Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts

Metathesis in Natural Product Synthesis Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer Related Ti...

Author: Janine Cossy | Stellios Arseniyadis | Christophe Meyer

102 downloads 2176 Views 8MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Metathesis in Natural Product Synthesis Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer

Related Titles Christmann, M., Bräse, S. (eds.)

Dalko, P. I. (ed.)

Asymmetric Synthesis – The Essentials

Enantioselective Organocatalysis

395 pages with 85 figures and 3 tables

Reactions and Experimental Procedures

2008

559 pages with 364 figures and 115 tables

Softcover

2007

ISBN: 978-3-527-32093-6

Hardcover ISBN: 978-3-527-31522-2

Hudlicky, T., Reed, J. W.

The Way of Synthesis

Cornils, B., Herrmann, W. A., Muhler, M., Wong, C.-H. (eds.)

Evolution of Design and Methods for Natural Products

Catalysis from A to Z

1018 pages with 285 figures and 15 tables 2007 Softcover ISBN: 978-3-527-31444-7

A Concise Encyclopedia approx. 1560 pages in 3 volumes with 2778 figures and 104 tables 2007 Hardcover ISBN: 978-3-527-31438-6

Sheldon, R. A., Arends, I., Hanefeld, U.

Green Chemistry and Catalysis 448 pages with 476 figures and 21 tables 2007 Hardcover ISBN: 978-3-527-30715-9

de Vries, J. G., Elsevier, C. J. (eds.)

The Handbook of Homogeneous Hydrogenation 1641 pages in 3 volumes with 422 figures and 254 tables

Enders, D., Jaeger, K.-E. (eds.)

Asymmetric Synthesis with Chemical and Biological Methods 470 pages with 84 figures and 57 tables 2007 Hardcover ISBN: 978-3-527-31473-7

2007 Hardcover ISBN: 978-3-527-31161-3

Metathesis in Natural Product Synthesis Strategies, Substrates and Catalysts

Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer With a Foreword by Robert H. Grubbs

The Editors Prof. Janine Cossy Laboratorie de Chimie Organique, ESPCI 10 Rue Vauquelin 75231 Paris Cedex 05 France Dr. Stellios Arseniyadis Laboratoire de Chimie Organique, ESPCI 10 Rue Vauquelin 75231 Paris Cedex 05 France Dr. Christophe Meyer Laboratoire de Chimie Organique, ESPCI 10 Rue Vauquelin 75231 Paris Cedex 05 France

Cover The gryffon painting being part of the front cover picture has been kindly provided by Dominique Escortell

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 http://dnb.d-nb.de.  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

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. Cover Adam Design, Weinheim Typesetting Laserwords Private Limited, Chennai, India Printing and Binding Strauss GmbH, Mo¨ rlenbach Printed in the Federal Republic of Germany Printed on acid-free paper ISBN: 978-3-527-32440-8

VII

Contents

Foreword V Preface XV List of Catalysts XIX List of Contributors XXI Abbreviations XXV 1

1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8

2

2.1 2.2 2.2.1 2.2.2 2.2.2.1 2.2.2.2 2.2.2.3 2.2.3 2.2.3.1

Synthesis of Natural Products Containing Medium-size Carbocycles by Ring-closing Alkene Metathesis 1 Nicolas Blanchard and Jacques Eustache Introduction 1 Formation of Five-membered Carbocycles by RCM 1 Formation of Six-membered Carbocycles by RCM 11 Formation of Seven-membered Carbocycles by RCM 22 Formation of Eight-membered Carbocycles by RCM 30 Formation of Nine-membered Carbocycles by RCM 33 Formation of 10-membered Carbocycles by RCM 34 Conclusion 39 References 40 Natural Products Containing Medium-sized Nitrogen Heterocycles Synthesized by Ring-closing Alkene Metathesis 45 Sebastiaan (Bas) A. M. W. van den Broek, Silvie A. Meeuwissen, Floris L. van Delft, and Floris P. J. T. Rutjes Introduction 45 Five-membered Nitrogen Heterocycles 47 Dihydropyrroles 47 Pyrrolidine Alkaloids 47 Pyrrolidines 47 Dipyrrolidines 49 Polyhydroxypyrrolidines 49 Indolizidine Alkaloids 52 Polycyclic Indolizidines 52

Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts. Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32440-8

VIII

Contents

2.2.3.2 2.2.4 2.3 2.3.1 2.3.1.1 2.3.1.2 2.3.1.3 2.3.1.4 2.3.2 2.3.3 2.4 2.5 2.6

Polyhydroxyindolizidines 55 Pyrrolizidine Alkaloids 59 Six-membered Nitrogen Heterocycles 61 Piperidine Alkaloids 61 Piperidines 61 Piperidine Carboxylic Acids 66 Piperidones 68 Polyhydroxypiperidines 69 Indolizidine Alkaloids 70 Quinolizidine Alkaloids 73 Seven-membered Nitrogen Heterocycles 78 Eight-membered Nitrogen Heterocycles 81 Conclusion 82 References 83

3

Synthesis of Natural Products Containing Medium-size Oxygen Heterocycles by Ring-closing Alkene Metathesis 87 Jon D. Rainier Introduction 87 General RCM Approaches to Medium Rings 89 Laurencin 95 Eunicellins/Eleutherobin 102 Helianane 104 Octalactin A 105 Microcarpalide and the Herbarums 106 Marine Ladder Toxins 109 Ciguatoxin 109 Brevetoxin 117 Gambierol, Gambieric Acid, Olefinic-ester Cyclizations 120 Conclusion 124 Acknowledgments 124 References 124

3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.8.1 3.8.2 3.8.3 3.9

4

4.1 4.2 4.3 4.4 4.5 4.6

Phosphorus and Sulfur Heterocycles via Ring-closing Metathesis: Application in Natural Product Synthesis 129 Christopher D. Thomas and Paul R. Hanson Introduction 129 Synthesis and Reactivity of Sultones Derived from RCM 129 Total Synthesis of the Originally Proposed Structure of (±)-Mycothiazole 132 Synthesis and Reactivity of Phosphates from RCM 134 Applications of Phosphate Tethers in the Synthesis of Dolabelide C 140 Conclusion 144

Contents

Acknowledgment 144 References 144 5

5.1 5.2 5.3 5.3.1 5.3.2 5.3.3 5.3.4 5.3.5 5.3.6 5.3.7 5.3.8 5.4 5.4.1 5.4.2 5.5 5.5.1 5.5.2 5.5.3 5.6 5.7

6

6.1 6.2 6.3 6.4

7 7.1 7.2 7.2.1 7.3

Synthesis of Natural Products Containing Macrocycles by Alkene Ring-closing Metathesis 149 Ana Gradillas and Javier Pérez-Castells Introduction 149 Organization of the Chapter 151 Macrocyclic Polyketides 152 Resorcinylic Macrolides 152 Salicylate Macrolides 155 Other Antibiotic Macrolides 158 Macrocyclic Musk 162 Epothilones 163 Amphidinolides 165 Other Polyketides 167 Natural Cyclophanes 168 Terpenoids 169 Diterpenoids 169 Macrocyclic Lipids 171 Macrocycles of Amino Acid Origin 172 Macrolactams 172 Cyclodepsipeptides 173 Alkaloids 174 Macrocyclic Glycolipids 175 Conclusions and Outlook 177 References 178 Synthesis of Natural Products and Related Compounds Using Ene–Yne Metathesis 183 Miwako Mori Introduction 183 Synthesis of Natural Products and Related Compounds Using Ene–yne Metathesis 185 Synthesis of Natural Products and Related Compounds Using Ene–yne Cross-metathesis (CM) 195 Synthesis of Natural Products Using Skeletal Reorganization 197 References 202 Ring-closing Alkyne Metathesis in Natural Product Synthesis 205 Paul W. Davies Introduction 205 Alkyne Metathesis 205 Background to Alkyne Metathesis 206 Ring-closing Alkyne Metathesis 207

IX

X

Contents

7.3.1 7.4 7.4.1 7.4.1.1 7.4.1.2 7.4.1.3 7.4.1.4 7.4.1.5 7.4.1.6 7.4.1.7 7.4.2 7.4.2.1 7.4.2.2 7.5

RCAM as a Synthetic Strategy 210 Applications of RCAM in Natural Product Synthesis RCAM/Hydrogenation Strategies 211 Macrocyclic Musks 211 Prostaglandin Lactones 212 Sophorolipid Lactone 213 Epothilone A 213 Cruentaren A 215 Latrunculins A, B, C, M, and S 216 Myxovirescin A1 218 RCAM and Alternative Alkyne Manipulations 218 Citreofuran 218 Amphidinolide V 221 Conclusions 221 References 221

8

Temporary Silicon–Tethered Ring–Closing Metathesis Reactions in Natural Product Synthesis 225 P. Andrew Evans Introduction 225 Temporary Silicon–Tethered Ring–Closing Metathesis Reactions 226 O–SiR2 –O Tethered Substrates: Symmetrical Silaketals 226 C2 -Symmetrical Silaketals and Applications 227 Achiral and Racemic Silaketals 229 Related Applications and Developments 229 O–SiR2 –O Tethered Substrates: Unsymmetrical Silaketals 230 Spiroketals 230 Long-range Asymmetric Induction 233 Annonaceous Acetogenins 237 Trisubstituted Alkenes 239 Related Applications and Developments 240 Dienyne TST-RCM: Symmetrical and Unsymmetrical Silanes 240 Macrolide Antibiotics 242 O–SiR2 –C Tethered Substrates: Allyl and Vinylsiloxanes 244 Lignans from Allylsiloxanes 245 Z-Trisubstituted Alkenes from Allylsiloxanes 248 Di- and Trisubstituted Alkenes from Vinylsiloxanes 250 Related Applications and Developments 255 Enyne TST-RCM: Tri- and Tetrasubstituted Acyclic Dienes 255 Illudins 255 Conclusions and Outlook 256 Acknowledgments 257 References 257

8.1 8.2 8.2.1 8.2.1.1 8.2.1.2 8.2.1.3 8.2.2 8.2.2.1 8.2.2.2 8.2.2.3 8.2.2.4 8.2.2.5 8.2.3 8.2.3.1 8.2.4 8.2.4.1 8.2.4.2 8.2.4.3 8.2.4.4 8.2.5 8.2.5.1 8.3

211

Contents

9

9.1 9.1.1 9.1.2 9.2 9.3 9.4 9.5 9.5.1 9.5.2 9.6

10 10.1 10.2 10.2.1 10.2.1.1 10.2.1.2 10.2.1.3 10.2.1.4 10.2.1.5 10.2.1.6 10.2.1.7 10.2.2 10.2.2.1 10.2.2.2 10.2.3 10.2.3.1 10.2.3.2 10.2.3.3 10.2.3.4 10.3 10.3.1 10.3.1.1 10.3.1.2 10.3.1.3 10.3.2

Metathesis Involving a Relay and Applications in Natural Product Synthesis 261 Thomas R. Hoye and Junha Jeon Introduction 261 The Relay Concept 261 Basic Tenets of RCM 262 Early Relay Metathesis Discoveries 263 Examples of Relay Metathesis Directed at Targets Other than Natural Products 268 Examples of Relay Metathesis Motivated by Natural Product Synthesis 269 Examples of Relay Metatheses Thwarted in Achieving the Desired Outcome 278 Interference from a Truncation Event 278 Interference from Premature Macrocyclization 279 Conclusion 281 Acknowledgments 283 References 283 Cross-metathesis in Natural Products Synthesis 287 Joëlle Prunet and Laurence Grimaud Introduction 287 Functionalization of Olefins 287 Cross-metathesis with Acrylate Derivatives 287 Acrylonitrile 287 Thioacrylates 288 Acrylic Acid 288 Acrylimides 289 Acrylates 290 Acrolein 291 Vinyl Ketones 292 Cross-metathesis with Vinyl Derivatives 292 Vinyl Boronates 292 Vinyl Silanes 293 Cross-metathesis with Allylic Derivatives 293 Allyl Silanes 294 Allyl Phosphonates 294 Allylic Alcohol Derivatives 294 Miscellaneous 295 Appending a Side Chain 296 With No Functional Group 297 A Simple Case 297 The Specific Case of Isopropylidene 297 Removing Part of a Side Chain 298 With Functional Groups 299

XI

XII

Contents

10.4 10.5 10.5.1 10.5.2 10.5.3 10.5.4 10.5.5 10.6 10.7 10.8

Couplings 300 Cascade Processes Involving CM 303 ROM/CM 303 ROM/CM/RCM 305 ROM/RCM/CM 305 CM/RCM 306 RCEYM/CM 307 Ene–yne CM 308 Alkyne CM 309 Conclusion and Perspectives 310 Acknowledgments 310 References 310

11

Cascade Metathesis in Natural Product Synthesis 313 Marta Porta and Siegfried Blechert Introduction 313 RCM–CM Sequences 314 Ene–ene RCM–CM 314 Synthesis of (3R,9R,10R)-Panaxytriol 314 Ene–yne–ene RCM–CM 315 Synthesis of (+)-8-epi-Xanthatin 316 Ene–yne–ene RCM–RCM 316 Synthesis of Bicyclic Structures 317 Synthesis of (−)-Securinine and (+)-Viroallosecurinine 317 Total Synthesis of ent-Lepadin F and G 318 Synthesis of Tricyclic Compounds 319 Synthesis of (±)-Guanacastepene A 319 Approach to Taxane Analogs 320 Synthesis of Natural Products Containing Tetracycles 321 Synthesis of Erythrina Alkaloids 321 ROM–CM Sequences 322 Synthesis of Bistramide A 323 RCM–ROM Sequences – Ring-rearrangement Metathesis (RRM) 325 RRM of Monocyclic Substrates 326 Synthesis of Tetraponerines 326 Synthesis of (−)-Swainsonine and (+)-Castanospermine 326 Synthesis of (+)-trans-195A 327 Synthesis of (−)-Centrolobine – Diastereoselective RRM (d-RRM) 328 RRM of Bicyclic Substrates 329 Synthesis of Indolizidine 251F, (±)-trans-Lumausyne and Aburatubolactam A 330 Synthesis of (+)-ent-Lepadin B 331

11.1 11.2 11.2.1 11.2.1.1 11.2.2 11.2.2.1 11.3 11.3.1 11.3.1.1 11.3.1.2 11.3.2 11.3.2.1 11.3.2.2 11.3.3 11.3.3.1 11.4 11.4.1 11.5 11.5.1 11.5.1.1 11.5.1.2 11.5.1.3 11.5.1.4 11.5.2 11.5.2.1 11.5.2.2

Contents

11.6 11.6.1 11.6.1.1 11.6.1.2 11.6.2 11.6.2.1 11.6.2.2 11.6.2.3 11.7

12

12.1 12.2

12.2.1 12.2.2 12.2.3 12.3 12.4 12.5

13

13.1 13.2 13.3 13.4 13.5 13.6 13.7

RCM–ROM Sequences Combined with Other Metathesis Reactions 332 RCM–ROM–RCM 332 RCM–ROM–RCM Cascades of Monocyclic Structures 333 RCM–ROM–RCM Cascades of Bicyclic Structures 336 RCM–ROM–CM 337 Synthesis of (−)-Lasubine II 337 Synthesis of (+)-Cylindramide A and Bicyclic Core of Geodin A 338 Total Synthesis of (+)-Mycoepoxydiene 338 Conclusions and Outlook 339 References 340 Catalytic Enantioselective Olefin Metathesis and Natural Product Synthesis 343 Amir H. Hoveyda, Steven J. Malcolmson, Simon J. Meek, and Adil R. Zhugralin Introduction 343 Total Synthesis of Natural Products with Enantiomerically Pure Chiral Olefin Metathesis Catalysts Bearing a C2 -symmetric Diolate Ligand 343 Total Synthesis of Coniine through Enantioselective RCM with Substrates Bearing a Tertiary Amine 343 Enantioselective Synthesis of Africanol by a Ring-opening/Ring-closing Metathesis Reaction 344 Enantioselective Synthesis of the Lactone Fragment of Anti-HIV Agent Tipranivir 345 Enantioselective Synthesis of Quebrachamine through an Exceptionally Challenging RCM Reaction 345 Synthesis of Baconipyrone C by Ru-catalyzed Enantioselective ROCM 347 Conclusions and Future Outlook 347 Acknowledgments 347 References 347 Metathesis Reactions in Solid-phase Organic Synthesis 349 Sofia Barluenga, Pierre-Yves Dakas, Rajamalleswaramma Jogireddy, Gaële Valot, and Nicolas Winssinger Introduction 349 Metathesis-based Cyclorelease Reaction 350 Ring-closing Metathesis (RCM) 354 Intraresin Dimerization 358 Restricting Peptide Conformation through Cyclization 359 Cross-metathesis on Solid Phase 363 Ene–yne Metathesis on Solid Phase 367

XIII

XIV

Contents

13.8

Conclusion 369 Acknowledgments References 370 Index 373

370

V

Foreword In the last few decades, metathesis has been among the key reactions that have revolutionized the synthesis of complex molecules. Many organic chemists in academic and industrial laboratories, in the field of natural products, have used this reaction as a very practical, versatile, and selective synthetic tool. Olefin metathesis has helped to elevate the art and science of chemical synthesis to its present high level. The examples in this book will demonstrate that organic chemists, with the metathesis reaction in hand, have a new way to consider the connections that are required for efficient access to natural products. This book assembles the most important and interesting examples in the synthesis of natural products using metathesis. Owing to the possibilities opened by olefin and acetylenic metathesis, a great variety of carbocyclic – nitrogen-, oxygen-, sulfur-containing heterocycles – natural products with small-, medium-, and macrocyclic size can be obtained rapidly. The synthetic transformations that couple metathesis steps in cascade reactions are particularly elegant. Emphasis has been put on the metathesis step showing the importance of the catalysts that are tolerant of a large variety of functional groups, very regio-, stereoselective, and even enantioselective. The power of the catalysts and of the metathesis reaction can be appreciated when alternative pathways are considered. Every reaction and catalyst can always be improved. In the area of metathesis, the development of more active and robust catalysts, catalysts that can control the E and Z stereoselectivity of the formed olefins, particularly the stereoselectivity of trisubstituted olefins, or catalysts that can control the enantioselectivity remains a challenge. As has been demonstrated in the past, improvements of the catalyts give rise to increasingly exciting applications in the field of complex molecules and particularly in the field of natural products synthesis. This book will be a good source of inspiration for those planning future developments of metathesis reactions in the field of natural and non-natural products. Robert H. Grubbs


XV

Preface In the 1960s, the ring-opening polymerization of cycloalkenes and the disproportionation of linear alkenes, both used by the polymer and the petroleum industry, were the first reported examples of ‘‘olefin metathesis reactions.’’ Whereas those transformations were generally carried out with ill-defined catalysts, the mechanism of olefin metathesis proposed by Chauvin and Hérisson in 1971 identified metal carbenes as catalytically active species with reactions proceeding through metallacyclobutane intermediates. The mid-1970s saw the emergence of the first well-defined alkylidene–metal complexes for olefin metathesis initially based on tantalum and tungsten. However, in the late 1980s, the quest for higher functional group tolerance resulted in the development of the molybdenum complex, also known as Schrock’s catalyst, which was later used by Grubbs and Fu in ring-closing metathesis (RCM) to access oxygen and nitrogen heterocycles. Up to now, several applications of RCM to natural products synthesis have been reported using Schrock’s catalyst as the initiator; however, its air- and moisture sensitivity, which implies the use of a glove box or Schlenk techniques, has certainly hampered its more widespread use by organic chemists. In 1992, Grubbs and coworkers reported the first stable vinylidene ruthenium catalyst to be active in both ring-opening metathesis (ROM) and RCM. In 1995, further refinements led to the development of an air- and moisture-stable as well as highly functional group-tolerant benzylidene ruthenium complex also known as Grubbs first-generation catalyst. The latter became the first user-friendly metathesis catalyst and has allowed numerous synthetic applications. The replacement of one phosphine by a strongly σ -donating N-heterocyclic carbene ligand to further improve the stability of the active species and accelerate the initiation phase stimulated the discovery of the second-generation catalysts. To date, many catalysts have been devised with the goal of improving the rate of initiation and the stability of the catalytic propagating species to enable the metathesis of sterically hindered substrates. This was attained by fine-tuning the steric and/or electronic properties of the benzylidene part or the N-heterocyclic carbene of the ruthenium complexes, and/or other subtle ligand exchange. For the tremendous impact of metathesis in the science of organic synthesis, Chauvin, Grubbs, and Schrock received the Nobel Prize in Chemistry in 2005. The aim of the book is to emphasize the impact of metathesis on the synthesis of natural products and/or biologically active compounds, and highlight how they Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts. Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32440-8

XVI

Preface

have provided new and elegant solutions to many synthetic puzzles. As RCM has been the first class of metathesis reactions routinely used in natural products chemistry, the first three chapters of the book will highlight its applications to the synthesis of small- to medium-size carbocycles (Chapter 1, N. Blanchard and J. Eustache), nitrogen heterocycles (Chapter 2, L. van Delft and Floris P. J. T. Rutjes), and oxygen heterocyles (Chapter 3, J. D. Rainier). Phosphorus and sulfur heterocycles synthesized via RCM also deserved a section since they have found useful applications in the stereoselective synthesis of acyclic subunits found in various natural products (Chapter 4, C.D. Thomas and P.R. Hanson). The use of RCM for the synthesis of macrocyclic compounds has also been covered (Chapter 5, A. Gradillas and J. Pérez-Castells) since it constitutes an attractive alternative to traditional routes such as macrolactonization or macrolactamization. Alkynes can also be used as reacting partners in metathesis reactions as illustrated in the two following chapters of the book. Indeed, while ene–yne metathesis catalyzed by alkylidene ruthenium complexes allows a convenient access to conjugated dienes (Chapter 6, M. Mori), ring-closing alkyne metathesis using a well-defined tungsten–alkylidyne complex or molybdenum precatalysts activated in situ offers a convenient route toward cycloalkynes (Chapter 7, P. Davies). As for many reactions, there are situations where a planned metathesis event was found to be either unsuccessful or did not operate with high efficiency, stereoselectivity, and/or chemoselectivity. Silicon-tethered metathesis (Chapter 8, P. A. Evans) and the use of an unsaturated relay allowing initiation of metathesis at an appropriate reactive site (Chapter 9, T. R. Hoye and J. Jeon) are two strategies that have been used to circumvent some of these problems. More recently, cross-metathesis (CM) has emerged as a useful catalytic and chemoselective alternative to traditional olefination methods. Applications in the context of natural product synthesis have therefore been covered (Chapter 10, J. Prunet and L. Grimaud). After disclosing the synthetic potential of each of the different metathesis reactions, it appeared important to illustrate how their combination in cleverly designed cascades has led to some impressive and elegant synthesis of structurally complex natural products (Chapter 11, M. Porta and S. Blechert). The development of chiral molybdenum or ruthenium catalyst has also enabled the achievement of enantioselective metathesis reactions whose applications yet reported to the synthesis of natural products have been listed in one chapter (Chapter 12, A. H. Hoveyda, S. J. Malcolmson, S. J. Meek, and A. R. Zhugralin). Finally, the last section of the book is devoted to solid-phase metathesis, which constitutes a useful tool in diversity-oriented synthesis for chemical biology while also simplifying the purification stages (Chapter 13, S. Barluenga, P.-Y. Dakas, R. Jogireddy, G. Valot, and N. Winssinger). We would like to warmly thank all the authors for contributing to this book and acknowledge their expertise on the different topics that have been covered. We also thank the team at Wiley-VCH and especially Stefanie Volk for her helpful assistance during the preparation of this book. We sincerely hope that this book will be a valuable source of information for researchers working in both academic and industrial laboratories and that it will

Preface

stimulate new applications and developments of metathesis in the field of natural product synthesis. Janine Cossy, Stellios Arseniyadis, and Christophe Meyer.

XVII

XIX

List of Catalysts

N F3C F3C

Cl

O Mo

Ru

Cl

PCy3

Ph

O

N Mes Mes N Cl Ru Ph Cl PCy3

PCy3 Ph

N Mes Mes N Cl Ru Cl O

CF3 CF3 [Mo]-I

Cl

[Ru]-I

PCy3 Ru

Cl

PCy3

Ph

Cl

Ph

Cl

[Ru]-II

PCy3 Ru PCy3

Ph

N Mes Mes N Cl Ru Ph Cl PCy3

[Ru]-V

[Ru]-IV

[Ru]-VI

PCy3 Cl Ru Cl O


[Ru]-VII

N

N

Cl Ru Cl O


Ph [Ru]-VIII

[Ru]-IX

[Ru]-X

N Br

Cl

N Mes Cl

Ru Ph N

N Mes Mes N Cl Ru Cl PCy3

O O S NMe2

[Ru]-XI Ph

Mes N

NO2

Me

Me N Mes Mes N Cl Ru Cl O

[Ru]-III

N

Ph i -Pr N

i -Pr Cl Ru Cl O

Br [Ru]-XII

[Ru]-XIII

(R,R )-[Ru]-XIV


XXI

List of Contributors Sofia Barluenga Université Louis Pasteur de Strasbourg Organic and Bioorganic Chemistry Laboratory Institut de Science et Ingénierie Supramoléculaire 8 Allée Gaspard Monge 67000 Strasbourg France

Pierre-Yves Dakas Université Louis Pasteur de Strasbourg Organic and Bioorganic Chemistry Laboratory Institut de Science et Ingénierie Supramoléculaire 8 Allée Gaspard Monge 67000 Strasbourg France

Nicolas Blanchard Université de Haute-Alsace Ecole Nationale Supérieure de Chimie de Mulhouse Laboratoire de Chimie Organique et Bioorganique associé au CNRS 3 Rue Alfred Werner 68093 Mulhouse Cedex France

Paul W. Davies University of Birmingham School of Chemistry Birmingham B15 2TT United Kingdom

Siegfried Blechert Technische Universität Berlin Institute of Chemistry Straße des 17. Juni 135 10623 Berlin Germany

Jacques Eustache Université de Haute-Alsace Ecole Nationale Supérieure de Chimie de Mulhouse Laboratoire de Chimie Organique et Bioorganique associé au CNRS 3 Rue Alfred Werner 68093 Mulhouse Cedex France


XXII

List of Contributors

P. Andrew Evans The University of Liverpool Department of Chemistry Crown Street Liverpool L69 7ZD UK Ana Gradillas Universidad CEU-San Pablo Departamento de Qu´ımica Facultad de Farmacia Urb. Montepr´ıncipe 28668 Boadilla del Monte Madrid Spain Laurence Grimaud Ecole Nationale Supérieure des Techniques Avancées Unité Chimie et Procédés 32 boulevard Victor 75739 Paris Cedex 15 France Paul R. Hanson University of Kansas Department of Chemistry 1251 Wescoe Hall Drive Malott Hall Lawrence, KS 66045 USA Amir H. Hoveyda Boston College Department of Chemistry Eugene F. Merkert Chemistry Center Chestnut Hill MA 02467 USA

Thomas R. Hoye University of Minnesota Department of Chemistry 207 Pleasant Street SE Minneapolis Minnesota 55455 USA Junha Jeon University of Minnesota Department of Chemistry 207 Pleasant Street SE Minneapolis Minnesota 55455 USA Rajamalleswaramma Jogireddy Université Louis Pasteur de Strasbourg Organic and Bioorganic Chemistry Laboratory Institut de Science et Ingénierie Supramoléculaire 8 Allée Gaspard Monge 67000 Strasbourg France Steven J. Malcolmson Boston College Department of Chemistry Eugene F. Merkert Chemistry Center Chestnut Hill MA 02467 USA


Simon J. Meek Boston College Department of Chemistry Eugene F. Merkert Chemistry Center Chestnut Hill MA 02467 USA

Javier Pérez-Castells Universidad CEU-San Pablo Departamento de Qu´ımica Facultad de Farmacia Urb. Montepr´ıncipe 28668 Boadilla del Monte Madrid Spain

Silvie A. Meeuwissen Radboud University Nijmegen Institute for Molecules and Materials Heyendaalseweg 135 6525 ED Nijmegen The Netherlands

Jon D. Rainier University of Utah Department of Chemistry 315 East 1400 South Salt Lake City UT 84112 USA

Miwako Mori Health Sciences University of Hokkaido Ishikari-Tobetsu Hokkaido 061-0293 Japan

Floris P. J. T. Rutjes Radboud University Nijmegen Institute for Molecules and Materials Heyendaalseweg 135 6525 ED Nijmegen The Netherlands

Marta Porta Technische Universität Berlin Institute of Chemistry Straße des 17. Juni 135 10623 Berlin Germany

Christopher D. Thomas University of Kansas Department of Chemistry 1251 Wescoe Hall Drive Malott Hall Lawrence, KS 66045 USA

Joëlle Prunet Ecole Polytechnique ` Laboratoire de Synthese Organique UMR CNRS 7652 DCSO 91128 Palaiseau France

Sebastiaan (Bas) A. M. W. van den Broek Radboud University Nijmegen Institute for Molecules and Materials Heyendaalseweg 135 6525 ED Nijmegen The Netherlands

XXIII

XXIV


Floris L. van Delft Radboud University Nijmegen Institute for Molecules and Materials Heyendaalseweg 135 6525 ED Nijmegen The Netherlands Gaële Valot Université Louis Pasteur de Strasbourg Organic and Bioorganic Chemistry Laboratory Institut de Science et Ingénierie Supramoléculaire 8 Allée Gaspard Monge 67000 Strasbourg France

Nicolas Winssinger Université Louis Pasteur de Strasbourg Organic and Bioorganic Chemistry Laboratory Institut de Science et Ingénierie Supramoléculaire 8 Allée Gaspard Monge 67000 Strasbourg France Adil R. Zhugralin Boston College Department of Chemistry Eugene F. Merkert Chemistry Center Chestnut Hill MA 02467 USA

XXV

Abbreviations Ac acac AIBN Ar AROM BBN 9-BBN Bn Boc BOM BQ Bz CAN Cat. Cbz CM COD Cp Cp* CSA Cy dba DBU DCC DCE DDQ DIAD DIPEA DMAP DMB dmdba DMDO DMF

acetyl acetylacetonato azobisiisobutyronitrile aryl asymmetric ring-closing metathesis 9-borabicyclononane 9-borabicyclo[3.3.0]nonan-9-yl benzyl tert-butyloxycarbonyl benzyloxymethyl 1,4-benzoquinone benzoyl ceric ammonium nitrate catalytic benzyloxycarbonyl cross-metathesis cycloocta-1,5-diene cyclopentadienyl 1,2,3,4,5-pentamethylcyclopentadienyl 10-camphorsulfonic acid cyclohexyl dibenzylideneacetone 1,8-diazabicyclo[5.4.0]undec-7-ene N,N -dicyclohexylcarbodiimide 1,2-dichloroethane 2,3-dichloro-5,6-dicyanobenzoquinone diisopropyl azodicarboxylate diisopropylethylamine N,N-dimethylaminopyridine 3 ,5 -dimethoxybenzoin bis(3,5-dimethoxybenzylidene)acetone dimethyldioxirane N,N-dimethylformamide


XXVI

Abbreviations

DMP DMP DMPU DMSO DPPA dppp EDC Et Fmoc Grubbs II HMDS HMPA HOBt i-pr IBX dIcr Im Ipc KHMDS LDA LiHMDS m-CPBA Me MEM Mes MOM Ms MS NaHMDS NBS NCS NIS NMO PCC Ph Phth Piv PMB PMP PPTS pTs Py Quant. RCAM

Dess-Martin periodinane dimethoxypropane 1,3-dimethyl-3,4,5,6-tetrahydro-2-(1H)-pyrimidinone dimethylsulfoxyde diphenylophosphoryl azide 1,3-bis(diphenylphosphino)propane 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide ethyl 9-fluorenylmethoxycarbonyl benzylidene[1,3-bis(2,4,6-trimethylphenyl)-2-imidazolidinylidene] dichloro(tricyclohexylphosphine) ruthenium hexamethyldisilazane hexamethylphosphoroamide hydroxybenzotriazole isopropyl 2-iodoxybenzoic acid (+)-?2-caranyl imidazole diisopinocampheyl potassium hexamethyldisilazide lithium N,N-diisopropylamide lithium bistrimethylsilylamide m-chloroperbenzoic acid methyl 2-methoxyethoxymethoxy mesityl methoxymethyl methanesulfonyl molecular sieves sodium hexamethyldisilazide N-bromosuccinimide N-chlorosuccinimide N-iodosuccinimide N-methylmorpholine N-oxide pyridinium chlorochromate phenyl phthalyl pivaloyl p-methoxybenzyl 4-methoxyphenyl pyridinium p-toluenesulfonate para-toluensulfonyl pyridine quantitative ring-closing alkyne metathesis

Abbreviations

RCEYM RCM ROM ROMP RRCM RRM rt t-Bu TBAF TBDPS TBS TCE Teoc TES Tf TFA 2-Th THF THP TIPS TMEDA TMS Tol-BINAP TPAP Tr Trt Ts TS

ring-closing ene-yne metathesis ring-closing metathesis ring-opening metathesis ring-opening metathesis polymerization relay ring-closing metathesis ring-rearrangement metathesis room temperature tertiary butyl tetrabutylammonium fluoride tert-butyldiphenylsilyl tert-butyldimethylsilyl trichloroethoxymethoxy 2-(trimethylsilyl)ethoxycarbonyl triethylsilyl trifluoromethanesulfonyl trifluoroacetic acid 2-thienyl tetrahydrofuran tetrahydropyran triisopropylsilyl N,N,N ,N -tetramethylethylenediamine trimethylsilyl 2,2 -bis(ditolylphosphino)-1,1 -binaphthalene tetra(n-propyl)ammonium perruthenate triphenylmethyl (trityl) trityl p-toluenesulfonyl transition state

XXVII

1

1 Synthesis of Natural Products Containing Medium-size Carbocycles by Ring-closing Alkene Metathesis Nicolas Blanchard and Jacques Eustache

1.1 Introduction

This chapter deals with the synthesis of naturally occurring molecules (or related models) and focuses on the construction of medium-size carbocycles by ring-closing metathesis (RCM). We have arbitrarily chosen to organize this chapter by increasing ring size. Strategic aspects and potential problems are discussed.

1.2 Formation of Five-membered Carbocycles by RCM

Strategic positioning of the double bonds of a 1,6-diene prior to RCM can be efficiently accomplished via [3,3]-sigmatropic rearrangements. Only selected examples are discussed below, based on originality and efficiency criteria. Catalytic asymmetric Claisen rearrangement was reported by Hiersemann et al. in the enantioselective synthesis of the C10–C18 segment of ecklonialactone B (1), a C18-oxylipin isolated from the brown algae Ecklonia stolonifera and Egregia menziessi [1]. The [3,3]-sigmatropic rearrangement of 2 using catalyst 3 led to the acyclic α-keto ester 4 that can be reduced and cyclized to the corresponding disubstituted cyclopentene 5 using [Ru]-II. Further functional group transformations led to the desired C10–C18 fragment 6 of the natural product (Scheme 1.1). A [3,3]-sigmatropic rearrangement/RCM sequence [2] was also used as a key step in the total synthesis of a bioactive spirobenzofuran 7 isolated from the mycelium cultures of Acremonium sp. HKI 0230 [3]. The two consecutive quaternary centers embedded in the 1,6-diene 9 are worthy of note in the context of cyclopentene formation by RCM, and the five-membered ring compound 10 was obtained in high yield (97%) using catalyst [Ru]-I. The same type of strategy was applied for the efficient construction of the two vicinal quaternary carbon atoms present in the herbertanes sesquiterpenes (Scheme 1.2) [4]. Besides the Ireland–Claisen [3,3]-sigmatropic rearrangement/RCM sequence, some examples of Johnson–Claisen/RCM were reported by Ghosh and Maity [5]. Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts. Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32440-8

1 Synthesis of Natural Products Containing Medium-size Carbocycles

2

2+

O

O N

N Cu t -Bu H O OH t -Bu 2 2

CO2Me O

2 SbF6− H

CO2Me

BnO

CF3CH2OH, CH2Cl2 rt, 3d 92%

2

CO2Me

H

H 4 (dr > 95 : 5, ee = 92%)

Et H OH OBn

O

16

13

O

1

O

( )7

O

H 6

5

2. [Ru]-II (1 mol%) Cl(CH2)2Cl 75% (2 steps)

Et

Steps

OH OBn

1. K-selectride (dr > 95 : 5)

O OBn

3 (7.5 mol%)

H 10 (−)-Ecklonialactone B (1)

Scheme 1.1 O 1. LDA TMSCl·Et3N O

Ar O

2. Reflux then HCl 3. CH2N2

[Ru]-I (5 mol%) CH2Cl2, rt CO2Me 97%

Ar

8

Steps

9

HO OH

Ar

O

CO2Me

Spirobenzofuran from Acremonium sp. HKI 0230 (7)

10

Ar = 2,5-Dimethoxy-4-methylphenyl

Scheme 1.2 OH O

O

CH3C(OEt)3 propionic acid

O

O

Xylenes, 140 °C 66%

OH 12

O

O

Steps

13

O

O

Steps

CH2Cl2, rt PMB 81%

PMB CO2Et

[Ru]-I (4 mol%)

HO

HO 14

HO

15

Sequosempervirin A (11)

Scheme 1.3

In the first total synthesis of sequosempervirin A (11), a norlignan isolated from Sequoia sempervirens, the γ , δ-unsaturated carbonyl derivative 13 was prepared from 12 by Johnson–Claisen rearrangement. Further steps led to 14, the precursor of the key RCM reaction catalyzed by [Ru]-I, which established the desired spiro structure of compound 15 (Scheme 1.3). An analogous Johnson–Claisen rearrangement/RCM sequence was used by Ghosh et al. in the synthesis of the carbocyclic core of the nucleoside (−)-BCA (16), a potential inhibitor of HIV reverse transcriptase (Scheme 1.4) [6]. The chiral cyclopentene 20 was obtained in excellent yield through a [Ru]-I-mediated RCM of 1,6-diene 19. Further epoxidation and functional group transformations led to (−)-BCA core structure 23.


3

NH2 N

HO

N

N

N

HO Bis(hydroxymethyl)cyclopentenyl adenine, (−)-BCA (16)

CH3C(OEt)3 propionic acid

OH 17

18 (dr = 1 : 1) BnO

O

O

Steps

H

O

[Ru]-I

H

O

140 °C, 6 h 68%

O

O

Steps

O

O

H

EtO2C

CO2Et

19 (dr = 3 : 1) BnO

HO

H

m -CPBA 85%

BnO

C6H6, 60 °C, 20 h 96%

O

Steps

NH2 HO

BnO

EtO2C 20

21

22

23

Scheme 1.4

1. LDA

O

2. HCl 3. CH2N2

O

Steps

OH

MeO2C

25

26

27 OH

[Ru]-II (15 mol%)

HO

Moist SiO2

Steps

p -Tol

CH2Cl2 28

29

Laurokamurene B (24)

Scheme 1.5

Another [3,3]-sigmatropic rearrangement/RCM sequence was developed by Srikrishna and coworkers in the first total synthesis of the aromatic sesquiterpene (±)-laurokamurene B (24) (Scheme 1.5) [7]. RCM of 27 led to cyclopentenol 28 bearing a labile hydroxyl moiety, both benzylic and allylic. The latter compound was found to be unstable and when treated with silica gel delivered the rearranged alcohol 29 in a one-pot operation. Derivatives of nitiol (30), a novel complex sesterterpenoid that acts as a modulator of IL-2 gene expression, were prepared using RCM of vinyl ketone 33 as a key step (Scheme 1.6) [8]. Worthy of note is the use of catalyst [Ru]-XIII [9] as [Ru]-II was found to increase intermolecular cross-metathesis at the expense of the desired

4

1 Synthesis of Natural Products Containing Medium-size Carbocycles OH

H

H HO

HO

Nitiol (30)

1,22-Dihydroxynitianes

O

O

O OPMB ( )4

O

LDA

( )2

HO

OPMB

Steps

( )2

OPMB

CH2Cl2, 40 °C 81%

TMSCl, Et3N 31

32

[Ru]-XIII (5 mol%)

33 OTBS SnBu3

OSO2CF3

O 1. LiHB(s-Bu)3 ( )2 OPMB

36

( ) OPMB 2

2. PhNTf2

CO2Me

OPMB OTBS ( ) 3

Cat. Pd(PPh3)4 CuCl, LiCl CO2Me

34

35

37

Scheme 1.6

RCM. Conjugate reduction of enone 34 using l-selectride followed by trapping of the resulting enolate with PhNTf2 delivered enol triflate 35. The latter was then submitted to a Stille cross-coupling reaction with vinyl stannane 36 leading to the advanced intermediate 37 in the synthesis of 1,22-dihydroxynitianes. In the total synthesis of (−)-allosamizoline (38), the carbocyclic fragment of the chitinase inhibitor allosamidin, Donohoe and Rosa reported the efficient elaboration of the cyclopentenyl moiety via the RCM of a 1,6-diene (Scheme 1.7) [10]. Wittig olefination of aldehyde 39, readily obtained from d-glucosamine, proved troublesome when basic phosphorus ylide 40 was used and the 1,6-diene 42 was isolated in 18% yield. Since the terminal substituents of the alkenes are not transferred into the cyclized products during RCM reactions, the reactivity of acrylate 43 was evaluated as the latter was obtained in high yield (91%) using the less basic ylide 41. RCM proceeds equally well, delivering cyclopentene 44 in 88% yield. Formation of cyclopentenes through the RCM of 1,6-dienes can prove itself difficult in very crowded environment. During synthetic efforts toward the tetracyclic skeleton of the sesquiterpenoid tashironin (47), Mehta and Maity reported the efficient RCM of 50 using [Ru]-I that led to 51 in 86% yield [11]. However, the RCM of 52 having a methyl substituent in allylic position led to 53 in considerably lower yield (25%) even in the presence of the more active [Ru]-II catalyst (Scheme 1.8). A rapid elaboration of the ABC core structure of the tetranortriterpene dumsin (54) was reported by Srikrishna et al. starting from the readily available (R)-carvone [12]. Two independent efficient RCM reactions, catalyzed by [Ru]-I, were used as


MOMO MOMO

Ph3P=CHR 40: R = H 41: R = CO2Me

O

MOMO MOMO

Toluene, 90 °C

NHCO2Me

[Ru]-II (10 mol%) R Toluene 90 °C

NHCO2Me

OMOM 44 80% (from 48) 88% (from 49)

39 42: R = H (18%) 43: R = CO2Me (91%) O Steps MOMO

NMe2

O NMe2 1. NIS

N H OMOM 45

NMe2

O Steps

N

2. AllylSnBu3 MOMO AIBN

NHCO2Me

MOMO

HO

N

HO

OMOM

OH

46

(−)-Allosamizoline (38)

Scheme 1.7 OH O

OH MeO

PhI(OCOCF3)2

MeO

O [4 +2]

O

MeO

Toluene reflux 76%

THF, rt 40% 48

50

[Ru]-II (20 mol%)

O

MeO

O

51

O

MeO

MeO

C6H6 reflux 86%

49

O

O

[Ru]-I (10 mol%)

O

BzO

O

O

OH O

RO

O O

OH

C6H6, reflux 25%

O

52

53

Tetracylic core structure

Tashironin (47)

Scheme 1.8 O

O

[Ru]-I (5 mol%)

Steps

O 1. AllylZnBr

CH2Cl2 99% (R )-Carvone

55

Li, liq NH3

2. PCC

Allyl bromide THF/t-BuOH

O

56

57 O RCM

AcO O AcO HO A

[Ru]-I (5 mol%) O 58

CH2Cl2 100%

O

C H B

O A′

O Dumsin (54)

C

A H B

O O

H 59

D

H

RCM

O

i Bu Dumsin ABC tricycle

Scheme 1.9

key steps to create the spiro cyclopentene in compound 56 and the trans-fused BC rings in compound 59 (Scheme 1.9). In challenging cases, simple transformations of the allylic substituents may promote a productive RCM of 1,6-dienes to the corresponding cyclopentenes as shown by Eustache et al. in the stereoselective synthesis of spiroepoxide 61, a

5

6

1 Synthesis of Natural Products Containing Medium-size Carbocycles O O O OMe CO2H

O O

Fumagillin (60)

O

i- Bu

Spiroepoxide analog (61)

62

NaBH4, CeCl3·7H2O MeOH, 0 °C

OH

i- Bu OMe 64

PMPO HO

O [Ru]-II (15 mol%) Toluene 70 °C 30%

OMe

OMe

i- Bu

CH2Cl2 40 °C 42%

OMOM

63

67

BF3·SMe2 CH2Cl2, −78 °C PMPO HO

i- Bu OMe 65

i- Bu OMOM

66

Dess–Martin periodinane CH2Cl2, 20 °C 55% (3 steps)

[Ru]-II HO (15 mol%) CH2Cl2 40 °C

i- Bu

OMe

PMPO [Ru]-II HO (15 mol%)

i- Bu OH 68

[Ru]-II (5 mol%) CH2Cl2 40 °C 80%

PMPO HO

i- Bu OH 69

Scheme 1.10

potential methionine aminopeptidase-2 inhibitor related to the natural product fumagillin (60) (Scheme 1.10) [13]. Actually, the RCM of vinyl ketone 62 was found to be sluggish using catalyst [Ru]-II (toluene, 70 ◦ C) yielding only 30% of the cyclopentenone 63. The enhanced reactivity of the corresponding allylic alcohol 64 allowed a significant yield increase (55%, over three steps from 62). However, the presence of a potential coordination site for the catalyst could also be detrimental to the efficiency of the RCM. Thus, RCM of the Methoxymethyl (MOM) ether 66 gave cyclopentene 67 in moderate yield (42%), whereas the corresponding free alcohol 68 cyclized to 69 in 80% yield with a threefold decrease in catalyst loading (Scheme 1.10). Cyclopentenes bearing a trisubstituted double bond could be efficiently prepared via the RCM of the corresponding 1,6-diene. Catalyst [Ru]-II is well suited for this transformation since [Ru]-I usually leads to inferior yields. The prochiral trisubstituted olefin could then be further reduced or oxidized, thus creating a new tertiary or quaternary stereogenic center. Trauner et al. used RCM as a key step to elaborate the cyclopentyl core of (−)-guanacastepene E (70) and (−)-heptemerone B (71) (Scheme 1.11) [14]. The cyclization precursor was prepared via an asymmetric ene reaction of glyoxylate 72. Further transformations led to 1,6-diene 74 that could be cyclized to cyclopentenone 75 in 86% yield using [Ru]-II (5 mol%). Conjugate addition of the organocuprate 76 then established the C11 quaternary stereocenter. The central seven-membered ring was then closed with an electrochemical oxidation. Fomannosin (79) is a sesquiterpene metabolite isolated from the wood-destroying Basidiomycetes fungus Fomes annonsus (Fr.) Karst. Its inherent instability and very congested structure have elicited a great deal of interest from the synthetic


O R*O

Steps R*O

SnCl4, −78 °C O

[Ru]-II (5 mol%)

OH O

72

O

73 (dr = 10 : 1)

R = (1S,2R )-2-Phenylcyclohexyl

OBn

Toluene, reflux 86%

OBn

O

74

75

Cu(CN)(2-Th)Li2 PO O 76: P = TBDPS

1. KHMDS 2. TBSOTf

11 PO

BF3·OEt2 THF, −40 °C

OBn

O

O

PO

3. Anodic oxidation

OBn

H

O

O

78

77

Steps

RO

OAc

H

O

O R = H: (−)-Guanacastepene E (70) R = Ac: Heptemerone B (71)

Scheme 1.11

PO

O

OMe

Cp2ZrBu2

OP

[Ru]-II (5 mol%)

Steps OP

OPMB

PMBO

80: P = TBDPS

OH

PMBO

81

OTBS

C6H6 reflux 91%

OP PMBO

82

OTBS 83

(dr = 2.4 : 1)

O Steps THF

O PMBO

OH

O

O SEt

1. IBX, DMSO 2. Pd/C, Et3SiH PMBO silica gel

Steps

O OH

O OH

84

85

O O OH (+)-Fomannosin (79)

Scheme 1.12

community. Paquette et al. reported a straightforward strategy for the elaboration of both antipodes of fomannosin. Compound 80 (prepared from α-d-glucose) underwent a zirconium-promoted ring contraction providing the four-membered ring 81. The challenging RCM of 82 catalyzed by [Ru]-II was used as a key step, and led to cyclopentene derivative 83 featuring a trisubstituted double bond adjacent to

7

8


OH

OH OMe

OHC

OBn

Br

Br2,

87

OBn

CH2Cl2 reflux, 20 h 88%

OMe

OBn

OMe

O

OMe

O

Bu3SnH

OBn

O HO2C

Steps

OH

AIBN 85%

OMe

91%

89

EtO

90

OEt

OMe

88

EtO Br

OMe

HO [Ru]-II

Steps

O

OMe

(+)-Puraquinonic acid (86)

91

Scheme 1.13

HO

HO O

OR

HO

OH HO

HO

HO

OH

Galactofuranosides (92)

OH

Carbagalactofuranose (93) BnO

BnO OH

BnO BnO 94

OH

OH

[Ru]-II (10 mol%) BnO Toluene, 60 °C, 24 h 24%

BnO 95

OH

BnO

BnO OAc

BnO BnO 96

OAc

[Ru]-II (8 mol%) Toluene, 60 °C, 24 h 89%

OAc BnO BnO

OAc

97

Scheme 1.14

a quaternary stereocenter (Scheme 1.12) [15]. The lactone ring was then elaborated by an intramolecular Knoevenagel reaction. RCM was used as a key step in the total synthesis of (+)-puraquinonic acid (86), a fungal metabolite that induces differentiation in HL-60 cells [16]. Starting from aromatic aldehyde 87, the RCM precursor 88 was prepared in a few steps. Following RCM leading to 89 (88%), the newly formed trisubstituted alkene was involved in a radical cyclization. Bromo acetal 90 was treated with Bu3 SnH and AIBN thus triggering a stereoselective 5-exo trig cyclization reaction. Further steps led to the natural product 86 (Scheme 1.13). Carbahexofuranoses and carbapentofuranoses syntheses have been intensively investigated and RCM reaction emerged as one of the most practical and efficient methods to this effect. Among the numerous syntheses of such derivatives,


the synthesis of 4α-carba-β-d-galactofuranose 93 is worthy of note [17]. A dramatic protecting group effect was observed during the RCM of 94 and 96. Whereas free allylic alcohols are known to enhance reactivity, RCM of 94 leads to the five-membered ring 95 in poor yield (24%). Under the same conditions, the diacetate 96 undergoes an efficient RCM providing 97 in 89% yield (Scheme 1.14). Two simultaneous RCM have been used in efforts toward the core structure of the elisabethin diterpenoids [18]. Trisallylcarveol (99), accessible from (R)-carvone, can be transformed in a single step under mild conditions albeit with a moderate yield into the tricyclic compound 100 possessing the core structure of the desired target molecule (Scheme 1.15). Tetrasubstituted double bonds embedded in a five-membered ring can also be efficiently prepared via RCM. In the total synthesis of (±)-spirotenuipesines A (101) H [Ru]-I (5 mol%)

Steps

(R )-Carvone

O


OH

O

H O

OH Me

99

100

OH

Elisabethin A (98)

Scheme 1.15 OAc [Ru]-II HO (5 mol%)

TBSO

OH 103 AcO

Steps OTBS

N2

104

Steps

H

O

O

O

Bu3SnH AIBN

H

Toluene

H

O

O

O

O 107

106

O

105

MeS(S)CO H

Cu(II) cat O

C6H6 reflux 82%

O 108 HO

HO I

1. KOH MeOH 2. HCl 3. I2, KI NaHCO3

O

Steps

O

OH

O

O O

HO 109

Scheme 1.16

O

O

Spirotenuipesine B (102)

HO Spirotenuipesine A (101)

9


10

and B (102), Danishefsky et al. simplified the preparation of a key intermediate, compound 104, from nine steps [19] to three steps when an RCM strategy was adopted (Scheme 1.16) [20]. The resulting tetrasubstituted double bond in 104 was then submitted to an intramolecular diazoester cyclopropanation reaction. Radical opening of the resulting strained cyclopropyl group in 107 delivered bicyclic lactone 108, which was converted to 109 by iodolactonization. The latter compound could be further transformed to (±)-spirotenuipesines A (101) and B (102). Another example of elaboration of tetrasubstituted double bonds by RCM was disclosed by Kotora et al. in a formal total synthesis of estrone (Scheme 1.17). Starting from 110, two zirconium-mediated cyclizations enabled the preparation of compound 112 whose RCM catalyzed by [Ru]-II (20 mol%) (toluene, 90 ◦ C) provided estrone precursor 113 in 82% yield [21].

MeO H F

1. Cp2ZrBu2 2. CuCl (10 mol%) F Cl

MeO 110

2. Methallyl chloride CuCl (10 mol%)

MeO 111

[Ru]-II (20 mol%)

H

1. Cp2ZrBu2

H

Steps Estrone

Toluene, 90 °C MeO 82%

MeO 112

113

Scheme 1.17

1. LDA O O

H

2. PhSe

O

3. Bu4NIO4

N

O OPMB

55% Bn XN*

[Ru]-I, Ti(Oi- Pr)4

OMe OPMB 117

Scheme 1.18

CH2Cl2, 55 °C 53%

Steps

XN* OH OPMB 116

115

O

O

O Steps OMe OPMB 118

O

2

3 8 7

1 1′ 4O 5

2′

3′

4′

5′

OMe OH Fumagillol (114)

6

1.3 Formation of Six-membered Carbocycles by RCM


Among the recently reported synthetic approaches to angiogenesis inhibitors such as ovalicin, fumagillin, and synthetic analogs, several use RCM to construct the six-membered ring characteristic of this family of molecules [22–24]. In the first metathesis-based synthesis of fumagillol (114) (Scheme 1.18) [22a], the nonfunctionalized C7–C8 bond was established through an RCM/hydrogenation sequence. In agreement with Fürstner’s observations, adding Ti(Oi-Pr)4 was required for the RCM of 117 to proceed with catalyst [Ru]-I [25]. The resulting advanced intermediate 118 was easily converted to fumagillol. The approach was used for the preparation of fumagillin analogs 121–123, (Scheme 1.19) [22b] Using the more active [Ru]-II eliminated the need for Ti(Oi-Pr)4 in the RCM of 119 to 120. The second approach to fumagillol (114) relies on a Claisen/RCM sequence (Scheme 1.20) [23c]. Starting from 1,2;5,6-di-O-isopropylidene d-mannitol (124), the ester 125 was prepared and subjected to a glycolate-Claisen rearrangement leading to 126 (77%). RCM of 126 afforded cyclohexene 127 in high yield (94%), which was converted to an advanced intermediate previously used by Sorensen et al. in their synthesis of fumagillol [26]. O

O

O

[Ru]-II (10 mol%) Toluene, 70 °C 78%

OMe OPMB 119

O OMe OCinn 121

OMe OPMB 120

O

O

OMe HO

OMe OCinn 122

Cinn =

OMe OCinn

O

123

Scheme 1.19

OH

O

O O

O O

Steps

1. KHMDS 2. TMSCl

TBDPSO

OH

O

124

[Ru]-I (10 mol%)

O O 127

Scheme 1.20

125

COOH

CH2Cl2, 40 °C 94%

PMBO

COOH O O

77%

O OPMB

PMBO

O

Steps Fumagillol

TBDPSO

126

11

12

1 Synthesis of Natural Products Containing Medium-size Carbocycles TBDPSO

O O O O

TBDPSO

O

OH Steps

O

[Ru]-II (2 mol%)

O

O Steps

Toluene, 80 °C 84% OMe

O

129

TBDPSO

TBDPSO

O O

O OMe OTES

O O

[Ru]-II (1 mol%)

Toluene, 80 °C 94% OMe

Steps 134

OMe 133 OH

OH

131

O

132 OH

OH +

130

OMe

O OH 1. Li 134

O OMe

2. TBAF 3. TPAP, NMO 4. VO(acac)2, t - BuOOH

O Ovalicin (128)

Scheme 1.21 Li Ar O

Ar

OHC

Steps

CuI, Bu3P

HO

OHC

62%

56%

137 HO [Ru]-II (5 mol%) CH2Cl2, rt 85%

138

139

Ar O

H

Ar

MgBr

Steps

OMe H OH

H

H

rac -Ottelione A (135)

140 Ar

HO

Ar

MgBr

OHC

HO

H

Ar

[Ru]-II (5 mol%)

138 62%

141 O Steps

CH2Cl2, rt 86%

H

142

143

OMe H OH

Ar =

OMe OTBDMS

H

rac -Ottelione B (136)

Scheme 1.22

In contrast to the two previous syntheses, the six-membered ring was constructed first in Takahashi’s et al. synthesis of ovalicin (128) (Scheme 1.21) [24]. Thus, 2,3;5,6-di-O-isopropylidene-d-mannose (129) was converted to the diastereomeric RCM substrates 130 and 131, which smoothly cyclized to compounds 132 and 133 in the presence of [Ru]-II catalyst. Clive et al. designed a synthesis of both otteliones A (135) and B (136), featuring an efficient cuprate conjugate addition/Grignard reaction/RCM sequence


COOH

CH2OAc

Steps

COOMe

Bu3SnH AIBN

CH2OAc CH2OAc

CH2OAc

146

147

Steps

148

[Ru]-II (5 mol%)

149 HO H N

Steps

CH2Cl2, 40 °C 95%

O 150

OH

145

O

O

HO

OH

COOH

Platencin (144)

Scheme 1.23

(Scheme 1.22) [27]. Starting from cyclopentenone, the cyclopentene carboxaldehyde 137 was prepared. Conjugate addition of the organocopper derived from 2-butadienyllithium generated 138 and subsequent treatment with vinylmagnesium bromide afforded the RCM substrate 139. This compound was cyclized using catalyst [Ru]-II to furnish intermediate 140, which was converted to rac-ottelione A in two steps. Using the same sequence, the epimeric trans-aldehyde 141 (obtained by equilibration of 138) afforded rac-ottelione B via 142 and 143. An asymmetric version of these syntheses was also developed. Very recently, two RCM-based but conceptually different formal syntheses of platencin 144 were disclosed [28, 29]. In both cases, the target was intermediate 145 that has been previously converted to platencin by Nicolaou et al. [30a] and Rawal et al. [30c]. In Lee’s et al. synthesis [28], the bicyclo[2.2.2]octane moiety was assembled through a radical cyclization/skeletal rearrangement [31]. Standard chemical manipulation then led to triene 149 whose RCM ([Ru]-II) afforded cyclohexenol 150 in 95% yield (Scheme 1.23). Mulzer’s et al. approach implies a nonobvious construction of the strained bicyclo[2.2.2]octane skeleton by RCM (Scheme 1.24) [29]. Starting from (−)-perillaldehyde (151), the decaline 152 was prepared. Wittig reaction afforded triene 153, which was submitted to RCM conditions ([Ru]-II, CH2 Cl2 , reflux). RCM completion required more than 24 hours and 8 mol% of catalyst but the yield of 154 was excellent (90%) and no side-products were reported. Compound 154b was then converted to intermediate 145 in two steps (Scheme 1.24). Polyprenylated acylphloroglucinols, such as garsubellin A (155) and hyperforin A (156), are bioactive molecules characterized by a polycyclic bridged system [32]. In 2005, Shibasaki et al. described a RCM-based total synthesis of (±)-garsubellin A (Scheme 1.25) [33]. The α,β-unsaturated ketone 157 was converted to cyclohexanone 158 and a Claisen rearrangement/RCM sequence was used to construct the characteristic bicyclo[3.3.1]nonane skeleton.1) Thus, enol ether 159 cleanly rearranged at high temperature to diene 160.2) Treatment of 160 by [Ru]-III [(20 mol%), toluene, reflux, 1) The Claisen/RCM sequence has been used

2) Ethylene carbonate was added to the reac-

repeatedly for constructing various ring systems.

tion mixture to compensate for carbonate hydrolysis in 183.

13

14

1 Synthesis of Natural Products Containing Medium-size Carbocycles NMe2

CHO

O

O

H

OTBDMS 68%

H

CH2

Ph3P CHO

80%

152

151

153

O H

[Ru]-II (8 mol%)

1. NBS

CH2Cl2, reflux 90%

2. CrCl2 O 154

145

Scheme 1.24 OEt

O

OTIPS

Steps

O Steps

OTIPS 200 °C

O O

O

96%

O 157

158

159

OMOM

O

OMOM

OTIPS

O O

Steps

O

Toluene, reflux 92%

O

OTIPS

O [Ru]-III (20 mol%)

O O

160

161

OMOM

OMOM

I O

O

HO

162

O O

O

O 1 step

HO

O

Garsubellin A (155)

Scheme 1.25

48 hours] led to intermediate 161 in excellent yield [34]. Finally, 161 was converted to 162 and then to (±)-garsubellin A (155). As shown in Scheme 1.26, the method could not be extended to hyperforin (156) as RCM of 163 led to 164 possessing a seven-membered ring (Scheme 1.26). Eight to ten-membered-rings, which are common in natural products [35], can be prepared by heterolytic fragmentation of bicyclic systems [36]. The latter can be formed by several methods among which RCM is one of the most efficient.


O

HO O

MeO

O

O

[Ru]-III

MOMO

O O

MeO

MOMO

163

164

Hyperforin A (156)

Scheme 1.26 1. Vinyl (or allyl) lithium OTBDMS 2. Allylbromide

COOMe O ( )n

3. TBAF

O

[Ru]-I (7 mol%) CH2Cl2, 20 °C

HO

Pb(OAc)4

( )n OH

166a : n = 0 166b : n = 1

165

O

167a : n = 0 (95%) 167b : n = 1 (87%)

MeOH ( )n O 168a : n = 0 (84%) 168b : n = 1 (78%)

Scheme 1.27

Mascare˜ nas et al. reported an RCM/ring fragmentation strategy to eightand nine-membered carbocycles (Scheme 1.27) [37]. Starting from 1,2cyclohexanedione, the dienes 166a and 166b were obtained. When submitted to RCM conditions ([Ru]-I, 20 ◦ C), only the diastereomers having the two unsaturated side chains in a syn relationship reacted. The resulting bridged compounds 167a and 167b were cleaved by lead tetraacetate to afford ketoesters 168a and 168b. By contrast, direct construction of the eight-membered ring 170 by RCM of 169 was unsuccessful (Scheme 1.28) [37a]. A related RCM/fragmentation protocol was used for preparing germacranolides that may exist as E,E-, E,Z-, or Z,Z-10-membered rings (Figure 1.1) [38]. Starting from (R)-carvone, the trienes 171 and 172 were prepared and RCM provided decalines 173 and 174, respectively (Scheme 1.29). Selective conversion of the secondary alcohols in 173 and 174 to the corresponding mesylates and Wharton fragmentation [39] led to (Z,Z)- and (E,Z)-germacratrienes 177 and 178. The more substituted germacrenes 179–182 were also prepared using this strategy [38]. A similar strategy was reported for a synthesis of (±)-periplanone C (183) (Scheme 1.30) [40]. The triene 185 was prepared in a few steps from 3-isopropylanisole and smoothly cyclized using [Ru]-I (3 mol%) to the trans-decaline TBDMSO

TBDMSO [Ru]-I or [Ru]-II

OH 169

Scheme 1.28

OH 170

15

16

1 Synthesis of Natural Products Containing Medium-size Carbocycles COOMe

HO

CHO

HO

O O

O

O

O

O

Figure 1.1

O

Various types of germacranolides. OH

HO

OMs

[Ru]-I (5 mol%)

OH

Steps

C6H6, 80 °C 94%

NaH 79%

HO

171

175

Steps

C6H6, 80 °C 78%

177

OMs

[Ru]-I (10 mol%) OH

O

HO

173

OH HO

O

NaH 67%

172

HO

HO

174

176

O 178

O

O

O

O

179

180

181

182

Scheme 1.29

OMe Steps

COOMe O Steps

OMs MeOOC [Ru]-I (3 mol%) HO

184

185

CH2Cl2, rt 81%

HO 186

Steps

KOH 74% O 187

O (±)-Periplanone C (183)

Scheme 1.30

186 (81%). Mono-O-mesylation and treatment with KOH afforded cyclodecanone 187, which was converted to (±)-periplanone C (183) in a few steps. Interestingly, using the same conditions, epi-185 did not cyclize. This was attributed to the favorable preorganization of 185, favoring RCM, whereas the more stable conformation of epi-185 places the two olefins away from each other (Scheme 1.31) [41].


17

MeOOC [Ru]-I

OH COOMe HO

No RCM

H

epi- 185 H

MeOOC

OH

H

OH

[Ru]-I HO

MeOOC

MeOOC 186

185 Scheme 1.31

R'O O

TMSO Steps OAc

O

TMSO

MsCl Et3N

[Ru]-I (30 mol%) OH

C6H6, reflux 90%

RO

OH 191

A

189

188

Scheme 1.32

TMSO

O

TMSO Steps

[Ru]-I (30 mol%) OH

C6H6, reflux 93%

OH

192

193

TMSO

194

TMSO Steps

[Ru]-I (30 mol%) OH 195

C6H6, reflux 83%

OH 196

OH

C H HO BzOAcO D O

75%

190

O B

197

Scheme 1.33

Access to anti-Bredt alkenes is possible using an RCM/fragmentation sequence [42]. The strategy has been used for preparing the bicyclic ketone 188 having a structure similar to the AB rings of taxanes 189. RCM of 190 catalyzed by [Ru]-I afforded 191 (90%) and subsequent fragmentation provided ketone 188 (Scheme 1.32). Application to the construction of other bridged ring systems has been examined. RCM of 192 delivered 193 in high yield (93%) and the cyclization of 195 to afford the strained cycloalkene 196 (83%) is noteworthy (Scheme 1.33). In 2006, Robichaud and Tremblay reported the first RCM-based formal synthesis of (+)-compactin (198) [43]. The functionalized cyclohexene 199 (ring A) was prepared and submitted to RCM conditions to construct the six-membered B

18

1 Synthesis of Natural Products Containing Medium-size Carbocycles HO

O O

O

OH OH

[Ru]-II (20 mol%)

TMS

TMS

OH H

OH

OH

OH H

O

H

A

CH2Cl2, reflux 85%

H 199

200

201

B

Compactin (198)

Scheme 1.34

OH Bu3Sn

OH

(S ) 204

I

OH

O

OH

BzO

OTBDMS O

Steps

LDA TBDMSCl

BzO

O

Cat. Pd(0) OH

OH

203

1. Ireland– Claisen 2. MeOH, DCC

OBz

205

OBz

206

O BzO

60%

MeOOC OBz

OMe

O Steps

[Ru]-II (5 mol%) OBz

207

CH2Cl2, reflux 91%

OBz

208

OH

(−)-Perrotinene (202)

Scheme 1.35

ring, thus completing the elaboration of the fully functionalized decaline system 200. Removal of the trimethylsilyl (TMS) group was difficult, and provided the target compound 201 in 31% yield, whose transformation to compactin had been previously described (Scheme 1.34). The first synthesis of (−)-perrotinene 202 was reported in 2008 [44]. The strategy is again characterized by an Ireland–Claisen rearrangement/RCM key sequence (Scheme 1.35). Alcohol 205 was prepared by Stille coupling between the aryl iodide 203 and the enantiomerically pure vinyl stannane 204. The ensuing, highly stereoselective, Ireland–Claisen rearrangement of the silyl ketene acetal derived from 206 allows a complete chirality transfer and the control of the configuration of the two stereogenic centers in 207. RCM of the latter substrate proceeded smoothly and gave cyclohexene 208 in high yield (91%). Recently, Li et al. described a synthesis of (−)-10α- and (−)-10β-hydroxy-4muurolen-3-one (209) [45]. Starting from (R)-carvone, compound 210 was prepared in a few steps and its RCM proceeded smoothly using catalyst [Ru]-II to afford intermediate 211 in quantitative yield (Scheme 1.36). Vannusal A (212) is a complex triterpene characterized by a pentacyclic bridged core. Nicolaou et al. have devised a synthetic approach to compound 216 possessing the core structure of vannusal A (Scheme 1.37) [46]. A Mn(III)-induced radical


O

O

Steps

(R )-Carvone

HO

210

OH

H

O

Steps

[Ru]-II (5 mol%) CH2Cl2, reflux 100%

H

HO

H

H

19

H

211

(−)-10a-hydroxy-4muurolen-3-one (209)

Scheme 1.36

O

Mn(OAc)3 Cu(OAc)2

MeOOC

H

OMe OAc

OMe OAc

76%

OMe

H

Steps

OTES H

COOMe 213

214

215 OH

1. [Ru]-II (33 mol%) CH2Cl2, 50 °C 84%

A

B H

C

OHC

OMe OAc

OH OAc

H

D

2. PPTS, MeOH 91%

OH

E

H

H

H OAc

216

OH

Vannusal A (212)

Scheme 1.37 O O

Cl

O

O

O

O

219, Pd(0)

Steps

82%

i -BuO

Cl

C6H6, 60 °C 97%

i -BuO 220 (ee = 87%)

218

[Ru]-X (5 mol%)

CF3

O

t -Bu N

Steps

i -BuO

B

Steps

Cl

Cl O

A

O

HO

(+)-Laurencenone B (222)

CF3

F3 C

Br 221

P

Cl

(+)-Elatol (217)

219

Scheme 1.38

cyclization of the β-ketoester 213 was used to prepare the bridged tricyclic system ABC 214. This compound was converted in a few steps to the RCM substrate 215 whose cyclization proceeded cleanly and in good yield (84%) using catalyst [Ru]-II (33 mol%). In 2008, Stoltz et al. reported the first synthesis of elatol 217 (Scheme 1.38) [47]. Starting from dimedone, the mixed carbonate 218 was prepared. Decarboxylative oxygen-to-carbon migration was effected using a Pd(0) complex based on the chiral

20

1 Synthesis of Natural Products Containing Medium-size Carbocycles 1. Et3B, O2, Bu3SnH 2. [Ru]-II (15 mol%) Ti(Oi -Pr)4 (15 mol%) CH2Cl2, 40°C 3. Zn, AcOH, Et3N

S O

l

lpc2B

MeS

OMe O

O

OMe O O

O

+

BOMO

224

OMe O H

H

O

225 TCE =

BOMO

O

H

HO

H

O

Durhamycine aglycone model (227)

226

CCl3

O

O

OMe O

O OH

OH BOMO

34% (3 steps)

OTCE

OTCE

BOMO 223

OMe OH

O OH OH

OH

OH

O

Durhamycine aglycone

Scheme 1.39

O HO

MeOOC OH Steps OH

OH D-Xylose

MeOOC

O Steps

[Ru]-I (15 mol%)

BnO

OBn OBn 228

CH2Cl2, 25 °C 92%

BnO

OBn OBn 229

HO HO

OH OH

(+)-Cyclophellitol (230)

Scheme 1.40

phosphine 219 [48] to afford the optically enriched ketone 220 (ee = 87%) in good yield (82%). The challenging RCM of 220, leading to a tetrasubstituted double bond, proceeded exceedingly well using catalyst [Ru]-X [49]. The RCM product 221 was then converted to (+)-laurencenone (222) then (+)-elatol (217) in a few steps. Roush et al. have reported the synthesis of a durhamycin aglycon model (Scheme 1.39) [50]. Allylation of aldehyde 223 with the allylic borane 224 provided 225 after the formation of a xanthate. Deoxygenation gave the requisite RCM precursor whose cyclization proceeded reasonably well provided that Ti(Oi-Pr)4 was used as an additive. As suggested by model studies, this additive may avoid the coordination of the ruthenium carbene by either the aryl ether (OBOM) or the neighboring ethers (OMe, OTCE). The mild conditions of the olefin metathesis reaction make it an ideal method for the preparation of cyclitols and carbasugars [51, 52]. In general, the syntheses start from readily available carbohydrate derivatives that are converted to dienes and then submitted to RCM conditions. The resulting functionalized cyclohexenes are then converted to the target molecules in a few straightforward steps [53b]. According to this principle, the syntheses of (+)-cyclophellitol (230) (Scheme 1.40) [54], valienamine (233) (Scheme 1.41) [53, 55], valiolamine (236) (Scheme 1.42) [56], conduritol derivatives 238–240 (Scheme 1.43) [57], phosphatidyl inositol [58], and analogs (243) (Scheme 1.44) [59] have been reported. Noncarbohydrate starting materials may also be used for preparing carbasugars. For example, Roush’s et al. asymmetric allylation of chiral aldehyde 244 gave diene 245. Subsequent RCM catalyzed by [Ru]-II and dihydroxylation allowed access to

1.3 Formation of Six-membered Carbocycles by RCM BnO Steps

OH Steps

[Ru]-I (14 mol%)

BnO

OBn

BnO

BnO

HO

OH

O

BnO

OBn

OBn

CH2Cl2, rt 58%

BnO

2,3,4,6-Tetra-O -benzylD-glucopyranose

NH2

HO HO

OBn

OBn

OH OH

OBn

231

21

232

(+)-Valienamine (233)

Scheme 1.41 OTMS

O

OH

BnO

Steps TMSO

2,3,4-Tri-O -benzylD-arabinofuranose

OH Steps

BnO

OBn 92% (with [Mo]-I) 42% (with [Ru]-I) OBn OBn

OBn

BnO

RCM

NH2

HO

BnO

HO OBn

234

235

(+)-Valiolamine (236)

Scheme 1.42 OH OH OH HO

Steps

BnO

OH

BnO

OBn

OH

OBn

Glucitol

BnO

RCM 32% (with [Ru]-I) 89% (with [Ru]-II) 92% (with [Mo]-I)

BnO Mannitol

BnO

Galactitol

OBn

Steps BnO

Tetra-O - benzyl conduritol A (240)

Scheme 1.43 OH OH

Steps

OH

HO

OBn

Methyl a-D-glucopyranoside

Steps

HO

242 OH

O

O

P O OH OH

OC16H33

OH Phosphatidylinositol analog (243)

Scheme 1.44

OBn OBn

241

OH HO

OH

[Ru]-I

OBn 83% BnO

BnO

OH

OBn OBn

Tetra-O -benzyl conduritol E (239)

OMe

OBn OBn

BnO

Steps OBn

O

BnO

Tetra-O - benzyl conduritol F (238)

237

OH OH


22

Bd Ipc2

PhMe2Si

OBn

OBn SiMe2Ph

CHO 244

HO

87%

OBn OH

OBn

SiMe2Ph OH OsO HO 4

[Ru]-II

OBn OBn

245 SiMe2Ph OH

HO

246

247

OBn OBn

OH OH

HO or

HO

SiMe2Ph OH

OBn

HO

OBn

HO

OH

OH

OH

OH

or HO

OH OH

(+)-Conduritol B (249)

(+)-Conduritol F (248)

247

OH

HO

D-(+)-chiro -inositol (250)

Scheme 1.45 OH HO

OH OH

O

HO

OH

Steps

BnO

OBn

NH

O

O

OH O Pancratistatin (251)

COOMe 252

OBn OBn

Ar

Steps BnO

OBn OBn

[Ru]-I

BnO

90 – 95%

Ar COOMe

Ar2CuMgBr 95 – 97%

OH

D-Xylose

OBn BnO

253

OBn OH

254

OBn Ar = Ph, p - anisyl

Ar 255

Scheme 1.46

the pivotal intermediate 247. This compound was converted to conduritols F (248) and B (249) by stereospecific syn (basic conditions) or anti (acidic conditions) Peterson elimination, respectively, and debenzylation. Alternatively, Tamao–Fleming oxidation and debenzylation provided d-(+)-chiro-inositol (250) (Scheme 1.45) [60]. Pancratistatin (251), an important anticancer product, is structurally related to carbasugars. In 2004, Kornienko and Nadein reported a synthesis of 1-aryl1-deoxyconduritols that were used as synthetic intermediates to prepare a series of simplified pancratistatin analogs. Compound 252, prepared from d-xylose, underwent conjugate addition of magnesium diarylcuprates to afford 253. Further steps led to 254 whose RCM catalyzed by [Ru]-I provided aryl deoxyconduritols 255 (Scheme 1.46) [61, 62]. 1.4 Formation of Seven-membered Carbocycles by RCM

RCM of 1,8-dienes constitutes an efficient access to cycloheptenes since di-, tri-, and tetrasubstituted olefins could be elaborated in good yields with catalyst loading as low as 1 mol%. In this context, the highly demanding field of natural product synthesis constitutes an efficient test for the development of more active and

1.4 Formation of Seven-membered Carbocycles by RCM OMe

A

O

N

OMe N E

O O O

OMe

C B F

OMe

D OH OH

Methyllicaconitine (256) [Ru]-I (10 mol%)

N OH EtO2C 257

CH2Cl2, rt, 24 h 97%

N

O Steps

N EtO2C

O OH

258

N

O

OH O

259

Scheme 1.47

functional group tolerant catalysts as shown by Stoltz in the (−)-cyanthiwigin F synthesis. The following section, by no mean exhaustive, covers the more recent applications in RCM of 1,8-dienes applied to the synthesis of biologically relevant targets. When possible, comparison between various catalysts and experimental conditions is also discussed. Challenging RCM was reported during the synthesis of the ABE tricyclic analogs 259 of the alkaloid methyllicaconitine (256), a competitive antagonist of nicotine acetylcholine receptors [63]. The B ring system was synthesized via a [Ru]-I catalyzed RCM of the 1,8-diene 257 possessing two contiguous quaternary stereogenic centers. Compound 258 having the characteristic trans AB ring fusion of methyllicaconitine was obtained in excellent yield (97%) (Scheme 1.47). A key step in the total synthesis of (±)-frondosin B (260) reported by Mehta and Likhite is RCM of 1,8-diene 262 using [Ru]-I (C6 H6 , reflux) [64]. The corresponding bicyclic 6,7-fused ring system 263 was obtained in good yield (80%). Cyclization of 264 mediated by boron trifluoride led to the corresponding benzofuran 265, and subsequent acid-catalyzed isomerization of the trisubstituted olefin then provided (±)-frondosin B (260) (Scheme 1.48). In synthetic studies toward rameswaralide (266), a promising anti-inflammatory agent, the cycloheptenone tricyclic core 269 was efficiently prepared through RCM of the acyclic ω-alkenyl acrylate 268 under very mild conditions using [Ru]-II (10 mol%) (CH2 Cl2 , rt) in high yield (90%) (Scheme 1.49) [65]. A related ω-alkenyl acrylate RCM was used in the first total synthesis of (+)-sundiversifolide (270) reported by Shishido et al. [66]. An excellent yield of chiral cycloheptenone 273 was obtained from 272 using catalyst [Ru]-II. Enone 273 was then chemoselectively reduced to the corresponding allylic alcohol 274, thus setting the stage for an Eschenmoser–Claisen rearrangement. Further iodolactonization/radical reduction led to bicyclic lactone 275, a late intermediate in the synthesis of the natural product (Scheme 1.50).

23


24

OMe

OMe MeO

MeO H

Steps

O

OMe

OTBS

C6H6, reflux 80%

OH

261

MeO H

[Ru]-I

Steps

OH

262

OH

OTBS

263

HO

A HO H

O

O

H

BF3·OEt2

TsOH

CH2Cl2, 0 °C 264

B

HO

C O

D

C6H6

(±)-Frondosin B (260)

265

Scheme 1.48 OH OTES

H O

O

H

OTES

[Ru]-II (10 mol%)

Steps O

O

OTES O

H

O

A

O

B

OTES

O OH

O

H

268

CO2Me H

H

O

CH2Cl2, rt 90%

H

267

H

269

D O

C

H Rameswaralide (266)

Scheme 1.49 OH

O

O N

O O

O

Steps

[Ru]-II (5 mol%) TBSO

CH2Cl2, reflux 95%

Bn 271 OH

Red-Al TBSO

272 1. Me2NC(OMe)2Me toluene

TBSO 2. I2, THF, H2O 3. Bu3SnH, AIBN 274

Et2O 273

H

Steps

O TBSO

HO

H O O

O H 275

H (+)-Sundiversifolide (270)

Scheme 1.50

Trisubstituted olefins embedded in seven-membered carbocycles are also synthesized in high yields from the corresponding 1,8-dienes. In their elegant total synthesis of (−)-cyanthiwigin F (276), Stoltz and Enquist used for the first time catalyst [Ru]-X [49] possessing an increased activity for the generation of trisubstituted olefins. A double Pd-catalyzed decarboxylative allylic alkylation of 277 in

1.4 Formation of Seven-membered Carbocycles by RCM

O O

O

Steps

t- Bu

O O O

O

N

Ph2P

Pd(dmdba)2 Et2O, 25 °C 78%

O

O

277 dmdba = Bis(3,5-dimethoxybenzylidene)acetone

[Ru]-X (10 mol%) C6H6, 60 °C

t- BuSH AIBN

O

O

B

C

H

Steps H

C6H6, 80 °C 57%

O

O 279

278 (dr = 4.4 : 1, ee = 99%)

B

A

O

O

O

O

then NaBO3 51%

280

281

(−)-Cyanthiwigin F (276)

Scheme 1.51

NaOMe

Cl

Steps

THPO

MeOH 0 °C

O 283

H

CO2Me

H TESO

H

AD-mix a

PO

OMOM

PO

OH

96% H TESO 286

OEt

H TESO 287

CH2Cl2 88%

OEt

285: P = TBDPS

284 OMOM

[Ru]-II (5 mol%)

PO

Hept H

OMOM

H

H

H

THPO

O Steps

O OAc O

O H

Pr O OH

O

O

OH

O Thapsigargin (282)

O

Scheme 1.52

the presence of a chiral ligand gave 278, which was converted to tetraene 279 [67]. RCM of 279 followed by a one-pot cross-metathesis with vinylpinacolboronate led after oxidation to the bicyclic aldehyde 280 possessing the BC ring system of the natural product. Radical cyclization of the latter compound (t-BuSH, AIBN) established the desired ABC tricyclic motif 281 that could be further transformed to (−)-cyanthiwigin F (276) (Scheme 1.51). The guaianolide class of natural products is characterized by a tricyclic 5,7,5-ring system [68]. Among these promising sesquiterpene lactones were identified the thapsigargins (such as 282, Scheme 1.52), potent inhibitors of sarco/endoplasmic reticulum ATPases (SERCAs). Recently, Ley reported a successful strategy to access these complex natural products based on the construction of the seven-membered ring by RCM. Compound 283, prepared from (S)-carvone, underwent Favorskii ring contraction and 284 was converted to enol ether 285. RCM of 285, using catalyst [Ru]-II (5 mol%) allows the formation of the bicyclic enol ether 286 in

25

26


excellent yield (88%) [69, 70]. The latter was then efficiently converted into the corresponding α-hydroxyketone 287 via asymmetric dihydroxylation. The same key step was used in the total synthesis of closely related guaianolides, trilobolide, nortrilobolide and thapsivillosin F [71]. Ingenol (288) is a diterpenoid possessing a fascinating bicyclo[4.4.1]undecane skeleton, with a highly strained intrabridgehead topology. RCM-based strategies for the construction of the B ring of ingenol were reported by Wood [72] and Kigoshi [73]. In Wood’s total synthesis, compound 289 (synthesized from 3-carene) underwent Diels–Alder cycloaddition with cyclopentadiene to afford 290 (Scheme 1.53). Ring-opening metathesis (ROM) with ethylene gave 291, which was further elaborated to the RCM precursors 292 and 293. Only 45% of

H2C CH2 [Ru]-I (2 mol%)

Cyclopentadiene BF3·OEt2 59%

O

98%

O

289

Steps O

290

291 D

O O

H O R

[Ru]-I (80 mol%) for 292

O

O

H

O [Ru]-III (25 mol%) for 293

292: R = H 293: R = CH2OPMB

C

Steps

H

A

H

O B

HO HO HO

R

294: R = H (45% from 292) Ingenol (288) 295: R = CH2OPMB (76% from 293)

Scheme 1.53

O Et3CONa O

Cl

Steps

Et3COH/xylene reflux, 53%

H

O

296

H

297

298

O [Ru]-II (25 mol%) Toluene, reflux 87%

O H

SeO2 Dioxane

H 299

Scheme 1.54

OH

H H CHO Winkler's intermediate (300)


27

the desired cycloheptenone 294 was obtained from 287 using [Ru]-I (80 mol%), whereas an excellent 87% yield of 295 was obtained from 293 with [Ru]-III (25 mol%). Kigoshi’s formal synthesis (Scheme 1.54) relies on the efficient spirocyclization reaction of ketone 296 also prepared from 3-carene. The AC ring system was then further elaborated into 1,8-diene 298 that could be cyclized to 299 in 87% yield using [Ru]-II (25 mol%). A dramatic temperature and solvent effect was observed as no reaction took place in dichloromethane. Allylic oxidation then delivered aldehyde 300, an intermediate in Winkler’s total synthesis of ingenol [74]. A rare example of formation of cycloheptadiene from 1,3,8-triene was reported by Hiemstra et al. in the enantioselective synthesis of the tetracyclic left-hand substructure 301 of solanoeclepin A (302) (Scheme 1.55) [75]. Starting from compound 303, a [4+2]-cycloaddition was used to elaborate the oxabicyclic motif. Nolan’s catalyst [Ru]-VI (15 mol%) was used to cyclize the 1,3,8-triene 305 to the desired cycloheptadiene 306 in quantitative yield. It is worth mentioning that [Ru]-I was quite inefficient in this case even if a stoichiometric quantity was used. A concise asymmetric total synthesis of (+)-cyanthiwigin U (307) was reported by Phillips and Pfeiffer relying on the clever use of tandem metathesis of bicyclo[2.2.2]octenes (Scheme 1.56) [76]. The two-directional tandem ROM–RCM of compound 308, using catalyst [Ru]-II under an atmosphere of ethylene, simultaneously generated the cyclopentenone and the cycloheptenone rings of Ph OH

N O

1. n- BuMgCl −60 °C

Ph O

2. Toluene reflux

O

OAc

[Ru]-VI (15 mol%) Toluene, reflux 99%

O HO

303 TBDPS O O

N

TBDPS O O

Steps

304

305 HO2C

OH

OAc

HO

O

Steps

O

O OMe 306

H

O O

O

301

Solanoeclepin A (302)

Scheme 1.55

O H2C CH2 [Ru]-II (20 mol%)

O

O

Toluene >43%

O

OH Steps H

H O

308 Scheme 1.56

309

O OH

Cyanthiwigin U (307)

28


309 with great efficiency. This synthetic strategy was later extended to the synthesis of cyanthiwigin W and Z [77]. An RCM was reported by Wicha et al. in a synthetic approach to the diterpenes guanacastepene A (310) and heptemerone G (311) (Scheme 1.57) [78]. The 1,8-diene 312 possessing a 1,1-disubstituted olefin was cyclized efficiently to 313 in high yield (96%) using catalyst [Ru]-II. Further functional group transformation including epoxidation and Dauben oxidation [79] led to the AB core structure 314 of the targeted diterpenes. This very efficient RCM should not mask the fact that trisubstituted double bonds of terpenoids, imposed by the biosynthesis pathway from isoprene units, could be challenging to prepare through the RCM of 1,8-dienes as subtle differences in the metathesis precursor can have a dramatic impact on the reaction rate and yields [80]. For example, [Ru]-I could efficiently catalyze the RCM of the substituted 1,8-diene 315 to the corresponding trisubstituted cycloheptene 316 as demonstrated by Tori et al. in the total synthesis of two sphenolobane-type diterpenes (Z,E)- and (E,E)-320 isolated from the liverwort Anastrophyllum aurium (Scheme 1.58) [81]. However, during a synthesis of (−)-tormesol (321), a diterpene isolated from Halimium viscosum, the cyclization of the related nonenolizable β-ketoester 322 proved sluggish as only 50% conversion was observed using catalyst [Ru]-I in combination with Ti(Oi-Pr)4 (1 equiv.). The more active [Ru]-II overcame this limitation and allowed the synthesis of cycloheptene 323 in excellent yield with a low O

H

[Ru]-II

1

R

O OPiv CH2Cl2, reflux 96% O

O 312

OPiv 313

OHC

O

Steps

i- Pr 314

A

B

OR2

C

i- Pr

HO

R1 = OAc, R2 = H : Guanacastepene A (310) R1 = H, R2 = Ac : Heptemerone G (311)

Scheme 1.57 TESO CO2Et

TESO

[Ru]-I (10 mol%)

CN

CO2Et Steps

Steps

CH2Cl2, 40 °C 94% 315 O H

316

H

H 1.

O

MgBr

2. PCC 318

Scheme 1.58

OH

317

CHO

H Steps

O

O 319

(Z,E )- and (E,E )-320 Sphenolobane-type diterpenoids


29

catalyst loading (1 mol%) [82]. Compound 323 was then converted to (−)-tormesol through intermediates 324 and 325 (Scheme 1.59) [81, 82]. A very challenging RCM was reported by Reiser et al. during the enantioselective synthesis of arglabin (326), a promising antitumor agent isolated from Artemisia glabella [83]. Condensation of cyclopropanecarbaldehyde 327 with allylic silane 328 led, after treatment under basic conditions, to adduct 329, which was further elaborated to 330. The tetrasubstituted double bond embedded into the seven-membered ring of 331 was efficiently prepared via the RCM of compound 330. Worthy of note is that an inert gas sparging of the reaction combined with catalyst [Ru]-II (15 mol%) had to be used to ensure complete conversion. Epoxidation from the α-face of the seven-membered ring proved troublesome since the more stable β-epoxide was obtained using dimethyldioxirane or peracids. A hydroxyl-directed vanadium-catalyzed epoxidation of the homoallylic alcohol 331 overcame the intrinsic bias of the system, leading to the desired epoxide 332 in good yield and selectivity (Scheme 1.60). [Ru]-I (30 mol%) Ti(Oi- Pr)4 (100 mol%) 40 °C, 50% conv.

O CO2Et

O CO2Et

or [Ru]-II (1 mol%), 45 °C 88 – 95%

322

O CN

323

324 HO

O H

Steps

Steps

H

Li Et2O, −78 °C

(−)-Tormesol (321)

325 Scheme 1.59 O CO2Me

TMS

BF3·OEt2, −78 °C then

O CO2Et + OHC 327 1. [Ru]-II (15 mol%) toluene, 95 °C sparging with Ar (70%)

H O

O

OH 331

Scheme 1.60

H 330 OPMB

OPMB AcO H

H Steps

O 78%

H

H O

329

O 2. DDQ (90%)

O

H

TBHP V(O)(acac)2 (2 mol%)

H

Steps

H O

Ba(OH)2·8H2O 0 °C OPMB 62% 328 AcO H

AcO H

CHO

O

H O

O H

332 OH (a/b = 90 : 10)

H O H

O

(+)-Arglabin (326)


30

1.5 Formation of Eight-membered Carbocycles by RCM

Cyclooctanoids are among the most difficult carbocycles to elaborate from acyclic precursors due to severe developing ring strain and restricted bond rotation in the precyclization conformer [84]. In the mid 1990s, RCM of 1,9-dienes emerged as a powerful strategy for the synthesis of carbocyclic eight-membered ring [85]. Steric and conformational effects were found to have a dramatic impact on the rate and efficiency of cyclization reactions. High catalyst loading (up to 50 mol%), long reaction times, and elevated temperatures were usually reported. Up to 2006, an excellent review compiles the recent advances in the synthesis of challenging di[86] and trisubstituted [87] double bonds embedded in cyclooctenyl rings [88]. The dicyclo[a,b]cyclooctane ring systems found in ophiobolin M (333) and fusicoccin A (334) terpenes have stimulated a great deal of interest. The groups of Williams [89] and Wicha [90] reported two different RCM approaches aiming at the rapid elaboration of the eight-membered B ring system. The first approach relied on a [Ru]-I catalyzed RCM of triene 335 leading to cyclooctene 336 in moderate yield (Scheme 1.61) [89]. Worthy of note is the fact that [Ru]-II proved inferior as a catalyst leading to lower conversion and partial isomerization of the terminal olefin(s). The second approach [90] reports the synthesis of the AB ring system of serpendione (337) via RCM reaction of 338 (Equation 1, Scheme 1.62) that contains O O

O

O H

[Ru]-I (6 mol%) ( )3

HO RO H

CH2Cl2, 45 °C 45%

H

OAc

H

335

OH

HO

336

OMe Ophiobolin M (333)

Fusicoccin A (334) (R = Carbohydrate)

Scheme 1.61 COOMe

O

COOMe MeO

H

[Ru]-II (3 mol%)

H

H

(1)

CH2Cl2, reflux 95% O

H

O

O

O 338

339

COOMe

COOMe [Ru]-II (3 mol%)

H

CH2Cl2 reflux 95%

O 340

Scheme 1.62

H

COOMe +

O 341 (34%)

Serpendione (337)

H (2)

O 342 (49%)

1.5 Formation of Eight-membered Carbocycles by RCM

31

OTBDPS OTBDPS

O

[Ru]-I (20 mol%)

Steps O

C6H6, reflux 83%

Me

Me

344

345

O O

H Me 346

AcO

O

OH

V(O)(acac)2 TBHP 76%

O

H Me 347

O

OH

Steps O

H Me

O

O

(±)-Mycoepoxydiene (343)

Scheme 1.63

both a geminal disubstituted and a monosubstituted olefin. The [Ru]-II catalyzed transformation was extremely efficient and delivered compound 339 possessing the AB ring system in 95% yield. When exposed to [Ru]-II, the epimeric compound 340 could be transformed to the desired cyclooctene 341, albeit in low yield (34%). Isomerization of the allylic double bond prior to cyclization accounts for the major cycloheptene product 342 (49%) (Equation 2, Scheme 1.62), thus once again stressing the importance of subtle steric and conformational effects in the formation of medium rings by RCM. By contrast, the use of [Ru]-I led only to dimerization. Imposing a conformational constraint that favors cyclization is a useful strategy for the RCM of 1,9-dienes. In the total synthesis of (±)-mycoepoxydiene (343), Tadano et al. reported the RCM of the diallylated tetrahydrofuran 344 using catalyst [Ru]-I under diluted conditions to prevent oligomerization [91]. A good yield of the desired oxygen-bridged cyclooctene 345 was obtained (83%). A dramatic solvent dependence was noted as CH2 Cl2 induced only decomposition of the starting diene 344. Compound 345 was converted to (±)-mycoepoxydiene 343 in several steps including oxidation of furylcarbinol 346 to dihydropyranone 347 (Scheme 1.63). Vinigrol (348) is a metabolite of the fungal strain Virgaria nigra that possesses a fascinating tricyclo[4.4.4.04a,8a ]tetradecane skeleton [92]. Its biological activity, including antihypertensive, platelet aggregation-inhibiting, and tumor necrosis factor antagonist properties, has fueled a considerable synthetic interest. The groups of Barriault [93] and Paquette [94] investigated two different approaches involving disconnections of the carbocyclic eight-membered ring through RCM at the C9–C10 or C10–C11 positions, respectively. Unfortunately, these two strategies met with failure. As observed in previous studies, [Ru]-I was not active enough to promote the RCM of 1,9-dienes 349, 350, and 351, and [Ru]-II led to double bond migration (Scheme 1.64). In order to increase conformational flexibility in the cyclization precursor, Paquette et al. considered the relocation of the double bond from the 3,4 to the 2,3 site to project axially the two C1 and C5 substituents. As shown by MM3 calculations on model systems, the closer proximity of these two substituents in the 2,3 isomer could allow a more RCM. Diene 352 was then prepared and

32

1 Synthesis of Natural Products Containing Medium-size Carbocycles 10

HO

HO

9

11

H HO Vinigrol (348) 10

9

H O

[Ru]-I or [Ru]-II X

O H

O H

O

10

O

[Ru]-I or [Ru]-II R1

9

10

O

11

H

X

R2

10

R1 R2

H 11

349

350: R1, R2 = OCH2CH2O 351: R1 = OTBDPS, R2 = H

Scheme 1.64

3

4

3 4

5 1

OH

H 5H H

1 H

H

OH

3

5

3 2

OH

1

OH H

∆3,4 isomer

H ∆2,3 isomer

AcO AcO

2

10 5

3

OH H 5 H H 2 1 H OH

HO

OH

1

[Ru]-II OTBDPS

X

AcO AcO

11

H OP

352

Scheme 1.65

subjected to catalyst [Ru]-II but unfortunately, only starting material was recovered (Scheme 1.65). Prunet et al. have extensively investigated the possibility of synthesizing the B  ring system of Taxol (353) via the RCM of 1,9-diene 355 (Scheme 1.66) [95]. When a diastereomeric mixture of the latter compound and epi-355 was subjected to catalyst [Ru]-I, only compound 355 cyclized to cyclooctene 356 (possessing the C9–C10 unsaturation) albeit slowly, emphasizing once again that RCM is highly dependent upon the conformational and steric effects in the cyclization precursor (Equation 1, Scheme 1.66). Recently, the formation of the C10–C11 unsaturation by RCM was reported and proved more efficient in terms of protecting group compatibility and catalyst efficiency (Equation 2, Scheme 1.66). Unprotected diol 357 led to the desired cyclooctene ring 358 in an impressive quantitative yield, thus suggesting that the BC ring system of taxol may be synthesized according to this strategy. These recent results are in stark contrast to the previous C9–C10

1.6 Formation of Nine-membered Carbocycles by RCM R2O R1HN

O OH

O

Ph

B O

OH

33

A HO

C DO

H

OBz OAc

R1 = Bz, R2 = Ac: Taxol (paclitaxel) (353) R1 = Boc, R2 = H: Taxotere (docetaxel) (354)

10 9

10 9

+

1 2

Bu

O

1 2

Bu

O O

O

1

C6H6, reflux, 8 d

O

Bu

O

[Ru]-II (5 mol%) CH2Cl2

1 2

Bu

O

(1)

O O

O

epi - 355 (46%)

355

356 (34%)

(2)

Reflux 100%

OH OH 357

+

2

O

O

epi - 355

Bu

10 9

[Ru]-I (10 mol%)

Bu OH OH 358

Scheme 1.66

RCM strategy that was thwarted by conformational constraints imposing a cyclic carbonate protecting group of the C1–C2 diol motif.

1.6 Formation of Nine-membered Carbocycles by RCM

The failed metathesis-based synthesis of pestalotiopsin A 359, compared to the successful synthesis of cornexistins, reported by Paquette et al. [96] and Clark et al. [97] illustrates well the uncertainty linked to RCM toward 8 to 10-membered rings. RCM of several substrates 360–362 was evaluated (Figure 1.2). For 360 and 361, regardless of the catalyst used ([Ru]-I, II, III, VII, and XI), the starting diene was either recovered (when the reaction was performed below 80 ◦ C) or decomposed (in refluxing toluene). The MOM ether 362 was isomerized to 363.

AcO

O

O O HO

H H

O O OMe OH

Pestalotiopsin A (359)

Figure 1.2

H H 360

O

O O O

OMe OH

H H

O

O O O

OMe OH

H H

361

Pestalotiopsin a synthesis:unteactive RCM substrates.

362

O O O

OMe OMOM

H H 363

OMe OMOM


34

HO

HO

O OH

5

RCM O

O

O O

Hydroxycornexistin (364) O

O

[Ru]-I, CH2Cl2 reflux (70%)

O

O

O 365

366

H

O

O H

+

or [Ru]-II, toluene 80 °C (61%) O 367 (dr = 1 : 1)

O 368

O 369 368/369 = 2 : 3

Scheme 1.67

In 2003, Clark et al. reported a novel strategy for the preparation of cornexistins. In model studies, attempts to form intermediate 364, featuring a trisubstituted olefin, by the RCM of 365 were unsuccessful regardless of the catalyst used ([Ru]-I or [Ru]-II). In contrast, the cyclization of 367 (a racemic mixture of two diastereomers) led to the isomeric mixtures 368 and 369 in good yield. In subsequent studies, the RCM reaction was examined more closely and shown to be significantly stereoselective (368/369 = 2 : 3) [98].3) Compound 368 was converted to (±)-5-epi-hydroxycornexistin in a few additional steps (Scheme 1.67).

1.7 Formation of 10-membered Carbocycles by RCM

In 2001 and 2002, Koskinen et al. reported what appeared to be the first examples of 10-membered ring carbocycle syntheses via RCM [99]. As can be seen in Table 1.1, the reaction proved difficult to perform and yields remained moderate even under optimized conditions. Using alcohol 370a (syn/anti = 2 : 1), a single cyclized product (371a) was obtained in low yield and some starting material was recovered. With the silyl ether 370b as substrate, the reaction was cleaner, affording a mixture of three diastereomers in fair yield. Addition of Ti(Oi-Pr)4 was slightly beneficial. The roughly 2 : 1 trans/cis ratio of the major RCM products 371b and 372b compares well with the corresponding 2 : 1 syn/anti ratio of the substrate 370b [99a]. When 3) [Ru]-I catalyst was used for this transforma-

tion (18 hours, CH2 Cl2 , reflux, 77%). [Ru]-II

did not bring any significant improvement (personal communication from Dr Clark).


35

O O O

O O

HO Diversifolin (374) O

O O

HO

[Ru]-II (10 mol%) BQ (20 mol%) Toluene reflux

TBDMSO

O

O

HO

(R )-376 (31%) (S )-376 (40%)

O

HO

[Ru]-II (10 mol%)

+

Toluene reflux

TBDMSO

O

O

HO

OTBDMS

TBDMSO

(R )-376 (27%) (S )-376 (trace)

375 dr = 1 : 1

377 (35%)

Scheme 1.68 Table 1.1

Cyclization of alcohol 370a and TBDMS ether 370b.

[Ru]-I

i- Pr O

OR

+

CH2Cl2 reflux, 48 h O

370a : R = H 370b : R = TBDMS (syn/anti : 2 : 1)

+

i- Pr OR 371a 371b

Substrate

Substrate recovery (%)

370a 370b 370b + Ti(Oi-Pr)4 (20 mol%)

25 6 0

i- Pr O

OR

i- Pr O

372a 372b

OR 373a 373b

Products 371a (11%) 371b (16%) 371b (21%)

– 372b (35%) 372b (47%)

– 373b (11%) 373b (8%)

the more active [Ru]-II catalyst was used, only the syn-370b precursor cyclized to afford a mixture of 371b and 373b in a 1 : 6 ratio. The anti-370b isomer was not recovered and was probably converted to polymeric material [99b]. A synthetic study toward diversifolin (374) was published in 2007 by Kobayashi et al. (Scheme 1.68) [100]. First, a long sequence (17 steps) led to the key RCM substrate 375, obtained as a mixture of epimers. RCM was difficult and required significant optimization. While using [Ru]-I no reaction was observed, whereas [Ru]-II in 1,2-dichloroethane (DCE) led to the formation of 377 in low yield (15%) in which one of the terminal olefin had shifted. When toluene was used as the solvent,


36

O

O O

Me

O

O KN(SiMe3)2

OMe O

O

81%

H

Me

O OMe O

H O

Me

O

2. MsCl, base

O

379

1. HCHO LiN(SiMe3)2

H O

+

OTf

OMe O

O

380

381 380/381 = 1 : 1

O

Me

O

O OMe O +

O Me

O

H O

H O

O 382 (28%)

O

OMe O

Me

O

[Ru]-III 383

OMe O HCl

H

54%

OH

O O H O

82%

O

O 383 (32%)

384

O (+)-Eremantholide A (378)

O

Scheme 1.69

1. BuLi OMe F

MeO

OMe

2. ClMg

Steps

O O

3. OMe 386

OHC 387

OMe OH OSiPh2t - Bu 65%

[Ru]-II (40 mol%) slow addition Tetrafluoro-1,4-benzoquinone (80 mol%) toluene, 80 °C 65%

O OMe O

OSiPh2t - Bu

388

389

O

O

O CAN

O

O O

O

O 390

O O

O ent -Clavilactone B (385)

Scheme 1.70

377 was obtained in 35% yield along with (3R)-376 (27%), and some starting material was recovered. Only (3R)-375 cyclizes rapidly and no (3S)-376 was formed. RCM of (3S)-375 is more difficult allowing the competing olefin isomerization to prevail. This may be avoided by adding 1,4-benzoquinone (BQ), whose role is probably to prevent the accumulation of ruthenium hydrides (decomposition products of [Ru]-II) that are thought to be responsible for olefin migration [101]. In Hale’s and Li’s synthesis of (+)-eremantholide A (378) (Scheme 1.69), the RCM step takes place almost at the end of the synthesis and involves a fairly


37

hindered substrate. Epimers 382 and 383 were prepared by alkylation of triflate 379 followed by hydroxymethylation and subsequent dehydration. Using [Ru]-III as catalyst, the reaction works well. Remarkably, only epimer 383 reacted, affording the desired bicyclic system 384 in 54% yield and (+)-eremantholide A, while the other diastereomer 382 cross-metathesized [102]. In Barrett’s synthesis of ent-clavilactone B 385 (Scheme 1.70), the benzyne intermediate generated from 386 was allowed to react with methallylmagnesium chloride and epoxy-aldehyde 387, leading to compound 388, which was converted to the RCM precursor 389. As expected, RCM proved to be difficult and had to be performed under special conditions: the catalyst [Ru]-II (40 mol%) was slowly added, along with tetrafluoro-1,4-BQ (to prevent olefin isomerization), and the formed ethylene was removed during the reaction. Under these conditions, the RCM product 390 was obtained in a satisfactory yield (65%) and converted to ent-clavilactone B [103]. During the period 2001–2006, Gennari et al. published a series of accounts describing the synthesis, via RCM, of simplified eleuthesides as potential anticancer agents [104–106]. In particular, these authors recently reported a novel formal synthesis of eleutherobin 391 (Scheme 1.71) [105, 106]. Starting from (R)-(−)-carvone, the half-protected dialdehyde 392 was prepared. Diastereoselective reagent-controlled oxyallyltitanations (with TADDOL as O O H N

O H

N

OMe O OAc OH

O Eleutherobin (391)

H

H

CHO

OMe

1. PMPO s - BuLi [(S,S )-TADDOLCpTiCl] 2. MOMCl, Et3N

Steps

H

OMe

392

1. PMPO s - BuLi [(R,R )-TADDOLCpTiCl]

MOMO H

OPMP

OMe

H

OMe

393

MOMO H

[Ru]-II OPMP (30 mol%) slow addition

2. PivCl, DMAP, i- Pr2NEt H OPiv 395

Scheme 1.71

MOMO H

OH

OH

Toluene reflux 64%

OPMP

CHO 394

MOMO H

H

OPMP H

H Steps

O

H H OPiv 396

OH

O

H

OMe OPiv 397


38

the chiral ligand) of both aldehydic functions led to the RCM precursor 395, which was cyclized using [Ru]-II catalyst to afford the 10-membered carbocycle 396. The latter was converted to the advanced intermediate 397 that had previously been converted to eleutherobine. Forcing RCM conditions were necessary ([Ru]-II [30 mol%], slow addition, toluene, reflux), and surprisingly, only the (E)-cycloalkene 396 was obtained [25, 106, 107]. Using simplified models of 395, Gennari et al. have shown by a variety of methods (molecular mechanics, DFT) that the trans-four-membered ruthenacycle intermediates 398 leading to (E)-cyclodecenes are significantly more stable than the corresponding cis-ruthenacycles 399, leading to (Z)-cyclodecenes (Scheme 1.72). RCM of diastereomer 400 was more difficult than that of 395, affording the ring-cyclized product 401 in low yield alongside with regioisomer 402 resulting from a metathetic ring rearrangement (RCM–ROM–RCM cascade). Excentricine 403 belongs to a rare class of isoquinoline alkaloids featuring a bridged, bicyclic ether moiety. A possible approach to this class of compounds was illustrated by the synthesis of the simplified model compound 408 (Scheme 1.73) [108]. The strategy relies on a Bischler–Napieralski isoquinoline synthesis and an RCM as key reactions. The site of the RCM reaction was chosen so as to avoid the formation of inactive ruthenium complexes (likely to be observed for γ ,δ- or δ,ε-unsaturated amides or esters). RCM of diene 406 proceeded smoothly using catalyst [Ru]-II to afford the desired 10-membered ring 407 in good yield (68%). However, the last deprotection step was unsatisfactory and proceeded only in low yield. The solution to this problem was provided by simply inverting the reaction sequence. Thus, a facile RCM in the presence of [Ru]-I afforded the 14-membered macrocyclic lactam 409. After hydrogenation of the olefin, a Bischler–Napieralski reaction led to the 2,3-dihydroisoquinoline 410 that was reduced to 408. Intermediates (relative energy) H

OMe OPh

H

H

H RuLn More stable than

H

H OPh OPiv 399 (+ 47.3 kJ) H

OPh OPiv 398 (+34.1 kJ) MOMO H

MOMO H

OPMP

OPMP

[Ru]-II

H H

H OPiv

OPMP

400

Scheme 1.72

OMe OPh H RuLn

401 (27%)

MOMO

H

OPMP H

H OPiv

+

OPMP

PMPO 402 (15%)

OPiv

1.8 Conclusion

39

MeO NH H H

MeO O

OH

H Excentricine (403) CHO

MeO Steps

MeO HN

MeO

O

MeO

95%

OH 404

Vanillin

405

MeO Steps

MeO N

MeO

Ac

MeO

[Ru]-II CH2Cl2 reflux 68%

406

N

MeO

Steps Ac

MeO

NH

MeO

408

407 MeO NH

[Ru]-I 404

N

MeO

POCl3

MeO CH2Cl2 rt 88%

O

409

Steps

N

MeO

H2, Pd/C

410

Scheme 1.73

1.8 Conclusion

In the last decade, the alkene/alkyne metathesis reaction has become a major synthetic tool for the preparation of complex organic molecules and, not surprisingly, RCM has proved to be particularly useful for the synthesis of medium-size (5–10-membered) rings. For ring sizes of five to seven, the reaction works well but for larger (8–10-membered) rings, the metathesis is less predictable and its outcome depends on still imperfectly understood factors. Remarkable sequences, in particular [3,3]-sigmatropic rearrangement/RCM or RCM/fragmentation combinations, have been classically used as synthetic strategies.

408

40


References 1 Wang, Q., Millet, A., and Hiersemann, 2

3 4

5 6

7

8

9

10 11 12

13

14

15

16 17 18

19

M. (2007) Synlett, 1683–1686. Probst, N.P., Haudrechy, A., and Plé, K. (2008) J. Org. Chem., 73, 4338–4341 and references cited therein. Srikrishna, A. and Laksmi, B.V. (2005) Tetrahedron Lett., 46, 7029–7031. (a) Srikrishna, A. and Lakshmi, B.V. (2005) Tetrahedron Lett., 46, 4879–4881; (b) Srikrishna, A. and Lakshmi, B.V. (2005) Synlett, 1173–1175. Maity, S. and Ghosh, S. (2007) Tetrahedron Lett., 48, 3355–3358. Banerjee, S., Ghosh, S., Sinha, S., and Ghosh, S. (2005) J. Org. Chem., 70, 4199–4202. Srikrishna, A., Khan, I.A., Babu, R.R., and Sajjanshetty, A. (2007) Tetrahedron, 63, 12616–12620. Wilson, M.S., Woo, J.C.S., and Dake, G.R. (2006) J. Org. Chem., 71, 4237–4245. Chatterjee, A.K., Morgan, J.P., Scholl, M., and Grubbs, R.H. (2000) J. Am. Chem. Soc., 122, 3783–3784. Donohoe, T.J. and Rosa, C.P. (2007) Org. Lett., 9, 5509–5511. Mehta, G. and Maity, P. (2007) Tetrahedron Lett., 48, 8865–8868. Srikrishna, A., Pardeshi, V.H., and Thriveni, P. (2008) Tetrahedron: Asymmetry, 19, 1392–1396. Rodeschini, V., van de Weghe, P., Tarnus, C., and Eustache, J. (2005) Tetrahedron Lett., 46, 6691–6695. Miller, A.K., Hughes, C.C., Kennedy-Smith, J.J., Gradl, S.N., and Trauner, D. (2006) J. Am. Chem. Soc., 128, 17057–17062. Paquette, L.A., Peng, X., Yang, J., and Kang, H.-J. (2008) J. Org. Chem., 73, 4548–4558. Clive, D.L.J. and Yu, M. (2002) Chem. Commun., 2380–2381. Frigell, J. and Cumpstey, I. (2007) Tetrahedron Lett., 48, 9073–9076. Srikrishna, A., Pardeshi, V.H., and Satyanarayana, G. (2007) Tetrahedron Lett., 48, 4087–4090. Ziegler, F.E., Metcalf, C.A.III, Nangia, A., and Schulte, G. (1993) J. Am. Chem. Soc., 115, 2581–2589.

20 (a) Dai, M. and Danishefsky, S.J.

21

22

23

24

25 26

27 28

29 30

31

32

(2007) J. Am. Chem. Soc., 129, 3498–3499; (b) Dai, M., Krauss, I.J., and Danishefsky, S.J. (2008) J. Org. Chem., 73, 9576–9583. Herrmann, P., Budˇesˇ´ınský, M., and Kotora, M. (2008) J. Org. Chem., 73, 6202–6206. (a) Boiteau, J.G., van de Weghe, P., and Eustache, J. (2001) Org. Lett., 3, 2737–2740; (b) Rodeschini, V., Boiteau, J.-G., van de Weghe, P., Tarnus, C., and Eustache, J. (2004) J. Org. Chem., 69, 357–373. (a) Picoul, W., Urchegui, R., Haudrechy, A., and Langlois, Y. (1999) Tetrahedron Lett., 40, 4797–4800; (b) Picoul, W., Bedel, O., and Langlois, Y. (2003) Pure Appl. Chem., 75, 235–249; (c) Bedel, O., Haudrechy, A., and Langlois, Y. (2004) Eur. J Org. Chem., 3813–3819. Takahashi, S., Hishinuma, N., Koshino, H., and Nakata, T. (2005) J. Org. Chem., 70, 10162–10165. Fürstner, A. and Langemann, K. (1997) Synthesis, 792–803. (a) Vosburg, D.A., Weiler, S., and Sorensen, E.J. (2003) Chirality, 15, 156–166; (b) Vosburg, D.A., Weiler, S., and Sorensen, E.J. (1999) Angew. Chem. Int. Ed., 38, 971–974. Clive, D.L.J. and Liu, D. (2008) J. Org. Chem., 73, 3078–3087. Yun, S.Y., Zheng, J.-C., and Lee, D. (2008) Angew. Chem. Int. Ed., 47, 6201–6203. Tiefenbacher, K. and Mulzer, J. (2008) Angew. Chem. Int. Ed., 47, 6199–6200. (a) Nicolaou, K.C., Tria, G.S., and Edmonds, D.J. (2008) Angew. Chem. Int. Ed., 47, 1780–1783; (b) Nicolaou, K.C., Tria, G.S., and Edmonds, D.J. (2008) Angew. Chem. Int. Ed., 47, 1780–1783; (c) Hayashida, J. and Rawal, V.H. (2008) Angew. Chem. Int. Ed., 47, 4373–4376. Toyota, M., Yokota, M., and Ihara, M. (2001) J. Am. Chem. Soc., 123, 1856–1861. Ciochina, R. and Grossman, R.B. (2006) Chem. Rev., 106, 3963–3986.

References 33 Kuramochi, A., Usuda, H., Yamatsugu,

34

35 36 37

38 39 40 41

42 43 44

45

46

47

48

49

50 51

K., Kanai, M., and Shibasaki, M. (2005) J. Am. Chem. Soc, 127, 14200–14201. Kingsbury, J.S., Harrity, J.P.A., Bonitatebus, P.J., Hoveyda, A.H., Yokota, M., and Ihara, M. (1999) J. Am. Chem. Soc., 121, 791–799. Mehta, G. and Singh, V. (1999) Chem. Rev., 99, 881–930. Grob, C.A. and Schiess, P.W. (1967) Angew. Chem., 79, 1–14. (a) Rodr`ıguez, J.R., Castedo, L., and Mascare˜ nas, J.L. (2002) Chem. Eur J., 8, 2923–2930; (b) Rodr`ıguez, J.R., Castedo, L., and Mascare˜ nas, J.L. (2000) Org. Lett., 2, 3209–3212. Mehta, G. and Kumaran, R.S. (2005) Tetrahedron Lett., 46, 8831–8835. Wharton, P.S. (1961) J. Org. Chem., 26, 4781–4782. Ivkovic, A., Matovic, R., and Saicic, R.N. (2004) Org. Lett., 6, 1221–1224. Matovic, R., Ivkovic, A., Manojlovic, M., Tokic-Vujosevic, Z., and Saicic, R.N. (2006) J. Org. Chem., 71, 9411–9419. Mehta, G. and Kumaran, R.S. (2002) Chem.Commun., 1456–1457. Robichaud, J. and Tremblay, F. (2006) Org. Lett., 8, 597–600. Song, Y., Hwang, S., Gong, P., Kim, D., and Kim, S. (2008) Org. Lett., 10, 269–271. Bi, F.Q., Fang, L.J., Zhang, C.X., and Li, Y.L. (2008) Chin. Chem. Lett., 19, 1265–1267. Nicolaou, K.C., Jennings, M.P., and Dagneau, P. (2002) Chem.Commun., 2480–2481. White, D.E., Stewart, I.C., Grubbs, R.H., and Stoltz, B.M. (2008) J. Am. Chem. Soc, 130, 810–811. Tani, K., Behenna, D.C., McFadden, R.M., and Stoltz, B.M. (2007) Org. Lett., 9, 2529–2531. Stewart, I.C., Ung, T., Pletnev, A.A., Berlin, J.M., Grubbs, R.H., and Schrodi, Y. (2007) Org. Lett., 9, 1589–1592. Pragani, R. and Roush, W.R. (2008) Org. Lett., 10, 4613–4616. For a recent review about carbasugars see: Arjona, O., Gómez, A.M., López, J.C., and Plumet, J. (2007) Chem. Rev., 107, 1919–2036.

52 For reviews about olefin metathesis in

53 54 55

56

57

58

59

60 61 62

63

64 65 66

67 68 69

carbohydrate chemistry see: (a) Roy, R. and Das, S.K. (2000) Chem. Commun., 519–529; (b) Madsen, R. (2007) Eur. J Org. Chem., 399–415. Kapferer, P., Sarabia, F., and Vasella, A. (1999) Helv. Chim. Acta, 82, 645–656. Ziegler, F.E. and Wang, Y. (1998) J. Org. Chem., 63, 426–427. Chang, Y.-K., Lee, B.-Y., Kim, D.J., Lee, G.S., Jeon, H.B., and Kim, K.S. (2005) J. Org. Chem., 70, 3299–3302. Sellier, O., van de Weghe, P., Le Nouen, D., Strehler, C., and Eustache, J. (1999) Tetrahedron Lett., 40, 853–856. Ackermann, L., El Tom, D., and Fürstner, A. (2000) Tetrahedron, 56, 2195–2202. Nishikawa, A., Saito, S., Hashimoto, Y., Koga, K., and Shirai, R. (2001) Tetrahedron Lett., 42, 9195–9198. Andresen, T.L., Skytte, D.M., and Madsen, R. (2004) Org. Biomol. Chem., 2, 2951–2957. Heo, J.-N., Holson, E.B., and Roush, W.R. (2003) Org. Lett., 5, 1697–1700. Nadein, O.N. and Kornienko, A. (2004) Org. Lett., 6, 831–834. Kireev, A.S., Nadein, O.N., Agustin, V.J., Bush, N.E., Evidente, A., Manpadi, M., Ogasawara, M.A., Rastogi, S., Togelj, S., Shors, S.T., and Kornienko, A. (2006) J. Org. Chem., 71, 5694–5707. Barker, D., Brimble, M.A., McLeod, M.D., and Savage, G.P. (2004) Org. Biomol. Chem., 2, 1659–1669. Mehta, G. and Likhite, N.S. (2008) Tetrahedron Lett., 49, 7113–7116. Mehta, G. and Lakshminath, S. (2006) Tetrahedron Lett., 47, 327–330. Yokoe, H., Sasaki, H., Yoshimura, T., Shindo, M., Yoshida, M., and Shishido, K. (2007) Org. Lett., 9, 969–971. Enquist, J.A. Jr. and Stoltz, B.M. (2008) Nature, 453, 1228–1231. Schall, A. and Reiser, O. (2008) Eur. J. Org. Chem., 2353–2364. Andrews, S.P., Ball, M., Wierschem, F., Cleator, E., Oliver, S., Högenauer, K., Simic, O., Antonello, A., Hünger, U., Smith, M.D., and Ley, S.V. (2007) Chem. Eur. J., 13, 5688–5712.

41

42

1 Synthesis of Natural Products Containing Medium-size Carbocycles 70 van de Weghe, P., Bisseret, P.,

71

72

73

74

75

76

77 78

79 80 81

82

83

84

Blanchard, N., and Eustache, J. (2006) J. Organomet. Chem., 691, 5078–5108. Oliver, S.F., Högenauer, K., Simic, O., Antonello, A., Smith, M.D., and Ley, S.V. (2003) Angew. Chem. Int. Ed., 42, 5996–6000. Nickel, A., Maruyama, T., Tang, H., Murphy, P.D., Greene, B., Yusuff, N., and Wood, J.L. (2004) J. Am. Chem. Soc., 126, 16300–16301. Watanabe, K., Suzuki, Y., Aoki, K., Sakakura, A., Suenaga, K., and Kigoshi, H. (2004) J. Org. Chem., 69, 7802–7808. Winkler, J.D., Rouse, M.B., Greaney, M.F., Harrison, S.J., and Jeon, Y.T. (2002) J. Am. Chem. Soc., 124, 9726–9728. Benningshof, J.C.J., Blaauw, R.H., van Ginkel, A.E., Rutjes, F.P.J.T., Fraanje, J., Goubitz, K., Schenk, H., and Hiemstra, H. (2000) Chem. Commun., 1465–1466. Pfeiffer, M.W.B. and Phillips, A.J. (2005) J. Am. Chem. Soc., 127, 5334–5335. Pfeiffer, M.W.B. and Phillips, A.J. (2008) Tetrahedron Lett., 49, 6860–6861. Michalak, K., Michalak, M., and Wicha, J. (2008) Tetrahedron Lett., 49, 6807–6809. Dauben, W.G. and Michno, D.M. (1977) J. Org. Chem., 42, 682–685. Kirkland, T.A. and Grubbs, R.H. (1997) J. Org. Chem., 62, 7319–7318. Nakashima, K., Inoue, K., Sono, M., and Tori, M. (2002) J. Org. Chem., 67, 6034–6040. Nakashima, K., Fujisaki, N., Inoue, K., Minami, A., Nagaya, C., Sono, M., and Tori, M. (2006) Bull. Chem. Soc. Jpn., 79, 1955–1962. Kalidindi, S., Jeong, W.B., Schall, A., Bandichhor, R., Nosse, B., and Reiser, O. (2007) Angew. Chem. Int. Ed., 46, 6361–6363. (a) Miller, S.J., Kim, S.-H., Chen, Z.-R., and Grubbs, R.H. (1995) J. Am. Chem. Soc., 117, 2108–2109; (b) Galli, C. and Mandolini, L. (2000) Eur. J. Org. Chem., 3117–3125; (c) Illuminati, G. and Mandolini, L. (1981) Acc. Chem. Res., 14, 95–102; (d) Mitchell, L.,

85 86

87

88 89

90

91

92

93 94 95

96

97 98

99

Parkinson, J.A., Percy, J.M., and Singh, K. (2008) J. Org. Chem., 73, 2389–2395. Maier, M.E. (2000) Angew. Chem. Int. Ed., 39, 2073–2077. (a) Wenz, M., Groβbach, D., Beitzel, M., and Blechert, S. (1999) Synthesis, 607–614; (b) Krafft, M.E., Cheung, Y.Y., and Abboud, K.A. (2001) J. Org. Chem., 66, 7443–7448; (c) Paquette, L.A. and Efremov, I. (2001) J. Am. Chem. Soc., 123, 4492–4501. (a) Paquette, L.A., Tae, J., Arrington, M.P., and Sadoun, A.H. (2000) J. Am. Chem. Soc., 122, 2742–2748; (b) Fürstner, A. and Langemann, K. (1996) J. Org. Chem., 61, 8746–8749. Michaut, A. and Rodriguez, J. (2006) Angew. Chem. Int. Ed., 45, 5740–5750. Ruprah, P.K., Cros, J.-P., Pease, J.E., Whittingham, W.G., and Williams, J.M.J. (2002) Eur. J. Org. Chem., 3145–3152. Michalak, K., Michalak, M., and Wicha, J. (2005) Tetrahedron Lett., 46, 1149–1153. Takao, K.-I., Watanabe, G., Yasui, H., and Tadano, K.-I. (2002) Org. Lett., 4, 2941–2943. Uchida, I., Ando, T., Fukami, N., Yoshida, K., Hashimoto, M., Tada, T., Koda, S., and Morimoto, Y. (1987) J. Org. Chem., 52, 5292–5293. Morency, L. and Barriault, L. (2005) J. Org. Chem., 70, 8841–8853. Paquette, L.A., Efremov, I., and Liu, Z. (2005) J. Org. Chem., 70, 505–509. (a) Ma, C., Schiltz, S., Le Goff, X.F., and Prunet, J. (2008) Chem. Eur. J., 14, 7314–7323; (b) Schiltz, S., Ma, C., Ricard, L., and Prunet, J. (2006) J. Organomet. Chem., 691, 5438–5443. Paquette, L.A., Dong, S., and Parker, G.D. (2007) J. Org. Chem., 72, 7135–7147. Clark, J.S., Marlin, F., Nay, B., and Wilson, C. (2003) Org. Lett., 5, 89–92. Clark, J.S., Northall, J.M., Marlin, F., Nay, B., Wilson, C., Blake, A.J., and Waring, M.J. (2008) Org. Biomol. Chem., 6, 4012–4025. (a) Nevalainen, M. and Koskinen, A.M.P. (2001) Angew. Chem. Int. Ed., 40, 4060–4062; (b) Nevalainen, M. and

References

100

101

102 103

Koskinen, A.M.P. (2002) J. Org. Chem., 67, 1554–1560. Nakamura, T., Oshida, M., Nomuira, T., Nakazaki, A., and Kobayashi, S. (2007) Org. Lett., 9, 5533–5536. Hong, S.H., Sanders, D.P., Lee, C.W., and Grubbs, R.H. (2005) J. Am. Chem. Soc., 127, 17160–17161. Li, Y. and Hale, K.J. (2007) Org. Lett., 9, 1267–1270. Larrosa, I., Da Silva, M.I., Gómez, P.M., Hannen, P., Ko, E., Lenger, S.R., Linke, S.R., White, A.J.P., Wilton, D., and Barrett, A.G.M. (2006) J. Am. Chem. Soc., 128, 14042–14043.

104 Castoldi, D., Caggiano, L., Bayón, P.,

105

106

107

108

Costa, A.M., Cappella, P., Sharon, O., and Gennari, C. (2005) Tetrahedron, 61, 2123–2139. Castoldi, D., Caggiano, L., Panigada, L., Sharon, O., Costa, A.M., and Gennari, C. (2005) Angew. Chem., 117, 594–597. Castoldi, D., Caggiano, L., Panigada, L., Sharon, O., Costa, A.M., and Gennari, C. (2006) Chem. Eur. J., 124, 51–62. Bourgeois, D., Mahuteau, J., Pancrazi, A., Nolan, S.P., and Prunet, J. (2000) Synthesis, 869–882. Larghi, E.L. and Kaufman, T.S. (2008) Tetrahedron, 64, 9921–9927.

43

45

2 Natural Products Containing Medium-sized Nitrogen Heterocycles Synthesized by Ring-closing Alkene Metathesis Sebastiaan (Bas) A. M. W. van den Broek, Silvie A. Meeuwissen, Floris L. van Delft, and Floris P. J. T. Rutjes

2.1 Introduction

The formation of carbon–carbon bonds is a crucial issue in organic synthesis. Out of all methodologies available, the metathesis reaction has, over the past decade, evolved as one of the most useful tools in this field to rapidly construct the skeleton of carba- and heterocyclic molecules [1]. Key step of the metathesis reaction entails the metal-catalyzed redistribution of two carbon–carbon double bonds by a scission–recombination process [2]. Among various applications of the metathesis reaction, ring-closing metathesis (RCM) has emerged as one of the most powerful tools for the construction of natural and unnatural cyclic compounds (Scheme 2.1) [3]. Other types of metathesis reactions developed so far include cross-metathesis (CM), ring-opening metathesis (ROM), ring rearrangement metathesis (RRM, combining ROM with RCM in a tandem sequence), and ene–yne metathesis [3]. The metathesis reaction has already been used for several decades in polymer chemistry for ring-opening metathesis polymerization (ROMP) [4]. Only since the development of well-defined molybdenum and ruthenium carbene complexes by Schrock [5] and Grubbs [6] in 1990 and 1992, respectively, an explosion of their application in organic synthesis has been witnessed. Initially, it was demonstrated that the Schrock catalyst [Mo]-I could serve as a homogeneous catalyst for olefin RCM [7]. Unfortunately, metal carbene complex [Mo]-I suffers from extreme sensitivity to air and moisture, which precluded widespread use among organic chemists. Not long thereafter, the more stable yet less reactive benzylidene ruthenium complex [Ru]-IV was developed [8]. In 1995, additional research in this field led to the ruthenium carbene catalyst [Ru]-I [9], which is currently commercially available as the Grubbs-I or first-generation Grubbs catalyst. This catalyst still appeared to be somewhat less reactive than catalyst [Mo]-I in olefin metathesis, but yet retained the remarkable air and water stability characteristics of ruthenium complex [Ru]-IV. In 1999, the effectiveness of the ruthenium-based metathesis catalysts was even further enhanced by replacing one of the phosphine ligands with a heterocyclic carbene, the 1,3-bis(2,4,6-trimethylphenyl)-2-imidazolidinylidene (IMesH2 ) ligand, to give Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts. Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32440-8

46

2 Medium-sized Nitrogen Heterocycles Synthesized by Ring-closing Alkene Metathesis

RCM

Scheme 2.1

+

Schematic illustration of the ring-closing metathesis reaction.

catalyst [Ru]-II [10]. This carbene complex exhibits high olefin metathesis activity in RCM reactions due to, amongst others, a higher thermal stability and wider functional group tolerance, and is nowadays commercially available as the Grubbs-II or second-generation Grubbs catalyst. Many (commercially available) modifications of these catalysts exist [11]; in particular, ones in which the tricyclohexylphosphine and benzylidene groups have been replaced by an ortho-isopropoxy-substituted benzylidene, the so-called Grubbs–Hoveyda catalysts [Ru]-III and [Ru]-IX [12]. Formation of heterocycles by RCM was first reported by Grubbs and Fu, who found that a di-N-allylated trifluoroacetamide underwent cyclization in the presence of Schrock’s catalyst [Mo]-I to give the corresponding 3-pyrroline [13] (Figure 2.1). At that time it was shown that the ruthenium–alkylidene complex [Ru]-IV gave also excellent yields for similar heterocyclic RCM reactions, even when performed in undistilled solvents and in the presence of air. Although it was well established that free amines are generally incompatible with the metathesis reaction [14], later work demonstrated that this problem could be overcome by the protection of the basic nitrogen atom as an amide or carbamate. In addition, protonation of the amine to give the corresponding ammonium salts and thus avoiding protecting groups is tolerated by [Ru]-I and [Ru]-II.

PCy3 Ru Cl PCy3 Ph

Cl

Mes N

iPr

O Me F3C

Mo

O Me CF3 [Mo]-I

PCy3 Ru Cl PCy3

[Ru]-II

Mes N Cl

N F3C F3C

Ru PCy3 Ph

Cl

[Ru]-I

Pri

N Mes

Me Ph Me

Cl Me

Ph

Cl

Cl

N Mes Ru O

[Ru]-IV

Cl Cl

PCy3 Ru O

Me

Me Me [Ru]-III

[Ru]-XI

Figure 2.1 Metal–alkylidene complexes as metathesis catalysts: Schrock’s catalyst [Mo]-I, the first benzylidene ruthenium complex [Ru]-IV, Grubbs-I catalyst (Cl2 (Cy3 P)2 Ru=CHPh) [Ru]-I, Grubbs-II catalyst (Cl2 (IMesH2 )(Cy3 P)Ru=CHPh) [Ru]-II, Grubbs–Hoveyda-I catalyst [Ru]-IX, Grubbs–Hoveyda-II catalyst [Ru]-III.

Ph

2.2 Five-membered Nitrogen Heterocycles

As a result of the compatibility of the Ru catalysts with nitrogen-based functional groups, RCM became a widely applied synthetic tool in the total synthesis of nitrogen-containing natural products. In particular, the formation of medium-sized (five- to eight-membered) heterocyclic systems, which frequently occurs in naturally occurring alkaloids, became a key target for metathesis-mediated synthesis. Hence, the aim of this account is to provide an overview of the majority of syntheses of alkaloid natural products in which a medium-sized nitrogen-containing ring was constructed using RCM. The examples are restricted to pathways that really have led to a natural product and cover the literature until August 2008.

2.2 Five-membered Nitrogen Heterocycles 2.2.1 Dihydropyrroles

Interesting candidates for the application of RCM are the linear marine peptides (aeruginosins) that share a common 1-aza[4.3.0]-bicyclic core unit. Hanessian et al. recently described the total syntheses of dysinosin A (1) [15] that was isolated from a new genus of sponge of the family Dysideidae and oscillarin (2) [16] that was found in the algal cultures of Oscillatoria agardhii. Both alkaloids were constructed via two peptide couplings between three common subunits: an indolizidine carboxylic acid (4 or 6) derived from l-glutamic acid, an acyclic peptide chain (5 or 7) and pyrrolino-ethylamine 3 (Scheme 2.2). The synthesis of the -3 pyrroline unit 3 is shown in Scheme 2.3 and started with readily available butyrolactone. Six steps involving ring opening to the hydroxy ester and substitution of the reduced ester with allylamine were required to form the metathesis precursor 8. Cyclization of N-Boc-protected 8 into pyrroline 9 employing [Ru]-I (10 mol%) went smoothly in a high yield of 90%. The six-membered carbocycle of dysinosin A fragment dihydroxyindolizidine 4 was prepared via RCM as well (99% yield) using the same catalyst (1 mol%) as for 8. 2.2.2 Pyrrolidine Alkaloids 2.2.2.1 Pyrrolidines (−)-(S)-Nicotine (13) is one of the many piperidine and pyrrolidine alkaloids isolated from the leaves of Nicotaiana tabacum. Welter and coworkers developed a synthesis route where 13 was obtained in only four steps from pyridinylallyl carbonate 10 involving RCM (Scheme 2.4) [17]. The synthesis commenced with an Ir-catalyzed asymmetric allylic amination of 10 to produce bis-alkene 11 in excellent enantiomeric purity when using phosphoramidate ligand L1. Subsequently, secondary amine 11 was N-protected to produce 12 prior to RCM in order to prevent catalyst

47

48

2 Medium-sized Nitrogen Heterocycles Synthesized by Ring-closing Alkene Metathesis H

HO HO

H

O Amide NH coupling

HN

BocN +

i

Pr

N NHBoc

3

N H H

+

HN

Ph Oscillarin (2)

+

CO2H O

CO2H N H H

HO

MeO OH

4

5

L-Glutamic acid

D-Leucine and D-Mannitol

HN

CO2H

Ph

H

H CO2Me

NH2

HO

H2N

OSO3 Dysinonin A (1)

O O

N

NH2

MeO

HO

H Ph

N

HO

N

HO

RCM O O

HN

Amide coupling

NH

H 2N

N

i Pr H

O

+

O

HN HO

Ph

6

7

L-Glutamic acid

D-Phenyllactic acid and D-Phenylalanine

Scheme 2.2 Retrosynthesis of the natural products dysinosin A (1) and oscillarin (2) sharing the RCM product 3.

deactivation. Best RCM results were achieved with [Ru]-II yielding the corresponding dihydropyrrole derivative in >90% yield. The synthesis of (S)-nicotine was completed through a deprotection step and finally reduction of the double bond. Over the past few decades, the biosyntheses of tropane alkaloids such as (−)-hyoscyamine, (−)-scopolamine, and (−)-cocaine have been studied extensively [18]. The skeletons of all these alkaloids feature the tropinone moiety, produced by oxidation and subsequent cyclization of (+)-hygrine (17) [19]. It was envisioned that RCM could be invoked in order to develop an enantioselective synthesis of 17 [20]. As depicted in Scheme 2.5, a phase-transfer catalytic allylation reaction with methallyl bromide converted t-butyl ester 13 into product 14 with 97% ee. Metathesis precursor 15 was then prepared in six more steps, including transformation of the t-butyl ester of 14 into a terminal olefin. Dihydropyrrole 17 was formed in quantitative yield by RCM employing [Ru]-II (8 mol%). Subsequent hydrogenation and deprotection of 16 led to (+)-hygrine 17 with overall yield of 29% and an ee of 97%. H2N 6 steps Butyrolactone

RO

[Ru]-I (10 mol%) N Boc 8: R = TBDPS

Scheme 2.3

CH2Cl2, rt 90%

RO

3 steps N Boc

9: R = TBDPS

N NHBoc

BocN 3


MeO2CO

R 2 steps

N

1. 1 M HCl Et2O, rt 2. [Ru]-I (5 mol%) CH2Cl2, reflux

N

N

3. TsNHNH2, NaOAc N DMF, 96 °C 4. LiAlH4 (S )-Nicotine (12) L1 THF, rt

N 10

Me

11

49

C6H4-o -OMe Me O P N Me O C6H4-o -OMe

Scheme 2.4

2.2.2.2 Dipyrrolidines A total synthesis of the dipyrrolidine alkaloid (−)-trans-dendrochrysine (22), isolated from the orchid endemic Dendrobium chrysanthum, was achieved by Blechert and coworkers utilizing RRM [21]. The asymmetric synthesis commenced with preparation of diolefin 20 from commercially available tropone (19, Scheme 2.6) in 11 steps involving a solid-phase supported Candida antarctica lipase B (Cal-B)-mediated desymmetrization. Different ruthenium catalysts were evaluated for the RRM process and it appeared that [Ru]-III (5 mol%) in refluxing toluene under an ethylene atmosphere [22] gave the best results providing bicycle 21 in 91% yield. Thus, (−)-trans-dendrochrysine (22) was synthesized with a yield of 5.9% over 18 linear steps. 2.2.2.3 Polyhydroxypyrrolidines Naturally occurring iminosugars (azasugars) have been prominent target molecules for the utilization of RCM, involving formation of a cyclic olefin, followed by dihydroxylation. The mulberry tree (Morus alba) alkaloid 1,4-dideoxy-1,4-imino-d-ribitol [(+)-DRB, 27] is a potent inhibitor of glucosidases and of eukaryotic DNA polymerases. The Riera synthesis of polyhydroxypyrrolidine 27 featured a five-membered ring formation via RCM as illustrated in Scheme 2.7 [23]. Divinylcarbinol 23 was converted in a number of steps into the enantiomerically pure RCM precursor 25 (ee = 99%) via a Sharpless epoxidation and subsequent Payne rearrangement as the key steps. RCM under the influence of the [Ru]-II catalyst (8 mol%) proceeded uneventfully, despite the two unprotected alcohols, to give intermediate 26 in a surprisingly high yield of 90%. Synthesis of (+)-DRB (27) t-BuO2C

Me

t -BuO2C

CH2Br

N

N

X (1 mol%), KOH Ph Ph PhMe·CHCl3 Ph −20 °C, 95% 13 Et Br

O N L2

Scheme 2.5

+

Br

O

14

N

[Ru]-II (8 mol%) rt

Me

CH2Cl2 99%

15

3 steps

−

O

N CO2Et

Ph

Et

−

N+

6 steps Me

O

N

Ref. 18

O (+)-Hygrine (17)

Me

Me

N Me HCl

O

O

16 N

Ref. 19

O N CO2Et

Tropinone (18)

(−)-Hyoscyamide (−)-Scopolamine (−)-Cocaine

50

2 Medium-sized Nitrogen Heterocycles Synthesized by Ring-closing Alkene Metathesis O

OTBS

[Ru]-III (5 mol%) H2C = CH2

11 steps N

N

Ns

Cbz

19

Toluene, reflux 91%

OTBS N

N

Ns

Cbz

20

21 6 steps

O N

N Me

O

(−)-trans -Dendrochrysine (22)

Scheme 2.6

was completed in several steps including a side chain–directed syn-dihydroxylation with OsO4 of the ring olefin. A similar metathesis strategy was utilized by Trost et al. for the total syntheses of the azasugars 2,5-dideoxy-2,5-imino-d-mannitol [(+)-DMDP, 28], (−)-bulgecinine (29), and the somewhat more complex (+)-broussonetine G (30) [24, 25]. (+)-DMDP (28) can be isolated from several sources, but was initially discovered in the leaves of Derris eliptica. (−)-Bulgecinine was found in bacterial cultures of Pseudomonas acidophila and Pseudomonas mesoacidophila and the broussonetine family including broussonetine G in branches of the deciduous tree Broussontia kazinoki. All three natural products were obtained from the same metathesis product 31 (Scheme 2.8),

HO

23

1. (−)-diisopropyl tartrate cumene hydroperoxide O CH2Cl2, −35 °C, 2. 0.5 N NaOH 84% ee = 99%

OH 2 steps

24

Boc N

25

OH OH

[Ru]-I (8 mol%) reflux

Boc OH N OH

CH2Cl2 90%

26 4 steps

H N HO

OH

OH

DRB (27)

Scheme 2.7


HO

H N

OH

O O

BnO

OH

H N

HO2C

51

N HO

OH

OH

DMDP (28)

OH

OH

H N

HO

RCM

31

(−)-Bulgecinine (29)

Grignard BnO

Cbz O N

O O

OH

(+)-Broussonetine G (30)

Br Me N OMe

+

O O

32

33

Scheme 2.8 Natural products (+)-DMDP (28), (−)-bulgecinine (29), and (+)-broussonetine G (30) sharing the same RCM-made building block 31.

from which broussonetine G was derived via a Grignard addition of 33 onto 32. Key steps in the synthesis of the central intermediate 31 were a palladium-catalyzed dynamic kinetic asymmetric transformation (DYKAT) and RCM. The synthesis of 31 started with a double dynamic asymmetric allylic amination for the enantioselective construction of diallylamine 36 from allylic epoxide (34, Scheme 2.9) involving ligand (R,R)-L3. While RCM with a free hydroxyl (36) appeared troublesome, the ring closure proceeded smoothly in high yield using [Ru]-II (1.2 mol%) after benzyl protection (36), which presumably prevents undesired coordination to ruthenium. Finally, seven steps including a Grignard reaction between alkyl bromide 33 and the corresponding Weinreb amide of 32 were required to afford broussonetine G (30) in 8.2% overall yield. HO O

Pd2dba3·CHCl3 (R,R )-L3 DBU (1 mol%), rt

34 O +

HO O RO

O

[Ru]-II (0.8 and 0.4 mol%) BnO reflux

N O

HN

CH2Cl2 90% ee = 93%

OH

H N

3 steps

O O

5 steps

N

O NH PPh2

OH

H N

HO2C

CH2Cl2 85% 36: R = H 37: R = Bn

OH

DMDP (28)

NH PPh2 O

OH (−)-Bulgecinine (29) L3

31

35 10 steps

OH

H N

OH

O HO

OH

O

(+)-Broussonetine G (30)

Scheme 2.9

52


2.2.3 Indolizidine Alkaloids 2.2.3.1 Polycyclic Indolizidines Alkaloids with polycyclic skeletal frameworks are, when it comes to their synthesis, excellent candidates for RCM. Illustrative are the indolizidines rhynchophylline (43) and its C(7)-epimer iso-rhynchophylline (44), both isolated from the plant Uncaria rhynchophylla (Rubiaceae) [26]. Deiters’ total synthesis of 43 and 44 started with the efficient construction of diallylamine 39 via amide formation between indole-3-acetic acid (38) and diallylamine (Scheme 2.10). One-pot RCM–carbomagnesation of 39 was smoothly achieved with only 1 mol% of [Ru]-I catalyst and 4 equiv. of EtMgCl to afford the 2-ethyl-3-butene-amine derivative in 71% yield. It appeared that the electron-withdrawing carbonyl moiety was critical to the success of the RCM–carbomagnesation steps. Amide reduction and subsequent treatment with acryloyl chloride delivered the second metathesis precursor 40. Cyclization with [Ru]-I (5 mol%) then furnished the α,β-unsaturated lactam 41 in a high yield of 91%. Continuation of the total synthesis of alkaloids 43 and 44 included a Bischler–Napieralski cyclization (42) and subsequent rearrangement into the oxindole framework. The erythrina alkaloids form a widely distributed family that lately has received considerable attention as a result of their unique tetracyclic skeleton and biological activities. Erythrocarine (46) [27, 28], erythravine (50) [29], and (+)-β-erythroidine (59) [30] are three members of this family that have been synthesized with metathesis as a key step. The molecular structures of erythrocarine, found in the seeds of Erythrina caribaea, and erythravine, isolated from Erythrina cochleata, are related. As a result, the total syntheses independently proposed by, respectively, Mori et al. and Hatakeyama et al. proceeded via the comparable intermediates 45 and 49, respectively, as shown in Scheme 2.11. Ene–yne 45 was synthesized in a number of steps. This substrate was subjected to HCl, followed by [Ru]-I catalyst (10 mol%) to induce a tandem enyne-RCM process to yield the acetylated natural product in a quantitative conversion as a 1 : 1 mixture of diastereoisomers. After

CO2H

Diallylamine EDC, rt

O N

CH2Cl2 88%

N H

1. [Ru]-I (1 mol%) THF, rt 2. EtMgBr, Cp2ZrCl2 THF, rt 3. i: LiAlH4, Et2O, rt ii: acryloyl chloride Et3N, rt

N H

38

39

N N H Et 41

Scheme 2.10

O

45% (3 steps)

3 steps

EtO2C 42

O

[Ru]-I (5 mol%) rt CH2Cl2 91%

N H Et 40

N

5 steps

N N H H

N

Et H

OMe

Et O N H

CO2Me

Rhynchophylline (43)

2.2 Five-membered Nitrogen Heterocycles NHBoc CHO

O O

Br

6 steps

O

Refs. 27, 28

O

SiMe3

CO2Me 45: R = Me

44

1. HCl, Et2O, rt 2. [Ru]-I (10 mol%) CH2Cl2, rt

O N

O

8 steps

RO

MeO 1. [Ru]-I (10 mol%) CH2Cl2, reflux

N

RO

3. K2CO3, MeOH, rt

N

MeO

2. K2CO3, MeOH, rt

HO

AcO 93% (3 steps)

HO 73% (2 steps)

49: R = CH2

Erythrocarine (46)

Erythravine (50) 8 steps

MeO 3 steps

N CO2Et CO2Et

MeO

Ref. 29

MeO NH2

MeO

48

47

Scheme 2.11

separation of the two isomers, acetyl cleavage provided erythocarine (46) in pure form. Intermediate 49 was prepared from 48 in a large number of steps. The crucial tandem ene–yne-RCM process was also performed with [Ru]-I catalyst (10 mol%), which afforded the tetracyclic product in 78% in a diastereomeric ratio of α-OAc-50: β-OAc-50 as 63 : 37. Once more, separation of the products, followed by deacetylation of α-OAc-50 afforded erythravine (50) in an overall yield of 7.0%. The efficient approach to erythravine paved the way for a second erythrina alkaloid [(+)-β-erythroidine (56)] synthesis by Hatakeyama et al. This alkaloid was found in several species of the Erythrina genus and was synthesized in 26 linear steps as depicted in Scheme 2.12. Starting from 54, tandem ene–yne RCM of the isomeric mixture 55 underwent cyclization with [Ru]-I catalyst (10 mol%) to OH

O OH

10 steps

Me O

3 steps

CCl3 6 steps

N

HO Me O MeO

OH 51 MeO2C MeO2C

53 O

O N

6 steps

N

O

54

[Ru]-I (10 mol%)

55

N

O CH2Cl2, rt 42%

MeO

MeO

Scheme 2.12

MeO 52

MeO (+)-b -Erythroidine (56)

53

54


OMe OMe

OMe MeO

5 steps

for 59:

OMe Br

MeO

R H

4 steps for 59

CO2H

R

5 steps for 60

57 +

MeO

p -Anisaldehyde

2N

MeO 58

1

2

59: R = H, R = CHPh2 1 2 60: R = Me, R = Cbz

1 1. [Ru]-I (5 mol%), CH2Cl2, rt

2. H2, 10% Pd/C, MeOH, rt 3. HCHO, HCl, EtOH, reflux

OMe MeO

for 60: 1. [Ru]-II (2 mol%), CH2Cl2, rt 2. H2, 10% Pd/C, MeOH, rt MeO 3. HCHO, HCl, EtOH, reflux 48% (from 59) 49% (from 60)

H N

(−)-Antofine (61)

Scheme 2.13

furnish (+)-β-erythroidine (56) in a moderate yield together with a small amount of a mono-ring-closed by-product. (−)-Antofine 61 is a member of the small group of phenanthroindolizidine alkaloids and was initially found in Vincetoxicum nigrum, belonging to the Asclepiadaeceae family. Recently, Kim et al. described two routes to enantiopure (−)-antofine [31, 32]. Both routes began with the coupling of homoveratric acid (57) and p-anisaldehyde in order to form phenanthryl alcohol, the precursor of bromide 58 (Scheme 2.13). In the first article, the synthesis was continued via enantioselective catalytic phase-transfer alkylation of a protected glycine derivative with 58. The glycine fragment was further transformed into the diallylamine of 59 in three steps. In the more recent article, enantiopure 60 was obtained via a Stille coupling with an enantiopure fragment and subsequent Overman rearrangement. Both routes converged to 61 after an RCM step that was high yielding in both cases, followed by hydrogenation and Pictet–Spengler cyclization. Interestingly, RCM of diolefin 60 took place 10 times faster and was somewhat higher yielding in comparison with 59, probably as a result of different catalysts. Overall, it can be concluded that the second approach to (−)-antofine was, though one step longer, more efficient and higher yielding. The tricyclic tetraponerines form an unusual class of alkaloids representing the major constituents of contact poison of the New Guinean ant Tetraponera sp. A flexible synthesis of tetraponerines T1–T8 is challenging as they differ in side chain, stereochemistry at C-9, and size of one of the rings. Stragies et al. proposed a combined ROM–RCM strategy that led to four of these alkaloids [33]. One of them was tetraponerine T6 (65), synthesized in only six steps. The synthesis commenced with a palladium-catalyzed domino allylic alkylation on dicarbonate 62 to deliver the disulfonamide 63. [Ru]-I-induced (5 mol%) ROM–RCM yielded an equilibrium between precursor 63 and product 64 in a 1 : 10 ratio. Subsequent elaboration of the side chain, followed by cyclization, finally provided tetraponerine T6 (65), in a good 30% overall yield (Scheme 2.14). One of the simple representatives of the indolizidine series isolated as a trace compound from the skin of the Neotropical frog Dendrobates speciosus is indolizidine 167B (70). Although neither the relative nor the absolute configuration of 70 was known, Blechert and coworkers synthesized the alkaloid in accordance with the literature data via RRM (Scheme 2.15) [34]. The two olefinic side chains of product 67 required for RRM process were introduced in four steps from cycloheptenediol-monoacetate (66). RRM under the influence of [Ru]-I (5 mol%)


MeO2CO

Ns Pd2dba3 (1.5 mol%)

OCO2Me

CH(OEt)2 ( )3 N [Ru]-I (5 mol%) H2C = CH2, rt

(S,S )-L4, −60 °C

CH2Cl2

Ns N 79%

Ns

N

CH(OEt)2 ( )3 4 steps

H

O

n -Bu N

N H Ns

NH PPh2

N

H

NH PPh2 O

89%

62

63

64

L4

Tetraponerine T6 (65)

Scheme 2.14

was in situ followed by TBAF-mediated silyl ether cleavage into alcohol 69 in an outstanding 92% yield. Final ring closure via reductive amination and hydrogenation delivered natural product 70 in a stereocontrolled way in an overall yield of 35%. Another member of the indolizidine family derived from the frog genera Dendrobatid is (−)-indolizidine 223A (73). Davis et al. discovered that an intramolecular Mannich reaction combined with metathesis provided a high yield and enantioselective solution for its synthesis [35]. Reaction of 4-heptanone with enantiopure sulfinimine 71, followed by treatment with crotonaldehyde, gave the tetrasubstituted piperidinone 72 via an intramolecular Mannich reaction (Scheme 2.16). Subsequent RCM with [Ru]-I (5 mol%), followed by hydrogenation furnished as the key steps, afforded (−)-indolizidine 223A 73 in an overall yield of 9.3%. 2.2.3.2 Polyhydroxyindolizidines Despite the availability of various existing synthetic procedures, application of RCM in the synthesis of polyhydroxyindolizidines recently became increasingly popular. The skeletons of all alkaloids described below were produced according to the same strategy involving N-alkenylation and RCM for the formation of five- and six-membered nitrogen heterocycles (Scheme 2.17). The total synthesis of lengtinosine, isolated from Australahus lentiginosus, was performed by the group of Spino. Uniflorine A, recently discovered in the tree Eugenia uniflora L., and initially proposed as structure 75, was independently synthesized by various groups including those of White, Pyne, and Dhavale. However, NMR data of 75 did not correspond with the data of natural uniflorine A.

HO

OAc

Me [Ru]-I (5 mol%) Si O Me CH2Cl2, reflux

4 steps N Cbz

O N Cbz

67

66

OH

TBAF (1 M in THF) THF, 0 °C to rt

2 steps N

N Cbz

n Pr

92% (2 steps) 69

Scheme 2.15

H

Indolizidine 167B (70)

55

68

Si

Me Me

56


4-Heptanone 2 steps

+

O

O

n Pr

N

S

1. [Ru]-I (5 mol%) CH2Cl2, reflux

Me

Et

Et

p-Tolyl

H

2. H2, 10% Pd/C, MeOH, rt N

3. NaBH4, MeOH, −78 °C to rt 4. PhOC(S)Cl pyridine, 0 °C to rt 5. Bu3SnH, AIBN, PhH, reflux

n Pr 72

71

Et

N

Et

n Pr (−)-Indolizidine 223A (73)

26% (5 steps)

Scheme 2.16

Synthesis of (+)-lentiginosine’s metathesis precursor 80 commenced with the stereoselective transformation of commercially available l-menthone into compound 79 (Scheme 2.18). RCM precursor 80 was efficiently cyclized with [Ru]-I (5 mol%) to dihydropyrrolidine 76, and the l-menthone auxiliary was recovered. From there, lentiginosine 74 was eventually synthesized in five steps. In 2004, White et al. were the first to report a synthesis for putative uniflorine 75 in which RCM was used as the key step (Scheme 2.19). A three-component boronic acid–Mannich reaction (Petasis reaction) of l-xylose, allylamine, and (E)-styrene boronic acid provided the amino-tetraol 81 as a single diastereoisomer in good yield. Application of [Ru]-I (10 mol%) on N-Boc- and O-trityl-protected 81 afforded metathesis product 77 in 86% yield. The route to tetrahydroxyindolizidine 75 was completed by catalytic syn-dihydroxylation of 74 and six-membered ring formation. Further investigations involving five-membered RCM in the quest to find the correct structure of uniflorine A provided 2-epi-75 and 1,2-diepi-75 [36] and 8a-epi-75 and 2,8a-diepi-75 [37]. Yet, all these putative structures delivered mismatching NMR dataas well. In 2008, White et al. suggested, therefore, that uniflorine A is 1,2,6,7-tetrahydroxy-3-hydroxymethylpyrrolizidine [36].

H N

HO

OH Ref. 36

OH

(+)-Lentiginosine (74)

RCM R2

OR1

R3

R4

R5

HO

Ref. 37

H

HO

H

OH

N

OH

Putative Uniflorine A (75)

N

76: R1 = TBDPS, R2 = 77: R1 = Tr, R2 = R3 = R3 = R4 = H, R5 = Boc R4 = OH, R5 = Boc Me Ph O

i Pr

Scheme 2.17 Retrosyntheses of (+)-lentiginosine (74) and the putative structure of uniflorine A (75) proceed via similar RCM precursors (76 for 74, 77 for 75).

L-Xylose

B(OH)2 NH2

2.2 Five-membered Nitrogen Heterocycles RO

Me

PPh3, DEAD, HN3

Me

RO

57

RO

Me 2 steps

PhH, 0 °C

i Pr

98%

OH

i Pr

78: R = TBDPS

i Pr

N3

H

5 steps

OTBDPS N Boc

CH2Cl2, reflux

H

OH OH

N

(+)-Lentiginosine (74)

76

100%

Boc

80: R = TBDPS

79: R = TBDPS

[Ru]-I (5 mol%)

N

Scheme 2.18

Next, Pyne et al. successfully utilized the synthetic methodology created for putative uniflorine 75 to construct the analogous castanospermine (84, Scheme 2.19) [38]. In this case, amino-tetraol 81 had to be protected with three different protecting groups resulting in compound 82. Treatment of 82 with [Ru]-II (10 mol%) yielded pyrrolo-oxazolone 83 in 88% yield. Diastereoselective syn-dihydroxylation of the olefin, followed by several steps, led then to castanospermine (84). Trihydroxyindolizidine (−)-swainsonine 89 was first isolated from the fungus Rhizoctonia leguminicolain and has since then attracted great attention from chemists and biologists. In 2002, both Blechert et al. [39] and Pyne et al. [40] reported a metathesis-based approach to alkaloid 89 via the creation of a five-membered heterocycle. The route by Blechert et al. began with the enantioselective formation of oxazolidinone 86 via palladium-catalyzed desymmetrization of the meso-bis-carbamate derivative of diol 85 (Scheme 2.20). From there, the metathesis precursor 84 was delivered in excellent yield. Tandem RRM employing [Ru]-I catalyst (5 mol%) under an ethylene atmosphere converted compound 87 into the desired 3-pyrroline 88 in yields up to 98%, while the stereochemical integrity of the two chiral centers was completely preserved during the reaction. Next, elaboration 1.(Boc)2O, Et2N MeCN, DMF, 0 °C to rt 2. TrCl, pyridine, rt 3. [Ru]-I (10 mol%) CH2Cl2, reflux OTr HO HO HO

H

HN

Ph

N Boc 77 32% (3 steps)

3 steps TrO

OH 81

OH H

Ref. 40

O

N

O 82

Scheme 2.19

HO HO

H OH OH

N

Putative Uniflorine A (75) TrO

BnO HH

BnO

HO 4 steps

[Ru]-II (10 mol%) BnO reflux Ph CH2Cl2 88%

OBn HH O O 83

N

HO 6 steps

HO HO

H OH N

Castanospermine (84)

58

HO


Pd2dba3 (2.5 mol%) (S,S )-L4, TsNC

OH 85

2 steps O N

Et3N, THF –50 °C to rt

O

Ts

Ts

N

CH2Cl2 89%

87: R = TBDPS

86

92%, 97% ee

RO

[Ru]-I (5 mol%) H2C=CH2

RO

Ts

H N

88: R = TBDPS 4 steps

HO

H

OH OH

N

(−)-Swainsonine (89)

Scheme 2.20

of the side chain, ring closure, and olefin dihydroxylation using commercially available AD-mix-α afforded swainsonine 89 in an impressive overall yield of 40%. Pyne and coworkers reported a completely different synthetic approach to (−)-swainsonine (89) starting from readily available 4-pentynol (Scheme 2.21). This was elaborated via cis-diol 90 into the enantiopure metathesis precursor 91 in a number of steps. Next, RCM with [Ru]-II (6.5 mol%) converted 91 into the cyclic product 92 in 95% yield. Intramolecular N-alkylation and stereoselective cis-dihydroxylation via the same methodology as applied by Blechert et al. completed the synthesis of swainsonine (89).

OR

HO

OR 3 steps

HO

4 steps

[Ru]-I (6.5 mol%)

H

4-Pentynol N OH 90: R = PMB

Boc

91: R = PMB

CH2Cl2, reflux 95%

RO

Boc

N

92: R = PMB 3 steps

HO

H N

OH OH

(−)-Swainsonine (89)

Scheme 2.21

2.2 Five-membered Nitrogen Heterocycles HO

H

R1

OH Ref. 44

O

OH

N

H

RCM Ref. 43

N

1-Epiaustraline (93)

OH

H

OH

N

O

OH

HO

59

OR2

OH

94: R1 = CH2CH2OPMB, R2 = Bn 95: R1 = CH=CH2, R2 = H

(+)-Australine (96)

Scheme 2.22 (+)-1-epi-Australine and (+)-australine, which were derived from a similar RCM precursor.

2.2.4 Pyrrolizidine Alkaloids

The family of polyhydroxyindolizidine alkaloids is not the only target of interest in the field of natural product synthesis when it comes to application of RCM. Several approaches for the formation of the hindered 5,5-fused bicyclic (poly)hydroxypyrrolidines have been reported as well. The RCM methodology is, in some of these cases, applied to construct the second pyrrole, while on the other hand a larger number of cases exist for both natural and nonnatural products where the first cycle is created this way. (+)-Australine (96) [41] and its epimer (+)-1-epi-australine (93) [42] constitute two examples in which RCM was used in the construction of the second pyrrole (Scheme 2.22). Both alkaloids were, together with other australine epimers, isolated from the seeds of Castanospermum australe. Pyne et al. were the first to use RCM in the synthesis of a tetrahydropyrrolizidine, namely, 1-epi-australine 93. Ring opening of epoxide 97, readily obtained from 3-butynol, with an enantiopure allylic amine provided the desired diallylamine in moderate selectivity (Scheme 2.23). Conversion of the diallylamine into oxazolidinone 98 opened the door to the first RCM cyclization. Portionwise addition of [Ru]-I catalyst (30 mol% in total) gave product 95 in 97% yield. When less catalyst was used, significantly lower yields of 95 were obtained. syn-Dihydroxylation of the 3-pyrroline 95 and an additional five more steps were required to obtain (+)-1-epi-australine (93) as a colorless oil. Unfortunately, an identical approach failed for the synthesis of (+)-australine (96) since a Mitsunobu ring closure of the second ring did not take place. A few years later, however, Trost and coworkers published a different route that made an RCM-mediated synthesis of (+)-australine (96) possible. In a similar procedure as PMBO

2 steps

H

PMBO

[Ru]-I (30 mol%) O

O 97

Scheme 2.23

O

CH2Cl2, reflux OBn

98

5 steps O

N

97%

HO

H

PMBO

N

O

N OBn

95

H

OH OH OH

(+)-Epi-australine (93)

60

2 Medium-sized Nitrogen Heterocycles Synthesized by Ring-closing Alkene Metathesis 1. i: Zn, THF, NH4Cl (aq) ii: C2H5OCO2C2H5 2. Phtalimide, (C3H5PdCl)2 (R,R )-L4, 5% Na2CO3 CH2Cl2

Acrolein

H O HO

31% (2 steps)

NPhth

N

O

CH2Cl2

O

99

H

[Ru]-II (1 mol%) rt

3 steps

N

O

OH

OH

77%

100

95 5 steps

HO

H

OH OH

N

OH (+)-Australine (96)

Scheme 2.24

described for the synthesis of compound 31 [24, 25], triolefin 99 was prepared in an enantiomerically pure form. Interestingly, application of RCM on unprotected 100 with only 1 mol% of [Ru]-II afforded the corresponding product 95 in 77% yield. Next, benzylation of the primary alcohol, followed by selective epoxidation and subsequent opening with benzyl alcohol, provided an improved precursor for the second cyclization. Thus, (+)-australine (96) was eventually successfully prepared and isolated as its hydrochloride salt (Scheme 2.24). In Martin’s synthesis of (+)-hyacinthacine A2 (103), one of the alkaloids occurring in the bulbs of Muscari armeniacum (Hyacintheceae), RCM was used for the second ring formation (Scheme 2.25) [43]. Starting from d-arabinofuranose, only four steps were required to prepare the metathesis precursor 102, including reductive amination with allylamine – an intramolecular displacement of the allylic benzoate ester. RCM was performed on the corresponding hydrochloride salt 102 in the presence of [Ru]-I (16 mol%) in toluene to provide the corresponding trihydroxypyrrolizine in 30% yield (75% based on recovered starting material). (+)-Hyacinthacine A2 (103) was eventually delivered by concomitant debenzylation and double bond hydrogenation.

OBn

BzO D-arabino-

3 steps OBn

furanose O

OBn

101

Scheme 2.25

Allylamine, AcOH NaBH3CN Mol. sieves 0 – 40 °C MeOH 78%

H N

OBn OBn

1. HCl, Et2O 2. [Ru]-I (16 mol%) toluene, 60 °C

3. H2, Pd/C, rt, MeOH/THF/HCl OBn

H N

OH OH OH

25% (3 steps) 102

(+)-Hyacinthacine A2 (103)

2.3 Six-membered Nitrogen Heterocycles OBn

104

OBn

EtO2C Br

H N 105

3 steps

HO N

Ts O

O 106

1. (COCl)2, DMSO Et3N, CH2Cl2 −78 °C to rt 2. Ph3PCH2Br, BuLi THF, −78 °C 3. [Ru]-II (5 mol%) PhH, reflux

OBn 2 steps

H N

H

61

OH

N O

107

Trachelanthamidine (108)

52% (3 steps)

Scheme 2.26

Chang et al. reported a synthesis of the monohydroxypyrrolidine (−)trachelanthamidine (108) also utilizing RCM to create the second cycle (Scheme 2.26) [44]. The synthesis of this alkaloid, isolated from the plant Trachelanthus korolkovi, commenced with a stereo- and regioselective stepwise [3+2]-annulation reaction of bromoacrylate 104 with sulfonylacetamide 105 leading in two more steps to compound 106. Preparation of the diene for the metathesis reaction was achieved by subsequent Swern oxidation and Wittig olefination. The pyrrolidizine skeleton 107 was formed upon ring closure, employing [Ru]-II (5 mol%), in 81%. Through hydrogenation and subsequent lactam reduction, the authors achieved a synthesis of (−)-trachelanthamidine (108) in a total of seven steps.

2.3 Six-membered Nitrogen Heterocycles

Alkaloids incorporating six-membered nitrogen heterocycles are abundantly present in nature and occur in a multitude of alkaloid natural product families. Many of these alkaloids exhibit a wide range of diverse biological activities. As a result, extensive research has been devoted to the synthesis of many of the natural products and derivatives. This has led to an enormous variety of synthetic methodology that has been efficiently applied to synthesize many of these alkaloids. Among these methods, metathesis-based processes have played an increasingly important role as is illustrated in the sequel. 2.3.1 Piperidine Alkaloids 2.3.1.1 Piperidines (−)-Allosedamine (113) and (+)-sedamine (119), two piperidine alkaloids isolated from Lobelia inflate, have been used for the treatment of respiratory disorders such as asthma, bronchitis, and pneumonia. While several racemic syntheses have been reported, there are much less asymmetric ones. An enantioselective route to (−)-allosedamine (113) was developed by Raghaven et al., which commenced with addition of lithium compound 110 to the imine 109 providing the separable allylic amine 111 as a 3 : 1 mixture of diastereoisomers (Scheme 2.27) [45, 46]. Six

62

2 Medium-sized Nitrogen Heterocycles Synthesized by Ring-closing Alkene Metathesis NHTs

NHR O LDA, −78 °C add 109, then add 110

Ph 109 + O S

S

Ph

THF 81%

S

Me HO N

6 steps

Ph

NHR O Ph

Tol

Tol

111s +

CO2Et 112

Tol

111a

Li 110

Me HO N

90%

Me HO N

3 steps

Ph (−)-Allosedamine (114)

[Ru]-I (5 mol%) reflux, 2 h toluene

Ph 113

Scheme 2.27

additional steps were required to form metathesis precursor 112 from the major isomer 111a. Straightforward RCM ([Ru]-I, 5 mol%) provided the corresponding heterocycle 113, which upon hydrogenation, deprotection, and methylation afforded (−)-allosedamine (114). Yadav et al. developed via a similar strategy stereoselective syntheses of both (−)-allosedamine (114) and (+)-sedamine (119), starting from the Sharpless epoxidation product of cinnamyl alcohol (114, Scheme 2.28) [47]. The latter compound was converted in a number of steps into a separable diastereomeric mixture of amino alcohols 116a and 116b. N-Allylation gave in both cases the corresponding RCM precursors, which on exposure to [Ru]-I catalyst (20 mol%, benzene, 50 ◦ C) yielded the cyclic compounds 117 and 118 in 90 and 92% yield, respectively. Finally, deprotection and methylation led to (−)-allosedamine (114) and (+)-sedamine (119) via identical procedures. A different approach to synthesize (+)-sedamine (119), reported by Cossy et al. [48, 49], was based on a double enantioselective allyltitanation of the aldehydes 120 and 122 (Scheme 2.29). Next, transformation of the resulting homoallylic alcohol 123 into amine 124 was accomplished via a Mitsunobu reaction, which was in a two-step procedure converted into the RCM precursor 124. Completion of the synthesis was achieved through RCM (catalyst [Ru]-I, benzene, reflux), followed by hydrogenation of the alkene and deprotection, in 52% over four steps.

O Ph

7 steps MOMO OH

NHBoc

Ph

115

116a + MOMO

NHBoc

Ph 116b

1. NaH, H2C=CHCH2Br Boc DMF, 50 °C MOMO N 2. [Ru]-I (20 mol%) benzene, 50 °C

Ph 117

1. NaH, H2C=CHCH2Br Boc MOMO N DMF, 50 °C

92%

Scheme 2.28

Ph

(−)-Allosedamine (114)

90%

2. [Ru]-I (20 mol%) benzene, 50 °C

Me OH N

2 steps

Ph

Me OH N

2 steps Ph

118

(+)-Sedamine (119)


O H

86%

THF / H2O

121

N

O

OPMB

H

Ph Ph

Ph

O O

122

Ph 124

Cp Ti O Ph Ph O

Ph

123

Me Me

(R,R )-I

75% (R,R )-I, −78°C (2 steps) Et2O 3 steps OH OPMB

Boc OPMB

94%

125 3 steps

OPMB OsO4 /NaIO4 Ph

[Ru]-I (5 mol%) Boc N OPMB benzene, reflux Ph

Cp Ti O Ph Ph O

1. (S,S )-I, Et2O, −78°C 2. t -BuOK, PMBBr, THF

Ph 120

63

Ph Ph

NH OH Ph 119: (+)-Sedamine

O O

Me Me

(S,S )-I

Scheme 2.29

A synthesis of (+)-allosedamine, described by Chang et al. in 2004, started with a hydrolytic kinetic resolution of 126, which in turn was prepared in three steps from (+)-styrene oxide (Scheme 2.30) [50]. Subsequent formation of the diallylamine 129 as shown above, followed by [Ru]-II-catalyzed RCM (10 mol%, benzene, 70 ◦ C), afforded (+)-allosedamine (131). α-Conhydrine, one of the alkaloids occurring in hemlock, was isolated from the seeds and leaves of the poisonous plant Conium maculatum. A synthesis of α-conhydrine (135) [51] started with transformation of prolinol derivative 132 into amino alcohol 133 via a regioselective Baeyer–Villiger reaction of ketone 132 1. L5, AcOH, H2O THF, 0 °C to rt 2. PPTS/MeCN/ OMOM MOMO O CH2Cl2, 25 °C

O +

129

N

OH OH

Ph

Ph

Ph

MOMO

130

131

44%

47%

N Co

t -Bu

O

L5

t -Bu

O

t -Bu

4 steps Me MOMO N

1. [Ru]-II (10 mol%) benzene, 70 °C 2. PtO2, H2, EtOAc 25 °C, 1 h

Ph

Me MOMO N Ph

132

133 54%

2 steps

Me HO N Ph (+)-Allosedamine (134)

Scheme 2.30

t -Bu

64

2 Medium-sized Nitrogen Heterocycles Synthesized by Ring-closing Alkene Metathesis OH

O Et 2 steps

N

Et

HN

Ts

OBn

Ts

135

5 steps N

OBn

Ts

136

137

1. [Ru]-II (15 mol%) CH2Cl2, reflux 2. 10% Pd/C, EtOH rt, 10 h Et 3. Na/Hg (6 mol%) OBn Na3PO4, MeOH, rt, 5 h

N H

138: a-Conhydrine

69% (3 steps) 42% (3 steps)

OMOM

OMOM

PdCl2(MeCN)2 HN

O CCl3

139

Toluene

i: 2 M NaOH ii: ClCOCH2CH=CH2 Et3N

OMOM

Et

Et HN

O

55%

+

HN

CCl3 140

O CCl3

1:16

141

CH2Cl2 52%

Et OH

1. [Ru]-I (10 mol%) CH2Cl2, reflux 2. 10% Pd/C, H2, EtOAc 3. i: BH3·THF ii: 6 M HCl OMOM

Et HN O 142

Scheme 2.31

(Scheme 2.31). Compound 133 was then subjected to oxidation and subsequent Wittig olefination to afford the RCM precursor 134. Smooth RCM under the influence of catalyst [Ru]-II (15 mol%, CH2 Cl2 , reflux) gave the corresponding unsaturated heterocycle in 92% yield, which upon hydrogenation and desulfonylation provided the target compound 135. Another approach to the synthesis of α-conhydrine (138) involved an Overman rearrangement of 136, which was prepared from (S)-glycidol in seven consecutive steps [52]. Deprotection of the amine 139, followed by acylation, provided the RCM precursor 139. Ring closure under the influence of catalyst [Ru]-I (CH2 Cl2 , reflux) afforded the corresponding unsaturated piperidine in quantitative yield. Finally, hydrogenation and subsequent MOM deprotection afforded 135 in 42% yield. An auxiliary-based synthesis of the related alkaloid (+)-β-conhydrine (144) commenced with conversion of enantiopure benzyl-protected 2-hydroxypropionaldehyde into hydrazone 140b (Scheme 2.32). Diastereoselective alkylation and subsequent acylation gave the RCM precursor 141b, which upon treatment with [Ru]-II (5 mol%, CH2 Cl2 , rt) led to the corresponding product in 75% yield. Subsequent hydrogenation and deprotection led to (+)-β-conhydrine (144) in a yield of 23% over six steps [53]. Fairly similar procedures were applied in the synthesis of (S)-(+)-coniine (143) [54]. A formal synthesis of (S)-(+)-coniine (143) was developed by Chang et al. in 1999 starting from l-norvaline (144), which was converted into diolefin 145 in several steps [55]. Finally, RCM ([Ru]-I, CH2 Cl2 , rt), followed by hydrogenation, led to the known intermediate 146 [56] thus completing the formal synthesis (Scheme 2.33). Another approach in the synthesis of functionalized piperidine alkaloids (−)-coniine (149a) and (−)-pipecoline (149b) started with the imines 147a and 147b using (R)-α-methylbenzylamine as a chiral auxiliary (Scheme 2.34) [57]. Consecutive diastereoselective vinylation and N-allylation afforded compounds 148a and 148b as two diastereoisomers (71% de, major isomer shown). RCM


65

.HCl

3 steps N H R N

1. CH2=CHCH2Li Et2O, −78 °C to rt

N

2. CH2=CHCOCl OMe

−78 °C

140a: R = Pr 140b: R = (CH2)2CHOBn

[Ru]-II (5 mol%) O

N N

R

CH2Cl2, rt OMe

49%

O

N

Pr

(S )-(+)-Coniine·HCl (143)

R

N OMe

75% 142

141a: R = Pr 141b: R = EtCHOBn

4 steps

N H H

Scheme 2.32

([Ru]-I, CH2 Cl2 , rt) gave rise to the corresponding cyclic products in good yields, of which the diastereoisomers were separated via silica gel chromatography. Eventually, straightforward hydrogenation led to the natural products 149a and 149b. An auxiliary-mediated approach reported by Agami et al. also led to a synthesis of (−)-β-conhydrine (156, Scheme 2.35) [58]. This time (S)-phenylglycinol (150) was used as the starting material, which via Weinreb amide 151 and N,O-acetal 152 was converted into oxazolidinone 153. The latter compound was then diastereoselectively allylated using the corresponding N-acyliminium ion, deprotected, and allylated at the nitrogen atom. The resulting diolefin 154 was then subjected to RCM [catalyst [Ru]-I (4 mol%), CH2 Cl2 , reflux, 79%] and another four consecutive steps to give (−)-β-conhydrine (156). Probably, the most well-known alkaloid in N. tabacum is (S)-nicotine (Section 2.2.2), which plays an important role in the modulation of nicotinic acetylcholine receptors (nAChRs). However, related alkaloids have also been isolated from the same source, such as the homologous analogs (S)-anabasine (160a) and (S)-anatabine (161a). In 2000, Felpin et al. published a synthesis of the latter alkaloids starting from an asymmetric allylboration of 3-pyridinecarboxaldehyde into compound 157 [59, 60]. Azide substitution of the alcohol, followed by reduction and functionalization, provided the RCM precursor 158. Subsequent protonation of the pyridine ring, followed by RCM, proceeded uneventfully ([Ru]-I catalyst, CH2 Cl2 ) to give the ring-closed product 159 that via hydrogenation was converted into both (S)-anabasine (160a) and its methylated counterpart 160b. 1. [Ru]-I (5 mol%) CH2Cl2, rt

H NH2 Pr

CO2H

N Boc

144

Scheme 2.33

145

Pr

2. Pd/C, H2, EtOH

Ref. 58 N Boc

Pr

.HCl N H

Pr

94% (2 steps) 146

Me

OH (+)-(b )-Conhydrine (144)

(S )-(+)-coniine·HCl (143)

66


N

1. CH2=CHCH2Br, Zn* CeCl3·7H2O, THF, rt

R

Me

1. [Ru]-I, CH2Cl2, rt N

2. CH2=CHCH2Br, NaH

Ph

THF, 0 °C

Me

147a: R = Pr 147b: R = Me

R

N H

2. Pd/C, H2, MeOH

R

Ph

148a: R = Pr (57%) 148b: R = Me (60%)

(−)-Coniine (R = Pr, 58%) (149a) (−)-Pipecoline (R = Me, 68%) (149b)

Scheme 2.34

Alternatively, (S)-anatabine (161a) and its methylated derivative 161b could be directly obtained from the metathesis intermediate 159 (Scheme 2.36). En route to the natural products ent-CP-999,994 (165) and ent-L-733,060 (166), in 2005 Nakano et al. reported a highly enantioselective palladium-catalyzed asymmetric allylic amination (>99% ee) of the allylic acetate 162 using the ligand L6 [61]. Conversion of the resulting allylic amine 163 with 3-butenoic acid in the presence of DCC followed by RCM (catalyst [Ru]-II, CH2 Cl2 , reflux, 94%) yielded the piperidinone core structure 164, which was readily converted into the target molecules 165 and 166 (Scheme 2.37). 2.3.1.2 Piperidine Carboxylic Acids A significant number of alkaloids contain the 2-piperidinecarboxylic acid (pipecolic acid) unit as a core structure. Being a proline homolog, the pipecolic acid moiety has been used in many modified peptides and synthetic drugs. For example, naturally occurring 4-hydroxy-2-pipecolic acids have been isolated from various green plants and constitute important chiral building blocks of biologically active molecules such as palinavir. A synthesis of 4-hydroxy-2-pipecolic acid derivatives described by Johnson et al. involved conversion of (S)-167 into the corresponding oxazolidinone, which was N-allylated using allyl bromide and then subjected to [Ru]-I catalyst NH2

1. EtMgBr, Et2O rt

O

O OH 4 steps

N OMe N H Me

150

151

O N H

2. NaBH4, EtOH −78 °C 49% (2 steps)

OH

152

Me

NaH, reflux THF 98%

Me N H

Me

[Ru]-I (4 mol%) reflux N

CH2Cl2

OH 155

O O

154

78% 3 steps N H

Me OH

(–)-b-Conhydrine (156)

Scheme 2.35

O AllylSiMe3, TiCl4 −78 °C CH2Cl2 98%

Me N O 153

O


OH 6 steps N

Cbz

N 157

1. HCl (g), CH2Cl2 2. [Ru]-I, CH2Cl2

N

79% (2 steps)

2 steps

N Cbz

N

N R N (S )-Anabasine (R = H) (160a) (S )-N -Methylanabasine (R = Me) (160b)

159

158

67

2 steps

N R

N

(S )-Anatabine (R = H) (161a) (S )-N -Methylanatabine (R = Me) (161b)

Scheme 2.36

(CH2 Cl2 , rt) to give the cyclic product 168 [62]. The metathesis product was either readily converted into the protected trans-4-hydroxypipecolic acid derivative 169, but upon elaboration of the side chain also to the natural product (2S,4S)-(−)-SS20846A (171) (Scheme 2.38). Baikiain (175), another example of a natural product containing the pipecolic acid core structure, was synthesized in 2000 by Rutjes et al. via a metathesis route [63]. Amidopalladation of methyl propadienyl ether with the enantiopure l-allylglycine derivative 172 provided the RCM precursor 173 in good yield. Cyclization ([Ru]-I catalyst, 10 mol%, CH2 Cl2 , rt) afforded the corresponding cyclic N,O-acetal 174, which was readily transformed into the natural product via N-acyliminium-mediated reduction with Et3 SiH, followed by deprotection (Scheme 2.39). A different approach to baikiain (175) was also reported two years later by Riera et al. [64]. Their synthesis commenced with a Sharpless epoxidation of 2,5-hexadienol to give the enantiopure epoxide 176. Subsequent epoxide ring opening with allylamine and N-Boc protection afforded the RCM precursor 177, which was readily ring closed ([Ru]-I, CH2 Cl2 ) in 72% to the tetrahydropyridine derivative. Finally, oxidative cleavage of the diol, followed by Pinnick oxidation, led to the N-Boc-protected natural product 178 (Scheme 2.40).

OAc Ph

L6, [Pd(Cl(h3-C3H5)]2 PhCH2NH2, 0 °C

Ph

90% ee = 99%

162

1. 3-Butenoic acid DCC, DMAP CH2Cl2, rt O Ph 2. [Ru]-I, CH2Cl2 reflux

H N

NHBn

Ph 163

N

94%

S

L6

Scheme 2.37

O H Ph2P

Ph

N H

Bn 164

O N H

Ph

ent -L-733,060 (166)

OMe

ent -CP-99,9994 (165)

CF3

O-DES-PS

Ph

CF3

68

2 Medium-sized Nitrogen Heterocycles Synthesized by Ring-closing Alkene Metathesis 1. i: NaH, THF ii: NaH, LiI 4-Bromo-1-butene DMF

HO

OMOM 4 steps

O

N

NHBoc 2. [Ru]-I CH2Cl2 167

HO2C

O 168

72% (2 steps)

N Boc 169

6 steps OH

OMOM AcCl MeOH

N Boc

N H

100% Me

Me 170

(2S,4S )-(−)-SS20846A (171)

Scheme 2.38 OBn Pd(OAc)2, dppp, Et3N sealed tube, MeCN HN Ns

CO2Me

Reflux

[Ru]-I (10 mol%) CH2Cl2

BnO

N Ns

50%

172

BnO

rt

CO2Me

N Ns

CO2Me

N H

CO2H

88%

173

174

Baikain (175)

Scheme 2.39

2.3.1.3 Piperidones The 2-(1H)-pyridone ring system is abundantly encountered not only in a wide variety of naturally occurring alkaloids, but also in the synthetic intermediates of piperidine, (iso)quinoline, indolizidone, and quinolizidine alkaloids. Plants belonging to genus Lycopodium are known to contain alkaloids possessing unique skeletal characteristics and biological activities, such as acetylcholine esterase (AChE) inhibition. One of them is the Phlegmarine-type alkaloid lycoposerramine-V (183) [65]. Its synthesis featured the initial formation of cyclohexene 179 from (R)-3-methylcyclohexanone (Scheme 2.41). t-Butyldiphenylsilyl removal, subsequent Swern oxidation and elaboration of the double bond provided aldehyde 180,

2 steps

O

N

Boc

OH OH 176

Scheme 2.40

177

1. [Ru]-I, CH2Cl2 2. NaIO4, THF/H2O OH 3. NaClO , t -BuOH/H O 2 2 NaH2PO4 54% (3 steps)

CO2H N Boc 178


EtO2C

1. NH2OMe·HCl, AcOH toluene, reflux 2. LAH, THF, 0 °C to rt

EtO2C 4 steps

CHO

OTBDPS O

Me

Me

3. (COCl)2, DMSO, Et3N CH2Cl2, −78 °C to rt 54% (3 steps)

179

OHC

181

5 steps N H H

Me

N

2. i: H2, Pd/C, EtOAc, rt ii: BH3·THF, THF reflux

O

N H H

Me

N

76% (2 steps) Lycoposerramine-V (183)

N

Me

180

1. [Ru]-I, CH2Cl2, rt

69

182

Scheme 2.41

which was then converted into quinoline derivative 181. The aldehyde function was in a number of steps transformed into diolefin 182, proceeding via an asymmetric allylation with B-allyldiisopinocampheylborane. RCM ([Ru]-II, CH2 Cl2 , rt) proceeded in near quantitative yield and was followed by hydrogenation and lactam reduction to afford the targeted lycoposerramine-V (183). Pipermethystine is a major constituent of the leaves of Piper methysticum, a large shrub indigenous to the Islands of the South Pacific [66]. Its synthesis commenced with butenamide 184, which upon subjection to [Ru]-I catalyst (4 mol%, CH2 Cl2 , reflux) gave the pyridinone 185 in good yield. Epoxidation of 185, followed by treatment with t-BuOK, afforded the allylic alcohol 186, which was enzymatically resolved with a lipase to give the fully protected enantiopure (S)-pipermethystine (187). The nonreacted alcohol (R)-187 was afterwards also converted into (R)-pipermethystine (188) upon esterification with acetic anhydride (Scheme 2.42). 2.3.1.4 Polyhydroxypiperidines In addition to five-membered ring azasugars (Section 2.2.2.3), a large number of different six-membered ring azasugars exist in nature. Since these compounds resemble the glycopyranosyl cation after protonation of the nitrogen atom, their structures show promising potential in the treatment of several diseases. Two syntheses of six-membered ring azasugars are described in Scheme 2.43, both starting from the protected diacetone glucose derivative 189. After inversion of the free hydroxyl group, compound 190 was converted into the RCM precursor 191. [Ru]-I-mediated ring closure (CH2 Cl2 , rt) led to the unsaturated bicyclic intermediate 192 in 90% yield. From there, d-1-deoxyallonojirimycin (193) was prepared via face-selective dihydroxylation of the cyclic olefin [67]. In addition, l-1-deoxyallonojirimycin (194) was directly prepared from diacetone glucose derivative 189 according to identical procedures.


70

A similar approach to synthesize six-membered ring azasugars was published by Takahata et al., which commenced with a Wittig reaction of the d-serine-derived Garner aldehyde 195. Facile elaboration into the metathesis precursor 196, followed by treatment with [Ru]-I (CH2 Cl2 , rt), afforded tetrahydropyridine (197) in virtually quantitative yield. The latter structure was shown to be a good starting point for the preparation of fagomine (200) and its congeners 199 and 201 (Scheme 2.44) [68]. 2.3.2 Indolizidine Alkaloids

Coniceine (205), a bicyclic alkaloid containing the simplest indolizidine skeleton, has attracted much attention as a building block of indolizidine alkaloids. An early RCM-based synthesis was described by Chang et al. [69]. It started with the ethyl ester of proline (202), which was converted in four steps into the RCM precursor 203. Subsequent metathesis under the influence of catalyst [Ru]-II (CH2 Cl2 , rt) provided the corresponding cyclic olefin (204), which is a known intermediate en route to coniceine (205) (Scheme 2.45) [70]. Another synthesis was reported in 2005 by Génisson et al. who constructed the six-membered indolizidine skeleton via RCM starting from the epoxide 206 (Scheme 2.46). Selective epoxide hydrolysis, followed by alkylation with the triflate of 3-butenol, led to the diolefin 207 in 72%. Subsequent RCM ([Ru]-II [5–10 mol%], toluene, 70 ◦ C) afforded the tetrahydropyridine 208 in 66% yield. Completion of the synthesis enclosed hydrogenolysis followed by ring closure to afford the targeted alkaloid (−)-lentiginosine (209) [71]. In 2006, Jung et al. reported another synthesis of (−)-lentiginosine (209) starting from the d-xylose-derived olefin 211 [72]. Subjection to chlorosulfonyl isocyanate (CSI) provided the allylic amine 212 with a high diastereoselectivity (syn/anti = HO

1. [Ru]-I (4 mol%) CH2Cl2, reflux, 12 h

O

2. i: n -BuLi, THF, −78 °C ii: Ph(CH2)2COCl

O

184

2 steps

O

N

N H

O

N Ph

O

185

Ph 186 Lipase, CH2=CHOAc 4Å MS, toluene

AcO

Ac2O, TMSOTf 0˚C N

HO

AcO N

O

O

O

Ph

(R )-Pipermethystine (188)

Scheme 2.42

+

O

N

CH2Cl2 98%

Ph

(R )-187 (46%)

O

O Ph

(S )-Pipermethystine (187) (47%)

2.3 Six-membered Nitrogen Heterocycles Me Me

71

Me Me

O O

2 steps

O

O O

O O

O

O HO

Me Me

O

N

O HO

189

O 190

191 [Ru]-I CH2Cl2

OH

OH HO

HO

OH N H

OH

Me Me

O

Cbz

Me Me

O

OH

5 steps

O N Cbz

OH

N H

90%

L-1-Deoxyallonojirimycin

D-1-Deoxyallonojirimycin

(194)

(193)

O

Me Me

192

Scheme 2.43 CHO

[Ru]-I

3 steps O Me

NBoc Me 195

Oxone, CF3COCH3 NaHCO3, aqueous Na2EDTA, MeCN

N Boc 196

OTBDPS CH2Cl2 97%

N Boc

O

OTBDPS

197

1. K2OsO4·H2O 4-methylmorpholine N -oxide acetone, H2O 2. i: HCl (10%), dioxane ii: Dowex 1 × 2 (OH−)-form

OTBDPS

N Boc

60%

198

92% (2 steps)

OH

N H

OH

OH OH OH

3-epi -Fagomine (199)

OH

OH N H

OH

Fagomine (200)

N H

OH

3,4-di-epi Fagomine (201)

Scheme 2.44

1 : 26) in 84% yield. With this compound in hand, the corresponding pyrrolidine was readily formed via intramolecular bromide displacement. Cbz-hydrogenolysis and introduction of the 3-butenyl moiety as described by Génisson led to the RCM precursor 213 (Scheme 2.47). Cyclization (catalyst [Ru]-II, toluene, 70 ◦ C) proceeded readily in 85% yield to give the bicyclic skeleton, which upon benzyl deprotection and dihydroxylation of the olefin gave (−)-lentiginosine (209). A different approach leading to ent-lentiginosine (77) was adopted by Schmidt et al. [73] who prepared pyrrolidine 214 in three steps from l-tartrate. N-Acyliminium ion–mediated allylation provided 215 in quantitative yield as a

72


diastereomeric mixture (1 : 1), which was separated via column chromatography. RCM ([Ru]-I [0.4 mol%], CH2 Cl2 , reflux, 12 hours) afforded the corresponding indolizidine in 78% yield, which was converted into the natural product via deacetylation and hydrogenation (Scheme 2.48). In 1994, the tricyclic perhydropyrrolo[2,1-j]quinolone lepadiformine (222) [74] was isolated from the tunicate Clavelina lepadiformis by Biard et al. and exhibits moderate cytotoxic activity against various tumor cell lines in vitro, as well as high cardiovascular effects in vitro and in vivo. Unfortunately, many synthetic efforts were initially directed to a wrongly assigned structure. However, since 2001, several elegant syntheses have been reported. One of them included RCM of 219 (Scheme 2.49), which was prepared via an Ireland–Claisen rearrangement of amino acid ester 217. Further elaboration provided diolefin 219, which upon RCM ([Ru]-II, CH2 Cl2 , 40 ◦ C, 1 hour) afforded the bicyclic skeleton 220 in 98% yield. Via five more additional steps, the tricyclic cyanide 221 was prepared from which (−)-lepadiformine (222) could be readily synthesized following known procedures [75]. Melodinus alkaloids, also known as meloquinolines, represent a group of monoterpenoid indole alkaloids isolated from certain Apocynaceae species, amongst which is (+)-meloscine (229). Although in 1989 Overman et al. described the first racemic synthesis of meloscine, it was not until recently that Bach et al. succeeded in synthesizing (+)-meloscine (229) in enantiopure form [76]. Its synthesis started with a [2+2]-photocycloaddition of enone 223 in the presence of the chiral auxiliary A (Scheme 2.50). Ring expansion of cyclobutane 224 into the α-hydroxycyclopentenone (225) was then accomplished by a retro-benzylic ester rearrangement. Further conversion of 225 via 226 provided diolefin 227, which after RCM (catalyst [Ru]-II [15 mol%], toluene, 65 ◦ C) gave the corresponding product 228 in an excellent yield (95%). The latter compound was then transformed in five steps into the targeted natural product (+)-meloscine (229). Daly et al. isolated numerous alkaloids such as steroidal batrachotoxins, histrionicotoxins, gephyrotoxins, and pumiliotoxins from the skin of Neotropical poison frogs. One of these alkaloids, (−)-205B (236), possessed an unusual aza-acenaphthylene ring system [77]. In 2005, Smith et al. published a first total synthesis of alkaloid 205B (236) based on a three-component Linchpin coupling of compounds 230–232, followed by a one-pot sequential cyclization of intermediate 233 to complete the indolizidine core structure 234. Upon treatment with Me3 SiCl in the presence of LHMDS, the corresponding kinetic silyl enol ether was formed, which was then directly subjected to metathesis catalyst [Ru]-II (benzene,

H N

4 steps CO2Et

202

Scheme 2.45

O

[Ru]-II

Ref. 72

N

N

CH2Cl2 74% 203

O 204

H N

(−)-Coniceine(205)

2.3 Six-membered Nitrogen Heterocycles 1. 3 M H2SO4, dioxane reflux, 2. Dimethoxypropane CSA, rt

O NHBn

TBDPSO

OH [Ru]-II (5-10 mol%) 70 °C, toluene

HO

3. 210, proton sponge CH2Cl2, rt

206

HO BnN

66% OH

207

38% (3 steps) HO HO BnN TfO 1. 10 % Pd/C, H2, MeOH 12 M HCl (cat.)

HO

210 HO

N

208

2. PPh3, CCl4, Et3N, DMF, rt

(−)-Lentiginosine (209)

61% (3 steps)

Scheme 2.46

65 ◦ C) to provide after an aqueous workup the cyclic ketone 235 in a satisfactory yield (81%). From this aza-acenaphthylene skeleton, (−)-205B (236) was readily synthesized in six steps (Scheme 2.51). 2.3.3 Quinolizidine Alkaloids

The structure of quinolizidine 233A (242), obtained from skin extracts of poisonous frogs of the genera Dendrobates and Mantella, was published in 1993. Following up on racemic syntheses, a formal synthesis of enantiopure 242 was reported by Rutjes et al. based on a cationic cyclization of allylic N,O-acetals (Scheme 2.52) [78]. OBn

CSI, Na2CO3, 0 °C then 25% Na2SO3

Br

Toluene

OBn OBn

OBn Br OBn NHCbz

84% 211

212 1. Et3SiH, Pd(OAc)2, Et3N CH2Cl2, reflux 2. CH2=CHCH2CH2OTf proton sponge, CH2Cl2 3. [Ru]-II, toluene, 70 °C HO HO

(−)-Lentiginosine (209)

Scheme 2.47

49% (3 steps)

N N

73

BnO

OBn 213

74


3 steps

N

O

L-Tartrate

AcO

AllylSiMe3 Me3SiOTf

NH

OAc

N

CH2Cl2

CCl3

AcO

96%

OAc

214

215 [Ru]-I (0.4 mol%) CH2Cl2, reflux

HO

AcO

H

HO

2 steps AcO

N

78% H N

216

(+)-Lentiginosine (77) Scheme 2.48

The approach commenced with CM ([Ru]-IX catalyst [5 mol%, CH2 Cl2 , reflux) of the enantiopure protected 2-amino-6-heptenoic ester 237 with allyltrimethylsilane to afford the modified side chain via unexpected double bond isomerization. Subsequent amidopalladation with benzyl propadienyl ether gave the allylic N,O-acetal 238, which smoothly cyclized (Sn(OTf)2 [2 mol%], CH2 Cl2 ) to give 239 as a 3 : 1 O OTBDPS n=3

O NBoc BnO

HO2C

LiHMDS TBSCl THF, −78 °C to rt 79%

NBoc BnO

217

OTBDPS

218

5 steps

[Ru]-II CH2Cl2, 40 °C

NH OTBDPS

BnO

98%

NBoc OTBDPS

BnO

220

219

5 steps Me

Ref. 77 N

N BnO

CN 221

Scheme 2.49

HO (−)-Lepadiformine (222)

2.3 Six-membered Nitrogen Heterocycles Bn

Boc

N

Bn hv, A silyl enol ether

Boc N

CO2Me OSiMe3

Toluene N H

O

76%

223

N H

H O

Bn

K2CO3 rt

Boc N O

OAc

H O

MeOH N H

98%

224

225 4 steps

CO2Me

N

[Ru]-II (15 mol%) 65 °C, 18 h H O

N H 228

7 steps

Toluene 95%

Boc OAc N

CO2Me

N

N H 227

H O

N H 226

H O

N 5 steps H N O H (+)-Meloscine (229)

H NO

N

O Me Me

Me A

Scheme 2.50

mixture of diastereoisomers (major one shown). Deprotection, introduction of the second alkene, and subsequent RCM (catalyst [Ru]-II [5 mol%], CH2 Cl2 , reflux) afforded 240 in 83% yield. Finally, hydrogenation and reduction of both carbonyl groups provided 241, thereby constituting a formal synthesis of quinolizidine 233A (242) [79]. In 2005, Honda et al. published a first total synthesis of a quinolizidine alkaloid using RCM as one of the key steps, namely, (−)-deoxynupharidine (246) [80]. The alkaloid was isolated from plants of the genus Nuphar and exhibited amongst others immunosuppressive activity. Its synthesis commenced with a stereoselective construction of the enantiopure 5,6-disubstituted piperidinone building block 244 by modification and ring expansion of the protected d-pyroglutamic acid derivative 243 (Scheme 2.53). Subsequent RCM of 244 (catalyst [Ru]-II, benzene, 60 ◦ C) afforded quinolizidinone 245 in 96% yield, whereafter three consecutive steps led to the natural product itself. Blaauw et al. described a first synthesis of the quinolizidine alkaloid (+)epiquinamide (250) [81]. This compound was isolated from the skin of the Ecuadorian frog Epipedobates tricolor and was claimed to show activity on the nicotinic

75

76

2 Medium-sized Nitrogen Heterocycles Synthesized by Ring-closing Alkene Metathesis O NTs

Me O

O

+ S

S

+

O

231

O NHTs OTBS

3 steps

OBPS

TBS 230

Me

Me

S

Me

S

232

OBPS

233 4 steps

Me

O

Me

H H

N Me

Me

6 steps S S

H

H

N

(−)-205B (236)

O

1. LHMDS, Me3SiCl −78 °C, THF

H

2. [Ru]-II, 65 °C benzene then acidic workup

H 235

81% (2 steps)

Me Me H

H N S

S

H 234

Scheme 2.51

1. AllylSiMe3, [Ru]-IX (5 mol%) CH2Cl2, reflux, 16 h HN Ns

CO2Me

SiMe3

2. Benzyl propadienyl ether Pd(OAc)2 /dppp (5 mol%) BnO Et3N, MeCN, rt, 16 h

237

N

CH2Cl2

CO2Me

Ns

86% 238

70% (2 steps)

Sn(OTf)2 (2 mol%) 0 °C to rt, 2 h CO2Me

N Ns

239 1. PhSK, MeCN, 50 °C 6 h, 59% 2. 3-butenoyl chloride 2,6-lutidine, CH2Cl2 0 °C, 2 h 3. [Ru]-II (5 mol%), CH2Cl2, reflux, 4 h

Et

Et N

Ref. 81

N

OH

1. PtO2 (cat.), H2 MeOH, 2 h

N

2. LiAlH4, THF 70 °C, 20 h Quinolizidine 233A (242)

241

88% (2 steps)

CO2Me O

240

Scheme 2.52

receptor. The approach started with the efficient conversion of commercially available l-allysine ethylene acetal (247) in the functionalized pipecolic acid derivative 248. RCM under the influence of catalyst [Ru]-II (toluene, rt), followed by hydrogenation, provided the bicyclic skeleton 249 in excellent yield. Further functional group manipulation then provided (+)-epiquinamide (250) in an overall yield of 22% from the RCM precursor 248 (Scheme 2.54). The Lythraceae plants produce a large variety of bioactive alkaloids. A number of them comprise the quinolizidine scaffold, such as (−)-lasubine I and II. Liao et al.


77

described in 2006 a synthesis of (−)-lasubine I (254) [82] involving RCM to construct one of the two rings (Scheme 2.55). Most effort were directed to an enantioselective synthesis of the functionalized piperidine and the RCM precursor 253. This was achieved via SN 2-type cyclization of amino alcohol 252, which in turn was prepared via two sequential Roush asymmetric allylboration steps from the protected amino alcohol 251. RCM under the influence of catalyst [Ru]-I (CH2 Cl2 , reflux) gave the corresponding ring-closed product in 92% yield, which upon catalytic hydrogenation in the presence of HF·Et3 N was efficiently converted into (−)-lasubine I (254). A synthesis of the highly conjugated pentacyclic quinolizidine alkaloid mitralactonine (257), isolated in 1999 by Takayama et al., shows the usefulness of RCM in constructing polycyclic frameworks [83]. It commenced with a protected tryptamine derivative, which amongst other steps via a Pictet–Spengler cyclization was converted into the RCM precursor 255. RCM (catalyst [Ru]-II [10 mol%], toluene, 80 ◦ C) proceeded smoothly both on the N-Ts-protected (87%) as well as on the unprotected compound. The tosyl protection, however, appeared crucial in the next steps involving osmium-catalyzed dihydroxylation and oxidation into the corresponding hydroxyl enone intermediate. TBAF-mediated Ts deprotection then gave rise to the free indole 256, which is a known precursor for preparing mitralactonine (257) (Scheme 2.56) [84]. The sparteine group of alkaloids, abundant in common papilionaceaous plants, is characterized by an asymmetrical exo–endo arrangement of bridgehead hydrogen atoms at C6 and C11. (−)-α-Sparteine, obtained from the commercially available (−)-sparteine by isomerization, has found widespread use as chiral ligand in asymmetric catalysis and asymmetric deprotonation reactions. Therefore, a practical stereocontrolled entry into the sparteine core structure would contribute to exo-ending the versatility of existing asymmetric methods. An elegant approach was detailed in 2005 by Blakemore et al. who identified tetraoxobispidine (260) as a synthetically versatile scaffold, which could be readily prepared starting from a Knoevenagel condensation of dimethyl malonate (258) with paraformaldehyde (Scheme 2.57) [85]. The intermediate was directly treated with ammonia (259), followed by cyclization and subsequent double allylation, to provide the key intermediate 260. Subsequent Grignard addition resulted in a remarkably regioselective diallylation affording the tetraene and the RCM precursor 261. Subjection to [Ru]-I catalyst (4 mol%, CH2 Cl2 , rt) provided the tetracyclic structure 262 in 82% yield. Me

Me

8 steps O

N

CO2Me N Boc

Scheme 2.53

[Ru]-II

N

O

H N

Benzene, 60 °C Me

243

O

Me 3 steps

H

244

(96%)

O Me 245

Me (−)-Deoxynupharidine (245)

78


O HO

O

HO

5 steps H2N

CO2H

Ph

O

247

[Ru]-II

CO2Me

N

AcHN

H

Toluene, rt 98%

248

CO2Me

N

6 steps

H N

O 249

(+)-Epiquinamide (250)

Scheme 2.54

Finally, racemic α-sparteine (266) was obtained upon global reduction of all olefin, aminal and carbonyl functions. Alkaloids isolated by Rakotoson et al. in 1998 from Galipea officinalis consisted of several 2-substituted quinolines among which is (+)-(S)-angustureine (266). An RCM-mediated synthesis of this compound started with a Mitsunobu alkylation of sulfonamide 265 with the enantiopure secondary alcohol 262 to give the desired α,ω-diene 264 in 78% yield. With the RCM precursor in hand, subjection to [Ru]-II catalyst (CH2 Cl2 , 50 ◦ C) gave quinoline 265 in 92% yield. Finally, hydrogenation, deprotection, and subsequent methylation provided (+)-(S)-angustureine (266) in 68% yield (Scheme 2.58) [86]. 2.4 Seven-membered Nitrogen Heterocycles

In contrast to the large abundance of RCM examples of five- and six-membered nitrogen heterocycles en route to natural products, RCM to prepare seven-membered rings occurs considerably less frequently. One of the examples concerns a formal synthesis of (−)-balanol (271, Scheme 2.59). Balanol was first isolated in 1993 from the fungus Verticillium balanoids and is a potent inhibitor of human protein kinase NHTs 6 steps

MeO MeO

TsHN OH

TBS OH

MeO MeO

251

O

252 3 steps

OH

OTBS

1. [Ru]-II CH2Cl2, reflux N MeO OMe (−)-Lasubine I (254)

Scheme 2.55

2. H2, Pd/C HF·Et3N 75% (3 steps)

N MeO OMe 253

2.4 Seven-membered Nitrogen Heterocycles

Me N N

1. [Ru]-II (10 mol%), toluene 80 °C, 3 h 2. K3Fe(CN)6, K2CO3, OsO4 (cat.), t -BuOH:H2O 1:1 rt, 24 h, 60%

255

Me

N N H

3. DMSO, (COCl)2, CH2Cl2 Et3N, 60 °C 4. TBAF, THF, reflux, 1.5 h

Ts

79

OH O 256

16% (4 steps)

Ref. 86

N

Me O

N H

O MeO2C

Mitralactonine (257)

Scheme 2.56 O

KOH, 90 °C then aq. NH3 MeO2C

+

(CH2O)n 56%

CO2Me

O

H2N

NH2

1. MeSO3H, 150 °C

H2N

NH2

2. CH2=CHCH2Br NaH, DMF, rt

O 258

O 259

O

O

N

N

O

O

17% 2 steps

260 81%

H

OH

2 steps

N

N H

(±)-a-isosparteine (263)

N O

O N

HO 262

CH2=CHCH2MgBr −78 °C Et2O, THF 77%

[Ru]-I (4 mol%) CH2Cl2, rt OH

N O

O N

HO 261

Scheme 2.57

C. Because of its high inhibitory activity, balanol has been a target for a number of synthetic efforts. In 1999, Cook and coworkers published a formal synthesis of balanol (271) based on RCM to form a seven-membered nitrogen heterocycle as a key step [87]. The balanol synthesis started with d-serine derivative 267, which via oxazoline 268 was transformed into the metathesis precursor 269 (Scheme 2.59). RCM using [Ru]-I catalyst (2 × 5 mol%, CH2 Cl2 , reflux) provided the corresponding unsaturated heterocycle in 77% yield, which upon acidic hydrolysis of the oxazoline and hydrogenation with Wilkinson’s catalyst afforded 270, thereby completing the formal synthesis [88].

80


HO

Me

DEAD, PPh3

262 +

THF, rt 78%

N

Me

Ts 264

NH Ts 263

[Ru]-II, CH2Cl2, 50 °C

92%

3 steps N Me

N Ts

Me

(+)-(S )-Angustureine (266)

Me

265

Scheme 2.58

Two years later, Fürstner et al. developed a synthesis of azepine 277 based on a selective Sharpless asymmetric epoxidation of divinylcarbinol (Scheme 2.60) [89]. The resulting epoxide 272 was then regioselectively opened with allylamine to give the corresponding diene. Protection of the secondary amine with an N-Boc group provided the precursor 275 for RCM. Cyclization proceeded in 94% yield using [Mo]-I catalyst (CH2 Cl2 , 30 min, reflux, 94%). Subsequent conversion of the OTBS CO2Me

TBSO

N OBn

NHBoc

N

O

267

N 268

HO2C O HO OH OH O

Cbz 1. [Ru]-I (2 × 5 mol%) CH2Cl2, reflux 2. 2N HCl, THF, rt then excess Et3N MeOH, rt 3. (Ph3P)3RhCl (5 mol%) H2 (25 psi), benzene

OH O Balanol (271)

Scheme 2.59

269

52% (3 steps)

OH O

Ref. 88

O NH HN

OBn

O

4 steps

N

N H OBn

Cbz 270

2.5 Eight-membered Nitrogen Heterocycles

OBn [Mo]-I, 30 min reflux OH

Allylamine 70 °C

OR O

N R

94%

3 steps OH

274 R = H 275 R = Boc

NaH, BnBr THF, rt

N H

N

OBn

Boc

94% 272 R = H 273 R = Bn

OH O

OBn N R

CH2Cl2

81

276

277

Boc2O, Et3N CH2Cl2, rt

95%

100%

Scheme 2.60

secondary alcohol into the azide, reduction, and acylation gave the Boc-protected derivative of 277, which also represents a formal synthesis of balanol [88]. An additional example involves the synthesis of (−)-stemoamide 282 (Scheme 2.61) [90]. In a five-step procedure, 278 was synthesized starting from the commercially available (S)-pyroglutaminol. Next, a vinyl iodide was installed via a chemoselective iodoboration of the alkyne substituent and subsequently subjected to the Reformatsky reagent derived from ethyl α-bromoacetate to yield the RCM precursor 279. By using [Ru]-II catalyst under high dilution conditions (c = 0.005 M), azepine 280 was formed in an excellent yield of 92%. Since installation of the C8-hydroxy group via hydroboration was unsuccessful, an alternative was sought in hydrolyzing the ester and exposure to CuBr2 on alumina, which gave the desired lactone 281. Conjugate hydride addition from the sterically most accessible β-face using NiCl2 ·6H2 O and NaBH4 , followed by methylation, completed the synthesis.

2.5 Eight-membered Nitrogen Heterocycles

Manzamine A (288) has attracted much attention over the past years because of its potent antitumor activity and since it was the first member of a group of O N

1. B-I-9-BBN, −20 °C EtO CH2Cl2 /hexane

O

O N

2. Pd(PPh3)4, 50 °C THF/DMPU 278

92%

O

H O

N H

H Stemoamide (282)

O 1. NiCl2, NaBH4 −30 °C, MeOH 2. MeI, LHMDS THF, −78 °C 71% (2 steps)

Scheme 2.61

N H 280

O Me

O

O

CH2Cl2 279

56% (2 steps)

[Ru]-II EtO 40 °C

N H O

O

H 281

1. LiOH·H2O THF/MeOH/H2O 2. CuBr2 on Al2O3 CHCl3, 65 °C then Et3N, rt 65% (2 steps)


82

Br

CH2=CHSnBu3 CO2Me Pd(PPh3)4

N

R

PhMe O

N

Boc

68%

R

CO2Me

H

H

H

N

N

O

H

N

N

O

Boc

OTBDPS

OTBDPS 283 (R = (CH2)5OTBDPS)

CHO

6 steps

Boc

OTBDPS

284 (R = (CH2)5OTBDPS)

285 2 steps

N

N H

H H

N

[Ru]-I then 1N HCl

OH

H

CH(OMe)2 H

N

OH

26% H

N

Manzamine A (288)

H

287

N

O

1. [Ru]-I CH2Cl2 2. i: KOH, MeOH reflux ii: CH2=CH(CH2)3COCl Et3N, CH2Cl2 50% (2 steps)

H

CH(OMe)2 H

N

O O

N O

286

Scheme 2.62

similar alkaloids to be isolated in 1986. The combination of the complex and unusual structure of manzamine A together with its promising biological activity has led to numerous synthetic efforts. One of these comes from Martin et al. involving two cyclizations via RCM (Scheme 2.62) [91]. The advanced intermediate 283 was converted by a domino Stille/Diels–Alder reaction into the tricyclic system 284. From there, eight sequential steps were required for installing the two olefinic moieties required for the macrocyclic ring formation (286). This very early RCM macrocyclization was carried out using [Ru]-I catalyst giving rise to the 13-membered ring containing predominantly the Z-olefin (Z/E = 8 : 1) in a respectable yield (67%). Then, hydrolysis of the cyclic carbamate followed by N-acylation provided diolefin 287, thereby setting the stage for the formation of the eight-membered ring. RCM with [Ru]-I catalyst took place in a moderate yield of 26%. Promotion of RCM with the Schrock catalyst ([Mo]-I) did not improve the yield. Nevertheless, in a few more steps the synthesis of manzamine A (293) could be successfully completed.

2.6 Conclusion

Only 16 years have elapsed since Grubbs and Fu reported that Schrock’s molybdenum catalyst ([Mo-I]) could be used to induce efficient cyclization of functionalized α,ω-diene-amines. Since then, a wide range of RCM reactions to form the nitrogen heterocyclic products have been reported, including applications in the synthesis of natural products. These applications have been greatly facilitated by the advent

References

of new and more reactive RCM catalysts, thereby continuously widening the range of substrates and increasing the scope of the metathesis processes. This overview shows that, especially, five- and six-membered heterocyclic rings are ideal targets for RCM and that applications in larger rings are lagging behind. On the one hand, this is due to the lower occurrence of the larger rings in natural products; on the other hand, especially, eight-membered rings are significantly more difficult to form using the metathesis reaction. However, it may be clear from the spectacular developments over the past few years that further progress in the years to come will also provide solutions for problems that we are currently dealing with.

References 1 (a) Chauvin, Y. (2006) Angew. Chem.

2

3

4 5

6

7 8

9

10

Int. Ed., 45, 3740–3747; (b) Grubbs, R.H. (2006) Angew. Chem. Int. Ed., 45, 3748–3759; (c) Schrock, R.R. (2006) Angew. Chem. Int. Ed., 45, 3760–3765. Grubbs, R.H. (ed.) (2003) Handbook of Metathesis, Wiley-VCH Verlag GmbH, Weinheim. For recent reviews on RCM, see e.g.: (a) Brown, R.C.D. and Satcharoen, V. (2006) Heterocycles, 70, 705–736; (b) Arisawa, M., Nishida, A., and Nakagawa, M. (2006) J. Organomet. Chem., 691, 5109–5121; (c) Martin, S.F. (2005) Pure Appl. Chem., 77, 1207–1212; (d) Deiters, A. and Martin, S.F. (2004) Chem. Rev., 104, 2199–2238. Ivin, K.J. (1983) Olefin Metathesis, Academic Press, London. Schrock, R.R., Murdzek, J.S., Bazan, G.C., Robbins, J., DiMare, M., and O’Regan, M. (1990) J. Am. Chem. Soc., 112, 3875–3886. Nguyen, S.T., Johnson, L.K., and Grubbs, R.H. (1992) J. Am. Chem. Soc., 114, 3874–3875. Fu, G.C. and Grubbs, R.H. (1992) J. Am. Chem. Soc., 114, 5426–5427. Nguyen, S.T., Grubbs, R.H., and Ziller, J.W. (1993) J. Am. Chem. Soc., 115, 9858–9859. Schwab, P., France, M.B., Ziller, J.W., and Grubbs, R.H. (1995) Angew. Chem. Int. Ed., 34, 2039–2041. Scholl, M., Ding, S., Lee, C.W., and Grubbs, R.H. (1999) Org. Lett., 1, 953–956.

11 Clavier, H., Grela, K., Kirschning, A.,

12

13 14 15

16

17

18

19 20

21 22 23

Mauduit, M., and Nolan, S.P. (2007) Angew. Chem. Int. Ed., 46, 6786–6801. (a) Harrity, J.P.A., Visser, M.S., Gleason, J.D., and Hoveyda, A.H. (1997) J. Am. Chem. Soc., 119, 1488–1489; (b) Harrity, J.P.A., La, D.S., Cefalo, D.R., Visser, M.S., and Hoveyda, A.H. (1998) J. Am. Chem. Soc., 120, 2343–2351. Fu, G.C. and Grubbs, R.H. (1992) J. Am. Chem. Soc., 114, 7324–7325. Philips, A.J. and Abell, A.D. (1999) Aldrichim. Acta, 32, 75–89. Hanessian, S., Margarita, R., Hall, A., Johnstone, S., Tremblay, M., and Parlanti, L. (2002) J. Am. Chem. Soc., 124, 13342–13343. Hanessian, S., Tremblay, M., and Petersen, J.F.W. (2004) J. Am. Chem. Soc., 126, 6064–6071. Welter, C., Moreno, R.M., Streiff, S., and Helmchen, G. (2005) Org. Biomol. Chem., 3, 3266–3268. Leete, E., Bjorklund, J.A., Couladis, M.M., and Kim, S.K. (1991) J. Am. Chem. Soc., 113, 9286–9292. Leete, E. and Kim, S.H. (1989) J. Chem. Soc., Chem. Commun., 1899–1900. Lee, J.-H., Jeong, B.-S., Ku, J.-M., Jew, S.-S., and Park, H.-G. (2006) J. Org. Chem., 71, 6690–6692. Dochnahl, M., Schulz, S.R., and Blechert, S. (2007) Synlett, 2599–2601. Mori, M., Sakakibara, N., and Kinoshita, A. (1998) J. Org. Chem., 63, 6082–6083. Murruzzu, C. and Riera, A. (2007) Tetrahedron: Asymmetry, 18, 149–154.

83

84

2 Medium-sized Nitrogen Heterocycles Synthesized by Ring-closing Alkene Metathesis 24 Trost, B.M., Horne, D.B., and

25

26

27

28

29

30

31 32

33 34

35 36 37

38

39

40 41

42 43

Woltering, M.J. (2006) Chem. Eur. J., 12, 6607–6620. Trost, B.M., Horne, D.B., and Woltering, M.J. (2003) Angew. Chem. Int. Ed., 42, 5987–5990. Deiters, A., Pettersson, M., and Martin, S.F. (2006) J. Org. Chem., 71, 6547–6561. Shimizu, K., Takimoto, M., Sato, Y., and Mori, M. (2006) J. Organomet. Chem., 691, 5466–5475. Shimizu, K., Takimoto, M., and Mori, M. (2003) J. Org. Chem., 5, 2323–2325. Fukumoto, H., Esumi, T., Ishihara, J., and Hatakeyama, S. (2003) Tetrahedron Lett., 44, 8047–8049. Fukumoto, H., Takahashi, K., Ishihara, J., and Hatakeyama, S. (2006) Angew. Chem. Int. Ed., 45, 2731–2734. Kim, S., Lee, J., Lee, T., Park, H.-G., and Kim, D. (2003) Org. Lett., 5, 2703–2707. Kim, S., Lee, T., Lee, E., Lee, J., Fan, G.-J., Lee, S.K., and Kim, D. (2004) J. Org. Chem., 69, 3144–3149. Stragies, R. and Blechert, S. (2000) J. Am. Chem. Soc., 122, 9584–9591. Zaminer, J., Stapper, C., and Blechert, S. (2002) Tetrahedron Lett., 43, 6739–6741. Davis, F.A. and Yang, B. (2005) J. Am. Chem. Soc., 127, 8398–8407. Davis, A.S., Ritthiwigrom, T., and Pyne, S.G. (2008) Tetrahedron, 64, 4868–4879. Karanjule, N.S., Markad, S.D., and Dhavale, D.P. (2006) J. Org. Chem., 71, 6273–6276. Machan, T., Davis, A.S., Liawruangrath, B., and Pyne, S.G. (2008) Tetrahedron, 64, 2725–2732. Buschmann, N., Rückert, A., and Blechert, S. (2002) J. Org. Chem., 67, 4325–4329. Lindsay, K.B. and Pyne, S.G. (2002) J. Org. Chem., 67, 7774–7780. Trost, B.M., Aponick, A., and Stanzl, B.N. (2007) Chem. Eur. J., 13, 9547–9560. Pyne, S.G. and Tang, M. (2004) Tetrahedron, 60, 5759–5767. Rambaud, L., Compain, P., and Martin, O.R. (2001) Tetrahedron: Asymmetry, 12, 1807–1809.

44 Chang, M.-Y., Wu, D.-C., and

45 46 47 48 49

50 51 52 53

54

55 56

57 58 59

60

61

62 63

Chang, N.-C. (2005) Heterocycles, 65, 2965–2971. Raghaven, S. and Rajender, A. (2004) Tetrahedron Lett., 45, 1919–1922. Raghaven, S. and Rajender, A. (2004) Tetrahedron, 60, 5059–5067. Yadav, J.S. (2006) Synthesis, 23, 4005–4012. Cossy, J., Willis, C., Bellosta, V., and Bouzbouz, S. (2000) Synlett, 1461–1463. Cossy, J., Willis, C., Bellosta, V., and Bouzbouz, S. (2002) J. Org. Chem., 67, 1982–1992. Kang, B. and Chang, S. (2004) Tetrahedron, 60, 7353–7359. Chang, M., Kung, Y., and Chen, S. (2006) Tetrahedron, 62, 10843–10848. Jamieson, A.G. and Sutherland, A. (2007) Org. Lett., 9, 1609–1611. Lebrun, S., Couture, A., Deniau, E., and Grandclaudon, P. (2008) Tetrahedron: Asymmetry, 19, 1245–1249. Lebrun, S., Couture, A., Deniau, E., and Grandclaudon, P. (2007) Org. Lett., 9, 2473–2476. Jo, E., Na, Y. and Chang, S. (1999) Tetrahedron Lett., 40, 5581–5582. (a) Enders, D. and Tiebes, J. (1993) Liebigs Ann. Chem., 106, 173–177; (b) Reding, M.T. and Buchwald, S.L. (1998) J. Org. Chem., 63, 6344–6347; (c) Davies, S.B. and McKervey, M.A. (1999) Tetrahedron Lett., 40, 1229–1232 and references cited therein. Pachamuthu, K. and Vankar, Y. (2001) J. Organomet. Chem., 1-2, 359–363. Agami, C., Couty, F., and Rabasso, N. (2000) Tetrahedron Lett., 41, 4113–4116. Felpin, F., Vo-Thanh, G., Robins, R., Villieras, J., and Lebreton, J. (2000) Synlett, 1646–1648. Felpin, F., Girard, S., Vo-Thanh, G., Robins, R., Villieras, J., and Lebreton, J. (2001) J. Org. Chem., 66, 6305–6312. Takahashi, K., Nakano, H., and Fujita, R. (2005) Tetrahedron Lett., 46, 8927–8930. Sabat, M. and Johnson, C. (2001) Tetrahedron Lett., 42, 1209–1212. Tjen, K.C.M.F., Kinderman, S.S., Schoemaker, H.E., Hiemstra, H., and Rutjes, F.P.J.T. (2000) Chem. Commun., 699–700.

References 64 Ginesta, X., Pericas, M.A., and Riera, A. 65

66

67

68

69

70

71

72 73

74 75

76 77 78

(2002) Tetrahedron Lett., 43, 779–782. Shigeyama, T., Katakawa, K., Kogure, N., Kitajima, M., and Takayama, H. (2007) Org. Lett., 9, 4069–4072. Arrayas, R.G., Alcudia, A., and Liebeskind, L.S. (2001) Org. Lett., 3, 3381–3383. Ghosh, S., Shashidhar, J., and Dutta, S.K. (2006) Tetrahedron Lett., 47, 6041–6044. Yasunori, B., Chiemi, A., Hideo, N., Atsushi, K., Isao, A., and Hiroki, T. (2001) Tetrahedron: Asymmetry, 12, 817–820. Park, S.H., Kang, H.J., Ko, S., Park, S., and Chang, S. (2001) Tetrahedron: Asymmetry, 12, 2621–2624. Nukui, S., Sodeoka, M., and Shibasaki, M. (1993) Tetrahedron Lett., 34, 4965. Ayad, T., Génisson, Y., and Baltas, M. (2005) Org. Biomol. Chem., 3, 2626–2631. Kim, I.S., Zee, O.P., and Jung, Y.H. (2006) Org. Lett., 8, 4101–4104. El-Nezhawy, A.O.H., El-Diwani, H.I., and Schmidt, R.R. (2002) Eur. J. Org. Chem., 24, 4137–4142. Marsden, S.P. (2005) Synlett, 2528–2530. (a) Sun, P., Sun, C., and Weinreb, S.M. (2002) J. Org. Chem., 67, 4337–4345; (b) Sun, P., Sun, C., and Weinreb, S.M. (2001) Org. Lett., 3, 3507–3510. Selig, P. and Bach, T. (2008) Angew. Chem. Int. Ed., 47, 5082–5084. Smith, A.B. and Kim, D. (2005) Org. Lett., 7, 3247–3250. Kinderman, S.S., de Gelder, R., van Maarseveen, J., Schoemaker, H.E.,

79 80 81

82

83

84

85

86

87

88

89 90

91

Hiemstra, H., and Rutjes, F.P.J.T. (2004) J. Am. Chem. Soc., 126, 4100–4101. Michel, P. and Rassat, A. (1999) Chem. Commun., 2281. Katoh, M., Mizutani, H., and Honda, T. (2005) Tetrahedron Lett., 46, 5161–5163. Wijdeven, M.A., Botman, P.N.M., Wijtmans, R., Schoemaker, H.E., Rutjes, F.P.J.T., and Blaauw, R.H. (2005) Org. Lett., 7, 4005–4007. Liu, S., Fan, Y., Peng, X., Wang, W., Hua, W., Akber, H., and Liao, L. (2006) Tetrahedron Lett., 47, 7681–7684. Chavan, S.P., Sharma, P., Sivappa, R., and Kalkote, R.T. (2006) Tetrahedron Lett., 47, 9301–9303. Takayama, H., Kurihara, M., Kitajima, M., Said, I.M., and Aimi, N. (1999) J. Org. Chem., 64, 1772–1773. Blakemore, P.R., Kilner, C., Norcross, N.R., and Astles, P.C. (2005) Org. Lett., 7, 4721–4724. Theeraladanon, C., Ariswa, M., Nakagawa, M., and Nishida, A. (2005) Tetrahedron: Asymmetry, 16, 827–831. Cook, G.R., Shanker, P.S., and Peterson, S.L. (1999) Org. Lett., 1, 615–618. Nicolaou, K.C., Bunnage, M.E., and Koide, K. (1994) J. Am. Chem. Soc., 116, 8402–8403. Fürstner, A. and Thiel, O.R. (2000) J. Org. Chem., 65, 1738–1742. Torssell, S., Wanngren, E., and Somfai, P. (2007) J. Org. Chem., 72, 4246–4249. Humphrey, J.M., Liao, Y., Ali, A., Rein, T., Wong, Y., Chen, H., Courtney, A.K., and Martin, S.F. (2002) J. Am. Chem. Soc., 124, 8584–8592.

85

87

3 Synthesis of Natural Products Containing Medium-size Oxygen Heterocycles by Ring-closing Alkene Metathesis Jon D. Rainier

3.1 Introduction

Medium-sized cyclic ethers are present in a number of interesting and important natural products. As illustrated in Figure 3.1, this list includes members of the polycyclic ether family (brevetoxins, ciguatoxins, etc.), the laurencins, heliannuol A, along with members of the eunicellin family. In addition to this are medium ring lactones, including octalactin A, and the 10-membered lactones, microcarpalide and herbarumin III. Recognizing the general difficulty associated with the synthesis of medium rings along with the specific synthetic challenges associated with the generation of the targets listed above, synthetic chemists have devised a number of creative methodologies to medium ring substrates [1]. As the title suggests, the focus of this review is on those methods that utilize alkene metathesis [2]. It was not long after the development of ring-closing metathesis (RCM) that scientists interested in organic synthesis became intrigued with the notion of medium ring generation [3]. However, these chemists also quickly came to realize that the medium rings are thermodynamically unstable, when coupled with their slower rates of formation, that RCM is reversible, and that undesired side products are often formed, for example, from isomerization and intermolecular reactions, and so meant that one needed to pay close attention to the product stability, starting material conformation, and reactivity. A daunting task indeed! This chapter attempts to define what is known about the scope and limitations of oxygenated medium ring synthesis via RCM [4]. As the focus here is on natural product synthesis and the utility of the reaction, the chemistry of the cyclization precursors and their cyclization products is also highlighted. As it is of particular interest to the author, examples of carbonyl-olefination cyclizations using a stoichiometric titanium reagent are also included. Obvious to even the casual follower of the metathesis literature is that a large number of catalysts have been developed and employed in RCM [5] (Figure 3.2). Highlighted here are cyclizations that employ both the Grubbs I ([Ru]-I) and II Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts. Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32440-8

3 Synthesis of Medium-size Oxygen Heterocycles by Ring-closing Alkene Metathesis

88

O HO

OH

O

OH

O O O OH Microcarpalide (2)

O Octalactin A (1) Br H

Me

O

Br

O

AcO

Me H H

H Asbestinin 12 (9)

HO

Me

O

H

O

H

H3C OO H HO

Me OH HO H

Me H H O

H3C H

Figure 3.1

N

PCy3 Ru Ph

Cl PCy3 [Ru]-I

O H CH3 HO H

H

O

O H H OH

OH O H HOH

O H H

HO OH

H OH TBSO Me HO

O CH 3 H Ciguatoxin (11) Me H

O

O

H

Me Me H H O O

Me H OH

CHO O

HH AcO Me OAc Ophirin B (10)

H

O

H

CHO O H

OH H

O O H H H Me Me Brevetoxin B (13)

Natural products conta ining medium ring oxygen heterocycles.

N

Cl Ru Ph Cl PCy3 [Ru]-II

Figure 3.2

H H O

H H O

H Me H O O

O H H

H

H H O

CH3 H O H

O H H Brevetoxin A (12)

Cl

O AcO

OH HH O Me

Me

Astrogorgin (8)

Me

Me H H

O AcO

HH Me AcO OAc

H O

(±)-Hellianuol A (7)

HH

Me

O AcO

OH

O

Me

AcO

OAc

HO

O H H H Br (−)-Isoprelaurefucin (6)

(+)-Laurencin (5)

Isolaurallene (4)

O Herbarumin III (3)

H

O O H

Br

Br

H

H

HO

OH

OH

N

N

Cl Ru Cl O [Ru]-III

PCy3 Cl Ru Cl PCy3 [Ru]-V

F3C F3C Ph

N

O Mo O F3C PCy3 F3C [Mo]-I

Ph

Catalysts/reagents that have been employed in RCM to medium oxygen rings.

([Ru]-II) catalysts [6, 7], along with reactions using the Hoveyda–Grubbs ([Ru]-III) [8], Fürstner ([Ru]-V) [9], or Schrock ([Mo]-I) [10] catalyst. For the purposes of this chapter, medium ring heterocycles are defined as ring systems that are 8-(oxocine)- to 9-(oxonene)- and 10-(oxecin)-membered-rings. Because their synthesis is often problematic, representative examples of seven-(oxepine)-membered rings are also included. While this chapter has mostly been divided into natural product class, also included are more general RCM reactions that lead to medium ring ethers. The reader will find that it is not hard to imagine these methods being used in total synthesis studies.

3.2 General RCM Approaches to Medium Rings


Clark et al. was one of the very first to recognize that RCM could be used to generate medium rings. In model studies aimed at the generation of the medium rings present in brevetoxin A, Clark and Kettle showed that allyl ethers having pendant olefins undergo RCM cyclizations to give oxocines and oxonenes (Table 3.1) [11]. Interestingly, in the generation of oxonenes, the stereochemistry of the ether dictated the efficiency of the reactions (Table 3.1, entries 3 and 4). Diastereomers 18 and 20 gave dramatically different yields of the cyclic products 19 and 21, respectively. This was one of the first studies that demonstrated the significant role of substituents and conformation on RCM cyclizations to medium rings. Clark and Kettle also demonstrated that [Mo]-I catalyst can be used to give dihydropyrans and oxepines from enol ether–olefin RCM (Table 3.2, entries 1–4) [12]. Table 3.1

Generation of oxocines and oxonenes.

7 steps

O O

Entry

H O Ar

CHO

O

H

Diene

H

O

R1 R n

HO

2

[Mo]-I (25 mol%) PhH (0.006 M), 60 °C Ar

Cyclic ether

H O Ar 1

H

H HO

14 Et

Ar

Ar

H

H

O

H

O

H

16

Ar

H

H

O

O

H

18

Ar

H

Et

H

15

97

H

17

86

H

O

H

19

86

Et H HO

O Ar

HO

O

Et H HO

4

H

H Et HO

O Ar

O

O

H Et HO

3

H

Et H HO

O 2

O

O O

O O

H

Ar 20

H

O

H

R2

n

Yield (%)

H

Et

HO

R1

21

14

89

90

3 Synthesis of Medium-size Oxygen Heterocycles by Ring-closing Alkene Metathesis Enol ether–olefin RCM to cyclic enol ethers.

Table 3.2

H O

H R″

O

H

R′

[Mo]-I (25 mol%) C5H12 , rt

n

Entry

Acyclic enol ether

H

1

O H 22

O 26

a Reaction

34

93

O

42

H 94

H

HO

O

O

O O

O

H 31 (27%)

H

H

O

O

30

O

72

HO

29 H

6a

H

O

H

H

O

O H 27

4

5a

H

O

H

O

Yield (%)

O H 25

H

O

H

O

3

28

n R″

O H 23

O H 24 H

H

R′

Product(s)

O

H

2

O

O

O

H 32 (13%)

O 33 (40%)

O

HO

O

O H 35

0

was run at 60 ◦ C in C6 H6 .

Cyclization reactions to the corresponding oxocines were plagued by competitive isomerization and/or dimerization (Table 3.2, entries 5 and 6). Electron-rich olefins, such as those studied by Clark and Kettle, continue to be problematic in RCM due to their slow cyclizations, thereby allowing competing reactions (isomerization


91

O PMP

H

O

1. MeO

H

PMP

Br

O H

O

HO H 37

3. (COCl)2, DMSO, Et3N, CH2Cl2

36

H

O

2. LiAlH4

OH

H

4. Ph3PCH3Br, n -BuLi, THF, rt 59% PMP

Catalyst

O

38

PMP

H

O

Catalyst

O

Conditions

Scheme 3.1

H

HO H

H Catalyst

PMP

Conditions

Scheme 3.2

H

H

[Ru]-I (10–20 mol%)

CH2Cl2 (95

H

45a

H

56 H O

n

H

O 54 H

H

57 H

O

O 52 H O

O

>95

O 50 H O

H

O

O 48 H

H

55 H O

O

m

H

46 H

H

53 H O

O

H

51 H O

O

H

H

44 H

H

49 H O

O

H

47 H O

O

H

45 H O

3

H

43 H O

2

Cyclic ether

O

O

H

58 H

42a

O

O

[Ru]-I (30 mol%), 0.005 M, CH2 Cl2 , 45 ◦ C.

63

93


94

1. KHMDS, −78 °C; TMSCl, −78 °C to rt

O

BnO

H OBn CO2Me H O

OBn

O

2. CH2N2, rt

O 59

Scheme 3.4

CO2Me [Ru]-I (10 mol%) CH2Cl2, 40 °C 87%

O

79%

60

61

Ireland–Claisen, RCM sequence to oxocines.

OPMP O

OPMP CO2Me

1. KHMDS, −78 °C; TMSCl, −78 °C to rt (81%) O

O

2. CH2N2, rt (81%)

53%

63

Scheme 3.5

O

CH2Cl2, 40 °C

O

62

H OPMP CO2Me

[Ru]-II (10 mol%)

H

64

Ireland–Claisen, RCM sequence: Z-enol ethers.

O HORO

O

O

OBn

BnO

O

1. KHMDS, TMSCl

MeO2C

2. TMSCHN2, MeOH

N

O N

OR

O 65a: R = (E )-CH2CH CHPr 65b: R = (Z )-CH2CH2CH CHEt 65c: R = (Z )-CH2CH2CH2CH CHMe

[Ru]-II (10 mol%)

O

MeO

(CH2Cl)2 (0.005 M) 80 °C O

n

BnO

66a: R = (E )-CH2CH CHPr (55%) 66b: R = (Z )-CH2CH2CH CHEt (56%) 66c: R = (Z ) -CH2CH2CH2CH CHMe (60%)

67a: n = 1 (89%) 67b: n = 2 (57%) 67c: n = 3 (58%)

Scheme 3.6 The synthesis of optically active medium rings using sequential Ireland–Claisen RCM technology.

O

NMe

i- Pr

S

Ph

O

1. BuLi; ClTi(Oi -Pr)3; i -PrCHO (81%) TESO 2. TESCl, imidazole, DMF (95%)

i -Pr

i-Pr

68

BuLi, allylbromide

S 69

92%

Ph NMe

2.

CO H n 2

DCC, DMAP, CH2Cl2

Scheme 3.7

n i -Pr

H O

O

NMe S

i -Pr

Ph

O

NMe S

[Ru]-II (8 mol%)

H

Ph

O PhCH3, (0.005 M) 110 °C

O

n

71a: n = 0 (47%)

72a: n = 0 (87%)

71b: n = 2 (67%)

72b: n = 2 (81%)

The synthesis of medium ring lactones.

O

NMe S

70 i -Pr

O 1. HF, pyridine (95%)

TESO

i -Pr

Ph

3.3 Laurencin

[Ru]-I, [Ru]-II, or [Ru]-V O

TBSO

CH2Cl2 (0.002 M), 40 °C 65%

73

Scheme 3.8

TBSO O 74

Triene cyclizations.

Linderman et al. have found that oxepines and oxocines can be efficiently generated from the RCM of acyclic tributylstannyl-substituted dienes using [Ru]-I catalyst [20]. As illustrated in Table 3.4, stannylated dienes gave very high yields of the corresponding oxocines regardless of the relative position of the olefins. In contrast to these results, attempts at employing t-butyl groups in the place of the stannane gave only polymeric material indicating that the stannane is not affecting the reaction through a steric effect. That the α-alkoxy stannane is synthetically useful is also beneficial (Scheme 3.9). Linderman et al. converted oxocene 83 into ketone 84 by first transmetallating with n-BuLi and then quenching the resulting organolithium with N,N-dimethyl benzamide to give 84 in 75% yield (+5% trans-isomer) [20].

3.3 Laurencin

In the first of a series of studies employing asymmetric enolate methodology to RCM precursors, Crimmins et al. employed an aldol-RCM strategy to synthesize medium ring heterocycles that are found in the red algae Laurencia sp. Central to the success of the strategy was the recognition that the vicinal diol, required for the synthesis of the natural product, would facilitate ring closure through a gauche effect placing the substrate in a favorable conformation for cyclization [21]. As outlined in Schemes 3.10–3.12, this strategy was used to effect the generation of oxepines, oxocines, and oxonenes. The synthesis of oxepine 88 is illustrated in Scheme 3.10 [22]. Using glyoxylate oxazolidinone 85 as a staging point, aldol condensation with acrolein provided 86 (de > 99%), and reductive removal of the auxiliary and RCM using [Ru]-I catalyst gave 88 in 95% yield. The use of this strategy in the synthesis of oxocines and a formal synthesis of laurencin is illustrated in Scheme 3.11 [22]. From acyloxazolidinethione 89, aldol condensation with 3-butenal provided 91 after reductive removal of the auxiliary. As with the synthesis of oxepine 88, the generation of oxocine 92 was successfully carried out using [Ru]-I catalyst. Subsequent manipulation of the protecting group enabled the interception of 93, an intermediate used in the synthesis of (+)-laurencin by Holmes et al. [23]. As an illustration of the importance of substitution on the ring-closing reaction, earlier attempts to carry out a related

95

96

3 Synthesis of Medium-size Oxygen Heterocycles by Ring-closing Alkene Metathesis Table 3.4

Stannyl ketals in RCM.

R

R

m

Bu3Sn

Bu3Sn

PhH (0.01 – 0.015 M), rt

n

Entry

Diene

Product

Bu3Sn

n Yield (%)

O

O 75

1

m

O

[Ru]-I (3 – 10 mol%)

O

Bu3Sn

76

74

78

38

80

84

82

96

H3C O

O Bu3Sn

77

2

Bu3Sn

H3C O

O Bu3Sn

Bu3Sn

3

79 H3C

O Bu3Sn 4

O 81

Bu3Sn

H3C

H3C

n -BuLi, THF

O

−78 °C; PhCONMe2 75%

Bu3Sn 83

Scheme 3.9

O Ph O

84

Stannyl ketal manipulations.

cyclization on a substrate lacking the substitution in 77 gave only dimers and other oligomers [24]. The aldol-RCM strategy was also used to generate oxonene 97 (Scheme 3.12) [22]. Crimmins et al. also used the strategy outlined above in the synthesis of isolaurallene [25]. The synthesis of the diene cyclization precursor is shown in

3.3 Laurencin O

O O

N

85

O

1. LiBH4, MeOH, THF, 0 °C (66%)

N

CH2Cl2, −78 °C

O

2. Ac2O, Et3N, CH2Cl2, DMAP 0 °C to rt (70%)

O

CHO 40%

Ph

O

OH O

Bu2BOTf, Et3N

86 Ph (de > 99%) OAc

OAc [Ru]-I (7 mol%)

OAc 87

88

Asymmetric enolate-RCM sequence: oxepines.

S

O O BnO

OAc O

CH2Cl2, 40 °C (0.003 M) 95%

O

Scheme 3.10

97

N

89

CHO 65%

O BnO

OAc 2. Ac2O, Et3N, CH2Cl2, DMAP

O

0 °C to rt (67%)

90

OAc

OAc

1. LiBH4, MeOH, Et2O, 0 °C (86%) O

O BnO

OAc

[Ru]-I (7 mol%) CH2Cl2, 40 °C (0.003 M) 97%

N

CH2Cl2, −78 °C

Ph

O

OH O

TiCl4, (−)-sparteine

O

Ph (de > 99%)

BnO

1. DDQ, CH2Cl2, H2O (60%) 2. TBDPSCl, imidazole, DMF (66%) 3. K2CO3, MeOH (97%)

O TBDPSO

92

OH

91

Ref. 23

OH

93

Br O OAc (+)-Laurencin

Scheme 3.11

Formal synthesis of (+)-laurencin.

S

O O

N

O

TiCl4, (−)-sparteine

OH O N

CH2Cl2, −78 °C

94

CHO 50–60%

Ph

O 95 (de > 99%)

1. LiBH4, MeOH, Et2O, 0 °C (70 – 80%) O

2. Ac2O, Et3N, CH2Cl2, DMAP 0 ˚C to rt (89%)

Ph OAc

OAc

[Ru]-I (7 mol%) OAc

O

O

CH2Cl2, 40 °C (0.003 M) 89% (+ 10% dimer)

96

Scheme 3.12

O

OAc

97

Asymmetric aldol, RCM reactions: oxonenes.

Scheme 3.13. From benzyl glycidyl ether 98, ring opening and esterification gave 100 after attachment of oxazolidinone 99. Aldol condensation gave 101 in 58% yield along with 20% of a mixture of two other isomers. Following the manipulatoin of the protecting group and oxidation, the second alkene was inserted into the substrate using Brown’s allyl borane chemistry to give the cyclization precursor 103.


98

O

O H

O

BnO

1. MeMgBr, CuI (100%) 2. BrCH2CO2H, NaH, THF (85%) 3. PivCl; n-BuLi, 99, THF (79%) O

98

HN

O

BnO

N

Me

O O

TiCl4, i-Pr2NEt CHO, −78 °C 58%

Bn 100

O

BnO

Xc OH

Me 101

Xc

O

Bn 99 1. LiBH4, Et2O, CH3OH, 0 °C (61%) 2. TBSOTf, 2,6-lutidine, CH2Cl2, 0 °C (83%) 3. Na, NH3, THF, −33 °C (74%)

Me

O

HO 102

Me

1. (COCl)2, DMSO, Et3N

OTBS

d

2. (2- Icr)2B-allyl, Et2O, −78 °C;

OTBS

OTBS

O

OTBS

TIPSO

103

H2O2 40% (2 steps)

Scheme 3.13

Me

O

OTBS OTBS

TIPSO

Synthesis of isolaurallene.

Me [Ru]-I (5 mol%)

OTBS

Me

OTBS

HO

O

TIPSO

O OH

CH2Cl2, (0.003 M), 40 °C 89% 104

103

105 Br

X

Me

H

O

H

H O

HO

H 106

Scheme 3.14

RCM cyclization to the isolaurallene oxonene.

RCM to isolaurallene’s oxonene was successful using [Ru]-I catalyst (Scheme 3.14) [25]. Unfortunately, the subsequent conversion of 105 into the requisite furan 106 for the completion of isolaurallene using bromonium ion activation was unsuccessful. Having been unsuccessful in the cyclization to 106, Crimmins et al. took an alternate approach to isolaurallene involving a hydroxy-epoxide cyclization to the furan [25]. To this goal, triene 107 was synthesized using an approach that was similar to that used to generate 103. Attempted cyclization of 107 to oxonene 108 again illustrated some of the difficulties associated with medium ring formation (Scheme 3.15). When 107 was subjected to [Ru]-I catalyst, a mixture of the desired 108 and dihydropyran 109 from competitive cyclization onto the disubstituted olefin was isolated. Attempts at optimization involving the examination of the effect of alcohol protecting group (R1 , R2 , R3 ) on the cyclization reaction were only moderately successful. Taking advantage of the mild nature of the RCM conditions and the ruthenium catalysts, Crimmins et al. were able to successfully complete the synthesis of isolaurallene by incorporating an epoxide rather than an alkene into the RCM [25]. This resulted in the generation of oxonene 114 in 94% yield (Scheme 3.16). Epoxide 114 was subsequently converted into isolaurallene.

3.3 Laurencin OR2

OR2 Me

O OR3

R1O

[Ru]-I (5 mol%)

Me

Me

O

+

OR3

R1O

99

OR3 O

R1O

CH2Cl2 (0.003 M) 40 °C 107

Scheme 3.15

108 (22 – 54%)

Oxonenes from trienes.

OBn Me

O

1. Ac2O, pyridine, CH2Cl2 (96%) 2. DDQ, CH2Cl2

OH

BnO

109 (45 – 51%)

OH Me

O

3. NaBH4, EtOH (67%)

OH O

(+)-Diethyl tartrate OAc

TESO

112

2. CH2Cl2, CH3OH, PPTS 81%

111

110 Me

1. Et3SiOTf, CH2Cl2, 2,6-lutidine OAc

HO

OH

O Me

O

O [Ru]-I (5 mol%)

OAc

TESO Ti(Oi - Pr)4, t-BuOOH 97%

CH2Cl2 (0.003 M), 40 °C 94%

Br

H

O OAc

TESO

114

113 (de > 98%) Me

Me

Br O

H

H O H

Isolaurallene

Scheme 3.16

Epoxy diene RCM to isolaurallene.

In addition to utilizing aldol-RCM sequences to access medium rings, Crimmins et al. have also generated medium rings by coupling glycolate oxazolidinone alkylations with RCM reactions [26]. As illustrated in Scheme 3.17, Crimmins and Emmitte’s synthesis of (+)-laurencin began with the allylation of 118 to give the cyclization precursor 119. Cyclization using [Ru]-I catalyst gave oxocine 120 in 94% yield. Subsequently, 120 was transformed into laurencin in an additional 10 steps. Other research groups have also employed asymmetric glycolate alkylation/RCM strategies to synthesize medium rings that are relevant to the laurencin family. For example, Lee et al. have utilized such a strategy in the synthesis of (–)-isoprelaurefucin [27]. As outlined in Scheme 3.18, the allylation or prenylation of the enolate derived from glycolate amide 121 provided the corresponding alkylated RCM precursor 122 with high diastereoselectivity. Of interest was that the nature of the secondary hydroxyl-protecting group on 121 was important in this reaction. With TIPS (triisopropyl silyl) rather than PMB (para-methoxy benfyl), the other diastereomer predominated. Subsequent to the alkylation, RCM using [Ru]-I catalyst gave the desired oxepine 123 in 65% overall yield for the two steps. Further manipulation of 123 resulted in the generation of (–)-isoprelaurefucin. Clark et al. have also employed a glycolate alkylation–RCM strategy to synthesize medium ring ethers (Scheme 3.19) [28]. Specifically, their work to (+)-laurenyne’s

OH


100

OH

NaH

HO2C

O

BrCH2CO2H THF 88%

OBn 115

OBn 116

LiN Bn

Bn

BnO

PivCl, Et3N, Et2O

O

N

O

O

O

O THF

O THF

118

117 76%

NaN(TMS)2, I

Bn

THF, PhCH3 −78 °C to −45 °C 71%

BnO

N OH O H 119

Br

Scheme 3.17

Bn

[Ru]-I (5 mol%) CH2Cl2 (0.005 M) 40 °C 94%

O

N BnO

OH H

O

O

O THF O

120

O H OAc H (+)-Laurencin

Synthesis of (+)-laurencin.

R LiHMDS, THF R Me2N

O O

OBn H

121

OPMB

I

R

−78 °C, 30 min

R

Me2N H

O

OBn H

O OPMB 122a: R = H (dr = 10 : 1) 122b: R = Me (dr = 15 : 1) Br

[Ru]-I (30 mol%)

10 steps

OBn CH2Cl2 (0.003 M), 40 °C Me2N O 0.5 h for R = H H H O OPMB 5 h for R = Me 123 65% (2 steps) Scheme 3.18

H O

O H H H Br (−)-Isoprelaurefucin

Synthesis of (–)-isoprelaurefucin.

oxocine ring involved the coupling of a diastereoselective enolate allylation to 125 with RCM to give 126 using either the [Ru]-I or [Ru]-II catalyst. With [Ru]-II catalyst, the yield for the RCM transformation was enhanced when the alcohol was protected as a TBDPS (tert-butyl diphenyl silyl) ether. For the cyclization using [Ru]-I catalyst, the yield was enhanced when the catalyst was added to the reaction in several small portions. Adsool and Pansare coupled diastereoselective morpholine dione addition chemistry with diene RCM using [Ru]-II catalyst to generate oxocine 133 (Scheme 3.20)

3.3 Laurencin

H O PMP

H

O

O O

H

Me Ph N Me OH

1. LiCl, LDA, THF, −78 °C; I (75%)

H O

2. LDA, BH3·NH3, THF, 0 °C (88%)

124

PMP

H

O

101

OH O

H

125

Catalyst conditions OTBDPS

1. TBDPSCl, imidazole, DMF, rt (98%)

HO O

2. [Ru]-II (2 mol%), PhCH3, 80 °C (99%) PMP

H

O

OH

HO TBAF, THF, rt O 91% PMP O H H 127

H 126

Entry

Catalyst

Conditions

Yield

1

[Ru]-I (5 mol%)

CH2Cl2, rt

48(%)

2

63(%) [Ru]-I (10 mol%) CH2Cl2, rt, catalyst added (5 × 2 mol%) [Ru]-II (5 mol%) PhCH3, 80 °C 75(%)

3

Scheme 3.19

Alkylation–RCM strategy to (+)-laurenyne. OH PMPO

Ph

N

m-CPBA, CH2Cl2

O

O

Ph

N

80%

O O

O

128

130

N

KH, THF 60%

O

129 Ph

N TMS BF3 • OEt2, CH2Cl2 60%

O

O

[Ru]-II (10 mol%) CH2Cl2 (0.017 M), rt 85% PMPO

O 132

Ph

N

O

O

O 133

OPMP

1. Na/NH3 (96%) 2. LiAlH4 (84%) 3. NaIO4, MeOH, H2O 4. Ce(NH4)2(NO3)6 85% (2 steps)

Br

O Ref. 30

O

Ph O OPMP OH O 131

O AcO

PMPO 134

(+)-Laurencin

Scheme 3.20

Formal synthesis of laurencin.

[29]. Oxocine 133 was converted into ketone 134, thus intercepting an intermediate that had previously been synthesized by Overman et al. on their route to laurencin [30]. In addition to the key RCM reaction, Adsool and Pansare also effectively utilized hemiketal 131 in a coupling reaction with allylsilane to give the RCM precursor.

102


O N

O

BnO

H O

H

OH

1. LiBH4, MeOH Et2O, 0 °C (92%)

O Me

HO

Me +

O 2. [Ru]-II (5 mol%) BnO conditions

H OBn 135 [Ru]-I or [Ru]-II (5 mol%) CH2Cl2 (0.01 M), 40 °C

BnO

H

BnO Me 136

BnO

OBn

Conditions

2

H HO

Me

2

Me OBn 138

[Ph3PCH2OBn]Cl O

t-BuOK, −78 °C

MeO2C TESO Me 139

Me

HH Me

O AcO

HH R O Me H R = OH, Asbestinin 12 Me

Me HH O

AcO

R = H, 11-Acetoxy-4 -deoxyasbestinin D

Scheme 3.21

+

O

MeO2C TESO Me 140 (140/141 = 3 : 1)

Me Me

BnO O

MeO2C TESO Me 141

rt, 2 h

Me AcO

Me Me

Me

OHC

O Me

PhH (0.002 M), 80 °C (89%, 136/137 > 15 : 1)

hn, PhSSPh

BnO N

CH2Cl2 (0.002 M), 40 °C (75%, 136/137 = 3 : 1)

137 [Ru]-II (5 mol%) PhH (0.002 M) 80 °C (89%, 136:137 = >15 : 1)

O O

H

Me

H O

HH AcO Me OAc Ophirin B

Me HH

Me

O BnO HH Me MeO2C TESO 142 80% (overall yield from 139)

Synthesis of ophirin B.

3.4 Eunicellins/Eleutherobin

In another beautiful illustration of the power of their RCM methodology, Crimmins et al. have coupled asymmetric glycolate alkylation/RCM with intramolecular Diels–Alder cycloadditions to synthesize members of the eunicellins and asbestinins [31]. As outlined in Scheme 3.21, the synthesis of ophirin B began with diene 135 that was obtained from Crimmins et al. previously discussed glycolate alkylation protocol [32]. When diene 135 was exposed to either [Ru]-I or [Ru]-II catalyst at 40 ◦ C, exclusive generation of dimer 138 was observed. Reasoning that an unfavorable conformation about the oxazolidinone might be hindering cyclization, Crimmins et al. reduced the oxazolidinone and studied the cyclization chemistry of the resulting alcohol. Pleasingly, the alcohol from 135 underwent the desired cyclization to oxonene 136 when [Ru]-II catalyst was used at 40 ◦ C but a large quantity of dimer 137 as it was also formed. The authors subsequently showed that the amount of dimer could be minimized by carrying out the reaction at 80◦ C in benzene. That compound 137 was the kinetic product was demonstrated through its conversion into the desired product 136 by resubjecting 137 to [Ru]-II catalyst. Having solved the oxonene generation, the synthesis of the eunicellin

3.4 Eunicellins/Eleutherobin

HO

BnO

H

H O

Me

OTIPS

OH

OTIPS

O OBn

BnO

Scheme 3.22

H

AcO

Me OBn 144

OH 145

HH Me OAc OAc Astrogorgin

Synthesis of astrogorgin.

RO O

OAc O

143

RO

Me H H

[Ru]-II (5 mol%) PhH (0.01 M), 80 °C 96%

O O

H

[Ru]-I (10 mol%) CH2Cl2 (0.03 M), 40 °C additive

O

OH

O O

146 Without additive R = TBS (30%) R = Me (33%) Additive: Ti(Oi -Pr)4 R = TBS (83%) R = Me (63%)

Scheme 3.23

103

Generation of Eleutherobin’s oxonene ring.

six-membered ring from a Diels–Alder strategy was examined next. To this goal, 136 was converted into enal 139. When 139 was subjected to benzyloxymethylenetriphenylphosphorane, a 3 : 1 mixture of alkene isomers 140/141 was formed from which only the E/E-stereoisomer 140 underwent the desired intramolecular Diels–Alder cycloaddition to give 142 upon standing at room temperature for 2 hours. Subsequently, it was found that the undesired Z/E-isomer 141 could be equilibrated into a 1 : 1 mixture of dienes and ultimately into the cycloadduct by exposing it to light and PhSSPh. Utilizing this protocol on the mixture until all the diene had reacted resulted in an 80% overall yield of adduct 142. Tricycle 142 was converted into ophirin B in an additional six steps. Crimmins and Ellis employed a very closely related approach in their generation of 11-acetoxy-4-deoxyasbestinin D and asbestinin-12 [33]. With the exception that an allylic ether was employed in the RCM reaction, Crimmins et al. employed a nearly identical strategy to generate astrogorgin (Scheme 3.22) [34]. Using the tubulin stabilizing natural product eleutherobin as an inspiration, Kaliappan and Kumar have investigated the generation of medium rings that are fused to ribose subunits using RCM [35]. Initially targeting eleutherobin’s oxonene ring, they exposed 145 to [Ru]-I catalyst and realized low yield of cyclic product (Scheme 3.23). Following the discovery that the alcohol protecting group had little impact on the overall efficiency of the reaction as substrates with R = TBS and Me performed similarly, the effect of additives on the reaction was examined and

104

3 Synthesis of Medium-size Oxygen Heterocycles by Ring-closing Alkene Metathesis R1 R2 HO n OH

O

R1 R2 HO

[Ru]-I (10 mol%) CH2Cl2 (0.03 M), 40 °C Ti(Oi-Pr)4 (10 mol%)

O

n OH

O O

147 148a: n = 0, R1 = H, R2 = OMe (0%, no reaction) 148b: n = 0, R1 = H, R2 = OH (84%) 148c: n = 1, R1 = OMe, R2 = H (68%) 148d: n = 3, R1 = H, R2 = OMe (82%)

Scheme 3.24

The generation of bicyclic fused ethers using RCM.

NO2

NO2

O2N MgBr n

n O 150a: n = 1 (78%) 150b: n = 2 (75%)

O 149

[Ru]-I (5 mol%) PhCH3 (0.004 M), 80 °C

n O 151a: n = 1 (60%) 151b: n = 2 (65%)

Scheme 3.25 Synthesis of benzo-fused medium rings using sequential Michael-RCM technology.

it was found that catalytic amounts of Ti(Oi-Pr)4 dramatically increased the yield of 146. Presumably, Ti(Oi-Pr)4 affects the conformation of the substrate through chelation [36]. Kaliappan and Kumar were also able to successfully synthesize oxocines and oxecines using this strategy (Scheme 3.24) [35]. Interestingly, the synthesis of the corresponding oxepine required that the alcohol not be protected, presumably a result of the influence of allylic substitution on RCM. As with the oxonenes, the use of Ti(Oi-Pr)4 as an additive was crucial to the success of these reactions. 3.5 Helianane

Medium-sized oxygenated rings that are fused to aromatic systems have also been targeted using RCM. Deb et al. have developed an approach to the oxocines and oxonenes that are present in the helianane family, which couple β-nitrostyrene Michael addition chemistry with RCM (Scheme 3.25) [37]. For example, the addition of Grignard reagents containing a terminal olefin to the nitrostyrene 149 resulted in the generation of the corresponding Michael adducts 150. RCM cyclization gave the corresponding medium rings 151 in reasonable yields. Macias et al. have developed a stereoselective route to heliannuol A using diene metathesis to generate the requisite oxocine (Scheme 3.26) [38]. The cyclization precursor was obtained from the sequential propargyl ether formation and allyl Grignard addition to the acetophenone derivative 152. Reductive removal of the

3.6 Octalactin A O

Cl

BnO

1. Et3SiH, BF3 •OEt2 (82%)

MgBr (83%)

2.

OH

OH

CuCl2, DBU (95%) BnO

1.

2. Pd /CaCO3, Pb, quinoline, H2 (92%)

O

152

105

153

O BnO

[Ru]-I (10 mol %)

BnO

CH2Cl2 (0.01 M), rt 91%

O 154

Scheme 3.26

CF3, NaHCO3, Oxone (72%) HO

1. 155

2. LiAlH4 (75%)

O

O

3. Pd /C, H2 (99%)

(±)-Hellianuol A

Synthesis of heliannuol A. O

PMBO

OPMB

O

1. TBSO

O

O P(OEt)2

156

H

157 Ba(OH)2, THF

HO

O

OPMB

O

2. HF, CH3CN 81%

159 (PhCHO)2SmI•SmI3 (30 mol%) 96%

158

O PMBO

OH

O

O OH

160

OPMB

[Ru]-II (20 mol%)

HO

PMBO

Ti(Oi -Pr)4 (10 mol%) CH2Cl2 (0.01 M), 40 °C

O

OH

OPMB

O

OH

OH O

161

Octalactin A

O O

PMBO

OH 162

Scheme 3.27 Synthesis of the octalactin A core using a sequential Evans–Tishchenko/RCM strategy.

benzylic alcohol and Lindlar reduction of the alkyne resulted in the formation of the cyclization precursor 154. RCM using [Ru]-I catalyst provided heliannuol A after epoxidation, reduction, and removal of the benzyl protecting group.

3.6 Octalactin A

Medium ring lactone containing natural products have also been the target of RCM reactions. In their work targeting octalactin A, Hulme et al. utilized a sequential Evans–Tishchenko ester formation/RCM sequence to generate octalactin’s oxocine ring (Scheme 3.27) [39]. From phosphonate 156, Horner–Wadsworth–Emmons coupling with methyl ketone 157 gave enone 158. The use of the Evans–Tishchenko reaction between 158 and 159 gave the ester cyclization precursor 160 as a single diastereomer in 96% yield. RCM using [Ru]-II catalyst resulted in the generation of lactone 161 and cyclopentene 162 (1 : 1 mixture). Although the yield of the desired cyclization product was not improved by using other RCM catalysts, the authors


106

O

6 steps

O O

O

O O 165 (69%) + 166 (9%)

O

HO O

HCl, H2O HO

O

[Ru]-V (10 mol%) CH2Cl2 (0.002 M), 40 °C

THF 90%

O

HO

O Herbarumin I

O

O

O 164

163

CH2Cl2 (0.002 M), 40 °C 86%

Scheme 3.28

HCl, H2O HO

O

[Ru]-II (10 mol%)

O

O O 166

THF 47%

O

HO

O 167

Synthesis of herbarumin I. CO2H

O

O

5 steps

O

HO

OH

O O

O 163

O O

O

CH2Cl2 (0.002 M), 40 °C

O O

O

OMOM O

170 (E/Z > 95:5)

OMOM O

168

[Ru]-V (10 mol%)

Scheme 3.29

OMOM DCC, DMAP, rt 67%

HCl, H2O

HO

MeOH, 60 °C 84%

HO

O

169

OH O

Herbarumin II

Synthesis of herbarumin II.

were able to overcome this problem by utilizing Ti(Oi-Pr)4 as a cocatalyst giving 161 in 70% yield. This additive presumably affects the conformation of the cyclization precursor through chelation [36]. 3.7 Microcarpalide and the Herbarums

Concerned with the generation of the thermodynamically less stable E-isomer of herbarumin I, Fürstner et al. examined the ability of both [Ru]-V and [Ru]-II catalysts to effect the cyclization of diene 164 (Scheme 3.28) [40]. Impressively, [Ru]-V catalyst gave E-alkene isomer 165 as the major product in 69% yield along with 9% of Z-isomer 166. The authors found that the amount of 166 did not increase with time indicating that the observed E : Z ratio was the result of a thermodynamic control. Interestingly, the more reactive [Ru]-II catalyst resulted in the exclusive generation of Z-isomer 166.

3.7 Microcarpalide and the Herbarums O

KHMDS, THF, −50 °C;

O

O

O

O

O

O

PhO2S

N

Ph

O

OH O

171

172

1. Sorbic acid chloride, pyridine (76%)

HO

2. HCl, MeOH, H2O (80%)

HO

O

O O

O

Pinolidoxin

Scheme 3.30

Conversion of the herbarumin I precursor into pinolidoxin.

Fürstner et al. attempted the generation of herbarumin II from herbarumin I via the oxidation of a herbarumin I enolate, which resulted in the generation of the C-2 herbarumin II epimer. For this reason, the hydroxyl group was incorporated into the RCM precursor and herbarumin II was generated using [Ru]-V catalyst as illustrated in Scheme 3.29 [40]. Of note was the formation of less than 5% of the corresponding Z-alkene isomer during the cyclization event. In addition to forming the herbarumins, Fürstner et al. were able to synthesize pinolidoxin by carrying out the aforementioned hydroxylation reaction on herbarumin I precursor 71 (Scheme 3.30) [40]. This work effectively corrected previously misassigned stereochemistry of pinolidoxin [41]. Davoli et al. have combined a diastereoselective homologation reaction that was initially developed by Matteson et al. with diene RCM to generate (– )-microcarpalide [42]. As illustrated in Scheme 3.31, the use of Matteson et al. chemistry resulted in the generation of alcohol 177 after oxidation of 176. Esterification with acid 178 and RCM using [Ru]-I catalyst gave a 2 : 1 mixture of alkene isomers favoring the E-isomer. Similar results were observed with [Ru]-V catalyst. Gurjar et al. have also utilized diene RCM in the generation of the medium ring lactones microcarpalide and herbarumin III (Scheme 3.32) [43]. To microcarpalide, esterification of 180 with 181 led to the cyclization precursor 182 in 76% yield. RCM using [Ru]-I catalyst gave the corresponding 10-membered lactone and microcarpalide after the removal of the benzyl and MEM (methoxy ethoxy methyl) protecting groups. The obvious difference between the work of Gurjar et al. and Davoli et al. lies in the choice of alcohol protecting groups, MEM and Bn, respectively, again demonstrating the dramatic effect of substrate substitution on the cyclization reaction. Although a related sequence of reactions was utilized to generate herbarumin III, of note was the presence of the free allylic hydroxyl group in the

107


108

O B O Cl

−78 °C to reflux 70%

173

Cl Cl

O B O

BnOH, BuLi, THF

OBn 174

Cl Li

ZnCl2, THF −100 °C to rt 64%

175

MgBr

B O OBn O

−78 °C to rt 74%

176

B O OBn O

O O

HO H2O2, NaOH

O

178

O

O

DCC, DMAP

0 – 45 °C 90%

OH 177

O

85%

OBn

OBn

1. [Ru]-I (17 mol%) CH2Cl2 (0.005 M), 40 °C (92%, E/Z = 2 : 1)

179

OH OH

2. TiCl4, CH2Cl2, 0 °C (66%)

Scheme 3.31

O

O O OH Microcarpalide

Synthesis of (–)-microcarpalide.

OBn OBn

DCC, DMAP

+

HO O

180

OBn

181

OH OMEM

76%

1. [Ru]-I (20 mol%) CH2Cl2 (0.003 M), 40 °C (67%) 2. TiCl4, CH2Cl2, 0 °C (76%)

Scheme 3.32

BnO O

182

O OMEM

OH OH O O OH Microcarpalide

Diene cyclization to microcarpalide.

cyclization reaction and that [Ru]-II catalyst was much more successful than [Ru]-I catalyst (Scheme 3.33) [43]. When compared with Fürstner et al. work (Scheme 3.28), it was interesting that [Ru]-II catalyst gave the desired E-olefin in this cyclization.

3.8 Marine Ladder Toxins

OH OPMB HO

2,4,6-Trichlorobenzoylchloride

+ O

184

183

109

PMBO O

DMAP, THF 82%

O 185

HO

1. DDQ, CH2Cl2, H2O (94%)

O

2a. [Ru]-I (20 mol%) CH2Cl2 (0.004 M), 40 °C (36%) or 2b. [Ru]-II (1 mol%) CH2Cl2 (0.004 M), 40 °C (78%) Scheme 3.33

H O

H

O

Diene cyclization to herbarumin III.

H

H

OBn

H OBn

[Ru]-I (1 mol%) CH2Cl2 (0.04 M), rt 97%

OBn

186 H O

Scheme 3.34

O Herbarumin III

H

O

H

H

OBn 1. Na/NH3, THF/EtOH, −78 °C OMe OBn O , p -TsOH 2. MeO H H OBn OMe 187 59% O

H O

O PMP H OH 188

A OH

O B

H

H OBn

I OPMB 189

Synthesis of the ciguatoxin A-ring.

3.8 Marine Ladder Toxins 3.8.1 Ciguatoxin

RCM has played an important role in the synthesis of members of the ladder toxin family of natural products. Arguably, the most impressive use of RCM to generate medium rings in the total synthesis of ladder toxins has come from the Hirama laboratory and their synthesis of ciguatoxin. Their efforts to synthesize ciguatoxin are outlined in Schemes 3.34–3.40, which began with the synthesis of the A-ring from C-glucoside 186 (Scheme 3.34) [44]. Cyclization using 1 mol% of the Grubbs catalyst [Ru]-I gave oxepine 187 in 97% yield. Reductive removal of the benzyl ethers was accomplished in 59% yield after the generation of PMP (para-methoxyphenyl) acetal 188. The allylic ether in 187 is sensitive and was partially cleaved (21%) when exposed to Na/NH3 . Additional protecting group manipulation gave the A,B-precursor 189.


110

OH H

HO HO

O

O

H

1. NaIO4, MeOH

PMP O

2. Ph3PMeBr, t-BuOK, THF (72%, 2 steps)

H 190

3.

t-BuO

O O

O

PMP

LDA, THF, −78 °C;

O

CHO

H

91%

191

O Br, NaH, THF, DMF

t-BuO

(99%) H O

HO

t-BuO2C

O

[Ru]-I (7 mol%) PMP

CH2Cl2 (0.006 M), 40 °C t-BuO2C 85%

O

H 192

Scheme 3.35

H

E

t-BuO2C H

H

O H 194

O

H

O B

O

E

t-BuO2C

O

H 193

H O

O H

PMP

O

H 194

I OPMB

H

PMP LDA, THF, HMPA, −78 °C 51%

O

H

H

O H

H [Ru]-I (7 mol%) O

H

O

O

H

H

O

O

H

HO HO OBn PMB

O

H

H O

H O H H

H

H

O

H O

O

CO2t-Bu O O H H OPMB H OBn 195 (dr = 6 : 1) H

Si(i-Pr)2 4 Sn O MeLi, Et2O, −78 °C Si(i-Pr)2 94%

O

H

H

O

H

198 H

2. Et3SiH, BF3 • OEt2, CH2Cl2, −78 °C to −30 °C (98%)

A O

H

H O

H

O B

H

H OBn

H

C O

H O D

H

PMP

H O H H

HO O H HO OBn PMB

O Si(i-Pr) 2 O Si(i-Pr)2 O

1. CSA, CH(OMe)3, CH2Cl2 (64%)

Scheme 3.36

H O

PMP

H

O H H 189 OBn

H O

CHO O H H OPMB OBn 196

CH2Cl2 (0.009 M), 35 °C 96%

H O

E

Synthesis of the ciguatoxin E-ring.

A H O

HO

O

O

Si(i-Pr)2 O Si(i-Pr)2

197 O

H

H

O H OH OBn H

E O H H

O

O

H O

H O H H

O Si(i-Pr) 2 O O Si(i-Pr)2

199

Si(i-Pr)2 O Si(i-Pr)2

200

Synthesis of the ciguatoxin C,D-ring.

The generation of the eight-membered ciguatoxin E-ring coupled glycolate aldol chemistry with RCM and began with protected d-glucose (Scheme 3.35) [44]. Oxidative cleavage and Wittig reaction gave 191 after glycolate ester formation. Aldol condensation gave 191 as a mixture of three diastereomers. RCM was carried out on the mixture using [Ru]-I catalyst to give the eight-membered ring substrate 193 in 85% yield. Following their separation, each of the diastereomers was independently processed to the E-ring coupling precursor 194. Using RCM, Hirama, Inoue, et al. generated the seven-membered ciguatoxin D-ring after coupling the A,B- and E-precursors (Scheme 3.36) [44]. Alkylation of the E-ring glyoxylate enolate from 194 with A,B-ring iodide 189 resulted in 195


CH3 NAPO CH3 H H O H H OH O H I A B E H O H CH3 C D J O O O O ONAP O H H H H H H K H H H SPh NAPO HO O L OEE H O CO2Me M H3 C 201 CH3

111

H

ONAP =

O

CH3 CH3 H NAPO H O H I H H O H H G H Bu3SnH, AIBN, PhCH3, 85 °C O H CH3 O O A B E J O C D ONAP OH H O H K H O H HO H H H H HO CH3 NAPO O L MeO C 202 CH H 2 3H O NAPO OEE O H M H H H3C I H H O H H G H O H CH3 O 1. i -Bu AlH, CH Cl , −80 °C O CH 2 2 2 A B E 3 J O C D ONAP 2. Ph PCH , 0 °C O H H O 3 2 H K H O H H O H H H H O H NAPO O 3. CSA, MeOH, rt (41%, 4 steps) H L O 4. SO3 pyridine, Et3N, DMSO M 203 H3C CH3 5. Ph3PCH2, 0 °C (100%, 2 steps) H

1. [Ru]-I (30 mol%) CH2Cl2, 40 °C (90%) 2. DDQ (63%)

Scheme 3.37

CH3 CH3 H HO O H I O H H OH H H G H O H CH3 O A B E J O C D OH OH H H F O K H O O H H H H H HO H HO O H L O Ciguatoxin M H3 C CH3 H

Radical cyclization–RCM sequence to ciguatoxin.

CH3 NAPO Me H O O H H OH H H I A B O E H O H CH3 C D J O O O ONAP O H H HO H H H SPh H K H H NAPO H O O L H O CO2R′ M H3 C 204a: R′ = Me CH3 R 204b: R′ = C6F5 H

R = H, ONAP Bu3SnH, AIBN, PhCH3, 85 °C

H H O D

H NAPO CH3 O OH H H H H H O H Me H H 205a: R′ = Me (27%) O

E

O

205b: R′ = C6F5 (7%) CO2R′ CH3 NAPO CH3 H O H H H I H H O H H G H O H CH3 O O A B E J O C D ONAP OH H H O K H O H HO H H H H O H NAPO O R′O2C H L O M 206a: R′ = Me (54%) H3C CH3 206b: R′ = C6F5 (74%) R 1. i-Bu2AlH, CH2Cl2, -90 °C 2. Ph3PCH2, 0 °C

CH3 NAPO CH3 H O H H I O H H OH H O H CH3 G H O A B E J O C D ONAP OH H H O K H O O H H H H H H O H NAPO O 92% (2 steps) H L O 207 M H3C CH3 CH3 R HO CH3 H 1. [Ru]-I (30 mol%), CH2Cl2, 40 °C (90%) O H H H H I H H 2. DDQ, CH2Cl2, H2O (63%) O H O H CH3 G H O O A B E J O C D OH OH H H F O K H O O H H H HO H H H HO O H L O R = OH: Ciguatoxin M H3 C (CTX1B) CH3 R = H: CTX3C R H

Scheme 3.38

Radical cyclization–RCM sequence to ciguatoxin.

112


NAPO CH3 H O H H G H O

CH3

H O H CH3 J O ONAP OH H H K H H H O L O O H 208 M H3 C CH3 ONAP 1. [Ru]-I, CH2Cl2, 40 °C (94%)

H

H H O H O E B C D O H H O H H H NAPO

A O

I

2. DDQ, CH2Cl2, H2O (33%)

Scheme 3.39

H

O

H

210

O

H HO HOH D B C A O OH H H OH

Radical cyclization–RCM sequence to ciguatoxin analogs.

H

H OBn

H OBn

Scheme 3.40

CH3 HO CH3 H O H I O H CH3 H OH G H J O E OH OH H H 8 K H H H O O H O H L H O 209 M H3 C CH3 OH

OBn

[Ru]-I (5 mol%) CH2Cl2 (0.01 M), rt 95%

O H

H

211

O

H

H OBn

OBn OBn

Diastereoselective RCM to the ciguatoxin A-ring.

as a 6 : 1 mixture of diastereomers. Following the manipulation of protecting group and oxidation state, vinyllithium addition to 196 gave the allylic alcohol RCM precursor 197 as a mixture of four diastereomers. Cyclization using [Ru]-I catalyst (7 mol%) provided a 96% yield of the D-ring oxepine as 198. All four diastereomers were converted into enone 199 that was then transformed into the A–E ciguatoxin subunit 200 using an acid-catalyzed cyclization and reduction sequence. Hirama et al. were able to combine the A–E and H–J precursors by using a thioacetal/free-radical cyclization strategy to give the G-ring oxepine 202 (Scheme 3.37) [45]. Additional manipulation resulted in the diene cyclization precursor 203. In what is arguably the most impressive demonstration of RCM to medium rings (four alkenes, nine-membered ring formation), the cyclization of 203 using [Ru]-I catalyst gave the F-ring and the entire ciguatoxin skeleton in 90% yield. In contrast to this result, the attempted cyclization of 203 with [Ru]-II catalyst resulted in a lower yield of the oxonene and the generation of significant quantities of a by-product came from a competitive RCM–ROM (ring-opening metathesis) reaction onto the E-ring alkene [46]. The synthesis of ciguatoxin was completed after the global removal of the napthyl protecting groups. Hirama et al. also utilized dienyl substrate 204a as a precursor to the F,G-ciguatoxin ring system (Scheme 3.38) [47]. Although the use of 204a enabled a more convergent coupling sequence, it suffered from a competitive free-radical cyclization reaction to generate the pyranyl substrate 205a. Subsequently,


H

O

H

O

H

H OBn

H

O

H

OBn

H O H

OBn

CH2Cl2 (0.009 M), rt 100%

212 Scheme 3.41

H

[Ru]-I (50 mol%)

A O

H

O B

H

H OBn

H

C O

H

113

OBn

H O D H

213

Generation of the ciguatoxin A-ring.

α,β-unsaturated pentafluorophenyl ester 205b was used as the radicalphile resulting in a 74% yield of the desired product 206b along with 7% of pyran 205b. Compounds 206a and 206b were converted into the cyclization precursor 207 as illustrated. RCM using [Ru]-I catalyst as before gave 51-hydroxy CTX3C and CTX1B (ciguatoxin) after the removal of the protecting group. Their late stage RCM reaction to the F-ring enabled Inoue, Hirama, et al. to examine the effect of the size of the F-ring on ciguatoxin’s toxicity and ability to bind to voltage-gated sodium channels (VGSCs) [48]. As illustrated in Scheme 3.39, the synthesis of an oxocine analog of the F-ring 209 was accomplished in 93% yield employing [Ru]-I catalyst. Interestingly, 209 showed a 173-fold lower affinity for VGSC binding than did CTX3C and was not toxic to mice. The importance of the nine-membered F-ring was further demonstrated when an acyclic variant of CTX3C also showed a much lower affinity for VGSC binding. In addition to their other accomplishments, Hirama et al. utilized an interesting diastereoselective RCM reaction in their synthesis of the ciguatoxin (CTX1B) A-ring [49]. When triene 210 (available from triacetyl-d-glucose) was subjected to [Ru]-I catalyst, oxepine 211 was isolated in 95% yield as a single diastereoisomer. In the presence of an external olefin and [Ru]-I catalyst, compound 211 subsequently underwent a competitive ROM reaction, illustrating the sensitivity of medium rings having olefins imbedded in them to RCM conditions. Sasaki et al. have utilized RCM to generate the seven-membered ciguatoxin A-ring (Scheme 3.41) [50]. From the intact B–D subunit 212 that was synthesized using a novel enol phosphonate alkyl Suzuki coupling sequence, RCM using [Ru]-I catalyst gave tetracycle 213. In the course of synthesizing a model of ciguatoxin F-ring, Sasaki et al. developed the radical cyclization, RCM sequence that was used by Hirama et al. in their ciguatoxin synthesis (Scheme 3.42) [51]. The conversion of acetal 214 into selenoand thio-mixed acetals 215a and 215b was accomplished using i-Bu2 AlSePh and i-Bu2 AlSPh, respectively. Interestingly, the bicyclic acetal in 214 was necessary for conversion into the mixed acetal using the Al reagent; when monocyclic acetals were used, mixtures of products were isolated. Both 216a and 216b were converted into the corresponding oxepine 217 via 7-exo-trig radical cyclizations. Elaboration of 217 gave the cyclization precursor 218, which when treated with [Ru]-I catalyst provided the ciguatoxin F-ring as 219 in 61% yield. By coupling a reductive isomerization sequence with RCM, Fujiwara et al. were also able to generate ciguatoxin’s F-ring oxonene (Scheme 3.43) [52]. From oxepine

OBn


114 BnO H O

O

BnO Me O

OBn

O O TIPS

H

OBn

H

i-Bu2AlX

214

X = SePh (76%) X = SPh (85%)

OBn Me O

H OH O H MOMO

O H H CO2Me

OBn

6 steps

O

CH2Cl2 (0.004 M), 35 °C 61%

F H

CO2Me

OH

BnO

220

O

BnO

PMe3 98%

O

O

BnO

CN BnO BnO

O O

CHO OTBS

O 222

OTBS

224

H

OBn OBn

OH 216a: X = SPh (70%) 216b: X = SePh (79%) OBn OBn

O H H 218 OBn

OBn

CO2Me OTBS

O 221

Me3Al, TMSCN

BnO

74%

BnO

BnO

O

O O 223 O

9 steps

BnO

O 225

Scheme 3.43

X O

OBn Me O

H

1. i -Bu2AlH, CH2Cl2 −78 °C (94%) 2. TPAP, NMO (84%)

CN BnO

Me O

Synthesis of the ciguatoxin F–H subunit.

TBSO BnO

O

O H

OBn Me O G H O H H 219

H H O

O

CO2Me

OBn

[Ru]-I (22 mol%)

BnO

H O

H

217

Scheme 3.42

3 steps

OBn

OBn X O O TIPS H OH 215a: X = SPh (70%) 215b: X = SePh (79%)

CH2Cl2, hexane, rt

Bu3SnH, AIBN, PhCH3, 80 °C

BnO

Me O

O

O

OTMS OTBS

Bu3SnH, BF3 • OEt2 63%

Me Me

[Ru]-I (20 mol%) CH2Cl2 (0.003 M), rt (97%)

H BnO BnO

O O

O

Me Me

O H 226

Synthesis of ciguatoxin F-ring.

220, conjugate addition and conversion of the resulting ester to the corresponding aldehyde gave 222. Cyanohydrin formation was followed by reduction with Bu3 SnH to give acrylonitrile derivative 224 as a 1 : 1 mixture of diastereomers. Conversion of 224 into the F-ring cyclization precursor 225 involved an epoxidation, Wittig sequence, and the reduction of an intermediate π-allyl complex. The diastereomers resulting from the reduction of 223 were separated after the formation of a 1,3-diol acetonide. Diastereomer 225 was cyclized and subjected to [Ru]-I catalyst (20 mol%) to provide the F-ring oxonene 226 (97%). Another example of some of the problems associated with the use of RCM in the formation of medium rings comes from the Fujiwara et al. work on ciguatoxin [52]. The generation of the E-ring oxepine with the F-ring intact resulted in the generation of not only the desired product (228) but also a significant quantity of bis-oxepene 229 from a competitive ROM/RCM tandem sequence (Scheme 3.44).


H

O O

OBn O

TBSO

O

O

Br t-BuLi, −78 °C Ph

(79%)

O

TBSO

Ph

O

O

233

OBn O

H

O

OBn +

O

OBn

H H

O

O H H 228 (67%)

H 229 (24%)

HO

TBSO

H

O

1. TBAF, THF (100%) 2. NaIO4, H2O

O

O OTBS 231

Ph 3. NaBH4, MeOH (99%, 2 steps)

1. KCN, DMSO (95%) 2. i -Bu2AlH, −78 °C; HCl, H2O (97%) TBSO 3. MeMgBr (89%) TBSO 4. DMP (100%) 5. Cp2TiCl2, AlMe3 (88%)

O O 234

O

HO

H

O

HO

O O

OH 232

1. PhB(OH)2, PhH (95%) 2. ClCH2SO2Cl, 2,6-lutidine (100%) Ph 3. 30% H2O2, EtOAc 4. TBSOTf, 2,6-lutidine (97%, 2 steps)

1. TBAF (100%) HO Ph 2. TBDPSCl TBDPSO imidazole (77%)

[Ru]-II (10 mol%) CH2Cl2 (0.005 M), 40 °C (97%)

HO TBDPSO

H2, [Ir(PCy3)(py)(cod)]PF6 O 236

O O

Ph O

235 Me

Me

Scheme 3.45

OBn

O

Ciguatoxin’s E,F-ring system.

O

O OTBS 230

H

H

O

H 227

OSO2CH2Cl

TBSO

O

CH2Cl2 (0.003 M), rt

Scheme 3.44

OHC

OBn

[Ru]-I (30 mol%)

115

O O

Ph

88%

HO TBDPSO O

O

O 237 (dr = 19 : 1)

RCM to ciguatoxin I-ring.

Fujiwara and Murai have used tri-O-acetyl-d-glucal as a precursor to both the ciguatoxin F-ring and the I-ring oxocine [53]. The coupling of aldehyde 230 with vinyl lithium gave 231 (Scheme 3.45). Oxidative cleavage of the pyran ring gave 232 after reduction. The manipulation of the protecting group and the formation of the disubstituted olefin gave the cyclization precursor 235. Impressively, RCM using [Ru]-II catalyst resulted in the requisite trisubstituted olefin and oxocine 236 in high yield. Interestingly, hydrogenation of 236 using Crabtree’s catalyst showed unusual stereoselectivity. Rather than the anticipated directed reduction, the reaction gave 237. Also interesting was that the use of PtO2 as a catalyst for the hydrogenation resulted in the formation of the epimer of 237 as the major product. A related sequence of reactions was used to generate the ciguatoxin F-ring oxonene as illustrated in Scheme 3.46 [53]. RCM using [Ru]-I catalyst gave oxonene 241 in 98% yield. In a related fashion to their aforementioned Ireland–Claisen/RCM sequence (Schemes 3.4–3.6), Goto et al. have coupled a stereoselective [2,3]-Wittig rearrangement with RCM as a mean to synthesize the ciguatoxin E,F-ring system (Scheme 3.47) [54]. From 242, conjugate addition provided 243 in 99% yield after one recycle. Stereoselective ketone reduction and alkylation with bromo-t-butyl acetate gave the rearrangement precursor 244. The [2,3]-rearrangement of 244 was

Ph


116

HO

H

O

HO

OSO2CH2Cl

1. PhB(OH)2, PhH (98%)

O O

Ph

OH 238

O 2. ClCH2SO2Cl, 2,6-lutidine PhB (98%) O

O Ph

O

O

239 [Ru]-I (16 mol%) O

TBSO O 240

TBSO

Scheme 3.46

O

Ph CH2Cl2 (0.003 M), 40 °C 98%

O

O

TBSO

Ph

O 241

TBSO

O

Synthesis of the ciguatoxin F-ring.

O O

BnO BnO

E O 242

OH

BnO

O Bu3P, CH2Cl2 99%

LHMDS, −78 °C to 0 °C 86%

BnO

O 243

O

O BnO BnO

O

O O 245

HO

CO2t-Bu

BF3 • OEt2, THF −78 °C to 0 °C 87%

BnO BnO

Scheme 3.47

O

O O

OH

BnO BnO

MeOH, −78 °C (100%) 2. BrCH2CO2t-Bu, Bu4NHSO4 50% aq NaOH CH2Cl2, rt (94%) 1. LiBH4, THF, 0 °C 2. Bu2SnO, TsCl, Et3N CH2Cl2, rt 3. K2CO3, MeOH, rt 77% (3 steps)

O Li, CuCN

O O 1. NaBH , CeCl •7H O 4 3 2

O

[Ru]-I (129 mol%)

O

O O 244

O CO2t Bu

O BnO BnO

BnO

CH2Cl2 (0.003 M), 40 °C BnO 92%

247

O

O O 246

O

E

H H O

O

F

O O OH

H 248

[2,3]-Wittig/RCM sequence to ciguatoxin F-ring.

induced with LHMDS (lithium hexamethyldisilazide) to give 245 in 86% yield with almost complete diastereoselectivity. Reduction of the ester, epoxide formation, and coupling with vinyl cuprate provided the F-ring precursor 247. Despite the presence of four olefins, [Ru]-I catalyst reacted chemoselectively with the terminal olefins to give the F-ring oxonene in 92% yield as 248. As part of their efforts targeting the A–E subunit of ciguatoxin CTX3C, Clark et al. utilized a two-directional RCM sequence to synthesize the A- and E-rings in a single flask (Scheme 3.48) [55, 56]. From glucal derivative 249, the functional group manipulation resulted in triene–yne cyclization precursor 250. Two-directional diene, ene–yne RCM using [Ru]-II catalyst gave A,B,C-tricycle 251 in 58% yield. Selective oxidation of the enol ether and enone formation gave the D-ring precursor 252. RCM using [Ru]-II catalyst gave oxepine 253 in 70% yield. Functional group

3.8 Marine Ladder Toxins H

HO

H

O

H

H OBn

H

O

O Si(t-Bu)2

O

H

249

H

H OBn

H

[Ru]-II (10 mol%) H2C CH2 PhMe (0.02 M), 70 °C 58%

O

O

H

250

1. DMDO, CH2Cl2, 0 °C;

H

BF3 • OEt2, Et3SiH, −40 °C O

2. NaH, ClCH2COCHPPh3 Bu4NI, THF, 65 °C

H

H

H

O

H

H

O

H

H OBn

O

H

[Ru]-II (5 mol%)

O

H

H

O H H OBn 254

H O

H OBn

O

H

H

O

O

H

H OBn

O

H

253

O O H

H

O

CH2Cl2 (0.007 M), 40 °C 70%

252

3. HCHO, Et2O 56% (2 steps)

O

O

O

251

O

(71%, dr = 5 : 2)

Scheme 3.48

O

117

H [Ru]-II (10 mol%) CH2Cl2 (0.007 M), 40 °C 50%

O

H

O

H

H OBn

H

O

H O

H

255

Two-directional synthesis of the ciguatoxin A–E precursor.

manipulation and alkylation provided dienyl ketone 254 as the E-ring precursor. A third RCM reaction gave the E-ring oxocine and completed the synthesis of the A–E ciguatoxin precursor 255. 3.8.2 Brevetoxin

Kadota, Yamamoto et al. have developed a novel subunit coupling strategy that combines an intramolecular allylstannane/acetal cyclization with diene RCM. This sequence of reactions was used to link the B–C- and F–G-rings of brevetoxin B (Scheme 3.49) [57]. Beginning with the B–C rings containing substrate 256, N-chlorosuccinimide (NCS)-induced oxidation and AgOTf-catalyzed coupling with F,G-rings containing alcohol 258 gave 259 in 81% yield for the two steps. Removal of the TBS group and incorporation of an allylstannane moiety using allylstannyl ether 260 gave the cyclization precursor 261 following the elimination of MeOH. Intramolecular allylation catalyzed by AgOTf gave the D-ring as 262 in 84% yield as a 78 : 22 mixture with epimer 263, respectively. Diene 262 was subjected to [Ru]-I catalyst to give oxepene 264 in 72% yield. The use of [Ru]-II catalyst in the cyclization of diene 263 gave a quantitative yield of oxepine 265. Interestingly, 262 did not undergo RCM under the conditions used to cyclize 263. Kadota, Yamamoto et al. took a related approach to the generation of the H–I brevetoxin B rings and the completion of the synthesis (Scheme 3.50) [57]. Esterification of A–G subunit 267 with J,K subunit 266 gave ester 268. Allylstannane generation and reduction of the ester gave the acetal cyclization precursor 269 after the formation of chloroacetate. As anticipated, the cyclization to the brevetoxin B I-ring was highly diastereoselective giving pyran 270 as a single diastereomer in

O O H


118

H O Ph

O

O B

Me H

H H 256

C O

H

OTBS

NCS SPh

CCl4, rt

H

O O

Ph

O

Me H

H H 257

O

Me Me H O OBn G F OBn, HO O H H Me 258

OTBS SPh

H

AgOTf, CH2Cl2, −78 °C to rt 81%

Cl

1. TBAF, THF, rt OMe H O Ph

O

H

Me H OTBS Me Me H O O OBn SPh H

O

H

O Me 259

H

H

H

AgOTf, 4Å MS CH2Cl2, rt 84%

O Ph

O

O

O

H

2. Bu3Sn

H

H

260

O

3. TMSI, HMDS, CH2Cl2, 0 °C (84%) Me Me H OBn O

Me H H O O

Bu3Sn

CSA, CH2Cl2, rt (81%, 2 steps)

OBn

Me

O H

H Me

H

H

O

Me

N

Ph

O

O

H

Me H O SPh H

O

O

H

Me

Me Me H O OBn H

H

O

OBn

Me

261

H

+

O

OBn

O

Ph

O

H

H

Me Me H O OBn

Me H OH O

O H Me

H

H

H

O

Me

OBn

262 263 262 / 263 = 78:22 [Ru]-I (50 mol%) PhH (0.002 M), 60 °C [Ru]-II (64 mol%) CH2Cl2 (0.008 M), rt 100% 72%

H O Ph

O

H

O B

Me Me H O OBn Me H H O E F G C D O OBn O O H H H Me H H Me 264

H O Ph

O

H

O

Me H OH H

O

H

Me Me H O OBn

O H H Me 265

H

O

Me

OBn

Scheme 3.49 The use of RCM in convergent subunit couplings to the brevetoxin B B–G subunit.

84% yield. RCM using [Ru]-I catalyst provided an 81% yield of the H-ring oxocine and the fully intact brevetoxin B skeleton as 271. Oxidation of the A-ring and the allylic ether side chain completed the synthesis of brevetoxin B. Crimmins et al. have used their sequential glycolate alkylation/RCM and glycolate aldol/RCM protocols to synthesize the marine natural product brevetoxin A. The synthesis of the B-ring began with the glycolate coupling of 272 with methallyl iodide to give adduct 273 (Scheme 3.51) [58]. Protecting group manipulation then led to the glycolate aldol precursor 274. Coupling of 274 with benzyl iodomethyl ether gave 275 after the reductive removal of the auxiliary. Oxidation and vinyl Grignard addition gave the cyclization precursor 276 (major diastereomer shown) as a 3 : 1 mixture of diastereomers. Cyclization using [Ru]-II catalyst delivered the B-ring oxocine as tricyclic alkene 277 in 61% yield for the three steps. The diastereomers arising from the Grignard addition were separated after cyclization and the minor diastereomer was recycled to 277 using an oxidation–reduction sequence. The alcohol in 277 was then used as a handle to direct the hydrogenation of the alkene to give the desired C6 brevetoxin A precursor 278 and ultimately 279 after modification. The E-ring was synthesized as outlined in Scheme 3.52 [58]. Beginning from glycolate aldol–derived adduct 280, alkylation with bromoacetonitrile gave 281 following the reductive removal of the auxiliary. Oxidation of the primary alcohol, diastereoselective addition of allyl stannane to the resulting aldehyde, and hydrolysis gave lactone 282. Protecting group and oxidation state manipulation gave 283 as


H HO2C TESO

O J

OTBS Me

H

H

K O

OTBDPS

266

Cl

Me H COCl, Et3N, THF, 40 °C 94%

Cl

A O

Cl

Me H

O

O O

H

H

O

H

H

Me Me H O O

Me H H O H Me

H

Me

O B

O

H H

O

H

Me Me H O OH F G

E O H Me

H

H

O

Me

267

OTBS Me

O TESO

O H Me 268

H H O C D

OTBDPS

O

H

1. TBAF, THF, 0 °C OMe 260

2. Bu3Sn

CSA, CH2Cl2, rt 3. TMSI, HMDS, CH2Cl2, 0 °C (71%, 3 steps) 4. i -Bu2AlH, CH2Cl2, −78 °C; (CH2ClCO)2O, DMAP pyridine, CH2Cl2, -78 °C (68%)

Me H

Me Me O

Me H H O O O

O

H

H

O

H Me

H

H

H

O

H

H

O OR

Me

O J

O

OTBS Me

H

K O

OTBDPS

269: R = COCH2Cl SnBu3

MgBr2•OEt2 CH3CN, 40 °C 82%

Me H

O

H Me Me H O O

Me H H O O

O

H

H

O

H Me

H

TBSO Me H O

H

H

O

Me

O O

H

H

O

H Me

H

TBSO Me H O

H Me Me H O O

H H O

H

O

Me Me O

Me H H O O

O

H

H

O

H

H Me

H

O H H

TBSO Me H O

2. HF•pyridine, CH2Cl2, 0 °C 3. MnO2, Et2O, rt (84%, 2 steps)

O

OTBDPS O H

O H Me 271

1. PCC, PhH, 80 °C (81%, 2 steps)

Me H

O H H

O H Me 270

[Ru]-I (150 mol%) PhH (0.002 M), 40 °C

Me

I

OTBDPS O H

H

O

H HO

CHO O H

O H H

Me

Brevetoxin B

Scheme 3.50

Completion of brevetoxin B.

119

120 O O

3 Synthesis of Medium-size Oxygen Heterocycles by Ring-closing Alkene Metathesis 1. NaHMDS, THF CH2 CH(CH3)CH2I

O N

PMBO

PMBO

OPMB −78 °C to −45 °C 2. LDA, EtOAc, −78 °C 272 i-Pr 66% (2 steps)

6 steps

EtO2C O

O 273

OBn

2. CH2

TBSO

CHMgBr, 0 °C

H

O

H 276 (dr = 3 : 1)

275 OH

PMBO H

[Ir(PCy3)(Py)(COD)]PF6

TBSO

H2, −50 °C, CH2Cl2 93%

Scheme 3.51

H

[Ru]-II (10 mol%) H TBSO OH H CH2Cl2 (0.002 M), 40 °C OBn 61% (3 steps)

H Me H OH OBn

O

4 steps

H 278 (dr > 98 : 2)

Me H H O

TIPSO PMBO

N OH

H

280

281 O Me H H O

O H 277

H OH OBn

H Me PMBO H TIPSO B Me OH H O H CHO 279

H

1. (COCl)2, DMSO; Et3N CH2Cl2, −78 °C 2. AllylSnBu3, AlMe3, −78 °C

HO H

283

O O H

282 (dr > 7 : 1)

O Me H H O

[Ru]-II (5 mol%) CH2Cl2 (0.005 M), 40 °C 75%

Me H H O

HO

3. HCl, MeOH, 65 °C 80%

(MeO)2(O)P OBn OBn H

TBSO

Scheme 3.52

Me

The alkylation/RCM approach to the brevetoxin A B-ring.

OTIPS 1. NaHMDS, −78 °C Me i -Pr BrCH2CN, THF H PMBO N O O 2. NaBH4, H2O, THF H 75% O O

(MeO)2(O)P

O PMBO H

PMBO H

1. (COCl)2, DMSO, NEt3 O

N

O

PMBO

2. LiBH4, MeOH, Et2O 72%

i -Pr

274 O

TIPSO

1. NaHMDS, THF (BnO)2CH2, TMSI, −78 °C

TIPSO

TBSO

E H

OBn OBn H

284

Synthesis of the brevetoxin A E-ring.

the E-ring cyclization precursor. Exposure of 283 to [Ru]-I catalyst resulted in the smooth conversion to the corresponding oxonene 284. A related strategy was used to build the brevetoxin A G-ring oxocine 288 (Scheme 3.53) [59]. In an exceptionally convergent fashion, the A-, E-, and G-ring fragments (279, 284, and 288, respectively) were coupled and carried forward to brevetoxin A [60]. 3.8.3 Gambierol, Gambieric Acid, Olefinic-ester Cyclizations

Johnson et al. synthesis of gambierol employed enol ether–olefin RCM to generate the seven-membered E- and H-rings [61]. The use of RCM to generate the H-ring is illustrated in Scheme 3.54. Olefinic ester 289 containing the intact the F- and G-rings was subjected to the two-step metathesis protocol that was initially described by Clark et al. (Table 3.2) where the ester was first converted into the corresponding

3.8 Marine Ladder Toxins O O

O

1. NaHMDS, THF BnOCH2I, −78 °C to −45 °C

OPMB O H

N

BnO

2. LiBH4, MeOH, Et2O, 0 °C 69%

i -Pr

H

O

OTIPS

Me

1. Dess–Martin, CH2Cl2 OPMB H

HO

121

2. VinylMgBr, THF, −78 °C

OTIPS 285

286 BnO

H Me O

OTIPS

H [Ru]-II (5 mol%)

BnO

OPMB CH Cl (0.01 M), 40 °C HO 2 2 H 71% (3 steps)

HO

O

Me

G

OTIPS OPMB

H 288

287 Me H O O

Me O H

B H

O

H

HO H

Me H OH

HO H H Me H O O

E O H H Brevetoxin A

Scheme 3.53

H PMBO TMSO

O H H

H

O

CHO

The synthesis of the brevetoxin A G-ring and completion.

O

Me

Scheme 3.54

G

H

O H

O Me H 289

TBSO OTBS O

1. TiCl4, Zn, PbCl2, CH2Br2, TMEDA, THF, 65 °C (71%) 2. [Mo]-I (11 mol%), hexanes (0.005 M), 65 °C (88%)

H PMBO TMSO

O

H

HO

O Me Me H 290

The synthesis of the gambierol H-ring.

acyclic enol ether and this substrate was then subjected to either [Mo]-I catalyst or [Ru]-II catalyst to give 290. For 289, [Mo]-I catalyst was more reactive. The significant advantage to enol ether–olefin RCM is that the precursor to the enol ether is an ester thus enabling the coupling of fairly elaborate substrates. An illustration of this is Johnson et al. completion of gambierol where the A–C and F–H precursors 291 and 292 were coupled and then subjected to the same two-step RCM sequence outlined for 290 (Scheme 3.55). In contrast to the protocol from 289, a titanium ethylidene reagent was used in the initial enol ether forming reaction to avoid the substrate decomposition that was seen with the use of the corresponding titanium methylidene. Surprisingly, this modification led to the generation of a 2 : 1 mixture of oxepine 294 and expected acyclic enol ether 293. After separation, enol ether 293 was also converted into 294 using [Ru]-II catalyst. This latter reaction also resulted in the generation of a small amount of dihydropyran 295. The overall yield to 294 was 80%. Oxepine 294 was converted into gambierol in an additional 12 steps.


122

TIPSO TBDPSO H

Me Me H OTES O O

O

H

H 291

Cl

1.

HOTES

TBDPSO O H

O H

H

H

H OH

H

OTBS OTBS Me

O Me Me H 292

0

Cl, Et3N, DMAP 2. CH3CH2Br2, CH2Cl2, Zn , PbCl2, TMEDA, THF, TiCl4 Cl CH2Cl2, 40 °C 90%

O H

OTBS TBDPSO + OTBS Me

Me O H 293 (30%)

H

TIPSO Me Me H OTES O

H H O

O Me

H

OH

H

Me HO

O

O

Cl

TIPSO Me Me O

O

H +

H

OH

O O Me Me O H H H 294 (60%)

OH

H H O

OTBS

OTBS Me

[Ru]-II (40 mol%) PhH (0.002 M), 80 °C H2C CH2 OH Me Me H H O H O O H A B C D E H H F O O O H G O H H Me H H H O Gambierol Me OH H Me

294 (65%) HO

+ OTIPS Me Me H OTES H H O O

TBDPSO H

O

H

H

O

Scheme 3.55

H OH

OTBS Me

O

O Me Me H 295 (20%)

H

OTBS

Completion of gambierol.

H H

Me BnO H Me

O

TIPSO +

O

H 296

HO

HO2C

H

Me Me

H

O

2,4,6-Trichlorobenzoyl chloride

O

Et3N, DMAP 100%

H

O

BnO

TIPSO

O

H

O H H Me

298: X = O 299: X = CH2

H

Hexanes (0.002 M) 60 °C TIPSO Me

BnO

Scheme 3.56

A OH H Me

B O

H

O C

Me Me

H

D O

O E

O

X H

O

H

Me

297

[Mo]-I (11 mol%)

Me Me

TiCl4, Zn, PbCl2, CH2Br2, TMEDA, THF, 65 °C 100%

+ inseparable oligomer

H

300 (99%

0.06 M LiOH H2O/dioxane

OH

O

OH P

O

3

SPh (4) 29

27

7

65–70%

(1)

No Rxn

10

LiAlH4, Et2O 0 °C

HCl Dioxane/H2O or TMSCl

O O

5

P O

3

O 1

Hard Nu

LiSPh, THF 98%

O

Soft Nu

OH P O

Et2Zn, CuCN·2LiCl 80%

O

10

(5)

Et 30

(S,S,PR )-26

(E /Z = 8 : 1)

Scheme 4.10

with allyl alcohol through a phosphate-mediated sequence [4a]. Basic hydrolysis (Equation 3, Scheme 4.10) also resulted in attack at phosphorus, however, due to the characteristics of the phosphate monoanion [21], an excess of LiOH only displaced one of the three P–O bonds, yielding phosphate acid 28. Variation from an oxygen nucleophile to a sulfur nucleophile such as LiSPh (Equation 4, Scheme 4.10) demonstrated the preference for addition onto the most sterically accessible C3 carbinol position, furnishing phosphate acid 29. Lastly, cuprate addition resulted in allylic displacement of the more sterically accessible olefin (Equation 5, Scheme 4.10), resulting in phosphate acid 30, along with preferential formation of the E-isomer, which is consistent with previous reports in the field of cuprate additions to allylic phosphates [22]. The preference of cuprate nucleophiles to add to the more sterically accessible olefin of 26 led into examining the possible functionalization of the endocyclic olefin [4a] After regioselective hydrogenation of the exocyclic olefin in the presence of Wilkinson’s catalyst, cuprate addition to phosphate 31 preferentially occurred at

4.4 Synthesis and Reactivity of Phosphates from RCM

O O

RhCl(PPh3)3 (10 mol%)

P

O Toluene, H2, 1 h 40%

O

5 4

(S,S,PR)-26

O O O P 1 O 2 31

1. Et2Zn,CuCN·2LiCl 2. Red-Al 69% (2 steps)

OH

OH

32

Et

(Regioselectivity > 99 : 1, dr > 99 : 1)

Scheme 4.11

the C5 position with high regioselectivity (Scheme 4.11). Removal of the phosphate acid with Red-Al produced diol 29 bearing an 1,3,4-anti,anti-stereotriad, as a single diastereomer over the three-step sequence from (S,S,PR )-26. The remarkable regioselectivity in this transformation is attributed to inherent stereoelectronic effects of the C4–C5 olefin and P–O bond of the corresponding allylic phosphate. For addition into the C5 position (path A, Figure 4.2, X-ray of structure 26), orthogonal alignment of the π * orbital of the C=C bond and σ * orbital of the P–O provide proper electronics for the allylic displacement to occur. However, in a potential C4 addition (path B, Figure 4.2), SN 2’ addition does not take place due to coplanar alignment of the aforementioned π * orbital of the C=C bond to the P–O σ * orbital [4a]. Facial selectivity was attributed to the overall geometry of the bicyclic skeleton of 26, whereby the bicyclo[4.3.1]phosphate system hinders approach of the cuprate from the concave face (Figure 4.3, X-ray of structure 26) [4a]. Coplanar C C/C–OP(O) alignment in path B Path A nucleophile Et2Zn, CuCN·2LiCl

Path B nucleophile Et2Zn, CuCN·2LiCl

Orthogonal C C/C–OP(O) alignment in path A

Figure 4.2

137

X-Ray structure of (S,S,PR )-26 displaying orbital alignment.

138

4 Phosphorus and Sulfur Heterocycles via Ring-closing Metathesis

Convex face accessible for Nu addition

Concave face blocked from Nu addition

Figure 4.3

X-Ray structure of (S,S,PR )-26 displaying facial bias.

O P

O

O

R, Ru-[III]

O

CH2Cl2

P

O

R

O

O

O

(R,R,PS )-26

R

Yield (%)

E/Z

Type

CH2OH CH2OTBS CH2NHBoc CH2OP(O)(OMe)2 CO2Me CO2t-Bu CHO

86 87 69 80 78 60 78 72

>99 : 1 >99 : 1 >99 : 1 2:1 8:1 5:1 >99 : 1 >99 : 1

I I I I II II II II

BnO HO Scheme 4.12

Another strong attribute of the bicyclic phosphate 26 is the ability to mediate cross-metathesis (CM) with various type I and II olefins [23] as shown in Scheme 4.12 [4b]. Studies performed with (R,R,PS )-26 (synthesized from (R,R)-1,6-diene-3,5-diol) provided stereoselective formation of the E-isomer in the presence of the catalyst [Ru]-III. When attempting to undergo CM with type III partners [23], very poor yields were observed in the presence of [Ru]-III (Table 4.1). In accordance with Grubbs

4.4 Synthesis and Reactivity of Phosphates from RCM Table 4.1

Cross partner

O

Conc.

Yield

0.05 M

< 5%

0.05 M

< 5%

neat

< 5%

neat

< 5%

0.05 M

no rxn.

MeO

NC

none

classification of CM partners, type III olefins do not undergo homodimerization, and are therefore ideal heteropartners with substrates bearing type I and II olefins and unreactive with substrates possessing type III olefins [23]. On the basis of the observations seen and the fact that 26 did not undergo homodimerization, the exocyclic olefin in 26 was deemed to be type III [4b]. By combining the CM and cuprate methodologies, an advanced intermediate 37 containing the key aspects of the dolabelide family was constructed (Scheme 4.13) [24]. Hence, phosphate-mediated coupling of (R,R)-24 with 1,1-dimethylallyl alcohol afforded the bicyclic phosphate 33 over a three-step sequence. CM of the latter with but-3-en-1-ol in the presence of [Ru]-III and subsequent hydrogenation of the exocyclic olefin of 34 furnished an extended side chain bearing the C15–C19 structure of dolabelide. After benzylation of the primary alcohol in 35 with p-methoxybenzyl trichloroacetimidate, addition of a methyl cuprate, methylation of the phosphate acid (TMSCHN2 ), and removal of the phosphate ester (Red-Al) yielded advanced intermediate 37, a key synthon toward the synthesis of dolabelide C [4b]. The differentiated 1,3,4-anti,anti-stereotriad seen in 37 is also present in a number of biologically active natural products, including dolabelide C, bitungolide E, (+)-discodermolide, dictyostatin, and salicylihalamide A (Figure 4.4) [10, 21–24].

139

140


OH

OH

O 1. POCl3, Et3N, DMAP (90%)

(R,R )-24

OLi , THF, −40 °C

2.

P

O

OH

O

[Ru]-III (10 mol%) 77%

O 33

3. [Ru]-III, (5 mol%) (50% over 2 steps)

O P

O

HO

O

[Ru]-II (10 mol%)

O

Et3N (50 mol%) CH2Cl2 (c = 0.005 M)

O 34

O

HO

P

O

NH CCl3 , PPTS

O

94%

35

1. Me2Zn, CuCN·2LiCl 2. TMSCHN2

O

PMBO

O

then H2 (300 psi), 37 °C 71%

OPMB

P

OPMB

OH

OH

3. Red-Al

O

65% (3 steps)

O 36

37

Scheme 4.13 OH

OAc OAc

O OAc

OH OH O

H2 N OAc

O

OH

OH

O HO HO (+)-Discodermolide

Dolabelide C

O O

H N

OH OH

OH O H

O

O

OH O

HO O

OH

O

O OH OH Bitungolide E

Figure 4.4

Salicylihalamide A

Dictyostatin

Cross metathesis of (S,S,PR )-26 with type III olefins.

4.5 Applications of Phosphate Tethers in the Synthesis of Dolabelide C

In 2008, Hanson et al. reported the synthesis of two key subunits of dolabelide C, which possess the anti-1,3-diol motif observed at the C7/C9 positions and the 1,3,4-anti,anti-stereotriad at the C19/C21–C22 positions. Retrosynthetic analysis segmented the 24-membered macrolide into a C1–C14 subunit 38 and

4.5 Applications of Phosphate Tethers in the Synthesis of Dolabelide C

11

OAc OAc

OH

9

3

7 OH

OAc

19

14

O

OH 21

11

O

OR

9

7

OPMB 3 Me O

OR

OAc n-Pr

23

OR

14

141

Me

1

2

2

OP

OP

OP

19

21

23

Me

2

OP

1 OH

+

27 n-Pr

Me 38

Dolabelide C

39

Scheme 4.14 OR

OR

9

OPMB

7

3

11

Me

OR 14

RCM

Me

O

1 O

P

OH

O

9

O 7

38

Me

OP1 OP2 19

21

PMBO

O

23

OP2 27 n-Pr

1

3 CM

OP2 Me

OP

O

3

P O

OP1 OP2

O

19

23 H

21

O 7

CM

41

OH

9

7

(R,R )-24

19 O O O P 21 O

RCM 39

OH

(R,R,PS )-26

40

R1

9 O

OH

OH

19

21

(S,S )-24

23 (S,S,PR)-26

Scheme 4.15

C15–C30 subunit 39 from disconnecting the lactone bond and C14–C15 olefin (Scheme 4.14) [7]. Each subunit was constructed by a phosphate-mediated desymmetrization of 24, followed by the aforementioned CM and allylic phosphate displacements. The C1–C14 subunit was envisioned to have the C1–C6 side chain appended via CM to yield advanced intermediate 40 from (R,R,PS )-26. Subsequent regioselective hydrogenation of the C5–C6 olefin and regioselective Pd(0)-mediated allylic hydride addition (to C10) would afford a terminal olefin armed for oxidation and subsequent installation of the C11 carbinol center (Scheme 4.15) [7a]. The C15–C30 subunit largely depended on the CM/cuprate approach (Scheme 4.13) and the advanced intermediate 41 would be synthesized from (S,S,PR )-26 [7b]. The synthesis of the C1–C14 subunit was initiated from readily available (R,R,PS )-26, which was coupled with olefin 42, possessing the necessary C2–C4 stereochemistry, by CM in the presence of [Ru]-III (Scheme 4.16). Regioselective hydrogenation of the external olefin in 43 was accomplished using in situ formation of diimide from o-nitrobenzenesulfonyl hydrazine [25] and Et3 N to provide partially reduced phosphate 44. This was followed by another regioselective ring opening, and a Pd(0)-mediated reduction of allylic phosphate 44 via formation of hydride from formic acid and Et3 N [26]. Overall, this transformation resulted in an allylic displacement of the endocyclic phosphate with hydride, which after methylation with TMSCHN2 yielded phosphate ester 45. Removal of the phosphate ester with LiAlH4 produced advanced intermediate 46, bearing the C1–C10 framework [7a].


142

11

11 O O

P

PMBO OTBS 4 2 3

9 +

O

O

[Ru]-III (6 mol%) CH2Cl2, 90 °C

O 7

9

O

O 7

72%

(R,R,PS )-26

PMBO

O

P

43

11

Pd(OAc)2 (5 mol%)

SO2NHNH2

O

NO2

P

Et3N, CH2Cl2, rt, 24 h

O

9 PMBO

O O 7

72%

1

3

42

OTBS

3

OTBS

HCO2H, Et3N DCE, 40 °C

OR

9

7

PMBO

OTBS 1

11

then TMSCHN2

1

OR

87%

45: R = P(O)OMe 46: R = H

44

LiAlH4 75%

Scheme 4.16 12

12 O P O

9 PMBO

O O 7

3

1

10

O

Pd(OAc)2 (5 mol%) HCO2H, Et3N

OTBS

DCE, 40 °C

O

PdLn

9 PMBO

O

P

44

11

O 7

OTBS 1

3 44a

(Regioselectivity C10/C12 = 37 : 1)

12

O OMe P O O 10

PMBO

OTBS

RO P

9

HCOOH, Et3N

7

1

O

45 TMSCHN2, MeOH

9 PMBO

O O 7

3

OTBS 1

R=H R = Me

Scheme 4.17

Introducing Pd(0) to phosphate 44 gave rise to π-allyl complex 44a (Scheme 4.17). The Pd(0)-mediated allylic hydride addition [26] to phosphate 44 gave the desired regioselectivity of hydride addition at the internal position (C10) of 44a (regioselectivity C10/C12 = 37 : 1), similar to the previous cuprate additions to 26. Both examples are illustrative of the preference in regioselectivity of allylic phosphate displacements with the endocyclic π bond. The C15–C30 subunit was synthesized by means of two approaches. Initially, regioselective hydroboration [27] of the exocyclic olefin in (S,S,PR )-26, followed by PMB protection of the primary alcohol, provided 47 (Scheme 4.18). Allylic displacement using a methyl cuprate derived from Me2 Zn and CuCN·2LiCl furnished the aforementioned 1,3,4-anti,anti-stereotriad of the C19/C21–C22 subunit as a single diastereomer. Methylation of the phosphate acid and cleavage of the resulting phosphate ester 48 with LiAlH4 afforded 1,3-diol 49 over the five-step sequence from bicyclic phosphate 26 [7b].

4.5 Applications of Phosphate Tethers in the Synthesis of Dolabelide C 1. 9-BBN-H, then H2O NaBO3·4H2O (10 equiv.)

19 O O 21 O P 22 O

2. PPTS, CH2Cl2 NH PMBO

(S,S,PR )-26

OPMB 19 O O 21 O P 22 O

CCl3

1. CuCN·2LiCl Me2Zn, THF

PMBO

MeOH 87%

19

47

71% (2 steps)

OR

2. TMSCHN2

143

OR 22 21

48: R = P(O)OMe (dr > 99 : 1) 49: R = H

LiAlH4 96%

Scheme 4.18 OPMB 3

OPMB 3

OPMB O O

P O

O

O

[Ru]-III (6 mol%)

O

DCE, 90 °C

P O

O

82%

(S,S,PR )-26

SO2NHNH2 NO2

O O

Et3N, CH2Cl2 75%

P O

O 1. CuCN·2LiCl, Me Zn 2 2. TMSCHN2, MeOH, 91%

51

50

15

OR OR 18 19 21 OPMB Me

LiAlH4 92%

52: R = P(O)OMe (dr > 99 : 1) 53: R = H

Scheme 4.19

OH

OH

9

7

PMBO

OH

OH

9

7

OTBS

PMBO

OTBS 1

11

1

OH

46

54

14

Dolabelide C

OH

OH 15

15

19 OPMB

21

53

TIPSO 19

OP

OP

OP

21

23

27

30

55: P = MOM

Scheme 4.20

An alternative route utilized CM of (S,S,PR )-26 and PMB-protected pent-4-en-1-ol to give phosphate 50 (Scheme 4.19). Selective hydrogenation of the exocyclic olefin gave 51, which possesses the C15–C18 side chain of the C15–C30 subunit. Again, allylic displacement with a methyl cuprate derived from Me2 Zn and CuCN·2LiCl occurred regioselectively and gave phosphate 52 (after methylation with TMSCHN2 ). Phosphate cleavage with LiAlH4 afforded 1,3-diol 53 in a five-step protocol from 26 in higher overall yields than the previous route, with the C15–C18 side chain [7b]. With the carbon framework in hand, advanced intermediates 46 and 53 were converted to the C1–C14 (54) [7a] and C15–C30 (55) [7b] subunits of dolabelide C respectively, setting the stage for final efforts en route to the total synthesis of the natural product (Scheme 4.20). The synthesis of bicyclic phosphate triester 26 has been accomplished via a desymmetrization of C2 -symmetric diol 24 using a phosphate-tether mediated

144


coupling/RCM sequence with allyl alcohol. The facile assembly of complex polyol synthons 54 and 55 can be realized by exploiting selective reactivity patterns within 26, namely, CM, cuprate additions, hydroboration, and hydrogenation. Overall, this work represents the first report of using a phosphate-mediated approach to constructing complex intermediates contained in a natural product. Additional biologically active natural products bearing the 1,3,4-anti,anti-stereotriad can be accessed through these phosphate-mediated processes [24].

4.6 Conclusion

The use of temporary tethers in synthesis represents a powerful approach to the coupling of two advanced structures en route to more advanced synthons. RCM plays a key role in the construction of sultone and phosphate heterocycles providing effective methods where both sulfur and phosphorus serve as temporary tethers mediating several selective transformations. The use of sultones to construct 1,3-dienols possessing an internal Z-olefin has been utilized in the synthesis of the originally proposed structure of (±)-mycothiazole and has the potential of expanding to other biologically active natural products. Phosphate tethers have served a central role in the construction of asymmetric 1,3-anti-diol subunits, which have been demonstrated in the construction of both the C1–C14 and C15–C30 subunits of dolabelide C [7]. Additional uses of these versatile heterocycles are in order and will be reported in future studies.

Acknowledgment

The authors would like to thank the National Science Foundation (NSF CHE-0503875) and the National Institute of General Medical Sciences (NIH RO1 GM077309) for their kind and generous support of their program. The authors also kindly acknowledge Materia, Inc., for supplying metathesis catalyst and helpful suggestions.

References 1 (a) Fensterbank, L., Malacria, M.,

and Sieburth, S. (1997) Synthesis, 8, 813–854; (b) Gauthier, D.R., Zandi, K.S., and Shea, K.J. (1998) Tetrahedron, 54, 2289–2338. 2 (a) White, J.D. and Carter, R.G. (2001) in Science of Synthesis: Houben-Weyl Methods of Molecular Transformations, vol. 4 (ed. I. Fleming), Georg Thieme

Verlag, New York, pp. 371–412 and references cited therein; (b) Evans, P.A., Cui, J., Gharpure, S.J., Polosukhin, A., and Zhang, H.R. (2003) J. Am. Chem. Soc., 125, 14702–14703; (c) Cox, L.R. and Ley, S.V. (2000) in Templated Organic Synthesis (eds F. Diederich and P.J. Stang), Wiley-VCH Verlag GMBH, Weinheim, pp. 275–395.

References 3 Nosse, B., Schall, A., Jeong, W.B., and

4

5

6 7

8

9

10

Reiser, O. (2005) Adv. Synth. Catal., 347, 1869–1874. (a) Whitehead, A., McReynolds, M.D., Moore, J.D., and Hanson, P.R. (2005) Org. Lett., 7, 3375–3378; (b) Waetzig, J.D. and Hanson, P.R. (2006) Org. Lett., 8, 1673–1676; (c) Whitehead, A., McParland, J.P., and Hanson, P.R. (2006) Org. Lett., 8, 5025–5028. (a) Le Flohic, A., Meyer, C., Cossy, J., Desmurs, J.-R., and Galland, J.-C. (2003) Synlett, 5, 667–670; (b) Le Flohic, A., Meyer, C., and Cossy, J. (2006) Tetrahedron, 62, 9017–9037. Le Flohic, A., Meyer, C., and Cossy, J. (2005) Org. Lett., 7, 339–342. (a) Waetzig, J.D. and Hanson, P.R. (2008) Org. Lett., 10, 109–112; (b) Whitehead, A., Waetzig, J.D., Thomas, C.D., and Hanson, P.R. (2008) Org. Lett., 10, 1421–1424. (a) Karsch, S., Schwab, P., and Metz, P. (2002) Synlett, 12, 2019–2022; (b) Karsch, S., Freitag, D., Schwab, P., and Metz, P. (2004) Synthesis, 10, 1696–1712. (a) Roberts, D.W. and Williams, D.L. (1987) Tetrahedron, 43, 1027–1062; (b) Burglass, A.J. and Tillett, J.G. (1991) in The Chemistry of Sulfonic Acids, Esters and their Derivatives (eds S. Pata¨ı and Z. Rappoport), John Wiley & Sons, Inc., New York, pp. 789–878; (c) Metz, P. (1998) J. Prakt. Chem., 340, 1–10. Palytoxin: (a) Uemura, D., Ueda, K., Hirata, Y., Naoki, H., and Iwashita, T. (1981) Tetrahedron Lett., 22, 2781–2784; Ostreocin: (b) Ukena, T., Satake, M., Usami, M., Oshima, Y., Naoki, H., Fujita, T., Kan, Y., and Yasumoto, T. (2001) Biosci. Biotechnol. Biochem., 65, 2585–2588; (c) (+)-Discodermolide: Gunasekera, S.P., Gunasekara, M., and Longley, R.E. (1990) J. Org. Chem., 55, 4912–4915; additions and corrections (d) Gunasekera, S.P., Gunasekera, M., and Longley, R.E. (1991) J. Org. Chem., 56, 1346; Dictyostatin: (e) Pettit, G.R., Chichacz, Z.A., Gao, F., Boyd, M.R., and Schmidt, J.M. (1994) J. Chem. Soc., Chem. Commun., 1111–1112; (f) Paterson, I., Britton, R., Delgado, O., Meyer, A., and Poullennec, K.G. (2004)

Angew. Chem. Int. Ed., 43, 4629–4633; (g) Shin, Y., Fournier, J.-H., Fukui, Y., Brückner, A.M., and Curran, D.P. (2004) Angew. Chem. Int. Ed., 43, 4634–4637; Neodihydrohistrionicotoxin: (h) Tokuyama, T., Yamamoto, J., Daly, J.W., and Highet, R.J. (1983) Tetrahedron, 39, 49–53; Mycothiazole: (i) Crews, P., Kakou, Y., and Qui˜ noà, E.J. (1988) J. Am. Chem. Soc., 110, 4365–4368. 11 Palladium/Nickel Cross Coupling: (a) Diederich, F. and Stang., P.J. (eds) (1998) Metal-Catalyzed Cross-Coupling Reactions, Wiley-VCH Verlag GmbH, Weinheim; (b) Nerenburg, J.B., Hung, D.T., Somers, P.K., and Schreiber, S.L. (1993) J. Am. Chem. Soc., 115, 12621–12622; (c) Sugiyama, H., Yokokawa, F., and Shioiri, T. (2003) Tetrahedron, 59, 6579–6593; (d) Porée, F.-H., Barbion, J., Dhulut, S., Betzer, J.-F., Pancrazi, A., and Ardisson, J. (2004) Synthesis, 18, 3017–3022; Nozaki-Hiyama-Kishii Coupling: (e) Armstrong, R.W., Beau, J.-M., Cheon, S.H., Christ, W.J., Fujioka, H., Ham, W.-H., Hawkins, L.D., Jin, H., Kang, S.H., Kishi, Y., Martinelli, M.J., McWhorter, W.W. Jr., Mizuno, M., Nakata, M., Stutz, A.E., Talamas, F.X., Taniguchi, M., Tino, J.A., Ueda, K., Uenishi, J., White, J.B., and Yonaga, M. (1989) J. Am. Chem. Soc., 111, 7525–7530; Wittig Olefination: (f) Stork, G. and Zhao, K. (1989) Tetrahedron Lett., 30, 2173–2174; Allylic Organometallic Species: (g) Ikeda, Y., Ukai, J., Ikeda, N., and Yamamoto, H. (1987) Tetrahedron, 43, 723–730; (h) Hodgson, D.M. and Wells, C. (1992) Tetrahedron Lett., 33, 4761–4762; (i) Tsai, D.J.S. and Matteson, D.S. (1981) Tetrahedron Lett., 22, 2751–2752; (j) Roush, W.R. and Grover, P.T. (1992) Tetrahedron, 48, 1981–1998; Cu-Promoted Coupling with Alkynyl Silane: (k) Marshall, J.A., Chobanian, H.R., and Yanik, M.M. (2001) Org. Lett., 3, 4107–4110. 12 Structure of mycothiazole was later determined that the C14-C15 was incorrectly assigned E instead of Z: Sonnenschein, R.N., Johnson, T.A.,

145

146


13

14

15 16

17

18

19

Tenney, K., Valeriote, F.A., and Crews, P. (2006) J. Nat. Prod., 69, 145–147. (a) Corey, E.J. and Boaz, N.W. (1984) Tetrahedron Lett., 25, 3063–3066; (b) Yanagisawa, A., Noritake, Y., Nomura, N., and Yamamoto, H. (1991) Synlett, 251–253; (c) Lipshutz, B. (1990) Synlett, 119–128; (d) Magid, R.M. (1980) Tetrahedron, 36, 1901–1930; (e) Piarulli, U., Daubos, P., Claverie, C., Roux, M., and Gennari, C. (2003) Angew. Chem. Int. Ed., 42, 234–235. (a) Takai, K., Oshima, K., and Nozaki, H. (1980) Tetrahedron Lett., 21, 2531–2534; (b) Sato, M., Takai, K., Oshima, K., and Nozaki, H. (1981) Tetrahedron Lett., 22, 1609–1612; (c) Fugami, K., Oshima, K., and Utimoto, K. (1987) Chem. Lett., 2203–2206; (d) Nicolaou, K.C. and Namoto, K. (1998) Chem. Commun., 1757–1758; (e) Lepifre, F., Clavier, S., Bouyssou, P., and Coudert, G. (2001) Tetrahedron, 57, 6969–6975; (f) Kadota, I., Takamura, H., Sato, K., and Yamamoto, Y. (2001) Tetrahedron Lett., 42, 4729–4731. Morin, M.D. and Rychnovsky, S.D. (2006) Org. Lett., 7, 2051–2053. (a) Bartlett, P.A. and Jernstedt, K.K. (1977) J. Am. Chem. Soc., 99, 4829–4830; (b) Bartlett, P.A. and Jernstedt, K.K. (1980) Tetrahedron Lett., 21, 1607–1610. (a) Letsinger, R.L., Finnan, J.L., Heavner, G.A., and Lunsford, W.B. (1975) J. Am. Chem. Soc., 97, 3278–3279; (b) Beaucage, S.L. and Caruthers, M.H. (1996) in Bioorganic Chemistry: Nucleic Acids (ed. S.M. Hecht), Chapter 2, Oxford University Press, New York, pp. 36–74. (a) Burke, S.D., Muller, N., and Beaudry, C.M. (1999) Org. Lett., 1, 1827–1829; (b) Burke, S.D. and Voight, E.A. (2001) Org. Lett., 3, 237–240; (c) Lambert, W.T. and Burke, S.D. (2003) Org. Lett., 5, 515–518. (a) Deslongchamps, P. (1983) Stereoelectronic Effects in Organic Chemistry, Pergamon Press, Oxford; (b) Vande Griend, L.J., Verkade, J.G., Pennings, J.F.M., and

20 21

22

23

24

Buck, H.M. (1977) J. Am. Chem. Soc., 99, 2459–2463. Pearson, R.G. (1963) J. Am. Chem. Soc., 85, 3533–3539. (a) Westheimer, F.H. (1987) Science, 235, 1173–1178; (b) Thatcher, G.R.J. and Kluger, R. (1989) Mechanism and catalysis of nucleophilic substitution in phosphate esters. Adv. Phys. Org. Chem., 25, 99–265; (c) Miyashita, K., Ikejiri, M., Kawasaki, H., Maemura, S., and Imanishi, T. (2003) J. Am. Chem. Soc., 125, 8238–8243; (d) Cox, J.R. Jr. and Ramsay, J.O.B. (1964) Chem. Rev., 64, 317–352; (e) Bunton, C.A. (1970) Acc. Chem Res., 3, 257–265; (f) Westheimer, F. (1968) Acc. Chem. Res., 1, 70–78; (g) Gorenstein, D.G., Rowell, R., Taira, K. (1981) Stereoelectronic Effects in Phosphate Esters, ACS Symposium Series, vol. 171, pp. 69–75; (h) Fanni, T., Taira, K., Gorenstein, D.G., Vaidyanathaswamy, R., and Verkade, J.G. (1986) J. Am. Chem. Soc., 108, 6311–6314 and references cited therein. (a) Yanagisawa, A., Noritake, Y., Nomura, N., and Yamamoto, H. (1991) Synlett, 4, 251–253; (b) Belelie, J.L. and Chong, M. (2001) J. Org. Chem., 66, 5552–5555; (c) Calaza, M.I., Hupe, E., and Knochel, P. (2003) Org. Lett., 5, 1059–1061; (d) Piarulli, U., Daubos, P., Claverie, C., Roux, M., and Gennari, C. (2003) Angew Chem. Int. Ed. Engl., 42, 234–236; (e) Piarulli, U., Claverie, C., Daubos, P., Gennari, C., Minnaard, A.J., and Feringa, B.L. (2003) Org. Lett., 5, 4493–4496; (f) Kacprzynski, M.A. and Hoyveda, A.H. (2004) J. Am. Chem. Soc., 126, 10676–10681; (g) Larsen, A.O., Leu, W., Oberhuber, C.N., Campbell, J.E., and Hoyveda, A.H. (2004) J. Am. Chem. Soc., 126, 11130–11131. Chatterjee, A.K., Choi, T.-L., Sanders, D.P., and Grubbs, R.H. (2003) J. Am. Chem. Soc., 125, 11360–11370. Dolabelides A-D: (a) Ojika, M., Nagoya, T., and Yamada, K. (1995) Tetrahedron Lett., 36, 7491–7494; (b) Suenaga, K., Nagoya, T., Shibata, T., Kigoshi, H., and Yamada, K. (1997) J. Nat. Prod., 60, 155–157; Salicylihalamides: (c) Wu, Y., Seguil, O.R., and

References De Brabander, J.K. (2000) Org. Lett., 2, 4241–4244; (d) Holloway, G.A., Hugel, H.M., and Rizzacasa, M.A. (2003) J. Org. Chem., 68, 2200–2204; (e) Snider, B. and Song, F. (2001) Org. Lett., 3, 1817–1820; Bitungolides A-F: (f) Sirirath, S., Tanaka, J., Ohtani, I.I., Ichiba, T., Rachmat, R., Ueda, K., Usui, T., Osada, H., and Higa, T. (2002) J. Nat. Prod., 65, 1820–1823; Rhizoxin D: (g) Lafontaine, J.A., Provencal, D.P., Gardelli, C., and Leahy, J.W. (1999) Tetrahedron Lett., 40, 4145–4148;

(+)-Discodermolide: (h) Smith, A.B.III, Kaufman, M.D., Beauchamp, T.J., LaMarche, M.J., and Arimoto, H. (1999) Org. Lett., 1, 1823–1826. 25 Myers, A.G., Zheng, B., and Movassaghi, M. (1997) J. Org. Chem., 62, 7507. 26 (a) Hughes, G., Lautens, M., and Wen, C. (2000) Org. Lett., 2, 107–110; (b) Chau, A., Paquin, J.-F., and Lautens, M. (2006) J. Org. Chem., 71, 1924–1933. 27 Lucas, B.S., Luther, L.M., and Burke, S.D. (2004) Org. Lett., 6, 2965–2968.

147

149

5 Synthesis of Natural Products Containing Macrocycles by Alkene Ring-closing Metathesis Ana Gradillas and Javier Pérez-Castells

5.1 Introduction

The impact of metathesis in modern synthetic chemistry is evidenced by the number of publications and reviews that have appeared in the last decade, some of them covering the specific field of macrocyclization in natural product synthesis [1]. It is therefore difficult not to repeat the reports done by other authors or duplicate their critical views on the chemistry. This chapter is exclusively devoted to the total syntheses of natural bioactive macrocycles involving diene and the diene–ene metathesis, and focuses on the reaction conditions used in the macrocyclization key step and on the stereochemical outcome of the reaction. As we shall see, the choice of the catalyst, solvent, temperature, concentration, and reaction time is crucial in these transformations. In addition, we show how subtle variations on the structure of the substrate lead to modifications in the conformational organization of the molecule, which are critical for the success of ring-closing metathesis (RCM). In particular, the substitution pattern, the steric hindrance of the substituents, the ring size to be formed, and the presence of coordinating heteroatoms are particularly important as they have an important influence on the outcome of the reaction. This chapter has been organized around the various families of natural products in order to compare the reaction conditions for compounds bearing structural similarities. We have deliberately decided to reveal only the information regarding the metathesis step and thus not disclose other details of the syntheses. Synthetic efforts toward macrocycles are often laborious and unpractical. The classical strategies involved are macrolactonizations, macrolactamizations, and macroaldolizations. The macrocyclization approach encounters several problems including the competition between the desired intramolecular reaction and intermolecular processes, which result in the formation of polymers. RCM has the advantage of being compatible with a wide variety of functional groups and forming double bonds that can eventually be converted into other functionalities. The main disadvantages are the control of the stereochemistry of the newly formed double bond and the optimization of the yield. On the other hand, the concomitant loss of ethylene makes this reaction highly atom economical and entropically Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts. Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32440-8

150

5 Synthesis of Natural Products Containing Macrocycles by Alkene Ring-closing Metathesis

driven. Regarding the prediction of the configuration of the newly formed double bond, cyclization of 1,6-, 1,7-, 1,8-, and 1,9-dienes to, respectively, 5-, 6-, 7-, and 8-membered rings generally gives the thermodynamic (Z)-alkenes, while larger rings are obtained as E/Z mixtures where the (E)-isomer is dominant. For cycle going from 10- to 14-membered rings, however, the outcome is not as trivial and the factors that establish the E/Z selectivity still remain unclear. In addition, the E/Z ratio of the product may be influenced by secondary isomerizations mediated by the metathesis complexes. In our study, macrocycles are defined as compounds containing at least one ring with 12 or more covalently connected atoms. It is worth noting that they constitute more than 3% of the total amount of natural products described to date. Approximately, 35–40% of all current trade drugs are natural products, or analogs [2]. The presence of macrocycles is particularly common in antitumorals, antibiotics, and antifungals. Nature has done a selection along millions of years to reach bioavailable compounds with good ability to interact with natural macromolecules. The balance between specificity and affinity allows these products, which come generally from bacteria and fungi, to interact efficiently not only with their natural molecules but also with certain receptors of mammals, thus becoming useful for humans. Probably there are not so huge differences between the structure of the natural targets and those present in humans, but, obviously, there is a great functional difference [3]. Using natural products to access leads is therefore highly desirable, in particular, naturally occurring macrocycles that have potent biological activities. As they usually have complex structures, with a variety of chemical motifs and ring sizes, their total synthesis plays a crucial role in order to confirm their structure and to develop derivatives that can modulate the biological activity or the pharmacokinetic properties. Macrocycles have equilibrium between flexibility and conformational preorganization to achieve optimal binding properties with respect to their biological target. However, they are many times toxic and cause undesired effects, which make necessary their derivatization. With regard to the structures, the most common ring size goes from 11- to 16-membered rings, with a special incidence for 14-membered rings. This group is dominated either by cycles containing only carbon atoms or by macrocyclic lactones and ethers. The first comes biogenetically from the mevalonate pathway and comprises compounds with around 20 carbons that correspond to diterpenoids. One particularly interesting group of macrocycles are those bearing aromatic rings in their structure including phenyl ethers, ansa cycles, and biphenyl structures. They generally have cycles going from 14- to 20-membered rings. The other abundant group is formed by oxygen containing compounds, including macrolactones with a possible polyketide origin. These compounds are generally named macrolides, and some of them have two lactone groups and are known as dienolides. In addition, some 13- to 15-membered oxygen containing macrocycles are not olides but ethers. It is presumed that they come from alkaloid origin. The majority of the nitrogen containing macrocycles are lactams, and generally have larger cycles reaching sometimes 30-membered cycles. Many of these are cyclopeptides,

5.2 Organization of the Chapter

possessing three to six amino acids. The presence of nonnatural amino acids and unnatural stereochemistry is frequent [4].

5.2 Organization of the Chapter

In this chapter, a special emphasis is given on the results of the RCM macrocyclization step. This key step is illustrated in schemes in which the starting material used for the metathesis, the resulting product, the yield and the eventual ratio of stereoisomers, the best reaction conditions used, the structure of the natural product, and the references are showed. The classification of the compounds follows the biogenetic one and the family group. The variety in natural macrocyclic structures is high but the biosynthetic pathways that give these products are limited to four main origins and several subsequent transformations, including the cyclization step. The biosynthesis of macrocycles can follow the polyketide pathway, that is, the acetate metabolism, which leads to iterative C2 unit couplings and an aldol or Claisen type cyclization. The second pathway is the mevalonic pathway, that is, isoprenoid metabolism, involving the coupling of C5 units and cyclization, which gives mostly diterpenes. A third group of macrocycles come from the metabolism of amino acids including cyclopeptides, cyclodepsipeptides, and alkaloids. Finally, the metabolism of sugars leads to cyclic oligosaccharides and glycolipids; the latter is a combination with the first pathway, as they include fatty acid units in their structure. Following the biogenetic classification, we first address macrolactones, divided into the main family groups that have received synthetic attention via RCM (resorcinylic, salycilates, other antibiotic macrolides, macrocyclic musks, epothilones, amphidinolides, other polyketides, and natural cyclophanes), then the terpenoids, followed by the macrocycles obtained from the amino acid metabolism (lactams, depsipeptides, alkaloids), and finally the glycolipids. From the historical point of view this story begins in the late 1990s, mainly with the RCM-based epothilone syntheses [5], and the early synthesis of the 18-membered α,β-unsaturated macrolide aspicilin [6]. It is worth noting that previously, some groups had disclosed RCM to form macrocycles without completing the total synthesis of the natural products [7]. The first example of regioselective diene–ene metathesis was disclosed in 1999, when the synthesis of simplified macrolide analogs of the immunosuppressant sanglifehrin was carried out [8]. With regard to the reaction conditions used, three generations of ruthenium complexes and the molybdenum Schrock catalyst have been used. Catalyst [Ru]-I is the cheapest, although it is thermally unstable and with low reactivity implying its failure with substituted olefins [9]. However, in some cases, this low reactivity is desirable and in the most recent contributions these complexes are still used. Thermal stability plays a critical role in the lifetime of the species and turnover number (TON), and it is known that [Ru]-I decomposes after 1 hour at 100 ◦ C in toluene-d8 up to 75%. Some derivatives of this first-generation catalyst such as [Ru]-IV and

151

152


[Ru]-V improve the kinetic initiation and have been used, although generally in early contributions. The second-generation catalysts [Ru]-II and [Ru]-VI give better results with substituted olefins, as does also the Hoveyda–Grubbs catalyst [Ru]-III. Because of the incorporation of the nitrogen containing heterocycle (NHC), such as in [Ru]-III, the stability of the ruthenium complex is increased (stable after 1 hour at 100 ◦ C); it can be recovered by chromatography and reused. This latter complex has been modified also by substitution of the aromatic ring, giving rise to a new family of complexes that can be named ‘‘third-generation’’ catalysts. It is worth noting that the newly described water compatible ruthenium catalysts [Ru]-VII and [Ru]-VIII have not yet found use in macrocyclization chemistry. The molybdenum species [Mo]-I is highly reactive but sensitive to water and oxygen and not compatible with several functionalities, thus it has been much less used than the other catalysts in macrocyclization [10].

5.3 Macrocyclic Polyketides

Many macrocyclic structures are polyketides. The majority of them have one or more lactone linkages (poliolides) and are generally referred to as macrolides. Since they present interesting biological activities, they are ideal targets for synthetic studies. These compounds are isolated from bacteria or fungi and are in general of marine origin [11]. 5.3.1 Resorcinylic Macrolides

The resorcinylic macrolides (also known as resorcinylic acid lactones or RALs) are a family of naturally occurring homologous macrolides. They are derived from polyketide biogenesis in different fungi and have been isolated from different fungal strains [12]. They have unique chemical structures and potent biological activities (antitumoral, antibiotic, and antimalarial activities) [13]. Among these compounds the most popular are radicicol (3) and its derivatives. In addition, aigialomycin D (7), zeranol (25), zeralenone (28), and the pochonins (16, 19, 22) share the basic skeleton consisting of a 14-membered ring lactone that includes a 1,2-phenylene. The major contribution in this field was achieved by Danishefsky et al. who, despite the relative similarity of the target molecules, have developed a wide variety of synthetic strategies and conditions. The disconnections and the studies on the RCM reaction conditions developed by this group were reviewed by Winssinger et al. [12]. Among the biological activities of the resorcinylic macrolides, the ability of these compounds to inhibit Hsp90 (heat shock protein 90) is outstanding. The syntheses of radicicol (3) are based on RCM of a diene–ene system, which avoids the problem of regioselectivity, compatibility with labile groups such as a vinyl epoxide, the dithiane protecting group, and the E/Z selectivity [14]. In these


O

O

O TBSO

H [Ru]-II (10 mol%) CH2Cl2 (2 mM), 42 °C, 2 h TBSO

H S

S

H

O

O

O

O

H

60%

H

HO

S

RCM (diene–ene system) X O

S

OTBDPS

OTBDPS

OH Radicicol (3), X = Cl Monocillin (4), X = H

2 (14-membered)

1

H

O

O

153

Scheme 5.1

initial syntheses, the role of both the temperature and the concentration appeared to have a tremendous impact on the outcome of RCM [15]. In compound 1, RCM occurred under mild conditions (2 mM, CH2 Cl2 , 42 ◦ C) to give the monomeric macrocyclic product 2 (Scheme 5.1). However, when similar conditions were applied to the cyclopropyl precursor (X = CH2 ), the major isolated product was a dimeric 28-membered ring macrocycle. On the other hand, performing the reaction at 110 ◦ C in toluene and quenching the reaction after a short period resulted in a clean conversion and a 54% yield of the desired macrocycle (Scheme 5.2). In a more recent and improved approach to cycloproparadicicol (synthetic cyclopropane analog of radicicol) the synthesis was achieved via a stable cobalt complex, which led to the cyclized products as a 2 : 1 mixture of isomers in 57% yield (Scheme 5.3). The complexation of the ethynyl group with Co2 (CO)8 avoids a competitive ene–yne metathesis and brings into proximity the double bonds favoring RCM. This so-called ynolide methodology was also applied to the synthesis of aigialomycin D (7), a natural product with antimalarial and antitumoral activity, and also a CDK1 and CDK5 inhibitor, isolated from marine mangrove fungus Aigialus parvus (Scheme 5.4) [16]. Oxidative metal decomplexation followed by Diels–Alder reaction using a disiloxydiene was used to elaborate the benzenic

O

X

O

H

H

TBSO S

[Ru]-II (5 –10 mol%)

S

H O

H

TBSO S

X = O: CH2Cl2, 42 °C, 2 h (60%) X = CH2: CH2Cl2, 42 °C,

OTBDPS

X

O

O

X

O

H

HO

S

H

O OH X = O, Radicicol (3) X = CH2, Cycloproparadicicol

OTBDPS

18 h (38% + dimer) X = CH2: PhCH3, 110 °C, 10 min (54%)

Scheme 5.2

O

O

O

R R' Co (CO)3 Co (CO)3

Scheme 5.3

OTBS

O

O [Ru]-II (25 mol%), CH2Cl2 R = Me, R' = H, 45 °C (57%) R = H, R' = Me, 25 °C (38%)

O R

(OC)3Co

Co (CO)3

O R' OTBS

O

Cycloproparadicicol (the ynolide approach)


154

O (OC)3Co

O O

(OC)3Co TBSO

O

[Ru]-II (25 mol%)

(OC)3Co

CH2Cl2 (0.2 mM), rt, 10 h

(OC)3Co

38%, (E )-isomer only

OH

O

O

O RCM

HO

TBSO

OH

O O

O 5

OH

6 (14 -membered)

Aigialomycin D (7)

Scheme 5.4

system. The main advantage of this route is the possibility of scaling up the synthesis producing, for example, multigram of cycloproparadicicol. More recently, Winssinger et al. reported a flexible polymer-supported synthesis of aigialomycin D (7) and a library of analogs thereof [17]. The syntheses of cis- and trans-resorcilides (10, 13), which are plant growth inhibitors isolated from Penicillium sp., were based on RCM of enolizable enones. For the cyclization of diene 8, a combination of high dilution in refluxing CH2 Cl2 (0.5 mM) and low catalyst loading (2 mol%) was necessary to reach reasonable yields of 9 (Equation 1, Scheme 5.5). Diene 11 gave trans-alkene 12 as the unique stereoisomer in high yield (Equation 2, Scheme 5.5) [18]. Pochonins are a new and interesting group of resorcinylic lactones isolated from Pochonia chlamydosporia. Pochonin C (16) is active against HSV (Herpes Simple Virus); however, it has less inhibitory activity than radicicol against Hsp90 due to a different conformational behavior of the macrocycle. On the other hand, pochonin D (22) can adopt the so-called L-shape bioactive conformation against Hsp90 with a small energy penalty, and is a stable and structurally simple Hsp90 inhibitor. More structurally related to radicicol, pochonin A (19) has both Hsp90 and HSV inhibition activity. Winssinger et al. have disclosed the syntheses of this family of RALs. Polyene RCM was used for the synthesis of pochocin C (16) (Equation 1, Scheme 5.6) [19]. Thus, treatment of intermediate 14 with catalyst [Ru]-II in refluxing toluene led exclusively to 15 in 10 minutes. The synthesis of pochonin A H

O

O

[Ru]-II (25 mol%) CH2Cl2 (0.5 mM), rt, 1 h

O Equation 1

O

O H

O

O

O

40%, (Z )-isomer only

O

O

OH

OH

cis -Resorcylide (10)

9 (12-membered)

O

BnO

[Ru]-II (10 mol%) O

Equation 2


BnO O

11

Scheme 5.5

OH O

O O

CH2Cl2 (1 mM), rt, 30 min

RCM

O

OH 8

BnO

O

H

O

BnO O

12 (12-membered)

HO

O RCM

trans -Resorcylide (13)

5.3 Macrocyclic Polyketides O

MOMO

O

O H

MOMO Equation 1

[Ru]-II (5 mol%) PhCH3 (2 mM), reflux, 10 min H

87%

MOMO

OH

O O

O H

Cl HO

HO Cl

O

O

O RCM (diene – ene system)

14 15 (14-membered) TBSO

O

[Ru]-II (10 mol%) O

TBSO

TBSO

Pochonin C (16)

O

OH O

PhCH3 (2 mM), 80 °C, 12 h 79% (E /Z > 95 : 5)

Cl

O

O TBSO

HO

O RCM

Cl

O

Equation 2

O

O H

MOMO

155

O 17

MOEO

18 (14-membered) O MOEO

[Ru]-II (10 mol%) O EOMO O

O

PhCH3 (2 mM), 120 °C, 10 min 90% (E /Z = 4 : 1)

Cl Equation 3

Pochonin A (19)

OH O

EOMO

O HO

Cl O

20

O

21 (14-membered)

Cl O Pochonin D (22)

Scheme 5.6

(19) [20] and D (22) [21] was achieved, respectively, from 17 and 20 using [Ru]-II in refluxing toluene (Equations 2 and 3, Scheme 5.6). For the synthesis of 19, the double bond resulting from RCM was epoxidized. Included in this resorcinylic group are a number of 12-membered ring lactones. On the basis of the RCM of ketone 23, Fürstner et al. developed a synthetic approach to zeranol (25), a nonsteroidal estrogenic growth promoter (Equation 1, Scheme 5.7). The same authors synthesized (S)-(−)-zearalenone (28) and (R)-lasiodiplodin (31), two inhibitors of the prostaglandin biosynthesis isolated from the fungus Botryosphaeria rhodina [22]. In all these syntheses, a complete lack of reactivity was observed when [Ru]-I was utilized. On the contrary, when [Ru]-II and [Ru]-V were used, the cyclized products were isolated in good to excellent yield (Equations 2 and 3, Scheme 5.7). 5.3.2 Salicylate Macrolides

One closely related family to resorcinylic macrolides are the salicylate enamide macrolides, which are secondary marine metabolites with a 12-membered ring benzolactone and a dienyl enamide side chain. Salicylihalamide A (34) is a potent inhibitor of the mammalian vacuolar ATPase and exerts great differential cytotoxicity in the National Cancer Institute (NCI) 60 cell lines [23]. Other members

RCM


156

OMe O

OH

OMe O

O

[Ru]-I (19 mol%) O

O

CH2Cl2, reflux, 3 d

O

O 79% (E/Z = 2.4 : 1) MeO

MeO

HO

Equation 1

O

23

24 (14-membered)

OMe O O

91%, (E )-isomer only MeO

O

Equation 2

Zeranol (25)

OH

OMe O

[Ru]-V (5 mol%) PhCH3, 80 °C, 4 h MeO

O

O

HO

27 (14-membered)

(S )-(−)-Zearalenone (28)

OMe O

OMe O [Ru]-V (5 mol%) PhCH3, 80 °C, 15 h

O


MeO

O

RCM

O

OMe O

O O

O

26

OH

RCM Hydrogenation

O

O HO

MeO

RCM Hydrogenation

Equation 3 29

30 (12-membered)

(R)-(+)-Lasiodiplodin (31)

Scheme 5.7

of this family are the apicularens, which bear an identical enamide side chain; the lobatamides A–F; and the oximidines II and III, which possess an enamide side chain terminated by an O-methyl oxime [24]. Mitsunobu/RCM strategy, Suzuki cross-coupling, Stille cross-coupling are the three general approaches for the synthesis of salicylihalamides A (34) and B (35) [25]. After a thorough optimization, the best reaction conditions involved the use of [Ru]-II (10 mol%) in toluene at 70 ◦ C for 2 hours (Scheme 5.8). However, the synthesis of 34 has been the subject of many efforts in order to obtain a good (E/Z) stereoselectivity and deserve some comments (Table 5.1). The remote phenolic hydroxy group is of importance as it favors the undesired NH O OPMB

OMe O O

OMOM

[Ru]-II (10 mol%) PhCH3 (1mM), 70 °C, 2 h

OPMB

OMe O O

OMOM

75%, (E )-isomer 12%, (Z)-isomer 32

O

OH

RCM 33 (12-membered)

Scheme 5.8

OH O

Salicylihalamide A: C17-(E ) (34) Salicylihalamide B: C17-(Z ) (35)

5.3 Macrocyclic Polyketides Table 5.1

157

Conditions for the synthesis of salicylihalamide precursors.

R2 R4 Remote controlling group

O

O

R4

OR3

R1O

R1

R2

R3

R4

Catalyst

Solvent

T (◦ C) Time

Yield (%)

(E/Z) Reference

Me TBS H Me Me MOM MOM H Me TBS MOM

CH=CH–CH2 OPMB CH=CH–COOMe CH=CH–COOMe CH2 –CH2 OBn CH2 OPMB CH2 OPMB CH2 OPMB CH2 OPMB CH2 OPMB CH2 OPMB CH2 OPMB

TBS H H TBDPS MOM MOM MOM MOM MOM MOM MOM

H H H H H H H Me Me Me H

[Ru]-I [Ru]-I [Ru]-I [Ru]-I [Ru]-I [Ru]-I [Ru]-I [Ru]-VI [Ru]-VI [Ru]-VI [Ru]-II

CH2 Cl2 CH2 Cl2 CH2 Cl2 CH2 Cl2 CH2 Cl2 CH2 Cl2 CH2 Cl2 PhCH3 PhCH3 PhCH3 CH2 Cl2

rt Rt Rt Rt 80 80 80 rt

85 57 Low 75 95 95 93 69 93 91 98

10/1 4/1 Z 9/1 9/1 9/1 5/1 Z 2/1 1/1.5 2/1

40 min 4d 4d 3.0 h 1.3 h 1.3 h 18 h 20 h 1.5 h 1.0 h 1.3 h

(Z)-isomer. If the phenolic hydroxy group is protected, the (E)-isomer is the major compound. The influence of the catalyst is also crucial, as with catalyst [Ru]-I, which was used by Smith et al. [26], Snider et al. [27], and Labreque et al. [28], a high (E)-selectivity was observed, especially if the phenolic hydroxy group was protected as a methyl ether (Table 5.1). De Brabander et al. [29] showed that this process was under kinetic control. On the other hand, with the second-generation catalysts used by Fürstner et al. [30] and de Brabander et al., the equilibrium is reached thus leading to a E/Z ratio of 2 : 1. An isomerization process can compete during the experiment. Oximidine II (38) and III (41), isolated from Pseudomonas, were constructed using a relay RCM strategy. Wang et al. and Porco et al. reported, in the synthesis of oximidine II (38), the first example of a successful diene–diene RCM to construct a macrocyclic conjugated triene. A well-defined substrate possessing two differentially functionalized RCM alkene partners was required for the RCM process. The presence of the terminal methyl group on the conjugated diene with an (E,Z) configuration and the use of [Ru]-II catalyst were crucial to avoid oligomeric products or the formation of smaller cycles resulting from the reaction with internal double bonds. Fast addition of the substrate to the catalyst solution resulted in reduced decomposition of the starting material (Scheme 5.9) [31].

[26] [27] [27] [28] [29] [29] [29] [30] [30] [30] [29]

158


OPMB TBSO

O O

Equation 1

H

H N

OPMB [Ru]-II (5 mol%) OMOM CH Cl (2 mM), TBSO 2 2 reflux, 70 min

O

OMOM O

O H

TBSO

OPMB

O O

RCM (bis–diene system) Oximidine II (38) H N

37 (12-membered)

OPMB

H

Equation 2

O

[Ru]-III (10 mol%) CH2Cl2 (2 mM), TBSO 50 °C, 30 min

N OMe

H

48%

36

O OH

OH O

O O

N

OH O

OMe O

71%

H

O

O

H

O

RCM (diene–ene system) 39

40 (12-membered)

Oximidine III (41)

Scheme 5.9

5.3.3 Other Antibiotic Macrolides

Numerous biologically active macrocycles are antibiotics isolated from various microorganisms, with a wide variety of structures and ring sizes ranging from 12to 28-membered rings. Fourteen natural antibiotic lactones and diolides synthesized using RCM are shown in Schemes 5.10–5.24. 10-Deoxymethynolide (44) is the aglycone of the methymycin family of macrolides and was constructed via an exclusive E selective O

O [Ru]-II CH2Cl2, rt, 12 h

O

Equation 1

OTBS

O

O


42

O RCM

O

O OTBS

43 (12-membered) O

O

OH

O

10-Deoxymethynolide (44) O RCM

Equation 2

[Ru]-II CH2Cl2, rt, 12 h

O

O

OPMB 88%, (E )-isomer only

O

OTBS 45

Scheme 5.10

O

OPMB OTBS

46 (14-membered)

O O

OH O

Narbonolide (47)

5.3 Macrocyclic Polyketides Ph

Ph

O

O [Ru]-II BHT, PhCH3, reflux, 6 h

O O

PMBO

OTBS

O

159

46%

48

OH O

O

PMBO O

OAc OH

RCM

O

O OTBS

49 (12-membered)

OH

O

R

Pladienolide B: R = H (50) Pladienolide D: R = OH (51)

Scheme 5.11

RCM using [Ru]-II in CH2 Cl2 at room temperature (Equation 1, Scheme 5.10). The same conditions were used for the synthesis of narbonolide (47), which possesses a 14-membered ring lactone instead of a 12-membered ring. Narbonolide (47) is the aglycone of the pikromycin family of antibiotic macrolides (Equation 2, Scheme 5.10) [32]. Recently, the synthesis of two members of methymycins, namely, pladienolides B (50) and D (51), possessing antineoplasic activity and isolated from Streptomyces platensis, was described. These examples used a less efficient RCM but a better catalyst loading was employed (Scheme 5.11) [33]. In addition to these four compounds, other similar 12-membered ring lactones were obtained such as sporiolode B (54), isolated from fungal strain Cladosporium sp. [34], and (10S,12R)-10-hydroxy-12-methyl-1-oxacyclododecan-2,5-dione (57) [35], which were formed by RCM leading to (E/Z) mixtures (Equations 1 and 2, Scheme 5.12). Two early examples disclosed by Fürstner et al., ricinelaidic acid (60) and gloeosporone (63) (germination self-inhibitors isolated from Colletotrichum gloeosporioide), were built by using [Ru]-I (Equations 1 and 2, Scheme 5.13). In the synthesis

OBn Equation 1

O

O O

O

O O

OMe

OTPS

O

[Ru]-II (10 mol%) CH2Cl2, reflux, 12 h

O

O OMe

Sporiolide B (54)

O

O

OTPS BnO

55

O O

72%, (E /Z )-mixture

BnO

OH

OMe

53 (12-membered)

O

Equation 2

Scheme 5.12

OBn

70% (E /Z = 2 : 1)

52

O

RCM Hydrogenation

[Ru]-I (30 mol%) CH2Cl2 (0.1 mM), reflux, 24 h

56 (12-membered)

OH O MCR Hydrogenation 57


160

O

O


O

Equation 1

O O

O

72% RCM 59 (13-membered)

58

(R)-(+)-Ricinelaidic Lactone (60)

[Ru]-I (3 mol%) CH2Cl2, Ti(OiPr)4 (30 mol%), 40 °C, 96 h

Equation 2

OTBS O

O OH OR O

80% (E/Z = 2.7 : 1)

O

61

O O

O

62 (14-membered)

O

(−)-Gloeosporone (63)

Scheme 5.13

of gloeosporone (63), RCM was performed in the presence of a Lewis acid in an attempt to avoid any interaction between the oxygen electron pairs and the catalyst. In both cases, yields were good – although low (Z/E) selectivities were observed [36]. Migrastatin (66), isolated from S. platensis, inhibits the migration of human tumor cells. This compound was synthesized via RCM of the fully functionalized tetraene 64 (Scheme 5.14). The reaction was highly chemoselective and proceeded in a highly (E)-selective fashion [37]. Among the three RCM-based syntheses of aspicilin (69) – a natural product isolated from Lecanoraceae lichen – only the most efficient one is reported herein (Scheme 5.15). Important variations in the (E/Z) selectivity were observed; however, this had no detrimental impact on the synthesis, as hydrogenation of the olefin was required in order to obtain the natural product [38]. Sch 351448 (72), isolated from Micromonospora, is a novel activator of low-density lipoprotein receptor (LDL-R) promoter with an IC50 of 25 nM, which features a 28-membered macrodiolide consisting of two identical hydroxy carboxylic acid units. In its synthesis, intramolecular olefin metathesis of 70 mediated by [Ru]-II proceeded smoothly, and the macrodiolide was obtained after a hydrogenation/hydrogenolysis sequence (Scheme 5.16) [39]. Other syntheses of this O

O

O [Ru]-II (20 mol%) PhCH3 (0.5 mM), reflux, 15 min

O O O

N H

O

O

O

O

OTBS OMe 64

Scheme 5.14

O

O


OTBS

N H

OMe 65 (14-membered)

O

O

N H

OH RCM

OMe

(+)-Migrastatin (66)

O

5.3 Macrocyclic Polyketides RCM Hydrogenation O

O [Ru]-II (5 mol%) CH2Cl2, 1.5 mM, rt

O

161

OH OH

O

OMOM 83%, (Z )-isomer only

OH

OMOM

O

O

O O

O

67

O

RCM

(+)-Aspicilin (69)

68 (18-membered)

Scheme 5.15

OH O

O

O

OH

OH

[Ru]-II (10 mol%) CH2Cl2 (3 mM),

O

O O

O

O

O

OH O

reflux, 10 h

R R

RCM Hydrogenation

R O

O

O

O

HO

O

O

HO

HO

70

O

OH O

O

O

CO2H 71 (28-membered)

(+)-Sch 351448 (72)

Scheme 5.16

compound have also appeared in the literature, the most recent one, in 2008, involved a templated olefin metathesis strategy to assemble the C2 -symmetric core of the natural product, and a stepwise fragment coupling to complete the target molecule [40]. Macrosphelides A (75) and B (76), isolated from Macrosphaeropsis sp. FO-5050 and Periconia byssoides, have antitumoral properties. For their synthesis, several RCM conditions were used, but 1 equiv. of [Ru]-II catalyst and long reaction time were necessary to obtain the cyclized product in reasonable yields (60–65%, Scheme 5.17). The authors pointed out that the addition of a Lewis acid under catalytic conditions was ineffective [41]. Zampanolide (79), a macrolide which exhibits significant activity against a variety of tumor cell lines, was recently isolated from the sponge Dactylospongia sp. The RCM cyclization step for the synthesis of this compound was done on substrate 77; however, as with other substrates bearing protected hydroxy groups, the reaction OMEM

O

O

O HO

[Ru]-II (100 mol%) CH2Cl2, rt, 2 – 5 d

O O

O 73

Scheme 5.17

O

O OMEM O

HO

O

R

O O

OH

RCM

O

74 (16-membered)

O

O O

CO2H

O

O

Macrosphelide A: R = C-OH (75) Macrosphelide B: R = CO (76)


162 O HO

O OH [Ru]-II (10 mol%) CH2Cl2 (1 mM), HO rt, 1 h

O

O

O OH

O OH

O

93%

O

N H

77

RCM

O

O

(−)-Zampanolide (79)

78 (20-membered)

Scheme 5.18

was totally ineffective (Scheme 5.18). In contrast, the conditions allowed the synthesis of the nonnatural enantiomer of the related macrolide dactylolide [42]. Palmerolide A (82), a recently disclosed marine natural product that was isolated from Synoicum adareanum, possesses potent and selective activity against the melanoma cancer cell line UACC-62. Nicolaou et al. reported the total syntheses of five stereoisomers of palmerolide A including the originally proposed (82) and the revised [ent-(19-epi-20-epi-82)] structures. The RCM cyclization proceeded smoothly under mild conditions (Scheme 5.19) [43]. An effective RCM of 87, containing no less than 10 double bonds, was successfully used for the synthesis of iejimalides’ group – a new group of cytotoxic compounds – recently isolated. RCM reaction gave the desired macrocycles in high yields and E-stereoselectivity (Scheme 5.20) [44]. 5.3.4 Macrocyclic Musk

Three classes of compounds, namely, nitroarenes, polycyclic benzenoids, and macrocyclic ketones and lactones, possess musky odors. The latter have been used for the longest time in fragrance formulations. Efficient synthetic methods that can be scaled up are necessary to decrease their cost of production. Muscone (93) [45] and civetone [46], are two of the most important classical sources of musk odors. Muscone (93), the principal musk odorant perfume ingredient from the male musk I

O

O I

O

OH HO

[Ru]-II (5 mol%) CH2Cl2 (5 mM), 23 °C , 1 h

O

HN

O

O

81% (E/Z > 95 : 5)

OH

HO O 80

Scheme 5.19

HO

O

NH2

O

O

O 81 (20-membered)

NH2

O O

Palmerolide A (82)

OH RCM NH2

5.3 Macrocyclic Polyketides OMe

OMe

OMe [Ru]-II (20 mol%) CH2Cl2 (5 mM), rt, 3 d OMe Equation 1

OMe

O

NH

84 (24-membered)

O O

TBSO

TBSO

HN H

HN O O

O O

H

OMe

O

88 (24-membered) NHTeoc

O O

O

87

NHTeoc

Iejimalide B; R = H (89) HN O Iejimalide D; R = SO3Na (90) O RO HN H

Scheme 5.20

deer (Moschus moschiferus) is nowadays produced synthetically by RCM using the conditions shown in Scheme 5.21. Other relatively simple macrolides synthesized by Fürstner et al. are recifeiolide (96) – a musk fragrance enhancer isolated from Cephalosporium recifei (Equation 1, Scheme 5.22) [47]; exaltolide (99) [48] – a musk odorant perfume isolated from Archangelica officinalis, which was the first macrolactone to be prepared by RCM (Equation 2, Scheme 5.22); and muscopyridine (102) (Equation 3, Scheme 5.22) [49], which is present in the secretion of the ventral glands of male musk deer. 5.3.5 Epothilones

The scope of RCM has been fully illustrated by the synthesis of epothilones (epothilones A–F) and has been reviewed recently [5b, c]. Epothilones are OH

OH

O

[Ru]-I (7 mol%) (ClCH2)2, 50 °C 74%

RCM Hydrogenation 91

Scheme 5.21

O O

HN

OMe RCM (diene–ene system)

OMe


RO

H

[Ru]-II (20 mol%) CH2Cl2 (5 mM), rt, 48 h OMe

Iejimalide A; R = H (85) HN Iejimalide C; R = SO3Na (86)

HN

OMe

OMe

O O

O

83

Equation 2

RCM (diene–ene system)

OMe

O


O

163

92 (15-membered)

(R)-(−)-Muscone (93)

164

5 Synthesis of Natural Products Containing Macrocycles by Alkene Ring-closing Metathesis [Ru]-IV (2 – 5 mol%) CH2Cl2, 40 °C

O Equation 1

O

O

O

O

80%, E /Z = 4.7 : 1

94

O

RCM Recifeiolide (96)

95 (12-membered)

O O Equation 2

O

O

[Ru]-I (5 mol%) CH2Cl2, rt, 30 h

O

O

62% (E/Z = 23 : 77)

97

+

98 (16-membered) [Ru]-V (10 mol%) CH2Cl2 (6 mM), reflux, 14 h

Cl−

N H

Equation 3

− + Cl N H

57% (E/Z = 1.8 : 1)

100

RCM Hydrogenation Exaltolide (99)

101 (13-membered)

N

RCM Hydrogenation (R)-(+)-Muscopyridine (102)

Scheme 5.22

16-membered ring macrolides, first isolated from Sorangium cellulosum. Since the discovery of their powerful cytotoxic properties and their taxol-like mode of action, much effort has been undertaken to develop efficient synthesis methodologies that would allow a straight access to these molecules as well as to a variety of structural analogs [50]. In October 2007, the first epothilone derivative, ixabepilone, received FDA approval for the treatment of metastatic breast cancer [51]. The structure of these macrocycles is suitable to RCM-based approaches, although macrolactonization has also been thoroughly used. Two strategies are possible, namely, the C9–C10 and the C12–C13 double bond disconnection. The stereocontrol of the C12–C13 double bond by using RCM is difficult; indeed, none of the catalyst used initially, [Ru]-I and [Mo]-I, produced a good Z selectivity as can be seen from the transformation of 103–104, which led to a 1 : 1 mixture of isomers. This strategy was used for the synthesis of epothilone C (105) (Scheme 5.23) [52]. More recently, a solution was provided by using a silicon-tethered RCM. Hence, a disiloxane was used for the construction of the (Z)-C12–C13 double bond via

S OTBS

N

[Ru]-I (6 mol%) CH2Cl2 (6 mM), 25 °C 94% (E /Z = 1 : 1)

O O OTBS O 103

Scheme 5.23

S

S OTBS

N O

C12-C13

RCM

OH

N O

O OTBS O 104 (16-membered)

O

OH O

Epothilone C (105)


S

R OTBS

N

[Ru]-II (30 mol%) CH2Cl2 (3 mM), 40 °C, 8 – 48 h 89% (E /Z = 1 : 1)

O

S

O

C9-C10 R

R S

R OH

N

RCM

R OH O

O

O

OH O

OH O

Epothilone A: R = H (108)

107: R = H, Me (16-membered)

106: R = H, Me

O

N

O

O OTBS O

165

Epothilone B: R = Me (109)

Scheme 5.24

RCM, which led to a nine-membered ring intermediate which was then opened and used for the synthesis of epothilone B and epothilone D [53]. The alternative approach involving the formation of the C9–C10 bond followed by a hydrogenation step was first proposed by Danishefsky et al. [52a]. However, it was not until the discovery of more active second-generation ruthenium catalysts, such as [Ru]-II, that satisfactory results were obtained. In 2002, the stereoselective total syntheses of epothilone A (108) and epothilone B (109) by RCM of the epoxy compound 106 in the presence of catalyst [Ru]-II was reported [54]. In addition, the epothilone skeleton was also obtained via a diene–ene cyclization using [Ru]-II leading to the macrolide in 75% yield (Scheme 5.24). In a similar approach, Danishefsky et al. used 110 as the RCM substrate to obtain epothilone D [55]. Consequently, RCM of 110, in refluxing toluene in the presence of [Ru]-II, gave exclusively the E-isomer 111 in high yield. This intermediate was ultimately transformed into epothilone D (112), (Scheme 5.25). An alternative to the previous strategies consisted of an RCM to form the C10–C11 bond of epothilone 490 (115), a naturally occurring metabolite. The conditions for RCM were critical in this case, and best results were obtained using [Ru]-II as the catalyst in refluxing toluene for a few minutes [56]. These conditions avoided the formation of by-products while increasing the yield of the desired product (Scheme 5.26). 5.3.6 Amphidinolides

Amphidinolides are marine natural products, isolated from symbiotic marine dinoflagellates Amphidinium sp. In general, their most outstanding biological activity is their cytotoxic effects. Various subfamilies of these compounds bearing

S OTBS

N

[Ru]-II (30 mol%) PhCH3 (3 mM), 110 °C, 10–20 min 78%, (E )-isomer only

O O OTES O 110

Scheme 5.25

C9-C10 MCR

S

S OTBS

N O

OH

N O

O OTBS O 111 (16-membered)

O

OH O

Epothilone D (112)

166


C10-C11 [Ru]-II (10 mol%) CH2Cl2 (2 mM), 35 °C, 5.5 h

S N

S

S OTBS

N

64%

O

O O

O OTBS O

O OTES O

OH

N

O

113

MCR

114 (16-membered)

OH O

Epothilone D (115)

Scheme 5.26

different macrocyclic ring sizes have been described to date. Amphidinolide A (118) was synthesized in 57. The highly unsaturated 20-membered ring lactone was formed at a late stage by RCM. To provide regio- and E-stereoselectivity, 0.5 equiv. of catalyst [Ru]-II was necessary; however, the yield of the cyclized product remained low (35%, Scheme 5.27). The amphidinolide T subfamily contains a saturated 19-membered ring lactone. The synthesis of a few members of this subfamily has been achieved by Fürstner et al. [58]. For the synthesis of amphidinolide T3 (121), for instance, good conversion could be attained only in refluxing toluene. The desired ring closure was faster than the decomposition of the catalyst, delivering cycloalkene 120 in 82% yield as an E/Z mixture of isomers in a 2 : 1 ratio, which was further hydrogenated (Equation 1, Scheme 5.28). Amphidinolides T1, T4, and T5 were prepared in high yield by applying RCM to diene 122 utilizing [Ru]-VI in CH2 Cl2 . The corresponding cycloalkene 123 (E/Z = 6 : 1) was then hydrogenated (Equation 2, Scheme 5.28). More recently, amphidinolide Y was discovered. In the total synthesis of this compound, the formation of the (E)-trisubstituted alkene via RCM appeared to be challenging. A critical effect of the remote substituents in RCM was demonstrated. Hence, the cyclization of the C6-ketone 125 in refluxing CH2 Cl2 , using [Ru]-II as the catalyst, which was added in portions up to 50 mol%, was successful (Scheme 5.29) [59]. The synthesis of amphidinolide E (130) is an example of RCM applied to a polyene. It was necessary to use [Ru]-I to avoid the metathesis of an internal bond and decomposition of the starting material, as was observed when using [Ru]-II (Scheme 5.30) [60]. In a recent contribution, amphidinolide H (133) was constructed from polyene 131 by using RCM cyclization as a key step with the use of [Ru]-II catalyst in

OH

[Ru]-II (50 mol%) CH2Cl2, reflux, 24 h

PMBO PMBO OO

TBSO O 116

Scheme 5.27

35%, (E ) -isomer only

RCM OH

OH HO

PMBO PMBO

HO O

TBSO

O O

O

O 117 (20-membered)

HO O

Amphidinolide A (118)

5.3 Macrocyclic Polyketides OMOM

OMOM

O

TBDPSO

TBDPSO

HO

[Ru]-V (20 mol%) PhCH3,110 °C, 10 min

O O

O O

82% (E /Z = 2 : 1)

Equation 1

O

O

O

O

O

O

119

120 (19-membered)

OMOM

RCM O Hydrogenation Amphidinolide T3 (121)

OMOM

O

TBDPSO

TBDPSO

HO

[Ru]-V (10 mol%) CH2Cl2, reflux, 2 h

O O

O O

86% (E /Z = 6 : 1)

Equation 2

167

O O

O O

O RCM Hydrogenation

O

122

123 (19-membered)

O

Amphidinolide T4 (124)

Scheme 5.28 OTBS

RCM

TESO

O O

[Ru]-II (50 mol%) CH2Cl2, reflux, 3 d

TBSO

O HO HO

TESO

O

O

O

O

Ketone at C6, other substituents not allowed

O O

O

O

O

125

126 (17-membered)

Amphidinolide Y (127)

Scheme 5.29

benzene at room temperature. The desired macrocycle 132 was obtained as a single E-isomer in 73% yield (Scheme 5.31) [61]. 5.3.7 Other Polyketides

Other macrocyclic compounds presumably of polyketide origin such as okilactomycin (136) and terpestacin (139), both possessing antitumoral activity, were

OTES

O O

73%, (E,E )-diene only

O

O

O

128

Scheme 5.30

[Ru]-I (20 mol%) CH2Cl2, reflux, 12 h

OTES

O O O

HO

O O

129 (19-membered)

OH

O

HO

O O RCM (diene–ene system ) Amphidinolide E (130)


168

OTBS O

OTBS O [Ru]-II (10 mol%) OTBDPS C6H6, rt

TBSO

O

TBSO

HO OH

O O

O

OH

OTBDPS

68 –72%, (E )-isomer only

O O

OH OTES

OTES

131

O O

O 132 (26-membered)

RCM

O

Amphidinolide H (133)

Scheme 5.31

synthesized by RCM utilizing [Ru]-II catalyst in benzene (Equations 1 and 2, Scheme 5.32) [62, 63]. In addition, the total synthesis of the IgE antagonist, aspercyclide (142), was reported [64]. The key step was a kinetically controlled RCM reaction to form the 11-membered ring cycle 141 (Equation 3, Scheme 5.32). 5.3.8 Natural Cyclophanes

Naturally occurring paracyclophanes are scarce, and cylindrocyclophanes were the first to be isolated. These 22-membered [7,7]-paracyclophanes were found to be the major cytotoxic components in three different strains of the terrestrial blue-green algae Cylindrospermum lichenforme. Biosynthetically, the [7,7]-paracyclophane skeleton of the cylindrocyclophanes is presumed to arise from the dimerization of two identical resorcinol fragments of polyketide-originated metabolite. Smith et al. O O Equation 1

O

O [Ru]-II PhCH3 (0.07 mM), 80°C, 24 h

O

O

O

O

RCM Hydrogenation

O

OBn 134

OBn 135 (13-membered)

[Ru]-II (10 mol%) C6H6 (1 mM), rt, 16 h

O PMBO Equation 2 OH

O

O OH (−)-Okilactomycin (136)

O

O

PMBO

HO

35 – 44%, (E )-isomer only HO

137

138 (15-membered)

[Ru]-V (20 mol%) PhCH3 (2 mM), reflux, 4 h

MeO Equation 3

O O OPMP O 140

Scheme 5.32

O O

69% (E / Z = 5 : 1)

MeO

RCM

HO O O

OPMP

O 141 (11-membered)

OH

RCM

Terpestacin (139)

O O

OH

O (+)-Aspercylide C (142)

5.4 Terpenoids

MeO MeO

OMe

OMe

[Ru]-I (6 mol%) CH2Cl2 (4 mM), 20 °C, 22 h 88%

MeO MeO

OMe

OMe

HO HO

169

OH

OH RCM Hydrogenation

143

144 (22-membered)

(−)-Cylindrocyclophane F (145)

Scheme 5.33

reported a huge synthetic work where they used several synthetic strategies to construct the macrocycle, including a double CM (cross-metathesis) dimerization, and RCM of substrate 143 using [Ru]-I, which was highly efficient and gave 144 as a single isomer (Scheme 5.33). [Mo]-I catalyst was also used and was shown to give almost the same results [65]. 5.4 Terpenoids

Terpenoids, especially those constituted by four mevalonate units, are among the most abundant macrocycles in Nature. There are few examples of synthetic approaches toward these compounds using RCM for the cyclization step. 5.4.1 Diterpenoids

One of the most compact and condensed carbon frameworks consisting of multiple interconnected ring systems, coleophomones B–C (148, 151) – isolated from a Stachybotrys cylindrospora fungal broth, has an antifungal activity and inhibits the serine protease enzyme, heart chimase. The 11-membered ring of these two compounds was constructed using an impressive olefin metathesis reaction to build the C16–C17 bond (Equations 1 and 2, Scheme 5.34). RCM reactions proceeded with excellent yields and selectivities for both members of the family [66]. The synthesis of clavirolide C (154), a member of the dolabellane family of diterpenes isolated from coral Clavularia viridis, was recently disclosed. Consequently, treatment of allylic ether 152 (dr = 4 : 1) with [Ru]-III (10 mol%) under high dilution conditions and with a slow addition of the substrate to a solution of catalyst gave the desired compound 153 in 70% yield as a single isomer (E/Z > 95 : 5), (Scheme 5.35). Interestingly, other catalysts such as [Ru]-II failed to give the cyclized product [67]. Several strategies to construct the macrocyclic diterpene tonantzitlolone (157) via RCM have been accomplished [68]. While the C1–C2 double bond could not be formed probably due to steric congestion created by the three allylic methyl groups present in the starting material, macrocycle 156 was eventually obtained from 155 with high E-selectivity due to less congested olefins (Scheme 5.36).


170

16 RCM OMe O

O Equation 1

O 17 O

OO

[Ru]-I (10 mol%) CH2Cl2, 40 °C, 3 h 86%, (E )-isomer only

O

OMe O

O

O

PBPO

OPBP 146

H O

Coleophomone B (148) D16,17 (E )-macrocycle

147 (11-membered)

16 O

O

O

RCM 17

[Ru]-II (10 mol%) CH2Cl2, 40 °C, 3 h

OO

O

O

80%, (Z )-isomer only MeO

Equation 2

O

OPBP

OMe

PBPO

149

150 (11-membered)

O

O

H O

Coleophomone B (151) D16,17 (Z )-macrocycle

Scheme 5.34 O O

H

OSiEt3

OH

[Ru]-III (10 mol%) (ClCH2)2 (1 mM), 83 °C, 6 h

HO O H

H

O

OSiEt3

70% (E /Z >95 : 5)

O RCM

152

153 (11-membered)

Clavirolide C (154)

Scheme 5.35

Floresolide B (160), a recently discovered natural product isolated from an ascidian of genus Aplidium that exhibits cytotoxicity against KB tumor cells, contains an aromatic ring connected to a [10]-metacyclophane and a seven-membered ring lactone. The lactone bridge present in substrate 158 was installed prior to the

RCM H HO HO

O

1

2

[Ru]-II CH2Cl2, reflux, 2 h 70% (E/ Z = 1 : 6)

O H 155

Scheme 5.36

HO HO

O

O

O O H

156 (15-membered)

O HO HO

O O H

Tonantzitlolone (157)

5.4 Terpenoids

O

O

O

[Ru]-II (10 mol%) (ClCH2)2 (0.5 mM), 40 °C, 15 min

O O 2N

89% (E /Z = 2.9 : 1)

171

O O

158

O

O O2N

RCM

159 (12-membered)

OH

O O

Floresolide B (160)

Scheme 5.37

metathesis, as substrates without any constraint gave either dimeric products or low yields (Scheme 5.37) [69]. 5.4.2 Macrocyclic Lipids

Large macrocyclic lipids are rarely found in bacterial membranes. 36- and 72-membered ring macrocyclic lipids 163 and 166 were found in Archaeabacteria, and built up using an olefin metathesis approach. Consequently, in the presence of [Ru]-I, RCM of 161 and 164 proceeded efficiently in 79% yield under high dilution conditions to give the 36-membered ring compound 162 (Equation 1, Scheme 5.38). The 72-membered ring compound 165 was in turn obtained in 45% yield (Equation 2, Scheme 5.38) [70].

RCM/Hydrogenation

[Ru]-I (29 mol%) CH2Cl2 (1 mM), reflux, 42 h

Equation 1 161 BnO O H

79% (E/Z = 7 : 1)

O O

H

BnO

O

Archaeal membrane lipid (163) O HO O H

162 (36-membered)

O OH O

HO

O OBn

OBn

H

O OH

RCM/Hydrogenation [Ru]-II (20 mol%) CH2Cl2 (3.8 mM), reflux, 72 h

Equation 2

45% (E /Z = 10 : 1) 164

165 (72-membered)

O

Scheme 5.38

H O OBn

Archaeal membrane lipid (166)

OH

O

OBn

OH

O OH


172

O H NH

O

O

N

O

O H

O

O S NH

[Ru]-I (30 mol%) PhCH3 (1 mM), 100 °C, 18 h

O S

NH O

N

O H

O NH

O

NH

O

N

O S O

NH

37– 42%, (E )-isomer only O

O

O

O

OH

Mes 168 (20-membered)

Mes 167

OH

RCM (diene –ene system )

(−)-Griseoviridin (169)

Scheme 5.39

5.5 Macrocycles of Amino Acid Origin 5.5.1 Macrolactams

Pharmacologically active macrolactams are highly functionalized. The reaction conditions used for the RCM key macrocyclization step are summarized in Schemes 5.39–5.41. The macrocyclic antibiotic griseoviridin (169), a compound with antibacterial properties isolated from Streptomyces griseus, has a 23-membered ring cycle. The synthesis of this natural product was described by Meyers et al. using a diene–ene RCM. Consequently, the cyclization of precursor 167 using catalyst [Ru]-I gave compound 168 with moderate yield but with total E-selectivity. A simple deprotection step was necessary to afford griseoviridin (169) (Scheme 5.39) [71]. Fluvirucin B1 (172) is one of the first natural macrolides being synthesized by RCM, and one of the few examples utilizing [Mo]-I efficiently. This antifungal agent, isolated from Actinomadura SCC 1777, was obtained through a high yielding and totally stereoselective RCM of compound 170 (Scheme 5.40) [72]. A new class of ansamycin compounds bearing a (E,E,E)-triene within a 21-membered ring lactam were recently described. In particular, cytotrienins A–D (175) that were isolated from Streptomyces sp. RK95-74 and are important antitumoral compounds. Their synthesis was disclosed by Panek et al. and the crucial RCM step was performed on polyenic substrate 173. Surprisingly, when using [Ru]-II, the insertion of the ruthenium took place on one of the disubstituted double bonds thus affording the 19-membered ring. On the other

O O

OAc

OAc O NHCOCF3 [Mo]-I (20 mol%) C6H6, 60 °C, 10 h

O

OAc NHCOCF3

NH2 O

O

O HN

171 (14-membered)

Scheme 5.40

OH

O OH

HN 170

RCM/Hydrogenation

O


HN

OAc

Sch 38516 (fluvirucin B1) (172)

5.5 Macrocycles of Amino Acid Origin

173

X OMe

OMe NH

NH OMe O

TBSO

X

[Ru]-I (20 mol%) CH2Cl2, reflux, 2 h 73%

NH OMe O

TBSO

OMe

TBSO

HO H N

R

OMe

O O

OMe

TBSO

O R=

173

O

174 (21-membered) R=

RCM (diene –ene system) X = OH: (+) -cytotrienin A1 X = O: cytotrienin C2 (quinone) X = OH: (+)-cytotrienin B3 X = O: cytotrienin D4 quinone) 175

Scheme 5.41

hand, when using [Ru]-I in refluxing CH2 Cl2 , the insertion took place on the terminal double bonds affording (E,E,E)-triene 174 with excellent selectivity and yield (Scheme 5.41) [73]. 5.5.2 Cyclodepsipeptides

The 13-membered ring lactone and lactam antifungal–antibiotic (−)-PF1163B (178), isolated from Streptomyces, was synthesized using 12 mol% of Ru-[II] (Scheme 5.42) in 60% yield [74]. The natural product was eventually obtained after hydrogenation. Five total syntheses of the antiproliferative marine metabolite (−)-spongidepsine (181) have been reported to date, and most of them use RCM. As a general trend, when RCM is performed on substrates of type 179, high yields are obtained but with variable E/Z ratios. This outcome is fortunately not detrimental as the resulting double bond is then hydrogenated (Scheme 5.43) [75]. An RCM approach to the total synthesis of antimitotic marine product arenastatin (184) using [Ru]-I catalyst is shown in Scheme 5.44 [76]. O

O

N O

O

HO

BnO

BnO O

O

[Ru]-II (12 mol%) (ClCH2)2 (2 mM), 82 °C, 6 h 60%, (E/Z )-mixture

O

N O

O

O

N O

O

RCM Hydrogenation 176

Scheme 5.42

177 (13-membered)

(−)-PF 1163B (178)


174

O

O NMe

Ph

O

O

O

[Ru]-II (10 mol%) PhCH3, 110 °C, 20 min

NMe

Ph

NMe

Ph O

80%, (E/Z )-mixture

O

RCM Hydrogenation

O

O PMBO

PMBO 179

180 (13-membered)

(−)-Spongidepsin (181)

Scheme 5.43 O Ph

O

[Ru]-I (10 mol%) CH2Cl2, Ph reflux, 6 h

O OO

HN N H

O

O

182

OMe

O O

OO O

O

HN N H

RCM

O

O

O

183 (16-membered)

OMe

O OO O

HN N H

O

OMe

Arenastatin A (184)

Scheme 5.44

5.5.3 Alkaloids

Roseophilin (186) is a 13-membered macrocycle, which was isolated from the culture broth of Streptomyces griseoviridis, and possesses high antitumoral activity. This product belongs to the pyrrole alkaloids of the prodigiosin family. The problem in the synthesis of this compound is the incorporation of the macrocycle into an ansa-bridged azafulvene. The synthetic efforts were first attempted on precursors bearing fulvalene cycles [77]. However, RCM was successful only when bulky groups were suitably situated to exert conformational constraints that favored the reactive conformation. Unfortunately, these early efforts led to the undesired enantiomer or to racemic products (Scheme 5.45a,b). Better results were obtained when forming the ansa-macrocycle prior to the formation of one or both five-membered ring cycles as the strain was partially avoided. Thus, Fürstner et al.’ acyclic precursor [78], led to the corresponding macrocycle in high yield by using [Ru]-I (Scheme 5.45c) while Boger et al. obtained ent-186 efficiently by treatment of a monocyclic triene with [Ru]-I (Scheme 5.45d) [79]. The first total synthesis of the enantiomerically pure 186 was achieved by Tius et al. from a monocyclic diene (185) (Scheme 5.45e) [80]. The pyrrole ring was formed subsequently using a Paal–Knorr reaction. In Scheme 5.46, the syntheses of different macrocyclic natural products classified as alkaloids are summarized. The rich variety of biological activities associated with both plant-derived and animal-derived alkaloids has stimulated many organic synthetic groups. The main problem from a metathesis point of view is the ability of the nitrogen to coordinate to metal-alkylidene complexes and to interfere unproductively with the catalytic activity. In the examples summarized herein, the use of [Ru]-I, [Ru]-II, and [Ru]-V has proven to be effective. Consequently, (−)-sarain A (189) (Equation 1, Scheme 5.46), isolated from sponge Reniera sarai [81]; ircinal

5.6 Macrocyclic Glycolipids

175

H TIPSO

N Ts

O (b)

[Ru]-I, CH2Cl2 40 °C, 16 h Yield (rac): 60%

O [78a]

[78b]

N H Ts

SO2Ph

(a) [Ru]-I, (10 mol%) CH2Cl2, 40 °C, 16 h Yield (ent): 91% N

[79]

PhO2S

CO2Me

OH OTBS

(c) [Ru]-I (10 mol%) CH2Cl2, reflux, 2 h

MeO

Macrocyclic intermediate

O

Yield (rac): 85%

Cl

(d) [Ru]-I, (20 mol%) CH2Cl2, 40 °C, 72 h Yield (ent ): 88% (E /Z 1 : 1)

[81] [80]

(e) [Ru]-I, (30 mol%) CH2Cl2, 40 °C, 30 h Yield : 91% (E/Z mixtures)

NH (−)-Roseophilin (186)

O O MeO2C

N SEM

BzO 185

Scheme 5.45

(192) and manzamine A (193) (Equation 2, Scheme 5.46), isolated from the marine sponges Haliclona and Pellina [82], respectively; nakadomanine (196) (Equation3, Scheme 5.46), isolated from the sponge Amphimedon [83]; nonylprodigiosyn (199) (Equation 4, Scheme 5.46), isolated from Actinomadura madurae [84]; and isooncinotine (202) (Equation 5, Scheme 5.46), isolated from Stem bark Oncinotis nitida and Oncinoris tenuiloba [85], have all been efficiently synthesized using RCM with [Ru]-I or [Ru]-V. In general, chelation of the catalyst with the nitrogen functionality is avoided by protonation or by RCM of bonds with a remote relative situation from the potentially interfering groups.

5.6 Macrocyclic Glycolipids

The performance and excellent application profile of RCM are illustrated by the total synthesis of various resin glycosides and sugar-based macrodiolides such as tricolorins A and G, woodrosin I, and ipomoeassins B and E (Scheme 5.47). The synthesis of these amphiphilic natural products has resulted in the advancement of carbohydrate chemistry and is due to Fürsner et al. who have recently reviewed their effort in this area [86]. In this review, they fully detail the syntheses of tricolorins A (208) (Equation 1, Scheme 5.47) and G (211) (Equation 2, Scheme 5.47), two natural products isolated from Ipomea tricolor, which are plant growth inhibitors

176


OTIPS N

Equation 1

O

N

RCM / Hydrogenation H O−

[Ru]-I (5 mol%) CH2Cl2 (0.25 mM), reflux

OTIPS N

+

O

N

O

O

OTBS

H

CH(OMe)2 [Ru]-I (13 mol%) CH2Cl2 (5 mM), reflux, 3 h

H

N Equation 2

O

N

H N

76% (E/Z = 8 : 1)

(−)-Sarain A (189)

CH(OMe)2

H

H N

OH HO

OTBS 188 (13-membered)

187

N

N

75–85%, (E /Z )-mixture

RCM OH

H

N

N

O

O

O O 191 (13-membered)

190

O Equation 3

[Ru]-I (15 mol %) CH2Cl2 (0.5 mM), 50 °C, 24 h

N O N

66% (E /Z = 5 : 3)

O

194

MeO

N

N H

65% (E/Z = 10 : 1)

NH

NH Cl− NH + H N

MeO

NH MeO

NH Nonylprodigiosin (199)

H

N + H2N 200

RCM Hydrogenation

N

H

H

N

(+)-Nakadomarin (196)

198 (21-membered)

N

H

H

[Ru]-V CH2Cl2, reflux, 16 h

197

Equation 5

O

O

195 (15-membered)

NH Cl− +NH

Equation 4

H

N H

Ircinal, R = CHO (192) Manzamine A, R = 1-b- carbolinyl (193) RCM

N

RCM

[Ru]-V CH2Cl2, reflux, 16 h HN + H2N

201 (22-membered) O

HN HN O

(−)-Isooncinotine (202) HN

O

Scheme 5.46

and protect sugarcane against invasive weeds [87] as well as the synthesis of woodrosin I (214) (Equation 3, Scheme 5.47) isolated from Ipomea tuberulosa L [88]. RCM was carried out with the first-generation catalysts and proceeded in high yields and stereoselectivities. More recently, the same group has disclosed the total synthesis of cytotoxic agents ipomoeassin B and E, isolated from Ipomoea squamosa. Consequently, exposure of diene 215 to [Ru]-II (10 mol%) in refluxing CH2 Cl2 afforded macrocycle 216 in excellent yield as a mixture of both geometrical isomers, which could be selectively hydrogenated using RhCl(PPh3 )3 without affecting the lateral unsaturated esters. The resulting products were transformed into ipomoeassin E (217) (Equation 4,

5.7 Conclusions and Outlook

177

RCM O

Equation 1

[Ru]-IV (5 mol%) CH2Cl2, reflux, 24 h

O O O O O O Ph O OH O O 203

O

OO

O O

O O

Ph O O BnO Equation 2 O BnO BnO O

[Ru]-IV (10 mol%) CH2Cl2, reflux, 22 h

O O H

O

OH 204 (19-membered)

O

O O

90%

O O Ph O O O

OH O O HO OH HO O O O O O O Tricolorin A O (205) O O O O O HO HO OH OH O O HO HO O O H HO HO O O HO RCM HO O Hydrogenation

O

O O

O Ph O BnO

O O H

O O

93% BnO

BnO

O

O RCM Tricolorin G (208) Hydrogenation

O 207 (21-membered)

O 206

OH OHO HO

[Ru]-IV OBn OBn (10 mol%) OBn OBn O O CH2Cl2, BnO OBn BnO OBn 40 °C, 16 h O O O BnO BnO O O O O BnO 94% BnO O O O O O O O Ph Ph (E/Z ) = 9 : 1) O O O O O OO O O O O O O O O O OH OH Equation 3 O O O O Cl Cl 210 (29-membered) 209

O O O O O O O

OH HO O OH O HO HOO O O O O OH O O O HO O O

Woodrosin I (211)

AcO AcO AcO PeMe2Si O O Ph O

O

OO

O O O OTBS

Equation 4 212

[Ru]-II (10 mol%) CH2Cl2, PeMe2Si reflux Ph

O O O

O O O O

O OO

O O

O O O O

OTBS

213 (26-membered)

O

Ph

O O

O O O O

RCM/Hydrogenation

O O O OH

O OH HO

Ipomoeassin E (214)

Scheme 5.47

Scheme 5.47). Extension of this methodology to the preparation of ipomoeassin B was straightforward [89].


Since the discovery of ruthenium- and molybdenum catalysts, the metathesis reaction has triggered the attention of synthetic chemists. Even in the early times, RCM process has been used in the construction of macrocycles with great success. Actually, a wide range of examples has been reported using olefin metathesis in the context of natural macrocycles. The works published before 2000 used the only

178


available first-generation catalysts and had some problems with efficiency. These complexes do not stand harsh reaction conditions and are generally unreactive with di- or trisubstituted double bonds. However, after the second-generation complexes were introduced, several examples have shown that some synthetic problems related to regioselectivity could be solved using the first-generation catalysts, especially with polyenic substrates. There is not a clear rule that indicates the best catalyst to use in each case. Subtle variations in the substrate structure and in the type of final product may lead to different outcomes with each type of carbene complexes. In particular, remote substitution and the presence of coordinating heteroatoms have an important influence on the results. In addition, the E/Z ratio of the product may be influenced by all these factors and by secondary isomerizations mediated by the metathesis catalysts. Unfortunately, there still exists a lack of prediction over the stereochemistry for unsaturated macrocycles obtained by RCM. The catalyst loading varies from one case to another (from 1 to 25 mol%) and is highly dependent on the species used. Other important aspects are the concentration (0.25–8 mM) and the addition time which sometimes have to be slow. The catalyst may be more effective if added in portions. Long reaction times are sometimes required when working at low temperatures. In conclusion, when planning a new synthesis, it is worth testing different catalysts, temperatures, and concentrations with additional fine tuning of the loading and addition rate. On the other hand, some aspects in the selection of reaction conditions for a macrocycle construction using RCM are quite straightforward. These are solvent selection [toluene, CH2 Cl2 , (ClCH2 )2 ] and the reaction temperature (from rt to refluxing toluene). In the near future, the discovery of more powerful and selective catalysts should solve the remaining problems.

References 1 Among the numerous references, we

will mention those that cover specifically or in a particular section the diene or ene-diene metathesis in the synthesis of macrocycles (a) Mori, M. (2007) Adv. Synth. Catal., 349, 121–135; (b) Schmidt, B. (2004) Eur. J. Org. Chem., 1865–1880; (c) Hoveyda, A.H., Gillingham, D.G., Van Veldhuizen, J.J., Kataoka, O., Garber, S.B., Kingsbury, J.S., and Harrity, J.P.A. (2004) Org. Biomol. Chem., 2, 8–23; (d) Grubbs, R.H. (2004) Tetrahedron, 60, 7117–7140; (e) Prunet, J. (2003) Angew. Chem. Int. Ed. Engl., 42, 2826–2830; (f) Trnka, T.M., Grubbs, R.H. (2001) Acc. Chem. Res., 34, 18–29; (g)

Fürstner, A. (2000) Angew. Chem. Int. Ed. Engl., 39, 3012–3043; (h) Maier, M.E. (2000) Angew. Chem. Int. Ed., 39, 2073–2077; (i) Yet, L. (2000) Chem. Rev., 100, 2963–3007; (j) Randall, M.L. and Snapper, M.L. (1998) Strem Chem., 17, 1–9; (k) Armstrong, S.K. (1998) J. Chem. Soc., Perkin Trans. 1, 371–388; (l) Grubbs, R.H. and Chang, S. (1998) Tetrahedron, 54, 4413–4450; (m) ¨ Mulzer, J. and Ohler, E. (2004) Topics Organomet. Chem., 13, 269–366; (n) Nicolaou,K.C., Bulger, P.G., and Sarlah, D. (2005) Angew. Chem. Int. Ed., 44, 4490–4527; (o) Gaich, T. and Mulzer, J. (2005) Curr. Topics Med.

References

2

3

4

5

6

7

8

9

10 11

Chem., 5, 1473–1494; (p) Van der Weghe, J. and Eustache, J. (2005) Curr. Topics Med. Chem., 5, 1495–1519; (q) Deiters, A. and Martin, S.F. (2004) Chem. Rev., 104, 2199–2238. (a) Newman, D.J., Cragg, G.M., and Snader, K.M. (2003) J. Nat. Prod., 66, 1022–1037; (b) Henkel, T., Brunne, R.M., Mueller, H., and Reichel, F. (1999) Angew. Chem. Int. Ed., 38, 643–647. (a) Wessjohann, L.A., Ruijter, E., Garc´ıa-Rivera, D., and Bradt, W. (2005) Mol. Divers., 9, 171–186; (b) Wessjohann, L.A. and Ruijter, E. (2004) Topics in Current Chemistry, vol. 243, Springer, Heidelberg, pp. 137–184. Gournelis, D.C., Laskaris, G.G., and Verpoorte, R. (1997) Nat. Prod. Rep., 14, 75–82. (a) Altmann K.-H. (2004) Org. Biomol. Chem., 2, 2137–2152; (b) Zhang, R. and Liu, Z. (2004) Curr. Org. Chem., 8, 267–290; (c) Rivkin, A., Cho, Y.S., Gabarda, A.E., Yoshimura, F., and Danishefsky, S.J. (2004) J. Nat. Prod., 67, 139–143; (d) Nicolaou, K.C., Wissinger, N., Pastor, J., Ninkovic, S., Sarabia, F., He, Y., Vourloumis, D., Li, T., Giannakakou, P., and Hamel, E. (1997) Nature, 390, 268–272; (e) Nicolaou, K.C., King, N.P., and He, Y. (1999) Top. Organomet. Chem., 1, 73–104. Nishioka, T., Iwabuchi, Y., Irie, H., and Hatakeyama, S. (1998) Tetrahedron Lett., 39, 5597–5600. Borer, B.C., Deeremberg, S., Bierfugel, H., and Pandit, U.K. (1994) Tetrahedron Lett., 35, 3191–3194. Cabrejas, L.M.M., Rohrbach, S., Wagner, D., Kallen, J., Zenke, G., and Wagner, J. (1999) Angew. Chem. Int. Ed., 38, 2443–2446. For a review on ruthenium based catalyst, see: Bieniek, M., Michrowska, A., Usanov, D.L., and Grela, K. (2008) Chem. Eur. J., 14, 806–818. Schrock, R.R. and Czekelius, C. (2007) Adv. Synth. Catal., 349, 55–77. (a) Katz, L. and Ashley, G.W. (2005) Chem. Rev., 105, 499–528; (b) Yeung, K.S. and Paterson, I. (2005) Chem. Rev., 105, 4237–4313; (c) McDaniel, R., Welch, M., and Hutchinson, C.R.

12

13

14 15

16

17

18

19

20

21

(2005) Chem. Rev., 105, 543–558; (d) Rychnovsky, S.D. (1995) Chem. Rev., 95, 2021–2040; (e) Norcross, R.D. and Paterson, I. (1995) Chem. Rev., 95, 2041–2114. For a recent review on the structure, biology and chemistry of RALs see: Winssinger, N. and Barluenga, S. (2007) Chem. Commun., 22–36. (a) Ayer, W.A. and Pena-Rodriguez, L. (1987) Phytochemistry, 26, 1353–1355; (b) Sugawara, F., Kim, K.W., Kobayashi, K., Uzawa, J., Yoshida, S., Murofushi, N., Takahashi, N., and Strobel, G.A. (1992) Phytochemistry, 31, 1987–1990; (c) Sharma, S.V., Agatsuma, T., and Nakano, H. (1998) Oncogene, 16, 2639–2645; (d) Yamamoto, K., Garbaccio, R.M., Stachel, S.J., Solit, D.B., Chiosis, G., Rosen, N., and Danshefsky, S.J. (2003) Angew. Chem. Int. Ed., 42, 1280–1284. Geng, X., Yang, Z.-Q., and Danishefsky, S.J. (2004) Synlett, 1325–1333. Yamamoto, K., Biswas, K., Gaul, C., and Danishefsky, S.J. (2003) Tetrahedron Lett., 44, 3297–3299. (a) Yang, Z.-Q., Geng, X., Solit, D., Pratilas, C.A., Rosen, N., and Danishefsky, S.J. (2004) J. Am. Chem. Soc., 126, 7881–7889; (b) Geng, X. and Danishefsky, S.J. (2004) Org. Lett., 6, 413–416. Barluenga, S., Dakas, P.Y., Fernandin, Y., Meijer, L., and Winssinger, N. (2006) Angew. Chem. Int. Ed., 45, 3951–3954. (a) Garbaccio, R.M., Stachel, S.J., Baeschlin, D.K., and Danishefsky, S.J. (2001) J. Am. Chem. Soc., 123, 10903–10908; (b) Couladouros, E.A., Mihou, A.P., and Bouzas, E.A. (2004) Org. Lett., 6, 977–980. (a) Barluenga, S., López, P., Moulin, E., and Winssinger, N. (2004) Angew. Chem. Int. Ed., 43, 3467–3470; (b) Barluenga, S., Moulin, E., Lopez, P., and Winssinger, N. (2005) Chem. Eur. J., 11, 4935–4952. Moulin, E., Barluenga, S., and Winssinger, N. (2005) Org. Lett., 7, 5637–5639. Moulin, E., Zoete, V., Barluenga, S., Karplus, M., and Wissinger, N. (2005) J. Am. Chem. Soc., 127, 6999–7004.

179

180

5 Synthesis of Natural Products Containing Macrocycles by Alkene Ring-closing Metathesis 22 (a) F¨ urstner, A., Seidel, G., and

23

24

25 26 27 28

29

30

31

32

33

34

Kindler, N. (1999) Tetrahedron, 55, 8215–8230; (b) Fürstner, A., Thiel, O.R., Kindler, N., and Bartkowska, B. (2000) J. Org. Chem., 65, 7990–7995. For a review on the chemistry and biology of salicylihalamide A and related compounds, see: Yet, L. (2003) Chem. Rev., 103, 4283–4306. (a) Kunze, B., Jansen, R., Sasse, F., Hofle, G., and Reichenbach, H. (1998) J. Antibiot., 51, 1075–1080; (b) Galinis, D.L., McKee, T.C., Pannell, L.K., Cardellina, J.H., and Boyd, M.R. (1997) J. Org. Chem., 62, 8968–8969; (c) Kim, J.W., Shin-ya, K., Furihata, K., Hayakawa, Y., and Seto, H. (1999) J. Org. Chem., 64, 153–155. Herb, C., Bayer, A., and Maier, M.E. (2004) Chem. Eur. J., 10, 5649–5660. Smith, A.B.III and Zheng, J. (2002) Tetrahedron, 58, 6455–6471. Snider, B.B. and Song, F. (2001) Org. Lett., 3, 1817–1820. Labrecque, D., Charron, S., Rej, R., Blais, C., and Lamothe, S. (2001) Tetrahedron Lett., 42, 2645–2648. (a) Wu, Y., Liao, X., Wang, R., Xie, X.S., and de Brabander, J.K. (2002) J. Am. Chem. Soc., 124, 3245–3253; (b) Wu, Y., Esser, L., and de Brabander, J.K. (2000) Angew. Chem. Int. Ed., 39, 4308–4310; (c) Wu, Y., Seguil, O.R., and de Brabander, J.K. (2000) Org.Lett., 2, 4241–4244. Fürstner, A., Dierkes, T., Thiel, O.R., and Blanda, G. (2001) Chem. Eur. J., 7, 5286–5298. (a) Wang, X. and Porco, J.A. (2003) J. Am. Chem. Soc., 125, 6040–6041; (b) Wang, X., Bowman, E.J., Bowman, B.J., and Porco, J.A. (2004) Angew. Chem. Int. Ed., 43, 3601–3605. Xuan, R., Oh, H.-S., Lee, Y., and Kang, H.-Y. (2008) J. Org. Chem., 73, 1459–1461. Kanada, R.M., Itoh, D., Nagai, M., Niijima, J., Asai, N., Mizui, Y., Abe, S., and Kotake, Y. (2007) Angew. Chem. Int. Ed., 46, 4350–4355. Chen, Q. and Du, Y. (2006) Tetrahedron Lett., 47, 8489–8492.

35 Krishna, P.R. and Srinivas, R. (2008)

Tetrahedron Asymmetry, 19, 1153–1160. 36 F¨ urstner, A. and Langemann, K. (1997)

J. Am. Chem. Soc., 119, 9130–9136. 37 (a) Gaul, C., Njardarson, J.T., Shan, D.,

38

39

40

41

42

43

44

45

Dorn, D.C., Wu, K.-D., Tong, W.P., Huang, X.-Y., Moore, M.A.S., and Danishefsky, S.J. (2004) J. Am. Chem. Soc., 126, 11326–11337; (b) Gaul, C., Njardarson, J.T., and Danishefsky, S.J. (2003) J. Am. Chem. Soc., 125, 6042–6043. (a) Dixon, D.J., Foster, A.C., and Ley, S.V. (2000) Org. Lett., 2, 123–125; See also: ref 6 and (b) Banwell, M.G., and McRae, K.J. (2000) Org. Lett., 2, 3583–3586. (a) Kang, E.J., Cho, E.J., Lee, Y.E., Ji, M.K., Shin, D.M., Chung, Y.K., and Lee, E. (2004) J. Am. Chem. Soc., 126, 2680–2681; (b) Kang, E.J., Cho, E.J., Ji, M.K., Lee, Y.E., Shin, D.M., Choi, S.Y., Chung, Y.K., Kim, J.S., Kim, H.J., Lee, S.G., Lah, M.S., and Lee, E. (2005) J. Org. Chem., 70, 6321–6329; (c) Soltani, O. and de Brabander, J.K. (2005) Org. Lett., 7, 2791–2793; (d) Bolshakov, S. and Leighton, J.L. (2005) Org. Lett., 7, 3809–3812; (e) Crimmins, M.T. and Vanier, G.S. (2006) Org. Lett., 8, 2887–2890. Cheung, L.L., Marumoto, S., Anderson, C.D., and Rychnovsky, S.D. (2008) Org. Lett., 10, 3101–3104. Kawaguchi, T., Funamori, N., Matsuya, Y., and Nemoto, H. (2004) J. Org. Chem., 69, 505–509. (a) Ding, F. and Jennings, M.P. (2008) J. Org. Chem., 73, 5965–5976; (b) Hoye, T.R. and Hu, M. (2003) J. Am. Chem. Soc., 125, 9576–9577. Nicolaou, K.C., Sun, Y.-P., Guduru, R., Banerji, B., and Chen, D.Y.-K. (2008) J. Am. Chem. Soc., 130, 3633–3644. Fürstner, A., Nevado, C., Waser, M., Tremblay, M., Chevrier, C., Teplý, F., A¨ıssa, C., Moulin, E., and Müller, O. (2007) J. Am. Chem. Soc., 129, 9150–9161. (a) Louie, J., Bielawski, C.W., and Grubbs, R.H. (2001) J. Am. Chem. Soc., 123, 11312–11313; For the RCM of related substrates and conditions see: (b) Fujimoto, S., Yoshikawa, K., Itoh,

References

46

47 48 49 50

51

52

53 54

M., and Kitahara, T. (2002) Biosci. Biotechnol. Biochemi., 66, 1389–1392; (c) Kamat, V.P., Hagiwara, H., Suzuki, T., and Ando, M. (1998) J. Chem. Soc., Perkin Trans 1, 2253–2254. Obtained via RCAM methodology: Fürstner, A. and Seidel, G. (2000) J. Organomet. Chem., 606, 75–78. Fürstner, A. and Lagemann, K. (1997) Synthesis, 792–803. Fürstner, A. and Langemann, K. (1997) J. Am. Chem. Soc., 119, 9130–9136. Fürstner, A. and Leitner, A. (2003) Angew. Chem. Int. Ed., 42, 308–311. Reviews on epothilones, see: (a) Feyen, F., Cachoux, F., Gertsch, J., Wartmann, M., and Altmann, K.H. (2008) Acc. Chem. Res., 41, 21–31; (b) Altmann, K.H., Pfeiffer, B., Arseniyadis, S., Pratt, B.A., and Nicolaou, K.C. (2007) ChemMedChem, 2, 396–483; (c) Nicolaou, K.C. (2005) J. Med. Chem., 48, 5613–5638; (d) Mulzer, J. (2000) Monatsh. Chem., 131, 205–238; (e) Nicolaou, K.C., Ritzén, A., and Namoto, K. (2001) Chem. Commun., 1523–1524. Mulzer, J., Altmann, K.-H., Höfle, G., Müller, R., and Prantz, K. (2008) C. R. Chim., 11, 1336–1368. (a) Meng, D., Bertinato, P., Balog, A., Su, D.S., Kamenecka, T., Sorensen, E.J., and Danishefsky, S.J. (1997) J. Am. Chem. Soc., 119, 10073–10092; (b) Nicolaou, K.C., He, Y., Vourloumis, D., Vallberg, H., Roschangar, F., Sarabia, F., Ninkovic, S., Yang, Z., and Trujillo, J.I. (1997) J. Am. Chem. Soc., 119, 7960–7973. Gaich, T. and Mulzer, J. (2005) Org. Lett., 7, 1311–1313. (a) Sun, J. and Sinha, S.C. (2002) Angew. Chem. Int. Ed., 41, 1381–1383; For previous syntheses, see: (b) Yang, Z., He, Y., Vourlumis, D., Vallberg, H., and Nicolaou, K.C. (1997) Angew. Chem. Int. Ed., 36, 166–169; For the synthesis of epothilone A derivatives see: (c) Meng, D., Su, D.S., Balog, A., Bertinato, P., Sorensen, E.J., Danishefsky, S.J., Zheng, Y.H., Chou, T.C., He, L., and Horwitz, S.B. (1997) J. Am. Chem. Soc., 119, 2733–2734.

55 Rivkin, A., Yoshimura, F.,

56

57

58

59 60 61

62

63

64 65

66

67 68

Gabarda, A.E., Chou, T.C., Dong, H., Tong, W.P., and Danishefsky, S.J. (2003) J. Am. Chem. Soc., 125, 2899–2901. Biswas, K., Lin, H., Njardarson, J.T., Chappell, M.D., Chou, T.C., Guan, Y., Tong, W.P., He, L., Horwitz, S.B., and Danishefsky, S.J. (2002) J. Am. Chem. Soc., 124, 9825–9832. (a) Malecka, E.R. Jr., Terrell, L.R., Geng, F., and Ward, J.S.III (2002) Org. Lett., 4, 2841–2844; (b) Trost, B., Chisholm, J., Wrobleski, S., and Jung, M. (2002) J. Am. Chem. Soc., 124, 12420–12421; (c) Lam, H.W. and Pattenden, G. (2002) Angew. Chem. Int. Ed., 41, 508–511; (d) Hoye, T.R. and Zhao, H. (1999) Org. Lett., 1, 1123–1126. (a) Fürstner, A., Aissa, C., Riveiros, R., and Ragot, J. (2002) Angew. Chem. Int. Ed., 41, 4763–4766; (b) A¨ıssa, C., Riveiros, R., Ragot, J., and Fürstner, A. (2003) J. Am. Chem. Soc., 125, 15512–15520. Jin, J., Chen, Y., Li, Y., Wu, J., and Dai, W.-M. (2007) Org. Lett., 9, 2585–2588. Va, P. and Roush, W.R. (2006) J. Am. Chem. Soc., 128, 15960–15961. Fürstner, A., Bouchez, L.C., Funel, J.-A., Liepins, V., Porée, F.-H., Gilmour, R., Beaufils, F., Laurich, D., and Tamiya, M. (2007) Angew. Chem. Int. Ed., 46, 9265–9270. Smith, A.B. III, Basu, K., and Bosanac, T. (2007) J. Am. Chem. Soc., 129, 14872–14874. Trost, B.M., Dong, G., and Vance, J.A. (2007) J. Am. Chem. Soc., 129, 4540–4541. Fürstner, A. and Müller, C. (2005) Chem. Commun., 5583–5585. Smith, A.B.III, Adams, C.M., Kozmin, S.A., and Paone, D.V. (2001) J. Am. Chem. Soc., 123, 5925–5937. Nicolaou, K.C., Montagnon, T., Vassilikogiannakis, G., and Mathison, C.J. (2005) J. Am. Chem. Soc., 127, 8872–8888. Brown, M.K. and Hoveyda, A.H. (2008) J. Am. Chem. Soc., 130, 12904–12906. Jasper, C., Wittenberg, R., Quitschalle, M., Jakupovic, J., and Kirschning, A. (2005) Org. Lett., 7, 479–482.

181

182

5 Synthesis of Natural Products Containing Macrocycles by Alkene Ring-closing Metathesis 69 Nicolaou, K.C. and Xu, H. (2006) Chem. 70 71

72

73 74

75

76 77

78 79 80 81

Comm., 600–602. Arakawa, K., Eguchi, T., and Kakinuma, K. (1998) J. Org. Chem., 63, 4741–4745. Dvorak, C.A., Schmitz, W.D., Poon, D.J., Pryde, D.C., Lawson, J.P., Amos, R.A., and Meyers, A.I. (2000) Angew. Chem. Int. Ed., 39, 1664–1666. Xu, Z., Johannes, C.W., Salman, S.S., and Hoveyda, A.H. (1996) J. Am. Chem. Soc., 118, 10926–10927. Evano, G., Schaus, J.V., and Panek, J.S. (2004) Org. Lett., 6, 525–528. Bouazza, F., Renoux, B., Bachmann, C., and Gesson, J.-P. (2003) Org. Lett., 5, 4049–4052. (a) Chen, H.-J. and Forsyth, C.J. (2004) Angew. Chem. Int. Ed., 43, 2148–2152; See also: (b) Ferrié, L., Reymond, S., Capdevielle, P., and Cossy, J. (2006) Org. Lett., 8, 3441–3443; (c) Chandrasekhar, S., Yaragorla, S.R., Sreelakshmi, L., and Reddy, C.R. (2008) Tetrahedron, 64, 5174–5183. Tripathy, N.K. and Georg, G.I. (2004) Tetrahedron Lett., 45, 5309–5311. (a) Kim, S.H., Figueroa, I., and Fuchs, P.L. (1997) Tetrahedron Lett., 38, 2601–2604; (b) Bamford, S.J., Luker, T., Speckamp, W.N., and Hiemstra, H. (2000) Org Lett., 2, 1157–1160. Fürstner, A., Gartner, T., and Weintritt, H. (1999) J. Org. Chem., 64, 2361–2366. Boger, D.L. and Hong, J. (2001) J. Am. Chem. Soc., 123, 8515–8519. Harrington, P.E. and Tius, M.A. (2001) J. Am. Chem. Soc., 123, 8509–8514. 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.

82 (a) Martin, S.F., Humphrey, J.M.,

83

84

85

86 87

88

89

Ali, A., and Hillier, M.C. (1999) J. Am. Chem. Soc., 121, 866–867; (b) 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. (a) Nagata, T., Nakagawa, M., and Nishida, A. (2003) J. Am. Chem. Soc., 125, 7484–7485; For other RCM conditions, see also: (b) Young, I.S. and Kerr, M.A. (2007) J. Am. Chem. Soc., 129, 1465–1469. Fürstner, A., Grabowski, J., and Lehmann, C.W. (1999) J. Org. Chem., 64, 8275–8280. Scheiper, B., Glorius, F., Leitner, A., and Fürstner, A. (2004) Proc. Natl. Acad. Sci. U. S. A., 101, 11960–11965. Fürstner, A. (2004) Eur. J. Org. Chem., 943–958. For the original total synthesis paper see: (a) Tricolorin A., Fürstner, A., and Müller, T. (1998) J. Org. Chem., 63, 424–425; (b) Tricolorin, G., Fürstner, A., and Muller, T. (1999) J. Am. Chem. Soc., 121, 7814–7821. For the original total synthesis of woodrosin I, see: (a) Fürstner, A., Jeanjean, F., and Razon, P. (2002) Angew. Chem. Int. Ed., 41, 2097–2101; (b) Fürstner, A., Jeanjean, F., Razon, P., Wirtz, C., and Mynott, R. (2003) Chem. Eur. J., 9, 320–326; (c) Fürstner, A., Jeanjean, F., Razon, P., Wirtz, C., and Mynott, R. (2003) Chem. Eur. J., 9, 307–319. Fürstner, A. and Nagano, T. (2007) J. Am. Chem. Soc., 129, 1906–1907.

183

6 Synthesis of Natural Products and Related Compounds Using Ene–Yne Metathesis Miwako Mori

6.1 Introduction

Metathesis is the most important and useful reaction in recent synthetic organic chemistry [1]. In ring-closing olefin metathesis (RCM), formally, bond fission of two double bonds occurs and the new double bond is formed at the same time to produce a cyclic compound (Equation 6.1). Ring-closing olefin metathesis

+

(6.1) Metathesis of ene–ynes having alkene and alkyne moieties in a molecule is an extremely interesting reaction [2]. In this reaction, the double bond of ene–yne 1 is cleaved and a carbon–carbon bond is formed between the double and triple bonds, and the cleaved alkylidene part of the double bond migrates onto the alkyne carbon to produce a cyclic compound 2 having a 1,3-diene moiety (Equation 6.2). R1

Ring-closing enyne metathesis R1

R R 1

2

(6.2)

The first ene–yne metathesis was reported by Katz et al. [3], who used a Fischer tungsten–carbene complex. Then, Mori et al. reported a chromium-catalyzed ene–yne metathesis [4]. It was later found that the ruthenium–carbene complexes [Ru]-IV and [Ru]-I were very effective for ene–yne metathesis [5]. The reaction would proceed via a [2 + 2]-cycloaddition of a ruthenium–carbene complex 3 with an alkyne part to produce ruthenacyclobutene 4, and ring opening of this latter species affords a ruthenium carbene complex 5, which reacts with an alkene part to produce ruthenacyclobutane 6. Subsequent ring opening of 6 gives a cyclized compound 2, and a ruthenium–carbene complex is regenerated (Scheme 6.1, Route 1). The other mechanism considered also involves at first reaction of the Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts. Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32440-8

184

6 Synthesis of Natural Products and Related Compounds Using Ene–Yne Metathesis

R1

R1

R1

[Ru] [Ru]

R1 Ph

R 4

[Ru]

R

1

R

1

Route 1

Route 2

3

R1 [Ru] 3′

R1 2

R

R

7

[Ru]

[Ru] R

R 8 Scheme 6.1

6

[Ru]

Ph [Ru]

R1

R

5

3 R1

[Ru]

R1

9

1

[Ru] R 10

R1

Ring-closing ene–yne metathesis.

alkene part of ene–yne 1 with a ruthenium carbene complex to afford a new ruthenium carbene complex 7. The latter species reacts with the alkyne part to produce ruthenacyclobutene 8, and its subsequent ring opening gives ruthenium carbene 9, which undergoes [2 + 2]-cycloaddition with the alkene part of ene–yne 1 to produce ruthenacyclobutane 10. Ring opening of 10 gives cyclic compound 2, and the ruthenium carbene complex 7 is regenerated (Scheme 6.1, Route 2). Later, the detailed study on the reaction mechanism was shown by Lippstreu and Staub, who described that the reaction would proceed via route 2, and ruthenacyclobutene 4 generated from the alkyne part of ene–yne 2 and ruthenium carbene complex 3 do not exist as local minimum in the catalytic cycle [6]. Using ruthenium carbene complexes [Ru]-IV and [Ru]-I, various carbo- and heterocycles could be synthesized from the corresponding ene–ynes [5]. Furthermore, diene–yne metathesis, cross ene–yne metathesis, and ring-opening ene–yne metathesis have been developed (Scheme 6.2). Novel routes for the synthesis of various structurally complex compounds, such as natural products and related biologically active substances, were thus available. Mori et al. reported the synthesis of heterocycles having a diene moiety using ene–yne RCM [5]. Ene–ynes 11a–c were treated by Grubbs’s catalyst [Ru]-IV (1 mol%, C6 H6 , rt) at room temperature to afford heterocycles 12a–c in high yields. Using this procedure, five- to nine-membered heterocycles could be synthesized (Scheme 6.3). In this reaction, ene–ynes having a terminal alkyne did not give satisfactory results [7]. For example, RCM of enyne 11d afforded cyclic compound 12d in only 21% yield. It is reasoned that an alkene part in product 12d further reacts with

6.2 Synthesis of Natural Products and Related Compounds Using Ene–yne Metathesis

· Ring-closing ene–yne metathesis R

R · Diene–yne metathesis R′

R′ · Cross ene–yne metathesis

+

R1

R1

R2 R2

· Ring-opening metathesis

Scheme 6.2

n+2

n

Various ene–yne metathesis.

R [Ru]-IV (1 mol%) ( ) n

( )n

C6H6, rt

N Ts

N Ts 11a n = 0 R = Me 11b n = 1 R = CH2OTBS 11c n = 2 R = CH2OTBS Scheme 6.3

R

12a 12b 12c

91% 88% 77%

Synthesis of heterocycles using ruthenium catalyst.

methylidene ruthenium carbene to afford ruthenium carbene 14d, which would be coordinated by the alkene part to produce 15d. Thus, the catalytic activity would decrease (Scheme 6.4). In fact, when the reaction of 11d using [Ru]-I was carried out under ethylene gas, the catalytic activity was improved to afford 12d in 90% yield even with the use of 1 mol% of the ruthenium catalyst [Ru]-I [7]. The higher reactivity observed in ene–yne RCM in the presence of ethylene gas has often been advantageously explored in applications to natural product synthesis.


The first application of ene–yne RCM was reported in the total synthesis of the natural product (−)-stemoamide [8]. (−)-Pyroglutamic acid was converted into ene–yne 16 having an ester group on the alkyne, and RCM of ene–yne 16 was carried out in the presence of ruthenium–carbene complex [Ru]-I (4 mol%) to afford bicyclic compound 17 in 87% yield. Conversion of 17 into 18 followed by

185

186


[Ru]-I (1 mol%) N

N

CH2Cl2, rt, 22 h

Ts

Ts

11d

12d Under Ar Under H2C

[Ru]

21% 90%

CH2 [Ru]

[Ru]

[Ru]

N

N

N

N

Ts

Ts

Ts

Ts

12d

Scheme 6.4

13

14

15

Metathesis of enyne having terminal alkyne under ethylene gas.

CO2Me MeO2C O N H (−)-Pyroglutamic acid

HO2C

N

H

O [Ru]-I (4 mol%) H

CH2Cl2, rt 87%

16 MeO2C

NaBH4

17 Me

Me

H

MeOH

N

O

1. NaOH 2. CuBr2 on Al2O3 3. Et3N

85% 18

NaBH4

O H

O

76% from 18

O H

O

N

O

H 19

Me H NiCl2·6 H2O

O

N

N

O

H (−)-Stemoamide

Scheme 6.5

Total synthesis of (−)-stemoamide.

halo-lactonization gave 19. From this compound 19, (−)-stemoamide could be synthesized by conjugate reduction (Scheme 6.5). Carbacephem 22a and carbapenem 22b were synthesized, respectively, from ene–ynes 21a and 21b, which were prepared from 4-acetoxy-azetidinone 20. The yield in compound 22b (29%) is lower compared to that in 22a (80%) because of the highly strained, fused 4,5-membered ring system (Scheme 6.6) [9].


OAc

Me

NH

O

OTBS

OTBS

OTBS Me

[Ru]-I (5 mol%)

Me

CH2CH2, reflux

N

O

20

Me O

Me

80%

21a

OH

N 22a

OH [Ru]-I (5 mol%)

Me N

O

CH2Cl2, reflux

O

29%

21b Scheme 6.6

Me N 22b

Construction of carbacephem and carbapenem skeleton.

O

O

O

23

[Ru]-I (10 mol%)

CDCl3, 50 °C

CH2Cl2, rt 40%

1d

O

O

24

91%

O O O (±)-Differolide

Scheme 6.7

Synthesis of (±)-differolide.

(±)-Differolide was easily synthesized by ene–yne metathesis [10]. Ene–yne 23 was reacted with catalyst [Ru]-I to give lactone 24, which was spontaneously dimerized to afford (±)-differolide (Scheme 6.7). An enantioselective biomimetic synthesis of (−)-longithorone A was accomplished on the basis of the proposed biosynthesis [11]. Two [12]-paracyclophanes 25 and 26 were synthesized from common intermediate 27 by applying ene–yne metathesis macrocyclization in 42 and 31% yields, respectively. Intermolecular Diels–Alder reaction of 25 and 26 provided 32. Deprotection followed by oxidation gave 33, which spontaneously gave longithorone A via transannular Diels–Alder reaction (Scheme 6.8). The total synthesis of (+)-anatoxin-a was achieved by Martin et al. 12a, b and Mori et al. [12c, d] by using the same strategy. The key step is the construction of an azabicyclo[4.2.1]nonene ring system. For that purpose, the 2,5-cis-disubstituted pyrrolidine derivative 34 was synthesized from (+)-pyroglutamic acid. Ene–yne metathesis applied to 34 was carried out using second-generation catalyst [Ru]-II to form the azabicyclic compound 35 [12a, b], which was chemoselectively dihydroxylated to provide 36. From the latter compound, anatoxin-a was synthesized (Scheme 6.9). Using a similar procedure, (+)-ferruginine was synthesized from (−)-pyroglutamic acid [13]. Construction of the azabicyclo[3.2.1]octene ring system was carried out by applying RCM to ene–yne 37. Wacker oxidation of the resultant

187

188


Me O

Me

OH

H

OR

MeO

Me

Me

O

O

OHC

OHC

RO

25

Longithorone A Me

OMe Me

26

Me Me TIPS

TBSO TBSO OMe

H

OMe O

Me

1. [Ru]-I (50 mol%) H2C CH2 CH2Cl2 2. TBAF 42%

OTBS 28

27 Me

TBSO Me

26

OMe OTBS 29 (dr > 20 : 1)

Me TBSO

[Ru]-I (50 mol%) H2C CH2

TBSO

27 OMe

MeO TBSO

31%

OTBS H

OTBS TBSO

31

30

(E/Z = 3.9 : 1, dr = 1 : 1.4) Me OR H

MeO Me2AlCl

25 + 26

CH2Cl2, −20 °C

Me OHC

RO OMe

Me 32: R = TBS (dr = 1 : 1.4)

Me O H

O 1. TBAF 2. PhI(O)

Me OHC 33

Scheme 6.8

RO OMe Me

Total synthesis of (−)-longithorone A.

(−)-Longithorone A 90%

25


Cbz N [Ru]-II O

CO2H N H (+)-Pyroglutamic acid

N Cbz

1. OsO4

OH OH

2. NaHSO3 76%

1. NaIO4 2. TMSI

35 H ·HCl N Me O

3. HCl 95%

36 Scheme 6.9

CH2Cl2, rt, 16 h Me 84%

34

Cbz N Me

Me

(+)-Anatoxin-a·HCl

Total synthesis of anatoxin-a.

Boc N [Ru]-I O

N H

N Boc

CO2H

37 1. PdCl2 – CuCl2 H2O, 95 °C 2. TFA 3. CH2O NaBH3CN 73% Scheme 6.10

Me N

Me

CH2Cl2 Reflux,10 h 86%

38

O

(+)-Ferruginine Synthesis of (+)-ferruginine.

diene 38 afforded methyl ketone, and then deprotection followed by methylation gave (+)-ferruginine (Scheme 6.10). Kozmin et al. developed a highly efficient synthesis of cyclic compound 41a bearing a methyl ketone functionality from ene–yne 39a having a silyloxy group on the alkyne using RCM [14a]. As an application of this method, they succeeded in the synthesis of eremophilanes [14b]. RCM of ene–yne 39b having a triisopropylsilyloxy group on the alkyne followed by treatment with HF gave cycloalkene 41b having the methyl ketone functionality. From this compound 41b, α- and β-eremophilanes were synthesized (Scheme 6.11). New allo-colchinoids functionalized at C10 or C11 in the C-ring were synthesized using the RCM of ene–yne 43 for the construction of the seven-membered ring [15]. The reaction proceeded smoothly to give 44 using catalyst [Ru]-II in 92% yield. Deprotection followed by PCC oxidation gave 45, which was subjected to a Diels–Alder aromatization sequence to form 46. Amination followed by acetylation gave allo-colchicines (Scheme 6.12).

189

190


39a

O

OTIPS

OTIPS

[Ru]-I

HF

60 °C, 15 min 98%

CH3CN 40a

41a H

H Me TIPSO

1. [Ru]-VI (5 mol%) C6H6, 50 °C 2. HF

H2, Pd/C 95%

O

85%

Me

39b

Me

Me

Me 41b

H

H H Me

O

H

1. Ph3P CH2 2. H2, Pd/C 90% Me

Me

Me Me a-Eremophilane

Me

42

Me 1. K2CO3 2. Ph3P CH2 3. H2, Pd/C

Me

H Me

H Me b-Eremophilane

77%

Scheme 6.11

Me

Total synthesis of eremophilane.

MeO

[Ru]-II (5 mol%)

MeO

MeO

CH2Cl2 Reflux, 4 h

MeO

OMe OTBS

2. PCC 52%

OMe OTBS

92%

43

1. TBAF

44 MeO

MeO O MeO

1.

CO2Me

2. DDQ OMe 85% 45

O MeO OMe MeO2C 46

MeO NHAc MeO MeO MeO2C

allo -Colchicine

Scheme 6.12

Synthesis of allo-colchicine.

1. NH4OAc NaBH3CN 2. Ac2O 69%

6.2 Synthesis of Natural Products and Related Compounds Using Ene–yne Metathesis TBDPSO Me

Me

Me

CO2H

OTBDPS OH Me

O Me

48

OPMP DCC

47

OTBDPS Me OPMP

Me 2. H2O then CH2N2 3. K2CO3 MeOH

49 OTBDPS Me OPMP

[Ru]-II

CO2Me Toluene 80 °C 73%

Me

Me

OPMP TMS

1. KHMDS Toluene TMSCl

O

CO2Me Me Me

50

51

69% from 47

OTBDPS Me OPMP

2. Ac2O, Sc(OTf)3

CO2Me 3. LiHMDS MOMCl TBDPSO

Me 52

Scheme 6.13

OTBDPS O Me O

1. PhSH, AlCl3

OPMP TBDPSO

47%

Me

Subunit B of quartromicin

Synthesis of subunit of quartromicin.

The agalacto-spirolactone B subunit of quartromicins has been synthesized using the Claisen–Ireland/RCM of ene–yne approach by Haudrechy et al. [16]. Compound 49 was converted into ene–yne 50, which was treated with catalyst [Ru]-II (toluene, 80 ◦ C) to afford 51 in 73% yield. From this compound 51, subunit B of quartromicin was synthesized (Scheme 6.13). An enantioselective synthesis of (−)-galanthamine was realized in 11 steps starting from isovanillin (Scheme 6.14). The ene–yne 53 (ee = 92%) underwent an efficient RCM reaction in the presence of catalyst [Ru]-I (3 mol%) to give diene 54 in 85% yield. Hydroboration and oxidation of the terminal alkene present in 54 gave an intermediate homoallylic alcohol in excellent yield, which underwent sequential Heck reaction with a palladium catalyst and allylic oxidation with SeO2 to give 56. From compound 56, (−)-galanthamine could be synthesized via mesylation of the hydroxyl group, cleavage of N-Boc, and intramolecular nucleophilic substitution [17]. For the synthesis of angucyclinone-type natural products, a concise and highly enantioselective route was developed [18]. Chiral vinylcyclohexene derivative 58 was synthesized using RCM applied to ene–yne 57 under an atmosphere of ethylene in high yield. Intermolecular Diels–Alder reaction of 58 and 59 followed by aromatization gave compound 60 having the benz[a]anthraquinone skeleton. By utilizing this strategy, the total synthesis of five natural products: YM-181741, (+)-ochromycinone, (+)-rubiginone B, (−)-tetragomycin, and MM-47755, was accomplished (Scheme 6.15).

191

192


Me NBoc

Me NBoc

I O OMe

9-BBN-H

[Ru]-I (3 mol%)

I

CH2Cl2, rt

O

H

OMe

85% 53

Me NBoc OH I

then H2O2 NaOH

H

O OMe

91%

54

55

Me OH NBoc

1. Pd(OAc)2 dppp Ag2CO3

H

Me N 1. MsCl

OH

2. TFA

2. SeO2

O

35%

OMe

H

OH

O H

23% OMe

56 (a /b = 4.8 : 1)

Scheme 6.14

(−)-Galanthamine

Total synthesis of galanthamine.

O OTBS

[Ru]-II H2C CH2

Toluene, 80 °C 12 h 88%

57

Br

OTBS 1.

OH

O OAc O

59

2. K2CO3, MeOH OH O

3. h n, O2 4. TBAF

58

60

34%

O

O

CH2OH

O

O

Me

(+)-Ochoromycinone

YM-181741 O

Me

O

OH

OH O

Scheme 6.15

Me

O

O

(+)-Rubiginone B2 Me OH

OMe O

Tetrangomycin

O

OMe O

OH O

OH O

O

MM-47755

Synthesis of benz[a]anthraquinone skeleton.


Synthesis of anthramycin derivative 69a was achieved using RCM and alkene cross-metathesis (CM) (Scheme 6.16) [19]. L-Methionine was converted to ene–yne 61, and subsequent RCM with catalyst [Ru]-I under an atmosphere of ethylene gave pyrrolidine derivative 62. After deprotection and condensation with the commercially available acid chloride 63, the resulting amide 64 underwent reductive cyclization using Zn/AcOH followed by treatment with diluted HCl to SMe MeO2C H HO H

Cbz N

NH2

61

1. TMSCl NaI

OBn Me

NO2

OBn

CO2Me H

[Ru]-VIII (10 mol%) H2C CHCO2Et

1. Zn, AcOH

OBn H N

Me

2. aq. HCl

O

COCl 76% 63

62

76%

N

NO2

MeO2C H Cbz N

CH2Cl2 rt, 24 h

O l -Methionine

2. Me

[Ru]-I H2C CH2

N

86%

O 65

64

BnO Me

H N

O H

RhCl3·3H2O

N

CH2Cl2, rt, 17 h

CO2Et

O

60%

O H

EtOH 110 °C 50%

66

Ph

BnO

H N

Me

O H

1. TFA BF3·Et2O

N CO2Et

O

O Me

2. PhCH(OMe)2 p-TsOH

H

NaBH4

N CO2Et

O 68

53%

67

O

N

Ph O Me

OH H

N

Me

OH H N

N O

Scheme 6.16

Synthesis of anthramycin derivative.

H N

R

69a: R = CO2Et 69b: R = CONH2

OH

O (+)-Anthramycin

NH2 O

193


194

Me

Me

Me

TBSO

TBSO OH Pd(OAc)2 Cu(OAc)2 68%

Me 70

H

H O

H BuO

O

71 (dr > 12 : 1)

72

Me [Ru]-II

H

O

O

O

H

H

Me

CD2Cl2, rt

O

2. LDA, MeI (90%)

Me

O

1. [Ru]-II CH2Cl2, rt (70%)

Me

66%

(−)-Dihydroxanthatin

73 Synthesis of (−)-dihydroxanthatin.

O

Me

AO

H O B

C

E

O Me

D

F

O

G

Me

H

O

O

OH

H Me

HO

O

H

Me

Scheme 6.17

O

H

OBu

H

O

Lancifodilactone G [Ru]-III H2C CH2

TMSO H BnO Me

CO2Me

CH2Cl2 TMSO 40 °C, 30 h 88%

BnO

74

Me

H

TMSO

CH2Cl2, 40 °C 42 h

BnO

72%

75

1. K3Fe(CN)6 K2OsO2(OH)4 (DHQD)2PHAL

MeO2C

OTBS HO

51% (three steps from 76)

BnO

OH

1. t -BuOK

H

A O

O B

2. TBAF H Me

H Me 76

O

HO

2. HCl/MeOH 3. TBSOTf

Scheme 6.18

MeO2C

[Ru]-III

88% BnO

77

Construction of ABC ring of lancifodilactone G.

H Me 78

C

6.3 Synthesis of Natural Products and Related Compounds Using Ene–yne Cross-metathesis (CM)

provide the pyrrolo-1,4-benzodiazepinone 65. To convert the vinyl group into an α,β-unsaturated ester group, CM with ethyl acrylate was carried out using catalyst [Ru]-VIII. The reaction proceeded smoothly to give compound 66 in 60% yield. Isomerization of the double bond in the pyrrolidine ring using RhCl3 ·H2 O afforded compound 67. Debenzylation followed by protection gave 68, the amido group of which was converted into aminal 69a. Conversion of 69b into (+)-anthramycin was previously described by Stille et al. [20]. Morken and Evans succeeded in the synthesis of (−)-dihydroxanthatin using RCM and CM [21]. Allylic alcohol 70 was converted into tetrahydrofuran 71 by the catalytic Oshima–Utimoto reaction [22]. From compound 71, lactone 72 was synthesized. Ene–yne metathesis of 72 using catalyst [Ru]-II followed by methylation gave bicyclic compound 73. CM of 73 and methyl vinyl ketone in the presence of [Ru]-II afforded (−)-dihydroxanthatin (Scheme 6.17). Lancifodilactone G has eight rings with complex cyclic connectivity. Paquette et al. synthesized the ABC rings of lancifodilactones G, compound 78, using ene–yne RCM and then CM as the key steps (Scheme 6.18) [23]. RCM of ene–yne 74 using catalyst [Ru]-III in CH2 Cl2 under an ethylene atmosphere gave cycloheptene derivative 75, which was subjected to CM with methyl acrylate and catalyst [Ru]-III to give α,β-unsaturated ester 76. From this functionalized seven-membered ring intermediate, the target compound 78 was obtained.

6.3 Synthesis of Natural Products and Related Compounds Using Ene–yne Cross-metathesis (CM)

A novel synthetic procedure of 1,3-diene from alkyne and ethylene using CM was developed in 1997 by Mori et al. [24]. When a solution of alkyne 79a in CH2 Cl2 was stirred under an atmosphere of ethylene at rt in the presence of catalyst [Ru]-I, 1,3-diene 80a was obtained in 62% yield. It is interesting that, formally, the double bond of ethylene is cleaved and each methylene part is introduced onto the alkyne carbon to produce 1,3-diene 80a (Scheme 6.19). The possible reaction course is shown in Scheme 6.20. Reaction of methylidene carbene complex 81, generated from 3 and ethylene, with alkyne 79 gives ruthenacyclobutene 82, ring opening of which gives ruthenium carbene 83. It reacts with ethylene to afford ruthenacyclobutane 84, which undergoes ring opening to give 1,3-diene 80, and 81 is regenerated. If 81 reacts with ethylene, ruthenacyclobutane would be produced, but this is a nonproductive process as 81 would be regenerated (Scheme 6.20). However, propargyl ester 79a or amide 79b gave good results, whereas homopropargyl amide 80c led to 1,3-diene 80c in only 11% yield (Scheme 6.19). Presumably, a heteroatom at the propargylic position is important, and the ruthenium catalyst would be coordinated by the heteroatom at first and then the reaction would proceed. When the second-generation ruthenium–carbene complex [Ru]-II was used for this reaction (toluene, 80 ◦ C), alkynes 79d and 79e lacking heteroatoms at the propargylic positions, gave 1,3-dienes 80d and 80e in good yield [25a, b].

195

196


H 2C

(1 atm)

CH2

R1

[Ru]-I or [Ru]-II

R2

R1

79

R2 80


Et

OBz

CH2Cl2, rt, 45 h 62%

79a Hex

Et

[Ru]-I (5 mol%) H2C CH2 N n Ts

OBz 80a

Hex

CH2Cl2, rt

79b: n = 1 79c: n = 2

N n Ts

80b (81%) 80c (11%) Me

[Ru]-II (5 mol%) H2C CH2

Me

Toluene, 80 °C 30 min 85%

79d

80d TMS

[Ru]-II (5 mol%) H2C CH2 TMS

MeO

Toluene, 80 °C 16 h 87%

79e

Scheme 6.19

80e

Synthesis of 1,3-diene using cross-metathesis. [Ru]

Ph [Ru]

MeO

3 H2C

Ph [Ru]

[Ru] 81

CH2

R

R′

80 [Ru] R

R′ 84

R′

R

H2C CH2

79

[Ru]

[Ru] R′

R 82

Scheme 6.20

R′

R 83

Possible reaction course for the formation of 1,3-diene.

6.4 Synthesis of Natural Products Using Skeletal Reorganization

Furthermore, the reaction was more rapid and several substituents on the alkyne were tolerated (TMS, CO2 Me). Synthesis of natural products using ene–yne CM is interesting because a 1,3-diene moiety can be constructed onto the alkyne carbons at a later stage. Anolignans were synthesized using CM of ene–yne as a key step. 1,3-Diene 86 could be synthesized from alkyne 85 by treatment with catalyst [Ru]-II under ethylene gas. Palladium-catalyzed deacetoxylation followed by deprotection gave anolignan A. Anolignan B could be synthesized in a similar manner. It is interesting to note that the two methylene parts of the anolignan skeleton were introduced at a later stage of the total synthesis using CM (Scheme 6.21) [26]. A short and efficient synthesis of highly substituted tetrahydropyridines 88 was achieved from a monosubstituted alkyne, a terminal alkene, and an imine by a combination of ene–yne CM and aza-Diels–Alder reaction under high pressure. CM of terminal alkyne and alkene afforded diene 87, which reacted with imine to give pipecolic acid derivative 88 in high yield (Scheme 6.22) [27]. The reaction was further extended to intramolecular Diels–Alder reaction, and cis-hexahydro-1H-indene 90 was synthesized from diene 89 and terminal alkyne in one operation. The possible intermediate 92 was spontaneously converted into 90. Deprotection of the silyl group followed by pyridinium chloro chromate (PCC) oxidation gave indanone 91 (Scheme 6.23) [28]. New vitamin D receptor antagonists 24,24-ethanovitamine D3-26,23-lactones 93a and 93b and their analogs were synthesized (Scheme 6.24) [29]. The CD-ring precursors 99a and 99b were efficiently prepared by using a ruthenium-catalyzed intermolecular ene–yne CM applied to 95 in the presence of ethylene as a key step. Cyclopropanation of the enone moiety in the resulting product 96, followed by treatment with DIBAL-H and then deprotection, gave compounds 98a and 98b, epimeric at C23. Chemoselective oxidation of 98a and 98b followed by palladium-catalyzed coupling with 100 and then deprotection afforded 93a and 93b, respectively. Lee et al. succeeded in the total synthesis of (−)-amphidinolide E, whose side chain was constructed using ene–yne CM [30]. Alkyne 101 was first reacted with ethylene in the presence of catalyst [Ru]-II to give 102, which was further engaged in situ in a chemoselective CM with 2-methyl-1,4-pentadiene (103) to produce triene 104 in 65% yield along with diene 102 in 19% yield. The isolated diene 102 was recycled and further reacted with 103 in a similar manner to afford triene 104, which was finally converted to sulfone 105. Condensation of sulfone 105 and aldehyde 106 afforded compound 107, which was further elaborated to seco-acid 108. Lactonization of 108 followed by deprotection completed the total synthesis of amphidinolide E (Scheme 6.25).


Transition-metal-catalyzed cycloisomerization of ene–ynes 109 occurring with skeletal rearrangement can produce Type I product 111 (similar to ene–yne RCM

197

198


OMs

OAc

OMs [Ru]-II (10 mol%) H2C CH2

O O

AcO AcO OMs

O

Toluene, 80 °C 86%

OAc OMs

O

85

86 OH

OH 1. Pd2dba3, PBu3 HCO2H 2. PhLi 89%

OH

OH

O

HO

O Anolignan A

Scheme 6.21

Synthesis of anolignan s using ene–yne CM.

[Ru]-I AcO

OBn AcO

87%

1.

Anolignan B

CO2Et

CO2Et

NTs

NTs

OBn

HO 2. NaOMe 90% Scheme 6.22

OBn 87

88 Synthesis of pipecolic acid derivatives.

OTBS

CN OTBS

CN 89 +

[Ru]-III (7.5 mol%) C6H6, 80 °C 24 h OMe 62%

H MeO

CN OTBS

MeO H

Scheme 6.23

91 Synthesis of cis-fused carbo-bicycles.

2. PCC 64%

90

CN O

MeO

1. TBAF

92


25

23S

23R

H O

H O

199

OH H

O

O

21

HO

HO

OH

HO

OH

1a, 25-Dihydroxyvitamin D3

OH

93a

93b OTBS

CHO

H

H O

1. 2. TPAP NMO 77%

H Br

OTBS

94

H O

CH2Cl2

H Br

[Ru]-VIII H2C CH2

92%

95

OTBS

H Br

96

23 Me3S(O)I NaI

1. DIBAL-H

H O

OTBS 2. TBAF 95%

85% H Br

O

98b Scheme 6.24

100 93a

2. HF

H Br

98a (23 S ), 98b (23 R ) (1 : 1.2)

TBSO OTBS Pd(PPh3)4, Et3N

H O

97%

OH

1.

23 98a

OH

H Br

97

MnO2

H

99a

62%

93b Synthesis of 24,24-ethanovitamin D3 lactones.

product) or Type II product 112 resulting from single and double cleavage from a cyclopropyl metal carbene intermediate (Scheme 6.26) 110 (31, 32). Fürstner et al. achieved the formal total syntheses of the antibiotics metacycloprodigiosin and streptorubin B by a platinum-catalyzed skeletal reorganization reaction (Scheme 6.27) [33]. The key step leading to the meta-bridged pyrrole core structures consisted of a ‘‘formal’’ metathesis reaction of electron-deficient

200


EtO2C

OMOM

O

OH

O HO

O

TIPSO

O

HO 101 [Ru]-II H2C

Amphidinolide E

CH2 then

EtO2C

EtO2C

OMOM

O

103

TIPSO

OMOM

O TIPSO

65% 104 (102 19%)

102 O O

H

O

O2S N N

O Ph TIPSO N N 105

OMOM

O O

OMOM

TIPSO

O

OTBDPS

OTBDPS 106 LiHMDS 74%

1. NaOH 2. IBX 3. NaClO2 4. TBAF 74%

107

OMOM

O O

HO CO2H

O 108 Scheme 6.25

Synthesis of amphidinolide E.

1. EtO H RuCl2(p -cymene)2 2. HCl 34%

Amphidinolide E

6.4 Synthesis of Natural Products Using Skeletal Reorganization R′ R

Type I XnM R′

R′

MXn

111

R

R 109

R R′

110 Type II 112

Scheme 6.26

Skeletal reorganization.

N

N

H

H N

MeO

N

Metacycloprodigiosin Ts N

Streptorubin B

PtCl2 TsN (5 mol%)

TsN Bu3SnH

42% O

O 113a

O 115

Excess KAPA

2. PhOCSCl 3. Bu3SnH AIBN

Ts N

Cat. Pd(0) 70%

114a

1. LiAlH4

53%

H N

MeO

H N

Metacycloprodigiosin

75% N 116

N

Ts

117

PtCl2 (5 mol%) TsN

H

TsN Streptorubin B

Toluene 50 °C O

79% 113b

Scheme 6.27

O 114b

118

Formal total syntheses of streptorubin B and metacycloprodigiosin.

201

202


OMe

Cl

O N H

O N

RN

119: R = SEM

Roseophilin

H

TBDMSO OTBDMS PtCl2 (5 mol%)

H

O O

Toluene, 80 °C 99%

120 Scheme 6.28

TBSO

121

122

Formal total synthesis of roseophilin.

ene–ynes 113a and 113b catalyzed by PtCl2 . The skeletal reorganization products 114a and 114b were converted into the respective target molecules. Trost et al. succeeded in the formal total synthesis of roseophilin [34]. Macrocyclic compound 121 was synthesized from ene–yne 120 by platinum-catalyzed skeletal reorganization. From 121, the 1,4-diketone 122 was synthesized and converted to pyrrole derivative whose transformation into roseophilin had previously been reported (Scheme 6.28).

References 1 (a) Grubbs, R.H. (ed.) (2003) Hand-

book of Metathesis, Wiley-VCH Verlag GmbH; (b) Fürstner, A. (ed.) (1998) Topics in Organometallic Chemistry, vol. 1, Springer-Verlag, Berlin, Heidelberg; (c) Trunk, T.M. and Grubbs, R.H. (2001) Acc. Chem. Res., 34, 18–29; (d) Fürstner, A. (2000) Angew. Chem. Int. Ed., 39, 3012–3043; (e) Herisson, J.-L. and Chauvin, Y. (1971) Macromol. Chem., 141, 161. 2 Reviews for Enyne Metathesis; (a) Mori, M. (1998) Top. Organomet. Chem., 1, 133–154; (b) Poulsen, C.S. and Madsen, R. (2003) Synthesis, 1–18; (c) Mori, M. (2003) in Handbook of Metathesis, vol. 2 (ed. R.H. Grubbs), Wiley-VCH Verlag GmbH, pp. 176–204; (d) Giessert, A.J. and Diver, S.T. (2004) Chem. Rev., 104, 1317–1382; (e) Mori,

M. and Kitamura, T. (2005) Ene–Yne and alkyne metathesis, in Comprehensive Organometallic Chemistry III, Transition Metal Organometallics in Organic Synthesis (ed. T. Hiyama), Elsevier, pp. 271–310. 3 (a) Katz, T.J. and Sivavec, T.M. (1985) J. Am. Chem. Soc., 107, 737–738; (b) Sivavec, T.M., Katz, T.J., Chiang, M.Y., and Yang, G.X.-Q. (1989) Organometallics, 8, 1620–1625; (c) Katz, T.J. and Yang, G.X.-Q. (1991) Tetrahedron Lett., 32, 5895–5898. 4 (a) Watanuki, S., Ochifuji, N., and Mori, M. (1994) Organometallics, 13, 4129–4130; (b) Watanuki, S., Ochifuji, N., and Mori, M. (1995) Organometallics, 14, 5062–5067; (c) Mori, M. and Watanuki, S. (1992) J. Chem. Soc., Chem. Commun., 1082–1084; (d) Watanuki,

References

5

6 7

8

9

10 11

12

13

14

15 16 17

S. and Mori, M. (1993) Heterocycles, 35, 679–682. (a) Kinoshita, A. and Mori, M. (1994) Synlett, 1020–1022; (b) Kinoshita, A., Sakakibara, N., and Mori, M. (1999) Tetrahedron, 55, 8155–8167. Lippstreu, J.J. and Straub, B.F. (2005) J. Am. Chem. Soc., 127, 7444–7457. Mori, M., Sakakibara, N., and Kinoshita, A. (1998) J. Org. Chem., 63, 6082–6083. (a) Kinoshita, A. and Mori, M. (1996) J. Org. Chem., 61, 8356–8357; (b) Kinoshita, A. and Mori, M. (1997) Heterocycles, 46, 287–299. (a) Barrett, A.G.M., Baugh, S.P.D., Braddock, D.C., Flack, K., Gibson, V.C., Procopiou, P.A., White, A.J.P., and Williams, D.J. (1998) J. Org. Chem., 63, 7893–7907; (b) Duboc, R., Henaut, C., Savignac, M., Genet, J.-P., and Bhatnagar, N. (2001) Tetrahedron Lett., 42, 2461–2464. Hoye, T.R., Donaldson, S.M., and Vos, T.J. (1999) Org. Lett., 1, 277–280. Layton, M.E., Morales, C.A., and Shair, M.D. (2002) J. Am. Chem. Soc., 124, 773–775. (a) Brenneman, J.B. and Martin, S.F. (2004) Org. Lett., 6, 1329–1332; (b) Brenneman, J.B., Machauer, R.M., and Martin, S.F. (2004) Tetrahedron, 60, 7301–7314; (c) Mori, M., Tomita, T., Kita, Y., and Kitamura, T. (2004) Tetrahedron Lett., 45, 4397–4399; (d) Tomita, T., Kita, Y., Kitamura, T., Sato, Y., and Mori, M. (2006) Tetrahedron, 62, 10518–10527. Aggarwal, V.K., Astle, J., and Rogers-Evans, M. (2004) Org. Lett., 6, 1469–1472. (a) Reddy, S. and Kozmin, S.A. (2004) J. Org. Chem., 69, 4860–4862; (b) Schramm, M.P., Reddy, D.S., and Kozmin, S.A. (2001) Angew. Chem. Int. Ed., 40, 4274–4277. Boyer, F.-D. and Hanna, I. (2007) Org. Lett., 9, 715–718. Bedel, O., Francais, A., and Haudrechy, A. (2005) Synlett, 2313–2316. Satcharoen, V., McLean, N.J., Kemp, S.C., Camp, N.P., and Brown, R.C.D. (2007) Org. Lett., 9, 1867–1869.

18 Kaliappan, K.P. and Ravikumar, V.

(2007) J. Org. Chem., 72, 6116–6126. 19 Kitamura, T., Sato, Y., and Mori, M.

(2004) Tetrahedron, 60, 9649–9657. 20 Pena, M.R. and Stille, J.K. (1989) J. Am.

Chem. Soc., 111, 5417. 21 Evans, M.A. and Morken, J.P. (2005) Org.

Lett., 7, 3371–3374. 22 (a) Fugami, K., Oshima, K., and

23 24

25

26 27 28 29

30

31

Utimoto, K. (1987) Tetrahedron Lett., 28, 809; (b) Fugami, K., Oshima, K., and Utimoto, K. (1989) Bull. Chem. Soc. Jpn., 62, 2050. Paquette, L.A. and Lai, K.W. (2008) Org. Lett., 10, 2111–2113. Kinoshita, A., Sakakibara, N., and Mori, M. (1997) J. Am. Chem. Soc., 119, 12388–12389. (a) Mori, M., Tonogaki, K., and Kinoshita, A. (2004) Org. Synth., 81, 1–13; (b) Tonogaki, K. and Mori, M. (2002) Tetraheron Lett., 43, 2235–2238; (c) Smulik, J.A. and Diver, S.T. (2000) J. Org. Chem., 65, 1788–1792; (d) Smulik, J.A. and Diver, S.T. (2000) Org. Lett., 2, 2271–2274; (e) Smulik, J.A., Giessert, A.J., and Diver, S.T. (2002) Tetrahedron Lett., 43, 209–211. Mori, M., Tonogaki, K., and Nishiguchi, N. (2002) J. Org. Chem., 67, 224–226. Schürer, S.S. and Blechert, S. (1999) Tetrahedron Lett., 40, 1877–1880. Mix, S. and Blechert, S. (2005) Org. Lett., 7, 2015–2018. Saito, N., Masuda, M., Saito, H., Takenouchi, K., Ishizuka, S., Namekawa, J., T-Kamimura, M., and Kittaka, A. (2005) Synthesis, 2533–2543. (a) Kim, C.H., An, H.J., Shin, W.K., Yu, W., Woo, S.K., Jung, S.K., and Lee, E. (2008) Chem. Asian J., 3, 1523–1534; (b) Kim, C.H., An, H.J., Shin, W.K., Yu, W., Woo, S.K., Jung, S.K., and Lee, E. (2006) Angew. Chem. Int. Ed., 45, 8019–8021. (a) Trost, B.M. and Tanoury, G.J. (1988) J. Am. Chem. Soc., 110, 1636–1638; (b) Trost, B.M. and Trost, M.K. (1991) J. Am. Chem. Soc., 113, 1850–1852; (c) Trost, B.M. and Trost, M.K. (1991) Tetrahedron Lett., 32, 3647–3650; (d) Trost, B.M., Ynai, M., and Hoogsteen, K. (1993) J. Am. Chem. Soc., 115, 5294–5295; (e) Trost, B.M., Hashmi, A.S.K. (1994)

203

204

6 Synthesis of Natural Products and Related Compounds Using Ene–Yne Metathesis J. Am. Chem. Soc., 116, 2183–2184; (f ) Trost, B.M., Hashmi, A.S.K. (1993) Angew. Chem. Int. Ed., 32, 1085–1087; (g) Trost, B.M. and Chang, V.K. (1993) Synthesis, 824–832. 32 (a) Chatani, N., Furukawa, N., Sakurai, H., and Murai, S. (1996) Organometallics, 15, 901–903; (b) Chatani, N., Morimoto, T., Muto, T., and Murai, S. (1994) J. Am. Chem. Soc., 116,

6049–6050; (c) Chatani, N., Kataoka, K., and Murai, S. (1998) J. Am. Chem. Soc. 120, 9104–9105; (d) Lee, S.I. and Chatani, N. (2009) Chem. Commun., 371–384. 33 F¨ urstner, A., Szillat, H., Gabor, B., and Mynott, R. (1998) J. Am. Chem. Soc., 120, 8305–8314. 34 Trost, B.M. and Doherty, G.A. (2000) J. Am. Chem. Soc., 122, 3801–3810.

205

7 Ring-closing Alkyne Metathesis in Natural Product Synthesis Paul W. Davies

7.1 Introduction

There is ever increasing societal and legislative pressure to minimize waste production and energy consumption while retaining or improving our current standard of living. The chemical community can contribute significantly toward achieving these goals, not least as the synthesis and manipulation of molecules is of fundamental importance across our developing societies. A significant contribution can therefore be achieved by performing molecular synthesis as efficiently as possible to diminish the associated energy consumption and waste production footprint. This can entail developing new and more rapid strategies to construct molecules, and to employ reactions which are inherently efficient; produce minimal waste; and use simple, readily introduced functionality so as to accelerate precursor preparations. The metal-catalyzed alkene metathesis processes discussed throughout this book are shining examples of such advances, and have rapidly been integrated and proven within the crucible of complex natural product synthesis. In this chapter we will look at the application of a sister reaction: alkyne metathesis.

7.2 Alkyne Metathesis

Since its discovery, alkyne metathesis has become a potent catalytic tool applied in the preparation of materials and organometallic complexes as well as bioactive organic molecules including natural products. In an analogous fashion to alkene metathesis, the process involves exchange of methine units between two carbon–carbon triple bond systems (Scheme 7.1). This chapter focuses on one aspect of alkyne metathesis: ring-closing alkyne metathesis (RCAM) in natural product synthesis. A brief background precedes an overview of RCAM, the catalyst systems used in RCAM-containing approaches to natural products, and a discussion of its potential as a synthetic strategy. The value Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts. Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32440-8

206

7 Ring-closing Alkyne Metathesis in Natural Product Synthesis

R2

R2 +

1

R2 Catalyst

1

R

R

Scheme 7.1

R2 +

R1

R1

Alkyne metathesis.

of alkyne metathesis in macrocyclizations is then explored through examples of its successful application in a range of natural product syntheses. 7.2.1 Background to Alkyne Metathesis

After the initial discovery of the first effective catalytic alkyne metathesis using a heterogeneous mixture of tungsten oxides and silica at very high temperature (about 200–450 ◦ C) [1], Mortreux et al. reported alkyne metathesis by a homogeneous catalyst system in 1974 [2a]. This catalytically active system derived from heating a mixture of Mo(CO)6 (or a related molybdenum source) and simple phenol additives such as resorcinol, in high boiling solvents [2]. This system is still employed, even in complex molecule synthesis, and there have been a number of significant refinements made to it. Additionally, well-defined and isolable metal alkylidynes and other precatalysts have been introduced. Progress across this area has been reviewed recently and will not be discussed here in further depth [3]. A major contribution to the field was provided by Katz and McGinnis when they proposed the now-accepted mechanism for alkyne metathesis at an early stage of the reactions development [4]. A metal carbyne-promoted metathesis of alkynes proceeds through a sequence of formal [2+2]-cycloaddition and cycloreversion steps via metallacyclobutadiene intermediates (Scheme 7.2). Experimental evidence for this proposal was provided by Schrock’s use of high-valent metal alkylidynes: with isolated metallacyclobutadiene intermediates acting as catalytically competent species for alkyne metathesis [5]. More recently, theoretical investigation of tungsten and molybdenum alkylidynes as catalysts for alkyne metathesis has been performed to shed light onto orders of relative reactivity [6]. These intertwined advances in mechanistic understanding and catalyst development have progressed the field to the stage that it is now applicable to molecules of significant complexity and fragility.

R2

M

R2

+ R1

R2

M R1

Scheme 7.2

R1

M R1

R1

R1

The mechanism of alkyne metathesis.

R2 M R1

+ R1

7.3 Ring-closing Alkyne Metathesis


RCAM is the reaction of an acyclic diyne to afford a cyclic alkyne and an acyclic alkyne side product (Scheme 7.3). As with ring-closing (alkene) metathesis (RCM), the reaction is attractive in synthetic terms, as it requires relatively simple and mild reaction conditions and precludes a need to prepare and use highly reactive functional groups. The first examples of catalytic RCAM for the synthesis of functionalized macrocycles were reported by Fürstner and Seidel [7]. Two catalyst systems were employed in this study: the ‘‘instant’’ system prepared by reaction of catalytic amounts of Mo(CO)6 1 with stoichiometric quantities of phenols, and the well-defined tungsten–alkylidyne 2 first prepared by Schrock in 1981 [5, 8]. Subsequent applications of RCAM, particularly in natural product synthesis, have either utilized these two systems or a molybdenum system developed by Fürstner that employs in situ reaction of trisamido molybdenum complex 3 [9] and CH2 Cl2 (Figure 7.1) [10]. The choice of a catalyst system for any particular reaction is generally dictated by substrate compatibility. Although not yet at the same level of applicability as the catalysts available for alkene metathesis, these catalyst systems show some excellent functional group tolerance [3d]. The ‘‘instant’’ system represents a very practical and easily applicable method as it employs cheap, commercially available, and stable reagents without the need to use rigorously purified solvents or an inert atmosphere. Since its original discovery, modifications have led to improved functional group tolerance and a reduced reaction temperature (∼140 ◦ C) [3b, d]. R

R RCAM

+

R = Me, Et

R

R

Scheme 7.3

Ring-closing alkyne metathesis (RCAM).

t- Bu

t- Bu CO OC

W CO

O

O

Mo

OC

N

O Ar

CO CO 1

Figure 7.1

Mo

N N

t- Bu Ar

Ar 2

3: Ar = 3,5-dimethylphenyl

Key metal complexes employed in RCAM.

207

208

7 Ring-closing Alkyne Metathesis in Natural Product Synthesis MeO OR

MeO OR

OR OR

OR

1 (cat), F3CC6H4OH Chlorobenzene

OR

135 °C, 6 h, 76% or Microwave 150 °C, 5 min, 71%

4: R = PMB

Scheme 7.4

5

Microwave heating in RCAM.

However, this process is generally less applicable in comparison to the other available systems, as the reaction conditions remain relatively harsh and the catalyst displays lower activity. In cases where a substrate is suitably robust, the ‘‘instant’’ Mortreux-type system of 1 and phenols can be ideal. A modified Mortreux-type reaction system was successfully employed in the synthesis of turrianes, which are naturally occurring cyclophane derivatives. The RCAM approach can also be combined with modern synthetic techniques; in the cyclization of 4 the use of microwave heating resulted in a reduction of the reaction time from 6 hours under conventional heating to 5 minutes (Scheme 7.4) [11]. The commercially available complex 2 shows higher activity and is metathesis active under mild conditions, typically ambient temperature up to 90 ◦ C [3c, h]. RCAM reactions with 2 are often run at 80 ◦ C in toluene or a related aromatic solvent. However, rigorously inert (anhydrous and oxygen-free) conditions are required. The core structure is amenable to tuning by ligand modifications, reflected in a recent report which demonstrated that the effective RCAM of diyne 6 can be performed at room temperature to afford cyclic alkyne 8 in excellent yield when imidazolin-2-iminato tungsten alkylidyne catalyst 7 was employed (Scheme 7.5) [12]. The same catalyst has also been employed in the preparation of cyclophanes [13]. The other commonly used system employs molybdenum amide 3 as a precatalyst [5c, 14]. In the presence of CH2 Cl2 , 3 reacts to form a mixture of molybdenum chloride 9 and terminal alkylidyne 10 (Scheme 7.5) [10, 14]. Interestingly, 9 is the catalytically competent component at 80 ◦ C, whereas the metal alkylidyne 10 effected only one turnover in alkyne metathesis. Alternatively, the use of a higher order alkyl chloride gives rise to catalytically competent nonterminal alkylidyne 11 (Scheme 7.6). This approach can be coupled with a ‘‘reductive-recycle’’ strategy to convert chloride 9 back to 3 [15, 16]. As with the tungsten analogues, molybdenum alkylidynes of type 11 can be further modified by ligand exchange to effect their properties and reactivity; the use of alcohols affords trisalkoxy molybdenum alkylidynes [15–17]. To date, applications of this system in natural product synthesis have employed the in situ generation of the catalytically active species. Molybdenum complex 3 is


t-Bu N

N N

W

O O

CF3 CF3

CF t-Bu CF3 3

7 (2 mol%)

O

O

O

O

Hexane, rt 95%

6 Scheme 7.5

209

8

Use of imidazolin-2-iminato-substituted tungsten alkylidyne 7 in RCAM.

R

t -Bu

t - Bu N Ar

Mo

N t - Bu N Ar Ar

3: Ar = 3,5-dimethylphenyl

Cl RCHCl2

t -Bu

t -Bu

Mo N

Ar 9

N t -Bu N Ar Ar

+

t -Bu

t -Bu

Mo N

Ar

N t -Bu N Ar Ar

10: R = H 11: R = Et

Scheme 7.6 Activation of the molybdenum trisamido complex 3 to form metathesis-active complexes.

typically reacted with the diyne in a toluene/CH2 Cl2 mixture at ∼80 ◦ C under inert atmosphere. Taken alongside comparative studies [3d, 7], the applications discussed below show that the catalyst formed from 3/CH2 Cl2 displays the widest general applicability and functional group tolerance of the three most used systems. Functionalities such as esters, isolated double bonds, silyl ethers, sulfones, aldehydes, nitro groups, ketones, alkyl chlorides, acetals, and nitriles can be present in reactions using 3/CH2 Cl2 [3d]. ‘‘Acidic’’ protons, such as those of secondary amides, are incompatible with 3, although tertiary amides are fully compatible. While the ‘‘instant’’ system is effective for robust substrates, the reaction conditions are too harsh for more labile units. Tungsten alkylidyne complex 2 tolerates units that include acetal, alkene, amide, carbonyl, enoate, ester, ether, furan, ketone, silyl ether, sulfonamide, and sulfone and is compatible with secondary amides. The high Lewis acidity of the tungsten center in 2 does, however, render substrates containing thioether, basic nitrogen groups, and polyethers incompatible [3d]. The decreased effective Lewis acidity and hence functional group tolerance of 3 is proposed to result from the close-packing amido ligands efficiently shielding the central molybdenum in

210


the catalytically active template and preventing coordination of potential donor substrates [14]. 7.3.1 RCAM as a Synthetic Strategy

RCAM is intrinsically less applicable than the equivalent RCM strategy, as the cyclic products of RCAM must be of sufficient size to be able to accommodate a linear alkyne unit. In such cases, however, the RCAM approach shares many of the positive characteristics of RCM, and may indeed be more suitable in certain situations. RCAM requires a reaction system of substrate, catalyst, and solvent only; it converts relatively straightforward and handleable functionality to make new carbon–carbon bonds; and it shows good functional group tolerance. Importantly, and in contrast to alkene metathesis, RCAM shows excellent chemoselectivity toward other carbon–carbon π-systems such as alkenes. A limitation of current RCAM methods is that end-capped alkynes must be used to achieve efficient catalysis. Terminal alkynes are incompatible with the catalyst systems in use [7, 8, 18]. This proviso can impact negatively on a synthetic strategy, as an additional step may be required to cap terminal alkynes. Furthermore, the RCAM of capped alkynes leads to less volatile side products (but-2-yne or hex-3-yne) which may be more difficult to purge from the system, particularly under the milder conditions associated with the more reactive catalysts. The reactions may therefore be performed under reduced pressure to aid the removal of side products. This helps drive the desired metathesis toward completion and reduces the possibility of a catalyst deactivation route known to occur with the polymerization of but-2-yne by ring expansion in the presence of Mo(VI) and W(VI) complexes [3a, 15]. Despite this current complication, the RCAM strategy provides a useful macrocyclization technique, which can also be considered as a solution to the vexing issue of stereoselectivity in ring-closing macrocyclization metathesis. When the stereochemistry of the macrocyclic alkene is important, the otherwise superb RCM strategy can become impractical because of the absence of any predictable control over alkene geometry. This problem has led to the formation of the undesired alkene isomer in significant quantities and even as the major product in several advanced syntheses [19]. The use of RCAM provides a route around this issue, yet employs a similar retrosynthetic logic to RCM, as stereoselective partial reduction of alkynes may be performed in a predictable fashion (Scheme 7.7). The potential of this strategy was recognized at the outset of RCAM development. Indeed, the combination of RCAM and Lindlar reduction to access (Z)-alkenes is the most widely employed alkyne metathesis strategy within natural product synthesis [3d, 7]. The combination of RCAM and semi-reduction to an (E)-alkene has been less widely applied to date, possibly because the semi-reduction strategy itself is less well established. However, a useful stereoselective semi-reduction method has been established, which employs a ruthenium-catalyzed trans-hydrosilylation followed by a protodesilylation [20–22].

7.4 Applications of RCAM in Natural Product Synthesis

H2, Lindlar Pd

(EtO)3SiH [Cp*Ru(MeCN)3]PF6

AgF Si(OEt)3

Scheme 7.7

THF/MeOH

Post RCAM entry to stereodefined macrocyclic alkenes.


The RCAM strategy has been employed toward a variety of biologically relevant synthetic targets including materials such as diaminosuberic acid derivatives [23], oligopeptides [16, 17], glycophanes [24], cyclic β-turn [25], and lantibiotic nisin Z mimics [26], and alkyne-linked glycoamino acids [27]. For the remainder of this chapter, we will only explore natural product applications with a focus on the RCAM macrocyclization step and the manipulation of the resulting alkyne. 7.4.1 RCAM/Hydrogenation Strategies 7.4.1.1 Macrocyclic Musks The macrocyclic musks and fragrances, yuzu lactone (12) [28], ambrettolide (13) [28], and civetone (14) [29], which all contain a (Z)-configured alkene, were chosen as early targets for application of an RCAM/Lindlar reduction strategy (Figure 7.2). In all cases, application of the Mortreux ‘‘instant’’ system was successful, and the desired (Z)-alkene isomers were obtained exclusively from hydrogenation [28, 29]. Good overall yields for the formation of cyclic (Z)-alkene 14 from 15 were achieved in the presence of an unprotected carbonyl functionality using either tungsten complex 2 or the user-friendly and low-tech ‘‘instant’’ RCAM catalyst system (Scheme 7.8). The insect repellent alkaloids epilachnene 19 and homologues as well as the cytotoxic sponge extract motuporamine C were prepared in an analogous fashion by an RCAM/Lindlar hydrogenation approach [28, 30]. All of these early examples demonstrate the complementary nature of the RCAM/semi-reduction manifold

211

212


O O

O

O O

Yuzu lactone (12)

Ambrettolide (13)

Civetone (14)

Figure 7.2 Macrocyclic musks, which were early natural product targets for the RCAM/Lindlar hydrogenation synthetic strategy.

O O 2 (Cat) Toluene, 80 °C (65%) or 1 (5 mol%), F3CC6H4OH Chlorobenzene, 140 °C (59%) 15 Scheme 7.8

H2, Lindlar Pd Quinoline CH2Cl2 94%

14

16 The RCAM/Lindlar hydrogenation manifold in the synthesis of civetone.

to an RCM strategy that often leads to a mixture of alkene geometries [31]. For instance, in the synthesis of epilachnene, high yielding formation of the macrocyclic (Z)-alkene 19 in a fully stereoselective manner is achieved by an RCAM/semi-reduction [28]. This contrasts with the RCM approach which gives the undesired (E)-isomer 21 as the major constituent when starting from diene 20 (Scheme 7.9) [31a, 32]. The same key processes to install macrocycles with a (Z)-alkene were recently employed in the preparation of some novel designed musk compounds [33]. 7.4.1.2 Prostaglandin Lactones While the ‘‘instant’’ system was suitable for the previous examples, it was found to be incompatible with more elaborate applications. For example, no product was obtained from the cyclization of diyne 22. In contrast, the 3/CH2 Cl2 catalyst system was found to be highly effective for the cyclization of 22 to 23 (Scheme 7.10). In this case, a slightly higher yield of 77% in the RCAM step could be achieved by modifying the aromatic unit on the amido ligand of 3 to 4-fluorophenyl [34]. As a comparison, the use of 2 required longer reaction times and resulted in lower conversion of material. Relative efficacies of the catalyst systems were compared in the cyclization of the C15 epimer of 22 and showed that for substrates of this complexity, the molybdenum amide systems 3 (81–87% in RCAM) were more effective than the tungsten alkylidyne 2 (65%) [34].


O 2 (5 mol%) (71%)

O NFmoc

17

1. H2, Lindlar Pd quinoline, CH2Cl2 (94%)

or 1 (5 mol%) p -ClC6H4OH (1 equiv.) (67%)

O

O NFmoc

18

O O

NH

2. TBAF·3H2O (62%) Epilachnene (19)

O

O [Ru]-I (5 mol%)

O NFmoc

O

89%

20

NFmoc

21 (Z /E = 1:2)

Scheme 7.9 A comparison of the RCAM/Lindlar hydrogenation manifold and the RCM approach in the synthesis of epilachnene.

7.4.1.3 Sophorolipid Lactone The synthesis of sophorolipid lactone 28 demonstrated the low Lewis acidity of the metal center in 3 and the resulting catalytically active species. A high yield of 26 was achieved in RCAM of diyne 25 despite the presence of acid-labile PMB ethers and glycidic linkages (Scheme 7.11) [35]. Once again, stereoselective semi-reduction resulted in quantitative yield of the required (Z)-olefin 27. 7.4.1.4 Epothilone A The potent biological activity of the epothilones rendered them hot targets for synthesis teams and it is significant that the first three syntheses of epothilone A (33) used RCM to form the macrocycle. This served to highlight the confidence of the synthetic community, with what was then a relatively new tool. However, despite the effective cyclizations, the absence of any significant selectivity for the desired (Z)-olefin geometry represented a major issue at a late stage of the synthesis [19a–c]. The epothilones serve as a similarly enlightening example of the potential applicability of a RCAM/Lindlar reduction; selective access into the (Z)-alkene 31

213

214


O

O 3 (7.5 mol%)

O

O TBSO

Toluene/ CH2Cl2, 80 °C 68–73%

O

O

TBSO

22

23 O

1. H2, Lindlar Pd (86%)

O O

2. HF, MeCN (88%)

HO

PGE2-1,15-lactone (24)

Scheme 7.10

Prostaglandin lactone synthesis. OR

RO O

O

RO RO

O

O

O RO

OR

O RO 3 (10 mol%)

OR

Toluene/CH2Cl2, 80 °C 78%

OR

25: R = PMB

26

OR RO H2, Lindlar Pd Quant.

O

O

RO RO

O RO

O

O OR

OH O

HO DDQ 93%

27

Scheme 7.11

O

O

O

O RO

O

O

RO RO

Synthesis of sophorose lipid lactone.

O

O

HO HO

O HO

O

O

O

OH

Sophorose lipid lactone (28)


S

S 3 (10 mol%)

OR

N

N

OR

Toluene/CH2Cl2 80%

O O

O O

OR O

29: R = TBS

H2, Lindlar Pd

O S OR

N

OH

N

O

O O

OR O

31 Epothilone C (32) Scheme 7.12

OR O 30

S

CH2Cl2

215

O aq HF 70% over two steps

OH O

Epothilone A (33)

RCAM-based synthesis of epothilones.

was achieved in high yield (Scheme 7.12). Aside from the issue of alkene geometry, this synthesis highlighted the capability of RCAM to be applied to molecules of similar complexity to those RCM is used for. Labile functionalities such as the aldol substructure were left intact, no epimerization of the stereogenic center α to the carbonyl was seen, and the reaction proceeded even in the presence of the basic nitrogen and sulfur atoms of the thiazole [36]. 7.4.1.5 Cruentaren A Complex RCAM processes are mediated by both 2 and 3. The macrocyclic salicylate lactone cruentaren A (36) has been prepared independently by two groups, Maier [37, 38] and Fürstner [39], using a RCAM approach. Their chosen cyclization precursors, 34a and 34b, differ only by the choice of a particular silyl protecting group and in the extent of functionalization on the side chain (Scheme 7.13). Both RCAM processes could be performed in high yield. Complex 3/CH2 Cl2 had to be used for the cyclization of 34b, as its acid-labile OTHP functionality was incompatible with complex 2. It is interesting to note that in the subsequent transformations to the target molecule 36, the Lindlar reduction was carried out immediately after the RCAM by Fürstner and coworkers, whereas it was applied at the final step of the synthesis by Vintonyak and Maier, thus demonstrating some flexibility in the RCAM/semi-reduction approach. The success of the key RCAM strategy and endgame led to its retention in a second-generation synthesis by Maier et al. [40].

216


R

R

TBSO

TBSO

OMe O

OMe O Conditions

O MeO

O MeO

1

R O

R1O 34

35

34a: R = CH2ODMB, R1 = TIPS 34b: R =

OTHP

2 (10 mol%) Toluene, 80 °C (91%)

, R1 = TIPS

3 (10 mol%) Toluene/CH2Cl2, 80 °C (87%) HO O NH HO

Steps

OH

O O

MeO HO Cruentaren A (36) Scheme 7.13

The synthesis of cruentaren A by RCAM/Lindlar reduction strategies.

7.4.1.6 Latrunculins A, B, C, M, and S A major application of RCAM is seen in the synthesis of a range of actin binding macrolides by Fürstner et al. First employed in the synthesis of latrunculin B (39) [41], the RCAM/Lindlar reduction manifold was subsequently utilized to prepare latrunculin C, M, and other analogs [42]. A mixed ketal, a thioazolidinone unit, and a branching α-substituent to one of the alkyne units of diyne 37 proved no obstacle to successful RCAM with either precatalyst 3 or the tungsten complex 2, with the former giving slightly higher yield of 38 (Scheme 7.14). The latrunculins A (42) and S, however, represented a new challenge for application of RCAM. The presence of a conjugated diene in the target compounds required the first use of a chemoselective enyne–yne metathesis in natural product synthesis [42, 43]. A model study employing tungsten complex 2 had shown that the metathesis of simple enyne–yne systems was viable [21b]. However, latrunculin A (42) contains a cyclic structure which is smaller that those previously prepared by RCAM (a


O

O H

O

O O

O

3 (5 mol%) Toluene/CH2Cl2, 80 °C 70%

OMe

PMBN

O H

OMe

1. H2, Lindlar Pd CH2Cl2 (quant) 2. CAN, MeCN/H2O (78%)

O H

37

O 38

Latrunculin B (39)

Synthesis of latrunculin B via RCAM.

O

O

O

O 3 (10 mol%)

O H

OMe

TeocN

O H


OMe

TeocN S

S O

O 40

41

O 1. H2, Lindlar Pd Quinoline, CH2Cl2 (82%) 2. TBAF, THF (62%) 3. aq HOAc, 60 °C (80%)

O O H

OH S

S O

Scheme 7.14

O

HN

PMBN S

O

217

OH

HN S O Latrunculin A (42)

Scheme 7.15 An RCAM/Lindlar hydrogenation approach to latrunculin A involving enyne–yne coupling.

16-membered ring). Despite this issue and the increased complexity of enyne–yne 40, no problems were observed in the effective RCAM of the conjugated alkyne, demonstrating the orthogonality of the reaction system derived from 3/CH2 Cl2 toward alkenes (Scheme 7.15). Although the PMB functionality is compatible with RCAM, a late-stage protecting group exchange on the thiazolidinone to Teoc had to

218


be performed before RCAM as a consequence of the problematic deprotection of the PMB group after RCAM. Finally, the required conjugated (Z)-olefin of 42 was obtained by Lindlar hydrogenation of 41 [43]. 7.4.1.7 Myxovirescin A1 The 28-membered ring macrocycle of the antibiotic macrolide myxovirescin A1 was also assembled using an enyne–yne RCAM/semi-reduction approach. However, the presence of a conjugated (E)-olefin in the target compound required an alternative manipulation of the cyclic alkyne to the Lindlar reduction approach that had previously been employed to access conjugated (Z)-alkenes. Although enyne–yne 45 features a methyl ether close to the reaction site in a potentially ligating position, the RCAM step proceeded smoothly with the 3/CH2 Cl2 system (Scheme 7.16). In this case, the tungsten alkylidyne 2 was less effective than the 3/CH2 Cl2 catalyst system. However, the presence of a secondary amide linkage meant that a high catalyst loading was required to achieve a high yield of the cyclic alkyne 46 [44]. The following step required the semi-reduction of the cyclic enyne to a conjugated (E)-olefin, which is much less well established than the proven technology for stereospecific semi-reduction to (Z)-olefins. Although model studies on the hydrosilylation of cyclic enyne system had given the desired outcome with very high stereoselectivity [21b], the reaction of 46 was poorly stereoselective. Indeed, three isomeric vinyl silanes 47, 48a, and 48b were obtained even when using the less sterically imposing [CpRu(MeCN)3 ]PF6 catalyst in place of the more commonly applied [Cp* Ru(MeCN)3 ]PF6 . The formation of the (Z,Z)-configured isomer 50 along with the desired (E,Z)-configured 49 major product, after protodesilylation of the vinyl silanes 47, 48a, and 48b, demonstrates that the hydrosilylation methodology requires further improvement for such highly complex substrates. 7.4.2 RCAM and Alternative Alkyne Manipulations

While macrocylic structures containing (Z)-olefins would now appear to suggest an obvious site for an RCAM-based disconnection strategy, the alkyne can also be employed for other means. Indeed, the use of alkynes as reactive units is growing rapidly [45] and application of such processes alongside an RCAM strategy may well be a fruitful approach in the future. The syntheses of two natural products in which the alkyne introduced by RCAM is employed in reactions other than semi-reduction of the alkyne are shown below. 7.4.2.1 Citreofuran In the synthesis of the polyketide metabolite citreofuran (57), RCAM was incorporated into the retrosynthetic analysis by setting an alkyne–one as the precursor for the furan ring. The diyne precursor 55 was rapidly assembled from acid 51 by its reaction with chloroenamine 52 to yield the alkoxyisocoumarin 53 upon treatment with Et3 N. The addition of homopropargylic Grignard 54 to unstable 53 resulted in


H2N

O

O

HO

219

O OH

HN O

TBDPSO

MeO

MeO

OH O O 43

1. HOBt, EDC·HCl i- Pr2NEt, CH2Cl2 2. TBAF, THF

O

O O

HOBt, EDC·HCl, Et3N, DMAP 65 – 75%

44

O

O O

O HN MeO

O

O

O

O 3 (40 – 50 mol% in 3 portions)

HN MeO


O O

O 46

45 O O

[CpRu(MeCN)3]PF6 (30 mol%) (EtO)3SiH O Toluene

O O HN

H

HN

+ O

OMe

O

O

SiR3

OMe

O

O

O

Si(OEt)3

O

H

48a/48b: R = OEt (53%)

47 (13%)

AgF aq THF/ MeOH

AgF aq THF/MeOH 94%

25%

23%

O

O O

O

HN

HN

O

O

O

MeO

O

MeO O

O O

O 49

Scheme 7.16 The enyne–yne RCAM/hydrosilylation/ protodesilylation approach to myxovirescin A1 .

50

220


the desired RCAM precursor diyne 55 when TMSCl was used as a mild activating agent [46]. RCAM of 55 using complex 2 then worked well despite the presence of an unprotected ketone. It was specifically noted that the precursors must be of high purity to ensure good conversion. In particular, traces of amine can act as catalyst poison. Once the macrocycle 56 was formed, an acid-catalyzed cycloisomerization between the ketone and the alkyne afforded the furan in good yield with the natural product 57 reached after demethylation of the bisphenol (Scheme 7.17) [46].

O

Cl

O 1.

MeO

NMe2

52

MeO

O O

2. Et3N

OH

51

54

TMSCl, THF 75%

OMe O

OMe O

O

BrMg

53

O

O

MeO

2 (cat) O

O

MeO

1. p -TsOH (85%)

78 – 81%

O

OMe

O HO

2. B -I-9-BBN (60%)

O

OMe 55

TBSO

OH 56

Scheme 7.17

Citreofuran (57)

An RCAM/cycloisomerization approach to citreofuran.

H O H

O OTBS O

O

3 (20 mol%)

OTBS


O O

58

[Ru]-II (2 mol%) H2C CH2 (1.8 atm) Toluene, 45 °C 90%

OTBS

59

H O H

H

O H

Steps OH

OTBS O O 60

Scheme 7.18

O

OTBS

O O Amphidinolide V (61)

The synthesis of amphidinolide V using complementary metathesis methods.

References

7.4.2.2 Amphidinolide V An alternative use of an alkyne installed by RCAM was demonstrated in the synthesis of the proposed structure of amphidinolide V when a complementary ene–yne metathesis reaction was employed (Scheme 7.18) [47]. RCAM of diyne 58 using 3/CH2 Cl2 proceeded smoothly despite the labile trans-configured vinyl epoxide, the allylic alcohol, and the apparent ring strain of the product. The resulting cyclic alkyne 59 was then subjected to a ruthenium-catalyzed ene–yne metathesis under ethylene atmosphere with [Ru]-II. This step installed the two exo-methylene units required in the natural product (Scheme 7.18). On completion of the synthesis, the NMR data of 61 corresponded to those of the isolated natural product, apart from a discrepancy with a singular resonance. Importantly, the RCAM/ene–yne metathesis synthetic approach was flexible enough to allow this issue to be explored through the preparation of several isomers of 61. All of these isomers matched the reported data less closely than the originally prepared isomer [47].

7.5 Conclusions

The examples shown throughout this chapter demonstrate that RCAM is a powerful methodology for the preparation of complex natural compounds. The technique is both robust and able to mediate transformations of complex structures bearing labile functionalities. A significant use of the RCAM involves its combination with a subsequent stereoselective semi-reduction of the alkyne. This approach provides a predictable alternative to the use of RCM in macrocyclizations, and has been verified by successful application in a range of natural products. The strategy accommodates a range of differing functionalities with a level of complexity closely comparable to that associated with RCM. Further advances in catalyst development are likely to improve the uptake and applicability of the RCAM strategy. In particular, the design of a catalyst system that mediates effective RCAM of terminal alkynes under mild conditions with good functional group tolerance is extremely desirable and would have great effect on the synthetic impact of this already highly employable reaction.

References 1 Pennella, F., Banks, R.L., and

Bailey, G.C. (1968) Chem. Commun., 1548–1549. 2 (a) Mortreux, A. and Blanchard, M. (1974) J. Chem. Soc., Chem. Commun., 786–787; (b) Mortreux, A., Dy, N., and Blanchard, M. (1975/76) J. Mol. Catal., 1, 101–109; (c) Mortreux, A.,

Petit, F., and Blanchard, M. (1978) Tetrahedron Lett., 19, 4967–4968; (d) Bencheick, A., Petit, M., Mortreux, A., and Petit, F. (1982) J. Mol. Catal., 15, 93–101; (e) Mortreux, A., Delgrange, J.C., Blanchard, M., and Lubochinsky, B. (1977) J. Mol. Catal., 2, 73–82; (f ) Mortreux, A., Petit, F., and

221

222


3

4 5

6 7 8

9

Blanchard, M. (1980) J. Mol. Catal., 8, 97–106. (a) Zhang, W. and Moore, J.S. (2007) Adv. Synth. Catal., 349, 93–120; (b) Mortreux, A. and Coutelier, O. (2006) J. Mol. Catal. A: Chem., 254, 96; (c) Schrock, R.R. and Czekelius, C. (2007) Adv. Synth. Catal., 349, 55–77; (d) Fürstner, A. and Davies, P.W. (2005) Chem. Commun., 2307–2320; (e) Fürstner, A. (2003) in Handbook of Metathesis, vol. 2 (ed. R.H. Grubbs), Wiley-VCH Verlag GmbH, Weinheim, pp. 432–462; (f ) Schrock, R.R. (2003) in Handbook of Metathesis, vol. 1 (ed. R.H. Grubbs), Wiley-VCH Verlag GmbH, Weinheim, pp. 173–189; (g) Gradillas, A., Pérez-Castells, J. (2006) Angew. Chem. Int. Ed., 37, 6086–6101; (h) Schrock, R.R. (2002) Chem. Rev., 102, 145–180. Katz, T.J. and McGinnis, J. (1975) J. Am. Chem. Soc., 97, 1592–1594. (a) Wengrovius, J.H., Sancho, J., and Schrock, R.R. (1981) J. Am. Chem. Soc., 103, 3932–3934; (b) Pedersen, S.F., Schrock, R.R., Churchill, M.R., and Wasserman, H.J. (1982) J. Am. Chem. Soc., 104, 6808–6809. Zhu, J., Jiu, G., and Lin, Z. (2006) Organometallics, 25, 1812–1819. Fürstner, A. and Seidel, G. (1998) Angew. Chem. Int. Ed., 37, 1734–1736. (a) Schrock, R.R., Clark, D.N., Sancho, J., Wengrovius, J.H., Rocklage, S.M., and Pedersen, S.F. (1982) Organometallics, 1, 1645–1651; (b) Freudenberger, J.H., Schrock, R.R., Churchill, M.R., Rheingold, A.L., and Ziller, J.W. (1984) Organometallics, 3, 1563–1573; (c) Listemann, M.L. and Schrock, R.R. (1985) Organometallics, 4, 74–83; (d) Fischer, H., Hofmann, P., Kreissl, F.R., Schrock, R.R., Schubert, U., and Weiss, K. (eds) (1988) Carbyne Complexes, Wiley-VCH Verlag GmbH, Weinheim; (e) Schrock, R.R. (1986) Acc. Chem. Res., 19, 342–348; (f ) Schrock, R.R. (1995) Polyhedron, 14, 3177–3195. (a) Laplaza, C.E., Odom, A.L., Davis, W.M., Cummins, C.C., and Protasiewicz, J.D. (1995) J. Am. Chem. Soc., 117, 4999–5000; (b) Laplaza, C.E.

10

11

12

13

14

15 16 17

18

19

20

and Cummins, C.C. (1995) Science, 268, 861–863; (c) Cummins, C.C. (1998) Chem. Commun., 1777–1786. Fürstner, A., Mathes, C., and Lehmann, C.W. (1999) J. Am. Chem. Soc., 121, 9453–9454. Fürstner, A., Stelzer, F., Rumbo, A., and Krause, H. (2002) Chem. Eur. J., 8, 1856–1871. Beer, S., Hrib, C.G., Jones, P.G., Brandhorst, K., Grunenberg, J., and Tamm, M. (2007) Angew. Chem. Int. Ed., 46, 8890–8894. Beer, S., Brandhorst, K., Grunenberg, J., Hrib, C.G., Jones, P.G., and Tamm, M. (2008) Org. Lett., 10, 981–984. Fürstner, A., Mathes, C., and Lehmann, C.W. (2001) Chem. Eur. J., 7, 5299–5317. Zhang, W., Kraft, S., and Moore, J.S. (2004) J. Am. Chem. Soc., 126, 329–335. Zhang, W., Kraft, S., and Moore, J.S. (2003) Chem. Commun., 832–833. (a) McCullough, L.G. and Schrock, R.R. (1984) J. Am. Chem. Soc., 106, 4067–4068; (b) McCullough, L.G., Schrock, R.R., Dewan, J.C., and Murdzek, J.C. (1985) J. Am. Chem. Soc., 107, 5987–5998; (c) Tsai, Y.-C., Diaconescu, P.L., and Cummins, C.C. (2000) Organometallics, 19, 5260–5262. For a recent report on the use of terminal alkynes in alkyne metathesis see Coutelier, O., Nowogrocki, G., Paul, J.-F., and Mortreux, A. (2007) Adv. Synth. Catal., 349, 2259–2263. For some examples (a) Bertinato, P., Sorensen, E.J., Meng, D., and Danishefsky, S. (1996) J. Org. Chem., 61, 8000–8001; (b) Nicolaou, K.C., He, Y., Vourloumis, D., Vallberg, H., Roschangar, F., Sarabia, F., Ninkovic, S., Yang, Z., and Trujillo, J.I. (1997) J. Am. Chem. Soc., 119, 7960–7973; (c) Schinzer, D., Limberg, A., Bauer, A., Böhm, O.M., and Cordes, M. (1997) Angew. Chem. Int. Ed. Engl., 36, 523–524; (d) Ono, K., Nakagawa, M., and Nishida, A. (2004) Angew. Chem. Int. Ed., 43, 2020–2023. Trost, B.M., Ball, Z.T., and Jöge, T. (2002) J. Am. Chem. Soc., 124, 7922–7923.

References 21 F¨ urstner, A. and Radkowski, K. (2002)

22 23

24 25

26

27

28

29 30 31

32 33 34

Chem. Commun., 2182–2183; (b) Lacombe, F., Radkowski, K., Seidel, G., and Fürstner, A. (2004) Tetrahedron, 60, 7315–7324. Trost, B.M. and Ball, Z.T. (2001) J. Am. Chem. Soc., 123, 12726–12727. Aguilera, B., Wolf, L.B., Nieczypor, P., Rutjes, F., Overkleeft, H.S., van Hest, J.C.M., Schoemaker, H.E., Wang, B., Mol, J.C., Fürstner, A., Overhand, M., van der Marel, G.A., and van Boom, J.H. (2001) J. Org. Chem., 66, 3584–3589. Doyle, D. and Murphy, P.V. (2008) Carbohydr. Res., 343, 2535–2544. IJsselstijn, M., Aguilera, B., van der Marel, G.A., van Boom, J.H., van Delft, F.L., Schoemaker, H.E., Overkleeft, H.S., Rutjes, F.P.J.T., and Overhand, M. (2004) Tetrahedron Lett., 45, 4379–4382. (a) Ghalit, N., Poot, A.J., Fürstner, A., Rijkers, D.T.S., and Liskamp, R.M.J. (2005) Org. Lett., 7, 2961–2964; (b) Ghalit, N., Rijkers, D.T.S., and Liskamp, R.M.J. (2006) J. Mol. Catal. A, 254, 68–77. Groothuys, S., van den Broek, S.A.M.W., Juijpers, B.H.M., IJsselstijn, M., van Delft, F.L., and Rutjes, F.P.J.T. (2008) Synlett, 111–115. Fürstner, A., Guth, O., Rumbo, A., and Seidel, G. (1999) J. Am. Chem. Soc., 121, 11108–11113. Fürstner, A. and Seidel, G. (2000) J. Organomet. Chem., 606, 75–78. Fürstner, A. and Rumbo, A. (2000) J. Org. Chem., 65, 2608–2611. (a) Fürstner, A. and Langemann, K. (1997) Synthesis, 792–803; (b) Fürstner, A. and Langemann, K. (1996) J. Org. Chem., 61, 3942–3943; (c) Fürstner, A., Koch, D., Langemann, K., Leitner, W., and Six, C. (1997) Angew. Chem. Int. Ed. Engl., 36, 2466–2469. Goldring, W.P.D. and Weiler, L. (1999) Org. Lett., 1, 1471–1473. Kraft, P. and Berthold, C. (2008) Synthesis, 543–550. (a) Fürstner, A. and Grela, K. (2000) Angew. Chem. Int. Ed., 39, 1234–1236; (b) Fürstner, A., Grela, K., Mathes, C., and Lehmann, C.W. (2000) J. Am. Chem. Soc., 122, 11799–11805.

35 (a) F¨ urstner, A., Radkowski, K.,

36 37

38

39

40 41

42

43 44

45

46

47

Grabowski, J., Wirtz, C., and Mynott, R. (2000) J. Org. Chem., 65, 8758–8762; (b) Fürstner, A. (2004) Eur. J. Org. Chem., 943–958. Fürstner, A., Mathes, C., and Grela, K. (2001) Chem. Commun., 1057–1059. (a) Vintonyak, V.V. and Maier, M.E. (2007) Angew. Chem. Int. Ed., 46, 5209–5211; (b) Vintonyak, V.V. and Maier, M.E. (2007) Org. Lett., 9, 655–658. Vintonyak, V.V., Calà, M., Lay, F., Kunze, B., Sasse, F., and Maier, M.E. (2008) Chem. Eur. J., 14, 3709–3720. Fürstner, A., Bindl, M., and Jean, L. (2007) Angew. Chem. Int. Ed., 46, 9275–9278. Maier, M.E. and Ritschel, J. (2008) Arkivoc, 314–329. Fürstner, A., De Souza, D., Parra-Rapado, L., and Jensen, J.T. (2003) Angew. Chem. Int. Ed., 42, 5358–5360. (a) Fürstner, A., Kirk, D., Fenster, M.D.B., Aissa, C., and De Souza, D. (2005) Proc. Natl. Acad. Sci. U.S.A., 102, 8103–8108; (b) Fürstner, A., De Souza, D., Turet, L., Fenster, M.D.B., Parra-Rapado, L., Wirtz, C., Mynott, R., and Lehmann, C.W. (2007) Chem. Eur. J., 13, 115–134; (c) Fürstner, A., Kirk, D., Fenster, M.D.B., Aissa, C., De Souza, D., Nevado, C., Tuttle, T., Thiel, W., and Müller, O. (2007) Chem. Eur. J., 13, 135–149. Fürstner, A. and Turet, L. (2005) Angew. Chem. Int. Ed., 44, 3462–3466. Fürstner, A., Bonnekessel, M., Blank, J.T., Radkowski, K., Seidel, G., Lacombe, F., Gabor, B., and Mynott, R. (2007) Chem. Eur. J., 13, 8762–8783. For an example, see the recent advances in platinum and gold activation of alkynes: Fürstner, A. and Davies, P.W. (2007) Angew. Chem. Int. Ed., 46, 3410–3449. Fürstner, A., Castanet, A.S., Radkowski, K., and Lehmann, C.W. (2003) J. Org. Chem., 68, 1521–1528. Fürstner, A., Larionov, O., and Flügge, S. (2007) Angew. Chem. Int. Ed., 46, 5545–5548.

223

225

8 Temporary Silicon–Tethered Ring–Closing Metathesis Reactions in Natural Product Synthesis P. Andrew Evans

8.1 Introduction

Temporary tethered reactions provide an important strategy for target-directed synthesis, since they generally circumvent the problems encountered with entropically unfavorable reactions (Scheme 8.1) [1]. For example, although Lewis-acid- and transition metal-catalyzed reactions have advanced considerably in the last few decades, the intermolecular reaction manifold is often considerably more challenging than the intramolecular version [2]. Nishiyama and Stork independently reported the concept of utilizing a temporary or disposal tether in the context of a free radical addition [3], which has resulted in the application of this strategy to an array of challenging synthetic transformations [4]. Interestingly, the temporary tether has been compared to the formation of an enzyme–substrate complex in enzyme catalysis [5], since it provides a high effective concentration of the reactants [6], which reduces the demands of a challenging intermolecular reaction through the provision of higher reaction rates under milder reaction conditions. Furthermore, this approach generally reduces the degrees of freedom in the transition state (TS), which generally results in improved selectivity and enables the functionalization of these derivatives into useful intermediates for target-directed synthesis.

X A

+

X

B

A–B A

Reactants

B

Tethered reactants

A–B Tethered products

X = Si, P, B, Zn, Mg, ... Scheme 8.1

Temporary tethered strategy in organic synthesis.


Products

226

8 Temporary Silicon–Tethered Ring–Closing Metathesis Reactions

Despite extensive studies with various temporary tethers, silicon has been proven the most versatile group since 1) the reactive components can be easily tethered; 2) the tether is generally stable to the reaction conditions, and; 3) it can be readily cleaved following the key transformation (Scheme 8.1) [7]. Alternatively, the silicon tether can be functionalized using a variety of methods, including oxidation, group transfer, and transmetallation, which further enhances the synthetic opportunities [8]. While the silicon variant of the temporary tether concept is now well established and has been the subject of several reviews [4], this chapter provides the first account of this approach in the context of ring-closing metathesis (RCM) reactions that are specifically applied to natural product synthesis.

8.2 Temporary Silicon–Tethered Ring–Closing Metathesis Reactions

Olefin metathesis represents one of the most important carbon–carbon bondforming reactions in synthetic organic chemistry. Ring-closing metathesis involves the metal-mediated exchange of a metal alkylidene complex with a tethered diene, which has proven highly effective for the construction of an array of carbo- and heterocycles using relatively mild reaction conditions. Hence, it is not surprising that this process has emerged as one of the most popular ring-forming reactions, which has featured so heavily in countless total syntheses [9]. However, a critical limitation with the reaction has been the ability to control alkene geometry, since the reaction generally prefers the formation of the thermodynamically more stable E-alkene. This issue is particularly evident in cross-metathesis (CM) and ring-closing reactions that produce large rings, where there is no formal conformational preference for the Z-isomer. The temporary silicon-tethered ring-closing metathesis (TST-RCM) overcomes this limitation using ring strain to control the formation of the Z-olefin in small and several medium rings, which provides intermediates that can then be functionalized and incorporated into synthetic targets (see below). This process also facilitates the cross-coupling of electronically similar alkenes (Type I), which remain challenging for CM, particularly for the formation of the Z-olefin, making this an important method in the context of total synthesis [10]. This chapter is organized chronologically by the type of tether, metathesis reaction, and relevant application to provide a unique prospective in the field over the last decade in the context of total synthesis of natural products. 8.2.1 O–SiR2 –O Tethered Substrates: Symmetrical Silaketals

The first TST-RCM sequence was reported by Grubbs and Fu for the construction of an achiral 1,4-diol, which circumvented the isolation of the silaketal intermediate in order to simplify purification (Equation 8.1) [11]. Treatment of the bis-alkoxysilane 1


with Shrock’s catalyst [Mo]-I furnished after treatment with tetra-n-butylammonium fluoride (TBAF) the (Z)-1,4-diol 2 in 71% overall yield. A particularly important feature with this approach is that it demonstrated that the sensitive bis-allylic siloxane is tolerated, since this functionality is often prone to Lewis acid activation and the decomposition of the catalyst through elimination of the metal alkylidene intermediate [12]. 1. Cat. [Mo]-I O PhH, rt Ph Si Me 2. TBAF, PhH O 71%

Me Me

(8.1)

OH OH

1

2

8.2.1.1 C 2 -Symmetrical Silaketals and Applications Evans and Murthy were the first to demonstrate the synthetic potential of silaketal intermediates in the TST-RCM with the construction of C2 -symmetrical derivatives (Table 8.1) [13]. A series of enantiomerically enriched allylic alcohols were treated with dichlorodiphenylsilane and 2,6-lutidine to furnish the bis-alkoxysilane intermediates, which were subjected to RCM utilizing Grubbs’ first-generation catalyst [Ru]-I, to provide the corresponding silaketals in excellent yield (entries 1–5). This study demonstrated that the diphenyl-substituted silaketals were stable and that the reaction tolerates a variety of allylic alcohol substituents to provide a general approach for the construction of C2 -symmetrical silaketals. The synthetic utility of this methodology was demonstrated with an application to a concise, total synthesis of the reduced carbohydrate d-altritol (Scheme 8.2) [13]. Dihydroxylation of the C2 -symmetrical silaketal 3 with catalytic osmium tetroxide and N-methylmorpholine (NMO), followed by deprotection of the silyl ethers and

Table 8.1

Synthesis of C 2 -symmetrical silaketals.

Ph2SiCl2

R

2,6-Lutidine

OH

Entry

R

1 2 3 4 5

TBDPSOCH2 PhCH2 c-Hex Ph Npth

R

R O

O

Si Ph Ph

BAS (%)

87 86 95 94 88

Cat. [Ru]-I

R

R O

O Si Ph Ph

CH2Cl2, ∆

Reaction Time (h) 32 26 26 42 25

DPSK (%)

88 95 90 93 84

BAS, bis-alkoxysilane; DPSK, diphenyl silaketal; Npth, 2-Naphthyl; TBDPS = t-BuPh2 Si.

227


228

1. Cat. OsO4 NMO, rt

Ph Ph Si O O

OAc OAc AcO

OTBDPS 2. TBAF, THF then Ac2O, Py 75%

TBDPSO 3

Scheme 8.2

Cat. NaOMe OAc

OH OH HO

OH

MeOH, rt 88%

OAc OAc

OH OH D-Altritol (5)

4

Asymmetric synthesis of D-altritol.

peracetylation of the polyol to expedite isolation, furnished the hexaacetate 4 in 75% overall yield. Saponification of 4 with sodium methoxide in methanol provided d-altritol (5) in 88% yield. Ariza et al. utilized the TST-RCM in conjunction with the Ireland–Claisen rearrangement to facilitate the total synthesis of (−)-phaseolinic acid (Scheme 8.3) [14]. The C2 -symmetrical silaketal 6 was prepared in 58% overall yield using an adaptation of the protocol described by Evans and Murthy [13], which employed enantiomerically enriched propargylic alcohols to form the symmetrical bis-alkoxysilane rather than allylic alcohols. Selective reduction of the bis-alkyne with Lindlar catalyst, followed by RCM with catalyst [Ru]-I, afforded the silaketal 6 in good overall yield. Cleavage of the silaketal 6 with TBAF followed by the acetylation of the diol furnished the bis-acetate 7 in 96% yield. This set the stage for the Ireland–Claisen rearrangement with the homotopic esters. Treatment of 7 with potassium hexamethyldisilazide followed by tert-butyldimethylsilyl chloride at −78 ◦ C provided the silylketene acetal, which underwent rearrangement in refluxing toluene to furnish the tert-butyldimethylsilyl ester. Hydrolysis of the secondary alcohol followed by acid-catalyzed cyclization provided the γ -lactones 8/9 in 85% overall yield from 7, which were epimeric at the α-methyl group (dr = 4 : 1). The mixture of stereoisomers originates from the formation of the E(O)- versus the preferred Z(O)-enolate, which undergo stereospecific rearrangement as the silyl ketene acetals (Scheme 8.3). The synthesis was then completed via the ozonolysis of 8 and oxidation of the resulting aldehyde to provide

C5H11

C5H11 O Si Ph Ph

1. TBAF, THF 2. (EtCO)2O

C5H11

O

C5H11 O

Et3N, DMAP 96%

O

O

6

O Me

7

Me

1. KHMDS, DMPU, C5H11 TBSCl 2. PhMe, ∆ 3. LiOH, H2O, ∆ 4. aq HCl, ∆ 85%

C5H11

Me

C5H11

Z(O)-KA

H

Scheme 8.3

C5H11

SiR3 O O C5H11 H O

E(O)-KA

Me

O

O

C5H11 (dr = 4 : 1)

8

Me

O

O

(−)-Phaseolinic acid (10)

Asymmetric synthesis of (−)-phaseolinic acid.

O 9

1. O3, –78 °C then Me2S 2. NaClO2, H2O2, NaH2PO4

HO2C C5H11

Me

+

93-98 %

SiR3 O O C5H11 Me O

C5H11

O


(−)-phaseolinic acid (10) in 40% overall yield from the commercially available propargylic alcohol. 8.2.1.2 Achiral and Racemic Silaketals Hoye and Promo examined the effect of ring size and alkene substitution on E/Z-selectivity and rate of reaction (Table 8.2) [15]. This study also employed the diphenylsilane as the linker for the construction of symmetrical bis-alkoxysilanes. The optimal reaction conditions involved the slow addition of catalyst [Ru]-I via a syringe pump at room temperature to avoid isomerization of the alkene, which led to truncated ring systems after RCM. Although the reaction favors the formation of the Z-isomer, the selectivity does diminish with increasing ring size (Table 8.2, entries 1–3), whereas substituted alkenes result in lower efficiency and selectivity (entry 1 vs 4). Additional studies examined the effect of a racemic secondary and achiral tertiary allylic alcohol in the TST-RCM process. Interestingly, the former affords an equal mixture of meso- and d,l-silaketals, indicating no obvious diastereoselectivity for this particular tether (entry 5), whereas the latter proved completely unreactive (entry 6). Synthesis of symmetrical silaketals.

Table 8.2

R1 R2 R

n n R R1

R1 R2

O Ph Si Ph O

Cat. [Ru]-I CH2Cl2

n n

O Ph Si Ph O

R1 R2

R2

Entry

n

R

R1

R2

Addition Time for [Ru]-I (h)a

Temperature (◦ C)

1 2 3 4 5 6

0 1 2 0 0 0

H H H Me H H

H H H H Et Me

H H H H H Me

1 10 10 1 1 1

25 25 25 45 55 80

a b

Z/Eb

97 : 3 95 : 5 92 : 8 96 : 4 ≥99 : 1 –

Yield (%)

84 85 78 61 91 NR

Catalyst added in the time indicated via syringe pump. GC–MS ratios.

8.2.1.3 Related Applications and Developments In related studies, Lobbel and Köll utilized this approach in the construction of C2 -symmetrical carbon-linked disaccharides [16]. A critical feature of this approach was the relatively poor efficiency of the RCM, which served to highlight the limitations of the first-generation catalyst [Ru]-I for challenging synthetic applications. Mioskowski et al. later demonstrated that [Ru]-VI provided a more efficient catalyst

229

230


for simple TST-RCM reactions [17]; however, the Grubbs’s second-generation catalyst [Ru]-II has proven the optimal catalyst for many transformations [18]. Nevertheless, the ability to utilize different catalysts provides a unique opportunity to control selectivity based on catalyst reactivity (see below). 8.2.2 O-SiR2 -O Tethered Substrates: Unsymmetrical Silaketals

Hoye and Promo described the first example of the TST-RCM reaction for the preparation of unsymmetrical silaketals (Scheme 8.4) [15]; however, the concept was independently described by Evans and Murthy in their approach to the annonaceous acetogenin (−)-mucocin [19]. The unsymmetrical bis-alkoxysilane 12 was prepared through the sequential reaction of equimolar amounts of dichlorodiphenylsilane, pyridine, and the secondary alcohol rac-11 followed by the addition of allyl alcohol. Treatment of the bis-alkoxysilane 12 in an analogous manner to the symmetrical derivatives provided the unsymmetrical silaketal 13 in 62% yield (Z/E ≥ 19 : 1). Although the construction of unsymmetrical bis-alkoxysilanes remains challenging, it has featured as an important cross-coupling in a number of important total syntheses (see below).

Et OH

Et

Ph2SiCl2, Py allyl alcohol 50%

rac -11

Scheme 8.4

O

Et

Cat. [Ru]-I CH2Cl2, rt 62%

O

Si Ph Ph 12

O

O

Si Ph Ph 13

Preparation of unsymmetrical silaketals.

8.2.2.1 Spiroketals Eustache et al. applied this strategy to the synthesis of the eastern spiro[5.5]ketal fragment of okadaic acid (Figure 8.1) [20]. Sequential coupling of the enantiomerically enriched secondary alcohol 14 and racemic allylic alcohol rac-15 with dichlorodimethylsilane at low temperature furnished the unsymmetrical bis-alkoxysilane 16 in excellent yield as an equal mixture of diastereoisomers (Scheme 8.5). Treatment of 16 with catalyst [Ru]-I furnished the silaketal 17 in 46% yield as a single diastereoisomer, favoring the cis-isomer. Interestingly, this tentative stereochemical assignment contradicts the later studies by Verdine and Evans Me HO2C Me OH

H

O

O O OH

Figure 8.1

Me

Okadaic acid.

H

H

OH Me O

O H

O

H

OH

Me

H

O

8.2 Temporary Silicon–Tethered Ring–Closing Metathesis Reactions OTBDPS Me

Me2SiCl2 n -BuLi, –78 °C

Me

TBDPSO

Me Me

O

O Si rac-15, imidazole Me Me THF, 20 °C TBDPSO 92% 16 TPDPS = t-BuPh2Si

OH 14 OH

[Ru]-I or [Ru]-II

OTBDPS Me

TBDPSO

65% from 16

OTBDPS

rac-15 Me HO Me

H

O

20

Scheme 8.5

O O Si Me Me 17

PhH, ∆

4

O

4

Me

4

1. TBAF, THF 2. TFA, PhMe 60% from 18

OTBDPS 1. MnO2 4 CH2Cl2/EtOAc

Me O

Me

TBDPSO

OH 19

Cat. TFA THF/MeOH 4

Me

OTBDPS OH

Me

2. H2, Pd/C TBDPSO

OH 18

TST-RCM approach to the spiro[5.5]ketal of okadaic acid.

[21, 22], which demonstrated that eight-membered silaketals favor the trans-isomer (see below). Nonetheless, the diastereomeric preference could be avoided using the more reactive [Ru]-II catalyst, which resulted in the formation of the epimeric diols 18 in 65% overall yield after acid-catalyzed deprotection of the tether. Selective oxidation of the allylic alcohol 18 with MnO2 followed by the reduction of the alkene of the α,β-unsaturated ketone affords the hydroxy ketone 19. Deprotection of the primary tert-butyldiphenylsilyl ethers in 19 with TBAF followed by acid-catalyzed spiroketalization provided the spiro[5.5]ketal 20 in 60% overall yield, which is directly applicable to okadaic acid. Additional studies by the same group utilized a similar sequence for the construction of the spiro[5,4]ketal present in attenol A (Scheme 8.6) [23]. The unsymmetrical bis-alkoxysilane 23 was prepared from the epimeric allylic alcohol 21 and the secondary alcohol 22 as a mixture of diastereoisomers. The mixture of epimers was again justified on the basis that this stereogenic center would be oxidized to the ketone, and the previous study had demonstrated that both stereoisomers could be converted to the silaketal. Surprisingly, treatment of the bis-alkoxysilanes 23 with catalyst [Mo]-I furnished the silaketal 24 as a single stereoisomer, favoring the trans-diastereoisomer. In contrast to the previous studies, the diastereomeric bias could not be overcome, which provided the diol 25 in 22% overall yield after acid-catalyzed deprotection of the silyl tether. Although the unreactive bis-alkoxysilane 23’ enriched in the opposite stereoisomer could be recycled, the deprotection of the tether to facilitate the stereochemical inversion was tedious. The diol 25 was then converted to the spiro[5,4]ketal 26 in 72% overall yield using a sequence similar to that described earlier (see above) [20], which allowed the completion of the total synthesis of attenol A (27) using a five-step reaction sequence. Overall, this work demonstrates that there is an inherent preference for the formation of a specific stereoisomer in the TST-RCM in eight-membered rings, which has significant implications for the implementation of this methodology in total synthesis. Kozmin and Marjanovic described an ingenious application of the TST-RCM strategy for controlling spiroketal formation in the total synthesis of spirofungin

231


232

Me OTBDPS O

Me Me2SiCl2 OH n -BuLi, –78 °C OPMB

O

22, imidazole THF, 20 °C 92%

21

I

O

O O Si Me Me

OPMB

Cat. [Mo]-I

O Si Me Me 23′ (dr = 7 : 3)

OPMB

O

OTBDPS I

+

OTBDPS

23 (dr = 1 : 1)

I

Me

PhH, rt

O

OPMB 24

Me

O O Si Me Me

I

OTBDPS

22% from 23

OH 22

TFA THF/MeOH Me

OH

OH O

5 steps

1. MnO2 CH2Cl2/EtOAc

O

O

O

O I Me

26 Me

OH

OTBDPS OH OH

2. H2, Pd/C 3. DDQ, CH2Cl2/H2O 72% OH I

O

OPMB 25

Attenol A (27)

Scheme 8.6

Total synthesis of attenol A.

H O

15

H

11

R1

O 19 H R2 Me H

28 Double anomeric stabilization sterically disfavored

Figure 8.2

H

H Me

O R2

19

Me

15

O H

Me 11

R1

H

H 29 Single anomeric stabilization sterically favored

15

H

O

O 19 Me H

Me 11

H O O Si R R

OSiR3

30 Temporary silicon-tethered assisted configurational lock

Rationale for the TST-RCM spiroketalization.

A over its spiroketal epimer spirofungin B (Figure 8.2) [24]. On the basis of structural analysis with the reveromycins, it was rationalized that spirofungin A was the most likely to display antiproliferative activity in mammalian cells through the inhibition of isoleucyl-tRNA synthase. Although the desired spiroketal 28 is stereoelectronically favored by double anomeric stabilization, there is a severe steric interaction between the axial substituent at C19 with the group at C11, whereas the undesired spiroketal 29 possesses single anomeric stabilization but lacks the steric congestion. This is presumably the reason why there is no thermodynamic preference for their formation. In order to circumvent this problem, a TST-RCM strategy was devised that would be able to take advantage of the relative orientation of the groups at C19 and C11 to lock the configuration in a tethered structure 30. The tethered trienone 33 was prepared in 65% overall yield via the sequential coupling of the primary and secondary alcohols 31 and 32 with dichlorodiisopropylsilane and imidazole, followed by the chemoselective removal of the 1,3-dioxane


233

O O

1. i -Pr2SiCl2, 31 imidazole then Me Me 32, imidazole

O BnO

+

Me

2. Oxalic acid BnO CH2Cl2, SiO2 65%

HO BnO

TBSO

OH 31

O Me BnO O

Si

i-Pr

32

[Ru]-II (30 mol%) Me PhH, ∆ 85%

O

Me

BnO

BnO O Si O i-Pr i-Pr

OTBS

i-Pr 33

34 98%

H O H

O Me

Me

H

Me

Me

OH

Me

H

OH

O

OH H Spirofungin A (37)

Scheme 8.7

O

10 steps

O H

O Me

H2, Pd/C

H Me CHO H OH

H 36

OTBS

1. TBAF, THF

O O

2. NaIO4, THF, H H2O Me H 91%

Me OTBS H

O O Si i-Pr i-Pr 35

Total synthesis of spirofungin A.

protecting group (Scheme 8.7). Treatment of 33 with catalyst [Ru]-II furnished the cyclic dienone 34 in 85% yield, with excellent chemoselectivity for the terminal olefins. Hydrogenation of the dienone 34 and concomitant hydrogenolysis of the secondary benzyl ethers resulted in spontaneous spiroketalization to afford the tethered spiroketal 35 in 98% yield as a single stereoisomer. Another striking feature of this approach is the ability to differentiate the substituents to facilitate the introduction of the side chains. Fluoride-mediated cleavage of the silyl ethers in 35, followed by oxidative cleavage of the 1,2-diol, furnished the hydroxy aldehyde 36 in 91% yield. An additional 10 steps were required to elaborate this intermediate and thus complete the stereoselective synthesis of spirofungin A (37). 8.2.2.2 Long-range Asymmetric Induction Evans et al. devised a novel approach to long-range asymmetric induction using the TST-RCM reaction [22]. It was envisioned that the TST-RCM reaction of the unsymmetrical bis-alkoxysilane 40 would preferentially provide the cis-diastereoisomer 41 (Scheme 8.8), since the reaction should proceed through the favored TS outlined in Figure 8.3. A critical feature of this approach was the ability to suppress intermolecular alkylidene exchange with the prochiral propenyl groups and to optimize the nonbonding interactions using the substituents on the silicon tether. This work demonstrated that, although the steric nature of the tether is important, the catalyst has the most significant influence. For example, catalyst [Ru]-I afforded optimum selectivity, whereas the more reactive [Mo]-I and [Ru]-II catalysts provided inferior yields and selectivity, which is consistent with related studies by Verdine et al. [21]. The diastereoselective TST-RCM of 40 with catalyst [Ru]-I proved remarkably tolerant to a range of aryl, alkyl, and hydroxymethyl derivatives, each of which afforded excellent stereocontrol in favor of the cis-diastereoisomer 41. This concept was also applied to the higher homologs to determine the scope and limitations of this process (Table 8.3) [22]. Interestingly, Grubbs’s

234


i-Pr2SiCl2 (excess) R 38, imidazole

+

R OH

39

Ln M

R

Me

R′

Disfavored TS

Proposed transition states for 1,4-stereocontrol.

1. Cat. [Ru]-II CH2Cl2, ∆

n Si

H

Diastereoselective TST-RCM reactions with homologated alkenyl alcohols.

Table 8.3

O

R

HO R′ O Si

Ln M

Favored TS

Bn

+

Me

O R′ O Si R H H R′

Figure 8.3

O

Diastereoselective TST-RCM for 1,4-stereocontrol.

Me

Me

O

CH2Cl2, 40 °C 54 – 90%

Si i-Pr i-Pr 40

71–88%

Scheme 8.8

O

O O Si Si i-Pr i-Pr i-Pr i-Pr 41 42 (dr up to ≥ 99 : 1) R = Npth (2-Naphthyl), Ph, Pr, c -Hex, BnOCH2,...

then 39, imidazole

OH

38

R

Cat. [Ru]-I O

O

n Bn

2. 10% Pd/C H2, PhH, rt

i-Pr i-Pr

O

O Si i -Pr i -Pr

Pr +

n Bn

Pr O

Si

O

i -Pr i -Pr

Entry

n

Yield of Silaketal (%)

cis : trans

Yield (%)

1 2 3 4

1 2 3 4

92 68 75 87

1 : 11 1 : 27 1:3 1:3

90 92 75 73

second-generation catalyst [Ru]-II proved optimal for the formation of the medium ring silaketals, which were hydrogenated to remove E/Z-isomers and simplify analysis. Although the higher homologs favor the formation of the trans-diastereoisomer (Table 8.3, entries 1–4), this observation is fully consistent with the model for diastereocontrol in which the pseudo-axial/equatorial positions are reversed (Figure 8.4). Hence, the preferred TS for these rings also has both substituents pseudo-equatorial to avoid the steric interaction derived from the axial isopropyl group on the silicon tether. Additional studies demonstrated that this process is fully reversible, in which the diastereoselectivity both increases and decreases for several substrates (Table 8.3, entries 1 and 2), thereby permitting the level of stereocontrol to be tuned accordingly.


Me

Me

Me H

L nM

X RH

235

HH

L nM

O i -Pr Si O

X R

i -Pr

O i -Pr Si O i -Pr

Me Disfavored TS

Favored TS

X = (CH2)n where n = 1 – 4

Proposed transition states for 1, (n + 4)-stereocontrol (where n = 1–4).

Figure 8.4

Pickett et al. utilized the long-range asymmetric induction strategy in an improved synthesis of the sex pheromone (2S,7S)-dibutyroxynonane produced by Sitodiplosis mosellana (Scheme 8.9) [25]. Preliminary studies demonstrated that the TST-RCM reactions with the diphenyl- and diisopropyl tethers furnished the trans-diastereoisomer with excellent stereocontrol (dr = 17 : 1 and 21 : 1, respectively), albeit contaminated with the symmetrical silaketals that were difficult to purify using standard chromatography. It was envisioned that a more sterically hindered silyl tether would reduce the necessity for a large excess of the coupling agent and minimize the formation of the symmetrical silaketal while further improving diastereocontrol. Treatment of the enantiomerically enriched secondary alcohol 43 with di-tert-butylsilyl bis(trifluoromethanesulfonate), pyridine, and dimethylaminopyridine at −78 ◦ C, followed by warming to room temperature and subsequent addition of divinylcarbinol, furnished the bis-alkoxysilane 44. RCM with catalyst [Ru]-I afforded the nine-membered silaketal 45 in 70% overall yield, with improved diastereoselectivity (dr = 32 : 1) and less than 4% of the symmetrical silaketal. Although the more sterically hindered tether successfully overcame the previous problems, the deprotection of the silaketal 45 proved significantly more challenging due to the increased steric congestion at silicon. Nonetheless, the synthesis was completed through the deprotection of silyl tether in 45 by heating with excess TBAF, followed by the hydrogenation of the diene and acylation of the diol to provide the sex pheromone (2S,7S)-dibutyroxynonane (46) in 22% overall yield from 43.

2

Me

OH

t-Bu2Si(OTf)2 Py, DMAP THF, –78 °C to rt (CH2 CH)2CHOH

43

Scheme 8.9

2

Me

O Si t- Bu t- Bu 44 O

1. TBAF (excess) THF, 4 Å MS, ∆

2

Cat. [Ru]-I Me CH2Cl2, ∆ 70% (from 43)

O

Si

O

t-Bu t-Bu 45 (dr = 32 : 1)

2. H2, PtO2 3. (PrCO)2O, Py, DMAP 55%

OCOPr Et

Me PrCOO

Dibutyroxynonane (46)

Total synthesis of the sex pheromone (2S,7S)-dibutyroxynonane.

Harvey et al. recently described an adaptation to the long-range asymmetric induction strategy for the construction of the C12–C24 fragment of peloruside A (Figure 8.5), which simultaneously controls the configuration of the trisubstituted Z-olefin and the C18 stereogenic center (Scheme 8.10) [26]. Sequential coupling of the racemic homoallylic alcohol rac-47 and allylic alcohol 48 with


236

O HO

OMe

O

18

15

HO HO HO O

MeO

OMe

12

OH

Figure 8.5

+

Me

Peloruside A.

Me Ph2SiCl2, rac- 47 Et3N, CH2Cl2 Me HO

Me

0 °C to ∆; then Me 48, Et3N, rt O 76% 48

OH

rac - 47

O

Cat. [Ru]-II Me

O Si O Ph Ph 49

Me

CH2Cl2, ∆ 35–53%

18

Me

12

15

Me

O O O Si Ph Ph 50 (dr ≥ 99 : 1)

Scheme 8.10 TST-RCM diastereoisomer-discrimination approach to the C12–C24 fragment of peloruside A.

dichlorodiphenylsilane furnished the unsymmetrical bis-alkoxysilane 49 in 76% yield as an equal mixture of diastereoisomers. Interestingly, the presence of the C18 ethyl group resulted in a more challenging RCM. Treatment of the unsymmetrical bis-alkoxysilane 49 with the slow addition of [Ru]-II catalyst in refluxing CH2 Cl2 afforded the silaketal 50 in 35–53% yield with ≥99 : 1 diastereoselectivity favoring the cis-diastereoisomer. The ability to utilize the β-hydroxy ketone 48 in the TST-RCM reaction is fairly significant given the propensity of this type of intermediate to undergo retro-aldol and β-elimination. The rationalization of the stereochemical outcome is consistent with the Evans’ model for 1,5-stereocontrol in an eight-membered silaketal (Figure 8.4), which places the C8 substituent in a pseudo-equatorial orientation to avoid nonbonding interactions with the pseudo-axial phenyl group on the silicon tether (Figure 8.6) [22]. Nonetheless, an interesting feature with this process is that the alkylidene exchange occurs with the homoallylic derivative, which is contrary to the previous cases that employ a prochiral alkene. Overall, this provides another example of the inherent versatility of this strategy, which enables remote diastereoselectivity to be controlled. Me

Me

LnRu

Et O

Me O

H

H

O Si Ph Ph

Favored TS

Figure 8.6

LnRu

H O

Me O

H

Et

Si

O Ph

Ph

Disfavored TS

Proposed transition states for diastereoisomer discrimination.


237

8.2.2.3 Annonaceous Acetogenins Evans et al. utilized the TST-RCM strategy in an expeditious total synthesis of the annonaceous acetogenins (−)-mucocin, which has potent antitumor activity (Scheme 8.11) [27]. The unsymmetrical bis-alkoxysilane 53 was obtained in 74% yield via the sequential coupling of the allylic alcohols 51 and 52 under standard conditions. The RCM was particularly challenging, due to the pseudo-axial orientation of one of the substituents. Unfortunately, reducing the steric bias of the tether introduced additional problems with respect to isolation and purification of the silaketal. Hence, treatment of the unsymmetrical bis-alkoxysilane 53 with the syringe pump addition of excess [Ru]-II in refluxing 1,2-dichloroethane (DCE) furnished the silaketal 54 in 83% yield. Fluoride-mediated global cleavage of the silyl ethers followed by a chemoselective diimide reduction of the alkyne and acyclic alkene furnished (−)-mucocin (55) in 12.5% overall yield. In a related study, Brown et al. employed a similar strategy to complete a total synthesis of a related non-adjacent tetrahydrofuran (THF) containing acetogenin, (+)-cis-sylvaticin (Figure 8.7) [28]. An interesting feature of this synthesis is that a catalytic amount of catalyst [Ru]-II may be utilized when the solvent is switched from 1,2-dichloromethane to toluene and the reaction heated at 75 ◦ C. Hoye et al. extended the synthetic utility of the TST-RCM reaction with an elegant and highly convergent total synthesis of (+)-gigantecin and its constitutional isomer (+)-14-deoxy-9-oxygigantecin (Scheme 8.12) [29]. This study examined the merit of combining the TST-RCM reaction in conjunction with CM to facilitate the rapid assembly of this class of nonadjacent THF containing annonaceous acetogenins. Me

Me TBSO +

O

9

51

OH

O OH

TIPSO

i -Pr2SiCl2, 51 TBSO O imidazole, CH2Cl2

TIPSO

O 5

52

0 °C to rt then 52, imidazole 74%

9

O O

Si

i-Pr

i-Pr

O

O

O

O

5

53

[Ru]-II (1.8 equiv.) DCE, ∆

83%

Me Me HO 9

OH O OH

Scheme 8.11

1. HF/MeCN CH2Cl2, rt

O

O 5 OH (–)-Mucocin (55)

2. TsNHNH2 NaOAc, ∆ 87%

O

Total synthesis of (−)-mucocin.

Me OH 5

8

O

O OH

Figure 8.7

OH

OH

(+)-cis-Sylvaticin.

O O

TIPSO

TBSO 9

O O

i-Pr

Si

O

i-Pr

O 54

O O

5


238

11

TIPSO

O 56

+

O OH 57

OH

Ph2SiCl2, 56, Py, PhMe then 57 0 °C to rt

52%

11

O TIPSO

Cat. [Ru]-III 67% PhMe, 80 °C

O

Si O Ph Ph 58

O Cat. [Ru]-III, 60 CH2Cl2, ∆

63%

Me

11

TIPSO

OH

O Ph O O Si Ph O 59

11

O TIPSO

O

Si O Ph Ph

1. Cat. [Ru]-III, 60, ∆ 2. TsNHNH2, NaOAc, ∆ 48% 3. HF/CH3CN, rt

O OH

OH

O

OH OH

14-Deoxy-9-oxygigantecin (61)

O 11

O

O OH Me

1.TsNHNH2, NaOAc H2O, DME, ∆ 2. HF/CH3CN, rt Me OH O ( )5 O O OH

OH

(+)-Gigantecin (63)

RCM/CM approach

Me TIPSO 60

Scheme 8.12

O

69%

( )4

11

O 62

O

CM/RCM approach

O O

Total synthesis of 14-deoxy-9-oxygigantecin and (−)-gigantecin.

The bis-alkoxysilane 58 was prepared through the sequential loading of the allylic alcohols 56 and 57 on to dichlorodiphenylsilane. Interestingly, the RCM of 58 with Hoveyda–Grubbs’ second-generation catalyst [Ru]-III furnished the 11-membered silaketal 59 rather than the desired 7-membered derivative. Although this anomaly was not immediately obvious, since it is a constitutional isomer of the desired product, it is presumably due to preferential reaction with the less hindered terminal Type I alkene [10]. Butenolide 60 was then installed with a CM with 59, followed by the chemoselective reduction and global cleavage of the silyl ethers to afford 14-deoxy-9-oxygigantecin (61) in 48% overall yield. Interchanging the order of the CM and TST-RCM reactions allows for the preferential functionalization of the less hindered and more reactive Type I alkene and thus permits the formation of the seven-membered silaketal without competitive RCM. A critical feature for the success of this strategy is that the formation of the cis-silaketal is preferred over the trans-derivative, which provides optimal rates of reaction [22]. Treatment of a mixture of the unsymmetrical bis-alkoxysilane 58 and butenolide 60 (1 : 4 ratio) with catalyst [Ru]-III provided the gigantecin skeleton 62 in 63% yield in a single step. Additionally, the formation of the homodimer of butenolide 60 is inconsequential since it can be recycled through


a standard CM with ethylene, and the formation of the 11-membered silaketal is significantly reduced (10%) albeit having undergone a CM with butenolide 60 to provide direct access to the constitutional isomer. The synthesis, of (+)-gigantecin (61) was completed in a manner analogous to that described for the constitutional isomer in 69% overall yield [29]. Overall, these total syntheses serve to highlight the strategic advantages of the TST-RCM/CM approach for the expeditious syntheses of these agents, and the importance of understanding the rates of reaction for these processes. 8.2.2.4 Trisubstituted Alkenes Mulzer et al. utilized the TST-RCM strategy to prepare the Z-trisubstituted C12–C13 alkene of the northern fragment of epothilone B and D (Figure 8.8) [30]. Preliminary studies demonstrated that acyclic dienyl esters were unsuitable substrates for the RCM that provide eight-membered lactones. It was envisioned that the strain in the TS could be relieved using a silicon tether, as the silicon atom is larger and has more polarizable soft d-orbitals that allow for more facile bond distortion. Hoye and Promo had previously demonstrated that a nine-membered silaketal is accessible in excellent yield and with 95 : 5 selectivity for the Z-isomer, thereby illustrating proof-of-concept (Table 8.2) [15]. The unsymmetrical bis-alkoxysilane 66 was prepared via the sequential silylation of the homoallylic alcohols 65 and 64 with dichlorodimethylsilane (Scheme 8.13). The RCM of the unsymmetrical bis-alkoxysilane 66 required the slow addition of either [Ru]-II or [Ru]-III second-generation catalysts, which afforded the silaketal 67 in 98% yield, as a 5 : 1 mixture favoring the Z-isomer. Interestingly, the addition of the catalyst in a single portion leads to a significant reduction in the yield due to its decomposition. Cleavage of the silyl tether in 67, followed by a standard differential protection sequence, provided the primary alcohol 68 in 88% overall yield. Sequential functionalized of the termini of 68 using a four-step sequence to homologate the alcohol and install the thiazole substituted E-trisubstituted alkene furnished 69 in 94% overall yield. The key aldehyde 70, which had previously undergone fragment coupling en route to epothilone B and D, was prepared in an additional four steps. Although the selectivity in the formation of the C12–C13 olefin is suboptimal, it serves to demonstrate the potential of the TST-RCM approach for the preparation of challenging trisubstituted olefins.

O S

12

S

13

OH

N O

OH

N O

O

Figure 8.8

12 13

OH O

Epothilone B and D.

O

OH O

239

240


+

Me OH PMBO 64

65, n-BuLi, THF, –78 °C Me

HO 65

Me

Me Cat. [Ru]-II or III

Me Me2SiCl2, rt O O Si PMBO then 64, DMAP, Me Me DMF, rt 84% 66

CH2Cl2, ∆ 98%

Me O Si O Me Me 67 (E /Z = 1 : 5)

PMBO

1. TBAF, THF 88% 2. TBSOTf, CH2Cl2 2,6-lutidine then HF Me S Me

Me 4 steps

Me

57%

N TBSO

Me

70

Scheme 8.13

CHO

S Me

Me 4 steps PMBO

Me

N

CN

94%

TBSO 69

Me TBSO

OH 68

Total synthesis of epothilone B and D.

8.2.2.5 Related Applications and Developments In related studies, the TST-RCM has been employed to prepare 2-ene-1,5-diols that can be converted to functionalized THF’s present in acetogenins using an electrophilic selenocyclization [31]. This strategy is also featured in a novel approach to the pyrrolo[3,2-c]quinoline core of martinelline and martinellic acid natural products [32]. Additionally, Verdine et al. utilized this methodology to prepare stereodiversified libraries of 1,4- and 1,5-enediols, which enabled the identification of a low nanomolar ligand for the mu opioid receptor (MOR) [21]. Finally, Postema and Piper revisited the construction of carbon-linked disaccharides using the TST-RCM with the more active Grubbs’s second-generation complex, which demonstrated that the RCM of hindered carbohydrates remained problematic for fully functionalized carbohydrates [33]. 8.2.3 Dienyne TST-RCM: Symmetrical and Unsymmetrical Silanes

Lee et al. examined the merit of the dienyne TST-RCM with symmetrical and unsymmetrical bis-alkoxyalkynylsilanes for the stereoselective construction of 1,3-butadienes which are not accessible through classical enyne metathesis reactions [34–36]. Table 8.4 outlines the key findings with the symmetrical bis-alkoxyalkynylsilanes, which were prepared through the base-induced alcoholysis of trialkynylsilanes with various racemic and enantiomerically enriched secondary alcohols [36], and subjected to the enyne TST-RCM using catalyst [Ru]-II to afford the silaketals in excellent yield (Table 8.4, entries 1–6). Removal of the silyl tether under relatively forcing conditions, furnished the 1,3-butadiene derivative with retention of the alkene configuration. This study demonstrated the importance of tailoring the reactivity and stability of the trialkynylsilane with the volatility of


241

Stereoselective synthesis of 1,3-dienes via the dienyne TST-RCM of symmetrical bis-alkoxyalkynylsilanes.

Table 8.4

CP Si

R1

Cat. NaH ROH (2 equiv.)

R2

R2 R

CP O Si

n

Hexanes, rt

2

O

R

R1

n

CH2Cl2, ∆

2

3

Cat. [Ru]-II

R1

O

CPO Si

R

R

OH TBAF

n

n

3

R

R1

R1

OH

1 n R

THF, 60 °C n

2

R2

CP = cyclopentyl

R1

R2

R3

n

BAAS (%)

Method

BCS (%)

Diol (%)

Me Ph(CH2 )2 (R)-BnOCH2 Ph(CH2 )2 Ph(CH2 )2 Me

CH2 OMe CH2 OMe CH2 OMe Me CH2 OMe Me

H H H H Me n-Pent

1 1 1 1 1 2

70 74 67 82 81 66

A A A A B B

75 89 83 88 88 76

68 78 68 65 83 50

Entry

1 2 3 4 5 6

Method A: CH2 Cl2 , 40 ◦ C; Method B: toluene, 110 ◦ C. BAAS, bis-alkoxyalkynylsilane; BCS, bicyclic siloxane.

the displaced alkyne to reduce the formation of the trialkoxysilane (entry 2 vs. 4). It also demonstrated that there is no diastereoisomer discrimination with the racemates and that the enantiomerically enriched derivatives furnish a mixture of diasteroisomers at silicon (entries 1/2 vs. 3). Additional alkene substitution required more forcing reaction conditions (entry 5) and, lastly, the process was amenable to the formation of larger rings (entry 6). Additional studies expanded the scope of this process to unsymmetrical bis-alkoxyalkynylsilanes since they are more likely to be of synthetic utility in the context of a total synthesis [36]. These systems represent a significant synthetic challenge with regard to the preparation of the unsymmetrical derivatives and the ability to initiate a selective enyne metathesis reaction to avoid the formation of regioisomers. The unsymmetrical bis-alkoxyalkynylsilane 72 was prepared through an adaptation of the protocol described for the symmetrical derivatives (Scheme 8.14). Sequential treatment of the trialkynylsilane 71 with a tertiary alcohol followed by a primary alcohol in the presence of a catalytic amount of 1. Cat. NaH, 50 °C OH

CP Si

Ph

Ph Ph

Ph CP O O Si

OMe HO OMe 71

Scheme 8.14

78%

OMe

79%

CP = Cyclopentyl

Ph CP Ph O O Si

72

Ph Ph TBAF

OH

THF, 60 °C

CH2Cl2, ∆

2. Cat. NaH, rt

OMe

Cat. [Ru]-II

OMe 73

Enyne TST-RCM of unsymmetrical bis-alkoxyalkynylsilanes.

77%

OMe 74

OH

242


base furnished the unsymmetrical bis-alkoxyalkynylsilane 72 in 79% overall yield. Critical to the success of this process is the addition of the tertiary alcohol followed by the primary derivative, which provides the necessary steric differentiation between the two pendant alkenes to facilitate initiation at the less hindered olefin. Hence, the sequential enyne metathesis of 72 furnished the bicyclic siloxane 73, which upon cleavage of the silyl tether provided the unsymmetrical 1,3-diene 74 in 60% overall yield from 72. Alternatively, the enyne TST-RCM of 75 provides a mixture of bicyclic siloxanes 76/77 in 81% yield, as 1.5 : 1 mixture (Equation 8.2), which illustrates the importance of being able to differentiate the alkene groups in the initiation step. Me

Me CP O O Si

Cat. [Ru]-II

Me

Me

CH2Cl2, ∆ 81% OMe

75

Me

O CPO Si

76

+

Me OCPO Si

(8.2) OMe

OMe

77

(1.5 : 1 mixture, not assigned)

8.2.3.1 Macrolide Antibiotics Kim and Lee utilized this strategy for the construction of the E/Z-butadiene motif at C14–C15 in the boron-containing C2 -symmetrical macrodiolide tartrolon B (Figure 8.9), which belongs to the class of ion-carrier antibiotics active against Gram-positive bacteria [37]. Interestingly, tartrolon A1–3 and B are all biologically active, which indicates that the presence of boron is not necessary for its antibiotic activity. Another striking feature of this metabolite is the homology of the C1–C7 segment, which is also present in the structurally related boron-containing antibiotics boromycin, aplasmomycin, and borophycin. In light of the problems encountered with group differentiation, the alkenes were differentially substituted to garner the necessary regiocontrol in the enyne TST-RCM reaction. The unsymmetrical bis-alkoxyalkynylsilane 80, applicable to the natural product, was prepared as a mixture of diastereoisomers at silicon in 58% yield by the addition of the secondary alcohol 78 to the bis-alkynylsilane 79 in the presence of catalytic sodium hydride (Scheme 8.15). Treatment of 80 with catalyst [Ru]-II furnished 17

17

Me 14

O

O Me

Me

OH O 9

14

Me

2

O

OH OH

O

Figure 8.9

OH

O

O

HCl

2

O

Me

2

Na2B4O7

HO O HO Me Me

Tartrolons A1–3 and B.

OH O

O

O Me O

O

9

Na B

O

O

Me

O

2

O

Me O

OH

O

Me Me

8.2 Temporary Silicon–Tethered Ring–Closing Metathesis Reactions O

O Ph

PMBO

PMBO

Si Me O

O

79 OBn

Me OMOM

Cat. NaH hexanes, rt 58%

Me

Me OMOM Me MOMO

O Ph Si O 80

OH

MOMO

O

78

Me OBn Cat. [Ru]-II

89%

CH2Cl2, ∆

O

O

2

PMBO

PMBO

O TBAF, THF

Me OMOM

60%

7 11

9

Me

MOMO

OH

OH

Me

Me MOMO

Me Ph O O Si

81

82

Scheme 8.15

O Me OMOM

Synthesis of C1–C21 fragment of tartrolon B.

the bicyclic siloxane 81, which was deprotected with TBAF to afford the C1–C21 fragment applicable to tartrolon B [37a]. Unfortunately, the stereochemical assignments at C2, C7, C8, C9, and C11 had the opposite stereochemistry to the natural product, since the incorrect enantiomer of B-methoxydiisopinocampheylborane was employed for the construction of 78 [37b]. Nonetheless, this method demonstrated the synthetic utility of this approach for the construction of stereodefined 1,3-butadienes. H OH

H

O H H

OAc Me

H

O O

Figure 8.10

(−)-Cochleamycin A.

More recently, Lee et al. have also employed this strategy for the construction of the E/Z-1,3-butadiene motif required for the synthesis of (−)-cochleamycin A using a biomimetic type transannular Diels–Alder reaction, (Figure 8.10) [38]. The enantiomer exhibits antimicrobial activity against Gram-positive bacteria and has modest cytotoxic activity. The unsymmetrical bis-alkoxyalkynylsilane 85 was prepared as a mixture of diastereoisomers at silicon by the addition of the secondary alcohol 84 to the bis-alkynylsiloxane 83 (Scheme 8.16). Although the

243

244

8 Temporary Silicon–Tethered Ring–Closing Metathesis Reactions PivO OEt

CP

EtO

Si

O S

Me

PivO 84 HO

OEt Me

S 83

EtO2C

85

O

Me TBSO TBSO

87

Scheme 8.16

1. Cat. [Ru]-II, DCE, ∆ 2. TBAF, THF, ∆

PivO 11 steps

OEt EtO

Me HO OH

S

TBSO

Me

S

61%

HO

Si

O S

O

CP

EtO

Cat. NaH PhMe, 60 °C 68%

Me

S 86

Synthesis of C1–C19 fragment of (−)-cochleamycin A.

bis-alkynylsiloxane 83 was prepared from the trialkynylsilane in 81% yield without the formation of the corresponding symmetrical silaketal, the preparation of the unsymmetrical silaketal 85 required elevated temperature, which also led to problems with the double addition of the secondary alcohol 84. Treatment of the unsymmetrical bis-alkoxyalkynylsilane 85 with catalyst [Ru]-II catalyst followed by the deprotection of the bicyclic siloxane using TBAF furnished the E/Z-butadiene 86 in 61% yield. Although the enyne TST-RCM proceeds with exquisite group selectivity, premature termination at the monocyclic siloxane remained problematic even with longer reaction times and higher catalyst loadings. The elaboration of this fragment using an 11-step sequence provided the same intermediate employed by Roush et al., which completes a formal synthesis of the enantiomer of the natural product. 8.2.4 O–SiR2 –C Tethered Substrates: Allyl and Vinylsiloxanes

The first TST metathesis sequence involving a mono-alkoxysilane was reported by Grubbs and Chang for the construction of achiral (Z)-1,n-diols and hydroxy aldehydes (Table 8.5 and Equation 8.3) [39]. The formation of the mono-alkoxysilanes using an alkyl or vinylsilane is well established and comparatively trivial as compared to the bis-alkoxysilanes described earlier in this chapter. However, the oxidative and reductive functionalization of the products from the RCM remains a significant problem. Table 8.5 outlines the preliminary work that inspired the subsequent applications to natural product synthesis. Although Grubbs’s first-generation catalyst [Ru]-I is effective for the allylsiloxane (Table 8.5, entries 1–2), the more sterically hindered vinylsiloxanes required the more reactive Schrock catalyst [Mo]-I (entries 3–4). Nevertheless, these represent the first examples of vinylsiloxanes undergoing RCM, and ironically the catalyst requirement has not changed despite the

8.2 Temporary Silicon–Tethered Ring–Closing Metathesis Reactions Table 8.5

TST-RCM reactions of acyclic mono-alkoxysilanes.

m

n Si O Me Me

BnO

Entry 1 2 3 4

Cat. [Mo]-I or [Ru]-II Solvent, rt

m BnO

O

n Si Me Me

m

n

Catalyst (mol%)

Solvent

Yield (%)

1 2 1 2

1 1 0 0

[Ru]-I (3) [Ru]-I (5) [Mo]-I (3) [Mo]-I (5)

CH2 Cl2 CH2 Cl2 PhH PhH

88 96 84 91

advent of new, more reactive ruthenium complexes. Another interesting feature is the necessity to slightly increase the amount of catalyst used upon increasing ring size; however, the process does not require high dilution for the formation of an eight-membered ring (entry 2).

n BnO

Si O Me Me 88 n = 0,1

KF, KHCO3 H2O2 MeOH/THF rt

OH

OH OH vs. BnO

BnO 89 n = 0 (80%)

CHO

(8.3)

90 n = 1 (93%)

In additional studies, this work also demonstrated the oxidative cleavage of these silacycles using the Tamao and Fleming oxidation (Equation 8.3). Interestingly, the allylsilane 88 (n = 1) provided the (Z)-1,5-diol 89, whereas the vinylsilane 88 (n = 0) furnished the γ -hydroxy aldehyde 90 [39]. 8.2.4.1 Lignans from Allylsiloxanes Cossy and Meyer independently described a similar TST-RCM of monoalkoxysilanes 91 for the preparation of a variety of allylsiloxanes 92 (Scheme 8.17) [40]. Although the results were analogous to those reported by Grubbs, this study demonstrated that the larger rings required more forcing reaction conditions and that allylic substituents significantly reduce the overall efficiency, illustrating the sensitivity of catalyst [Ru]-I to steric effects. Nonetheless, a striking feature of this work is the ability to combine the TST-RCM sequence with a Sakurai reaction using aldehydes and ketones to facilitate the stereoselective construction of triand tetrasubstituted THF’s and THP’s 93/94 (n = 1–2). In concurrent studies, Marsden et al. reported an analogous reaction sequence, albeit using a different Lewis acid, in a study that focused entirely on the construction of trisubstituted THF’s 93/94 (n = 1) [41]. Interestingly, these studies demonstrated that the Sakurai reaction can be effected with either boron trifluoride etherate or trimethylsilyl triflate, wherein straight chain alkyl aldehydes provide optimal stereocontrol over

245

246

8 Temporary Silicon–Tethered Ring–Closing Metathesis Reactions Cat. [Ru]-I, CH2Cl2

n Si Me O Ph Me 91 n = 0,1,2

Cat. Lewis acid

n

rt or ∆ 85 – 95%

Ph

+

n

RCHO Si Me O CH2Cl2, –78 °C Me 65–83% 92

Ph

O 93

R

n Ph

n = 1,2

O

R

94

Scheme 8.17 Stereoselective construction of cyclic ethers using the TST-RCM in conjunction with the Sakurai reaction. SiR3 Ph H

O +

R

H Favored TS

Figure 8.11

vs.

SiR3 Ph H

O +

H

R Disfavored TS

LA

Ph

O H

+

LA

Ph

R

H

H

O +

H R

Epimerization

Origin of selectivity and equilibration in the Sakurai reactions.

branched alkyl and aryl derivatives (dr = 3.3 : 1–49 : 1). Figure 8.11 illustrates the model for stereocontrol, which involves the initial condensation of silyl ether with the aldehyde to provide the (E)-oxonium ion that undergoes cyclization through a chair-like TS (favored TS). Marsden et al. demonstrated that the formation of the minor diastereoisomer is a result of the ionization of the cyclic ether, which places the C2 substituent pseudo-axial to avoid the gauche interaction with the vinyl group. This insight explained that the extent of equilibration could be attributed to the reaction temperature and the nature of the Lewis acid, which provided reaction conditions that facilitate the selective formation of either diastereoisomer [41b]. Additional studies by Marsden et al. demonstrated the versatility of this strategy for the construction of aryltetralins and lignans [42]. The highlight of this work was a concise total synthesis of the lignan (+)-virgatusin, which is a 2,5-diaryl-3,4-di(alkoxymethyl)THF and the enantiomer of the natural product (Scheme 8.18). Although Cossy and Meyer had demonstrated that this substitution pattern undergoes the Sakurai reaction, they also demonstrated that the presence of allylic substituents were challenging for the RCM [40]. Treatment of the syn-aldol adduct 95 with allyldimethylsilyl chloride and imidazole afforded the allylsiloxane, which was subjected to RCM using Grubbs’s first-generation catalyst [Ru]-I to provide the allylsiloxane 96 in 70% overall yield. The origin of the improved efficiency of the metathesis reaction can be ascribed to the anti- versus the syn-stereochemical arrangement in the product, which minimizes unfavorable gauche interactions. Treatment of the cyclic allylsiloxane 96 with piperonal in the presence trimethylsilyl triflate at −78 ◦ C furnished the tetrasubstituted THF 97 in 63% yield as an inseparable mixture of diastereoisomers (dr = 10 : 1). Conversion of this intermediate to (+)-virgatusin (98) using a four-step sequence provided the enantiomeric natural product with a significantly reduced diastereomeric ratio (dr = 3 : 1), which was attributed to chromatographic resolution rather than equilibration. Rodr´ıguez-Garcia et al. extended the TST-RCM to the pterocarpans, which have a cis-fused benzofuran–benzopyran skeleton (Scheme 8.19) [43]. These agents are


O

1. AllylMe2SiCl imidazole DMF, rt

OH Ar1

Xc

2. Cat. [Ru]-I CH2Cl2, ∆ 70%

95

Ar 2CHO, TMSOTf

COXc Me Si

O Me

Ar1

COXc

CH2Cl2, –78 °C 63%

Ar 2

Ar1

O

96

97

Ar = 3,4-(MeO)2C6H3 1

Ar 2 =

O 4 steps

O

O Xc = O

247

N

MeO

Bn

OMe O OMe

O O

OMe

(+)-Virgatusin (98)

Scheme 8.18 Concise total synthesis of (+)-virgatusin using the TST-RCM and Sakurai reaction. 1. OsO4, NaIO4 2. LiAlH4 MeO 3. DIAD, Ph3P THF, rt 49%

1. Cat. [Ru]-II MeO OMe CH2Cl2, ∆ Me

2. 100, BF3·OEt2 61%

Si O Me 99

MeO 100

101 CHO OPiv

Scheme 8.19

H H

OMe

O

cis -Pterocarpan (102)

OMe O OPiv

O

48% 1. OsO , NaIO 4 4 2. NaBH4 3. DIAD, Ph3P THF, rt

O H MeO

H

O

OMe

trans- Pterocarpan (103)

Stereoselective synthesis of cis- and trans-pterocarpans.

part of a large class of isoflavonoids that have an array of biological activity, which makes them attractive synthetic targets. Although the synthetic strategy is directly analogous to the work of Cossy and Marsden, there are a number of interesting observations that enhance the understanding of this process [40, 41]. For example, the RCM of the allylsiloxane 99 proceeds more efficiently with catalyst [Ru]-II and the Sakurai condensation of the resulting cyclic allylsiloxane with the aldehyde 100 in the presence of boron trifluoride etherate furnished the cis-2,3-disubstituted benzofuran 101 with excellent stereocontrol (dr ≥19 : 1) irrespective of the reaction temperature. Although the improved efficiency of the RCM is expected with the newer catalyst; the level of stereocontrol is somewhat intriguing given the work of Marsden et al., which demonstrated that these derivatives are prone to ionization and equilibration [41b]. The synthesis was completed through a three-step sequence that culminates with a Mitsunobu cyclization of the hydroxyphenol derived from

248


101 via the oxidative cleavage of the vinyl group and reduction of the aldehyde with concomitant reductive cleavage of the pivaloyl group to afford the cis- and trans-pterocarpans (102) and (103) in modest overall yield. Although the oxidative cleavage of the vinyl group does not equilibrate the aldehyde, the chromatographic purification leads to a mixture of epimers favoring the trans-diastereoisomer (dr = 6 : 1). The epimerization can be avoided by reducing the crude aldehyde, which provides a stereodivergent synthesis. Finally, the origin of the stereocontrol was rationalized by the direct allylation of the aldehyde through a closed TS, followed by displacement of the secondary carbinol with the phenoxide, which is similar to the mechanism initially proposed by Marsden et al. [41a]. 8.2.4.2 Z-Trisubstituted Alkenes from Allylsiloxanes Barrett et al. employed the TST-RCM strategy to prepare a series of C19 analogs from 1α-hydroxy-5,6-trans-vitamin D2 (Figure 8.12) [44]. Although the hormonally active form of vitamin D3 promotes cell differentiation and inhibits tumor cell proliferation, it also leads to hypercalcemia at therapeutically significant dose levels. Interestingly, several studies have demonstrated that structural changes to this agent provide analogs that discriminate between these activities, albeit there are relatively few reports on C19 modifications. Me Me Me

HO

Me Me Me

OH

H

H

19

19

OH

Figure 8.12

HO

1α, 25-Dihydroxyvitamin D3 and Vitamin D2 .

Treatment of the allylic alcohol 104 with allyldimethylsilyl chloride followed by catalyst [Ru]-II furnished the cyclic allylsiloxane 105 in good overall yield (Scheme 8.20). The level of chemoselectivity is outstanding in this process, since the reaction occurs exclusively between the two terminal alkenes of the pentaene, Me Me Me

H 19

HO

OTBS 104

Scheme 8.20

Me

Me

Me Me 1. AllylMe2SiCl Et3N, CH2Cl2, rt 2. Cat. [Ru]-II 19 CH2Cl2, ∆ 62% Me Si O Me

H

Me Me A. H2O2, KF DMF, rt B. TBAF, THF rt

OTBS 105

H 19

OR A: 106 R = TBS Conditions B: 107 R = H

Synthesis of vitamin D2 analogs via the TST-RCM.


which is presumably the result of group differentiation in the initiation step. This study also examined the merit of the phosphonate tethers, which behave analogously in the RCM. Unfortunately, the attempted Tamao–Fleming oxidation of the allylsiloxane 105 to the diol provided the pentaene 106, whereas the treatment of 105 with TBAF furnished the deprotected pentaene 107 in 50% and 64% yield, respectively, through a preferential Peterson olefination. Finally, the product pentaene 107 provides a versatile intermediate for cycloaddition reactions to demonstrate the utility of this strategy, despite the fact that it also highlights the limitations with the functionalization of trialkysiloxanes. Miller and Li utilized the TST-RCM reaction to overcome the poor selectivity that had previously been obtained in the installation of the challenging exocyclic ethylidene using a Wittig reaction in the total syntheses of (+)-streptazolin (Scheme 8.21) [45]. This alkaloid, isolated by Drautz and Zahner from cultures of Streptomyces viridochromogenes, has both antibiotic and antifungal activity and possesses an unusual skeleton due to an internal carbamate unit, which requires the partial hydrogenation to allow purification and characterization due to its propensity to undergo polymerization. Preliminary studies demonstrated that the internal urethane unit impacts the ability to affect the oxidative cleavage of the cyclic allylsiloxane in a manner similar to that described by Barrett et al. for the vitamin D2 analogs [44]. Furthermore, the formation of the unsymmetrical bis-alkoxysilane led to the formation of dimers with either [Ru]-I or [Ru]-II catalysts, which was attributed to a less ordered TS. Saponification of the secondary allylic acetate 108 followed by the silylation of the resulting allylic alcohol with allyldimethylsilyl chloride furnished the allylsiloxane 109 in 74% overall yield. RCM using catalyst [Ru]-II, furnished the six-membered ring-fused silacycle 110 in quantitative yield, which was treated with potassium fluoride and KHCO3 to afford the exocyclic ethylidene 111 in 50% yield. Although the yield is somewhat modest, this represents one of a few examples of the protodesilylation of a conjugated and fused allylsilane. 1. MeOH, K2CO3

Si Me O Me

2. AllylMe2SiCl

OAc N O

H OPiv OEt

TEA, DMAP CH2Cl2, rt 74%

N O

108

H OPiv OEt

Si Me O Me

Cat. [Ru]-II CH2Cl2, ∆ 99%

N O

109

H OPiv OEt

110 KF, KHCO3 50% MeOH/THF (1 : 1)

OAc N

H

O

O 113

Scheme 8.21

1. H2, Pd/C 2. Ac2O, Pyr. 90%

NaOMe/MeOH

OH N

H

O O (+)-Streptazolin (112)

Total synthesis of (+)-streptazolin.

OH

∆ 76%

N O

H OPiv OEt

111

249

250


Interestingly, the oxidation of the allylsiloxane 110 provided the allylic alcohol in 55% yield, however the conversion to 111 proved problematic. The synthesis was completed with the removal of the pivaloyl group in 111 followed by in situ carbamate formation to afford (+)-streptazolin (112) in 76% yield, which was converted to the crystalline allylic acetate 113 for characterization. The phytotoxic agents cornexistin and 14-hydroxycornexistin were independently isolated from the fungus Paecilomyces variotii Bainier strain SANK 21086, and are members of the nonadride family that are biosynthetically derived from the combination of nine maleic anhydride groups (Figure 8.13). These agents display potent herbicidal activity at the same concentration against a number of grass and broadleaf weed species with minimal phytotoxicity toward maize, making them significantly important as potential herbicides. Additionally, cornexistin and 14-hydroxycornexistin display complementary activity, in which the former tends to be more active against grass weeds, while the latter is significantly more active against the broadleaf weeds. OH Me OH

O O

OH

O O

O O

n -Pr

OH

Figure 8.13

O O

n -Pr

OH

Cornexistin and 14-hydroxycornexistin.

These features inspired Taylor et al. to develop a unified strategy for the preparation of these agents using the TST-RCM reaction to install the critical Z-trisubstituted alkene at C7 (Scheme 8.22) [46]. Interestingly, the relative stereochemistry at C8 is critical for the success of this strategy, since the allylsiloxane derived from the secondary alcohol 114 furnished dimeric products irrespective of the catalyst employed. Inversion of the C8 stereogenic center using a two-step oxidation/reduction sequence provided the epimer 115 (dr = 4 : 1), whereas the attempted Mitsunobu reaction led to elimination. Treatment of the secondary alcohol 115 with allyldimethylsilyl chloride provided the allylsiloxane, and subsequent exposure to [Ru]-II catalyst gave the cyclic allylsiloxane 116 in modest overall yield due to the relatively modest silylation yield (50%). Although the attempted protodesilylation of 116 required for cornexistin was unsuccessful due to the lack of reactivity, the Tamao–Fleming oxidation afforded the diol 117 in quantitative yield. 8.2.4.3 Di- and Trisubstituted Alkenes from Vinylsiloxanes The advantage of the TST-RCM strategy is the ability to rationally prepare sterically hindered and inactivated Z-olefins, which remain the most challenging substrates for CM. Barrett et al. outlined a series of novel ring-opening reactions that permit the stereoselective construction of E- and Z-olefins (Scheme 8.23) [47]. The vinylsiloxane 118 was prepared in a manner analogous to that described by Grubbs using the Schrock catalyst [Mo]-I in excellent yield [39].

8.2 Temporary Silicon–Tethered Ring–Closing Metathesis Reactions OH

Me Si Me R1 R2

8

MeO2C

O

1. AllylMe2SiCl Et3N, CH2Cl2

KHCO3

2. cat. [Ru]-II OPMB

MeO2C OAc

n-Pr

CH2Cl2, ∆

OPMB

MeO2C

OH

KF, H2O2

8

MeO2C

THF/MeOH 99%

114 R1 = OH, R2 = H

OPMB

MeO2C

OAc n-Pr 116

50%

8

MeO2C

n-Pr 117

OAc

115 R1 = H, R2 = OH

Scheme 8.22 Stereoselective synthesis of Z-trisubstituted alkene in cornexistin and 14-hydroxycornexistin using the TST-RCM strategy.

OH

n-Hex

c -Hex Me 119

Scheme 8.23

TBAF, DMSO n -Hex

Me Me Si O

80 °C 78%

MeLi, THF

c -Hex Me 118

0 °C to rt 58%

OH

n -Hex Me3Si

c -Hex Me 120

Stereoselective construction of E- and Z-homoallylic alcohols.

Treatment of the vinylsiloxane 118 with TBAF in DMSO at 80 ◦ C furnished the E-homoallylic alcohol 119, whereas the analogous substrate with methyllithium (or other organolithium reagents) afforded the Z-homoallylic alcohol 120 as the exclusive products. Barrett et al. employed the organolithium ring-opening strategy for the synthesis of E-alkene in a total synthesis of d,l-glucosylceramide, which is a member of the glycosphingolipid class of natural products (Scheme 8.24) [48]. These agents are associated with formation of many structures in biological systems, as they provide for the shape and support for cell membranes. Hence, they are intimately involved in cellular recognition, which has implications for many biological processes, including cancer, allergy, and inflammation. The preparation of these agents present several challenges, the most notable being the regio- and stereoselective construction of the polysaccharide unit through the O-glycosylation and the stereoselective installation of the E-alkene. The examination of the TST-RCM was prompted after preliminary studies demonstrated that CM, was unsuccessful due to the presence of an allylic substituent [10]. The acyclic vinylsiloxane 122 was prepared from the secondary alcohol 121 using the silyl triflate formed in situ from the silyl chloride with silver triflate. Treatment of the vinylsiloxane 122 with catalyst [Mo]-I followed by phenyllithium afforded the Z-vinylsilane 123 in 83% overall yield, which was converted to d,l-glucosylceramide (124) using a four-step sequence. The ability to chemoselectively manipulate the cyclic vinylsiloxane intermediate in the presence of additional silyl ethers illustrates the inherent versatility of this strategy.

251

252


Me OR O

RO RO

OH

C13H25

O

RO

SiMe2OTf

2,6-Lutidine CH2Cl2, 0 °C to rt 79%

MOMO

121

RO RO

OR O

Me C13H25 Si

O O

RO

MOMO

122 1. Cat. [Mo]-I heptane, ∆

83%

2. PhLi, THF, 0 °C

O

HO HO

OH O OH

HN

14

O

C13H25

RO RO

OH

D,L-Glucosylceramide

Scheme 8.24

4 steps

OR O

OH O

C13H25

RO

(124)

MOMO

SiMe2Ph

123

Stereoselective synthesis of

D,L-glucosylceramide

(RO = TBSO).

Denmark and Yang demonstrated that vinylsiloxanes participate in a series of intermolecular silicon-assisted metal-catalyzed cross-couplings [49a,c]. Interestingly, the RCM of the alkoxysilane 125 (n = 0–2) with [Mo]-I furnished the vinylsiloxane 126, whereas the analogous process with Grubbs’s second-generation catalyst [Ru]-II was ineffective (Scheme 8.25). Sequential treatment of the vinylsiloxane 126 with TBAF followed by the requisite aryl halide and the palladium catalyst furnished the Z-alkenyl alcohols 127 in excellent yield. Interestingly, electron-poor and electron-rich aryl halides exhibit similar reactivity in the cross-coupling reactions and the steric effect from ortho-substituents is minimal, albeit for groups that are able to chelate the metal to slow the rate of reductive elimination. Finally, this process is general for five- through seven-membered vinylsiloxanes (n = 0–2), albeit the α-substituted derivatives (R1 ) undergo cross-coupling with significantly reduced reaction rates. 1 Me Me R Si O

Ph

n

PhH, rt 81–95%

R2

125

Me Me R1 Si O

Cat. [Mo]-II

Ph

2 n R

OH

Cat. Pd(dba)2 ArI, TBAF, rt 81–93%

Ph

126

R2 R1

n Ar 127

n = 0–2

Scheme 8.25

Palladium-mediated cross-coupling of vinylsiloxanes.

Denmark and Yang also reported the intramolecular variant, which provides a novel approach to the construction of a variety of medium ring carbocycles that contain a homoannular 1,3-diene (Equation 8.4) [49b,c]. This is a particularly interesting transformation given that cyclization reactions that afford medium-sized rings are generally disfavored entropically. The cyclic vinylsiloxanes 128 were prepared in a manner analogous to that described earlier using [Mo]-I catalyst, albeit with a pendant vinyl iodide. Preliminary studies demonstrated that conditions that


generally favor the intra- over the intermolecular reaction manifold, namely, high dilution or slow addition, were not optimal for this process. Additional studies demonstrated that the quantity of TBAF and the palladium catalyst were critical for optimal yields. The intramolecular palladium-catalyzed cross-coupling of the vinylsiloxanes 128 (n = 1, m = 0–6) in the presence of excess TBAF (10 equiv.) furnished the carbocyclic alcohols 129 in 55–72% yield. I

Me Me Si O m

Cat. [allylPdCl]2

n

m

TBAF, rt 55–72%

(8.4)

n

OH

128 m = 0–6, n = 1

129

Additional studies demonstrated that the TST-RCM in conjunction with the intramolecular silicon-assisted metal-catalyzed cross-coupling could be readily adapted to prepare cyclic ethers, which have proven extremely challenging to prepare using conventional methods. The process provided the key transformation in the total synthesis of the antifeedant (+)-brasilenyne, which is a nine-membered homodiene containing cyclic ether that was isolated from the digestive gland of the sea hare Aplysia brasiliana (Scheme 8.26) [50]. Silylation of homoallylic alcohol 130 followed by RCM with [Mo]-I catalyst furnished the cyclic vinylsiloxane 131 in 84% overall yield. Intramolecular silicon-assisted palladium-catalyzed cross-coupling of the vinylsiloxane 131 in the presence of excess TBAF afforded the cyclic ether 132 in 61% yield, which was converted to (+)-brasilenyne (133) in an additional six steps. I HO

1. VinylMe2SiCl Et3N, CH2Cl2

O Me

PMBO

2. Cat. [Mo]-II PhH, rt 84%

130

I Me Si

O Me

Cat. [allylPdCl]2

O Me

PMBO

HO O

TBAF, rt 61%

Me

PMBO 132

131

6 steps

Cl O Me (+)-Brasilenyne (133)

Scheme 8.26

Total synthesis of (+)-brasilenyne.

Vilarrasa et al. combined the aforementioned methodologies to install the E-trisubstituted alkene present in the cytotoxic macrolide amphidinolide X, which was isolated from Amphidinium dinoflagellate and possesses an unusual asymmetric diolide structure (Scheme 8.27) [51]. Although it was envisioned that the challenging trisubstituted C12–C13 olefin could be introduced via RCM, the attempted cyclization failed to directly provide the 16-membered macrocycle with a variety of

253

254


O

n -Pr

PMBO

HSiMe2Cl, CH2Cl2

Me

2. 135, Et3N, CH2Cl2, rt

Me Si

Me

1. [Cp*Ru(CH3CN)3]PF6

O

O

O O PMBO

Me Me

76%

134 O

Me

n -Pr

136

O OH

Cat. [Mo]-I PhH, rt

78%

Me 135 Me Me O

Me

SiMe3

Me

O O OTBS

Me

n -Pr

PMBO 138

1. MeLi, THF, 0 °C

O

2. TBSOTf, CH2Cl2

Me

O O

2,6-lutidine, –78 °C

Me Si

O PMBO

Me

90%

Me

n-Pr

137

1. NIS, (CF3)2CHOH, 0 °C 2. Me2Zn, Pd(Ph3P)4

89%

THF, 0 °C to rt

Me Me Me O

O O OTBS PMBO

Me

Me

Me 5 steps

Me

O O

O

O

n -Pr

Me

n-Pr

O O

Me 139

Scheme 8.27

Amphidinolide X (140)

Total synthesis of amphidinolide X.

catalysts and substrates with stereochemical variations. This prompted the introduction of the challenging trisubstituted alkene using the TST-RCM strategy in a fragment-coupling step. Hydrosilylation of the alkyne 134 with chlorodimethylsilane and Trost catalyst [Cp* Ru(CH3 CN)3 ]PF6 afforded the triorganochlorosilane, which was coupled with the secondary alcohol 135 to provide the acyclic vinylsiloxane 136 in 76% overall yield. Treatment of the acyclic vinylsiloxane 136 with Schrock catalyst [Mo]-I provided the cyclic variant 137 in 78% yield. The direct conversion of the vinylsiloxane 137 into the trisubstituted alkene 139 proved more problematic than anticipated, which prompted the development of a novel reaction sequence. Ring opening of the vinylsiloxane 137 with methyllithium followed by the protection of the secondary alcohol as the tert-butyldimethylsilyl ether afforded the vinylsilane 138 in excellent yield. Iododesilylation of the vinylsilane 138 with N-iodosuccinimide in (CF3 )2 CHOH followed by a Negishi coupling with dimethylzinc afforded the E-trisubstituted alkene 139 in 89% overall yield. The elaboration of this fragment using a five-step sequence completed the total synthesis of amphidinolide X (140). Overall, this provides another excellent example of the successful manipulation of a cyclic vinylsiloxane to afford substituted alkenes in a highly selective manner.


8.2.4.4 Related Applications and Developments In related studies, the TST-RCM of allylsiloxanes was featured as a homologation step in a novel approach to trans-fused oxepane polyethers [52]. Additionally, the enantioselective TST-RCM of prochiral alkenes has been developed using molybdenum and ruthenium chiral complexes with excellent enantioselectivities [53]. Finally, Yao developed a polymer-bound ruthenium carbene complex, which was both robust and recyclable enough to be utilized for the TST-RCM of eight-membered allylsiloxanes after being employed in an alternative RCM reaction [54]. Despite the advent of several new catalysts, Schrock catalyst remains the optimal metal-complex for vinylsiloxane RCM. 8.2.5 Enyne TST-RCM: Tri- and Tetrasubstituted Acyclic Dienes

Yao described the stereoselective construction of acyclic tri- and tetrasubstituted 1,3-dienes using the enyne TST-RCM to circumvent the problems with controlling chemo-, regio- and stereoselectivity (Scheme 8.28) [55]. Treatment of the enyne siloxane 141 with catalyst [Ru]-I furnished the diene 142, which was subjected to an oxidative cleavage to afford the diene 143 in excellent overall yield. Interestingly, this study demonstrates that a variety of substituents are tolerated, particularly on the alkyne (R2 ), in which the most challenging examples utilize ethylene gas to overcome low reactivity that extends the reaction times. Additional studies demonstrated that the method was also applicable to the preparation of seven-membered siloxanes and a tandem dienyne metathesis under high dilution conditions to provide a tetrasubstituted diene with Grubbs’s second-generation catalyst [Ru]-II. Although Lee et al. have continued to extend this methodology, it has still not been applied to natural product synthesis [56]. Me O Me Si

R1

Cat. [Ru]-I

Me O Me Si

R1

HO R2

KF, KHCO3, H2O2

CH2Cl2, ∆ 141

R2

142

MeOH/THF 68–88% (2 steps)

R1 R2

HO

143

R1 = Ph(CH2)2, CH2OH; R2 = Me, CH2OPiv, ...

Scheme 8.28

Enyne TST-RCM of enynesiloxanes.

8.2.5.1 Illudins Movassaghi et al. employed this strategy to prepare the cyclohexadiene moiety present in the semisynthetic analogs of illudin S, namely, (−)-acylfulvene and (−)-irofulven (Figure 8.14) [57]. The semisynthetic analogs were developed to overcome the severe toxicity exhibited by the naturally occurring agents. For example, (−)-irofulven is currently in phase II clinical trials for the treatment of ovarian, prostrate, and thyroid cancer. Although these targets were ultimately prepared using a bis-alkoxyalkynylsilane enyne TST-RCM, this example serves

255

256


HO

O

OH

Me

HO

O

Me

Me

Me Me

OH

Figure 8.14

Me

R

Illudin S, (−)-acylfulvene (R = H), and (−)-irofulven (R = CH2 OH).

to demonstrate the potential for total synthesis despite the inherent limitations associated with the functionalization of the siloxanes in complex molecules. Treatment of the enynesiloxane 144, which had been prepared through the silylation of the corresponding propargylic alcohol, with catalyst [Ru]-II afforded the cyclic diene 145 in 52–76% yield (Scheme 8.29). A critical feature of this approach is that it provides the 1,3-diene motif in good yield, and it is tolerant to a variety of substituents: for example, alkenes, hydroxyesters, and alkyl iodides. Unfortunately, the dienes 145 were particularly sensitive to chromatography on silica gel and alumina, which is evident from the attempted protodesilylation, which led to significant decomposition. Although the oxidation of the dienylsiloxane 145 provides the triol 146, the poor yield of this step in addition to the sensitivity of the substrates prompted the examination of the bis-alkoxyalkynylsilane to circumvent the oxidation reaction and thus provide a more viable route for the synthesis of these agents.

HO Me

Me Me Si O Cat. [Ru]-II R

HO Me

Me Me Si O KF, KHCO3, H2O2

PhH, 65 °C 52–76%

Me

OH

OH

MeOH/THF R Me

(R = CH2CHMe2, 27%)

145

144

HO Me

R Me 146

R = CH2CHMe2, (CH2)2CH CMe2, CH2OPiv, (CH2)4I, ...

Scheme 8.29

Enyne TST-RCM for the synthesis of (−)-acylfulvene and (−)-irofulven.


The TST-RCM provides a versatile synthetic strategy for target-directed synthesis as exemplified by the examples outlined herein. The TST-RCM provides homo- and cross-coupling reactions in both a stereospecific and stereoselective fashion, which is particularly useful for the construction of E- and Z-substituted alkenes, in addition to dienes. The advent of more reactive RCM catalysts that have exquisite functional group tolerance has increased the versatility of this process, which will dramatically increase the number of applications to the challenging motifs present in total synthesis applications. Furthermore the recent development of the enantioselective TST-RCM is likely to provide a very useful method for the asymmetric synthesis of challenging alkene-containing substrates for target-directed synthesis. The ability to circumvent the limitations associated with the assembly and functionalization of

References

the unsymmetrical bis-alkoxyalkynylsilanes and the organosiloxanes, respectively, will inspire the development of more creative solutions to these problems. Finally, in the absence of catalysts that can control the formation of E- and Z-alkenes, particularly Z-di- and E/Z-trisubstituted versions, the TST-RCM will provide a versatile alternative that will continue to gain favor in natural product synthesis. Acknowledgments

The author would like to acknowledge the intellectual and experimental contributions of his former coworkers in the metathesis subgroup. In particular, he would like to highlight the contributions of Jian Cui, Srinivasa Murthy, and Alexei Polosukhin. He is also grateful to Alen Cusak for his assistance in the preparation of this chapter. Finally, the National Institutes of Health (GM54623) is thanked for generous financial support, in addition to the numerous pharmaceutical companies for unrestricted funds. References 1 For a recent review on disposable teth-

2

3

4

5

6

7

ered transformations, see: Gauthier, D.R. Jr., Zandi, K.S., and Shea, K.J. (1998) Tetrahedron, 54, 2289–2338. (a) Bender, M.L. and Neveu, M.C. (1958) J. Am. Chem. Soc., 80, 5388–5391; (b) Page, M.I. (1973) Chem. Soc. Rev., 2, 295–323; (c) Kirby, A. (1980) J. Adv. Phys. Org. Chem., 17, 183–278. (a) Nishiyama, H., Kitajima, T., Matsumoto, M., and Itoh, K. (1984) J. Org. Chem., 49, 2298–2300; (b) Stork, G. and Kahn, M. (1985) J. Am. Chem. Soc., 107, 500–501. For a review on temporary silicon-tethered reactions, see: (a) Bols, M. and Skrydstrup, T. (1995) Chem. Rev., 95, 1253–1277; (b) Fensterbank, L., Malacria, M., and Sieburth, S.McN. (1997) Synthesis, 813–854. For a direct comparison of the temporary silicon tethered strategy to enzyme catalyzed reaction, see: Xin, Y.C., Mallet, J.-M., and Sinaÿ, P. (1993) J. Chem. Soc., Chem. Commun., 864–865. Fersht, A. (1999) Structure and Mechanism in Protein Science, W. H. Freeman and Co., New York. For selective deprotection of silyl ethers see: (a) Nelson, T.D. and Crouch, R.D. (1996) Synthesis, 9, 1031–1069; (b)

8

9

10

11

Crouch, R.D. (2004) Tetrahedron, 60, 5833–5871. Brook, M.A. (2000) Silicon in Organic, Organometallic and Polymer Chemistry, Wiley-Interscience, New York. For review on olefin metathesis see: (a) Grubbs, R.H., Miller, S.J., and Fu, G.C. (1995) Acc. Chem. Res., 28, 446–452; (b) Schuster, M. and Blechert, S. (1997) Angew. Chem. Int. Ed., 36, 2036–2056; (c) Grubbs, R.H. and Chang, S. (1998) Tetrahedron, 54, 4413–4450; (d) Armstrong, S.K. (1998) J. Chem. Soc., Perkin Trans. 1, 371–388; (e) Blechert, S. (1999) Pure Appl. Chem., 71, 1393–1399; (f) Phillips, A.J. and Abell, A.D. (1999) Aldrichim. Acta, 32, 75–89; (g) Fürstner, A. (2000) Angew. Chem. Int. Ed., 39, 3012–3043; (h) Schrock, R.R. and Hoveyda, A.H. (2003) Angew. Chem. Int. Ed., 42, 4592–4633; (i) Deiters, A. and Martin, S.F. (2004) Chem. Rev., 104, 2199–2238; (j) Nicolaou, K.C., Bulger, P.G., and Sarlah, D. (2005) Angew. Chem. Int. Ed., 44, 4490–4527; and pertinent references therein. Chatterjee, A.K., Choi, T.-L., Sanders, D.P., and Grubbs, R.H. (2003) J. Am. Chem. Soc., 125, 11360–11370. Fu, G.C. and Grubbs, R.H. (1992) J. Am. Chem. Soc., 114, 5426–5427.

257

258

8 Temporary Silicon–Tethered Ring–Closing Metathesis Reactions 12 Levisalles, J., Rudler, H., and Villemin,

13 14

15 16 17

18

19 20

21

22

23

24 25

26

D. (1979) J. Organomet. Chem., 164, 251–257. Evans, P.A. and Murthy, V.S. (1998) J. Org. Chem., 63, 6768–6769. Amador, M., Ariza, X., Garcia, J., and Ortiz, J. (2004) J. Org. Chem., 69, 8172–8175. Hoye, T.R. and Promo, M.A. (1999) Tetrahedron Lett., 40, 1429–1432. Lobbel, M. and Köll, P. (2000) Tetrahedron: Asymmetry, 11, 393–396. (a) Huang, J., Stevens, E.D., Nolan, S.P., and Petersen, J.L. (1999) J. Am. Chem. Soc., 121, 2674–2678; (b) Scholl, M., Trnka, T.M., Morgan, J.P., and Grubbs, R.H. (1999) Tetrahedron Lett., 40, 2247–2250; (c) Briot, A., Bujard, M., Gouverneur, V., Nolan, S.P., and Mioskowski, C. (2000) Org. Lett., 2, 1517–1519. Scholl, M., Ding, S., Lee, C.W., and Grubbs, R.H. (1999) Org. Lett., 1, 953–956. Evans, P.A. and Murthy, V.S. (1999) Tetrahedron Lett., 40, 1253–1256. Boiteau, J.-G., Van de Weghe, P., and Eustache, J. (2001) Tetrahedron Lett., 42, 239–242. (a) Gierasch, T.M., Chytil, M., Didiuk, M.T., Park, J.Y., Urban, J.J., Nolan, S.P., and Verdine, G.L. (2000) Org. Lett., 2, 3999–4002; (b) Harrison, B.A. and Verdine, G.L. (2001) Org. Lett., 3, 2157–2159; (c) Harrison, B.A., Gierasch, T.M., Neilan, C., Pasternak, G.W., and Verdine, G.L. (2002) J. Am. Chem. Soc., 124, 13352–13353. Evans, P.A., Cui, J., and Buffone, G.P. (2003) Angew. Chem. Int. Ed., 42, 1734–1737. Van de Weghe, P., Aoun, D., Boiteau, J.-G., and Eustache, J. (2002) Org. Lett., 4, 4105–4108. Marjanovic, J. and Kozmin, S.A. (2007) Angew. Chem. Int. Ed., 46, 8854–8857. Hooper, A.M., Dufour, S., Willaert, S., Pouvreau, S., and Pickett, J.A. (2007) Tetrahedron Lett., 48, 5991–5994. Casey, E.M., Teesdale-Spittle, P., and Harvey, J.E. (2008) Tetrahedron Lett., 49, 7021–7023.

27 Evans, P.A., Cui, J., Gharpure, S.J.,

28

29

30 31

32 33 34

35 36

37

38 39 40 41

42

Polosukhin, A., and Zhang, H.-R. (2003) J. Am. Chem. Soc., 125, 14702–14703. Brown, L.J., Spurr, I.B., Kemp, S.C., Camp, N.P., Gibson, K.R., and Brown, R.C.D. (2008) Org. Lett., 10, 2489–2492. Hoye, T.R., Eklov, B.M., Jeon, J., and Khoroosi, M. (2006) Org. Lett., 8, 3383–3386. Gaich, T. and Mulzer, J. (2005) Org. Lett., 7, 1311–1313. Van de Weghe, P., Bourg, S., and Eustache, J. (2003) Tetrahedron, 59, 7365–7376. Hara, O., Sugimoto, K., and Hamada, Y. (2004) Tetrahedron, 60, 9381–9390. Postema, M.H.D. and Piper, J.L. (2002) Tetrahedron Lett., 43, 7095–7099. (a) Diver, S.T. and Giessert, A.J. (2004) Chem. Rev., 104, 1317–1382; (b) Maifeld, S.V. and Lee, D. (2005) Chem. Eur. J., 11, 6118–6126; (c) Van de Weghe, P., Bisseret, P., Blanchard, N., and Eustache, J. (2006) J. Organomet. Chem., 691, 5078–5108; (d) Chattopadhyay, S.K., Karmakar, S., Biswas, T., Majumdar, K.C., Rahaman, H., and Roy, B. (2007) Tetrahedron, 63, 3919–3952. Hansen, E.C. and Lee, D. (2006) Acc. Chem. Res., 39, 509–519. (a) Grimm, J.B. and Lee, D. (2004) J. Org. Chem., 69, 8967–8970; (b) Grimm, J.B., Otte, R.D., and Lee, D. (2005) J. Organomet. Chem., 690, 5508–5516; (c) Kim, Y.J., Grimm, J.B., and Lee, D. (2007) Tetrahedron Lett., 48, 7961–7964. (a) Kim, Y.J. and Lee, D. (2006) Org. Lett., 8, 5219–5222; (b) Kim, Y.J. and Lee, D. (2007) Org. Lett., 9, 3859. Mukherjee, S. and Lee, D. (2009) Org. Lett., 11, 2916–2919. Chang, S. and Grubbs, R.H. (1997) Tetrahedron Lett., 38, 4757–4760. Meyer, C. and Cossy, J. (1997) Tetrahedron Lett., 38, 7861–7864. (a) Cassidy, J.H., Marsden, S.P., and Stemp, G. (1997) Synlett, 1411–1413; (b) Miles, S.M., Marsden, S.P., Leatherbarrow, R.J., and Coates, W.J. (2004) J. Org. Chem., 69, 6874–6882. (a) Miles, S.M., Marsden, S.P., Leatherbarrow, R.J., and Coates, W.J.

References

43

44

45 46

47

48

49

(2004) Chem. Commun., 2292–2293; (b) Akindele, T., Marsden, S.P., and Cumming, J.G. (2005) Org. Lett., 7, 3685–3688. ´ (a) Jiménez-González, L., Alvarez-Corral, M., Mu˜ noz-Dorado, M., and Rodr´ıguez-Garc´ıa, I. (2005) Chem. Commun., 2689–2691; (b) Jiménez-González, L., Garc´ıa-Mu˜ noz, ´ S., Alvarez-Corral, M., Mu˜ noz-Dorado, M., and Rodr´ıguez-Garc´ıa, I. (2007) Chem. Eur. J., 13, 557–568. Ahmed, M., Atkinson, C.E., Barrett, A.G.M., Malagu, K., and Procopiou, P.A. (2003) Org. Lett., 5, 669–672. Li, F. and Miller, M.J. (2006) J. Org. Chem., 71, 5221–5227. Tung, J.C., Chen, W., Noll, B.C., Taylor, R.E., Fields, S.C., Dent, W.H. III, and Green, F.R. III (2007) Synthesis, 2388–2396. Ahmed, M., Barrett, A.G.M., Beall, J.C., Braddock, D.C., Flack, K., Gibson, V.C., Procopiou, P.A., and Salter, M.M. (1999) Tetrahedron, 55, 3219–3232. Barrett, A.G.M., Beall, J.C., Braddock, D.C., Flack, K., Gibson, V.C., and Salter, M.M. (2000) J. Org. Chem., 65, 6508–6514. (a) Denmark, S.E. and Yang, S.-M. (2001) Org. Lett., 3, 1749–1752; (b) Denmark, S.E. and Yang, S.-M. (2002) J. Am. Chem. Soc., 124, 2102–2103; (c) Denmark, S.E. and Yang, S.-M. (2004) Tetrahedron, 60, 9695–9708.

50 (a) Denmark, S.E. and Yang, S.-M.

51

52 53

54 55 56

57

(2002) J. Am. Chem. Soc., 124, 15196–15197; (b) Denmark, S.E. and Yang, S.-M. (2004) J. Am. Chem. Soc., 126, 12432–12440. Rodr´ıguez-Escrich, C., Urp´ı, F., and Vilarrasa, J. (2008) Org. Lett., 10, 5191–5194. Brouard, I., Hanxing, L., and Martin, J.D. (2000) Synthesis, 6, 883–892. (a) 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; (b) Kiely, A.F., Jernelius, J.A., Schrock, R.R., and Hoveyda, A.H. (2002) J. Am. Chem. Soc., 124, 2868–2869; (c) Funk, T.W., Berlin, J.M., and Grubbs, R.H. (2006) J. Am. Chem. Soc., 128, 1840–1846. Yao, Q. (2000) Angew. Chem. Int. Ed., 39, 3896–3898. Yao, Q. (2001) Org. Lett., 3, 2069–2072. (a) Miller, R.L., Maifeld, S.V., and Lee, D. (2004) Org. Lett., 6, 2773–2776; (b) Maifeld, S.V., Miller, R.L., and Lee, D. (2004) J. Am. Chem. Soc., 126, 12228–12229; (c) Park, S., Kim, M., and Lee, D. (2005) J. Am. Chem. Soc., 126, 9410–9415; (d) Park, S. and Lee, D. (2007) Synthesis, 2313–2316. Movassaghi, M., Piizzi, G., Siegel, D.S., and Piersanti, G. (2006) Angew. Chem. Int. Ed., 45, 5859–5863.

259

261

9 Metathesis Involving a Relay and Applications in Natural Product Synthesis Thomas R. Hoye and Junha Jeon

9.1 Introduction

As each of the chapters in this monograph makes clearly evident, alkene (and alkyne) metathesis is one of the most powerful and versatile synthetic methodologies in all of organic chemistry. In this chapter discussion is focused on (ruthenium–alkylidene-mediated) ring-closing metathesis (RCM) reactions involving a relay process that is initiated at a remote alkene site within the substrate. The chapter is organized as follows: the concept is introduced in Section 9.1; the earliest discoveries and reports that established the viability of the strategy are presented in Section 9.2; uses of relay metathesis in settings not directly motivated by natural product synthesis are presented in Section 9.3; applications in natural product synthesis are discussed in Section 9.4; instructive examples in which the desired relay event is thwarted are presented in Section 9.5 and concluding comments are offered in Section 9.6. 9.1.1 The Relay Concept

RCM (1 to 2, Scheme 9.1) is one of the most common and important types of olefin metathesis reaction. Accommodation of considerable structural complexity within ‘‘A’’ in RCM substrates 1 is one of the hallmarks of this powerful and broadly applicable reaction. Nonetheless and despite significant advantages available by modification of external factors (for example, catalyst selection, additive use, solvent choice, or ethylene concentration), not every diene substrate 1 can be successfully cyclized to cycloalkene 2. Because of the advent of relay ring-closing metathesis (RRCM, 4a to 4b to 3 to 2, Scheme 9.1), chemists now have an alternative for resuscitating/redirecting these otherwise lifeless/recalcitrant systems when such a limitation is encountered. More specifically, altering the parent diene substrate 1 by installation of an extender arm that includes a tether element ‘‘T’’ provides the opportunity to Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts. Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32440-8

262

9 Metathesis Involving a Relay and Applications in Natural Product Synthesis

A

A

RCM

R1 R2

RRCM

R1 R2 R1

1

A

RRCM R2

2

[Ru] 3

A R1 R2

T X

T

4a: X = CH2 4b: X = [Ru] Scheme 9.1 Generic RCM (1 to 2) versus relay RCM (RRCM, 4a to 2 via 4b and 3) processes.

1) load ruthenium metal and its associated ligands (designated herein as [Ru]) to a reactive terminal alkene (4a to 4b); 2) relay [Ru] by an initial RCM event that ejects a disposable cycloalkene (generally five-membered) to give 3; and 3) complete the cyclization to the desired product 2. In practice, the relay tether (T) has been C(CO2 R)2 , O, CH2 , OSiR2 , or C(=O). As the examples discussed in this chapter show, there is a variety of reasons why a relay-modified substrate might be used and/or why it might lead to a successful cyclization outcome. 9.1.2 Basic Tenets of RCM

A few fundamental principles that directly impact the level of efficiency associated with every specific metathesis reaction are worth stating at the outset of this discussion. As with any catalytic process, the rate of the primary reaction versus the lifetime of the working catalyst system – in effect, the turnover number (TON) – is a critical parameter. In the field of alkene/alkyne metathesis, there has been considerable evolution in the design and implementation of the catalysts (or precatalysts) that are used. Often, these advances have been driven by the desire to increase the TON by improving the longevity of the catalytic intermediates (catalyst robustness toward destructive side reactions) or speeding up the rate of the slow step(s) in each metathesis cycle (or both). The rate-limiting event in substrate turnover is often the rate of loading of the metal onto a substrate alkene, although the subsequent rate of cyclization of the initially loaded species is sometimes the slowest stage. An alternative to catalyst (and reaction condition) alteration for improving either the TON and/or the selectivity of a given metathesis reaction is substrate modification. Of course, requirements dictated by the structure of the desired product place obvious constraints on this approach. On the other hand, two of the four alkene carbon atoms in a simple RCM substrate are expendable (see • and in 1, Scheme 9.1), since they are absent in the product. This means that, in principle, those two terminal atoms can bear substituents other than hydrogen (and, correspondingly, that the by-product of the

9.2 Early Relay Metathesis Discoveries

263

reaction can be a second alkene other than ethylene). In retrospect, it was this seemingly simple notion that opened our eyes to the possibility of using a relay fragment to advantage as we set out to demonstrate RRCM and some of its strategic benefits. A few comments about yields of reaction product(s) are in order. A reported chemical yield can represent several different things; it often, necessarily, has a substantial error bar, especially when the experiment has been performed on a small scale. This is often the case with complex and precious substrates: for example, those in hand at the closing stage of a total synthesis effort. Even when the yield value is accurate, it is often a combination of inherent imperfections in both the reaction chemistry and the purification/handling routines. Similar issues affect the way one interprets the catalyst loading in metal-mediated reactions. Although we have provided the yields and catalyst loadings reported for the metathesis reactions described here, we encourage readers to be mindful of the above points and not to permit yield vagaries to overshadow the more important conceptual and strategic lessons that are revealed and taught by each of the case studies described. Importantly, in every example the product shown represents the major (and often sole) reaction path that was reported by the original investigators.

9.2 Early Relay Metathesis Discoveries

The earliest demonstration of a RRCM reaction in our research group at the University of Minnesota was carried out by Hongyu Zhao. He took up the challenge presented by the low reactivity of 1,1-disubstituted ethylenes toward the early generation ruthenium alkylidenes [1], most commonly represented by diene 5, and its failure to cyclize to 6 (Scheme 9.2) [2]. The problem was successfully solved by preparing and then cyclizing the relay-armed triene substrate 8 [3–5]. In this RRCM experiment, initiator [Ru]-I was used, since the experiment predated the report of E Me E E

Me

Me E E

[Ru]-I

[Ru]

CH2Cl2, 45 °C

5 : E = CO2Et

Me 6

Me

CH2Cl2, 45 °C

Me

66%

8 Scheme 9.2

Me E E

[Ru]-I (26 mol%)

Me [Ru]

Our first example of RRCM: closure of 8 to 7.

9

Me 7

Relay Me E E

E

Me

Me

264


the discovery of [Ru]-II. (Although these ruthenium complexes are commonly referred to as catalysts, it is more correct to refer to them as metathesis initiators or precatalysts, since they are not recovered unchanged once they have performed their duty.) This showed that by use of a relay strategy, net loading of the metal to a substrate that is otherwise inert, in this case to produce 6 (via 9), is achievable. Independent efforts by the Piscopio, Lee, and Porco groups at Array Biopharma, the University of Wisconsin, Madison, and Boston University, respectively, also laid the base for further development of relay metathesis. In the first of these reports, the RRCM concept was presented as shown in Scheme 9.3 [6]. The example shown in Scheme 9.4 constitutes the first case of a relay cross-metathesis (relay-CM) process [7]. Reaction to give 16 is faster and more efficient for the relay-activated substrate 17 than for the ‘‘parent’’ CM partner 14. In the synthesis of oximidine III (23), Porco et al. faced the need to cyclize a polyene like 18 to access macrolide 19, but this process failed for either (E)- or (Z)-18 (Scheme 9.5) [8]. Use of the relay-activated substrate 20 and the second-generation O HO R1

H

() n R3 C R2

A′ D

B′

Grubbs or Schrock's catalyst

C′ A

O RO [M] R1

B

10

O

A′ H

() n

RO R3

R2

D C

R1 12

Scheme 9.3

() n R3 C R2

A′ D

B′ C′ A B

11

B′ C′

H

O +

A H

RO R3

R2

() n

B

D C′

Diastereoselective RCM

C

R1 13

Generic representation of prospects for RRCM processes.

O Ph [Ru]-II (10 mol%) CH2Cl2, rt BnO

14

O

OBn 15

O

OBn

Ph 16

Ph

O 17

Scheme 9.4 The first relay cross-metathesis (relay-CM) reaction: cross-coupling of donor ene–yne 17 with acceptor alkene 15.

(E /Z = 1 : 7, 34% from 14) (E /Z = 1 : 4, 63% from 17)

9.2 Early Relay Metathesis Discoveries OPMB TBSO

OPMB

O

TBSO OH

265

O

[Ru]-II or [Ru]-III

OH

O

O

19

Me 18 OPMB TBSO

O OH

[Ru]-III (10 mol%) TBSO CH2Cl2 50 °C O 71%

OPMB

OPMB TBSO

O OH

OH

O

Me

Me 20

O

Me 21

[Ru]

O [Ru]

22

H N O TBSO

O OH

N OMe

O Oximidine III (23)

Scheme 9.5

The first successful application of RRCM in a natural product synthesis.

Hoveyda–Grubbs initiator [Ru]-III nicely solved this problem, presumably because ruthenium loading adjacent to the oxirane was now viable via the relay of 21 to 22. This constituted the first report of a successful application of an RRCM reaction in the context of a natural product total synthesis. Other early studies of RRCM applications in our labs were carried out by Danielson [9], Tennakoon [4, 10], Wang [4, 11], and Jeffrey [4, 12]. RCM of the silicon-tethered substrates 25 and 28 to give the hindered (Z)-trisubstituted alkene, an intermediate for use in our synthesis of peloruside A (29), was successful in both cases (Scheme 9.6) [5, 10]. However, it was difficult to prepare 25 in high diastereomeric purity. More specifically, the two enantiomers of the secondary alcohol precursor (±)-24 were barely discriminated by several lipases that we examined. By contrast, citronellene-derived alcohol 27, a 1 : 1 mixture of epimers at C15, could be kinetically resolved quite efficiently. Thus, the relay tether served an auxiliary but nonetheless important role, merely because of its size. In a model study of a potential application of RRCM for formation of the C4–C5-trisubstituted alkene in oocydin A (33), the substrates 30a and 30b were each successfully cyclized to the 13-membered lactones 32a and 32b, respectively

266

9 Metathesis Involving a Relay and Applications in Natural Product Synthesis [Ru]-II (45 mol%) CH2Cl2, 65 °C 19%

Et Me

O

OH

O SiPh2 CN

Et

CN

Me Me

CN

27

HO OH OH O

HO

Me

CN 28

O 1

O SiPh2

15

CN

OMe

5 8

OMe

[Ru]-II (10 mol%) CH2Cl2, 65 °C 58%

(+)-Peloruside A (29)

RRCM permitting the use of an otherwise inefficient enzymatic resolution.

[Ru]-II (10 mol%) N2 sparge O 1–5 min

O Cl 5

12

Me

15

Scheme 9.6

26

O OH

Me

O

11

Et Me

15

MeO

O SiPh2

Me

O

Me

18

25 versus

OH

18

15

(±)-24

Me

Et

Me

15

O Me 4 X

Y

Me

Toluene 110 °C 50–70%

30a,b

O

O

Cl Me

5

O Me

[Ru] 4 X

Cl

O

5 O

O versus

E

Y

Me 4

H

O O

O

O

5 Z

Me 4

32a (from 30a)

31a,b

a: X Y = b: X Y =

O

Cl

32b (from 30b)

O H OH Me

Cl

H 5

CO2H

O Me 4 OAc

Scheme 9.7

Oocydin A (33)

RRCM through alkylidenes that are otherwise difficult to access (31a and 31b).

(Scheme 9.7) [5, 11]. This demonstrated for the first time the ability to initiate a key cyclization through either an electron-deficient β-ketoalkylidene (31a) or an allylidene (31b) complex by the strategic design of the relay substrate. Interestingly, the E/Z selectivities with respect to the newly formed 4,5 -bond were absolute and opposite for the two substrates. Another example of a successful cyclization via an alkylidene adjacent to an electron-deficient functional group is the cyclization of 34 to 35 with [Ru]-I (Scheme 9.8) [4, 5]. The complementary closure of 36 to the same product

9.2 Early Relay Metathesis Discoveries O

[Ru]-I (10 mol%) O

O Me Me

CH2Cl2 45 °C

Me

59%

34

Scheme 9.8

O

Me

[Ru]-I O (10 mol%) Me

O

Me Me CH2Cl2 45 °C Me 59%

35

O

267

O Me Me

Me

Me

36

37

Access to a carboalkoxyalkylidene complex (from 34).

demonstrates versatility, and it is noteworthy that the parent, non-relay-activated diene 37 was unreactive under the conditions employed. Presumably, in the absence of a relay tether, ruthenium loading to either alkene in 37 was simply too slow relative to the initiator/catalyst lifetime. Finally, a different type of complementarity is clearly evident from the RRCM results summarized in Scheme 9.9. Ene–yne–ene cascade cyclization (for a non-relay, directed example see [13]) of either of the isomeric, relay-activated substrates 38 or 41 gave either isomeric bicyclic ether 39 or 40, respectively, each with very high regiochemical selectivity [5, 12]. This example shows the ability to use RRCM to dictate the formation of different constitutional isomers merely by relocation of the relay tether. That is, the directionality (‘‘endedness’’) of the process from ‘‘right-to-left’’ versus ‘‘left-to-right’’ can be controlled at will. Interestingly, the parent, non-relay-activated ene–yne–ene substrate (not shown) gave a nearly 1 : 1 product ratio, and the use of [Ru]-I rather than [Ru]-II gave greater regioselectivity for each of 38 and 41. Although the above earliest examples were described by various terminologies (‘‘RRCM’’ [4], ‘‘metathesis trigger’’ [6], ‘‘RCM–RCM’’ [8], or ‘‘catalyst delivery vehicle’’ [7]), these words all converge to the same concept. Relay metathesis ‘‘involves the design of substrates that, in effect, permit[s] one to dictate the sequence of metathesis events by choreographing the metal atom . . . through the individual steps’’ [5] of the metathesis cascade. The above, first-generation relay metathesis experiments were summarized in the earliest review of the subject [14]. O

Me O Left -to-right

Me

39/40 = 26 : 1 (77%) O

38 [Ru]-I (13 mol%)

O

O

+

Me

Me Me

Me 39

O

[Ru]-I (14 mol%)

O

40

39/40 = 1 : 45 (78%)

Me 41

Scheme 9.9 Complementary product constitution by controlling the ‘‘directionality’’ of ene–yne–ene RRCM.

Right u -to-left l

Me O

268


9.3 Examples of Relay Metathesis Directed at Targets Other than Natural Products

In their efforts to implement a strategy by which the chirality in the menthol-derived auxiliary in 42 could be transmitted into nonracemic heterocycles like 43, Spino et al. sought to cyclize the very hindered parent substrate 42a via RCM (Scheme 9.10), but the latter compound proved to be unreactive [15]. Substrate modification to 42b and its subsequent RRCM closure using the [Ru]-VI initiator resulted in formation of lactam 43 (via an electron-deficient alkylidene intermediate); the remainder of the product was the parent diene 42a. We refer to this type of process as truncation [5], and it results when the RCM cyclization rate of the relayed ruthenium intermediate is slow relative to the competing bimolecular process by which another molecule of substrate, 42b here, cross-metathesizes with net transfer of a methylene group. That obstacle was overcome here by an insightful modification of the relay tether involving the addition of one more substituent. Specifically, the use of the substrate analog 42c resulted in the formation of 43, with impressively high efficiency. Arguably, substrates 42b and 42c, whose cyclization was absolutely enabled by the relay strategy, might be the most demanding ever to have been successfully ring-closed by metathesis. Fujiwara et al. conceived a plan for the synthesis of nonracemic oxacycles like 47 that capitalized on sequential chirality transfer and then RRCM (Scheme 9.11) [16]. In the first stage, an Ireland–Claisen rearrangement of the serine-derived allylic ester 44 produced 45, thereby simultaneously repositioning both alkene and stereogenic carbon centers. This RRCM substrate was cyclized using [Ru]-II to give 47 (via 46). Grandbois and Collins have used the asymmetric initiator (R,R)-[Ru]-XIV to kinetically resolve a racemic sample of the doubly armed, C2 -symmetric tetraene (±)-48 (Scheme 9.12) [17]. Interestingly, the enantiomeric excess of the [7]helicene BnO

BnO OTIPS

Me

Me H NH R

Me

O

OTIPS

[Ru]-VI (10 mol%) Toluene (0.005 M) D, 15 min

Me

NH O

Me 43

R 42a

H

63% (from 42b)

42b 42c

0% (from 42a)

Me

Scheme 9.10

91% (from 42c) RRCM activation to allow closure of the remarkably hindered substrates 42.

9.4 Examples of Relay Metathesis Motivated by Natural Product Synthesis

Ireland – Claisen

O RO

O

O [Ru]-II (10 mol%) MeO2C

BnO Pr

BnO N

O O

44 Scheme 9.11

MeO2C

269

O

N 45

DCE 80 °C 89%

MeO2C

O

BnO

BnO [Ru]

O

O

Pr

46

47

Sequential chirality transfer and RRCM of a serine derivative.

Ph

CH2Cl2, rt, 2 h 53%

(±)-48

Ph

i -Pr N N i- Pr Cl Ph Ru H Cl PCy3

(R,R )-[Ru]-XIV (4 mol%)

49

(R,R )-[Ru]-XIV

Scheme 9.12 Use of RRCM and an asymmetric initiator to kinetically resolve the helically chiral substrate (±)-48.

49 that was produced was higher when the RRCM cyclization was used compared to the case of the parent, non-relay-activated, divinylbiphenanthrene substrate.


In the course of studying the effects of noncovalent interactions on various RCM reactions to produce macrocycles, Collins et al. showed that the relay-armed substrate 50 closed efficiently to the macrocycle 52, whereas the parent diene 53 gave mostly dimer 54 (Scheme 9.13) [18]. The ability to direct the main cyclization event via alkylidene 51 (vs that derived by loading the metal onto the more reactive monosubstituted alkene in 53) substantially affected the outcome. Motivated by the structures of various monomeric and dimeric members of the longithorone family [for example, longithorone C (57) and longithorone I (60)], these investigators expanded their studies (Scheme 9.14) [19]. Hence, the RRCM closure of pentaene 55 by use of the [Ru]-VIII initiator gave the triene 56 [19b], which contains many of the core structural elements of longithorone C (57). The ene–yne RRCM of substrate 58a or 58b using [Ru]-III gave diene 59a or 59b, respectively [19c]. Trienomycin A (64) and the related geldanamycin (65) are ansamycin antibiotics containing metacyclophane–macrolactams. With an eye toward preparation of structurally simpler analogs, Peng and Blagg synthesized trienes 61a and 61b and studied their RCM reactivities (Scheme 9.15) [20]. Upon being exposed to [Ru]-I,

270


O F F

O F

Me

F O

F O O

Me

[Ru]-II (10 mol%) CH2Cl2 reflux, 15 h 68%

F F

F

F

O F

Me

F O

53 Scheme 9.13

F

F O

[Ru]-II (10 mol%)

F O O

F F

Me

F

F O O

52

O F

O Me

Me

F O

Me

O

F O

51

CH2Cl2 reflux, 15 h 27 – 67%

F O O

F

Me

50

F

[Ru]

O

F

F F

F O O FO

F O O

54 Use of RRCM to divert dimerization (53 to 54) to cyclization (50 to 52).

the former gave only the RRCM product 62, whereas the methylated analog 61b gave a preponderance of the macrocyclic diene 63b. A combination of factors likely contributes to this different outcome: first, the metal relay event is expected to be slower in 61b versus 61a and, second, since either of these two products can also arise from initial alkylidene formation at C8 in the nonenyl amide moiety, the non-relay mode of cyclization to give 62 should also be slower from 61b than from 61a. In the course of their total synthesis of (−)-mucocin (71), Crimmins et al. targeted intermediate dihydrofurans like 67 (Scheme 9.16) [21]. Earlier, they had established that trienes like 66 gave poor regioselectivity for formation of the desired 67 versus the isomeric six-membered dihydro-2H-pyran 68 [22]. This result is consistent with the competitive cyclizations, in either direction, suggested by arrows a versus b in 66. This problem was solved by the cleverly conceived use of substrate 69, which underwent RRCM with high efficiency (via c) to give the desired dihydrofuran 70. The key step in the total synthesis of dihydro-epi-deoxyarteannuin B (73) reported by Dudley et al. involved metathesis closure of the hindered alkene in 72 [23]. This was first attempted, unsuccessfully, with the parent diene 72 (Scheme 9.17). CM, presumably with benzylidene fragments from the initiator, was observed. It is likely that obligate loading of the metal to either of the hindered alkene carbons (• or ) in 72 was simply too difficult. This problem was nicely and efficiently solved through the RRCM of analog 74. Here, the relay event provided for insertion of the metal onto carbon and the subsequent RCM. In their total synthesis of (−)-archazolid B (76), Trauner et al. successfully ring-closed the relay-activated polyene 75 at a late stage via an intermediate ruthenium allylidene complex (at carbon ) (Scheme 9.18) [24]. They designed this RRCM event into the synthesis plan from the outset, in part to minimize the


Me

Me Me

C6F5 Me

[Ru]-VIII (10 mol%) Ti(Oi- Pr)4 (5 equiv.)

C6F5

Me

Me

Me

O O


O

Me O

O O

271

O Me

55

Me

56

Longithorone C (57)

Me O

O

[Ru]-III (20 mol%)

Me

Aryl

Toluene (0.4 × 10−4 M) reflux, 2 h

O O O F F 58a:

F

Aryl =

O

Me

Aryl

O

O

Me O

OHC

H

O O 59 35% (from 58a) 59% (from 58b)

Me

O

F Me F

O

CF3 58b:

Longithorone I (60)

Aryl = CF3

Scheme 9.14

RRCM cyclizations relevant to longithorones 57 and 60.

possible complication of initial activation at C21 (carbon •) in 75, from which they speculated that undesired cyclization to excise a δ-valerolactone might occur. Kim et al. used the relay-CM strategy as a key step in the construction of the conjugated ene–yne in (+)-scanloenyne (79) (Scheme 9.19) [25]. The relay-activated partner 77, under the action of [Ru]-XII, was successfully cross-metathesized with the terminal alkene in 78 to give the ene–yne in 79 with high Z-selectivity, a type of stereocontrol first demonstrated by Lee [7a]. In a ploy to distinguish the pair of diastereotopic vinyl groups in substrate 80, Burke et al. used RCM to convert that triene to the bicyclic ketal 81, a key intermediate in their synthesis of (+)-didemniserinolipid B (85) (Scheme 9.20) [26]. They were badgered by the secondary formation of 82, via subsequent CM of the primary product 81 with the stoichiometric amount of styrene that was coproduced by the initial RCM event. By insightful incorporation of the designer relay auxiliary seen in substrate 83, they changed the profile of the metathesis

272


R

NH O

[Ru]-I (6 mol%)

NH

+

O

CH2Cl2, rt

61a: R = H

R

NH

O

63a: R = H (0%)

62 (86% from 61a)

61b: R = Me

63b: R = Me (37%)

(26% from 61b) OH

O OMe

O Me

Me

NH Me

HO

O

O Me HN

O O

O

Me

OMe HO

OMe O H2N

Trienomycin (64)

Scheme 9.15

N H

OMe Me

O

Me

Geldanamycin (65)

Competitive RCM and RRCM cyclizations to trienomycin A (64) analogs.

by-products to ethylene and indene 84, neither of which eroded the vinyl group in 81. The efficiency of this RRCM reaction was excellent. Njardarson et al. recently reported the RRCM cyclization of 86 to the inside–outside bicyclic triene 87 (Scheme 9.21), an analog of 89 that, in turn, is an atropisomer of the natural product hypoestoxide (88) [27]. This is another example in which the workers wisely designed the RRCM into their de novo plans, presumably envisioning that closure of the disubstituted ruthenium alkylidene at C6 to the less substituted alkene carbon at C5 (product numbering) had a greater likelihood of success than the alternative. Parker et al. developed an effective method for the stereocontrolled synthesis of a key intermediate in a projected discodermolide (94) synthesis that involved RCM closure to the dihydrooxasiline 91 and its subsequent conversion to the acyclic vinyl iodide 93, the C9–C14 subunit of 94 (Scheme 9.22) [28]. Attempts to cyclize the hindered triene 90 with either [Mo]-I or [Ru]-II failed. The steric hindrance in 90 is a direct consequence of the strategy used for the otherwise efficient synthesis of this nonracemic substrate, which involved a [2,3]-Wittig rearrangement. A solution to the metathesis cyclization was discovered in the form of the [Ru]-II-promoted RRCM of tetraene 92, which was also formed via a [2,3]-Wittig rearrangement and which produced 91 with extremely high efficiency. The success of this RRCM was due at least in part to the fact that the site of initiation (carbon •) strongly favored formation of only the six-membered oxasilacycle. Initiation at the vinyl


273

b

O

R2O

RHC=[Ru]

a

O R1

+

R1

OR2

OR2

67 via a

66

R1

O 68 via b

O c

[Ru]-II (10 mol%)

TIPS O

TIPS

C6H6, 80 °C

TESO

O OTES

69

70 87% (after the removal of TES)

OH

OH O

O

Me O

C10H21

OH

OH

O 71

Scheme 9.16

Me

RRCM as a key step in the synthesis of (−)-mucocin (71).

Me

RRCM [Ru]-II (10 mol%)

Me

RCM attempts H

O

H

Me

O

Me

72 Scheme 9.17

Me

O O

CH2Cl2 40 °C 74 – 90%

Me

Me s

H

O

Me

O

Dihydro-epi - deoxyarteannuin (73)

74

RRCM in the synthesis of dihydro-epi-deoxyarteannuin B (73).

silane (carbon ) would provide the opportunity for competitive formation of fiveversus six-membered cyclization products. Meyer, Cossy, et al. recently described the use of an RRCM to close the dihydropyranone ring present in phoslactomycin B (97) [29]. Specifically, the tetraene 95 was efficiently converted to 96 using [Ru]-II (Scheme 9.23), a key step in their formal total synthesis of 97. In this case, cyclization of the non-relay armed parent triene could be achieved, albeit much more slowly than the closure of 95. As a consequence, a substantially higher initiator loading [Ru]-II was required to achieve comparable conversion efficiency in the absence of the relay activation unit. This represents another type of advantage that can accrue from use of RRCM.


274

Me

Me

Me

MeO

Me OTBS Me Me

O

O O

O S Me Me

(−)-Archazolid B (76)

75 Scheme 9.18

O N

MeHN

Me

OH

Me Me

2. HCO2H/H2O/THF (84%)

S Me

O

Me

Me

N O

Me OH

1. [Ru]-II (20 mol%) toluene (27%)

O

Me

MeHN

OH Me

21

Me

MeO

RRCM closure via an allylidene in the synthesis of (−)-archazolid B (76).

In parallel with the Parker example shown in Scheme 9.22, the stereocontrolled construction of 95 also capitalized on a [2,3]-Wittig rearrangement to create the C4–C5 bond. Lee et al. have used a relay metathesis trigger to further enable their studies of metallatropic [1,3]-shifts. For example, both of the relay-activated diene–diynes 98 underwent CM with (Z)-1,4-diacetoxy-2-butene (99) to give ene–diyne 100 (Scheme 9.24) [30]. This process was designed to involve insertion of ruthenium onto the relay subunit in 98 to give 101, followed by rapid migration of [Ru] to the isomeric species 102 and 103 and drainage of the most reactive of these, 103, from the equilibrium manifold by final CM with 99. More recently, the same investigators have reported the successful implementation of this process in the total synthesis of three structurally related natural products tricholomenyn A (106), asperpentyn (109), and harveynone (110) [30, 31]. The tandem relay ene–yne RCM/metallatropic shift reactions of 104 gave 105, which lacks the dihydrofuran relay by-product but is otherwise isomeric with 104 (Scheme 9.25). Product 105 was then efficiently elaborated to tricholomenyn A (106). Interestingly, the pair of related, non-relay-activated substrates 107a and 107b had different fates upon treatment with [Ru]-II. The former did not undergo a tandem

TIPS

TBSO

Br

+

O O

Et

1. [Ru]-XII (30 mol%) C6H6, 70 °C (50%)

HO

Br O

Et

O

2. TBAF (77%)

(Z/E = 5 : 1)

O 77

78 Scheme 9.19

(+)-Scanloenyne (79)

Z-Selective relay-CM in the synthesis of (+)-scanloenyne (79).


275

F

[Ru]-I (10 mol%) CH2Cl2 O

MsO

()

O

15


H

O O

F

Ph

81 (53%)

O ( ) 14 MsO 80 MsO

()

O

15

O ( ) 14 MsO 84

83

H

O OH 82 (10%)

Ph NaO3SO

O

()

O

15

H

O

NH2

EtO2C (+)-Didemniserinolipid B (85) Scheme 9.20 RRCM substrate 83 altering the profile of RCM by-products in the synthesis of (+)-didemniserinolipid B (85).

n -Pr

5

HO H HO

Ti(Oi- Pr)4 (10 equiv.) then [Ru]-II (20 mol%) Me

HO

H 5

Toluene, reflux, 8 min 95%

H

HO

6

Me 6

86

Me

87

H 6

O

Me H

O

Me O

H

O AcO

AcO

H

Hypoestoxide (88) Scheme 9.21 RRCM to control the directionality of ring closure in the synthesis of macrocycle 87.

H

H

5

O 6 Me

Me O

89

Me

276

9 Metathesis Involving a Relay and Applications in Natural Product Synthesis Me Me Si O Me

[Ru]-II

Me

or [Mo]-I

Me

Me Me Me Si O

Me

90

[Ru]-II (10 mol%)

Me Me Si O Me

Me Me

CH2Cl2, rt 92%

Me

Me

91

92

Me Me

I

Me

OH

Me

Me

O

OH Me l

Me

NH2

93

OH

O

Me

O Me

OH

Discodermolide (94)

Fully regiocontrolled RRCM closure of 92 to 91.

O

O O

O

OH

l Me

O

Scheme 9.22

Me

MOMO

4

O

OTBDPS [Ru]-II (5 or 18 mol%)

MOMO

OTBDPS

5

CH2Cl2, reflux 75 or 88%

OH

Et TrO 95

OH

Et TrO

96

O

HO O

Et

HO

O P

O

OH

OH H2N Phoslactomycin B (97)

Scheme 9.23 RRCM enhancing the efficiency of cyclization, permitting a lower level of initiator loading.

cyclization analogous to that of 104, but the latter did to give 108. This intermediate was readily carried forward to both asperpentyn (109) and harveynone (110). In a related and more mechanistically motivated study, Lee et al. were able to structurally characterize an intermediate product of metallatropic rearrangement. Specifically, upon treatment of 112 with 1 equiv. of [Ru]-I they isolated crystalline 111 as the sole product (Scheme 9.26) [32]. Additionally, they observed a divergence of reactivity depending on the choice of the initiator. The action of [Ru]-II on substrate 112 gave the truncation product 113 exclusively. In light of additional strong evidence they accumulated for the reversible nature of the metal migration, the authors rationalized that the more hindered nature of the heterocyclic carbene ligand in [Ru]-II sterically destabilized ruthenium complex 114.


277

T AcO

OAc

99

C6H13

OAc

[Ru]-II (10 mol%) CH2Cl2, 40 °C

O O

C6H13

100 61% (from 98a, Z/E = 5 : 1) 40% (from 98b)

98a: T = O 98b: T = Si(Me)2O

Relay

CM

C6H13

O O

O

Metallatropic [1,3]-shift

101 Scheme 9.24

[Ru]-II (10 mol%)

O

102

Metallatropic [1,3]-shift

O

OTES O

CH2Cl2, 40 °C 58%

103

OAc

OAc

105

R1

107a: R1 = OTHP, R2 = MOM 107b: R1 = H, R2 = Ac

X Y

Me

O

O CH2Cl2, 40 °C OR2

Tricholomenyn A (106)

OTES

[Ru]-II (10 mol%)

O

O O

104

OTES

O O

Cascade of relay initiation, metallatropic rearrangement, and CM.

OTES

OAc

[Ru]

[Ru]

[Ru]

C6H13

O O

OH

OAc

108 62% (from 107b)

Scheme 9.25 Integration of RRCM with metallatropic shift in the syntheses of tricholomenyn A (106), asperpentyn (109), and harveynone (110).

Asperpentyn (109): X = OH, Y = H Harveynone (110): X,Y = O

278


O Me Me

Me

Me

N

PCy3

Ru

N Mes

Cl Oi- Pr [Ru]-II (100 mol%)

Cl

[Ru]-I (100 mol%) O 49% i- Pr

111 Scheme 9.26

Oi- Pr

Ru

Cl Cl

O

i -Pr 112

113

114

Relay activation to access pathways that lend mechanistic insight.

9.5 Examples of Relay Metatheses Thwarted in Achieving the Desired Outcome

Not every attempt to implement a relay metathesis strategy has resulted in the intended cyclization to a desired cyclic alkene. The obstructions fall mostly into one of two categories: 1) interference from a truncation event, 2) interference from premature macrocyclization. 9.5.1 Interference from a Truncation Event

As briefly introduced earlier (see the transformation of 42b to 42a in Scheme 9.10), in some instances the main RCM event turns out to be slower than that of bimolecular CM with other alkene-containing, methylene-donor molecules in the reaction medium. These donors are nearly always either the terminal alkene in an additional molecule of the substrate or the stoichiometric by-product of metathesis cyclization – most often ethylene. This truncation event was observed in early studies of RRCM with the substrate 8 (Scheme 9.2). In that experiment, approximately 2% of the parent (and unreactive) 5 was obtained in competition with closure to the RRCM product 7. As with all cases of truncation, this is most likely the result of competitive reaction processes of the metal-relayed intermediate 6, which can evolve to 7 by a unimolecular cyclization or to 5 by a bimolecular reaction with 8 and/or ethylene. This problem was again encountered soon thereafter in the attempted RRCM cyclization of substrates 115a–d to give 117 (Scheme 9.27), a key intermediate in the synthesis of callipeltoside A (118) [9]. In every instance, the outcome was the formation of the truncation product and parent diene 116, a substrate type we already knew to be unresponsive to the RCM protocols available at the time [33]. In this hindered and heavily branched system, truncation was always faster than cyclization, even when the experiment was performed at high dilution and with

9.5 Examples of Relay Metatheses Thwarted in Achieving the Desired Outcome

OH O

O

Me Me

OH O

Attempted RRCM

OPMB OTBS Me

MeO R

R1

Me

O

OH Me

O

O

Me

OPMB

OPMB OTBS Me MeO

MeO

O

2

Me

115a –d a

O

Me

116

b

c

279

O OTBS

117

d

O

L-Callipeltose

Me R1 =

R2 =

O H

Me

H

O

O

H

H

Me

O

MeO

Me O

OH O O

Me

Callipeltoside A (118) Scheme 9.27 is small.

Truncation: a problem when the rate constant for the final RCM step

the relay tethers present in 115c and 115d, which are less reactive CM partners compared to 115a and 115b, respectively. There are a few additional reports of truncation reaction products [11c, 20]. Rychnovsky et al. envisioned a possible double RCM within the coordination sphere of the Ca(II) biscarboxylate salt of the diene substrate 120 or of its relay-tethered analog 119 as an innovative approach to a synthesis of the dimeric C2 -symmetric macrodiolide natural product SCH 351448 (121) (Scheme 9.28) [34]. Again, metal loading and relay within 119 were effective, but the internal cyclization was sufficiently slow that truncated 120 was the only product observed. Under slightly more forcing conditions, the parent diene 120 produced a monomeric macrocyclic lactone RCM product. In studies aimed at a synthesis of characiol (124), Helmboldt and Hiersemann observed that it was possible to close triene 123b by RCM to give 17-norcharaciol, but that the methyl-substituted analog 123a was unreactive (Scheme 9.29) [35]. They then attempted to cyclize the relay-activated analog 122, but observed that it was converted to 123a rather than the macrocyclic dienedione present in 124. This reinforces the tenet that the rate of the primary unimolecular cyclization event of interest must be greater than that of the competing bimolecular CM reaction with another of the substrate molecules in order to RRCM to suceed. 9.5.2 Interference from Premature Macrocyclization

Another potential pathway that can interfere with the intended RRCM outcome is the internal macrocyclization of the initial ruthenium alkylidene. This becomes

Cl


280

O

OH OSEM O

O Ca

O

[Ru]-II or [Ru]-III

O Ca

O

2

OH O

O NaO2C

Me Me

OH O

O Me Me O

O

O

O Me Me

Me

119

OH OSEM O

O

O Me Me

O

120

Me

2

Me CO2H Me

OH O

O

HO

SCH 351448 (121) Scheme 9.28 is small.

Truncation: a problem when the rate constant for the final RCM step

a serious problem when the large-ring cyclization process is faster than the relay event (for example, in the generic RRCM process depicted in Scheme 9.1, closure of the alkylidene in 4b directly onto the R1 -bearing alkene prior to relay to the R2 -bearing alkene). Such a process was first observed as a competitive event during the RRCM cyclization of 28 to give 26, which was described earlier (Scheme 9.6); that is, the macrocyclic diene 127 was formed along with the desired 26 (Scheme 9.30). It is not clear whether this macrocyclization product arises via 126a and its reaction through path a in competition with relay contraction, or from competitive initial formation of 126b and its cyclization via path b. In the course of their synthesis of a ring-expanded bryostatin analog 131, Trost et al. studied the RCM cyclization of substrate 128, which was armed for a potential RRCM reaction (Scheme 9.31) [36]. This complex polyene cyclized very cleanly under the action of [Ru]-III to produce the cyclic diene 129. There was no evidence of formation of the smaller RRCM product. The rate of the relay event here is presumably lowered by the pair of quaternized carbons allylic (•) and homoallylic () to C17 in the alkylidene derived from 128.

9.6 Conclusion

O HO

Me

O HO Me Me

Me HO H

Me Me Me

Me HO H

O

122 Toluene, rt 90%

O HO

O

[Ru]-II or [Ru]-III (10 mol%)

Me

281

Me

R

Me

R HO

Me Me

H

O

123a: R = Me

R = Me : Characiol (124)

123b: R = H

R = H : 17-Norcharaciol (125)

Scheme 9.29 Substitution on the alkylidene intermediate slowing cyclization en route to 124.

X

Et

Et

b Y

9

O

a

Me 8 Me

O SiPh2

Et

[Ru]-II (10 mol%) CH2Cl2, 65 °C

O

Me O

+

SiPh2

O SiPh2

Me

Me

O

NC NC

NC

28: X = CH2, Y = CH2 126a: X = [Ru], Y = CH2 126b: X = CH2, Y = [Ru]

26 (58%)

127 (24%) (E/Z mixture)

Scheme 9.30 Competitive rates of relay versus direct cyclization resulting in premature ring closing.

Fürstner et al. attempted to close 132a to afford the skeleton of spirastrellolide A (134) by an RRCM reaction [the corresponding parent, non-relay-activated diene (not shown) was unreactive]. When exposed to initiator [Ru]-II, triene 132a gave, instead, the non-relay-contracted macrocyclic diene 133 (Scheme 9.32) [37]. The cumulative steric hindrance from several branched sites on the distal end of the initially formed ruthenium alkylidene 132b again presumably slowed the rate of the internal relay event to the point where cyclization within 132b to C25 of the remote vinyl group superceded.

9.6 Conclusion

We have reviewed here the discovery, evolution, and applications of ‘‘metathesis involving a relay.’’ While most of the examples fall into the category of RRCM, the handful comprising relay-CM processes further reinforces both the power and

282


Me Me MeO H3CO2C O 16

Me Me MeO

OAc

H3CO2C

O

H

O

[Ru]-III (17 mol%)

H

OAc

O

H

H

OTES O

17

OH H O

Me Me AcO

OTES C6H6 50 – 80 °C 80%

O

O

Me

Me Me AcO

OTES

CO2CH3

H3CO2C O

O

Me

OTES

CO2CH3 129 (E /Z = 1 : 1)

128

Me Me HO

OH H O

Me Me MeO OAc

H3CO2C O

O

OAc

O

O Me Me

OH H O

OH

O

O

O

n-Pr

OH

Me O

CO2CH3

OH

Me Me

OH H O

AcO Bryostatin (130)

Me 131

O OH

CO2CH3

Scheme 9.31 Macrocyclization of the relay-armed RRCM substrate 128 occurring via direct RCM (to 129) en route to the bryostatin analog 131.

potential of relay metathesis. As this collection of examples demonstrates, relay metathesis has been implemented at the strategic level for a number of different reasons and to address a variety of different types of reactivity and selectivity problems and issues. The feature common to each application is the opportunity that a relay metathesis strategy provides to capitalize on a redirecting effect that can alter, and thereby potentially complement, the outcome of the parent (or non-relay) metathesis reaction. Given the huge (and still growing) impact that olefin metathesis has had on the science of organic synthesis, it is reasonable to assume that limitations in classical metathesis reactions will continue to be encountered, especially so in the often more complex settings of natural product synthesis. Hopefully, the relay metathesis results archived here will serve to stimulate readers’ new ideas not only for solutions to their specific problems but also for new and innovative relay-based conceptual advances as well. For relevant

References Me

Me OMe

O O

BnO H

O

O O

O OH

25

HO O O

HO

O

BnO [Ru]-II (10 mol%)

Cl

X

Me

H

O

Cl

H O

CH2Cl2 reflux 64%

O

O

O OH

OH OMe

Me

OMe

O

H O

283

HO O O

Me

HO

O

Me OH OMe

O

132a: X = CH2 132b: X = [Ru]

133

Me OMe

O O

O

O

OH

Cl

O

OH OHO O

HO O O

HO

Me

Me OH OMe

Spirastrellolide A (134)

Scheme 9.32 Premature macrocyclization, which is faster than the relay event, precluding RRCM.

recent studies appearing after submission of this chapter to the publisher and added ca. five months later at the stage of review of pageproof see Refs 38–43. Acknowledgments

The authors thank Matthew J. Jansma and the editors of this book for their critical reviews of the manuscript, and the National Institutes of Health of the Department of Health and Human Services for research grants (GM65597 and CA76497) that have supported various aspects of our studies in this area here at the University of Minnesota. References 1 Ulman, M. and Grubbs, R.H. (1998)

Organometallics, 17, 2484–2489. 2 (a) Kirkland, T.A. and Grubbs, R.H. (1997) J. Org. Chem., 62, 7310–7318;

(b) refs in Stewart, I.C., Ung, T., Pletnev, A.A., Berlin, J.M., Grubbs, R.H., and Schrodi, Y. (2007) Org. Lett., 9, 1589–1592.

284

9 Metathesis Involving a Relay and Applications in Natural Product Synthesis 3 (a) Hoye, T.R. and Zhao, H. (1999)

4

5

6

7

8

9

10

11

Abstracts of Papers, 218th National Meeting of the American Chemical Society, August 22–26, 1999, New Orleans, LA, American Chemical Society, Washington, DC, ORGN-620; (b) Hoye, T.R. and Zhao, H. (1999) Abstracts of Papers, 218th National Meeting of the American Chemical Society, August 22–26, 1999, New Orleans, LA, American Chemical Society, Washington, DC, ORGN-303. Zhao, H. (2000) PhD Thesis, Studies towards a total synthesis of callipeltoside A, University of Minnesota, Minneapolis, MN. Hoye, T.R., Jeffrey, C.S., Tennakoon, M.A., Wang, J., and Zhao, H. (2004) J. Am. Chem. Soc., 126, 10210–10211. Robinson, J., Piscopio, A.D., and Zhu, L. (2003) Abstracts of Papers, 226th National Meeting of the American Chemical Society, September 7–11, 2003, New York, American Chemical Society, Washington, DC, ORGN-118. (a) Hansen, E.C. and Lee, D. (2004) Org. Lett., 6, 2035–2038; (b) Hansen, E.C. (2006) PhD Thesis, Ruthenium-catalyzed enyne couplings: reactivity, selectivity and synthetic utility, University of Wisconsin, Madison, WI. (a) Wang, X., Bowman, E.J., Bowman, B.J., and Porco, J.A. (2004) Angew. Chem. Int. Ed. Engl., 43, 3601–3605; (b) Wang, X. (2005) PhD Thesis, Studies toward the synthesis of chondriamides A and C, oximidines II and III, and the tetrapetalon A core, Boston University, Boston, MA. Danielson, M.E. (2003) PhD Thesis, A total synthesis of callipeltoside A, University of Minnesota, Minneapolis, MN. Tennakoon, M.A. (2001) PhD Thesis, efforts toward the total synthesis of scyphostatin; synthetic studies towards peloriside A, University of Minnesota, Minneapolis, MN. (a) Hoye, T.R. and Wang, J. (2003) Abstracts of Papers, 226th National Meeting of the American Chemical Society, September 7–11, 2003, New York, American Chemical Society, Washington, DC, ORGN-670; (b) Wang, J. (2005) PhD Thesis, 1. The first total synthesis of

12

13 14 15

16

17 18

19

20 21 22

haterumalide NA (oocydin A); 2. Deciphering the coupling constants of spiruchostatins by NMR spectroscopy; 3. Design and synthesis of analogs of latminculin A and B, Actin polymerization inhibitors, University of Minnesota, Minneapolis, MN; (c) Hoye, T.R. and Wang, J. (2005) J. Am. Chem. Soc., 127, 6950–6951. Jeffrey, C.S. (2004) Master thesis, Part I: The epoxidation of 1.0 mg of geraniol. Part II: Relay ring closing metathesis (RRCM): Selectivity in tandem enyne synthesis, University of Minnesota, Minneapolis, MN. Maifeld, S.V. and Lee, D. (2005) Chem. Eur. J., 11, 6118–6126. Wallace, D.J. (2005) Angew. Chem. Int. Ed. Engl., 44, 1912–1915. (a) Spino, C., Bolsvert, L., Douville, J., Roy, S., Lauzon, S., Minville, J., Gagnon, D., Beaumier, F., and Chabot, C. (2006) J. Organomet. Chem., 691, 5336–5355; (b) Lauzon, S., Tremblay, F., Gagnon, D., Godbout, C., Chabot, C., Mercier-Shanks, C., Perreault, S., DeSeve, H., and Spino, C. (2008) J. Org. Chem., 73, 6239–6250. Sato, D., Fujiwara, K., Kawal, H., and Suzuki, T. (2008) Tetrahedron Lett., 49, 1514–1517. Grandbois, A. and Collins, S.K. (2008) Chem. Eur. J., 14, 9323–9329. El-Azizi, Y., Schmitzer, A., and Collins, S.K. (2006) Angew. Chem. Int. Ed. Engl., 45, 968–973. (a) El-Azizi, Y., Zakarian, J.E., Bouillerand, L., Schmitzer, A.R., and Collins, S.K. (2008) Adv. Synth. Catal., 350, 2219–2225; (b) Zakarian, J.E., El-Azizi, Y., and Collins, S.K. (2008) Org. Lett., 10, 2927–2930; (c) Collins, S.K., El-Azizi, Y., and Schmitzer, A.R. (2007) J. Org. Chem., 72, 6397–6408. Peng, W. and Blagg, B.S.J. (2006) Org. Lett., 8, 975–978. Crimmins, M.T., Zhang, Y., and Diaz, F.A. (2006) Org. Lett., 8, 2369–2372. She, J. (2004) PhD Thesis, Cost efficient asymmetric aldol reactions: application towards the total synthesis of (+)-gigantecin, University of North Carolina, Chapel Hill, NC.

References 23 Dudley, G.B., Engel, D.A., Ghiviriga, I.,

24

25

26

27

28 29

30 31 32

33

Lam, H., Poon, K.W., and Singletary, J.A. (2007) Org. Lett., 9, 2839–2842. Roethle, P.A., Chen, I.T., and Trauner, D. (2007) J. Am. Chem. Soc., 129, 8960–8961. Lee, H., Kim, K.W., Park, J., Kim, H., Kim, S., Kim, D., Hu, X., Yang, W., and Hong, J. (2008) Angew. Chem. Int. Ed. Engl., 47, 4200–4203. Marvin, C.C., Voight, E.A., Suh, J.M., Paradise, C.L., and Burke, S.D. (2008) J. Org. Chem., 73, 8452–8457. McGrath, N.A., Lee, C.A., Araki, H., Brichacek, M., and Njardarson, J.T. (2008) Angew. Chem. Int. Ed. Engl., 47, 1–5. Xie, Q., Denton, R.W., and Parker, K.A. (2008) Org. Lett., 10, 5345–5348. Druais, V., Hall, M.J., Corsi, C., Wendeborn, S.V., Meyer, C., and Cossy, J. (2009) Org. Lett., 11, 935–938. Cho, E.J. and Lee, D. (2008) Org. Lett., 10, 257–259. Li, J., Park, S., Miller, R.L., and Lee, D. (2009) Org. Lett., 11, 571–574. Yun, S.Y., Kim, M., Lee, D., and Wink, D.J. (2009) J. Am. Chem. Soc., 131, 24–25. Hoye, T.R. and Zhao, H. (1999) Org. Lett., 1, 169–171.

34 Cheung, L.L., Marumoto, S., Anderson,

35 36

37

38

39

40 41

42

43

C.D., and Rychnovsky, S.D. (2008) Org. Lett., 10, 3101–3104. Helmboldt, H. and Hiersemann, M. (2009) J. Org. Chem., 74, 1698–1708. Trost, B.M., Yang, H., Thiel, O.R., Frontier, A.J., and Brindle, C.S. (2007) J. Am. Chem. Soc., 129, 2206–2207. Fürstner, A., Fasching, B., O’Neil, G., Fenster, M.D.B., Godbout, C., and Ceccon, J. (2007) Chem. Commun., 3045–3047. Tannert, R., Hu, T.-S., Arndt, H.-D., and Waldmann, H. (2009) Chem. Commun., 1493–1495. Donohoe, T.J., fishlock, L.P., Basutto, J.A., Bower, J.F, Procopiou, P.A. and Thompson, A.l (2009) Chem. Commun., 3008–3010. Crimmins, M.T, Jacobs, D.L. (2009) Org. lett., 11, 2695–2698. Ashworth, I.W., Miles, J.A.L., Nelson, D.J., Percy, J.M., singh, K. (2009) Tetrahedron, 65, 9637–9646. Donohoe, T.J., Bower, J.F., Basutto, J.A., fishlock, L.P., procopiou, P.A, and callens, C.K.A. (2009) Tetrahedron, 65, 8969–8980. Siegel, D.S., Piizzi, G., Piersanti, G., and Movassaghi, M. (2009) J. Org. Chem., 2009, ASAP.

285

287

10 Cross-metathesis in Natural Products Synthesis Joëlle Prunet and Laurence Grimaud

10.1 Introduction

This chapter is divided into six sections. The first three sections loosely reflect the role of alkene cross-metathesis (CM) in the general plan of the synthesis of natural products, which can be the functionalization of terminal olefins appending a side chain to the core of a complex compound, or coupling two fragments to build the entire skeleton of the target molecule. Afterwards, tandem processes involving CM will be presented, followed by a few examples of ene–yne and alkyne CM in natural product synthesis [1]. This review is not comprehensive, but we have tried to select significant examples in each of the categories of CM reactions cited above, while at the same time emphasizing to which types [2] the olefins involved in the metathesis reactions belong.

10.2 Functionalization of Olefins 10.2.1 Cross-metathesis with Acrylate Derivatives

Electron-deficient olefins such as acrylate derivatives fall into the type II category. Cross-metatheses of these substrates (used in excess) with type I terminal olefins are usually easy and diastereoselective in favor of the E-isomer, except for acrylonitrile, which leads to the Z-products. 10.2.1.1 Acrylonitrile Stockman, Fuchs, and coworkers employed a two-directional CM reaction between dienone 1 and acrylonitrile to synthesize the core of histrionicotoxin (Scheme 10.1), a noncompetitive inhibitor of nicotinic acetylcholine receptors [3]. Thermal conditions for this metathesis reaction required refluxing in tetrahydrofuran Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts. Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32440-8

288

10 Cross-metathesis in Natural Products Synthesis

CN

O 1

O

[Ru]-III (30 mol%)

+

CH2Cl2, 120 °C Microwave 5 h 62%

CN (3 equiv.)

2 CN NH2OH·HCl NaOAc 89% MeOH/MeCN, 50 °C

NC HN HO

Ref. 3

Histrionicotoxin Scheme 10.1

N O

4

NC Toluene, 180 °C Sealed tube 95%

CN

O N

NC H

3

CM with acrylonitrile in the synthesis of histrionicotoxin.

(THF) for two weeks and adding 5 mol% of catalyst every two days. Fortunately, CM could be accelerated by microwave irradiation and compound 2 could be obtained in 62% yield after only 5 hours at 120 ◦ C. Kinetic tricyclic product 3 was then obtained by an oxime formation/Michael addition/1,4-prototropic shift/[3+2]-cycloaddition cascade, before being isomerized to the thermodynamic tricycle 4 via a retro-[3+2]/[3+2]-cycloaddition sequence. 10.2.1.2 Thioacrylates The synthesis of faranal, the trail pheromone of the common ant Monomorium pharaonis, was recently reported by Minnaard, Feringa, and coworkers [4]. Hence, CM of 5 with ethyl thioacrylate furnished the unsaturated thioester 6 in good yield along with 5% of the homodimer of 5 (Scheme 10.2). The second stereocenter was installed by an enantioselective conjugate addition, which led to 7 in both high yield and high selectivity. 10.2.1.3 Acrylic Acid Ruthenium catalysts are tolerant to many functional groups including carboxylic acids. BouzBouz and Cossy took advantage of this property for the construction of the seco acid of colletol, a bis-lactone isolated from the fermentation broth of Collectotrichum capsici [5]. Conjugated hexenoic acid derivative 9 was obtained in 78% from olefin 8 and acrylic acid, while the C9–C10 double bond was formed with an E/Z selectivity better than 20 : 1 (Scheme 10.3). Macrolactonization under Yamaguchi’s conditions followed by the cleavage of the methoxymethyl (MOM)

10.2 Functionalization of Olefins

O

O BnO

[Ru]-III (10 mol%)

+

SEt (2 equiv.)

5 (ee = 94%)

CH2Cl2, reflux 83%

BnO

SEt 6

Cul (3 mol%) MeMgBr (4.0 equiv.) (R )-TolBINAP (3.3 mol%) t -BuOMe, −70 °C 91%

O

O

Ref. 4

BnO

H

7 (dr = 98 : 2, ee > 99.5%)

Faranal

Scheme 10.2

O

SEt

Synthesis of faranal involving CM with ethyl thioacrylate.

MOMO

O

OH Acrylic acid (3 equiv.) [Ru]-III (5 mol%)

O

CH2Cl2, 25 °C 78%

8

O

MOMO

10

OH

CO2H 9

9

(E/Z > 20 : 1) 2,4,6-Cl3C6H3COCl Et3N, toluene, 25 °C 60% DMAP, toluene, 110 °C

O

O O

O 2 N HCl, THF 79%

HO

MOMO O

O

Colletol Scheme 10.3

9

O

10

O

10

CM with acrylic acid toward colletol.

ether led to colletol in good overall yield. It is interesting to note that an alternative route involving acylation of 8 with acrylic acid followed by ring-closing metathesis (RCM) led to macrocycle 10 as a 2.8 : 1 mixture of E- and Z-isomers. 10.2.1.4 Acrylimides During their total synthesis of cylindramide A [6], a macrocyclic tetramic acid isolated from Halicondria cylindrata, which is cytotoxic to B16 melanoma cells, Hart and Phillips performed a highly diastereoselective Diels–Alder reaction between acrylimide 13 and cyclopentadiene, applying the methodology developed

289


290

O

O +

TIPSO

O

O [Ru]-II (10 mol%)

N

11

O

12

CH2Cl2, 40 °C 59%

TIPSO

N

O

13 Bn

Bn O H

Me

O

96%

HN

Et2AlCl CH2Cl2, −78 °C

HO

H

O

O Xc =

N

O

NH

Bn

OH

OTIPS

O

Cylindramide A Scheme 10.4

Xc 14 (dr = 45 : 1)

CM with an acrylimide derivative en route to cylindramide A.

by Evans et al. [7]. Dienophile 13 was prepared via a CM between olefin 11 and acryloyl oxazolidinone 12 using [Ru]-II (Scheme 10.4). 10.2.1.5 Acrylates There are numerous examples of CM reactions with ethyl or methyl acrylate. An interesting case was uncovered by Trost et al. during their formal synthesis of isoaltholactone, a cytotoxic agent selective for HT-29 colon cancer cells [8]. CM of 15 was regioselective, with metathesis only occurring on the side of the allylic alcohol to give unsaturated ester 16 in good yield (Scheme 10.5). The authors invoke the O

OAr

O +

MeO

CH2Cl2, reflux, 2 h 69%

15 OH

(5 equiv.)

MeO 16

O

OH

1. TBSCl, Im 81% 2. CAN 3. PhSH, Et3N

Ar = 4-methoxy-2,3-dimethylphenyl O O H O

OAr

[Ru]-II (2 mol%)

H

OH

Ph Isoaltholactone

1. Otera's catalyst (85%)

Ph

PhS 18

OTBS

SPh OH

O

O

Ref. 10

2. Styrene (5 equiv.) [Ru]-II (5 mol%) CH2Cl2, reflux 16 h (90%)

Scheme 10.5 Regioselective CM with methyl acrylate for the formal synthesis of isoaltholactone.

MeO 17

OTBS

10.2 Functionalization of Olefins

291

deactivating effect of the electron-rich arylether to explain this selectivity. Further functionalization of compound 16 led to ester 17, which was lactonized in the presence of Otera’s catalyst [9]. CM of the resulting lactone with styrene then furnished lactone 18, while elimination of the benzenethiol moiety completed the formal synthesis of isoaltholactone [10]. 10.2.1.6 Acrolein Cossy and BouzBouz designed a retrosynthesis of the C1–C14 fragment of amphidinol 3 with three CM reactions between homoallylic alcohols and acrylic derivatives [11]. The first metathesis between 19 and acrolein was straightforward, giving the expected product 20 in 79% yield (Scheme 10.6). Aldehyde 20 was then transformed into bis-acetate 21 by treatment with the allyltitanium complex (S,S)-24, first reported by Hafner, Duthaler, and coworkers [12], followed by acetylation. Interestingly, the second CM with acrolein leading to 22 was very chemoselective and only the terminal olefin reacted. The internal double bond did not undergo metathesis, probably because the corresponding metallacyclobutane was deactivated by complexation of the ruthenium by the carbonyl moiety of the acetate group. It had been shown on model systems that the same reaction on the corresponding alcohols was not selective [11]. Aldehyde 22 was then transformed into homoallylic acetate 23 as previously, while the C1 and C2 carbons were installed by CM with ethyl acrylate. In each case, the E/Z ratio obtained after metathesis was better than 50 : 1.

CHO

OH

OH

(3 equiv.)

PMBO

[Ru]-III, 25 °C CH2Cl2, 36 h 79%

19

OAc

2 steps

PMBO

CHO

OAc

PMBO

20

21 CHO [Ru]-III, 25 °C CH2Cl2, 12 h (3 equiv.) 63%

OAc

OAc

OAc

OAc 2 steps

PMBO

OAc

PMBO

CHO

23

22

COOEt [Ru]-III, 25 °C CH2Cl2, 24 h 61%

(3 equiv.)

OAc PMBO

OAc

OAc

(S,S )-24 COOEt

14

1

Ti O

O

Ph Ph O

Ph Ph O

C1−C14 fragment of Amphidinol 3 Scheme 10.6

Successive CM for the synthesis of the C1–C14 fragment of amphidinol 3.

292


10.2.1.7 Vinyl Ketones In their synthesis of nupharamine [13], isolated from the rhizome of the water lily Nuphar japonica, Gebauer and Blechert effected a CM of homoallylic protected amine 26 (prepared in three steps from the corresponding phthalimide 25) with furyl vinyl ketone, leading to the desired compound 27 in good yield (Scheme 10.7). Hydrogenation of the double bond followed by deprotection of the amine and reductive amination then furnished nupharamine. The vinyl ketone partner used in the CM reaction can be of a size comparable to that of the other partner. For example, Cossy et al. recently described the synthesis of the monomeric counterpart of marinomycin A, isolated from the saline culture of Marinospora strain CNQ-140, which shows significant antibiotic activity against Staphylococcus aureus or Enterococcus faecium, using a CM between a vinyl ketone and a type I olefin [14]. Another case of CM between a complex vinyl ketone and a type I olefin was reported by Rychnovsky and coworkers for the construction of the C1–C52 segment of amphidinol 3 [15]. 10.2.2 Cross-metathesis with Vinyl Derivatives

CM reactions can also be used to prepare partners for cross-coupling reactions such as vinyl boronates, vinyl silanes, or vinyl halides, which are type II olefins. 10.2.2.1 Vinyl Boronates CM reactions with vinyl boronates were first reported by Morril and Grubbs [16]. They are usually diastereoselective in favor of the E-isomer. Vincent and Prunet used this type of CM for the synthesis of the C1–C15 fragment of dolabelide C, a cytotoxic macrolactone isolated from the sea hare Dolabella auricularia [17]. Hence, 1,1-disubstituted olefin 28 was reacted with 4 equiv. of vinyl boronate 29, and the resulting product was converted to the corresponding vinyl iodide in 89% yield for the two steps (Scheme 10.8). As it is often the case with 1,1-disubstituted 1. TFA (30%) 2. MeNH2

NPhth

3. Boc2O, NaHCO3 81%

25

NH OH

26 Boc

[Ru]-III (10 mol%) CH2Cl2 40 °C, 72 h 74%

O O (1.2 equiv.)

O

OH

1. H2, Pd/C 2. TFA (10%)

N H

O

Nupharamine Scheme 10.7

3. NaBH4, EtOH 75%

OH

NH 27 Boc

O

Synthesis of nupharamine by a CM/reductive amination sequence.

10.2 Functionalization of Olefins Ph O

O

OTBS

OPMB OTBS

28 OAc OAc OB O

(1)

29

[Ru]-II, CH2Cl2, reflux

55%

OAc

OH

O OH OH O

OAc

(2) I2, NaOH Ph O

O

Dolabelide C OTBS

15

OPMB 1

I

OTBS C1−C15 fragment

Scheme 10.8 CM with a vinyl boronate for the synthesis of the C1–C15 fragment of dolabelide C.

olefins with no strong steric bias, the selectivity of the metathesis reaction was modest (E/Z = 2 : 1) [2]; however, the desired E-olefin could be obtained in 55% overall yield after separation of the two diastereomers. A CM reaction between vinyl boronate 29 and a terminal olefin was also employed by Rahn and Kalesse during their synthesis of chlorotonil A (see Scheme 10.11 for the structure of this compound), which has been isolated from the myxobacterium Sorangium cellulosium [18]. The resulting product was then engaged in a Suzuki coupling. 10.2.2.2 Vinyl Silanes The same kind of strategy was employed by Kirschning and coworkers for their synthesis of thuggacin B, a polyketide isolated from the myxobacterium Sorangium cellulosium, which shows promising activity against Mycobacterium tuberculosis [19]. CM of terminal olefin 30 (encompassing a diol motif) with vinyl triethoxysilane proceeded in satisfactory yield (Scheme 10.9). When the two hydroxy groups present in 30 were protected as TBS ethers, the yield of the metathesis was poor, probably due to steric hindrance. After protection of the diol moiety as an acetonide, vinyl silane 31 was transformed into vinyl iodide 32, which was later engaged in a Sonogashira coupling. 10.2.3 Cross-metathesis with Allylic Derivatives

This category of coupling usually involves two type I olefins, and proceeds in good yield but with variable stereoselectivity.

293


294

O

HO 1. (EtO)3SiCH CH2 [Ru]-II, CH2Cl2, reflux (61%)

HO N

30

O N

2. 2,2-DMP Cat PPTS, rt, 2 h (72%)

S

X

S

CO2Et H13C6

CO2Et H13C6

31: X = Si(OEt)3 32: X = I

HO

MeOH, KHF2, rt, 12 h then I2, rt, 4 h 94%

HO OH N

O

OH

OH

S

O

Thuggacin B Scheme 10.9

CM with a vinyl silane en route to thuggacin B.

10.2.3.1 Allyl Silanes Allyl trimethylsilane has been used as a CM partner by Fukuyama and coworkers in the synthesis of the 3,8-diazabicyclo[3.2.1] core of lemonomycin, a tetrahydroisoquinoline alkaloid isolated from Streptomyces candidus with interesting antibiotic activity against Staphylococcus aureus or Enterococcus faecium, as well as cytotoxicity against HCT-116 human colon cancer cell line [20]. CM between disubstituted olefin 33 and allyl trimethylsilane led to the desired product 34 in 51% yield as a mixture of E- and Z-isomers (Scheme 10.10). This compound, when treated with BF3 ·OEt2 , yielded bicycle 35 via the corresponding acyl iminium. When the metathesis reaction was performed with the allyl analog of 33 without the terminal methyl group, a low yield was obtained because of partial decomposition. 10.2.3.2 Allyl Phosphonates Rahn and Kalesse employed a diastereoselective CM reaction during their synthesis of chlorotonil A between 36 and allyl phosphonate 37, which led to phosphonate 38 as a single isomer (Scheme 10.11) [18]. This phosphonate was then used in an olefination with aldehyde 39 which occurred without elimination of the allylic OPMB group to afford diene 40. 10.2.3.3 Allylic Alcohol Derivatives Roulland recently described the total synthesis of Oocydin A, a chlorinated macrolide isolated from the bacterium Serratia marcescens [21]. CM of diol 41 with allyl acetate proceeded to give compound 42 in good yield and excellent selectivity (E/Z = 95 : 5) (Scheme 10.12). This compound was then engaged in a palladium-catalyzed allylic substitution that led to THF 43 in a highly diastereoselective fashion.

10.2 Functionalization of Olefins OAc

OAc

SiMe3

SiMe3

NBoc Ar

NBoc [Ru]-II (2 mol%)

N

Ar

CH2Cl2, reflux 51%

O OAc

N O OAc

33

BF3·OEt2 CH2Cl2, −78 °C

HO

MeO Ar = O MeO

OH

H

34

95%

H

MsO Boc N O R=

NH

N

MeO

OH

O

N

Ar

OR

OH

O OAc 35

Lemonomycin NMe2

Scheme 10.10

CM with allyl silane for the construction of the lemonomycin core.

O P(OCH2CF3)2 37

36

O P(OCH2CF3)2 [Ru]-II

+

H O

+

CH2Cl2 64%

OPMB 38

MeO

OPMB KHMDS Et2O, −80 °C

Cl

O

O

Chlorotonil A

Scheme 10.11

H H O

H O

39

H

H Cl

O

H

55%

H H

H

H

Ref. 18

MeO

H H O

OPMB 40 (Z,E/E,E = 3 : 1)

Formation of an allyl phosphonate by CM in the synthesis of chlorotonil A.

10.2.3.4 Miscellaneous Lasonolide A, a macrolide isolated form the Caribbean marine sponge Forcepia sp., exhibits potent antitumor activity against A-549 human lung carcinoma and Panc-1 human pancreatic carcinoma. The synthesis of this molecule described by Ghosh and Gong involves a CM between olefin 44 and homoallyltetrazolylsulfone 45, which proceeds in good yield (Scheme 10.13) [22]. After protection of the hydroxy group present in 46 as a methylthiomethyl (MTM) ether, the resulting sulfone was

295

296


OH OMPM

AcO

[Ru]-II

OMPM

CH2Cl2, reflux

OH

41

OH

AcO

OH

77%

42 (E/Z = 95 : 5) Pd2(dba)3 P(p -MeOC6H4)3 THF, 40 °C

99%

OH

O Cl

CO2H

H Ref. 21

O

OAc O

96:4 O

OH 43

Oocydin A

Scheme 10.12

Synthesis of oocydin involving a CM with allyl acetate.

OH

HO

O +

S

N N N N

O 44

O

Ph

O

[Ru]-II CH2Cl2, reflux

O

Me O

H OMPM

O

Me

N

81%

O

45

O

N N 46 N

S

O

Ph

OH O H

O

15

O

Me HO

Me 12

O O Lasonolide A Scheme 10.13

Ref. 22

H H

O

H Me OH

CM with homoallyl tetrazolyl sulfone for the construction of lasonolide A.

engaged in a Julia–Kocienski coupling with an unsaturated aldehyde to form the C12–C15 dienic system of lasonolide A.

10.3 Appending a Side Chain

Appending a side chain to the core of a complex molecule very often involves a type II olefin (the core) and a type I olefin (the side chain). However, in some cases the side chain can be of type II or III as exemplified in this section.


297

10.3.1 With No Functional Group 10.3.1.1 A Simple Case Reiser and coworkers reported the enantioselective synthesis of five paraconic acids with interesting antibiotic and antitumor properties [23]. These 3-carboxylic acid-substituted γ -butyrolactones differ inter alia by their side chain at C4, which were attached by CM (Scheme 10.14). For the synthesis of roccellaric acid, lactone 47a was treated with 1.5 equiv. of 1-dodecene and catalyst [Ru]-I. Compound 48a was obtained in 57% yield as a 3.5 : 1 mixture of E- and Z-isomers. In an analogous way, CM of lactone 47b with 1-decene furnished 48b, a precursor of nephrosteranic acid, in 53% yield. Even though a large excess of 1-decene (4 equiv.) and the more active catalyst [Ru]-II were used, the conversion was not complete (27% recovered 47b). However, the selectivity was much higher in the latter case (E/Z = 7 : 1). 10.3.1.2 The Specific Case of Isopropylidene 1,1-Disubstituted alkenes do not usually dimerize in the presence of catalyst [Ru]-II but when isobutylene is used as solvent, a small amount of background dimerization is observed. Grubbs and coworkers reported the superior behavior of 2-methyl-2-butene for this kind of reactions [24], while Porco and Qi employed this reagent for the synthesis of clusianone, a potential anti-HIV therapeutic agent isolated from the floral resin of Clusia species [25]. The isoprenyl side chain was installed in good yield from 49 by allylic acetate reduction followed by CM in neat 2-methyl-2-butene in the presence of catalyst [Ru]-II. Compound 50 was isolated and then transformed into the target molecule by nucleophilic demethylation (Scheme 10.15).

O

9

(1.5 equiv.) [Ru]-I (5 mol%) 57% 3

O

O

O

R

O

CO2H

1. H2, Pd/C

O 9

48a (E/Z = 3.5 : 1)

2. CrO3, H2SO4 3. NaHMDS, MeI 84%

O

O

12

Roccellaric acid

4

47a: R = O O 47b: R = CHO

CO2H

CHO

7

(4 equiv.) [Ru]-II (5 mol%) 53%

Ref. 23

O

O

7

48b (E/Z = 7 : 1)

Scheme 10.14 CM with simple alkenes for the syntheses of roccellaric and nephrosteranic acids.

O

O

10

Nephrosteranic acid

298


O

O

O

O

1. Pd(PPh3)4, HCO2NH4 toluene,105 °C (90%)

O

AcO

2. [Ru]-II,

OMe

O

(89%)

49

OMe

50

O

O LiOH, dioxane, reflux

O

77%

OH

Clusianone

Scheme 10.15 Transforming an allyl into an isoprenyl side chain using CM with 2-methyl-2-butene.

10.3.1.3 Removing Part of a Side Chain

Wipf et al. converted an allyl substituent into an ethyl group during their synthesis of tuberostemonine (Scheme 10.16) [26], a compound from the family of the Stemona alkaloids, which have been used against tuberculosis and bronchitis in Eastern folk medicine. Isomerization of the terminal double bond in 51 was

O

O H

O

[Ru]-II, toluene

H N H O

O

H

O

Allyltritylamine DIPEA 85%

H H N H O H

51

52 TsOH, [Ru]-III, CH2Cl2 reflux, H2C CH2

O O

O

81%

O H H N H O H

H2, Pd/C

O

O

H H

97%

N H O H

Tuberostemonine Scheme 10.16 Isomerization/CM sequence for the synthesis of the side chain of tuberostemonine.

53

O


299

accomplished with catalyst [Ru]-II in refluxing toluene in the presence of allyltritylamine and diisopropylethylamine (DIPEA) [27]. CM reaction of 52 with ethylene in the presence of catalyst [Ru]-III and para-toluenesulfonic acid (TsOH) in refluxing CH2 Cl2 furnished compound 53 in 81% yield. Interestingly, the use of the phosphine-free catalyst for this step was important in order to avoid lengthy chromatographic separation that led to decomposition of the product. Hydrogenation of the vinyl substituent finally provided tuberostemonine in excellent yield. Franck and coworkers had previously employed a similar isomerization/CM sequence for the preparation of sugar derivatives used in the synthesis of α-C-galactosyl ceramide immunostimulants, but in their case the isomerization reaction was catalyzed with PdCl2 [28]. 10.3.2 With Functional Groups

During their synthesis of prostaglandins F2α and J2 , Mulzer and coworkers studied the effect of allylic and homoallylic substituents on CM [29]. Hence, reaction of lactone 54a (R = Bz) with allylic silylether 55, in the presence of complex [Ru]-II added via syringe pump, furnished compound 56a in only 40% yield and 7 : 1 E/Z selectivity along with 25% of the corresponding debenzoylated product 56c (R = H) (Scheme 10.17). With catalyst [Ru]-III, the yield dropped to 25%. These poor results are probably due to the deactivation of the catalyst by complexation of the ruthenium by the carbonyl moiety of the benzoate. Indeed, when lactone 54b (R = TBS) was submitted to CM, the desired product 56b was obtained in 84% yield and 17 : 1 E/Z selectivity. The result with lactone 54c (R = H) was also satisfactory, as the corresponding coupled product was isolated in 70% yield and 17 : 1 E/Z selectivity (in all cases, the catalyst was added over a period of O

O O

[Ru]-II (6 mol%)

+

O

CH2Cl2, 40 °C

OTBS RO

RO

OTBS 56a: R = Bz (40%, E:Z = 7 : 1) 56b: R = TBS (84%, E:Z = 17 : 1) 56c: R = H (70%, E:Z = 14 : 1)

55

54a: R = Bz 54b: R = TBS 54c: R = H HO

CO2H

CO2H HO

Scheme 10.17

O OH PGF2α

OH PGJ2

Effect of homoallylic substituents on CM en route to prostaglandins.

300


58

TBSO

OH

[Ru]-II (10 mol%)

Me

O

CH2Cl2 Reflux, 20 h 98%

H

57 (dr = 2 : 1)

TBSO Me

O

OH

H

59 Bu4NF 93% THF

HO Me

O

OH

H

b-H: Heliannuol G a-H: Heliannuol H Scheme 10.18 Installing the side chain of heliannuols G and H using CM with a hindered allylic alcohol.

8–12 hours). Compounds 56b and 56c were then converted into prostaglandins F2α and J2 , respectively. It is noteworthy that when catalyst [Ru]-III was used with 54b or 54c, only the homodimers of each partner were obtained. Shishido and coworkers have used CM with a hindered allylic alcohol for appending the side chain of heliannuols G and H, two sesquiterpenes isolated from the leaves of Helianthus annus L. SH-22, and YPP, which exhibit interesting phytotoxic allelopathic activity [30]. Reaction of TBS ether 57 with 2-methyl-3-buten-2-ol 58 led to the direct precursor of heliannuols G and H in excellent yield (Scheme 10.18). However, when the phenolic analog of 57 was submitted to similar reaction conditions, a mixture of heliannuols G and H was obtained in only 32% yield, thereby showing the importance of the phenol protecting group on the CM. 10.4 Couplings

There are numerous examples of CM being used to assemble the skeleton of natural products. For example, in the last step of the synthesis of melithiazole C, a fungicide extracted from myxobacterium Melittangium lichenicola, Cossy and coworkers optimized a CM between vinyl thiazole 60 and allylic methylether 61 (Scheme 10.19) [31]. While catalyst [Ru]-I was ineffective, CM of an equimolar amount of 60 and 61

O

OMe OMe N

+

S

60 (2 equiv.) Scheme 10.19

CO2Me

61

OMe OMe [Ru]-II (30 mol%) CH2Cl2 40 °C, 60 h 56%

O

N CO2Me

S Melithiazole C

CM with a vinyl thiazole for the last step of melithiazole C synthesis.

10.4 Couplings

with [Ru]-II or [Ru]-III resulted in the formation of the homodimer of thiazole 60. Since compound 61 is a type II olefin that dimerizes slowly, more forcing conditions were employed (30 mol% of catalyst) and the target compound was obtained in 56% yield (68% conversion) with an excellent E/Z ratio. Higher catalyst loading or heating the reaction in benzene at 60 ◦ C did not improve the yield. Marshall and Sabatini recently reported the synthesis of asimicin, an Annonaceus acetogenin exhibiting a broad range of biological activities [32]. Two successive CM reactions were envisaged for the construction of 64 (Scheme 10.20). When C2 -symmetric bis-alkene 62a was reacted with 1-decene, the product resulting from a bis-metathesis reaction 62b was obtained in 80% yield. No selective mono CM was possible since 62a is a type II olefin as opposed to a type I partner. However, further metathesis of type III olefin 62b with type I olefin 63 furnished the desired product 64 in 38% yield along with 28% of recovered 62b and 9% of the bis-metathesis product. A single recycle of the side products furnished an additional 25–30% yield in 64. Hydrogenation of the two alkene moieties and acidic removal of the protecting groups led to asimicin in good yield. Bis-metathesis of 62a with 63 followed by CM with 1-decene was also performed, but the overall process was slightly less efficient.

MOMO

H

O

H H

O

H

OMOM

62a 80% [Ru]-II, CH2Cl2,

H17C8 MOMO

H

O

H H 62b

O

C8H17

C8H17 H

TBSO

O

7 63

O

+

OMOM 38%

[Ru]-II CH2Cl2

TBSO C8H17 MOMO

H

O

H H

O

7

H

OMOM

O O

64

80%

1. TsNHNH2 2. HCl

OH C8H17 HO

H

O

H H

O

H

7

OH

Asimicin Scheme 10.20

Two successive CMs for the construction of asimicin.

O O

301

302


Two other syntheses of acetogenins involving CM were described in the same year by Crimmins et al. (mucocin) [33] and Hoye et al. (gigantecin) [34]. CM involving a diene partner can be difficult when the other olefin is hindered. Morita and Kuwahara had to employ an allylic pivalate as a diene precursor for the preparation of the dienoic fragment of elaiolide [35]. Koide and coworkers resorted to the use of the nitro-substituted [Ru]-VII [36] for the metathesis of a diene leading to FR901464, which is a potent anticancer product isolated from Pseudomonas sp. No. 2663 [37]. However, the CM of diene 66 with 2 equiv. of allylic alcohol 65 performed by Crimmins et al. led to diene 67 in 63% yield along with 31% of recovered 66 and the homodimer of 65, which could be recycled (Scheme 10.21) [38]. The E-selectivity on the other hand was total. Compound 67 was then transformed into apoptolidinone, the aglycone of apoptolidine, which induces cell death by inhibiting the mitochondrial F0 F1 -ATPase. Quinn et al. designed a tandem RCM/CM process for the synthesis of three members of the Annonaceous acetogenin family, muricatacin [39], rollicosin [40], and Me

Me

Me Me

HO

OTBS

66

Me

Me O OEt

[Ru]-II (10 mol%)

+

MeO

Me

HO

O

65

Me

Me

OEt

CH2Cl2 25 °C, 3 h 63%

O

O Me

O OMe H O

O Me

OTBS O OMe H O

MeO O OMe OTBS Me

Me

67

OTBS

OMe OTBS Me OTBS

1. TBSCl, Im (75%) 2. LiOH – H2O (77%) 3. 2,4,6-Cl3C6H2C(O)Cl, Et3N then PhMe, DMAP (68%) 4. H2SiF6 (61%)

Me

Me

HO Me Me

OH Me H

MeO HO Apoptolidinone

O O OH H O

Me OH

Scheme 10.21

OMe OH Me

CM with a diene partner for the synthesis of apoptolidinone.

10.5 Cascade Processes Involving CM

OBn

PhS 9

+

O O

68

OBn

O O

[Ru]-III (10 mol%) CH2Cl2, 40 °C 77%

69

303

PhS

O

9

O

O 70

O

OH

1. H2, Pd/C, EtOAc 56%

2. m -CPBA, CH2Cl2 3. PhCH3, 110 °C

Muricatacin

O O

OH 9

O O Scheme 10.22

OH

O

OH Rollicosin

O

O 9

O O Squamostolide

Tandem RCM/CM in the synthesis of squamostolide.

squamostolide [41] (Scheme 10.22). Formation of the unsaturated five-membered ring lactone by RCM and appending the side chain by CM of the other alkene occurred in one pot (Scheme 10.22). The optimized conditions necessitated slow addition of the catalyst by syringe pump to a 0.01 M solution of 68 and 69 in refluxing CH2 Cl2 over 6 hours and further heating for an additional 14 hours. The simple RCM product was observed as an intermediate when catalyst [Ru]-II was used (confirming that the RCM process is faster than the CM reaction), but not with complex [Ru]-III which proved to be superior in this case. Since homodimerization of type I olefin 69 was very fast, the CM leading to 70 likely involved this dimer. Interestingly, the other possible RCM product (the six-membered ring unsaturated lactone) was never observed. Simple transformations ultimately led to squamostolide in good overall yield.


Since olefin metathesis reactions involve alkenes both as starting materials and products, they are potential candidates for cascade processes. 10.5.1 ROM/CM

Parrain and coworkers described an efficient cascade involving the ring opening of cyclopropene ketal of type 71 followed by CM with a variety of alkenes in the presence of catalyst [Ru]-I for the synthesis of protected divinyl ketones [42]. This sequence was employed by Kozmin and coworkers for the preparation of the linear center portion of bistramide A, a marine metabolite isolated from Lissoclinum bistratum which exhibits potent cytotoxicity against several cell lines [43], as well as

O

304


for the synthesis of spirofungin A, an inhibitor of isoleucyl-tRNA synthase isolated from Streptomyces violaceusniger Tü [44], and routiennocin, an antibiotic agent extracted from Streptomyces routienii Huang sp. nov. (ATCC39446) [45]. For the latter case, ring opening of cyclopropene acetal 71 with catalyst [Ru]-II followed by CM with alkene 72 produced the elongated acetal, which proved inert to subsequent metathesis (Scheme 10.23). The corresponding dienone obtained by in situ acidic hydrolysis was subjected to a second CM with alkene 73 to furnish 74 in 59% yield. Subsequent hydrogenation and saponification of the methyl ester led to the target molecule. Asymmetric ring-opening metathesis (AROM) followed by CM proved very efficient for the synthesis of baconipyrone C realized by Hoveyda and Gillingham (Scheme 10.24) [46]. Hence, desymmetrization of compound 75 was best accomplished by ROM/CM with styrene in the presence of [Ru]-XV generated

Me

Me

O

O OH

MeO2C O

H N

OBn

71

OBn N

+ Me

O

Me

72

73 1. [Ru]-II (10 mol%) dil. aq. HClO4 (50%) 2. [Ru]-II (15 mol%) THF, 20 °C (59%) MeO2C

H N

O

OBn

OH

OBn N

O

O Me

Me

74 1. H2, Pd /C (62%) 2. aq LiOH, THF (82%)

O O Me H Me

N H H

O

CO2H OH

O NH Routiennocin

Scheme 10.23

Cascade ROM/CM of cyclopropene ketal for the synthesis of routiennocin.


Me

OPMB Me

+

O 75

Me

[Ru]-XV (2 mol%)

Ph

No solvent, −15 °C, 20 h 62%

(8 equiv.)

Ph

Ph Et

O

Me Me

Me

O

Me O

O

OPMB Me O 76 (ee = 88%)

Ph N

MesN I

O Me

OH O

Et

O

Ru

Me

O Oi -Pr

Et Baconipyrone C Ph

Scheme 10.24

[Ru]-XV

AROM/CM in the synthesis of baconipyrone C.

in situ. Interestingly, when the corresponding isolated catalyst was used [47], higher catalyst loading was required for a lower yield. 10.5.2 ROM/CM/RCM

A three-step process was reported by Tadano and coworkers for the construction of the 9-oxabicyclo[4.2.1]nona-2,4-diene core of mycoepoxydiene [48]. The reactions were first examined in a stepwise manner. Ring opening of 77 followed by CM of the carbene intermediates with 1,3-butadiene furnished a mixture of regioisomeric products 78 and 79 (Scheme 10.25). Since only the Z-stereomers can undergo subsequent ring closure, this reaction was optimized for the formation of the Z-olefins. Catalyst [Ru]-I gave the best ratio, while complex [Ru]-II always favored the thermodynamic E-isomers. Formation of cyclooctadiene 80 necessitated the more active catalyst [Ru]-II; therefore, for the one-pot procedure this carbene was added to the reaction after consumption of 77 and 1,3-butadiene degassing, and the mixture was heated at reflux. Desilylation was then effected to enable separation of 80 from the by-products probably resulting from the unproductive E-isomers of 78 and 79. 10.5.3 ROM/RCM/CM

Hart and Phillips have used a CM for the preparation of a direct precursor of vinyl ketone 81, see Section 2.1 (Scheme 10.4) [6]. Their synthesis of cylindramide A then involved an ROM/RCM cascade of norbornene derivative 81, followed by CM with alkene 82, which led to the fused 5,5-bicyclic compound 83 in 59% yield as a 2 : 1 mixture of separable E- and Z-diastereomers (Scheme 10.26).

305

306


Me O

Me O

(4 equiv.)

OTBDPS 77

[Ru]-I Benzene 20 °C

OTBDPS

78 (E /Z = 1 : 1.5) + Me O

OTBDPS

79 (E /Z = 1 : 1.5)

O

Me H O

Then [Ru]-II, reflux TBAF

O

23%

Me

Ref. 48

O

OH

AcO Mycoepoxydiene Scheme 10.25

80

ROM/CM/RCM process for the preparation of mycoepoxydiene.

OTIPS O

O +

O

O

OTIPS 81

82

Scheme 10.26

[Ru]-I (4 mol%)

O

CH2Cl2, 40 °C 59%

H O H

83 (E /Z = 2 : 1)

O O

ROM/RCM/CM cascade en route to cylindramide A.

10.5.4 CM/RCM

Smith et al. reported an elegant synthesis of cytotoxic agents cylindrocyclophanes A and F, which were isolated from the terrestrial blue-green algae Cylindrospermum lichenforme [49]. Dimerization of bis-olefin 84 with [Ru]-I, followed by RCM of the resulting dimer furnished exclusively the [7,7]-paracyclophane 85 in 61% yield as a single E,E-isomer (Scheme 10.27). The yield could not be improved with catalyst [Ru]-II; however, the use of [Mo]-I gave better results (72% yield). In addition, none of the ‘‘head-to-head’’ dimerization products was observed during this reaction. In an interesting experiment, these dimers, prepared by another route, were independently submitted to [Mo]-I and compound 85 was obtained as


[Ru]-I, CH2Cl2, 20 °C, 75 h (61%)

MeO

OMe

[Ru]-II, benzene, 40 °C, 27 h (58%) [Mo]-I, benzene, 20 °C, 2 h (72%)

84

MeO MeO

OMe

85 (E,E only)

R

84%

OH

HO HO

(1) H2, Pd/C (2) BBr3

OH R

Cylindrocyclophane F: R = H Cylindrocyclophane A: R = OH Scheme 10.27

Synthesis of cylindrocyclophane F using a CM/RCM sequence.

the exclusive product in each case, as the [8,6]-cyclophanes were not detected. The dimers probably undergo a cascade of metathesis reactions leading to the most stable [7,7]-paracyclophane 85. This compound was then converted in two steps into cylindrocyclophane F. 10.5.5 RCEYM/CM

Finally, metathesis between an alkene and an alkyne, first reported by Kinoshita and Mori [50], is a very efficient way of producing 1,3-dienes [51]. The ring-closing version of this reaction (ring-closing ene–yne metathesis (RCEYM)) leads to semicyclic 1,3-dienes, which can undergo CM with another alkene. Clark and coworkers utilized this sequence of reactions for the construction of polyether substructures found in the dinoflagellates toxins gambierol and hemibrevetoxin B [52]. Compound 86 was thus transformed into diene 87 by RCEYM, followed by CM with the dimer of allyl acetate in 62% overall yield (Scheme 10.28). When they used the one-pot process first reported by Grimaud, El Ka¨ım, and coworkers [53], 87 was produced in 54% yield along with 25% of the ene–yne RCM product. Hoveyda and coworkers also employed this sequence of reactions for the synthesis of the antimycobacterial agent erogorgiaene [54]. The stepwise operation was efficient but the yield for the one-pot operation was never over 20%.

307

OMe


308

BnO

BnO

H

H

OBn H O

1. [Ru]-II (5 mol%) H2C CH2 Toluene, 80 °C (82%)

O

OAc

2. AcO

H

Scheme 10.28

CH2Cl2, 20 °C, 24 h 76%

88

H

OBn H O

H

87

4 steps

MeOOC

Me

OBn H N

N Cbz

OH H N

O H

N

89

N

Ref. 55

COOEt

O Anthramycin Scheme 10.29

O H

O 90 COOEt [Ru]-III,CH2Cl2 20 °C, 17 h

Me

OAc O

Synthesis of a polycyclic ether subunit by an ene–yne RCM/CM sequence.

[Ru]-I, H2C CH2

N Cbz

BnO

(3 equiv.) Toluene, 80 °C (75%)

86

MeOOC

BnO

H

Me

60%

OBn H N

O H N

O 91

COOEt

Synthesis of an anthramycin derivative.

A derivative of the antitumor agent anthramycin was synthesized by Mori and coworkers by performing a CM reaction on a diene that was formed by ene–yne metathesis, but several steps separated the two operations [55]. [Ru]-I efficiently catalyzed the formation of five-membered diene 89 (Scheme 10.29). After four steps, the resulting product 90 was submitted to CM in the presence of ethyl acrylate and [Ru]-III to furnish compound 91. The dienamine system could be isomerized into the one present in anthramycin by heating 91 with rhodium trichloride (RhCl3 ·3H2 O) in ethanol at 110 ◦ C. 10.6 Ene–yne CM

Very few examples of ene–yne CM have been reported in the field of natural products synthesis. Mori et al. employed this reaction for the synthesis of anolignans A and B, two HIV-1 reverse transcriptor inhibitors extracted form Anogeissus acuminate [56]. CM of internal alkyne 92 with ethylene in the presence of [Ru]-II

10.7 Alkyne CM

309

OMs OAc

OMs

O O 92

OAc OMs

OH

O

93

Pd2(dba)3.CHCl3, Bu3P HCO2H, Et3N, THF, reflux

PhLi 66%

O Anolignan A Scheme 10.30

OMs

O

Toluene, 80 °C, 36 h 86%

OH

O

AcO AcO

H2C CH2 [Ru]-II (10 mol%)

89%

OMs

OMs

O O 94

Ene–yne CM with ethylene toward anolignan A.

led to the desired 2,3-substituted diene 93 in 86% yield (Scheme 10.30). The same reaction with [Ru]-I afforded 93 in 65% yield, along with 15% of recovered 92. Hydrogenolysis under palladium catalysis then furnished 94, which was converted to anolignan A by removal of the mesylates. A CM between a terminal alkyne and an allylsilylether was successfully performed by Rodriguez and coworkers for the preparation of the dienic system of mycothiazole [57]. Diver and coworkers also described an elegant ene–yne CM with 1,5-hexadiene followed by RCM en route to a cyclohexadienic intermediate of scabrosins [58].

10.7 Alkyne CM

Alkyne metathesis has only been recently applied to the synthesis of natural products [59]. The development of a well-defined molybdenum catalyst by Fürstner et al. largely contributed to this success. Up to now, few examples of the CM version have been reported. However, Fürstner et al. described the syntheses of dehydrohomoancepsenolide [60] and PGE2 methyl ester [61] using this strategy. The triaryl amido molybdenum complex [Mo]-V did not induce metathesis, but when it was activated by CH2 Cl2 , the corresponding chloro complex was formed, and CM of alkyne 95 with internal alkyne 96 led to compound 97 in 51% yield (Scheme 10.31). This alkyne was then hydrogenated and the silylether hydrolyzed to furnish PGE2 methyl ester in good overall yield. Another example of alkyne CM can be found in the synthesis of terpestacin reported by Chan and Jamison [62].

310


O MeO 3

+ TBSO

O

3

OTES 95

[Mo]-V (10 mol%) CH2Cl2

O OMe

O

Toluene, 80 °C 51% TBSO

96

H11C5

COOMe

OTES 97 1. H2 (1 atm), CH2Cl2 Lindlar catalyst (87%)

O N Mo

N

COOMe

2. aq. HF, THF (88%)

N HO [Mo]-V

Scheme 10.31

OH PGE2 methyl ester

Synthesis of PGE2 methyl ester using alkyne CM.

10.8 Conclusion and Perspectives

CM was developed later than RCM, but can now be found in numerous syntheses of natural products. This reaction has largely benefited from the discovery of second-generation ruthenium catalysts, and in most cases, high yielding chemoand E-stereoselectivities are obtained. More importantly, the outcome of CM can be predicted using the classification of olefins according to their ability to dimerize proposed by Grubbs and coworkers [2], making it a reliable reaction when planning the synthesis of complex targets. However, CM involving hindered olefinic partners can be difficult, and in many cases require addition of fresh catalyst after a certain reaction time or syringe pump addition of the catalyst. The construction of trisubstituted double bonds [63] and Z double bounds [64] by CM is also not an easy task, but progress is being made in this direction with the development of new Ru or Mo catalysts.

Acknowledgments

Financial support for this work was provided by the CNRS, the Ecole Polytechnique, and the Ecole Nationale Supérieure des Techniques Avancées.

References 1 For a previous review on the application

of olefin CM to the synthesis of natural products, see: Prunet, J. (2005) Curr. Top. Med. Chem., 5, 1559–1577.

2 (a) Chatterjee, A.K., Choi, T.-L.,

Sanders, D.P., and Grubbs, R.H. (2003) J. Am. Chem. Soc., 125, 11360–11370; (b) Blackwell, H.E., O’Leary, D.J.,

References

3

4

5 6 7

8

9

10

11 12

13 14 15

16 17 18 19

20

21

Chatterjee, A.K., Washenfelder, R.A., Bussmann, D.A., and Grubbs, R.H. (2000) J. Am. Chem. Soc., 122, 58–71. Karatholuvhu, M.S., Sinclair, A., Newton, A.F., Alcaraz, M.-L., Stockman, R.A., and Fuchs, P.L. (2006) J. Am. Chem. Soc., 128, 12656–12657. Van Zijl, A.W., Szymanski, W., Lopez, F., Minnaard, A.J., and Feringa, B.L. (2008) J. Org. Chem., 73, 6994–7002. BouzBouz, S. and Cossy, J. (2006) Tetrahedron Lett., 47, 901–904. Hart, A.C. and Phillips, A.J. (2006) J. Am. Chem. Soc., 128, 1094–1095. Evans, D.A., Chapman, K.T., and Bisaha, J. (1984) J. Am. Chem. Soc., 106, 4261–4263. Trost, B.M., Aponick, A., and Stanzl, B.N. (2007) Chem. Eur. J., 13, 9547–9560. Orita, A., Sakamoto, K., Hamada, Y., and Otera, J. (1999) Tetrahedron, 55, 2899–2910. For the synthesis of isoaltholactone, see: Harris, J.M. and O’Doherty, G.A. (2001) Tetrahedron, 57, 5161–5171. BouzBouz, S. and Cossy, J. (2001) Org. Lett., 3, 1451–1454. Hafner, A., Duthaler, R.O., Marti, R., Rihs, G., Rothe-Streit, P., and Schwarzenbach, F. (1992) J. Am. Chem. Soc., 114, 2321–2336. Gebauer, J. and Blechert, S. (2005) Synlett, 18, 2826–2828. Amans, D., Bellosta, V., and Cossy, J. (2007) Org. Lett., 9, 1453–1456. Huckins, J.R., de Vicente, J., and Rychnovsky, S.D. (2007) Org. Lett., 9, 4757–4760. Morril, C. and Grubbs, R.H. (2003) J. Org. Chem., 68, 6031–6034. Vincent, A. and Prunet, J. (2006) Synlett, 14, 2269–2271. Rahn, N. and Kalesse, M. (2008) Angew. Chem. Int. Ed., 47, 597–599. Bock, M., Dehn, R., and Kirschning, A. (2008) Angew. Chem. Int. Ed., 47, 9134–9137. Rikimaru, K., Mori, K., Kan, T., and Fukuyama, T. (2005) Chem. Commun., 394–396. Roulland, E. (2008) Angew. Chem. Int. Ed., 47, 3762–3765.

22 Ghosh, A.K. and Gong, G. (2007) Org.

Lett., 9, 1437–1440. 23 Chhor, R.B., Nosse, B., S¨ orgel, S.,

24

25 26

27

28

29 30

31 32 33 34

35 36

37

38

39 40

Böhm, C., Seitz, M., and Reiser, O. (2003) Chem. Eur. J., 9, 260–270. Chaterjee, A.K., Sanders, D.P., and Grubbs, R.H. (2002) Org. Lett., 4, 1939–1942. Qi, J. and Porco, J.A. (2007) J. Am. Chem. Soc., 129, 12682–12683. (a) Wipf, P., Rector, S.R., and Takahashi, H. (2002) J. Am. Chem. Soc., 124, 14848–14849; (b) Wipf, P. and Spencer, S.R. (2005) J. Am. Chem. Soc., 127, 225–235. Hu, Y.-J., Dominique, R., Das, K.S., and Roy, R. (2000) Can. J. Chem., 78, 838–845. Chen, G., Schmieg, J., Tsuji, M., and Franck, R.W. (2004) Org. Lett., 6, 4077–4080. Sheddan, N., Arion, V., and Mulzer, J. (2006) Tetrahedron Lett., 47, 6689–6693. Morimoto, S., Shindo, V., Yoshida, M., and Shishido, K. (2006) Tetrahedron Lett., 47, 7353–7356. Gebauer, J., Arseniyadis, S., and Cossy, J. (2007) Org. Lett., 9, 3425–3427. Marshall, J.A. and Sabatini, J.J. (2006) Org. Lett., 8, 3557–3560. Crimmins, M.T., Zhang, Y., and Diaz, F.A. (2006) Org. Lett., 8, 2369–2372. Hoye, T.R., Eklov, B.M., Jeon, J., and Khoroosi, M. (2006) Org. Lett., 8, 3383–3386. Morita, A. and Kuwahara, S. (2007) Tetrahedron Lett., 48, 3163–3166. Michrowska, A., Bujok, R., Harutyunyan, S., Sashuk, V., Dolgonos, G., and Grela, K. (2004) J. Am. Chem. Soc., 126, 9318–9325. Albert, B.J., Sivaramakrishnan, A., Naka, T., Czaicki, N.L., and Koide, K. (2007) J. Am. Chem. Soc., 129, 2648–2659. Crimmins, M.T., Christie, H.S., Chaudhary, K., and Long, A. (2005) J. Am. Chem. Soc., 127, 13810–13812. Quinn, K.J., Isaacs, A.K., and Arvary, R.A. (2004) Org. Lett., 6, 4143–4145. Quinn, K.J., Isaacs, A.K., De Christopher, B.A., Szklarz, S.C., and Arvary, R.A. (2005) Org. Lett., 7, 1243–1245.

311

312

10 Cross-metathesis in Natural Products Synthesis 41 Quinn, K.J., Smith, A.G., and

42 43

44 45 46

47

48

49

50 51

Cammarano, C.M. (2007) Tetrahedron, 63, 4881–4886. Michaut, M., Parrain, J.-L., and Santelli, M. (1998) Chem. Commun., 2567–2568. Statsuk, A.V., Liu, D., and Kozmin, S.A. (2004) J. Am. Chem. Soc., 126, 9546–9547. Marjanovic, J. and Kozmin, S.A. (2007) Angew. Chem. Int. Ed., 46, 8854–8857. Matsumoto, K. and Kozmin, S.A. (2008) Adv. Synth. Catal., 350, 557–560. Gillingham, D.G. and Hoveyda, A.H. (2007) Angew. Chem. Int. Ed., 46, 3860–3864. Van Veldhuizen, J.J., Campbell, J.E., Giudici, R.E., and Hoveyda, A.H. (2005) J. Am. Chem. Soc., 127, 6877–6882. Takao, K.-I., Yasui, H., Yamamoto, S., Sasaki, D., Kawasaki, S., Watanabe, G., and Tadano, K.-I. (2004) J. Org. Chem., 69, 8789–8795. Smith, A.B.III, Adams, C.M., Kozmin, S.A., and Paone, D.V. (2001) J. Am. Chem. Soc., 123, 5925–5937. Kinoshita, M. and Mori, M. (1994) Synlett, 1020–1022. For reviews on ene-yne metathesis, see: (a) Poulsen, C.S., Madsen, R. (2003) Synthesis, 1, 1–18; (b) Diver, S.T., Giessert, A.J. (2004) Chem. Rev., 104, 1317–1382; (c) Mori, M. (2007) Adv. Synth. Catal., 349, 121–135.

52 (a) Clark, J.S. (2006) Chem. Commun.,

53

54

55 56 57

58

59 60 61 62 63 64

3571–3581; (b) Clark, J.S., Elustondo, F., and Kimber, M.C. (2004) Chem. Commun., 2470–2471. Royer, F., Vilain, C., Elka¨ım, L., and Grimaud, L. (2003) Org. Lett., 5, 2007–2009. Cesati, R.R.III, de Armas, J., and Hoveyda, A.H. (2004) J. Am. Chem. Soc., 126, 96–101. Kitamura, T., Sato, Y., and Mori, M. (2004) Tetrahedron, 60, 9649–9657. Mori, M., Tonogaki, K., and Nishiguchi, N. (2002) J. Org. Chem., 67, 224–226. Rodriguez-Conesa, S., Candal, P., Jimenez, C., and Rodriguez, J. (2001) Tetrahedron Lett., 42, 6699–6702. Middleton, M.D., Peppers, B.P., and Diver, S.T. (2006) Tetrahedron, 62, 10528–10540. Fürstner, A. and Davies, P.W. (2005) Chem. Commun., 2307–2320. Fürstner, A. and Dierkes, T. (2000) Org. Lett., 2, 2463–2465. Fürstner, A. and Mathes, T. (2001) Org. Lett., 3, 221–223. Chan, J. and Jamison, T.F. (2004) J. Am. Chem. Soc., 126, 10682–10691. Stewart, I.C., Douglas, C.J., and Grubbs, R.H. (2008) Org. Lett., 10, 441–444. Ibrahem, I., Yu, M., Schrock, R.R., and Hoveyda, A.H. (2009) J. Am. Chem. Soc., 131, 3844–3845.

313

11 Cascade Metathesis in Natural Product Synthesis Marta Porta and Siegfried Blechert

11.1 Introduction

Alkene metathesis is well established as a valuable synthetic tool in organic chemistry, and has proven to be a unique technique for the construction of complex structures in a rapid and effective way [1]. Since all metathesis transformations (except diyne metathesis) can be promoted by the same carbene catalysts, it is possible to combine them in a sequential way and enable a metathesis cascade in one pot. A domino reaction has been defined as ‘‘a process involving two or more bond-forming transformations which take place under the same reaction conditions without additional reagents and catalysts, and in which the subsequent reactions result as a consequence of the functionality formed in the previous step’’ [2]. These domino processes, also called tandem, provide a tremendous increase in molecular complexity within the use of one single catalyst in one pot. By carefully adjusting the conditions and judicious choice of the catalytic complex, it is often possible to conduct the process with excellent regio-, chemo-, diastereo-, and even enantioselectivity. Domino processes are especially attractive not only because they provide elegance to the synthetic route but also for being ecologically and economically favorable since they minimize the amount of waste, solvents, reagents, and energy compared to stepwise reactions. Much effort has been invested in developing new methodologies involving metathesis domino processes that have been successfully applied to the preparation of biologically active and naturally occurring molecules. In this chapter, a selection of synthetic approaches with ruthenium-catalyzed domino reactions that include exclusively metathesis steps will be discussed, whereby syntheses involving tandem reactions connecting a metathesis reaction with a reaction of a different type [3] or several metathesis transformations occurring independently at the same time will not be described [4]. Through this collection of natural product syntheses, we would like to highlight the high efficiency and utility of metathesis domino reactions in organic synthesis, discuss its driving forces, limits, and scope, and ultimately illustrate the enormous synthetic potential of this transformation. Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts. Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32440-8

314

11 Cascade Metathesis in Natural Product Synthesis

i -Pr N F3C F3C O Mo O F3C

i -Pr Ph

Cl

PCy3

N Cl

Ru Cl

Ph PCy3

N Cl

N

Ru

Ru Ph

Cl

N

PCy3

Cl

O

R

i -Pr

CF3 [Mo]-I

[Ru]-I

[Ru]-II

[Ru]-III : R = H [Ru]-VII : R = NO2

Scheme 11.1

Carbene complexes used in metathesis cascades.

At the heart of advancement in metathesis and its establishment as a unique methodology for carbon–carbon bond formation lies the development of effective catalysts. A great variety of olefin metathesis catalysts have been developed over the past 20 years. Currently, there are up to 20 ruthenium catalysts commercially available; several reviews have been published with an overview of the most significant ones, where their reactivity, stability, and selectivity have been also addressed [5]. Despite the large number of these carbene complexes, catalysts [Mo]-I [6], [Ru]-I [7], [Ru]-II [8], and [Ru]-III [9] are the most widely applied in domino metathesis with just a few exceptions (Scheme 11.1). Catalyst [Mo]-I exhibits high olefin metathesis activity but lacks general functional group tolerance and suffers from strong air and moisture sensitivity. Functional group compatibility is crucial for the synthesis of complex molecules, and therefore commercially available catalysts [Ru]-I, [Ru]-II, and [Ru]-III have been utilized in synthesis to a greater extent than [Mo]-I. Catalysts of the first generation tend to furnish preferentially kinetic compounds, while [Ru]-II and [Ru]-III usually drive the reaction toward thermodynamically stable products. 11.2 RCM–CM Sequences

There are many combinations possible of metathesis reactions that can be used in domino metatheses, increasing process complexity with the number of transformations included. Natural product syntheses with ring-closing metathesis (RCM)–cross-metathesis (CM) sequences are described in this section. Ene–ene and ene–yne–ene cascade reactions in a two-metathesis step process will also be discussed. 11.2.1 Ene–ene RCM–CM 11.2.1.1 Synthesis of (3R,9R,10R)-Panaxytriol This type of metathesis cascade can be considered as an example of relay ring-closing metathesis (RRCM) (Chapter 9)[10]. This methodology is classically used in RCM of

11.2 RCM–CM Sequences AcO

[Ru] AcO

O

OAc

315

OAc 4 [1,3]-Shifts

[Ru]-II (10 mol%) + Et3Si 1

O C6H13

O

[Ru]

O

CH2Cl2, 40 °C 61%

C6H13 3

C6H13 O

O

O

O

O

C6H13

C6H13

C6H13

O

2

O

O

5 Cy3P Cl Ru Cl N N

6

OH

HO 7

C6H13

OH Panaxytriol

Scheme 11.2

Synthesis of panaxytriol.

sterically hindered olefins, which remains unreactive to standard metathesis conditions because of the difficulty of initiating the catalytic process. The relay approach involves incorporation of a temporary tether, containing a sterically accessible olefin, where the catalytic cycle is initiated. The purpose of this strategy is the formation of a kinetically favorable five-membered ring with concomitant delivery of a ruthenium carbene onto the originally less accessible position to facilitate the following metathesis step. In 2005, Lee et al. published an innovative tandem methodology combining metathesis and metallotropy for the construction of enediynes and oligoenynes [11a] and have recently applied this strategy in the synthesis of panaxytriol [11b], a component of Red Ginseng. Panaxytriol was first isolated in 1983 and exhibits inhibitory activity against a range of tumor cell types, such as human gastric adenocarcinoma, human breast carcinoma, and mouse lymphoma. Diyne 3 was synthesized in eight steps from 1 and 2 using a ruthenium-catalyzed Alder-ene reaction and a Cadiot–Chodkiewicz reaction as key transformations (Scheme 11.2). By treatment of 3 with catalyst [Ru]-II (10 mol%), propagating carbene 5 is generated, followed by two consecutive metallotropic [1,3]-shifts to form a new ruthenium species that undergoes CM with alkene 4, to finally complete the catalytic cycle. The desired product is obtained in 61% yield as a mixture of isomers with an unusual high E/Z selectivity (1 : 5). Interestingly, under this reaction conditions, it was possible to isolate the ruthenium complex 7 in 10% yield, suggesting that the carbene species derived from 7 is a viable intermediate in the catalytic cycle. 11.2.2 Ene–yne–ene RCM–CM

Enyne metathesis is widely used for its capacity to carry out multiple bond formations in tandem fashion, which is due to the ability of the alkyne moiety to

316

11 Cascade Metathesis in Natural Product Synthesis [Ru]-III (20 mol%) O (10 equiv.)

MeO

8

Scheme 11.3

O

O

O CH2Cl2, reflux 83%

O

O

O

9

O

OH

O

10

(+)-8- epi -Xanthatin

Synthesis of (+)-8-epi-xanthatin through a RCM–CM cascade.

form a conjugated ruthenium carbene as result of the reaction with an olefin, which ultimately can react with a second double bond to form a substituted 1,3-diene [12]. 11.2.2.1 Synthesis of (+)-8-epi-Xanthatin There is only one example in the literature where an enyne RCM coupled to an intermolecular CM in a sequence process has been applied in the synthesis of a natural product. The synthesis of (+)-8-epi-xanthatin, isolated from the leaves extracts from Xanthium canadense, was proposed by Martin et al. through an enyne RCM–CM cascade combining ring closing of 8 with a CM with 9 [13]. The special challenge for this transformation consists in the use of an electron-poor olefin for the final metathesis reaction. The use of [Ru]-III (20 mol%), the catalyst of choice for this type of transformations [14], and an excess (10 equiv.) of the methyl vinyl ketone 9 resulted in the formation of the desired product in 83% yield (Scheme 11.3).

11.3 Ene–yne–ene RCM–RCM

When an olefin happens to be attached to the ring that has been formed after an initial ene–yne RCM, a new bicyclic moiety can be formed by a second RCM. Such metathetical transformation is highly useful for the construction of polycyclic ring systems and it has been applied in the synthesis of natural products and derivatives to a great extent. Herein, we report a selection of the most significant examples (Scheme 11.4).

X

Ene–yne–ene RCM - RCM

R2

X

X = CH2, N

R1

Scheme 11.4

R2

General principle of ene–yne–ene RCM–RCM domino reactions.

11.3 Ene–yne–ene RCM–RCM O O H

O

O [Ru]-VII (10 mol%) CH2Cl2, rt 69%

NBoc

H

O H

O

O

RCM

H

RCM

NBoc

Ru

O

12

13

14 O

H

O

H

O [Ru]-VII (2 mol%)

OH NH

NBoc Et

(+)-Pipecolinic acid

CH2Cl2, rt 74%

O H

HO

NBoc

15

N (−)-Securinine

16

O H

O [Ru]-VII (1 mol%) CH2Cl2, rt

NBoc Et 17

O H NH 18 (20%)

+

O H

HO

NBoc

N

19 (80%)

RCM RCM

NBoc

Ru

NH

11

317

(+)-Viroallosecurinine

Scheme 11.5 Synthesis of (−)-securinine and (+)-viroallosecurinine through an ene–yne–ene metathesis cascade.

11.3.1 Synthesis of Bicyclic Structures 11.3.1.1 Synthesis of (−)-Securinine and (+)-Viroallosecurinine A suitable example of how to steer selectivity in an ene–yne–ene metathesis cascade is illustrated in the synthesis of securinine [15] and viroallosecurinine [16] published by Honda et al. Naturally occurring securinega alkaloids comprise about 20 tetracyclic compounds, isolated from Securinega suffruticasa in 1956, which show a great scope of biological activity. Securinine, the most abundant alkaloid of this group, has been used clinically since 1968 as a central nervous system (CNS) stimulating drug, acting as a stereospecific antagonist at the gamma aminobutyric acid (GABA) binding site of the GABA–receptor complex. First synthetic attempts with compound 11 using catalyst [Ru]-VII [17] delivered the undesired bicyclic structure 12, which results from the initial attack of the ruthenium carbene complex at the terminal olefin via intermediate 13 (Scheme 11.5). To direct the initiation of the catalytic cycle into the other double bond, through intermediate 14, Honda et al. decided to introduce steric hindrance using a disubstituted olefin in the alkyl chain. Their results showed, however, that the resulting compound was not reactive enough to undergo metathesis under the chosen reaction conditions. With the use of a more electron-rich allyl ether moiety such as in 15, it was possible to isolate the desired product in 74% yield and successfully complete the synthesis of securinine. Ruthenium complexes [Ru]-I and [Ru]-II were

318


also tested, leading to lower yields and requiring higher reaction temperatures as well as longer reaction times [18]. The same synthetic strategy was applied in the synthesis of its analog viroallosecurinine with the epimeric compound 17. Under the same catalytic conditions, the metathesis domino reaction delivered the desired product 19 in 80% yield together with compound 18 in 20% yield. 11.3.1.2 Total Synthesis of ent-Lepadin F and G Lepadins are members of the decahydroquinoline alkaloids, which are found in different marine sources, with a normally pentasubstituted cis-fused ring. Blechert et al. have recently reported the total synthesis of ent-lepadin F and G by a well-directed dienyne RCM cascade where aforethought functionalization of the corresponding olefins has permitted obtaining selectively the desired product [19]. When substrate 20 (Scheme 11.6) is subjected to metathesis conditions, two different reaction pathways can be envisioned: metathesis may initiate at the terminal double bond to produce ruthenium carbene intermediate 23, which then reacts with the alkyne moiety to furnish the desired bicyclic compound 21. Metathesis may also initiate on the disubstituted alkenyl moiety to produce ruthenium carbene species 24, which would then lead to the bicyclic structure 22. Selectivity in the metathesis cascade was achieved by a coordinative effect of the unprotected alcohol, which favors the initial catalyst attack of the allylic alcohol olefin, together with the use of a disubstituted alkene as a final cross-partner, which is a common strategy for the selective catalytic cycle initiation at the monosubstituted alkene. For the metathesis domino reaction, several conditions were tested with different catalysts at different temperatures. When the metathesis precursor 20 was exposed to catalyst [Ru]-II in CH2 Cl2 both at room temperature and at 40 ◦ C, no reaction occurred. The use of the more thermally stable

BnO OH

BnO [Ru]-I (10 mol%) H

N PMB OH

DCE, 60 °C 90%

OH

OH

H

O

N

H

PMB 21

20

O

N H H

BnO

BnO

OH H

[Ru]

N N PMB

22

Scheme 11.6

H

PMB OH 23

Synthesis of lepadine alkaloids.

O

N H H

ent -Lepadin G OBn

O H

[Ru] H 24

N PMB OH

ent -Lepadin F


catalyst [Ru]-III in toluene at 80 ◦ C resulted in complete conversion of 20 after 3 hours and the desired product 21 was obtained in 62% yield. Optimized conditions were found with the use of the more reactive but less stable catalyst [Ru]-I at 60 ◦ C in 1,2-dichloroethane (DCE). Under these conditions, 21 was isolated without significant formation of by-products and with an increased yield of 90%. 11.3.2 Synthesis of Tricyclic Compounds 11.3.2.1 Synthesis of (±)-Guanacastepene A The first application of an ene–yne–ene RCM–CM cascade in natural product synthesis was published by Hanna and Boyer in 2002 with their studies on the tricyclic ring systems related to guanacastepene A [20a]. Originally isolated from a fungus collected in the Guanacaste Conservation Area in Costa Rica, this diterpene became an attractive synthetic target because of its novel highly functionalized 5,7,6-ring fused structure and an interesting antibacterial activity revealed by biological studies. Hanna et al.’s synthetic approach started with the construction of trienes with different substitution patterns on the alkyne moiety (Scheme 11.7). In a first attempt, the use of catalyst [Ru]-II (10 mol%) in refluxing CH2 Cl2 resulted in a mixture of products. By lowering the reaction temperature, the desired product 29 was obtained in 40–60% yield along with remaining starting material. In their efforts to optimize the reaction conditions, the authors found that by stirring the

R

R [Ru]-II (7.5 mol%)

OSiEt3

CH2Cl2, rt

25 : R = H 26 : R = Me 27 : R = SiMe3 28 : R = CO2Me

OSiEt3 29 : R = H (80%) 30 : R = Me (70%) 31 : R = SiMe3 (0%, no reaction) 32 : R = CO2Me (93%) [Ru]-II (10 mol%), CH2Cl2, reflux MeOOC

COOMe

OHC O

O

OH

[Ru]-II (12 mol%) CH2Cl2, reflux 82 %

33

34

Scheme 11.7 Metathesis studies on (±)-guanacastepene A tricyclic core and its total synthesis.

(±)-Guanacastepene A

319

320


reaction mixture under a stream of nitrogen, so that the solvent slowly evaporated, a complete conversion of the starting material was achieved and 29 could be isolated in 80% yield. The use of other ruthenium complexes did not show any improvement and the starting diene–ynes were recovered unchanged. Methyl substitution on the alkyne moiety was also tolerated, but a bulkier substituent such as trimethylsilane (TMS) impeded the diene–yne metathesis cascade. In the presence of an electron-withdrawing group on the alkyne (compound 28), conversion to the desired product 32 was achieved in 93% yield, although harsher conditions were required. The total synthesis of guanacastepene A with a diene–yne RCM domino reaction as the key step was published two years later by the same group [20]. In this case, triene–yne 33 was efficiently converted to 34 in 82% yield by the second-generation catalyst [Ru]-II. In this unprecedented synthetic approach to this terpene skeleton, it is quite remarkable that the tandem metathesis reaction initiates at the relatively unreactive conjugated diene and also that the second RCM, despite steric hindrance at the trisubstituted olefin, can occur even at room temperature, probably because of the formation of a thermodynamically favored conjugated triene.

11.3.2.2 Approach to Taxane Analogs As shown above, domino RCM–CM can be successfully applied in the synthesis of tri- and tetracyclic structures. Granja et al. have recently reported a diene–yne domino metathesis to construct the tricyclic carbon framework characteristic for taxanes [21]. Their approach evidences that success of the metathesis cascade largely depends upon structural strain, and thus the sterical alignment of the reactant olefins, implying selective transformations. Their studies are based on an array of different metathesis substrates that differ in the grade of functionalization and in the configuration of each chiral center. Tactically, and in the same way as in 33, the terminal isopropylidene favors the initial incorporation of the ruthenium to the less-substituted double bond (Scheme 11.8). Their preliminary results showed that only the isomer 35 permitted obtaining the desired tricyclic core, while all other diastereoisomers yielded bicycle 39 resulting from the RCM between the alkyne and the terminal olefin. When the structural complexity of the skeleton was increased with the introduction of some relevant functional groups for the biological activity of taxanes, unfortunately, neither the hydroxyketones (R1 = OH and Y = CO) (structure 36) nor the corresponding 1,3-diols (R1 = OH and Y = CH–OH) were transformed into the desired tricyclic compound under the applied metathesis conditions. Instead, complete decomposition of the dienynes was observed. With the corresponding protected alcohols in compound 37 (R1 = OMEM and Y = CH–OMEM), it was possible to obtain the desired tricyclic structure, whereas all other stereoisomers at C2, C3, and C8 yielded again bicyclic structures of type 39 as a result of the interrupted domino reactions. When these trienes were subjected to metathesis conditions, they could not be converted to the tricyclic product, presumably because of steric hindrance of both olefins and structural strain in the metallacyclic intermediate.


R2

R2

8 3 *

8 * 3

OH

OH [Ru]-II (10 mol%) C6H6, reflux 70%

H

2

H

*

* * Y

* Y

2

R1

H

35

321

36

R1

39

AcO H

H

H [Ru]-II (10 mol%) * H OMEM OMEM

(S)-37 (R)-37

Scheme 11.8

*

C6H6, reflux

H

H

HO

OMEM OMEM

H

H OMEM OMEM

(S)-38 (70%) (R)-38 (70%)

Metathesis results on the synthetic approach to taxane analogs.

11.3.3 Synthesis of Natural Products Containing Tetracycles 11.3.3.1 Synthesis of Erythrina Alkaloids Erythrina alkaloids were isolated from several species of the Erythrina genus and display a variety of pharmacological effects, such as sedative, hypotensive, neuromuscular-blocking, and CNS activity. Hatakeyama et al. [22] and Mori et al. [23] published almost simultaneously a synthetic route to erythravine and erythrocarine, respectively, using a diene–yne RCM domino metathesis as the key step. For the synthesis of erythravine, protection of the alcohol functionality was necessary, whereby the choice of the protecting group had a very significant influence on the stereoisomer distribution, even though a 1 : 1 epimeric mixture was used as a starting material (Scheme 11.9). With the use of an ester as protecting group, the desired product was obtained as a 63 : 37 mixture of α and β epimers in 78% yield. The discrepancy between the initial and the obtained epimeric ratio may arise from different rates of cyclization and/or decomposition rates under the metathesis conditions. In Mori et al.’s approach, metathesis precursor 41 was synthesized with a nickel-mediated alkyne carboxylation as a key step. A better yield was achieved by converting the amino group into its corresponding hydrochloride. First-generation Grubbs catalyst [Ru]-I was again the catalyst of choice, which afforded the desired product 42 quantitatively as a 1 : 1 mixture of α and β isomers after 18 hours at room temperature. More challenging was the synthesis of the nonaromatic derivate erythroidine published by Hatakeyama et al. in 2006 (Scheme 11.10) [24]. RCM of 43, 46,

322


MeO

MeO N

MeO

N

MeO RO

MeO

RO

Substrate

Conditions

Yield (%)

a

b

0

–

–

R = TES

[Ru]-I or [Ru]-II (10–20 mol%) CH2Cl2 or toluene, rt to reflux [Ru]-II (10 mol%), CH2Cl2, reflux

63

27

73

R = Ac

[Ru]-II (10 mol%), CH2Cl2, reflux

78

63

37

R=H

N

MeO MeO

Erythravine

O N

O HO O OTBS 40

N

O

CH2Cl2, rt R 41 : R = a,b-OAc

Scheme 11.9

HCl [Ru]-I (10 mol%)

Quantitative

O N

O

Erythrocarine R

42 : R = a,b-OAc

Metathesis domino reaction toward erythravine and erythrocarine.

(S)- or (R)-48, and 51 using catalysts [Ru]-I and [Ru]-II, the last being generally more effective, afforded product mixtures containing starting material, bicyclic structures resulting from the RCM of the double bonds, and only 43 and (S)-48 delivered the desired tricyclic products 44 and 49, respectively, in moderate yields. In the case of 43, compound 45 resulting from the initiation of the metathesis cascade on the allylic ether olefin was also generated. Unfortunately, attempts toward the synthesis of erythroidine with 44 and 49 failed, presumably because of their instability under basic conditions. With compound 53 as substrate, which is structurally closer to the target molecule, erythroidine was furnished in 42% yield with [Ru]-I (10 mol%) (CH2 Cl2 , rt) together with 4% of 54, along with unconverted starting material (31%). In this case, [Ru]-II turned out to be less effective and gave erythroidine in less than 30% yield.

11.4 ROM–CM Sequences

Because of the reversible character of the metathesis itself, it is possible to perform ring-opening metathesis (ROM) of endocyclic olefins if the process is thermodynamically favored. This particular type of tandem process is especially atom-economical (as no ethylene is released), although the negative entropy

11.4 ROM–CM Sequences

MeO2C

CO2Me

MeO2C

MeO2C

MeO2C

MeO2C MeO MeO

N

MeO2C

MeO2C

+

N

43

N

HO MeO

44

MeO

45

46

47 H

OTBS

OTBS

O HO

O

HO

N O

N

O MeO

+

N

O

N

O

MeO

MeO 48

49

Substrate

50

51

Conditions

52

Products (Yield %)

43

[Ru]-I (10 mol%), CH2Cl2, reflux, 14 h

43 (27)

44 (0)

45 (26)

43

[Ru]-II (10 mol%), CH2Cl2, reflux, 1 h

43 (57)

44 (15)

45 (0)

46

[Ru]-I (10 mol%), CH2Cl2, reflux, 12.5 h

46 (55)

44 (7)

—

(S )-48

[Ru]-I (20 mol%), CH2Cl2, reflux, 14 h

48 (0)

49 (27)

50 (9)

(R )-48

[Ru]-II (20 mol%), CH2Cl2, rt, 4 h

48 (66)

49 (0)

50 (0)

[Ru]-II (20 mol%), CH2Cl2, reflux, 14 h

51 (0)

52 (55)

—

51

N

O

MeO

MeO

N

HO

N

MeO

OTBS

323

H MeOOC

O

NH2

[Ru]-I (10 mol%) N

O MeO

CH2Cl2, rt

O

O O

N

+

MeO MeO

MeO 53

Scheme 11.10

N

O

Erythroidine (42%)

Metathesis studies on the synthesis of erythroidine.

resulting from two molecules being combined into one must be overcome. Despite that, natural product syntheses has been reported using ROM–CM sequences where ethylene is used as cross-partner [25]; herein the focus will be on the use of more complex olefins as cross-partners (Scheme 11.11) [26]. 11.4.1 Synthesis of Bistramide A

Bistramide A is a novel marine metabolite of Lissoclinum bistratum which possesses a profound effect on cell cycle regulation, with highly selective

54 (4%)


324

X X

ROM-CM R

R

ROM CM X = CH2, N, O Scheme 11.11

General principle of ROM–CM domino reactions.

1. [Ru]-II (10 mol%) C6H6, 60 °C

O OBn O

2. 1 M H2SO4, MeCN

+

O

O OBn

O

N

N 63%

O

O 55 (1.5 equiv.)

57 (E / Z = 3 : 2)

56 (1 equiv.)

[Ru]-II (10 mol%) CH2Cl2, 36 °C 68%

H O

O

OBn

Me

O B n

58 (1 equiv.)

H H O HN

O

OBn

OBn

O

OBn N

OH 59

NH

HO Bistramide A

O

O

H

H O

H O

Scheme 11.12

Synthesis of bistramide A.

activation of a unique isotype from protein kinase C. Kozmin et al.’s inventive approach for the synthesis of the spiroketal domain involves a ROM–CM of cyclopropenone acetal 55 with alkene 56 (Scheme 11.12) [27]. This transformation was catalyzed by catalyst [Ru]-II (10 mol%), and only a slight excess (1.5 equiv.) of 55 was needed to afford dienone 57 in 63% overall yield after acidic hydrolysis. Interestingly, the corresponding intermediate acetal proved to be inert toward subsequent metathesis, which also justifies the absence of the formation of double-cross-coupled product and also impedes the performance of a ROM-double CM domino process. The next envisaged CM step had to be performed separately, and the desired product 59 was isolated in 68% yield after treating 57 with 10 mol% catalyst [Ru]-II and 1 equiv. of olefin 58.

11.5 RCM–ROM Sequences – Ring-rearrangement Metathesis (RRM)


So far the synthetic potential of RCM and ROM has been exemplified when combined with CM in a domino process to construct demanding structures, in terms of complexity and functionalization, in a rapid manner starting from relatively easily accessible substrates. When RCM and ROM themselves are sequentially combined, this domino process is classified as ring-rearrangement metatheses (RRM). Among all the possible combinations of metathesis reactions, this methodology has proven its extraordinary synthetic potential when it comes to constructing carbocycles and heterocycles, serving as a key step in an increasing number of natural product syntheses since 1999, when the first application of ruthenium-catalyzed RRM in the synthesis of (−)-halosaline was published [28]. By definition, in these domino reactions a strained carbocyclic alkene is transformed into a new cycle by intramolecular ROM–RCM with an exocyclic double bond. Since stereocenters remain unchanged during the metathesis reaction, probably the most important feature of RRM is the capacity to transfer any stereochemical information from the starting material into the product. While the application of RCM can be limited by the accessibility of the chiral acyclic precursor, RRM offers the possibility of transferring stereocenters, easily accessible in a ring system, to the side chain or vice versa, and thus avoids complicated multistep syntheses of chiral structures which are otherwise difficult to access (Scheme 11.13). Ring-rearrangement domino processes can be applied to mono- and polycyclic structures of different sizes. Because of the reversibility of this metathesis cascade, driven by thermodynamic factors (loss of ring strain) as well as kinetic effects (such as the formation of a less reactive carbene complex), the strategic selection of protecting groups, reaction conditions, and electronic properties of the reacting parts is crucial in order to shift the equilibrium to the desired product. Bicyclic systems deserve special attention among the different types of substrates suitable for metathesis reactions. Owing to the highly strained ring system, these substrates are highly reactive, and the driving force for the reaction can be clearly attributed to the release of energy. The introduction of an external olefin, usually ethylene, is necessary in most of the cases to avoid oligomerization, a common side reaction in RRM.

R * ()n

RCM R *

()

n ROM

X *

RRM

X * () m

() m X = CH2, N, O

Scheme 11.13

General principle of the RRM cascade.

325

326

11 Cascade Metathesis in Natural Product Synthesis CH(OEt)2

[Ru]-I (5 mol%) H2C CH2 (EtO)2HC AcO

CH2Cl2, rt

R

3

N

N

R N 3

R N

N

R Tetraponerine T7

n

60 : n = 2, R = p -Ns 61 : n = 2, R = Cbz 62 : n = 1, R = p -Ns [Ru]-I (5 mol%) H2C CH2 CH2Cl2, rt 89%

Scheme 11.14

N H

63 : R = p -Ns (71% conversion) 64 : R = Cbz ( 97%) AcO

H

CH(OEt)2

p -Ns

N

H

3

N H

N

N

R 65

Tetraponerine T6

Synthesis of tetraponerines.

11.5.1 RRM of Monocyclic Substrates 11.5.1.1 Synthesis of Tetraponerines An early application of RRM in target-oriented synthesis consisted in a skeletal reorganization of enantiomerically pure five-membered carbocycles (Scheme 11.14), which were readily accessible with a palladium-catalyzed allylic substitution of cyclopentene derivatives. Four tetraponerine alkaloids T4, T6, T7, and T8 isolated from the New Guinean ant Tetraponera were successfully synthesized by Blechert et al. with a flexible approach that permitted constructing both five- and six-membered heterocycles by varying the length of the N-alkenyl substituent on the cyclopentene [29]. As a representative example, subjecting metathesis precursors 60 and 62 to catalyst [Ru]-I (5 mol%) resulted in 71 and 91% conversion, respectively, after two days under ethylene atmosphere. By changing the N-protecting group to benzyloxycarbonyl (Cbz), the transformation proceeded quantitatively in 12 hours at room temperature and tetrahydropyridine 64 was isolated in 97% isolated yield. A plausible explanation might be attributed to a change in the complexation behavior with the ruthenium carbene during the course of the reaction. 11.5.1.2 Synthesis of (−)-Swainsonine and (+)-Castanospermine Members of the polyhydroxylated indolizidine family have received continuous, intense scrutiny from synthetic chemists owing to their densely functionalized, stereochemically rich structures. Swainsonine, isolated from the fungus Rhizoctonia leguminicolain, has especially attracted great attention mainly because of its important antimetastatic, antitumor-proliferative, anticancer, and immunoregulating activities, being the first compound to be selected for testing as an anticancer drug, reaching Phase I in clinical trials. Blechert et al. presented a synthetic approach

11.5 RCM–ROM Sequences – Ring-rearrangement Metathesis (RRM) Ts

O


TBSO

N O

Ts

N

TBSO

CH2Cl2, reflux 98%

Ts

66

Ac AcO

N

[Ru]-II (5 mol%) H2C CH2 CH2Cl2, reflux 97%

N

NHAc BnO Ac

OTBS N

[Ru]-II (5 mol%) H2C CH2 CH2Cl2, reflux 93%

70

OTBS H Ac 69

OBn H Ac 71

OH

(−)-Swainsonine H

OH OH

N

(−)-Swainsonine

TBSO

N

H OH

N

OH

BnO

68

OAc

OH

N

67

OTBS

BnO

H

327

HO HO

OH OH H N

(+)-Castanospermine

Scheme 11.15 Synthesis of (−)-swainsonine and (+)-castanospermine by RRM of cyclopentene derivatives.

to this polyhydroxylated indolizidine alkaloid based on RRM of a cyclopentene derivative obtained by palladium-catalyzed desymmetrization of cyclopentenediol [30]. When the metathesis precursor 66 (Scheme 11.15) was treated with commercially available [Ru]-I (5 mol%) under ethylene atmosphere, the desired rearranged product could be isolated in a yield of up to 98%, whereas rearrangement of its benzyl ether analog afforded a starting material/product ratio of 18 : 1. The sterically demanding tert-butyldimethylsilyl (TBS) ether proved to be essential to shift the rearrangement equilibrium completely toward the product, apparently due to an increase of the steric interactions between the substituents, thus facilitating ring opening and also preventing product dimerization. Mariano et al. focused on the RRM reaction of unsymmetrically substituted aminocyclopentenediol derivatives to investigate the regiochemical and stereochemical control of the process [31]. Tetrahydropyridine 68 rearranged to 69, in which the benzyloxy group is positioned next to the N-acyl group, whereas reaction of diastereomer 70 selectively furnished the regioisomer 71 in 97% yield. The high selectivity of this process was ascribed to an energy difference between the ruthenacyclobutane intermediates, thus suggesting the influence of kinetic factors. Tetrahydropyridine 69 was further converted into (−)-swainsonine, whereas 71 served as starting material in the synthesis of (+)-castanospermine (Scheme 11.15) [32]. 11.5.1.3 Synthesis of (+)-trans-195A RRM is not limited to cyclopentene derivatives, although applications of such metathesis domino process in six-membered cyclo-olefins have seldom been reported. Even though ring rearrangement of cyclohexenes to five-membered heterocycles is thermodynamically disfavored, synthesis of tetrahydropyridine

328

11 Cascade Metathesis in Natural Product Synthesis O

O

Br Cl +

p-Ns

N

H

[Ru]-I (5 mol%) H2C CH2 CH2Cl2, reflux 96%

N

N H H

p -Ns 72

Scheme 11.16

73

(+)-trans -195A

Application of RRM of cyclohexenes in the synthesis of (+)-trans-195A.

derivatives can be achieved, thereby indicating that the resulting heterocycle is thermodynamically more favored than the corresponding cyclohexene. This concept was used in the synthesis of the poison frog alkaloid (+)-trans-195A, where 72 was selectively converted to the cis-2,6-disubstituted tetrahydropyridine 73 in 96% yield, with catalyst [Ru]-I (5 mol%) under an ethylene atmosphere (Scheme 11.16) [33]. 11.5.1.4 Synthesis of (−)-Centrolobine – Diastereoselective RRM (d-RRM) There are two possible pathways for the generation of a new chiral center by means of metathesis: asymmetric induction by a chiral complex itself (asymmetric metathesis, Chapter 12), or diastereoselective induction where the chiral information present in the substrate influences the sterics of the newly formed chiral center. Metathesis domino processes can also be performed in a diastereoselective manner generating new stereocenters with high selectivity. The challenge for such a process is to shift the reaction to the desired isomer, whereby the reaction can be kinetically or thermodynamically controlled. The selectivity is mainly catalyst dependent; thus electronic and, specially, steric effects of the remaining ligands in the initiated catalytic complex are decisive for the reaction outcome. Blechert and Böhrsch presented in 2006 the single application in natural product synthesis of this methodology with the construction of (−)-centrolobine [34], a tetrahydropyranic antibiotic that has shown activity against leishmania amazonensis promastigotes, a major health problem in Brazil. The key features of its synthetic approach involve not only a d-RRM but also three consecutive transformations catalyzed by complex [Ru]-II. The enantiopure metathesis precursor 74, obtained by an asymmetric iridium-catalyzed allylic etherification, undergoes a ROM–RCM domino reaction in which the stereocenter at C5 is established diastereoselectively. Different conditions and precatalysts were tested: [Ru]-I (5 mol%) led to poor diastereoselectivity (cis : trans = 2 : 1), whereas the employment of [Ru]-II (3 mol%) in benzene provided a significant improvement of the diastereoisomeric ratio to 4 : 1 (cis : trans). Optimized reaction conditions were found to be under an ethylene atmosphere in saturated CH2 Cl2 at 50 ◦ C in a pressure vessel. A selective double bond isomerization of the terminal alkene with in situ generated ruthenium hydride species followed the metathesis cascade. A one-pot CM-hydrogenation procedure completed the synthesis of centrolobine in five steps, with 22% overall yield and 98% ee (Scheme 11.17).


329

Naph O P O

MeO

MeO

Naph

OH

[Ru]-II (3 mol%) H2C CH2

[Ir(COD)Cl]2 +

O

O CH2Cl2, 50 °C

THF, 0 °C to rt 87% BocO

MeO

(cis/trans = 4 : 1)

74 (ee = 98%) 1. [Ru]-II (10 mol%) toluene, rt OH

MeO

OH

NaBH4 (0.4 equiv.) 55% (2 steps from 74)

MeO

O

O 2. Pd/C 5%, H2 50%

(−)-Centrolobine

Scheme 11.17

d-RRM cascade in the synthesis of (−)-centrolobine.

X1 ROM

RRM

X2

X2 RCM

X1

n

n X1, X2 = CH2, N, O

Scheme 11.18

General principle of RRM of norbornene derivatives.

11.5.2 RRM of Bicyclic Substrates

Norbornene derivatives are common substrates for ring-rearrangement domino processes. Because of their highly strained structure, equilibrium is strongly shifted to the product and, owing to the highly stereocontrolled norbornene synthesis, it is granted easy access to new architectures which otherwise would be difficult to synthesize (Scheme 11.18). In general, different mechanistic pathways can be taken into account for these domino reactions (Scheme 11.19). The first would involve initial attack at the terminal exocyclic double bond, which would then undergo metallocycloaddition onto the endocyclic olefin. Alternatively, ring opening may occur first leading to two possible intermediates A and D. As observed in early studies [35], favored formation of A or D, and therefore B or E, depends on the ring size of the final product as well as on the presence of ethylene in the reaction mixture. Under these conditions, it is possible to convert A into C and shift the equilibrium toward


330

R

Ru

R

Ru

R

n

n

n

A

B

R

R

n

n

C H

R

R Ru

R

n Ru

n

n

H E

D

R

R

Ru

R

Ru

n

n

n

Ru Scheme 11.19

Mechanistic aspects of RRM domino processes.

the most stable compound E. Moreover, dimeric or oligomeric materials can be reverted back to monomeric ruthenium-alkylidene, thereby avoiding the formation of the major by-products of these type of domino metathesis. Attack at the less hindered and kinetically favored olefin cannot be excluded, although it seems more plausible that the metathesis cascade starts with ROM, with loss of ring strain being the main driving force. 11.5.2.1 Synthesis of Indolizidine 251F, (±)-trans-Lumausyne and Aburatubolactam A Following the pioneering work on the synthesis of [x.3.0] bicyclic systems by RRM from norbornene derivatives [35], Aubé et al. applied this strategy to synthesize indolizidine alkaloid 251F, where enone 76 was obtained from the rearrangement of norbornene 75 (Scheme 11.20) [36]. Early attempts afforded the desired product in only poor yield (30%) mainly because of olefin oligomerization. Under an atmosphere of ethylene, it was possible to optimize the metathesis cascade and the bicyclic compound 76 was isolated in 93% yield with the use of catalyst [Ru]-I (5 mol%). An equivalent approach was used by Chandler and Phillips in the synthesis of trans-kumausyne, isolated from red algae of the genus Laurencia in Japan [37].

11.5 RCM–ROM Sequences – Ring-rearrangement Metathesis (RRM) [Ru]-I (5 mol%) H2C CH2

O

Me H

H

H OH

93%

O

H

O


O

O

H

H

N3

Indolizidine 251F AcO

H

CH2Cl2, rt

H

O

O

H

90% 77

N

Me

76

75

H OH

CH2Cl2, rt O

331

H

Br

(±)-trans-Kumausyne

78

H H

O

[Ru]-I (2.5 mol%) H2C CH2

O

H

90%

H

H H

CH2Cl2, rt

79

H H

H SnBu 3

H

O O

80

NH

O N

O

O O

HO Aburatubolactam A

Scheme 11.20 Rearrangement of norbornene derivatives in the synthesis of indolizidine 251F, (±)-trans-kumausyne and aburatubolactam A.

Subjecting the oxanorbornene derivative 77 to catalyst [Ru]-I (5 mol%) in CH2 Cl2 under an atmosphere of ethylene resulted in smooth ROM–RCM cascade, providing tetrahydrofuran 78 in 83% yield. The latest of the applications of this type of RRM cascade has been recently published by the same group in the synthesis of aburatubolactam A [38]. This macrolactam, isolated from bacteria of a marine mollusc collected in Japan, is a member of a growing class of mixed polyketide amino acid metabolites that displays a diverse range of biological activities including cytotoxicity, antimicrobial activity, and the inhibition of superoxide generation. Phillip et al.’s approach, which included Stille coupling and Lacey–Dieckmann cyclization as key features, started with the synthesis of 80 through a ROM–RCM domino process. When treated with the first-generation catalyst [Ru]-I under an ethylene atmosphere, enone 79 was rearranged to the desired bicyclo[3.3.0]octenone 80 in 90% yield. The absence of an ethylene atmosphere decreased the efficiency of the transformation and led to 80 in 41% yield. 11.5.2.2 Synthesis of (+)-ent-Lepadin B Also very recently, Barbe and Charette published another domino metathesis in the synthesis of ent-lepadin B [39]. This lepadine was isolated from the tunicate of Clavelina lepadiformis and was shown to exhibit significant in vitro cytotoxicity

332


H Bz N

N Ph

[Ru]-II (2 mol%) Toluene, reflux 79%

N OMe

OBn 81

Scheme 11.21

H

Bz

H H N

N H H OBn

OH

ent- Lepadin B

82

Synthesis of ent-lepadin B by RRM cascade.

against several human cancer cell lines. Lepadin B has been recently identified as a potent blocker of neuronal nicotinic acetylcholine receptors, which are implicated in several neurological disorders including nicotinic addiction, epilepsy, as well as Parkinson and Alzheimer diseases. Subjecting 81 to catalyst [Ru]-II (2 mol%) (toluene, 80 ◦ C) afforded the desired rearranged product 82 in 79% yield after only 2 minutes (Scheme 11.21). Interestingly, performing the reaction under an ethylene atmosphere decreased the reaction rate and did not provide any yield improvement.

11.6 RCM–ROM Sequences Combined with Other Metathesis Reactions

The synthetic potential of RRM can be amplified with the introduction of a third olefin, permitting the access to much more complex structures in one single transformation, by a further intermolecular CM or a second RCM. As discussed above, two mechanistic pathways can be envisaged. The metathesis cascade may involve an initial formation of a ruthenium–alkylidene complex at the terminal olefin, or alternatively, the sequence may begin with [2+2]-cycloaddition of the ruthenium carbene onto endocyclic double bond, followed by ROM–RCM or CM. Neither of these possibilities can be ruled out, and the occurrence of one or the other may be influenced according to ring strain, steric and electronic properties of the corresponding olefins, and also the reaction conditions employed (Scheme 11.22). 11.6.1 RCM–ROM–RCM

These types of transformations were first explored by Grubbs et al. [40] and have often been applied in the synthesis of many natural products. From a mechanistic point of view, these metathesis cascades possess an additional driving force, apart from the release of ring strain, which consists in an entropy gain with the release of a molecule of ethylene as volatile by-product. Like other RRM reactions discussed, oligomerization is an important side reaction, which may be suppressed by the use of an ethylene atmosphere and by performing the reaction at higher dilution.


RCM CM R2

R1

R2 ROM Intermolecular

m X2

k X1

RCM

n

m R1

k X1

R1 = X2

k Intramolecular

n

X1

n

X1, X2 = CH2, N, O

Scheme 11.22

General concept of the RCM–ROM–RCM and RCM–ROM–CM sequences.

11.6.1.1 RCM–ROM–RCM Cascades of Monocyclic Structures Synthesis of (−)-Anaferine, (+)-Dihydrocuscohygrine, Cuscohygrine Synthetic applications of RCM–ROM–RCM cascades have been mainly championed in Blechert et al. in the elegant synthesis of a variety of structurally diverse alkaloids. With a flexible synthetic approach, ring rearrangement of the seven-membered rings 83 and 84 led to the core of (−)-anaferine [41], (+)-dihydrocuscohygrine, and cuscohygrine [42] by just modifying the N-alkenyl chain length (Scheme 11.23). Tandem RRM was carried out under an inert atmosphere with catalyst [Ru]-I in refluxing CH2 Cl2 and provided bis-tetrahydropyridine derivative 85 after 48 hours in 87% yield. However, in the case of 83, a one-pot metathesis cascade-hydrogenation was performed in order to avoid ruthenium-promoted decomposition of the rearranged products, and the dipyrrolidine derivative 86 was isolated in a satisfactory 72% yield. Synthesis of (+)-Dumetorine A and (+)-Astrophylline The first enantioselective synthesis of (+)-dumetorine was also achieved through an RCM–ROM–RCM domino reaction as a key transformation in only six steps and a 27% overall yield from the functionalized cyclopentene 87 [43]. Preliminary investigation with catalyst [Ru]-I (10 mol%) allowed obtaining the desired product 89 in only 7% yield. The low yield was attributed to the preferential formation of a stable cyclic carbene complex 88 (Scheme 11.24) which inhibited further reaction and interrupted the catalytic cycle. With the use of Ti(Oi-Pr)4 (30 mol%) as a co-additive, the formation of the chelated intermediate was inhibited and the desired ring-rearranged product 89 was obtained in good yield (80%) with only 5 mol% of catalyst [Ru]-I. The first synthesis of (+)-astrophylline was performed in a similar manner by Blechert and Schaudt [44]. The functionalized cyclopentene 91, in which the stereocontrolled generation of the 1,2-trans relative configuration was achieved via a n-BuLi-mediated [2,3]-Wittig–Still rearrangement of 90, was successfully transformed to 92 with the use of the second-generation catalyst [Ru]-II (1 mol%).

333

334

11 Cascade Metathesis in Natural Product Synthesis OTBS

O

N

N

Cbz

Cbz

N H

85 [Ru]-I (5 mol%) CH2Cl2, reflux

(−)-Anaferine

CO2Et

N OAc

N H

87%

EtO2C HO

O

N

n

n

OTBS 1. [Ru]-I (5 mol%) CH2Cl2, reflux

83 : n = 1 84 : n = 2

OTBS

O

72%

2. H2, Pd /C, EtOH

N Cuscohygrine

OTBS N

OH

N

EtO2C

N

CO2Et 86

N

N

(+)-Dihydrocuscohygrine

Scheme 11.23 RRM domino processes of seven-membered rings in the synthesis of naturally occurring alkaloids.

C22–C34 Subunit of Halichondrin B A more sterically challenging rearrangement of this type was employed by Burke et al. in the synthesis of a known synthetic precursor of halichondrin B [45]. This polyether macrolide, isolated from marine sponges, is a tubulin-interactive antimitotic agent that has been recommended by the National Cancer Institute for stage A preclinical development because of potent activity against various chemoresistant human solid tumor xenografts. In initial investigations, treatment of triene 93 with [Ru]-I (25 mol%) led to the formation of a mixture of diastereomeric dihydropyrans 95 and 96 in 21% yield (Scheme 11.25) as a result of CM at the catalytic cycle initiation. The corresponding propagating carbene species failed to undergo intramolecular ring closing, presumably because of steric hindrance in the [2+2]-cycloaddition, resulting in interruption of the metathesis sequence. While efforts to optimize the transformation with an ethylene atmosphere yielded only oligomerized products, the use of the more reactive, but also more sensitive, Schrock catalyst [Mo]-I effected the desired transformation and provided the C22–C34 halichondrin B precursor 94 in 79% yield. Synthesis of (−)-Halosaline and (−)-Indolizidine 167B The synthesis of halosaline presents an early application of domino metathesis in natural product synthesis


335

O O

Ru

[Ru]

N

O

N

CH2Cl2, reflux

O

Boc

Boc

7%

N Boc

O

O

[Ru]-I (10 mol%)

87

88

Ti(Oi -Pr)3 [Ru]

[Ru]-I (5 mol%) Ti(Oi-Pr)4 (30 mol%)

O

N

CH2Cl2, reflux 80%

O

N

Boc

N

O

O

O

O

Boc

Boc

(+)-Dumetorine

89 Bu3Sn O

H

[Ru]-II (1 mol%) N Cbz

N N

H

H

N

Boc

CH2Cl2, reflux 82%

N H

Boc

H

N

N

Cbz

Boc 90

91

Ph

O (+)-Astrophylline

92

Scheme 11.24 RRM of cyclopentene derivatives in the synthesis of (+)-dumetorine A and (+)-astrophylline.

O O

O H H

O

14 17

O O

O

TBSO

O

O OTBS

C6H6, 60 °C 79%

O

1 O 30

[Mo]-I (25 mol%) O

O

H

OTBS

H Ph

Ph

O

H 34 H

H

O

OTBS 94

93

22 O

38 O

[Ru]-I (25 mol%) +

Toluene, 80 °C O 21% OTBS

O OTBS

O

95

H

O

OTBS

OTBS 96

Halichondrin B

H O 44 H

O

O

H

H 53 OH

HO HO

Scheme 11.25 Synthesis of a halichondrin B precursor via ring-opening double ring-closing metathesis.


336

TBSO OTBS

OTBS

[Ru]-I

N

N

H Ts

Ts N 97 Si O

O

CH2Cl2, rt

N

OH

TBAF 78%

H Ts

100

99

Si

N

SiMe3

H

98


Ts N

Ts

101

OH

N

N H H

H Ts

(−)-Halosaline

102

Cbz [Ru]-I (5 mol%) H2C CH2 CH2Cl2, rt

O Si 103

Scheme 11.26

N Cbz

O

Si

OH

TBAF 78%

N Cbz

104

H N

(−)-Indolizidine 167B

RCM–ROM–RCM cascades of silylethers in the synthesis of alkaloids.

by Blechert et al. [28]. Synthetic studies toward this alkaloid began with a tandem RCM–ROM–CM sequence using allyltrimethylsilane as cross-partner to generate 99. Performing the transformation with [Ru]-I as the catalyst led to the desired metathesis product 99 in less than 10% yield. The main product obtained was 98, suggesting that steric hindrance hampered the CM step. The use of catalyst [Mo]-I, which is known to give higher yields for sterically hindered substrates (as seen in the synthesis of 94), only resulted in CM of the terminal double bond of cyclopentene 97 with allyltrimethylsilane. A clever modification of the metathesis substrate was the addition of a tethered functionality, which enabled the synthesis of 102 by a one-pot RRM desilylation procedure [46]. Replacement of the CM step by an additional RCM overcame the problem of steric hindrance and allowed the use of a cheaper and less sensitive ruthenium catalyst. Compound 102 was finally obtained in an increased overall yield of 78% with catalyst [Ru]-I (5 mol%) (Scheme 11.26). A similar approach served the same group in the synthesis of (−)-indolizidine 167B, an alkaloid isolated from a neotropical frog species. RRM of 103 with [Ru]-I (5 mol%), followed by cleavage of the silyl ether functionality, furnished 104 in 78% yield. With this procedure, Blechert et al. also exemplified that RCM–ROM–RCM domino processes are not restricted to the formation of same sized bicyclic systems. 11.6.1.2 RCM–ROM–RCM Cascades of Bicyclic Structures Synthesis of (+)-Cyanthiwigin U Phillips and Pfeiffer have also combined the RRM metathesis of bicyclic compounds described in Section 4.2 with another metathesis


337

OPiv [Ru]-II (20 mol%) H2C CH2

CHO O

+ CHO

O OH

O 105

Toluene, reflux 43% (3 steps from 105)

H O

106

H

O H

O

ROM-RCM Two–directional tandem ROM-RCM

H HO

H O

(+)-Cyanthiwigin U

ROM-RCM

Scheme 11.27

O H

RCM–ROM–RCM cascade in the synthesis of (+)-cyanthiwigin U.

step [47]. In this case, the substituted bicyclo[2.2.2]octane 106 served as a precursor for the synthesis of cyanthiwigin U, a diterpenoid isolated from both fungal and marine sources. An elegant directional domino ROM–RCM sequence enabled the construction of the desired tricyclic core in the presence of catalyst [Ru]-I (20 mol%) and CH2 Cl2 atmosphere. Four further transformations completed a concise synthesis of this complex compound in 12 steps and 17% overall yield (Scheme 11.27). 11.6.2 RCM–ROM–CM

The domino processes discussed so far concern intramolecular reactions. Extension of the intramolecular domino metathesis processes described in previous sections to an intermolecular reaction, namely, CM, was also successfully applied in organic synthesis (Scheme 11.22). Although CM is afflicted with difficulties in E/Z selectivity and chemoselectivity (three different products are possible in a reaction with two different olefins), the choice of suitable substrates and catalysts alleviates the limitation of these problems. Both mono- and bicycles are suitable substrates. In this case, however, the loss of ring strain, which usually shifts the domino metathesis process toward the desired product in RRM of bicyclic structures, does not guarantee a successful reaction, as CM displays the main origin of side reactions. 11.6.2.1 Synthesis of (−)-Lasubine II Following the studies on the synthesis of halosaline, Blechert and Zaja presented the synthesis of (−)-lasubine II [48]. Cyclopentenone 107 was chosen as the metathesis substrate in order to avoid the failure of the CM step observed in 98 with the incorporation of the bulky TBS group. Rearrangement of 107 in the presence

338


O

+ N Boc

OMe

CH2Cl2, reflux 48%

H

O

OMe [Ru]-II (5 mol%)

N

OMe

N

OH

Boc

OMe MeO

107

108

Scheme 11.28

109

OMe (−)-Lasubine II

Synthesis of (−)-lasubine II by RCM–ROM–CM.

of dimethoxystyrene and [Ru]-I (5 mol%) resulted only in traces of CM of the terminal double bond of 107 with 108. Fortunately, utilization of 5 mol% of either [Ru]-II or [Ru]-III furnished the desired E-configured product 109 in a gratifying yield of 48 and 44%, respectively. The main by-product isolated resulted, again, from the CM between 107 with 108, which demonstrates the main concurrent reaction and, therefore, the main difficulty of these types of transformations (Scheme 11.28). 11.6.2.2 Synthesis of (+)-Cylindramide A and Bicyclic Core of Geodin A Phillips et al. have also extended the domino metathesis of norbornene derivatives with a CM step and employed this strategy in the construction of the bicyclic fragment of geodin A [49] and the synthesis of (+)-cylindramide A [50]. In both cases, the bicyclic compound 110 served as the metathesis precursor, using different olefins as cross-partners (Scheme 11.29). Metathesis cascade with 111 led to the formation of 54% of product 113 (E/Z = 1.5 : 1) along with 30% of 112, indicating the partial failure of the final CM step. Interestingly, when olefin 114 was employed, none of the rearranged product lacking CM was found and the desired bicyclic structure 115 was formed in 59% yield with a selectivity of E/Z = 2 : 1. 11.6.2.3 Total Synthesis of (+)-Mycoepoxydiene Although the domino metathesis performed in the synthesis of the fungal metabolite (+)-mycoepoxydiene [51] does not really involve a ring rearrangement by definition, it demonstrates the synthetic utility of ROM–CM–RCM processes for the expansion (or contraction) of macrocyclic ring systems. This type of sequence, previously explored by the groups of Diver and Grubbs among others, was first studied on the oxabicyclic compound 116 separating the first ROM–CM from the RCM step. ROM and CM with 1,3-butadiene (4 equiv.) furnished a mixture of regioisomeric trienes 117 and 118 in 75% yield when catalyzed by of [Ru]-I (5 mol%). The ring closure of these two structures was found to be poorly effective, and the desired product 119 was obtained in only 29% yield. The one-pot procedure, however, permitted obtaining 120 in a better overall yield than in a stepwise manner with the sequential use of catalysts [Ru]-I and [Ru]-II (Scheme 11.30).

11.7 Conclusions and Outlook O

O

OTIPS H

O [Ru]-I (4 mol%) O

+

CH2Cl2, rt

OTIPS

H

OTIPS

H

339

H

H

H

111 + 112 (30%)

113 (54%)

COOMe

COOMe

Bicyclo[3.3.0]octene core of geodin A

(E /Z =1.5 : 1) O OTIPS

110 +

O

O

OTIPS H

H

HN

O

Me O

H

[Ru]-I (4 mol%) H

HO O

CH2Cl2, rt 59%

O

NH O O 115 (E /Z = 2 : 1)

114

Scheme 11.29

O

OH

O

(+)-Cylindramide A

RCM–ROM–CM cascades on bicyclic systems.

[Ru]-I (5 mol%) O

C6H6, rt 75%

+

[Ru]-II (1.2 mol%)

O

O

OTBDPS C6H6, reflux 29%

OTBDPS

117

118

OTBDPS

119

1. [Ru]-I (5 mol%) C6H6, rt 2. [Ru]-II (4 × 5 mol%) C6H6, reflux

O

OTBDPS

TBAF, THF 22%

H O

OH AcO

120

116

Scheme 11.30

O O

Mycoepoxydiene

Synthesis of mycoepoxydiene by ROM–CM–RCM.


Alkene metathesis has revolutionized modern approaches to the synthesis of a wide variety of useful organic molecules with far-reaching applications in natural product synthesis. The aforementioned examples are testimony to the utility of catalytic

O

340


olefin metathesis in domino reactions and show how these domino processes permit multi C–C bond-forming–bond-breaking reactions allowing rapid access to structures of impressive complexity which have proved otherwise very challenging. ROM–CM, RCM–CM, ene–yne RCM–RCM, and RCM–RCM cascades have been described, which have proven to be very efficient and versatile methods for the construction of carbocycles, heterocycles, and polycyclic structures. The extension of the RCM–RCM sequence by further metathesis steps has been also established in organic synthesis and has led to even more complex structures, although the catalysts with higher activities are still required. The transfer of stereoinformation in RRM processes is an attractive feature, but chiral starting materials are needed and further development of asymmetric processes is therefore essential. Given the advantages of metathesis cascades in terms of atom economy and catalytic efficiency, we can only expect the interest in these processes to grow further.

References 1 (a) Grubbs, R.H. (ed.) (2003) Handbook

of Metathesis, Vols. 1, 2, 3, Wiley-VCH Verlag GmbH, Weinheim; For reviews of the alkene-metathesis reaction, see: (b) Holub, N. and Blechert, S. (2007) Chem. Asian J., 2 (9), 1064–1082; (c) Schmidt, B. and Hermanns, J. (2004) Top. Organomet. Chem., 7, 223–267; (d) Connon, S.J. and Blechert, S. (2004) Top. Organomet. Chem., 7, 93–124; (e) Fürstner, A. (2000) Angew. Chem. Int. Ed., 39, 3012–3043; (f ) Connon, S.J. and Blechert, S. (2003) Angew. Chem. Int. Ed., 42, 1900–1923; (g) Grubbs, R.H. and Chang, S. (1998) Tetrahedron, 54, 4413–4450; (h) Schuster, M. and Blechert, S. (1997) Angew. Chem. Int. Ed., 36, 2037–2056; For its application in total synthesis, see: (i) Nicolaou, K.C., Bulger, P.G., and Sarlah, D. (2005) Angew. Chem. Int. Ed., 44, 4490–4527. 2 Tietze, L.F. (1996) Chem. Rev., 96, 115–136. 3 For examples on natural product synthesis see: (a) Moreno-Manas, M., Pleixats, R., and Santamaria, A. (2001) Synlett, 1784–1786; (b) Limanto, J. and Snapper, M.L. (2000) J. Am. Chem. Soc., 122, 8071–8072. 4 For examples see: (a) Piper, J.L. and Postema, M.H.D. (2004) J. Org. Chem., 69, 7395–7398; (b) Wallace, D.J., Bulger, P.G., Kennedy, D.J., Ashwood,

5

6

7

8

9

10

11

12

M.S., Cottrell, I.F., and Dolling, U.-H. (2001) Synlett, 357–360. For a recent review, see: Deshmukh, P.H. and Blechert, S. (2007) Dalton Trans., 24, 2479–2491. Schrock, R.R., Murdzek, J.S., Bazan, G.C., Robbins, J., DiMare, M., and O’Regan, M. (1990) J. Am. Chem. Soc., 112, 3875–3886. Nguyen, S.T., Johnson, L.K., Grubbs, R.H., and Ziller, J.W. (1992) J. Am. Chem. Soc., 114, 3974–3975. Scholl, M., Ding, S., Lee, C.W., and Grubbs, R.H. (1999) Org. Lett., 1, 953–956. (a) Garber, S.B., Kingsbury, J.S., Gray, B.L., and Hoveyda, H. (2000) J. Am. Chem. Soc., 122, 8168–8179; (b) Gessler, S. and Blechert, S. (2000) Tetrahedron Lett., 41, 9973–9976. (a) Hoye, T.R., Jeffrey, C.S., Tennakoon, M.A., Wang, J., and Zhao, H. (2004) J. Am. Chem. Soc., 126, 10210–10211; (b) Wallace, D. (2005) Angew. Chem. Int. Ed., 44, 1912–1915. (a) Kim, M. and Lee, D. (2005) J. Am. Chem. Soc., 127, 18024–18025; (b) Cho, E.J. and Lee, D. (2008) Org. Lett., 10, 257–259. For specific reviews of the enyne-metathesis reaction, see: (a) Villar, H., Fringsa, M., and Bolm, C. (2007) Chem. Soc. Rev., 36, 55–66;

References

13

14 15 16

17

18 19 20

21

22

23 24

25

26

27

28 29

(b) Diver, S.T. and Giessert, A.J. (2004) Chem. Rev., 104, 1317–1382. (a) Kummer, D., Brenneman, J., and Martin, S.F. (2005) Org. Lett., 7, 4621–4623; (b) Kummer, D.A., Brenneman, J.B., and Martin, S.F. (2006) Tetrahedron, 62, 11437–11449. Dewi, P., Randl, S., and Blechert, S. (2005) Tetrahedron Lett., 46, 577–580. Honda, T., Namiki, H., Kaneda, K., and Mizutani, H. (2004) Org. Lett., 6, 87–89. Honda, T., Namiki, H., Watanabe, M., and Mizutani, H. (2004) Tetrahedron Lett., 45, 5211–5213. Michrowska, A., Bujok, R., Harutyunyan, S., Sashuk, V., Dolgonos, G., and Grela, K. (2004) J. Am. Chem. Soc., 126, 9318–9325. Honda et al., unpublished results. Niethe, A., Fischer, D., and Blechert, S. (2008) J. Org. Chem., 73, 3088–3093. (a) Boyer, F.-D. and Hanna, I. (2002) Tetrahedron Lett., 43, 7469–7472; (b) Boyer, F.D., Hanna, I., and Ricard, L. (2004) Org. Lett., 6, 1817–1820. Aldegunde, M.J., Castedo, L., and Granja, J.R. (2008) Org. Lett., 10, 3789–3792. Fukumoto, H., Takahashi, K., Ishihara, J., and Hatakeyama, S. (2003) Tetrahedron Lett., 44, 8047–8049. Shimizu, K., Takimoto, M., and Mori, M. (2003) Org. Lett., 5, 2323–2325. Fukumoto, H., Takahashi, K., Ishihara, J., and Hatakeyama, S. (2006) Angew. Chem. Int. Ed., 45, 2731–2734. For examples see: (a) Limanto, J. and Snapper, M.L. (2000) J. Am. Chem. Soc., 122, 8071–8072; (b) Bassindale, M.J., Hamley, P., and Harrity, J.P.A. (2001) Tetrahedron Lett., 42, 9055–9057. For another application of a ROM-CM cascade see: Schrader, T.O. and Snapper, M.L. (2002) J. Am. Chem. Soc., 124, 10998–11000. Alexander, A., Statsuk, V., Liu, D., and Kozmin, S.A. (2004) J. Am. Chem. Soc., 126, 9546–9547. Stragies, R. and Blechert, S. (1999) Tetrahedron, 55, 8179–8188. Stragies, R. and Blechert, S. (2000) J. Am. Chem. Soc., 122, 9584–9591.

30 Buschmann, N., R¨ uckert, A., and

31

32 33 34 35 36

37 38 39 40

41 42 43

44 45 46

47 48 49

50 51

Blechert, S. (2002) J. Org. Chem., 67, 4325–4329. Song, L., Duesler, E.N., and Mariano, P.S. (2004) J. Org. Chem., 69, 7284–7293. Zhao, Z., Song, L., and Mariano, P.S. (2005) Tetrahedron, 61, 8888–8894. Holub, N., Neidhöfer, J., and Blechert, S. (2005) Org. Lett., 7, 1227–1229. Böhrsch, V. and Blechert, S. (2006) Chem. Commun., 1968–1970. Stragies, R. and Blechert, S. (1998) Synlett, 169–170. Wrobleski, A., Sahasrabudhe, K., and Aubé, J. (2004) J. Am. Chem. Soc., 126, 5475–5481. Chandler, C.L. and Phillips, A.J. (2005) Org. Lett., 7, 3493–3495. Henderson, J.A. and Phillips, A.J. (2008) Angew. Chem. Int. Ed., 47, 1–4. Barbe, G. and Charette, A.B. (2008) J. Am. Chem. Soc., 130, 13873–13875. Zuercher, W.J., Hashimoto, M., and Grubbs, R.H. (1996) J. Am. Chem. Soc., 118, 6634–6640. Blechert, S. and Stapper, C. (2002) Eur. J. Org. Chem., 16, 2855–2858. Stapper, C. and Blechert, S. (2002) J. Org. Chem., 67, 6456–6460. Rückert, A., Deshmukh, P.H., and Blechert, S. (2006) Tetrahedron Lett., 47, 7977–7981. Schaudt, M. and Blechert, S. (2003) J. Org. Chem., 68, 2913–2920. Burke, S.D., Quinn, K.J., and Chen, V.J. (1998) J. Org. Chem., 63, 8626–8627. Zaminer, J., Stapper, C., and Blechert, S. (2002) Tetrahedron Lett., 43, 6739–6741. Pfeiffer, M.W.B. and Phillips, A.J. (2005) J. Am. Chem. Soc., 127, 5334–5335. Zaja, M. and Blechert, S. (2004) Tetrahedron, 60, 9629–9634. Phillips, A.J., Hart, A.C., and Henderson, J.A. (2006) Tetrahedron Lett., 47, 3743–3745. Hart, A.C. and Phillips, A.J. (2006) J. Am. Chem. Soc., 128, 1094–1095. Takao, K., Yasui, H., Yamamoto, S., Sasaki, D., Kawasaki, S., Watanabe, G., and Tadano, K. (2004) J. Org. Chem., 69, 8789–8795.

341

343

12 Catalytic Enantioselective Olefin Metathesis and Natural Product Synthesis Amir H. Hoveyda, Steven J. Malcolmson, Simon J. Meek, and Adil R. Zhugralin

12.1 Introduction

In this chapter, a brief account of various applications of catalytic enantioselective olefin metathesis [1] to the synthesis of biologically active molecules is presented. The availability of chiral catalysts for olefin metathesis offers schemes for enantioselective synthesis of natural products that are more concise than those entailing the use of achiral Ru- or Mo-based complexes and enantiomerically pure substrates [2]. Enantioselective olefin metathesis plays a critical role in the development of more effective catalysts. A less appreciated role of investigations directed toward chiral catalyst development is that such efforts can lead to the identification of better catalysts. In designing a chiral catalyst, the structure – the steric and electronic characteristics – of the promoter molecule might be altered to the extent that a fundamentally new type of catalyst, one that likely has different reactivity and/or selectivity profiles, emerges.

12.2 Total Synthesis of Natural Products with Enantiomerically Pure Chiral Olefin Metathesis Catalysts Bearing a C 2 -symmetric Diolate Ligand 12.2.1 Total Synthesis of Coniine through Enantioselective RCM with Substrates Bearing a Tertiary Amine

In 2005, the first examples of enantioselective olefin metathesis involving substrates that bear Lewis basic amines were reported [3]. Catalytic activity of number of chiral Mo-based diolates (such as 2, Scheme 12.1) were probed; enantioselective ring-closing metathesis (RCM) reactions deliver various azacyclic structures in up to 99/1 enantiomeric ratio (er). It should be noted that Ru-based catalysts can be largely inhibited in the presence of the same class of amine substrates [4]. Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts. Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32440-8

12 Catalytic Enantioselective Olefin Metathesis and Natural Product Synthesis

344

Me

Me [Mo]-II (10 mol%) Me C H , 22 °C, 24 h 6 6

N Bn

83%

1

1. ClRh(PPh3)3 (10 mol%) C6H6, 400 psi H2, 48 h

H Me

N Bn

2 (er = 93.5/6.5)

2. Pd(OH)2/(C) (20%) EtOH, 400 psi H2, 48 h 72% (Overall yield)

Me N

H N H

O Mo O

Me

Ph

(R )-Coniine

[Mo]-II

Scheme 12.1

Total synthesis of (R)-coniine through enantioselective RCM.

OTBS

TBSO Me

Me

Me Me

[Mo]-III (3 mol%) pentane, 22 °C, 6 h 97%

H 4

3

(er = 93.5/6.5)

(dr > 98/2)

i -Pr Ar

i -Pr N

O

O Mo O Me

Me

OH

Me Me

OH

CH2I2, Et2Zn H Me (+)-africanol

Cl(CH2)2Cl, 0 °C 81%

Me

Ar Me 8 steps

H

Ph

[Mo]-III

Me

5

Scheme 12.2 Synthesis of Africanol through an enantioselective ring-opening/ring-closing metathesis (ROM/RCM) reaction. Ar = 2,4,6-Me3 C6 H2 .

The utility of the above-mentioned class of catalytic transformations has been demonstrated in the context of preparation of enantiomerically enriched coniine, the neurotoxin found in the hemlock plant. Thus, Mo-catalyzed RCM of benzylamine 1 in the presence of 10 mol% of chiral [Mo]-catalyst ([Mo]-II) gives rise to the formation of unsaturated piperidine 2 in 93.5/6.5 er and 83% yield (Scheme 12.1). A two-step sequence furnishes the poisonous alkaloid. 12.2.2 Enantioselective Synthesis of Africanol by a Ring-opening/Ring-closing Metathesis Reaction

As illustrated in Scheme 12.2, treatment of meso tertiary silyl ether 3, prepared diastereoselectively through alkylation of the corresponding norbornenone, with 3 mol% of chiral alkylidene [Mo]-III gives rise to 4 in high enantiomeric purity (er = 93.5/6.5) and 97% yield. A bicyclic structure containing the desired stereochemistry is thus accessed through a single, efficient catalytic process. Transformation to 5,

12.3 Enantioselective Synthesis of Quebrachamine

345

Ph O

O

[Mo]-III (5 mol%) Ph

C6H6, 50 °C, 6 h 95%

6

Me O

Ph OH

Et Tipranavir

i -Pr N

Ar

6 steps 49% (Overall yield)

Me O

i -Pr

7 (er = 95.5/4.5)

O

O Mo O Ar

O

O O S N N H

O

Ph CF3

[Mo]-III O 8

Scheme 12.3 Enantioselective synthesis of the lactone fragment of tipranavir through a ring-opening/ring-closing metathesis (ROM/RCM) reaction. Ar = 2,4,6-Me3 C6 H2 .

bearing the trisubstituted cyclic alkene and conversion of the vinyl unit to the desired methyl substituent require eight additional steps. The natural product is then accessed through a hydroxyl-directed cyclopropanation reaction [5]. 12.2.3 Enantioselective Synthesis of the Lactone Fragment of Anti-HIV Agent Tipranivir

Unsaturated pyran 6 (Scheme 12.3) has been utilized (via 8) in the synthesis of the lactone fragment of tipranavir, a molecule discovered to possess potent activity against the HIV-protease enzyme [6]. Synthesis of enantiomerically enriched 7 (er = 95.5/4.5), a chiral product that would otherwise be difficult to access in the nonracemic form, is therefore achieved through a catalytic reaction of an easily available achiral substrate. Enantioselective preparation of 7 by an RCM process that involves the use of an achiral catalyst demands an effective protocol for preparation of the requisite O-substituted quaternary carbon stereogenic center.

12.3 Enantioselective Synthesis of Quebrachamine through an Exceptionally Challenging RCM Reaction

An enantioselective synthesis of Aspidosperma alkaloid quebrachamine, summarized in Scheme 12.4, has been recently accomplished [7]. The late-stage RCM requires ring-closure onto one of two sterically hindered vinyl groups at a congested all-carbon quaternary center [8] in the presence of a Lewis basic tertiary amine. [Mo]–diolate complexes (e.g., [Mo]-II, Scheme 12.1) do not promote any reaction even under forcing conditions (up to 50 mol% catalyst, up to 80 ◦ C). When

Ph

12 Catalytic Enantioselective Olefin Metathesis and Natural Product Synthesis

346

N

N [Mo]-IV (1 mol%) (prepared in situ ) C6H6, 22 °C, 1 h

N H

N H

9

i -Pr

10 >98% conv., 84% (er = 98/2)

i -Pr N

N

N

Mo O

Cl PtO2 (5 mol%), H2 (1 atm)

Et

Ph

Cl

TBSO

[Mo]-IV

EtOH, 22 °C, 1 h 97%

N H (+)-Quebrachamine

Scheme 12.4 A stereogenic-at-Mo complex as chiral catalyst for enantioselective RCM in a stereoselective synthesis of quebrachamine.

Me

OPMB Me

Me

O

[Ru]-XV (2 mol%)

11

Styrene (8 equiv.) No solvent, −15 °C, 20 h 62%

Et

O

Me

Me

O

O Me Et

OH O

O

Me

N

MesN Na/NH3, t-BuOH Et2O, −78 °C, 3 min 70%

I Ru O Oi-Pr [Ru]-XV

Me

Ph Ph

O

(+)-Baconipyrone C

Ph

Ph

12 (er = 94/6; E/Z > 98/2)

Me

O

Me O

Ph

OPMB Me

Me

PMBO

OH

Et 13 Site-selectivity > 98/2 E/Z > 98:2; > 98% unconjugated olefin

Scheme 12.5 Total synthesis of baconipyrone C by an enantioselective Ru-catalyzed ring-opening/cross-metathesis (ROCM) reaction. PMB = p-methoxybenzyl; Mes = 2,4,6-Me3 C6 H2 .

a chiral [Ru]-catalyst (e.g., the chiral complex illustrated in Scheme 12.5) is used, rac-10 is obtained. To address the above reactivity and selectivity problems, a new class of stereogenic-at-[Mo] complexes, represented by monopyrrolide [Mo]-IV, was designed [7a]. The stereogenic-at-[Mo] complex [Mo]-IV, prepared and used in situ from the corresponding achiral Mo-bispyrrolide and a chiral aryl alcohol, readily promotes the difficult RCM. Tetracyclic 10 is obtained efficiently in 84% yield and an enantiomeric ratio of 98/2[9].

References

12.4 Synthesis of Baconipyrone C by Ru-catalyzed Enantioselective ROCM

Total synthesis of baconipyrone C, summarized in Scheme 12.5 [10], is the first and only application of Ru-catalyzed enantioselective olefin metathesis to natural product synthesis. Treatment of oxabicycle 11 with styrene and 2 mol% chiral Ru complex [Ru]-XV [11] leads to the formation of pyran 12 in 62% yield and with an enantiomeric ratio of 94 : 6. Ru-carbene [Ru]-XV is generated in situ by subjecting the corresponding Ag-based N-heterocyclic carbene (NHC) to an achiral Ru–PCy3 complex and NaI. It is also worthy of note that the diketone fragment of baconipyrone C was synthesized through a tandem double-allylic alkylation process promoted by a chiral NHC–Cu complex that is structurally related to carbene [Ru]-XV. 12.5 Conclusions and Future Outlook

Discovery and development of chiral catalysts has an important relationship with natural product synthesis; it is not only that the utility of such catalysts are best demonstrated in a total synthesis setting, but also that efforts to prepare natural products with maximum efficiency and selectivity illustrates the need for more effective catalysts. Therefore, catalyst development driven by demands presented in a total synthesis (problem in search of a solution) may prove to be a more valuable approach than the strategy of catalyst discovery followed by search for an appropriate application (solution in search of a problem). There is little doubt that more active and robust olefin metathesis catalysts need to be discovered and developed for this important class of transformations to achieve their maximum potential [12]. One avenue of research that can lead to identification of such superior catalysts likely involves natural product synthesis, since challenges faced in the context of such efforts often demand innovation in catalyst design. Acknowledgments

We are grateful to the National Science Foundation (Grants CHE–0213009 and CHE–0715138) and the National Institutes of Health (Grant GM-59426) for generous support of our efforts toward the development of effective chiral olefin metathesis catalysts and applications of the derived methods to target-oriented synthesis.

References 1 Schrock, R.R. and Hoveyda, A.H. (2003)

Angew. Chem. Int. Ed., 42, 4592–4633. 2 For reviews on applications of catalytic olefin metathesis reactions (largely promoted by achiral catalysts) to natural

product synthesis, see: (a) Deiters, A. and Martin, S.F. (2004) Chem. Rev., 104, 2199–2238; (b) Nicolaou, K.C., Bulger, P.G., and Sarlah, D. (2005) Angew. Chem. Int. Ed., 44, 4490–4527.

347

348

12 Catalytic Enantioselective Olefin Metathesis and Natural Product Synthesis 3 Sattely, E.S., Cortez, G.A., Moebius,

4

5 6

7

Hoveyda, A.H. (2009) J. Am. Chem. Soc., D.C., Schrock, R.R., and Hoveyda, 131, 943–953. 8 (a) Christophers., J. and Baro, A. (eds) A.H. (2005) J. Am. Chem. Soc., 127, (2006) Quaternary Stereocenters: Chal8526–8533. For representative examples, see: (a) Lee, lenges and Solutions for Organic Synthesis, K.L., Goh, J.B., and Martin, S.F. (2001) Wiley-VCH Verlag GmbH, Weinheim; Tetrahedron Lett., 42, 1635–1638; (b) (b) Denissova, I. and Barriault, L. (2003) Wipf, P., Rector, S.R., and Takahashi, Tetrahedron, 59, 10105–10146. 9 Formation of an all-carbon quaternary H. (2002) J. Am. Chem. Soc., 124, stereogenic center through enantioselec14848–14849; (c) Wipf, P. and Spencer, tive olefin metathesis is relatively rare. S.R. (2005) J. Am. Chem. Soc., 127, For the first example of such a transfor225–235; (d) Cortez, G.A., Schrock, mation, see: Lee, A.-L., Malcolmson, S.J., R.R., and Hoveyda, A.H. (2007) Angew. Puglisi, A., Schrock, R.R., and Hoveyda, Chem. Int. Ed., 46, 4534–4538. Hoveyda, A.H., Evans, D.A., and Fu, A.H. (2006) J. Am. Chem. Soc., 128, G.C. (1993) Chem. Rev., 93, 1307–1370. 5153–5157. Cefalo, D.R., Kiely, A.F., Wuchrer, 10 Gillingham, D.G. and Hoveyda, A.H. M., Jamieson, J.Y., Schrock, R.R., and (2007) Angew. Chem. Int. Ed., 46, Hoveyda, A.H. (2001) J. Am. Chem. Soc., 3860–3864. 11 Van Veldhuizen, J.J., Campbell, J.E., 123, 3139–3140. (a) Malcolmson, S.J., Meek, S.J., Giudici, R.E., and Hoveyda, A.H. (2005) Sattely, E.S., Schrock, R.R., and J. Am. Chem. Soc., 127, 6877–6882. 12 Zhugralin, A.R. and Hoveyda, A.H. Hoveyda, A.H. (2008) Nature, 456, (2007) Nature, 450, 243–251. 933–937; (b) Sattely, E.S., Meek, S.J., Malcolmson, S.J., Schrock, R.R., and

349

13 Metathesis Reactions in Solid-phase Organic Synthesis Sofia Barluenga, Pierre-Yves Dakas, Rajamalleswaramma Jogireddy, Gaële Valot, and Nicolas Winssinger

13.1 Introduction

As is evident from the different chapters of this book, the mildness and versatility of the metathesis reaction [1–3] have had a tremendous impact on organic chemistry, which enabled it to be rapidly embraced by the community of synthetic chemists as it offered a new manifold of disconnections in retrosynthetic analysis [4] and new opportunities in diversity-oriented synthesis [5]. The discovery of the first efficient and tolerant soluble catalyst was reported at a time of high activity in the field of combinatorial chemistry. The potential of this enabling reaction was therefore rapidly evaluated in the context of solid-phase organic synthesis (see introduction for the most relevant catalysts). Importantly, it was swiftly established that the most popular support for solid-phase organic synthesis (polystyrene-based resins) is compatible with the ruthenium-based metathesis catalysts, and the past 10 years have shown that conditions developed for solution-phase chemistry generally translate well to solid-phase synthesis. An important consideration for carrying out a ring-closing metathesis (RCM) is the rate of cyclization versus oligomerization which will dictate the concentration at which the reaction should be performed to maximize the desired outcome. Analogously, the outcome of metathesis reactions on polymer-bound alkenes will depend on the density of the alkenes (loading of the resin) and mobility or conformational flexibility of the alkenes (cross-linking of the resin and nature of the linker). From the first reports in 1996/1997 to the early 2000s, all reactions on solid phase were carried out with the first-generation ruthenium catalyst [Ru]-II bearing phosphine ligands. Following the commercial availability of second-generation catalysts with a carbene ligand, its superior reactivity was frequently favored. As for solution-phase chemistry, subtle differences in the outcome of the reaction were sometimes observed depending on the choice of catalyst ([Ru]-II versus [Ru]-III). While both catalysts lead to the same intermediate following the first step of the catalytic cycle, differences in the rate and reversibility of this first step have been found to lead to significantly improved performances in some circumstances. This is particularly true for cross-metathesis Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts. Edited by Janine Cossy, Stellios Arseniyadis, and Christophe Meyer Copyright  2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-32440-8

350

13 Metathesis Reactions in Solid-phase Organic Synthesis

(CM) reaction with α,β-conjugated alkenes for which [Ru]-III often performs best. A welcomed addition to this family is the development of a fluorous-tagged catalyst that dramatically facilitates catalyst isolation, a critical point if the metathesis is used in the cleavage of high-throughput synthesis programs. This chapter is divided into six subsections which cover the most prominent applications of the metathesis reaction in solid-phase organic chemistry: RCM for cleavages from the resin; synthesis of small rings to macrocycles on the resin via RCM; the use of the metathesis reaction for dimerization of polymer-bound small molecules; RCM to constrict peptide conformations (while this is formally an RCM, it is treated separately because of the unique considerations of peptide chemistry); CM between a polymer-bound alkene and an alkene in solution; and ene–yne metathesis. As the field evolved relatively rapidly with major advances in the catalysts, each subsection is organized in chronological order for the sake of clarity. 13.2 Metathesis-based Cyclorelease Reaction

Following the first publications reporting the broad functional group compatibility of the new ruthenium-based metathesis catalyst, it became apparent that an alkene functionality could serve as a linker to immobilize the small molecules of interest. The first applications of this principle exploited the metathesis reaction to perform an RCM with concomitant cleavage of the substrate from the resin, the so-called cyclorelease reaction, to afford an endocyclic alkene product (Figure 13.1, Equation 1). Conversely, if a terminal olefin is sought rather than an endocyclic alkene, the linker could be designed to release the terminal alkene while leaving the ring-closing product polymer bound (Figure 13.1, Equation 2. An important consideration in the former case is that the catalyst remains bound to the resin and an alkene must be within reacting distance for the catalyst to turn over. The efficiency of the turnover is strongly dependent on the substrate, the type of resin, and the conditions used. Nonetheless, it has been shown that external alkenes can be used to promote efficient catalyst turnover in some cases [6, 7]. As the reaction is reversible and under thermodynamic control, whether the external alkene will also compete for CM and ring-opening metathesis (ROM) depends on the strain of the cyclized product. In the second case, release of a terminal alkene upon RCM on the polymer, the catalyst remains associated with the alkene in solution and “Ru”

Ru

+ R

R R

Figure 13.1

“Ru”

+ Ru

R

Metathesis-based cyclorelease reaction.

Further catalytic cycles

Further catalytic cycles

R Endocyclic product R Terminal alkene

13.2 Metathesis-based Cyclorelease Reaction

351

can readily turn over. An important consideration in these reactions is that the product will inevitably be contaminated with the catalyst, which will require some form of purification prior to biological screenings. The more recent development of practicable methods to remove the catalyst such as oxidation with dimethyl sulfoxide (DMSO) and Ph3 PO [8], which converts the catalyst to easily removable polar by-products, are important advances. Alternatively, the light fluorous-tagged Grubbs–Hoveyda catalyst reported by Curran (f-[Ru]-III) [9] which can be removed by fluorous solid-phase extraction offers another solution which is compatible with automation. Pioneering examples of the cyclorelease cleavage to obtain biologically relevant products include the preparation of tetrapeptide macrocycles reported by Blechert et al. [10], the synthesis of epothilones by Nicolaou et al. [11], and lactams [12] as well as β-turn mimetics by Piscopio et al. [13] (Figure 13.2). In the case of the OAc AcHN MeO

O

O

n

N H

O

OAc OAc

O

AcHN MeO

O

[Ru]-I (15 mol%)

OAc O HN

O N H

CH2Cl2, 40 °C, 2 h 30%

O

O

OAc

O

O OAc

O HN

N

O

N

FmocHN

FmocHN 85% purity

Trityl resin, 0.5 mmol/g Blechert (1997)

Wittig

O

O

S

H O

N

Aldol

OH

S S

HO

N

Esterification

OH

[Ru]-I (3 × 10 mol%) HO CH2Cl2, 48 h 52%

O

N O O OTBS O

O OTBS O

O OTBS O

>85% purity

Nicolaou (1997) NHBoc O OH

O

N

[RU]-I (5 mol%)

NHBoc O N

ClCH2CH2Cl 80 °C, 16 h

OMe

Ph

O O Ph

Piscopio (1997) O

H

O

t-BuO

N

O

C

Ph NH2

H N HO

O

O NHBoc

HN

NHBoc O

4 Component Ugi reaction

O N

Ph

O N H

Ot-Bu O

Cinnamyl amine resin 0.5–0.7 mmol/g

Piscopio (1999)

Figure 13.2

Pioneering examples of the RCM cyclorelease.

[Ru]-I (5 mol%) ClCH2CH2Cl 80°C, 24 h 55%

NHBoc

HN O N

Ph

O N H

Ot-Bu O


352

epothilones, the chemistry was used to make over 100 analogs, shedding important light on the structure activity of this pharmacophore [14]. More recently, this strategy was used by Overkleeft et al. to prepare a pilot library of nine pyranofurans from mannitol using [Ru]-II for the cyclorelease (Figure 13.3). The reaction was found to be extremely efficient, affording over 90% yield in several examples with just 5 mol% catalyst using a moderately high loading resin (0.55 mmol g−1 ) [15]. The effect of an external alkene in the cyclization to facilitate catalyst turnover was revisited by Brown et al. in the context of the cyclorelease of seven-membered sulfonamide on a fairly high loading resin (0.7–1.2 mmol g−1 ). The cyclorelease proceeded with 30% yield using as little as 1 mol% catalyst and afforded nearly quantitative yield with 5% catalyst; however, the addition of octene did not improve the outcome [16]. A cyclorelease strategy was also used recently in a medicinal chemistry effort following the identification of a potent modulator of the human Motilin receptor (hMOT-R) from a high-throughput program [17]. To this end, Marsault et al. prepared a focused library of macrocycles containing H N

O

HO HO HO

O R OH OH OH

NH

2

O

9

O

[Ru]-II (5 mol%)

O O HN 1 R

D-(+)-Mannitol

R

NH

2

CH2Cl2, 40 °C 16 h 74−99%

O

O

H

O H O O HN 1 R

Rink amide resin 0.55 mmol/g

Overkleeft (2003)

S N R

O

O O

O

[Ru]-I (2.5−5 mol%)

O R S N

CH2Cl2, 40 °C 3−15 h 45−100%

2-Carboxyethyl polystyrene 0.76−1.24 mmol/g or Merrifield resin Brown (2004) O O

R

H N

HN R

3

2

N H

O

Trityl resin

O

O R1

HN

O

HN R

Bts

R

H N 3

2

R

O

O

N H

N X

Bts = benzothiazole-2-sulfonyl

2

O R

O

1

R

[Ru]-II

R1 RCM Bts HN Cyclative release

3

NH HN O NH X

Marsault (2006) X

Enyne RCM Y n F17C8

N Ts

n

Y

n

CH2Cl2, 40 °C

O

F17C8

f- [Ru]-III n

X RCM X = O or N Y = NNs or Oi -Pr2SiO n = 1, 2 or 3

RCM/ROM RCM

Nelson (2009)

Figure 13.3

Selected examples of RCM cyclorelease.

Y

n Y

n

Ts N

O

+ F17C8

[Ru] O

13.2 Metathesis-based Cyclorelease Reaction

variations of the tripeptide motif and the alkene linkage. Over 30 analogs were prepared on a fairly high loading resin (1.2 mmol g−1 ) to afford the macrocycles which were individually purified. Nelson et al. reported a very elegant extension of the metathesis-based cyclorelease reaction leveraged on combinations of RCM and ROM in conjunction with ene–yne metathesis to afford over 80 different polycyclic scaffolds [18]. Rather than using a polymer matrix to immobilize the substrates, Nelson opted for the light fluorous tag technology developed by Curran et al. [19–21], wherein molecules covalently linked to a perfluorinated chain of three to nine carbons can be adsorbed on fluorinated silica gel in solid-phase extractions (F-SPE). Though the metathesis was performed in solution phase, this elegant example has been included, as the logic of linker connectivity and the underlying motivation (ease of automation/purification) are the same as for solid-phase chemistry. The final metathesis cascade was performed with a fluorous-tagged metathesis catalyst [9] (Figure 13.1) such that the final F-SPE not only removed the linker by-product but also the catalyst. This work represents a significant landmark in the degree of skeletal diversity achieved by a sequence involving only three bond-building steps prior to the metathesis cascade and attests to the power of the olefin metathesis to generate molecular diversity in so-called diversity-oriented synthesis. Regarding the second mode of RCM-based cleavage affording terminal alkenes (Figure 13.1, Equation 2), Schmidt and Knerr investigated the release of polymer-bound oligosaccharides by RCM to afford O-allyl glycosides which represented versatile protected intermediates for further manipulation (Figure 13.4) [22, 23]. An alkene linker is particularly attractive for oligosaccharide synthesis, as it is orthogonal to the acidic nature of most glycosylation methods and ester type protecting groups which are required to control β-glycosylations. Alternatively, Seeberger has used a CM to release O-alkenyl glycosides (Figure 13.11, Section 13.6). While a direct comparison of both systems has not been reported, it can be anticipated that the linker reported by Schmidt and Knerr would cleave faster on the basis of the fact that the kinetics of the metathesis reaction is strongly influenced by the level of substitution of the alkene and that RCM for five-membered rings are extremely fast. In 2002, Waldmann et al. published the solid-phase synthesis of a library of dysidiolide analogs (a protein phosphatase inhibitor) leveraging a similar cleavage strategy that yielded compounds with improved biological activity [24, 25]. In this case, the final step of the synthesis is a cyclorelease reaction yielding a highly substituted alkene. It is interesting to note that the use of cyclorelease cleavage kinetically relayed the catalyst to only one of two trisubstituted alkenes. In this case, the use of [Ru]-II did not improve the yield of the reaction relative to [Ru]-I. The synthesis of the natural product was achieved in 11 steps in an overall yield of 14% representing an average yield of 84% for each step. The analogs were synthesized in a total of 8–11 steps on solid phase with an overall yield of 6–27% based on a Merrifield resin with a loading of 1.1 mmol g−1 . The breadth of reaction employed in the synthesis (organometallic additions, Wittig and Diels–Alder reactions) testifies to the versatility and robustness of the alkene linker. This concept was further extended by van der Marel et al. to a

353

354

13 Metathesis Reactions in Solid-phase Organic Synthesis OBn

OBn O O BnO OBn OBn

O

BnO

OBn

O [Ru]-I (2 × 9 mol%)

OBn

Merrifield resin 0.2 mmol/g

CH2Cl2, 23 °C, 18 h 90%

O

OBn BnO O O BnO OBn OBn

O

O O

+

OBn

Schmidt (1999)

Wittig O

[Ru]-I (2 × 10 mol%) Diels−Alder

CH2Cl2, 23 °C, 16 h 14% (11 steps)

H

HO

OH Furan addition

O

O

+ H

HO

OH

O

O Merrifield resin 1.1 mmol/g

Dysidiolide

O

Waldmann (2002)

O O

[Ru]-II (10 mol%) CH2Cl2, 40 °C, 12 h 67%

+

+

[Ru]

CH2

O Glycoside

O Glycoside Merrifield resin, 1% DVB 0.38 mmol/g Van der Marel (2004)

Figure 13.4

Selected examples of cyclorelease.

tandem RCM reaction in the context of a linker for carbohydrate synthesis. In the cleavage modes previously described (Figure 13.1, equation 2), the catalyst is regenerated after a catalytic cycle in the form of a polymer-bound carbene or a carbene linked to the product of interest. This latter strategy bypasses this issue, as the tandem metathesis produces a very reactive methylene carbene. Indeed, good yields (>68%) were achieved in just 2 hours for the cleavage of a disaccharide [26].

13.3 Ring-closing Metathesis (RCM)

While the first examples of RCM on solid phase exploited the opportunity to simultaneously release the substrate from the resin, RCM without concomitant cleavage has been even more extensively used in solid-phase synthesis and has the advantage that the catalyst is recovered in the solution at the end of the catalytic cycle and thus unhindered to turn over. However, the proportion of cyclization


355

versus CM with another polymer-bound alkene will still depend on the nature of the reaction and the loading of the resin. Early work on solid-phase organic synthesis had already investigated the possibility of using polymeric support to ‘‘isolate’’ reactive intermediates and create a quasi-infinite dilution environment. It was concluded that even at low loading (>0.1 mmol g−1 ), interaction between polymer-bound intermediates is still possible [27]. Nevertheless, the likelihood and frequency of such intraresin interactions will decrease at lower loading and is dependent not only on the loading but also on the level of cross-linking, the solvent, and the swelling properties of the polymer as well as the linker. The number of parameters and the heterogeneous nature of all polymeric supports make a general conclusion difficult. As will be discussed in the next section, the intraresin interaction can be favored with highly loaded resins (typically higher than 1 mmol g−1 ) if there is no competition for thermodynamically more favorable RCM. One of the first examples of RCM in solid-phase organic synthesis is the modular assembly of functionalized azacycles described by Blechert et al. (Figure 13.5) [28]. 2

R N TrO

R1

n

Trityl resin 0.5 mmol/g

Blechert (1998)

1

2

[Ru]-I (8

−15

R

R mol%)

N

CH2Cl2, 40 °C 12−15 h 85–89%

TrO

n

O

O N H

N H

OMe O

O

O N

Ph

MeO O 200 µm Tentagel

O

OMe O

O

[Ru]-I (15 mol%) CH2Cl2, 40 °C

Ph

N MeO

16 h 77%

O

Schreiber (1999) Ns O

N

R2

Ns N

[Ru]-I (2 × 5 mol%)

O

23 °C, 1 h

2

O

O AP1867

O

CH2Cl2

R1 Cl

R

O R1

N H

O HO R1

Cl

Chloro trityl resin 0.16 mmol/g

Verdine (2001)

N O

1. [Ru]-II (20 mol%) CH2Cl2, 40 °C 48 h

O H N O

2. HF–py, 2 h O

N

O Br

Pri

Si

i Pr

Silyl linker based polystyrene beads (500–560 µm)

Schreiber (2000)

Figure 13.5

RCM of polymer-bound dienes or polyenes.

OH O N

H

N O

O H

H H N

O Br

N

R

2

356


The approach was based on a RCM to form heterocycles of different ring sizes. On the basis of the recognition that many bioactive secondary metabolites have macrocyclic structure, Schreiber et al. evaluated the possibility of using a RCM to access diverse macrocycles. As only [Ru]-I was available at the time, the study was performed with the first-generation catalyst and as a proof of concept. Though only one example of a macrocycle was reported on solid phase, it suggests that this approach should be compatible with split-pool synthesis of larger libraries [29]. Indeed, a small molecule emanating from this chemistry was recently found to bind to the Sonic hedgehog protein and inhibit its signaling pathway [30]. Beyond these early proofs of principle regarding the efficacy of the RCM on solid phase, Verdine’s lab developed the solid-phase synthesis of tetrahydrooxazepines in which the critical step of the synthetic scheme was a RCM to form the cis-endocyclic olefin (Figure 13.5). This was also the first example of a RCM being performed on a molecule containing an N–O bond. The resin of choice was a polystyrene-based trityl resin. The metathesis reaction was performed at room temperature for 1 hour using [Ru]-I as the catalyst. For all the substrates examined, the metathesis reaction was found to produce a single compound with complete conversion [31], thereby enabling the preparation of a library of 320 compounds. In a very elegant example that stands as an early milestone in the development of diversity-oriented synthesis, Schreiber et al. reported a synthetic pathway leverage on one of the most powerful reactions in combinatorial chemistry, the Ugi multicomponent reaction, which delivers a scaffold bearing two terminal alkenes and a strained internal alkene. Exposure of this polymer-bound intermediate to [Ru]-II (20 mol%, CH2 Cl2 , 40 ◦ C, 60 hours) afforded the thermodynamically more favorable rearranged product having a seven-five-five-seven fused polycyclic structure (Figure 13.5) [32]. On the basis of their investment in HF-mediated cleavages for library preparation, the work was performed using a silyl linker. Arya et al. have made extensive use of RCM for the synthesis of natural product-like polycyclic compounds on solid phase. They have prepared a wide variety of tetrahydroquinoline derivatives and polycyclic compounds including 6-, 8-, and 10-membered rings as shown in Figure 13.6. The phenolic hydroxyl group was utilized as an anchorage point to a Wang resin which can be conveniently cleaved with triflouroacetic acid (TFA) [33–35]. An extension of this work was published more recently in which an aminoindoline-based library was prepared. In this case 6-, 7- and 10-membered rings were generated via RCM using a silyl-based resin with macrobeads [36, 37]. Inspired by the plethora of biological activities found in polyketides bearing recurring 1,3-diols with α,β-unsaturated lactones, Waldmann et al. reported the synthesis of natural product-derived libraries via iterative stereoselective allylations and ozonolyses. Esterification of the final allylation product and RCM afforded α,β-unsaturated δ-lactones with multiple oxygenated side chains (Figure 13.6). The stereoselective synthesis on a Wang resin was achieved in a 40–60% overall yield in 10 steps and with a 63–80% diastereoselectivity. The metathesis reaction was carried out using [Ru]-III as the catalyst of choice. After the RCM, the resin turned brown as a result of the absorption of the metal; however, after cleavage the ruthenium content of the product was still significant and a


O

O

O OAc [Ru]-II (60 mol%)

N H O

O

OAc

N

5% TFA

O

H HO

O

O

OAc

N

H

CH2Cl2, 40 °C, 14 h

357

O O

O

Wang resin 1.4 mmol/g O 1. [Ru]-I (2 × 6 mol%) 2. Conjugate addition

N

O 3. 5% TFA

O

SPh

O N

OH

H

Arya (2004) O

2. HF–Py O

H OAc NH PhCO

PhCO

Silyl linker based microbeads (500–560 µm), 1–1.3 mmol/g

O

Ph

O

1

COR OAc N

N N

OH O

OH O

N

CO2Et

CO2Et

OH O

O

O O NHCOPh

2

R

N OH O

H OAc NH

O

O

O

O

1. [Ru]-II (40–50 mol%) CH2Cl2, 40 °C, 24 h

N

OTBS

N

2. 5% TFA

O

O Wang resin 1.7 mmol/g

O

OH

Wang resin 1.19 mmol/g

1. [Ru]-II, CH2Cl2 40 °C, 6 h

OTBS N

HO

O

O

O

NH O

O

O

Arya (2006) O

O O Silyl linker 0.6 mmol/g

1. [Ru]-II (40–50 mol%) CH2Cl2, 40 °C, 24 h

OR′ O

R

OH OR′ O R

2. HF–Py, 40 °C, 18 h

n

n

Waldmann (2007) HO

EOMO

[Ru]-II (3 × 6 mol%) CH2Cl2, 120 °C 3 × 25 min

O O

micro waves

EOMO

EOMO

EOMO

R

O

1. H2O2, HFIP 2. PS-SO3H

O

O

HO R

O HO

S R 1. Bu3SnH, AIBN 2. PS-SO3H, MeOH

S

O O

HO

Thiophenol resin 0.39 mmol/g

R Aigialomycin D analogs

Winssinger (2006) EOMO

2

O

R O

EOMO O

O

R1 N

2. HFIP

HO

Trityl linker 0.6 mmol/g Winssinger (2008)

Selected examples of RCM.

2

O

R O

HO O

R O

OH

O

Figure 13.6

1. [Ru]-II (3 × 6 mol%) CH2Cl2, 120 °C 3 × 45 min micro waves

1

N

358


purification step by high-performance liquid chromatography (HPLC) was required [38]. Our group has also relied on the use of RCM for the solid-phase synthesis of natural products and libraries thereof. We have been attracted to the resorcylic acid lactones (RALs) on the basis of the observation that the small library available from natural sources contains several potent kinase and ATPase inhibitors [39]. For the solid-phase synthesis of aigialomycin D and analogs, a RCM with [Ru]-II provided the 14-membered macrocycle in excellent yield. However, we had noted that at lower temperatures and with shorter reaction times, mixtures of (E)- and (Z)-isomers were obtained [40]. To drive the reaction to the desired (E)-isomer, the reaction was carried out at elevated temperatures (120 ◦ C), which was most conveniently achieved under microwave irradiation. As the catalyst is unstable at that temperature, it was added in three portions over 75 minutes. It is noteworthy that the presence of the thioether linker did not impede the reaction. More recently, RCM was used to refine the biological activity of Hsp90 (heat shock protein 90) inhibitors through the synthesis of a focused pochoxime library. The reaction conditions were the same as for aigialomycins and also proceeded in excellent yields (HPLC analysis of crude cleavage product showed a complete and clean conversion) [41].

13.4 Intraresin Dimerization

As previously discussed, the fact that polymer-bound intermediates can react with each other had been observed since the first investigations into solid-phase organic synthesis in the early 1970s [27]. Motivated by the knowledge that many signal transduction pathways are triggered by receptor dimerization, Schreiber et al. reasoned that the metathesis could be harnessed to dimerize libraries of polymer-bound small molecules [42]. Indeed, using a high-capacity (1–2 mmol g−1 ), lightly cross-linked (1% DVB) polystyrene with a silyl linker, they achieved over 90% yield in a number of examples suggesting that this methodology could be general to obtain libraries of monomers and dimers from the same synthetic pathways. As anticipated, the metathesis afforded a mixture of (E) and (Z) isomers. This methodology was used by Liao et al. for the synthesis of benzofuran-based dimeric molecules synthesized by a Sonogashira reaction followed by a palladium-carbonylative annulations. Using the same resin and conditions as previously reported by Schreiber, the dimers were obtained in 70–80% yield as E/Z mixtures [43]. Dervan and coworkers also ingeniously capitalized on the ability to dimerize polymer-bound intermediates to generate a library of H-pin polyamides via on-resin cross-linking (Figure 13.7), as the dimeric nature of these polyamides is important for their DNA binding properties [44]. The reaction was achieved on highly loaded PAM resin (1.2 mmol g−1 ) using [Ru]-I. To avoid ambiguities steaming from alkene isomers, the double bond was hydrogenated post cleavage. DNA binding affinities were analyzed for

13.5 Restricting Peptide Conformation through Cyclization

Cleavage

Dimerization

O

O FmocHN

FmocHN

n

O

n

O

n

O

[Ru]-I (2 × 10 mol%)

O

O

n

O O

O

O NHFmoc

359

O

CH2Cl2 (0.03 M) 40 °C, 24 h

NHFmoc

Silyl linker, 1–2 mmol/g, 1% DVB

Schreiber (2001) O

O

1. [Ru]-I (10 mol%) CH2Cl2, 40 °C, 24 h

O

O

2. HF-Py, THF, 1 h 3. TMSOMe, 0.5 h

O

Silyl linker 1–2 mmol/g, 1% DVB

O

O

R

R O

O

HO

OH

O

R

O

O O

Liao (2003)

O

O O

HN N HN

O

n

O

HN

O

N O

N NH

N

O

NH

HN

N

CH2Cl2, 40 °C 48 h

O

HN

O

HN

O

O PAM resin

O

O

NH

n N

O

[Ru]-I (2 × 10 mol%)

N

NH

O O

HN

N O

N

O NH

O

NH

N O

N

N

NH

HN

NH

O

O

O PAM resin

0.88 mmol/g, 1% DVB

Dervan (2003)

Figure 13.7

Selected examples of intraresin dimerization.

each member of the library, thereby enabling a rapid optimization of the optimal linker length.


While proteinic ligands are typically endowed with high affinity and specificity for their target, attempts to reconstitute these activities with the peptidic fragment

360


involved in the interaction are usually disappointing because of the fact that short peptides do not form the stable secondary structures which are involved in the interactions. To this end, a number of efforts have been reported to stabilize secondary structures and in particular α-helical structures. One of the most used technologies is the formation of disulfide bonds that are readily formed between natural cysteine residues but which suffer from being not pharmacologically stable. Grubbs et al. were the first to explore the potential of the metathesis reaction to form a stable covalent bond between residues on an adjacent helical turn of an α-helix. One helical turn being on average 3.6 amino acid residues, a serine or a homoserine bearing an allyl group was introduced in a heptameric peptide selected for its capacity to form a helix and therefore position the terminal residues within reaction distance from the second residue and the sixth residue (i + 4). Treatment of this peptide with [Ru]-I afforded the cross-linked peptide in high yield. While it was obtained as a mixture of olefin geometry (about E/Z = 5 : 1), a single compound was readily obtained after hydrogenation of the alkene. Circular dichroism spectra showed that the cross-linked peptides were indeed helical and demonstrated that the metathesis reaction with tolerant catalysts was uniquely suited to rigidify peptide motifs [45]. This first example carried out in solution phase after cleavage of the peptide from the resin inspired a number of other research groups to harness the robustness of this chemistry with a variety of unnatural amino acids bearing terminal alkenes. As the side chain of the residue in a helix points in a parallel direction rather than toward each other, Verdine et al. investigated the cross-linking of helices using residues having opposite stereochemistry. To neutralize the detrimental impact on helical formation of a d-amino acid residue, quaternary amino acids were used (Figure 13.8). The metathesis proceeded in high yield in just 2 hours using a 10 mM concentration of catalyst. As expected, high yields were obtained only for alkenes with sufficient distance to reach over and react with the second alkene. For the ‘‘stapling’’ of a single turn using homochiral amino acids (pointing in a parallel direction), 8 atoms were sufficient while 11 were required to ‘‘staple’’ two helical turns bearing opposite stereochemistry. It is noteworthy that the stapling of a double turn with an 11-carbon spacer represents a 33-membered ring. The ‘‘stapled’’ peptides were shown to have enhanced helical conformation and were over 40-fold more resistant to proteolytic degradation relatively to their unstapled counterpart [46]. This methodology has been elegantly applied to two therapeutically relevant problems. In the first one, a 23-mer peptide with a single helical turn staple was used to agonize the BH3 domain, which is an important regulator to apoptosis and is generally disabled in cancer cell lines. Importantly, the stable peptide had improved pharmacological properties (nearly 10-fold extension in half-life in serum) and induced apoptosis in a series of cancer lines. This peptide was shown to be effective in prolonging the life span of leukemic mice [47]. In the second application, a 12-mer peptide containing a double helical turn staple was used to agonize p53, a transcription factor playing an important role in protecting cells from malignant transformations. The designed peptide was shown to agonize the p53 pathway in cells [48]. The success of these examples suggests that this stapling


361

methodology by RCM may enable the development of a wide array of peptidic compounds to probe protein interactions and has a good therapeutic potential. In an alternative approach to helical stabilization by side chain stapling, Arora et al. have shown that artificial α-helical turns with a short peptide can be obtained by mimicking the key C=O:H–N hydrogen bond with a carbon–carbon bond accessed by metathesis, thereby affording conformationally stable structures at physiological conditions (Figure 13.8) [49]. They later showed that microwave irradiation provided significant improvements in the efficiency of this transformation [50]. As a general principle, the biological activity of linear peptides can be enhanced by cyclization if such cyclized products favor the bioactive conformation. To this end, Kessler and Schmiedeberg investigated the efficacy of a metathesis on peptides containing two allyl glycine residues or homoallyglycine residues at various distances and

HN HN

O (S )

n

(S )

O n

NH

NH HN O

(S )

n

HN

(R )

O n

NH

NH [Ru]-I ClCH2CH2Cl

High yield for 8 carbon linker using homochiral crosslinking amino acids

[Ru]-I ClCH2CH2Cl High yield for >11 carbon linker using heterochiral crosslinking amino acids Verdine (2000) O R O N H

R HN

O N H

O

R N

R

O

O N H

O N

R HN O

O

N H

R

R HN

[Ru]-II or [Ru]-III (15–25 mol%)

O

R

N H

N

R

O HN

Cleavage O

HN

R

a-helix

Rink amide resin or preloaded Wang resin

R O R

HBS a-helix

bis-olefin

Arora (2004) O

O

N Peptide O 2,3

[Ru]-I Peptide 2,3

CH2Cl2, 40 °C 12 h

N Peptide O 2,3

Peptide 2,3

18–24 membered rings

Kessler (2002)

Figure 13.8

Restricting peptide conformation through cyclization.

362


noted that the cyclization efficiency in longer peptides was significantly reduced [51]. However, this could be overcome if a pseudoproline (acetonide-protected serine) was incorporated in the sequence, thereby providing the necessary level of preorganization for efficient cyclization. Nature has relied extensively on disulfide bridges between cysteine residues to restrict peptide conformation, and a number of naturally occurring biologically active peptides contain this motif. For the aforementioned reasons, an alkene surrogate is an attractive alternative. Vederas et al. have relied on this approach to prepare analogs of the antibiotic leucosin A, which has nanomolar activity against pathogens such as Listeria monocytogenes (Figure 13.9) [52]. In this case, treatment of the polymer-bound fully protected peptide (37-mer) bearing two allyl glycine residues did not afford any cyclization. It was found that precomplexation of the peptide by brief treatment with LiCl in dimethylformamide (DMF) prior to the metathesis afforded the desired cyclization product. Interestingly, cyclization of shorter peptides did not require similar pretreatment [53]. The corresponding carbocyclic analog still displayed nanomolar activity but was about 10-fold less potent than the parent compound. They applied the same methodology to the synthesis of a carbocyclic antibiotic analog bearing three macrocycles, using sequential on-resin RCM [54]. It is interesting to note that the second and third RCM did not result in any scrambling in the precedent ring-closure products, attesting to the lower reactivity of endocyclic alkenes and the

1

R Ser

O 1. 0.8 M LiCl/DMF 5 min

H N Gly Ser Lys Thr

1

R Ser

O H N Gly Ser Lys Thr

2. [Ru]-II (20 – 70 mol%) ClCH2CH2Cl, reflux 36 h

His NH O 2 Novasyn TGA resin R low loading

His NH 2 R

O

Vederas (2006)

O

NH O

O

H N

HN

NHFmoc [Ru]-II (20 mol%) O CH2Cl2, 40 °C 12 h

D Leu Ala

Ile

D Ala

Pro Dhb Ala

NH

Pro

Gly

Gly

Thr

Thr

PmcHN

Arg

Pro Thr

Lys

Asn

[Ru]-II (20 mol%) NH-D-Gly-Phe-D-Gly-Tyr-Fmoc CH2Cl2, 40 °C 48 h

Rink amide resin

Schiller (2007)

Figure 13.9

N H

O

Ala

Gly

Gly

Vederas (2007) NH-D-Gly-Phe-D-Gly-Tyr-Fmoc

Dhb

Tyr Ile

HN

O Ala

Ala

O Wang resin 0.7 mol/g

Ile

Restricting peptide conformation through cyclization.

Gly

YY

CO2H

13.6 Cross-metathesis on Solid Phase

thermodynamic stability of the first and second macrocycles. The replacement of the sulfide bridges with other groups not only could shed light on structure–activity relationships but also generate variants that are resistant to aerobic oxidation and have better metabolic stability. Although the preliminary biological evaluation of this larger ring analog showed no antimicrobial activity at comparable concentrations to the natural lactin A2, the methodology described has found wide application for the preparation of rigidified peptidomimetics containing multiple carbocyclic rings. In fact, shortly after, Schiller et al. synthesized the dicarba analogs of the cyclic opioid peptide H-Tyr-c[d-Cys-Gly-Phe-d(or l)-Cys]NH2 by RCM on Rink resin using [Ru]-II (20 mol%, reflux 48 hours) [55]. Among the dicarba analogs, the trans isomer of H-Tyr-c[d-Allylgly-Gly-Phe-l-Allylgly]NH2 proved to be an excellent µ and δ agonist with subnanomolar potency for both receptors. While the kinetic product of a metathesis generally is the cis alkene, as the reaction is reversible, equilibration typically leads to the thermodynamically most favorable alkene geometry. In macrocycles, it is not uncommon to obtain a mixture of alkene geometries. This has led many researchers to reduce the alkene by catalytic hydrogenation using palladium on charcoal after cleavage from the resin. Robinson et al. have shown that this reduction could perhaps be performed more conveniently on solid phase using Wilkinson’s catalyst and medium hydrogen pressure (60 psi) [56]. As for many transition-metal-catalyzed reactions, microwave irradiation can be a particularly effective method to rapidly reach high temperatures resulting in short reaction times. This is also the case in the metathesis for peptide cyclization. For example, the dicarba analog of α-conotoxin was obtained in just 1 hour with microwave irradiation while 48 hours were required with conventional heating [57].


The practical and general implementation of CM [58] is challenging for alkenes of comparable reactivity, as a statistical mixture of homo and heterocoupling is typically observed. Blechert et al. quickly recognized that immobilization of one of the olefinic partners would provide a means to isolate the desired heterocoupling product albeit without preventing some homocoupling in solution. To this end, different olefins were reacted in the presence of [Ru]-I to tritylpolystyrene-bound N-Boc-N-allylglycinol and N-Boc-C-allylglycinol (Figure 13.10) [59]. Of course, the reaction was most productive with an excess of the alkene in solution, as the latter was partially consumed in the formation of a homodimer product, which in principle can react with the polymer-bound alkene albeit more slowly. It was also shown that CM could be performed on an allyl silane resin which offers mild acidic cleavage conditions (TFA results in rapid protodesilylation) in addition to cleavage with concomitant allylation using an acetal with a Lewis acid (TiCl4 ) [60, 61]. CM has also been used in conjunction with ROM of strained ring systems. Cuny et al. capitalized on strained norbornene to synthesize highly functionalized

363


364

R (3–6 equiv.) [Ru]-I (4 mol%)

O

N Boc

O

R

N Boc

12–48 h

Trityl-resin Blechert (1996) R (1–2 equiv.) [Ru]-I (5 mol%)

Si

Si

R

R

3% TFA

CH2Cl2, 40 °C, 18 h

CH2Cl2

Allyldimethylsilylpolystyrene 1.3 mmol/g Blechert (1997) R 1. O O

(10 equiv.)

X

+

2. TFA:CH2Cl2 (1 : 1) MeO2C

X= N

Wang resin 0.85–1.01 mmol/g

N

R

R

[Ru]-I (10 mol%) CH2Cl2, 23 °C, 18 h

O

X

MeO2C ; -NH(CH2)3NH- ;

X

MeO2C

2.7 : 1 to 1 : 1

H

O

H

O

-NH(CH2)2O(CH2)2O(CH2)2NH-

Cuny (1997) R O O

(5 equiv.)

O N H

X

O

[Ru]-I (10 mol%)

N H

O

CH2Cl2, 23 °C, 18 h

Wang resin MeO2C 0.69 mmol/g

X = -(CH2)3- ; -(CH2)2O(CH2)2O(CH2)2-

4 resins

O

O N H

X

N H

HN

MeO2C

R

X

N H R 2N

O 4608 compounds

32 resins

R

Cuny (1998) O

H N

O

Fmoc

H N

O

CH2Cl2, 40 °C, 16 h

n

n = 1, 2

O

C10H21 (4 equiv.) [Ru]-I (10 mol%)

Fmoc

n

Wang resin 0.6–0.8 mmol 90% capped with SiMe2t-Bu

n = 1, 2

C10H21

Gibson (1998)

O

O P

O

HO

R

2

O

OMe

[Ru]-I (20 mol%) R

1

CH2Cl2, 25 °C, 48 h

R

2

O O P OMe OH

O R 1

R

2 3

R

R

1

Nicolaou (1998)

Figure 13.10

Pioneering examples of cross-metathesis on solid phase.

cyclopentanes after a ring opening/CM (Figure 13.10) [62]. In this case, the reaction was performed at room temperature with [Ru]-I. Under these conditions, metathesis with the terminal alkene produced in the ring-opening reaction is very slow, the ROM being the driving force for the reaction. The fact that the strained alkene is polymer bound reduces the efficacy of its polymerization and favors the CM pathway. However, this will of course depend to some extent on the loading of the resin. The unsymmetrical norbornenes generated two regioisomeric


products [63]; nevertheless, the methodology was robust enough to be used in the generation of a library of 4608 highly functionalized cyclopentanes by using different diamine linkers and different terminal aryl olefins [64]. In an attempt to apply the CM for the derivatization of amino acids, Gibson et al. studied the reaction of polymer-bound unsaturated amino acids with different alkenes. The reaction of polymer-bound homoallylglycine (Hag) was achieved successfully for aryl- and alkyl-substituted alkenes when very low loading of Wang resin was used (about 0.06–0.1 mmol g−1 ), although the efficiency of the reaction dropped dramatically for vinylglycine and dehydroalanine [65]. It is important to note that this study was performed with [Ru]-I and the lack of reaction in the latter two examples should be overcome using second-generation catalysts. Nicolaou et al. used a CM in a strategic bond-forming reaction to access muscone via a Horner–Wadsworth–Emmons cyclorelease using a ketophosphonate linker. In this case, the mildness and functional group compatibility of the CM were essential for this carbon–carbon bond formation. The CM between unprotected unsaturated alcohols and terminal alkenes using [Ru]-I proceeded in 60–70% yield [66]. Concurrently to the aforementioned efforts by Schmidt et al. using a cyclorelease reaction to cleave an oligosaccharide from the resin, Seeberger et al. relied on a CM under ethylene atmosphere to cleave the oligosaccharide (Figure 13.11). This latter linker is readily prepared from Merrifield resin and 4-octene-1,8-diol in one step, and affords the n-4-pentenyl glycosides which can be engaged in further glycosylations promoted by electrophilic reagents such as NIS. This linker has been applied to the synthesis of different polysaccharides [67–69], including the automated synthesis of antigens Gb-3 and Globo-H [70]. This linker has also been used for amino sugars such as the amino glycoside antibiotics and heparin. In this case, it was necessary to use [Ru]-II and pent-1-ene for the cleavage [71]. However, an alkenic linker is incompatible with two types of glycosyl donors (thioglycosides and of course n-pentenyl glycosides) because of the fact that both of them require an electrophilic activation which cannot be performed in presence of the olefinic double bond. To this end, it was shown that the alkene could be masked as a vicinal dibromide which was compatible with NIS-mediated glycosylation and that the latter vicinal dibromide functionality could be converted back to the alkene under the action of tetrabutylammonium iodide (TBAI). After regeneration of the olefin, the linker could be cleaved, as previously described, by CM with ethylene [72, 73]. Other examples of CM on solid phase catalyzed by [Ru]-I include the formation of stilbenoids [74]. The development of the more reactive second-generation catalysts proved particularly beneficial for CM. This reaction was applied to the solid-phase synthesis of a library of 3-(aryl)alkenyl β-lactams as cholesterol absorption inhibitors. The CM step was carried out in high yields using [Ru]-II (5 mol%), and the whole synthetic sequence gave good overall isolated yields and the E-alkenes exclusively [75]. In an alternative strategy to the aforementioned use of the RCM to access peptidic macrocycle surrogate of disulfide, Liskamp et al. used [Ru]-II to perform a CM between two allylglycine units which were subsequently macrolactamized en route to the DE fragment of nisin Z [76]. Having the bicyclic intermediate as a reference, the authors investigated

365


366

BnO O O O O O BnO PivO BnO PivO PivO O BnO

BnO

[Ru]-I (20 mol%)

O

BnO

BnO

H2C CH2

BnO O O O O BnO PivO BnO PivO PivO BnO

O

36 h, 23 °C

BnO

O

Merrifield resin 0.45–0.55 mmol/g Seeberger (1999)

1. LiBr, CuBr2 MeCN, THF

O DMTO

2. Cl2CHCO2H

AcO O 1. SEt BnO OPiv BnO TBAI Br NIS, TMSTf 4-butanone O Dioxane, 95 °C R′O 2. Guanidine Br MeOH 48 h

O HO

[Ru]-I (20 mol%) H2C CH2 CH2Cl2, rt, 36 h R′O R′ = oligosaccharide

Reiterate Seeberger (2000) O O

O

n R2 [Ru]-II (2 × 5 mol%)

N

O

N

n

CH2Cl2, 40 °C, 20 h 35–78% yield

Wang resin

R2

R1

Mata (2006)

R1

[Ru]-II Fmoc-Alg-OH H BocN

OH BocHN

O

1,1,2-trichloroethane FmocHN overnight, 60 °C O 56% ArgoGel-OH resin

O

H BocN

O

O

O N H

O

O

N H

O

O

H N

N H

HN

O CONHTrt

D-E fragment of nisin Z (11% overall yield)

Liskamp (2007) O OBz

5

O

O O

TsN

OBz

5

TsN

[Ru]-I (4 cycles)

O

O

CH2Cl2, 35 °C

Merrifield resin Couladouros (2007) Cl

[Ru]-VII (5 mol%) quinine

O O

N H

O

3

O N

O

Ti (Oi-Pr)4 CH2Cl2 40 °C, 24 h

O

OH

RO

Polystyrene resin

N

60% yield Cl

Koide (2007)

O

O S

N H

[Ru]-III (50 mol%) Cy2BCl

OtBu H N Rink O

FmocHN

Toluene, 80 °C, 16 h

O FmocHN

(5 equiv.) Liskamp (2008)

Figure 13.11

Selected examples of cross-metathesis on solid phase.

O S

N H

OtBu H N O

Rink

O

13.7 Ene–yne Metathesis on Solid Phase

the more challenging prospect of obtaining the desired bicyclic system in one step by RCM of the linear peptide containing the four allyl glycine units, which proved feasible. The lower reactivity of first-generation catalyst can be an asset when performing a CM on a terminal alkene in the presence of an internal alkene. For the synthesis of prosophylline, a natural product which belongs to the family of the piperidine alkaloids known to have medicinal properties, Couladouros et al. relied on a CM of a terminal alkene in the presence of an endocyclic alkene. After extensive experimentation, it was found that four consecutive cycles at a fairly low temperature (35 ◦ C) using [Ru]-I gave optimal yield [77]. Koide and Garner exploited the CM as a general method to derivatize natural products with the addition of an epoxide moiety [78]. Having a heteroatom in the allylic position can be problematic in a metathesis reaction, as it can coordinate to the ruthenium carbene intermediate thereby slowing the catalytic cycle. To this end, the addition of a Lewis acid such as Ti(OiPr4 ) to compete for chelation with the epoxide led to higher yields. In the latter case, introduction of a spacer between the trityl linker and the resin was critical for obtaining higher yields. It was also noted that [Ru]-VII [79] afforded better results. The optimized conditions (longer linker, [Ru]-VII, and Lewis acid) enable derivatization of complex natural products such as quinine. Another application of the CM is Liskamp’s synthesis of alkene dipeptidosulfonamide as isosteres of the native peptide bond [80], in which a polymer-bound sulfonamide is reacted with readily available allylic amines. The reactions were found to require fairly high catalyst loading and the use of a Lewis acid (Cy2 BCl) to afford only modest yields. Nevertheless, the peptide incorporating this isostere was found to have improved biological activity thus warranting this method. Other examples of CM using second-generation catalysts also include the preparation of porphyrin–RGD peptide conjugate [81]. As it is the case for solution-phase chemistry, it has been noted that CM with α,β-conjugated systems performed better with [Ru]-III rather than [Ru]-II [82].

13.7 Ene–yne Metathesis on Solid Phase

The ene–yne metathesis, as its name implies, is a bond-forming and reorganization reaction between an alkyne and an alkene to yield a 1,3-diene. The reaction is mechanistically akin to the alkene metathesis and has been carried out both in the intermolecular and intramolecular version. The first examples of the reaction on solid phase were reported by Blechert et al. using the intermolecular version of the reaction. In the first example, a polymer-bound propargyl ester was reacted with a variety of highly functionalized alkenes in the presence of [Ru]-I, affording an allylic diene which could be released from the resin by taking advantage of the reactivity of allylic esters under palladium catalysis [83]. It is interesting to note that, as the cleavage is leveraged on the reactivity of allylic esters, only compounds having successfully participated in the ene–yne metathesis will be released. In the second example, the reaction was achieved between an allylic

367


368

Nu:, Pd(PPh3)4

R1 O O

O

[Ru]-I (7−10 mol%)

n O n = 2, 8

CH2Cl2, 40 °C, 18 h

THF 50–60 °C, 8 h

O

O O

R1

n O

Nu

R1

Et3N+HCO2−

Merrifield resin 0.6–0.8 mmol/g

R

Pd(PPh3)4 THF, 70 °C

1

(45–86% overall yield)

Blechert (1998) R Si

R

[Ru]-I (5.5 mol%)

1.5% TFA

Si

CH2Cl2, 0 °C, 18 h

R

CH2Cl2 (39–62% overall yield)

Polystyrene based resin (1% DVB), 0.9 mmol/g Blechert (1998) R1 O

CH2Cl2, 40 °C

O

R1

O

[Ru]-I (2 × 5 mol%) O

24 h Ene–yne metathesis

HMBA-AM resin 0.4-1.0 mmol/g Blechert (1999)

O [Ru]-I (5 mol%)

N O

R2

Wang resin 1.7 mmol/g

R

O

2

N 3

O

Diels−Alder

R1

O

R

O

CH2Cl2/toluene −35 °C

R1

R1

1

O

R2 MeAlCl2

R N

O

R Cyclative cleavage

O

O

R1

O

N

N

Benzene 75 °C, 18 h

R2

(14–28% overall yield)

Toluene 105 °C, 18 h

O

NR

O

O

R2

R2

Heerding (1998) O

i-Pr

MgBr

1.

O

O

THF, 0 °C to rt

Si i-Pr

Polystyrene based resin 1.4 mmol/g

2.

Si

Oi-Pr B

i-Pr

OH B

OH

OH H O

O

N Ph

i-Pr

H O

i-PrO [Ru]-I (15 mol%) CH2Cl2, 40 °C, 19 h

Schreiber (2002)

(i-PrO)2B

i-Pr

R Si

O i-Pr

R

[Ru]-III (15 mol%) Toluene, 80 °C, 24 h

OH

i-Pr

Allyl boronate additions Vinyl boronate coupling reactions

R Si

O i-Pr

Polystyrene based resin 1.3 mmol/g

O

B OH R

Diels–Alder reactions Chemoselective oxidations

Schreiber (2005)

Figure 13.12

Ene–yne metathesis on solid phase.

silane polystyrene resin and several highly functionalized alkynes, thus affording polymer-bound dienes which could be released from the resin under mild acidic condition (protodesilylation) to afford an isomerized diene product as mixture of (E)- and (Z)-isomers. The broad scope of the reaction was demonstrated with a variety of substrates such as Fmoc-protected amino acids bearing an alkene ester and protected sugars (Figure 13.12) [84]. Blechert et al. recognized early on that the ene–yne reaction was particularly attractive for diversity-oriented synthesis. Indeed, it allowed bringing diversity in a complexity-building reaction. The resulting

13.8 Conclusion

1,3-diene product could indeed be engaged in further powerful complexity-building reactions such as a Diels–Alder or palladium cross-couplings. The application of an ene–yne metathesis followed by a Diels–Alder reaction was demonstrated a year later by Blechert et al. The product of the Diels–Alder was further harnessed as a handle for cyclization/cleavage with amines to afford octahydrobenzapinones with high purity [85]. The potential of intramolecular ene–yne metathesis coupled to a Diels–Alder reaction was elegantly used in a library synthesis by Heerding et al. for the preparation of 4200 isoindolines using split-and-mix synthesis. Analysis of a sample of the library showed that the final products were typically obtained in over 70% purity [86]. Schreiber et al. further elaborated the potential of this reaction sequence in the context of diversity-oriented synthesis by using boronic esters in the ene–yne annulations, thus providing an extra handle to divergent reaction pathways [87, 88]. In this case, it was observed that the efficacy of the reaction in solution could not be directly transferred to solid phase using highly loaded macrobeads. A significant difference in performance was indeed observed between [Ru]-II and [Ru]-III despite the fact that the key intermediate in the catalytic cycle was the same in both cases. It was also noted that marked improvement in the reaction could be obtained when the reaction was carried out under an ethylene atmosphere. This suggests that regeneration of the catalyst following the ene–yne annulation may be critical for good performance of the reaction. Indeed, if the intermediate from the ene–yne annulations has the potential to participate in a subsequent RCM thus releasing the catalyst in solution, the performance of both [Ru]-II and [Ru]-III was comparable.

13.8 Conclusion

The functional group tolerance of the ruthenium-based metathesis catalysts has had a tremendous impact on solid-phase organic synthesis. The efficacy of the reaction in solution generally translates directly to solid-phase transformation and its potential has been harnessed in a number of library syntheses, solid-phase syntheses of natural products, or diversity-oriented syntheses. It enables the use of chemically robust alkenes as linkers which can be cleaved by RCM or CM. It, of course, provides new manifolds of diversification in diversity-oriented synthesis as has been elegantly shown in landmark examples by Schreiber and Nelson. Another metathesis application of paramount importance is in peptide chemistry where solid-phase synthesis is omnipresent. The ability to stabilize secondary structures in short peptide motifs and replace pharmacologically unsuitable disulfide bonds or simply restrict the conformation of a peptidic library has already been successfully implemented in a number of important examples. The orthogonality of the metathesis reaction to peptide chemistry provides a really powerful tool in this regard.

369

370


Acknowledgments

We thank our collaborators who have contributed to our work in the area of diversity-oriented synthesis and solid-phase organic synthesis as well as the funding agencies (ANR, ARC, ERC, CNRS). While this review covers development of the olefin metathesis in solid-phase organic chemistry since the first reports in 1996, it is not intended to be comprehensive and we apologize for the inevitable arbitrary omissions.

References 1 Schrock, R.R. (2006) Angew. Chem. Int. 2 3 4

5 6

7

8 9 10

11

12 13 14

Ed., 45, 3748–3759. Grubbs, R.H. (2006) Angew. Chem. Int. Ed., 45, 3760–3765. Chauvin, Y. (2006) Angew. Chem. Int. Ed., 45, 3740–3747. Nicolaou, K.C., Bulger, P.G., and Sarlah, D. (2005) Angew. Chem. Int. Ed. Engl., 44, 4490–4527. Burke, M.D. and Schreiber, S.L. (2004) Angew. Chem. Int. Ed. Engl., 43, 46–58. Veerman, J.J.N., Van Maarseveen, J.H., Visser, G.M., Kruse, C.G., Schoemaker, H.E., Hiemstra, H., and Rutjes, F.P.J.T. (1998) Eur. J. Org. Chem., 11, 2583–2589. van Maarseveen, J.H., den Hartog, J.A.J., Engelen, V., Finner, E., Visser, G., and Kruse, C.G. (1996) Tetrahedron Lett., 37, 8249–8252. Ahn, Y.M., Yang, K., and Georg, G.I. (2001) Org. Lett., 3, 1411–1413. Matsugi, M. and Curran, D.P. (2005) J. Org. Chem., 70, 1636–1642. Pernerstorfer, J., Schuster, M., and Blechert, S. (1997) Chem. Commun., 1949–1950. Nicolaou, K.C., Winssinger, N., Pastor, J., Ninkovic, S., Sarabia, F., He, Y., Vourloumis, D., Yang, Z., Li, T., Giannakakou, P., and Hamel, E. (1997) Nature, 387, 268–272. Piscopio, A.D., Miller, J.F., and Koch, K. (1997) Tetrahedron Lett., 38, 7143–7146. Piscopio, A.D., Miller, J.F., and Koch, K. (1999) Tetrahedron, 55, 8189–8198. Nicolaou, K.C., Vourloumis, D., Li, T., Pastor, J., Winssinger, N., He, Y., Ninkovic, S., Sarabia, F., Vallberg, H., Roschangar, F., King, N.P.,

15

16

17

18

19

20

21 22 23 24

25

Finlay, M.R.V., Giannakakou, P., Verdier-Pinard, P., and Hamel, E. (1997) Angew. Chem. Int. Ed. Engl., 36, 2097–2103. Timmer, M.S., Verdoes, M., Sliedregt, L.A., van der Marel, G.A., van Boom, J.H., and Overkleeft, H.S. (2003) J. Org. Chem., 68, 9406–9411. Moriggi, J.-D., Brown, L.J., Castro, J.L., and Brown, R.C.D. (2004) Org. Biomol. Chem., 2, 835–844. Marsault, E., Hoveyda, H.R., Peterson, M.L., Saint-Louis, C., Landry, A., Vezina, M., Ouellet, L., Wang, Z., Ramaseshan, M., Beaubien, S., Benakli, K., Beauchemin, S., Deziel, R., Peeters, T., and Fraser, G.L. (2006) J. Med. Chem., 49, 7190–7197. Morton, D., Leach, S., Cordier, C., Warriner, S., and Nelson, A. (2009) Angew. Chem. Int. Ed., 48, 104–109. Luo, Z., Zhang, Q., Oderaotoshi, Y., and Curran, D.P. (2001) Science, 291, 1766–1769. Curran, D.P. (2004) Handbook of Fluorous Chemistry, John Wiley & Sons, 101–127. Zhang, W. and Curran, D.P. (2006) Tetrahedron, 62, 11837–11865. Knerr, L. and Schmidt, R.R. (1999) Synlett, 1802–1804. Knerr, L. and Schmidt, R.R. (2000) Eur. J. Org. Chem., 15, 2803–2808. Brohm, D., Philippe, N., Metzger, S., Bhargava, A., Muller, O., Lieb, F., and Waldmann, H. (2002) J. Am. Chem. Soc., 124, 13171–13178. Brohm, D., Metzger, S., Bhargava, A., Muller, O., Lieb, F., and Waldmann, H.

References

26

27

28 29

30

31

32 33

34

35

36

37

38

39 40

41

(2002) Angew. Chem. Int. Ed. Engl., 41, 307–311. Timmer, M.S.M., Codee, J.D.C., Overkleeft, H.S., van Boom, J.H., and van der Marel, G.A. (2004) Synlett, 2155–2158. Crowley, J.I., Harvey, T.B.III, and Rapoport, H. (1973) J. Macromol. Sci., Chem., 7, 1117–1126. Pernerstorfer, J., Schuster, M., and Blechert, S. (1999) Synthesis, 138–144. Lee, D., Sello, J.K., and Schreiber, S.L. (1999) J. Am. Chem. Soc., 121, 10648–10649. Stanton, B.Z., Peng, L.F., Maloof, N., Nakai, K., Wang, X., Duffner, J.L., Taveras, K.M., Hyman, J.M., Lee, S.W., Koehler, A.N., Chen, J.K., Fox, J.L., Mandinova, A., and Schreiber, S.L. (2009) Nat. Chem. Biol., 5, 154–156. Koide, K., Finkelstein, J.M., Ball, Z., and Verdine, G.L. (2001) J. Am. Chem. Soc., 123, 398–408. Lee, D., Sello, J.K., and Schreiber, S.L. (2000) Org. Lett., 2, 709–712. Arya, P., Durieux, P., Chen, Z.-X., Joseph, R., and Leek, D.M. (2004) J. Comb. Chem., 6, 54–64. Khadem, S., Joseph, R., Rastegar, M., Leek, D.M., Oudatchin, K.A., and Arya, P. (2004) J. Comb. Chem., 6, 724–734. Arya, P., Couve-Bonnaire, S., Durieux, P., Laforce, D., Kumar, R., and Leek, D.M. (2004) J. Comb. Chem., 6, 735–745. Prakesch, M., Sharma, U., Sharma, M., Khadem, S., Leek, D.M., and Arya, P. (2006) J. Comb. Chem., 8, 715–734. Reddy, P.T., Quevillon, S., Gan, Z., Forbes, N., Leek, D.M., and Arya, P. (2006) J. Comb. Chem., 8, 856–871. Umarye, J.D., Lessmann, T., Garcia, A.B., Mamane, V., Sommer, S., and Waldmann, H. (2007) Chem. Eur. J., 12,13, 3305–3319. Winssinger, N. and Barluenga, S. (2007) Chem. Commun., 22–36. Barluenga, S., Dakas, P.-Y., Ferandin, Y., Meijer, L., and Winssinger, N. (2006) Angew. Chem. Int. Ed., 45, 3951–3954. Barluenga, S., Wang, C., Fontaine, J.G., Aouadi, K., Beebe, K., Tsutsumi, S., Neckers, L., and Winssinger, N.

42

43 44

45 46

47

48

49

50 51 52

53

54

55

56

57

58 59

(2008) Angew. Chem. Int. Ed. Engl., 47, 4432–4435. Blackwell, H.E., Clemons, P.A., and Schreiber, S.L. (2001) Org. Lett., 3, 1185–1188. Liao, Y., Fathi, R., and Yang, Z. (2003) J. Comb. Chem., 5, 79–81. Olenyuk, B., Jitianu, C., and Dervan, P.B. (2003) J. Am. Chem. Soc., 125, 4741–4751. Blackwell, H.E. and Grubbs, R.H. (1998) Angew. Chem. Int. Ed., 37, 3281–3284. Schafmeister, C.E., Po, J., and Verdine, G.L. (2000) J. Am. Chem. Soc., 122, 5891–5892. Walensky, L.D., Kung, A.L., Escher, I., Malia, T.J., Barbuto, S., Wright, R.D., Wagner, G., Verdine, G.L., and Korsmeyer, S.J. (2004) Science, 305, 1466–1470. Bernal, F., Tyler, A.F., Korsmeyer, S.J., Walensky, L.D., and Verdine, G.L. (2007) J. Am. Chem. Soc., 129, 2456–2457. Chapman, R.N., Dimartino, G., and Arora, P.S. (2004) J. Am. Chem. Soc., 126, 12252–12253. Chapman, R.N. and Arora, P.S. (2006) Org. Lett., 8, 5825–5828. Schmiedeberg, N. and Kessler, H. (2002) Org. Lett., 4, 59–62. Derksen, D.J., Stymiest, J.L., and Vederas, J.C. (2006) J. Am. Chem. Soc., 128, 14252–14253. Stymiest, J.L., Mitchell, B.F., Wong, S., and Vederas, J.C. (2003) Org. Lett., 5, 47–49. Pattabiraman, V.R., Stymiest, J.L., Derksen, D.J., Martin, N.I., and Vederas, J.C. (2007) Org. Lett., 9, 699–702. Berezowska, I., Chung, N.N., Lemieux, C., Wilkes, B.C., and Schiller, P.W. (2007) J. Med. Chem., 50, 1414–1417. Whelan, A.N., Elaridi, J., Harte, M., Smith, S.V., Jackson, W.R., and Robinson, A.J. (2004) Tetrahedron Lett., 45, 9545–9547. Robinson, A.J., Elaridi, J., Van Lierop, B.J., Mujcinovic, S., and Jackson, W.R. (2007) J. Pept. Sci., 13, 280–285. Testero, S.A. and Mata, E.G. (2008) J. Comb. Chem., 10, 487–497. Schuster, M., Pernerstorfer, J., and Blechert, S. (1996) Angew. Chem. Int. Ed. Engl., 35, 1979–1980.

371

372

13 Metathesis Reactions in Solid-phase Organic Synthesis 60 Schuster, M., Lucas, N., and Blechert, S. 61

62 63 64 65

66

67

68

69 70

71 72

73 74

75

(1997) Chem. Commun., 823–824. For a review covering the scope of the metathesis at the time: Grubbs, R.H. and Chang, S. (1998) Tetrahedron, 54, 4413. Cuny, G.D., Cao, J., and Hauske, J.R. (1997) Tetrahedron Lett., 38, 5237–5240. Arjona, O., Csaky, A.G., and Plumet, J. (2000) Synthesis, 6, 857–861. Cao, J., Cuny, G.D., and Hauske, J.R. (1998) Mol. Diversity, 3, 173–179. Biagini, S.C.G., Gibson, S.E., and Keen, S.P. (1998) J. Chem. Soc., Perkin Trans. 1, 2485–2500. Nicolaou, K.C., Pastor, J., Winssinger, N., and Murphy, F. (1998) J. Am. Chem. Soc., 120, 5132–5133. Andrade, R.B., Plante, O.J., Melean, L.G., and Seeberger, P.H. (1999) Org. Lett., 1, 1811–1814. Palmacci, E.R., Plante, O.J., Hewitt, M.C., and Seeberger, P.H. (2003) Helv. Chim. Acta, 86, 3975–3990. Castagner, B. and Seeberger, P.H. (2007) Top. Curr. Chem., 278, 289–309. Werz, D.B., Castagner, B., and Seeberger, P.H. (2007) J. Am. Chem. Soc., 129, 2770–2771. Kanemitsu, T. and Seeberger, P.H. (2003) Org. Lett., 5, 4541–4544. Melean, L.G., Haase, W.-C., and Seeberger, P.H. (2000) Tetrahedron Lett., 41, 4329–4333. Seeberger, P.H. and Haase, W.C. (2000) Chem. Rev., 100, 4349–4394. Chang, S., Na, Y., Shin, H.J., Choi, E., and Jeong, L.S. (2002) Tetrahedron Lett., 43, 7445–7448. Testero, S.A. and Mata, E.G. (2006) Org. Lett., 8, 4783–4786.

76 Ghalit, N., Kemmink, J., Hilbers,

77

78 79

80

81

82

83 84 85 86

87

88

H.W., Versluis, C., Rijkers, D.T.S., and Liskamp, R.M.J. (2007) Org. Biomol. Chem., 5, 924–934. Couladouros, E.A., Strongilos, A.T., and Neokosmidis, E. (2007) Tetrahedron Lett., 48, 8227–8229. Garner, A.L. and Koide, K. (2007) Org. Lett., 9, 5235–5238. Michrowska, A., Bujok, R., Harutyunyan, S., Sashuk, V., Dolgonos, G., and Grela, K. (2004) J. Am. Chem. Soc., 126, 9318–9325. Brouwer, A.J., Elgersma, R.C., Jagodzinska, M., Rijkers, D.T.S., and Liskamp, R.M.J. (2008) Bioorg. Med. Chem. Lett., 18, 78–84. Sol, V., Chaleix, V., Granet, R., and Krausz, P. (2008) Tetrahedron, 64, 364–371. Poeylaut-Palena, A.A., Testero, S.A., and Mata, E.G. (2008) J. Org. Chem., 73, 2024–2027. Schuerer, S.C. and Blechert, S. (1998) Synlett, 166–168. Schuster, M. and Blechert, S. (1998) Tetrahedron Lett., 39, 2295–2298. Schurer, S.C. and Blechert, S. (1999) Synlett, 1879–1882. Heerding, D.A., Takata, D.T., Kwon, C., Huffman, W.F., and Samanen, J. (1998) Tetrahedron Lett., 39, 6815–6818. Micalizio, G.C. and Schreiber, S.L. (2002) Angew. Chem. Int. Ed., 41, 152–154. Brittain, D.E.A., Gray, B.L., and Schreiber, S.L. (2005) Chem. Eur. J., 11, 5086–5093.

373

Index

a (+)-trans-195A 327 aburatubolactam A 330 (–) -acylfulvene 256 africanol 344 agalacto-spirolactone B 191 aigialomycin D 152 aigialomycin D analog 357 Alder-ene reaction 315 aldol condensation 110 alkene linker 350–353 alkylation – catalytic phase-transfer 54 – conjugate addition of organocuprate 6 – intramolecular N-alkylation 58 – Mitsunobu 78 – of imine with functionalized alkyl lithium 62 – palladium-catalyzed domino allylic 54 alkyne carboxylation, nickel-mediated 321 allo-colchicine 190 allylation 118 – N-allylation 62, 64 – asymmetric 20, 69 – coupling of hemiketal with allylsilane 101 – diastereoselective via N-acyliminium ion 65, 71 – double palladium-catalyzed decarboxylative 24 allylboration, asymmetric 65, 77 allylic amination, asymmetric 48, 66 allylic hydride addition 142 D-altritol 227 ambrettolide 211 aminopalladation 67, 74 amphidinolide 165f – (−)-amphidinolide E 197, 200

– amphidinolide V 221 – amphidinolide X 254 amphinidol 3 291 (−)-anaferine 333 (+)-anatoxin-a 187 angucyclinone 191 annonaceous acetogenin 237f anolignan 197f, 308 ansamycin derivative 172 anthramycin 308 anthramycin (+)-anthramycin 193 (−)-archazolid B 270, 274 arenastatin 173 asimicin 301 aspercylide C 168 asperpentyn 274, 277 (+)-aspicilin 160 (+)-astrophylline 333 attenol A 231f auxiliary-based synthesis 64, 72 aza-Diels-Alder reaction 197 azasugar 49ff, 69ff

b baconipyrone C 305, 348 Bayer-Villiger reaction of ketone, regioselective 63 bicyclic acetal 113 bicyclo[4.3.1]phosphate 135f bis-alkoxysilane 227 bistramide A 303, 325 1,3-bis(2,4,6-trimethylphenyl)− 2-imidazolidinylidene 45 bitungolide E 140 (+)-brasilenyne 253 brevitoxin 88f, 117ff bryostatin, ring-expanded analog of 280


374

Index

c Cadiot-Chodkiewicz reaction 315 callipeltroside A 278 carbapenem 186 carpacephem 186 cascade metathesis 303, 313ff – asymmetric ROM/CM 304 – catalytic carbene complexes for 314 – CM/RCM 306 – ene-ene RCM/CM 314 – ene-yne-ene RCM/CM 315f, 321 – ene-yne-ene RCM/RCM 316ff – for preparation of tricyclic structures 319ff – metallotropic [1,3]-shift/CM 315 – ring-closing ene-yne metathesis/CM 307 – ROM/RCM see ring-rearrangement metathesis – ROM/CM of cyclopropene ketal 303, 322ff – ROM/CM/RCM 305 – ROM/RCM/CM 305 – RRM/CM 337 – RRM/RCM 332ff (+)-castanospermine 326 (−)-centrolobine 328 characiol 279 chlorotonil A 293f ciguatoxin 109ff citreofuran 218 civetone 162, 211 clavirolide 169 cleavage reaction – of α-hydroxy ketone 15 – of silyl ether 55 clusianone 297 CM see cross-metathesis (−)-cochleamycin A 244 coleophomone 169 colletol 289 conformation constraint 31 (R)-coniine 346 cornexistin 250 cross-metathesis (CM) 25, 45, 74, 138ff, 193, 239 – alkyne CM 309f – control of double bond configuration 226 – cross-coupling of side chains 300f – for appending functionalized side chains 299f – for appending unfunctionalized side chains 296ff – for partly removing a side chain 298f – of ene-yne systems 184f, 195ff, 308f – on solid phase 362ff – promoted by microwave heating 288, 361

– relay cross-metathesis (relay CM) 264, 271, 274 – two-directional 287 – with 1,1-disubstituted alkenes in 2-methyl-2-butene 297 – with acrolein 291 – with acrylate 290 – with acrylic acid 288 – with acrylimide 289 – with acryonitrile yielding (Z)-alkene 287 – with allyl phosphonate 294 – with allyl silane 294 – with allylic alcohol derivative 294f, 299 – with thioacrylate 288 – with vinyl boronate 292 – with vinyl ketone 292 – with vinyl silane 293 cruentaren A 215 cuprate addition 136, 139f cuscohygrine 333 (+)-cyanthiwigin U 336 cyclization – by electrochemical oxidation 6 – Pictet-Spengler 54, 77 cyclization see also macrocyclization – Bischler-Napieralski 38, 52 – boron trifluoride-mediated 23 – cationic, of allylic O, N-acetals 73 – electrophilic selenocyclization 240 – fused heterocycle system via 1,4prototropic shift/[3 + 2] cycloaddtion 288 – intramolecular allylstannane/acetal 117 – Mitsunobu 59, 247 – of 1,6-dienes via RRCM 263 – of allylidene via RRCM 266, 274 – of electron-deficient β-ketoalkylidene via RRCM 266f – of hindered substrates via RRCM 268ff – of hydroxy-epoxide to furan ring system 98 – of olefinic esters using Tebbe reagent 123 – of trienes 99 – of triene-yne precursor 116f – one-pot sequential 72 – radical 25 – stereoselective 5-exo trig 8 – to lactams 66 – via intramolecular allylation 117 – via reductive amination 55 – zirconium-mediated 10 cyclodepsipeptides 173f cyclopeptide 150

Index cyclophanes 168 cycloproparadiciol 153 cyclorelease reaction 350f – of dysidiolide analogs 352 – of epothilone 351 – of lactams 351 – of O-alkenyl glycosides 352 – of polymer-bound oligosaccharides 352 – of tetrapeptide macrocycle 351 – of β-turn mimetics 352 cylindramide A (+)-cylindramide A 338 cylindramide A 289, 305 cylindrocyclophane 306

d Dauben oxidation 28 decarboxylative oxygen-to-carbon migration 19 dehydrohomoancepsenolide 309 10-deoxymethynolide 158 (+)-14-deoxy-9-oxygigantecin 238 desymmetrization – lipase-mediated 49 – of C2 -symmetric diol using phosphate tether 135 – palladium-catalyzed, of meso-bis-carbamate 57 – via ROM 304 – via RRCM 275 diallylamine formation 47, 52, 62f – diastereoselective 65 – enantioselective 51, 54, 59, 67 diallylation, regioselective 77 diazoester cyclopropanation, intramolecular 10 (2S, 7S)-dibutyroxynonane 235f dictyostatin 132, 140 (+)-didemniserinolipid B 271, 275 Diels-Alder cycloaddition 26f, 102f, 189, 289 – transannular, biomimetic 243 1,6-diene 1 dienolide 150 (±)-differolide 187 (+)-dihydrocuscohygrine 335 dihydro-epi-deoxyarteannuin B 270, 273 dihydrooxasiline 272 dihydroxylation 26, 49, 56, 58f, 77 – diastereoselective syn-dihydroxylation 57 – face-selective, of cyclic olefin 69 – side chain-directed syn-dihydroxylation 50 discodermolide – (+)-discodermolide 132, 140 272, 276

diterpenoides 150, 169f dolabelide C 139ff, 292 double enantioselective allyltitanation 62 double-bond preorganization in cyclic systems 15f (+)-dumetorine 333 dynamic kinetic asymmetric transformation (DYKAT) 51

e enantioselective synthesis 49, 54f, 58, 61 – enantioselective olefin metathesis 343ff – of functionalized piperidine 77 – of oxazolidinone derivative 57 ene reaction, asymmetric 6 ene-yne metathesis 45, 92 – on solid phase 369ff ene-yne precursor 52 enzymatic resolution 69 epilacnene 211 epothilone 163ff, 213, 239f, 351 epoxidation – hydroxyl-directed of homoallylic alcohol 29 – selective 28, 60, 69 – Sharpless 62, 67, 80 epoxide hydrolysis, selective 70 ergorgiane 307 eromophilane 189 erythocarine 322 erythravine 322 erythroidine 323 24,24-ethanovitamine D3 lactones 197 exaltolide 163

f faranal 289 (+)-ferruginine 187 Fischer tungsten-carbene complex 183 floresolide B 170 fluvirucin B1 172 FR901464 302 fumagillin 6 Fürstner catalyst see metathesis catalyst, [Ru-V]

g (−)-galanthamine 191 gambieric acid 122 gambierol 120f, 307 gauche-effect in vicinal diol 95 geldanamycin 272 geodin A 340 (+)-gigantecin 238

375

376

Index gloeosporone 159 glycopyranosyl cation 69 d,l-glucosylceramide 252 Grignard reaction 77, 104, 218 – of alkyl bromide and Weinreb amide 51 griseoviridin 172 Grubbs catalyst see metathesis catalyst, [Ru-I] and [Ru-II] 45 Grubbs-Hoveyda catalyst see metathesis catalyst, [Ru-III] (±)-guanacastepene A 319

– acid-catalyzed 23 – of double bonds 31, 74 isooncinotine 175

h

l

halichondrin B 334 (−)-halosaline 335 harveynone 274, 277 heliannuol 300 hemibrevitoxin B 307 heterocyclic carbene 45 heterolytic fragmentation of bicyclic systems 14 histrionicotoxin 288 homoallyl alcohol 252 – with conjugated diene moiety 131 homologation 255 – diastereoselective 107 Horner-Wadsworth-Emmons coupling to enones 105 hydrogenation 55, 79, 115 – of olefin bond 60, 62 – regioselective, of bicyclic phosphate 141f – selective 176 hydrosilylation 254 14-hydroxycornexistin 250 (10S, 12R)-10-hydroxy-12-methyl-1oxacyclododecan-2,5-dione 159 1α-hydroxy-5,6-trans vitamin D2 248 hypoestoxide 272, 275

lancifodilactone G 194 (R)-lasiodiplodin 155f lasonolide A 295f (−)-lasubine II 339 latrunculin 216 lejimalide 163 lemonomycin 294 lepadine alkaloids 318 (+)-ent-lepadin B 331 Linchpin coupling, three-component longithorone – (−)-longithorone 187 – longithorone C 271

i illudin 256 indolizdine – (−)-indolizidine 167B 335 – indolizidine 251F 330 intramolecular displacement of allylic benzoate ester 60 iodoboration, chemoselective 81 iododesilylation 254 iodolactonization 10 ipomoeassin E 176 (−)-irofulvene 256 isoaltholactone 290 isomerization

j Julia-Kocienski coupling

295

k kinetic resolution, hydrolytic 63 Knoevenagel condensation 8, 77 (±)-trans-kumausyne 330

72

m macrocyclic glycolipids 175 macrocyclic lipids 171 macrocyclic musk 162, 211f macrocyclic natural products – biogenetic classification of 151 – structural classification of 150 macrocyclic polyketide see also macrolide 152ff, 167f macrocyclization – stereoselectivity of double-bond formation 150 – using Otera’s catalyst 291 – via relay ring-closing metathesis 265ff – via Yamaguchi reaction 288 macrolactams 172f macrolide 140f, 150, 265ff – antibiotic 158 – resorcinylic 152ff – salicylate 155ff macrosphelide 161 Mannich reaction, intramolecular 55 manzamine A 175 marine ladder toxins 109ff marine peptide 47 medium size carbocycle 1 melithiazole C 300

Index metacycloprodigiosin 199 metal-alkylidene complex 46 metallatropic [1,3]-shift 274 metathesis catalyst – [Mo]-I 46, 80, 89f, 121, 183, 227, 252, 306 – [Ru]-I 1f, 4ff, 46, 305, 308, 321 – [Ru]-II 1f, 6ff, 30ff, 46, 166, 231f, 241, 280, 293, 308 – [Ru]-III 46, 92, 138f, 141, 195, 239, 280, 288, 291, 299, 308, 316 – [Ru]-IV 46, 183 – [Ru]-IX 46, 74 – [Ru]-V 46, 88, 106f, 155 – [Ru]-VI 27, 46 – [Ru]-VII 302, 319, 369 – [Ru]-VIII 183, 271 – [Ru]-XI 46 – [Ru]-XII 271 – [Ru]-XIII 3 – bearing a C2 -symmetric diolate ligand 345ff – chiral [Mo]-II 344 – chiral [Mo]-III 344f – chiral [Ru]-XV 304, 346 – comparison of, for oxepine synthesis 91 – fluorous-tagged f -[Ru]-III 351 – for cascade metathesis reaction 314 – for RCAM 206f – imidazolin-2-iminato tungsten alkylidene catalyst 208 – polymer-bound ruthenium carbene complex 256 – removal of, from target molecules 351 – stereogenic-at [Mo]-IV complex 346 – triaryl amido molybdenum complex for alkyne CM 310 metathesis reaction type 45 migrastatine 160 MM-47755 192 molybdenum hexacarbonyl 206 monoterpenoid indole alkaloid 72 Mortreux-type reaction 208 motuporamine C 211 (−)-mucocin 237f, 270 muricatacin 303 muscone 162 muscopyridine 163 mycoepoxydiene 305 – (+)-mycoepoxydiene 338 mycothiazole 309 – structure of (±)-mycothiazole 132 – total synthesis of (±)-mycothiazole, via sulfonylation-RCM sequence 132f

n nakadomanine 175 narbonolide 158 Negishi coupling 254 neodihydrohistrionicotoxin nephrosteranic acid 297 nonylprodigiosyn 175 17-norcharaciol 281 nupharamine 292

132

o (+)-ochoromycinone 192 okadaic acid 230f okilactomycin 167 olefin isomerization inhibitor 37 oocydin A 266, 294, 296 oxazolidinone preparation 65, 66 oxecin skeleton 88 oxepine skeleton 88, 92, 96, 100, 109 oxidation – of metathesis catalyst 351 – oxidative cleavage of diol 110 – Pinnick 67 – Swern 61, 68 – Tamao-Fleming 22, 245, 249 oximidine 157 oximidine III 265 oxocine skeleton 88, 92, 97f, 101, 104f oxonene ring system skeleton 88, 97f, 102ff oxyallyltitanation of aldehyde, diastereoselective 37

p Paal-Knorr reaction 174 palmerolide 162 palytoxin 132 (3R, 9R, 10R)-panaxytriol 314f peloruside A 236, 265 peptide coupling 47 perhydropyrrolo[2,1-j]quinolone 72 Petasis reaction 56 Peterson elimiation 22 (−)-PF1163B 173 PGE2 methyl ester 309 (−)-phaseolinic acid 228 phenanthroindolizidine alkaloid 54 phlegmarine-type alkaloid 68 phoslactomycin B 276 phosphate – allylic displacement 134, 141f – as tether in multipodal coupling reactions 134 – cyclic phosphate from allylic alcohol and Z-dienol 134f

377

378

Index – limitations 278 resorcylide 154 ricinelaidic acid lactone 159 ring contraction – Favorskii reaction 25 – zirconium-promoted 7 ring expansion 75 ring opening of epoxide 59 ring-closing alkyne metathesis (RCAM) 205f – complementary ene-yne metathesis 221 – mechanism of 206 – promoted by microwave heating 208 – scope and limitations 210, 212 – synthesis of stereodefined macrocyclic alkenes 211ff ring-closing metathesis (RCM) 1ff, 45 – chelation effect 106 – competing side reactions 90, 114, 157 – conformational constraint 174 q – cyclorelease reactions based on 351f quartomicin 191 – diastereoselectivity 112f quebrachamine 346 – effect of protective groups 107 – for preparation of oxygenated medium-sized r ring system 87ff radicicol 152 – formation of (E)-trisubstituted alkene 166 RCAM see ring-closing alkyne metathesis – formation of tricyclic 5,7,5-ring system 25f RCM see ring-closing metathesis – in combination with intramolecular rearrangement reaction 52 allylstannane/acetal cyclization 117f – cycloisomerization of ene-yne system 197f – in formation of (sesqui)terpenoid structures – Eschenmoser-Claisen rearrangement 23 4ff – glycolate-Claisen rearrangement 11 – in formation of carbahexofuranose 8f – Ireland-Claisen 1, 18, 72, 92, 94, 115 – in formation of carbapentofuranose 8f – Johnson-Claisen rearrangement 1f – in formation of cylcopentenes 1f – metathetic ring rearrangement 38 – in presence of Lewis-acid 160 – Overman 54, 64 – in synthesis of ciguatoxin ring system – [2,3]-rearrangement 116, 272, 335 110ff – [3,3]-sigmatropic rearrangement – of allylic acetate 9 recifeiolide 163 – of allylic alcohol 6 reduction – of diene-diene system to conjugated triene – N-acyliminium-mediated 67 157 – conjugate reduction of enone 4 – of diene-ene system via cobalt complex – of ketone, stereoselective 115 153f – of lactams 61, 69 – of diene-yne systems 184f reductive amination with allyl amine 60 – of electron-rich olefins 90 Reformatsky reaction 81 – of ene-yne systems 183ff, 189 relay concept 261f relay ring-closing metathesis (RRCM) 251f, – of enol ether-olefins 121f – of epoxy-diene 99 316 – control of directionality in RRCM 267, 275 – of ether-olefins 89ff – of kinetic silyl enol ether 72 – ene-yne RRCM 269f – of olefinic sulfonates 130f – interference from premature – of phosphates 136 macrocyclization 279f, 283 – of polymer-bound dienes 355ff – interference from truncation reaction 278ff – of polymer-bound polyenes 355ff phosphate (contd.) – in organic synthesis 134 – mediation of CM 138 – multivalent activation by phosphate tether 135f – reactivity of bicyclic phosphate from RCM 136ff [2 + 2]-photocycloaddition 72 pipecolic acid see 2-piperidine carboxylic acid 2-piperidine carboxylic acid 66 pladienolide 159 pochonin 152, 154f proline homolog 66 prosophylline 367 prostaglandin 299 prostaglandin PGE2 -1, 15-lactone 212 pterocarpan 247f 2-(1H)-pyridone ring system 68

Index – of precursor with protected hydroxyl 51, 56f – of precursor with unprotected hydroxyl 9, 51, 60 – of stannyl ketals 96 – of tosyl-protected precursor 77 – of trienes to oxepines 96 – of vinyl ketone 3f, 6 – regioselectivity 105, 166 – relay strategy 157, 261f – simultaneous multiple RCM 9 – stereoselectivity of double-bond formation 82, 106, 226f – strategies for preparing epothilones 164, 353 – two-directional sequence 116 – two-step protocol using titanium ethylidene reagent 121 – under ethylene gas atmosphere 185, 191 – under high dilution conditions 81 – use of tetra(isopropyl) titanium as co-catalyst 106, 333 – ynolide approach to resorcinylic macrolides 153 ring-opening metathesis (ROM) 26, 346 – of ene-yne systems 184f ring-opening metathesis polymerization (ROMP) 45 ring-rearrangement metathesis (RRM) 45, 49, 54, 57, 325f – diastereoselective 328 – enantioselective 344f – mechanism of 330 – of bicyclic substrates 329f – of monocyclic substrates 326f – of norbornene derivatives 329ff – transfer of stereocenter 325 rocellaric acid 297 rollicosin 303 ROM see ring-opening metathesis ROMP see ring-opening metathesis polymerization roseophilin 173, 202 Roush reaction see allylboration, asymmetric routiennocin 304 RRCM see relay ring-closing metathesis RRM see ring-rearrangement metathesis (+)-rubiginone B2 192 ruthenacyclobutene 183ff

s Sakurai reaction 246 salicylate lactone 215 salicylihalamide A 140, 156f

(−)-sarain A 174 scabrosin 309 (+)-scanloenyne 271, 274 SCH 351448 279 – (+)-SCH 351448 161 Schrock catalyst see metathesis catalyst, [Mo-I] (−)-securinine 317 silicon tether – achiral silaketal tether 229 – allylsiloxane 245ff – C2 -symmetrical silaketal 227f – long-range asymmetric induction 233f – racemic silaketal tether 229 – symmetrical silaketal tether 226f – unsymmetrical bis-alkoxyalkynylsilane 242f – unsymmetrical silaketal tether 230 – vinylsiloxane 245ff, 251ff skeletal reorganization reaction 197ff, 201 solid-phase catalyst extraction 351 solid-phase metathesis reaction see also cyclorelease reaction 350ff – cross-linking of α-helices 360 – cross-metathesis 364ff – formation of 1,3-dienes via ene-yne metathesis 367ff – intraresin dimerization 358f – restricting peptide conformation through cyclization 359ff Sonogashira coupling 293 sophorose lipid lactone 213 sparteine alkaloids 77 spirastrellolide A 281 spirocyclization 27 spirofungin A 231, 304 (−)-spongidepsine 173 sporiolide B 159 squamostolide 303 stannyl ketal 96 (−)-stemoamide 185 stereoselective synthesis 56, 252 Stille cross-coupling 4, 18 (+)-streptazolin 250 streptorubin B 199 sulfoximine derivative 92 sulfur dioxide extrusion 133 sultone – leaving group ability 130f – preparation of, from olefinic sulfonyl chlorides and alkenols 130ff – synthesis of (±)-mycothiazole 132f (−)-swainsonine 328 (+)-cis-sylvaticin 237

379

380

Index synthesis – of achiral (Z)-1, 4-diols 226, 234f, 238 – of acyclic tri-/tetrasubstituted 1,3-dienes 255 – of (Z)-alkenes via CM with acrylonitrile 287ff – of (E)-alkenes via CM with electron-deficient terminal olefin 287ff – of D-altritol via TST-RCM 228 – of ansa-bridged azafulvene skeleton 174 – of aza-acenaphthylene skeleton 72f – of 1-aza[4.3.0]-bicyclic skeleton 47 – of azabicyclo[3.2.1]octene ring 187 – of azepines 80f – of benz[a]anthraquinone skeleton via ene-yne RCM 191f – of bicyclo[3,3.1]nonane skeleton 13 – of bicyclo[2.2.2]octane skeleton 13 – of bridged carbocycle – – ingenol 26 – – solanoeclepin A 27 – – vinigrol 32 – of carbacephem skeleton 187 – of carbapenem skeleton 187 – of carbon-linked disaccharides 240 – of cis-fused benzofuran-benzopyran skeleton 246f – of cis-hexahydroxy−1H-indene skeleton 197 – of conjugated Z-dienols 131f – of cyclic alkynes 207 – of cyclic ether via TST-RCM/cross-coupling sequence 254 – of cyclophane derivatives 208 – of 1,3-dienes via CM 196 – of 1,3-dienes via RCAM/hydrosilylation/dehydrosilylation 219 – of 1,3-dienes via stereoselective TST-RCM 241f – of dihydropyrroles 47 – of dihydropyrrolidines 56 – of (anti,anti)-1, 3-dihydroxy-4-methyl stereotriad 140 – of anti-1, 3-diols using phosphate tether 140ff – of dipyrrolidines 49 – of eight-membered carbocycle 14f – – fusicoccin A 30 – – (±)-mycoepoxydiene 31 – – ophiobolin M 30 – – serpendione 30 – – taxol B ring 33 – – vinigrol analog 32

– of eight-membered nitrogen heterocycle 81f – – manzamine A 81f – of eight-membered oxygen heterocycle (+)-laurencin 95f, 101 – – (+)-laurenyne 101 – – brevetoxin G-ring 121 – – ciguatoxin E-ring 110f – – heliannuol A 104f – – octalactin A 105 – of 2-ene-1,5-diols 240f – of exocyclic ethylidene moiety via TST-RCM 249f, 257 – of five-membered carbocycle 1ff – – (−) -allosamizoline 4 – – (−) -bis(hydroxymethyl)cyclopentenyl adenine 2f – – 4a-carba-ß-D-galactofuranose 8 – – 1,22-dihydroxynitianes 4 – – dumsine from (R)-carvone 4f – – ecklonialactone 1f – – elisabethin A from (R)-carvone 9 – – estrone 10 – – (+)-fomannosin 6 – – (−)-guanacastepene E 6 – – (−)-heptemerone B 6 – – (±)-laurokamurene B 3 – – nitiol 3f – – (+)-puraquinoic acid 8 – – sequosempervirin A 2 – – spiroepoxide structure 6 – – spirotenuipesine 9 – – tashironin 4 – – tetrasubstituted cyclopentene 9f – – trisubstituted cyclopentene 6f – – via RRCM 263 – of five-membered nitrogen heterocycle 47ff – – (−)-antofine 54 – – (+)-australine 59 – – (+)-1-epi-australine 59 – – (+)-broussonetine G 51 – – (−)-bulgecinine 50 – – castanospermine 57 – – (−)-trans-dendrochrysine 49f – – 1,4-dideoxy-1,4-imino-D-ribitol (+)-DRB 49f – – 2,5-dideoxy-2,5-imino-D-mannitol (+)-DMDP 50 – – dysinosin A 48 – – (+)-β-erythroidine 52f – – erythravine 52f – – erythrocarine 52f – – (+)-hyacinthacine A2 60

Index – – – – – – – – – – – – – – – – – – – – – – – – – –

–

– – – – – – – – – – – –

– (+)-hygrine 48 – (−)-indolizidine 55 – lentiginosine 55ff, 70f, 73f – ent-lentiginosine 71 – (−)-(S)-Nicotine 47ff – oscillarin 47ff – (−)-phaseolinic acid 228 – rhynchophylline 52 – iso-rhynchophylline 52 – (−)-swainsonine 57f – (−)-trachelanthamidine 61 – using ene-yne RCM 185 of five-membered oxygen heterocycle 273 of functionalized medium-sized azacycles 355 of fused carbocycle (+)-arglabin 29 – compactin 18 – durhamycine aglycone model 20 – guanacastepene 28 – heptemerone G 28 – (−)-10α-hydroxy-4-muurolen-3-one from (R)-carvone 19 – (−)-perrotinene 18 – sphenolobane-type diterpenoids 28 – (−)-tormesol 28 – vannusal A 19f – via RCM 37f, 188, 190, 192, 198, 306, 319, 337, 344 of fused nitrogen-containing polycyclic systems via RCM 39, 51, 53f, 56f, 59f, 61, 68, 72, 74ff, 164, 172f, 175f 186f, 189, 201f, 272, 317, 339, 346, 357 of fused oxygen-containing polycyclic systems via RCM 36f, 51, 53, 57, 59f, 68, 89ff, 101, 103ff, 109ff, 121f, 135ff, 140ff, 153ff, 161f, 167ff, 173, 176f, 194, 214, 217, 306, 316f, 321, 339, 351 of fused ring system 104, 109f of heliananes 104f of heterocycle containing a N–O-bond 355ff of hindered 5,5-fused bicyclic nitrogen heterocycle 59 of hindered (Z)-trisubstituted alkene via RRCM 265 of γ -hydroxy aldehyde via TST-RCM 245 of indolizidines 52ff, 56, 70ff of isoprenyl side chain via CM 297 of lignans via TST-RCM 246f of macrolactams 351f of macrolactones 353, 359 of medium-sized carbocycles via TST-RCM/cross-coupling sequence 253f

– of medium-sized cyclic ether systems 89f, 93ff – of medium-sized lactams 351, 357 – of medium-sized lactones via sulfoximines 92, 94 – of medium-sized sulfonamides 354 – of metacyclophane-macrolactams via RRCM 269f – of monohydroxypyrrolidine 61 – of nine-membered carbocycle 14 – – hydroxycornexistin 34 – – pestalotiopsin 33 – of nine-membered oxygen heterocycle 97 – – astrogorgin 103 – – benzo-fused bicyclic systems 104 – – (+)-brasilenyne 254 – – ciguatoxin F-ring 113f – – eleutherobin 102f – – eunicellin derivatives 102f – – isolaurallene 98 – – ophirin B 102 – – ribose-fused bicyclic systems 103 – of nitrogen heterocycles via ene-yne RCM 185ff – of octahydrobenzapinones via ene-yne metathesis/Diels-Alder reaction 369 – of Z-olefin by TST-RCM 226f – of oligoenynes 317 – of oxygen heterocycle via ene-yne RCM 187 – of pipecolic acid derivatives via CM 197 – of piperidine carboxylic acids 66f – of piperidines 61ff, 346 – of piperidones 68ff – of polycyclic indolizidines 52ff – of polyhydroxyindolizidines 55ff – of polyhydroxypiperidines 69f – of polyhydroxypyrrolidines 49 – of protected divinyl ketone from cyclopropene ketal 303 – of pyrrolidines 47ff, 193, 308 – of pyrrolizidines 59ff – of pyrrolo-oxazolones 57 – of quinolizidines 73 – of semicyclic 1,3-diene 307 – of seven-membered carbocycle 14 – – allo-colchicine 190 – – (−)-cyanthiwigin F 25 – – (−)-dihydroxanthatin 194 – – (±)-frondosin B 23 – – lancifodilactone G 194 – – methyllicaconitine analog 23 – – rameswaralide 24 – – (+)-sundiversifolide 24 – – thapsigargin 25

381

382

Index synthesis (contd.) – – trisubstituted cycloheptene 24 – – (+)−8-epi-xanthatin 318 – of seven-membered nitrogen heterocycle 78ff – – azepine derivative 80, 186 – – (−)-balanol 78, 80 – – (−)-stemoamide 81, 186 – – using ene-yne RCM 185 – of seven-membered oxygen heterocycle 92, 95 – – (−)-isoprelaurefucin 100 – – brevetoxin E-ring 118 – – ciguatoxin A-ring 109, 113 – – ciguatoxin D-ring 110f – – gambieric acid B-ring 122f – – gambierol H-ring 121 – of silaketals 227ff – of six-membered carbocycle – – asperpentyn 277 – – bicyclic systems 14f – – carbasugar 20f – – conduritols 21f – – cyclitol 20 – – (+)-cyclophellitol 20 – – dihydro-epi-deoxyarteannuin B 273 – – eremophilane 190 – – fumagillol 10f – – (−)-galanthamine 192 – – garsubbelin A 14 – – harveynone 277 – – ovalicin 12 – – rac-ottelione B 12 – – pancratistatin analogs 22 – – phosphatidylinositol analog 21 – – platencin 13 – – polyprenylated acylphloroglucinol 13f – – quartromicin 191 – – tricholomenyn A 277 – – (+)-valienamine 21 – of six-membered nitrogen heterocycle 61ff – – (−)-205B 72, 76 – – (−)-allosedamine 61ff – – (+)-(S)-angustureine 78, 80 – – (S)-anabasine 65, 67 – – (S)-anabatine 65, 67 – – baikiain 67f – – α-conhydrine 63 – – (−)-β-conhydrine 65f – – coniceine 70, 72 – – (S)-(+)-coniine 64f – – ent-CP-999,994 66f – – 1-deoxyallonojirimycin 69, 71 – – (−)-deoxynupharidine 75, 77

– – – – – – – – – – – – – – – – – – – – – – – – – – – – –

– – – – – – – – – – – – – – – – –

– (+)-epiquinamide 75, 78 – fagomine 71 – 4-hydroxy-2-pipecolic acid derivatives 66 – (±)-α-isosparteine 79 – ent-L-733,060 66f – (−)-lasubine I 76, 78 – lepadiformine 72, 74 – lycoposerramine-V 68f – (+)-meloscine 72, 75 – mitralactonine 77, 79 – (−)-pipecoline 66 – pipermethystine 69f – quinolizidine 233A 73ff – (+)-sedamine 61ff – α-sparteine 78 – (2S, 4S)-(−) -SS20846A 67f – tetrahydropyridine derivative 70 – tetraoxobispidine 77 – using ene-yne RCM 185 of six-membered oxygen heterocycle 269, 273, 347f of spiro-carbocycle – (+)-elatol 19 – (+)-laurencenone B 19 of spiroketals via TST-RCM 230ff of stereodefined macrocyclic alkenes by RCAM/semi-reduction 211ff stereoselective, of 1,n-diol via TST-RCM 234ff stereoselective, of cyclic ethers 246 stereoselective, of homoallylic alcohols 251f of 2-substituted furans via RCAM/cycloisomerization of alkyn-4one skeleton 220 tandem ene-yne RCM/metallatropic shift reaction 274ff tandem enyne-RCM process 52f tandem RCM/CM process 302 tandem ROM/RCM sequence 114 of ten-membered carbocycle 14f, 34f – bridged polycyclic systems 17 – ent-clavilactone B 36f – diversifolin 35 – eleutherobine precursor 37f – (+)-eremantholide A 36f – excentricine 38f – germacranolides 15 – (±)-periplanone C 16 – stereoselectivity of double-bond formation 38 of ten-membered lactones 106ff of ten-membered oxygen heterocycle – herbarumin I 106

Index – – – – – – – – – – –

– herbarumin II 106 – herbarumin III 109 – microcarpalide 107f – pinolidoxin 107 of tetrahydrooxazepines 356 of tetrahydropyrrolizidine 59 of trans-fused oxepane polyethers 255f of tricyclic ring systems 82 of (E, E, E)-triene skeleton 172 of trihydroxyindolizidines 57 of Z-trisubstituted alkenes via TST-RCM 240, 248ff – of unsaturated cyclic phosphates 134ff – of α,β-unsaturated lactams 52, 268, 357 – of α,β-unsaturated lactones 187, 357 – of unsaturated sultones 130ff synthetic strategy – aldol/RCM sequence 97f – chirality transfer/RRCM sequence 268f – CM/reductive amination sequence 292 – combined ROM/B935RCM strategy 54 – cuprate addition/Grignard reaction/RCM sequence 12 – domino Stille/Diels-Alder reaction 82 – Evans-Tishchenko/RCM sequence 105 – glycolate alkylation-RCM sequence 99, 102 – intramolecular Mannich reaction/B886 RCM process 55 – iodolactonization/radical reduction process 23 – Ireland-Claisen rearrangement/TST-RCM strategy 228 – isomerization/CM sequence 298 – radical cyclization/skeletal rearrangement 13 – radical cyclization/RCM sequence 111f – RCAM/cycloisomerization strategy 219 – RCAM/hydrosilylation/protodesilylation strategy 219 – RCAM/Lindlar hydrogenation strategy 211 – RCM/hydrogenation sequence 11 – RCM/ring fragmentation sequence – – for preparation of anti-Bredt alkenes 17 – – for preparation of carboycle 15 – RCM-carbomagnesation process 52 – RCM/ROM/RCM cascade 38 – reductive isomerization/RCM sequence 113 – scission/recombination process 45 – [3,3]-sigmatropic rearrangement/RCM sequence 1ff, 11, 13, 18, 94, 191 – stereoselective [2,3]-Wittig rearrangement/RCM sequence 115

– sulfonylation/B959RCM sequence 131ff – thio acetal/free-radical cyclization strategy 112 – TST-RCM/CM strategy 239 – TST-RCM/organolithium ring-opening strategy 251f – TST-RCM/palladium-mediated cross-coupling strategy 253f – TST-RCM/Sakurai reaction strategy 247

t tartolon B 242 taxane analog 322 Tebbe reagent 123 temporary silicon-tethered RCM (TST-RCM) 226ff – chemoselectivity in 248 – diastereoselective reaction with alkenols 234f – enantioselective, of prochiral alkene 255f – for preparation of 2-ene-1,5-diols 240f – for preparation of Z-trisubstituted alkenes 240f – for stereoselective preparation of 1,n-diols 234ff, 245 – long-range asymmetric induction 233f – of acyclic mono-alkoxysilane 245 – of diene-yne systems 241ff – of enynesiloxane 256 – spiroketalization 232 terpenoides 169ff terpestacin 167, 309 tethered metathesis reaction see also RRCM 225ff tetrangomycin 192 tetraponerine 54, 328 thuggacin B 293 tipranavir 347 titanium alkylidene 121f tonantzilolone 169 trichloromenyn A 274, 277 tricolorin A 175 triene-yne precursor 116 trienomycin A 272 tropane alkaloid 48 TST-RCM see temporary silicon-tethered RCM tuberostemonine 298 turriane 208

u uniflorine A 55ff

383

384

Index

v vinylation, diastereoselective (+)-virgatusin 247 (+)-viroallosecurinine 317 vitamin D2 249

y 64

w Weinreb amide 65 Wittig olefination 4, 13, 61, 64, 70, 110 woodrosin I 176

x xanthatin derivative – (−)-dihydroxanthatin 194 – (+)-8-epi-xanthatin 316

YM-181741 192 yuzu lactone 211

z zampanolide 161 zeralenone 152, 156 zeranol 152

Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts

Cp2TiCl in Natural Product Synthesis

Alkene Metathesis in Organic Synthesis

Combinatorial synthesis of natural product-based libraries

Combinatorial Synthesis of Natural Product-Based Libraries

Synthesis of Solid Catalysts

Olefin Metathesis and Metathesis Polymerization, Second Edition

Catalysts for Fine Chemical Synthesis. Microporous and Mesoporous Solid Catalysts

Olefin Metathesis and Metathesis Polymerization

Catalysts for Fine Chemical Synthesis

Catalysts for Fine Chemical Synthesis, Microporous and Mesoporous Solid Catalysts

Catalysts for fine chemical synthesis

Fischer-Tropsch Synthesis, Catalysts and Catalysis

Studies in Natural Product Chemistry, Volume 18: Stereoselective Synthesis, Part K (Studies in Natural Products Chemistry)

Natural Product Research

Strategies and Tactics in Organic Synthesis, Volume 6 (Strategies and Tactics in Organic Synthesis)

Strategies and Tactics in Organic Synthesis, Volume 7 (Strategies and Tactics in Organic Synthesis)

Strategies and Tactics in Organic Synthesis, Volume 4 (Strategies and Tactics in Organic Synthesis)

Catalysts for Fine Chemical Synthesis, Hydrolysis, Oxidation and Reduction (Catalysts For Fine Chemicals Synthesis) (Volume 1)

Advances in Fischer-Tropsch Synthesis, Catalysts, and Catalysis (Chemical Industries)

Palladium reagents and catalysts. Innovations in organic synthesis

Transition metal reagents and catalysts: innovations in organic synthesis

Transition metal reagents and catalysts: innovations in organic synthesis

System Synthesis: Product and Process Design

Natural Product Isolation (Methods in Biotechnology)

Metathesis Polymerization

Asymmetric Organocatalysis in Natural Product Syntheses

Catalysts for Fine Chemical Synthesis, Catalysts for Carbon-Carbon Bond Formation (Catalysts For Fine Chemicals Synthesis) (Volume 3)

Natural Product Chemistry at glance

Improving Product Reliability: Strategies and Implementation

Catalysts For Fine Chemical Synthesis, Vol. 1

Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts

Cp2TiCl in Natural Product Synthesis

Alkene Metathesis in Organic Synthesis

Combinatorial synthesis of natural product-based libraries

Combinatorial Synthesis of Natural Product-Based Libraries

Synthesis of Solid Catalysts

Olefin Metathesis and Metathesis Polymerization, Second Edition

Catalysts for Fine Chemical Synthesis. Microporous and Mesoporous Solid Catalysts

Olefin Metathesis and Metathesis Polymerization

Catalysts for Fine Chemical Synthesis

Catalysts for Fine Chemical Synthesis, Microporous and Mesoporous Solid Catalysts

Catalysts for fine chemical synthesis

Fischer-Tropsch Synthesis, Catalysts and Catalysis

Studies in Natural Product Chemistry, Volume 18: Stereoselective Synthesis, Part K (Studies in Natural Products Chemistry)

Natural Product Research

Strategies and Tactics in Organic Synthesis, Volume 6 (Strategies and Tactics in Organic Synthesis)

Strategies and Tactics in Organic Synthesis, Volume 7 (Strategies and Tactics in Organic Synthesis)

Strategies and Tactics in Organic Synthesis, Volume 4 (Strategies and Tactics in Organic Synthesis)

Catalysts for Fine Chemical Synthesis, Hydrolysis, Oxidation and Reduction (Catalysts For Fine Chemicals Synthesis) (Volume 1)

Advances in Fischer-Tropsch Synthesis, Catalysts, and Catalysis (Chemical Industries)

Palladium reagents and catalysts. Innovations in organic synthesis

Transition metal reagents and catalysts: innovations in organic synthesis

Transition metal reagents and catalysts: innovations in organic synthesis

System Synthesis: Product and Process Design

Natural Product Isolation (Methods in Biotechnology)

Metathesis Polymerization

Asymmetric Organocatalysis in Natural Product Syntheses

Catalysts for Fine Chemical Synthesis, Catalysts for Carbon-Carbon Bond Formation (Catalysts For Fine Chemicals Synthesis) (Volume 3)

Natural Product Chemistry at glance

Improving Product Reliability: Strategies and Implementation

Catalysts For Fine Chemical Synthesis, Vol. 1

Recommend Documents