scheme 10.4) could possibly involve base-catalyzed anti-elimination of C-terminal allo-Thr 14 in a suitably O-activated form (e.g., 14 → 15). This might then deliver the enamide with correct (Z)-geometry on concomitant decarboxylation (15 → 16). While biochemical evidence is limited, chemical reactivity suggests this to be a viable process (see below). Successful total syntheses of natural TMC-95 compounds have been achieved using a combination of new and established chemical techniques – some in the end more biomimetic than others. To form the critical biaryl bond, mostly Pd-mediated coupling has been used, whereas biomimetic oxidative coupling processes have yet to be explored (Figure 10.3). Additionally, considerable synthetic work has
O
O
OH
O-activation (e.g. by kinase)
O
OH
N H
N H
Me
OH
O
Me O
O
H
O
base-induced anti-elimination 3− −CO2, −H2PO4
H
O N H
X
H Me
16
Potential biosynthetic explanation for the peculiar (Z)-enamide side-chain.
Modified Julia coupling Williams et al., PNAS 2004, 101, 11949 Z-selective Mizoroki-Heck Inoue, Hirama et al., ACIE 2003, 42, 2654 Aldol condensation Ma et al., TL 2000, 41, 9089 Danishefsky et al., JACS 2004, 126, 6347
O HO H
7-endo epoxide ring opening Inoue, Hirama et al., ACIE 2003, 42, 2654 Sharpless asymmetric dihydroxylation Ma et al., TL 2000, 41, 9089 Danishefsky et al., JACS 2004, 126, 6347 Williams et al., PNAS 2004, 101, 11949 Suzuki-Miyaura coupling Ma et al., TL 2000, 41, 9089 Williams et al., PNAS 2004, 101, 11949 Danishefsky et al., JACS 2004, 126, 6347 Stille coupling Williams et al., OL 2003, 5, 197
Figure 10.3
Me H
RHN
P OH 15 O
14
Scheme 10.4
O
HO
* O N HO
O
NH
Me
NH
O H NH CONH2 O Me * Me N 35 H O
1a-d: TMC-95A-D
Decarboxylative Grob-type anti-elimination Inoue, Hirama et al., ACIE 2003, 42, 2654 Williams et al., PNAS 2004, 101, 11949 Rearrangement-hydrolysis of a-silylallyl amides Danishefsky et al, JACS 2004, 126, 6347 Macrolactamization Inoue, Hirama et al., ACIE 2003, 42, 2654 Danishefsky et al., JACS 2004, 126, 6347 Williams et al., PNAS 2004, 101, 11949 Late-stage oxidation Inoue, Hirama et al., ACIE 2003, 42, 2654 Separation of Epimers Danishefsky et al., JACS 2004, 126, 6347 Williams et al., PNAS 2004, 101, 11949
Summary of synthetic approaches to TMC-95A-D (1a–d).
10.1 Indole-Oxidized Cyclopeptides
been dedicated to the synthesis of unnatural analogs [5], which in some cases has produced more selective proteasomal inhibitors. 10.1.2.1 Formation of the Trp-Tyr Biaryl Bond by Metal-Catalyzed Cross Coupling Palladium catalysis has been used to mediate the intermolecular coupling between appropriately activated oxindole (Scheme 10.5, 17 and 18) and phenol fragments (19 and 20) to form the biaryl (21) in a relatively mild and selective manner. Although less biomimetic, this highly efficient and flexible approach was well suited for analog studies. The more biomimetic intramolecular cross coupling strategy gave some success during the synthesis of simplified TMC-95 analogs [3, 4]. Interestingly, ring closure was found in this case to be most effective for substrates bearing oxindoles that were sp3 -hybridized at the C3 position (Scheme 10.5, 21). However, yields were typically lower compared with intermolecular cross-couplings.
Me
Me
O
N
R
1
H
+
R 2O
O
OMe NHR3
N H 1
17: R = Boc 18: R1 = Cbz
O
N
i
R1 H
3
N 2 H O
75-90%
O I
Me
H
Bpin
H
Me
R 2O
O
OMe 2
3
19: R = Bn, R = Cbz 20: R2 = MOM, R3 = Boc
21
NHR3
Scheme 10.5 Conventional biaryl coupling. Reagents and conditions: (i) 17 + 19, cat. PdCl2 (dppf ), K2 CO3 , DME, 80 ◦ C [6]; or 18 + 20, cat. PdCl2 (dppf ), K2 CO3 , aq. DME [7]. Bpin = 4,4,5,5-tetramethyl-1,3,2-dioxaborolane-2-yl; DME = 1,2-dimethoxyethane; dppf = 1,1 -bis(diphenylphasphino)ferrocene.
10.1.2.2 Stereocontrolled Oxidation of the Oxindole Fragment Metal catalysts are frequently used to perform selective oxidation reactions in a biomimetic fashion and often rival enzymatic processes in terms of their mildness and stereocontrol. The electron-poor oxindole-based vinylogous amides 22 and 24 were dihydroxylated using catalytic osmium(IV) tetraoxide in two structurally similar cases in excellent yield by inherent substrate control (Scheme 10.6). Inoue, Hirama and coworkers employed substrate-controlled oxidation to stereoselectively introduce the requisite diol, but with a twist to install the correct stereochemistry (Scheme 10.7). After substrate-controlled epoxidation of enamide 26 with dimethyldioxirane (DMDO), treatment of epoxide 27 with a strong Lewis acid initiates an inspired transcarbamoylation via a 7-endo epoxide ring opening by the pendant Boc-carbonyl to form oxazinanone 28. Despite the elegance and efficiency of these oxidations, they may not fully reflect biosynthesis. The Trp β-hydroxyl group is isolated from Nature as a 1 : 1
361
362
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids Me
Me
N
O
R1
O
H
HO HO
H O N R2O
N Me H R1 O OR4 N O H O NH CONH2
OR4
O
R2O
H O NH
Me
CONH2
3
NHR3
NHR
22 (R1 =Boc, R2 = Bn, R3 = Cbz, R4 = tBu)
i, 88% (d.r. = 5:1) ii, 87%
24 (R1 =Cbz, R2 = MOM, R3 = Boc, R4 = Bn)
23 R1 = Boc, R2 = Bn, R3 = Cbz, R4 = tBu) 25 (R1 = Cbz, R2 = MOM, R3 = Boc, R4 = Bn)
(d.r. = 20:1)
Scheme 10.6 Reagents and conditions: (i) OsO4 , NMO, (DHQD)2 PHAL, t-BuOH–H2 O, room temp. (rt) [6]; (ii) OsO4 , py, 0 ◦ C, then sat. NaHSO3 [7].
Me
O
O
Me N Boc
i
O Br
26
Br
O
N O
dr > 95%
N Boc
Me Me
F3B
N Boc
O O
27
O
ii
H HO
N O O
87% 2 steps
O Br
single diastereomer
Me Me
N Boc 28
Scheme 10.7 Reagents and conditions: (i) DMDO, CH2 Cl2 , rt; (ii) BF3 ·OEt2 , CH2 Cl2 , −78 → 0 ◦ C [8].
epimeric mixture, whereas the C3 hydroxyl group exists as a single epimer. The two alcohol functional groups might therefore be introduced in two separate enzymatic oxidation steps. 10.1.2.3 Late-Stage Stereoselective (Z)-Enamide Formation Two contrasting synthetic methods have been described for the installation of the (Z)-enamide: one distinctly biomimetic, the other arguably more synthetic (Scheme 10.8). The first method (Route A) [8] started from l-allo-threonine and operates under mild reaction conditions (0 ◦ C). After hydrogenolysis of benzyl ester 29, a Grob-type anti-elimination (29 → 30) of carbon dioxide was induced under Mitsunobu conditions in a good yield over the two steps. A more engineered approach (Route B) [6] – a stepwise dyotropic rearrangement involving sequential 1,4-silyl and 1,4-hydrogen shifts – necessitated a higher temperature to drive the reaction. Nevertheless, these forcing conditions were well tolerated; the intermediate silyl-enimidate (32 → 33) readily hydrolyzed to form the target (Z)-enamide (34).
10.1 Indole-Oxidized Cyclopeptides ROUTE A O
R1 =
O
ROUTE B
OBn H OH
N H
HO
H
HO O
Me
O
MOMO
H
O H
N H
PPh3
Me −OPPh3, CO2
59% 2 steps
O N H
Me
Me
30 anti-elimination
ii
Me O
TESO H 2 R TESO NH O O N H CONH2 NH R3O O O Me Me N H 3 O R = H or TES
SiEt3 N H 31
O CONH2
N H TESO O
O
R2 =
NH
NH O
i O
R1
N
29
O
363
SiEt3
OSiEt3
∆
N H 32
N Me 33 49% 2 steps
hydrolysis
O N H
Me 34 thermally-driven rearrangement
Scheme 10.8 Reagents and conditions: (i) (1) H2 , Pd(OH)2 /C, THF–H2 O (1 : 2) and (2) DEAD, PPh3 , mol. sieves, 0 ◦ C [8]; (ii) (1) xylene, 140 ◦ C and (2) HF–py, THF–py, then Me3 SiOMe [6].
10.1.3 Celogentin C
The bicyclic peptides celogentins A–H, J, and K [Figure 10.4 shows celogentin C (2) and K (35)] were isolated together with the analogous moroidin [9] from the seeds of Celosia argentea [10]. These structurally unique secondary metabolites are inhibitors of tubulin polymerization, with 2 being the most active congener. all L-configured peptide
Me Me O
Me
HN pyroglutamate
O
H N
N H O
HN
H
Me H H N O A
N O
Me
N H
H
Me
Me Me O
NH2 HN
NH
N H O
O H H N O
N H N-linked imidazolyl-indole
O
H N
HN O H Me
N Me
N H
H
2: celogentin C -inhibition of tubulin polymerization, cytotoxic-
Kobayashi et al., JOC 2001, 66, 6626 (celogentin A-C) Kobayashi et al.,TH 2003, 59, 5307 (celogentin D-H & J) Kobayashi et al., TH 2004, 60, 2489 (celogentin K & moroidin)
Isolation, bioactivity, and structures of celogentin C (2) and K (35).
N
O HN
O
H
COOH N
35: celogentin K (seeds of tropical plant Celosia argentea)
NH2
H N
O
H
unusual Leu-Trp linkage
Figure 10.4
H N
Me
OH COOH
N
O H Me
N
HN O N
B
O
N
364
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids
Common to all isolated celogentins is the 17-membered A-ring (Figure 10.4), whereas the B-ring is found in different sizes and with a varying peptide sequence. The most distinguishing features of the celogentins though are the highly unusual sp2 –sp3 Leu-Trp and sp2 –sp2 His-Trp linkages. Whereas the peptide origin of the celogentins is rather obvious, (Scheme 10.9) the biosynthetic logic behind the highly unusual side chain modifications is not. In this case, structural differences evident between naturally occurring variants may provide some useful insights. For example, epoxidation across the C2−C3-bond would generate iminium ion 37 and as such activate the indole nucleus for nucleophilic attack (Scheme 10.10). An intramolecular attack by His (37 → 38) may then be facilitated by the presence of a turn-inducing Pro in the peptide chain, for example, to deliver the B-ring of celogentin C (2) after desiccation. In the absence of such conformational preorganization, hydrolysis, followed by further enzymatic oxidation (37 → 35) could in principle become more prominent. Indeed, celogentin K (35) was isolated together with the corresponding ring-closed product moroidin [9]. The biosynthesis of the unprecedented Leu-Trp linkage is currently unclear, but we can propose a likely pathway. In peptide alkaloids, oxidases are known to frequently introduce hydroxyl functionality at the β-position of amino acids
H2N Me Me O
Me
N H O
HN H N
O
Me H N O A
HN
H N
HN
H
O
HN O B O N
NH COOH
N
O H Me
N
N H
Me
2: celogentin C oxidative side -chain coupling
H O
O Me
Me O
H N
H2N
H
OH
H H Me
N H
Me H O H N O Me
H N
N
NH N H
Me 36
Scheme 10.9
NH2
N O
O
H N H
H N
O OH
O
H N
N
Simplified retrobiosynthetic proposal for celogentin.
10.1 Indole-Oxidized Cyclopeptides H N Me Me O
Me
HN
N H O
O H Me
Me Me O
Me [O]
Me
Me H H O N
N H O
HN
3 N2 N H
O HN
OH N
H
HN
H N
O
O H Me
N H
Me
−H2O
N
Me Me O N H O
HN
N N H H
O
N
N HN O O B N
HN
H N
O
H
O A
N
N H
Me H H N
H NH2
N H NH COOH
N
O H Me
O CO2H
O
35
Me
N H
Me
38
Scheme 10.10
H N
N H
O A
37
HO
NH2 NH
O HO H2O:
HN
H N
O
Me H H N
2
Biosynthetic proposal for the formation of the B-ring Trp-His linkage.
(Scheme 10.11, 39 → 40) [11]. Enzymatic activation of the alcohol (40 → 41) might then induce N-selective attack by the vicinal α-amide group (41 → 42) and lead to an N-acyl aziridine intermediate [12]. This could suffer a nucleophic attack by the Trp indole ring to close the A-ring (42 → 43 → 44), potentially guided by a Lewis-acidic enzyme. β-Selective ring-openings of aziridines with arenes are known [13], even in an intramolecular fashion [14]. HN
cytochrome P450 oxygenase
O
N H Me
HN
O
N H Me
Me
O-activation
HN
OH
O O P O OH Me 41 O
N H Me
Me
40
39
−H2PO4
SN2
HN N H Me
O
A rearomat.
N H Me 44
Scheme 10.11
−H+
HN N H Me
O H
A SN2
N H Me 43
HN
O
α
β
N H
N Me
Me
42
Proposal for the biosynthetic origin of the A-ring Trp-Leu linkage.
Biosynthetic reasoning aside, celogentin C (2) with its most promising bioactivity has received much synthetic attention, culminating in several elegant total syntheses. In Nature, the closure of the A and B rings is thought to result from
365
366
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids
Knoevenagel condensation Castle et al., JACS, 2010, 132, 1159
Me
HN
Radical conjugate addition Castle et al., JACS, 2010, 132, 1159
H N
O
Me Me Me O N H O
H H H N
O
A
O H Me
O
N O
HN
HN
B
HN
NCS-mediated oxidative coupling Jia et al., OL, 2010, 12, 956 Chen et al., ACIE, 2010, 49, 958 Castle et al., JACS, 2010, 132, 1159
O N
NH
N H
NH2
COOH
N
Sequential asymmetric Michael addition/bromination/azidation Jia et al., OL, 2010, 12, 956 C-H activation–indolylation Chen et al., ACIE, 2010, 49, 958
N H
Me
2: celogentin C
Figure 10.5
Synthetic approaches to celogentin C.
intramolecular oxidative tailoring of the peptide side chains. Naturally, the unusual Leu-Trp and His-Trp coupling arrangements provided a stiff synthetic challenge, and were preferentially formed in an inter-molecular fashion prior to ring closure by conventional macrolactamization (Figure 10.5). 10.1.3.1 Intramolecular Knoevenagel Condensation/Radical Conjugate Addition The difficult Leu-Trp linkage was first addressed by Castle and coworkers in their total synthesis of celogentin C (2) [15]. A Knoevenagel condensation between titanated α-nitro amide tautomer 45 and Trp aldehyde 46 formed the acyclic chain and gave nitro olefin 47 in good yield and as a single isomer (Scheme 10.12). At the critical bond-forming step, substrate-controlled 1,4-addition of an isopropyl radical was executed, but gave only slight preference for the desired diastereomer. Subsequent SmI2 -mediated NO2 -group reduction then gave amine 48 in excellent yield [15]. A more stereoselective synthesis inspired by Evans’s auxiliary chemistry was used by Jia to form the key Leu-Trp arrangement as a single diastereomer (Scheme 10.13) CO2tBu CbzHN Me Cl4Ti
TES
O
N H
68% (single isomer)
+
BnO2C
Me O N H Me Me 45
H N
N O
O Ti Cl Cl Cl
O
Me
NO2
H N
N BnO2C H Me i
46
Me
Me O
H
N H
t
BuO2C
TES NHCbz
47
Scheme 10.12 Reagents and conditions: (i) TiCl4 , NMM, THF–Et2 O (2 : 1); (ii) Et3 B/O2 , Zn(OTf )2 (2.0 equiv.), iPrI, Bu3 SnH; and (iii) SmI2 , THF–MeOH [15]. NMM = N-methyl morpholine.
90% (1:3:2:1 d.r. in favor of 4)
NH2 Me
H N
N BnO2C H Me ii, iii
O Me
Me O
Me O
Me
N H
t
BuO2C
48
TES NHCbz
10.1 Indole-Oxidized Cyclopeptides O
O
CO2Me N
O
(Boc)2N
HH
i
Ph Me
Me
N Boc
RCu M
49
O
N H
H +
i
50
51 65%, 2 steps
CO2Me (Boc)2N O
Ph
N Boc Me
CO2Me
(Boc)2N
N
ii, iii 78%, 2 steps
Me
O
O
N3
N Boc
Pr H
"Br "
HO
CO2Me
(Boc)2N O M O
O N Boc
Br Me
Me
53
52
Scheme 10.13 Reagents and conditions: (i) (1) 50 from CuBr-Me2 S (in situ), THF, −30 ◦ C, then 49; (2) NBS; (ii) (1) NaN3 ; (iii) LiOH, H2 O2 [16].
[16]. Here, the chiral oxazolidinone controls the initial Michael-type 1,4-cuprate addition (49 + 50 → 51) and the subsequent enolate capture by N-bromosuccinimide (NBS) (51 → 52). The α-amino group is introduced in a latent form as an azide – via SN 2 reaction of the intermediate bromide prior to auxiliary cleavage (52 → 53) – to be unveiled at a later stage for the coupling chemistry. 10.1.3.2 C−H Activation–Indolylation Using methodology developed first by Daugulis [18] and then Corey [19], Chen demonstrated that N-phthaloyl protected 8-aminoquinoline amide 54 couples stereoselectively to iodotryptophane 55 under Pd(II) catalysis (Scheme 10.14).
CO2tBu
N O
HN
O
Me
N
i
+ N
O
BocHN
Me O
54 (2.0 eq.)
O
N Pd N H OAc
O Me
N Ts 55 (1.0 eq.)
I
H
Me 56 85%
CO2tBu
CO2tBu BocHN HO
HN
O
2 N3 Me
BocHN
N
Me
N Ts
O N Ts
PhthN
58
Scheme 10.14 Reagents and conditions: (i) Pd(OAc)2 (cat.), AgOAc, t-BuOH, 110 ◦ C, 36 h [17].
Me
Me 57
367
368
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids
The quinoline serves as a chelating auxiliary for palladium coordination, and promotes the formation of trans-palladacycle intermediate 56. This palladium(II) intermediate then undergoes a cross-coupling with 55 to deliver the desired product 57 as a single diastereomer. The phthaloyl group is critically involved in the arylation step, providing bis-protection of the α-amino group and a steric bias. Just like in the previous example (Scheme 10.13), an azide 58 is used as masked amine functionality in the later stages of the synthesis.
10.1.3.3 NCS-Mediated Oxidative Coupling A serendipitous discovery by Castle and coworkers was crucial for the intermolecular side chain coupling between Trp and His [15]. They initially found that treating 59 with N-chlorosuccinimide (NCS) afforded an undesirable dichlorinated product of unknown structure, which was unreactive toward 60 (Scheme 10.15). When, however, Pro-OBn was preincubated with 59 and NCS, monochlorination to a presumed α-chloro-imminium ion was achieved, which led to the desired adduct 61 after hydrogenolysis. It is noteworthy that Castle’s unique conditions were also used in subsequent total synthesis efforts [16, 17].
Me
Me Me O HN
O
H N
Me H H N N
N H O
N HN
NHPbf
O
O NHCbz
HN O H Me
H
H CO2Bn
+ Me
H N
N H
N
HN
O
CO2tBu
59
60 64% (2 steps)
i, ii
H
H Me
Me Me O HN
N H O
Me H H N N
CO2H HN
N NHPbf
O
O NH2
O
H N
HN O H Me
Me
N H 61
Scheme 10.15 Reagents and conditions: (i) Pro-OBn (2.0 equiv), NCS (3.0 equiv.), 1,4-dimethyl piperazine, CH2 Cl2 then 60 (5.0 equiv.); (ii) 10% Pd/C, HCO2 NH4 , MeOH–H2 O (5 : 1) [15]. Pbf = 2,2,4,6,7-pentamethyl-2,3-dihydrobenzofuran-5-sulfonyl.
N
N
HN
O
CO2tBu
10.1 Indole-Oxidized Cyclopeptides
10.1.4 Himastatin and Chloptosin
The antibiotic and antitumor bacterial metabolite himastatin (Figure 10.6, 3) is a rather exceptional example of a peptide alkaloid given that it is a fully symmetric dimer coupled via a pyrroloindole [20]. The more recently isolated compounds chloptosin (62) [21] and kutzneride 2 (63) – one of nine structurally related kutznerides [22] – are structurally similar to 3 in many respects, hinting at a common biosynthetic pathway. What distinguishes these metabolites from other indole-oxidized peptide alkaloids is that the enzymatic oxidative processes apparently do not mediate formation of the macrocycles, which appear to be typical NRPS (non-ribosomal peptide synthesis) cyclo(depsi)peptide products. However, oxidative tailoring of Trp and Orn side chains forms five- and six-membered rings at the periphery of the macrocycle, and seems to cause dimerization of the pyrroloindoles. Taking chloptosin (62) as an example, the dimeric structure can be severed across the biaryl bond (Scheme 10.16, 62 → 64) and at the peripheral pyrroloindole D/L-depsipeptidemacrocycle Me Me Me HO Me biaryl H OH O Me N Me linkage HN HN H H H N H H N O HO O N O O N O O O O N O O N O OH OH N O H H N H H H NH NH N bispiperazic acid Me HO Me O H pyrroloindole Me OH Me Me Me 3: himastatin Me (Actinomycete ATCC 53653 strain) Me
HO
-antibiotic, anticancerLam et al., J. Antibiot. 1990, 43, 956 Leet et al., J. Antibiot. 1990, 43, 961 Leet et al., J. Antibiot. 1996, 49, 299
D/L-peptide-macrocycle OH H H N HN Me N Cl H H O N O O O O N OH N O H H 2 NH N bis-pyrroloindole H NH Me Me piperazic acids
D/L-depsipeptide-macrocycle
MeO
62: chloptosin (Streptomyces MK498-98F14 strain) -antibiotic, anticancerUmezawa et al., JOC 2000, 65 ,459
Me
monomeric pyrroloindole H
HN
HO N Cl Cl b-hydroxyglutamic acid
N HO H 3 O
H
tBu piperazic acid H N N O O H O O O Cl H NH N H O-methyl serine OMe O
H
O
OH 63: kutzneride 2 (Actinomycete Kutzneria sp. 744) -antimicrobial-
Broberg et al., J. Nat. Prod. 2006, 69, 97 Pohanka et al., J. Nat. Prod. 2006, 69, 1776
Figure 10.6 Structural features and biological activities of pyrroloindole-based peptide alkaloids.
369
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids
370
MeO
HO H N Me H HN H N H O N O O H O O N OH N O H H NH N H NH Me Me 62
Me HN H N
Me
kutzneride-like Cl
oxidative dimerization
X 2
H
HO
HN
O O
N N H H Me
Cl
H N H
OH
alternate D/L-peptide
Me
H2 N
H N
H N
N H
O Me
OH H H O N
O N H
O
O HO
Me
H
OMe
Me H H N
HN
Cl aromatic halogenation, O-methylation
Me
macrolactamization
H2N
O
O O
HN
HN
OH
Me
NH
H N
ring formation
X
NH2 O
O
N
O O
64 side chain oxidation
HN
O
O
O O
NH NH2 NH
HN OH H
66
NH2
H
OMe
65
Scheme 10.16
Retrobiosynthetic simplification of chloptosin (62).
and piperazic acid rings (64 → 65), which quickly exposes a modified alternating d/l cyclic peptide (66). The identification of and subsequent studies into the gene cluster of the kutznerides [23] has helped to elucidate the order of steps and the function of some constituent enzymes. While the precise order of biosynthetic events remains unclear, oxidative dimerization and pyrroloindole formation are presumed to occur after assembly of the cyclic (depsi)peptide monomer (Scheme 10.17a). Little is also known MeO
H
N H
H H OH H N Me N
HN O O
N
O O
O O
N H NH Me
(a)
Me Cl H HN
NH
HO
OH
HN
NH H
N Me H
Cl Me
OH
H N
NH2 flavin-dependent monooxygenase
OH
H2N (b)
O
H
OH O
O O H N H
O O
O O
N
NH
H N
H
OMe
B
O H -H2O
H2N
Me HN H N
O H HN HN
OH
Scheme 10.17 Simplified proposal for dimerization (a) and piperazic acid formation (b) during the biosynthesis of dimeric pyrroloindole peptide alkaloids [23].
10.1 Indole-Oxidized Cyclopeptides
371
about the biosynthesis of the unusual (though not uncommon) piperazic acids (Scheme 10.17b). Feeding experiments [24, 25] suggest that their synthesis occurs prior to incorporation into the peptide chain. Similar to the hydrazo linkage in valanimycin [26], the biosynthesis of piperazate might be initiated with N-hydroxylation of the α-amino group of ornithine by a flavin-dependent monooxygenase (KtzI in the case of the kutznerides), followed by nucleophilic attack of the α-amino group to close the ring. The O-methylation of serine, by contrast, most likely occurs during assembly. Characterization of the flavin-dependent halogenase enzymes KtzQ, KtzR, and the NADH-utilizing FAD reductase KtzS has indicated that regioselective double halogenation in kutzneride biosynthesis occurs on the free l-Trp (Scheme 10.18, 67 → 68 → 69) through the sequential action of the KtzQR enzyme pair. It is likely, therefore, that dichlorinated Trp is the monomer incorporated into the growing kutzneride assembly line by the KtzH adenylation domain (69 → 70) [27]. The final conversion into the pyrroloindole moiety in the mature kutzneride is postulated to occur via epoxidation of the indole 2,3-double bond by the heme protein KtzM, followed by intramolecular capture of the epoxide by the amide nitrogen (71 → 63) [23]. The biosynthetic origin of the other non-proteinogenic non-ribosomal amino acids found in the kutzneride family is the subject of ongoing studies. To date, three successful bi-directional syntheses have been reported in the literature, one for 3 and two for 62 (Scheme 10.19). Bi-directional synthesis was pursued earlier for polyketides, which are inherently more symmetric [28]. In light of the brief discussion on the biosynthesis however, this innovative strategy eventually R O H
O H
OH KtzQ
NH2
OH NH2
(KtzS)
X
N H 67
Cl
KtzH
Cl
N H
(adenylation domain)
N H
H HN HO O
Cl O H O O NH Me O N H H NH O N O NH2 H H t Bu O MeO
OH
68: X = H
KtzR (KtzS)
S
Cl
70
69: X = Cl
Me H HN H HO N H H
Cl Cl
N OH
O
O O N H H
Me
O O O
O O
H
t Bu H N N
NH
H
H HN H
Cl KtzM
Cl
(epoxidation/syn-cis heteroannulation)
OMe OH
Cl
HN
HN H HO
O O N H
O
Scheme 10.18 Proposed biosynthesis of 6,7-dichlorotryptophan and its incorporation into the kutzneride 2 (63) assembly according to Walsh et al. [23, 27]; R = NRPS-thiolation domain.
O O
O O
H
t Bu H N N
NH
H
OMe OH
63: kutzneride 2
O
71
Cl
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids
372
Macrolactamization
Oxidative cyclization of L-tryptophan
Esterification Me
Me Me HN
H
HO N H Me
H N H
O
O O
N
Me H H N
Me
O O
N
H N H
N H H Me
H
NH
O
HO OH
O
O
HN
H
N
O O
O
Me
HO
O
N
O O
NH
Me
Me
H
OH
Me
OH
Me
Me H N
3: himastatin
Stereoselective bromolactonization
Me Evans's chiral auxiliary
Transition metal-mediated coupling - Danishefsky et al., CEJ 2001, 7, 41 -
BI-DIRECTIONAL TOTAL SYNTHESIS
Zinin rearrangement Yao et al., OL 2010, 12, 1124 Transition metal-mediated coupling Ley et al., ACIE 2010, 49, 6139 Palladium-catalyzed indole formation Yao et al., OL 2010, 12, 1124
Macrolactamization Yao et al., OL 2010, 12, 1124 Ley et al., ACIE 2010, 49, 6139
MeO
H
N H
HN
N
O O
O
H N H O O
Me
HO Me H H N
NH H Me
HN
N
OH
Selenocyclization Yao et al., OL 2010, 12, 1124 Ley etal., ACIE 2010, 49, 6139
Scheme 10.19
H HO
N
O
N H NH Me
Cl
Cl 62: chloptosin
N H H Me
Me HN H N O O O
O H N H OH MeO
Biotransformation Ley et al., ACIE 2010, 49, 6139
O O
N
NH
H N
H
Organocatalytic tandem transformation Ley et al., ACIE 2010, 49, 6139
Summary of the synthetic approaches to himastatin (3) and chloptosin (62).
adopted by three groups in order to complete their respective syntheses cannot be considered truly biomimetic. For peptide alkaloid biosynthesis, it is difficult to see how Nature might use bi-directional synthesis to her own advantage. Even so, the success stories described below indicate what is currently feasible in the laboratory. 10.1.4.1 Synthesis of the Himastatin Pyrroloindole Core Danishefsky and coworkers were successful in installing the biaryl group after pyrroloindole cyclization [29]. Starting from protected L-tryptophan 72 (Scheme 10.20), the authors prepared a fully-protected pyrroloindole core (74)
10.1 Indole-Oxidized Cyclopeptides CO2tBu
CO2tBu
HO i
NHTr
N
55%, 2 steps
N H H
N H 72
Cbz tBuO2C
N
H
N H
74 ii
CO2tBu
TBSO
N
N
OTBS
N H Cbz
73% CO2tBu
TBSO iv
Cbz
Cbz
Cbz
73 Cbz H N
CO2tBu
TBSO
steps
X
83%
N N H Cbz
77
Cbz
75 (X = I) iii 87%
76 (X = SnMe3)
Scheme 10.20 Himastatin synthesis [29]. Reagents and conditions: (i) (1) DMDO, CH2 Cl2 , −78 ◦ C, then (2) HOAc, MeOH, CH2 Cl2 ; (ii) ICl, 2,6-di-tert-butylpyridine, CH2 Cl2 ; (iii) Me6 Sn2 , Pd(PPh3 )4 , THF, 60 ◦ C; and (iv) Pd2 dba3 , 76, Ph3 As, DMF.
via a DMDO-mediated syn-cis oxidative annulation (72 → 73). The trityl- and tert-butyl protecting groups were critical for achieving the syn-cis selectivity. A regioselective aromatic iodination was then followed by stannylation and Pd-mediated dimerization under Stille coupling conditions (74 → 77), to create the biaryl bond in excellent yield and clear a path for the bi-directional synthesis. 10.1.4.2 Synthesis of the Chloptosin Pyrroloindole Core The first synthesis of chloptosin (62) was performed by Yao and built on Danishefsky’s seminal himastatin work. However, the added complexity of two 6-chloro-substituents on the biaryl pyrroloindole core had to be mastered [30]. This was achieved through formation of the biaryl bond via a Zinin rearrangement – predetermining the bi-directional synthesis – followed by a regioselective aromatic iodination to set up metal-catalyzed indole formation (Scheme 10.21, 78 → 79). After condensation of 79 with two equivalents of l-pyroglutamic acid derivative 80, treatment of the intermediate bis-imine with palladium(II) diacetate effected a Heck-type heteroannulation (80 → 82) which left the two aryl chlorides untouched. Finally, a highly stereoselective selenocyclization mediated the formation of the pyrroloindole into the correct syn-cis configuration (83), followed by an obligatory oxidative work-up to give 85. Ley’s very recent synthesis of the dimeric pyrroloindole core en route to 2 is even more direct [31]. In this case, enzyme catalysis was used to prepare 6-chloro-l-Trp as a key building block. 10.1.4.3 Macrolactamization All three highlighted total syntheses successfully close the macrocycle using bi-directional macrolactamization, only at different positions along the 18-membered ring (Scheme 10.19). Yao elaborated the pyrroloindole dimer (Scheme 10.21, 85) to the macrocyclization precursor 86 (Scheme 10.22) using
373
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids
374
Cl
Cl
I
steps
NH2 + HO
H2N NO2
I
CO2tBu N Boc
Cl
78
79
H
Cl
Cl
H Boc N
2
tBuO2C
80
H HO
N
N
OH
tBuO2C
42%
E
81 (R = H)
Boc H N
Boc
R N
i
N H Boc
Cl
ii 85%
Boc N
CO2tBu iv
Boc
tBuO2C
85
Cl iii
H Boc N
75%
82 (R = Boc)
SePh
2
89%
syn-cis annulation
83
Scheme 10.21 Chloptosin core synthesis [30]. Reagents and conditions: (i) Pd(OAc)2 (10 mol. %), DABCO (6.0 equiv), DMF (0.2 M), 120 ◦ C, 48 h; (ii) Boc2 O, DMAP; (iii) N-(phenylseleno)phthalimide (84), PPTS, Na2 SO4 , CH2 Cl2 , rt; and (iv) m-CPBA, aq. NaHCO3 , i-PrOH, rt. HO
O
MeO H
FmocHN NH O O
N N Cbz H
O O
H
NH H
NCbz Me
Cl
H
2 45%, 2 steps
N
O O
N
Cbz H
Me 86
H H N
HN
i
OH
O
N
MeO
OTBS H Me N H N
O
O O
H Me N H N
NCbz Me
Cl
NH H
H HO
OH
O
N
Me
OTBS
Cl
Me 87
Me CbzN N
HN O O
O
N O H NMe N H H H TBSO
O O
H Cbz N
NH
N
H
OMe
Scheme 10.22 Chloptosin synthesis [30]. Reagents and conditions: (i) (1) iPr2 NH, CH3 CN, rt and (2) HOAt, PyBOP, iPr2 NEt, DMF, CH2 Cl2 , −5 ◦ C to rt.
a sequence of advanced solution-phase peptide couplings, which was then cyclized to afford 87 in an excellent yield. This complex example exceptionally showcases bi-directional macrocyclization for total synthesis (>2500 Da intermediate heavily-laden with heteroatom-rich coordinating protecting groups) [30]. Danishefsky was the first to demonstrate this in his total synthesis of 3 [29]. The Ley group has since accomplished the same feat in their total synthesis of 62 [31]. These synthetic efforts define the limits of biomimetic approaches in peptide alkaloid synthesis, and highlight the methodological needs for more selective coupling between large peptide subunits and/or individual cyclopeptide monomers. Apparently, attempts by Danishefsky and coworkers to couple two cyclopeptides to form the biaryl were unsuccessful [26]. However, in their synthesis of chloptosin (62), Ley and coworkers reported the successful dimerization of a fully formed macrocycle monomer using Stille- and Suzuki-cross-couplings [31]. Unfortunately, deleterious over-oxidation could not be suppressed. Nevertheless, this insightful chemical ‘‘hint’’ suggests that dimerization of the fully-formed monomer by oxidative coupling might indeed be biosynthetically feasible within a favorable enzymatic environment.
10.1 Indole-Oxidized Cyclopeptides
10.1.5 Diazonamide
Diazonamides A and B (Figure 10.7, 4 and 89) are secondary metabolites isolated from the colonial ascidian Diazona angulata [32, 33]. Diazonamide A (4) features a unique, densely functionalized quaternary carbon (C10) located at the center of a bicyclic ring system, which is surrounded by bis-indole, 2,4-bisoxazole, and α-hydroxyvaline structural motifs. The minor isolate, diazonamide A (4) was initially assigned the structure 88 by analogy to a misassigned X-ray crystal structure of the p-bromobenzamide derivative of the major isolate diazonamide B (not shown), [32]. Recently, three additional diazonamide family members (C–E, Figure 10.7, 90–92, respectively) were isolated alongside 4 and 89 [33]. Biological data for these new analogs has shown the importance of the side chain valine residue and the need for an α-hydroxy in place of an α-amino group for compound potency [33]. Compound 4 is endowed with high in vitro cytotoxicity toward human tumor cell lines, where it blocks cell division during mitosis through a unique interaction with the mitochondrial enzyme ornithine δ-amino transferase [34]. The synthesis of the originally proposed structure (Figure 10.7, 88) by Harran and coworkers [35] prompted a structural reassignment of both diazonamide A from 88 to 4 and diazonamide B to 89 based on mismatching analytical and biological data between the natural product and the synthetic material. Despite this turn of events, several elegant approaches to the revised structure (4) of this structurally impressive and biologically significant secondary metabolite were quickly produced, notwithstanding the diverse and noteworthy routes to the originally-proposed diazonamide A structure (88) [36].
R2 1
R HN
Me
30 10
G
R
R3 N
N A O O
4
Me
Me
H N
O
11
N H
16
O
H2N
Cl
B E
F H
29
24
Cl
Me H N
Me
H
O
26
N N
C NH
18
N
O O HO
Cl Cl
O
NH
O
D
OH
4: diazonamide A (R1 = X, R2 = Me, R3 = H, R4 = H) R2 = Me, R3 = H, R4 = Br) 89: diazonamide B (R1 = H, 90: diazonamide C (R1 = L-Val, R2 = Me, R3 = H, R4 = H) 91: diazonamide D (R1 = H, R2 = Et, R3 = Cl, R4 = H) 92: diazonamide E (R1 = H, R2 = Et, R3 = Cl, R4 = H) (from colonial ascidian Diazona angulata) - anti-mitotic, anticancer -
88
Me
X=
Me OH
O
Fenical & Clardy et al., JACS 1991, 113, 2303 Reyes et al., TL 2008, 49, 2283
Figure 10.7 (89–92).
Diazonamide A (4), the initially proposed structure (88), and natural analogs
375
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids
376
Me Me HO
Me H N
Me
H
H
N N O O
O
N
Cl
O
Me H N
H 2N
N H
N H
O
Me Cl
Me O
HO
Me
HO
N O O
O
Me
Me
H N
N H
O
N
COOH
oxidative coupling of side -chains
Me H N
H2N
O
Me O
N
N
N H
O
N H
N H
HO
N
HO O
H Me oxidative coupling
N NH OO HO
94
of side -chains H H COOH H N
O HO N H
O
N H
dehydrative cyclization oxidation
Me H
H
95 dehydrative cyclization oxidation
Me H
COOH
O N
93
Me
O
Me
O O
OH
97
halogenation decarboxylation
Me H N
O
H N
H N
4
Me
Me
N H
O
N
Me HO
Me H N O
N H
Me O
N
Me H H N
N H
O
O
OH
N H H O OH
OH
H N HO
96
Scheme 10.23 Simplified retrobiosynthetic analysis of diazonamide A (4) reveals its pentapeptide origin.
At first sight, the peptide origin of 4 appears hidden (Scheme 10.23). In a retrobiosynthetic sense, pruning the aromatic chlorides is a logical first step (4 → 93). Regioselective chlorination may be achieved on the mature polycyclic core in the later stages of a synthesis. At this juncture, disconnection across the central C10-quaternary amino acetal–biaryl bonds (93 → 95 or 94 → 96) and hydrolysis of the 2,4-bisoxazole ring system (93 → 94 or 95 → 96) reverts the complex core back to a linear chain (96). Removal of the apparent oxidative tailoring eventually reveals the anticipated all-l-Val-Tyr-Val-Trp-Trp peptide 97. How all the reactions are really executed in Nature remains unclear. Synthesis could at least make a contribution in this regard, to execute potentially feasible biomimetic strategies, and help to substantiate hypotheses. The peculiar central aminal moiety is certainly the most striking structural element of diazonamide A. Its biosynthesis is anything but clear, although it could
10.1 Indole-Oxidized Cyclopeptides
result from SN 1-type interception of an advanced, highly-oxidized indolyl-oxazole by the tyrosine phenol group (Scheme 10.24, 98 → 100) followed by desiccation. For this, the ‘‘natural’’ reactivity of indole must be reversed, probably by oxidizing past the indolinone stage. On the other hand, the Tyr residue might, after oxidation, serve as an electrophile for nucleophilic attack by the indole C3 (98 → 99). Here, steric hindrance at C3 and the susceptibility of Trp to undergo oxidation might complicate matters. Synthetic approaches based on both of these biomimetic postulates have been explored [37, 38]. Me H H
Me H N
N
Me
H N
Me
Me
N
H
N O
Tyr oxidation
N O O
Trp oxidation
3
C-3 nucleophilic attack
OH
N H
O
O
N
Me
H N
N O O
C-3 electrophilic attack
HN
O O
N H
H 99
98
Scheme 10.24
100
Biosynthetic proposal for the formation of the aminal core.
The constrained bis-indole biaryl might be formed via a biradical coupling mechanism (Scheme 10.25, 101 → 102) that is thought to take place once the aminal has been established, or at least when the central Trp is in a higher oxidized state (e.g., as an oxindole like in TMC-95A-D-vide supra) [37, 38]. The α-hydroxyvaline is believed to result from oxidative deamination of valine (Scheme 10.26, 103 → 104), followed by reduction of the subsequent α-oxo functionality by a reductase (104 → 105) [39].
N
N
N
O
O
O
O
NH
N H
Me
Me
O
O 103
Scheme 10.26
102
Simplified biosynthetic proposal for biaryl coupling.
amino acid oxidase
OH
H2N
NH
N H
101
Scheme 10.25
N
O
biradical coupling
Me
Me
Me
Me
reductase
OH
O O 104
OH
HO O 105
Biosynthetic proposal for the formation of the α-hydroxyvaline.
377
378
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids Yonemitsu oxidation-cyclodehydration Harran et al., ACIE 2003, 42, 4961 Robinson-Gabriel cyclodehydration Nicolaou et al., JACS 2004, 126, 12888 Nicolaou et al., JACS 2004, 126, 12897 Cyclodehydrative oxidation Magnus et al., JACS 2007, 129, 12320 Macrolactamization Nicolaou et al., JACS 2004, 126, 12888 Nicolaou et al., JACS 2004, 126, 12897
Late-stage NCS-mediated chlorination Harran et al., ACIE 2003, 42, 4961 Nicolaou et al., JACS 2004, 126, 12888 Nicolaou et al., JACS 2004, 126, 12897
Me Me HO
Me H N
H N
Me N
O O
O
Cl
N
Cl
O Peptide coupling Harran et al., ACIE 2003, 42, 4961
O
N H
N H
SmI2-hetero pinacol coupling and oxime cleavage cascade Nicolaou et al., JACS 2004, 126, 12897
4: diazonamide A
Figure 10.8
Highlights of synthetic approaches to 4 in the peripheral regions.
Me Me HO
Oxidative annulation Harran et al., ACIE 2003, 42, 4961 Electrophilic aromatic substitution Nicolaou et al., JACS 2004, 126, 12888 Nicolaou et al., JACS 2004, 126 ,12897 Selective O-to-C migration Magnus et al., JACS 2007, 129, 12320
Me H N O
Me
H N
N O O
N O
O
N H
Cl
Nucleophilic 1,2-addition to isatin Nicolaou et al., JACS 2004, 126, 12888 Magnus et al., JACS 2007, 129, 12320 Lewis acid-mediated hydroxymethylation Nicolaou et al., JACS 2004, 126, 12897
Cl NH
Witkop-type photoinduced macrocylization Harran et al., ACIE 2003, 42, 4961 Suzuki-Miyaura coupling Nicolaou et al., JACS 2004, 126, 12897
4: diazonamide A
Figure 10.9
Highlights of synthetic approaches to 4 at the polycyclic core of 4.
Diazonamide A (4) is by far the most complex indole-oxidized peptide alkaloid discussed in this subsection. It has generated considerable attention in the synthetic community, not least for the significant challenge posed by the dense polycyclic core, which has elicited several unique approaches to its construction. Unfortunately, due to space restrictions, not all of disconnections (Figures 10.8 and 10.9) can be discussed in detail here. Instead, greater priority is given to more biomimetic aspects. 10.1.5.1 Late-Stage Aromatic Chlorination Early model studies on simplified substrates highlighted the potential of installing the chlorine atoms at the later stages of the synthesis (Scheme 10.27a) [40]. This strategy could be transferred to more complex intermediates and was used in all successful total syntheses of 4. The highlighted chlorination step used toward the first total synthesis of the correct structure of diazonamide A (4) (Scheme 10.27b) [38] was employed with a similar outcome later as well [37].
10.1 Indole-Oxidized Cyclopeptides Me Boc
Me
Me N
N H
O
N
N H
i
O
Me
Boc
N
Cl
N
Cl O
O
86%
NH
379
NH
(a) Me H N Teoc
Me Me
H
H N Teoc
N N O
(b)
OO
30-50%
O
Cl
N
Cl
O
NH
N H
O
N
ii
N
O
Me
H N
O
NH
N H
Scheme 10.27 Late stage chlorination reactions: (a) model studies [40]; (b) total synthesis of 4 [38]. Reagents and conditions: (i) N-chlorosuccinimide (4.0 equiv.), CCl4 , 40 ◦ C, 18 h; and (ii) 2,3,4,5,6,6-hexachloro-2,4-cyclohexadien-1-one (2.5 equiv.), N, N -dimethylformamide, rt, 24 h.
10.1.5.2 Bisoxazole Ring System via Oxidative Dehydrative Cyclization Moody showed that a diazonamide precursor peptide could be directly oxidized to a putative intermediate (Scheme 10.28) [41]. Though low yielding, this approach nonetheless enables direct access to the key 2,4-bisoxazole core 107 in a single step starting from the readily accessible peptide 106. This interesting result hints at a biosynthetic route whereby the polycyclic core may be formed in a single enzyme-mediated oxidation step.
H Cbz
N
Me O N H
Me H N O
N
O N H
OMe O
H
Cbz
i
H N
Me O
Me
O
N H N
N H 106
CO2Me
O
17% HO
HO
N
N
N H
H
107
Scheme 10.28 Biomimetic synthesis of the 2,4-bisoxazole core [41]. Reagents and conditions: (i) 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ), THF, reflux.
10.1.5.3 Oxidative Annulation Hypervalent iodine reagents were used by Harran to form the critical aminal core by oxidative activation of the Tyr residue (Scheme 10.29) [38]. Indolyl-oxazole 108 was treated with a slight excess of PhI(OAc)2 in the presence of LiOAc, which presumably activates at the phenol OH group (109) for nucleophilic attack by the
380
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids
Me Ns
H N
O N H
Me N CO2Me
O
i
Ns
H N
O
Me
Me
Me N
N H
CO2Me
O
HO
AcO I
Br
108
N H Ph
109
Me
Me H N
N
N H 113 (7-8%) O
CO2Me
Ns
Route A
N O
CO2Me
O
20-25%
H Ns N
Me
H N
N OO
N O H
Br
Br
Me
Me
N
N H 110
Me
OO
CO2Me
O
Br
Route B 15%, 1:1 d.r.
H N
N
OO H
O
N H
H Ns N
Me
H H N N Ns
Br
112
O
CO2Me N H
Br
111
Scheme 10.29 Formation of the aminal core by oxidative annulation [38]. Reagents and conditions: (i) PhI(OAc)2 (1.1 equiv.), LiOAc (2.0 equiv.), TFE, inverse addition, −20 ◦ C, 10 min; Ns = 4-nitrotoluenesulfonyl (nosyl), TFE = 2,2,2-trifluoroethanol.
electron-rich indole (Route A, 109 → 111). Formation of the ‘‘shunt’’ product 112 (Route B) and the undesired diastereomer 113 could easily be suppressed were the oxidation to occur in an enzymatic setting. 10.1.5.4 Sequential Nucleophilic 1,2-Addition, Electrophilic Aromatic Substitution In a distinct approach by Nicolaou and coworkers, organolithium reagent 114 was reacted with isatin 115 to produce tertiary alcohol 116 as a mixture of diastereomers (Scheme 10.30) [37]. Treatment with acid then enabled an SN 1 reaction between 116 and the phenol ring of tyrosine 118, to form the critical quaternary carbon bond, presumably via intermediate 117. In this case, the Boc protecting group was lost during the aromatic substitution step and had to be reinstalled to facilitate the separation of the two diastereomers. Interestingly, attempts to form the ring inter-molecularly under various acidic conditions failed (120 → 121). This result could suggest that strong enzymatic assistance might be needed, were the cationic intermediate to be formed during biosynthesis. 10.1.5.5 Reductive Aminal Formation Despite promising studies on simpler substrates (typically 4.0 equiv. BH3 ·SMe2 ), aminal formation from the sterically encumbered indolinone 122 (Scheme 10.31)
10.1 Indole-Oxidized Cyclopeptides Me Boc
Me Me
N
N H
O
Boc
OBn
O
Me
Me
O
N
OBn
N MOM Br 115
H
O
H
N
N
CBz
MOM
CO2Me
Me
Me N
H2N O
Cbz N
−H+
OBn
−H2O
i Me N H
O OH 120
Br
116
118
O OH
OBn
N
Br
MOM
conditions Me
Me H N Cbz
N O
Me
H N
N
OBn
OO
Br
O OH 121
47%
H
N OO HO
HO O
O
Me
H
H Cbz N
N
N H
Li
114
MeO2C
381
OBn
O N H
N Br
119
117
N Br MOM
Scheme 10.30 Nucleophilic 1,2-addition followed by intramolecular electrophilic aromatic substitution and an attempted intramolecular transformation [37]. Reagents and conditions: (i) 118 (4.0 equiv.), p-TsOH (4.0 equiv.), ClCH2 CH2 Cl, 83 ◦ C, 25 min. Me H Cbz N
Me Me
H N
N
Me
N
OO
O O O 122
Cl
Cbz
i Cl
−H2O
N O
56%
H N
H N N OO
N H
N H
O H
N O
N H
123
Scheme 10.31 Reductive aminal formation [37]. Reagents and conditions: (i) DIBAL-H (1.0 M in PhCH3 , 100.0 equiv), THF, −78 → 25 ◦ C, 3 h.
proved to be highly challenging [122 → 123, 100.0 equiv. DIBAL-H (diisobutylaluminum hydride)!] [37]. Forged only under strong laboratory conditions, this bond construction is perhaps less likely to occur in Nature. 10.1.5.6 Indole–Indole Coupling The coupling of the Trp residues is thought to proceed via a biradical mechanism. In the flask, these radicals were photochemically generated from indolyl bromide 124 (Scheme 10.32), leading to 126 via intermediate 125 in an excellent yield and as a single atropisomer. Apparently, conformational constraints control the
Cl Cl N H
382
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids Me
H H N Teoc N
N
OO
O
N i
O N H 124
Me
bi-radical coupling then aromatization
Me
Br
N H
N H Li
Br
H H N N Teoc
N H LiBr O
N
OO
125
N O
72% O
OAc
Me
N H 126
N H OH
Scheme 10.32 Witkop-type photo-induced macrocyclization [38]. Reagents and conditions: (i) hv (300 nm) argon-purged CH3 CN–H2 O (3 : 1, 3.0 mM), LiOH (2.0 equiv.) rt, 3 h; Teoc = 2-(trimethylsilyl)ethoxycarbonyl.
stereochemical outcome of this reaction and the difference in electron density between the two aryl rings facilitates the biaryl coupling. The conditions for this transformation were originally developed by Harran [38]. Nicolaou applied a similar strategy before aminal formation after slight modification to the reaction conditions [37]. All the synthesis data indicate that in Nature an enzymatically derived radical would react in a similar fashion. Both Nicolaou and Harran have charted the intrinsic reactivity of many putative biosynthesis intermediates. By inference, the radical coupling path way could be the more accessible one for biosynthesis. Future research on the producer organism and biosynthesis genes will hopefully clarify this point, and provide insight into the enzymatic machinery that enables the formation of this highly peculiar molecule in Nature. 10.2 A Complex Peptide Alkaloid: Ecteinascidin 743 (ET 743)
The ecteinascidins (Figure 10.10, 127–132) are a family of alkaloids derived from the sea squirt Ecteinascidia turbinata, which quickly raised considerable interest due to their potent antiproliferative properties [42–44]. Their activity was uncovered in 1969 during a natural product screening program of the American National Cancer Institute (NCI) [45]. Follow-up studies led to their isolation and structural identification [42], including ecteinascidin 743 (ET 743, trabectidin, 128) as the most abundant marine alkaloid of the family – still at a very low content of 0.5–4 ppm in the biomass. The remarkable three-dimensional architecture, combined with its meager availability from natural sources rendered ET 743 a very attractive synthetic target. This led to key examples of biomimetic synthesis, which were instrumental in advancing therapeutic studies. Semi-synthetic ET 743 was recently approved in Europe for the treatment of advanced soft tissue sarcoma and is marketed as Yondelis [46]. The ecteinascidins feature a piperazine-bridged bis(tetrahydroisoquinoline) framework and a pentacyclic A–E ring system, which they share with the bacterial safracin and saframycin alkaloids. Linked to this platform is a bridged ten-membered lactone ring that incorporates a benzylic thioether and a
10.2 A Complex Peptide Alkaloid: Ecteinascidin 743 (ET 743) OMe HO MeO
HO O AcO
Me A
OMe
NH HO S O H B
C
O Me
N
O
N
O
Me
O
O
R5
NH
X
O
NH2
Figure 10.10
Me
MeO R2
O
127: ET 729 (R1 = H, X = OH) 128: ET 743 (R1 = Me, X = OH) 129: ET 745 (R1 = Me, X = H) 130: ET 759A (R1 = Me, X = = O, lactam) 131: ET 759B (R1 = Me, X = OH, S-oxide) 132: ET 770 (R1 = Me, X = CN)
Me
H
Me
Me
MeO
R1 D
O
O O
E N
OMe Me
H
Me
383
R3 R4 NH O Me
133: Safracin A (R2 = H) 134: Safracin B (R2 = OH) 135: Cyano-safracin B (R2 = CN)
136: Saframycin A (R3 = CN, R4 = H, R5 = H) 137: Saframycin B (R3 = R4 = R5 = H) 138: Saframycin C (R3 = R4 = H, R5 = OMe) 139: Saframycin G (R3 = CN, R4 = H, R5 = OH) 140: Ranieramycin C (R3 = R4 = O, R5 = OH) 141: Ranieramycin D (R3 = R4 = O, R5 = OEt)
Some of the most representative ecteinascidins and related compounds.
spiro-tetrahydroisoquinoline unit, rendering the ecteinascidins significantly more complex than the safracins (133–135) and saframycins (136–141). The absolute stereochemistry was determined by X-ray crystal structure analysis of the natural N12 -oxide of ET 743 [47] and a derivative of ET 729 [48]. 10.2.1 Biosynthesis and Biomimetic Strategy
It may come as a surprise that ET 743 is derived from a peptide precursor. However, feeding experiments on both enzymatic preparations and living organisms using radiolabeled amino acids showed that tyrosine and cysteine are employed in the biosynthesis of ecteinascidins [49, 50]. The two-carbon unit C1–C22 is presumably derived from glyoxylate, which probably originates from glycine [48, 51]. The gene cluster of the structurally related saframycin A was characterized and gives good precedence for ET 743 (Scheme 10.33). A core tetrapeptide 142 (Ala-Gly-Tyr-Tyr) is thought to be assembled first by an NRPS machinery. The conversion of Tyr into the observed 3-hydroxy-5-methyl-Omethyltyrosine could take place either before its incorporation into the NRPS Me
MeO
OMe
Me O
HO Me
NH2 H N
MeO Me O N
O
O
N H
NRPS S OH
O
OMe Me 142
Scheme 10.33
HO
A
O
OH
Me N E OBN C D
OMe Me
Me
O
MeO
Me E
H B
A
OH
HN
O
1 N 22
O
C
N
O Me
H CN NH
O O
NH2
Me 143
Gene cluster-based proposal for saframycin A biosynthesis.
136: saframycin A
384
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids
machinery or during assembly [52, 53]. Reductive cleavage likely generates keto-piperazinol intermediate 143, which then becomes the subject of an intricate maturation phase that ultimately delivers saframycin A (136) via several oxidoreduction, cyclization, N-methylation, and nitrile introduction steps. The order of ring closures remains to be clarified, but as the peptide chain is apparently released as an aldehyde it seems likely that ring C is closed first, followed by annelating E through D and then ring A through B later (arrows on Scheme 10.33). The structural similarity between ET 743 (128) and saframycin A (136) (Figure 10.10) strongly suggests a common biosynthetic pathway for the pentacyclic core, which helps to guide a speculative retrobiosynthetic simplification of ET 743 (Scheme 10.34). Logical disconnections removing l-dopa (128 → 144) and cysteine (144 → 145) reveal a similar core structure 145. However, only a complex sequence of post-assembly modifications can transform a Tyr-Tyr-Gly-based tripeptide 147 into 146: reductive cleavage, carbinolamine formation, ring cyclization, N-methylation, oxidative deamination (NH2 → C = O), amide reduction (C = O → OH), esterification (with Cys), benzyl cation formation, intramolecular cysteine addition, oxidative deamination, and a stereoselective Pictet–Spengler condensation! OMe HO MeO
NH2 NH O AcO
Me
OMe
HO
O HO
Me
O SH
Me N
N
N O
OH
Me
H N
•Me
N
O
•Me
O
H OH O
O
OH
128: Ecteinascidin 743
O
Me
Me
N
O
O O
HO O SH
•Me
H
OMe
HO
NH2 HS
144
145
Me
O
Me
OH
O O H2N
N H
Me O N
OH O
O S
NRPS OH
O
OMe
147
OH N
O
O H2N
N
Me
OMe Me
OH 146
Me
Scheme 10.34
Proposed biosynthesis for ET 743.
While the individual steps remain to be further illuminated, their formal necessity has led to the development of an array of elegant methods and total syntheses, all of which feature strongly biomimetic elements. Total syntheses were accomplished first by the group of Corey et al. [54], then by Fukuyama [55] and Zhu [56]. Williams [57] and Danishefsky [58] have contributed syntheses of the A–E pentacycle (Figure 10.11). Based on this precedent, a semi-synthesis starting from cyanosafracin B has been developed, which is used to produce
10.2 A Complex Peptide Alkaloid: Ecteinascidin 743 (ET 743) Pictet Spengler cyclization Pomerantz-Fritsch reaction Corey et al., JACS 1996, 118, 9202 Zhu et al., JACS 2006, 128, 87 Fukuyama et al., JACS 2002, 124, 6552 Danishefsky et al., ACIE 2006, 45, 1754 Zhu et al., JACS 2006, 128, 87 Spontaneous phenol-aldehyde cyclization Transamination Fukuyama et al., JACS 2002, 124, 6552 Corey et al., JACS 1996, 118, 9202 Fukuyama et al., JACS 2002, 124, 6552 Zhu et al., JACS 2006, 128, 87 Ortho-quinone methide capture Corey et al., JACS 1996, 118, 9202 Fukuyama et al., JACS 2002, 124, 6552 HO Zhu et al., JACS 2006, 128, 87 H G
NH
MeO
O AcO
Me
F
O
A
B
S
C
OR1
E
N
•Me
Me A
OH
O Pictet-Spengler cyclization Corey et al., JACS 1996, 118, 9202
Figure 10.11
Me
B
N
C
R2
Carbinolamine formation Corey et al., JACS 1996, 118, 9202 •R Amide coupling Corey et al.,OL 2000, 2, 993 Williams et al., OL 2003, 5, 2095 Strecker reaction Zhu et al., JACS 2006, 128, 87
E
D
N
O
O O
OMe HO
Me
D
N
Mannich bisannulation Corey et al., JACS 1996, 118, 9202 Pictet-Spengler cyclization Corey et al., OL 2000, 2, 993 Williams et al., OL 2003, 5, 2095 Zhu et al., JACS 2006, 128, 87 Danishefsky et al., ACIE 2006, 45, 1754 Intramolecular Heck reaction Fukuyama et al., JACS 2002, 124, 6552
OMe
HO
385
CN
Ugi four component reaction Fukuyama et al., JACS 2002, 124, 6552
Summary of synthetic approaches to ET 743.
ET 743 on scale for drug use [59]. All the syntheses can be divided into three key phases: ABCDE-pentacycle formation, bridge installation, and the endgame. For the pentacycle several different access pathways have been developed, while the later stages of the synthesis have been realized in a fairly comparable manner by Corey, Fukuyama, and Zhu – all very close to anticipated biosynthetic events. 10.2.2 Pentacycle Formation
The seminal and optimized studies [54, 60] are based on biomimetic reactivity. Carbinolamine formation from amino acid ester 148 and a Mannich bis-annulation were used to effectively form the ABC ring system 149 (Scheme 10.35). An amino acid was appended to yield lactone 150, which was reduced and the phenol protecting groups cleaved to yield lactol 151. This was engaged in a biomimetic, diastereoselective Pictet–Spengler reaction via iminium-ion 152 to install the CD rings (→ 153). Williams et al. (Scheme 10.36) [57] employed an auxiliary β-lactam to control the stereochemical course of the first of two planned Pictet–Spengler reactions (154 → 155). After coupling of an unnatural N-methyltyrosine (156), the β-lactam was skillfully brought into play as a masked aldehyde: after chemoselective reduction with LiEt3 BH, the intermediate metalated aminal opened and eliminated BnNH2 . The resulting α, β-unsaturated aldehyde condensed with the free secondary amine to form iminium ion 157, which underwent an intramolecular Pictet–Spengler cyclization cascade to furnish the pentacycle 158. In Danishefsky’s approach (Scheme 10.37) [58], a Pomeranz–Fritsch isoquinoline synthesis was used to prepare AB-ring precursor 159, which carried an aldehyde on ring E. Acid-mediated Boc-group cleavage initiated the formation of
386
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids
O OH Me
H
OH
O OMe
O NHCbz
O
OMe
O
Me
i-iii 73%
O
Me
2 steps
H NH
81% O
O
O 148
H 149
OMe OTBS TBSO O OAll O H NH Alloc N H O H O 150 iv,v
HO OAll Me
OH
HO OAll
H N
4 steps
Alloc
N
O O HO
OMe
OMe
OMe
Me
68%
H
OAll
H
O HO
153
Me
vi
Alloc
N N
O
CN
OH
57% O
HO O
H
OH NH Alloc
N
O
H
HO
O
H
H
O
151
152
Scheme 10.35 Corey’s approach to the pentacycle core 153 of ET 743. Reagents and conditions: (i) BF3 ·OEt2 , H2 O; (ii) BF3 ·OEt2 , 4 A˚ molecular sieves; (iii) H2 , Pd/C; (iv) LiAlH2 (OEt)2 , Et2 O, −78 ◦ C; (v) KF, MeOH; and (vi) 0.6 M TfOH, H2 O–CF3 CH2 OH (3 : 2), BHT, 45 ◦ C. BHT = 2,6-difert butyl-4-methyl phenol. Bn OMe N
Bn OMe N
O
Me
O 18% N
MeO
Me
Ph OBn
7 steps
OBn
Ph 154
O
NH
MeO
O
OMe
82% 2 steps
Bn N
H O
Me
NHMe N
MeO
O OBn
OBn
OBn
155
156
Me
OH OMe
+
LiEt3BH, then H OMe HO
Me
OMe
MeO
Me
N N
MeO OBn
O OBn 158
Scheme 10.36
H
Me
Me 49%
OH NMe
N
MeO OBn
OMe Me
O OBn 157
Williams’ synthesis of the pentacyclic core 158 of ET 743.
the D-ring imminium ion 160, which engaged in a diastereoselective vinylogous Pictet–Spengler cyclization to form the CD-rings of pentacycle 161. Fukuyama’s synthesis (Scheme 10.38) [55] featured an atom-economic Ugi reaction to assemble starting material 162, which was transformed into diketopiperazine 163 in four steps. Chemoselective reduction of the secondary amide allowed transformation to enamine 164, which was subjected to a (non-biomimetic) Heck cyclization to efficiently generate the CDE-ring system 165. Multiple functional group
10.2 A Complex Peptide Alkaloid: Ecteinascidin 743 (ET 743)
O
BnO
OMe
Me
OH
N
O O
O OBn
Me
Me
BnO H
OH 58%
N
N
O O
Boc
Me
H
Me
Me i N
OMe
OMe
OBn
H
OH
387
O
Me
Me
N N
O
H
O
O OBn
OBn
159
Me
160
161
Scheme 10.37 Danishefsky’s synthesis of ET 743’s pentacyclic core (161): Reagents and conditions: (i) CHF2 CO2 H, MgSO4 , benzene. OMe
Me
Me PMP O NH NH
OMOM Me N
OMe OBn
Me OMOM
I Boc 4 steps 44%
O
Me
Me
O
Me
OMe 4 steps Me
NH
O H AcO
O
OBn O
162
I
Me
61%
N O
O H TBDPSO
OBn
OMs N N
O H O AcO
I
Boc
O
163
164 i 83%
OMe HO Me
OH OH H 4
N
Troc
ii 84%
O H AcO
BnO
OMe Me
BnO
OBn O
N O
OMe Me
CN 167
Me
N
Troc
N
14 steps
Me
N N
8%
O
O O
Me
OMs H
H AcO
CN 166
O
H AcO
O 165
Scheme 10.38 Synthesis of the pentacyclic core (167) of ET 743 developed by Fukuyama. Reagents and conditions: (i) Pd2 (dba)3 (5 mol. %), P(o-tol)3 , (20 mol. %), TEA, CH3 CN, reflux and (ii) Pd/C, H2 , THF.
interconversions gave the advanced aldehyde 166, which smoothly ring-closed in a phenol–aldehyde condensation to the advanced ABCDE pentacycle 167 after hydrogenolytic debenzylation. This effort was based in part on previous work on saframycin A [61]. Notably, the intramolecular electrophilic substitution of the phenol by the aldehyde yielded the requisite oxidation state at C4 for the construction of the ten-membered bridge (vide infra). In Zhu’s approach (Scheme 10.39) [56], the synthesis of the pentacyclic core starts with a rapid construction of D–E fragment 170 using a highly diastereoselective Pictet–Spengler condensation of Garner’s aldehyde [62] with substituted phenylalanol 168 (prepared from 3-methylcatechol in eight steps) [63]. After masking the secondary amine of 169 as the N-allyloxycarbamate and releasing the amino alcohol, the subsequent D–E fragment is coupled with 171 (prepared from sesamol in six steps) [56] to afford intermediate 172 as the major diastereomer. Ring C was
Boc
O
4
N
176
Me
Alloc
OMe
Me
84%
iv O O
MOMO Me N
175
CN OAc
N
Me
Alloc
OMe
170
38%
O
Me Me +
O O
MOMO Me
OMe
4 steps
AcO
AllylO O
AllylO
H2N
HO
HO 169
33%
4 steps
AllocN
Me
HN
HO
95%
N
O
Boc
Me
i
CN OAc
N
AllylO OH H
168
Me
OMe
174
O
171
N
CN OEt
N
Br
Alloc
Me
ii 91%
OMe
COOEt
AllylO OTBS
O
OMOM
78%
iii
O 172
O
O O
MOMO Me
O
Me
MOMO
O
N
173
Me
Me
Alloc
OMe
Alloc
N
OH OEt
NH
AllylO OTBS
2 steps 95%
OAc OEt
NH
OMe
Scheme 10.39 Synthesis of the pentacyclic core (176) of ET 743 developed by Zhu. Reagents and conditions: (i) N-Boc-L-serinal-N,O-dimethylacetal, AcOH, CH2 Cl2 /CF3 CH2 OH (7 : 1), 3 A˚ molecular sieve; (ii) TEA, MeCN, 0 ◦ C; (iii) Dess–Martin reagent then TMSCN, ZnCl2 ; and (iv) TFA, CH2 Cl2 .
O
MOMO Me
HO
H2N
HO
OMe
AllylO OH
388
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids
10.2 A Complex Peptide Alkaloid: Ecteinascidin 743 (ET 743)
389
constructed into the D–E segment using a decisive zinc chloride-catalyzed Strecker reaction to give intermediate 174. A Pomeranz–Fritsch-type cyclization installed ring B, with the corrected oxidation state at C4 affording the pentacycle 176. Overall, many of these key transformations were conceived at oxidation stage levels similar to the natural substrates and hence are considerably biomimetic, but their strategic placement within the scaffold deviates. Compared with their final strategic ring closures, Corey’s and Williams’ syntheses used the D-ring for a biomimetic annelation of a nucleophilic aromatic ring, which may be formed in biosynthesis during earlier steps. Danishefsky employed an imminium ion cyclization to close the C-ring, which in biosynthesis will be formed first. Fukuyama and Zhu close the B-ring late in synthesis, comparable to the anticipated biosynthesis, but by forming a C–C bond that in biosynthesis is part of an amino acid. 10.2.3 Bridge Formation
The centerpiece of the ecteinascidin scaffold synthesis is certainly the construction of the ten-membered-ring thioether bridge. Saliently, retrobiosynthetic analysis strongly suggested a cysteine-thiol capture, which was put into synthetic practice in two different forms (Scheme 10.40). Corey managed to selectively oxidize the A-ring arene to 177, which was eliminated to give a highly reactive quinone methide 178. Base-induced Fm-deprotection then unleashed the nucleophilic thiolate 179 −
OMe MOMO Me
Me O
O OH H Me
N
i-iii
N
O O
O
O O SFm
O
(a)
177
Me
AllO OH OH H
OMe
HO R
1
i
Me
O
N
O
O O
CN
H
v
Me
79% O O
CN
180 (R =Mom)
H
NHAlloc HS
O
O
Me
+
O
O AcO
O H ii N
O O
182
NHAlloc OMe Me RO O SH N Me N
179
H
O
SR NHAlloc
O
O
N
O O
O
NHAlloc
+
H
O AcO
N
SFm NHAlloc
HS
Me
Me
O O 178
NHAlloc
N
iv
N
CN
O
O
H
Me
H
NHAlloc
S
71%
Me O
NHAlloc OMe AllO Me O SH 1 N R N
O 183
CN 184
1
(b)
181a (R = H; R = Troc) 1 181b (R = CAr3; R = Alloc)
Ar =
OMe
Scheme 10.40 (a) Corey’s strategy for bridge formation via an ortho-quinone methide. Reagents and conditions: (i) DMSO, Tf2 O, −40 ◦ C; (ii) i-Pr2 NEt, 0 ◦ C; (iii) t-BuOH, 0 ◦ C; (iv) (Me2 N)2 C=N-t-Bu, 23 ◦ C;
and (v) Ac2 O, 23 ◦ C. Fm = fluorenylmethyl. (b) Fukuyama’s and Zhu’s approaches for bridge formation via a benzyl cation. Reagents and conditions: (i) TFA, CF3 CH2 OH; and (ii) Ac2 O, pyridine, DMAP.
390
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids
in situ, which added to the quinine methide and formed the sought after thioether 180 in very good yield (Scheme 10.40a). On the other hand, Fukuyama accessed the D-ring in the correct oxidation stage at the benzylic carbon (Scheme 10.40b). Hence, simple acid treatment of 181a led to the highly stabilized benzyl cation 182 (isoelectronic to protonated quinine methide 183), which captured the free thiol chain and delivered the thioether bridge in 184 [55]. Zhu [56] employed an efficient one-pot deprotection/cyclization of the S-trityl protected precursor 181b to convert it into 184. In all cases the product was isolated as the more stable phenol acetate. The efficient biomimetic synthesis of this bridge both by nucleophilic and by electrophilic chemistry indicates that this ring closure will readily occurs in Nature, where – according to typical oxidative biosynthesis chemistry – a cationic pathway may be in operation. 10.2.4 Endgame
To finish the synthesis of ET 743, an l-dopa derivative must be incorporated. Here, Fukuyama and Zhu essentially follow Corey’s precedent (Scheme 10.41) [56, 61]. Allyloxycarbamate 185 is deprotected to the free amine, and a pyridinium salt closely related to the amino-transferase cofactor pyridoxal pyrophosphate is then used to execute an oxidative deamination to ketone 186. After this astonishingly biomimetic move, the ketone was condensed with 5-(2-aminoethyl)-2-methoxyphenol in a Pictet–Spengler-like tetrahydroisoquinoline synthesis to afford 187. Notably, this condensation was exquisitely facile and completely stereoselective even without the aid of an enzyme. Apparently, the steric bulk of the E-ring pointing in the same direction leads to an exo-attack of the methoxyphenol ring on the intermediate imminium ion. Final protecting group manipulations delivered ET 743 (128). Overall, among peptide alkaloids the ecteinascidines have arguably stimulated some exquisite biomimetic synthesis plans, and the methods developed for this particularly intriguing scaffold will continue to stimulate the field of biomimetic synthesis. Notably, a large set of these transformations was adapted to the semi-synthesis of ET 743 for drug use. This illustrates that biomimetic HO O NHAlloc OMe NH OMe OMe MeO O O RO Me O RO Me RO Me AcO O S AcO O S AcO O S H H H Me Me Me N Me i,ii N Me iii N Me 77% 82% 2 steps 70% N N N O O O O O CN CN O CN 185 186 187 H
128
Scheme 10.41 Biomimetic completion of ET 743 total synthesis. Reagents and conditions: (i) n-Bu3 SnH, PdCl2 (PPh3 )2 , AcOH; (ii) N-methylpyridinium-4-carboxyaldehyde iodide, DBU, (CO2 H)2 ; and (iii) 5-(2-aminoethyl)-2-methoxyphenol, silica gel. R = Mom.
10.3 Outlook
391
syntheses can be competitive on a larger scale, and can even positively impact process research.
10.3 Outlook
In this and the preceding chapter we could neither cover all peptide alkaloid structures nor possibly refer to all the excellent work executed beyond the material we were able to discuss. The world of peptide alkaloids is certainly much larger. For example, we had to omit other glycopeptides antibiotics such as teicoplanin, complestatin (188/189, Figure 10.12), or kistamycin, the indole-crosslinked alkaloids of the kapakahine family (190), or the instructive phallotoxins (191) and amatoxins from the death cap mushroom family – to mention just a few. However, from all the knowledge available and the selection of chemistry illustrated above, it becomes increasingly apparent that peptide alkaloid structures make use of a general toolbox of enzymatic manipulations that show similarities even between very distinguished RPS and NRPS biosynthetic pathways and different species in the kingdoms of life (bacteria, fungi, plants). Especially, the specific control Nature exerts over assembly, maturation, and tailoring of a peptide OH
HO2C
H N O
H
O
H N
O N Me
O
Cl
N H H
HN H N
O O
Cl Cl
OH
N HH
OH
HO2C
O
H N
O O
Cl Cl
OH
H N
OH
6
O
Cl OH
188: chloropeptin I
H
N Me
O
7
H N
O O
Cl
N HH
O
Cl Cl
OH
OH
O
H N
N HH
Cl Cl
N
H
189: chloropeptin II/complestatin
HN N
O
O
N HH Me
N
H N
S N
HN
Me
H O O Me Me O NH NH2 H O H N N H O Me Me 190: kapakahine A
Figure 10.12 Other peptide alkaloid structures. Oxidation-derived connector bonds are highlighted.
N H H
HO H H H N
OH
OOO N O H HH O NH N H N O OHO
HO
191: phalloidin
Cl OH
OH
OH
O
O
392
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids
alkaloid is highly stimulating. For more precise control of reactions on peptide precursors in the chemical sense and for targeted planning, more specific oxidation and crosslinking chemistry needs to be developed. In such synthesis endeavors, chemo- and regioselectivity will be much more necessary (and may be more difficult to achieve) than the central stereoselectivity that dominates the assembly of carbon chains. These efforts may eventually contribute to a different approach to synthesis – namely to more manipulating of oligomers, and to less assembling of building blocks.
References 1. (a) Koguchi, Y., Kohno, J., Nishio, M.,
2.
3.
4.
5. 6.
7.
8.
Takahashi, K., Okuda, T., Ohnuki, T., and Komatsubara, S. (2000) J. Antibiot., 53, 105–109; (b) Kohno, J., Koguchi, Y., Nishio, M., Nakao, K., Kuroda, M., Shimizu, R., Ohnuki, T., and Komatsubara, S. (2000) J. Org. Chem., 65, 990–995. Kaiser, M., Siciliano, C., Assfalg-Machleidt, I., Groll, M., Milbradt, A.G., and Moroder, L. (2003) Org. Lett., 5, 3435–3437. Berthelot, A., Piguel, S., Le Dour, G., and Vidal, J. (2003) J. Org. Chem., 68, 9835–9838. Sattely, E.S., Fischbach, M.A., and Walsh, C.T. (2008) Nat. Prod. Rep., 25, 757–793. Coste, A., Couty, F., and Coste, G. (2008) C. R. Chim., 11, 1544–1573. (a) Lin, S. and Danishefsky, S.J. (2001) Angew. Chem., 113, 2021–2024; Angew. Chem. Int. Ed., 40, 1967–1970; (b) Lin, S. and Danishefsky, S.J. (2002) Angew. Chem., 114, 530–533; Angew. Chem. Int. Ed., 41, 512–515; (c) Lin, S., Yang, Z.-Q., Kwok, B.H.B., Koldobskiy, M., Crews, C.M., and Danishefsky, S.J. (2004) J. Am. Chem. Soc., 126, 6347–6355. (a) Albrecht, B.K. and Williams, R.M. (2003) Org. Lett., 5, 197–2000; (b) Albrecht, B.K. and Williams, R.M. (2004) Proc. Natl. Acad. Sci. U.S.A., 101, 11949–11954. (a) Inoue, M., Furuyama, H., Sakazaki, H., and Hirama, M. (2001) Org. Lett., 3, 2863–2865; (b) Inoue, M., Furuyama, H., Sakazaki, and H., Hirama, M. (2003)
9.
10.
11.
12.
13.
14.
15.
Angew. Chem., 115, 2758–2761; Angew. Chem. Int. Ed., 42, 2654–2657. (a) Leung, T.-W.C., Williams, D.H., Barna, J.C.J., Foti, S., and Oelrichs, P.B. (1986) Tetrahedron, 42, 3333–3348; (b) Kahn, S.D., Booth, P.M., Waltho, J.P., and Williams, D.H. (1989) J. Org. Chem., 54, 1901–1904; correction: Kahn, S.D., Booth, P.M., Waltho, J.P., Williams, D.H. (2000) J. Org. Chem., 65, 8406. (a) Kobayashi, J., Suzuki, H., Shimbo, K., Takeya, K., and Morita, H. (2001) J. Org. Chem., 66, 6626–6633; (b) Suzuki, H., Morita, H., Iwasaki, S., and Kobayashi, J. (2003) Tetrahedron, 59, 5307–5315; (c) Suzuki, H., Morita, H., Shiro, M., and Kobayashi, J. (2004) Tetrahedron, 60, 2489–2495. Puk, O., Bischoff, D., Kittel, C., Pelzer, S., Weist, S., Stegmann, E., S¨ussmuth, R.D., and Wohlleben, W. (2004) J. Bacteriol., 186, 6093–6100. Example for aziridine biosynthesis: Ogasawara, Y. and Lu, H.-W. (2009) J. Am. Chem. Soc., 131, 18066–18068. Reviews: (a) Hu, X.E. (2004) Tetrahedron, 60, 2701–2743; (b) Lu, P. (2010) Tetrahedron, 66, 2549–2560. Bergmeier, S.C., Katz, S.J., Huang, J., McPherson, H., Donoghue, P.J., and Reed, D.D. (2004) Tetrahedron Lett., 45, 5011–5014. (a) Ma, B., Litvinov, D.N., He, L., Banerjee, B., and Castle, S.L. (2009) Angew. Chem., 121, 6220–6223; Angew. Chem. Int. Ed., 48, 6104–6104; (b) Ma, B., Banerjee, B., Litvinov, D.N., He, L., and Castle, S.L. (2010) J. Am. Chem. Soc., 132, 1159–1171.
References 16. Hu, W., Zhang, F., Xu, Z., Liu, Q.,
17.
18.
19. 20.
21.
22.
23.
24.
25.
26.
27.
28. 29.
Cui, Y., and Jia, Y. (2010) Org. Lett., 12, 956–959. Feng, Y. and Chen, G. (2010) Angew. Chem., 122, 970–973; Angew. Chem. Int. Ed., 49, 958–961. (a) Zaitsev, V.G., Shabashov, D., and Daugulis, O. (2005) J. Am. Chem. Soc., 127, 13154–13155; (b) Shabashov, D. and Daugulis, O. (2005) Org. Lett., 7, 3657–3659. Reddy, B.V.S., Reddy, L.R., and Corey, E.J. (2006) Org. Lett., 8, 3391–3394. (a) Lam, K.S., Hesler, G.A., Mattei, J.M., Mamber, S.W., Forenza, S., and Tomita, K. (1990) J. Antibiot., 43, 956–960; (b) Leet, J.E., Figroeder, D.R., Krishnan, B.S., and Matson, J.A. (1990) J. Antibiot., 43, 961–966; (c) Leet, J.E., Figroeder, D.R., Golik, J., Matson, J.A., Doyle, T.W., Lam, K.S., Hill, S.E., Lee, M.S., Whitney, J.L., and Krishnan, B.S. (1996) J. Antibiot., 49, 299–311. Umezawa, K., Ikeda, Y., Uchihata, Y., Naganawa, H., and Kondo, S. (2000) J. Org. Chem., 65, 459–463. (a) Broberg, A., Menkis, A., and Vasiliauskas, R. (2006) J. Nat. Prod., 69, 97–102; (b) Pohanka, A., Menkis, A., Levenfors, J., and Broberg, A. (2006) J. Nat. Prod., 69, 1776–1781. Fujimori, D.G., Hrvatin, S., Neumann, C.S., Strieker, M., Marahiel, M.A., and Walsh, C.T. (2007) Proc. Natl. Acad. Sci. U.S.A., 104, 16498–16503. Arroyo, V., Hall, M.J., Hassall, C.H., and Yamasaki, K. (1976) J. Chem. Soc., Chem. Commun., 845–846. Umezawa, K., Ikeda, Y., Kawase, O., Naganawa, H., and Kondo, S. (2001) J. Chem. Soc., Perkin Trans. 1, 1550–1553. (a) Parry, R.J. (1994) J. Chem. Soc., Chem. Commun., 995–996; (b) Tao, T., Alemany, L.B., and Parry, R.J. (2003) Org. Lett., 5, 1213–1215. Heemstra, J.R. and Walsh, C.T. Jr. (2008) J. Am. Chem. Soc., 130, 14024–14025. Poss, C.S. and Figreiber, S.L. (1994) Acc. Chem. Res., 27, 9–17. (a) Kamenecka, T.M. and Danishefsky, S.J. (2001) Chem. Eur. J., 7, 41–63; (b) Kamenecka, T.M. and Danishefsky, S.J. (1998) Angew. Chem., 110,
30.
31.
32.
33.
34.
35.
36. 37.
3164–3166; Angew. Chem. Int. Ed., 37, 2993–2995; (c) Kamenecka, T.M. and Danishefsky, S.J. (1998) Angew. Chem., 110, 3166–3168; Angew. Chem. Int. Ed., 37, 2995–2998 (a) Yu, S.-M., Hong, W.-X., Wu, Y., Zhong, C.-L., and Yao, Z.-J. (2010) Org. Lett., 12, 1124–1127; (b) Hong, W.-X., Chen, L.-J., Zhong, C.-L., and Yao, Z.-J. (2006) Org. Lett., 8, 4919–4922. Oelke, A.J., France, D.J., Hofmann, T., Wuitschik, G., and Ley, S.V. (2010) Angew. Chem., 122, 6275–6278; Angew. Chem. Int. Ed., 49, 6139–6142. Lindquist, N., Fenical, W., Vanduyne, G.D., and Clardy, J. (1991) J. Am. Chem. Soc., 113, 2303–2304. Fernandez, R., Martin, M.J., Rodriguez-Acebes, R., Reyes, F., Francesch, A., and Cuevas, C. (2008) Tetrahedron Lett., 49, 2283–2285. Wang, G., Shang, L., Burgett, A.W.G., Haran, P.G., and Wang, X. (2007) Proc. Natl. Acad. Sci. U.S.A., 104, 2068–2073. (a) Li, J., Chen, X., Burgett, A.W.G., and Harran, P.G. (2001) Angew. Chem., 113, 2754–2757; Angew. Chem. Int. Ed., 40, 2682–2685; (b) Li, J., Jeong, S., Esser, L., and Harran, P.G. (2001) Angew. Chem., 113, 4901–4906; Angew. Chem. Int. Ed., 40, 4765–4770; (c) Li, J., Burgett, A.W.G., Esser, L., Amezcua, C., and Harran, P.G. (2001) Angew. Chem., 113, 4906–4909; Angew. Chem. Int. Ed., 40, 4770–4773. Lachia, M. and Moody, C.J. (2008) Nat. Prod. Rep., 25, 227–253. (a) Nicolaou, K.C., Chen, D.Y.K., Huang, X.H., Ling, T.T., Bella, M., and Snyder, S.A. (2004) J. Am. Chem. Soc., 126, 12888–12896; (b) Nicolaou, K.C., Hao, J.L., Reddy, M.V., Rao, P.B., Rassias, G., Snyder, S.A., Huang, X.H., Chen, D.Y.K., Brenzovich, W.E., Giuseppone, N., Giannakakou, P., and O’Brate, A. (2004) J. Am. Chem. Soc., 126, 12897–12906; (c) Nicolaou, K.C., Hao, J.L., Reddy, M.V., Rao, P.B., Rassias, G., Snyder, S.A., Huang, X.H., Chen, D.Y.K., Brenzovich, W.E., Giuseppone, N., Giannakakou, P., and O’Brate, A. (2004) J. Am. Chem. Soc., 126, 15316–15316.
393
394
10 Biomimetic Synthesis of Indole-Oxidized and Complex Peptide Alkaloids 38. Burgett, A.W.G., Li, Q.Y., Wei, Q., and
39.
40. 41.
42.
43.
44.
45. 46.
47.
48.
49. 50. 51.
Harran, P.G. (2003) Angew. Chem., 115, 5111–5116; Angew. Chem. Int. Ed., 42, 4961–4966. Magarvey, N.A., Ehling-Schulz, M., and Walsh, C.T. (2006) J. Am. Chem. Soc., 128, 10698–10699. Magnus, P. and McIver, E.G. (2000) Tetrahedron Lett., 41, 831–834. (a) Sperry, J. and Moody, C.J. (2006) Chem. Commun., 22, 2397–2399; (b) Sperry, J. and Moody, C.J. (2010) Tetrahedron, 66, 6483–6495. Rinehart, K.L., Holt, T.G., Fregeau, N.L., Keifer, P.A., Wilson, G.R., Perun, T.J., Sakai, R., Thompson, A.G., Stroh, J.G., Shield, L.S., Seigler, D.S., Li, L.H., Martin, D.G., Grimmelikhuijzen, C.J.P., and G¨ade, G. (1990) J. Nat. Prod., 53, 771–792. Rinehart, K.L., Holt, T.G., Fregeau, N.L., Stroh, J.G., Keifer, P.A., Sun, F., Li, L.H., and Martin, D.G. (1990) J. Org. Chem., 55, 4512–4515. Wright, A.E., Forleo, D.A., Gunawardana, G.P., Gunasekera, S.P., Koehn, F.E., and McConnel, O.J. (1990) J. Org. Chem., 55, 4508–4512. Cuevas, C. and Francesch, A. (2009) Nat. Prod. Rep., 26, 322–337. European Medicines Agency, Meeting highlights from the Committee for Medicinal Products for Human Use, 16-19 July 2007, Press release 19 July 2007; http://www.ema.europa.eu/pdfs/human/ press/pr/43140707en.pdf (accessed 19 July 2007) Sakai, R., Rinehart, K.L., Guan, Y., and Wang, A.H.-J. (1992) Proc. Natl. Acad. Sci. U.S.A., 89, 11456–11460. Guan, Y., Sakai, R., Rinehart, K.L., and Wang, A.H.-J. (1993) J. Biomol. Struct. Dyn., 10, 793–818. Russel, G.K. and Miranda, N.F. (1995) J. Nat. Prod., 58, 1618–1621. Jeedigunta, S., Krenisky, J.M., and Kerr, R.G. (2000) Tetrahedron, 56, 3303–3307. Kapadia, G.J., Rao, G.S., Leete, E., Fayez, M.B.E., Vaishnav, Y.N., and
52.
53.
54.
55.
56.
57. 58.
59.
60. 61.
62.
63.
Fales, H.M. (1970) J. Am. Chem. Soc., 92, 6943–6951. Li, L., Deng, W., Song, J., Ding, W., Zhao, Q.-F., Peng, C., Song, W.-W., Tang, G.-L., and Liu, W. (2008) J. Bacteriol., 190, 251–263. Velasco, A., Acebo, P., Gomez, A., Schleissner, C., Rodriguez, P., Aparicio, T., Conde, S., Munoz, R., de la Calle, F., Garcia, J.L., and Sanchez-Puelles, J.M. (2005) Mol. Microbiol., 56, 144–154. Corey, E.J., Gin, D.Y., and Kania, R.S. (1996) J. Am. Chem. Soc., 118, 9202–9203. Endo, A., Yanagisawa, A., Abe, M., Tohma, S., Kan, T., and Fukuyama, T. (2002) J. Am. Chem. Soc., 124, 6552–6554. Chen, J., Chen, X., Bois-Choussy, M., and Zhu, J. (2006) J. Am. Chem. Soc., 128, 87–89. Jin, W., Metobo, S., and Williams, R.M. (2003) Org. Lett., 5, 2095–2098. Zheng, S., Chan, C., Furuuchi, T., Wright, B.J.D., Zhou, B., Guo, J., Danishefsky, S.J. (2006) Angew. Chem., 118, 1786–1791; Angew. Chem. Int. Ed., 45, 1754–1759. Cuevas, C., P´erez, M., Martin, M.J., Chicharro, J.L., Fernandez-Rivas, C., Flores, M., Francesch, A., Gallego, P., Zarzuelo, M., de la Calle, F., Garcia, J., Polanco, C., Rodriguez, I., and Manzanares, I. (2000) Org. Lett., 2, 2545–2548. Martinez, E.J. and Corey, E.J. (2000) Org. Lett., 2, 993–996. Fukuyama, T., Yang, L., Ajeck, K.L., and Sachleben, R.A. (1990) J. Am. Chem. Soc., 112, 3712–3713. (a) Garner, P. and Park, J.M. (1987) J. Org. Chem., 52, 2361–2364; (b) Garner, P. and Park, J.M. (1988) J. Org. Chem., 53, 2979–2984; (c) Garner, P., Park, J.M., and Malecki, E. (1988) J. Org. Chem., 53, 4395–4398. De Paolis, M., Chen, X., and Zhu, J. (2004) Synlett, 729–731.
395
Part II Biomimetic Synthesis of Terpenoids and Polyprenylated Natural Compounds
Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
397
11 Biomimetic Rearrangements of Complex Terpenoids Bastien Nay and Laurent Evanno
11.1 Introduction
‘‘Rearrangements form an integral part of terpene chemistry and sometimes take a truly spectacular course.’’ So was introduced the Simonsen lecture delivered by Ourisson in February 1964 at the Imperial College of Science and Technology, London [1]. This lecture gave an interesting overview dealing with some impressive terpene rearrangements, the most spectacular course of which, according to the author, had truly biomimetic likeness. Besides biomimetic polyolefin cyclizations to give (poly)terpenes, which have been extensively reviewed by others [2],1) complex rearrangements offer a second aspect of terpene biomimetic chemistry. Both aspects have been intensively studied since the nineteenth century, beginning in 1802 with Kindt’s experiments on turpentine oil transformation into artificial camphor [3]. The amount of work carried out is huge and obviously will not be treated exhaustively here. Recent examples have rather been privileged although we have tried to keep ‘‘classics’’ close to the discussion.
11.2 Beginning with Monoterpene Rearrangements 11.2.1 Historical Overview of Monoterpene Rearrangements: A Century since Wagner’s Structure of Camphene
In 1899, Georg Wagner described the pinacol-related rearrangement of borneol (1a) into camphene (2) [4] and of α-pinene (5) into bornyl chloride (4a) [5]. The
1) An example of polyolefin cyclization is
used to introduce Section 11.5 dealing with triterpene rearrangements. It will be
the only one described in this chapter, which deals with rearrangements of advanced biomimetic structures.
Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
398
11 Biomimetic Rearrangements of Complex Terpenoids Biosynthetic interrelationships in the monoterpene series
Wagner's and Meerwein's works Cl OH HCl H HCl ref. [4]
1) AcOAg 2) NaOH ref. [9c] HCl
camphene hydrochloride (3) HCl
WM shift
+
H
HCl
bornyl cation
+
Cl
menthyl cation
isobornyl chloride (4b)
cationic cyclization
+
ref. [9c]
bornyl chloride (4a)
isocamphyl cation
ref. [9c]
H
Cl
ref. [5]
a-pinene (5)
+
ref. [9c]
camphene (2)
borneol (1a) isoborneol (1b)
cationic cyclization
+ WM shift
pinyl cation
fenchyl cation
Scheme 11.1 Wagner’s and Meerwein’s work on the rearrangement of borneols (1), α-pinene (5), and camphene (2) (stereochemistry not specified) [4, 5, 9c], and relation to the biosynthetic routes of the corresponding cation (in the box) [6, 7].
transformations were performed in the presence of a strong acid and if we consider the general biosynthetic pathways for monoterpenes [6, 7] they are probably the earliest biomimetic transformations in the terpene series (Scheme 11.1). Not only did this work put an end to the eighteenth century controversy over the structure of camphene [8] but it opened the way to useful chemical transformations in organic synthesis. In 1914, Hans Meerwein started a series of discussions on the mechanism of this reaction [9].2) He finally suggested that the rearrangements were initiated by the ionization of the substrates into the corresponding carbocations [9c]. Meerwein’s carbocation concept and ionic mechanisms stand as an important breakthrough in organic chemistry. Strictly, the Wagner–Meerwein rearrangement can thus be defined as the carbocation-promoted 1,2-carbon shift of an alkyl or a hydrogen. Many related processes have been described involving longer, especially transannular, migration events. Wagner–Meerwein rearrangements can proceed in tandem as a succession of hydrogen or methyl (or alkyl) shifts. As shown in the following, the reaction is classical in terpene synthesis and biosynthesis [7]. Considering the economic importance of terpenes in the chemical, pharmaceutical, and fragrance industries, the rearrangement of α-pinene (5) into camphene (2) is a pivotal reaction in terpene manufacture, being the basis of the synthesis of camphor. The process has been improved and now relies on heterogeneous catalysis (Scheme 11.2). The isomerization of α-pinene (5) can be carried out over TiO2 , giving large amounts of camphene (2) accompanied by monocyclic terpenes resulting from ring opening–elimination reactions (terpinenes, cymene, limonene) [11, 12]. Further acid-catalyzed isomerization gives isobornyl acetate (6), which yields camphor (7) after hydrolysis and oxidation. 2) This was complementary to Meerwein’s
own work on the mechanism of the pinacol
rearrangement published a year before: see Reference [10].
11.2 Beginning with Monoterpene Rearrangements
TiO2
+
∆
a-pinene (5)
camphene (2)
+
+ ...
+
a-terpinene cymene
399
limonene
AcOH 1) NaOH 2) oxidation
camphor (7)
OAc 6 Scheme 11.2
O
Industrial process for the production of camphor (7) [11, 12].
11.2.2 Kinetics of the Monoterpene Rearrangement and Relation with the Catalytic Landscape in Terpene Biosynthesis
The Wagner–Meerwein rearrangement of a substrate R1 X results from a combination of thermodynamic and kinetic conformational events. It is conditioned by the formation of the carbocation R1 + , which is the rate-determining step, after which structural changes occur (R1 + → R2 + ). Formation of the carbocation is favored under acid catalysis and solvolytic conditions [13]. Depending on the complexity of R1 + , the reaction should not be restricted to a single mode of rearrangement (Figure 11.1). Knowing the activation barrier G‡ of the carbocation rearrangement is important for understanding the overall reaction mechanism from R1 X. Especially, this barrier ‘‘dictates’’ the choice of experimental conditions used to carry out these reactions [13]. Sorensen has studied the kinetics of terpene rearrangements and the thermodynamic stability of various carbonium ions under acidic and superacidic conditions [13, 14]. For a given reaction, it was possible to determine which of these processes has the lowest barrier by noting the products formed under varying experimental conditions. In particular, Sorensen studied the rearrangement of TS1+
TS2+
Case of complex organic molecules k1 R1+
k2 k3
R2+ R3+
k4
∆G‡
R5+
E
R1+
products
R4+
R1X
R2+ R3+
R5+
R4+ Products: R2X ... R5X
Figure 11.1 Reaction diagram of Wagner–Meerwein rearrangements involving complex intermediates: the reaction is not restricted to a single product.
400
11 Biomimetic Rearrangements of Complex Terpenoids
7
6
H3C
H3C
4 3 CH3
5
H
+
2 6 H3C 1 9
∆G‡= 11.2 kcal/mol
multiple WM processes
CH3 +
‡
∆G = 22.2 kcal/mol
13
Scheme 11.3
H3C
+
12
H3C
+
− 92°C
10 ∆G‡= 13.1 kcal/mol
H
‡
∆G = 18.5 kcal/mol
CH3
CH3
− 15°C
25°C
8
CH3
− 130°C
1 + 2 CH3 ∆G‡ < 4 kcal/mol
8
CH3
− 130°C
H3C
+
11
CH3 +
CH3 H s-delocalized carbonium intermediate
H3C
CH3 CH3
9
Rearrangement of the fenchyl cation (8) according to Sorensen [13].
the fenchyl cation (8) generated from α-fenchene (Scheme 11.3). In this system, the cation to cation barriers were readily overcome in formic acid as the solvolysis medium. The 1,2-shifts could involve partially σ-delocalized carbonium ions (box in Scheme 11.3) [15]. It was not possible to slow down the formation of the cation 9, and even at −130 ◦ C the combination of complex Wagner–Meerwein processes leading to the β-fenchene cation 10 was too rapid to be measured by NMR. However at −92 ◦ C the mixture evolved toward the ε-fenchene cation 11. The final ions 12 and 13 were only observed on raising the temperature to −15 and 25 ◦ C, respectively. These observations and related works allowed the activation energies to be inferred for each step (Scheme 11.3). Typically, the Wagner–Meerwein alkyl 1,2-shift of a tertiary cation into a parent tertiary cation has a low activation barrier <4 kcal mol−1 [13]. When the resulting cation is secondary instead, the activation energy increases by 5.5 kcal mol−1 . We complete this discussion with the interesting computer-assisted work of Johnson and Collins, who analyzed the multiple-rearrangement products of the norbornyl and fenchyl (8) cations [16]. They elaborated road maps for determining sequences of Wagner–Meerwein rearrangements, 6,2-endo-hydride shifts, 3,2-exo-hydride shifts, 3,2-exo-methyl shifts, and the ‘‘double Wagner–Meerwein’’ rearrangement which interconverts, given the rearranged cations. Important mechanistic information resulted from the use of graph models and computer programs. Applying this to Sorensen’s work on the rearrangement of the fenchyl cation (8, Scheme 11.3), there would be 330 theoretically distinguishable cationic intermediates, reduced to a minimum of 165 racemic ‘‘nodes’’ [16b]. The system appeared to seek out those structures that are thermodynamically most stable at a given temperature. We can bring these observations together with more recent works on terpene biosynthesis. The role of enzymes has been pointed out not only for the formation of a reactive carbocation but also in providing a stabilizing template for the cyclization reaction as well as controlling water access. In these terpene synthases, one cationic species only, or a few among the possible tens or hundreds, would be preferred and stabilized in an unreactive aprotic environment by way of interactions with the π-systems of aromatic residues [17]. By the controlled mutation of specific amino
11.3 Biomimetic Rearrangements of Sesquiterpenes
acids in the active site, it has been possible to define catalytic landscapes, in work used to understand the cyclization potential of linear C15 -precursors [18].
11.3 Biomimetic Rearrangements of Sesquiterpenes
With more important structural complexity, the sesquiterpenes offer larger rearrangement capabilities and higher chemical diversity than monoterpenes. Historical work on longifolene, as an analog of camphene, showed a very similar rearrangement pattern under acidic conditions, yet with additional reactivity or with limitations imposed by the shielding effect provided by the large transannular bridge [1, 19]. In this section, we will see how complex polycyclic and densely functionalized compounds can be obtained not only by cationic rearrangements but also by oxidative cleavage and cyclization cascades. Special attention will be devoted to caryophyllane polycyclic derivatives, which offer a rich and diversified chemistry. 11.3.1 Caryophyllenes in Sesquiterpene Biosyntheses
Caryophyllene (14) (trans isomer) and isocaryophyllene (15) (cis isomer) are the biosynthetic starting point for numerous polycyclic sesquiterpenes [20]. They are available on a large scale for use in the flavoring and fragrance industry. The chemistry of caryophyllenes has been well reviewed by Collado et al. [21]. We present herein a short description of relevant information about these compounds with regard to the biomimetic synthesis of sesquiterpenes. The caryolanyl (16) and clovanyl (17) cations are derived from the cationic cyclization and rearrangement of 14 (Scheme 11.4), whereas the presilphiperfolanyl cation (18) and its rearranged forms (silphinyl 19 and terrecyclanyl 20) are derived from the isomer 15. The co-occurrence of these different sesquiterpene skeletons in common living organisms shed light on their biogenetic relationships. 13 C NMR studies and MM1 calculations [22] have shown that there are four possible conformations of trans-caryophyllene (14) distinguished by the relative positions of the exocyclic methylene and olefinic methyl groups (Scheme 11.4). A 3 : 1 ratio has been measured between the main conformations, βα and ββ, having a low inversion barrier (16.25 ± 0.11 kcal mol−1 ). Treatment of caryophyllene (14) in acidic media (acetic, formic, sulfuric acids, and superacidic media) led to numerous rearranged compounds after 30 min, the relative yields of which reflect the relative population of the possible conformers. For longer reaction times (several days), the major products were caryolan-1-ol (from cation 16) and clov-2-ene (from cation 17). These results suggested reversible interconversion between intermediates that eventually evolved toward the more stable isomers [21]. In superacidic conditions (SO2 FCl) at −124 ◦ C, the distribution of products changed [23]. At this temperature, the various conformers of caryophyllene
401
11 Biomimetic Rearrangements of Complex Terpenoids
402
H
15 14 10
12
3 1 9
H
13 8
5
5
1
7
13
5
αβ
caryophyllene (14)
∆H kcal.mol−1 16.94 % (13C NMR) % MM1 0.09 % MM3 -
4
10
1
9 8
+
C5 C9+
4
H 1
5
9
C4
C9
+
5 H 1 9+ 8 10 13
5 8
+
10
terrecyclanyl cation (20)
siphinyl cation (19)
presylphiperpholanyl cation (18)
βα
13.69 25 21.22 26
14.78 3.37 44
12.94 75 75.33 29
15
H
H+ + C4 C8+
H
1 13 9 4 5 + 8
H
caryolanyl cation (16)
1 9 8 13 +
4 5
clovanyl cation (17)
H C8+ C9+
then H9
C4+
10 5
7
C5 C13 C1
12
3 1 9
H+ C4+
+
C13 C10 C5
14 10
1
9
4
αα
H
4
H
+
8
ββ
C5 C13 C1 C10
13
H 13 isocaryophyllene (15)
8 13 5
9 1 4
αα
βα
Scheme 11.4 Conformational equilibrium of caryophyllene (14) and isocaryophyllene (15) and general biosynthetic origin of polycyclic sesquiterpenes (numbering according to 14 and 15).
(14) are not rapidly interconverting, providing evidence for a significant transition barrier between the αα and ββ conformations. This gave experimental support for the results of population predictions based on MM3 calculations, suggesting proportions of 44, 29, and 26% for the αα, βα, and ββ conformers, respectively [24]. The interconversion between caryophyllene (14) and isocaryophyllene (15) has been reported under thermal conditions (pyrolysis at >240 ◦ C); it proceeds through a series of Cope [3,3]-sigmatropic rearrangements [25]. The presylphiperpholane (type 18), silphinane (type 19), and neoclovane skeletons (not shown) were only observed on treating isocaryophyllene (15) in acidic media, and were accompanied by other rearrangement products. To better control access to the different skeletons from caryophyllenes, the use of activated derivatives, for example, epoxides or tosylates, has been employed (see below) [26, 27]. 11.3.2 Biomimetic Studies in the Caryolane and Clovane Series
As discussed above, under acidic and superacidic conditions, caryophyllene (14) can undergo non-controlled rearrangements, mainly leading to caryolane and clovane
11.3 Biomimetic Rearrangements of Sesquiterpenes
403
derivatives, yet with low selectivity. Transformations performed on the α- and β-oxides of 14 afforded the caryolane and clovane frameworks, respectively, with much higher selectivity. In the presence of hydrated bismuth triflate utilized as a proton donor, caryophyllene-β-oxide 21, having an αα conformation, underwent cyclization on the α-face of the molecule, generating the bridgehead carbocation 22 (Scheme 11.5). The good overlap between the cation orbital and the transannular cyclobutane bond promoted Wagner–Meerwein rearrangement into 23. After β-elimination of a proton, 2-clovene (24) was obtained [27]. H
4
O
13
H
O
13
OH
Bi(OTf)3•xH2O CH2Cl2
H+
OH 8
60%
4
1
H
13 4
1
OH − H+
13
H
8 9
aa−21
21
1,2-shift
22
24
23
Scheme 11.5 Biomimetic synthesis of 2-clovene (24) through acid-catalyzed rearrangement of caryophyllene-β-oxide 21.
The Brønsted acid (PhO)2 P(O)OH promoted the same rearrangement of 21 towards the clovanyl cation 23, which was readily trapped by magnolol (25) available in the reaction mixture (Scheme 11.6a). This afforded clovanemagnolol (26), a natural product with neuroregenerative properties isolated from Magnolia obovata [28]. A similar reaction was performed with caryophyllene α-oxide 27, involving a ββ conformation. In this case, a non-matching alignment of the cation orbital with the transannular cyclobutyl C–C bond of cation 28 prevented the Wagner–Meerwein shift. The carbocation 28 was instead directly trapped by magnolol (25) to give caryolanemagnolol (29, Scheme 11.6b). The low yields OH OH OH H
4
H (PhO)2P(O)OH CH2Cl2 O
1,2-shift
22
HO O
25
23
15%
23°C
H 13
H
+
clovanemagnolol (26)
21
magnolol 25
(a) OH H H
4
O H 13
27
H
(PhO)2P(O)OH CH2Cl2
+
O bb-27
OH
+
H
25
X +
23°C
OH
(b)
Scheme 11.6 Acid-catalyzed rearrangement of caryophyllene α- and β-oxides (21 and 27): two-step biomimetic synthesis of (a) clovanemagnolol (26) and (b) caryolanemagnolol (29).
28
HO HO
10%
H caryolanemagnolol (29)
404
11 Biomimetic Rearrangements of Complex Terpenoids
(10–15%) of these transformations are balanced by the limited number of steps for this biomimetic route [two steps from caryophyllene (14)]. Better trapping yields were obtained when 4-bromophenol was used instead of 25, releasing a synthetic intermediate, but the two-step biomimetic synthesis makes this route particularly competitive. 11.3.3 Biomimetic Studies in the Triquinane Series
Isocaryophyllene (15) holds the αα conformation and underwent under acidic conditions (H2 SO4 , ether) a Wagner–Meerwein 1,2-shift of C10 from C9 onto C8 (Scheme 11.7a) [29]. Subsequent transannular electrophilic cyclization of 31 created the C5–C9 bond of the presilphiperfolane skeleton (32), which was neutralized as presilphiperfolene (33) in 35% global yield. Better control of the rearrangement in aqueous conditions was obtained when the process was performed on the tosylate 35, which originated from the caryophyllene β-oxide 21 (Scheme 11.7b) [30]. In this route, the methyl at C8 was missing after ozonolysis of the exocyclic double bond of 21. Furthermore, the endocyclic trans-olefin (34) was regenerated by the reduction of the epoxide. Stereochemical considerations of the rearrangement of 35 ruled out a concerted pathway involving departure of the tosylate group. Steric hindrance of the tosylate would favor a αβ-conformation and the solvolytic formation of a short-lived caryophyllenyl carbocation. Then, rapid rearrangement resulted in the formation of the norpresilphiperfolanol (36). According to the authors, although the methyl at C8 was missing in this route, the reaction provided a chemical precedent for a biogenetic connection between the trans caryophyllene series and the presilphiperfolanols. H
H2SO4 ether
10
8
9 8+
9
αα−30
15
9 +
5
5
35%
H (a)
10 +
31
−H+
32
H
H
33
H
Me OH
H O
65%
H (b)
1) O3 2) Zn-Cu AcOH
21
H
1) LiAlH4 2) nBuLi, TsCl 8
H O 34
OTs H 8
83%
H2O, acetone 75°C 61%
ab−35
H
H
H
8
36 H
Scheme 11.7 Biomimetic rearrangements of isocaryophyllene and caryophyllene β-oxide into presilphiperfolanes.
Presilphiperfolanol (37) can easily undergo ring contraction in the presence of H2 SO4 –SiO2 in benzene to generate silphiperfol-6-ene (39, Scheme 11.8) [31]. After the formation of the carbocation 18, contraction of the cyclohexane ring into 38 was followed by a methyl 1,2-shift. Subsequent return to neutrality by β-elimination of a proton afforded silphiperfolene (39), accompanied by α-terrecyclene (40) as
11.3 Biomimetic Rearrangements of Sesquiterpenes
OH H
H
H2SO4-SiO2 benzene 70°C
H
18
37
405
H
−H+
+
38
39 (79%) silphiperfolane skeleton
presilphiperfolane skeleton
40 (1%) terrecyclene skeleton
Scheme 11.8 Biomimetic rearrangement of presilphiperfolanol (37) into silphiperfolene (39) and the secondary product terrecyclene (40).
a minor product (1%). It was presumed to be formed by the Wagner–Meerwein rearrangement of an intermediary silphinyl cation (not shown). In another experiment, the formation of α-terrecyclene (40) was rationalized by the specific transformation of silphinyl mesylate 41 through double Wagner–Meerwein rearrangement (80% yield) (Scheme 11.9) [32]. Formolysis promoted the abstraction of the mesylate group from 41 to give cation 42. The migration of methine C7 from C11 to C1 afforded the bicyclo[3.3.1]nonane system 43. Then, migration of the methylene C5 from C4 to C11 gave the cation 44, achieving at the same time the complex transformation of 41 and biomimetic synthesis of α-terrecyclene (40). 9 7 10 OMs 11 1 4
H 6 5
41 silphinane skeleton
HCOOH HCOONa
80%
6 5
1
7 4 9
11
7 1
71 5
4
11
10
42
5 9
9
4 11
10 11 4
10
43
5
7 1
−H+
44
Scheme 11.9 Biomimetic rearrangement of the silphinyl mesylate (41) into terrecyclene (40).
11.3.4 Oxidative Rearrangements in the Silphinane Series: the Penifulvins
Penifulvins A–C (45–47) have been isolated from the fungus Penicillium griseofulvum [33]. Among these compounds, 45 showed significant insecticide activity against the fall armyworm Spodoptera frugiperda. The originality of penifulvins lies in a complex dioxafenestrane core in which four fused cycles share a common quaternary center (Scheme 11.10). The biosynthetic hypothesis has been secured by the co-occurrence of 12-hydroxy-silphinen-15-oic acid (48) in the fungus, which was extracted along with 45–47. Compound 48 would be the precursor of 50. In 2009, Gaich and Mulzer reported the first total synthesis of 45 [34a], which was followed in 2010 by those of 46 and 47 [34b]. The work features a biomimetic step consisting of the oxidative rearrangement of a silphinene intermediate (52), derived from the photochemical rearrangement of the aromatic compound 51, into the dioxafenestrane core. Ozonolysis of the olefin led to the dialdehyde intermediate 53,
40 terrecyclene skeleton
406
11 Biomimetic Rearrangements of Complex Terpenoids Biosynthetic proposal for penifulvins
O O
O O O
HO2C
O
O
OH O
[O]
R2 R1 R1=R2=H: penifulvin A (45) R1=OH, R2=H: penifulvin B (46) R1=H, R2=OH: penifulvin C (47)
HO H
46% (3 steps)
Me 51
[O]
H HO 12-hydroxy-silphinen15-oic acid (48)
1) hn 2) Li HO2C 3) IBX NaClO2
O
OH O
O O
O
H HO
HO
49
50
O O
O3 thiourea
H O O
OH O
H O O
O
PDC 82%
78%
H 52
H 53
54
penifulvin A (45)
Scheme 11.10 Biosynthetic proposal for penifulvins (box) and Mulzer’s biomimetic synthesis of penifulvin A (45).
which underwent domino cyclization of both oxygenated cycles. The lactol 54 was not isolated, being oxidized in the presence of PDC (pyridinium dichromate) to give the natural product 45. Using this key-transformation, penifulvin A was synthesized racemically in five steps from o-tolylacetic acid (14% overall yield), or enantioselectively in eight steps (8% overall yield). 11.3.5 Miscellaneous Sesquiterpene Rearrangements
Neopupukeananes (structure type 57) and trachyopsanes (structure type 58) are coexisting marine sesquiterpenes derived from amorphane (Scheme 11.11). According to Srikrishna’s biosynthetic proposal and synthetic work, the trachyopsanes would be derived from neopupukeananes via a Wagner–Meerwein shift leading to biomimetic rearrangement CSA (50 equiv) PhH O
O
2
O
O
2
O
87%
55
CN H
HO
TsCl pyridine
OHC HN H
Scheme 11.11
4 58 trachyopsane skeleton
H
EtOH
94%
92%
trachyopsane (62)
4 57 neopupukeanane skeleton OHC 1) (CH2SH)2, I2 HN 2) Raney Ni
56
61
Biomimetic synthesis of trachyopsane A (62).
59
H2SO4 AcOH TMSCN
H
89%
O 60
O 59
11.3 Biomimetic Rearrangements of Sesquiterpenes
407
C2–C4 bond formation from cation 57 [35]. The neopupukeanane substrate 56 was constructed in a very short and enantioselective sequence from (R)-carvone (55) [36]. Cation 57 was generated by treatment of 56 in the presence of 50 equiv of CSA (camphorsulfonic acid) in refluxing benzene. Rearrangement into the trachyopsane cation 58 was followed by β-elimination to give trachyops-2(14)-ene-7one (59) in 87% yield. Compared to other acidic systems [HCO2 H, TFA (trifluoroacetic acid), MsOH, p-toluenesulfonic acid (PTSA), and BF3 ·OEt2 /TFA or MsOH], CSA gave the best results for this rearrangement. The synthesis of 2-isocyanotrachyopsane (61) was completed by reduction of the ketone and installment of the isonitrile group. The panasinsane skeleton (63) is found in the roots of Panax ginseng (Scheme 11.12). From this plant were also isolated neoclovane and ginsenane (64) compounds, all showing substantial variations within the tricyclic structure. Collado studied the rearrangements of panasinsane derivatives prepared from caryophyllene oxide [37]. Novel structures were discovered and the 4,5,6-fused tricyclic skeleton of a panasinsane precursor was converted into ginsenol derivatives under tetracyanoethylene-catalyzed solvolysis. Diels-Alder HO
Wagner-Meerwein rearrangements
H H panasinsane (63)
Scheme 11.12
ginsenane (64)
α-methylene lactone
OH
H
Diels-Alder O O
O
H
O
H
H
HH O O O (+)-absinthin (65)
dienophile O
eudesmanolide skeleton (66)
O xanthipungolide (67)
Biomimetic synthesis of miscellaneous sesquiterpenes.
Diene-containing terpenoids can react with parent dienophile structures during biomimetic Diels–Alder reactions, affording dimers or heterodimers. The triterpene (+)-absinthin (65, Scheme 11.12) is actually a dimeric sesquiterpenoid isolated from Artemisia absinthium. It has been synthesized in nine steps and 19% overall yield from O-acetylisophotosantonic lactone, which is readily available from the photolysis of santonin and which holds a reactive cyclopentadiene moiety [38]. The biomimetic dimerization of the sesquiterpene precursor via regio- and stereospecific Diels–Alder reaction proceeded spontaneously in 72% yield upon leaving it neat at room temperature for ten days. Five additional steps were needed to obtain absinthin. Importantly, (+)-absinthin was previously found to afford the monomeric sesquiterpene artabsin through a retro-Diels–Alder reaction upon heating [39], suggesting a biosynthetic relationship between both monomeric and dimeric series. Natural α-methylene lactones are particularly interesting dienophiles that have been used in biomimetic Diels–Alder reactions. Plagiospirolides A and B are C35 terpenic natural products, the biosynthesis of which was supposed to arise from the Diels–Alder reaction between an eudesmanolide sesquiterpene (66) (containing an α-methylene lactone dienophile, Scheme 11.12) and a fusicoccane
408
11 Biomimetic Rearrangements of Complex Terpenoids
diterpene (containing a cyclopentadiene). The biomimetic Diels–Alder reaction has been reproduced experimentally by heating both precursors in benzene at 60 ◦ C [40]. Similar transformations of eudesmanolides have been reported although the required conditions could sometimes hardly be regarded as physiological [41]. Finally, the sesquiterpene xanthipungolide (67) (Scheme 11.12) was synthesized by irradiating an EtOH solution of an α-methylene lactone precursor found in the same plant and bearing a dienone moiety [42]. The reaction involved a tandem process starting with the electrocyclization of the dienone to reveal a diene that spontaneously underwent intramolecular [4 + 2] cyclization with the proximate α-methylene lactone dienophile.
11.4 Diterpene Rearrangements
Polycyclic diterpenes arise from the electrophilic cyclization of the linear geranylgeranyl diphosphate precursor 68. Important biosynthetic intermediates have been described (Scheme 11.13), among which are verticillene (70), casbene (71), and both enantiomers of copalyl diphosphate [(−)-72 shown, and (+)-72] leading to kaurenes (73) [7]. All are precursors of various structures of great interest. Carbon skeleton reorganization and oxygenation lead to important chemical diversity and have inspired many synthetic chemists to use biomimetic strategies. H
H OPP H
OPP geranylgeranyl phosphate (68)
verticillene (70) a
a H
H H
ent-kaurene (ent-73)
OPP H copalyl pyrophosphate ((-)-72)
Scheme 11.13
H b
H
b
H 69
casbene (71)
Examples of early chemical diversity in diterpene biosynthesis.
11.4.1 Dead End Products in the Biomimetic Synthesis of Antheridic Acid from Gibberellins
Antheridic acid (78, Scheme 11.14) is an antheridium-inducing factor (antheridiogen) isolated from the fern Anemia phyllitis [43]. Close similarities between 78 and fungal gibberellins, for example, gibberellins A7 (74) and A9 (75), suggested
6
15
1
H 7 CO H 2
OH
gibberelin A7 (74)
H CO H 2
CO
OH
gibberelic acid (79)
H CO H 2
CO
OH
H CO H 2
OH
O
H CO Me 2 83
CO
I O
gibberellenic acid (80)
HO HO2C
MOMO
30-40%
(NH2)2 H2O 120°C
HO2C 76
15
9 H 8
O CO
81
H CO Me 2
O
HO2C 77
10 98
15
O 10 steps
82
CO2Me
H
CO CO2H
HO O
H
CO
O
O Ac
antheridic acid (78)
HO
biomimetic AcO step
1, 2-shift
H CO H 2
CO
HO O
antheridic acid (78)
HO
OAc acid-catalyzed
1, 2-shift or C8-C15 bond cleavage, respectively
H CO2Me 84 cyclogibbane skeleton
MOMO
80%
KH
AcO HO2C
8 steps
[O]
or
OH
Scheme 11.14 Unsuccessful biomimetic strategy for synthesis towards antheridic acid (78) (a) and an alternative synthetic route (b).
(b)
HO
(a)
HO
4 18
gibberellin A7: R = OH (74) gibberellin A9: R1 = H (75)
R
10
H O
proposed biosynthetic hypothesis for antheridic acid
1) elimination H 1 O at C9-C10 13 9 2) epoxidation 2 10 8 14 19 CO 3 5 16 17 3) C9-C15 bonding 1
11 12
11.4 Diterpene Rearrangements 409
410
11 Biomimetic Rearrangements of Complex Terpenoids
a common biosynthetic relationship.3) A Wagner–Meerwein 1,2-shift induced by acid-catalyzed opening of an epoxide (76) was initially proposed by Nakanishi [43a]. This hypothesis also raised the prospect of achieving the biomimetic synthesis of 78 from readily available precursors derived from gibberellins. Mander undertook this task and prepared epoxide 81 from gibberellenic acid 80, which is accessible by treatment of gibberellic acid (79) in hot water (Scheme 11.14a) [45]. Unfortunately, the reactions of epoxide 81 with a range of Lewis acids did not produce any evidence of the desired rearrangement into the product 82 having the antheridic acid skeleton. In fact, the carbocyclic core remained unchanged and only products from lactonization or elimination were observed (not shown). Failure was attributed to the presence of the 17-methylene group in the D-ring, which would be responsible for the reduced migration aptitude of C15, and also to the proximity of the C19 carboxylic acid in the epoxide 81. A more secure strategy was finally elected to perform the synthesis of antheridic acid (78) from gibberellin A7 (74, Scheme 11.14b). The approach consisted in the construction of the cyclogibbane derivative 84 by means of intramolecular enolate alkylation. Then, a controlled fragmentation of the cyclopropane ring achieved construction of the tetracyclic core of 78, the synthesis of which was completed after additional steps. Several antheridiogens based on the 9,15-cycloderivative 77 have been isolated from natural sources, which make this approach particularly attractive by revealing biogenetic likeness [46]. However, the synthetic intermediates in this route are not biomimetic. 11.4.2 Biomimetic Synthesis of Marine Diterpenes from Pseudopterogorgia elisabethae
The gorgonian octocoral Pseudopterogorgia elisabethae is the source of interesting rearranged diterpenes, including colombiasin A (85) [47], elisabethin A–D (A: 86) [48, 49], elisabanolide (87) [48, 49], and elisapterosin A and B (B: 88) [49] (Scheme 11.15), which are biogenetically linked to amphilectane diterpenes. Many of them have shown significant antituberculosis and/or antitumor activities, but they are famous to organic chemists because of the important work undertaken using biomimetic approaches. Although colombiasin A and elisapterosin B have been successfully worked out, the case of elisabethin A remains unresolved [50]. According to Rodr´ıguez, the biosynthesis of the Pseudopterogorgia diterpenes would start with the formation of the serrulatane bicyclic skeleton (Scheme 11.15) [47–49]. The complex polycyclic systems of colombianes and elisapteranes would be formed by [4 + 2] and [5 + 2] cycloadditions from serrulatane, respectively, although electrophilic cyclizations and Wagner–Meerwein rearrangements have 3) Gibberellins are biogenetically derived
from the ent-kaurane skeleton ent-73 (Scheme 11.13), the rearrangements of which have been well described in the literature; for example, see Reference [44]. The biogenetic-like interconversions of the tetracyclic diterpene in the
beyerane–kaurane series (including atiserane, hibaane, trachylobane, and phyllocladane) were achieved using carbonium ion rearrangements very similar to those of bridged bicyclic monoterpenes.
O Me
Me
OH
Me
O O Me O O elisabanolide (87)
H
H
H H
O
Me OH
Me
Me
O
elisabethin A (86)
O
Me
OH
elisapterosin B (88)
Me
Me
H
Me O H
12
14
1
14 15
1
serrulatane
10
9
C9-C14
amphilectane
9 12
C1-C9
1, 2-shift
15
14
7 1 2
[5+2]
[4+2]
3
C2-C12
C10-C15
elisabethane
10
9
12
1 2
15
1
elisapterane
9 10
colombiane
9
Biosynthetic relationships in the Pseudopterogorgia diterpene series
Scheme 11.15 Representative examples of marine diterpenes from Pseudopterogorgia elisabethae and biosynthetic hypotheses [49] (atom numbering according to the elisabethane core).
Me Me
Me
H colombiasin A (85)
Me
H
Me O
11.4 Diterpene Rearrangements 411
11 Biomimetic Rearrangements of Complex Terpenoids
412
also been proposed. The elisabane skeleton would arise from oxidative cleavage of the quinonoid ring of elisapteranes. The first synthesis of colombiasin A (85) was reported in 2001 by Nicolaou [51]. The strategy featured two Diels–Alder reactions, the first one to build the bicyclic moiety of a serrulatane-type intermediate and the second to complete the construction of the colombiane skeleton in a biomimetic manner (Scheme 11.16). The bicyclic serrulatane intermediate 93 was build from the diene 89 and the quinone precursor 90 through intermolecular Diels–Alder cyclization while a Claisen rearrangement promoted by Pd(PPh3 )4 allowed for branching of the side chain on 92. The intramolecular Diels–Alder reaction was performed on sulfone 94 following cheletropic extrusion of SO2 to unmask the diene. Only the endo product was observed after 20 min of reaction at 180 ◦ C in toluene (sealed tube), affording the colombiane skeleton 95 in 89% yield. The cycloaddition did not work when performed directly on the free diene, although it was successfully carried out on 7-epi-93 for the synthesis of 7-epi-colombiasin A. After deoxygenation and deprotection, the total synthesis of colombiasin A (85) was completed. Although the work was initially done racemically, Nicolaou was able to perform an asymmetric Diels–Alder reaction between substrates 89 and 90, using the chiral Makami catalyst [(R)-BINOL-TiCl2 ]. This allowed the absolute configuration of the natural product to be determined. Retrosynthetic disconnexions Me O OH Diels-Alder Me biomimetic H Claisen Diels-Alder O Me Me H colombiasin A (85) 3 steps Me
O H
OMe
95
4. LiHMDS, then crotyl chloroformate (94%) O Diels-Alder
Me O 89
Me
OMe PhMe, 180°C 20 min (89%) HO H 7 Me cheletropic Me H extrusion Me of SO2 then [4+2] O2S giving the endo product only
90
Me
O OMe 1. SO2 (97%) 2. HNO3, AgO TBSO Me (85%)
H Me
O
Me
Me
1. EtOH, 25°C, 2h (83%) 2. K2CO3, MeI (83%) 3. TFA
+ TBSO
O
HO H Me 7
O
OMe Me
O
94
OMe
O
Me 91
Pd(PPh3)4 (58%)
OMe OMe
7
OMe
Me OMe
Me
OMe
13 steps O
H Me
Me
93
OMe
Me 7
OMe 92
biomimetic Diels-Alder
Scheme 11.16
Nicolaou’s synthesis of colombiasin A (85).
Kim and Rychnovsky developed a unified strategy for the asymmetric synthesis of (−)-colombiasin A (85) and elisapterosin B (88) [52], which supported Rodr´ıguez’s biosynthetic proposal. Here again, the cis-decalin serrulatane skeleton was synthesized by an intermolecular Diels–Alder reaction between the quinone 90 and the enantiomerically pure diene 97 made from the asymmetric alkylation of the propionamide (96) of Myer’s pseudoephedrine auxiliary (Scheme 11.17). The Diels–Alder reaction was carried out in the presence of LiClO4 in Et2 O, which provided the only successful conditions, giving the endo product 98 as the major stereoisomer. Finally, functionalization of the side chain led to the
11.4 Diterpene Rearrangements OAc stereocontrolled alkylation OH Me 4 steps N Me Me Me O TIPSO 96 97
AcO
O OMe
H
O
Me
90 O
H Me
LiClO4
H
Diels-Alder
98
AcO OMe Me
O
+
H Me
OTIPS 1.7:1
H
H
413
O OMe Me O OTIPS
99
11 steps Me O Me OH O Diels-Alder H
O
Me biomimetic [5+2] Me
Me H stereocontrolled alkylation elisapterosin B (88)
Scheme 11.17
OMe 1. 180°C, PhMe (83%) 2. AlCl3, PhNMe2 CH2Cl2 (73%) Me
BF3-OEt2 (25 equiv) CH2Cl2, −78°C
H Me
biomimetic [5+2]
O
biomimetic [4+ 2]
85
100
Rychnovsky’s synthesis of colombiasin A (85) and elisapterosin B (88).
diene moiety on the serrulatane skeleton (100) for the biomimetic intramolecular cycloaddition. When the serrulatane substrate 100 was submitted to thermal [4 + 2] cycloaddition (toluene, 180 ◦ C), and after an additional step of demethylation, (−)-colombiasin A (85) was obtained in 83% yield. The natural product was prepared in 17 steps and 3.9% overall yield from 96. When submitted to a large excess of BF3 ·OEt2 (25 equiv.) in CH2 Cl2 at −78 ◦ C, 100 underwent [5 + 2] cycloaddition to yield (−)-elisapterosin B (88) in 41% yield as a mixture of separable diastereoisomers. The proposed [5 + 2] cycloaddition had good precedent from Joseph-Nathan’s work on BF3 ·OEt2 -catalyzed [5 + 2] cycloadditions of naturally occurring sesquiterpenes [53].4) Elisapterosin B (88) was obtained in 16 steps and 2.6% overall yield from 96. Other studies on the biomimetic synthesis of Pseudopterogorgia diterpenes have been reported by others, differing mainly in the strategy used to synthesize the serrulatane precursor. Harrowven used the Moore rearrangement to make a dihydroquinone advanced intermediate, synthesizing 85 and 88 in 12 and 11 steps, respectively, from (−)-dihydrocarvone [55]. Jacobsen used chiral chromium complexes as catalysts to promote the intermolecular Diels–Alder reaction leading to the serrulatane core from quinone 90 [56]. Davies used the combined C–H 4) In his work published in 1987, Joseph
Nathan described the [5 + 2] cycloaddition of the sesquiterpene perezone (i) into αand β-pipitzol (ii and iii). The reaction was highly stereoselective in the presence of BF3 ·OEt2 , yielding a 9 : 1 mixture of both isomers. It was also strongly influenced by electronic factors. This reaction had been
described in 1885 under thermal conditions [54]. OH
OH O
BF3-OEt2 CH2Cl2 0°C
O
O
+
O
i
OH O
O
H ii
(9:1)
H iii
11 Biomimetic Rearrangements of Complex Terpenoids
414
activation/Cope rearrangement between a 2-diazoester and dihydronaphthalene derivatives, using a chiral dirhodium complex catalyst, to set up the side chain [57]. 11.4.3 Biomimetic Relationships among Furanocembranoids
The furanocembranoids and their rearranged derivatives are diterpenes isolated from marine sources, most of them from gorgonian corals [58]. Remarkable bioactivities have been associated with these structurally interesting compounds and, thus, they have been of great interest to the synthetic chemist community. Furanocembranoids feature a 14-membered carbocycle usually encompassed by furan and butenolide heterocycles associated with a high oxidation level at variable positions. Rearrangements of the furanocembrane skeleton and oxidative cleavages further increase the chemical diversity, as shown by the photochemical transformation of bipinnatin J (101) into the related pseudopterane (102) and gersolane (103, 104) skeletons by Rodr´ıguez (Scheme 11.18), respectively, formed from the photochemically allowed [σ 2s + π2s ] (or 1,3-sigmatropic rearrangement) and [σ 2a + π2a ] cycloadditions [59]. Both skeletons were formed by irradiation of 101 in acetonitrile. This experiment provided evidence for the biosynthetic connection of pseudopteranes and gersolanes to furanocembranes. More complex polycyclic skeletons have also been encountered, resulting from [4 + 2], [5 + 2], [2 + 2], or [4 + 3] cycloadditions, as exemplified by intricarene, bielschowskysin, and rameswaralide described below.
O
O
OH
OH
O
hn CH3CN 100% (120:1:6)
O bipinnatin J (101) furanocembrane skeleton
H O
+
H
OH +
OH O O
O O
kallolide A (102) pseudopterane skeleton from [σ 2s+π2s] cycloaddition
O
O
O O pinnatin A (103) pinnatin C (104) gersolane skeleton from [σ 2a+π 2a] cycloaddition
Scheme 11.18 Photochemical rearrangement of bipinnatin J (101) according to Rodr´ıguez [59].
Bipinnatin J (101) is a pivotal biosynthetic intermediate isolated from Pseudopterogorgia bipinnata [60] and was first synthesized independently by Trauner [61a], Rawal [62], and soon after by Pattenden [63]. In their work, Trauner [61b] and Pattenden figured out the biosynthetic origin of intricarene (111), a polycyclic compound issued from a transannular 1,3-dipolar cycloaddition (Scheme 11.19). In all cases, bipinnatin J was obtained by the macrocyclization of the linear precursor 105 under Nozaki–Hiyama conditions (CrCl2 /NiCl2 or CrCl2 ). Trauner reduced 101 into rubifolide (106), a putative biosynthetic precursor of the series, which in turn was oxidized into isoepilophodione B (107) in the presence of mCPBA (m-chloroperbenzoic acid) or alternatively by the selective and biomimetic addition
11.4 Diterpene Rearrangements
O
O
O
CrCl2
OH
Et3SiH CF3CO2H
Br O
O
O RO
O
mCPBA [61b] or VO(acac)2, tBuOOH [63] Ac2O
H
O O 108 (R=H)
Scheme 11.19
mCPBA
81% [61b] 30% [63] (2 steps)
109 (R=Ac)
O O
88% O
O
O bipinnatin J (101)
O 105
O
99%
415
O rubifolide (106)
O isoepilophodione B (107) coralloidolides (see scheme below)
TMP, DMSO 150°C (26%) [61b] or DBU, MeCN reflux (10%) [63]
+
O
O− H O
O 110
[5+2]
H
O
O
H O O intricarene (111)
Bipinnatin J (101) as a pivotal natural precursor of furanocembranoids.
of singlet oxygen to the furan moiety of 106 [61b]. In the same work, Trauner used mCPBA to oxidize 101 into the sensitive hydroxypyranone 108, the acetylation of which furnished the direct precursor of intricarene 109. Treatment of 109 with the hindered base 2,2,6,6-tetramethylpiperidine (TMP) in dimethyl sulfoxide (DMSO) at 150 ◦ C afforded of the oxidopyrylium species 110, which underwent transannular 1,3-dipolar cycloaddition to yield 111 in 26% yield [64]. According to the authors, temperatures in excess of 150 ◦ C cannot be deemed biomimetic. The requirement of such conditions thus opens up the possibility that an enzyme (possibly a monooxygenase) could mediate the reaction. Similarly, Pattenden oxidized bipinnatin J (101) to the hydroxypyranone 108 in the presence of VO(acac)2 and tBuOOH [63]. After acetylation to 109 and treatment under basic conditions [DBU (1,8-diazabicyclo[5.4.0]undec-7-ene), MeCN, reflux], the pyrone intermediate 110 was generated, the [5 + 2] cycloaddition of which gave intricarene 111 through an endo transition state, in 10% yield (not optimized). Rubifolide (106), synthesized by reduction of bipinnatin J (101), was the starting material used by Trauner for the biomimetic synthesis of coralloidolides A, B, C, and E (112–115, Scheme 11.20) [65]. The work used Nature’s strategy of oxidative diversification to obtain this family of diterpenes, which originated from an alcynocean coral. Chemoselective nucleophilic epoxidation at C11–C12 of 106 provided coralloidolide A (112) as a single diastereoisomer. Oxidative cleavage of the furan ring of 112 in the presence of mCPBA afforded coralloidolide E (113) with a similar degree of chemoselectivity. The reactivity of this compound under various reaction conditions was then explored. After extensive experimentation, the authors found that treatment of 113 with Sc(OTf)3 hydrate in dioxane led to clean conversion into the rearranged product coralloidolide B (114) in 63% yield. This biomimetic conversion would proceed in a stepwise manner, with the catalyst taking part both in hydration of the enedione and in the subsequent transannular attack of the epoxide at C11. Alternatively, coralloidolide C (115) was obtained by treatment of 113 in the presence of DBU in excess. The base catalyzed
11 Biomimetic Rearrangements of Complex Terpenoids
416
rubifolide (106)
H2O2 NaOH
11
95%
O
O O
mCPBA O
73%
63%
H
11
O
O coralloidolide A (112)
O
Sc(OTf)3 .x H2O
11 O
12
O
OH
3
7
O
O
OH O coralloidolide B (114)
O coralloidolide E (113) O 3
7
OH
DBU 20%
O
O O coralloidolide C (115)
Scheme 11.20 (112–115).
Conversion of rubifolide (106) into coralloidolides A, B, C, and E
the transannular aldol addition of C7 onto C3 to close the cyclopentenone ring. Remarkably, no protecting groups were used during these biomimetic syntheses. Bielschowskysin (117) is a complex furanocembranoid isolated by Rodr´ıguez from Pseudopterogorgia kallos [66], with a new skeleton that was initially supposed to originate from the verrillane core during biogenesis (Scheme 11.21). However, some authors have made the hypothesis that it could be formed photochemically by the [2 + 2] cycloaddition of a precursor derived from the oxygenation of bipinnatin Biosynthetic hypothesis for bielschowskysin Me Me OH Me H HO [O] [2+2] HO O H
Me OH
Me O
O
O O bipinnatin J (101)
O
O
hn
118a
O
O
O
HO
Me H
O
O O
H 119a
O
O
120a
120a:120b = 5:1 (50% from 118)
hn HO Me O
Me
O hn Me
Me H HO
118b Scheme 11.21
Me O
O
H
O O
Me
O
H
O
OH
O
Me Me H HO
O
H H O O OAc bielschowskysin (117)
O OAc 116
Me
HO Me
Me OH H
O
O
HO
MeH O
H 119b
Biomimetic studies toward bielschowskysin (117).
O
O
O
120b
11.4 Diterpene Rearrangements
J (101) [61a, 67, 68]. Doroh and Sulikowski provided the bases toward the total synthesis of 117 [69]. The stereoselective intramolecular [2 + 2] photocycloaddition of the 5-alkylidene-2(5H)-furanone 118a led to the tetracyclic intermediate 120a as the major cycloadduct (120a : 120b = 5 : 1). The substrate was irradiated in acetone, leading to isomerization into 118b and to the 1,4-biradical intermediates 119a and 119b. This furnished a model study validating the [2 + 2] cycloaddition biomimetic route toward 117. In a related work, Lear described the synthesis of the tricyclo[3.3.0]oxoheptane core of 117 by a [2 + 2] cycloaddition of an allene-butenolide substrate [68]. More recently, Pattenden and Winne proposed an intramolecular [4 + 3]cycloaddition approach to rameswaralide (122, Scheme 11.22) inspired by biosynthetic speculations [70]. Rameswaralide is a polycyclic furanocembrane-related diterpene isolated from the soft coral Sinularia dissecta [71]. According to Pattenden and Winne’s proposal, 122 would be derived from rubifolide (106) or a related compound. From this precursor, oxidative transformations would lead to furan opening to give an activated intermediate 121, the [4 + 3]-cycloaddition of which would result in rameswaralide (122). In the same report, an alternative route from plumarellide (124), another furanocembranoid, would involve a 1,2-rearrangement. This speculation was tested on the synthetic seco-furanobutenolide 126, which was submitted to acid-catalyzed rearrangement. Treatment of 126 in TFA, in the presence of small amounts of water at room temperature, soon afforded the rearranged polycyclic product 127 in quantitative yield. The exact sequence of this intriguing rearrangement could not be unraveled during this work and the authors postulated that either a [4 + 3]-cycloaddition or a [4 + 2]-cycloaddition followed by 6 → 7 ring expansion would occur. Yet another type of rearrangement was also discussed, involving a concerted [6 + 4]-cycloaddition. 11.4.4 Miscellaneous Diterpenes
Pallavicinolide A (128) is a rearranged seco-labdane diterpene isolated from a Japanese liverwort (Scheme 11.23) [72], and was recently synthesized in 32 steps from 2-methyl-1,3-cyclohexanedione [73]. The synthesis featured a Grob fragmentation performed on intermediate 129, the singlet oxygen oxidation of a furan ring, and an intramolecular Diels–Alder cycloaddition on substrate 130 as the biomimetic step. All retrosynthetic disconnections in this synthesis are reminiscent of the hypothetical biogenetic route for the natural product. In the proposal for the biosynthesis of the icetexane-based diterpene perovskone (136), it was suggested that the skeleton may arise from the addition of geranyl diphosphate to an icetexone precursor (Scheme 11.24). Majetich used this hypothesis for his total synthesis of this intricate natural product [74]. The tricyclic precursor 132 (rings A–C) was synthesized from 1-bromo-2,3,5-trimethoxybenzene. It was engaged in a Diels–Alder reaction, catalyzed by Eu(fod)3 at 45 ◦ C, with trans-α-ocimene (133), making ring D of intermediate 134. When pursuing the reaction at 110 ◦ C, ring E was closed by an ene-reaction while ring F was formed by
417
O
O
121
OH OH CO2Me O
[O]
O rubifolide (106)
O
O
[4 +3]
O 123
H H
OH O CO2Me H H OH
O
O
O rameswaralide (122)
O
HO
OH
OH
1,2-migration
[4 +2]
Biosynthetic proposal for rameswaralide
O
O
H
O
OH
CH2OH
O
O
HO
H 125
H
H H O
[O]
O
CO2Me
plumarellide (124)
H
H
OH
418
11 Biomimetic Rearrangements of Complex Terpenoids
O
O
O
Scheme 11.22
O
O
HO
HO
O
O O
O
OH
HO
O
O
O
O
O
HO
O
HO
CO2Et
H2O
CO2Et
CO2Et
O
O
OH
[6 + 4] route
O
O
CO2Et
O O
[4 + 2] route O
O
O
H
O
H
OH O H
H
OH
H
OH O H
ring expansion
OH CO2Me O [4 + 3] route
Biomimetic studies toward rameswaralide (122).
126
TFA H2O
CO2Et
O
H
CO2Et
OH
CO2Et
O
CO2Et
O
O
H
H
O 127
OH CO2Et
11.4 Diterpene Rearrangements 419
11 Biomimetic Rearrangements of Complex Terpenoids
420
Biosynthetic origin of pallavicinolide A O
16
1
20
13 16
11 9
1
8
15
15 2
[O]
7 18
19
labdane
17
H 15
O
O
4 steps
O
O O H
O H
H
H
OH
OMOM
H
OMOM
O
H
OMs Grob 129 fragmentation
Scheme 11.23
9
H
pallavicinolide A (128)
seco-labdane
10 steps
O
1
2
O
O
O
H
7
TMS
TBSO
8
9
O fragmentation
O H
13
13
17
H O
130
131
Biomimetic synthesis of pallavicinolide A (128). OH
O OH
Biosynthetic hypothesis for perovskone
H
133
O O
Eu(fod)3, 45 °C, 72h
O +
H O:
H
134 110 °C, 48h
OPP
O F
O
O A
G C E
O B
D
Amberlyst(R) 15-ion exchange resin
82%
O F
HO O
E
CH2Cl2 reflux, 30 min (90%)
perovskone (136)
Scheme 11.24
D
132 O
135
Biomimetic synthesis of perovskone (136).
subsequent addition of the hydroxyl on the remaining olefin. Final acid treatment of 135 completed the synthesis of perovskone (136) by closing ring G.
11.5 Triterpene Rearrangements
Triterpenes are derived from the cationic cyclization of squalene-based precursors. One of the most beautiful biomimetic transformations of this type, the pentacyclization of the linear polyene 137 towards sophoradiol (140, Scheme 11.25), was published by Johnson in 1994 [75]. The outcome of this cyclization relied on the presence of a fluorine atom on C13, which was use to control the regioselectivity of the cyclization, while being easily and regioselectively removed in the presence of
11.5 Triterpene Rearrangements
SiMe3 13
F
TFA 22
1
HO 137
F
22
H
31%
1) RuCl3, NaIO4 (88%) 2) SnCl4 (92%)
13
H
H H
421
O
H 138
O
H
139
DIBALH (87%)
22
H H HO
OH 4:1
H sophoradiol (140)
Scheme 11.25 Example of polyolefin cyclization: the biomimetic synthesis of sophoradiol (140).
tin tetrachloride. In the absence of fluorine on C13, a five-membered ring would have been formed. The biomimetic cyclization was performed in TFA, giving the pentacyclic intermediate 138 in 31% yield. After the cyclization process, return to neutrality was provided by TMS-elimination from the terminal propargylsilane. The polycyclization could also be performed in 50% yield in the presence of the Lewis acid SnCl4 , but was accompanied by fluorine elimination at C13, a side reaction yet incompatible with the planned oxidative cleavage of the exocyclic double bonds. Sophoradiol (140) was then obtained in three high-yielding steps. This impressive achievement, representing the first controlled polycyclization towards a pentacyclic system, stands as a magnificent conclusion to Johnson’s career. Important biologically relevant compounds bear, most frequently, the tetracyclic or the pentacyclic C30 framework of this series, which can even be truncated with possible loss of several carbons. There are numerous examples of polycyclizations of linear intermediates toward such structures, performed with synthetic or methodological purposes that cannot be described in this chapter. The reader is invited to consult excellent reviews published by others on this field [2]. Further biosynthetic rearrangements of the polycyclic C30 framework can lead to modified skeletons. For example, cyclopamine (148, Scheme 11.26) is a teratogenic steroidal alkaloid isolated from the lily corn Veratrum californicum [76]. It strongly inhibits hedgehog signaling during embryogenesis. The name of the compound originates from the Cyclops-like malformations of lambs, observed when pregnant sheep ate the plant. Cyclopamine consists of a 14(13 → 12)abeo-cholestane core (or C-nor D-homo) arising from a transposition in the cholestane skeleton [77]. The total synthesis of 148 was accomplished in 2009 by Giannis from a commercially available steroid (Scheme 11.26) [78]. It was presumed that the cationic rearrangement of 12β-hydroxy steroids into their C-nor-D-homo counterparts would be a reliable approach, as demonstrated by others [79]. Since 12β-hydroxy steroids are rare and a properly functionalized one resembling the cyclopamine ABCD skeleton was not available, it was envisioned that the hydroxyl could be introduced
H
H
O
HN
H 142
H
12
N
147
H
H
H BnO
8 steps
BnO
O
BsN
(48%)
[Cu(MeCN)4]PF6 acetone, O2 then AcOH, MeOH
O
146
H
H 14
12
H 13
143
H H
OH 12
+
O
BnO ratio 7:3
O
HO
145
H
O
O
H
O
O
BIOMIMETIC STEP
144
H
H 13 14
12
Tf2O, pyridine 0 50 °C 75%
BnO
4 steps
Semi-synthetic approach to cyclopamine (148); Bn-OPT: 2-benzyloxy-1-methylpyridinium triflate.
BnO
45% (3 steps)
1) N-sulfinylbenzene sulfonamide, then Raney Ni, H2 2) Raney Ni 3) sodium naphthalenide
BnO
(90% 2 steps)
2) picolyl amine pTsOH
O 1) Bn-OPT, MgO
cyclopamine (148)
H
141
Scheme 11.26
HO
HO
H
12
N
422
11 Biomimetic Rearrangements of Complex Terpenoids
11.5 Triterpene Rearrangements
423
at position 12 by selective C–H activation of dehydroepiandrosterone (141). This was undertaken by treating the 2-picolylimine 142 with tetrakis(acetonitrilo)copper(I) hexafluorophosphate and molecular oxygen in acetone, furnishing the 12β-hydroxy derivative 143. The reaction was performed with complete regioselectivity and stereoselectivity. After installment of the spiro lactone moiety of intermediate 144, the biomimetic Wagner–Meerwein rearrangement of the CD ring system was performed by exposure of the alcohol to trifluoromethanesulfonic anhydride in pyridine. The 14(13 → 12)abeo-cholestane core was isolated in nearly quantitative yields as a 3 : 7 mixture of regioisomers 145 and 146, respectively. Even though it would be necessary to convert the exocyclic double bond into the endocyclic isomer, the compound 146, as the dominant isomer, was engaged in the next steps. The fused piperidine cycle was installed in eight steps before double bond isomerization using an Alder-ene reaction in the presence of a sulfurizing agent. Final deprotections led to cyclopamine (148). The synthesis was undertaken in 20 steps and 1% overall yield from the steroid 141. At the same time several isomeric compounds were obtained for structure–activity studies. Betulin (149, R = CH2 OH), an abundant triterpene isolated from the birch tree, shares the same lupane pentacyclic framework as betulinic acid (150, R =CO2 H) and lupeol (153, R = CH3 ) (Scheme 11.27). The expansion of ring E through Wagner–Meerwein rearrangement of 149 was first reported in 1922 by Schulze and Pieroh by use of formic acid [80]. This biomimetic transformation afforded allobetulin (151) with the oleane framework, a compound that exhibits many biological H+
H
E R
H A
19
96%
H
HO
H R = CH2OH: betulin (149) R = CO2H: betulinic acid (150)
(a)
O
H
R = CH2OH Bi(OTf)3, DCM or TfOH, DCM
H H
HO
H
94%
H
H allobetulin (151) oleane skeleton
152
lupane skeleton
H
E H A HO (b)
O
H Bi(OTf)3, DCM (longer reaction times)
lupeol (153)
CF3SO3H (20 mM) CDCl3, 23 °C
mixture of rearrangement products: germanicol, d-amyrin, 18-epi-b-amyrin, taraxasterol, Ψ-taraxasterol (154), a-amyrin
Scheme 11.27 Biomimetic conversion of lupane into oleane skeletons: (a) Bi(III) salt catalysis by Salvador [27]; (b) Corey acid-catalyzed rearrangement of lupeol [82].
H H HO
H 154 (major product)
424
11 Biomimetic Rearrangements of Complex Terpenoids
activities (e.g., antifeedant, anti-inflammatory, cytotoxic). During the reaction, a Wagner–Meerwein 1,2-shift was accompanied by tandem formation of an ether bridge following a step-by-step carbocation mechanism. The reaction has been well studied and can be catalyzed by various acids. For example, montmorillonite K10 and kaolinite have been reported to convert quantitatively 149 into 151 [81]. More recently, the use of ‘‘eco-friendly’’ bismuth(III) salts as catalysts has been reported (Scheme 11.27a) [27]. In fact, the acid protons released from the hydrolysis of Bi(OTf )3 ·xH2 O were shown to be the true catalytic species during the formation of compounds 151 since no reaction occurred when a proton scavenger was added to the reaction mixture. For longer reaction times in the presence of the bismuth(III) salt, a second contraction at the A-ring of 151 provided the end product 152. Recently, Corey reported the total synthesis of lupeol (153) [82]. On this occasion, the one-step conversion of 153 into the naturally occurring pentacyclic triterpenes germanicol, δ-amyrin, 18-epi-β-amyrin, taraxasterol, ψ-taraxasterol (154), and α-amyrin via cationic intermediates was studied (Scheme 11.27b). Treating lupeol with a 20 mM solution of triflic acid in CDCl3 at 23 ◦ C resulted in a mixture of the six rearrangement products, monitored by 1 H NMR, the major product of the reaction being ψ-taraxasterol (154). In this rearrangement the main drawback was the non-controlled return to neutrality, generating a complex mixture of isomeric products.
11.6 Some Examples of the Biomimetic Synthesis of Meroterpenoids
Meroterpenoids are natural products that have their terpenic origin mixed in with another biosynthetic pathway. Compounds mixing terpene and polyketide features, including phenols, have provided many examples of biosynthetic rearrangements. In prenylated phenols it is common to observe covalent interactions between the terpene part and the phenol ring. This chemistry has been described elsewhere in this book. The biomimetic conversion of a geranylated pyrone (157) towards the skeleton of transtaganolides (155) and basidiolides (156), biologically active compounds isolated from the Mediterranean plant Thapsia garganica, has been reported recently (Scheme 11.28) [83]. This involved a sequence of Ireland–Claisen rearrangement of 157 into 158 and an intramolecular Diels–Alder reaction into 159, following the biosynthetic disconnections. Erinacine E (165) is a rearranged diterpene glycoside, isolated from the fungus Hericium erinaceum, with neuritogenic activities [84]. Its biosynthesis would result from a double C–C linkage between the glycoside part and the cyathane diterpene nucleus of an intermediate derived from erinacine P. The cyathane glycoside derivative 160, analogous to erinacine P, has been synthesized by Nakada (Scheme 11.29) [85]. After Swern oxidation into 161, a conjugate addition–elimination occurred to make the C2 –C13 bond of compound 162, the structure of which was close to the natural products striatals A–D. Subsequent treatment with DBU installed the
11.7 Conclusion
425
Biosynthetic disconnections for transtaganolides and basidiolides OMe OMe O 18 18 O 18 O O or O 19 O Ireland-Claisen 19 19 9 O 9 rearrangement 9 2
MeO O
1 2 10
2
8
O
O O 157
O
1 10
8
O
5 Diels-Alder R reaction R = Me: transtaganolides (155) R = CO2Me: basidiolides (156)
I
1 10
BTMSA Et3N (0.1 equiv) toluene 84%
O
I
O
O O 158
OH
8
O O
5
NaHCO3 water, 50 °C 3 days
O
5
I
O
CO2H
61%
159
Scheme 11.28 Johansson’s biomimetic study towards transtaganolides (155) and basidiolides (156) (BTMSA: N,O-bistrimethylsilylacetamide).
C4 –C15 bond in the strained hexacyclic intermediate 164, and deprotection led to erinacine E (165). The biomimetic Cope-rearrangement of globiferin (167), isolated from the antimalarial root extract of Cordia globifera, led to cordiachrome C (166) (Scheme 11.30) [86]. The co-occurrence of both metabolites in the same plant provided substantial information on the biogenetic relationship between the natural products. Although harsh conditions were employed for the rearrangement of 167, the racemic character of cordiachrome C (166) has suggested a non-enzymatic biosynthetic pathway. The asymmetric biomimetic synthesis of (−)-longithorone A (172), a cytotoxic marine natural product, featured a double Diels–Alder reaction of properly functionalized ansa-farnesylhydroquinone derivatives 168 and 169 (Scheme 11.31) [87]. The stereoselective intermolecular cycloaddition of both atropoisomeric substrates resulted in heterodimerization to give 170. After activation as the bis(quinone) 171, the intramolecular cycloaddition afforded the optically pure natural product 172. This work provided a unique example of chirality transfer in complex molecule synthesis involving chiral atropoisomers.
11.7 Conclusion
This chapter has illustrated how complex terpenoid compounds can be obtained using Nature’s biosynthetic strategy. Striking examples were discussed, showing the efficiency of this strategy, sometimes under conditions having physiological likeness. Some reactions occurred spontaneously, but most of them were undertaken with the use of a physical or chemical activation. The Wagner–Meerwein
Scheme 11.29
DBU
H
13
O
OH
4' OH
13
163
H
O
2'
4' OBz
15
O
−O OTES
O
1) Bz2O + 2) H 3) DMSO (COCl)2 NEt3
H
H
13
H O O
162
H
13
2'
4' OBz
15
CHO
O OTES
O
1) benzyl migration 2) NaBH4 3) K2CO3, MeOH
H
O
5) Me4NBH(OAc)3
OR 4' O O Ph 15 4) IBX, DMSO O−
2'
164 (R=TES)
H
O
O
CHO 15
OBz
13
2'
4' OBz
OTES
−
161
O
O
Nakada’s biomimetic synthesis of erinacine E (165).
H
160
15
OTES OTES
2'
OBz
O
H
13
H
15
OH
OH 4' OH
O OH 2'
erinacine E (165)
H
O
426
11 Biomimetic Rearrangements of Complex Terpenoids
11.7 Conclusion O
O
biosynthetic relationship
H O cordiachrome C (166)
Scheme 11.30
xylene, 130 °C, 2h
O globiferin (167)
85%
Biomimetic conversion of globiferin (167) into cordiachrome C (166).
MeO
MeO OTBS
CHO
OTBS
Me2AlCl 70% d.r = 1:1.4
TBSO
OHC
OMe
168
TBSO OMe
170
169
1) TBAF 2) PHIO 90% (2 steps)
O O
OH
H OHC
O O O
(−)-longithorone A (172)
Scheme 11.31
OHC
O O
171
Biomimetic synthesis of longithorone A (172).
rearrangement, which was discovered more than a century ago, has been used extensively in terpene biomimetic synthesis, using weak to strong acid catalysis. It is also the key reaction for diversification in terpene biogenesis. Furthermore, many biomimetic cycloadditions were undertaken by photochemical or thermal activation. The photochemical conditions are present as sunlight activation in Nature and could proceed without the involvement of an enzyme. The thermal conditions, which can be very harsh in the laboratory to reach the proper activation barrier for the cycloaddition or the rearrangement, usually do not occur in Nature. In such cases the activation barrier might be overcome by enzymes during biosynthesis, with somewhat better stereoselectivity for chiral natural products. For the most strained reactant systems of biosynthetic hypotheses, quantum chemical calculations have provided theoretical catalysts called theozymes [88]. This work predicts the catalytic system to be used in a reaction, with important applications in biological systems. Salvileucalin B (175) is a polycyclic rearranged diterpene related to the neoclerodane structure isolated from Salvia leucantha (Scheme 11.32) [89]. The strained tricyclo[3.2.1.02,7 ]octane moiety of this natural product caused chemists to question its biosynthetic origin. Fortunately, it was isolated at the same time as the biogenetically correlated salvileucalin A (173), so
427
428
11 Biomimetic Rearrangements of Complex Terpenoids
O
O O
O O salvileucalin A (173)
O
‡
O
O
O O
Scheme 11.32
O
O
O O
[O]
O
O 174
O O ‡ 174
O
O
salvileucalin B (175)
Salvileucalin A (173), a biosynthetic precursor of salvileucalin B (175).
that a biosynthetic scheme could be postulated. Quantum chemical calculations have revealed how functional group arrays present in a known enzyme active site could accelerate the intramolecular Diels–Alder reaction proposed to occur during the biosynthesis of 175 [90]. Several theozymes were screened to find that, after the oxidation of 173 to obtain the activated compound 174, the cycloaddition could be promoted by selective binding of the transition state 174‡ to the active site of a catalytic antibody species. The theozyme would lower the activation barrier of the cycloaddition from circa 25 to 20 kcal mol−1 , a value that is at the high end of the range of typical barriers for enzyme-catalyzed reactions. The work also pointed out that a biomimetic total synthesis of 175 might benefit from performing the Diels–Alder reaction prior to installing the ‘‘bottom’’ lactone. It was shown that an organocatalyst (thiourea) or a Lewis acid (AlMe2 Cl) could promote this transformation. This synthesis has yet to be performed.
References 1. Ourisson, G. (1964) Proc. Chem. Soc., 2.
3. 4. 5.
6.
274–282. For example, see the following early or late reviews: (a) Johnson, W.S. (1976) Bioorg. Chem., 5, 51–98; (b) Yoder, R.A. and Johnston, J.N. (2005) Chem. Rev., 105, 4730–4756. Kindt, H.H. (1802) Trommsdorff J. Pharm., 11, 132. Wagner, G. (1899) J. Russ. Phys. Chem. Soc., 31, 680. (a) Wagner, G. and Brickner, W. (1899) Chem. Ber., 32, 2302–2325; (b) Wagner, G. and Slawinski, K. (1899) Chem. Ber., 32, 2064–2086. Banthorpe, D.V., Charlwood, B.V., and Francis, M.J.O. (1972) Chem. Rev., 72, 115–155.
7. Dewick, P.M. (2002) Medicinal Natural
8. 9.
10. 11.
Products: A Biosynthetic Approach, 2nd edn, John Wiley & Sons, Ltd, Chichester. Birladeanu, L. (2000) J. Chem. Ed., 77, 858–863. (a) Meerwein, H. (1914) Liebigs Ann. Chem., 405, 129–175; (b) Meerwein, H. and van Emster, K. (1920) Chem. Ber., 53, 1815–1829; (c) Meerwein, H. and van Emster, K. (1922) Chem. Ber., 55, 2500–2528. Meerwein, H. (1913) Liebigs Ann. Chem., 396, 200–225. Swift, K.A.D. (2001) in Fine Chemicals Through Heterogeneous Catalysis (eds R.A. Sheldon and H. van Bekkum), Wiley-VCH Verlag GmbH, Weinheim, pp. 242–246.
References 12. Gscheidmeier, M. and H¨aberlein, H. 13. 14.
15.
16.
17. 18.
19.
20.
21.
22.
23.
(1998) US Patent US005826202A. Sorensen, T.S. (1976) Acc. Chem. Res., 9, 257–265. (a) Huang, E., Ranganayakulu, K., and Sorensen, T.S. (1972) J. Am. Chem. Soc., 94, 1779–1780; (b) Huang, E., Ranganayakulu, K., and Sorensen, T.S. (1972) J. Am. Chem. Soc., 94, 1780–1782; (c) Haseltine, R., Huang, E., Ranganayakulu, K., Sorensen, T.S., and Wong, N. (1975) Can. J. Chem., 53, 1876–1890; (d) Haseltine, R., Ranganayakulu, K., Wong, N., and Sorensen, T.S. (1975) Can. J. Chem., 53, 1901–1914. Olah, G.A., de Member, J.R., Lui, C.Y., and Porter, R.D. (1971) J. Am. Chem. Soc., 93, 1442–1446. (a) Johnson, C.K. and Collins, C.J. (1974) J. Am. Chem. Soc., 96, 2514–2523; (b) Collins, C.J., Johnson, C.K., and Raaen, V.F. (1974) J. Am. Chem. Soc., 96, 2524–2531. Allemann, R.K. (2008) Pure Appl. Chem., 80, 1791–1798. O’Maille, P.E., Malone, A., Dellas, N., Hess, B.A., Smentek, L., Sheehan, I., Greenhagen, B.T., Chappell, J., Manning, G., and Noel, J.P. Jr. (2008) Nat. Chem. Biol., 4, 617–623. For an acid-catalyzed ring contraction of longifolene into isolongifolene, see: Ranganathan, R., Nayak, U.R., Santhanakrishnan, T.S., and Dev, S. (1970) Tetrahedron, 26, 621–630. Bohlmann, F., Zdero, C., Jakupovic, J., Robinson, H., and King, R.M. (1981) Phytochemistry, 20, 2239–2244. Collado, I.G., Hanson, J.R., and Macias-Sanchez, J. (1998) Nat. Prod. Rep., 15, 187–204. (a) Shirahama, H., Osawa, E., Chhabra, B.R., Shimokawa, T., Yokono, T., Kanaiwa, T., Amiya, T., and Matsumoto, T. (1981) Tetrahedron Lett., 22, 1527–1528; (b) Guella, G., Chiasera, G., N’Diaye, I., and Pietra, F. (1994) Helv. Chim. Acta, 77, 1203–1221. Gatilova, V.P., Korchagina, D.V., Rybalova, T.V., Gatilov, Yu.V., Dubovenko, Zh.V., and Barkhash, V.A. (1989) Zh. Org. Khim., 25, 320–332.
24. Fitjer, L., Malich, A., Paschke, C., Kluge,
25.
26.
27.
28.
29.
30. 31.
32.
33.
34.
35.
36. 37.
38.
39.
S., Gerke, R., Rissom, B., Weiser, J., and Noltemeyer, M. (1995) J. Am. Chem. Soc., 117, 9180–9189. Olhoff, G., Uhde, G., and Schulte-Elte, K.H. (1967) Helv. Chim. Acta, 50, 561–570. Collado, I.J., Hanson, J.R., Hernandez-Galan, R., Hitchcock, P.B., Macias-Sanchez, A.J., and Racero, J.C. (1998) Tetrahedron, 54, 1615–1626. Salvador, J.A.R., Pinto, R.M.A., Santos, R.C., Le Roux, C., Matos Beja, A., and Paix˜ao, J.A. (2009) Org. Biomol. Chem., 7, 508–517. Cheng, X., Harzdorf, N.L., Shaw, T., and Siegel, D. (2010) Org. Lett., 12, 1304–1307. Cameron, A.F., Hannaway, C., Roberts, J.S., and Robertson, J.M. (1970) Chem. Commun., 248–249. Shankar, S. and Coates, R.M. (1998) J. Org. Chem., 63, 9177–9182. Coates, R.M., Ho, J.Z., Klobus, M., and Wilson, S.R. (1996) J. Am. Chem. Soc., 118, 9249–9254. Coates, R.M., Ho, J.Z., Klobus, M., and Zhu, L. (1998) J. Org. Chem., 63, 9166–9176. (a) Shim, H.S., Swenson, D.C., Gloer, J.B., Dowd, P.F., and Wicklow, D.T. (2006) Org. Lett., 8, 1225–1228; (b) Shim, H.S., Gloer, J.B., and Wicklow, D.T. (2006) J. Nat. Prod., 69, 1601–1605. (a) Gaich, T. and Mulzer, J. (2009) J. Am. Chem. Soc., 131, 452–453; (b) Gaich, T. and Mulzer, J. (2010) Org. Lett., 12, 272–275. Srikrishna, A., Ravi, G., and Venkata Subbaiah, D.R.C. (2009) Synlett, 32–34. Srikrishna, A. and Gharpure, S.J. (1999) Tetrahedron Lett., 40, 1035–1038. Amigo, C.F.D., Collado, I.G., Hanson, J.R., Hern´andez-Gal´an, R., Hitchcock, P.B., Mac´ıas-S´anchez, A.J., and Mobbs, D.J. (2001) J. Org. Chem., 66, 4327–4332. Zhang, W., Luo, S., Fang, F., Chen, Q., Hu, H., Jia, X., and Zhai, H. (2005) J. Am. Chem. Soc., 127, 18–19. Vok´acˇ , K., Samek, Z., Herout, V., and ˇ Sorm, F. (1968) Tetrahedron Lett., 9, 3855–3857.
429
430
11 Biomimetic Rearrangements of Complex Terpenoids 40. Kato, N., Wu, X., Nishikawa, H.,
41. 42.
43.
44.
45.
46.
47. 48.
49.
50.
51.
52.
Nakanishi, K., and Takeshita, H. (1994) J. Chem. Soc. Perkin Trans. 1, 1047–1053. Matusch, R. and H¨aberlein, H. (1987) Liebigs Ann. Chem., 455–457. Ahmed, A.A., Jakupovic, J., Bohlmann, F., Regaila, H.A., and Ahmed, A.M. (1990) Phytochemistry, 29, 2211–2215. (a) Nakanishi, K., Endo, M., N¨af, U., and Johnson, L.F. (1971) J. Am. Chem. Soc., 93, 5579–5581; (b) Zanno, P.R., Endo, M., Nakanishi, K., N¨af, U., and Stein, C. (1972) Naturwissenschaften, 59, 512. Coates, R.M. and Bertram, E.F. (1969) J. Chem. Soc. D, 797–798; (b) Coates, R.M. and Bertram, E.F. (1971) J. Org. Chem., 36, 3722–3729; (c) Coates, R.M. and Kang, H.Y. (1987) J. Chem. Soc., Chem. Commun., 232–233; (d) McGrindle, R. and Roy, R.G. (1971) J. Chem. Soc. C, 1018–1020. (a) Furber, M. and Mander, L.N. (1987) J. Am. Chem. Soc., 109, 6389–6396; (b) Furber, M. and Mander, L.N. (1988) Tetrahedron Lett., 29, 3339–3342. Yamauchi, T., Oyama, N., Yamane, H., Murofushi, N., Schraudolf, H., Pour, M., Furber, M., and Mander, L.N. (1996) Plant Physiol., 111, 741–745. Rodr´ıguez, A.D. and Ram´ırez, C. (2000) Org. Lett., 2, 507–510. Rodr´ıguez, A.D., Gonz´alez, E., and Huang, S.D. (1998) J. Org. Chem., 63, 7083–7091. Rodr´ıguez, A.D., Ram´ırez, C., Rodr´ıguez, I.I., and Barnes, C.L. (2000) J. Org. Chem., 65, 1390–1398. For a non-biomimetic total synthesis of elisabethin A, see: (a) Heckrodt, T.J. and Mulzer, J. (2003) J. Am. Chem. Soc., 125, 4680–4681; for a critical highlight on the total synthesis of elisabethin A, see: (b) Zanoni, G. and Franzini, M. (2004) Angew. Chem. Int. Ed., 43, 4837–4841. (a) Nicolaou, K.C., Vassilikogiannakis, G., M¨agerlein, W., and Kranich, R. (2001) Angew. Chem. Int. Ed., 40, 2482–2486; (b) Nicolaou, K.C., Vassilikogiannakis, G., M¨agerlein, W., and Kranich, R. (2001) Chem. Eur. J., 7, 5359–5371. Kim, A.I. and Rychnovsky, S.D. (2003) Angew. Chem. Int. Ed., 42, 1267–1270.
53. Joseph-Nathan, P., Garibay, M.E., and
54. 55.
56.
57.
58. 59.
60. 61.
62. 63. 64.
65.
66.
67. 68. 69. 70. 71.
Santillan, R.L. (1987) J. Org. Chem., 52, 759–563. Ansch¨utz, R. and Leather, W. (1885) Chem. Ber., 18, 715–717. Harrowven, D.C., Pascoe, D.D., Demurtas, D., and Bourne, H.O. (2005) Angew. Chem. Int. Ed., 44, 1221–1222. Boezio, A.A., Jarvo, E.R., Lawrence, B.M., and Jacobsen, E.N. (2005) Angew. Chem. Int. Ed., 44, 6046–6050. Davies, H.M.L., Dai, X., and Long, M.S. (2006) J. Am. Chem. Soc., 128, 2485–2490. Roethle, P.A. and Trauner, D. (2008) J. Nat. Prod., 25, 298–317. (a) Rodr´ıguez, A.D. and Shi, J.-G. (1998) J. Org. Chem., 63, 420–421; (b) Rodr´ıguez, A.D., Shi, J.-G., and Huang, S.D. (1998) J. Org. Chem., 63, 4425–4432. Rodr´ıguez, A.D. and Shi, J.-G. (1998) J. Org. Chem., 63, 420–421. (a) Roethle, P.A. and Trauner, D. (2006) Org. Lett., 8, 345–347; (b) Roethle, P.A., Hernandez, P.T., and Trauner, D. (2006) Org. Lett., 8, 5901–5904. Huang, Q. and Rawal, G.H. (2006) Org. Lett., 8, 543–545. Tang, B., Bray, C.D., and Pattenden, G. (2006) Tetrahedron Lett., 47, 6401–6404. For a recent review on cycloadditions involving oxidopyrylium species, see: Singh, V., Krishna Vikrant, U.M., and Trivedi, G.K. (2008) Tetrahedron, 64, 3405–3428. Kimbrough, T., Roethle, P.A., Mayer, P., and Trauner, D. (2010) Angew. Chem. Int. Ed., 49, 2919–2621. Marrero, J., Rodr´ıguez, A.D., Baran, P., Raptis, R.G., S´anchez, J.A., Ortega-Barria, E., and Capson, T.L. (2004) Org. Lett., 6, 1661–1664. Bray, C.D. and Pattenden, G. (2006) Tetrahedron Lett., 47, 3937–3939. Miao, R., Gramani, S., and Lear, M.J. (2009) Tetrahedron Lett., 50, 1761–1733. Doroh, B. and Sulikowski, G.A. (2006) Org. Lett., 8, 903–906. Pattenden, G. and Winne, J.M. (2009) Tetrahedron Lett., 50, 7310–7313. Ramesh, P., Reddy, N.S., Venkateswarlu, Y., Reddy, M.V.R., and Faulkner, D.J. (1998) Tetrahedron Lett., 39, 8217–8220.
References 72. Toyota, M., Sata, T., and Asakawa, Y. 73.
74. 75. 76. 77.
78.
79.
80. 81.
(1998) Chem. Pharm. Bull., 46, 178–180. (a) Dong, J.-Q. and Wong, H.N.C. (2009) Angew. Chem. Int. Ed., 48, 2351–2354; (b) Dong, J.-Q. and Wong, H.N.C. (2009) Angew. Chem., 121, 2387–2390. Majetich, G. and Zhang, Y. (1994) J. Am. Chem. Soc., 116, 4979–4980. Fish, P.V. and Johnson, W.S. (1994) J. Org. Chem., 59, 2324–2335. Keeler, R.F. (1968) Phytochemistry, 7, 303–306. Heretsch, P., Tzagkraroulaki, L., and Giannis, A. (2010) Angew. Chem. Int. Ed., 49, 3418–3427. Giannis, A., Heretsch, P., Sarli, V., and St¨oβel, A. (2009) Angew. Chem. Int. Ed., 48, 7911–7914. Hirschmann, R., Snoddy, C.S., and Wendler, N.L. (1952) J. Am. Chem. Soc., 74, 2693–2694. Schulze, H. and Pieroh, K. (1922) Chem. Ber., 55, 2332–2346. Li, T.-S., Wang, J.-X., and Zheng, X.-J. (1998) J. Chem. Soc., Perkin Trans. 1, 3957–3965.
82. Surendra, K. and Corey, E.J. (2009)
J. Am. Chem. Soc., 131, 13928–13929. 83. Larsson, R., Sterner, O., and Johansson,
M. (2009) Org. Lett., 11, 657–660. 84. Kawagishi, H., Shimada, A., Hosokawa,
85. 86.
87.
88.
89.
90.
S., Mori, H., Sakamoto, H., Ishiguro, Y., Sakemi, S., Bordner, J., Kojima, N., and Furukawa, S. (1996) Tetrahedron Lett., 37, 7399–7402. Watanabe, H. and Nakada, M. (2008) J. Am. Chem. Soc., 130, 1150–1151. Dettrakul, S., Surerum, S., Rajviroongit, S., and Kittakoop, P. (2009) J. Nat. Prod., 72, 861–865. Layton, M.E., Morales, C.A., and Shair, M.D. (2002) J. Am. Chem. Soc., 124, 773–775. Tantillo, D.J., Chen, J., and Houk, K.N. (1998) Curr. Opin. Chem. Biol., 2, 743–750. Aoyagi, Y., Yamazaki, A., Nakatsugawa, C., Fukaya, H., Takeya, K., Kawauchi, S., and Izumi, H. (2008) Org. Lett., 10, 4429–4432. Tantillo, D.J. (2010) Org. Lett., 12, 1164–1167.
431
433
12 Polyprenylated Phloroglucinols and Xanthones Marianna Dakanali and Emmanuel A. Theodorakis
12.1 Introduction
Guttiferae (Clusiaceae) is a family of plants that includes more than 37 genera and 1600 species [1]. Characteristic to this family is its large variation in plant morphology, which makes it an important group of plants for the study of floral diversification and evolutionary plasticity. Although mainly confined to the tropical areas, this family also includes the genus Hypericum, a plant that is found widely around the Mediterranean area. Several plants of the Guttiferae family have a rich history in ethnomedicine for their broad-spectrum antibacterial and healing properties. For instance, the antibacterial and antidepressant activities of Hypericum perforatum (St. John’s wort) have been noted in traditional European medicine. In fact, Hypericum extracts have been tested in various clinical trials and are currently used in certain countries for the treatment of depressive, anxiety, and sleep disorders [2–11]. On the other hand, members of the Garcinia genus of tropical trees have considerable value as sources of medicines, pigments, foodstuffs, and lumber [12, 13]. Chemically, the Clusiaceae family of plants constitutes a rich source of polyprenylated acylphloroglucinols and xanthones. Both chemical classes have generated substantial interest due to their fascinating chemical structures and potent bioactivities [10]. This chapter summarizes the chemical classification, biosynthesis, and synthetic approaches toward these compounds.
12.2 Polycyclic Polyprenylated Phloroglucinols 12.2.1 Introduction and Chemical Classification
The chemical structures of all known polycyclic polyprenylated acylphloroglucinols (PPAPs) can be classified into three types that are related to the biosynthesis Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
434
12 Polyprenylated Phloroglucinols and Xanthones R1 O
R1 HO
5
A
9
R1
R2
7 8
R1
B 3
O
R3
R5
O
O
1
R2
7
5
9
HO
R1
1
R1
R1
O
HO
R3
5
8
6 1
3
O
R4
R1
O
HO
R3
Type B
Type A
7
O
3
O
R1
R2
O
5
9
O
1
3
O
R3
R1 Type C
R1 = Me, C5H9, or C10H17 R2 = H or prenyl R3 = i -Pr, i -Bu, s -Bu, Ph, 3-(OH)C6H4, or 3,4-(OH)2C6H3 R4 = Me, R5 = OH or R4, R5 = CH2CHR6 R6 = H, C(CH3) = CH2, or C(CH3)2OH
Figure 12.1
Classification of polycyclic polyprenylated acylphloroglucinols (PPAPs).
proposal (Scheme 12.2 below). Types A, B(I), and C PPAPs are distinguished by the presence of a highly oxygenated bicyclo[3.3.1]nonane-2,4,9-trione motif, while type B(II) PPAPs contain a bicyclo[3.2.1]octane-2,4,8-trione carbon framework (Figure 12.1). In all types, the bicyclic motif is further decorated with prenyl, geranyl, and related side chains. Certain family members contain additional rings, formed by cyclizations between the β-diketone and an alkene, leading to adamantanes, pyrano-fused, or other cyclic substructures. Type A PPAPs have an acyl-substituent at C1 adjacent to a quaternary C8 carbon, while the type B compounds have the acyl-substituent at C3. The more rare type C PPAPs have the acyl group at C1 but the quaternary carbon is located at the distant C6 (Figure 12.1) [14]. For the purpose of this chapter, the carbon numbering of these molecules is based on the nemorosone numbering [14]. Figure 12.2 shows representative members of the PPAPs. Hyperforin (1), one of the bioactive ingredients of Hypericum perforatum [15, 16], belongs to the type A PPAPs and is also known for its antibacterial [17] and anticancer properties [9, 18]. Another type A phloroglucinol is garsubellin A (2), a natural product noted for its activities against neurodegenerative diseases. In fact, recent studies have shown that garsubellin A induces biosynthesis of acetylcholine, a neurotransmitter that at low concentrations can lead to Alzheimer’s disease [19]. Nemorosone (3) exhibits antimicrobial [20, 21], cytotoxic, and antioxidant activities [20], while, clusianone (4), a type B(I) PPAP, is known for its anti-HIV activity [22]. The type B(II) enaimeone A (5) was isolated from Hypericum papuanum, the leaves of which are used in the traditional medicine of Papua New Guinea for treating sores [23]. Garcinielliptone M (6), a type C PPAP, isolated from Garcinia subelliptica, shows potential anti-inflammatory activity [24]. 12.2.2 Biosynthesis of PPAPs
PPAPs derive biosynthetically from the less complex monocyclic polyprenylated acylphloroglucinols (MPAPs), a class of natural products isolated from plants of the
12.2 Polycyclic Polyprenylated Phloroglucinols
O
O
O
O
OH
HO
hyperforin (1) Type A
O
O
O
O
O
garsubellin A (2) Type A
O
OH
nemorosone (3) Type A
OH 7
O
O
O
O
OH
OH O
O
clusianone 7-epi-clusianone (4) Type B(I)
Figure 12.2
O
O
O OH
enaimeone A (5) Type B(II)
garcinielliptone M (6) Type C
Representative members of the PPAP family. OH
O
OH
O
R HO
O
R HO
O
HO
a-acids (7)
Figure 12.3
O
R = i Pr or CH2i Pr or n Bu or, CH2i Bu etc
b-acids (8)
Monocyclic polyprenylated acylphloroglucinols (MPAPs) from Humulus lupulus.
Myrtaceae and Cannabinaceae families [10]. There are two main classes of MPAPs: the diprenylated α-acids (7) and triprenylated β-acids (8) (Figure 12.3). α-Acids are responsible for the flavor and bitter taste of beer [25] while β-acids show, among other properties, free radical scavenger activity [7, 26]. Labeling and enzymological experiments have provided evidence that bitter acids are biosynthesized via condensation of three malonyl-CoA and one acyl-CoA, such as isobutyryl-CoA (9) (Scheme 12.1) [27–30]. The intermediate polyketide 10 can then cyclize via an intramolecular Dieckmann condensation to produce acylphloroglucinol 11 [31, 32]. Subsequent prenylations (or geranylations) occur via an enzymatic process that involves prenyltransferase-catalyzed reactions
435
12 Polyprenylated Phloroglucinols and Xanthones
436
O
O
SACo
HO
HO
OH
3 x malonyl-CoA
OH
DMAPP
CoAS
PT
O
O 10
isobutyryl-CoA (9)
CoAS
O
OH
O
O
12
phlorisobutyrophenone (11)
O O
OH
DMAPP PT
proposed
O
malonyl-CoA DMAPP =
OPP
HO
OH
O
O
colupulone (15)
Scheme 12.1
OH
HO
OH
HO
dimethylallyl diphosphate PT = prenyltransferase
HO proposed
O
O
cohumulone (14)
OH
O
deoxycohumulone (13)
Biosynthesis of acylphloroglucinol 11 and bitter acids 14 and 15.
of the appropriate diphosphates with phloroglucinol [25, 33–37]. This stepwise prenylation process is illustrated in Scheme 12.1 for the construction of 12 and 13 (deoxycohumulone) from 11. It has been shown that chemical and/or enzymatic oxidation of 13 can lead to cohumulone (14), a representative α-acid. It has also been proposed that additional prenylation of 13 can form colupulone (15), a typical β-acid [25]. Type A and type B PPAPs are proposed to arise via reaction of acylphloroglucinol 16 with prenyl diphosphate. The resulting carbocation 17 can be attacked by the C1 or C5 enol to produce compounds 18 or 19, respectively, that represent type A and type B PPAPs (Scheme 12.2) [14]. Acylphloroglucinol 20, containing a prenylated C1 center, is the proposed intermediate of type C PPAPs. Specifically, reaction of 20 with prenyl diphosphate can form carbocation 21 and, after cyclization, bicyclic 22, a type C PPAP [10]. 12.2.3 Biomimetic Synthesis of PPAPs
A 2006 review summarizes the synthetic efforts toward PPAPs [10]. These studies set the stage for the first total synthesis of garsubellin A, reported by the Shibasaki group [38], and were followed a few months later by a synthesis from the Danishefsky group [39]. Since then, additional total syntheses have been published of either type A [hyperforin (1), garsubellin A (2), nemorosone (3)] [40–42], or type B [clusianone (4)] [41, 43–45] PPAPs. Among all published approaches there are only two strategies built upon biomimetic considerations that involve: (i) formation of a fully prenylated B ring and (ii) construction of the A ring via a cation-based alkylative dearomatization. Both strategies departed from the
12.2 Polycyclic Polyprenylated Phloroglucinols
437
O R 1′
5′
C1 attack
HO R
O8 5
O
HO
HO R 5′
O O
HO
7
O
1′
HO O
18
HO R
OPP
1′
5′
17
7
O
O 5
C5 attack
path toward type A, B PAPPs
HO
3
O
7
1′
HO R
O
HO R
O
20
6
O O
O
O
1
9
R HO
O 22
21
OPP
5
O
HO R
R
O Type B
R1
O
OH
1
9
O
O
19
OH
R O Type A
3
O
HO
16
O 1
9
O
3
Type C
path toward type C PAPPs
Scheme 12.2
Michael additionElimination
Biosynthesis of type A, type B, and type C PPAPs from MPAPs.
Intramolecular Michael addition
Alkylation
OHC
OH
O
25
HO
OH R
O
O
23
R1
X
OH
O 24 OAc R = Ph
O
R2 HO
R = i-Pr R1 = H or CH2OTBS R2 = H or prenyl X = Cl or Br
clusianone (4)
Scheme 12.3
Annulation based on R1 cation R2cyclization O R OH
O
26
Biomimetic approaches toward type A and type B PPAPs.
bis-prenylated acylphloroglucinol 23, a key intermediate in the biosynthesis of these natural products (Scheme 12.3). The first one, reported by the Porco group, resulted in the total synthesis of clusianone (type B PPAP) [45] using a double Michael reaction as a key step. More recently, the Couladouros group produced compound 26, representing the fully functionalized bicyclic core of the type A PPAPs, by a double alkylation of 23 with allyl electrophile 25 [46].
12 Polyprenylated Phloroglucinols and Xanthones
438
12.2.3.1 Biomimetic Total Synthesis of (±)-Clusianone The Porco synthesis of clusianone (4) is summarized in Scheme 12.4 and features a double Michael addition of clusianophenone B (27) with α-acetoxy enal 24. The bicyclic product 28 was methylated to form 29 as a mixture of regioisomers [45]. The employment of enal 24 proved to be a useful handle for the ensuing installation of the prenyl group at C7 (Scheme 12.4). Addition of vinyl magnesium bromide to aldehyde 29 and acetylation of the resulting alcohol gave access to acetate 30. Palladium-mediated formate reduction followed by cross metathesis using Grubbs second-generation catalyst (31) yielded methylated clusianone 32 (81% over two steps). Finally, demethylation of 32 afforded (±)-clusianone (4) as a mixture of enol tautomers.
AcO
CHO 7
OH
O
O KHMDS, 24, 65 °C
HO
1.
OH
O
O
OR O
clusianophenone B (27)
OHC 24
N
TMSCHN2, i Pr2EtN 54% (2 steps)
O
OMe O 30
28: R = H
1. Pd(PPh3)4, HCO2NH4, 90% 2. Grubbs 2nd generation cat (31) 2-methyl-2-butene, 89%
29: R = Me
OAc
O
O
N Cl
LiOH, ∆, 77%
Ru Cl
MgBr
2. Ac2O, i Pr2EtN, DMAP 74%
Ph P(Cy)3
Grubbs 2nd generation catalyst (31)
Scheme 12.4
O
OH
or LiCl, ∆, 69%
O clusianone (4)
O
OMe O
32
Biomimetic synthesis of clusianone by Porco, Jr. et al.
The double Michael reaction, used for the conversion of 27 into 28, deserves an additional comment. The authors reported that heating of this reaction at 65 ◦ C led to desired compound 28 via epimerization of the C7 aldehyde stereocenter. Interestingly, performing this reaction at 0 ◦ C formed the adamantane-like compound 34 via an intramolecular aldol reaction. When 34 was treated with potassium hexamethyldisilazane (KHMDS) at 65 ◦ C compound 28 was synthesized, presumably via a retro-aldol epimerization process (Scheme 12.5). This observation allows the synthesis of adamantane-like compounds that are structurally related to hyperibone K (36).
12.2 Polycyclic Polyprenylated Phloroglucinols O CH
H
27
H OH O
O
KHMDS, 24, 0 °C
439
O O
O
O O
O 34
33
1 eq. KHMDS, 65 °C retro-aldol epimerization
O HC
CHO
H O
H
O H3O+
O O
O
O
OH
O O
O
O
hyperibone K (36)
O
35 28
Scheme 12.5
Observations related to the double Michael reaction 27 → 28.
12.2.3.2 Biomimetic Approach to the Bicyclic Framework of Type A PPAPs The Couladouros approach towards the bicyclic core of type A PPAPs is highlighted in Scheme 12.6 [46]. C-alkylation of deoxycohumulone (13) with chloride 37, under a two-phase solvent system at pH 14, produced compound 38 as a mixture of two diastereoisomers. Acetylation of the C4 hydroxyl group of 38 followed by mesylation OTBS
OTBS
HO
8
OH
OH
KOH (3 N), 37, PhCl/H2O aliquat 336, pH=14
OH
O
1. Ac2O, pyridine
* HO 4
99%
b
8 O
OH
O 38
2. MsCl, Et3N 65–73%
O 9
O
1
a
b
AcO
b
OH
39
deoxycohumulone (13)
OTBS OTBS
Cl HO
TBSO O
37
O
O
O
+ AcO
Scheme 12.6
O
AcO
41
40
via path b
via path a
Biomimetic formation of the type A carbon framework.
O
12 Polyprenylated Phloroglucinols and Xanthones
440
of the C8 alcohol produced bicyclic motifs 40 and 41 in 1.5 : 1 ratio, presumably via common intermediate 39. Compound 41 is generated via O-alkylation of the C9 enol to the intermediate C8 carbocation formed during the reaction (path β). Notably, the presence of the double bond, allylic to the tertiary alcohol, is of importance for the stabilization of the cation 39. The authors have also evaluated a Michael addition for construction of the type A skeleton, in a similar manner as that presented above, in the synthesis of clusianone [45]. This approach was only successful for the synthesis of compounds non-substituted at C8 (Scheme 12.7). These results illustrate the difficulty in synthesizing the fully functionalized carbon framework of type A PPAPs, where the quaternary bridgehead C1 is located next to the fully substituted C8. O
OTBS
13
H H TBSO Br KOH (3 N), PhCl/H2O
O
aliquat 336, pH=14, 38%
O OH
HO
O
1. Ac2O, pyridine, 89% 2. HF⋅pyr 3. IBX 69% (2 steps)
O
8
O NaHCO3
AcO
OH
42
Scheme 12.7
O O
1
71%
AcO
O
43
44
Construction of a type A PPAP motif via an intramolecular Michael addition.
12.2.3.3 Biomimetic Synthesis of (±)-Ialibinone A and B and (±)-Hyperguinone B Very recently a biomimetic synthesis of PPAPs isolated from Hypericum papuanum was reported. Scheme 12.8 depicts the total synthesis of racemic ialibinones A and O
O
O
O
PhI(OAc)2 TEMPO
73%
O
O
OH
OH
V hyperguinone B O 13
O
O
O
PhI(OAc)2
NaOMe, MeI
OH
HO
78%
O
O
O
O
OH
OH
H I
45
O
O H
O
OH
O
O
O
OH
H
O
II
O
O
+
−H
58%
O
O
OH
H
Scheme 12.8
ialibinone B
OH
H IV
ialibinone A
O
+
Biomimetic synthesis of PPAPs via oxidative cyclization reactions.
III
12.2 Polycyclic Polyprenylated Phloroglucinols
B and hyperguinone B [47]. The synthesis of all three natural products started with compound 13, which is the same starting material used in Couladouros’ approach and similar to that used in Porco’s synthesis of clusianone. The C-methylation of 13 was carried out by treatment with NaOMe–MeI to give 45 in very good yield. Reaction of 45 with PhI(OAc)2 gave a 1 : 1 mixture of (±)-ialibinones A and B in 58% combined yield. The reaction most likely proceeds via an initial single-electron oxidation of 45 to give intermediate I. A stereoselective 5-exo-trig cyclization of this radical onto the pendant prenyl group would then give tertiary radical II, which can undergo a second cyclization onto the other prenyl group to give the tertiary radical III. Finally, an additional single-electron oxidation of III would lead to the tertiary carbocation IV and then to the final racemic natural products. In contrast, treatment of 45 with PhI(OAc)2 in the presence of TEMPO (2,2,6,6-tetramethylpiperidine-1-oxyl) gave (±)-hyperguinone B in 73% yield. The reaction presumably proceeds via hydride abstraction by the in situ generated TEMPO cation to give intermediate V, which undergoes a 6π-electrocyclization leading to the pyran ring of hyperguinone B. 12.2.4 Non-biomimetic Synthesis of PPAPs
In addition to the biomimetic approaches toward PPAPs, presented above, there are also a few non-biomimetic total syntheses that have been reported in the literature in the past few years. In this section we present briefly the total synthesis of garsubellin A, nemorosone, and hyperforin – representative members of type A PPAPs – and the synthesis of clusianone, a type B PPAP. 12.2.4.1 Total Synthesis of Garsubellin A The Shibasaki synthesis of garsubellin A (2) is summarized in Scheme 12.9 [38]. An interesting structural feature of this natural product is the additional ring (C-ring) that is formed by oxidative cyclization of a prenyl group. Key to the synthesis was the conversion of compound 47 into 51 via two steps: (i) a stereoselective Claisen rearrangement of 47 → 49 via intermediate 48, which installs the alkene substituents at C1 and C5 syn to each other, and (ii) a ring-closing metathesis using the Hoveyda–Grubbs catalyst 50. The fused tetrahydrofuran ring was constructed after allylic oxidation, hydrolysis of the carbonate, and Wacker oxidative cyclization. Finally, Stille coupling of 53 with tributyl(prenyl)tin completed the total synthesis of (±)-garsubellin A. The Danishefsky synthesis of garsubellin A (2) is highlighted in Schemes 12.10 and 12.11 [39]. Compound 54, representing the B ring of the target molecule, was converted into acetonide 55 in five steps in 43% overall yield (Scheme 12.10). Treatment of 55 with HClO4 at 80 ◦ C led to a mixture of bicyclic adducts 57 and 58 that, upon further heating, produced initially 59 and ultimately adduct 60 (71% isolated yield). Iodocarbocyclization of 61 provided 62 using standard iodolactonization conditions (Scheme 12.11). Conceptually, this reaction is similar to the Se-mediated
441
12 Polyprenylated Phloroglucinols and Xanthones
442
OEt
O 14 steps
O
A yield
46
O
NaOAc, ∆
A
O O MOMO
O 6% combined
O O
5
O O MOMO
96%
MOMO 47
O
O COi Pr
48
A 1
49 50 92%
HO O
O A
HO
1. I2, CAN
B
C O
MOMO
C O
2. pTsOH⋅H2O 80% (2 steps)
O I 53
SnBu3 PdCl2⋅dppf
O
O A
HO
1. (PhSe)2, PhIO2, pyr 2. CSA, 70% (2 steps)
B 52
O
B
3. LiOH 4. Na2PdCl4, TBHP 71% (2 steps)
O
AO O
O 51
O
Mes N N Mes Cl Cl Ru O
20%
garsubellin A (2) 50
Scheme 12.9
Total synthesis of garsubellin A by the Shibasaki group.
O
O
OTIPS B
5 steps
B
MeO
MeO
OMe
43% combined yield
OMe
B O
60 (71%)
Scheme 12.10
C H O 59
56
OMe
OMe HO
HO
HO C H O
HO
OMe
HO
OMe
HO
55
O
O
MeO
H2O/dioxane 80 °C
O
54
B
HClO4
B O
C H O
B
OMe 58
O
C H O
B O OMe 57
Formation of the B-C ring system of garsubellin A by Danishefsky et al.
cyclization approach reported by Nicolaou during his efforts towards the synthesis of garsubellin [48, 49]. Compound 62 underwent additional iodination to form triiodide 63. Treatment of 63 with excess isopropylmagnesium chloride led to compound 64 via an intramolecular Wurtz cyclopropanation and subsequent allylation. Reaction of 64 with TMSI (trimethylsilyl iodide) formed 65, containing the tricyclic framework of garsubellin A. Finally, the synthesis of garsubellin A (2) was completed after decoration of 65 with the appropriate prenyl and acyl substituents.
12.2 Polycyclic Polyprenylated Phloroglucinols
443
I
60
Grubbs 2nd generation cat. (31) 2-methyl-2-butene
O
O
O
HO
68%
I
I2, KI, KHCO3
H O
O
HO
85%
H O
61
I
I2, CAN
O
77%
HO
I H O
62
O 63 I
i PrMgCl then Li2CuCl4
67%
Br
I
I 1. AIBN,
O HO
SnBu3 82%
H H O
O
O
O HO
2. Grubbs 2nd generation cat. 2-methyl-2-butene
H O
H
TMSI then 1 N HCl
O
98%
HO H O
O
73%
66 LDA, TMSCl, then I2
65
64
25–36%
−i PrMgCl then
O TMSO
O
I
O
OH 1. Dess-Martin
TMSO
H O
O 67
Scheme 12.11
garsubellin A (2)
2. Et3N(HF)3
72%
H O
O
88% (2 steps)
68
Total synthesis of garsubellin A by the Danishefsky group.
12.2.4.2 Total Synthesis of Nemorosone and Clusianone through Differentiation of ‘‘Carbanions’’ More recently, the Danishefsky group has extended the above strategy to the synthesis of nemorosone (3) and clusianone (4) [41]. These natural products proved to be more demanding targets than garsubellin A, due to the lack of the furano-fused ring that provides further stability to the molecule. A significant complication arose during the iodonium-induced carbocyclization of compound 70 (Scheme 12.12). In addition to the desired product 73, the authors obtained substantial quantities of 71 and 72 (53% combined yield). Most likely, these side products were formed via O-alkylation of an iodonium intermediate. Gratifyingly, treatment of 71 and 72 with zinc in aqueous THF could regenerate 70 in high yield, allowing recycling of the starting material. Reaction of 73 with isopropylmagnesium chloride produced cyclopropane adduct 74 that, in turn, upon treatment with TMSI gave rise to bicyclic motif 75. Radical allylation then produced compound 76, a common intermediate in the synthesis of nemorosone and clusianone. Completion of the synthesis of 3 and 4 required differentiation of carbons C1 and C3 of common intermediate 76. To this end, treatment of 76 with an excess
69
3
O 1
O
MeO 70
O 1
SnBu3 84%
AIBN
O
I2, KI, KHCO3
Zn 78%
O
70
75
O
I
O
I +
71 (29%)
MeO
MeO
O
Zn 87%
MeO
O I
98%
TMSI
70
O 72 (24%)
I
+
MeO
MeO
I
O
I
74
O
I
O
i PrMgCl 93%
73 (32%)
O
Formation of the common intermediate 70 towards the synthesis of nemorosone and clusianone.
76
32%
3 steps
MeO
OH
Scheme 12.12
MeO
OMe
I
444
12 Polyprenylated Phloroglucinols and Xanthones
12.2 Polycyclic Polyprenylated Phloroglucinols
445
of LDA (lithium diisopropylamide) and TMSCl (trimethylsilyl chloride) followed by oxidative quenching with iodine provided 77 in moderate yield along with 78, which can in turn be converted into 77 by applying the same reaction conditions (Scheme 12.13). The total synthesis of 3 was completed after, first, acylation of C1 and then lithium-mediated allylation at C3.
O
O 1
MeO
51% for 77 12% for 78
MeO
O
15% combined yield
O
OH
TMS
76
nemorosone (3)
77: R1 = I
LDA, TMSCl then I2, 41%
Scheme 12.13
O
R1
then I2
O
3
O
5 steps
LDA, TMSCl,
78: R1 = H
Total synthesis of nemorosone.
On the other hand, treatment of 76 with LDA without the presence of TMSCl and quenching with benzaldehyde gave rise to 79, after acylation at C3 (Scheme 12.14). Conversion of 79 into its C1 iodo derivative and transformation of the iodo group into an allyl functionality concluded the synthesis of clusianone (4).
O
MeO
3
76
O
1. LDA, then PhCHO
1
O
2. Dess-Martin
57%
O 4 steps
MeO
O
20% combined yield
O 79
Scheme 12.14
OH
O O
clusianone (4)
Total synthesis of clusianone by the Danishefsky group.
12.2.4.3 Total Synthesis of (−)-Hyperforin (+)-Hyperforin (1), a type A PPAP, contains an additional chiral quaternary center compared to the other herein mentioned acylphloroglucinols, thus presenting greater difficulties towards its total synthesis. In early 2010 a catalytic asymmetric total synthesis of (−)-hyperforin was published by Shibasaki and coworkers [42]. Notably, this is the first catalytic asymmetric synthesis of any PPAP, since the previously reported asymmetric synthesis of clusianone involved a late-stage kinetic resolution (Section 12.2.4.4) [50]. Compound 83, having the desired stereochemistry at C7 and C8 carbons, was constructed via a catalytic asymmetric Diels–Alder reaction between 80 and 81 (Scheme 12.15). Compound 83 was then converted into allyl ether 84, which underwent a stereoselective Claisen rearrangement to form ketone 86. This rearrangement proceeded with high selectivity from the β face, most likely due to the
12 Polyprenylated Phloroglucinols and Xanthones
446
OTIPS
OTIPS O
N
OTIPS O +
FeBr3, AgSbF6, 82
O 80
OTIPS
A 8
93%, 96% ee
O 17 steps
N O
A
33%
7
81
MOMO
O O 83
84 >99% (d.r. = 12:1)
∆ O O
O
1. (Sia)2BH, H2O2, NaOH 2. Dess-Martin
A
O
O
O
H
MOMO
A
74% (2 steps)
i PrCO
R Me
O
MOMO
MOMO 87
H
8
86
85
1. NaOEt 2. Dess-Martin 86% (2 steps)
OMe
MOMO
O O
O A B
O A
6 steps
B
53% combined yield
O
O 88
Scheme 12.15
Ar
O
N N
N
O
82
Ar
Ar = 4-EtOC6H4
89
Catalytic asymmetric synthesis of the bicyclic framework of hyperforin.
pseudo-axial orientation of the methyl group at C8, which blocks the α face, as shown in intermediate 85. Initial attempts to form the B ring of hyperforin via an olefin metathesis approach proved unsuccessful, prompting the development of an alternative strategy using an intramolecular aldol reaction [51]. With this in mind, hydroxylation and oxidation of the terminal alkene of 86 produced aldehyde 87 that after intramolecular aldol reaction and oxidation gave rise to bicyclic adduct 88. Peripheral decoration of 88 afforded ketone 89 in six steps (53% combined yield). Scheme 12.16 shows the completion of the (−)-hyperforin synthesis from 89. Oxidation of C2 proved to be more difficult than anticipated since any attempt at nucleophilic addition in 90 failed. Furthermore, efforts to induce a [3.3] sigmatropic rearrangement of xanthate 91 led to dithionate 92 after a [1.3] rearrangement. This finding, however, allowed the use of a vinylogous Pummerer rearrangement for the oxidation of C2. To this end, dithionate 92 was converted into methylsulfoxide 93 and, after Pummerer rearrangement, to the desired allylic alcohol 94 (three steps, 61% combined yield). Finally, the prenyl group at C3 was installed by intramolecular allyl transfer via a π-allyl-palladium intermediate and cross metathesis to give (−)-(1).
O
A
OH
O
B
O 2
O
7 steps
90
A
4% combined yield
O
MeS
2. CS2, NaH, MeI, >99%
A
1. NaBH4, 95% (d.r.>33:1)
Total synthesis of ent-hyperforin.
(−)- hyperforin (1)
B
O
2. Pd(OAc)2, O2 >99%
1. TMSCl, Et3N DMAP, 84%
Scheme 12.16
89
OMe
3
94
B
O
OH
O
S
O
B
O
OMe
91 SMe
A
∆
(d.r.>33:1)
TFA, 2,6-di-tertbutylpyridine, 65%
O
OMe
MeS O
O
S
B
O
A
B
O
SMe
A
93
O
OMe
1. EtSLi, MeI, Et3N 98% (2 steps) 2. NaBO3·4H2O 95% (d.r. = 1.3:1)
92
O
OMe
12.2 Polycyclic Polyprenylated Phloroglucinols 447
448
12 Polyprenylated Phloroglucinols and Xanthones
12.2.4.4 Total Synthesis of Clusianone In 2006 the Simpkins laboratory reported the first total synthesis of (±)-clusianone using as a key step a regioselective lithiation of enol ether derivatives [43]. The construction of the bicyclic core was inspired by the approach described by Spessard and Stoltz (Scheme 12.17) [52]. These authors accomplished a diastereoselective conversion of enol ether 95 into bicyclic trione 96, using dichloromalonate as the electrophile. Notably, construction of the bicyclo[3.3.1]nonane by the use of malonyl dichloride was reported initially by Effenburger [53].
7 5
1
OTBS 95
O
1. Cl
Scheme 12.17 Model studies toward the synthesis of bicyclo[3.3.1]nonane-2,4,9-triones.
7
O
O
Cl
2. KOH, BnEt3NCl 36% 87% brsm
OH
O 96
Scheme 12.18 summarizes Simpkins’ synthesis of clusianone [43]. Notably, enol ether 98 was pre-functionalized with a prenyl group at the C1 center to overcome problems deriving from the steric hindrance of this center after the formation of the bicycle. In fact, Danishefsky had already reported similar problems during the installation of such a substituent in the synthesis of garsubellin A (Section 12.2.4.1). Interestingly, prenylation of 100 at the bridgehead position afforded compound 101 in high yield. The same efficiency was observed for the acylation of C3 by the action of lithium tetramethyl piperidine (LTMP), leading, after hydrolysis, to racemic clusianone. Importantly, prenylation of racemic 100 using a chiral base, such as 103, led to selective alkylation of the (−) isomer via a kinetic resolution process. This reaction allowed assignment of the absolute configuration of clusianone, isolated from Clusia torresii, as the (+) isomer. Interestingly, the (−) isomer of 4 matches the data of a compound reported by a Brazilian group [54]. Based on this, it is possible that clusianone exists in Nature in either enantiomeric form [50]. The above strategy was then extended to a formal synthesis of garsubellin A [40]. Specifically, application of the Effenburger-type cyclization to 105 afforded 110, a compound isolated by the Danishefsky group en route to the synthesis of garsubellin A (Scheme 12.19) [39]. Marazano and coworkers have also published their synthesis of (±)-clusianone incorporating a similar approach [44]. In their synthesis the starting enol ether 114 contains all three prenyl groups, yielding, after reaction with dichloromalonate, compound 115, which has both quaternary bridgehead centers (Scheme 12.20). Notably, formation of the desired product 115 was accompanied with side product 116 and desilylated compounds 112/113. Under different Lewis acid catalysis, this reaction produced significant amounts of bicyclic adduct 117. In principle the double alkylation of 114 to 115 has some biosynthetic relevance. However, the sequence of ring construction (formation of the B ring at the end) is not in accordance with the biosynthesis scenario in which the B ring is formed at the
103
O
O
O clusianone (4)
Ph
67%
5 steps
1
OH
OMe 98
7
O
O Cl
LiOH
O
O
O 102
O
2. KOH, BnEt3NCl
Cl
1.
Total synthesis of clusianone by the Simpkins group.
N
Li Li
N
Scheme 12.18
Ph
Ph
Ph
OMe 97
O 1
OH
OMe
99
O
7
91%
PhCOCl
LTMP
pTsOH 24% from 98
HC(CH3)3
O
LDA then
O
5
OMe
OMe
91%
101
3
O
Br
100
O
12.2 Polycyclic Polyprenylated Phloroglucinols 449
Scheme 12.19
garsubellin A (2)
104
O
HO
110
H O
105
O
O Cl
O
O
H 93%
Et3N-HF
2. KOH, BnEt3NCl
Cl
1.
O
OH
109
H O
106
TMSO
O
Formal synthesis of garsubellin A by the Simpkins group.
see Scheme 12.10
47%
3 steps
OTBS
O
O
H
X H O
*
50%
O
H
108: X=OTMS
107: X=OH
O
LTMP, Li(2-Th)CuCN then Br
TMSCl, DMAP imidazole, 90%
22% from 105
mCPBA
450
12 Polyprenylated Phloroglucinols and Xanthones
12.2 Polycyclic Polyprenylated Phloroglucinols OH
O 6 steps
A
O
A
+
A
112 7
:
113 3
451
OTMS TMSOTf, Et3N
A
O 111
1. O
114
O
Cl
BF3·Et2O 2. DMAP
1. O Cl
Cl
TMSOTf 2. DMAP
OH
O A
+
O
+
O 112/113
B O
OH 115 35% benzoyl cyanide Et3N 65%
O
O 116 25%
22%
117 38%
clusianone (4)
Scheme 12.20
O
Total synthesis of clusianone by Marazano and coworkers.
beginning. Based on this, such an approach cannot be considered as a biomimetic strategy. 12.2.5 Concluding Remarks
In short, biomimetic approaches toward PPAPs are based on double alkylation on a functionalized B ring with an electrophile. These strategies require fewer steps in a linear sense and can potentially produce the natural product target in higher yield. The total synthesis of racemic clusianone, by the Porco group [45], was completed in seven synthetic steps, starting from the biosynthetic intermediate 27, with a total yield of 25%. The approach is based on biosynthetic hypotheses that involve alkylative dearomatization of phloroglucinols and carbocation-mediated cyclization. This method can be applied with minor changes for the synthesis of other type B PPAPs and can also be implemented to the synthesis of adamantane-like PPAPs. The strategy presented by Couladouros et al. [46] can give access in few steps and satisfactory yields to several compounds possessing the bicyclic framework of natural products for SAR (structure–activity relationship) studies. In contrast to other reported methods, the two quaternary centers (C1 and C8) are connected during the cyclization step, thereby eliminating steric hindrance problems encountered in other reported approaches. A similar approach, and the most recently published biomimetic synthesis of PPAPs [47], starts with alkylation of a functionalized B ring, yet formation of the second ring occurs after oxidative cyclization. On the other hand, the non-biomimetic strategies require many steps,
O
Cl
452
12 Polyprenylated Phloroglucinols and Xanthones
as shown by the Danishefsky group syntheses of garsubellin A, nemorosone, and clusianone that were completed in 17, 14, and 12 steps respectively, or by the synthesis of (−)-hyperforin by the Shibasaki group, which required 44 steps. The advantage of the non-biomimetic approaches is that they are more flexible in terms of the synthetic route followed to the target and can lead to asymmetric final products. In such a way the asymmetric total synthesis of (−)-hyperforin was achieved. In addition to the synthetic strategies described herein, there is still ongoing interest towards the synthesis of PPAPs and their analogs. New approaches have appeared in the literature, aimed at a facile method for the construction of the bicyclic core of PPAPs that will allow easier access to those compounds for their biological evaluation and SAR studies [55–61].
12.3 Polyprenylated Xanthones 12.3.1 Introduction and Chemical Classification
Polyprenylated xanthones constitute a subclass of a larger class of compounds known as xanthones, all bearing a dibenzo-γ -pyrone scaffold [62]. Polyprenylated xanthones can further be divided, according to their oxidation degree, into mono-, di-, tri-, and so on, oxygenated compounds. Recently, the classification, synthesis, and biological evaluation of simple xanthones have been reviewed extensively [62–71]. In this chapter we focus on the so-called caged Garcinia xanthones (CGXs), owing to their similarities (isolation, biosynthesis) to the aforementioned polyprenylated phloroglucinols. Caged xanthones are natural products isolated from plants of the genus Garcinia (Guttiferae family) that are found in lowland rainforests of India, Indochina, Indonesia, West and Central Africa, and Brazil [1, 72]. The most studied member of this family is gambogic acid (118), a compound isolated from gamboge, the resin of Garcinia hanburyi (Figure 12.4). Common to the chemical structure of all CGXs is a xanthone backbone in which the C ring has been converted into an unusual 4-oxa-tricyclo[4.3.1.03,7 ]dec-8-en-2-one ring (caged) scaffold (see structure 120) [73, 74]. This general motif can be further decorated with different substituents on the aromatic ring A and/or can be oxidized to yield a wide range of compounds, representative members of which are shown in Figure 12.4. This concept is exemplified by the structure of forbesione (125), a natural product isolated from Garcinia forbesii [75] and Garcinia hanburyi [76]. Specifically, prenylation at the C5 center of forbesione (gambogic acid numbering) gives access to the gaudichaudione scaffold [77], represented here by deoxygaudichaudione A (126) [78]. Alternatively, prenylation of 125 at C5 followed by cyclization with the pendant phenol gives access to the morellin scaffold, represented here by desoxymorellin (123) [79]. Progressive oxidations at the C29 center of 123 produce morellinol (122), morellin
12.3 Polyprenylated Xanthones OH R O O
29
OH R O
O A
A
O
B
O
B
O
O
29 28
5
27
O
453
OH O
HO O
O
CO
O
CO
O
O
O
120: general motif of caged xanthones
118: R = CO2H: gambogic acid 119: R = CH3: gambogin
121: R = CHO: 122: R = CH2OH: 123: R = CH3: 124: R = CO2H:
OH
OMe O
O
5
HO
6
HO
HO O O
O
17
O O
O
17
morellin 125: forbesione morellinol desoxymorellin morellic acid
OMe O O O
O A16
O O
126: deoxygaudichaudione A 127: 6-O-methylbractatin
Figure 12.4
O
O
128: 6-O-methylneobractatin
O O
O
B C O 11
129: lateriflorone
Representative members of the caged Garcinia xanthone family.
(121), and morellic acid (124) [80]. Geranylation at the C5 center of forbesione affords, after formation of the pyran ring, gambogin (119) [81]. Further oxidation at C29 leads to the structure of gambogic acid (118) [76]. Compounds arising from isomerization around the C27=C28 double bond have also been isolated. Thus, morellin (121), having the cis configuration about the C27=C28 double bond, is known to isomerize to the trans isomer, isomorellin [82]. Similar observations have been reported for gambogic acid [83]. In contrast, the bractatin subfamily (127, 128) [84, 85] provides examples of forbesione-type natural products that contain a reverse prenyl group at the C17 center. Although the vast majority of the CGXs contain the general motif 120, there are a few examples of natural products with alternative cage structures or with additional oxidations of the xanthone motif. For instance, 6-O-methylneobractatin (128) is the only natural product known to contain a modified caged scaffold, referred to as the neo-motif [84, 85]. In addition, in the structure of lateriflorone (129) [86], the caged motif is attached to a spiroxalactone core, which is likely a product of oxidation of the xanthone B ring. Biologically, CGXs are known for their antimicrobial and anticancer activities and are widely used as herbal medicines in traditional Eastern medicine [83, 87–89]. Initial biological studies with semi-purified gamboge extracts documented its antiprotozoal activities, thus lending support for its indigenous use in the treatment of enteric diseases [90–94]. It has also been shown that morellin (121) and gambogic acid (118) exhibit a high specific growth inhibitory effect on Gram-positive bacteria in vitro and a protective action against experimental staphylococcal infections in mice [95–98]. In addition to their antimicrobial activity, most CGXs have received a great deal of attention for their anticancer activity [99, 100]. An ever increasing body of evidence indicates that these compounds are cytotoxic against various cancer cell lines at low micromolar concentrations [79].
12 Polyprenylated Phloroglucinols and Xanthones
454
12.3.2 Biosynthesis of Polyprenylated Xanthones
Biosynthetically, the xanthone backbone of the caged compounds is assumed to derive from common benzophenone intermediates that are synthesized in a similar manner to that described for PPAPs. Their oxygenation patterns indicate a mixed shikimate (formation of C ring)–acetate (formation of A ring) pathway [101–106]. The proposed biosynthesis is exemplified with the synthesis of maclurin (134) and 1,3,5,6-tetrahydroxyxanthone (135) in Scheme 12.21 [107–110]. Shikimic acid, derived from the shikimic pathway, can be converted into protocatechuic acid (131) after oxidation, dehydration, and enolization. Reaction of 131 with coenzyme A (HSCoA) can produce activated ester 132 that can further react with three units of malonyl-coenzyme A to yield intermediate 133. A Dieckmann condensation gives rise to benzophenones, such as maclurin (134). Depending upon the benzophenone produced, this is a branch point in the biogenesis of other benzophenone-type natural products. It is generally accepted that xanthones such as 1,3,5,6-tetrahydroxyxanthone (135) are formed by means of phenolic coupling of the benzophenone precursors [109, 110]. +
O OH
HO
NADP [O] dehydration enolization
O
O
HO
OH OH shikimic acid (130)
CoAS
HSCoA
OH OH protocatechuic acid (131)
OH OH 132 3 x malonyl-CoA
OH
O
A HO
OH C
O
A OH
OH
tetrahydroxy xanthone (135)
Scheme 12.21
HO
O
O Dieckmann condensation
C OH
OH OH
maclurin (134)
O −
O
OH O
SCoA 133
OH
Biosynthesis of the backbone of xanthones in higher plants.
Different hypotheses have been proposed for the biosynthetic conversion of simpler xanthones such as 135 into the more complex caged structures. The first proposal, illustrated in Scheme 12.22, requires at an early stage the prenylation of a xanthone, as 136, at C11 and C13 positions to produce 137 [111]. An oxidation–reduction–oxidation sequence of reactions is then required to form the final caged structure 141. Essential to this hypothesis is a presumed nucleophilic attack by the C13 tertiary alcohol of 138 on the pendant prenyl group that could initiate a cyclization cascade leading to the caged structure 140. Nonetheless, as shown with structure 139, neither the molecular geometry nor the reactivity required for this cascade is optimal, making this proposal unlikely.
12.3 Polyprenylated Xanthones O
O
O
O oxidation reduction
O
11
11
13
13
OH
13
OH O
137
O
138
O
O
O O
O
OH
136
O O
13 O
O O
13 O
141
455
13
HO
140
139
Scheme 12.22 Proposed biosynthesis of the CGX motif via a cascade of nucleophilic attacks.
A more plausible biosynthetic scenario stems from the pioneering work of Quillinan and Scheinmann [112]. In their work, they proposed that the caged motif can be formed via a Claisen rearrangement followed by a Diels–Alder reaction on the intermediate dienone (Scheme 12.23). The authors also provided experimental evidence in support of the Claisen/Diels–Alder reaction cascade: upon heating of compound 143, prepared by allylation of mesuaxanthone B (142), at 190 ◦ C for 14 h they observed products showing NMR signals characteristic of cage structure 145. At present, there have been no labeling experiments testing the biosynthetic feasibility of the Claisen/Diels–Alder reaction cascade. Nonetheless, additional support for the validity of this proposal was obtained by recent studies in which retro-Diels–Alder fragments have been detected in mass spectroscopy studies of several CGXs [113, 114]. O OH O
OH O
O
OH
OH mesuaxanthone (142)
Claisen rearrangement O 190 °C, 14 h
O
O
Diels–Alder reaction O
O O
O O
O 143
144
145
Scheme 12.23 Proposed biosynthesis of the CGX motif via a Claisen/Diels–Alder reaction cascade.
12.3.3 Biomimetic Synthesis of Caged Garcinia Xanthones
Inspired by Quillinan and Scheinmann’s proposed biosynthesis, both the Nicolaou [115] and Theodorakis [116] groups evaluated the tandem Claisen/Diels–Alder sequence for the synthesis of representative members of the CGX family. Their work provided further support for the proposed biosynthetic hypothesis.
O
12 Polyprenylated Phloroglucinols and Xanthones
456
Starting from the tris-allylated xanthone 146 both groups investigated the possibility of synthesizing forbesione (125) in one pot. In principle, exposure of such motif to heat could produce four products arising from a combination of two competing C-ring Claisen/Diels–Alder reactions, leading to regular and neo-caged motifs, and two A-ring Claisen migrations, producing C17 and C5 prenylations. Working with methoxy xanthone 146c, the Nicolaou group was the first to describe its conversion into methyl forbesione 147c and methyl neoforbesione (148c) in a 2.4 : 1 ratio and 89% combined yield (Scheme 12.24) [115]. On the other hand, studies by the Theodorakis group showed that heating of xanthone 146a led only to the isolation of forbesione 147a and isoforbesione (149a). The neo-C-ring isomers
OH
5 6
O A
HO
OR O
O B O
17
C
O 13
135
HO
OH
146a: R = H 146b: R = Ac 146c: R = Me
8
O
12
23 22
O 28 27
26
21
1. C-ring Claisen / Diels–Alder 1. C-ring Claisen/ Diels–Alder Claisen migration (C26–C28 unit) Claisen migration (C21–C23 unit) then Diels–Alder using C26–C27 then Diels–Alder using C21–C22 alkene as the dienophile alkene as the dienophile 2. A-ring Claisen reaction migration at C17 or C5 28
OR 5
OR
5
O
26
HO
12
17
O
2. A-ring Claisen reaction migration at C17 or C5
13 O
O O
HO O
17
O
21
147a: R = H forbesione (125) (49%) 147b: R = Ac (79%) 147c: R = Me methyl forbesione (63%)
O
HO 17
5 26 12
O
148a: R = H neoforbesione (ND) 148b: R = Ac (ND) 148c: R = Me methyl neoforbesione (26%)
28
OR
13 O 23
O
26
23
5
12 13
OR O
HO O
21
149a: R = H isoforbesione (35%) 149b: R = Ac (ND) 149c: R = Me (ND)
O 17
O
12 13
O
26
150a: R = H isoneoforbesione (ND) 150b: R = Ac (ND) 150c: R = Me (ND)
Scheme 12.24 Biomimetic synthesis of forbesione (125) and related structures via a Claisen/Diels–Alder/Claisen reaction cascade.
12.3 Polyprenylated Xanthones
148a and 150a were not detected in this case. More impressively, the O6-acetylated xanthone 146b afforded, upon heating, solely acetyl forbesione (147b) [116]. Similar observations have been reported more recently by other groups [117]. The above-described results towards the synthesis of forbesione along with the results from several model studies [118] can be summarized as follows: • The C-ring Claisen/Diels–Alder rearrangement proceeds first and is followed by an A-ring Claisen reaction. • The site-selectivity of the A-ring Claisen rearrangement (C17 versus C5 prenylation) is controlled by the steric and electronic effects of the C6 phenolic substituent. • The site-selectivity of the C-ring Claisen/Diels–Alder reaction is attributed to and governed by the electronic density of the C8 carbonyl-group. Being para to the C12 allyloxy unit, the electron-deficient C8 carbonyl carbon polarizes selectively the O–C28 bond and facilitates its rupture. In turn, this leads to a site-selective Claisen rearrangement of the C12 allyloxy unit onto the C13 center, thereby producing exclusively the regular caged motif found in the structure of forbesione (147a). • Substitution of the C6 phenol can regulate the electronic density of the C8 carbonyl group, thus affecting the site selectivity of the C-ring Claisen/Diels–Alder reaction. The experimental findings on the tandem Claisen/Diels–Alder/Claisen reaction cascade provide useful insights regarding the biosynthesis of all known CGXs [118]. All these natural products (representative examples shown in Figure 12.4) share a common caged motif, exemplified by structure 120, except for 6-O-methylneobractatin (128), which contains the neo-caged motif. The remote electronic effects of the seemingly innocuous 6-O-methyl group may explain the concomitant biosynthesis of both 6-O-methylbractatin (127) and 6-O-methylneobractatin (128). Studies by the Nicolaou group have shown that the Claisen/Diels–Alder reaction can be accelerated in the presence of polar solvents [119]. For instance, as depicted for the synthesis of gambogin (Scheme 12.25), the conversion of allyl ether 159 into caged structure 160a and the neo-isomer 160b was dramatically accelerated upon changing the solvent from benzene to DMF to a MeOH–water (1 : 2) mixture. It has been proposed that polar aprotic solvents, such as DMF, and more impressively protic solvents, such as water, can accelerate the Claisen rearrangement by stabilizing its polar transition state [120–124]. The concurrent acceleration of the Diels–Alder component of this cascade may be due to the hydrophobic effect of water [125] rather than to a polarity or hydrogen-bonding phenomena [126–128]. Computational studies on the above-mentioned reaction have also concluded that the Claisen rearrangement is reversible and the energetics of the irreversible Diels–Alder cyclization can determine the product formation [129].
457
12 Polyprenylated Phloroglucinols and Xanthones
458
OMOM A
MOMO
1. n BuLi 2. TBAF 3. MnO2
Br
156 OMOM O H
6
MOMO
18
A
C
1. NaHMDS, 154 2. TBAF C
HO 157
OBn
MOMO
O R2O
Br
6
O
MOMO +
17
O
MeOH/H2O, ∆ MOMO
O O
18
100%
O
OMOM O
O O
160a (75%)
160b (25%)
OR1
1. t BuOK, 154 74% 2. Ph3P=CH2
MOMO 6
OMOM O
158a: R1 = H, R2 = CMe2CH=CH2 158b: R1 = CMe2CH= CH2, R2 = H
H 154
OMOM O O O
MOMO
18
3. Ph3P=CH2 57% OH O
OBn 152
6
6
8
B O
4. KOH 5. H2, Pd /C 44%
TBSO
OMOM O
O
159 1. HCl
1. Ac2O, pyr 2. DMF, ∆ 3. DBU, CuI (cat), 163 4. DMF, ∆
OH O O O O
12%
2. DBU, 161 CuI (cat) 3. H2, Lindlar
48% OCOCF3 161 OH O
OCOCF3
O 17
O
O O
O
163
162 gambogin (119)
Scheme 12.25
Biomimetic synthesis of gambogin by the Nicolaou group.
12.3.3.1 Nicolaou Approach to Forbesione and Gambogin Scheme 12.26 depicts the synthetic strategy developed by Nicolaou and Li [115] for the synthesis of 6-O-methylforbesione (147c). Xanthone 153 was generated in five steps and in 78% combined yield starting from the aryl bromide 151 and the benzaldehyde 152. Treatment of 153 with α-bromoisobutyraldehyde (154) under basic conditions followed by Wittig olefination produced a mixture of the diallylated compounds 155a and 155b that, after reiteration of the alkylation/olefination reactions, yielded the triallylated xanthone 146c. Heating of 146c in DMF at 120 ◦ C induced the Claisen/Diels–Alder/Claisen reaction cascade to produce compound 147c along with its neo-isomer 148c in 89% combined yield. In a similar manner, the total synthesis of gambogin was achieved starting from the partially protected xanthone 157 (Scheme 12.25) [119]. This time the Claisen/Diels–Alder reaction proceeded quantitatively in refluxing MeOH–H2 O (1 : 2) to produce the regular caged motif 160a along with its neo-isomer 160b in a 3 : 1 ratio. Methoxymethyl (MOM) deprotection of 160a followed by propargylation with alkyne 161 at C18 and partial reduction with Lindlar catalyst gave rise to compound 162. Gambogin was then synthesized after a sequence of four reactions
12.3 Polyprenylated Xanthones
459
OMe A
BnO
Br
151 OMe
O H
C
TBSO
OBn
1. n BuLi 18 A 2. TBAF HO 3. MnO2 4. KOH 5. H2, Pd/C 78%
OMe O 8
B O
13
HO
1. t BuOK, 154 C 12
OH
154 6
HO
OMe O O O
6
+ O
R2O
Br
1. t BuOK, 154 86% 2. Ph3P=CH2
OMe O
HO O O
O
DMF, ∆ 89%
O
18
OMe O
O O
148c: 6-O-methylneoforbesione (26%)
Scheme 12.26
OR1
155a: R1 = H, R2 = CMe2CH=CH2 155b: R1 = CMe2CH = CH2, R2 = H
O H
OMe O O
2. Ph3P=CH2 64%
153
OBn 152
O
18
147c: 6-O-methylforbesione (63%)
O
146c
Biomimetic synthesis of 6-O-methylforbesione (147c).
that included: (i) acetylation of C6 phenol; (ii) Claisen rearrangement to install the prenyl group at C17; (iii) propargylation of the resulting phenol with alkyne 163; and (iv) Claisen rearrangement to form the dihydropyran ring of the natural product. 12.3.3.2 Theodorakis’ Unified Approach to Caged Garcinia Xanthones The common structural motif of most CGXs suggests that they can be synthesized by functionalizing the A ring of forbesione. Along these lines, the Theodorakis group developed a strategy that uses forbesione (125) to gain access to representative members of the gaudichaudiones, morellins, and gambogins [118]. As illustrated in Scheme 12.27, ZnCl2 -mediated condensation of phloroglucinol (164) with benzoic acid 165 produced xanthone 135. Propargylation of 135 with the propargyl chloride 166 followed by partial reduction using Lindlar catalyst and acetylation of phenol at C6 gave rise to compound 146b. Heating of 146b (DMF, 1 h, 120 ◦ C) set the stage for a site-selective Claisen/Diels–Alder/Claisen reaction cascade that produced, after deprotection of the C6 acetate, forbesione (125) in 72% combined yield. Further decoration of the A ring of forbesione gave access to more functionalized CGX family members. Specifically, propargylation of the C18 phenol of forbesione with chloride 166 afforded, after Lindlar reduction and Claisen rearrangement, deoxygaudichaudione A (126). On the other hand, propargylation of 125 and immediate Claisen rearrangement formed desoxymorellin (123). Finally, condensation of forbesione (125) with citral (168) in Et3 N produced gambogin (119). 12.3.3.3 Synthesis of Methyllateriflorone It has been proposed that the unprecedented spiroxalactone motif of lateriflorone (129) could be formed by condensation of two fully functionalized fragments, 169 and 170 (Scheme 12.28) [86]. An alternative and likely more biosynthetically
460
12 Polyprenylated Phloroglucinols and Xanthones OH
HO
6
A ZnCl2,
164 OH
POCl3
O HO
HO
18
O B C
HO
C
OH
O
OH
O
146b Cl 166
DMF, ∆
OH O
O
6
O
1. K2CO3, 166 18 2. H2, Lindlar HO 3. DMF, ∆
O O
O
O
135
OH HO
O
OAc
18
3. Ac2O, pyr 16%
OH 165
18
6
1. K2CO3, CuI (cat), 166 2. H2, Lindlar
A
O
46%
HO
OH
O O
34%
O
79%
OAc O
O
K2CO3, MeOH
O
91%
O deoxygaudichaudione A (126)
167
forbesione (125)
1. K2CO3, 166 61% 2. DMF, ∆
75% Et3N, 168
O H
OH
OH
168
O O
O
O O
O O
O
desoxymorellin (123)
Scheme 12.27
O O
O
gambogin (119) (C2 epimers)
Unified biomimetic synthesis of CGXs by the Theodorakis group.
relevant hypothesis could involve conversion of xanthone (172) into dioxepanone (171) that, upon hydrolysis and spirocyclization at the C16 center, could form the spiroxalactone ring system of lateriflorone. Quite recently, the Nicolaou group has reported a synthesis of C11methyllateriflorone (178) (Scheme 12.29) [130]. Key to the strategy was the coupling of orthogonally protected hydroquinone 173 with acid 174 that after selective deprotection of the C7 MOM ether produced compound 175 (61% combined yield). Oxidation of 175 in the presence of iodosobenzene bis(trifluoroacetate) in methanol, followed by heating under acidic conditions formed spiroxalactone 177. Acid-catalyzed hydrolysis of 177 gave rise to C11-methyllateriflorone (178) in 66% yield. 12.3.3.4 Non-biomimetic Synthesis of the Caged Garcinia Xanthones An alternative non-biomimetic synthesis of CGXs relies on a tandem Wessely oxidation/Diels–Alder reaction cascade. Yates and coworkers applied this strategy to the synthesis of caged structures reminiscent of the CGX motif. Thus, treatment
12.3 Polyprenylated Xanthones
O O
O
HO A
O
16
OH
+
O O
O HO
O
A16 OH
O
O
169
170
B
O
O O
O
OH
129: lateriflorone B-ring hydrolysis and reorganization
OH O
O HO
B 16 O
HO
OH
C
O HO
16 O
O
OH
OH
172
Scheme 12.28
O O
prenylation B-ring oxidation
A
171
Proposed biosynthesis of lateriflorone (129).
of phenol 179 with Pb(OAc)4 in acetic acid produced 2,4-cyclohexadienone 180 that, upon heating at 140 ◦ C, formed compound 181 (gambogic acid numbering) (Scheme 12.30) [131]. In a previous study, xanthene 182 was treated with lead tetraacrylate [formed in situ by Pb(OAc)4 and acrylic acid] to produce dienone 183 and, after an intramolecular Diels–Alder reaction, caged compound 184 [132]. Theodorakis and coworkers [133] investigated the application of the Wessely/Diels–Alder strategy for the synthesis of a more hydroxylated caged motif related to the structure of lateriflorone (129) (Figure 12.4). Treatment of 185 with Pb(OAc)4 in acrylic acid–dichloromethane produced, after heating in refluxing benzene (80 ◦ C), tricyclic lactone 187 in 82% combined yield (Scheme 12.31). Crystallographic studies established that 187 is a constitutional isomer of the desired structure 190 and is reminiscent of the so-called neo-caged structure. The connectivity of compound 187 suggested that during the Wessely oxidation the acrylate unit was attached exclusively at the more electronically rich C11 center of 185, instead of the desired C13 carbon. In turn, this produced dienone 186 that subsequently underwent an efficient Diels–Alder cycloaddition with the pendant acrylate dienophile. To alter the connectivity of the caged structure, one could have the acetoxy group preinstalled at the C13 center and promote the migration of the prenyl group. Along these lines, heating of allyl ether 188 in m-xylene (140 ◦ C) gave rise exclusively to caged motif 190 via a Claisen rearrangement and Diels–Alder cycloaddition. The selectivity of the Claisen rearrangement at the C13 center can be explained by considering that intermediate 189 has the necessary geometry that allows it to be trapped as the Diels–Alder adduct.
461
7
A
+
O
O
178
O O
O
HO
HO
B
O
174
O
O 11 OMe
O
O 11 OMe
66%
2. PPTS, ∆
1. H2O, H+
61% O
1. EDC, DMAP MeO O 2. HCl, MeOH A
O
MeO
175
A
OH HO
O
Total synthesis of C11-methyllateriflorone (178).
A
OMOM
OH
Scheme 12.29
173
O
MeO
O
177
O O
9
O
B
O
O
11 OMe
O
PhI(TFA)2, MeOH O 60% O 11 OMe
176
O HO
O
69%
PPTS, PhH, ∆
A
OMe
O
MeO
O
O 11 OMe
O
462
12 Polyprenylated Phloroglucinols and Xanthones
12.3 Polyprenylated Xanthones O O
Pd(OAc)4 13
ACOH
12
13
C6H4Me2
OH
O
Wessely oxidation
179
140 °C
12
O
O
Diels–Alder reaction
180
O
O 80 °C
O 12
13
OH
O
181
O Pd(OAc)4
O
12
13
O
12
OH O
O
13 O
OH
O
182
12
O
13
183
184
Scheme 12.30 Representative examples of caged structures, 181 and 184, formed via a Wessely oxidation/Diels–Alder reaction cascade.
O 11
MeO
OMe
12
13
Pd(OAc)4 MeO
O
OH
O 11 12 OMe 13 O
OH
185
MeO O
11
14
O
13
12
O
188
Scheme 12.31
MeO O 80 °C 82%
13
186
OMe
140 °C 92%
MeO O
11
14 13
O
O 187
OMe
12
MeO
11 OMe
12
O
12 O OMe 13 O 11 14
O 189
190
Synthesis of caged structures 187 and 190.
12.3.3.5 Concluding Remarks CGXs are a family of polyprenylated xanthones that have a remarkable chemical structure, inspiring biosynthesis, and significant medicinal potential. Their chemical structure is represented by an unusual xanthone backbone in which the C ring has been converted into a 4-oxa-tricyclo[4.3.1.03,7 ]dec-8-en-2-one (caged) scaffold. Their biosynthesis is proposed to involve a cascade of Claisen and Diels–Alder reactions and has provided the inspiration for the development of efficient laboratory syntheses of the parent molecules and designed analogs. Their medicinal value stems from their use in ethnomedicine and remains still largely unexplored [134]. The recent advances in the synthesis of these compounds have paved the way for the generation of analogs with the desired pharmacological and biological profile. In particular, the biosynthetically inspired Claisen/Diels–Alder reaction
463
464
12 Polyprenylated Phloroglucinols and Xanthones
cascade can reliably produce the caged motif of CGXs in excellent yields. On the other hand, the non-biomimetic Wessely/Diels–Alder strategy can form analogs of the caged motif that cannot be made by fragmentation of the natural products. It is very likely that these strategies will be used for the development of more potent CGX analogs. One limitation of both strategies is that, at present, they both deliver racemic mixtures of the caged structures. Thus, the development of an enantioselective variant of the Claisen/Diels–Alder and Wessely/Diels–Alder reaction cascades still needs to be addressed. References 1. Gustafsson, M.H.G., Bittrich, V., and
2. 3.
4. 5.
6. 7. 8.
9.
10. 11.
12.
13. 14.
15.
Stevens, P.F. (2002) Int. J. Plant Sci., 163, 1045–1054. Woelk, H. (2000) Br. Med. J., 321, 536–539. Di Carlo, G., Borrelli, F., Ernst, E., and Izzo, A.A. (2001) Trends Pharmacol. Sci., 22, 292–297. Miller, A.L. (1998) Altern. Med. Rev., 3, 18–26. Barnes, J., Anderson, L.A., and Phillipson, J.D. (2001) J. Pharm. Pharmacol., 53, 583–600. M¨uller, W.E. (2003) Pharmacol. Res., 47, 101–109. Verotta, L. (2002) Phytochem. Rev., 1, 389–407. Rodriguez-Landa, J.F. and Contreras, C.M. (2003) Phytomedicine, 10, 688–699. Medina, M.A., Marti’nez-Poveda, B., Amores-Sa’nchez, M.I., and Quesada, A.R. (2006) Life Sci., 79, 105–111. Ciochina, R. and Grossman, R.B. (2006) Chem. Rev., 106, 3363–3386. Linde, K., Berner, M.M., and Kriston, L. (2008) Cochrane Database Syst. Rev. (Art. No.: CD000448). doi: 10.1002/14651858.CD000448.pub3. Mabberley, D.J. (1997) The Plant-Book: A Portable Dictionary of the Vascular Plants, 2nd edn, Cambridge University Press, New York. Kumar, P. and Baslas, R.K. (1980) Herba Hung., 19, 81–91. Cuesta-Rubio, O., Valez-Castro, H., Frontana-Uribe, B.A., and Cardenas, J. (2001) Phytochemistry, 57, 279–283. M¨uller, W.E., Singer, A., Wonnemann, M., Hafner, U., Rolli, M., and
16. 17.
18.
19.
20.
21.
22.
23.
24.
25.
26. 27.
28.
Sch¨afer, C. (1998) Pharmacopsychiatry, 31 (Suppl. 1), 16–21. Mennini, T. and Gobbi, M. (2004) Life Sci., 75, 1021–1027. Schempp, C.M., Pelz, K., Wittmer, A., Sch¨opf, E., and Simon, J.C. (1999) Lancet, 353, 2129. Quiney, C., Billard, C., Salanoubat, C., Fourneron, J.D., and Kolb, J.P. (2006) Leukemia, 20, 1519–1525. Fukuyama, Y., Kuwayama, A., and Minami, H. (1997) Chem. Pharm. Bull., 45, 947–949. Cuesta-Rubio, O., Frontana-Uribe, B.A., Ramirez-Apan, T., and Cardenas, J. (2002) Naturforsch. Teil C, 57, 372–378. Lokvam, J., Braddock, J.F., Reichardt, P.B., and Clausen, T.P. (2000) Phytochemistry, 55, 29–34. Piccinelli, A.L., Cuesta-Rubio, O., China, M.B., Mahmood, N., Pagano, B., Pavone, M., Barone, V., and Rastrelli, L. (2005) Tetrahedron, 61, 8206–8211. Winkelmann, K., Heilmann, J., Zerbe, O., Rali, T., and Sticher, O. (2001) Helv. Chim. Acta, 84, 3380–3392. Weng, J.-R., Tsao, L.-T., Wang, J.-P., Wu, R.-R., and Lin, C.-N. (2004) J. Nat. Prod., 67, 1796–1799. Zuurbier, K.W.M., Fung, S.-Y., Scheffer, J.C., and Verpoorte, R. (1998) Phytochemistry, 49, 2315–2322. Gerh¨auser, C. (2005) Eur. J. Cancer, 41, 1941–1954. Adam, P., Arigoni, D., Bacher, A., and Eisenreich, W. (2002) J. Med. Chem., 45, 4786–4793. Drawert, F. and Beier, J. (1974) Phytochemistry, 13, 2149–2155.
References 29. Drawert, F. and Beier, J. (1976) Phyto30. 31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
chemistry, 15, 1693–1694. Drawert, F. and Beier, J. (1976) Phytochemistry, 15, 1695–1696. Klingauf, P., Beuerle, T., Mellenthin, A., El-Moghazy, S.A.M., Boubakir, Z., and Beerhues, L. (2005) Phytochemistry, 66, 139–145. Liu, B., Falkenstein-Paul, H., Schmidt, W., and Beerhues, L. (2003) Plant J., 34, 847–855. Zuurbier, K.W.M., Fung, S.-Y., Scheffer, J.C., and Verpoorte, R. (1995) Phytochemistry, 38, 77–82. Hecht, S., Wungsintaweekul, J., Rohdich, F., Kis, K., Radykewicz, T., Schuhr, C.A., Eisenreich, W., Richter, G., and Bacher, A. (2001) J. Org. Chem., 66, 7770–7775. Eisenreich, W., Rohdich, F., and Bacher, A. (2001) Trends Plant Sci., 6, 78–84. Gabrielsen, M., Rohdich, F., Eisenreich, W., Grawert, T., Hecht, S., Bacher, A., and Hunter, W.N. (2004) Eur. J. Biochem., 271, 3028–3035. Boubakir, Z., Beuerle, T., Benye, L., and Beerhues, L. (2005) Phytochemistry, 66, 51–57. Kuramochi, A., Usuda, H., Yamatsugu, K., Kanai, M., and Shibasaki, M. (2005) J. Am. Chem. Soc., 127, 14200–14201. Siegel, D.R. and Danishefsky, S.J. (2006) J. Am. Chem. Soc., 128, 1048–1049. Ahmad, N.M., Rodeschini, V., Simpkins, N.S., Ward, S.E., and Blake, A.J. (2007) J. Org. Chem., 72, 4803–4815. Tsukano, C., Siegel, D.R., and Danishefsky, S.J. (2007) Angew. Chem. Int. Ed., 46, 8840–8844. Shimizu, Y., Shi, S.-L., Usuda, H., Kanai, M., and Shibasaki, M. (2010) Angew. Chem. Int. Ed., 49, 1103–1106. Rodeschini, V., Ahmad, N.M., and Simpkins, N.S. (2006) Org. Lett., 8, 5283–5285. Nuhant, P., David, M., Pouplin, T., Delpech, B., and Marazano, C. (2007) Org. Lett., 9, 287–289. Qi, J. and Porco, J.A. Jr. (2007) J. Am. Chem. Soc., 129, 12682–12683.
46. Couladouros, E.A., Dakanali, M.,
47.
48.
49.
50.
51.
52. 53.
54.
55. 56.
57. 58. 59. 60. 61.
62. 63. 64.
65.
Demadis, K.D., and Vidali, V.P. (2009) Org. Lett., 11, 4430–4433. George, J.H., Hesse, M.D., Baldwin, J.E., and Adlington, R.M. (2010) Org. Lett., 12, 3532–3535. Nicolaou, K.C., Pfefferkorn, J.A., Kim, S., and Wei, H.X. (1999) J. Am. Chem. Soc., 121, 4724–4725. Nicolaou, K.C., Pfefferkorn, J.A., Cao, G.-Q., Sanghee, K., and Kessabi, J. (1999) Org. Lett., 1, 807–810. Rodeschini, V., Simpkins, N.S., and Wilson, C. (2007) J. Org. Chem., 72, 4265–4267. Shimizu, Y., Kuramochi, A., Usuda, H., Kanai, M., and Shibasaki, M. (2007) Tetrahedron Lett., 48, 4173–4177. Spessard, S.J. and Stoltz, B.M. (2002) Org. Lett., 4, 1943–1946. Sch¨onw¨alder, K.-H., Kollatt, P., Stezowski, J.J., and Effenburger, F. (1984) Chem. Ber., 117, 3280–3296. de Oliveira, C.M.A., Porto, A.M., Bittrich, V., Vencato, I., and Marsaioli, A.J. (1996) Tetrahedron Lett., 37, 6427–6430. Abe, M. and Nakada, M. (2007) Tetrahedron Lett., 48, 4873–4877. Abe, M., Saito, A., and Nakada, M. (2010) Tetrahedron Lett., 51, 1298–1302. Kraus, G.A. and Jeon, I. (2008) Tetrahedron Lett., 49, 286–288. Mehta, G. and Bera, M.K. (2009) Tetrahedron Lett., 50, 3519–3522. Mitasev, B. and Porco, J.A. Jr. (2009) Org. Lett., 11, 2285–2288. Takagi, R., Inoue, Y., and Ohkata, K. (2008) J. Org. Chem., 73, 9320–9325. Pouplin, T., Tolon, B., Nuhant, P., Delpech, B., and Marazano, C. (2007) Eur. J. Org. Chem., 5117–5125. Sousa, M.E. and Pinto, M.M.M. (2005) Curr. Med. Chem., 12, 2447–2479. Vieira, L.M.M. and Kijjoa, A. (2005) Curr. Med. Chem., 12, 2413–2446. Silva, A.M.S. and Pinto, D.C.G.A. (2005) Curr. Med. Chem., 12, 2481–2497. Gales, L. and Damas, A.M. (2005) Curr. Med. Chem., 12, 2499–2515.
465
466
12 Polyprenylated Phloroglucinols and Xanthones 66. Pinto, M.M.M., Sousa, M.E., and
67.
68.
69. 70. 71.
72. 73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
Nascimento, M.S.J. (2005) Curr. Med. Chem., 12, 2517–2538. Riscoe, M., Kelly, J.X., and Winter, R. (2005) Curr. Med. Chem., 12, 2539–2549. Pinto, M.M.M. and Castanheiro, R.A.P. (2009) Curr. Org. Chem., 13, 1215–1240. Pouli, N. and Marakos, P. (2009) Anticancer Agents Med. Chem., 9, 77–98. Han, Q.-B. and Xu, H.-X. (2009) Curr. Med. Chem., 16, 3775–3796. El-Seedi, H.R., El-Ghorab, D.M.H., El-Barbary, M.A., Zayed, M.F., Goransson, U., Larsson, S., and Verpoorte, R. (2009) Curr. Med. Chem., 16, 2581–2626. Sultanbawa, M.U.S. (1980) Tetrahedron, 36, 1465–1506. Ollis, W.D., Ramsay, M.V.J., Sutherland, I.O., and Mongkolsuk, S. (1965) Tetrahedron, 21, 1453–1470. Ahmad, S.A., Rigby, W., and Taylor, R.B. (1966) J. Chem. Soc. (C), 772–779. Leong, Y.-W., Harrison, L.J., Bennett, G.J., and Tan, H.T.-W. (1996) J. Chem. Res. (S), 392–393. Wang, L.L., Li, Z.L., Xu, Y.P., Liu, X.Q., Pei, Y.H., Jing, Y.K., and Hua, H.M. (2008) Chin. Chem. Lett., 19, 1221–1223. Cao, S.-G., Sng, V.H.L., Wu, X.-H., Sim, K.-Y., Tan, B.H.K., Pereira, J.T., and Goh, S.H. (1998) Tetrahedron, 54, 10915–10924. Han, Q.-B., Wang, Y.-L., Yang, L., Tso, T.-F., Qiao, C.-F., Song, J.-Z., Xu, L.-J., Chen, S.-L., Yang, D.-J., and Xu, H.-X. (2006) Chem. Pharm. Bull., 54, 265–267. Tao, S.-J., Guan, S.-H., Wang, W., Lu, Z.-Q., Chen, G.-T., Sha, N., Yue, Q.-X., Liu, X., and Guo, D.-A. (2009) J. Nat. Prod., 72, 117–124. Sukpondma, Y., Rukachaisirikul, V., and Phongpaichit, S. (2005) Chem. Pharm. Bull., 53, 850–852. Asano, J., Chiba, K., Tada, M., and Yoshii, T. (1996) Phytochemistry, 41, 815–820. Nair, P.M. and Venkataraman, K. (1964) Indian J. Chem., 2, 402–404.
83. Lin, L.-J., Lin, L.-Z., Pezzuto, J.M.,
84.
85.
86.
87.
88. 89.
90. 91.
92. 93. 94. 95. 96. 97. 98.
99.
100.
Cordell, G.A., and Ruangrungsi, N. (1993) Magn. Reson. Chem., 31, 340–347. Thoison, O., Fahy, J., Dumontet, V., Chiaroni, A., Riche, C., van Tri, M., and S´evenet, T. (2000) J. Nat. Prod., 63, 441–446. Thoison, O., Cuong, D.D., Gramain, A., Chiaroni, A., Hung, N.V., and S´evenet, T. (2005) Tetrahedron, 61, 8529–8535. Kosela, S., Cao, S.-G., Wu, X.-H., Vittal, J.J., Sukri, T., Masdianto, M., Goh, S.-H., and Sim, K.-Y. (1999) Tetrahedron Lett., 40, 157–160. Jiangsu New Medical College (1997) Dictionary of Chinese Traditional Medicines, Shanghai Scientific and Technical Publishers, Shanghai. Jinxiang, Y. (1989) Chin. J. Cancer Res., 1, 75–78. Panthong, A., Norkaew, P., Kanjanapothi, D., Taesotikul, T., Anantachoke, N., and Reutrakul, V. (2007) J. Ethnopharmacol., 111, 335–340. Rao, P.L.N. and Verma, S.C.L. (1951) J. Sci. Ind. Res. B, 10, 184–185. Gupta, V.S., Rao, A.V.S.P., and Rao, P.L.N. (1963) Indian J. Exp. Biol., 1, 146–147. Sani, B.P. and Rao, P.L.N. (1966) Indian J. Exp. Biol., 4, 27–28. Santhanam, K. and Rao, P.L.N. (1968) Indian J. Exp. Biol., 6, 158–159. Puttanna, C.R. and Rao, P.L.N. (1968) Indian J. Exp. Biol., 6, 150–152. Santhanam, K. and Rao, P.L.N. (1969) Indian J. Exp. Biol., 7, 34–36. Sani, B.P. and Rao, P.L.N. (1969) Indian J. Chem., 7, 680–684. Verma, S.C.L. and Rao, P.L.N. (1967) Indian J. Exp. Biol., 5, 106–109. Rao, D.R., Gupta, T.R., Gupta, V.S., Rao, K.V.N., and Rao, P.L.N. (1963) Indian J. Chem., 1, 276–277. Batova, A., Lam, T., Wascholowski, V., Yu, A.L., Giannis, A., and Theodorakis, E.A. (2007) Org. Biomol. Chem., 5, 494–500. Chantarasriwong, O., Cho, W.C., Batova, A., Chavasiri, W., Moore, C., Rheingold, A.L., and Theodorakis, E.A.
References
101. 102. 103.
104. 105. 106.
107.
108.
109. 110.
111.
112.
113. 114.
115. 116.
(2009) Org. Biomol. Chem., 7, 4886–4894. Dewick, P.M. (1998) Nat. Prod. Rep., 15, 17–58. Knaggs, A.R. (2003) Nat. Prod. Rep., 20, 119–136. Herrmann, K.M. and Weaver, L.M. (1999) Annu. Rev. Plant Biol., 50, 473–503. Beerhues, L. and Liu, B. (2009) Phytochemistry, 70, 1719–1727. Gottlieb, O.R. (1968) Phytochemistry, 7, 411–421. Dewick, P.M. (2009) Medicinal Natural Products: A Biosynthetic Approach, 3rd edn, John Wiley & Sons, Inc., Hoboken, New Jersey. Locksley, H.D., Moore, I., and Scheinmann, F. (1967) Tetrahedron, 23, 2229–2234. Carpenter, I., Locksley, H.D., and Scheinmann, F. (1969) Phytochemistry, 8, 2013–2025. Bennett, G.J. and Lee, H.-H. (1988) J. Chem. Soc., Chem. Commun., 619–620. Bennett, G.J., Lee, H.-H., and Das, N.P. (1990) J. Chem. Soc., Perkin Trans. I, 2671–2676. Kartha, G., Ramachandran, G.N., Bhat, H.B., Nair, P.M., Raghavan, V.K.V., and Venkataraman, K. (1963) Tetrahedron Lett., 4, 459–472. Quillinan, A.J. and Scheinmann, F. (1971) J. Chem. Soc., Chem. Commun., 966–967. Yemul, S.S. and Rama Rao, A.V. (1974) Org. Mass Spectrom., 9, 1063–1072. Han, Q., Yang, L., Liu, Y., Wang, Y., Qiao, C., Song, J., Xu, L., Yang, D., Chen, S., and Xu, H. (2006) Planta Med., 72, 281–284. Nicolaou, K.C. and Li, J. (2001) Angew. Chem. Int. Ed., 40, 4264–4268. Tisdale, E.J., Slobodov, I., and Theodorakis, E.A. (2003) Org. Biomol. Chem., 1, 4418–4422.
117. Li, N.-G., Wang, J.-X., Liu, X.-R., Lin,
118.
119.
120. 121. 122. 123.
124. 125.
126. 127. 128. 129.
130.
131.
132. 133.
134.
C.-J., You, Q.-D., and Guo, Q.-L. (2007) Tetrahedron Lett., 48, 6586–6589. Tisdale, E.J., Slobodov, I., and Theodorakis, E.A. (2004) Proc. Natl. Acad. Sci. U.S.A., 101, 12030–12035. Nicolaou, K.C., Xu, H., and Wartmann, M. (2005) Angew. Chem. Int. Ed., 44, 756–761. Gajewski, J.J. (1980) Acc. Chem. Res., 13, 142–148. Ganem, B. (1996) Angew. Chem. Int. Ed., 35, 936–945. Ganem, B. (1996) Angew. Chem., 108, 1014–1023. Severance, D.L. and Jorgensen, W.L. (1992) J. Am. Chem. Soc., 114, 10966–10968. Gajewski, J.J. (1997) Acc. Chem. Res., 30, 219–225. Tanford, C. (1980) The Hydrophobic Effect, 2nd edn, John Wiley & Sons, Inc., New York. Breslow, R. (2004) Acc. Chem. Res., 37, 471–478. Lindstr¨om, U.M. (2002) Chem. Rev., 102, 2751–2772. Grieco, P.A. and Kaufman, M.D. (1999) J. Org. Chem., 64, 6041–6048. Hayden, A.E., Xu, H., Nicolaou, K.C., and Houk, K.N. (2006) Org. Lett., 8, 2989–2992. Nicolaou, K.C., Sasmal, P.K., and Xu, H. (2004) J. Am. Chem. Soc., 126, 5493–5501. Bhamare, N.K., Granger, T., John, C.R., and Yates, P. (1991) Tetrahedron Lett., 32, 4439–4442. Bichan, D.J. and Yates, P. (1972) J. Am. Chem. Soc., 94, 4773–4774. Tisdale, E.J., Chowdhury, C., Vong, B.G., Li, H., and Theodorakis, E.A. (2002) Org. Lett., 4, 909–912. Chantarasriwong, O., Batova, A., Chavasiri, W., and Theodorakis, E.A. (2010) Chem. Eur. J., 16, 9944–9962.
467
469
Part III Biomimetic Synthesis of Polyketides
Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
471
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings Gr´egory Genta-Jouve, Sylvain Antoniotti, and Olivier P. Thomas
Out of intense complexities intense simplicities emerge. Winston Churchill
13.1 Introduction
Polyketides represent one of the most complex groups of natural products, exemplified by renowned families of compounds such as polyphenols, macrolides, polyethers, and polyenes. To date, around 10 000 polyketides have been characterized and they are recognized as useful templates for the discovery and the design of new drugs [1]. Despite the fact that highly complex structures are frequent, a striking characteristic of this family of compounds arises from the structural simplicity of their biosynthetic precursors and of their assembly. The understanding of polyketide biosynthesis during the last two decades was facilitated by the relative simplicity of these iterative biosynthetic steps. Because many polyketides are produced by microorganisms their biosynthetic enzymes and genes were among the first to be identified [2, 3]. Polyketide synthases (PKSs) belong nowadays to the best known and described polyenzymatic complexes and their huge potential to afford polyketides diversity was reviewed in 2009 by Hertweck [4]. A few years earlier, a comprehensive review by Staunton et al. related the history and key achievements in the discovery of the logic of polyketide biosynthesis [5] and also in 2001 Whiting reviewed the chemistry of natural phenolic compounds across the twentieth century [6]. In the first part of this chapter, we will therefore focus on biomimetic polyketide synthesis. As we will see in the second part of this chapter, the original idea that some aromatic compounds may be derived from linear polyketone chains first came from Collie in the late nineteenth century [7]. Even if the main interest of Robinson’s laboratory in Oxford focused on alkaloid and terpene biosynthesis, this idea was later supported by Robinson himself and particularly by a former fellow of his laboratory, Arthur Birch, in the 1950s. He was the first to postulate and demonstrate, using isotopically enriched precursor, that acetic acid was the sole Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
472
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings
molecule responsible for the construction of linear polyketone chains giving rise to aromatic natural products such as 6-methylsalicylic acid [8]. As a consequence, the Claisen condensation appeared as the key reaction leading to the creation of the C–C connections between acetic acid units. Birch’s discovery triggered a high number of scientific reports that all showed that acetic acid not only lead to aromatic compounds but also to structurally more complex polyketides. Indeed, a succession of reductive, oxidative, and cyclization processes of the linear polyketone chain would be able to afford the high diversity of the polyketide family. These simple chemical processes inspired a flurry of synthetic organic methodologies trying to mimic the natural strategy to synthesize complex bioactive molecules. The first work reported by Claisen in 1881 described the condensations of ketones with aldehydes under strongly basic conditions [9], but the Claisen condensation is now well recognized as a C–C bond formation between two esters leading to a β-keto ester. Because these ‘‘synthetic’’ conditions were not consistent with ‘‘natural’’ enzymatic conditions it was further discovered that two additional steps were selected during evolution to lower the pKa at the α-position of the acid derivative to facilitate this condensation. First, acetic, propionic, or butyric acids are linked to coenzyme A (CoA) or to the thiol residue of a polyenzymatic complex through a thioester bond, which induces a decrease of approximately two pKa units compared with an ester. In this chapter, we will detail in-depth the results of chemical studies using thioesters for the Claisen condensation. A second development of biological systems to facilitate this C–C bond formation was a key carboxylation at the α position of the acyl derivatives, leading to highly reactive malonyl derivatives. This activation was found to be catalyzed by a carboxylase and biotin appeared as the cofactor transferring carbon dioxide to this position [10]. Here also we will focus mainly on synthetic strategies using malonyl derivatives. In the infancy of polyketide biomimetic chemistry, pioneer chemists developed reactions mainly aimed at mimicking Nature and improving the understanding of metabolic pathways. Later on, the logic of biomimetic synthesis also inspired organic chemists in the design of new synthetic methodologies often applied to the total synthesis of complex bioactive products. The examples presented in this chapter illustrate these two aspects of biomimetic polyketides and polyaromatic natural products.
13.2 Polyketide Assembly Mimics
The mechanism of a polyketide assembly between two acetyl units has been studied extensively and is closely related to fatty acid assembly. The C–C connection originates from a decarboxylative Claisen condensation between one nucleophilic malonyl unit linked to the polyenzymatic complex by a thioester bridge and a second electrophilic acetyl derived thioester leading to a β-keto thioester (Scheme 13.1). Nevertheless, the exact sequence of the mechanistic steps is not fully accepted and
13.2 Polyketide Assembly Mimics O
O R
O
SCoA
O
O R
SCoA
AT
O OH
MAT
KS
TE
ACP
KS
ACP
S
S
CO2 S R
S O O
Scheme 13.1
O
O O
R
O
Mechanism of the key C–C connection between two units of a polyketide.
diverse experimental proof supported three distinct mechanisms: concerted, stepwise addition–decarboxylation, and stepwise decarboxylation–addition. We will detail later in this chapter how biomimetic approaches can help in understanding the real biosynthetic events. PKSs are complex polyenzymatic systems closely related to fatty acid synthases (FASs) [4]. The larger chemical diversity produced by PKS is due to a broader tolerance toward the precursors but also to the absence of some reductive steps always present for FASs. PKSs have been classified in several types depending on their global architecture and function. Iterative and non-iterative type I PKS mainly produce non-aromatic compounds, including a large family of bioactive macrolactones represented by erythromycin (1) (Figure 13.1). In type II PKS, a unique module composed of three domains is used in an iterative manner to yield most of the aromatic and polyaromatic polyketides exemplified by doxorubicin (2). Finally, type III PKS can metabolize a broad range of starter units followed by various iterative steps and cyclization, often leading to aromatic compounds like tetrahydrocannabinol (3). For type I multimodular PKS, each module is responsible for the addition of an acetyl or propionyl unit. Several domains are constitutive of a module: a ketosynthase (KS), an acyl carrier protein (ACP), and optionally an acyl transferase (AT) O O
OH
HO OH
N HO O
O O
O O
Erythromycin A (1)
Figure 13.1
O
OH
OH OH
H
O O
O
O
H
OH O O
OH Doxorubicin (2)
OH
O NH2 OH
Tetrahydrocannabinol (3)
Structures of three polyketides produced by the three types of PKS.
473
474
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings
and malonyl acyl transferase (MAT). Iteration of the process leads to polyketones and finally a chain termination domain (TE) allows the release of the product as a carboxylic acid or a lactone. The high diversity of polyketides arises from the presence of additional domains in some modules of the PKS. The simplest modules with no reductive step lead to a β-keto acyl derivative (Scheme 13.2). To classify the biomimetic reactions further reported we will name this transformation a ‘‘type-a reaction.’’ A ‘‘type-b reaction’’ will give directly the β-hydroxy acyl derivative, involving an additional reductive process due to the presence of a ketoreductase (KR) domain in the PKS module. Furthermore, the presence of a dehydratase (DH) domain will allow the biosynthesis of an α, β-unsaturated acyl derivative, classified as a ‘‘type-c reaction’’ in biomimetic approaches. Finally, an enoyl reductase module (ER) could be present to saturate the chain by a formal hydrogenation. This last process is mostly representative of the FAS and biomimetic access to this connection type is beyond the scope of this chapter. Type a reaction
O R
+ − SCoA O
O
O
KS
O
ACP
AT
O
R
SCoA
OH
Type b reaction KR
O R
+ − O SCoA
O
O
KS
OH
ACP
AT
O
R
SCoA
OH
Type c reaction KR DH
O R
+ − SCoA O
O
O
R
SCoA
KR O R
+ − SCoA O
Scheme 13.2
O
O
KS AT SCoA
O
ACP
KS AT
DH
OH
ER O
ACP R
OH
The four types of product formation for a PKS module.
As a consequence, this section is divided according to the type of PKS assembly the synthetic chemist plans to mimic. In the case of a type-a mimic, the electrophile entity can be chosen among the acyl derivatives, whereas for type-b and -c mimics the acyl derivative should be replaced by an aldehyde or an equivalent to keep the corresponding oxidation state in a one-step procedure. In all these cases, several nucleophiles were tested; we will first focus on the most ‘‘biomimetic’’ reactions corresponding to a decarboxylative Claisen condensation using malonic acid half-thioesters (MAHTs).
13.2 Polyketide Assembly Mimics
13.2.1 Type-a Mimics
As early as 1967, Lynen proposed a biochemical process based on an acyl transfer from a MAHT to an enzyme bound thioester as the key step, explaining a polyketide unit assembly [11]. Nucleophilic activation of the acyl through biotin carboxylation was proven to occur in the biosynthetic assembly of a polyketide unit. We have thus decided to first detail biomimetic studies using malonyl derivatives as nucleophiles and, in a second part, condensations that do not require a decarboxylative step. 13.2.1.1 Malonyl Activation The first experiments designed to mimic a type-a PKS module were reported in 1975 by Scott et al. The acyl transfer was tested on a chemical template to mimic the enzymatic machinery and to facilitate the transfer through an intramolecular process. To this end, Scott et al. used catechol derivative 4 with both hydroxyl groups acylated, one by an acetyl and one by a malonyl group [12]. Strongly basic conditions in the presence of isopropylmagnesium bromide allowed them to isolate the modified acyl-transferred catechol derivative 5 in 30% yield (Scheme 13.3).
MgBr
O O
O
O
4
(2 eq.)
HO
THF, r. t., 3 h.
5 30%
O
O
O + CO2 O
HO
Scheme 13.3
Scott’s conditions for the intramolecular acyl transfer.
The intramolecular nature of the acyl transfer was demonstrated and chelation of the magnesium cation with the malonate in 6 was assumed to be essential for control of the C-acylation over O-acylation (Scheme 13.4).
O
6
O O
O O−
O
Scheme 13.4
Mg Br
O− Mg Br 7
+ CO2
O O O
Proposed mechanism for Scott’s acyl transfer.
Three years later, Kobuke and Yoshida studied an analogous intermolecular reaction using the acyl and malonyl thioesters 8 and 9, respectively, which could be considered closer to AcylCoA and MalonyCoA and therefore more biomimetic than the catechol derivative 4 [13]. They were able to isolate and characterize the β-ketoacyl derivative 10 (60% yield), the product of a type-a condensation between both units, using catalytic Mg(II) salts and imidazole as a soft organic base (Scheme 13.5). This study evidenced the major role played by the thioester present in the polyenzymatic systems. By lowering the pKa of the α hydrogen, this
475
476
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings Mg(OAc)2 (cat.) N
O
O
+
O SnBu
HO
SPh 8
O
N H THF, r. t., 87 h.
O SnBu
+ CO2
10
9
60%
Scheme 13.5 Biomimetic conditions of Kobuke and Yoshida, using thioesters, imidazole, and catalytic Mg(II) salts.
acyl derivative allows the use of a soft organic base and a catalytic amount of metal, avoiding harsh conditions and getting closer to real biosynthetic mimics. Worth noting is the use of imidazole as a base, which improves the biomimetic character of the process by emulation of histidine-rich PKS. Even if the catalytic sites of PKS were not described as containing metals, the presence of Mg(II) salts can help in the decarboxylation step through coordination with the malonyl group (Scheme 13.6). PKS
PKS
(a)
N
N N H
HN O
ACP
N H
O
S
O
O ACP
H −CO2
S
Mg
HN
HN O
N
O
n
Bu
S
O
H
+
−
KS S
(b)
NH
O
S
O KS
HN
N
O n
−CO2
Bu
O
NH +
S
O Ph
S
Ph
S
−
Scheme 13.6 Proposed mechanism for the condensation between thioesters in: (a) the active site of a PKS and (b) the biomimetic conditions of Kobuke and Yoshida.
At the same time, Masamune extended the scope of this reaction to other bi-substituted malonyl and imidazolyl acyl derivatives [14]. Subsequently, no significant advance in this field was made during the next 20 years until Matile and coworkers found that, under similar conditions, the self-condensation of a malonyl thioester was possible when derived from a para-methoxythiophenol such
13.2 Polyketide Assembly Mimics
as 11, increasing at the same time the acidity of the α-proton [15]. In this case, even if Mg(OAc)2 remained the most potent metal catalyst, 5-nitrobenzimidazole was more efficient than imidazole as an organic base and could be used in catalytic amounts to afford the self-condensed product 12 in 34% yield, along with the non-decarboxylated precursor 13 (37%), both issuing from the Claisen self-condensation (Scheme 13.7). Mg(OAc)2 (cat.), O2N O
O
O S
HO 11
N N (cat.) H
THF, r. t., 48 h.
O
O
O SAr
12 34%
+
O
HO
O SAr
13 37%
Scheme 13.7 Matile’s biomimetic conditions for the self-condensation of malonyl thioesters.
Given the frequency and the amounts of some by-products of decarboxylation or hydrolysis of the thioesters isolated, the authors concluded that ‘‘Claisen self condensation toward polyketide was questionable’’ under similar conditions. Mimicking a long evolutionary polyenzymatic complex could not be an easy task! Decarboxylative intramolecular Claisen condensations were also studied by changing the template. The group of Harisson published several reports on the use of glycoluril to promote the acyl transfer. One of the reports focused on the description of specific conditions for a malonyl condensation [16], but most of their work showed that activation by a malonyl was not necessary and we will therefore detail their results in the following section. The use of MAHT as nucleophiles was particularly adequate for soft and biomimetic reaction conditions, nevertheless their preparations were sometimes troublesome and synthetic chemists mostly developed methodologies based on non-activated acyl derivatives. 13.2.1.2 Without Malonyl Activation Because the presence of a carboxylic acid substituent increases the acidity of the α-proton of the malonyl by around two pKa units, an acyl transfer with a simple acyl nucleophile requires strong basic conditions. Important advances in this field were obtained by the group of Harisson, who used glycoluril 14 as a template for an intramolecular acyl transfer [17]. The original properties of this template allowed them to repeat the acyl transfer in several subsequent iterative steps, culminating in the addition of four C2 units in the example of adduct 15 (Scheme 13.8) [18]. This approach was applied to isotopically 2 H and 13 C labeled acyl derivatives, which allowed the synthesis of these important labeled precursors used for biosynthetic studies [19]. Kinetic and mechanistic studies were undertaken to understand the regioselectivity of the reaction during the acyl transfer [20]. All these approaches were dedicated to the construction of β-polyketones and they belong to the type-a mimic class. One of the conclusions that can be underlined
477
478
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings
N
N O
t
O N
N O
O 14
BuOLi
THF, 0 to 25 °C, 20 min.
N
N
O
O N
O O
N H
N
N
N
N
O
O
O
O
16 93%
17 50% t
BuOLi
N
N
N H
N
O
O O O
18 60%
N
N
N H
N
O
O
15 5%
O O
Scheme 13.8 Harisson’s intramolecular iterative acyl transfer using glycoluril 14 as a template.
is the subtle balance of the conditions necessary to mimic the catalytic site of an enzyme. The detailed work of the group of Harisson in this field is worth noting, which also lead to labeled polyketones. Another challenging issue was to obtain β-hydroxyacyl derivatives in a unique one-pot procedure, thus mimicking a type-b PKS module. 13.2.2 Type-b Mimics
In this case the electrophile entity is an aldehyde or an equivalent that can undergo a nucleophilic addition under mild conditions to afford the desired β-hydroxyacyl product. In this case also, activation of the nucleophilic acyl moiety has been studied. A key point of these studies came from the control of the stereochemistry of the nucleophilic addition, an enzyme catalysis leading to a unique stereoisomer during the formation of the stereogenic center. Aldol-type reactions using acid
13.2 Polyketide Assembly Mimics
derivatives have been greatly studied and reviewed [12]. In this chapter we will only detail biomimetic reactions using malonyl half thioesters (MAHT) or unactivated thioesters that were recognized as very soft nucleophile for an aldol addition. 13.2.2.1 Malonyl Activation A first catalytic procedure that did not require in situ aldolate functionalization, or an excess of nucleophile, was developed by the group of Shair with MAHT in 2003 [22]. In this case, catalysis with the magnesium salts previously used for type-a mimics were not efficient and after a small screening Cu(II) salts were identified as the best catalysts for the addition of MAHT 19 on aldehyde 20, affording the β-hydroxythioester 21 in good yield. Benzimidazole derivatives were found to be necessary as an additional soft organic base for this condensation (Scheme 13.9). Cu(2-ethylhexanoate)2 (cat.), O 2N N O H
O
+
O
HO
20
Scheme 13.9
S 19
N (cat.) H
O S
OH + CO2
THF, r. t., 3.5 h. 21 85%
Shair’s catalytic aldol condensation of aldehydes with a MAHT.
Worth noting are the exceptionally mild conditions of this reaction, which was performed at room temperature and which did not require dry solvents and vessels usually necessary for a catalytic aldol reaction. Increasing the electrophilicity of the aldehyde with electron-withdrawing substituents even improved the yields of the desired products. Interestingly, the reaction was successfully applied to MAHT substituted at the α-position by a methyl, thus mimicking a propionate insertion in a polyketide chain. In this case, the diastereoselectivity was largely in favor of the syn products, which renders this reaction even more interesting. Subsequent studies on this important reaction focused on the enantioselectivity that could be induced by the presence of a chiral ligand coordinating the metal catalyst. The first report of a small enantiomeric excess using Cu(II) triflate salts and a chiral bisbenzimidazole derived from tartric acid was reported by the group of Cozzi [23]. At the same time, the group of Shair greatly improved diastereoselective and enantioselective excesses of 22 with methyl MAHT 23 as the nucleophile and a bisoxazoline ligand associated with the copper(II) triflate catalyst (Scheme 13.10) [24]. The presence of the thioester proved to be critical for the nucleophilic addition to proceed under very mild condition and allow further modifications to access to more complex structures. Shair and coworkers underlined a significant compatibility of the reaction conditions with several substrates, including substrates bearing protic functional groups that are often troublesome for such type of additions. Mechanistic insights into polyketide biosynthesis were also obtained working in these biomimetic conditions [25]. It appeared indeed much easier to detail and interpret all the kinetics parameters of such a simplified reaction than to study these
479
480
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings Cu(OTf)2 (cat.), O
O
O
O
+
H
N
O
Ph
S
HO
24
O
N
(cat.)
Ph
Toluene/acetone 9:1, r. t., 24 h.
23
OH
S
+ CO2
89% syn/ anti 11:1 93% ee
22
Scheme 13.10 Shair’s asymmetric catalytic aldol condensation of aldehydes with methyl MAHT.
parameters in a complex polyenzymatic system. Kinetic experiments using labeled precursors suggested that decarboxylation of the malonyl moiety occurred after a preliminary addition and not before, as usually stated for FASs and polyketases (Scheme 13.11) [26]. Fatty acid and Polyketide Biosynthesis: Enolate Formation by Decarboxylation His O
S ACP H N His
His H
NH
N
S
O
O
NH
HN
ACP O
H N His
N
+ CO2
N
MAHT aldol condensation: Enolate Formation prior to Decarboxylation H O
S O
S
O
B
+
O O
Cu L
Scheme 13.11
+ BH
O Cu
L
L
L
Two distinct mechanisms of enolate formation before nucleophilic addition.
A metal-free decarboxylative addition of MAHT to activated carbonyl compounds was further developed by Fagnou and coworkers [27]. For the first time, they were able to extend a decarboxylative Claisen condensation to malonic acid half oxoesters (MAHOs) 25 without the use of a metal catalyst, and the ketol product 26 was formed in good yield in the presence of triethylamine starting from ethyl pyruvate (27) at room temperature (Scheme 13.12).
O O 27
+ HO
O
Scheme 13.12
O
O
O
Et3N
O
THF, r. t., 4 h.
OH O
O
+ CO2
O 25
26
82%
Fagnou’s metal-free decarboxylative condensation with a MAHO.
13.2 Polyketide Assembly Mimics
481
These metal-free conditions also allowed a deeper understanding of the reaction mechanism in particular by diffusion ordered spectroscopy (DOSY) NMR experiments. Identifying the post-addition/pre-decarboxylation adduct 28 by NMR for both MAHT and MAHO, the authors undoubtedly demonstrated that the stepwise addition–decarboxylation was the unique mechanism possible for this reaction, leading to the same conclusions as the group of Shair. They were also able to demonstrate the reversibility of the first addition step and kinetic studies revealed the higher reactivity of MAHT over MAHO (Scheme 13.13). As a consequence, these results shed some light on the real biosynthetic mechanism. PKS
NH
N
O
O
H
ACP
S
O KS
NH HN
HN
N
O
S
N
H
HN O
PKS
NH
N
H ACP
PKS
NH
O
OH O ACP
O
−CO2
S
O KS
S
KS S
S
(a)
N H O Ph
O
O O
H
O Ph
O
O
O O
O
O
HNEt3 Ph −CO2
HNEt3 O
28
HNEt3
O O
O
O
O (b)
N
O
O
HNEt3
Scheme 13.13 Stepwise addition–decarboxylation for the C–C connection: (a) in polyketide biosynthesis; (b) under Fagnou’s conditions.
This result is highly significant as it is well known that PKS enzymes do not require a metal ion to perform the C–C condensation (see, for example, Reference [28]). As extensions of these important catalytic decarboxylative Claisen condensations with MAHT to organic synthesis, the carbonyl group has been replaced by other unsaturated electrophiles that undergo, in the same manner, nucleophilic addition for the creation of important C–C bonds. As striking examples, a catalytic decarboxylative condensation of MAHT to imines has been found to be catalyzed by chiral Cinchona alkaloids, leading to important chiral β-aminothioesters [29]. Other similar conditions were developed for the asymmetric 1,4-decarboxylative addition of MAHT to nitroolefins, also catalyzed by chiral Cinchona alkaloids [30]. Indeed, the group of Wennemers was inspired by the cysteine, histidine, and asparagine triad described in the catalytic active site of
NH
482
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings Cys His
O HN
N
N H
R
N O O
NH
O
H N H
Asn
S
H O
O
O
N
O
S
O
N
H
O S
R
O
ACP
(a)
(b)
F3C
CF3
Scheme 13.14 Rational design behind an asymmetric and organocatalytic addition of MAHT to nitroolefins by Wennemers: (a) the cysteine, histidine, and asparagine triad catalytic active site of some PKS enzymes; (b) the designed organocatalyst.
some PKS enzymes to develop an organocatalytic and asymmetric version of the addition of MAHT to an appropriate electrophile (Scheme 13.14). In ethyl vinyl ether as solvent the reaction is performed at room temperature to afford the addition product in around 60% yield and with an enantiomeric excess of ca. 80%. 13.2.2.2 Without Malonyl Activation Thioesters alone have long been recognized as important enolizable substrates for nucleophilic addition to several functional groups, and malonyl activation is not always necessary if strong bases are used [31]. In a biomimetic context, the group of Coltart developed an aldol addition of non-activated thioesters 29 under very mild conditions, leading to β-hydroxythioester 30 [32]. Here also, choosing a thioester was justified by the increased acidity of the α-proton, which avoids the use of strong bases like lithium diisopropylamide (LDA) or t-BuOK. The method involved an organic amine base and Mg(II) salts (Scheme 13.15). O
MgBr2.OEt2 i-Pr2NEt
O H +
31
Scheme 13.15
S 29
CH2Cl2, r. t., 30 min.
OH O S 30 96%
Coltart’s aldol addition with non-activated thioester.
Importantly, Mg(II) and Cu(II) salts emerged as the best Lewis acids to promote the condensation. Even if catalytic conditions were not developed at this stage, the wide scope of this reaction allowed several extensions to C–C bond forming reactions for organic chemists. Type-a mimics have also been studied under these conditions, but to a lesser extent, and to activate the nucleophile N-acylbenzotriazoles and pentafluorophenyl esters are necessary to give access to 1,3-diketones in good yields [33]. As another extension of this work, an asymmetric version of the Mannich condensation of thioesters was developed by this group using Cinchona alkaloids as chiral organocatalysts [34]. Also avoiding enolate
13.2 Polyketide Assembly Mimics
formation through malonic activation, the group of Barbas III developed an asymmetric version of Michael and Mannich additions of thioesters to α, β-unsaturated aldehydes catalyzed by chiral proline derivatives [35, 36]. To increase the acidity of the α-proton they used the S-trifluoroethyl thioester 32, which reacted with the Michael acceptor 33 to yield the addition product 34 with a high enantioselectivity (Scheme 13.16). F3C
CF3 CF3
N H
O F3C
O
+
S
Cl
O CF3
F3C
S
O
PhCO2H (10 mol%)
33
32
O
(10 mol%)
MeOH, r. t., 17h.
34 70% Cl anti :syn 65:35 90% ee (anti)
Scheme 13.16 Barbas III asymmetric and organocatalytic addition of thioesters to Michael acceptors.
The aldol reaction is not restricted to the use of thioester as nucleophiles and numerous other acyl derivatives, even in a chiral environment, have been developed. The Evans oxazolidinones belong to the most renowned examples in this field, but, in this case, the biomimetic nature of the addition is not so relevant [20–37]. 13.2.3 Type-c Mimics
The last challenge in the biomimetic C–C assembly of two acyl units leading to polyketides was to obtain a dehydrated acyl product (type-c mimics). The condensation product can be obtained starting from an aldehyde and a subsequent dehydration affords the unsaturated acyl derivative. These studies are particularly challenging as, in a one-pot procedure, the synthetic chemist would mimic a complex module consisting of five domains. As above, we will mainly address reports on biomimetic conditions using thioesters as nucleophiles. With a malonyl diester electrophile in the presence of piperidine under strong conditions the reaction is known as the Knoevenagel condensation, leading to an α-olefinic diester moiety [38]. The Doebner modification includes the use of pyridine with a mono- or dicarboxylic acid and a subsequent decarboxylation step yields an unsaturated acyl derivative [39]. Application of this strategy to MAHO often leads to a mixture of (Z) and (E) stereoisomers as well as α, β and β, γ unsaturated regioisomers. Control of the selectivity is troublesome and the harsh conditions required for this condensation prompted new studies for this reaction [40]. A green process using simultaneous microwave and ultrasonic irradiation was reported for such applications [41].
483
484
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings
Very few studies focused on the development of mild biomimetic conditions to promote this condensation. A very interesting catalytic version using 4-dimethylaminopyridine (DMAP) as the sole catalyst was developed by List and coworkers in 2005, featuring high selectivity in favor of the α, β unsaturated 35 compound with an (E) configuration, starting with aldehyde 36 and MAHO 37 (Scheme 13.17) [42]. O
O O H
+ HO2C
36
O
DMAP (cat.)
O
35
DMF, r. t., 5-48 h.
37
Scheme 13.17
+ CO2
91% E :Z 95:5
List’s condensation of MAHO in a type-c mimic.
To obtain very soft conditions for this condensation, a catalytic version of this procedure in wet THF was proposed with phenylacetaldehyde (38) as electrophile and MAHT 39 as nucleophile [43]. After a rapid screening of several metallic catalysts, Yb(OTf)3 appeared as the most effective to obtain the desired dehydrated product, with Cu(II) salts giving mostly the aldol product as described by Shair. The use of 5-methoxybenzimidazole as the base afforded a very high regioselectivity in favor of the β, γ unsaturated thioester 40 with an (E) configuration. Interestingly, the regioselectivity was inverted in the case of aliphatic aldehyde 41, suggesting thermodynamic control of the reaction (Scheme 13.18). H
38
S
O
HO2C O +
S
Bn Yb(OTf)3 (cat.)
39
O
O 40
Bn + CO2
75%
N N H
41
Scheme 13.18
O
H HO2C +
THF, r. t., 16 h.
O S
S
Bn
39
42
32%
O
Bn
+ CO2
A type-c reaction mimic with MAHT.
In this case also, the mechanism was suggested to proceed via a first addition step followed by decarboxylation of the reactive intermediate. To get further insight into the mechanism of this condensation, the role of the metal cation should be clearly assessed but as usually described the decarboxylation could be concomitant with the dehydration step. To date, no work has been reported on a condensation of type-c without malonyl activation. This first part has focused on the biomimetic methods developed by synthetic chemists to mimic the assembly of acyl units for the construction of important natural products. In biological systems, the long alkyl chain is released after a cyclization or a simple hydrolysis. Historically, the first biosynthetic studies
13.3 Biomimetic Access to Aromatic Rings
on polyketides were performed on aromatic compounds formed mainly by type II and III PKS of non-reduced β-polyketone chains. As a first consequence, a retrobiosynthetic approach would consider these aromatic end-products as masked 1,3-dicarbonyl precursors. Indeed, Birch reduction of a mono- or disubstituted benzene derivative 43 affords a 1,4-cyclohexadiene 44 in a regioselective manner, further leading to the 1,3-dicarbonyl compound 45 after subsequent ozonolysis of both double bonds (Scheme 13.19). Applications of this approach to the biomimetic syntheses of natural products were reviewed by Hilt et al. in 2009 [44]. R
R
R Li
1. O3
liq. NH3
43
O
2. Pd/C, H2
44
Scheme 13.19
O 45
O +
O 46
Birch reduction–ozonolysis reaction sequence leading to 1,3-dicarbonyls.
We will detail in the second part of this chapter the direct biomimetic constructions of aromatic mono- and polycyclic polyketides starting from β-polyketones. 13.3 Biomimetic Access to Aromatic Rings
The first biomimetic reaction leading to an aromatic polyketide dates back to the late nineteenth century when James Collie obtained orcinol (47) by boiling 4-pyranone 48 in a barium hydroxide solution (Scheme 13.20) [7]. For the first time, linear polyketone 49 was proposed as an intermediate in the cyclization and aromatization into 47 but this result remained mostly ignored by the scientific community at the time of its disclosure.
−H2O
Ba(OH)2
O 48 Scheme 13.20
OH
O
O
O O 49
OH Orcinol (47)
First biomimetic access to an aromatic ring by Collie and coworkers.
More than 50 years after, the group of Birch suggested that acetic acid could be the unique precursor of the alkyl chains of β-polyketones. He was further able to prove the role of these alkyl chains in the biosynthesis of some aromatic compounds like 6-methylsalicylic acid using feeding experiments with 14 C labeled acetic acid [45]. In 1955, Robinson was able to present a general biosynthetic scheme for the construction of aromatic polyketides in his benchmark book The Structural Relations of Natural Products [46]. The biosynthesis of polyketides was further confirmed to be the result of iterative condensations of acetyl and malonyl units to form a polyketoacid that could undergo two types of cyclization: an aldol condensation (A) or a Claisen condensation (B) (Scheme 13.21) [47].
485
486
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings
SCoA
+ 3x
CO2H SCoA
O
SCoA O
O
O
O
O
aldol reaction A
B
Claisen reaction
Base O
O O O
O
SCoA
−
O
O
SCoA O
H
SCoA
O
OH
O
O −
H+
O −
O
O
O
SCoA
−SCoA
−H2O
O
O
O
SCoA O enolization hydrolysis
O
O −SCoA
O enolization
OH O
HO OH OH O orsellinic acid
Scheme 13.21
HO
OH
phloracetophenone
Two biosynthetic pathways leading to aromatic polyketides.
In 2001, Thomas presented a biosynthetic classification of fused-ring aromatic polyketides based on their mode of cyclization [48]. Indeed, fungi and streptomycetes were found to utilize distinct polyketide-folding strategies for the construction of polycyclic systems. The mode F folding utilized by fungi leads, for example, to islandicin (50) produced by Penicillium islandicum, whereas the mode S folding yields actinorhodin (51) produced by Streptomyces coelicolor (Scheme 13.22).
Mode F folding
Islandicin (50)
Mode S folding
Scheme 13.22
Actinorhodin (51)
Two modes of folding leading to polycyclic aromatic polyketides.
13.3 Biomimetic Access to Aromatic Rings
Differences appear in the number of acetate units present in the first aromatic ring (left-hand side): two units for the mode F folding and three units for the mode S. No exception to this observation has been reported so far. Eukaryotes and prokaryotes are then characterized by distinct biosynthetic pathways and the mode F folding was more recently identified in plants [49]. These observations would have to be taken into consideration for biomimetic approaches. The group of Bringmann reported the first natural product proceeding from these two distinct modes of folding in two organisms. For the first time, convergence of these two distinct biosynthetic pathways was observed for chrysophanol (52) [50]. It was even shown that a third metabolic pathway was involved in the formation of this secondary metabolite. The anthracenoid 53, previously synthesized by a biomimetic approach, was proved to be an intermediate towards compound 52 following the mode S folding in a second strain of Streptomyces. These modes of cyclization appeared to be organism-specific, a fact of high evolutionary interest (Scheme 13.23) [51]. O Chrysophanol (52) OH O
OH Mode F, eukaryotes
O
O
O
O
O
O
O
O SACP
Mode S O
OH O 52
Mode S' Prokaryotes
OH
Scheme 13.23
O
OR OH O 53
O
OH O
OH
52
A third mode of cyclization towards chrysophanol (52).
At the same time, the first studies on biomimetic access to simple aromatic polyketides were reported. The major difficulties were closely associated with the synthetic routes leading to the preparation of long β-polyketones. Figure 13.2 presents an overview of important findings in this field. Since the 1970s, the synthetic effort moved from tri- up to dodeca-β-ketoacids by improvement of the acylation methods (Figure 13.2). 13.3.1 Biomimetic Access to Benzenoid Derivatives
Even though the cleavage of pyrones appeared as a first access to tri-β-carbonyl entities, two major routes were designed to obtain these important precursors (Scheme 13.24) [52]:
487
Figure 13.2
1893
Collie
O
O
O
R
O O O
Pentacarbonyl
O
1971 O R
R
O O O O
1973
Heptacarbonyl
O O O R
O
O O
O O
O O
OH OH
O
putative decacarbonyl acyclic precursor of tetracyclic ring system
O
O
???
Historical achievements toward the synthesis of β-polyketones and biomimetic aromatic polyketides.
1966
Ph OH Harris studies with triketoacids
O
488
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings
13.3 Biomimetic Access to Aromatic Rings O
O
O
1. Excess base
R
2. CO2
O OH
54 O
O
R 55
O
O
1. Excess base
OMe 56
Scheme 13.24
2. RCO2Me
O
O OMe
R 57
Alternatives for the synthesis of tri-β-carbonyl entities.
1) Carboxylation of β-diketones such as 54 into diketoacids of type 55; 2) acylation of the dianion of acetoacetic acid ester 56 into 57. One result of particular biogenetic significance was the cyclization of the thioacid derivative 58 into pyrone 59 obtained by the group of Harris (Scheme 13.25) [53]. Birch has been able to prove that thioacid 58 was much more prone to cyclize into resorcinol derivative 59 than the corresponding β-triketone [54]. O
O
O 2 NaNH2 liq. NH3
O
O
O
O
1. COS −
Na
+
−
Na
+
2. H+, H2O
SH 58
OH
O
O
59 42%
Scheme 13.25
Harris’ biomimetic access to pyrones.
Tetra-β-carbonyl compounds have attracted much more attention as they are the smallest polycarbonyl systems that can provide reasonable models of biological systems, giving access to benzenoid natural products. The use of LDA proved to be much more efficient in yielding the reactive trianion intermediates [55]. As described in Scheme 13.21, these intermediates can lead to a resorcinol or a phloroglucinol compound in addition to two other pyrones. The group of Harris was able to observe the formation of 6-phenyl-β-resorcylic acid (60) by cyclization of 7-phenyl-3,5,7-trioxoheptanoic acid (61) under mild ‘‘physiological’’ conditions at an appropriate acidic pH (Scheme 13.26) [56]. Esterifying the carboxylic acid resulted in another mechanism of cyclization. Indeed, when treated in a pH 8.5 buffered aqueous solution, ester 62 led exclusively to resorcinol derivative 63, while in an aqueous solution of potassium hydroxide at −5 ◦ C the same ester 62 gave a mixture of phloroglucinol 64 and resorcylic derivatives 63, 65, and 66. These results are in accordance with the fact that deprotonation of the acid derivative 61 can only lead to aldol products and that Claisen condensations require esterification of the carboxylic acid. Notably, in these
489
490
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings
O
O
O
OH
O
Ph
Aqueous buffer, pH 5
CO2H
25 °C,16 h.
OH 61
HO
CH2N2
O
O
O
OH
Phosphate buffer, pH 8.5
O
CO2Me
OMe
Ph
Ph 60 86%
HO
62
Ph 63 92%
KOH 2 M −5 °C, 19 h.
OH O
OH Ph +
HO OH 64 66%
HO
R Ph 34% R = CO2H 65 R = CO2Me 63 R = H 66
Harris’ biomimetic cyclization of β-triketoacids.
Scheme 13.26
cases, Mg(II) salts were not efficient acid catalysts for promoting the cyclization of the polyketide intermediates. An interesting study, developed by the groups of Scott and Money, and reviewed in 1970, used pyrones as masked tetraketides [57]. Under similar conditions, Crombie and James underlined the role of Mg(II) in performing the self-condensation of bispyrone 67 into phloroglucinol derivative 68, without reporting the yield (Scheme 13.27) [58]. Chelate formation and geometry prevents aldol-type cyclization and Claisen condensation provides benzenoid derivative 68. MeO Mg O
OH
O
O
O
O
O
O
Mg(OMe)2
OH O 67
O
CO2H
MeOH
OMe O
MeO2C O Mg
O
OMe
OH O MeO2C HO
OH 68
Scheme 13.27
Use of pyrones as masked tetraketide.
13.3 Biomimetic Access to Aromatic Rings
491
A similar synthetic strategy using 1,3-dioxin-4-one as masked tetra-β-carbonyl synthons was developed recently by the group of Barrett. They first reported the biomimetic syntheses of bioactive resorcylate lactones produced by the marine fungus Hypoxylon oceanicum LL-15G256 [59]. The strategy was based on a late-stage biomimetic aromatization into resorcylate derivatives. As an initial application, thermolysis of the 1,3-dioxin-4-one 69 and in situ trapping of the transient ketene 70 by the chiral alcohol 71 gave the tri-β-ketoester 72, which was not isolated due to a high instability (Scheme 13.28). Because the use of previous Harris conditions or analogous conditions at pH 9 were unsuccessful [60], aromatization was performed using base-catalyzed aldol condensation and subsequent addition of a strong acid. Ring-closing metathesis was thus applied to cyclize ester 73 into the natural (S)-(−)-zearalenone (74). This approach was further slightly modified to give access to (+)-montagnetol and (+)-erythrin [61]. O
O
HO O
O
O
O
O
O
O
O
71 O
O
O PhMe, 110 °C 70
69
72
O O
HO 73
(cat.) OH O
O
O Toluene, 80 °C
2. HCl, MeOH
Scheme 13.28
72 NMes
Cl Ru Cl O
OH O
O O
O
MesN
1. KOMe, MeOH
O
O HO S-(−)-zearalenone (74)
Barrett’s biomimetic total synthesis of zearalenone (74).
The outcome of the cyclization was even more difficult to control with non-protected pentacarbonyl intermediates, which were first synthesized by the group of Harris [62]. Coumarin derivative 75 was obtained in good yield starting from the unprotected compound 76 in slightly acidic or basic conditions. Upon using more basic conditions (KOH) this compound led to the other major aldol condensed product 77 (Scheme 13.29). Notably, no Claisen condensation product was isolated under these conditions even with the methyl ester of 76. At the same time, Scott and Money undertook similar studies with a three-rings fused pyranone [63]. Several other region-isomers were obtained but yields did not exceed 15%. An alternative approach towards protected hexaketides was developed by the group of Schmidt, where the methyl ether of triacetic lactone 78 underwent an acylation by the trianion of heptane-2,4,6-trione 79 to yield the monoprotected hexaketide-derived intermediate 80 [64]. In these conditions only the substituted chromone 81 was obtained, albeit in low yield (Scheme 13.30). Even though the transformation of the above-mentioned benzenoid compounds into naphthalenoids has been performed [65], no direct biomimetic transformations from penta-β-carbonyl intermediates could be performed, and other
O
492
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings
O
O OH
NaHCO3
O
O
O
O
OH
O
OH
HO
HO
O
O
OH
75 84%
76 OH
KOH
O
77 67%
HO HO
Scheme 13.29
Harris’ biomimetic cyclizations of pentacarbonyl derivative 76. O
O O + O 78
O
O
Li
− +
O −
O
O
O
−
Li + Li +
O
O
O
O
O O
80
O 81 19%
79
Scheme 13.30
Schmidt’s condensation of masked hexaketides.
strategies using hexa-β-carbonyl derivatives were necessary to afford naphthalenoid derivatives. 13.3.2 Biomimetic Access to Naphthalenoid Derivatives
A tris(pyrone), analogous to the bispyrone 67, was used by the group of Money as a masked hexacarbonyl derivative but neither benzenoid nor naphthalenoid products could be isolated, which definitely defined the limit of their approach. At the same time, the group of Harris was able to synthesize β-hexaketone 82 by double acylation of acetylacetone with the aryl β-ketoester 83 (Scheme 13.31) [66]. O
O
2x
O
+
O
O
1. n -C4H9Li (7 eq.) +
2. H
O
O
O
O
O
O
82 19%
83
Scheme 13.31 Preparation of β-hexaketones by polyanion acylation with β-ketoester monoanions.
Cyclization of hexaketide 84 could give rise to four naphthyl derivatives, but the reaction appeared to be highly regioselective. Depending on the basic or acidic conditions, naphthalenoid region-isomers 85 or 86 were the major products (Scheme 13.32). The approach of Harris culminated into the biomimetic total syntheses of 6-hydroxymusizin (87), as well as the related heterocyclic metabolites barakol (88)
13.3 Biomimetic Access to Aromatic Rings OH Ph
493
O Ph
HO
1. KOH 2. K2CO3
O
O
O
O
O
OH 85 70%
O OH OH O
84
Ph
SiO2
HO
Ph 86 92%
Regioselectivity in the naphthyl cyclization of β-hexaketones.
Scheme 13.32
and eleutherinol (89), using terminally protected hexaketone 90 and heptaketone 91 (Scheme 13.33) [67]. The mild conditions of all these reactions are of particular interest, confirming that biomimetic approaches could be useful methods by which to access natural products in good yields.
H2SO4 25 °C, 1 h.
O
O
O
O
O
O
O
OH O O i -Pr2NH
O
90
HO
HCl, acetone
O
O
92 80%
O
O
O
O HO 93 87%
O O Barakol (88) 80%
O
OAc OH O
1. Ac2O, pyridine 2. HCl, acetone
OH OH O
KOH 25 °C, 25 min.
AcO
HO 6-hydroxymusizin (87) 70%
94 55%
O OH O O
O
O
O
O
O
O
O
O
O
HO O
O
O Et3N
O
HO
91
O
O
95 57%
HO O
96 O
OH OH O 96
1. HCl, acetone 25 °C, 8 h. 2. i -Pr2NH HO Benzene, 25 °C, 5 min.
O OH O
TFA 25 °C, 30 min.
97
O
HO Eleutherinol (89) 19%
Scheme 13.33 Harris’ biomimetic syntheses of 6-hydroxymusizin (87), barakol (88), and eleutherinol (89).
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings
494
The group of Yamaguchi addressed this issue by a closely similar approach, accessing modified β-polyketide diester from the dianion of methyl acetoacetate (98) and hydroxyglutarate diester 99 (Weiler reaction) [68]. As an example, 1,8-naphthalenediol derivative 100 was obtained in good yield after cyclization of 101 in the presence of Ca(II) salts (Scheme 13.34) [69]. O
O OMe +2
HO O
OH OH OMe
OMe
OMe Li
THF, r. t., 2 h.
Na
O
HO
CO2Me
Ca(OAc)2
CO2Me
MeOH, reflux, 2 h.
O
100 50%
OMe
O
98
99
O
O
O
O
101
Scheme 13.34
Yamaguchi’s aromatic cyclization of polyketides into naphthalenoids.
An interesting application of this methodology led to the biomimetic synthesis of the naphthopyran polyketide (±)-nanaomycin A (102) (Scheme 13.35) [70]. An additional acetyl was introduced by a Claisen condensation between 100 and AcOtBu to yield 103, which underwent a reduction–lactonization–protection sequence leading to 104. Transformation of the lactone into the methyl ketone 105 was followed by deprotection and oxidation to give 102. O
OH OH CO2Me
LDA
MOMO
OH OH O
t-Bu
100
1. LDA
104
MOMO
O
O
O
OH O
O 105 92% HO
t-Bu
104 98%
1. HCl, MeOH
O
Scheme 13.35
t-Bu
OMOM O
2. Pd(OAc)2, PPh3, Et3NHCO2H
O
O
2. Et3N 3. MOMCl
O 103 96%
O
1. NaBH4
CO2Me
CO2Me THF, −78 °C to r. t.
OMOM O
O
2. DMDO, 18-crown-6 3. TFA, Et3SiH t-Bu 4. H2SO4 aq.
O
O
OH O Nanaomycin (102) 45%
Yamaguchi’s biomimetic synthesis of nanaomycine (102).
These biomimetic approaches have been further extended to the synthesis of important polycyclic aromatic polyketides, some of which exhibit high potential as antimicrobial compounds. 13.3.3 Biomimetic Access to Anthracenoid Derivatives
For longer and regular β-polyketones, control of the regioselectivity of the cyclization proved to be highly challenging and modification of the protection or modification of some carbonyl moieties was found to be necessary. The group of Harris generalized their approach with terminally protected heptapolyketones
13.3 Biomimetic Access to Aromatic Rings
495
such as 106 to perform the syntheses of the anthracenoids emodin (107) and chrysophanol (52) using an alternative approach for the synthesis of precursor 109 (Scheme 13.36) [71]. All these results could now been interpreted in the light of the biosynthetic considerations developed by the group of Thomas and Bringmann (Section 13.3).
O
O O
O
O
O
O
O
O
O
O
O
OH O
2. HCl, acetone
then CH2N2 O
106
O 1. NaOMe
O i -Pr2NH
O
110 10%
O
O
111
O
1. Me2SO4, K2CO3 Acetone, reflux, 2 h. 2. NaOMe, MeOH
O
3. HI, CH3CO2H 4. CrO3, CH3CO2H
OH O
OH
HO O Emodin (107) 20%
O O
N
O +
O
O
Li
OH OH O
O
2
O
O
O
OH OH O
O
112 24%
1. HCl CH3CO2H, 100 °C, 1 h.
O OH O
OH
2. CrO3
25 °C, 45 min.
114 88%
Scheme 13.36
O
109
NaOH
112
O
Li
108
113
N
OH CH CO H, 90 °C, 5 min. 3 2
O Chrysophanol (52) 89%
Harris’ biomimetic syntheses of emodin (107) and chrysophanol (108).
The biomimetic formation of the key aromatic intermediates 111 and 112 was followed by additional steps using conventional chemistry to achieve the biomimetic total syntheses. In both cases, the quinone functionality was introduced in the last step of the syntheses upon CrO3 oxidation of the central phenol. 13.3.4 Biomimetic Access to Tetracyclic Derivatives 13.3.4.1 Biomimetic Access to Tetracenoid Derivatives Tetracyclines have been recognized as very important antimicrobials and the group of Harris applied their methodology to the synthesis of pretetramide (115) and its methyl derivative, which are key biosynthetic intermediates toward
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings
496
O
O
O CO2tBu
O
O
O
1. TFA
O
OtBu
O
CO2t Bu 2. CH3COCl
O
then
118 92%
117
O
Li
N
O O
O
O OtBu
N
N
119 30%
N O N
O
O
O
OH O N
OH OH OH OH O
OH HI, P
119
NH2
CH3CO2H
120 70% O
OH
OtBu OH O
Pretetramide (115) 74% OH O OH
O NH2
HO
H
H
OH N
Tetracycline (116)
Scheme 13.37
Harris’ biomimetic synthesis of pretetramide (115).
tetracycline (116) (Scheme 13.37) [72]. They started from naphthenoid 117, a close analog of previously used glutarates whose reactivity was enhanced by using the anhydride form 118 obtained after dealkylative cyclization promoted by trifluoroacetic acid (TFA). A first addition of AcOtBu anion and subsequent dehydration was followed by the addition of the dianion of hydroxymethylisoxazole as a β-ketoamide equivalent. Acidic cyclization in the presence of phosphorus led to the desired pretetramide (115). The biosynthesis of pretetramide, featuring an amide function, which is unusual in the polyketides realm, involves the initial co-condensation of nine malonylCoA units. The elongation is followed by the action of an amidotransferase, introducing the amide function, to yield an enzyme-bound polyoxoamide further sequentially cyclized by several enzymes [73]. 13.3.4.2 Biomimetic Access to Tetraphenoid Derivatives An extension of the previous work of the group of Yamaguchi led to the biomimetic synthesis of the anthracenoid intermediate 121 starting from the diester 122 (Scheme 13.38). The use of aromatic diesters appeared as a simple procedure, based on Ca-promoted cyclizations, to allow the synthesis of the functionalized anthracenoid 121 following acetylation of the hydroxyl groups to obtain a stable product. After the partial synthesis of important naphthoquinones [74], one of the most prominent landmark accomplishments of this methodology led to the biomimetic synthesis of (−)-urdamycinone B (123), a potent antibiotic produced by Streptomyces fradie (Scheme 13.39) [75]. Access to the naphthol derivative 124 was performed following the same cyclization by Ca(II) salts. An additional acetyl unit was
13.3 Biomimetic Access to Aromatic Rings O CO2Me
O
OH OH
O
O +2
CO2Me
CO2Me Li
Na
OMe THF, r. t., 2 h.
O
122
CO2Me
Ca(OAc)2
CO2Me
MeOH, r. t., 2 h.
O OMe
O
497
Ac2O, DMAP Pyridine, r. t., 2 h.
OAc OAc CO2Me CO2Me 121 36%
Scheme 13.38
Yamaguchi’s access to anthracenoids.
O R HO O
OMe
O
1. THF, r. t., 2 h.
R
Li
R=
CO2Me
MeOH, reflux, 2 h.
Na
MOMO
CO2Me
OMe 2. Ca(OAc) 2
O
124 40%
R' =
OH OH
O LDA O
126 88% O MOMO
O
R' 3. MOMCl, Et N 3 O
H
THF, −78 °C, 1 h.
S
S
1. LDA
MOMO
OMOM
R S
S
2. K2CO3, MeOH
128 60%
R
128
O
OH O
OH O
2. O2, triton B 3. NBS
O 125 83%
127 80% O
1. HCl, MeOH
R
2. K2CO3, MeOH
OMOM O
R
DIBALH
1. Pd(OAc)2, PPh2, Et3NHCO2H
O
THF, -78 °C to r. t., 12 h
OMOM O
MOMO
CO2Me
R
R'
124
125
OMOM
OH OH O
OMe +2
O
NaOH
R
MeOH, −25 °C, 2.5 h.
129 40% O O
O
OH O
(−)-urdamicynone B (123) 34%
Scheme 13.39
Yamaguchi’s biomimetic synthesis of (−)-urdamicynone (123).
then added by Claisen condensation of 124 with dimethylallyl acetate in the presence of LDA. Deallyloxycarbonylation catalyzed by a Pd(II) complex followed by lactonization and methoxymethyl (MOM)-protection of the hydroxyl groups yielded 125 in 83% from 126. Controlled reduction of the enol lactone moiety of 125 by diisobutylaluminum hydride (DIBALH) led to the reactive aldehyde 127, which was submitted to a double inter/intra aldolization reaction leading
498
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings
to the protected diketone 128. Deprotection and oxidation by molecular oxygen in the presence of triton B afforded the quinone intermediate 129 in 40% yield, before a final base-induced aldol annelation yielded the target (−)-urdamicynone (123). Angucycline antibiotics were also the main targets of Krohn and coworkers. An alternative biomimetic approach based on a late biomimetic cyclization was developed to synthesize these compounds in good yields [76]. For example, they used dibromonaphthoquinone 130 as the starting material to synthesize tetrangomycin (131) by a combination of biomimetic and conventional chemistries (Scheme 13.40) [77]. CO2Me
O
O
O Br
O
O
Br
O
CO2Me
Bu3Sn
CO2Me
Br O
O
O 130
1. (Bu3Sn)2O 2. OsO4, NaIO4 3. K2CO3, MeOH
O
O
O
O
O
O
O
O
132 O
132
O
O
O
OH
O OH
1. NMO
O O
O
O 2. NaOH 3. AlCl3
O
O Tetrangomycin (131)
Scheme 13.40
Krohn’s biomimetic synthesis of tetrangomycin (131).
The same group was able to perform the syntheses of the closely related angucyclinone, aquayamycin, and WP 3688-2 using a SmI2 -mediated cyclization of ketides [78, 79]. 13.3.4.3 Biomimetic Access to Benzo[a]tetracenoid Derivatives The last step towards complex aromatic polyketides was access to benzo[a]tetracenoid derivatives, which necessitated dodecaketide equivalents. The group of Krohn undertook synthetic studies to construct these very important antibiotics, which culminated in the biomimetic synthesis of 5,6-dideoxypradione (133) closely, which is related to pradimicin A (134) (Scheme 13.41) [80]. While the synthesis of the anthracenequinone 135 began with a Diels–Alder reaction between the appropriate naphthoquinone and diene, construction of the remaining bicycle was performed by a biomimetic approach using very mild conditions. Beyond the scope of this chapter, but worth noting for those interested in such phenolic compounds, is the biomimetic synthesis of aquaticol from hydroxycuparene, a C15 phenolic derivative, through phenol dearomatization and dimerization by Diels–Alder cycloaddition by the group of Quideau [81].
References CO2Me O
O
MeO
OMe
O MeO
O
K2CO3 Propan-2-ol, CH2Cl2, 45 °C
OH O
135
O
HO OH
OH O
O
133 (57%)
CO2H NH
O O
HO OH
MeO O OH O
OH
O NH
HO
Pradimicin A (134)
Scheme 13.41
O OH O
OH OH
Krohn’s biomimetic synthesis of 133.
13.4 Conclusion
Throughout this chapter we have demonstrated the parallel advances in the biochemical understanding of the polyketide biosynthetic pathways and the development of powerful chemical methods aimed at mimicking these pathways. We first noticed that pioneering work begun in the nineteenth century. But is this field still alive and what are the prospects of biomimetic synthesis in the field of polyketides? Two distinct answers can be given to these questions. There is renewed interest in the search for polyketide assembly mimics, as evidenced by recent synthetic methodologies developed in this field. Our increasing knowledge of the enzymatic active sites is now opening the way for new powerful asymmetric organocatalytic methods for the construction of key C–C connections. In terms of the second question, the biological pathways inspired a flurry of biomimetic total syntheses of complex aromatic polyketides and the pioneering work is now used by several groups of chemists worldwide. Tetra- and pentacyclic highly bioactive substances are now accessible by straightforward routes mimicking the cyclization/aromatization biosynthetic strategies. Very powerful chemical routes, inspired by the functioning of biological systems optimized during millions of years of evolution, have today given rise to renewed interest in the context of a synthetic chemistry at the heart of sustainable development. Without doubt we have not yet reached the apogee of biomimetic chemistry but we may have just passed its infancy. References 1. Rohr, J. (2000) Angew. Chem. Int. Ed.,
39, 2847–2849.
2. Smith, S. and Tsai, S.-C. (2007) Nat.
Prod. Rep., 24, 1041–1072.
499
500
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings 3. Hill, A.M. (2006) Nat. Prod. Rep., 23, 4. 5. 6. 7. 8.
9. 10.
11. 12.
13. 14.
15. 16. 17. 18.
19.
20.
21.
22.
256–320. Hertweck, C. (2009) Angew. Chem. Int. Ed., 48, 4688–4716. Staunton, J. and Weissman, K.J. (2001) Nat. Prod. Rep., 18, 380–416. Whiting, D. (2001) Nat. Prod. Rep., 18, 583–606. Collie, J.N. and Myers, W.S. (1893) J. Chem. Soc., 63, 122–128. Birch, A.J., Massy-Westropp, P.A., and Moye, C.J. (1955) Aust. J. Chem., 8, 539–544. Claisen, L. and Claparede, A. (1881) Ber. Deut. Chem. Ges., 14, 2460–2468. Zempleni, J., Wijeratne, S.S., and Hassan, Y.I. (2009) Biofactors, 35, 36–46. Lynen, F. (1967) Biochem. J., 102, 381–400. Scott, A.I., Wiesner, C.J., Yoo, S., and Chung, S.-K. (1975) J. Am. Chem. Soc., 97, 6277–6278. Kobuke, Y. and Yoshida, J. (1978) Tetrahedron Lett., 19, 367–370. Brooks, D.W., Lu, L.D.-L., and Masamune, S. (1979) Angew. Chem. Int. Ed., 18, 72–74. Sakai, N., Sord´e, N., and Matile, S. (2001) Molecules, 6, 845–851. Chen, H. and Harrison, P.H. (2002) Can. J. Chem., 80, 601–607. Sun, S. and Harisson, P.H. (1992) Tetrahedron Lett., 33, 7715–7718. Sun, S. and Harisson, P.H. (1994) J. Chem. Soc., Chem. Commun., 2235–2236. Sun, S., Edwards, L., and Harisson, P.H. (1998) J. Chem. Soc., Perkin Trans. 1, 437–448. Chen, M., Won, K., McDonald, R.S., and Harisson, P.H. (2006) Can. J. Chem., 84, 1188–1196. (a) Carreira, E.M. (1999) in Comprehensive Asymmetric Catalysis, vol. 3 (eds E.N. Jacobsen, A. Pfaltz, and H. Yamamoto), Springer, Heidelberg, pp. 997–1065; (b) Mahrwald, R. (2004) Modern Aldol Reactions, vol. 2, Wiley-VCH Verlag GmbH, Weinheim. Lalic, G., Aloise, A.D., and Shair, M.D. (2003) J. Am. Chem. Soc., 125, 2852–2853.
23. Orlandi, S., Benaglia, M., and Cozzi, F.
(2004) Tetrahedron Lett., 45, 1747–1749. 24. Magdziak, D., Lalic, G., Lee, H.M.,
25. 26. 27.
28.
29.
30. 31.
32. 33. 34.
35.
36.
37.
38.
39.
40.
41.
Fortner, K.C., Aloise, A.D., and Shair, M.D. (2005) J. Am. Chem. Soc., 127, 7284–7285. Fortner, K.C. and Shair, M.D. (2007) J. Am. Chem. Soc., 129, 1032–1033. Davies, C., Heath, R.J., White, S.W., and Rock, C.O. (2000) Structure, 8, 185–195. Blaquiere, N., Shore, D.G., Rousseaux, S., and Fagnou, K. (2009) J. Org. Chem., 74, 6190–6198. Zhang, Y.-M., Hurlbert, J., White, S.W., and Rock, C.O. (2006) J. Biol. Chem., 281, 17390–17399. Ricci, A., Pettersen, D., Bernardi, L., Fini, F., Fochi, M., Perez Herrera, R., and Sgarzani, V. (2007) Adv. Synth. Catal., 349, 1037–1040. Lubkoll, J. and Wennemers, H. (2007) Angew. Chem. Int. Ed., 46, 6841–6844. Benaglia, M., Cinquini, M., and Cozzi, F. (2000) Eur. J. Org. Chem., 563–572. Yost, J.M., Zhou, G., and Coltart, D.M. (2006) Org. Lett., 8, 1503–1506. Lim, D., Fang, F., Zhou, G., and Coltart, D.M. (2007) Org. Lett., 9, 4139–4142. Kohler, M.C., Yost, J.M., Garnsey, M.R., and Coltart, D.M. (2010) Org. Lett., 12, 3376–3379. Alonso, D.A., Kitagaki, S., Utsumi, N., and Barbas, C.F. III (2008) Angew. Chem. Int. Ed., 47, 4588–4591. Utsumi, N., Kitagaki, S., and Barbas, C.F. III (2008) Org. Lett., 10, 3405–3408. Le Sann, C., Munoz, D.M., Saunders, N., Simpson, T.J., Smith, D.I., Soulas, F., Watts, P., and Willis, C.L. (2005) Org. Biomol. Chem., 3, 1719–1728. Jones, G. (1967) in Organic Reactions, vol. 15, John Wiley & Sons, Inc., New York, pp. 204–599. Augustine, J.K., Naik, Y.A., Poojari, S., Chowdappa, N., Sherigara, B.S., and Areppa, K. (2009) Synthesis, 2349–2356. Lopez Herrera, F.J. and Pino Gonzalez, M.S. (1986) Carbohydr. Res., 152, 283–291. Peng, Y. and Song, G. (2003) Green Chem., 5, 704–706.
References 42. List, B., Doehring, A., Hechavarria
43.
44. 45.
46.
47. 48. 49.
50.
51.
52.
53. 54.
55.
56.
57. 58.
Fonseca, M.T., Wobser, K., van Thienen, H., Rios Torres, R., and Llamas Galilea, P. (2005) Adv. Synth. Cat., 347, 1558–1560. Berrue, F., Antoniotti, S., Thomas, O.P., and Amade, P. (2007) Eur. J. Org. Chem., 1743–1748. Hilt, G. and Weske, D.F. (2009) Chem. Soc. Rev., 38, 3082–3091. Birch, A.J., Massy-Westropp, P.A., and Moye, C.J. (1955) Aust. J. Chem., 8, 539–544. Robinson, R. (1955) The Structural Relations of Natural Products, Clarendon Press, Oxford. Birch, A.J. and Donovan, F.W. (1953) Aust. J. Chem., 6, 360–368. Thomas, R. (2001) ChemBioChem, 2, 612–627. Bringmann, G., Wohlfarth, M., Rischer, H., R¨uckert, M., and Schlauer, J. (2000) Angew. Chem. Int. Ed., 39, 1464–1466. Bringmann, G., Noll, T.F., Gulder, T.A.M., Gr¨une, M., Dreyer, M., Wilde, C., Pankewitz, F., Hilker, M., Payne, G.D., Jones, A.L., Goodfellow, M., and Fiedler, H.-P. (2006) Nat. Chem. Biol., 2, 429–433. Bringmann, G., Gulder, T.A.M., Hamm, A., Goodfellow, M., and Fiedler, H.-P. (2009) Chem. Commun., 6810–6812. Reviews: (a) Harris, T.M. and Harris, C.M. (1977) Tetrahedron, 33, 2159–2185; (b) Harris, T.M. and Harris, C.M. (1986) Pure Appl. Chem., 58, 283–294. Harris, T.M. and Harris, C.M. (1969) Tetrahedron, 25, 2687–2691. Birch, A.J., Cameron, D.W., and Richards, R.W. (1960) J. Chem. Soc., 4395–4400. Harris, T.M., Murphy, G.P., and Poje, A.J. (1976) J. Am. Chem. Soc., 98, 7733–7741. (a) Harris, T.M. and Carnay, R.L. (1966) J. Am. Chem. Soc., 88, 5686–5687; (b) Harris, T.M. and Carney, R.L. (1967) J. Am. Chem. Soc., 89, 6734–6740. Money, T. (1970) Chem. Rev., 70, 553–560. Crombie, L. and James, A.W.G. (1966) Chem. Commun., 357–359.
59. Navarro, I., Basset, J.-F., Hebbe, S.,
60.
61.
62. 63. 64. 65. 66. 67. 68.
69.
70.
71.
72.
73.
74.
Major, S.M., Werner, T., Howsham, C., Br¨ackow, J., and Barrett, A.G.M. (2008) J. Am. Chem. Soc., 130, 10293–10298. Barrett, A.G.M., Morris, T.M., and Barton, D.H.R.J. (1980) J. Chem. Soc., Perkin Trans. 1, 10, 2272–2277. Basset, J.-F., Leslie, C., Hamprecht, D., White, A.J.P., and Barrett, A.G.M. (2010) Tetrahedron Lett., 51, 783–785. Harris, T.M. and Murphy, G.P. (1971) J. Am. Chem. Soc., 93, 6708–6709. Comer, F.W., Money, T., and Scott, A.I. (1967) Chem. Commun., 231–233. Stockinger, H. and Schmidt, U. (1976) Liebigs Ann. Chem., 1617–1625. Baker, P.M. and Bycroft, B.W. (1968) Chem. Commun., 71–72. Wittek, P.J. and Harris, T.M. (1973) J. Am. Chem. Soc., 95, 6865–6867. Harris, T.M. and Wittek, P.J. (1975) J. Am. Chem. Soc., 97, 3270–3271. Review: Yamaguchi, M. (1992) in Studies in Natural Products Chemistry, vol. 11 (ed. A.U. Rahman), Elsevier Science BV, Amsterdam, pp. 113–149. Yamaguchi, M., Hasebe, K., and Minami, T. (1986) Tetrahedron Lett., 27, 2401–2404. Yamaguchi, M., Nakamura, S., Okuma, T., and Minami, T. (1990) Tetrahedron Lett., 31, 3913–3916. Harris, T.M., Webb, A.D., Harris, C.M., Wittek, P.J., and Murray, T.P. (1976) J. Am. Chem. Soc., 98, 6065–6067. (a) Gilbreath, S.G., Harris, C.M., and Harris, T.M. (1988) J. Am. Chem. Soc., 110, 6172–6179; (b) Harris, T.M., Harris, C.M., Oster, T.A., Brown L.E. Jr., and Lee, J.Y.-C. (1988) J. Am. Chem. Soc., 110, 6180–6186; (c) Harris, T.M., Harris, C.M., Kuzma, P.C., Lee, J.Y.-C., Mahalingam, S., and Gilbreath, S.G. (1988) J. Am. Chem. Soc., 110, 6186–6192. Pickens, L.B., Kim, W., Wang, P., Zhou Watanabe, K., Gomi, S., and Tang, T. (2009) J. Am. Chem. Soc., 131, 17677–17689. Yamaguchi, M., Hasebe, K., Higashi, H., Uchida, M., Irie, A., and Minami, T. (1990) J. Org. Chem., 55, 1611–1623.
501
502
13 Polyketide Assembly Mimics and Biomimetic Access to Aromatic Rings 75. Yamaguchi, M., Okuma, T.,
76. 77.
78. 79.
Horiguchi, A., Ikeura, C., and Minami, T. (1992) J. Org. Chem., 57, 1647–1649. Review: Krohn, K. (2002) Eur. J. Org. Chem., 1351–1362. Krohn, K., Boker, N., Florke, U., and Freund, C. (1997) J. Org. Chem., 62, 2350–2356. Krohn, K., Frese, P., and Florke, U. (2000) Chem. Eur. J., 6, 3887–3896. Krohn, K., Vidal, A., Tran-Thien, H.T., Florke, U., Bechthold, A., Dujardin, G.,
and Green, I. (2010) Eur. J. Org. Chem., 3080–3092. 80. (a) Krohn, K. and Bernhard, S. (1999) Eur. J. Org. Chem., 3099–3103; (b) Krohn, K., Bernhardt, S., Florke, U., and Hayat, N. (2000) J. Org. Chem., 65, 3218–3222. 81. Gagnepain, J., Castet, F., and Quideau, S. (2007) Angew. Chem. Int. Ed., 46, 1533–1535.
503
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides Bastien Nay and Nassima Riache
14.1 Introduction
The biosynthesis of polyketides is extraordinarily orchestrated by proteins called polyketide synthases (PKSs) which are always highly organized modular or iterative enzymatic complexes [1]. They are analogous to the fatty acid synthases that catalyze the decarboxylation of the malonate precursor to drive the elongation of acyl chains. Depending on the functional attributes of the PKS enzymatic complex, the elongated carbon chain will be more or less functionalized, with a tuned reactivity for specific rearrangements, especially cyclizations, toward the natural product. The ‘‘minimal PKS’’ is made of an acyl carrier protein (ACP), an acyl transferase (AT), and a ketosynthase (KS), which lead to highly reactive poly-β-ketoacyl chains. When the chain reaches the appropriate length (this can be controlled by a chain length factor) it undergoes Claisen condensations to furnish phenolic compounds such as actinorhodin, tetracycline, or doxorubicin [1c, 2]. Additional modules can be involved within the PKS, catalyzing ketone reduction, dehydration, and enoyl reduction. In such a complete PKS, the carbon chain is saturated, as in fatty acids. These additional modules are optional and can be missing, which means that the functionalization can stop at the β-hydroxy or the enoyl stage, thus increasing the variability of functional groups within the elongated chain. In many cases, the linear product of the PKS will be able to undergo complex rearrangement towards polycyclic structures. This is the biosynthetic path for numerous polyketide natural products, some of which are described in this chapter. The science of synthesis was soon interested in polyketide complexity. It has been obvious to many authors that mimicking the biosynthetic route would provide efficient access to this complexity. The biomimetic synthesis of polyketides embraces a wide range of organic reactions. One of the first biomimetic syntheses of aromatic polyketides was reported by Collie and Myers in 1893 [3], consisting of the condensation of a trienone into the phenolic compound orcinol. Since then, this strategy has been used to synthesize numerous polyaromatic natural products [4]. To obtain the poly-β-ketoacyl precursors, some authors embarked Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
504
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides
on the biomimetic decarboxylative condensation of malonate onto acetate units. Especially successful attempts were realized by use of enzyme mimics [5].1) Studies of the biomimetic synthesis of non-aromatic polycyclic polyketides have grown impressively over the past 20 years. This has brought important keys to the comprehension of biosynthetic steps and of the reactivity of biosynthetic intermediates. Rapid and efficient synthetic methods have been collected for the construction of biologically interesting polyketides. In this chapter, we give a comprehensive overview of those works, describing significant examples with variable complexity and reactivity.
14.2 Biomimetic Studies in the Nonadride Series 14.2.1 Dimerization Process towards Isoglaucanic Acid
Glaucanic acid (2) is a bis(anhydride) derived from the formal [6π+4π] cycloaddition of two C9 -units (1) (Scheme 14.1). It was first isolated in 1931 from Penicillium glaucum [6]. With regard of its C9 structural features the name nonadride was suggested by Sutherland and coworkers [7, 8]. The biosynthesis of 2 was studied by Sutherland, who demonstrated the acetate origin of glaucanic acids [9]. When the synthetic [3 H]-labeled C9 -precursor 1 was fed to the growing mould Penicillium purpurogenum, another nonadride producer, 51% of the [3 H]-activity was incorporated into glauconic acid (4). The mould was thus capable of effecting the dimerization of 1 into 2 and 4. There were numerous possible mechanisms for this condensation, but an attractive one was the [6π+4π] cycloaddition of the anion of 1 with the cisoid form of the same anhydride to give 2. The reaction via an endo transition state would give the correct relative stereochemistry. This dimerization was attempted in vitro by treating anhydride 1 with triethylamine in dimethylformamide (DMF) [9b, 10]. Yet it only gave 4% of the epimeric exo product, iso-glaucanic acid (3). No other bis(anhydrides) of this type were formed during the reaction. The difference between the biochemical and chemical routes leading to 2 and 3, respectively, was O O O
O O
1b
H biomimetic route: NEt3, DMF, 4% leading to 3 via 1b
1a O
1) For
O 1
O
O
O
Me O
OH
O 2 1
O
anion of 1
Scheme 14.1
O
biogenetic hypothesis: [6π+4π] cycloaddition leading to 2 via 1a
O H
5 [O]
2 1
O
9
O O 2 (5-α): glaucanic acid 3 (5-β): iso -glaucanic acid
5 9
Me O
O O 4: glauconic acid
Nonadride biogenetic hypothesis and Sutherland’s biomimetic studies.
a comprehensive discussion on polyketide enzyme mimics and on
the biomimetic synthesis of aromatic polyketides, see Chapter 13.
14.2 Biomimetic Studies in the Nonadride Series O MgCl2 NEt3, DMSO O
O O
HO2C
O H O
+
H
O O
O 5
O
+O O
6 (8.5%)
Scheme 14.2
O
HO2C
505
O 7 (6%)
O
O
O
8 (<2%)
Dimerization route toward the nonadride 6 and side-products 7 and 8.
explained by postulation, in the latter case, of an isomerization of the anion 1a to the anion 1b, which would then attack the neutral form 1. This isomerization would not happen in vivo. More recently the mechanism of this reaction was reinvestigated by Baldwin, who suggested a stepwise dimerization of the anhydride unit (Scheme 14.2) rather than a [6π+4π] cycloaddition [11]. After testing different reaction conditions (solvent, temperature, base, addition of a metal salt) to optimize the dimerization of the homo-derivative 5, it was found that a mixture dimethyl sulfoxide (DMSO)/NEt3 (0.66 equiv.)/MgCl2 (0.5 equiv.) gave the best yield (8.5%) of iso-glaucanic acid analog 6. It was accompanied with two regioisomers of cyclization (7 and 8), the presence of which suggested a stepwise mechanism via Michael addition for the formation of 6. Tethering both anhydride units through their lateral alkenyl chain, giving dimeric bis(anhydrides) with various tether lengths, allowed for 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU)-catalyzed intramolecular cyclizations that occurred in better yields (14–17%) and with variable stereoselectivities at the cyclononene substitutions [12]. 14.2.2 The Unresolved Case of CP-225917 and CP-263114
The CP-225917 and CP-263114 molecules, or phomoidrides A and B (10, 11), have been isolated from an unidentified fungus (Scheme 14.3) [13]. They are oxygenated and bridged nonadrides with inhibitory activities against squalene synthase and ras-farnesyltransferase, thus providing lead structures for the development of anticancer and cholesterol-lowering agents. Compared to glaucanic acid (2), these compounds share longer alkyl chains and a trans double bond embedded within the common cyclononane ring, which is also bridged by a new C–C bond between C10 and C26. New structural challenges thus arise with regard to a biomimetic strategy. Biogenetically, 10 would be formed by the controlled dimerization and functionalization of anhydride 9. S
O
S O
26
10
9 ()
O 13
17 16
()
5
9b
Scheme 14.3
9a
O
16 17
10
HO2C O
5
O 26 OH 14
O 14 O
O O
O
13
O
stepwise mechanism
O
O
O HO2C
9
O
O
HO
O CP-225917 = phomoidride A (10)
O
O O CP-263114 = phomoidride B (11)
Biosynthetic origin of the CP-225917 (10) and CP-263114 (11).
506
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides
With regard to polyketide biosyntheses, Sulikowski and coworkers speculated that both monomers (9) were covalently attached to an enzyme active site (thioester linkages) during the dimerization process [14]. This would serve to govern the topology of the reaction toward the cyclononene ring through a stepwise mechanism involving two Michael additions between the anhydride units 9 (C13 → C14, C17 → C9). Conformational restrictions (exo topology) for the cyclization would be imposed by the thioester linkages. Then, a transannular Dieckmann cyclization (C10 → C26) and a decarboxylation at C13 [15] would lead to the CP–225917 skeleton 10. Sulikowski attempted to use this biosynthetic hypothesis in a biomimetic strategy for the synthesis of 10 and 11 [16]. The thioester templating effect was mimicked by the use of 1,n-diols with variable chain lengths (n = 1–6) to connect both anhydrides units through their acetyl part. Unfortunately, when basic cyclization conditions were applied to these dimeric anhydrides (DBU, MeCN, 80 ◦ C, 0.001 M), the desired carbon connection C13–C14 was not observed [16a] – instead the C13–C17 bond was formed. Condensation derivatives were described for longer tethers (n = 3–6). The biomimetic synthesis of phomoidrides thus remains unresolved.
14.3 Biomimetic Syntheses Involving the Diels–Alder Reaction
First reported in 1928, the Diels–Alder reaction [17] is, to date, one of the mostly used methodologies in organic synthesis and often constitutes a key step in the total synthesis of natural products [18], among which polyketides are far from being an exception. Additionally, even though few ‘‘Diels–Alderases’’ have yet been identified, the enzymatic Diels–Alder reaction stands as the biogenetic hypothesis of numerous polyketides (see also Chapter 21) [19]. Chemists have attempted to demonstrate the relevance of biosynthetic Diels–Alder reactions in numerous biomimetic syntheses of natural products, some of which are described in the following section. 14.3.1 Biomimetic Diels–Alder Reactions Affording Decalin Systems
Lovastatin (14) (= mevinolin, Scheme 14.4) was isolated from a culture of Aspergillus terreus [20] and from Monascus ruber [21]. It is a potent inhibitor of cholesterol biosynthesis in humans, after hydrolysis of the lactone into the active β-hydroxy-acid form. The biosynthesis of 14 originates from the polyketide metabolism, through the triene (12), which undergoes an endo-selective Diels–Alder addition to form the decalin core 13 [22]. It was then suggested that this cycloaddition would be driven by a biological process that involves a Diels–Alderase [23].2) 2) For a discussion on Diels–Alderases see
Chapter 21.
14.3 Biomimetic Syntheses Involving the Diels–Alder Reaction HO
Biosynthesis of lovastatin (14) O H
O O HO2C
+ S-ACP
S-ACP
O
S-ACP
O O
H 13
lovastatin (14)
products from thermal or Lewis acid conditions
COR
H
COR COR 16
H
15a
endo
endo
COR
exo
COR
R = OEt, OH or S(CH2)2NHAc
15c
H
Scheme 14.4
15b
COR
COR 18
exo
H H
15 17
Diels– Alder
H
S-KS
product obtained under enzymatic conditions (LNKS)
H
O
O
IMDA
12
H
507
COR 15d
Diels–Alder cyclization of trienes 15.
Vederas and coworkers studied the relevance of this biosynthetic Diels–Alder cyclization by testing the reactivity of various synthetic trienes (15) (Scheme 14.4) under thermal or Lewis acid conditions [24]. It only resulted in the formation of decalins 18 and 19, respectively, by endo and exo addition from chair like transition states 15c and 15d with the methyl substituent in an equatorial position. The endo compound 16, which contains the required stereochemistry, was not formed under these conditions. This suggested implication of an enzymatic Diels–Alder reaction to promote the endo transition state 15a in vivo towards the desired cycloadduct 16, with the methyl substituent axially positioned. In 2000, Vederas isolated and characterized the lovastatin nonaketide synthase (LNKS), a type I iterative PKS and the first naturally occurring Diels–Alderase [25]. The desired decalin (16), with the stereochemistry of lovastatin 14, was formed exclusively when the linear substrate 15 was submitted to LNKS. This Diels–Alderase would indeed stabilize the transition state 15a. Solanapyrones (23, 27, Scheme 14.5) are phytotoxic polyketides and inhibitors of DNA polymerase β and λ, produced by the fungi Alternaria solani and Ascochyta rabiei [26]. Their biosynthesis would involve an IMDA (intramolecular Diels–Alder reaction) reaction leading to the decalin system [27, 28]. Studies indicated that the biosynthesis of most of these compounds proceeds via an exo selective Diels–Alder cycloaddition that would be catalyzed by an oxidase involved in the biosynthetic route [29]. The exo cycloadduct corresponds to the cis-decalin ring of solanapyrones A and B, while the endo cycloadduct holds the trans-fusion of solanapyrones D and E. The first biomimetic synthesis of (±)-solanapyrone A (23) was achieved in 1987 by Ichihara and coworkers (Scheme 14.5a) [30]. Later it was shown by Oikawa
19
H
COR
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides
508
S
S
S
O
OMe O
PhMe 180 °C
S
O
S OMe
O
+
O
1h, 71%
O
O
H
21
CHO OMe
H 20
(a)
S
endo:exo 2:1
OMe O
H
H
H 22
H solanapyrone A (23) CHO
O
Ph
CHO
(b)
24
O
Me N
N H (20%)
Me Me Me
5 °C, MeCN 71% >20:1 dr 90% ee
OHC Me
H
H 25
TMSO
OTMS
CO2Me
HO
OMe Me TiCl4, CH2Cl2 −78 °C 75%
O
OMe O
H
H
H 26
H solanapyrone D (27)
Scheme 14.5 Biomimetic syntheses of solanapyrones: (a) Ichihara’s synthesis of solanapyrone A (23); (b) MacMillan’s organocatalytic synthesis of solanapyrone D (27).
and coworkers that thermal conditions and A. solani cell-free extracts promoted different stereoselectivity. The same team succeeded in the enantioselective and exo-selective synthesis of (−)-23 utilizing a crude enzyme preparation [31]. Recently, the purification and identification of the Diels–Alderase solanapyrone synthase has been achieved [32]. The facile cyclization of the linear prosolanapyrone intermediate in enzyme-free aqueous buffer suggested that this Diels–Alderase attends only to control the stereochemistry of the cyclization rather than serving as a true enzymatic catalyst. In 2005, the enantioselective synthesis of solanapyrone D (27) involving an organocatalytic biomimetic IMDA was achieved by MacMillan and coworkers (Scheme 14.5b) [33]. The cycloaddition proceeded with 71% yields (>20 : 1 dr, 90% ee). Nargenicin A1 (28, Scheme 14.6) is an antibiotic isolated from Nocardia argentinensis [34]. Cane and coworkers established that nargenicin is derived from four propionate and five acetate units [35]. Later, intact incorporation of a series of postulated chain elongation intermediates into nargenicin demonstrated that the oxygen atom at C13 is not derived from propionate. This implied that the C4–C13 bond is not formed by an aldol-type condensation [36]. Cane suggested that the cis-decalin system of 28 may arise from an IMDA cycloaddition of a linear precursor. In 1989 Roush and coworkers investigated the Diels–Alder reaction towards 28 from decatrienone substrates such as 29 [37]. The stereoselectivity of the IMDA reaction was improved in the transannular version of this reaction (Scheme 14.6) [38]. The authors succeeded in the stereoselective synthesis of the tricyclic lactone core (31) of nargenicin A1 thanks to the transannular Diels–Alder (TADA) reaction of the transient 18-membered macrolide 30.
14.3 Biomimetic Syntheses Involving the Diels–Alder Reaction
Br
Br
O
O
O
O
8 4
1) Cl3C6H2COCl NEt3, THF, 23 °C
11
O
HO2C
2) DMAP PhMe, 100 °C
13
OH
O
4 13
O O
TBDPSO TBDPSO
29
O H MeO O
Br 8
H 4 13
H HO
O O
N H
Scheme 14.6
4
11
O HO
30
O
H 13
O
O
H
H
O
O nargenicin A1 (28)
TBDPSO
31
Roush’s studies toward the biomimetic synthesis of nargenicin A1 (28).
Superstolide A (34) is a cytotoxic macrolide isolated from the New Caledonian marine sponge Neosiphonia superstes [39]. In 2008, Roush and coworkers reported the total synthesis of (+)-superstolide A by stereocontrolled TADA reaction of macrolactone 32 (Scheme 14.7) [40]. On this occasion, the comparison between IMDA and TADA strategies showed that higher diastereoselectivity was obtained in the transannular version. Indeed, while the IMDA reaction gave a mixture of three cycloadducts, the TADA cycloaddition of 32 provided the sole cycloadduct 33. This highly diastereoselective cycloaddition was attributed to conformational effects within the macrocycle. Himbacine (38), himbeline (39), and himandravine (40) are members of Galbulimina alkaloids, isolated from the bark of Galbulimina baccata (Scheme 14.8) [41]. Their biosynthesis has polyketide origins and would involve a Diels–Alder reaction of the linear precursor 36, where an iminium ion would activate the reaction as postulated by Baldwin and coworkers [42]. In 2005, this hypothesis was supported by Baldwin’s biomimetic synthesis of 38–40 (Scheme 14.8) [43]. The one-pot N-Boc deprotection of intermediate 35 and subsequent iminium formation (36) induced IMDA cycloaddition to give the endo cycloadduct 37. This hypothetical intermediate in the biosynthesis was then converted into Galbulimina alkaloids 38–40. 14.3.2 Biomimetic Diels–Alder Reactions Affording Tetrahydroindane Systems
Spiculoic acid A (42) is a cytotoxic metabolite isolated from the Caribbean marine sponge Plakortis angulospiculatus [44]. The polyketide biogenetic origin of 42 would involve four butyrate and one propionate units incorporated into the linear
509
Scheme 14.7
MeO
32
O
O O NBoc
or 80 °C PhMe, 2h 30–35%
CDCl3 MeO 23 °C, 5 days
H
OTBS H
33
O
O
Roush’s total synthesis of superstolide A (34).
OTBS
O NBoc
3) TFA, CH2Cl2 4) Ac2O, Et3N, THF 42%
1) TBAF, THF 2) O=N=CCOCCl3 MeO CH2Cl2 then Al2O3
H 2N
O
O
superstolide A (34)
H
O H
O
NHAc
OH
510
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides
14.3 Biomimetic Syntheses Involving the Diels–Alder Reaction Me
Me Me
NHBoc
Me
NH TFA CH2Cl2 0 °C
O
NH
N H
0 °C-rt
Me
Me
O
O
H
H
Me
17
H
H
H
H
O H 35
Scheme 14.8
511
O
O
36
Me O
O 37
R
O
17α-H, R = Me: himbacine (38) 17α-H, R = H: himbeline (39) 17β-H, R = H: himandravine (40)
Baldwin’s biomimetic synthesis of Galbulimina alkaloids 38–40. Biosynthetic origin of spiculoic acid (42) H O
IMDA
ent-42 O
HO
Enz-S
1) DDQ 2) Pd(PPh3)4 3) IBX
H
O spiculoic acid (42)
O 41
Ph + Ph
Scheme 14.9
PMBO 43
O
Ph3P O
O 44
H OPMB
toluene 100 °C 25%
42% (3 steps)
O Ph
PMBO 45
O
H
O O
Baldwin’s biomimetic synthesis of ent-spiculoic acid (42).
precursor 41, the IMDA reaction of which would afford the spiculane skeleton (Scheme 14.9). In 2006 Baldwin and coworkers reported the biomimetic total synthesis of ent-spiculoic acid 42 [45]. This work revealed the absolute configuration of the natural product. In fact, the Wittig reaction between the aldehyde 43 and the phosphoranylidene 44 furnished directly the cycloadduct 46, through IMDA reaction of the presumed linear intermediate 45. ent-Spiculoic acid (ent-42) was finally obtained upon three steps of deprotection and oxidation. This achievement, as well as Perkin’s conclusions regarding the stereoselectivity of the thermal IMDA cyclization towards spiculoic acids [46], argues in favor of a biosynthetic Diels–Alder step. (−)-Galiellalactone (49, Scheme 14.10) is a hexaketide isolated from the ascomycete Galiella rufa [47], with selective and potent inhibition of IL-6 [48] and potential antitumor activities [49]. The biosynthesis would involve an IMDA reaction from the biosynthetic intermediate (−)-pregaliellalactone (47) into (+)-deoxygaliellalactone (48), followed by enzymatic hydroxylation into 49 [50]. This unusual cycloaddition with inverse electron demand was supported by feeding experiments of the substrate 47 to the fungus [51]. On this occasion, Sterner and coworkers showed that the reaction was selective, providing the
46
512
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides
natural (+)-deoxygaliellalactone (48) as a sole isomer. Since spontaneous or thermally induced cyclizations also provided stereoselectively the natural product, it was suggested that hypothetical galiellalactone Diels–Alderase would only serve to enhance the cyclization rate [51]. In 2007, a successful biomimetic synthesis of 48 from 47 was reported by Lebel and Parmentier (Scheme 14.10) [52]. The precursor 47 was involved in IMDA cyclization in the presence of AlCl3 under microwave irradiation, providing (+)-deoxygaliellalactone (48). A one-pot sequence towards 48 from the aldehyde 51 was also reported by the authors. Biosynthesis of galiellalactones
H
IMDA
O
O H (−)-galiellalactone (49)
H (+)-deoxygaliellalactone (48)
IBX, DMSO
O
O
80%
PPh3, TMSCHN2
AlCl3, µW
IMesCuCl dioxane, 60 °C
dioxane 140 °C
O
77%
O
51
H
H
O O
56%
O
OH
H
O O
(−)-pregaliellalactone (47)
50
OH
oxidation
O
O
O
H
(−) pregaliellalactone (47)
O H (+)-deoxygaliellalactone (48)
1) PPh3, TMSCHN2, IMesCuCl, dioxane, 60 °C 2) AlCl3, 140 °C
52%
Scheme 14.10
Biosynthesis of galiellalactones and Lebel’s biomimetic synthesis.
14.3.3 Biomimetic Diels–Alder Reactions Affording Spiro Systems
Abyssomicin C (54) is an antibiotic isolated in 2004 from the marine actinomycete Verrucosispora AB 18-032 [53]. Sorensen and coworkers suggested in 2005 that the biosynthesis of 54 involves an IMDA reaction [54]. Their biomimetic enantioselective synthesis of (−)-abyssomicin C via a tandem β-elimination/diastereoselective Diels–Alder macrocyclization from the linear intermediate 52 provided an argument in favor of the postulated biosynthesis (Scheme 14.11a) [54]. The reaction was accomplished in the presence of a catalytic amount of La(OTf)3 . Abyssomicin C (54) was obtained from the cycloadduct 53 in three steps: stereoselective epoxidation, demethylation, and intramolecular epoxide opening. In 2007, Nicolaou reported another total synthesis of 54 [55]. On this occasion, he demonstrated that it could be equilibrated with the atropisomer atrop-abyssomicin C (55) in the presence of ethereal HCl (Scheme 14.11b). After L-Selectride reduction of 55 and acidic workup, abyssomicin D (57) was obtained. These experiments gave strong support to the biosynthetic conversion of 54 into 57 through 55 in the microorganism.
H
52
O O
O
2:1 ratio
O H
OMe
O
O H O
50%
La(OTf3) toluene 100 °C
O O
O
O
O
O L-Selectride
OH abyssomicin C (54)
O
O
OH atrop-abyssomicin C (55)
unstabilized CDCl3
O
O
O
OMe
O
OH 56
O
O H
O
O
60%
Michael addition
1) DMDO, acetone (67%) 2) LiCl, DMSO, 50 °C (quant) 3) pTsOH, LiCl, MeCN (50%)
O
O
H
O
O
OH
O
HO O
53
O
+ +
O
OMe
O
IMDA
O
OMe
abyssomicin D (57)
O
O
H
O
O
Biomimetic synthesis of abyssomicins C and D according to Sorensen (a) and Nicolaou (b).
OH abyssomicin C (54)
O
O H O
TBSO
Scheme 14.11
(b)
(a)
14.3 Biomimetic Syntheses Involving the Diels–Alder Reaction 513
514
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides
Last but not least, atrop-abyssomicin C (55) was finally isolated in 2007 as the major atropisomer in the strain Verrucosispora AB 18-032 [56], giving a successful conclusion to this synthetic work. Gymnodimine (58) is a member of a large class of natural products bearing a cyclic imine fused to a cyclohexene ring and a macrocarbocycle, isolated from oysters contaminated by a Dinoflagellate [57]. Kishi and coworkers suggested that a biosynthesis apparented to pinnatoxin A [58], that is, an intramolecular reaction involving an iminium dienophile, would lead to 58 [59]. They conducted a synthetic study using the biomimetic precursor 59, which was converted into the gymnodimine core 61 in a two-step procedure, consisting of the formation of a cyclic imine (60) and then IMDA reaction (Scheme 14.12). O
OH O
OH
OH
OH
OH
H
Pd(PPh3)4
O
O
O IMDA N
NHAlloc
N
O
60
59 gymnodimine (58)
pH 6.5 sodium citrate/HCl buffer
OH
OH
H
OH O
OH
H
+
N
36 °C 48h, 60 µM
O N
61 (exo)
(1:1)
62 (endo)
Scheme 14.12 Kishi’s synthetic studies toward the biomimetic synthesis of gymnodimine (58).
This reaction proceeded under aqueous conditions at pH 6.5 at 36 ◦ C, giving the exo product 61 with the desired diastereoselectivity and the endo product 62 (dr 1 : 1). This result is in contrast with the cycloaddition made on enone dienophiles similar to 59, which only gave undesired endo products. These results suggested that the hypothetical biosynthesis of gymnodimine (58), wherein the Diels–Alder precursor reacts to form the natural product, could occur spontaneously without the aid of an enzyme. 14.3.4 Biomimetic TADA Reactions toward FR182877, Hexacyclinic Acid and the Parent Cochleamycin A and Macquarimicin A
The compound (−)-FR182877 (64, also named cyclostreptin) was isolated by Sato and coworkers from a Streptomyces strain (Scheme 14.13) [60, 61]. It showed potent cytotoxicity with irreversible microtubule stabilizing properties. Soon after this, the very closely related hexacyclinic acid (65) was reported by Zeeck and coworkers, who investigated its biosynthetic origin [62]. Using 13 C labeling, it was demonstrated that 65 originated from the type I polyketide biogenetic route, incorporating six acetate
14.3 Biomimetic Syntheses Involving the Diels–Alder Reaction O H
HO Me
HO Me
biogenetic condensation
Me
H O
O
Me H
A
HO
515
O H OH Me H H H OO D
H
B
C
O HO HO
Me
Me + endo -63 +
O Me
O
Me
O Me
HO
HO Me
Me
O
FR182877 (64)
double biosynthetic TADA cyclization + functionalization
Me
H
H
O
Me H
A
AcO H
63 Me
O H OH H HO Me H H H O O D
B
C
H Me
H CO2H
Me + exo -63 +
Scheme 14.13
H Me
H
hexacyclinic acid (65)
Common biosynthetic origin for FR182877 (64) and hexacyclinic acid (65).
and four propionate units. It was supposed that an IMDA reaction and further aldol reaction, lactonization, and ketalization would provide all the functionalities of 65. The parent FR182877 (64) would have the same biosynthetic origin. In fact both products would be formed from the double TADA reaction of a simpler macrocyclic polyketide precursor. Evans and Starr reasoned that they could even share the common precursor 63, the cyclization of which would lead to 64 or 65, depending on whether the transition state of the TADA reaction is endo or exo, respectively [63]. Macrocyclic intermediates incorporating a β-keto-δ-lactone like that in 63 have been invoked in the biosynthesis of other complex natural products such as cochleamycin A and macquarimicin A, the biomimetic synthesis of which is discussed below. The synthesis of FR182877 (64) has been investigated by several authors, giving support to the biosynthetic considerations discussed above. Sorensen and coworkers first performed the biomimetic synthesis of (+)-64 (Scheme 14.14) after TESO O
OTES
O O
N
12 steps
1
Bn +
1) MeOCOCl, pyr. 2) TMSCl, imid. 3) Pd2dba3
Me
O OTES
TESO
1
71% (3 steps)
TBSO
CHO
Me
19
CO2tBu 66
OH
OTMS O tBu
19
OH
O
O 67 1) KHMDS PhSeBr (91%) 2) mCPBA
(+)-64
Me 1) PPTS, MeOH 2) TFA, CH2Cl2 TESO 3) EDC, DMAP, CH2Cl2 H
62% (3 steps)
Scheme 14.14
OTES CO tBu Me 2 Me OTMS NaHCO3 O H H CH2Cl2 OTMS OTES O 40 °C TESO 19 O tBu H 1 40% H Me (2 steps) O H Me Me 69 68E
Sorensen’s biomimetic synthesis of (+)-FR182877 [(+)-64].
516
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides
considerable work that highlighted the particular reactivity of some intermediates [64]. The intermediate 66 was obtained after 12 steps, beginning with an asymmetric aldol reaction with an Evans chiral auxiliary. This linear compound bears all the needed functionalities, that is, an allylic alcohol and a β-ketoester, to envisage the macrocyclic ring closure. After methoxycarbonylation of the primary alcohol at C1 and protection of the secondary one, a Tsuji–Trost reaction in the presence of Pd2 dba3 allowed for the C1–C19 ring closure of 66 smoothly and in very good yield and diastereoselectivity. The macrocyclic biomimetic precursor [(E)-68] of FR182877 was finally revealed by selenylation and oxidative deselenylation at C19 from 67, giving an equimolar ratio of (E) and (Z) isomers [(E)-68 and (Z)-68]. The high reactivity of these macrocyclic pentaenes precluded all isolation attempts. The transannular cycloaddition was also observed at ambient temperature. Complete conversion of the isomeric pentaenes into three major products was observed on warming a chloroform solution buffered with sodium bicarbonate for 4 h at 40 ◦ C. The major product of the sequence was the pentacycle 69, bearing the same relative stereochemistry as FR182877. To complete the synthesis, deprotection of silyl ethers and final lactone formation were performed, giving (+)-64 in 62% yield over three steps. However, it had been reported that the correct structure of the natural product FR182877 was the opposite enantiomer, (−)-64 [61]. Therefore in the same final report [64d], Sorensen and coworkers described the total synthesis of (−)-64, a work made on a large scale for biological purposes. It was possible to synthesize 24 g of the cyclic intermediate ent-67, which was transformed into 5.4 g of the lactone 70 (Scheme 14.15), a stable compound on storage. The conversion of 70 into the less stable and non-storable natural product (−)-64 was undertaken on a 100-mg scale. Evans and Starr undertook a total synthesis of (−)-64 using a similar double TADA process [63]. This work confirmed the absolute configuration of FR182877. According to the authors, ‘‘the macrocyclic conformation of the reactive intermediate appears to predispose the reacting olefin/carbonyl faces to a single orientation, and the resulting high stereoselectivity of the double TADA sequence, exclusively leading to FR182877, stands in stark contrast to the results obtained in similar acyclic systems (71).’’ Me HO H
OH CO2H H H
O
Me
Me OH
O
H H Me
H H
N-methyl-2-chloropyridinium iodide NEt3, CH2Cl2/MeCN, 23 °C
HO H
Scheme 14.15 Final step in the large-scale synthesis of the natural enantiomer of FR182877 (b.r.s.m.: based on recovered starting material).
Me OO
H
60% b.r.s.m.
H Me 70 (5.4 grams stock)
H
OH
Me
H
H Me
(−)-64
14.3 Biomimetic Syntheses Involving the Diels–Alder Reaction O MeO
N Me
TBSO H
OTBS Me Me
Br Br
60 °C, CDCl3
H
H CHO
OTBS Me
OTBS Me OHC
OTBS 71
Me
TBSO H
Me
TBSO 72
Br
+ Me N OMe
H
H CHO
517
Me OTBS Me
Me
O
TBSO 72:73 = 1:1.7
O 73
Scheme 14.16 Stereoselectivity in the IMDA reaction of an acyclic system related to FR182877.
In the acyclic IMDA reaction of 71, only 1 : 1.7 diastereofacial selectivity (72 : 73) was indeed realized (Scheme 14.16). Similar results were reported by Sorensen and coworkers [64b]. During these works, no product from an exo transition state resembling exo-63‡ (Scheme 14.13) was observed, leaving the synthesis of hexacyclinic acid 65 as still elusive at this stage. Internal control elements such as stereocenters within the molecular architecture might be responsible for the orientation toward FR182877 (64). However, 65 has, as an extra particularity, a carboxylic acid function on ring B. This may add some electronic effects in the stereocontrol of the first TADA reaction, disfavoring the endo transition state. This carboxylic acid could thus be present at early stages during the biosynthesis of 65, especially on the biogenetic macrocyclic precursor. Cochleamycin A (77) was isolated in 1992 from Streptomyces DT136 by Shindo [65]. It showed antimicrobial activity against Gram-positive bacteria and was cytotoxic against various tumor cell lines [66]. Shindo suggested a biosynthetic route to cochleamycin A (77) and B (80) from a polyketide linear product (74) by an IMDA reaction (Scheme 14.17) [67]. This IMDA would give the 5,6-fused ring system (75) with the desired stereochemistry, through an endo transition state of a (9E,11Z)-diene, or an exo transition state of a (9Z,11E)-diene (not shown), giving in both alternative routes the cis junction of the natural products. An intramolecular condensation and further functionalization would achieve biosynthesis of the natural product. It is remarkable that macquarimicins (78, 79, 81) share the same carbocyclic skeleton and stereochemistry as cochleamycins, except the methyl substituent at C14 (Scheme 14.17). A high degree of stereocontrol may be necessary to achieve such an invariable selectivity. This could be controlled thanks to a biosynthetic TADA process on intermediate 76, which arises from intramolecular condensation of 74. Paquette and coworkers drew inspiration from Shindo’s biosynthetic hypothesis and succeeded in the highly enantioselective biomimetic IMDA reaction of the 5,6-fused bicyclic system of cochleamycin A (77) [68]. They used the (E,Z,E)-1,6,8-nonatriene (82), which reacted in toluene at 195 ◦ C via an endo transition state to furnish the bicyclic intermediate 83 (Scheme 14.18a). One
Me N OMe
10
11
OH
9
12
O
−H2O
IMDA
Y
O
H
O
O
76
17
16
H 75
O
H
[O]
O
H
9
12
[O]
OH
10
O
H O
R4 H reduction
OH
H R1=R3=H, R2=Me
O
H YH
R2 R3
R1=Me, R2=H, R3=R4=O, Y=H macquarimicin A (78)
R4=OAc, Y=H cochleamycin A (77)
TADA 11
−H2O
R1
O
R4 H
O
7
H
R2 R3
R4 H
R4=OAc, Y=H cochleamycin B (80)
H R1=R3=H, R2=Me
O
H H O Y
R1
R1=Me, R2=H, R3=R4=O, Y=acetonyl macquarimicin C (81)
alkylation
OH
H R1=Me, R2=H,
O
H
R2 R3
R3=R4=O, Y=acetonyl macquarimicin B (79)
HO
H Y H
R1
Biosynthetic hypothesis for cochleamycins and macquarimicins.
H 74 Y=H or acetonyl
O
O
16
Scheme 14.17
O
OHC
17
Y
[O]
H H OHC
[O]
518
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides
14.3 Biomimetic Syntheses Involving the Diels–Alder Reaction (a)
OPMB
HO OPMB
1) BHT (20%) PhMe, 195 °C, 26h 2) NaBH4
PivO O
PMP (b)
H
O
O 82
OMOM
OHC
OMOM OMOM
70%
H
H H H
SmI2
H THF H
EtO2C
MOMO
TBDPSO
84 OTBDPS
OMOM
H
EtO2C H MOMO
MOMO
OH
H cochleamycin A (77)
H Br
OHC MOMO
O
OMOM
H H
OAc H
O
PMP
OMOM
Yb(fod)3, BHT xylene, 140 °C 4h
H
O
O
83
Me H
H H
HO H PivO
66% (2 steps)
H
519
HO
OMOM OMOM
OHC 85
77
86
87
Scheme 14.18 Paquette’s partial synthesis of cochleamycin A (77) (a) and Tatsuta’s total synthesis (b).
year later, Tatsuta and coworkers achieved the total synthesis of 77 also using an IMDA reaction to form the bicyclic framework [69]. The triene (84) was heated at 140 ◦ C in the presence of Yb(fod)3 and BHT (butylated hydroxytoluene) in xylene, providing a single cycloadduct (85) towards the synthesis of 77 (Scheme 14.18b). Roush and coworkers reported in 2004 the biomimetic total synthesis of cochleamycin A (77) in 2.4% yield and 23 steps from 3-butene-1-ol, accomplished by TADA reaction of the macrocyclic intermediate 88 (Scheme 14.19) [70]. The substrate was heated at 125 ◦ C in toluene, providing the carbon skeleton 90 of the natural product as the sole cycloadduct in 69% yield, through the endo transition state 89. TESO
BHT
Me PhMe TESO O O
O
88
Scheme 14.19
TESO
125 °C 21h 69%
H
H O O
O
H
TESO 89
1) PPTS
H
Me 2) Ac O 2 H
Me
OTES
H
O
OTES
61% (2 steps)
O O
90
Roush synthesis of (+)-cochleamycin A (77).
Macquarimicins A–C (78, 79, 81, respectively, Scheme 14.17) were isolated in 1995 from Micromonospora chalcea [71]. Interesting biological properties have been associated with these compounds, such as selective inhibition of membrane-bound sphingomyelinase, anti-inflammatory, or cytotoxic activities [72]. Tadano and coworkers have reported extensive investigations on biomimetic IMDA and TADA
77
520
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides
reactions applied to macquarimicins [73]. To construct the tetrahydroindane ring with cis-anti-cis ring fusion by IMDA cycloaddition, the reaction of a (E,Z,E)-1,6,8-nonatriene proceeding through an endo mode or the reaction of a (E,E,Z)-1,6,8-nonatriene through an exo mode are two possible routes. In this context, Tadano and coworkers embarked on the synthesis of various IMDA and TADA substrates. A model study comparing the reactivity of (E,Z,E)- and (E,E,Z)-1,6,8-nonatrienes (91, 92) was performed (Scheme 14.20a). It implied the potential advantage of (E,Z,E)-triene 91 in the construction of the required cis-anti-cis ring junction (93). The (E,E,Z)-triene 92 gave indeed very poor selectivity in the IMDA reaction compared to 91 under thermal conditions. Moreover, four types of (E,Z,E)-trienes (94–97) were evaluated in the synthesis of macquarimicins, being linear or macrocyclic and incorporating a preformed lactone moiety or not (Scheme 14.20b). (a) Model IMDA studies OMPM
BHT, 150 °C sealed tube toluene
OTr
E Z
E
OHC
75%
91 OMPM
OHC
Z
H
OTr
93 + 5 isomers (1:1:1:0.1:0.1:0.1)
71%
92
OTr
H H OHC 93
BHT, 185 °C sealed tube toluene, 11d
E
E
OMPM H
(b) Scope of Diels–Alder substrates toward the synthesis of macquarimicins OMPM
CHO TBSO O
OMPM
O OTBS
OMPM
O
CHO O
OTBS
MeO2C
OP
OP
O
O
O
O 94
OMPM
MOMO
95
96
OH
97
O
Scheme 14.20 Tadano’s model study (a) and Diels–Alder substrates (b) for the synthesis of macquarimicins.
The best result was obtained with macrocyclic substrate 97 with the alkylidene lactone in place. The cycloaddition of 97 proceeded at 130 ◦ C in toluene, providing the desired diastereoisomer 98 as the sole cycloadduct in 47% yield (Scheme 14.21). Control of diastereofacial selectivity was favored by the presence of the lactone cycle. Subsequently, macquarimicin A (78) was synthesized after deprotections and oxidation of the alcohol at C13. Then one-step installment of the acetonyl moiety at C17 furnished macquarimicin B (79). Macquarimicin C (81) was, finally, formed in quantitative yield by acid-catalyzed intramolecular dehydrative alkylation of macquarimicin B (79).
14.3 Biomimetic Syntheses Involving the Diels–Alder Reaction OMPM
OMPM
OTBS
OTBS
OMPM H Me
1) HF-pyr, pyr 2) iPr2NEt, MeOH 3) PhSeCl, NEt3 4) mCPBA
MeO2C
BHT, toluene 130 °C sealed tube
O
96
O
H H O
OH
H H
H
OH 98 (47% from 96) O
H O O O H H (+)-macquarimicin C (81)
H
H CSA CH2Cl2
O
quant.
HO
OMe
H H
H H H
H O O
O O
O
H H
O
O + 97+
97
O
Scheme 14.21
H OMPM O
OH
O
O
O
521
THF/H2O 83%
OH
H (+)-macquarimicin B (79)
O
O O
OH
H (+)-macquarimicin A (78)
Tadano’s biomimetic synthesis of macquarimicins.
14.3.5 Biomimetic TADA Reactions toward Spinosyns
Spinosyns are produced by Saccharopolyspora spinosa originally collected in the Caribbean islands [74]. They are tetracyclic and contain a 12-membered macrocycle fused to a 5,6,5-tricyclic ring system. More than 25 spinosyns have been isolated to date, which constitute an important class of insecticidal polyketides. Their biosynthesis has been well studied and numerous analogs have been synthesized [75]. More recently, the entire spinosyn biosynthetic gene cluster was identified through gene sequencing and functional analysis of the gene products in S. spinosa [76]. The biosynthesis of spinosyns is expected to be initiated by the oxidation of the 15-OH group of the mature macrocyclic polyketide precursor 99 followed by conjugated dehydration (Scheme 14.22). The activated intermediate can then be involved in a TADA cycloaddition followed by a vinylogous Morita–Baylis–Hillman reaction, which would be mediated by an enzyme nucleophile, to deliver the tetracyclic spinosyn aglycone 100. Subsequent glycosylation steps install the sugars of spinosyns. The biomimetic synthesis of spinosyn A (101) has been studied by Roush and coworkers. It was known from Evans and Black that the diastereofacial selectivity of the IMDA reaction of linear substrates would favor an incorrect C7–C11 trans-fused diastereoisomer [77]. Thus, Roush first envisaged performing the TADA cyclization on a glycosylated macrocyclic intermediate 103a bearing an additional bromine substituent at C6 (Scheme 14.23) and arising from the Wittig–Horner–Emmons reaction of phosphonate 102a [78]. The glycosyl at C9 would allow control of the stereoselectivity at C7–C11 during cycloaddition, thanks
O
O
99
15 11 6
HO Et
oxidation, dehydration TADA reaction vinylogous MBH reaction
OH
SpnJ, L, M, F
Biosynthesis of spinosyns.
Me
Scheme 14.22
Et
HO
OH
O O
O
OH
100
HH
Me
11 7
HH
H
OH
glycosylation steps
Et
SpnG, I, K, H, P
O
O
O
O
HH
NMe2 Me
H spinosyn A (101)
H H
Me
O
O
MeO Me O OMe OMe
522
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides
X
EtO
EtO
P
O
O O O Et
Scheme 14.23
101 H
HH HH
O
O O
OH
6
Et
X
7
11
O O
O
OPMB
Roush’s total synthesis of spinosyn A (101).
RO
RO
0.005M, 23 °C; 2) (TMS)3SiH, AIBN; 3) DDQ
104b: 1) PMe3, t-amyl alcohol
104a: 1) PMe3, t-amyl alcohol 0.005M, 23 °C; 2) TFA
Et
Me
103a (75%) 103b (58%)
105 (64%): R = trimethylrhamnosyl (spinosyn A pseudoaglycon) 100 (40%): R = H (spinosyn A aglycon)
RO
Me
i Pr2NEt, LiCl CH3CN OPMB 0.001M RO 9 23 °C
O 102a: R = trimethylrhamnosyl, X = Br 102b: R = PMB, X = H
RO
Me
9
9
H
H
H
O
Et
O
(most abundant
O
O Et
OPMB
OPMB
O
O
Me
X stereoisomer) 104a (73:12:9:6 d.r.) 104b (70:18:12)
6
7 11
HH
Me
+ 103a + + 103b+
X 6
7
11
+ +
14.3 Biomimetic Syntheses Involving the Diels–Alder Reaction 523
524
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides
to steric interactions with bromine at C6 during the transition state 103a‡ . The TADA cyclization thus gave the cycloadduct 104a in 75% yield and good selectivity (dr 73 : 12 : 9 : 6), as a consequence of strain interactions involving the C6 and C9 substituents. Eventually, the presence of a silyloxy substituent on C8 proved to be effective in increasing the steric interactions and the selectivity (dr > 95 : 5) [79]. Although the total synthesis of 101 was accomplished from 102a, the same authors later demonstrated that a simpler substrate such as 103b can give a very similar distribution of TADA cycloadducts. Consequently, the synthesis of the spinosyn aglycon 100 was achieved from macrolactone 103b by TADA cycloaddition and PMe3 -catalyzed vinylogous Morita–Baylis–Hillman reaction on intermediate 104b, followed by PMB (p-methoxybenzyl) deprotection.
14.4 Biomimetic Cascade Reactions 14.4.1 A Metalated Ionophore Template for the Biomimetic Synthesis of Tetronasin
Tetronasin (108) is an ionophoric antibiotic isolated from Streptomyces longisporoflavus (Scheme 14.24) [80]. It shows strong affinity for sodium ions, which may be linked to its antibiotic activity. This affinity would be involved in the biosynthesis of the natural product 108 from a linear polyoxygenated polyketide 106 bearing the tetronic acid and tetrahydrofuran residues at the opposite termini. The biosynthesis and the total synthesis of tetronasin have been studied by Ley and Staunton [81, 82]. Their work culminated with Ley’s realization of the biomimetic step, that is, the one-step cyclization of a linear precursor analogous to 106 into the complete polycyclic core of tetronasin, using a metal-promoted folding of the chain as the directing process.
HO
O +
O O
Na O O Me
H O
H
O H O
H
HO
HO
O
H
H O
O H
H Na O O Me
O H
H
Scheme 14.24
107
Na
O O
O O
linear tetronic acid precuror (106)
H
H
O
O
O Me
H
tetronasin (108)
Biosynthetic origin of tetronasin (108).
From the late biosynthetic precursor 106, both tetrahydropyran and cyclohexane rings would indeed be formed simultaneously during a unique biosynthetic step [81d], which is mechanistically plausible according to stereoelectronic analysis of the cyclization process. The linear intermediate 106 would fold around the sodium salt of the tetronic acid residue with assistance of the various chelating
14.4 Biomimetic Cascade Reactions
oxygens of the polyketide. This would render the centers linked within bonding distance and the relevant orbitals correctly aligned for simultaneous cyclization of both six-membered rings. This pre-cyclization complex would adopt a low energy conformation with the metal in a central cavity, resembling that found in the natural product 108. This biosynthetic postulate has inspired Ley’s biomimetic synthesis of 108 [82]. The polyene substrate for the cyclization was synthesized from 109 and 110 (Scheme 14.25). It was designed with an electron-deficient diene system (C10–C13) to facilitate the pyran ring formation via Michael addition. The methoxymethyl-furan ring was left intact to enhance chelation. The tetronic acid part was not installed at this stage but simplified as a methyl carboxylate. Treatment of the polyene 111 with potassium hexamethyldisilazide (KHMDS) in toluene for 30 min at 0 ◦ C generated the potassium salt 111-K, the conformation of which was guided by intramolecular chelation, inducing the cascade cyclization. This challenging reaction gave one diastereoisomeric product (112) in very good yields (67%) by two subsequent conjugate additions, O17 → C13 then C10 → C5. The reaction generated two new rings and created four stereogenic centers simultaneously and with complete control. However, the configuration of the C4-methyl substituent in 112 was opposite to that found in tetronasin. Other conditions (metal bases, solvents, and temperature) were less satisfactory for the cyclization. Functional group interconversions finally gave the aldehyde 113, the epimerization of which was achieved in 85% yield in the presence of morpholine and catalytic p-TSA (p-toluenesulfonic acid). The product 114 was identical to that previously prepared by Yoshii [83] as a late-stage precursor in his total synthesis of tetronasin. In Yoshii’s total synthesis of tetronasin, the reaction of 114 with the diazoacetoacetate 115 in the presence of ZrCl4 (Scheme 14.26) gave the β-ketoester 116 [83]. Deprotection and Dieckmann cyclization of 116 finally installed the tetronic acid system as the sodium salt of tetronasin (108). This last cyclization step is also truly biomimetic. 14.4.2 The 6,5,6-Fused System and Macrocycle of Hirsutellones: Work Yet to Be Done?
The hirsutellones (e.g., 117 and 118) constitute a group of fungal metabolites with interesting medicinal properties (Figure 14.1). They were isolated in 2005 from Hirsutella nivea BCC 2594, an entomopathogenic fungus, and showed significant antimycobacterial activities compared to the reference drug isoniazid [84]. The group is related to other methylated analogs sharing the same carbocyclic core (119–121), yet with stereochemical variations: pyrrocidines [85], pyrrospirones [86], and GKK1032 compounds [87]. Structurally, hirsutellones feature a 6,5,6-fused tricyclic system and a 12- or 13-membered p-cyclophane macrocycle, the construction of which may be particularly difficult. Furthermore, a γ -lactam or a succinimide motif is embedded in the macrocycle.
525
CHO
4
H
109
O
O
H
H
O
O H
H
6
10
CH2Cl2 85%
H
morpholine TBSO pTSA
OMe
OMe
110
H
10 steps
CHO
4
4
H
17
O H
CO2Me
HO
13
113
H
H
O
O
H
111
H
21 KHMDS
OMe
6 steps
67%
OMe PhMe
26
Ley’s formal synthesis of tetronasin (108) (carbon numbering of tetronasin).
114
P(O)(OMe)2
Scheme 14.25
H
TBSO
BnO
+ OHC
EtO2C
H
EtO2C
H
O
K
O
H
wrong configuration
O
CO2Me
4
H
111-K Me
Ph
O
EtO2C
H
H
112
O
H O
OMe
526
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides
14.4 Biomimetic Cascade Reactions CHN2 O
TBSO
TBSO
O
H
CO2Me
(115)
114
ZrCl4, CH2Cl2 80%
O
1) TBAF O 2) HF, MeCN; H H OMe NaHCO3
H
CO2Me O
O H
H
O
H
H
O−Na+
H O
116
O
O
H
108
O
Final steps in Yoshii’s total synthesis of tetronasin (108). O
O
H
92%
H O O
Scheme 14.26
527
OH
OH
OH
NH
H
NH H
O
O H
NH
O H
H
O
O
H
H
O
NH
O O HH H
O
O H OH
H H
H H
hirsutellone A (117)
H
H
hirsutellone B (118)
pyrrocidine A (119)
pyrrospirone A (120)
OH O H
NH
O H
H O
H GKK1032A2 (121)
Figure 14.1
Structure of hirsutellones (117, 118) and related compounds (119–121).
The biosynthesis of this family of fungal products was described by Oikawa in 2003 [88]. The work was undertaken on the GKK1032 series (121) produced by Penicillium sp., using 13 C and 2 H labeling (Scheme 14.27). Oikawa demonstrated the nonaketide origin of the tricyclic moiety. Moreover, he showed that l-tyrosine was a precursor of the γ -lactam part while l-methionine stood for the methyl donor. A mixed PKS/NRPS (non-ribosomal peptide synthetase) enzyme would be involved. Of particular interest was the biogenetic cyclization mechanism suggested by Oikawa to make the polycyclic system from the linear precursor 122. After oxidation of 122, a cationic polycyclization process on intermediate 123 would occur. This may eventually be concerted. However, a stepwise mechanism including an intramolecular Diels–Alder reaction to close the cycle C is not excluded. Several synthetic works toward hirsutellones have been reported recently. On the occasion of the isolation of the dimeric hirsutellone F (125), Isaka and coworkers obtained the putative biosynthetic intermediate 126 by cleavage of 125 in a 1-M NaOH solution (Scheme 14.28) [89]. This monomeric compound was poorly stable, tending to rearrange into hirsutellone A (117) during basic treatment, by
OMe
528
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides OH
CH3CO2Na NH2
NH
O
O
H
H
O D
CO2H
S HO
NH2
D
H
CO2H
121
OH
OH OH
NH
HO
NH
HO
O
[O]
O
O
O
H
O
B
A 122
[H]
121
O C
H 124
123
Scheme 14.27
NH
O H
Biosynthetic origin of the GKK1032 compounds and cyclization mechanism. a) insitu 1,2-shift OH
HH
OH
1′
O H
O H
NH O
O
H
H O N
H O
O H
O
H
H
NaOH 1N
O
H O
2′
NH O
H O
H
O H
β-elimination
HH
HH
hirsutellone A (117)
OH NH
O hirsutellone C (127)
Scheme 14.28
O
HH
biosynthetic intermediate (126)
hirsutellone F (125)
O
NH H
OH b) epoxidation H2O2
c) reduction
NH
NaBH4
H O hirsutellone B (118)
Biomimetic conversion of intermediate 126 into hirsutellones A–C.
1,2-migration of the benzyl group to form the succinimide ring. In the presence of a reducing agent (NaBH4 ), 126 afforded good yields of hirsutellone B (118), while epoxidation with H2 O2 gave hirsutellone C (127). This work confirmed Oikawa’s biogenetic hypotheses on the formation of the various lactam derivatives [88]. There is still no biomimetic total synthesis of hirsutellones with regard of the construction of the 6,5,6-fused tricyclic moiety. However, Nicolaou’s total synthesis of hirsutellone B (118) [90] is remarkably reminiscent of Oikawa’s cascade cyclization leading to the GKK1032A2 carbocyclic core (Scheme 14.29) [88]. Starting from the epoxide intermediate 131, electrophilic activation in the presence of Et2 AlCl provided diastereoselectively the tricyclic core 134 of hirsutellones in 50% yield. The reaction proceeded by intramolecular epoxide opening and
O
TMS
Ph3P CHI
[O]
128
O OH
AH O
isomeriz
O
129
O
NH H
O
132
TMS
CO2Me
Et2AlCl
131
TMS
CO2Me
Cl
−78 °C 25 °C 50%
Et2AlCl, CH2Cl2
H
Et2Al
H 133
CO2Me
H
18 steps
O
NH
OH
130
O
H CO2Me
134
HH
O
H
OH
(b) Nicolaou's total synthesis of hirsutellone B (118) O
Et2Al O
O
NH
OH
H
HH
O
H
124
O
118
121
O
NH
OH
Scheme 14.29 (a) Oikawa’s alternative hypothesis for the biosynthesis of GKK1032A2 (121) in relation to (b) Nicolaou’s total synthesis of hirsutellone B (118).
nBu3Sn
O
122
HO
OH
(a) Oikawa's alternative biogenetic hypothesis for GKK1032A2 compound 121
14.4 Biomimetic Cascade Reactions 529
530
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides
electrophilic cyclization into 133, followed by Diels–Alder reaction through an exclusive endo transition state leading to 134. Interestingly, in this mechanistic scheme, the electrophilic activation is inversed in comparison to the oxidative activation performed on the biogenetic precursor 122 (Scheme 14.27).
14.5 Conclusion
Theoretical analysis of polyketide biosynthesis has suggested that over a billion possible structures could be synthesized from PKS enzymatic models [91]. As yet, however, only 10 000 polyketide structures have been discovered. The present chapter deals only with biomimetic rearrangements in the carbocyclic non-aromatic series. Many important studies around polyketides are described in other chapters of this book. The structural complexity is an important character of polyketides and we know that it is related to chemical diversity. Considering the important biological properties of polyketides and the potential number of polyketide ‘‘privileged structures’’ [92], synthetic analogs will be valuable drug candidates for medicinal research of the twenty-first century. The application of diversity oriented methods to polyketide synthesis is an important issue in meeting this objective [93]. The power of this method has already been demonstrated, especially when using the biomimetic strategy [94]. Yet, an important alternative methodology emerged with the advent of the combinatorial biosynthesis of polyketides. The manipulation of gene clusters coding for each PKS modules and their combination or hybridization has provided unprecedented polyketide libraries [95]. Joining both chemical and biochemical technologies promises important discoveries in the next few decades.
References 1. (a) Chan, Y.A., Podevels, A.M., Kevany,
3. (a) Collie, J.N. and Myers, W.S. (1893)
B.M., and Thomas, M.G. (2009) Nat. Prod. Rep., 26, 90–114; (b) Smith, S. and Tsai, S.-C. (2007) Nat. Prod. Rep., 24, 1041–1072; (c) Hertweck, C., Luzhetskyy, A., Rebets, Y., and Bechthold, A. (2007) Nat. Prod. Rep., 24, 162–190; (d) Fischbach, M.A. and Walsh, C.T. (2006) Chem. Rev., 106, 3468–3496; (e) Staunton, J. and Weissman, K.J. (2001) Nat. Prod. Rep., 18, 380–416; (f) Hertweck, C. (2009) Angew. Chem. Int. Ed., 48, 4688–4716. 2. Keatinge-Clay, A.T., Maltby, D.A., Medzihradszky, K.F., Khosla, C., and Stroud, R.M. (2004) Nat. Struct. Biol., 11, 888–893.
J. Chem. Soc., 63, 122–128; (b) Collie, J.N. (1907) Proc. Chem. Soc., 23, 230–231. 4. Harris, T.M. and Harris, C.M. (1986) Pure Appl. Chem., 58, 283–294. 5. (a) Kobuke, Y. and Yoshida, J. (1978) Tetrahedron Lett., 19, 367–370; (b) Sakai, N., Sord´e, N., and Matile, S. (2001) Molecules, 6, 845–851; (c) Chen, H. and Harrison, P.H.M. (2002) Can. J. Chem., 80, 601–607; (d) Ji, Q., Williams, H.J., Roessner, C.H., and Scott, A.I. (2007) Tetrahedron Lett., 48, 8026–8028. 6. Wijkman, N. (1931) Liebigs Ann. Chem., 485, 61–73.
References 7. Baldwin, J.E., Barton, D.H.R.,
8.
9.
10.
11.
12.
13.
14.
15.
Bloomer, J.L., Jackman, L.M., Rodriguez-Hahn, L., and Sutherland, J.K. (1962) Experientia, 18, 345–388. For Sutherland’s work on structure elucidation, see: (a) Sutherland, J.K. and Barton, D.H.R. (1965) J. Chem. Soc., 1769–1771; (b) Barton, D.H.R., Jackman, L.M., Rodriguez-Hahn, L., and Sutherland, J.K. (1965) J. Chem. Soc., 1772–1778; (c) Barton, D.H.R., Godinho, L.D.S., and Sutherland, J.K. (1965) J. Chem. Soc., 1779–1786; (d) Baldwin, J.E., Barton, D.H.R., and Sutherland, J.K. (1965) J. Chem. Soc., 1787–1798. For Sutherland’s work on structure elucidation, see: (a) Bloomer, J.L., Moppett, C.E., and Sutherland, J.K. (1968) J. Chem. Soc. C, 588–591; (b) Huff, R.K., Moppett, C.E., and Sutherland, J.K. (1972) J. Chem. Soc., Perkin Trans. 1, 2584–2590; (c) Moppett, C.E. and Sutherland, J.K. (1966) J. Chem. Soc., Chem. Commun., 772–773. Huff, R.K., Moppett, C.E., and Sutherland, J.K. (1968) J. Chem. Soc., Chem. Commun., 1192–1193. Baldwin, J.E., Beyeler, A., Cox, R.J., Keats, C., Pritchard, G.J., Adlington, R.M., and Watkin, D.J. (1999) Tetrahedron, 55, 7363–7374. Baldwin, J.E., Adlington, R.M., Roussi, F., Bulger, P.G., Marquez, R., and Mayweg, V.W. (2001) Tetrahedron, 57, 7409–7416. (a) Dabrah, T.T., Harwood, H.J., Huang, L.H., Jankovich, N.D., Kaneko, T., Li, J.C., Lindsey, S., Moshier, P.M., Subashi, T.A., Therrien, M., and Watts, P.C. (1997) J. Antibiot., 50, 1–50; (b) Dabrah, T.T., Kaneko, T., Massefski, W., and Whipple, E.B. (1997) J. Am. Chem. Soc., 119, 1594–1598. Spencer, P., Agnelli, F., Williams, H.J., Keller, N.P., and Sulikowski, G.A. (2000) J. Am. Chem. Soc., 122, 420–421. Sulikowski, G.A., Agnelli, F., Spencer, P., Koomen, J.M., and Russell, D.H. (2002) Org. Lett., 4, 1447–1450.
16. (a) Sulikowski, G.A., Agnelli, F., and
17. 18.
19.
20.
21. 22.
23. 24. 25.
26.
Corbett, R.M. (2000) J. Org. Chem., 65, 337–342; (b) Sulikowski, G.A., Liu, W., Agnelli, F., Corbett, R.M., Luo, Z., and Hershberger, S.J. (2002) Org. Lett., 4, 1451–1454. Diels, O. and Alder, K. (1928) Justus Liebigs Ann. Chem., 460, 98–122. Takao, K.-I., Munakata, R., and Tadano, K.-I. (2005) Chem. Rev., 105, 4779–4807. (a) Oikawa, H. and Tokiwano, T. (2004) Nat. Prod. Rep., 21, 321–352; (b) Stocking, E.M. and Williams, R.M. (2003) Angew. Chem. Int. Ed., 42, 3078–3115. Alberts, A.W., Chen, J., Kuron, G., Hunt, V., Huff, J., Hoffman, C., Rothrock, J., Lopez, M., Joshua, H., Harris, E., Patchett, A., Monaghan, R., Currie, S., Stapley, E., Albers-Sch¨onberg, G., Hensens, O., Hirschfield, J., Hoogsteen, K., Liesch, J., and Springer, J. (1980) Proc. Natl. Acad. Sci. U.S.A., 77, 3957–3961. Endo, A. (1979) J. Antibiot., 32, 852–854. (a) Chan, J.K., Moore, R.N., Nakashima, T.T., and Vederas, J.C. (1983) J. Am. Chem. Soc., 105, 3334–3336; (b) Moore, R.N., Bigam, G., Chan, J.K., Hogg, A.M., Takashima, T.T., and Vederas, J.C. (1985) J. Am. Chem. Soc., 107, 3694–3701; (c) Yoshizawa, Y., Witter, D.J., Liu, Y., and Vederas, J.C. (1994) J. Am. Chem. Soc., 116, 2693–2694. Endo, A. and Hasumi, K. (1993) Nat. Prod. Rep., 10, 541–550. Witter, D.J. and Vederas, J.C. (1996) J. Org. Chem., 61, 2613–2623. Auclair, K., Sutherland, A., Kennedy, J., Witter, D.J., van den Heever, J.P., Hutchinson, C.R., and Vederas, J.C. (2000) J. Am. Chem. Soc., 122, 11519–11520. (a) Ichihara, A., Tazaki, H., and Sakamura, S. (1983) Tetrahedron Lett., 24, 5373–5376; (b) Alam, S.S., Bilton, J.N., Slawin, A.M.Z., Williams, D.J., Sheppard, R.N., and Strange, R.N. (1989) Phytochemistry, 28, 2627–2630; (c) Oikawa, H., Yokota, T., Sakano, C., Suzuki, Y., Naya, A., and Ichihara, A.
531
532
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides
27.
28.
29.
30.
31.
32.
33.
34.
35.
(1998) Biosci. Biotechnol. Biochem., 62, 2016–2022; (d) Mizushina, Y., Kamisuki, S., Kasai, N., Shimazaki, N., Takemura, M., Asahara, H., Linn, S., Yoshida, S., Matsukage, A., Koiwai, O., Sugawara, F., Yoshida, H., and Sakaguchi, K. (2002) J. Biol. Chem., 277, 630–638. (a) Oikawa, H., Yokota, T., Abe, T., Ichihara, A., Sakamura, S., Yoshizawa, Y., and Vederas, J.C. (1989) J. Chem. Soc., Chem. Commun., 1282–1284; (b) Oikawa, H., Yokota, T., Ichihara, A., and Sakamura, S. (1989) J. Chem. Soc., Chem. Commun., 1284–1285. (a) Oikawa, H., Suzuki, Y., Naya, A., Katayama, K., and Ichihara, A. (1994) J. Am. Chem. Soc., 116, 3605–3606; (b) Oikawa, H., Suzuki, Y., Katayama, K., Naya, A., Sakano, C., and Ichahara, A. (1999) J. Chem. Soc., Perkin Trans. 1, 1225–1232. (a) Oikawa, H., Katayama, K., Suzuki, Y., and Ichahara, A. (1995) J. Chem. Soc., Chem. Commun., 1321–1322; (b) Katayama, K., Kobayashi, T., Oikawa, H., Honma, M., and Ichihara, A. (1998) Biochim. Biophys. Acta, 1383, 387–395. Ichihara, A., Miki, M., Tazaki, H., and Sakamura, S. (1987) Tetrahedron Lett., 28, 1175–1178. Oikawa, H., Kobayashi, T., Katayama, K., Suzuki, Y., and Ichihara, I. (1998) J. Org. Chem., 63, 8748–8756. Katayama, K., Kobayashi, T., Chijimatsu, M., Ichihara, A., and Oikawa, H. (2008) Biosci. Biotechnol. Biochem., 72, 604–607. Wilson, R.M., Jen, W.S., and MacMillan, D.W.C. (2005) J. Am. Chem. Soc., 127, 11616–11617. (a) Celmer, W.D., Chmurny, G.N., Moppett, C.E., Ware, R.S., Watts, P.C., and Whipple, E.B. (1980) J. Am. Chem. Soc., 102, 4148883; (b) Celmer, W.D., Cullen, W.P., Moppett, C.E., Jefferson, M.T., Huang, L.H., Shibakawa, R., and Tone, J. (1979) US Patent 4, 148,883 (a) Cane, D.E. and Yang, C.-C. (1984) J. Am. Chem. Soc., 106, 784–787;
36.
37. 38.
39.
40.
41.
42.
43.
44.
45.
46. 47.
(b) Cane, D.E. and Yang, C.-C. (1985) J. Antibiot., 38, 423–426. (a) Cane, D.E., Tan, W., and Ott, W.R. (1993) J. Am. Chem. Soc., 115, 527–535; (b) Cane, D.E. and Ott, W.R. (1988) J. Am. Chem. Soc., 110, 4840–4841; (c) Cane, D.E. and Luo, G. (1995) J. Am. Chem. Soc., 117, 6633–6634. Coe, J.W. and Roush, W.R. (1989) J. Org. Chem., 54, 915–930. Roush, W.R., Koyama, K., Curtin, M.L., and Moriarty, K.J. (1996) J. Am. Chem. Soc., 118, 7502–7512. (a) D’Auria, M.V., Debitus, C., Paloma, L.G., Minale, L., and Zampella, A. (1994) J. Am. Chem. Soc., 116, 6658–6663; (b) D’Auria, M.V., Paloma, L.G., Minale, L., Zampella, A., and Debitus, C. (1994) J. Nat. Prod., 57, 1595–1597. (a) Tortosa, M., Yakelis, N.A., and Roush, W.R. (2008) J. Am. Chem. Soc., 130, 2722–2723; (b) Tortosa, M., Yakelis, N.A., and Roush, W.R. (2008) J. Org. Chem., 73, 9657–9667. Brown, R.F.C., Drummond, R., Fogerty, A.C., Hughes, G.K., Pinhey, J.T., Ritchie, E., and Taylor, W.C. (1956) Aust. J. Chem., 9, 283–287. Baldwin, J.E., Chesworth, R., Parker, J.S., and Russell, A.T. (1995) Tetrahedron Lett., 36, 9551–9554. (a) Tchabanenko, K., Chesworth, R., Parker, J.S., Anand, N.K., Russell, A.T., Adlington, R.M., and Baldwin, J.E. (2005) Tetrahedron, 61, 11649–11656; (b) Tchabanenko, K., Adlington, R.M., Cowley, A.R., and Baldwin, J.E. (2005) Org. Lett., 7, 585–588. Huang, X.-H., van Soest, R., Roberge, M., and Andersen, R.J. (2004) Org. Lett., 6, 75–78. Kirkham, J.E.D., Lee, V., and Baldwin, J.E. (2006) Chem. Commun., 2863–2865. Crossman, J.S. and Perkins, M.V. (2008) Tetrahedron, 64, 4852–4867. (a) Hautzel, R. and Anke, H. (1990) Z. Naturforsch., Teil C, 45, 68–73; (b) K¨opcke, B., Johansson, M., Sterner, O., and Anke, H. (2002) J. Antibiot., 55, 36–40; (c) Johansson, M., K¨opcke, B., Anke, H., and Sterner, O. (2002) J. Antibiot., 55, 104–106.
References 48. Weidler, M., Rether, J., Anke, T., and 49.
50.
51.
52. 53.
54.
55.
56.
57.
58.
59.
Erkel, G. (2000) FEBS Lett., 484, 1–6. Hellsten, R., Johansson, M., Dahlman, A., Dizeyi, N., Sterner, O., and Bjartell, A. (2008) Prostate, 68, 269–280. Steglich, W., Eizenh¨ofer, T., Casser, I., Steffan, B., Rabe, U., Boeker, R., Knerr, H.J., Anke, H., and Anke, T. (1993) DECHEMA Monograph, vol. 129, (ed. T. Anke and U. Onken), VCH, Weinheim, pp. 3–13. (a) Johansson, M., K¨opcke, B., Anke, H., and Sterner, O. (2002) Angew. Chem. Int. Ed., 41, 2158–2160; (b) Johansson, M., K¨opcke, B., Anke, H., and Sterner, O. (2002) Tetrahedron Lett., 58, 2523–2528. Lebel, H. and Parmentier, M. (2007) Org. Lett., 9, 3563–3566. (a) Bister, B., Bischoff, D., Str¨obele, M., Riedlinger, J., Reicke, A., Wolter, F., Bull, A.T., Z¨ahner, H., Fielder, H.-P., and S¨ussmuth, R.D. (2004) Angew. Chem. Int. Ed., 43, 2574–2576; (b) Riedlinger, J., Reicke, A., Z¨ahner, H., Krismer, B., Bull, A.T., Maldonado, L.A., Ward, A.C., Goodfellow, M., Bister, B., Bischoff, D., S¨ussmuth, R.D., and Fiedler, H.-P. (2004) J. Antibiot., 57, 271–279. Zapf, C.W., Harrison, B.A., Drahl, C., and Sorensen, E.J. (2005) Angew. Chem. Int. Ed., 44, 6533–6537. (a) Nicolaou, K.C. and Harrison, S.T. (2007) J. Am. Chem. Soc., 129, 429–440; (b) Nicolaou, K.C., Harrison, S.T., and Chen, J.S. (2009) Synthesis, 129, 33–42. Keler, S., Nicholson, G., Drahl, C., Sorensen, E.J., Fiedler, H.-P., and S¨ussmuth, R.D. (2007) J. Antibiot., 60, 391–394. Seki, T., Satake, M., Mackenzie, L., Kaspar, H.F., and Yasumoto, T. (1995) Tetrahedron Lett., 36, 7093–7096. Uemura, D., Chou, T., Haino, T., Nagatsu, A., Fukuzawa, S., Zheng, S.Z., and Chen, H.S. (1995) J. Am. Chem. Soc., 117, 1155–1156. Johannes, J.W., Wenglowsky, S., and Kishi, Y. (2005) Org. Lett., 7, 3997–4000.
60. (a) Sato, B., Muramatsu, H.,
61.
62.
63.
64.
65. 66.
67.
68. 69.
70. 71.
Miyauchi, M., Hori, Y., Takase, S., Hino, M., Hashimoto, S., and Terano, H. (2000) J. Antibiot., 53, 123–130; (b) Sato, B., Nakajima, H., Hori, Y., Hino, M., Hashimoto, S., and Terano, H. (2000) J. Antibiot., 53, 204–206; (c) Yoshimura, S., Sato, B., Kinoshita, T., Takase, S., and Terano, H. (2000) J. Antibiot., 53, 615–622. Yoshimura, S., Sato, B., Kinoshita, T., Takase, S., and Terano, H. (2002) J. Antibiot., 55, C1. H¨ofs, R., Walker, M., and Zeeck, A. (2000) Angew. Chem. Int. Ed., 39, 3258–3261. (a) Evans, D.A. and Starr, J.T. (2002) Angew. Chem. Int. Ed., 41, 1787–1790; (b) Evans, D.A. and Starr, J.T. (2003) J. Am. Chem. Soc., 125, 13531–13540. (a) Vanderwal, C.D., Vosburg, D.A., Weiler, S., and Sorensen, E.J. (1999) Org. Lett., 1, 645–648; (b) Vanderwal, C.D., Vosburg, D.A., and Sorensen, E.J. (2001) Org. Lett., 3, 4307–4310; (c) Vosburg, D.A., Vanderwal, C.D., and Sorensen, E.J. (2002) J. Am. Chem. Soc., 124, 4552–4553; (d) Vanderwal, C.D., Vosburg, D.A., Weiler, S., and Sorensen, E.J. (2003) J. Am. Chem. Soc., 125, 5393–5407. Shindo, K. and Kawai, H. (1992) J. Antibiot., 45, 292–295. (a) Shindo, K., Matsuoka, M., and Kawai, H. (1996) J. Antibiot., 49, 241–243; (b) Shindo, K., Iijima, H., and Kawai, H. (1996) J. Antibiot., 49, 244–248. Shindo, K., Sakakibara, M., and Kawai, H. (1996) J. Antibiot., 49, 249–252. Chang, J.Y. and Paquette, L.A. (2002) Org. Lett., 4, 253–256. Tatsuta, K., Narazaki, F., Kashiki, N., Yamamoto, J., and Nakano, S. (2003) J. Antibiot., 56, 584–590. Dineen, T.A. and Roush, W.R. (2004) Org. Lett., 6, 2043–2046. (a) Jackson, M., Karwowski, J.P., Theriault, R.J., Rasmussen, R.R., Hensey, D.M., Humphrey, P.E., Swanson, S.J., Barlow, G.J., Premachandran, U., and McAlpine, J.B.
533
534
14 Biomimetic Synthesis of Non-Aromatic Polycyclic Polyketides
72.
73.
74. 75.
76.
77. 78.
79.
80.
81.
(1995) J. Antibiot., 48, 462–466; (b) Hochlowski, J.E., Mullally, M.M., Henry, R., Whittern, D.M., and McAlpine, J.B. (1995) J. Antibiot., 48, 467–470. Tanaka, M., Nara, F., Yamasato, Y., Masuda-Inoue, S., Doi-Yoshioka, H., Kumakura, S., Enokita, R., and Ogita, T. (1999) J. Antibiot., 52, 670–673. (a) Munakata, R., Katakai, H., Ueki, T., Kurosaka, J., Takao, K., and Tadano, K. (2003) J. Am. Chem. Soc., 125, 14722–14723; (b) Munakata, R., Katakai, H., Ueki, T., Kurosaka, J., Takao, K., and Tadano, K. (2004) J. Am. Chem. Soc., 126, 11254–11267. Mertz, F.P. and Yao, R.C. (1990) Int. J. Syst. Bacteriol., 37, 19–22. For a review on the biochemistry of spinosyns, see: Huang, K.-X., Xia, L., Zhang, Y., Ding, X., and Zahn, J.A. (2009) Appl. Microbiol. Biotechnol., 82, 13–23. Waldron, C., Matsushima, P., Rostek, P.R. Jr., Broughton, M.C., Turner, J., Madduri, K., Crawford, K.P., Merlo, D.J., and Baltz, R.H. (2001) Chem. Biol., 8, 487–499. Evans, D.A. and Black, W.C. (1993) J. Am. Chem. Soc., 115, 4497–4513. Mergott, D.J., Franck, S.A., and Roush, W.R. (2004) Proc. Natl. Acad. Sci. U.S.A., 101, 11955–11959. Winbush, S.M., Mergott, D.J., and Roush, W.R. (2008) J. Org. Chem., 73, 1818–1829. Davies, D.H., Snape, E.W., Suter, P.J., King, T.J., and Falshaw, C.P. (1981) J. Chem. Soc., Chem. Commun., 1073–1074. For biosynthetic studies of tetronasin, see: (a) Bulsing, J.M., Laue, E.D., Leeper, F.J., Staunton, J., Davies, D.H., Ritchie, G.A.F., Davies, A., Davies, A.B., and Mabelis, R.P. (1984) J. Chem. Soc., Chem. Commun., 1301–1302; (b) Doddrell, D.M., Laue, E.D., Leeper, F.J., Staunton, J., Davies, A., Davies, A.B., and Ritchie, G.A.F. (1984) J. Chem. Soc., Chem. Commun., 1302–1304; (c) Demetriadou, A.K., Laue, E.D., Staunton, J., Ritchie, G.A.F., Davies, A., and Davies, A.B.
82.
83.
84.
85.
86.
87.
88. 89.
(1985) J. Chem. Soc., Chem. Commun., 408–410; (d) Hailes, H.C., Jackson, C.M., Leadlay, P.F., Ley, S.V., and Staunton, J. (1994) Tetrahedron Lett., 35, 307–310; (e) Hailes, H.C., Handa, S., Leadlay, P.F., Lennon, I.C., Ley, S.V., and Staunton, J. (1994) Tetrahedron Lett., 35, 311–314; (f) Hailes, H.C., Handa, S., Leadlay, P.F., Lennon, I.C., Ley, S.V., and Staunton, J. (1994) Tetrahedron Lett., 35, 315–318; (g) Boons, G.-J., Clase, J.A., Lennon, I.C., Ley, S.V., and Staunton, J. (1995) Tetrahedron, 51, 5417–5446; (h) Less, S.L., Leadlay, P.F., Dutton, C.J., and Staunton, J. (1996) Tetrahedron Lett., 37, 3519–3520. For biomimetic synthesis studies of tetronasin, see: (a) Boons, G.-J., Lennon, I.C., Ley, S.V., Owen, E.S.E., Staunton, J., and Wadsworth, D.J. (1994) Tetrahedron Lett., 35, 323–326; (b) Ley, S.V., Brown, D.S., Clase, J.A., Fairbanks, A.J., Lennon, I.C., Osborn, H.M.I., Stokes, E.S.E., and Wadsworth, D.J. (1998) J. Chem. Soc., Perkin Trans. 1, 2259–2276. Hori, K., Kazuno, H., Nomura, K., and Yoshii, E. (1993) Tetrahedron Lett., 34, 2183–2186. Isaka, M., Rugseree, N., Maithip, P., Kongsaeree, P., Prabpai, S., and Thebtaranonth, Y. (2005) Tetrahedron, 61, 5577–5583. He, H.Y., Yang, H.-Y., Bigelis, R., Solum, E.H., Greenstein, M., and Carter, G.T. (2002) Tetrahedron Lett., 43, 1633–1636. Shiono, Y., Shimanuki, K., Hiramatsu, F., Koseki, T., Tetsuya, M., Fujisawa, N., and Fimura, K.-I. (2008) Bioorg. Med. Chem. Lett., 18, 6050–6053. Koizumi, F., Hasegawa, A., Ando, K., Ogawa, T., Hara, M., and Yoshida, M. (2001) Jpn. Kokai Tokkyo Koho JP 2, 001,247,574; Chem. Abstr., 135 (2001) 209979. Oikawa, H. (2003) J. Org. Chem., 68, 3552–3557. Isaka, M., Prathumpai, W., Wongsa, P., and Tanticharoen, M. (2006) Org. Lett., 8, 2815–2817.
References 90. Nicolaou, K.C., Sarlah, D., Wu, T.R.,
94. (a) Pelish, H.E., Westwood, N.J.,
and Zhan, W. (2009) Angew. Chem. Int. Ed., 48, 6870–6874. 91. Gonz´ales-Lergier, J., Broadbelt, L.J., and Hatzimanikatis, V. (2005) J. Am. Chem. Soc., 127, 9930–9938. 92. Breibauer, R., Vetter, I.R., and Waldmann, H. (2002) Angew. Chem. Int. Ed., 41, 2878–2890. 93. Burke, M.D. and Figreiber, S.L. (2004) Angew. Chem. Int. Ed., 43, 46–58.
Feng, Y., Kirchhausen, T., and Shair, M.D. (2001) J. Am. Chem. Soc., 123, 6740–6741; (b) Lindsley, C.W., Chan, L.K., Goess, B.C., Joseph, R., and Shair, M.D. (2000) J. Am. Chem. Soc., 122, 422–423. 95. Xue, Q., Ashley, G., Hutchinson, C.R., and Santi, D.V. (1999) Proc. Natl. Acad. Sci. U.S.A., 96, 11740–11745.
535
537
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening Ivan Vilotijevic and Timothy F. Jamison
15.1 Introduction
The simple ether, a C–O–C motif, appears in nearly all families of oxygencontaining natural products, which are synthesized by organisms in all kingdoms of life. A subgroup of natural products characterized by the regular occurrence of multiple C–O–C motifs is designated the polyether family, and these can be broadly divided into linear and polycyclic polyethers. The latter group is of special interest due to its structural diversity and the biological activity of its members, which ranges from antibiotic, antifungal, and anticancer properties to extreme toxicity. Distinctive structural elements of polycyclic polyethers can be used as a foundation for their classification. Depending on these particular structural features, which can generally be traced back to the biosynthetic pathways for the synthesis of these molecules, polycyclic polyethers are divided into two main groups. The first group includes molecules with multiple fused cyclic ethers that are postulated to be formed in nature via all-endo cascades of epoxide openings. The other group consists of molecules that are produced via all-exo biosynthetic cascades of epoxide-opening reactions and normally feature multiple rings that are interconnected by a carbon–carbon bond. Some polyethers are produced via cascades that feature both endo and exo epoxide openings. Further distinction between different classes of polycyclic polyethers is made based on their biosynthetic origin; these natural products are either polyketide- or terpene-derived. For the purposes of this chapter, polycyclic polyethers will be classified in three major groups: polyether ionophores [1], squalene-derived polyethers [2], and the ladder polyethers [3]. Each of these groups will be discussed in the context of epoxide-opening cascades. Such reactions are postulated to be involved in their biosynthesis [4] and have been utilized as a method to rapidly construct polyether frameworks in the total synthesis of polycyclic polyethers [5].
Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
538
15.2 Synthetic Considerations: Baldwin’s Rules
Regioselectivity and stereospecificity in epoxide-opening reactions are the main determinants of the product composition in epoxide-opening cascades and intramolecular epoxide-opening reactions in general. The vast majority of epoxide ring-opening reactions proceed with inversion of configuration. Regioselectivity, however, is far more variable and case-dependant. Baldwin’s rules of ring closure are a three-criteria classification of ring-closing reactions based on both empirical results and theoretical considerations, and are used to predict the outcome of intramolecular ring-forming reactions, including epoxide openings. The criteria are: the size of the formed ring, the position of the bond that is broken relative to the smallest formed ring, and the geometry of the electrophile [6]. If the position of the bond broken during the ring closing reaction is exocyclic, outside of the formed ring, then the reaction is classified as exo. If the broken bond is within the smallest formed ring, the reaction is classified as endo (Figure 15.1a). In Baldwin’s classification, reactions involving sp3 hybridized electrophiles are described as tet due to the tetragonal geometry of the electrophile, sp2 hybridized electrophiles are trig, and sp electrophiles are diagonal or dig. With such classification in mind, Baldwin formulated a simple set of guidelines to predict the relative feasibility of different ring closing reactions [6]. Although empirical, Baldwin’s rules are conceptually based on stereoelectronic considerations [7]. Favored ring-closing reactions are those in which the length and nature of the linking chain enable the terminal atoms to achieve the proper geometries for the reaction. Disfavored ring closings, on the other hand, generally require severe distortions of bond angles and bond distances. For instance, 4-exo-trig reactions are predicted to be favored over 5-endo-trig ring closing reactions (Figure 15.1a). With few exceptions, intramolecular epoxide-opening reactions favor the smaller heterocycle (e.g., tetrahydrofuran 5, likely arising from a spiro transition state, Figure 15.1b), not the larger one (tetrahydropyran 6, from fused transition state, Figure 15.1b) [8]. Baldwin designates products that arise via fused and spiro 4-exo -trig
‡
‡ Y X
A 1
Y X
spiro 5-exo -tet
5-endo -trig Y X
or
A
A outside of the formed ring
inside the formed ring
fused 6-endo -tet ‡
O OH 4
O
‡ or
O H
O O H
outside of the formed ring
outside of the formed ring
OH A Y X (a)
2 favored
A X
Y
3 disfavored
O (b)
Figure 15.1 (a) Baldwin’s rules: general classification; (b) Baldwin’s rules in intramolecular epoxide-opening reactions [6].
5 favored
OH
O 6 disfavored
15.3 Polycyclic Polyethers: Structure and Biosynthesis
transition states as endo and exo, respectively. However, because the epoxide C–O bond that breaks is outside of the newly formed ring in both cases, each can be considered an exo process under the same construct (Figure 15.1b). To avoid potential confusion, the distinct terms ‘‘fused’’ and ‘‘spiro’’ can be used to describe the transition states in epoxide-opening reactions [9]. Intramolecular epoxide-opening reactions tend to follow the rules that lie between those for tetrahedral and trigonal systems, generally favoring what are usually termed the exo processes, that is, those that proceed via a spiro transition state [6]. 15.2.1 Control of Regioselectivity in Intramolecular Epoxide-Opening Reactions
The development of efficient methods for enantioselective epoxidation such as the Sharpless asymmetric epoxidation [10, 11], Jacobsen epoxidation [12, 13], and the Shi epoxidation [14–16] make epoxides attractive intermediates in asymmetric synthesis [17]. These methods have enabled syntheses of many of the polyepoxides that will be discussed herein, and have thus accelerated investigations of epoxide-opening cascades. For epoxides to be versatile synthetic intermediates, effective ways to control the regioselectivity in epoxide-opening reactions are necessary. The exo mode of cyclization is typically preferred; therefore, methods to facilitate endo cyclization have constituted a particularly active area of research. Most of the approaches to promote the desired endo outcome of intramolecular epoxide openings use directing groups covalently attached to the epoxides. These directing groups either stabilize (relative to an H atom) the desired transition states, enabling regioselective nucleophilic attack, or make the undesired cyclization route less energetically favorable by changing the electronic properties of the epoxide. Currently available methods for endo cyclization of epoxides rely on alkenyl [20, 23–26], alkynyl [27–29], alkyl [19, 30, 31], and silyl [18, 32, 33] substituents that stabilize partial positive charge within the desired, fused transition state in the Lewis or Brønsted acid-catalyzed reactions (Scheme 15.1a). The directing groups that promote endo cyclization via destabilization of the undesired spiro transition state include sulfones [21, 34, 35], as well as methoxymethyl substituents in combination with a lanthanide Lewis acid [22, 36–38] (Scheme 15.1b). Catalytic antibodies [39–42] and transition-metal complexes [43, 44] can also be particularly effective in promoting endo cyclization by lowering the energies of fused transition states in certain cases. 15.3 Polycyclic Polyethers: Structure and Biosynthesis 15.3.1 Polyether Ionophores
Polyether ionophores are lipophilic carboxylic acids that contain multiple fiveand six-membered cyclic ethers organized either as spiroketals or as linked cyclic
539
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
540
O SiMe3 OH 7 Me
Me3Si
80%
H OTBS
O 8
OTBDPS
O
TIPSO
Me Me
CH3NO2
86% endo/exo = 3 : 1 CSA
O HO 11
O
Me Me
CHCl3
80%
H
O
H
H
O
12
OTBDPS O
14 OMe La(OTf)3
O Me
HO
H
O
H
OMe
10
75% endo/exo = 82 : 18
p -TsOH· H2O
13
Me Me
CH2Cl2 −40 to 25 °C
O SO2Ph
H
OH 2, 6-lutidine
9
(a)
H
CH2Cl2
TIPSOTf
O Me
HO
BF3·Et2O
OH 15
CH3OH, CH2Cl2
96% endo/exo = 91 : 9
HO Me H O 16
(b)
Scheme 15.1 Endo cyclizations via fused transition states: (a) stabilized by directing groups [18–20]; (b) enabled by deactivation of exo pathway [21, 22]. TIPS = triisopropylsilyl, Tf = trifluorosulfonyl, CSA = camphorsulfonic acid, Ts = toluenesulfonyl, TBS = tert-butyldimethylsilyl, and TBDPS = tert-butyldiphenylsilyl.
ethers (Figure 15.2). The first members of this family, X–206, nigericin, and lasalocid A, were isolated in 1951, but due to their toxicity did not initially draw much attention [45, 46]. It was not until 1967 – when a crystal structure of monensin A (17, Figure 15.2) was disclosed [47] and the cation binding abilities of these molecules were first examined [48] – that this family of natural products was thrust back into the spotlight. Subsequent discoveries of their ability to control coccidiosis [49], a devastating poultry disease, and their action as growth promoters in ruminant animals [50], both of which capitalize upon the antibiotic activity of these structures, inspired several research groups to pursue the isolation of novel members of this family, study their biosynthesis, and put effort into their total synthesis. The biological function of polyether ionophores is directly related to their ability to selectively bind metal cations via coordination with multiple oxygen atoms and, due to their lipophilic nature, transport them through biological membranes [48]. By doing so, polyether ionophores disturb the delicate dynamic equilibria of cations across the cell membrane and thus disrupt regular cell function [51], resulting in diverse effects, including antibiotic, antimalarial, anti-obesity, and insecticide activity. Since the isolation of the first polyether ionophores in early 1950s, well over 100 members of this family have been isolated and characterized [52–54]. Although most members of this family are produced by the Streptomyces genus, polyether ionophores have also been isolated from other actinomycetes [4]. A large body of experimental data on their biosynthesis [55–60] and earlier speculation by Westley [55] led the groups of Cane, Celmer, and Westley to propose a unified stereochemical model of polyether antibiotic structure and biogenesis in 1983 [61]. According to the Cane–Celmer–Westley hypothesis, in the biosynthesis of monensin A an all-(E) polyene precursor 25, produced in classic type I polyketide
O
R
CO2H
R
HO
H O
H O H Me
O
Me
Me
Me
Me
OH
Me
Me
O
Me H
HO2C
OMe
OMe
Me Me
Me
Me
O Me H 22, lonomycin A O H H O HO Me Me
OMe
OMe Me
Me
21, cationomycin
O O H O OH H Me Me Me H
Me
HO
Me
O Me HO
Me O H H
MeO
O Me O H Me Me HO O OH HO2C O
HO Me
H
Structures of representative polyether ionophores.
20, isolasalocid, R=
19, lasalocid A, R=
OH
OH O
H
Me
17, monensin A, R = Me 18, monensin B, R = H
Me Me
H
Me
OH Me
O H O MeH O H H OMe O HO CO2H
Figure 15.2
Me
Me
Me
Me
HO
HO
Me O O
H Me
OMe
HO2C
Me
O Me H O 24, etheromycin H H O HO Me
Me
OMe
Me
23, ionomycin
Me Me Me
OMe
OH O Me Me
OH
O O H MeH Me OH OH
O O H O OH H Me Me Me H
Me HO2C
Me
Me
Me
15.3 Polycyclic Polyethers: Structure and Biosynthesis 541
542
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
synthase fashion from five acetates, seven propionates, and one butyrate unit, is oxidized to the corresponding polyepoxide 26 (Figure 15.3a). Nucleophilic addition of the C5 hydroxyl in 26 to the C9 ketone forms a hemiketal that triggers a cascade of all-exo epoxide-opening events leading to the formation of monensin A. Cane, Celmer, and Westley further extended this proposal to the biosynthesis of all polyether ionophores known at the time and described the requisite polyene precursors and pathways to each of these natural products [61]. Failure of the producing organism to incorporate synthetic all-(E) premonensin triene 25 into the biosynthetic pathway and convert it into monensin A encouraged alternative biosynthetic proposals [62, 63]. Townsend suggested that monensin A may be produced from an all-(Z) isomer of 25 through a series of oxidative cyclizations proceeding via a [2+2] mechanism involving an Fe-containing monooxygenase (Figure 15.3b). Related proposals were considered by McDonald [64, 65] and Leadlay [66]. Upon sequencing the monensin biosynthetic gene cluster [66, 67], Leadlay found that deletion of monCI from the producing organism caused accumulation of all-(E) premonensin triene 25. This suggested that a single oxidase enzyme, the product of monCI, is involved in the production of triepoxide 26 [68]. Disruption of monBI and monBII genes led to the production of partially cyclized, chemically competent biosynthetic intermediates [69]. The Oikawa group provided direct evidence for the involvement of an enzyme-catalyzed cascade of epoxide-opening reactions in the biosynthesis of polyether ionophore lasalocid A (19, Figure 15.2) [70]. Significant homology of monBI and monBII with lsd19 from the lasalocid A biosynthetic genes identified the Lsd19 as the putative epoxide hydrolase. Lsd19 was obtained in nearly pure form by cloning las19 and expressing it in Escherichia coli. This enzyme was then utilized in the efficient transformation of synthetic prelasalocid diepoxide into lasalocid A in vitro [70, 71]. Further in vivo studies by Leadlay and coworkers demonstrated that the presence of Lsd19 changes the stereochemical course of polyether ring formation, channeling the polyepoxide intermediate to lasalocid A as the major product [72]. When Lsd19 is not present, as in lsd19 mutant, the formation of the second ring proceeds exclusively by the kinetically favored pathway to form isolasalocid (20, Figure 15.2), thus demonstrating that Lsd19 is responsible for the final biosynthetic cyclization to form lasalocid A. 15.3.2 Polyethers Derived from Squalene
Polycyclic polyethers derived from squalene have been isolated from diverse sources, including marine sponges, red algae, and tropical plants [2]. Some of these natural products, often referred to as oxasqualenoids, resemble polyether ionophores due to the regular occurrence of 2,5-linked oligotetrahydrofuran segments, as in glabrescol (33, Figure 15.4). Other oxasqualenoids, typically isolated from marine sources, such as red algae of the Laurencia and Chrondria genera, resemble ladder polyether structures and feature multiple fused rings such as those found in armatol A (38). The third group of oxasqualenoids, exemplified by enshuol (36),
O
HO
Fe O 27 Ln
Me
Me
Me
HO
OH
7
OH Me
Me
1
O
Fe O Ln 28
17, monensin A
Me
OH
H
Me
O
OH Me
O O MeH O H O H H OMe O HO O
Me
O
H
O H O FeLn 29
cyclization
HO
[O]
Me
5
H
Me
25
Me O
Me
O H O FeLn 30 O
26
Me
O Me
epoxidation
Me
9 O Me O
OH
OH
Me
HO
Me
Me (Me)O O
HO
O
Me HO
HO
O
H
Me
31
H
Me
Me
O H O
H
(OH)
O
Me O
Me
O
Figure 15.3 (a) Cane–Celmer–Westley hypothesis: a model of monensin A biogenesis [61]; (b) Townsend–McDonald hypothesis: biosynthesis via metal-mediated oxidative cyclization [62, 64].
(b)
(a)
5
Me
O
polyketide synthase
FeLn
15.3 Polycyclic Polyethers: Structure and Biosynthesis 543
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
544 Me
Me
Me
Me
Me
OH Me H
Me H
O Me HO
H O Me
O
Me OH
H Me O
H
H H 32, teurilene
Me Me
O
Me H
O
H
Br
Me O OH Me H HO Me
Me H
Me
Me HO
Me
H
AcO
O
O Me H
Me Me
Me OH Me
Br
H O
O
O Me H
HO
MeO H
Me OH Me
O O O O Me Me H H Me H Me H 36, enshuol Me Me
O
HO H Me H
O
H O
OH HO Me 38, armatol A Me H
Me
Me
HO Me Me
H O
H
Me
Me OH Me
Me
Br
39, dioxepandehydrothyrsiferol
Figure 15.4
Me OH
H O
H
H 37, predehydrovenustatriol acetate
Me H O
H O HO HO
Br
Me Me Me
Br
Me Me 34, intricatetraol
35, dehydrovenustatriol
Me H O HO
O
H Me O
33, glabrescol
Me H O
Me Cl Me
Me Cl Me
Br
O H O Me
Me
OH Me
Me Me
H OH
O Me 40, abudinol B H Me
Structures of representative oxasqualenoids.
incorporates both types of polycyclic structures. In addition to these highly oxygenated triterpenes, related squalene-derived polycycles such as abudinol B (40) have been isolated from marine sponges. These molecules typically feature two separate cyclic systems and contain up to two cyclic ethers that are fused to other carbocycles and not to each other [2]. That these oxasqualenoids could be efficiently derived from squalene polyepoxide precursors was recognized soon after similar proposals were put forward for the polyether ionophores and ladder polyethers [2, 73, 74]. Reports of isolation of new oxasqualenoids are often accompanied by a biosynthetic proposal for the specific carbon skeleton [75–78]. The common ground for all proposals is the involvement of epoxide-opening cascades [2, 74, 79]. Biosynthetic studies on these molecules are, unfortunately, scarce. Their triterpene origin is evident due to the absence of skeletal rearrangements, but the proposed oxidation and cyclization steps remain speculative. The important distinction between biosynthetic epoxide-opening cascades leading to polyether ionophores and those that are proposed for oxasqualenoids is the regioselectivity of epoxide-opening events. While polyether ionophores, with few exceptions, arise via all-exo epoxide-opening cascades, cascades that are proposed in the biosynthesis of oxasqualenoids have to incorporate both endo and exo epoxide-opening reactions depending on the structure of the oxasqualenoid natural product.
15.3 Polycyclic Polyethers: Structure and Biosynthesis
The presence of halogens in oxasqualenoid molecules (e.g., bromine at position 3 of the oxasqualenoids isolated from red algae, such as dehydrovenustatriol or armatol A) is an interesting structural feature that has itself drawn biosynthetic speculation. Formation of a bromonium species by the action of an electrophile could initiate the epoxide-opening cascade reaction, producing the complete tricycle of 39 in a single step (electrophilic initiation, Figure 15.5) [80]. However, isolation of dehydrovenustatriol and predehydrovenustatriol acetate (35 and 37, Figure 15.4) from the same producing organism suggests that at least two discrete steps are involved in the cyclization of the polyepoxy precursor. Predehydrovenustatriol acetate lacks the halogenated cyclic ether of dehydrovenustatriol in its otherwise fully cyclized structure, suggesting that it may be its direct precursor. If this is the case, dioxepandehydrothyrsiferol could be derived from the intermediate 42 produced in the preceding cascade (acid/base initiation, Figure 15.5). 15.3.3 Ladder Polyethers
The group of ladder polyether natural products consists of molecules featuring anywhere from 4 to 32 five- to nine-membered cyclic ethers, fused to each other in a trans-syn-trans arrangement. This creates a repeating C–C–O sequence that stretches throughout the polycyclic core of these molecules (Figure 15.6). The first isolated member of this family, brevetoxin B (44), was reported by Nakanishi and Clardy in 1981 [81] and was followed by numerous others, including maitotoxin [82–84], the largest nonpolymeric molecule isolated from natural sources to date. The minimal availability combined with the unprecedented size of ladder polyethers have inspired herculean endeavors in the isolation and structural characterization of these compounds and have pushed the limits of analytical methods, including chromatography, mass spectrometry, NMR, and X-ray diffraction [85]. The structural challenges associated with synthesizing these molecules have also stimulated development of many novel synthetic methodologies [86–89]. Ladder polyethers are notorious for their association with harmful algal blooms commonly referred to as red tides [90, 91]. A rapid increase in the concentration of dinoflagellate algae, for example, the brevetoxin-producing Karenia brevis, leads to the increased production of red tide toxins, some of which are members of ladder polyether family. The effects of red tide are devastating killings of fish and marine mammals. However, some marine species not affected by red tides accumulate and, occasionally, further elaborate the toxins [92, 93], transferring them up the food chain, resulting in human poisoning by ingestion of shellfish exposed to a red tide [94]. Despite their uniform structure, ladder polyethers exhibit diverse biological activities ranging from extreme toxicity [95–97] to anticancer [98–100] and antifungal [101, 102] properties. More recently, a member of this family, brevenal (51, Figure 15.6), has been shown to protect fish from the neurotoxic effects of brevetoxins [103, 104] and has been identified as a potential therapeutic for cystic fibrosis
545
Me O
Figure 15.5
Me Br
O
H O
O
]
Me O
Me O O
Me Me Me
Me OH Me
H [Br
]
41, pentaepoxysqualene
O Me
OH O Me H 39, dioxepandehydrothyrsiferol
Me H O
[Br
Me
Me
O
Me
O
H HO
Me
Me
HA, B
Me O
42
Me H O
A H
Me
O Me H
acid/ base initiation
Proposed epoxide-opening cascades in the biosynthesis of dioxepandehydrothyrsiferol.
Br
Me Me
Me
Me Me
electrophilic initiation
Me
O
B
H O
Me
CH2
546
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
15.3 Polycyclic Polyethers: Structure and Biosynthesis
H
OHC
HO H
H O
O
O Me H
H
H
H
O
O
H
H
H
H H O
H
O
O H H
Me
Me H H O
Me H O
Me substituents at ring junctions
H H O
HO H
HO2C
O
H
H
O Me Me
OH H
O H Me OH HO
OHC
Me Me O
H
H
O
H
H
O
H O H
HO H O H
OHC Me
Me
H
O
OH
Me H O
H
H
O
O
H
H
Me O H
H
H
O
Me
H OH
O
H
O Me Me H 45, yessotoxin
syn Me HO
OH Me
O
H
Me O
H
H H O
H OH
O O H H H OHH
H Me
H
H O H Me O H H
O H H
O H H
H O H Me
47, ciguatoxin CTX3C
O
H OH
HO Me H O H
O
Me H OH
O
Me
OH Me
Me
O H H
Me
H
H
49, brevisamide
OH
Me O Me H
OH
Me
OHC OH H H O
OH H O O
NHAc
Me H O
H
O Me Me
H H O
H H O
H
H H O Me O H
OH
H
H
H O Me OH H
H O
trans -syn -trans geometry
Me
HO Me H O H
OH
OH
50, brevisin
Figure 15.6
H
O
NaO3SO
HO Me
H
H O H
HO
O Me H
O
H O H
NaO3SO
repeating C-C-O motif
H O 48, gambieric acid A Me
H
Me Me O
H O H
trans trans
Me
Me
H O H
H OH
44, brevetoxin B
H O H HO HO H
H
O
OH
46, gambierol Me
OH Me O H
H
43, gymnocin B
Me
H H O
O
547
Me OHC Me
Me
H H O H H O
Me
H
O
Me
Me OH
OH H
O Me H 51, brevenal OH
Structures of representative members of the ladder polyether family.
[105, 106]. While their mode of action is not well understood on the molecular level, it is known that brevetoxins and ciguatoxins bind and disrupt voltage-sensitive sodium channels [107–111], gambierol blocks voltage-gated potassium channels [112], and maitotoxin causes an influx of calcium ions into cells that in turn causes uncontrolled secretion of neurotransmitters and severe muscle contractions [113–116]. Binding of yessotoxin (45) to the transmembrane α-helix of glycophorin A causes the dissociation of oligomeric protein [117]. Soon after the structure of brevetoxin B was reported, Nakanishi [118] and Shimizu [119] hypothesized that the structural and stereochemical similarities among ladder polyethers are a direct consequence of their biosynthetic origin. Such similarity was proposed to arise through the transformation of a polyepoxide into a ladder polyether via a series or cascade of epoxide-opening events (Figure 15.7a). The oxygen and two carbon atoms of each epoxide constitute the C–C–O backbone, and, with the proviso that all of the ring openings proceed with
548
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
inversion of configuration at each epoxide derived from an (E)-alkene, the trans-syn topography is explained by this mechanism. All alkenes in a hypothetical polyene precursor would require identical stereoselectivity of epoxidation to produce either an all-(S,S) or all-(R,R) polyepoxide, suggesting that a single promiscuous oxidase could be sufficient [120]. Despite its intellectual appeal, the hypothesis relies upon a ring-opening process generally regarded to be disfavored. As discussed earlier, according to Baldwin’s rules [6], epoxide-opening reactions of this type typically favor the smaller heterocycle, for example, THF (tetrahydrofuran) over THP (tetrahydropyran). In the case of the proposed precursor to brevetoxin B, the biosynthetic cascade would have to overcome ten consecutive disfavored epoxide openings. In an effort to shed some light on the validity of Nakanishi’s hypothesis, labeling studies have been reported for brevetoxin A, brevetoxin B, and yessotoxin [119, 122–124]. These studies corroborated the polyketide origin of ladder polyethers, which is also supported by genetic studies [125–128], but did not illuminate any subsequent epoxidation or cyclization steps. Some remote evidence in support of this hypothesis can be taken from biosynthetic studies on a related natural product, okadaic acid [129, 130], and isolation of 27,28-epoxybrevetoxin B from Karenia brevis [131]. The intriguing new structures of brevisamide [132] and brevisin [133] (49 and 50, respectively, Figure 15.6) have recently been isolated from K. brevis. These structures differ from most known polyethers and may help further the understanding of biosynthetic pathways involved in production of ladder polyethers. Wright and coworkers have demonstrated that the proposed epoxide intermediates in biosynthesis of all polyethers isolated from Karenia brevis, including the aberrant polyethers brevisamide (49) and brevisin (50), feature identical (S,S) configuration [134]. This may suggest that a single promiscuous epoxidase can be involved in the biosynthesis of not only a single ladder polyether natural product but in the biosynthesis of several related natural products within the same organism. These authors also propose that the ring-forming cascade from a polyepoxide precursor flows in the opposite direction of nascent polyketide chain biosynthesis [132, 134]. Nakanishi’s hypothesis remains speculative due to the lack of strong experimental support in its favor. However, the stereochemical uniformity inferred from the polyene to polyepoxide to ladder polyether pathway has served as the basis for structural reassignment of brevenal [135, 136] and the speculative structural reassignment of the largest known natural product, maitotoxin [120, 137, 138]. According to a modification of Nakanishi’s hypothesis by Spencer, biosynthesis of ladder polyethers might also proceed in an iterative fashion through the repeated action of a monooxygenase and an epoxide hydrolase with broad specificities [120]. In this scenario, rings of a ladder polyether molecule would be formed sequentially, each being formed immediately after the epoxidation of the appropriate (E)-alkene in the biosynthetic precursor, thus avoiding the polyepoxide intermediate proposed by Nakanishi. If Wright’s proposal for the direction of the polyketide chain extension is correct, such oxidation and cyclization steps would occur after the
Me epoxidation
Me HO Me
O
H
Me H
O
H
O
Me
H
H H O O H H Me H
O Me
H Me Me H O O
O H H
O
H+
44, brevetoxin B
O H H
O proposed polyepoxide intermediate, 53 Me Me O Me Me O O O O O O O Me HO OH Me epoxide-opening Me cascade H O
OH
Me
Me Me
Me
CHO
CHO
CHO
(b)
H
H
H
H
H
H
O
57
H
H
H
H
54
H
O
O
H
O
R'
O
58
O
H
H
R
O
OH RO
O
O
H+ O R
R'
R'
H
H
H
H
O
56
O
O
O
R
OH
R'
R'
O
R
HO H+
H
55
HO
H H
Figure 15.7 (a) Nakanishi’s hypothesis: a model of brevetoxin B biosynthesis [118]; (b) Giner’s proposal for biosynthesis of ladder polyethers via an epoxy ester pathway [121].
(a)
O
O
O
Me
proposed polyene precursor, 52
HO
15.3 Polycyclic Polyethers: Structure and Biosynthesis 549
550
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
synthesis of polyene precursor, forming the polyether molecule in iterative fashion, one cyclic-ether at a time. Giner has suggested that ladder polyethers may be derived from an all-(Z) polyene precursor [121, 139]. Giner hypothesized that an epoxy ester intermediate may undergo cyclization with the carbonyl group of the ester as nucleophile, leading to the formation of an orthoester intermediate 56 (Figure 15.7b). Upon collapse of the orthoester, attack of the alcohol nucleophile on what used to be the second electrophilic site of the starting cis epoxide then produces the ring of a ladder polyether and regenerates an ester for the next ring-closing reaction. The Townsend–McDonald hypothesis (Figure 15.3b) can also be extended to a proposal for biogenesis of ladder polyether via a similar all-(Z) polyene precursor. As yet, these hypotheses have not been tested experimentally.
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers
The first epoxide-opening cascades were disclosed in the early 1950s [140]. These early reports typically involved the rearrangement of 1,5-diepoxides that, under appropriate conditions, react with an external nucleophile to undergo a cascade of epoxide openings. Depending on the reaction conditions, these cascades may involve direct epoxide opening or formation of epoxonium ion intermediates, in either case producing tetrahydrofuran products in agreement with Baldwin’s rules [6]. Other early reports on epoxide-opening cascades focused on rearrangements of topologically interesting polyepoxides [141–146], as well as the transannular epoxide-opening cascades of conformationally flexible polyepoxide substrates [147–150]. 15.4.1 Epoxide-Opening Cascades in the Synthesis of Polyether Ionophores
The Cane–Celmer–Westley proposal [61] for the biosynthesis of polyether ionophores via a sequential epoxide-opening cascade quickly sparked great interest within the synthetic community, as emulation of such a biosynthetic pathway could in theory provide a rapid, straightforward approach to several natural products in this family. Especially encouraging was the agreement of the proposed cascade reactions with empirical guidelines [6] for regioselectivity in epoxide-opening reactions. For example, in their efforts toward aurodox, the Nicolaou group utilized a common approach to construct the THF backbone via epoxide-opening cascade of 1,5-diepoxide 59 and extended this methodology to triepoxide 61. Cyclization of 61 to 62 constitutes the first epoxide-opening cascade that affords the 2,5-linked bis-tetrahydrofuran motif common for polyether ionophores (Scheme 15.2a) [151]. While working on the synthesis of uvaricin, the Hoye group reported studies on triepoxides 63–65 [152, 153]. Their strategy involved release of the alcohol nucleophile via ester hydrolysis followed by a base-promoted Payne rearrangement
O
65
O
64
O
63
AcO O
AcO O
AcO O
83%
2. Ac2O, Py
1. aq. NaOH
OAc H
OAc H
Me
O
H
OAc H AcO
AcO
AcO
60
MeO2C
90%, two steps
2. TBSCI, imid. DMF
1. KCH2SOCH3, PhMe/DMSO
67
H H
66
H H
O
O
AcO H
AcO H
68
O Me
O
OAc
OAc
OAc
61
MeO2C
OTBS
Me
O
H
AcO H H OH O
O
O
O O >90%
conditions
Me
MeO2C similar
HO Me H
71
O
H
70
Me
O
H H
O OH
O
HO Me H
72
O
1, 4-dioxane
NaOH, 18OH2
Me
Me
O
H
OH H18O
OH Me
75%
O O OH 69 HO
Me
H18O (c)
Me
O
O H
H
Me
OH
Me OH
O HO
62
O
Scheme 15.2 (a) Construction of the central backbone of aurodox and related bis-tetrahydrofuran 62 [151]; (b) cascades cyclization of disubstituted 1,5,9-triepoxides reported by Hoye [152, 153]; (c) support for the Payne rearrangement mechanism [154]. imid. = imidazole; Py = pyridine.
(b)
OAc O
OAc O
OAc O
Me
59 Me
(a)
O
O
MeO2C O
O
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers 551
552
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
to unveil a new secondary alcohol nucleophile that triggers a cascade of two consecutive epoxide-opening reactions to afford 2,5-linked bis-tetrahydrofuran products (66–68, respectively, Scheme 15.2b). To distinguish between the pathway that operates via Payne rearrangement and other possible pathways involving epoxide opening by water, Hoye and Jenkins conducted experiments on a diastereomeric mixture of diepoxides 69 and 70 in 18 O labeled water (Scheme 15.2c) [154]. In these experiments 18 O was incorporated only in the primary alcohols of 71 and 72, which is consistent with the proposed Payne rearrangement pathway. Although the desired bis-tetrahydrofurans were formed via what appeared to be stereospecific epoxide opening, enantioenriched starting materials were transformed into racemic products due to initiation at both ends of the triepoxide substrates. This problem was addressed with application of appropriate ester protecting groups, which allowed desymmetrization of related diepoxides via unidirectional cascades enabled by the two competing pathways of different rate: hydrolysis of the ester protecting group and epoxide-opening cascade [155]. The first emulations of the alkene epoxidation/epoxide-opening cascade sequence from the proposed biosynthetic pathway to polyether ionophores came from the Schreiber [156] and Still [157] laboratories. Taking advantage of the powerful stereocontrol effects of allylic chiral centers on epoxidations of macrocyclic alkenes by peroxyacids described by Vedejs [158], Schreiber synthesized two diastereomeric diepoxides derived from a cyclic diene 73 (Scheme 15.3a). Diepoxides 74 and 75 were then subjected to a one-pot sequence of base-promoted ester hydrolysis followed by an acid-induced epoxide-opening cascade to form, respectively, 2,5-linked bis-tetrahydrofuran 76, corresponding to C9–18 fragment of monensin B, and its diastereomer 77 after acetonide formation. [156]. In a concurrent report, the Still group described the preparation of a tricycle closely related to the C9–C23 fragment of monensin B [157]. Still used the resident chirality at C18 and C22 of 78 (carbon numbering as in premonensin B) to achieve good stereocontrol in the epoxidation reaction (Scheme 15.3b). Triepoxide 79, isolated in 59% yield (74% corrected for purity of 78), was then taken through a one-pot ester hydrolysis and acid-catalyzed epoxide-opening cascade sequence to afford 80, a diastereomer of the C9–C23 fragment of monensin B. Remarkable yields in these cascades, observed irrespective of the stereochemistry of the epoxides, are likely ensured by directing effects of the appropriately positioned Me groups at each epoxide in polyepoxide precursors. Shortly after the pioneering work of Still and Schreiber, Paterson reported initial studies on epoxide-opening cascades that afford fragments of polyether ionophores [159]. To enable more flexibility in choice of the cascade substrates, the Paterson group relied on Sharpless asymmetric epoxidation to control the stereoselectivity of epoxidation in an acyclic substrate. Under acidic conditions, diepoxyesters 81 and 83 were converted into the corresponding bis-tetrahydrofurans 82 and 84 in good yields (Scheme 15.4a). Similar to Schreiber and Still, Paterson studied trisubstituted epoxides with methyl substituents in positions where their electronic effects promote the desired outcome of the reaction. However, the stereochemical outcome of these cascades may be compromised when an electronic preference for 6-endo cyclization exists. Jaud and coworkers observed inversion of stereochemistry
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers
Me
O
Me Me
O
9.5 : 1dr 74 major
O
Me
73
O
5
4
O
Me O Me
O O
22
m-CPBA
Me Me
NaHCO3 CH2Cl2
18
4 : 1 dr Me
(b)
Me
O
2. Me2CO, H+
O
90-95%
Me Me
O O Me H Me H 76
O
O O Me H Me H 77 Me
i. NaOH, CH3OH/H2O
O
ii. AcOH O
O
O
O
Me
O
59%
78
O
1. KOH, HCl
O Me Me
Me
+
90-95%
O 75
(a)
Me
2. Me2CO, H
O
Me
1. KOH, HCl
O 74
m-CPBA
Me
O
5
4
Me O Me
O
94%
Me H
79
O
Me H
553
Me Me
Me O
OH H
OH
80
Scheme 15.3 Epoxide-opening cascades for preparation of 2,5-linked tetrahydrofurans reported by (a) Schreiber [156] and (b) Still [157]. m-CPBA = meta-chloroperbenzoic acid.
at C5 in acid-catalyzed cyclizations of 85 [160]. It was proposed that initial cyclization via a 6-endo pathway is operative, to produce an orthoester intermediate with inverted configuration at C5 (Scheme 15.4b). Elimination of 2-methylpropene results in the formation of a final five-membered lactone and liberates a tertiary alcohol, which is the nucleophile in the following epoxide opening. Epoxide-opening cascades described so far have typically relied on acid or base catalysis. In these regimes, a cascade can be initiated either via activation of an alcohol nucleophile by deprotonation or, alternatively, via epoxide activation with Brønsted or Lewis acids. While nucleophile activation allows for good control over the direction of the cascade, it is limited to polyepoxide substrates with protic nucleophiles such as alcohols. Activation of the epoxide with acid, on the other hand, is typically unselective. Under these conditions any or all of the epoxides in Me O CH2Cl2
Me Ot-Bu
(a)
81 Me O
(b)
O
76%
HO
O
MeH
O
O
O Me OH
Me O
Me O
CSA
OH
CH2Cl2
83
OO OH t-BuO
t-BuO 86
HO
OH O 87
85, R= TBDPS
Scheme 15.4 (a) Epoxide-opening cascades for preparation of bis-tetrahydrofurans by Paterson [159]; (b) stereochemical outcome of cascades with preference for 6-endo cyclization [160].
88
O
O
70%
Me H
O Me OH
84
OH Me H 5 O
Me 5
O 5
CSA CH2Cl2
Me Ot-Bu
82
5
RO t-BuO2C
HO
CSA
O
O
Me O
HO
OH
OH O
5
O
O H Me H 89, 72%
OH Me OTBDPS
554
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
the polyepoxide substrate can be activated and the cascade may proceed in both directions, with varying points of initiation. This problem is increasingly pronounced as more epoxides are added to the polyepoxide chain, limiting acidcatalyzed cascades to di- or triepoxide substrates. Several research groups have offered alternative means of initiation in epoxideopening cascades that address some of these issues in activation using acid and/or base. Perhaps inspired by the cleavage of the thioester linkage between the polyepoxide substrate and the acyl carrying protein in the proposed biosynthesis of polyether ionophores, several cascades have been initiated by the release of a carboxylic acid nucleophile from an ester through the action of esterase enzymes under mild conditions. Robinson and coworkers subjected esters 90 and 91 to pig liver esterase in slightly basic aqueous phosphate buffers (Scheme 15.5a) [161]. Efficient cascades afforded 82 and 92, respectively, in good yields after prolonged exposure. These cascades appeared to proceed in a stepwise fashion, allowing detection of partially cyclized intermediates consistent with the hypothesis that the anionic carboxylate of hydrolyzed 90 initiates the cascade and is a stronger nucleophile than is a secondary alcohol in the second step. The concept of the selective generation of a reactive epoxonium intermediate on one side of a polyepoxide substrate was introduced by the Murai group in their work towards the synthesis of ladder-type polyethers (see the discussion in Section 15.4.4.2.) [164]. Introducing the same concept for control of cascade direction to the arena of polyether ionophores, Floreancig and coworkers have demonstrated that mesolytic benzylic carbon–carbon bond cleavage in the benzylic position of the radical cations of homobenzylic ethers, such as 93, 95, 97, and 99, forms oxonium ions. These react with pendent epoxides to form epoxonium ions, which can undergo further cyclization [162, 163]. The Floreancig group has shown that both cis and trans substituted epoxides, 93 and 95, are suitable for these reactions and that stereochemical information is preserved during the cascade (Scheme 15.5b). Several examples, in which mono- and diepoxides with a pendent acetal nucleophile are efficiently transformed into bis-tetrahydrofuran products 94, 96, 98, and 100, have established single-electron oxidation as another effective method for the initiation epoxide-opening cascade cyclizations. Starting with bromomethylpolyepoxides (101, Scheme 15.6), the Marshall group has successfully initiated epoxide-opening cascades through transient formation of allylic alkoxyzinc species (Scheme 15.6) [165–167]. These reactions are reminiscent of Nicolaou’s initial reports but avoid the basic conditions used in early cascades, replacing them with the generally mild metallic zinc in alcoholic solvents. Diastereomeric triepoxyfarnesyl bromides 104–112 all undergo cyclization, demonstrating the utility of this approach (Scheme 15.6). 15.4.2 Applications of Epoxide-Opening Cascades in the Synthesis of Ionophores
Epoxide-opening cascade reactions described so far have been inspired by the Cane–Celmer–Westley biosynthetic proposal for polyether natural products.
93
O
95
OC8H17
O
OC8H17
77%
78%
t-BuPh, DCE
OTHP hn, NMQPF6
83%
t-BuPh, DCE
O Me
1:1 C8H17O
1:1 C8H17O
82
O
O
O
HO
96
H H
94
H H
H O Me
O
O
C8H17HO
C8H17HO
OH
O Bn
O Bn
O
O
hn, NMQPF6
70%
50 mM phos. buffer pH 7.5-8.0, 21 °C
pig liver esterase
O
Me
79%
t-BuPh, DCE
hn, NMQPF6
HO
H
100
OE
OE
OH HO
OH HO H
O
HO
H H 98
H O Me
O
92
O Me
1:1 C8H17O
1:1 t-BuPh, DCE C H O 8 17 64% (82% brsm)
Me
OEt
99
O
97
O OEt
Me O
CO2Me 91
Me O
O
Scheme 15.5 Epoxide-opening cascades initiated by (a) enzymatic ester hydrolysis [161] and (b) single-electron oxidation of homobenzylic ethers [162, 163]. NMQ = N-methylquinolinium, DCE = 1,2-dichloroethane; brsm = based on recovered starting material.
(b)
Bn
Bn
(a)
50 mM phos. buffer pH 7.5-8.0, 21 °C
pig liver esterase
OTHP hn, NMQPF6
Me OH O
CO2Me 90
Me O
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers 555
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
556
X H
Me H
O
O
Me OH
Me 102
Me
Zn, TBAI
Me
O
104, R= TBS 79% 105, R= TBDPS 80%
Me
O Me O Me 109, 45%
EtOH
O Me O
OR Me
Me
O
Me
Me
H O Me 103
O Me O
O
O
OR
Me Br
106, R= TBS 81% 107, R= TBDPS 77% Br
Y
X= Br, Y= H or OR
Me
Br
H
Me HO
Zn, TBAI
101
Me
O
X
O Me O Me
Y
OR
Br
Me
O
EtOH X= OR, Y= Br
O Me O
O
Me
O Me O
108, R= TBDPS 83% Br Me
O Me O
Br
Me
Me 110, 63%
O
RO
111, R= Boc 112, R= TBS
52% 71%
Scheme 15.6 Synthesis of bis-tetrahydrofurans via cascades of the triepoxyfarnesyl bromides reported by Marshall [165, 166]. Boc = tert-butoxycarbonyl, TBA = tetrabutylammonium.
Their development was typically driven by the search for chemical evidence in favor of this biosynthetic pathway. Equally important for the synthetic community, these reactions were developed with specific synthetic targets in mind, and they represent a classic example of the rapid generation of complexity from relatively simple starting materials. Although inspired by nature and developed for the synthesis of natural products, epoxide-opening cascades were used in the preparation of artificial ionophores even before the biosynthetic proposal for polyether ionophores was put forth [168, 169]. Several research groups were successful in extending these reactions to the total synthesis of polyether ionophores or fragments thereof. Notable contributions came from Paterson’s group in their investigations toward etheromycin (24, Figure 15.2) [159]. In their first-generation approach, Paterson and coworkers extended studies on model diepoxides 81 and 83 (Scheme 15.4a) to mixtures of more complex diastereomeric triepoxides 114A and 114B. Triepoxides 114A and 114B, both featuring tert-butyl ester as the trapping nucleophile, underwent acid-promoted cascade reaction to afford the CDE ring system of etheromycin, 115A, and its diastereomer 115B (Scheme 15.7a). A more elaborate polyepoxide cyclization in the second-generation approach to etheromycin also allowed the formation of the BC spiroacetal during the cascade [170]. Exposure of diepoxide 116 to acidic conditions triggered deprotection of the secondary alcohol, which formed the hemiketal. The hemiketal nucleophile initiated the cascade of epoxide openings, resulting in formation of 117 (Scheme 15.7b). Paterson and coworkers also attempted to incorporate the appropriate substitution pattern in the starting diepoxide to install the hydroxyl at C4 of the tetrahydropyran ring in the BC spiroketal. However, when diepoxide 118 was exposed to the acidic promoter, elimination of the secondary alcohol occurred to form a trisubstituted alkene in 119 (Scheme 15.7c). In contrast, when diketone 120 was used, the desired BCD fragment of etheromycin (122, Scheme 15.7d) was produced in a single step [171].
O
O
HO
TBSO
Me
Me O
116
HO
Me
Me O
Me Me
120
Me
Me
Me
OTBS
O
O
O
113
O
Me O
O
O
O
65%
Me2CO
CSA
65%
Me2CO
CSA
93%
CH2Cl2
m-CPBA NaHCO3
Me
O
Me
TBSO
117
O
O Me
O O Me H
HO 114 A,B
t-BuO
O
MeH 121
O
O
O
O Me
OTBS
Me
Me
58%
Me OH
(c)
O
HO Me
CSA CH2Cl2, 0 °C
aq. HF, CH3CN; 47%
Me
O
Me Me
O
O
Me Me
Me O
79%
CH3CN
aq. HF
118
O
O
MeH
HO
Me
Me O
O MeH
B
H
Me
O
119
O Me
Me
OH
O O Me H
OH
O Me Me
O
O D C Me O O O MeH Me O Me 122
Me
Me
4
O
65%
Me2CO
CSA
115 A,B
O
Me
Scheme 15.7 (a) First-generation, (b) second-generation, and (c) and (d) third-generation cascade approaches to etheromycin reported by Paterson [159, 170, 171].
(d)
Me
(b)
Me
(a)
t-BuO
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers 557
558
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
The Evans group elegantly incorporated an epoxide-opening cascade to construct the CD ring system of lonomycin A (22, Figure 15.2), by way of macrocyclic diepoxide 123 (Scheme 15.8) [172]. Upon lactone hydrolysis, a cascade of epoxide openings afforded 125 in a straightforward manner. The E ring of lonomycin A was then constructed in a fashion that mimics the following step of the proposed biosynthetic cascade to produce tricycle 126. A cascade starting from triepoxide 127 would, in principle, be a more direct route to the lonomycin A backbone, but stereocontrolled synthesis of such a substrate would have been considerably more complex. Expanding on their zinc-initiated epoxide-opening cascades of terminal iodomethylepoxides [165, 166], Marshall and colleagues constructed the bis-tetrahydrofuran motif of ionomycin and transformed it into a fully elaborated C17–C32 fragment of ionomycin [167]. 15.4.3 Epoxide-Opening Cascades in the Synthesis of Squalene-Derived Polyethers
As discussed earlier, oxasqualenoids feature both fused cyclic ethers and 2,5-linked oligotetrahydrofurans. This means that epoxide-opening cascades that would produce such diverse structures have to incorporate both endo and exo selective epoxide openings. Despite the development of various methods for control of regioselectivity in intramolecular epoxide-opening reactions, cascades that successfully achieve this goal are scarce. Instead, most epoxide-opening cascades are either all-exo, like those reported in the synthesis of polyether ionophores, or all-endo, like those that are to be discussed in the context of ladder polyether synthesis. Both types have been used in various ways to produce fragments of squalene derived polyethers. In their studies toward the total synthesis of glabrescol [174], the Corey group reported a rapid synthesis of the originally proposed structure of glabrescol. They prepared pentacycle 130 from the corresponding pentaepoxide 129 in a single step under acidic conditions (Scheme 15.9a) only to find that this material had physical and spectral properties different from natural glabrescol. The authors also prepared three other diastereomers of pentaepoxide 129, all of which cyclized to corresponding CS -symmetric pentacyclic polyethers under the same conditions described for 129 (e.g., 131 to 132). However, none of the produced polyethers were identical to natural glabrescol. The correct structure of glabrescol, which is in fact a C2 -symmetric molecule, was disclosed in a subsequent report by Morimoto [173]. The Corey group also investigated the possibility that glabrescol is a C2 -symmetric molecule [175]. Their synthesis of the revised structure of glabrescol relies on a bidirectional double cyclization of a tetraol tetraepoxide 133 (Scheme 15.9b). The choice of acidic reaction conditions was crucial in this case to ensure that cyclization to form the AB and A B rings of glabrescol via epoxide opening at more substituted positions is faster than the rate of cyclization to form the C ring (via exo-opening at a less substituted position). Upon treatment of 133 with camphorsulfonic acid (CSA), the bidirectional cascade formation of AB and A B was able to outcompete the undesired unidirectional cascade that would form a diastereomer of the ABCB
Me
O
Me
127 Me
O
Me Me
Me
Me
123
O
O Me
79%
ii. AcOH
OMe O
O Me
OMe O
i. KOH, H2O CH3OH
O Me
124
O O 128
O
O
Me Me
MeO
Me
HO Me
Me Me O
Me Me O
CO2H
Me O
O Me
O Me
OMe O 85%
O
CH2Cl2
O
C O
MeH
Me
D O
O Me
OMe O
OH Me Me E Me OH MeH O O HO MeO 126
Me
81%
O Me OH HO Me 125 Me
MeH
1. MMPP, CH2Cl2, 4 Å mol sieves 2. AcOH, CH2Cl2
4 Å mol. O sieves
Scheme 15.8 Synthesis of the CDE ring system of lonomycin A reported by Evans [172]. MMPP = magnesium monoperoxyphthalate.
O Me
O
Me Me O
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers 559
560
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
tetracycle and afforded 134 in good yield. Tetracyclic intermediate 134 was then rapidly converted into the natural product in two steps (Scheme 15.9b). The Morimoto group used intramolecular epoxide-openings to construct cyclic ethers of numerous oxasqualenoids. For example, a base-promoted epoxide-opening cascade on diepoxide 136 [176, 177] afforded the C ring of aurilol and enshuol (137, Scheme 15.10). After elaboration to the diepoxide 138, a Brønsted acid-catalyzed epoxide-opening cascade afforded the D and E rings of enshuol. Finally, reagent-controlled, silyl triflate-promoted opening of the trisubstituted epoxide 140 with a tertiary alcohol nucleophile via 6-endo cyclization [19] efficiently formed 141 and 142, containing the B ring of enshuol (Scheme 15.10) [178]. In their latest work on oxasqualenoids, the Morimoto group reported a route to omaezakianol, which includes two consecutive epoxide-opening cascades [179]. The first of the two cascades is similar to the cascade cyclization used to construct 147, the THF-containing subunit of C2 -symmetric oxasqualenoid intricatetraol [180]. Diepoxides 143 and 144 rearranged to the functionalized tetrahydrofurans 146 and 147, respectively, under basic conditions. Interestingly, a single bicyclic side product 145 was produced in this reaction, possibly via Payne rearrangement followed by 5-exo opening of the C6–C7 epoxide and 8-endo nucleophilic attack of the resulting tertiary alkoxide at the C7 position onto the terminal epoxide formed in the initial Payne rearrangement. Elaboration of 146 to triepoxide 148 followed by an acid-catalyzed epoxide-opening cascade afforded 149, which was then rapidly elaborated into the natural product (Scheme 15.11a). Each of the THF rings in Morimoto’s synthesis of omaezakianol is formed via an epoxide opening of a trisubstituted epoxide. Interestingly, these cascades proceed in good yield, one under basic and the other under acidic conditions, despite the potentially adverse effects of methyl substituents of the trisubstituted epoxides. By modifying the pentaepoxide substrate 129 used in their synthesis of the originally proposed structure of glabrescol, the Corey group recently completed a short total synthesis of omaezakianol [181]. Pentaepoxidation of the racemic chlorohydrin 151 afforded the cascade substrate 152 that, upon treatment with a Brønsted acid, cyclized to the pentacyclic product 153, which was converted into omaezakianol in a single step (Scheme 15.11b). Elegant work on ent-abudinol B and the related terpenes ent-durgamone and ent-nakorone capitalizing on studies of epoxide-opening cascades directed toward synthesis of ladder polyether structures was reported by McDonald [182, 183]. In their first-generation approach to ent-abudinol B, McDonald and coworkers devised a convergent synthetic scheme featuring a late-stage coupling of fragments derived from ent-durgamone and ent-nakorone (156 and 159, Scheme 15.12a) [182]. In the synthesis of subunit 155, a cascade of epoxide openings on diepoxide 154 was employed. Use of tert-butyldimethylsilyl triflate as a Lewis acid, two endo-selective cyclizations directed by methyl substituents, with an enol-silane as trapping nucleophile, led to formation of bicyclic compound 155 that could be further elaborated to ent-durgamone. An analogous strategy was utilized in the synthesis of the more complex ent-nakorone, but a hybrid cascade of oxacyclizations and carbocyclizations was required. Diepoxide 157, with a terminating propargyl silane nucleophile,
Me
O
Me
Me
OH
Me
Me OH
O
133
Me Me
Me
Me HO
O
OH
O Me 129 Me OH
Me OH
Me Me
O
Me
O
Me
O
44%
CH2Cl2
CSA
31%
PhMe
CSA
Me
O
H
OH 130
134
Me Me OH
Me
Me
O
Me
Me OH
O
Me
Me
Me
O
135
Me Me
Me
O
Me
Me
OMs
OH
O O H H Me Me O O H H
O 131 Me
Me Me
OH
HO
OH
50% yield, brsm at 60% conversion
MsCl, Py DMAP
Me
Me
OH
OH Me
Me Me
O H H Me Me O O H H
O
H
O O H H Me Me O O H H OH
HO Me
Me
Me
65%
HOAc
NaOAc
PhMe
CSA
Me
O
H
C O
H
OH
Me
Me
Me
Me Me
OH
Me 33, glabrescol
OH
O B' B O H H Me Me O A' A O H H
H
132
Me Me
O O H H Me Me O O H H
H
OH
Me
Me
Me
Scheme 15.9 (a) Synthesis of proposed structure of glabrescol (Corey, 2000) [174]; (b) bidirectional cyclization to form glabrescol (Corey, 2000) [175]. Ms = methanesulfonyl.
(b)
(a)
O
O
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers 561
O OH
Me
Me OH Me
88%
1, 4-dioxane
aq. NaOH
OH
Me
O H H HO Me 140
C O
HOH HO Me Me OH 137
SEMO
Total synthesis of enshuol by Morimoto [178].
D E O O H Me H Me H 139
136
Scheme 15.10
H HO Me
C O
Me
O
OSEM
SEMO
CH3NO2
TIPSOTf, 2, 6-lutidine
138
Me
O
B C O O Me MeH H 141, 22% R = H 142, 49% R = TIPS
RO
HO H HO Me Me OH
SEMO
CSA
53%
CH2Cl2
45, enshuol
Me Me
O
562
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
O
Me
OH
Me
7 Me 2
O
Me
Me
152
151
O
Me
Me
O Me HO MeH OH
1
Me
O Me
O
Me
Me
O
aq. LiOH 1,4-dioxane
Me
Me
O
OH
Shi epoxidation
148
144, R , R = PMP, H
143, R1, R2= Me, Me
O
O
aq. NaOH 1,4-dioxane
acetone
CSA
30-40%
PhMe
CSA
O
O
Me
Me Cl Me
Me
Me Me
Me
Me
O
Me O
21% (two steps)
Me
Me
O
Me
7
145, 23%
Me
Me
Me
O
O Me Me
O
PMP O H HO Me
O Me OH
OH
72%, 146, R1, R2= Me, Me 67%, 147, R1, R2= PMP, H
Me Me
O
R2
H
O
H Me
Me H
HO
O
Me H
Me H
O O H Me H Me H 153
O
O
Me OH Me
Me OH Me
Me OH Me
O O H Me H Me H 150, omaezakianol Na, Et2O 60 °C, 4 h 76%
O
O OH O O Me H Me H HO MeH Me H 149
Me
OH
R1
Scheme 15.11 Synthesis of omaezakianol via epoxide-opening cascades reported by (a) Morimoto’s group [179] and (b) Corey’s group [181]. PMP = p-methoxyphenyl.
(b)
Cl Me
Me
OH
Me
Me Me (a)
Cl Me
O
Me Me
O
Me
R2
O
R1
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers 563
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
564
O Me
TBSO
O Me
CH2Cl2 60%
TMSOTf DTBMP
Me O Me
H
75%
157
(a)
TBSO
O Me
Me
Me
(b)
H H
H
Me
Me
Me
H
TMSOTf, DTBMP
H OH
O Me H Me ent-40, ent-abudinol B
CH2Cl2 then TBAF 15%
Me H
HO
Me
Me H O Me
Me
O Me H Me
H
TMSOTf, DTBMP
H
O H Me Me O Me H 159, ent-nakorone
H
then TBAF 50%
Me
HO O
O
CH2Cl2
160
Me Me
H
O
HO H H
Me
Me
H
Me
158
Me O Me
ent-40 ent-abudinol B •
Me Me
Me OH Me 156, ent-durgamone H
TMSO
TMS CH2Cl2
Me
Me Me
O H 155
Me
O
OTBS
O
154
O Me
H
Me
TBSOTf DTBMP
Me
161
Me
HO OH Me Me
Me
H
H
Me O Me
Me
O Me
162
Scheme 15.12 Syntheses of ent-abudinol B via hybrid oxa/carbocyclizations: (a) convergent approach [182]; (b) biomimetic approach [183]. DTBMP = 2,6-di-tert-butyl-4-methylpyridine.
underwent efficient TMSOTf (trimethylsilyl trifluoromethanesulfonate)-promoted cyclization to tricyclic allene 158. As in the previous cascade, the regioselectivity of cyclizations was directed by methyl substituents. Further elaboration of fragments 155 and 159 into their corresponding vinyl triflates and subsequent modified Suzuki–Miyaura coupling produced ent-abudinol B. A second-generation approach was based on the proposed biosynthetic pathway to ent-abudinol [2, 73, 74], which involves a hybrid cascade of epoxide openings and carbocyclizations [183]. Similar to the first-generation approach, diepoxide 160 was treated with TMSOTf to produce 161, containing the tricyclic fragment of ent-abudinol (Scheme 15.12b). A two-step elaboration of the cascade product 161 via Wittig methylenation and Shi epoxidation afforded diepoxide 162. Diepoxide 162, carrying a terminal alkene instead of an enol ether trapping nucleophile (as in 154), was subjected to the same conditions as in the transformation of 158 into 159 to produce ent-abudinol B, along with several isomeric products resulting from pathways enabled by the relatively low nucleophilicity of the terminating alkene. Despite the linear nature of this route to ent-abudinol B, structural complexity is generated quickly, making this approach very efficient. In efforts to further improve the synthesis of abudinol B, the McDonald group has prepared (3R,6R,7R,18R,19R,22R)-squalene tetraepoxide, a putative biosynthetic precursor to various oxasqualenoids, but this material failed to cyclize to ent-abudinol B [184].
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers
The Jamison group recently reported an epoxide-opening cascade-based approach to the synthesis of the enantiomer of the oxasqualenoid natural product dioxepandehydrothyrsiferol (Scheme 15.13) [80]. The presence of bromine in dioxepandehydrothyrsiferol presented the opportunity to initiate the epoxide-opening cascade with electrophilic bromine reagents. This approach promised good control over the direction of the cascade. Relying on the well-documented bromoetherification strategy to construct bromooxanes and bromooxepanes [76, 177, 178, 185], Jamison and coworkers first examined diepoxide substrates 163, 164, and 167 featuring various terminating nucleophiles (Scheme 15.13a). When carried out in highly polar, non-nucleophilic solvents such as hexafluoro-iso-propanol, cyclization reactions of 163, 164, and 167 afforded the desired tricyclic products 165, 166, and 168, respectively, in good yields via all-endo epoxide-opening cascades, albeit as 1 : 1 mixtures of diastereomers resulting from non-stereoselective bromonium formation. A polar solvation environment facilitates a cationic cascade, thus maximizing the directing effects of the appropriately positioned methyl substituents, and securing good endo selectivity in each cyclization. The Jamison group extended these findings to the cyclization of triepoxide 169. Treatment of 169 with N-bromosuccinimide (NBS) in hexafluoro-iso-propanol afforded the tetracycle 170, which contains the fully elaborated tricycle of ent-dioxepandehydrothyrsiferol (Scheme 15.13b). 15.4.4 Epoxide-Opening Cascades in the Synthesis of Ladder Polyethers
Epoxide-opening cascades were initially, and nearly exclusively, explored in the context of the synthesis of polyether ionophores and other natural products that could arise from exo opening of epoxides. This is not surprising considering the breadth of data supporting Baldwin’s rules. Successful endo-selective cascades for preparation of ladder polyether-like fragments require circumventing the inherent selectivity for smaller rings in epoxide-opening reactions. 15.4.4.1 Iterative Approaches As discussed previously, most methods for regioselective endo epoxide opening rely on the effects of directing groups directly attached to the epoxide. These directing groups are typically not present in the target ladder polyethers; the fact that they are incorporated in products of such epoxide-opening reactions presents a major challenge for their successful utilization in total synthesis of ladder polyethers because of the need for their removal or extensive synthetic elaboration. If such reactions are extended to cascades of epoxide openings, multiple directing groups would be incorporated at the ring junctions of the final product, thus creating the need for selective elaboration of each of the groups into H or Me groups, the exclusive substituents found at the ring junctions of ladder polyethers. As they are good directors of regioselectivity, methyl groups would appear to be the exception; however, they are typically present at only a few ring junctions in each ladder polyether and are rarely distributed in a uniform substitution pattern.
565
Me O
Me O
X
Me
O
O
Me Me
163, X = O 164 X = CH2
O Me
169
O
O
NBS
Br
Me Me
72%
4Å MS
(CF3)2CHOH
Ot -Bu NBS
4Å MS
(CF3)2CHOH
Ot-Bu
O O Me H
Me O X
Br
Me Me 3
1:1
H O
O
O H Me
Me H O
66%, 165, X = O 73%, 166, X = CH2
H O
170
O
Me O
Me
O
Me O
Br
Me Me
167
O Me
Me
O
Me H O
4Å MS 58%
Br
Me Me
Me
O
H
O
Me OH Me
H
O Me H 168
Me O
OH OH Me ent-39, ent-dioxepandehydrothyrsiferol
H
NBS (CF3)2CHOH
OH
Scheme 15.13 (a) Bromonium-initiated epoxide-opening cascades; (b) synthesis of ent-dioxepandehydrothyrsiferol reported by Jamison [80]. NBS = N-bromosuccinimide.
(b)
Me
(a)
Me
Me
566
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers
Despite the problems associated with the use of directing groups in cascades of epoxide-opening reactions, they have been of tremendous value in iterative approaches to ladder polyether synthesis. Such approaches depend on the type of the directing group used in the epoxide-opening reaction and require efficient removal of this group following each iteration. If all requirements are met, a sequence of endo cyclization, removal of the directing group, and homologation to a new epoxide bearing the appropriate directing group for the next cyclization results in the formation of one cyclic ether per iteration. The Nicolaou research group was the first to explore and report a successful iterative approach to ladder polyethers based on endo-selective epoxide opening (Scheme 15.14a) [24]. The epoxy alcohol 171 bearing an alkenyl directing group underwent Brønsted acid-catalyzed cyclization with excellent endo-selectivity due to the ability of alkenyl substituent to stabilize partial positive charge in the transition state for the desired cyclization. Upon elaboration of the tetrahydropyran 172 to epoxy alcohol 173, another acid-catalyzed opening of alkenyl epoxide afforded diad 174 with excellent efficiency. The Mori group reported a complementary approach to ladder polyethers relying on endo-selective opening of epoxysulfones (Scheme 15.14b) [21]. Exposure of epoxysulfone 175 to Brønsted acid led to 6-endo cyclization and subsequent loss of phenyl sulfonate to yield ketone 176. A sequence involving alkylation of the sulfone-stabilized cis-oxiranyl anion completed the homologation process to 177, which contains an epoxide with the appropriate directing group for the next iteration. Repeating this protocol three times furnished tetracycle 178. Mori and coworkers also developed methods for larger oxygen heterocycles [186] and ring junction substitution patterns (Me and H) [35, 187] present in ladder polyether natural products. Capitalizing on the effects of a silyl group attached directly to an epoxide, studied in detail by Hudrlik [188, 189] and Paquette [190], Jamison developed an iterative approach to the synthesis of trans-fused oligo(tetrahydropyran) fragments (Scheme 15.14c) [18]. In contrast to cyclization of epoxysulfones, in which the sulfone deactivates the undesired site of epoxide opening, silyl groups stabilize positive charge in the transition state leading to 6-endo epoxide opening. The directing group can easily be removed after cyclization by treatment with TBAF (tetrabutylammonium fluoride). The utility of this approach was demonstrated by synthesis of THP triad 182. 15.4.4.2 Epoxide-Opening Cascades Leading to Fused Polyether Systems Early work on epoxide-opening cascades by the Murai group led to the development of methods for the endo-selective lanthanide-promoted opening of methoxymethyl substituted epoxides [22]. Murai and coworkers prepared polyepoxides 183 and 189, which incorporate a methoxymethyl directing group at each epoxide [191]. Under conditions described for substrates containing one epoxide, diepoxide 183 was converted into a THP diad 184 with methoxymethyl groups present at the ring junctions (Scheme 15.15). The side products isolated in this reaction are suggestive of a pathway that proceeds in a stepwise fashion from the primary alcohol, initially
567
H
CH2Cl2 −40 to 25 °C 95%
179
SiMe3
CH2Cl2 80%
BF3·Et2O
H
O
H
H
O
OR
O SiMe3 180
HO
OH
O
H
172
O
O
H
H
176
HO
175
OR p -TsOH·H2O O CHCl3 SO2Ph 80%
H OTBS
171
HO
CSA
O
H
O H 173
H
O
H
CH2Cl2 −40 to 25 °C 100%
CSA
SO2Ph O OR O H H OTBS 177 H HO O O H Me3Si SiMe3 181
HO HO H
H O
O
O
H
H
O
Me
H
O
H
182
H
O H H 178 H H H HO O H
H
O H 174
H
O
H
O
OR
Scheme 15.14 Iterative synthesis of oligo(tetrahydropyran) fragments via 6-endo cyclization of (a) alkenyl epoxides [24], (b) epoxysulfones [21], and (c) epoxysilanes [18]. R = TBPDS.
(c)
(b)
O
(a)
O
568
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
O
OH
OH
MeO
HO
H2O, CH2Cl2
La(OTf)3, La2O3
H2O, CH2Cl2
La(OTf)3, La2O3
Me
H
O
OMe
Me
O H MeO
OMe H
184, 52%
O
HO
H
OMe
Me
H
O
OMe
O Me
OMe
185, 4.5%
O H 190, 9.3%
O
HO
OMe O
OMe
O
H
O
OMe
OH
O
191, 15%
OMe
OMe 186, 0.3%
Me HO O
Murai’s epoxide-opening cascades directed by a methoxymethyl group [191].
189
OMe
183
OMe
O
O
OMe
Scheme 15.15
O Me
OMe
OMe
Me O
M
Me OMe
O
O OH
188
OMe
O
187, 15%
OMe
H HO
OMe
Me O
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers 569
570
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
forming intermediate 186, and then 184 and 185. The authors also reported that the other diastereomer of 183 fails to afford any of the corresponding THP diad. The postulated intermediate 188 does not react further due to strain in the requisite boat-like transition state and steric repulsions between the two methoxymethyl substituents. Murai has also demonstrated that cascades directed by methoxymethyl groups in combination with an appropriate Lewis acid can be extended to larger ladder polyether type fragments such as triad 190, albeit in low yield. McDonald reported the first cascade reactions that produce oxepane and trans-fused bisoxepane motifs via endo-selective epoxide opening [30, 192]. A range of terminating nucleophiles such as ketones, esters, carbonates, and acetals were examined and demonstrated to be a factor in determining regioselectivity in these Lewis acid-promoted polyepoxide cyclizations (Scheme 15.16a). McDonald and coworkers extended these reactions to polyepoxides 202–205 for the synthesis of polyoxepane systems [30, 192]. The efficiency of these reactions tends to drop as the number of epoxides in the polyepoxide precursor increases (Scheme 15.16b). A possible reason for a nonlinear decrease of yield in cascades that involve more than two epoxides may be unselective activation of any or all of the epoxides in the starting materials. If selective activation of the epoxide distal to the terminating nucleophile could be achieved, cascades would presumably proceed in one direction, and higher yields should be observed. With this in mind, the McDonald group prepared substrates 210 and 211 that feature a vinyl and a methyl substituent at the terminal epoxide of the polyepoxide chain (Scheme 15.16c) [193]. Based on Nicolaou’s work on alkenyl epoxides [20, 23, 24], it was expected that the stabilization by the vinyl substituent would not only improve selectivity in epoxide-opening reactions but also lead to selective activation of the alkenyl epoxide over the interior epoxides under finely tuned conditions. Optimization revealed Gd(OTf)3 and Yb(OTf )3 as the best reaction promoters. Indeed, desired oxepane ring-containing products 212 and 213 were produced in higher yield than in the corresponding reactions of substrates 202 and 203, which lack vinyl substituents. The synthetic utility of these impressive cascades is, however, limited by the requirement for an alkyl directing group on each epoxide, resulting in the incorporation of the directing groups at every ring junction of the final cascade product. Were these cascades to be used in the synthesis of natural products, they would have to accommodate polyepoxides without directing groups and allow for various substitution patterns that would install methyl groups only at the desired positions of the final products. The McDonald group has offered two approaches to address these concerns. The first is based on the similar directing effects of silyl groups relative to alkyl substituents and the opportunity to remove them after the cascade. For example, 214 and 215 were converted into the corresponding cyclization product 216, with efficiencies comparable to those of cascades with only methyl-substituted epoxides (Scheme 15.17) [194]. The McDonald group has also investigated cascades that would incorporate disubstituted epoxides with no directing groups present, as in 217, 218, 220, and 221 (Scheme 15.17) [194]. The difference between the electronic properties of disubstituted and trisubstituted epoxides may have worked in favor of the desired
Me
O
(c)
(b)
Me O
O
Me
O
t-BuO
O
196
Me O
O
Me
O
O
Me
Me
210, n=1 211, n=2
Me
O
n
60%
2. Ac2O, Et3N
1. BF3·Et2O CH2Cl2
2. Ac2O, Et3N
1.BF3·Et2O CH2Cl2
202, n=0 203, n=1 204, n= 2 205, n=3
Me O
Me O
O
192, X= CH2 193, X= O
Me
t-BuO
O
O
X
n
O
O
X
H O OAc
Me Me O
Me
n = 1, 2
CH2Cl2
LA
n= 0, 1, 2, 3
CH2Cl2
BF3·Et2O
O
OH
H O
Me Me
O
Me O
Me O
O
Me O O H Me
H
OH
OH O H H Me 207, n=1, 52%
Me O O
H
Me O Me H 206, n = 0, 60%
O
Me O
200
Me O
198
Me O
n=1 LA = Gd(OTf)3, 212, 77%
n=1 LA = BF3·Et2O, 212, 65%
O
O
HO Me O Me 27%, 194, X= CH2 60% 195, X= O Me HO Me O O OAc O O t-BuO O Me 197
Me
O
O
O
Me O O H Me
H Me O
H
O O H H Me 209, n= 2, 27%
OH
OH
Me Me
H
H
n = 2 LA = Yb(OTf)3, 213, 56%
O
O
H Me O
O
201
Me
OH
OAc
OAc
Me
Me
Me
Me Me
O
O
199
H
O O H Me H Me 208, n= 3, 12% Me O
Me O
O
O
O
O HO
O
Me
H Me O
2. Ac2O, Et3N 65%
1. BF3·Et2O CH2Cl2
2. Ac2O, Et3N 65%
1. BF3·Et2O CH2Cl2
Scheme 15.16 Lewis acid catalyzed alkyl-directed cascades of (a) 1,5-diepoxides, (b) CH2 –CH2 interrupted polyepoxides [30, 192], and (c) CH2 –CH2 interrupted polyepoxides with terminal alkenyl epoxides reported by McDonald [193].
(a)
t-BuO
Me
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers 571
O
Me O
O
Me O
O O
H O
O Me
H SiMe3 216, 40% 216, 45% O Me
Me O
220, R=Ot-Bu 221, R= Me2N
O
O OAc RCO2
2. Ac2O, Py
1. BF3·Et2O CH2Cl2
H
Me Me Me
O O
Me O H
O
O
H
H H 222, 20% 222, 25%
H
O
2. Ac2O,Py
Me 1. BF3·Et2O CH2Cl2 O
217, R= Ot - Bu 218, R= Me2N
O
Me O
Me O
OAc
O
Me Me
H
O
O
Me Me O OAc H H H 219, 30% 219, 35%
H
Scheme 15.17 Synthesis of polyoxepane ring systems, via cascades of polyepoxide precursors without alkyl directing groups at internal epoxides, reported by McDonald [194].
R
O
215, R= Me2N
2. Ac2O,Py
Me 1. BF3·Et2O CH2Cl2 O
Me3Si RCO2 214, R= Ot - Bu
Me
572
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers
cascade through preferential activation of the epoxide distal to the terminating nucleophile in a fashion similar to cascades on alkenyl epoxides 210 and 211. Cascades of both triepoxides 217 and 218, and tetraepoxides 220 and 221, under standard Lewis acid activation proceeded to form the desired tricyclic polyether 219 and tetracyclic polyether 222 (Scheme 15.17). It was proposed that once the first epoxonium ion is formed at the distal end the transition states leading to endo and exo opening of the disubstituted epoxonium ion are different in energy, with a higher degree of ring strain associated with the bicyclo[3.1.0] intermediate than for the bicyclo[4.1.0] intermediate formed as the product of endo opening. A directing group was required on the epoxide proximal to the trapping nucleophile, as there is minimal strain associated with either five- or six-membered carbonates formed at the end of the cascade when carbonate or carbamate nucleophiles were used. A directing group was, therefore, necessary to ensure endo regioselectivity in the opening of this last epoxide. In addition to their studies on the use of epoxide-opening cascades in the construction of oxepanes, the McDonald group has, in similar fashion, also explored cascades directed toward synthesis of oligo(tetrahydropyran)s. The effects of the terminating nucleophile on epoxide-opening cascades of 1,4-diepoxides 223–225 were examined first [31]. Depending on the nucleophile, these reactions can proceed with either retention or inversion of configuration at the ring junction (Scheme 15.18a). Stronger nucleophiles at elevated temperatures favored the inversion of stereochemistry in the opening of internal epoxide, thus setting a trans geometry at the ring junction of 226. In contrast, less nucleophilic carbonates favored the production of diastereomeric product 227, corresponding to retention of configuration. These were explained by the mechanism outlined in the Scheme 15.18b, in which the terminating nucleophile intercepts a cationic intermediate at different points in the continuum between the extremes of epoxonium ion 230 and tertiary alkyl carbocation 234. McDonald proposed that cis-fused products arise from fast nucleophilic addition to the tertiary carbocation, whereas trans-fused products are favored with a stronger nucleophile, which intercepts a tight ion pair intermediate structurally related to the epoxonium ion [31]. The McDonald group extended these findings to epoxide-opening cascades of triepoxides 235 and 236, which carry directing groups on each of the epoxides. When activated by a Lewis acid at an appropriate temperature, triepoxide 235 with the carbamate terminating nucleophile was transformed into the ladder polyether-like tricycle 238 in 31% yield. Triepoxide 236 with a carbonate nucleophile, however, failed to afford any of the desired products and, instead, at low temperatures gave 237 (Scheme 15.18c). To summarize, the choice of the terminal nucleophile not only dictates whether cyclization will proceed with retention or inversion but in the case of 1,4,7-triepoxides it also determines the regioselectivity of the cyclization onto the epoxide proximal to the carbonyl nucleophile [31]. A conceptually novel way of promoting epoxide-opening cascades to ladder polyethers was investigated by Murai et al. [164]. They envisioned that activation of a polyepoxy halide with a silver salt would selectively generate an epoxonium ion at one end of the polyepoxide chain. This epoxonium would then serve as an
573
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
574
R
Me O
O
Me O
O Me
(a)
R
H BF3·Et2O
O
CH2Cl2 −40 °C, 10 min.
O
Me O
O
O
O Me 230
(b)
Me Me O O O (c)
O
H
O
237
H
Me
OH
Me Me
OH O 226
R
O BF3
BF3·Et2O CH2Cl2,−40 °C 23% R = t -BuO
O
O
Me
R
O Me 231
O
Me O
O Me
Me
Me Me
H
H O
OH O 227
O
H
H
O
H O
O BF3
R
Me
O
O Me 233 H O
R
O BF3
H
O
O Me 234
O Me
BF3·Et2O CH2Cl2 , 20°C 31% R = Me2N
235, R= Me2N 236, R= t-BuO
H
Ot-Bu 228
Me Me
O O
O BF3
Me Me
H H
O Me
Me Me
12% -
Me Me
d+ Me O Me
d+ O Me 232 O
O
O
56% 70% 10%
H
Me O BF3
H
H
H O
<4% 35%
R
O
Me Me
Me
223, R = t-BuO 224, R = PhNH 225, R = Me2N
O Me 229
R
O
O
O BF3
O
O Me
Me
H
OH
O Me H Me 238
Scheme 15.18 (a) Stereochemical outcome of epoxide-opening cascades as function of the type of terminating nucleophile; (b) mechanistic rationale; (c) effects of the terminating nucleophile on the outcome of cascades leading to poly(tetrahydropyran) systems [31].
electrophile for nucleophilic attack by the neighboring epoxide, forming a new ring and a new epoxonium intermediate, thus propagating the cascade. The direction of the cascade in these reactions is therefore controlled by the position of the halide, and the need for selective activation of only one of many epoxides in polyepoxide substrates is eliminated. As Murai focused on the trans-disubstituted epoxide substrates without directing groups, these studies constituted the first efforts toward epoxide-opening cascades in the synthesis of ladder polyether fragments. While this mode of activation proved to be somewhat effective in reactions involving a single epoxide, cascade cyclizations of 1,4-diepoxides uniformly failed to produce any of the desired trans-fused oligo(tetrahydropyran)s [164, 195]. As described in Section 15.4.1, Floreancig and coworkers have demonstrated that mesolytic benzylic carbon–carbon bond cleavage of the radical cations of homobenzylic ethers, such as 93, 95, 97, and 99 (Scheme 15.5b), forms oxonium ions that react with pendent epoxides to form epoxonium ions capable of undergoing further nucleophilic attack. This strategy is conceptually similar to Ag-promoted reactions of halo epoxides reported by Murai [164] and Jamison [195]. After their initial success in the synthesis of bis-tetrahydrofuran fragments of polyether ionophores, the Floreancig group published their experimental and computational studies, in collaboration with Houk and coworkers, on the structure–reactivity relationships for intramolecular additions to bicyclic epoxonium ions [196]. They observed that
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers
ring size has a significant impact on these processes, with endo-cyclizations being preferred for bicyclo[4.1.0] epoxonium ions bearing an alkyl directing group and exo-cyclizations being preferred for bicyclo[3.1.0] epoxonium ions, despite the presence of a directing group (Scheme 15.19a). The authors propose that these effects can be attributed to the ability of the larger ring to accommodate a looser transition state with significant SN 1 character, thereby promoting the endo-process regardless of solvent polarity. As they had clearly demonstrated that the epoxonium ion structure is a significant determinant of regioselectivity under these kinetic cyclization conditions, Floreancig and coworkers then designed several extended substrates that underwent cascade cyclizations to form fused tricyclic systems under the oxidative conditions described earlier (Scheme 15.19a). Gagn´e and coworkers have reported another method for initiation of epoxide-opening cascades that provides good control over direction of the cascade [197]. In this work, allenyl epoxides 252 and 253 were treated with a cationic gold(I) phosphite catalyst to form epoxonium intermediates that were, in turn, opened by the pendant alcohol nucleophiles to afford bicyclic products 254 and 255 (Scheme 15.19b). Appropriately positioned methyl groups were required for the reactions to proceed with good regioselectivity. As described earlier, the Jamison laboratory has successfully utilized the directing effects of trimethylsilyl groups to develop an iterative approach to the synthesis of oligo(tetrahydropyran) fragments (Scheme 15.14c). However, when a cascade reaction under the same Lewis acid conditions was attempted on diepoxide 256, with suitably positioned silyl groups, the only isolable product was bis-tetrahydrofuran 257 (Scheme 15.20a) [198]. Thorough evaluation of reaction conditions revealed that the outcome of this reaction was very different when a Brønsted base in alcoholic solvents was used. Under these conditions diepoxide 256 underwent a cascade to produce THP diad 258. Surprisingly, the trimethylsilyl directing group was absent from the ring junction in the product. Further modification to the design of polyepoxide substrates and reaction conditions resulted in the development of epoxide-opening cascades directed by ‘‘disappearing’’ silyl groups (Scheme 15.20b) [198]. These modifications included the pre-formation of one THP ring in substrates 259, 262, and 265 prior to the cascade and the addition of CsF to the reaction mixture. It was proposed that these cascades proceed as a sequence of silyl-directed epoxide opening followed by protiodesilylation via a homo-Brook rearrangement pathway. After each Brook rearrangement, removal of the silyl group by fluoride reveals the alkoxide nucleophile to propagate the cascade. While disappearing directing groups address problems related to the removal of substituents not present in the natural targets, these reactions developed by Jamison and coworkers suffer from the inability to incorporate the methyl substituents found frequently at ring junctions. In their efforts to develop a directing group-free endo-selective intramolecular epoxide opening and pave the way to a directing group-free cascade capable of incorporating all types of epoxide substitution, the Jamison group reasoned that pre-organization of the substrate in an appropriate fashion could encourage the cyclization toward the endo pathway by altering the approach of the alcohol nucleophile to the epoxide. When designing
575
2
7
O 6 3
O 2
O
OMe
MeO2C
O
OH 254
CO2Me Me O
249
Me
O O
Ot -Bu
Ot-Bu
Ot-Bu O
9
5
O Me
O O
10
5 12
Me H O 4 O Me
O O
•
X O
Me
MeO
OH
O
H O O Me
O O
X= NTs
AgOTf, CH2Cl2
(PhO)3PAuCl
H 250, 30%
H
MeO
252, X= C(CO2Me)2 253, X= NTs 59% (trans:cis =4:1)
t-BuPh, DCE
hn, NMQPF6
O H MeO 247 (4aS, 5aS, 10aR, 12aR ), 79% not shown 248 (4aR, 5aS, 10aR, 12aS ), 54%
t-BuPh, DCE
hn, NMQPF6
AgOTf, CH2Cl2 X = C(CO2Me)2 40%
MeO
11
Me H O 4 2'
2 5
4
O
O
Me H
O H H 255
Me O
251, 25%
H H
TsN
O
O
O O
MeO O O O O Me H Me H H 241 (4aR, 5aS, 9aR, 11aS ), 6% 243 (4S )-4-((2S, 2'R, 5R ), 34% not shown 242 (4aS, 5aS, 9aR, 11aR ), 9% 244 (4R )-4-((2S, 2'R, 5S), 52%
t-BuPh, DCE
hn, NMQPF6
(PhO)3PAuCl
O
Me Me O 245 (2S,3S,6R,7R ) not shown 246 (2R,3R,6R,7R)
OMe
6
O
239 (2R,3R,6R,7R ) not shown 240 (2S,3S,6R,7R )
OMe
3
7
O
Scheme 15.19 Endo-selective epoxide-opening cascades initiated by: (a) oxidative cleavage of homobenzylic ethers [196] and (b) cationic gold(I) phosphite catalyst with allenyl epoxides [197].
(b)
Ph2CH
(a)
Ph2CH
Ph2CH
Me O
Me O
576
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers HO H
HO O Me
Me3Si
O Me SiMe3 257 BF3·Et2O 50% CH2Cl2 HO SiMe3 O Me O SiMe3 256 14%
HO
H
Me
O Me
Me O H Me3Si 258 (a)
Me
O SiMe3
(b)
H
O
H
Cs2CO3, CsF
O O
R SiMe3 >95:5 dr 259, R = TMS >95:5 dr 260, R = H H HO SiMe3 O
Cs2CO3 CH3OH H
H
577
MeOH
1. Cs2CO3, CsF MeOH
2. Ac2O, DMAP O R Py, CH2Cl2 SiMe3 90: 10 dr 262, R = TMS 92: 8 dr 263, R = H H 1. Cs2CO3, CsF HO SiMe3 MeOH O 2. Ac2O, DMAP O O R Py, CH2Cl2 SiMe3
O
92: 8 dr 265, R = TMS 90: 10 dr 266, R = H
HO
H
H
O
55% 261 62% 261 H H H R'O O Me
O O H H H 39% 264, R' = Ac 35% 182, R' = H (1 step)
Me AcO
H
H
O
H
H
H
O
O
H
15% 267 20% 267
Scheme 15.20 Epoxide-opening cascades directed by a disappearing silyl group by Jamison [198].
substrates to test this hypothesis, Jamison proposed that such a situation may exist in cyclizations of disubstituted epoxy-alcohol 268, where one THP is formed prior to the epoxide-opening reaction (Figure 15.8). Analysis of the transition states in cyclization reaction of 268 revealed that trans-bicyclo[4.4.0]decane derivatives are typically less strained than the corresponding trans-bicyclo[4.3.0]nonanes. If the greater stability of the endo-product 261 compared to 269 was reflected in the energies of the transitions states leading to these products, increased endo selectivity would be observed under kinetic control. While investigating this hypothesis, Jamison discovered that the regioselectivity of epoxide opening in epoxy-alcohol 268 is dependent on the pH of the aqueous medium used to promote cyclization [199]. The selectivity for the desired THP product 261 increases substantially as the pH of the reaction environment approaches neutrality (Figure 15.8a). Increased water content in the reaction mixture maps well with the increase in selectivity and rate of conversion of reactions examined in water–THF mixtures (Figure 15.8b). Aqueous cascade reactions of templated diepoxide 270 and triepoxide 271 that lack directing groups were also examined and found to proceed with good efficiency in water at elevated temperatures, affording the THP triad 182 and tetrad 272 [199]. Hydrogen-bonding interactions between the THP template, epoxide, and water molecules were proposed as the origin of endo selectivity in the described reactions. Kinetic studies of the cyclization reactions of epoxy alcohol 268 and its carbocyclic analog featuring a cyclohexane in place of the THP template suggest existence of at least two competing mechanisms that are first- and second-order in water, respectively [200]. It was proposed that the selective pathway is second-order in water and operable only for the THP-templated epoxy alcohol 268. Jamison hypothesized that epoxy alcohol cyclizations in water occur for hydrated conformations that
H
H
O
O
O
0.0
HO
H
H
2.0
O
4.0
6.0
70 °C 60%
H2O
maximum selectivity near pH7
270, 4:1 dr
0.0
2.0
4.0
6.0
8.0
10.0
12.0
O H 268
O
Me
HO
pH
8.0
H
H
HO
Me
O
O
H
H H 182
H
O
Me
10.0 12.0 14.0
Dependence of THP:THF (261:269) Selectivity on pH
Me
H
H
H
O
(b)
O H 269
H
0
10
20
30
40
50
60
40
60
selectivity
O 271, 3:1 dr
O
HO
H
H
O
70 °C 53%
H2O
HO
Me
H
H
80 % H2O in tetrahydrofuran
20
O
H
H
O
O
272
H
H
1.0 100
2.0
3.0
4.0
5.0
6.0
7.0
8.0
70
80 conversion
Dependence of Conversion and Selectivity on Water Content
Me
HO H O
9.0
0
O H 261
H
90
O
H
H
O
THP : THF (261:269) selectivity
Figure 15.8 (a) THP : THF selectivity in reactions of THP-templated epoxides as a function of pH; (b) effect of water on conversion and selectivity in cyclizations of 268 in THF–water mixtures; (c) directing group-free epoxide-opening cascades promoted by water [199].
(c)
Me
(a)
THP : THF (261:269) selectivity
HO
conversion (%)
578
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers
579
possess the appropriate geometry, which is possibly attained in the form of a twist-boat conformation that is stabilized by hydrogen-bonding interactions with water molecules. The electron-withdrawing effects of the oxygen in the THP template may also provide electronic bias for endo cyclization via destabilization of the positive charge at the exo position. Methyl groups are commonly encountered at ladder polyether ring junctions, with approximately one in five ring junctions bearing a methyl substituent. Therefore, the putative polyepoxide precursors to ladder polyethers, in addition to disubstituted epoxides, regularly feature two types of trisubstituted epoxides. A methyl group on a trisubstituted epoxide can be situated in a position where its directing effect promotes endo-opening under acidic conditions (as in 273, Scheme 15.21a) or in a position that would normally promote exo-selective epoxide opening under the same conditions (as in 274, Scheme 15.21a). To incorporate methyl substituents at the ring junctions of the final products in epoxide-opening cascades, the Jamison group, in an immediate extension of their work on aqueous directing group-free endo-selective epoxide openings, prepared and evaluated cyclization reactions of templated epoxy alcohols 273 and 274, which feature both types of methyl substitution on the epoxide (Scheme 15.21a) [201, 202]. Acid-catalyzed cyclizations of epoxy alcohol 273 proceeded with high endo selectivity due to the electronic effect of the methyl substituent at the endo site of attack on the epoxide. Base-promoted cyclizations of the same molecule, however, predominantly produced the exo product, bicycle 277. In contrast, base-promoted cyclizations of the more challenging epoxy alcohol 274 proceeded with moderate endo selectivity, while acid-promoted reactions afforded exo product 278. Both 273 and 274 produced endo products 275 and 276 with good selectivity when cyclized in deionized water. In addition to achieving high endo selectivity, aqueous cyclizations circumvent side reactions associated with other activation conditions, such as the rearrangement of epoxide 273 to a iso-propyl ketone via 1,2-hydride shifts under acidic conditions.
Me
HO 2 O R
H
Me
O H R1 1 273, R = Me; R2= H 274, R1= H; R2= Me
HO
R1 H O R2
H
HO
Me R2
O
Me
HO H 3 R O
O H R2 279, R1= Me; R2= Me; R3= H 280, R1= H; R2= Me; R3= H (b) 281, R1= H; R2= H; R3= H O
H
H
Reaction Cs2CO3 CSA BF3•OEt2 H2O CH2Cl2 CH2Cl2 conditions MeOH O
275, R = Me; R = H 277, R = Me; R2= H 276, R1= H; R2= Me 278, R1= H; R2= Me 1
2
1
(a) R1
R1 O
HO Me
R2
H
R1
O
O
R3
H
H
O
275:277 ratio
1 : 17
5.8 : 1 >20 : 1 >20 : 1
276:278 ratio
3.0 : 1
1 : 5.2
Isolated yield, 282
0%
43%
61%
54%
Isolated yield, 283
0%
46%
63%
67%
0%
0%
0%
32%
282, R1= Me; R2= Me; R3= H Isolated yield, 284 283, R1= H; R2= Me; R3= H 281, R1= H; R2= H; R3= Me
Scheme 15.21 (a) Cyclizations of templated, trisubstituted epoxides 273 and 274; (b) water overcomes methyl group directing effects in epoxide-opening cascades [201].
1 : 11 4.9 : 1
580
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
Extending this work further, the Jamison group demonstrated that methyl substituents on epoxides are also tolerated in aqueous cascades (Scheme 15.21b) [201, 202]. Under aqueous conditions, diepoxides 279 and 280, which incorporate methyl substituents at the positions of endo attack afforded the triads 282 and 283 (Scheme 15.21b). Diepoxide 281, with a methyl substituent at the exo site, afforded the THP triad 284. While diepoxides 279 and 280 also produced some of the desired products upon acidic activation, only aqueous conditions afforded any of the THP triad 284 from diepoxide 281. Aqueous epoxide-opening cascades overcome the need for methyl substituents to be uniformly distributed on each epoxide and can accommodate both types of trisubstituted epoxides in combination with disubstituted epoxides [203]. 15.4.4.3 Applications of Epoxide-Opening Cascades in the Synthesis of Ladder Polyethers The endo-selective opening of alkenyl epoxides has become a standard tool for synthesis of tetrahydropyran rings that has been used by the groups of Nicolaou, Yamamoto, Nakata, Mori, and Sasaki in syntheses of hemibrevetoxin B [204–207], brevetoxin B [208–210], brevetoxin A [211], gambierol [212–216], and brevenal [135, 136]. This method is generally not amenable to cascades of more than one epoxide. Nevertheless, it has been used in iterative syntheses of ladder polyethers. For example, Nicolaou’s approach to the FG ring system of brevetoxin B includes an acid-catalyzed opening of alkenyl epoxides 285 and 287 to form both rings in the FG fragment (288, Scheme 15.22a) [208]. Other iterative approaches to oligo(tetrahydropyran)s have also been used in the synthesis of ladder polyether natural products and their fragments. Mori and coworkers have reported a total synthesis of hemibrevetoxin B (289, Scheme 15.22b) that relies solely on their iterative strategy for construction of trans-fused tetrahydropyran rings [217]. In combination with methods that allow for ring expansion of tetrahydropyran to oxepane systems [186], hemibrevetoxin B was prepared in an iterative fashion using endo-selective intramolecular opening of epoxysulfones (Scheme 15.22b). The Mori group was also successful in the preparation of gambierol and the ABCDEF-ring system of yessotoxins and adriatoxins using analogous strategy [35, 187, 216]. An epoxide-opening cascade was utilized in the synthesis of hemibrevetoxin B by Holton [218]. Although only one epoxide is involved in this reaction, two cyclic ethers of the natural product are produced in a single operation. Computational studies by Houk [40, 42] that suggest that alkyl group-directed 6-endo cyclization normally requires a loose, SN 1-like transition state prompted Holton to carry out the cascade in a strongly polar solvent. In a fashion similar to the work of the Murai [164], Jamison [80, 195], and Floreancig [196], the alkene in 294 was activated with N-(phenylseleno)phthalimide, and the cascade leading to formation of the 7,6-fused BC ring system of hemibrevetoxin B proceeded in high yield (Scheme 15.23). In efforts to expand the utility of aqueous epoxide-opening cascades for synthesis of ladder polyether fragments, the Jamison group investigated oxygen-containing
HO
O
Me
H
285
O H H H
Me H O
CH2Cl2 0 to 25 °C 83%
CSA
O H
OH
BOMO
R
289, hemibrevetoxin B
Cl
H
Me Me OH O
286
H
CHO
H
Me Me O
H
290
291
PhO2S O RO
287
O OH H Me
Me Me O
O TolO2S
BOMO
TBDPSO
Cl OH
R
292
OTBDPS Me O TolO2S
85%
CH2Cl2, 0 °C
PPTS
HO
HO
H
H
293
O
H
OBn OBn
Me Me H O OH F G BOMO O H H Me 288 R
Scheme 15.22 (a) Iterative synthesis of FG fragment of brevetoxin B by Nicolaou [208]; (b) retrosynthetic analysis for the iterative synthesis of hemibrevetoxin B by Mori [217]. R = prenyl, BOM = benzyloxymethyl, PPTS = pyridinium p-toluenesulfonate.
(b)
(a)
BOMO
R
15.4 Epoxide-Opening Cascades in the Synthesis of Polycyclic Polyethers 581
582
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening O BnO Me BnO
OTIPS
HO Me
OMOM O
Me
O
O
H
Me
Me O
O
H
294 O (CF3)2CHOH 83%
NSePh O
O BnO Me BnO
OTIPS Me H OMOM O H
289, hemibrevetoxin B
Me
O Me
Me
O
O H H SePh
H
O
O H
295
Scheme 15.23 Cascade approach to BC rings of hemibrevetoxin B reported by Holton [218].
templates that could produce fragments suitable for coupling and further elaboration to natural products [219]. Evaluation of a benzylidene acetal template revealed significant differences in the reactivity of epoxy alcohols templated by this motif (e.g., 296, Scheme 15.24) compared to those templated by a THP. Slow cyclization rates in water and the instability of benzylidene acetals prevented these substrates from being used in aqueous reactions. However, silica gel was found to promote cyclization of benzylidene acetal templated epoxy alcohol 296 to 297 in good yield and with high endo selectivity. Elaboration of 297 to the triepoxy-alcohol 298, with a functionalized THP template, set the stage for an epoxide-opening cascade. Incubation of 298 in water at 60 ◦ C for five days afforded some of the desired THP tetrad 300 and a larger quantity of 299, in which two THP rings had formed but the final epoxide remained intact (Scheme 15.24). More forceful conditions (80 ◦ C, nine days) drove the cyclization reaction of 298 to completion and allowed the isolation of 300, the HIJK fragment of gymnocin A, in 35% yield upon acetylation. THP HO
H
H O
O H 296
O
SiO2
Ph
CH2Cl2, 40°C 78%, >9 : 1 endo : exo selectivity
H
BnO
O
HO H
O
H H
O
O
reaction conditions:
H
i. H2O, 60 °C, 5 days; ii. Ac2O, Et3N i. H2O, 80 °C, 9 days; ii. Ac2O, Et3N
O
H H 299
O
O
OH
AcO
HO
+
H
BnO
23% --
Scheme 15.24 Cascade synthesis of the HIJK fragment of gymnocin A reported by Jamison [219].
H
H
O
H
O
H OMe
O 298
OMe Me
BnO
Ph
O H 297 H
O AcO
H
H
O
H
H
O
H H 300 14% 35%
H O
Me
OH
H OMe OH Me
15.5 Summary and Outlook
tetrad 300 features four differently substituted hydroxyl groups ready for further synthetic elaboration.
15.5 Summary and Outlook
We have presented examples of the many uses of biomimetic epoxide-opening cyclizations in the synthesis of polycyclic polyethers. Epoxide-opening cascades are, however, in no way limited only to the synthesis of polycyclic polyether natural products. Cascade cyclizations that involve epoxide-opening steps have, in fact, found use in many syntheses of natural products outside the polyether families discussed herein (i.e., schweinfurthins and wortmannin) [220–224]. Epoxide-opening cascades that create 2,5-linked oligotetrahydrofurans proceed in agreement with Baldwin’s rules and almost always with high selectivity for the smaller rings. While regioselectivity in these reactions is not a major challenge, further improvements are needed to accommodate more diverse substrates and better address challenges of total synthesis. Development of mild conditions to enable better functional group compatibility and methods for selective activation of specific epoxides are imperative for these reactions to proceed with higher yields. This is especially true for the cascades that involve three or more epoxides. Epoxide-opening cascades leading to the formation of fused polyethers are harder to achieve, burdened by empirical rules of regioselectivity that generally regard endo-selective epoxide-opening reactions to be disfavored. Despite considerable work toward epoxide-opening cascades that would produce a ladder polyether in a single synthetic operation, this goal remains elusive. For a cascade to be successful it must be sufficiently flexible to accommodate various ring sizes and epoxide substitution patterns, among other significant challenges, thus making the design of such reactions all the more difficult. While many problems have been addressed in creative ways, further advances are necessary in several directions. For example, construction of seven-, eight-, and nine-membered rings without directing groups on the epoxide remains challenging. Furthermore, accommodating various ring sizes in a single cascade is difficult. Polyepoxide cyclization reactions for the synthesis of ladder polyethers should ideally be able to incorporate a larger number of epoxides (greater than 3–4 epoxides at a time, which is the current state of the art) to construct large polyether fragments. Avoiding side reactions and maintaining efficiency is currently a substantial hurdle for these ambitions. Finally, any such methodology needs to provide products that are amenable to rapid functionalization and elaboration into entire natural products. Reagents and catalysts that will enable efficient control of regioselectivity in epoxide-opening cyclization hold great promise. These promoters could be either small molecules or the enzymes that are postulated to be involved in the biosynthesis of the polycyclic polyether natural products. Several reports testify to viability of this approach and suggest that research in this area may be worthwhile [39, 41, 43, 44, 70, 72]. The challenges posed by the synthesis of the fascinating family of
583
584
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
polycyclic polyether natural products will continue to stimulate research and bring exciting new developments to the field of organic synthesis.
References 1. Dutton, C.J., Banks, B.J., and Cooper,
2.
3. 4. 5. 6. 7. 8.
9.
10. 11. 12.
13.
14. 15. 16. 17.
18. 19.
C.B. (1995) Nat. Prod. Rep., 12, 165–181. Fernandez, J.J., Souto, M.L., and Norte, M. (2000) Nat. Prod. Rep., 17, 235–246. Rein, K.S. and Borrone, J. (1999) Comp. Biochem. Physiol. Part B, 124, 117–131. Gallimore, A.R. (2009) Nat. Prod. Rep., 26, 266–280. Vilotijevic, I. and Jamison, T.F. (2009) Angew. Chem. Int. Ed., 48, 5250–5281. Baldwin, J.E. (1976) J. Chem. Soc., Chem. Commun., 734–736. Johnson, C.D. (1993) Acc. Chem. Res., 26, 476–482. Coxon, J.M., Hartshorn, M.P., and Swallow, W.H. (1973) Aust. J. Chem., 26, 2521–2526. Danishefsky, S., Dynak, J., Hatch, E., and Yamamoto, M. (1974) J. Am. Chem. Soc., 96, 1256–1259. Katsuki, T. and Sharpless, K.B. (1980) J. Am. Chem. Soc., 102, 5974–5976. Katsuki, T. and Martin, V.S. (1996) Org. React., 48, 1–299. Zhang, W., Loebach, J.L., Wilson, S.R., and Jacobsen, E.N. (1990) J. Am. Chem. Soc., 112, 2801–2803. Jacobsen, E.N., Zhang, W., Muci, A.R., Ecker, J.R., and Deng, L. (1991) J. Am. Chem. Soc., 113, 7063–7064. Tu, Y., Wang, Z.-X., and Shi, Y. (1996) J. Am. Chem. Soc., 118, 9806–9807. Shi, Y. (2004) Acc. Chem. Res., 37, 488–496. Wong, O.A. and Shi, Y. (2008) Chem. Rev., 108, 3958–3987. Xia, Q.H., Ge, H.Q., Ye, C.P., Liu, Z.M., and Su, K.X. (2005) Chem. Rev., 105, 1603–1662. Heffron, T.P. and Jamison, T.F. (2003) Org. Lett., 5, 2339–2342. Morimoto, Y., Nishikawa, Y., Ueba, C., and Tanaka, T. (2006) Angew. Chem. Int. Ed., 45, 810–812.
20. Nicolaou, K.C., Prasad, C.V.C., Somers,
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34. 35.
P.K., and Hwang, C.K. (1989) J. Am. Chem. Soc., 111, 5335–5340. Mori, Y., Yaegashi, K., and Furukawa, H. (1996) J. Am. Chem. Soc., 118, 8158–8159. Fujiwara, K., Tokiwano, T., and Murai, A. (1995) Tetrahedron Lett., 36, 8063–8066. Nicolaou, K.C., Duggan, M.E., Hwang, C.K., and Somers, P.K. (1985) J. Chem. Soc., Chem. Commun., 1359–1362. Nicolaou, K.C., Prasad, C.V.C., Somers, P.K., and Hwang, C.K. (1989) J. Am. Chem. Soc., 111, 5330–5334. Suzuki, T., Sato, O., and Hirama, M. (1990) Tetrahedron Lett., 31, 4747–4750. Matsukura, H., Morimoto, M., Koshino, H., and Nakata, T. (1997) Tetrahedron Lett., 38, 5545–5548. Mukai, C., Ikeda, Y., Sugimoto, Y.-i., and Hanaoka, M. (1994) Tetrahedron Lett., 35, 2179–2182. Mukai, C., Sugimoto, Y.-i., Ikeda, Y., and Hanaoka, M. (1994) Tetrahedron Lett., 35, 2183–2186. Mukai, C., Sugimoto, Y.-i., Ikeda, Y., and Hanaoka, M. (1998) Tetrahedron, 54, 823–850. McDonald, F.E., Wang, X., Do, B., and Hardcastle, K.I. (2000) Org. Lett., 2, 2917–2919. Bravo, F., McDonald, F.E., Neiwert, W.A., Do, B., and Hardcastle, K.I. (2003) Org. Lett., 5, 2123–2126. Hudrlik, P.F., Holmes, P.E., and Hudrlik, A.M. (1988) Tetrahedron Lett., 29, 6395–6398. Adiwidjaja, G., Fl¨orke, H., Kirschning, A., and Schaumann, E. (1995) Tetrahedron Lett., 36, 8771–8774. Mori, Y. (1997) Chem. Eur. J., 3, 849–852. Mori, Y., Furuta, H., Takase, T., Mitsuoka, S., and Furukawa, H. (1999) Tetrahedron Lett., 40, 8019–8022.
References 36. Fujiwara, K., Mishima, H., Amano, A.,
37. 38. 39. 40.
41.
42. 43.
44.
45.
46.
47.
48.
49.
50.
51. 52.
53.
54.
Tokiwano, T., and Murai, A. (1998) Tetrahedron Lett., 39, 393–396. Fujiwara, K., Saka, K., Takaoka, D., and Murai, A. (1999) Synlett, 1037–1040. Tokiwano, T., Fujiwara, K., and Murai, A. (2000) Chem. Lett., 272–273. Janda, K.D., Shevlin, C.G., and Lerner, R.A. (1993) Science, 259, 490–493. Na, J., Houk, K.N., Shevlin, C.G., Janda, K.D., and Lerner, R.A. (1993) J. Am. Chem. Soc., 115, 8453–8454. Janda, K.D., Shevlin, C.G., and Lerner, R.A. (1995) J. Am. Chem. Soc., 117, 2659–2660. Na, J. and Houk, K.N. (1996) J. Am. Chem. Soc., 118, 9204–9205. Tokunaga, M., Larrow, J.F., Kakiuchi, F., and Jacobsen, E.N. (1997) Science, 277, 936–938. Wu, M.H., Hansen, K.B., and Jacobsen, E.N. (1999) Angew. Chem. Int. Ed., 38, 2012–2014. Berger, J., Rachlin, A.I., Scott, W.E., Sternbach, L.H., and Goldberg, M.W. (1951) J. Am. Chem. Soc., 73, 5295–5298. Harned, R.L., Hidy, P.H., Corum, C.J., and Jones, K.L. (1951) Antiobiot. Chemother., 1, 594–596. Agtarap, A., Chamberlin, J.W., Pinkerton, M., and Steinrauf, L.K. (1967) J. Am. Chem. Soc., 89, 5737–5739. Pressman, B.C., Harris, E.J., Jagger, W.S., and Johnson, J.H. (1967) Proc. Natl. Acad. Sci. U.S.A., 58, 1949–1956. Shumard, R.F. and Callender, M.E. (1967) Antimicrob. Agents Chemother., 7, 369–377. Raun, A.P., Cooley, C.O., Potter, E.L., Rathmacher, R.P., and Richardson, L.F. (1976) J. Anim. Sci., 43, 670–677. Russell, J.B. and Houlihan, A.J. (2003) FEMS Microbiol. Rev., 27, 65–74. Westley, J.W. (1977) in Advances in Applied Microbiology, vol. 22 (ed. D. Perlman), Academic Press, New York, pp. 177–223. Westley, J.W., Evans, R.H. Jr., Sello, L.H., Troupe, N., Liu, C.M., and Miller, P.A. (1981) J. Antibiot., 34, 1248–1252. Westley, J.W., Evans, R.H. Jr., Sello, L.H., Troupe, N., Liu, C.-M., Blount,
55.
56.
57.
58. 59.
60. 61.
62. 63. 64. 65.
66.
67.
68.
69.
J.F., Pitcher, R.G., Williams, T.H., and Miller, P.A. (1981) J. Antibiot., 34, 139–147. Westley, J.W., Evans, R.H. Jr., Harvey, G., Pitcher, R.G., Pruess, D.L., Stempel, A., and Berger, J. (1974) J. Antibiot., 27, 288–297. Westley, J.W., Blount, J.F., Evans, R.H. Jr., Stempel, A., and Berger, J. (1974) J. Antibiot., 27, 597–604. Cane, D.E., Liang, T.-C., and Hasler, H. (1981) J. Am. Chem. Soc., 103, 5962–5965. Westley, J.W. (1981) Antibiotics, 4, 41–73. Cane, D.E., Liang, T.C., and Hasler, H. (1982) J. Am. Chem. Soc., 104, 7274–7281. Hutchinson, C.R. (1983) Acc. Chem. Res., 16, 7–14. Cane, D.E., Celmer, W.D., and Westley, J.W. (1983) J. Am. Chem. Soc., 105, 3594–3600. Townsend, C.A. and Basak, A. (1991) Tetrahedron, 47, 2591–2602. Koert, U. (1995) Angew. Chem., Int. Ed. Engl., 34, 298–300. McDonald, F.E. and Towne, T.B. (1994) J. Am. Chem. Soc., 116, 7921–7922. McDonald, F.E., Towne, T.B., and Schultz, C.C. (1998) Pure Appl. Chem., 70, 355–358. Leadlay, P.F., Staunton, J., Oliynyk, M., Bisang, C., Cortes, J., Frost, E., Hughes-Thomas, Z.A., Jones, M.A., Kendrew, S.G., Lester, J.B., Long, P.F., McArthur, H.A.I., McCormick, E.L., Oliynyk, Z., Stark, C.B.W., and Wilkinson, C.J. (2001) J. Ind. Microbiol. Biotechnol., 27, 360–367. Oliynyk, M., Stark, C.B.W., Bhatt, A., Jones, M.A., Hughes-Thomas, Z.A., Wilkinson, C., Oliynyk, Z., Demydchuk, Y., Staunton, J., and Leadlay, P.F. (2003) Mol. Microbiol., 49, 1179–1190. Bhatt, A., Stark, C.B.W., Harvey, B.M., Gallimore, A.R., Demydchuk, Y.A., Spencer, J.B., Staunton, J., and Leadlay, P.F. (2005) Angew. Chem. Int. Ed., 44, 7075–7078. Gallimore, A.R., Stark, C.B.W., Bhatt, A., Harvey, B.M., Demydchuk, Y., Bolanos-Garcia, V., Fowler, D.J.,
585
586
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
70.
71.
72.
73.
74. 75.
76.
77. 78.
79.
80.
81.
82.
83.
84.
85. Yasumoto, T. and Murata, M. (1993) Staunton, J., Leadlay, P.F., and Chem. Rev., 93, 1897–1909. Spencer, J.B. (2006) Chem. Biol., 13, 86. Inoue, M. (2005) Chem. Rev., 105, 453–460. 4379–4405. Shichijo, Y., Migita, A., Oguri, H., Watanabe, M., Tokiwano, T., Watanabe, 87. Nakata, T. (2005) Chem. Rev., 105, 4314–4347. K., and Oikawa, H. (2008) J. Am. Chem. 88. Sasaki, M. and Fuwa, H. (2008) Nat. Soc., 130, 12230–12231. Migita, A., Shichijo, Y., Oguri, H., Prod. Rep., 25, 401–426. 89. Nicolaou, K.C., Frederick, M.O., and Watanabe, M., Tokiwano, T., and Aversa, R.J. (2008) Angew. Chem. Int. Oikawa, H. (2008) Tetrahedron Lett., 49, Ed., 47, 7182–7225. 1021–1025. Smith, L., Hong, H., Spencer, J.B., and 90. Sellner, K.G., Doucette, G.J., and Leadlay, P.F. (2008) ChemBioChem, 9, Kirkpatrick, G.J. (2003) J. Ind. Micro2967–2975. biol. Biotechnol., 30, 383–406. Rudi, A., Yosief, T., Schleyer, M., and 91. Schrope, M. (2008) Nature, 452, Kashman, Y. (1999) Tetrahedron, 55, 24–26. 92. Murata, M., Legrand, A.M., Ishibashi, 5555–5566. Kashman, Y. and Rudi, A. (2005) Y., Fukui, M., and Yasumoto, T. (1990) Phytochem. Rev., 3, 309–323. J. Am. Chem. Soc., 112, 4380–4386. Sakemi, S., Higa, T., Jefford, C.W., and 93. Flewelling, L.J., Naar, J.P., Abbott, J.P., Bernardinelli, G. (1986) Tetrahedron Baden, D.G., Barros, N.B., Bossart, Lett., 27, 4287–4290. G.D., Bottein, M.-Y.D., Hammond, Hashimoto, M., Kan, T., Nozaki, K., D.G., Haubold, E.M., Heil, C.A., Yanagiya, M., Shirahama, H., and Henry, M.S., Jacocks, H.M., Leighfield, Matsumoto, T. (1990) J. Org. Chem., 55, T.A., Pierce, R.H., Pitchford, T.D., 5088–5107. Rommel, S.A., Scott, P.S., Steidinger, Matsuo, Y., Suzuki, M., and Masuda, K.A., Truby, E.W., Van Dolah, F.M., M. (1995) Chem. Lett., 1043–1044. and Landsberg, J.H. (2005) Nature, 435, Manriquez, C.P., Souto, M.L., Gavin, 755–756. 94. Lewis, R.J. (2001) Toxicon, 39, 97–106. J.A., Norte, M., and Fernandez, J.J. 95. Murata, M., Legrand, A.M., and (2001) Tetrahedron, 57, 3117–3123. Domingo, V., Arteaga, J.F., Moral, Yasumoto, T. (1989) Tetrahedron Lett., J.F.Qd., and Barrero, A.F. (2009) Nat. 30, 3793–3796. 96. Murata, M., Legrand, A.M., Ishibashi, Prod. Rep., 26, 115–134. Tanuwidjaja, J., Ng, S.-S., and Jamison, Y., and Yasumoto, T. (1989) J. Am. T.F. (2009) J. Am. Chem. Soc., 131, Chem. Soc., 111, 8929–8931. 97. Murata, M., Naoki, H., Iwashita, T., 12084–12085. Lin, Y.-Y., Risk, M., Ray, S.M., Matsunaga, S., Sasaki, M., Yokoyama, Van Engen, D., Clardy, J., Golik, J., A., and Yasumoto, T. (1993) J. Am. James, J.C., and Nakanishi, K. (1981) Chem. Soc., 115, 2060–2062. 98. Satake, M., Shoji, M., Oshima, Y., J. Am. Chem. Soc., 103, 6773–6775. Murata, M., Naoki, H., Matsunaga, S., Naoki, H., Fujita, T., and Yasumoto, Satake, M., and Yasumoto, T. (1994) T. (2002) Tetrahedron Lett., 43, J. Am. Chem. Soc., 116, 7098–7107. 5829–5832. Sasaki, M., Matsumori, N., Maruyama, 99. Ferrari, S., Ciminiello, P., T., Nonomura, T., Murata, M., Dell’Aversano, C., Forino, M., Tachibana, K., and Yasumoto, T. Malaguti, C., Tubaro, A., Poletti, R., (1996) Angew. Chem., Int. Ed. Engl., 35, Yasumoto, T., Fattorusso, E., and 1672–1675. Rossini, G.P. (2004) Chem. Res. ToxiNonomura, T., Sasaki, M., Matsumori, col., 17, 1251–1257. 100. Ronzitti, G., Callegari, F., Malaguti, C., N., Murata, M., Tachibana, K., and and Rossini, G.P. (2004) Br. J. Cancer, Yasumoto, T. (1996) Angew. Chem., Int. 90, 1100–1107. Ed. Engl., 35, 1675–1678.
References 101. Nagai, H., Murata, M., Torigoe, K.,
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114. 115.
116.
Satake, M., and Yasumoto, T. (1992) J. Org. Chem., 57, 5448–5453. Nagai, H., Mikami, Y., Yazawa, K., Gonoi, T., and Yasumoto, T. (1993) J. Antibiot., 46, 520–522. Bourdelais, A.J., Campbell, S., Jacocks, H., Naar, J., Wright, J.L.C., Carsi, J., and Baden, D.G. (2004) Cell. Mol. Neurobiol., 24, 553–563. Bourdelais, A.J., Jacocks, H.M., Wright, J.L.C., Bigwarfe, P.M. Jr., and Baden, D.G. (2005) J. Nat. Prod., 68, 2–6. Abraham, W.M., Bourdelais, A.J., Sabater, J.R., Ahmed, A., Lee, T.A., Serebriakov, I., and Baden, D.G. (2005) Am. J. Respir. Crit. Care. Med., 171, 26–34. Baden, D.G., Abraham, W.M., Bourdelais, A.J., and Michelliza, S. (2009) Fused pentacyclic polyethers US 7638500. Poli, M.A., Mende, T.J., and Baden, D.G. (1986) Mol. Pharmacol., 30, 129–135. Bidard, J.N., Vijverberg, H.P.M., Frelin, C., Chungue, E., Legrand, A.M., Bagnis, R., and Lazdunski, M. (1984) J. Biol. Chem., 259, 8353–8357. Gawley, R.E., Rein, K.S., Jeglitsch, G., Adams, D.J., Theodorakis, E.A., Tiebes, J., Nicolaou, K.C., and Baden, D.G. (1995) Chem. Biol., 2, 533–541. Gawley, R.E., Rein, K.S., Kinoshita, M., and Baden, D.G. (1992) Toxicon, 30, 780–785. Trainer, V.L., Baden, D.G., and Catterall, W.A. (1994) J. Biol. Chem., 269, 19904–19909. Cuypers, E., Abdel-Mottaleb, Y., Kopljar, I., Rainier, J.D., Raes, A.L., Snyders, D.J., and Tytgat, J. (2008) Toxicon, 51, 974–983. Takahashi, M., Ohizumi, Y., and Yasumoto, T. (1982) J. Biol. Chem., 257, 7287–7289. Gusovsky, F., Yasumoto, T., and Daly, J.W. (1989) FEBS Lett., 243, 307–312. Murata, M., Gusovsky, F., Yasumoto, T., and Daly, J.W. (1992) Eur. J. Pharmacol., 227, 43–49. Sinkins, W.G., Estacion, M., Prasad, V., Goel, M., Shull, G.E., Kunze, D.L., and
117.
118. 119. 120.
121. 122.
123.
124.
125.
126.
127. 128.
129.
130.
131.
132.
133.
Schilling, W.P. (2009) Am. J. Physiol. Cell Physiol., 297, C1533–C1543. Mori, M., Oishi, T., Matsuoka, S., Ujihara, S., Matsumori, N., Murata, M., Satake, M., Oshima, Y., Matsushita, N., and Aimoto, S. (2005) Bioorg. Med. Chem., 13, 5099–5103. Nakanishi, K. (1985) Toxicon, 23, 473–479. Chou, H.N. and Shimizu, Y. (1987) J. Am. Chem. Soc., 109, 2184–2185. Gallimore, A.R. and Spencer, J.B. (2006) Angew. Chem. Int. Ed., 45, 4406–4413. Giner, J.-L., Li, X., and Mullins, J.J. (2003) J. Org. Chem., 68, 10079–10086. Lee, M.S., Repeta, D.J., Nakanishi, K., and Zagorski, M.G. (1986) J. Am. Chem. Soc., 108, 7855–7856. Lee, M.S., Qin, G., Nakanishi, K., and Zagorski, M.G. (1989) J. Am. Chem. Soc., 111, 6234–6241. Satake, M. (2000) Tennen Yuki Kagobutsu Toronkai Koen Yoshishu, 42, 259–264. Snyder, R.V., Gibbs, P.D.L., Palacios, A., Abiy, L., Dickey, R., Lopez, J.V., and Rein, K.S. (2003) Mar. Biotechnol., 5, 1–12. Snyder, R.V., Guerrero, M.A., Sinigalliano, C.D., Winshell, J., Perez, R., Lopez, J.V., and Rein, K.S. (2005) Phytochemistry, 66, 1767–1780. Rein, K.S. and Snyder, R.V. (2006) Adv. Appl. Microbiol., 59, 93–125. Perez, R., Liu, L., Lopez, J., An, T., and Rein, K.S. (2008) Mar. Drugs, 6, 164–179. Murata, M., Izumikawa, M., Tachibana, K., Fujita, T., and Naoki, H. (1998) J. Am. Chem. Soc., 120, 147–151. Izumikawa, M., Murata, M., Tachibana, K., Fujita, T., and Naoki, H. (2000) Eur. J. Biochem., 267, 5179–5183. Chou, H.-N., Shimizu, Y., Van Duyne, G., and Clardy, J. (1985) Tetrahedron Lett., 26, 2865–2868. Satake, M., Bourdelais, A.J., Van Wagoner, R.M., Baden, D.G., and Wright, J.L.C. (2008) Org. Lett., 10, 3465–3468. Satake, M., Campbell, A., Van Wagoner, R.M., Bourdelais, A.J., Jacocks, H., Baden, D.G., and Wright,
587
588
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening
134.
135.
136.
137.
138.
139. 140. 141.
142. 143. 144.
145.
146.
147. 148.
149.
J.L.C. (2009) J. Org. Chem., 74, 989–994. Van Wagoner, R.M., Satake, M., Bourdelais, A.J., Baden, D.G., and Wright, J.L.C. (2010) J. Nat. Prod., 73, 1177–1179. Fuwa, H., Ebine, M., and Sasaki, M. (2006) J. Am. Chem. Soc., 128, 9648–9650. Fuwa, H., Ebine, M., Bourdelais, A.J., Baden, D.G., and Sasaki, M. (2006) J. Am. Chem. Soc., 128, 16989–16999. Nicolaou, K.C. and Frederick, M.O. (2007) Angew. Chem. Int. Ed., 46, 5278–5282. Nicolaou, K.C., Frederick, M.O., Burtoloso, A.C.B., Denton, R.M., Rivas, F., Cole, K.P., Aversa, R.J., Gibe, R., Umezawa, T., and Suzuki, T. (2008) J. Am. Chem. Soc., 130, 7466–7476. Giner, J.-L. (2005) J. Org. Chem., 70, 721–724. Wiggins, L.F. and Wood, D.J.C. (1950) J. Chem. Soc., 1566–1575. Benner, S.A., Maggio, J.E., and Simmons, H.E. (1981) J. Am. Chem. Soc., 103, 1581–1582. Simmons, H.E. III and Maggio, J.E. (1981) Tetrahedron Lett., 22, 287–290. Paquette, L.A. and Vazeux, M. (1981) Tetrahedron Lett., 22, 291–294. Paquette, L.A., Williams, R.V., Vazeux, M., and Browne, A.R. (1984) J. Org. Chem., 49, 2194–2197. Weitemeyer, C., Preuss, T., and de Meijere, A. (1985) Chem. Ber., 118, 3993–4005. Liang, S., Lee, C.-H., Kozhushkov, S.I., Yufit, D.S., Howard, J.A.K., Meindl, K., Ruehl, S., Yamamoto, C., Okamoto, Y., Schreiner, P.R., Rinderspacher, B.C., and de Meijere, A. (2005) Chem. Eur. J., 11, 2012–2018. Person, G., Keller, M., and Prinzbach, H. (1996) Liebigs Ann., 507–527. Guiard, S., Giorgi, M., Santelli, M., and Parrain, J.-L. (2003) J. Org. Chem., 68, 3319–3322. Alvarez, E., Manta, E., Martin, J.D., Rodriguez, M.L., Ruiz-Perez, C., and Zurita, D. (1988) Tetrahedron Lett., 29, 2097–2100.
150. Alvarez, E., Diaz, M.T., Perez, R.,
151. 152. 153. 154. 155. 156.
157. 158. 159.
160.
161.
162.
163.
164.
165. 166.
167. 168.
169.
170.
Ravelo, J.L., Regueiro, A., Vera, J.A., Zurita, D., and Martin, J.D. (1994) J. Org. Chem., 59, 2848–2876. Dolle, R.E. and Nicolaou, K.C. (1985) J. Am. Chem. Soc., 107, 1691–1694. Hoye, T.R. and Suhadolnik, J.C. (1985) J. Am. Chem. Soc., 107, 5312–5313. Hoye, T.R. and Suhadolnik, J.C. (1986) Tetrahedron, 42, 2855–2862. Hoye, T.R. and Jenkins, S.A. (1987) J. Am. Chem. Soc., 109, 6196–6198. Hoye, T.R. and Witowski, N.E. (1992) J. Am. Chem. Soc., 114, 7291–7292. Schreiber, S.L., Sammakia, T., Hulin, B., and Schulte, G. (1986) J. Am. Chem. Soc., 108, 2106–2108. Still, W.C. and Romero, A.G. (1986) J. Am. Chem. Soc., 108, 2105–2106. Vedejs, E. and Gapinski, D.M. (1983) J. Am. Chem. Soc., 105, 5058–5061. Paterson, I., Boddy, I., and Mason, I. (1987) Tetrahedron Lett., 28, 5205–5208. Nacro, K., Baltas, M., Zedde, C., Gorrichon, L., and Jaud, J. (1999) Tetrahedron, 55, 5129–5138. Russell, S.T., Robinson, J.A., and Williams, D.J. (1987) J. Chem. Soc., Chem. Commun., 351–352. Kumar, V.S., Aubele, D.L., and Floreancig, P.E. (2002) Org. Lett., 4, 2489–2492. Kumar, V.S., Wan, S., Aubele, D.L., and Floreancig, P.E. (2005) Tetrahedron: Asymmetry, 16, 3570–3578. Hayashi, N., Fujiwara, K., and Murai, A. (1997) Tetrahedron, 53, 12425–12468. Marshall, J.A. and Chobanian, H.R. (2003) Org. Lett., 5, 1931–1933. Marshall, J.A. and Hann, R.K. (2008) J. Org. Chem., 73, 6753– 6757. Marshall, J.A. and Mikowski, A.M. (2006) Org. Lett., 8, 4375–4378. Schultz, W.J., Etter, M.C., Pocius, A.V., and Smith, S. (1980) J. Am. Chem. Soc., 102, 7981–7982. Iimori, T., Still, W.C., Rheingold, A.L., and Staley, D.L. (1989) J. Am. Chem. Soc., 111, 3439–3440. Paterson, I. and Boddy, I. (1988) Tetrahedron Lett., 29, 5301–5304.
References 171. Paterson, I. and Craw, P.A. (1989) 172.
173.
174. 175. 176.
177.
178.
179.
180.
181. 182.
183. 184.
185. 186.
187.
188.
189.
Tetrahedron Lett., 30, 5799–5802. Evans, D.A., Ratz, A.M., Huff, B.E., and Sheppard, G.S. (1995) J. Am. Chem. Soc., 117, 3448–3467. Morimoto, Y., Iwai, T., and Kinoshita, T. (2000) J. Am. Chem. Soc., 122, 7124–7125. Xiong, Z. and Corey, E.J. (2000) J. Am. Chem. Soc., 122, 4831–4832. Xiong, Z. and Corey, E.J. (2000) J. Am. Chem. Soc., 122, 9328–9329. Morimoto, Y., Iwai, T., Nishikawa, Y., and Kinoshita, T. (2002) Tetrahedron Asymmetry, 13, 2641–2647. Morimoto, Y., Nishikawa, Y., and Takaishi, M. (2005) J. Am. Chem. Soc., 127, 5806–5807. Morimoto, Y., Yata, H., and Nishikawa, Y. (2007) Angew. Chem. Int. Ed., 46, 6481–6484. Morimoto, Y., Okita, T., and Kambara, H. (2009) Angew. Chem. Int. Ed., 48, 2538–2541. Morimoto, Y., Okita, T., Takaishi, M., and Tanaka, T. (2007) Angew. Chem. Int. Ed., 46, 1132–1135. Xiong, Z., Busch, R., and Corey, E.J. (2010) Org. Lett., 12, 1512–1514. Tong, R., Valentine, J.C., McDonald, F.E., Cao, R., Fang, X., and Hardcastle, K.I. (2007) J. Am. Chem. Soc., 129, 1050–1051. Tong, R. and McDonald, F.E. (2008) Angew. Chem. Int. Ed., 47, 4377–4379. Tong, R.B., Boone, M.A., and McDonald, F.E. (2009) J. Org. Chem., 74, 8407–8409. Corey, E.J. and Ha, D.C. (1988) Tetrahedron Lett., 29, 3171–3174. Mori, Y., Yaegashi, K., and Furukawa, H. (1997) Tetrahedron, 53, 12917–12932. Furuta, H., Takase, T., Hayashi, H., Noyori, R., and Mori, Y. (2003) Tetrahedron, 59, 9767–9777. Hudrlik, P.F., Arcoleo, J.P., Schwartz, R.H., Misra, R.N., and Rona, R.J. (1977) Tetrahedron Lett., 18, 591–594. Hudrlik, P.F., Hudrlik, A.M., and Kulkarni, A.K. (1982) J. Am. Chem. Soc., 104, 6809–6811.
190. Fristad, W.E., Bailey, T.R., Paquette,
191. 192.
193.
194.
195. 196.
197.
198.
199. 200. 201. 202. 203.
204.
205.
206.
207. 208.
L.A., Gleiter, R., and Boehm, M.C. (1979) J. Am. Chem. Soc., 101, 4420–4423. Tokiwano, T., Fujiwara, K., and Murai, A. (2000) Synlett, 335–338. McDonald, F.E., Bravo, F., Wang, X., Wei, X., Toganoh, M., Rodriguez, J.R., Do, B., Neiwert, W.A., and Hardcastle, K.I. (2002) J. Org. Chem., 67, 2515–2523. Bravo, F., McDonald, F.E., Neiwert, W.A., and Hardcastle, K.I. (2004) Org. Lett., 6, 4487–4489. Valentine, J.C., McDonald, F.E., Neiwert, W.A., and Hardcastle, K.I. (2005) J. Am. Chem. Soc., 127, 4586–4587. Heffron, T.P. and Jamison, T.F. (2006) Synlett, 2329–2333. Wan, S., Gunaydin, H., Houk, K.N., and Floreancig, P.E. (2007) J. Am. Chem. Soc., 129, 7915–7923. Tarselli, M.A., Zuccarello, J.L., Lee, S.J., and Gagn´e, M.R. (2009) Org. Lett., 11, 3490–3492. Simpson, G.L., Heffron, T.P., Merino, E., and Jamison, T.F. (2006) J. Am. Chem. Soc., 128, 1056–1057. Vilotijevic, I. and Jamison, T.F. (2007) Science, 317, 1189–1192. Byers, J.A. and Jamison, T.F. (2009) J. Am. Chem. Soc., 131, 6383–6385. Morten, C.J. and Jamison, T.F. (2009) J. Am. Chem. Soc., 131, 6678–6679. Morten, C.J. and Jamison, T.F. (2009) Tetrahedron, 65, 6648–6655. Morten, C.J., Byers, J.A., Van Dyke, A.R., Vilotijevic, I., and Jamison, T.F. (2009) Chem. Soc. Rev., 38, 3175–3192. Nicolaou, K.C., Reddy, K.R., Skokotas, G., Sato, F., and Xiao, X.Y. (1992) J. Am. Chem. Soc., 114, 7935–7936. Morimoto, M., Matsukura, H., and Nakata, T. (1996) Tetrahedron Lett., 37, 6365–6368. Mori, Y., Yaegashi, K., and Furukawa, H. (1997) J. Am. Chem. Soc., 119, 4557–4558. Kadota, I. and Yamamoto, Y. (1998) J. Org. Chem., 63, 6597–6606. Nicolaou, K.C., Duggan, M.E., and Hwang, C.K. (1989) J. Am. Chem. Soc., 111, 6676–6682.
589
590
15 Biomimetic Synthesis of Polyether Natural Products via Polyepoxide Opening 209. Nicolaou, K.C. (1996) Angew. Chem., 210.
211.
212. 213.
214.
215.
216.
Int. Ed. Engl., 35, 589–607. Matsuo, G., Kawamura, K., Hori, N., Matsukura, H., and Nakata, T. (2004) J. Am. Chem. Soc., 126, 14374–14376. Nicolaou, K.C., Yang, Z., Shi, G., Gunzner, J.L., Agrios, K.A., and Gartner, P. (1998) Nature, 392, 264–269. Fuwa, H., Sasaki, M., and Tachibana, K. (2001) Org. Lett., 3, 3549–3552. Fuwa, H., Kainuma, N., Tachibana, K., and Sasaki, M. (2002) J. Am. Chem. Soc., 124, 14983–14992. Kadota, I., Takamura, H., Sato, K., Ohno, A., Matsuda, K., Satake, M., and Yamamoto, Y. (2003) J. Am. Chem. Soc., 125, 11893–11899. Kadota, I., Takamura, H., Sato, K., Ohno, A., Matsuda, K., and Yamamoto, Y. (2003) J. Am. Chem. Soc., 125, 46–47. Furuta, H., Hasegawa, Y., and Mori, Y. (2009) Org. Lett., 11, 4382–4385.
217. Mori, Y., Yaegashi, K., and Furukawa,
218.
219.
220.
221.
222.
223.
224.
H. (1998) J. Org. Chem., 63, 6200–6209. Zakarian, A., Batch, A., and Holton, R.A. (2003) J. Am. Chem. Soc., 125, 7822–7824. Van Dyke, A.R. and Jamison, T.F. (2009) Angew. Chem. Int. Ed., 48, 4430–4432. Nicolaou, K.C., Edmonds, D.J., and Bulger, P.G. (2006) Angew. Chem. Int. Ed., 45, 7134–7186. Mente, N.R., Neighbors, J.D., and Wiemer, D.F. (2008) J. Org. Chem., 73, 7963–7970. Topczewski, J.J., Neighbors, J.D., and Wiemer, D.F. (2009) J. Org. Chem., 74, 6965–6972. Topczewski, J.J., Callahan, M.P., Neighbors, J.D., and Wiemer, D.F. (2009) J. Am. Chem. Soc., 131, 14630–14631. Sato, S., Nakada, M., and Shibasaki, M. (1996) Tetrahedron Lett., 37, 6141–6144.
591
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products James Burnley, Michael Ralph, Pallavi Sharma, and John E. Moses
16.1 Introduction
The term biomimetic derives from the Greek ‘‘bios,’’ meaning life, and mimetic, the adjective for ‘‘mimesis’’ or mimicry, and is the application of methods observed in Nature to the design and development of synthetic systems. This methodology transfer is desirable, not only because evolutionary pressures typically force natural systems to become highly optimized and efficient, but also because such systems boast an inherent elegance [1]. When applied to chemical syntheses, biomimetic approaches often facilitate rapid access to complex structures that may otherwise require inconceivable conventional synthetic pathways. Inspired by key transformations in the known (or proposed) biosynthetic pathway, biomimetic syntheses are perhaps most applicable to those systems that do not strictly require enzymatic catalysis or control. Instead, the structure itself is predisposed to the biomimetic chemical change, following a defined reaction pathway flowing energetically ‘‘downhill.’’ Heathcock neatly describes the biomimetic strategy: ‘‘The basic assumption of this approach is that Nature is the quintessential process development chemist. We think that the molecular frameworks of most natural products arise by intrinsically favorable chemical pathways – favorable enough that the skeleton could have arisen by a non-enzymatic reaction in the primitive organism. If a molecule produced in this purely chemical manner was beneficial to the organism, enzymes would have evolved to facilitate the production of this useful material’’ [2]. In recent times, the training of synthetic chemists has been dominated by the powerful retro-synthetic approach developed by Corey [3]. In this logical method, subskeletal functionalization generally occurs early on in the synthetic plan, and these ‘‘handles’’ are used as junctions for connecting and building up complex frameworks. There seems to be no limit in what can be achieved using this powerful system [4], and with the development of new reactions and catalysts, the future Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
592
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products
of synthesis is very exciting. However, in certain cases problems may arise in assembling together such highly functionalized sub-units into structures resembling the targeted natural products [5]. In contrast, Nature generally assembles the molecular scaffolding first with relatively few, but key, functional groups in place. Through the process of secondary metabolism, these core frameworks are then functionalized in a highly specific manner, often enzymatically [6]. Biomimetic chemistry offers alternative and complementary strategies toward total synthesis. They are often elegant and efficient processes, enlightening novel pathways to Nature’s most complex architectures. From the early days of Sir Robert Robinson [7], to modern day practitioners, the biomimetic chemist is driven not only by the desire to evolve new chemistries but also at developing a deeper understanding of the processes that actually make natural products, which themselves, due to evolutionary factors, have been pre-selected for their beneficial properties and elegance of synthetic approach. As such, biomimetic syntheses are relevant, and indeed have proven to be powerful in the syntheses of a huge array of complex secondary metabolites, including, steroids [8], terpenes [9], alkaloids [2, 10], and polyketides [11]. Of the known chemical transformations available to the chemist (and indeed Nature), it is fair to say that in recent times pericyclic reactions have taken center stage in biomimetic syntheses. In particular, cycloadditions have been a cornerstone, with the Diels–Alder reaction proving to be the most powerful and well-studied transformation [12]. This is no doubt due to the remarkable efficiency and complexity-generating properties of this amazing reaction that, in one step, creates two new σ -bonds and up-to four new stereocenters. However, when considering complexity-generating processes, electrocyclization reactions have also proven valuable tools in biomimetic chemistry, not least because they often participate in reaction cascades [13]. Such cascades may involve several electrocyclizations and, in combination with cycloadditions, for instance, they can deliver spectacular molecular complexity and diversity. Although electrocyclization reactions have played a central role in the biomimetic syntheses of a wide range of natural products, this chapter is restricted to polyketide-derived natural products, where electrocyclization reactions have been a key transformation and, in some cases, have provided essential validation of biosynthetic pathways. Rather than provide a comprehensive review of all polyketide-derived natural product syntheses involving electrocyclizations, we have chosen to limit the number of case studies to provide a more detailed analysis of the thought processes behind the biomimetic strategy. Electrocyclization reactions that have featured in total syntheses but have not been proposed, or are unlikely, to be biomimetic are not discussed.
16.2 Electrocyclic Reactions
Electrocyclic reactions are pericyclic processes that involve the cyclizations of conjugated polyenes. In such reactions, the π-bonds move in a cyclic manner and
16.3 Polyketides
a new σ -bond is formed at the terminal carbon in the polyene system, leading to a ring structure [14, 15]. Thermodynamically, electrocyclizations are driven by the formation of stronger σ -bonds at the expense of weaker π-bonds. In principle, there is no limit to the number of π-bonds that can be involved in a given electrocyclization, but systems containing 6π and 8π electrons leading to six- and eight-membered rings are most common, particularly in biomimetic syntheses. Electrocyclizations are not restricted to all-carbon systems, and both oxa- and aza-systems can also participate. The most striking and synthetically useful feature of electrocyclization reactions is that their stereochemical outcome can be accurately predicted and controlled, depending upon the substrate choice and reaction conditions. For example, (2E,4Z,6E)-octa-2,4,6-triene (1) yields cis-5,6-dimethyl-1,3-cylohexadiene (4) selectively under thermal conditions, whereas (2E,4Z,6Z)-octa-2,4,6-triene (3) yields trans-5,6-dimethyl-1,3-cylohexadiene (2) (Scheme 16.1). This trend is reversed under photochemical conditions, and is dependent upon whether the reaction proceeds via a conrotatory or disrotatory pathway, and can be readily predicted according to the rules formulated by Woodward and Hoffman [14]. Another key feature of electrocyclization is the requirement for the reacting termini to be brought into close proximity. As such, reactive polyenes must normally contain internal double bond(s) with (Z)-configuration. This is an essential consideration when planning a biomimetic synthesis involving an electrocyclization reaction. con hν
H3 C
1
CH3
2
6p e− system
dis ∆
dis ∆
con hν
CH CH3 3 4
Scheme 16.1
3
Stereocontrol in 6π electrocyclizations.
Biomimetic syntheses, by their very nature, require the practitioner to develop an understanding of or, perhaps more correctly, a rationale about the biosynthetic origins of the target structure. It is therefore pertinent to briefly discuss polyketide biosynthesis, which will place biomimetic-inspired syntheses into context, and also reveal a fundamental paradox about the involvement of enzyme catalysis in biomimetic pericyclic reactions. 16.3 Polyketides
The polyketide-derived family of natural products has, for many years, been a rich source of biologically and chemically intriguing molecules. These diverse
593
594
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products
secondary metabolites, which are often obtained from moulds and other soil and airborne microorganisms and microbes, have provided us with many pharmaceutically significant targets. For example, one of the longest standing antibiotics erythromycin [16] has been in use for over 50 years, whereas neocarzinostatin [17], of the enediyne family of natural products, demonstrates potent anticancer properties, and epitomizing the diversity of polyketides is maitotoxin [18], one of the largest, most complicated, and toxic compounds known to man (Figure 16.1). The struggle by scientists to manufacture and uncover the biological mechanisms for the production of these compounds has provided great insights into cell biology and synthetic methodology [19]. At first glance, structures so dissimilar seem unlikely to come from a common biosynthetic pathway. For example, comparing erythromycins’ highly saturated macrolactone skeleton with doxorubicin’s polyaromatic center piece exemplifies the variety of structures that originate via this pathway.
16.4 Fatty Acid Biosynthesis
By the 1950s it was realized that the polyketide biosynthetic pathway was similar to the biosynthesis of fatty acids. Through a series of radiolabeling experiments it was clear that iterative Claisen condensations followed by a series of reductive transformations were responsible for the formation of long chain fatty acids [19]. Polyketide biosynthesis takes place in the polyketide synthase (PKS), and for each turn of the cycle the fatty acid chain length is extended by two carbons. Throughout the cycle thionyl bonds are used to attach the growing fatty acid, incoming monomers and intermediates to the enzyme complexes. The fatty acid chain is attached initially to a ketosynthase (KS), and the process is initiated when an unbound KS enzyme attacks acetyl co-enzyme A (AcCoA). Once initiated, a free acyl carrier protein (ACP) acquires a malonyl group from malonyl-co-enzyme A (MCAT) obtained from the organism’s primary metabolism, a process mediated by acyl transferase (AT). Claisen condensation takes place with the loss of CO2 , and by exchanging the growing fatty acid onto the ACP. The β-ketothioester is then carried through a reductive pathway until the β-ketone is fully reduced to the saturated methylene. The cycle is continued until the fatty acid chain has reached a specific length, then the thionyl esterase (TE) ejects the molecule as a long-chain fatty acid (Scheme 16.2). After condensation, further reductive manipulations may take place as the ketide is passed around the enzyme complex or megasynthase via the ACP. First the keto-reductase (KR) reduces the β-ketone to an alcohol, which then eliminates water as a result of the dehydratase (DH) to give an olefin, and the enoyl reductase (ER) then reduces it further to the saturated thioester. Rather than rigidly following the full reductive pathway to give saturated products polyketide synthesis can omit certain transformations to leave a more oxidized
Figure 16.1
OH
OH
O O S OH
O
O O
OH
HO
H O
O H H OH
OH
OH
OH
O
H
H
H
O
H
O
OH
N OH HO O O O
Erythromycin
O
OH
H
O
OH
O
H
O
H
H
O H H
O
OH
H
O
H
O OH
O O
H O
O
O
O
O
H
H
O
H
H
O
H
H
O
H O
OH
H OH OH
H
OH H H O
H O
OH
O H
OH
O
OH H
N OH H Neocarzinostatin
O H H
O
O NaO S O OH H H OH O O OH
Maitotoxin
H
H
O
H OH
Examples of structurally diverse complex polyketides.
ONa
O
OH
HO
O
O
O
H
H O O
OH OH
H
H O
OH H OH
OH
H O
H O
OH
OH
H H O H OH
16.4 Fatty Acid Biosynthesis 595
596
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products O O
n
ACP
KS
S
SH O
n
OH HO
Fatty Acid
KS
S
CoA
MCAT
O S
n
O
ACP
HS
SH
CoA O n
Reductive Pathway O KS
SH
n
O
O
CO2 S
ACP
ACP
S
KS
S O
O Claisen
β-ketothioester
Scheme 16.2
General polyketide biosynthesis.
skeleton (Scheme 16.3). For example, the cycle can be terminated after the ketoreductase step giving a β-hydroxy ketone. The TE has a more important role, rather than just hydrolyzing the thionyl to give a carboxylate, since it can facilitate lactonization to give macrolides. Further to this, polyketide synthesis has access to a greater variety of starting materials (rather than solely MCAT) and aromatizing enzymes that produce polyaromatic structures. Often the initial product of the megasynthase is not the natural product, just the basic compound skeleton; further functionalization can be achieved through oxidation and glycosylation. There are three classes of PKSs (the collective term for the group of enzymes), aptly named I, II, and III. Type I PKS is the most complicated of the PKSs and can be split further into two subcategories. Type I modular PKS utilizes a series of individual catalytic modules to provide each transformation. These modules will add one C2 unit of varying oxidation state and consist of individual catalytic domains responsible for condensation and reduction, often leading to reduced macrolide-type products. Type I iterative PKS exploits a single multi-enzyme to carry out the iterative steps and commonly gives reduced aromatic polyketides. This iterative complex consists of two KS, α and β, and an ACP, which associate to give the active complex. A defined number of catalytic cycles produce a poly β-ketone chain that is then transformed to give aromatic products. Most relevant to biomimetic syntheses employing electrocyclization reactions are the products resulting from the action of the DH, which leads to the formation of polyenes (step 3). The DH that catalyzes the transformation of the β-hydroxythioester into the (E)-α,β-unsaturated species consists of histidine and asparagine residues in the active site. The histidine acts as a base removing the enolic proton and the asparagine provides a proton to release water and the unsaturated polyketide. It is these polyenes that will form the basis of discussion for
16.5 Biomimetic Analysis O ACP/ CoA Step 1: Claisen condensation of a monomer
O
S
O OH
R
S
KS
R′ KS
Step 1
from the primary metabolism on to the growing O
fatty acid chain. ACP/ CoA Step 2: Reduction of the β-ketothioester by
597
O
O
O R
S R′
HO n
R R′
Type I/III PKS R′ = Me, H, Et etc
KR
Step 2
the ketonereductase to the corresponding β-hydroxythioester.
O ACP
OH
S
R R′
Step 3: Dehydration by the dehydratase
O HO
OH
n
R
Type I PKS R′ = Me, H, Et etc
R′
DH
Step 3
yielding the unsaturated thioester. O ACP
O R
S
HO
n
R′ Step 4: The reductase reduces the olefinic
ER
Step 4
bond to give the fully saturated fatty acid
O
O
backbone. ACP
R R′
HO
S R′
n
R R′
Fatty Acid R′ = H
Scheme 16.3
Stepwise illustration of polyketide biosynthesis.
the rest of this chapter, focusing on the biomimetic syntheses on complex natural products. One major concern of natural products chemistry is the uncertainty about whether the isolated structures were Nature’s initial intention! In other words, the true natural product produced by the given organism may not always be ‘‘static,’’ and the isolated compounds may be the product of a post-biosynthetic opportunity. As such, the complexity of the given structure has simply been endowed upon the system, as a consequence of its original function. These structures are still worthy targets for synthesis, since they are often biologically active and structurally fascinating. Since pericyclic reactions frequently require little activation, it is compelling to propose that these transformations are ideal candidates for post-biosynthetic activity and, as such, would not have pressurized the producing organism to develop enzymatic machinery. This may explain the rarity, or indeed absence, of the elusive ‘‘Diels–Alderases’’ and ‘‘electrocyclases,’’ and may also explain why in many cases the isolated natural products, proposed to arise through electrocyclizations, are obtained in racemic form.
16.5 Biomimetic Analysis
Through interrogation of biosynthetic pathways and by considering each individual step, one may identify a key transformation that could form the basis of a biomimetic
598
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products Table 16.1
Polyketide biosynthesis broken down into sub-stages.
Stage in polyketide biosynthesis
Description
Comment
Stage 1
Claisen condensation/formation of the carbon chain Post-condensation transformation, for example, reduction/dehydration Complexity generation Aromatization Cyclization Electrocyclization
Under enzymatic control (ACP/KS) Under enzymatic control (KR/DH/ER)
Stage 2 Stage 3 (or 4)
Stage 4 (or 3)
Oxidation
Enzymatic control Enzymatic control Unknown, but most likely non-enzymatic control Both enzymatic and non-enzymatic control
synthesis. Table 16.1 summarizes an approach to aid in the planning and execution of a biomimetic synthesis, which involves breaking down polyketide-derived biosynthesis mechanism into key stages. Stages 1 and 2 are classified as lower order, and are common to all polyketide derived natural products. They involve stitching together the key carbon segments, which eventually lead to fragments ripe for conversion into more complicated higher order structures through complexity generating reactions. Stages 3 or 4 are later stages and often provide the inspiration for biomimetic syntheses. Although planning a synthesis based upon one critical step may be deemed high risk, the potential rewards of developing novel transformations and validating biosynthetic hypotheses are worthwhile. To demonstrate the power of the biomimetic approach, a selection of case studies now follows. Unfortunately, and necessarily, the work of many groups around the world who have made significant contributions to this field will be omitted. We can only apologize for this and encourage the reader to research the primary literature in detail for other examples. 16.6 6π Electrocyclizations
Thermal 6π electrocyclizations of conjugated trienes are probably the most common of all electrocyclizations employed in the biomimetic synthesis of polyketide-derived natural products. Such reactions lead to conjugated cyclohexadiene units (Scheme 16.1) and, although thermodynamically favorable, often require activation by relatively high temperatures. The cyclohexadiene products can also participate in further tandem reactions such as Diels–Alder cycloadditions, and such cascades can lead to extremely complex architectures.
16.6 6π Electrocyclizations
16.6.1 Tridachiahydropyrones
However, we begin by examining a family of natural products whose synthesis relied upon a photochemical 6π electrocyclization. Although very powerful, photochemical electrocyclizations are rarely utilized in total synthesis compared to their thermal counterparts. In biomimetic systems, however, one may intuitively realize that Nature would in fact take full advantage of the essentially endless supply of photons at her disposal, and there is perhaps no better example of the power of photochemical electrocyclizations in polyketide synthesis than in the tridachiahydropyrone family. Adopting a biomimetic strategy, researchers in our own laboratories were able to provide convincing evidence to corroborate the biosyntheses of oxytridachiahydropyrone (5) [formerly tridachiahydropyrone B (5 ) and C (5 )], and tridachiahydropyrone (6). The successful approach to these complex metabolites also led to their structural reassignment [20]. (−)-Tridachiahydropyrone (6) (corrected structure shown in Scheme 16.4) is a marine-derived natural product isolated in 1996 by Cimino et al. from a sacoglossan mollusc (Tridachia crispata) [21], whereas tridachiahydropyrone B (5 ) and C (5 ) [later reassigned as oxytridachiahydropyrone (5)] were isolated a few years later from P. ocellatus (order Sacoglossa, family Elysioidea) along with several other propionate-derived metabolites [22]. Such opisthobranch molluscs characteristically produce polyketide (polypropionate)-derived metabolites, which commonly incorporate a γ -pyrone unit. Metabolites 5 and 6 clearly share a common core framework of differing oxidation state, providing a strong indication about their biosynthetic relationship. Scheme 16.4 illustrates the biomimetic analysis of tridachiahydropyrones: retro-oxidation of oxytridachiahydropyrone (5) (stage 4) leads to tridachiahydropyrone (6). As previously mentioned, late-stage oxidation is common in biosynthesis, and may or may not be governed by enzymes. In the present case, it was proposed that the oxidation was photochemically driven, and therefore highly unlikely to be enzyme-catalyzed. The inspiration behind this conclusion came from the studies of Ireland and Scheuer, who proposed that metabolites of this family are synthesized by the shell-less molluscs to act as natural sunscreens [23]. Understanding the larger biological system helped unravel the complexities behind the biogenesis, making it truly biomimetic. It was proposed that tridachiahydropyrone (6) itself could be derived from the (2Z,4E,6E)-triene 7, through a photochemical 6π conrotatory electrocyclization (stage 3), the key biomimetic complexity generating step. The lower order biosynthetic stages, 2 and 1, lead to polyene 8, derived from the building block propionate 9, which, in Nature, would be stitched together using the PKS machinery. In the forward synthesis, triene 7 became the primary synthetic target, with the key question being, could 7 be converted into tridachiahydropyrone (6) through a photochemical 6π electrocyclization? The biomimetic rationale did not account for the fact that (−)-tridachiahydropyrone (6) was isolated as a single enantiomer, and it was not certain whether
599
O
OMe
hν
O2
Scheme 16.4
O
O
OMe
O
O
OMe
O O O
O
OMe
(7)
O
O
OMe
Biomimetic analysis of the tridachiahydropyrones.
O O
(8)
O
O
O
O
OMe
O
OH
Enzymatic control
OH
(9)
O
Stage 1: Poly-condensation
tridachiahydropyrone C (5′′) (proposed structure)
Enzymatic control
Stage 2: Post-condensation transformation
tridachiahydropyrone B (5′) (proposed structure)
hν
6π e− con
tridachiahydropyrone (6) (revised structure)
tridachiahydropyrone (6′) (proposed structure)
oxytridachiahydropyrone (5) (revised structure)
O O
1
Non-enzymatic / photochemical
Non-enzymatic / photochemical
O
Stage 3: Electrocyclization
Stage 4: Oxidation
Biomimetic Analysis
600
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products
16.6 6π Electrocyclizations
the biosynthesis was enzymatically controlled. To test the photochemical electrocyclization hypothesis, the researchers first had to synthesize the proposed (2Z,4E,6E)-polyene precursor 7. In biosynthetic terms, 7 would be a product of stages 1 and 2, which is driven by the powerful PKS machinery. Nature is the master of such chemistry, and although a similar iterative approach could in principle be developed in the laboratory, the efficiency and overall yield would most likely be compromised. The research team opted for a convergent strategy employing a Suzuki cross coupling reaction between the boronate ester 10 and the γ -pyrone 11, which gave the biomimetic precursor 7 in good yield (Scheme 16.5). To plan the photochemical reaction, the research team considered the marine habitat of the producing molluscs to optimize the chance of success. The practical lower limit for UV radiation in seawater is approximately 290 nm, and it was reasoned that placing the substrate in a glass vial in direct sunlight would be a good model. After exposure of a 0.018 mmol sample of 7 in methanol on a window ledge for three days, the researchers obtained target 6 in 29% yield (along with 57% starting material), and were able to corroborate the proposed reassigned structure using X-ray crystallography. When the corresponding all-(E)-polyene 13 was irradiated, a tandem reaction sequence also led to the target 6. Monitoring of the reaction by 1 H NMR spectroscopy, specific (E)–(Z) isomerization about the C2=C3 double bond to give 7 was observed, which was followed by 6π electrocyclization to give 6. It was concluded that the all-(E)-polyene 13 was perhaps the true biosynthetic precursor. Interestingly, an unexpected product 14, later named phototridachiahydropyrone, was also co-isolated from the reaction mixture, whose origin was thought to arise from 6 through a photochemically allowed suprafacial [1,3]-sigmatropic shift. In fact, prolonged exposure of 6 led to the complete conversion into 14, thus indicating that this photochemical end point of the tandem reaction sequence may also be a natural product yet to be isolated, but may also depend upon the natural habitat [24]. The laboratory synthesis of compounds that later turn out to be natural products is not an uncommon event in biomimetic chemistry, and examples will be discussed below. The final step in the study was to determine if 6 could be converted into oxytridachiahydropyrone (5) [formerly tridachiahydropyrone B (5 ) and C (5 ]. Both compounds were reported as an inseparable mixture of isomers differing in geometry about the C10=C11 double bond, in a ratio of 4 : 5, as determined by 1 H NMR spectroscopic analysis. The relative, or absolute, configuration of the chiral centers was not elucidated [22]. It was reasoned that tridachiahydropyrone B (5 ) and C (5 ) were derived biosynthetically from 6, via a photochemical [4+2] cycloaddition with singlet oxygen. A mixture of diastereoisomeric compounds would be expected to arise from a facially selective addition of singlet oxygen to either the concave or convex face of tridachiahydropyrone (6). Tridachiahydropyrone C (5 ) would arise either through isomerization of the C10=C11 double bond of tridachiahydropyrone B (5 ) or its precursor. It was known that α-methoxy-γ -pyrones function as photosensitizers, and perhaps the fused pyrone ring of tridachiahydropyrone would also function analogously. When a solution of 6 was irradiated with a UV
601
Scheme 16.5
10
[20]
12[32]
O
O
THF, 80 °C
Pd(PPh3)4, KOH,
O
THF, 80 °C
OMe
OMe
3
13
2
hν MeOH
7
O
O
O
O
OMe
OMe
O
O
OMe
hν
1
O2
hν CHCl3
O
OMe
O
O
OMe
oxytridachiahydropyrone (5) (revised structure)
O O
O
phototridachiahydropyrone (14)
tridachiahydropyrone (6) (revised structure)
6π e− con
hν MeOH
Biomimetic synthesis of the tridachiahydropyrones by a key photochemical 6π electrocyclization.
Br
O B
O 11[32] Pd(PPh3)4, KOH,
Br
O
602
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products
16.6 6π Electrocyclizations
source under a continuous flow of molecular oxygen for 4 h a new product was obtained, whose spectral data agreed with that reported for the non-separable mixture of tridachiahydropyrone B (5 ) and C (5 ), and in identical ratio (4 : 5). However, further analysis revealed that the mixture was in fact two rotameric forms of the related diastereoisomer, which was renamed oxytridachiahydropyrone (5) (Scheme 16.6) [20b]. Compound 5 was the product of exclusive endo addition of oxygen via the less hindered face of the diene. The biomimetic studies of the tridachiahydropyrones led not only to the successful syntheses and structural correction of two complex natural products, it also provided great insight into the occurrence of these interesting compounds. The photochemical transformations delivered unusually high levels of control. The specific (E)–(Z) isomerization of the C2=C3 double bond of polyene 13, followed by a 6π conrotatory electrocyclization is a powerful and remarkable tandem sequence, yet the ease with which the transformation occurs clearly demonstrates that Natures pathways can be replicated with great success. Rotameric H H 7
O O
H
11
v's
O
Exo-adduct
11
O
O O O
O MeO
H
7
O O
O MeO
= observed nOe
Endo-adduct
O
OMe
oxytridachiahydropyrone (5)
Scheme 16.6 Biomimetic synthesis of oxytridachiahydropyrone (5) via photochemical reaction with singlet oxygen.
16.6.2 Tridachione Family
Several other cyclohexadiene natural products isolated from sacoglossan molluscs can be envisaged as being derived from conjugated polyene systems through 6π electrocyclization [25, 26]. For example, 9,10-deoxytridachione (15), tridachiapyrone I (16), and tridachiapyrone A (17) can be readily traced back to a tetraene of type 18, whereas deoxyisotridachione (19) can be thought of as arising through a photochemical 6π conrotatory electrocyclization of the tetraene 20 (Scheme 16.7) [27–29]. In some cases, the cyclohexadiene moiety of the molluscan natural products acts as staging points for further transformations. For example, pioneering photochemical studies by Ireland and Scheuer demonstrated that 9,10-deoxytridachione (15) could be converted in vitro and in vivo into the complex isomer photodeoxytridachione (22) [25]. The transformation was presumed to occur through a [σ 2a + π 2a ] pericyclic mechanism since the non-enzymatic reaction led exclusively to enantiomerically pure products, indicating that an open chain polyene was not involved. Furthermore, epoxidation of 15 gave tridachione (21) [28], which was the parent compound of the series (Scheme 16.8).
603
OMe
O
O
OMe
O
OMe
Scheme 16.7
O
∆
6π e− dis
E E Z
O
18
E
OMe
O
O
O
hν
6π e− con
OMe
O OMe
O
deoxyisotridachione (19)
Electrocyclic pathways to cyclohexadiene natural products.
tridachiapyrone A (17)
O
tridachiapyrone I (16)
O
O
9, 10-deoxytridachione (15)
O
∆
6π e− dis
E
E
Pyr
Z
20
Z
O
O
OMe
604
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products
16.6 6π Electrocyclizations
605
OMe
O
O
O
[O]
O
O OMe
tridachione (21)
Scheme 16.8
O
hν [σ 2a + π 2a]
H
O
OMe 9, 10-deoxytridachione (15)
photodeoxytridachione (22)
Chemical transformations of 9,10-deoxytridachione (15).
The biomimetic synthesis of 9,10-deoxytridachione (15) was independently achieved by Baldwin et al. [29] and Trauner et al. [30]. Both research groups utilized a convergent cross-coupling approach to assemble the (2E,4Z,6E,8E)-polyene-pyrone 18, which was perfectly set up to undergo a 6π disrotatory electrocyclization to give the target cyclohexadiene 15. Baldwin et al. opted for a Suzuki coupling to unite the boronate 23 to the vinyl iodide 24, which gave the desired polyene 18 in 65% yield. Subsequent heating of 18 in benzene initiated thermal 6π electrocyclization to afford 15 in 31%, along with the unexpected bicyclo[4.2.0]octadiene 27 product, which was itself later isolated as a natural product and named ocellapyrone A (Scheme 16.9a). The formation of 27 occurs through an elegant cascade of electrocyclizations and will be discussed in more detail below. In Trauner’s synthesis, a modified Stille coupling [31] was employed to unite the (Z)-vinyl stannane 25 to the (E)-iodo alkene 26, to yield the polyene 18, which then underwent electrocyclization under microwave conditions (150 ◦ C, toluene) to give 15 and 27 in 32% and 38% yield respectively (Scheme 16.9b). In a broader study of sacoglossan mollusc-derived polyketides, Moses et al. [32] proposed a general biosynthetic scheme to explain the formation of a range of natural products, including 9,10-deoxytridachione (15), from the all-(E)-tetraene pyrone 28. The scheme could be expanded to account for all possible diastereoisomers, but for presentation purposes only a selection of compounds are illustrated. Through a sequence of selective (E)–(Z) double bond isomerizations and electrocyclizations, it was realized that several core structures could in principle be obtained from 28. Scheme 16.10 illustrates several possible pathways, which involve chemistry at stage 3 and 4 of polyketide biosynthesis. Clearly, 9,10-deoxytridachione (15) plays an important role in this general biosynthetic scheme. To test the hypothesis, the researchers again used a Suzuki coupling to synthesize the all (E)-polyene 28. Several isomerization–electrocyclization conditions were examined, each of which lead to interesting outcomes. Studies with 28 carried out in the dark under palladium-catalyzed isomerization conditions lead to a complex mixture of products. Purification afforded a reasonable quantity of the racemic bicyclo[4.2.0]octadiene 34, corresponding to the endo-isomer of ocellapyrone A (27) (Scheme 16.11). The formation of the bicyclo[4.2.0]core will be discussed later, since its origin also lies in electrocyclic chemistry. Interestingly, compound 34 was itself incorrectly proposed as the original structure of ocellapyrone A (27) [33], before revision and confirmation of the correct structure by synthesis. Notably,
O
Scheme 16.9
25
+
I
I 26
O
O
24
O
O
OMe
OMe
∆
(b) 70%, 1.0:1.2 :: 15:27
18
O
OMe
45 °C, 89%
E
E O
(a) 46%, 2.0:1.0 :: 15:27
E
Z
DMF,1.5 h,
Pd(PPh3)4, CsF,CuI
THF, KOH, rt, 2h, 65%
Pd(PPh3)4
Biomimetic syntheses of 9,10-deoxytridachione (15).
SnMe3 +
(b) Trauner's Approach
23
O B
(a) Baldwin's Approach
OMe
O OMe
O
ocellapyrone A (27)
H
+
9, 10-deoxytridachione (15)
O
O
606
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products
E
O
H O
E
hν or ∆
OMe
∆
∆ 8π con / 6π dis
OMe
O
O
O
Z
E
Z 29
E
18
Z
O
E
O
O
Z
20
Z
E
E
E
O
OMe
OMe
O
hν
hν 6π con
[π4a + π2s]
∆ 8π con / 6π dis
[π4s + π2a]
hν
hν 6π con
∆ [π4a + π2a]
∆ 6π dis
O
Proposed biosynthetic pathways from polyene 28.
ocellapyrone B (32)
O
28
E
Scheme 16.10
E
Selective isomerization
hν
E
OMe
OMe [O]
H
O
O OMe
O
ocellapyrone A (27)
H
photodeoxytridachione (22)
O
OMe
[ σ2a + π 2a]
hν
[O]
[O]
9,10-deoxytridachione (15)
O
O
[O]
O
O
O OMe
OMe
H
O
OMe
H
O
[O]
O
OMe
tridachiapyrone F (31)
O
tridachiapyrone E (30)
O
O tridachiapyrone I (16)
O
O
tridachione (21)
O
O
16.6 6π Electrocyclizations 607
608
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products
34 could be a precursor in the biosynthesis of ocellapyrone B (32). However, this intermediate has not yet been isolated from a natural source. Thermal cyclization of 28 was carried out in xylene at 150 ◦ C in the dark for 1.5 h. These reaction conditions afforded multiple products, including ocellapyrone A (27), along with the related diastereoisomer 33 and cyclohexadiene 9,10-deoxytridachione (15) (Scheme 16.11). Photolysis of 28 again afforded a complex mixture of crude products, which could be separated to give photodeoxytridachione (22) and deoxyisotridachione (19). Presumably, 22 was formed through 15 in a manner consistent with the experiments of Ireland and Scheuer, although the latter was not isolated in the photochemical experiments. Alternative non-electrocyclic reaction mechanisms have been considered, including a photochemical Diels–Alder [34] and a highly specific diradical process [35]. These studies clearly demonstrate the multiple pathways that are possible from one common intermediate, but also emphasize the power of employing biomimetic strategies. Although the yields were not optimized, the work addressed the bigger question about the biosynthetic origins of these complex natural products. It is highly likely that the pathways unraveled by the researchers are similar to those that operate in Nature. However, the racemic syntheses again raise fundamental questions about the origins of the enantiopurities of the isolated compounds. 16.6.3 Pseudorubrenoic Acid A
The possible involvement of 6π electrocyclic ring closures, mediated by electrocyclase enzymes of polyunsaturated acyclic polyketide intermediates, has been raised by Rickards et al. in the biosynthesis of pseudorubrenoic acid A (35). This aromatic fatty acid is an antimicrobial carboxylic acid isolated from the soil bacterium Pseudomonas fluorescens [36]. Rickards et al. noticed that the lack of oxygen functionality may indicate that the compound is constructed through a pathway distinct from the common biosynthesis of aromatic polyketides. The biosynthetic pathway was proposed to involve the formation of cyclohexadiene 36 through 6π electrocyclization of (7E,9Z,11E,13Z)-tetraene 37, followed by oxidation to 35 (Scheme 16.12). To evaluate their biomimetic hypothesis, the tetraene 43 was synthesized from phenyl sulfide 38 and (Z)-1-bromopent-2-ene (39), which under basic conditions afforded the enyne 40. Oxidation of 40 to the corresponding sulfoxide and elimination next gave a mixture of (E/Z) diastereoisomers of 41. Palladium-catalyzed cross coupling of the sp and sp2 centers present in 41 and 42, followed by a reduction gave a mixture of the tetraenes 43a–d. Upon heating, this isomeric mixture of tetraenes underwent thermally induced transformation to give cyclohexadiene 44 as the major product, which was followed by subsequent aromatization to the o-disubstituted benzene 45. Rickards’ synthesis provided support for his electrocyclization hypothesis, and serves as a model for the possible
O
OMe
O
O
OMe
Scheme 16.11
O
O
OMe
deoxytridachione (15), 4%
hν Cyclohexane 22 h
Biomimetic conversion of 28.
photodeoxytridachione (22), 9%
H
+
deoxyisotridachione (19), 4%
O
+
O
O
OMe
OMe
ocellapyrone A (27), 10%
H
Dark Xylenes, 150 °C, 1.5 h
28
O
O
+
O
33, 9%
H
Dark PdCl2(MeCN)2, DMF, 50 °C, 2 h
O
OMe
O
34, 6%
H
O
OMe
16.6 6π Electrocyclizations 609
610
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products
CO2H pseudorubrenoic acid A (35)
Scheme 16.12
CO2H 36
CO2H 37
Retrosynthetic biomimetic analysis of pseudorubrenoic acid A (35).
involvement of an electrocyclase in the biosynthesis of pseudorubrenoic acid A (35) (Scheme 16.13) [36]. 16.6.4 Torreyanic Acid
As briefly mentioned earlier, electrocyclization reactions can also occur with hetero-carbon polyene systems, and it is not surprising that this reactivity has been exploited in biomimetic syntheses. In particular, oxa-6π electrocyclic ring closures are common, with such reactions being essentially thermoneutral processes with low activation barriers. The relative equilibrium concentration of the dienal/dienone 46 and the (2H)-pyran 47 (Scheme 16.14) depends upon the electronic properties of the system [37], and despite such reactions being common in Nature, (2H)-pyrans that are not trapped by further transformations are rare motifs in natural products. Among the wide variety of polyketide-based natural products discovered to date, the synthesis of the epoxyquinone (+)-torreyanic acid (51) provides an excellent example of a biomimetic oxa-6π electrocyclization. Isolated from the endophytic fungus Pestalotiopsis microspora and characterized by Clardy and coworkers in 1996 [38], the complex acid 51 displayed potent biological activity. In the original report of its isolation and structural characterization a biosynthetic scheme toward torreyanic acid (51) was proposed, involving Diels–Alder dimerization of 2H-pyran monomers epimeric at C9 (C9 ) (50/50 ). In this endo-selective [4+2] cycloaddition the two pentyl side chains are oriented away from one another in the transition state. The isolation of the monomeric epoxyquinol ambuic acid (48) [39] from P. microspora further supports the proposed biosynthesis of torreyanic acid via oxidative dimerization of monomeric intermediates (Scheme 16.15). The first total synthesis of torreyanic acid in racemic form was successfully executed by Porco et al. [40]. This research group later used a complementary strategy to complete the first asymmetric synthesis of 51, following the same general biomimetic route [41]. Beginning with the 1,3-dioxane derivative 52, the homoallylic alcohol 53 was synthesized in several steps. An interesting enantioselective epoxidation of 53, developed by Porco, was achieved using a tartrate-mediated nucleophilic method to give 54 in excellent yield and ee (91% yield, 91% ee). Dess–Martin periodinane (DMP) oxidation of the homoallylic alcohol followed by a two-carbon homologation next afforded 55. A cross-coupling reaction then realized 56, which was subsequently deprotected to afford the biomimetic precursor 57. Finally, treating 57 with DMP initiated the tandem oxidation/6π-electrocyclization/Diels–Alder dimerization to give (+)-torreyanic acid tert-butyl ester (58), which was hydrolyzed
39
38
SPh
Scheme 16.13
100%
OTBDMS
MnO2, xylene,
42
5
41, 35%
TBDMSO
TMS
120 °C
44, 11%
2. Base, xylene
− 6π e con
40, 57%
1. mCPBA, NaHCO3 DCM
toluene, 150 °C, 18 h
TMS
PhS
Biomimetic synthesis of a pseudorubrenoic acid A model system (45).
84%
OTBDMS
HMPA, THF
LDA, 0 °C
43a (7E, 9Z, 11E, 13Z) 43b (7E, 9Z, 11Z, 13Z) 43c (7Z, 9Z, 11E, 13Z) 43d (7Z, 9Z, 11Z, 13Z)
Br
TMS
I
45, 100%
OTBDMS
2. Zn (Cu/Ag), MeOH-Water
MeCN, 50% aq. NaOH
1. Pd(Ph3)4, CuI, TEBACl
16.6 6π Electrocyclizations 611
612
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products Scheme 16.14
X
X O
6π eX = H, Alkyl, Aryl
46
Oxa-6π electrocyclizations.
O
47
under acidic conditions to afford free (+)-torreyanic acid (51). Analytical techniques confirmed the absolute configuration as that of the natural (+)-torreyanic acid (51) (Scheme 16.16). Interestingly, the biomimetic complexity-generating cascade demonstrated by Porco occurred without the need for catalysis. The optical activity associated with 51 was carried through from the monomeric precursor ambuic acid (48), which is most likely to be the case in the natural system. Mehta et al. have reported a related biomimetic approach to (±)-torreyanic acid, starting from 2-allyl-p-benzoquinone to access the tert-butyl ester of ambuic acid 48, which underwent the analogous biomimetic tandem oxidation/6π electrocyclization/Diels–Alder dimerization sequence [42]. A similar dimerization approach accounts for the formation of the polyketidederived metabolites epoxyquinol A (63), B (62), and C (64), isolated from a fungi by Osada et al. [43]. A biosynthetic proposal for 62–64 from the dienal 59 was proposed by Osada, which centered upon a tandem oxa-6π electrocyclization to give the corresponding C5 epimers 60 and 61, followed by a [4+2] cycloaddition/dimerization sequence as depicted in Scheme 16.17. Interestingly, a formal [4+4] cycloaddition of 60 produced the corresponding homodimer epoxytwinol (65) (Scheme 16.17) [44]. Several groups have reported asymmetric biomimetic syntheses of the epoxyquinols, including, Hayashi [45], Porco [46], Kuwahara [47], and Mehta [48], with each team developing unique routes to the key biomimetic dienal precursor 59.
16.7 8π Systems and the Black 8π–6π Electrocyclic Cascade 16.7.1 Endiandric Acids
The 8π–6π electrocyclic cascade (sometimes referred to as Black’s cascade) is probably the most elegant and classical display of the power of electrocyclization reactions in Nature. This tandem reaction sequence, which involves a cascade of thermally induced 8π, then 6π electrocyclizations, has proven its worth in biomimetic synthesis. The most well-known example of this amazing transformation was demonstrated by K.C. Nicolaou et al. in their elegant syntheses of endiandric acids A–G (66–72) (Figure 16.2) [49, 50]. Detailed accounts of this landmark study already exist [4a], and although we will refer to aspects of this pioneering work we will also focus on other compounds where this cascade has been a key transformation.
O
O
Bun
O
H H
oxidation
H
O
nBu
O
O intermolecular Diels-Alder
C5H11
O
49
O
O O
COOH
6π e
COOH
O
HOOC
O
O
O
O
9′
O
C5H11
O
50/50′
electrocyclization
−
Lee et al. biosynthetic hypothesis for the formation of (+)-torreyanic acid (51).
torreyanic acid (51)
O
O
HO2C
Scheme 16.15
HO2C
O
O
Ambuic acid (48) COOH
C5H11
HO
OH
9
O
O O
50/50′ COOH
O
O
16.7 8π Systems and the Black 8π–6π Electrocyclic Cascade 613
52
OMe
OMe
Scheme 16.16
C5H11
HO
O
O
O
O O
O
O L-DIPT, Toluene, −40 °C, 50 h
TFA, DCM 0 °C, 2h
ROOC O O
O
H
O
H
H
O
O
O
O
OH
R = H, torreyanic acid (51), 100%
COOtBu
O
O
56, 94%
O
Bu O
n
O
O
TBDPSO C5H11
O
54, 91%, 91% ee
Br
TBDPSO
Bun R = tBu, 58, 80%
O
ROOC
Pd(PPh3)4, toluene, 110 °C, 2 h
DCM, 1.5 h
COOtBu
OH
Ph3COOH NaHMDS
(E)-tributyl-1-heptenylstannane
53
O
O
Dess-Martin periodinane
COOtBu
Br
TBDPSO
55, 94%
O
57, 71%
Br
TBDPSO
Steps[41]
Porco’s biomimetic synthesis of (+)-torreyanic acid (51).
2. 48% aq. HF MeCN, 15 min
1. TBAF / AcOH THF, 20 h
DCM, −78 to −5 °C, 4 h
2. PPh3 = C(CH3)COO Bu,
t
1. Dess-Martin periodinane, DCM, 35 min
O
O
614
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products
16.7 8π Systems and the Black 8π–6π Electrocyclic Cascade
615
O HO
OH
O homodimerization
O HO
O 60
6π
O
O epoxyquinol B, (62) O
OH O
O OH
heterodimerization
+
O
H O epoxyquinol A, (63)
O O
O
O
[4 + 2]
OH 59
O
H
OH O
O
O
[4 + 2]
O
O
O 61
HO O
OH
homodimerization
O
O
[4 + 2]
O
H
O epoxyquinol C, (64) OH
O O
O
homodimerization
O
HO
[4 + 4]
O 60
HO O O
O
O epoxytwinol (RKB-3564D), (65)
Scheme 16.17 Biosynthetic hypothesis for the epoxyquinols A–C (62–64) and epoxytwinol (65).
Ph H
H
Ph H H
H
H CO2H
H
H
H
H
H
HO2C H
H
H
CO2H
H H H
H
endiandric acid A (66)
H HO2C
H
H
Ph Ph
H endiandric acid D (69)
H Ph
H
H CO2H
H endiandric acid E (70)
H
endiandric acid F (71)
Figure 16.2
endiandric acid C (68)
endiandric acid B (67)
H CO2H
HO2C
H
H endiandric acid G (72)
Structures of endiandric acids A–G (66–72).
Ph
Ph
616
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products
The endiandric acids were isolated from the Australian plant Endiandra introrsa (Lauraceae) by Black’s group in 1980 [49]. Despite containing several stereocenters, the endiandric acids are found in Nature as racemates, which is unusual for chiral center containing natural products. To explain this observation, Black proposed an intriguing hypothesis for the biosynthesis of these molecules from achiral poly-unsaturated precursors through a series of non-enzymatic electrocyclizations (Scheme 16.18). The Black hypothesis suggests a cascade of reactions, shown in Scheme 16.18, by which endiandric acids A–G are formed in Nature. Thus endiandric acids E (70), F (71), and G (72) were proposed as intermediate precursors to the tetracyclic endiandric acids A (66), B (67), and C (68), respectively; the conversion being facilitated by an intramolecular Diels–Alder reaction. Endiandric acid D (69) cannot undergo an intramolecular Diels–Alder reaction, and so it does not form a corresponding tetracycle. An additional, striking feature of the Black hypothesis is that endiandric acids D–G could arise from sequential electrocyclizations of achiral polyenes (73–76). Intrigued by the structures of the endiandric acids and by Black’s hypothesis for their biogenetic origin, Nicolaou et al. initiated a program directed towards their total synthesis. The strategy they used relied upon three sequential pericyclic reactions, two of which were electrocyclizations (8π and 6π) and one of which was an intramolecular Diels–Alder reaction. The two electrocyclic reactions are thermally allowed by the Woodward–Hoffmann rules, and proceed in a stereospecific manner as shown in Scheme 16.18b. Thus, to obtain the desired trans disubstituted [4.2.0] bicyclic product, the (E,Z,Z,E) or the (Z,Z,Z,Z) tetraenes must be used. Through a display of a series of electrocyclization reactions, Nicolaou et al. demonstrated the biomimetic, one-step synthesis of the endiandric acids involving the cascade of reactions proposed by Black. For example, the acetylenic precursor 81 was assembled and converted into the conjugated polyene 82 using a diastereospecific Lindlar catalyst reduction to introduce the central (Z,Z) geometry. Compound 82 could not be isolated but instead underwent spontaneous 8π–6π electrocyclic cascades to yield endiandric acid methyl esters D (83) and E (84). Upon heating 84 in toluene at 100 ◦ C, an intramolecular Diels–Alder reaction then gave endiandric acid A methyl ester (85) (Scheme 16.18). Interestingly, although 83 cannot undergo a cycloaddition reaction it can equilibrate with 84. In fact, a one-pot reduction of 81 followed immediately by heating the crude products gave endiandric acid A methyl ester (85) in 30% yield [50]. The tendency of cyclooctatrienes like 77 to undergo further pericyclic reactions means that they are not common motifs in natural products. One rare example, however, is compound 87 isolated from the brown algae Cutleria multifida [51]. Boland et al. proposed that cyclooctatriene 87 is derived in vivo from (3Z,5Z,7E)-nona-1,3,5,7-tetraene (86), via a thermal 8π electrocyclization, and then confirmed his proposal through synthesis. The octatriene 87 was found to be relatively stable at room temperature and did not readily undergo 6π electrocyclization, which required heating to above 50 ◦ C. The corresponding 6π electrocyclization product 88, and the octatriene isomer 89, itself a product of [1,5]-hydrogen shift, were both subsequently found in Nature (Scheme 16.19) [52].
E
E Ph
8π con
Ph
8π con
Ph
HO2C n
82 Ph
CO2Me
n = 0, 79 n = 1, 80
6π dis
n = 0, 77 n = 1, 78
H
H
H CO2H n
H
H Ph
6π e− dis
8π e− con
H
H CO2Me H endiandric acid E methyl ester (84)
Ph
H
H endiandric acid D methyl ester (83)
MeO2C
H
n = 0, endiandric acid D (69) n = 1, endiandric acid G (72)
n
H
n = 0, endiandric acid E (70) n = 1, endiandric acid F (71)
Ph
HO2C
Ph
nCO2H
6π dis
[4+2]
∆
H CO H n 2
H
H
H
H H
H
H
H
Ph
CO2Me
H
endiandric acid A methyl ester (85)
H
Ph H
endiandric acid C (68)
HO2C H
Ph
H
H
n = 0, endiandric acid A (66) n = 1, endiandric acid B (67)
H
H
Ph H
(a) Black’s electrocyclization cascade hypothesis; (b) Nicolaou’s biomimetic synthesis of the endiandric
81
2. PhMe, 100 °C
1. H2, Pd/BaSO4, quinoline
CO2Me
n = 0, 75 n = 1, 76
Z
or
Z
n
Z
n CO2H
Z
n = 0, 73 n = 1, 74
Scheme 16.18 acids.
(b)
(a)
HO2C
Ph
Z
Z
16.7 8π Systems and the Black 8π–6π Electrocyclic Cascade 617
618
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products
Me
8π con
Me
H Me
> 50 °C
86
Scheme 16.19
7-methylcycloocta1,3,5-triene (87)
Me
+
6π dis
88
H
89
Biomimetic synthesis of 7-methylcycloocta-1,3,5-triene (87).
The methodology for the construction of bicyclo[4.2.0] ring systems that evolved from studies on the endiandric acids has now become firmly established in organic synthesis. Most notably, the tandem electrocyclization sequence has been observed by Widmer et al., using synthetic analogs of (9Z,11Z)-vitamin A, (9Z,11Z)-vitamin A acetate, and palmitate [53]. The thermal instability of conjugated tetraenes with similar geometry had previously been reported [54], and the research of Nicolaou and Widmer has demonstrated the strict requirement for the internal double bonds to attain the (Z,Z) configuration, so that the terminal double bonds may be brought into close proximity to initiate the cascade reaction sequence. 16.7.2 Nitrophenyl Pyrones: SNF4435 C and D
Natural products containing, or derived from, the bicyclo[4.2.0]octadiene ring systems are, however, rare, and only in the past few years have compounds been isolated with this motif in place. Two such compounds are the diastereoisomers SNF4435 C (90) and SNF4435 D (91), which were isolated in 2001 from the culture broth of a strain of Streptomyces spectabilis in a [2.3 : 1.0] ratio [55]. Both compounds displayed impressive immunosuppressant activity and multidrug resistance reversal effects, selectively suppressing B-cell proliferation versus T-cell growth. This mode of action was different from known immunosuppressants such as cyclosporin A (CsA) and FK-506, which work by inhibiting T-cell activation. The SNF compounds 90 and 91 belong to the nitrophenyl pyrone family of natural products, and are identified as consisting of a bicyclo[4.2.0]octadiene core linked to a spiro-fused tetrahydrofuran ring. This tricyclic core is substituted with a γ -pyrone moiety, and contains five stereocenters, two of which are quaternary. A further distinguishing structural feature is the presence of a para-nitrophenyl group, which is also found in other natural products such as orinocin (92) [56], aureothin (93) [57], and spectinabilin (94) [58] (Figure 16.3). The interesting stage 1 and 2 biosynthesis of the nitrophenyl pyrones 90–94 involves combining polypropionate units with p-aminobenzoic acid, and has been the topic of much investigation [59]. The proposed biogenesis of the endiandric acids suggested by Black, and the supporting experimental evidence provided by Nicolaou et al., inspired the groups led by Trauner [60] and Baldwin [61] to propose biomimetic routes to the SNF compounds 90 and 91 based upon stage 3 transformations. Both research groups realized that the origins of SNF4435 C (90) and SNF4435 D (91) could also be embedded in conjugated polyene chemistry,
16.7 8π Systems and the Black 8π–6π Electrocyclic Cascade O
O2N
O
O2N
O
O O
O
OMe
SNF4435 C (90)
OMe
SNF4435 D (91) O
O
O2N
O
O
OMe
O
O2N
orinocin (92)
O2N
Figure 16.3
O
OMe
aureothin (93) O
O spectinabilin (94)
O
OMe
Structures of p-nitrophenyl pyrones.
with Baldwin and one of the authors suggesting that spectinabilin (94) could be their biosynthetic precursor in Nature. Providing some weight to this hypothesis was an observation by Rinehart et al. in their report on the isolation of spectinabilin (94). It was noted that 94, a constitutional isomer of SNF4435C (90) and SNF4435D (91), ‘‘is not very stable, about 50% being converted to other substances during one month at room temperature’’ [58]. Although the ‘‘other substances’’ were not identified, it was not unreasonable to propose that spectinabilin (94) may have undergone various thermally/photochemically-induced double bond isomerizations, which could lead to a range of products derived from pericyclic chemistry. Thus, SNF4435 C (90) and SNF4435 D (91) were envisaged as having originated from an isomer of spectinabilin (94), namely, the (−Z,2Z,4Z,6E)-tetraene (95), which could itself arise from (94) by two consecutive internal double bond (E–Z)-isomerizations. The ‘‘internal’’ (Z,Z)-double bond geometry of tetraene 95 would allow the two ‘‘outer’’ double bonds of the tetraene back bone to meet and undergo the reaction cascade (stage 3), beginning with a diastereoselective 8π conrotatory electrocyclization of the tetraene 95 giving rise to both the major octatriene 96 and the minor diastereoisomer 97. An endo-selective 6π disrotatory electrocyclization of the octatrienes 96 and 97 then give rise to SNF4435 C (90) and SNF4435 D (91) respectively (Scheme 16.20). Model studies by the Trauner et al. and Baldwin et al. provided validation for this biomimetic hypothesis. The Trauner group opted to target the polyene precursor 101, with the internal double bonds set up with (Z,Z)-geometry, utilizing a convergent Stille cross coupling of the vinyl iodide 98 with the stannane 99. The intermediate 101 was not isolated, but instead immediately underwent a 8π–6π electrocyclic cascade leading to the cyclohexadiene 102, with a fused bicyclo[4.2.0]
619
620
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products O
E
E O2N
E
Z O
O
94
OMe
E-Z isomerization E Ar
Z Z
Z Pyr
Ar H
95
O H O Pyr
O H Pyr
H Ar
8π conrotatory electrocyclization Pyr H
H
H O
Ar O 97
Ar 96
Pyr H
endo -selective 6π disrotatory electrocyclization SNF4435 C, (90)
Scheme 16.20
SNF4435 D, (91)
Proposed biosynthesis of SNF compounds 90 and 91.
core (Scheme 16.21a). Baldwin et al., on the other hand, based on the spectinabilin (94) hypothesis, prepared the all-(E)-polyene 100, which upon treatment with Pd(II) resulted in (E–Z)-isomerization of the two internal double bonds to give the (2E,4Z,6Z,8E)-tetraene 101, which underwent the Black cascade in the same fashion as Trauner’s example (Scheme 16.21b). (a) Trauner's approach SnMe3 +
I
O2N 98
CO2Me
Pd(MeCN)2Cl2 DMF, rt
99 O2N
(b) Baldwin's approach
E
E
E
E CO Me 2
Pd(MeCN)2Cl2 DMF, rt
O2N
E
Z Z
E CO Me 8π – 6π 2
101
O2N 100
Scheme 16.21
CO2Me
Model studies toward SNF4435 C (90) and D (91).
102
16.7 8π Systems and the Black 8π–6π Electrocyclic Cascade
Following these studies, several total syntheses of the SNF-compounds were reported. Parker et al. reported the first total synthesis, having used a very similar coupling strategy as Trauner to access the (−E,2Z,4Z,6Z)-tetraene 95 (Scheme 16.22a) [62]. The vinyl iodide 98 and the vinyl stannane 103 underwent Pd-catalyzed cross coupling to give the tetraene 95, which then underwent the expected electrocyclization cascade to 90/91. Trauner’s synthesis involved a similar approach to his model studies, but in this case the stannane 104 and the iodide 105 were coupled to give the target structures in excellent yield (89%) (Scheme 16.22b) [63]. Baldwin’s biosynthetic hypothesis was developed using palladium catalysis as a means of isomerizing spectinabilin (94) to the required (−Z,2Z,4Z,6E)-tetraene geometry [64]. The tetraene 94/94 was constructed by sequential Suzuki and Negishi coupling reactions, beginning with the boronate ester 107 and the 1,1-dibromide 108 (Scheme 16.23). The isomerization followed by 8π–6π electrocyclic cascade was then initiated by 25 mol.% PdCl2 (MeCN)2 to furnish 90 and 91 in a [2.5 : 1.0] in 22% yield. This isomer ratio was close to the [2.3 : 1.0] ratio observed in Nature, and supports the biosynthetic hypotheses. Interestingly, diastereoisomers 110 and 111 were also isolated in 18% yield as a [2.1 : 1.0] mixture. This is consistent with the 8π–6π-electrocyclization cascade of the (−Z,2Z,4Z,6Z)- and (−E,2Z,4Z,6E)-tetraene isomers 112 and 113 (Scheme 16.24). The double bond isomerization hypothesis to 90 and 91 has since been established as being photo-induced in Nature, by Trauner and Hertweck et al. [65]. The biosynthetic cascade is initiated by two (E–Z)-internal double bond isomerization reactions of spectinabilin (94) to provide the helical geometry required for thermal 8π-conrotatory and 6π-disrotatory electrocyclizations. 16.7.3 Ocellapyrones
The powerful sequence involving double bond isomerization followed by electrocyclization cascades has been further developed by the research groups of Baldwin [29] and Trauner [30] in their syntheses of the molluscan polypropionate ocellapyrone A (27) [33]. Although the original synthetic target of both groups was 9,10-deoxytridachione (15) (Scheme 16.9), the key (2E,4Z,6E,8E)-tetraene 18 arrived at by both research groups was found to undergo further double bond isomerization to give the (2E,4Z,6Z,8E)-polyene 29. Intermediate 29 led to both ocellapyrone A (27) and its isomer (34) in varying amounts, and can be explained by examining the transition state for the 6π electrocyclization. Diastereomer 34 is the product arising from exo-orientation of the ethyl group, whereas compound 27 arises from endo-orientation (Scheme 16.25). To access the (2E,4Z,6Z,8E)-polyene 29 directly, Trauner [30] synthesized the vinyl stannane 116 and vinyl iodide 24, which were coupled using the conditions of Mee et al. [31]. Polyene 29 was not isolated, and underwent the electrocyclic cascade directly to give isomers 27 and 34. Interestingly, a stage 4 type Diels–Alder reaction of isomer 34 with singlet oxygen gave the corresponding ocellapyrone B (32) in excellent yield (Scheme 16.26).
621
103
O
105
+
104
Scheme 16.22
I
O2N
O
+
98
O OMe
O
I
OMe
Pd(PPh3)4, CsF, CuI, DMF
O2N
Pd(MeCN)2Cl2, DMF, rt
O OMe
[(b) 89%, 3.0:1.0 :: 90:91]
[(a) 53%, 3.8:1.0 :: 90:91]
95
O
O
O2N
8π– 6π
O2N O O
O OMe
O
OMe
O
SNF4435 D (91)
O
+
SNF4435 C (90)
Synthesis of SNF4435 C (90) and D (91): (a) Parker’s approach; (b) Trauner’s approach.
O
O
SnMe3
(b) Trauner's approach
SnMe3
O2N
(a) Parker's approach
622
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products
OMe
Scheme 16.23
THF, rt
O 2N
O B O
OMe
94/94'
O
108
+
O
O
Br
107, 98%, 1:1.2 E /Z
O
O
70 °C, dark, 1 day
25 mol% PdCl2(MeCN)2
OMe
Br
TlOEt, aq. THF
10 mol% Pd(PPh3)4
Baldwin’s syntheses of the SNF compounds 90 and 91.
O 2N
[57b]
steps
Me2Zn t 2 mol% Pd( Bu3P)2
106
O
O
O
O Pyr
Ar
109, 64% 35% Z-isomer 29% E-isomer
O
O
OMe
Pyr
Ar
110
O
Ar
18% [2.1:1.0 :: 110:111]
Pyr
111
O Pyr
22% [2.5:1.0 :: 90:91] SNF4435 C (90) SNF4435 D (91)
Ar
O2N
Br
O
O
16.7 8π Systems and the Black 8π–6π Electrocyclic Cascade 623
624
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products O O2N
Z
Z
O
Z
Z
Z
OMe
E
O
Z
E
O
O O 113
112
MeO
O2N H H O
Ar O
H Pyr
Ar
H Pyr
110
Scheme 16.24
111
Formation of diastereoisomers 110 and 111. O
E
O
E-Z isomerization
O
E
E Z O
E
Z
O
E
OMe
OMe
Z O
OMe
29
18
O 8π con
O
O
MeO O 114
MeO
OMe
115
6π dis
6π dis
H
H
O
O
O OMe O
Scheme 16.25
34
ocellapyrone A (27)
Electrocyclic formation of ocellapyrone A (27).
16.7.4 Elysiapyrones
The γ -pyrone functionality is clearly a very common motif in polypropionate-derived natural products from marine molluscs [66]. It is therefore not surprising that
16.7 8π Systems and the Black 8π–6π Electrocyclic Cascade O
O
E
Pd(PPh3)4, CsF CuI, DMF
I SnMe3
+
O
116
625
OMe
Z Z
E O
1 h, 45 °C
24
OMe
29 O O
H
H
O +
8π – 6π O
MeO
O
Scheme 16.26
O2 methylene blue
MeO
O
CHCl3, hn reflux, 30 min
OMe ocellapyrone A (27) (8%)
H
1
O 34 (78%)
O ocellapyrone B (32) (89%)
Trauner’s synthesis of ocellapyrones A (27) and B (32).
elysiapyrones A (125) and B (126), which were isolated from the mollusc Elysia diomedea, also have this privileged functionality. The elysiapyrones share common structural features with the ocellapyrones, consisting of a familiar bicyclo[4.2.0]octane core, but with the exception of epoxide ring decoration [67]. Darias proposed a biosynthetic hypothesis for these optically active metabolites, which involved an enzyme-mediated Black 8π–6π electrocyclization cascade (stage 3), followed by a [4+2] cycloaddition of singlet oxygen and subsequent rearrangement of the corresponding endoperoxide (stage 4) [67]. Trauner et al. used their experience in biomimetic electrocyclizations to investigate Darias’ hypothesis, and developed a total synthesis of elysiapyrones A (125) and B (126) in racemic form [68]. They again called upon the Stille reaction to access the key (2E,4Z,6Z,8E)-polyene 118, from fragments 117 and 24, which was not isolated and underwent the 8π–6π electrocyclization cascade to afford the diastereoisomers 121 and 122. Diels–Alder cycloaddition of 121 and 122 with singlet oxygen next gave endo-peroxides 123 and 124, respectively, which upon treatment with a ruthenium catalyst then gave the target isomerized epoxide products 125 and 126. This elegant synthesis followed the biosynthetic rationale closely, but then led to the products in racemic form (Scheme 16.27). 16.7.5 Shimalactones
Shimalactones A (127) and B (128) are another selection of natural products possessing a bicyclo[4.2.0]octadiene core. These interesting lactones were isolated from a marine-derived fungus, Emericella variecolor GF10, found at depths of 70 m off the coast of Gokasyo Gulf, Mie Prefecture, Japan [69], and display promising cytotoxic properties. Interesting structural features include a novel oxabicyclo[2.2.1]heptane moiety linked to a bicyclo[4.2.0]octadiene core by a trisubstituted alkene. These structural features make the shimalactones a challenging molecular target for organic synthesis. In 2006, Trauner et al., proposed that the biosynthesis of the shimalactones
6π
6π
24
O
O
Scheme 16.27
I
+
OMe
OMe
methyleneblue, CHCl3, reflux, 30 min
O2, hν
methyleneblue, CHCl3, reflux, 30 min
O2, hν
118
MeO
MeO
O
O
O 124
O
H
O 123
O
O O
OMe
O O
H
MeO
O O
DCM, 2 h
RuCl2(PPh3)3
DCM, 2 h
RuCl2(PPh3)3
Trauner’s biomimetic synthesis of elysiapyrones (125) and (126).
122
H
O
O
121
H
O
O
CuI, DMF
Pd(PPh3)4, CsF
OMe
SnMe3
117
O
O
8π
O
120
O
OMe
elysiapyrone B (126)
O
H
O
O
119
OMe
O
OMe
O
+
O
elysiapyrone A (125)
O
H
O
MeO
626
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products
16.7 8π Systems and the Black 8π–6π Electrocyclic Cascade
627
could involve the heptaenyl β-ketolactone 135 undergoing enzymatic epoxidation at the penultimate double bond to render 134, followed by acid-catalyzed epoxide opening to give undecapentaheptenyl cation 133 [70]. This cation could then isomerize to cation 132, which is attacked by the enol form of the β-ketolactone. Notably, structures 133 and 132 are probably not true resonance structures as A1,3-strain induced by the methyl groups prevent the π-system from being planar. Polyene 132 would then need to undergo isomerization to the (E,E,Z,Z,E)-polyene 131 to set up the 8π–6π electrocyclizations cascade to obtain shimalactones A (127) and B (128) (Scheme 16.28).
H+
H
H
OH
OH
OH
OH + O
O O
O
O shimalactone A (127)
O
O
132
O
6π e− dis shimalactone B (128)
[O] OH +
OH
O
O
133 OH
OH
O
O O
O
O 8π e− con
129
OH
H+ O
O
130
O
O
134 O
OH
E
Z E
E
Z O 131
Scheme 16.28
O
O O
135
Proposed biosynthesis of shimalactones A (127) and B (128).
Trauner et al. later described the biomimetic synthesis of the shimalactones using the acid-catalyzed cyclization 8π–6π electrocyclization strategy mentioned [70]. The electrocyclization precursor was assembled from the two fragments 145 and 148. The synthesis of compound 145 began with a Heathcock anti-aldol addition of the boron enolate of Evan’s reagent 136, followed by acylation of the resulting alcohol 137 with propionic anhydride to give 138, which underwent a Stille coupling with tributylisopropenylstannane to afford 139 (Scheme 16.29). Dieckmann cyclization using KHMDS (potassium hexamethyldisilazane) next gave the β-ketolactone 140 as a single diastereoisomer. After careful optimization it was found that camphorsulfonic acid (CSA) catalyzed the cyclization to give the oxabicyclo[2.2.1]heptane core found in the natural product. Allylic bromination of
O
628
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products
141 under radical bromination conditions gave rise to 142 and 142 as a 1 : 1 mixture. Compound 142 underwent smooth oxidation with IBX (2-iodoxybenzoic acid) to yield the corresponding aldehyde 143. Despite being very sensitive to basic conditions, a three-step sequence was realized to afford 144, which after a Stork–Zhao olefination gave fragment 145 (Scheme 16.29). Fragment 148 was prepared using an asymmetric addition of 2-butenylmethyl zinc to aldehyde 146, followed by iodine–tin exchange. In a spectacular display of the power of biomimetic synthesis, when fragments 145 and 148 were exposed to Pd(0), shimalactones A (127) and B (128) were accessed directly in 66% yield as a 5.0 : 1.0 mixture, with the intermediate 131 not isolated (Scheme 16.30).
16.8 Biological Electrocyclizations and Enzyme Catalysis
The involvement of enzyme catalysis in biological electrocyclizations is uncertain. Many of the natural products that are believed to be derived through electrocyclizations have been isolated in racemic form (endiandric acids, etc.). Clearly, in such cases one can rule-out the involvement of electrocyclases. On the other hand, where optically active precursors undergo electrocyclization this would, in principle, lead to optically active products without the need for enzyme assistance. For example, the SNF compounds 90 and 91 described above are derived from enantiomerically pure spectinabilin (Scheme 16.20), and it is highly unlikely that an electrocyclase is involved. This is more important when considering photochemical electrocyclizations, which would not require enzyme assistance since they proceed through high-energy excited states. However, there are cases that cannot be explained so easily. For instance, ocellapyrones A (27) and B (32) and elysiapyrones A (125) and B (126) have all been isolated in optically active form, despite their proposed precursors being achiral. Although it is possible that a particular enantiomer may be consumed selectively through the process of metabolism, such a ‘‘kinetic resolution’’ theory is unlikely to explain every case. It has been proposed that the course of the 8π–6π electrocyclization may be guided in the active site of the PKS that generates the polyketide precursors (Figure 16.4) [71]. The chiral environment could in principle influence the torquoselectivity of the electrocyclization, leading to optically enriched products. Furthermore, the synthase may also accelerate the cyclization by preorganizing the conformation of the polyene precursor, thus acting as a catalyst. This hypothesis has some weight, when one considers the biosynthesis of aromatic polyketides. In this event, polycarbonyl intermediates undergo intramolecular condensation and aromatization when covalently linked to the type II PKS [72]. A similar argument applies to several cyclohexadienes, such as 9,10-deoxytridachione (15), which have been isolated in optically active form. Such compounds are proposed to arise through 6π thermal electrocyclizations from achiral polyenes (Schemes 16.7–16.10). However, it has been shown that elevated temperatures
O
136
N
O
Scheme 16.29
IBX
KHMDS
O
O
O
O
I
O
O
(EtCO)2O
+ O
I
Et
O
141
O
O
O
141'
144, 80%
O
O
O
O
138, 96%
O
65%, [2.0:1.0 :: 141:141']
P OEt OEt
O
O
1. NaH 2. LiEt3BH 3. Dess-Martin
EtO
O
150 °C
CSA, Benzene
O
O
N
137, 79%
OH O O
I PPh3
SnBu3
O
Br
Pd(PPh3)4 CuI, CsF
NBS, AIBN
N
O
Trauner’s synthesis of key fragment 145, en route to the shimalactones.
143, 84%
O
O
140, 93%
O
O
nBu2BOTf DIPEA
I
CHO
O
O
O
+
139, 92%
O N
O
Br
O
O
I
142
145, 32%
O
O
O
142'
O O 97%, [1.0:1.0 :: 142:142']
Et
O
16.8 Biological Electrocyclizations and Enzyme Catalysis 629
630
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products n
Cy2BH, Me2Zn 2-Butyne
I OHC
OH
I
BuLi Me3SnCl
OH
SnMe3
Pd(PPh3)4 CuTC 145
Ph
146
Ph
147, 64%, 95% ee
N
148, 23%
OH
E
H
H
OH
Z E
E +
OH
Z
OH
O
O
O
O
O
131
O
O
O shimalactone A (127)
O
shimalactone B (128)
66%, [5.0:1.0 :: 127:128]
Scheme 16.30 Trauner’s biomimetic synthesis of 148 and shimalactones A (127) and B (128).
O S
Figure 16.4 PKS-bound tetraene undergoing an asymmetric-torquoselective 8π electrocyclization.
are often required to initiate these cyclizations, and it could be the case that the reactions are actually photochemically derived, although the relative double bond configurations of the polyene precursor would need to be different. This still does not explain the origin of optical activity in these compounds. In fact, compounds that are chemically derived pose a more intriguing question, since it is highly unlikely that photoactive enzymes have evolved to effect such processes, since they are very scarce [73]. The authors would suggest an alternative explanation for the case of the mollusc metabolite tridachiahydropyrone (6) and other related photochemical electrocyclization products.1) It has been suggested that some polypropionate metabolites are biosynthesized and then translocated to the tissue of the digestive diverticula of the molluscs, where they serve as sunscreens to protect the molluscs from damaging UV-radiation 1) The proposal was first put forward by one
of the authors, J. E. Moses, during a research lecture at The Chemistry Research
Laboratory, University of Oxford on 15 October 2009.
16.9 Conclusion Local chirality influences the torquoselectivity of 6π electrocyclization
Local chirality of lipid membrane Polyene embeded in lipid bilayer Polar Head
Cholesterol
OMe
OH OMe
O O Hydrophobic Tail
O H H H
O
hn E-Z isomerization
13
O
hn Release from lipid
7
O
OMe
Enantiomerically enriched 6 (arbitarily assigned)
Model lipid bilayer
Scheme 16.31
631
Photochemical electrocyclizations in membranes.
[23, 25]. This hypothesis is reasonable, since these sacoglossans live in shallow lagoons where they are exposed to direct sunlight. In this respect, perhaps the complexity of the natural products becomes a consequence of the function of their predecessor’s (e.g., polyene 13, Scheme 16.5) to act as a sunscreen, and, as such, would not necessarily have evolved enzymatic assistance for their syntheses. As sunscreens, optimal protection would result from the compounds being located in the outer region of the cell, and most likely in the membrane. Given that the key polyene 13 has similar structural properties to that of a lipid, that is, a fatty chain and a polar head group, then it is reasonable that 13 would sit comfortably in the lipid bilayer of the surface cell membrane (Scheme 16.31). The cell membrane consists of a thin layer of amphipathic phospholipids that forms a continuous, spherical lipid bilayer. The composition is complex and includes proteins, carbohydrates, and cholesterol among other components. The immediate chiral environment of the membrane could in principle influence the resulting enantiomeric purity of the tridachiahydropyrone family (and related compounds); that is, polyene 13 becomes embedded in the bilayer, and upon irradiation is transformed into (−)-tridachiahydropyrone (6). The surrounding chiral environment of the membrane may influence the reaction in a torquoselective fashion, favoring the formation of one enantiomer over the other. Studies by Hailes in the 1990s provide some support for this hypothesis. In their work, Hailes et al. demonstrated that chiral micellar media can be used to induce modest levels of enantiomeric enrichment in aqueous Diels–Alder reactions [74]. 16.9 Conclusion
Considering the volume of evidence, it is highly likely that electrocyclic reactions play an important role in the biosynthesis of polyketide-derived natural products.
632
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products
The above discussion is not exhaustive or complete, and no doubt the future will provide many more interesting examples. With very few exceptions, most notably 9,10-deoxytridachione (15), most of the electrocyclizations discussed occur readily without need for catalysis or excessive temperatures, which is supportive of the biomimetic hypothesis in general [1, 2]. As such, electrocyclization reactions have proven valuable and effective tools in the synthesis of highly complex and structurally interesting natural products. The identification of enzymes that mediate electrocyclizations is still an important goal and, if they do indeed exist, such enzymes may provide inspiration for biomimetic catalysts to achieve asymmetric induction in these processes. To date, there are very few examples of asymmetric electrocyclizations [75], providing a wealth of opportunity for discovery and development [1].
Acknowledgments
We are very grateful to Professor G. Pattenden, FRS (University of Nottingham) and Dr. R. Rodriguez (University of Cambridge) for their helpful comments and suggestions during the preparation of this manuscript. We thank the EPSRC (P.S.), Pfizer (M.R.), and University of Nottingham (J.B.) for financial support.
References 1. Moses, J.E. and Adlington, R.M. (2005) 2. 3.
4.
5.
6.
7.
Chem. Commun., 5945–5952. Heathcock, C.H. (1996) Proc. Natl. Acad. Sci. U.S.A., 93, 14323–14327. Corey, E.J. and Cheng, X.-M. (1995) The Logic of Chemical Synthesis, John Wiley & Sons, Inc., New York. (a) Nicolaou, K.C. and Sorensen, E.J. (1996) Classics in Total Synthesis: Targets, Strategies, Methods, Wiley-VCH Verlag GmbH, Weinheim; (b) Nicolaou, K.C. and Snyder, S. (2003) Classics in Total Synthesis II: More Targets, Strategies, Methods, Wiley-VCH Verlag GmbH, Weinheim. Sierra, M.A., de la Torre, M., and Nicolaou, K.C. (2004) Dead Ends and Detours: Direct Ways to Successful Total Synthesis, Wiley-VCH Verlag GmbH, Weinheim. The biosynthesis of Taxol provides an excellent example: Rhor, J. (1997) Angew. Chem. Int. Ed., 36, 2190–2195. Sir Robert Robinson’s synthesis of tropinone is generally credited as the first biomimetic synthesis: Robinson,
R. (1917) J. Chem. Soc., Trans., 111, 762–768. 8. (a) Johnson, W.S., Gravestock, M.B., Parry, R.J., Myers, R.F., Bryson, T.A., and Miles, D.H. (1971) J. Am. Chem. Soc., 93, 4330–4332; (b) Johnson, W.S., Gravestock, M.B., and McCarry, B.E. (1971) J. Am. Chem. Soc., 93, 4332–4334. 9. Examples of biomimetic terpene synthesis: (a) Yoder, R.A. and Johnston, J.N. (2005) Chem. Rev., 105, 4730–4756; (b) Tang, B., Bray, C.D., and Pattenden, G. (2009) Org. Biomol. Chem., 7, 4448–4457; (c) Trauner, D., Elliott, G.I., Maimone, T.J., and Malerich, J.P. (2005) J. Am. Chem. Soc., 127, 6276–6283. 10. Examples of biomimetic Alkaloid synthesis: (a) Scholz, U. and Winterfeldt, E. (2000) Nat. Prod. Rep., 17, 349–366; (b) Amat, M., Griera, R., Fabregat, R., Molins, E., and Bosch, J. (2008) Angew. Chem. Int. Ed., 47, 3348–3351; (c) Schwartz, M.A. and Zoda, M.F. (1981) J. Org. Chem., 46, 4623–4625.
References 11. Examples of biomimetic polyketide
12.
13. 14. 15.
16. 17.
18.
19. 20.
21.
22. 23. 24. 25. 26. 27.
synthesis that are non-electrocyclic: Gademann, K., Bethuel, Y., Locher, H.H., and Hubschwerlen, C. (2007) J. Org. Chem., 72, 8361–8370. Moses, J.E., Adlington, R.M., Baldwin, J.E., Commeiras, L., Cowley, A.R., Baker, C.M., Albrecht, B., and Grant, G.H. (2006) Tetrahedron, 62, 9892–9901; (b) Moses, J.E., Commeiras, L., Baldwin, J.E., and Adlington, R.M. (2003) Org. Lett., 5, 2987–2988; (c) Nicolaou, K.C., Snyder, S.A., Montagnon, T., and Vassilikogiannakis, G. (2002) Angew. Chem. Int. Ed., 41, 1668–1698. For example see: Tietze, L.F. (1996) Chem. Rev., 96, 115–136. Woodward, R.B. and Hoffmann, R. (1964) J. Am. Chem. Soc., 87, 395–397. Fleming, I. (2010) Molecular Orbitals and Organic Chemical Reactions: Reference Edition, Wiley-Blackwell. Staunton, J. (1997) Chem. Rev., 97, 2611–2629. Shida, N., Miyazaki, K., Kumagai, K., and Rikimaru, M. (1965) J. Antibiot., 18, 68–76. Murata, M., Naoki, H., Iwashita, T., Matsunaga, S., Saski, M., Yokoyama, A., and Yasumoto, T. (1993) J. Am. Chem. Soc., 115, 2060–2062. Weissman, K.J. (2009) Methods Enzymol., 459, 3–16. (a) Sharma, P., Griffiths, N., and Moses, J.E. (2008) Org. Lett., 10, 4025–4027; (b) Sharma, P., Lewis, W., Lygo, B., and Moses, J.E. (2009) J. Am. Chem. Soc., 131, 5966–5972. Ortea, J., Cimino, G., Mollo, E., and Gavagnin, M. (1996) Tetrahedron Lett., 37, 4259–4262. Fu, X., Hong, E.P., and Schimitz, F.J. (2000) Tetrahedron, 56, 8989–8993. Ireland, C. and Scheuer, P.J. (1979) Science, 205, 922–923. Sharma, P. and Moses, J.E. (2010) Synlett, 4, 525–529. Ireland, C. and Faulkner, J.D. (1981) Tetrahedron, 9, 233–240. Schmitz, F.J. and Ksebati, M.B. (1985) J. Org. Chem., 50, 5637–5642. Gavagnin, M., Spinella, A., Castelluccio, F., Cimino, G., and
28.
29.
30. 31.
32.
33.
34.
35.
36. 37.
38.
39.
Marin, A. (1994) J. Nat. Prod., 57, 298–304. Epoxidation of deoxytridachione: Ireland, C., Faulkner, D.J., Solheim, B.A., and Clardy, J. (1978) J. Am. Chem. Soc., 100, 1002–1003. (a) Rodriguez, R., Adlington, R.M., Eade, S.J., Walter, M.W., Baldwin, J.E., and Moses, J.E. (2007) Tetrahedron, 63, 4500–4509; (b) Moses, J.E., Adlington, R.M., Rodriguez, R., Eade, S.J., and Baldwin, J.E. (2005) Chem. Commun., 1687–1689. Miller, A.K. and Trauner, D. (2005) Angew. Chem. Int. Ed., 44, 4602–4606. Mee, S.P.H., Lee, V., and Baldwin, J.E. (2004) Angew. Chem. Int. Ed., 43, 1132–1136. Eade, S.J., Walter, M.W., Byrne, C., Odell, B., Rodriguez, R., Baldwin, J.E., Adlington, R.M., and Moses, J.E. (2008) J. Org. Chem., 73, 4830–4839. Manzo, E., Ciavatta, M.L., Gavagnin, M., Mollo, E., Wahidulla, S., and Cimino, G. (2005) Tetrahedron Lett., 46, 465–468. (a) Br¨uckner, S., Baldwin, J.E., Moses, J.E., Adlington, R.M., and Cowley, A.R. (2003) Tetrahedron Lett., 44, 7471–7473; (b) Moses, J.E., Baldwin, J.E., Marquez, R., Adlington, R.M., Claridge, T.D.W., and Odell, B. (2003) Org. Lett., 5, 661–663. Zuidema, D.R., Miller, A.K., Trauner, D., and Jones, P.B. (2005) Org. Lett., 7, 4959–4962. Rickards, R.W. and Skropeta, D. (2002) Tetrahedron, 58, 3793–3800. (a) Marvell, E.N. (1980) Thermal Electrocyclic Reactions, vol. 43, Academic Press, New York; (b) Ansari, F.L., Qureshi, R., and Qureshi, M.L. (1999) Electrocyclic Reactions, Wiley-VCH Verlag GmbH, Weinheim; (c) Woodward, R.B. and Hoffmann, R. (1970) The Conservation of Orbital Symmetry, Verlag Chemie, Weinheim. Lee, J.C., Strobel, G.A., Lobkovsky, E., and Clardy, J. (1996) J. Org. Chem., 61, 3232–3233. Li, J.Y., Harper, J.K., Grant, D.M., Tombe, B.O., Bashyal, B., Hess, W.M., and Strobel, G.A. (2001) Phytochemistry, 56, 463–468.
633
634
16 Biomimetic Electrocyclization Reactions toward Polyketide-Derived Natural Products 40. Li, C., Lobkovsky, E., and Porco, J.A.
41.
42. 43.
44.
45.
46. 47. 48. 49.
50.
51. 52. 53.
Jr. (2000) J. Am. Chem. Soc., 122, 10484–10485. Li, C., Johnson, R.P., and Porco, J.A. Jr. (2003) J. Am. Chem. Soc., 125, 5095–5106. Mehta, G. and Pan, S.C. (2004) Org. Lett., 6, 3985–3988. (a) Kakeya, H., Onose, R., Koshion, H., Yoshida, A., Kobayashi, K., Kageyama, S.I., and Osada, H. (2002) J. Am. Chem. Soc., 124, 3496–3497; (b) Kakeya, H., Onose, R., Yoshida, A., Koshino, H., and Osada, H. (2002) J. Antibiot., 55, 829–831. Kakeya, H., Onose, R., Koshino, H., and Osada, H. (2005) Chem. Commun., 2575–2577. Shoji, M., Imai, H., Mukaida, M., Sakai, K., Kakeya, H., Osada, H., and Hayashi, Y. (2005) J. Org. Chem., 70, 79–91. Li, C. and Porco, J.A. Jr. (2005) J. Org. Chem., 70, 6053–6065. Kuwahara, S. and Imada, S. (2005) Tetrahedron Lett., 46, 547–549. Mehta, G. and Islam, K. (2004) Tetrahedron Lett., 45, 3611–3615. (a) Bandaranayake, W.M., Banfield, J.E., Black, D.St.C., Fallon, G.D., and Gatehouse, B.M. (1980) J. Chem. Soc., Chem. Commun., 162–163; (b) Bandaranayake, W.M., Banfield, J.E., and Black, D.St.C. (1980) J. Chem. Soc., Chem. Commun., 902–903. (a) Nicolaou, K.C., Petasis, N.A., Zipkin, R.E., and Uenishi, J. (1982) J. Am. Chem. Soc., 104, 5555–5557; (b) Nicolaou, K.C., Petasis, N.A., Uenishi, J., and Zipkin, R.E. (1982) J. Am. Chem. Soc., 104, 5557–5558; (c) Nicolaou, K.C., Zipkin, R.E., and Petasis, N.A. (1982) J. Am. Chem. Soc., 104, 5558–5560; (d) Nicolaou, K.C., Petasis, N.A., and Zipkin, R.E. (1982) J. Am. Chem. Soc., 104, 5560–5562. Pohnert, G. and Boland, W. (1994) Tetrahedron, 50, 10235–10244. Pohnert, G. and Boland, W. (2002) Nat. Prod. Rep., 19, 108–122. Vogt, P., Schlageter, M., and Widmer, E. (1991) Tetrahedron Lett., 32, 4115–4116.
54. Huisgen, R., Dahmen, A., and Huber,
55.
56.
57.
58.
59.
60. 61.
62. 63. 64.
H. (1967) J. Am. Chem. Soc., 89, 7130–7131. (a) Kurosawa, K., Takahashi, K., and Tsuda, E.J. (2001) J. Antibiot., 54, 541–547; (b) Takahashi, K., Tsuda, E., and Kurosawa, K. (2001) J. Antibiot., 54, 548–553. Pancharoen, O., Picker, K., Reutrakul, V., Taylor, W.C., and Tuntiwachwuttikul, P. (1987) Aust. J. Chem., 40, 455–459. (a) Hirata, Y., Nakata, H., Yamada, K., Okuhara, K., and Naito, T. (1961) Tetrahedron, 14, 252–274. (b) synthesis: Jacobsen, M.F., Moses, J.E., Adlington, R.M., and Baldwin, J.E. (2005) Org. Lett., 7, 641–644. Kakinuma, K., Hanson, C.A., and Rinehart, K.L. Jr. (1976) Tetrahedron, 32, 217–222. (a) Hertweck, C. (2009) Angew. Chem. Int. Ed., 48, 4688–4716; (b) Buscha, B. and Hertweck, C. (2009) Phytochemistry, 70, 1833–1840; (c) He, J. and Hertwick, C. (2004) J. Am. Chem. Soc., 126, 3694–3695; (d) Taniguchi, M., Watanabe, M., Nagai, K., Suzamura, K., Suzuki, K., and Tanaka, A. (2000) J. Antibiot., 53, 844–847; (e) Yamazaki, M., Katoh, F., Ohnishi, J., and Koyama, Y. (1972) Tetrahedron Lett., 26, 2701–2704; (f) Yamazak, M., Maebayashi, Y., Katoh, H., Ohishi, J., and Koyama, Y. (1975) Chem. Pharm. Bull., 23, 569–574; (g) Cardillo, R., Fuganti, C., Giangrasso, G.D., Grasselli, P., and Santopietro-Amisano, A. (1974) Tetrahedron, 30, 459–461; (h) Cardillo, R., Fuganti, C., Ghiringhelli, D., and Giangrasso, D. (1972) Tetrahedron Lett., 48, 4875–4878. Beaudry, C.M. and Trauner, D. (2002) Org. Lett., 4, 2221–2224. Moses, J.E., Baldwin, J.E., Marquez, R., and Adlington, R.M. (2002) Org. Lett., 4, 3731–3734. Parker, K.A. and Lim, Y.H. (2004) J. Am. Chem. Soc., 126, 15968–15969. Beaudry, C.M. and Trauner, D. (2005) Org. Lett., 7, 4475–4477. (a) Jacobsen, M.F., Moses, J.E., Adlington, R.M., and Baldwin, J.E.
References
65.
66.
67.
68.
(2006) Tetrahedron, 62, 1675–1689; (b) Jacobsen, M.F., Moses, J.E., Adlington, R.M., and Baldwin, J.E. (2005) Org. Lett., 7, 2473–2476. Muller, M., Kusebauch, B., Liang, G., Beaudry, C.M., Trauner, D., and Hertweck, C. (2006) Angew. Chem. Int. Ed., 45, 7835–7838. (a) Wetzel, S., Wilk, W., Chammaa, S., Sperl, B., Roth, A.G., Yektaoglu, A., Renner, S., Berg, T., Arenz, C., Giannis, A., Oprea, T.I., Rauh, D., Kaiser, M., and Waldmann, H. (2010) Angew. Chem. Int. Ed., 49, 3666–3670; (b) Wilk, W., Waldmann, H., and Kaiser, M. (2009) Bioorg. Med. Chem., 17, 2304–2309. Cueto, M., D’Croz, L., Mate, J.L., San-Martin, A., and Darias, J. (2005) Org. Lett., 7, 415–418. Barbarow, J.E., Miller, A.K., and Trauner, D. (2005) Org. Lett., 7, 2901–2903.
69. Wei, H., Itoh, T., Kinoshita, K.,
70. 71.
72.
73.
74. 75.
Kotoku, N., Aoki, S., and Kobayashi, M. (2005) Tetrahedron, 61, 8054– 8058. Sofiyev, V., Navarro, G., and Trauner, D. (2008) Org. Lett., 10, 149–152. Beaudry, C.M., Malerich, J.P., and Trauner, D. (2005) Chem. Rev., 105, 4757–4778. Korman, T.P., Hill, J.A., Vu, T.N., and Tsai, S.C. (2004) Biochemistry, 43, 14529–14538. Heyes, D.J., Ruban, A.V., Wilks, H.M., and Hunter, C.N. (2002) Proc. Natl. Acad. Sci. U.S.A., 99, 11145–11150. Diego-Castro, M.J. and Hailes, H.C. (1998) Chem. Commun., 1549–1550. Maciver, E.E., Thompson, S., and Smith, M.D. (2009) Angew. Chem. Int. Ed., 48, 9979–9982.
635
637
Part IV Biomimetic Synthesis of Polyphenols
Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
639
17 Biomimetic Synthesis and Related Reactions of Ellagitannins Takashi Tanaka, Isao Kouno, and Gen-ichiro Nonaka
17.1 Introduction
Tannins are defined as a specific group of plant polyphenols that precipitate proteins and heavy metals from aqueous solutions. The scientific term ‘‘tannin’’ etymologically originated from leather tanning, in which collagen proteins of animal raw hides are precipitated by complexation with tannins to produce durable leather. Tannins are also responsible for the astringent taste of food and beverages caused by precipitating salivary proteins [1]. This property is related to the biological dysfunction of proteins such as inhibition of enzymes in animal digestive tracts; therefore, tannins are thought to accumulate as feeding deterrents against herbivores [2]. Ironically, it has recently been recognized that the inhibition of the digestive enzymes is linked to a decrease in the incidences of common diseases caused by diets rich in carbohydrates and fats because the uptake of sugars and fats is decreased [3]. In addition, many biological activities, such as antioxidative, anticancer, and immunomodulation activities are reported [4–6]. Tannins are basically classified into two major groups based on their biosynthetic origins: hydrolyzable tannins and condensed tannins [1]. The condensed tannins, a synonym of ‘‘proanthocyanidins,’’ are C–C linked oligomers of flavan-3-ols (catechins). The hydrolyzable tannins are esters of gallic acid (3,4,5-trihydroxybenzoic acid) or related phenol carboxylic acids with polyalcohols, usually d-glucose. There are also some exceptional tannins possessing both structural units, which are called complex tannins [7]. In contrast to the wide distribution of condensed tannins in Angiospermae, Gymnospermae, and even in ferns [8], hydrolyzable tannins are found in relatively limited Dicotyledoneae plant families [9]. Nevertheless, the structural variation of hydrolyzable tannins is wider than proanthocyanidins and therefore attractive from the viewpoint of organic chemistry, because the oxidative metabolism of simple galloyl esters with glucopyranose produces a wide range of structurally different molecules, as seen in this chapter. The oxidation products are called ellagitannins, which typically possess 3,4,5,3 ,4 ,5 -hexahydroxydiphenoyl (HHDP) esters derived by oxidative C-C coupling between two galloyl esters (Scheme 17.1) [10]. The coupling reactions proceed under complete stereochemical control. The Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
2
O
HO OH
O O O
3
O
O
R
OH
R
OH
O
HO
HO
HO
HO
O O R
O
R O
3
O
O R
R
OH
OH
O R
O
6 O
OO
O
HO OH
R
O
O
3,6-(R)-HHDP-glucose
HO
HO
ellagitannin
OH
O
O
OH (S)-HHDP group
Biosynthesis and decomposition of ellagitannins.
2,3-(S)-HHDP-glucose
HO
HO
R O R O O
R
O
oxidative coupling
R = Horgalloyl groups
O R
O
O
O
galloyl group
gallotannin
O
OH
Scheme 17.1
HO
HO
HO OH
HO hydrolysis
HO
O
O
O
O
OH
OH
O
O O O R
4
O
OH
R
6
OH +
4,6-(S)-HHDP-glucose
HO
HO
HO
HO
ellagic acid (1)
HO
O
O O
R
CH2
R
HO O R
HO
O
O O
R
640
17 Biomimetic Synthesis and Related Reactions of Ellagitannins
17.2 Biosynthesis of Ellagitannins
term ellagitannins originated from the observation that the hydrolysis of usual ellagitannins affords ellagic acid (1). Structural diversity of ellagitannins comes from the presence of positional and stereochemical isomers of the HHDP group on a glucopyranose core and further oxidative modifications of the molecules including production of quinone derivatives and oligomeric compounds. This chapter provides a concise overview of the total synthesis of ellagitannins and reactions related to semi-synthesis.
17.2 Biosynthesis of Ellagitannins
It is commonly recognized that different plant families accumulate different types of ellagitannins. This fact apparently implies that the biosynthesis of ellagitannins is controlled by enzymes specific to each plant, and this is the reason why the structural variation of ellagitannins is also important from the viewpoint of chemotaxonomy [11]. Although the details of ellagitannin biosynthesis remain unresolved, quite recent pioneering studies have been achieved by Gross and coworkers [12]. Oxidation of 1,2,3,4,6-pentagalloyl-β-d-glucose (2) with an enzyme preparation obtained from Tellima grandiflora (Saxifragaceae) furnished tellimagrandin II (3) [13] (syn. eugeniin [14]) as a major product (Scheme 17.2). This was the first and only successful enzymatic preparation of ellagitannins. The reaction proceeds not only regioselectively but also diastereoselectively. It was demonstrated that the enzyme belongs to the laccase subgroup of phenol oxidases (EC 1.10.3.2). In addition, another enzyme catalyzing regioselective intermolecular C–O coupling of 3 to cornusiin E (4) [15] was characterized (Scheme 17.2). This enzyme also belongs to the same subgroup of polyphenol oxidases. The reactions, however, only represent the biosynthesis of limited members of the vast ellagitannin class. Plants exemplify various intramolecular C–C couplings of galloyl groups on a glucopyranose core; that is, coupling between 1,6-; 1,3-; 2,3-; 2,4-; 3,6-; 3,4-; and 4,6-positions are known [16, 17]. Regiospecificity of intermolecular couplings in oligomerization also differs with plant species [9, 11]. Furthermore, the mechanisms of further oxidative metabolism of the HHDP groups (vide infra) remains a matter of speculation. If the tannins are accumulated simply as defensive substances to disable functions of a herbivore’s proteins [2] it may not be necessary to convert the pentagalloylglucose into ellagitannins because galloylglucose exhibits sufficient activity to precipitate proteins. However, the wide distribution of ellagitannins in the plant kingdom is suggestive of particular benefits of the oxidation of the gallotannins. Higher water solubility of ellagitannins compared to pentagalloylglucose may have some advantage in the accumulation of the compounds in the plant cell vacuoles [18], because protein precipitation is only effective when ellagitannins are present in high concentrations [2]. In addition, there are interesting reports from another viewpoint of plant–herbivore interactions. It is known that many herbivorous caterpillar midgut fluids have a high pH (e.g., pH 10),
641
17 Biomimetic Synthesis and Related Reactions of Ellagitannins
642
OH
HO
HO
HO
O
HO HO O
O
HO O HO
OH
6
4
O
O
HO
OH
1/2 O2
O
O OH
O
O
O O
O
O
OH
OH OH
OH
OH
OH
2
3 3 HO
OH
HO
HO
3 HO
OH
O O
HO HO
HO
O
O
O
O O
OH
HO
HO
HO HO
H2O
OH O
O
HO
1/2 O2
O
O O
OH
OH
OH
6
4
HO HO
H2O
OH
O
S
HO
OH
O
O
HO
HO
2O
O HO
O O
O O HO
O
HO HO
OH
4
O
O
O
OH
O O
OH
O
OH
O
HO OH
OH
O
O
OH
OH
OH
OH 4
Scheme 17.2
Production of monomeric and dimeric ellagitannins from pentagalloylglucose.
and ellagitannins react to form high concentrations of harmful semiquinone radicals under alkaline conditions compared to proanthocyanidins and simple gallotannins [19].
17.3 Biomimetic Total Synthesis of Ellagitannins 17.3.1 Chemical Synthesis of Ellagitannins by Biaryl Coupling of Galloyl Esters
The first total synthesis of a natural ellagitannin was achieved by Feldman et al. in a manner similar to the enzymatic HHDP formation (Scheme 17.3) [20]. A synthetic precursor 5 having two partially protected galloyl groups at glucose 4,6-positions was treated with Pb(OAc)4 to give a cyclized product 6 and subsequent deprotection furnished tellimagrandin I (7) [21]. The reaction was diastereoselective and the atropisomerism of the biphenyl bond was exclusively the (S)-configuration. In addition, an analogous reaction using substrate 8 having the same acyl groups
BnO
O BnO
Scheme 17.3
Ph Ph
O
Ph HO O
Ph O
O
4
O
O
O
O
O
Ph Ph
O
6
O
O
HO
Ph
O
Ph HO O O
OH
O Ph
OBn
OBn
Ph O
OBn
O
O O O
8
4
O
O
O
6
73%
Pb(OAc)4
O
O
O Ph
O
H, CPh2
H, CPh2
Ph
O
OH
O
O O
O
S
O
O
O
6
2) H2/Pd
1) Pb(OAc)4
6
OBn
O 4 O O O
BnO
BnO
O
O
First chemical synthesis of natural ellagitannin, tellimagrandin I (7).
5
O
OBn
O
OH
O
O
7
OBn
O
(29%)
OBn
OBn
HO
O
HO
HO
OH
OH
O
O
OH
O
7
O
O
O
+ 2,3,4,6-tetra-O-galloylglucose (29%)
82%
H2/Pd
HO HO
HO
O
O
OH
OH
OH
OH
17.3 Biomimetic Total Synthesis of Ellagitannins 643
644
17 Biomimetic Synthesis and Related Reactions of Ellagitannins
at glucose 2,3,4,6-positions furnished 7 along with another natural gallotannin, 2,3,4,6-tetragalloyl glucose [22]. The reaction was not only diastereoselective but also regioselective, that is, coupling between the C2 and C3 galloyl groups was not observed. Tellimagrandin II (3), the 1-O-β-galloyl form of 7, was also synthesized using a similar methodology [23]. Molecular mechanics based conformational studies suggested that the precedence of the 4,6-galloyl coupling over the 2,3- or 3,4-galloyl couplings in the reaction was due to the shorter distance between the aromatic rings [20, 24]. As for the atropisomerism of the HHDP group, the exclusive formation of the (S)-biphenyl bond in this synthetic study was also supported by computational chemistry calculations. The first synthesis of an ellagitannin with 2,3-HHDP glucose, sanguiin H5 (11), was also accomplished by Feldman and coworkers by similar Pb(OAc)4 coupling of partially protected galloyl groups attached to glucose 2,3 positions. The reaction also yielded (S)-HHDP groups diastereoselectively [25]. Furthermore, the biaryl coupling strategy was adapted for the synthesis of 2,3-; 4,6-bis-(S)-HHDP glucose [26]. The results were consistent with the naturally occurring ellagitannins, that is, the HHDP groups bridged over glucose 4,6- or 2,3-positions are almost exclusively adapted to the (S)-configuration, with only a few exceptions [27, 28]. More recently, another synthetic study of 11 using a different atropdiastereoselective biaryl coupling strategy was reported (Scheme 17.4) [29]. Su et al. employed an organocuprate oxidative intramolecular biaryl-bond forming reaction to a halogenated benzoyl derivative (9). Treatment of 9 with isopropylmagnesium bromide, followed by transmetalation with CuBrSMe2 and subsequent cuprate oxidation afforded 10 [30]. The reaction proceeded with complete diastereoselectivity without any dimeric side products. Developments in chromatographic and spectroscopic methods during the 1970–1980s expanded the structural and chemotaxonomical knowledge of natural ellagitannins [10]. The stereochemical regularity found in the structures led to a biogenetical postulation that the regulation of the HHDP atropisomerism in natural ellagitannins reflects the conformational preferences of pentagalloylglucose (2), a pivotal biosynthetic precursor. In the case of 4,6- and 2,3-HHDP glucopyranoses, the postulation was supported by molecular mechanics based conformation searches [24, 28], and it was further substantiated by the chemical synthesis achieved by the Feldman’s group. Dai and Martin also demonstrated experimental results that were in line with the postulation [31]. They examined Ullmann-type biaryl couplings between methylated galloyl esters of α-methyl aldopyranosides (Scheme 17.5). The 4,6-coupling of 12 furnished a permethylated derivative of tellimagrandin I (13) with an (S)-biphenyl bond, a result in accordance with that of Feldman and coworkers. In addition, the coupling between 2,3-cis-located galloyl groups on d-mannopyranose (14) afforded an (R)-HHDP group (15), whereas the 3,4-cis-galloyl groups on d-galactopyranose (16) yielded product 17 with an opposite configuration. Although the products 15 and 17 do not represent natural ellagitannins, these diastereoselective biaryl couplings strongly suggest the importance of conformational effects in the substrate. It may also imply that the enzyme catalyzes
17.3 Biomimetic Total Synthesis of Ellagitannins OBn
OBn OBn
Ph
O O O
BnO
O O O
BnO
I
O
3) oxidant
OBn
O I
OBn
OBn
1) i-PrMgBr 2) CuBr SMe2
O O O
Ph
OBn O
645
OBn O
N N
OBn
9
O
O O
O
O
O2N
OBn
O
NO2
OBn
BnO
BnO BnO OBn OBn
65 %
10 Pd/C 99% H , THF 2
OH HO HO O
OH O
O
OH O
O OH
HO HO
O
O
HO OH
OH
11
Scheme 17.4
Synthesis of sanguiin H5 (11).
only the oxidation of the galloyl groups in the ellagitannin biosynthesis and the stereochemical regulation of the subsequent coupling reaction occurs spontaneously in the most stable conformation. However, this hypothesis is probably not applicable to the ellagitannins with glucopyranose cores with a 1 C4 or related boat conformation [10, 17]. In the biosynthesis, it is required that the enzymes force the glucopyranose core to adopt an unstable conformation prior to the biaryl coupling. 17.3.2 Ellagitannins with 1 C4 Glucopyranose Cores
Dai and Martin [31] demonstrated lower stereoselectivity of the coupling between the 2,4-digalloyl groups in a derivative of 3,6-anhydro-d-glucose (18) adopting 1 C4 conformation (Scheme 17.5). The result indicated the greater conformational flexibility of the two galloyl groups in 18 compared to those of proximally located esters in 12, 14, and 16. In natural ellagitannins, only one example of 2,4-HHDP-glucose, phyllanemblinin B (20), was found in an extract of Phyllanthus emblica (Euphorbiaceae) and its atropisomerism was found to be the (R)-configuration [32]. The 2,4-HHDP group in ellagitannins is found as an oxidized form commonly called the dehydrohexahydroxydiphenoyl (DHHDP) ester [33, 34], such as the (S)-DHHDP group in granatin B (21) [35, 36] and the (R)-DHHDP group in geraniin (22) (Scheme 17.6) [34].
17 Biomimetic Synthesis and Related Reactions of Ellagitannins
646
OMe OMe
MeO
MeO O
MeO I O
I
4
O
O O
Cu(0)
O OMe
O
MeO
MeO MeO
6
O
MeO
O
D-glucose (4C1)
OMe
OMe
O
MeO
OMe
O
O O
O O
MeO
O OMe
OMe
50%
OMe
OMe
13 OMe
MeO
O
OMe
OMe O
I
OMe I OMe
O OBn
BnO MeO
O OMe
R
O
77%
BnO
O
D-mannose
2
O
OMe
3
O O OBn
MeO
OMe
OMe 15
14 MeO
OMe
O MeO
MeO
O OBn
MeO
OMe OMe
Cu(0) 2 3
OMe
OMe OMe
12
O
OMe
O
MeO
OMe
OMe
O
S
I
I
Cu(0)
O
4
O
MeO
69%
3 BnO
O
MeO
BnO OMe
OMe 17 D-glucose (1C4)
2
O
I
OMe OMe
OMe 18
O
Cu(0)
OMe
O
I
O O
OMe
4
O
O
MeO
16
O O O
O OBn
S O
O
D-galactose
MeO
MeO MeO
OMe
MeO OMe
O
OMe
O
O
O
90%
OMe
MeO
MeO MeO OMe OMe 19 mixture of diastereomers (73:27)
Scheme 17.5
Ullmann-type biaryl couplings between methylated galloyl groups.
HO
HO
HO
HO
R
O
O
+
25
O
OH
OH
OH
O
OH
OH
O
OH
OH
OH
OH
O
ellagic acid (1)
OH
O
O
O
O O
HO
20
OH
O O
2
OH
OH
OH
OH
DHHDP
HO
Scheme 17.6
O
O O
HO
O
O
HO
O
O O O
OH
OH
OH
CO2H
OH
O
HO
OH
HO
OH
OH
OH
OH
24
R
O
OH
21
O
O O
HO
O OH
OH
2
OH
O
OH
O
HO
HO
HO O
H S
O
OH
HO
OH
O
4
O
R O O
HO
Ellagitannins with 2,4-HHDP and 2,4-DHHDP esters.
22, 24, and 25 were isolated from Geranium sp. 23 is a hypothetical biosynthetic intermediate.
HO
O
4
OH
HO
HO
HO
OH
O
OH
OH
OH
O
S
H O
unstable
OH
HO
O HO O
HO
O
O
OH
O
HO
O
O
O
23
HO
O
O
O
OH
O
OH
OH
OH
OH
HO
O
OH
OH
OH
H2, Pd/C
OH
O
R
O
O O
OH
22
OH
O
OH
OH
O
HO
O
OH
HO
HO
HO
O
OH
RH
O
O
R O O
HO
HO
O
HO
OH
OH
OH
O
R
H
OH
O OH
17.3 Biomimetic Total Synthesis of Ellagitannins 647
17 Biomimetic Synthesis and Related Reactions of Ellagitannins
648
These ellagitannins are called dehydroellagitannins. Occurrence of both (R)- and (S)-DHHDP groups may reflect the above-mentioned low diastereoselectivity of the coupling between 2,4-digalloyl derivatives. The (R)-DHHDP group is more common in Nature, and geraniin and related ellagitannins are widely distributed in Euphorbiaceous plants [9, 10]. Catalytic hydrogenation of the (R)-DHHDP group of 22 yields the corresponding (R)-HHDP ester (23) [10b]. Although 23 has not been found in Nature, its partial hydrolysis product, geraniinic acid A (24), was found in Geranium thunbergii together with 22 [37]. The 2,4-HHDP ester attached to the 1 C4 glucopyranose is unstable and easily hydrolyzed in aqueous solution to give corilagin (25) [38] and ellagic acid (1). The stability of phyllanemblinin B (20) is apparently contradictory; however, this can be ascribed to the conformational flexibility of the molecule due to the absence of the 3,6-HHDP ester, which allows the glucopyranose to adopt a skew boat conformation and mitigate the strain of the macrocyclic ester ring [32]. Corilagin (25) has been shown to adopt a 1 C4 conformation in acetone-d6 and a skew-boat conformation in DMSO-d6 [39, 40]. A persuasive explanation for the rare occurrence of 2,4-HHDP glucoses in Nature was made by Feldman and coworkers [41]. They applied the aforementioned Pb(OAc)4 oxidative cyclization to a derivative 26 of 2,4-digalloyl 1,6-anhydroglucose and obtained a mixture of regioisomers of 2,4-HHDP glucose (27). However, subsequent deprotection by hydrogenation gave the unexpected 2,4-digalloyl product (30) (Scheme 17.7). This unusual result was explained by formation of a cyclohexadienone tautomer 28. Molecular mechanics calculations O OMe O
O OMe O
O OMe O
Pb(OAc)4
O
O
O
O
-38°C 69%
O
O
HO
H2, Pd/C
O
O
O O Ph
O
HO
O O
Ph
Ph Ph
O O
O H, CPh2
HO OH
HO
OH OH
30
O OMe O O
O O
O
O
O
O HH O O 28
O OMe O
H2,Pd/C
O O
O
HO
O
27
Ph Ph
O O
H, CPh2
26
Scheme 17.7
O
100%
O
Ph O
Ph
Ph Ph
O O
O
H
H O
H OH O H+ 29
Reductive cleavage of the biphenyl bond of a 2,4-HHDP derivative.
O Ph O
Ph
17.3 Biomimetic Total Synthesis of Ellagitannins
indicated that the non-aromatic tautomer 28 is substantially lower in energy than 2,4-HHDP-1,6-anhydroglucose 27. Reduction of the tautomer 28 afforded an intermediate 29 and subsequent elimination of the aromatic rings yielded 30. Although the biosynthetic mechanism of the DHHDP esters is unknown, this reaction mechanism strongly suggests that tautomerization between the aromatic ring and the cyclohexadienone structure explains the predominant disposition of the DHHDP esters at the glucose 2,4-positions in natural ellagitannins. The bis-cyclohexadienone structure of 28 is apparently related to the acyl group found in carpinin B (31) isolated from Carpinus japonica (Beturaceae) [42]. Hydrogenolysis of 31 yielded the DHHDP ester (32) and finally tellimagrandin I (7), indicating that the 4,6-acyl group of 31 is an oxidation metabolite of the HHDP group (Scheme 17.8). Compound 32 is a desgalloyl analog of trapain [43] (1,2,3-tri-O-galloyl-4,6-(S)-DHHDP-β-d-glucose, syn. isoterchebin [44]). In addition, the coexistence of 31 and 32 with 7 in the same plant source supports their biosynthetic relationship. The DHHDP hydrated quinone structure of 22 shows close similarity to dehydrotheasinensin A (34) produced by the enzymatic oxidation of (−)-epigallocatechin 3-O-gallate (33) during black tea fermentation (Scheme 17.9) [45]. The quinone dimer 34 is unstable and moderately undergoes oxidation–reduction dismutation O
O
O HO
HO H H
O O HO
O
HO
HO
OH
O O O O
O
O
OH
O O
O
OH
OH
O
O
OH
HO
O O O
HO HO
HO
O
S H
O HO
H2, Pd/C
O
OH OH
O
HO
OH
OH
HO
OH
31
OH
32 or Na2S2O4
H2, Pd/C
OH HO O O
HO HO
O
O O
O
O
HO
OH
O O
OH HO
OH OH
HO
OH
OH
7
Scheme 17.8
Reduction of carpinin B (31) and production of DHHDP and HHDP esters.
649
OH
OH
A
O
O
O
OH
OH
x2
HO
OH
HO
O
H
34
O
OH
O
O
HO
O
O
OH
OH
OH
OH
OH OH
OH
OH OH
OHO
O
Oxidation of epigallocatechin gallate during tea fermentation.
O
OH
OH OH
OH
G
OH OH
OH
polyphenol oxidase
O
O
33
33a
O C
O
Scheme 17.9
HO
HO
B
OH
reduction
HO
HO
HO
HO
HO
HO
OH
OH
O
O
O
O
O
35
oxidation products
O
OH
OH
OH
OH
OH OH
OH
OH
650
17 Biomimetic Synthesis and Related Reactions of Ellagitannins
17.3 Biomimetic Total Synthesis of Ellagitannins
to furnish a stable biphenyl dimer, theasinensin A (35), along with a mixture of oxidation products. Theasinensin A seems to be a final product because enzymatic oxidation of 35 did not afford 34. Despite their structural similarity, there has been no experimental evidence of similar reductive conversion of the DHHDP ester into the HHDP group (e.g., 32 to 7) in plant metabolism. Oxidation of the galloyl group of 33 scarcely occurred due to the higher redox potential of the galloyl group than that of the B-ring. Only small amounts of the trimers were produced by the oxidative coupling between the galloyl group of 35 and the B-ring quinone 33a [46, 47]. Although total synthesis of ellagitannins bearing the DHHDP esters has not been accomplished, it was reported that oxidation of methyl gallate (36) with o-chloranil produced a product 38 equivalent to an imaginary DHHDP dimethyl ester (37, 37a) (Scheme 17.10) [48]. The product 38 is formed by the intramolecular addition within the quinone dimer 37. Treatment of 38 with o-phenylenediamine and Na2 S2 O4 affords phenazine (39) and HHDP derivatives (40), respectively, which are also expected to be derived from 37a by similar treatment. 17.3.3 Synthesis of an Allagitannin with 3,6-(R)-HHDP Group
Corilagin (25) usually coexists with the dehydroellagitannins, such as 21 and 22, and may be a partial hydrolysate of the biosynthetic precursor of the dehydroellagitannins (Scheme 17.6). Total synthesis of the 3,6-(R)-HHDP glucose structure was achieved by a biaryl coupling methodology; however, the strategy is different from those of preceding studies. Yamada et al. [49] first cleaved the pyranose ring of a protected glucopyranose 41 by Wittig olefination at the anomeric position to give 42. Partially protected galloyl groups were then introduced at the 3,6-positions (43) and CuCl2 -amine mediated oxidative coupling between the acyl groups furnished the desired 3,6-bridged macrocyclic bi-ester 44 with an (R)-biphenyl bond. Subsequent regeneration of the aldehyde at the anomeric position by oxidative cleavage of the olefinic bond formed the glucopyranose ring (46). Finally, acylation at the anomeric position and deprotection yielded 25 (Scheme 17.11). This is the first success in constructing a 3,6-bridged (R)-HHDP glucose by the biphenyl coupling of non-proximal galloyl groups. 17.3.4 Synthesis of Ellagitannins by Double Esterification of Hexahydroxydiphenic Acid
The aforementioned biaryl coupling methodology is ‘‘biomimetic’’ in the sense that the key reaction is apparently the same as the plant biosynthesis. In addition, there is another successful strategy, in which firstly a protected hexahydroxydiphenic acid is prepared; secondly, the dicarboxylic acid is docked onto a diol derivative of the glucopyranose; lastly, the protecting groups are removed. Nelson and Meyers synthesized an enantiomerically pure HHDP derivative (50) from bromide 48 by
651
36
OH
OH
Scheme 17.10
HO
CO2Me
37
HO O HO
O
O O
37a
OH
OH
OH
CO2Me
+ H2O
OH O
CO2Me H
O
MeO2C MeO2C
N
38
N
Na2S2O4
OH OH
CO2Me
o-phenylenediamine
HO OH
39
O OH
O
64 %
OH O
MeO2C
O
MeO2C MeO2C
Synthesis of dehydrohexahydroxydiphenic acid esters from methyl gallate.
o-chloranil
OH
HO HO
40
OH
CO2Me
HO OH
MeO2C
OH
652
17 Biomimetic Synthesis and Related Reactions of Ellagitannins
HO
O
COCl
(α/β = 33/67)
DMAP, CH2Cl2, 73%
BnO
BnO
BnO
HO
BnO
Total synthesis of corilagin (25).
46
OH
OBn
OBn OBn
43
O O
O
O
OBn
76%
CuCl2 n-BuNH2 MeOH
2) NaH, PMBCl 3) HCl, 65%
1 )Ph3P=CH2, THF, 82 %
OPMB
O
O
OH
OBn OBn
BnO
O
OBn
O
BnO O
Scheme 17.11
BnO
BnO
BnO
HO
BnO
HO
41
HO
BnO MOMO
OMOM
42
BnO
44
MOMO
BnO
47
OBn OBn
OBn
OBn OBn
2) BnBr, 91%
1) K2CO3 acetone
OBn
OBn
OBn
BnO
BnO
45
90%
HO
OPMB
O
OBn OBn
O
O
HO
OBn
OBn OBn
H2, Pd(OH)2/C
O
BnO
100%
CO2H, EDCI, DMAP, CH2Cl2
2) conc. HCl, i-PrOH/TH, 86%
1)
O O O O O O O
BnO
OBn OBn
OPMB
OH
OBn
OBn
OPMB
O
OH
O
O
R
BnO
O
HO
BnO HO
OH
MOMO
OH
O
OH
O
OH
O
OH 25
O
OH
O
O O
HO
2) cat. OsO4, NaIO4, 69%
1) DDQ, CH2Cl2, 64%
OH
OH OH
17.3 Biomimetic Total Synthesis of Ellagitannins 653
654
17 Biomimetic Synthesis and Related Reactions of Ellagitannins OMe
S
OMe
MeO MeO
O
N Br
Cu Pyr DMF, reflux
N
MeO O O
S
60%
MeO
OMe
MeO
N
OMe
1) TFA, H2O 2) Ac2O, pyr.
MeO MeO
3) KOt-Bu, H2O, THF
MeO
quantitative
MeO
CO2H CO2H
S
OMe
OMe 48
50
49
HO O HO O TMB O OMe TMB
DCC, DMAP 38%
TMB = 3,4,5-(MeO)3C6H2CO-
51
OMe MeO O MeO MeO
O
MeO
O
O O
O O OMe
O OMe O
MeO
OMe OMe
OMe OMe
OMe
13
Scheme 17.12
Synthesis of a permethylated derivative of tellimagrandin I (13).
oxazoline-mediated asymmetric Ullmann coupling (Scheme 17.12) [50]. Further coupling of 50 with diol 51 furnished a permethylated derivative of tellimagrandin I (13). Lipshutz et al. synthesized enantiopure 50 in a different manner (Scheme 17.13) [51]. The aryl bromide precursor 54 was prepared by coupling of bromide 52 with (S,S)-stilbene diol 53. Dilithiation of 54 and subsequent treatment with CuCN forms diarylcuprate 55. Oxidation of 55 with molecular oxygen yielded a single diastereomer (56). Finally, hydrogenation and oxidation produced (S)-50. The acid 50 was used for the synthesis of a permethyl derivative of tellimagrandin II (3). The synthesis using the enantiomerically pure HHDP acid was followed by a double esterification strategy using ‘‘racemic’’ HHDP dicarboxylic acid. Itoh et al. demonstrated diastereoselective esterification of racemic biaryl acid chloride 57 with a 2,3-diol of the glucose derivative (58) [52]. The diastereoselectivity was dependent on the base and solvent. A NaH–toluene combination predominantly afforded 2,3-(R)-diastereomer 60, which is thermodynamically less stable (Scheme 17.14). It was shown that the esterification occurred first at the C2 position (59), and the product 60 was produced by kinetically-controlled intramolecular ester cyclization.
17.3 Biomimetic Total Synthesis of Ellagitannins MeO HO
OMe
MeO
Br HO
Br
Ph
53
MeO
OMe
Ph
OMe
O
Br
NaH 97%
Ph
1) t-BuLi, THF, −78 °C 2) CuCN
O
Br
Ph
MeO MeO
O
Ph
O
Ph
Li2CuCN MeO
OMe MeO
MeO
52
OMe MeO
OMe
55
54 3 O2 −78 °C
77%
OMe MeO MeO MeO
O
Ph
O
Ph
MeO OMe 56 1) H2, Pd/C 2) KMnO4, 99% 87%
50
Scheme 17.13
Preparation of (S)-hexamethoxydiphenic acid by Lipshutz et al.
In contrast, when Et3 N was used as the base in THF, the (S)-diastereomer was mainly produced (S : R = 4.4 : 1). The results were applied to the synthesis of a permethyl derivative (61) of pedunculagin [2,3-; 4,6-bis-(S)-HHDP glucose] [53]. The total synthesis of the corresponding unprotected ellagitannin is based on the double esterification strategy established by Khanbabaee et al. [54]. The synthesis used racemic HHDP dicarboxylic acid and succeeded in complete diastereoselective esterification with a glucose 4,6-diol derivative; that is, esterification of the hexabenzyl HHDP acid (62) to the 4,6-diols of a partially protected glucose (63) yielded 64 with (S)-chirality. Subsequent deprotection and acylation at the anomeric position followed by hydrogenolysis of the benzyl protective groups furnished the natural ellagitannin strictinin (65) (Scheme 17.15). Esterification of the same racemic dicarboxylic acid (62) to a 2,3-diol derivative of glucose (66) afforded an almost equal amount of diastereomers (S)-67 and (R)-67, which were separated using silica-gel chromatography (Scheme 17.15). Total synthesis of two naturally occurring ellagitannins, praecoxin B [2,3-(S)-HHDP-4,6digalloyl-glucose] and pterocarinin C [1,4,6-trigalloyl-2,3-(S)-HHDP-β-glucose], was accomplished from (S)-67 [55]. In addition, unnatural ellagitannin analogs, that is, 4,6-digalloyl- and 1(β),4,6-trigalloyl-2,3-(R)-HHDP-β-glucoses, were synthesized
655
Scheme 17.14
57
racemic
+
Ph
O O HO
58
HO
THF
Et3N
OMe
O
O
Ph
S
O O O
O
O O HO O O OMe
OMe
O
O O OMe
OMe
O
59
(S)-60 (S:R = 4.4:1)
2) (S)-57, Et3N, toluene
1) HCl, MeOH
MeO MeO OMe OMe
MeO
Cl
Ph
MeO MeO OMe OMe
MeO
PhCH3
NaH
Kinetic resolution of HHDP derivatives.
COCl
MeO
COCl
MeO
OMe
OMe
MeO
MeO
MeO
MeO MeO
MeO
O
R
O O O O
O OMe
OMe
O
O O
O
S
O O
O
O O
OMe
OMe
O
61
MeO MeO OMe OMe
MeO
OMe
S
OMe
(R)-60
MeO MeO OMe OMe
MeO
Ph
656
17 Biomimetic Synthesis and Related Reactions of Ellagitannins
OBn
Scheme 17.15
62
racemic
63
OBn
O O
O 2N
O O HO
66
HO
O O
O 2N
24%
DCC, DMAP, DMAP·HCl, CH2Cl2
HO HO BnO
O OBn
O
O2N
BnO
+
COCl
O
Ph
82%
BnO
3) H2, Pd/C 69%
BnO
BnO
2)
O O O
HO
HO HO
R
O O
O
OH
OBn
O
O2N
HO
65
HO O O O HO
(R)-67 35%
O
O
O
1) h ν 86%
O
(S)-67 34%
S
O O O
64
O O O BnO BnO
O
OH
BnO BnO OBn OBn
O
Ph
OBn
S
O
NO2
HO
BnO BnO OBn OBn
BnO
BnO
BnO BnO
BnO
OBn
Synthesis of strictinin (65) and diastereomers of 2,3-HHDP-glucoses.
DCC, DMAP, CH2Cl2
Ph
CO2H
BnO
BnO
CO2H
BnO
BnO
OBn
OH
O
O
O
OH
17.3 Biomimetic Total Synthesis of Ellagitannins 657
658
17 Biomimetic Synthesis and Related Reactions of Ellagitannins
from diastereomer (R)-67 [56]. Furthermore, Khanbabaee et al. applied the methodology to the synthesis of gemin D [3-O-galloyl-4,6-(S)-HHDP-d-glucose], hippomenin A [2-O-galloyl-4,6-(S)-HHDP-d-glucose] [57], and a synthetic precursor of 5-O-galloyl-4,6-(S)-HHDP-d-gluconic acid [58]. A simple total synthesis was demonstrated by Khanbabaee et al. (Scheme 17.16). The 2,3,4,6-tetrahydroxy derivative of glucose (68) was directly esterified with enantiomerically pure (S)-62 to give 2,3;4,6-bisdiphenoyl esters in good yields (60%). Undesired products (35%) were shown to be produced by an intermolecular esterification. Subsequent removal of the protecting groups furnished the natural ellagitannin pedunculagin (69) [59]. OH HO
O
OBn BnO BnO BnO
S
COOH COOH
HO
+ HO
HO
O 2N O OH
O
HO HO
O
S
1) DCC, DMAP, CH2Cl3 60% 2) H2 Pd/C 67%
O
HO OH
O O
O
68
(S)-62
S
HO HO
OH O
BnO OBn
O O
HO OH
OH OH
69
Scheme 17.16
Synthesis of pedunculagin (69).
17.3.5 Biomimetic Synthesis of Dimeric Ellagitannin
There is a distinctive class of ellagitannins called oligomeric ellagitannins, which are produced by intermolecular oxidative couplings of simple HHDP esters of glucopyranoses [9, 11]. Besides a few C–C linked dimers [60], the oligomers are produced by intermolecular C–O couplings between two pyrogallol rings of galloyl or HHDP esters, as exemplified by cornusiin E (4) (Scheme 17.2) [15]. As the sole success of ellagitannin total synthesis of this class, Feldman et al. [23, 61] reported the synthesis of coriariin A (70), which is biosynthesized by oxidative coupling of two tellimagrandin II molecules (3) (Figure 17.1) [62]. From a different viewpoint, the molecule can be deemed to be an ester composed of dehydrodigallic acid and two tellimagrandin I units. The synthetic study started from the preparation of the dehydrodigallic acid unit. The authors applied a hetero-Diels–Alder cycloaddition/reductive rearrangement sequence to the synthesis of the dehydrodigalloyl unit (Scheme 17.17) [63]. The o-quinone 71 was treated with a Lewis acid to give Diels–Alder adducts (72a and 72b) and subsequent elimination with NaOAc/HOAc to give a mixture of o-quinones (73a and 73b). Reduction of the quinones with hydrosulfite yielded a mixture of meta- and para-isomers 74a and 74b of dehydrodigallic acid. Upon the
17.4 Conversion of Dehydroellagitannins into Related Ellagitannins
HO
tellimagrandin II
tellimagrandin II
OH
OH OH
O
HO
OH HO
OH
HO HO
OH
O O
HO OH
O
O O O
C O
O
O
O
O O C
HO
O
HO HO OH
OH
O
OH OH
OH
OH OH
O
O O O
HO
HO
O O
OH OH
O
O dehydrodigalloyl unit
tellimagrandin I
OH OH
tellimagrandin I coriariin A (70)
Figure 17.1
Structure of coriariin A.
usual benzylation, the para-isomer 74b underwent a Smiles rearrangement and the meta-derivative 75 was obtained as the sole product. Unfortunately, direct application of this methodology to the synthesis of 70 by dimerization of the protected tellimagrandin II units failed due to the instability of the glucose-linked o-quinone derivative corresponding to 71. Therefore, an alternative three-component coupling strategy was adopted to synthesize 70 (Scheme 17.18) [61a,b]. Firstly, two equivalents of protected glucose 77 were connected to dicarboxylic acid 76. Secondly, benzyl-protected galloyl residues were introduced at the C2 and C3 positions of the glucose residue, and lastly the (S)-HHDP esters at the 4,6-positions were constructed according to the Feldman’s diastereomeric coupling method (Scheme 17.18).
17.4 Conversion of Dehydroellagitannins into Related Ellagitannins 17.4.1 Reduction of DHHDP Esters
As mentioned before, dehydroellagitannins are characterized by the DHHDP esters derived from HHDP groups (Scheme 17.6). These tannins are widely distributed in the plant families of Euphorbiaceae, Geraniaceae, Elaeocarpaceae, Aceraceae, Combretaceae, Punicaceae, Trapaceae, and Leguminosae [9], and particular tannins are important as constituents of oriental medicinal plants. The most typical and widely distributed dehydroellagitannin is geraniin (22); this tannin is sometimes accompanied by structurally related tannins probably produced by further metabolism of the DHHDP group. Some ellagitannins are chemically derived from 22. Heating of 22 with pyridine in CH3 CN [36] yielded mallotusinin (81) [64], 1-O-galloyl-2,4; 3,6-bis-(R)-HHDP-β-d-glucose (23), and acalyphidin M1 (82) [65]
659
74a
OBn
+
72b
O
O
+
72a
O
74b
OH OH
H
O
OBn
OBn
OH
O
CO2Bn
CO2Bn
OBn
O
CO2Bn
BnO
BnO2C
BnO
BnO2C
BnO
O
H O O
Synthesis of benzyl-protected dehydrodigallic acid.
BnO2C
OH
OH
HO
O
CO2Bn
O
B(OAc)3 CHCl3, 58 °C
BnO
71
O
Scheme 17.17
BnO
CO2Bn
BnO2C
O
O
O
75
+
OBn
OBn
73b
44% from 71
OBn
O
OH O
CO2Bn
BnO2C
BnO
CO2Bn
OBn
O
CO2Bn
BnO2C
BnO
BnO
73a
BnO2C
HO
(Smiles rearrangement)
BnCl, K2CO3
NaOAc HOAc
BnO
OBn
O
Na2S2O4
660
17 Biomimetic Synthesis and Related Reactions of Ellagitannins
17.4 Conversion of Dehydroellagitannins into Related Ellagitannins
661
Ph 1) TBAF, HOAc, 75% O O TBSO
Ph
OBn
HO2C
O TBSO
OBn
O TBSO O O
BnO
CO2H
BnO
CO2H
OBn
O
BnO
CCl3
77
O
OBn
TBSO
NH
BnO
2)
O
C O BnO
benzene, reflux 67%
O
3) I2, CH3OH, 77%
O OTBS
O
BnO
BnO
DCC, DMAP, 97%
O C
BnO
BnO 76
O 78
O
OTBS
Ph 1)
BnO
OBn
OBn O
HO O
O
OBn OBn
OBn
HO C O O BnO
O
O
O C
OBn BnO
BnO
OBn O OH
O
CO2H
DCC, DMAP, 87% 2) TBAF, HOAc, 87%
OH
BnO BnO
O
OBn
O O
O
O
TBSO
O
O
Ph
Ph
BnO
OBn BnO
79
Ph
Ph
O O BnO
Ph
O
HO O
O
O O
O
O
O HO
BnO
O
BnO OBn
OBn OBn
OBn
O
Ph
OBn
OBn
O
C O O BnO
O
O O C
O
OBn
O
O
O BnO
−40 °C, 74% 2) H2, Pd/C, 80%
O Ph
O O
O
Ph
O
BnO 80
Scheme 17.18
1) Pb(OAc)4,
OBn OH
O O
BnO
BnO
OBn
O
HO
O
Ph Ph
Total synthesis of dimeric ellagitannin coriariin A.
(Scheme 17.19), of which 81 and 82 were isolated from Euphorbia species. In this experiment, the production of 81 and 23 indicated the occurrence of reduction/oxidation dismutation; however, the oxidation products could not be identified. This reaction resembles decomposition of the quinone dimer 34 during production of black tea polyphenols (Scheme 17.9). A similar reductive reaction of quinone derivatives was also reported by Feldman et al. [63]. Treatment of a mixture of Diels–Alder adducts of gallic acid quinone 83 with DBU (1,8-diazabicyclo [5.4.0]undec-7-ene) afforded dehydrodigallic acid derivative 84 without the
coriariin A (70)
17 Biomimetic Synthesis and Related Reactions of Ellagitannins
662
R O
R H
O
R
HO OH O
O HO
OH
O
O pyridine
OH
CH3CN, 80 °C
O
R O
OH + HO
HO HO
O
O
OH
O
OH HO
HO
OH
OH
H HO
O HO
81 (35%)
OH OH
OH O
23 (2%)
+
Isolated from Euphorbia sp.
R
DHHDP R = 1-galloyl-3, 6-(R )-HHDP-β-D-glucose
O
22
HO
OH OH
O
O
O 82 (4%) Isolated from Euphorbia sp.
Scheme 17.19
Production of mallotusinin (81) and acalyphidin M1 (82) from geraniin (22).
CO2Me
MeO
O O 83
benzene reflux MeO2C
O O MeO
HO O CO2Me
MeO2C
OMe MeO
+ MeO O O O
MeO
Scheme 17.20
H
O
MeO2C
O
CO2Me
O H
O
MeO2C
MeO
CO2Me
H
OMe CO2Me O
O DBU
O CO2Me
O MeO O
20%
MeO
OMe
OH CO2Me
OH (no external reductant)
OH
84
O
Synthesis of the dehydrodigallic acid derivative 84.
requirement of an external reducing agent (Scheme 17.20). These results indicate the importance of redox dismutation in the reactions of non-protected derivatives. 17.4.2 Reaction with Thiol Compounds and the Biomimetic Synthesis of Chebulagic Acid
Thiol compounds readily react with DHHDP esters in aqueous solution at room temperature. Reaction of geraniin (22) with l-cysteine methyl ester afforded
17.5 Reactions of C-Glycosidic Ellagitannins
adduct 85, which retains the cyclohexene ring [66]. In contrast, cleavage of the cyclohexenone ring structure occurred on reaction with glutathione, a physiologically important tripeptide (Scheme 17.21) [67]. The products were determined to be sulfide compounds 86 and 87, and an α-ketocarboxylic acid 88. Reductive desulfurization of products 86 and 87 yielded natural ellagitannins chebulagic acid (89) and neochebulagic acid (90), respectively [68]. The reactions with thiol compounds may have some physiological significance because the thiol groups of proteins and peptides play important roles in biological systems. The structure of 88 resembles the natural ellagitannin euphormisin M2 (91) [69]. The production mechanism of 88 was revealed by the reaction of 22 with N-acetyl-l-cysteine, which afforded an isolable intermediate 92. This intermediate was decomposed to give 90 through the carboxylic acid 93 (Scheme 17.22) [67]. The reaction may be related to the biosynthesis of 91. 17.4.3 Other Reactions of DHHDP Esters
Other semi-synthetic reactions of the DHHDP esters have been reported. Brief treatment of 22 with aqueous sodium hydroxide (pH 9) yielded repandusinic acid A (94) [70] along with acalyphidin M1 (82) (Scheme 17.23) [36]. More interestingly, simple 1,4-addition of ascorbic acid at the α,β-unsaturated carbonyl group occurs in aqueous solution to give elaeocarpusin (95) [71], which is often found in plants together with 22. The reaction is reversible [38] and, probably, 22 and 95 exist as an equilibrium mixture in plant cells. Phylanthusiin D (96) [72], a simple 1,2-adduct of acetone, was shown to be prepared in aqueous acetone solution in the presence of ammonium formate or ammonium acetate [73]. This reaction serves as a method to prepare stable derivatives of dehydroellagitannins. This is because the equilibrium between six- and five-membered ring hemiacetal structures hampers the interpretation of spectroscopic data.
17.5 Reactions of C-Glycosidic Ellagitannins
C-Glycosidic ellagitannins with an open-chain glucose core are distributed in plant families of Fagaceae, Myrtaceae, Combretaceae, Casuarinaceae, and Rosaceae. Although biosynthesis remains unresolved, the close biogenetical relationship with of 2,3;4,6-bis-(S)-HHDP glucoses is apparent from the structural similarities and their co-occurrence in plants. The most important C-glycosidic ellagitannins are probably vescalagin (99) and castalagin (100) found in oak wood [74]. Because the barrels used for the aging of wine, brandy, and whisky are made from oak wood the tannins and their degradation products contribute to the taste and flavor of the liquors [75]. A plausible biogenesis of these tannins through aldehydes 97 and 98 is shown in Scheme 17.24 [76, 77]. Precursor 98 was isolated from Liquidambar formosana [76] and galls of Carpinus tschonoskii [78].
663
HO
R
H
H
H
O H
R
Scheme 17.21
OH OH
O
SO
O
O
HO2C H HO
O
89
H O
H
R
R
OH OH
O
Raney Ni
H O
H
86 (30%)
O
HO2C H H HO NH2
H
NH O
OH
OH
SO
O
O
O O 90
H
HO H
R
OH
OH
O
OO
HO2C O
+ HO2C
O
91
O
H
R
O
OH OH
O
OH OH 88 (29%)
O
H
R
R = 1-galloyl-3,6-(R )-HHDP-β-D-glucose
OH
OH
Raney Ni
87 (8%)
O
HO H
R
HO2C H H HO C H 2 NH2 O
NH
HO2C H HO2C
HO2C
+
O
H HN
HO2C
Semi-synthesis of chebulagic acid (89) and neochebulagic acid (90) from geraniin (22).
85
O HO O
S
r.t.
HO2C
HN
HO2C
glutathione O
NH2
SH
OH
OH
OH
O
OH
O
MeO2C
22
OH O
HO
H
O
OH
R H
MeO2C HN
HO
O
O
O HO
O
R
664
17 Biomimetic Synthesis and Related Reactions of Ellagitannins
17.5 Reactions of C-Glycosidic Ellagitannins R
665
R
SH H MeO2C
NHAc
22 H2O (pH 6.5) 10%
H AcHN HO2C H
CO2H
SO O
O
pH 7
H
H AcHN
OH
S NHAc HO
O OH
CO2H
OH
HO
O
O OH
OH
Production mechanism of the unnatural ellagitannin 88 from 22.
HO H
R
O
O
OH OH
HO2C
88 (80% from 92)
OH
93
R
HO2C
H
HO2C
92
Scheme 17.22
O
SO
OH− 17% (oxidation)
R H
HO HO
OH O
O
HO
O OH O
O H
O OH
O
O O
OH
O OHOH
(1, 4-addition)
OH
O
O
H HO OH ascorbic acid
O
HO O HO
R
H
94
OH
95
O
R = 1-galloyl-3,6-(R )-HHDP-β-D-glucose
H
O
HO
HO
OH O
R
R
acetone aq. HCO2NH4 87%
O OH
(1, 2-addition)
OH
22
O O HO
HO
O H OH O
OH
O 96
Scheme 17.23 Production of repandusinic acid A (94), elaeocarpusin (95), and phyllanthusiin D (96) from geraniin (22).
17.5.1 Conversion between Pyranose-Type Ellagitannins and C-Glycosidic Ellagitannins
Conversion of pyranose-type ellagitannin pedunculagin (69) into the corresponding C-glycosidic ellagitannins, 5-desgalloyl stachyurin (101) and casuariin (102), occurs in aqueous solutions at pH 7.5 [26]. The reverse reaction, C-glycosidic ellagitannin to pyranose-type ellagitannin, was observed upon acid hydrolysis of a punicacortein C (103) [79] to produce punicalagin (104) [80] (Scheme 17.25). 17.5.2 Reaction at the C1 Positions of C-Glycosidic Ellagitannins
In the center of the heartwood of Castanea crenata, degradation products of C-glycosidic ellagitannins, castacrenins A (106), B (107), and C (108), were detected. These products are produced from castalin (105) under acidic conditions by intramolecular cyclization or dehydration at the glucose C1 position (Scheme 17.26) [81].
HO
HO
OH O
O
O
Scheme 17.24
HO
HO HO
HO
OH
O
OH
OH
OH
O
O
HO
HO HO
HO
OH
OH H
97
CHO O H O H OO H OH O CH2
O
O
OH
O
OH OH
OH
OH
Possible biogenesis of vescalagin (99) and castalagin (100).
69
HO OH
O O
O
HO
HO
HO HO
HO
HO
HO HO
HO
OH
OH
H
R1
OH
O
OH
OH
OH
OH
OH
OH
OH
OH OH
OH
OH
OH
OH OH
O O H O H OO O H O O CH2 O
O
HO R2
O
OH
99: R1 = H, R2 = OH 100: R1 = OH, R2 = H
OH
OH
98
CHO O H O H O H OO H O O CH2 O
HO
666
17 Biomimetic Synthesis and Related Reactions of Ellagitannins
17.5 Reactions of C-Glycosidic Ellagitannins OH
OH
OH HO
HO
O O O
O
HO
HO HO
O OH
75 °C, 2.5 h
O
OH HO
HO
HO OH
O
O
HO
69
O
OH
HO OH
OH
HO OH
OH
102 (6%) OH HO
O
O
HO O
HO O O
OH
O HO O
HO
O
O
O
(7%)
H
O
OH
O O
O
O HO
OH
O
OH
O
HO HO HO OH
O
O O
OH
HO HO
OH
HO HO
O
O
OH
O 1% H2SO4
OH OH
O O
OH
HO
H OH
HO
OH
HO
O O
HO
101 (34%)
HO
OH OH
O O
O
OH
HO
OH
+
OH
O
HO
OH
HO
O
OH
OH H
O O
O
O
HO HO
O
pH 7.5
O
O
HO
O
O
HO HO
667
OH
HO HO OH
OH
103
OH
104
Scheme 17.25 Conversion between pyranose-type ellagitannins and C-glycosidic ellagitannins. OH
OH
H O O H O HO H H O O HO CH2 O
OH
OH
OH OH
HO HO H
OH
HOH2C 1M HCl 80 °C
HO H O
HO H HO
OH
OH OH
OH H
O
OH +
OO
O
OH
O
OH
HO
HO
H OH
O
OO
O O
O
O
HO OH
OH 105
HOH2C
OH
HO OH
OH
106
107
+ OH
HO HOH2C
OH OH
OH
OH O
HO
OO
OH
O O
O
HO OH 108
Scheme 17.26
Conversion of castalin (105) into castacrenins.
17 Biomimetic Synthesis and Related Reactions of Ellagitannins
668
HO HO
HO
OH
HO HO H
O
H H
C3 benzylic cation
OH H O H O O H O O H OH O O O CH2 O
HO HO HO
109
OH
O OH
HO OH
HO
OH
HO HO
OH OH
HO
OH
HO HO HO
O
HO
O
HO H O H O O H O O H OH O O O CH2 O
O
OCH3
110
OH
HO HO H
HO OH OH
OH
H2O-TFA (100:1) 20 mg/0.2 ml
HO
HO
OH
OH
O
OH HO
glutathione
HO
HO HO HO OH
113
H2N
OH OH OH OH OH
O H
H HN N
OH OH OH OH
CO2H
H
O HO S
O
H H
OH OH OH
OH
OH OH OH OH
OH
H H
111
HO2C
OH
OH H O H O O H O O H OH O O O CH2 O
OH
O O H O O H OH O O O CH2 O
HO HO
OCH3 OH OH OH
OH
H
OH HO
OH
TFA , THF or wine model solution
OH OH
HO
37°C, 11 days 10%
OCH3 OH
O HO
OH OH
OH
OH
O
OH
O O H O O H OH O O O CH2 O
HO HO
OH OH
O
oenin Glc
OH
OH
0.1 M AcOH
O-Glc OH
OH
OH
OCH3 OH HO
O O H O O H OH O O O CH2 O
ascorbic acid
99
OH
O
OH
H H
H
p -TsOH/ dioxane
HO
OH OH
HO OH
HO
H
OH
(+)-catechin
O OH HO
HO OH
HO
OH
HO
C3
OH
OH
OH
HO
HO
OH HO
O O
100 O OH
OH OH OH OH
O HO
HO
OH
OH
OH
OH
O
H
99
HO
H hydrogen O H bonding H
O
O C3
H HO O
OH
HO
OH
O O H O O H OH O O O CH2 O
HO HO HO
OH OH OH OH
OH 112
17.5 Reactions of C-Glycosidic Ellagitannins
A group of natural ellagitannins are produced from 99 and 100 by substitution of the C1 hydroxyl group with various nucleophilic compounds, and semi-synthetic studies have been demonstrated (Scheme 17.27). Substitution with (+)-catechin and ascorbic acid afforded acutissimin A (109) [82] and grandinin (110) [83], respectively, which are found in some Fagaceous plants together with 99 and 100. Dimerization of 99 under acidic conditions produces roburin A (111) [84]. The high substrate concentration in water (100 mg ml−1 ) was critical for the dimerization reaction, suggesting the possible involvement of hydrophobic self-association in the regioselective coupling. Roburin A and related oligomeric C-glycosidic ellagitannins are found in the wood of Quercus robur [85] and Castanea crenata [83] and the tannins accumulate at the sapwood–heartwood transition region in high concentrations. The oligomers may be produced by non-enzymatic self-dimerization. The substitution reactions of ellagitannins in liquors leached from the oak wood barrel are also important in food chemistry. More recently, S-glutathionyl vescalagin (112) was synthesized [82b] and the presence of the product in red wine aged in oak barrels was confirmed by LC/MS [86]. Reaction with wine anthocyanidins was also expected, and malvidin-8-C-vescalagin and oenin-8-C-vescalagin (113) were synthesized [86]. Experiments commonly showed the same stereochemical features: substitution at the C1 position of 99 exclusively gave products with a vescalagin-type configuration, and this stereoselectivity also seems to be the same in Nature. This selectivity was explained by the stabilization of the C1 hydroxyl group of castalagin (100) by the formation of hydrogen bonding with an adjacent phenolic hydroxyl group [82b, 86]. 17.5.3 Oxidation of C-Glycosidic Ellagitannins
From a commercial whisky, an oxidized form of ellagitannin originating from oak wood castalagin (100) was isolated and named whiskytannin B (115) [87]. When oak wood was extracted with 60% EtOH for three months, tannin 115 was gradually generated together with 114. The reaction was accelerated by raising the pH to 7.0, and pure 99 and 100 were converted into 114 and 115, respectively, within 24 h (Scheme 17.28). The reaction involves autoxidation and subsequent rearrangement along with the addition of an ethanol molecule. Comparison of the spectral data suggested that these products are identical to previously reported autoxidation products of 99 and 100 [88]. The rate of the generation and subsequent degradation of 114 was much faster than the rates for 115; this may be the reason why only 115 was detected in the commercial whisky. The difference in stability is probably due to the formation of intramolecular hydrogen bonding between the C1 hydroxyl group and the adjacent carbonyl group. ←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− Scheme 17.27 Biomimetic synthesis of acutissimin A (109), grandinin (110), roburin A (111), S-glutathionyl vescalagin (112), and 1-deoxyvescalagin-(1β → 8)-oenin (113).
669
670
17 Biomimetic Synthesis and Related Reactions of Ellagitannins O O OH
CH3CH2OO
100
60% EtOH
H O OH
O
99 autoxidation (-H2) R1 H H
R1 O O
HO
+
O O
OH
HO HO
OH
HO
C
O CH2CH3 OH OH H
R1 H H
R2 O O O O O H O H O O CH2 O
O
OH
OH OH OH OH
OH
114: R1 = H, R2 = OH (fast) 115: R1 = OH, R2 = H (slow)
Scheme 17.28
Autoxidation of vescalagin (99) and castalagin (100) in 60% EtOH.
The ellagitannins also undergo pyrolytic degradation at the process stage of charring or toasting in barrel production, and the degradation products were shown to impact on the astringent sensations of the liquors [75, 87]. Interestingly, the pyrolysis of vescalagin (99) and castalagin (100) furnished different products, namely, deoxyvescalagin (116) and dehydrocastalagin (117), respectively (Scheme 17.29). On pyrolysis of 99, dehydration and subsequent reduction furnished 116. In contrast, oxidation of the C1 aromatic ring of 100 occurs predominantly to yield the ketone 117.
17.6 Conclusions and Perspectives
Beginning with the initial coupling between two galloyl groups in a molecule of the same biosynthetic precursor, pentagalloyl-β-d-glucose (2), diverse structures of ellagitannins are biosynthesized. The products are classified into several subgroups, such as simple HHDP esters, dehydroellagitannins, oligomeric ellagitannins, and C-glycosidic ellagitannins, and the structural differences have sometimes had chemotaxonomical significance. The structures indicate that the biosynthesis is both regio- and stereoselectively regulated by enzymes specific to each plant species. However, most of the biosynthetic mechanisms are still unresolved and many of the current biogenetic discussions are speculative. Nonetheless, biogenetic consideration and many findings observed during biomimetic total synthesis described in this chapter suggest that ellagitannin molecules are constructed as a consequence of reasonable chemical requirements. Hopefully, biomimetic synthetic studies will lead to an understanding of the mechanism for the regulation of the regio- and stereoselective intramolecular and intermolecular oxidative couplings in ellagitannin biosynthesis. Taking the distribution of these compounds in Nature and their importance in herbal or food chemistry into account, production of dehydroellagitannins with a 1 C4 -glucopyranose core
H
H
OH
OH
OH
OH
180 °C − H2 O
180 °C − H2O
H H
HO
HO
O O O
O
OH
O
H O OH
+
OH
OH
OH
O
OH
H H O O
O OH
HO H H
+
OH
Pyrolytic degradations of vescalagin and castalagin.
OH OH
OH OH
100
O
O
OH
99
O
O
Scheme 17.29
H
O O
O O
HO H HO
H
HO HO H
OH
HO
HO
+H2 O
OH
OH
O
HO
HO HO
HO
HO
HO HO
HO
OH
OH
OH
OH
OH
OH
117
H O O H O O H OH O O O CH2 O
O
O
HO
116
H O O H O O H O O H O O CH2 O
O
HO H H
OH
OH OH
OH
OH OH
OH
OH
OH
OH
OH
OH
OH
OH
17.6 Conclusions and Perspectives 671
672
17 Biomimetic Synthesis and Related Reactions of Ellagitannins
and C-glycosidic ellagitannins may represent the most challenging organic synthesis and plant biochemistry targets. In addition, the biomimetic conversion of non-protected 2 into ellagitannins in aqueous solution should be considered in future studies. References 1. Haslam, E. (1996) J. Nat. Prod., 59, 2. 3.
4. 5. 6.
7.
8.
9. 10.
11.
205–215. Zucker, W.V. (1983) Am. Nat., 121, 335–365. Guy, K., Jaekyung, K., Klaus, H., Yanyan, C., and Xiaozhuo, C. (2007) Evid. Based Complement Altern. Med., 4, 401–407. Okuda, T. (2005) Phytochemistry, 66, 2012–2031. Heber, D. (2008) Cancer Lett., 269, 262–268. (a) Feldman, K.S. (2005) Phytochemistry, 66, 1984–2000; (b) Coca, A., Feldman, K.S., and Lawlor, M.D. (2009) in Chemistry and Biology of Ellagitannins, an Underestimated Class of Bioactive Plant Polyphenols (ed. S. Quideau), World Scientific, Hackensack NJ, pp. 203–272. Nonaka, G., Nishimura, H., and Nishioka, I. (1985) J. Chem. Soc., Perkin Trans. 1, 163–172. Porter, L.J. (1988) in Flavonoids (ed. J.B. Harborne), Chapman & Hall, London, pp. 21–62. Okuda, T., Yoshida, T., and Hatano, T. (2000) Phytochemistry, 55, 513–529. (a) Gupta, R.K., Al-Shafi, S.M.K., Layden, K., and Haslam, E. (1982) J. Chem. Soc., Perkin Trans. 1, 2525–2534; (b) Haddock, E.A., Gupta, R.K., and Haslam, E. (1982) J. Chem. Soc., Perkin Trans. 1, 2535–2545; (c) Haslam, E. (1982) Prog. Chem. Org. Nat. Prod., 41, 1–46; (d) Haddock, E.A., Gupta, R.K., Al-Shafi, S.M.K., Layden, K., Haslam, E., and Magnolato, D. (1982) Phytochemistry, 21, 1049–1062; (e) Haslam, E. and Cai, Y. (1994) Nat. Prod. Rep., 11, 41–66. Yoshida, T., Amakura, Y., and Yoshimura, M. (2010) Int. J. Mol. Sci., 11, 79–106.
12. (a) Niemetz, R. and Gross, G.G.
13. 14.
15.
16.
17.
18.
19.
20.
21. 22.
23.
(2005) Phytochemistry, 66, 2001–2011; (b) Niemetz, R. and Gross, G.G. (2003) Phytochemistry, 64, 1197–1201; (c) Niemetz, R., Schilling, G., and Gross, G.G. (2003) Phytochemistry, 64, 109–114; (d) Niemetz, R. and Gross, G.G. (2003) Phytochemistry, 62, 301–306; (e) Niemetz, R., Gross, G.G., and Schilling, G. (2001) Chem. Commun., 35–36. Wilkins, C.K. and Bohm, B.A. (1976) Phytochemistry, 15, 211–214. Nonaka, G., Harada, M., and Nishioka, I. (1980) Chem. Pharm. Bull., 28, 685–687. Hatano, T., Yasuhara, T., and Okuda, T. (1989) Chem. Pharm. Bull., 37, 2665–2669. Tanaka, T., Nonaka, G., Ishimatsu, M., Nishioka, I., and Kouno, I. (2001) Chem. Pharm. Bull., 49, 486–487. Lee, S.-H., Tanaka, T., Nonaka, G., and Nishioka, I. (1991) Chem. Pharm. Bull., 39, 630–638. Tanaka, T., Zhang, H., Jiang, Z., and Kouno, I. (1997) Chem. Pharm. Bull., 45, 1891–1897. (a) Barbehenn, R.V., Jones, C.P., Hagerman, A.E., Karonen, M., and Salminen, J.-P. (2006) J. Chem. Ecol., 32, 2253–2267; (b) Barbehenn, R.V., Jaros, A., Lee, G., Mozola, C., Weir, Q., and Salminen, J.-P. (2009) J. Insect Physiol., 55, 297–304. Feldman, K.S., Ensel, S.M., and Minard, R.D. (1994) J. Am. Chem. Soc., 116, 1742–1745. Okuda, T., Yoshida, T., and Ashida, M. (1981) Heterocycles, 16, 1681–1685. Gramshaw, J.W., Haslam, E., Haworth, R.D., and Searle, T. (1962) J. Chem. Soc., 2944–2947. Feldman, K.S. and Sahasrabudhe, K. (1999) J. Org. Chem., 64, 209–216.
References 24. Quideau, S. and Feldman, K.S. (1996) 25. 26. 27.
28.
29.
30.
31. 32.
33.
34.
35.
36.
37.
38.
39. 40.
41.
Chem. Rev., 96, 475–503. Feldman, K.S. and Sambandam, A. (1995) J. Org. Chem., 60, 8171–8178. Feldman, K.S. and Smith, R.S. (1996) J. Org. Chem., 61, 2606–2612. Nonaka, G., Ishimatsu, M., Ageta, M., and Nishioka, I. (1989) Chem. Pharm. Bull., 37, 50–53. The structures of cercidinins A and B in this reference were later revised. See Reference 88. Tanaka, T., Kirihara, S., Nonaka, G., and Nishiokoa, I. (1993) Chem. Pharm. Bull., 41, 1708–1716. Su, X., Surry, D.S., Spandl, R.J., and Spring, D.R. (2008) Org. Lett., 10, 2593–2596. Surry, D.S., Su, X., Fox, D.J., Franckevicius, V., Macdonald, S.J.F., and Spring, D.R. (2005) Angew. Chem. Int. Ed., 44, 1870–1873. Dai, D. and Martin, O.R. (1998) J. Org. Chem., 63, 7628–7633. Zhang, Y.-J., Abe, T., Tanaka, T., Yang, C.-R., and Kouno, I. (2001) J. Nat. Prod., 64, 1527–1532. Schmidt, O.T., Schanz, R., Wurmb, R., and Groebke, W. (1967) Liebigs Ann. Chem., 706, 154–168. Okuda, T., Yoshida, T., and Hatano, T. (1982) J. Chem. Soc., Perkin Trans 1, 9–14. Okuda, T., Hatano, T., Nitta, H., and Fujii, R. (1980) Tetrahedron Lett., 21, 4361–4364. Tanaka, T., Nonaka, G., and Nishioka, I. (1990) Chem. Pharm. Bull., 38, 2424–2428. Okuda, T., Yoshida, T., Hatano, T., Ikeda, Y., Shingu, T., and Inoue, T. (1986) Chem. Pharm. Bull., 34, 4075–4082. Schmidt, O.T., Schmidt, D.M., and Herok, J. (1954) Liebigs Ann. Chem., 587, 67–74. Seikel, M.K. and Hillis, W.E. (1970) Phytochemistry, 9, 1115–1128. Jochims, J.C., Taigel, G., and Schmidt, O.T. (1968) Liebigs Ann. Chem., 717, 169–185. Feldman, K.S., Iyer, M.R., and Liu, Y. (2003) J. Org. Chem., 68, 7433–7438.
42. Nonaka, G., Mihashi, K., and
43.
44. 45.
46.
47. 48. 49.
50. 51.
52. 53.
54.
55. 56. 57.
58.
Nishioka, I. (1990) Chem. Commun., 790–791. Nonaka, G., Matsumoto, Y., Nishioka, I., Nishizawa, M., and Yamagishi, T. (1981) Chem. Pharm. Bull., 29, 1184–1187. Okuda, T., Hatano, T., and Yasui, T. (1981) Heterocycles, 16, 1321–1324. (a) Tanaka, T., Watarumi, S., Matsuo, Y., Kamei, M., and Kouno, I. (2003) Tetrahedron, 59, 7939–7947; (b) Tanaka, T., Mine, C., Watarumi, S., Fujioka, T., Mihashi, K., Zhang, Y.-J., and Kouno, I. (2002) J. Nat. Prod., 65, 1582–1587. Tanaka, T., Matsuo, Y., and Kouno, I. (2005) J. Agric. Food Chem., 53, 7571–7578. Li, Y., Tanaka, T., and Kouno, I. (2007) Phytochemistry, 68, 1081–1088. Quideau, S. and Feldman, K.S. (1997) J. Org. Chem., 62, 8809–8813. (a) Yamada, H., Nagao, K., Dokei, K., Kasai, Y., and Michihata, N. (2008) J. Am. Chem. Soc., 130, 7566–7567; (b) Ikeda, Y., Nagao, K., Tanigakiuchi, K., Tokumaru, G., Tsuchiya, H., and Yamada, H. (2004) Tetrahedron Lett., 45, 487–489. Nelson, T.D. and Meyers, A.I. (1994) J. Org. Chem., 59, 2577–2580. Lipshutz, B.H., Liu, Z.-P., and Kayser, F. (1994) Tetrahedron Lett., 35, 5567–5570. Itoh, T. and Chika, J. (1995) J. Org. Chem., 60, 4968–4969. Itoh, T., Chika, J., Shirakami, A., Ito, H., Yoshida, T., Kubo, Y., and Uenishi, J. (1996) J. Org. Chem., 61, 3700–3705. Khanbabaee, K., Schulz, C., and L¨otzerich, K. (1997) Tetrahedron Lett., 38, 1367–1368. Khanbabaee, K. and L¨otzerich, K. (1997) Liebigs Ann. Recl., 1571–1575. Khanbabaee, K. and L¨otzerich, K. (1998) J. Org. Chem., 63, 8723–8728. Khanbabaee, K., L¨otzerich, K., Borges, M., and Großer, M. (1999) J. Prakt. Chem., 341, 159–166. Khanbabaee, K. and L¨otzerich, K. (1999) Eur. J. Org. Chem., 11, 3079–3083.
673
674
17 Biomimetic Synthesis and Related Reactions of Ellagitannins 59. Khanbabaee, K. and Großer, M. (2003) 60.
61.
62.
63.
64.
65.
66.
67. 68.
69.
70.
71.
72.
Eur. J. Org. Chem., 2128–2131. (a) Jiang, Z.-H., Tanaka, T., and Kouno, I. (1995) Chem. Commun., 1467–1468; (b) Tanaka, T., Jiang, Z.-H., and Kouno, I. (1997) Chem. Pharm. Bull., 45, 1915–1921. (a) Feldman, K.S., Lawlor, M.D., and Sahasrabudhe, K. (2000) J. Org. Chem., 65, 8011–8019; (b) Feldman, K.S. and Lawlor, M.D. (2000) J. Am. Chem. Soc., 122, 7396–7397. Hatano, T., Hattori, S., and Okuda, T. (1986) Chem. Pharm. Bull., 34, 4092–4097. Feldman, K.S., Quideau, S., and Appel, H.M. (1996) J. Org. Chem., 61, 6656–6665. Saijo, R., Nonaka, G., and Nishioka, I. (1989) Chem. Pharm. Bull., 37, 2063–2070. Amakura, Y., Miyake, M., Ito, H., Murakaku, S., Araki, S., Itoh, Y., Lu, C.-F., Yang, L.-L., Yen, K.-Y., Okuda, T., and Yoshida, T. (1998) Phytochemistry, 50, 667–675. Tanaka, T., Fujisaki, H., Nonaka, G., and Nishioka, I. (1992) Heterocycles, 33, 375–383. Tanaka, T., Kouno, I., and Nonaka, G. (1996) Chem. Pharm. Bull., 44, 34–40. (a) Schmidt, O.T. and Nieswandt, W. (1950) Liebigs Ann. Chem., 568, 165–173; (b) Schmidt, O.T., Hensler, R.H., and Stephan, P. (1957) Liebigs Ann. Chem., 609, 186–191; (c) Yoshida, T., Fujii, R., and Okuda, T. (1980) Chem. Pharm. Bull., 28, 3713–3715. Yoshida, T., Amakura, Y., Liu, Y.-Z., and Okuda, T. (1994) Chem. Pharm. Bull., 42, 1803–1807. Saijo, R., Nonaka, G., and Nishioka, I. (1989) Chem. Pharm. Bull., 37, 2624–2630. Tanaka, T., Nonaka, G., Nishioka, I., Miyahara, K., and Kawasaki, T. (1986) J. Chem. Soc., Perkin Trans. 1, 369–376. (a) Yoshida, T., Itoh, H., Matsunaga, S., Tanaka, R., and Okuda, T. (1992) Chem. Pharm. Bull., 40, 53–60; (b) Foo, L.Y. and Wong, H. (1992) Phytochemistry, 31, 711–713.
73. Tanaka, T., Fujisaki, H., Nonaka, G.,
74.
75.
76.
77.
78. 79.
80.
81.
82.
83.
and Nishioka, I. (1992) Chem. Pharm. Bull., 40, 2937–2944. (a) Mayer, W., Gabler, W., Riester, A., and Korger, H. (1967) Liebigs Ann. Chem., 707, 177–181; (b) Nonaka, G., Sakai, T., Tanaka, T., Mihashi, K., and Nishioka, I. (1990) Chem. Pharm. Bull., 38, 2151–2156 (a) Glabasnia, A. and Hofmann, T. (2007) J. Agric. Food Chem., 55, 4109–4118; (b) Glabasnia, A. and Hofmann, T. (2006) J. Agric. Food Chem., 54, 3380–3390. Okuda, T., Hatano, T., Kaneda, T., Yoshizaki, M., and Shingu, T. (1987) Phytochemistry, 26, 2053–2055. (a) Vivas, N., Laguerre, M., Glories, Y., Bourgeois, G., and Vitry, C. (1995) Phytochemistry, 39, 1193–1199; (b) Quideau, S., Varadinova, T., Karagiozova, D., Jourdes, M., Pardon, P., Baudry, C., Genova, P., Diakov, T., and Petrova, R. (2004) Chem. Biodivers., 1, 247–258. Ono, T. and Shigemori, H. (2009) Heterocycles, 78, 1993–2001. Tanaka, T., Nonaka, G., and Nishioka, I. (1986) Chem. Pharm. Bull., 34, 656–663. (a) Tanaka, T., Nonaka, G., and Nishioka, I. (1986) Chem. Pharm. Bull., 34, 650–655; (b) Mayer, W., Goerner, A., and Andrae, K. (1977) Liebigs Ann. Chem., 1976–1986; (c) Schilling, G. and Schick, H. (1985) Liebigs Ann. Chem., 2240–2245. Tanaka, T., Ueda, N., Shinohara, H., Nonaka, G., Fujioka, T., Mihashi, K., and Kouno, I. (1996) Chem. Pharm. Bull., 44, 2236–2242. (a) Ishimaru, K., Nonaka, G., and Nishioka, I. (1987) Chem. Pharm. Bull., 35, 602–610; (b) Quideau, S., Jourdes, M., Lefeuvre, D., Montaudon, D., Saucier, C., Glories, Y., Pardon, P., and Pourquier, P. (2005) Chem. Eur. J., 11, 6503–6513. Nonaka, G., Ishimaru, K., Azuma, R., Ishimatsu, M., and Nishioka, I. (1989) Chem. Pharm. Bull., 37, 2071–2077.
References 84. Viriot, C., Scalbert, A., Herv´e du
Penhoat, C., and Moutounet, M. (1994) Phytochemistry, 36, 1253–1260. 85. Herve du Penhoat, C.L.M., Michon, V.M.F., Peng, S., Viriot, C., Scalbert, A., and Gage, D. (1991) J. Chem. Soc., Perkin Trans. 1, 1653–1660. 86. (a) Jourdes, M., Lefeuvre, D., and Quideau, S. (2009) in Chemistry and Biology of Ellagitannins, an Underestimated Class of Bioactive Plant Polyphenols (ed. S. Quideau), World Scientific, Hackensack, NJ, pp. 320–365;
(b) Chassaing, S., Lefeuvre, D., Jacquet, R., Jourdes, M., Ducasse, L., Galland, S., Grelard, A., C´edric Saucier, C., Teissedre, P., Dangles, O., and Quideau, S. (2010) Eur. J. Org. Chem., 55–63. 87. Fujieda, M., Tanaka, T., Suwa, Y., Koshimizu, S., and Kouno, I. (2008) J. Agric. Food Chem., 56, 7305–7310. 88. Puech, J.-L., Mertz, C., Michon, V., Le Guernev´e, C., Doco, T., and Herv´e du Penhoat, C. (1999) J. Agric. Food Chem., 47, 2060–2066.
675
677
18 Biomimetic Synthesis of Lignans Craig W. Lindsley, Corey R. Hopkins, and Gary A. Sulikowski
18.1 Introduction to Lignans
Lignans was a term coined by Haworth in 1936 to describe a group of phenylpropanoid dimeric natural products linked through the β –β (8–8 ) carbons of the propanyl side-chains [1, 2]. Biosynthetically, lignans are derived from the shikimate pathway, and these secondary metabolites from the Plant Kingdom form the basis of chemical defense in plants and have been found to possess a broad spectrum of biological activities, including anticancer, antimitotic, antiangiogenesis, and antiviral, to list a few [3–10]. In Nature, oxidases (such as laccases, peroxidases, and other oxidases) initiate a one-electron oxidation of an electron-rich cinnamyl residue 1, wherein the number and location of ethers and phenol moieties is varied and R1 can be CH2 OH, COOH, COOR, or CONHR, which ultimately generates a reactive C8 quinone methide radical 2 [3–13]. Coupling of two of these 8–8 radicals often occurs with an anti- (or trans) orientation, to diminish steric interactions, delivering the β –β coupled intermediate 3 (Scheme 18.1). Based on the functionalization of the aromatic rings (Ar and Ar ) and the oxidation state of R1 , enormous structural diversity exists for the resulting lignan natural products 4–14 (stereochemistry not shown) upon addition of oxygen nucleophiles into the quinone methide core and/or cyclization (Figure 18.1) [3–14]. Notable examples of lignans include podophyllotoxin (15), etoposide (16), (+)-pinoresinol (17), and carpanone (18) [3–15]. In 1972, Gottlieb extended this family to include phenylpropanoid dimers with linkages other than 8–8 , which were named ‘‘neolignans’’ [16]. An attempt by Gottlieb to restrict the term neolignan to natural products only derived from the oxidative dimerization of allyl and/or propenylphenols met with resistance and, in 2000, IUPAC recommended defining neolignans as dimers between two phenylpropanoid (cinnamyl residues 1) units different from 8–8 [17, 18]. Indeed, Scheme 18.1 was an oversimplification. For example, (E)-coniferyl alcohol (19), upon one-electron oxidation by endogenous oxidases (Scheme 18.2), generates the coniferyl alcohol radical 20, which subsequently exists in several mesomeric forms (the C5 radical 21, the C1 radical 22, and the C8 radical 23) [12, 13]. All of Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
18 Biomimetic Synthesis of Lignans
678
8 8
R1 oxidase
R1
8'
+
+
anti
(OR)n
(OR)n 2
1
R1 (Ar')
8 8'
(OR)n + (OR)n
(Ar) R1
R1
+
3
(OR)n
C8–C8' or b–b' coupling
Scheme 18.1 One-electron oxidation of electron-rich cinnamyl residue 1 to generate the quinone methide C8 radical 2, which undergoes a β–β (8–8 ) coupling to deliver the trans-3. Ar Ar
Ar
O
Ar
Ar'
Ar' 4
5
O
O
Ar'
Ar'
6
O
Ar'
7
Ar
O
Ar'
8
9
CO2R O
O
CO2R
CO2R
CO2R
O
Ar'
O
Ar' 10
11
12
13
14 CH3
OR
H3C
O
O O
H
O O H3CO
OCH3
HO
OH H
O
OCH3
H3CO
H
O
O O
H O
O
O
OCH3 podophyllotoxin 15, R = H CH3 O O O etoposide 16, R = OH HO
(+)-pinoresinol, 17
carpanone, 18
Figure 18.1 Representative chemotypes 4–14 of lignans derived from β–β oxidative phenolic couplings (stereochemistry not shown); structures of well-known lignans 15–18.
these mesomeric forms can combine with either themselves or with any of the other mesomeric radicals 21–23, leading to diverse neolignan chemotypes 24–29 (Figure 18.2). This chapter will focus on the biomimetic synthesis of lignans. In Nature, lignan chemotypes 4–14 are produced as either racemates (though usually a single diastereomer) or, in some cases, as single enantiomers [3–14]. In
18.1 Introduction to Lignans
OH
9
OH
OH
OH
OH
8
8
1 6
2
5
3 OCH3
4
679
oxidase 1 e− oxidation
1 5
OCH3
OH
OCH3
O
19
20
OCH3
OCH3
O
O
O
21
22
23
Scheme 18.2 One-electron oxidation of (E)-coniferyl alcohol (19) to multiple quinone methide mesomeric radicals 21–23, which combine to give rise to neolignans (coupling products other than 8–8 ).
HO OH
Pr
O Ar 5'
RO2C
8
3 3'
8
Pr
Pr Ar
25
3 O 4'
O 4'
3'
8
Pr
CO2R 24
1'
Ar
26
27 OCH3
Pr
8
Ar
Pr
Pr 28
OCH3
O HO
OH 5'
8
OH 29
(+)-dehydrodiconiferyl alcohol, 30
Figure 18.2 Representative chemotypes 24–29 of neolignans derived from oxidative phenolic couplings (stereochemistry not shown) other than 8–8 ; structure of the well-known neolignan 30.
the laboratory, biomimetic radical coupling approaches for lignan synthesis are characterized by a lack of stereochemical control, and provide racemic products (though usually a single diastereomer). This lack of stereocontrol is observed with both metal-catalyzed (AgI , MnII , IIII , CuII , CoII , FeII ) oxidative phenolic couplings as well as enzyme-mediated [horse radish peroxidase (HRP), laccases, oxidases] oxidative phenolic couplings [3–14]. In Nature, the stereochemical outcome of these radical couplings is somehow controlled to provide enantiopure lignans. In 1997, Davin and coworkers identified the source – a 78-kDa protein, devoid of catalytic activity – which they termed a ‘‘dirigent protein,’’ a protein that dictates the stereochemistry of a compound synthesized by other enzymes [19]. This so-called auxiliary ‘‘dirigent’’ protein, in the presence of either a one-electron oxidant, (NH4 )2 S2 O8 , or an oxidase, flavin mononucleotide (FMN), effected an enantioselective β –β phenolic coupling of (E)-coniferyl alcohol (19) to (+)-pinoresinol
680
18 Biomimetic Synthesis of Lignans
H3CO
H3CO OH FMN or (NH4)2S2O8 1 e- oxidation
OH
H
O
8
+
5'
O HO
H3CO
H
OCH3
4'
+
O
O
ratio of racemic OCH3 17:30:31 is 2:3:1
OH
OH
OH
OH
8
H3CO
OH
OH
OH
erythro/threo-(±)guaiacyglycerol, 31
(±)-dehydrodiconiferyl alcohol, 30
(±)-pinoresinol, 17
19
OCH3
HO
H+ O FMN or (NH4)2S2O8 1e- oxidation
dirigent protein OCH3 OH
19
8 H
re
OH O H3CO H3CO O
si
H
OH
si H
OCH3
OH
H
8
re
OH OCH3
8'
OH
OH
O
O H
8' H
putative enzyme-bound intermediate 23
Scheme 18.3 One-electron oxidation of (E)-coniferyl alcohol (19) in the absence of a dirigent protein affords racemic (±)-pinoresinol (17) along with neolignans 30 and 31 in a 2 : 3 : 1 ratio. One-electron oxidation of (E)-coniferyl alcohol (19) in the presence of a dirigent protein provides
H3CO O
OCH3 H+ 32
OH (+)-17
enantiopure (+)-pinoresinol (17) via a putative enzyme-bound intermediate wherein radicals 23 are oriented ‘‘Si-face to Si-face’’ for the oxidative phenolic coupling. Only trace amount of neolignans 30 and 31 are detected in the presence of the dirigent protein.
(17) in high yield and with only traces of other lignan and neolignan products. In contrast, without the auxiliary ‘‘dirigent’’ protein, racemic (±)-pinoresinol was obtained in both cases along with neolignans 30 (C8–C5 ) and 31 (8–O–4 ) in 2 : 3 : 1 ratio. The mechanism of action of the ‘‘dirigent’’ protein is postulated as capture of the free radical 23 into enzyme binding sites and orientation of two 23 ‘‘Si-face to Si-face’’ to direct enantioselective coupling to provide 32 and, upon conjugate addition into the quinone methide, (+)-pinoresinol 17 (Scheme 18.3) [19]. Despite the similarity in nomenclature, lignans should not be confused with lignins. Lignins are heterogeneous cell-wall biopolymers (>10 000 amu) that consist of various electron-rich cinnamyl residues 1 and account for over 30% of all non-fossil organic carbon [20]. Thus, lignins are second only to cellulose as the most abundant biopolymer and the second most abundant natural product on earth. Lignins play critical roles in conducting water in plant stems and in taking atmospheric carbon into the living plant tissues and highly lignified wood is a valuable material for multiple applications [20].
18.1 Introduction to Lignans
18.1.1 Biomimetic Synthesis of Lignans
While numerous synthetic routes have been developed for the total synthesis of lignans representing all the key chemotypes 4–14, biomimetic approaches here are restricted to oxidative phenolic coupling reactions that are either metal-catalyzed (AgI , MnII , IIII , CuII , CoII , FeII , TlIII ) or enzyme-mediated (HRP, laccases, oxidases). Of these, Ag2 O and Ag2 CO3 are the most commonly employed metal catalysts, and HRP–H2 O2 is the most common enzyme-mediated system [3–14]. 18.1.1.1 Biomimetic Synthesis of Podophyllotoxin-Like Lignans Podophyllotoxin (15) is unique as an aryltetralin lactone lignan due to its unique structure and as a precursor to clinical oncology drugs such as etoposide (16) [21]. While many synthetic strategies have been developed to circumvent the growing demand on the plant resource, few are truly biomimetic. Indeed, most synthetic routes toward 15 rely on tandem conjugate addition approaches [22–24], Diels–Alder approaches [25–27], or, more recently, enantioselective sequential conjugate addition–allylation reactions [28]. It is well established that 15 is biosynthesized by the oxidative cyclization of the dibenzylbutyrolactone yatein (33), itself a lignan derived from 8–8 heterocoupling [29]. Upon treatment of 33 with Tl2 O3 in trifluoroacetic acid (TFA) a C1 epimeric, des-4-hydroxypodophyllotoxin (34) results (Scheme 18.4). Under hypervalent iodine or ruthenium catalysis, an unexpected eight-membered ring (35) forms exclusively [30]. Thus, with the proper metal catalysis, a biomimetic approach to podophyllotoxin is possible, but other routes have proven more direct and, importantly, enantioselective [28]. O O
O O O O H3CO
OCH3 OCH3 34
Tl2O3
O O
O
42%
O H3CO
IIII or Ru 65%
OCH3 OCH3 33
O
O H3CO H3CO
OCH3
35
Scheme 18.4 Biomimetic approaches to podophyllotoxin (15). Oxidative cyclization of the biosynthetic precursor of 15, yatein (33), yields the basic epimeric core of 15 under TlIII Lewis acid catalysis; however, hypervalent iodine or ruthenium catalysis leads to the eight-membered ring congener 35.
18.1.1.2 Biomimetic Synthesis of Furofuran Lignans The furofuran class of lignans, exemplified by (+)-pinoresinol 17 discussed earlier, represents one of the largest groups of lignan chemotypes, and is characterized by a 3,7-dioxabicyclic core with diverse electron-rich aromatics in the 2- and
681
18 Biomimetic Synthesis of Lignans
682
OCH3 OCH3
O O H O
O
O H
H H3CO
O
O
OCH3 O
OCH3 H
H H3CO
O
H3CO
OCH3 H
O
H3CO OCH3
(+)-sesamin, 36
epimagnolin, 37 O
O O H O
O
O HO
H O
O
fargesin, 38
(+)-diasesamin, 39 Figure 18.3
H
OH O
O
O
O
O H3CO
OCH3 H
OH
O
HO (–)-wodeshiol, 40
(+)-epi-pinoresinolin, 41
Representative furofuran lignans 36–41.
6-positions [12, 31]. Further diversity emanates from the degree of core oxidation and the stereochemistry of pendant aryl groups (exo or endo face). Representative furofurans 36–41 (Figure 18.3) are highlighted, and several excellent reviews have been written [32–37]. This class of lignans has demonstrated a wide range of biological activities with therapeutic relevance, and is found in traditional medicines [32, 38]. Like podophyllotoxin (15) and related compounds, numerous synthetic routes have been developed to access furofurans based on acyl anion additions, aldol chemistry, cycloaddition/rearrangements, and radical/photochemistry [30–36]. Here, we discuss the biomimetic, oxidative (β –β ) dimerization of cinnamyl derivatives, which follows the synthetic route illustrated in Scheme 18.3 for the biomimetic synthesis of (+)-pinoresinol (17). Enzyme-mediated biomimetic syntheses have been reported for racemic (±)-pinoresinol (17) and (±)-syringaresinol (43, Scheme 18.5) [32, 39]. In the latter case, exposure of (E)-coniferol (19) to Caldariomyces fumago open to air for 16 h provided a 1 : 1 ratio of racemic 17 (8–8 ) and racemic 30 (8–5 ) in low yield (10%). Exposure of 42 to a crude emulsion open to air for 11 days generated racemic 43 in low yield (12%), but without the production of the neolignan 8–5 product [32, 39]. Metal-catalyzed approaches have also been employed. For example, prior to 1945, both Haworth [40] and Erdtman [41] demonstrated that treatment of ferulic acid 44 with either FeCl3 /O2 or Fe2 (SO4 )3 followed by an acidic work-up delivered the bis-lactone 45 in 20–22% yields. This powerful biomimetic sequence to arrive at the bis-lactone precursor drove furofuran chemistry, and in 1955 the first demonstration of lithium aluminium hydride (LAH) reduction of a bis-lactone 47 led to the total synthesis of (±)-43 (Scheme 18.6) [42]. Subsequently, biomimetic
18.1 Introduction to Lignans
OH
OH
H3CO
H3CO
OH
OH
H
8
O C. fumago air, 16 h 1 e− oxidation
OCH3
O
O
5'
+
H3CO
H
ratio of racemic 17:30 is 1:1
H3CO
OH
OH
HO
(±)-pinoresinol, 17
19
(±)-dehydrodiconiferyl alcohol, 30 OH
H3CO
OH crude emulsion air, 11 days 1 e− oxidation
OCH3
H O H
OCH3
H3CO
O
OH H3CO
OCH3 OH
42
(±)-syringaresinol, 43
Scheme 18.5 Enzyme-mediated, biomimetic synthesis of racemic (±)-pinoresinol (17) and (±)-syringaresinol (43).
efforts moved from the use of cinnamyl acids, such as 44 and 46, to cinnamyl alcohols, such as 42, to avoid the reduction step. In this instance, the application of CuSO4 under an O2 atmosphere in aqueous acetone perfected the reaction to deliver racemic (±)-43 in 90% yield (Scheme 18.6) [43]. In recent years, TlIII and MnIII metal species have proven equally effective in promoting high yielding β –β phenolic couplings for the biomimetic synthesis of racemic furofurans [3–14, 32]. 18.1.1.3 Biomimetic Synthesis of Benzoxanthenone Lignans The benzoxanthenone class of lignans was first discovered in 1969 by Brophy and coworkers, from the light petroleum extract of the Carpano tree, Cinnamomum sp., and is exemplified by a rigid tetracyclic core with five contiguous stereocenters [15]. To date, all benzoxanthenones identified have been isolated as single diastereomers, and produced in Nature as racemates. The first member of this class, carpanone (18) was isolated along with an ortho-methoxystyrene named carpacin (48). Brophy and coworkers then proposed a biosynthesis for capanone that involved demethylation
683
684
18 Biomimetic Synthesis of Lignans OH H3CO COOH
H O
FeCl3, O2 or Fe2(SO4)2
O
H+ work-up 20-22%
OCH3
O
O H H3CO
OH
OH ferulic acid, 44
45 OH
H3CO
OH OCH3
OCH3
H3CO
COOH H O FeCl3, O2 H+ work-up OCH3 73%
H3CO
O
H LiAlH4
O
O
THF 34%
O H
O H
OH H3CO
OCH3
H3CO
OH 46
OCH3 OH
(±)-syringaresinol, 43
47
OH
CuSO4, O2 aq. acetone OCH3
H3CO
OCH3
H3CO
OH
H+ work-up 90%
H O
O H
OH H3CO
OCH3 OH
42
(±)-syringaresinol, 43
Scheme 18.6 Metal-catalyzed, biomimetic synthesis of racemic bis-lactone 45 and (±)-syringaresinol (43).
of carpacin to produce 49, β –β phenolic coupling to deliver trans-ortho-quinone methide 50, followed by an endo-selective inverse electron demand Diels–Alder to afford carpanone (18) (Scheme 18.7). Now considered a classic in total synthesis, Chapman and coworkers validated the biosynthetic proposal with the first total synthesis of carpanone (18) [44]. Following the rationale in Scheme 18.4, Chapman decided that the β –β -phenolic coupling must occur trans and that this configuration would dictate the subsequent inverse-electron demand Diels–Alder reaction. While must phenolic couplings employed one-electron oxidants, there was precedent for two-electron oxidants; therefore, Chapman utilized PdII to bring the two styrenyl phenols 49 together, delivering 51 en route to 50, and then carpanone 18 (Scheme 18.8). Chapman’s
18.1 Introduction to Lignans
685
CH3 H3C
CH3
H
O H O
O
O
O
O O
carpanone, 18
carpacin, 48
H3C CH3 8'
8
OH
O
[O] β−β' coupling
+
O
OCH3
O
OH
O
O
49 CH3 H3C
CH3 H3C
8 8'
O
O O
O
O
[4+2]
H
O H O
O O
O
O
O
carpanone, 18
50
Scheme 18.7 Structures of carpanone (18) and carpacin (48); biosynthetic proposal by Brophy and coworkers to account for the synthesis of lignan carpanone.
H3C
H3C
CH3 O O
PdCl2, NaOAc OH MeOH, H O 2
+ OH
β-β' coupling
O O 49
Scheme 18.8
46%
CH3 H
H
HH O O O
O
O O
H3C
CH3
Pd O
O
O
O O O
51
50
The now classical biomimetic total synthesis of carpanone (18).
strategy worked, providing carpanone (18) in 46% yield as a single diastereomer, as confirmed by single-crystal X-ray [44]. From starting materials devoid of stereocenters, the one-pot construction of a tetracyclic scaffold with complete stereocontrol of five contiguous stereocenters highlights the power of biomimetic synthesis. After this initial report, several laboratories disclosed additional oxidative systems, both stoichiometric and catalytic, to produce carpanone, including metal(II) salen/O2 (metal = Co, Mn, Fe), O2 (hv, Rose Bengal), AIBN (azobisisobutyronitrile), dibenzoyl peroxide, and AgO in yields ranging from 14 to 94% [45, 46]. In 2001, Ley reported on the total synthesis of carpanone employing only solid-supported
18
686
18 Biomimetic Synthesis of Lignans CH3 H3C
H
O
O O
H O
H O
CH3
CH3
H3C
H3C H
H3CO H O
H3CO O
OCH3
sauchinone, 52
O
H
O
O
O
OCH3
H O
O
O
OCH3
polemannone A, 53
O
OCH3 O
polemannone B, 54
CH3 H3C
CH3 H
H3CO
O
H O
H3CO OMe O CH3O H C OCH3 3 polemannone C, 55
O O
OCH3 OCH3 isoapiol, 56
Figure 18.4 Representative benzoxanthone lignans 52–55, and a biosynthetic precursor to 53–55, isoapiol (56).
reagents and scavengers [47]. Around the same time, Lindsley and Shair described a hetero-β –β -phenolic coupling reaction, facilitated by IPh(OAc)2 , to deliver hetero-tetracyclic analogs of carpanone; however, this oxidant system was unable to produce carpanone itself, but it was able to produce less electron-rich homodimers [48]. Since the initial account of the discovery of carpanone (18) and the benzoxanthone class of lignans, several other congeners have been reported, including the therapeutically relevant sauchinone (52) [49], a 1 ,7 -dihydro-stereoisomer of carpanone, and the highly oxygenated polemannones A–C (53–55, respectively) [50]. Interestingly, along with 53–55, Jakupovic also isolated a more highly oxygenated congener of carpacin, isoapiol (56), suggesting a similar biosynthetic pathway (Figure 18.4). Lindsley and coworkers recently developed a novel, catalytic CuCl2 /(−)-sparteine oxidative β –β -phenolic coupling reaction of styrenyl phenols which, after a rapid inverse-electron demand Diels–Alder reaction, affords the benzoxanthanone natural product carpanone (18) and related unnatural congeners in yields exceeding 85% (Scheme 18.9) as single diastereomers [51]. Less than 5% ee was observed under these conditions, suggesting that the reaction does not take place in the copper coordination sphere due to rapid dissociation of the intermediate keto-radical leading to no enantioselection [49]. However, the catalytic CuCl2 /(−)-sparteine oxidative system enabled the first biomimetic total syntheses of polemannones B and C, 54 (79% yield) and 55 (90% yield), from presumed biosynthetic precursors 57 and 58, respectively [52]. 18.1.1.4 Biomimetic Synthesis of Benzo[kl]xanthene Lignans Benzo[kl]xanthenes such as rufescidride (59), mongolicumin A (60), and yunnaneic acid H (61, Figure 18.5) represent a small and rare class of lignans [53–55], but
18.1 Introduction to Lignans H3C
CH3
H O
O
O OCH3
HO H
O
O
OCH3
O
H3CO
O
CH3 H3C
OCH3
H3CO 10 mol % CuCl2 10 mol % (−)-sparteine 4 Å MS, MeOH,−20 °C air, 24 h
57 79%
CH3 OH OCH3 58 90%
H3C H
H3CO H3CO
H 3C
Polemannone B, 54
OMe O OCH3 CH OCH3 3 Polemannone C, 55
O O
HO
49 CH3 H3C H
O
O H O
O
O
O
Carpanone, 18
Scheme 18.9 Application of a CuCl2 /(−)-sparteine oxidative catalysts system for the total synthesis of carpanone (18), polemannone B (54) and polemannone C (55).
CO2H
O O
O HO O
CO2H
rufescidride, 59 Figure 18.5
HO O
O
OH OH
CO2H
O
O
O
O
mongolicumin A, 60
CO2H
OH
OH OH
O
H O
O
HO
687
OH yunnaneic acid H, 61
Representative benzo[kl]xanthene lignans 59–61.
recent biomimetic synthesis reports have renewed interest in them [56]. In 2009, Tringali and coworkers initiated a study of the products from the biomimetic, metal-mediated oxidative phenolic coupling of caffeic acid phenethyl ester (CAPE, 62) [56]. Treatment of CAPE (62) with MnO2 in CH2 Cl2 at room temperature for 4 h generated two lignan products, 63 (48%) and 64 (16.5%), with the remaining mass balance being un-reacted CAPE (62). This represents the first biomimetic synthesis of the benzo[kl]xanthene class of lignans (Scheme 18.10) [56]. The mechanisms illustrated in Scheme 18.11 to account for the formation of 63 and 64 differ only in the exo versus endo orientation of the meta-OCH3 group in B (exo) and B (endo). MnO2 catalyzes a one-electron oxidation and, ultimately, the C8 radical quinone methides (65). An 8–8 oxidative phenolic coupling provides regioisomers A and A . Elimination provides either B or B . A 2–7 intramolecular
688
18 Biomimetic Synthesis of Lignans
CO2R O OR
MnO2 CH2Cl2, rt, 4 h
HO
HO
CO2R
CAPE, 62
HO
O
CO2R OH
OH R = CH2CH2Ph
CO2R +
OH
OH
OH
OH
3:1 63
64
Scheme 18.10 Manganese-mediated, biomimetic synthesis of benzo[kl]xanthenes 63 and 64.
cyclization of the electron-rich aromatic into the quinone methide B accounts for the formation of 64; note that B and B are conformational isomers and rapidly interconvert. A similar 2–7 intramolecular cyclization with B leads to C . Two additional oxidation steps lead to D and E . Intramolecular nucleophilic 3–6 attack of the meta-OH into the quinone methide affords F, which can then tautomerize, accounting for the formation of 63 [56]. Interestingly, an earlier report by Lemiere and coworkers found that methyl caffeate (66) afforded predominantly 71, a 5–8 neolignan (Scheme 18.12), as opposed to the two lignans 63 and 64 produced under MnO2 catalysis [56, 57]. Mechanistically, a 5–8 phenolic coupling between 67 and 68 provides 69, which can eliminate to generate 70, followed by 4–7 intramolecular cyclization to deliver racemic 71. Tringali and coworkers then treated CAPE (62) with Ag2 O in CH2 Cl2 at room temperature for 2 h and similarly formed the analogous 8–5 product 71. To determine if the Mn2+ ions are responsible for the observed divergent product pathways, CAPE (62) was again treated with Ag2 O, but this time in the presence of Mn(acac)2 in CH2 Cl2 at room temperature for 5 h, affording a 54% conversion of CAPE (62) into the benzo[kl]xanthene lignan 63 (51% yield), without any detectable 8–5 neolignan 71 (Scheme 18.13). These results highlight the influence of the metal in directing the formation of either lignans or neolignans; moreover, these data suggest Mn2+ ions can somehow stabilize quinone methide radicals in a manner that favors 8–8 oxidative phenolic couplings. It will be interesting to see if this ion selectivity (Ag+ versus Mn2+ ) is general for other cinnamyl substrates, providing control for the formation of various lignan and neolignan chemotypes [56].
18.2 Conclusion
While this chapter could not cover the full spectrum of lignan biomimetic syntheses reported, we focused on key syntheses and mechanistic understandings from the major classes of lignans. In the 75 years since ‘‘lignans’’ was coined by Haworth
O
H+
HO
O H
O
O
3
OH
HO
CO2R
E'
OH O H+
6'
OH
CO2R
OH
CO2R
CO2R
65
O
CO2R
8'
CO2R
O OH
HO
HO
H+
O
HO
D'
O
H
O
OH
CO2R
CO2R
OH
CO2R
CO2R
OH
CO2R
CO2R
B
7'
H
O
2
A
H
H+
Proposed mechanism for the formation of benzo[kl]xanthenes 63 and 64.
F'
OH
CO2R
64
63
CO2R
HO
OH
OH
OH
CO2R
HO
O
OH
CAPE, 62
MnO2 1-e oxidation
CO2R
OR
R = CH2CH2Ph
OH
Scheme 18.11
HO
O
8
H
O
HO
H+
O
2
O A'
OH
7'
C'
OH
CO2R
CO2R
OH
CO2R
CO2R
OH
CO2R
H
OH
H+ O B'
OH
OH
H
CO2R
18.2 Conclusion 689
690
18 Biomimetic Synthesis of Lignans OH MeO2C 5
OH MeO2C
O
O
H+
O Ag2O
OMe
67
CH2Cl2
H MeO2C
8'
RO2C
HO OH
O
HO HO 66
68
O
69
OH MeO2C OH
4
OH
OH
O OH MeO2C
MeO2C
7'
CO2Me HO 71
70
Scheme 18.12 Ag2 O-mediated oxidative phenolic coupling of methyl caffeate (66) forms racemic neolignan 71 through the proposed mechanism. CO2R HO
CO2R O
O
Ag2O Mn(acac)2
OR
CH2Cl2
HO OH
OH
OH
R = CH2CH2Ph
Ag2O
CH2Cl2
CAPE, 62
63
OH
OH O OH
RO2C CO2R 71
Scheme 18.13 Differential product formation in the presence of Mn2+ ions in the Ag2 O-mediated oxidative phenolic coupling of CAPE (62) to form either racemic neolignan 71 or benzo[kl]xanthene 63.
O
H+
References
to describe a group of phenylpropanoid dimeric natural products linked through the β –β (8–8 ) carbons of the propanyl side-chains, our understanding has grown tremendously in terms of the origin of enantioselectivity, of biomimetic approaches for β –β oxidative phenolic couplings (both metal- and enzyme-mediated), and ‘‘metal-tuning’’ to access distinct chemotypes via stabilization of certain radical pathways. Lignans truly highlight the power of biomimetic synthesis, accessing diverse stereochemically rich scaffolds from starting materials devoid of stereocenters. With a more complete arsenal of metal and enzyme catalysts available, the next decade holds even greater promise for the biomimetic synthesis of lignans.
References 1. Haworth, R.D. (1936) Heterocyclic
2.
3.
4.
5.
6.
7.
8.
compounds. Annu. Rep. Prog. Chem., 33, 310–334. Haworth, R.D. (1942) Chemistry of the lignan group of natural products. J. Chem. Soc., 448–456. Ayres, D.C. and Loike, J.D. (1990) Lignans – Chemistry, Biology and Clinical Properties, Cambridge University Press, Cambridge. Ward, R.S. (1995) Lignans, neolignans, and related compounds. Nat. Prod. Rep., 12(2), 183–205. Wallis, A.F.A. (1998) Lignin and Lignan Biosynthesis (eds N.G. Lewis and S. Sarkanen) ACS Symposium Series, vol. 697, American Chemical Society, Washington DC, pp. 323–333. Bose, J.S., Gangan, V., Prakash, R., Jain, S.K., and Manna, S.K. (2009) A dihydrobenzofuran lignin induces cell death by modulating mitochondrial pathway and G2/M cell cycle arrest. J. Med. Chem., 52(10), 3184–3190. Van Miert, S., Van Dyck, S., Schmidt, T.J., Brun, R., Vlietinck, A., Lemiere, G., and Pieters, L. (2005) Antileishmanial activity, cytotoxicity and QSAR analysis of synthetic dihydrobenzofuran lignans and related benzofurans. Bioorg. Med. Chem., 13(3), 661–669. Apers, S., Paper, D., Buergermeister, J., Baronikova, S., Van Dyck, S., Lemiere, G., Vlietinck, A., and Pieters, L. (2002) Antiangiogenic activity of synthetic dihydrobenzofuran lignans. J. Nat. Prod., 65(5), 718–720.
9. Charlton, J.L. (1998) Antiviral activ-
10.
11.
12.
13.
14.
15.
16.
17.
ity of lignans. J. Nat. Prod., 61(11), 1447–1451. Syah, Y.M. and Ghisalberti, E.L. (1996) Biologically active cyanogenetic, iridoid and lignan glycosides from Eremophila maculata. Fitoterapia, 67(5), 447–451. Ward, R.S. (2000) in Bioactive Natural Products (Part E), (ed. A.U. Raman, Studies in Natural Products Chemistry, vol. 24, Elsevier, Amsterdam, pp. 739–798. Spatafora, C. and Tringali, C. (2007) in Targets in Heterocyclic Synthesis, vol. 11 (eds O.A. Attanasi and D. Spinelli), Societa Chimica Italiana, Italy, pp. 284–312. Dewick, P.M. (2009) Medicinal Natural Products: A Biosynthetic Approach, 3rd edn, John Wiley & Sons, Ltd, Chichester. Pan, J.-P., Chen, S.-L., Yang, M.-H., Wu, J., Sinkkonen, J., and Zou, K. (2009) An update on lignans: natural products and synthesis. Nat. Prod. Rep., 26, 1251–1292. Brophy, G.C., Mohandas, J., Slaytor, M., Sternhell, M., Watson, T.R., and Wilson, L.A. (1969) Novel lignans from a Cinnamomum sp. from Bougainville. Tetrahedron Lett., 10, 5159–5162. Gottlieb, O.R. (1972) Plant chemosystematics and phylogeny. III. Chemosystematics of the Lauraceae. Phytochemistry, 11(5), 1537–1570. Moss, G.P. (2000) Nomenclature of Lignan and Neolignan. IUPAC-IUB Joint Commission on Biochemical
691
692
18 Biomimetic Synthesis of Lignans
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
Nomenclature, LG-1.2. Available online at http://www.chem.qmul.ac.uk/iupac/ lignan/LG0n1.html#p12 (accessed 22 November 2010). Ward, R.S. (1993) Lignans, neolignans, and related compounds. Nat. Prod. Rep., 10(1), 1–28. Davin, L.B., Wang, H.-B., Crowell, A.L., Debgar, D.L., Martin, D.M., Sarkanen, S., and Lewis, N.G. (1997) Stereoselective bimolecular phenoxy radical coupling by an auxiliary (dirigent) protein without an active center. Science, 275, 362–366. Davin, L.B., Jourdes, M., Pattern, A.M., Kim, K.-W., Vassao, D.G., and Lewis, N.G. (2008) Dissection of lignin macromolecular configuration and assembly: comparison to related biochemical processes in allyl/propenyl phenol and lignin biosynthesis. Nat. Prod. Rep., 25, 1015–1090. Ward, R.S. (1992) Synthesis of podophyllotoxin and related compounds. Synthesis, 8, 719–730. Ward, R.S. (2003) Different strategies for the chemical synthesis of lignans. Phytochem. Rev., 2, 391–400. Ziegler, F.E. and Schwartz, J.A. (1978) Synthetic studies on lignan lactones: aryl dithiane route to (±)-podorhizol and (±)-isopodophyllotoxin and approaches to the stagnane skeleton. J. Org. Chem., 43, 985–991. Pelter, A., Ward, R.S., Pritchard, M.C., and Kay, I.T. (1988) A short versatile synthesis of aryltetralin lignans including deoxypodophyllotoxin and epiisopodophyllotoxin. J. Chem. Soc., Perkin Trans. 1, 1603–1614. Rodrigo, R. (1980) A stereo- and regiocontrolled synthesis of Podophyllum lignans. J. Org. Chem., 45, 4538–4540. Rajapaksa, D. and Rodrigo, R. (1981) A stereocontrolled synthesis of antineoplastic Podophyllum lignans. J. Am. Chem. Soc., 103, 6208–6209. Forsey, S.P., Rajapaksa, D., Taylor, N.J., and Rodrigo, R. (1989) Comprehensive synthetic route to eight diastereomeric Podophyllum lignans. J. Org. Chem., 54, 4280–4290. Wu, Y., Zhao, J., Chen, J., Pan, C., and Zhang, H. (2009) Enantioselective
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
sequential conjugate addition-allylation reactions: a concise total synthesis of (+)-podophyllotoxin. Org. Lett., 11, 597–600. Broomhead, A.J., Rahman, M.M.A., Dewick, P.M., Jackson, D.E., and Lucas, J.A. (1991) Matairesinol as a precursor of Podophyllum lignans. Phytochemistry, 30, 1489–1492. Cambie, R.C., Craw, P.A., Rutlegde, P.S., and Woodgate, P.D. (1988) Oxidative coupling of lignans III. Non-phenolic oxidative coupling of deoxypodorhizon and related compounds. Aust. J. Chem., 41, 897–918. Ward, R.S. and Hughes, D.D. (2001) Oxidative cyclisation of cis- and trans-2,3-dibenzyl-butyrolactone using phenyl iodonium bis(trifluoroacetate) and 2,3-dichloro5,6-dicyano-1,4-benzoquinone. Tetrahedron, 57, 5633–5639. Brown, R.C.D. and Swain, N.A. (2004) Synthesis of furofuran lignans. Synthesis, 811–827. MacRae, W.D., Hudson, J.B., and Towers, G.H. (1989) The antiviral action of lignans. Planta Med., 55(6), 531–535. Anjaneyulu, A.S.R., Ramaiah, P.A., Row, L.R., Venkateswarlu, R., Pelter, A., Ward, R.S. (1981) New lignans from the heartwood of Cleistanthus collinus. Tetrahedron, 37(21), 3641–3652. Whiting, D.A. (1985) Lignans, neolignans, and related compounds. Nat. Prod. Rep., 2, 191–212. Whiting, D.A. (1987) Lignans, neolignans, and related compounds. Nat. Prod. Rep., 4, 499–525. Whiting, D.A. (1990) Lignans, neolignans, and related compounds. Nat. Prod. Rep., 7, 349–360. Whiting, D.A. (2001) Natural phenolic compounds 1900–2000: a bird’s eye view of a century’s chemistry. Nat. Prod. Rep., 18(6), 583–606. Sih, C.J., Ravikumar, P.R., Huang, F.C., Buckner, C. and Whitlock, H. Jr. (1976) Isolation and synthesis of pinoresinol diglucoside, a major antihypertensive principle of Tu-Chung
References
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
(Eucommia ulmoides Oliver). J. Am. Chem. Soc., 98(17), 5412–5413. Cartwright, N.J. and Haworth, R.D. (1944) Constituents of natural phenolic resins. XIX. Oxidation of ferulic acid. J. Chem. Soc., 535–537. Erdtman, H. (1935) Dehydrogenation of phenols. Svensk Kemisk Tidskrift, 47, 223–230. Freudenberg, K. and Schraube, H. (1955) Sinapyl alcohol and the synthesis of syringaresinol. Chem. Ber., 88, 16–23. Vermes, B., Seligmann, O., and Wagner, H. (1991) Synthesis of biologically active tetrahydrofurofuranlignan (syringin, pinoresinol) mono- and bisglucosides. Phytochemistry, 30(9), 3087–3089. Chapman, O.L., Engel, M.R., Springer, J.P., and Clardy, J.C. (1971) The total synthesis of carpanone. J. Am. Chem. Soc., 93, 6696–6698. Matsumoto, M. and Kuroda, K. (1981) Transition metal (II) Schiff’s base complexes catalyzed oxidation of trans-2(propenyl)-4,5-methylenedioxyphenol to carpanone by molecular oxygen. Tetrahedron Lett., 22, 4437–4440. Iyer, M.R. and Trivedi, G.K. (1992) Silver(I) oxide catalyzed oxidation of o-allyl- and o-(1-propenyl)phenols. Bull. Chem. Soc. Jpn., 65, 1662–1664. Baxendale, I.R., Lee, A.I., and Ley, S.V. (2001) A concise synthesis of the natural product carpanone using solid-supported reagents and scavengers. Synlett, 1482–1484. Lindsley, C.W., Chan, L.K., Goess, B.C., Joseph, R., and Shair, M.D. (2001) Solid-phase biomimetic synthesis of carpanone-like molecules. J. Am. Chem. Soc., 122, 422–423. Sung, S.H. and Kim, Y.C. (2000) Hepatoprotective diastereomeric lignans from Saururus chinensis herbs. J. Nat. Prod., 63, 1019–1021.
50. Jakupovic, J. and Eid, F. (1987)
51.
52.
53.
54.
55.
56.
57.
Benzoxanthenone derivatives from Polemannia montana. Phytochemistry, 26, 2427–2429. Daniels, N.R., Fadeyi, O., and Lindsley, C.W. (2008) A new catalytic Cu(II)/sparteine oxidant system for β,β-phenolic couplings of styrenyl phenols: total synthesis of carpanone and unnatural analogues. Org. Lett., 10, 4097–4100. Fadeyi, O., Daniels, R.N., DeGuire, S., and Lindsley, C.W. (2009) Total synthesis of polemannone B and C. Tetrahedron Lett., 50, 3084–3087. Da Silva, A.A.S., Souto, A.L., Agra, M.F., da-Cunha, E.V.L., Barbosa-Filho, J.M., da Silva, M.S., and Braz-Filho, R. (2004) A new arylnaphthalene type lignan from Cordia rufescens A. DC. (Boraginaceae). Arkivoc, 6, 54–58. Shi, S., Zhang, Y., Huang, K., Liu, S., and Zhao, Y. (2008) Application of preparative high-speed counter-current chromatography for separation and purification of lignans from Taraxacum mongolicum. Food Chem., 108, 402–406. Tanaka, T., Nishimura, A., Kouno, Y., Nonaka, G., and Yang, C.-R. (1997) Four new caffeic acid metabolites, yunnaneic acids A-H, from Salvia yunnanensis. Chem. Pharm. Bull., 45, 1596–1600. Daquino, C., Rescifina, A., Spatafora, C., and Tringali, C. (2009) Biomimetic synthesis of natural and ‘unnatural’ lignans by oxidative coupling of caffeic esters. Eur. J. Org. Chem., 6289–6300. Pieters, L., Van Dyck, S., Gao, M., Bai, R., Hamel, E., Vlietinck, A., and Lemiere, G. (1999) Synthesis and biological evaluation of dihydrobenzofuran lignans and related compounds as potential antitumor agents that inhibit tubulin polymerization. J. Med. Chem., 42(26), 5475–5481.
693
695
19 Synthetic Approaches to the Resveratrol-Based Family of Oligomeric Natural Products Scott A. Snyder
Look deep into nature, and then you will understand everything better. Albert Einstein
19.1 Introduction
Although the natural product resveratrol (1, Figure 19.1) is quite simple from the standpoint of molecular architecture, its two aromatic rings, double bond, and three phenol residues confer an ability to accomplish a profound array of biochemistry [1]. Indeed, recent screens in various in vivo, mouse-based assays have shown that it can serve as a potent anti-oxidant and antitumor agent as well as improve neuronal functioning, slow the aging process, and even effect weight loss [2]. In addition, although unconfirmed, many have extended these preliminary findings into a hypothesis that resveratrol impacts human health in similar ways, with its main source in our diets being the consumption of wine, particularly red variants [3]. The goal of this chapter, however, is not to evaluate these claims nor focus on the biological properties of resveratrol, profound as they may be. Rather, it is to discuss the main reason that many plants create this intriguing natural product and explore what can then be done chemically with its atoms in terms of bond constructions. Like most natural products, resveratrol (1) is produced largely to enable organismal survival. In this case, it is a phytoalexin whose protective role is potent antifungal activity that enables the over 70 species of plants throughout the world that synthesize it to combat (and hopefully survive) a number of exogenous attacks [3]. Grapevines, for instance, utilize resveratrol to manage the effects of the Botrytis fungus, a disease that many vintners welcome since it leads to grapes with intensified flavor but which comes with the risk of total harvest loss if not properly managed [4]. Resveratrol, however, is just the tip of the iceberg in terms of any plant’s total antifungal arsenal. Indeed, once infected, plants will also use resveratrol as a powerful synthetic building block, one that can be joined in myriad ways to create new and architecturally distinct polyphenolic natural products such Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
HO
O
O
OH
OH
H O
OH HO
OH
O
OH
OH
13: vaticanol C
HO
H
H HO
H
9: α-viniferin
O
H H H
OH
5: ε-viniferin
OH
O
1: resveratrol
OH
OH
H
OH
HO OH
HO
OH HO
H
H
H
OH
OH
OH
OH
HO
O
OH
HO OH
H
HO
H
OH
OH
OH
HO OH
OH
OH
OH
OH
15: vaticanol A
H
O
11: pallidol
HO
H
OH
7: quadrangularin A
HO
OH
3: paucifloral F
HO
OH
HO
HO
HO
HO
OH
HO
OH
OH
OH
14: vaticanol G OH
HO
H
OH
10: ampelopsin F
H
6: ampelopsin D
HO OH
O
OH
2: caraphenol B
HO
HO
HO
HO
HO
HO
HO
HO
OH
OH
HO
OH
O
H
O
H
OH
H
O
OH
HO
H
OH
OMe
OH
16: hopeanol
O HO
O
HO
H
O
OH
O
OH
OH
OH
OH
8: paucifloral E
HO
HO
12: hopeaphenol
HO
HO
O
OH
4: diptoindonesin D
HO
HO
HO
HO
O
HO
OH
Figure 19.1 Structures of resveratrol (1) and selected oligomers (2–16) representative of the architectural complexity of the family.
HO
HO
HO
HO
HO
HO
HO
696
19 Synthetic Approaches to the Resveratrol-Based Family of Oligomeric Natural Products
19.2 Biosynthetic Approaches
as 2–16 [5]. These structures, along with the approximately 500 others [6] that have been isolated to date, encompass diverse carbocyclic and heterocyclic systems that display fascinating stereochemical complexity; they also possess broad spectrum biological activity that includes antifungal properties, of course, but also leads to materials that could, in some cases, fight cancer, HIV, and bacteria as well [5]. In this way, resveratrol is the lynchpin for a cohesive, immune-like, chemical response wherein an array of structurally unique natural products (anywhere from ten or more from one plant) are marshaled concurrently to thwart an invading pathogen in the hope that at least one of the molecules has enough power to enable organismal survival [7]. In the ensuing sections, we present the current level of understanding regarding how these oligomers are formed in Nature as well as the synthetic efforts that have been directed toward fashioning their fascinating chemical diversity. As will be seen, much work remains to be done, but, hopefully, as several studies over the past few years have demonstrated, chemists are finally beginning to gain a handle on how to tackle the chemical complexity presented by the family, complexity that the parent molecule’s structure belies. In fact, in our opinion, that degree of diversity is without parallel in terms of any other oligomeric natural product family in existence.
19.2 Biosynthetic Approaches
Resveratrol is the product of an enzyme-based synthesis, one that sequentially adds three molecules of malonyl-CoA to 4-coumaroyl-CoA with a terminating enzymatic cyclization using ‘‘stilbene synthase’’ to fashion its 3,5-dihydroxybenzene ring [8]. Beyond the building block itself, however, far less is known with regard to how the dimeric and higher-order structures are fashioned; enzymatic participation is often invoked because the oligomers are obtained as single enantiomers, though at what point that involvement occurs, and how it occurs (i.e., whether to create proper reactive intermediates, ensure correct presentation of monomers to create a specific dimer, etc), is unknown. Heightening that mystery is the fact that certain plants will produce the antipode of a given natural product forged enantioselectively by another species [9]. Nevertheless, despite such uncertainty, the structures within Figure 19.1 hint at several possibilities for their general mode of construction based on certain homologies. For instance, if resveratrol (1) was dimerized in an uncontrolled way into a range of structures such as 5–8, these molecules could then serve as launching points for the generation of many of the higher-order and/or more unique structures due to their possession of reactive functionality. For instance, oxidative cleavage of the double bond within ampelopsin D (6) could lead to the triaryl-containing natural product paucifloral F (3), while regioselective oxidation of that function followed by enol equilibration could afford caraphenol B (2). Alternatively, electrophilic activation of the double bond within 6, followed by a Friedel–Crafts reaction with
697
698
19 Synthetic Approaches to the Resveratrol-Based Family of Oligomeric Natural Products
the neighboring 3,5-dihydroxybenzene system, could lead to a structure such as ampelopsin F (10). These cores, along with all of the others, could then be further elaborated with additional resveratrol monomers; for instance, vaticanol C (13) formally has two resveratrol units attached stereospecifically onto ampelopsin F (10). Thus, if we were to encapsulate these ideas within the context of the opening discussion of this section, the higher-order structures could potentially result as single enantiomers through relative stereocontrol if Nature had prepared its given dimeric precursor with optical purity. The pathway by which most of the bond-forming chemistry occurs in Nature appears to be through radical bond constructions. More specifically, only partially controlled, or even non-controlled, radical-based constructions [3]. Scheme 19.1 shows a typical biosynthetic proposal for these higher-order structures, using ε-viniferin (5) for purposes of illustration. As indicated, initial oxidation of resveratrol (1) gives rise to two different radicals, one resonance stabilized (17) and one on the 3,5-dihydroxybenzene system (18). These species then unite to form 19, which, following rearomatization and phenol attack onto the remaining quinone methide, leads to the target structure (5). The trans-orientation of the two rings, a hallmark of nearly all the dihydrofuran rings within the resveratrol family, is likely the result of equilibration through ring-opening to produce the most thermodynamically stable isomer [10]. Of course, though the mechanism as shown suggests high control in radical positioning (as many biosynthetic hypotheses for this molecule class similarly implicate), such control is quite difficult to achieve in practice. Indeed, 18 is more difficult to form than 17 and the mechanistic pathway assumes that only the specific radicals drawn, and not their varied resonance alternatives, react. Thus, many pathways in addition to that shown for 5 are likely to occur concomitantly, thereby leading to many products. This statement is reflected further in several interesting biosynthetic explorations undertaken during the past three decades, studies that have revealed that (i) synthesizing resveratrol-based natural products in a controlled way is nearly impossible to achieve through biosynthetic approaches (in line with Nature’s outcome) and (ii) when a single structure is generated in the laboratory with some measure of control (i.e., in yields greater than 50%) it is often an analog/structural isomer of a natural product rather than a natural product. O O
HO
OH
O
OH
O [O]
1: resveratrol
OH
H
OH
OH
O HO
HO HO
OH
O
17
HO
OH 18
OH OH
19
Scheme 19.1 Representative biosynthetic scheme showing the putative generation of ε-viniferin (5) from resveratrol (1).
OH 5: ε-viniferin
19.2 Biosynthetic Approaches
HO
Horse radish peroxidase, H2O2 or
OH
AgOAc or Mn(OAc)3 or K3Fe(CN)6
O
O
OH
OH
OH
O OH
HO OH 17a
HO
O
O
HO
OH 1: resveratrol
699
OH
OH 17b
20a
HO
O HO
HO
OH
HO
OH
O
(22-97%)
OH
HO
OH
OH
OH 5: ε-viniferin
OH
21: non-natural analog
OH 20b
Scheme 19.2 Efforts to effect direct resveratrol (1) dimerization typically result in non-natural analogs of resveratrol dimers (such as 21).
The first reported exploration into the reactivity of resveratrol was disclosed by Langcake and Pryce in 1977 [11], investigations that evaluated the products generated from exposure of resveratrol (1, Scheme 19.2) to horseradish peroxidase in the presence of H2 O2 . Of the materials formed, only compound 21, the likely product of O- and C-centered radicals uniting through the most readily oxidized phenol within resveratrol, was characterized. As can be seen with a cursory inspection, though its dihydrofuran ring is reminiscent of the family, the architectural connectivity does not match that of ε-viniferin (5), the form most commonly found in Nature. A more recent study of the same process by Sako and coworkers using single-electron transfer reagents as initiators instead of an enzymatic system obtained similar results [12]. In fact, certain reagents led to 21 quantitatively, but no ε-viniferin (5) was ever characterized from these experiments. Of course, in most investigations of this type, the yield is not quantitative, and as such it is certainly possible that within the collection of additional materials produced several natural products can be found. For instance, in a study of resveratrol oligomerization using Anthromyces ramosus in the presence of H2 O2 , the investigators were able to obtain pure 22 (Scheme 19.3) as well as the natural product pallidol (11), albeit in low yield relative to the common, major product (i.e., 21) presented above [13]. To date, only one report describes a chemical synthesis of ε-viniferin (5) with any measure of efficacy; that work, shown in Scheme 19.4, found that if resveratrol was exposed to FeCl3 ·6H2 O in methanol as solvent, 5 could be obtained in a 30% yield [14]. Thus, these examples (Schemes 19.2–19.4) collectively reveal that with enough trial and error it is possible to forge a given scaffold with some preference over others through radical-based chemistry starting with resveratrol, but, nevertheless, not with any real degree of control or de novo predictability based on first principles of chemical reactivity.
700
19 Synthetic Approaches to the Resveratrol-Based Family of Oligomeric Natural Products OH HO Anthromyces ramosus, H2O2
HO
OH
OH acetone (4.6%)
HO OH
OH
HO
1: resveratrol (7.4%)
(18.4%)
OH
HO
OH
22
OH
O HO
H
H
OH
HO HO
HO 11: pallidol
OH
OH OH
21
Scheme 19.3 A second example of enzyme-induced oxidative dimerizations of resveratrol (1). O
HO
OH
HO FeCl3•6H2O
OH
MeOH (30%)
HO OH
OH 1: resveratrol
Scheme 19.4
5: ε-viniferin
OH
Conditions that are reported to provide ε-viniferin (5) from resveratrol (1).
More recent efforts, however, have taken a different approach to addressing the challenge of controlled radical generation on the resveratrol framework. Rather than utilizing resveratrol itself as the starting material, these investigations have asked what products would arise from derivatives of resveratrol and/or selectively protected forms through similar oxidation chemistry. From the standpoint of biomimetic chemistry, these investigations might make more inherent sense in that Nature could certainly temporarily engage a given phenol in resveratrol with an acetate, phosphate, sulfonate, or carbohydrate group, thereby altering the radicals that could be generated and thus leading to different product distributions [15]. Indeed, since each plant seems to synthesize different arrays of resveratrol-based oligomers (in terms of specific structures and their respective amounts), such tuning might reflect the mechanism whereby plants have evolved their compound collections in certain directions based on selective evolutionary pressure. Natural product isolation procedures may be overly masking the actual presence of these groups by effecting their excision from the phenol subunits prior to characterization.
19.2 Biosynthetic Approaches HO
OH
t-Bu OH Horse radish peroxidase, H2O2
O
t-Bu HO OH t-Bu
acetone/H2O, 25 °C
t-Bu
O
t-Bu
HO
OH 23
t-Bu
24 (35%)
OH
t-Bu
OH
t-Bu HO
HO
OH
HO OH OH
HO OH
t-Bu
+ other products O
7: quadrangularin A
OH
t-Bu
25
Scheme 19.5 Use of resveratrol derivative 23 to achieve a total synthesis of quadrangularin A (7).
The first landmark study along these lines was reported by the Hou group in 2006, in which the t-butylated resveratrol analog 23 (Scheme 19.5) was exposed to horseradish peroxidase and H2 O2 in aqueous acetone [16]. Such a design took advantage of the more facile generation of radicals at the para-disposed phenol that the earlier reports had consistently mentioned, but used the bulky t-butyl groups to ensure that it would be the carbon-centered resonance-forms of the original radical that would be the primary reactive species. As a result, they were able to obtain indane 25 in 35% yield. Subsequently, these investigators were able to remove the t-butyl ‘‘protective’’ groups through exposure to strong acid and ultimately access the natural product quadrangularin A (7). Whether this design will enable access to other resveratrol-based cores remains the subject of additional investigations, but it certainly illustrates the power of a different approach as relates to the controlled resveratrol oligomerization problem, which did not appear possible starting from resveratrol itself. Similarly, work by Velu and coworkers sought to examine the reactivity of dimethylated resveratrol analog 26 (Scheme 19.6), a material that can only form radicals from one phenol [17]. Their study showed a remarkable ability to fashion unique chemical diversity as a result of this structural alteration, principally just by changing the solvent in which they used a given oxidant. For instance, if 26 was exposed to FeCl3 ·6H2 O in a mixture of CH2 Cl2 –MeOH (7 : 3), the unique tetrahydrofuran 28 was produced as the predominant product, while exposure of the same material to the same oxidant in CH2 Cl2 alone gave rise to a mixture that contained both 29 and 30, protected forms of the natural products ampelopsin
701
19 Synthetic Approaches to the Resveratrol-Based Family of Oligomeric Natural Products
702
AgOAc
OMe
MeO
CH2Cl2 or
OH
CH2Cl2/MeOH (7:3)
H
H
O HO
(18-36%)
OMe
MeO HO
MeO
HO
OMe
29: protected pallidol
OMe
FeCl3•6H2O, CH2Cl2
OMe 27 OMe
(10% 29) (7% 30)
HO OH
HO
26
OH O
OMe
R OMe MeO OMe
(29%) FeCl3•6H2O, CH2Cl2/MeOH (7:3)
MeO
30: R = OMe, protected ampelopsin F
OMe MeO OMe 28
Scheme 19.6 Explorations into the effect of partial phenol protection of resveratrol (i.e., 26) on dimeric product formation.
F (10) and pallidol (11, cf. Figure 19.1). In general, however, most compounds formed are not natural products, though perhaps the most important conclusion from such an outcome to date is that there are many frameworks in addition to those isolated from Nature that can be formed from resveratrol’s atoms and functional groups. In addition to altered resveratrol fragments, an intriguing biomimetic approach has been to examine what can be produced from a given higher-order structure. In other words, the question is not what materials can be made from resveratrol itself but rather what frameworks could result from a given dimer through exposure to different conditions. The Niwa group has been the main contributor to such endeavors, and Scheme 19.7 illustrates some of their seminal experiments. Here, ε-viniferin (5) was viewed as a critical starting point, and, in fact, upon exposure to various acid sources, these investigators obtained isoampelopsin D (33), ampelopsin D (6), and ampelopsin F (10) along with several other natural products in differing amounts [18]. Mixtures were always obtained, largely as a result of the different stereochemistry that can result from the ring-opening and reclosing process proposed (i.e., access to either 32a or 32b), but their study highlights the provocative notion that ε-viniferin (5), not resveratrol itself, could be the real lynchpin to many of the higher-order structures of the family since much of the rest of the family, as noted above, could derive from this array of new dimeric structures. In addition to exploring the reactions of ε-viniferin (5) under rearrangement conditions, the Niwa group has also evaluated what it can do when oxidized in the presence of resveratrol [9b]. As one might expect, complicated mixtures are
OH
HO
O
31
OH
OH
6: ampelopsin D
HO
HO
HO
HO
OH
OH
OH
OH
HO
HO
Base
A
B
H
HO HO
HO
C
32a
B
32
OH
HO
C
OH
A
OH
OH
OH
OH
H
HO
32b
B
A
C
C
OH
10: ampelopsin F
B
A
H
HO HO
HO
HO
HO
HO
Biogenetic explorations using ε-viniferin (5) instead of resveratrol as the starting material.
OH
OH
acid source
OH
OH
33: isoampelopsin D
HO
OH
O
5: ε-viniferin
HO
HO
Scheme 19.7
HO
HO
H
OH
OH
OH
OH
19.2 Biosynthetic Approaches 703
19 Synthetic Approaches to the Resveratrol-Based Family of Oligomeric Natural Products
704
OH
OH
O
OH
O
HO
O
HO Horse radish peroxidase, H 2O 2
+
HO
HO
aq. acetone
OH
OH OH RO
5: ε-viniferin
OR
OH
O
O RO
34
1 or 26
OH
OH
O
HO
OR 35
O
HO H
H H
H
HO OR OH
(R = H, 1.1%) (R = Me, 2.7%)
HO OR OH
OR OH
37: R = Me 38: R = H, davidiolA
OR OH 36
Scheme 19.8 Biomimetic total synthesis of davidiol A (38) starting from ε-viniferin (5) and resveratrol (1).
formed. Nevertheless, out of this milieu they were recently able to isolate the natural product davidiol A (38, Scheme 19.8), albeit in only 2.7% overall yield if a protected form of resveratrol (26) was used in the process and in 1.1% yield if it was resveratrol itself. This natural product is one of many diastereomers possessing a fused 7,5-bicyclic system, of which vaticanol A (15, cf. Figure 19.1) is another example. Finally, outside of radical chemistry, there are other ways to envision the biogenetic synthesis of resveratrol-like frameworks. Carbocation-induced reactions, not so unlike the work discussed above in the context of Scheme 19.7, are one such option. The key results of this approach are shown in Scheme 19.9, wherein select examples of stilbene derivatives of general structure 39 have been shown to lead to various indane (43) and tetralin (42) ring systems reminiscent of the resveratrol family following regioselective cation generation formed through a retro-Ritter reaction [19]. To date, however, the extension of such studies to the exact resveratrol phenol patterning has yet to afford the same carbon skeletons. Thus, it remains to be seen whether such acid-induced rearrangements can afford pertinent resveratrol-based oligomers [20]. There is at least, in our opinion, one other mechanistic possibility for resveratrol oligomer formation: photochemical resveratrol dimerization followed by subsequent rearrangements. To our knowledge this particular approach has never been pursued, though it is one additional alternative that, at least based on simple mechanistic pictures, could afford many of the desired frameworks as well. Recent synthesis work targeting a marine-based family of varied dimeric structures provides the impetus behind this idea [21].
19.3 Stepwise Synthetic Approaches R1
R1
R1
R2
R2
R Retro-
Ethyl poly-
NHCOR phosphate
R2
Ritter reaction
N
80 °C, 8 h
R4 39
R3
40
R3 R1
R4
R4
R3
41
R4
R3
R1 R2
+ other products
R2
R3
R1
+
R2 +
R2
R4 R2
R1 R4
R4
R3 44
R3 43
R1
R3 R4 42
Scheme 19.9 Example of non-radical-based dimerizations; the use of this chemistry with appropriately functionalized resveratrol derivatives has not yet proven effective.
19.3 Stepwise Synthetic Approaches 19.3.1 Work toward Single Targets within the Resveratrol Family
While much has clearly been learned from the biogenetic approaches above, the resveratrol family has also served as a testing ground for the power of various synthetic methods, as well as more stepwise, non-biomimetic approaches. In this section, we will present three recent studies along such lines [22]. The first was performed by Jeffrey and Sarpong, wherein various palladium-catalyzed processes were evaluated for their ability to produce resveratrol-like dimeric structures [23]. As indicated in Scheme 19.10, materials possessing a brominated 3,5-dimethoxybenzene ring system provided the critical handle by which to add additional carbogenic complexity, with a Heck-type reaction cascade between 45 and 46 affording indene 47 in 53% yield and a Larock annulation between 49 and the same alkyne (46) leading to a 1 : 1 mixture of 50 and 51. Although none of these frameworks themselves are natural products, 47 is an oxidized form of quadrangularin A (7, cf. Figure 19.1) and could be converted into a similarly overoxidized form of pallidol (11, cf. Figure 19.1), while 51 could be transformed into paucifloral F (3) in two additional steps. The second approach was executed by Kim and Choi in an effort to prepare the seven-membered carbocyclic ring system possessed by various members of the resveratrol family [24]. These investigators targeted the natural product
705
O
49
Br
OMe
+
Br
MeO
46
OMe
OMe
50
O
OMe
DMF, 135 °C,15 h (53%)
Pd(OAc)2, Ph3P, i -Pr2NEt
OMe
OMe
MeO
+
MeO
MeO
47 MeO MeO
MeO
51
O
OMe
OMe
OMe
(38%)
CH2Cl2,
FeCl3•6H2O
MeO
OMe 25 °C, 18 h
OMe
HO
MeO
O
OH
OMe
3: paucifloral F
HO
HO
48
MeO
OMe
Use of palladium-based reactions to forge resveratrol-based cores as well as to synthesize paucifloral F (3).
MeO
MeO
DMF, 100°C, 17 h (81%) (1:1)
46, Pd(OAc)2, OMe NaOAc, n-Bu4NCl
OMe 45
Scheme 19.10
MeO
MeO
OH
OMe
706
19 Synthetic Approaches to the Resveratrol-Based Family of Oligomeric Natural Products
19.3 Stepwise Synthetic Approaches MeO
MeO
MeO O
Br
O Bi(OTf)3
O
MeO2C
OMe
OMe Pd(OAc)2, KOAc
MeO2C
CH2Cl2, ∆ (88%)
MeO 52
DMA, 80 °C (65%)
OMe
MeO
O
54
Dess-Martin oxidation
(100%)
MeO O
O Bi(OTf)3
HO
(98%) OMe
OMe MeO
OMe OMe
MeO
MeO
O
MeO
O MeO2C
53
MeO
OMe
57: permethylated shoreaphenol
MeO
707
OMe
MeO 56
MeO
O OMe OMe
MeO 55
Scheme 19.11 Synthesis of a permethylated form of shoreaphenol (57) based on two key Bi(OTf )3 -catalyzed reactions.
shoreaphenol specifically [25], which they were able to access in protected form (57, Scheme 19.11) through a series of Bi(OTf )3 -catalyzed reactions [26]. The first of these Lewis acid-induced operations proved critical in forging the furan ring of the target through a Friedel–Crafts cyclization onto the ketone group within 52 followed by dehydration. Then, following the attachment of an aryl ring onto the open C2 position of the resultant benzofuran (i.e., 53) and the execution of several additional steps to generate 55, Bi(OTf )3 was utilized again to promote ring-opening of the lone epoxide and final C–C bond construction. This final step is certainly biomimetically inspired as it reflects a possible mode for Nature’s formation of seven-membered rings within the family. Our final example targeted a very unique set of natural products within the resveratrol family, molecules whose biogenetic origin is unclear as their architectural core is unlike any other structures in the family. These compounds are hopeanol (16) and hopeahainol A (64, Scheme 19.12), first isolated by Tan and coworkers in 2008 and shown to possess modest antitumor activity profiles [27]. Structurally, these polycycles differ primarily in terms of their aryl oxidation patterning, a difference that then determines either the presence or absence of a lactone function as well as an additional C–C bond. In early 2009, Nicolaou, Chen, and coworkers accomplished the total synthesis of these two targets, illustrating in the process that hopeahainol A (64) is likely Nature’s precursor to hopeanol (16), as a final exposure of the first natural product to NaOMe in MeOH afforded the other smoothly in 80% yield [28]. Critical to the sequence were two sequential cascade-based reactions, the first starting from 59, which used p-TsOH to effect a multistep conversion affording the rearranged polycycle 61, and the second being a ring-opening/lactonization sequence which then converted that new material into 63. Epoxidation of the lone double bond within 63, followed by Friedel–Crafts cyclization on the resultant
HO
B
R
D O
Scheme 19.12
(80%)
NaOMe
HO O O
D
A
64: hopeahainol A
B
O
OH
OMe
OH
O
OMe p -TsOH
OMe
HO
HO
O
B O
MeO
OMe
63
O
60
OH
O
D
A
OMe
OMe
OMe
OMe
OMe
OMe
(76%)
aq. NH4Cl
(65%)
HO
A
O
B
O
O
OH 62 OMe
O D
OMe
OMe
OMe
OMe
61 OMe
OH
KOt-Bu OMe
HO
OMe
Nicolaou and Chen’s synthesis of hopeahainol A (64) and hopeanol (16) via iterative cascade-based rearrangements.
16: hopeanol, R = CO2Me
HO
O
A
OH
HO
OH
OH
O
HO
HO
HO
O
59
OTBS
OH
OMe
58
TBSO
OMe
708
19 Synthetic Approaches to the Resveratrol-Based Family of Oligomeric Natural Products
19.3 Stepwise Synthetic Approaches
product, forged the final bonds needed to reach hopeahainol A (64), much in the manner described above for the Kim and Choi synthesis of protected shoreaphenol. 19.3.2 Towards a Universal, Controlled Synthesis Approach
Despite the elegance of the biosynthetic studies described earlier and the utility of the stepwise approaches just presented in the preceding section, neither of these avenues proved capable of selectively delivering all of the main resveratrol-derived frameworks in a controlled manner. Over the course of the past three years, our research group has sought to develop a solution to this challenge. Specifically, we wondered if an additional synthetic approach might exist that, even if stepwise in design, could enable access to every carbocyclic framework within the family. Given the failure of biosynthetic studies to take resveratrol itself, or variously protected forms thereof, into single natural product structures in high yields, we felt that a different starting material was required, one which would possess some of the reactivity of the parent molecule for the family, but which could be tempered appropriately to allow for controlled reactions upon exposure to different reagents in ways that the other approaches could not achieve. Our search for that alternate starting material began by examining all the known structures of the family, looking for any anomalies that might serve as a potential clue for its design. Those anomalies were found in the form of molecules such as paucifloral F (3, cf. Figure 19.1) in that it, like several others in the family, has an odd number of aromatic rings. Although these structures likely result from normal resveratrol dimers, as mentioned earlier, it was these structures that prompted the idea that perhaps the key building block needed to be one that possessed three aromatic rings; specifically, something like compound 65 (Scheme 19.13), OH
HO
R Br
HO H
Br
OH H
HO
HO R 65
1,3-diol on any ring
HO Br
O HO
Br
HO Dimerization
OH
HO OH
O
2
Scheme 19.13 General approach to target the entire carbocyclic diversity of the resveratrol class based on the use of key intermediate 65.
OH
H
HO
OH
709
710
19 Synthetic Approaches to the Resveratrol-Based Family of Oligomeric Natural Products
with alteration of the bottom, resveratrol-like, ring coupled with an additional functional group and modifiable third aryl system. Our hope was that this piece and its derivatives, following exposure to various different electrophiles (bromine, oxygen, and proton), could enable the controlled generation of all of the frameworks indicated, carbon cores that reflect much of the diversity of the entire family. As the remainder of this section will reveal, those efforts have borne fruit as nearly all of the pathways defined within Scheme 19.13 have been reduced to practice. In fact, over 15 natural products and dozens of analogs encompassing eight distinct architectures have been prepared from this critical building block to date [29]. We begin our discussion of these efforts by describing our approach to indane-containing resveratrol oligomers. As shown in Scheme 19.14, compound 66 was exposed to an acid source under controlled temperature conditions, which led to the cyclized carbocation 68 via initial ionization followed by nucleophilic attack of the proximal olefin. What happened next, however, depended on the nature of the acid counterion. When that ion was trifluoroacetate, that species itself served as a nucleophile, attacking 68 to forge, following basic work-up, the secondary alcohol 69 in 75% yield. This material could then be converted into paucifloral F (3) in just two additional steps. However, if an acid with a less nucleophilic counterion, such as p-TsOH, was used instead in the cyclization leading to 68, then it proved possible to add other nucleophiles and access different complexity. For instance, with chloride present, molecules such as 73 were accessed smoothly while the addition of excess p-methoxybenzyl sulfide afforded compound 70. Use of a Ramberg–B¨acklund olefination [30] with the latter of these materials (i.e., 70), followed by global methyl ether cleavage, completed a synthesis of ampelopsin D (6) in 18% overall yield for these final steps. From here, a final treatment with HCl in MeOH at elevated temperature (80 ◦ C) provided the means to access isoampelopsin D (39) as well; intriguingly, only the specific tetra-substituted olefin was formed through this isomerization process. Thus, simple alteration of reaction conditions allowed for the controlled synthesis of three different indane-containing resveratrol oligomers. Pleasingly, application of the same chemistry to alternative forms of the key building block such as 71 led to the smooth formation of related natural products with their pendant aryl rings switched (i.e., 72 and 7) as well despite the fact that the electron density of the phenols is directing differently.1) With these syntheses in hand, we next turned our attention to accessing the complexity posed by molecules such as ampelopsin F (10) and pallidol (11), compounds that are, in principle, one intramolecular Friedel–Crafts cyclization away from ampelopsin D (6) and quadrangularin A (7), respectively. Indeed, the idea for executing such a transformation was revealed by the earlier work from the Niwa group in the context of Scheme 19.7 [18]. As noted, however, the reaction 1) Notably, use of a protecting group other
than methyl for the phenols prevented the initial addition of the third aryl ring, likely due to the added steric bulk. Also, the ketone within molecules such as paucifloral F (3) could not be olefinated under any
set of conditions attempted; isomerization of the adjacent stereocenter was observed instead. Thus, only the sulfide nucleophile incorporated through the cascade provided a means to incorporate a fourth aryl ring to access the desired targets.
OH
71
OMe
OH
OMe
HCl, MeOH, 80°C
OMe
HO
67
OH
OMe
HO
O
OH
OH
OH
OH
MeO
MeO
1. mCPBA 2. t-BuOH, aq. KOH, CCl4, 80°C
and HO
HO
MeO
MeO
MeO
MeO
OH
70
S
OH
OMe
p -TsOH
68 SH acid =
OMe
K2CO3,
acid = TFA;
OMe
OMe
MeO
MeO
O
OH
73
Cl
OMe
OMe
OH
OMe
3: paucifloral F
HO
HO
OMe
(63% overall)
69
OH
1. [O] 2. BBr3
MeO
MeO
MeO
HO
(75%)
OMe MeOH MeO
OH OH 7: quadrangularin A (10% overall)
(18% MeO overall)
OH 3. BBr 3
OMe
6: ampelopsin D
HO
HO
OMe
OMe
72: “isopaucifloral F” (35% overall)
MeO
OH (95%) HO
OH 39: isoampelopsin D
HO
HO
OMe
acid source, −30→ −20°C
OMe
Scheme 19.14 Cascade-based approach to fashion indane-containing members of the resveratrol family of natural products from key triaryl intermediates.
MeO
OH
OMe
OMe 66
MeO
HO
MeO
19.3 Stepwise Synthetic Approaches 711
712
19 Synthetic Approaches to the Resveratrol-Based Family of Oligomeric Natural Products
was not selective in their case. Their electrophile added to both faces of the alkene and afforded access to many products. Our hope was that if we used an alternative electrophile that could add reversibly to the double bond (unlike simple proton), then perhaps higher control could be achieved if the desired C–C bond construction terminated the sequence. Our choice for this purpose was bromine, and, as shown in Scheme 19.15, that approach proved successful. In the event, we separately treated both 74 and 77 with an electrophilic bromine source [either Br2 or N-bromosuccinimide (NBS)] and found that the reaction commenced in both cases with smooth and regioselective bromination of the A-ring, followed by a second halogenation of the pendant 3,5-dimethoxybenzene ring system.2) Both of the respective mono-brominated and di-brominated species could be isolated in near quantitative yield. With the addition of a little more halogen, however, the double bond was finally engaged by the electrophile, and cyclization to either 76 or 79 proceeded smoothly to deliver the desired carbocyclic cores with three halogens attached. Importantly, the reversibility of bromonium formation is critical to the end of these sequences, as in the lower reaction (i.e., 78 → 79) product generation requires the addition of bromine onto the more hindered olefin face based on the stereochemistry of its direct aryl C-ring neighbor; thus, though this mode of addition is less favored on steric grounds, the reversibility of bromonium formation [31] in advance of the terminating cyclization was likely the reason why this process proceeded as smoothly as was observed. Pleasingly, we were able to perform similar chemistry with more highly oxidized forms of the same starting materials, accessing two members of the cararosinol family of natural products (i.e., 80 and 81) [32] as well. In addition, although one can argue from the standpoint of atom economy [33] that adding extra bromine atoms into a structure only to remove them later is not ideal [34], it is worth noting that the aryl halides with both 76 and 79 are positioned perfectly to incorporate the extra carbon frameworks needed to access natural products such as nepalensinol B (82) [35] or vaticanol C (13, Figure 19.1). Work to address these more complex tetramers through such strategies is ongoing, and hopefully will be reported soon. The final major element of carbogenic complexity within the resveratrol family is seven-membered rings as typified by such structures as diptoindonesin D (4), paucifloral E (8), hopeaphenol (12), and both vaticanols G (14) and A (15). We have recently found that we can access such systems as well from our common precursors; the key, however, was to adjust that intermediate slightly through an oxidation; subsequent exposure to electrophilic oxygen in the form of 1,1,1-trifluorodimethyldioxirane [36] afforded the means to access the seven-membered carbocycle of fully protected hemsleyanol E (83, Scheme 19.16). 2) As a general trend based on a broad reading
of the literature, electrophilic bromine will add para to protected phenols; the observed
chemoselectivity here follows that trend exclusively, with the order of halogenation likely reflecting steric biases.
HO
HO
A
MeO
C
H
OH
OMe
OMe
Br2 or NBS
OMe OMe
OMe
C
Br2 OMe or NBS
OH
HO
HO
MeO
HO
Br
A
78
B
75
H
OH
C
OH
(5389%)
MeO
MeO
Br
C
H
OMe
HO
79
HO HO
A Br OMe
H
MeO 76
Br
HO
B
Br
MeO
MeO
MeO
(80%)
FriedelCrafts alkylation
OH
OMe
OMe
OMe OMe
OMe
OMe Br
Br
OH HO 81: cararosinol D (19% overall)
H
Br
MeO
Br Br
MeO
MeO
MeO
OMe
H O
H
OH
HO
OH
OH
OH
OH
OH
OH
10: ampelopsin F
H
HO 11: pallidol
H
OH OH
HO
HO
82: nepalensinol B
HO
H
OH
2. BBr3 (80%)
HO
HO
HO
1. (TMS)3SiH, AIBN
OMe
Br
O
2. BBr3 (63%)
OMe
OMe
Br
1. H2, Pd/C
OMe
Bromine-induced cyclization cascades to fashion pallidol (11) and ampelopsin F (10) from appropriate precursors.
OH HO 80: cararosinol C (26% overall)
H
77
B
74
HO
MeO
Scheme 19.15
MeO
MeO
A
MeO
B
OMe
19.3 Stepwise Synthetic Approaches 713
714
19 Synthetic Approaches to the Resveratrol-Based Family of Oligomeric Natural Products OMe MeO
OMe OMe
MeO
1. [O] 2. O O
OMe
F3C
OH
OMe
MeO
p -TsOH, ∆
[O]
(96%)
(96%)
MeO
MeO
OMe
OMe O
MeO 85
OMe
83: protected hemsleyanol E
66
MeO
O
HO
FriedelCrafts alkylation MeO (34%)
OMe
O
O
MeO MeO OMe 84: protected diptoindonesin D
Scheme 19.16 Preparation of seven-membered rings for protected natural products such as diptoindonesin D (84) from key intermediate 66.
Although this reaction appears simple, it proved exceedingly difficult to achieve. Indeed, no other epoxidation reagent afforded the desired product, and only through the use of in situ generated 1,1,1-trifluorodimethyldioxirane was the reaction successful. We surmise that methyl ether protection of the phenols may be responsible, as efforts with biologically-derived materials have proven easier to epoxidize [18]. Nevertheless, through further simple oxidation chemistry we have been able to produce a protected form of diptoindonesin D (84) and are well on the way to many of the other seven-membered ring natural products mentioned above. Intriguingly, if one exposes molecules like 83 to strong acid in non-polar solvents, the alcohol can be ionized and the neighboring aryl group will rearrange through a phenonium shift to afford natural product analog structures such as 85; to date, no such structure has been isolated from Nature even though the free-phenol form of 83 is a known natural product.3) Thus, the sequence affords access to additional structural diversity that, at least at present, Nature does not seem to sample. Current work within our group is directed towards evaluating what additional frameworks can be built starting from the same intermediates, and, interestingly, some other research teams have already begun to explore such pathways with our key materials. For instance, the Kraus group [37] has recently used the oxidized form of 66 (i.e., 86, Scheme 19.17) to fashion the dihydrofuran ring of amurensin H (90) [38] through a phosphine-base catalyzed condensation reaction. 3) This phenonium shift only occurs at ele-
vated temperatures in non-polar solvents, which may explain why Nature does not
appear to have such rearranged materials though the presence of the alcohol itself implies the potential for facile ionization.
19.3 Stepwise Synthetic Approaches MeO OMe
OMe
OMe
MeO
MeO
O
OMe O
OMe O
1. BBr3 2. NaH,
Br
MeO (76% overall)
OMe
OMe
86
87
P4-t -Bu, benzene, 170 °C, sealed tube
RO
MeO
O
O
OR
OMe OH
OR
(42%)
OMe
OR OR 89: R = Me 90: R = H, amurensin H
OMe OMe BBr3 (67%)
88
Scheme 19.17 Use of the same key intermediate by the Kraus group to accomplish a total synthesis of the furan-containing natural product amurensin H (90).
Beyond resveratrol-based oligomers, however, we have also initiated several programs seeking to determine what other polyphenolic natural products can be prepared from the same key starting materials (i.e., structures of general form 66, cf. Scheme 19.13). Schemes 19.18 and 19.19 present some of our more advanced work, efforts that have been directed toward three non-resveratrol-based natural products. In the first of these schemes, a series of Friedel–Crafts cyclizations provided the means to convert 91 into the natural product cassigarol B (97) [39] in two to four steps, while other chemoselective sequences afforded the means to access two other [3.2.2]-bicycles (98 and 99) [29b]. In the second, the dalesconols (104 and 105) [40], two potent immunosuppressive natural products, were synthesized in several steps from key building block 100; this work featured a cyclization cascade that forged the entire polycyclic core from that critical starting material using the same carbon atom on the B-ring as nucleophile and electrophile, respectively [41]. Alteration in reaction conditions with the same types of starting material afforded the means to access other unique structures, such as 106, in a controlled manner as well. Thus, based on this work it could be argued that intermediates of general form 66 are ‘‘privileged’’ materials for the creation of much structural diversity, both of the resveratrol class and of many other scaffolds of potential biochemical significance [42].
715
B
RO
HO
A
RO
A
OR
98
OH
OH
Scheme 19.18
BBr3 (87%)
OR
OR
OH
and
A
MeO
MeO MeO
HO
HO
HO
95
B
OH
OH
OMe
OMe
(91%)
A OH
OMe
MeO
MeO 94
MeO
OMe OMe 91
C
MeO
HBr in AcOH or PBr3 (58-74%)
OMe
99
C
MeO
B
OMe 1. [O]
OMe
OMe
LiAlH4; aq. HCl
(82%)
OMe 2. p -TsOH
MeO MeO
MeO
MeO
MeO
92
MeO
MeO
A
B
OMe
OMe
O
OMe
OMe
B
OMe 93 [X-ray]
A
C
O
p -TsOH (83%)
C
Total synthesis of cassigarol B (97) and allied structures (98 and 99) from key intermediate 91.
96: R = Me 97: R = H, cassigarol B
RO
HO
C
HO
OH
716
19 Synthetic Approaches to the Resveratrol-Based Family of Oligomeric Natural Products
19.4 Conclusions X
X OMOM OMe
MeO OH
OMOM
A MeO OH
H2, Pd/C EtOH/ EtOAc
BnO
D
H
OMe
102 PhI(OAc)2
MeO
O
(32% overall)
O
OMe O
H HO O
A X
101
HO
OMe
E
MOMO
C
X = OMe or H MeO
B
C
MeO
MeO
100
OMe
D
MeO HO
TFA, TFE, −45 °C, 15 min
B HO
MeO
E
717
OMe
HO 106 [X-ray obtained]
OH O R 104: R = H, dalesconol A 105: R = OH, dalesconol B
MeO
OMe
MOMO 103
Scheme 19.19 Use of key precursor 100 to access dalesconols A and B via a cascade-based cyclization sequence, as well as other unique non-natural frameworks (such as 106).
19.4 Conclusions
As this chapter has, hopefully, demonstrated, there is a wealth of chemistry that can be performed with the reactive functionality of resveratrol, chemistry that leads to an impressive array of stereochemical and architectural complexity. Biomimetic studies have offered insights into how that complexity is made in Nature, and rational synthetic approaches have begun to afford various perspectives into how that complexity can be selectively achieved, largely based on the selection of appropriate building blocks. As such, the future for the synthesis of resveratrol-based oligomers certainly seems bright. In addition, what may even be brighter is the wealth of biochemical knowledge garnered from the thorough evaluation of resveratrol’s larger and arguably more complex oligomeric forms which those synthetic approaches can now fuel.
Acknowledgments
The author would like to thank all of the undergraduate, graduate, and postdoctoral students within his group (past, present, and future) for their work on this exciting family of natural products and for their many contributions, both experimental and intellectual, which has made this project a true joy to pursue. Special thanks
X
718
19 Synthetic Approaches to the Resveratrol-Based Family of Oligomeric Natural Products
are noted to Mr. Adel ElSohly, Ms. Yunqing Lin, and Mr. Daniel Wespe for their comments on the manuscript as well as assistance in its preparation. Financial support for our endeavors in the resveratrol field from the National Institutes of Health (R01-GM84994), the Research Corporation for Science Advancement (Cottrell Scholar award to S.A.S.), the Alfred P. Sloan Foundation (Research Fellowship to S.A.S.), Eli Lilly (Grantee award to S.A.S.), and Bristol-Myers Squibb are gratefully acknowledged.
References 1. Resveratrol was originally isolated
from the roots of white hellebore, and later, from the roots of the Japanese knotweed: (a) Takaoka, M. (1940) Proc. Imp. Acad. Tokyo, 16, 405–407; (b) Hillis, W.E. and Inoue, T. (1967) Phytochemistry, 6, 59–67. 2. For seminal and representative work, see the following, and references cited therein: (a) Jang, M., Cai, L., Udeani, G.O., Slowing, K.V., Thomas, C.F., Beecher, C.W.W., Fong, H.H.S., Farnsworth, N.R., Kinghorn, A.D., Mehta, R.G., Moon, R.C., and Pezzuto, J.M. (1997) Science, 275, 218–220; (b) Howitz, K.T., Bitterman, K.J., Cohen, H.Y., Lamming, D.W., Lavu, S., Wood, J.G., Zipkin, R.E., Chung, P., Kisielewski, A., Zhang, L.-L., Scherer, B., and Sinclair, D.A. (2003) Nature, 425, 191–196; (c) Walle, T., Hsieh, F., DeLegge, M.H., Oatis, J.E., and Walle, U.K. (2004) Drug Metab. Dispos., 32, 1377–1382; (d) Wood, J.G., Rogina, B., Lavu, S., Howitz, K., Helfand, S.L., Tatar, M., and Sinclair, D. (2004) Nature, 430, 686–689; (e) Szewczuk, L.M., Forti, L., Stivala, L.A., and Penning, T.M. (2004) J. Biol. Chem., 279, 22727–22737; (f ) Szewczuk, L.M., Lee, S.H., Blair, I.A., and Penning, T.M. (2005) J. Nat. Prod., 68, 36–42; (g) Chen, G., Shan, W., Wu, Y., Ren, L., Dong, J., and Ji, Z. (2005) Chem. Pharm. Bull., 53, 1587–1590; (h) Baur, J.A., Pearson, K.J., Price, N.L., Jamieson, H.A., Lerin, C., Kalra, A., Prabhu, V.V., Allard, J.S., Lopez-Lluch, G., Lewis, K., Pistell, P.J., Poosala, S., Becker, K.G., Boss, O., Gwinn, D., Wang, M., Ramaswamy, S.,
Fishbein, K.W., Spencer, R.G., Lakatta, E.G., Le Couteur, D., Shaw, R.J., Navas, P., Puigserver, P., Ingram, D.K., de Cabo, R., and Sinclair, D.A. (2006) Nature, 444, 337–342; (i) Milne, J.C., Lambert, P.D., Schenk, S., Carney, D.P., Smith, J.J., Gagne, D.J., Jin, L., Boss, O., Perni, R.B., Vu, C.B., Bemis, J.E., Xie, R., Disch, J.S., Ng, P.Y., Nunes, J.J., Lynch, A.V., Yang, H., Galonek, H., Israelian, K., Choy, W., Iffland, A., Lavu, S., Medvedik, O., Sinclair, D.A., Olefsky, J.M., Jirousek, M.R., Elliott, P.J., and Westphal, C.H. (2007) Nature, 450, 712–717; (j) Boocock, D.J., Faust, G.E.S., Patel, K.R., Schinas, A.M., Brown, V.A., Ducharme, M.P., Booth, T.D., Crowell, J.A., Perloff, M., Gescher, A.J., Steward, W.P., and Brenner, D.E. (2007) Cancer Epidemiol. Biomarkers Prev., 16, 1246–1252; (k) Kim, D., Nguyen, M.D., Dobbin, M.M., Fischer, A., Sananbenesi, F., Rodgers, J.T., Delalle, I., Baur, J.A., Sui, G., Armour, S.M., Puigserver, P., Sinclair, D.A., and Tsai, L.-H. (2007) EMBO J., 26, 3169–3179; (l) Heiss, E.H., Schilder, Y.D.C., and Dirsch, V.M. (2007) J. Biol. Chem., 282, 26759–26766. 3. For selected reviews, see: (a) Saiko, P., Szakmary, A., Jaeger, W., and Szekeres, T. (2007) Rev. Mutation Res., 658, 68–94; (b) Athar, M., Back, J.H., Tang, X., Kim, K.H., Kopelovich, L., Bickers, D.R., and Kim, A.L. (2007) Toxicol. Appl. Pharmacol., 224, 274–278; (c) Kopp, P. (1998) Eur. J. Endocrinol., 138, 619–620; (d) Soleas, G.J., Diamandis, E.P., and Goldberg, D.M. (1997) Clin. Biochem., 30, 91–113.
References 4. For recent work, see: (a) Favaron, F.,
Lucchetta, M., Odorizzi, S., Pais da Cunha, A.T., and Sella, L. (2009) J. Plant Pathol., 91, 579–588; (b) van Baarlen, P., Legendre, L., and van Kan, J.A.L. (2004) Botrytis: Biology, Pathology and Control (eds Y. Elad, B. Williamson, P. Tudzynski, and N. Delen), Kluwer Academic Publishers, Dordrecht, pp. 143–161 5. For selected references regarding the isolation and biological activity of the molecules in Figure 19.1, see: (a) Langcake, P. and Pryce, R.J. (1977) Experientia, 33, 151–152; (b) Diyasena, M.N.C., Sotheeswaran, S., Surendrakumar, S., Balasubramanian, S., Bokel, M., and Kraus, W. (1985) J. Chem. Soc., Perkin Trans. 1, 1807–1809; (c) Oshima, Y., Ueno, Y., Hikino, H., Yang, L.L., and Yen, K.Y. (1990) Tetrahedron, 46, 5121–5126; (d) Kurihara, H., Kawabata, J., Ichikawa, S., and Mizutani, J. (1990) Agric. Biol. Chem., 54, 1097–1099; (e) Kitanaka, S., Ikezawa, T., Yasukawa, K., Yamanouchi, S., Takido, M., Sung, H.K., and Kim, I.H. (1990) Chem. Pharm. Bull., 38, 432–435; (f ) Dai, J.-R., Hallock, Y.F., Cardellina, J.H., and Boyd, M.R. (1998) J. Nat. Prod., 61, 351–353; (g) Ohyama, M., Tanaka, T., Ito, T., Iinuma, M., Bastow, K.F., and Lee, K.-H. (1999) Bioorg. Med. Chem. Lett., 9, 3057–3060; (h) Tanaka, T., Ito, T., Nakaya, K., Iinuma, M., and Riswan, S. (2000) Phytochemistry, 54, 63–69; (i) Tanaka, T., Iliya, I., Ito, T., Furusawa, M., Nakaya, K., Iinuma, M., Shirataki, Y., Matsuura, N., Ubukata, M., Murata, J., Simozono, F., and Hirai, K. (2001) Chem. Pharm. Bull., 49, 858–862; (j) Luo, H.-F., Zhang, L.-P., and Hu, C.-Q. (2001) Tetrahedron, 57, 4849–4854; (k) Tanaka, T., Ito, T., Nakaya, K., Iinuma, M., Takahashi, Y., Naganawa, H., and Riswan, S. (2001) Heterocycles, 55, 729–740; (l) Huang, K.-S., Lin, M., and Cheng, G.-F. (2001) Phytochemistry, 58, 357–362; (m) Takaya, Y., Yan, K.-X., Terashima, K., Ito, J., and Niwa, M. (2002) Tetrahedron, 58, 7259–7265; (n) Ito, T., Tanaka, T., Iinuma, M., Iliya, I.,
Nakaya, K., Ali, Z., Takahashi, Y., Sawa, R., Shirataki, Y., Murata, J., and Darnaedi, D. (2003) Tetrahedron, 59, 5347–5363; (o) Ito, T., Tanaka, T., Iinuma, M., Nakaya, K., Takahashi, Y., Sawa, R., Naganawa, H., and Chelladurai, V. (2003) Tetrahedron, 59, 1255–1264; (p) Ito, T., Akao, Y., Yi, H., Ohguchi, K., Matsumoto, K., Tanaka, T., Iinuma, M., and Nozawa, Y. (2003) Carcinogenesis, 24, 1489–1497; (q) Ito, T., Tanaka, T., Iinuma, M., Nakaya, K., Takahashi, Y., Sawa, R., Murata, J., and Darnaedi, D. (2004) J. Nat. Prod., 67, 932–937; (r) Supudompol, B., Likhitwitayawuid, K., and Houghton, P.J. (2004) Phytochemistry, 65, 2589–2594; (s) Sahadin Hakim, H.H., Juliawaty, L.D., Syah, Y.M., bin Din, L., Ghisalberti, E.L., Latip, J., Said, I.M., and Achmad, S.A. (2005) Z. Naturforsch., Teil C: J. Biosci., 60, 723–727; (t) Ohguchi, K., Akao, Y., Matsumoto, K., Tanaka, T., Ito, T., Iinuma, M., and Nozawa, Y. (2005) Biosci. Biotechnol. Biochem., 69, 353–356; (u) Wang, S., Ma, D., and Hu, C. (2005) Helv. Chem. Acta, 88, 2315–2321; (v) Guebailia, H.A., Chira, K., Richard, T., Mabrouk, T., Furiga, A., Vitrac, X., Monti, J.-P., Delaunay, J.-C., and Merillon, J.-M. (2006) J. Agric. Food Chem., 54, 9559–9564; (w) Yamada, M., Hayashi, K., Ikeda, S., Tsutsui, K., Tsutsui, K., Ito, T., Iinuma, M., and Nozaki, H. (2006) Biol. Pharm. Bull., 29, 1504–1507. 6. For selected examples of recent isolations, see: (a) He, S., Jiang, L., Wu, B., Li, C., and Pan, Y. (2009) J. Org. Chem., 74, 7966–7969; (b) Ito, T., Abe, N., Oyama, M., Tanaka, T., Murata, J., Darnaedi, D., and Iinuma, M. (2009) Helv. Chim. Acta, 92, 1203–1216. 7. For a recent review discussing the generation of varied structures in the terpene collection of structures, see: Fischbach, M.A. and Clardy, J. (2007) Nat. Chem. Biol., 3, 353–355. 8. For leading references, see: (a) Zhang, Y., Li, S.-Z., Li, J., Pan, X., Cahoon, R.E., Jaworski, J.G., Wang, X., Jez, J.M.,
719
720
19 Synthetic Approaches to the Resveratrol-Based Family of Oligomeric Natural Products
9.
10.
11. 12.
13.
14. 15.
16.
17.
18.
Chen, F., and Yu, O. (2006) J. Am. Chem. Soc., 128, 13030–13031; (b) Lee, M.S. and Pyee, J.H. (2004) Nat. Prod. Sci., 10, 248–251. For example, investigators have been able to separately isolate both antipodes of the natural product davidiol A from different plant species: (a) Tanaka, T., Ito, T., Ohyama, M., Ichise, M., and Tateishi, Y. (2000) Phytochemistry, 53, 1009–1014; (b) He, Y.-H., Takaya, Y., Terashima, K., and Niwa, M. (2006) Heterocycles, 68, 93–100. For selected examples of the proposed equilibration step, see: (a) Kurosawa, W., Kobayashi, K., Kan, T., and Fukuyama, T. (2004) Tetrahedron, 60, 9615–9628; (b) Lian, Y. and Hinkle, R.J. (2006) J. Org. Chem., 71, 7071–7074. Langcake, P. and Pryce, R.J. (1977) J. Chem. Soc., Chem. Commun., 208–210. Sako, M., Hosokawa, H., Ito, T., and Iinuma, M. (2004) J. Org. Chem., 69, 2598–2600. Takaya, Y., Terashima, K., Ito, J., He, Y.-H., Tateoka, M., Yamaguchi, N., and Niwa, M. (2005) Tetrahedron, 61, 10285–10290. Yao, C.-S., Lin, M., and Wang, Y.-H. (2004) Chin. J. Chem., 22, 1350–1355. For one example of a resveratrol-derived oligomer with carbohydrates attached, see: Ito, T., Abe, N., Oyama, M., and Iinuma, M. (2009) Tetrahedron Lett., 50, 2516–2520. (a) Li, W., Li, H., and Hou, Z. (2006) Angew. Chem. Int. Ed., 45, 7609–7611; for a recent attempt to further explore such starting materials in the synthesis of resveratrol oligomers, see: (b) Li, W., Li, H., Luo, Y., Yang, Y., and Wang, N. (2010) Synthesis, 1247–1250; (c) Li, W., Li, H., Wang, A.X., Luo, Y., and Zang, P. (2010) J. Chem. Res. (S), 118–120. Velu, S.S., Thomas, N.F., Buniyamin, I., Ching, L.K., Feroz, F., Noorbatcha, I., Gee, L.C., Awang, K., Abd. Wahab, I., and Weber, J.F.F. (2008) Chem. Eur. J., 14, 11376–11384. (a) Takaya, Y., Yan, K.-X., Terashima, K., He, Y.-H., Niwa, M. (2009) Tetrahedron, 58, 9265–9271; (b) Niwa, M., Ito, J.,
19.
20.
21.
22.
23.
24. 25.
Terashima, K., Koizumi, T., Takaya, Y., and Yan, K.-X. (2000) Heterocycles, 53, 1475–1478. (a) Aguirre, J.M., Alesso, E.N., and Moltrasio Iglesias, G.T. (1999) J. Chem. Soc., Perkin Trans. 1, 1353–1358; (b) Aguirre, J.M., Aleso, E.N., Iba˜ nez, A.F., Tombari, D.G., and Moltrasio Iglesias, G.Y. (1989) J. Heterocycl. Chem., 26, 25–27. For selected examples of other, resveratrol-like, biomimetic dimerization attempts, see: (a) Li, X.-C. and Ferreira, D. (2003) Tetrahedron, 59, 1501–1507; (b) Thomas, N.F., Lee, K.C., Paraidathathu, T., Weber, J.F.F., Awang, K., Rondeau, D., and Richomme, P. (2002) Tetrahedron, 58, 7201–7206; (c) Thomas, N.F., Lee, K.C., Paraidathathu, T., Weber, J.F.F., and Awang, K. (2002) Tetrahedron Lett., 43, 3151–3155; (d) Engler, T.A., Gfesser, G.A., and Draney, B.W. (1995) J. Org. Chem., 60, 3700–3706; (e) Engler, T.A., Draney, B.W., and Gfesser, G.A. (1994) Tetrahedron Lett., 35, 1661–1664. (a) Seiple, I.B., Su, S., Young, I.S., Lewis, C.A., Yamaguchi, J., and Baran, P.S. (2010) Angew. Chem. Int. Ed., 49, 1095–1098; (b) Su, S., Seiple, I.B., Young, I.S., and Baran, P.S. (2008) J. Am. Chem. Soc., 130, 16490–16491; (c) O’Malley, D.P., Yamaguchi, J., Young, I.S., Seiple, I.B., and Baran, P.S. (2008) Angew. Chem. Int. Ed., 47, 3581–3583. For selected syntheses of resveratrol itself, see: (a) Botella, L. and N´ajera, C. (2004) Tetrahedron, 60, 5563–5570; (b) Takaya, Y., Terashima, K., Ito, J., He, Y.-H., Tateoka, M., Yamaguchi, N., and Niwa, M. (2005) Tetrahedron, 61, 10285–10290; (c) Chen, G., Shan, W., Ren, L., Dong, J., and Ji, Z. (2005) Chem. Pharm. Bull., 53, 1587–1590 (a) Jeffrey, J.L. and Sarpong, R. (2009) Org. Lett., 11, 5450–5453; (b) Jeffrey, J.L. and Sarpong, R. (2009) Tetrahedron Lett., 50, 1969–1972. Kim, I. and Choi, J. (2009) Org. Biomol. Chem., 7, 2788–2795. (a) Saraswathym, A., Puroshothaman, K.K., Patra, A., Dey, A.K., and Kundu,
References
26.
27.
28.
29.
30.
31.
32. 33.
A.B. (1992) Phytochemistry, 31, 2561–2562; (b) Ge, H.M., Huang, B., Tan, S.H., Shi, D.H., Song, Y.C., and Tan, R.X. (2006) J. Nat. Prod., 69, 1800–1802. For a recent review on bismuth(III) chemistry, see: Salvador, J.A.R., Pinto, R.M.A., and Silvestre, S.M. (2009) Curr. Org. Synth., 6, 426–470. (a) Ge, H.M., Zhu, C.H., Shi, D.H., Zhong, L.D., Xie, D.Q., Yang, J., Ng, S.W., and Tan, R.X. (2008) Chem. Eur. J., 14, 376–381; (b) Ge, H.M., Xu, C., Wang, X.T., Huang, B., and Tan, R.X. (2006) Eur. J. Org. Chem., 5551–5554. Nicolaou, K.C., Wu, T.R., Kang, Q., and Chen, D.Y.-K. (2009) Angew. Chem. Int. Ed., 48, 3440–3443; (b) for the full account of this work, see: Nicolaou, K.C., Kang, Q., Wu, T.R., Lim, C.S., and Chen, D.Y.-K. (2010) J. Am. Chem. Soc., 132, 7540–7548. (a) Snyder, S.A., Zografos, A.L., and Lin, Y. (2007) Angew. Chem. Int. Ed., 46, 8186–8191; (b) Snyder, S.A., Breazzano, S.P., Ross, A.G., Lin, Y., and Zografos, A.L. (2009) J. Am. Chem. Soc., 131, 1753–1765. For a recent review on this reaction, see: Taylor, R.J.K. (1999) Chem. Commun., 217–227. For a leading introduction to the reversibility of bromonium formation, see: Brown, R.S. (1997) Acc. Chem. Res., 30, 131–137. Yang, G., Zhou, J., Li, Y., and Hu, C. (2005) Planta Med., 71, 569–571. For a review on the concept of atom economy, see: Trost, B.M. (1991) Science, 254, 1471.
34. For an interesting commentary on utiliz-
35.
36.
37. 38.
39.
40.
41.
42.
ing bromine as a protective device, see: Effenberger, F. (2002) Angew. Chem. Int. Ed., 41, 1699–1700. Yamada, M., Hayashi, K., Hayashi, H., Ikeda, S., Hoshino, T., Tsutsui, K., Tsutsui, M., Iinuma, M., and Nozaki, H. (2006) Phytochemistry, 67, 307–313. These conditions were taken from past efforts to synthesize the epothilones: Nicolaou, K.C., He, Y., Vourloumis, D., Vallberg, H., Roschanger, F., Sarabia, F., Ninkovic, A., Yang, Z., and Trujillo, J.I. (1997) J. Am. Chem. Soc., 119, 7960–7973. Kraus, G.A. and Gupta, V. (2009) Tetrahedron Lett., 50, 7180–7183. Li, Y., Yao, C., Bai, J., Lin, M., and Cheng, G. (2006) Acta Pharm. Sinica, 27, 735–740. Baba, K., Maeda, K., Tabata, Y., Doi, M., and Kozawa, M. (1988) Chem. Pharm. Bull., 36, 2977–2983. (a) Zhang, Y.L., Ge, H.M., Zhao, W., Dong, H., Xu, Q., Li, S.H., Li, J., Zhang, J., Song, Y.C., and Tan, R.X. (2008) Angew. Chem. Int. Ed., 47, 5823–5826; (b) Wen, L., Cai, X., Xu, F., She, Z., Chan, W.L., Vrijmoed, L.L.P., Jones, E.B.G., and Lin, Y. (2009) J. Org. Chem., 74, 1093–1098. Snyder, S.A., Sherwood, T.C., and Ross, A.G. (2010) Angew. Chem. Int. Ed., 49, 5146–5150. For a recent review on this concept, see: Welsch, M., Snyder, S.A., and Stockwell, B.R. (2010) Curr. Opin. Chem. Biol., 14, 347–361.
721
723
20 Sequential Reactions Initiated by Oxidative Dearomatization. Biomimicry or Artifact? Stephen K. Jackson, Kun-Liang Wu, and Thomas R.R. Pettus
20.1 Overview
Robert Robinson first introduced the concept of biomimetic synthesis in 1917 with a one-pot preparation of tropinone [1]. This remarkable transformation joins three reactants together, capitalizing upon the reactive proclivities of polyketides. Because aromatic compounds themselves are the end point of polyketide biosynthetic sequences, which are followed by the action of a few specific tailoring decarboxylative and oxidative enzymes, it is hard to conclude that sequences commencing with dearomatization are indeed ‘‘biomimetic’’ [2]. However, sequential reactions initiated by oxidative phenol dearomatization can often appear as ‘‘biomimetic,’’ reaching very complex molecular architectures in short order because the functionality and the stereo- and regio-chemical tendencies that are programmed into the polyketide intermediates are still efficacious in the corresponding phenol derivatives and in the reactions stemming from these compounds. In this chapter, we examine some of the sequential one-pot reactions initiated by oxidative dearomatization. This discussion is by no means a comprehensive treaty on this topic. However, we attempt to untangle this subject for the benefit of interested chemists by focusing on one-pot oxidative dearomatization sequences commencing from phenols that result in the formation or cleavage of three or more C–C, C–O, or C–N bonds. In all of the following schemes, the intermediate formed by the initial oxidative dearomatization undergoes further chemical reactions involving the formation and breakage of two or more bonds. Thus, the examples presented are different from most rudimentary oxidative dearomatization reactions that most often result in the usual quinol and ketal adducts [3].
20.2 Oxidative Dearomatization Sequences and the Initial Intermediate
The processes that we will describe involve some type of phenoxonium cation [4], a p-quinone methide [5], an o-quinone methide [6], or their corresponding p- and Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
724
20 Sequential Reactions Initiated by Oxidative Dearomatization. Biomimicry or Artifact?
O + + R δ δ R
O O R
δ+ R
phenol
oxidation
R B p -quinone methide
A phenoxonium cation
C o -quinone methide two or more
O
products
bonds formed or broken
O O
R O D p -quinone
Scheme 20.1 events.
R E o -quinone
The initial species formed by phenol oxidation that can lead to further
o-quinones (Scheme 20.1, reactive intermediates A–E). These species can lead to a lengthy sequence of reactions, just so long as the subsequent intermediates are themselves compatible with the initial oxidative conditions. The reactivity of each species can be further expanded by an attached π-framework of conjugated sp2 R-substituents so as to augment the availability of reaction manifolds. In the following sections, we discuss some sequences leading from the above five species that follow oxidative dearomatization. These subsequent transformations include intermolecular dimerizations, as well as successive intermolecular [4+2], [3+2], and [5+2] cycloadditions (Sections 20.3 and 20.4). In addition, there are numerous examples of intramolecular cascade sequences including 1,4-additions, [5+2], [4+2], [3+2] cycloadditions, and subsequent aldol-like processes (Sections 20.5 and 20.6). There are also several examples where oxidative dearomatization to one of the above species is then followed by a series of tautomerizations, sigmatropic shifts (Section 20.7), and further reaction. We will also present examples where oxidation is followed by ring rupture, resulting in either smaller rings or larger rings (Sections 20.8 and 20.9), which can even incorporate additional intramolecular and intermolecular events (Section 20.10). Finally, we present some natural products presumed to form through phenol oxidative pathways (Section 20.11). Our concluding opinions on the biomimetic authenticity of presented schemes and future of this area of research are provided in the last section.
20.3 Intermolecular Dimerizations
The most memorable example of a phenol oxidative cascade is likely Chapman’s 1971 synthesis of carpanone (4) [7]. The cascade sequence can be viewed as commencing with a palladium-promoted oxidation of the starting vinyl phenol 2
20.3 Intermolecular Dimerizations
O
O
HO
O 1
O
O
O
O
O
O
O
O
O O
O
H 46%
O
O
H
O
2 3
Scheme 20.2
725
carpanone (4)
Chapman’s synthesis of carpanone.
to its corresponding resonance stabilized phenoxonium cation 1, which succumbs to a vinylogous diastereoselective aldol-like process with the original phenol 2 (Scheme 20.2). This reaction affords the dimeric o-quinone methide intermediate 3, which undergoes a subsequent diastereoselective intramolecular Diels–Alder reaction to produce carpanone (4). The cascade is somewhat unusual in that it involves both inter- and intra-molecular components: a phenoxonium undergoing remote addition and, subsequently, the dimerization of two o-quinone methides resulting in the formation of two carbon–carbon bonds and one carbon–oxygen bond. In a more recent related example, Antus has shown that oxidative dimerization of isoeugenol (5) leads to dehydrodiisoeugenol (9) in a single operation (Scheme 20.3) [8]. Treatment of 5 with phenyl iodine diacetate (PIDA) results in the formation of p-phenoxonium 6, the resonance structure (7) of which is intercepted by a second equivalent of isoeugenol (5). Nucleophilic ring closure then gives rise to the formal [3+2] addition product, dehydrodiisoeugenol (9). OMe O
OH
PIDA
OH
CH2Cl2, rt 35 %
OMe
OMe dehydrodiisoeugenol (9)
isoeugenol (5)
OMe 5
OMe
OH
OH
O
O
OMe
OMe
H 6
Scheme 20.3
7
OH
8
Antus’ carpanone related net [3+2] sequence for dehydrodiisoeugenol.
The more common oxidative dimerization cascade processes involve subsequent Diels–Alder dimerization of an intermediate formed by addition of a nucleophile to the phenoxonium species. In particular, these products capitalize upon the inherent preference for the p-quinol and o-quinol intermediate to undergo subsequent dimerization of the same side as the polar C–O bond that is formed from dearomatization. This trait is likely imbued through asymmetric orbital development. Within
OMe
726
20 Sequential Reactions Initiated by Oxidative Dearomatization. Biomimicry or Artifact? OH HO
O
neither isolable HO
OH
oxidation
OH
Pb(OAc)4 then saponification
13
14
OH
O sorbicillin (10)
O
diene: reacts from bottom o - quinol
dienophile: reacts from bottom p - quinol [4 +2]
saponification O
X
O
O HO
O
R
O
OH O
O
O
PhI(OCOCF3)2 H3O+
O
11
OH O
O
HO
oxidation
Scheme 20.4
O
O
R=
12
HO O
R OH
bisorbicillinol (15)
Nicolaou and Pettus’ syntheses of bisorbicillinol from sorbicillin.
this class, bisorbicillinol (15), a dimer of the oxidative dearomatization of sorbicillin (10), is perhaps the most recognized putative natural product involving this hypothetical oxidative dearomatization biomimicry (Scheme 20.4). In the Nicolaou synthesis, sorbicillin undergoes oxidation and saponification to form a tautomeric mixture of non-isolable o- and p-quinols (13 and 14), which then succumb to a regioselective Diels–Alder reaction on the same side of the quinol as the newly installed tertiary alcohol to afford bisorbicillinol (15) in 16% yield from sobicillin (10) [9]. Pettus, on the other hand, installs a glycolic acid derived disposable linker to produce 11 [10]. Oxidation and sequential treatment with aqueous base and acid affords the manageable spiroketal intermediate 12. Further saponification proceeds to the same tautomeric quinol mixture of 13 and 14, producing bisorbicillinol in a somewhat improved 33% yield. Other examples of this kind of dearomatization and [4+2] dimerization can be found throughout the literature [11]. For example, Pettus has shown that ortho oxidation of xylenol 16 with the I(V) reagent 17, also known as o-iodoxybenzoic acid (IBX), affords the I(III) o-quinol derivative 18 (Scheme 20.5) [12]. This o-quinol intermediate undergoes [4+2] dimerization to afford the corresponding dione 19, which succumbs to further reduction to produce 20 in 51% overall yield. O O
Me
I
OH
17 HO O
Me O
I(V)
I
O reductant 18
Scheme 20.5
Me
O I
Me 16
O
Me O
(III)
Me
OR
Me
Me O OR 19: R = C7H4IO2 20: R = H, 51% overall
Pettus’ oxidative dearomatization and dimerization of 2,6-xylenol with IBX.
20.4 Successive Intermolecular Reactions
727
Although it remains debatable as to whether products of oxidative dimerization actually constitute legitimate natural products or are themselves merely artifacts of isolation, many examples exist in the literature. For instance, Takeya has shown that the o-quinone 22 generated from oxidation of demethylsalvicanol (21) affords grandione (23) when heated in the solid state, through a [4+2] cycloaddition pathway (Scheme 20.6) [13]. O
O
OH
O
O
OH HO
OH [O]
OH
O
H
74%
H
O
H
solid state
O O
∆
HO
HO
H (+)-demethylsalvicanol (21)
H
H 22
Scheme 20.6
(+)-grandione (23)
Takeya’s o-quinone dimerization.
20.4 Successive Intermolecular Reactions
Intermediates arising from oxidative dearomatization can also serve as reacting partners in intermolecular cycloadditions. For example, Singh has demonstrated that the Adler–Becker oxidation product 25 of 24 undergoes smooth cycloaddition with methyl acrylate to yield the Diels–Alder adduct 26 in 67% yield in a regio- and diastereoselective fashion (Scheme 20.7) [14]. Further manipulation of adduct 26 resulted in assembly of the diquinane skeleton of ptychanolide. Me Me
O
OH OH
O Me
Me Me 24
Scheme 20.7
aq. NaIO4 aq. MeCN 0 °C to rt 67 %
Me
O
CO2Me
Me
O Me
O
[4+ 2]
Me
Me Me 25
H CO2Me
Me Me ptychanolide O
26
Singh’s application of Adler–Becker oxidation and successive cycloaddition.
A novel oxidative cascade involving an o-quinone methide intermediate was reported recently by Sigman [15]. Exposure of 27 to catalytic oxidation with Pd and the i-PrQuinox ligand affords o-quinone methide 29 through allylic oxidation (Scheme 20.8). The o-quinone methide generated under these very mild conditions then undergoes a [4+2] cycloaddition with enol-ether 28, furnishing adduct 30 with excellent dr and er. Phenoxonium species generated from oxidative dearomatization can often enter multiple reaction pathways depending on the given conditions. For example, Yamamura has reported that anodic oxidation of phenol 31 in the presence of
728
20 Sequential Reactions Initiated by Oxidative Dearomatization. Biomimicry or Artifact? OEt
OH
Me
Pd(MeCN)2Cl2 (4 mol %) CuCl (8 mol %)
Me OH
+ 27
28 O
N
N i -PrQuinox i -Pr
Z:E = 10:1
Scheme 20.8
O
i-PrQuinox (14 mol %) KHCO3 (40 mol %) PhMe, O2, rt
OEt
OEt
H
O 29
Me
O
[4+ 2]
Me Me
O
30
72 % dr = 33.5:1 er = 93:7
Me Me
Sigman’s enantioselective oxidative dearomatization and cycloaddition.
cis-isosafrole (32) affords the [5+2] exo-product 33 in 25% yield (Scheme 20.9) [16]. Interestingly, the isolation of a second product, futoenone (34), from the reaction mixture implies a tandem cationic cascade sequence as a competing event. The use of the trans form of olefin 32, however, affords the corresponding [5+2] product exclusively in 81% yield. O
O
OMe
O
OMe
Me
anodic oxidation
+
TFA/AcOH
O 31
Scheme 20.9
Me
OMe
O
O
32
OMe
+
O
O
OH
O
33 (25%)
O futoenone (34) (15%)
Yamamura’s [5+2] cycloaddition.
Several examples of intermolecular [3+2] reactions initiated by oxidation are known. Recently, Morrow reported that phenyl iodine(bis)trifluoroacetate (PIFA) could promote a [3+2] reaction between the chromene 35 and phenol 36 (Scheme 20.10) [17]. Only para-substituted phenols can be used by this method to generate pterocarpans, such as 37, in appreciable yield. This is a rather unfortunate limitation of this chemistry in light of the fact that naturally occurring pterocarpans typically posses both C8 and C9 oxidation.
OMe
+ MeO
O 35
Scheme 20.10
O
9 8
OMe
PIFA MeCN
HO
49 %
36
MeO
O 37
Morrow’s oxidative dearomatization and successive [3+2] reaction.
Successive oxidation of a heavily substituted phenol was reported by Stoltz in the total synthesis of the norditerpenoid, dichroanone (40) [18]. In the final step of the synthesis, a sequence of oxidations takes place, initiated by IBX. During these events, an o-quinone intermediate stemming from 38 is trapped by pentafluorothiophenol to generate catechol 39, which undergoes further oxidation and addition of hydroxide to afford the p-quinone natural product (Scheme 20.11).
20.5 Intramolecular Cycloadditions OH
OH
HO
OH
O
IBX, CHCl3, rt
[O] KOH
then, C6F5SH, rt 35 %
O
SC6F5
38
39
Scheme 20.11
729
(+)-dichroanone (40)
Stoltz’ successive phenol oxidation.
A related successive oxidation of a catechol substrate has been reported by Sarpong in the synthesis of abrotanone (43) from brussonol (41) [19]. This process involves a four-electron oxidation with the addition of two methanol equivalents in a regioselective fashion. Thus, treatment of brussonol with Cu(NO3 )2 in the presence of morpholine affords the mixed quinone amino acetal 42, which upon subsequent exposure to methoxide gives rise to the natural product 43 (Scheme 20.12). HO
O Cu(NO3)2•3H2O
O
N
H
H brussonol (41)
42
O NaOMe
O
morpholine MeOH
Scheme 20.12
OH
OH
O
OMe H
O
OH
OMe OMe
abrotanone (43)
Sarpong’s successive catechol oxidation.
20.5 Intramolecular Cycloadditions
Liao was among the first to popularize successive oxidation and intramolecular cycloaddition. For instance, the penicillones were accessed through this strategy, beginning with phenol 44 [20]. Treatment of the latter with PIDA in the presence of trans-crotyl alcohol affords o-quinone acetal 45, which slowly undergoes an intramolecular Diels–Alder reaction, giving rise to adduct 46 (Scheme 20.13). The latter was then elaborated to the penicillone B natural product (47). OH Me
OMe
44 Me
Scheme 20.13
PIDA
trans -crotyl alcohol CH2Cl2, rt 87 %
Me Me
O
Me 45
OMe O
2 days
Me Me
O OMe
Me
O 46
Me HO Me O Me
OH O
penicillone B (47)
Liao’s oxidative dearomatization and intramolecular cycloaddition.
Wood has also reported an intramolecular cycloaddition prompted by oxidative dearomatization in the context of studies toward CP-263,115 [21]. Phenol 48, upon oxidation with PIDA, forms a phenoxonium that is intercepted by propargyl alcohol to form the intermediate diene 49, which undergoes a subsequent intramolecular Diels–Alder reaction to afford 50 (Scheme 20.14).
730
20 Sequential Reactions Initiated by Oxidative Dearomatization. Biomimicry or Artifact?
MeO2C
OH
MeO2C n-Bu
O
MeO2C MeO2C n-Bu
PIDA
O
OH
MeO2C MeO2C n-Bu
O
O O
O
95 %
48
49
Scheme 20.14
O
50
Wood’s oxidative dearomatization and intramolecular cycloaddition.
The oxidative dearomatization step may also proceed intramolecularly, as demonstrated by Danishefsky in studies toward the synthesis of the sesquiterpenoid tashironin [22]. Treatment of phenol 51 with PIDA in toluene presumably affords the intermediate o-quinone acetal 52, which undergoes a transannular Diels–Alder reaction to give 53 as a single diastereomer (Scheme 20.15). Interestingly, the benzylic stereocenter in 51 controls the diastereofacial selectivity of the oxidative dearomatization step and, thereby, secures the stereoselectivity of the cycloaddition. Me
OH
O OMe
OH
Me
51
PIDA PhMe, rt 66 %
Me
Scheme 20.15
Me
Me 52
OTs
O OMe O
O
OMe
OTs 53 Me single diastereomer
OTs
Danishefsky’s oxidative dearomatization and transannular cycloaddition.
A remarkable intramolecular cascade was reported recently by Njardarson during studies toward the synthesis of vinigrol [23]. Exposure of phenol 54 to PIDA in hexafluoro-isopropanol (HFIP) led to the isolation of the unexpected compound 55 and the desired Wessely product 56 (Scheme 20.16). Formation of the bicyclic compound 55 is postulated to arise from a polar cationic [5+2] reaction, facilitated by the electron-donating methoxy moiety of the phenol. O
O OH
EtO2C OH
PIDA
CO2Et OMe 54
Scheme 20.16
HFIP, 0 °C 65 % 55/56 = 1:0.9
O MeO O
H O 55
+
O O CO2Et OMe 56 (dr = 2.5:1)
Njardarson’s [5+2] intramolecular cascade.
Other intramolecular oxidative dearomatization cascades involving a [3+2] manifold have also been reported. For example, Ciufolini has shown that aldoximes lead to [3+2] adducts via phenol oxidative dearomatization due to concurrent nitrile oxide formation [24]. Employing an elaborate substrate, Sorensen utilized this method for the construction of the pentacyclic core of the cortistatins [25]. Treatment of compound 57 with PIDA results in the formation of presumed nitrile oxide 58, which undergoes an intramolecular [3+2] cycloaddition, affording adduct 59 in good yield (Scheme 20.17).
20.6 Other Successive Intramolecular Cascade Sequences
Me OTBS
Me OTBS
Me OTBS
731
PIDA, TFE
HO
N
rt, then 50 °C 80 %
H
HO
O
N
H
O
57
O
N O
58
59
O
O
H
OH
Scheme 20.17
Sorensen’s oxidative dearomatization and intramolecular cycloaddition.
20.6 Other Successive Intramolecular Cascade Sequences
Pettus has reported a cascade involving the oxidation of various resorcinol amide derivatives, such as 60 with PIFA [26]. The sequence proceeds to the lactone intermediate 61, presumably by addition of the amide carbonyl to the phenoxonium followed by hydrolysis of the intermediate iminium species by addition of water at the conclusion of the reaction (Scheme 20.18). In this instance, however, an oxygen atom belonging to the neighboring nitro substituent adds in 1,4-fashion to the vinylogous ester to produce the propeller structure 62 [27]. O
N O
TIPSO
then H2O
NO2
OH
–
O 1,4-add.
O
TIPSO
O
TIPSO
O N O O O
O
13%
61
60
Scheme 20.18
O O2N
PhI(OCOCF3)2
62
Pettus’ oxidative dearomatization with successive C–O bond formations.
Very recently, Pettus reported a remarkable intramolecular [5+2] cycloaddition initiated by oxidative dearomatization (Scheme 20.19). The synthesis commences HO
O
O Pb(OAc)4 61 %
O
O 2 steps
(±)-cedrene (68)
Scheme 20.19
65
64
(±)-curcuphenol (63)
OAc
67
66
Pettus’ dearomatization with successive [5+2] and C–O bond formation.
O
732
20 Sequential Reactions Initiated by Oxidative Dearomatization. Biomimicry or Artifact?
with a one-pot synthesis of curcuphenol (63) from 4-methylsalicaldehyde via an o-quinone methide intermediate [28]. Subsequent exposure of phenol 63 to the oxidant lead tetraacetate [Pb(OAc)4 ] proceeds through the intermediacy of phenoxonium 64, the tertiary cation 65 and the allyl cation 66 to afford the tricyclic compound 67 by eventual steric controlled interception by a residual acetate ion. This material has been successfully converted into cedrene (68) in two further operations. Heathcock has reported a facile synthesis of styelsamine B (71) from aniline 69 and catechol 70 (Scheme 20.20) [29]. The sequence, which undoubtedly resembles the biosynthesis, involves an o-quinone intermediate that succumbs to successive imine formation and intramolecular 1,4-addition of the aniline nitrogen atom. Further oxidation, aldol reaction, and dehydration produce 71 (Scheme 20.20). This compound undergoes a further series of oxidations to produce cystodytin J (72). OH
BrH•H2N
OH
O NH2•HBr 69
OH
1,4-addition, aldol reaction
70
N
NH oxidation, imine formation
O oxidations
N
N H
NHAc
NHAc
NHAc styelsamine B (71)
Scheme 20.20
cystodytin J (72)
Heathcock’s synthesis of styelsamine B.
The oxidative dearomatization of symmetrical p-substituted phenols produces meso-cyclohexadienones. Various two-step methods have been reported to desymmetrize these compounds into enantioenriched products [30]. An impressive one-pot, catalytic protocol was reported by Gaunt that makes use of a pyrrolidinecatalyzed intramolecular desymmetrizing 1,4-addition to the meso-intermediate [31]. Phenolic substrates of type 73 are treated with PIDA in the presence of pyrrolidine catalyst 74 to afford the dearomatized dienone 75 (Scheme 20.21). Subsequent enamine formation and Michael addition affords annulated products of type 76, many of which in high yield and stereoselectivity. Rodr´ıguez and coworkers have recently shown that coatline B (77) undergoes a fast and irreversible reaction in slightly alkaline water to afford matlaline (82) in near quantitative yield [32]. This fascinating process – the historic backdrop of
N H
OH
OTMS Ar Ar
O
O
74 (10 mol %)
CHO X 73
n
PIDA (1 eq.) MeOH, 0 °C
n = 0, 1, 2 X = CH2, O, N-Ts
Scheme 20.21
H CHO MeO
MeO X
n 75
X
CHO
n
76 52 - 84 % dr up to 20:1 ee up to 99 %
Gaunt’s oxidative dearomatization and in situ desymmetrization.
20.7 Successive Tautomerizations and Rearrangements O
OH HO
HO
O
OH [O]
OH
R
OH
R
OH
R HO O
OH O
OH O
OH
O
H2O OH 79
78
coatline B (77) HO R = HO HO
O OH OH
O
O
OH
OH O
HO R
matlaline (82)
O CO2H
O O
HO R
[O]
OH
OH HO
CO2H 81
OH CO2H
R 80
Scheme 20.22 Rodr´ıguez’s oxidative biaryl formation followed by furan and pyran formation.
which is equally intriguing – is thought to occur with a molecular oxygen induced oxidation of the catechol moiety in 77 to the o-quinone 78 (Scheme 20.22). This oxidation event was shown to be extremely facile, an unusual observation given that catechols are generally oxidized either enzymatically or with strong oxidant. Nucleophilic attack on the reactive quinone of 78 affords putative spiro-annulated intermediate 79, which undergoes a retro-Friedel–Crafts reaction to give 80. A second oxidation would generate o-quinone 81, leading to the blue fluorescent ‘‘natural’’ product matlaline (82). This process is also entirely stereoselective, apparently driven by the α-hydroxy stereogenic center of coatline B (77) and not by the glucopyranosyl (R) residue.
20.7 Successive Tautomerizations and Rearrangements
There are several examples in the literature where the first intermediate derived from phenolic dearomatization undergoes further rearrangement prior to succumbing to a second reaction event. For example, Porco has shown that treatment of phenoxide 83 with a sparteine/Cu reagent produces o-quinol intermediate 84, which undergoes a stereospecific 1,2-shift to isomeric o-quinol 85 (Scheme 20.23) [33]. The latter then spontaneously dimerizes in a [4+2] sense affording bicyclo[2.2.2]octenone 86 with superb enantioselectivity. In this case, the monomer 84 resists dimerization until rearrangement to the more reactive unsubstituted olefin in 85 takes place. During studies on a biomimetic ring closure towards the synthesis of the morphinan alkaloids, Feldman observed an interesting series of tautomeric shifts of the oxidized products of stilbene derivatives, such as 87 [34]. Rather than succumbing
733
20 Sequential Reactions Initiated by Oxidative Dearomatization. Biomimicry or Artifact?
734 OLi
O Me
OH
[(−)-sparteine]2Cu2O2(PF6)2
Me
3 Å MS, O2, −78°C 51 %, 99 % ee
Me
OH HO Me O
t-Bu Me
O
suprafacial 1,2-shift
O
OH
t-Bu 83
84
85
86
Scheme 20.23 Porco’s oxidative dearomatization, suprafacial shift, and [4+2] dimerization.
to a desired 6π electrocyclization leading to the morphinan architecture, it was proposed that the bis-o-quinone ketal 88 suffers first a reversible [1,7] sigmatropic H-shift leading to intermediate 89, followed by a reversible [1,5] H-shift affording intermediate 90. Tautomer 91 then cyclizes irreversibly to afford compounds 92 and 93 (Scheme 20.24). MeO HO
MeO Pb(OAc)4
Me
O
CH2Cl2
OH MeO
OAc MeO [1,7] H-shift
O
Me
H
O MeO AcO OMe
OMe 87
OAc
O MeO AcO OMe
88
89 [1,5] H-shift
MeO
OH
MeO
OAc
MeO
OAc
MeO HO (a)
MeO
+
O OMe
AcO 93 (47 %)
Scheme 20.24
O
(b)
O H
(b) (a)
MeO
O OAc OMe
92 (21 %)
OH MeO AcO OMe
O MeO AcO OMe
91
90
Feldman’s successive [1,7] and [1,5] H-shifts and electrocyclization.
Canesi has recently shown that the phenoxonium generated from phenol 94 can be intercepted by a tethered olefin leading to further rearrangement of functionality about the carbon chain to give 95 (Scheme 20.25) [35]. The term ‘‘homo-Wagner–Meerwein’’ transposition has been applied to this type of rearrangement, in what is in essence a heretofore unreported 1,3-allylic transposition, likely occurring via a concerted process. The use of IBX to convert phenols into their corresponding o-quinones was first reported by Pettus [36]. It was further shown that o-quinone 97, which is afforded by oxidation of 96, could be tautomerized into the corresponding p-quinone methide 98 (Scheme 20.26) [37]. This intermediate undergoes intramolecular arylation to produce the benzylated brazilin derivative 99, deprotection of which affords brazilin
20.7 Successive Tautomerizations and Rearrangements OH
O PIDA HFIP/CH2Cl2, −17°C 2 min, 50 %
94
O
O
OH
Scheme 20.25 HO
O
Canesi’s dearomatization, allylic transposition, and ketal formation. OH
O O
O
O
OH
IBX
96
OH
tautomerization
HO
O
97 OBn
O OH
ring rupture
O
98
OBn
O
95
OBn
O
OH
HO
O
air
HO
OH
HO
O
O O
O
two 1,4additions
postulated OH brazilide A (102)
OH 101
OR 99: R = Bn 100: R = H, brazilin
Scheme 20.26 Pettus’ oxidative dearomatization followed by tautomerization, cyclization, and presumed oxidative cleavage to brazilide A.
(100). The catechol in this compound undergoes further oxidation upon standing to form p-quinone methide 101, the oxidized form of brazilin that has often been used as a stain. It has been speculated that further oxidative rupture within this p-quinone methide gives rise to brazilide A (102) by a successive series of lactonizations [38]. Trauner has exploited the equilibrium between a p-quinone and its o-quinone methide tautomer in the synthesis of rubioncolin B (106) [39]. Exposure of advanced intermediate 103 to PIDA, in the presence of the desilylating reagent tris(dimethylamino)sulfonium difluorotrimethylsilicate (TASF), is presumed to initially gives rise to a p-quinone that is in equilibrium with o-quinone methide 104 (Scheme 20.27). Density functional theory calculations have shown the latter to be the energetically favored tautomer, facilitating the ensuing [4+2] cycloaddition that affords rubioncolin B derivative 105. In a biomimetic synthesis of icetexane-based diterpenes, Majetich reinvestigated Takeya’s unusual solid state [4+2] dimerization of the o-quinone derived from (+)-demethylsalvicanol (21) (Scheme 20.6) [40]. It is thought that the high negative activation entropy associated with the [4+2] dimerization of o-quinone 22 is overcome by pre-organization of the components in the solid state, thereby facilitating the regioselective cycloaddition. Majetich reported that the use of water
735
Scheme 20.27
MeO2C TBSO
O
103
O
OMe
MeCN/H2O 60 %
TASF; PIDA
O
O
O
MeO
MeO2C 104
MeO2C HO O
Trauner’s oxidative dearomatization, tautomerization, and [4+2] cycloaddition.
OTBS
O
MeO2C
[4+2]
BBr3 95 %
MeO2C HO
H
O O
OR
rubioncolin B (106): R = H
105: R = Me
MeO2C
O
O
736
20 Sequential Reactions Initiated by Oxidative Dearomatization. Biomimicry or Artifact?
20.8 Sequential Ring Rupture and Contraction O HO
O
OH
OH
O
OH
[O]
H
H
22
(+)-demethylsalvicanol (21)
O
OH [4+2] solid state 50 °C, 60 h
O
H
72 % or H2O, 50 °C 51 %
737
O HO
H
70 % 60 °C, Et2O 40 h
(+)-grandione (23) HO
HO
O
OH
OH O H
H 107
(−)-brussonol (41)
Scheme 20.28 Majetich’s oxidation, tautomerization, and 1,6-addition synthesis of brussonol.
as a solvent was found to enable the dimerization, in what is speculated to be due to a high effective-molarity between reacting partners, a consequence of hydrophobic interactions that mimic the solid state. The use of ether as solvent, however, favors a tautomeric shift to enol form 107 – due to the lower dielectric constant of the solvent – which upon nucleophilic ring closure leads to (−)-brussonol (41) as the favored product (Scheme 20.28). In a beautiful example of a biomimetic synthesis, Trauner has shown that hydantoin 108 can be converted into two natural products through a postulated common intermediate [41]. Exposure of 108 to AgO produces exiguamine A (111) or its oxygenated derivative, exiguamine B (116), depending on the number of equivalents of oxidant employed. The proposed mechanism for this transformation begins with the o-quinone oxidation product 109 (Scheme 20.29). Tautomerization would afford quinone methide 110, which is poised for an oxa-6π electrocyclization, affording exiguamine A. o-Quinone 109 can also tautomerize to the isomeric quinone methide 112, which can then undergo an oxa-6π electrocyclization to intermediate 113. The latter is then intercepted by an irreversible oxidation to quinone 114. Further tautomerization to 115 and electrocyclization gives rise to exiguamine B. This proposed mechanism is consistent with the observation that exposure of exiguamine A to excess AgO fails to garner its oxygenated derivative.
20.8 Sequential Ring Rupture and Contraction
The general instability of some phenol oxidation products can lead to further rearrangement through ring rupture followed by either contraction or expansion. For
N H
N H
114
O
N+
N O
N
O
OH
O
OH
O
O
[O]
20 eq.
AgO
H2N N H
H2N
MeOH/ 2 % HCO2H
AgO (10 or 20 eq.)
O HO
O N
O
N
+
N O
N
N H O
N
+
O
N
O
O
109
O
exiguamine B (116)
N H
113
O
O HO
H2N
O O
6p
N+
O
N
N
OH
Trauner’s biomimetic synthesis of the exiguamines.
N+
O
N
N
OH
N
N
O
115
O HO
O HO
O O
N
O
Scheme 20.29
N H
108
H2N
H2N
H2N
O O
H2N
O
O
O
112
excess AgO
N H
O HO
N H
110
O
O
O
O N
O
OH
O
N
+
O
N
AgO 10 eq.
N+
N O
N
exiguamine A (111)
O
O
N H
H2N
N+
O
N
N
H2N
O
O OH
738
20 Sequential Reactions Initiated by Oxidative Dearomatization. Biomimicry or Artifact?
20.9 Sequential Ring Rupture and Expansion O
OH
OH
SIBX THF 95 %
MeO
O
MeO
+
OMe
MeO
O
O
OH OMe
O 1:1
117
118
(+)-wasabidienone B1 (119) CDCl3, 10 days then, SiO2 column
O
+ MeO
O O
O O
O
MeO
MeO O
122 (19 %)
Scheme 20.30
OMe
OMe
O
HO
(-)-wasabidienone B0 (121) (17 %)
120
Quideau’s oxidative dearomatization and successive ring contraction.
example, Quideau observed the instability of wasabidienone B1 (119) in solution and found it slowly underwent ring contraction to wasabidienone B0 (121) [42]. Subjection of phenol 117 to a ‘‘stabilized’’ iodoxybenzoic acid (SIBX) derivative afforded an equimolar mixture of o-hydroxylated products 118 and (+)-wasabidienone B1 (Scheme 20.30). Upon standing in deuteriochloroform for ten days and then purification on silica gel, wasabidienone B0 , and cyclopentadienone 122 were isolated, presumably through the intermediacy of enol 120. Wasabidienone B0 could also be obtained in higher yield by heating a solution of 119 in benzene for five days, since by-product 122 likely arises due to the acidity of the deuteriochloroform.
20.9 Sequential Ring Rupture and Expansion
The coupling of catechins of type 123 and 124 has been known for some time to yield the benzotropolone nucleus common to the theaflavins. However, it was not until Nakatsuka isolated the putative bicyclo[3.2.1] intermediate 126 that the exact nature of the mechanism was elucidated [43]. Although the details of the oxidative coupling of 123 and 124 is a matter of debate, the rearrangement and ring expansion of the adduct 125 would give rise to compound 126 (Scheme 20.31). Hydrolysis affords ene-diol 127 and further oxidation leads to diketone 128, which undergoes a decarboxylative tautomerization affording benzotropolone 129. During the course of studies on the copper-mediated oxidation of catechols, Rogi´c reported the fascinating oxidation of catechol 130 by methoxy(pyridine)copper(II) chloride under strict anaerobic and anhydrous conditions [44]. The immediate oxidation product is an o-quinone of type 131, which undergoes carbon–carbon bond cleavage in an extradiol sense and is reminiscent of the reaction manifold
739
740
20 Sequential Reactions Initiated by Oxidative Dearomatization. Biomimicry or Artifact?
OH O +
OH
O
O
O [O] [4 + 2]
OH O
O
68 %
O
OH
O H
OH 124
123
OH OH2 OH
125
126 H2O 30 min
O
HO
OH
–CO2
OH OH
O
OH OH
air
OH H OH
129
Scheme 20.31
HO
O
O
94 %
O
OH H OH
128
127
Nakatsuka’s dearomatization and successive ring expansion.
many enzyme oxygenases follow (Scheme 20.32). Rogi´c further suggests that the fact that this oxidation takes place in the absence of water may lead one to speculate that it may be no coincidence that the redox active sites in metalloenzymes are typically found in hydrophobic regions of the enzyme. X = OMe, Cl X OH OH
CO2Me O
PyCuClOMe (6 eq.) 64 %
O OMe OMe
O
130
131
Scheme 20.32
132
Rogi´c’s Cu-mediated extradiol catechol cleavage.
In an attempt to mimic the proposed biosynthetic pathway of rosmanol and isorosmanol from carnosol (133), Tejera observed an intriguing ring expansion of the catechol moiety [45]. Treatment of carnosol with meta-chloroperbenzoic acid (m-CPBA) and base led to the isolation of anhydride 135 (Scheme 20.33). The elucidated mechanism involves an initial oxidation to an o-quinone followed by peracid addition to give intermediate 134. A subsequent Baeyer–Villiger rearrangement then affords anhydride 135. OH HO O
O
Ar O
m-CPBA
O
O
O
O OH
O
O
O
H
H
O
carnosol (133) H
Scheme 20.33
O
O
135 134
Tejera’s oxidative ring expansion.
20.11 Natural Products Hypothesized to Conclude Phenol Oxidative Cascades
20.10 Successive Intramolecular and Intermolecular Reactions
Some interesting combinations of intra- and inter-molecular reactions have also been reported. Pettus observed that oxidation of resorcinol derivative 136 with 2 equiv. of PIFA produces the epoxide 138 in a single pot (Scheme 20.34). This compound was further elaborated in two more operations so as to complete the synthesis of epoxysorbicillinol (139) [10]. The cascade sequence is thought to first produce intermediate 137, which displays an extremely electron-deficient double bond. Furthermore, PIFA is thought to be in equilibrium with iodosylbenzene (PhIO) and the corresponding trifluoroacetic anhydride. The PhIO serves as a nucleophilic source of oxygen and subsequently produces the epoxide and iodobenzene in the same pot. Pettus has shown this reactivity of PhIO in various electron-deficient systems and the reagent was subsequently employed by MacMillan for the catalytic epoxidation of unsaturated iminium intermediates [46, 47].
20.11 Natural Products Hypothesized to Conclude Phenol Oxidative Cascades
Besides some of the cascades previously described for the natural products carpanone (Scheme 20.2) – dehydrodiisoeugenol (Scheme 20.3), bisorbicillinol (Scheme 20.4), grandione (Scheme 20.6), futoenone (Scheme 20.9), cedrene (Scheme 20.19), matlaline (Scheme 20.22), brazilide A (Scheme 20.26), rubioncolin B (Scheme 20.27), brussonol (Scheme 20.28), exiguamine B (Scheme 20.29), wasabidienones (Scheme 20.30), theaflavins (Scheme 20.31), carnosol (Scheme 20.33), and epoxysorbicillinol (Scheme 20.34) – there are undoubtedly other natural products that are the conclusion of cascades and sequences initiated by phenol oxidation. For example, in the biosynthesis of sclerocitrin (144), Steglich proposes a catechol dimerization to 140, followed by oxidative ring cleavage to the anhydride 141 and ring contraction to the δ-lactone 142 (Scheme 20.35). Further retro-electrocyclization produces 143, whereupon a vinylogous aldol ring closure is proposed to complete the biosynthesis of sclerocitrin (144) [48]. Zhao has proposed that przewalskin A (147) emerges from catechol 145 by oxidation to the corresponding o-quinone, followed by addition of acetoacetyl-CoA to both carbonyl residues to afford cyclopropane 146 (Scheme 20.36) [49]. This material is speculated to undergo ring expansion to produce the corresponding seven-membered carbocycle, which upon further oxidation affords przewalskin A (147). Hertweck has proposed that oxidative tailoring enzymes are encoded in gene clusters along with specific polyketide synthetases. For example, the hexacyclic aromatic 148 is thought to emerge from 13 acetyl Co-A fragments (Scheme 20.37). Further phenol oxidation is speculated to result in dehydrocollinone (149), which undergoes successive oxidative rupture of the center ring so as to provide the aryl-oxy spiroketal motif found in griseorhodin (150) and related natural products [50].
741
OH
Me 2 equiv
PIFA
137
O Me
O O
O
O
Me PhI=O
O Me
O
Pettus’ oxidative dearomatization and successive epoxidation.
O 136
O
N
Scheme 20.34
Me
O
138
O
O
O
O
Me
2-steps
O
O
OH
Me
epoxysorbicillinol (139) 28 %
O
HO Me
742
20 Sequential Reactions Initiated by Oxidative Dearomatization. Biomimicry or Artifact?
20.11 Natural Products Hypothesized to Conclude Phenol Oxidative Cascades
Pu H
O O
Pu H
[O]
O
XOC
O O O
H H
Pu H
O
743
O O
O O
Pu
140
H H Pu O 141
O H H Pu 142
Pu = 4-hydroxypulvinic acid O
O
H OHO
H O
Pu
OH O
O
HH
Pu
O H Pu 143
O sclerocitrin (144)
Scheme 20.35
Pu
Steglich’s proposed biosynthetic pathway for sclerocitrin.
OH HO HO [O]
O H
CH3COCH2COCoA
O
O−
O HO HO
HO
[O]
O
O
OH
H
H
OH 146
145
Scheme 20.36
O
HO
O
R
O O
HO OH O OH
O
OH O
MeO
OH OH OH
Scheme 20.37
OH
przewalskin A (147)
Zhao’s proposed biosynthetic pathway for przewalskin A.
HO
148
OH
O
O
OH
(dehydro)collinone (149)
O
Me O
OH
HO
HO
O
OH
O Me
O O OHO
griseorhodin A (150)
Hertweck’s proposed biosynthetic pathway for griseorhodin A.
The fredericamycins are thought to emerge from the identical hexacyclic compound 148 (Scheme 20.38) [51]. In this instance, oxidation affords fredericamycin C1 (151), which is intercepted with ammonia to afford fredericamycin B (152). Further oxidation and rupture of the central ring in this instance proceeds through the intermediacy of the diones 153–155, whereupon ring contraction affords 156 and loss of CO2 provides the 5,5-spirocyclic array found in fredericamycin A (157). Although to the best of our knowledge it has never before been postulated in the chemical literature, we suspect that many of the compounds in the frondosin family of natural products are themselves the consequence of various phenol oxidative dearomatization cascades (Scheme 20.39) [52]. For example, we believe that frondosin A (158) likely oxidizes to o-quinone methide 159, which succumbs to a vinylogous aldol reaction to produce frondosin C (160). Alternatively, the
OH
OH
O
HO O
OH
O
H N
fredericamycin C1 (151)
OH O
HO OH O HO
[O]
NH3
−CO2
O
OH
[O]
155
OH O−
156
O
H N
O
O
fredericamycin B (152)
OH O
O
HO CO2H
MeO
HO OH O HO
Shen’s proposed biosynthetic pathway for fredericamycin A.
fredericamycin A (157)
O
O
Scheme 20.38
MeO
148
MeO
O
O
O
O
154
O
O
153
OH2
O
744
20 Sequential Reactions Initiated by Oxidative Dearomatization. Biomimicry or Artifact?
20.11 Natural Products Hypothesized to Conclude Phenol Oxidative Cascades Me
Me
Me H
Me
Me
[O]
Me
H HO
OH
HO
Me
H
H
Me
Me
O
OH O
frondosin A (158)
frondosin C (160)
159 6p [O]
Me
Me
Me
Me Me
H
H
Me O
O OR
H
retro-[4+2]
Me
−CH2O
O O
HO
frondos in D (161): R = H frondos in E (162): R = Me
Scheme 20.39
Me
Me
163
O HO frondos in B (164)
Pettus’ proposed biosynthetic pathway for frondosins.
o-quinone methide intermediate 159 can undergo a 6π electrocyclization and further oxidation in the presence of water or methanol to produce frondosin D (161) and E (162), respectively. A remote 1,6-conjugate addition in 161 to give 163, followed by a retro [4+2] cycloaddition to liberate formaldehyde, would then produce frondosin B (164). Porco theorized that the tetracyclic tetrapetalone skeleton would emerge from a cascade initiated by phenol oxidation of an ansa phenolic system 165 (Scheme 20.40) [53]. It was speculated that a phenoxonium would be intercepted by a neighboring olefin to generate the allyl cation 166. The nitrogen atom was anticipated to intercept its conformer 167 to afford the tetracycle 168, which would undergo further oxidation to afford tetrapetalone C (169). However, this proposal failed in the laboratory; the hydroquinone was simply converted into the corresponding quinone. In addition to the previously described natural products, there are many others that appear to be the result of some sort of phenolic oxidation and subsequent series of reactions. The isariotins (170 and 171) [54], TK-57-164A (172) [54, 55], scyphostatin (not shown) [56], and aranorosin (173) [57] all seem to arise from various sorts of tyrosine oxidations (Figure 20.1). Hugonone A (174) appears to stem from a dimerization of an o-quinol that would be accessible from the corresponding α-methylated phenol [58], whereas malettinin A (175) would seem to arise from oxidation of the properly outfitted naphthol [59]. On the other hand, the heliespirones, such as 176, emerge from oxidation of a properly attenuated hydroquinone followed by a 1,4-conjugate addition reaction [60]. Crews has proposed that curcuphenol (63) (Scheme 20.19) is the progenitor of many of the heliananes, such as 180, through a sequence of reactions that presumably involves phenolic oxidation [61]. We speculate that kushecarpin (177) is the product of phenol
745
20 Sequential Reactions Initiated by Oxidative Dearomatization. Biomimicry or Artifact?
746
Et
Me Et Me OH RO
OH
Me
OH H N
[O]
Me
RO
H N
166
O
Me Me H RO
O
O
[O]
Me
N
O
O
O
O
tetrapetalone C (169)
Scheme 20.40
167
Et OH OH
RO
Me
O
=R
Me Me H OH
OH N
Me
O
Me HO
Et O
OH
OH H Et N
RO
O
O
165
Me Me H +
OH
Me Me H +
OH
168
Porco’s proposed biosynthetic pathway for tetrapetalone C.
O
O O RHN
O
OH OH
O H OH
isariotin E (170)
O
O O
O
Cl O
OH
O
OMe H
OH
H
HN
OH O
NHR NHR
isariotin F (171)
TK-57-164A (172)
aranorosin (173)
O OH
O O HO
O
O
HO CH3
O
OH
HO
OMe Ph
OH O heliespirone C (176)
O O
O
mimosifolenone (178)
helianane (180)
OH
OMe
OMe
O HO
HO
O MeO
hugonone A (174)
malettinin A (175) O
O
OMe
kushecarpin (177)
Figure 20.1
Ph mimosifoliol (179)
Additional natural compounds that may arise from phenolic oxidation.
References
oxidation of some appropriately substituted pterocarpan. Finally, the coincident isolation of mimosifolenone (178) along with mimosifoliol (179) from the root wood of Aeschynomene mimosifolia would lead us to suppose that mimosifolenone is the result of some as yet undetermined oxidative dearomatization cascade that begins with mimosifoliol [62].
20.12 Conclusion
We hope that the foregoing discussions have given the reader pause so as to cautiously consider biomimetic claims involving oxidative phenol dearomatization as well as any future ‘‘biomimetic’’ suppositions on this topic. In our opinion, in the absence of actual isotopic feeding studies, these kinds of speculations are merely fanciful postulations, even if based upon enzymes that are conjectured to exist because of specific genomic sequences. Since time immemorial, man has ascribed any event leading from a simple beginning to a complex end as the result of some grand design. We believe that in the case of many of the foregoing molecules, Nature’s role may be overstated. Indeed, many of these compounds may in fact be the result of specific post transformations outside of Nature’s omniscient tutelage. However, no one can deny that these reaction sequences, irrespective of the origins of their creation, are beautiful, even if they are indeed artifacts invented by man’s mind and brought into being by his hands. In this regard, we have been particularly inspired by the work of Rodr´ıguez and Trauner, among others, in the area of cascades emanating for oxidative dearomatization and anticipate that these researchers will continue to inspire us in the future with their creative synthetic designs.
References 1. Robinson, R. (1917) J. Chem. Soc.,
5. Angle, S.R., Rainier, J.D., and
Trans., 111, 762–768. 2. Leeper, F.J. and Vederas, J.C. (eds) (2000) Biosynthesis: Aromatic Polyketide Isoprenoids, Alkaloids, Topics in Current Chemistry, vol. 209, Springer-Verlag, Berlin, Heidelberg, New York. 3. Magdziak, D., Meek, S.J., and Pettus, T.R.R. (2004) Chem. Rev., 104, 383–1429. 4. Novak, M., Brinster, A.M., Dickhoff, J.N., Erb, J.M., Jones, M.P., Leopold, S.H., Vollman, A.T., Wang, Y.-T., and Glover, S.A. (2007) J. Org. Chem., 72, 9954–9962.
Woytowicz, C. (1997) J. Org. Chem., 62, 5884–5892. 6. (a) Van de Water, R.W. and Pettus, T.R.R. (2002) Tetrahedron, 58, 5367–5405; (b)Ferreira, S.B., da Silva, F.deC., Pinto, A.C., Gonzaga, D.T.G., and Ferreira, V.F. (2009) J. Heterocycl. Chem., 46, 1080–1097. 7. (a) Chapman, O.L., Engel, M.R., Springer, J.P., and Clardy, J.C. (1971) J. Am. Chem. Soc., 93, 6696–6698; (b) Lindsley, C.W., Chan, L.K., Goess, B.C., Joseph, R., and Shair, M.D. (2000) J. Am. Chem. Soc., 122, 422–423.
747
748
20 Sequential Reactions Initiated by Oxidative Dearomatization. Biomimicry or Artifact? 8. Juh´asz, L., K¨ urti, L., and Antus, S. 9.
10.
11.
12.
13.
14. 15.
16.
17.
18.
19.
20. 21.
22.
(2000) J. Nat. Prod., 63, 866–870. Nicolaou, K.C., Vassilikogiannakis, G., Simonsen, K.B., Baran, P.S., Zhong, Y.-L., Vidali, V.P., Pitsinos, E.N., and Couladouros, E.A. (2000) J. Am. Chem. Soc., 122, 3071–3079. Pettus, L.H., Van de Water, R.W., and Pettus, T.R.R. (2001) Org. Lett., 3, 905–908. (a) Mehta, G. and Maity, P. (2007) Tetrahedron Lett., 48, 8865–8868; (b) Ganepain, J., Castet, F., and Quideau, S. (2007) Angew. Chem. Int. Ed., 46, 1533–1535. Magdziak, D., Rodriguez, A.A., Van De Water, R.W., and Pettus, T.R.R. (2002) Org. Lett., 4, 285–288. Aoyagi, Y., Takahashi, Y., Satake, Y., Fukaya, H., Takeya, K., Aiyama, R., Matsuzaki, T., Hashimoto, S., Shiina, T., and Kurihara, T. (2005) Tetrahedron Lett., 46, 7885–7887. Singh, V., Chandra, G., and Mobin, S.M. (2008) Synlett, 3111–3114. Jensen, K.H., Pathak, T.P., Zhang, Y., and Sigman, M.S. (2009) J. Am. Chem. Soc., 131, 17074–17075. (a) Yamamura, S., Shizuri, Y., Shigemori, H., Okuno, Y., and Ohkubo, M. (1991) Tetrahedron, 47, 635–644; (b) Yamamura, S. and Nishiyama, S. (2002) Synlett, 533–543. Mohr, A.L., Lombardo, V.M., Arisco, T.M., and Morrow, G.W. (2009) Synth. Commun., 39, 3845–3855. McFadden, R.M. and Stoltz, B.M. (2006) J. Am. Chem. Soc., 128, 7738–7739. Simmons, E.M., Yen, J.R., and Sarpong, R. (2007) Org. Lett., 9, 2705–2708. Hsu, D.-S. and Liao, C.-C. (2007) Org. Lett., 9, 4563–4565. (a) Njardarson, J.T., McDonald, I.M., Spiegel, D.A., Inoue, M., and Wood, J.L. (2001) Org. Lett., 3, 2435–2438; (b) B´erub´e, A., Drutu, I., and Wood, J.L. (2006) Org. Lett., 8, 5421–5424. (a) Cook, S.P., Gaul, C., and Danishefsky, S.J. (2005) Tetrahedron Lett., 46, 843–847; (b) Polara, A., Cook, S.P., and Danishefsky, S.J. (2008) Tetrahedron Lett., 49, 5906–5908;
23.
24.
25. 26.
27.
28.
29. 30.
31.
32.
33.
34.
(c) Cook, S.P. and Danishefsky, S.J. (2006) Org. Lett., 8, 5693–5695. Morton, J.G.M., Kwon, L.D., Freeman, J.D., and Njardarson, J.T. (2009) Synlett, 23–27. Mendelsohn, B.A., Lee, S., Kim, S., Teyssier, F., Aulakh, V.S., and Ciufolini, M.A. (2009) Org. Lett., 11, 1539–1542. Frie, J.L., Jeffrey, C.S., and Sorensen, E.J. (2009) Org. Lett., 11, 5394–5397. (a) Wenderski, T.A., Huang, S., and Pettus, T.R.R. (2009) J. Org. Chem., 74, 4104–4109; (b) Mejorado, L.H. and Pettus, T.R.R. (2006) J. Am. Chem. Soc., 128, 15625–15631; (c) Mejorado, L.H. and Pettus, T.R.R. (2006) Synlett, 3209–3214; (d) Wang, J. and Pettus, T.R.R. (2004) Tetrahedron Lett., 45, 5895–5899; (e) Wang, J., Pettus, L.H., and Pettus, T.R.R. (2004) Tetrahedron Lett., 45, 1793–1796; (f) Mejorado, L.H., Hoarau, C., and Pettus, T.R.R. (2004) Org. Lett., 6, 1535–1538; (g) Van de Water, R.W., Hoarau, C., and Pettus, T.R.R. (2003) Tetrahedron Lett., 44, 5109–5113. Marsini, M.A., Huang, Y., Van de Water, R.W., and Pettus, T.R.R. (2007) Org. Lett., 9, 3229–3232. Green, J.C. and Pettus, T.R.R. (2010) J. Am. Chem. Soc., asap, doi: 10.1021/ja109925. Skyler, D. and Heathcock, C.H. (2001) Org. Lett., 3, 4323–4324. (a) Imbos, R., Minnaard, A.J., and Feringa, B.L. (2002) J. Am. Chem. Soc., 124, 184–185; (b) Hayashi, Y., Gotoh, H., Tamura, T., Yamaguchi, H., Masui, R., and Shoji, M. (2005) J. Am. Chem. Soc., 127, 16028–16029; (c) Liu, Q. and Rovis, T. (2006) J. Am. Chem. Soc., 128, 2552–2553. Vo, N.T., Pace, R.D.M., O’Hara, F., and Gaunt, M.J. (2008) J. Am. Chem. Soc., 130, 404–405. Acu˜ na, A.U., Amat-Guerri, F., Morcillo, P., Liras, M., and Rodr´ıguez, B. (2009) Org. Lett., 11, 3020–3023. Dong, S., Zhu, J., and Porco, J.A. Jr. (2008) J. Am. Chem. Soc., 130, 2738–2739. Feldman, K.S. (1997) J. Org. Chem., 62, 4983–4990.
References 35. Gu´erard, K.C., Chapelle, C., Giroux,
36.
37. 38.
39.
40. 41.
42.
43.
44. 45.
46. 47. 48.
49.
50.
M.-A., Sabot, C., Beaulieu, M.-A., Achache, N., and Canesi, S. (2009) Org. Lett., 11, 4756–4759. Magdziak, D., Rodriguez, A.A., Van de Water, R.W., and Pettus, T.R.R. (2002) Org. Lett., 4, 285–288. Huang, Y., Zhang, J., and Pettus, T.R.R. (2005) Org. Lett., 7, 5841–5844. Yang, B.O., Ke, C.-Q., He, Z.-S., Yang, Y.-P., and Ye, Y. (2002) Tetrahedron Lett., 43, 1731–1733. Lumb, J.-P., Choong, K.C., and Trauner, D. (2008) J. Am. Chem. Soc., 130, 9230–9231. Majetich, G. and Zou, G. (2008) Org. Lett., 10, 81–83. Volgraf, M., Lumb, J.-P., Brastianos, H.C., Carr, G., Chung, M.K.W., M¨unzel, M., Mauk, A.G., Andersen, R.J., and Trauner, D. (2008) Nat. Chem. Biol., 4, 535–537. Pouys´egu, L., Marguerit, M., Gagnepain, J., Lyvinec, G., Eatherton, A.J., and Quideau, S. (2008) Org. Lett., 10, 5211–5214. (a) Yanase, E., Sawaki, K., and Nakatsuka, S.-I. (2005) Synlett, 2661–2663; (b) Takino, Y., Imagawa, H., Horikawa, H., and Tanaka, A. (1964) Agric. Biol. Chem., 28, 64–71. Demmin, T.R. and Rogi´c, M.M. (1980) J. Org. Chem., 45, 4210–4214. Marrero, J.G., Tejera, L.S.A., Luis, J.G., and Rodr´ıguez, M.L. (2002) Synlett, 1517–1519. McQuaid, K.M. and Pettus, T.R.R. (2004) Synlett, 2403–2405. Lee, S. and MacMillan, D.W.C. (2006) Tetrahedron, 62, 11413–11424. Winner, M., Gim´enez, A., Schmidt, H., Sontag, B., Steffan, B., and Steglich, W. (2004) Angew. Chem. Int. Ed., 43, 1883–1886. Xu, G., Hou, A.-J., Wang, R.-R., Liang, G.-Y., Zheng, Y.-T., Liu, Z.-Y., Li, X.-L., Zhao, Y., Huang, S.-X., Peng, L.-Y., and Zhao, Q.-S. (2006) Org. Lett., 8, 4453–4456. Xu, Z., Schenk, A., and Hertweck, C. (2007) J. Am. Chem. Soc., 129, 6022–6030.
51. Wendt-Pienkowski, E., Huang, Y.,
52. 53.
54.
55.
56.
57.
58.
59.
60.
61. 62.
Zhang, J., Li, B., Jiang, H., Kwon, H., Hutchinson, C.R., and Shen, B. (2005) J. Am. Chem. Soc., 127, 16442–16452. Pettus, T.R.R. and coworkers (2008) NSF Grant Application, CHE-0806357. (a) Wang, X. and Porco, J.A. Jr. (2005) Angew. Chem. Int. Ed., 44, 3067–3071; (b) Wang, X. and Porco, J.A. Jr., corrigenda (2006) Angew. Chem. Int. Ed., 45, 6607. Bunyapaiboonsri, T., Yoiprommarat, S., Intereya, K., Rachtawee, P., Hywel-Jones, N.L., and Isaka, M. (2009) J. Nat. Prod., 72, 756–759. (a) Nakagawa, A., Nishikawa, N., Takahashi, S., and Yamamoto, K. (2004) Patent WO2004074269; (b) Cha, J.Y., Huang, Y., and Pettus, T.R.R. (2009) Angew. Chem. Int. Ed., 48, 9519–9521. Tanaka, M., Nara, F., Suzuki-Konagai, K., Hosoya, T., and Ogita, T. (1997) J. Am. Chem. Soc., 119, 7871–7872. Roy, K., Mukhopadhyay, T., Reddy, G.C., Desikan, K.R., Rupp, R.H., and Ganguli, B.N. (1988) J. Antibiot., 41, 1780–1784. Mdee, L.K., Waibel, R., Nkunya, M.H.H., Jonker, S.A., and Achenbach, H. (1998) Phytochemistry, 49, 1107–1113. (a) Angawi, R.F., Swenson, D.C., Gloer, J.B., and Wicklow, D.T. (2005) J. Nat. Prod., 68, 212–216; (b) Angawi, R.F., Swenson, D.C., Gloer, J.B., and Wicklow, D.T. (2003) Tetrahedron Lett., 44, 7593–7596. Mac´ıas, F.A., Galindo, J.L.G., Varela, R.M., Torres, A., Molinillo, J.M.G., and Fronczek, F.R. (2006) Org. Lett., 8, 4513–4516. Harrison, B. and Crews, P. (1997) J. Org. Chem., 62, 2646–2648. Fullas, F., Kornberg, L.J., Wani, M.C., Wall, M.E., Farnsworth, N.R., Chagwedera, T.E., and Kinghorn, A.D. (1996) J. Nat. Prod., 59, 190–192.
749
751
Part V Frontiers in Biomimetic Chemistry: From Biological to Bio-inspired Processes
Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
753
21 The Diels–Alderase Never Ending Story Atsushi Minami and Hideaki Oikawa
21.1 Introduction
The Diels–Alder reaction is one of the most widely used reactions in organic synthesis, forming a six-membered ring from a 1,3-diene and a dienophile with high regio- and stereoselectivity under relatively mild conditions [1]. In addition, by creating four chiral centers or quaternary stereogenic centers in organic synthesis, the Diels–Alder reaction is a powerful tool that has been applied to the synthesis of complex pharmaceutical and biologically active compounds [2]. Natural products presumably biosynthesized via a [4+2] cycloaddition are frequently encountered in the literature [3]. Several reviews on natural Diels–Alder type cycloadducts have covered more than 300 cycloadducts, including polyketides, terpenoids, phenylpropanoids, alkaloids, and natural products, formed via mixed biosynthetic pathways. Several observations indicate that natural products may be biosynthesized by biological Diels–Alder reactions: 1) 2) 3) 4)
co-isolation of an adduct and of the corresponding precursor; co-occurrence of adducts and their regio- and diastereoisomers; dimeric structure of a single component; unusual combination of biosynthetic components such as monoterpene and triterpene.
Careful examination of the structures of natural [4+2] adducts sometimes provides useful information on their biosynthesis. However, retrosynthetic analysis of plausible adducts incorporating a cyclohexene moiety or its equivalent sometimes misled their biosynthetic pathway. For example, the flavonolignan silydianin [4] is a plausible [4+2] adduct that could be biosynthesized from coniferyl alcohol and an ortho-quinone derived from the flavanone taxifolin (Scheme 21.1, [4+2] route). Usually, radical coupling of phenylpropanoids affords various lignan products (radical coupling route) [5]. Along this line, the skeleton of silydianin is therefore most likely constructed by radical coupling followed by nucleophilic addition of the resultant enolate (path A). Co-occurrence of the lignan silychristin in the same plant supported this biosynthetic pathway (path B). In previous reviews, Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
21 The Diels–Alderase Never Ending Story
754
OH MeO
OH
HO oxidation
HO
+
OH
O
radical coupling
O R
O R
silychristin
Scheme 21.1
O
MeO HO
OH
diene
[4+2 ]
HO
OH
OH O
O O
O
HO
HO
path A
OMe
O
OH R dienophile
OH
A
path B
HO
preceding reaction
MeO
OH O
OH
OH
oxidase
B
MeO H+
OH
OH OH O taxifolin
coniferyl alcohol
MeO
O
O
OH O
OH OH O silydianin
Biosynthetic pathway of silydianin and its related metabolite.
we can find several representative examples of plausible [4+2] adducts that could be biosynthesized via alternative routes [3]. Because it is difficult to distinguish between the involvement of a Diels–Alderase or of alternative enzymes, we have collected candidates with intriguing natural [4+2] adducts and will discuss their enzymatic formation in this chapter.
21.2 Diels–Alderases Found in Nature
We reported the enzymatic activity of solanapyrone synthase in 1995 as the first Diels–Alderase [6]. To date, two additional Diels–Alderases, lovastatin nonaketide synthase [7] and macrophomate synthase (MPS) [8], have been purified and characterized. The first two catalyze intramolecular Diels–Alder (IMDA) reactions while the last one catalyses an intermolecular Diels–Alder reaction. We have reported the detailed reaction pathway [9] of MPS and its catalytic mechanism based on the crystal structure [9]. However, only three Diels–Alderases have been characterized up to now. The reason why the number of characterized Diels–Alderase is limited may be attributed to the difficulty in identifying the actual substrate and the enzyme responsible for the biosynthesis of the target molecule. In this section, we describe these three natural Diels–Alderases and discuss the mechanism of their catalysis.
21.2 Diels–Alderases Found in Nature
755
21.2.1 Lovastatin Nonaketide Synthase
The biosynthesis of the cholesterol-lowering drug lovastatin isolated from Aspergillus terreus has been investigated extensively by Vederas and coworkers [10]. Incorporation experiments with multiple labeled acetate and 18 O-oxygen suggested that it is biosynthesized via the polyketide pathway. Based on feeding studies and co-occurrence of 4a,5-dihydromonacolin L (3), this compound was speculated as an intermediate. This was confirmed by the successful conversion of 3 into lovastatin, using a blocked mutant of A. terreus (Scheme 21.2) [11]. Because lovastatin does not have an electron-withdrawing group in the dienophile moiety, it was proposed that the requisite Diels–Alder reaction occurred at the hexaketide stage. However, all efforts to convert 13 C-labeled hexaketide precursor 1b into lovastatin were unsuccessful due to non-enzymatic cycloadditions affording a 1 : 1 mixture of undesired diastereomers 4b (endo) and 4c (exo) in aqueous media (half life of 1b: two days) (Scheme 21.2). This clearly showed significant rate acceleration for the cycloaddition of 1b in aqueous media [12]. HO preceding
9 acetate reaction + C1-unit from LovB, LovC methionine polyketide synthase
SR O
[4+2 ]
hexaketide precursor 1a R = Enz 1b R = NAC SNAC
H 4a (endo)
H 2
O
5
SNAC
H 4b (endo)
O H
4a
O H
O O
O
H 4a,5-dihydromonacolin L (3) (DHM-L) O H
HO
O
SEnz
H
O H
Scheme 21.2
O H
lovastatin
SNAC
H 4c (exo)
Biosynthetic pathway of lovastatin.
In 1999, the biosynthetic gene cluster of lovastatin was cloned by Hutchinson’s group [13]. They succeeded in achieving heterologous expression of polyketide synthase (PKS) genes (lovB and lovC) in Aspergillus nidulans to produce 3. Collaborating with the Hutchinson group, the Vederas group started enzymatic studies using lovastatin nonaketide synthase (LovB), which is responsible for the construction of the lovastatin backbone. Using the purified LovB, hexaketide triene precursor 1b was incubated without co-factors and substrates to give the three adducts 4a–c in a ratio of 1 : 15 : 15. Minor endo-product 4a was confirmed to be the one with the same stereochemistry as natural 3 [7]. In lovastatin biosynthesis, hexaketide precursor 1b should load on the corresponding ketosynthase domain of LovB; it is then processed downstream to yield 3. Since adducts 4a–c were obtained as N-acetyl-cysteamine (NAC) thioesters, the obligatory thioester exchange did not occur in the Diels–Alder reaction.
756
21 The Diels–Alderase Never Ending Story
Recently, LovB and LovC were successfully overexpressed in yeast cell and their catalytic functions were fully reconstituted in the presence and absence of co-factors to produce 3 and various aberrant products [14]. These results clearly demonstrated that PKS LovB is responsible for the corresponding Diels–Alder reaction. 21.2.2 Macrophomate Synthase
The phytopathogenic fungus Macrophoma commelinae has the ability to transform 2-pyrone 5 into the corresponding benzoate analog macrophomic acid (Figure 21.1a) [15]. Later, we succeeded in the purification of MPS, which is responsible for this complex aromatic conversion with oxaloacetate as a substrate for the C3-unit precursor [16]. Throughout all studies on MPS, the most difficult task was the identification of the C3-unit precursor. As in the case of lovastatin, establishing substrate and enzymatic activity are always tough work in a Diels–Alderase study. The catalytic mechanism of the MPS reaction was investigated extensively, showing that the reaction involves three separate steps: (i) decarboxylation; (ii) two carbon–carbon bond formations; and (iii) decarboxylation with concomitant dehydration [8, 17]. Recently, it was found that the third step of the MPS reaction is decarboxylation to afford intermediate 10 and that the subsequent dehydration is not catalyzed by MPS [18]. Unstable intermediate 10 was released from active site and dehydrated to give macrophomate in a non-enzymatic manner. In the absence of 5, MPS simply acts as a decarboxylase with high catalytic efficiency (Figure 21.1a). The crystal structure of the MPS complexed with pyruvate and Mg2+ was determined with a resolution of 1.70 A˚ (Figure 21.1b) [9]. On the basis of this structural information, the pathway of the MPS reaction can be outlined as follows (Figure 21.1a): oxaloacetate is incorporated into the active site of MPS in a similar way to that of pyruvate. Lewis acidity of the magnesium promotes decarboxylation to form the enolate anion, which is stabilized by an electron sink provided by the divalent cation [8, 9]. As shown in Figure 21.1b, the 2-pyrone molecule is fixed in place through two hydrogen bonds between the carbonyl oxygen of 2-pyrone, Arg101, the C5-acyl oxygen, and Tyr169. These hydrogen bonds act not only in substrate recognition but also enhance reactivity in the inverse electron demand Diels–Alder reaction by reducing the lowest unoccupied molecular orbital (LUMO) energy of the diene. The binding model explains the substrate specificity and stereochemical course of whole reaction pathway [16]. Based on formation of the aberrant adduct with pyrone 8 and the observation that dehydration proceeds formally in an anti-sense [8, 16], it was proposed that the higher energy [4+2] adducts 9 and 7 are transformed into either macrophomic acid or the rearranged product 8 as shown in Figure 21.1a. Extensive point mutation experiments on MPS identified essential amino acid residues [19] and remarkable tolerance for mutation. Mixed quantum and molecular mechanics calculations on the MPS-catalyzed reaction suggested that the transition state of an alternative Michael–aldol route is more stable than that of the concerted
NH2
H N
O
9
OCH3 O
O
O
HO
HO2C
−CO2
O
OH 10
CH3O
O O
R1
O
O − H2O
CH3O
8
O
macrophomic acid
O
O
OH
CO2−
Tyr169 O R2
HO2C
H3CO2C
6 R1 = H, R2 = OCH3
OCH3
O
1
O
–
5 R = OCH3, R2 = CH3
NH2
NH Arg101
7
step 3
O
O
Mg2+ –
− CO2
step 1
(b)
Trp52
Trp68 Phe275
2-pyrone
pyruvate
Arg101
Trp278
Pro151
3 6
Phe149 Tyr169
Asp211 Mg(II)
Glu185
Detailed reaction mechanism of macrophomate synthase (a) and model of MPS active site (b).
O
HO
N
O
NH2
2+ Mg –O
N
O
–
Arg101
NH2
NH2
NH
Figure 21.1
(a)
Arg101
NH
[4+2 ]
step 2
Arg
101
O
’
Mg2+ –O
’
H2 N
’
O
’
– Mg2+ O
21.2 Diels–Alderases Found in Nature 757
758
21 The Diels–Alderase Never Ending Story
Stepwise mechanism O
CH3O
O
O
O HO2C
CH3O
CO2H
O
− CO2
O
1) Michael reaction
2) aldol reaction
O
9
macrophomic acid
HO2C O O CO2H
HO2C
OHC
−CO2
HO2C
aldol reaction MPS
O
OH (44% ee)
Scheme 21.3 Alternative reaction mechanism (Michael–aldol route) and the aldol reaction catalyzed by macrophomate synthase.
Diels–Alder route [20], indicating that the two-step route is the energetically preferred process. Although intriguing observations that MPS itself can catalyze the aldol reaction have been reported (Scheme 21.3) [21], experimental evidence must be provided to evaluate the validity of calculation results. The recent finding that the first decarboxylation, the second C–C bond formation, and the third decarboxylation into 10 are rapid processes [18] suggested that it is difficult to distinguish a concerted from a stepwise process experimentally. 21.2.3 Solanapyrone Synthase
Solanapyrones were isolated as phytotoxic substances from various fungi [22]. This family consists of diastereomers A (15), D (16), and their reduced forms B (17) and E (18) [23]. A series of feeding experiments with simple isotopically labeled precursors established that solanapyrones are biosynthesized via a polyketide pathway and that the diastereomers 15 and 16 are produced at the later stage [23, 24]. Isolation of these substances as optically active forms strongly indicates that 15 and 16 are biosynthesized from the achiral linear triene precursor prosolanapyrone III 14 via an enzyme-catalyzed Diels–Alder reaction. Although there are many candidates such as lovastatin hexaketide for the intriguing cycloaddition, incorporation of isotopically labeled biosynthetic precursors, prosolanapyrones I (12) and II (13) into 15 and 16 confirmed the biosynthetic pathway of solanapyrones as shown in Scheme 21.4 [25]. To support the involvement of a Diels–Alderase in this reaction, enzymatic conversion of 14 was examined. Using crude enzyme, we found that both 13 and 14 were converted into 15 and 16 [6]. Although purification of solanapyrone synthase (SPS) was hampered by its instability, it has been purified as a single
21.2 Diels–Alderases Found in Nature prosolanapyrone synthase
8 acetate (PSS) + 2 C1-unit from methionine
OCH3
O
OCH3 HOH2C
H3C
R O O hexaketide precursor prosolanapyroneI (12) 11 (PSP-I)
O
O O
R
O
prosolanapyrone II (13) (PSP-II) OHC
OCH3
O
O
[4+2 ]
OCH3
OCH3
O
O
R
prosolanapyrone III (14) (PSP-III)
OCH3
OHC
[4+2 ]
O O
solanapyrone B (17) (SP-B)
HOH2C
O O
OCH3
OCH3
OHC
O
endo
Scheme 21.4
H
solanapyrone A (15) (SP-A)
exo
OHC
HOH2C
H
H3C
solanapyrone synthase (SPS)
R=
R
EnzS
OHC
preceding reaction
759
O H3C
H
H solanapyrone D (16) (SP-D)
OCH3
O O
solanapyrone E (18) (SP-E)
Biosynthetic scheme of solanapyrones.
band on SDS-PAGE (sodium dodecyl sulfate polyacrylamide gel electrophoresis) [26]. Recently, the gene cluster responsible for biosynthesis of solanapyrones has been identified by homology based PCR (polymerase chain reaction) and genome walking [27]. The gene cluster contains PKS gene (sol1) for backbone construction and four genes for modification enzymes, including FAD-dependent monooxygenase gene (sol5). This monooxygenase Sol5 possessing a flavin-binding domain was overexpressed in yeast and purified as a single band on SDS-PAGE. The enzyme thus obtained showed enzymatic activity catalyzing both oxidation (13 to 14) and Diels–Alder reaction (14 to 15 and 16) with high optical activity and exo/endo ratio (7 : 1). The character of this enzyme was identical to that of crude solanapyrone synthase from A. solani. To assess the diastereoselectivity and the intrinsic reactivity of prosolanapyrones, Diels–Alder reactions were examined under various conditions [28]. In less polar solvents, heating was required for the effective cycloaddition of 12 to 14. An increase in the oxidation level of the 3-substituent in the prosolanapyrones enhances rate acceleration. This can be rationalized in terms of the LUMO energy of the dienophile moiety in the pyrone precursors. Endo/exo-selectivities with 12–14 were essentially the same in various organic solvents while the endo-selectivity was increased with increasing solvent polarity. The slight preference for endo-selectivity in less polar solvents suggested that there is little steric congestion in both endo- and exo-transition states (Scheme 21.4) as reported in the reactions of simple decatriene systems [29].
760
21 The Diels–Alderase Never Ending Story
In aqueous medium, the non-enzymatic reactions of 12–14 were accelerated and gave endo-adducts with high selectivity (15 : 16 = 3 : 97) [28]. These effects were observed significantly in the reaction of 14 but not in that of 13, indicating that the oxidation of 13 significantly enhanced the reactivity for the Diels–Alder reaction. Similar rate accelerations and the predominant formation of endo-adducts have been reported as a hydrophobic effect [30]: water forces the substrate to form the more compact endo-transition state, reducing its molecular surface exposed to the aqueous medium. In addition, molecular orbital calculations indicated the importance of hydrogen bonding between the water and dienophile carbonyl group to reduce the LUMO energy of the dienophile [31]. Owing to the effects described above, the background reaction could not be ignored in the enzymatic reaction under standard conditions. Contrary to the non-enzymatic reaction, the enzymatic conversion of 13 provided preferentially exo-adduct 15. These observations indicate that the major role of solanapyrone synthase is the oxidation of 13 to the more reactive 14 and the stabilization of the exo-transition state. Considering the significant rate acceleration for the cycloaddition of aldehyde 14 in aqueous medium [28], hydrogen bonding between the carbonyl group and some amino acid residues in the active site might contribute to catalyzing the cycloaddition as observed in the case of Diels–Alder catalytic antibodies [32]. SPS is an example of a Diels–Alderase that has a bifunctional role. This finding is important if we are to discuss the nature of the Diels–Alderases described in this chapter.
21.3 Intramolecular Diels–Alder Reactions Possibly Catalyzed by Dehydratase or DH-Red-Domain of PKS or Hybrid PKS-NRPS
Type-I PKS can catalyze the carbon-chain elongation process using several extender units such as malonyl-CoA and methylmalonyl-CoA to afford a β-ketoacyl intermediate [33]. The resultant β-ketoacyl product is subjected to additional modifications toward β-hydroxy, enoyl, or saturated compounds prior to the next elongation step. Because these modification reactions stop in the arbitrary stage, both (E,E)conjugated diene and dienophile such as α,β-unsaturated carbonyl system can easily be biosynthesized from this elongation process. This indicates that natural products produced by type-I PKS and PKS-non-ribosomal peptide synthethases (NRPS) hybrid are better candidates as potential substrates for the IMDA reaction than other types of natural products (Scheme 21.5). Actually, polyketides can display many [4+2] cycloadducts such as decalin, spirotetronate, as-indacene, indane, and isoindolone skeletons [3]. Among these, the most common carbocycle is decalin in the metabolites produced by fungi and actinomycetes, while the feasibility of such biosynthetic cycloadditions has been supported by several total syntheses of decalin polyketides. Actual involvement of lovastatin PKS as Diels–Alderases has been proven, as discussed in Section 21.2.1 [7]. Recently identified biosynthetic gene clusters revealed plausible candidates to catalyze those Diels–Alder reactions. As discussed in the previous section, we
21.3 Intramolecular Diels–Alder Reactions
O
preceding reaction
OH
O
O
PKS (DH domain) or dehydratase
761
[4+2 ]
diene
dienophile
Scheme 21.5 Typical biosynthetic pathway of [4+2] adducts via type-I polyketide with PKS (dehydratase domain) or dehydratase.
assume that these putative Diels–Alderases might be bifunctional enzymes, catalyzing not only the formation of reactive species but also their [4+2] cycloaddition the same active site. In this section, we discuss recent advances on the biosynthetic machinery of polyketide natural products incorporating [4+2] cycloadducts. 21.3.1 Equisetin and Chaetoglobosin (Compactin, Lovastatin, Solanapyrone)
Equisetin, a fungal metabolite, is also a typical polyketide [4+2]-adduct with a decalin scaffold. An incorporation study with isotope labeled precursors established that the molecular skeleton of equisetin is constructed by an octaketide and l-serine (Scheme 21.6) [34]. Recently, the biosynthetic gene cluster of equisetin has been identified, which is similar to the highly reducing type-I modular PKS of various fungal polyketides [35]. An intriguing cycloaddition with PKS EqiS is proposed to occur at the heptaketide intermediate stage. At the final stage of equisetin biosynthesis, the reduction (Red) domain of EqiS catalyzes an unusual intramolecular Claisen condensation to give a tetramate moiety. The cytochalasins constitute one of the largest family of intramolecular biological Diels–Alder adducts, which possess a characteristic isoindolone moiety fused to a 11-, 13-, and 14-membered macrocycle. Chaetoglobosin A, a specific inhibitor of myosin microtubule formation, is a representative example of a [4+2]-macrocyclic polyketide with an isoindolone moiety [36]. Based on feeding experiments of labeled precursors and accumulation of intermediates with treatment of cytochrome
8 acetate + L-serine
EnzS
preceding reaction
SEnz
EqiS PKS-NRPS
diene A
O dienophile X
O H
[4+2 ] EqiS
H
EqiS
EqiS
O O
diene A
O dienophile X
NMe OH
preceding reaction
O
HO NMe OH
reduction domain
MeN EqiS
EqiS
SEnz
O
OH
O
[4+2 ]
O
OH H
H equisetin
Scheme 21.6
Proposed biosynthetic pathway of equisetin.
762
21 The Diels–Alderase Never Ending Story
9 acetate + 3 C1-unit from methionine + L-tryptophan
H
preceding reaction
O
CheA? condensation dehyration
HN
CheA, CheB
OO
N H
HN OO
O CheA?
H
[4+2 ]
H HN HN OO N H prochaetoglobosin I
OO
N H
O
OH
chaetoglobosin A
Scheme 21.7
Proposed biosynthetic pathway of chaetoglobosin.
P-450 inhibitor [37], it was proposed that the biosynthetic pathway goes via the plausible [4+2] adduct prochaetoglobosin I (Scheme 21.7). Recently, the biosynthetic gene cluster of chaetoglobosin has been identified [38]. It consists of the PKS-NRPS hybrid gene (cheA), stand-alone enoyl reductase gene (cheB), and a series of oxidation enzyme genes (cheD, cheG, and cheE), which correlate with the transformation proposed in the biosynthetic pathway. This suggested that the backbone of chaetoglobosin is constructed by CheA in cooperation with CheB, probably through hypothetical intramolecular endo-selective cycloaddition, which was supported by the retro-Diels–Alder reaction of prochaetoglobosin I [19]. Biomimetic synthesis of cytochalasin D [39] using a similar IMDA via a putative deoxytetramate intermediate provides further support for this hypothesis (Scheme 21.8).
toluene, 80°C
O
BzN O
53%
OH
H
H O
BzN O
HN O OAc
OH O
cytochalasin D
Scheme 21.8
Biomimetic construction of cytochalasan skeleton.
Accumulation of genetic and biochemical data has now allowed us to discuss Diels–Alderase in the biosynthesis of the fungal polyketides lovastatin [13], compactin (a demethylated analog of lovastatin) [40], solanapyrone [27], equisetin [9], and chaetoglobosin [38]. Figure 21.2 shows the domain organization of the PKSs. Heterologous expression of PKSs, LovB, and the solanapyrone PKS (PSS) led to the production of dihydromonacolin L [13] and desmethylprosolanapyrone I [27], respectively. These observations clearly indicated that the cycloaddition occurred during chain extension in LovB (Scheme 21.2). However, PSS did not catalyze the
21.3 Intramolecular Diels–Alder Reactions LovB
KS
MAT
DH
MT
ER0
KR
ACP
CON
LovC
ER
MlcA
KS
MAT
DH
MT
ER0
KR
ACP
CON
MlcG
ER
PSS
KS
MAT
DH
MT
ER
KR
ACP
CON
EqiS
KS
MAT
DH
MT
?
KR
ACP
CON
A
T
R
Eqi9
ER
CheA
KS
MAT
DH
MT
?
KR
ACP
CON
A
T
R
CheB
ER
Figure 21.2
Domain organization of several [4+2] adducts.
Diels–Alder reaction (Scheme 21.4), although its domain architecture is almost identical to the corresponding PKSs, LovB, and compactin PKS MlcA, and common hexaketide precursors are involved in their chain-elongation steps. The findings on PSS raised a question on the timing of the enzymatic IMDA reaction. In the biosynthesis of lovastatin and compactin, the carbonyl group of the hexaketide intermediate is activating the dienophile and is then lost by further chain extension steps after cycloaddition. Thus, cycloaddition timed at just after introduction of the dienophile moiety is considered to be required in these PKS reactions. Domain architectures of both PKS Che and Eqi are nearly identical and showed significant similarity to those of decalin PKS shown above. Thus, it is reasonable to speculate that decalin PKS and cytochalasin PKS are closely related to each other. The structures of other [4+2]-decalin and cytochalasan polyketides indicate that these adducts usually retain carbonyl groups adjacent to the dienophile moieties. In these compounds, [4+2] cycloaddition could proceed either during the extension steps or after the extension steps. Interesting examples are found in the biosynthesis of the octaketides equisetin [35] and aspochalasin Z [41], the cycloaddition precursors of which are closely related (Scheme 21.9). While the timing of cycloaddition in equisetin may be optional, the cycloaddition of aspochalasin Z obviously must proceed in a step following chain extension. Since these metabolites have carbonyl groups adjacent to putative dienophiles, it is likely that the intriguing cycloaddition occurs after the extension steps. This proposal can explain the formation of both aspochalasin-type and equisetin-type cycloadducts, depending on the dienophile being involved (A or B in Scheme 21.9) and selected by the corresponding Diels–Alderase. No experimental evidence is currently available on the timing and players involved in the cycloaddition in these biosyntheses. 21.3.2 Kijanimicin, Chlorothricin, and Tetrocarcin A
The spirotetronate antibiotics such as kijanimicin [42], chlorothricin [43], and tetrocarcin A [44] possess not only decalin ring scaffold but also the spirotetronate moiety, which could be assembled by IMDA. The construction of decalins is proposed to proceed in a similar way to those of lovastatin biosynthesis. In this section, we thus focus on the unique spirotetronate scaffold that is proposed to be biosynthesized from the IMDA between a diene moiety and a γ -methylene-acyltetronate
763
764
21 The Diels–Alderase Never Ending Story
Diels-Alderase
H
(dienophile B ) (diene B )
HN OO
diene A
diene B
O
A
O
Diels-Alderase
NH
aspochalasin Z
B
H N
O O H
H
(dienophile A) (diene A)
equisetin-type adduct
Scheme 21.9
Proposed biosynthetic pathway of aspochalasin Z.
dienophile on the putative intermediate 21 (Scheme 21.10). An efficient synthesis of this spirotetronate scaffold found in the chlorothricin-type skeleton, using IMDA, supports the proposed biosynthetic pathway. In tetronate moiety formation (19 to 20), a three-carbon unit derived from glycerol is necessary to attach the linear polyketide chain as shown in the Scheme 21.10. Comparative gene analysis between kijanimicin (kij) [45], chlorothricin (chl) [46], and tetrocarcin A (tca) [47] biosynthetic gene clusters indicates that the tetronate moiety might be biosynthesized by specifically conserved enzymes. A recent study on the tetronate biosynthetic pathway [48] allowed speculation that the tetronate formation is catalyzed by the FkbH homolog KijC, the stand-alone acyl carrier protein (ACP) KijD, the acyltransferase KijE, and the ketoacyl ACP synthase III homolog KijB. Currently, it is proposed that the dehydrative dienophile formation (20 to 21) would possibly proceed with a flavin-dependent monooxygenase/oxidoreductase homolog KijA, followed by the [4+2] cycloaddition (21 to 22) occurring spontaneously to form the spirotetronate moiety. 21.3.3 Indanomycin
Indanomycin, produced by Streptomyces antibioticus NRRL 8167, is a hybrid compound of polyketide and nonribosomal peptide with an indane scaffold (Scheme 21.11) [49]. Feeding experiments with 13 C-labeled precursors such as acetate, propionate, and butyrate have supported the involvement of a [4+2] cycloaddition for the indane ring formation [50]. Because the diene part is installed by normal PKS functions, the formation of a reactive dienophile is a key step for cycloaddition. Bioinformatic data on the identified biosynthetic gene cluster suggested that PKS lacks a key dehydratase (DH) domain to construct the dienophile moiety [51]. Therefore, two routes can be considered (Scheme 21.11):
21.3 Intramolecular Diels–Alder Reactions
765
R O
O ACP + HO S
O
19
S KijD
S KijC
HO
KijD ?
OH
OH
H
[4+2 ]
O
O (KijA?) dehydratase
O
O
O
20
21
O
OH
22
OH
O
O
O H O
H O
O
HO O
H H
OH O
HO
HO
O OH O
HO O O
HO H
NH
MeO
O O
HO O
NO2
OH
O O
O O
H O O O
H H
H O
HO
O
O
O
HO
HO
O O
OH O
O
OMe
Cl
O
HO O
NO2
O H
H
kijanimicin
O P O OH OH
R
preceding reaction
KijB ?
NH
O
R
R OH O
MeO
O P HO O OH KijC ?
HO
O
H O O O
O
O
chlorothricin
tetrocarcin A
Scheme 21.10 Proposed biosynthetic pathway of kijanimicin, chlorothricin, and tetrocarcin A.
NH
NH O
HN
O preceding OH reaction
O H
[4+2 ]
19 O HO
O dehydratase (IdmH ?)
O
Scheme 21.11
HO
19
O
H indanomycin
Proposed biosynthetic pathway of indanomycin.
21 The Diels–Alderase Never Ending Story
766
1) Dehydration at C19 occurs during the polyketide chain elongation, catalyzed by an adjacent DH domain such as module 3, and the cycloaddition takes place to give the adduct. 2) Alternatively, after completion of chain assembly, a putative DH (IdmH) would give the dehydration product followed by the cycloaddition at the same active site to give indanomycin, as proposed in the kijanimicin biosynthesis. Since the former route requires an unusual participation of neighboring DH domain on the downstream module, the latter route is more likely. Again, this fits our proposal that enzymes producing a reactive intermediate catalyze the Diels–Alder reaction. Currently, the biosynthetic gene clusters for other indane-containing antibiotics such as stawamycin, plakotenin, amaminol, cafamycin, and cochleamycin have not been identified yet [50]. Thus, additional biochemical or genetic characterization of IdmH is required for detailed analysis of indane ring formation. 21.3.4 Spinosyn
A different type of Diels–Alder adduct exists in the core skeleton of spinosyn. Spinosyn possesses a 22-membered macrolide with unusual as-indacene scaffold, d-forosamine, and methylated l-rhamnose [52]. Bioinformatic data on the biosynthetic gene cluster suggested the biosynthetic pathway of spinosyn (Scheme 21.12) [53]. Successful bioconversion of aglycone using the strains disrupting four genes (spnF, J, L, M) firmly established biosynthetic transformations at the later stage, suggesting that these genes are involved in as-indacene formation. The detailed function of SpnJ, a flavin-dependent dehydrogenase, was identified by in vitro analysis to catalyze the oxidation of the hydroxy group at C15 in macrocyclic intermediate 23 [54]. Because the possible linear polyketide precursor was not accepted by SpnJ, this oxidation (23 to 24) occurs after the macrolactone ring formation. The remaining transformations into spinosyn may require OH
HO O
OH 15
OH OH
oxidation SpnJ
O
C-C bond formation
OH
OH
OH
O 23
Me2N
OH
O O
24
dehydratase
preceding reaction
O O
MeO O H H
O O O
H H
O
HH
OMe OMe H
OH H
H
spinosyn
Scheme 21.12
Proposed biosynthetic pathway of spinosyn.
[4+2 ]
OH
21.4 Diels–Alder Reactions after Formation of Reactive Substrates by Oxidation Enzymes
C–C bond formation, then dehydrative isomerization forming reactive diene, followed by the cycloaddition. The exact sequence for as-indacene formation must await further experimental characterizations of the remaining three genes (spnF, L, M).
21.4 Diels–Alder Reactions after Formation of Reactive Substrates by Oxidation Enzymes
Considering that an oxidase can convert an unactivated precursor into a reactive Diels–Alder substrate in the case of the Diels–Alderase solanapyrone synthase (Section 21.2.3), we speculated that a similar mechanism can operate in the biosynthesis of other Diels–Alder adducts. Among highly diverse structures of natural [4+2] adducts, there are significant numbers of adducts that are derived by the oxidative transformation preceding the Diels–Alder reaction [3]. For example, oxidation of catechol provides highly reactive ortho-quinones that have two dienes and two dienophiles in the same molecule (Scheme 21.13). Either non-enzymatic or enzymatic cycloaddition proceeds in various modes to give structurally unique adducts as either racemates or optically active forms. Similarly, chiral ortho-quinols derived from phenols are transformed into structurally complex molecules. Besides phenol oxidation, oxidative transformation of oligoprenyl side chain and terpene gives reactive conjugated dienes that react with suitable dienophiles to furnish [4+2] adducts. In this section, we introduce several putative examples of Diels–Alder reactions found in Nature, along with biomimetic syntheses and biological conversions of putative Diels–Alder precursors. O HO
OH
preceding reaction
O
O
s-cis-diene
O [4+2 ]
O
s-cis-diene
oxidase
O
dienophile dienophile reactive [4+2 ] substrate O
O O
O O
HO
preceding reaction
O
OH [4+2 ]
s-cis-diene
oxidase
dienophile
see main text
dienophile reactive substrate
Scheme 21.13
Model of ortho-quinone and ortho-quinol formation as a Diels–Alderase.
767
768
21 The Diels–Alderase Never Ending Story
21.4.1 Oxidation of Phenol and Catechol to Reactive Dienone and Orthoquinone
Quinones, probably derived from catechol or hydroquinone, are common precursors of natural [4+2] adducts. Bazan et al. proposed the involvement of a Diels–Alder reaction in the biosynthesis of phenylphenalenone lachanthocarpone based on the conversion of 25 into lachanthocarpone, with NaIO4 , via the ortho-quinone 27 [55] (Scheme 21.14). To examine the intermediacy of the diarylheptanoid in the biosynthesis of anigorufone, H¨olscher and Schneider fed the 13 C-labeled 26 to the cultured root of Anigozanthos preissii [56]. The anigorufone isolated showed significant incorporation, establishing involvement of oxidation of 26 to 27, followed by a Diels–Alder reaction. In this case, an oxidase would provide the reactive precursor 27 for a cycloaddition. Then, the cycloaddition proceeded after releasing it from the active site of the oxidase to give achiral products. Involvement of a Diels–Alderase is not essential in this case. OH HO
OH HO
A. preissii
NaIO4
25 X = O 26 X = H, H
CO2H CO2H
X
O
preceding reaction
C1-unit
OH
O O [4+2 ]
oxidase
27
Scheme 21.14
lachanthocarpone R R = OH anigorufone R=H
Biosynthetic pathway of lachanthocarpone and anigorufone.
Grandione from Torreya grandis is a dimer formed by an endo-selective heteroDiels–Alder reaction between two ortho-quinones derived from the diterpene monomer demethylsalvicanol. Biomimetic oxidation of demethylsalvicanol in the solid state at room temperature afforded a single [4+2] adduct that was identical to grandione (Scheme 21.15) [57]. In grandione biosynthesis, enzymatic oxidation would convert the catechol moiety of the monomer
HO
OH
preceding reaction
O
O
O [4 +2 ]
OH
oxidase or solid 50 - 70°C
O
OH
O
O O
H
H
H O HO
demethylsalvicanol
Scheme 21.15
grandione
Proposed biosynthetic pathway of grandione.
H
21.4 Diels–Alder Reactions after Formation of Reactive Substrates by Oxidation Enzymes
demethylsalvicanol into a reactive ortho-quinone that could afford a [4+2] adduct non-enzymatically. A Diels–Alder reaction between different types of terpene units is plausible in several cases. Perovskone from the heartwood of Chamaecyparis obtusa [58] is regarded as an adduct between a chiral para-quinone (28) derived from the modified diterpene barbatusol and the linear monoterpene trans-β-ocimene from geranyl diphosphate [59] (Scheme 21.16). In the total synthesis of perovskone [60], the Diels–Alder reaction of 28 and trans-α-ocimene with Lewis acid Eu(fod)3 proceeded in good regio- and diastereoselectivity to afford the desired diastereomer 29 as a major product. Since this biomimetic [4+2] cycloaddition is accompanied by the formation of the unnatural endo-adduct, a Diels–Alderase that could catalyze phenol oxidation into quinone 28 might be responsible for the construction of the perovskone skeleton during biosynthesis. A similar example involving phenol oxidation was reported in the biosynthesis of heliocides H1 and H4 [61]. Hemigossypol could be oxidized to a para-quinone that underwent cycloaddition to afford regioisomers H1 and H4 in an endo-selective manner. The involvement of ortho-quinones in the biosynthesis of natural products is recognized in many cases (Figure 21.3). The ortho-quinone dimers of the plant diterpenes hongencaotone from Salvia prionitis [62] and actephilol A from Actephila excelsa are regarded as hetero-Diels–Alder adducts [63]. In addition, endo- and exo-triterpene dimers xuxuarine Eα and Eβ [64] and also trimer triscutin A [65] are found in the South American medicinal plant Maytenus blepharodes and M. scutioides, with several related metabolites. DDQ (2,3-dichloro-5,6-dicyanobenzoquinone) oxidation of the monomer pristimerin gave the diastereomeric dimers xuxuarines Eα and Eβ. This biomimetic synthesis provides circumstantial evidence for the involvement of a Diels–Alder reaction in the biosynthesis though an alternative mechanism involving ortho-quinone 30. Ortho-quinol, another reactive species, is oxidatively derived from phenols. The dimer of the neolignan asatone was isolated from the plant Asarum teitonense (Scheme 21.17) [66]. Later, two closely related trimers, 33a and 33b, were also isolated [67]. Based on the oligomeric structure of neolignans, biosynthetic pathways of these metabolites were proposed as shown in Scheme 21.17. Oxidation of phenol 31 produced a modified ortho-quinol (32) that dimerized to give asatone through a Diels–Alder reaction. Further cycloadditions of asatone with 32 yield 33a and b. This proposal was supported by the result that the anodic oxidation of phenol 31 produced a reactive quinone methide that underwent cycloaddition to give asatone quantitatively [68]. Having no optical activity, all these lignans would be formed by spontaneous cycloadditions of the ortho-quinol 32. Observations on asatone biosynthesis strongly suggest that an oxidase produces the reactive ortho-quinol, which is released from the enzyme active site before dimerizing to afford racemic [4+2] adducts. Similar oxidase-catalyzed phenolic oxidation could produce a [4+2] adduct in the biosynthesis of monoterpene aquaticol (Scheme 21.18) [69]. A recent total synthesis of aquaticol [70] showed that a mixture of ortho-quinols 35a and b prepared by oxidation of chiral monomer 34 with SIBX (stabilized 2-iodoxybenzoic
769
OH
oxidase
preceding reaction
oxidase
preceding reaction
HO
HO
28
H
O
O
HO
HO
110°C, 48 h
45°C, 72 h;
Eu(fod)3
[4+2 ]
[4+2 ]
trans-α-ocimene
CHO O
+
28
OH O
29
O
OH
heliocide H1
O
H
HO
O
CHO O
Proposed biosynthetic pathways of perovskone and heliocides.
hemigossypol
Scheme 21.16
HO
CHO OH
barbatusol
HO
H
HO
O
+
O
HO
HO
perovskone
O
heliocide H4
O
CHO O H
Amberlyst 15 CH2Cl2, 30 min
cyclization
O O
770
21 The Diels–Alderase Never Ending Story
Figure 21.3
MeO2C
O
O HO
O HO O
O
HO
O
xuxuarine Eα
O
H
MeO2C
pristimerin
O
O
O
O HO O
O
DDQ
O
O
HO
O
O
DDQ
[4+2]
O
O
O HO O
O
O
30
O
OH
CO2Me
H
actephilol A
O OMe
O
O
H
xuxuarine Eα + xuxuarine Eβ
HO
dimerization
triscutin A
CO2Me
pristimerin
MeO2C
O
O O OMe
O oxidation product
xuxuarine Eβ
CO2Me
hongencaotone
CO2Me
O
O
Representative ortho-quinone oligomers of natural [4+2] adducts.
O
O
HO
CO2Me
OMe
OH
21.4 Diels–Alder Reactions after Formation of Reactive Substrates by Oxidation Enzymes 771
772
21 The Diels–Alderase Never Ending Story preceding reaction
MeO OH
O 32
O MeO
asatone (racemic)
Scheme 21.17
OMe 1′
32
OMe OMe O
OMe
31
[4+2 ]
OMe
OMe
MeO
or anodic ox. MeO OMe CH3OH MeO
OMe
O
[4+2 ]
oxidase
OMe
OMe
H OMe O OMe + 1′-epimer OMe OMe 33b OMe O 33a
Proposed biosynthetic pathway of asatone and its analogs.
OH
OH preceding reaction
R
H
HO R
O [4+2 ]
O
O
H
oxidase
34
OH
R
35a
aquaticol HO 35a +
1) SIBX
OH
O
S
HO H
S
S aquaticol +
2) TFA; aq. NaOH;
O
O
H
aq. Na2S2O4 (2 steps, 49%)
35b −160.4°
−160.4°
OH
4′
O
−170.7°
OH 1.93 5′ 5
6′
HO
6
4
7′
O 7
2
TS-A (35a + 35a)
Scheme 21.18
36
−162.3°
HO
3.34 3.34 2′
(1:1)
1.99 5′ 5
6′
4′
O 2′
6
4
O
3.38 3.27 7′
2 7
TS-B (35a + 35b)
Proposed biosynthetic pathway of aquaticol.
acid) provided only two diastereomers, aquaticol and 36, out of four expected products (Scheme 21.18). This indicated that chiral ortho-quinol 35a gave aquaticol as a single diastereoisomer. On the basis of computational results, the energy difference between transition states A (TS-A, aquaticol-like) and B (TS-B, 36-like) is more than 9.9 kcal mol−1 , which was explained by hyperconjugative effects. In this case, reactivity and remarkable diastereoselectivity originates from the substrate, suggesting that involvement of a Diels–Alderase is not essential in this case. The bisorbicillinoids are a family of structurally diverse fungal metabolites represented by bisorbicillinol, which shows DPPH (1,1-diphenyl-2-picrylhydrazyl) radical scavenging activity [71]. Based on extensive studies of the structure elucidation and the biosynthesis of the bisorbicillinoids, Abe et al. proposed that the
21.4 Diels–Alder Reactions after Formation of Reactive Substrates by Oxidation Enzymes O
O O
OH O preceding reaction
HO
HO OH
37a
OH
[4+2 ] 37a + 37b
oxidase
sorbicillin
HO
O
OH O OH O
O
OH
HO OH
sorbicillinol 37b
sorbiquinol
[4+2 ]
OH
O HO O
HO
O
O
37a + 37a
bisorbicillinol OH O
O
O
OK O
O
+
KO
OAc 38
[4+2 ]
H+
KOH
37a + 37b O
OK 37a-K
(2:3)
bisorbicillinol
OK 37b-K
OH O OHC
chiral bisorbicillinoids OPMB
O
39
OPMB PMB-protected sorbicillinol O
OH
HO O
O OH
OH
–
37ab
OH O O
HO O
Scheme 21.19
O
O O OH HO trichodimerol
Proposed biosynthetic pathway and biomimetic synthesis of bisorbicillinoids.
stable aromatic monomer sorbicillin could be enantioselectively oxidized into the reactive ortho-quinol sorbicillinol (37a and b), which dimerizes via two modes of [4+2] cycloadditions to provide bisorbicillinol and sorbiquinol (Scheme 21.19) [72]. During the purification of 37ab, it was found that the concentration of a solution of 37ab caused a [4+2] cycloaddition to give sorbiquinol, indicating that 37ab is highly reactive and that the conversion is non-enzymatic and occurs under mild conditions with complete regio- and stereoselectivity [73]. In synthetic studies of the bisorbicillinoids [74], basic hydrolysis of acetate 38 gave two discrete quinolates (bis-deprotonated forms of 37a-K and b-K) that underwent cycloaddition after subsequent acidification. More recently, enantioselective total synthesis of the bisorbicillinoids has been achieved via protected sorbicillinol, which was prepared by a versatile route using readily available intermediate 39 [75]. In addition, a sequential
773
774
21 The Diels–Alderase Never Ending Story
Michael addition–ketalization sequence of reactive ortho-quinols 37ab gave the highly complex cyclization product trichodimerol by either simple evaporation [72] or base-catalyzed condensation [76]. Involvement of a Diels–Alderase is not necessary in this case because a non-enzymatic reaction provided a single product and because enantioselective oxidation of sorbicillin introduced the chirality in 37ab, thus determining the stereochemistry in bisorbicillinol. Detection of a significant amount of the monomer sorbicillin in the bisorbicillinol-producing fungus indicated that an oxidase would provide the chiral reactive substrate 37ab, and that the Diels–Alder reaction of 37ab would be promoted in aqueous medium without a Diels–Alderase. This example suggests that conversion of an achiral substrate into a chiral cycloadduct is not conclusive evidence for distinguishing non-enzymatic and enzymatic Diels–Alder reactions. A literature search allowed us to find many natural [4+2] adducts derived from the oxidation of phenol to reactive ortho-quinols described above. Dimeric abietane diterpenoid maytenone from Crossopetalum rhacoma, a dimeric phenanthrenoid from Juncus acutus [77], obtunone (aromatic monoterpene–linear monoterpene) from the heartwood of Chamaecyparis obtusa [78], and vulbilide [79] (triterpene–abietane diterpene) are representative examples (Figure 21.4). To date, there is no report dealing with the corresponding oxidase. However, in the case of fungal metabolites bisorbicillinoids, the biosynthetic gene could be identified by genomic analysis, a bioinformatic search and its functional analysis because the substrate and products have been unambiguously determined. Thus, the detailed mechanism of the enzymatic Diels–Alder reaction could be elucidated in the near future.
O OH
HO
HO
O HO
OH
HO O
O
H O
O
H
H
obtunone
maytenone
O O OH
OO HO
HO O
H H
O
HO O
HO O
H Ar OH
O vulbilide
Figure 21.4
Representative ortho-quinol [4+2] adducts.
phenanthrenoid dimer
21.4 Diels–Alder Reactions after Formation of Reactive Substrates by Oxidation Enzymes
21.4.2 Conjugated Diene Derived from Dehydrogenation of Prenyl Side Chain
Prenyl side chains are frequently converted into conjugated dienes that serve as a glue to combine reactive dienophiles such as chalcone, stilbene, coumarin, and substituted prenyl diene (Scheme 21.20). Thus, in this case, a dehydrogenase providing prenyl diene is a candidate of Diels–Alderase. preceding reaction
Ar
Scheme 21.20
dehydrogenase
[4+2 ]
Ar s-cis-diene reactive substrate
see main text
Model of conjugated diene formation for a Diels–Alderase.
Kuwanons X, Y, I, and J, constituents of moraceous plants, are phytoalexins consisting of pairs of diastereomers derived from chalcones and stilbenes [80]. Based on their structures, kuwanons I and J are regarded as dimeric adducts of a chalcone precursor with the diene derived from the prenyl side chain (Scheme 21.21). Isolation of these diastereomers as optically active forms strongly indicated that the achiral precursors afford endo- and exo-adducts via an enzymatic Diels–Alder reaction [79]. The chemical feasibility of the corresponding [4+2] cycloaddition between chalcone and flavonoid with prenyl diene was confirmed [81]. In a feeding experiment with the non-natural methoxychalcone 40 into callus tissue of Morus alba, the dimeric adducts 42a and b were obtained, indicating that dehydrogenation of the prenyl group followed by [4+2] cycloaddition between 40 and 41 yielded the non-natural adduct 42b (Scheme 21.21) [82]. A plausible dehydrogenase involving kuwanon biosynthesis may convert 40 into the reactive diene 41 but in this case the same enzyme may provide an active site for the intermolecular Diels–Alder reaction between 40 and 41 to afford chiral 42b. Although a cell-free system from Morus bombycis showed the catalytic activity of a similar transformation observed in cell culture [83], unfortunately no characterization of the corresponding enzyme has been reported. Plausible dehydrogenases could produce various elaborated structures such as multicaulisin [84], sanggenon O [85], palodesagren A [86], and sorocein L [87]. Various dimeric coumarins and quinonolones derived from the corresponding monomers with prenyl dienes are found (Figure 21.5). Representative coumarin dimers are thamnosin [88], isothamnosin A [89], and toddasin [90]. In addition, dimeric quinolones and mixed dimers of coumarin–quinolone, toddacoumalone [91], mexolide [92], microcybin [93], and vepridimerine A [94] have been found from various plants (Figure 21.6). While kuwanon-type adducts were obtained as optically active forms, dimeric coumarines and quinonolones were isolated as racemic forms, indicating that dehydrogenases would release the products before the cycloaddition occurs. [4+2]-Adducts from plants are indeed frequently obtained in racemic forms [3].
775
HO
HO kuwanon I 3''-Hα kuwanon J 3''-Hβ
OH
OH
OH O O
Biosynthetic formation of kuwanons.
kuwanon X 3''-Hα kuwanon Y 3''-Hβ
OH
OH O
OH
Scheme 21.21
HO
HO
3''
3''
O
flavonoid HO HO
OH
HO
stilbene HO OH HO
41
Ar
OH O 40
chalcone
Morus alba or cell-free system from Morus bombycis
OH
OMe
RO
HO
HO
42a: R = H 42b: R = Me
OH
OH
OH O O
chalcone
OH
OMe
776
21 The Diels–Alderase Never Ending Story
21.4 Diels–Alder Reactions after Formation of Reactive Substrates by Oxidation Enzymes
777
geranyl HO HO
Ar
OH HO O
HO O
O
Ar
HO
OH O
Ar
O
O OH
O OH O
OH
HO OH
OH multicaulisin
HO
OH sanggenon O
modified chalcone
OH HO
OH OH
HO HO O OMe
H OH
O
O
O
OH
H
OH
OH
H
O coumarine palodesagren A
sorocein L
Figure 21.5 Representative examples of [4+2] adducts derived from chalcone and various metabolites with prenyl dienes.
Similar plausible Diels–Alder type cycloadditions following dehydrogenation are found in the biosynthesis of natural products with polyprenyl side chains. The unique prenylated [12]-paracyclophane quinone dimer longithorone A was isolated from the tunicate Aplidium longithorax as a cytotoxic agent, and is proposed to originate from sequential cycloaddition reactions between precursors 44 and 45 (Scheme 21.22) [95]. Later, a series of novel farnesylated benzoquinone monomers possessing meta-bridged structures were isolated from the same tunicate, indicating the possible involvement of a Diels–Alderase. Again, it is possible that either dehydrogenase forming the diene system on the bridged farnesylated core 44 or oxidase forming the α,β-unsaturated aldehyde 45 may proceed prior to the intriguing inter- and intramolecular cycloadditions. A biomimetic total synthesis of longithorone A [96] indicated that the intermolecular reaction shows inherent low reactivity and facial selectivity, suggesting the involvement of a Diels–Alderase. The enantiomer of ircinianin, a plausible intramolecular endo-adduct [97], was isolated from the marine sponge Ircinia sp. Based on isolation of linear furanosesterterpene-tetronic acids such as variabilin and the successful biomimetic synthesis of ircinianin via 46 [98], it was suggested that the linear oligo-isoprenyl precursor would be dehydrogenated to give a conjugated triene, serving as a
OH
O
O
X
R
O
PDN
O O
O
O
O
O
O
OMe
OMe
thamnosin
O
coumarine
N O quinolone Me
O
OMe
toddacoumalone
MeO
PDN =
R = H or OMe; X = O or NMe
R
R
OH
O
OMe
O
OMe
MeO
N OMe Me
O
O isothamnosin A
O
H O
H
O
O
OMe mexolide
MeO OMe
vepridimerine A
O
Me OMe N OMe
MeO
O
O
Representative examples of dimeric [4+2] adducts derived from coumarines and quinolones with prenyl dienes.
O
X
R
X
R
PDN
R
Figure 21.6
R
R
PDN
R
basic units with prenyl dienes
O
778
21 The Diels–Alderase Never Ending Story
21.5 Summary O
O [O]
O
O
dehydrogenase
O
CHO 43
44
H O
O 45
O
[4+2 ]
O H
H [4+2 ]
H OHC
O
O OHC
O
O O
longithorone A
Scheme 21.22
Proposed biosynthetic pathway of longithorone.
substrate for a biosynthetic Diels–Alder reaction (Scheme 21.23). Isolation of metabolites without a diene system suggested that the dienophile part is likely introduced prior to diene formation [98]. Similar transformation could be involved in the biosynthesis of methyl isosarcotortuoate [99] and methyl sarcophytoate [100]. 21.4.3 Cyclopentadiene Formation Derived from Dehydrogenation
Sesquiterpenes with a cyclopentadiene system are plausible key intermediates in the biosynthesis of related plant metabolites. Although putative intermolecular [4+2] adducts are frequently found in plant terpenoids (Scheme 21.24) [3], the corresponding monomer with cyclopentadiene is relatively rare, suggesting that dehydrogenation or hydroxylation–dehydration that installs a reactive diene precedes the Diels–Alder reaction (Figure 21.7). Along this line, the biogenetic pathway of plagiospirolide A was proposed [101]; this pathway was supported by the synthesis featuring a Diels–Alder reaction of diplophyllolide A and diene 47 [102]. Biogenetically related adducts between an α-methylene butenolide and cyclopentadiene are plagiospirolide E [103], biennin C [104], arteminolide B [105], ornativolide A [106], artemisolide [107], and stolonilactone [108]. Cyclopentadiene dimers absinthin [109] and vielanin A [110] and guaiane sesquiterpene dimer [111] are also found.
21.5 Summary
In this chapter, we introduced the concept of natural Diels–Alderases and provided reasonable speculation on potential Diels–Alderases such as PKS, DH, oxidase, and dehydrogenase. Diels–Alderases and their candidates described in this chapter
779
O
H H
O
H
O
O CO2Me
O
O
H
O
preceding reaction
O
MeO2C
O
O 46
OH
O
O
H
H OH
O
H
O
HO
H
O methyl sarcophytoate
MeO2C
O
[4+2 ]
Representative examples of [4+2] adducts derived from terpenoid dienes.
OH
dehydrogenase
OH methyl isosarcotortuoate
O
variabilin
CO2Me
Scheme 21.23
O
OH
H OH
ircinianin
OH O
780
21 The Diels–Alderase Never Ending Story
21.5 Summary preceding reaction
[4+2 ]
dehydrogenase or hydroxylasedehydratase
Scheme 21.24
781
see main text
s-cis-diene reactive substrate
Model of cyclic diene formation as a Diels–Alderase. preceding reaction
H O H O
or hydroxylasedehydratase
H
O
[4+2 ]
dehydrogenase
H H
O
H
H
plagiospirolide A 47
dihydrofusicoccadiene
H H diplophyllolide A O
O
H O O
O H
O
H H
OH H
H HO
O O arteminolide B
plagiospirolide E
O
H
O
biennin C
H
O
H
O
artemisolide H
O
H
absinthin
OH
H O
H
O
OAc O
H O
H
O
stolonilactone
OAc
H
H
O
OH
O
O vielanin A
O O
H
H
O
H
O O
ornativolide A
O
HO
H O
O
OH
HO H
O O
O
H H
H O
H O
O
O
guaiane sesquiterpene dimer
Figure 21.7 Proposed biosynthetic pathway of plagiospirolide A, and various [4+2] adducts derived from terpenes with reactive cyclopentadiene.
are not categorized as a single enzyme group because they do not show any characteristic catalytic mechanism. Therefore, we propose that any enzyme providing reactive dienes and dienophiles could be a Diels–Alderase. In our extensive survey of natural [4+2] adducts we have frequently encountered reactions related to oxidation (phenol oxidation, dehydrogenation) and dehydration (PKS or other) forming reactive substrates (Scheme 21.25). Currently, only limited information is available on the Diels–Alderases. However, innovative methodologies recently developed will provide information on gene clusters of many natural [4+2] adducts.
782
21 The Diels–Alderase Never Ending Story A
preceding reaction
B
precursor dehydrogenase oxidase PKS
preceding reaction
precursor PKS or PKS-NRPS dehydratase oxidase decarboxylase
C D reactive diene
A [4+2 ]
Diels-Alderase
G
G
B
E X
F
C D
E
F X reactive dienophile
Scheme 21.25 General biosynthetic pathway involving skeletal construction with putative Diels–Alderase.
This allows biochemical investigation of Diels–Alderases, and we anticipate that the outline of this intriguing enzyme should be elucidated within the next ten years.
References 1. Carruthers, W. (1990) Cycloaddition Re-
2.
3.
4.
5.
actions in Organic Synthesis, Pergamon, Oxford. (a) Nicolaou, K.C., Edmonds, D.J., and Bulger, P.G. (2006) Angew. Chem. Int. Ed., 45, 7134–7186; (b) Takao, K., Munakata, R., and Tadano, K. (2005) Chem. Rev., 105, 4779–4807. (a) Ichihara, A. and Oikawa, H. (1998) Curr. Org. Chem., 2, 365–394; (b) Oikawa, H. and Tokiwano, T. (2004) Nat. Prod. Rep., 21, 321–352; (c) Oikawa, H. (2005) Bull. Chem. Soc. Jpn., 78, 537–554; (d) Oikawa, H. (2010) in Comprehensive Natural Products II: Chemistry and Biology, vol. 8 (eds L. Mander and H.-W. Liu), Elsevier, Oxford, pp. 277–314; (e) Stocking, E.M. and Williams, R.M. (2003) Angew. Chem. Int. Ed., 42, 3078–3115. (a) Alikaridis, F., Papadakis, D., Pantelia, K., and Kephalas, T. (2000) Fitoterapia, 71, 379–384; (b) Lee, J.I., Hsu, B.H., Wu, D., and Barrett, J.S. (2006) J. Chromatogr. A, 1116, 57–68. Dewick, P.M. (2008) Medicinal Natural Products: A Biosynthetic Approach,
6.
7.
8.
9.
10.
11.
3rd edn, John Wiley & Sons, Ltd., Chichester. Oikawa, H., Katayama, K., Suzuki, Y., and Ichihara, A. (1995) J. Chem. Soc., Chem. Commun., 1321–1322. Auclair, K., Sutherland, A., Kennedy, J., Witter, D.J., Van den Heever, J.P., Hutchinson, C.R., and Vederas, J.C. (2000) J. Am. Chem. Soc., 122, 11519–11520. Watanabe, K., Mie, T., Ichihara, A., Oikawa, H., and Honma, M. (2000) J. Biol. Chem., 275, 38393–38401. Ose, T., Watanabe, K., Mie, T., Honma, M., Watanabe, H., Yao, M., Oikawa, H., and Tanaka, I. (2003) Nature, 422, 185–189. (a) Moore, R.N., Bigam, G., Chan, J.K., Hogg, A.M., Nakashima, T.T., and Vederas, J.C. (1985) J. Am. Chem. Soc., 107, 3694–3701; (b) Yoshizawa, Y., Witter, D.J., Liu, Y., and Vederas, J.C. (1994) J. Am. Chem. Soc., 116, 2693–2694. Auclair, K., Kennedy, J., Hutchinson, C.R., and Vederas, J.C. (2001) Bioorg. Med. Chem. Lett., 11, 1527–1531.
References 12. Sorensen, J.L., Auclair, K., Kennedy,
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
J., Hutchinson, C.R., and Vederas, J.C. (2003) Org. Biol. Chem., 1, 50–59. Kennedy, J., Auclair, K., Kendrew, S.G., Park, C., Vederas, J.C., and Hutchinson, C.R. (1999) Science, 284, 1368–1372. Ma, S.M., Li, J.W.H., Choi, J.W., Zhou, H., Lee, K.K.M., Moorthie, V.A., Xie, X.K., Kealey, J.T., Da Silva, N.A., Vederas, J.C., and Tang, Y. (2009) Science, 326, 589–592. Sakurai, I., Miyajima, H., Akiyama, K., Shimizu, S., and Yamamoto, Y. (1988) Chem. Pharm. Bull., 36, 2003–2011. (a) Watanabe, K., Oikawa, H., Yagi, K., Ohashi, S., Mie, T., Ichihara, A., and Honma, M. (2000) J. Biochem., 127, 467–473; (b) Oikawa, H., Watanabe, K., Yagi, K., Ohashi, S., Mie, T., Ichihara, A., and Honma, M. (1999) Tetrahedron Lett., 40, 6983–6986. (a) Watanabe, K., Mie, T., Ichihara, A., Oikawa, H., and Honma, M. (2000) Biosci. Biotechnol. Biochem., 64, 530–538; (b) Watanabe, K., Mie, T., Ichihara, A., Oikawa, H., and Honma, M. (2000) Tetrahedron Lett., 41, 1443–1446. Serafimov, J.M., Westfeld, T., Meier, B.H., and Hilvert, D. (2007) J. Am. Chem. Soc., 129, 9580–9581. Serafimov, J.M., Lehmann, H.C., Oikawa, H., and Hilvert, D. (2007) Chem. Commun., 1701–1703. Guimaraes, C.R.W., Udier-Blagovic, M., and Jorgensen, W.L. (2005) J. Am. Chem. Soc., 127, 3577–3588. Serafimov, J.M., Gillingham, D., Kuster, S., and Hilvert, D. (2008) J. Am. Chem. Soc., 130, 7798–7799. (a) Ichihara, A., Tazaki, H., and Sakamura, S. (1983) Tetrahedron Lett., 24, 5373–5376; (b) Alam, S.S., Bilton, J.M., Slawin, M.Z., Williams, D.J., Sheppard, R.N., and Strange, R.M. (1989) Phytochemistry, 28, 2627–2630. (a) Oikawa, H., Yokota, T., Ichihara, A., and Sakamura, S. (1989) J. Chem. Soc., Chem. Commun., 1284–1285; (b) Oikawa, H., Yokota, T., Sakano, C., Suzuki, Y., Naya, A., and Ichihara, A. (1998) Biosci. Biotechnol. Biochem. 62, 2016–2022
24. (a) Oikawa, H., Yokota, T., Abe, T.,
25.
26.
27.
28.
29.
30. 31.
32.
33. 34.
35. 36.
37.
Ichihara, A., Sakamura, S., Yoshizawa, Y., and Vederas, J.C. (1989) J. Chem. Soc., Chem. Commun., 1282–1284. (a) Oikawa, H., Suzuki, Y., Naya, A., Katayama, K., and Ichihara, A. (1994) J. Am. Chem. Soc., 116, 3605–3606; (b) Oikawa, H., Suzuki, Y., Katayama, K., Naya, A., Sakano, C., and Ichihara, A. (1999) J. Chem. Soc., Perkin Trans. 1, 1225–1232. (a) Katayama, K., Kobayashi, T., Oikawa, H., Honma, M., and Ichihara, A. (1998) Biochim. Biophys. Acta, 1384, 387–395; (b) Katayama, K., Kobayashi, T., Chijimatsu, M., Ichihara, A., and Oikawa, H. (2008) Biosci. Biotechnol. Biochem., 72, 604–607. Kasahara, K., Miyamoto, T., Fujimoto, T., Oguri, H., Tokiwano, T., Oikawa, H., Ebizuka, Y., and Fujii, I. (2010) ChemBioChem, 11, 1245–1252. Oikawa, H., Kobayashi, T., Katayama, K., Suzuki, Y., and Ichihara, A. (1998) J. Org. Chem., 63, 8748–8756. Raimondi, L., Brown, F.K., Gonzalez, J., and Houk, K.N. (1992) J. Am. Chem. Soc., 114, 4796–4804. Breslow, R. (1991) Acc. Chem. Res., 24, 159–164. Ruiz-L´opez, M.F., Assfeld, X., Garc´ıa, J.I., Mayoral, J.A., and Salvatella, L. (1993) J. Am. Chem. Soc., 115, 8780–8787. Xu, J.A., Deng, Q.L., Chen, J.G., Houk, K.N., Bartek, J., Hilvert, D., and Wilson, I.A. (1999) Science, 286, 2345–2348. Fischbach, M.A. and Walsh, C.T. (2006) Chem. Rev., 106, 3468–3496. Sims, J.W., Fillmore, J.P., Warner, D.D., and Schmidt, E.W. (2005) Chem. Commun., 186–188. Schumann, J. and Hertweck, C. (2006) J. Biotechnol., 124, 690–703. Sekita, S., Yoshihira, K., Natori, S., and Kuwano, H. (1973) Tetrahedron Lett., 14, 2109–2112. (a) Oikawa, H., Murakami, Y., and Ichihara, A. (1992) J. Chem. Soc., Perkin Trans. 1, 2955–2959; (b) Oikawa, H., Murakami, Y., and Ichihara, A.
783
784
21 The Diels–Alderase Never Ending Story
38. 39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50. 51. 52.
(1992) J. Chem. Soc., Perkin Trans. 1, 2949–2953. Schumann, J. and Hertweck, C. (2007) J. Am. Chem. Soc., 129, 9564–9565. Thomas, E.J. and Watts, J.P. (1999) J. Chem. Soc., Perkin Trans. 1, 3285–3290. Abe, Y., Suzuki, T., Ono, C., Iwamoto, K., Hosobuchi, M., and Yoshikawa, H. (2002) Mol. Genet. Genom., 267, 636–646. Holtzel, A., Schmid, D.G., Nicholson, G.J., Krastel, P., Zeeck, A., Gebhardt, K., Fiedler, H.P., and Jung, G. (2004) J. Antibiot., 57, 715–720. Mallams, A.K., Puar, M.S., Rossman, R.R., McPhail, A.T., and Macfarlane, R.D. (1981) J. Am. Chem. Soc., 103, 3940–3943. Muntwyler, R. and Keller-Schierlein, W. (1972) Helv. Chim. Acta, 55, 2071–2094. Hirayama, N., Kasai, M., Shirahata, K., Ohashi, Y., and Sasada, Y. (1982) Bull. Chem. Soc. Jpn., 55, 2984–2987. Zhang, H., White-Phillip, J.A., Melancon, C.E., Kwon, H.J., Yu, W.L., and Liu, H.W. (2007) J. Am. Chem. Soc., 129, 14670–14683. Jia, X.Y., Tian, Z.H., Shao, L., Qu, X.D., Zhao, Q.F., Tang, J., Tang, G.L., and Liu, W. (2006) Chem. Biol., 13, 575–585. Fang, J., Zhang, Y.P., Huang, L.J., Jia, X.Y., Zhang, Q., Zhang, X., Tang, G.L., and Liu, W. (2008) J. Bacteriol., 190, 6014–6025. Sun, Y.H., Hahn, F., Demydchuk, Y., Chettle, J., Tosin, M., Osada, H., and Leadlay, P.F. (2010) Nat. Chem. Biol., 6, 99–101. Westley, J.W., Evans, R.H.J., Sello, L.H., Troupe, N., Liu, C.-M., and Blount, J.F. (1979) J. Antibiot., 32, 100–107. Roege, K.E. and Kelly, W.L. (2009) Org. Lett., 11, 297–300. Li, C.X., Roege, K.E., and Kelly, W.L. (2009) ChemBioChem, 10, 1064–1072. Kirst, H.A., Michel, K.H., Martin, J.W., Creemer, L.C., Chio, E.H., Yao, R.C., Nakatsukasa, W.M., Boeck, L.D., Occolowitz, J.L., Paschal, J.W., Deeter, J.B., Jones, N.D., and Thompson,
53.
54.
55. 56.
57.
58.
59.
60. 61.
62. 63.
64.
65.
66.
G.D. (1991) Tetrahedron Lett., 32, 4839–4842. Waldron, C., Matsushima, P., Rosteck, P.R. Jr., Broughton, M.C., Turner, J., Madduri, K., Crawford, K.P., Merlo, D.J., and Baltz, R.H. (2001) Chem. Biol., 8, 487–499. Kim, H.J., Pongdee, R., Wu, Q.Q., Hong, L., and Liu, H.W. (2007) J. Am. Chem. Soc., 129, 14582–14584. Bazan, A.C., Edwards, J.M., and Weiss, U. (1978) Tetrahedron, 34, 3005–3015. H¨olscher, D. and Schneider, B. (1995) J. Chem. Soc., Chem. Commun., 525–526. Aoyagi, Y., Takahashi, Y., Satake, Y., Fukaya, H., Takeya, K., Aiyama, R., Matsuzaki, T., Hashimoto, S., Shiina, T., and Kurihara, T. (2005) Tetrahedron Lett., 46, 7885–7887. Parvez, A., Choudhary, M.I., Akhter, F., Noorwala, M., Mohammad, F.V., Hasan, N.M., Zamir, T., and Ahmad, V.U. (1992) J. Org. Chem., 57, 4339–4340. Dudareva, N., Martin, D., Kish, C.M., Kolosova, N., Gorenstein, N., Faldt, J., Miller, B., and Bohlmann, J. (2003) Plant Cell, 15, 1227–1241. Majetich, G. and Zhang, Y. (1994) J. Am. Chem. Soc., 116, 4979–4980. Bell, A.A., Stipanovic, R.D., O‘Brien, D.H., and Fryxell, P.A. (1978) Phytochemistry, 17, 1297–1305. Li, M., Zhang, J.S., and Chen, M.Q. (2001) J. Nat. Prod., 64, 971–972. Ovenden, S.P.B., Yew, A.L.S., Glover, R.P., Ng, S., Rossant, C.J., Regalado, J.C., Soejarto, D.D., Buss, A.D., and Butler, M.S. (2001) Tetrahedron Lett., 42, 7695–7697. (a) Gonzalez, A.G., Rodriguez, F.M., Bazzocchi, I.L., and Ravelo, A.G. (2000) J. Nat. Prod., 63, 48–51; (b) Jacobsen, N.E., Wijeratne, E.M.K., Corsino, J., Furlan, M., Bolzani, V.D., and Gunatilaka, A.A.L. (2008) Bioorg. Med. Chem., 16, 1884–1889. Gonzalez, A.G., Alvarenga, N.L., Bazzocchi, I.L., Ravelo, A.G., and Moujir, L. (1999) J. Nat. Prod., 62, 1185–1187. Yamamura, S., Terada, Y., Chen, Y., Hong, M., Hsu, H., Sasaki, K., and
References
67.
68.
69. 70.
71.
72. 73.
74.
75.
76. 77.
78.
79.
80. 81.
82.
83.
Hirata, Y. (1976) Bull. Chem. Soc. Jpn., 49, 1940. Niwa, M., Terada, Y., Nonoyama, M., and Yamamura, S. (1979) Tetrahedron Lett., 49, 813–816. Nishiyama, A., Eto, H., Terada, Y., Iguchi, M., and Yamamura, S. (1983) Chem. Pharm. Bull., 31, 2820. Su, B.N., Yang, L., Gao, K., and Jia, Z.J. (2000) Planta Med., 66, 281–283. Gagnepain, J., Castet, F., and Quideau, W. (2007) Angew. Chem. Int. Ed., 46, 1533–1535. Abe, N., Murata, T., and Hirota, A. (1998) Biosci. Biotechnol. Biochem., 62, 661–666. Abe, N. and Hirota, A. (2004) J. Synth. Org. Chem. Jpn., 62, 584–597. Abe, N., Sugimoto, O., Tanji, K., and Hirota, A. (2000) J. Am. Chem. Soc., 122, 12606–12607. Nicolaou, K.C., Vassilikogiannakis, G., Simonsen, K.B., Baran, P.S., Zhong, Y.L., Vidali, V.P., Pitsinos, E.N., and Couladouros, E.A. (2000) J. Am. Chem. Soc., 122, 3071–3079. Hong, R., Chen, Y., and Deng, L. (2005) Angew. Chem. Int. Ed., 44, 3478–3481. Barnes-Seeman, D. and Corey, E.J. (1999) Org. Lett., 1, 1503–1504. Falshaw, C.P. and King, T.J. (1983) J. Chem. Soc., Perkin Trans. I, 1749–1752. Kuo, Y.H., Chen, C.H., and Huang, S.L. (1998) Chem. Pharm. Bull., 46, 181–183. Alvarenga, N.L., Ferro, E.A., Ravelo, A.G., Kennedy, M.L., Maestro, M.A., and Gonzalez, A.G. (2000) Tetrahedron, 56, 3771–3774. Nomura, T. and Hano, Y. (1994) Nat. Prod. Rep., 11, 205–218. Nomura, T., Fukai, T., Narita, T., Terada, S., Uzawa, J., Iitaka, Y., Takasugi, M., Ishikawa, S., Nagao, S., and Masamune, T. (1981) Tetrahedron Lett., 22, 2195–2198. Hano, Y., Nomura, T., and Ueda, S. (1990) J. Chem. Soc., Chem. Commun., 610–613. Nomura, T. (2001) Yakugaku Zasshi, 121, 535–556.
84. Ferrari, F., Delle Monache, F., Suarez,
85. 86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
A., and Compagnone, R.S. (2000) Fitoterapia, 71, 213–215. Shi, Y.Q., Fukai, T., and Nomura, T. (2001) Heterocycles, 54, 639–646. Shirota, O., Takizawa, K., Sekita, S., Satake, M., Hirayama, Y., Hakamata, Y., Hayashi, T., and Yanagawa, T. (1997) J. Nat. Prod., 60, 997–1002. Ferrari, F., Filho, V.C., Cabras, T., and Messana, I. (2003) J. Nat. Prod., 66, 581–582. Kutney, J.P., Inaba, T., and Dreyer, D.L. (1970) Tetrahedron, 26, 3171–3184. Gonzalez, A.G., Cardona, R.J., Diaz, C.E., Lopez, D.H., and Rodriguez, L.F. (1977) An. Quim., 73, 1510–1514. Sharma, P.N., Shoeb, A., Kapil, R.S., and Popli, S.P. (1980) Phytochemistry, 19, 1258–1260. Ishii, H., Kobayashi, J., and Ishikawa, T. (1991) Tetrahedron Lett., 32, 6907–6910. Chakraborty, D.P., Roy, S., Chakraborty, A., Mandal, A.K., and Chowdhury, B.K. (1980) Tetrahedron, 36, 3563–3564. Hasan, C.M., Kong, D.-Y., Gray, A.I., Waterman, P.G., and Armstrong, J.A. (1993) J. Nat. Prod., 56, 1839–1842. Ngadjuri, T.B., Ayafor, J.F., Sandengam, B.L., Connolly, J.D., Rycroft, D.S., Khalid, S.A., Waterman, P.G., Brown, N.M.D., Grundon, M.F., and Ramachandran, V.N. (1982) Tetrahedron Lett., 23, 2041–2044. (a) Fu, X., Hossain, B., Helm, D., and Schmitz, F.J. (1994) J. Am. Chem. Soc., 116, 12125–12126; (b) Fu, X., Hossain, M.B., Schmitz, F.J., and VanderHelm, D. (1997) J. Org. Chem., 62, 3810–3819. Layton, M.E., Morales, C.A., and Shair, M.D. (2002) J. Am. Chem. Soc., 124, 773–775. Hofheinz, W. and Schonholzer, P. (1977) Helv. Chim. Acta, 60, 1367–1370. Uenishi, J., Kawahama, R., and Yonemitsu, O. (1997) J. Org. Chem., 62, 1691–1701.
785
786
21 The Diels–Alderase Never Ending Story 99. Jingyu, S., Kanghou, L., Tangsheng,
100.
101.
102.
103.
104. 105.
P., Cun-heng, H., and Clardy, J. (1986) J. Am. Chem. Soc., 108, 177–178. Kusumi, T., Igari, M., Ishitsuka, M.O., Ichikawa, A., Itezono, Y., Nakayama, N., and Kakisawa, H. (1990) J. Org. Chem., 55, 6286–6289. Sp¨orle, J., Becker, H., Gupta, M.P., Veith, M., and Huch, V. (1989) Tetrahedron, 45, 5003–5014. Kato, N., Wu, X., Nishikawa, H., Nakanishi, K., and Takeshita, H. (1994) J. Chem. Soc., Perkin Trans. 1, 1047–1053. Sporle, J., Becker, H., Allen, N.S., and Gupta, M.P. (1991) Phytochemistry, 30, 3043–3047. Gao, F., Wang, H.P., and Mabry, T.J. (1990) Phytochemistry, 29, 3875–3880. Lee, S.H., Kim, H.K., Seo, J.M., Kang, H.M., Kim, J.H., Son, K.H., Lee, H.,
106. 107.
108.
109.
110.
111.
Kwon, B.M., Shin, J., and Seo, Y. (2002) J. Org. Chem., 67, 7670–7675. Zdero, C. and Bohlmann, F. (1989) Phytochemistry, 28, 3105–3120. Kim, J.H., Kim, H.K., Jeon, S.B., Son, K.H., Kim, E.H., Kang, S.K., Sung, N.D., and Kwon, B.M. (2002) Tetrahedron Lett., 43, 6205–6208. Iguchi, K., Fukaya, T., Takahashi, H., and Watanabe, K. (2004) J. Org. Chem., 69, 4351–4355. Beauhaire, J., Fourrey, J.L., Vuilhorgne, M., and Lallemand, J.Y. (1980) Tetrahedron Lett., 21, 3191–3194. Kamperdick, C., Phuong, N.M., Van Sung, T., and Adam, G. (2001) Phytochemistry, 56, 335–340. Martins, D., Osshiro, E., Roque, N.F., Marks, V., and Gottlieb, H.E. (1998) Phytochemistry, 48, 677–680.
787
22 Bio-Inspired Transfer Hydrogenations Magnus Rueping, Fenja R. Schoepke, Iuliana Atodiresei, and Erli Sugiono
22.1 Introduction
The demand for chiral molecules, particularly those with hydrogen as part of the stereogenic center, has increased in recent years, and intensive research has been carried out in both industry and academia to develop efficient methods for synthesizing such compounds [1]. In this context, significant attention has been devoted to the asymmetric hydrogenation of unsaturated compounds, for example, olefins, carbonyls, and imines, which is recognized as one of the most important and convenient routes to the corresponding optically active products [2]. So far, most of the enantioselective reductions rely on biological processes or transition metal catalyzed high-pressure hydrogenations, hydrosilylations, and transfer hydrogenations. Beside their high substrate specificity, the enzymatic processes sometimes suffer from undesired by-products, poor catalyst stability under the operational conditions, substrate and/or product inhibition, and problems with catalyst recovery. Likewise, despite the high reactivity and selectivity exhibited by the organometallic complexes employed in the metal-catalyzed processes, most of these protocols suffer from a limited number of substrates and difficulties with catalyst separation and recycling. Hence, an alternative approach to these chiral compounds would be of great value.
22.2 Nature’s Reductions: Dehydrogenases as a Role Model
Enzymes are catalysts that evolve in Nature and one of their characteristics is high selectivity. During biochemical transformations, enzymes are assisted by non-proteinogenic molecules called co-factors. For instance, nicotinamide adenine dinucleotide (NADH), one of the essential co-factors in Nature, serves as a hydride source for various biological reductions. The synthesis of the amino acid glutamate by the glutamate dehydrogenase (GDH) catalyzed reductive amination of 2-ketoglutarate represents one example (Equation 22.1). The reaction is reversible Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
788
22 Bio-Inspired Transfer Hydrogenations
K89 H N H D165 H O S380
H
O
O O Q110
H N H
O
O O H N H
NH2
N
O H H N H K113
(a)
O H H
HO N H2N
N N
O O O P O OH NADH O P O O O
O
N HO
OH
(b)
Figure 22.1 Proposed activation mechanism of GDH: (a) crystal structure of the GDH active site; (b) activation of α-imino glutarate by protonation from aspartate D165 with subsequent hydride transfer from the NADH to provide the amino acid glutamate.
although the equilibrium favors glutamate formation: 2-ketoglutarate + NH3 + NAD(P)H ←→ glutamate + NAD(P)+
(22.1)
On the basis of kinetic studies and X-ray crystal structure analysis, a possible mechanism for the mode of action of GDH has been proposed (Figure 22.1) [3]. It was envisaged that the aspartate D165 residue present in the catalytic active pocket of the enzyme is essential for the reductive amination of 2-ketoglutarate, being involved in the activation of α-iminoglutarate by protonation and conversion into a highly reactive iminium ion. Subsequent hydride transfer from NADH leads to the formation of the amino acid glutamate. Site-directed mutagenesis studies in which the putative catalytic aspartil residue, Asp-165, was replaced by serine support this mechanism since, despite conservation of the native folding pattern, the catalytic activity of the modified enzyme considerably decreased [4]. This clearly emphasizes the role of the aspartate in the transfer hydrogenation and indicates protonation as the key step for the reductive amination. Based on the aforementioned activation mechanism, several groups have carried out intensive research to disclose efficient non-enzymatic systems capable of inducing asymmetry during the catalytic reduction of unsaturated substrates in a similar manner to Nature’s dehydrogenases.
22.3 Brønsted Acid Catalyzed Transfer Hydrogenation of Imines, Imino Esters, and Enamines
The metal-catalyzed enantioselective reduction of olefin and carbonyl compounds represents a versatile transformation providing valuable intermediates for the
22.3 Brønsted Acid Catalyzed Transfer Hydrogenation of Imines, Imino Esters, and Enamines
N
R2 R1
R
"Hydride"
HN
Brønsted acid
1
NADH
H
H
NH2
H H
H H H
N H2N
N
N N
HO
N HO
O HO
O
O O HO
H
H H O O P O
P O O
O H H O R'O
R2
R R1 rac-3
Potential NADH analogues:
O
Potential Brønsted acids: O H H
H2N N R
AcOH HCl TFA HBF4 CSA DPP
R3SiH OR"
789
NaBH4
N H 2 Hantzsch-ester R' = R" R' = R"
Figure 22.2 Towards a biomimetic transfer hydrogenation of ketimines with NADH analogs.
synthesis of natural products and biologically active substances or, when applied in the last steps, optically active target molecules. Recent comprehensive overviews on the hydrogenation of alkenes, carbonyls, and amines, respectively, are given in the chapters of books given in References [1b] and [2a,b]. Conversely, the asymmetric reduction of imines, which is a convenient route to chiral amines, still represents a challenging field. No widely applicable metal-based system has been developed to date. Therefore, as an alternative protocol, the asymmetric transfer hydrogenation of imines was one of the first reactions performed in a bio-inspired fashion, by employing a combination of NADH analog as hydride source and catalytic amounts of Brønsted acids (Figure 22.2). The strategy is based on activating the imine by catalytic protonation and formation of an iminium ion. Subsequent hydride transfer from a suitable NADH mimic yields the corresponding amine. 1,4-Dihydropyridines, also known as Hantzsch esters (HEH) [5], are synthetic analogs of NADH. Their potential as a hydrogen source was acknowledged for the first time in 1955 by Mauzerall and Westheimer [6], who showed that dihydropyridines can reduce carbonyl compounds by a direct hydrogen transfer to the substrate. Since then, a broad range of transfer hydrogenations conducted with HEH in combination with different Lewis acids and additives has been reported [7, 8]. Interestingly, whereas considerable efforts have been made to design effective HEH reagents that operate in conjunction with different metal catalysts, few metal-free asymmetric versions had been explored until recently. In the late 1980s Singh and Batra described the use of HEH as reducing agents in the metal-free acid catalyzed reduction of imines to give products in up to 62% ee [7h]. Furthermore,
790
22 Bio-Inspired Transfer Hydrogenations O
H
H
O
EtO
N
N H 2a (1.4 equiv)
R2 R1
R
OEt
5 mol% DPP 40 °C, CH2Cl2
1
Scheme 22.1
HN
R2
R R1 rac-3 14 examples
Brønsted acid catalyzed reduction of ketimines.
List and MacMillan reported their usefulness in the organocatalytic enantioselective transfer hydrogenation of enals with chiral imidazolidinone catalysts [9, 10]. Subsequently, in 2005, Rueping described their use in the metal-free catalyzed transfer hydrogenation of imines in both an achiral as well as a chiral fashion [11]. The first attempts were devoted to the development of a viable protocol for the reduction of imines with catalytic amounts of achiral Brønsted acids (Figure 22.2) [11a]. Promising results were obtained when propiophenone ketimine 1 [R = Ph, R1 = Et, R2 = p-methoxyphenyl (PMP)] was treated with HEH 2a in the presence of various Brønsted acids, to give the corresponding amine rac-3 in moderate to good yields (Scheme 22.1). This preliminary study revealed diphenyl phosphate (DPP) as the best catalyst for this transformation. Further investigations focused on the effect of temperature and solvent on the yield and showed that the reaction could be performed well even at higher temperatures in non-polar aromatic solvents or at milder temperatures in halogenated solvents. Accordingly, a large variety of aromatic ketimines 1, including α-imino esters, have been reduced, under the optimized reaction conditions, to give products 3 in good to excellent yields (67–92%) (Figure 22.3). Next, as the reduction of ketimines leads to the formation of a stereocenter, a chiral proton source [12] has been applied as catalyst. Since in the achiral version of the reaction DPP performed best in terms of yields, chiral phosphoric acids based on the BINOL core structure have been examined [13]. These had previously been used as chiral resolving agents [14], as ligands in metal catalysis [15], and more recently as organocatalysts in the enantioselective Mannich reaction [16, 17]. Notably, the steric demand and the acidity of the chiral BINOL phosphoric acids can easily be adjusted by introducing substituents with different steric
HN
PMP
HN
Ph
HN
Et
Bu
PMP
rac-3b: 70%
Figure 22.3
PMP
HN
Me
Me
Ph Me
HN
PMP CO2Me
F
MeO
rac-3a: 92%
HN
rac-3c: 76%
rac-3d: 67%
rac-3e: 75%
Scope of the Brønsted acid catalyzed transfer hydrogenation.
rac-3f: 77%
22.3 Brønsted Acid Catalyzed Transfer Hydrogenation of Imines, Imino Esters, and Enamines
bulk and electronic properties in the 3,3 -positions of the backbone. The synthesis of the chiral BINOL phosphate derivatives is straightforward [11c–e, 16–22]. For example, Rueping reported the synthesis of 5 starting from the protected (R)-3,3 -dibromo-1,1 -binaphthyl-2,2 -diol and its octahydro analog derivative 4a, which are obtained from BINOL and H8 -BINOL, respectively [11c-e] (Scheme 22.2). Suzuki coupling of 4a with various aryl boronic acids afforded the corresponding 3,3-aryl-substituted BINOL derivatives, which were converted into the desired catalysts 5 via deprotection and subsequent phosphorylation. A similar approach towards this class of compounds was reported by Akiyama and involves Suzuki coupling of a bis(boronic acid) binaphthol derivative with different aryl halides [18], deprotection, and phosphorylation [16b]. The synthesis of the 3,3 -bis(2,4,6-triisopropylphenyl) BINOL derivative required for the preparation of 5e can be accomplished according to Schrock [20]. The silylated Brønsted acids 6 can be synthesized following a procedure reported by Yamamoto [21] and MacMillan [22], which involves a Brook-type-rearrangement of the BINOL-derivative 4b using t-BuLi and subsequent phosphorylation. The chiral BINOL-phosphoric acid derivatives 5 were evaluated in the enantioselective transfer hydrogenation of ketimines 1 (Scheme 22.3). The best results were obtained when the reduction was carried out with 1.4 equiv. of dihydropyridine 2a at 60 ◦ C in benzene in the presence of 20 mol.% 5a as catalyst [11c,d]. Under optimized conditions, a broad range of substituted aromatic ketimines could be reduced to the corresponding amines 3 in moderate to high yields (46–91%) and with good enantioselectivities (68–84% ee) (Figure 22.4). Moreover, for two selected substrates the enantiomeric excess could be increased to 94% and 98%, respectively, by simple recrystallization. This represents the first example of a highly enantioselective Brønsted acid catalyzed transfer hydrogenation. List described the catalytic enantioselective imine reduction employing the same combination of HEH/BINOL-phosphoric acid [23]. Use of phosphoric acids with sterically demanding residues in the 3,3 -positions of the BINOL skeleton allowed shorter reaction times, milder temperatures, and lower catalyst loadings. Accordingly, a large variety of aromatic ketimines 1 were reduced in the presence of ent-5e as catalyst to give products ent-3 in high yields (85–98%) and selectivities (80–92% ee) (Figure 22.5). The protocol is also suitable for the reduction of in situ generated imines as well as for the reduction of more challenging aliphatic substrates. The protocol has also been extended to the reductive amination of α-branched aldehydes 7, yielding chiral β-branched amines 9 via a dynamic kinetic resolution process (Scheme 22.4) [24]. The necessary prerequisites to this approach are (i) racemization of the in situ formed imine under the reaction conditions and (ii) one of the two enantiomers undergoes a much faster reaction than the other one (Scheme 22.5). Screening of reaction conditions, including different phosphoric acid derivatives, various solvents, and temperatures, in the reaction of hydratopic aldehyde 7 (R = Ph, R1 = Me) with p-anisidine in the presence of HEH 2b identified optimum conditions for a highly selective process. Use of 5 A˚ molecular sieves was needed to achieve a high degree of asymmetric induction. Furthermore, handling
791
Scheme 22.2
4a 5
Ar
O O P O OH
Ar
5a Ar: 3,5-(CF3)-Phenyl 5b Ar: 9-Phenanthryl 5c Ar: Anthracenyl 5d Ar: 2-Naphthyl 5e Ar: 2,4,6-(i-Pr)3-Phenyl 5f Ar: [H8]-9-Phenanthryl 5g Ar: [H8]-Phenyl
3. POCl3, pyr. HCl, H2O
2. HCl conc.
1. Pd(PPh3)4 ArB(OH)2
4b
Synthesis of BINOL-phosphoric acid derivatives 5 and 6.
Br
OMOM OMOM
Br
Br
OSiR3 OSiR3
Br
2. POCl3, pyr. HCl, H2O
1. t-BuLi, THF
SiR3 6a R: SiPh3 6b R: [H8]-SiPh3 6c R: SiPh2Me 6d R: SiPhMe2
6
O O P O OH
SiR3
792
22 Bio-Inspired Transfer Hydrogenations
22.3 Brønsted Acid Catalyzed Transfer Hydrogenation of Imines, Imino Esters, and Enamines O
H
H
O
EtO
N
OEt N H 2a (1.4 equiv)
R2 R1
R
HN
R1 3 13 examples R
20 mol % 5a 60 °C, C6H6
1
Scheme 22.3
Metal-free asymmetric transfer hydrogenation of ketimines.
PMP
HN
HN
PMP
HN
Me
Me
PMP Me
OMe
F 3d: 82%, 84% ee HN
R2
3g: 72%, 76% ee 3h: 91%, 78% ee PMP PMP HN HN
Ph
Me
Me
Me
Ph 3i: 71%, 72% ee a
3j: 71%, 74% ee (98)a
3k: 82%, 70% ee (94)a
After one recrystallization from methanol
Figure 22.4
Scope of the enantioselective biomimetic transfer hydrogenation of ketimines.
HN
PMP Me
ent-3d: 95%, 85% ee PMP Me
ent-3l: 96%, 88% ee
Figure 22.5
PMP
HN
PMP Me
Me OMe
F
HN
HN
ent-3g: 92%, 80% ee ent-3h: 88%, 92% ee HN
PMP Me
ent-3k: 85%, 84% ee
HN
PMP Me
ent-3m: 80%, 90% ee
Asymmetric biomimetic transfer hydrogenation of ketimines according to List.
the reaction under oxygen-free conditions was required to circumvent formation of undesired by-products such as acetophenone and p-formylanisidine. The reaction proved general with respect to aldehyde and amine structure. Substrates bearing aromatic residues with different substitution patterns and electronic properties performed best in this reaction (80–96% yield, 94–98% ee) (Figure 22.6). Aliphatic aldehydes are also tolerated although with lower selectivities
793
794
22 Bio-Inspired Transfer Hydrogenations O
H
O
H
Ot-Bu
MeO N H
O R
+ R2NH2
H 1
R
7
R ∗
2b (1.2 equiv) 5 mol% 5e 5Å MS, 6 °C, C6H6
8
Scheme 22.4
NHR2
R1 9 16 examples
Organocatalytic asymmetric reductive amination of α-branched aldehydes.
O
H H
Ot-Bu
MeO
2
O
R NH2 N H 2b
8
O R
5 mol% 5e, 5Å MS, C6H6, 6°C
R ∗
H R1 7
NHR2
R1 9 8
8 2b
(ii) 2
N R
R
HN
(i)
R
H
R2
N
(i)
R
H R
R
H R1
1
1
R2
Scheme 22.5 Dynamic kinetic resolution in the reductive amination of α-branched aldehydes.
NHPMP
NHPMP
NHPMP
Me
Me
Me F
9a: 87%, 96 % ee
9b: 86%, 94% ee
9c: 89%, 94% ee
NHPMP
NHPMP
NHPMP
Et
Me
Me
9d: 92%, 98% ee
Figure 22.6
9e: 81%, 78% ee
9f: 77%, 80% ee
Scope of the asymmetric reductive amination of α-branched aldehydes.
22.3 Brønsted Acid Catalyzed Transfer Hydrogenation of Imines, Imino Esters, and Enamines O
H H
O
EtO
O R1
R
+ R2NH2
10
8
Scheme 22.6
OEt N H 2a (1.2 equiv)
10 mol% 6a, 5Å MS C6H6, 40-50 °C
HN R
R2 R1
3 25 examples
Organocatalytic asymmetric reductive amination of ketones.
PMP
HN
HN
PMP
HN
Me
Me
PMP Me
MeO 3c: 77%, 90% ee HN
PMP
3l: 87%, 94% ee HN
PMP
3n: 79%, 91% ee HN
PMP
Me O2N 3o: 71%, 95% ee
Figure 22.7
3p: 75%, 85% ee
3q: 71%, 83% ee
Scope of the asymmetric biomimetic reductive amination of ketones.
(40–80% ee). The absolute configuration of 9a was determined by conversion into the known carbobenzyloxy (Cbz)-protected amine. In addition, recent mechanistic studies have rationalized the stereochemical outcome of the reaction [25]. MacMillan reported a similar catalytic system for the reductive amination of ketones (Scheme 22.6) [22]. Use of 5 A˚ molecular sieves during the in situ imine generation was crucial to attain fair levels of conversion and selectivity. Under optimized conditions a large variety of aromatic ketones (10) with different electronic properties were treated with p-anisidine [8, R = p-MeO-C6 H4 ] in the presence of HEH 2a and catalyst 6a to give good yields (60–87%) and excellent selectivities (83–97% ee) (Figure 22.7). The protocol has also been extended to cyclic aryl and aliphatic ketones. Furthermore, diverse aromatic as well as heteroaromatic amines are tolerated under the same reaction conditions. Recently, Akiyama described the transfer hydrogenation of ketimines with a system based on BINOL-phosphoric acid derivatives as catalysts and benzothiazolines as alternative reducing agents [26]. You and Antilla independently reported the enantioselective reduction of α-imino esters yielding enantio-enriched α-amino acid derivatives [27, 28]. Whereas the system reported by You is based on catalysts derived from BINOL as backbone [27], Antilla reported a novel catalyst based on VAPOL as chiral backbone [28]. The best results were obtained when VAPOL derivative 13 was employed in the reduction of ethyl ester derivatives at 50 ◦ C in non-polar solvents (Scheme 22.7).
795
796
22 Bio-Inspired Transfer Hydrogenations O
H H
O
EtO
OEt N H
PMP
N
PMP
HN ∗ R COOR1
2a (1.4 equiv) 1
COOR
R 11
5 mol% 13 50 °C, Toluene
12
Ph Ph
O O P OH O
13
11 examples
Scheme 22.7
HN ∗
PMP
Organocatalytic asymmetric transfer hydrogenation of α-imino esters.
HN ∗
COOEt
PMP
HN ∗
COOEt
PMP COOEt
Cl a
12a: 93%, 96% ee
12b: 98%, 96% ee
HN ∗
12c: 95%, 98% ee
PMP COOEt
HN ∗
PMP COOEt
MeO 12d: 96%, 94% ee
12e: 88%, 99% eeb,c
a
Absolute configuration was assigned as R Imino ester was formed in situ (yield after two steps) c Absolute configuration was assigned as S b
Figure 22.8 Scope of the asymmetric biomimetic transfer hydrogenation of α-imino ethyl esters.
Under the optimized conditions, various α-imino ethyl esters (11) bearing aromatic groups with different substitution patterns and electronic properties were reduced to give products in high yields (93–98%) and excellent selectivities (94–98% ee) (Figure 22.8). The methodology was also extended to the reductive amination of alkyl substituted α-keto esters formed in situ from the appropriate keto esters and p-anisidine. Screening of different conditions (various catalysts, solvents, and HEH) led You and coworkers to develop an improved protocol that allows reactions with low catalyst loadings (1 mol.% ent-5c) in diethyl ether at room temperature and with the same HEH (2a) as hydrogen source [27]. The selectivity proved to be highly dependent on the residue of the ester group and superior enantioselectivities were obtained by increasing the steric bulk (R1 : Ot-Bu ≈ Oi-Pr > OBn > OEt >> OMe). A wide variety of α-imino esters were hydrogenated under these conditions to give products in high yields (78–95%) and selectivities (84–98% ee) (Figure 22.9). Only in one case was the product obtained in lower yield (R = cyclohexyl, R1 = i–Pr 46%, 88% ee). To show the usefulness of the method, a representative substrate was selected and the reaction was performed on gram scale. Product, 12f, was obtained in 85% yield and 96% ee. Surprisingly, the catalyst employed by MacMillan in the reductive amination of ketones was inactive under the conditions developed by You and Antilla. Moreover, when β, γ -alkynyl α-imino esters were subjected to similar conditions (Scheme 22.8), both alkyne and imine moieties were reduced to give the
22.3 Brønsted Acid Catalyzed Transfer Hydrogenation of Imines, Imino Esters, and Enamines
HN ∗
PMP
HN ∗
COOEt
PMP
HN ∗
COOi-Pr
PMP
HN ∗
COOi-Pr
PMP
797
HN ∗
COOi-Pr
PMP COOi-Pr
Br 12a: 88%, 92% ee
a
a
12f: 87%, 97% ee 85%, 96% eeb
12g: 89%, 98% ee
a
12h: 92%, 97% (>99%) eea,c,d
12i: 46%, 88% eea
a
Reactions were performed on a 0.2 mmol scale with 1 mol% catalyst Reaction was performed on a 4 mmol scale with 0.1 mol% catalyst c After recrystallization d Absolute configuration was assigned as S b
Figure 22.9
Scope of the asymmetric biomimetic transfer hydrogenation of α-imino esters.
O
H H
O
EtO
N
N H
PMP
HN ∗
2a (2.2 equiv)
COOR1 R
OEt
14
1 mol% ent-5c Et2O, rt
R
PMP COOR1
15 10 examples
Scheme 22.8 Organocatalytic asymmetric transfer hydrogenation of β, γ -alkynyl α-imino esters.
HN ∗
PMP COOEt
15a: 34%, 92% eea HN ∗
F
PMP
COOt-Bu
HN ∗
COOt-Bu
15b: 58%, 94% ee
PMP
15d: 64%, 95% ee a
HN ∗
HN ∗
PMP COOt-Bu
15c: 60%, 93% ee PMP COOt-Bu
15e: 27%, 83% ee
Absolute configuration was assigned as S
Figure 22.10 Scope of the asymmetric biomimetic transfer hydrogenation of β, γ -alkynyl α-imino esters.
corresponding β, γ -alkenyl α-amino esters 15 with very high selectivities, albeit in low yields (Figure 22.10) [29]. Nevertheless, the method is worth considering since the presence of a double bond in the products offers the possibility of further derivatization. With regard to the mechanism, control experiments have shown that reduction of the carbon–carbon triple bond is faster than that of the carbon–nitrogen double bond.
798
22 Bio-Inspired Transfer Hydrogenations O
H
H
O
EtO
NHAc R
OEt N H 2a (1.1 equiv)
NHAc
5 mol% ent-5c or 1 mol% ent-5c + 10 mol% AcOH 50 °C, Toluene
16
Scheme 22.9
R 17 11 examples
Organocatalytic asymmetric transfer hydrogenation of enamides.
More recently, Antilla reported the enantioselective reduction of aromatic enamides with chiral phosphoric acid catalysts [30]. The product 17a (R = Ph) was obtained in 92% yield and 91% ee when the reaction was carried out with 5 mol.% ent-5c as catalyst in toluene at 50 ◦ C (Scheme 22.9). Decreasing the catalyst loading had a detrimental effect on the yield and reaction time. It was assumed that the reaction takes place through an iminium ion intermediate whose formation is affected; therefore, a dual catalytic chiral–achiral acid system was considered. The role of the achiral acid co-catalyst was to enable iminium ion formation while being inactive in the asymmetric hydrogenation step. Accordingly, the reaction times were substantially reduced and the yields considerably increased when employing an achiral acid as co-catalyst, while the selectivities remained essentially unaffected. Various enamides 16 were reduced under the optimized reaction conditions to give products 17 in moderate to excellent yields (43–99%) and moderate to high enantioselectivities (41–92% ee) (Figure 22.11). Regarding the influence of the aryl moiety, lower selectivities were obtained with increasing steric bulk. With two exceptions (α-naphthyl and o-MeO-C6 H4 derivatives), the absolute configuration of the products has been assigned as (R). Although impressive results were obtained for aromatic enamides, the protocol proved unsuitable for aliphatic analogs. Regarding the mechanism of the imine reduction, it is assumed that the Brønsted acid catalyzed transfer hydrogenation proceeds similarly to the reaction of GDH (Scheme 22.10). In the first step the imine substrate is activated by the chiral Brønsted acid catalyst, which generates an iminium ion with formation of the chiral-ion pair A. NHAc
NHAc
NHAc
NHAc
Me
Me
Me
Me
NHAc MeO
Me
F A:
17a: 39%, 91% ee
17b: 53%, 91% ee
17c: 90%, 90% ee
17d: 91%, 94% ee
17e: 90%, 74% ee
B:
97%, 91% ee
93%, 90% ee
96%, 89% ee
99%, 92% ee
98%, 71% ee
A: 5 mol% ent-5c B: 1 mol% ent-5c + 10 mol% AcOH
Figure 22.11
Scope of the asymmetric biomimetic transfer hydrogenation of enamides.
22.3 Brønsted Acid Catalyzed Transfer Hydrogenation of Imines, Imino Esters, and Enamines
N R
1
R2 R1
ArO O P + R2 ArO O− H N A R R1
O H
O ArO P ArO OH 5
H O
R'O
799
H
OR'' N H 2
N
R
R2 H + R'O R1
O
O OR'' N
3
Scheme 22.10 Proposed mechanism of the organocatalyzed transfer hydrogenation of ketimines.
Subsequent hydride transfer from the HEH 2 yields the desired chiral amine 3 and regenerates the catalyst. The (R) absolute configuration of the amines depicted in Figure 22.4 was explained by a stereochemical model built on the basis of the X-ray crystal structure of the chiral Brønsted acid catalyst 5d. In the transition state the Brønsted acid catalyst activates the ketimine in such a way that the Re-face is effectively shielded, leaving only the Si-face free for the nucleophilic attack (left-hand side of Figure 22.12). Based on theoretical investigations Goodman and Himo independently proposed a bifunctional activation mechanism in which the imine and the dihydropyridine are simultaneously activated by the same catalyst molecule: whereas the ketimine is protonated by the acidic proton, the dihydropyridine is activated via a hydrogen bond formed from the Lewis basic oxygen of the phosphoryl group (right-hand side of Figure 22.12) [31]. The phosphoric acid catalyst thereby acts as a Brønsted acid/Lewis base bifunctional catalyst.
Ar
O O P O O H
Nu
−
Ar Ar
O O P O O H
N Ph
Nu−
Ph
Figure 22.12 BINOL-phosphoric catalyzed activation of ketimines – a simplified stereochemical model.
Ar H N
N Ph Ph
R
R H
H
800
22 Bio-Inspired Transfer Hydrogenations
22.4 Asymmetric Organocatalytic Reduction of N-Heterocycles 22.4.1 Asymmetric Organocatalytic Reduction of Quinolines
Hydrogenation of quinolines is another challenging field since it allows formation of the corresponding tetrahydroquinolines, which are common structural units in alkaloids and biologically active compounds. Given the importance of this class of molecule, considerable efforts have been made to develop improved methods for their synthesis. Although different approaches are available and well-established methods including homogeneous and heterogeneous metal-catalyzed hydrogenations, hydroborations, and transfer hydrogenations have been reported, these methods suffer from a limited substrate scope [32]. Thus, the development of efficient alternative protocols would be desirable. Given that Brønsted acid catalyzed enantioselective transfer hydrogenation proved general for the reduction of carbon–nitrogen double bonds in several acyclic systems, its application to the related cyclic analogs appeared feasible. The first reports on metal-free quinoline hydrogenation stem from Rueping [33, 34]. Since Brønsted acids were successful in catalyzing the asymmetric reduction of imines, it was expected that the reduction of substituted quinolines would proceed in a similar manner. In the first step it was envisaged that the Brønsted acid protonates the quinoline to form a chiral ion pair A and D, respectively (Scheme 22.11a,b). A 1,4-addition was anticipated, with formation of the enamines B and E, respectively. Subsequent isomerization should yield the corresponding (a) key step: enantioselective hydride transfer R2
R2 ∗
H− 1,4-hydride
O
N H
P O RO OR
R1
addition
O
P O RO OR
A
R2 ∗
protonation 1
N H
R1
N H
R
O
P O RO OR
B
H−
R2 ∗ 1,2-hydride addition
∗ N H
R1
C
(b) key step: enantioselective proton transfer H− R
O
P O RO OR
R
1,4-hydride addition
N H
O
D
P O RO OR
∗ R
Brønsted acid catalyzed protonation
N H
O E
addition
N H
P O RO OR
Scheme 22.11 Quinoline reduction – proposed mechanism as a function of quinoline substitution pattern: (a) enantioselective hydride transfer and (b) enantioselective protonation.
F
∗ R
1,2-hydride
N H
22.4 Asymmetric Organocatalytic Reduction of N-Heterocycles
iminium ions C and F, respectively, and a second hydride transfer would give the desired tetrahydroquinolines. Obviously, the stereo-determining step of this hydrogenation cascade differs according to the quinoline substitution pattern. In the case of 2- and 4-substituted quinolines, enantioselectivity will be induced in one of the hydride transfer steps (Scheme 22.11a). In the case of 3-substituted quinolines the enantio-determining step has to be the asymmetric Brønsted acid catalyzed protonation (Scheme 22.11b). Initial research focused on the development of an efficient achiral version with identification of suitable reaction conditions for the reduction of 2-methylquinoline as model substrate. Different Brønsted acids were capable of catalyzing the reaction; DPP gave optimum yield and reaction rate (Scheme 22.12) [33]. Overall, the best results were obtained when 2-methylquinoline (18a) was treated with 2.4 equiv. of dihydropyridine 2a in the presence of 1 mol.% DPP at 60 ◦ C in benzene for 12 h. Under these conditions a large variety of alkyl, aryl, and heteroaryl 2-substituted quinolines gave products in good to high yields (75–95%). Furthermore, 3- and 4-substituted quinolines as well as 2,3- and 2,4-disubstituted quinolines were reduced under the same conditions to give products in high yields (82–94%). The excellent results obtained in the achiral version were the starting point for the development of an asymmetric version (Scheme 22.13) [34]. Screening of different BINOL based chiral phosphoric acids in the reduction of 2-phenylquinoline 18b (R = Ph) revealed that large substituents in the 3,3 positions of the backbone are essential to attain high levels of enantioselectivity. Furthermore, comparable results were obtained in chlorinated as well as aromatic solvents. O H H O OEt
EtO N H 2a (2.4 equiv)
N
1 mol% DPP 60 °C, C6H6
N H 19a: 85%
18a
Scheme 22.12
First organocatalyzed reduction of quinolines.
O H H O EtO
OEt N H 2a (2.4 equiv)
N
R
18
Scheme 22.13
1-5 mol% 5b 60 °C, C6H6
N H
R
19 16 examples
Brønsted acid catalyzed transfer hydrogenation of 2-substituted quinolines.
801
802
22 Bio-Inspired Transfer Hydrogenations Me
F
N H
N H
N H
N H
N H CF3
Me 19b: 92%, 97% ee 19c: 93%, 98% ee
O
N H
19d: 65%, 97% ee
N H
19e: 93%, 99% ee
Cl
N H
19f: 91%, 99% ee
N H
OMe 19g: 93%, 91% ee
Figure 22.13
19i: 91%, 88% ee
19h: 90%, 98% ee
19j: 91%, 87% ee
Scope of the transfer hydrogenation of 2-substituted quinolines 18.
With optimum conditions in hand, the scope of the reaction was explored and the results are summarized in Figure 22.13. Notably, a wide range of 2-substituted quinolines (18) bearing aliphatic, aromatic as well as heteroaromatic residues readily reacts with dihydropyridine 2a, and the corresponding tetrahydroquinolines 19 can be isolated in moderate to good yields (54–93%) and excellent enantioselectivities (87–99% ee). The usefulness of this highly enantioselective transfer hydrogenation process was demonstrated in a two-step synthesis of biologically active natural products with a tetrahydroquinoline core, namely, galipinine, cuspareine, and angustureine [35]. For this purpose, Brønsted acid catalyzed hydrogenation of the appropriate 2-substituted quinolines, which were available by simple alkylation of 2-methylquinoline, generated the tetrahydroquinoline derivatives with excellent enantioselectivities. Subsequent N-methylation afforded the desired natural products in good overall yields (Scheme 22.14).
2a, 1 mol% 5b
N
R
88-95%
18
N Me (+)-cuspareine 88%, 90% ee
1. CH2O, AcOH 2. NaBH4
N H
R
90-95%
19
OMe OMe
20
N Me
O O
(+)-galipinine 89%, 91% ee
Scheme 22.14 Application of the enantioselective Brønsted acid catalyzed transfer hydrogenation of 2-substituted quinolines in the synthesis of alkaloids.
N Me
R
N Me (−)-angustureine 79%, 90% ee
22.4 Asymmetric Organocatalytic Reduction of N-Heterocycles
Additionally, Metallinos reported a Brønsted acid catalyzed enantioselective reduction of the structurally related di-substituted 1,10-phenanthrolines [36]. Du reported a similar catalytic system, which made use of double axially chiral phosphoric acids 21 in the reduction of 2-substituted quinolines to give the corresponding tetrahydroquinolines with comparable enantioselectivities [37]. For 2-phenylquinoline, aromatic solvents and diethyl ether gave similar results, but for the n-butyl analog diethyl ether proved superior in terms of selectivity. Notably, lowering the catalyst loading did not affect the selectivity, and the reactions could be performed with HEH 2c at 35 ◦ C in diethyl ether with only 0.2 mol.% catalyst (Scheme 22.15). Compared to Rueping’s system, a slight increase in the enantiomeric excess has been reported when alkyl substituted quinolines were reduced under these conditions. 2,3-Disubstituted quinolines have also been tested and products obtained with high levels of diastereo- and enantiocontrol (Figure 22.14). Along the same lines, Rueping and coworkers also investigated the reduction of 4-substituted derivatives 23. To date, no direct approach toward the optically active 4-substituted tetrahydroquinolines 24 has been described. Multistep sequences have been employed to synthesize these valuable chiral 4-substituted tetrahydroquinolines [38], which exhibit biological activity in vivo [39]. Therefore, the development of a catalytic asymmetric route would be a substantial improvement (Scheme 22.16). O
H
H
O
t-BuO
Ot-Bu N H 2c (2.4 equiv)
N
R
O O N H
0.2 mol% 21 35 °C, Et2O
18
R
OR O
P
OH OR
19 18 examples 21: R = cyclohexyl
Scheme 22.15 Double axially chiral phosphoric acids in the transfer hydrogenation of 2-substituted quinolines.
N H
N H 22b
22a >99%, cis/trans = >20:1 cis = 82% ee
Figure 22.14
93%, trans/cis = 94:6 trans = 91% ee
N H 22c >99%, trans/cis = >20:1 trans = 92% ee
Substrate scope of the transfer hydrogenation of 2,3-substituted quinolines.
803
804
22 Bio-Inspired Transfer Hydrogenations O
O
H H
RO
OR N H 2
R2
R2
2a: R = Et, 2c: R = t-Bu
R1
R1
5 mol% 6b
N
50 °C, C6H6
23
Scheme 22.16
N H 24 17 examples
Brønsted acid catalyzed transfer hydrogenation of 4-substituted quinolines.
Evaluation of different catalysts showed a strong correlation between the steric bulk of the substituents on the chiral backbone of the catalyst and the selectivity of the reaction, whereby larger residues lead to superior enantioselectivities [40]. This result is explained by the position of the stereocenter which is now further away from both the protonated nitrogen and the catalytic center of the phosphoric acid catalyst. After extensive optimization of the reaction parameters, the best results, with respect to both reactivity and selectivity, have been obtained with a combination of catalyst 6b and t-butyl HEH 2c. Figure 22.15 depicts selected examples, demonstrating that several 4-substituted tetrahydroquinolines can be prepared in good yields and with good to excellent enantioselectivities. To obtain a complete picture of the transfer hydrogenation of quinolines, the reduction of 3-substituted quinolines 25 was also addressed [41]. This reaction represents the first direct access to optically active 3-substituted tetrahydroquinolines 26 (Scheme 22.17), and proceeds via an enantioselective Brønsted acid
O
N H
Cl
N H
24a: 96%, 92% ee
Cl
24b: 81%, 90% ee
OMe
Ph
Cl
N H
24c: 67%, 75% ee
N H
24d: 96%, 84% ee
Figure 22.15 Scope of the organocatalyzed transfer hydrogenation of 4-substituted quinolines. O
H H
O
O
R N 25
Scheme 22.17
O N H 2d (1.4 equiv) 5 mol% 6b 60 °C, C6H6
R N H
26 10 examples
Metal-free asymmetric reduction of 3-substituted quinolines.
22.4 Asymmetric Organocatalytic Reduction of N-Heterocycles
OMe N H
N H
26a: 76%, 84% ee
26b: 67%, 85% ee
26d: 50%, 86% ee
26c: 77%, 85% ee Ph
S
N H
F
N H
N H 26e: 58%, 82% ee
N H
22c
87%, cis/trans = 1:8 22c: cis = 99% ee 22c: trans = 94% ee
Figure 22.16 Scope of the Brønsted acid catalyzed asymmetric protonation of 3-substituted quinolines.
catalyzed protonation (Scheme 22.11b). As in the case of related 4-substituted quinolines, application of Brønsted acid catalyst 6b gave the best chiral induction. Among the various HEH 2 tested (R = R = Et, i-Pr, t-Bu, Bn, allyl), the allyl ester 2d proved to be a superior hydride source, resulting in slightly better enantioselectivities. Various 3-substituted quinolines bearing aryl and heteroaryl residues with different electronic properties could be reduced under the optimized reaction conditions to provide the corresponding tetrahydroquinolines in good yields and with high enantioselectivities (Figure 22.16). As an extension of this protocol, the octahydroacridine 22c was obtained by reduction of the appropriate 2,3-disubstituted quinoline substrate. In contrast to the above asymmetric transfer hydrogenations of 2- and 4-substituted quinolines, which proceed through an enantioselective hydride transfer as a key step, this latest cascade involves an enantioselective Brønsted acid catalyzed proton transfer as the enantiodifferentiating step.
22.4.2 Asymmetric Brønsted Acid Catalyzed Hydrogenation of Indoles
Indolines 28 represent a common structural feature of many natural alkaloids that possess widespread biological and pharmaceutical activity. To date only the metal-catalyzed reduction of trimethyl-3H-indole has been achieved with high levels of enantiocontrol [42]. Given the structural similarities between indoles and quinolines, the application of a Brønsted acid catalyzed transfer hydrogenation to the reduction of indoles 27 seemed reasonable (Scheme 22.18). Screening of different catalysts revealed the same Brønsted acid 5c to be highly effective in the reduction of indoles. The reaction proved to be general and the products were obtained in high yields (54–99%) and with excellent enantioselectivities
805
806
22 Bio-Inspired Transfer Hydrogenations O
H H
O
EtO
R3
R
2
R2 R1
N 27
Scheme 22.18
OEt N H 2a (1.4 equiv) 1 mol% 5c Toluene, rt
R3
R2 R2 R1 N H
28 15 examples
Metal-free asymmetric transfer hydrogenation of indoles.
(70–99% ee) (Figure 22.17) [43]. In addition, lowering the catalyst loading from 1 to 0.1 mol.% resulted in a lower conversion without any significant detrimental influence on the selectivity. 22.4.3 Asymmetric Brønsted Acid Catalyzed Hydrogenation of Benzoxazines, Benzothiazines, Benzoxazinones, Quinoxalines, Quinoxalinones, Diazepines, and Benzodiazepinones
Heterocyclic compounds like dihydro-2H-benzoxazines 29, dihydro-2H-benzothiazines 30, dihydro-2H-quinoxalines 31 as well as dihydro-2H-benzoxazinones 32, dihydro-2H-quinoxalinones 33, and dihydro-2H-benzodiazepinones 34 attract considerable attention since they are common structural motifs in a large number of natural products and possess interesting biological activities (Figure 22.18). Moreover, they have often been employed as chiral building blocks in the synthesis of pharmaceuticals, for example, as promising anti-depressants, calcium antagonists, as well as anti-inflammatory, anti-nociceptive, anti-bacterial, anti-microbial agents, and as potential non-nucleoside HIV-1 reverse transcriptase inhibitors [44]. The catalytic enantioselective reduction of imine-containing heterocycles represents a direct and efficient approach to these important classes of compounds. Despite their utility, only a few catalytic systems have been reported for the reduction of these cyclic imines and they are restricted to alkyl-substituted derivates, particularly methyl- and ethyl-substituted ones [45]. Considering the success of the organocatalytic enantioselective transfer hydrogenation of imines [11a-c], quinolines [33, 34], and indoles [43], it became apparent that the bio-inspired strategy could be applicable to the transfer hydrogenation of the whole set of imine containing heterocycles (Figure 22.18). Analogous to the previously reported procedure, it was anticipated that the chiral Brønsted acid would activate the substrates through catalytic protonation, thus enabling hydride transfer from the dihydropyridine to occur (Scheme 22.19). In the case of quinoxalines 37, a double 1,2-hydride addition has to take place to produce the desired dihydro-2H-quinoxalines 31. Subsequent reaction optimization showed that the highest enantioselectivities were obtained when catalyst 5b was applied in the reduction of benzoxazines, benzothiazines, and benzoxazinones [46]. Furthermore, detailed investigation of the catalyst’s activity allowed a decrease in the catalyst loading from 10 to 0.01 mol.%
Me N H
MeO N H
N H
N H
N H
OEt
O
28e: 93%, >99% ee
Substrate scope of the Brønsted acid catalyzed transfer hydrogenation of 2,3,3-substituted indoles.
28i: 87%, 90% ee
28d: 86%, 99% ee
N H
28h: 92%, 96% ee
28c: 92%, 97% ee
28g: 93%, 98% ee
Reaction performed with 0.1 mol% catalyst
F
28b: 84%, >99% ee
N H
28f: 86%, 75% ee
Figure 22.17
a
N H
28a: 99%, 97% ee 81%, 96% eea
N H
OMe
CF3
22.4 Asymmetric Organocatalytic Reduction of N-Heterocycles 807
808
22 Bio-Inspired Transfer Hydrogenations H N
S
O 1
3
R
R N H 29
R
R2
5
R
R4
N H
N H
30
Figure 22.18
O 7
O
9
R
R6
N H
31
H N
O
R
R8
N H
32
33
H N
O
N H
R12
11
R10 34
Challenging amine-containing heterocyclic classes.
(substrate : catalyst ratio = 10 000 : 1). Remarkably, the reduction of phenylbenzoxazine 35a with only 0.01 mol.% of catalyst proceeded without much loss in reactivity and selectivity, and the corresponding 2H-dihydro-benzoxazine 29a could be isolated in 90% yield and 93% ee. To date this is the lowest catalyst loading reported for an organocatalytic enantioselective transformation, corresponding to a turnover number (TON) of 9000 and a turnover frequency (TOF) of 500 h−1 (Table 22.1). Best results in terms of yield and selectivity were found in chloroform at room temperature. Accordingly, a large variety of benzoxazines were reduced at room temperature in chloroform with 0.1 mol.% catalyst to give products in high yields and excellent enantioselectivities (Figure 22.19a, 29a–f). The excellent results obtained in this metal-free transformation along with the remarkable low catalyst loadings render this approach as a competitive alternative process to enantioselective metal-catalyzed reductions. X R' N
R
35 X: O 36 X: S 37 X: N
R' Ar
Ar
O O P O OH
− O O H N+ P O O
Ar
R Ar chiral ion pair A
5
HEH 2
X R'
X
* N H
R 29 X: O 30 X: S 31 X: NH
Scheme 22.19 Mechanism of the asymmetric reduction of various imine-containing heterocycles.
22.4 Asymmetric Organocatalytic Reduction of N-Heterocycles Table 22.1
Influence of catalyst loading on the reduction of benzoxazine 35a. O HH O EtO
O N 35a Entry 1 2 3 4 5 6
Ph
OEt N H 2a (1.25 equiv)
O
Brønsted acid 5b 60 °C, C6H6
N H 29a
Ph
Loading of 5b (%)
Yield (%)
ee (%)
10 5 2 1 0.1 0.01
91 95 93 94 95 90
96 96 96 96 96 93
Under similar conditions (1.25 equiv. 2a, 1 mol.% 5b, chloroform, room temperature) various 2-aryl-substituted benzothiazines and benzoxazinones were reduced to give products in moderate to high yields and excellent enantioselectivities (Figure 22.19b and d, 30a–f, 32a–f). The reduction of benzoxazinones leading to the corresponding cyclic substituted amino acid derivatives 32a–f is of high synthetic value since the products can easily be converted into the optically active open-chain amino acids. For example, treatment of dihydrobenzoxazinone 32c with benzylamine in the presence of pyridin-2-ol yielded the glycine derivative 38 without loss of enantiomeric excess (Scheme 22.20). Regarding the reduction of quinoxalines, superior results were obtained when the reactions were conducted with 2.4 equiv. HEH 2a and 10 mol.% of the Brønsted acid 5c at 35 ◦ C in chloroform [47]. Remarkably, a broad range of substrates, with different substitution patterns on both aryl and tetrahydroquinoxaline cores, is tolerated in the reaction (Figure 22.19c, 31a–f). Owing to the lower solubility of quinoxalinones, a slightly modified procedure had to be applied. Products were obtained with excellent selectivities despite moderate yields (Figure 22.19e, 33a–e). Unfortunately, the reduction of the benzodiazepinones 39 proved to be problematic and proceeded with low conversions under standard conditions (HEH 2d, phosphoric acid diester catalysts, toluene, 50 ◦ C) [48]. However, this could be circumvented by employing BINOL-based N-triflylphosphoramides which, due to the strongly electron-withdrawing triflyl-group, possesses a higher reactivity (Scheme 22.21) [49a]. So far these catalysts have been used to promote asymmetric Diels–Alder reactions, dipolar cycloadditions, ene reactions, and Nazarov cyclizations in a highly efficient manner [49].
809
(e)
(d)
O
Br
O
33a: 42%, 98% ee
N H
H N
32a: 92%, >99% ee
N H
O
(c) 31a: 88%, 96% ee
O
O
O
33b: 75%, 98% ee
N H
H N
32b: 91%, >99% ee
N H
O
31b: 73%, 94% ee
Br
O
N H
S Br
Cl
O
N H
H N Cl
Cl
O
N H
O
N H
S
O
O
N H
O
Me
O
S
N H
H N
O S
32f: 81%, 90% ee
O
33e: 23%, 90% ee
Br
N H
O
31f: 93%, 82% ee
N H
H N
30f: 54%, 93% ee
N H
S
29f: 95%, >99% ee
32e: 90%, 98% ee
N H
O
33d: 60%, 98% ee
N H
O
31e: 93%, 98% ee
N H
H N
F
29e: 92%, 98% ee
H N
32d: 55%, 96% ee
N H
O
31d: 95%, 92% ee
33c: 56%, 92% ee
N H
H N
Br
29d: 93%, 98% ee
N H
O
30d: 51%, 94% ee 30e: 70%, >99% ee
32c: 90%, >99% ee
N H
O
Cl Cl 31c: 98%, 96% ee
N H
N H
N H OMe
H N
H N
Br
Ph
Ph 30c: 78%, 94% ee
H N
30b: 50%, 96% ee
N H
N H
N H Me
S
S
Br
N H
O
29c: 94%, 98% ee
S
29a: 95%, 98% ee
(b) 30a: 87%, >99% ee
(a)
Br 29b: 93%, >99% ee
N H
N H
Cl
O
O
810
22 Bio-Inspired Transfer Hydrogenations
22.4 Asymmetric Organocatalytic Reduction of N-Heterocycles Figure 22.19 Substrate scope of the transfer hydrogenation of (a) benzoxazines, (b) benzothiazines, (c) quinoxalines, (d) benzoxazinones, and (e) quinoxalinones.
←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−
O
O
OH H N
Pyridin-2-ol (3 equiv)
O N H
BnNH2 (5 equiv) THF, rt
N H
32c
Bn
38
> 99% ee
Scheme 22.20
H N
90%, 99% ee
Synthesis of the glycine derivative 38.
O
R N 39
1. 2d, 5 mol% 41 MTBE, 50 °C, MW
H N
O O P O N Tf H
R
2. DMF, AcCl, Py
Ar
O
N
Ar
O 40 15 examples
41
Scheme 22.21 N-Triflylphosphoramide-catalyzed asymmetric transfer hydrogenation of benzodiazepinones.
Application of 5 mol.% of the N-triflylphosphoramide 41 allowed full conversion for a limited number of substrates. A further improvement has been achieved by performing the reactions at 50 ◦ C in methyl t-butyl ether (MTBE) under microwave irradiation. Accordingly, a broad range of benzodiazepinones 39 were reduced and subsequently acetylated to give protected dihydro-benzodiazepinones 40a–f in moderate to excellent yields (51–95%) and with excellent enantioselectivities (83–99% ee) (Figure 22.20). Chiral dihydro-2H-benzodiazepines have been obtained from racemic benzodiazepines via a dynamic kinetic asymmetric transfer hydrogenation process [50]. Treatment of racemic 42 with HEH 2d in the presence of 5g afforded products with moderate level of diastereo- and good levels of enantiocontrol (Scheme 22.22). Interestingly, the minor diastereomer has been obtained with slightly better enantioselectivity as compared to the major one. The absolute stereochemistry of the products has been assigned by X-ray crystal structure analysis. The result has been rationalized by the pathway depicted in Scheme 22.23. It was proposed that whereas the (S) component of the initial racemic mixture undergoes fast reduction reaction (i) the (R) component participates in two distinct processes: a slow transfer hydrogenation reaction (ii) with concomitant racemization of the substrate via a sequence involving retro-Mannich and Mannich reactions (iii). The racemization was confirmed by a control experiment that showed that the starting material recovered after the reaction was racemic.
811
812
22 Bio-Inspired Transfer Hydrogenations
H N Cl
H N
O Cl
N Ac
H N
O Cl
N Ac F
O
N Ac NO2
40a: 93%, 95% ee H N
Cl Cl
O
40b: 83%, 95% ee H N
Cl Cl
N Ac
O
40c: 87%, 99% ee H N
Br
N Ac
N Ac
NO2
O
40d: 94%, 99% ee
O
O
NO2
40e: 52%, 97% ee
40f: 53%, 95% ee
Figure 22.20 Scope of the N-triflylphosphoramide catalyzed reduction of benzodiazepinones.
H N N
Ph
H N
2d (1.3 equiv) 10 mol% 5g
H N
Ph
+
CHCl3,−10 ˚C, Na2SO4
N H
Ph
42
Ph
N H
Ph
43a
Ph
43a'
Scheme 22.22 Dynamic kinetic resolution in the transfer hydrogenation of benzodiazepines.
H Ph N N
Ph N N H
S -42 Mannich
+
Ph
44 N
R - 42
+
(i)
H Ph N
Ph
H Ph N
H Ph N fast reaction
43a
N H
Ph
43a′
(ii)
Ph
Retro-Mannich (iii)
Scheme 22.23 Proposed mechanism for the dynamic kinetic transfer hydrogenation of benzodiazepines.
Ph
H Ph N
H Ph N slow reaction
N H
+ N H Ph ent -43a major diastereomer
N Ph H ent - 43a' minor diastereomer
22.4 Asymmetric Organocatalytic Reduction of N-Heterocycles
O R
H N
S
1
3
R R2
N H
HEH: Cat:
R4
R6
N H
R8
N H
H N
O
N H
R12
11
R
R
R
R N H
O
9
7
5
H N
O
O
813
N H
R10
29
30
31
32
33
34
2a 5b
2a 5b
2a 5c/ 5e
2a 5b
2a 5c/ 5e
2d 41
Figure 22.21 Reaction parameters for the metal-free asymmetric reduction of various imines.
Evidently, careful selection of the catalyst and reaction conditions offers access to different classes of N-heterocycles in an optically active form. Typically, enhanced reactivities and high levels of selectivities are achieved with the Hantzsch ethyl ester in combination with phenanthryl, anthracenyl, or triisopropylphenyl-substituted BINOL-phosphoric acid catalysts (Figure 22.21). Throughout, the reductions proceed smoothly under mild conditions, providing differently alkyl-, aryl-, and heteroaryl-substituted products in good yields and with excellent enantioselectivities (Figures 22.19 and 22.20). 22.4.4 Asymmetric Organocatalytic Reduction of Pyridines
Piperidine alkaloids and derivatives belong to another important class of N-heterocycles. Piperidine is the structural core of many natural products with relevant biological and pharmaceutical properties. From a synthetic point of view, the catalytic asymmetric hydrogenation of pyridines represents the most convenient and efficient access to these compounds. However, the enantioselective reduction of substituted pyridines is a great challenge and only a few metal-catalyzed reductions are known [51]. In this context, the highly enantioselective organocatalytic protocol developed recently by Rueping constitutes a major breakthrough [52]. Screening of different Brønsted acids allowed identification of an effective catalyst for the reduction of pyridine derivatives 45 and 47 (Scheme 22.24).
O
O
NC
2a (4 equiv)
N 45 (a)
R1
5 mol% 5c 50 °C, C6H6
N H
R1
Me
46 6 examples
2a (4 equiv)
N 47
(b)
Scheme 22.24 Brønsted acid transfer hydrogenation of pyridines to the corresponding piperidines.
R2
5 mol% 5c 50 °C, C6H6
NC Me
N H 48
4 examples
R2
22 Bio-Inspired Transfer Hydrogenations
814
O
O
O
(a) 46a: 84%, 91% ee
46b: 72%, 91% ee
N H
N H
46d: 66%, 92% ee
NC N H
48a: 73%, 90% ee
N H
46c: 69%, 89% ee
NC
NC
(b)
N H
N H
N H
O
O
46e: 83%, 87% ee
NC N H
48b: 55%, 84% ee
N H
48c: 47%, 86% ee
48d: 68%, 89% ee
Figure 22.22 Substrate scope of the transfer hydrogenation of pyridine-derivatives (a) 45 and (b) 47.
A broad range of different azadecalinones 46 as well as tetrahydropyridines 48 were synthesized in good yields (47–84%) and with high enantioselectivities (84–92% ee) by employing 5 mol.% of Brønsted acid 5c as catalyst and dihydropyridine 2a as the hydrogen source (Figure 22.22). Furthermore, transfer hydrogenation of pyridine 45c with ent-5c as catalyst lead to the corresponding hexahydroquinolinone ent-46c, a valuable intermediate in the synthesis of the alkaloid diepi-pumiliotoxin C (Scheme 22.25). O
O + NH2
O ent - 5c
(i) O
(ii)
Me H
HEH 2a
Me
N 45c
N H ent -46c
N H H diepi -Pumiliotoxin C
Scheme 22.25 Transfer hydrogenation as key step in the synthesis of diepi-pumiliotoxin C: (i) EtOH, 50 ◦ C, 12 h, then 140 ◦ C, 2 h [53]; (ii) Reference [54e].
This new Brønsted acid catalyzed cascade reduction of pyridines gives the corresponding products in good yields and with excellent enantioselectivities. It is particularly noteworthy since it provides a simple and straightforward route to decahydroquinoline or piperidine alkaloid natural products [54, 55].
22.5 Asymmetric Organocatalytic Reductions in Cascade Sequences
As illustrated above, several efficient protocols that are able to mimic Nature’s reduction by simply replacing the dehydrogenase and NADH system with a combination of a readily available Brønsted acid and a dihydropyridine were developed recently. In addition to promoting the reactions with high levels of stereocontrol, enzymes can also build up sophisticated structures starting from simple compounds. In Nature, the multistep sequences are often realized in domino- and multicomponent reactions, and are, thus, blueprints for organic synthesis [56].
22.5 Asymmetric Organocatalytic Reductions in Cascade Sequences
As part of their ongoing studies on chiral Brønsted acid catalysis, Rueping and coworkers have developed recently an asymmetric organocatalytic cascade reaction in which multiple steps are catalyzed by the same chiral Brønsted acid catalyst. This provides valuable tetrahydropyridines 48 and azadecalinones 46 with high enantioselectivities [57]. Taking advantage of their experience in the field of chiral ion pair catalysis and based on the initial biomimetic strategy, they envisioned a new organocatalytic multiple reaction cascade consisting of a one pot Michael addition–isomerization–cyclization–elimination–isomerization–transfer hydrogenation sequence in which every single step is catalyzed by the same chiral Brønsted acid (Scheme 22.26). It was assumed that exposure of a mixture of enamine 49 and enone 50 to catalytic amounts of Brønsted acid would afford the corresponding 1,4-addition products A and B (Scheme 22.27). Subsequent Brønsted acid catalyzed cyclization R3
O
+ R2
R1
NH2 49
R3
2a (1.1 equiv) 5 mol% 5c, 50 °C C6H6 or CHCl3
R2
50
Scheme 22.26
Brønsted acid catalyzed cascade reaction.
R3
R3
O
R2
+
NH2
5c
R1
49
R2
N R1 H 46 or 48
HEH 2a
50 H+
Michaeladdition
R1
R1 O
R3
O H+
R3
Transfer hydrogenation
A
B H 2N
R1 N H 46/48
R2
R2
NH2
H+ HEH 2a
Isomerization H+
Cyclization
H+
R3 R2 C
N H
OH R1
H2O
Elimination
Scheme 22.27
H+
R3 R2 D
N H
R1
R3 R2
R1 N − E H B*
Isomerization/ Protonation
Mechanism of the Brønsted acid catalyzed multistep reaction sequence.
815
816
22 Bio-Inspired Transfer Hydrogenations
of A, which under the reaction conditions is in equilibrium with B, would give the hemiaminal C, which is expected to readily eliminate water and form the dihydropyridine (D), an intermediate observed also in the asymmetric pyridine reduction. In the next step, Brønsted acid catalyzed protonation should generate an iminium ion, a chiral ion pair E enabling the final step, namely, enantioselective hydride transfer to give the desired product 46 or 48, to take place. To promote this multiple reaction cascade as a powerful strategy for the asymmetric synthesis, it is necessary to obtain a high level of stereocontrol in the last step of the sequence. Remarkably, the transformation could be accomplished by applying the ideal combination of Brønsted acid 5c and dihydropyridine 2a [57a]. Notably, the same chiral Brønsted acid catalyzes all six steps in this new three component reaction, allowing rapid, direct, and efficient access to valuable tetrahydropyridines and azadecalinones with excellent levels of enantiocontrol (89–99% ee) from simple readily available starting materials (Figure 22.23). This proves that the organocatalytic hydrogenation protocol is amenable to complex molecular cascading. Another concept making use of both amine and Brønsted acid catalysis was developed by List and coworkers [58]. Starting from linear diketones 51 and achiral amines 52, various substituted cyclohexylamines 53 were obtained with high diastereo- and enantioselectivities (2 : 1 to 99 : 1 dr, 82–96% ee) (Figure 22.24). The reaction proceeds via an amine-catalyzed intramolecular aldol condensation O
O
O
O
N H
N H
N H
OMe 46a: 66%, 89% ee
46f: 74%, 97% ee
N H
Br 46g: 73%, 99% ee
46h: 78%, 99% ee O
O
O NC
NC N H
S
N H
Me
N H
48e: 89%, 96% ee
Figure 22.23
HN
PEP
n -Bu 53a: 75% 90% ee, 10:1dr
Figure 22.24
F
MeO
OMe
46i: 47%, 92% ee
CF3
Br 48f: 77%, 97% ee
N H
N H
CF3
48g: 55%, 99% ee
F 48h: 52%, 97% ee
Scope of the Brønsted acid catalyzed cascade reaction.
HN
PMP
i-Pr 53b: 76% 92% ee, 3:1dr
HN
PEP
Naphth 53c: 73% 82% ee, 2:1dr
HN O
PEP
Me
53d: 72% 92% ee, 99:1dr
Scope of the Brønsted acid catalyzed cascade reaction to substituted amines.
22.6 Conclusion O
NH2 +
O X
HN HEH 2a
R 52
51
PhOR'
5e X
R 53
OR'
X = CH2, O, S
2a
HN
PhOR'
−B* H + PhOR' N
O X
R
X
R
−B*
+ H N PhOR'
2a
X
R
Scheme 22.28 Synthesis of amine derivatives via a Brønsted acid catalyzed cascade sequence.
and subsequent Brønsted acid catalyzed conjugate reduction–imine reduction sequence (Scheme 22.28). Recent mass spectrometry studies in which all crucial intermediates were detected strongly support the proposed catalytic cycle [59]. Additionally, this study showed that electrospray ionization mass spectroscopy (ESI-MS) is a suitable technique for the determination of reaction pathways in Brønsted acid catalysis and is likely to be used in future mechanistic investigations.
22.6 Conclusion
As illustrated in this chapter, the bio-inspired transfer hydrogenation developed by Rueping represents the key starting point for the development of many powerful reductions, and has led to the rapidly growing field of organocatalyzed transfer hydrogenation. The generality of the BINOL-phosphate/dihydropyridine combination, when compared to Nature’s dehydrogenase/NADH system, renders this newly developed method as particularly noteworthy and allows a broad range of diverse cyclic and acyclic amines as well as heterocycles to be obtained with impressive levels of enantioselectivity. Notably, with these new protocols even the synthesis of challenging substrates can be successfully addressed. The mild reaction conditions and the operational simplicity and practicability render these methods as essential tools in the chemist’s toolbox. Moreover, analogous to Nature’s multicomponent domino reactions, the catalytic asymmetric Brønsted acid catalyzed transfer hydrogenation is effective in multistep reaction sequences, showing that this protocol can be used in complex molecular cascades. Although high to exceptional levels of selectivity and reactivity are obtained for a broad range of substrates, there is still room for improvement. It is desirable that the hydride source dihydropyridine employed in these transformations can be recycled or used in catalytic amounts in a similar way to Nature’s NADH. We are confident that this metal-free transfer
817
818
22 Bio-Inspired Transfer Hydrogenations
hydrogenation procedure will find widespread application in organic synthesis and will, at the same time, inspire the chemistry community to design more powerful organocatalytic systems.
References 1. (a) Ojima, I. (2000) Catalytic Asymmetric
2.
3.
4.
5. 6.
Synthesis, 2nd edn, Wiley-VCH Verlag GmbH, Weinheim; (b) Jacobsen, E.N., Pfaltz, A., and Yamamoto, H. (eds) (2000) Comprehensive Asymmetric Catalysis, vols 1–3, 2nd edn, Springer, Berlin; (c) Asymmetric catalysis special feature issues: (2004) Proc. Natl. Acad. Sci. U.S.A., 101(15–16); (d) Blaser, H.U. and Schmidt, E. (eds) (2004) Asymmetric Catalysis on Industrial Scale, Wiley-VCH Verlag GmbH, Weinheim; (e) Beller, M. and Bolm, C. (eds) (2004) Transition Metals for Organic Synthesis Building Blocks and Fine Chemicals, vols 1–2, 2nd edn, Wiley-VCH Verlag GmbH, Weinheim; (f) Enders, D. and Jaeger, K.-E. (eds) (2007) Asymmetric Synthesis with Chemical and Biological Methods, Wiley-VCH Verlag GmbH, Weinheim; (g) Mikami, K. and Lautens, M. (eds) (2007) New Frontiers in Asymmetric Catalysis, John Wiley & Sons, Inc., Hoboken, NJ. (a) Andersson, P.G. and Munslow, I.J. (eds) (2008) Modern Reduction Methods, Wiley-VCH Verlag GmbH, Weinheim; (b) de Vries, J.G. and Elsevier, C.J. (eds) (2007) The Handbook of Homogeneous Hydrogenation, Wiley-VCH Verlag GmbH, Weinheim; (c) Tang, W. and Zhang, X. (2003) Chem. Rev., 103, 3029–3070. Stillman, T.J., Baker, P.J., Britton, K.L., and Rice, D.W. (1990) J. Mol. Biol., 234, 1131–1139. Dean, J.L.E., Wang, X.G., Teller, J.K., Waugh, M.L., Britton, K.L., Baker, P.J., Stillman, T.J., Martin, S.R., Rice, D.W., and Engel, P.C. (1994) Biochem. J., 301, 13–16. Hantzsch, A. (1882) Justus Liebigs Ann. Chem., 215, 1–82. Mauzerall, D. and Westheimer, F.H. (1955) J. Am. Chem. Soc., 77, 2261–2264.
7. (a) Steevens, J.B. and Pandit, U.K.
(1983) Tetrahedron, 39, 1395–1400; (b) Nakamura, K., Fujii, M., Ohno, A., and Oka, S. (1984) Tetrahedron Lett., 25, 3983–3986; (c) Vanniel, J.C.G. and Pandit, U.K. (1985) Tetrahedron, 41, 6005–6011; (d) Watanabe, M., Fushimi, M., Baba, N., Oda, J., and Inouye, Y. (1985) Agric. Biol. Chem., 49, 3533–3538; (e) Vanniel, J.C.G., Kort, C.W.F., and Pandit, U.K. (1986) Recl. Trav. Chim. Pays-Bas, 105, 262–265; (f) Fujii, M. (1988) Bull. Chem. Soc. Jpn., 61, 4029–4035; (g) Fujii, M.Y., Aida, T., Yoshihara, M.K., and Ohno, A.Y. (1989) Bull. Chem. Soc. Jpn., 62, 3845–3847; (h) Singh, S. and Batra, U.K. (1989) Indian J. Chem. Sect. B: Org. Chem. Incl. Med. Chem., 28, 1–2; (i) Zhu, X.Q., Liu, Y.C., and Cheng, J.P. (1999) J. Org. Chem., 64, 8980–8981; (j) Zhu, X.Q., Wang, H.Y., Wang, J.S., and Liu, Y.C. (2001) J. Org. Chem., 66, 344–347; (k) Itoh, T., Nagata, A., Kurihara, A., Miyazaki, M., and Ohsawa, A. (2002) Tetrahedron Lett., 43, 3105–3108; (l) Itoh, T., Nagata, K., Miyazaki, M., Ishikawa, H., Kurihara, A., and Ohsawa, A. (2004) Tetrahedron, 60, 6649–6655; (m) Liu, Z.G., Han, B., Liu, Q., Zhang, W., Yang, L., Liu, Z.L., and Yu, W. (2005) Synlett, 1579–1580; (n) Liu, Z.G., Liu, Q., Zhang, W., Mu, R.Z., Yang, L., Liu, Z.L., and Yu, W. (2006) Synthesis, 771–774; (o) Menche, D. and Arikan, F. (2006) Synlett, 841–844; (p) Menche, D., Hassfeld, J., Li, J., Menche, G., Ritter, A., and Rudolph, S. (2006) Org. Lett., 8, 741–744; (q) Zhang, Z.G. and Schreiner, P.R. (2007) Synlett, 1455–1457; (r) Wang, D.W., Zeng, W., and Zhou, Y.G. (2007) Tetrahedron: Asymmetry, 18, 1103–1107; (s) Shen, X.X., Liu, Q., Xing, R.G., and Zhou, B. (2008) 126, 361–366; (t) Goswami, P., Ali, S., Khan, M.M.,
References
8.
9.
10.
11.
and Das, B. (2008) Lett. Org. Chem., 5, 659–664; (u) Liu, Q., Li, J., Shen, X.X., Xing, R.G., Yang, J., Liu, Z.G., and Zhou, B. (2009) Tetrahedron Lett., 50, 1026–1028; (v) Liu, X.Y. and Che, C.M. (2009) Org. Lett., 11, 4204–4207; (w) Richter, D. and Mayr, H. (2009) Angew. Chem., Int. Ed., 48, 1958–1961. For reviews on transfer hydrogenation performed with HEH, see: (a) Ouellet, S.G., Walji, A.M., and MacMillan, D.W.C. (2007) Acc. Chem. Res., 40, 1327–1339; (b) You, S.L. (2007) Chem. Asian J., 2, 820–827; (c) Connon, S.J. (2007) Org. Biomol. Chem., 5, 3407–3417; (d) Wang, C., Wu, X.F., and Xiao, J.L. (2008) Chem. Asian J., 3, 1750–1177. (a) Ouellet, S.G., Tuttle, J.B., and MacMillan, D.W.C. (2005) J. Am. Chem. Soc., 127, 32–33; (b) Huang, Y., Walji, A.M., Larsen, C.H., and MacMillan, D.W.C. (2005) J. Am. Chem. Soc., 127, 15051–15053; (c) for related enone reduction: Tuttle, J.B., Ouellet, S.G., and MacMillan, D.W.C. (2006) J. Am. Chem. Soc., 128, 12662–12663. (a) Yang, J.W., Fonseca, M.T.H., and List, B. (2004) Angew. Chem., Int. Ed., 43, 6660–6662; (b) Yang, J.W., Fonseca, M.T.H., Vignola, N., and List, B. (2005) Angew. Chem., Int. Ed., 44, 108–110; for related reductions: (c) Yang, J.W., Fonseca, M.T.H., and List, B. (2005) J. Am. Chem. Soc., 127, 15036–15037; (d) Martin, N.J.A. and List, B. (2006) J. Am. Chem. Soc., 128, 13368–13369. Rueping, M., Azap, C., Sugiono, E., and Theissmann, T. (2005) Synlett, 2367–2369; (b) Rueping, M., Sugiono, E., Azap, C., Theissmann, T., and Bolte, M. (2005) Org. Lett., 7, 3781–3783; (c) Rueping, M., Sugiono, E., Azap, C., and Theissmann, T. (2007) in Catalysts for Fine Chemical Synthesis, vol. 5 (eds S.M. Roberts and J. Whittall), John Wiley & Sons, Ltd, Chichester, pp. 162–170; (d) Rueping, M. and Sugiono, E. (2008) in Ernst Schering Foundation Symposium Proceedings, vol. 2 (eds M.T. Reetz, S. List, S. Jaroch, and H. Weinmann), Springer, Berlin, Heidelberg, pp. 207–253; (e) Rueping, M.,
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
Sugiono, E., and Schoepke, F.R. (2010) Synlett, 852–865. (a) Schreiner, P.R. (2003) Chem. Soc. Rev., 32, 289–296; (b) Bolm, C., Rantanen, T., Schiffers, I., and Zani, L. (2005) Angew. Chem., Int. Ed., 44, 1758–1763. (a) Akiyama, T., Itoh, J., and Fuchibe, K. (2006) Adv. Synth. Catal., 348, 999–1010; (b) Akiyama, T. (2007) Chem. Rev., 107, 5744–5758; (c) Connon, S.J. (2006) Angew. Chem., Int. Ed., 45, 3909–3912; (d) Terada, M. (2008) Chem. Commun., 4097–4112; (e) Yamamoto, H. and Payette, N. (2009) in Hydrogen Bonding in Organic Synthesis (ed. P.M. Pihko), Wiley-VCH Verlag GmbH, Weinheim, pp. 73–140. (a) Wilen, S.H., Qi, J.Z., and Williard, P.G. (1991) J. Org. Chem., 56, 485–487; (b) Fujii, I. and Hirayama, N. (2002) Helv. Chim. Acta, 85, 2946–2960. (a) Inanaga, J., Sugimoto, Y., and Hanamoto, T. (1995) New J. Chem., 19, 707–712; (b) Furono, H., Hanamoto, T., Sugimoto, Y., and Inanaga, J. (2000) Org. Lett., 2, 49–52. (a) Akiyama, T., Itoh, J., Yokota, K., and Fuchibe, K. (2004) Angew. Chem., Int. Ed., 43, 1566–1568; (b) Akiyama, T. (2004) PCT Int. Appl. WO 200409675. (a) Uraguchi, D. and Terada, M. (2004) J. Am. Chem. Soc., 126, 5356–5357; (b) Terada, M., Uraguchi, D., Sorimachi, K., and Shimizu, H. (2005) PCT Int. Appl. WO 2005070875. Simonsen, K.B., Gothelf, K.V., and Jørgensen, K.A. (1998) J. Org. Chem., 63, 7536–7538. Bartoszek, M., Beller, M., Deutsch, J., Klawonn, M., Kockritz, A., Nemati, N., and Pews-Davtyan, A. (2008) Tetrahedron, 64, 1316–1322. Zhu, S.S., Cefalo, D.R., La, D.S., Jamieson, J.Y., Davis, W.M., Hoveyda, A.H., and Schrock, R.R. (1999) J. Am. Chem. Soc., 121, 8251–8259. Maruoka, K., Itoh, T., Araki, Y., Shirasaka, T., and Yamamoto, H. (1988) Bull. Chem. Soc. Jpn., 61, 2975–2976. Storer, R.I., Carrera, D.E., Ni, Y., and MacMillan, D.W.C. (2006) J. Am. Chem. Soc., 128, 84–86.
819
820
22 Bio-Inspired Transfer Hydrogenations 23. Hoffmann, S., Seayad, A.M., and List, B.
24.
25. 26. 27.
28.
29. 30. 31.
32.
33.
34.
35.
(2005) Angew. Chem., Int. Ed., 44, 7424–7427. Hoffmann, S., Nicoletti, M., and List, B. (2005) J. Am. Chem. Soc., 128, 13074–13075. Marcelli, T., Hammar, P., and Himo, F. (2008) Adv. Synth. Catal., 351, 525–529. Zhu, C. and Akiyama, T. (2009) Org. Lett., 11, 4180–4183. Kang, Q., Zhao, Z.A., and You, S.L. (2007) Adv. Synth. Catal., 349, 1657–1660. Li, G.L., Liang, Y.X., and Antilla, J.C. (2007) J. Am. Chem. Soc., 129, 5830–5831. Kang, Q., Zhao, Z.A., and You, S.L. (2007) Org. Lett., 10, 2031–2034. Li, G.L. and Antilla, J.C. (2009) Org. Lett., 11, 1075–1078. (a) Marcelli, T., Hammar, P., and Himo, F. (2008) Chem. – Eur. J., 14, 8562–8571; (b) Simon, L. and Goodman, J.M. (2008) J. Am. Chem. Soc., 130, 8741–8747. (a) Glorius, F. (2005) Org. Biomol. Chem., 3, 4171–4175; (b) Wang, W.B., Lu, S.M., Yang, P.Y., Han, X.W., and Zhou, Y.G. (2003) J. Am. Chem. Soc., 125, 10536–10537; (c) Lu, S.M., Han, X.W., and Zhou, Y.G. (2004) Adv. Synth. Cat., 346, 909–912; (d) Yang, P.Y. and Zhou, Y.G. (2004) Tetrahedron: Asymmetry, 15, 1145–1149; (e) Xu, L.K., Lam, K.H., Ji, J.X., Wu, J., Fan, Q.H., Lo, W.H., and Chan, A.S.C. (2005) Chem. Commun., 1390–1392; (f) Reetz, M.T. and Li, X.G. (2006) Chem. Commun., 2159–2160; (g) Han, Z.Y., Xiao, H., Chen, X.H., and Gong, L.Z. (2009) J. Am. Chem. Soc., 131, 9182–9183. Rueping, M., Theissmann, T., and Antonchick, A.P. (2006) Synlett, 1071–1074. (a) Rueping, M., Antonchick, A.P., and Theissmann, T. (2006) Angew. Chem., Int. Ed., 45, 3683–3686; (b) Rueping, M., Theissmann, T., and Antonchick, A.P. (2007) in Catalysts for Fine Chemical Synthesis, vol. 5 (eds S.M. Roberts and J. Whittall), John Wiley & Sons, Ltd, Chichester, pp. 170–174. (a) Rakotoson, J.H., Fabre, N., Jacquemond-Collet, I., Hannedouche, S., Fouraste, I., and Moulis, C.
36. 37. 38. 39.
40.
41.
42.
43.
44.
(1998) Planta Med., 64, 762–763; (b) Jacquemond-Collet, I., Hannedouche, S., Fabre, N., Fouraste, I., and Moulis, C. (1999) Phytochemistry, 51, 1167–1169; (c) Houghton, P.J., Woldemariam, T.Z., Watanabe, Y., and Yates, W. (1999) Planta Med., 65, 250–254. Metallinos, C., Barrett, F.B., and Xu, S. (2008) Synlett, 720–724. Guo, Q.S., Du, D.M., and Xu, J. (2008) Angew. Chem., Int. Ed., 47, 759–762. Mani, N.S. and Wu, M. (2000) Tetrahedron: Asymmetry, 11, 4687–4691. (a) Higuchi, R.I., Edwards, J.P., Caferro, T.R., Ringgenberg, J.D., Kong, J.W., Hamann, L.G., Arienti, K.L., Marschke, K.B., Davis, R.L., Farmer, L.J., and Jones, T.K. (1999) Bioorg. Med. Chem. Lett., 9, 1335–1340; (b) Hamann, L.G., Mani, N.S., Davis, R.L., Wang, X.N., Marschke, K.B., and Jones, T.K. (1999) J. Med. Chem., 42, 210–212; (c) Edwards, J.P., Higuchi, R.I., Winn, D.T., Pooley, C.L.F., Caferro, T.R., Hamann, L.G., Zhi, L., Marschke, K.B., Goldman, M.E., and Jones, T.K. (1999) Bioorg. Med. Chem. Lett., 9, 1003–1008. Rueping, M., St¨ockel, M., and Theissmann, T. manuscript submitted for publication. Rueping, M., Theissmann, T., Raja, S., and Bats, J.W. (2008) Adv. Synth. Catal., 350, 1001–1006. (a) Liu, D., Li, W., and Zhang, X. (2004) Tetrahedron: Asymmetry, 15, 2181–2184; (b) Qiu, L.Q., Kwong, F.Y., Wu, J., Lam, W.H., Chan, S., Yu, W.Y., Li, Y.M., Guo, R.W., Zhou, Z., and Chan, A.S.C. (2006) J. Am. Chem. Soc., 128, 5955–5965; (c) Blaser, H.U., Buser, H.P., Hausel, R., Jalett, H.P., and Spindler, F. (2001) J. Organomet. Chem., 621, 34–38. Rueping, M., Brinkmann, C., Antonchick, A.P., and Atodiresei, I. (2010) Org. Lett., 12, 4604–4607. (a) Belattar, A. and Saxton, J.E. (1992) J. Chem. Soc., Perkin Trans. 1, 679–683; (b) Krohn, H., Kirst, H.A., and Maag, H. (eds) (1993) Antibiotics and Antiviral Compounds, Wiley-VCH Verlag GmbH, Weinheim; (c) Kleemann, A., Engel, J., Kutscher, B., and Reichert, D. (eds) (2001) Pharmaceutical Substances, 4th edn, Thieme, Stuttgart, New York;
References
45.
46.
47.
48.
49.
(d) Achari, B., Mandal, S.B., Dutta, P.K., and Chowdhury, C. (2004) Synlett, 2449–2467; (e) Fantin, M., Marti, M., Auberson, Y.P., and Morari, M. (2007) J. Neurochem., 103, 2200–2211; (f) Tenbrink, R.E., Im, W.B., Sethy, V.H., Tang, A.H., and Carter, D.B. (1994) J. Med. Chem., 37, 758–768; (g) Borrok, M.J. and Kiessling, L.L. (2007) J. Am. Chem. Soc., 129, 12780–12785; (h) Cass, L.M., Moore, K.H.P., Dallow, N.S., Jones, A.E., Sisson, J.R., and Prince, W.T. (2001) J. Clin. Pharmacol., 41, 528–535. (a) Satoh, K., Inenaga, M., and Kanai, K. (1998) Tetrahedron: Asymmetry, 9, 2657–2662; (b) Noyori, R. (1996) Acta Chem. Scand., 50, 380–390; (c) Zhou, Y.G., Yang, P.Y., and Han, X.W. (2005) J. Org. Chem., 70, 1679–1683; (d) Krchnak, V., Smith, J., and Vagner, J. (2001) Tetrahedron Lett., 42, 2443–2446; (e) Lee, J., Murray, W.V., and Rivero, R.A. (1997) J. Org. Chem., 62, 3874–3387; (f) Morales, G.A., Corbett, J.W., and DeGrado, W.F. (1998) J. Org. Chem., 63, 1172–1177; (g) Zaragoza, F. and Stephensen, H. (1999) J. Org. Chem., 64, 2555–2557; (h) Ilas, J., Anderluh, P.S., Dolenc, M.S., and Kikelj, D. (2005) Tetrahedron, 61, 7325–7348. Rueping, M., Antonchick, A.P., and Theissmann, T. (2006) Angew. Chem., Int. Ed., 45, 6751–6755. Rueping, M., Tato, F., and Schoepke, F.R. (2009) Chem. Eur. J., 16, 2688–2691. Rueping, M., Merino, E., and Koenigs, R.M. (2010) Adv. Synth. Catal., 352, 2629–2634. (a) Nakashima, D. and Yamamoto, H. (2006) J. Am. Chem. Soc., 128, 9626–9627; (b) Rueping, M., Ieawsuwan, W., Antonchick, A.P., and Nachtsheim, B.J. (2007) Angew. Chem., Int. Ed., 46, 2097–2100; (c) Enders, D., Huttl, M.R.M., Runsink, J., Raabe, G., and Wendt, B. (2007) Angew. Chem., Int. Ed., 46, 467–469; (d) Jiao, P., Nakashima, D., and Yamamoto, H. (2008) Angew. Chem., Int. Ed., 47, 2411–2413; (e) Rueping, M., Nachtsheim, B.J., Moreth, S.A.,
50.
51.
52.
53.
54.
55.
56.
and Bolte, M. (2008) Angew. Chem., Int. Ed., 47, 593–596; (f) Rueping, M., Theissmann, T., Kuenkel, A., and Koenigs, R.M. (2008) Angew. Chem., Int. Ed., 47, 6798–6801; (g) Rueping, M. and Ieawsuwan, W. (2009) Adv. Synth. Catal., 351, 78–84. Han, Z.-Y., Xiao, H., and Gong, L.-Z. (2009) Bioorg. Med. Chem. Lett., 19, 3729–3732. (a) Legault, C.Y. and Charette, A.B. (2005) J. Am. Chem. Soc., 127, 8966–8967; (b) Lei, A.W., Chen, M., He, M.S., and Zhang, X.M. (2006) Eur. J. Org. Chem., 4343–4347; (c) Glorius, F., Spielkamp, N., Holle, S., Goddard, R., and Lehmann, C.W. (2004) Angew. Chem., Int. Ed., 43, 2850–2852. Rueping, M. and Antonchick, A.P. (2007) Angew. Chem., Int. Ed., 46, 4562–4565. (a) Bohlmann, F. and Rahtz, D. (1957) Chem. Ber., 90, 2265–2272; (b) Bagley, M.C., Brace, C., Dale, J.W., Ohnesorge, M., Phillips, N.G., Xiong, X., and Bower, J. (2002) J. Chem. Soc., Perkin Trans. 1, 1663–1671. Synthesis of gephyrotoxin and pumiliotoxin: (a) Fujimoto, R., Kishi, Y., and Blount, J.F. (1980) J. Am. Chem. Soc., 102, 7154–7156; (b) Ito, Y., Nakajo, E., Nakatsuka, M., and Saegusa, T. (1983) Tetrahedron Lett., 24, 2881–2884; (c) Pearson, W.H. and Fang, W.-K. (2000) J. Org. Chem., 65, 7158–7174; (d) Wei, L.-L., Hsung, R.P., Sklenicka, H.M., and Gerasyuto, A.I. (2001) Angew. Chem., Int. Ed., 40, 1516–1518; (e) Sklenicka, H.M., Hsung, R.P., McLaughlin, M.J., Wie, L.-L., Gerasyuto, A.I., and Brennessel, W.B. (2002) J. Am. Chem. Soc., 124, 10435–10442. (a) Daly, J.W. (1998) J. Nat. Prod., 61, 162–172; (b) O’Hagan, D. (2000) Nat. Prod. Rep., 17, 435–446; (c) Daly, J.W., Spande, T.F., and Garraffo, H.M. (2005) J. Nat. Prod., 68, 1556–1575; (d) Michael, J.P. (2005) Nat. Prod. Rep., 22, 603–626. (a) Tietze, L.F. (1996) Chem. Rev., 96, 115–136; (b) Tietze, L.F., Brasche, G., and Gericke, K. (eds)
821
822
22 Bio-Inspired Transfer Hydrogenations (2007) Domino Reactions in Organic Synthesis, Wiley-VCH Verlag GmbH, Weinheim; (c) Enders, D., Grondal, C., and H¨uttl, M.R.M. (2007) Angew. Chem., Int. Ed., 46, 1570–1581; (d) Alba, A.N., Companyo, X., Viciano, M., and Rios, R. (2009) Curr. Org. Chem., 13, 1432–1474; (e) Grondal, C., Jeanty, M., and Enders, D. (2010) Nat. Chem., 2, 167–178. 57. (a) Rueping, M. and Antonchick, A.P. (2008) Angew. Chem., Int. Ed., 47, 5836–5838; For further domino reactions from the same group: (b) Rueping, M., Sugiono, E., and Merino, E. (2008)
Angew. Chem., Int. Ed., 47, 3046–3049; (c) Rueping, M., Sugiono, E., and Merino, E. (2008) Chem. – Eur. J., 14, 6329–6332; (d) Rueping, M., Merino, E., and Sugiono, E. (2008) Adv. Synth. Catal., 350, 2127–2131; (e) Rueping, M., Kuenkel, A., Tato, F., and Bats, J.W. (2009) Angew. Chem., Int. Ed., 48, 3699–3702. 58. Zhou, J. and List, B. (2007) J. Am. Chem. Soc., 129, 7498–7499. 59. Schrader, W., Handayani, P.P., Zhou, J., and List, B. (2009) Angew. Chem., Int. Ed., 48, 1463–1466.
823
23 Life’s Single Chirality: Origin of Symmetry Breaking in Biomolecules Michael Mauksch and Svetlana B. Tsogoeva
` C’est la dissym´etrie qui cr´ee le ph´enomene. P. Curie, J. Phys. 1894, 3, 393.
23.1 Introduction
Life on earth is based on chiral molecules: amino acids (as constituents of the proteins coded by RNA), sugars, and achiral nucleobases that form together the polymeric nucleic acids (RNA and DNA) as carriers of the genetic code [1]. Both sugars and amino acids are C-chiral molecules with one or more chiral carbon centers. The absolute configuration of all the amino acids and sugar molecules employed in the molecules of life on earth is almost exclusively uniform: ‘‘l’’ for amino acids and ‘‘d’’ for sugars – a fact called ‘‘biological homochirality’’ [2]. As an exception, d-amino acids are used as camouflage by bacteria in their cell walls, but never for functional biopolymers that are involved in proteogenesis or replication as these amino acids could be self-poisoning [3]. Many biological receptors on cell walls of various tissues, for example, those involved in the olfactory senses [4], have an explicit sensitivity for molecules with their specific handedness: for example, the cyclic monoterpene limonene has a fresh orange-like smell in the (R)-form, whereas the (S)-enantiomer has a harsh lemon-like odor (Figure 23.1) [5]. One of the biggest challenges in biomimetic chemistry is therefore to emulate Nature’s selection for a single handedness of biomolecules in their neogenesis, to make single enantiomer drugs and fragrances more readily accessible. It appears widely accepted now that life as we know it would not be conceivable without the homochirality of biomolecules [6–9]. Both left- and right-handed versions of biomolecules would, however, in principle be capable of supporting complex life. This leads us to the question whether the observation of only one form of chiral molecules is due to a deterministic process or is, alternatively, accidental, where both forms might have been present in the initial stages of life and one form had become extinct later on in evolution. While fossilic amino acids with a preponderance of (not necessarily natural) l-amino acids had been found Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
824
23 Life’s Single Chirality: Origin of Symmetry Breaking in Biomolecules CH3
CH3
HO
HO O
H2N O H3C
CH2
(R )-limonene (orange)
Figure 23.1
H3C
CH2
(S )-limonene (lemon)
O H2N O
NH2 (S )-asparagine (sweet)
NH2 (R )-asparagine (bitter)
Enantiomers having different smell or taste.
in carbonaceous meteorites [10], no fossilic traces of polymeric peptides or nucleic acids with an excess of the unaccustomed sense of chirality have been discovered yet. The same holds for amino acids or carbohydrates. Hence, it might appear that the origin of homochirality in the small molecules that form the building blocks of life had predated the advent of life on earth [11]. In this context, the exogenous hypothesis of organic material (enclosed in meteorites), enantioenriched by, for example, circularly polarized light in interstellar star-clouds [12], and impacting on the surface of the early earth, has some charm, because it would explain the conundrum of how life could have a both a disparate geographic origin and a deterministic single chirality. However, it is also possible that all those early monomeric chiral molecules of life, for example, those stemming possibly from outer space as well as those that resulted from hypothetical early life with opposite handedness, would mostly become fully deracemized, before the replication apparatus of life successfully sustained the now observed sense of chirality (in light of the racemization times of amino acids, which are short in geological – and probably also evolutionary – time scales [13]). This would explain today’s absence of biomolecules with a greater abundance of the opposite handedness (e.g., d-amino acids) in fossils [14]. Moreover, all these precursor molecules of biopolymers and the peptides or nucleobases themselves are susceptible to oxidation [15]. The change from a probably initially reducing to an oxidizing atmosphere had presumably occurred at about the same period when life began [16]. These oxidation processes would leave hardly traces of a hypothetical extinct ‘‘antipode-life,’’ which leaves the question of its early existence still unanswered. Another important aspect of the homochirality of biomolecules is the distinction between the origin of the initial chiral imbalance and the mechanism of its amplification to enantiomeric purity. To explain the former, several theories have been suggested, the most popular being the proposal of forced symmetry breaking by parity violating energy differences, which are actually known to cause miniscule energetic differences between enantiomers [17]. However, these energy differences are tiny and can even be much smaller than the noise level caused by statistical (thermal) fluctuations that cause accidental chiral imbalances in the enantiomeric composition [18]. In the following we will therefore assume that an initial enantio-imbalance already exists, regardless of its cause, and concentrate on the mechanism of its amplification. Another possibility involves enantioenrichment in biopolymers through polymerization reactions (without overall asymmetric amplification) [19]. While it appears
23.2 Autocatalytic Enantioselective Reactions
Thermodynamic
Kinetic
Phase Equilibria
Asymmetric Autocatalysis
Geometric Adsorption/ Surface Reaction
Conglomerate Deracemization Polymerization/ Aggregation
Figure 23.2 Overview of the enantiodifferentiating chemical and physicochemical processes employed in approaches to the conundrum of biological homochirality.
possible that polymerization itself led to the selection of homochiral strands of peptides or DNA – fed on a pool of racemic monomeric precursor molecules, it is unlikely that this could have occurred only at a single unique spot and only at a unique moment in time. In fact, if life itself – as an autopoietic process – could have generated its own homochiral environment we would assume that present lower organisms could feed on a racemic pool of chiral nutrient molecules. However, d-amino acids have a poisonous or at least inhibiting effect on life processes based on l-amino acids in most organisms [20]. Moreover, and because of the statistic nature of the polymerization (and depolymerization) processes it is also difficult to imagine how polymerization alone could have resulted in the deterministic formation of only left-handed or, alternatively, of only right-handed (‘‘homochiral’’) DNA. This situation has resulted in a multitude of theories that were proposed to explain either the deterministic or the accidental deracemization or, alternatively, the enantioselective formation of amino acids or carbohydrates, or, as a further alternative, the formation of homochiral biopolymers from racemic precursors. In the following, we will discuss the different approaches to the conundrum of biological homochirality, employing enantiodifferentiating chemical and physicochemical processes (see Figure 23.2 for an overview). We focus first on the processes that could have either led to deracemization of a preformed mixture of chiral biomolecules or to the enantioselective formation of these molecules starting from an accidental miniscule chiral bias.
23.2 Autocatalytic Enantioselective Reactions
Asymmetric amplification and spontaneous mirror symmetry breaking in chemical stereoselective or stereospecific reactions appear to be the key to the solution of the homochirality problem, regardless of whether the amplification occurs during polymerization or not [21]. Spontaneous enantioenrichment (when the ee value is defined as the degree of enantiopurity of a polymer or of an ensemble of
825
826
23 Life’s Single Chirality: Origin of Symmetry Breaking in Biomolecules
polymers) might also occur in the formation of homochiral polymers in stereospecific reactions acting on a racemic precursor pool of chiral monomers [22]. Asymmetric amplification requires less stringent mechanistic conditions than symmetry breaking in stereoselective reactions of prochiral reactants, and denotes the (usually limited) increase of enantiomeric excess from non-zero starting values, for example, in the catalyst, whereas spontaneous mirror symmetry breaking denotes the amplification of enantiomeric excess from even the tiniest initial enantio-imbalances (e.g., those caused by statistical fluctuations or parity violating energy differences) through nonlinear reaction schemes [23]. While several reaction schemes have been invented and studied that exhibit positive nonlinear effects in the ratio of product to catalyst ee (Figure 23.3) – for example, those of Noyori (with non-specific mutual inhibition due to statistical formation of homoand heterochiral aggregates of catalyst) [24], and of Kagan (e.g., in the reservoir model of catalyst aggregation, or with catalytic ML2 complexes built from a chiral ligand L) [25], and several examples (e.g., in the epoxidation of chalcone) have been reported – the phenomenon of true asymmetric amplification due to spontaneous mirror symmetry breaking is more elusive. In the presumably prochiral or racemic prebiotic environment, asymmetric amplification appears a prerequisite for an assumed deracemization of the biomolecule precursors. As asymmetric amplification is implied in spontaneous symmetry breaking, and if one assumes that sizable enantio-imbalances (e.g., through enantioenriched molecular material from outer space) had not existed a priori, we concentrate in the following discussion on spontaneous mirror symmetry breaking as the crucial phenomenon and how it could have occurred endogenously on the primordial earth. MLRLR + MLSLS x y kRR kSS
(a) M + LR + LS
eeligand
eemax
−eemax
racemic β = z/(x + y) g = kRS /kRR
eeprod = eemax * eeligand * (1 + β)/(1 + gβ) (b)
S
S +
S R R monomeric catalysts, active
S +
R
MLRLS z kRS
dimeric catalysts inactive
R
Figure 23.3 Kagan (a) and Noyori (b) models of asymmetric amplification through catalyst aggregation. Dimer formation in the Noyori model is statistic. In (a) g denotes the branching ratio for heterodimeric versus homodimeric catalyst activity, β is proportional to the equilibrium constant for the
dimer equilibria, kRS and kRR are the rate constants for the formation of chiral product (not shown here) by heterochiral and homochiral catalysts, respectively, M is a metal center to which chiral ligands LR and LS could coordinate.
23.2 Autocatalytic Enantioselective Reactions
A possible solution to the homochirality conundrum that meets these requirements could involve complex reaction mechanisms with nonlinear kinetics. Since a landmark paper by the British crystallographer Charles Frank, published in 1953, who presented a mathematical model reaction network capable of producing homochiral solutions (with 100% ee) [26], chemists began to consider spontaneous mirror symmetry breaking reaction mechanisms in chemistry. The Frank mechanism (Figure 23.4) involves linear and irreversible autocatalytic steps followed by an irreversible nonlinear inhibition reaction in which the product enantiomers formed in the autocatalytic step recombine to give an optically and catalytically inactive heterochiral product dimer. Thereby, the enantiomeric purity of the autocatalytic product is constantly increased, once a racemic initialization phase is passed (Figure 23.4). Frank did not specify the physical boundary conditions of the system this mechanism is operated in and did not comment on thermodynamic requirements, but states confidently that a ‘‘well stirred mixture, like, for example, a reaction bulb’’ would be sufficient to demonstrate the process in Reference [26]. For decades, the Frank mechanism and its modifications served as a hypothetical example of spontaneous chiral symmetry breaking in homogeneous mixtures, (a)
A +R A +S R+S
∆G
R + R S + S P
A
A R,S
R S
P
P
r.c. (b) A +R A +S R+S
∆G
R + R S + S A + A
A
A
RS
R S
R,S r.c.
(c) A + R+ R A +S+ S
R + R+ R S+S+S
A
for (a), (b): conc.
R+R S+S
RR SS
for (c): conc. R
R S
B.P. racemic nonracemic
R S
R S
S time
time R S R S R S
Figure 23.4 Basic schematic autocatalytic mechanisms of asymmetric amplification: the original Frank mechanism, depicted here with reversible reaction steps (a), and the mechanism with recycle Frank kinetics
(b), which both involve mutual inhibition, and the hypercompetitive mechanism without mutual inhibition (c). A = prochiral reactant, B.P. = bifurcation point.
827
828
23 Life’s Single Chirality: Origin of Symmetry Breaking in Biomolecules
mostly in open flow systems. Kondepudi and Nelson investigated, for example, in 1983 a modification of the Frank mechanism with reversible autocatalytic steps and an irreversible inhibition step under flow conditions [27]. Gutman et al. studied the Frank mechanism and a modification with enantiomer racemization analytically [28]. However, with limited resources and in a closed system, complete homochirality (with 100% ee) cannot be achieved by the original Frank scheme, where a minor enantiomer recycling pathway is absent. While Avetisov first considered a hypothetical fully reversible Frank-type mechanism (Figure 23.4) [29], and Plasson proposed a complex autocatalytic recycling reaction network in 2004 [30], a simple, fully reversible and recycle modification of the Frank mechanism was first proposed in 2007 to apply in reversible organic reactions (Figure 23.4) [31]. Such a non-productive reaction cycle can, of course, not simply turn by itself – without violating the law of energy conservation, because the recycling step is necessarily non-spontaneous. Every turn of the reaction cycle is therefore accompanied by an inevitable loss in chemical energy, which has to be compensated through coupling to an irreversible process. Biochemical examples are so-called ‘‘futile’’ metabolic cycles, like phosphorylation/dephosphorylation or the ATP-consuming process of glycolysis combined with gluconeogenesis [32]. The proposed Frank type reaction scheme with specific minor enantiomer recycling (Figure 23.4b), and variations thereof, was further studied under the conditions of a closed system by Sugimoro et al. (using the master equation from statistical thermodynamics) [33], by Ribo and Hochberg (employing stability analysis) [34], and by Blackmond (imposing equilibrium thermodynamics) [35], who all came to the conclusion that – under assumed reversible (i.e., non-dissipative) conditions – the recycle Frank kinetics alone (with its closed reaction loops) results – in a rate equation formalism – in asymmetric depletion, rather than asymmetric amplification, because a unidirectional cyclic operation of the mechanism without a source of energy is, of course, precluded by thermodynamic restrictions [35]. This result is unsurprising because, due to the symmetry of the reaction network, the principle of microscopic reversibility demands that the concentrations of the enantiomers are equal at equilibrium: [R]eq = [S]eq , which imposes a restriction of generality at situations far from equilibrium [35, 36], letting the implication of asymmetric depletion appear as a tautology [36]. Most real physical and chemical processes are irreversible (i.e., dissipative), though. Nicolis and Prigogine suggested already decades ago that a chiral bias could be spontaneously generated and amplified in dissipative far-from-equilibrium systems [37, 38]. Indeed, important biological functions (e.g., enzyme regulation, action of molecular motor proteins) could, for example, be understood by resorting to the concept of the open system non-equilibrium steady state and implying closed reaction loops [39]. The possibility of a dissipative non-productive reaction cycle, operating at non-equilibrium concentrations under open system steady-state conditions and consisting of a sequence of reversible reaction steps, was not considered in earlier proposals for the origin of homochirality – and has apparently been overlooked; recently, the reversible and recycle modification of the original Frank mechanism
23.2 Autocatalytic Enantioselective Reactions
(Figure 23.4a,b) was investigated under open flow conditions and it was suggested that it might lead to mirror symmetry breaking when coupled to an energy consuming side- or follow-up reaction running inside the experimentally closed system [36], or, alternatively, to an external source of energy [30]. Both scenarios realize the open system conditions through exchange of chemical energy with the environment. The latter possibility was already explored in depth by Plasson and coworkers for a similar mechanism invented earlier by the same group, inspired by the hypothetical experimental prebiotic chemistry system of W¨achtersh¨auser et al. [40]. Recently, Ribo and coworkers have demonstrated that a fully reversible Frank mechanism with a weakly exergonic inhibition step could result in spontaneous mirror symmetry breaking and (temporary) asymmetric amplification in a reversible homogeneous system, that is even closed to matter flow [41]. However, to date, there is no verified laboratory example for either the classical or recycle Frank mechanism. Nevertheless, the Frank model and its modifications have been frequently invoked to explain the observation of asymmetric autocatalysis in the Soai reaction [21], despite the absence of experimental evidence for a bifurcation mechanism. The main drawback of the Frank model though is the absence of a plausible mechanism of chirality induction in organic reactions. As one of several discussed alternatives to the Frank scheme with its linear order autocatalytic steps [42], Decker proposed in 1975 [43] that hypercompetitive mechanisms could also lead to mirror symmetry breaking. In this class of reactions, one of the autocatalytic enantiomers is practically outrun by its antipode, the amount of which is growing faster (Figure 23.4). In these reaction schemes, the autocatalytic step has a higher than linear reaction order with respect to the monomeric product. No mutual inhibition as in the Frank mechanism is involved, and no bifurcation in the ee values occurs: the enantio-imbalances increase monotonously from a tiny starting value. It has been proposed that reactions in the coordination sphere of metal complexes could provide a plausible experimental realization of this mechanism [43]. It appeared therefore promising to look for such examples among organometallic reactions where the chiral catalyst could act as a reactant (e.g., in group transfer reactions) and which is afterwards regenerated by stereoselective oxidative addition of the transferred group from the reactant pool. The first remarkable experimental demonstration of such spontaneous chiral symmetry breaking was achieved by Soai, who reported in 1995 the formation of chiral pyrimidyl alcohols in an irreversible organometallic reaction with high product enantiomeric excess of more than 99%, starting from very low ee values and in a sequential batch reaction protocol (Scheme 23.1) [44]. The mechanism of the Soai reaction was not elucidated until recently, when Schiaffino and Ercolani found, through density functional theory (DFT) calculations and a fit of computed and experimental data, that dimeric (at higher temperatures) or heterochiral 2 : 2 tetrameric aggregates (at lower temperatures) of the chiral Zn-alkoxide primary reaction product form the actual ‘‘autocatalytic’’ species (Figure 23.5) [45]. The homochiral (RR) dimer (1) constitutes a bidentate ligand that coordinates to two molecules of iPr2 Zn, forming a dinuclear metal complex 2 (step a, Figure 23.5). Complex 2 then rearranges to 3 through coordination of
829
830
23 Life’s Single Chirality: Origin of Symmetry Breaking in Biomolecules
OH
N
R
OZni Pr
N Zni Pr2
consecutive asymmetric autocatalytic reactions
CHO
N
N
N
i Pr
N
c H
N
H
Zn Zn
iPr
i Pr
H N
Zn
i Pr iPr
Zn N
i Pr
N
i Pr Zn O O Zn i Pr 2
H
i Pr
N
5
e N
i Pr
iPr
N
b N
i Pr
HH
a
N
Zn O
2 i Pr2Zn
N
i Pr i Pr
O Zn
N N
i Pr
1 i Pr
i Pr i Pr
N O
Zn N
N
Zn O
N
i Pr
i Pr Zn N i Pr O O N i Pr O Zn i Pr H N Zn Zn N O i Pr i Pr f N i Pr 6 H N
i Pr N Zn i Pr
i Pr
i Pr
Figure 23.5 Mechanism of the Soai reaction according to DFT computations [45]. The dominant catalytic species is a homochiral dimer. Lower case letters (a–f) refer to the different reaction steps as discussed in the text. Species 1 is the homochiral dimer (RR) catalyst.
H
O
O
H
N
N
Zn
Zn
O
i Pr
i Pr
O
O
Zn
3 i Pr
i Pr
i Pr
Zn i Pr
N
i Pr
O
i Pr
N
d
Zn
Zn
H
N
H
O
N
4
N
N
O
O
i Pr i Pr
i Pr
O
N
i Pr
i Pr N Zn
H
i Pr
N
i Pr
i Pr Zn
i Pr N Zn
N
N
H N
2
+ Zni Pr2
N
R
O
N (with high ee)
H3O+
N
R
H
OH
N
R N (with low ee)
Scheme 23.1 Soai reaction: a first absolute asymmetric synthesis.
N
23.2 Autocatalytic Enantioselective Reactions
an oxygen atom of one dimer part toward the Zn metal center of the second monomer unit. Taking up two further pyrimidyl aldehyde molecules produces complex 4, in which the aldehyde is coordinated via N and O atoms to zinc. In step d, one of the alkyl moieties (encircled) is transferred onto the carbonyl atom of one of the aldehyde molecules. Transfer of a further isopropyl unit to the second aldehyde results in complex 6 (step e), which is a homochiral tetramer (R4 ). Finally, the tetramer dissociates to release the dimeric catalyst 1. The formation of (RS)2 heterodimers is further proposed to explain the observed high extent of asymmetric amplification at lower temperature. In the year of Soai’s seminal report, P. Bailey suggested the spontaneous self-amplification of enantiomeric excess in the generation of a chiral product through modification of Kagan’s ML2 model [25] to account for autocatalysis [46]. A few years later, Blackmond and Brown [47] and Buono and Blackmond [48] thoroughly investigated the kinetics of the Soai reaction with microcalorimetry, and proposed that the reaction rate depends on the enantiomeric composition of initially added product [47, 48]. They observed that the reaction proceeded twice as fast when the initially added product was enantiopure, as compared to racemic added product [48]. Hence they deduced that the catalyst should be a homochiral product dimer, rather than the monomer itself, and that the heterodimer is catalytically inactive (Figure 23.6) [47]. As a consequence, the change of the ratio of enantiomer concentrations d[R]/d[S] depends quadratically on the ratio of enantiomer concentration at a given point in time: d[R]/d[S] = ([R]/[S])2 (R and S are denoted LR and LS , respectively, in Figure 23.6 for consistency with S
R R R
LR + LR LS + LS
fast
LRR
fast
LR + LS
fast
LSS LRS
A + N + LRS
S
inactive
R
dimeric catalyst
+
inactive
A + N + LSS
dimeric catalyst
+
S
A + N + LRR
S
k1 k1 k2
LRR + LR
dLR/dt ~ k1*A*N*LR2
LSS + LS
dLS/dt ~ k1*A*N*LS2
LRS + ½*LS + ½*LR
Figure 23.6 Blackmond–Brown model of asymmetric autocatalysis in the Soai reaction. Only the homochiral dimers are catalytically active. The reaction proceeds via a second-order autocatalysis. The mechanism is a combination of Kagan’s concept
k1 >> k2 of dimer catalysis and Noyori’s model with monomer/dimer equilibria. LR and LS are monomeric enantiomers from which dimers LRR , LSS (homochiral), and LRS (heterochiral) can form. ‘‘A’’ is a prochiral reactant, N denotes a nucleophile.
831
832
23 Life’s Single Chirality: Origin of Symmetry Breaking in Biomolecules
Figure 23.3). The product enantiomeric excess is therefore continuously increasing (see also Figure 23.4c). These mechanistic insights do not explain fully the dynamics of the Soai reaction, though. There is, for example, evidence from stochastic analysis (i.e., fitting of the likelihood in outcome for a certain enantiomeric excess to a parameterized kinetic model) that the Soai reaction proceeds via three cooperatively coupled autocatalytic cycles [49]. The Soai reaction serves as a paradigm example of absolute asymmetric synthesis in the absence of external asymmetric physical forces [50], and as a proof-of-principle for the possible role of autocatalytic processes in the solution of the homochirality conundrum. However, the reaction requires non-aqueous media to run in, which might not be that plausible in light of the probably aqueous conditions on the primordial earth. Fully organic examples of asymmetric autocatalysis [51], which, similar to the Soai reaction, can also show the behavior of an absolute asymmetric synthesis, were first reported by Mauksch and Tsogoeva in 2007 [52]: an asymmetric Mannich-type product – a functionalized amino acid – was able to replicate itself and with the same absolute configuration (Figure 23.7). DFT (density functional theory) calculations showed that the formation of homochiral product dimers is kinetically preferred with respect to formation of heterochiral dimers in the autocatalytic steps [52]. Figure 23.7 shows the proposed catalytic cycle: the Mannich product is OMe O HN
OMe
OMe CO2Et
+
O
N EtO2C
(15 mol%, 99% ee)
H
4 days, RT
O HN CO2Et 42% y, 94% ee (R)
+ Reactant
Product
Reactant-Product Complex +
Reagent TS
2x Product
Product-Dimer
Figure 23.7 Asymmetric autocatalysis in an organocatalytic reaction. The proposed mechanism involves only a monomeric product autocatalyst as the active catalytic species.
23.3 Autocatalysis and Self-replication
supposed to form hydrogen bonded adducts with the reactant imine via specific ‘‘recognition sites,’’ providing an explanation for the experimentally found chirality induction, which is similarly effective as with well-known external organocatalysts, like proline. Even without added catalyst, product with 9.4% ee at 31% yield was obtained after four days reaction time [31]. In principle, a conceivable enantioenrichment in certain amino acids might have led to a corresponding enantioenrichment in the sugars, because it has been shown first by Pizzarello and Weber [53] and later by Cordova [54] that amino acids might be efficient catalysts in the aldol reactions involved in producing carbohydrates. The connection of amino acid handedness and those of sugars has also been recently established by Nanita and Cooks [55]. Very recently, it was also suggested that asymmetric autocatalysis in the aldol reaction might play a role in gluconeogenesis via the formose reaction [56].
23.3 Autocatalysis and Self-replication
Self-replication appears to be an indispensable feature of the molecules of life (RNA and DNA) and necessary to sustain their homochirality–once achieved–against the relentless trend to racemization. Eigen and Schuster have theoretically studied self-replicating and autocatalytic systems [57]. Philp [58] and von Kiedrowski [59] have elegantly shown the connection of autocatalysis and self-replication. While von Kiedrowski reported studies on the replication and autocatalysis of oligomeric nucleotide sequences [59], his group also found that autocatalysis does not necessarily have to be self-accelerating – a property often attributed to autocatalytic or autoinductive processes – because the initial product of the autocatalytic step in the reaction of, for example, non-polymeric organic molecules can be considered a, for instance hydrogen-bonded, dimer of product template molecules (Figure 23.8), rather than isolated monomers themselves, as, for example, is assumed in the original Frank scheme [26]. This mechanism, which provides a reasonable mechanism of chiral induction in chiral autocatalysis, requires the presence of appropriate recognition sites (e.g., hydrogen bond donor/acceptor pairs) in the product (or substrate) molecules. Such a reaction scheme, which involves subsequent dissociation of the dimer product to release the monomeric catalytic template molecules [58], appears more plausible in organo-autocatalytic reactions [52, 56] than schemes that require the regeneration of the catalyst or those that give only monomeric initial product. However, the mechanism is not nonlinear and therefore cannot give asymmetric amplification, because the ratio of formation rates of the enantiomers is proportional to the ratio of their concentrations: d[R]/d[S] = [R]/[S]. Very recently, Rebek, a pioneer of self-replicating simple organic non-nucleotide molecules, elegantly demonstrated an experimental example where the product of the autocatalytic step is even able to catalyze organocatalytically a different reaction in which the self-replicator is reproduced (Scheme 23.2) [60].
833
23 Life’s Single Chirality: Origin of Symmetry Breaking in Biomolecules
834
O O
A N H
N O
T OH
N H O
O O N O
H
B
N O
CO2H
O
O
H O
N + + N N
N N H N O H N O
N H
N
−
H N
H N NN
N
N N H N O H N O O O O H H H N N N N O N O H N
O
N N H N O N H H O N O O O O H H − N N N N N+ O O N
N H O
N H O
O
T
[T·T]
Figure 23.8
[A·B·T]
Example of a minimal organic self-replicator with the autocatalytic template T.
t -Bu
NH2 N N
t -Bu
N O
NH2
N N
5% Et3N
N
CH2Cl2
BnN
+
O
−78°C
NH3+Cl−
t -Bu
NH2
O
N H
N H N
N H
S
t -Bu
O
H 3C
BnN
H3C
S C N
self-complementary recognition sites
NH2
N N H
O
implanted organocatalysis site
O
H N H
O
N
N N
H N
N N H H
O
CH3
Scheme 23.2 Organocatalytic synthetic self-replicator of Rebek [60]; recognition sites are hydrogen bond donor/acceptor pairs.
23.4 Polymerization and Aggregation Models of Enantioenrichment
W¨urthner et al. [22] have demonstrated recently that in an ensemble of polymeric dyes homochiral strands can evolve from the initial aggregates even if the aggregation is fed on a near-racemic pool of monomeric precursors, employing a
23.5 Phase Equilibria low ee
low ee fast
RR RS self-assembly SS
chiral monomers
Figure 23.9
dimers
low ee
high ee ee amplification
A
A′
aggregates
enantioenriched aggregates
Wurthner’s ¨ polymerization model of enantioenrichment.
kinetic (autocatalytic) explanation for the ‘‘majority rules’’ effect of asymmetric amplification in supramolecular chemistry (Figure 23.9) [61]. Very recently, it was even shown that supramolecular aggregation on surfaces could result in drastic symmetry breaking and asymmetric amplification [62]. Lahav and coworkers have, for example, very recently and elegantly shown that homochiral oligopeptides can show spontaneous enantioenrichment during their formation reaction, due to the additional unexpected formation of racemic beta-sheets [19]. A further example is the theoretical recycling reaction model of Plasson et al. [30], which is actually a model for polymerization with epimerization and has been studied in depth by Brandenburg et al. [63]. A non-racemic non-equilibrium steady state is assumed to evolve from a racemic state far from equilibrium [64]. Notably, the ‘‘recycling’’ in this model does refer to the reactivation of the chiral monomers resulting from depolymerization, rather than to a ‘‘minor enantiomer recycling’’ that leads back to the prochiral reactant, discussed earlier [31]. In contradistinction to the original Frank mechanism, homochiral product dimers are assumed to be lower in energy than heterochiral dimers, but higher in energy than monomers [30]. The Plasson mechanism is, however, nevertheless a kinetic bifurcation mechanism (like Frank’s), at least when only dimer species are involved. With longer chains, next-nearest neighbor interactions could be taken into account, which would render the polymerization and epimerization steps nonlinear with a reaction order slightly larger than one. The enantioenrichment is here principally twofold: first, individual polymers might partially depolymerize to be rebuilt with a higher degree of enantiopurity; secondly, the proportion of homochiral polymers of the same sense of handedness can grow in an ensemble of polymeric strands of different lengths. This scenario might conceivably also apply to biopolymers. In this case, the origin of homochirality in biomolecules would not be due to symmetry breaking and true asymmetric amplification acting on a pool of near-racemic precursor molecules (like, e.g., amino acids), but would rather be the consequence of the autopoietic processes of life itself. It is conceivable that the formation of homochiral biopolymers and the development of the replication mechanism of these polymers had developed in sync. 23.5 Phase Equilibria
Enantioenrichment through asymmetric autocatalysis is based on nonlinear reaction kinetics under non-equilibrium thermodynamic conditions. Another
835
836
23 Life’s Single Chirality: Origin of Symmetry Breaking in Biomolecules (a)
(b)
solution solid
high ee R
RS
S
R
S
R
RS
eutectic
eutectic
low ee 100% ee
0% ee
100% ee
Figure 23.10 (a) Situation in solution; (b) an example binary melting diagram. RS denotes the racemic compound (marked by a dystectic point on the liquidus curve) and R and S are pure enantiomers.
possibility of phase-specific enantioenrichment through – thermodynamically, rather than kinetically – driven enantiospecific phase re-distribution processes has been put forward first by Hayashi [65] and later by Blackmond [66] and by Breslow [67]. The ternary system, composed of the two amino acid enantiomers and water, has a ternary phase diagram with eutectic points corresponding to a non-racemic distribution of the enantiomers in the solution and the solid phases when the amino acid forms a racemic compound, while those amino acids that form conglomerates have only racemic eutectics as thermodynamic fixed points, in accord with the Gibbs’ phase rule [65, 66]. The racemic compound is here assumed to have a higher melting point than the enantiopure solids. Figure 23.10 depicts the situation in solution (a) with an exemplary binary melting diagram (b), RS denotes the racemic compound (marked by a dystectic point on the liquidus curve), and R and S are pure enantiomers. The ee value of, for example, serine in solution increased to the equilibrium value of 99.5% in a heterogeneous mixture of scalemic solid serine under its saturated solution, while the solid phase showed a compensating asymmetric depletion [66]. Soloshonok made the remarkable observation that a similar self-disproportionation of enantiomers with enantioenrichment in one of the two phases (accompanied by asymmetric depletion in the other phase) may also occur during chromatography or sublimation, re-crystallization, distillation, or other phase transitions [68]. Soloshonok explained this by the difference in stability of homochiral versus heterochiral aggregates. Suhm and coworkers found that the self-disproportionation behavior observed by Soloshonok may be shown only by compounds that form true racemates (i.e., racemic compounds), and when a certain critical ee value in the initial scalemic mixture is exceeded [69]. As most biochemical reactions take place in solution, such solution phase enantioenrichment is possibly relevant in the origin of homochirality scenarios. Moreover, about 90% of all chiral compounds form true racemates. A potential drawback could be the unexplained fate of the racemic or asymmetrically depleted solid amino acid phases. Even properties of liquid crystal phases have been suggested earlier as possible cause for the biological homochirality [70].
23.7 Spontaneous Symmetry Breaking in Conglomerate Crystallizations Figure 23.11 Achiral reactant A ‘‘sees’’ only one of the two enantiotopic faces and, therefore, reacts stereospecifically.
a
A
A c b
As an alternative, Cooks reported in 2007 that solid serine at 3% ee shows remarkable genuine asymmetric amplification to 69% ee on sublimation [71]. The mechanism of this amplification is still not clear.
23.6 Adsorption on Chiral Surfaces
Even achiral compounds or ions may crystallize in chiral space groups. Hence, apart from the above thermodynamic or kinetic schemes, a method of chiral resolution (separation of enantiomers) in the biochemical context is suggested by the possibility of enantiospecific adsorption of small chiral molecules like amino acids on chiral crystal surfaces of, for example, calcite, as a special case of chiral recognition [72]. Such mechanisms could probably have had a high chance of realization in prebiotic scenarios. The enantiotopic faces of crystals have also recently been employed in such a ‘‘geometric’’ approach to absolute asymmetric synthesis in a stereoselective reduction reaction by Kuhn [73], acting on similar observations by Holland and Richardson made 20 years earlier [74]. An achiral reactant A ‘‘sees’’ only one of the two enantiotopic faces and reacts stereospecifically (Figure 23.11). Lahav further proposed that chiral additives possessing a greater affinity or structural resemblance to one of the enantiomers in their crystalline state may hamper kinetically the crystallization of that enantiomer, resulting in accumulation of the opposite enantiomer in the solid phase (‘‘Lahav’s rule of reversal’’) [75]. Acting on these predictions, Vlieg et al. demonstrated recently that this approach can be used to deracemize scalemic conglomerates of chiral compounds to enantiopurity in the solid state by grinding them under their saturated solution in the presence of a chiral additive, while the opposite enantiomer stays in solution [76].
23.7 Spontaneous Symmetry Breaking in Conglomerate Crystallizations
Pincock [77] and later Kondepudi [78] observed that a conglomerate forming inorganic salt (NaClO3 ) spontaneously crystallizes with random enantiomeric
837
838
23 Life’s Single Chirality: Origin of Symmetry Breaking in Biomolecules
excess from supersaturated solutions. Kondepudi found that vigorous stirring led to a fat-tailed (leptokurtic) histogram distribution of crystal enantiomeric excesses, while Pincock earlier observed a Gaussian distribution with a maximum likelihood for the racemic outcome [77, 78]. For an organic racemizing compound in unstirred supersaturated solutions, a similar observation was already made in the 1940s by Havinga [79, 80]. Rib´o more recently pointed out that primary nucleation during crystallization from the supersaturated solution phase through cooperative ‘‘chiral recognition’’ interactions could result in sizable enantiomeric excesses in the crystalline phase [81]. Alternatively, secondary nucleation has been invoked as the cause of high cee (crystal enantiomeric excess) values in vigorously stirred supersaturated solutions under kinetic control (and when crystallization and dissolution are balanced), because secondary nucleation implies the autocatalytic nature of the crystallization process in the presence of a crystal surface, which results in self-amplification of enantiomeric excess in the solid phase [82]. In 2005, Viedma demonstrated in a legendary experiment that a racemic conglomerate mixture of an inorganic salt with achiral ions in the (merely saturated) solution can be fully deracemized when crystal abrasion through vigorous stirring is assisted by added glass beads [83]. Viedma explained his observation qualitatively by thermodynamically controlled dissolution of the smallest crystals (due to the Gibbs–Thomson rule) in conjunction with the kinetically controlled nonlinear autocatalytic crystal growth, which was already proposed earlier by Uwaha as the cause for the conglomerate deracemization [84]. Afterwards, Rib´o [81] and, later, Blackmond [85] developed a conceptual expansion of the Viedma experiment to the deracemization of intrinsically chiral compounds that racemize rapidly in solution. In 2008, Vlieg, Blackmond, and associates reported a first stunning experimental realization of this idea: the stirred slurry of a racemic or nearly racemic chiral proteinogenic amino acid compound was fully deracemized to a chiral solid phase with 100% ee and with almost complete conversion rates (Figure 23.12) [86]. The compound racemized rapidly in the solution phase in the presence of a strong base at room temperature (with a half-life of about 10 min), while the deracemization process itself took weeks to complete. The increase in crystal ee value with time was exponential. Ostwald ripening, a thermodynamic mechanism of crystal growth, was employed by the authors to explain the deracemization process [86, 87]. The remarkable experimental process was later optimized further by Vlieg, Kaptein, Kellogg, and coworkers to allow complete deracemization in days or even hours, rather than weeks [87]. Soon after the first experimental reports, again Uwaha explained also these newer observations on the basis of a rate equation model that involves nonlinear kinetics of crystal growth, extending his earlier model [84] to account for solution phase racemization [88]. Uwaha proposed that, instead of monomeric molecules, subcritical clusters are incorporated into the growing bulk crystals. The connection of these fascinating results to the origin of homochirality is not at once apparent. Nevertheless, wind- or water-tossed sand or pebbles might conceivably have provided for the grinding in half-dried puddles, for example, at sea-shores.
23.7 Spontaneous Symmetry Breaking in Conglomerate Crystallizations
Crystallization
Solution
Enantiomerization R
S
Dissolution
(R )-enantiomorph
Crystallization
Solution (S )-enantiomorph
Crystal Crushing Solid
Solid
NH2
N O
Figure 23.12 Complete conversion of racemic crystal conglomerates into a single chiral state. Crystallization, crystal crushing, and dissolution form two competing process cycles for the R and S species,
respectively – coupled by the enantiomerization reaction. The example shown is based on Reference [86]. The minor enantiomer is recycled through dissolution and enantiomerization.
Most recently, Tsogoeva, Mauksch, and coworkers reported that the slurry of the chiral product of a (reversible) Mannich reaction stirred under its saturated toluene solution can also be deracemized, and without the presence of a strong base (and without the presence of glass beads), combining conglomerate deracemization and asymmetric synthesis [89]. The Mannich product is here proposed to enantiomerize via the reactant imine – instead of directly; Mannich and retro-Mannich reaction steps are catalyzed by racemic or achiral organocatalysts. Functionalized amino acids can be formed by the Mannich-type reaction. A solution phase excess of the solid phase’s minor enantiomer was observed during the initial stages of the deracemization experiment [89, 90]. This surprising effect was explained in accord with Uwaha’s autocatalytic crystal growth model, in which such an observation is implied [88]: nonlinear growth leads to faster deposition of the major enantiomer from solution [90]. Notably, the incorporation of clusters is not the only possibility for achieving a nonlinear growth rate: monomers could conceivably also aggregate and interact at neighboring sites at the crystal growth front. Paradoxically, even at mildly elevated temperatures and in presence of a catalyst, the racemization half-life of the dissolved Mannich product was days in solution, while the total deracemization of the solid took only hours, which is a behavior opposite to the observations of Vlieg and associates, where direct and solution phase enantiomerization was faster, rather than slower, than deracemization [86, 87]. To explain this, a mechanism of autocatalytic enantiomerization at the crystal surface was invoked, a mechanism recently proposed theoretically by Saito et al. to explain deracemization of stirred conglomerate slurries [91]. The surface-assisted racemization in Saito’s model, however, plays a dual role: first, the implied
839
840
23 Life’s Single Chirality: Origin of Symmetry Breaking in Biomolecules
nonlinearity of the enantiomerization process (due to its dependence on the respective crystal surface areas of the enantiomorphic bulk crystals) is employed to explain the physical cause for the deracemization and, second, it provides a different mechanism for the chemical process of enantiomerization, independent from a solution phase racemization [90]. Asymmetric autocatalysis, a process known from homogeneous systems [44, 51], might therefore be extended conceptually to heterogeneous systems [90]. Slowly racemizing chiral compounds have not been considered before in concepts for complete conglomerate deracemization [81, 85–87]. Very recently, Bolm was able to employ a similar process in the deracemization of an aldol reaction product [92]. Viedma, Blackmond, and coworkers observed for the deracemization of a proteinogenic amino acid, applying the base-catalyzed direct enantiomerization method in heated slurries, that in absence of grinding the increase in crystal ee value was less than exponential [93]. They also observed the temperature dependence of the deracemization process and found that heating accelerates the deracemization process, which was explained by the temperature-induced activation of the solution-phase enantiomerization.
23.8 Symmetry Breaking in Reaction–Diffusion Models, Collision Kinetics, and Membrane Diffusion
Nonlinear kinetics, required to break the mirror symmetry spontaneously, are not only exhibited by chemical reaction kinetics, or in the autocatalytic physical process of crystal growth, but also by the combination of chemical reactions with, for example, the physical process of diffusion, which has a nonlinear time-dependence on the concentration gradients, as shown by Gayathri and Rao in a recent theoretical paper [94]. A further idea was proposed by Shinitzky, stating that the diffusion and, hence, permeability of phase boundaries or membranes for amino acids could differentiate between the enantiomers due to chiroselective hydration via ortho-H2 O [95]. Toxvaerd showed theoretically, by employing kinetics based on collision theory in molecular dynamics simulations, that a simple chiral carbohydrate, glyceraldehyde, may deracemize at high particle densities via reaction-step dependent activation barriers for the keto–enol tautomerization [96].
23.9 Concluding Remarks and Outlook
As we have seen, several competing theories are vying to explain the origin of biological homochirality, which appears so central to life in the forms we are familiar with. The race between these different approaches is far from decided, as more
References
theories and observations are reported. While some theories for the endogenous origin of homochirality stress a thermodynamic origin of enantioenrichment in the solution phase, others put more weight on mirror symmetry breaking kinetic mechanisms of asymmetric amplification building up upon initial imbalances in the enantiomeric compositions in homogeneous ensembles of chiral molecules. Supramolecular aggregation and polymerization could also result in a homochiral world of biosystems. A further suggestion involves symmetry breaking crystallizations or the exploitation of chiral influences already present in Nature, for example, on the surface of chiral crystals. We have apparently not reached yet a level of understanding of prebiotic chemistry that allows us to decide between alternative explanations. Whatever the specific cause for the initial chiral imbalance, and for its amplification to homochirality, life itself is preserving the once developed dominance of one sense of handedness through the eternal process of reproduction through self-replication.
References 1. Losick, R., Watson, J.D., Baker, T.A.,
2.
3.
4.
5.
6. 7.
Bell, S., Gann, A., and Levine, M.W. (2008) Molecular Biology of the Gene, Pearson/Benjamin Cummings, San Francisco. Meierhenrich, U. (2008) Amino Acids and the Asymmetry of Life, Springer, Berlin. Lam, H., Oh, D.-C., Cava, F., Takacs, C.N., Clardy, J., de Pedro, M.A., and Waldor, M.K. (2009) d-amino acids govern stationary phase cell wall remodeling in bacteria. Science, 325, 1552–1555. Polak, E.H., Fombon, A.M., Tilquin, C., and Punter, P.H. (1989) Sensory evidence for olfactory receptors with opposite chiral selectivity. Behav. Brain Res., 31(3), 199–206. Fahlbusch, K.-G., Hammerschmidt, F.-J., Panten, J., Pickenhagen, W., Schatkowski, D., Bauer, K., Garbe, D., and Surburg, H. (2002) Flavors and fragrances, Ullmann’s Encyclopedia of Industrial Chemistry, Wiley-VCH Verlag GmbH, Weinheim. Calvin, M. (1969) Chemical Evolution, Oxford University Press, London. Siegel, J.S. (1998) The homochiral imperative of molecular evolution. Chirality, 10, 24–27.
8. Mason, S. (1985) Chemical evolution:
9.
10.
11.
12.
13. 14.
15.
origin of biomolecular chirality. Nature. 314, 400–401. Blackmond, D.G. (2010) The origin of biological homochirality. Cold Spring Harbor. Perspect. Biol., 2, a002147. Cronin, J.R. and Pizarello, S. (1997) Enantiometric excesses in meteoritic amino acids. Science, 275, 951–955. Bonner, W.A. (1994) Enantioselective autocatalysis, spontaneous resolution and the prebiotic generation of chirality. Orig. Life Evol. Biosph., 24, 63–78. Bailey, J. (2001) Astronomical sources of circularly polarized light and the origin of homochirality. Orig. Life Evol. Biosph., 31, 167–183. Brown, R.H. (1985) Amino acid dating. Origins, 12(1), 8–25. Florkin, M. (1969) in Organic Geochemistry – Methods and Results (eds G. Eglinton and M.T.J. Murphy), Springer-Verlag, New York and Amsterdam, pp. 498–520. Bada, J. and Miller, S.L. (1985) The composition of the primitive atmosphere and the synthesis of organic compounds on the early Earth. Workshop on the Early Earth: The Interval from Accretion to the Older Archean (eds L.D. Ashwal and K. Burke) (SEE N85-33062 21-91), Lunar and Planetary Institution, Houston, pp. 8–10.
841
842
23 Life’s Single Chirality: Origin of Symmetry Breaking in Biomolecules 16. Miller, S.L. and Orgel, L.E. (1974) The
17.
18.
19.
20.
21.
22.
23.
24.
25.
Origins of Life on Earth, Prentice Hall Inc., Englewood Cliffs, NJ. Wesendrup, R., Laerdahl, J.K., Compton, R.N., and Schwerdtfeger, P. (2003) Biomolecular homochirality and electroweak interactions. I. The Yamagata hypothesis. J. Phys. Chem. A, 107, 6668–6673. Tranter, G.E. (1987) Parity violation and the origins of biomolecular handedness. Biosystems, 20, 37–48. Rubinstein, I., Clodic, G., Bolbach, G., Weissbuch, I., and Lahav, M. (2008) Racemic beta-sheets as templates for the generation of homochiral (isotactic) peptides from aqueous solutions of (RS)-valine or –leucine N-carboxyanhydrides: relevance for biochirogenesis. Chem. Eur. J., 14, 10999–11009. Beardsley, R.E. (1962) Amino acid cross resistance in Agrobacterium tumefaciens. J. Bacteriol., 84, 1237–1240. Blackmond, D.G. (2004) Asymmetric autocatalysis and its implications for the origin of homochirality. Proc. Natl. Acad. Sci. U.S.A., 101, 5732–5736. Lohr, A. and W¨urthner, F. (2008) Evolution of homochiral helical dye assemblies: involvement of autocatalysis in the ‘‘majority rules’’ effect. Angew. Chem. Int. Ed., 120(7), 1252–1256. Rivera Islas, J., Lavabre, D., Grevy, J.M., Lamoneda, R.H., Cabrera, H.R., Micheau, J.C., and Buhse, T. (2005) Mirror symmetry breaking in the Soai reaction: a kinetic understanding. Proc. Natl. Acad. Sci. U.S.A., 102, 13743–13748. Kitamura, M., Suga, S., Niwa, M., and Noyori, R. (1995) Self- and non-self recognition of asymmetric catalysts. Nonlinear effects in the amino alcohol-promoted enantioselective addition of dialkylzinc to aldehydes. J. Am. Chem. Soc., 117, 4832–4842. Puchot, C., Samuel, O., Dunach, E., Zhao, S., Agami, C., and Kagan, H.B. (1986) Non-linear effects in asymmetric synthesis. J. Am. Chem. Soc., 108, 2353–2357.
26. Frank, C.F. (1953) On spontaneous
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
asymmetric synthesis. Biochim. Biophys. Acta, 11, 459–464. Kondepudi, D.K. and Nelson, G.W. (1983) Chiral symmetry breaking in non-equilibrium systems. Phys. Rev. Lett., 50, 1023–1026. Gutman, I., Todorovic, D., Vuckovic, M., and Jungwirth, P. (1992) Modelling spontaneous chiral stereoselection: the Frank mechanism with racemization. J. Chem. Soc., Faraday Trans., 88, 1123–1127. Avetisov, V. and Goldanskii, V. (1996) Mirror symmetry breaking at the molecular level. Proc. Natl. Acad. Sci. U.S.A., 93, 11435–11442. Plasson, R., Bersini, H., and Commeyras, A. (2004) Recycling Frank: spontaneous emergence of homochirality in noncatalytic systems. Proc. Natl. Acad. Sci. U.S.A., 101, 16733–16738. Mauksch, M., Tsogoeva, S.B., Wei, S., and Martynova, I. (2007) Demonstration of spontaneous chiral symmetry breaking in asymmetric Mannich and aldol reactions. Chirality, 19, 816–825. Boiteux, A. and Hess, B. (1981) Design of glycolysis. Philos. Trans. R. Soc. London Ser. B., 293, 5–22. Sugimoro, T., Hyuga, H., and Saito, Y. (2008) Fluctuation induced homochirality. J. Phys. Soc. Jpn., 77, 064606. Ribo, J.M. and Hochberg, D. (2008) Stability of racemic and chiral steady states in open and closed chemical systems. Phys. Lett. A, 373(1), 111–122. Blackmond, D.G. (2009) Challenging the concept of ‘‘recycling’’ as a mechanism for the evolution of homochirality in chemical reactions. Chirality, 21, 359–362. Mauksch, M. and Tsogoeva, S.B. (2008) Spontaneous emergence of homochirality via coherently coupled antagonistic and reversible reaction cycles. ChemPhysChem, 9, 2359–2372. Nicolis, G. and Prigogine, I. (1981) Symmetry breaking and pattern selection in far-from-equilibrium systems. Proc. Natl. Acad. Sci. U.S.A., 78(2), 659–663. Iwamoto, K. (2003) Spontaneous appearance of chirally asymmetric steady
References
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
states in a reaction model including Michaelis-Menten type catalytic reactions. Phys. Chem. Chem. Phys., 5, 3616–3621. Qian, H. (2006) Open-system nonequilibrium steady state: statistical thermodynamics, fluctuations, and chemical oscillations. J. Phys. Chem. B, 110, 15063–15074. Huber, C. and W¨achtersh¨auser, G. (1998) Peptides by activation of amino acids with CO on (Ni,Fe)S surfaces: implications for the origin of life. Science, 281, 670–672. Crusats, J., Hochberg, D., Moyano, A., and Ribo, J.M. (2009) Frank model and spontaneous emergence of chirality in closed systems. ChemPhysChem, 10, 2123–2131. Seelig, F.F. (1971) System-theoretic model for the spontaneous generation of optical antipodes in strongly asymmetric yield. J. Theor. Biol., 31, 355–361. Decker, P. (1975) Evolution in bioids: hypercompetivity as a source of bistability and a possible role of metal complexes as prenucleoprotic mediators of molecular asymmetry. Orig. Life, 6, 211–218. Soai, K., Shibata, T., Morioka, H., and Choji, K. (1995) Asymmetric autocatalysis and amplification of enantiomeric excess of a chiral molecule. Nature, 378, 767–768. Schiaffino, L. and Ercolani, G. (2009) Amplification of chirality and enantioselectivity in the asymmetric autocatalytic Soai reaction. ChemPhysChem, 10(14), 2508–2515. Bailey, P.D. (1995) On the self-replication of chirality. J. Chem. Soc., Chem. Commun., 1797–1798. Blackmond, D.G., McMillan, C.R., Ramdeehul, S., Schorm, A., and Brown, J.M. (2001) Origins of asymmetric amplification in autocatalytic alkylzinc additions. J. Am. Chem. Soc., 123, 10103–10104. Buono, F.G. and Blackmond, D.G. (2003) Kinetic evidence for a tetrameric transition state in the asymmetric autocatalytic alkylation of pyrimidyl aldehydes. J. Am. Chem. Soc., 125, 8978–8979.
49. Barabas, B., Caglioti, L., Micskei, K., and
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
Palyi, G. (2009) Data-based stochastic approach to absolute asymmetric synthesis by autocatalysis. Bull. Chem. Soc. Jpn., 82, 1372–1376. Mislow, K. (2003) Absolute asymmetric synthesis: a commentary. Collect. Czech. Chem. Commun., 68, 849–863. Wynberg, H. (1989) Asymmetric autocatalysis: facts and fancy. J. Macromol. Sci. A, 26(8), 1033–1041. Mauksch, M., Tsogoeva, S.B., Martynova, I.M., and Wei, S. (2007) Evidence of asymmetric autocatalysis in organocatalytic reactions. Angew. Chem. Int. Ed., 46, 393–396. Pizzarello, S. and Weber, A.L. (2004) Prebiotic amino acids as asymmetric catalysts. Science, 303, 1151. Cordova, A., Sunden, H., Xu, Y., Ibrahem, I., Zou, W., and Engquist, M. (2006) Sugar-assisted kinetic resolution of amino acids and amplification of enantiomeric excess of organic molecules. Chem. Eur. J., 12, 5446–5451. Nanita, S.C. and Cooks, R.G. (2006) Serine octamers: cluster formation, reactions, and implications for biomolecule homochirality. Angew. Chem. Int. Ed., 45, 554–569. Mauksch, M., Wei, S.-W., Freund, M., Zamfir, A., and Tsogoeva, S.B. (2010) Spontaneous mirror symmetry breaking in the aldol reaction and its potential relevance in prebiotic chemistry. Orig. Life Evol. Biosph., 40(1), 79–91. Eigen, M. and Schuster, P. (1978) The hypercycle. A principle of natural self-organization. Part B: the abstract hypercycle. Naturwissenschaften, 65, 7–41. Quayle, J.M., Slawin, A.M.Z., and Philp, D. (2002) A structurally-simple minimal self-replicating system. Tetrahedron Lett., 43, 7729–7733. von Kiedrowski, G.A. (1986) Self-replicating hexadeoxynucleotide. Angew. Chem. Int. Ed., 25, 932–935. Kamioka, S., Ajami, D., and Rebek, J. (2009) Synthetic autocatalysts show organocatalysis of other reactions. Chem. Commun., 7324–7326.
843
844
23 Life’s Single Chirality: Origin of Symmetry Breaking in Biomolecules 61. Green, M.M., Park, J.-W., Sato, T.,
62.
63.
64.
65.
66.
67.
68.
69.
Teramoto, A., Lifson, S., Selinger, R.L.B., and Selinger, J.V. (1999) The macromolecular route to chiral amplification. Angew. Chem. Int. Ed., 38, 3138–3154. Haq, S., Liu, N., Humblot, V., Jansen, A.P.J., and Raval, R. (2009) Drastic symmetry breaking in supramolecular organization of enantiomerically unbalanced monolayers at surfaces. Nat. Chem., 1, 409–414. Brandenburg, A., Lehto, H.J., and Lehto, K.M. (2007) Homochirality in an early peptide world. Astrobiology, 7(5), 725–732. Plasson, R. and Bersini, H. (2009) Energetic and entropic analysis of mirror symmetry breaking processes in a recycled microreversible chemical system. J. Phys. Chem. B, 113, 3477–3490. Hayashi, Y., Matsuzawa, M., Yamaguchi, J., Yonehara, S., Matsumoto, Y., Shoji, M., Hashizume, D., and Koshino, H. (2006) Large nonlinear effect observed in the enantiomeric excess of proline in solution and that in the solid state. Angew. Chem. Int. Ed., 45, 4593–4597. Klussmann, M., Iwamura, H., Mathew, S.P., Wells, D.H., Pandya, U., Alan Armstrong, A., and Blackmond, D.G. (2006) Thermodynamic control of asymmetric amplification in amino acid catalysis. Nature, 441, 621–623. Breslow, R. and Levine, M.S. (2006) Amplification of enantiomeric concentrations under credible prebiotic conditions. Proc. Natl. Acad. Sci. U.S.A., 103, 12979–12980. Soloshonok, V.A. (2006) Remarkable amplification of the self-disproportionation of enantiomers on achiral-phase chromatography columns. Angew. Chem. Int. Ed., 118, 780–783. Albrecht, M., Soloshonok, V.A., Schrader, L., Yasumoto, M., and Suhm, M.A. (2009) Chirality-dependent sublimation of α-(trifluoromethyl)-lactic acid: Relative vapor pressures of racemic, eutectic, and enantiomerically pure forms, and vibrational spectroscopy of isolated (S,S) and (S,R) dimers. J. Fluor. Chem., 131, 495–504.
70. Thiemann, W. and Teutsch, H. (1990)
71.
72.
73.
74.
75.
76.
77.
78.
79.
Possible amplification of enantiomer excess through structural properties of liquid crystal – a model for origin of optical activity in the biosphere. Orig. Life. Evol. Biosph., 20, 121–126. Perry, R.H., Wu, C., Nefliu, M., and Cooks, R.G. (2007) Serine sublimes with spontaneous chiral amplification. Chem. Commun., 1071–1073. Hazen, R.M., Filley, T.R., and Goodfriend, G.A. (2001) Selective adsorption of l- and d-amino acids on calcite: implications for biochemical homochirality. Proc. Natl. Acad. Sci. U.S.A., 98, 5487–5490. Kuhn, A. and Fischer, P. (2009) Absolute asymmetric reduction based on the relative orientation of achiral reactants. Angew. Chem. Int. Ed., 48, 6857–6860. Chenchaiah, P.C., Holland, H.L., and Richardson, M.F. (1982) A new approach to the synthesis of chiral molecules from nonchiral reactants. Asymmetric induction by reaction at one surface of a single (nonchiral) crystal. J. Chem. Soc., Chem. Commun., 436–437. Addadi, L., Berkovitch-yellin, Z., Weissbuch, I., Lahav, M., and Leiserowitz, L. (1983) Morphology engineering of organic crystals with the assistance of ‘‘tailor-made’’ growth inhibitors. Mol. Cryst. Liq. Cryst., 96, 1–17. Noorduin, W.L., van der Asdonk, P., Meekes, H., van Enckevort, W.J.P., Kaptein, B., Leeman, M., Kellogg, R.M., and Vlieg, E. (2009) Complete chiral resolution using additive-induced crystal size bifurcation during grinding. Angew. Chem. Int. Ed., 48, 3278–3280. Pincock, R.E., Perkins, R.R., Ma, A.S., and Wilson, K.R. (1971) Probability distribution of enantiomorphous forms in spontaneous generation of optically active substances. Science, 174, 1018–1020. Kondepudi, D.K., Kaufman, R.J., and Singh, N. (1990) Chiral symmetry breaking in sodium chlorate crystallization. Science, 250, 975–976. Havinga, E. (1954) Spontaneous formation of optically active substances. Biochim. Biophys. Acta, 13, 171–174.
References 80. Kostyanovsky, R.G., Kostyanovsky,
81.
82.
83.
84.
85.
86.
87.
V.R., Kadorkina, G.K., and Lyssenko, K.A. (2001) Wedekind-Fock-Havinga salt Me(Et)N+ (AlI)PhI– * CHCl3 as historically the first object for absolute asymmetric synthesis: spontaneous resolution, structure and absolute configuration. Mendeleev Commun., 11, 1–42. Crusats, J., Veintemillas-Verdaguer, S., and Rib´o, J.M. (2006) Homochirality as a consequence of thermodynamic equilibrium? Chem. Eur. J., 12, 7776–7781. Asakura, K., Nagasaka, Y., Osanai, S., and Kondepudi, D.K. (2005) Kinetic model for chiral symmetry breaking transition in growth front of conglomerate crystal phase. J. Phys. Chem. B, 109, 1586–1592. Viedma, C. (2005) Chiral symmetry breaking during crystallization: complete chiral purity induced by nonlinear autocatalysis and recycling. Phys. Rev. Lett., 94(6), 065504. Uwaha, M. (2004) A model for complete chiral crystallisation. J. Phys. Soc. Jpn., 73, 2601–2603. Blackmond, D.G. (2007) ‘‘Chiral Amnesia’’ as a driving force for solid-phase homochirality. Chem. Eur. J., 13, 3290–3295. Noorduin, W.L., Izumi, T., Millemaggi, A., Leeman, M., Meekes, H., van Enckevort, W.J.P., Kellog, R.M., Kaptein, B., Vlieg, E., and Blackmond, D.G. (2008) Emergence of a single solid chiral state from a nearly racemic amino acid derivative. J. Am. Chem. Soc., 130, 1158–1159. Noorduin, W.L., Vlieg, E., Kellogg, R.M., and Kaptein, B. (2009) From Ostwald ripening to single chirality. Angew. Chem. Int. Ed., 48, 9600–9606.
88. Uwaha, M. (2008) Simple models for
89.
90.
91.
92.
93.
94.
95.
96.
chirality conversion of crystals and molecules by grinding. J. Phys. Soc. Jpn., 77, 083802–083805. Tsogoeva, S.B., Wei, S., Freund, M., and Mauksch, M. (2009) Generation of highly enantioenriched crystalline products in reversible asymmetric reactions with racemic or achiral catalysts. Angew. Chem. Int. Ed., 48, 590–594. Wei, S., Mauksch, M., and Tsogoeva, S.B. (2009) Autocatalytic enantiomerisation at the crystal surface in deracemization of scalemic conglomerates. Chem. Eur. J., 15, 10255–10262. Saito, Y. and Hyuga, H. (2008) Chiral crystal growth under grinding. J. Phys. Soc. Jpn., 77, 113001–113004. Flock, A.M., Reucher, C.M.M., and Bolm, C. (2010) Enantioenrichment by iterative retro-aldol/aldol reaction catalyzed by an achiral or racemic base. Chem. Eur. J., 16, 3918–3921. Viedma, C., Ortiz, J.E., de Torres, T., Izumi, T., and Blackmond, D.G. (2008) Evolution of solid phase homochirality for a proteinogenic amino acid. J. Am. Chem. Soc., 130, 15274–15275. Gayathri, V.S. and Rao, M. (2007) Fluctuation-induced chiral symmetry breaking in autocatalytic reaction-diffusion systems. Europhys. Lett., 80, 28001–28006. Scolnik, Y., Portnaya, I., Cogan, U., Tal, S., Haimovitz, R., Fridkin, M., Elitzur, A.C., Deamer, D.W., and Shinitzky, M. (2006) Subtle differences in structural transitions between poly-l- and poly-d amino acids of equal length in water. Phys. Chem. Chem. Phys., 8, 333–339. Toxvaerd, S. (2009) Origin of homochirality in biosystems. Int. J. Mol. Sci., 10, 1290–1299.
845
847
Part VI Conclusion: From Natural Facts to Chemical Fictions
Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
849
24 Artifacts and Natural Substances Formed Spontaneously Pierre Champy
M’enfin! Gaston Lagaffe In the supreme horror of that second I forgot what had horrified me, and the burst of black memory vanished in a chaos of echoing images. Howard Phillips Lovecraft Ceci n’est pas une pipe. Ren´e Magritte 24.1 Introduction
A definition of natural product artifacts can be easily constructed on etymological foundations. The word ‘‘artifact (/artefact)’’ originates from Latin: its prefix ‘‘art-’’ arises from ‘‘ars, artis,’’ a very general word for ‘‘way of being,’’ a meaning that latter evolved to ‘‘ability acquired through study or practice.’’ ‘‘Art-’’ thus embraced, in numerous variations, ‘‘human activities tending toward order.’’ Complemented with the suffix ‘‘-fact’’ (‘‘facere’’ – to do), it expressed the concept of craft, with positive (‘‘artifex’’ – profession) or negative (‘‘artificium’’ – cunning) aspects. ‘‘Artificial,’’ ‘‘artis factum,’’ thus in opposition with ‘‘natural,’’ was taken up in medical English as ‘‘artifact’’ to designate a living tissue alteration provoked by scientific intervention [1]. ‘‘Artifact’’ now bears several meanings in the common language or within specific lexical fields. For the anthropologist, an artifact is ‘‘a human-made object that gives information about the culture of its creator and users,’’ or ‘‘a product transformed by Man, distinguishable from another provoked by a natural phenomenon.’’1) For any scientist, an artifact is an ‘‘undesired alteration in data, introduced by a technique and/or technology.’’ In our context, ‘‘data’’ will be transposed to a molecular level, and considered through the natural products chemist’s 1) Note that both definitions can apply to
extraction artifacts. Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
850
24 Artifacts and Natural Substances Formed Spontaneously
looking glass. After this primary focus, precision is needed. Indeed, both ‘‘ars’’ and ‘‘factum’’ shall prevent us from considering some natural products as artifacts: • Inconstant metabolites (e.g., inducible metabolites or trace compounds arising from unspecific metabolism) can barely be called artifacts. • Taxonomic mistakes prior to chemical study induce artifactual results, mainly at a chemotaxonomic level (Figure 24.1). Metabolites originating from symbiotic or pathogen microorganisms may also cause chemotaxonomical misinterpretations [2].2) • Similarly, isolating a phthalate or describing the trifluoroacetic acid (TFA) salt of an alkaloid as a new natural compound is a virtual artifact only.3) We will consider here an artifact as a natural product chemically altered while manipulated, with the exception of deliberate derivatization and degradation procedures. This molecule can be modified in its matrix or under purified form. The event leading to its formation can take place as soon as harvest, treatment, storage, and processing of the producer organism. It more frequently occurs during the extraction phase or isolation process. Purely analytical artifacts, though commonly observed, will not be dealt with in this chapter. An artifact is not always identified as such. Its true nature may be evidenced on the basis of several clues: • appearance during isolation, generally noticed by TLC or HPLC-UV/MS; the precursor is sometimes monitored; • absence when a new procedure is applied (e.g., solvent change); • obtention of a racemic form or of several enantiomers when enzymatic, stereospecific biosynthesis is expected; • an unexpected oxidation state for the molecule; • a dubious structure regarding biosynthetic origin or a lack of match with chemotaxonomic standards; • work with a notoriously unstable group of compounds; • critical analysis of the protocol used (i.e., pH and redox conditions, light exposure, energy transfer, time spent in solution, etc.) or inappropriate storage and shelf-life. In case of doubt, authors frequently try to investigate the artifactual nature of the compounds they isolate (modification of isolation protocol; deliberate conversion of presumed precursor; etc.). Such demonstrations will be presented, if available, for the examples displayed throughout this chapter. In some interesting cases, a molecule can definitely look like an artifact but prove to be natural. Small molecules isolated from the bryozoan Biflustra perfragilis (Membraniporidae; methanol, chloromethane, dichloromethane, methanethiol, 2) Marine products are often subject to cau-
tion, and endophytic compounds obtained during the study of higher plants are now largely considered.
3) Several examples can be found in the recent
literature.
O
O OH OH
Figure 24.1
Possible or proven botanical misidentifications [3, 4].
O
NO2 Aristolochic acid I (2)
COOH
highly nephrotoxic upon chronic exposure
O
O
In the 1990s, Aristolochia fangchi roots (Aristolochiaceae) were mistaken with those of Stephania tetandra (Menispermaceae) because of near-homonymy in Chinese, and incorporated in weight loss dietary supplements.
O
O
"To our knowledge this is the first report of acetogenins occurring in the Vitaceae", state Pettit et al. Isolation of (1), highly interesting from a chemotaxonomical viewpoint, is indeed peculiar and requires confirmation, as Annonaceous acetogenins were thought to be exclusively produced in the unrelated Annonaceae taxa. The plant material was apparently poorly identified: Was the right species harvested? [J. Nat. Prod. 2008, 71, 130-133]
22-Epicalamistrin B (1)
OH
O
"This is not an acetogenin of Annonaceae" Ampelocissus sp., Vitaceae, root
24.1 Introduction 851
852
24 Artifacts and Natural Substances Formed Spontaneously
dimethyl sulfide, dimethyl disulfide) have been identified as genuine products rather than laboratory contaminants [5]. Nevertheless, many spontaneously self-assembled compounds may constitute borderline cases. Examples have been reviewed by Gravel and Poupon [6]. Artifacts of natural molecules are feared and abhorred by the common pharmacognosist. Although often perceived as nebulous entities, they are frequently encountered [7],4) , 5) echoing throughout the halls of literature: the keywords ‘‘artifact/artefact’’ yield about 2000 answers in a general search on the website of the journal Phytochemistry. An extensive review of published data is therefore not proposed here. However, relevant, generally recent examples were chosen, so as to give an overview of the generation of artifacts in regard to the reactivity of natural products. They will be presented according to their main causes (enzymatic, oxidative, photo- and thermally-induced processes, basic and acidic-catalysis, working solvents), and, secondarily, stigmata. 24.2 Glucosidases as Triggers for Formation of By-products
Some plant secondary metabolites are classified as phytoanticipins. When plant tissue in which they are present is disrupted they are bio-activated by the action of enzymes – generally β-glucosidases. These binary systems – two sets of components that when separated are relatively inert – provide plants with an immediate chemical defense against herbivores and pathogens [17]. Hydrolysis products can be viewed as artifacts (e.g., aldehydes produced by hydrolysis of cyanogenic glycosides and spontaneous release of HCN [18]). They often are reactive and can spontaneously rearrange, as in the case of ranunculine (3), a hemiterpenic glucoside from Ranunculaceae (Ranunculus, Clematis spp.). It yields protoanemonine (4), a reactive compound susceptible to nucleophilic attacks and responsible for contact dermatitis in humans (Scheme 24.1) [19]. Glucosinolates are inert thio-S-(β-d-glucopyranosyl)-(Z)-N-hydroximinosulfate esters with a variable side chain. They are found almost exclusively in Brassicaceae spp. Upon injury of plant tissues they are hydrolyzed by myrosinases (β-thioglucoside hydrolases) to yield unstable thiohydroxamate sulfonates that spontaneously rearrange to insecticidal isothiocyanates or to other compounds, depending mostly on pH conditions. Glucosinolates occur in many vegetables, and are of interest for human health: by-products and the conditions of their 4) Notably, R. Verpoorte and coworkers re-
cently published a review about extraction artifacts due to commonly used solvents [7]. 5) Notably, the whole history of culinary traditions is marked by chemical processing in search of palatability or taste [8], yielding ‘‘artifactual’’ molecules (e.g., Maillard reaction [9], fermentation of Camellia sinensis (Theaceae) leaves [10–12], and interactions of polyphenols
during wine maturation [13]). Medical traditions include preparation methods aimed at equilibrating the risk/benefit ratio of herbal remedies (e.g., oxidation of anthrones glycosides of bark of Rhamnus purshiana (Rhamnaceae) [14]; hydrolysis of norditerpenic alkaloid esters of Aconitum spp. tuber and roots (Ranunculaceae) [15, 16]).
24.3 Oxidation Processes
HO HO
OH O O
β-D-glucosidase hydrolysis upon mechanical injury
O
OH
HO
O
O
O spontaneous dehydration
Ranunculine (3)
allergizing, toxic
O O
O
Protoanemonine (4) Scheme 24.1
O
O
O
Anemonine (5)
Hemiterpenes from Ranunculaceae that cause irritation.
formation (including cooking conditions) have thus been extensively studied and reviewed (Scheme 24.2) [20]. 24.3 Oxidation Processes
Spontaneous oxidative processes are common causes of artifact formation. The newly formed compounds gain stability, with extended conjugation. These artifacts frequently occur during the isolation process and can often be efficiently tracked down. Many marine products are involved, some appearing to be unstable in atmospheric rather than aqueous environment, especially when exposed to light. However, plant products of various biogenetic origins also yield examples, as in the aforementioned case of Rhamnus hydroxyanthracenes (see footnote 5). Oxidized positions are quite various, and it can be noticed that enzymatic intervention would seem likely in several cases. Therefore, authors sometimes appear to be very suspicious about artifactual origins, and discussions are frequent (Scheme 24.3). 24.3.1 Thiol Oxidation
The numerous oxidation and rearrangement products of 2-propenesulfenic acid (9) and of allicin (10), spontaneously formed from 9 after enzymatic hydrolysis of (+)-S-2-propenyl-l-cysteine-S-oxide (8, a phytoanticipin from garlic, Allium sativum, Asparagaceae, ex-Alliaceae; Scheme 24.4), have been reviewed extensively in an elegant and comprehensive article by Erick Block, who raises the following question in regard to this horde of organic sulfur compounds: ‘‘What is a natural product?’’ [22].6) 6) Block and, more recently, Rose et al. also
present compounds from other Alliaceae species. [22].
853
Scheme 24.2
R
Rearrangement products following hydrolysis of glucosinolates.
Spontaneous rearrangements pH > 8
S
N
thiocyanates
21, 425-447; Trends Plant Sci. 2006, 11, 1360-1385 ]
S C isothiocyanates main insecticidal compounds pH 6-7 R N spontaneous cyclisation R S C R R S NH R SH N −H2SO4 Glc O OH β-OH-isothiocyanates pH 3-4 N N OSO − S OSO3− N R 3 unstable -S nitriles 5-oxazolidine-2-thiones enzymes? glucosinolates thiohydroxamateO-sulfonates n N epithionitriles S unstable enzymes? [Nat. Prod. Rep. 2004,
Myrosinase hydrolysis upon mechanical injury
854
24 Artifacts and Natural Substances Formed Spontaneously
24.3 Oxidation Processes
Thorecta choanoides (sponge, Dictyoceratida: Thorectidae)
OH
O OCH3
sesquiterpene quinol 6
in vivo
OCH3
7 OH atmospheric conversion:
O
unsuccessful 6/7: Typical redox coenzyme
no ratio proposed
Artifactual nature discussed: "Attempts to convert 6 to 7 through aerial oxidation proved unsuccessful. Despite this, it could not be entirely discounted that 7 was an oxidative artifact of 6 brought on during storage and/or handling." Is arguing really necessary? Scheme 24.3
Oxidized coenzyme Q-3 (7): the redox status quandary [21].
Allium sativum, Alliaceae
O S
NH2
alliinase
COOH
(+)-S-2-Propenyl-Lcysteine-S-oxide (8)
spontaneous
SOH +
COOH
2-Propenesulfenic acid (9)
x2
−H2O
O SS Allicin (10)
Scheme 24.4
NH2
Too numerous to mention! "Depending on the Allium species, under differing conditions, thiosulfinates can decompose to form [...] diallyl, methyl allyl, diethyl mono-, di-, tri-, tetra, pentaand hexasulfides, vinyldithiins and ajoenes. " [Nat. Prod. Rep. 2005, 22, 351–368]
Allicin (10) as the first step in a long series of artifacts.
Unrelated species are similarly problematic; for example, oxidation and rearrangement of the volatile compounds in the pungent (and thus controversial) durian fruit, Durio zibethinus, Malvaceae (ex-Bombacaceae/Durioniaceae) [23]. Dithiorhodysinin disulfide (12), a furanosesquiterpene isolated from the nudibranch Ceratosoma brevicaudatum, was reported to arise from a thiol monomer (11) upon exposure to air. Compound 11 was previously obtained from the sponge (and prey) Dysidea avara, but not 12. It is hypothesized that 12 might originate from the nudibranch’s diet, corresponding to an in vivo storage form rather than to an artifact (Scheme 24.5) [24].
855
856
24 Artifacts and Natural Substances Formed Spontaneously
sponge: Dysidea avara
H
diet/ sequestration
O
HS
nudibranch: Ceratosoma brevicaudatum
O
H S S
H
in vivo?
O
ox.
H
H 11
H
11
Dithiorhodysinin disulfide (12) spontaneous quantitative dimerization during extraction/purification?
Scheme 24.5
Ambiguous fate of a marine thiol group.
Pyrroloiminoquinolines from a sponge of the Latrunculia genus (Latrunculiidae) yielded another example of such dimerization. Discorhabdin B (13) was shown to undergo degradation upon light exposure. Among formed products, discorhabdin W (14) was identified as a disulfide dimer of 13; compound 14 converts, under reducing conditions, into the unstable monomer 15, which in turn spontaneously cyclizes to form 13 (Scheme 24.6) [25]. Latrunculia sp. (sponge)
H N
O H N S N Br
O
H O N
H N
hn
H N
S S N Br
N Br O
O
dithiothreitol
O Discorhabdin B (13)
H N
Discorhabdin W (14) other unidentified degradation products H N
O
H N SH
N Br O unstable intermediate (15) (detected by LC-MS) spontaneous Scheme 24.6
Dimerization and recovery of discorhabdin B (13).
24.3 Oxidation Processes
24.3.2 Oxidation Processes of Oxygenated Functions
Oxidation of conjugated oxygenated positions can occur within a crude extract, but are more frequently observed during isolation. A precursor is sometimes obtained along with the newly formed metabolite, as in the case of clathrins B (16) and C (17), from an Australian sponge (Clathria sp.). Here, the authors clearly identified the oxidation process of (16). Complete conjugation is remarkable for clathrin C (17) (Scheme 24.7) [26].
OH
Clathrin B (16) optically active, stereochemistry undetermined because of rapid degradation air
O Clathrin C (17) formation observed during isolation Scheme 24.7
Clathrins from Clathria sp.
Oxidation of hydroperoxide-bearing carbons is also exemplified in the literature, with an obvious gain of stability for the artifactual derivatives. Among clerod-3,4-ene diterpenes found in Aristolochia spp. (Aristolochiaceae), two presenting a peroxide function at C2 (18 and 20) were isolated, accompanied by carbonyl analogs (19 and 21) (Scheme 24.8). Compounds 18 and 20 were shown to undergo rapid decomposition to the corresponding enones 19 and 21, respectively, suggesting that 2-oxo-3-clerodenes could be artifacts [27]. This might indeed be the case for 2-oxopopulifolic acid (22) [28] and rel-(5S,8R,9S,10R)-2-oxo-ent-3-cleroden-15-oic acid (23) [29], the precursors of which were not obtained. However, apparent decomposition products of hydroperoxides cannot always be regarded as artifacts, especially when the stability of the hydroperoxide permits a full purification process and structural analysis (e.g., musambin A (24) from Markhamia lutea (Bignoniaceae), for which no degradation to musambin C (25) was observed [30]). Lambertellols A (26) and B (27) were isolated from the filamentous fungus Lambertella corni-maris under conditions of co-culture with the pathogenic fungus Molinia fructigena, along with lambertellin (30). Isomerization between 26 and 27 was observed during structural studies. It occurred under mild conditions:
857
H COOCH3
O
H COOCH3
H COOCH3
Scheme 24.8
COOH
H
COOCH3
HO COOH
H
COOH
H COOH
Artifacts?
Musambin C (25) 20 mg
H H
O
enone
rel-(5S,8R,9S,10R )-2-Oxo-ent3-cleroden-15-oic acid (23)
O
A. brasiliensis
oxidation during processing/isolation or normal biogenetical filiation?
H OH
H
2-Oxopopulifolic acid (22)
O
A. galeata, A. cymbifera
no degradation observed during isolation/analyzes
Musambin A (24) 200 mg
H H
Possible artifacts from hydroperoxides.
HO
H OH
H
2-Oxopopulifolic acid methyl ester (21)
O
Markhamia lutea, Bignoniaceae, leaves isolation in mild conditions OOH
Hydroperoxy-ent-3-cleroden-15-oic acid methyl ester (20)
HOO
A. esperanzae, roots
Hydroperoxy-ent-3-cleroden-3,132-Oxokolavenic acid methyl ester (19) diene-15-oic acid methyl ester (18) Artifactual oxidation observed
HOO
Aristolochia brasiliensis, Aristolochiaceae, stems
858
24 Artifacts and Natural Substances Formed Spontaneously
24.3 Oxidation Processes
chromatography on silica gel, standing in methanol or within fungus culture medium. Biosynthesis of 26 and 27 is induced in response to acidification of the medium. It is therefore likely that both molecules co-exist in ‘‘real-life’’ conditions. Nevertheless, conversion of 26 and 27 into 30 was observed under the same conditions as isomerization. To explain this interconversion, the authors proposed a retro-Michael type opening of the butenolide ring (26/27) to a carboxylic intermediate (28), followed by oxidation of its semiquinone moiety into a quinone (29). An intramolecular Michael-type ring closure at position C2 furnishes a pyrone, with a final oxidative aromatization to give 30. Interestingly, this artifact (30) is the active substance responsible for growth inhibition of M. fructigena during mycoparasitism by L. corni-maris (Scheme 24.9) [31]. 24.3.3 Newly Oxygenated Products
Spontaneous oxygenations of unsaturated positions due to air exposure are frequently described, and are most likely produced by photocatalyzed free-radical mechanisms. Some compounds proposed as being artifacts may also appear to be enzymatically formed: an epoxy derivative (32) of radicinol (31; Scheme 24.10a), occurring in aging but not in fresh cultures of the phytopathogen Alternaria chrysanthemi (Deuteromycetes), is supposed to be due to oxidation by 1 O2 [32]7) but a change in the metabolic profile of the organism could also explain its appearance. Spontaneous oxygenation can also be quickly followed by dehydration. This pathway can be hypothesized for formation of ochrone B (34), isolated from Coelogyne ochracea (Orchidaceae), along with its 9,10-dihydro analog ochrone A (33). Despite a possible biosynthetic relationship, 34 proved to be an artifact due to aerial oxidation, as it was found in pure 33 after standing for several days (Scheme 24.10b) [33]. Dysidea herbacea (Aplysillidae), a common sponge of tropical seas, allowed isolation of the known furanosesquiterpene herbacin (35) and of a new hydroxylbutenolide 36, the relative stereochemistry of which was further confirmed by conversion of 35 in quantitative yield upon photochemical oxygenation. A complementary search for a putative one-ringed furanic biogenetic precursor of 36 (37) proved negative [34]. Interestingly, the closely related compound 38, previously isolated from Dysidea pallescens, thus appears to be artifactual (Scheme 24.11). During isolation of N-methylwelwitindolinone C isonitrile (39) from the terrestrial cyanobacteria Fischerella muscicola and F. major, formation of three minor oxidized derivatives (42–44) was observed, and could be reproduced under UV irradiation. The authors proposed a pathway from 39, with photocatalyzed peroxidation leading to the hydroperoxide (40), the reduction of which yields 43. The isonitrile group in 43 slowly hydrolyses to give 44 during fractionation; 44 is readily obtained by photocatalysis from pure 43. Alternatively, from 7) The authors did not try to convert isolated
59 into 58.
859
OH
OH
O
OH
OH oxidation
retro-Michael
Michael
28 (not detected)
OH
O O
Scheme 24.9
Isomerization and decomposition of lambertellol.
O
O
OH
Lambertellin (30)
O
identified as responsible for growth inhibition of M. fructigena
29 (not detected )
O
O OH 1. Michael OH 2 2. oxidation O
Silica gel, MeOH, PSA culture medium...
[J.Org. Chem. 2008, 73, 5039-5047]
Lambertellol B (27) Silica gel, MeOH, PSA culture medium...
O
H O
Silica gel, MeOH, PSA medium ...
OH
Michael
retro-Michael
OH O Lambertellol A (26) O
H
O
Lambertella spp.
860
24 Artifacts and Natural Substances Formed Spontaneously
24.3 Oxidation Processes
Alternaria chrysanthemi (Deuteromycetes) aged cultures only
O
O
O OCH3 OH
OCH3 OH
OO
O
O (a)
Epoxy-radicinol (32)
Radicinol (31)
non enzymatic conversion?
Coelogyne ochraceae, Orchidaceae, whole plant
O
O
O 10 Ochrone A (33) (b) Scheme 24.10
HO
OCH3
HO
OCH3 O
9
Ochrone B (34)
spontaneous aromatization after purification
Miscellaneous oxidation artifacts.
Dysidea herbacea, Aplysillidae 1O
O
2
Herbacin (35) Scheme 24.11
O
O
CH3CN light, 500 W, 3h
D. pallescens
O 36
O
OH putative biosynthetic
racemic
precursor 37
O 38
OH artifact?
Artifactual sesquiterpenic lactones from Dysidea spp.
intermediate 40, an intramolecular epoxidation of the chloroalkene group would lead to the putative chloroepoxide 41. Cyclization of the newly formed hydroxyl (positions C3 to C14), with concomitant opening of the epoxide ring (C14/C15), induces formation of a keto group at C15, with Cl as leaving group, to give 428) (Scheme 24.12) [35]. Spontaneous oxygenation of the α-position of carbonyl functions is also stated to be easy by reaction with molecular oxygen from air under various conditions.
8) Alternatively, 42 could be formed by re-
arrangement of the epoxy intermediate (41) to a chloroketone prior to intramolecular cyclization (not shown). According to the authors, this latter route is much
less likely, as chloroepoxides require relatively high temperatures to rearrange to chloroketones. Notably, during fractionation, the authors employed mild conditions.
861
O2, hn
O H (90 W, O 4 d) N
H
Cl H
CN
Scheme 24.12
+H2O
−HCl
OHCHN
O OH 3 O N
H H
observed during isolation
O OH O N slow conversion
H
41
CN
Cl O
Pathways to oxidized welwitindolinones from Fischerella spp.
Cl
O O OH 3 O N
14
both undetected
3-hydroxy-N-methylwelwitindolinone C isonitrile (43)
Fischerella spp. (cyanobacteria)
40
CN
N -methylwelwitindolinone C isonitrile (39)
CN
15
Cl
Cl
CN
H H
[J. Nat. Prod. 1999, 62, 569-572]
Nmethylwelwitindolinone D isonitrile (42) O
O OH O 3-hydroxy-N-methylN welwitindolinone C formamide (44)
H
N
O O
15
O
862
24 Artifacts and Natural Substances Formed Spontaneously
24.3 Oxidation Processes
863
Re-investigation of the roots of Pueraria mirifica (Fabaceae), led to isolation of deoxymiroestrol (45), a potent phytoestrogen, along with the known hydroxylated analog miroestrol (46). The authors proposed miroestrol to be an artifact of oxidation with atmospheric O2 : indeed, they observed facile aerial conversion of deoxymiroestrol (45) into miroestrol (46) and isomiroestrol (47) under mild conditions (methanolic solution, room temperature, one week; Scheme 24.13). The rate of conversion was not given by the authors. Noticeably, 46 showed lower affinity for estrogen receptor than its precursor, challenging conservation of the drug [36].
HO H H OH H
air, MeOH, r.t., 1 week
8
H O HO O7 (+)-Deoxymiroestrol (45) Scheme 24.13
HO H H OH H
14
+
HO H H OH H
14 8
OH OH O HO O O H O (+)-Miroestrol (46) (+)-Isomiroestrol (47)
HO
Artifactual oxidation of deoxymiroestrol from Pueraria mirifica.
Stemona spp. (ex-Roxburghia, Stemonaceae, Dioscoreales) and related genera are a source of alkaloids characterized by a pyrrolo-azepine nucleus. Among lactonic representatives, tuberostemonine (48) and oxo-tuberostemonine (49) were described in the 1970s. With regard to 48, 49 shows relocation of lactone ring from C11 to C1, while keeping the same configuration at C11, and displays a double bond at C9=C9a. The possibility that 49 might be an artifact formed by air oxidation of 48 was proposed, since it was also obtained from 48 by oxidation with mercuric acetate (Scheme 24.14) [37]. However, a biosynthetic mechanism for formation of 49, with a sequential mechanism implicating double bond formation/lactone opening/closure from a putative hydroxylated precursor does not seem impossible. In this way, the semi-synthetic procedure applied for 48 would be biomimetic.
Stemona spp., Stemonaceae
O HO
H 11 H H 99a 1 H HN H O
HO H HgAcO2
O Tuberostemonine (48) Scheme 24.14
N
O O H
artifact?
H O air oxidation?
O Oxo-tuberostemonine (49)
Putative artifactual oxidation of a Stemona alkaloid.
864
24 Artifacts and Natural Substances Formed Spontaneously
24.3.4 Oxidative Coupling
Spontaneous dimerization of alkaloids is described as the result of nucleophilic attack or of oxidative coupling. The artifactual nature of bipowine (51) and bipowinone (52), two rare bis-aporphines from Popowia pisocarpa (Annonaceae) was investigated. If spontaneous oxidation of 51 to 52 was observed, dimerization of their putative precursors wilsonirine (50) and pancoridine (53), respectively, was not. However, the authors were able to obtain 7,7 -bisdehydronorglaucine (56) from dehydronorglaucin (55, from 54) in solution at room temperature (Scheme 24.15). Some bisacridones resulting from a C–C linkage between an aromatic moiety and a terpenic dihydrofuran (e.g., 57–59) [38] or dihydropyran (60) [39] were obtained from Taiwanese Citrus spp. (Rutaceae). Being optically inactive, they were suggested to be either artifacts or non-enzymatically formed native compounds. Nevertheless, bisacridones with ether linkages between isoprene units and phenols were isolated from Glycosmis spp. (e.g., 61, 62) [40]. Extractions were conducted using acetone under long reflux conditions (30–40 h), while authors working with ethanol as an extraction solvent seem to obtain monomers only (e.g., 63, 64) [41]. A Norrish type I reaction of the solvent could be hypothesized as a source of free radicals,9) but, notably, more ‘‘classical’’ artifacts were not obtained (see Section 24.9). However, it can be noticed that the species investigated appear to have intense oxidative metabolism, with frequent isolation of biscoumarins. A parallel can be drawn with bis-ether flavanolignans from milk thistle (Silybum marianum, Asteraceae): silybin (65) and isosilybin (66) are mixtures of trans diastereoisomers, which exemplify non-stereospecific biosynthetic radical coupling (Figure 24.2). It is thus difficult to conclude whether achiral bisacridones from Rutaceae are artifacts or chemomarkers of part of this family. 24.3.5 The N-Oxide and Oxoalkaloid Cases
Alkaloids of diverse classes are encountered in their N-oxide forms, in various taxa (isoquinolines from Magnoliales, indolomonoterpenes from Gentianales, pyrrolizidines from Asterales, various alkaloidic skeletons from Rutales, etc.), but can be considered collectively. The N-oxides are generally minor compounds in comparison to their counterparts. To our knowledge, there is no clade-specialization for N-oxygenation in higher plants. Such N-oxides can sometimes be artifacts due to air oxidation – especially in the case of poor storage conditions. If such compounds appear to be artifactual, most authors reckon their formation is biosynthetically deliberate, and might be linked to plant stress-conditions. Indeed, N-oxygenation is catalyzed by classical monooxygenases. Notably, mamallian phase I metabolites produced by liver CYPs sometimes are N-oxides [42]. The artifactual nature of a 9) Photochemical cleavage of ketone into two
free radical intermediates.
HO
H3CO
Scheme 24.15
H3CO
OCH3
NH
N
7
NH
[J. Nat. Prod. 1986, 49, 1028-1035]
2 OCH3 7,7'-Bisdehydronorglaucine (56)
H3CO
H3CO
spontaneous dimerization in MeOH/CH2Cl2, r.t.
Possible artifactual formation of bis-aporphines.
Norglaucine (54)
N
H3CO OCH3 OCH3 O
spontaneous 7-7' dimerization: not observed
O N H3CO OCH3 OCH3 OCH3 Pancoridine (53) Bipowinone (52)
Dehydronorglaucine (55)
H3CO
H3CO
OCH3
1) NH3CO chlorosuccinimide NH 2) sodium H3CO ethanolate
H3CO
H3CO
O
H3CO
Sequence 1 spontaneous oxidation: not observed
OH HN OCH3 OCH3 Bipowine (51)
7 7'
OCH3 OCH3 NH
H3CO
Sequence 1
H3CO
Sequence 1
NH
OCH3 Wilsonirine (50)
H3CO
HO
H3CO
spontaneous oxidation in DMSO, CHCl3, CH2Cl2, r.t., 24 h
Popowia pisocarpa, Annonaceae
24.3 Oxidation Processes 865
O
N R3
R O 1OH
O
OH
O
Glycofolinine (63)
O
HO O
Figure 24.2
HO
O O
O
OH O
N
N
O
OH
O
HO
O
O
N
N
O
OH
O
OH
O
O OH
O
OH
[Chem. Pharm. Bull. 2004, 52, 362-364]
OH
O
Buntanbismine C (60) [Phytochemistry 1996, 1, 221-223]
O
[a]D not measured; Configuration undetermined. → to be regarded as an artifact?
Achiral: Radical coupling?
O
OH
Isosilybin A (66a) Isosilybin B (66b) Isosilybin (66) - diastereoisomeric pair
O
O
O
N
N OH O OH
O
OH O Glycobismine-G (62) OH O
O
O HO Glycobismine-F (61)
OH OH OH Silybin A (65a) Silybin B (65b) Silybin (65) - diastereoisomeric pair
O
O
OH
Achiral bisacridones from Rutaceae.
OH O
O
Silybum marianum, Asteraceae
O
Acrifoline (64)
N
O
N
[J. Nat. Prod. 1995, 58, 1629-1631]
HO
O OH
O OH
Glycosmis citrifolia, Rutaceae
Glycosmis citrifolia
Infra-familial specificity is likely
O
Citrus grandis f. buntan, stem bark
Trans Achiral: In plantae dimerization or artifact?
[Chem. Pharm. Bull. 1995, 43, 1340-1345]
O
R2
R1
R2 R3 H H O OH Citbismine A (57) OH Citbismine B (58) OMe OH H Citbismine C (59) OMe OH Me N O
Citrus grandis, C. paradisi, Rutaceae, roots
866
24 Artifacts and Natural Substances Formed Spontaneously
HN
H
hn [O]
Scheme 24.16
H
H N H H H N
N
H
O
H N
H
H2O
(+)-C-curarine I (73)
H
N
N
N N-oxide form (70)
H
Oxidized indolomonoterpenes.
(−)-C-dihydrotoxiferine (72)
N
H N H H HN H H
N
N Bisnordihydrotoxiferine (69)
H
H N H H H N
O N
(under dry form)
COOCH3 hn [O]
H
Strychnos spp., Loganiaceae peroxides from aged ethyl-ether N
Vobparicine (67) N H
N H
N4
Tabernaemontana chippii, Apocynaceae
N
enhanced toxicity in mammals!
oxidation products predominate in traditional curare preparations
[Nat. Prod. Commun. 2009, 4, 447-454]
N carbinolamine form (71)
O
H N H H N H H
N H
[J. Nat. Prod. 1985, 48, 400-423]
(+)-C-calebassine I (74) [Hesse M., 2002]
N
apparently non artifactual
Vobparicine-N4-oxide (68)
COOCH3
H
H N OH H H H HO N H H
N
OH
N
HN
N H H H N H H
N H
N H
O N4
24.3 Oxidation Processes 867
868
24 Artifacts and Natural Substances Formed Spontaneously
trace indolomonoterpenic N-oxide from Tabernaemontana chippii (Apocynaceae) root bark was discussed as follows: ‘‘Whether or not vobparicine-N4-oxide (68) is an artifact is not clear. The facts suggest that it may be a genuine product because, following column chromatography, 68 was isolated from a different fraction than vobparicine (67) and also fractions containing vobparicine never showed any accompanying spot of the N4-oxide’’ (Scheme 24.16) [43]. However, it was shown that some N-oxidized indolomonoterpenes such as 70 were of artifactual origin, due to peroxides formed in diethyl ether after long exposure of the solvent to air and light [7]. Note that spontaneous oxidation of indolomonoterpenes can also take place at other positions, as observed for conopharyngine (75; Tabernaemontana spp. and other Apocynaceae) and its 3-hydroxy analog 77, which, respectively, yield hydroxyindolenine derivatives 76 and 78 upon standing for a long time (Scheme 24.17) [44].
3
R
minor artifacts, apparition on long standing of pure 75 and 77
N
H3CO N H3CO H H3COOC
H3CO
Conopharyngine (75): R=H 3R/S-Hydroxyconopharyngine (77): R=OH Scheme 24.17
HO
H3CO
7
R
N
N H3COOC
Conopharyngine-hydroxyindolenine (76): R=H 3R/S-Hydroxyconopharynginehydroxyindolenine (78): R=OH
Formation of hydroxyindolenine derivatives.
Oxoalkaloid derivatives raise similar questions about their true nature. Their artifactual formation can be very swift, as in the air oxidation of vasicoline (79) to vasicolinone (80) during fractionation or in its purified form (Scheme 24.18) [45]. Oxoaporphines and dioxoaporphines seem to appear spontaneously during long and uncareful storage of extracts or of isolated compounds, with an extension
Adhatoda vasica, Acanthaceae
full conjugation of quinazoline system
O standing in air for a few hours
N N Vasicoline (79) Scheme 24.18
N
N
Vasicolinone (80) Oxidation of vasicoline (79).
apparition during isolation of 79
N N
OCH3
O
7
OCH3
CH3
OCH3
O
CH3
O
O
N
CH3
OCH3 Corunnine (86)
H3CO
O
H3CO
[Accounts Chem. Res. 2006, 39, 293-300]
Pontevedrine (85)
H3CO
N CH H CO 3 3
N
O H3CO
O O
[O]
Dioxoglaucine (84)
H3CO
H3CO
H3CO
N
OCH3 Deshydroglaucine (82)
H3CO
[O]
[O]
H3CO
H3CO
Scheme 24.19
In plantae oxidative pathways toward oxoaporphinoids.
The "origin of oxoaporphines" puzzle: Monooxygenation/oxidation? H2O2 and other peroxides under cellular oxidative stress conditions? Peroxides as solvent impurities? Air oxidation?
3
N CH
Oxoglaucine (83)
H3CO
H3CO
H3CO
5
N CH3 6a H
4
OCH3 S(+)-Glaucine (81)
H3CO
H3CO 1
H3CO
24.3 Oxidation Processes 869
870
24 Artifacts and Natural Substances Formed Spontaneously
of their conjugated domains. However, most of the derivatives obtained probably originate from a specific oxidation by light and singlet oxygen prior to isolation, as part of plant defense against pathogens, with some oxoalkaloids being proposed as phototoxic phytoalexins. For example, upon mechanical injury of Magnoliales or Ranunculales, the aporphine glaucine (81) is rapidly oxidized into oxoglaucine (83) (major product), ponteverdine (85) and other oxidized derivatives (82, 84, 86; Scheme 24.19) [46]. In planta spontaneous oxidation of enzymatic monooxygenation products can also be hypothesized. It was noticed that dioxoglaucine (84) affords an iminium (i.e., 6,6a-dehydro-84) under UV irradiation. Similar examples of formation of iminium ions are proposed below.
24.4 Exposure to Light
As previously seen, light is a necessary condition for oxygenation by 1 O2 to give detectable artifacts. Oxidative light degradation is sometimes a quickly occurring phenomenon, as in the case of rotenoids [47].10) Many different compounds have been already shown to undergo conversion and degradation by various photochemical reactions. Photochemical epimerization of stereogenic centers and isomerization of double bonds are also observed in several natural series. 24.4.1 Isomerization and Epimerization
A remarkable instance is that of the essential oil of aniseed fruit (Pimpinella anisum L., Apiaceae), which is used in aromatherapy and in the food industry. It contains 80–95% (E)-anethole (87), which is poorly toxic. However, hepatic toxicity of the (Z)-isomer (88) was observed in humans.11) Both drug and essential oil must be stored away from light, as (E)-anethole isomerizes into its (Z)-isomer under UV irradiation [14]. Stilbenoids are also well known for undergoing light-induced isomerization. The olefinic photo-isomerization process, in the excited state, involves twisting (about the former double bond) of stilbene fragments relative to one another. The (E)-form is favored, as in resveratrol (90), in regard to (Z)-form (89). Combretastatin A4 (91), bearing a (Z)-configuration, is a cytotoxic derivative that inspired semi-synthetic derivatives undergoing clinical 10) This was studied with the neurotoxic pes- 11) The European Pharmacopoeia sets an up-
ticide rotenone, a prenylated isoflavonoid mainly produced by Derris, Milletia, Tephrosia spp. (Fabaceae). Rotenone is a champion of photochemical degradation: It undergoes O-demethylation, epimerization, epoxidation, hydroxylation, and dehydration [47].
per limit of 0.5% (Z)-anethole in the essential oil, and the acceptable limit for dietary intake is 2.5 mg kg –1 According to Bruneton, a heavy drinker of aniseed flavored beverages is likely to ingest more than 250 mg total anethole a day! [14].
24.4 Exposure to Light
Essential oils of Pimpinella anisum, Foeniculum vulgare (Apiaceae), Illicium verum (Illiaceae)
Trans-anethole (E-anethole, 87) LD50, per os, Mouse : O 2-3 g/kg
Cis-anethole (Z-anethole, 88)
light
O
LD50: 0.24 g/kg
Vitis vinifera, Vitaceae; Polygonum cuspidatum, Polygonaceae... favoured E-isomers
HO
OH
light
HO OH OH
OH
Cis-resveratrol (89)
major isomer
phytoalexin various biological activities
Trans -resveratrol (90)
Combretum cafrum, Combretaceae inhibitor of tubuline polymerization
H3CO H3CO
OCH3
Combretastatin A4 (91)
OH OCH3 light
native isomer
OH OCH3 H3CO H3CO Scheme 24.20
OCH3
inactive isomer
E-combretastatin A4 (92)
Light-induced isomerization of anethole and stilbenoids
trials. It is readily isomerized to its inactive (E)-form 92 (Scheme 24.20) [48]. Many other cases exist in various chemical series, some being hypothetical but likely artifacts in regard to biosynthetic processes and known analogs (e.g., histrionicotoxins from South American frogs (Dendrobates spp.) with trans-enynes [49]). Epimerization processes are also observed following photo-excitation. Oxidation of reserpine (93) under light exposure leads to the iminium 3-dehydroreserpine (94), and also yields the 3-(S) isomer isoreserpine (95). The reaction is favored in chloroformic solutions [7]. (+)-Milnamide A (96), a metabolite of marine sponges
871
872
24 Artifacts and Natural Substances Formed Spontaneously
Rauvolfia serpentina, Apocynaceae
Various marine sponges, e.g.: Auletta sp.
OCH3 H O
3N
H3CO
O
N H H H H3COOC
OCH3
N
OCH3
Reserpine (93) (isolated)
N
N H O
COOH
N
(+)-Milnamide A (96)
light
H O
N
O N H H H3COOC OCH3 3-Dehydroreserpine (94) + H3CO
H O
N H3CO
O H
OCH3
N H H H H3COOC
O
light
OCH3 OCH3 OCH3
H N
OCH3 OCH3 OCH3
OCH3
slow
O
N N H O
N
COOH
(+)-Milnamide D (97)
"true" natural product?
Isoreserpine (95) Scheme 24.21
Reserpine (93) and milnamide A (96): photoinduced artifacts.
(e.g., Auletta spp.), was shown to auto-oxidize to (+)-milnamide D (97), suggesting a non-biogenetic origin for this compound. The rate of oxidation was similarly accelerated in aged CHCl3 and CDCl3 solutions (Scheme 24.21) [50]. 24.4.2 Rearrangements
Rearrangements due to UV exposure in the course of purification of plant metabolites can be exemplified with sesquiterpenes such as ent-bicyclogermacrene (98), which converts into ent-spathulenol (99) after peroxidation by atmospheric oxygen (Scheme 24.22) [51]. A complex example was provided by Majetich et al., who studied the conversion of (+)-komaroviquinone (102) into (+)-komarovispirone (103). Both compounds had been previously isolated from Dracocephalum komarovi (Lamiaceae), a plant used in Uzbek traditional phytotherapy. It appears that 103 is rapidly formed upon light exposure of 102. The authors proposed a radical mechanism (Scheme 24.23), and suggested formation from plant material [52], that is, a true artifactual nature for 103 [53].12) 12) Note that the authors underline the anal-
ogy with light-induced rearrangement of (+)-verbenone into (+)-chrysantenone.
Scheme 24.22
guaiyl cation (101)
Biosynthetic pathway
HO H
H
HO
H
H
[Phytochemistry 2001, 41, 1347-1350]
(−)-ent-spathulenol (99)
Photoinduced oxidation and rearrangement of ent-bicyclogermacrene (98).
H germacrene A (100)
(−)-ent-bicyclogermacrene (98)
H
HO H
[H]
DicranoIejeunea yoshinagana, Lejeuneaceae (bryophyte) total conversion
H
allowed to stand exposed to air, r.t., 4 days 1O H 2 hn OO
24.4 Exposure to Light 873
O
O O hn
HO
OH
O
Scheme 24.23
O OH
isolation artifact
O
O 6
9
O
O
OH
O
O OH
7
O 20 9
H
H
O
O
7
O
O
O OH
20 H6
O
H
O
O
Photoinduced rearrangement of komaroviquinone (102).
common acids and bases (+)-Komarovispirone (103)
O
O
hn, r.t., C6H12 1h: 90%
O H OH (+)-Komaroviquinone (102)
stable in the dark
Dacrocephalum komarovi, Lamiaceae
OH
O
O
O
[Org. Lett. 2008, 10, 89-91]
O
7
O
O H
O H6
H
O
O
874
24 Artifacts and Natural Substances Formed Spontaneously
O N hn
O
H N
Scheme 24.24
O
O
N
N O
O
O
N
HN O N H Brevianamide D (107)
hn
HN
O
O Brevianamide C (106)
H N
O Brevianamide B (105) (minor)
N
Photoinduced rearrangements of brevianamide A (104).
O
hydrogen transfer
H HN
N H HN
O
bond rupture
thermal treatment
[Tetrahedron 1972, 28, 2999-3008; adapted from: Eur. J. Org. Chem. 2008, 27-42]
Norrish-type cleavage
O HN O Brevianamide A (104)
H N
Penicillium brevi-compactum
24.4 Exposure to Light 875
876
24 Artifacts and Natural Substances Formed Spontaneously
Brevianamides A (104), B (105), C (106), and D (107) were isolated from Penicillium brevi-compactum. A thorough experimental showed that 105–107 are photochemical isomerization or cleavage products of 104, probably resulting from a Norrish-type reaction (Scheme 24.24). These compounds were not isolated when the fungus was cultivated in the dark and when the purification procedure was conducted with low light [54]. 24.4.3 Photocycloaddition and Photodimerization
Biomimetic synthesis literature offers delightful accounts of auto-assemblies under UV irradiation, the [2 + 2] photocycloaddition of α, β-unsaturated ketones or esters to alkenes or alkynes being largely used [6, 55]. Natural products obtained using this strategy do not necessarily correspond to artifactual compounds, but rather illustrate peculiarities in the field of biosynthesis. Reactions of α, β-unsaturated ketones can be induced by simple daylight absorption [55]. Lactones and pyrones are thus considered as ‘‘photoactive,’’ and artifactual intra/homo/hetero-photocycloadditions of natural products often implicate such moieties. In addition, artifactual cyclizations of polyenes are reported. Generally, these reactions can take place without a solvent, but acetone can act as a sensitizer [55]. From the African shrub Ethulia vernonioides (Asteraceae), two terpenic 5-methylcoumarins, 108 and 109, were isolated. Compound 109, a hexacyclic compound, was evidenced to be an easily formed artifactual intramolecular cycloadduct of 108 (Scheme 24.25) [56].
Ethulia vernonioides (Hoenalia vernonioides), Asteraceae, aerial parts observed during optical rotation [2+2] cycloaddition measurement of 108 at λ<540 nm O O C6H6, O O O O H 6′ O hn, 3h 7′ 6′
O
5′ 10′O
O
4
O Hoenalia coumarin (108) Scheme 24.25
7′ 3
O
10′
O
10′
O O Cyclohoenalia coumarin (109)
Intramolecular [2 + 2] cycloaddition of ‘‘hoenalia coumarin’’ (108).
Numerous marine organisms have yielded cycloaddition products. Their artifactual nature outside of the marine environment is hard to discriminate. Sharma et al. explored the photoinduced auto-cyclization of polypropionates from shell-devoid molluscs of the order Sacoglossa, studying the influence of UV absorption by sea-water on the rate of formation of tridachiahydropyrone (112) and of phototridachiahydropyrone (113) from a putative precursor 110. Note that 112 only is a natural product. Compound 113 was the photochemically preferred product in the
O
O O air /water
b : UV
MeOH
a : hn
+
O +
O
112 → 113 air: complete at 13h sea water: 45% only
O O O O O O 111 Tridachiahydropyrone (112) Phototridachiahydropyrone (113) unreported compound
O
hn [1-3]-sigmatropic rearrangement
Scheme 24.26
Influence of medium on photocatalyzed synthesis of marine polypropionates.
a : natural light, 3 days b :UV light, 125 W, 27 h Yield Ratio (110/111/112/113) 75% conversion, 20:7:1 (111/112/113) path length: no strong effect prolonged exposure: complete formation of 113 air 70% 1 : 2 : 2 : 3.5 sea water 58% 1 : 1.7 : 1 : 0.3 [Synlett. 2010, 4, 525-528]
supposed precursor (110)
Possible in natural habitat?
natural daylight
Trichodachia crispata (marine mollusc)
24.4 Exposure to Light 877
878
24 Artifacts and Natural Substances Formed Spontaneously
total synthesis of 112 from 111, and has not been described in a natural source so far. Its formation appears to be partly inhibited in sea-water,13) which is in keeping with its absorption spectrum (Scheme 24.26) [57]. From the bark of an Aniba sp. (Lauraceae), the styrylpyrone 114 was obtained, along with two dimers (115, 116) whose origin was uncertain, as Lauraceae spp. are prone to oxidative coupling. Found in much lower amounts than 114, both 115 and 116 were shown to be photoinduced dimerization artifacts (Scheme 24.27) [58].14) Biyouyanagin A (120) and B (121) are minor compounds isolated from Hypericum chinense (Clusiaceae). Their total synthesis was performed by Nicolaou et al., through [2 + 2] photocycloaddition, from ent-zingiberene (118) and hyperolactone C (119) under conditions compatible with extraction conditions (Scheme 24.28) [59]. Biyouyanagins 120 and 121 were not demonstrated to appear during purification, and are likely to form in planta without enzymatic intervention from readily available precursors. They do though appear as potential artifacts. Finally, orinocin (124), a short pyrone isolated from light-exposed cultures of Streptomyces orinoci, was linked to the aureothin family of Streptomyces metabolites: indeed, an isomer of neoaureothin (122) was also obtained, and two relatively stable diastereoisomers (123a,b) were identified in light-working conditions only. Pure 122 was exposed to daylight, to partially afford 123a,b by a 8π − 6π electrocyclic rearrangement cascade. This step was followed by photoinduced retro-[2 + 2]-cycloaddition, giving 124. This entire artifactual sequence was termed ‘‘polyene splicing’’ (Scheme 24.29) [60]. The literature sometimes depicts the photochemical synthesis of natural cycloadducts with protocols that are far from real-life15) conditions (monochromatic light, high intensity irradiation, sensitizers, etc.) [61]. However, in several cases, compounds are readily formed: determination of their genuine natural character requires specifically conducted studies, and is a matter of debate.
24.5 Heat and Pressure
Heat and pressure often provide the energy necessary for artifactual formation through various mechanisms, including oxidation processes. 13) Owing to the presence of organic com- 15) Or ‘‘real-lab’’ – from an extractor’s perspec-
pounds; distillated water had no major influence. 14) To our knowledge, this is not described from the well-studied Piper methysticum (kawa, Piperaceae).
tive.
O
Scheme 24.27
O H CO 3
O
Styrylpyrone photodimers.
H3CO
H3CO
styrylpyrone 117
O
OCH3
no homo or heterodimer identified
O
O
H3CO
H3CO H3CO
O
H3CO O
O
O
OCH3
OCH3
light
O
O O OCH3
O
O
O
O
OCH3
O
O
O
OCH3
OCH3
OCH3
OCH3
styrylpyrone-dimer 116
H3CO
H3CO
light
OCH3 OCH3
H3CO
O
O
OCH3
styrylpyrone-dimer 115 = Aniba dimer-A (A.gardneri)?
H3CO
H3CO
H3CO
dimerization reactions could be performed from 114, using a UV lamp
H3CO
H3CO
styrylpyrone 114 OCH3 (methoxy-yangonine)
Aniba sp., Lauraceae, bark
24.5 Heat and Pressure 879
880
24 Artifacts and Natural Substances Formed Spontaneously
To our knowledge, the authors did not express concern about the hypothetic artifactual nature for 149 and 150.
H
H
H
O
H +
O
O O
Ent-zingiberene (118)
Hyperolactone C (119)
H H H
H
O
H O Biyouyanagin A (120, 51%)
H
O
HO H
O
O
Biyouyanagin B (121, 3%)
CH2Cl2, hn, 320 nm, sentisizer (2'acetophenone), 5°C, 12 h [Chem. Eur. J. 2010, 16, 7678-7682] Scheme 24.28
Biomimetic total synthesis of biyouyanagins A (120) and B (121).
24.5.1 Epimerization and Isomerization In or Out of Solutions
Seeds of Zizyphus vulgaris var. spinosus (Zizyphus jujuba var. spinosa, Rhamnaceae) are traditionally used in China and in the Middle East as a sedative remedy. Cyclopeptides such as sanjoinine A (frangufoline, 125) were identified as active principles, with GABA (γ -aminobutyric acid) agonist activity. Chinese materia medica describes roasting of this material to improve hypnotic activity. Compound 125 was shown to isomerize at its exo-amino acid (dimethyl-phenylalanine) to give sanjoinine Ah1 (126), its heat induced artifact, which exerts a more potent pharmacological activity in vivo, probably gaining affinity for GABA receptors (Scheme 24.30) [62]. Epimerization at C2 of the major catechins in a crude extract of green tea (Camellia sinensis) was investigated in aqueous solutions. The epimers of (−)-epicatechin and (−)-derivatives (e.g., 127, 128) can be observed under conditions of tea brewing (90 ◦ C, 3 min, tea sample). After 30 min, about 40% conversion is observed (Scheme 24.31) [63]. Such epimerization processes are thus expected for related molecules, casting doubt on the stereochemistry of tannins or comparable compounds obtained at boiling temperature or using a Soxhlet apparatus. 24.5.2 Hydrodistillation
Calorific energy provided by hydrodistillation can be a cause of rearrangement of natural volatile compounds. Confirmation of artifactual nature is not easy, as gas chromatography conditions are likely to induce the same reactions. Hydrodistillation of the aerial parts of Laggera tomentosa (Asteraceae) was shown to yield (−)-chrysanthenone (134), (−)/(+)-filifolone (132/135) and (Z)-isogeranic acid (133), all likely to be artifacts. The main constituent of the oil, (+)-chrysanthenone
O
6
O
Scheme 24.29
O
O2N
light: partial conversion of 122
2 cis-diastereoisomers
OCH3 O2N
H
O O2 N
light
OCH3
O
O orinocin (124)
O
6
mesitylene
OCH3
O
"polyene splicling"
[Phytochemistry 2009, 70, 1833–1840]
retro-[2+2]-cycloaddition
(123a, 123b)
H
O
conrotational 6p-8p electrocyclic rearrangements
123, 124: formation upon light exposure of cultures only
Non enzymatic ‘‘polyene splicing.’’
neoaureothin isomer (122)
O2N
O
Streptomyces orinoci
24.5 Heat and Pressure 881
OO
Scheme 24.30
di-Me-Phe
N
NH (R )-D-
Sanjoinine Ah1 (126) [Pure Appl. Chem. 1989, 61, 443-448]
Effects on hexobarbital-induced sleeping time in mice (1mg/kg, i.p.) Control: 18 min NH 125: 26 min 126: 33 min
improved sedative activity
N
Traditional roasting of Zizyphus jujuba seeds.
hypnotic, GABA agonist
OO HN HN O
O
epimerization occurring during roasting, but not with hot aqueous extraction
Sanjoinine A (125)
HN HN (S )-LO
O
Zizyphus jujuba, Rhamnaceae, seeds
882
24 Artifacts and Natural Substances Formed Spontaneously
O
2-S
native
OH OH
OH
Scheme 24.31
Epimerization of gallocatechins during tea brewing.
5 min 60 min 30°C: 0.2% 1.6% 60°C: 2.2% 18.1%
[J. Agric. Food Chem. 2003, 51, 510-514]
OR OH (+)-derivatives(129, 130)
HO H2O,
OR OH (-)-derivatives (127, 128)
HO
Epimerization: R = H: (+)-epigallocatechin 129 5 min 60 min OH 30°C: 0.0% 0.1% OH 60°C: 4.1% 28.4% 2-R O R= gallate: OH (+)-epigallocatechin-3-O -gallate 130
Tea (Camellia sinensis aqueous leaves extract)
24.5 Heat and Pressure 883
O
−H+
OH
hydrodistillation or GC analysis
H
O H
+H+
Scheme 24.32
−H+
H2O
COOH
[2 +2] cycloaddition
(Z )-Isogeranic acid (133)
C O
8%
hydrodistillation only
car-3-ene / carane skeleton 137
"termination step [of biosynthetic pathway] involving loss of a proton" [Dewick, P.M. 2009]
racemization [Phytochemistry 2001, 58, 489-492]
Artifacts formed from (+)-chrysanthenone (131) during hydrodistillation.
menthyl /a-terpinyl cation 136
H
3-carene synthase
5%
O
100°C
(+)-Chrysanthenone (131) parent compound, 53%
H (+)-Filifolone (135) (−)-Chrysanthenone (134) O
H
Biosynthetic pathway:
O
92/8 8%
H (−)-Filifolone (132)
H
Laggera tomentosa, Asteraceae, essential oil
884
24 Artifacts and Natural Substances Formed Spontaneously
24.5 Heat and Pressure
(131) is believed to be their precursor, undergoing thermal and solvolytic reactions (Scheme 24.32) [64]. Even though not occurring in the gas phase, hydrolysis of glucosides was observed during hydrodistillation, and can be mentioned here: (Z)- and (E)-5-ethylidene-2(5H)-furanones (142 and 143) were obtained from foliage of Halocarpus biformis (Podocarpaceae) using this technique [65], but were absent from an hexane extract. Aqueous cold extraction yielded two potential glycosylated precursors (138, 139). After treatment with a β-d-glucosidase, they gave the aglycones 140 and 141, which easily dehydrated to 142 and 143 (Scheme 24.33) [66].16) From Osmunda japonica (Osmundaceae, whole plant), (4R,5S)-osmundalactone (145) and 146 (diastereoisomer of 140 and 141) were both identified as artifactual products of the acid hydrolysis of osmundaline (144) during extraction [67]. Hydrated and saturated derivatives 146 and 148 were later isolated, following a non-drastic protocol (Scheme 24.33) [68].17) Notably, dehydration of 146 was not reported. 24.5.3 Decarboxylation Processes
The most famous example of heat-induced decarboxylated products is that of azulenes from Asteraceae, which has impacted the field of European phytotherapy: matricin (149), a sesquiterpenic lactone from Chamaemelum and Chamomilla spp., can be considered as an unstable prodrug, easily yielding the cyclooxygenase (COX) inhibitors 150 and 151 during infusion of the plant material [14]. Hydrodistillation induces similar reactions, which explain the blue color of some Asteraceae essential oils. This case is evocative of the very unstable pyrroloquinoline carboxylic pigment sanguinone A (152), isolated along with sanguinone B (153) from the fruiting bodies of the mushroom Mycena sanguinolenta (Marasmiaceae, Trichomatales); 153 was identified as a highly conjugated decarboxylation artifact from 152, formed under milder conditions than 151 (Scheme 24.34) [69]. 24.5.4 Supercritical CO2
In a convincing chemical study of hop strobiles (Humulus lupulus, Cannabaceae) having undergone previous supercritical CO2 extraction [70], the authors obtained the well known and abundant xanthohumol (154) along with new chalcone derivatives under racemic forms (155–158). These derivatives might be artifacts from 154 formed during pretreatment of the plant material, with probable oxidation to epoxy intermediates that are likely to be unstable under high pressure 16) See compound 3 (Scheme 24.1). 17) The role of glucosidases in the formation
of these molecules appears plausible, and might be artifactual.
885
O
Scheme 24.33
O
O
O
H
H O O 145
H HO
+
acidic hydrolysis
O
H H HO CH 146 3
O
O
HO H H O
147
noticed
O 148 H H HO CH3 + no dehydration
unlikely to dehydrate
H O O Osmundaline (144)
GluO
O
[Phytochemistry 1996, 42, 453-459]
H H [Tetrahedron 1974, 30, 2307-2316] H H HO CH 140 HO CH 141 3 3 [Chem. Pharm. Bull., 1984, 32, 2815-2820]
O
O
O O + H H H H GlcO CH GlcO CH3 3 138 b-D-glucosidase 139
O
extraction : H2O, cold
Osmunda japonica, Osmundaceae
Artifactual lactones of Halocarpus biformis formed during hydrodistillation.
−H2O
O E + Z H CH3 H3C H 142: minor 143: major
O
O
[Phytochemistry 1986, 26, 649-653] hydrodistillation extraction: Halocarpus biformis, C6H12, cold Podocarpaceae, leaves
886
24 Artifacts and Natural Substances Formed Spontaneously
HOOC H N
herbal tea preparation
-AcOH −2 H2O
oxidative decarboxylation -CO2
HN
N
blue
yellow
HN
N
in MeOH or H2O at r.t.
Heat-induced degradation of matricin (149) and sanguinone A (152).
blue
N
N
Chamazulene (151)
absent in fresh crude extract
Chamazulene carboxylic acid (150)
HOOC
- CO2
COX-2 inhibitors
N N N N O O H O H Sanguinone A (152) stable for a few hours only Sanguinone B (153)
Mycena sanguinolenta, Marasmiaceae, HN fruiting body
Scheme 24.34
O H
human metabolism
H OH O Matricin (149) O
Roman chamomile, Chamaelum nobile, Asteraceae
O
24.5 Heat and Pressure 887
888
24 Artifacts and Natural Substances Formed Spontaneously
conditions. Interestingly, some of the compounds were identical to microbial bioconversion products of 154. A cyclic derivative (160) of desmethylxanthohumol was also obtained, but not its parent compound (159). It is also noteworthy that (±)-prenylnaringenin (161), described as an isomerization artifact of 159 [71], was not retrieved (Scheme 24.35) [70]. 24.6 Alkaline Media 24.6.1 Amination Processes
During alkaloid extraction and purification, alkalinization of plant material or of aqueous extracts is performed. Researchers eventually use diluted NaOH, but ammonia is much preferred, with subsequent artifacts identified throughout the literature. The newly formed compounds often have interesting and original structures. For example, from the bark of Desmos dumosus (Annonaceae), desmosine (163) was quantitatively obtained from 5-hydroxy-6,7-dimethoxyflavone (162). The artifactual nature of the newly formed compound, though obvious, was verified by authors by modifying their working conditions [72]. Similarly, from Desmos dunalii, dunaliine A (165) was obtained along with the chalcone 164. The authors did not challenge the artifactual nature of 165, but proposed a pathway from 164 involving a putative, secondary oxidation step (Scheme 24.36) [73]. The secoiridoid base gentianine (167), mainly isolated from Gentianaceae and Loganiaceae spp., similarly arises from treatment with ammonia [74]. After glucose hydrolysis of gentiopicroside (166) or swertiamarine (168),18) nucleophilic addition of NH3 on the newly-formed dialdehyde system, followed by dehydration, generates the pyridine ring of 167 [75].19) Compound 167 was sometimes obtained with carbonates in the alkalinization steps, suggesting it could be a natural compound. Hesse considers that ‘‘it is conceivable that the action of Na2 CO3 might bring about the release of NH3 from certain plant materials’’ [77]. However, 167 and derivatives were isolated in neutral conditions (e.g., Goodeniaceae [78]). A biosynthetic scheme is known, with early introduction of a nitrogen atom in the secoiridoid skeleton (Scheme 24.37) [75]. 18) This reaction was observed with other iri-
doids. 19) Note the importance of spontaneous nucleophilic addition on an aldehyde (and of the Pictet–Spengler reaction) in the non-enzymatic synthesis of alkaloids from biological primary amines ex vivo (e.g., Maillard reactions in meat [20]) [76a] or in vivo. For example, the tetrahydro-isoquinolines
(R/S)-salsolinols were demonstrated to arise from dopamine (a catecholaminergic neurotransmitter) and acetaldehyde (the metabolite of ethanol) in the brain of alcoholics, causing parkinsonism [76b]. Tetrahydro-β-carbolines are similarly formed from serotonin (5-hydroxytryptamin) [76c]. Such reactions can also implicate endogenous carboxylates [76].
OH
OH
(±) OH
Scheme 24.35
OH
OH
O
O
OH
OH
(±) OH
OH O Desmethylxanthohumol J (160)
O
HO
Chalcones obtained from hops after supercritical CO2 treatment.
O
OH OH
[J. Nat. Prod. 2004, 67, 2024-2032]
previously identified as a microbial transformation product of xanthohumol (154)
O O XanthohumolI (158)
HO
(±)
O
OH
(±) OH O (±)-prenylnaringenin (161)
8
most potent phytoestrogen in hop
HO
O O 1",2"- Dihydroxanthohumol C (157)
OH
HO
O
154 + new compounds 155 to 158, 160
epoxy intermediates? OH
O O Xanthohumol (154)
HO
"Spent hop"
Humulus lupulus, Cannabaceae, strobiles Extraction with "classical" isolation procedure supercritical CO2
Artifactual formation described [J. Chromatogr. A. 1999, 832, 97-107]
OH O Desmethylxanthohumol (159)
OH
OH
O O Xanthohumol G (156)
OH
O O Xanthohumol H (155)
HO
HO HO
HO
OH
previously identified as a metabolite of xanthohumol OH (154) in rat feces
24.6 Alkaline Media 889
OH O NH3
O
H3CO
H3CO
Scheme 24.36
O
H O NH2
O
O NH2 Desmosine (163)
H3CO
NH4OH H3CO
Na2CO3
OH NH2 HO
O
O
OH
OH
strangely enough, HO O NH 2 optically active: artifactual Dunaliine A (165) unlikely to be artifactual [Nat. Prod. Res. 2009, 23, 652-658]
H3CO
NH4OH
NH3
OCH3
dihydrochalcone (164) OH O
Desmos dunalii precursor H3CO proposed by authors
Artifactual alkaloidic phenylchromane derivatives from Desmos spp.
[Phytochemistry 1998, 49, 2191-2192]
162
H3CO
H3CO
Desmos dumosus, Annonaceae
H3CO
H3CO
890
24 Artifacts and Natural Substances Formed Spontaneously
O
O tautomerism
OH
O O O NH3
O
nucleophilic addition
Scheme 24.37
O
O H
Gentianine (167) as an artifactual alkaloidic compound.
stereoisomer of iridoidal
H
H
transamination
NH3
OH
OGlc
O
NH2
H H
imine formation [O]
actinide
N
167
adapted from [Dewick, M. P. 2009]
N
H
ring cleavage: seco-serie
N O O NH −H O O 2 O OH O O Gentianine (167) Swertiamarine (168)
OH
O O O O Gentiopicroside hydrolysis (166) (acid or base-catalyzed) Biosynthetic pathway to pyridine ring : iridoid-like
O
OGlc
Gentiana spp., Gentianaceae
sometimes obtained without NH4OH (e.g. Gentiana fetisowii): not always artifactual! Swertzia spp., Loganiaceae
24.6 Alkaline Media 891
892
24 Artifacts and Natural Substances Formed Spontaneously
24.6.2 Other Base-Catalyzed Reactions
Agujasterone C (169), an ecdysteroid bearing a hydroxyl-group at C11, was obtained from Leuzea carthamoides (Asteraceae) along with 5-deoxykaladasterone (170), an analog with a dienone system. As 170 was absent from the crude extracts of the plant, it was identified as an artifactual dehydration product of 169, arising from alumina fractionation of the extract, as shown with pure 169. This finding challenges the status of kaladasterone (172), isolated along with muristerone A (171) from seeds of various Ipomea species (Convolvulaceae) and proposed as being a genuine natural product (Scheme 24.38) [79].20) Leuzea carthamoides, Asteraceae
OHOH
R=H
170 absent from crude extracts (fresh and dry plant)
HO 11 HO HO
OH OH
11
HO
8
OH
6 7
RO
Agujasterone C (169) Ipomea spp., Convolvulaceae R=OH
isolation process using alumina HO
RO
C6H6, alumina, room temperature
Muristerone A (171)
Scheme 24.38
9
?
OH 5-Deoxykaladasterone (170)
Kaladasterone (172)
described as natural
Alumina-catalyzed dehydration of ecdysteroids.
From Delphinium spp. roots (Ranunculaceae), after basic treatment (NH4 OH) for alkaloid extraction, methyllycaconitine (173) by-products (174, 175), resulting from pyrrolidine-2,5-dione aminolysis, were retrieved. Similarly, avhadaridine (177), originating from lycaconitine (176), was also obtained. Surprisingly, Shamma et al. isolated 177 from Delphinium cashmirianum in neutral conditions. They considered the compound as an artifact. Nevertheless, the artifactual nature of its methyl-ester counterpart, cashmiradelphine (178), was not determined: this product was apparently present in the crude ethanolic extract, but could be readily obtained from 176 in methanol at boiling temperature or at room temperature with basic (alumina) or acid catalysis (silica gel) (Scheme 24.39) [80]. Lounasmaa et al. have published a comprehensive review on the artifactual formation of a series of indolomonoterpenic alkaloids from Rauvolfia spp. (Apocynaceae) in the sarpagine series, implicating the Cannizzaro reaction21) that might take place during alkaloid purification steps [81]. For example, O-acetylpreperakine (183) can be expected to originate from the ring-opened form (Z)-vellosimine 20) Note that the authentic nature of other 21) Redox disproportionation of non-enoliz-
9–11 -ecdysteroids has repeatedly been questioned [79].
able aldehydes to carboxylic acids and alcohols, catalyzed by strong bases.
OCH3
O
O
O
H N
H N
O
+
O
Scheme 24.39
N
OO
By-products of imide-bearing aconitine-like alkaloids.
suspected to be natural but obtained by CC on silica gel with MeOH / CHCl3
* SiOH or Al2O3 MeOH, 16h * MeOH, b.t., 1h
?
O
O
no use of NH4OH!
O Lycaconitine (176)
O
Delphinium spp.
NH2 175
O
NH2 174
O
Delphinium tricorne, D. semibarbatum, Ranunculaceae
O Methyllycaconitine (173)
N
OH NH OH 4 O OH O OO OCH3
N
H3CO
OH OCH3 H N
O
Delphinium spp.
H N
O
D. cashmirianum
yield: 100%
OCH3 O Cashmiradelphine (178)
O
NH2 O Avhadaridine (177)
O
24.6 Alkaline Media 893
N H
N
H
N
H H CHO
OCOCH3
H H 20 CHO
17 H CHO
180
Cannizzaro reaction*
CHO
N
H
17
H
H OH
OH
N H
H
HO
H OH
Peraksine (186)
N
H
Artifacts from (Z)-vellomisine (179) via Cannizzaro reactions.
Cannizzaro reaction*
N 21-Epi-dihydro-peraksine (187) H
Rauvolfia caffra, Apocynaceae
Scheme 24.40
179a
H H 21 OH
OH
N H
HN
H CHO
N H
H CHO 20 21 H 181 N
H CHO
N H
17
21
185
N
H
21
H OH
Cannizzaro reaction*
H H CHO
OH
H CHO
N H 182
N
H
strong base: Cannizzaro reaction
N N CH3COOH: H H Dihydroperaksine (184) O -Acetylpreperakine (183) esterification
N
H
Z-Vellosimine (179)
* 17-carboxy-182, 21-carboxy-185, 17-carboxy-187: not observed
N H
H CHO H tautomery OH N
894
24 Artifacts and Natural Substances Formed Spontaneously
24.7 Acidic Conditions during Purifications
(179a). Note the acetylation of 182, a product of the Cannizzaro reaction, during isolation procedures (Scheme 24.40).22) In the ajmaline series, similar reactions were observed (Scheme 24.41). The two series are in close relationship, via interconversions: under hydrolytic conditions (acidic or basic) the acetyl group of vomilenines (188, 190) may be cleaved. In the case of (Z)-vomilenine (190) this leads to deacetyl-(Z)-vomilenine (194), which is in equilibrium with 16-epi-(Z)-vellosimine (195) and (Z)-vellosimine (179) (Scheme 24.42) [81].
24.7 Acidic Conditions during Purifications
Natural products nearly always encounter acidic catalysts in the course of isolation. This is obvious in the case of liquid/liquid extraction steps during purification of alkaloids, which require aqueous acidic media. However, acid-catalyzed reactions can be due to trace impurities in organic solvents. They also might take place on silica gel, as in the retro-Michael conversion of the instable alkaloid stephacidin B (197; Aspergillus ochraceus) into its monomer avrainvillamide (196) (Scheme 24.43) [82]. The various classes of acid-induced artifacts observed are summarized below, and in Section 24.8 on protic solvents. 24.7.1 Epimerization
Acidic conditions can promote epimerization. An illustration published recently for a p-hydroxybenzyl-substituted furanic ring in the stilbene tetramer kobophenol A (Caragana sinica, Fabaceae) challenges the conservation of medicinal extracts [83]. Similarly, radix of Asarum heterotropoides (Aristolochiaceae), a Chinese medicinal plant, contains furofuranic lignans, among which sesamin (198) and asarinin (199) are identified as main active principles. Variable amounts of these molecules are obtained using different extraction methods, and several reports indicate that 198 epimerizes to 199 under acidic conditions, with ring opening similar to that previously shown. Sesamin (198) is supposed to be more stable, as it bears two substituents in exo positions, against one in exo and the other in endo positions for 199. A study of the epimerization process was published, showing an unexpected ratio of 198 and 199 (Scheme 24.44) [84]. The lignan cubebin (200), isolated from Aristolochia lagesiana and Aristolochia pubescens (Aristolochiaceae) provides a similar example. It was shown to be a mixture of 200/201 in a 3 : 2 proportion, suggesting the presence of two epimers (at C9 and/or C8) or a conformational equilibrium (Scheme 24.45). It is well known that lactols can undergo ring-opening and closing in solution: acid-catalyzed ring-opening of 200, in CDCl3 solution, leads to inversion of the stereogenic 22) Use of 10% CH3 COOH.
895
N
H
OH
N
H H CHO
Scheme 24.41
H N
never detected
188b
N
H3COCO N
N 190
H3COCO
N
H OH
N
H H CHO H 20 N CHO H Pekarine (191) 20-Epi-perakine (192) Cannizzaro reaction
H3COCO
Recyclization (attack from the b-side)
H H CHO N CH2OH Raucaffrinoline (193)
H3COCO
NH CHO
N Z-Vomilenine (190a)
H3COCO NH CHO slow
N 188a
H3COCO
Formation of raucaffrinoline (193) via the Cannizzaro reaction.
N 19-Epi-pekarine (189)
H3COCO
N E-Vomilenine (188)
H3COCO
Rauvolfia biauriculata, Apocynaceae
896
24 Artifacts and Natural Substances Formed Spontaneously
24.7 Acidic Conditions during Purifications
acidic or basic catalysis
O O
H N
N
H2O
OH
HO
-COOCH3
Z-Vomilenine (190)
H N
OH
N Deacetyl-Z-vomilenine (194)
OHC H N
OH
N H 16-epi-Z-vellosimine (195) Scheme 24.42
CHO H OH N N H Z-vellosimine (179)
Artifactual relationship between the ajmaline/sarpagine series.
center at C9. An intermediary aldehyde was detected by NMR under appropriate conditions (temperatures above 26 ◦ C, long period of storage in CDCl3 solution). However, it is not possible to establish whether both anomers are natural compounds [85]. 24.7.2 Hydrolysis
Ester hydrolysis is sometimes observed in relatively mild conditions.23) Bromophenols coupled to pyroglutamic acids 202 and 203 were isolated from the red alga Rhodomela confervoides (Rhodomelaceae). Compound 203 is a possible artifact since its conversion from 202 was observed on standing in aqueous methanol (Scheme 24.46). However, the methylation of 203 did not occur upon heating a methanolic solution either with or without silica gel at 45 ◦ C for 48 h [87]. Artifactual hydrolysis in a heterosidic series generally occurs in strong acidic conditions. However, the quinolinic alkaloid 205, from the marine cyanobacterium Lyngbya majuscula, is supposed to be an artifact from its glycosidic counterpart 204, possibly catalyzed by silica gel (Scheme 24.47a) [88]. Extraction with hot water (or protic solvents) is suspected to induce hydrolysis of salicylic glucosides, as evidenced with Populus deltoides (Salicaceae) (Scheme 24.47b) [89]. Freeze-drying (i.e., low pressure) in external flasks, without temperature control, caused the same phenomenon in fresh Salicaceae leaves, due to thawing [90].24) 23) Note
also the occurrence of transacetylation reactions on vicinal diol systems, observed in various solvents (e.g., salvinorins D and E [86]). 24) In comparison to air-dried leaves, the content in phenolic glycosides (e.g.,
salicortin and 2 -cinnamoylsalicortin) dropped by about 50%, giving rise to the normally undetectable aglycon salicin.
897
N
O N
N O
O O
OH N
N O
NHH O O H N
N
O
N O
O O O
Stephacidin B (197) on silica gel.
N
O
O N
N
O O
OH N O Stephacidin B (197)
NHH O N H
Spontaneous dimerization Michael addition
Silica gel chromatography : Retrodimerization
O Avrainvillamide (196)
NH H H H N
Scheme 24.43
O O
N
O
Michael addition
898
24 Artifacts and Natural Substances Formed Spontaneously
H
O
O
O
H
O
O Asarinin(199)
O
O H O
O
Which one is the artifact?
Scheme 24.44
H
H R R R1 O H H H HH
RR H O H H R1
R1= methylenedioxyphenyl, R = furan moiety
H H R1 O H R R HH
Epimerization of furofuranic lignans under various extraction conditions.
Asarum heterotropoides var. mamdshuricum, Aristolochiaceae
H
more stable in silico
O Sesamin (198)
O
O
ratio 198/199 1:1 Extraction by soaking in acetone, room temp.: Vacuum distillation (10mmHg,140°C, 15min): 1:2 Steam distillation (110°C, 3h): 1:2 198 or 199 in EtOH/HCl (10%), boiling temp. 4.5:5.5* (*equilibrium at 1.5h) +
24.7 Acidic Conditions during Purifications 899
900
24 Artifacts and Natural Substances Formed Spontaneously
HO H H 8 9O
O O
8'
H O
3:2
O
H
in CDCl3
O O
(8R,8'R,9R )-cubebin (200) Scheme 24.45
H H OH O
O
O
(8R,8'R,9S )-cubebin (201)
Cubebin anomers from Aristolochia spp.
Rhodomela confervoides
O
Br Br
O Br
N
HO
O OH
O
Br H2O / MeOH 3d., r.t.
OH
N
HO
O
202
OH
203
extraction with EtOH; fractionation on silica gel, EtOAc /MeOH 1:0 to 0:1
MeOH with or without silica gel, 45°C, 48h Scheme 24.46
Methyl-ester hydrolysis.
artifact?
O HO (a)
HO
O
O O
N H3O+? 204
N 205
Lyngbya majuscula (cyanobacterium) hot water extraction
CH2OH HO O O OH HO OH Salicoside (206)
CH2OH OH
artifact?
Salicylic alcohol (207)
(b)
Populus deltoides, Salicaceae
Scheme 24.47
Hydrolysis of heterosides.
24.7.3 Other Acid-Catalyzed Reactions
Acid-catalyzed formation of iminiums from norrhoeadines (208) via acetal hydrolysis, followed by imine formation and dehydration, is particularly well known (Scheme 24.48) [91].
24.7 Acidic Conditions during Purifications
RO NH RO
H3O+
RO
O 208
RO
NH CHO
RO
OR
HO RO
209
OR
OR −H2O
RO N RO
OR
OR red iminium salt 210 Scheme 24.48
Norrhoedanines in acidic conditions.
Several indoloquinoline-type alkaloids were found in the African shrub Cryptolepis sanguinolenta (Apocynaceae, ex-Asclepiadaceae), generating a debate on the origin of alkaloids cryptolepinone (212) and hydroxycryptolepine (213), which appear to be the keto and enol forms of the same compound [92]. Cryptolepinone (212/213), obtained from roots of the plant under acidic conditions, was not detected in neutral extracts. Forced formation of 212 from 211, dissolved in organic solvents (MeOH or CH3 CN), was challenged in the absence or presence of acid (HCl 0.1 M), light, and/or air over 19 days. Increasing amounts of cryptolepinone (212) were detected, with the highest concentration (2.5%) in acid conditions under air exposure. Nevertheless, treatment of 211 with m-chloroperbenzoic acid (m-CPBA) gave a 16.5% yield of 212 (Scheme 24.49). Acid-catalyzed formation of an acetal during silica chromatography was observed for laurencione (215), from the red alga Laurencia spectabilis. The molecule occurs as a mixture of two interconverting forms: (±)-2-hydroxy-2methyldihydrofuran-3(2H)-one and 5-hydroxy-2,3-pentanedione. Chromatography on silica gel produces an artifactual dimeric spiroacetal 216 [93] previously isolated from Laurencia pinnatifida [94]. Regeneration of 215 is readily observed by acid hydrolysis (Scheme 24.50). From Rutaceae spp., two groups recently isolated an acetonic acetal (218) similar to the prenylated coumarin phebalosin (217). Acidic hydrolysis of the epoxide group in 217 appears probable, and would lead to a vicinal diol (not isolated) likely to react with acetone. However, the authors claimed not to have used acetone as solvent, while considering [95] – or not – 218 to be an artifact! Notably, the accompanying murralongin (219) [96] – obviously a rearrangement product, as its formation was observed in the course of purification – could have a similar origin (Scheme 24.51). Acid-catalyzed cyclizations of natural products are relatively frequent. Aposphaerin B (221), a minor achiral metabolite obtained from cultures of the fungus Paraphaeosphaeria quadriseptata, is thought to arise from cytosporone F (220) during extraction and chromatographic separation using methanol, by
901
O2, H3O+
m-CPBA
Scheme 24.49
compound, low yield
N
O Cryptolepinone (212)
isolation under acidic conditions only
Cryptolepine (211) and cryptolepinone (212) from Cryptolepis sanguinolenta.
N OH Hydroxycryptolepine (213) N Quindoline (214) H apparently unaffected by acidic treatment
N
N H
N
Cryptolepis sanguinolenta, Apocynaceae, roots
N Cryptolepine (211) experiments on pure
N
isolation under acidic or neutral conditions
902
24 Artifacts and Natural Substances Formed Spontaneously
24.8 Protic Solvents
Laurencia spectabilis (±)-2-hydroxy-2-methyl-dihydrofuran3(2H)-one
O
HO O O 85%
5-hydroxy-2,3-pentanedione
OH O Laurencione (215)
15%
p -TSA, H2O
silica gel
Laurencia pinnatifida
O O
Scheme 24.50
O
O racemate OH OH 216
Artifactual formation of a spiroketal from laurencione (215).
a Michael-type addition of a phenol to the enone moiety.25) This was further suggested by the treatment of 220 with p-toluenesulfonic acid, resulting in the formation of 221 (Scheme 24.52a) [97]. Likewise, the minor chromane 223 from the Australian marine brown alga Peritbalia caudata (Sporochnaceae) might be an artifact from precursor 222. Formation of 223 was described via acid- or UV-catalyzed cyclization of 222 by Blackman et al. [98].26) However, latter attempts to induce this cyclization through exposure of 222 to silica gel at room temperature in the presence of sunlight, in common extraction solvents over prolonged periods, proved unsuccessful, rendering the issue of artifactual formation unclear (Scheme 24.52b) [99]. Finally, Thalictrum (Ranunculaceae) saponins (224) can be cited. Acidic hydrolysis of their sugar moieties also causes hydrolysis of their lactol rings, followed by a nucleophilic attack of a free vicinal hydroxy group to give a furanic ring. Dehydration would yield squarrofuric acid (225).27) Similarly, acidic hydrolysis of thalicoside A (226) yields a furanic artifact (227) [100]. Notably, in this case, acidic treatment induces opening of the cyclopropane ring – a classical degradation process in the cycloartane series (Scheme 24.53). 24.8 Protic Solvents
Protic solvents are extensively used in the isolation of natural products, either for extraction (MeOH, EtOH, less frequently H2 O) or purification by normal25) Note that 220 and 221, both ethyl esters at 26) See compound 157 (Scheme 24.35). the C8 position, are possibly natural (no 27) Surprisingly, a saturated analog
ethanol used during isolation).
obtained.
is
903
O
O
O
O
O
O
O
Scheme 24.51
OH
−H+
O
O
Murralongin (219) artifactual
H3CO O
Murraya omphalocarpa, Rutaceae [Molecules 2008,13, 122-128]
according to published experimental conditions: no acetone used in the course of isolation!
intermediate diol: not obtained → full consumption to 218?
extraction: MeOH
OH
OH2+ HO
appearence in the course of purification
Artifactual coumarins from Rutaceae spp.
Minumicroline acetonide (218) considered or not as artifactual
H3CO
H+ Phebalosin (217)
H2O
H3CO
extraction: AcOEt
Galipea panamensis, Rutaceae
[J. Nat. Prod. 2010, 73, 1012-1014]
904
24 Artifacts and Natural Substances Formed Spontaneously
24.8 Protic Solvents
Paraphaeosphaeria quadriseptata (fungus)
O
MeOH extraction, silica gel H+ with MeOH HO
O 8
HO
1 1' 3
O
O
?
3'
OH O
O
O H
p-TSA
Cytosporone F (220) (a)
Aposphaerin B (221) Peritbalia caudata, Sporochnaceae (brown alga)
OH
UV or H3O+ (1)
silica gel (AcOEt), RP HPLC (MeOH/H2O)(2)
O
various solvents, light, air, r.t. (2)
222 (b)
223 artifact?
(1) [Aust. J. Chem. 1979, 32, 2783]; (2) [J. Nat. Prod. 1994, 57, 849-851]
Scheme 24.52
Acid-catalyzed intramolecular Michael additions.
or reversed-phase chromatography.28) Alcohols easily react with several functions, especially under heating conditions or during long storage of wet extracts or raw material. Participation of neighboring groups is frequently possible. Reactions depicted below are often acid or base catalyzed. 24.8.1 Lactonic Compounds: Epimerization, Transesterification
Lactonic compounds are especially sensitive to protic solvents, under various conditions. For example, acetogenins of Annonaceae bear a S-γ -methyl butyrolactone, which is frequently α,β-unsaturated. Epimerization of this stereogenic center is easy in methanolic solutions containing traces of alkali.29) Nevertheless, a series of oxopropyl-substituted lactonic ‘‘isoacetogenins’’ have been frequently obtained, as inseparable mixtures of cis/trans isomers. Duret et al. showed that these compounds occur by translactonization of natural acetogenins [101] bearing a hydroxyl group at the β-position relative to the unsaturated lactone [e.g., annonacin (228)], and proposed pathways involving acidic or basic catalysis [102]. However, silica gel is not a strong enough catalyst to induce this reaction, which could be observed upon thorough heating in alcoholic solution or in water (Scheme 24.54) [103]. 28) Not to forget deuterated solvents in NMR 29) After extraction of isoquinolinic alka-
experiments – in which case, artifact formation is easy to detect.
loids – the other major class of interest in the family.
905
O OH H+
H
Scheme 24.53
CH2O-b-D-Gal
H
OH
OH
OH
HO
O OH
CH2OH unnamed artifact 227
H
O
saturated analogs
HOOC
Squarrofuric acid (225)
HO HO
-H2O -oses [O]
acidic hydrolysis to obtain genines
O
OH
H+
Artifactual cyclization of Thalictrum saponins during acidic hydrolysis.
Gal-b-D-O
Thalicoside A (226)
Thalictrum minus, aerial parts
R = H, osidic chain Talichtrum saponins (squarrogenins, 224)
RO HO
HO
Thalictrum squarrosum, Ranunculaceae
906
24 Artifacts and Natural Substances Formed Spontaneously
3 OH
Annonacin (228)
OH
H
O OH
OH
OH
OH
Scheme 24.54
HO
H
O H2O
H
O
O O
O
O
H
O
O
O
H O
O
H 2O
OH
4
O
O O
3
H
O
O
O
OH 9
O
O
γ - epimers
(2,4-cis /trans)isoacetogenins isolated as mixtures
O
O H O
no longer detectable using Kedde reagent less cytotoxic than acetogenins counterparts
[Tetrahedron Lett. 1997, 38, 8849-8852]
O
H
OH O
OH
Cis/trans-isoannonacin (229)
O
OH
translactonization
H HO
epimerization
OH O
O
R = H, OH
Epimerization and translactonization of annonaceous acetogenins.
O
R = H alkaline treatment
H
O
O
1
9
H
acid or base, protic solvent,
C-34 /36: S most Annonaceous acetogenins bear an a,b-unsaturated- g-lactone O O R
O
R = OH acidic treatment + heat / (heat, 24h)
HO
O
R = OH alkaline treatment
unsaturated lactone: revelation with Kedde reagent (dinitrobenzoic acid, then KOH)
O
O
Annonaceae, e.g. Annona spp.
24.8 Protic Solvents 907
908
24 Artifacts and Natural Substances Formed Spontaneously
Interestingly, Li et al. isolated isoannonacin (229) in its trans-form only, from seeds of Annona muricata, suggesting a possible enzymatic pathway in planta [104]. Complete lactone hydrolysis or alcoholysis can be induced under drastic conditions. However, partial conversion of lactones into esters can occur when traces of acid are present, as exemplified with lactonic eudesmanes 230–233, the methyl ester derivatives of which (234–237) were observed in a methanolic extract prepared at boiling temperature, as well as during fractionation with silica gel. Their artifactual nature was challenged (Scheme 24.55) [105]. Surprisingly, such reactions are not frequently mentioned in the numerous studies on Asteraceae lactonic sesquiterpenes. Note that methanolysis of macrolactones has also been described [106].
Hieracium intybaceum, Asteraceae CH2Cl2 extract
MeOH extract
kept in solution, r.t., 7d. SiO2 CC using MeOH
R1
H R2 O
(CH3)2CO extract: no detection by HPLC-UV-MS
OCH3 R1
H2SO4 in H O R2 OH MeOH r.t., 4d. incomplete conversion R = O or a-O-b-D-Glc/b-H 230 to 233 R1 = a-Me/b-H or CH 234 to 237 2 2 Scheme 24.55
O
Methanolysis of lactonic sesquiterpenes.
24.8.2 Esterification, Transesterification
Under weak acidic conditions, in alcoholic solvents, esterification or transesterification reactions are reported, especially during extraction at boiling temperature, as previously shown in the case of cyclic esters. Artifacts (e.g., 238–242, methyl ester of 243, 244; Figure 24.3) arise in solubilized extracts upon long standing or, more frequently, during fractionation on silica gel. Traces of HCl from chloroform can also act as catalyst. Unexpected methyl and ethyl esters are thus quite commonly encountered, in various natural series. Adducts with butanol are scarce, as this solvent is less used (mostly for liquid/liquid partitions). Noteworthy examples can also be found in the iridoid series, with asperuloside (245), acetyl-asperulosidic acid (247), and its carboxylic derivatives at C10 (248). The authors challenged the obvious artifactual nature of ethyl esters (246, 249) of these molecules, avoiding ethanol and using methanol throughout their protocols. However, unclear results were obtained, evidencing the possible implication of solvent impurities. Nevertheless, silica was not the catalyst according to the protocols used (Scheme 24.56) [108].
O O O O
O O O O
O
n
O O
H
N
N
N
H
H
O
N H N
NH2
facile exchange with CD3OD observed during NMR experiments
H
specimen stored in NH EtOH Dehydrobatzelladine C (243):
O
N
N
H O
N
O
NH O
Figure 24.3
Miscellaneous esterification and transesterification artifacts [107].
native compound [Nat. Prod. Rep. 2002,19, 617-649] [J. Nat. Prod., 2000, 63, 193-196] Crambescidin-431 (244): Artifact
Neofolitispate 1 (240): n=14 Neofolitispate 2 (241): n=13 Neofolitispate 3 (242): n=12 Artifacts
H3CO
O
obtained from fresh specimen obtained from a specimen stored Didemnaketal A (239): R1=H, R2=CH3: Artifact? during 10 years in MeOH Monanchora arbuscula (sponge) Neofolistipa dianchora (sponge) M. unguiculata specimen stored H H easy guanidine catalyzed in MeOH H H H exchange with MeOH O O
O N
[Nat. Prod. Rep. 2004, 21, 50-76]
transesterification
OR2
note absence of transesterification of "classical" alkyl esters
Didemnaketal C (238): R1=H, R2=CH2CH2SO3Na: Parent compound?
R1O
O
O
Didemnum sp. (ascidia)
esterification O
24.8 Protic Solvents 909
910
24 Artifacts and Natural Substances Formed Spontaneously
Hedyotis chrysotricha, Rubiaceae non artifactual according to authors, MeOH claiming not to have used EtOH
O H
O EtOH?
O HO H
O
H
O EtOH?
O
H
OGlc
Asperuloside (245)
OH
O
O O
O HO H
O
O
OGlc
Asperulosidic acid ethyl ester (246)
O
O
H
OGlc
Acetyl asperulosidic acid (247)
[J. Nat. Prod. 1999, 62, 611-612]
Tarenna attenuata, Rubiaceae artifactual MeOH
O HO H
OR EtOH
O H O OGlc OR 248: R = H, cinnamoyl, etc.
O HO H
O
OH
O H O OH Tarennin (249)
[J. Nat. Prod. 2006, 69, 971-974] Scheme 24.56
Esterification and transesterification of Rubiaceae iridoids.
24.8.3 ‘‘Apparent Alkylations’’
Substitutions of hydroxyl, phenol, or ether groups are frequent, under conditions similar to that of transesterifications, by SN 1 or SN 2 type reactions. These reactions preferentially occur on hemiacetals to form acetals. They are also reported when strongly inducing neighboring groups are present, or for resonance-stabilized intermediate carbocations. Ethanol or butanol ‘‘adducts’’ can easily be recognized as non-natural. Artifactual methylated compounds are less likely to be identified, and might therefore be largely underestimated: they are generally evidenced when chromatographic profiles of extracts prepared with different solvents are compared. Scheme 24.57 presents examples of artifactual methyl ethers (251, 253–255) caused by the use of methanol [109]. The recently published isolation of xanthepinone (256) from a fungus of the Rhizina genus provides a nice example of methanolysis of an acetal finally yielding a rearranged methyl-acetalic artifact (257, Scheme 24.58) [110]. Nucleophilic substitution of a halogen can also constitute a mechanism for generating methoxy-bearing artifactual marine products. This is proposed for methoxydechlorochartelline A (259) in relation to chartelline A (258) (Scheme 24.59) [111].
OH O
O
OH
O BuOH
extraction: MeOH
O O
Scheme 24.57
Miscellaneous methylation artifacts.
[J. Nat. Prod. 1990, 53, 1619-1622]
Authors state that 254 "may be an artifact produced during MeOH extraction [from Aplysiadiol (R=H)]"
OH
HO HO
OH Pouzolignan A methylether (253)
HO
Br
OCH3
O
[Phytochemistry Lett. 2010, 3, 29–32]
Pouzolignan A (252)
OH
O
O
silica CC w/ MeOH
OH extraction: cold MeOH, OH
OH
255
O
facilitated by bromine
[J. Nat. Prod. 1999, 62, 882-884]
Br atom attracting effect?
OCH3
OH
OCH3 Br
identified as an artifact due to extraction with MeOH; hydroxylated derivative obtained with other solvents
Odonthalia corymbifera, Rhodomelaceae (red alga)
SN1/SN2 facilitated by conjugation?
OH R = CH3: Aplysiadiol methyl ether (254)
OR
Apiysia kurodai, Aplysiidae (mollusc)
Reported as a genuine metabolite in an unidentified saprophytic ascomycete [J. Nat. Prod. 2008, 71, 1973-1976]
[J. Nat. Prod. 2010, 73, 1156-1159] no butanolic acetal observed
Br
O
hemiacetalic
Pouzolzia occidentalis, Urticaceae, aerial parts
OH O OCH3 HO HO 1-Methoxydehydroherbarin (251)
HO
Corynespora sp. (endolichenic fungus)
1-Hydroxydehydroherbarin (250)
HO
hemiacetalic
24.8 Protic Solvents 911
O O
Scheme 24.58
O COOCH3 O
CH3OH OH OCH3 O O H H3COOC O O OH O OH OH OH O OH O COOCH3 O O O O CH3
O
H3COOC O OH OH
257
O
O OH O O COOCH3 OCH3
proposed pathways [Phytochemistry Lett. 2010, 3, 152-155]
O OH OH
OH
complete conversion
O COOCH3 O OCH 3 257: likely to be an artifact
O OH O
minor compound only
O O O
H3COOC OH O
MeOH, p-TsOH, r.t., 16h
MeOH, SiO2, r.t., 4d
Methanol induced rearrangement of xanthepinone (256).
methanolysis of the ketene acetal
O OCH3 OH O OHO b a O 256 c O b c O H CH3
a
intramolecular hetero Michael addition/elimination
Xanthepinone (256)
methanolysis at the enol ether carbon
isolation:
O OCH3 SiO2 CC using MeOH OH O OHO
wood-decay fungus Rhizina sp. BCC 12292
912
24 Artifacts and Natural Substances Formed Spontaneously
24.8 Protic Solvents
Chartella papyracea, Flustridae (bryozoa) Unusual substitution; Artifactual nature suspected
Br Br
O Br N
Cl N
Br NH
?
N
Br Br
Chartelline A (258) Scheme 24.59
MeOH
MeO N Br O NH Br N N
Methoxydechlorochartelline A (259)
Methanol substitution of chartelline A (258).
Methoxylation artifacts are sometimes due to nucleophilic addition on conjugated systems, such as for puupehenone (260), which bears a quinone-methide system. Under fractionation of the chloromethylenic extract of a marine sponge (Hyrtios sp.) with methanol, 260 easily yielded 15α-methoxypuupehenol (261), thus becoming a trace compound [112]. Interestingly, 260 could be regenerated from 261 by filtration on silica gel (Scheme 24.60) [113].30) Only a few reports depict such reactions, although they possibly occur frequently. Ethylation artifacts – also found in various chemical series – are easily distinguished as ethyl-ether groups are rare in natural products. These artifacts are generally generated during extraction; ethanol is barely used in chromatographic procedures. However, traces of ethanol used as a stabilizer (e.g., in chloroform) can also yield such artifacts (Scheme 24.61) [114]. Such reactions also occur in the case of acetoxy acetals, the instability of which was noted by several authors (Scheme 24.62) [115]. Butanol adducts are described in the literature, such as 4-O-butylpaeoniflorin (273) from Paeonia suffruticosa (Paeoniaceae), formed from paeniflorin (272) during fractionation of a crude methanolic extract with butanol [116]. No evidence of a catalyst can be found in the protocol used, but the hemiacetalic position (C4) in Paeonia monoterpenes is very reactive. Indeed, the roots of Paeonia lactiflora, which are used in Chinese medicine, are sometimes traditionally fumigated with sulfur to preserve their white color (white-peony root) and to prevent bacterial growth. This processing method induces a fall in the amount of 272 and related hemiacetalic compounds to yield sulfated analogs at C4 (e.g., sodium paeoniflorin sulfonate 271) (Scheme 24.63) [117]. 24.8.4 Formation of Acetals
Formation of acetals from aldehydes can be observed under mild conditions: methanolic extraction of secoiridoids (e.g., secologanin 279) easily yields dimethyl 30) Interestingly, 261 was less cytotoxic than
260 (this is quite understandable in regard to reactivity), while showing improved
in vitro activity against Plasmodium falciparum.
913
Scheme 24.60
H
15
O
OH
cooling to ice temperature
regeneration on SiOH gel
O
OH OH
[J. Nat. Prod. 1999, 62, 1304-1305]
final yield: 25%
15 a-Methoxypuupehenol (261)
H
Sephadex LH-20 gel-filtration H (CHCl3-MeOH 2:8) H3CO 15
warming: quantitative regeneration [J. Org. Chem. 1993, 58, 6565-6569]
O
Hyrtios sp., CH2Cl2 extract
Methanol adduct on puupehenone (260).
final yield: 0.1%
Puupehenone (260)
major metabolite in CH2Cl2 crude extract, according to NMR
914
24 Artifacts and Natural Substances Formed Spontaneously
N OH
extraction: EtOH
O N
facilitation by mesomery
Scheme 24.61
265
O
O O
EtO 266 O
O O
from hemiacetal
HO
H
H OEt O
unknown precursor
O
extraction: EtOH
Dysidea arenaria (sponge)
[J. Nat. Prod. 1982, 46, 441-452]
Dubirheine (O-ethylrhoeganine) (264)
H O H EtO
N H
[Molecules 2005, 10, 1292-1297]
9-Hydroxyfurodysinin-O -ethyl lactone (267)
OH
[Chem. Biodiv. 2007, 4, 2210-2217]
from ester
O
Miscellaneous ethylation artifacts.
O
O
O
OH
Ligularia villosa, Asteraceae, roots extraction: EtOH room temp. AcOEt
[J. Asian Nat. Prod. Res. 2005, 7, 127-130]
O
O
Papaver americanum L., Papaveraceae
chiral: SN2?
OEt Polygonatine A (262) Polygonatine B (263)
O
Polygonatum sibiricum, Liliaceae
24.8 Protic Solvents 915
916
24 Artifacts and Natural Substances Formed Spontaneously
Lepidolaena clavigera, Lepidolaenaceae (livewort)
extraction: EtOH
OO
O
269
O O
OO O O Clavigerin A (268)
OO
O 270
Hydrolysis products: RP-HPLC using MeOH O not obtained All compounds active as antifeedants: Active form = hydrolysis product? Scheme 24.62
both undetected with non protic extraction solvents or with CH3CN used in RP-HPLC
[J. Nat. Prod. 2008, 71, 258-261]
Artifactual alcoholysis of acetoxy-acetalic sesquiterpenes.
acetal artifacts (275), as shown with Lonicera korolkovii [118] and Lonicera japonica (Caprifoliaceae) [119]. Conversion of secologanic acid (274) into 276 was also observed. Similarly, extraction of L. japonica aerial parts with water, followed by partition with n-butanol, yielded low amounts of secologanin dibutyl acetal (277) and butylsecologanic acid (278). Both molecules were obviously absent in crude extracts [120]. A complex artifact, korolkoside (280), showing both acetal and ester artifacts, was also isolated from L. korolkovii (Scheme 24.64) [118]. Surprisingly, it displays acetalization of the aldehydic group of a secologanin moiety by glucose of a second monomer (C4/C6 diol system). This feature, atypical for a natural product, was probably acid-catalyzed in anhydrous conditions during fractionation. However, the authors state that the methyl esters and dimeric nature of 280 are of natural origin, according to an unpublished study of Lonicera morrowii conducted ‘‘without acidic conditions.’’ In the isocedrene series, a peculiar acetal (282) was obtained from Pleocarphus revolutus (Asteraceae, aerial parts) after methanolic extraction, along with its dialdehydic counterpart 281 (Scheme 24.65). However, the artifactual nature of 282 was not proved [121]. 3,5-Dibromo-1-hydroxy-4-oxo-2,5-cyclohexadien-l-acetamide (283), from the sponges Aplysina fistularis forma fulva [122] and Aplysina caissara (Verongidae) [123], is also of interest: the presence of two bromine atoms in ortho positions to the ketone facilitates acetal formation during extraction with methanol, to give 284.
24.9 Acetone-Derived Artifacts
Use of acetone in acidic conditions can quite easily give rise to acetals on vicinal diol systems, as previously exemplified (compound 218, Scheme 24.51).
traditional sulfur fumigation
O OSO3Na
O O OH
4
O
Scheme 24.63
Nucleophilic substitutions on paeoniflorin (272).
Samples from Chinese drug stores: variable amounts of 271 nd 272; → 272 << 271 [Helv. Chim. Acta 2010, 93, 565-572]
OO
O
[J. Nat. Prod. 2009, 72, 1465-1470]
O 4-O-Butylpaeoniflorin (273)
O
GlcO
P. suffruticosa, cortex absent from MeOH extract according to HPLC-UV
BuOH fractionation
extraction: MeOH
O Paeoniflorin (272)
O
GlcO
Paeonia lactiflora, Paeoniaceae, roots
O Sodium paeoniflorinsulfonate (271)
O
GlcO
absent from unprocessed roots
24.9 Acetone-Derived Artifacts 917
OGlc
O
O
OGlc Secologanic acid (274) O H COOCH 3 H
H
O H
H
RO O H O
O O
OH
O
artifactual
artifactual?
Lonicera spp., Caprifoliaceae
Scheme 24.64
O
O O HO
O OH
OH
O O H
[J. Nat. Prod. 2001, 64, 1090-1092]
O H OGlc Korolkoside (280)
H
O
artifactual? artifactual?
Secoiridoid acetals from Lonicera spp.
a: extraction with MeOH [Chem. Pharm. Bull. 1988, 36, 3664-3666; J. Nat. Prod. 2001, 64, 1090-1092] b: partition with BuOH of H2O extract; CC silica using CHCl3/MeOH [J. Nat. Prod. 1995, 58, 1756-1758]
OGlc R = Me: SecologaninR = Me :7O -Methylsecologanic acid (276) (a, b) O dimethylacetal (275) (a, b) R = Bu: 7O -Butylsecologanic acid (278) (b) H R = Bu: SecologaninOGlc dibutylacetal (277) (b) Secologanin (279)
H
O
RO ORCOOCH 3 H
?
HO
918
24 Artifacts and Natural Substances Formed Spontaneously
24.10 Halogenated Solvents
Pleocarphus revolutus, Asteraceae extraction: MeOH
O
Aplysina spp. (sponge)
O
artifact?
Br
Br MeOH Br
OO
Br
O
O
O
281 Scheme 24.65
919
O O 282
HO 283
O HO 284
NH2
NH2
Acetals from Pleocarphus revolutus and Aplysina spp.
Aristolochia rodriguesii
OH HOOC
285
OH
CC silica gel, (CH3)2CO
O HOOC
286
O
likely artifact Scheme 24.66
An acetal caused by acetone under acidic conditions.
The kauranoid 286 from Aristolochia rodriguesii (Aristolochiaceae), an acetonide of ent-16β,17-dihydroxy-(−)-kauran-19-oic acid (285), is probably an artifact of purification using acetone on silica gel (Scheme 24.66) [29]. Acetone can also form specific artifacts with alkaloids: under acidic conditions, adducts on enamines are observed, for example, for indolomonoterpenes [1,2-dehydrobeninine (287) and its acetonide derivative 288 (Scheme 24.67) or for lysine-derived alkaloids [lycodine (289) and its artifact hydroxypropyllycodine (290)]. Hesse points out that the presence of an acetonyl side chain in an alkaloid does not dictate that the compound is an artifact [77], as in the case of (−)-pelletierine (291, Punica granatum, Punicaceae) [75].
24.10 Halogenated Solvents
Chloroform and methylene chloride are often used in natural products fractionation and analysis. Halogen-containing solvents are unstable. Chloroform, depending on purity and age, tends to decompose into HCl and phosgene (COCl2 ) when exposed to light and oxygen. Phosgene easily yields artifacts, especially with amines [e.g., (−)-verbaskine (292) and ovihernangenine (293) (Figure 24.4)] [77], and is a source of free radicals. Ethyl chloroformate, which is also formed in aging chloroform containing ethanol used as a stabilizer to prevent the previous reaction, is also able to react with amines. These aspects have been reviewed [7, 124, 125]. Notably, methylene chloride also tends, to a lesser extent, to produce HCl.
H H O
H+ H N O
Scheme 24.67
N H
H H
N
S
H N OH N H Hydroxypropyllycodine (290)
iminium intermediate
Acetone adducts on alkaloids under acidic conditions.
N H Lycodine (289)
H
O
hydrolysis, decarboxylation
O N H N H H Mannich reaction on (-)-Pelletierine (291)
Lys
CoASOC
Biosynthetic pathway: acetoacetyl-CoA
HO
N H R/S 288
N
H O
Lycopodium obscurum, Lycopodiaceae
N H O H+ O O 1,2-Dehydrobeninine (287)
N
H O
Hedranthera barteri, Apocynaceae
920
24 Artifacts and Natural Substances Formed Spontaneously
24.11 Protoberberines, a ‘‘Cabinet de Curiosit´es’’
921
O HN O H N
N Carbon atom from phosgene
O N
(−)-Verbaskine (292) Verbascum nobile, Scrophulariaceae Figure 24.4
O O
O N N H H O
O O
O O O Ovihernangenine (293) Hernandia nymphaeifolia, Hernandiaceae
Phosgene adducts.
Some natural compounds can be halogenated in the course of isolation. Artifacts from terrestrial plants are easily recognizable: even though organochlorines and organobromines have been detected across all phyla, including higher plants, they are scarcely encountered as genuine molecules in plants [126]. Terpenoids exemplify several mechanisms of chlorination. The sesquiterpene madolin Q (294) from Aristolochia heterophylla is considered to be an artifact formed during extraction and separation with chloroform, by substitution with a • CCl3 radical, according to authors [29]. A chlorinated triterpene glycoside 296 was obtained from Cimicifuga racemosa (Ranunculaceae) roots, along with its hydroxylated counterpart 295 (Scheme 24.68). Compound 296, undetected in the crude methanol extract, apparently appeared during partition using chloroform and n-butanol. The authors proposed a reaction of HCl with the tertiary hydroxyl function at C25 and conducted a series of experiments in which a dilute chloroformic solution of 295 was subjected to HCl vapor: 295 partly converted into 296, confirming it is an isolation artifact. Triterpenic saponosides from Securidaca longepedunculata (Polygalaceae) obtained by extraction with methylene chloride have two types of genine, namely, senegenine (298) and the chlorinated presenegenine (297) [127]. Artifactual chlorinated alkaloids are also observed (e.g., 299) [128]; chloroform often reacts with quartenarized compounds by nucleophilic addition (e.g., 300) [129]. Finally, cases of tertiary amines reacting with alkyl halides (dichloromethane, chloroform, and their impurities) during extraction to form quaternary ammonium salts were identified [e.g., macrosalhine (301, Scheme 24.69), and see Scheme 24.40]: ‘‘Caution is therefore needed when quaternary ammonium derivatives are isolated using these solvents’’ [81].
24.11 Protoberberines, a ‘‘Cabinet de Curiosit´es’’
Protoberberines are quaternary isoquinolines encountered mostly in Magnoliales and Ranunculales (e.g., Berberidaceae, Papaveraceae, Ranunculaceae). The ease
O O
Madolin Q (294)
H H
O
HO
OH HO Presenegenine (297)
COOH
H
N+
OH O
COOH
O
Cl
NH
O
7-chloro-norcepharadione B (299)
H3CO
H3CO
extraction CHCl3, NaOH 20 %; sep. silica gel using CHCl3/MeOH Dihydroxy-1-methyl-3-oxo-2-(trichloromethyl)-3H indolinium (300)
CCl3
Cl
unaffected
O
partial conversion Houttuynia cordata, Saururaceae
H Senegenine (298) COOH
H
Artifactual chlorinated terpenoids and alkaloids.
artifactual nature not discussed by authors
HO
Zanthoxylum nitidum, Rutaceae
H COOH
H
H
CH2Cl
HCl vapors, CHCl3 solution Securidaca longepedunculata, Polygalaceae
extraction: CH2Cl2
H
O HO H HO OH Chlorodeoxycimicigenol-3-O-b-D-xyloside (296)
O
H
Cimicifuga racemosa, Ranunculaceae undetected in crude extract (LC-MS) Hpartition of extract with CHCl3 H 24: a H H 25 H O O unaffected OH
HO H HO OH Cimicigenol-3-O-b-D-xyloside (295)
O
Scheme 24.68
HO
HO
O
Cl3C
Aristolochia heterophylla
922
24 Artifacts and Natural Substances Formed Spontaneously
24.11 Protoberberines, a ‘‘Cabinet de Curiosit´es’’
Rauvolfia caffra
N
artifactual quaternarization due to CHCl3 and basic conditions
OH H CHO H
N H 182' (from intermediate 182) Scheme 24.69
N N
O H OH
H H Macrosalhine (301)
Quaternization of the indolomonoterpenic alkaloid macrosalhine.
with which these alkaloids undergo artifactual transformations justifies their presentation as the concluding part of this chapter. The above-mentioned oxidation, dimerization, amination, addition of acetone, and chlorination reactions can be exemplified with the two major representatives of this class, berberine (302) and palmatine (314, Scheme 24.72 below), the reactivity and common artifacts of which have been reviewed [130, 131]. First, these native quaternary salts are easily converted into their base forms (303 for berberine) by strong bases. Both diastereomers of bimolecular aminoacetals, formed by the condensation reaction between two molecules of the base form, were detected as minor products (304 for berberine). They are immediately converted back into salts by traces of acid. Nevertheless, quaternary protoberberine salts are unstable in the presence of concentrated alkali: the alkaloid bases formed are subsequently transformed into 7,8-dihydro- or 8-keto-derivatives by disproportionation of the pyridine ring (305, 306, Scheme 24.70). Analogically, nucleophilic addition of alcohols is easy, inducing formation of 8-methoxy and 8-ethoxy-7,8-dihydroderivatives (308 and 309, respectively, from berberine), during the isolation or separation procedures in alkaline media, in the presence of methanol and ethanol. These unstable artifacts are easily converted back into their quaternary forms in an acidic environment. 8-Amino-derivatives (307) are formed with primary amines in methanol or ethanol (Scheme 24.71). When chloroform is used during extraction under alkaline conditions, 8-trichloromethyl adducts are also easily formed; trichloromethylberberine (310) and trichloromethylpalmatine (315) are, thus, frequently observed. The behavior of 310 on a silica gel column was studied: oxidized products 311–313 are obtained. A similar study of 315 led to isolation of 316–318 (Scheme 24.72) [7]. In a protic medium, 7,8-dihydroprotoberberine derivatives (e.g., 305) are in equilibrium with their iminium forms (305a). The latter species are unstable and undergo rapid disproportionation to give a mixture of corresponding protoberberines (e.g., 302) and tetrahydroprotoberberines (e.g., 319). Notably, natural tetrahydroprotoberberines are natively (S)-enantiomers (Scheme 24.73) [75]. Finally, berberine (302) was shown to undergo facile nucleophilic attack by acetone during fractionation, yielding the heptacyclic karachine (320, Scheme 24.74) [77].
923
N HCl
NaOH
O
Scheme 24.70
8
O
O
N 7 OH −H2O
O
O
Alkaline treatment of protoberberines.
N7 8 O
O
O
O
O
O
O
O 8' H N O N 8 H O
O
8-oxo-7,8-dihydroberberine (306)
Berberine base (303) (8-hydroxy-7,8-dihydroberberine)
O
O
Cl−
7,8-Dihydroberberine (305)
concentrated alkali
Berberine (302) chloride
O
O
O
8
N7 O
O
O
304
O
O
924
24 Artifacts and Natural Substances Formed Spontaneously
24.12 Conclusion
O
O O
N
NH2 O
NH3
OH−,
302
O 8-Amino-7,8dihydroberberine (307) Scheme 24.71
925
ROH O
N
OR O
O R = Me: 8-Methoxy-7,8-dihydroberberine (308) R = Et: 8-Ethoxy-7,8-dihydroberberine (309)
Nucleophilic additions on berberine in alkaline conditions.
Similar reactions can be expected for chemo-equivalents. For example, nucleophilic attacks also occur on benzophenanthridines (Papaveraceae), with these artifacts being more stable due to lower pK. 6-Hydroxy-5,6-dihydrobenzophenanthridines also yield dimerization products [130]. This reactivity was exemplified by easy conversion of chelilutine (321; Sanguinaria canadensis, Eschscholtzia californica, Papaveraceae) into 322–324 under basic conditions (Scheme 24.75) [132]. It is also noteworthy that a parallel reactivity exists for dehydroaporphines (e.g., 55), which are considered as key intermediates in the biosynthesis of 7-oxygenated and 7-alkylated aporphines. Their enamine group is nucleophilic enough to readily react during isolation. Obviously, such compounds should be studied or purified using adapted precautions.
24.12 Conclusion
The variety and often poor-description of favorable ‘‘conditions + reactivity’’ combinations leading to artifacts render the topic intricate. In most cases, formation of an artifact is due to uncontrolled elements, such as unidentified solvent impurities (carelessness and fumbling are rarely stated in ‘‘Material and Methods’’ sections). Some practices appear to be at risk for the phytochemist, with several solvents to be considered with caution, especially with thermal treatment.31) Surprisingly, some artifact-prone protocols do not seem, in numerous reports, to yield any by-product. Nevertheless, discrepancies in the literature are frequent for identical or chemo-equivalent natural products, without any apparent clue as to why. The nearly virtual fractionation of crude extracts allowed by development of hyphenated techniques is now a frequent starting point in phytochemical studies. It permits a reduction in duration and number of manipulation steps and, obviously, in formation of artifacts – while also facilitating their detection and 31) The use of Soxhlet apparatus only scarcely
appears in recent literature, but long
extractions at boiling temperature are frequently performed.
N+
Scheme 24.72
O
CCl3
N O
CCl3
O O
N O
O
N
O
O
O
O
N
O O
O Polyberbine (313)
CHO N OH O
O OH
O Puntarenine (317)
O
O
OH O Magallanesine (318)
N
O
O
O
O berberrubine (312)
O
O
O Saulatine (316)
O
O
Silica gel CC, CHCl3 /MeOH 99:1
O O 8-Trichloromethylpalmatine (315)
O
O
O
Silica gel CC, CHCl3 /MeOH 99:1 CHCl3
N
O Oxyberberine (311)
O
O
Chloroformic adducts and their oxidation artifacts.
Palmatine (314)
O
O
N
O 8-Trichloromethylberberine (310)
O O
926
24 Artifacts and Natural Substances Formed Spontaneously
N7 8
O
Scheme 24.73
14
N+ O
disproportionation
O
O H
N
Berberine (302)
(R,S)-Tetrahydroprotoberberine (canadine, 319)
O iminium form (305a)
O
O
Disproportionation of dihydroprotoberberines.
O 7,8-Dihydroberberine (305)
O O
ROH or H+
O
O
302 * (S )-tetrahydroxyprotoberberine oxidase
Biosynthesis: (S )-canadine (S-319) O2 oxidase* H2O2 305a O2 oxidase* H2O
24.12 Conclusion 927
+
N
H O
Acetone adducts of berberine.
O O H
O N H H
O
O
O
O
N O
O
O
H O +
O
OH
O Karachine (320)
Scheme 24.74
[aD=0]
Berberine (302) O N OO H
O
O
Berberis aristata, Berberidaceae
H3O+
H O
O
O
O
O
N H
H
O
O
O
O
[Hesse M., 2002]
O HO
HO
N+ H H
928
24 Artifacts and Natural Substances Formed Spontaneously
Scheme 24.75
N5
O
+
Na2CO3, H2O O
O
NH4OH
Chelilutine (321)
6
N
Reactivity of benzophenanthridines.
OH 322
6
Obtained in a NMR tube, on shaking with aqueous Na2CO3 in CDCl3
O
O
O
O
O O
O
N
N 6'
O
6
O
O
O
O
N O
O O
O 324
O NH O N
O
[Nat. Prod. Rep. 2000, 17, 247-268]
O 323
O
O
O
24.12 Conclusion 929
930
24 Artifacts and Natural Substances Formed Spontaneously
structural identification. Recognizing the artifactual nature of an isolated natural product can be rather difficult. Measuring the ratio between biosynthesis and artifactual formation is yet another exciting challenge. Regarding this particular aspect, dereplication32) can help in defining metabolic ‘‘priorities,’’ especially in chemotaxonomic or ecological perspectives.33) The reactivity of natural products from various series sometimes affords artifacts of complex structures34) under mild conditions. Parts of these examples are evocative of biogenetic processes, and are generally compatible with simple, ‘‘green’’ procedures. With several drawbacks (selectivity, stereocontrol, yield, etc.) but genuine spontaneity, the chemistry of artifacts can thus offer helpful pieces – readily possible and accessible synthesis steps – in the jigsaw of rationale and efficient biomimetic strategies.
References 1. Rey, A., Tomi, M., and Hord´e, T.
2.
3.
4.
5.
6. 7.
8.
(2006) Dictionnaire Historique de la Langue Fran¸caise, Le Robert-Sejer, Paris. (a) Gunatilaka, A.A.L. (2006) J. Nat. Prod., 69, 509–526; (b) Strobel, G., Daisy, B., and Castillo, U. (2004) J. Nat. Prod., 67, 257–268. Pettit, G.R., Mukku, V.J.R.V., and Cragg, G. (2008) J. Nat. Prod., 71, 130–133. (a) Wu, K.M., Farrelly, J.G., and Upton, R. (2007) Phytomedicine, 14, 273–279; (b) Heinrich, M., Chan, J., and Wanke, S. (2009) J. Ethnopharmacol., 125, 108–144. Blackman, A.J., Davies, N.W., and Ralph, C.E. (1992) Biochem. Syst. Ecol., 20, 339–342. Gravel, E. and Poupon, E. (2008) Eur. J. Org. Chem., 27–42. Maltese, F., Van der Kooy, F., and Verpoorte, R. (2009) Nat. Prod. Commun., 4, 447–454. Barham, P., Skibsted, L.H., and Bredie, W.L.P. (2010) Chem. Rev., 110, 2313–2365.
32) And constitution of a correctly stored
extract library for later comparisons. 33) For example, compounds 26, 30 (Scheme 24.9); 57–62 (Figure 24.2); 110–113 (Scheme 24.26).
9. Davidek, T., Clety, N., and Devaud, S.
10.
11. 12. 13.
14.
15.
16. 17.
18.
19.
(2003) J. Agric. Food Chem., 51, 7259–7265. Balentine, D.A., Wiseman, S.A., and Bouwens, L.C. (1997) Crit. Rev. Food Sci. Nutr., 37, 693–704. Tanaka, T., Mine, C., and Watarumi, T. (2002) J. Nat. Prod., 65, 1582–1587. Li, Y., Shibahara, Y., and Tanaka, T. (2010) J. Nat. Prod., 73, 33–39. Quideau, S., Jourdes, M., and Lefeuvre, D. (2005) Chem. Eur. J., 11, 6503–6513. Bruneton, J. (2009) Pharmacognosie, Phytochimie, Plantes M´edicinales, 4th edn, Lavoisier, Paris. Singhuber, J., Zhu, M., and Prinz, S. (2009) J. Ethnopharmacol., 126, 18–30. Yue, H., Pi, Z.F., and Li, H.L. (2008) Phytochem. Anal., 19, 141–147. Moran, A.V., Jørgensen, K., and Jørgensen, C. (2008) Phytochemistry, 69, 1795–1813. (a) Jones, D.A. (1998) Phytochemistry, 47, 155–162; (b) Vetter, J. (2000) Toxicon, 38, 11–36. Bruneton, J. (2005) Plantes Toxiques: V´eg´etaux Dangereux Pour L’homme et les
34) Or key intermediates such as, for
example, 194 (Scheme 24.42).
References
20.
21. 22.
23.
24. 25. 26. 27.
28.
29. 30. 31.
32. 33.
34.
35. 36.
´ Animaux, 3rd edn, Editions Tec et Doc, Paris. (a) Holst, B. and Williamson, G. (2004) Nat. Prod. Rep., 21, 425–447; (b) Kliebenstein, D.J., Kroymann, J., and Mitchell-Olds, T. (2005) Curr. Opin. Plant Biol., 8, 264–271; (c) Grubb, C.D. and Abel, S. (2006) Trends Plant Sci., 11, 1360–1385. Bonny, M.L. and Capo, R.J. (1994) J. Nat. Prod., 47, 539–540. (a) Block, E. (1992) Angew. Chem. Int. Ed. Engl., 31, 1135–1178; (b) see also: Rose, P., Whiteman, M., and Moore, P.K. (2005) Nat. Prod. Rep., 22, 351–368; (c) Lanzotti, V. (2006) J. Chromatogr. A, 1112, 3–22. Voon, Y.Y., Abdul Hamida, N.S., and Rusul, G. (2007) Postharvest Biol. Technol., 46, 76–85. Ksebati, M.B. and Schmitz, F.J. (1988) J. Nat. Prod., 51, 857–861. Lang, G., Pinkert, A., and Blunt, W.G. (2005) J. Nat. Prod., 68, 1796–1798. Capon, R.J., Miller, M., and Rooney, F. (2000) J. Nat. Prod., 63, 821–824. Lopes, L.M.X., Bolzani, Vda.S., and Trevisan, L.M.V. (1987) Phytochemistry, 26, 2781–2784. Leitao, G.G., Kaplan, M.A.C., and Galeffi, C. (1992) Phytochemistry, 31, 3277–3279. Wu, T.-S., Damu, A.M., and Su, C.-R. (2004) Nat. Prod. Rep., 21, 594–624. Lacroix, D., Prado, S., and Deville, A. (2009) Phytochemistry, 70, 1239–1245. Nomiya, M., Murakami, T., and Takada, N. (2008) J. Org. Chem., 73, 5039–5047. Sheridan, H. and Canning, A.-M. (1999) J. Nat. Prod., 62, 1568–1569. Bhaskar, M.U., Rao, L.J.M., and Rao, N.S.P. (1991) J. Nat. Prod., 54, 386–389. Venkateswarlu, Y., Bibiani, M.A.F., and Reddy, M.V.R. (1994) J. Nat. Prod., 57, 827–828. Jimenez, J.I., Huber, U., and Moore, R.E. (1999) J. Nat. Prod., 62, 569–572. Chansakaow, S., Ishikawa, T., and Seki, H. (2000) J. Nat. Prod., 63, 173–175.
37. Pilli, R.A. and Ferreira de Oliveira,
38.
39. 40.
41.
42.
43.
44. 45. 46. 47.
48. 49. 50.
51.
52. 53. 54. 55.
Mda.A. (2000) Nat. Prod. Rep., 17, 117–127. (a) Takemura, Y., Matsushita, Y., and Nagareya, N. (1995) Chem. Pharm. Bull., 43, 1340–1345; (b) Ju-Ichi, M., Takemura, Y., and Nagareya, N. (1996) Heterocycles, 42, 237–240. Wu, T.-S., Huang, S.-C., and Wu, P.-L. (1996) Phytochemistry, 42, 221–223. Negi, N., Jinguji, Y., and Ushijima, K. (2004) Chem. Pharm. Bull., 52, 362–364. (a) Ono, T., Ito, C., and Furukawa, H. (1995) J. Nat. Prod., 58, 1629–1631; (b) Michael, J.P. (1997) Nat. Prod. Rep., 6, 605–618. Desrivot, J., Edlund, P.-O., and Svensson, R. (2007) Toxicology, 235, 27–38. Van Beek, T.A., Verpoorte, R., and Baerheim, A. (1985) J. Nat. Prod., 48, 400–423. Hootele, C., Levy, R., and Kaisin, M. (1967) Bull. Soc. Chim. Belg., 76, 300. Hesse, M. (1981) Alkaloid Chemistry, John Wiley & Sons, Ltd, Chichester. Flors, C. and Nonell, S. (2006) Acc. Chem. Res., 39, 293–300. (a) Cheng, H.M., Yomamoto, I., and Casida, J. (1972) J. Agric. Food Chem., 20, 850–855; (b) Cabras, P., Caboni, P., and Cabras, M. (2002) J. Agric. Food Chem., 50, 2576–2580; (c) Draper, W.M. (2002) Analyst, 127, 1370–1374. Waldeck, D.H. (1991) Chem. Rev., 91, 415–436. Spande, T.F., Jain, P., and Garraffo, H.M. (1999) J. Nat. Prod., 62, 5–21. Liu, C., Masuno, M.N., and MacMillan, J.B. (2004) Angew. Chem. Int. Ed., 43, 5951–5954. Toyota, M., Koyama, H., and Asawaka, Y. (1996) Phytochemistry, 41, 1347–1350. Majetich, G. and Yu, J. (2008) Org. Lett., 10, 89–91. Hurst, J.J. and Whitham, G.H. (1960) J. Chem. Soc., 2864–2869. Birch, A.J. and Russell, R.A. (1972) Tetrahedron, 28, 2999–3008. Hoffman, N. (2008) Chem. Rev., 108, 1052–1103.
931
932
24 Artifacts and Natural Substances Formed Spontaneously 56. Schuster, N., Christiansen, C., and
57.
58.
59.
60. 61. 62. 63.
64.
65.
66.
67. 68.
69. 70.
71.
72.
Jakupovic, J. (1993) Phytochemistry, 34, 1179–1181. Sharma, P. and Moses, J.E. (2010) Synlett, 525–528. See references citations therein for further examples. Rossi, M.H., Yoshida, M., and Maia, J.G.S. (1997) Phytochemistry, 45, 1263–1269. (a) Nicolaou, K.C., Sanchini, S., and Wu, T.R. (2010) Chem. Eur. J., 16, 7678–7682; (b) Tanaka, N., Okasaka, M., and Ishimaru, Y. (2005) Org. Lett., 7, 2997–2999; (c) Tanaka, N., Kashiwada, Y., and Kim, S.Y. (2009) J. Nat. Prod., 72, 1447–1452; (d) Nicolaou, K.C., Wu, T.R., and Sarlah, D. (2008) J. Am. Chem. Soc., 130, 11114–11121. Busch, B. and Hertweck, C. (2009) Phytochemistry, 70, 1833–1840. Hoffman, N. (2008) Chem. Rev., 108, 1052–1103. Han, B.H., Park, M.H., and Park, J.H. (1989) Pure Appl. Chem., 61, 443–448. Suzuki, M., Sano, M., and Yoshida, R. (2003) J. Agric. Food Chem., 51, 510–514. Asfaw, N., Storesund, H.J., and Skattebøl, L. (2001) Phytochemistry, 58, 489–492. Hayman, A.R., Perry, N.B., and Weavers, R.T. (1986) Phytochemistry, 26, 649–653. Perry, N.B., Benn, M.H., and Foster, L.M. (1996) Phytochemistry, 42, 453–459. Hollenbeak, K.H. and Kuehne, M.E. (1974) Tetrahedron, 30, 2307–2316. Numata, A., Hokimoto, K., and Takemura, T. (1984) Chem. Pharm. Bull., 32, 2815–2820. Peters, S. and Spiteller, P. (2007) J. Nat. Prod., 70, 1274–1277. Chadwick, L.R., Nikolic, D., and Burdette, J.E. (2004) J. Nat. Prod., 67, 2024–2032. Stevens, J.F., Taylor, A.W., and Deinzer, M.L. (1999) J. Chromatogr. A., 832, 97–107. Sulaiman, M., Martin, M.-T., and Paies, M. (1998) Phytochemistry, 49, 2191–2192.
73. Awang, K., Abdullah, Z., and Mukhtar,
74.
75.
76.
77.
78. 79. 80.
81. 82. 83. 84. 85.
86. 87. 88. 89. 90. 91.
92.
M.R. (2009) Nat. Prod. Res., 23, 652–658. Koch, M. (1965) Gentianine et swertziamarine de l’ Anthocleista procera Leprieur ex-Bureau (Loganiac´ees). Universit´e Paris dissertation. Dewick, P.M. (2009) Medicinal Natural Products: A Biosynthetic Approach, 3rd edn, John Wiley & Sons, Ltd, Chichester. (a) Herraiz, T. (2000) J. Chromatogr. A, 881, 483–499; (b) Nagatsu, T. (1997) Neurosci. Res., 29, 99–111; (c) Fekkes, D., Tuiten, A., and Bom, I. (2001) Neurosci. Lett., 303, 145–148. Hesse, M. (2002) Alkaloids: Nature’s Curse or Blessing? John Wiley & Sons, Ltd, Chichester. Ghisalberti, E.L. (2004) Fitoterapia, 75, 429–446. Szendrei, K., Varga, E., and Hadju, Z. (1988) J. Nat. Prod., 51, 993–995. Shamma, M., Chinnasamy, P., and Miana, G.A. (1979) J. Nat. Prod., 42, 615–623. Lounasmaa, M. and Hanhinen, P. (2004) J. Nat. Prod., 63, 1456–1460. Escolano, C. (2005) Angew. Chem. Int. Ed., 44, 7670–7673. Cheng, C., Liang, G., and Hu, C. (2008) Molecules, 13, 938–942. Li, C.-Y., Chow, T.J., and Wu, T.-S. (2005) J. Nat. Prod., 68, 1622–1624. de Pascoli, I.C., Nascimento, I.R., and Lopes, L.M.X. (2006) Phytochemistry, 67, 735–742. Kutrzeba, L.M., Li, X.-C., and Ding, Y. (2010) J. Nat. Prod., 73, 707–708. Zhao, J., Ma, M., and Wang, S. (2005) J. Nat. Prod., 68, 691–694. Orjala, J. and Gerwick, W.H. (1997) Phytochemistry, 45, 1087–1090. Picard, S. and Chenault, J. (1994) J. Nat. Prod., 57, 808–810. Orians, M. (1995) J. Chem. Ecol., 21, 1235–1243. Montgomery, C.T., Cassels, B.K., and Shamma, M. (1982) J. Nat. Prod., 46, 441–452. Fort, D.M., Litvak, J., and Chen, J.L. (1998) J. Nat. Prod., 61, 1528–1530; Erratum in: (1998) J. Nat. Prod., 61, 1577.
References 93. Aelterman, W., De Kimpe, N., and
94.
95. 96.
97.
98. 99. 100.
101.
102.
103.
104.
105. 106.
107.
108.
Kalinin, V. (1997) J. Nat. Prod., 60, 385–386. De Kimpe, N., Georgieva, A., and Boeykens, M. (1995) J. Org. Chem., 60, 5262–5265. Chia, Y., Chang, F., and Wang, J. (2008) Molecules, 13, 122–128. Arango, V., Robledo, S., and S´eon-M´eniel, B. (2010) J. Nat. Prod., 73, 1012–1014. Paranagama, P.A., Wijeratne, E.M.K., and Gunatilaka, A.A.L. (2007) J. Nat. Prod., 70, 1939–1945. Blackman, A.J., Dragar, C., and Wells, R.J. (1979) Aust. J. Chem., 32, 2783. Rochefort, S.J. and Capon, R.J. (1994) J. Nat. Prod., 57, 849–851. Khamidullina, E.A., Gromova, A.S., and Lutsky, V.I. (2006) Nat. Prod. Rep., 23, 117–129. Bermejo, A., Figad´ere, B., and Zafra-Polo, M.-C. (2005) Nat. Prod. Rep., 22, 269–303. Duret, P., Figad´ere, B., and Hocquemiller, R. (1997) Tetrahedron Lett., 38, 8849–8852. Duval, R. (2003) H´emisynth´ese et e´ valuation biologique d’analogs de la squamocine, une ac´etog´enine d’Annonaceae. Contribution a` l’´etude du m´ecanisme d’action. Universit´e Paris-Sud 11 dissertation. Li, D.-Y., Yu, J.-G., and Zhu, J.-X. (2001) J. Asian Nat. Prod. Res., 3, 267–276. Grass, S., Zidorn, C., and Ellmerer, E.P. (2004) Chem. Biodiv., 1, 353–360. (a) Luesch, H., Yoshida, W.Y., and Moore, R.E. (2002) Tetrahedron, 56, 7959–7967; (b) Sone, H., Kondo, T., and Kiryu, M. (1995) J. Org. Chem., 60, 4774–4781. (a) Braekman, J.C., Daloze, D., and Tavares, R. (2000) J. Nat. Prod., 63, 193–196; (b) Roberto, G. and Berlinck, S. (2002) Nat. Prod. Rep., 19, 617–649; (c) Faulkner, D.J., Newman, D.J., and Cragg, G.M. (2004) Nat. Prod. Rep., 21, 50–76. (a) Peng, J.N., Feng, X.-Z., and Liang, X.-T. (1999) J. Nat. Prod., 62, 611–612; (b) Yang, X.-W., Ma, Y.-L.,
109.
110.
111.
112. 113.
114.
115. 116. 117. 118. 119.
120.
121.
122.
and He, H.-P. (2006) J. Nat. Prod., 69, 971–974. (a) Okija, M., Yoshida, Y., and Okumura, M. (1990) J. Nat. Prod., 53, 1619–1622; (b) Kurihara, H., Mitani, T., and Kawabata, J. (1999) J. Nat. Prod., 62, 882–884; (c) Opatz, T., Kolshorn, H., and Thines, E. (2008) J. Nat. Prod., 71, 1973–1976; (d) Mohammed, M., Maxwell, A.R., and Ramsewak, R. (2010) Phytochem. Lett., 3, 29–32; (e) Wijeratne, E.M.K., Bashyal, B.P., and Gunatilaka, M.K. (2010) J. Nat. Prod., 73, 1156–1159. Isaka, M., Sappan, M., and Auncharoen, P. (2010) Phytochem. Lett., 3, 152–155. Anthoni, U., Bock, K., and Chevolot, L.C. (1987) J. Org. Chem., 52, 5638–5639. Hamann, M.T. and Scheuer, P. (1993) J. Org. Chem., 58, 6565–6569. Bourguet-Kondracki, M.L., Lacombe, F., and Guyot, M. (1999) J. Nat. Prod., 62, 1304–1305. (a) Sun, L.-R., Li, X., and Wang, S.-X. (2005) J. Asian Nat. Prod. Res., 7, 127–130; (b) Piggott, A.M. and Karuso, P. (2005) Molecules, 10, 1292–1297; (c) Kuroda, C., Kiuchi, K., and Torihata, A. (2007) Chem. Biodiv., 4, 2210–2217. Perry, N.B., Burgess, E.J., and Foster, L.M. (2008) J. Nat. Prod., 71, 258–261. Ha, D.T., Ngoc, T.M., and Lee, I.-S. (2009) J. Nat. Prod., 72, 1465–1470. Cheng, Y., Cheng, P., and Zhang, H. (2010) Helv. Chim. Acta., 93, 565–572. Kita, M., Kigoshi, H., and Uemura, D. (2001) J. Nat. Prod., 64, 1090–1092. Kawai, H., Kuroyanagi, M., and Ueno, A. (1988) Chem. Pharm. Bull., 36, 3664–3666. Tomassini, L., Cometa, M.F., and Nicoletti, M. (1995) J. Nat. Prod., 58, 1756–1758. Zedro, C., Bohlmann, F., and Niemeyer, H.M. (1988) J. Nat. Prod., 51, 509–512. Ciminiello, P., Costantino, V., and Fattorusso, E. (1994) J. Nat. Prod., 65, 705–712.
933
934
24 Artifacts and Natural Substances Formed Spontaneously 123. Saeki, B.M., Granato, A.C., and
124.
125.
126. 127.
Berlinck, R.G.S. (2002) J. Nat. Prod., 65, 796–799. Cone, E.J., Buchwald, W.F., and Darwin, W.D. (1982) Drug Metabol. Dispos., 10, 561–567. Mauden, K.E., Will, S.M.R., and Lambert, W.E. (2007) J. Chromatogr. B, 848, 384–390. Gribble, G.W. (1999) Chem. Soc. Rev., 28, 335–346. Neuwinger, H.D. (1996) African Ethnobotany: Poisons and Drugs, Chapman & Hall, Weinheim.
128. Guinaudeau, H., Leb´euf, M., and
129. 130.
131. 132.
Cav´e, A. (1994) J. Nat. Prod., 57, 1033–1135. Hu, J., Wang, W.-D., and Shen, Y.-H. (2007) Helv. Chim. Acta, 90, 720–722. Marek, R., Seˇck`arova` , P., and Hulova` , D. (2003) J. Nat. Prod., 66, 481–486. Grycov`a, L., Dost`al, J., and Marek, R. (2007) Phytochemistry, 68, 150–175. Bentley, K.W. (2000) Nat. Prod. Rep., 17, 247–268.
935
Index
a abyssomicins 512–513 acalyphidin production 662 acetals formation 913–916 – by acetone under acidic conditions 919 – secoiridoid acetals from Lonicera spp 918 acetone-derived artifacts 916–919 acetoxy-acetalic sesquiterpenes, artifactual alcoholysis of 916 acetyl co-enzyme A (AcCoA) 594 acid-catalyzed reactions 900–903 – coumarins from Rutaceae spp 904 – cryptolepine from 902 – intramolecular Michael additions 905 – of natural products 901 – norrhoedanines in acidic conditions 901 – protic solvents 903–906 – spiroketal from laurencione 903 acrolein scenario 182–200 – endo-intramolecular Diels–Alder reaction 198–199 – keramaphidin skeleton conversion into ircinal/manzamine skeleton 200 – transannular hydride transfers 199–200 Actephila excelsa 769 actinorhodin 486 acyl carrier protein (ACP) 473 acyl transfer – Scott’s conditions for 475 – using glycoluril 478 acylphloroglucinols 434–436 Adler–Becker oxidation 727 Aeschynomene mimosifolia 747 Agelas wiedenmayeri 231 agelastatins 250–253 – Wardrop’s synthesis 252 ageliferins 254–255 ajmaline/sarpagine series 897
(±)-akuammicine 102 aldol condensation 360 aldolization–crotonization process 286 aldotripiperideine 12 alkaline media 888–895 – amination processes 888–892 alkaloids, I–X – arginine-derived: see in Chapter 1 – FR-901483: see in Chapter 2 – guanidinium alkaloids: chapter 7 – lysine-derived: see in Chapter 1 – manzamine alkaloids: Chapter 6 – non aminoacid derived alkaloids: Chapter 8 – ornithine-derived: see in Chapter 1 – peptide alkaloids: – – aryl-peptide alkaloids: see in Chapter 9 – – azole-peptide alkaloids: see in Chapter 9 – – complex peptide alkaloids: see in Chapter 10 – – indole-oxidized peptide alkaloids: see in Chapter 10 – pyrrole-2-aminoimidazole: Chapter 7 – TAN-1251: see in Chapter 2 – tryptophan-derived – – dioxopiperazine alkaloids: Chapter 4 – – indolemonoterpene alkaloids: Chapter 3 – – modified indole nucleus: Chapter 5 – tyrosine-derived: Chapter 2 3-alkylpiperidines 181–182, see also manzamine alkaloids, biomimetic synthesis alkylpyridines with unusual linking patterns 194–195 – pyrinadine A, biomimetic synthesis 195 – pyrinodemin A, biomimetic synthesis of 194 3-Alkylpyridiniums 191 allicin 855 Alternaria solani 507
Biomimetic Organic Synthesis, First Edition. Edited by Erwan Poupon and Bastien Nay. 2011 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2011 by Wiley-VCH Verlag GmbH & Co. KGaA.
936
Index amauromine, Danishefsky’s total synthesis 126 amide coupling 385 amination processes 888–892 amino acids – in primary metabolism 4 – from primary metabolism to alkaloid biosynthesis 5–6 – structure of 4 4-aminobutyraldehyde 9 aminopentadienals 190, 213–215 aminopentadienimines 189 ampelopsin D 710–711 ampelopsin F 713 amurensin H 715 anchinopeptolides 26 – synthesis 26–29 anethole 871 anhydrovinblastine series 110–113 anigorufone 768 Anigozanthos preissii 768 Annona muricata 908 annonaceous acetogenins 907 antheridic acid from gibberellins 408–410 anthracenoids 494–495 – biomimetic access to 494–495 – chrysophanol, Harris’ biomimetic synthesis 495 – emodin, Harris’ biomimetic synthesis 495 – nanaomycine, Yamaguchi’s biomimetic synthesis 494 – Yamaguchi’s access to 497 Anthromyces ramosus 699 anticancer vinblastine series 110–113 antirhine derivatives 99 Aplidium longithorax 777 alkylations, apparent 910–913 – chartelline A, methanol substitution of 913 – puupehenone, methanol adduct on 914 – xanthepinone, methanol induced rearrangement of 912 aquaticol 772 L-arginine 3 arginine, alkaloids derived from 18–30 Aristolochia heterophylla 921 Aristolochia lagesiana 895 Aristolochia pubescens 895 Aristolochia rodriguesii 919 Aristotelia alkaloids 93–96 (+)-aristoteline (16) 95 aromatic polyketides 485–499, see also anthracenoid derivatives; benzenoid
derivatives; naphthalenoid derivatives; tetracyclic derivatives – Collie and coworkers 485 – cyclization towards chrysophanol 487 – folding leading to 486 – pathways leading to 486 aromatic rings, biomimetic access to 471–499 Artemisia absinthium 407 artifacts and natural substances formation 849–930, see also hydrodistillation; hydrolysis, artifactual; oxidation processes – acetals formation 913–916 – acetone-derived artifacts 916–919 – acid-catalyzed reactions 900–903 – acidic conditions during purifications 895–903 – ajmaline/sarpagine series 897 – alkylations 910–913 – alcoholysis 916 – base-catalyzed reactions 892–895 – botanical misidentifications 851 – chloroformic adducts 926 – decarboxylation processes 885 – ecdysteroids 892 – esterification, transesterification 908–910 – by exposure to light 870–878, see also light – gentianine 891 – glucosidases 852–853 – halogenated solvents 919–921 – heat and pressure in 878–888 – – biyouyanagins A and B 880 – – matricin degradation 887 – – sanguinone A degradation 887 – – styrylpyrone photodimers 879 – hemiterpenes 853 – imide-bearing aconitine-like alkaloids 893 – methyl-ester hydrolysis 900 – miscellaneous methylation artifacts 911 – miscellaneous oxidation artifacts 861 – phenylchromane derivatives from Desmos spp 890 – protic solvents 903–906 – protoberberines 921–925 – raucaffrinoline 896 – Rubiaceae iridoids 910 – supercritical CO2 885–888 – transesterification artifacts 909 – (Z)-vellomisine 894 aryl-containing peptide alkaloids 317–350 – cyclopeptides containing biaryls 344–345 Asarum heterotropoides 895 Asarum teitonense 769 asatone 772
Index Ascochyta rabiei 507 Aspergillus nidulans 755 Aspergillus ochraceus 135, 158 Aspergillus terreus 506, 755 Aspidosperma alkaloids 106 – fragmentation 106–109 – rearrangements of 106–109 aspochalasin Z 764 autocatalytic enantioselective reactions 825–833 – of asymmetric amplification 827 – density functional theory (DFT) calculations 829 – dissipative non-productive reaction cycle 828 – Kagan and Noyori models of 826 – reversible Frank-type mechanism 828 – Soai reaction 830 – – Blackmond–Brown model in 831 avrainvillamide 135–136 Axinella vaceletti 233 axinellamines A/B 257, 262–263 azaspiracids 291–293 – azaspiracid-1, 3D structure of 292 – FGHI rings of 292 – HI rings of, Forsyth’s biomimetic approach to 293 – structure of 292 aziridinium mechanism 259 – for massadine formation 259 – for tetramer stylissadine A formation 259–261 azole-containing peptide alkaloids 321–336, see also lissoclinamides; thiangazole; thiostrepton – biomimetic cyclodehydration 324 – enzymatic heterocycle formation 324 – GE2270A 334–336 – structural features 321–323 – synthesis of, completion 333
b Baeyer–Villiger rearrangement 740 Baldwin’s hypothesis development 195–200 – from cyclostellettamines to keramaphidin-type alkaloids 195–200 – pyridinium alkaloids and manzamine A-type alkaloids, linking 195–197 barakol, Harris’ biomimetic synthesis of 493 barbaline 294 basidiolides 425 bastadin biosynthesis 342–343
benzenoids 487–492 – biomimetic access to 487–492 – masked hexaketides, Schmidt’s condensation of 492 – pentacarbonyl derivative, Harris’ biomimetic cyclizations of 492 – β-polyketones synthesis 488 – pyrones 489 – tetra-β-carbonyl compounds 489 – tri-β-carbonyl entities synthesis 489 – β-triketoacids 490 – zearalenone, Barrett’s synthesis of 491 benzo[a]tetracenoid derivatives 498–499 benzo[kl]xanthenes 686–688 – manganese-mediated biomimetic synthesis 688 benzodiazepinones, hydrogenation 806 benzophenanthridines 929 benzosceptrin A 266 benzothiazines, hydrogenation 806 benzoxanthenone lignans 683–686 – biomimetic synthesis of 683–686 – carpanones 683 – mechanism 689–670 benzoxazines, hydrogenation 806–813 benzoxazinones, hydrogenation 806 berberine 923 betulin 423 biaryls, cyclopeptides containing 344–345 – biphenomycins A–C 344 bicyclo[2.2.2]diazaoctanes 126–141 – hetero-Diels–Alder formation of 127 – Sammes’ model study 127 bicyclo[3.3.1]nonane-2,4,9-triones 448 bielschowskysin 416 BINOL-derivative, Brook-type-rearrangement 791 bio-inspired transfer hydrogenations 787–818 – benzodiazepinones 806 – benzothiazines 806 – benzoxazines 806–813 – benzoxazinones 806 – BINOL-phosphoric acid derivatives, synthesis 792 – Brønsted acid catalyzed transfer hydrogenation 788–799 – cascade sequences 814–817 – dehydrogenases as model 787–788 – diazepines 806 – enamides 798 – α-imino esters 796
937
938
Index bio-inspired transfer hydrogenations (contd.) – indoles 805–806 – ketimines 793 – piperidines 813 – pyridines 813–814 – quinoxalines 806 – quinoxalinones 806 biological homochirality 823–841, see also autocatalytic enantioselective reactions – chiral surfaces, adsorption on 837 – collision kinetics, symmetry breaking in 840 – conglomerate crystallizations, spontaneous symmetry breaking in 837–840 – enantioenrichment, polymerization and aggregation models of 834–835 – membrane diffusion, symmetry breaking in 840 – phase equilibria 835–837 – Plasson mechanism 835 – reaction–diffusion models, symmetry breaking in 840 – self-replication and 833–834 biphenomycins biaryls 344–345 bipinnatin J, photochemical rearrangement of 414 bisacridones from Rutaceae 866 bis(2-oxo-3-oxazolindinyl)phosphinic chloride (BOPCl) 133 bis-aporphines 864 bisorbicillinoids 772–773 bisorbicillinol 726 – Nicolaou and Pettus’ synthesis 726 – from sorbicillin 726 bis-steroidal pyrazines 299–300 – biomimetic pseudo-combinatorial approach to 299 – unsymmetrical approaches to 300 biyouyanagins A and B 880 Black’s cascade 612 Blackmond–Brown model 831 (±)-borreverine 173–175 – biosynthesis proposal by Koch et al. 173 botanical misidentifications 851 bouvardin synthesis 342 brazilide A 735 brevetoxin B 545 – Nakanishi’s hypothesis 549 brevetoxins 580–582 (±)-brevianamides 155–158 – brevianamide A 875–876 – brevianamide B, biosynthesis – – by Williams et al. 157, 159
brevianamides 118 – biosynthesis of 118 – – brevianamide B 133 – – brevianamide E 119 – – brevianamide F 119 – biosynthetic proposal for 128 – brevianamide F 118 – – Danishefsky’s total synthesis of 120 – – Kametani’s total synthesis of 120 Brønsted acid catalyzed transfer hydrogenation 788–799 bukittinggine 303 bukittinggine, Heathcock’s synthesis of 303 Burgess cyclodehydration 325
c caffeic acid phenethyl ester (CAPE) 687 caged xanthones 452 – Garcinia xanthones, biomimetic synthesis 455–464 – – non-biomimetic synthesis of 460–463 – – Theodorakis’ unified approach to 459 Callyspongia species 186 calycanthines 168–171 – biomimetic synthesis – – by Scott et al. 169 – – by Stoltz et al. 169–170 – biosynthesis proposal by Woodward and Robinson 169 camphene 397–399 camphorsulfonic acid (CSA) 627 camptothecin 93, 111, 152 – biosynthesis by Hutchinson et al. 153 – (±)-camptothecin 150–154 – synthesis by Winterfeldt et al. 153 Cane–Celmer–Westley hypothesis 540–541, 554 cararosinol C and D 713 carbinolamine formation 385 carpanone 683–685, 724 – Chapman’s synthesis of 725 – related sequence for dehydrodiisoeugenol 725 Carpinus tschonoskii 663 caryolane, biomimetic studies in 402–404 caryophyllenes in sesquiterpene biosyntheses 401–402 cascade reactions, biomimetic 524–530 – hirsutellones 525–530 – tetronasin synthesis 524–525 cascade sequences – hydrogenation 814–817
Index – – Brønsted acid catalyzed 815 – oxidative dearomatization 731–733 cassiarins A and B 284–285 – Yao’s biomimetic synthesis of 286 cassigarol B 716 castalagin 666, 670 – pyrolytic degradations of 671 Castanea crenata 669 catechol oxidation 768–775 (±)-cedrene 731 celogentin C 357, 363–368 – A-ring Trp-Leu linkage origin 365 – bioactivity 363 – B-ring Trp-His linkage formation 365 – C–H activation–indolylation 367–368 – isolation 363 – retrobiosynthetic proposal for 364 – structures of 363 – synthetic approaches to 366 cephalostatins 294–298 – structures 295 Ceratosoma brevicaudatum 855 cermizine C 51–52 C-glycosidic ellagitannins 663–670 – oxidation of 669–670 – reactions at C1 positions 665–669 chaetoglobosin 761–763 Chamaecyparis obtusa 769, 774 Chartella papyracea 160 chartelline A, methanol substitution of 913 chartelline C – biosynthesis proposal for 162 – (±)-Chartelline C 160–164 – endgame of synthesis 164 – synthetic route by Baran et al. 163 chebulagic acid synthesis 662–664 Chichibabin synthesis of pyridines 188 chimonanthines 168–171 chiral surfaces, adsorption on 837 chloptosin 369–374 – chloptosin pyrroloindole core synthesis 373 – retrobiosynthetic simplification of 370 – synthetic approaches to 372 chloroformic adducts 926 chloropeptin I 391 chloropeptin II/complestatin 391 chlorothricin 763–764 Chrossopetalum rhacoma 774 Claisen/Diels–Alder/Claisen reaction cascade 456–457 clathrins 857 clathrodins 227 – biogenetic hypothesis for 229–233
– post-clathrodin in P-2-AIs biogenesis 228 – pre-clathrodin in P-2-AIs biogenesis 228 clovane series, biomimetic studies in 402–404 clusianone, total synthesis of 448–451 – (±)-clusianone 438–439 – – double Michael reaction 438–439 – – Porco synthesis 438 – by Marazano and coworkers 451 – by Simpkins group 449 – through ‘carbanions’ differentiation 443–445 cochleamycins 514–521 – biosynthetic hypothesis for 518 – isolation 517 – Paquette’s partial synthesis of 519 – Roush synthesis of 519 collision kinetics, symmetry breaking in 840 colombiasin A – (−)-colombiasin A 412 – Nicolaou’s synthesis of 412 – Rychnovsky’s synthesis 413 communesins 168–171 – biomimetic synthesis by Stoltz et al. 169–170 – study by Funk et al. 171 complanadine A, total synthesis of 53–54 complex peptide alkaloids 357–392 complex terpenoids, biomimetic rearrangements of 397–428, see also monoterpene rearrangements – miscellaneous diterpenes 417–420 conglomerate crystallizations, spontaneous symmetry breaking in 837–840 Cordia globifera 425 cordiachrome C 427 coriariin A 659 corilagin 653 Corynanthe alkaloids 95–99 Corynanthe skeleton into Strychnos skeleton 99–102 CP-225917 505–506 CP-263114 505–506 cryptolepine from 902 Cryptolepis sanguinolenta 901 CuI-mediated cyclization 347 curcuphenol 731–732 Cutleria multifida 616 (±)-cycloanchinopeptolide D 28 cyanophenyloxazolopiperidine 15 cyclic imine marine alkaloids 275–284
939
940
Index cyclic peptides containing aryl-alkyl ethers 336–339 cyclic peptides containing biaryl ethers 339–343 – bastadin biosynthesis 342–343 – bouvardin synthesis 342 – eurypamide B synthesis 341 – isodityrosine subunit 340 – – synthesis strategies 340 – K-13 synthesis 342 cyclooroidin 238 cyclopamine 422 cyclostellettamine alkaloids 215–217 – cyclostellettamine B 191 cyclostreptin 514 cytochalasin D 762
dehydrohexahydroxydiphenoyl (DHHDP) ester 645 dehydropiperidines 332 Delphinium cashmirianum 892 density functional theory (DFT) 829, 832 deoxybrevianamide E, Kametani’s total synthesis of 120 Desmos dumosus 888 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) 335 diazepines, hydrogenation 806 diazonamides 375–382 – aminal core formation 377 – biaryl coupling 377 – 2,4-bisoxazole core, biomimetic synthesis of 379 – bisoxazole ring system via oxidative d dehydrative cyclization 379 dalesconols 715 – diazonamide A 357, 375 Danishefsky’s synthesis 120 – α-hydroxyvaline formation 377 – of amauromine 126 – indole–indole coupling 381–382 – of spirotryprostatins 122–123 – late-stage aromatic chlorination daphniglaucine A 294 378–379 daphnilactone A 304 – oxidative annulation 379–380 Daphniphyllum alkaloids 298–305 – reductive aminal formation 380–381 – daphnilactone A 304 – retrobiosynthetic analysis of 376 – Heathcock group in 300 – sequential nucleophilic 1,2-addition, – methyl homodaphniphyllate 304 electrophilic aromatic substitution 380 – methyl homosecodaphniphyllate dibromoagelaspongin 238–243 synthesis 301–302 – biomimetic hypotheses for 241 – proto-daphniphylline biosynthesis 301 – rac-dibromoagelaspongin dearomatization, see oxidative dearomatization – – Al-Mourabit’s approach to 244 debromodispacamides B, Al-Mourabit’s – – Feldman’s total synthesis of 242 biomimetic synthesis 236–237 dibromophakellin 243–247 debromodispacamides D, Al-Mourabit’s – Nagasawa’s enantioselective synthesis of biomimetic synthesis 236–238 249 decahydroquinoline alkaloids 285–288 – Romo’s enantioselective synthesis – biosynthetic origin of 287 of 248 – cis-195A, Amat’s biomimetic synthesis of – synthesis from dihydrooroidin 233–234 287 dibromophakellstatin 243–247 – Dendrobates histrionicus 286 – Feldman’s synthesis of 245 – Dendrobates pumilio 285 – Lindel’s synthesis of decalin systems 506–509 (−)-dibromophakellstatin 246 – Diels–Alder reactions affording – Romo’s synthesis of 506–509 (+)-dibromophakellstatin 246 decarboxylation processes 885 2,3-dichloro-5,6-dicyano-1,4-benzoquinone decarboxylative Grob-type anti-elimination (DDQ) 158, 292 360 (+)-11,11-Dideoxyverticillin A 141, 144, dehydrodiisoeugenol 725 171–173 dehydroellagitannins 648 – dimerization synthesis strategy by dehydroellagitannins conversion into related Movassaghi et al. 172 ellagitannins 659–663 – endgame of synthesis by Movassaghi et al. dehydrohexahydroxydiphenic acid esters synthesis from methyl gallate 652 173
Index Diels–Alder reaction/cycloaddition 124–131, 461, 506–524, 753–782 – [4 + 2] adducts derived from terpenoid dienes 780 – after reactive substrates formation by oxidation enzymes 767–779 – anigorufone 768 – aspochalasin Z 764 – bicyclo[2.2.2]diazaoctanes formation 127 – biomimetic Diels–Alder reaction 129 – biomimetic syntheses involving 506–524 – catechol oxidation 768–775 – chaetoglobosin 761–763 – chlorothricin 763–764 – cyclization of trienes 507 – cyclopentadiene formation 779 – cytochalasin D 762 – decalin systems 506–509 – equisetin 761–763 – indanomycin 764–766 – intramolecular, catalysis of 760 – kijanimicin 763–764 – lachanthocarpone 768 – longithorone 779 – lovastatin nonaketide synthase (LovB) 755–756 – macrophomate synthase 756–758 – nargenicin A 508–509 – in nature 754–760 – ortho-quinol formation 767 – ortho-quinone formation 767 – phenol oxidation 768–775 – prenyl side chain dehydrogenation, conjugated diene derived from 775–779 – solanapyrone synthase (SPS) 758–760 – solanapyrones 507–508 – spinosyn 766–767 – spiro systems 512–514 – superstolide A 509 – tetrahydroindane systems 509–512 – tetrocarcin A 763–764 – transannular Diels–Alder (TADA) reaction 508 diethyl azodicarboxylate (DEAD) 135 diffusion ordered spectroscopy (DOSY) 481 dihydrooroidin, dibromophakellin synthesis from 233–234 dihydropyridines – biomimetic synthesis 188–189 – chemistry, Baldwin’s hypothesis 186 – isomerization of 189 – Marazano biomimetic synthesis of 188
dihydropyridinium salts, biomimetic synthesis 188–189 diisopropyl-carbodiimide (DIC) mediated amide bond formation 152 dimerization 10–11, 173–174 – dimeric ellagitannin synthesis 658–659 – dimeric indolomonoterpene alkaloids 110–113 – intermolecular 724–727 – process towards isoglaucanic acid 504–505 – in pyrrolidine series 10 – in pyrroloindole peptide alkaloids 370 dimers (and oligomers) 184 dimethylallyl pyrophosphate (DMAPP) 129 dioxepandehydrothyrsiferol, epoxide-opening cascades in 546 dioxoaporphines 868 dioxopiperazine alkaloids 117–147, see also bicyclo[2.2.2]diazaoctanes; Prenylated indole alkaloids – derived from tryptophan and amino acids 122–126 dioxopiperazines – synthesis 131 – from tryptophan and proline 119–122 1,1-diphenyl-2-picrylhydrazyl (DPPH) 772 diptoindonesin D 714 dirigent protein 679–680 discorhabdins 154–155 – biosynthesis proposed by Munro et al. 155 – discorhabdin B 856 – (±)-discorhabdin C and E 154–155 – – synthesis by Heathcock et al. 156 dispacamide A, Al-Mourabit’s synthesis of 238 dissipative non-productive reaction cycle 828 diterpene rearrangements 408–420 – antheridic acid biomimetic synthesis from gibberellins 408–410 – early chemical diversity in 408 – miscellaneous diterpenes 417–420 (−)-ditryptophenaline 143 double Michael reaction 438–439 Durio zibethinus 855 Dysidea herbacea 859 Dysidea pallescens 859
e ecdysteroids, alumina-catalyzed dehydration of 892 ecteinascidin 743 (ET 743) 382–391 – biomimetic strategy 383–385 – biosynthesis 383–385
941
942
Index ecteinascidin 743 (ET 743) (contd.) – bridge formation 389–390 – – Corey’s strategy for 389 – pentacycle formation 385–389 – – Corey’s approach to 386 – – Danishefsky’s synthesis of 387 – – Fukuyama’s work 387 – – Williams’ synthesis of 386 – – Zhu’s approach 387 – proposed biosynthesis for 384 – saframycin A biosynthesis 383 – synthetic approaches to 385 Elaeocarpus alkaloids, biomimetic syntheses of 19–22 electrophilic aromatic substitution 380 6π electrocyclizations, polyketides 598–612, see also tridachiahydropyrones eleutherinol, Harris’ biomimetic synthesis of 493 elisapterosin B 412 – Rychnovsky’s synthesis 413 ellagitannins 639–640, 642–659 – synthesis with 3,6-(R)-HHDP group 651 – biosynthesis 640–642 – with 1 C4 glucopyranose cores 645–651 – corilagin 653 – decomposition 640 – dehydroellagitannins conversion into 659–663 – – acalyphidin production 662 – – benzyl-protected dehydrodigallic acid synthesis 660 – – castalagin 666 – – C-glycosidic ellagitannins 665 – – chebulagic acid synthesis 662–663 – – dehydrodigallic acid derivative synthesis 662 – – DHHDP esters reduction 659–662 – – DHHDP esters 663 – – dimeric ellagitannin coriariin A 661 – – mallotusinin production 662 – – pyranose-type ellagitannins 665 – – thiol compounds reaction 662–663 – – vescalagin 666 – dehydrohexahydroxydiphenic acid esters synthesis from methyl gallate 652 – 2,4-DHHDP esters 647–649 – dimeric ellagitannin synthesis 658–659 – epigallocatechin gallate oxidation during tea fermentation 650 – hexahydroxydiphenic acid, double esterification of 651–658 – 2,4-HHDP ester 647–649 – pedunculagin 658
– synthesis by biaryl coupling of galloyl esters 642–645 – tellimagrandin I 654 – Ullmann-type biaryl couplings 646 ellipticine 103 ellipticine-type alkaloids 102–105 elysiapyrones 624–626 emodin, Harris’ biomimetic synthesis 495 – chrysophanol, Harris’ biomimetic synthesis 495 enamides – Brønsted acid catalyzed transfer hydrogenation of 788–799 – hydrogenation 798 enantioenrichment – polymerization and aggregation models of 834–835 – W¨urthner’s model 835 endiandric acids 612–618 – Nicolaou’s biomimetic synthesis 617 endocyclic enamines 12–13, 39 7-endo epoxide ring opening 360 endo-intramolecular Diels–Alder reaction 198–199 enshuol 562 ent-17-epialantrypinone 142 ent-alantrypinone 118, 142 epidithiodioxopiperazines 141–146 – ent-alantrypinone synthesis 142 epimerization 895–897 – lactonic compounds 905–908 – light-induced 870–872 – – furofuranic lignans under 899 – – of gallocatechins during tea brewing 883 – – in or out of solutions 880 epinitraramine 37 epoxide-opening cascades in polycyclic polyethers 550–583 – bis-tetrahydrofurans synthesis via 556 – enshuol 562 – ent-abudinol B synthesis via 564 – enzymatic ester hydrolysis 555 – first-generation approach to 557 – glabrescol 561 – ladder polyethers synthesis 565–583 – omaezakianol synthesis via 563 – polyether ionophores synthesis 550–554 – – applications of 554–558 – – bis-tetrahydrofurans 553 – – 2,5-linked tetrahydrofurans 553 – second-generation approach to 557 – single-electron oxidation of homobenzylic ethers 555
Index – in squalene-derived polyethers synthesis 558–565 – third-generation approach to 557 epoxide-opening reactions – Baldwin’s rules in 538–539 – regioselectivity control in 539 epoxyquinols A–C 615 epoxysorbicillinol 741 epoxytwinol 615 equisetin 761–763 erinacine E, Nakada’s biomimetic synthesis of 426 ervatamine alkaloids 102–105 ervitsine alkaloids 102–105 erythromycin 595 eurypamide B synthesis 341 eusynstyelamide A 26 exiguamines 737–738
– bipinnatin J as precursor 415 furofuran lignans, biomimetic synthesis of 681–683 – enzyme-mediated 682–683 – metal-catalyzed approaches 682
g
Galbulimima alkaloids 271–275, 509 – Baldwin’s biomimetic synthesis of 511 – Class I 272–273 – – Baldwin’s biosynthetic hypothesis for 272, 274 – Class II 273–275 – – Movassaghi’s biosynthetic hypothesis for 275–276 – Class III 273–275 – – Movassaghi’s biosynthetic hypothesis for 275–276 galiellalactones 511–512 f galloyl esters 642–645 fastigiatine, total synthesis of 52–53 gambogin synthesis by Nicolaou group fatty acid biosynthesis 594–597 458–459 fatty acid synthases (FASs) 473 Garcinia forbesii 452 ficuseptine 25–26 Garcinia hanburyi 452 Fischerella muscicola 859 Garcinia subelliptica 434 (−)-Fischerindole I 164–166 Garcinia xanthones, biomimetic synthesis fissoldhimine 22–25 455–464 – biogenetically inspired heterodimerization gardenamide 293–294 toward 24 garsubellin A, total synthesis of 441–443 – biosynthetic hypotheses 23 – by Danishefsky et al. 442 – structures 23 – by Shibasaki group 442 flavin mononucleotide (FMN) 679 – by Simpkins group 450 forbesione, biomimetic synthesis of 456 GE2270A 334–336 – Nicolaou approach to 458–459 – Nicolaou and Bach works 335 – via Claisen/Diels–Alder/Claisen reaction geissoschizine 101 cascade 456 (−)-Gelselegine 166–168 FR182877 514–521 – biosynthesis proposal by Sakai et al. 167 – acyclic system related to 517 Gelsemium elegans 166 – biosynthetic origin for 515 gentianine 891 – large-scale synthesis of 516 Geranium thunbergii 648 – Sorensen’s biomimetic synthesis of 515 Gibbs’ phase rule 836 FR-901483 compounds 61–86 Gibbs–Thomson rule 838 – Ciufolini synthesis of 80–86 GKK1032 compounds – Snider synthesis of 64–67 – biosynthetic origin of 528 – – aldol step in 66 – cyclization mechanism 528 – Sorensen synthesis of 78–79 – Oikawa’s hypothesis for 529 – synthesis via oxidative amidation chemistry glabrescol 561 77–86 gliotoxin 118, 145 – total syntheses of 63–71 – Kishi’s total synthesis of 146 – Wardrop approach to 77 globiferin, biomimetic conversion into fredericamycins 743–744 cordiachrome C 427 frondosins 745 glucosidases, as by-products formation furanocembranoids, biomimetic relationships triggers 852–853 among 414–417 glucosinolates, hydrolysis of 854
943
944
Index glutaconaldehydes 213–215 glutacondialdehydes 190 glutamate dehydrogenase (GDH) 787 glutaraldehyde – alkaloid skeletons from 13–15 – condensations of 14 glycosidation 347 grandione 768 griseorhodin A 743 guanidinium alkaloids 225–267 – biomimetic synthesis of 225–267 gymnodimines 282–284, 514 – Kishi’s biomimetic approach to 283 – plausible biosynthetic origin of 283 – structure 283 gypsetin 118, 126–127 – synthesis of 127
h halicyclamines 201–203 – Baldwin–Marazano concepts 207 – biomimetic models toward 205–208 – first generation approach to 206 – halicyclamine A, biomimetic synthesis – second generation approach to 206 Halocarpus biformis 885 Haloxylon salicornicum 12 heliocides 770 hemibrevetoxins 580–582 hemiterpenes 853 Hericium erinaceum 424 hetero-Diels–Alder formation of bicyclo[2.2.2]diazaoctanes 127 heterosides hydrolysis 900 Heteroyohimbines 95–99 hexacyclinic acid 514–521 – biosynthetic origin for 515 hexafluoro-isopropanol (HFIP) 730 hexahydroxydiphenic acid (HHDP) double esterification of 651–658 himandravine 509 himastatin 357, 369–374 – himastatin pyrroloindole core synthesis 372–373 – synthetic approaches to 372 himbacine 272, 509 himbeline 509 Hirsutella nivea 525 hirsutellones 525–530 – 6,5,6-fused system of 525–530 – macrocycle of 525–530 – Nicolaou’s total synthesis of 529 – structure of 527 (±)-hobartine 95
homo-Wagner–Meerwein transposition 734 hopeahainol A 708 hopeanol 708 Horner–Wadsworth–Emmons type olefinations 341 horse radish peroxidase (HRP) 679 Humulus lupulus, MPAPs from 435 Husson’s strategy (modified Polonovski reaction) 188 Hutchinson’s biosynthesis 152 hydrodistillation 880–885 – artifacts from (+)-chrysanthenone 884 – lactones of Halocarpus biformis formed during 886 – polyene splicing 881 – Zizyphus jujuba seeds 882 hydrolysis, artifactual 897–900 – cubebin anomers from Aristolochia spp 900 – of heterosides 900 – methyl-ester hydrolysis 900 – stephacidin B on silica gel 898 hydroperoxides, artifacts from 858 207 6-hydroxymusizin, Harris’ biomimetic synthesis 493 hymenialdisines 247–250 hymenin 249 (−)-hyperforin, total synthesis 445–448 – catalytic asymmetric synthesis of 446 – ent-hyperforin 447 hyperguinone B 440–441 Hypericum chinense 878 Hypericum papuanum 434, 440 Hypericum perforatum 434
i Iboga alkaloids 106 (±)-ialibinone A and B 440–441 imidazole, biomimetic conditions using 476 imide-bearing aconitine-like alkaloids 893 imines, Brønsted acid catalyzed transfer hydrogenation of 788–799 imino esters, Brønsted acid catalyzed transfer hydrogenation of 788–799 α-imino esters, hydrogenation 796 – organocatalytic asymmetric transfer 797 indanomycin 764–766 indole alkaloids 149–175, see also modified indole nucleus alkaloids – indole nucleus conversion into first derivatives 150 indole–indole coupling 381–382 – Witkop-type photo-induced macrocyclization 382
Index indolemonoterpene alkaloids 91–113 – botanical distribution 91–93 – classification 91–93 indole-oxidized cyclopeptides 357–382, see also chloptosin; himastatin – celogentin C 357, 363–368, see also individual entry – indolyl–phenyl coupling 359 – NCS mediated oxidative coupling 368 – TMC-95A-D 357–363 indoles, hydrogenation 805–806 – asymmetric brønsted acid catalyzed 805–806 indolomonoterpenes 867 indolomonoterpenic alkaloid macrosalhine, quaternization of 923 intramolecular Diels–Alder (IMDA) cycloaddition 138, 273, 506, 754 intramolecular Heck reaction 385 iodotrimethylsilane (TMSI) 126 ircinal A, biogenesis 210 ircinal alkaloids 200–201 – (4 + 2) cycloaddition strategy towards an ircinal model 203 isatisine A 174 islandicin 486 isoacetogenins 905 isoampelopsin D 711 isoanhydrovinblastine 112 (±)-isoborreverine 173–175 – biosynthesis proposal by Koch et al. 173 isocaryophyllene 404 isoglaucanic acid, dimerization process towards 504–505 isomerization, light-induced 870–872 – anethole 871 – in or out of solutions 880 – stilbenoids 871
j jasminiflorine 174 juliprosine 25–26 juliprosopine 25–26 Juncus acutus 774
k K-13 synthesis 342 Kametani’s total synthesis 120 kapakahine A 391 Karenia brevis 545, 548 keramaphidin alkaloids 200–201 keramaphidin B, biomimetic total synthesis 197–198 – Baldwin’s hypothesis validation 198
– model studies 197 keramaphidin model, selective oxidation of 205 ketimines, hydrogenation 793 – asymmetric biomimetic transfer 793 – enantioselective biomimetic transfer 793 – metal-free asymmetric transfer 793 ketosynthase (KS) 473 kijanimicin 763–764 Knoevenagel condensation 366–367, 483 komaroviquinone 874 kutzneride 371 kuwanons 776
l lachanthocarpone 768 lactonic compounds 905–908 – annonaceous acetogenins 907 – epimerization 905–908 – methanolysis of 908 – Thalictrum saponins cyclization during acidic hydrolysis 906 – transesterification 905–908 ladder polyethers 545–550 – dioxepandehydrothyrsiferol 546 – Giner’s proposal for biosynthesis 549 – Nakanishi’s hypothesis 548 – structures 547 – synthesis 565–583 – – applications 580–583 – – 6-endo cyclization 567 – – fused polyether systems 567–580 – – iterative approaches 565–567 – – Jamison proposal 577 – – McDonald group 570–573 – – Murai’s work 569 – – THP : THF selectivity in 578 – trans-syn-trans arrangement 545 Laggera tomentosa 880 Lahav’s rule of reversal 837 lambertellol 860 lateriflorone biosynthesis 461 Laurencia pinnatifida 901 Laurencia spectabilis 901 Leuzea carthamoides 892 life’s single chirality 823–841, see also biological homochirality light, see also electrocyclizations – in artifacts and natural substances formation 870–878 – – epimerization 870–872 – – isomerization 870–872 – – milnamide A 872 – – reserpine 872
945
946
Index light, see also electrocyclizations (contd.) – photochemical reactions: see Chapter 16 – photocycloaddition 876–878 – photodimerization 876–878 – rearrangements by 872–876 – – brevianamide A 875–876 – – ent-bicyclogermacrene 873 – – komaroviquinone 874 lignans 677–691 – biomimetic synthesis of 681 – – benzo[kl]xanthenes 686–688 – – benzoxanthenone lignans 683–686 – – furofuran lignans 681–683 – – podophyllotoxins 681 – chemotypes of 678–679 Liquidambar formosana 663 lissoclinamides 326–328 – to heptapeptide 327 – entry into secondary metabolism 5–6 – metabolism toward alkaloids 6 Lobelia alkaloids 44 longithorone 779 – longithorone A 427 Lonicera japonica 916 Lonicera korolkoviii 916 Lonicera morrowii 916 – entry into secondary metabolism 5 lovastatin 507, 755–756 – lovastatin nonaketide synthase (LovB) 755–756 lupine alkaloids 31–34 – biomimetic conversion into oleane skeletons 423 – biomimetic synthesis of 32–33 – – oxidative deamination step 32 Lycopodium-like alkaloids 50–51 – Lycopodium alkaloids 44–54 lycoposerramine series 49–50 Lyngbya majuscula 897 lysine-derived alkaloids 3–54 – biomimetic synthesis of 30–42 – L-arginine 3 – L-lysine 3 – L-ornithine 3 lysine-derived reactive units 11–15 – oxidative degradation of free L-lysine 11–12 – Sch¨opf’s pioneering works 13 – tetrahydropyridine 12
m macquarimicins 514–521 – biosynthetic hypothesis for – structure 517
518
– Tadano’s biomimetic synthesis 521 – Tadano’s model study 520 macrocyclic complex alkaloids 183 macrocyclization 340 macrolactamization 338, 347, 360, 373–374 macrophomate synthase (MPS) 754, 756–758 – catalytic mechanism of 756–757 – Michael–aldol route 756 macroxine 174 madangamine alkaloids 208–210 – madangamine C type alkaloids – – biogenesis 209 – – biomimetic synthesis 210 maitotoxin 595 (+)-makomakine 95 Malbranchea aurantiaca 140 malbrancheamides 118, 136 – malbrancheamide B 136 – proposed biosynthesis of 138 – total syntheses of 138 mallotusinin production 662 malondialdehyde scenario 182–191, 200–203 – aminopentadienal connection 202 – halicyclamine connection 201–203 – keramaphidin/ircinal connection 200–201 malonic acid half-thioesters (MAHTs) 474 malonyl activation 475–477 malonyl acyl transferase (MAT) 474 malonyl half thioesters (MAHT) 479 – asymmetric and organocatalytic addition to nitroolefins 482 – Shair’s catalytic aldol condensation with 479 – – asymmetric 480 Mannich bisannulation 385 manzamine A – ABC-ring system synthesis of 204 – AB-ring system synthesis of 204 – biomimetic models toward 203–204 manzamine alkaloids synthesis 181–221, see also Baldwin’s hypothesis development; pyridinium marine sponge alkaloids – acrolein scenario 182–191, see also individual entry – 3-alkylpyridiniums biosynthetic hypotheses based on pachychaline series 187 – aminopentadienals 190, 213–215 – Baldwin’s hypothesis, dihydropyridine chemistry 186 – biomimetic C5 reactive units from Zincke reaction 189–191 – Chichibabin synthesis of pyridines 188
Index – cyclostellettamine A type 184–185 – from cyclostellettamines to keramaphidin and halicyclamine/haliclonamine alkaloids 218 – dimers (and oligomers) 184 – from fatty acids to long-chain aminoaldehydes and sarain alkaloids 215 – from fatty aldehydes precursors to simple 3-alkyl-pyridine alkaloids 182–187 – from ircinal and pro-ircinals to manzamine A alkaloids 218 – glutaconaldehydes 213–215, see also glutacondialdehydes – glutacondialdehydes 190, see also glutaconaldehydes – Husson’s strategy (modified Polonovski reaction) 188 – ircinal pathway, spinal cord of manzamine metabolism 218 – macrocyclic complex alkaloids 183 – madangamine alkaloids 208–210 – malondialdehyde scenario 182–191, see also individual entry – manzamine alkaloid chemistry, milestones in 185 – Marazano biomimetic synthesis of dihydropyridine 188 – Marazano modified hypothesis, pyridinium chemistry 186 – modified hypothesis testing in laboratory 203–208 – – biomimetic models toward halicyclamines 205–208 – – biomimetic models toward manzamine A 203–204 – – (4 + 2) cycloaddition strategy towards an ircinal model 203 – monomers 184 – nakadomarine A, biomimetic model of 210–211 – from pro-ircinals to madangamine alkaloids 218–219 – pyridine ring formation 186 – theonelladine A type 184–185 – total syntheses of 219–220 – towards a universal scenario 215–219 – xestospongins 184–185, 191–193 Marazano biomimetic synthesis of dihydropyridine 188 Marazano’s hypothesis 201 marcfortine C 135 – total synthesis of 137 marcfortines 155–158
marine diterpenes biomimetic synthesis from Pseudopterogorgia elisabethae 410–414 – colombiasin A 412 marine polypropionates 877 marine pyrrole-2-aminoimidazole alkaloids, See pyrrole-2-aminoimidazole (P-2-AI) marine alkaloids marine thiol group 856 Markhamia lutea 857 masked hexaketides, Schmidt’s condensation of 492 massadine chloride 263–265 massadine 257, 263–265 – Baran’s biogenetic hypothesis 264 – formation, intramolecular aziridinium mediated mechanism for 259 – Romo’s biosynthesis proposal for 261 mauritiamine 253–254 meleagrine 174 meloscandonine 174 membrane diffusion, symmetry breaking in 840 meroterpenoids, biomimetic synthesis of 424–425 mersicarpine 174 metal-catalyzed cross coupling, Trp-Tyr biaryl bond formation by 361 methanolysis of lactonic sesquiterpenes 908 methoxymethyl (MOM)-protection 497 6-O-methylforbesione synthesis 458–459 methyl homodaphniphyllate 304 methyl homosecodaphniphyllate 301 methyllateriflorone synthesis 459–460 methyllateriflorone, total synthesis of 462 7-methylcycloocta-1,3,5-triene 618 Mg(II) salts, biomimetic conditions using catalytic 476 milnamide A 872 minfiensine 174 modified indole nucleus alkaloids 149–175, see also camptothecin; discorhabdins – biomimetic synthesis of 149–175 – – monoterpenoid indole alkaloids 150 modified Julia coupling 360 Monascus ruber 506 mongolicumin A 686–687 monocyclic polyprenylated acylphloroglucinols (MPAPs) 434 – from Humulus lupulus 435 monomers 184 monoterpene rearrangements 397–401 – century since Wagner’s structure of camphene 397–399 monoterpenoid indole alkaloids 150
947
948
Index Montmorillonite K-10 (MK-10) Morus bombycis 775 Myrioneuron alkaloids 34–39
165
n nakadomarine A, biomimetic model of 210–211 nakamuric acid, Baran’s synthesis of 255 nanaomycine, Yamaguchi’s biomimetic synthesis 494 naphthalenoid derivatives 492–494 – barakol, Harris’ biomimetic synthesis of 493 – biomimetic access to 492–494 – eleutherinol, Harris’ biomimetic synthesis of 493 – 6-hydroxymusizin, Harris’ biomimetic synthesis 493 – naphthyl cyclization of β-hexaketones 493 – polyketides into, Yamaguchi’s aromatic cyclization 494 nargenicin A 508–509 N-chlorosuccinimide (NCS) mediated oxidative coupling 368 Negishi-cross coupling 340 nemorosone, total synthesis through ‘carbanions’ differentiation 443–445 neocarzinostatin 595 neolignans 677 neopupukeananes 406 neoselaginellic acid 174 neosymbioimine 276–279 nepalensinol B 713 N-heterocycles, asymmetric organocatalytic reduction 800–814 – enantioselective hydride transfer 800 – enantioselective protonation 800 Nicotiana tabacum 10 nicotinamide adenine dinucleotide (NADH) 787 nitraramine, biomimetic synthesis of 35–37 nitraria alkaloids 14, 34–39 nitrophenyl pyrones 618–621 N-methylcytisine conversion into kuraramine 33–34 N-methyltriazolinedione (MTAD) 125 Nocardia argentinensis 508 nonadride series 504–506 – biomimetic studies in 504–506 – CP-225917 505–506 – CP-263114 505–506 – dimerization process towards isoglaucanic acid 504–505 – Sutherland’s biomimetic studies 504
non-amino acid origin alkaloids, biomimetic synthesis 271–307, see also cyclic imine marine alkaloids; Galbulimima alkaloids non-aromatic polycyclic polyketides 503–530, see also nonadride series non-prenylated indole alkaloids 141–146, see also epidithiodioxopiperazines non-ribosomal peptide synthesis (NRPS) 319–320, 346 norrhoedanines in acidic conditions 901 norzoanthamine 290 notoamide J synthesis 121 – Williams’ biomimetic synthesis of 124 nucleophilic 1,2-addition 380
o ocellapyrones 621–624 – ocellapyrone A, electrocyclic formation of 624 o-iodoxybenzoic acid (IBX) 726 – dimerization of 2,6-xylenol 726 – Pettus’ oxidative dearomatization 726 okaramine N 118, 125 oligomeric ellagitannins 658 oligomers 695–718, see also resveratrolbased family of oligomers – synthetic approaches to 695–718 olivacine alkaloids 102–105 omaezakianol synthesis 563 o-quinone dimerization 727 L-ornithine 3 ornithine alkaloids 18–30 – reactive units 9–11 – – 4-aminobutyraldehyde 9 oroidin 237–238 – Al-Mourabit’s synthesis of 239 – Lindel’s conversion into rac-cyclooroidin 240 orsellinic acid 486 ortho-quinone methide capture 385 Osmunda japonica 885 oxasqualenoids 542–544, 558–560 oxidation processes 853–870 – achiral bisacridones from Rutaceae 866 – allicin 855 – bis-aporphines 864 – dioxoaporphines 868 – discorhabdin B 856 – hydroperoxides, artifacts from 858 – indolomonoterpenes 867 – lambertellol 860 – marine thiol group 856 – newly oxygenated products 859–864 – N-oxide and oxoalkaloid cases 865–870
Index – oxidative coupling 865 – oxoaporphines 868–869 – of oxygenated functions 857–859 – pyrroloiminoquinolines 856 – Tabernaemontana spp. 868 – thiol oxidation 853–856 – vasicoline 868 – welwitindolinones from Fischerella spp 862 oxidative cyclization 347, 440 oxidative dearomatization 723–747 – Adler–Becker oxidation, Singh’s application of 727 – Canesi’s 735 – Danishefsky’s 730 – Diels–Alder dimerization 725 – Feldman’s 734 – Gaunt’s 732 – Heathcock’s synthesis of styelsamine B 732 – initial intermediate 723–724 – intermolecular dimerizations 724–727 – intramolecular cascade sequences 731–733 – intramolecular cycloadditions 729–731 – Liao’s 729 – Majetich’s 737 – Morrow’s 728 – Nakatsuka’s 740 – Njardarson’s 730 – Pettus’ 731, 735 – phenol oxidative cascades 741–747 – Porco’s 734 – Quideau’s 739 – rearrangements 733–737 – Rodr´ıguez’s 733 – Rogi´c’s 740 – Sarpong’s 729 – sequences 723–724 – sequential reactions initiated by 723–747 – sequential ring rupture – – and contraction 737–739 – – and expansion 739 – Sigman’s enantioselective 728 – Sorensen’s 731 – Stoltz’ 729 – successive intermolecular reactions 727–729, 741 – successive intramolecular reactions 741 – successive tautomerizations 733–737 – Takeya’s o-quinone dimerization 727 – Tejera’s 740 – Trauner’s 736, 738 – Wood’s 730 – Yamamura’s 728
oxidative diversification 415 oxindole fragment, stereocontrolled oxidation of 361–362 oxindoles synthesis 139 oxoaporphines 868–869 oxysceptrins 254–255 – Baran’s synthesis of 255
p P-2-AIs simple dimers, biomimetic synthesis 253–255 – ageliferins 254–255 – mauritiamine 253–254 – oxysceptrins 254–255 – sceptrins 254–255 Pachychalina species 186 Paeonia lactiflora 913 Paeonia suffruticosa 913 paeoniflorin 917 palau’amine 255–257, 265 – Al-Mourabit’s biogenetic proposal for 260 – axinellamine A 257 – axinellamine B 257 – massadine 257 – synthesis – – first proposal based on Diels–Alder key step 257 – – Kinnel’s biogenetic proposal for 258 – – Scheuer biogenetic proposal for 258 paliurine F synthesis 339 pallavicinolide A 417, 420 pallidol 702, 713 paraherquamide A, biosynthesis of 130 paraherquamides 155–158 Paraphaeosphaeria quadriseptata 901 paucifloral F 711 pedunculagin 658 pelletierine – based metabolism 42–54 – biomimetic synthesis 43–44 Penicillium brevicompactum 117 Penicillium glaucum 504 Penicillium islandicum 486 Penicillium purpurogenum 504 penifulvins 405–406 pentacarbonyl derivative, Harris’ biomimetic cyclizations of 492 pentacycle formation 385–389, see also under Ecteinascidin 743 (ET 743) pentacyclization 305–306 peptide alkaloids 317–318, see also aryl-containing peptide alkaloids; azole-containing peptide alkaloids – aryl-alkyl ether peptide alkaloids 337
949
950
Index peptide alkaloids (contd.) – – ring-closing strategies in 338 – – ring formations in biosynthesis of 337 – biosynthesis, key features 319–321 – covalent folding of peptide chains into 321 – cyclic peptides containing biaryl ethers 339–343 – cyclized by aryl side chains oxidation 336–350 peptide fragment coupling 347 perovskone 420, 770 phalarine 174 phalloidin 391 phenol oxidation 768–775 phenol oxidative cascades 741–747 – additional natural compounds arising 746 – Hertweck’s 743 – Pettus’ 742, 745 – Porco’s 746 – Shen’s 744 – Steglich’s 743 – Zhao’s 743 phenols, oxidative amidation of 71–77 – Honda oxidative cyclization 75 – Knapp iodocyclization 72 – oxidative spirocyclization 72 – stereoselective cyclization of 73 – Wardrop oxidative cyclization 76 phenoxonium species 727 phenyl iodine diacetate (PIDA) 725 phenyl iodine(bis)trifluoroacetate (PIFA) 728 phlegmariurine series 49–50 phloracetophenone 486 phosgene adducts 921 photochemical reactions, see also light, electrocyclizations photocycloaddition 876–878 – hoenalia coumarin 876 – marine polypropionates 877 photodimerization 876–878 Phyllanthus emblica 645 Pictet Spengler cyclization 121, 385 pinnatoxins 279–282 – (−)-pinnatoxin A, Kishi’s biomimetic synthesis 281 – pinnatoxin A, biosynthetic origin of 281 L-pipecolic acid 6 pipecolic acids 15–18 – biomimetic access to 15–18 – biosynthesis 15–16 – – by photocatalysis 18 – containing secondary metabolites 16 – importance 15–16
– Rossen’s biomimetic synthesis of 17 – Yamada’s biomimetic access to 17 piperidines, hydrogenation 813 Plakortis angulospiculatus 509 p-nitrophenyl pyrones 619 podophyllotoxins 681–682 polyamine alkaloids 7–8 – polyamine backbones in 7 polycyclic polyethers, see also ladder polyethers; polyether ionophores; squalene – biosynthesis 539–550 – epoxide-opening cascades in 550–583, see also individual entry – structure 539–550 polycyclic polyprenylated acylphloroglucinols (PPAPs) 433–452 – biomimetic synthesis of 436–441 – biosynthesis of 434–436 – classification of 434 – from MPAPs 437 – non-biomimetic synthesis of 441–451 – – Garsubellin A 441–443 – synthesis via oxidative cyclization reactions 440 – Type A PPAPs 439–440 – – via an intramolecular Michael addition 440 polyene/polyene splicing 607, 878, 881 polyepoxide opening, polyether natural products synthesis via 537–584 – synthetic considerations, Baldwin’s rules 538–539 – – 4-exo-trig reactions 538 – – 5-endo-trig reactions 538 polyether ionophores 539–542 – Cane–Celmer–Westley hypothesis 540 – endo cyclizations 540 – structures of 541 polyketide assembly mimics/polyketide synthases (PKSs) 472–485, see also aromatic polyketides – C–C connection mechanism in 473 – – addition–decarboxylation for 481 – structure 473 – Type-a mimics 475–478 – – acyl transfer, Scott’s conditions for 475 – – catalytic Mg(II)salts, biomimetic conditions using 476 – – imidazole, biomimetic conditions using 476 – – malonyl activation 475–477 – – thioesters, biomimetic conditions using 476 – – without malonyl activation 477–478
Index – Type-b mimics 478–479 – – Coltart’s aldol addition with non-activated thioester 482 – – enolate formation before nucleophilic addition 480 – – Fagnou’s metal-free decarboxylative condensation 480 – – malonyl activation 479–482 – – without malonyl activation 482–483 – Type-c mimics 483–485 – – Barbas III asymmetric and organocatalytic addition of thioesters 483 – – Birch reduction–ozonolysis reaction 485 – – List’s condensation of MAHO in 484 – – reaction mimic with MAHT 484 polyketides (PK) 284–293, 485–499, 503–530, 591–632 – aromatic polyketides 485–499 – biological electrocyclizations 628–631 – biomimetic analysis 597–598 – biosynthetic origin proposed by Morita 285 – Black’s electrocyclization cascade hypothesis 617 – cassiarins A and B 284–285 – decalin systems 506-509 – electrocyclic reactions 592 – 6π electrocyclizations, 598–612, see also individual entry – electrocyclization reactions toward 591–632 – elysiapyrones 624–625 – endiandric acids A–G 615 – enzyme catalysis 628–631 – epoxyquinols A–C 615 – epoxytwinol 615 – fatty acid biosynthesis 594–597 – general biosynthesis 596 – nitrophenyl pyrones 618–621 – nonadrides 504-506 – non aromatic polyketides 503–530 – p-nitrophenyl pyrones 619 – shimalactones 625–628 – structure 285 – 8π systems and black 8π –6π electrocyclic cascade 612–628 – torreyanic acid 614 polyolefin cyclization 421 polyprenylated phloroglucinols 433–464, see also polycyclic polyprenylated acylphloroglucinols (PPAPs) polyprenylated xanthones 452–464 – biosynthesis of 454–455 – – CGX motif via cascade of nucleophilic attacks 455
– – CGX motif via Claisen/Diels–Alder reaction cascade 455 – caged xanthones 452 – Diels–Alder cycloaddition 461 – Wessely/Diels–Alder strategy 461 Pomerantz-Fritsch reaction 385 Popowia pisocarpa 865 Populus deltoides 897 Porco synthesis of clusianone 438 potassium hexamethyldisilazide (KHMDS) 525 prenyl side chain dehydrogenation 775–779 prenyl-9-borabicyclo[3.3.1]nonane (prenyl-9-BBN) 119 prenylated indole alkaloids 117–141 – notoamide J synthesis 121 presilphiperfolanol 404 pretetramide, Harris’ biomimetic synthesis of 496 pro-ircinal alkaloids 218–219 – to madangamine alkaloids 218–219 – to nakadomarine alkaloids 219 L-proline 6 see also pyrrole-2aminoimidazoles proline, dioxopiperazines derived from 119–122 protoberberines 921–925 – acetone adducts of 928 – alkaline treatment of 924 – – nucleophilic additions on 925 – dihydroprotoberberines 927 – indolomonoterpenic alkaloid macrosalhine, quaternization of 923 proto-daphniphylline 301 – pentacyclization of 305 przewalskin A 743 Pseudomonas fluorescens 608 pseudopelletierine 43–44 Pseudopterogorgia bipinnata 414 Pseudopterogorgia elisabethae, marine diterpenes biomimetic synthesis from 410–414 Pseudopterogorgia kallos 416 pseudorubrenoic acid A 611 pteriatoxins 279–282 Pueraria mirifica 863 (−)-pumiliotoxin C, Amat’s biomimetic synthesis of 287 Pummerer oxidative cyclization 240 Punica granatum 919 purifications, acidic conditions during 895–903 – epimerization 895–897 putrescine N-methyltransferase (PMT) 5
951
952
Index puupehenone, methanol adduct on 914 pyranose-type ellagitannins 665–667 pyridine alkaloids 215–217 – Chichibabin synthesis of 188 – hydrogenation 813–814 pyridinium chemistry, Marazano modified hypothesis 186 pyridinium marine sponge alkaloids, biomimetic synthesis 191–195, see also xestospongins – 3-alkylpyridiniums 191 – alkylpyridines with unusual linking patterns 194–195 – cyclostellettamine B 191 – upenamides, synthetic approaches to 193 – Zincke-type pyridine ring-opening 193–194 pyridinium salts 181–182, see also manzamine alkaloids synthesis pyrinadine A, biomimetic synthesis 195 pyrinodemin A, biomimetic synthesis of 194 pyrones – Harris’ biomimetic access to 489 – as masked tetraketide 490 pyrrole-2-aminoimidazole (P-2-AI) marine alkaloids 225–267, see also clathrodins – Al-Mourabit’s retro-biogenetic proposal for 232 – biomimetic synthesis of 225–267 – George B¨uchi’s work 233–234 – new challenging P-2-AI synthetic targets and perspectives 266–267 – P-2-AI biosynthesis, common chemical pathway for 256–257 – P-2-AI linear monomers, biomimetic synthesis 237–238 – P-2-AI polycyclic monomers, biomimetic synthesis 234–253, see also cyclized monomers – P-2-AIs simple dimers, biomimetic synthesis 253–255 – synthetic achievements 261–265 – tautomerism in building blocks of 229 pyrrolizidine ring, biomimetic access to 18–19 pyrroloiminoquinolines 856 pyrroloindole-based peptide alkaloids 369 – dimeric 370
q quadrangularin A 701, 710 Quercus robur 669 quinolines, asymmetric organocatalytic reduction 800–805
– asymmetric biomimetic transfer 798 – Brønsted acid catalyzed transfer 801 – organocatalytic asymmetric transfer 798 – 2,3-substituted quinolines 803 – 3-substituted quinolines 804 – 4-substituted quinolines 804 quinolinic acid 194 quinoxalines, hydrogenation 806 quinoxalinones, hydrogenation 806
r rameswaralide 417–419 raucaffrinoline via Cannizzaro reaction 896 reaction–diffusion models, symmetry breaking in 840 red tides 545 reductive aminal formation 380–381 reserpine 872 resveratrol-based family of oligomers 695–718, see also oligomers – biosynthetic approaches 697–705 – davidiol A from 704 – indane-containing members of 711 – palladium-based reactions 706 – quadrangularin A 701 – stepwise synthetic approaches 705–717 – – work toward single targets within 705–709 – synthetic approaches to 695–718 – universal, controlled synthesis approach 709–717 – ε-viniferin from 698–700 rhazinilam 93 Rhodomela confervoides 897 ribosomal peptide synthesis (RPS) 319–320 ritterazines 294–298 – structures 295 Robinson-Gabriel cyclodehydration 325 Rubiaceae iridoids 910 rubifolide conversion into coralloidolides A, B, C, and E 416 rufescidride 686–687 Ru-mediated SN Ar-cyclization 340
s Saccharopolyspora spinosa 521 saframycin A biosynthesis, gene cluster-based proposal for 383 Salvia leucantha 427 Salvia prionitis 769 salvileucalins A and B 428 Sammes’ model study of cycloaddition 127 sanguiin H5, synthesis of 645 sanjoinine G1 synthesis 339
Index sarains – biomimetic models of, side branch of manzamine tree 211–213 – biomimetic synthesis 212 – – first sarain A model 213 – – second sarain A model 213 – sarain A-type alkaloids – – biogenesis 212 sceptrins 254–255 secologanin 150 – derived indolomonoterpene alkaloids 95–109 – derived quinoline alkaloids 109–110 Securidaca longepedunculata 921 Securiflustra securifrons 162 Sedum alkaloids 44 self-replication 833–834 senepodine G 51–52 serratezomine A 47 serratinine 47 – into lycoposerramine B 47–49 – into serratezomine A 48 sesquiterpene rearrangements 401–408 – caryophyllenes in 401–402 – miscellaneous sesquiterpene rearrangements 406–408 shimalactones 625–628 shoreaphenol 708 silphinane series, oxidative rearrangements in 405–406 silphinyl mesylate 405 Silybum marianum 865 silydianin 754 siomycin A 331 SNF4435 C and D 618–621 – Baldwin’s approach 620, 623 – Parker’s approach 622 – Trauner’s approach 620 Soai reaction 830–831 sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) 759 solanapyrone synthase (SPS) 758–760 – endo/exo-selectivities 759 solanapyrones 507–508 solasodine 294 Sophora flavescens 33 sophoradiol, biomimetic synthesis of 421 sorbicillin 726 spinosyns 766–767 – biomimetic TADA reactions toward 521–524 – biosynthesis of 522 – Roush’s total synthesis of 523
spiro systems, Diels–Alder reactions 512–514 – abyssomicin C 512–513 – gymnodimine 514 spirolactam formation 73 spirotryprostatin A, Danishefsky’s synthesis 123 spirotryprostatin B 118 – Danishefsky’s synthesis 122 (±)-sporidesmin A, total synthesis of 145 spontaneous phenol-aldehyde cyclization 385 squalene, polyethers derived from 542–545, 558–565 ‘stabilized’ iodoxybenzoic acid (SIBX) 739, 769 Strecker reaction 385 Stemona spp. 863 Stenus comma 39–42 stenusine 39–42 – natural versus biomimetic 41 – putative biosynthetic pathway 40 – stereochemical particulars 40 – structure 40 stephacidins – biosynthesis proposal for 160 – stephacidin A 134–136 – – conversion to stephacidin B 136 – – improved biomimetic synthesis of 135 – stephacidin B 136 – (+)-Stephacidin A 158–160 – – biosynthesis through notoamide S 136 – – total synthesis by Baran et al. 161 – (−)-Stephacidin B 158–160 – – synthesis by Baran et al. 162 – – biosynthesis through notoamide S 136 stilbene synthase 697 stilbenoids 871 strellidimine 113 Streptomyces antibioticus 764 Streptomyces coelicolor 486 Streptomyces fradie 496 Streptomyces longisporoflavus 524 Streptomyces orinoci 878 strictosidine alkaloids 95–99 strictosidine 92, 150 strychnine 103 strychnochromine 174 styelsamine B 732 – Heathcock’s synthesis of 732 stylissadine A formation – aziridinium mechanism for 259–261 – from massadine, Baran and K¨ock’s proposal 261
953
954
Index stylissazole C 266 styrylpyrone photodimers 879 supercritical CO2 treatment 885–888 – chalcones obtained from 889 superstolide A 509 – Roush’s total synthesis of 510 Suzuki-Miyaura coupling 345, 360, 564 symbioimine 276–279 – biomimetic synthesis of 277 – – Snider’s approach 278 – – Thomson’s approach 278 – biosynthetic origin of 278 – Chruma’s contribution to 279
t Tabernaemontana spp. 868 TAN-1251 compounds 61–86 – Ciufolini synthesis of 80–86 – Honda synthesis of 79–83 – – aldol cyclization 85 – Snider synthesis of 68–71 – – solvent effects in 69 – synthesis via oxidative amidation chemistry 77–86 – total syntheses of 63–71 – Wardrop approach to 77 tangutorine 37–39 tannins 639–672, see also ellagitannins – condensed tannins 639 – hydrolyzable tannins 639 tautomerism in building blocks of P-2-AI monomer clathrodin 229 Teichaxinella morchella 231 tellimagrandin I 643–644 terengganesine B 174 terpene precursors alkaloids 293–305, see also Daphniphyllum alkaloids – barbaline 294 – cephalostatins 294–298 – daphniglaucine A 294 – gardenamide 293–294 – ritterazines 294–298 – solasodine 294 terrecyclene 405 tetracyclic derivatives 495–499 – anthracenoids, Yamaguchi’s access to 497 – benzo[a]tetracenoid derivatives 498–499 – biomimetic access to 495–499 – – tetracenoid derivatives 495–496 – pretetramide, Harris’ biomimetic synthesis of 496 – tetrangomycin, Krohn’s synthesis of 498 – tetraphenoid derivatives 496–498 – (−)-urdamicynone 497
tetrahydroanabasine chemistry 12–13 tetrahydrofuran ring 548 tetrahydroindane systems 509–512 – Diels–Alder reactions affording 509–512 – galiellalactones 511–512 – spiculoic acid A 509 – superstolide A, Roush’s total synthesis of 510 tetrahydropyran ring 548 tetrahydropyridine 12 tetrangomycin, Krohn’s synthesis of 498 tetrapetalone C 746 2,2,6,6-tetramethylpiperidine (TMP) 415 tetrocarcin A 763–764 tetronasin – biosynthetic origin of 524 – Ley’s formal synthesis 526 – synthesis 524–525 – Yoshii’s total synthesis of 527 thallium trinitrate (TTN) mediated cyclization 341 Thapsia garganica 424 theonelladine alkaloids 215–217 theozymes 427 thiangazole 324–326 – to pentapeptide, hypothetical biomimetic simplification of 325 – strategic disconnections in total syntheses 325 thioesters – biomimetic conditions using 476 – condensation between 476 – malonyl thioesters, self-condensation of 477 thiol compounds reaction 662–663 thiol oxidation 853–856 thiostrepton 328–334 TMC-95A-D 357–363 – (Z)-enamide side-chain 360 – late-stage stereoselective (Z)-enamide formation 362–363 – 3-methyl-2-oxopentanoic side-chain T origin 360 – retrobiosynthetic analysis of 359 – synthetic approaches to 360 topaquinone 32–33 Torreya grandis 768 torreyanic acid 610–612, 614 Townsend–McDonald hypothesis 550 trachyopsane A, biomimetic synthesis of 406 transamination 385 transannular Diels–Alder (TADA) reaction 508 transannular hydride transfers 199–200
Index transesterification artifacts 909 transesterification, lactonic compounds 905–908 transtaganolides 425 tridachiahydropyrones 599–603 – biomimetic analysis of 600 – biomimetic synthesis of 602 tridachione family 603–608 – 9,10-deoxytridachione 604–606 – oxytridachiahydropyrone 603 – polyene 607 – pseudorubrenoic acid A 608–610 trienes, cyclization of 507 trimethylsilyl trifluoromethanesulfonate (TMSOTf) 564 triquinane series, biomimetic studies in 404–405 tris(dimethylamino)sulfonium difluorotrimethylsilicate (TASF) 735 triterpene rearrangements 420–424 tropinone chemistry 29–30 Trp-Tyr biaryl bond formation by metal-catalyzed cross coupling 361 tryptamine 150 tryptophan alkaloids 91–113, see also indolemonoterpene alkaloids – biomimetic synthesis of 117–147, see also bicyclo[2.2.2]diazaoctanes; non-prenylated indole alkaloids; prenylated indole alkaloids – – dioxopiperazines derived from 119–122 – – tryprostatin B, biomimetic total synthesis of 121 TTN-oxidative coupling 340 tyrosine alkaloids 61–86
u Ugi four component reaction 385 Ullmann-coupling 340–341, 646 upenamides, synthetic approaches to 193 (−)-urdamicynone, Yamaguchi’s synthesis 497 usambarine 93
vellomisine 894 Veratrum californicum 421 (+)-versicolamide B 118 versicolamides, asymmetric synthesis 140 vescalagin 666, 670 – pyrolytic degradations of 671 vincadifformine 109 vincamine 109 vincorine 174 vincoside alkaloids 95–99 ε-viniferin – biogenetic explorations using 703 – davidiol A from 704 – from resveratrol 698–700 VM55599 129, 155–158 – biosynthesis of 130 – (−)-VM55599, asymmetric total synthesis of 132 – Williams’ biomimetic total synthesis of 130 Vorbr¨uggen condensation 72
w Wagner–Meerwein rearrangement 398–400, 410 Wardrop oxidative cyclization 76 welwitindolinones – from Fischerella spp, oxidation 862 – welwitindolinone A – – biosynthesis proposal for 165–166 – – synthesis by Baran et al. 165 – – (+)-Welwitindolinone A 164–166 Wieland–G¨umlich aldehyde 101 Williams’ biomimetic synthesis – of notoamide J 124 – of VM55599 130 (+)-WIN 64821 synthesis 143 Winterfeldt-Witkop cyclization 153 Witkop-type photo-induced macrocyclization 382 Woodward–Hoffmann rules 616 W¨urthner’s polymerization model 835
x v vancomycin 345–350 – biaryl-ether formation during biosynthesis of 347 – Evans’ synthesis of 349 – Nicolaou’s synthesis of 348 – structure of 346 vasicoline 868
xanthepinone, methanol induced rearrangement of 912 xanthones 433–464, see also polyprenylated xanthones xestospongins 191–193, 215–217 – biomimetic synthesis by the Baldwin group 193 – xestospongins A 192
955
956
Index
y yohimbine 95–99 yunnaneic acid H 686–687 Yuzuriha 298
z zamamidine C, retrobiosynthesis 220 zearalenone, Barrett’s synthesis of 491
Zincke reaction 189–191 – pyridine ring-opening 193–194 Zizyphus jujuba 882 zoanthamine alkaloids 288–291 – biosynthetic origin proposed by Uemura 289 – cyclization substrate synthesis 290 – Kobayashi’s biomimetic approach to 291 – proposed biosynthetic route for 290
