Organicof Synthesis via Examination Selected Natural Products
7815tp.indd 1
2/28/11 1:33 PM
This page intentionally left blank
Organicof Synthesis via Examination Selected Natural Products
David J Hart The Ohio State University, USA
World Scientific NEW JERSEY
7815tp.indd 2
•
LONDON
•
SINGAPORE
•
BEIJING
•
SHANGHAI
•
HONG KONG
•
TA I P E I
•
CHENNAI
2/28/11 1:33 PM
Published by World Scientific Publishing Co. Pte. Ltd. 5 Toh Tuck Link, Singapore 596224 USA office: 27 Warren Street, Suite 401-402, Hackensack, NJ 07601 UK office: 57 Shelton Street, Covent Garden, London WC2H 9HE
British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library.
Cover image credits: Sponge: Reproduced from Office of Ocean Exploration and Research, National Oceanic and Atmospheric Administration, U.S. Department of Commerce http://oceanexplorer.noaa.gov/explorations/03bio/logs/hirez/lasonolide1_hires.jpg Rauwolfia serpentina: Reproduced with permission from Dr. Aruna Radhakrishnan
ORGANIC SYNTHESIS VIA EXAMINATION OF SELECTED NATURAL PRODUCTS Copyright © 2011 by World Scientific Publishing Co. Pte. Ltd. All rights reserved. This book, or parts thereof, may not be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording or any information storage and retrieval system now known or to be invented, without written permission from the Publisher.
For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA. In this case permission to photocopy is not required from the publisher.
ISBN-13 978-981-4313-70-4 ISBN-10 981-4313-70-X
Typeset by Stallion Press Email:
[email protected]
Printed in Singapore.
XiaoLing - Organic Synthesis.pmd
1
3/9/2011, 2:11 PM
b1026_FM.qxd
12/21/2010
10:54 AM b1026
Page v
Organic Synthesis via Examination of Selected Products
This book is dedicated to
My Teachers and Mentors Mark Green Richard Lawton William Dauben David Evans John Swenton;
My Parents Harold and Geraldine Hart;
My Students and Colleagues at The Ohio State University;
My Family;
and Especially My Wife Rose, the Spice of My Life.
b1026_FM.qxd
12/21/2010
10:54 AM b1026
Page vi
Organic Synthesis via Examination of Selected Products
This page intentionally left blank
b1026_FM.qxd
12/21/2010
10:54 AM b1026
Page vii
Organic Synthesis via Examination of Selected Products
Contents
Preface
ix
Chapter 1
Introduction
1
Chapter 2
Steroids
15
Chapter 3
Prostaglandins
71
Chapter 4
Pyrrolizidine Alkaloids
137
Chapter 5
Juvabione and the Vicinal Stereochemistry Problem
155
Chapter 6
Functional Group Reactivity Patterns and Difunctional Relationships
203
Chapter 7
Some Unnatural Products — Twistane and Triquinacene
245
Chapter 8
Alkaloids — Difunctional Relationships and the Importance of the Mannich Reaction
279
Chapter 9
Alkaloids from “Dart-Poison” Frogs
335
Chapter 10 Morphine and Oxidative Phenolic Coupling vii
403
b1026_FM.qxd
12/21/2010
10:54 AM b1026
viii
Page viii
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
Chapter 11 Olefin Synthesis and Cecropia Juvenile Hormone
445
Chapter 12 A Recent Example of Structure Determination Through Total Synthesis and Convergent Syntheses: Lasonolide A
479
Chapter 13 Ionophores: Calcimycin
503
Chapter 14 Erythromycin A Aglycone
535
Concluding Remarks
559
Index
563
b1026_FM.qxd
12/21/2010
10:54 AM b1026
Page ix
Organic Synthesis via Examination of Selected Products
Preface
This book is based on a graduate level course I taught at The Ohio State University in the autumn of 2006. The course consisted of twenty-eight 50-minute classes over a period of ten weeks (Chemistry 941). Students enrolled in the course were largely in their second or third years of graduate school. All had taken a three-quarter “organic reactions” sequence and a two-quarter “physical organic chemistry” sequence. I expected students to bring a sizeable toolbox of reactions, and a sound understanding of mechanistic organic chemistry, with them to the classroom. My goal was to introduce students to the field of organic synthesis with a focus on my own interest in natural products synthesis. This book follows the sequence of topics I discussed in Chemistry 941. I have done little to modify the slides that I used as the basis of lectures. I have merely added text to accompany each slide. Several homework assignments were presented during the quarter, and I have added many more problems that I hope readers will find interesting and instructive.a An index has also been appended.b My view of organic synthesis has naturally been influenced by my own experiences. I have been influenced by my teachers and, if they read this book, they will see themselves reflected on many of these pages. I have also been influenced by other books on the topic of organic synthesis, a few of which appear below: •
Ireland, R. E. Organic Synthesis, Prentice-Hall, 1969 (147 pages)
a
A partial answer key is available to course instructors. Please contact
[email protected] The index within the book is abbreviated and only provides information found in the text pages (odd-numbered pages). A thorough index of the slides (even-numbered pages) with sections organized by compound, reagent type, reaction type, and subject is available online at http://www.worldscibooks.com/chemistry/7815.html
b
ix
b1026_FM.qxd
12/21/2010
10:54 AM b1026
x
• • • • • •
Page x
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
Fleming, I. Selected Organic Syntheses: A Guidebook for Organic Chemists, John Wiley and Sons, 1973 (227 pages) Warren, S. Organic Synthesis: The Disconnection Approach, John Wiley and Sons, 1982 (391 pages) Wyatt, P.; Warren, S. Organic Synthesis — Strategy and Control, John Wiley and Sons, 2007 (909 pages) Corey, E. J.; Cheng, X-M. The Logic of Chemical Synthesis, John Wiley and Sons, 1989 (436 pages) Nicolaou, K. C.; Sorensen, E. J. Classics in Total Synthesis: Targets Strategies, Methods, VCH, 1996 (798 pages) Nicolaou, K. C.; Snyder, S. A. Classics in Total Synthesis II: Targets, Strategies, Methods, John Wiley and Sons, 2003 (560 pages)
I was most influenced by the Ireland and Fleming books (now out of print), because they were published at the time I was developing an interest in organic chemistry and organic synthesis. To those who have read the Ireland and Fleming books, their influence will be apparent in my selection of topics. Also, my selection of topics is simply a reflection of my own interests. In no way do I mean to slight the many chemists not cited herein who have made landmark contributions to the field of natural products synthesis.c I have provided references on the slides for the papers that form the basis of this book.d These references are not repeated in the text, but additional references have been provided at points where I think they could be useful. The reader can always refer to the papers that form the basis of this book for additional details and citations. I have also tried to provide reaction yields when they were easily gleaned from the papers. I did not make any attempt to extract yield information from experimental data and have taken yields reported by authors at face value. Throughout the text I will refer to the “slides” by topic and number. For example there are six slides associated with the introduction. When I refer to Introduction-1, the reader should look at the first slide associated with the introduction. Slides appear on left-hand (even) pages with the accompanying text located on adjacent right-hand (odd-numbered) pages. A Table of Contents is provided that should help the reader move from one topic to another in a non-linear manner. Now let us begin. c
For some compilations of syntheses that you might find interesting see: Anand, N.; Bindra, J. S.; Ranganathan, S. Art in Organic Synthesis, John Wiley and Sons, 1970 (427 pages). Bindra, J. S.; Bindra, R. Creativity in Organic Synthesis: Volume 1, Academic Press, Inc., 1975 (322 pages). d A PowerPoint presentation of each chapter, with selected structures, bonds and comments highlighted in color, is available online at http://www.worldscibooks.com/chemistry/7815.html
b1026_Chapter-01.qxd
2
Steroids OH
H
H
H H
H
HO
H
H
Cortisone
O
HO
O CO2H
CO2H
HO
HO
OH PGF1
OH
PGA1
O CO2H
HO
OH
PGE1
HO
CO2H
OH PGF2
CO2H HO
OH PGE2
Introduction-1
Five-membered Rings Olefin Synthesis Acyclic Diastereoselection
Page 2
Prostaglandins
Six-membered Rings Five-membered Rings Stereocontrol in Cyclic Systems Acyclic Diastereoselection Biomimetic Synthesis
10:54 AM
Progesterone
Organic Synthesis via Examination of Selected Products
Cholesterol
H
O
O
b1026
H
12/21/2010
O
H
Organic Synthesis via Examination of Selected Natural Products
O OH
O
b1026_Chapter-01.qxd
12/21/2010 b1026
10:54 AM
Page 3
Organic Synthesis via Examination of Selected Products
Introduction
3
Introduction-1 One of my objectives is to provide a sense of the history of organic synthesis.1 I am not a historian, so the reader should understand that this is only my perspective of the field. It is my own sense of this history, however, that leads me to select steroids as the first family of natural products for discussion (Introduction-1). These compounds were clearly of interest to early practitioners of organic synthesis because of their biological activity, structural complexity, and central role they played in the development of the field of biosynthesis. On the other hand, one might argue that steroids were early targets for synthesis because some steroids, such as equilinen and estrone (Steroids-2), were not too complex, and thus were achievable synthetic goals given the tools available at the time. Our discussion will begin with the Woodward synthesis of cholesterol (circa 1950; the steroid we all love to eat) because it is one of the earliest examples of complex natural products synthesis. It also introduces a number of strategies that are still widely used in carbocycle synthesis (synthesis of 6-membered rings, synthesis of 5-membered rings, stereocontrol in cyclic systems). This synthesis will also touch upon other topics, such as acyclic diastereoselection, that are still of contemporary interest. We will next move to the topic of biomimetic synthesis to see how thinking about a biosynthetic pathway to a natural product can influence the design of synthetic pathways to that natural product. It will also illustrate how persistence and attention to detail can play an important role in realizing a given synthetic strategy in the laboratory (see Introduction-1 for targets). The second broad topic for discussion will be prostaglandins (Introduction-1). Interest in this family of natural products was once again stimulated by their importance in biology. The focus here will be on methods for 5-membered ring synthesis, olefin synthesis, and once again we will encounter the topic of acyclic diastereoselection. Although steroids and prostaglandins have remarkably different structures, we will see that they have some structural features in common, and there are strategic parallels to be found in their synthesis when these structural features are addressed.
b1026_Chapter-01.qxd
4
Return to Steroids: The Sidechain Stereochemistry Problem 20
Pyrrolizidine Alkaloids
C15 of Prostaglandins CO2Me
CO2Me
The Juvabiones
O
*
*
*
O *
H threo
H erythro
Difunctional Relationships
X
X
Y
X
2 1
1
1
3
Y 1,2-Difunctional Relationship
X
Y
1
5
4
Y 1,3-Difunctional Relationship
Even Difunctional Relationship
Introduction-2
Odd Difunctional Relationship
Page 4
H
H
10:54 AM
C20 of Cholesterol
OH
Organic Synthesis via Examination of Selected Products
RO
HO
7
N
12/21/2010
1
15
H
OH
H
b1026
H
HO
R
Organic Synthesis via Examination of Selected Natural Products
O
H H
b1026_Chapter-01.qxd
12/21/2010 b1026
10:54 AM
Page 5
Organic Synthesis via Examination of Selected Products
Introduction
5
Introduction-2 I will next move from a family of compounds to a general problem in organic synthesis that I call “the terpenoid sidechain stereochemistry problem” (Introduction-2). This is really a relatively simple example of a problem in acyclic diastereoselection. The central question here is: When a stereogenic center does not reside in a ring, how can one control stereochemistry at this center relative to other stereogenic centers in the molecule? Acyclic diastereoselection is a general problem presented to the synthetic chemist by many families of natural products. For example, it arises in the steroids at C20, in the prostaglandins at C15, and at the exocyclic carbon of the sesquiterpene ester juvabione. It arises, in a less obvious manner, at C1 and C7 of the pyrrolizidine alkaloids. We will revisit this problem again and again throughout this book. At this point we will examine a topic that can provide insight into strategies that have been adopted for the synthesis of a host of molecules. The topic here is that of “difunctional relationships”. For example, the juvabiones have three functional groups: ketone, ester and alkene. The central question here is: When two functional groups are present in a target, does their spatial relationship provide a clue as to what chemistry can be used to construct that relationship and/or the ease with which the relationship can be constructed? The ideas set forth in this section are derived entirely from a concept I learned as a postdoc in the laboratories of David Evans (then at the California Institute of Technology), although any misrepresentation of these ideas are my responsibility alone.
b1026_Chapter-01.qxd
6
Some Unnatural Products
Me N
Me
Me N
H N
Porantherine
O
H
Me
H
O
Me Difunctional Relationships
Difunctional Relationships
OH
HO
N
H HO
H
N CH3
H3C N
OH H
Biomimetic Synthesis
Introduction-3
H Histrionicotoxin
O OH
Morphine
O
Me
Luciduline
N
H O
Page 6
N
10:54 AM
Alkaloids
Organic Synthesis via Examination of Selected Products
Triquinacene
Where to we start?
b1026
H
12/21/2010
H
H
Organic Synthesis via Examination of Selected Natural Products
H Twistane
b1026_Chapter-01.qxd
12/21/2010 b1026
10:54 AM
Page 7
Organic Synthesis via Examination of Selected Products
Introduction
7
Introduction-3 Targets to be addressed in the next section are twistane and triquinacene (Introduction-3). These are unnatural products. Twistane consists of 6-membered rings and triquinacene of 5-membered rings. Both have low levels of functionality. How can the concept of examining “difunctional relationships” be used to develop strategies (or analyze syntheses) of these compounds? We will also look at the alkaloids porantherine and luciduline with the same question in mind. I think that it is important that aspiring practioners of synthesis lose any fear of heteroatoms they might have early in the game. Thus we will next spend some time talking about alkaloids, nitrogen-containing natural products. We will already have seen such compounds (the pyrrolizidine alkaloids) earlier in the book, so this section will serve in part to revisit strategies within the context of “new” targets. For example we will revisit the notion of “biomimetic synthesis” with morphine (arguably the most important painkiller used in medicine) and the concept of difunctional relationships within the context of histrionicotoxin and pumiliotoxin-C, neurotoxins used as a defense secretion by a certain frog species.
b1026_Chapter-01.qxd
8
More Alkaloids
H
H
N
H
HO Stereoelectronic Effects
H
N H O
H MeO2C
OMe
O OMe
OMe OMe
Reserpine
Introduction-4
HO
H
Page 8
A Classic in Alkaloid Synthesis
N H
N H
Kinetics vs. Thermodynamics for Stereocontrol
MeO
Gephyrotoxin
10:54 AM
H
Me
H
12/21/2010
Pumiliotoxin-C N H H H
H
Organic Synthesis via Examination of Selected Products
H
b1026
N
H
H
H
Organic Synthesis via Examination of Selected Natural Products
Me H
b1026_Chapter-01.qxd
12/21/2010 b1026
10:54 AM
Page 9
Organic Synthesis via Examination of Selected Products
Introduction
9
Introduction-4 We will continue with alkaloids by discussing a personal favorite (gephyrotoxin) and a classical target (reserpine) (Introduction-4). As with all of the natural (and unnatural) product targets discussed in this book, numerous total syntheses of gephyrotoxin and reserpine have been reported. It is instructive to compare syntheses, and look for strategic differences and similarities, and this will be done here as well as throughout this book.
Lasonolide A
N
O HO O
O
O
O O
CO 2H H N
O OH Calcimycin (A-23187)
Total Synthesis and Structure Determination Olefin Synthesis Macrolide Synthesis
Introduction-5
Acyclic Diastereoselection Thermodynamics and Stereocontrol
Page 10
O
10:54 AM
NHMe
O
12/21/2010
OH O
b1026_Chapter-01.qxd
Macrocyclic Natural Products and Ionophores
Organic Synthesis via Examination of Selected Products
Cecropia JH
b1026
OMe
O
Synthesis and Structure Determination Claisen Rearrangements Fragmentations Methodology Driven Synthesis Sigmatropic Rearrangements
Organic Synthesis via Examination of Selected Natural Products
O
10
Cecropia Juvenile Hormone: Stereoselective Olefin Synthesis
b1026_Chapter-01.qxd
12/21/2010 b1026
10:54 AM
Page 11
Organic Synthesis via Examination of Selected Products
Introduction
11
Introduction-5 We will then leave alkaloids and move back in time to a target that became important in the 1960’s, Cecropia juvenile hormone (Introduction-5). Approaches to this deceptively simple structure constitute a study in stereoselective tri-substituted olefin synthesis. Given the importance of olefins (both synthesis and chemistry of) to “modern” organic synthesis, I think that a visit to this “old” topic will be instructive, and will help set the stage for a discussion of targets of more contemporary interest. In addition, it will focus on the important role synthesis plays in structure determination, and the stimulus natural products can provide for the development of new synthetic methodology. We will continue with olefin chemistry by considering lasonolide A (Introduction-5). Lasonolide A is a target that will reinforce how synthesis is still an important tool for determining structure. This target will allow us to briefly look at some modern organometallic chemistry as applied to the problem of stereoselective olefin synthesis. In addition it will be used to introduce the topic of macrolide synthesis (macrocyclic lactones). We will move on to the ionophores, a large family of natural products that present many synthetic challenges. From the many targets one might discuss in this chapter, I have chosen the historically significant calcimycin (A23187). This target will provide us with a look at several strategies for synthesizing molecules with multiple stereogenic centers.
b1026_Chapter-01.qxd
12
A Polypropionate Target O
O
O O OH
MeO Erythromycin A
Strategy and Tactics Strategy = Plans
Tactics = Execution of Strategy
Assembling Carbon Skeleton
Reagent Selection
Oxidation States at Carbon
Functional Group Compatibility
Stereochemical Issues
Functional Group Interconversions
Tactical Flexibility
Stereochemical Issues
Introduction-6
Page 12
Macrolide Synthesis Acyclic Diastereoselection
10:54 AM
OH
Organic Synthesis via Examination of Selected Products
O
O O
12/21/2010
OH
b1026
HO
Organic Synthesis via Examination of Selected Natural Products
NMe 2
HO
b1026_Chapter-01.qxd
12/21/2010 b1026
10:54 AM
Page 13
Organic Synthesis via Examination of Selected Products
Introduction
13
Introduction-6 Although macrolides are of contemporary interest, the first major steps toward successful macrolide syntheses were reported in the 1970’s. Thus, for historical reasons, we will look at two approaches to the classical target erythromycin A and its aglycone, erythronolide-A. The erythromycins are examples of “polypropionates”, natural products biosynthetically derived largely from propionic acid units via a series of condensation reactions. Many natural products, broadly called polyketides, share this biosynthetic origin. These compounds are decorated with multiple stereogenic centers, and acyclic diastereoselection problems that are much more complex than “the terpenoid sidechain stereochemistry problem” will surface with erythromycin, including the problem of asymmetric synthesis. Finally, throughout this book I will use the terms “strategy” and “tactics” when discussing syntheses. These terms are not new and, in fact, there is a series that bares the title Strategies and Tactics in Organic Synthesis.2 In a broad sense what I will mean by strategy is “the plan” and by tactics I mean “execution of the plan”. Some of the general features of strategy and tactics are outlined on the slide (Introduction-6). Whereas there are some distinct differences between the two terms, there is also some overlap (for example both deal with stereochemical issues), so I feel it is important not to get too rigid with definitions here. Nonetheless I hope this will help the reader keep these issues clear when I begin to use these terms. Let’s have a look at the steroids as targets for synthesis.
b1026_Chapter-01.qxd
12/21/2010 b1026
14
10:54 AM
Page 14
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
References 1. For a retrospective view see: Nicolaou, K. C.; Vourloumis, D.; Winsdsinger, N.; Baran, P. S. “The art and science of total synthesis at the dawn of the twenty-first century” Angew. Chem., Int. Ed. 2000, 39, 44–122. 2. Strategies and Tactics in Organic Synthesis, Lindberg, T., Ed.; Academic Press; 1984, Vol. 1 (370 pages). Strategies and Tactics in Organic Synthesis, Lindberg, T., Ed.; Academic Press; 1989, Vol. 2 (469 pages). Strategies and Tactics in Organic Synthesis, Lindberg, T., Ed.; Academic Press; 1991, Vol. 3 (544 pages). Strategies and Tactics in Organic Synthesis, Harmata, M. Ed.; Elsevier; 2004, Vol. 4 (415 pages). Strategies and Tactics in Organic Synthesis, Harmata, M. Ed.; Elsevier; 2004, Vol. 5 (486 pages). Strategies and Tactics in Organic Synthesis, Harmata, M. Ed.; Academic Press; 2007, Vol. 7 (532 pages).
OH
10:54 AM
Page 16
Organic Synthesis via Examination of Selected Products
12/21/2010 b1026
Steroids-1
Fusidic Acid
Organic Synthesis via Examination of Selected Natural Products
H
Testosterone Cholic Acid
H HO
O
H
H H H
H OH HO
OAc
H H
CO2H H HO
H
OH CO2H OH
Cortisone Progesterone Cholesterol
H H H H H H
O O HO
H H H
O OH O H
O
b1026_Chapter-02.qxd
16
Steroids
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 17
Organic Synthesis via Examination of Selected Products
Steroids
17
Steroids-1 The steroids we will consider as targets for synthesis are shown here (Steroids-1). Cholesterol (and derivatives) is an important component of cell membranes. Gallstones are mainly cholesterol. It is one of the steroids most familiar to the layperson. Other steroids of much interest include the sex hormones (structurally related to progesterone) and the corticosteroids, represented here by cortisone. Most steroids share a common tetracyclic ring system, but are adorned differently in terms of oxidation state at various carbons. The three targets we will consider are only the tip of the iceberg. Fused ring systems with different ring juncture stereochemistry (cis vs trans), different sidechains, and different oxidation patterns are common. A few examples are shown here (see problems for some questions about fusidic and cholic acids).
b1026_Chapter-02.qxd
18
Woodward, R. B.; Sondheimer, F.; Taub, D.; Heusler, K.; McLamore, W. M. "Total Synthesis of Steroids", J. Am. Chem. Soc. 1952, 74, 4223. O
O
H H
H
B
H
HO
HO
H
HO Estrone
Equilenin
Cholesterol
1
9 10
HO
H 13
3
H 5
8
CO2Me
14
9
H
H
3
12
D H
9
O
H
16
O
2 Cyclohexene as Latent Cyclopentene Deconjugative Annulation
O
Directed Annulation H
H
O 8
H
H O
5 trans-decahydronaphthalene more stable than cis-decahydronaphthalene
O
O
4
11
O
10
1 Progesterone (oxidation state) Testosterone (sidechain degradation) Cholesterol (sidechain extension) Corticosteroids (∆9,11 oxidation)
C
20
H
1
16
H
5
O
6
17
17
11
O
4
Steroids-2
3
Page 18
11
20
H
10:54 AM
20 17
Miescher and Anner (1948)
Bachman, Cole and Wilds (1939)
Stucture by Winhaus and Wieland (1932)
Organic Synthesis via Examination of Selected Products
9 stereogenic centers = 512 stereoisomers
12/21/2010
A
H
D
b1026
H
A
C
D
Organic Synthesis via Examination of Selected Natural Products
H
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 19
Organic Synthesis via Examination of Selected Products
Steroids
19
Steroids-2 The first steroids to be prepared by total synthesis were equilenin and estrone (Steroids-2). These targets have progressively more complex stereochemistry. Equilenin has only two stereogenic centers and thus, four stereoisomers are possible. Estrone has four stereogenic centers (16 possible isomers).1 Cholesterol has 9 stereogenic centers and thus, in principle, 29 = 512 stereoisomers (including enantiomers) are possible. Although some of these are geometrically impossible given the bonding requirements of carbon, it is easy to see that cholesterol presents stereochemical problems that are at least an order of magnitude more complex that equilenin and estrone. The Woodward group (Harvard) reported the first total synthesis of cholesterol in the early 1950s. Their objective was not simply to prepare cholesterol, but to pass through intermediates that could be used to access a host of other steroids including progesterone and the corticosteroids. They eventually settled on tetracyclic compound 1 as a key target. At this point I will digress a bit. Throughout this book, I will undoubtedly interpret or present certain things in a manner that may seem inaccurate to the reader. Well, the reader may be correct. For example, I have no idea how Woodward and his students settled on 1 as a key intermediate. In hindsight, it is an excellent choice, but whether or not this was part of the initial design of the synthesis, or “evolved” as a key intermediate as research proceeded, I have no clue. The bottom line is that the reasons behind making a choice (for example of an intermediate or a reagent) are not always clear in print, and can even be inaccurate in print due to the tendency we all have as human beings to want to sound smarter than we are. Certainly more important, choice of a key intermediate can be critical if the synthetic objective is a family of molecules. “Can I get to the target structure(s) from this intermediate?” is a question that one asks regularly when designing (or developing) a synthetic strategy. In this case, 1 is very close to progesterone and testosterone. Reaching both targets calls for reduction chemistry at the C9-C11 and C16-C17 double bonds, and some D-ring modifications, including an oxidative degradation at C17 to reach testosterone. To reach cholesterol from 1, deconjugative reduction of the A-ring enone must be accomplished, as well as the aforementioned olefin reductions, and installation of the C17 side chain with control of stereochemistry at C20. One can imagine using the C9-C11 double bond as a functional handle for reaching cortisone from 1. Finally, one can imagine a number of tactics for accomplishing all of the aforementioned transformations. It is important to avoid
b1026_Chapter-02.qxd
12/21/2010 b1026
20
10:54 AM
Page 20
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
strategies that have to be scrapped if a single tactic fails. Having lots of options is desireable. Working backwards to develop a synthetic strategy is also common. It is called retrosynthetic analysis.2 As an example, one path to 1 might be through an intermediate of type 2. In a forward direction, hydrolysis of the acetal, cleavage of the resulting diol, aldol dehydration of the resulting dialdehyde, and adjustment of oxidation state at C20 would accomplish the necessary transformation. Of course this would have to be done in a manner that was compatible other functional groups. The details of this will emerge when we examine the actual synthesis. Working backwards again, the D-homosteroid nucleus of 2 [the term “D-homosteroid” is loosely derived from “homologous series” which generally means one more CH2] was to be prepared from tricyclic enone 3 via a deconjugative alkylation to establish the C1-C10 bond, followed by an aldol dehydration to afford the A-ring enone. Enone 3 was to be derived from 4 via an aldol dehydration, and 4 was to come from dienone 5, the first key intermediate in this synthetic strategy. There are several critical aspects to selection of 5 as an early intermediate. One is that cyclohexenes can always be converted to 1acylcyclopentenes via oxidative cleavage of the double bond, followed by an aldol-dehydration reaction sequence. Since a cyclohexene has a lower level of functionality than an acylcyclopentene (one functional group vs two functional groups), it is often useful to package an acylcyclopentene as a cyclohexene (or cyclohexene derivative such as an acetonide). This is part of the generally useful advice to “keep functionality at as low a level as possible throughout a synthesis”. Why? Problems of functional group compatibility associated with a given transformation are minimized. Another reason why 5 was chosen as a key intermediate is that, at the time, little was known about how to reliably prepare trans-fused hexahydroindans (the CD ring system of the steroids we have considered thus far). In fact, it was known that cis-hexahydroindan (the parent 6/5 fused ring hydrocarbon) was more stable than trans-hexahydroindan. On the other hand, it was known that trans-decalin (the parent 6/6 fused ring hydrocarbon) was more stable than cis-decalin. Thus, it was decided to prepare the 6/6 ring system, hoping that thermodynamics could be used to establish stereochemistry at the ring juncture, and knowing that the cyclohexene could serve as a latent 5-membered ring. Now let’s examine the synthesis.
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 21
Organic Synthesis via Examination of Selected Products
Steroids
21
Steroids-3 The synthesis began with a Diels-Alder reaction between 1,3-butadiene and quinone 6 to give cis-hexahydronaphthalene 7 (Steroids-3).3 The nature of the Diels-Alder reaction gave the cis-cycloadduct, but base-mediated epimerization via an intermediate enolate, provided trans-hexahydronaphthalene isomer 8. In spite of the aforementioned plan, however, it turned out there was little difference in thermodynamic stability between 7 and 8. Nonetheless, seeding a basic solution of 7 with 8, accomplished selective crystallization of the desired trans isomer in near quantitative yield (an application of LeChatelier’s Principle). Adjustment of oxidation states in the C-ring was accomplished by lithium aluminum hydride reduction to give an intermediate diol, followed by acid promoted enol ether hydrolysis and β-elimination of the C12 hydroxyl group to introduce the C11-C12 double bond. The C8 hydroxyl group was then removed via a reduction procedure that is now common, but in the 1950s represented new synthetic methodology. Note that the conversion of 8 to 5 involves only reduction chemistry. Reduction-oxidation chemistry permeates syntheses and thus, a general consideration of the topic is worthwhile. Let’s look in detail at the changes in oxidation state that occur during this transformation. One convenient way of comparing oxidation states at carbon is to count the number of bonds from carbon to oxygen. For example, methane (an alkane) has no bonds to oxygen, methanol (an alcohol) has one bond to oxygen, and formaldehyde (an aldehyde), formic acid (an acid) and carbon dioxide have 2–4 bonds to oxygen, respectively. Of course not all functional groups contain oxygen. But functional groups that can be converted to alcohols by substitution reactions (alkyl halides, sulfides, selenides, amines, azides, to mention a few) are really at the alcohol oxidation state. The same goes for functional groups at the ketone or aldehyde oxidation state (imines, oximes, hydrazones, thioacetals). Alkenes are a bit trickier. One way to look at alkenes is that hydration of an alkene (a reaction that does not involve redox chemistry) converts one carbon to the alkane oxidation state and the other to the alcohol oxidation state. Conversely, dehydration of an alcohol introduces an alkene (without redox chemistry). So in alkenes, the oxidation state at one carbon is “alkane” and at the other carbon “alcohol”. It is not always possible to decide which carbon one should regard at which oxidation state. That depends on substituents. Now let’s look at the conversion of 8 to 5. In 8, C8 and C12 are clearly at the ketone oxidation state. Carbon-9 is best classified as being at the ketone oxidation state, because the “natural” course of hydration of an enol ether converts it to a ketone (or aldehyde). Conversely, enol ethers are derivatives
b1026_Chapter-02.qxd
22
O
O
1. NaOH, dioxane
100 °C, 96 h
6
7
O 86%
2. Seed with transisomer; add HCl to neutralize
MeO 9
12 11 8
8
O 9 H HO 60% 9
H
O 8 90%
H 8
O
H 55-60% 11
Stereochemistry based on thermodynamic arguements. The "guess" turned out to be correct. This was about the same time that Barton published his seminal papers on conformatinal analysis of cyclic systems.
O
O
H O
HO
8
H 10
2. KOH, dioxane
O
94%
1. NaOMe, HCO2Et
11
2. H2SO4
O 9
8
H 5
45% from trans-cycloadduct
4
Notice the use of Claisen condensation, conjugate addition, and aldol dehydration reactions to construct the six-membered ring (annulation chemistry). We will return to possible reasons for choice of this strategy for construction of the B-ring at a later date. Note that the same fundamental strategy is used for construction of the A-ring.
Steroids-3
Page 22
12
1.t-BuOK, t-BuOH
10:54 AM
1. Ac2O 2. Zn in xylene or hot AcOH
Organic Synthesis via Examination of Selected Products
An interesting application of LeChatelier's Principle. Selective crystallization provides the trans-isomer. Studies suggest that cis and trans isomers are equally stable.
12/21/2010
H
12
b1026
O
MeO
11
Organic Synthesis via Examination of Selected Natural Products
PhH, hydroquinone
MeO
1. LiAlH4 2. dioxane, aq. H2SO4
O
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 23
Organic Synthesis via Examination of Selected Products
Steroids
23
of ketones (or aldehydes). This leaves C11 at the alkane oxidation state. In 5, the oxidation states at C9 and C8 are “ketone” and “alkane”, respectively. The substituent at C9 (a carbonyl group) renders C12 electron-deficient relative to C11. Thus it is reasonable to regard C12 to be at the “alcohol” oxidation state and C11 at the “alkane” oxidation state. Note that hydration of conjugated enones normally occurs to provide β-hydroxy ketones. So, in the conversion of 8 to 5, the oxidation state at C12 changes from “ketone” to “alcohol” and the oxidation state at C8 changes from “ketone” to “alkane”. This transformation requires “one reduction” at C12, “two reductions” at C8, and no redox chemistry at either C9 or C11. Whereas this analysis is “after the fact” in this case, a useful question to ask as one plans a functional group transformation (or series of transformations) is: What oxidation state changes occur? This can reveal whether redox chemistry is needed or not, and guide the practitioner in choice of reagents (reducing agents vs oxidizing agents). Whereas we have only considered oxidation states at carbon here, similar analyses can be applied to oxidation states at heteroatoms. Let’s continue with the synthesis. The conversion of enone 5 to dienone 11 is an example of an “annulation” or “annelation” reaction.4 The words mean the same thing: to build one ring onto another. The only difference is that annulation is derived from the Greek “annulus”, and annelation is derived from the French “anneler”.5 The type of reaction used here is commonly called a “Robinson Annelation”. It involves conjugate addition of the enolate of α-formylketone 10 to ethyl vinyl ketone (a Michael Reaction),6 followed by an aldol-dehydration sequence. This added what will become the B-ring of the steroids onto a pre-existing C-ring. Why was 5 converted to 10 prior to conducting the annulation? Both 5 and ethyl vinyl ketone (the annulation reagent) are ketones of comparable acidity. Conversion of 5 to 10 “activates” C8 relative to the annulation reagent, defining 10 as the nucleophile and ethyl vinyl ketone the electrophile in the Michael reaction. Deformylation of the intermediate non-enolizable 1,3dicarbonyl compound formed in the Michael addition, gave conjugate adduct 4, which continued on to 11 under the basic reaction conditions. It is notable that the Woodward group had only limited analytical tools available at the time this synthesis was conducted. Nowadays one would use NMR methods, or perhaps crystallography, to assign stereochemistry. In this case, it was presumed that C8 would be under thermodynamic control and this alone was the basis for assignment of stereochemistry at the newly formed stereogenic center. The “guess” turned out to be correct. The next portion of the plan called for removal of the C11-C12 double bond and deconjugative annulation of the A-ring to the pre-existing B-ring.
b1026_Chapter-02.qxd
24
12
O
O 9
H
HO
8
H 5
10
45% from trans-cycloadduct
1. OsO4, ether 2. mannitol, KOH, H2O OH H OH H O
O
CuSO4
H
12
57% + isomer (stereochemistry unknown)
H O
1. HCO2Et NaOMe, PhH
O H
H O
acetone
O
H2, Pd PhH
O 6
13
O
H 3
O
2. aq. KH2PO4 O HO
83%
Steroids-3 (Continued )
6
H 14
Page 24
Notice the use of Claisen condensation, conjugate addition, and aldol dehydration reactions to construct the six-membered ring (annulation chemistry). We will return to possible reasons for choice of this strategy for construction of the B-ring at a later date. Note that the same fundamental strategy is used for construction of the A-ring.
10:54 AM
94% 4
Organic Synthesis via Examination of Selected Products
2. KOH, dioxane
O
Stereochemistry based on thermodynamic arguements. The "guess" turned out to be correct. This was about the same time that Barton published his seminal papers on conformatinal analysis of cyclic systems.
2. H2SO4
8
O
H
b1026
O
O
H 55-60% 11
11
12/21/2010
H 8
1. NaOMe, HCO2Et
Organic Synthesis via Examination of Selected Natural Products
1.t-BuOK, t-BuOH
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 25
Organic Synthesis via Examination of Selected Products
Steroids
25
First, however, the D- and C-ring double bonds had to be differentiated. This was accomplished by vicinal dihydroxylation of the D-ring olefin using osmium tetroxide (stoichiometric amounts) followed by protection of the resulting diol as acetonide 13 (an acetal).7 The vicinal dihydroxylation proceeded with modest stereoselectivity and good regioselectivity, presumably due to steric effects. The stereochemistry of the diol was never established, but this did not thwart completion of the synthesis. A practical point that I want to make here is that although both diastereomers of diol 12 could have been carried through the synthesis in principle, this was not done in practice. Only the major isomer was used. Although one is sometimes tempted to say that “stereocontrol will not matter because the stereochemistry will be destroyed by the time I reach my target”, this is generally not a good approach to take. In reality, characterization of mixtures is difficult and most successful syntheses do not carry mixtures through multiple steps. This does not mean that one should not attempt a reaction if there is some doubt about its stereochemical course. It is good, however, to have options available so the required control can be achieved. In reality, however, separations of stereoisomer mixtures are frequently needed to execute a synthesis of a complex molecule. Continuing with the synthesis, hydrogenation of the sterically less hindered of the remaining double bonds proceeded smoothly to provide 3. Annulation of the A-ring onto the B-ring first required that C6 be “blocked”. This was accomplished by conversion of 3 to vinylogous amide 15 (Steroids-4) via intermediate C6-hydroxymethylene ketone 14.
b1026_Chapter-02.qxd
26
O
O
N-methylaniline
H
H
MeOH, ∆
O
14
HO
O
10
H 16
O 15
N Ph
33% (46% of C 10 isomer)
CHO 17
O
1. HIO4 , dioxane
H
O
MeMgI, ether
H 15
2.
AcO
O H
65%
H
O
N
18
O
H O
O
2
H
17
58%
(catalyst)
1. H2 , PtO2 , AcOH 1. Na 2Cr2 O7 , AcOH 2. CH 2N 2
2. CrO3 , AcOH CO2 Me
CO2 Me NaBH4
H H O 58%
1
Compared with material prepared from natural source.
Oppenhauer Oxidation
CO2 Me
H
H H
HO
H 19
Reduction gives mixture. β-Isomer resolved with digitonin.
Steroids-4
O
H 20
H
This marks the end of de novo synthesis. This material was prepared from cholesteryl acetate in 14% overall yield. This is an example of a relay synthesis .
Page 26
H
H O
10:54 AM
Ac 2O, NaOAc, ∆
Blocking Groups: Birch, Robinson (1944)
Organic Synthesis via Examination of Selected Products
80% from dienone
Note cyclohexene as latent 5-membered ring. We will return to this later.
b1026
O H
O
2. KOH, H2 O, ∆
Organic Synthesis via Examination of Selected Natural Products
H
H
O H
HO 2C
12/21/2010
O
1. CH 2=CHCN Triton B t-BuOH, PhH H2 O
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 27
Organic Synthesis via Examination of Selected Products
Steroids
27
Steroids-4 Treatment of 15 with base and acrylonitrile, followed by hydrolysis of the intermediate nitrile, gave conjugate adduct 16. This reaction was complicated by formation of the C10 epimer of 16 as the major product. Woodward indicates that little selectivity was expected in this step. We will see (Steroids-9) that a modification of this approach that solved this problem was eventually developed by others. Annulation of the A-ring was completed by treating 16 with acetic anhydride, followed by opening of the resulting enol ester 17 with methyl magnesium iodide to give an intermediate 1,5-diketone. Aldol-dehydration occurred under the basic reaction conditions to provide tetracycle 2. Keeping in mind that 1 was projected as a key intermediate for the synthesis of a variety of steroids, the Woodward group next turned to contraction of the D-ring to the required acyl cyclopentene. The diol was liberated and cleaved to a dialdehyde using periodic acid. The dialdehyde was then subjected to piperidinium ion mediated intramolecular aldol-dehydration to provide 18. The aldol-dehydration also provided the regioisomeric enal as a minor product. The regioselectivity of this reaction was attributed to the apparent steric accessibility of the C17 methylene in the intermediate dialdehyde relative to the C15 methylene. The synthesis of 1 was completed in a straightforward manner. This synthesis provided racemic 1. To conclusively establish its structure, 1 was reduced with sodium borohydride to provide 19 (this reaction gave a separable mixture of diastereomers), which was resolved and oxidized to return a single enantiomer of 1. This material was identical in all respects to a sample of material prepared by degradation of a corticosteroid. The Woodward group conducted most of their remaining work with material prepared by degradation of cholesteryl acetate to tetracycle 20, which was also prepared from synthetic 1 as shown in (Steroids-4). A synthesis of a natural product from an intermediate prepared by degrading a natural product, prepared in turn by total synthesis, is known as a relay synthesis. This is a less common practice nowadays than in the early days of natural products synthesis, but it is not unheard of even today. We will see this practice again.
b1026_Chapter-02.qxd
28
Endgame for Several Steroids CO2 Me
CO2 Me
H AcO
H
others
H
H
H
H
O
H 22 androsterone
23 testosterone
20
CO2 Me
H H
2. Ac2 O, pyridine
H
H
AcO 3 24
17
H
H
1. SOCl2 2. Me2 Cd
H 2, PtO2
3. RMgBr 4. Ac2 O, ∆
H
H HO
30
H
H 27
Ac 2O
H
MeOH-H2 O
H AcO
4
AcO
1. Hydrolysis 2. CrO3 3. Br2 , pyridine NaBH4
H
reduction not selective
Cholestanol Acetate
β -alcohol is major diastereomer (axial attack)
H
H
3
H 26
25
H
4
H H
3
H
H
O 29
Cholesterol Standard Deconjugation Move
Steroids-5
28
Page 28
1. KOH, MeOH
H
HO
CO2 H
10:54 AM
Barbier-Wieland Degradation
EtOH
Organic Synthesis via Examination of Selected Products
NaBH4
H HO
H 21
20
H
3. O3 ; Me2 S
H
OH
17
b1026
O
H
2. Ac2 O
H
17
12/21/2010
H H
1. H2 , catalyst
Organic Synthesis via Examination of Selected Natural Products
17
O 1. PhMgBr 2. H3 O+
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 29
Organic Synthesis via Examination of Selected Products
Steroids
29
Steroids-5 The Woodward group converted this key intermediate (20) into a variety of natural steroids including progesterone (Steroids-1). Key intermediate 20 was also converted to androsterone (22) by reduction of the C3 ketone followed by Barbier-Wieland degradation of the D-ring ester to a C17 ketone.8 Androsterone had previously been converted to testosterone (23) by others. The cholesterol synthesis was accomplished by initial reduction of 20 with sodium borohydride to provide 24. Ester hydrolysis and esterification of the resulting C3 alcohol gave 25. The C17 acid was converted to a mixture of olefins 26, followed by unselective reduction of the olefins to provide cholestanyl acetate (27) after separation of the C20 diastereomers. The lack of diastereoselectivity in this transformation begs the question: How can stereochemistry at an acyclic stereogenic center be controlled relative to other stereogenic centers? We will return to this question later in detail when we consider syntheses of a natural product called juvabione, but for now let us continue. Hydrolysis of 27 provided cholestanol, which had previously been converted to cholesterol by other research groups. This conversion involved the preparation of 28 via bromination-dehydrobromination of a C3 ketone. Enone 28 was converted to cholesterol by what is now a standard deconjugation move in terpenoid synthesis. Dienol acetate 29 was reduced with sodium borohydride, the intermediate enolate protonated at C4, and the resulting β,γ-unsaturated ketone was then reduced by the NaBH4 (pseudoaxial delivery of hydride) to provide the homoallylic alcohol substructure of cholesterol (30).
b1026_Chapter-02.qxd
30
Corticosteroids
5
H
2. NaOMe, MeOH HO
AcO
4
H
1
O
12
H H
3
O
cortisone H
O H H
34 OAc O
36
OAc 21
O
CN
1. Pyr, Ac2O 2. KCN 3. POCl3
H
H
H
3
H
H
37
Steroids-6
H
3
O H
17
O
1. OsO4 2. CrO3
O
HO
H
1. Br2 2. 2,4-DNP 3. pyruvic acid 4. HCl, H2O
H
OH 20
33
CO2Me
4
H 38
H
OH
Page 30
HBr Br
3. Ac2O, pyr 4. CrO3, AcOH
10:54 AM
H 35
Organic Synthesis via Examination of Selected Products
1. Zn, AcOH 2. NaBH4
1. SOCl2 2. CH2N2 3. AcOH 4. NaOH
HO
32
Cr+6 Oxidation
17
H
H
3
H
CO2H
11
H
AcO
H
31
At this point the synthesis intersects with others on the production of cortisone
O
H OH
b1026
3
O
CO2Me
11
1. PhCO3H
H 9
3. Ac2O, pyridine
H
HO
12/21/2010
H
CO2Me 11
Organic Synthesis via Examination of Selected Natural Products
CO2Me 17 1. H2, Pd/SrCO3 2. NaBH4 16
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 31
Organic Synthesis via Examination of Selected Products
Steroids
31
Steroids-6 As a final target, we will examine the conversion of intermediate 1 to cortisone (Steroids-1) as outlined in Steroids-6. The critical transformation was conversion of 1 to 31, which had previously been “connected” with cortisone by a series of transformations reported from other laboratories. Hydrogenation of 1 resulted in reduction of the C4-C5 and C16-C17 olefins to provide material with the required C17 stereochemistry and a mixture of cis- and trans-fused AB-ring systems. Sodium borohydride reduction of the C3 ketone and acetylation of the resulting alcohol provided 31. The overall yield of this transformation was approximately 15% due to lack of stereoselectivity in the hydrogenation. The rest of the cortisone synthesis involves a series of transformations reported by other groups over the period extending from approximately 1940–1950, but we will review these here for the purpose of comparison when we consider other approaches to cortisone. The C11 oxygen, that is so characteristic of the corticosteroids, was introduced by epoxidation of the sterically most accessible face of the olefin followed by ring opening to provide 32. Oxidation of the secondary alcohols gave hemiacetal 33. Treatment of 33 with HBr gave dione 34, a transformation that is interesting from the standpoint of mechanism. Reduction of the C12 bromide (recall the conversion of 8 to 9) was followed by reduction of the ketones, esterification of the sterically most accessible C3 alcohol, and oxidation of the C11 alcohol to provide 35. The C17 side chain was then introduced. Acid 35 was first converted to 36 using a four-reaction sequence. The primary alcohol was acylated, the C20 ketone was converted to a cyanohydrin, and dehydration of the C20 alcohol then provided 37. Vicinal dihydroxylation of the unsaturated nitrile, followed by elimination of HCN to reveal the C20 ketone, and oxidation of the C3 secondary alcohol, gave 38. The A-ring stereochemistry and oxidation state was then adjusted by initial bromination at C4, dehydrohalogenation to provide the 2,4-DNP of the A-ring enone, and subsequent hydrolysis of the 2,4-DNP and C21 acetate to provide cortisone (Steroids-1).
b1026_Chapter-02.qxd
32
Wieland, P.; Ueberwasser, H.; Anner, G.; Miescher, K. "Uber die Herstellung des ∆8,14-1,7-Dioxo-8,11-dimethyldodecahydrophenanthrens. Totalsynthetische Versuche in der Steroidreihe I", Helv. Chim. Acta 1953, 50, 376. Cl
PhH
Et3N
O
O
Et3N
(and cis isomer)
OH
O
O O H
or
H
or H
H
O O
H
NaBH4
O
O
O O 1. EtMgBr 2. HCl; then NaOH
O Wiland, P.; Anner, G.; Miescher, K. "Die sterische Verknupfung eines ∆8,14-1,7-Dioxo-8,11-dimethyldodecahydrophenanthrens mit den Steroiden. Totalsynthetische Versuche in der Steroidreihe II" Helv. Chim. Acta 1953, 50, 646.
O
H
O
O
NaOH O
NMe2
Wieland, P. Ueberwasser, H.; Anner, G.; Miescher, K. "Totalsynthese von D-Homo-steroiden. Totalsynthetische Versuche in Steroidreihe III" Helv. Chim. Acta 1953, 50, 1231.
O O
Ac2O O
H
O
H
HO2C
O O
1. Ac2O 2. MeMgI 3. NaOH
H H
H
H O
CO2H
Steroids-7
O
Page 32
(CH2OH)2, TsOH
1. (CH2OH)2, TsOH 2. Me2NMgI
10:54 AM
CO2Me
CO2Et
H
H
Organic Synthesis via Examination of Selected Products
O CO2Et
O
b1026
O
1. H2, Pd 2. NaOH 3. CH2N2
12/21/2010
O
O
O EtO2C
Organic Synthesis via Examination of Selected Natural Products
O
O
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 33
Organic Synthesis via Examination of Selected Products
Steroids
33
Steroids-7 I picked the Woodward approach to steroids as the first synthesis to discuss because I want to revisit strategies that appear in that work within the context of other targets. But it is important to recognize that a number of other groups were working in the area, using conceptually similar approaches (in part). I will mention a few of these approaches here. The A- and B-rings in the synthesis we have just considered were introduced by annulation reactions. This type of chemistry was introduced by Robinson and ultimately used in his approach to steroids. It was also explored in detail by Wieland during the course of his studies directed toward the structure determination and synthesis of steroids. Some reactions reported by the Wieland group appear in Steroids-7 without much comment. These approaches resemble the Woodward approach in that the steroid ring system is assembled starting with the D-ring and working toward the A-ring. Other similarities involve use of a deconjugative Michael addition to introduce the incipient A-ring, application of the keto acid to cyclohexenone transformation to construct the A-ring, and synthesis of a D-homosteroid that could be ring contracted to provide the required 5-membered ring. Finally, if this work were being done today, additional tactics would be available that might provide better control of stereochemistry at the CD-ring juncture.
O
O 39
SOCl2
OMe CH2=CH2, AlCl3
Cl O
O
O
O
40
OMe
Cl O
41
42
O
OH
OH
HCl O HO2C
O
Rhizopus arrhizus
O O
45
OAc
OAc Ac2O
O
H
1.
OAc
THF O O
H
2. KOH, MeOH
O
H 10
H
O
O 48 minor
O
H
+ O
NaOAc O
47
O
MgBr
OH
H
O 43
H2, Pd/C
O HO2C
OMe
pyridine toluene, ∆
O CO2Me 44
O CO2Me 46
Et3N
49 major
Steroids-8
50
Page 34
O
10:54 AM
OMe
HO
12/21/2010
MeOH
Organic Synthesis via Examination of Selected Products
O
b1026
Strategy: (1) Build from D-ring toward the A-ring using annulation chemistry. (2) Obtain desired enantiomer by enzymatic reduction of a pro-chiral diketone. (3) Use A-ring enone rather than AB-cis system throughout introduction of C11 ketone of corticosteroids. (4) Use C17 ketone as handle for introduction of D-ring sidechain.
Organic Synthesis via Examination of Selected Natural Products
For some related chemistry see: Velluz, L.; Nomine, G.; Mathieu, J.; Toromanoff, E.; Bertin, D.; Tessier, J.; Pierdet, A. "Sur l'acces stereospecifique, par synthese totale, a la serie 19-nor-steroide. La 19-nor-testosterone de synthesis" Comptes Rend. Acad. Sci.1960, 250, 1084. The following material was abstracted from "Selected Organic Syntheses" by Ian Fleming with permission from Wiley-Blackwell. The synthesis relies in part on annulation chemistry developed by Wieland and Miescher.
O
b1026_Chapter-02.qxd
34
An Industrial Scale Synthesis of Steroids
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 35
Organic Synthesis via Examination of Selected Products
Steroids
35
Steroids-8 An industrial scale approach to steroids was developed by both Merck (US)9 and Ciba (Swiss) during the 1950s. The Ciba approach is summarized in Steroids 8–10. The critical aspects of the strategy were: (1) to build from the D-ring toward the A-ring using annulation chemistry, (2) to obtain the desired enantiomer by enzymatic reduction of a pro-chiral diketone, (3) to use an A-ring enone rather than an AB-cis system throughout introduction of the C11 ketone of the corticosteroids, and (4) to use a C17 ketone as a handle for introduction of the D-ring sidechain. The synthesis began with the preparation of annulation reagent 43 from glutaric anhydride (39). Conjugate addition of 2-methylcyclopentan-1,3dione to 43 provided trione 44. This trione has a plane of symmetry bisecting the 1,3-dicarbonyl substructure. The two carbonyl groups are identical and reduction of one ketone with an achiral reducing agent provides a racemic mixture of 45 (an equal mixture of enantiomers). But use of an enantiopure reducing agent can lead to an unequal mixture of enantiomers. In this case, reductases (enzymes) produced by the bacterium Rhizopus arrhizus led to enantioselective reduction of 44 to provide 45. Acid promoted aldol-dehydration of 45 was accompanied by hydrolysis of the methyl ester to provide hexahydroindan 46. Catalytic hydrogenation gave 47. The hydrogenation of enones of type 46 to give either a cis- or trans-fused ring system is known. It is a catalyst and substrate dependent process and, in practice, one must evaluate a number of tactics to determine how to best accomplish this transformation with the desired stereoselectivity.10 Continuing with the synthesis, the B-ring was assembled using the enol lactone annulation procedure we saw in the Woodward synthesis. The difference between 50 and the BCDring system intermediate in the Woodward synthesis (see 3 in Steroids-3), is that the subsequent deconjugative alkylation reaction must install the C10 methyl group rather than the remaining carbons of the A-ring. There is a small general lesson that can be discussed here. If you have to introduce two different groups to the same carbon during the course of a synthesis, you have to pick an order-of-introduction. If introducing the groups in one order shows a stereochemical preference, introducing the groups in the opposite order can often produce the other stereochemical result. In the Woodward synthesis, installation of the C10 methyl group preceeds introduction of the A-ring (via a deconjugative conjugate addition). In the Ciba approach, installation of the C10 methyl group follows introduction of the A-ring (precursor group). Whereas the “methyl first” approach gave a mixture of stereoisomers, the “methyl second” approach gave the desired stereochemical result.
b1026_Chapter-02.qxd
36
OCOPh
OH O
O
17
H
1. AcOH, H
H
H
H
O
O 55
54
testosterone
estradiol
Synthesis of Annulation Reagent for Introduction of AB-Ring System
HBr
O
O
H H O
OCOPh HO
OH
O
O
H
TsOH
H
MgBr O O
1. RONa, CH3I 2. H3O+
OCOPh 11
O
H 10
9
O
O 52
1. (CH2OH)2 TsOH 2. Mg
O
O
OCOPh O
Br
57
Steroids-9
58
H
Page 36
O
O EtO2C
10:54 AM
H
H
2. HO
HO
H
12/21/2010
H
Organic Synthesis via Examination of Selected Products
H
19
b1026
OH
OCOPh 1. Pd/C/EtOH
Organic Synthesis via Examination of Selected Natural Products
56
PhMe
OH
H
H
O
52
O
H
2. H+
H O
51
H
OCOPh 1. H2, Pd/C
17
H
2. PhCOCl
H O
O
+
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 37
Organic Synthesis via Examination of Selected Products
Steroids
37
Steroids-9 The A-ring was introduced by hydrolysis of the acetate to provide D-ring alcohol 51. Hydrolysis of the acetal and esterification of the C17 alcohol gave benzoate 52. As we shall see, 52 proved to be a key intermediate in the synthesis of corticosteroids. But first let’s take a little side trip to some simpler steroids. Treatment of 52 with t-amyloxide gave dienone 54 via an aldoldehydration reaction. This steroid-like molecule is actually at the phenol oxidation state, and thus, Pd-mediated isomerization of the dienone followed by benzoate hydrolysis gave estradiol. On the other hand, catalytic hydrogenation of enone 52 followed by an acid-promoted aldol-dehydration gave enone 56. This is the benzoate of 19-nortestosterone. Thus, this chemistry provided an efficient route to a number of steroid analogs. Continuing with the path to cortisone, 52 was converted to acetal 57 (also available from 51). Deconjugative alkylation of 57 gave 58 with the correct stereochemistry at C10 and the C9-C11 double bond in place for use in corticosteroid synthesis. It appears that the stereochemical course of this alkylation involves “axial” attack of the electrophile (methyl iodide) on the intermediate dienolate. One might wonder why this stereochemistry was not also observed in Woodward’s conversion of 15 to 16. One possible explanation is that the conjugate addition might be reversible, and thus the reaction is controlled by thermodynamics, whereas the alkylation of 57 should clearly be contolled by kinetics. It is also possible that the blocking group in 15 exerts some influence that is not clearly understood. Finally, I note that it is possible that these reactions are not totally selective, and that information remains a trade secret.
b1026_Chapter-02.qxd
38
OCOPh O
17
H
O
O HO
1. NaOH 2. CrO3
17
H
H
HOBr
Br
H
1. HBr 2. KOAc OAc
O O
O
H
OH
OsO4
H H
H
Ph-I(OAc)2
OAc
H
26 step synthesis of cortisone. Note the minimal use of protecting groups and the selectivity in carbonyl addition reactions.
H
O
O 63
64 cortisone
For a review with a good discussion of reactivity principles for 6-membered rings see: Velluz, L.; Walls, J.; Nomine, G. Angew Chem. Int. Ed. 1965, 4, 181−200
Steroids-10
Page 38
60
61
62
10:54 AM
H O
12/21/2010
17
H
Organic Synthesis via Examination of Selected Products
O
2. H2/Pd
H
O
60
O
1. t-amyl-O HC CH
17
H
1. CrO3 2. Zn, AcOH
b1026
59
HO
H
O
58
O
H
Organic Synthesis via Examination of Selected Natural Products
O
H
9
H
O
11
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 39
Organic Synthesis via Examination of Selected Products
Steroids
39
Steroids-10 The synthesis of cortisone continues with an aldol-dehydration reaction, ester hydrolysis, and oxidation of the resulting C17 alcohol to provide 59. Treatment of 59 with HOBr results in addition to the most electrophilic of the two double bonds. Bromonium ion formation from the most accessible face of the olefin (recall the stereochemistry of epoxidation of 31) followed by anti-periplanar opening of the intermediate with water afforded 60. Oxidation of the C11 alcohol and reductive removal of the C9 bromide gave 61. Addition of acetylide to the C17 ketone followed by partial hydrogenation of the alkyne gave allylic alcohol 62. The stereochemical course of the acetylide addition is typical for C17 steroidal ketones. Allylic rearrangement accompanied treatment of 63 with HBr and the resulting primary allylic bromide was converted to acetate 63. Finally, treatment of 63 with osmium tetroxide and iodobenzene diacetate resulted in vicinal dihydroxylation of the most electrophilic olefin, and oxidation of the intermediate secondary alcohol, to provide cortisone acetate. Two remarkable features of this synthesis are the minimal use of protecting groups and the use of low-tech chemistry.
b1026_Chapter-02.qxd
40
Biosynthesis of Steroids
R R O
cyclize
HO
H H
squalene oxide (R = prenyl group)
Steroids-11
Page 40
Labelling studies show that squalene is derived from acetic acid. 18 Carbons come from methyl groups and 12 carbonscome from carboxyl carbons. Whereas squalene has 30 carbons, cholesterol has only 27 carbons, and other steroids have even fewer carbons. Therefore “degradation” must accompany (or follow) cyclization. Robinson was the first to propose a manner in which squalene might cyclize to provide the tetracyclic ring system of the steroids. This was followed by an alternate hypothesis by Bloch and Woodward. The two proposals suggest different labelling patterns in tetracyclic products. For example, if one thinks about cholesterol as the product, the two proposals suggest that the “black” methyl groups would have to be lost. These are different for the two proposals. Experiments eventually disproved the Robinson hypothesis and supported the Bloch-Woodward hypothesis. For an early experiment see Woodward, R. B.; Bloch, K. “The Cyclization of Squalene in Cholesterol Synthesis” J. Am. Chem. Soc.1953, 75, 2023-2024. The following (abbreviated) pathway from squalene oxide to lanosterol is now standard in undergraduate textbooks.
10:54 AM
Woodward and Bloch (1953)
12/21/2010
Sir Robert Robinson (1934)
Organic Synthesis via Examination of Selected Products
squalene
b1026
tail-to-head tail-to-tail
Organic Synthesis via Examination of Selected Natural Products
tail-to-head
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 41
Organic Synthesis via Examination of Selected Products
Steroids
41
Steroids-11 Natural products syntheses of numerous compounds have been inspired by nature. In other words, the pathway taken by nature to a given natural product often suggests a synthetic strategy that one might follow in the laboratory. Such syntheses are sometimes refered to as “biomimetic syntheses” and numerous examples permeate the literature. The notion of mimicing nature is quite old, but it is time-tested, often used, and still always stimulating to think about this approach to designing a laboratory synthesis of a new natural product. There are at least two fundamental approaches that have been used to design biomimetic syntheses: (1) determine the biosynthetic pathway and mimic it as best possible and (2) imagine the biosynthetic pathway and develop the “biomimetic synthesis”. The second approach is frequently followed because it is a lot of work to determine a biosynthetic pathway. There are a lot of well-established biosynthetic transformations (much like wellestablished laboratory reactions), however, and one can frequently propose a reasonable biosynthesis. Steroids, and the terpenoids from which they are biosynthetically derived, were among the first natural products to be tackled in the laboratory via a biomimetic approach. Lanosterol, a triterpene (30 carbons), is the tetracyclic precursor of steroids such as cholesterol. Squalene is the biosynthetic precursor of lanosterol. Labelling studies show that squalene is derived from acetic acid, as are all terpenoids. Eighteen of the 30 carbons of squalene come from methyl groups (of acetic acid) and 12 carbons come from carboxyl carbons. Whereas squalene has 30 carbons, cholesterol has only 27 carbons, and other steroids have even fewer carbons. Therefore “degradation” must accompany (or follow) the cyclization of squalene to the tetracyclic triterpenes. Sir Robert Robinson was the first to propose a manner in which squalene might cyclize to provide the tetracyclic ring system of the steroids.11 This was followed by an alternate hypothesis by Bloch and Woodward. The two proposals suggest different labelling patterns for tetracyclic products. For example, if one thinks about cholesterol as the product, the two proposals suggest that the “black” methyl groups would have to be lost (Steroids-11). These are different for the two proposals. Experiments eventually disproved the Robinson hypothesis and supported the Bloch-Woodward hypothesis (stereochemical details added by others). The abbreviated pathway from squalene to lanosterol is now standard in undergraduate textbooks (see Steroids-11 and Steroids-12).
b1026_Chapter-02.qxd
42
R
R H
HO
H R
HO
H
H
HO
H B
A H
Stork, G.; Burgstahler, A. W. " The Stereochemistry of Polyene Cyclization" J. Am. Chem. Soc. 1955, 77, 5068. Eschenmoser, A.; Ruzicka, L.; Jeger, O.; Arigoni, D. "On Triterpenes: A Stereochemical Interpretation of the Biogenetic Isoprene Rule for the Triterpenes" Helv. Chim. Acta 1955, 38, 1890-1904 (original in German). For an English translation and an interested update see Eschenmoser, A.; Arigoni, D. " Revisited after 50 Years: The Stereochemical Interpretation of the Biogenetic Isoprene Rule for the Triterpenes " Helv. Chim. Acta 2005, 88, 3011−3050.
Steroids-12
Page 42
H
H
10:54 AM
This biosynthetic pathway led a number of groups to suggest that polyolefin cyclizations might be used to prepare fused carbocycles including steroids. Of course the biosynthesis is mediated by enzymes that may predispose squalene oxide (or other isoprenoids) for cyclizations. Stork and Eschenmoser hypothesized that the stereoselective synthesis of terpenoids and steroids via polyolefin cyclizations should be possible if such cyclizations proceeded via synchronous processes. In a seminal paper Stork and Burgstahler said " It is the purpose of this paper to draw attention to the important conclusion that concert ed cyclization to structurally identical hydronaphthalenes will result in a tr ans system if the bicyclic precursor is an open chain triene convertible to a cation of type B and a cis product if the precursor is monocyclic (type A). If the reactions are not concerted, mixtures of products will result".
Organic Synthesis via Examination of Selected Products
Stork-Eschenmoser Hypothesis
b1026
lanosterol
12/21/2010
H
Organic Synthesis via Examination of Selected Natural Products
H H
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 43
Organic Synthesis via Examination of Selected Products
Steroids
43
Steroids-12 This biosynthetic pathway to the steroids led a number of groups to suggest that polyolefin cyclizations might be used to prepare fused carbocycles including steroids. Of course the biosynthesis is mediated by enzymes that may predispose squalene oxide (or other isoprenoids) for cyclizations. In 1955, Stork and Eschenmoser independently hypothesized that the stereoselective synthesis of terpenoids and steroids via polyolefin cyclizations should be possible if such cyclizations proceeded via synchronous processes. There were many other players in the game and I will not attempt to tell the story in any detail. Suffice it to say, in a seminal paper, Stork and Burgstahler (then at Harvard University) said “It is the purpose of this paper to draw attention to the important conclusion that concerted cyclization to structurally identical hydronaphthalenes will result in a trans system if the bicyclic precursor is an open chain triene convertible to a cation of type B, and a cis product if the precursor is monocyclic (type A). If the reactions are not concerted, mixtures of products will result”. At the same time, Eschenmoser, Ruzicka, Jeger and Arigoni (ETH-Switzerland) made comparable suggestions about the steric course of acid-catalyzed cyclizations in their landmark paper that discussed the biogenetic origins of triterpenes from polyolefins (squalene). This is a fascinating story that has been elegantly described by Eschenmoser and Arigoni on the 50th anniversary of these papers. What I will present next is an attempt by one group (that of William S. Johnson at Wisconsin and Stanford) to translate into practice the notion that one might make steroids in the lab by polyolefin cyclizations. During the course of the next few pages we will see how an idea can evolve from simple beginnings to a high level of complexity. Whereas the syntheses we will ultimately see do not follow the precise pathway nature takes to the steroids, they do revolve around stereoselective polyolefin cyclizations, and in that manner, can be called “biomimetic syntheses”.12
b1026_Chapter-02.qxd
44
Polyolefin Cyclization Route to Steroid Total Synthesis
S O
O 65
H
1. HCO2H (80%) 2. saponification
10
1
1
ONs
trace
H 68 2.9%
5.4%
66
H
OH
10
OH 1.6% H
OH
10
HO 14%
OH Only trans-decalins
1
1
H 69 2.2%
67 51%
70
H 6.7%
Johnson, W. S.; Lunn, W. H.; Fitzi, K. "Cationic Cyclizations Involving Olefinic Bonds. IV. The Butenylcyclohexenol System" J. Am. Chem. Soc. 1964, 86, 1972. H anhydrous HCO2H
ONs
H
OH +
OH +
80% of single alcohol isolated (92:6:2 by GC)
Steroids-13
dienes
Hint of a new cyclization initiation group
Page 44
10
10:54 AM
OH
12/21/2010
O
1. HCO2H 2. saponification
Organic Synthesis via Examination of Selected Products
NO2
b1026
Johnson, W. S.; Bailey, D. M.; Owyang, R.; Bell. R. A.; Jaques, B.; Crandall, J. K. "Cationic Cyclizations Involving Olefinic Bonds. II. Solvolysis of 5-Hexenyl and trans-5,9-Decadienyl p-Nitrobenzenesulfonates" J. Am. Chem. Soc. 1964, 86, 1959.
Organic Synthesis via Examination of Selected Natural Products
Many groups have contributed to this research area, but the most extensive contributions come from the group of William S. Johnson (Wisconsin and Stanford) and we will focus on those contributions. His research had modest beginnings and it is worth a look at the "evolution" of his approach to steroids.
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 45
Organic Synthesis via Examination of Selected Products
Steroids
45
Steroids-13 The Johnson group began by examing the solvolysis of 5-hexenyl nosylate (65) in formic acid. This reaction gave only small amounts of cyclohexanol and cyclohexene, but illustrated that that cyclization was possible. Solvolysis of diene 66 gave cyclohexanol 67 as the major product. This product results from anti-addition of the electrophilic carbon (C1) and oxygen nucleophile (formate) across the trans-olefin. Decalins 68–70 were produced in low yields, but it was notable that only trans-fused decalins were produced. This observation was consistent with an anti-addition of electrophilic carbon (C1) and nucleophilic carbon (C10) across the cis-olefin (see B in Steroids-12).
b1026_Chapter-02.qxd
46
Johnson, W. S.; Crandall, J. K. "Cationic Cyclizations Involving Olefinic Bonds. V. Solvolysis of cis-5,9-Decadienyl p-Nitrobenzenesulfonate" J. Am. Chem. Soc. 1964, 86, 2085.
H 69 5%
H 70 14%
HO 10%
3% of several other products, but no cis-bicyclics
10 6 1
ONs
0.02 M in anhydrous HCO2H containing 0.04 M pyridine
5 1
1h, 75 oC 3%
HO 8%
71
6
72 56%
OH
H
OH 10
5 1
10
OH
6
16%
H 73 13% of inseparable mixture
is not a common intermediate
Johnson, W. S.; Owyang. R. "Olefinic Cyclizations. VI. Formolysis of Some Branched-Chain Alkenyl p-Nitrobenzenesulfonates" J. Am. Chem. Soc. 1964, 86, 5593. Johnson, W. S.; Crandall, J. K. "Olefinic Cyclizations. VII. Formolysis of cis- and trans-5,9-Decadienyl p-Nitrobenzenesulfonate and of Some Isomeric Monocyclic Esters" J. Org. Chem. 1965, 30, 1785−1790 (full paper of earlier work).
Steroids-14
Page 46
5
10:54 AM
66
67 57%
OH
12/21/2010
3%
H
OH
Organic Synthesis via Examination of Selected Products
1h, 75 oC
NsO
H
b1026
OH
Organic Synthesis via Examination of Selected Natural Products
0.02 M in anhydrous HCO2H containing 0.04 M pyridine
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 47
Organic Synthesis via Examination of Selected Products
Steroids
47
Steroids-14 Finally, solvolysis of diene 71 gave cyclohexanol 72 and cis-decalins 73 as the most notable products (Steroids-14). Once again, the stereochemistry of these products suggest that anti-addition of electrophile and nucleophile across the cis-olefin was dictating the stereochemical course of cyclization reactions. This experiment also showed that the solvolyses of 66 and 71 do not pass through a common monocyclic intermediate.
b1026_Chapter-02.qxd
48
More Cation Cyclization Initiators
H
H
30%
OH
OH
35%
22%
1. HCO2H, rt, 5 min 2. saponify 3. H2, Pd/C
HO
OH
OH
O
[O]
+ 75
H 78
H 77
OH
H 79
H
H HO
76 major product
Steroids-15
Page 48
Alkene converted to diols under reaction conditions. Note that cyclization occurs at secondary center of allylic system.
10:54 AM
OH
74
12/21/2010
H 2. saponification
Organic Synthesis via Examination of Selected Products
OH
H
H
1. HCO2H, 23 oC, 5 min
b1026
HO
OH
Organic Synthesis via Examination of Selected Natural Products
Johnson, W. S.; Neustaedter, P. J.; Schmiegel, K. K. "Olefinic Cyclizations. VIII. The Butenylmethylcyclohexanol System" J. Am. Chem. Soc. 1965, 87, 5148−5157.
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 49
Organic Synthesis via Examination of Selected Products
Steroids
49
Steroids-15 The Johnson group also examined other cyclization initiators. For example the use of allylic alcohols gave high yields of monocyclization products (bottom of Steroids-13) via a presumably symmetrical allylic carbocation. Unsymmetrical carbocations derived from allylic alcohols 74 and 75 led to mixtures of products, but once again in good yield and with good stereocontrol. For example, solvolysis of 75 gave enol 76 as the major product after formate ester saponification. Catalytic hydrogenation of the olefin and oxidation of the secondary alcohol converted all cyclization products to cis-decalone 79.
b1026_Chapter-02.qxd
50
Advances Based on Improved Tactics
O 81
80
O
HO
H 82
H
3. Zn dust 4. CrO3, H2SO4 acetone, H2O
H O
H 83
cyclopentene as precursor of cyclohexenone and cyclohexene as acyl-cyclopentene precursor O
88
O
O 20
H 3
H
H
16
H
O
H O
16,17-dehydroprogesterone (racemic) 84
H
CHO
16
H
H 3
O 85
4
5
H 86
OH
87
Notice elements of the annulation methodology finding its way into the strategy.
Johnson, W. S.; Semmelhack, M. F.; Sultanbawa, M. U. S.; Dolak L. A. "A New Approach to Steroid Total Synthesis. A Nonenzymatic Biogenetic-Like Olefinic Cyclization Involving the Stereospecific Formation of Five Asymmetric Centers" J. Am. Chem. Soc. 1968, 90, 2994−2996.
Steroids-16
Page 50
Synthesis of a Steroid
10:54 AM
Johnson, W. S.; Kinnel, R. B. "Stereospecific Tricyclization of a Polyolefinic Acetal" J. Am. Chem. Soc. 1966, 88, 3861−3862
Organic Synthesis via Examination of Selected Products
trans-anti-trans First example of tricycle formation with stereocontrol across two ring fusions
b1026
H O
1. TsCl 2. NaI
H
89%
O
12/21/2010
HO
Organic Synthesis via Examination of Selected Natural Products
SnCl4, PhH 0-5 oC, 17 min
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 51
Organic Synthesis via Examination of Selected Products
Steroids
51
Steroids-16 A series of studies were conducted to develop new tactics for conducting polyolefin cyclizations. New initiating groups, solvents and acid promotors were examined. For example, acetal 80 was efficiently converted to 82, whose tricyclic structure resembles the ABC-ring system of the steroids, presumably via an intermediate cation of type 81 (Steroids-16). Advances of this type set the stage for the total synthesis of steroids via polyolefin cyclizations. We will focus largely on progesterone (see Steroids-1 or Steroids-19 for structure). Let’s look at the Johnson group’s plan in a retrosynthetic manner. The penultimate (next to last) intermediate in one approach to progesterone was to be 16,17-dehydroprogesterone (84). This compound was to be derived from tetracarbonyl compound 85 via a double aldol-dehydration. This highly functionalized compound was to be packaged, in a relatively benign, manner as diene 86. This strategy contains some familiar elements: (1) use of a cyclohexene precursor to the D-ring (2) use of aldol-dehydration chemistry to prepare the A-ring enone. The cyclopentene to cyclohexenone transformation is also interesting. We have previously seen a cyclohexene used as a precursor of a 5-membered ring, and now we see (or will see) that a cyclopentene can be used to rapidly prepare a 6-membered ring. This is general transformation. At this point the retrosynthetic plan diverges from what we have previously seen, and converges with Johnson’s polyolefin cyclization studies. The plan was to use a symmetrical allylic carbocation initiator (to be generated from 87). Cyclization was expected to occur via a poly-chairlike conformation to add the BCD-rings to the cyclopentene (the A-ring precursor). Termination of the cyclization by loss of a proton from C16 was expected to provide 86. It is notable that the regiochemistry of this final proton loss was not guaranteed. Tertiary alcohol 87 was expected to come from cyclopentenone 88. We will not consider the strategy for construction of 88 in detail, but simply note that the key issues are construction of the cyclopentenone (which was to be accomplished by aldol-dehydration of a 1,4-diketone) and construction of the E-1,2-disubstituted and E-trisubstituted olefins. Let’s go directly to the synthesis.
b1026_Chapter-02.qxd
52
BnO
Na enolate
Na, NH3
1. MeLi, THF
O
2. TsO 89
X
91
BnO
92
O
87
Note symmetrical cation initiator
H O
CHO
4
O
88
95
Most highly substituted cyclopentenone
O 2.5% aqeous KOH
H
H
H
OCHO
HCO2H, rt, 3h H OH
O 85
3
84
60−75%
29% ModelStudy Generality
Steroids-17
Page 52
OH
1 2
10:54 AM
H2O-EtOH
O
O
2% NaOH
2. LiAlH4
1. OsO4 2. H2S 3. Pb(OAc)4 THF, 0 °C
1. Ba(OH)2, EtOH 2. H3O+ 3. HCl, MeOH O
MeLi
30% 86
H
O
X = Br
12/21/2010
H
O
Organic Synthesis via Examination of Selected Products
H
93
acetoacetic ester synthesis of substituted acetones
1. CF3CO2H CH2Cl2, -78 °C
H
O
X = OTs LiBr
b1026
Other cyclization products formed. Tetracycle crystallized from hydrocarbon fraction.
TsCl Use of alkyne to control olefin geometry
94
CO2Et
Organic Synthesis via Examination of Selected Natural Products
X = OH 90
EtO2C O
O
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 53
Organic Synthesis via Examination of Selected Products
Steroids
53
Steroids-17 We have seen that in polyolefin cyclizations, olefin stereochemistry translates to fused-carbocycle stereochemistry (for example 80 to 82). Thus, this approach to steroids (and fused carbocycles in general) requires methods for stereoselective olefin synthesis. We will see some examples of such methodology on the next few pages, and will revisit this topic throughout the book (in particular see the Cecropia Juvenile Hormone syntheses in Chapter 11). In the current case, the E-disubstituted olefin was prepared using the wellknown dissolving metal reduction of alkynes. Terminal alkyne 89 was converted to the corresponding acetylide which reacted with homoallylic tosylate 90 to give 91. Sodium in ammonia was then used to reduce the alkyne and remove the benzyl protecting group to provide 92 (X = OH). The alcohol was converted to the corresponding bromide 92 (X = Br) via an intermediate tosylate, and alkylation of 93 with this bromide then gave 94. Ester hydrolysis, decarboxylation, and ketal hydrolysis provided 1,4-diketone 95. The conversion of 92 to 95 is a variation of the “acetoacetic ester synthesis of substituted acetones”. This tactic uses a β-dicarbonyl as the nucleophile, rather than an unactivated ketone, to control enolate regiochemistry and problems with proton transfers that often complicate alkylations of simple ketone enolates (also compare with the conversion of 5 to 4 on Steroids-3).13 Aldol-dehydration of 95 provided 88 as the major product (C4 as nucleophile and C2 as electrophile) with possibly a trace of the product derived from the alternate aldol-dehydration. Addition of methyllithium to 88 gave 87. Treatment of 87 with trifluoroacetic acid gave a variety of products from which 86 was crystallized in 30% yield (after conversion of trifluoroacetates to alcohols for separation purposes; additional 86 was available by dehydration of tetracyclic alcohols obtained from the cyclization of 87; tricyclic products were also formed). The synthesis of 84 was completed by cleavage of the olefins and aldol-dehydration of the intermediate 85. This biomimetic route to progesterone was elegant, but there was room for improvement. For example, could a cyclization terminating group be developed that would lead directly to a 5-membered D-ring? One solution to this problem was to replace the terminating olefin with an alkyne. The idea was that if the alkyne cyclized to provide a 5-membered ring, capture of an intermediate vinyl cation would directly provide the C17-acetyl group of progesterone (see cyclization substrate 99 in Steroids-18).
b1026_Chapter-02.qxd
54
Total Synthesis of dl-Progesterone
BrMg
1. LiAlH4 (90%)
138 oC, 2.5 h 90
89
Johnson-Faulkner Claisen
O 93
2. Br(CH2)4Br (2.9 equiv)
Br
O 94
92
1. (CH2OH)2, PhH, TsOH 2. NaI 3. Ph3P, PhH
73%
O
O
O
O
1. PhLi (1 eq) 2. RCHO 3. PhLi (1 eq) 4. MeOH PPh3 I
Furans are latent 1,4-dicarbonyl compounds. Metalated furans are acyl anion equivalents.
61% from 2-methylfuran
O MeLi 98 OH
99
O
NaOH
HCl
H2O-EtOH
MeOH
40% from RCHO
O
Schlosser modification of Wittig reaction gives trans-olefin
95
97
O
O
O
O 96
Johnson, W. S.; Gravestock, M. B.; McCarry, B. E. "Acetylenic Bond Participation in Biogenetic-Like Olefinic Cyclizations. II. Synthesis of dl-Progesterone" J. Am. Chem. Soc. 1971, 93, 4332−4334
Steroids-18
Page 54
1. n-BuLi
CHO
2. CrO3-2pyr (86%) 91 55%
10:54 AM
HO
EtO2C
Organic Synthesis via Examination of Selected Products
CHO
12/21/2010
98%
b1026
CH3(OEt)3 CH3CH2CO2H (0.3 mol%)
Organic Synthesis via Examination of Selected Natural Products
This synthesis addresses problems associated with construction of the D-ring in the synthesis of dl-16,17-dehydroprogesterone. Can one terminate the polyolefin cyclization to directly afford a 5-membered ring?
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 55
Organic Synthesis via Examination of Selected Products
Steroids
55
Steroids-18 The synthesis of the cyclization substrate (99) began with addition of Grignard reagent 89 to methacrolein to provide allylic alcohol 90. A variation of the Claisen rearrangment, developed by the Johnson group for the synthesis of squalene and other terpenoids, was used to establish stereochemistry in trisubstituted olefin 91. Once again I will digress. Pattern recognition plays a big role in synthesis design. Realization that 91 is a δ,γ-unsaturated carbonyl compound, and knowing that such substructures can always be prepared by a Claisen rearrangement (although this does not mean it is always the best method to use) probably played a role in the choice of this approach to 99. There is actually another way to recognize that a Claisen rearrangement might be considered as a path from 90 to 91. Do you want to do an SN2′ reaction on an allylic alcohol? Always consider a sigmatropic rearrangment as a possibility. This process “enforces” the regiochemistry of such a substitution reaction. The Claisen rearrangement is like an “enforced” SN2′ reaction of an enolate on an allylic alcohol derivative. Returning to the synthesis, an oxidation state adjustment gave aldehyde 92, which was then coupled with phorphorane 95 to provide E-olefin 96. Wittig reactions between “unstabilized phorphoranes” such as 95 and aldehydes usually provide Z-alkenes. The Schlosser modification, however, is a nice variation that provides E-olefins with good stereoselectivity, and that is what was used in this case.14 The origin of 95 is also interesting. The masked 1,4-dicarbonyl group in 95 has its origin in the choice of 2-methylfuran (93) as the point of departure. Hydrolysis of the two “enol ethers” present in furan liberates the 1,4-dicarbonyl. Finally, the organolithium intermediate derived from 93 is a nice example of how 2-metalated furans can function as acyl anion equivalents.15 Conversion of 96 to 99 followed chemistry we have already seen.
b1026_Chapter-02.qxd
56
O 1. CF3 CO2H O ClCH 2CH2 Cl
O 17
H
O H H
H
99
100
101
71%
45%
β:α= 85:15 (recrystallized to purity)
OH
O
O
O
O
5:1 17
OCOCF3 H 104
H 103
H 101
Additive Effects: Introduction of 17-Hydroxyl Group O O
O N O
TFA
O
N H
CH3 CH 2NO2
17
H H
30%
OH 99
105
106
N
H
H
H
O
107
Morton, D. R.; Gravestock, M. B.; Parry, R. J.; Johnson, W. S. “Acetylenic Bond Participation in Biogenetic-Like Olefinic Cyclizationos in Nitroalkane Solvents. Synthesis of the 17-Hydroxy-5β-pregnan-2-one System” J. Am. Chem. Soc. 1973, 95, 4418−4419.
Steroids-19
Page 56
H 102
O
O
10:54 AM
O
Organic Synthesis via Examination of Selected Products
dl-progesterone Termination Events
b1026
O
12/21/2010
H
Organic Synthesis via Examination of Selected Natural Products
H
2. K2CO3 -MeOH OH
1. O3 , 2 min 2. Zn, AcOH 3. KOH, H2 O
5:1
O
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 57
Organic Synthesis via Examination of Selected Products
Steroids
57
Steroids-19 The cyclization of 99, initiated using trifluoroacetic acid and terminated using an excess of ethylene carbonate, gave 100 in 71% as a 5:1 mixture at C17. The termination events presumably involve capture of an intermediate vinylic carbocation to 103 via highly stabilized carbocation 102. Hydrolysis of the trifluoroacetate, an intramolecular acyl transfer, and tautomerization of the resulting enol 104 provided the products. The synthesis of progesterone (101) was then completed using the now familiar transformation of the cyclopentene to the A-ring enone. Use of nitroethane as a solvent also led to an interesting termination event with introduction of a C17 hydroxyl group (as an oxime ether and mixture of stereoisomers), presumably via the sequence shown at the bottom of Steroids-19.
b1026_Chapter-02.qxd
58
Corticosteroids
H H
H 109
66%
110 C11 methyl group occupies pseudo-equatorial site
Condition dependent: TFA-ClCH2CH2Cl gives only 16%
O HO
11
HO
11
CF3CH2OH, TFA o
25 C, 42h
17
Johnson, W. S.; Escher, S.; Metcalf, B. W. "A Stereospecific Total Synthesis of Racemic 11α-Hydroxyprogesterone via a Biomimetic Polyene Cyclization" J. Am. Chem. Soc. 1976, 98, 1039−1041.
H H
H
OH 111
112
29-35%
Steroids-20
Page 58
Johnson, W. S.; DuBois, G. E. "Biomimetic Polyene Cyclizations. Asymmetric Induction by a Chiral Center Remote from the Initiating Cationic Center. 11α-Methylprogesterone" J. Am. Chem. Soc. 1976, 98, 1038−1039.
10:54 AM
108
Organic Synthesis via Examination of Selected Products
0 oC, 3h OH
11
12/21/2010
CF3CH2OH, TFA
91:9
b1026
11
Organic Synthesis via Examination of Selected Natural Products
O 11
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 59
Organic Synthesis via Examination of Selected Products
Steroids
59
Steroids-20 The Johnson group next adapted their polyolefin cyclizations to accommodate preparation of corticosteroids. Initial work directed at this goal involved the preparation and cyclization of rac-108. This material contains one stereogenic center (C11 methyl group). The idea was to see if this single stereogenic center would induce relative stereochemistry at the six stereogenic centers formed in the cyclization. Indeed this was the case and, in accord with prediction, the C11 methyl group was equatorially disposed in the product (109) (and presumably the cyclization transition state 110). When the C11 methyl group was replaced with a hydroxyl group (rac-111), the cyclization gave rac-112 with an α-hydroxyl group at C11. The cyclization of 111 was much slower than the cyclization of 108, suggesting a build-up of positive charge near the electron-withdrawing hydroxyl group during the rate-determining events of the cyclization process.
b1026_Chapter-02.qxd
60
Asymmetric Synthesis of C11-Oxygenated System
SiMe3
SH
BF3-Et2O AcOH 15 oC, 15 min
SiMe3
1. Tl(NO3)3-trihydrate MeOH, 23 oC, 5 min
S
2. Cl
O 11
2. 5% aq. HCl, 23 oC
86%
SiMe3
84%
SiMe3
113 1. (+)-α-pinene/9-BBN THF, 20 oC, 43 h (Midland Method) 2. KOH, MeOH
O HO
HO
HO
11R
O
H H
114
O
alkylation
H
75% (96%ee) +
OH
O
112
O 115
111
Corticosteroids
11
O
O Johnson, W. S.; Frei, B.; Gopalan, A. S. "Improved Asymmetric Total Synthesis of Corticoids via Biomimetic Polyene Cyclization Methodology" J. Org. Chem. 1981, 46, 1512−1513. Forfurther improvements see: Johnson, W. S.; Lyle, T. A.; Daub, G. W. "Corticoid Synthesis via Vinylic Fluoride Terminated Biomimetic Polyene Cyclizations" J. Org. Chem. 1982, 47, 161−163.
Steroids-21
O Br O
Page 60
HO
10:54 AM
59% overall
Organic Synthesis via Examination of Selected Products
SiMe3
1. n-BuLi, THF -20 oC
b1026
SH
PCC
S
S
12/21/2010
S
CHO
Organic Synthesis via Examination of Selected Natural Products
CH2OH
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 61
Organic Synthesis via Examination of Selected Products
Steroids
61
Steroids-21 The synthesis of 112 was then modified to provide a single enantiomer. This called for an asymmetric synthesis of cyclization substrate 111. This was accomplished by Midland reduction of ketone 113 to provide 114 with excellent enantioselectivity (Steroids-21). Alkylation of 114 with the appropriate bromide (prepared from 2-methylfuran according to the procedures described on Steroids-18), followed by a few well-precedented reactions, gave 115, and thence 111 and 112. Application of the Midland reduction is notable. This is a relatively early application of a reagent-controlled asymmetric synthesis. It is also notable that the Midland method works extremely well on alkyl alkynyl ketones (because they look like aldehydes to the reagent) and thus, is well-suited to this application.16
b1026_Chapter-02.qxd
62
Effect of Cation Stabilizing Groups
H
CHO
H
pentane, 15 min O
R=H
116b
R=
116a
?
HO
O HO
117
H 118a 118b
R = H (30%) R = CH=CMe 2 (77%)
(7 steps) Suggests enzymes may operate similarly
C 8 cation-stabilizing group increases yield in cyclization
CO2 iPr SiEt 3
HO
CO2 iPr AcO
9:1
X
1. 5% TFA-CH2 Cl2 , -20 o C, 1 min X
H
2. Ac2 O, 4-DMAP 3. Et3 N, PhH
OH 119 X = CH=CMe2
H 80−83%
120 X = CH=CMe2
C 8-substituent provides enormous rate acceleration
Steroids-22
When X = H, the yield is 20% if the cyclization is conducted in 20% TFA in CF3 CH 2OH-CH 2Cl2 at − 20 °C for 24 h. The reaction gives only 1−2% of product after 1 h under these conditions.
Page 62
Johnson, W. S. Telfer, S. J.; Cheng, S.; Schubert, U. "Cation-Stabilizing Auxilliaries: A New Concept in Biomimetic Polyene Cyclization" J . Am. Chem. Soc. 1987, 109, 2517−2518. An extension to another cyclization system and the corticosteroids is shown below: Johnson, W. S.; Lindell, S. D.; Steele, J. J. Am. Chem. Soc. 1987, 109, 5852.
10:54 AM
116a
O
H
Organic Synthesis via Examination of Selected Products
O
H
12/21/2010
R
b1026
R
SnCl4, 0 oC
Organic Synthesis via Examination of Selected Natural Products
R
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 63
Organic Synthesis via Examination of Selected Products
Steroids
63
Steroids-22 We will end this section with the series of studies shown on Steroids-22. These studies were designed to determine the effect of a carbocation-stabilizing group at C8 on the rate of steroid-targeted polyolefin cyclizations. The effects were remarkable. The influence on yield is seen in the cyclizations of 116a and 116b. The substrate with carbocation-stabilizing 2-methylpropenyl group at C8 (116b) gave a much higher yield of tetracycle (118b) than did the “parent” substrate 116a. This observation was translated to a system relevant to the corticosteroids. Thus 119 was converted to 120 in excellent yield. This is notable because of the low yield (slow cyclization rates) encountered with 11-hydroxy substrate 111, and the poor performance of the “parent” substrate related to 119 (see Steroids-22). The conversion of 119 to 120 also illustrates yet another termination tactic, the use of an allylsilane to direct the regiochemical course of D-ring formation.
b1026_Chapter-02.qxd
12/21/2010 b1026
64
10:54 AM
Page 64
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
References and Notes 1. Anner, G.; Miescher, K. “Steroids. LXXIII. Total Synthesis of Natural Estrone” Experientia 1948, 4, 25–26. Bachmann, W. E.; Cole, W.; Wilds, A. L. “Total Synthesis of the Sex Hormone Equilenin” J. Am. Chem. Soc. 1939, 61, 974–975. Johnson, W. S.; Petersen, J. W.; Gutsche, C. D. “A New Method of Producing Fused Ring Structures Related to the Steroids. Synthesis of Equilenin” J. Am. Chem. Soc. 1945, 67, 2274–2275. 2. For an introduction to “retrosynthetic analysis” see Corey, E. J.; Cheng, X-M. “The Logic of Chemical Synthesis”, John Wiley and Sons, 1989 (pages 5–16) 3. For some early reviews that the author (DJH) has found useful when teaching the fundamentals of the Diels-Alder reaction see: Sauer, J. “Diels-Alder Reactions. II. Reaction Mechanism” Angewandte Chem. Int. Ed. Eng. 1967, 6, 16–33. Sauer, J.; Lang, D.; Wiest, H. “Diels-Alder Reaction. II. The Addition Capacity of Cis-Trans Isomeric Dienophiles in Diene Additions” Chem. Ber. 1964, 97, 3208–3218. Sauer, J.; Wiest, H.; Mielert, A. “Diels-Alder Reaction. I. Reactivity of Dienophiles Towards Cyclopentadiene and 9,10-Dimethylanthracene” Chem. Ber. 1964, 97, 3183–3207. 4. Jung, M. E. “A Review of Annulation” Tetrahedron 1976, 32, 3–31. 5. du Feu, E. C.; McQuillin, F. J.; Robinson, R. “Synthesis of Substances Related to the Sterols. XIV. A Simple Synthesis of Certain Octalones and Ketotetrahydrohydrindenes which may be of Angle-Methyl-Substituted Type. A Theory of the Biogenesis of the Sterols” J. Chem. Soc. 1937, 53–60. 6. Bergmann, E. D.; Ginsburg, D.; Pappo, R. “The Michael Reaction” Organic Reactions 1959, 10, 179–555. 7. Van Rheenan, V.; Kelly, R. C.; Cha, D. Y. “An Improved Catalytic Osmium Tetroxide Oxidation of Olefins to cis-1,2-Glycols using Tertiary Amine Oxides as the Oxidant” Tetrahedron Lett. 1976, 17, 1973–1976. 8. Wieland, H. “Hydrogenation and Dehydrogenation” Ber. 1912, 45, 484–493. Barbier, P.; Locquin, R. “Method of Decomposing Various Saturated Mono- and Dibasic Acids” Comptes Rend. 1913, 156, 1443–1446. 9. Hirschmann, R. “The Cortisone Era: Aspects of its Impact. Some Contributions of the Merck Laboratories” Steroids 1992, 57, 579–592. 10. Hajos, Z. G.; Parrish, D. R. “Stereocontrolled Synthesis of trans-Hydrindan Steroidal Intermediates” J. Org. Chem. 1973, 38, 3239–3243. 11. Robinson, R. “Structure of Cholesterol” Chem. Ind. (London) 1934, 1062–1063. 12. Johnson, W. S. “A Fifty-Year Love Affair with Organic Chemistry” in Profiles, Pathways, and Dreams, American Chemical Society (Washington DC), 1998 (229 pages)
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 65
Organic Synthesis via Examination of Selected Products
Steroids
65
13. House, H. O.; Trost, B. M. “The Chemistry of Carbanions. IX. The Potassium and Lithium Enolates Derived from Cyclic Ketones” J. Org. Chem. 1965, 30, 1341–1348. House, H. O.; Trost, B. M. “The Chemistry of Carbanions. X. The Selective Alkylation of Unsymmetrical Ketones” J. Org. Chem. 1965, 30, 2502–2512. 14. Vedejs, E.; Peterson, M. J. “The Wittig Reaction: Stereoselectivity and a History of Mechanistic Ideas (1953–1995)” Advances in Carbanion Chemistry, 1996, 2, 1–85. Vedejs, E.; Peterson, M. J. “Stereochemistry and Mechanism in the Wittig Reaction” Topics in Stereochemistry 1994, 21, 1–157. Maryanoff, B. E.; Reitz, A. B. “The Wittig Olefination Reaction and Modifications Involving PhosphorylStabilized Carbanions. Stereochemistry, Mechanism, and Selected Synthetic Aspects” Chemical Reviews 1989, 89, 863–927. 15. Meyers, A. I. “Heterocycles in Organic Synthesis” Wiley-Interscience, 1974 (332 pages). DegI’Innocenti, A.; Pollicino, S.; Capperucci, A. “Synthesis and Stereoselective Functionalization of Silylated Heterocycles as a New Class of Formyl Anion Equivalents” Chemical Communications 2006, 4881–4893. Yus, M.; Najera, C.; Foubelo, F. “The Role of 1,3-Dithianes in Natural Product Synthesis” Tetrahedron 2003, 59, 6147–6212. Albright, J. D. “Reactions of Acyl Anion Equivalents Derived from Cyanohydrins, Protected Cyanohydrins and α-Dialkylaminonitriles” Tetrahedron 1983, 39, 3207–3233. Seebach, D.; Corey, E. J. “Generation and Synthetic Applications of 2-Lithio-1,3-dithianes” J. Org. Chem. 1975, 40, 231–237. 16. Midland, M. M.; Greer, S.; Tramontano, A.; Zderic, S. A. “Chiral Trialkylborane Reducing Agents. Preparation of 1-Deuterio Primary Alcohols of High Enantiomeric Purity.” J. Am. Chem. Soc. 1979, 101, 2352–2355. Midland, M. M.; McDowell, D. C.; Hatch, R. L.; Tramontano, A. “Reduction of α,βAcetylenic Ketones with B-3-Pinanyl-9-borabicyclo[3.3.1]nonane. High Asymmetric Induction in Aliphatic Systems” J. Am. Chem. Soc. 1980, 102, 867–9.
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 66
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
66
Problems 1. Draw the most stable conformations of cholic acid and fusidic acid (Steroids-1). Describe the stereochemistry at each ring junction and the conformation of each six-membered ring. (Steroids-1) 2. The energy difference between the chair conformations of methylcyclohexane is 1.8 kcal mol−1. Use this information to estimate the energy difference between (a) cis-decalin and trans-decalin and (b) the two chairchair conformations of each of the cis-decalins shown below. (Steroids-2)
H
H
H
H
H
3. Provide a mechanism for the reduction of the acetate of ketol 9 to ketone 5 with Zn/AcOH. (Steroids-3) 4. The following transformation is called a Pummerer rearrangement. Identify the oxidation state changes that occur at specific atoms during this reaction. Explain why no oxidizing agents are needed to accomplish this transformation. (Steroids-3)
S O
Ph
OAc
(CF3CO)2O, Ac2O (excess)
S
Ph
2,6-lutidine, 30 min, rt
Tanikaga, R.; Yabuki, Y.; Ono, N.; Kaji, A. “Facile Pummerer rearrangement of sulfoxide in an acetic anhydride-trifluoroacetic anhydride mixture” Tetrahedron Lett. 1976, 2257–2258. 5. Another reaction sequence that would be expected to accomplish the conversion of 17 to 2 follows: O H
O (MeO)2 P CH2Li
O H
O H O
O
O H
or Ph3P=CH2
O
Provide a mechanistic interpretation of this transformation. See Henrick, C. A.; Boehme, E.; Edwards, J. A.; Fried, J. H. “Reaction of phosphoranes and phosphonate anions with enol lactones. A new method for the preparation of cyclic α,β-unsaturated ketones” J. Am. Chem. Soc. 1968, 90, 5926–5927 for more information. (Steroids-4)
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 67
Organic Synthesis via Examination of Selected Products
Steroids
67
6. Suggest a mechanism for the conversion of 33 to 34. (Steroids-6) 7. Provide structures (most stable conformation) for the intermediates en route from 34 to 35. (Steroids-6) 8. Provide structures for the intermediates en route from 38 to cortisone. Provide a mechanism for the step that introduces the C4-C5 double bond. (Steroids-6) 9. Certain cationic transition metal complexes, for example [Ir(cod) (PCy3)py]+ PF6−, can be used to “direct” hydrogenations. Two examples are shown below. [Ir(cod)(PCy)3py] PF6
96:4
H2 OH
OH [Ir(cod)(PCy)3py] PF6
96:4
H2 OH
OH
Show how this observation, in conjunction with other chemistry, can be used to accomplish the following “trans-perhydroindan-producing” transformation. (Steroids-8) ? O
H
O
Crabtree, R. H.; Davis, M. W. “Directing effects in homogeneous hydrogenation with [Ir(cod)(PCy3)(py)]PF6” J. Org. Chem. 1986, 51, 2655. See also Evans, D. A.; Morrissey, M. M. “Rhodium(I)-catalyzed hydrogenation of olefins. The documentation of hydroxyl-directed stereochemical control in cyclic and acyclic systems” J. Am. Chem. Soc. 1984, 106, 3866. 10. Alkylation of enolates derived from cyclohexanone enolates occurs largely with “axial” entry of the electrophile. Propose stereoselective syntheses of A and B. (Steroids-8)
O
O A
B
b1026_Chapter-02.qxd
12/21/2010 b1026
68
10:54 AM
Page 68
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
11. Provide the structures isolated after each of the four steps used to convert 82 → 83. (Steroids-16) 12. Show the conformation of the allylic carbocation (derived from 87) that would lead to diene 86. What other tetracyclic dienes might be anticipated? (Steroids-16) 13. The Claisen rearrangement is known to proceed largely through a chairlike transition state. Explain why the rearrangement of C is more selective (in terms of olefin geometry) than D. Predict the product expected from “Ireland-Claisen” rearrangment of E. (Steroids-16) E:Z = 98:2 138°C O
EtO2C
C
OEt
E:Z = 86:14
98°C O
OHC
D
LiN(i-Pr)2, THF, rt O (aqueous workup) O
?
E
Perrin, C. L.; Faulkner, D. J. “Cis-Trans ratios in Claisen and Cope rearrangements” Tetrahedron Lett. 1969, 2783. Ireland, R. E.; Willard, A. K. “Stereoselective Generation of Ester Enolates” Tetrahedron Lett. 1975, 16, 3975–3978. Ireland, R. E.; Mueller, R. H.; Willard, A. K. “Ester Enolate Claisen Rearrangement. Construction of the Prostanoid Skeleton” J. Org. Chem. 1976, 41, 986–996. Chillous, S.; Hart, D. J.; Hutchinson, D. K. “A Non-resolutive Approach to the Preparation of Configurationally Pure Difunctional Molecules” J. Org. Chem. 1982, 47, 5418–5420.
b1026_Chapter-02.qxd
12/21/2010 b1026
10:54 AM
Page 69
Organic Synthesis via Examination of Selected Products
Steroids
69
14. Propose a mechanism for the following transformation. (Steroids-22)
CF3CO2H
OH
HO
Volkmann, R. A.; Andrews, G. C.; Johnson, W. A. “Novel Synthesis of Longifolene” J. Am. Chem. Soc. 1975, 97, 4777. 15. Propose a synthesis of the allylic alcohol used in Problem 14. (Steroids-22) 16. Propose a mechanism for the following transformation. (Steroids-22) O O
AcOSO2Me CH2Cl2 H
Corey, E. J.; Balanson, R. D. “Simple synthesis of (±)-cedrene and (±)-cedrol using a synchronous double annulation process” Tetrahedron Lett. 1973, 14, 3153–3156. 17. Propose syntheses of the following compounds. O O O
O
b1026_Chapter-03.qxd
72
Prostaglandins O
HO
O CO2H
CO2H
PGF1α
CO2H
CO2H HO
OH
HO
OH
PGE2
PGA2
O
HO
CO2H
CO2H HO
OH
O
OH
PGF2β
PGE3
OH PGD2
HO2C
OH
O
CO2H
O O
CO2H
CO2H HO
O
OH
OH
TXA2
TXB2
Prostaglandins-1
HO
OH PGI2
Page 72
PGF2α
OH
10:55 AM
O
O CO2H
HO
PGA1
Organic Synthesis via Examination of Selected Products
HO
HO
OH
PGE1
12/21/2010
OH
b1026
HO
OH
Organic Synthesis via Examination of Selected Natural Products
HO
CO2H
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 73
Organic Synthesis via Examination of Selected Products
Prostaglandins
73
Prostaglandins-1 Prostaglandins are hormones derived from the C20 fatty acid arachidonic acid. The compounds were first isolated from the prostate gland and hence their name.1 A key structural feature of the prostaglandins is the presence of a fivemembered ring with the carbons in a variety of oxidation states, adorned with two adjacent (carbon) sidechains. The thromboxanes are a closely related family of fatty acid derived hormones, two examples of which are shown here. We will focus on the prostaglandins.
b1026_Chapter-03.qxd
74
Prostaglandins: Some History and a Brief Perspective From HART/CRAINE/HART/HADAD, Organic Chemistry, 12E, Copyright 2007 Brooks/Cole, a part of Cengage Learning, Inc. Reproduced by permission, www.cengage.com/permissions
O2 11
12
CO2H
CO2H 11
12
OR PGG2 R = OH PGH2 R = H
HO
OH PGE2
Prostaglandins are widespread in tissue, and elicit a wide variety of physiological responses. They play important roles in many life processes such as digestion, blood circulation, and reproduction. It has also been established that injured cells in the body produce and release prostaglandins that, in this situation, give rise to the phenomena of inflammation and pain. In 1969, Dr. John Vane and his collaborators at the Royal College of Surgeons in London discovered that aspirin inhibited production of prostaglandins by injured tissue. Eventually it was shown that this happens because aspirin binds to cyclooxygenase, inhibiting the conversion of arachidonic acid to PGG2. This stops the production of prostaglandins, resulting in a reduction of inflammation and pain. One side effect of taking aspirin is stomach irritation and this is also a result of inhibition of prostaglandin synthesis. In turns out that PGE2 protects the cells of the stomach wall by stimulating formation of a protective layer of mucous. PGE2 also helps regulate acid levels in the stomach and, in its absence, hydrochloric acid production rises. Thus, it is understandable how inhibition of PGE2 release in the stomach by orally ingested aspirin could lead to an upset stomach. Of course, this does not keep aspirin from being useful. The discovery of how aspirin works has inspired medicinal chemists to search from new COX inhibitors that might behave as painkillers. Celebrex and (the now infamous) Vioxx are COX inhibitors that were developed for this purpose. Because prostaglandins were only isolable in small amounts from natural sources, synthesis played a key role in the determination of their structures (including absolute stereochemical assignments) and in furnishing material for biological studies. In response to this need (in part), many methods for the preparation of highly decorated cyclopentanes were developed beginning in the 1960's. Of course developments in the area of terpenoid chemistry also contriubuted to this burst of activity (do a search for cyclopentane-containing terpenoids and see what you find). This section will be devoted to selected methods that were developed for the preparation of prostaglandins with a focus on DJH's favorites.
Prostaglandins-2
Page 74
arachidonic acid
O O
10:55 AM
cyclooxygenase
O 8
12/21/2010
9
CO2H
Organic Synthesis via Examination of Selected Products
8
b1026
9
Organic Synthesis via Examination of Selected Natural Products
Prostaglandins are hormones derived from the C20 fatty acid arachidonic acid. The compounds were first isolated from the prostate gland and hence their name (Samuelsson and Bergstrom). Prostaglandins are made in cells by a series of enzyme-catalyzed reactions. A critical step is the reaction of arachidonic acid with oxygen to produce PGG2, a highly reactive peroxide-containing prostaglandin. In this reaction, promoted by an enzyme called cyclooxygenase (COX), C9 and C11 of arachidonic acid bond to oxygen, and C8 and C12 bond to one another, to form the cyclopentane ring that is characteristic of all prostaglandins. Once PGG2 is formed, it is reduced to PGH2, which is subsequently converted to PGE2 and a host of other prostaglandins.
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 75
Organic Synthesis via Examination of Selected Products
Prostaglandins
75
Prostaglandins-2 A critical step in the biosynthesis of prostaglandins is the reaction of arachidonic acid with oxygen to produce PGG2, a highly reactive peroxide-containing prostaglandin. In this reaction, promoted by an enzyme called cyclooxygenase (COX), C9 and C11 of arachidonic acid bond to oxygen, and C8 and C12 bond to one another, to form the cyclopentane ring that is characteristic of all prostaglandins.2 Once PGG2 is formed, it is reduced to PGH2, which is subsequently converted to PGE2 and a host of other prostaglandins. Prostaglandins are widespread in tissue, and elicit a wide variety of physiological responses. They play important roles in many life processes such as digestion, blood circulation, and reproduction. It has also been established that injured cells in the body produce and release prostaglandins that, in this situation, give rise to the phenomena of inflammation and pain. In 1969, Dr. John Vane and his collaborators at the Royal College of Surgeons in London discovered that aspirin inhibited production of prostaglandins by injured tissue.3 Eventually it was shown that this happens because aspirin binds to cyclooxygenase, inhibiting the conversion of arachidonic acid to PGG2. This stops the production of prostaglandins, resulting in a reduction of inflammation and pain. One side effect of taking aspirin is stomach irritation, and this is also a result of inhibition of prostaglandin synthesis. In turns out that PGE2 protects the cells of the stomach wall by stimulating formation of a protective layer of mucous. PGE2 also helps regulate acid levels in the stomach and, in its absence, hydrochloric acid production rises. Thus, it is understandable how inhibition of PGE2 release in the stomach by orally ingested aspirin could lead to an upset stomach. Of course, this does not keep aspirin from being useful. The discovery of how aspirin works has inspired medicinal chemists to search from new COX inhibitors that might behave as painkillers. Celebrex and (the now infamous) Vioxx are COX inhibitors that were developed for this purpose.4 Because prostaglandins were only isolable in small amounts from natural sources, synthesis played a key role in the determination of their structures (including absolute stereochemical assignments) and in furnishing material for biological studies. In response to this need (in part), many methods for the preparation of highly decorated cyclopentanes were developed beginning in the 1960’s. Of course, developments in the area of terpenoid chemistry also contributed to this burst of activity (do a search for cyclopentane-containing terpenoids and see what you find). This section will be devoted to selected methods that were developed for the preparation of prostaglandins with a focus on some of my favorites.
b1026_Chapter-03.qxd
76
Prostaglandin Synthesis: Some Contributions from the Corey Group Carbonyl Surrogate Poor Leaving Group
Strategy
14
12
HO
OH
X
S
Br 170-190 °C S O
6
O
1 torr 60-75%
S
9
Br
(CH2)6CN
O2N
4
15
THF-pentane 8
7
S
S
60-75%
5
5
Acyl Anion Equivalent
S
S
Synthesis of Dienophile 1. KOH (cat), MeOH, CH3NO2 2. Ac2O, H2SO4 (cat) CHO
NC 10
3. NaHCO3, EtOAc, ∆ 60%
NC NO2
NC
1. Al(Hg), H2O 90% Et2O-EtOH
4 diene
S
"high yield" O2N
S 11
2.
O
O H
O
CH3 85%
Corey, E. J.; Andersen, N. H.; Carlson, R. M.; Paust, J.; Vedejs, E.; Vlattas, I.; Winter, R. E. K. "Total Synthesis of Prostaglandins. Synthesis of the Pure dlE1, -F1 , F1 , A1, and B1 Hormones" J. Am. Chem. Soc. 1968, 90, 3245-3247.
Prostaglandins-3
Page 76
66%
O
R2
3
Li
NBS O
X
Recall Woodward Approach to Steroid D-ring
Synthesis of Diene
S
15
2 O
10:55 AM
PGE1
PGE1
O
OR
1
Organic Synthesis via Examination of Selected Products
13
R1
R2
12
11
HO
13
11
R2
15
ozonolysis
12/21/2010
9
R1
aldol
b1026
R1
CO2H
Organic Synthesis via Examination of Selected Natural Products
8
NO2
OHCHN
O
O
Good EWG for Diels-Alder which sets stereochemistry and regiochemistry
Protected Ketone
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 77
Organic Synthesis via Examination of Selected Products
Prostaglandins
77
Prostaglandins-3 We will start with some contributions to prostaglandin synthesis from the Corey group, perhaps the major contributors to this area of natural products synthesis. The structures of many prostaglandins were initially not on firm ground. Early synthetic objectives in this area were tied to a desire to establish structure, including stereochemical details. Thus, early approaches to prostaglandins did not address all stereochemical issues in a selective manner. Rather they were designed to provide a number of compounds (whose stereochemistry could be assigned using chemical and spectroscopic methods) that could be compared with natural materials to establish both structure and possibly structure-activity relationships from the standpoint of biology. The first synthesis reported by the Corey group afforded members of the PGE, PGF and PGA families of prostaglandins (see Prostaglandins-1). The strategy is outlined here within the context of PGE1 (short-handed as structure 1). The plan was to prepare this compound from 1,6-dicarbonyl compound 2 via an intramolecular aldol condensation. The details behind the choice of 2 as an intermediate are interesting. We can speculate that the choice of a formamide as a precursor to the C9 ketone, and the inclusion of a protected ketone at C15, were designed to minimize potential problems with β-elimination reactions during the aldol condensation. Protection of the C15 ketone would also differentiate it from the C13 ketone and facilitate reduction chemistry that would be needed at C13 subsequent to formation of the 5-membered ring. From the standpoint of tactics, mild conditions would be required to avoid dehydration (to an acylcyclopentene) during intramolecular aldol condensation. Intermediate 2 was to be prepared from cyclohexene 3, which was to come from a Diels-Alder reaction between nitroalkene 4 and 2-substituted-1,3-butadiene 5. This diene-dienophile pair is well-matched to provide the required regiochemistry. It is notable that this strategy is simply a variation of the cyclohexene-to-cyclopentane strategy we saw several times during our examination of steroid syntheses (see the Woodward D-ring synthesis on Steroids-4, and the Johnson D-ring synthesis on Steroids-17). In the forward direction, diene 5 was prepared by alkylation of metallated 1,3-dithiane 9 with allylic bromide 8. In this reaction, 9 plays the role of an “acyl anion equivalent”.5 We will talk about equivalencies in more detail in Chapter 6, but at this point it is worth noticing that the dithiane will eventually emerge as the C15 protected ketone. Dienophile 4 was prepared by an “aldol-dehydration” reaction between nitromethane and aldehyde 10, a reaction known as the Henry reaction.6 The Diels-Alder reaction between 4 and
OHCHN
1. (CH2OH)2, THF HgCl2 NC(H2C)6 2. OsO4, pyridine
O O
3. Pb(OAc)4, acetone OHCHN
(CH2)6CN
1. DBN, CH2Cl2 11
O
C5H11
2. Ac2O, pyridine
AcO
14 45% crystalline with minor amounts of C11 epimer
N
DBN =
(CH2)6CO2 C5H11
THPO
OHCHN (CH2)6CN 1:1
3. KOH, MeOH H2O, 110- 125 °C
AcO
17
C5H11
Zn(BH4)2
OHCHN (CH2)6CN
diglyme, 25 °C
C5H11 AcO
HO 16
15
This synthesis provided lots of information about intermediate stability, provided stereoisomers (not abstracted here) that helped with structure assignments.
1. NBS (NHBr) 2. base (imine) 3. H2O, pH 2 (THP hydrolysis, C=NH to C=O)
O
O (CH2)6CO2H C5H11 HO 25%
HO
dl-PGE1
0.5 N HCl in H2O-THF (1:1)
60-80% O
(CH2)6CO2H
60 h at 25 °C
C5H11
dl-PGE1
NaBH4 MeOH
HO (CH2)6CO2H 1:1
0 °C
C5H11 HO
HO
dl-PGA1
Prostaglandins-4
dl-PGF1α
HO
Page 78
1:1
THPO
1. KOH, MeOH 30 °C, 15 min 2. TsOH, DHP, dioxane
10:55 AM
H3N
Organic Synthesis via Examination of Selected Products
1. NaBH4, EtOH 2. H2SO4, THF 3. DCC, CuCl2 Et2O
N
12/21/2010
33% overall
O
b1026
13
O O
Organic Synthesis via Examination of Selected Natural Products
CHO
12
b1026_Chapter-03.qxd
S
OHCHN
78
S
NC(H2C)6
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 79
Organic Synthesis via Examination of Selected Products
Prostaglandins
79
5 gave 11 in good yield. Reduction of the nitro group to an amino group was accomplished using aluminum amalgam. The resulting amine was protected as the formamide (12) using formic acetic anhydride. The selective reduction of the nitro group in the presence of a nitrile, an olefin, and a dithiane is notable. One reason why it is useful to have an arsenal of reactions for accomplishing any single functional group transformation, is to be able to have at least one transformation that can be used in the presence of any other functional group.
Prostaglandins-4 The synthesis continued with conversion of the thioacetal to the corresponding acetal, vicinal dihydroxylation of the olefin, and oxidative cleavage of the resulting diol with lead tetraacetate to provide keto-aldehyde 13, a specific example of generic intermediate 2 (see Prostaglandins-3). The intramolecular aldol condensation was accomplished using mild basic conditions (amidine base DBN) to provide β-acetoxy ketone 14 as the major stereoisomer after acylation of the intermediate alcohol. One might imagine that thermodynamics are responsible for the stereoselectivity observed in the ring-forming reaction. Protected 1,3-diketone 14 was converted to enone 15 using a standard reduction-hydrolysis-dehydration reaction sequence. Reduction of the C15 ketone provided a mixture of diastereomeric alcohols (16). The acetate was hydrolyzed under mild conditions to provide the C11 alcohol. The two hydroxyl groups were then protected as THP ethers. The nitrile was then converted to the carboxylate salt with concomitant hydrolysis of the formamide under vigorous basic conditions to provide 17. Oxidation of the amino group and hydrolysis of the resulting imine gave dl-PGE1 along with equal amounts of dl-15-epi-PGE1. Dehydration of dl-PGE1 (a β-elimination) gave dl-PGA1 and reduction of dl-PGE1 with sodium borohydride provided dl-PGF1α and dl-PGF1β (the C9 diastereomer). This synthesis gave racemic material, but it also revealed a lot about intermediate stability, afforded stereoisomers that helped with structure assignments of the natural products, and provided materials for biological evaluation. Needless to say, numerous other syntheses shortly followed this report. We will next examine a so-called “second generation aldol approach” reported by the Corey group.
b1026_Chapter-03.qxd
80
Second Generation Aldol Approach
8
OMe
Ph3P=CHCHO
MeO
MeO
NO2 (CH2)6CN NaH
OMe CHO
MeO
80%
HO OH TsOH, PhH
9 11
HO
Y
(CH2)6CN
12
9
+
8 15 C
5H11
O
11
HO 85% combined
27 X = NHCHO Y = H 28 X = H Y = NHCHO
X
(CH2)6CN 8
12
15 C
5H11
NHCHO (CH2)6CN
TsOH, acetone 25 °C, 40 h
O
C5H11
O
O
2.
O
O H
O
29 X = NHCHO Y = H 30 X = H Y = NHCHO
Advanced to PGE1, 15-epi-PGE1, 11-epi-PGE1, 11,15-epi-PGE1, PGA1 and 15-epi-PGA1
O
26
NO2
1. Al(Hg), H2O Et2O-EtOH
(CH2)6CN O
CH3
O
C5H11
O
O
O
89% 25
Separation schemes for 27-30 were developed. Materials were advanced using methodology related to that described in the Diels-Alder approach. Whereas this synthesis appears to be shorter, it is truly stereorandom. The upside is that it provided stereoisomers for biological evaluation.
Corey, E. J.; Vlattas, I.; Andersen, N. H.; Harding, K. "A New Total Synthesis of Prostaglandins of the E1 and F1 Series Including 11Epiprostaglandins" J. Am. Chem. Soc. 1968, 90, 3247-3248.
Prostaglandins-5
Page 80
X
Y
10:55 AM
20
O 24
Organic Synthesis via Examination of Selected Products
OHC (CH2)6CN
C5H11
OMe
Horner-WadsworthEmmons reaction
22
19
18
9
23
O
12/21/2010
DMSO
MeO
OHC
O MeO P (CH2)6CN MeO
b1026
OMe
NO2
NaNO2
NO2
Organic Synthesis via Examination of Selected Natural Products
Br
base (CH2)6CN 21
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 81
Organic Synthesis via Examination of Selected Products
Prostaglandins
81
Prostaglandins-5 The key feature of this synthesis is the use of an acid-promoted vinylogous aldol condensation to provide compounds 27–30 from key intermediate 26. The synthesis of 26 begins with a conjugate addition between nitro-acetal 19 and α,β-unsaturated aldehyde 21 to provide a mixture of diastereomeric nitro-aldehydes 22. A Horner-Wadsworth-Emmons (HWE) reaction installed what will become the C12 sidechain. Protection of the C15 ketone was followed by conversion of the nitro group to a C9 formamide (26), as a racemic mixture of diastereomers at C8 and C9. Treatment of 26 with p-toluenesulfonic acid in acetone (presumably with some water present) gave a mixture of vinylogous aldol products 27–30 in excellent yield. Separation schemes were developed and the individual isomers were advanced to a host of prostaglandins (and prostaglandin stereoisomers) using methodology related to Diels-Alder approach we examined above. The downside of this synthesis is that it was truly stereorandom. The upside is that it was truly stereorandom; it provided stereoisomers for biological evaluation. Finally, this approach does a better job at introducing C13-C15 at the desired oxidation states. We will see that the HWE reaction became a popular method for introduction of the C12 sidechain.
A Practical Synthesis O
Wittig
reduction
resolvable
12
OR'
11
HO
HO PGF2α 31
HO
oxidation
oxidation
OR' 35 (5 equiv) CN MeO
OMe
ClCH2OMe
Cu(BF4)2
-55 oC
0 oC
37
THF
Cl CN endo-exo mixture 40
Keep below 0 oC or ...
+
CH2 C O
Ketene is not a good dienophile. Therefore there is a need for "new chemistry" ... the development of a “ketene equivalent” for use in Diels-Alder reactions.
1,5-H shifts
OMe
OMe 39
38
+ minor
Corey, E. J.; Weinshenker, N. M.; Schaaf, T. K.; Huber, W. "Stereo-Controlled Synthesis of Prostaglandins F2α and E2 (dl)" J. Am. Chem. Soc. 1969, 91, 5675.
major
Prostaglandins-6
Page 82
Cl Hot Stuff
O 34
directed hydration
Execution of the Plan
Na
11
33
10:55 AM
HWE
32
Organic Synthesis via Examination of Selected Products
OH
OR'
b1026
8
CO2H
9
36
12
R'O
8
12/21/2010
HO2C
O
HO
Organic Synthesis via Examination of Selected Natural Products
Goals
b1026_Chapter-03.qxd
82
1. All primary PGs from single precursor 2. Control stereochemistry 3. Resolution at early stage
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 83
Organic Synthesis via Examination of Selected Products
Prostaglandins
83
Prostaglandins-6 As the structures of prostaglandins were established, a need for practical syntheses developed. The Corey group set as a goal the synthesis of all primary prostaglandins from a single precursor, with control over stereochemistry, and providing the required enantiomer at an early stage of the synthesis. The strategy that was followed is illustrated here within the context of an approach to PGF2α (31). It was felt that lactone 32 might provide access to 31 (and many of the prostaglandins shown at the beginning of this chapter). Attachment of the “upper” C8 sidechain might be accomplished by reduction of the lactone to a lactol followed by an olefination reaction. Attachment of the “lower” C12 sidechain might be accomplished by oxidation of C13 to an aldehyde followed once again by olefination methodology. Lactone 32 was to be derived from acid 33. The carboxyl group was to provide a handle for resolution and also help in the “directed hydration” of the olefin needed to convert 33 to 32. Hydroxy acid 33 was to be derived from 34 by a BaeyerVilliger oxidation, and 34 was to come from a Diels-Alder reaction between a cyclopentadiene of type 35 and an appropriate dienophile. This was a very daring strategy, particularly in the early stages of the proposed synthesis. It was known that ketene, the “dienophile” that would directly lead from 35 to 34, did not partake in Diels-Alder reactions. It was also known that 5-substituted cyclopentadienes of type 35 easily isomerize to the more stable 1- and 2-substituted isomers via 1,5-hydride shifts (sigmatropic rearrangements). Thus, the proposed Diels-Alder strategy could not be accomplished directly and required use of some “indirect” tactics. In the first Corey prostaglandin synthesis we saw the use of an “acyl anion equivalent”. Now we will see another application of the “equivalency concept”, use of a “real” reactant that, after some functional group manipulations, can be used to accomplish a transformation that could not be accomplished otherwise. Let’s look at the tactics that were ultimately used by the Corey group. Sodium cyclopentadienide was alkylated with chloromethyl methyl ether to provide 37. Chloromethyl methyl ether is extremely reactive and thus, this reaction could be accomplished at a low temperature, below the temperature at which 37 isomerizes to cyclopentadienes 38 and 39. Cyclopentadiene 37 was then reacted with an excess of 2-chloroacrylonitrile, an extremely reactive dienophile. Under the influence of Lewis acid promotion (copper tetrafluoroborate) this diene-dienophile pair underwent cycloaddition, to give 40 at temperatures where diene isomerization did not occur. Although 40 was obtained as a mixture of endo and exo diasteromers, it is notable that the dienophile approached 37 from the sterically least
b1026_Chapter-03.qxd
84
-55 °C 36
THF
MeO KOH (2.5 eq), 14 h
Cu(BF4)2 37 Keep below 0 oC or ...
0 °C
Cl CN endo-exo mixture 40
DMSO, 25-30 °C
12 8
11
41 O 80% overall
MeO m-CPBA (1.25 eq) CH2Cl2 NaHCO3
O
O
42 95%
Corey, E. J.; Weinshenker, N. M.; Schaaf, T. K.; Huber, W. "Stereo-Controlled Synthesis of Prostaglandins F2α and E2 (dl)" J. Am. Chem. Soc. 1969, 91, 5675.
Prostaglandins-6 (Continued )
Page 84
ClCH2OMe
(5 equiv) MeO CN
10:55 AM
OMe
12/21/2010
Hot Stuff
Na
Organic Synthesis via Examination of Selected Products
Cl
b1026
Lactone can be hydrolyzed to hydroxyacid and resolved. (vide infra).
Organic Synthesis via Examination of Selected Natural Products
A Practical Synthesis
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 85
Organic Synthesis via Examination of Selected Products
Prostaglandins
85
hindered face (opposite the methoxymethyl group) to provide the required relative stereochemical relationship between what was to become C8, C11 and C12 of PGF2α. Cycloadduct 40 was almost there. All that had to be done was to convert the α-chloronitrile to the corresponding cyanohydrin. The cyanohydrin would then presumably afford the desired ketone 41. This transformation was accomplished in good yield by treating 40 with KOH in DMSO. Whereas it seemed likely that a cyanohydrin should be an intermediate in the conversion of 40 to 41, it did seem unlikely that the transformation of 40 to the corresponding cyanohydrin would take place via an intermolecular SN2 type of process. The mechanistic details of this reaction were eventually revealed 14 years after the Corey group reported their synthesis!7 Before proceeding, it is important to note that it is the conversion of the α-chloronitrile to a ketone (a functional group transformation) that establishes the α-chloroacrylonitrile as a ketene equivalent in Diels-Alder chemistry. α-Chloroacrylonitrile is not the only ketene equivalent that has been used in Diels-Alder chemistry, nor in this strategy for prostaglandin synthesis. Reviews have been written on this topic.8 What is perhaps most important is the development of the concept of “equivalencies” that was occuring at the time this synthesis was undertaken.9 We will see this concept again. The synthesis continued with Baeyer-Villiger oxidation of 41 to provide 42. Notice the fortunate selectivity here between the Baeyer-Villiger oxidation and olefin epoxidation.
Ac 2O
8 11
OMe AcO
43
OMe
AIBN, rt, PhH AcO
44
O DME, 1h, 25 o C
O
DME, 25 °C, 30 min
O
(Collins)
13
48
97% (Separate C 15 isomers by preparative TLC and recycyle the β-isomer by MnO2 /CH 2Cl2 or DDQ-dioxane oxidation, followed by reduction.)
70% from alcohol
crystalline
CO 2H
PGE2 RO
Ph3 P=CH(CH2 )3 CO 2
i-Bu2 AlH (2 eq) 15
OTHP
HO CO 2H
dimsyl sodium in DMSO
-60 °C, 30 min
OR 1. H2 Cr 2 O7, H2 O-water 2. purify on acid washed SiO2
OH O
50
(unstable oil) O
O O
THPO
46 90%
47
1. K2CO3 , MeOH, 15 min 2. CH 2Cl2 , TsOH (0.01 eq) DHP
O
AcO
AcO
O
15
THPO
OTHP
RO
OR
51 AcOH-H2 O (2:1) 37 o C, 3h
80%
Prostaglandins-7
R = THP (52) R = H (dl-PGF 2α) (31)
Page 86
AcO
OH
DHP =
OH
CHO
1:1
49
O
CH2 Cl2
14
15
AcO
O
CrO 3-2pyr
O
10:55 AM
O
Zn(BH4 )2
BBr3 CH2 Cl2 , 0 °C
(OMe 2)2 Na
Organic Synthesis via Examination of Selected Products
O O
P
b1026
O
45
99% overall
80%
12/21/2010
HO
n-Bu3SnH
10
OMe pyridine
12
O
I
O
Organic Synthesis via Examination of Selected Natural Products
O 42
9
I
O
O
b1026_Chapter-03.qxd
MeO
O
O
86
O
1. NaOH, H 2O (2.5 eq), 0 °C 2. CO 2 (neutralize) 3. KI3 (3 eq), 0 °C
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 87
Organic Synthesis via Examination of Selected Products
Prostaglandins
87
Prostaglandins-7 Conversion of 42 to intermediate 45 (a specific example of strategic structure 32) required regio- and stereoselective hydration of the olefin, and a transesterification. This was accomplished by lactone hydrolysis, careful neutralization using carbon dioxide to sequester excess sodium hydroxide, and an electrophilic addition (using an I+ source) to provide iodolactone 43. The alcohol was then protected and the iodide was reduced to complete the synthesis of 45. To extend the “equivalency concept” we have been developing, we could say that the iodine serves as an “equivalent” to a proton during the conversion of 42 to 45. Whereas a proton would certainly have been more direct, the use of iodine (which presumably forms an iodonium ion intermediate rather than the carbocation expected from a protonation) may help control regiochemistry and stereochemistry.10 The tactic used to reduce the iodide is also notable. This process occurs via a radical intermediate at C10, thus avoiding β-elimination reactions that would most likely have plagued metal-mediated reductions (for example Zn/AcOH), or troubles associated with the sterics (2o iodide) and selectivity (lactone and ester incompatibility) that would have accompanied hydride reductions (for example LiAlH4 or LiEt3BH). The synthesis continued with ether cleavage, oxidation of the resulting primary alcohol 46, and olefination of the intermediate aldehyde 47. Reduction of the ketone 48 gave 49 as a mixture of diastereomers at C15. These were separated and the “wrong” C15 diastereomer could be recycled via an oxidationreduction sequence. Adjustment of protecting groups (in preparation for a carbonyl reduction) gave 50. The lactone was reduced to the corresponding lactol 51 using reliable methodology and a Wittig olefination installed the C8 sidechain with control of olefin geometry. Removal of the alcohol protecting groups completed the synthesis of dl-PGF2α (31).
b1026_Chapter-03.qxd
88
Tactical Improvements
O
OMe
2. acidify O
NH2Me
HO 54
67% of theory (33.5%)
55
Corey, E. J.; Schaaf, T. K.; Huber, W.; Koelliker, U.; Weinshenker, N. M. "Total Synthesis of Prostaglandins F2α and E2 as the Naturally Occuring Forms" J. Am. Chem. Soc. 1970, 92, 397-398.
Diels-Alder Problem Cl OMe
Tl
Tl Tl2(SO4)4 KOH
Isolable and storable; Less basic than sodium salt and thus, fewer problems with isomerization
MeO CN
ClCH2OMe -20 °C Et2O
37
41
O
50-55% overall from cyclopentadienide Can also use ClCH2OBn
Corey, E. J.; Koellinker, U.; Neuffer, J.; "Methoxymethylation of Thallous Cyclopentadienide. A Simplified Preparation of a Key Intermediate for the Synthesis of Prostaglandins" J. Am. Chem. Soc. 1971, 93, 1489.
Prostaglandins-8
Page 88
53
95%
OMe
2. crystallize
HO
R S
10:55 AM
42
OH CO2
1. (+)-ephedrine
12/21/2010
CO2H
1. hydrolysis
Organic Synthesis via Examination of Selected Products
MeO
b1026
Use of (-)-ephedrine provides the enantiomer
The Resolution
Organic Synthesis via Examination of Selected Natural Products
The synthesis of prostaglandins via the "Corey Lactone" was adopted by chemical industry and used to prepare kilogram quantities of materials. The synthesis met most of the stated goals, but there was room for tactical improvements. Some problems to be addressed were: (1) How was the key hydroxy acid to be resolved? (2) Could asymmetry be introduced into the Diels-Alder reaction such that resolution would be unnecessary? (3) Could the need for stoichiometric use of n-Bu3SnH be avoided? (4) Could the C15 stereochemistry problem be solved? Research directed toward solving these tactical problems led the the development of generally useful chemistry.
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 89
Organic Synthesis via Examination of Selected Products
Prostaglandins
89
Prostaglandins-8 Synthesis of prostaglandins via the “Corey Lactone” (45) was adopted by chemical industry and used to prepare kilogram quantities of materials. The synthesis met most of the stated goals, but there was room for tactical improvements. Some problems to be addressed were: (1) How was the key hydroxy acid to be resolved? (2) Could asymmetry be introduced into the Diels-Alder reaction such that resolution would be unnecessary? (3) Could the need for stoichiometric use of n-Bu3SnH be avoided? (4) Could the C15 stereochemistry problem be solved? Research directed toward solving these tactical problems led to the development of generally useful chemistry. Resolution of 42 was accomplished by traditional methods. Conversion of 42 to hydroxy acid 53 was followed by resolution as its ephedrine salt. Variants of the Diels-Alder reaction were developed. For example, use of thallium cyclopentadienide (rather than sodium cyclopentadienide) provided a more stable, less basic nucleophile for use in generation of 37.
b1026_Chapter-03.qxd
90
Another Ketene Equivalent Cl
Cl 58 COCl
0 oC, 18h
Et2O
56
3. ∆, reflux 4. AcOH-H2O (2:1)
59
O 87%
77%
Asymmetry in the Diels-Alder Reaction: An Auxiliary Approach OBn (2.5 eq) O
O
Ph
BnO
56 AlCl3 (0.7 eq), -55 °C, 1h
H 89%
O
OR*
1. LDA, THF, -78 oC 2. O2; (EtO)3P 3. LiAlH4
BnO
BnO NaIO4 OH 62
t-BuOH-H2O OH
95% 60
61
59
O
97%
94% recovery of ROH
Corey, E. J.; Ensley, H. E. "Preparation of an Optically Active Prostaglandin Intermediate via Asymmetric Induction" J. Am. Chem. Soc. 1975, 97, 6908
Prostaglandins-9
Page 90
O 83%
Corey, E. J.; Ravindranathan, S.; Terashima, S. "A New Method for the 1,4-Addition of the Methylenecarbonyl Unit (-CH2CO-) to Dienes" J. Am. Chem. Soc. 1971, 93, 4326.
10:55 AM
30-gram laboratory scale
Acid chlorides are extremely hot dienophiles.
Organic Synthesis via Examination of Selected Products
90%
99%
O
O
b1026
-20 °C
BnO
12/21/2010
1. NaN3, 25 °C, DME, 1.5 h 2. Filter
BnO 57 COCl
ClCH2OBn
Organic Synthesis via Examination of Selected Natural Products
OBn
Tl
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 91
Organic Synthesis via Examination of Selected Products
Prostaglandins
91
Prostaglandins-9 The temperature at which the Diels-Alder could be conducted was lowered by developing α-chloroacryloyl chloride (57) as the ketene equivalent. Acid chlorides are much more reactive than their nitrile or ester counterparts in Diels-Alder reactions, and this is an excellent example of putting this kinetic fact to use.11 A Curtius rearrangment was used as the key functional group transformation in establishing the ketene equivalency. Ester 60 was then developed as a chiral auxiliary for use with diene 56. This diene-dienophile pair reacted at a very low temperature, under the influence of a Lewis acid promoter, to provide 61 in excellent yield and with high diastereoselectivity. The equivalency of 60 with ketene was established by sequential α-hydroxylation of the enolate of 61, reduction of the ester to give diol 62, and periodiate cleavage to afford 59 as a single enantiomer.
b1026_Chapter-03.qxd
92
Synthesis of Auxiliary
O
O
OH
Ph 85:15
66
Me
ent -60
CH2 =CHCOCl
Et3N
O 99% O
O
Ph
O
BnO
Xc
Ph
60
65
The Tin Hydride Problem O
O O I
O n-Me3 SnCl (0.1-0.3 eq) OMe NaBH4 , EtOH, 100 W Hg-lamp
OMe AcO
AcO 44
Can also use n-Bu3 SnCl
45 73-94%
Corey, E. J.; Suggs, J. W. "A Method for Catalytic Dehalogenations via Trialkyltin Hydrides" J. Or g. Chem. 1975, 40, 2554-2555
Prostaglandins-10
Page 92
ent -61
10:55 AM
Top-f ace endo-attack
Organic Synthesis via Examination of Selected Products
87%
1:1 kinetic 85%
1. Potential for π-stacking 2. Minimize dipole-dipole repulsion in conformation of O-acyl bond 3. Minimze steric effects in complexation of Lewis acid 4. The non-catalyzed process shows little asymmetric induction
b1026
O
64
(-)-Pulegone
M δ O
Me Me
toluene, ∆
2. KOH EtOH 63
Ph
12/21/2010
Na, i-PrOH
Organic Synthesis via Examination of Selected Natural Products
1. PhMgBr CuCl
Model for Asymmetric Induction (for enantiomer)
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 93
Organic Synthesis via Examination of Selected Products
Prostaglandins
93
Prostaglandins-10 Chiral ketene equivalent 60 was prepared from pulegone (63), a common monoterpene. Both enantiomers of 63 are known and thus, both enantiomers of 60 are available. A model that rationalizes the observed diastereoselectivity follows. The model emphasizes three points: (1) The ester reacts from a conformation that minimizes dipole-dipole repulsion in terms of conformation around the O-acyl bond. This is normally the lowest energy conformation for any ester. (2) Steric effects are minimized in the presumed reactive complex between 60 and the Lewis acid. The metal complexes opposite the large ester alkyl group, and the vinyl dienophile reacts from an s-trans conformation to minimize metal-vinyl group interactions. (3) π-Stacking contributes to shielding of one face of the olefin from the diene. It is notable that the non-catalyzed process shows little asymmetric induction.12 The use of stoichiometric amounts of tin hydride in the conversion of 44 to 45 could be perceived as a problem due to the toxicity of organotin compounds. The problem was addressed by developing a method that was catalytic in tin. Basically the trimethyltin iodide produced during this free radical reduction was recycled through trimethyltin hydride via reduction with sodium borohydride.
b1026_Chapter-03.qxd
94
The C15 Stereochemistry Problem
8
9 12
HO
O
OH Li
68
67 Both s-cis (5.90 µ) and s-trans (5.97 µ) isomers visible in IR.
H Separate at stage of PGF2α or PGE2 where it is simple.
B 71
(from limonene and thexylborane)
O
O N C O
O 15
HO
11
O 69
O 15
Et3N O
H N
IR suggest s-cis enone is stacking with π-framework of biphenyl unit
O
O 70
Corey, E. J.; Becker, K. B.; Varma, R. V. "Efficient Generation of the 15S Configuration in Prostaglandin Synthesis. Attractive Interactions in Stereochemical Control of Carbonyl Reduction" J. Am. Chem. Soc. 1972, 94, 8616.
Prostaglandins-11
Page 94
1. 71, -115 °C 2. LiOH, 120 °C, 72h (100%) 3. acidify
10:55 AM
RO
Organic Synthesis via Examination of Selected Products
With R = p-phenylbenzoyl, boron hydride 71 gives 82:18 at -120 °C in Et2O-pentane-THF.
92:8
15 11
b1026
With R = SiR3 (TBDMS or IPDMS), Zn(BH4)4 and other boron hydride give 1:1 mixture.
O
?
12/21/2010
O
Organic Synthesis via Examination of Selected Natural Products
O O
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 95
Organic Synthesis via Examination of Selected Products
Prostaglandins
95
Prostaglandins-11 The C15 stereochemistry problem represents a “general” stereochemical problem in synthesis. Consider the conversion of enone 67 to allylic alcohol 68. Compound 67 has four stereogenic centers (C8, C9, C11 and C12). Reduction of C15 to an alcohol could generate two diastereomers. In principle, the aforementioned stereogenic centers could influence the reduction, resulting in formation of one diastereomer (for example 68) in preference to the other (for example 15-epi-68). This type of diastereoselection does not have to rely on the use of chiral reducing agents and is called “relative asymmetric induction”. It is notable that this approach can be applied to one enantiomer of 67, or racemic 67. If complete stereoselectivity is observed with one enantiomer of 67, a single enantiomer of 68 would be obtained. The same reaction starting with racemic 67 would afford a racemic mixture of 68. A second approach to this problem involves use of a single enantiomer of a chiral reducing agent. This approach involves “reagent control” of stereochemistry. If we use a reagent that will reduce an enone of type 67 to the S-alcohol, regardless of stereogenic centers elsewhere in the molecule, then reduction of 67 will give 68. Of course if we start with the enantiomer of 67, we will not obtain the enantiomer of 68. We will obtain the enantiomer of 15-epi-68. This type of diastereoselection must use a single enantiomer of a chiral reducing agent. This approach is effective if one is working with single enantiomers of reaction substrates. It is not effective if one is working with racemic material. How do these approaches work with prostaglandin substrates of type 67? The bottom line is that the first approach is only marginally successful. The four ring stereogenic centers in 67 are remote from C15. The two faces of the ketone are only marginally different (review stereochemical definitions such as re and si from your earlier studies of stereochemistry). The only functional handle for differentiating the diastereotopic faces of the carbonyl group is the C11 protecting group. A number of protecting groups were surveyed in conjunction with a variety of reducing agents. The idea was clearly to use a “big” protecting group that would selectively block one diastereotopic face of the C15 carbonyl group. Ultimately it was found that the combination of urethane 70 and reducing agent 71 (single enantiomer derived from (+)-limonene) gave the desired isomer 68 with 92:8 diastereoselectivity. The explanation invoked for success was π-π stacking of the biphenyl unit with the enone as shown here.13 IR spectroscopy suggested that not only did π-complexation occur, but it occurred selectively with the enone in the s-cis conformation. It
b1026_Chapter-03.qxd
96
Reagent Control of Stereochemistry
O
O
O
OH
Ph Ph H
73
72
75
Inversion of C15 Stereochemistry 1. MsCl, Et3N, -20 °C 2. KO2 (4 eq), 18-C-6 (4.5 eq) DMSO-DMF-DME (1:1:1) 0 oC, 20 min
O O
RO 77
OH
3. Ph3 P (to reduce ROOH) 4. EtOCOCl, LiOH, H 2O-DME (1:1) (to relactonize hydroxy acid)
O O
Other typical nucleophiles, such as tetraalkylammonium acetates, fail.
RO 78
OH
75% Corey, E. J.; Nicolaou, K. C.; Shibasaki, M.; Machida, Y.; Shiner, C. S. "Superoxide ion as a Synthetically Useful Oxygen Nucleophile" Tetr ahedr on Lett . 1975, 3183-3186.
Prostaglandins-12
Page 96
O 76 N B Me H 3B O RS Corey, E. J.; Bakshi, R. R.; Shibata, S.; Chen, C.-P; Singh, V. K. " A Stable and Easily Prepared Catalyst for the R L Enantioselective Reduction of Ketones. Applications to Multistep Syntheses" J. Am. Chem. Soc. 1987, 109, 7925. See also Itsuno, S.; Ito, K.; Hirao, A.; Nakahama, S. "Asymmetric Reduction of Aliphatic Ketones with the Reagent Prepared from (S)-(-)-2-Amino-3-methyl-1,1-diphenylbutan-1-ol and Borane" J. Or g. Chem. 1984, 49, 555-557. For a review see Corey, E. J.; Helal, C. J. Angew . Chem. Int. E d. 1998, 37, 1986-2012. Complex
10:55 AM
BH3 -THF (60 mol%)
O N B H 3B Me
Organic Synthesis via Examination of Selected Products
O
O (10 mol%) N B 74 Me
91:9
THF, 23 °C , 2 min
b1026
catalyst system
O
Ph Ph H
Ph Ph H
O
O
Catalyst
Catalyst System O
12/21/2010
Use of enantiomeric catalyst gives opposite diastereomer with 90:10 selectivity.
Organic Synthesis via Examination of Selected Natural Products
O
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 97
Organic Synthesis via Examination of Selected Products
Prostaglandins
97
would be interesting to see if computational chemistry (not really developed in 1972) could have guided the synthetic effort associated with this somewhat Edisonian approach to the problem. Finally, note that the reducing agent (71) is chiral (and a single enantiomer) and thus, some “reagent control” may also be operating in this example.
Prostaglandins-12 Now let’s consider an example of reagent control of stereochemistry. Treatment of 72 with a catalytic amount of 74 (derived from proline) and “stoichiometric” amounts of borane-THF complex gives 73. Use of ent-74 gives 15-epi-73 with the same magnitude (but opposite sense) of asymmetric induction. It is clear that the reduction reagent ignores everything but the carbonyl group (and its environs) and controls the sense of the reduction. In this reaction, 74 really plays the role of a pre-catalyst with 75 being the actual catalyst. An explanation for the stereochemical sense of the reduction involves: (1) initial complexation the Lewis acidic boron with the Lewis basic carbonyl group. Complexation occurs on the convex face of the azaboraoxabicyclo[3.3.0]octane and opposite RL of the carbonyl compound (the vinyl group in this case); (2) intramolecular transfer of hydride to the carbonyl group, and (3) “ligand exchange” at nitrogen and boron to regenerate the catalyst. Note that if a reagent selects for the undesired stereoisomer, as in the conversion of 72 → 73, this can sometimes be corrected by inverting configuration. In the present case the Corey group develop potassium superoxide as a reagent for accomplishing this task (inversion of 2o alcohol 77 to 78).
b1026_Chapter-03.qxd
98
Additional Strategies O
O
O conjugate
R1
R2
component
O conjugate
R1
R1
12
addition
R2
11
HO
R2
RO
HO
80
79
RO 81
79
R2
R2
addition
R2
HO
HO
82
83
HO CO2Et
Br(CH2)6CO2Et
NaOCl, H2O2
CO2Et
O
84 100%
O
O CO2H
CO2Et
Baker’s Yeast
HO
90 52%
85 20-40% (separated)
1. DHP, TsOH 2. Bu3P-CuI, Et2O Li
O CO2Et
3. HOAc-H2O HO
86 HO
89 60%
1. i-Bu2AlH 2. I2 3. n-BuLi
15-deoxy-PGE1 87
Li 88
Sih, C. J.; Salomon, R. G.; Price, P.; Peruzotti, G. Sood, R. "Total Synthesis of dl-15-Deoxyprostaglandin E1" J. Chem. Soc., Chem. Commun. 1972, 240
Prostaglandins-13
Page 98
-10 oC, EtOH
THF, 25 oC
10:55 AM
Li
Organic Synthesis via Examination of Selected Products
Conjugate Addition
b1026
three
R1
O
O
R1
Organic Synthesis via Examination of Selected Natural Products
9
R1
12/21/2010
8
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 99
Organic Synthesis via Examination of Selected Products
Prostaglandins
99
Prostaglandins-13 Many other strategies have been adopted for the synthesis of prostaglandins. Three are shown here. One approach involves conjugate addition of the C12 sidechain (R2) to a cyclopentenone of type 80 (80 → 79). Presumably the C11 substitutent would control the stereochemistry at C12 (addition of nucleophile from opposite side of ring) and thermodynamics could then be used to control stereochemistry at the epimerizable C8 center. If the absolute stereochemistry at C11 is fixed, and the absolute stereochemistry at C15 in the nucleophile (R2) is also set, control of stereochemistry between the ring stereogenic centers in the PGE series of prostaglandins can be controlled based on choice of the coupling partners. A related “three-component coupling” approach involves conjugate addition of the C12 sidechain to enone 81 followed by trapping of the enolate with the C8 sidechain as an electrophile to provide 79. Stereochemistry would presumably be controlled by kinetics, that is addition of nucleophile and electrophile from the face of the five-membered ring opposite vicinal substitutents. A third approach involves preparation of an α-methylene cyclopentanone (83) and introduction of the remainder of the C8 sidechain via a conjugate addition. We will look at the first two strategies in some detail, followed by a glance at the α-methylene cyclopentanone strategy. All of these syntheses have convergency in common. For that matter, the “Corey Lactone” approach to prostaglandins is also convergent. In a convergent synthesis, several fragments are prepared and then assembled using key bond constructions. Convergency miminizes the longest linear sequence of reactions and can increase overall yield. For example, a 10-step sequence with a 90% yield at each step would give a 35% yield of final product. If the same target could be prepared by conducting a 5-step sequence, 4-step sequence, and a 1-step coupling reaction, each proceeding in 90% yield, the overall yield would be 53%. Some targets are amenable to convergent approaches and others are not. But if possible, convergency is desireable. Let’s start with a synthesis of racemic 15-deoxy-PGE1 (90) developed by the Sih group at the University of Wisconsin. The key reaction was the addition of vinyllithium 88 to the THP ether derived from enone 86. The vinyllithium reagent was prepared by hydroalumination-iodination of 1-octyne (87) followed by a transmetallation. Enone 86 was prepared in two steps from cyclopentadiene and ethyl 7-bromoheptanoate. Notice that the alkylation of lithium cyclopentadienide with this bromoester provided 1-substituted cyclopentadiene 84 in quantitative yield. The isomerization that had to be avoided in the Corey lactone approach was a necessary part of these tactics.
b1026_Chapter-03.qxd
100
Adaptation for Asymmetric Synthesis
91
O
O
OEt
OEt 93
H (cat)
1. Bu3P-CuI O CO2Et 2. 86
20%
3. HOAc-THF-H2O 4. Baker’s Yeast
O
HO
O
O CO2H
OH 96 (5%) 11,15-epi-ent-PGE1 minor
CO2H
+ HO
OH 95 (22%)
CO2H
+ HO
15-epi-ent-PGE1
OH 94 (20%) PGE1
Selectivity for addition trans to C11-OH is good (10:1).
Sih, C. J.; Price, P.; Sood, R.; Salomon, R. G.; Peruzzotti, G.; Casey, M. "Total Synthesis of Prostaglandins. II. Prostaglandin E1" J. Am. Chem. Soc. 1972, 94, 3643
Prostaglandins-14
Page 100
THPO
10:55 AM
25%
Organic Synthesis via Examination of Selected Products
92
b1026
OH
Li
12/21/2010
Li metal
I
3. CH2=CHOEt
Organic Synthesis via Examination of Selected Natural Products
1. i-Bu2AlH 2. I2
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 101
Organic Synthesis via Examination of Selected Products
Prostaglandins
101
Treatment of 84 with singlet oxygen, generated from bleach and hydrogen peroxide, gave 85 and 86 as a separable mixture. This transformation actually takes place by a cycloaddition between 1O2 and the 2-substituted cyclopentadiene, presumably in equilibrium with 84. The intermediate cyclic peroxide presumably undergoes fragmentation to provide the isomeric cyclopentenones. Protection of 86 as its THP ether was followed by addition of 88 as the corresponding cuprate, and protonation of the enolate to provide 89. The synthesis of 90 was completed by enzymatic hydrolysis of the ethyl ester.
Prostaglandins-14 The Sih approach was readily adapted to an asymmetric synthesis of PGE1. Alcohol 91 was prepared as a single enantiomer and converted to 92 using standard methodology. Note that installation of the ethoxyethyl ether (EE) generates a new stereogenic center. Vinyl iodide 92 was actually a mixture of diastereomers. This is a practical problem associated with EE and THP protecting groups that can complicate characterization of intermediates. Nontheless, vinyllithium 93 was prepared (once again as a presumed mixture of diastereomers), converted to the corresponding cuprate, and reacted with racemic 86 (2 equivalents of cuprate). Hydrolysis of the THP protecting group and enzymatic hydrolysis of the ester provided PGE1 (94) and 15-epient-PGE1 (95) as the major products and 11,15-epi-ent-PGE1 (96) as a minor product. The origin of PGE1 (94) and 96 was (R)-86, while the origin of 95 was (S)-86. The cuprate addition occurred trans to the C11 hydroxyl group with good selectivity. This synthesis is short, but is complicated by formation of mixtures because of the use of (rac)-86. So let’s look at a strategically related synthesis that addresses this problem.
O
O
O CO2H
(CH2)7CH3
99
R
O
OH
(methyl oleate)
11
O
MeO2C
O
LDA-THF-HPMA
101
D-Glyceraldehyde
CH3OCH2Cl (MOMCl)
R
O
OMOM
i-Pr2NEt (DIPEA) (Hunig’s Base)
O 102
CN
R
1. EVE, HCl
CN
HO TsO
R
9
10 2. NaHMDS OMOM OMOM PhH, ∆ , 6h OH 104 105 65% 85% EE = ethoxyethyl metalated cyanohydrin as EVE = ethyl vinyl ether acyl anion equivalent NaHMDS = NaN(SiMe3)2
O 1. TsCl, pyridine (73%) 2. i-Bu2AlH, PhMe (-40 °C) O 3. HCN, EtOH, NH3 (cat)
R OMOM
EEO
1. NaIO4, KMnO4 (cat) 2. H3O+ 3. CH2N2
HO
CN
CO2Me
1. NaOH, THF-H2O 0 °C
O
2. HCl, H2O
107 Me 80% Me OMe
R
OMOM 106 46%
3. CH2=C(OMe)Me H+
103 72%
O CO2H
9
8
HO
several steps
OH
11
O
HO
OH 1 PGE1
Stork, G.; Takahashi, T. "Chiral Synthesis of Prostaglandins (PGE1) from D-Glyceraldehyde" J. Am. Chem. Soc. 1977, 99, 1275-1276
Prostaglandins-15
Page 102
EEO
10% aq. H2SO4-THF
86%
76%
Note: Acetal of formaldehyde is too difficult to hydrolyze in presence of other acetals
1. cyclopentane via acyl anion alkylation 2. C8-C12 bond via aldol 3. C10-C12 from glyceraldehyde
MeO2C
10:55 AM
O
98
12/21/2010
CHO
OR’
Organic Synthesis via Examination of Selected Products
MeO2C(H2C)7
12
11
100
PGE1
10
RO
97
CO2Me
12
10
b1026
RO
OH
1
TsO
Organic Synthesis via Examination of Selected Natural Products
HO
8
CO2Me
9
10
b1026_Chapter-03.qxd
102
Asymmetry from the Chiral Pool
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 103
Organic Synthesis via Examination of Selected Products
Prostaglandins
103
Prostaglandins-15 The Stork group developed several prostaglandin syntheses, one of which is outlined here. The convergent endgame (97 → 1) is related to the Sih synthesis we just reviewed. The synthesis of the cyclopentenone (97), however, is quite different. The plan was to prepare an acyl anion equivalent of type 98.14 An intramolecular alkylation would construct the 5-membered ring. This was to be followed by a β-elimination of the C12 substituent to introduce the double bond. There is a general strategic message here. When making an unsaturated ring (particularly one that will not accommodate a trans-double bond), make the ring first (make the σ-bond) and then introduce the double bond (make the π-bond). A double bond places geometric constraints on a ring-forming reaction (the substituents on the double bond involved in ring formation must have a Z-relationship). This can be problematic. We have seen this strategy before. For example, the aldol-dehydration route to cycloalkenes makes the ring first (σ-bond formation in an aldol reaction) and introduces the double bond second (π-bond formation via dehydration). “Chiral pool” is a term used loosely to describe all enantiomerically pure materials that are available from natural sources in such abundance that they can be used as starting materials for synthesis.15 Glucose (a carbohydrate), citronellol (a terpene), and proline (an amino acid) are all members of the “chiral pool”. The Stork synthesis begins with D-glyceraldehyde acetonide (99), a common triose derivative. The chiral center in 99 is ultimately to become C11 of the prostaglandins. An aldol condensation between 99 and the enolate derived from methyl oleate (100) provided 101 as a mixture of diastereomers. Protection of the hydroxyl group and acetonide hydrolysis was accompanied by lactonization to give 103. To convert 103 to an acyl anion equivalent of type 98, the primary hydroxyl group had to be activated for displacement, the lactone had to be reduced to an aldehyde, and the “polarity” of the aldehyde had to be inverted because deprotonation of the aldehyde to directly provide 98 is not possible. These tasks were accomplished in sequence by formation of the primary tosylate, reduction of the lactone to the lactol (a cyclic hemiacetal form of the aldehyde) and cyanohydrin formation. We saw one application of the relationship between carbonyl compounds and their cyanohydrin derivatives in an earlier prostaglandin synthesis (Prostaglandins-6). Now we will see another. Cyanohydrin 104 was converted to the corresponding ethoxyethyl (EE) ether. Deprotonation of the nitrile was followed by an intramolecular alkylation to give 105. Hydrolysis of the EE ether followed by reverse cyanohydrin formation would reveal the C9 ketone and establish the metalated cyanohydrin derivative as an acyl anion
O
O CO2Me
MsCl, 4-DMAP
Zn, AcOH i-PrOH
O
OH
O CO2H
CO2Me Rβ
112 83%
HO
PGE1
86%
CO2Me 95:5
THPO
OH 1
O
111 84%
PLE = Pig Liver Esterase
OTHP
1. PhC(=S)Cl 2. n-Bu3SnH, (t-BuO)2 117 O CO2Me
TBSO
OTBS
113 70%
1. H2, Pd-BaSO4 quinoline (87%)
O
O
O H Al Li O OEt
O
CO2H
2. HF-pyridine (98%) 3. PLE (82%) HO
114
OH
PGF2
Noyori, R.; Suzuki, M. "Prostaglandin-Synthesen durch Dreikomponenten-Kupplung" Angew. Chem. 1984, 96, 854-882 (review with 152 references)
Prostaglandins-16
O 115
-100 °C, THF
HO 116 87% (90% ee)
Many other routes to this material have been developed.
Page 104
TBSO
1. AcOH-H2O-THF 2. PLE
10:55 AM
1. RβLi, Bu3P-CuI 2. RαCHO
OTHP 110 92%
Organic Synthesis via Examination of Selected Products
109 83%
b1026
THPO
OTHP
12/21/2010
CO2Me
THPO
108
OH
Organic Synthesis via Examination of Selected Natural Products
THPO
O
1. RβLi, Bu3P-CuI 2. RαCHO
b1026_Chapter-03.qxd
104
Three-Component Coupling Approach
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 105
Organic Synthesis via Examination of Selected Products
Prostaglandins
105
equivalent, but this was not done immediately. First the C8 sidechain olefin was oxidatively cleaved to afford an acid, the EE protecting groups were removed, and the acid was converted to the corresponding methyl ester using diazomethane. We can now see the reason for selecting methyl oleate as the nucleophile in the starting aldol condensation. The olefin was serving as an equivalent of the sidechain carbonyl group (recall olefins serving as carbonyl precursors in several of the steroid synthesis seen in Chapter 2). Notice that the olefin was benign throughout the transformation of 99 → 105. Consider the difficulties that might have been encountered if an attempt had been made to carry C1 through as an acid derivative. The synthesis of 107 (a specific example of 97) was accomplished by transforming the cyanohydrin into a ketone, β-elimination of the C12 substitutent, and protection of the C11 hydroxyl group. Note that the acetal protecting group in 107 does not introduce a new stereogenic center (unlike EE and THP protecting groups). The synthesis of PGE1 (1) was completed in several steps by addition of the lower sidechain in a manner similar to what we have already seen.
Prostaglandins-16 The Noyori group championed the three-component coupling approach to prostaglandins. This is illustrated here with syntheses of PGE1 (1) and PGF2 (114). Cuprate additions were used to introduce the C12 sidechain and aldol condensations of the resulting enolate were used to attach the C8 sidechain. Use of an aldol condensation leaves the C8 sidechain in too high an oxidation state for PGE1. This was corrected by dehydration of 109 to 110, followed by reduction of the electron-deficient alkene (without disturbing the relatively electron-rich C13-C14 olefin). The PGF2 synthesis also required some adjustments in oxidation state after the three-component coupling step. The aldol product (112) hydroxyl group was reduced using the BartonMcCombie reduction (radical intermediates).16 Lack of complications from potential propargyl-allenyl radical isomerization is notable. Semi-hydrogenation of the alkyne was used to establish the C5-C6 Z-olefin. One of several enantioselective routes to 108 (via 116) is shown without comment.
b1026_Chapter-03.qxd
106
A Variation on the Three-Component Theme
120 72%
81%
O
S
HO 122
121 75%
97% syn-3% anti racemate O O
I
3. I2, imidazole, Ph3P
1
+ OMe
O
OMe
N
TBSMSCl, DMF
126
125
2. n-Bu3SnH, AIBN 130 °C, 2 h
OH
1. n-BuLi (2.4 eq) -42 2 °C C, 40 min 2. CuCN (1.2 eq) -78 °C, 60 min
Bu3Sn
o
23 C, 1.5 h
TBSO Bu3Sn
HO
R
S-enantiomer converted to R-enantiomer using Mitsunobu reaction and 3,5-dinitrobenzoic acid in 91% yield. Total yield of R-enantiomer is 61% with 95% recovery of cyclodextrin.
OTBS
3. oxime 124 + BF3-Et2O (1eq) -78 °C OMe N 4. R-I (126) (2.7 eq), HMPT 8 Et3N (2 eq) 7
OTBS 128
123
imidazole (4 eq)
128
127
N
TBSO
O 1
12
OTBS
O O
1. NaHSO4 DME-H2O (5:1) 2. CH2N2 3. H2, Lindlar’s catalyst, PhH
129 (60%) + 16% of recovered oxime ether Corey, E. J.; Niimura, K.; Konishi, Y.; Hashimoto, S.; Hamada, Y. "A New Synthetic Route to Prostaglandins" Tetrahedron Lett. 1986, 27, 2199.
Prostaglandins-17
Page 106
124 1. TBSCl, imidazole DMF, 23 °C, 18 h (94%)
precipitates
10:55 AM
1
HO
Organic Synthesis via Examination of Selected Products
O O
80 °C
O
b1026
119
N
12/21/2010
DBU, PhH
CH2Cl2, 0 °C
1. n-BuLi 2. (CH2O)n (94%)
treat racemate with α-cyclodextrin
Organic Synthesis via Examination of Selected Natural Products
pyridine, MeOH 23 °C, 30 min
OMe
N
m-CPBA
H2NOMe-HCl (1.6 eq)
118
OMe
MeO N
MeO N
O
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 107
Organic Synthesis via Examination of Selected Products
Prostaglandins
107
Prostaglandins-17 Another example of the three-component coupling approach to PGs is shown here. This variation uses oxime ether 124 as the cyclopentenone component, a mixed cuprate derived from vinylstannane 128 as the C12 sidechain, and introduces the C8 sidechain via an alkylation reaction (rather that the aldol condensation used by Noyori) with iodide 126 serving as the electrophile. The preparation of 124 began with conversion of cyclopentenone (118) to oxime ether 119. Deconjugation of the double bond was observed in this reaction, presumably because the carbonyl group of the β,γ-unsaturated cyclopentonone is more electrophilic (and thus undergoes oxime formation at a faster rate) than the carbonyl group of 118 itself. Deconjugation is a fairly common observation when derivatizing a conjugated enone. Epoxidation of 119 was followed by a β-elimination reaction to give 121 (and its enantiomer 122) which were resolved by selective precipitation of 122 using α-cyclodextrin. The R-enantiomer (123 = 121) was converted to 124. The S-enantiomer (122) was also kept in play using a Mitsunobu alcohol inversion-protection sequence. Iodide 126 was prepared from terminal alkyne 125 in a straightforward manner. Use of an orthoester protecting group for the incipient C1 carboxyl group was unusual. This so-called OBO (triOxaBicycloOctane) protecting group was developed for this purpose as shown on Prostaglandins-18 without comment. Alcohol 127 was protected and converted to vinylstannane 128 using a free radical hydrostannylation. Stannane 128 was converted to the corresponding organolithium reagent by a transmetalation reaction, and subsequently to a mixed cuprate. Reaction of the cuprate with oxime ether 124, followed by trapping of the resulting anion with iodide 126, completed the three-component coupling to provide 129. This version of the three-component coupling provides the “carbon skeleton” of the prostaglandins with the required oxidation state at C7. This is a consequence of introducing the C8 sidechain via an alkylation reaction rather than an aldol reaction. One can speculate that the “enolate” derived from oxime ether 124 is better behaved in alkylation reactions than the enolate derived from ketone 108 (Prostaglandins-16), permitting this change in tactics. This may be because it is a better nucleophile or undergoes proton-transfer reactions (after alkylation) more slowly due to reduced acidity of the cyclopentanone oxime ether (from 124) relative to the cyclopentanone (from 108). There are trade-offs here with the differing tactics. The Corey approach eliminates the need for an oxidation state adjustment at C7 (needed in the Noyori approach), but requires the use of more protecting group chemistry.
40% aq. HF
CO2H
9
CO2Me
CH3CN HO
HO
OTBS 131
OH 132
73%
Ti reagent (black) effects reduction of N-O bond: C=NOMe to C=NH
PGE2 98%
KOH, MeOH HO
OH
140 °C
O
O
pyridine, CH2Cl2, 0 °C
OH
(-EtOH)
BF3-Et2O (0.25 eq)
O O
4-12 h, -15 °C, CH2Cl2
O 125 (90%)
135
134
133
O
O
HC CCH2CH2CH2COCl
sidechain = CH2CH2CH2Br (91%), n-C5H11 (90%), CH2CH2Ph (75%)
α-Methylenecyclopentane Strategy OH
O Br
O Br PGs
O 136
RO
HO
R 137
R
RO
138
RO R 139
R 140
Stork, G.; Isobe, M. "A General Approach to Prostaglandins via Methylenecyclopentanones. Total Synthesis of dl-Prostaglandin F2α" J. Am. Chem. Soc. 1975, 97, 4745-4746.
Prostaglandins-18
Page 108
(EtO)2C=O (1 eq)
10:55 AM
OH
Organic Synthesis via Examination of Selected Products
Corey, E. J.; Raju, N. "A New General Synthetic Route to Bridged Carboxylic Ortho Esters" Tetrahedron Lett. 1983, 24, 5571.
OBO-Ester Protecting Groups
b1026
130
12/21/2010
OTBS
O
O
Organic Synthesis via Examination of Selected Natural Products
TBSO
1. LiOH, THF-H2O 2. TiCl3 - 3 THF i-Bu2AlH, PhMe 3. acidification Na-citrate,H2O NaOAc
b1026_Chapter-03.qxd
CO2Me
108
OMe N
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 109
Organic Synthesis via Examination of Selected Products
Prostaglandins
109
Prostaglandins-18 The Corey group version of the three-component coupling approach was completed by hydrolysis of orthoester 129, conversion of the C1 acid to the corresponding methyl ester, and semi-hydrogenation of the alkyne to give 130. The ester was hydrolyzed and the oxime ether was then converted to the C9 carbonyl group (131). The protecting groups were then removed to give PGE2 (132). The third strategy mentioned on Prostaglandins-13 was the αmethylenecyclopentane strategy. This route was introduced by the Stork group and followed the reaction sequence outlined at the bottom of this page. This is clearly a variation of the three-component coupling approach, but it is less efficient because two carbon-carbon bond-forming reactions are used to transform 139 to the PGs.
b1026_Chapter-03.qxd
110
Organopalladium Chemistry and Prostaglandins
Unactivated olefin requires: (1) activation and (2) direction to control stereochemistry
HO
EtO2C Cl
M 9
Pd-H
Na 9
8 8
THF, Li2PdCl4 EtO2C
143
142
CO2Et
144 M = Pd Directed anti-addition of Pd and malonate across carbon-carbon double bond
Me2N 8 9
CO2Et CO2Et
H2
Me2N
CO2Et
9 8
145
146
+
93%
1. MeI 2. KOH, DMF 3. ∆
O O 9
CO2Et
8
147 85%
Pd0
Prostaglandins-19
elimination
Page 110
Me2NH, THF, -78 °C
HCl (g)
NMe2
CO2Et
Me2N
10:55 AM
141
Model Studies
Organic Synthesis via Examination of Selected Products
R
b1026
R
Organic Synthesis via Examination of Selected Natural Products
HO
12/21/2010
Holton, R. A. "Prostaglandin Synthesis via Carbopalladation" J. Am. Chem. Soc. 1977, 99, 8083-8085
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 111
Organic Synthesis via Examination of Selected Products
Prostaglandins
111
Prostaglandins-19 We will next focus on the ambitious strategy described by structure 141. The idea is to develop an equivalent to 141 that would allow addition of one sidechain to an unactiviated olefin, and capture the reactive intermediate with the other sidechain. For the sake of simplicity, I have presented this as though one sidechain is introduced as a nucleophile and the other as an electrophile, but as we will see the strategy is not restricted to polar intermediates. The critical obstacles in this strategy are (1) identifying suitable reactions for use with an “unactivated” olefin and (2) using cyclopentane substitutents (for example the two hydroxyl groups) to control addition stereochemistry and regiochemistry. We will look at two approaches, one that involves organometallic intermediates, and the other involving free radical intermediates. The Holten group (then at Purdue University) reported carbopalladation chemistry that was adaptable to this approach to prostaglandins. In model studies, it was found that allylic amine 143 reacted with diethyl sodiomalonate, in the presence of stoichiometric amounts of lithium tetrachloropalladate, to give 146 after exposure to hydrogen. This reaction apparently involves an “amine-directed” anti-electrophilic addition of electrophile (PdII) and nucleophile (malonate) across the olefin (143 → 144), followed by a synelimination of a palladium hydride (Pd-H) to afford 145, concluding with hydrogenation of the double bond to provide 146. Quaternization of the amine, conversion of the malonate to a carboxylate, and an intramolecular displacement reaction, gave lactone 147. The similarity between 147 and the Corey lactone is clear.
EtO2C 1.
Me2N
149 NMe2
CO2Et Me2N
12
CH(CO2Et)2
DMF-PhH (2:1)
145
24h, rt Cl
92%
148
Pd0 precipitates
Me2N
CH(CO2Et)2
CH(CO2Et)2 9
11
Cl
O
8 12
O
O
Me2N
2. acetone, (NH4)2CO3 (to quench XS boron hydride)
151
150 20% (origin not clear)
CH(CO2Et)2 1:1
Cl
15
O
HO 152
50% O
1. NaCN, DMF, 75 °C (Cl to CN) 2. MeI, CH3CN
Me3N
I
O CH(CO2Et)2
3. KOH; neutralize, ∆
Prostaglandins
1:1
NC
O
1:1
HO
HO
HO 154
153
75-80%
Cyanoethyl is protecting group for phosphodiester linkage in nucleic acid synthesis
Prostaglandins-20
Page 112
+
1. LiEt3BH (1.5 eq) Et2O-THF (4:1) -115 °C, 2h
10:55 AM
Me2N
Organic Synthesis via Examination of Selected Products
Stoichiometric in Pd+2
b1026
O
DIPEA (5 eq)
O 8
12/21/2010
11
Organic Synthesis via Examination of Selected Natural Products
2. Hunig’s base, ∆
9
HOCH2CH2Cl-DMSO (4:1)
CO2Et
THF, Li2PdCl4
M
THF, Li2PdCl4
CO2Et
Na
143
b1026_Chapter-03.qxd
112
The Real Thing
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 113
Organic Synthesis via Examination of Selected Products
Prostaglandins
113
Prostaglandins-20 The key to adapting this chemistry to the “real thing” was harnessing electrophilic addition chemistry of aminoalkene 145. Thus, omission of hydrogen allowed isolation of 145 in excellent yield. Repeating the “aminedirected” addition chemistry, with 2-chloroethanol as the nucleophile, presumably gave anti-oxypalladation product 148. This presumed intermediate was unstable, but reacted with enone 149 to provide a 50% overall yield of enone 151 along with some 150 as a side product. The desired product (151) was presumably formed by a carbopalladation-dehydropalladation reaction that occurred with retention of configuration at C12 of intermediate 148. This is a variation of the now well-known Heck reaction.17 It is possible that 150 is derived from 145 via a π-allyl palladium intermediate, followed by the Heck-type reaction. The synthesis of a PG intermediate was accomplished from 151 by reduction of the C15 carbonyl group to give a mixture of diastereomeric alcohols 152. This was followed by conversion of the chloride to a nitrile, to facilitate removal of this C11 hydroxyl protecting group. Quaternization of the amine, conversion of the malonate to a carboxylate, and lactone formation as in the model study, completed the synthesis of 154. It is notable that deprotection of the C11 hydroxyl group via a retro-1,4-addition accompanied the malonate hydrolysis and decarboxylation. The Holten synthesis of prostaglandins was truly ground-breaking research as it established the potential of organopalladium chemistry as a tool for the synthesis of structurally complex natural products.
12
RO
156
158
157
1. O2, sensitizer, hν 2. reduction
1. Ac2O, pyridine CH2Cl2, rt 2. lipase
HO
HO 160
1. TBSMSCl, rt imidazole, CH2Cl2
EVE, N-iodosuccinimide CH2Cl2, -20 °C
HO
O
I
2. KCN, EtOH, rt TBSO
161
TBSO 163
162 n-Bu3SnCl (0.1 eq) NaBH3CN (2 eq) t-BuN=C (20 eq) AIBN (0.1 eq) t-BuOH, ∆, 5 h
For cyclopentadiene oxidations see Kaneko, C.; Sugimoto, A.; Tanaka, S. Synthesis 1974, 876.
OEt
OEt
OEt HWE Reation
O
O
O i-Bu2AlH
Prostaglandins CHO TBSO
O
TBSO 165
166 93%
84%
8
PhMe, -20 °C
12
CN
TBSO 164 71%
Stork, G.; Sher, P. M.; Chen, H-L. "Radical Cyclization-Trapping in the Synthesis of Natural Products. A Simple, Stereocontrolled Route to Prostaglandin F2α" J. Am. Chem. Soc. 1986, 108, 6384.
Prostaglandins-21
Page 114
159
AcO
10:55 AM
OEt HO
Organic Synthesis via Examination of Selected Products
155
RO
RO
RO
b1026
12 11
12/21/2010
8
8
9
RO
O
O
Organic Synthesis via Examination of Selected Natural Products
O
b1026_Chapter-03.qxd
114
Stork's Free Radical Approach
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 115
Organic Synthesis via Examination of Selected Products
Prostaglandins
115
Prostaglandins-21 A strategically related, but tactically different, synthesis of prostaglandins was developed by the Stork group.18 This synthesis features the development of stereo- and regioselective free radical addition reactions for the functionalization of “unactivated olefins”. The idea was to prepare a generic PG precursor of type 155 from a diol derivative of type 158. It was imagined that one of the alkoxy groups could “direct” a free radical addition to C8 of 157, and that capture of the resulting C12 radical (156) from the convex face would give 155. The initial free radical was to be derived from iodide 163. Iodide 163 was prepared from cyclopentadiene (159). A singlet oxygen cycloaddition was followed by reduction of the peroxide to provide diol 160. Esterification and enzyme-mediated hydrolysis of the diester gave 161 as a single enantiomer. Protection of the C11 hydroxyl group, hydrolysis of the C9 acetate, and functionalization of the resulting alcohol gave 163. Treatment of 163 with tri-n-butyltin radicals (generated using a variation of the CoreySuggs method we saw earlier), in the presence of a large excess of tert-butylisonitrile, gave nitrile 164. This reaction presumably involves an initial 5-hexenyl radical cyclization, followed by addition of the resulting C12 radical to the isonitrile carbon. The resulting iminoyl radical expels a tert-butyl radical to generate 164. The fate of the tert-butyl radical is most likely to continue chains (abstract H-atom from Sn) and to isomerize the tert-butyl isonitrile to tert-butyl nitrile (hence the use of excess isonitrile). Conversion of nitrile 164 to aldehyde 165 and HWE olefination provided intermediate 166, whose potential as an intermediate in PG synthesis should now be obvious.
O
OEt I
O
SiMe3 12
TBSO
O
163
OSiMe3 169
168
1. 1.5% aq. HCl-THF (3:2)
6
(S)-BINAL-H, THF
O
PGF2α
14
-100 °C
15
2. Wittig (62%) 11
12
TBSO
OH
60% overall
13
O
170
171
58% overall
89% (90% de)
Another Use for α-Trialkylsilyl Enones
SiEt3
CH3CHO
SiMe3
MgBr
Jones
SiMe3
THF, -78 °C
Oxidation OH
OLi
O
O 5% NaOMe Et3Si
O
MeOH
O 80% overall
Stork, G.; Ganem, B. "α-Silylated Vinyl Ketones. A New Class of Reagents for Annelation of Ketones" J. Am. Chem. Soc. 1973, 95, 6152-6153.
OSiMe3 MeLi
Prostaglandins-22
Methyl Vinyl Ketone (MVK) gives less than 5% under the same conditions.
Page 116
TBSO
172
10:55 AM
OEt
OEt O
Organic Synthesis via Examination of Selected Products
Pd(OAc)2, CH3CN, rt
b1026
TBSO
12/21/2010
12
Organic Synthesis via Examination of Selected Natural Products
TBSO
OEt 140 °C, neat
O
167
O
n-Bu3SnCl (0.1 eq) NaBH3CN (2 eq) THF, rt, 10h, hν (254 nm)
12
b1026_Chapter-03.qxd
116
SiMe3 OEt
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 117
Organic Synthesis via Examination of Selected Products
Prostaglandins
117
Prostaglandins-22 The “isonitrile trap” approach requires two carbon-carbon bond-forming steps to introduce the C12 sidechain. A more convergent approach (one step for sidechain introduction) is shown here. Free radical generation from 163 in the presence of α-trimethylsilyl enone 167 provided (presumably) cyclizationtrapping product 168. Thermolysis of 168 resulted in “tautomerization” to α-trimethylsilyl enol ether 169. Exposure of 169 to stoichiometric amounts of PdII introduced the C13-C14 olefin in a “Saegusa oxidation” (α-palladation of the enol ether followed by a β-hydride elimination, much as in the Holton carbopalladation seen in Prostaglandins-20).19 Application of a reagentcontrolled reduction of the C15 ketone, hydrolysis of the C6 acetal, deprotection at C11, and Wittig olefination completed a synthesis of PGF2α (172).
b1026_Chapter-03.qxd
118
The Acyclic Diastereoselection Problem
trans ANTItrans
SYNtrans
173
174
175
H
H
H
H
H
H cis 176
ANTIcis
SYNcis SYNtrans = ANTIcis
174
175
SYNcis = ANTItrans
1. 9-BBN 2. NaOH, H2O2, H2O
1. 9-BBN 2. NaOH, H2O2, H2O
OH
syn-addition 177
OH
syn-addition 178
179
(racemic)
Prostaglandins-23
180 (racemic)
Page 118
anti-addition
syn-addition
10:55 AM
H
H
H
H
12/21/2010
H
Organic Synthesis via Examination of Selected Products
anti-addition
H
b1026
syn-addition
Organic Synthesis via Examination of Selected Natural Products
Control of Vicinal Stereochemistry Through Control of Olefin Geometry
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 119
Organic Synthesis via Examination of Selected Products
Prostaglandins
119
Prostaglandins-23 Let’s look again at the problem of establishing the relative stereochemistry between the C15 and ring stereogenic centers in the prostaglandins. In the syntheses examined thus far, this problem has been handled in two ways: (1) use of ring stereogenic centers to control diastereoselectivity in reduction of a C15 ketone (Prostaglandins-11) and (2) independent control of absolute stereochemistry at C15 either before (Prostaglandins-12, 16, 17) or after (Prostaglandins-22) introduction of the C12 sidechain. Let’s look at another approach to this acyclic diastereoselection problem that has proved useful well beyond the context of prostaglandin synthesis (as have the two strategies we have already studied). We will start by examining two general topics in stereocontrol: (1) control of vicinal stereochemistry through control of olefin geometry and (2) transfer of chirality. There are a number of approaches to controlling vicinal stereochemistry. One useful method involves olefin addition chemistry. For example, antiaddition of two “groups” (white balls) to trans-olefin 173 affords 174, whereas syn-addition of the same two groups to 173 gives 175. Adducts 174 and 175 are diastereomers. Therefore the stereochemistry of the addition, “syn” or “anti”, dictates vicinal stereochemistry in the product. If we start with the olefin geometrical isomer (176), the anti-addition provides 175 and the syn-addition gives 174. Thus, starting olefin geometry also dictates vicinal stereochemistry in the product. The mechanism by which the addition takes place is of no real consequence, as long as it is a stereospecific process (syn or anti). Electrophilic addition reactions and cycloaddition reactions, for example, are two very different families of reactions that can be used in this approach to control of vicinal stereogenic centers. The hydroboration-oxidation of olefins is one example of this approach. The process involves a stereospecific syn-addition of “H” and “OH” across the olefin π-bond. This reaction sequence is regioselective with unsymmetrical olefins. The diastereoselective conversion of E-olefin 177 to 178, and the isomeric Z-olefin 179 to 180, illustrates this process. Note that this discussion applies to relative control of stereochemistry. Whereas use of an enantioselective hydroborating agent might afford 178 and 180 as single enantiomers (or enriched in one enantiomer), 177 would still provide 178 and the diastereomeric olefin (179) would provide the diastereomeric alcohol (180).20
S
R
R
O 183
182
181
S
O
O
O
O
185
184
186
R2
R2 O S
S
R1
S Ph
S
R Ph
188
S
R1 Ph
S 190
R2 O S
R1 191
Ph R2 R1
R2
O
189
R2
R1
O
Ph
187
O S
S
R1
R2
O
Ph R2 R1
S
O
(MeO) 3P
R
OH
Ph R1 192
In principle, diastereomeric sulf oxides 187 and 188 can interconvert via 191 and also both af ford a single allylic alcohol (192).
191
For a review on transf er of chirality see: Hill, R. K. "Chirality Transfer via Sigmatropic Rearrangements" in "Asymmetr ic Sy nt hesis", Morrison, J. D., Ed.; Academic Press, 1984, 3, 503-572 (266 references).
Prostaglandins-24
Page 120
Stereoisomeric starting materials lead to stereoisomeric products. Understand that the absolute configuration of starting materials and products are not important. For example it is possible to have an and R/ S pair of starting materials that rearrange to an R/ S pair of products with the R starting materials providing the R product and the S starting material provides the S product. Many sigmatropic rearrangements take place with excellent transfer of chirality (not always 100% due to stereochemical leakage due to competing rearrangment transition states). The Claisen rearrangement (above) is one example. The Mislow-Evans rearrangement (sulfoxide to sulfenate ester as shown below) is another.
10:55 AM
O
Organic Synthesis via Examination of Selected Products
S
R R
b1026
S
12/21/2010
O
O
O
Organic Synthesis via Examination of Selected Natural Products
O
b1026_Chapter-03.qxd
120
Transfer of Chirality
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 121
Organic Synthesis via Examination of Selected Products
Prostaglandins
121
Prostaglandins-24 What do we mean by transfer of chirality? Reactions in which the stereoselective formation of one stereogenic center is connected to the destruction of another stereogenic center involve transfer of chirality.21 Sigmatropic rearrangements are a large family of intramolecular reactions that often can be used to transfer chirality. We will look at two examples: (1) the Claisen rearrangement (of allyl vinyl ethers to β,γ-unsaturated carbonyl compounds) and (2) the Mislow-Evans rearrangement (of allylic sulfoxides to allylic alcohols via sulfenate esters). Thermal rearrangement of 181 provides 183, presumably via chairlike transition state 182. This reaction involves transfer of chirality from a carbonoxygen bond to chirality in a carbon-carbon bond. The transfer of chirality is high. Stereoisomeric starting materials lead to stereoisomeric products. For example 184 (the enantiomer of 181) affords 186 (the enantiomer of 183). The designation of absolute configuration in starting materials and products are not important. For example it is possible to have an and R/S pair of starting materials that rearrange to an R/S pair of products where the R starting material provides the R product and the S starting material provides the S product. Many sigmatropic rearrangements take place with excellent transfer of chirality (not always 100% due to stereochemical leakage due to competing rearrangment transition states). Claisen rearrangements (including many variations of the reaction shown here) are one example. The Mislow-Evans rearrangement (sulfoxide to sulfenate ester) is another. For example, in the presence of thiophilic reagents such as trimethyl phosphite, allylic sulfoxides of type 187 or 188 are converted to allylic alcohols 192 with excellent transfer of chirality. This reaction involves transfer of chirality from a carbon-sulfur bond to a carbon-oxygen bond. As with the Claisen rearrangement, the transfer of chirality is high because the transition state geometry in the critical [2,3]-sigmatropic rearrangement is well defined. The Evans-Mislow rearrangment is, in principle, a bit more complex than the Claisen from the standpoint of analysis. The sulfoxide sulfur represents a stereogenic center. Thus, 187 and 188 are diastereomers. Nonetheless, it is the stereogenicity at carbon that controls the rearrangment and thus, both diastereomers lead to 192 (largely) as single enantiomer. Note that the enantiomers of 187 and 188 should both provide the enantiomer of 192. So what does this have to do with the acyclic diastereoselection problem in posed by the prostaglandins?
O
OH
193
Br
195 81%
N2 O
O
p-TsN3, Et3N
EtO
O
O
O
196 55%
toluene, ∆, 2h
H PhSH, KO-t-Bu CO2Et H
50%
m-CPBA -78 °C, 5 min CH2Cl2
O
EtOH, rt, 5 min
CO2Et 12
198
vicinal stereochemistry established
Ph
CO2Et
O CO2Et
13
S
199 (69%)
Ph
S
O
O S Ph
200
vicinal stereochemistry retained (SN2 reaction) O
Vicinal stereochemical relationship translated to 1,4-stereochemical relationship
(MeO)3P MeOH ∆, 30 min
O
O CO2Et
CO2H
PGA2
15 12
203
OH
Prostaglandins-25
202 (63%)
OH
201
Page 122
197 Cu(bronze)
O
EtO
CH3CN, 2h, rt
O
EtO
10:55 AM
0 °C to rt, 20 min
Organic Synthesis via Examination of Selected Products
194 84%
12/21/2010
PBr3, CaH Et2O, 0 °C
b1026
LiAlH4 Et2O, 0 °C
Organic Synthesis via Examination of Selected Natural Products
Taber, D. F. "Cyclopentanone Ring Formation with Control of Side Chain Stereochemistry. A Simple Stereoselective Route to the Prostaglandins" J. Am. Chem. Soc. 1977, 99, 3513-3514. For another application of the Mislow-Evans rearrangement to this problem see: Stork, G.; Raucher, S. J. Am. Chem. Soc. 1976, 98, 1583 and Kondo, K.; Umemoto, T.; Takahatake, Y.; Tunemoto, D.Tetrahedron Lett. 1977, 113. H
b1026_Chapter-03.qxd
122
Application to Prostaglandin Problem
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 123
Organic Synthesis via Examination of Selected Products
Prostaglandins
123
Prostaglandins-25 Within the context of a synthesis of PGA2 (203), Taber united the two stereochemical topics we have just reviewed to establish the stereochemical relationship between ring stereogenic centers and the acyclic stereogenic center (C15). There are three key steps in his plan: (1) Starting with diazo compound 197, a syn-addition of a carbene to the C12-C13 trans double bond established a well-defined vicinal stereochemical relationship between the incipient C12 and C13 carbons of PGA2. (2) An SN2 opening of 198, using potassium thiophenoxide as a nucleophile, modified the vicinal stereochemical relationship (198 → 199). The overall conversion of 197 to 199 accomplished the anti-addition of a carbon electrophile and a sulfur nucleophile across a trans-olefin. (3) Oxidation of sulfide 199 provided a mixture of sulfoxides which underwent Mislow-Evans rearrangement to provide allylic alcohol 202 with transfer of chirality. The conversion of 199 to 202 involved translation of a 1,2-stereochemical relationship into a 1,4-stereochemical relationship with control of diastereoselectivity. The sequence of establishing 1,2-stereochemistry, followed by transfer of chirality to establish 1,4- (or other) stereochemical relationships is one we will see again.
b1026_Chapter-03.qxd
124
Application to the Terpene Sidechain Stereochemical Problem
206 (75% overall) Me2CuLi 0 °C
H Net anti-addition of two carbons across a carbon-carbon π-bond.
∆ O
H t-Bu-O2C O
208 86% overall
Prostaglandins-26
207
Page 124
O 205
204
10:55 AM
PhMe
12/21/2010
O
t-Bu-O2C
t-Bu-O2C N N O
Organic Synthesis via Examination of Selected Products
Et3N, 60 °C
b1026
TsN3, CH3CN
t-Bu-O2C
H H
Cu-bronze
Organic Synthesis via Examination of Selected Natural Products
Trost, B. M.; Taber, D. F.; Alper, J. B. "An Approach to the Stereocontrolled Creation of an Acyclic Side Chain of Some Natural Products" Tetrahedron Lett. 1976, 3857-3860.
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 125
Organic Synthesis via Examination of Selected Products
Prostaglandins
125
Prostaglandins-26 Prostaglandins-26 illustrates how the vicinal stereocontrol exercise just examined within the context of prostaglandins has been applied to the steroid (or terpenoid) sidechain problem we encountered in Chapter 2. Diazoketone 205 (derived from 204) was converted to 206 using an intramolecular carbenoid cyclopropanation of the cis-olefin. Opening of the cyclopropane with lithium dimethyl cuprate gave 207 which was then converted to 208. The relationship between 208 and the D-ring of steroids (with appended C17 sidechain) is clear.
MeO 2C CO2 Me
209
211
210
NaH, THF
CHO
resolved as brucine salt
MeO 2C CO2 Me
1. NaBH4
(CH2 )6 CO 2Me
MeO 2C CO2 Me
O
MeO2 C
2. DHP, H +
CO 2Me
MeO2 C
NaH
OTHP
OTHP
O
CO 2Me
216
214
213
O
O CO 2H
For additional details see: Abraham, N. A. Tetr ahedr on Lett . 1973 , 451.
CO 2H +
OH
OH
218
217
Prostaglandins-27
Page 126
215 MeO2 C
10:55 AM
2. Me2 S
212 C5 H11
12/21/2010
MeO 2C CO2 Me
(MeO)2
Organic Synthesis via Examination of Selected Products
HO 2C CO2 H
1. O3 , MeOH, -75 oC
b1026
CH2 N2
O
P
Organic Synthesis via Examination of Selected Natural Products
Abraham, N. A. "Prostaglandin IX. A Simple Synthesis of Optically Active 11-Deoxyprostaglandins" T etr ahedr on Lett . 1974, 1393-1394. O
b1026_Chapter-03.qxd
126
Geminally Activated Cyclopropanes as Carbon Electrophiles
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 127
Organic Synthesis via Examination of Selected Products
Prostaglandins
127
Prostaglandins-27 Compounds of type 198 (Prostaglandins-25) and 206 (Prostaglandins-26) are known as geminally activated cyclopropanes. They are cyclopropanes with two electron-withdrawing groups positioned on one of the cyclopropane carbons. Geminally activated cyclopropanes are carbon electrophiles that react with nucleophiles in SN2-like processes in which a carbon (positioned between the electron-withdrawing groups) behaves as a leaving group. It is this reactivity pattern that makes geminally activated cyclopropanes useful synthetic intermediates. In our last look at prostaglandins, we can see this put to use in an approach to some prostaglandin derivatives. The plan was to construct PG derivatives from an intermediate of type 216. This β-ketoester was to be derived from a Dieckmann condensation (construction of the C9-C10 bond). The Dieckmann precursor was to be derived from the reaction of malonate 215 with geminally activated cyclopropane 214. Note that control of absolute stereochemistry in this route leads back to 209, which was resolved as its brucine salt. It is also notable that 214 reacts selectively with malonate 215 at the more hindered secondary (but allylic) carbon of the cyclopropane with clean inversion of configuration. This brings to a close to our examination of routes to prostaglandins. I have chosen the geminally activated cyclopropane route as the last route to provide an entrée to our next topic, synthesis of pyrrolizidine alkaloids discussed in Chapter 4. This is not the only connection between these two topics. Prostaglandins and pyrrolizidines both contain 5-membered rings as important substructures, and the acyclic diastereoselection problem rears its head in both families of natural products. Thus, there will be some overlap in chemistry as well as some new strategies to study as we move forward.
b1026_Chapter-03.qxd
12/21/2010 b1026
128
10:55 AM
Page 128
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
References and Notes 1. Bergström, S.; Samuelsson, B. “Prostaglandins and Related Factors. XI. Isolation of Prostaglandin E1 from Human Seminal Plasma” J. Biol. Chem. 1962, 237, 3005–3006. Bergström, S.; Ryhage, R.; Samuelsson, B.; Sjovall, J. “Prostaglandins and Related Factors. XV. The structures of Prostaglandin E1, F1α, and F1β” J. Biol Chem. 1963, 238, 3555–3564. Samuelsson B.; Stallbert, G. “Prostaglandins and Related Factors. XVI. Structure and Synthesis of a Derivative of Prostaglandin E1” Acta Chem. Scand. 1963, 17, 810–816. Bergström, S. “Prostaglandins: Members of a New Hormonal System” Science 1967, 157, 382–391. 2. Samuelsson, B. “Biosynthesis of Prostaglandins” Progr. Biochem. Pharmacol. 1969, 5, 109–128. Bakhle, Y. S. “Structure of COX-1 and COX-2 Enzymes and Their Interaction with Inhibitors” Drugs of Today 1999, 35, 237–250. 3. Vane, J. R.; Botting, R. M. “The Mechanism of Action of Aspirin” Thrombosis Research 2003, 110, 255–258. Moncada, S.; Vane, J. R. “Mode of Action of Aspirin-like Drugs” Advances in Internal Medicine 1979, 24, 1–22. 4. Pennisi, E. “Building a Better Aspirin” Science 1998, 280, 1191–1192. 5. Yus, M.; Najera, C.; Foubelo, F. “The Role of 1,3-Dithianes in Natural Product Synthesis” Tetrahedron 2003, 59, 6147–6212. Groebel, B. T.; Seebach, D. “Umpolung of the Reactivity of Carbonyl Compounds Through SulfurContaining Reagents” Synthesis 1977, 357–402. Seebach, D.; Corey, E. J. “Generation and Synthetic Applications of 2-Lithio-1,3-dithianes” J. Org. Chem. 1975, 40, 231–237. 6. Luzzio, F. A. “The Henry Reaction: Recent Examples” Tetrahedron 2001, 57, 915–945. 7. Shiner, C. S.; Fisher, A. M.; Yacoby, F. “Intermediacy of α-Chloro Amides in the Basic Hydrolysis of α-Chloro nitriles to Ketones” Tetrahedron Lett. 1983, 24, 5687–5690. 8. Aggarwal, V. K.; Ali, A.; Coogan, M. P. “The Development and Use of Ketene Equivalents in [4+2] Cycloadditions for Organic Synthesis” Tetrahedron 1999, 55, 293–312. Ranganathan, S.; Ranganathan, D.; Mehrotra, A. K. “Ketene Equivalents” Synthesis 1977, 289–296. 9. Seebach, D. “Methoden der Reaktivitätsumpolung” Angewandte Chem. 1979, 91, 259–278. Evans, D. A.; Andrews, G. C. “Allylic Sulfoxides. Useful Intermediates in Organic Synthesis” Acct. Chem. Res. 1974, 7, 147–155. 10. For an excellent description of how to determine whether an electophile adds to an alkene to give a bridged intermediate or a carbocation see: Alder, R. W.; Baker, R.; Brown, J. M. “Mechanism in Organic Chemistry” Wiley-Interscience, 1971, 297–302 11. For some very useful rate data see: Sauer, J. “Diels-Alder Reactions. II. Reaction Mechanism” Angewandte Chem. Int. Ed. Eng. 1967, 6, 16–33. Sauer, J.; Lang, D.;
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 129
Organic Synthesis via Examination of Selected Products
Prostaglandins
12.
13.
14.
15.
16. 17. 18. 19.
20.
21.
129
Wiest, H. “Diels-Alder Reaction. II. The Addition Capacity of Cis-Trans Isomeric Dienophiles in Diene Additions” Chem. Ber. 1964, 97, 3208–3218. Sauer, J.; Wiest, H.; Mielert, A. “Diels-Alder Reaction. I. Reactivity of Dienophiles Towards Cyclopentadiene and 9,10-Dimethylanthracene” Chem. Ber. 1964, 97, 3183–3207. Evans, D. A.; Helmchen, G.; Ruping, M.; Wolfgang, J. “Chiral Auxiliaries in Organic Synthesis” Asymmetric Synthesis, Christmann, M.; Braese, S., Eds.; Wiley-VCH, 2007, 3–9. Jones, G. B. “π-Shielding in Organic Synthesis” Tetrahedron 2001, 57, 7999–8016. Jones, G. B.; Chapman, B. J. “π-Stacking Effects in Asymmetric Synthesis” Synthesis 1995, 475–497. Albright, J. D. “Reactions of Acyl Anion Equivalents Derived from Cyanohydrins, Protected Cyanohydrins and α-Dialkylaminonitriles” Tetrahedron 1983, 39, 3207–3233. Scott, J. W. “Readily Available Chiral Carbon Fragments and Their Use in Synthesis” in Asymmetric Synthesis, Morrison, J. D., Scott, J. W., Eds.; Academic Press, 1984, 4, 1–226. Barton, D. H. R.; McCombie, S. W. “New Method for the Deoxygenation of Secondary Alcohols” J. Chem. Soc., Perkin 1 1975, 1574–1585. Link, J. T. “The Intramolecular Heck Reaction” Organic Reactions 2002, 60, 156–534. Keck, G. E.; Burnett, D. A. “β-Stannyl Enones as Radical Traps: A Very Direct Route to PGF2α” J. Org. Chem. 1987, 52, 2958–2960. Ito, Y.; Hirao, T.; Saegusa, T. “Synthesis of α,β-Unsaturated Carbonyl Compounds by Palladium(II)-Catalyzed Dehydrosilylation of Silyl Enol Ethers” J. Org. Chem. 1978, 43, 1011–1013. Matteson, D. S. “The Use of Chiral Organoboranes in Organic Synthesis” Synthesis 1986, 973–985. Masamune, S.; Kim, B. M.; Petersen, J. S.; Sato, T.; Veenstra, S. J.; Imai, T. “Organoboron Compounds in Organic Synthesis. 1. Asymmetric Hydroboration” J. Am. Chem. Soc. 1985, 107, 4549–4591. Egger, M.; Keese, R. “MNDO Analysis of Regio- and Stereoselectivity in Hydroboration” Helv. Chim. Acta 1987, 70, 1843–1854. For another interesting application of the Masamune reagent see: Imai, T; Tamura, T.; Yamamuro, A.; Sato, T.; Wollmann, T. A.; Kennedy, R. M.; Masamune, S. “Organoboron Compounds in Organic Synthesis. 2. Asymmetric Reduction of Dialkyl Ketones with (R,R)- or (S,S)-2,5-Dimethylborolane” J. Am. Chem. Soc. 1986, 108, 7402–7204 (and the following paper which discusses mechanistic details of these reactions). Nubbemeyer, U. “Recent Advances in Asymmetric [3.3]-Sigmatropic Rearrangements” Synthesis 2003, 961–1008. Hill, R. K. “Chirality Transfer via Sigmatropic Rearrangements” Asymmetric Synthesis 1984, 3, 503–572.
b1026_Chapter-03.qxd
12/21/2010 b1026
130
10:55 AM
Page 130
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
Problems 1. Propose a reaction sequence that will convert PhCHO to PhCDO via (a) a sequence in which the deuterium label is introduced as an electrophile and (b) a sequence in which the deuterium label is introduced as a nucleophile. (Prostaglandins-3) 2. Suggest alternative tactics for the conversion of 11 → 12 (Prostaglandins3). What problems might be anticipated if this conversion was attempted using (a) H2, Pd/C or (b) LiAlH4. (Prostaglandins-3) 3. Provide the structures of intermediates en route from 14 → 15. Suggest a mechanism for the dehydration step (DCC, CuCl2, Et2O). (Prostaglandins-4) 4. Suggest a mechanism for the conversion of 26 → 27–30. (Prostaglandins-5) 5. Propose a mechanism for the conversion of 40 → 41 that does not rely on an intermolecular SN2 reaction at C6. (Prostaglandins-6) 6. Illustrate how the following reagents could serve as ketene equivalents for use in Diels-Alder reactions with 1,3-cyclopentadiene. (Prostaglandins-6) S Ph
O
NO2
PPh3 Br
C
EtO2C
O2Et
For references see: Williams, R. V.; Lin, X. “New ketene equivalents for the Diels-Alder reaction. Vinyl sulfoxide cycloaddition.” J. Chem. Soc., Chem. Commun. 1989, 1872–1873. Ruden, R.; Bonjouklian, R. “Cycloaddition of vinyl triphenyl-phosphonium bromide. New synthesis of cyclic phosphonium salts.” Tetrahedron Lett. 1974, 15, 2095–2098. Ranganathan, S.; Ranganathan, D.; Mehrotra, A. K. “Nitroethylene as a versatile ketene equivalent. Novel one-step preparation of prostaglandin intermediates by reduction and abnormal Nef reaction.” J. Am. Chem. Soc. 1974, 96, 5261–5262. Kozikowski, A. P.; Floyd, W. S.; Kuniak, M. P. “1,3-Diethoxycarbonylallene: an active dienophile and ethoxycarbonylketene equivalent in the synthesis of antibiotic C-nucleosides.” J. Chem. Soc., Chem. Commun. 1977, 582–583. 7. Illustrate how the following reagents could serve as “CH2=CHOH” and “CH2=CHNH2” equivalents in Diels-Alder reactions with 1,3cyclopentadiene. (Prostaglandins-6) O
B(OBu)2 Cl
O
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 131
Organic Synthesis via Examination of Selected Products
Prostaglandins
9. 10.
11. 12. 13.
131
For a relevant review see: Vaultier, M.; Lorvelec, G.; Plunian, B.; Paulus, O.; Bouju, P.; Mortier, J. “Recent Developments in the Use of α , β -Unsaturated Boronates as Partners in Diels-Alder Cycloadditions” Royal Society of Chemistry Special Publication 253 (Contemporary Boron Chemistry, 2000, 464–471. For another relevant article see: LeBel, N. A.; Cherluck, R. M.; Curtix, E. A. “Improved synthesis of amides from the Curtius reaction. Reaction of isocyanates and organolighium compounds” Synthesis 1973, 678–679. Provide the structures of intermediates in the conversion of 58 → 59. (Prostaglandins-9) Ethyl acetate is not miscible with water whereas γ-butyrolactone is freely soluble in water. Explain this observation and describe its relevance to the asymmetric induction model developed in Prostaglandins-10. Suggest a mechanism for the conversion of 44 → 45. (Prostaglandins-10). Provide the structures of intermediates in the conversion of 77 → 78. (Prostaglandins-12) The Mitsunobu reaction is another method frequently used to invert alcohol stereochemistry [for a review see Mitsunobu, O. “The use of diethyl azodicarboxylate and triphenylphosphine in synthesis and transformation of natural products” Synthesis 1981, 1–28]. An example is shown below.
EtO2C
O
OH
C6H13
EtO2C
O
C6H13 NO2
p-nitrobenzoic acid, PhH, ∆ EtO2CN=NCO2Et,Ph3P 94%
O O
Provide a mechanism for this reaction. (Prostaglandins-12). Also see: Bose, A. K.; Lal, B.; Hoffman, W. A, III; Manhas, M. S. “Steroids IX. Facile inversion of unhindered sterol configuration” Tetrahedron Lett. 1973, 14, 1619–1622. 14. Explain how the reaction of aqueous NaOCl with hydrogen peroxide generates singlet oxygen. (Prostaglandins-13) 15. Provide mechanistic details for the conversion of 84 → 85 + 86. (Prostaglandins-13) 16. Propose at least three approaches to the asymmetric (enantioselective) synthesis of propargylic alcohol 91. (Prostaglandins-14)
b1026_Chapter-03.qxd
12/21/2010 b1026
132
10:55 AM
Page 132
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
17. Propose syntheses of the following butenolides. Keep in mind the benefits of late introduction (vs early introduction) of the carbon-carbon π-bond. (Prostaglandins-15) O
O
O
O
For several relevant papers see: Bartlett, P. A. “Synthesis of β-acylacrylic esters and α,β-butenolides via β-keto sulfoxide alkylation” J. Am. Chem. Soc. 1976, 98, 3305–3312. Kende, A. S.; Toder, B. H. “Stereochemistry of deconjugative alkylation of ester dienolates. Stereospecific total synthesis of the litsenolides” J. Org. Chem. 1982, 47, 163–167. Yao, Z.-J.; Wu, Y.-L. “Total Synthesis of (10ξ,15R,16S,19S,20S,34R)-corossoline” Tetrahedron Lett. 1994, 35, 157–160. 18. Provide a mechanistic explanation for the following observation. (Prostaglandins-17) HOCH2CH2OH O O
p-TsOH O
PhH, ∆
For references see: Babler, J. H.; Malek, N. C.; Coghlan, M. J. “Selective hydrolysis of α,β- and β,γ-unsaturated ketals: method for deconjugation of β,β-disubstituted α,β-unsaturated ketones” J. Org. Chem. 1978, 43, 1821–1823. 19. Provide a mechanistic explanation for the following observation [House, H. O.; Trost, B. M. “The chemistry of carbanions. X. The selective alkylation of unsymmetrical ketones.” J. Org. Chem. 1965, 30, 2502–2512. House, H. O.; Trost, B. M. “The chemistry of carbanions. IX. The potassium and lithium enolates derived from cyclic ketones” J. Org. Chem. 1965, 30, 1341–1348.]: H OAc
1. MeLi (2.0 equiv), DME 2. add CH3I (3.7 equiv), -5 °C, 2 min 3. quench with aq. HCl
(1.0 equivalent)
O CH3 H
H
H3 C
3% O
CH3 CH3
H 28%
H
O CH3
H
H3C
CH3 53% O CH3
H3C
CH3 16%
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 133
Organic Synthesis via Examination of Selected Products
Prostaglandins
133
Describe the relevance of this observation to the choice of electrophile for introducing the C8 sidechain in the three-component coupling approach to the prostaglandins. (Prostaglandins-17) 20. Provide a mechanism for the conversion of 153 → 154. (Prostaglandins-20) 21. Provide a mechanism for the isomerization of t-BuN=C to t-BuC≡N during the conversion of 163 → 164. (Prostaglandins-21) 22. Predict the products expected from the following hydroboration-oxidation reactions. (Prostaglandins-23) (Masamune, S.; Kim, B.; Peterson, J. S.; Sato, T.; Veenstra, S. J. “Organoboron compounds in organic synthesis. 1. Asymmetric hydroboration” J. Am. Chem. Soc. 1985, 107, 4549–4951). B H
B H
NaOH, H2O2, H2O
NaOH, H2O2, H2O
23. Each if the following reactions involves transfer of chirality via a sigmatropic rearrangement. Predict the stereochemistry of each product, or explain the stereochemical course of each reaction sequence, using some mechanistic rationale. (Prostaglandins-24) O
O CO2 Et
CO2 Et 1. Et3N, p-TolSCl 2. (MeO) 3P, MeOH
C5 H11
Mechanism ?
OH
OH C 5H 11
Miller, J. G.; Kurz, W.; Untch, K. G.; Stork, G. “Highly Stereoselective Total Syntheses of Prostaglandins via Stereospecific Sulfenate-Sulfoxide Transformations. 13-cis-15β-Prostaglandins E1 to Prostaglandins E1” J. Am. Chem. Soc. 1974, 96, 6774–6775. Me
1. LDA, THF-HMPA Me
2. (i-Pr)3SiCl 3. 200 °C, dodecane,140 min
H
Me
Me
Me O
Me O
4. KF, H2O 5. CH2N2
Mechanism ?
MeO2C
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 134
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
134
Raucher, S.; Chi, K. W.; Hwang, K. J.; Burks, J. E. “Synthesis via Sigmatropic Rearrangement. 12. Total Synthesis of (±)Dihydrocostunolide via Tandem Cope-Claisen Rearrangement” J. Org. Chem. 1986, 51, 5503–5505. CMe3 Me 3C
1. Tf OCH 2 CO 2Et N
N
2. DBU, THF Ph
Tf = triflate (CF3 SO2 )
CO2Et
Ph
Mechanism and Stereochemistry?
Vedejs, E.; Arco, M. J.; Renga, J. M. “Conformational Control of Olefin Geometry in 2,3-Sigmatropic Ring Expansion” Tetrahedron Lett. 1978, 19, 523–526. O
(MeO)3 CCH3
O
MeO2 C
Product ?
O
CH 3CH2 CO2H (cat) ∆
OH
Stork, G.; Raucher, S. “Chiral Synthesis of Prostaglandins from Carbohydrates. Synthesis of (+)-15-(S)-prostaglandin A2” J. Am. Chem. Soc. 1976, 98, 1583–1584. 24. Outline a mechanism for the following transformation (note that the sequence must involve 1,4-asymmetric induction following by transfer of chirality to afford a 1,2-stereochemical relationship. (Prostaglandins-24)
H N
NH2 O
SOCl2 imidazole
H
H
CH2Cl2
N
N O
1. PhMgBr, THF, CH2Cl2 S O
2. (MeO)3P, MeOH, ∆
OH
H
H
N
NH O
Heintzelman, G. R.; Fang, W. K.; Keen. S. P.; Wallace, G. A.; Weinreb, S. M. “Stereoselective total synthesis of the cyanobacterial hepatotoxin 7-epicylindrospermopsin: revision of the stereochemistry of cylindrospermopsin” J. Am. Chem. Soc. 2001, 123, 8851–8853. Heintzelman, G. R.; Parvez, M.; Weinreb, S. M. “A model study on
b1026_Chapter-03.qxd
12/21/2010 b1026
10:55 AM
Page 135
Organic Synthesis via Examination of Selected Products
Prostaglandins
135
the synthesis of the marine hepatotoxin cylindrospermposin” Synlett 1993, 551–552. 25. Outline two syntheses of the diastereomer of 208 that use the strategy outlined in Prostaglandins-26. 26. Suggest a mechanism for the conversion of 214 → 216. (Prostaglandins-27)
b1026_Chapter-04.qxd
138
Pyrrolizidine Alkaloids
OH
HO
1
H
OH
HO
H
OH
HO
OH
H
7a
N
N
N
1
2
3
4
dihydroxyheliotridane
turnef orcidine
platynecine
The pyrrolizidine alkaloids pose a number of synthetic challenges and have served as a testing ground for synthetic methodology. One problem that appears within the targets shown above is the problem of vicinal stereocontrol. Since all of the stereogenic centers are located on ring carbons, the relationship between the pyrrolizidine vicinal stereochemistry and the steroid side chain problem might not be obvious. The following "plan" for the synthesis of pyrrolizidine alkaloids, however, reveals this relationship.
H
OH
HO
OH
H
HO
OH
H
HO
1
N 1
4
7a
N
5
3
O
Amides are frequently used as amine precursors in alkaloid synthesis. They are easy to prepare and they are more "friendly" than amines. For example amines are easily oxidized and amides are not easily oxidized.
N
OH E
H 7
CO2R
N
H
E OH
1
N H
H E E = CO2R
7
6
E
HO H 7a
H 8
If we ignore the C 7-OH, this problem reduces to adding an acetic acid residue and an amine "anti" across a cis double bond. E E
One way of accomplishing the require "anti-addition" is via the geminally activated cyclopropane chemistry that was used to address the vicinal stereochemistry problem within the context of terpenoids and the prostaglandins.
E
H HO
OH
H 7a
1
NH2
NH2 10
Pyrrolizidines-1
HO
H 7a 1
E
H
9
OH
Page 138
HO
10:55 AM
hastanecine
Organic Synthesis via Examination of Selected Products
N
12/21/2010
H
7
b1026
HO
Organic Synthesis via Examination of Selected Natural Products
Pyrrolizidine alkaloids are a large f amily of compounds that contain the pyrrolizidine substructure, usually decorated with a variety of hydroxyl groups and frequently with a macrocyclic structure bridging the C 1 and C 7 positions. Derivatives have been studied f or use in cancer chemotherapy, but these studies have been discouraging due to toxicity problems. Four isomeric alkaloids that are representative of this f amily of compounds are shown below.
b1026_Chapter-04.qxd
12/21/2010 b1026
10:55 AM
Page 139
Organic Synthesis via Examination of Selected Products
Pyrrolizidine Alkaloids
139
Pyrrolizidines-1 Pyrrolizidine alkaloids are a large family of compounds that contain the pyrrolizidine substructure, usually decorated with a variety of hydroxyl groups, and frequently with a macrocycle bridging the C1 and C7 positions.1 Derivatives have been studied for use in cancer chemotherapy, but these studies have been discouraging due to toxicity problems. Four isomeric alkaloids that are representative of this family of compounds (1–4) are shown in Pyrrolizidines-1. The pyrrolizidine alkaloids pose a number of synthetic challenges and have served as a testing ground for synthetic methodology. One problem that appears within these targets is that of vicinal stereocontrol. Since all of the stereogenic centers are located on ring carbons, the relationship between the pyrrolizidine vicinal stereochemistry problem and the steroid side chain problem might not be obvious. The following “plan” for the synthesis of pyrrolizidine alkaloids, however, reveals this relationship. Consider hastanecine (1). A “last step” in a possible synthesis of this alkaloid would be reduction of lactam 5. It is notable that amides are frequently used as amine precursors in alkaloid synthesis. They are easy to prepare and are more “user friendly” than amines. For example, amines are easily oxidized and amides are not. The “friendliness” is directly related to the nitrogen lone pair being involved in bonding to the adjacent carbonyl carbon (in amides) and thus, not being as nucleophilic as the nitrogen lone pair of the corresponding amine. Continuing with the plan, one precursor of lactam 5 would be amino ester 6. The relationship between the C1-C7a vicinal stereochemistry problem and the terpenoid sidechain problem emerges when we consider this structure. The stereogenic center at C1 is exocyclic to the ring stereogenic center at C7a. If we ignore the C7-OH stereogenic center, and we recall the discussion presented in Prostaglandins-25, one solution to this vicinal stereochemistry problem is addition of an acetic acid residue and an amine “anti” across a cis double bond. One way of accomplishing this required “anti-addition” would be via the geminally activated cyclopropane chemistry used to address the vicinal stereochemistry problem within the context of terpenoids and the prostaglandins. Thus, cyclopropanation of 9 (or derivative thereof) followed by amine-opening of geminally activated cyclopropane 10 would give 8 (7) which could be then moved on toward hastanecine. Of course, relative asymmetric induction would be needed in the conversion of 9 to 10. Let’s see how this problem was handled in practice.
PhthN
Pd/BaSO4
OTHP
7
13
PhthN
HO
H2, quinoline
OTHP
7
PhthN
14
54%
Amine protected as imide racemic MeO2CCH2COCl
H
O
MeO2C H
7a
18
O H
MeO2C H
O
H
N2
Cu-bronze O
7a
OTHP
7
NPhth
48% overall
O H
7
Not formed
1
110 oC
OTHP NPhth
1
O O
17
MeO2C TsN3, Et3N
1
PhthN
OTHP
O
16
OTHP
7
PhthN 15
PhthNCH2CH2 on convex face of [3.1.0]-system
C7-OH used to direct cyclopropanation and establish vicinal C7-C7a relationship
PhthNCH2CH2 on concave face of [3.1.0]-system
NH2NH2 HO
OTHP
H
1
7
N
3
O
4
O 19
NHNH2
1. 10% aq. HCl ∆, 12 h 2. NaOMe, MeOH 3. Ac2O, pyridine
AcO
OAc
H N
HO LiAlH4
3
20 O 70%
Pyrrolizidines-2
81%
OH
H 1 7
7a
N 1 dl-hastanecine
Use of trans-olefin provides dihydroxyheliotridane
Page 140
MeO2C
10:55 AM
12 72%
11
O
1
Organic Synthesis via Examination of Selected Products
MeOH, -78 oC
HO
ClMg C C
12/21/2010
N
CHO
b1026
1
OTHP O3, CH2Cl2
Organic Synthesis via Examination of Selected Natural Products
Danishefsky, S.; McKee, R.; Singh, R. K. "Stereospecific Total Synthesis of dl-hastanecine and dl-dihydroxyheliotridane" J. Am. Chem. Soc. 1977, 99, 7711. Also see: Danishefsky, S.; McKee, R.; Singh, R. K. "Kinetically Controlled Total Syntheses of dl-Trachelanthamidine and dl-Isoretronecananol" J. Am. Chem. Soc. 1976, 98, 4783-4788.
O
b1026_Chapter-04.qxd
140
Danishefsky Synthesis of Hastanecine
b1026_Chapter-04.qxd
12/21/2010 b1026
10:55 AM
Page 141
Organic Synthesis via Examination of Selected Products
Pyrrolizidine Alkaloids
141
Pyrrolizidines-2 Relative stereochemistry between C7 and C7a-C1 was to be established using an intramolecular cyclopropanation. The hope was that the carbene derived from diazoester 16 would undergo cyclopropanation to provide 18 rather than 17. Cyclopropanes 17 and 18 are diastereomers and 18 was expected to be the more stable isomer. Why? These isomers differ in the relationship of the C7 β-imidoethyl chain to the “fold” of the oxabicyclo[3.1.0]hexane ring system. In 18, the sidechain is on the convex face of this ring system. In 17, the sidechain is on the sterically more conjested concave face of the [3.1.0] ring system. For this reason it was assumed that 18 would be more stable than 17, and it was hoped that some of the product energy difference would be felt in the transition states leading to these products, favoring formation of 18. Carbene precursor 16 was prepared from phthalimide 11. Ozonolysis of the olefin gave aldehyde 12. An acetylide addition gave 13 as a racemic mixture. Semihydrogenation of the alkyne, esterfication of 14 with methyl malonyl chloride, and diazotization of mixed malonate 15 gave 16. The carbenoid insertion reaction proceeded according to plan to give 18. The phthalimido group was removed using hydrazine, liberating N4 as a nucleophilic amine. This was accompanied by hydrazinolysis of the lactone to liberate the C7 hydroxyl group. Opening of the geminally activated cyclopropane and cyclization of the intermediate aminoester provided 19. This material was treated with aqueous acid to remove the THP protecting group, subjected to methanolysis and decarboxylation (retro crossed-Claisen condensation) to transform the malonic acid substructure to an acetic acid unit, and esterification to provide lactam 20. The synthesis of hastanecine (1) was completed by LAH reduction of the lactam.
b1026_Chapter-04.qxd
142
Cycloaddition Approaches to Pyrrolizidine Alkaloids
OH
H
O
1
7 2
N
O
1 2
7
OH
2
7
OH
X
Reduce
2
OH
24
2 3
N O nitrone
OH
25
retronecine
Regiochemistry? Stereochemistry? 1
(MeO)3CH
7
N
HCl, MeOH
MeO
OMe 1. H2O2
7
N
OMe
MeO
HgO
MeO
N
HO
CO2Me Me
7
7
2. ∆
OMe N
OH
2
7
OH
O
N
O
CHCl3, 40 oC
O
3
O E
Me 28
27
MeO H MeO 7
CO2Me 1
7a
N
2
OH
Cope Elimination via N-oxide
MeO H MeO NH
3
84%
CO2Me Pd/C, H2
MeO H MeO
OH MeOH
3
OMs
34
E = CO2Me
30
29
33
CO2Me 3
N O 32
OMs
MsCl, Et3N
MeO H MeO
99%
CO2Me
N O 31 86%
Pyrrolizidines-3
3
1
7a
2
OH
Page 142
26 O
26
10:55 AM
23
22
7a
7
7a
N O
3
21
1
1
7a
NH
CO2R
O
CO2R
H
1
7a
N
O
CO2R
H
Organic Synthesis via Examination of Selected Products
7a
CO2R
H
b1026
HO
12/21/2010
dipolarophile Reduce
Organic Synthesis via Examination of Selected Natural Products
Nitrone Cycloadditions
b1026_Chapter-04.qxd
12/21/2010 b1026
10:55 AM
Page 143
Organic Synthesis via Examination of Selected Products
Pyrrolizidine Alkaloids
143
Pyrrolizidines-3 Now let’s take a little side journey into the field of alkaloid synthesis. Alkaloids are loosely defined as natural products that contain a basic nitrogen. Some peptides that contain basic nitrogen (lysine and imidazole residues for example) are not classifed as alkaloids, and some amides that do not contain basic nitrogen (colchicine for example) have been called alkaloids, and that is why I use the phrase “loosely defined”.2,3 I have already mentioned that the pyrrolizidine alkaloids are popular synthetic targets and thus, at this point we will examine a few other approaches to these natural products. Retronecine (21) is a pyrrolizidine alkaloid that has only two stereogenic centers, but does contain unsaturation in one of the 5-membered rings. It has been a popular synthetic target, in part because of its biological activity, as has been its C7 epimer heliotridine (see 38 on Pyrrolizidines-4). We will look at three approaches, all of which revolve around cycloaddition chemistry. The first of these revolves around nitrone 1,3-dipolar cycloaddition chemistry.4 The plan was to construct 21 from key intermediate 22. This transformation requires reduction of a ketone at C7 and an ester appended to C1, and a dehydration reaction to introduce the C1-C2 double bond. Intermediate 22 was to be prepared from 23 via an intramolecular N-alkylation. Notice that the synthesis of the unsaturated ring of retronecine was to follow the aforementioned strategy: make the σ-bond first and then introduce the π-bond (Prostaglandins-15). Intermediate 23 contains a 1,3-NO relationship: C7a is bonded to a nitrogen and C2 is bonded to an oxygen. This “difunctional relationship” is a structural feature common to many alkaloids (we will see why this is so in Chapter 8). Thus, synthetic methods that establish a 1,3-NO relationship are of “general” importance to the field of alkaloid synthesis. One such method is the dipolar cycloaddition of a nitrone (an azomethine oxide) to an olefin. This is the reaction around which the Tufariello group designed their approach to the pyrrolizidine alkaloids.5 The specific plan was to examine the reaction of nitrone 25 with dipolarophile 26 with the hope of obtaining cycloadduct 24 as the major product. Reduction of the weak nitrogen-oxygen bond was then to reveal an intermediate of type 23. The regiochemical and stereochemical course of the cycloaddition was critical to the success of the plan. It turned out that 25 was an impractical 1,3-dipole for implementation of the plan, but ketal 30 served admirably. This 1,3-dipole was prepared by conversion of 27 to the corresponding ketal 28. Oxidation of 28 to the amine oxide, followed by a Cope elimination, gave hydroxylamine derivative 29.
b1026_Chapter-04.qxd
144
MeO H MeO
CO2Me MsCl, Et3N
MeO H MeO
OH
N
1
O
CO2Me H2O, DME 2
N
HO
CO2Me
H
NaBH4 7
N
HCl
N
MeOH
37
81%
HO
21 Retronecine
RO
OR
OR
H
RO
7 4
NH 4
3
O
38
39
N
3
O
X
7a 2
7a
7
N
OR RO
H1
O
7
N
3
O
40
Heliotridine
THPO 1. (CH2=CH)2CuLi
Diastereoselectivity? Hope C7-OR will induce stereochemistry by sitting on convex face of [4.3.0] system
THPO
CO2Me
OH
THPO
THPO
i-Bu2AlH
MnO2
Et2O
Celite
CHO labile and used immediately
2. MeOH CO2Me 42
O 41
benzene
0 oC 43
44
100%
Pyrrolizidines-4
45
Page 144
OH
H
OH 1
N
Acylnitroso Diels-Alder
HO
H
10:55 AM
No yields for last 2 steps. Note use of nitrone cycloaddition to construct 1,3-NO relationship. This is an important one in alkaloids and we will return to this later.
Organic Synthesis via Examination of Selected Products
Tufariello, J. J.; Lee, G. E. "Functionalized Nitrones. A Highly Stereoselective and Regioselective Synthesis of dl-Retronecine" J. Am. Chem. Soc. 1980, 102, 373.
b1026
AlH3 THF
12/21/2010
36
Organic Synthesis via Examination of Selected Natural Products
35 98%
34
CO2Me
H 7
b1026_Chapter-04.qxd
12/21/2010 b1026
10:55 AM
Page 145
Organic Synthesis via Examination of Selected Products
Pyrrolizidine Alkaloids
145
Oxidation of 29 provided nitrone 30 which, in the presence of methyl γ-hydroxycrotonate underwent cycloaddition with the desired regiochemistry and stereochemistry to provide 31. The C3 hydroxyl group was then activated for the subsequent intramolecular N-alkylation by formation of mesylate 32. Hydrogenolysis of the N–O bond was followed by in situ N-alkylation to give 34 (a structural equivalent of 22).
Pyrrolizidines-4 The synthesis of retronecine was completed using a 4-reaction sequence. The C3 hydroxyl group of 34 was converted to a mesylate which underwent β-elimination to introduce the C1-C2 olefin. The acetal of 35 was hydrolyzed, and sodium borohydride reduction of the C7 ketone from the convex face provided 37. 1,2-Reduction of the unsaturated ester using alane finished the synthesis of retronecine (21). The Keck group described an approach to the pyrrolizidine alkaloids that revolved around acylnitroso Diels-Alder chemistry. The plan, outlined within the context of an approach to heliotridine (38), was to once again use an intramolecular N-alkylation to construct the N4–C3 bond. N-Alkylation substrate 39 was to be prepared by reduction of the N–O bond of 40, which was to result from an intramolecular cycloaddition of 41. It was hoped that the C7 substituent might induce relative stereochemistry at C7a by occupying a site on the convex face of the incipient azaoxabicyclo[4.3.0]nonane ring system in the cycloaddition transition state. Note that the cycloaddition of 41 establishes the olefin geometry needed to support the intramolecular N-alkylation. One of the key features of this synthesis was the development of tactics for the generation of the highly functionalized N-acylnitroso compound 41. It was well known that such compounds behaved as dienophiles in 4+2 cycloaddition reactions, but it was also known that they were extremely reactive and had to be generated in the presence of the diene partner, ready for immediate use. The synthesis began with alkynoate 42, which was converted to 43 using a cuprate conjugate addition. Reduction of the ester gave 44 and oxidation of the allylic alcohol using manganese dioxide provided aldehyde 45, which was used immediately.
b1026_Chapter-04.qxd
146
THPO O CH3CONHOH
OTBS O
1. LDA, THF-HMPA (4:1) 2. RCHO (45)
O N
O
6
7
N
1. MsCl, Et3N CH2Cl2, 0 oC TBSO 5 min
N
NH 2. LDA, THF
O
H
OTHP
N
1. MeOH, ∆ PPTS, 3h
HO
N O
53
N H
TsO
7
6
N
O
O
O
O 49
50 86% as 1.3:1 mixture of diastereomers. Early transition state?
OH
HO
H 7
N
THF
54
38
100% PPTS =
7a
N
LiAlH4
2. TBAF (2 eq)
O
H
86%
Heliotridine
TBAF = n-Bu4N F
Same route applied to other diastereomer to afford retronecine
Pyrrolizidines-5
OH
Keck, G. E.; Nickell, D. G. "Synthetic Studies on Pyrrolizidine Alkaloids. 1. (dl)-Heliotridine and (dl)-Retronecine via Intramolecular Dienophile Transfer" J. Am. Chem. Soc. 1980, 102, 3632.
Page 146
1. Lactam to amine difficult due to lability of functionality to vigorous conditions 2. Allylic activation NOT enjoyed by system (SN2 trajectory)
OTHP TBSO
H 7
90%
51
+
OTHP TBSO
EtOH, 0 oC, 5h OH
O
52
TBSO
OTHP 6% Na/Hg Na2HPO4
H
10:55 AM
OTHP
Organic Synthesis via Examination of Selected Products
H
12/21/2010
PhH, 80 oC 4.5 h
Competing ene reaction
b1026
64% from dienol
47
Organic Synthesis via Examination of Selected Natural Products
46
TBSO
48
3. t-BuMe2SiCl, DMF imidazole, 23 oC, 12h
Bu4N IO4
b1026_Chapter-04.qxd
12/21/2010 b1026
10:55 AM
Page 147
Organic Synthesis via Examination of Selected Products
Pyrrolizidine Alkaloids
147
Pyrrolizidines-5 The reactivity of the acylnitroso group was “packaged” in the form of 9,10dimethylanthracene cycloadduct 47.6 Conversion of this hydroxamic acid derivative to the corresponding enolate, followed by reaction with aldehyde 45 and protection of the resulting alcohol, gave 48 as a stable and easily characterized intermediate. Warming a benzene solution of 48 gave 9,10-dimethylanthracene and the desired intramolecular acylnitroso DielsAlder cycloadduct 50 as a mixture of diastereomers at C7a. The key intermediate 49 was presumably generated by a retro-Diels-Alder reaction. The level of diastereoselection in the cycloaddition was disappointing, and perhaps indicative of an early cyclization transition state in which the aforementioned steric effects were not well developed. Reduction of the N–O bond in 50 was followed by mesylate formation and N-alkylation to afford 52 and 53 as a separable pair of diastereomers. The synthesis of heliotridine (38) was completed by hydrolysis of the THP group, removal of the silicon protecting group from the C7 alcohol, and reduction of lactam 54. Subjecting 52 to the identical reaction sequence provided retronecine (21). Notice that as in the Danishefsky synthesis (and unlike the Tufariello synthesis) the basic pyrrolizidine nitrogen was packaged as a lactam until the last step of the synthesis.
b1026_Chapter-04.qxd
148
no C=C because of potential B-ring aromatization
Azomethine Ylid Cycloaddition
57 OH
RO
CO2 R
21
N
B
OMe
CO2 R 58
N
N 56
55
O H
N
N
NaHCO 3 60
N
DABCO SiMe3
TfO
O
62
1. LDA 2. MoO 5-pyr-HMPA
RO O N
68%
63
SiMe3
SiMe3
61 BnO
1. NBS 2. Et4N OAc 3. Me 3SiCH 2 OTf 4. DABCO
CO2 Me
70% overall (more practical on large scale) 8%
N
BnO H
R=H R = Bn
MeOSO2 CF3
+ 1. LDA 2. (PhSe)2
BnBr, NaH DME 90%
CH2 Cl2 , 20 oC, 18h CO2 Me
N
1. H2 , Pd/C EtOAc 2. 48h, neat
BnO
CO2 Me N
83%
51%
67
66
BnO OR CO2 R N R = Me 65
CH2 =CHCO 2Me
BnO OMe N
CsF, DME 20 o C
TfO
SiMe3
64
Vedejs, E.; Martinez, G. R. "Stereospecific Synthesis of Retronecine by Imidate Methylide Cycloaddition" J. Am. Chem. Soc. 1980, 102, 7993.
Pyrrolizidines-6
Page 148
N 59
N
OMe
OMe Me SiCH OTf 3 2
(MeO) 3O BF4
10:55 AM
N
Organic Synthesis via Examination of Selected Products
azomethine ylid
Retronecine
b1026
A
N
RO
BnO OR CO2 R
12/21/2010
H
Organic Synthesis via Examination of Selected Natural Products
HO
b1026_Chapter-04.qxd
12/21/2010 b1026
10:55 AM
Page 149
Organic Synthesis via Examination of Selected Products
Pyrrolizidine Alkaloids
149
Pyrrolizidines-6 The final pyrrolizidine synthesis we will consider is Vedejs’ approach to retronecine. This synthesis relies on a dipolar cycloaddition reaction for direct construction of the pyrrolizidine nucleus. The key reaction was to be a 1,3dipolar cycloaddition between an azomethine ylid of type 57 and an acrylate of type 58. It was projected that this would provide 56 (or 57 after loss of methanol), which would then be converted to retronecine (21) via a sequence involving late-stage introduction of the C1-C2 double bond. The synthesis began with the conversion of lactam 59 to imidate 60. N-Alkylation using trimethylsilylmethyl triflate, followed by demethylation of 61 with DABCO, gave 62. Hydroxylation of the enolate derived from 62, followed by Williamson etherification, gave 63 (R=Bn). Alternative tactics for moving forward from 60, that were more amenable to scale-up, were also developed and are shown without comment. O-Alkylation of 63 provided 64, the immediate precursor of azomethine ylid 57 (where R=Bn). Treatment of 64 with CsF in the presence of excess methyl acrylate gave 66, presumably via intermediate cycloadduct 65. The regiochemical course of the cycloaddition was precedented and stereochemistry was not an issue, given that the C1 and C7a stereogenic centers in 65 are destroyed in the subsequent elimination of methanol. Catalytic hydrogenation of 66, followed by epimerization at C1, gave 67 with good control of stereochemistry. The C1-C2 olefin was then introduced. Enolate formation and selenenylation gave 68.
10:55 AM
Page 150
Organic Synthesis via Examination of Selected Products
12/21/2010 b1026
Retronecine
Organic Synthesis via Examination of Selected Natural Products
90%
85%
70%
2. NH4Cl
OH
b1026_Chapter-04.qxd
150
69
21 selenoxide elimination
N
70 68
Pyrrolizidines-7
N N 3. ∆, CCl4 N
1. Li, NH3
HO H OH BnO H BnO H
CO2Me
i-Bu2AlH SePh
1. MCPBA (2 eq) 2. Me2S CO2Me BnO H
b1026_Chapter-04.qxd
12/21/2010 b1026
10:55 AM
Page 151
Organic Synthesis via Examination of Selected Products
Pyrrolizidine Alkaloids
151
Pyrrolizidines-7 Oxidation of 68, followed by elimination of the intermediate selenoxide, provided 69. Reduction of the ester and removal of the benzyl protecting group from the C7 alcohol completed the synthesis of retronecine (21). I note that my own personal interest in the Vedejs synthesis arises in part because the student who did the laboratory work (Greg Martinez) was an undergraduate student at UC Berkeley that I worked with closely for about a year when I was a graduate student. The aforementioned syntheses are examples of research that was designed to develop and test synthetic methodology. My opinion is that the lasting value of these syntheses is the chemistry and stategies that were developed within the context of a target-oriented excercise, not the supply of material for biological evaluation or the determination of a structure. These strategies have all seen use in other settings. Now let’s return to the “terpenoid sidechain problem”.
b1026_Chapter-04.qxd
12/21/2010 b1026
152
10:55 AM
Page 152
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
References and Notes 1. Yoda, H. “Recent Advances in the Synthesis of Naturally Occuring Polyhydroxylated Alkaloids” Current Organic Chemistry 2002, 6, 223–243. Casiraghi, G.; Zanardi, F.; Rassu, G.; Pinna, L. “Recent Advances in the Stereoselective Synthesis of Hydroxylated Pyrrolizidines. A Review” Org. Prep. Procedures Int. 1996, 28, 641–682. Stevens, R. V. “General Methods of Alkaloids Synthesis” Acc. Chem. Res. 1977, 10, 193–198. Stevens, R. V. “Alkaloid Synthesis” Total Synth. Nat. Prod. 1977, 3, 439–554. 2. For one view of what constitutes an alkaloid see: Pelletier, S. W. “The Nature and Definition of an Alkaloid” Alkaloids: Chemical and Biological Perspectives, Pelletier, S. W., Ed.; John Wiley and Sons, 1982, Vol. 1, pp 1–31. 3. There are many excellent monographs on alkaloid chemistry. For a thorough series see “Alkaloids” Academic Press; 1950–2007 (68 volumes). 4. Huisgen, R.; “1,3-Dipolar Cycloadditions” Angew. Chem. 1963, 75, 604–637. 5. Banerji, A.; Bandyopadhyay, D. “Recent Advances in the 1,3-Dipolar Cycloaddition Reactions of Nitrones” J. Ind. Chem. Soc. 2004, 81, 817–832. Black, D. St. Clair.; Crozier, R. F.; Davis, V. C. “1,3-Dipolar Cycloaddition Reactions of Nitrones” Synthesis 1975, 205–221. 6. Kibayashi, C.; Aoyagi, S. “Nitrogenous Natural Products via N-Acylnitroso Diels-Alder Methodology” Synlett 1995, 873–879. Corrie, J. E. T.; Kirby, G. W.; Sharma, R. P. “Formation of Aryl Isocyanates by Deoxygenation of Nitrosocarbonylarenes” J. Chem. Soc., Chem. Commun. 1975, 915–916. Kirby, G. W.; Sweeny, J. G. “Nitrosocarbonyl Compounds as Intermediates in the Oxidative Cleavage of Hydroxamic Acids” J. Chem. Soc., Chem. Commun. 1973, 704–705.
b1026_Chapter-04.qxd
12/21/2010 b1026
10:55 AM
Page 153
Organic Synthesis via Examination of Selected Products
Pyrrolizidine Alkaloids
153
Problems 1. Use of the trans-isomer of 14 in this reaction sequence gave dldihydroxyheliotridane (2). Outline a synthesis of trans-14 and provide the structure of intermediates en route from this olefin to 2. This synthesis of 2 also gave a small amount of dl-platynecine (4). What is the origin of this minor product? (Pyrrolizidines-2) 2. Propose mechanisms for the conversions of (1) 28 → 29 and (2) 29 → 30. (Pyrrolizidines-3) 3. Provide the structures of all dipolar cycloaddition products that would have been produced had the reaction of 26 with 30 shown no regioselectivity or stereoselectivity. (Pyrrolizidines-3) 4. Explain the stereochemical course of the conversion of (1) 36 → 37 and (2) 42 → 43. (Pyrrolizidines-4) 5. The conversion of 51 → 52 did not enjoy the “allylic activation” (N-alkylation rate enhancement) that might have been anticipated. Provide an explanation for this observation. (Pyrrolizidines-5) 6. Aside from being an appropriate nucleophile for the demethylation of 61, DABCO is nice for following the course of amine-promoted reactions in an NMR tube [compared with triethylamine, DBU, DBN or Hunig’s base (i-Pr2EtN)]. Why? (Pyrrolizidines-6)
N N DABCO
N N
N
DBN
N
N DBU
Hunig’s Base (DIPEA)
7. Provide the structure of intermediates en route from 60 → 63 via the “alternative tactics” that begin with NBS. (Pyrrolizidines-6) 8. Provide a rationale for the stereochemical course of the conversion of 66 → 67. (Pyrrolizidines-6) 9. What is a problem that one might have anticipated in the reduction of 69 → 70? (Pyrrolizidines-7) 10. Suppose you made the following observation [Heiba, E-A. I.; Dessau, R. M. “Free-radical isomerization. I. Novel rearrangement of vinyl radicals” J. Am. Chem. Soc. 1967, 89, 3772–3777]. How might you
b1026_Chapter-04.qxd
12/21/2010 b1026
154
10:55 AM
Page 154
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
develop this into an approach to the pyrrolizidine alkaloids (see text and papers cited therein for some targets)? Cl (PhCO)2O
Cl
CCl4 HC C(CH2) 2CH3
40%
∆
11. Suppose you made the following observation [Schoemaker, H. E.; Dijkink, J.; Speckamp, W. N. “Biomimetic α-acylimmonium cyclizations of unactivated olefins” Tetrahedron 1978, 34, 163–172. How might you develop this into an approach to the pyrrolizidine alkaloids? OEt HCO2H
OHCO
H
H 72%
N
N
O
O
12. Predict the product of the following reaction [Smith, R.; Livinghouse, T. “An Expedient Synthetic Approach to the Physostigmine Alkaloids via Intramolecular Formamidine Ylide Cycloadditions” J. Org. Chem. 1983, 48, 1554–1555]. MeO2C Me3Si
CO2Me N SPh
p-NO2C6H4COF CH3CN 60 °C, 3h
b1026_Chapter-05.qxd
156
A Solution to the Cholesterol Side Chain Problem: Ester Alkylation Wicha, J. "Synthesis of 25-Hydroxycholesterol from 3β-Hydroxyandrost-5-en-17-one. A Method for Stereospecif ic Construction of a Sterol Side-chain" J . Chem. Soc., Chem. Commun. 1975, 968-969
H
H
2. RBr, HPMA
H
H
O
H 3. LiAlH4
THPO
H
O
H
THPO 2
1
3 1. Hydrolysis 2. Ac2 O, pyridine
66% (clean stereochemistry) 20
H
OH
MeMgBr
H
H
H
HO
AcO 5
4
53% overall from starting ester
Explanation
General Model RL
RL H
13 17
6
H H
H
O OMe
13 17
H
1. Minimize Allylic Strain 2. Minimize Torsional Strain (product born with everything staggered)
Ra RM
O
(OR) O OR (O ) RS
OMe
H R
R X
O H
8 7 Electrophile
Juvabione-1
CO2 R
Rα RM
E
RS
9 Werner, B.; Fleming I.; Waterson, D. "Diastereoselectivity in the Alkylation of Enolates Having an Adjacent Silyl Group" J. Chem. Soc., Chem. Commun. 1984, 28. McGarvey, G. J.; Williams, J. M. "Stereoelectronic Controlling Features of Allylic Asymmetry. Application to Ester Enolate Alkylations" J. Am. Chem. Soc. 1985, 107, 1435.
Page 156
H
25-Hydroxycholesterol is a key intermediate in the synthesis of 1α,25-dihydroxyvitamin-D3
10:55 AM
THPO
O
Organic Synthesis via Examination of Selected Products
H
16
12/21/2010
H
1. LiAlH4 2. TsCl, pyridine
O
b1026
17 13
MeO2 C 1. LDA (3 eq) THF, -78 °C
Organic Synthesis via Examination of Selected Natural Products
CO2 Me
20
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 157
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
157
Juvabione-1 In this chapter we will take a look at several syntheses of a sesquiterpene ester called juvabione. We will see that it became a target for synthesis because of structural and biological issues, and became a “test molecule” for the evaluating methodology for controlling vicinal stereochemistry in acyclic systems. But first let’s once again examine the steroid sidechain problem. One of the simplest solutions was developed by Jerzy Wicha (Poland). He found that alkylation of the enolate derived from 1 gave 2 with a high level of stereocontrol at C20. The explanation for the high level of stereoselectivity was not obvious at the time, but work from the groups of Fleming and McGarvey provided an explanation several years later. The explanation is based on the following notions: (1) There is a size difference between the three substituents on the carbon β to the enolate carbonyl group. In the case of 1 this is C17. One group is small (RS=H), another medium (RM=CH2 at C16) and the third group is large (RL=C13 mimics a tert-butyl group in size). (2) The rate of alkylation is fastest from conformation 6. This conformation is arguably the most stable conformation of the enolate. Allylic strain is minimized. The large group (C13) is orthogonal to the plane defined by the enolate π-system. The medium-sized group (C16) is positioned adjacent to the vicinal hydrogen. The small group (H) is positioned nearly in the same plane as the enolate oxygen. (3) Alkylation takes place such that as the new carbon-carbon bond is formed, everything is staggered along the C17-C20 bond (torsional strain is minimized in the alkylation transition state). Thus 6 leads directly to 7. In a more general sense, systems of type 8 will provide products of type 9.
b1026_Chapter-05.qxd
158
Additional Examples of Side Chain via Alkylation
11
H
H
THPO
H HO
12
H 13
80%
C
D
OH
H 10 A HO
OH
1α,25-dihydroxyvitamin-D3
Juvabione-2
Page 158
Kurek-Tyrlik, A.; Minksztym, K.; Wicha, J. "Synthesis of (23R)- and (23S)-23H-Isocalysterols. The First Synthesis of a Representative of Marine Sterols with a Cyclopropene Moiety in the Side Chain" J. Am. Chem. Soc. 1995, 117,1849.
10:55 AM
H
H
12/21/2010
OMe
H
2. RBr, HPMA
Organic Synthesis via Examination of Selected Products
H
Br
b1026
H
MeO2C 1. LDA (3 eq) THF, -78 °C
Organic Synthesis via Examination of Selected Natural Products
CO2Me
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 159
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
159
Juvabione-2 The original work of Wicha was used to prepare a precursor of 1α,25dihydroxyvitamin-D3 (10), but it has been applied to other steroids (for example 13) as well. This is a simple and effective strategy for controlling C17-C20 relative stereochemistry in steroids. The solution relies on what is called 1,2-asymmetric induction. One stereogenic center induces stereochemistry in the formation of an adjacent stereogenic center. Whereas this “solution” works in a well-defined set of situations, it is not a universal solution. Let’s see why.
H R R
erythro-Juvabione
threo-Juvabione
Mori: First Synthesis
CO2Me
O
OMe Anisole will serve as source of cyclohexanone. Ketone will serve as functional handle for side chain.
OH
O
O 16 Not Stereoselective
17
18
Carbethoxylation will generate diastereomeric pair
Juvabione-3
Page 160
Mori, K.; Matsui, M.; "Synthesis of dl-Juvabione (Methyl dl-Todomatuate), A Sesquiterpene Ester with Juvenile Hormone Activity" Tetrahedron Lett. 1967, 2515-2518.
10:55 AM
H 15
14
Organic Synthesis via Examination of Selected Products
H
O
12/21/2010
O
b1026
Sidechain Problem: Control of Vicinal Stereochemistry
H R S
b1026_Chapter-05.qxd
CO2Me
Organic Synthesis via Examination of Selected Natural Products
CO2Me
160
Juvabione: Vicinal Stereocontrol Problem
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 161
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
161
Juvabione-3 Synthetic organic chemistry became an important tool to the field of entomology in the 1960s when scientists began to learn about small molecules that played an important role in the life cycle of insects.1 Of course, the “practical goal” was to use these chemicals to control insect populations. One family of molecules that became of interest are the “juvenile hormones”. These are compounds that regulate larval development. The “juvabiones” (14 and 15) are diastereomeric sesquiterpene methyl esters that were first isolated from the balsam fir. They contain vicinal cyclic and exocyclic stereogenic centers. Establishing this stereochemical relationship is an issue that must be addressed in any stereoselective synthesis. Whereas initial efforts were developed to supply material for biology, and to help resolve stereochemical issues, these targets became a popular testing-ground for the development and application of reactions that generate vicinal stereochemical relationships. We will spend some time looking at a selection (not all inclusive) of syntheses that I have chosen to make points of some general interest. Let’s start with a question. Can the Wicha strategy be successfully applied to the juvabiones? I am not aware that anyone has tried, but I suspect the answer is “no”. I will leave it for you to explain “why not” (or “why yes” if you disagree with me). The point I want to make here is that no one solution to a given problem will work for all variations of that problem. Thus there is usually a need for multiple solutions to a given problem (in this case a synthesis problem). I hope this “study” of a set of approaches (in some cases solutions) to the same problem is educational. Let’s start with the first total synthesis of the juvabiones reported by the Mori group in the 1960s. The goal here was to produce material. Thus, the synthesis was not stereoselective. The plan was to prepare a mixture of diastereomers and separate at some point. Thus 16 was to be prepared from 17 via dehydration of an intermediate cyanohydrin, tactics that were destined to lack selectivity.
BrCH2CO2Et Zn
3. SOCl2 4. Me2 NH
OAc
1. H2 , Pd/C
oxalic acid
2. AcCl, pyridine
OH
MeOH-H2 O
OH
KOH diglyme-H 2O
Full paper: Mori, K.; Matsui. "Synthesis of Compounds with Juvenile Hormone Activity-I. dl-Juvabione (Methyl dl-Todomatuate)" Tetr ahedr on 1968, 24, 3127-3138.
∆ CO 2H
CO 2H
OH
and
H
H 2O, acetone
O
H O
O H
H 26
CO 2H
CO 2Me CH2 N2
CrO 3, H 2 SO 4
27
14
15
dl-Todomatuic Acid Simple chemistry; Solve stereochemical problem by separation; Correlate stereochemistry to known compounds (todomatuic acid)
11.4 g from 100 g of 18 separated via crystallization of semicarbazones
Juvabione-4
Page 162
22
23
not a stereogenic center
10:55 AM
25
24
Organic Synthesis via Examination of Selected Products
2. POCl3
OMe
O
O
Li, NH 3
b1026
21
Organic Synthesis via Examination of Selected Natural Products
NMe 2
diastereomeric mixture
1. HCN OAc
OH
20
p-methoxyacetophenone
CN
OMe
3. (CH3 )2 CHCH2 MgBr
O
19
18
1. Li(EtO) 3AlH 2. H3 O+
12/21/2010
CO2Et
O
OMe
1. W-7 Raney Ni 2. NaOH
b1026_Chapter-05.qxd
OMe
162
OMe
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 163
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
163
Juvabione-4 The Mori synthesis began with p-methoxyacetophenone (18 = PMA). The acetyl group was converted into the “terpenoid sidechain” using a reliable reaction sequence. Birch reduction of 21, followed by enol ether hydrolysis, reduction of the olefin, and protection of the sidechain alcohol as an acetate, gave 24 as a mixture of diastereomers. Note that the critical vicinal stereochemical relationship was established in the sequence leading from 21 to enone 23, but without any diastereoselectivity. The conversion of 23 to 24 destroyed the vicinal stereochemical relationship. Conversion of 24 to the corresponding cyanohydrin, followed by dehydration to unsaturated nitrile 25, once again introduced the vicinal stereochemistry, but again without selectivity. Hydrolysis of the mixture of nitriles to acids 26, and oxidation of the sidechain alcohol, provided a racemic mixture of the diastereomeric keto acids 27. Ketones 27 were converted to a separable mixture of semicarbazones (upon reaction with NH2NHCONH2). The separated semicarbazones were hydrolyzed and converted to the corresponding methyl esters 14 and 15. The stereoisomers were correlated with known compounds (todomatuic acid). This synthesis did not address the vicinal stereochemistry problem in a selective manner. The synthesis did provide an impressive amount of 27, largely because it involved simple and reliable chemistry. The stereochemistry problem was solved using separation science.
b1026_Chapter-05.qxd
164
Related Approaches to the Juvabiones Ayyar, K. S.; Rao, G. S. K. "Studies in Terpenoids. IV. Synthetic Studies in Juvabiones and Analouges. Conversion of ar-(+)-turmerone to ar-(+)-juvabione" Can. J. Chem. 1968, 46, 1467.
28
Mori Juvabiones
31
30
O
32 (97%)
O
Feature is use of aldol condensation to introduce side chain as title implies. Synthesis does not address stereochemical issues.
Birch Approach: First Attempt to Seriously Address Diastereomer Problem CO2Me
enone to desymmetrize cyclohexanone
O
retro-aldol to reveal vicinal stereochemistry
OH
OH O
H
H
O
O
O 34
H
H 14
34
33
erythro-Juvabione
Juvabione-5
Page 164
CHO
10:55 AM
Ferrino, S. A.; Maldonado, L. A. "Further Extensions of the Kinetic Enolate Method for Terpenoid Synthesis" Syn. Commun. 1984, 14, 925-931. O O O Me2CuLi 2 steps O O O
Organic Synthesis via Examination of Selected Products
14/15
29
O
b1026
mixture of diastereomers O
OAc
12/21/2010
CO2Me
O
Organic Synthesis via Examination of Selected Natural Products
OMe
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 165
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
165
Juvabione-5 Two additional syntheses that did not address the vicinal stereochemistry problem are outlined here. They differ from the Mori synthesis in regard to the chemistry used to introduce the sidechain, but are similar in that they rely on reactions of a 4-substituted cyclohexanone to introduce the vicinal stereochemical relationship. The first clear attempt to address the stereochemical problem presented by the juvabiones was reported by Birch. The plan was to use enone 33 to desymmetrize the 4-substituted cyclohexanone intermediates encountered thus far. The vicinal stereocenters in enone 33 were to be derived from 34 via a retro-aldol condensation. As we will see, 34 was to be prepared via a DielsAlder reaction, a reaction that frequently has been used to establish vicinal stereochemistry with a high degree of selectivity.
OMe O
H
4R
H
O 1. NaBH4
H R
37
H O
2. MnO2
O
H H
H 40
endo
5:1 separable by chromatography
CO2Me
CO2Me
H
H
+
39 cis = 3 Hz
O 42
O 44
O
+
H H
50%
67%
CO2Me
Ca/NH3 H OH 45
H O
H 43
1. HCN 2. HCl, MeOH 3. POCl3, pyridine
H
short synthesis of mixture of diasteromers via unsaturated nitrile
major H H
24% H2SO4, CrO3 acetone, H2O
70%
CO2Me
CO2Me
H
H O
O
H
H 14
50%
Juvabione-6
H 14
erythro-Juvabione
Page 166
80%
R
O
R
CO2Me 41
10:55 AM
HClO4, AcOH
3. POCl3, pyridine 4. MeOH, HCl
1. H2, Pd/C, AcOH 2. acetone cyanohydrin K2CO3
O
R
38
Organic Synthesis via Examination of Selected Products
separation by distillation or GC
OMe O
O 1'
b1026
37 endo
Organic Synthesis via Examination of Selected Natural Products
(1:1)
4
12/21/2010
1'R
36 exo
Birch reduction of anisole (PhOMe)
HClO4, AcOH
80%
4R
35
O
OMe O +
1'S
180 °C, 4 days
b1026_Chapter-05.qxd
O
OMe
166
Birch, A. J.; Macdonald, P. L.; Powell, V. H. "Reactions of Cyclohexadienes. Part VIII. Stereoselective and Non-Stereoselective Syntheses of dlJuvabione" J. Chem. Soc. (C) 1970, 1469.
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 167
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
167
Juvabione-6 The Birch group synthesis began with a cycloaddition between diene 35 and trans-6-methyl-2-hepten-4-one to give a mixture of endo and exo cycloadducts. This reaction established the critical vicinal stereochemical relationship between C4 and C1′. In the endo cycloadduct, this relationship was rel-1′R,4R, and in the exo cycloadduct it was rel-1′S,4R.* I imagine the hope was that the endo cycloadduct would predominate (the norm for Diels-Alder reactions) leading to selective formation of 37. One can imagine that given the technological advances that have taken place since 1970 (high pressure, Lewis acid catalysis),2,3 it would now be possible to accomplish this reaction (or a variation using another dienophile) with high levels of diastereoselection. Nonetheless, in this instance, the diastereomers were either carried forward as a mixture in a non-selective synthesis, or separated before moving forward in a stereocontrolled manner. The non-selective synthesis involved treatment of 36/37 with acid to effect a retro-aldol condensation to provide 38. Reduction of the double bond, followed by a Mori-like endgame, gave 39 (14/15) as a mixture of diastereomers. When the synthesis was continued with endo-cycloadduct 37, the retroaldol condensation gave 40 as a single diastereomer. Reduction of the saturated ketone was accompanied by conjugate addition of the resulting alcohol to the enone to give 41 and 42 as a separable 5:1 mixture of diastereomers (“not necessarily respectively” in the words of the authors). We will just continue as though 42 was the major isomer as it is of little consequence to the outcome of the synthesis. Cyanohydrin formation and dehydration provided a separable mixture of unsaturated esters 43 and 44. Treatment of 44 with calcium in ammonia gave 45 and oxidation of the alcohol completed a synthesis of erythro-juvabione (14). In principle, 41 could have been put through the same paces to provide additional material. Formation of 43 represents a more serious loss of material, as bringing it through to either 14 or 15 would require considerably more effort than conversion of 44 to 14. Overall the Birch strategy is creative and directly addresses the vicinal stereochemistry problem that is the focus of this chapter. The execution of the plan, however, is problematic. I imagine that with more time, and tools
*Cycloadducts 36 and 37 are produced as racemic mixtures. Thus the RS descriptors only indicate relative stereochemistry. Furthermore, I have assigned RS stereochemistry not according to the Cahn-Ingold-Prelog (CIP) convention, but to indicate the diastereomeric juvabione that would be produced from the indicated cycloadduct.
O
H
NEt2
O NEt2
H
H 2O H
49
top-face protonation
H
H Br
4
H
53
OH
52
4
HCl
CO 2H H
H
CO 2H H
4
50
51
CN base
EEO (1.5 eq) OEE
1. (CH2 OH)2 PhH, TsOH
O H 4
2. CrO3 -2pyr H
54 50% overall
O
O
1. (MeO)2 C=O NaH, PhH
O
H 4
OH O
H H
O
MeO2 C
2. NaBH4 , MeOH DMF, -10 °C
4
H
1. TsCl, pyridine 2. NaOMe, MeOH 10% DMSO 3. hydrolysis
56
55 80%
60% MeO2 C
O H
er y thr o-Juvabione
4
14 55%
Juvabione-7
H
Page 168
H
OH EVE, Et2O
LiAlH 4
10:55 AM
OEE
OEE
H 4
PtO2 (Adam’s Catalyst) H 2, dioxane, trace HCl
Organic Synthesis via Examination of Selected Products
Ph3 P, CBr 4 Et2 O
CO 2H H
4
H 48
47
OEE
60%
b1026
46
H
12/21/2010
AcOH
Organic Synthesis via Examination of Selected Natural Products
H 3C C C NEt2
b1026_Chapter-05.qxd
O
O
168
Ficini, J.; Touzin, A. M.; "Cycloaddition of an ynamine to cyclohexenones. Synthesis of aminobicyclo[4.2.0]octenones" Tet rahedron Lett. 1972, 2093. Ficini, J.; Touzin, A. M.; " Stereochemical control of an asymmetric center f ormed a to the carboxyl f unction by hydrolysis of bicyclic enamines. Stereoselective synthesis of diastereoisomeric γ -keto acids" Tetr ahedron Lett. 1972, 2097. Ficini, J.; d’Angelo, J.; Noire, J. "Stereospecific Synthesis of dl-Juvabione" J. Am. Chem. Soc. 1974, 96, 1213.
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 169
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
169
available today (40 years later), tactics could be identified that would realize the potential of this strategy. Finally, I note the Birch synthesis is one example of a general strategy for controlling vicinal stereochemistry in which one (or both) of the stereogenic centers is not in a ring: (1) set the stereogenic centers in a ring and (2) open the ring. Let’s look at another synthesis of erythro-juvabione that follows this general strategy.
Juvabione-7 Jacqueline Ficini’s group (France) was very interested in ynamine chemistry and applied some of their discoveries to a synthesis of erythro-juvabione (14) as shown here. A formal [2+2]-cycloaddition of cyclohexenone with ynamine 46 gave 47. This bicyclo[4.2.0]octene derivative has very distinct convex and concave faces. Thus, a kinetically controlled protonation of the enamine double bond occured from the less hindered (convex) face to presumably give iminium ion 48, which underwent subsequent hydrolysis. A retro-Claisen condensation of the resulting strained-1,3-carbonyl compound provided 49. The strategy for controlling vicinal stereochemistry is truly (1) set it in a ring and (2) open the ring. In this case the ring opening relied on the strain inherent in an intermediate 2-acylcyclobutanone. The synthesis continued with reduction of the cyclohexanone to the alcohol oxidation state, taking it out of play for a series of reactions that constructed the sidechain (49 → 54). The sidechain ketone was then protected as an acetal, and the cyclohexanone was reinstalled by deprotection and oxidation of the cyclohexanol. Regioselective acylation of 55 under conditions of thermodynamic control, followed by reduction of the intermediate β-ketoester, gave 56 (for comparison see 3 → 14 on Steroids-3). Formation of the tosylate, a β-elimination, and ketal hydrolysis completed the synthesis of 14. One of the distinguishing features of this synthesis (relative to those we have examined thus far) is the use of a 3-substituted cyclohexanone as an intermediate rather than a 4-substituted cyclohexanone. This choice eliminates problems associated with passing through a symmetrical (24, 29, 32) or pseudosymmetrical (42) intermediates that either destroy stereochemistry at C4 (for example see 23 → 24 or hydrogenation of 42) or make maintaining control of this center difficult (see conversion of 42 → 45).
O H
H
14 erythro-Juvabione
59
OH S
OR
*
* OR OH
OR OH
S
S
*
S
S
O
4
60 H
1. n-BuLi
MeO
2. ZnCl2
H
HO OMe *
CH3 C
HO OMe H2, Pd/BaSO4 quinoline
ZnCl 63
64
62 65:35 mixture
76% (cis:trans = 97:3) (2 diastereomers)
Anion acceleration of oxy-Cope rearrangement: Evans, D. A.; Golob, A. M. "[3,3] Sigmatropic Rearrangements of 1,5-diene Alkoxides. Powerful accelerating effects of the alkoxide substituent" J. Am. Chem. Soc. 1975, 97, 4765.
LiAlH4 ∆
NaOMe
HO OMe
HO C
+ 65 85% (2 diastereomers)
Juvabione-8
66 trace
Page 170
61
*Stereochemistry not important (in theory) if rearrangement proceeds only via chair TS. O
Substrate Synthesis
MeO
59
60
15 threo-Juvabione
10:55 AM
S
Organic Synthesis via Examination of Selected Products
OR 58
R=H H
4
H
57
CO2Me
4
4
R
b1026
H
S
12/21/2010
O
4
OR S
H
Organic Synthesis via Examination of Selected Natural Products
4
O
OH
oxy-Cope
OH R=H
H
b1026_Chapter-05.qxd
CO2Me
170
Evans, D. A.; Nelson, J. V. "Stereochemical Study of the [3,3] Sigmatropic Rearrangement of 1,5-Diene-3-Alkoxides. Application to the Stereoselective Synthesis of dl-Juvabione" J. Am. Chem. Soc. 1980, 102, 774.
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 171
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
171
Juvabione-8 Recall our discussion of the Claisen rearrangement. This usually takes place via a well-defined chair-like transition state. We have already seen that this can translate to high transfer of chirality (Prostaglandins-24). It can also translate to excellent control of vicinal stereochemistry in an Csp3-Csp3 bond-forming reaction. If the olefin geometry of the allyl and vinyl portions of the rearrangement substrate are well defined, this translates to excellent control over vicinal stereochemistry. The same is true for the Cope rearrangement, the [3.3]-sigmatropic rearrangement of a 1,5-hexadiene.4 In 1980 the Evans group reported a synthesis of erythro-juvabione (14) during the course of studies designed to reveal the nature of transition states in the anion accelerated version of the oxy-Cope rearrangment. The idea was that 14 could be prepared from an intermediate of type 57. We have already seen this endgame in action during the Ficini synthesis of 14 (Juvabione-7). Compound 57 is a 1,6-dicarbonyl compound. Any 1,6-dicarbonyl compound can, in principle, be derived from the corresponding bis-enol tautomer (or a derivatives thereof). Thus, one precursor of 57 (where R = H) would be substituted 1,5-hexadiene 58, which could be derived in turn from an oxyCope rearrangement of isomeric 1,5-hexadiene 59. The key point is that the rearrangement of 59 to 58 establishes the vicinal stereochemical relationship of the juvabiones. On the other hand, the exocyclic Z-olefin (60) would provide the vicinal stereochemical relationship present in threo-juvabione (15). Note that (in theory), stereochemistry at the center marked with an asterisk should not play a major role in determining the stereochemical course of the rearrangments of 59. In other words, 59 and its C* diasteromer should both serve as precursors to 14, while 60 and its C* diastereomer should both serve as precursors to 15. We will come back to this point. The synthesis of the oxy-Cope substrates was accomplished in a direct, but non-stereoselective manner. Thus, lithiation of 61 and transmetallation to provide presumed allenylzinc reagent 62, followed by reaction with cyclohexenone, gave 63 as a mixture of diastereomers. Semihydrogenation of 63 gave Z-olefin 64 (compare with 60), and reduction of 63 with lithium aluminum hydride gave E-olefin 65 (compare with 59) along with a trace of allene 66.5 Both sets of diastereomers were separated by chromatography over silica gel impregnated with silver nitrate, and their rearrangements were examined.
b1026_Chapter-05.qxd
172
KH diglyme
HO S
OMe
S
110 °C
O
S
OMe H
OMe
OMe
O
H
H
O
H
O
S
O
O
MeO
MeO
(96)
O
MeO
S
OMe 110 °C
H
H
O OMe
O OMe
H S
H
H
O H OMe
37h
H 68
59 (R=Me)
(4)
Chair TS lower energy that Boat TS
Boat TS Endgame
H
1. H2 SO4 , acetone
H
3. MsCl, Et3N 4. NaOMe, MeOH H
OMe
OMe
H
1. (COCl)2 , PhH
H
2. H2 Cr 2 O7, H2 SO4 H2 O-acetone
CO2H H
67b
67a
CO 2Me
CO 2Me
CO 2Me 1. (MeO)2 C=O 2. NaBH4
O
69 38% overall
See paper for an interesting variation of this transformation via hydrazone chemistry (anion accelerated chelatropic extrusion)
2.
H O
Cd 2
H 14 60% overall erythro-Juvabione
Juvabione-9
Page 172
S
H
O OMe
S
10:55 AM
KH diglyme
HO
Organic Synthesis via Examination of Selected Products
77%
b1026
67
Chair TS
12/21/2010
OMe
H
Organic Synthesis via Examination of Selected Natural Products
37h 59 (R=Me)
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 173
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
173
Juvabione-9 Let’s begin with 59 (R=Me). Treatment of this rearrangement substrate with KH provided a 96:4 ratio of 67 + 68. These diastereomers differ in terms of vicinal stereochemistry and enol ether geometry. The major product (67) presumably was formed via rearrangement of the intermediate alkoxide via a chairlike transition state as anticipated. The minor product (68) was presumed to arise from rearrangement via a competing boat-like transition state. This nicely explains the differences in stereochemistry between the observed products (vicinal stereochemistry and olefin geometry). The synthesis was completed by installation of the unsaturated ester (67a → 67b), hydrolysis of the vinyl ether, oxidation of the intermediate aldehyde to acid 69, and installation of the rest of the sidechain.
b1026_Chapter-05.qxd
174
Rearrangement of Diastereomeric 1,5-Dienes
Methoxy axial in chair
74 boat (30)
73 chair (70)
H
OH S
S
OMe
KH
OMe
H 71
+
H
H OMe
73 boat (2)
O
O
S
R
KH
H
OMe
+
H
Methoxy axial in chair
OMe H
H 72
H
H
74 chair (98)
OH
Methoxy equatorial in chair
68 chair (70)
Juvabione-10
67 boat (30)
OMe
Page 174
O
O
10:55 AM
70
OMe
H
H
Organic Synthesis via Examination of Selected Products
+
OMe
12/21/2010
H
OMe
b1026
S
KH R
H
Organic Synthesis via Examination of Selected Natural Products
O
O HO
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 175
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
175
Juvabione-10 The remaining three diastereomeric rearrangement substrates [epi-59 (R=Me) and the two diastereomers of 65] were also subjected to rearrangement conditions. The epimer of 59 (70) gave 73, with the vicinal stereochemistry required for erythro-juvabione (14), as the major product. This material was accompanied, however, by significant amounts of 74, with the vicinal stereochemistry of threo-juvabione (15). The origin of these products can be explained exactly as shown for 59 (R=Me) [Juvabione-9]. The erosion in stereoselectivity presumably reflects a boat TS competing more successfully with a chair TS. Note that the methoxy group now occupies an “axial” site in the chair TS and is no longer sterically as differentiated from the boat TS. In accord with this observation substrate 71 (one of the two diastereomers of 65) rearranges with high selectivity (the methoxy group occupies an equatorial site in the chair TS) and 72 (the other diastereomer of 65) suffers considerable erosion of stereocontrol (the methoxy group occupies an axial site in the chair TS). This study provided insight into mechanistic aspects of the anion accelerated oxy-Cope rearrangment. It also illustrates how control of terminal olefin stereochemistry in [3.3]-sigmatropic rearrangements can be used to establish vicinal stereochemistry. Note that both 59 (R=Me) and 71 give superb control. Only when there are overriding steric factors (70 and 72) does erosion of stereocontrol begin to occur.
CO2Me
X O
O
H
76
75
78
77
O
O 1. LDA
1. MeMgBr
2. I(CH2)3Cl
2. H3O+
Cl
80
79
81
1. (CH2OH)2, TsOH
H O H
O
84
89% overall Ficini Intermediate
2. LDA, THF -78 °C to rt
OTBS
EtOH 25:1
77
50% overall
Stork-Danheiser Cyclohexenone Synthesis O
O
O H
76
83%
O O
2. TBDMSCl, Et3N imidazole 3. RMgCl 75
83 90% overall
100%
m-CPBA O
1. MeOH, TsOH
H
2. TBAF 3. PCC, CH2Cl2
O
H2, Pd/C
58%
+
O
82 6:1 peroxyacetic acid gives a 12:1 ratio of lactones
dl-erythro-Juvabione (14)
Schultz, A. G.; Dittami, J. P. "The Bicyclo[3.3.1]nonane Solution to the Problem of Vicinal Stereochemical Control at a Substituted Cyclohexane ring. A Total Synthesis of dl-erythro-Juvabione" J. Org. Chem. 1984, 49, 2615
Juvabione-11
Page 176
OEt
Cl
O
1. NaI, acetone
10:55 AM
Steric Approach Control in Catalytic Hydrogenation
Organic Synthesis via Examination of Selected Products
14
b1026
O
Organic Synthesis via Examination of Selected Natural Products
O
12/21/2010
O
O
H
OEt
b1026_Chapter-05.qxd
176
Control Stereochemistry in Ring - Cleave Ring
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 177
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
177
Juvabione-11 Here is another synthesis that relies on setting stereochemistry in a ring and then opening the ring. The strategy was to prepare sterically biased bicyclic olefin 77. It was imagined that reduction of the olefin would occur selectively from the face flanked by the methano bridge, to give 76. Baeyer-Villiger oxidation of the ketone was expected to occur in the normal manner, with migration of the most highly substituted carbon, to provide lactone 75. There are a number of reasonable ways to proceed from 75 to 14. Execution of the plan first called for a synthesis of 77. This was to be accomplished by an intramolecular alkylation of a compound of type 78 (where X = some leaving group). In the forward direction, chloride 81 was assembled using a reliable method for the preparation of substituted cyclohexenones. Alkylation of 79 (notice the differentiation of electrophilic carbons by use of different halides), followed by introduction of the methyl group, and hydrolysis–dehydration of the intermediate β-hydroxyketone derivative, gave 81. Conversion of the chloride to a better leaving group and kinetic enolate generation using LDA provided 77. Hydrogenation of 77 took place with 25:1 diastereoselectivity to provide 76. Baeyer-Villiger oxidation with m-CPBA gave the desired lactone along with the regioisomeric oxidation product 82. Regioselectivity improved to 12:1 when peroxyacetic acid was used as the oxidant. Hydrolysis of lactone 75 was followed by protection of the secondary alcohol as a TBDMS ether. Treatment of the ester with isobutylmagnesium chloride under controlled conditions gave ketone 83. The ketone was protected as an acetal. The alcohol was liberated with TBAF and oxidized with PCC to afford 84, an intermediate in the Ficini synthesis of erythro-juvabione (14). Before moving on, let’s compare the Schultz (this work) and Ficini approaches. Whereas both of these syntheses use the “set stereochemistry in a ring and then open the ring” approach, the Schultz synthesis places oxidation in the side chain precisely where it is needed and the Ficini approach does not. Both syntheses, however, add the rest of the sidechain (4 carbons in the Schultz synthesis and 5 carbons in the Ficini synthesis) in a single reaction. My own bias is that the Schultz plan is nicer with respect to the sidechain, but the Ficini synthesis does accomplish the task. It is notable that up to now, the Ficini synthesis is the only synthesis we have considered that does not place the sidechain oxygen directly in the desired location. Of course, the Ficini approach leads to an examination of ynamine chemistry whereas the other approaches would not. There is give and take in the planning of any synthesis.
96:4
74%
Me
75:25
2:1
2:3
55%
TMS
96:4
99:1
---
92%
TsOH, Et2O, rt
H
CO2tBu
OMe O
(OC)3Cr
EtOAc
H 94%
H
OtBu
R Me
CO2tBu H 88
rt, 3h
87 For a paper that describes a related approach and suggests a manner in which the Pearson approach might be adapted for asymmetric synthesis see: Miles, W. H.; Brinkman, H. R. "A Formal Synthesis of (+)-Juvabione" Tetrahedron Lett. 1992, 33, 589.
CO2H H 89 72% 8 steps
erythro-Juvabione
H
91
100% HCO2H
Possible Origin of Selectivity
14
Total synthesis can serve as a template for reaction development. If we look back this was the case for this synthesis and the Evans synthesis (and several that will follow). Asymmetric synthesis is also worth considering at the point. All of the preceeding syntheses afforded racemic material. At what point could you introduce asymmetry into each of these syntheses? Which strategies would be most amenable to asymmetric synthesis? Of course this problem has been addressed (not necessarily within the context of the strategies we have seen thus far) and we will now turn to this issue.
Juvabione-12
Page 178
H2, Pd/C
O
R H vs.
O
O
O Me
10:55 AM
H
OtBu
(OC)3Cr
90
OtBu R H 86 R = TMS
OMe
Yield
---
12/21/2010
R 85
O
(OC)3Cr
3. CF3CO2H, -60 oC, 30 min 4. NH4OH, rt, 30 min
Selectivity o
---
Organic Synthesis via Examination of Selected Products
m
Selectivity m
b1026
o (OC)3Cr
m:o
R
OMe 1. CH3CH2CO2tBu, LDA 2. HMPA
Organic Synthesis via Examination of Selected Natural Products
Pearson, A. J.; Paramahamsan, H.; Dudones, J. D. "Vicinal Stereocontrol during Nucleophilic Addition to Arene Chromium Tricarbonyl Complexes: Formal Synthesis of dl-erythro-Juvabione" Org. Lett. 2004, 6, 2121-2124.
OMe
b1026_Chapter-05.qxd
178
A Recent Approach to the Problem
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 179
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
179
Juvabione-12 Let’s now look at a recent approach to the problem. It is actually a study wherein the “problem” was used to learn something about the stereochemical course of a new reaction. The Pearson group was studying reactions of ester enolates with arene chromium tricarbonyl complexes. As part of this study, the reaction of anisole derivatives (85) with the enolate derived from tert-butyl propionate was examined. The result was formation of a chromium tricarbonyl complex of a 1-methoxy-1,3-cyclohexadiene of type 86. Conversion of the chromium-diene complex into a cyclohexenone would provide material that could serve as an intermediate in a juvabione syntheses if the reaction took place with good diastereoselectivity. There are both regiochemical and stereochemical issues here. The enolate (presumably largely of E-geometry based on the work of Ireland) could add “ortho” or “meta” to the methoxy group, “syn” or “anti” to the metal, and provide either “erythro” or “threo” vicinal stereochemistry. A series of compounds of type 85 were examined. It was well known that nucleophiles add anti to the metal in such complexes. Good “meta” selectivity was achieved with R=H or R=TMS. Excellent diastereoselectivity was observed for the meta adduct in the case of R=TMS. Whereas regioselectivity was good when R=H, diastereoselectivity was poor. The stereochemistry of 86 (R=TMS) was established by conversion to 89, an intermediate in the Ficini synthesis. Thus, metal decomplexation and protonolysis of the TMS group was achieved using p-toluenesulfonic acid in ether. The resulting enone 87 was hydrogenated to give 88, and the tert-butyl ester was cleaved using acid to provide 89. It was suggested that the observed stereoselectivity reflected a minimzation of steric interactions between the methyl group of the enolate and the R-group of 85 in a transition state that leads to a staggered array of substituents along the forming sp3-sp3 bond. This is consistent with lower selectivity observed in the meta adduct for 85 where R=H or Me. The Pearson synthesis is a nice example of total synthesis serving as a template for reaction development. If we look back, this was also the case for the Ficini and Evans syntheses (and several that will follow). Asymmetric synthesis is also worth considering at this point. All of the preceeding syntheses afforded racemic material. At what point could you introduce asymmetry into each of these syntheses? Which strategies would be most amenable to asymmetric synthesis? Of course this problem has been addressed (not necessarily within the context of the strategies we have seen thus far) and we will now turn to this issue.
b1026_Chapter-05.qxd
180
Terpenes as Starting Materials
O
OH
CN H
H 94
(R)-(+)-limonene
95 1. O2, hν, pyridine hematoporphyrin
mixture separated by crystallization of 3,5-dinitrobenzoates
2. CrO3, H2SO4 AcOH, PhH
2. KI, AcOH, H2O Singlet oxygen allylic oxidation is not regioselective. The reaction proceeds through an allylic alcohol that undergoes oxidative rearrangement in the Cr(VI) oxidation. For a full account see Pawson, B. A.; Cheung, H.-C.; Gurbaxani, S.; Saucy, G. J. Am. Chem. Soc. 1970, 92, 336. The synthesis established absolute configuration of natural product by comparison of optical properties with natural material.
CO2Me 1. Ag2O, NaOH
H O H 14
CHO H O
2. CH2N2 H 96 33%
Juvabione-13
Page 180
H 93
10:55 AM
H
H
2. NaCN
2. NaOH, H2O2, H2O
92
Li
1. TsCl, pyridine
H
12/21/2010
2
H
Organic Synthesis via Examination of Selected Products
BH
b1026
1.
Organic Synthesis via Examination of Selected Natural Products
Pawson, B. A.; Cheung, H.-C.; Gurbaxani, S.; Saucy, G. "Stereospecific Synthesis and Absolute Stereochemistry of Natural (+)-Juvabione" J. Chem. Soc., Chem. Commun. 1968, 1057-1058.
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 181
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
181
Juvabiones-13 One approach is to begin with a single enantiomer of a chiral starting material. This is the “chiral pool” approach (Prostaglandins-15). The Hoffman-LaRoche group began their synthesis with (R)-limonene (92). The challenges in moving from limonene to juvabione involved (1) introduction of the sidechain with control of stereochemistry at the exocyclic stereogenic center and (2) oxidation of the methylcyclohexene to the unsaturated ester. The first issue was not addressed with control over stereochemistry. Hydroboration-oxidation of 92 using a hindered borane gave a mixture of alcohols that were separated after derivatization. Once pure 93 was in hand, the rest of the sidechain was introduced in a straightforward manner (93 → 94 → 95). The methylcyclohexene was then oxidized using singlet oxygen, followed by oxidation of the mixture of intermediates. This reaction was not regioselective, but did give unsaturated aldehyde 96 in a modest yield. Oxidation of the aldehyde and esterification completed the synthesis of 14. This synthesis may not seem elegant, but it established the absolute configuration of the natural product.
2. DMF, hexanes
4
CHO
R
(S)-(-)-limonene used as starting material for diastereomeric target
95:5 olefin geometry (E :Z)
92:8
5:95 olefin geometry (E :Z)
61:39
S-substrate, 86:14 olefin geometry (E:Z)
80:20 (Ring = S)
MeO2 C
3 steps
MeO2 C
MeO2 C
HWE
N H
MeO2 C
O 100
CO2 Me MeO2 C
110 °C, toluene O
101
102
37%
74% (E:Z = 9:1)
Racemic synthesis that follows similar bond construction at step that introduces relative stereochemistry
Juvabione-14
14/15
O
97% erythro:threo = 14:86
Page 182
Fujii, M.; Aida, T.; Yoshinara, M.; Ohno, A. "NAD(P)+ - NAD(P)H Models. 71. A Convenient Route to the Synthesis of Juvabione" Bull. Chem. Soc. J pn. 1990, 63, 1255-1257.
10:55 AM
(R)-(+)-limonene
OH
H 99
98
97
4
54%
CHO
Organic Synthesis via Examination of Selected Products
92
Pawson chemistry completes synthesis
H 4 days
12/21/2010
(Activity I)
b1026
R
H
b1026_Chapter-05.qxd
Baker’s Yeast
Alumina H
Organic Synthesis via Examination of Selected Natural Products
1. n-BuLi H
182
Fuganti, C.; Serra, S.; "Baker’s Yeast mediated enantio selective synthesis of the bisabolene sesquiterpenes (+)-epijuvabione and (-)-juvabione "J. Chem. Soc., Perkin Trans. 1, 2000, 97-101.
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 183
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
183
Juvabione-14 Here is another approach from R-limonene (92). The idea was to use reagent control (see Prostaglandins-12) to establish the exocyclic stereogenic center. Substrate 98 was prepared from 92 via formylation of an intermediate allylic lithium reagent, and conjugation of intermediate enal 97. The chiral reagent in this case was a collection of enzymes secreted by fermenting Baker’s yeast. Thus, when the E:Z olefin ratio in 98 was 95:5, the reduction gave 99 as a 92:8 mixture of diastereomers (at the exocyclic center). Alcohol 99 has the stereochemistry required for erythro-juvabione (14). When the E:Z olefin ratio in 98 was 5:95, the reduction was less selective, affording a 61:39 mixture of stereoisomers with 99 as the major isomer. Finally, if S-limonene was the starting point for the synthesis, and the E:Z olefin ratio was 86:14, an 80:20 mixture of the C4 diastereomer of 99 and ent-99 was obtained. The bottom line is that the enzymes responsible for the reduction were capricious. They were reasonably selective when E-98 was the substrate, but showed lower selectivity (or more sensitivity to the C4 stereogenic center) with other substrates. Nonetheless the approach is interesting. A racemic synthesis that follows a similar bond construction in the step that introduces relative stereochemistry is shown here without much comment. It is a very direct synthesis, but it lacks selectivity obtained in many of the syntheses we have previously examined.
b1026_Chapter-05.qxd
184
Trost, B. M.; Tamaru, Y. "2-Methylthioacetic Acid and Diethyl Malonate as Acyl Anion Equivalents. Synthesis of Juvabione" T etr ahedr on Let t. 1975, 3797-3800 105 O
CHO
SMe
O
O
O
CO2 H
NaCN, MnO2 MeOH, AcOH
CO2 H acyl anion equivalent
85% No control of relative stereochemistry
O
14/15 Negishi, E.; Sabanski, M.; Katz, J.-J.; Brown, H. C. "An Efficient Synthesis of Juvabione and Todomatuic Acid via Hydroboration-Oxidation" Tet rahedron 1976, 32, 925-926.
CHO H
1. NH 2OH (to oxime) 2. Ac2 O (to CN)
CO 2Me H
3. KOH, EtOH (to CO2 H) 4. CH 2N 2 (to ester) 103
1. BH3 (1 eq), Me 2C=CMe 2 (1 eq), THF 2. Me2 C=CH2 (1 eq), Me2 C=CMe2 (0.5 eq) 3. 108 (1 eq), H2 O (2 eq)
CO 2Me H
4. CO (70 atm), 50 °C , overnight 5. NaOAc, H 2O, H2 O2 , 50 °C
1:1
H
108
78% O
14/15
68% overall Prepared 6.2 g as 1:1 mixture of diastereomers. Separated by hydrolysis to acid (KOH), semi carbazone preparation, crystallization, semicarbazone hydrolysis (H2SO4), and esterification (CH2N2).
Juvabione-15
Page 184
For another approach from perillaldehyde see: Craveiro, A. A.; Vieira, I. G. P. "Synthesis of (-)-Juvabione and (-)-Epi-Juvabione" J. Braz. Chem. Soc. 1992, 3, 124.
CO 2Me
10:55 AM
(+)-Perillaldehyde
107 Transformation that establishes equivalency
SMe
Organic Synthesis via Examination of Selected Products
106
104
103
O
b1026
LDA (2.4 eq) HPMA-THF
95%
2. HCl, H 2O
SMe I
CHO
1. NCS, MeOH NaHCO3
46%
12/21/2010
4 Steps
Organic Synthesis via Examination of Selected Natural Products
CO2
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 185
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
185
Juvabione-15 One shortcoming of limonene as a starting material is that the cyclohexene substituent (methyl group) is in the wrong oxidation state, and the allylic oxidation procedure that was used lacked regioselectivity. Perillaldehyde (103) does not have this oxidation state problem. Several groups have used this terpene as a point of departure in syntheses of the juvabiones. Two examples are presented here. The Trost synthesis illustrated the development of a new acyl anion equivalent (105). The Brown synthesis presented methodology for “stitching” alkenes together to produce unsymmetrically substituted ketones (108 + isobutylene + CO → 14/15). Neither synthesis addresses the vicinal stereochemistry problem. The Brown synthesis, however, is very efficient and provided substantial amounts of the natural products as a mixture.
13 Steps
(+)-Norcamphor
15 111
110
O
112
O
14 Steps
O 113
H
11% overall
Baker's Yeast
AcOH, H2SO4
O
O
95% ee
O
115
HO
114
O
CH3CO3H, NaOAc
O H 116 55%
Rigid system used to control relative stereochemistry 1. LDA, THF 2. MeI
Bu3Sn
OH n-BuLi 120
H 74 % overall
O
1. n-BuLi, THF 2. n-Bu3SnCH2I
119 H 7 steps (less than 10%)
OH H
threo-Juvabione (15)
118
Juvabione-16
1. i-Bu2AlH (89%) 2. Ph3P=CH2 (63%)
H O O H 117 81%
Page 186
Nagano, E.; Mori, K. "Synthesis of (+)-Juvabione, a Compound with Juvenile Hormone Activity" Biosci. Biotech. Biochem. 1992, 56, 1589-1591
10:55 AM
(+)-erythro-Juvabione (14) For an equally tortuous route from racemic norcamphor that involves a lipase-mediated resolution of the alcohols derived from the enones shown above see: Nagata, H.; Taniguchi, T.; Kawamura, M.; Ogasawara, K. "A Lipase-mediated Route to (+)-Juvabione and (+)-Epijvabione from Racemic Norcamphor" Tetrahedron Lett. 1999, 40, 4207-4210.
Organic Synthesis via Examination of Selected Products
(commercially available)
b1026
109
Organic Synthesis via Examination of Selected Natural Products
(+)-threo-Juvabione
Wharton
12/21/2010
MeMgI, CuCN LiCl, THF (92%)
3 Steps
O
b1026_Chapter-05.qxd
4 Steps
O
186
Kawamura, M.; Ogasawara, K. "Stereo- and Enantio-controlled Synthesis of (+)-Juvabione and (+)-Epijuvabione from (+)-Norcamphor" J. Chem. Soc., Chem. Commun. 1995, 2403-2404.
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 187
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
187
Juvabione-16 Two more enantioselective syntheses are described here. The first begins with norcamphor. This served as a starting material for the synthesis of 110 and its enantiomer 111. These compounds were used to prepare both 14 and 15 using an organocuprate conjugate addition to set vicinal stereochemistry. A Baeyer-Villiger oxidation was used to “liberate” the sidechain. A weak point in the choice of 109 is that a 5-membered ring must be expanded to a 6-membered ring. That is one reason why this approach is lengthy. The second synthesis began with a nifty reduction of diketone 113 using Baker’s yeast. Conversion of 114 to 115 was followed by Baeyer-Villiger oxidation and rearrangment to lactone 116 (compare with the Corey approach to prostaglandins; Prostaglandins-6). Alkylation of the enolate derived from 116 gave 117 and established the vicinal stereochemical relationship. Reduction of the lactone to a lactol was followed by a Wittig olefination to give 118. A [2,3]-sigmatropic rearrangement was then used to prepare 120, which was carried on to threo-juvabione. Both of these approaches use the now-familiar “set stereochemistry in a ring, then open the ring” strategy. They are effective in establishing vicinal stereochemistry, but are long because of the choice of starting material.
H
H
O
4 1’
H
OAc
H O
OH
OH H
O
122
116
121
O
123
(+)-(4R,1’R)-Juvabione (CO2Me)2CH2 base PdL *
Highest JH Activity OPiv MeO2C CO2Me
n
H
OPiv 88%
H
H
CO2H
H 125 1. LiAlH4 2. NaIO4
(catalyst)
CO2Me H O
NaH Pd(OAc)2 dppe, THF, ∆
O
O
117
116
CO2Me 124
Eur. J. Chem. 2000, 419-423
3 Steps H
MeO2C
5 steps overall (47%)
3. CH2N2 4. NaOMe MeOH CO2Me
54% overall
H O
(+)-(4R,1’R)-Juvabione (15) OH
H 126
Juvabione-17
Bergner, E. K.; Helmchen, G. "Enantioselective Synthesis of (+)-Juvabione" J. Org. Chem. 2000, 65, 5072-5074
Page 188
MeO2C
10:55 AM
racemic
Organic Synthesis via Examination of Selected Products
15
12/21/2010
H
SN2’
b1026
CO2Me
Organic Synthesis via Examination of Selected Natural Products
CO2Me
b1026_Chapter-05.qxd
188
Helmchen Synthesis: Catalytic Asymmetric Induction
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 189
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
189
Juvabione-17 Another enantioselective route to 116, that features catalytic asymmetric induction, is shown here (see Prostaglandins-12 for another example). This synthesis begins with racemic allylic acetate 123. A palladium-mediated allylation of dimethyl malonate in the presence of chiral ligands (for the Pd) provided 124 with excellent enantioselectivity. This material was converted to 116. Alkylation as per the Mori synthesis (Juvabione-16) gave 117, which was converted to 125 using another Pd-mediated malonate allylation. Malonate 125 was converted to 126 via an intermediate tetraol. The synthesis of threo-juvabione (15) was then completed using a short reaction sequence. This synthesis was designed to showcase organopalladium chemistry developed in the Helmchen laboratories. The stereochemistry of the ring stereogenic center was handled using reagent-based control. The bowlshaped nature cis-bicyclo[4.3.0]nonane was used to control stereochemistry of what becomes the exocyclic stereogenic center of the juvabiones.
OMe 130
O H
HO
O
OH
O
H
H
65% (15:85)
Bu3 P-CuI 127
O
TsOH 100%
Corey, E. J.; Ulrich, P. Tetr ahedr on Lett . 1975, 3685
OMe
OMe
OMe
131
132
minor
major
133 ketals separated 1. Hg(OAc)2 , THF-H2 O 2. NaBH4
94% yield (64% conversion)
CO 2Me H
O
O H
O
H
O
O OH H
15
H
135
Juvabione-18
AcOHg
CHO 134
Page 190
Li O
10:55 AM
Morgans, D. J. Jr.; Feigelson, G. B. "Novel Approach to Vicinal Stereocontrol during Carbon-Carbon Bond Formation. Stereocontrolled Synthesis of dl-t hr eo-Juvabione" J. Am. Chem. Soc. 1983, 105, 5477-5479
Organic Synthesis via Examination of Selected Products
Relative Stereochemistry? Absolute Stereochemistry?
b1026
129
127
12/21/2010
O
128
Organic Synthesis via Examination of Selected Natural Products
O
b1026_Chapter-05.qxd
190
Conjugate Addition Reactions of Cyclohexenone
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 191
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
191
Juvabione-18 A conceptually different approach to the juvabiones would be to develop reactions of cyclohexenone (127) with nucleophiles of type 128 that afford conjugate adducts of type 129 with good control over relative (vicinal) stereochemistry and absolute stereochemistry. This plan is conceptually related to the Pearson approach (Juvabione-12) but uses the electrophile needed to directly connect with the now familiar 3-substituted cyclohexanone endgame for juvabione synthesis. We will look at three different approaches that adapt this general strategy. Morgans examined reactions of 127 with cuprates derived from 130. The choice of 130 as a “real” version of 128 was based (in part) on the hope that the stereogenic center adjacent to the carbon-lithium bond might influence the stereochemical course of the conjugate addition. It is notable that this is really a “relative asymmetric induction” approach to the problem. If the solution worked in a relative sense, then all that would be needed to accomplish an enantioselective synthesis would be to start with a single enantiomer of 130. After surveying a number of organometallic derivatives of rac-130, conversion to racemic 131 and 132 was accomplished with a reasonable degree of diastereoselectivity. The mixture of ketones was protected and the major ketals were separated to provide 133. Treatment of 133 with mercuric acetate, followed by reduction of the presumed intermediate organomercurial 134, gave 135. This material was carried through to threo-juvabione (15) in the usual manner.
O
Me Me Me Me Ph Ph Me Me OEt
O
137
SiR3
Me
R
Si R R
O
H
HO
TiCl4, CH2Cl2 127
TiCl4 O
(MeO)2C=O MeO2C 138 140 erythro Me Me Me 88% (11:1) NaBH4 Me Ph Ph 78% (16:1) i-PrOH
-78 °C, 1h
H
139 threo Me Me Me 80% (3:1) Me Ph Ph 77% (3:1) Me Me OEt 65% (11:1)
H TiCl4
Me
O
H
MeO2C 143
R
Si R R
144
145
E-TS
Z-TS
H
MeO2C 14
Juvabione-19
CO2H
O
MeO2C 141
1. MsCl, pyridine 2. NaOMe, MeOH
1. Cy2BH 2. Brown oxidation
1. (COCl)2 2. RMgBr, THF Fe(acac)2
H
H
MeO2C 142
Page 192
Me Me Ph Ph Me OEt Me OPri
Transition State Models
H
H
Z-olefin
Hayashi, T.; Kabeta, K.; Kumada, M. Tetrahedron Lett. 1984, 25, 1499
H
O
10:55 AM
Me Me Me Me
NaH, THF
Organic Synthesis via Examination of Selected Products
SiR1R2R3
-78 °C, 1h
H
b1026
TiCl4, CH2Cl2 127
O
12/21/2010
E-olefin
SiR3
Organic Synthesis via Examination of Selected Natural Products
SiR1R2R3
136
b1026_Chapter-05.qxd
Allylsilane Reagents
192
Tokoroyama, T.; Pan, L.-R. "Efficient Stereoselective Synthesis of Both dl-Juvabione and dl-epi-Juvabione by New Extracyclic Stereocontrol Methodology" Tetrahedron Lett. 1989, 30, 197.
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 193
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
193
Juvabione-19 The addition of allyl silanes to α,β-unsaturated ketones to give δ,ε-unsaturated ketones is known as the Sakurai reaction. Thus, reaction of cyclohexenone (127) with allylsilanes of type 136 and 137 provides conjugate adducts of type 138 and 139. The relationship of these stereoisomers to erythro- and threo-juvabiones (14 and 15, respectively) is clear. The Tokoroyama group surveyed a series of E-crotylsilanes (136) and Z-crotylsilanes (137) in this process. They determined that E-crotylsilanes provided the erythro adduct (138) as the major diastereomer, where as the Z-crotylsilanes gave the threo adduct (139) as the major diastereomer. Selectivities were very good with proper selection of silicon substituents. Both conjugate adducts could be transformed to the respective juvabiones. For example, the unsaturated ester was introduced to 138 using the standard acylation-reduction-dehydration sequence. The terminal olefin of 142 was then chain-extended via acid 143 to provide erythro-juvabione 14. These stereochemical observations led to the proposal of transition-state models for the reaction of the E- and Z-crotylsilanes as shown in structures 144 and 145, respectively. This is another nice example of synthesis being used to learn something about a specific reaction. Of course, whether or not this mechanistic proposal is correct would need to be determined by further studies. The nice feature of this study is that it clearly demonstrates a relationship between reagent olefin geometry and reaction diastereoselectivity.
OH
t-BuOOH CH2Cl2
146
OH O S
+ Ph
147
74% (1:3)
148 1. Ms2O, Et3N CH2Cl2 2. DBU
50% after crystallization (97% ee)
H
O
+
+ 152
S O
Ph
S O
151
(82)
(10)
H
O
1. LiHMDS
Ph 150
S O
Ph 2.
(8)
O S
O 127
47% combined yield
76% + Z,R-isomer
1. Zn, AcOH (76% after crystallization) 2. KH, (MeO)2C=O, THF (77%)
O
H
S
MeO2C 153
H
1. NaBH4 Ph
2. MsCl 3. DBU
1. HCl, PhH H2O 2. HgCl2 Ph aq. THF S
MeO2C 154
CHO
MeO2C
63%
Juvabione-20
H
2. PDC (85%)
155
71%
1. RMgBr (64%)
H
Ph
149
O
MeO2C 15 100% ee
Page 194
H
10:55 AM
80% (91% ee)
60%
O
Ph
Organic Synthesis via Examination of Selected Products
145
144
OH O S
VO(acac)2 SPh
12/21/2010
OMe
Baker’s Yeast
b1026_Chapter-05.qxd
SPh
b1026
O
PhSCH2Li
Organic Synthesis via Examination of Selected Natural Products
O
194
Watanabe, H.; Shimizu, H.; Mori, K. "Synthesis of Compounds with Juvenile Hormone Activity. XXXI. Stereocontrolled Synthesis of (+)-Juvabione from a Chiral Sulfoxide" Synthesis 1994, 1249. For relevant preliminary work see: Hua, D. H.; Venkataraman, S.; Coulter, M. J.; Sinai-Zingde, G. J. Org. Chem. 1987, 52, 719. Binns, M. R.; Haynes, R. K.; Katsifis, A. G.; Schober, P. A.; Vonwiler, S. C. J. Am. Chem. Soc. 1988, 110, 5411.
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 195
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
195
Juvabione-20 Mori was clearly intrigued by juvabione. Nearly 25 years after his first synthesis, his group described the third “conjugate addition” approach we will consider. This synthesis addresses both relative and absolute stereochemistry problems. The plan was to prepare the anion derived from vinyl sulfoxide 149 and examine its reaction with cyclohexenone (127). There was precedence for the γ-carbon of this type of allylic anion to behave as the nucleophile in conjugate additions (Hua). The hope was that this addition would take place with good diastereoselectivity and that the sulfoxide would influence the absolute stereochemistry of the process. The first task was to prepare the chiral sulfoxide. The synthesis began with the conversion of methyl propionate (144) to keto-sulfide 145. Enzymatic reduction of the ketone using Baker’s Yeast gave 146 with decent enantioselectivity. A “directed oxidation” of the sulfide provided an unequal mixture of sulfoxides 147 and 148 (and presumably minor amounts of material derived from the 4–5% of ent-146 present in the starting material) from which 148 could be isolated in 50% yield. Dehydration of the alcohol provided 149 (along with some of the Z-isomer). Notice that Mori decided to place the alcohol beta to the sulfoxide in the precursor of 149. There might be a number of reasons for this, but one is that it facilitated the elimination reaction (dehydration) because of the electron-withdrawing properties of the sulfoxide. The planned conjugate addition gave adducts 150–152. Both the vicinal diastereoselection (150+152:151 = 9:1) and diastereoselection relative to the sulfoxide (152:150 = 10:1) were good. Vinyl sulfoxide 152 was carried on to unsaturated ester 154 using a straightforward reaction sequence. Vinyl sulfide 154 was hydrolyzed to aldehyde 155, establishing the aforementioned homoenolate anion equivalency. The synthesis was completed in two steps to provide a single enantiomer of threo-juvabione (15). A comparison of Mori’s first synthesis (Juvabione-4) and this synthesis provides an indication of how synthesis changed over this period of time.
b1026_Chapter-05.qxd
196
A Completely Different Approach Soldermann, N.; Velker, J.; Vallat, O.; Stoekli-Evans, H.; Neier, R. "Application of the Novel Tandem Process Diels-Alder Reaction/Ireland-Claisen Rearrangement to the Synthesis of rac-Juvabione and rac-Epijuvabione" Helv. Chim. Acta 2000, 83, 2266-2276.
O
CO2Me
endo vs exo
chair vs boat
159
157
158
O O
CH3CH2COCl KOtBu
160
-78 oC
161
O
1. NaHMDS
CH2=CHCO2Me (10 eq)
O CO2Me
140 °C
2. t-BuMe2SiCl, THF DMPU, -110 °C
163
162 56%
I MeO2C
MeO2C OR
CO2Me
O O
vs
CH3CN
O
O
I2
165 166
RO 167
chair
boat (major)
40% (one isomer) Synthesis finished by Arndt-Eistert homologation of the acid followed by the usual stuff.
Juvabione-21
CO2Me
O HO 164 60%
47:29:24 mixture of diastereomers
Page 196
CHO
OTBS
OTBS
10:55 AM
This paper presents a nice "overview" of other approaches (13 key intermediates). The "plan" involves an interesting disconnection that is related to the Evans solution to the vicinal stereochemistry problem. The plan is great, but lack of selectivity in the orbital symmetry controlled reactions are problematic.
Organic Synthesis via Examination of Selected Products
156
O
Diels-Alder
CO2Me
12/21/2010
RO
Claisen
b1026
CO2Me
O
OR
Organic Synthesis via Examination of Selected Natural Products
OR
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 197
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
197
Juvabione-21 The last synthesis we will consider follows a completely different plan. The idea was that a compound of type 156 would serve as a late intermediate in the synthesis. The vicinal stereochemistry was to be set using a Claisen rearrangement of an allyl vinyl ether of type 157. Being a cyclohexene, 157 was to be prepared by a Diels-Alder reaction between a diene of type 158 and a methyl acrylate (159). Given that Claisen and Diels-Alder reactions are thermally mediated, it was hoped that this might all happen in one pot merely by heating the diene and dienophile. For this plan to succeed, the Diels-Alder reaction would have to show endo-selectivity. This is the normal expectation for such a Diels-Alder reaction. The Claisen rearrangement would also have to occur via a well-defined transition state (166 or 167). This was a reasonable expectation based on the Evans-Nelson studies of a related Cope rearrangement (Juvabione-8). The required dienophile was prepared from crotonaldehyde (160). O-Acylation of the derived dienolate gave 161. Enolate generation and silylation gave 162. The tandem Diels-Alder/Claisen sequence gave 164 in 60% yield as a mixture of diastereomers. The structure of the major diastereomer was established by conversion of the mixture to a mixture of iodolactones from which 165 was isolated and analyzed by X-ray crystallography. Working backwards, it was deduced that the major stereoisomer came from an endoDiels-Alder followed by Claisen rearrangement from boat-like TS 167. Perhaps this is not surprising given the steric hinderance present in the chairlike conformation 166. From the standpoint of the synthesis, the mixture of acids 164 was carried on to the end to provide a mixture of juvabiones 14 and 15. This was a great plan, but it just did not deliver the stereoselectivity of other syntheses.
b1026_Chapter-05.qxd
12/21/2010 b1026
198
10:55 AM
Page 198
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
References 1. For early history of the topic see Trost, B. M.; “The Juvenile Hormone of Hyalophoria Cecropia” Acc. Chem. Res. 1970, 3, 120 and references cited therein. See also, Rees, H. H.; Goodwin, T. W. “Molting Hormones” Biochem. Soc. Trans. 1974, 2, 1027–1032. Karlson, P.; Sekeris, C. E. “Ecdysone, an Insect Steroid Hormone, and its Mode of Action” Recent Progress in Hormone Research 1966, 22, 473–495. 2. Giguere, R. J. “Nonconventional Reaction Conditions: Ultrasound, High Pressure, and Microwave Heating in Organic Synthesis in “Organic Synthesis: Theory and Applications” 1989, 1, 103–172. 3. Corey, E. J. “Catalytic Enantioselective Diels-Alder Reactions: Methods, Mechanistic Fundamentals, Pathways, and Applications” Angew. Chem. Int Ed. 2002, 41, 1560–1567. Fringuelli, F.; Piermatti, O.; Pizzo, F.; Vaccaro, L. “Recent Advances in Lewis Acid Catalyzed Diels-Alder Reactions in Aqueous Media” Eur. J. Org. Chem. 2001, 439–455. Evans, D. A.; Johnson, J. S. “Diels-Alder Reactions” Comprehensive Asymmetric Catalysis 1999, 3, 1177–1235. 4. Vittorelli, P.; Hansen, H. J.; Schmid, H. “Kinetics and Stereochemical Course of the Thermal Rearrangement of the Four Stereoisomers of Propenyl But-2′-enyl Ether” Helv. Chim. Acta 1975, 58, 1293–1309. 5. Corey, E. J.; Katzenellenbogen, J. A.; Posner, G. H. “New Stereospecific Synthesis of Trisubstituted Olefins. Stereospecific Synthesis of Farnesol” J. Am. Chem. Soc. 1967, 89, 42454247. Molloy, B. B.; Hauser, K. L. “Effects of Metal Alkoxides on the Lithium Aluminum Hydride Reduction of Substituted Prop-2-ynyl Alcohols” J. Chem. Soc., Chem. Commun. 1968, 1017–1019.
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 199
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
199
Problems 1. Outline a series of reactions that shows how the Wicha strategy could be used to prepare the C20 epimer of 5. (Juvabione-1) 2. Consider the following reactions. Predict the major product of each reaction. Which reactions you expect to take place with good diastereoselectivity or poor diastereoselectivity? Explain the basis for your predictions. (Juvabione-3)
OEt
1. LDA, THF
OEt
? O
O
?
2. CH3I
OEt 1. (PhMe2Si)2CuLi, THF
1. LDA, THF 2. CH3I
O
OEt ?
? O
2. CH3I
1. LDA, THF
O
OEt 1. (PhMe2Si)2CuLi, THF
OEt
2. CH3I
1. Ph2CuLi
? O
3. 4. 5. 6.
2. CH3I
? O
2. CH3I
For lead references see: Yamamoto, Y.; Maruyama, K. “The Opposite Diastereoselectivity in Alkylation and Protonation of Enolates” J. Chem. Soc., Chem. Commun. 1984, 904–905. Fleming, I.; Lewis, J. J. “A Paradigm for Diastereoselectivity in Electrophilic Attack on Trigonal Carbon Adjacent to a Chiral Center: The Methylation and Protonation of Some Open-Chain Enolates” J. Chem. Soc., Chem. Commun. 1985, 149–151. McGarvey, G. J.; Williams, J. M. “Stereoelectronic Controlling Features of Allylic Asymmetry. Application to Ester Enolate Alkylations” J. Am. Chem. Soc. 1985, 107, 1435–1437. Hart, D. J.; Krishnamurthy, R. “Investigation of a Model for 1,2-Asymmetric Induction in Reactions of α-Carbalkoxy Radicals: A Stereochemical Comparison of Reactions of α-Carbalkoxy Radicals and Ester Enolates” J. Org. Chem. 1993, 57, 4457–4470. What isomers of 23 would you expect to be present following the Birch reduction-hydrolysis sequence? (Juvabione-4) What are the structures of the diastereomeric semicarbazones derived from 27? (Juvabione-4) Suggest a mechanism for the conversion of 44 → 45. (Juvabione-6) Suggest some tactics that might (1) improve diastereoselectivity in the Diels-Alder cycloaddition and/or (2) provide a stereoselective
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 200
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
200
alternative to the cyanohydrin-dehydration route from 40 (or a related compound) to either 14 or 15. (Juvabione-6) 7. Propose a modification of the Ficini synthesis that might provide threojuvabione (15) [the C1’-epimer of 14] in a stereoselective manner. (Juvabione-7) 8. Explain the regioselectivity of the acylation of 55. (Juvabione-7) 9. Predict the stereochemical course of the following reactions. (Juvabione-10)
∆
∆
∆
?
?
∆
?
?
Doering, W. v. E.; Roth, W. R. “The Overlap of Two Allyl Radicals or a Four-Centered Transition State in the Cope Rearrangement” Tetrahedron 1962, 18, 67–74. O O O
O
∆
∆
∆
?
?
∆
?
?
Vittorelli, P.; Hansen, H. J.; Schmid, H. “Kinetics and Stereochemical Course of the Thermal Rearrangement of the Four Stereoisomers of Propenyl But-2’-enyl Ether” Helv. Chim Acta 1975, 58, 1293–1309. NMe 2
NMe 2 OEt HO
xylene, ∆
OEt ?
HO
xylene, ∆
?
Sucrow, W.; Richter, W. “Stereochemistry of the Claisen Rearrangement with 1-Dimethlamino-1-Methoxy-1-Propene” Chem. Ber. 1971, 104, 3679–3688.
b1026_Chapter-05.qxd
12/21/2010 b1026
10:55 AM
Page 201
Organic Synthesis via Examination of Selected Products
Juvabione and the Vicinal Stereochemistry Problem
201
10. Explain why the Eschenmoser-Claisen rearrangement and the Bartlett variation of the Eschenmoser-Claisen shown below give different stereochemical results. (Juvabione-10) O OH
MeC CNEt2
O NEt2
NEt2
benzene add catalytic BF3-Et2O to alcohol; stir 4 days at rt
10
(50%)
1
add ROH to ynamine over 18 h period
1
(56%)
10
Bartlett, P. A.; Hahne, W. F. “Stereochemical Control of the YnamineClaisen Rearrangement” J. Org. Chem. 1979, 44, 882–883. 11. Design a synthesis of pilocarpine that relies on [3.3]-sigmatropy to establish vicinal stereochemistry. (Juvabione-10) O N
O
N Me pilocarpine
12. Based on the following observation, propose a synthesis of threojuvabione (15). How might your plan be modified to afford erythro-juvabione (14)? (Juvabione-11)
N
Me
SePh
Me
Me H2O
O CH2Cl2
O
N
O
Cl
O
References: Byeon, C.-H.; Hart, D. J.; Lai, C.-S.; Unch, J. “Reactions of cyclohexanone enamines with α,β-unsaturated thioesters and selenoesters” Synlett 2000, 119–121. Hickmott, P. W.; Miles, G. J.; Sheppard, G.; Urbani, R.; Yoxall, C. T. “Enamine chemistry. XVII. Reaction of α,β-unsaturated acid chlorides with enamines. Further mechanistic investigations. Effect of triethylamine on the reaction path.” J. Chem. Soc., Perkin Trans. 1 1973, 1514–1519. Harding, K. E.; Clement, B. A.; Moreno, L.; Peter-Katalinic, J. “Synthesis of some
b1026_Chapter-05.qxd
12/21/2010 b1026
202
13.
14. 15. 16. 17. 18. 19. 20. 21.
22. 23.
24. 25. 26.
10:55 AM
Page 202
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
polyfunctionalized bicyclo[3.3.1]nonene-2,9-diones and bicyclo[4.3.1]decane-1,10-diones.” J. Org. Chem. 1981, 46, 940–948. Gelin, R.; Gelin, S.; Dolmazon, R. “Acylation of 1-morpholinecyclohexene by ethylenic acid chlorides.” Bull Soc. Chim. Fr. 1973, 1409–1416. Consider the Birch, Ficini, Evans, Schultz and Pearson syntheses and suggest how each synthesis might be modified to address enantioselectivity issues. (Juvabione-12) Suggest a mechanism for the conversion of 95 → 96. (Juvabione-13) Suggest tactics that could be used to proceed from 100 → 102 via 101. (Juvabione-14) Compare 111 (Juvabione-16) and 77 (Juvabione-11) as starting points for juvabione synthesis. (Juvabione-16) Propose an alternative synthesis of 115. (Juvabione-16) Suggest a mechanism for the conversion of 119 → 120. (Juvabione-16) Suggest a reaction sequence that will convert 124 → 116. (Juvabione-17) Provide a reaction sequence that will convert 116 to the C1’ diastereomer of 117. (Juvabione-17) Discuss how the malonic acid derivative used to convert 117 → 126 serves as a “carbomethoxy anion equivalent”. Suggest other tactics for accomplishing this transformation (other than those used in Juvabione16). (Juvabione-17) Provide a mechanism for the conversion of 133 → 135. (Juvabione-18) The reaction of 149 with 127 might be a direct conjugate addition. An alternative mechanism would be a 1,2-addition to the enone from the α-carbon of the metallated sulfoxide followed by an anion accelerated oxy-Cope rearrangment. Evaluate this possibility using a stereochemical analysis of each step of the reaction. (Juvabione-20) Discuss (in general terms) how the anion derived from 149 served as a homoenolate anion equivalent. (Juvabione-20) Propose a reaction sequence that would convert 164 to a mixture of 14 and 15. Juvabione-21) Recall the geminally activated cyclopropane chemistry we saw in Chapters 3 and 4. Propose diastereoselective syntheses of 14 and 15 that revolve around this methodology. (Juvabione-21)
b1026_Chapter-06.qxd
204
General Comments on Target-Oriented Synthesis
10:56 AM Page 204
Difunctional Relationships-1
12/21/2010
I will not comment in detail on the third and fourth steps. These are largely practitioner dependent. An understanding of functional group compatiblities, however, is clearly an important aspect of these steps.
Organic Synthesis via Examination of Selected Products
The second step can be approached in a number of ways. A process commonly called “retrosynthetic analysis” is often used. In this process, one begins with the target structure and systematically works backwards through (usually) less complex intermediates that one feels can be moved to the desired targets by application of certain tactics. In the last four chapters we have seen a number of examples of this process as applied to steroids, prostaglandins, pyrrolizidine alkaloids, and juvabione. It is obvious that this process (retrosynthetic analysis) can generate a series of pathways from simpler materials to the final target. Development of a given strategy can be a function of the practioner's imagination, knowledge of tactics, understanding of mechanistic and stereochemical principles. A quite different approach to stategy development involves “starting material recognition”. The syntheses of juvabione that begin with limonene and perillaldehyde are good examples of such strategies. Recognition of whether or not a synthesis can be accomplished in a covergent manner (rather than stepwise manner) is also important when developing (or comparing) strategies. For example the brevity of the three-component approach to the prostaglandins is a result of its convergent nature.
b1026
The first step can be quite personal. Compounds that are biologically important, but in scarce supply, are frequently selected as targets for synthesis. The notion that a given compound will have valuable properties (biological or otherwise) can serve as an impetus for synthesis. For example, process research groups are often faced with the synthesis of compounds that are likely to have practical importance. On occasion synthesis still serves as a means for proving the structure of a natural product. An interest in testing a particular type of chemistry can dictate target selection. A fascinating structure (which is personal opinion more than fact) can draw one to attempting a synthesis. The bottom line, however, is that this exercise is target-oriented and thus, a specific target or small family of targets must be selected before one can develop a synthetic strategy.
Organic Synthesis via Examination of Selected Natural Products
Target-oriented organic synthesis is a complex exercise that requires at least the following steps: (1) Selection of the target (2) Development of a synthetic plan (commonly called the strategy) (3) choice of reagents for accomplishing the plan (commonly called tactics) (4) execution of the plan in the laboratory (a process that frequently involves revisiting the selected tactics).
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 205
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
205
Difunctional Relationships-1 Target-oriented organic synthesis is a complex exercise that requires at least the following steps: (1) selection of the target (2) development of a synthetic plan (commonly called the strategy) (3) choice of reagents for accomplishing the plan (commonly called tactics) (4) execution of the plan in the laboratory (a process that frequently involves revisiting the selected tactics). A discussion of these four points is presented on page 204 (Functional Groups-1) Since strategy development is presumably logical, one goal of a number of research efforts has been to develop a set of guidelines that, if followed, will allow one to generate plans for the synthesis of any target. The efforts of the Corey group to develop computer-aided methods for strategy development are perhaps most famous in this regard. “The Logic of Chemical Synthesis” is a readable book that presents a distillation of many of the ideas developed in the Corey group. A number of other authors have written books (or articles) that address the development of synthetic strategies and some of these (that DJH has found particularly interesting, entertaining and useful) are shown on the next page. The Warren books are particularly well-organized and include a detailed discussion of difunctional relationships, the topic of this chapter. Although books are clearly useful, opinions and ideas expressed by teachers (mentors/advisors/colleagues/students) also are bound to have an effect on how a given practitioner approaches strategy development. For example, I was strongly influenced by Richard G. Lawton (University of Michigan), William G. Dauben (UC Berkeley) and David A. Evans (Caltech) and I am very grateful to them for the insights they shared with me, many of which I hope I am passing along to you. With these general comments, I want to now move to the topic of functional group reactivity patterns and difunctional relationships. An understanding of the first topic is essential to synthesis. The second topic can be a helpful consideration when developing synthetic strategies, and is revealing when analyzing published syntheses. For example, we will see how such an analysis shows why cyclohexenes and cyclopentenes can serve as precursors of 5- and 6-membered rings, respectively (recall the Johnson approach to progesterone). We will start with some general considerations and then move to examples from syntheses we have already seen.
b1026_Chapter-06.qxd
206
Useful Books
Functional Groups and Charge
δ C E
δ H3C X
δ H3C OR
X = Cl, Br, I
R = H, Ms, Ts, Tf
halogens
Alcohols and derivatives. This list is not exhaustive.
δ R2C O
δ R2C NR
carbonyl compounds
imines
Thus one can generate a set of functional groups that impart partial positive charge to carbon. Using terminology introduced by David Evans, I will call these E-functions (E for Electrophile). It is sometimes important to consider the protonated versions of E-functions to fully appreciate the charge they impart on carbon.
Difunctional Relationships-2
Page 206
Since most carbon-carbon bond forming reactions are polar in nature, it is useful to classify functional groups with regard to the polarity they impose on carbon. For example, a halogen imposes partial positive charge on the carbon to which it is bonds. So does a hydroxyl group in the form of a mesylate (or many other derivatives). So does an oxygen that is doubly bonded to carbon (carbonyl compounds). These "groups" are all more electronegative than carbon.
10:56 AM
Nicolaou, K. C.; Sorensen, E. J. "Classics in Total Synthesis: Targets, Strategies, Methods", Wiley-Verlag, 1996 (798 pages)
Organic Synthesis via Examination of Selected Products
Corey, E. J.; Cheng, X.-M. "The Logic of Chemical Synthesis", John Wiley and Sons, 1989 (436 pages)
12/21/2010
Warren, S.; "Organic Synthesis: The Disconnection Approach", John Wiley and Sons, 1982 (391 pages)
b1026
Fleming, I. "Selected Organic Syntheses: A Guidebook for Organic Chemists", John Wiley and Sons, 1973 (227 pages)
Organic Synthesis via Examination of Selected Natural Products
Ireland, R. E. "Organic Synthesis", Prentice-Hall, 1969 (147 pages)
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 207
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
207
Difunctional Relationships-2 When I first started to learn organic chemistry, I searched for ways to visualize bond-forming reactions. It seemed to me that most reactions were much like bringing magnets together. Oppositely charged atoms formed bonds and like-charged atoms did not, just like the different poles of magnets.* Although many reactions are not best classified in terms of polar intermediates (pericyclic reactions, free radical reactions, some transition metal mediated reactions), many carbon-carbon bond forming reactions are polar in nature. Therefore it is useful to classify functional groups in regard to the polarity they impose on carbon. For example a halogen imposes partial positive charge to the carbon on which it resides. It renders the carbon “electrophilic”. So does a hydroxyl group in the form of a mesylate (or many other derivatives). So does an oxygen that is doubly bonded to carbon (carbonyl compounds). These groups are all more electronegative than carbon. One can generate a set of functional groups that impart partial positive charge to carbon. Using terminology related to that introduced by David Evans, we will call these groups E-functions (E for Electrophilic).2 Note that it is sometimes important to consider the protonated versions of E-functions, or their complexes with other Lewis acids, to fully appreciate the charge they impose on carbon.
*In the mid-1970s I spent two years as a postdoctoral fellow with David Evans’ group, then at the California Institute of Technology where I was exposed to a well-organized way to think about polar coupling reactions. The conceptual framework that I will use was developed by Evans, as was much of the terminology used in this chapter. Whereas this presentation is much less detailed than the concepts developed by Evans, I hope it will be clear enough to students of organic synthesis to be useful. Any shortcomings in this presentation are due to me and not the concept. I also note that a somewhat related analysis of functional group behavior and relationships can be found in Stuart Warren’s book (vide supra).
C A Na
C N
R' Br
R' C N
C E
C N
This seems simple, but is worth recognizing and can be helpful when deciding how to construct certain bonds. Here are some other A-functions. Note that it is important to consider tautomers of some functional groups, as well as their conjugate acids and conjugate bases, to appreciate the polarity that they impart on carbon.
O H3C N O
OH H2C N O
1
2
nitro compound
tautomer
base
O H2C N O 3
O H2C N O
conjugate base
resonance structure
Difunctional Relationships-3
Page 208
Triply Bonded Nitrogen as an A-function
NH C Ar H
10:56 AM
ArH H C N H
Organic Synthesis via Examination of Selected Products
The majority of functional groups, however, can impose either E or N characteristics to carbon. These can be classifed as A-functions (for Alternate). A simple example of an A-function is triply bonded nitrogen:
b1026
main group organometallics
12/21/2010
δ C Si (Sn)
b1026_Chapter-06.qxd
δ C MgBr
Organic Synthesis via Examination of Selected Natural Products
δ C Li (Na, K)
δ C N
208
Some functional groups impart negative characteristics to a carbon (N-functions for Nucleophilic, not to be confused with nitrogen). These groups are largely Group IA and IIA metals (groups that are less electronegative that carbon).
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 209
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
209
Difunctional Relationships-3 Some functional groups impart negative characteristics to carbon (N-functions for Nucleophilic, not to be confused with nitrogen). These groups are largely Group IA and IIA metals (groups that are less electronegative than carbon), for example organolithium reagents, Grignard reagents, and even some silanes and stannanes. The majority of functional groups, however can impose either E or N characteristics to carbon. These can be classified as A-functions (for Alternate). An example is triply bonded nitrogen. Consider hydrogen cyanide (H-C≡N). It can behave as an electrophile when protonated. One example would be the Gattermann reaction in which HCN can be used to introduce a formyl group to an aromatic ring via an electrophilic aromatic substitution reaction.3 On the other hand, the triply bonded nitrogen is a strong enough electronwithdrawing group that HCN can be easily deprotonated to generate cyanide, which is a good nucleophile. This seems simple, but is worth recognizing and can be helpful when deciding how to construct certain bonds. Here are some other A-functions. Consider nitroalkanes with nitromethane (1) as an example.4 The nitro group is a good electron-withdrawing group (note the positive charge on the nitrogen) and thus can behave as an E-function. Nitroalkanes with α-hydrogens can tautomerize to the corresponding nitronic acids (2). This tautomer is electrophilic (note the presence of the iminium ion within 2). The Nef reaction involves the conversion of nitroalkanes to the corresponding aldehydes or ketones (with aqueous acid).5 This transformation is a direct result of the electrophilic nature of carbon bonded to a nitro group. Deprotonation of 1 (or 2) gives the corresponding conjugate base (3). This structure places negative charge on the α-carbon and illustrates how the nitro group can behave as an N-function. The Henry reaction, wherein the conjugate base of a nitroalkane behaves as a nucleophile toward an aldehyde or ketone, illustrates the nucleophilic properties imparted on carbon by the nitro group.6
C N OR
C N NR2
C N OR
C N NR2
H C N N
C N N
H N Br
Ar
H
H
H N 7
6
Br
5
4
Ar
H
H N
Can one capture this with nucleophiles? O
Some other A-functions
H H
R H C S R
sulfonium salts
O
O
H C S R
H C NR3
8
quarternary ammonium salts
sulfoxides
C : alkyating agents
C : conjugate base is nucleophile
C : conjugate base is nucleophile
C : conjugate base (S-ylid) is nucleophile
C : protonation generates electrophile (Pummerer intermediate)
C : alkylating agents (recall Holton PG synthesis)
Difunctional Relationships-4
Page 210
Ar
H N
10:56 AM
Br
H
H N
H
H
H
12/21/2010
ArNHNH2
H O
O
O
O H
Organic Synthesis via Examination of Selected Products
O
b1026
Note that oximes and hydrazones have the potential for inverting the "normal" electophilic nature of a carbonyl carbon. Recall the following "unusual" reaction encountered in the steroid syntheses.
Organic Synthesis via Examination of Selected Natural Products
diazoalkane
hydrazones
oximes and oxime ethers
b1026_Chapter-06.qxd
210
acid
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 211
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
211
Difunctional Relationships-4 Oximes and hydrazones have the potential for inverting the “normal” electrophilic nature of a carbonyl carbon. Consider the two resonance structures of an oxime (or oxime ether if R = alkyl) shown here. One implies that the carbonyl carbon should be an electrophile and the other implies that it might behave as a nucleophile. The same goes for hydrazones. Consider a diazoalkane and its conjugate acid. One is nucleophilic at carbon and the other electrophilic. Work the problems suggested for this section for examples where =NOR and =NNR2 and diazonium groups behave as E-functions and/or N-functions. An old-fashioned but interesting method for dehydrohalogenation of an α-haloketone relies on the “charge reversal” or “umpolung” or “charge affinity inversion” that occurs upon converting a ketone to a hydrazone derivative (4 → 8). Some additional A-functions include sulfonium salts, sulfoxides and quaternary ammonium salts. For example, sulfonium salts can be used to generate sulfur ylids (N-function) or as leaving groups in substitution reactions (E-function). Sulfoxides can be deprotonated to generate nucleophilic carbon (N-function) or activated by electrophiles to generate electrophilic carbon after loss of an α-proton (E-function). Quaternary ammonium salts are good leaving groups (E-function) and can be deprotonated to generate nucleophilic nitrogen ylids (N-function).
b1026_Chapter-06.qxd
212
Difunctional Relationships
X
X
Y
1
5
4
1
1
3
1
Y 1,3-Difunctional Relationship
Odd Difunctional Relationship
Even Difunctional Relationship
Carbonyl Chemistry and the Principle of Vinylogy
H
H
C
C
O C
O C
conjugate acid
acid
H
H
C
C
O C
O C
tautomer
H C
H C
O C
O C
conjugate base
H base C
H C
O C
O C
The natural polarity imparted by carbon is "carbonyl carbon positive" and "α-carbon negative". Also note that carbonyl groups can be converted into almost any other functional group, or can be prepared from almost any other functional group. They are extremely versatile and important in synthesis.
Difunctional Relationships-5
Page 212
carbonyl
10:56 AM
1,2-Difunctional Relationship
Organic Synthesis via Examination of Selected Products
Y
12/21/2010
Y
2
b1026
X
X
Organic Synthesis via Examination of Selected Natural Products
Much of organic synthesis consists of establishing relationships between two functional groups (difunctional relationships). Such relationships can be classified as "odd" or "even".
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 213
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
213
Difunctional Relationships-5 Given this introduction to the polarity that functional groups impart on carbon, let’s consider difunctional relationships. Consider a compound with two functional groups. They can be on adjacent carbons (1,2-difunctional relationship) or separated by a carbon (1,3-difunctional relationship). Any difunctional relationship can be classified as either even (1,2 or 1,4 or 1,6) or odd (1,3 or 1,5 or 1,7).* Since most functional groups can be derived from a carbonyl compound (or converted to a carbonyl compound) we will focus on this functional group. The natural polarity imparted by a carbonyl group on proximal carbons is “carbonyl carbon positive” and “α-carbon negative”. This is a natural consequence of the acid-base properties of carbonyl compounds as illustrated here: carbonyl carbons and their conjugate acids are electrophilic at the C=O carbon; the tautomers of carbonyl compounds and their conjugate bases are nucleophilic at the α-carbon.
*Evans has refered to even and odd difunctional relationships as “dissonant” and “consonant” for reasons that will become apparent.
O C
acidic
O C
base
C
C positive
If one allows for conjugation, the natural polarity imparted by a carbonyl group on a carbon chain is:
O
O
11 1,3
O
O
12 1,5
Odd difunctional relationships "reinforce" one another.
O
O
O
O
13 1,2
14 1,4
O
O
15 1,6
Even difunctional relationships "interfere" with one another.
A consequence of this relationship is that odd difunctional relationships can be constructed using the normal polarity of the carbonyl group. Construction of an even difunctional relationship will require the use of an A-function.
Difunctional Relationships-6
Page 214
10 How will two carbonyl groups interact in terms of polarity?
10:56 AM
O
Organic Synthesis via Examination of Selected Products
negative
b1026
C
H
12/21/2010
H 9
H
conjugate base
Organic Synthesis via Examination of Selected Natural Products
C
O C
acid
b1026_Chapter-06.qxd
O C H
tautomer
conjugate acid
214
carbonyl
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 215
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
215
Difunctional Relationships-6 What is the principle of vinylogy? Consider the acid-base behavior of carbonyl compounds. When a vinyl group is inserted between the carbonyl group and α-carbon of a carbonyl compound, the acidity of the α-CH bond is extended to the γ-CH bond (if not overridden by stereoelectronic considerations).* In addition, the β-carbon of the unsaturated charge associated with the C=O carbon is extended to the γ-carbon. This is simply conjugation. The vinyl group allows the α-carbon (now gamma in 9) to “talk” with the carbonyl. It wires them together. From the standpoint of charge distribution imparted by a carbonyl group to surrounding atoms, a conjugated π-bond extends the alternating charge pattern by two carbons. A second π-bond would extend the alternating charge pattern by two more carbons as shown in structure 10. How will two carbonyl groups interact in terms of polarity? Odd difunctional relationships “reinforce” one another.** Even difunctional relationships “interfere” with one another (see bold bonds in 13–15).*** A consequence of this relationship is that odd difunctional relationships can be constructed using the normal polarity of the carbonyl group (see bold bonds in 11 and 12). Construction of an even difunctional relationship will require use of an A-function.
*The C-H bond must still be able to align with the carbonyl π-bond. In other words, stereoelectronic considerations still apply to unsaturated systems. **Evans described these relationships as “consonant”. ***Evans described these relationships as “dissonant”.
b1026_Chapter-06.qxd
216
Construction of a 1,3-Difunctional Relationship
a
b
O
O OEt
b
OEt
O H3C
OEt 18
17
O
22
21
OEt
O
Cl 23
EtO 20
19
O
26
O CH3
OEt
Cl-Cl
25
O
O
Cl
NaCN 24
27
H2C N N
28
1,2-difunctional compound Whereas you can construct a 1,3-difunctional relationship in any number of ways, it will always be possible to construct this relationship using the normal polarity of carbonyl groups (no A-functions needed). This is a valuable consideration, BUT NOT A RULE. Sometimes the use of an A-function (in a complex setting) will have advantages.
Difunctional Relationships-7
Page 216
OEt
O
10:56 AM
O
Organic Synthesis via Examination of Selected Products
16
12/21/2010
OEt
O
b1026
O a
Organic Synthesis via Examination of Selected Natural Products
O
O
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 217
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
217
Difunctional Relationships-7 Let’s consider possible syntheses of compound 16 from monofunctional compounds. Polar disconnection of bond “a” leads to an “acyl cation” and an “enolate”. Real world versions of these polar species would be benzoate 17 and ester 18 (which would be deprotonated to provide the enolate). Thus, a logical way to contruct 16 might use a crossed Claisen condensation of 17 and 18. Disconnection of bond “b” leads to a crossed condensation between acetophenone (19) and diethyl carbonate (20). Both of these approaches to construction of the 1,3-difunctional relationship in 16 rely on the normal polarity of carbonyl groups. No A-function is needed. We could have disconnected bond “b” to a cation such as 21 and acyl anion 22. This disconnection, however, does not rely on the normal polarity of the carbonyl group. This does not mean that the synthesis could not be done this way. Chloroketone 23 and cyanide (24) are real world versions of 21 and 22, respectively. A displacement reaction followed by ethanolysis of the α-cyanoketone might provide 16. Chloroketone 23, however, is a difunctional compound. Two methods for preparing 23 are shown. The synthesis from ketone 25 makes use of the normal polarity of the carbonyl group. The synthesis from carbon fragments 27 and 28 makes use of an A-function.
b1026_Chapter-06.qxd
218
Construction of a 1,2-Difunctional Relationship O
O
O
H
H
O 30
29
30
31
H or
O
O
32
31
33
The need for an "acyl anion equivalent" surfaces
O
O Cl
O
O
N2
CH2N2
Cl
DMSO, ∆
H
23
Cl 27
Kornblum Oxidation Functional Group Transformation (FGT)
Construction of Difunctional Relationship
O n-BuLi S
O S
PhCO2Et S
S
S
S
H
HgCl2, CH3CN O
Ca(CO3)2, H2O
36
Li 35
34
29
29
FGT Construction of Difunctional Relationship
O OEt
H2C
O S
O
O S
CH3
37 17
(dimsyl anion)
38
CH3
1. Ac2O, ∆ 2. H3O+ FGT
Difunctional Relationships-8
O H O 29
FGT = Pummerer
Page 218
O
10:56 AM
Some Solutions
Organic Synthesis via Examination of Selected Products
Cannot construct without use of A-functions
b1026
need
O
Organic Synthesis via Examination of Selected Natural Products
??? H
12/21/2010
O
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 219
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
219
Difunctional Relationships-8 Let’s look at construction of a 1,2-dicarbonyl compound within the context of ketoaldehyde 29. There is no simple way to construct the bond between the carbonyls without use of an A-function. An analysis of this problem reveals the need for acyl anion equivalents of type 32 or 33. Several possibilities are shown here. We have already seen that a monofunctional compound (27) can be converted to a 1,2-difunctional compound (23) using diazomethane (A-function chemistry). The problem now simply becomes one of oxidation states. The chloride has to be oxidized to the aldehyde oxidation state. One method that could be used is the Kornblum oxidation (carbon is oxidized and sulfur is reduced).7 This is an example of a functional group transformation (FGT). It does not involve construction of a difunctional relationship. That work was done in the first step of the sequence (27 → 23). It is the FGT that establishes the equivalency of diazomethane with the “unreal” formyl anion 31 (see Steroids-18 and Prostaglandins-3, 6 and 15). Lithiated dithiane 35 is another example of a formyl anion equivalent.8 Acylation would provide 36 and thioacetal hydrolysis would provide 29. The thioacetal hydrolysis is the FGT that establishes the equivalency. Acylation of dimsyl anion 37 (from metallation of DMSO) would afford β-ketosulfoxide 38. Once again an oxidation is needed to establish the desired equivalency. This can be accomplished using a Pummerer rearrangement wherein oxidation occurs at carbon with reduction at sulfur.9 Hydrolysis of the resulting S,O-acetal would complete the required FGT. This is a small sampling of potential tactics for accomplishing the desired transformation. In fact, 29 is a very reactive compound because it has two adjacent carbonyl carbons (repelling magnets). Whereas it is easy to draw 29 on the page, it is more difficult to put it in a bottle because of the reactivity of the carbonyl groups with nucleophiles (including water). Nonetheless this exercise illustrates what must be considered if the task is construction of a 1,2-difunctional relationship.
b1026_Chapter-06.qxd
220
Construction of a 1,4-Difunctional Relationship Case #2
A
O
O 39
O
O
39
D
NO2
HO
O
N
A O
O
KOH
43
42
NO2
NO2 OLi
O 44
C
H2SO4
H2O
NO2
D
O
47
41
O
FGT
C=O
A
O
O 45
46
A-group must stabilize negative charge, but be transformed into carbonyl group (supports positive charge at carbon to which A is bonded)
Difunctional Relationships-9
Page 220
O
40
O
10:56 AM
or
Organic Synthesis via Examination of Selected Products
or
C
B
A
b1026
O B
12/21/2010
O
O Case #1
Organic Synthesis via Examination of Selected Natural Products
O
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 221
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
221
Difunctional Relationships-9 Now let’s consider 1,4-dicarbonyl compound 39. One polar disconnection leads to charged fragments A + B or C + D (Case #1). The C + D pair can easily be related to an acyl anion equivalent and methyl vinyl ketone (by application of the principle of vinylogy). The acyl anion equivalent can be generalized by structure 40 in which an A-function behaves as a N-function. An alternative disconnection is shown as Case #2. In this case it is a coin toss between A + B or C + D. Both require an enolate (from a ketone) and an α-acylcation equivalent. We will consider the use of an enolate derived from acetone (43) and a generalized α-ketocation equivalent 42, in which an A-function must behave as an N-function, rendering the terminal olefin electrophilic. What A-function might one use? One choice could be a nitro group. Working through Case #1 in the forward direction, deprotonation of 44 followed by conjugate addition to 41 might give 45. For Case #2, reaction of the enolate of acetone (or an appropriate derivative) with nitroalkene 46 might be expected to provide 45. A Nef reaction would then accomplish the FGT needed to convert 45 to 39.
b1026_Chapter-06.qxd
222
More Solutions O
O
48
OH
OH
2. H3 O+
O
O 49
50
O 39
51
O
Requires C=C reduction
O
Requires C=C reduction
52 54
53
O
O
O
59
60 O
OH
O dienol:
O
O
O
vs
dienolate: 56 55
Activating Group May Help
58 57
Difunctional Relationships-10
vs
Page 222
O
10:56 AM
Construction of a 1,5-Difunctional Relationship
Organic Synthesis via Examination of Selected Products
Here is another solution to the Case #2 approach. It is not a particularly good solution relative to other approaches , but it can be used to illustrate some points: (1) Ketone activation can be usef ul. In this case if PhCOMe (acetophenone) was used as the nucleophile, aldol condensations would surely compete with the desired epoxide opening because the epoxide opening will surely be slow. (2) You can go up and down in oxidation state along your reaction pathway (alcohol to ketone). (3) An epoxide can be regarded as a 1,2-difunctional compound. This is the actual origin of the ultimate even (1,4) dif unctional relationship in the target. (4) Epoxides are derived from alkenes. Thus, alkenes are wonderful precursors of 1,2-difunctional relationships. So here is a problem. What other tactics can you suggest that would get you from 48 to 39?
12/21/2010
O
O Jones
b1026
2.
1. NaOH
Organic Synthesis via Examination of Selected Natural Products
CO 2Et
O
O
1. NaOMe
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 223
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
223
Difunctional Relationships-10 Here is another solution to the Case #2 approach. It is not necessarily a good solution relative to other approaches, but it can be used to illustrate some points: (1) Ketone activation can be useful. In this case if PhCOMe (acetophenone) were used as the nucleophile, aldol condensations would surely compete with the desired epoxide opening because the epoxide opening is slow. (2) You can go up and down in oxidation state along the reaction pathway (alcohol to ketone). (3) An epoxide can be regarded as a 1,2-difunctional compound. This is the actual origin of the ultimate even (1,4) difunctional relationship in the target. (4) Epoxides are derived from alkenes. Thus, alkenes are wonderful precursors of 1,2-difunctional relationships. So here is a problem. What other tactics can you suggest that would get you from 48 to 39? The normal polarity imparted by carbonyl groups, in conjunction with the principle of vinylogy, can be used to construct 1,5-difunctional relationships. Disconnection of each of the four bonds connecting the carbonyl groups in 52 affords fragments that could (in principle) be coupled to give the target structure. For example, the dienol or dienolate derived from 54 might react with a carbonyl compound of type 53 to afford 52. Dienols tend to react with electrophiles at the γ-carbon so this transformation should be possible. The same comments apply to 59 and 60. Another possibility would be the conjugate addition of 55 (or 58) to enone 56 (or 57). Since both partners in these reactions could behave as both nucleophile or electrophile, the use of an activating group might help control the chemistry (vide supra). The bottom line is that construction of this 1,5-difunctional relationship can be accomplished without the use of A-functions through use of the normal polarity of carbonyl compounds.
b1026_Chapter-06.qxd
224
Application to Cyclohexanones O
O
O
O
5
R
HO
R
5
R
O 61
O
O
O
R
O
65 O
O 69
OH 3
R
O
62
70
and
3
R
67
R 66
O
O
R
O OR
71
68 R
O
R
O OH
OR OH
R
R 59
72
73
O
R
O
O 74
75
O
O
O
CO 2R
CO 2R
O
O
R
81
and
78
82
77 80
CO 2R
79
O 76
These disconnections describe annulation (annelation) routes to cyclohexanone derivatives. They are not all-inclusive. Notice that the acyclic precursors are 1,5-dicarbonyl compounds and that the ring-f orming reaction constructs a 1,3-difunctional relationship.
Difunctional Relationships-11
Page 224
59
10:56 AM
60
Organic Synthesis via Examination of Selected Products
59
b1026
and
Organic Synthesis via Examination of Selected Natural Products
64
63
12/21/2010
O
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 225
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
225
Difunctional Relationships-11 Let’s apply these concepts to the synthesis of 3-alkylcyclohexanones (59) from acyclic precursors. Cyclohexanone 59 is a monofunctional compound. It could be prepared from aldol 60 (a difunctional compound) by dehydration and reduction of the intermediate enone. Aldol 60 could be prepared from 61 by an intramolecular aldol condensation. Diketone 61 contains a 1,5-difunctional relationship. Thus it can be prepared using the normal polarity of the carbonyl group, for example by conjugate addition of 62 to enone 63 or conjugate addition of 64 to enal 65. Enal 65 could be prepared, in turn, by aldol-dehydration chemistry. A critical point in the “development” of this plan is placement of the second functional group, working backwards from 59 at a position that generates an odd (consonant) difunctional relationship. It is this choice that leads to a plan that relies on the polarity of the carbonyl group. Placement of the second functional group at the 3-position of 59 takes one back through 1,5-dicarbonyl 67 to enone 68 and ketone 69, or enone 70 and ketone 71. Another plan that works back through odd (consonant) difunctional relationships passes through vinylogous ester 74 and 1,3-diketone 76 to ketone 78 and unsaturated ester 79 or ester 80 and unsaturated ketone 81. See if you can work back from 59 through 1,3-diketone 82 in a similar manner before moving on with this chapter. All of these disconnections describe annulation (annelation) routes to cyclohexanone derivatives. There are tactical issues involved with executing the strategies in the lab that we will not discuss here. Of course these are not the only ways to construct simple cyclohexanone derivatives. The problems explore several other approaches. Nonetheless these are versatile strategies that have been widely used in organic synthesis. One point to take away from this exercise is that 6-membered rings can be prepared without resorting to use of A-functions. Although we have seen this only for carbocycles, it turns out this is true for heterocycles as well (see the problems).
b1026_Chapter-06.qxd
226
Translation to the Laboratory
CO2Et
H3O+
O
O 85
86
Cl O
O
O
O EtO2C
O
PhH O CO2Et
Et3N
87
Et3N
O
89
90 CO2Et
Cyclopentanones S O
O
O
O
O
S Br 91
HO
92
O
1,4-Difunctional Relationship
93
91
94 1,5-Difunctional relationship, but must use A-function to construct 5-membered ring
Difunctional Relationships-12
Br
95
Page 226
88
O
10:56 AM
Steroids Revisited
Organic Synthesis via Examination of Selected Products
2. Neutralize
83
O CO2Et
NaOEt, EtOH, ∆
12/21/2010
O
Me
b1026
O
84
O
Organic Synthesis via Examination of Selected Natural Products
OEt 1.
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 227
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
227
Difunctional Relationships-12 Let’s see how this can translate to the laboratory. The reaction of ethyl acetoacetate (83) with ethyl crotonate (84) provides 85 in 55% yield.10 This sequence involves initial formation of the enolate of 83. Enolate formation is insured by the use of an “activating group” that is removed in a later operation. Enolate formation is followed by a conjugate addition to 84. A subsequent proton transfer and (ester enolate to ketone enolate) is followed by a nucleophilic acyl substitution whose driving force is presumably formation of a stable 1,3-diketone enolate. Protonation, ester hydrolysis and decarboxylation complete the synthesis of 86. This synthesis is a variation of 78 + 79 → 77 → 76 (Functional Groups-11). Let’s revisit the Wieland approach to steroids (Steroids-7). The reaction of 87 with the enone derived from in situ dehydrohalogenation of 88 gave 89. This was followed by an intramolecular aldol-dehydration to give 90. This is variation of 68 + 69 → 67 → 66 (followed by hydration to a cyclohexenone) as seen in Functional Groups-11. There are many more tactics that have been developed to accomplish this fundamental strategy in the laboratory. How about the synthesis of cyclopentanones from acyclic precursors? This could also be accomplished by aldol-dehydration strategies. For example 93 will undergo an intramolecular aldol to give 92, and dehydration, followed by reduction of the resulting olefin, would give 91. This strategy starts with a 1,4-difunctional relationship. We have already seen that construction of this relationship will require the use of an A-function. We could prepare the cyclopentanone from a 1,5-difunctional compound. For example metallated 95 might function as an equivalent of 94 and provide 91 via an intramolecular alkylation. It is interesting to note that this transformation also requires the use of an A-function. This is because the 1- and 5-positions of a 1,5-dicarbonyl compound are both inherently positive. The polarity at one carbon must be reversed to facilitate bond construction. A useful lesson that falls out of this is that the construction of 5-membered rings generally requires the use of Afunctions. It is also interesting to note that aldol-dehydration routes to 5-membered rings frequently purchase (rather than construct) the critical 1,4-difunctional relationship!
b1026_Chapter-06.qxd
228
Progesterone Revisited
BrMg
1. LiAlH4 (90%)
HO
EtO2C
o
55%
Johnson-Faulkner Claisen
O 96
2. Br(CH2)4Br (2.9 equiv)
Br
O 97
1. (CH2OH)2, PhH, TsOH 2. NaI 3. Ph3P, PhH
73%
O
O
O
O
1. PhLi (1 eq) 2. RCHO 3. PhLi (1 eq) 4. MeOH PPh3 I
61% from 2-methylfuran
1,4-Difunctional Relationship is Purchased
O NaOH
MeLi 101
OH 102
Schlosser modification of Wittig reaction gives trans-olefin
98
O
40% from RCHO
O
O
O
O
HCl MeOH
H2O-EtOH O
100
99
Johnson, W. S.; Gravestock, M. B.; McCarry, B. E. "Acetylenic Bond Participation in Biogenetic-Like Olefinic Cyclizations. II. Synthesis of dl-Progesterone" J. Am. Chem. Soc. 1971, 93, 4332-4334
Difunctional Relationships-13
Page 228
1. n-BuLi
CHO
2. CrO3-2pyr (86%)
10:56 AM
138 C, 2.5 h
Organic Synthesis via Examination of Selected Products
CHO
12/21/2010
98%
b1026
CH3(OEt)3 CH3CH2CO2H (0.3 mol%)
Organic Synthesis via Examination of Selected Natural Products
This synthesis addresses problems associated with construction of the D-ring in the synthesis of dl-16,17-dehydroprogesterone. Can one terminate the polyolefin cyclization to directly afford a 5-membered ring?
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 229
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
229
Difunctional Relationships-13 Consider the Johnson synthesis of progesterone. The cyclization precursor (102) was derived from cyclopentenone 101, which was prepared by an intramolecular aldol-dehydration of 1,4-diketone 100. Johnson did not construct the 1,4-difunctional relationship present in 100. It was purchased in the form of 2-methylfuran (96). This simple heterocyclic compound is at the same oxidation state as the 1,4-dicarbonyl compound, released by hydrolysis of the furan, first in a protected form (98 and 99) and finally ready for use in the aldol dehydration (99 → 100).
b1026_Chapter-06.qxd
230
HO
RO
RO
OH
CO2Me
12
10 11
OR’ 105
104
12
10
CHO
O
MeO2C(H2C)7
(CH2)7CH3
CN
OH
O LDA-THF-HPMA
RO
R
NaHMDS PhH, ∆, 6h
RO
CN
R
TsO
(methyl oleate)
11
O
R
MeO2C
OMOM RO
O
108
107 106
76%
D-Glyceraldehyde
OMOM RO 109
Use Acyl Anion Equivalent [complex "A-function"] to construct 5-membered ring from 1,5-difunctional compound
Construct 1,3-Difunctional Relationship Stork, G.; Takahashi, T. "Chiral Synthesis of Prostaglandins (PGE1) from D-Glyceraldehyde" J. Am. Chem. Soc. 1977, 99, 1275-1276
Difunctional Relationships-14
Page 230
103 PGE1
10:56 AM
TsO
12/21/2010
8
CO2Me
CO2H
Organic Synthesis via Examination of Selected Products
O
O
O
b1026
Package C=O as alkene
Organic Synthesis via Examination of Selected Natural Products
Prostaglandins Revisited
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 231
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
231
Difunctional Relationships-14 The prostaglandins contain a 5-membered ring. In the syntheses we examined, how was this ring constructed within the context of our current discussion? Several of the syntheses purchase the 5-membered ring (Corey lactone approach, Sih synthesis, the three-component coupling syntheses, Holton’s approach, the Stork free radical cyclization approach). The Stork “glyceraldehyde” approach to PGE1 (103) proceeded through cyclopentenone 104. The enone was introduced by dehydration of a β-alkoxycyclopentanone. This 1,3-difunctional relationship was established by addition of the enolate of methyl oleate to glyceraldehyde (106 → 107). The 5-membered ring was established from a 1,5-difunctional compound making use of a metallated cyanohydrin as an A-function (108 → 109). Note that the 1,2-difunctional relationships, required by intermediates such as 107 and 108, were purchased (as 106).
b1026_Chapter-06.qxd
232
The Importance of Alkenes
O
OH R
O
O
OH R
HO R
O
O R
R 115
117
116
Prostaglandins Revisited OHCHN (CH2)6CN
OHCHN (CH2)6CN
C5H11 AcO
O O 118
O
5-Membered Ring
C5H11 O
O O
(CH2)6CN OHCHN O
119 1,6-Difunctional Relationship
O C5H11
O 120 Cyclohexene
Corey, E. J.; Andersen, N. H.; Carlson, R. M.; Paust, J.; Vedejs, E.; Vlattas, I.; Winter, R. E. K. "Total Synthesis of Prostaglandins. Synthesis of the Pure dl-E1,F1α, F1β, A1, and B1 Hormones" J. Am. Chem. Soc. 1968, 90, 3245-3247.
Difunctional Relationships-15
Page 232
114
If R = Me
10:56 AM
113 O
R
R
R 112
Organic Synthesis via Examination of Selected Products
R 111
110
If R = Me
12/21/2010
HO R
R
b1026
R
Organic Synthesis via Examination of Selected Natural Products
Alkenes are latent 1,2-difunctional compounds. Cyclohexenes are latent 1,6-difunctional compounds (and thus 5-membered rings). Cyclopentenes are latent 1,5-difunctional compounds (and thus 6-membered rings).
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 233
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
233
Difunctional Relationships-15 Alkenes are latent 1,2-difunctional compounds. Cyclic alkenes are latent odd (consonant) or even (dissonant) dicarbonyl compounds depending on the their ring size. For example a cyclohexene is a latent 1,6-dicarbonyl, and a cyclopentene is a latent 1,5-dicarbonyl. It now becomes clear why cyclohexenes can be transformed to 5-membered rings and cyclopentenes to 6-membered rings. We first saw these transformations in the area of steroid synthesis. We can also see that the first Corey syntheses of prostaglandins and the Woodward steroid synthesis use the same fundamental strategy to construct the 5-membered rings (illustrated in Functional Groups-15 with the Corey PG synthesis). A Diels-Alder reaction was used to prepare a cyclohexene (120). This was followed by oxidative cleavage of the olefin to give a 1,6-dicarbonyl compound (119), that was then converted to a 5-membered ring (118), relying on the normal behavior of the carbonyl group.
b1026_Chapter-06.qxd
234
The Importance of Atom Insertion Reactions Insertion of single atoms into carbon-carbon sigma bonds change difunctional relationships from odd to even (or even to odd).
1,6-difunctional relationship
Preparable from 1,4difunctional relationship
124 1,5-difunctional relationship
H
Beckman
121
O
O N
H
Beckman
123
125
126 O
F
O
N
O CH2-insertion
O F
127
134 O
129 1,5
1,4
131 1,6
1,3
1,4
1,5 O
cyclopropanation
F X 128
F O
O
enone formation
retro-aldol O
X 130
O
OR 132
Difunctional Relationships-16
OR 133
Page 234
O
O
10:56 AM
123
O
Organic Synthesis via Examination of Selected Products
122
121 Preparable from 1,5difunctional relationship
Baeyer-Villiger
12/21/2010
Baeyer-Villiger
b1026
O
O O
Organic Synthesis via Examination of Selected Natural Products
O
O
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 235
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
235
Difunctional Relationships-16 Atom insertions constitute another important family of reactions. Insertion of single atoms into carbon-carbon σ-bonds change difunctional relationships from odd to even (or even to odd). Cyclohexanones (for example 121) are preparable from 1,5-dicarbonyls (see Functional Groups-11) or 1,7-dicarbonyls. Baeyer-Villiger and Beckman rearrangments convert cyclohexanones into 1,6-difunctional compounds. Conversely, cyclopentanones (for example 123) can be prepared from 1,4-dicarbonyls or 1,6-dicarbonyls (vide supra) and can be similarly converted to 1,5-difunctional compounds. This is also true for acyclic compounds (127 → 128 and 129 → 130). Carbon insertion reactions accomplish the same change of difunctional relationships. For example insertion of a methylene between the carbonyl groups of 131 converts this 1,3-dicarbonyl compound to a 1,4-dicarbonyl compound. This relationship can be useful from time to time. In concluding this section I want to emphasize that consideration of difunctional relationships is only one way to think about designing a synthesis. Entirely different approaches can also be useful. Nonetheless this is clearly one analytical “tool” that can be used to guide decision-making when designing a synthesis of a given target molecule.
b1026_Chapter-06.qxd
12/21/2010 b1026
236
10:56 AM
Page 236
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
References 1. For an introduction to “retrosynthetic analysis” see Corey, E. J.; Cheng, X-M. “The Logic of Chemical Synthesis”, John Wiley and Sons, 1989 (pages 5–16). See also Wyatt, P.; Warren, S. “Organic Synthesis — Strategy and Control” John Wiley and Sons, 2007. 2. Evans, D. A.; Andrews, G. C. “Allylic Sulfoxides: Useful Intermediates in Organic Synthesis” Acct. Chem. Res. 1974, 7, 147–155. 3. Gattermann, L. Chem. Ber. 1898, 31, 1149–1152. 4. Seebach, D.; Colvin, E. W.; Lehr, F.; Weller, T. “Nitroaliphatic Compounds — Ideal Intermediates in Organic Synthesis?” Chimia 1979, 33, 1–18. Henning, R.; Lehr, F.; Seebach, D. “α,β-Doppeldeprotoierte Nitroalkane: Super-Enamine?” Helv. Chim. Acta 1976, 59, 2213–2217. 5. Nef, J. U. “Ueber die Constitution der Salze Nitroparaffine” Annalen 1894, 280, 263. Noland, W. E. “The Nef Reaction” Chem. Rev. 1955, 55, 137–155. Pinnick, H. W. “The Nef Reaction” Organic Reactions 1990, 38, 655–792. Ballini, R.; Petrini, M. “Recent Synthetic Developments in the Nitro to Carbonyl Conversion (Nef Reaction)” Tetrahedron 2004, 60, 1017–1047. 6. Henry, L. “Nitro-alcohols” Compt. Rend. 1895, 120, 1265–1268. 7. Kornblum, N.; Powers, J. W.; Anderson, G. J.; Jones, W. J.; Larson, H. O.; Levand, O.; Weaver, W. M. “A New and Selective Method of Oxidation” J. Am. Chem. Soc. 1957, 79, 6562. 8. Groebel, B. T.; Seebach, D. “Umpolung of the Reactivity of Carbonyl Compounds through Sulfur-Containing Reagents” Synthesis 1977, 357–402. Seebach, E.; Corey, E. J. “Generation and Synthetic Applications of 2-Lithio1,3-dithianes” J. Org. Chem. 1975, 40, 231–237. 9. Pummerer, R. “Ueber Phenylsulfoxy-essigsaure. II.” Chem. Ber. 1910, 43, 1401–1412. Russell, G. A.; Mikol, J. G. “Acid-catalyzed Rearrangements of Sulfoxides and Amine Oxides. The Pummerer and Polonovski Reactions” Mech. Mol. Migrations 1968, 1, 157–207. 10. Blanchard, J. P.; H. L. Goering “5-Methyl-2-cyclohexenone” J. Am. Chem. Soc. 1951, 73, 5863–5864.
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 237
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
237
Problems 1. Provide mechanisms for the following transformations. (Difunctional Relationships-4) Br NOH
MeONa, NBS
C
NOH
N
O
DMF
Grundmann, C.; Richter, R. “Nitrile Oxides. X. Improved Method for the Preparation of Nitrile Oxides from Aldoximes” J. Org. Chem. 1968, 33, 476–478. O NH 2NH2 HOCH 2CH2 OH
H
H
190 °C, 6 h
HO 2C
HO 2C
90%
Piers, E.; Yeung, B. W. A.; Rettig. S. J. “Methylenecyclohexane Annulation. Total Synthesis of (±)-Axamide-1, (±)-Axisonitrile-1, and the Corresponding C-10 Epimers” Tetrahedron 1987, 43, 5521–5535. O NH 2NH2 -H 2 O O
AcOH (0.2 eq) 75%
OH
Wharton, P. S.; Bohlen, D. H. “Hydrazine Reduction of α,β-Epoxy Ketones to Allylic Alcohols” J. Org. Chem. 1961, 26, 3615–3616. For a mechanistic study see Stork, G.; Williard, P. G. “Five- and Six-MemberedRing Formation from Olefinic α,β-Epoxy Ketones and Hydrazine” J. Am. Chem. Soc. 1977, 99, 7067–7068. For a recent application of the Wharton rearrangement see Liu, J.; Hsung, R. P.; Peters, S. D. “Total Syntheses of (+)-Cylindricines C-E and (−)-Lepadiformine through a Common Intermediate Derived from an aza-Prins Cyclization and Wharton’s Rearrangement” Org. Lett. 2004, 6, 3989–3992. O I MeO
O N2
HBr, Et2 O
I
72% MeO
Br
b1026_Chapter-06.qxd
12/21/2010 b1026
238
10:56 AM
Page 238
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
Hart, D. J.; Hong, W. P.; Hsu, L. Y. “Total Synthesis of (±)Lythrancepine II and (±)-Lythrancepine III” J. Org. Chem. 1987, 52, 4665–4673. 2. Provide a mechanism for the following reaction that illustrates how sulfonium and oxosulfonium groups behave as both N-functions and an E-functions in the following reactions. (Difunctional Relationships-4)
O
Me
(a)
O S
O
Me CH2
MeO MeO OMe
MeO MeO OMe
O
O H2C=SMe2
(b)
3. Provide mechanisms for the following reactions and discuss in terms of how sulfinyl [S(=O)R] or trialkylammonium groups influence charge on carbon. (Difunctional Relationships-4)
Br NaNH 2, NH 3
N Me
Me
N
83%
Me
Lednicer, E.; Hauser, C. R. “A Novel Ring Enlargement Involving the Ortho Substitution Rearrangement by Means of Sodium Amide in Liquid Ammonia” J. Am. Chem. Soc. 1957, 79, 4449–4451. CO2Et Br
DBU, THF, 30 min
N Ph
N Ph
CO2 Et 90%
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 239
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
239
Vedejs, E.; Arco, M. J.; Powell, D. W.; Renga, J. M.; Singer, S. P. “Ring Expansion of 2-Vinyl Derivatives of Thioane, N-Benzylpiperidine, and Thiepane by [2,3] Sigmatropic Shift” J. Org. Chem. 1978, 43, 4484–4485. O
O 1. NaCH 2 S(=O)CH3
OEt
(2 equivalents)
O S
98%
2. acidfication
Corey, E. J.; Chaykovsky, M. “New Synthesis of Ketones” J. Am. Chem. Soc. 1964, 86, 1639–1640. 4. The terms “charge reversal”, “umpolung” and “charge affinity inversion” have been used to describe processes wherein (1) a group is operated upon in a manner that reverses the normal polarity of the group (2) the operation that inverts the normal polarity of the group can be reversed to reveal the original functionality. For this process to have any practical importance, the derivative in which the normal polarity must be usable in some bond-forming reaction. Predict the products of the following reaction sequences and identify the individual steps that consitute the aforementioned process. (Difunctional Relationships-4) O H
morpholinium perchlorate A KCN, H2O
1. KOH, t-BuOH, CH2=CHCN B 2. AcOH, H2O
Leete, E.; Chedekel, M. R.; Bodem, G. B. “Synthesis of Myosmine and Nornicotine using an Acylcarbanion Equivalent as an Intermediate” J. Org. Chem. 1972, 56, 4465–4466. CH2=CHCOCH3 NO2
i-Pr2NH, 60 °C CHCl3
1. NaOMe, MeOH, 0 °C C
D 2. O3, MeOH, -78 °C
b1026_Chapter-06.qxd
12/21/2010 b1026
240
10:56 AM
Page 240
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
McMurry, J. E.; Melton, J. “Conversion of Nitro to Carbonyl by Ozonolysis of Nitronates: 2,5-Heptanedione” Organic Syntheses 1977, 56, 36–39. 1. n-BuLi 2. i-Pr-I S
HgCl2, CaCO3 E
S
F
3. n-BuLi
H2O
4. i-Pr-I
Corey, E. J.; Seebach, D. “Carbanions of 1,3-Dithianes. Reagents for C-C Bond Formation by Nucleophilic Displacement and Carbonyl Addition” Angew. Chem. Int. Ed. 1965, 4, 1075–1077. (MeO)3P, ∆
1. LDA G O
S
Ph
H
2. Me2C=CHCH2CH2I
MeOH
Evans, D. A.; Andrews, G. C.; Fujimoto, T. T.; Wells, D. “Stereoselective Synthesis of Trisubstituted Olefins” Tetrahedron Lett. 1973, 15, 1389–1394. 5. Explain why A can be deprotonated by lithium diisopropylamide (LDA) where as B cannot. Explain why this is not a breakdown of the principle of vinylogy. (Difunctional Relationships-6) O
O
O
O A
B
6. Compound C is commonly called a vinylgous acid. As the name implies, it has a pKa close to that of a carboxylic acid (5.25). Explain using the principle of vinylogy. Provide the structure of the vinylogous methyl ester, vinylogous acid chloride, vinylogous amide that can be derived from C. Suggest methods for their preparation (reason by analogy with the chemistry of carboxylic acids). (Difunctional Relationships-6) O
OH C
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 241
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
241
Fehnel, E. A.; Paul, A. P. “Thiapyran Derivatives. V. The Monosulfinyl and Monosulfonyl Analogs of Phlorogulcinol” J. Am. Chem. Soc. 1955, 77, 4241–4244. 7. Which of the two approaches to the triketone shown below is likely to give the higher yield? Why? (alkylations with α-bromoacetone vs propargyl bromide) (Difunctional Relationships-7) O
O
O
O
O
1. NaH, DMF
1. NaH, DMF
2. HC CCH2Br
2. BrCH2COCH3
O
O
3. H2O, H2SO4 (cat) HgSO4 (cat), MeOH
8. Provide the structure of intermediates and mechanism for the conversion of 38 to 29. (Functional Groups-8) 9. Show how a phensulfinyl group [PhS(=O)] might play the role of the A-group in the approaches to 39 outlined in Functional Groups-9. 10. The benzoin condensation is an example of a reaction that proceeds via acyl anion equivalent intermediates. Provide mechanisms for the following reactions. (Functional Groups-10) H3C
N S Cl
H2N CHO
N N H3C
O NaOH, EtOH, H2O OH
NaCN (cat)
CH2CH2OH
(cat)
CHO
NaOH, EtOH, H2O
11. Treatment of D with E affords 1,4-dicarbonyl compound F in 77% yield. Provide a mechanism for this reaction. (Functional Groups-10) H 1. THF, O Ph D
+
N Et Br
Me HO (30 mol%)
O SiMe3
S
Ph
Ph E
DBU O Ph
Ph i-PrOH (4 eq) 2. H2O
F
Ph
O
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 242
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
242
Mattson, A. E.; Bharadwaj, A. R.; Scheidt, K. A. “The ThiazoliumCatalyzed Sila-Stetter Reaction: Conjugate Addition of Acylsilanes to Unsaturated Esters and Ketones” J. Am. Chem. Soc. 2004, 126, 2314–2315. 12. Diketones 76 and 82 are almost aromatic. To what aromatic compound would addition of one mole of hydrogen provide 76 or 82? (Functional Groups-11) 13. Provide an analysis of 2-substituted cyclohexanone synthesis that is related to Functional Groups-11. Repeat the exercise for 4-substituted cyclohexanones, 2,4-disubstituted cyclohexanones, and so on. (Functional Groups-11) 14. Propose intermediates for the following reaction sequences and think about difunctional relationships along the transformation pathways. (Stevens pyridine, Danishefsky steriods) (Functional Groups-11) OMe
OMe
MeO
MeO
NH 2OH hydrochloride N Me H
(OMe) 2 CHO
EtOH 55%
N Me H
N
Sceletium Alkaloid A-4
Stevens, R. V.; Lesko, P. M.; Lapalme, R. “General Methods of Alkaloid Synthesis. XI. Total Synthesis of the Sceletium Alkaloid A-4 and an Improved Synthesis of (±)-Mesembrine” J. Org. Chem. 1975, 40, 3495–3498. OH H
O
OH 1. Na,NH3,Et2O
2. NaOH, H 2 O O
H 3. 10% aq HCl
N
OH
H O
H
1. TsOH, AcOH H 2. KOH, MeOH
H O 68% overall
O
Danishefsky, S.; Cain, P. “Optically Specific Synthesis of Estrone and 19-Norsteroids from 2,6-Lutidine” J. Am. Chem. Soc. 1976, 98, 4975–4983.
b1026_Chapter-06.qxd
12/21/2010 b1026
10:56 AM
Page 243
Organic Synthesis via Examination of Selected Products
Functional Group Reactivity Patterns and Difunctional Relationships
243
15. A synthesis of furans from 1,3-dicarbonyl derivatives is shown below. Provide the structure of possible intermediates and discuss in terms of difunctional relationship transformations. [Garst, M. E.; Spencer, T. A. “General Method for the Synthesis of 3- and 3,4-substituted Furans. Simple Syntheses of perillene and Dendrolasin” J. Am. Chem. Soc. 1973, 95, 250–252.] (Difunctional Relationships-13) O
1. PhSH, TsOH (cat)
O CHO
2. Me 2S=CH 2 3. HgSO4 , Et2O
Garst, M. E.; Spencer, T. A. “General Method for the Synthesis of 3- and 3,4-Substituted Furans. Simple Syntheses of Perillene and Dendrolasin” J. Am. Chem. Soc. 1973, 95, 250–252. 16. Propose a synthesis of the following compounds from acyclic starting materials. (Difunctional Relationships-15) CO2H
CO2H CO2H O
and
O O
Bartlett, P. A.; Green, F. R., III “Total Synthesis of Brefeldin A” J. Am. Chem. Soc. 1978, 100, 4858–4865.
b1026_Chapter-07.qxd
246
Twistane Analysis: One-Bond Disconnections 1,5-Relationship
1,6-Relationship
8
9
3
MsO
O *
A
B
* *
4
Twistane (1)
OMs 2
C O
O
O
H
O *
OMs
OMs
OMs
OMs
H 1,5-Relationship
*
5
6
7
1,5-Relationship
For a nice paper that includes multiple-bond disconnections see Hamon, D. P. G.; Young, R. N. "The Analytical Approach to Synthesis: Twistane" Aust. J. Chem. 1976, 29, 145-161.
Unnatural Products-1
Page 246
*
10:56 AM
MsO
D
Organic Synthesis via Examination of Selected Products
CO2R
12/21/2010
O RO2C
b1026
O
Organic Synthesis via Examination of Selected Natural Products
* *
RO2C CO2R
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 247
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
247
Unnatural Products-1 In Chapter 6 we saw that an analysis of difunctional relationships within a synthetic target could be used to guide development of a synthetic strategy. In this chapter we will examine syntheses of two unnatural products, with an eye on difunctional relationships. We will first focus on twistane (1), a hydrocarbon of theoretical interest because it contains only six-membered rings, fused such that each ring is constrained in a twist-boat conformation. Twistane is quite symmetrical, but is also chiral, and thus enantioselective synthesis is also an issue with this molecule. Twistane has no functional groups. Development of a synthetic strategy requires passing through functionalized intermediates, and thus we are faced with decisions regarding where to place functionality as we develop a plan. We will restrict our “retrosynthetic analysis” of twistane to one-bond disconnections. For a more thorough analysis of this target that includes multiple-bond disconnections, the reader should consult the excellent paper by Harmon and Young cited in Unnatural Products-1.1 Twistane has four different sigma bonds and thus, there are four possible one-bond disconnections that can be used to convert it to a difunctional intermediate. These disconnections are labelled as Paths A-D in Unnatural Products-1. Path A requires construction of a bond between the two carbons marked with asterisks in structure 2. We can imagine making this bond in a number of ways, one of which involves passing through 1,5-difunctional intermediate 3. An enolate alkylation might be used to construct the required C–C bond. Compound 3 is easily recognized as a compound that could be constructed by a Diels-Alder reaction. Path B requires the bond construction required by structure 4. Once again this might be accomplished via a 1,5difunctional compound of type 5. The bond disconnection depicted in Path C (1 → 6) can also be accomplished from a compound containing a 1,5difunctional relationship (7). Bond disconnection via Path D does not directly lead to a 1,5-dicarbonyl compound. Instead it suggests construction via a 1,6-difunctional compound of type 9. Let’s now look at syntheses that follow each of these paths.
b1026_Chapter-07.qxd
248
First Synthesis of Twistane Whitlock, H. W. Jr. "Tricyclo[4.4.0.03,8]decane" J. Am. Chem. Soc. 1962, 84, 3412.
12
100%
DMSO 13
OH
14
OMs
CN
85%
ethylene glycol
I
pyridine HO
I2
Et3N, EtOAc
MsO
O 16 68%
O 17 79%
HO 18 90%
19
O
O
HO
KOH NH2NH2
NaH (XS) O
Twistane
DMF, 60 °C O MsO 20
87%
15 92%
separate from exo cycloadduct
Brown Oxidation
O
CO2H
HO
23%
90% 21
∆
OH
22
Unnatural Products-2
1
Page 248
H2, Pt
LiAlH4
MsCl
10:56 AM
KOH 110 °C 12 h
85% endo
Organic Synthesis via Examination of Selected Products
10
EtO2C 11
12/21/2010
pyridine Chem. Ber. 1942, 75, 1379
b1026
NaCN
Organic Synthesis via Examination of Selected Natural Products
MsCl
LiAlH4
CH2=CHCO2Et
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 249
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
249
Unnatural Products-2 The first synthesis of twistane, reported by Howard Whitlock (University of Wisconsin) followed Path A. A Diels-Alder reaction between ethyl acrylate and 1,3-cyclohexadiene (10) gave a mixture of endo and exo cycloadducts 11. A standard homologation sequence was used to convert 11 to carboxylic acid 15 (and its exo isomer). An iodolactonization reaction, with six-membered ring lactone formation predominating over seven-membered ring lactone formation, gave 16 and established the desired 1,5-difunctional relationship.2 In addition, 16 was easily separated from the unreactive exo isomer of 15. A series of adjustments in oxidation state gave keto-mesylate 20. An intramolecular alkylation gave 22 and a Wolf-Kishner reduction provided twistane (1).
b1026_Chapter-07.qxd
250
Twistane Analysis: One-Bond Disconnections 1,5-Relationship
1,6-Relationship
8
9
3
MsO
O *
A
B
* *
4
Twistane (1)
OMs 2
C O
O
O
H
O *
OMs
OMs
OMs
OMs
H 1,5-Relationship
*
5
6
7
1,5-Relationship
For a nice paper that includes multiple-bond disconnections see Hamon, D. P. G.; Young, R. N. "The Analytical Approach to Synthesis: Twistane" Aust. J. Chem. 1976, 29, 145-161.
Unnatural Products-3
Page 250
*
10:56 AM
MsO
D
Organic Synthesis via Examination of Selected Products
CO2R
12/21/2010
O RO2C
b1026
O
Organic Synthesis via Examination of Selected Natural Products
* *
RO2C CO2R
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 251
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
251
Unnatural Products-3 One-bond disconnection via Path B leads a symmetrical species of type 4. The strategy introduced in Chapter 6 suggests that making one of the carbons in the disconnected bond an electrophile (place a leaving group on the carbon), and the other a nucleophile (place a carbonyl adjacent to the carbon). After stereochemical considerations (stereoelectronic requirements of an SN2 reaction), keto mesylate 5 can be proposed as a 1,5-difunctional intermediate that might undergo the required bond construction.
HO
OH
H
O
O
HC(OMe)3
H
O
OMe
OMe MsO
OMe MsCl pyridine
OMe
OMe
Li, EtOH
HO
OMe
O
THF, NH3 25A
hydrolysis OMe
O
*
MeO
O MsO
*
*
Dioxane NaH
chair-chair
chair-chair * OMs
O
28B
28A
via twist boat-twist boat
25B
Prepared 450 mg in 9 steps
O Gauthier, J.; Deslongschamps, P. "A New Synthesis of Twistane" Can. J. Chem. 1967, 45, 297
1. HSCH2CH2SH AcOH
* *
Twistane
2. RaNi 22
Unnatural Products-4
1
Page 252
26
27
10:56 AM
OMe
Organic Synthesis via Examination of Selected Products
1,5-Difunctional Relationship
b1026
H 25
H 24
23
Organic Synthesis via Examination of Selected Natural Products
OMe TsOH
2. oxidation
12/21/2010
1. H2, RaNi
b1026_Chapter-07.qxd
252
The Cis-Decalin Route to Twistane
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 253
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
253
Unnatural Products-4 Delongschamps was the first to report a synthesis via a difunctional intermediate of type 5. The synthesis began with naphthol 23. Catalytic hydrogenation of the naphthalene, followed by oxidation of intermediate alcohols, gave 1,5-diketone 24 and established the cis-decalin ring fusion required for the synthesis. Monoprotection of diketone 24 gave acetal 25. Dissolving metal reduction of 25 provided equatorial alcohol 26. This tactic is known to afford largely equatorial alcohols from rigid cyclohexanone substrates.3 Since acetal 25 presumably prefers conformation 25A over 25B for steric reasons, the reduction led to formation of 26 rather than the epimeric alcohol. Alcohol 26 was converted to 28 via mesylate 27, and the enolate derived from 28 underwent the desired intramolecular SN2 reaction, presumably via a twist-boat/twist-boat conformation derived from 28B, to give ketone 22. Reduction of 22, by desulfurization of an intermediate dithiane, completed the synthesis of twistane (1).
24
OH
MsCl, pyridine
H
O
OMs
H 28
H 29 42% crude NaH
recrystallized from hexanes
dioxane
O Wolf-Kishner
Nakazaki, M.; Chikamatsu, H.; Tarriguchi, M. "Horse Liver Alcohol Dehydrogenase (HLADH) Mediated Chemicoenzymatic Asymmetric Synthesis of (+)-Twistane from cis-Decalin-2,7dione" Chem. Lett. 1982, 1761-1764 77% 1
Unnatural Products-5
55% 22
Page 254
phosphate buffer, NAD+
H
10:56 AM
O
HLADH, 123 h, 25 °C
12/21/2010
H
O
Organic Synthesis via Examination of Selected Products
H
b1026
O
(S)-1
Organic Synthesis via Examination of Selected Natural Products
Enantiomers
(R)-1
b1026_Chapter-07.qxd
254
Asymmetric Synthesis of Twistane
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 255
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
255
Unnatural Products-5 Fifteen years after the Delongschamps group completed their synthesis, an enantioselective synthesis of (R)-twistane was reported using the same key bond-construction. Diketone 24 was enzymatically reduced (horse liver alcohol dehydrogenase) to give ketol 29. The synthesis was completed in a straightforward manner.
b1026_Chapter-07.qxd
256
Another Asymmetric Synthesis of Twistane
O
31
O
32
resolved +
Wolf-Kishner
13%
90% (R)-1
O
O 22
22
Tichy, M. "On the Absolute Configuration of Tricyclo[4.4.0.03,8]decane" Tetrahedron Lett. 1972, 2001-2004
Unnatural Products-6
Page 256
30
87%
H2
3. hν
10:56 AM
HO2C
Pd/CaCO3
12/21/2010
1. SOCl2 2. CH2N2
Organic Synthesis via Examination of Selected Products
(S)-1
b1026
(R)-1
Organic Synthesis via Examination of Selected Natural Products
Enantiomers
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 257
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
257
Unnatural Products-6 Another enantioselective synthesis of (R)-twistane was reported by Tichy. This synthesis began with Diels-Alder cycloadduct 30 which was resolved and then converted to cyclopropylketone 31 via an intramolecular carbene addition reaction. Hydrogenolysis of 31 provided ketones 32 and 22. Wolf-Kishner reduction of the minor product (22) gave twistane (1). The hydrogenolysis of 31 occurred at the cyclopropane σ-bond best alligned with π-bond of the carbonyl group. Other than poor regioselectivity in the hydrogenolysis, this was a very direct synthesis of twistane.
b1026_Chapter-07.qxd
258
Twistane Analysis: One-Bond Disconnections 1,5-Relationship
1,6-Relationship
8
9
3
MsO
O *
A
B
* *
4
Twistane (1)
OMs 2
C O
O
O
H
O *
OMs
OMs
OMs
OMs
H 1,5-Relationship
*
5
6
7
1,5-Relationship
For a nice paper that includes multiple-bond disconnections see Hamon, D. P. G.; Young, R. N. "The Analytical Approach to Synthesis: Twistane" Aust. J. Chem. 1976, 29, 145-161.
Unnatural Products-7
Page 258
*
10:56 AM
MsO
D
Organic Synthesis via Examination of Selected Products
CO2R
12/21/2010
O RO2C
b1026
O
Organic Synthesis via Examination of Selected Natural Products
* *
RO2C CO2R
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 259
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
259
Unnatural Products-7 One-bond disconnection via Path C generates bicyclo[3.3.1]nonane intermediates. Keto-mesylate 7 is a 1,5-difunctional intermediate one might convert to twistane via yet a third intramolecular alkylation. This is the pathway pursued by Hamon and Young.1
b1026_Chapter-07.qxd
260
The Bicyclo[3.3.1]nonane Route to Twistane Hamon, D. P. G.; Young, R. N. "The Analytical Approach to Synthesis: Twistane" Aust. J. Chem. 1976, 29, 145-161.
1. Na-K, Et2O
CHO
2. oxidation O
35
O
isomerization problems
OMe
36 NaBH4
O
1. (Cl3CCO)2O 2. OsO4 (cat), NaIO4
O OH
3. (CH2OH)2, TsOH 4. KOH, H2O
+
OH 39
38 84%
37 75%
42
1. 9-BBN 2. H2O2, NaOH H2O
Separated by Preparative GLC
O
O
O
1. CrO3-2Pyr (97%) 2. Ph3P=CH2 (90%)
O
O (3)
KOtBu, t-BuOH (1)
+
OMs
O
1. hydrolysis 2. MsCl, pyridine
OH
73% 41
7 OMs
22
95% overall
Unnatural Products-8
85:15
85:15
40
Page 260
O
O
CH2
10:56 AM
34 75%
Organic Synthesis via Examination of Selected Products
33
Geluk, H. W.; Schlamann, J. L. M. A. Tetrahedron 1968, 24, 5369
b1026
H3O+
Ph3P=CHOMe
12/21/2010
CH2
CH2
Organic Synthesis via Examination of Selected Natural Products
Cl
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 261
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
261
Unnatural Products-8 The Hamon and Young synthesis began with reductive fragmentation of the known adamantanone derivative 33 to provide 34. Conversion of 34 to key intermediate 7 required a bit of work. This transformation required that the ketone of 34 be converted to a hydroxymethyl group via a reductive homologation reaction, and that the olefin be oxidatively cleaved to a ketone. In practice, Wittig olefination of 34 provided enol ether 35, but hydrolysis of the enol ether was complicated by isomerization of the olefin to the more stable endocyclic isomer 36. Thus it was decided to reverse the homologation and oxidative cleavage operations. Reduction of 34 gave alcohol 37. Protection of the alcohol (through esterification), oxidative cleavage of the exocyclic methylene, protection of the resulting ketone as an acetal, and removal of the trichloroacetate protecting group, gave 38. Oxidation of the alcohol was followed by Wittig olefination to provide 39. Hydroborationoxidation of 39 gave alcohol 40 as the major stereoisomer. Hydrolysis of the ketal and mesylate formation completed the synthesis of key intermediate 7 contaminated with 15% of stereoisomer 41. Treatment of 7 with potassium tert-butoxide provided a separable mixture of materials that included 41 (unreacted from the impure starting material), and a 2:1 ratio of 22:42. We have already witnessed the conversion of 22 to twistane. It is notable that 7 must undergo energetically uphill conformational changes to meet the stereoelectronic requirements of the SN2 reactions leading to either 22 or 42. Apparently there is little energy difference between the transition states leading to the intramolecular alkylation products. Despite the problems associated with the last step, this is an interesting approach to twistane that provides one with the opportunity to think of alternatives for construction of the final carbon-carbon bond. I hope the problems will stimulate you in this regard.
b1026_Chapter-07.qxd
262
Twistane via a 1,6-Difunctional Relationship
H2, PtO2
CO2Me
CH2=CHCO2Me
(Adam’s Catalyst)
43
44 MeO2C CO2Me
CO2Me CO2Me
meso-compound not resolvable
N
O O N N
2. H2, PtO2 9
48
47 49%
40% crystallized before resolution
80% (MeO)3P, ∆
diacid resolved as brucine salt
H2, PtO2 Absolute configuration determined by hydroboration-oxidation of twistene to twistone, followed by optical rotation measurements and application of the Octant Rule. (S)-1
49 50%
Unnatural Products-9
Page 262
46
S N
1. Na, NH3 +
S
45 HO HO
10:56 AM
7:3
Organic Synthesis via Examination of Selected Products
+
12/21/2010
CO2Me
MeO2C CO2Me
b1026
CO2Me
Organic Synthesis via Examination of Selected Natural Products
Tichy, M.; Sicher, J. "Synthesis and Absolute Configuration of Tricyclo[4.4.0.03,8]dec-4-ene (Twistene)" Tetrahedron Lett. 1969, 4609
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 263
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
263
Unnatural Products-9 The final twistane synthesis we will consider originates from the one-bond disconnection indicated by Path D in Unnatural Products-1. Just as with Path A, the synthesis uses a Diels-Alder reaction to establish the bicyclo[2.2.2]octane substructure of key intermediate 9. The cycloaddition of cyclohexadiene 43 and methyl propiolate afforded a mixture of regioisomeric cycloadducts 44 and 45. Catalytic hydrogenation of the mixture occured from the sterically most accessible face of the olefin to afford mesocompound 46 and its diastereomer 9. Diester 9 was subjected to an acyloin condensation, and catalytic hydrogenation of the resulting α-hydroxyketone gave diol 47. A Corey-Winter reaction was used to convert 47 to 49 via thionocarbonate 48.4 Catalytic hydrogenation completed the synthesis.
b1026_Chapter-07.qxd
264
Woodward Synthesis of Triquinacene
H
H
H
H
X
H
H
X
H
X
H
H
H
H
O 54
OH
HO
53
52
Triquinacene
50
O
O
56 Winstein, S. Chem. Ind (London) 1960, 405
92% 55
O
51
57 74%
CH3CO3H AcOH NaOAc CH2Cl2
58 90%
dodecahedrane
Unnatural Products-10
OH
54 70% Pb(OAc)4 PhH, ∆
H2SO4 Et2O O
HO
O
CrO3-H2O
KOtBu Et2O-THF CrO3-pyridine
HO
O
O O
59 36%
Page 264
OH
H
H
10:56 AM
X
H
Organic Synthesis via Examination of Selected Products
H 50
H
b1026
H
Organic Synthesis via Examination of Selected Natural Products
H
12/21/2010
Woodward, R. B.; Fukunaga, T.; Kelly, R. C. "Triquinacene" J. Am. Chem. Soc. 1964, 86, 3162
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 265
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
265
Unnatural Products-10 Triquinacene (50) is another hydrocarbon that has long been of interest. One reason is that an appropriate dimerization of triquinacene would provide dodecahedrane (51). From the standpoint of synthesis, triquinacene forces one to address the problem of five-membered ring synthesis. In addition, it is interesting to examine the difunctional relationships used in syntheses by practitioners in the field. I will start with the Woodward synthesis. Triquinacene is symmetrical and thus, it is not surprising that the synthesis (plan) passes through a series of symmetrical intermediates. For example, it was felt that triquinacene could be prepare by elimination of two moles of H–X from an intermediate such as 52 or 53. At this point it is difficult to know what provided Woodward with the insight that led back to 54 as a projected intermediate. Perhaps it was recognition that the known compound 56, derived from norbornadiene and cyclopentadiene, was only one carbon-carbon bond away from containing the triquinacene carbon skeleton as a substructure. The synthesis began with 56, which had been described by the Winstein group at UCLA.5 Oxidation to ketone 57, and epoxidation from the sterically most excessible face of the olefin, provided 55. Base mediated intramolecular opening of the epoxide established the final carbon-carbon bond of the triquinacene skeleton, as alluded to above. Oxidation of 58 to the diketone was accompanied by hydrate formation to provide 54. It was now necessary to break two carbon-carbon bonds to reveal the tricyclic nucleus of triquinacene. This was accomplished by treatment of 54 with lead tetraacetate to reveal anhydride 59, with the first of three double bonds present in triquinacene in place.
b1026_Chapter-07.qxd
266
1. NaOMe CO2Me 2. NaOH, MeOH
MeO2C
1. SOCl2 2. NaN3, toluene
2. CH2N2
3. ∆ 4. MeOH
84% LiAlH4
H
H
H 50 78%
1. 30% H2O2, MeOH
HCO2H
2. 125-140 °C (10-30 mm)
NMe2
Me2N
37% aqueous CH2O
Cope Elimination
84%
Triquinacene
Unnatural Products-11
NHMe
MeHN
64
63 Eschweiler-Clark Methylation
84%
Page 266
H
10:56 AM
62
67% (crystalline)
(45% overall without isolation of anhydride)
NHCO2Me
MeO2CHN
61
Organic Synthesis via Examination of Selected Products
60 100%
59
CO2H
HO2C
b1026
1. MeOH, ∆
12/21/2010
O O
Organic Synthesis via Examination of Selected Natural Products
O
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 267
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
267
Unnatural Products-11 Anhydride 59 was next converted to diester 60. Epimerization to the more stable diester, with the carbomethoxy groups on the convex face of the triquinacene nucleus, followed by saponification, gave diacid 61. The next task was to convert the carboxyl groups to groups suitable for introducing the remaining double bonds using an appropriate elimination reaction. It was decided to use a bis-Cope elimination and thus, diamine 64 became the next target. A “double” Curtius rearrangement of 61 provided bis-carbamate 62. Reduction of the carbamate to diamine 63 was followed by Eschweiler-Clark methylation to give the target diamine 64. Oxidation of 64 to the bis-amine oxide was followed by thermal syn-elimination of N,N-dimethylhydroxylamine to provide triquinacene (50). The origin of the three 5-membered rings in this synthesis is notable. Two of the rings were purchased in the form of starting materials (cyclopentadiene and norbornadiene). The final ring was constructed from a 1,6-difunctional intermediate, keto-epoxide 55.
HO 2C
H 66
NaN 3, H 2 SO 4
CF3CO2 H O H
H
Jones
H
H
CHO
H
O
H
3N HCl-acetone O
O (450 W Hanovia)
H 69 40% from dione
H
hν, MeOH, Pyrex O
H 65
H 68
O
H 66 40%
95%
LiAlH 4
OMs
OH H
H
H
MsCl
H
H
H
H
H
OMs
OH H
H
70
71
70%
Al2O 3
H
H 50 60% from diol
Unnatural Products-12
Mercier, C.; Soucy. P.; Rosen, W.; Deslonschamps, P. "A Convenient Synthesis of Triquinacene" Sy n. Commun. 1973, 3, 161-164. See also Russo, R. Lambert, Y.; Deslongschemps, P. Can. J. Chem. 1971, 49, 531. For derivatives and variations on this theme see Delongschamps, P.; Cheriyan, U. O.; Lambert, Y.; Mercier, J.-C.; Ruest, L.; Russo, R.; Soucy, P. "Synthesis of Triquinacene and some of its derivatives" Can. J. Chem. 1978, 56, 1687.
Page 268
H
OH
10:56 AM
Thiele's Acid
1,6-Dicarbonyl Compound
Organic Synthesis via Examination of Selected Products
H 67
b1026
O
H 65
50
CO2H
O
O H
H
H
H
12/21/2010
CHO
H
H
Organic Synthesis via Examination of Selected Natural Products
H
b1026_Chapter-07.qxd
268
Deslongschamps Synthesis of Triquinacene
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 269
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
269
Unnatural Products-12 Ten years after appearance of the Woodward synthesis, a second synthesis was reported by the Delongschamps group (University of Sherbrooke in Canada). The Delongschamps strategy involved some classical disconnections along the line of carbonyl group chemistry. Thus, it was imagined that triquinacene (50) could be prepared from 65 (a 1,6-dicarbonyl compound) via an intramolecular aldol condensation, followed by some oxidation state adjustments to introduce the final two double bonds. It was recognized that 65 might be prepared from 66 by cleavage of the indicated bond. Diketone 66 was to be derived from Thiele’s acid (67), the Diels-Alder dimer of 1,3cyclopentadiene 2-carboxylic acid.6 Moving forward from Thiele’s acid (67), degradation to diketone 66 was followed by photochemical conversion of 66 to 65.7 An acid-promoted aldol condensation was used to construct the third five-membered ring (65 → 68). The alcohol was then oxidized to a ketone (69) and reduction of the diketone, with hydride delivery from the convex face of the molecule, provided diol 70. Formation of the bis-mesylate (71) and a double elimination reaction completed the synthesis of triquinacene. Once again it is interesting to see that two of the five-membered rings were purchased (cyclopentadiene) and the third ring was constructed from a 1,6difunctional compound (65). Once again, the latter stages of the synthesis make good use of symmetry.
76
OH H
H
H
H
H
O
O
O
H
50
72
73
O
O H 74
1,2-Dicarbonyl
K2CO3 , MeOH O
H
RO aqueous NaHCO3
OR
O
(CH 2O) n
H
79
R
H
83 Prepared 950 mg in one run. Good for analogs.
80
HCl, THF
H 2O, 1 week
H
H
HMPA, 230 °C
H
HO
1:1
H
HO R
H 81%
Johnson-Lemieux good on small scale but poor on large scale.
R = Me
H
O
2. Me2 S
2% diallylation
HO H
3:1
O
R=H CH2 N2 (90%)
H
1. O3 , EtOAc
3:1
O
2. HCl, AcOH, 85-90 °C
tBuO2 C H CO2 tBu 78 93%
77
H
1. KH, allyl iodide, -78 °C
H OH
H
H
50
82
80%
93%
H
BH3 -THF or i-Bu2 AlH, CH 2Cl2
1:1
H
H
O
O H 85% 81
Least stable aldehyde is trapped
Gupta, A. K.; Lannoye, G. S.; Kubiak, G.; Schkeryantz, J.; Wehrli, S.; Cook, J. M. "General Approach to the Synthesis of Polyquinenes. 8. Syntheses of Triquinacene, 1,10-Dimethyltriquinacene, and 1,10-cyclohexanotriquinacene" J . Am. Chem. Soc. 1989, 111, 2169-2179. Jim Cook was DJH’s teaching assistant for an undergraduate laboratory course at the University of Michigan. Thanks Jim!
Unnatural Products-13
Page 270
tBuO 2C
OHC
CO2 tBu
10:56 AM
tBuO2 C
O O
1,6-Dicarbonyl tBuO 2C
O
O
Organic Synthesis via Examination of Selected Products
H
H
75
H
b1026
O
CHO H
12/21/2010
H
Organic Synthesis via Examination of Selected Natural Products
H
b1026_Chapter-07.qxd
270
Cook Synthesis of Triquinacene
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 271
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
271
Unnatural Products-13 Fifteen years after the Delongschamps synthesis, Jim Cook’s group (University of Wisconsin-Milwaukee) described a versatile route to triquinacene. The synthesis revolved around a three-component reaction known as the Weiss reaction. The “plan” was to prepare 50 from tricyclic intermediate 72 which was to be prepared from 1,6-dicarbonyl compound 73. The precursor of 73 was to be diketone 74, to be prepared in turn from (CHO)2 (76) and acetone (75) or a derivative thereof. It is notable that the union of 75 with 76 involves the construction of only 1,3- and 1,5-difunctional relationships. It is the 1,2-difunctional relationship present in glyoxal (76) that facilitates construction of the two five-membered rings. The conversion of 74 to 73 would necessarily involve construction of a 1,4-difunctional relationship. Finally, it is clear that stereochemistry must be considered when introducing the acetaldehyde unit going from 74 to 73. Only the endo isomer (shown) is disposed properly for the intramolecular aldol condensation (73 → 72). Note that this was also an issue in the Delongschamps synthesis, who handled the problem by opening a ring constrained with the required stereochemistry (66 → 65). In the laboratory, acetone equivalent 77 was treated with glyoxal and base to afford 78 (R=H), followed by conversion to the bis-enol ether (R=Me) using diazomethane. Compound 78 (R=Me) is a vinylogous malonate and thus, reaction with base and an alkylating agent (allyl iodide), followed by ester hydrolysis and decarboxylation, gave 79 as a 3:1 mixture of diastereomers. The enolate derived from the monoalkylated intermediate is presumably more hindered than the enolate derived from 78 and thus, polyalkylation of 78 is not a major problem. Introduction of the acetaldehyde residue was completed by ozonolysis of 79 to provide 80, once again as a mixture of isomers. Treatment of 80 with acid promoted the desired aldol condensation to give 81 in 85% yield. Clearly the exo-isomer of 80 can be epimerized to the endo-isomer needed for the aldol. Thus, incorporation of the “extra carbonyl group”, compared to the Delongschamps synthesis, removed the need to establish stereochemistry at the sidechain stereogenic center. The rest of the synthesis is straightforward. Reduction of the ketones provided 82 and a triple-dehydration completed the synthesis of triquinacene (50). This approach is excellent from the standpoint of analog synthesis. For example using 2,3-butanedione as the 1,2-dicarbonyl compound in the Weiss reaction ultimately provided 83 where R=Me.
CO2Et
85
CO2Et
86
87
generated in situ
64-79%
after sublimation
hν (450 W Hanovia)
b
b
+
N
a
+
+
H
H
N 87 N2
90
91
3%
10%
H 92 25-30%
50 58-61%
c b b
88
Individual hydrocarbons purified by GC
or
a
89
50
Unnatural Products-14
H
Page 272
0.0025-0.01 M in pentane
10:56 AM
84
N N
2. MnO2 or CuCl2
Organic Synthesis via Examination of Selected Products
N
1. KOH, i-PrOH, ∆
12/21/2010
N
b1026
EtO2CN NCO2Et
I2
Organic Synthesis via Examination of Selected Natural Products
Wyvratt, M.; Paquette, L. A. "Domino Diels Alder Reactions. II. A Four-Step Conversion of Cyclopentadienide to Triquinacene" Tetrahedron Lett. 1974, 2433-2436
Na
b1026_Chapter-07.qxd
272
Paquette's Remarkable Synthesis
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 273
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
273
Unnatural Products-14 The final synthesis I want to present is a short and imaginative one described by Leo Paquette. Let’s jump right in. The synthesis begins with oxidative coupling of sodium cyclopentadienide (84) to afford 85. This is a highly reactive compound that undergoes reaction with diethyl azodicarboxylate to provide 86 via two sequential Diels-Alder reactions. The initial cycloaddition occurs largely from the face of the cyclopentadiene opposite the 1,3cyclopentadien-5-yl group (recall the Corey approach to prostaglandins), setting up the second Diels-Alder reaction. Hydrolysis of the carbamates and oxidation of the intermediate hydrazine provided azo compound 87. Examination of the azo compound (87) reveals that homolysis of the C–N bonds (with loss of dinitrogen) would provide a diradical (88) that would, upon homolysis of the a-sigma bond, result in formation of triquinacene (50). In the lab, irradiation of 87 accomplished this transformation in good yield. A number of other products reveal that this mechanistic pathway to triquinacene is reasonable. For example, formation of minor product 92 can be rationalized by intramolecular coupling of biradical 88. Homolysis of one of the b-sigma bonds in 88 would provide carbene 89 which could lead to 90 (C-H insertion). A 1,2-shift of the c-sigma bond gives 91. This synthesis is remarkable in its brevity and its use of symmetry. I hope that this chapter has served as a pleasant diversion from natural products synthesis, and has also helped (to a degree) place the discussion of difunctional relationships in a practical context.
b1026_Chapter-07.qxd
12/21/2010 b1026
274
10:56 AM
Page 274
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
References 1. Hamon, D. P. G.; Young, R. N. “The Analytical Approach to Synthesis: Twistane” Aust. J. Chem. 1976, 29, 145–161. 2. Baldwin, J. E. “Rules for Ring Closure” J. Chem. Soc., Chem. Commun. 1976, 734–736. Baldwin, J. E. “Approach Vector Analysis: A Stereochemical Approach to Reactivity” J. Chem. Soc., Chem. Commun. 1976, 738–741. 3. Huffman, J. W.; Charles, J. T. “The Metal-Ammonia Reduction of Ketones” J. Am. Chem. Soc. 1968, 90, 6486–6492. 4. Corey, E. J.; Winter, R. A. E. “A New, Stereospecific Olefin Synthesis from 1,2Diols” J. Am. Chem. Soc. 1963, 85, 2677–2678. 5. Bruck, P.; Thompson, D.; Winstein, S. “Dechlorination of Isodrin and Related Compounds” Chem. Ind. (London) 1960, 405–406. 6. Thiele, J. “Ueber Abkömmlinge des Cyclopentadiens” Chem Ber. 1901, 34, 68–71. 7. Matsui, T. “On the Mechanism of the Photolysis of Strained Cycloalkanones” Tetrahedron Lett. 1967, 9, 3761–3765. 8. Gupta, A. K.; Fu, X.; Snyder, J. P.; Cook, J. M. “General Approach for the Synthesis of Polyquinenes via the Weiss Reaction” Tetrahedron 1991, 47, 3665–3710.
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 275
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
275
Problems 1. Compound 20 is a 1,5-difunctional compound. Therefore one can design a synthesis that proceeds via intermediates that only contain odd difunctional relationships. Propose such a synthesis and indicate potential problems with your approach. If you can’t get started, have a look at the following reference: Ballester, P.; Costa, A.; Raso, A. G.; GomezSolivellas, A.; Mestres, R. “Dienediolates from Unsaturated Carboxylic Acids. Michael Addition of Dilithium 1,3-Butadiene-1,1-diolate (from Crotonic Acid) to Unsaturated Ketones” J. Chem. Soc., Perkin 1: Organic and Bio-Organic Chemistry 1988, 1711–1717. (Unnatural Products-2) 2. Propose another synthesis of 1 via the Path A disconnection and a 1,5-difunctional intermediate that differs from 20 in regard to placement of the functional groups. (Unnatural Products-2) 3. Suggest a 1,6-difunctional intermediate, and a reaction of your proposed intermediate, that might be used to construct the bond disconnected in Path B. (Unnatural Products-3) 4. Suggest an alternative bicyclo[3.3.1]nonane derivative that might be used to “connect” the carbons marked by asterisks in structure 6. Suggest tactics that will transform your proposed intermediate into twistane. (Unnatural Products-7) 5. Provide a mechanism for the conversion of 33 to 34. (Unnatural Products-8) 6. Propose an alternative synthesis of 7 that passes through intermediates with “odd” difunctional relationships. (Unnatural Products-8) 7. Explain the stereoselectivity associated with the conversion of 39 → 40. (Unnatural Products-8) 8. Review olefin metathesis and the Ramburg-Bachlund reaction (see examples below). Suggest alternative tactics for converting 9 → 49 based on these reactions. (Unnatural Products-9) O
O
O 1. t-BuOK, CCl4 2. acetal hydrolysis
S O
O
69%
Matsuyama, H.; Miyazawa, Y.; Takei, Y. “Tetrahydrothiopyran-4-one. A Useful 5-Carbon Synthon for the Synthesis of 3-Cyclopentenones”
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 276
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
276
Chem. Lett. 1984, 833–836. For another variation of the RamburgBacklund reaction see Matusuyama, H.; Miyazawa, Y.; Kobayashi, M. “Regioselective Alkylation and the Ramberg-Baecklund Type Reaction of α-(p-tolylsulfonyl)thiane S,S-Dioxide. A New Route to the Synthesis of 3-Alkyl-3-Cyclopentenones” Chem. Lett. 1986, 433–436.
10 mol% Shrock catalyst N O
9. 10. 11. 12. 13. 14.
15.
16.
N 73%
O
Ph N O Mo F3C CF3 O CF3 CF3 Shrock Catalyst
Martin, S. F.; Chen, H-J.; Courtney, A. K.; Liao, Y.; Patzel, M.; Ramser, M.; Wagman, A. S. “Ring-Closing Olefin Metathesis for the Synthesis of Fused Nitrogen Heterocycles” Tetrahedron 1996, 52, 7251–7264. For a review and other catalysts see Grubbs, R. H.; Chang, S. “Recent Advances in Olefin Metathesis and its Application to Organic Synthesis” Tetrahedron 1998, 54, 4413–4450. Outline a synthesis of 56 from 1,3-cyclopentadiene and norbornadiene. (Unnatural Products-10) Suggest a mechanism for the conversion of 54 → 59. (Unnatural Products-10) Provide the mechanism of the reactions used to convert 59 → 60. (Unnatural Products-11) Provide the structure of the products from each reaction along the path from 60 to 62. (Unnatural Products-11) Suggest a mechanism for the Eschweiler-Clarke methylation (63 → 64). (Unnatural Products-11) Illustrate how 1,3-cyclopentadiene 2-carboxylic acid behaves as a “2hydroxy-1,3-cyclopentadiene equivalent” in the synthesis of 67. Suggest intermediates along the path from 67 → 66 under the conditions cited in Unnatural Products-12? (Unnatural Products-12) Suggest a mechanism for the conversion of 66 → 65. Suggest a reason for the observed regioselectivity (reaction of only one of the ketones) based on your mechanism. (Unnatural Products-12) Provide a mechanism for the conversion of 77 to 78 (R=H). (Unnatural Products-13)
b1026_Chapter-07.qxd
12/21/2010 b1026
10:56 AM
Page 277
Organic Synthesis via Examination of Selected Products
Some Unnatural Products — Twistane and Triquinacene
277
17. Provide the structure of the triquinacene derivative that could be prepared from cyclohexan-1,2-dione using this strategy. (Unnatural Products-13) 18. Predict the triquinacene one might prepare from PhC(=O)CHO using this strategy. (Unnatural Products-13) 19. Suggest a mechanism for the formation of 85 from 84. Provide the structure of the intermediate in the conversion of 85 → 86. Provide the structure of an “unproductive” cycloadduct obtained from the reaction of 85 with diethyl azodicarboxylate. (Unnatural Products-14) 20. Suggest a mechanism for the formation of 92 from photolysis of 87. (Unnatural Products-14)
b1026_Chapter-08.qxd
280
Alkaloids: Importance of the Mannich Reaction Corey, E. J.; Balanson, R. D. "Total Synthesis of dl-Porantherine" J. Am. Chem. Soc. 1974, 96, 6516.
O
1
2
O
1,3-Dif unctional Relationship
Mannich Reaction
Porantherine NH 2
N
CHO
NH
CHO
CHO
6
O
Symmetrical intermediate. 1,5-Dif unctional Relationships. Mask aldehyde as alkene.
O
OH
O
O
O
CrO 3-pyr (6 equivalents)
8 95%
O
1. Mg 2. HCO2 Et
O
O 4
5
O
O
O
O
9 93%
O
CH3 NH 2
O
O
NMe
O
toluene, 4A sieves 18 h
10
In this case, symmetry and the availability of a 1,4-difunctional starting material makes the use of a γ -alkylation equivalent workable. Cl
7
Alkaloids-1
O
Page 280
O O
3
10:56 AM
1
12/21/2010
Me
Organic Synthesis via Examination of Selected Products
H 1
b1026
N
N
N N
Organic Synthesis via Examination of Selected Natural Products
Me N
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 281
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
281
Alkaloids-1 In this chapter we will focus on alkaloid total synthesis. We will start with several syntheses that rely on the addition of an iminium ion (or imine) to an enol (or enolate). This process is commonly refered to as the Mannich reaction.1 It is really the nitrogen counterpart of the aldol condensation. The Mannich reaction is important not only in the laboratory synthesis of many alkaloids, but also in the biosynthesis of many alkaloids. The fact that nature assembles alkaloids using this process is clearly what renders it useful in developing synthetic strategies for alkaloid synthesis. We will start with porantherine (1). This alkaloid is a difunctional molecule. A simple “last step” in any projected synthesis of porantherine would be introduction of the olefin from a ketone. There are two ketone precursors that could be considered. Ketone 2 is one of these. It is also a difunctional compound with a 1,3-difunctional relationship between the ketone and amine functional groups. Note that in the “other ketone”, the ketone and amine functional groups would have a vicinal or 1,2-difunctional relationship. One can imagine 2 being available from 4 (via 3) via an intramolecular Mannich reaction. Compound 4 might be prepared from 5 via another intramolecular Mannich reaction. And 5 would come from 6 via intramolecular imine formation. Going in the forward direction there would be several points to address: (1) Regioselective imine formation going from 6 → 5 would be an issue. This could be addressed by masking the aldehyde in some manner. (2) The stereochemistry of the acetyl group in 4 might be an issue in moving forward to 2. For example, the epimer of 2 would not be disposed to undergo the intramolecular Mannich reaction. It is notable that 6 contains only 1,5-difunctional relationships and thus, should be available via traditional carbonyl group chemistry. It is also notable that this plan makes good use of symmetry. This plan has been executed in two somewhat related manners by the Corey and Stevens groups. The Corey-Balanson synthesis began with the synthesis of symmetrical ketone 9 from 1,4-difunctional chloro-ketal 7. Imine formation provided 10.
O
4-pentenyllithium
O
O
O
NHMe
O
O
2. neutralize 3. ether extraction
PhH 11
12
90%
O
OAc 1.3 eq. TsOH
O O
O
O
1. OsO4 , NaIO4 t-BuOH, H 2O
NCHO
2. (CH2 OH)2 , TsOH, ∆, PhH
85%
O
NMe
CrO 3-2pyr
mixture
80%
3N KOH-EtOH 110 oC, 72 h, ∆
NH
O O
O
O 85%
16
1. 10% aq. HCl
N
N
TsOH-H 2O (10 eq) toluene, ∆, 8 h
2. neutralize 3. ether extraction O
85%
O
45%
N
The synthetic approach (among others) was suggested by LHASA-10, the Harvard program for computer-assisted synthetic analysis. For some reading see Corey, E. J. Quar t. Rev., Chem. Soc. 1971, 25, 455.
1
Alkaloids-2
2
1. NaBH4 2. SOCl2 (5 eq) pyridine
17
55%
Page 282
15
45%
O 13
14
10:56 AM
NCHO
Organic Synthesis via Examination of Selected Products
PhH, ∆,5A sieves
12/21/2010
65% from ketone
b1026
Aldehyde protection is important because the "first" imine formation must be between the amine and one of the methyl ketones.
Organic Synthesis via Examination of Selected Natural Products
(4 cycles needed because imine deprotonation rears its head as a problem)
10
NCH3
1. 10% aq. HCl
b1026_Chapter-08.qxd
NMe
O
282
O
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 283
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
283
Alkaloids-2 Reaction of 10 with 4-pentenyllithium gave 11. Enolization of 10 competed with imine addition and thus, “recycling” of this material was needed to achieve a respectable overall yield of 11 from ketone 9. Acid hydrolysis of 11 furnished enamine 12 after neutralization. An acid-mediated Mannich reaction converted 12 to a mixture of epimeric ketones 13. Oxidation of the N-methyl group provided formamide 14. The next stage of the synthesis called for conversion of the terminal olefin to an aldehyde. This was accomplished using a Johnson-Lemieux oxidation. The resulting ketoaldehyde was then converted to bis-ketal 15. Basic hydrolysis of the formamide provided 16 and hydrolysis of the acetals afforded a mixture of diastereomeric enaminoketones 17. An acid-promoted intramolecular Mannich reaction gave 2. Presumably epimerization of the ketone allowed both diastereomers of 17 to follow the path to 2. Reduction of ketone 2 with sodium borohydride provided an alcohol, which underwent formal dehydration upon treatment with thionyl chloride and pyridine, to give porantherine (1) as a racemic mixture. Overall this synthesis is quite direct. It is notable that the approach was suggested by a computer-assisted synthetic analysis developed at Harvard. It is also notable that the choice of 1,3-difunctional aminoketone 2 as a late intermediate may have been responsible for recognition of this “Mannich reaction” strategy.
b1026_Chapter-08.qxd
284
Related Approach to Porantherine Porantherine is a Euphorbaceae alkaloid isolated from the woody shrub Poranthera corymbosa Brogn. It is known to poison cattle wherever it grows. It came to light for poisoning cattle in New South Wales and Queensland, Australia. Some other members of this family of alkaloids are shown below.
H
H
N
a
AcO
H
N
a
HO
H
N H
N
N
N
H 1
O Me
Me
O
2
O
3 Me
Me
O
22
O
23
Stevens' notion was that enol (or enolate) addition to an iminium ion (or imine) occurs such that the developing carbon-carbon bond and nitrogen lone pair have an anti-periplanar relationship (a stereoelectronic arguement). Stevens' approach is a variation of the Corey-Balanson synthesis and it takes advantage symmetry. 1. Li (Na), THF O
O HCl
O
OH OH
O
Cl
O
O
PhH, ∆ 80%
24
25
26
2. Change to PhH Cl
O
O
NH
O
O
1. 10% aq. HCl 2. 10% aq. NaOH
3.
N OMe
Alkaloids-3
27
28
80%
isolated as oxalate salt
Page 284
N
N
10:56 AM
Porantheriline Poranthericine Porantherilidine Porantheridine Bz = benzoyl 21 20 1 19 18 Ryckman, D. M.; Stevens, R. V. "Stereorational Total Synthesis of dl-Porantherine" J. Am. Chem. Soc. 1987, 109, 4940-4948. The critical analysis is contained in Stevens' synthesis of coccinelline and precoccinelline: Stevens, R. V.; Lee, A. W. M. J. Am. Chem. Soc. 1979, 101, 7032. Porantherine
Organic Synthesis via Examination of Selected Products
BzO
Me
a
12/21/2010
O
N
b1026
H
H N
Organic Synthesis via Examination of Selected Natural Products
H
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 285
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
285
Alkaloids-3 Porantherine is one member of a family of structurally (and presumably biosynthetically) related alkaloids shown here. One might imagine that a late step in the synthesis (or biosynthesis) of alkaloids 19–21 could also involve capture of an iminium ion by an appropriate nucleophile (to construct the bonds marked “a”). Even porantherilidine (18) contains the 1,3-difunctional relationship needed to develop a synthetic strategy that takes advantage of the Mannich reaction. A related approach to porantherine was reported by the Stevens group about 10 years after publication of the Corey synthesis. The Stevens synthesis emphasized the notion that the stereoelectronic requirement for addition of a nucleophile to an imine (or iminium ion) was development of an antiperiplanar relationship between the nucleophile and the developing lone pair on nitrogen.2,3 This notion also leads to the disconnection of 2 → 3 → 22, and iminium ion 22 was to be prepared by oxidation of piperidine 23. Thus, the major difference between the Stevens and Corey approaches is the choice of a surrogate for the aldehyde required for generation of iminum ion 3. In the forward direction, the lithium reagent derived from 26 was treated with iminoether 27 to provide 28. Hydroysis of 28 gave enamine 29.
O
1. PhSO2NHNH2
N
Me 1
2
87%
PhH, ∆, 24 h
N Me
N Me
85%
32
57-65%
David, M.; Dhimane, H.; Vanucci-Bacque, C.; Lhommet, G. “Efficient Total Synthesis of Enantiopure (-)-Porantheridine” J. Org. Chem. 1999, 64, 84028405.
H
H N
OH N H
O
H2N
O a
a
MeO
a
N
b
O
19
33
Porantheridine
1,3- and 1,5-Difunctional Relationships
34
Alkaloids-4
O 1,2-Difunctional relationship selected because α-amino acid can serve as source of enantiopure material
H2N 35
CO2H
Page 286
2. t-BuLi, ether-hexane
N Me
1
TsOH
10:56 AM
O
Organic Synthesis via Examination of Selected Products
2. aqueous NaOH
12/21/2010
1. Hg(OAc)2 (5 equiv) 5% aqueous HOAc
b1026
31
30b
30a
29
N Me 95%
Organic Synthesis via Examination of Selected Natural Products
N Me
N Me
2. 10% aqueous NaOH Me
b1026_Chapter-08.qxd
N
O
OAc
AcO
1. Isopropenyl acetate (5 equivalents) TsOH, ∆, 48 h
286
O
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 287
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
287
Alkaloids-4 As in the Corey synthesis, an intramolecular Mannich reaction converted 29 to 31. Mercuric acetate oxidation of 31 to the corresponding iminium ion, followed by basification, gave enaminoketone 32. This intermediate does not have the proper stereochemistry for an intramolecular Mannich reaction, but under acidic conditions, epimerization is followed by cyclization to provide 2. The synthesis of porantherine (1) was completed using a Bamford-Stevens reaction to introduce the olefin.4 Let’s briefly look at a synthesis of porantheridine (19) reported by the Lhommet group. Disconnection of the N,O-acetal provided 33 as a key intermediate. The notion was that 33 would “collapse” to provide the target alkaloid. There are stereochemical issues associated with this process that we will address in time. Compound 33 has only 1,3- and 1,5-difunctional relationships and thus, should be straightforward to prepare via intermediates that also contain such relationships. For example, the “a” bonds in 33 could all be constructed using carbonyl addition or Mannich-type reactions. Indeed this was the strategy followed for introduction of the 2-hydroxypentyl sidechain. Disconnection of the “b” bond in structure 33 generated a 1,2difunctional compound. This disconnection did, however, have an advantage. It suggested lysine (35) as an enantiopure starting material.
H H2 N
H
anodic oxidation MeO
N
CO 2H
O
OEt
H N
EtO2 C
38
2. LiEt3BH
94%
OH
N H
CH2 Cl2
OH
2. NaOH, H 2O MeOH (91%)
HO
N CHO CBz
O
43 OBn
Ph 3P
O N CBz
generated with n-BuLi from phosphonium bromide
O anomeric effect controls new stereogenic center
44
85% (2 steps) 42
1,2-Dif unctional relationship to 1,5-Dif unctional relationship with the help of an A-function
1. H2 , Pd/C 2. TsOH 3. 4A sieves
N *
H
H N
O
vs
* N O
O 19 47%
Alkaloids-5
45
Page 288
separated and structure determined by X-ray
O H
O
O
39 91%
Swern oxidation
OBn
H N
4:1
40
41 (COCl)2 , DMSO; then ROH, Et 3N
1. NaH, PhCH2 Br DMF (86%)
H
10:56 AM
N CBz
OBn
ClCO2 Bn, Na 2CO3
H
Organic Synthesis via Examination of Selected Products
OBn
b1026
1. K2CO3 , H 2O EtOH
Lysine
Organic Synthesis via Examination of Selected Natural Products
34
O
O
90%
12/21/2010
TMSOTf CH2 Cl2
O
O
36
35
37
N
MeOH, LiClO4
O
O
b1026_Chapter-08.qxd
H
H 2N
O
288
TMSO
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 289
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
289
Alkaloids-5 Lysine (35) was converted to lactam 36. Anodic oxidation of the lactam provided N,O-acetal 34, chemistry pioneered by the Shono group in Japan.5 An acid-promoted intermolecular Mannich reaction between 34 and 37 provided β-ketoester 38. The stereochemical course of this reaction, at the new stereogenic center in the piperidine ring, follows from the stereoelectronic analysis suggest by Stevens for iminium ion addition reactions (Alkaloids-3). Hydrolysis and decarboxylation of the β-ketoester, followed by reduction of the resulting ketone, gave 39 as a separable mixture of diastereomers. The alcohol was protected as a benzyl ether and the cyclic carbamate was hydrolyzed to give amino alcohol 40. The amine was protected with a Cbz group (40 → 41) and the alcohol was then oxidized to aldehyde 42. Wittig olefination with β-acyl anion equivalent 43, provided 44. Catalytic hydrogenation of the olefin followed by hydrolysis of the acetal in the presence of a “water sponge” gave porantheridine (19) with excellent control of stereochemistry. From the standpoint of steric effects, 19 and its N,O-acetal diastereomer (45) look similar. But 19 enjoys double anomeric stabilization [nitrogen lone pair anti-periplanar C–O bond and oxygen lone pair antiperiplanar to C–N bond], whereas 45 enjoys only “single” anomeric stabilization. It is this feature that most likely controls stereochemistry in the final step of the synthesis.6
b1026_Chapter-08.qxd
290
Biomimetic Synthesis Revisited van Tamelen, E. E.; Foltz, R. L. "The Biogenetic-Type Synthesis of dl-Sparteine" J. Am. Chem. Soc. 1960, 82, 2400
H
N
Oxidation
N
N
N
49
O CH2 O
N
N
AcOH
Hg(OAc) 2
X
H
H
O
H
N N
H
49
Wolff-Kishner
Alkaloids-6
X=O
47
X = H2
46
Page 290
48
46 X = H 2 Sparteine 47 X = O 8-Ketosparteine
HN
N
8
10:56 AM
Mannich
Organic Synthesis via Examination of Selected Products
H
N
NH
O
8
12/21/2010
O
X
b1026
8
Organic Synthesis via Examination of Selected Natural Products
H
H
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 291
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
291
Alkaloids-6 The notion of “biomimetic synthesis” was introduced in Chapter 2 within the context of polyolefin cyclizations. It was suggested early in this chapter that alkaloid syntheses that employ Mannich reactions may also incorporate (in part) a biomimetic strategy. The synthesis of sparteine shown here is an elegant example of such a strategy. The idea was that sparteine (46) might arise from the double intramolecular Mannich reaction suggested by structure 48 (followed by reduction of the 8-keto group). A precursor of 48 might be 49. In the laboratory, 49 was assembled from piperidine, acetone and formaldehyde via a double intermolecular Mannich reaction. Mercuric acetate oxidation of 49 gave 8-ketosparteine (47), presumably via intermediates resembling 48 (stepwise oxidation-Mannich events seem most likely).7 A Wolff-Kishner reduction of 47 completed the synthesis of 46. Yields were not given for this synthesis and thus, the efficiency of this route (in terms of yield) cannot be evaluated. Efficiency in terms of the number of steps cannot be denied.
b1026_Chapter-08.qxd
292
First Total Synthesis of Luciduline Luciduline is a member of the Ly copodium f amily of alkaloids. For a short review see: Inubushi, Y.; Harayama, T. "Total Synthesis of Lycopodium Alkaloids" Heterocycles 1981, 15, 611-635.
N
H
N N
H
51
52
Lycopodine
Annotinine
Annotine
OH OH O
53 Annopodine
54
Serratinine
N
Me
Me
N
H
O MeHN
Me
Me
Me OH
O
Me
H 55
O
O
O
1,6-Dicarbonyl
Luciduline
NC 1. Cl 2. KOH OMe 59
57
56
55
MeO 58 Oxy-Cope
MgBr
H 2. (CH2 OH)2 , H +
250 °C OH MeO 60
O
OR
OH
O
MeO 58
61 R = Me
Alkaloids-7
O
O H
3:2
62 65% from ketone
Page 292
Scott, W. L.; Evans, W. A. "The Total Synthesis of dl-Luciduline" J . Am. Chem. Soc. 1972, 94, 4779.
10:56 AM
50
H
12/21/2010
N
O
Organic Synthesis via Examination of Selected Products
H
HO
O
b1026
O N
CO2 Me
O
HO
Organic Synthesis via Examination of Selected Natural Products
O O
H
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 293
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
293
Alkaloids-7 Let’s look at another alkaloid whose structure screams “Mannich reaction” for an endgame. Luciduline is one member of the Lycopodium family of alkaloids. This is a large family of natural products. A few structures are shown here (50–54). Luciduline (55) is a β-aminoketone. This is precisely the difunctional relationship that results from the Mannich reaction. Thus, a strategy that passes through 56 en route to 55 seems likely to succeed. Aminoketone 56 is a 1,5-difunctional compound and, in principle, should be available using “normal” carbonyl chemistry. We will see an example of this later, but the first synthesis of luciduline approached 56 from 1,6-difunctional intermediate 57. One versatile method for the preparation of 1,6-dicarbonyl compounds is the oxy-Cope rearrangement. We saw this in the Evans synthesis of juvabione (Chapter 5) and indeed, this is the methodology used by Evans and Scott in their synthesis of luciduline (58 → 57). The synthesis began with the preparation of rearrangement substrate 58. A Diels-Alder reaction between α-chloroacrylonitrile (a ketene equivalent) and cyclohexadiene 59 gave 60 after “hydrolysis” of the intermediate cycloadduct (recall the Corey synthesis of prostaglandins). Reaction of ketone 60 with isopropenylmagnesium bromide provided 58. Thermal rearrangement of 58 gave the ketone derived from tautomerization of 61. Regioselective ketal formation provided 62, the enol ether reacting faster that the ketone.
Me
1. TsNHNH 2
O
O O
O 64a
64b
∆
crystallization-equilibration to get material for throughput
NaSPh, MeOH
HO Me
Me
RaNi, EtOH
O
Me
PhS
O
O 82%
O
67
65a
O
85% 65b
O
66
Curtin-Hammett Principle
1. TsCl, pyridine 2. HCl, acetone OTs Me
MeNH 2
MeHN
Me NH
82%
68
MeHN
Me
PhH, ∆, 24 h O
Me
OTs
Me
N
(CH 2O) n isoamyl alcohol
O 69 Azide displacement fails. Possible involvement of N,N-acetal and boat conformation?
Alkaloids-8
56
20 h, ∆
O 55
94% Mannich Reaction
Me
Page 294
O
OH Me
PhS OH
O
10:56 AM
SPh
Organic Synthesis via Examination of Selected Products
12 h OH
b1026
63
Organic Synthesis via Examination of Selected Natural Products
O
62
O Me
H
O
O
12/21/2010
2. MeLi, Et2 O
H
H
m-CPBA
b1026_Chapter-08.qxd
Me
O O
294
O Me
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 295
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
295
Alkaloids-8 The diastereomers of 62 were separated by crystallization, and equilibration of the mother liquors provided more material. It was next required to move the ketone functionality to the adjacent carbon to establish the required 1,5difunctional relationship. Whereas how this was done is very interesting, this adjustment is sort of the bottle-neck in the synthesis. A Bamford-Stevens reaction was used to convert 62 to olefin 63. Epoxidation of 63 occurred from the convex face of the cis-octalin to provide 64. The epoxide was then opened by sodium thiophenoxide to provide 65 and Raney-Ni was used to reduce the C–S bond and provide 67. The epoxide opening is interesting and illustrates how mechanistic thinking must be an integral part of synthesis planning. It is well-known that epoxides are opened by nucleophiles to initially provide an anti-relationship between the nucleophile and the developing alcohol. In cyclohexene oxide this means that products are born with a trans-diaxial relationship (the Furst-Plattner rule).8 Epoxide 64 is likely to exist as an equilibrating mixture of conformations 64a and 64b. One might imagine that 64b would predominate for steric reasons (note the axial ketal oxygen-CH2 interaction in 64a). Anti-periplanar opening of 64b, however, would provide 66 and this was not observed. On the other hand, antiperiplanar opening of 64a would provide the observed product, born as 65a but sure to prefer conformation 65b once formed. An inspection of models shows that attack of thiophenoxide on 64b results in an interaction of the nucleophile with the two axial C–H bonds indicated in the structure. On the other hand, only one axial C–H provides cause for concern as thiophenoxide approaches 64a. Thus, the rate of reaction of the less stable conformation appears to be faster than the rate of reaction of the more stable conformation. This is a lovely example of the Curtin-Hammett principle in operation.9 Recall that the goal was to arrive at aminoketone 56. The projected Mannich reaction requires that the amino group be on the concave face surface of the cis-decalin ring system. Notice that the alcohol stereochemistry in 67 is set such that an SN2 reaction at the carbinol center would establish the required stereochemistry in 56 [We will see another approach to establishing this stereochemistry shortly]. Tosylate formation followed by acetal hydrolysis provided 68, but treatment of this material with azide failed to give any of the desired SN2 product. Treatment of 68 with methylamine, however, gave 56 in excellent yield. Given the results with azide, it is probable that this displacement occurs with intramolecular delivery the nucleophile via involvement of an N,N-acetal (69). The final Mannich reaction proceeded as anticipated to provide luciduline (55).
O
O
Me
H
Me 73
72
71
74
1,3-N,O-Relationship
H
O
H
O NH2 OH-HCl
+ SnCl4 (1 eq) H
CH3 CN
75
74
Me
H 67% (2:3)
H
N
OH
MeOH, NaOAc
Me
Me
H
77 40% overall
76
NaBH3 CN, MeOH Me
Me
N
N
Me
Jones oxidation
Me
1. MeOSO 2F, ether
O
55
78
98%
97%
70 70%
Alkaloids-9
H Me
NHOH
(CH 2O) n
toluene molecular sieves
2. LiAlH4
HO O
N
H 72 100%
Me
Page 296
73
O
10:56 AM
70
55
NHOH +
Me
H
O
H
12/21/2010
Me
Organic Synthesis via Examination of Selected Products
Me
H 2C O N H
b1026
N
N
Organic Synthesis via Examination of Selected Natural Products
Oppolzer, W.; Petrzilka, M. "A Total Synthesis of dl-Luciduline by a Regioselective Intramolecular Addition of an N-Alkenylnitrone" J. Am. Chem. Soc. 1976, 98, 6722-6723. Oppolzer, W.; Petrzilka, M. "An Enantioselective Total Synthesis of Natural (+)-Luciduline" Helv. Chim. Acta 1978, 61, 2755.
Me
b1026_Chapter-08.qxd
296
Oppolzer’s Nitrone Cycloaddition Route to Luciduline
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 297
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
297
Alkaloids-9 I hope it is clear by now that 1,3-N,O relationships appear in many alkaloids because of the manner in which they are made by nature. Since the Mannich reaction generates a 1,3-N,O relationship, it is a very useful reaction for alkaloid synthesis. Taking this a step further, it is probable that any reaction that establishes a 1,3-N,O relationship may find broad use in alkaloid synthesis. If we return to the pyrrolizidine alkaloid syntheses we briefly visited in Chapter 4, it is apparent that nitrone-alkene cycloadditions represent one such reaction. Let’s see how this was applied to a synthesis of luciduline by the Oppolzer group. The plan was to prepare 55 from 70, which would be derived from an intramolecular nitrone cycloaddition of 71. Cis-octalin 71 was to be prepared from hydroxylamine 72 and formaldehyde, a standard method for nitrone preparation. The cycloaddition required that the hydroxylamine reside on the concave face of the cis-octalin. It was felt that this stereochemistry could be established by hydride reduction of an oxime, which would be derived in turn from the corresponding ketone. A cycloaddition between 1,3-butadiene (73) and cyclohexenone 74 was to serve as the starting point of the synthesis. The cycloaddition proceeded smoothly, when promoted by acid, to provide ketones 75 and 76. Whereas 76 was surely the kinetic product, acid promoted enolization-isomerization provide 75 as well. Treatment of the mixture with hydroxylamine gave the desired oxime (77) and Borch reduction from the convex face provided 72.10 The rest of the synthesis proceeded as expected. The intramolecular nitrone cycloaddition gave 70. Methylation of the amine and reduction of the weak N–O bond provided 78. Jones oxidation (CrVI in acid) of 78 completed the synthesis of luciduline (55). When a single enantiomer of 74 was used, a single enantiomer of luciduline was obtained.
6
N H
7
MeO
MeO2C
2. Lactamization H
MeO2C
OMe OMe
OR +
O
MeO 2C OMe
HO 86
Tautomerization slow enough that cycloadduct can be isolated.
HO2C
O
R2
H H
HO2 C
O
87
83
OR OMe
R1
endo-cycloaddition
R2 R1
H HO2C
O
HO 2C
H
Tri-O-methylgallic acid
O H
CHO
MeO2C
OMe 85
H
An exercise in substituted cyclohexane synthesis
OMe
81
O
NH2 N 82 H 6-Methoxytryptamine 6
OMe
+
O
MeO
87
MeO2C 85
Alkaloids-10
84
Page 298
OMe
N H
80
OMe
O
Reserpine
N H
O
H 79
1. Reductive Amination O
10:56 AM
MeO
2
Stereochemical issues N H
12/21/2010
H
1
Organic Synthesis via Examination of Selected Products
3
b1026
4
Organic Synthesis via Examination of Selected Natural Products
Reserpine was isolated in 1952 from the Indian snake-root, Rauwolf ia ser pentina Benth. The structure was determined by 1955 and the compound quickly became important for the treatment of nervous and mental disorders. The first total synthesis of reserpine was reported a year later and the compound has remained a popular target for synthesis throughout the past half century. We will examine six syntheses that span this time period. We will begin with the Woodward group synthesis: Woodward, R. B.; Bader, F. E.; Bickel, H.; Frey, A. J.; Kierstead, R. W. "The Total Synthesis of Reserpine" J. Am. Chem. Soc. 1956, 78, 2023-2025. Woodward, R. B.; Bader, F. E.; Bickel, H.; Frey, A. J.; Kierstead, R. W. "A Simplified Route to a Key Intermediate in the Total Synthesis of Reserpine" J. Am. Chem. Soc. 1956, 78, 2657. Woodward, R. B.; Bader, F. E.; Bickel, H.; Frey, A. J.; Kierstead, R. W. "The Total Synthesis of Reserpine" T et rahedron 1958, 2, 1-57. 5
b1026_Chapter-08.qxd
298
Reserpine
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 299
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
299
Alkaloids-10 I will next move to reserpine. One reason is that, in all approaches to this alkaloid, the final reaction used to assemble the pentacyclic skeleton is formally a Mannich-type reaction (a Bischler-Napieralski reaction or a Pictet-Spengler reaction).11,12 A more honest reason is that this alkaloid is a classical target in natural product synthesis that has retained the interest of synthetic organic chemists for more than a half-century, including myself. As we will see, there are some good lessons to be learned from approaches to this alkaloid. Reserpine (79) was isolated in 1952 from the Indian snake-root, Rauwolfia serpentina Benth. The structure was determined by 1955 and the compound quickly became important for the treatment of nervous and mental disorders. The first total synthesis of reserpine was reported a year later and the compound has remained a popular target for synthesis throughout the past half century. We will examine six syntheses that span this time period. We will begin with the Woodward group synthesis. It is notable that this synthesis was being conducted about the same time that this group was developing their approach to steroids. In fact, the steroid work and the reserpine synthesis share some features, particularly in the early stages. The plan was to disconnect reserpine into indole 80 and tri-O-methylgallic acid (81). Indole 80 was to be converted to reserpine using a variant of the well-known Bischler-Napieralski synthesis of isoquinoline derivatives. This transformation (80 → 79) generates a stereogenic center and thus, there are some stereochemical issues associated with this portion of the synthesis. They are rather fascinating and, in time, we will see how various groups addressed this issue. Lactam 80 was projected to come from 6-methoxytryptamine (82) and aldehyde 83. It was anticipated that reductive amination of the aldehyde would be followed by cyclization of the intermediate aminoester to provide the lactam. Aldehyde 83 is a densely functionalized cyclohexane. Disconnection of the vicinal OR groups in 83 led to cyclohexene 84 as a possible intermediate, which was to be prepared by a Diels-Alder reaction between pentadienoic acid 85 and an appropriate cis-1,2-disubstituted ethylene. Whereas there might be regiochemical issues associated with the Diels-Alder reaction, the expected endo-cycloaddition would generate the proper relative stereochemistry at the three contiguous stereogenic centers in 84. Furthermore, the regiochemical issue disappeared when p-benzoquinone (86) was selected as the dienophile. As anticipated, the initial cycloaddition proceeded smoothly to provide 87. The next stages of the synthesis were two-fold in nature: (1) the olefin had to be converted to a vicinal diol derivative with control over stereochemistry and regiochemistry and (2) the enedione
b1026_Chapter-08.qxd
300
Some Chemistry of the Cycloadduct H
H
H
R O
O OH R
HO 89
R O
O
87
OAc
R = CO 2H
87a
Al(OiPr)3
H
87b
92
O
O
HO
O O
i-PrOH H H
O
O 91
i-PrOH
MeO HO
H MeO2 C
H O
NaOMe
H R OH
OH CO2 H
95a
88
O
HO 95b
93
O
O
MeOH
MeO
O
94
Br2 , MeOH OAc MeO MeO2 C
NBS
NaOMe H
Br H O
CHO OMe 83
HO
OH CO2 H
H H
96
O Br 97
O
O
Alkaloids-11
MeOH
O MeO 94
O
O
H 2SO4 , H 2O
O
O
Page 300
R H
O
10:56 AM
Al(OiPr)3 H
H O
H
H
H
90b
Organic Synthesis via Examination of Selected Products
O
90a
Ac 2O, NaOAc O
O
O OH R
H H
O
NaBH4 H
HO2 C
O
R O
89b
89a
MeOH, H+
H
benzene dioxane
b1026
MeO2 C
HO
12/21/2010
PhCO3 H
Organic Synthesis via Examination of Selected Natural Products
HO OH
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 301
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
301
substructure had to be degraded in an unsymmetrical manner to reveal the aldehydo-ester needed for reaction with 6-methoxytryptamine (82).
Alkaloids-11 The plan for accomplishing these two tasks, and partial execution of the plan, is outlined here. It was hoped that 87 could be converted to 88, which has the vicinal diol problem solved and is poised to reveal the aldehydo-ester in 83 (R = Ac) via oxidative cleavage of the enone. Enedione 87 was a sensitive compound. It is at the same oxidation state as p-hydroquinone 89 and indeed, acid promoted this transformation. Reduction of 87 with sodium borohydride, however, differentiated the two carbonyl groups to provide 89 (R = CO2H). The stereochemistry of the reduction can be rationalized by attack of hydride from the convex face of 87. The regiochemistry can be rationalized by reduction of the sterically more accessible carbonyl group, if one assumes that reduction occurs from the presumably more stable conformation 87b. Introduction of the differentiated vicinal diol came next. This transformation required regioselective anti-periplanar addition of two oxygens to the more electron-rich olefin in 89. One oxygen had to be delivered from the convex face of 89, and the other from the concave face. This was accomplished by epoxidation of 89 to provide 90, stereochemistry being controlled by the bowl-shape of 89. The next task was to regioselectively open the epoxide with “hydroxide” from the concave face of 90. This required several steps, but was eventually accomplished by intramolecular delivery of the nucleophilic oxygen as follows (recall the conversion of 68 to 56 in the luciduline synthesis). Dehydration of 90 gave δ-lactone 91. Reduction of the ketone in 91, from the only accessible face of the carbonyl group was followed by γ-lactone formation. This froze the substrate in a conformation related to 90b, in which the hydroxyl group was properly disposed for anti-periplanar opening of the epoxide to provide 92. Base-promoted elimination of the alcohol gave unsaturated ester 93, and treatment of 93 with sodium methoxide introduced the methoxyl group with proper stereochemistry (addition of the nucleophile to the convex face of unsaturated ester). A streamlined route from 87 to 94 was also developed. Thus, MeerweinPondorf-Verley reduction of 87 provided diol 95. Treatment of 95 with bromine in methanol gave β-bromolactone 97, presumably via intermediate bromonium ion 96. Reaction of 97 with sodium methoxide gave 94. Note that this route takes advantage of the same stereochemical principles followed by the epoxide route to 94.
O
O
98
AcOH
O
O
100
99
OH
MeO
O
OH MeO
O
101a
OH
O
O
OH
O
1. CH 2N 2
OAc
MeO2C
OMe
CO2 H
CO2 Me OAc
2. Ac2 O, pyridine
OMe
103
OMe OAc
102 1. HIO4 , H 2O
OMe OH 101b
102
2. CH 2N 2 O
OMe CHO
1. 6-Methoxytryptamine 2. NaBH4
MeO2C
OAc OMe 83
MeO
N H MeO2 C
HN
MeO
MeO2C 104
OAc OMe
Alkaloids-12
N H
O
POCl3
N
MeO2C 80 (R = Ac)
OAc OMe
Page 302
H
KClO 3 MeO2C
O
O
H
OsO4
10:56 AM
O
Organic Synthesis via Examination of Selected Products
OH
12/21/2010
O MeO
b1026
O
AcOH
OH
O
Organic Synthesis via Examination of Selected Natural Products
O
β-elimination
Zn
CrO 3 MeO
b1026_Chapter-08.qxd
302
Br
Br
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 303
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
303
Alkaloids-12 The next task was to convert 94 to unsaturated ester 102 (same as 88) in preparation for the oxidative cleavage reaction to provide 83. This transformation began by treating 94 with aqueous bromine to provide bromohydrin 98. The regiochemistry and stereochemistry of this transformation can once again be rationalized using a variation of the Furst-Plattner rule.8 Antiperiplanar opening of the bromonium ion derived from addition to the convex face of 94, provides 98. Oxidation of the alcohol, followed by reductive cleavage of the α-C–O and C–Br bonds in 99, followed by an acid-promoted β-elimination from 100, provided 101. Notice that this sequence springs 99 open to 101a, most likely the least stable conformation of 101. Conversion of the acid to a methyl ester (with diazomethane) and esterification of the secondary alcohol completed the synthesis of 102 (same as 88). Oxidative cleavage of the enone was accomplished in two steps (diol formation followed by periodate cleavage of 103) and esterification of the resulting acid gave 83. The synthesis continued as expected. 6-Methoxytryptamine reacted with 83 to provide the expected imine, which was reduced to provide lactam 80, presumably via amino ester 104.
H
N CO2Me
N
OMe
105 105
OMe
106
OAc anti-relationship of entering hydride and developing nitrogen lone pair
OMe OAc
1. KOH 2. Dicyclohexylcarbodiimide (DCC)
MeO
N H
N
1. MeOH N H
2. ArCOCl O
H MeO2C
OMe
O
O
N
O OMe
107
OMe
79 OMe
3
MeO
H1 108
OMe Reserpine
O H MeO
N H1
(CH3)3CCO2H
ON H
OMe Thermodynamics control this isomerization. There are several possible mechanisms. One begins with protonation at C-3 of the indole, followed by loss of the “H1” proton. The other involves a retro-electrophilic aromatic substitution reaction followed by an electrophilic aromatic substitution.
Alkaloids-13
Page 304
H
10:56 AM
OMe OAc
OMe
Organic Synthesis via Examination of Selected Products
MeO2C
N H
CO2Me
b1026
N H
12/21/2010
N
N H
Organic Synthesis via Examination of Selected Natural Products
MeO
b1026_Chapter-08.qxd
304
isoreserpine stereochemistry
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 305
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
305
Alkaloids-13 Now we come to the critical Bischler-Napieralski cyclization. Treatment of 104 with POCl3 gave iminium ion 105 via amide activation and an intramolecular electrophilic aromatic substitution. Reduction of this iminium ion with sodium borohydride gave 106. This material has the opposite stereochemistry (isoreserpine stereochemistry), at the new stereogenic center, to that required by reserpine. There are at least two explanations for this observation: (1) the reduction proceeds from the convex face of 105 and (2) the stereoelectronics of the iminium ion addition favor this reduction (recall the Stevens synthesis of porantherine). How was this stereochemical “mistake” fixed? Hydrolysis of the esters and lactonization of the resulting hydroxy acid converted 106 to 107 (note that conformations available to cis-decahydroisoquinolines can be treated in the same way that one analyzes conformations of cisdecalins). The transformation generates some serious 1,3-diaxial interactions that can be relieved by epimerization of 107 to 108 (see “H1” hydrogen). One would expect 108 to be thermodynamically more stable than 107. It turns out that this epimerization was accomplished by heating 107 with an acid (pivalic acid). There are several possible mechanisms one can imagine for this transformation. One begins with protonation at C3 of the indole, followed by loss of the “H1” proton. The other involves a retro-electrophilic aromatic substitution reaction followed by an aromatic substitution reaction. Opening of the lactone with methanol, followed by esterification of the secondary alcohol, completed the synthesis of reserpine (79).
O
RO
OMe Reserpine
MeO2 C
OMe
O
110 MeO2 C
OR 83
OMe
OR OMe
OMe
MeO2 C
OR OMe
109
Address stereochemistry and regiochemistry issues by making the cycloaddition intramolecular
OMe
THF
113
112
+ OH
HO 2C
2. reflux, 1 h 3. H2 O
HO 2C
114
94%
OH
HO 2C
OH
OH
major
minor O
O
O
O MeO 2C
O
119 30%
O
116
115
2. H2 SO4 , MeOH ∆(95%)
O
O
CaSO4 , Ag2 O + MeOH, ∆
MeO 2C 118
OMe
MeO 2C 100% (1:1)
117
Alkaloids-14
OMe
OH
6% 1. Ag2 O, MeI CaSO4 (70%)
Br
O
0.003M OMe
O
O hν, acetone
O
180 °C
OH
MeO 2C OMe 110
Page 306
O 1. HCO3 H, rt, 30 h
Li, NH 3
10:56 AM
MeO2C
HO2 C
111
RO 2C H
79
O
Organic Synthesis via Examination of Selected Products
N H
RO 2C
12/21/2010
MeO
N H
H
b1026
O
OH CHO
Organic Synthesis via Examination of Selected Natural Products
Pearlman, B. A. "A Method for Effecting the Equivalent of a deMayo Reaction with Formyl Acetic Ester" J. Am. Chem. Soc. 1979, 101, 6398. Pearlman, B. A.; "A Total Synthesis of Reserpine" J. Am. Chem. Soc. 1979, 101, 6404.
H
b1026_Chapter-08.qxd
306
Another Route to the Functionalized Cyclohexane
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 307
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
307
Alkaloids-14 We will now look at several other syntheses of reserpine, some in more detail than others. Here we see an approach reported by Pearlman. The plan was related to the Woodward approach at a late stage, as 83 was projected to be an intermediate. The approach to 83, however, was very different. The plan was to prepare strained β-hydroxyester 109 and then spring it loose to 83 by a retro-aldol reaction. The cyclobutane was to be prepared by a photocycloaddition between cyclohexene 110 and enol 111.13 The stereochemical and regiochemical issues associated with the cycloaddition were to be handled by making the photocycloaddition an intramolecular process. It is notable that this is an example of a synthesis that was designed to develop and illustrate new synthetic methodology, in this case a method for the “directed” syn-addition of two functionalized carbons across a carbon-carbon π-bond. The synthesis began with reduction of benzoic acid (112) to provide 113. Epoxidation of 113, followed by opening, provided diols 114 (major) and 115 (minor). Heating the mixture of hydroxy acids provided lactone 116 in low yield. Alkylation of the free hydroxyl group followed by methanolysis of the lactone gave 110, the desired photocycloaddition substrate. The desired course of the photocycloaddition was “cis to the hydroxyl group with formyl group vicinal to the hydroxyl group”. This is demanding (does not work) without the use of intramolecular direction. In the end, 110 was converted to a mixture of diastereomeric acetals 117 and 118, the latter of which was disposed to undergo the desired photocycloaddition to provide 119.
b1026_Chapter-08.qxd
308
via dimethyl acetal O
O
O
O
OMe
MeO 2C
MeO 2C
RO 2C
2. ArCOCl 121
OMe
OMe
MeO2 C
OCOAr
3. AcOH, rt, 23 h
OMe
59%
91%
119
CHO 1. MeOH, H 2 SO 4
MeO 2C
CH2 Cl2 120
O
83
3. (ArCO)2O
MeO
N H
N H
MeO2C
OAc
O
H
79 MeO 2C
MeO
OMe
OMe Reserpine
123
O
N 3:2
122
OMe
O
N H
MeO2C
OCOAr OMe
26%
OMe OMe
Recall MeO
MeO 2C
N OAc
61
OH
Luciduline Intermediate
124
OMe
MeO 2C
Alkaloids-15
N 125
+ CO2 Me
MeO2 C 126
OAc
Page 308
H
CO 2Me N
epimerization problem
10:56 AM
2. NaBH4 , MeOH
Wender, P. A.; Schaus, J. M.; White, A. W. "General Methodology for cis-Hydroisoquinoline Synthesis: Synthesis of Reserpine" J. Am. Chem. Soc. 1980, 102, 6157.
Organic Synthesis via Examination of Selected Products
1. 6-methoxytryptamine PhH, MeOH
b1026
H 2SO4
O
CF3CO3 H
12/21/2010
O
AcO
O
Organic Synthesis via Examination of Selected Natural Products
MeOH
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 309
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
309
Alkaloids-15 The conversion of 119 to retro-aldol substrate 121 was accomplished by hydrolysis of the acetal and esterification of the resulting acid to give 120. This process was accompanied by epimerization of both the ester and the resulting methyl ketone, presumably to relieve strain. Baeyer-Villiger oxidation of 120 completed the transformation to 121.14 The sensitive aldehyde was then converted to 122 using chemistry analogous to that developed by the Woodward group. The bottom line is that this is a creative plan that had trouble when it came to the details (tactics). The Wender group (then Harvard and now Stanford) reported another approach. The idea was to prepare reserpine from an enamide of type 123. Modification of the N-carbethoxy group and protonation of the enamine/enamide would afford the required pentacyclic skeleton. Reduction of the cyclohexene, with control of stereochemistry, would also be required to establish the five contiguous stereogenic centers of reserpine. Notice that 123 is a heterocyclic analog of luciduline intermediate 61. It is a 1,5-hexadiene. This is a recognition key for a Cope rearrangment. Working backward leads to azabicyclo[2.2.2]octane 124, which was to be prepared by a Diels-Alder reaction between N-acyldihydropyridine 125 and dienophile 126. Let’s see what happened in the lab.
N
CO 2 Me
OAc
MeO 2C
N
THF, -78 °C
127
rt MeO 2C
N 89%
Et 2O, 0 °C
OMe
65%
123 78%
129
Loss of oxidation state difference is unfortunate
CO 2Me N
CO 2Me N
H N
EtOAc
OMe 14% hydrogenolysis
OAc OMe
133
79%
AcO 134
OAc
Hydrogenation of ketone gives 2:1 mixture of diols
O
OMe
OMe 130
78%
95%
131
NaOMe gives 6:1 Therefore stereochemistry is a result of kinetic protonation. 4% each of epimeric diols at C18.
MeO
N H 135
OMe 90%
2. AcCl, -78 °C
18 17
MeOH, ∆, 25 h
18
AcO
HO
OAc
RBr, K2 CO 3
Me3 Si-I
H 2, Pd/C
132
MeO 2C
85%
N
AcO
OAc OMe
Next task is oxidative cyclization to provide the pentacyclic indole skeleton. At this point it is apparent why loss of the enamide is a shame.
Alkaloids-16
Page 310
OAc
10:56 AM
LiAlH 4, 2 min
H 2, Pd/C
MeO 2C
CO 2Me N 1. LiN(TMS) 2 -10°C, THF
CO 2Me N
Organic Synthesis via Examination of Selected Products
CO 2Me N
b1026
N
OAc
3. CH2N 2
For method see Wender, P.A .; Schaus, J. M.; Torney, D. C.; Tetrahedron Lett. 1979, 2485.
OMe
AcO
2. CF3 CO 2H
128
12/21/2010
60%
EtOAc MeO 2C
OtBu 1. Ac2 O, Et 3 N, 4-DMAP
endo-cycloadduct
CO 2 Me Xylene, ∆, 243 °C
AcO
O
2:1
Organic Synthesis via Examination of Selected Natural Products
toluene, 56 h
from pyridine, methyl chloroformate and NaBH4
HO
CO 2 Me LiCH 2CO 2 tBu
hydroquinone,120 °C
125
2:1
b1026_Chapter-08.qxd
MeO2 C
AcO
310
O 126
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 311
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
311
Alkaloids-16 The initial Diels-Alder reaction worked well to provide a mixture of endo (127) and exo (not shown) cycloadducts. A crossed Claisen condensation provided 128, and a series of functional group manipulations provided Cope substrate 129. The Cope rearrangement proceeded well to give 123 in excellent yield. Unfortunately a number of problems were encountered thereafter. It was not possible to reduce the enediol derivative without also reducing the enamide. For example, catalytic hydrogenation of 123 provided 130. The enamide is what differentiated oxidation states at the carbons adjacent to nitrogen and thus, we will see, it is unfortunate that differentiation was lost at this point of the synthesis. Continuing from 130, lithium aluminum hydride reduction converted the carbomethoxy group to a primary alcohol, and cleaved the enol acetate to afford an enolate. Kinetic protonation of the enolate afforded 131, although thermodynamics also favored 131 over its C17 diastereomer. Catalytic hydrogenation of 131 gave a 2:1 mixture of alcohols at C18. On the other hand, conversion of 131 to enol acetate 132, followed by catalytic hydrogenation, provided 133 with the desired C18 stereochemistry. The next stage of the synthesis required introduction of the indolylethyl group and oxidative cyclization to afford the pentacyclic skeleton of reserpine. First the carbamate was removed from decahydroisoquinoline 133 to give 134. Alkylation then gave 135. The next task was oxidative cyclization to provide the pentacyclic indole skeleton. At this point it becomes clear why loss of the enamide was unfortunate.
b1026_Chapter-08.qxd
312
MeO
N H
1. Hg(OAc)2
N
30%
45%
MeO
2. NaBH4
H
H H OCOAr
137
OMe
OMe
MeO
1. NaBH4
N H
MeO
N
2. acid
N H H
H 85%
1. Me2 C(CN)(OH), Et 3N H
MeO2C
139
OCOAr
R
140
2. DMSO, (COCl)2 , Et3N 3. MeOH
138
R
OCOAr OMe
R = CO 2Me (33%)
R = CHO (65%)
R = CHO (43%)
R = CH 2OCH 2SMe (20%)
N
N H H 79
H H
A Mechanism Problem
MeO
N
N H H
OMe
OMe isoreserpine
OCOAr
MeO
formic acid, formamide, ∆
H H
MeO2C
OCOAr
Useful reaction for use in relay syntheses.
MeO
N
N H 141
H H
MeO2C
OCOAr OMe
OMe
Alkaloids-17
Page 312
N H H
10:56 AM
OH
Organic Synthesis via Examination of Selected Products
DMSO, DCC H 3PO4 (Moffat Oxidation)
12/21/2010
HO
via diester
OH
N
N H H
2. KOH, MeOH, 5 min (65%)
b1026
136
OMe
135
HO
45%
OAc
MeO
H
Organic Synthesis via Examination of Selected Natural Products
AcO
1. ArCOCl (XS) (53%)
N
N H H
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 313
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
313
Alkaloids-17 Oxidative cyclization of 135 occurred with little regioselectivity. Oxidation at C3 afforded isoreserpine analog 136, and a nearly equal amount of material derived from oxidation at C21 was obtained (so-called inside isomers). Based on earlier work, it is probable that the sequence of events leading to 136 involve oxidation of 135 to an iminium ion, cyclization via an electrophilic aromatic substitution reaction, oxidation of the resulting cyclized products to an iminium ion related to 105 (see Alkaloids-13) and finally, reduction (NaBH4) of that iminium ion to provide 136. Conversion of 136 to isoreserpine (140) was relatively straightforward. I will leave the details to you (see problems for guidance). Since isoreserpine had previously been converted to reserpine (see Wooward synthesis for example), this constituted a synthesis of 79.
18 17
MOMO
seco-loganin
O 143
142 Cl
O
Bn
Bn O
NH
N
O O
149
147
OMOM 60% overall 148
Bn
O
O
5 Steps
HO
CO 2Me
OMOM 89% 146
Alkaloids-18
N
xylene, ∆ -CO2
O
MOMO
18 17
144
m-CPBA
Page 314
O
O
10:56 AM
N
Biosynthetic precursor of pentacyclic indole alkaloids
12/21/2010
O
H
Organic Synthesis via Examination of Selected Products
MOMO
O OHC
Bn
N
OGlu
b1026
Bn
H
Organic Synthesis via Examination of Selected Natural Products
Martin, S. F.; Rueger, H.; Williamson, S. A.; Grzejszczak, S. "General Stategies for the Synthesis of Indole Alkaloids. Total Syntheses of dl-Reserpine and dl-α-Yohimbine" J. Am. Chem. Soc. 1987, 109, 6124-6134. R R R N O N O N O N MeO N O H H O H O 20 H MOMO 79 MOMO O 18 17 MeO2C OTMBz OMOM 146 144 145 OMe
Original plan abandoned because equilibrium is greater than 9:1 on the side of the furan.
b1026_Chapter-08.qxd
314
Intramolecular Diels-Alder Route to Reserpine
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 315
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
315
Alkaloids-18 Martin (University of Texas) described a synthesis that followed the plan depicted here. The ultimate idea was that perhydroisoquinoline 144 could be converted to 79 using the conjugated diene as a handle for introducing the C17-C18 diol and C20 stereogenic center. This diene was to come from 146 via a Diels-Alder (to provide 145) retro-Diels-Alder (to provide 144) sequence. Actually, the original plan was to prepare 142 via an intramolecular Diels-Alder of 143. It is apparent that this plan took into account the needed substituents at C17 and C18, but it was abandoned because the equilibrium for this reaction was well to the side of the furan (143). The synthesis began with the preparation of amine 148 from propargyl alcohol (147). Acylation of the amine with 149 gave 146, and the proposed Diels-Alder reaction proceeded smoothly to give 144. The remote double bond of 144 was the most electrophilic double bond and thus, epoxidation from the sterically most accessible face gave 150 as the major product.
O
O
17
17
* OCOR
98%
OMe 90%
Pearlman’s Catalyst
Bn
N
N
N
O
AlH3
MeO2C
OMe
HO
OCOR
OCOR OMe
OMe
157 80%
156
75%
155
AcO 154
OAc OMe
Wender Intermediate
1. TsOH, MeOH (81%) 2. TMBzCl (84%) (TMBz = 3,4,5-trimethoxybenzoyl) Bn N
H N
Pd(OH)2 , H2
RBr, DMSO
MeO
AcOH
Endgame similar to Wender (Wenkert) transformations but analyzed in some detail
N
N H
H H
MeO2C
OTMBz OMe 158
MeO2C 159
OTMBz OMe
160
MeO2C
OTMBz OMe
69%
18 steps from propargyl alcohol overall yield = 17%
Alkaloids-19
Page 316
OCOR
N
O
1. PDC, DMF 2. CH 2N 2
MeO2C
Bn
Bn
10:56 AM
TsOH, MeOH, 45 °C Bn
Organic Synthesis via Examination of Selected Products
1. AlH3 2. H3 O+ 3. Ac2 O, 4-DMAP
*6% of other cis-perhydroisoquinoline
b1026
153
OMe
12/21/2010
MeOH
MOMO
Organic Synthesis via Examination of Selected Natural Products
OCOR
152
O *
Pd(OH)2 /C
Lots of experiments were required to overcome problems such as acyl transfer and regiochemistry
88% + 5% diepoxide
N
H 2 (1800 psi)
MOMO
OCOR 90% OH 151
DME
150
O
MeI (solvent) MOMO
18
O
N
Ag2 O, CaSO 4
CO2 H MOMO
Bn
Bn
N
b1026_Chapter-08.qxd
Bn
CO2 Li
N
316
Bn
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 317
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
317
Alkaloids-19 Epoxide 150 was then opened with lithium 2-ethylhexanoate to provide 151. This was not the first choice of nucleophile! Lots of experiments were required to overcome problems such as acyl transfer and regioselectivity. There is no doubt that “search” is an important part of the word “research”. Williamson etherification of the C17 alcohol gave 152, and catalytic hydrogenation of the olefin provided 153 as the major product. This was converted to the Wender intermediate 154 (133) in a straightforward manner. As an alternative, 153 was converted 157 by a series of oxidation state adjustments. The ester 157 was then converted to 158 and hydrogenolysis of the N-benzyl group provided 159. Introduction of the 6-methoxytryptophyl group provided 160 to set the stage for the now-familiar endgame.
N H
MeO
140
MeO2C
8%
OTMBz OMe
161
2. Isoreserpine is quickly oxidized by Hg2+ to give an iminium ion that is reduced by Zno to both reserpine and isoreserpine.
This observation suggests that the initial iminium ion cyclization takes place with some stereoelectronic control
BzMTO
BzMTO OMe
BzMTO OMe
N
OMe
N major MeO 2C
N
H 3
+ MeO 2C
3
N
N H
N H
79
161 reserpine, not easily oxidized OMe
OMe
Alkaloids-20
H H
MeO 2C
140
isoreserpine, easily oxidized
Page 318
1. Reserpine is only slowly oxidized to the corresponding iminium ion by Hg2+.Therefore the reserpine (or a portion of the reserpine) represents a kinetic cyclization product.
H
10:56 AM
Observations about Oxidation-Reduction Sequence
H
MeO2C
isoreserpine
Organic Synthesis via Examination of Selected Products
N
N H
OTMBz OMe
OMe
starting material (10%) inside isomers (18% + 4%)
MeO
H MeO2C
OTMBz
35%
reserpine
H
b1026
OMe
N
N H H
H
79
3. Zn, HClO4 -acetoneH2 O (1:1:1), ∆
MeO
H
12/21/2010
OTMBz
N
N H H
2. H2 S
H MeO2C
1. Hg(OAc)2 , AcOH H2 O, 85-90 °C
Organic Synthesis via Examination of Selected Natural Products
160
H
b1026_Chapter-08.qxd
MeO
318
inside N
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 319
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
319
Alkaloids-20 The Martin group used an oxidation-reduction sequence that was similar to that used by Wender. The reaction conditions, however, were slightly different and Zn-perchloric acid was used in the “reduction” step rather than sodium borohydride. The results differed from those obtained by Wender in that reserpine (79) was the major product and isoreserpine (140) the minor product. “Inside” isomers were also obtained due to lack of regioselectivity in the mercuric acetate oxidation. Several observations were made during the course of this transformation. First, reserpine is only slowly oxidized to the corresponding iminium ion by mercuric acetate (see 105 on Alkaloids-13 for a related example where the TMBz group = Ac). Therefore it was proposed that the reserpine isolated from this reaction (or a portion of the reserpine) represented the kinetically prefered product formed from cyclization of iminium ion 161. This is consistent with the stereoelectronic arguments for iminium ion-nucleophile reactions set forth by Stevens. It was also shown that isoreserpine was quickly oxidized by mercuric acetate to provide the corresponding iminium ion (see 105 on Alkaloids-13 for a related iminium ion), which was reduced by Zn0 to give a mixture of both reserpine and isoreserpine. Examination of the most stable conformations of 79 and 140 (at least toward C3) illustrates why one might expect to see such a difference in oxidation rates between these isomers. The C3-H and nitrogen lone pair (the site of “attack” by the Hg2+) have an anti-periplanar arrangement in isoreserpine, but not in reserpine! The bottom line is that the Martin synthesis begins to address the issue of directly obtaining reserpine, rather than having to rely on an isomerization of isoreserpine. We will revisit this later in this chapter.
b1026_Chapter-08.qxd
320
Enantioselective Synthesis Starting from the Chiral Pool Hannessian, S.; Pan, J.; Carnell, A.; Bouchard, H.; Lusage, L. "Total Synthesis of (-)-Reserpine Using the Chiron Approach" J. Or g. Chem. 1997, 62, 465-473
3
O
MeO2C
O
3
O MeO 2C OTMBz
OMe
O
OTBS
OTBS
O
OMe
OMe
OTBS
CO2 H
HO
OH
OMe
OH
164
163
162
HO
165 quinic acid
O
CO2 H 1. TsOH, DMF, PhH
HO
OH
2. Bu2 SnO, BnBr
OH
O
KH, MeI
HO BnO
165
O
O O
MeO
THF
166 81%
O
2. NaIO 4, RuO2-H2 O EtOAc, acetone , H 2O (90%)
BnO
OH
1. Pd(OH) 2 on C H2 , MeOH (98%) MeO O
OMe
OMe
168
167 90%
KHCO3 , MeOH
CO 2Me
I
CO 2Me
1. ClCH 2 CO 2H, CH2 Cl2
O
OTBS
2. NaI, CH 3CN
TBSMSOTf
CH2 =CHMgBr
DCC, 4-DMAP
O
HO
CO 2Me
CO 2Me
OTBS
THF
O
OTBS
PhH
O
OH OMe
OMe
OMe
171
170
164
169
80% overall
90%
90%
85-90%
OMe
Alkaloids-21
2,6-lutidine
Page 320
HO
10:56 AM
79
H 20
H
20
CO 2Me
carbonyl addition
Organic Synthesis via Examination of Selected Products
N H H
CO 2Me
X
radical cyclization
12/21/2010
MeO
21
b1026
CO 2Me N
Organic Synthesis via Examination of Selected Natural Products
21
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 321
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
321
Alkaloids-21 The syntheses we have examined thus far do not take into account enantioselectivity. The Hannesian group described a synthesis that used quinic acid (165) as an enantiopure starting material. The idea was to prepare reserpine (79) from penta-substituted cyclohexane 162. The incipient C3 was to come from an intermediate with that carbon at the aldehyde oxidation state (to be derived from reduction of the lactone). The incipient C21 was to be introduced at the carboxylic acid oxidation state. This plan avoids issues with formation of “inside” isomers as we saw in the Wender and Martin syntheses. Lactone 162 was to come via a radical cyclization of an intermediate of type 163. Reduction of an intermediate radical from the convex face of the cisbicyclic ring system was to control stereochemistry at C20. Compound 163 was to be prepared from 164, which was to come from quinic acid (165) through a series of protections and oxidation state adjustments. The protection of quinic acid was accomplished by γ-lactone formation, benzyl ether formation at the least hindered secondary equatorial alcohol, and methylation of the remaining two alcohols. The secondary alcohol of the resulting 167 was deprotected (hydrogenolysis) and the alcohol was oxidized to provide ketone 168. A transesterification followed by a β-elimination provided 169. Protection of the secondary alcohol provided intermediate 164. Axial delivery of a vinyl group to ketone 164 gave 170. Esterification of the tertiary alcohol and a Finkelstein reaction gave free radical cyclization substrate 171.15
O O
OTBS
AIBN, PhH, ∆
disiamylborane, THF
O
O O
OTBS OMe 162
90% (10 steps, 24% overall)
HO OTBS
N H H
6-methoxytryptamine
OMe
N
O
+
MeO2 C
"iso"
173 98%
174
HO
3
MeO
N H H
"reserpo"
MeO2 C
OTBS
2. BH 3-THF; then HPMA
HO
175
OMe
N
1. TMSOTf, CH2 Cl2 2,6-lutidine O
OTBS OMe
92% iso:reserpo = 1:1.4
MeO
N H H 84% 176
1. HF, CH3 CN
N
2. TBDMSOTf 2,6-lutidine
MeO2 C TMSO
OTBS OMe
3. SmI2, HMPA ethylene glycol THF
MeO
N H H
1. HF, CH3 CN
N
MeO
N
80%
30-32% 177
N H H
2. TMBzCl
MeO2 C
OTBS OMe
79
MeO2 C
OTMBz OMe
20 steps, 2.6% overall
Alkaloids-22
Page 322
O MeO2 C
MeO
3
10:56 AM
Me 3CCO2H
Organic Synthesis via Examination of Selected Products
Me 3CCO2H CO2 Me
12/21/2010
172
73%
radical reduction from convex face
O MeO2 C
b1026
3. NaClO2, t-BuOH Me2 C=CHMe NaHPO 4, H2 O 4. CH2 N2
OTBS OMe
Organic Synthesis via Examination of Selected Natural Products
OMe 171
CO2 Me
1. O3 2. Me 2S
b1026_Chapter-08.qxd
CO2 Me n-Bu3 SnH
322
CO2 Me
I
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 323
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
323
Alkaloids-22 The projected free radical cyclization proceeded as planned to give 172. Ozonolysis of the vinyl group, oxidation of the resulting aldehyde to an acid, and alkylation with diazomethane provided projected intermediate 162. Reduction of the lactone provided 173. Treatment of 173 with 6-methoxytryptamine and pivalic acid then provided a nearly equal mixture of lactams 174 (isoreserpine stereochemistry at C3) and 175 (reserpine stereochemistry at C3). The correct C3 stereoisomer was moved forward to 176 (protection of the tertiary alcohol followed by reduction of the lactam). The silyl ethers were removed, the secondary ether was re-protected, and reaction with samarium iodide accomplished reduction of the α-hydroxy ester to provide 177. Removal of the TBS group and esterification of the alcohol completed the synthesis of reserpine.
NH2 RO
MeO 2C
MeO 2C
N H
N H
N H 79
+
OMe
TsO OHC
178C
179 MeO 2C
OMe Key Intermediate
OMe
6-methoxytryptamine
RO OMe C-ring
N
RO H
H
NH
RO NH OMe
OMe N 140
H
MeO 2C
N
N
H isoreserpine
MeO 2C
140
MeO 2C
178B
The plan revolves around the notion that the indole should kinetically add to an iminium ion such that carbon-carbon bond and lone-pair are developed with an anti-periplanar relationship. This can occur in two ways. If the C-ring is born as a chair (with the requisite anti-periplanar relationship), the resulting product is reserpine. If the C-ring is born as a boat, the result is isoreserpine. Although isoreserpine is more stable than reserpine, the presumed conformations in which they would be born would f avor reserpine over isoreserpine. This assumes that product partitioning is controlled by kinetics (is not reversible). The "key intermediate" also deals nicely with regiochemical issues. Three syntheses of the key intermediate were described. We will examine only one of these.
Alkaloids-23
Page 324
reserpine OMe
10:56 AM
C-ring
Organic Synthesis via Examination of Selected Products
H
12/21/2010
OMe N
b1026
OMe
Organic Synthesis via Examination of Selected Natural Products
Stork, G.; Tang, P. C.; Casey, M.; Goodman, B.; Toyota, M. "Regiospecific and Stereoselective Syntheses of dl-Reserpine and (-)-Reserpine" J. Am. Chem. Soc. 2005, 127, 16255-16262. RO RO
N
b1026_Chapter-08.qxd
324
Regioselective and Stereoselective Route
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 325
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
325
Alkaloids-23 The final synthesis of reserpine (79) that we will examine was accomplished by the Stork group (Columbia). The plan revolves around the notion that the indole should kinetically add to an iminium ion such that the carbon-carbon bond and lone-pair are developed with an anti-periplanar relationship. This can occur in two ways. If the C-ring is born as a chair (with the requisite antiperiplanar relationship), the resulting product is reserpine (178C to 79). If the C-ring is born as a boat, the result is isoreserpine (178B to 140). Although isoreserpine is more stable than reserpine, the presumed conformations in which they would be born would favor reserpine over isoreserpine. This assumes that product partitioning is controlled by kinetics (is not reversible). Iminium ion 178 was to be generated from 179 via Schiff base formation followed by N-alkylation. Key intermediate 179 also deals nicely with regiochemical issues. Three syntheses of the 179 were described. We will examine only one of these.
O
MeO 2C
OTs
O
OHC
OTs
MeO2 C X
186
185
184
O
O
2. ClCH2 OBn
1. LiAlH4
OBn
2. H3 O+
EtO
OBn
1. LDA 2. acrylate, 0 °C
O
88% SiMe 2Ar 190
189
188
187
O
MeO 2C OBn
Need to introduce tosylate early for timing reasons (avoid γ-lactone f ormation from incipient δ -lactone).
n-Bu4 N F (TBAF)
O MeO 2C Ag2 O
O
MeO 2C
OTs
O
m-CPBA
MeI
Na2 HPO4
OH 193
60%
O
MeO 2C
OTs 1. H2 , Pd/C, EtOAc 2. TsCl, pyridine
SiMe 2F 60%
SiMe 2F 192
Tamao-Fleming Oxidation is mechanistically similar to the Baeyer-Villiger Oxidation
Alkaloids-24
191
88%
OBn
Page 326
1. LDA EtO
CO2 Me Si
O
10:56 AM
2. ClSiMe 2H
methyl acrylate
Organic Synthesis via Examination of Selected Products
SiMe2H
O
One could imagine the key reaction taking place by either Michael-Michael or Diels-Alder mechanisms. Both would be expected to occur from the face of the diene opposite the tosylate. Endo selectivity would be required for "stereochemical success". The choice of "X" is important. A methoxy group would complicate the Michael-Michael process by β-elimination, and the Diels-Alder process by reducing dienophilicity through electron donation.
b1026
Co2 (CO)8
1. n-BuLi O
183
12/21/2010
181
180
179
LiO
182
Organic Synthesis via Examination of Selected Natural Products
OMe
OMe
OTs
+
OH OMe
b1026_Chapter-08.qxd
MeO 2C
MeO2 C
326
O
OTs
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 327
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
327
Alkaloids-24 The approach to 179 that we will examine proceeded through lactone 180. This intermediate is a lactone reduction away from 179. The lactone was to be generated by Baeyer-Villiger oxidation of bicyclic ketone 181 which was to come, in turn, from a formal “cycloaddition” between dienolate 183 and β-hydroxyacrylate equivalent 182 (X=OH). One could imagine the key cycloaddition taking place by either Michael-Michael or Diels-Alder mechanisms. Both would be expected to occur from the face of the diene opposite the tosylate (or an appropriate tosylate precursor). Endo selectivity would be required for “stereochemical reasons”. The choice of “X” in 182 was important. A methoxy group would complicate the Michael-Michael process by β-elimination, and the Diels-Alder process by reducing the dienophilicity through electron donation. Furan derivative 186 was ultimately selected as the “dienophile”. The plan was to convert the C–Si bond to a C–O bond using a Tamao-Fleming oxidation.16 The dienolate derived from kinetic deprotonation of cyclohexenone 189 was ultimately selected as the “diene”. Cyclohexenone 189 was prepared using the classical “Stork-Danheiser synthesis” of 4-substituted cyclohexenones.17 The reaction between the enolate of 189 and 186 was very efficient, giving 190 in excellent yield. The 2-furyl group was replaced by fluoride using tetra-n-butylammonium fluoride (190 → 191). The benzyl ether was converted to the corresponding tosylate (192). Baeyer-Villiger oxidation of 192 was accompanied by Tamao-Fleming oxidation of the C–Si bond to provide 193. The timing of the conversion of the benzyl ether to the tosylate is notable. Had this been left for later in the synthesis, translactonization might have been a serious problem at the stage of intermediates such as 193, where the OTs group is replaced by an OH group.
b1026_Chapter-08.qxd
328
O MeO 2C
OTs
O
H
OTs i-Bu2 AlH
6-methoxytryptamine
OHC
MeO2C
OH OMe
84%
6-methoxytryptamine KCN, CH 3CN
RO
MeO
MeO 2C
CN N H What is the problem? Perhaps tight ion pairing (between iminium ion and cyanide) prevent the "plan" f rom operating and encourage SN2 or boat-line cyclization transition states to dominate. Possible solution to the problem? Use solvents (conditions) that will preclude ion pairing ... protonate the cyanide after it "leaves".
H 10% 1N HCl in THF
H
3
N H H
MeO2C
90%
OH OMe
CH3 CN, ∆
esterification
H MeO
N H
197
87%
196
MeO
N H
3
65% 195 MeO2C
Reserpine
N H H OH OMe
Alkaloids-25
Preparation of a single enantiomer of the 2benzyloxymethylcyclohex-2-ene-1-one provided an enantioselective route to the natural product.
Page 328
OMe N
10:56 AM
What is the problem? (1) imine formation (2) Pictet-Spengler cyclization (3) amine alkylation. This sequece of events would generate C 3 stereochemistry during the Pictet-Spengler and thus, the "plan" is not relevant to the stereochemical outcome. Possible solution to the problem? Intercept imine and change the sequence of events (2) and (3).
Organic Synthesis via Examination of Selected Products
Sequence of events? (1) imine formation (2) trap with cyanide (3) amine alkylation. α-Cyano amine formation reversible?
179
isoreserpine stereochemistry
b1026
194
N H
12/21/2010
195
OMe 100%
3
H
OH
OMe
N H
Organic Synthesis via Examination of Selected Natural Products
MeO2 C
MeO
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 329
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
329
Alkaloids-25 Etherification of 193 gave 194 and i-Bu2AlH reduction of the lactone provided key intermediate 179. Treatment of 179 with 6-methoxytryptamine provided 195 with “isoreserpo” stereochemistry at C3! What was the problem? It was suggested that the sequence of events leading to 195 might be: (1) imine formation, (2) Pictet-Spengler cyclization and (3) N-alkylation of the resulting secondary amine. This sequence would generate C3 stereochemistry during the Pictet-Spengler reaction, and thus the “plan” would not be relevant to the stereochemical outcome. What is a possible solution to the problem? Intercept the imine and change the sequence of events (2) and (3). It was projected that cyanide might be an appropriate imine trap. Thus, when 179 was treated with 6-methoxytryptamine in the presence of KCN, α-cyanoamine 196 was obtained in excellent yield. Note that the stereochemistry of the α-cyanoamine is that predicted by “stereoelectonic considerations”. It was known that α-cyanoimines could serve as precursors to iminium ions. Thus, warming 196 in a good ionizing solvent gave a cyclization product, but once again the product was 195 with the undesired C3 stereochemistry. What was the problem? It was surmised that tight ion pairing between the iminium ion and cyanide might be preventing the “plan” from operating, and might be encouraging SN2 or boat-like cyclization transition states to dominate. If this was the problem, it was hoped that conditions that “break up” ion pairs might give the desired stereochemical result. The tactic that was adopted was to protonate the cyanide after it “left” the α-carbon. The ploy worked and 197 was obtained as the major product in excellent yield. Esterification of the C19 alcohol completed the synthesis. This synthesis was also accomplished in an enantioselective manner when a single enantiomer of 189 was prepared and moved through the synthesis. This synthesis provides a marvelous example in which mechanistic reasoning was used to “rationally” identify tactics for accomplishing a desired result. Other syntheses of reserpine have been described, but I hope this sampling has been enjoyable. We will now move to alkaloid syntheses where the focus on iminium ion chemistry is somewhat reduced, although it will not disappear completely.
b1026_Chapter-08.qxd
12/21/2010 b1026
330
10:56 AM
Page 330
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
References 1. For some recent reviews see: Maryanoff, B. E.; Zhang, H-C.; Cohen, J. H.; Turchi, I. J.; Maryanoff, C. A. “Cyclizations of N-Acyliminium Ions” Chem. Rev. 2004, 104, 1431–1628. Hajicek, J. “Mannich and Related Reactions in Total Synthesis of Alkaloids” Chem. Listy 2004, 98, 1096–1112. Arend, M.; Westermann, B.; Risch, N. “Modern Variants of the Mannich Reaction” Angew. Chem. Int. Ed. 1998, 37, 1045–1070. 2. Stevens, R. V.; Lee, A. W. M. “Stereochemistry of the Robinson-Schopf Reaction. A Stereospecific Total Synthesis of the Ladybug Defense Alkaloids Precoccinelline and Coccinelline” J. Am. Chem. Soc. 1979, 101, 7032–7035. Stevens coccinelline 3. For more examples see: Delongschamps, P. “Stereoelectronic Effects in Organic Chemistry” Pergamon Press, 1983. 4. Shapiro, R. H. “Alkenes from Tosylhydrazones” Organic Reactions 1976, 23, 405–507. Chamberlin, A. R.; Stemke, J. E.; Bond, F. T. “Vinyllithium Reagents from Arenesulfonylhydrazones” J. Org. Chem. 1978, 43, 147–154. 5. Shono, T.; Matsumura, Y.; Tsubata, K. “Anodic Oxidation of N-Carbomethoxypyrrolidine: 2-Methoxy-N-carbomethoxypyrrolidine” Organic Syntheses 1985, 63, 206–213. 6. For an excellent introduction to stereoelectronic effects including the anomeric effect see: Deslongschamps, P. “Stereoelectronic Effects in Organic Chemistry” Pergamon Press, 1983 (375 pages). 7. Leonard, N. J.; Hay, A. S.; Fulmer, R. W.; Gash, V. W. “Unsaturated Amines. III. Introduction of α,β-Unsaturation by Means of Mercuric Acetate: ∆1(9a)Dehydroquinolizidine” J. Am. Chem. Soc. 1955, 77, 439–444. 8. Plattner, Pl. A.; Fürst, A. “Ueber Steroide und Sexualhormone: uber eine Ergiebige Method zur Herstellung des Epi-cholesterins und uber das 3α,5Dioxy-cholestan” Helv. Chem. Acta 1948, 31, 1455–1463. Fürst, A.; Plattner, Pl. A. “Ueber Steroide und Sexualhormone: 2α,3α- und 2β,3β-Oxidocholestane; Konfiguration der 2-Oxy-cholestane” Helv. Chem. Acta 1949, 32, 275–283. 9. Seeman, J. I. “The Curtin-Hammett Principle and the Winstein-Holness Equation. New Definition and Recent Extensions to Classical Concepts” J. Chem. Educ. 1986, 63, 42–48. Winstein, S.; Holness, N. J. “Neighboring Carbon and Hydrogen. XIX. tert-Butylcyclohexyl Derivatives. Quantitative Conformational Analysis” J. Am. Chem. Soc. 1955, 77, 5562–5578. Curtin, D. Y. “Stereochemical Control of Organic Reactions. Differences in Behavior of Diastereoisomers. I. Ethane Derivatives. The Cis Effect” Rec. Chem. Progr. 1954, 15, 111–128. An alternative to the Curtin-Hammett argument presented on page 295 is that the epoxide reacts with thiophenoxide from conformation 64b such that the product is born in the B-ring boat (or twist-boat) conformation of 65b.
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 331
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
331
10. Borch, R. F.; Bernstein, M. D.; Durst, H. D. “Cyanohydridoborate Anion as a Selective Reducing Agent” J. Am. Chem. Soc. 1971, 93, 2897–2904. 11. Bischler, A.; Napieralski, B. “Zur Kenntniss einer Neuen Isochinolinsynthese” Chem. Ber. 1893, 26, 1903–1908. Whaley, W. M.; Govindachari, “The Preparation of 3,4-Dihydroisoquinolines and Related Compounds by the Bischler-Napieralski Reaction” Organic Reactions 1951, 6, 74–150. 12. Pictet, A.; Spengler, T. “Über die Bildung von Isochinolin-Derivaten durch Enwirkung von Methylal auf Phenyl-üthylamin, Phenyl-alanin und Tyrosin” Chem. Ber. 1911, 44, 2030–2036. Whaley, W. M.; Govindachari, “The Preparation of 3,4-Dihydroisoquinolines and Related Compounds by the Bischler-Napieralski Reaction” Organic Reactions 1951, 6, 151–206. 13. Challand, B. D.; Hikino, H.; Kornis, G.; Lange, G.; De Mayo, P. “Photochemical Synthesis. XXV. Photochemical Cycloaddition. Some Applications of the use of Enolized β-Diketones” J. Org. Chem. 1969, 34, 794–806. 14. Krow, G. R. “The Baeyer-Villiger Oxidation of Ketones and Aldehydes” Organic Reactions 1993, 43, 251–798. Baeyer, A.; Villiger, V. “Ueber die Einwirkung des Caro’schen Reagens auf Ketone.” Chem. Ber. 1900, 33, 858–864. 15. Finkelstein, H. “Darstellung Organischer Jodid aus den Entsprechenden Bromiden und Chloriden” Chem. Ber. 1910, 43, 1528–1532. 16. Fleming, I. “Silyl-to-Hydroxy Conversion in Organic Synthesis” Chemtracts: Organic Chemistry 1996, 9, 1–64. Tamao, K.; Ishida, N.; Kumada, M. “(Diisopropoxymethylsilyl)methyl Grignard Reagent: A New, Practically Useful Nucleophilic Hydroxymethylating Agent” J. Org. Chem. 1983, 48, 2120–2122. 17. Stork, G.; Danheiser, R. L. “Regiospecific Alkylation of Cyclic β-Diketone Enol Ethers. General Synthesis of 4-Alkylcyclohexenones” J. Org. Chem. 1973, 38, 1775–1776.
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 332
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
332
Problems 1. What is the “other ketone” that could be converted to porantherine (1) in the same manner as ketone 2? (Alkaloids-1) 2. Outline a synthesis of 9 that passes through only “odd” difunctional intermediates. (Alkaloids-1) 3. Suggest mechanisms for the cyclization of 12→13 and for the oxidation of 13 → 14. (Alkaloids-2) 4. Suggest alternative tactics for the demethylation of 13. Do your alternative tactics call for a change in the sequence of steps leading from 13 → 15? (Alkaloids-2) 5. Suggest mechanisms for the oxidation of 31 → 32 and for the conversion of 2 → 1. (Alkaloids-4) 6. Suggest alternative tactics for the oxidation of 31 → 32. (Alkaloids-2) 7. Apply the stereoelectronic principles suggested by Stevens (Alkaloids-3) to the conversion of 34 → 38. Show how this notion rationalizes the stereochemical course at the newly-formed stereogenic center in the piperidine ring of 38. (Alkaloids-5) 8. Explain why oxidation of the nitrogen was not problematic in the last step of this synthesis. (Alkaloids-9) 9. Predict the product expected from the following reactions (BischlerNapieralski, Pictet-Spengler and related reactions). (Alkaloids-10) H N
Ph polyphosphoric acid (PPA) ∆
O
MeO
NH2
MeO
3,4-diethoxybenzaldehyde 24% aq. HCl
O MeO BnO
N
HCOOH Ph
Whaley, W. M.; Hartung, W. H. “Synthesis of Isoquinoline Derivatives” J. Org. Chem. 1949, 14, 650–654. Weinbach, E. C.; Hartung, W. H.
b1026_Chapter-08.qxd
12/21/2010 b1026
10:56 AM
Page 333
Organic Synthesis via Examination of Selected Products
Alkaloids — Difunctional Relationships and the Importance
333
“Synthesis of Tetrahydroisoquinoline Derivatives” J. Org. Chem. 1950, 15, 676–679. Burnett, D. A.; Hart, D. J. “Conformational Effects on the Oxidative Phenolic Coupling of Benzyltetrahydroisoquinolines to Morphinan and Aporphine Alkaloids” J. Org. Chem. 1987, 52, 5662–5667. 10. Draw the two chair-chair conformations for each cis-decalin shown below and indicate which conformation will be more stable. (Alkaloids-11) Me
H Me
Me
Me
Me
H Me
H Me
Me
Me
H
Me
H
Me
Me
H
Me Me
H Me Me
H
11. Rationalize the stereochemical course of the addition of methoxide to unsaturated ester 93 using stereoelectronic arguments. (Alkaloids-11) 12. Provide a mechanism for the conversion of 99 → 100. (Alkaloids-12) 13. Outline a synthesis of 6-methoxytryptamine. (Alkaloids-12) 14. Provide mechanistic details for the conversion of 107 → 108 by each of the mechanisms suggested in Alkaloids-13. (Alkaloids-13) 15. Provide a mechanism for the conversion of 112 → 113. (Alkaloids-14) 16. Can you provide a mechanistic rationale for the stereochemical course of the conversion of 113 → 114 + 115? for the conversion of 110 → 117 + 118? (Alkaloids-14) 17. Provide a mechanistic rationale for the stereochemical course of the protonation in the conversion of 130 → 131. Explain why thermodynamics favor 131 over its C17 diastereomer. (Alkaloids-16) 18. A Moffat oxidation was used to convert 137 → 138 (R=CHO). A minor product in this reaction was 138 (R=CH2OCH2SMe). Provide a mechanistic explanation for formation of this product. (Alkaloids-17) 19. Provide structures of intermediates in the conversion of 138 → 139. (Alkaloids-17) 20. Propose a mechanism for the conversion of 139 → 140. (Alkaloids-17)
b1026_Chapter-08.qxd
12/21/2010 b1026
334
10:56 AM
Page 334
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
21. Propose a reaction sequence that might convert 142 (were it available) to the following compound. (Alkaloids-18) Bn N
O
HO
OH OMe
22. 2-Ethylhexanoic acid is a commodity chemical (very inexpensive). Propose a synthesis of this acid that you think might be amenable to synthesis on a ton-scale. (Alkaloids-19) 23. Note that the sequence going from racemic 150 → 157 necessarily will involve diastereomeric mixtures if racemic 2-ethylhexanoic acid is used in the conversion of 150 → 151. Explain. Note that this is also true even if one enantiomer of 2-ethylhexanoic acid is used in this step. Explain. (Alkaloids-19) 24. Provide the products along the way from 153 → 154. (Alkaloids-19) 25. Lactones have some interesting properties relative to their their acyclic cousins (esters). Why is the lactone reduced faster than the esters in the reduction of 162? Why is ethyl acetate water insoluble, but γ-butyrolactone is completely water miscible? (Alkaloids-21) 26. Suggest a mechanism (or several mechanisms) for the conversion of 173 → 174 + 175. (Alkaloids-22) 27. Suggest a mechanism for the Tamao-Fleming portion of the conversion of 192 → 193. (Alkaloids-24) 28. Elaborate on the use of “stereoelectronic considerations” to explain the conversion of 179 → 196. Discuss the possibility that the gauche effect also plays a role in the stereochemical course of this transformation. (Alkaloids-25)
b1026_Chapter-09.qxd
336
Alkaloids from "Dart-Poison" Frogs From HART/CRAINE/HART/HADAD, Organic Chemistry, 12E Copyright Brooks/Cole, a part of Cengage Learning, Inc. Reproduced by permission. www.cengage.com/permissions.
H
R
S
H N H H H
S
H
O N
HO Me O HO
H
N O
O
HO
3
2
H
4
Gephyrotoxin
Pumiliotoxin-C
Histrionicotoxin
N
1 Batrachotoxin S
H
N R
H S
N
O
H
H
H H
H N
S
H
Me H
Histrionicotoxin-1
H HO
H
Page 336
O
Me H
S
10:56 AM
H N
12/21/2010
H
Organic Synthesis via Examination of Selected Products
Because many of these compounds are produced in only microgram quantites by the frogs, laboratory syntheses have often been developed to confirm structures and also to provide a supply of material for pharmacological studies. It turns out that most of these toxins act on the nervous system by affecting the manner in which ions are transported across cell membranes. As a result of this property, several of these alkaloids are now used as research tools in the field of neuroscience. We will spend this chapter examining selected syntheses of the aforementioned toxins (with the exception of batrachotoxin). Let us begin with histrionicotoxin.
b1026
Scientists from the National Institutes of Health have isolated and determined the structures of many of the compounds present in these skin secretions using the techniques of mass spectrometry and NMR spectroscopy. So far the structures of well over 200 different alkaloids have been determined. The most potent of these toxins is batrachotoxin (from Phyllobates terribilis), an example of a steroidal alkaloid. Other toxins include histrionicotoxin (from Dendrobates histrionicus), pumiliotoxin C (from Dendrobates pumilio), and gephyrotoxin.
Organic Synthesis via Examination of Selected Natural Products
One prolific source producer (or maybe not ... read on) of alkaloids are dart-poison frogs, a family of colorful and petite amphibians native to Costa Rica, Panama, Ecuador, and Colombia. The intense coloration of these frogs serves as a warning to potential predators who would have an unpleasant experience if they tried to make a meal of one of these tiny creatures. This is because the frogs secrete toxic alkaloids from glands on the surface of their skin. These secretions are so toxic that they have been used by locals to poison blowgun darts used in hunting, hence the name dart-poison frogs. It has recently been shown that, in many cases, the frogs actually injest the alkaloids from a dietary source (ants) before secreting them as a defense substance. It has also been shown that frogs raised in captivity do not secrete these alkaloids.
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 337
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
337
Histrionicotoxin-1 One prolific producer (or maybe not ... read on) of alkaloids are dart-poison frogs, a family of colorful and petite amphibians native to Costa Rica, Panama, Ecuador, and Colombia. The intense coloration of these frogs serves as a warning to potential predators who would have an unpleasant experience if they tried to make a meal of one of these tiny creatures. This is because the frogs secrete toxic alkaloids from glands on the surface of their skin. These secretions are so toxic that they have been used by locals to poison blowgun darts used in hunting, hence the name dart-poison frogs. It has recently been shown that, in many cases, the frogs actually injest the alkaloids from a dietary source (ants) before secreting them as a defense substance. It has also been shown that frogs raised in captivity do not secrete these alkaloids. Scientists from the National Institutes of Health (John Daly and coworkers) have isolated and determined the structures of many of the compounds present in these skin secretions using the techniques of mass spectrometry and NMR spectroscopy. So far the structures of well over 200 different alkaloids have been determined.1 The most potent of these toxins is batrachotoxin (1) (from Phyllobates terribilis), an example of a steroidal alkaloid. Other toxins include histrionicotoxin (2) (from Dendrobates histrionicus), pumiliotoxin C (3) (from Dendrobates pumilio), and gephyrotoxin (4). Because many of these compounds are produced in only microgram quantities by the frogs, laboratory syntheses have often been developed to help assign structures and also to provide a supply of material for pharmacological studies. It turns out that most of these toxins act on the nervous system by affecting the manner in which ions are transported across cell membranes. Due to this property, several of these alkaloids are now used as research tools in the field of neuroscience. In this chapter we will examine selected syntheses of the aforementioned toxins (with the exception of batrachotoxin). We will begin with histrionicotoxin (2).
O
H
R
R
S S
2
H
N
S
H
Perhydrohistrionicotoxin
O
O
H N
S
S
2
5
Corey Approach O
C5 H
O
H N
C=N
N
OH
reduce
OH N Beckman
OH
Barton
hydroborationOH
oxidize
addition
oxidation
C4
C4 6
OH
HN
C4
11
8
7
thermodynamic and stereochemical issues
C4
12
Kishi Approach
C4
C4
13 addition-dehydration
O NH2
O C4
OH OH
C4 OR 10
O 9
Histrionicotoxin-2
15
pinacol
O 14
Page 338
5
10:56 AM
S
S
S
H N
12/21/2010
R
H
Histrionicotoxin
O
Organic Synthesis via Examination of Selected Products
H
N
b1026
S
Organic Synthesis via Examination of Selected Natural Products
A number of histrionicotoxins have been characterized. These differ largely in the position of unsaturation on the two "side chains". Whereas the parent natural product eluded synthesis f or many years, the saturated version of this compound, perhydrohistrionicotoxin, was f irst prepared by Kishi [Aratani, M.; Dunkerton, L. V.; Fukuyama, T.; Kishi, Y.; Kakoi, H.; Sugiura, S.; Inoue, S. "Synthetic studies on histrionicotoxins. I. A stereocontrolled synthesis of dl-perhydrohistrionicotoxin" J . Or g. Chem. 1975, 40, 2009-2011; Fukuyama, T.; Dunderton, L. V.; Aratani, M.; Kishi, Y. "Synthetic Studies on Histrionicotoxins. II. A Practical Synthetic Route to dl-Perhydro- and dl-Octahydrohistrionicotoxin" J. Or g. Chem. 1975, 40, 2011-2012] and Corey [Corey, E. J.; Arnett, J. F.; Widger, G. N. "A simple total synthesis of dl-perhydrohistrionicotoxin" J. Am. Chem. Soc. 1975, 97, 430-431] and indeed, by many other groups over the years. We will simply look at the Corey and Kishi approaches retrosynthetically and will only examine two syntheses of perhydrohistrionicotoxin in detail. The f irst will introduce the use of Nacyliminium ions as electrophiles and the second will revisit N-acylnitroso compounds as reactive species f or use in alkaloid synthesis.
H
b1026_Chapter-09.qxd
338
Histrionicotoxin
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 339
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
339
Histrionicotoxin-2 A number of histrionicotoxins have been characterized. These differ largely in the position of unsaturation on the two “side chains”. Whereas the parent natural product eluded synthesis for many years, the saturated version of this compound, perhydrohistrionicotoxin (5), was first prepared by Kishi and Corey, and indeed, by many other groups thereafter. We will simply look at the Corey and Kishi approaches retrosynthetically and will then examine two syntheses of perhydrohistrionicotoxin in detail. The first synthesis will introduce the use of N-acyliminium ions as electrophiles, and the second will revisit N-acylnitroso compounds as reactive species for use in alkaloid synthesis. Both the Kishi and Corey approaches to 5 proceeded through imine 7 and lactam 8, the pentyl sidechain being introduced by addition of an organometallic reagent to the imine. Kishi approached 8 using an intramolecular conjugate addition strategy (9 → 8) This approach was plagued by thermodynamic and stereochemical issues, and several “corrections” were needed along the way. Cyclohexenone 9 was prepared from 10. It is notable that this strategy passes through intermediates with 1,3- and 1,5-difunctional relationships, and relies largely on normal chemistry of the carbonyl group. Corey approached 8 via a Beckman rearrangement of oxime 11, which came, in turn, from 12 via a Barton reaction.2 Alcohol 12 was prepared by hydroboration-oxidation of 13. The sequence from 13 → 12 → 11 → 8 nicely addresses the relative stereochemistry at the three contiguous stereogenic centers in perhydrohistrionicotoxin (5). Alkene 13 was prepared from 15 using a pinacol rearrangement to establish the spiro[4.5]decane ring system.
b1026_Chapter-09.qxd
340
N-Acyliminium Ion Cyclizations Some reactive intermediates
N
R
N-acyliminium ion
iminium ion
OMe
H
H
Me N
O
EAS
H N H
O
Me
O
N H
H
Me 17
16
Me
N O
17
OMe
OMe
18
N-Acyliminium Ions as Olefin Cyclization Substrates
OEt
O
HCO 2H
NaBH4 N
EtOH,
H+
N
O 19
OHCO
OHCO
O 20
Other Reducing Agents: NaBH4 /MeOH or i-Bu2 AlH/PhMe
N
N
N O 21
22
O
23
O
Initial work from group of Nico Speckamp. Looks like "anti addition" of electrophile and nucleophile across double bond. Work of Johnson served as inspiration for initial studies. Notice the construction of a 1,3-N,O relationship.
Histrionicotoxin-3
Page 340
OMe
H
10:56 AM
Early Results from Stork Lycopodine Synthesis
Organic Synthesis via Examination of Selected Products
oxocarbenium ion
N
12/21/2010
O
R
R
b1026
R
Organic Synthesis via Examination of Selected Natural Products
O R
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 341
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
341
Histrionicotoxin-3 We will examine two very different approaches to perhydrohistrionicotoxin. The first one revolves around the chemistry of N-acyliminium ions, a field largely developed by the Speckamp group in the Netherlands.3,4 In Chapter 8 we saw that iminium ions (the nitrogen analogs of oxocarbenium ions) are excellent electrophiles (Mannich reaction). N-Acyliminium ions are even more electrophilic than iminium ions. They react with both heteroatom and carbon nucleophiles as expected. An early use of an N-acyliminium ion in synthesis is the cyclization of enamide 16 to 18 in an electrophilic aromatic substitution reaction. This reaction was an early step in Stork’s approach to lycopodine.5 The reaction presumably proceeds through N-acyliminium ion 17, generated by protonation of the enamide. N-Acyliminium ions are “hot” and will even add to unactivated olefins. For example, N,O-acetal 20 reacts with formic acid to provide 23. The stereochemistry of the process is relatively clean and can be rationalized by an anti-periplanar addition of electrophilic carbon and nucleophilic oxygen across the carbon-carbon double bond with a transition state that resembles a chair N-acylpiperidine. In reality, this process is mechanistically more complex than this simple model (for example, possible carbocation intermediates and in some cases, some underlying sigmatropic rearrangements), but the model is simple, easy to remember, and has excellent predictive value. How was this chemistry used to prepare perhydrohistrionicotoxin?
H N
HO
C4
5e
24
CH2=CHMgBr
25
CH3C(OEt)3
CHO OH
26
Br
29
30
1. Mg, THF (2.2 eq RBr) Kishi Endgame
Cl O
Cl Mg N
SMe
SMe 1. MgCl2, CH2Cl2
OH
N
2. glutarimide 3. Remove solvent 4. HCO2H, 44 oC, 8 days
S Me3O BF4
OH
H N
1. P2S5 2. aq. NaOH MeOH
2. CH3(CH2)4MgCl C4
33
C4
88%
C4
32
C5
OH
35
N
C4
AlH3, toluene
OH
5
HN
6:1
perhydrohistrionicotoxin C4
Histrionicotoxin-4
70%
C4 31 30%
34 C5
H N
OHCO
O
Page 342
2. MsCl, Et3N 3. LiBr
OEt
∆
28
27
C4
O
1. LiAlH4
O
CH3CH2CO2H
H N
10:56 AM
C4
O
O
Organic Synthesis via Examination of Selected Products
C4 5a
MgBr
H N
C5
12/21/2010
OH
b1026
C5
Organic Synthesis via Examination of Selected Natural Products
Schoemaker, H. E.; Speckamp, W. N. "Stereocontrolled Synthesis of Functionalized 1-Azaspirans. Efficient Synthesis of Perhydrohistrionicotoxin" Tetrahedron 1980, 36, 951-958. For a related approach see Evans, D. A.; Thomas, E. W. "A formal Synthesis of dl-Perhydrohistrionicotoxin via αAcylimonium Ion-Olefin Cyclizations" Tetrahedron Lett. 1979, 411-414. H N
b1026_Chapter-09.qxd
342
Synthesis of Perhydrohistrionicotoxin
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 343
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
343
Histrionicotoxin-4 The first thing to recognize is that the N-acyliminium ion cyclization we just examined generates a 1,3-N,O relationship, precisely the kind of relationship that appears in lactam 8 (see Histrionicotoxin-2). This raises the question, can 8 be prepared by an N-acyliminium ion cyclization? Two chair-chair conformations are available to perhydrohistrionicotoxin (5a and 5e). Examination of 5e suggested that an N-acyliminium ion of type 24 might undergo cyclization to provide lactam 8 (see Histrionicotoxin-2). The key issue here is the regiochemical course of addition to the olefin. Will cyclization occur to afford a 5-membered ring or a 6-membered ring? It was felt that 24 could be obtained from Grignard reagent 25 and glutarimide (26). The bromide precursor of 25 was prepared from pentanal (27) in a straightforward manner. Notice the use of a Claisen rearrangement to carry out an “enforced SN2′ reaction” in the preparation of γ,δ-unsaturated ester 29 (see Prostaglandins-24). Treatment of glutarimide with 2.2 equivalents of Grignard reagent 25 (one equivalent to deprotonate the imide), and treatment of the resulting mixture of materials with formic acid, gave 31 albeit in modest yield. The synthesis was completed using the endgame developed by Kishi. Conversion of 31 to thiolactam 32 was followed by transformation to thioimidate 33. Exposure of 33 to magnesium chloride (to presumably generate complex 34) followed by pentylmagnesium chloride gave 35. Alane reduction of 35 completed the synthesis of 5.
b1026_Chapter-09.qxd
344
Perhydrohistrionicotoxin via Acylnitroso Ene Reactions Keck, G. E.; Yates, J. B. "A Novel Synthesis of dl-Perhydrohistrionicotoxin" J. Am. Chem. Soc. 1982, 47, 3591-3593.
N
O
O toluene, ∆
N
H
HO
Ph3 P=CHCH 3 on derived RCHO gives only a 40% yield of alkene
1. OsO4 (cat) NaIO 4 (89%)
N
O
O
N
CH2 =CHCH 2SnBu3
OAc
3. Ac2 O
AIBN, PhH, 41
88%
Pd(OAc)2 (0.01 eq) O
O
O O
N
6% Na/Hg
OH
HN
i-PrOH, rt
EtOAc 43
40
Br
What a shame! Can you think of a way to possibly correct this?
AcOH
Pt2O, H2
N
O
71% overall
44
45
HN
Me2 SO-(COCl) 2 C4
94%
O
Et3N
4:1
46
O
C4
Apply Kishi Endgame
Histrionicotoxin-5
Page 344
O
2. CH 2=CHMgBr
N
100%
NBS, CH 2Cl2
O
O
39
10:56 AM
38
Organic Synthesis via Examination of Selected Products
37
42
N
40 min
2. R-I 36
O
O
N
12/21/2010
O
1. LDA, THF-HMPA -20 C, 12 h
b1026
O
Organic Synthesis via Examination of Selected Natural Products
O O
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 345
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
345
Histrionicotoxin-5 The next perhydrohistrionicotoxin synthesis we will examine was reported by the Keck group (Utah). This synthesis featured chemistry of N-acylnitroso compounds, just as did the Keck synthesis of pyrrolizidines (Chapter 4). Rather than approaching this in a “retrosynthetic manner”, suffice it to say that this work intersects with the Kishi synthesis at the point of keto-lactam 46. Now we will jump right into the synthesis. The key reaction was an intramolecular ene reaction of acylnitroso compound 38. Recall that acylnitroso compounds are very reactive and thus, the plan (as in the Keck synthesis of pyrrolizidines) was to generate this species by a retro-Diels-Alder reaction of 37, in turn prepared by alkylation of the enolate derived from 36. The ene reaction worked well to provide 39. An electrophile initiated cyclization of 39 provided bromide 40. The next task was introduction of the 4-carbon chain. The plan was to accomplish this using another reaction developed largely in the Keck laboratories, the allylation of free radicals using allyl tri-n-butylstannane.6 This reaction also proceeded in good yield, but the sole product was 41. The stereochemistry of the pendant chain was opposite to that required by the target. Addition of the fourth carbon took considerable effort, but was accomplished in good yield as shown, to ultimately provide 44. Reduction of the N—O bond gave 45. Swern oxidation of the alcohol followed by epimerization provided 46 (along with 20% of the diastereomeric ketone), and the synthesis was completed using the Kishi endgame. This synthesis featured some innovative chemistry, but suffered in terms of stereochemistry. It is a shame that the allylation did not proceed with the desired stereochemistry because this would have rendered the synthesis extremely efficient.
b1026_Chapter-09.qxd
346
First Synthesis of Racemic Histrionicotoxin Carey, S. C.; Aratani, M.; Kishi, Y. "A Total Synthesis of dl-Histrionicotoxin" T etr ahedr on Let t. 1985, 26, 5887-5890.
H 2N
SiMe 3
S
O
O 47
50
49
48
(see 9) Synthesis of the azaspirane follows the 1975 path to octahydrohistrionicotoxin.
Cl
1. (C 5H 10 )2 Cd
CO2 Me
2. KOtBu, Et2O
1. H2 NCOCH 2CO2 Me NaOMe
O
O 1. EtOH, H+ 2. CH 2=CHMgBr
O
51
3. HCl (neutralize) 4. dioxane, 100 oC
53
52
O
2. H2 O, NaOH
54
H 2N
O 45% overall
1. HC(OEt) 3, EtOH, CSA
Epimeric alcohol recycled by Jones oxidation-reduction
1. Br2 , MeOH, CH 2 Cl2 56 2. DBU-DMSO, 140
H N
O
oC
O
1. OsO4 -NaIO4 dioxane-H2 O, rt 2. NaOH, MeOH, rt
O OAc
3. Ac2 O, pyr, rt 55
4. 180 oC (11 mm Hg)
2. HOAc, H2 O (workup) 3. NaOMe, MeOH (equilibrate) O
HN
HN 1. Li, NH 3 (separate) 49
50%
2. Ac2 O
4:1
O 100%
Histrionicotoxin-6
48
Page 346
O
10:56 AM
2
Organic Synthesis via Examination of Selected Products
S
R
12/21/2010
O
H N
b1026
H
O
O
H N OAc
H N OR
Organic Synthesis via Examination of Selected Natural Products
O
O S
H N
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 347
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
347
Histrionicotoxin-6 Many of the dart-poison frog alkaloids carry conjugated eneyne groups. This is part of the challenge for laboratory syntheses of these alkaloids. Thus, histrionicotoxin (2) is more of a challenge than perhydrohistrionicotoxin (5). As a result, fewer syntheses have been reported. We will look at three very different approaches to 2, beginning with the Kishi synthesis of racemic 2. The beginning of this synthesis resembles the Kishi approach to 5. The plan was to roughly follow a path from 50 → 49 → 48. The 3-buten-1-yl group in 48 was to serve as a handle for introducing the 4-carbon eneyne, and the lactam was to be a handle for introduction of the 5-carbon enyne, not necessarily in that order. The preparation of 54 (same as 50 and similar to 9) passed through intermediates with only odd-difunctional relationships and used “normal” carbonyl chemistry (1,2-carbonyl addition, aldol-dehydration, conjugate addition, hydrolysis-decarboxylation). Conditions were also found to coax 54 to undergo the intramolecular conjugate addition to provide 48 (and its epimer α to the ketone). An unselective reduction of the ketone provided 49 after esterification of the intermediate alcohol. The strategy for controlling stereochemistry of the 4-carbon enyne was to establish the cis-olefin in a ring, and then open the ring. Thus, the butenyl sidechain was degraded to an aldehyde and then converted to cyclic enol ether 55. Haloetherification of the olefin, followed by dehydrohalogenation of the secondary bromide, gave unsaturated acetal 56 (see Histrionicotoxin-7). Note that this transformation is formally just a halogenation-dehydrohalogenation of an aldehyde.
H N
O
1. AcOH, H2O, 60°C
H N
AcO
1. P2S5, pyridine, 80 °C
O
56
AcO
35-40% overall
Eschenmoser Sulf ide Extrusion Reaction
57
CO2Et NaBH3 CN cyclohexane, rt
60 37% from allylic acetate
38% of each diastereomer 1. LiAlH4 2. Ac2 O, pyridine 3. NaOH, MeOH
AcO
3. MeLi, Me3 SiCl, THF
OHC 60%
59
trouble with azetidine formation
OH 1. TBAF, THF
Cl H2 N
AcO
OMs
Br 1. LiBr, DMF, 50 °C 2. Ph3P, CH3CN, 160 °C
2. MsCl, Et3N, CH 2Cl2
H N
AcO
3. HCl, Et2O, CH2 Cl2
Me3 Si 62
CO2Et O
H N
AcO
2. NaOEt, EtOH, 50 °C
Me3 Si
61
H N
CO2Et 1. Ph3 PCH 2Cl Cl n-BuLi
64 31%
It is interesting that sulfonamide formation does not occur ... allylic strain issues?
63
35% overall 1. LDA, THF, rt CHO 2. TBS 3. TBAF, THF
Histrionicotoxin-7
PPh 3
Page 348
1:1
Me3 Si
H N
AcO
10:56 AM
H N
Organic Synthesis via Examination of Selected Products
1. NaOH, H2O, MeOH, -20 °C 2. PCC, CH2 Cl2 , rt
Borch reduction usually run at pH 4
AcO
58
60%
b1026
AcO
Organic Synthesis via Examination of Selected Natural Products
2. ethyl 2-bromoacetoacetate, K2CO3, MeOH
2. NaBH4,THF,H2O 3. Ac2O,pyridine
OMe
CO2Et O
H N
AcO
12/21/2010
O
b1026_Chapter-09.qxd
348
When reduced, Grob f ragmentations complicate oxidation of C 7 side chain.
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 349
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
349
Histrionicotoxin-7 With the double bond in place, the ring was “opened” to provide 57. Completion of the 4-carbon eneyne would eventually require homologation of the 3-carbon sidechain to a terminal alkyne via carbonyl addition chemistry to an aldehyde. It is notable that the aldehyde was “stored” at the alcohol oxidation state, presumably to avoid problems of olefin isomerization that might have been encountered in the aldehyde. Next, work began on the 5-carbon eneyne. The plan for this portion of histrionicotoxin was to introduce a two-carbon unit (β-hydroxyethyl group) that could ultimately be reacted with an appropriate propargaldehyde (3carbon unit) via a Wittig reaction. The first steps were toward introduction of the 2-carbon unit. Lactam 57 was converted to corresponding thiolactam, and an Eschenmoser sulfide contraction was used to prepare 58.7 Returning to the 4-carbon eneyne, the primary acetate was hydrolyzed. The resulting alcohol was oxidized to 59. Wittig olefination of 59 followed by sequential removal of the acetyl group and conversion of the intermediate vinyl chloride to a TMS-protected terminal acetylene, provided 60 (Corey-Fuchs reaction).8 Installation of the rest of the 5-carbon eneyne came next. Sodium cyanoborohydride reduction of 60 gave a 1:1 mixture of diastereomeric β-aminoesters 61. Lithium aluminum hydride reduction of 61 gave a diol which was converted to 62 using an esterification-hydrolysis sequence. It is interesting that the esterification reaction does not acylate the piperdine nitrogen. Perhaps allylic strain in the N-acyl derivative (not observed) is responsible for this unusual selectivity.9 Removal of the TMS group and conversion of the primary alcohol to a mesylate gave 63 (as the amine hydrochloride). This amine salt was then converted to phosphonium salt 64 in the standard manner. Generation of the phosphorus ylid derived from 64, Wittig olefination using TMS-propargaldehyde, and removal of the TMS group, gave 65 with excellent selectivity for the Z-geometrical isomer. The acetate was then hydrolyzed to give histrionicotoxin (2) (see Histrionicotoxin-8).
b1026_Chapter-09.qxd
350
1% of trans 24:1
H N
NaOH, MeOH
H N
HO
OH
18% histrionicotoxin
OR
HN
Br
OTBS MeO
O
O
+ H H
I 66
2
67
O 69
68
Brown Allylation
B
OH
2
MeO
O
CHO
71
MeO
OTBS
O
TBSCl, imidazole
-78 °C to rt 2h 70
CH2 Cl2 , 4-DMAP 44% (86% ee) 72
MeO
1. LDA
O
O
73
Br 68
2. J. Am. Chem. Soc. 1990, 112, 1661
Histrionicotoxin-8
3. LDA
Page 350
OH
HN
Br
O
I
10:56 AM
Stork, G.; Zhao, K.; "Total Syntheses of (-)-Histrionicotoxin and (-)-Histrionicotoxin 235H" J . Am. Chem. Soc. 1990, 112, 5875.
Organic Synthesis via Examination of Selected Products
An Enantioselective Synthesis of Histrionicotoxin
b1026
2
Organic Synthesis via Examination of Selected Natural Products
2
65
HN
12/21/2010
AcO
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 351
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
351
Histrionicotoxin-8 We will now examine an enantioselective route developed by Stork and Zhao. The plan was to use a “double Sonogashira” coupling to move from bis-vinyl iodide 66 to histrionicotoxin (2).10 The bis-vinyl iodide was to be installed by a “double Wittig” reaction of a dialdehyde. The two olefins in 67 were to serve as “stable” precursors of this dialdehyde. The piperidine ring of 66 was to be installed by an intramolecular SN2 reaction of a primary amine, wherein the bromide in 67 was to serve as the alkylating agent, and the lactone carbonyl group was to serve as a precursor to the amine. Lactone 67 was to be assembled by a “double alkylation” of ester 69 with bis-electrophile 68. The hope was that the allylic position of vinyloxirane would be more reactive than the primary bromide, thus dictating the sequence of events associated with the double alkylation. This strategy offers an opportunity for enantioselectivity and, in fact, demands that both 68 and 69 be prepared as single enantiomers to avoid problems with diastereomer formation in the double alkylation. Also note that the regioselectivity projected for the oxirane opening is related to selectivity we saw in the opening of a geminally activated vinylcyclopropane in the Abraham approach to prostaglandins (Prostaglandins-27). So this part of the plan was not so far-fetched. Aldehyde 73 was prepared from aldehyde 70 using a “Brown Allylation” to control absolute stereochemistry in the preparation of 72.11 Bromide 68 was prepared using a Sharpless epoxidation to control absolute stereochemistry.12 Conversion of 73 to the corresponding enolate, alkylation with 68, and addition of more LDA to generate a new enolate (74) gave a reasonable yield of 75 (see Histrionicotoxin-8/9).
OTBS
O
O
O 1. O3
O
OTBS CHO
78
79
TMS acetylene 31% 80 (see 66)
HN
1. TBAF, THF TMS TMS
OH
2. K2CO3 , MeOH 40%
81
HN
2
Histrionicotoxin-9
I
PhH
Sonogashira Coupling
Page 352
I
10:56 AM
I
3. PhI(OCOCF3 )2 CH 3CN, H 2O 4. Et3 N, ClCH2 CH2 Cl 65-70 °C
Pd(Ph3P) 4, CuI
12/21/2010
2. Ph3 P, CBr 4 (53%)
I
HN
Organic Synthesis via Examination of Selected Products
I
I
OAc
b1026
O
1. Me3 Al, NH 4 Cl 40 °C, ∆ 2. Ac 2O
Organic Synthesis via Examination of Selected Natural Products
1. 5% HCl-THF
O
THF
43%
76
Br
O
HMPA CHO
75
OTBS
O
77
NaN(TMS)2
O
43% 74
OAc
Ph3 PCH 2I I
2. Ph3 P
Br
b1026_Chapter-09.qxd
OTBS
352
O
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 353
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
353
Histrionicotoxin-9 The design of the double alkylation is interesting. It was projected that the first alkylation would afford a lactone (enolate). The stereochemistry of the three stereogenic centers in 74 is derived from the stereochemistry of the starting materials. The stereochemistry of the reaction leading from 74 to 75 is dictated by the lactone stereogenic center carrying the 3-bromopropyl group. If lactonization preceeded the second alkylation, the stereochemical course of this transformation was guaranteed. Continuing with the synthesis, ozonolysis of the bis-olefin gave dialdehyde 76. Bis-Wittig olefination of 76 using the ylid derived from phosphonium salt 77 provided 78 with good control of Z-olefin geometry, a reaction developed largely in the Stork laboratories. The TBS protecting group was removed and the resulting alcohol was converted (with inversion of configuration) to bromide 79 (same as 67). The lactone was opened to provide a primary amide. The secondary alcohol was protected as an acetate, a Curtius-type rearrangement was used to introduce the primary amine, and an intramolecular N-alkylation gave piperidine 80 (see projected intermediate 66). The synthesis was completed by introducing the two cis-eneynes (Sonogoshira coupling), and removing the alcohol and acetylene protecting groups.
N
O
R2
R1
O N
R2
O N
R2
R2 82
2
83 83
nitrone
84
85
2. BnO (CH2)5I, 50 °C, THF
RO
TBDPSO 86
TBDPSCl 90% imidazole CH 2Cl2
HO 2C
1. BCl3 -Me 2S, CH2 Cl2 (97%)
TBDPSO
2. Jones oxidation (98%)
90%
87
R=H R = TBDPS
1. Me3 CCOCl O
O S N Li H THF
Cl 2.
N=O
O S N H 70%
88
1. NaHMDS, THF, -78 °C O
2. 1-chloro-1-nitrosocyclohexane
NHOH OTBDPS
O
O
90
3. HCl, H 2O 4. NaHCO3 , H 2O
toluene, 80 °C, styrene
S N H
Oppolzer Auxiliary
Histrionicotoxin-10
O OTBDPS O
84% 89
Page 354
BnO 1. n-BuLi, -30 °C, THF
10:56 AM
or
12/21/2010
O
R1
N
Organic Synthesis via Examination of Selected Products
H
O
b1026
R1
R1
Organic Synthesis via Examination of Selected Natural Products
Davison, E. C.; Fox, M. E.; Holmes, A. B.; Roughley, S. D.; Smith, C. J.; Williams, G. M.; Davies, J. E.; Raithby, P. R.; Adams, J. P.; Forbes, I. T. Press, N. J.; Thompson, M. J. "Nitrone dipolar cycloaddition routes to piperidines and indolizidines. Part 9. Formal Ssynthesis of (-)-pinidine and total synthesis of (-)histrionicotoxin, (+)-histrionicotoxin and (-)-histrionicotoxin 235A" J. Chem. Soc., P er kin 1 2002, 1494-1514. See also Smith, C. J.; Holmes, A. B.; Press, N. J. "The total synthesis of alkaloids (-)-histrionicotoxin 259A, 285C and 285E" J . Chem. Soc., Chem. Commun. 2002, 1214-1215. For a related approach see Karatholuvhu, M. S.; Sinclair, A.; Newton, A. F.; Alcaraz, M-L.; Stockman, R. A.; Fuchs, P. L. "A Concise Total Synthesis of DL-Histrionicotoxin" J. Am. Chem. Soc. 2006, 128, 12656-12657.
H N
b1026_Chapter-09.qxd
354
Nitrone Cycloaddition Route to Histrionicotoxin
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 355
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
355
Histrionicotoxin-10 The final synthesis we will examine was reported by Holmes (Cambridge, UK) and a closely related synthesis has been reported by Fuchs (Purdue). We will examine the Holmes synthesis. This synthesis keys off the 1,3-N,O relationship in histrionicotoxin (2). Recall that this should trigger “nitrone cycloaddition” as a possible route to consider to this target. This suggests 82 as a reasonable target that could be moved forward to histrionicotoxin. The issue here was one of regiochemistry in the key cycloaddition. Would nitrone 83 undergo cycloaddition to provide 82 or 84? The key here turned out to be the choice of R2 and a fine balance of steric and electronic effects. Whereas I refer you to the primary literature for the gory details, it turns out that when R2 = C≡CTMS, the cycloaddition gave 84, while when R2 = C≡N (a more electron withdrawing group) the desired path was followed to provide a compound of type 82. The synthesis is a nice example of how persistence, attention to detail, and a willingness to explore options resulted in achieving a synthesis via a plan that was initially discouraging in the lab. Let’s go through some of the details. The nitrone that led to completion of the synthesis is structure 97, shown on Histrionicotoxin-11. This was prepared as a single enantiomer using asymmetric hydroxylamination chemistry developed by Oppolzer.12 Thus alcohol 85 was converted to N-acylsulfonamide 89 via a reaction sequence that we will not bother to discuss. Generation of the enolate of 89, followed by hydroxylamination using 1-nitrosochlorocyclohexane as a key reagent, gave hydroxylamine 90 in good yield and with a high diastereomeric excess.
b1026_Chapter-09.qxd
356
styrene
Xc
N O
OTBDPS
Xc
OTBDPS
O
O
Ph
85%
OTBDPS
BnO
2. NaH, THF 3. BnBr
O 92
91
1. LiAlH4
N
N O Ph
93
94%
OH
2. IBX, DMSO (100%)
9:1
O
N
BnO
190 °C
Ph
B(OiPr)2
C C N H
BnO
CHO
N O
94%
Ph
1. n-BuLi, THF, -78 oC 2. B(OiPr)3
96
94%
95
TMSCH2 CN Yamamoto-Peterson Olef ination
CN
OBn
O
N 1. BCl3 -Me 2S
O
N
1. i-Bu2AlH, PhMe; aqueous MeOH
2. MsCl, Et3N (99%) CN 80%
98
3. NaCN, DMSO 50 o C, 96 h (84%)
2. Ph3 PCH 2I I
CN 95%
O
100
KHMDS, THF 99
I
N
1. Sonogashira
I
2. Zn, AcOH 3. K2CO3 , MeOH
95% olefin homogeneity
H
O
H N
2 85% overall
(see 82) Improved Olefination Conditions (Stork)
Histrionicotoxin-11
crystalline X-ray
Page 356
97
TMS
N O
CN
CN
10:56 AM
OBn
Organic Synthesis via Examination of Selected Products
O
12/21/2010
I
94
+
1. HF-CH 3 CN (91%)
O
IBX =
b1026
PhCH=CH2
Organic Synthesis via Examination of Selected Natural Products
O
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 357
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
357
Histrionicotoxin-11 When 90 was heated in toluene, the hydroxylamine added to the alkyne to afford nitrone 91 (after tautomerization of a presumed intermediate Nhydroxyenamine) which was trapped by styrene in an intermolecular 1,3-dipolar cycloaddition to provide 92. An oxidation state adjustment (with removal of the chiral auxiliary) gave 93. The TBDPS protecting group was removed and the resulting primary alcohol was oxidized with IBX (94) (related to the Dess-Martin periodinane) to give aldehyde 95. Application of the Yamamoto variation of the Peterson olefination gave 96 with decent control over olefin geometry.13 Heating of 96 gave the desired nitrone cycloaddition product (98) in excellent yield. This reaction involved a retro-1,3-dipolar cycloaddition, followed by the key intramolecular cycloaddition, another nice example of the use of a retro-cycloaddition to generate a reactive intermediate for use in a pericyclic reaction (recall the generation of acylnitroso compounds via a retro-Diels-Alder reaction). The synthesis was completed by removal of the benzyl protecting group, homologation of the resulting primary alcohol, conversion of bis-nitrile 99 to a bis-aldehyde, Wittig olefination to give bis-vinyl iodide 100, a double Sonogoshira reaction to install the eneynes, and reduction of the N—O bond. It is notable that this strategy was also used to prepare a number of other histrionicotoxins that differed by virtue of oxidation state along the 4- and 5-carbon sidechains. This completes our look at histrionicotoxin. Let’s move on to pumiliotoxin-C (3) a simpler target that has also received lots of attention.
H
N
H
Me H
Path A
Path B
2
Me H 3
6
10
H
H
N H H 111 H
3
101
cis-perhydroquinoline with substituents equatorial
N H H
Steric Approach Control
Path C
Me H
1
Me
Me
O 103
104
t 5
2
N H
Me
4
3
8
N
N H H
H
102
106
6
tautomerization issues
regiochemical and stereochemical issues
109
Me t
O
1
CHO
2
Me
O
4
t 3
NH O 105
N H H 112
9
7
108
O
R
5
NH 110
108
Pumiliotoxin-1
R
t
7
6 8
N 9
O
Me
107
113
O
H R
Page 358
asymmetric induction?
10:56 AM
N H H
Organic Synthesis via Examination of Selected Products
Pumiliotoxin-C
12/21/2010
Me H
3
b1026
H
Organic Synthesis via Examination of Selected Natural Products
H
Me
b1026_Chapter-09.qxd
358
Diels-Alder Approaches to Pumiliotoxin C
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 359
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
359
Pumiliotoxin-1 Pumiliotoxin-C (3) has been an extremely popular target for synthesis. This alkaloid is a simple cis-decahydroquinoline with four stereogenic centers. The most stable conformation of 3 places the two ring substituents (propyl and methyl groups) on equatorial sites. One of the rings is a cyclohexane and thus, it is tempting to develop strategies that revolve around Diels-Alder reactions. We examine several syntheses that adopted this approach, and then glance at a few syntheses that use alternative approaches. Working backwards from 3, there are six places one could “retrosynthetically” place a double bond in the cyclohexane substructure. Two are shown here. Path A leads back to cyclohexene 101. Disconnection of 101 leads back to dienophile 102 and diene 103 (piperylene). This is not a good dienedienophile pair. Enamine 102 is electron-rich, whereas 103 is also electron-rich. Furthermore, enamine 102 would exist in its tautomeric imine form. Even if 102 were to undergo cycloaddition with 103, reaction would likely occur opposite the n-propyl group and afford an unusable diastereomer. One way to deal with both the tautomerization and stereochemical problems would be to use enamide 104 as the dienophile. This would require installation of the propyl sidechain at a later stage of the synthesis. Enamide 104, however, is still not an electron-deficient dieneophile and regiochemical problems would be likely to plague any cycloaddition that might occur. This problem can be fixed by moving to 105 as the dienophile. The nitrogen could be installed with retention of stereochemistry using a Beckman rearrangement, but cycloadditions between 105 and piperylene would surely give the wrong regiochemistry. The regiochemistry problem can be solved by changing the diene to 1,3-butadiene and introducing the methyl group at a later stage of the synthesis. Overall this plan requires a lot of “fixing” to enable construction of the ring system using a Diels-Alder approach. Path B works back through cyclohexene 111 and an intramolecular DielsAlder reaction of dienamine 112. Once again, enamine tautomerization is likely to present big problems, but once again this can be overcome by using dienamide 113. This plan provides an opportunity to control absolute stereochemistry if 113 could be prepared in enantiopure form. The dienophile olefin geometry would afford the proper relative stereochemistry at C2 and C3, but stereochemistry relative to C6 and C10 is not guaranteed. Path C is an intermolecular variant of Path B. Thus, reduction of imine 106 would be expected to occur from the convex face of the perhydroquinoline to give the proper stereochemistry at C6. Imine 106 might be expected to result from a endo-cycloaddition between dienamide 108 and
2. n-PrCN
Et 2O
117 70%
NH 2
Me H
Me H
1. H2 , Pd/C, 5 h MeOH, rt
3
215
N H H 120 CO2 Me 22%
oC,
20 h 119
N CO2 Me
2. ClCO2 Me
N 118
49%
80%
Chiral Pool Synthesis
CO 2H
Toluene, 230 °C
7 Steps
H2 N
N
norvaline
O 121
122
OTMS TMS
123
O
(BSA) via Grignard opening of N-Ts aziridine
N
N
N
60% 124
O 3 steps
Avoidance of allylic strain? Minor amounts of diastereomers obtained in Diels-Alder.
Pumiliotoxin-2
pumiliotoxin-C
3
Page 360
2. HCl, AcOH, H 2O (1:1:1), ∆, 30 h
N H H H
1. NaN(SiMe3 )2
5% solution in PhMe
10:56 AM
crotonaldehyde
Organic Synthesis via Examination of Selected Products
116 NOH 52%
115 O 62%
3. H3 O+
114
LiAlH 4
12/21/2010
NH2 OH
b1026
1. Mg, Et2 O
Organic Synthesis via Examination of Selected Natural Products
Oppolzer, W.; Frostl, W.; Weber, H. P. "The Total Synthesis of dl-Pumiliotoxin-C" Helv. Chim. Acta 1975, 58, 593. Oppolzer, W.; Flaskamp, E. "An Enantioselective Synthesis and the Absolute Configuration of Natural Pumiliotoxin-C" H elv . Chim. Act a 1977, 60, 204. Oppolzer, W.; Flaskamp. E.; Bieber. L. W. "Ef ficient Asymmetric Synthesis of Pumiliotoxin C via Intramolecluar [4+2] Cycloaddition" Helv . Chim. Act a 2001, 84, 141-145. For a related approach to pumiliotoxin-C and other Dendrobatid alkaloids see Banner, E. J.; Stevens, E. D.; Trudell, M. L.; Tetr ahedr on Lett . 2004, 45, 4411-4414.
Br
b1026_Chapter-09.qxd
360
First Total Synthesis of Pumiliotoxin C
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 361
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
361
trans-olefin 107, followed by deprotection of the nitrogen to liberate an amine. As shown here, the Diels-Alder would be plagued by regiochemical problems and poor dienophile electronic properties. But this could be fixed by using crotonaldehyde (109) as the dienophile followed by introduction of C5-C9 using carbonyl addition chemistry. In fact this looks like the most “secure” route when it comes to ensuring relative stereochemistry throughout the synthesis. Let’s begin with the first synthesis of pumiliotoxin-C, reported by the Oppolzer group in 1975.
Pumiliotoxin-2 The initial target in the Oppolzer synthesis was dienamide 119. This was prepared starting with homoallylic bromide 114. Formation of the Grignard reagent and reaction with propionitrile gave ketone 115. Preparation of oxime 116 was followed by reduction to provide 117. Imine formation with crotonaldehyde was followed by deconjugative N-acylation to provide 119. The critical Diels-Alder reaction proceeded to give 120 in 22% yield at 215 oC. The other major product (37%) was derived from “formal” hydrolysis of the dienamide. The synthesis was completed by catalytic hydrogenation of the olefin and removal of the carbamate protecting group under acidic conditions. The synthesis was modified, to provide an improved yield in the cycloaddition and introduce stereoselectivity at the impending C6 stereogenic center, by using norvaline (121) as the starting material and dienamide 122 as the cycloaddition substrate. Using BSA as a water scavenger improved the yield of the cycloaddition product (124) to 60%. The only isolated side products were minor amounts of diastereomeric cycloadducts. One might imagine that the major product is derived from the “approach of dienophile to diene” shown in structure 123. Allylic strain with the amide is minimal in this cyclization transition state. The synthesis was completed in much the same manner as from 120.
b1026_Chapter-09.qxd
362
Intermolecular Diels-Alder Route to Pumiliotoxin-C Overman, L. E.; Jessup. P. J. "Synthetic Applications of N-Acylamino-1,3-dienes. An Efficient Stereospecific Total Synthesis of dl-Pumiliotoxin C, and a General Entry to cis-Decahydroquinoline Alkaloids" J. Am. Chem. Soc. 1978, 100, 5179.
NHCO2 Et
2. NaN3
neat, 110 o C, 2.5 h
CHO
H
OHC
126
3. EtOH, ∆
Me
NHCO2 Et
129
128
O
O
Cu (bromine sponge)
NH3 Br
NHCO2 Et 131
132 NaHCO 3 H 2, Pd/C HCl, EtOH 95% H pumiliotoxin-C N
HCl, EtOH
N H H H
133
O H 2, Pd/C
HBr, HOAc
H 2, PtO2
O
3 83% overall
Pumiliotoxin-3
If CO2 Et = CO2Bn
NHCO2 Et
130 83%
Page 362
(MeO) 2
O P
10:56 AM
61% endo cycloaddition
Organic Synthesis via Examination of Selected Products
125
EtO2 CHN
12/21/2010
CO2 H
b1026
H
CHO 1. SOCl2
Organic Synthesis via Examination of Selected Natural Products
127
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 363
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
363
Pumiliotoxin-3 The Overman group reported a synthesis that proceeded along the disconnection shown in Path C of Pumiliotoxin-1. The key cycloaddition between dienamide 126 (from 125) and crotonaldehyde (127) gave 129 with excellent control over regiochemistry and stereochemistry. The remaining carbons were introduced via a Horner-Wadworth-Emmons reaction. Catalytic hydrogenation of the olefins converted cycloadduct 130 to keto-carbamate 131. Acid promoted “hydrolysis” of the carbamate gave ammonium salt 132. Conversion of 132 to its free base resulted in formation of imine 133. Catalytic hydrogenation of the imine provided pumiliotoxin-C (3). When the reaction sequence was performed using the benzyl carbamate corresponding to 126, conversion of 130 (CO2Et → CO2Bn) to 3 was accomplished in a single step! This synthesis provides a nice example of an intermolecular DielsAlder reaction having some advantages over its intramolecular counterpart.
H
O
H
135
N H
O
136
4 steps
N H 137
N H H 3
H Eschenmoser
H
53%
Me
Me 1. CH 2=CHCN O TMSO
OTMS
2. 10% aq HCl
CN
OTMS 141
170 °C xylene
TMSO
OTMS
5 steps
4 steps N H
O
H
N H H
3
2. NaOMe, MeOH
1. 1% HCl
CO 2Et
CH3 CH=CHCO 2Et
H 140
139 37%
TMSO
O
HO
138
H
H
4 steps
Note that difunctional relationships are better placed in these routes than in the perhydroindanone route
O O
142
H 2N 143
Pumiliotoxin-4
O
Page 364
Ibuka, T.; Mori, Y.; Inubushi, Y. "A New Stereoselective Synthesis of dl-Pumiliotoxin C Using Novel 1,3-Bis(trimethylsilyloxy)-1,3-dienes" Tetr ahedron Lett. 1976, 3169-3172.
10:56 AM
N H
H O
Organic Synthesis via Examination of Selected Products
2. TsCl H O 134
H 3 steps
12/21/2010
H 10 steps
b1026
H 1. NH 2OH
Organic Synthesis via Examination of Selected Natural Products
Ibuka, T.; Inubushi, T.; Saji, I.; Tanaka, K.; Masaki, N. "Total Synthesis of dl-Pumiliotoxin C Hydrochloride and its Crystal Structure" Tetr ahedr on Lett. 1975, 323-326. See also Inubushi, Y.; Ibuka, T. "Synthesis of Pumiliotoxin C, A Toxic Alkaloid from Central American Arrow Poison Frog, Dentrobates Pumilio and D. Auratus" Heter ocy cles 1977, 8, 633-660 (this is a brief review). H
b1026_Chapter-09.qxd
364
Additional Diels-Alder Approaches to Pumiliotoxin-C
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 365
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
365
Pumiliotoxin-4 The synthesis shown on this panel follows Path A as delineated in Pumiliotoxin-1. The initial cycloadduct (from cyclopentenone and 1,3butadiene) was nicely converted to lactam 135 by formation of the oxime and subsequent reaction with p-toluenesulfonyl chloride. The conversion of 135 to 136 required 10 steps. Not being able to carry the methyl group in the diene (see Pumiliotoxin-1) was certainly costly. Another 7 steps were needed to introduce the propyl side chain. A somewhat related approach that makes good use of odd-difunctional relationships in its design involved the Diels-Alder reaction between 138 and acrylonitrile to provide 139 after hydrolysis of an intermediate silyl enol ether. Bromination of 139, reduction of the α-bromoketone to an alcohol, treatment of the resulting bromohydrin with zinc in acetic acid, and treatment of the resulting olefin with 15% HClO4 in acetic acid gave lactam 140. The synthesis of 3 was completed in a manner similar to the synthesis from 136. A related synthesis that proceeded through 140 (derived from 141 and ethyl crotonate) was also reported by the Inubushi group.
b1026_Chapter-09.qxd
366
A Biomimetic Approach to Pumiliotoxin-C O
4
H
HO
histrionicotoxin
gephyrotoxin
R1
N H
R1
O 144a
144b
Bonin, M.; Royer, J.; Grierson, D. S.; Husson, H.-P. "Asymmetric Synthesis VIII: Biogenetically Patterned Approach to the Chiral Total Synthesis of (-)Pumiliotoxin-C" Tetr ahedr on Lett . 1986, 27, 1569-1572. For a review discussing this "general approach" to alkaloids see Husson, H.-P. "A New Approach to the Asymmetric Synthesis of Alkaloids" J . Nat . Pr od. 1985, 6, 894-906. Ph NC
N
O
1. LDA 2. R-I
144c
Ph CN N O
O
LiBF4 , CH3 CN H2O, 60 °C, 30min
O
Ph CN N
Ph O
4:1
NC
146
147
95%
66%
55%
Ph
Ph N
O
N
O 148
Pumiliotoxin-5
O PrMgBr, Et2O
N
∆, 1.5 h
O
145
Al2O 3, CH 2Cl2
149
O
15 °C, 1.5h
Page 366
R2
N
10:56 AM
O
R2
Organic Synthesis via Examination of Selected Products
pumiliotoxin C
2
12/21/2010
N H
3
b1026
N H
Organic Synthesis via Examination of Selected Natural Products
H
N
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 367
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
367
Pumiliotoxin-5 We have already seen that speculation (or hard information) about the biosynthesis of natural products can suggest strategies for laboratory syntheses. It has been speculated that the dart-poison alkaloids under consideration here arise from acyclic compounds (most likely polyketide-derived). For example enamine 144a (perhaps derived from an acyclic aminoketone) might undergo an intramolecular aldol-type reaction to afford the decahydroquinoline skeleton of pumiliotoxin C (3) or gephyrotoxin (4). Depending on the nature of R1 and R2, one or the other of these alkaloids could result “downstream”. On the other hand, the imine tautomer of 144a (144b) might undergo an intramolecular Mannich reaction to afford the azaspiro[5.5]undecane core of histrionicotoxin (2). Let’s look at an enantioselective approach to pumiliotoxin-C that mimics this biosynthetic proposal. The synthesis begins with 144c, prepared from phenylglycinol and glutaraldehyde. Deprotonation of 144c and alkylation of the resulting anion provided 145. Hydrolysis of the acetal gave 146 which was converted to perhydroquinoline 147 (as a mixture of stereoisomers) upon treatment with alumina. The transformation of 146 to 147 follows a mechanistic pathway that mimics the biosynthetic proposal. Ionization of the nitrile (recall the Stork approach to reserpine) and loss of a proton might give enamine 148. An intramolecular “aldol-dehydration” might give iminium ion 149. 1,4Addition of cyanide to the iminium ion would provide 147.
O Na, NH 3
Ph
H H N
H 2, Pd(OH)2 /C
N
H H N 7 : 3
-78 °C
95%
H
NC 151
H 3
152
H N
H
H
H
H
Acylnitroso Diels-Alder Approach
Naruse, M.; Aoyagi, S.; Kibayashi, C. "Total Synthesis of (-)-Pumiliotoxin C by Aqueous Intramolecular Acylnitroso Diels-Alder Approach" T etr ahedr on Lett . 1994, 35, 9213-9216. wrong stereochemistry BnO HN
O
OH
Pr4 N IO4 H 2O-MeOH (6:1)
BnO
BnO
H
H
N O
O
3 steps
H
5
7 steps
N
8a
O
N
H
Bz
156
0 oC 153
H
O
154
155
11 steps from malic acid allylic strain dictates conf ormational preference of benzamide
Pumiliotoxin-6
N O
Bz
2
7 steps
H
H 3
N H
H
Page 368
separated by chromatography
10:56 AM
N
Organic Synthesis via Examination of Selected Products
H
Very little insight into factors governing stereochemistry presented
b1026
100%
12/21/2010
nitrilere duction
Organic Synthesis via Examination of Selected Natural Products
150
b1026_Chapter-09.qxd
Ph N
368
O
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 369
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
369
Pumiliotoxin-6 Treatment of 147 with propylmagnesium bromide gave 150, and sodiumammonia removal of the nitrile provided 151. Hydrogenolysis of the benzylic C–N bond, ionization of the N,O-acetal to an iminium ion (and possibly tautomerization to an enamine) followed by hydrogenation, gave 152 and pumiliotoxin-C (3). The yields through this sequence were very good, but the route suffered in terms of diastereoselectivity. Nonetheless, the final product was produced as a single enantiomer. Notice that this is a “chiral auxiliary” approach to the problem of enantioselectivity, but the chiral auxiliary is sacrificed in the process. We will spend the rest of our time focusing on selected steps of a few syntheses, all of which were designed to provide single enantiomers of pumiliotoxin-C. The first one uses malic acid (a common dicarboxylic acid found in apples) as the source of enantioselectivity, and features the now familiar nitroso-Diels-Alder chemistry. Thus, hydroxamic acid 153 was prepared and oxidized to the key intermediate. Cycloaddition occurred to provide 154 which was then converted to pumiliotoxin via 155 and 156 in a rather laborious sequence of reactions. One of the more interesting problems encountered in this sequence was the conversion of keto-benzamide 156 to 3. Notice that 156 has the “wrong” stereochemistry for the pendant methyl group. This is presumably because 156 prefers a conformation in which C2 and C8a are axially disposed, and thus when the methyl group is equatorially disposed (thermodynamically most stable site) it is cis (not trans) to the propyl group. Can you suggest how the stereochemistry at C5 was “fixed”?
b1026_Chapter-09.qxd
370
Some Enantioselective Approaches to Pumiliotoxin-C Directed Hydrogenation
H
O
PF6 [Ir(cod)py(PCy)3 ]
N
N
H
O N
Pumiliotoxin-C
10 steps
+
t -BuOH H
H O
N H
N H H 159
intersects with Overman intermediate
H N H O H 100% (99:1) 160
H 2, CH 2Cl2 Crabtree’s catalyst
from anthranilic acid and proline
O
O
O CN
2 steps
H
RuH 2(PPh 3) 4 (3 mol%)
O H
N
H 2O, DME 161
via enamine
164
56% (1:9)
163
162
N
+
H 2, 5% Pd/C (K-type) O H H
(-)-Pumiliotoxin C (3)
O H
N + H
165
H
N H
90% (98:2)
166
Murahashi, S.-I.; Sasao, S.; Saito, E.; Naota, T. "Ruthenium-Catalyzed Hydration of Nitriles and Transf ormation of δ-Ketonitriles to Ene-Lactams: Total Synthesis of (-)-Pumiliotoxin C" T et rahedr on, 1993, 49, 8805.
Pumiliotoxin-7
Page 370
Schultz, A. G.; McCloskey, P. J.; Court, J. J. "Enantioselective Conversion of Anthranilic Acid Derivatives to Chiral Cyclohexanes. Total Synthesis of (+)Pumiliotoxin-C" J. Am. Chem. Soc. 1987, 109, 6493. Also see Schultz, A. G.; McCloskey, P. J. "Carboxamide and Carbalkoxy Group Directed Stereoselective Iridium-Catalyzed Homogeneous Olefin Hydrogenations" J. Org. Chem. 1985, 50, 5905.
10:56 AM
49% (1:5)
158
H O
Organic Synthesis via Examination of Selected Products
H N O H 157
O
12/21/2010
N
H
b1026
K, NH3 , THF
Organic Synthesis via Examination of Selected Natural Products
O
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 371
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
371
Pumiliotoxin-7 The next synthesis uses a chiral auxiliary approach in which the auxiliary is not sacrificed. The synthesis begins with anthranilic acid derivative 157. A potassium-ammonia reduction provided a mixture of diastereomers 158 and 159. A directed hydrogenation of the major diastereomer (159), using Crabtree’s cationic iridium catalyst, provided 160.14 This material was moved forward to pumiliotoxin-C through intermediates we encountered in the Overman approach (compare the substitution pattern and relative stereochemistry of 160 with 131 on Pumiliotoxin-3). Murahashi reported a synthesis that used the monoterpene pulegone (161) as a source of chirality and featured novel ruthenium catalyzed chemistry, retro-aldol-dehydration and nitrile hydrolysis reactions. Thus 161 was converted to 162 via reaction of its enamine with acrylonitrile. Treatment of 162 with an appropriate ruthenium catalyst, in aqueous dimethoxyethane, gave enamides 163 (minor) and 164 (major). This transformation featured the aforementioned nitrile hydrolysis accompanied by cyclization of the intermediate keto-amide to the enamide. In the case of 164, this was accompanied by the aforementioned retro-aldol-dehydration. Catalytic hydrogenation of the mixture of enamides gave 166 (minor) and 165 (major). The latter compound was transformed to pumiliotoxin-C (3) as we have already seen in previous syntheses. This synthesis was clearly performed to highlight methodology, and it does so quite effectively. Notice that the 1,5-difunctional relationship present in 162 was constructed using “normal” conjugate addition chemistry.
1.
Li
N
169 H
CO 2Me
4 Steps
Ph
CO 2Me
167
H Ph
168
from pulegone
N
Ph
H
NHTs
61% overall
3
171
170
OH
LiAlH 4
3 Steps
(-)-PTX
2:1
172
2. 2 mol% Pd(PPh3 )4 1.1 eq CH2 =CHCH2 OAc
84% 174 tr ans-cis = 9:1 tr ans = 96% ee
Ph Ligand =
O P N O Ph
H 175
NHTs 3
176
46% pure after chromatography
173
Dijk, E. W.; Panella, A.; Pinho, P.; Naasz, R.; Meetsma, A.; Minnaard, A. J.; Feringa, B. L. "The Asymmetric Synthesis of (-)-Pumiliotoxin C using Tandem Catalysis" T etr ahedr on 2004, 60, 9687-9693.
For a review of 7 recent approaches to pumiliotoxin-C [Habermehl (1998), Bach (1998), Comins (1993), Kunz (1999), Mori (2001), Stille (1993), Padwa (2000)] see Sklenicka, H. M.; Hsung, R. P. "Recent Approaches to cis-Azadecalins: Synthesis of Dendrobatid Alkaloid Pumiliotoxin C" Chemtr acts-Organic Chemistr y 2002, 15, 391-401.
Pumiliotoxin-8
Page 372
H O
10:56 AM
O
1. 0.5 mol% Cu (OTf)2 1.2 eq Me2Zn toluene, -30 °C, 3h 1 mol% ligand
Organic Synthesis via Examination of Selected Products
Schultz Intermediate Davies, S. G.; Bhalay, G. T etr ahedr on: Asy mmet ry 1996, 7, 1595-1596
12/21/2010
2. TsCl, Et3N CH 2Cl2
CO 2Me
b1026
H
Organic Synthesis via Examination of Selected Natural Products
1. 10% Pd/C MeOH, H 2
(-)-PTX 2. 2,6-di-t er t-butylphenol
O
b1026_Chapter-09.qxd
372
Ph
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 373
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
373
Pumiliotoxin-8 Davies has developed methodology for the preparation of β-aminoesters. One application of this methodology was to the synthesis of 171, related to the Schultz and Overman intermediates, carried on to pumiliotoxin-C in the usual manner. Pulegone once again served as the source of chirality for the methyl-bearing carbon. This was converted to 168, which then reacted with chiral lithium amide 169 to provide 170 after protonation of the intermediate enolate with 2,6-di-tert-butylphenol. Hydrogenolysis of the benzylic C–N bonds and the olefin, followed by formation of the sulfonamide, gave 171. The final synthesis we will consider uses asymmetric catalysis to establish absolute stereochemistry. Thus, treatment of cyclohexenone (172) with dimethylzinc and catalytic Cu(I) in the presence of chiral ligand 173 proceeded with good asymmetric induction. Alkylation of the intermediate enolate using allyl acetate in the presence of Pd(0) provided 174 with 96% ee. This ketone was reduced to provide a mixture of alcohols which were separated and converted to 171 by degrading the allylic side chain to a carboxyl group, and displacing the alcohol with a nitrogen nucleophile. This ends our look at pumiliotoxin-C. Other syntheses have been reported and some of the more recent approaches have been reviewed in a comparative manner, suggested reading for those interested (see reference at bottom of Pumiliotoxin-8). We will next move to the structurally related alkaloid gephyrotoxin (4).
b1026_Chapter-09.qxd
374
First Synthesis of dl-Gephyrotoxin Fujimoto, R.; Kishi, Y.; Blount, J. F. "Total Synthesis of dl-Gephyrotoxin" J. Am. Chem. Soc. 1980, 102, 7154-7156.
CO2Et
O
O
180
1. EtOC CMgCl THF, rt
O
separable
OH
1. LiBH4, THF, rt 2. KH, THF, rt 3. BnBr 4. Ba(OH)2
O 178
H N 186
A
H
H
N
12
RO Directed Hydrogenation
H
1. MsCl, Et3N CH2Cl2, ∆
H
H2, Pd/C HClO4 MeOH, rt
2. LiBr, DMF, rt
N Cbz
88% (c:t = 12:1)
N Reduction
179
CO2Et
183
48% O
H
EtO2C
2. ClCO2Bn, pyridine CH2Cl2, rt
182
HO
HO
N Bn
2. 5% aq. HCl
H
187
1. H2 (60 psi), Pd/C MeOH, HClO4, rt
CO2Et
EtO2C
H
H
RO
177
BnO
R = Bn R = H (83% overall)
H 185 85%
O TsOH, PhH, ∆
HO
N H 184 65%
H H
N
1. Pd/C, EtOAc, 60 psi H2: A is the major product (51%) along with 19% of material missing the 2o hydroxyl group 2. Pt/Al2O3, EtOAc, 60 psi H2: B is the major product (61% as the acetate) with about 6% of A. The C12 acetate, SiR3 ethers, and alkane (OR = H), gave no reduction under the same conditions.
HO 188
B
Gephyrotoxin-1
OBn
Page 374
N Bn 181
2. 5% aq. HCl
4
N
178
H
HO
H
+
O
10:56 AM
4
N
Organic Synthesis via Examination of Selected Products
HO
1. EtOC CMgCl THF, rt
HO
b
2
H
HO
N Bn
N
12/21/2010
H H
H
H
Gephyrotoxin
2
b1026
N
OR
a
5
5
H
O
O
H
H
Organic Synthesis via Examination of Selected Natural Products
O H
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 375
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
375
Gephyrotoxin-1 I am going to indulge myself a little with this section. Gephyrotoxin (4) was the first alkaloid my group worked on when I began my independent career in 1978 and thus, it holds a special place for me in the world of natural products chemistry. This alkaloid resembles pumiliotoxin-C to the extent that it has a cis-perhydroquinoline substructure with substituents at C2 and C5 (quinoline numbering). The stereochemistry at the ring stereogenic centers differs at C2 from pumiliotoxin-C. The fundamental structure and absolute stereochemistry was determined by X-ray crystallography (anomolous dispersion technique). We will look at three approaches to this alkaloid with care, and then take a quick look at the critical portions of 5 additional syntheses (not an exhaustive list). The first synthesis (of racemic material) was reported by Kishi and Fujimoto in 1980. Tricyclic vinylgous amide 177 was projected to be a key intermediate in the Kishi synthesis. Two of the five stereogenic centers of gephyrotoxin are set in this intermediate. The carbonyl group was to serve as a handle to introduce the 5-carbon cis-eneyne and stereoselective reduction of the olefin was to handle stereochemistry at the cis-perhydroquinoline ring fusion. Local difunctional relationships in 177 are “odd” and thus, one might anticipate that there should be a “carbonyl chemistry route” to this compound if one purchases the 5-membered ring. Working backward from 177 by disconnection of the bonds “a” and “b”, the plan was to couple 178 and 179 along the polar lines indicated in Gephyrotoxin-1. The symmetrical nature of 179, and a consideration of difunctional relationships, leads back to a succinimide derivative as a starting point for the synthesis. The synthesis began with N-benzylsuccinimide (180). Treatment of this imide with the Grignard reagent derived from ethoxyacetylene gave 181 after an acidic workup. This reaction nicely illustrates the equivalence of this acetylide with the enolate of ethyl acetate. It has practical advantages, however, over the enolate. It is stable at elevated temperatures and is less sterically hindered. On the other hand, ethoxyacetylene is a less friendly reagent to handle (and is more expensive) than ethyl acetate. Another important point is that imides are ketone-like in their reactivity. They should not be confused with amides! Treatment of 181 with the same acetylide gave 182. Once again it is important to recognize that the carbonyl group in 181 is imide-like (rather than amide-like). In fact it is a vinylogous imide (see Functional Groups-5 and the Principle of Vinylogy). Catalytic hydrogenation of 182, with concomitant hydrogenolysis of the N-benzyl group, followed by acylation of the pyrroldine nitrogen, gave 183. The major stereoisomer had cis
b1026_Chapter-09.qxd
12/21/2010 b1026
376
10:56 AM
Page 376
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
stereochemistry, the reduction of the first olefin presumably controlling the stereochemical course of the second olefin reduction. Modification of the two acetic acid sidechains converted 183 to 184. Reaction of pyrrolidine 184 with dione 178 gave enamine 185. The hydroxyl group was activated as the mesylate, transformed to the bromide, and an intramolecular alkylation of the enamine ensued to give 186 (R = benzyl). The next task was diastereoselective reduction of the olefin. It turns out that 186 (R = Bn, Ac, OSiR3) gave no reduction or selectivity under typical conditions for catalytic hydrogenations. In the presence of acid, it was possible to remove the benzyl group to provide 186 (R = H; same as 177). Hydrogenation of this substrate over Pd/C in ethyl acetate gave mainly 187 (A) with the wrong stereochemistry at the ring fusion. This is the result one might have expected based on steric accessibility of the enone to the surface of a heterogeneous (or for that matter a homogeneous) catalyst. When the catalyst was changed to Pt on alumina, however, the desired stereoisomer (188 = B) was obtained in good yield. The notion is that the hydroxylethyl group (via an interaction of the alcohol with alumina) directed the catalyst to the desired surface of the enone. Such directing effects had been previously noted, but this is a dramatic application of this phenomenon to a complex problem in synthesis.15 The primary alcohol was then selectively acylated to give 189 (see Gephyrotoxin-2).
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 377
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
377
Gephyrotoxin-2 The next stage of the synthesis called for “reductive homologation” of the C5 alcohol by two carbons, to ultimately serve as a handle for introduction of the cis-eneyne unit. Oxidation of 189 provided the C5 ketone. Application of the aforementioned acetylide addition chemistry, and cleavage of the acetate (MeMgCl), gave 191 as a mixture of geometrical isomers. Protection of the alcohol as either an acetate or TBDPS ether provided 192 and 193, respectively. As a prelude of things to come, reduction of 190 with either “big” (LiEt3BH) or “little” (NaBH4) hydrides gave only 189. It appears that 190 adopts a conformation in which only the “axial” face of the ketone is exposed for nucleophilic addition. The same seems to be qualitatively true of unsaturated esters 191 and 192. For example, catalytic hydrogenation of 191 over rhodium on alumina provided an equal mixture of 194 and 195 (R = CO2Et). It is surprising that this reaction provided as much of the desired stereoisomer (195) as it did! Dissolving metal reduction of 192 gave mainly the product derived from protonation of the presumed carbanionic intermediate from the sterically most accessible (or perhaps stereoelectronically prefered) axial site at C5. Heterogeneous catalytic hydrogenation of 193, however, did provide the desired stereochemistry at C5. It was suggested that the “huge” tert-butyldiphenylsilyl (TBDPS) group simply discouraged the catalyst from approaching the “top” face of the olefin, relative to the alternative face. This remote “blocking” effect is interesting, and is related to the strategy used by Corey to control stereochemistry at C15 in his approach to the prostaglandins (Prostaglandins-11). Continuing with the synthesis, reduction of 195 (R = CO2Et) gave the corresponding primary alcohol (R = CH2OH). Oxidation of the alcohol gave the aldehyde (195 where R = CHO). A Wittig reaction between the interesting ylid derived from 196, and careful acid hydrolysis of the intermediate allene, gave unstable cis-enal 197. Application of the Corey-Fuchs terminal acetylene synthesis completed the synthesis of gephyrotoxin (4). The synthesis described above produced racemic gephyrotoxin. An enantioselective synthesis was performed using pyrroglutamic acid as the point of departure. The starting material clearly delivered the absolute stereochemistry that had been assigned to the natural product based on crystallography. The specific rotation, however, of this synthetic material was opposite in sign (same in magnitude) to that reported for the natural material. So this produced a dilema. Something is wrong somewhere, or perhaps the frogs used to isolate the material used for optical measurements differed from those used to isolate material used in the crystallographic studies, and produced enantiomeric alkaloids. The production of one enantiomer of a natural product by one plant source, and the
Swern
5
5
H N
E:Z = 1:1
H H
H H
190
H
3. Workup
AcO
HO
89%
189
H
or Ac 2O, CH 2Cl2 , 4-DMAP
N
H
N
RO
89% 191
R = TBDPS (193) R = Ac (192)
H
H
190
unstable
HO
EtO2 C H
[H]
N H
R=H 192 R = Ac 193 R = TBDPS
N
H
1. Li, NH 3 2. LiAlH4 3. TBDPSCl 1. 5% Rh/Al2 O3 , H2 (1 atm), hexane 2. LiAlH4
+ H
H
NaOEt
N
H
H
194
Br 196
5
TBDPSO 1. H2 , Rh/Al2O 3 2. TBDPSCl
1. PCC 2. Ph3 P
H
R
5
H
RO 191
H
TBDPSO
H
1
35
1
H
3. PTS, acetone H2 O
195
1
H
OHC
OEt H
N 197
TBDPSO 1. Ph3 PCH 2Cl Cl 2. MeLi, THF; TMSCl 3. TBAF H
1
10
Gephyrotoxin
H R
H
N
about 45% overall R’O R = TMS R’ = TBDPS (198) R = R’ = H (4)
Gephyrotoxin-2
Page 378
R
H
10:56 AM
N
H
Organic Synthesis via Examination of Selected Products
H O
b1026
AcO
2. MeMgCl
N
TBDPSCl, imidazole, DMF
12/21/2010
NaBH4 or L-Selectride
EtO 2C
H
Organic Synthesis via Examination of Selected Natural Products
H
EtO 2C 1. EtOC CMgCl THF, rt
H
b1026_Chapter-09.qxd
O
H
378
HO
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 379
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
379
other by another plant source, is known. Thus this suggestion is not without some merit. Still, this “stereochemical problem” remains unresolved.16
Gephyrotoxin-3 The next synthesis features stereochemical aspects of N-acyliminium ion cyclizations and extends the Kishi-Fujimoto work on remote control of hydrogenation stereochemistry. The idea was that an intermediate of type 199 could serve as a precursor to gephyrotoxin (4). The amide would serve as a handle for introducing the hydroxyethyl side chain, while an appropriate C5 substitutuent would serve as a precursor to the cis-eneyne unit. It was felt that 199 might be obtained by cyclization of N-acyliminium ion 200. We have already seen that such cyclizations can be used to prepare indolizidinones, the azabicyclo[4.3.0]nonane substructure present in 199 (see Histrionicotoxin-3). The cyclization was to set stereochemistry at C2 relative to three stereogenic centers already set in the cyclization substrate. The critical question was whether an N-acyliminium ion of type 200 would cyclize from conformation 201M, which would provide the needed stereochemistry at C2, or from conformation 201m, which would provide stereochemistry opposite to that required at C2. Whereas 201M has two axial substituents on the cyclohexane, it was anticipated that allylic strain in conformation 201m would render cyclization from that conformation higher in energy. A vinyl group was selected as the choice of “R” because, as will be seen, it introduced a useful element of pseudo-symmetry into the synthetic plan. The synthesis of the projected N-acyliminium ion precursor (208) began with a cycloaddition between cyclohexenone (202) and 1,3-butadiene. This acid-promoted cycloaddition initially gave the cis-cycloadduct, but the acid also promoted epimerization of the ring juncture to a thermodynamic 9:1 mixture of products from which 203 was isolated in reasonable yield. Reduction of 203 occurred with the expected “axial delivery” of hydride to provide largely 204. A Mitsunobu reaction gave succinimide 205. The next task was to degrade the cyclohexene to a pair of vinyl groups. This was accomplished by ozonolysis of the double bond and a reductive workup to provide diol 206. Chemistry developed by Grieco and Sharpless was then used to formally dehydrate 206 to 207, and a DIBAL reduction gave 208 and set the stage for the key cyclization.17 Of course the story has a good ending. The N-acyliminium ion cyclization gave tricyclic formate 209 in good yield with no other stereoisomers being detected. So does allylic strain control the stereochemistry in the cyclization? If it is a factor, it is certainly not the only factor. It turns out that NMR analysis of
b1026_Chapter-09.qxd
380
Gephyrotoxin: An Example of Allylic Strain as a Stereocontrol Element Hart, D. J.; Kanai, K. "Total Synthesis of dl-Gephyrotoxin and dl-Dihydrogephyrotoxin" J . Am. Chem. Soc. 1983, 105, 1255-1263.
H
5
H
N
H
N
2
H
N
H
201M
O
201m
H
O
H
O
OH succinimide
H
OHCO
HCO 2H N O
211
205
62%
H
H
HO HO
N
O
H 206 60-80%
Both hydroxyethyl groups are equatorial. The imide is really big!
O-alkylation competes 80%
N
210
O 1. O3 , MeOH 2. NaBH4
H
83% (9% isomer)
78% OH
THF
H 204
203
202
O
N
Ph3 P, DEAD
-78 oC
AlCl3 , toluene
H
NO2
Is allylic strain really a control element ?
O
SeCN
1. Bu3 P 2. H2 O2 , THF, rt H 1. NaOH (99%) 2. NaH; CS2 ; MeI (92%) 3. n-Bu 3SnH, AIBN PhH, ∆ (68-78%)
4
H 209
O
OCHO H
N
O
H R H
O HCO 2H
N
H
N
OH
O i-Bu2 AlH
H
N
O
H 79%
201M
H 208
Gephyrotoxin-3
80%
H 207
76%
Page 380
LiAlH 4, Et2 O
10:56 AM
allylic strain
N-acyliminium ion
O
O
R
H
199
4
N
N
H
Organic Synthesis via Examination of Selected Products
200
O
HO
R
2
H
H
?
H
b1026
5
H
O
H
12/21/2010
R
H
Organic Synthesis via Examination of Selected Natural Products
R
H
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 381
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
381
206 showed that both hydroxyethyl groups were axially disposed (in CDCl3). Thus, the imide is huge. If one extrapolates to iminium ion 200, this factor alone would surely favor conformation 201M over conformation 201m in the cyclization. Is allylic strain part of the picture? Most certainly yes. Compound 210 has no intrinsic bias influencing conformational preferences other than allylic strain, and it is cleanly converted to 211.18 The acyliminium ion cyclization resulted in unwanted oxidation at C4. Thus, this was removed by hydrolysis to the secondary alcohol, followed by a Barton-McCombie deoxygenation to provide 212 (Gephyrotoxin-4).19
2 BH
H
H
S
215 77%
CO2Et
S P S
TBDPSCl imidazole DMF
OMe S
216
H N
H
A CO2Et
CO2Et
R=H R = TBDPS (95%)
+
Reduction Results OR
OH
221
TMS
1. CH2Cl2 TMS
O B
OHC H
H
H
N
2. H2O, NaHCO3
100%
H2, 50 psi, Pt/Al2O3
84%
(A:B = 68:32)
H2, 50 psi, Pt/Al2O3
96%
(A:B = 4:96)
H
NaBH3CN, pH4, MeOH
90%
(A:B = 33:67)
217
TBDPS
NaBH3CN, pH4, MeOH
92%
(A:B = 35:65)
B
H 5
H
O
H 222
H TBDPS
H
N
H
CO2Et
218 CO2Et
CO2Et
1. TBAF (92%) 2. Swern (78%)
1. i-Bu2AlH (83%) 2. KH, THF (60%) 219
9:1
H
H
N
H
TBS
H
H H
H
N
1. i-Bu2AlH
N
H TMS
2. TBAF
H
OH Dihydro-GTX
4
OH
Gephyrotoxin-4
H H
N
t-BuLi CO2Et
TBS TMS
220 Gephyrotoxin (with 5% trans isomer)
H
C C C H TMS
N
94% overall
223
OHC Li
Yamamoto
CO2Et 218
Page 382
H
10:56 AM
213
Lawesson's reagent
N
Organic Synthesis via Examination of Selected Products
P
MeO
H
12/21/2010
214 84%
212
H
2. H2O2, NaOH
N
5
b1026
O
H
Reduction
5
Organic Synthesis via Examination of Selected Natural Products
H
2. BrCH2CO2Et 3. Ph3P, Et3N
H
H
1.
1. Lawesson's reagent N
OR
OR
b1026_Chapter-09.qxd
H
H 5
382
disiamylborane
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 383
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
383
Gephyrotoxin-4 Lactam 212 was converted to the corresponding thiolactam, and an Eschenmoser sulfide contraction was used to prepare vinylogous urethane 214. Hydroboration-oxidation gave 215 (R = H) and 215 (R = TBDPS) after silylation of the primary alcohol. The fifth stereogenic center was then introduced by reduction of the vinylogous urethane. Sodium cyanoborohydride reduction at pH4 (Borch conditions) gave a nasty mixture of 216 and 217, regardless of the nature of the C5 substituent. Catalytic hydrogenation of 215 (R = H) was equally ineffective in terms of stereocontrol. The KishiFujimoto synthesis, which was published at this time, provided inspiration for completion of the synthesis. When 215 (R = TBDPS) was subjected to catalytic hydrogenation, excellent diastereoselectivity was observed to afford 217 as the major product. Presumably this is another example of long-range control of stereochemistry (by blocking one face of the olefin) in a heterogenous hydrogenation. With all of the stereochemistry established, the C5 sidechain was converted to aldehyde 218. A Yamamoto-Peterson olefination using reagent 219 provided 220 with good control over olefin geometry. An oxidation state adjustment provided gephyrotoxin (4). When the olefination was conducted using vinylboronate 221, β-hydroxysilane 222 was obtained. Reduction of the ester and a syn-elimination completed a synthesis of another dart-poison alkaloid, dihydrogephyrotoxin (223).20
b1026_Chapter-09.qxd
384
Overman's Approach to Gephyrotoxin Overman, L. E.; Lesuisse, D.; Hashimoto, M. "Importance of Allylic Interactions and Stereoelectronic Effects in Dictating the Steric Course of the reaction of Iminium Ions with Nucleophiles. An Efficient Total Synthesis of dl-Gephyrotoxin" J. Am. Chem. Soc. 1983, 105, 5373-5379.
H
N
8
4
R2
N
8a
H
8a
N
2
H
226
R2 2
8
224
225
NHCbz
H
MOMO
2. PCC, NaOAc CH2Cl2
229
228
MOMO
56%
CHO
110 oC, toluene
MOMO
227
CHO
230
NHCbz 231
NHCbz
60%
81% (9:1)
9:1 mixture of endoand exo- cycloadducts
Horner Wadsworth Emmons
Preparation of HWE Reagent
MOMO
O
H
1. EtOH, H2SO4 (cat) 2. PCC HO
OHC
O
N
233
O O NHCbz
2. NaOH
O
O
OH
10-camphorsulfonic acid (10-CSA or CSA)
O
O
(MeO)2POCH2Li
238
O
CO2Et
236 55% overall
n-BuLi (MeO)2POCH3
Gephyrotoxin-5
O 232
89% (separate at this stage)
237
CO2Et 235
MOMO 1. H2, Pd/C CF3CO2H (XS)
H
O
234
77%
O
P O
(OMe)2
Page 384
2. n-BuLi; (CH2O)n
1. 5% Pd/BaSO4 pyridine
OH
10:56 AM
1. MeOCH2Cl EtN(i-Pr)2 (Hunig's base)
Organic Synthesis via Examination of Selected Products
227
OH
HO
A1,2 Strain will control stereochemistry ?
CHO
b1026
H
R5
Organic Synthesis via Examination of Selected Natural Products
R5
12/21/2010
R5
H
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 385
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
385
Gephyrotoxin-5 The next approach we will examine eminates from the Overman group (UC Irvine). Given the similarity in structure between pumiliotoxin-C (3) and gephyrotoxin (4), it is not surprising that their approach to these alkaloids are similar. The major difference between 3 and 4 is in stereochemistry at C2 of the perhydroquinoline. In the Overman synthesis of pumiliotoxin-C, a C2 imine (or iminium ion) was reduced by catalytic hydrogenation with the bowl-shaped nature of the substrate directing reduction to the convex face of the ring system. Gephyrotoxin called for delivery of hydride from the concave face of the ring system. It was hoped that an iminium ion of type 225 might predominate in solution. Why? To minimize A1,2-strain between the nitrogen substituent and C8-C8a bond (see Problem 40). If this was the case, product formation with a trans-diaxial relationship between the N-lone pair and incoming hydride would give the required stereochemistry at C2 (see structure 224 and recall the Stevens’ stereoelectronic analysis of imine addition reactions). The rest of the strategy is the same as that used for pumiliotoxin C, except C5 and C2 were changed to accommodate differences in the targets. In the forward direction, aldehyde 230 was prepared from alcohol 228 in a straightforward manner. The Diels-Alder reaction between 230 and dienamide 227 proceeded to give a mixture of endo and exo cycloadducts with 231 as the major diastereomer. Phosphonate 238 was also prepared in a straightforward manner from butyrolactone 234. A Horner-WadsworthEmmons reaction between 231 and the anion derived from 238 gave 232 at which stage the mixture of diastereomers was separated. Hydrogenation of the olefins and hydrogenolysis of the Cbz group under acidic conditions gave imine 233 after basification.
b1026_Chapter-09.qxd
386
MOMO
MOMO
H
MOMO
H
Reduction
O
O
N H H H
239 O
4
H 2, Pd/C, 25 o C
98
2
LiAlH 4, Et2 O, -19 o C*
8
92
NaEt 3BH, Et2 O, -10 o C
95
5
O
MOMO
MOMO
H
1. 1N HCl, THF 2. NaOMe 3. NaBH4
1,2,2,6,6-tetramethylpiperidine MOMO
H
2. KOH, H2 O i-PrOH,
N H H H
H O
OMe 243
242 OMe
N
H O
Cl3C 75%
PPTS =
H
1. HClO4 , H 2O, THF
1. MeOH, PPTS
TsO N H
Gephyrotoxin-6
2. Ph3 P=CHCHO
CHO
H O 241
N
H O
Cl3C 85% overall
O O
Page 386
Cl3CCH 2OCOCl *Added Lewis acids did not help: AlH3 , LiClO4 , Et3B, Et 3Al
10:56 AM
96
240
Organic Synthesis via Examination of Selected Products
NaBH4 , MeOH, 25 oC
O
b1026
237
N H H H
12/21/2010
O
N
Organic Synthesis via Examination of Selected Natural Products
H
H
+
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 387
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
387
Gephyrotoxin-6 Next came the search part of research. A sampling of reducing agents eventually revealed that the desired transformation could be accomplished by lithium aluminum hydride in ether at low temperature. One can only speculate that other reagent systems did not have the presumed stereoelectronic requirements of a hydride reduction (H2, Pd/C for example) or did not meet the conformational requirements needed to generate 240 in preference to 239. For example, if the imine nitrogen is not tightly complexed to a metal (NaBH4) or if the hydride delivery agent is too large (NaEt3BH), reactions from other conformations (or via boat-like conformations) might begin to compete with 225 (Gephyrotoxin-6). Regardless of the reason for success, the mechanistic analysis probably provided the hope to stimulate exploration of this route. With the perhydroisoquinoline stereochemistry established, efforts moved to modifying the C2 side chain to facilitate formation of the 5-membered ring. The nitrogen of 240 was protected with a Troc-group. Hydrolysis of the acetal and a Wittig reaction converted 241 to 242. Protection of the aldehyde as an acetal and basic hydrolysis of the Troc-group provided 243. Acetal hydrolysis and basification of the reaction appended the 5-membered ring via an intramolecular conjugate addition. Sodium borohydride reduction of the resulting aldehyde gave alcohol 244 (Gephyrotoxin-7). It is possible that reversibility of the conjugate addition allowed thermodynamics to control stereochemistry in the cyclization, but this point is not really clear.
b1026_Chapter-09.qxd
388
TIPS
H H
4-DMAP Et 3N
H
245
3.
TIPS
N
55%
TIPS
OTBDPS
246 TIPS = (i-Pr) 3Si
56%
For reagent see: Corey, E. J.; Rucker, C. Tetr ahedr on Lett . 1982, 719-722
Synthesis proceeds in 15 steps and 6.5% overall yield from diene. For synthesis of perhydro-GTX via this strategy see: Overman, L. E.; Freerks, R. L. "Short Synthesis of dl-Perhydrogephyrotoxin" J. Or g. Chem. 1981, 46, 2833.
4
TBAF
H H
Gephyrotoxin H
N
58% pure + 41% of mixture OH
Gephyrotoxin-7
Page 388
Stereocontrol not understood
OTBDPS
H
247
Li
86% OH
H
2. Swern oxidation
N
10:56 AM
H 244
5
Organic Synthesis via Examination of Selected Products
H N
H
1. 24% HBr, DME (86%)
TBDBSCl
12/21/2010
MOMO
H
b1026
MOMO
Organic Synthesis via Examination of Selected Natural Products
9:1
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 389
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
389
Gephyrotoxin-7 The hydroxyethyl group was protected as TBDPS ether 245 before moving to completion of the C5-eneyne. The robust MOM group was removed using HBr in DME (notice that this protecting group survived several acidic conditions upon conversion of 232 to 243). Swern oxidation provided an aldehyde. Reaction of the aldehyde with Corey-Rucker reagent 246 gave Peterson olefination product 247 with 9:1 olefin geometry favoring the Z-isomer. Fluoride was used to remove the acetylene and alcohol protecting groups to provide gephyrotoxin (4). Due to the efficiency with which the perhydroisoquinoline core was prepared, I think this synthesis is the most efficient and versatile route to this alkaloid.
b1026_Chapter-09.qxd
390
Other Approaches to Gephyrotoxin Ibuka, T.; Chu, G. N.; Yoneda, F. "A Novel Synthetic Route to dl-Perhydrogephyrotoxin" J. Chem. Soc., Chem. Commun. 1984, 597.
O
O
2. Et3 N, MeOH
OTMS
2
N H H H CO2 Et
251
95%
CO2 Et
252
majar + minor 99% (3:1)
H
H
7 Steps 254 Comins, D. L.; Kuethe, J. T.; Miller, T. M.; Fevrier, F. C.; Brooks, C. A. "Diels-Alder Reactions of N-Acyl-2-alkyl(aryl)-5-vinyl-2,3-dihydro-4-pyridones" J. Or g. Chem. 2005, 70, 5221-5234. For addition work see Comins, D. L.; Joseph, S. P.; Peters, D. D. "Preparation of 2,6-Disubstituted 2,3-Dihydro4-pyridones: Dehydrogenation of Trimethylsilyl Enol Ethers with Palladium(II) Acetate" T et rahedr on Let t. 1995, 36, 9499-9452.
H
OBn H N
?
Cbz
255
6
2
N H Cbz 256
H
O
S S O
O
toluene,
257
OBn
OBn
258 H
O H 5
Cbz
CO2 Et
253 OH
O N
N H H H
conjugate addition
5:1
O
2
N
O
O
O
H
H O2 S
N H Cbz
H
O H +
SO2 3:1
H O2 S
N H Cbz SO2
Allylic Strain Arguements Other dienes give better selectivity, but this diene is the surrogate for ethylene.
Gephyrotoxin-8
259
79%
260
Page 390
This synthesis is patterned after the "Ibuka" pumiliotoxin-C synthesis. It suffers from lack of stereocontrol in the key sodium cyanoborohydride reduction.
10:56 AM
N H H
250
248
H
1. NaBH3 CN, pH4
12/21/2010
TMSO
OTMS
H
12 Steps
Organic Synthesis via Examination of Selected Products
TMSO
CO2 Et
175 C, 48 h
b1026
CO2 Et
Organic Synthesis via Examination of Selected Natural Products
249
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 391
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
391
Gephyrotoxin-8 We will now take a brief look at several additional approaches to gephyrotoxin. The Ibuka group followed a sequence (to perhydrogephyrotoxin) that resembled their approach to pumiliotoxin-C (Pumiliotoxin-4). This resulted in the synthesis of perhydroisoquinoline 251 from 248 and 249. Reduction of vinylogous amide 251 suffered from lack of stereocontrol at C2, providing largely 252 with the required 253 as a minor product. This result is in accord with the Overman studies. Ketoester 253 was eventually converted to perhydrogephyrotoxin (254). The Commins group (North Carolina State) took an approach based on the prediction that 255 would prefer the conformation shown here to avoid allylic strain between the “gray ball” and the N-Cbz group. If this were true it was felt that a cycloaddition between 255 and ethylene (or more likely an ethylene equivalent) would give the required “trans” stereochemistry across the 2- and 6-positions of the product piperidone (256). In practice, when 257 was reacted with bis-sulfonylethylene 258, this result was realized with about 3:1 facial selectivity, with a mixture of 259 (desired stereochemistry) and 260 (undesired) being obtained. The idea for proceeding from here was to reduce the C—S bonds to establish the equivalency between 258 and ethylene, and to use the enone to introduce the C5 sidechain via a conjugate addition.
OMe
OMe
OMe
262
3. Bu4N OAc 4. LiAlH4
263
H
+
N 264
10%
N 265
45%
HO
Br
Ito, Y.; Nakajo, E.; Nakatsuka, M.; Saegusa, T. "A New Approach to Gephyrotoxin" Tetrahedron Lett. 1983, 24, 2881-2884.
NH2
TMSO N
5 Steps
NMe2
TMS NMe3
OMe
OMe
HO
TMSO N
CsF (6 eq)
N
N
71%
H
CH3CN, 65 oC OMe
high dilution
269
270
OMe
OMe
268
267
266
4.5:1 not separable
1. BBr3, CH2Cl2 2. 5% Rh/Al2O3
Wei, L.-L.; Hsung, R. P.; Skelnicka, H. M.; Gerasyuto, A. I. "A Novel and Highly Stereoselective Intramolecular Formal [3+3] Cycloaddition Reaction of Vinylogous Amides Tethered with α,βUnsaturated Aldehydes: A Formal Synthesis of (+)-Gephyrotoxin" Angew. Chem. Int. Ed. 2001, 40, 1516. O
O
O
EtO
OTBS
HO Kishi Intermediate
1. 4 Steps 48%
271
5:1
NH2 HO
OH
H
O O
2. MnO2 272
N
3. OH
Gephyrotoxin-9
OTBDPS
75% 186
Page 392
TMSO
Ito and Kishi Intermediate
10:56 AM
22% for 9 Steps
Organic Synthesis via Examination of Selected Products
HO
via
b1026
O
N
12/21/2010
Br
OMe
Organic Synthesis via Examination of Selected Natural Products
H
N3
261
OMe
1. TfOH 2. LiEt3BH
5 Steps
b1026_Chapter-09.qxd
392
Pearson, W. H.; Fang, W. "Synthesis of Benzo-Fused 1-Azabicyclo[m.n.0]alkanes via the Schmidt Reaction: A Formal Synthesis of Gephyrotoxin" J. Org. Chem. 2000, 65, 7165-7174.
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 393
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
393
Gephyrotoxin-9 The Pearson group (University of Michigan) took an approach that relied on an N-insertion to prepare the tricyclic nucleus of gephyrotoxin. Thus, indanone 261 was converted to unsaturated azide 262. Treatment of 262 with triflic acid gave a mixture of iminium ions (including 265) that were reduced with lithium triethylborohydride, followed by conversion of the primary bromide to a primary alcohol. This sequence gave 264 as the major product (via 265). The relationship between 264 and intermediates in the Kishi approach should be clear. A different approach to the same tricyclic intermediate was described by the Ito group. The key step in this approach was an intramolecular cycloaddition of 268/269, generated by a novel elimination reaction of ammonium salt 267. Conversion to the Kishi intermediate 186 completed this partial synthesis of gephyrotoxin. The Hsung group (Wisconsin) reported an enantioselective synthesis of Kishi-like intermediate 186 that began with the preparation of 272. Formation of the enamine with cyclohexan-1,3-dione, followed by oxidation of the allylic alcohol and protection of the remaining primary alcohol, gave 273 (Gephyrotoxin-10).
O
Piperidinium acetate (0.1 eq), CHO
H 2, Pd/C
H
EtOAc, 80 C, 1-2 h
O TBAF
H N
186
TBDPSO
TBDPSO
273 60%
HO
Br
OTBS
4 Steps
NHBoc 276
O
Br
277
N H
OTBS
NaI, Et 3N DMF, 110 C
H 2, Pt/C, AcOH
N H 278
OH H
H 279
OTBS N H
98%
Note the relationship of this synthesis to the Kishi approach to gephyrotoxin. OH O Lepadin A
H
N H 280 H
Gephyrotoxin-10
O
Page 394
O
10:56 AM
Pu, X.; Ma, D. "Facile Entry to Decahydroquinoline Alkaloids. Total Syntheses of Lepadins A-E and H" J. Or g. Chem. 2006, 71, 6562-6572.
Organic Synthesis via Examination of Selected Products
28% overall separated
A Recent Application of the Kishi Strategy
O
80%
12/21/2010
275
2
b1026
274
H N
2
Organic Synthesis via Examination of Selected Natural Products
N H
1:1
N
b1026_Chapter-09.qxd
O
OTBDPS
394
O
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 395
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
395
Gephyrotoxin-10 An intramolecular conjugate addition, followed by an enamine “aldoldehydration” reaction provided 274 as a mixture of stereoisomers at C2. Catalytic hydrogenation of the isolated olefin gave 275 and TBAF removal of the silicon protecting group provided 186 after separation from the C2 diastereomer. By now I imagine you recognize that the key bond constructions in the Huang synthesis resemble those in the Kishi approach. This appears again in a synthesis of lepadin A (280) reported in 2006 by the Ma group, which I show here without comment. Why does this approach appear again and again? One cannot say for sure, but I suggest it is that the natural reactivity pattern of the carbonyl group is what attracts researchers to this bond disconnection.
b1026_Chapter-09.qxd
12/21/2010 b1026
396
10:56 AM
Page 396
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
References 1. Daly, J. W.; Spande, T. F.; Garraffo, H. M. “Alkaloids from Amphibian Skin: A Tabulation of Over Eight-Hundred Compounds” J. Nat. Prod. 2005, 68, 1556–1575. Daly, J. W. “Ernest Guenther Award in Chemistry of Natural Products. Amphibian Skin: A Remarkable Source of Biologically Active Arthropod Alkaloids” J. Med. Chem. 2003, 46, 445–452. 2. Majetich, G.; Wheless, K. “Remote Intramolecular Free Radical Functionalizations: An Update” Tetrahedron 1995, 51, 7095–7129. Barton, D. H. R.; Beaton, J. M.; Geller, L. E.; Pechet, M. M “A New Photochemical Reaction” J. Am. Chem. Soc. 1960, 82, 2640–2641. 3. Speckamp, W. N.; Moolenaar, M. J. “New Developments in the Chemistry of N-Acyliminium Ions and Related Intermediates” Tetrahedron 2000, 56, 3817–3856. Speckamp, W. N.; Hiemstra, H. “Intramolecular Reactions of N-Acyliminium Intermediates” Tetrahedron 1985, 41, 4367–4416. 4. Maryanoff, B. E.; Zhang, H-C.; Cohen, J. H.; Turchi, I. J.; Maryanoff, C. A. “Cyclizations of N-Acyliminium Ions” Chem. Rev. 2004, 104, 1431–1628. 5. Stork, G. “Polycyclic Natural Products: Total Synthesis of Lycopodine” Pure Applied Chem. 1968, 17, 383–401. Stork, G.; Kretchmer, R. A.; Schlessinger, R. H. “The Stereospecific Total Synthesis of dl-Lycopodine” J. Am. Chem. Soc. 1968, 90, 1647–1648. 6. Keck, G. E.; Yates, J. B. “Carbon-Carbon Bond Formation via the Reaction of Trialkylallystannanes with Organic Halides” J. Am. Chem. Soc. 1982, 104, 5829–5831. 7. Roth, M.; Dubs, P.; Goetschi, E.; Eschenmoser, A. “Synthetic Methods. 1. Sulfide Contraction via Alkylative Coupling: Method for Preparation of βDicarbonyl Derivatives” Helv. Chim. Acta 1971, 54, 710–734. 8. Corey, E.; J.; Fuchs, P. L. “Synthetic Method for Conversion of Formyl Groups into Ethynyl Groups” Tetrahedron Lett. 1972, 12, 3769–3772. 9. Hoffmann, R. W. “Allylic 1,3-Strain as a Controlling Factor in Stereoselective Transformations” Chem. Rev. 1989, 89, 1841–1860. Johnson, F. “Allylic Strain in Six-Membered Rings” Chem. Rev. 1968, 68, 375–413. 10. Sonogashira, K.; Tohda, Y.; Hagihara, N. “Convenient Synthesis of Acetylenes. Catalytic Substitutions of Acetylenic Hydrogen with Bromo Alkenes, Iodo Arenes and Bromopyridines” Tetrahedron Lett. 1975, 15, 4467–4470. 11. Brown, H. C.; Bhat, K. S.; Randad, R. S. “Chiral Synthesis via Organoboranes. 21. Allyl- and Crotylboration of α-Chiral Aldehydes with Diisopinocampheylboron as the Chiral Auxiliary” J. Org. Chem. 1989, 54, 1570–1576. Roush, W. R.; Grover, P. T. “N,N′-Bis(2,2,2-trifluoroethyl)-N,N’ethylenetartramide: An
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 397
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
12.
13.
14.
15.
16. 17.
18.
19. 20.
397
Improved Chiral Auxiliary for the Asymmetric Allyboration Reaction” J. Org. Chem. 1995, 60, 3806–3813. Oppolzer, W.; Cintas-Moreno, P.; Tamura, O.; Cardinaux, F. “Enantioselective Synthesis of α-N-Alkylamino Acids via Sultam-Directed Enolate Hydroxyamination” Helv. Chim. Acta 1993, 76, 187–196. Oppolzer, W.; Merifield, E. “Synthesis of (-)-Pinidine via Asymmetric, Electrophilic Enolate Hydroxy-amination/Nitrone Reduction” Helv. Chim. Acta 1993, 76, 957–962. Yamakado, Y.; Ishiguro, M.; Ikeda, N.; Yamamoto, H. “Stereoselective Carbonyl-Olefination via Organosilicon Compounds” J. Am. Chem. Soc. 1981, 103, 5568–5570. Crabtree, R. H.; Davis, M. W. “Directing Effects in Homogeneous Hydrogenation with [Ir(cod)(PCy3)(py)]PF6” J. Org. Chem. 1986, 51, 2655–2661. Stork, G.; Kahne, D. E. “Stereocontrol in Homogeneous Catalytic Hydrogenation via Hydroxyl Group Coordination” J. Am. Chem. Soc. 1983, 105, 1072–1073. Crabtree, R. H.; Felkin H.; Fillebeen-Khan, T.; Morris, G. E. “Dihydridoirridium Diolefin Complexes as Intermediates in Homogeneous Hydrogenation” J. Organometallic Chem. 1979, 168, 183–195. McMurry, J. E. “Total Synthesis of Copacamphene” Tetrahedron Lett. 1970, 11, 3731–3734. Thompson, H. W. “Stereochemical Control of Reductions. The Directive of Carbomethoxy vs. Hydroxymethyl Groups in Catalytic Hydrogenation” J. Org. Chem. 1971, 36, 2577–2581. Fujimoto, R.; Kishi, Y. “On the Absolute Configuration of Gephyrotoxin” Tetrahedron Lett. 1981, 22, 4197–4198. Grieco, P. A.; Gilman, S.; Nishizawa, M. “Organoselenium Chemistry. A Facile One-Step Synthesis of Alkyl Aryl Selenides from Alcohols” J. Org. Chem. 1976, 41, 1485–1486. Sharpless, K. B.; Young, M. W. “Olefin Synthesis. Rate Enhancement of the Elimination of Alkyl Aryl Selenoxides by ElectronWithdrawing Substituents” J. Org. Chem. 1975, 40, 947–949. Hart, D. J. “Effect of A(1,3) Strain on the Stereochemical Course of N-Acyliminium Ion Cyclizations” J. Am. Chem. Soc. 1980, 102, 397–398. Hart, D. J.; Tsai, Y-M. “N-Acyliminium Ions: Detection of a Hidden 2-Aza-Cope Rearrangement” Tetrahedron Lett. 1981, 22, 1567–1570. Barton, D. H. R.; McCombie, S. W. “New Method for the Deoxygenation of Secondary Alcohols” J. Chem. Soc., Perkin 1 1975, 1574–1585. Tsai, D. J. S.; Matteson, D. S. “A Stereocontrolled Synthesis of Z and E Terminal dienes from Pinacol E-1-Trimethylsilyl-1-propene-3-boronate” Tetrahedron Lett. 1981, 22, 2751–2752. Also see reference 13 above.
b1026_Chapter-09.qxd
12/21/2010 b1026
398
10:56 AM
Page 398
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
Problems 1. What are the structures of products that would result from addition of N-acyliminium ion 24 to the “other end” of the olefin (formation of a 5-membered ring). (Histrionicotoxin-4) 2. Given that perhydrohistionicotoxin has a 1,3-N,O relationship, propose a Mannich reaction that might be used to contruct the spirocyclic core of 5. What stereochemical issues would have to be addressed in your proposed Mannich reaction? (Histrionicotoxin-4) 3. Here is a practical problem. Perhydrohistrionicotoxin adopts largely conformation 5a in CDCl3. How can one distinguish between 5a and 5e by 1H NMR spectroscopy (in CDCl3)? [Hint: How would you distinguish between cis- and trans-4-tert-butylcyclohexanols by 1H NMR?] (Histrionicotoxin-4) 4. Consider using 2-[2-(1-butyl)cyclohexenyl]ethyl iodide in the intial alkylation of 36. Develop a plan that might convert the resulting product to the epimer of 44 at the position of n-butyl substitution, and not call for the stereochemical corrections seen in Histrionicotoxin-5. Discuss the “big question marks” associated with your plan. (Histrionicotoxin-5) 5. Provide the structure of all isolable intermediates proceeding from 49 → 56. 6. Provide a mechanism for the second step (Eschenmoser sulfide contraction) of the transformation of 57 → 58. (Histrionicotoxin-7) 7. The unsaturated ketone/ester unit of 58 is amide/urethane-like in its behavior. Thus it is not easily hydrolyzed by methoxide. Explain this “unreactive” behavior. (Histrionicotoxin-7) 8. Provide structures of intermediates and a mechanistic explanation for the conversion of aldehyde 59 to TMS-acetylene 60. (Histrionicotoxin-7) 9. Provide the structure of isolable intermediates en route from 61 to 62. Elaborate on the comments in the text regarding allylic strain relative to the “lack of reactivity” of the piperidine in this transformation, and the conversion of 62 → 63. (Histrionicotoxin-7) 10. Suppose the enolate of 73 had reacted with the bromide faster that the vinyloxirane. How might you change tactics to overcome this problem? (Histrionicotoxin-8) 11. Show the structure of the expected product after each step going from 78 → 80. (Histrionicotoxin-9) 12. Supply the structures of intermediates after each step in the conversion of 89 → 90. Provide a “model” that explains the stereochemical course of the hydroxylamination reaction. (Histrionicotoxin-10)
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 399
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
399
13. Provide a mechanistic explanation for the Z-olefin selectivity observed in the conversion of 95 to 96. (Histrionicotoxin-11) 14. Given that a trialkylsilylacetylene was used in the Sonogoshira reaction, provide the structure of intermediates en route from 98 to histrionicotoxin (2). (Histrionicotoxin-11) 15. Provide an estimate of the energy difference between the two chair-chair conformations of pumiliotoxin-C. (Pumiliotoxin-1) 16. Develop a Diels-Alder approach to pumiliotoxin-C that involves construction of the bonds labeled “a” (below) in a Diels-Alder reaction. Discuss problems and potential solutions (where available). (Pumiliotoxin-1) H a a
H
N H
17. Describe tactics for the conversion of 121 to 122. (Pumiliotoxin-2) 18. Explain why the most stable conformation of cis-N-benzoyl2,6-methylpiperidine has the two methyl groups in axial sites. (Pumiliotoxin-2) O
O N Ph
O
Quick, J.; Mondello, C.; Humora, M.; Brennan, T. “Synthesis of 2,6-Diacetonylpiperidine. X-Ray Difraction Analysis of its N-Benzoyl Derivative” J. Org. Chem. 1978, 43, 2705–2708. 19. Explain the regiochemical course of nitrogen insertion in the conversion of 134 to 135 via a Beckman rearrangement. (Pumiliotoxin-4) 20. Provide the products expected after each step of the following reaction sequence. (Pumiliotoxin-4) 1. NaH, PhCH2Br H
2. m-CPBA 3. 48% aq.HBr N
H
H
O
4. H2Cr2O7,H2SO4,acetone 5. LiBr-Li2CO3, DMF, ∆
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 400
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
400
21. Suggest tactics for accomplishing the following transformation. (Pumiliotoxin-4) H
O
22. 23. 24. 25.
H
H
? N Bn
O
Several Steps
N H
H
O
Suggest tactics for converting 136 → 137 → 3. (Pumiliotoxin-4) Provide intermediates en route from 139 → 140. (Pumiliotoxin-4) Outline syntheses of dienes 138 and 141. (Pumiliotoxin-4) Provide a mechanism for the following interesting reaction used in the conversion of 140 to 3. (Pumiliotoxin-4) H
H
H CaO (H2 O), ∆
H O
N
O
H 2 , PtO2 H
N
H
Mechanism ?
N H
26. Provide a mechanism for the following transformation. Propose the most stable conformation of 144c and the subsequent alkylation product 145. (Pumiliotoxin-5) CHO CHO NC H 2N
OH
N
O
144c
KCN
27. Provide a mechanism for the conversion of 147 → 150, and also for the conversion of 150 → 151 (you can ignore stereochemistry). (Pumiliotoxin-6) 28. Propose a synthesis of 153 from malic acid or any other commercially available starting material of your choice. (Pumiliotoxin-6) OH HO2 C
CO 2H
(S)-malic acid
29. Provide tactics for the conversion of 167 → 168. (Pumiliotoxin-8)
b1026_Chapter-09.qxd
12/21/2010 b1026
10:56 AM
Page 401
Organic Synthesis via Examination of Selected Products
Alkaloids from “Dart-Poison” Frogs
401
30. Provide a rationale for the stereochemical course of the enolate protonation in the conversion of 168 → 170. (Pumiliotoxin-8) 31. Suggest tactics that would convert 175 → 176. (Pumiliotoxin-8) 32. Explain the following stereoselectivity trend. (Gephyrotoxin-1) HO
HO
HO H 2 , Pd/C solvent
+ H
H
Solvent
33. 34. 35.
36.
hexane
61
39
ethanol
6
94
Thompson, H. W.; McPherson, E.; Lences, B. L. “Stereochemical Control of Reductions. 5. Effects of Electron Density and Solvent on Group Haptophilicity” J. Org. Chem. 1976, 41, 2903–2906. Provide a detailed mechanism for the Li-NH3 reduction of 192. Explain the stereochemical course of the reaction. (Gephyrotoxin-2) Provide a mechanism, including a stereochemical explanation, for the conversion of 195 (R = CHO) → 197. (Gephyrotoxin-2) The reaction of 204 with succinimide, triphenylphosphine, and diethyl azodicarboxylate (EtO2CN=NCO2Et) gave (after chromatography) a 20% yield of the diastereomer of 204 with inverted alcohol stereochemistry. Provide a mechanistic explanation for this observation. (Gephyrotoxin-3) Provide a stereochemical analysis for the conversion of 210 → 211. Predict the stereochemical course of the following reaction. (Gephyrotoxin-3) OH HCO2 H N
OHCO
H N
O
Stereochemistry ?
O
37. Suggest how NMR was used to determine the conformational preference of 206. (Gephyrotoxin-3) 38. Provide the structure of intermediates en route from 206 → 207. (Gephyrotoxin-3) 39. Provide mechanistic explanations for the observed stereocontrol in the conversions of 218 → 220 + 223. (Gephyrotoxin-4)
b1026_Chapter-09.qxd
12/21/2010 b1026
402
10:56 AM
Page 402
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
40. Provide the structure of the other half-chair conformation available to cis-perhydroquinolinium ion 225. Predict the product that would be obtained by a “stereoelectronically controlled reduction” from this conformation. (Gephyrotoxin-5) 41. What is the role of the trifluoroacetic acid in the conversion of 232 → 233? (Gephyrotoxin-5) 42. Provide a reaction sequence that might convert 250 → 251. (Gephyrotoxin-8) 43. Provide a mechanism for the conversion of 262 → 265. Provide the structure of the minor iminium ion obtained from this reaction. Rationalize the stereochemical course of the reduction of 265 to ultimately provide 264. (Gephyrotoxin-9) 44. Outline a synthesis of 272 beginning with preparation of 271 by construction of the bond indicated in red. (Gephyrotoxin-9)
OH Morphine
O
H
From HART/CRAINE/HART/HADAD, Organic Chemistry, 12E Copyright Brooks/Cole, a part of Cengage Learning, Inc. Reproduced by permission. www.cengage.com/permissions.
Many compounds similar to various parts of the morphine structure have been synthesized and tested for their analgesic properties. For example, demerol is an effective analgesic with a relatively simple structure compared to that of morphine. Notice that it still retains the piperidine ring present in morphine. Methadone, which retains the nitrogen of morphine but is no longer heterocyclic, was synthesized and used as an analgesic by Germans during World War II, when natural sources of morphine were in short supply. Later it was used in substitution therapy for heroin addiction, but it, too, is addictive. The search for a perfect analgesic continues. With this background, it should be no surprise that morphine has been a popular target for synthesis, with a new synthesis appearing as late as the 50th anniversary of the 1st synthesis. We will spend the next few pages comparing and contrasting several of these syntheses. MeO
O
O H AcO 2 Heroin
N CH3
H
CH3
HO 3
O
O
N
N O 4 Demerol
Codeine
NMe2
CH3 5 Methadone
For a thorough review of approaches to morphine see: Zezula, J.; Hudlicky, T. "Recent Progress in the Synthesis of Morphine Alkaloids" Synlett 2005, 388-405. Novak, B. H.; Hudlicky, T.; Reed, J. W.; Mulzer, J.; Trauner, D. "Morphine Synthesis and Biosynthesis - An Update" Current Organic Chemistry 2000, 4, 343-362. Also see Magnus, P.; Sane, N.; Fauber, B. P.; Lynch, V. J. Am. Chem. Soc. 2009, 131, 16045-16047.
Morphine-1
Page 404
The first attempts to find a substance with morphine's benefits but without its side effects involved minor modification of its structure. Acetylation with acetic anhydride gave its diacetyl derivative (heroin), which is a good analgesic with less of a respiratory depressant effect than morphine. But heroin is severely addictive, and its abuse is a serious problem. Methylation of the phenol of morphine gave codeine, which is useful as an anticough (antitussive) agent. Unfortunately, it is less that one-tenth as effective as morphine as an analgesic.
10:57 AM
Pain is a major problem in medicine and relief of pain has long been a medical goal. Morphine is an analgesic, a substance that relieves pain without causing unconsciousness. It was used for large-scale relief of pain from battle wounds during the American Civil War (largely as a consequence of the invention, at about that time, of the hypodermic syringe). But morphine has serious side effects. It is addictive and also can cause nausea, a decrease in blood pressure, and a depressed breathing rate that can be fatal to the very young or the severely debilitated.
Organic Synthesis via Examination of Selected Products
Morphine (named after Morpheus, the Greek god of dreams) is the major alkaloid present in opium. Opium is the dried sap of the unripe seed capsule of the poppy Papaver somniferum. High grade opium contains 9-14% of this alkaloid. Its medical properties have been known since ancient times. Morphine was not isolated in pure form until 1805. Its correct structure was not established until 1925, and it was not synthesized in the laboratory until 1952.
12/21/2010
H3C N
b1026
CH3
Organic Synthesis via Examination of Selected Natural Products
H
O OH
1
N
HO
AcO
b1026_Chapter-10.qxd
404
Morphine HO
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 405
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
405
Morphine-1 Morphine (1) (named after Morpheus, the Greek god of dreams) is the major alkaloid present in opium. Opium is the dried sap of the unripe seed capsule of the poppy Papaver somniferum. High grade opium contains 9–14% of this alkaloid. Its medical properties have been known since ancient times. Morphine was not isolated in pure form until 1805. Its correct structure was not established until 1925, and it was not synthesized in the laboratory until 1952. Pain is a major problem in medicine and relief of pain has long been a medical goal. Morphine is an analgesic, a substance that relieves pain without causing unconsciousness. It was used for large-scale relief of pain from battle wounds during the American Civil War (largely as a consequence of the invention, at about that time, of the hypodermic syringe). But morphine has serious side effects. It is addictive and also can cause nausea, a decrease in blood pressure, and a depressed breathing rate that can be fatal to the very young or the severely debilitated. The first attempts to find a substance with morphine’s benefits but without its side effects involved minor modification of its structure. Acetylation with acetic anydride gave its diacetyl derivative [heroin (2)], which is a good analgesic with less of a respiratory depressant effect than morphine. But heroin is severely addictive, and its abuse is a serious problem. Methylation of the phenol of morphine gave codeine (3), which is useful as an anticough (antitussive) agent. Unfortunately, it is less that one-tenth as effective as morphine as an analgesic. Many compounds similar to various parts of the morphine structure have been synthesized and tested for their analgesic properties. For example, demerol (4) is an effective analgesic with a relatively simple structure compared to that of morphine. Notice that it still retains the piperidine ring present in morphine. Methadone (5), which retains the nitrogen of morphine but is no longer heterocyclic, was synthesized and used as an analgesic by Germans during World War II, when natural sources of morphine were in short supply. Later it was used in substitution therapy for heroin addiction, but it, too, is addictive. I remember a line from the Woody Allen film “Take the Money and Run” where one of the characters says “I used to be a heroin addict. Now I’m a methadone addict”.
b1026_Chapter-10.qxd
406
First Synthesis of Morphine Gates, M.; Newhall, W. F. "Synthesis of Ring Systems Related to Morphine. I. 9,10-Dioxo-13-cyanomethyl-5,8,9,10,13,14-hexahydrophenanthrene" J. Am. Chem. Soc. 1948, 70, 2261. Gates, M. "Synthesis of Ring Systems Related to Morphine. III. 4,5-Dimethoxy-4-cyanomethyl-1,2-naphthoquinone and its Condensation with Dienes" J. Am. Chem. Soc. 1950, 72, 228. Gates, M.; Tschudi, G. "The Synthesis of Morphine" J. Am. Chem. Soc. 1955, 78, 1380.
CH3
O 8
7
6
1
O
MeO NC
1. H2 , Pd/C OH
AcOH
10
HO
9
NaNO2
OBz
AcOH
11
2. FeCl3
HO
O
12 O
NO
Bz = benzoyl 71%
88%
85%
SO2 MeOH-H2 O
O O
OH
1. NaNO 2, AcOH
16
MeO OMe
3. FeCl3
OBz
NaOH
2. H2 , Pd/C, AcOH MeO
MeO OMe
OBz
80%
80%
K2CO3 OMe
15
OBz
Me2 SO4 HO OH
14
82%
13
91%
NCCH2 CO2Et NaOH, K3 Fe(CN)6 O
O O MeO
17 MeO NC
MeO
8 MeO
CO 2Et 82%
O O
KOH
CN OH
1,3-butadiene AcOH, 85 °C 48 h
CN 97%
MeO OMe 50%
Morphine-2
H 2 (27 atm), Cu-chromite 7
130 °C
Page 406
dioxane pyridine
HO
OBz
PhCOCl
10:57 AM
H
CH3
HO
CN OH
N
Organic Synthesis via Examination of Selected Products
H
N
MeO
O
MeO
12/21/2010
MeO
O
MeO
b1026
reductive amination
Organic Synthesis via Examination of Selected Natural Products
MeO
HO
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 407
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
407
Morphine-2 The search for a perfect analgesic continues. With this background, it should be no surprise that morphine has been a popular target for synthesis, with a new synthesis appearing as late as the 50th anniversary of the first synthesis. We will spend the next few pages comparing and contrasting several of these syntheses. The first synthesis of morphine was reported by Marshall Gates (University of Rochester). It was a remarkable achievement. The plan is outlined briefly here. The idea was that if compound 6 could be obtained, it would be possible to move forward to morphine. In parallel studies it was shown that morphine could be degraded to 6, thus providing a “relay point” for any synthesis effort, since morphine was in abundant supply and synthetic 6 would no doubt be difficult to obtain in large quantities. Structure 6 could clearly be obtained through a cycloaddition of an appropriate dienophile with 1,3butadiene. The question is which dienophile? The choice ulitimately became o-naphthoquinone 8. A cycloaddition between this electron-deficient dienophile and 1,3-butadiene would provide 7, which one might convert to 6 via a reductive amination and additional oxidation state adjustments. The dienophile synthesis began with 2,6-dihydroxynaphthalene (9). Protection of one of the phenolic groups as a benzoate was followed by o-nitrosation to provide 11. Reduction of the nitroso group to a primary amine was followed by oxidation of the o-aminophenol to o-naphthoquinone 12. The quinone was reduced to the hydroquinone (13) and the hydroxyl groups were protected. The benzoate was hydrolyzed and the resulting phenol (15) was subjected to the same oxidation sequence to provide o-naphthoquinone 16. Treatment of this “enone” with ethyl cyanoacetate and base under oxidative conditions gave 17. Ester hydrolysis followed by decarboxylation gave the desired dienophile 8. The key cycloaddition was accomplished in the presence of acetic acid to give the desired cycloadduct 7.
MeO
MeO
MeO
ethylene glycol, ∆
N
16 Steps
MeO
2. LiAlH4 , Et2O
N
H
H 6
H 19
HO
O
H
CH3
N CH3
O
HO 1 Morphine
MeO
CH3
21
NH2 NH 2
MeO
MeO
MeO
HO
KOtBu, t -BuOH HO
KOH N
H
CH3
N
H
CH3 HO
HO 6
N
HO
20
18% aqueous H 2SO 4
H
H
β-Thebainone (available from nature)
MeO MeO
2. PhMe3 N OH MeI (76%)
MeO
28%
21
benzophenone
N
H
CH3
2. 2,4-DNP
N CH3
O 22
Fortuitous demethylation observed during WK. Some cleavage of other methoxy also noted.
Morphine-3
1. Br2 , AcOH
89%
23
Basic oxidation (Oppenhauer) because of nitrogen?
Page 408
H
N
10:57 AM
MeO 1. H2 , Pt2O (60%)
Organic Synthesis via Examination of Selected Products
MeO
2. collidine,
b1026
1. TsCl, pyridine
50%
How?
CH3
Resolved with O,O-dibenzoyl tartrate
Degradation of Morphine for use in Relay Synthesis
HO
N
Organic Synthesis via Examination of Selected Natural Products
H 18
50%
1. NaH, MeI, toluene
O
12/21/2010
H
NH 2NH2 , KOH
b1026_Chapter-10.qxd
MeO O O
408
MeO
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 409
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
409
Morphine-3 The conversion of 7 to keto-amide 18 required that the cyano group be hydrolyzed to a primary amide, and that reductive ring closure occur from nitrogen to the ketone generated upon protonation of the enol group present in 7. This was accomplished using copper chromite as a hydrogenation catalyst with ethanol as the solvent. This transformation is remarkable for its selectivity. Notice that the olefin is not disturbed. The intimate details of the mechanism of this process are not clear and I welcome your speculation.* Continuing with the synthesis, a Wolf-Kishner reduction provided amide 19. The N-methyl group was installed, followed by reduction of the lactam to provide key intermediate 6. This material was available from morphine (1) and was used as a relay point for this synthesis. Hydration of 6 using aqueous sulfuric acid gave a low yield of 21. This is an example of sophomore organic chemistry being applied to a complex molecule. Indeed Gates used what was available in the day, but this is clearly a weak point in the synthesis. An unexpected demethylation, observed as a minor reaction during the conversion of 18 → 19, was applied to 21 to selectively remove one phenolic methyl group and provide 22. Oxidation of 22 under basic conditions gave ketone 23. Bromination of 23, followed by 2,4-DNP promoted dehydrohalogenation, gave hydrazone 24 (see Morphine-4). Epimerization accompanied this transformation to establish the stereochemistry required by morphine γ to the enone.
*I recall a story from graduate school days (told by Gordon Rivers): The postdoc presents a remarkable result at group meeting. A graduate student in the audience asks “how does that work?”. The postdoc replies “very well!”. Reminds me of one of my favorite radio shows: “Whattaya Know? Not Much!”.
2. 2,4-DNP
N
H
CH3
24
tribromination
25
26
O
A remarkable synthesis given tools available at the time. Efforts to make better use of the alkene were not productive.
H
(Henry Rapoport)
N
O H
CH3
CH3
HO 1
Codeine
morphine 3
Biosynthesis of Morphine 28 HO
CO2 H HO decarboxylation NH 2
HO
oxidation
HO
NH 2 CHO
(f rom tyrosine) 27
PictetSpengler
Methylation HO HO
HO HO
29
MeO
HO
HO
N
30 laudanosine
Morphine-4
H
N
HO HO
Me
31
MeO reticuline
Page 410
HO
N
10:57 AM
HO pyridinium hydrochloride
Organic Synthesis via Examination of Selected Products
MeO
26%
CH3
b1026
N NHAr
O 41%
2. LiAlH4 , ∆ (60%)
N
12/21/2010
H
CH3
Organic Synthesis via Examination of Selected Natural Products
N NHAr
1. HCl, acetone (27%) O
2. H2 , Pt
N
Br
1. Br2 , AcOH
HO
b1026_Chapter-10.qxd
1. HCl, acetone
HO H
MeO
MeO
Br
410
MeO
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 411
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
411
Morphine-4 Hydrolysis of the hydrazone followed by catalytic hydrogenation of the olefin and hydrogenolysis of the aryl bromide gave keto-phenol 25. Treatment of 25 with bromine in acetic acid, followed by reaction of the intermediate tribromide with 2,4-dinitrophenylhydrazine, gave 26. Hydrolysis of the hydrazone was followed by reduction of the enone to provide codeine (3). Demethylation of codeine provided morphine (1). This is a remarkable synthesis given tools that were available at the time. It has been established that the biosynthesis of morphine follows a path from tyrosine derivative 27 to laudanosine (30) via a Pictet-Spengler type cyclocondensation of phenethylamine 28 and aldehyde 29. Dimethylation of 30 then provides reticuline (31).
HO
MeO Reduction
HO
Coupling
N
O N
CH 3 MeO
31
MeO
32
O
CH 3 MeO
33
34
OH thebaine
salutaridinol
hydrolysis
MeO
MeO Reduction
Demethylation O
O
HO
O
N
H
CH 3 1
morphine
HO
N
H
CH 3 O
3 codeine
N CH 3 35
codeinone
The First Biomimetic Synthesis of Morphine: Isotope Dilution Method MeO Isotope Dilution Method very low yield A
B
Suppose the yield is so low there is no chance of isolating B and determining yield.
1. Prepare radiolabeled A 2. Run reaction 3. Add cold B 4. Isolate B 5. Purify to constant radioactivity 6. Determine % radiolabel in B 7. Calculate yield of B
HO N CH 3 MeO
Tritiate aromatic ring with acid and tritiated water. Check for position of substitution using heavy water (all positions of aromatic ring deuterated).
31 OH reticuline
Barton, D. H. R.; Bhakuni, D. S.; James, R.; Kirby, G. W. "Phenol Oxidation and Biosynthesis. Part XII. Stereochemical Studies Related to the Biosynthesis of Morphine" J. Chem. Soc. ( C) 1967, 128.
Morphine-5
Page 412
H
10:57 AM
HO
Organic Synthesis via Examination of Selected Products
Nature’s route to the morphinane alkaloids is very efficient.There have been many attempts to mimic this in the laboratory. We will look at a few and look at the problems associated with translating nature’s plan to reality in the hands of man.
b1026
salutaridine
reticuline
N CH 3
12/21/2010
OH
SN2’
HO
Organic Synthesis via Examination of Selected Natural Products
MeO
MeO
N
CH 3
b1026_Chapter-10.qxd
MeO Oxidative Phenolic
412
MeO
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 413
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
413
Morphine-5 The key step in the biosynthesis of morphine involves the “oxidative phenolic coupling” of reticuline (31) to salutaridine (32). This step can be viewed mechanistically as (1) oxidation of the two aromatic rings to phenoxy radicals followed by an intramolecular radical-radical coupling or (2) oxidation of one ring to a radical cation or cation, followed by an intramolecular electrophilic aromatic reaction. This process is very important in the biosynthesis of a number of natural products, and is a process that nature has used to “crosslink” peptides containing aromatic residues. The biosynthesis of morphine continues with reduction of salutaridine (32) to salutaridinol (33) followed by an intramolecular SN2′ reaction to give thebaine (34). Dienol ether hydrolysis to codeinone (35), reduction of the ketone to codeine (3) and O-demethylation completes the biosynthesis of morphine (1). Nature’s route to the morphinane alkaloids is very efficient. There have been many attempts to mimic this in the laboratory. We will look at a few and look at problems associated with translating nature’s plan to reality in the hands of chemists. The first biomimetic synthesis of morphine was described by Sir Derek Barton (Imperial College in London). Barton showed that oxidation of reticuline (31) with potassium ferricyanide gave salutaridine (32) in 0.015% yield. Since salutaridine had been converted to morphine, this constituted a synthesis of the target alkaloid. How does one isolate a product in 0.015% yield? Not easily. In this case, the “isotope dilution” method was used to establish yield. Reticuline was prepared in radioactive “hot” form (tritiation in the aromatic ring). The oxidation reaction was run and a known amount of “cold” salutaridine was added to the reaction mixture. The salutaridine was then isolated and purified to a constant level of radioactivity. The amount of “hot” salutaridine derived from reticuline was calculated based on the radioactivity that had been incorporated, and the percentage yield of salutaridine (from reticuline) was thus determined. The details are given on Morphine-6. The bottom line is that this demonstrated the feasability of this route to morphine, but it did not provide a practical route to the natural product.
K3Fe(CN)6
HO p'
aqueous NaHCO3 CHCl3
N 0'
CH3 MeO 32
O
31
MeO
MeO
p
morphinane
0'
p
36
0'
CH3
aporphines
CH3 MeO
N
HO
CH3
0
37
MeO
O
38
MeO OH
ortho-ortho'
para-para'
para-ortho'
A number of other chemical oxidizing agents have been applied to this transformation (MnO2-silica gel, AgCO3-Celite, VOCl3) using either the tertiary amine or reticuline derivatives in which the N-methyl group is replaced with a carbamate. None of these deal adequately with the aforementioned problems. Electrochemical oxidations sometimes provide reasonable yields of morphinanes, but the para-para' isomers dominate, possibly for steric reasons. Here is one attempt to get around the regiochemical problem. MeO
MeO NaNO2
BnO H2N
H2SO4
N CH3
MeO OMe
HO
39
N CH3
AcOH MeO O
32
1.1%
Morphine-6
Kametani, T.; Ihara, M.; Fukumoto, K.; Yagi, H. “Studies on the Syntheses of Heterocyclic Compounds. Part CCC. Syntheses of Salutaridine, Sinoacutine, and Thebaine. Formal Total Syntheses of Morphine and Simomenine” J. Chem. Soc. (C) 1969, 2030-2033.
Page 414
N
p'
N
HO HO
10:57 AM
OH MeO
Organic Synthesis via Examination of Selected Products
There are problems associated with conducting this reaction in the laboratory: (1) ortho-para' coupling vs para-ortho' coupling or ortho-ortho' coupling (both of which give aporphines or para-para' coupling (which gives a less hindered morphinane) (2) oxidation regiochemistry and oxidation of the amine.
b1026
salutaridine reticuline This paper also describes results of significance in determining stereochemical aspects of the conversion of salutaridine to morphine (stereochemical course of SN2')
12/21/2010
OH
N
p'
Organic Synthesis via Examination of Selected Natural Products
MeO
CH3
1. Use 52 mg of hot racemic reticuline 2. Run reaction 3. Add 50 mg of cold (+)-salutaridine 4. Isolate 45 mg of (+)-salutaridine after crystallization to constant activity 5. Yield = 0.015% (0.03% of racemate)
0
HO
b1026_Chapter-10.qxd
MeO
p 0
414
MeO
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 415
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
415
Morphine-6 What are the problems associated with this reaction in the laboratory? The substrate could undergo a variety of oxidative-phenolic couplings, each from a different conformation (or from several different conformations). If you examine the structure of reticuline (31) in the upper left-hand corner of the page, para′-ortho coupling provides salutaridine (32). However, para-para′ coupling would provide 36, ortho-ortho′ coupling would provide 37 and para-ortho′ coupling would provide 38 (the latter two compounds belong to a family of alkaloids called aporphines). The site of oxidation (which aromatic ring is oxidized) can also influence coupling selectivities. Finally, oxidation of the tertiary amine can cause problems, although this is solved by N-acylation. A number of chemical oxidizing agents have been applied to this transformation, using either the tertiary amine or reticuline derivatives in which the N-methyl group is replaced with a carbamate. None of these adequately deal with the aforementioned problems. Electrochemical oxidations sometimes provide reasonable yields of morphinanes, but the para-para′ isomers dominate, possibly for steric reasons. One indirect solution to the site-of-oxidation problem was addressed by Kametani (Japan). His group reported the conversion of 39 to 32 in 1.1% yield (almost a 100-fold improvement over the Barton conditions), by performing the oxidation “site-selectively” using aryl diazonium chemistry, but this was still quite a low yield.
b1026_Chapter-10.qxd
416
Oxidative Phenolic Coupling Route to Morphinanes
HO MeO
K2CO3
Tl(OCOCF3 )3
R
+
HO N
40
R
41
MeO
42
MeO
R = COCF3 (7%) R = CO 2Et
R = COCF3 (11%) R = CO 2Et (26%) K2CO3 , MeOH
MeO
R = COCF3
MeO
MeO HO Morphine
HO
1N HCl, rt, 1 h
O
N
N
1:1
CH 3
1. ClCO2 Et, Et 3 N
MeO
MeO 34 Thebaine
Morphine-7
CH 3
2. LiAlH4
N H MeO
OH
O
44
43
82%
86%
Page 416
O
OH
R = COCF3 R = CO 2Et
31
N
HO
CH2 Cl2 -78 oC to -20 o C
HO
HO
R
10:57 AM
(CF3 CO)2 O
12/21/2010
H
Organic Synthesis via Examination of Selected Products
N
HO MeO
MeO
MeO N
b1026
MeO
MeO
Organic Synthesis via Examination of Selected Natural Products
Schwartz, M. A.; Mami, I. S. "A Biogenetically Patterned Synthesis of the Morphine Alkaloids" J . Am. Chem. Soc. 1975, 97, 1239-1240.
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 417
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
417
Morphine-7 Here is one of the better attempts to accomplish oxidative phenolic coupling in the lab. Schwartz (Florida State University) reported that oxidation of carbamates of type 40 derived from reticuline (31) gave morphinanes of type 42 in 11–26% yield upon oxidation with thallium trifluoroacetate. Aporphines of type 41 were also obtained in lower yields. Whereas the carbamate forces the “benzylic group” to be pseudo-axial on the tetrahydroisoquinoline core (a requirement for morphinane formation), the reason for regioselectivity is not clear to this author. Nonetheless this provides access to morphine via 42 (R = CO2Et). Reduction of the carbamate to a methyl group was accompanied by conversion of the enone to allylic alcohol 44. Treatment of 44 with acid provided thebaine (34), and the transformation of thebaine to morphine is known (see Morphine-5).
b1026_Chapter-10.qxd
418
Alternatives to Oxidative Phenolic Coupling 1: The Fuchs Approach
45
X
X
46
X
O SO2 Ph
H HO
47
MeO
MeO OH 48
40% overall
(COCl)2
Br
OH
OH 49
MeO
Br
TBA-bisulfate PhH, 40 h, 60 oC
1. TBDMSOTf Et3 N, CH 2Cl2 2. m-CPBA, CH 2Cl2
OH
TBSO
ArOH, Bu3 P SO 2Ph 55
SO 2Ph 52 90%
82%
Br
SO 2Ph 95%
Cl 51
O HF, CH3 CN
56
CHCl3 ∆
Br
O HO
PhSO2 Na
50
Br MeO
O
O
O Br
1
TBSO
DEAD
DEAD = EtO2CN=NCO2Et
80%
Morphine-8
NaBH4 CeCl3
O TBSO
SO 2Ph MeOH-CH 2 Cl2 54 95%
SO 2Ph 53 62%
Page 418
Morphine
Z = SO2 Ph
CHO 6 Steps
NMe
10:57 AM
Z
Z
Br
12/21/2010
Br
Organic Synthesis via Examination of Selected Products
O
O
O
HO
b1026
MeO
MeO
MeO
Organic Synthesis via Examination of Selected Natural Products
The following approaches revolve around construction of same key bond formed in the biomimetic approaches. They depart from the biomimetic approach, however by not "insisting" that a biaryl coupling be used to construct that bond. This "thought process" leads to a number of interesting approaches. We will examine three such approaches that involve carbanions, free radicals, and organopalladium species as key reactive intermediates for constructing this bond. Although it will not be covered here, an approach that involves cationic intermediates has also been described [T et rahedr on Let t. 1967, 4055]. We w ill start with an approach developed by the Fuchs group that revolves around vinyl sulf one addition chemistry. (Toth, J. E.; Hamann, P. R.; Fuchs, P. L. "Studies Culminating in the Total Synthesis of (dl)-Morphine" J. Org. Chem. 1988, 53, 4694-4708).
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 419
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
419
Morphine-8 Now we will turn to approaches that revolve around construction of the same key bond formed in the biomimetic approaches. They depart from the biomimetic approach, however, by not “insisting” that a biaryl coupling be used to construct that bond. This “thought process” leads to a number of interesting approaches. We will examine three such approaches that involve carbanions, free radicals, and organopalladium species as key reactive intermediates for constructing this bond. Although it will not be covered here, an approach that involves cationic intermediates has also been described.1 We will start with an approach developed by the Fuchs group (Purdue University) that revolves around vinyl sulfone addition chemistry. Fuchs imagined that a carbanion of type 45 might undergo an intramolecular conjugate addition to provide 46. A subsequent intramolecular alkylation would give 47, which then might be converted to morphine in a sequence of steps that would require degradation of the allyl sidechain and construction of the C–N bond required to complete the synthesis of morphine (1). This plan nicely addresses the regiochemical problems posed by the oxidative phenolic coupling approach through use of intramolecular reactions. The requirements for “Z” were that it “supports” the proposed cyclization-alkylation and that it be converted to an “H”. As an application of vinyl sulfone chemistry developed over a period of years in the Fuchs group, a phenylsulfonyl group was chosen to serve as “Z”. Thus, the first task was the preparation of a precursor of 57 (Morphine-9), a precursor of 45 where an aryl bromide was to be the source of the carbanion, and a secondary alcohol was to play the role of =X. The synthesis began with the preparation of dibromide 49 from isovanillin (48). This bromide was to be coupled to allyic alcohol 54 using an SN2 reaction. Alcohol 54 was prepared from cyclohexan-1,3-dione derivative 50 as follows. Treatment of this vinylogous acid with oxalyl chloride provide vinylogous acid chloride 51. A vinylogous nucleophilic acyl substitution reaction, using sodium benzenesulfinate as the nucleophile, provided 52. The adjacent hydroxyl group was introduced by reaction of an intermediate silyl enol ether with m-chloroperoxybenzoic acid. Reduction of the ketone from the least hindered face gave 54. A Mitsunobu reaction was used to couple 54 with phenol 49 to provide 55 in excellent yield.2 Removal of the TBS protecting group gave 56.
Br O
2. i-Bu2AlH
HO
MeO
Br 1. n-BuLi (2.2 eq)
Br O
MeO
Br
HO
+
59
90%
10%
60%
NMeCO2 R
O
7
62
KOtBu
NMeCO2 R
THF, 3 h 25 o C
MeO
O
63
NMeCO2 R O
CHCl3-H2 O
NMe 2. NaHCO3 , H 2O CHCl3 (60%)
8
MeO
DDQ, TsOH
O SO 2Ph
MeO 60
61
40% overall
R = CH 2CH 2SiMe3
HCl, Et 2 O, CH2 Cl2 ; NaOH (95%)
MeO
MeO NaBH4 , MeOH (95%)
O H
HO
O
NMe
O
H
NMe
HO 35 (codeinone)
morphine
BBr3, CHCl3 O
H
NMe
HO 3 (codeine)
1 50% from enone mixture
Morphine-9
Page 420
O
O
MeO
MeO 1. CF3 CO2H, 5 min o C (90%) 25
10:57 AM
MeO
54%
Organic Synthesis via Examination of Selected Products
6 Steps
12/21/2010
58
b1026
56
PhO2 S
SO 2Ph
57
Organic Synthesis via Examination of Selected Natural Products
SO 2Ph
SO 2Ph
O OH
O
2. NH 4Cl, H 2O
HO
OH
b1026_Chapter-10.qxd
MeO
1. CrO3 , H 2SO4 acetone
420
Br
Br
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 421
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
421
Morphine-9 An oxidation-reduction sequence adjusted stereochemistry to provide key intermediate 57. Treatment of 57 with an excess of n-BuLi gave the desired addition-alkylation product 59 in an excellent yield (45% or 60% if one accounts for recovered 57) along with a small amount of diene 58. This minor product presumably is the result of γ-deprotonation of the vinyl sulfone followed by α-protonation of the resulting carbanion. In reality it took a lot of substrate adjustment to finally achieve the desired annulation reaction. For example, 56 has all of the structural elements that would be needed to proceed with the plan, so what happened? Treatment of 56 with n-BuLi gave none of the desired cyclization product. Evidence indicated that γ- and γ ′-deprotonation of 56 was one cause of problems with this substrate. In retrospect (or most likely recognized ahead of time by Fuchs), the proposed cyclization required a unique set of rates to succeed: (1) metal-halogenation exchange (ArBr to ArLi) had to be faster than sulfone deprotonation and (2) cyclization had to be faster than intramolecular and/or intermolecular deprotonation of the sulfone by the resulting dianion (ArLi and ROLi). One can imagine other complications as well. For example, carbanion 46 could undergo an intramolecular dehydrohalogenation. Regardless, the “search” part of research brought this part of the synthesis to a successful conclusion. Moving forward from 59, six steps were required to convert this compound to 60. Vicinal dihydroxylation of the olefin was followed by oxidative cleavage of the intermediate diol using lead tetraacetate. Reductive amination of the resulting aldehyde with methylamine, followed by acylation of the intermediate secondary amine gave the desired carbamate. Swern oxidation of the secondary alcohol,3 followed by enol ether formation gave 60. Elimination of p-toluenesulfinic acid from 60 provided 61. Oxidation of this dienol ether to dienone 62 was followed by release of the secondary amine, followed by a conjugate addition reaction to establish the critical C–N bond. The remainder of the synthesis followed known chemistry. The mixture of enones 63 was converted to codeinone (35), codeine (3) and then morphine (1).
b1026_Chapter-10.qxd
422
Alternatives to Oxidative Phenolic Coupling 2: The Parker Approach Parker, K. A.; Fokas, D. "Convergent Synthesis of dl-Dihydroisocodeine in 11 Steps by the Tandem Radical Cyclization Strategy. A Formal Total Synthesis of dl-Morphine" J. Am. Chem. Soc. 1992, 114, 9688-9689.
NMeTs
NMeTs O
O
O
NMe
65 addition-addition-elimination
64
NH2
Li, NH3
TsCl
t-BuOH
Et3N, THF
MeO
68
66
NHTs
NHTs aq. HCl
acetone
72 96%
81%
97%
Directed Lewis Acid promoted reaction
NMeTs
ArOH, Bu3P, DEAD THF
HO TBSO 76 82%
Hunig's base -78 oC
HO
PhH 75 85%
MeOH
NMeTs
m-CPBA HO
HO 74 92%
J. Am. Chem. Soc. 1981, 103, 462 (Sharpless)
Morphine-10
NaBH4, CeCl3
O
Ti(O-i-Pr)4
HO
Luche Reduction
NMeTs
NMeTs
TBDMSOTf
O
71
70
69
NMeTs
K2CO3, MeI O
MeO
SPh
67
73 97%
Page 422
MeO
RO
RO
NH2
HO Br
RO 1
+
10:57 AM
HO
HO
SPh
H
Organic Synthesis via Examination of Selected Products
H
MeO
NMeTs
12/21/2010
MeO
b1026
MeO
Organic Synthesis via Examination of Selected Natural Products
HO
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 423
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
423
Morphine-10 Here is another alternative to oxidative phenolic coupling. Kathy Parker and her group (Brown University) described a synthesis of morphine that focuses on the same key bonds as seen in the Fuchs synthesis, but using totally different chemistry. The Parker plan was, once again, to construct the key C–N bond late in the synthesis from an intermediate of type 64. Parker also focused on initial construction of the same C–C bonds, but this time using a free radical addition-addition-elimination sequence beginning with a radical of type 65. The precursor of this radical (see 78 on Morphine-11) was to be an aryl bromide. This bromide was to be assembled from diol 66 and aryl bromide 67, structures that are similar to the pieces used in the Fuchs synthesis (49 and 54). The synthesis of 66 began with m-methoxyphenethylamine (68). Birch reduction of 68 gave 69, which was converted to sulfonamide 70. Enol ether hydrolysis was accompanied by conjugation of the olefin to provide enone 71. Alkylation of the nitrogen gave 72 and reduction of the ketone using the Luche conditions gave allylic alcohol 73.4 Hydroxyl-directed epoxidation of 73 gave 74, and a titanium-mediated opening of the epoxide, directed by the alcohol, gave diol 75.5 Protection of the homoallylic alcohol gave 76 (a derivative of 66). A Mitsunobu coupling of 76 with phenol 67 (whose preparation will not be described here) gave 77 (Morphine-11).
b1026_Chapter-10.qxd
424
Li, NH 3, t-BuOH
PhH, 130 C
SPh
O
+
o
78
72
79 ( 64 where R = H)
11%
35%
source = H atom transf er-fragmentation
(COCl)2 , DMSO; ROH, Et3 N
O H HO
NMe
O
morphine and other morphinanes
(Swern Oxidation)
80
H O
NMe
81 83%
85% An unexpected but pleasant surprise
dihydrocodeinone
Morphine-11
Page 424
MeO
MeO
10:57 AM
77
R = TBS
HO
Organic Synthesis via Examination of Selected Products
H O
RO
R=H
NMeTs
12/21/2010
n-Bu3 SnH [35 mM], AIBN
b1026
NMeTs O Br
10% HF CH3 CN
MeO
NMeTs
Organic Synthesis via Examination of Selected Natural Products
MeO
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 425
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
425
Morphine-11 The TBS group was removed to provide 78. The anticipated free radical gave 79 (64 where R = H) as had been hoped.6 This reaction involves sequential formation of the aryl radical, addition to the olefin to give a secondary radical (5-exo cyclization rather than 6-endo cyclization), a second radical cyclization to give a benzylic radical (6-endo cyclization rather than 5-exo cyclization), and a radical fragmentation. The expelled thiophenoxy radical was reduced by the tri-n-butyltin hydride to provide thiophenol and tri-nbutyltin radicals to continue the chain. Formation of 72 reveals that a 1,5-hydrogen atom transfer competes with the initial radical cyclization. This key reaction (as in the Fuchs synthesis) depends on a fine balance of competing reaction rates. It is notable that in the first cyclization, the 5-exo process wins over a possible 6-endo cyclization as expected, but in the second cyclization the reverse is true. The last stage of the synthesis was a pleasant surprise. Reductive remove of the N-tosyl group (Li, NH3, t-BuOH) gave 80 in outstanding yield. Presumably this reaction generated a nitrogen-centered radical that cyclized to afford the critical C–N bond. A Swern oxidation provided dihydrocodeinone which had previously been taken on to morphine and other morphinanes.
I
MeO 2C N H
CH2
H
MeO 2C N H
83
82
N
MeO
2. MeOCOCl, Et 3N
H
I
3. EtSH, BF3 -Et2O TBDMSOTf, 2,6-lutidine
84
MeO2 C
CrO 3, CH 2Cl2
N
MeO2 C
H 85
MeO
I
75%
86 MeO
I
Et3N, toluene, 110 oC
59%
OTBS
OTBS
OBn
H
N N H
Pd(OCOCF3) 2(PPh 3) 2
N
1. Ph3 P=CH2 (52%)
2. TBAF (90%)
DBS = CH2 OMe MeO2 C
N H
OH O
83 MeO
87
MeO 2C N H
I OH
48%
Morphine-12
Pd(OCOCF3) 2(PPh 3) 2 (20 mol%) 1,2,2,6,6,-pentamethylpiperidine toluene, 120 oC
Page 426
DBS
CO2 Me
83
O 1. HCO2 H, 75 o C
N
H 2C
H
Morphine (1)
I
10:57 AM
H 3C N
HO
OH CH2
12/21/2010
O
Organic Synthesis via Examination of Selected Products
O OH
MeO
OMe
b1026
OMe
Organic Synthesis via Examination of Selected Natural Products
Hong, C. Y.; Overman, L. E. "Preparation of Opium Alkaloids by Palladium Catalyzed Bis-Cyclizations. Formal Total Synthesis of Morphine" Tetr ahedr on Lett. 1994, 35, 3453. Also see: Hong, C. Y.; Kado, N.; Overman, L. E. "Asymmetric Synthesis of Either Enantiomer of Opium Alkaloids and Morphinans. Total Synthesis of (-)- and (+)-Dihydrocodeinone and (-)- and (+)-Morphine" J. Am. Chem. Soc. 1993, 115, 11028-11029. Heerding, E. A.; Hong, C. Y.; Kado, N.; Look, G. C.; Overman, L. E. "Simple Method for Controlling Stereoselection in Mannich Cyclization Reactions of Aldehydes" J. Org. Chem. 1993, 58, 6847-6948. OH
b1026_Chapter-10.qxd
426
Alternatives to Oxidative Phenolic Coupling 3: The Overman Approach
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 427
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
427
Morphine-12 Here is another approach whose biomimetic nature is more apparent than the carbanion and free radical routes we just examined. Overman (UC Irvine) anticipated that 82 could be taken to morphine through a series of standard transformations. This olefin was to be prepared from 83 via an interesting palladium-mediated reaction (a variation of the Heck arylation).7 Substrate 83 can clearly be recognized as an isoquinoline that is more highly reduced than in reticuline (31), the natural substrate used in morphine biosynthesis. In essence, the oxidation states of the two aromatic rings that nature couples have been purposely differentiated. I will not go through the synthesis of 84 (related to 83) as this can be an exercise for you. Removal of the N-DBS protecting group, allylic oxidation (Corey-Fleet reagent)8 of the alkene to enone 86, and a Wittig methylenation, provided cyclization substrate 83.
O
CH2
Pd
56%
87
H
Kenner Rice
Mulzer’s Conjugate Addition Route to Morphine
O OH
H
1
morphine OMe
OH Benzylic Handle
OMe
O OH
O
Use conjugate addition to set quaternary center
H 3C N H
MeO
89 H
1 Cl
MeO
1. HCO2 Me, NaOMe PhH (95%) 2. MVK, Et 3N
O 90
Resolved by chromatography over cellulose triacetate
MeO MeO
Cl
H
MeO
2. Me3 SiCl
MeO Br
3. NBS, THF
3. KOH, dioxane H2 O
Cl
77%
Morphine-13
O
DMF, 140 °C (minor bromide does not react)
3:1
H 91
O MVK = methyl vinyl ketone
1. (CH2 =CH)2 CuMgCl THF
92
84% (63% of β-isomer)
Page 428
H 3C N Mulzer, J.; Durner, G.; Trauner, D. "Formal Total Synthesis of (-)-Morphine by Cuprate Conjugate Addition" Angew . Chem. Int. E d. 1996, 35, 2830-2832.
10:57 AM
OH
Organic Synthesis via Examination of Selected Products
Iijima, I.; Rice, K. C.; Silverton, J. V. Het er ocy cles 1977, 6, 1157
b1026
82
88
12/21/2010
MeO 2C N H
Organic Synthesis via Examination of Selected Natural Products
H
O
MeOH, H2 O
MeO 2C N
MeO 2C N
O
b1026_Chapter-10.qxd
OsO4 , NaIO4
OH
428
OMe
OMe
OMe
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 429
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
429
Morphine-13 The projected Pd-mediated “double cyclization” was accomplished to give 82 in a good yield through a presumed intermediate of type 88. JohnsonLemieux oxidative cleavage of the double bond gave ketone 87 which had previously been converted to morphine by the Rice group (NIH). We will now move away from biomimetic synthesis and examine a perhydrophenanthrene-based approach reported by Mulzer. The plan was to prepare tetrahydrophenanthrene 89, use conjugate addition chemistry to set the quaternary center, and use the indicated benzylic position as a handle for construction of the same C–N bond central to the Gates, Fuchs and Parker syntheses. Tetralone 90 will serve as our point of departure. Annulation chemistry similar to that we encountered in the steroid area was used to provide 91. The enantiomers of 91 were separated by chromatography over a chiral support. A vinylic cuprate was used to generate the quaternary center. The resulting enolate was trapped as a trimethylsilyl enol ether, bromination of which provided α-bromoketone 93 as a mixture of diastereomers. Heating 92 converted the major diastereomer to dihydrobenzofuran 93. The minor stereoisomer did not react.
(CH 2OH)2
O
Cl
MeO 1. BH3 -Me 2S
O
Me3 SiCl
H
NMeSO2Ph O
95
O
H 96
O O O N C N N C N
ADDP =
81%
MeO
MeO 3N aq HCl
O
O
HO
H O
1
N
Me
O
99
O
95%
morphine
H
Me 98
79%
O
t-BuOH
NMeSO2Ph O
H 97
O 65%
11.5% overall f rom the tetralone
A Recent Synthesis MeO
MeO Br
? MeO N
OHC H 1,5-dicarbonyl from cyclopentene
O
100
N
SO2Ph
H O
SO2Ph
The critical bond construction is the same as in the Mulzer synthesis, but with the opposite polarity. This deposits residual functionality (after bond formation) at a diff erent location.
101
Taber, D. F.; Neubert, T. D.; Rheingold, A. L. "Synthesis of (-)-Morphine" J. Am. Chem. Soc. 2002, 124, 12416-12417. This article contains excellent references to other morphine syntheses.
Morphine-14
Page 430
Me
H
Li, NH 3, THF
O N
N
10:57 AM
MeO
HO
Organic Synthesis via Examination of Selected Products
NBS, CCl4 (PhCO) 2O, ∆
b1026
68%
92%
O
12/21/2010
94
O
93 100%
O
3. RaNi, MeOH KOH, H2
Bu3 P, ADDP
Organic Synthesis via Examination of Selected Natural Products
O
H
PhSO2 NHMe OH
2. H2 O2 , NaOH
O
CH2 Cl2
H
MeO
O
b1026_Chapter-10.qxd
MeO
Cl
430
MeO
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 431
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
431
Morphine-14 The ketone was protected and the resulting 94 was put through some standard paces to give first alcohol 95 and then sulfonamide 96. You can see that this route is headed for an intersection with the Parker route. All that is missing is the olefin needed for the N-centered radical cyclization. This was introduced using a free radical bromination-dehydrohalogenation. The resulting olefinic sulfonamide (97) cyclized to 98 upon treatment with lithium in ammonia. Ketal hydrolysis provided dihyrocodeinone (99) which we have encountered before. This synthesis is quite efficient, handles regiochemical issues nicely, and provides either enantiomer of the natural product. Taber (University of Delaware) described an approach that focuses on the same bond construction used by Mulzer to set the quaternary center, but uses the opposite sense of polarity in construction of this bond. The key transformation was projected to be 100 to 101 via a combination of aldoldehydration and intramolecular alkylation reactions (not in any particular order). Compound 100 is highly functionalized. Perhaps in an attempt to reduce the level of functionality in precursors, it was to be generated in a manner we first saw in Chapter 2, by the oxidative cleavage of a cyclopentene.
O OH
O MeO
MeO
OMe
Br
OMe
Br
105
104 80%
Recycleable
Dianion Alkylation
+
86%
Separable Ph
O
Ph
MeO
107
OMe carbene insertion into benzylic C-H bond
77%
Br 106
DPPA = (PhO)2 PON3 Br
MeO Br
NHSO 2Ph MeO 109
OMe 43%
1. O3 , CH2 Cl2 -78 oC
NSO2 Ph
TBAB, NaOH, H 2O MeO
BrCH 2CH2 Br
H
110
OMe
toluene, ∆
N
OHC
2. Ph3 P
O 83%
TBAB = n-Bu4 N Br
SO2Ph K2CO3 , TBAB toluene, ∆
100 MeO
MeO
MeO
1 morphine
BBr3
1. NaBH4
8 steps O N
86% H HO
MeO N
O Me
N
34% overall
3 (codeine)
H 111
Morphine-15
SO2Ph
H
2. BBr 3 101
O
SO2Ph
92%
Page 432
3. PhSO2 Cl
KHMDS, Et 2O
OMe
108
2. LiAlH4
Ph
O MeO
2. L-Selectride (LiEt3BH)
OMe
1. DPPA, DEAD Ph3P, THF
O
10:57 AM
1. H2 O, AcOH (80%) MeO
Ph
O
OH
Organic Synthesis via Examination of Selected Products
Krapcho Decarboxylation
b1026
50%
Alcohol will become the amine
Ph
12/21/2010
103
102
HO
Organic Synthesis via Examination of Selected Natural Products
2. RBr 3. LiCl, DMSO H2 O, ∆
OMe
Br
O
1. LDA (2 eq)
MeO
HO
Ph
Ph
b1026_Chapter-10.qxd
5 steps
432
Br
Ph
CO2 Me O
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 433
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
433
Morphine-15 We will not go through every detail of the synthesis. Suffice it to say that the known 1,6-dibromo-2-naphthol (102) was first converted to β-ketoester 103. The dianion derived from 103 was then alkylated with 1,3-dibromo-2methylpropene. Krapcho decarbomethoxylation of the β-ketoester gave 104.9 Ketone 104 was then resolved via chromatographic separation of diastereomeric acetals 105 and 106. The undesired acetal (105) could be recycled. The desired acetal could was converted to 107 via insertion of a carbene (α-elimination of HBr from 106) into the benzylic C–H bond. This is an unusual and creative approach to this compound that illustrates the power of such insertion reactions.10 Hydrolysis of the acetal followed by reduction of the intermediate ketone from the sterically most accessible face gave 108. The alcohol was converted to the corresponding azide (with inversion of configuration), which was reduced to the amine and converted to sulfonamide 109. Alkylation of 109 under phase transfer catalysis conditions provided 110, the penultimate intermediate in the projected synthesis of 101. Ozonolysis of the olefin and a reductive workup with triphenylphosphine provided 100. Treatment of 100 with base provided 101 as planned. It was shown (by isolation of an intermediate product) that the sequence of events was intramolecular alkylation of the aldehyde enolate followed by the aldoldehydration. Sodium borohydride reduction of 101, followed by treatment of the allylic alcohol with boron tribromide, gave 111 with the morphinane ring system intact. A lengthy, but reasonably efficient sequence of reactions converted 111 to morphine (1) via codeine (3).
MeO
CH2
H
CH3 H 2C
N
14
CH3
H
112 O
112
CH3
113
stereochemistry ? OMe Br
Evans, D. A.; Mitch, C. H. "Studies Directed Towards the Total Synthesis of Morphine Alkaloids" Tetr ahedr on Lett . 1982, 23, 285-288.
OMe
H 2C
Br
Br
H 2C
ZnBr 2
OMe TsOH, toluene
117 HO
116
Br
80%
Br
OMe n-BuLi
N
118
OMe N
CH3
114
N
OMe
14
N
CH3 H 112-α
OMe
H 2C kinetic product OMe
MeOH
OMe
H 2C 14
60% (95:5)
H
N
HClO4
OMe
H 2C
MeOH-Et2 O
N
CH3
112-β
Morphine-16
113
N
K2CO3
ClO4 OMe
2. RBr OMe
OMe
OMe ClO4
1. n-BuLi, -10 °C, THF
CH3
CH3
Li
H 2C
115
O +
CH3
114
95%
40-50%
OMe
OMe
OMe
CH3
CH3 CN NaI
Br
119
CH3
Page 434
N OMe
Br
10:57 AM
CH2
N
CH3
Organic Synthesis via Examination of Selected Products
1
N
12/21/2010
N
OMe
H 2C
b1026
H
OMe
H 2C
Organic Synthesis via Examination of Selected Natural Products
14
HO
OMe
OMe
MeO
O
b1026_Chapter-10.qxd
HO
434
Approaches Focusing on Late Stage Introduction of Benzylic Methylene
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 435
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
435
Morphine-16 The final approach we will examine revolves around construction of an angularly arylated perhydroisoquinoline of type 112, functionalized to allow introduction of the benzylic methylene group. This was to be accomplished using a zwitterionic CH2 equivalent. It was imagined that such an equivalent would behave as a nucleophile, and add to the iminium ion, and then as an electrophile toward the aromatic ring. Iminium ion 112 was to be generated by protonation of enamine 113, which was to be assembled by alkylation of the anion derived from 114 with bis-electrophile 115. Replacement of the ketone in morphine with a methylene was expected to control regioselectivity in the bis-alkylation. Note that this also reduces the number of electrophilic sites in the alkylating agent. There are stereochemical issues associated with this approach. Attack of the nucleophile on iminium ion 112 would have to generate a cis-relationship between the entering nucleophile and the aryl group, otherwise the anticipated electrophilic aromatic substitution reaction would be impossible. In addition, protonation of 113 would set stereochemistry at what becomes C14 of morphine and thus, stereoselectivity here is also an issue. The alkylating agent (117 = 115) was prepared from 116 via a modified Julia olefin synthesis. 4-Aryl-1-piperidinol 118 was prepared by o-metallation of catechol dimethyl ether followed by reaction of the organolithium reagent with N-methyl-4-piperidinone. Dehydration of 118 provided 114. Metallation of 114 with n-BuLi, followed by reaction with 117, gave 119. Conversion of the primary bromide to an iodide was followed by intramolecular alkylation of the enamine to provide 113. Kinetic protonation of 113 gave iminium perchlorate 112-β (trans-fused ring system). Upon dissolving in methanol this equilibrated with cis-fused iminium perchlorate 112-α. This set the stage for the key reaction.
ClO4
CHO
ClO4 CH2 N2 , CH2 Cl2
OMe
H 2C
cis ring fusion gives stereocontrol in aziridinium ion formation
H
120
121
95%
95%
H
O
N CH3
124
HO H
N CH3
2. LiEt 3BH H 3C N
H
122
BF3 -Et2 O -10 οC
H CH2
H 2C 123
90% overall
OMe
1. MsCl
MeO
THF-H 2O-AcOH (3:1:1)
OMe OMe O H 3C N
H H
80%
Intermediate in Gates synthesis of morphine. Drawback is lack of stereocontrol at C 14 .
OMe OMe
MeO
H
A strategically related synthesis of O-methylpallidinine (left) follows: McMurry, J. E.; Farina, V. "Total Synthesis of O-Methylpallidinine" T et rahedron Let t. 1983, 24, 4653-4656.
N
1. B2H 6, THF
O OMe
O 126
CH3
MeO O
Synthesis?
125
Morphine-17
2. H2 O2 , NaOH H2 O
CH2
Page 436
14
OMe
10:57 AM
OsO4 (cat), NaIO4
CHO
OMe
MeO
MeO
OMe
H 2C
CH3
Organic Synthesis via Examination of Selected Products
MeO
N
CH3
12/21/2010
H
b1026
112-α
N
CH3
H
Organic Synthesis via Examination of Selected Natural Products
N
OMe
H 2C
25 οC
25 °C
H
DMSO
OMe
H 2C
CH3
N
OMe
b1026_Chapter-10.qxd
OMe
436
OMe
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 437
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
437
Morphine-17 Treatment of 112-α with diazomethane gave aziridinium ion 120. The cis ring fusion apparently directs attack of the diazomethane to its convex face. In addition, rather than the aromatic ring behaving as the “nucleophile” toward the methylene cation equivalent, the perhydroisoquiniline nitrogen intervened. This problem was corrected by oxidizing the aziridinium ion to aldehyde 121 using a Kornblum-type oxidation.11 This substrate was suitable for the electrophilic aromatic substitution as reaction with boron trifluoride etherate provided 122, via an intramolecular EAS reaction from the less stable chair-chair conformation available to 121. The benzylic hydroxyl group was removed via reduction of the derived mesylate. A Johnson-Lemieux oxidation converted 123 to 124. Ketone 124 was an intermediate in the Gates synthesis and thus, this completed a formal synthesis of morphine. One drawback of this synthesis is that it initially provides the wrong stereochemistry at C14. There are morphinanes, however, that have C14α stereochemistry, so this is not all bad. One such morphinane is Omethylpallidinine (125), and a strategically similar approach to this alkaloid will be the last synthesis we will consider in this chapter. John McMurry (Cornell University) described a synthesis of 125 that begins with preparation of aryl cyclohexene 126 (see Problem 21). Hydroboration-oxidation of this alkene gave ketone 127.
O OMe
O
OMe
PCC NaOAc CH2 Cl2
O
NaH methallyl chloride
O
O 130
79%
Lemieux-von Rudolph Oxidation
LiAlH 4
O
O N
Me
134
133
100%
40%
O
pyridine H 2O
Me
OMe MeNHOH
O N Me
O
O NaOAc
O
O 131
132 100%
80%
HClO4
OMe
OMe
OMe
OMe
MeO
MeO
OMe
MeO
N2
O
CH2 N2 N H 135 78%
Me
1. NaH, Ph, HCO 2Et 2. TsS(CH 2 )3STs, KOAc
O
O N
CH3 CN H 136
Me
N H 137
Me
3. m-CPBA 4. HCl, H 2O 5. TsOH, MeOH
MeO O
125 N Me H 20% overall
30% O-Methylpalladinine
Morphine-18
Page 438
N
OMe OMe
TsCl
O O
O
OMe OMe
OMe
10:57 AM
OMe
OMe
Organic Synthesis via Examination of Selected Products
KOH, EtOH
b1026
84%
100%
dioxane
O 129
Note use of allylic choride instead of α-chloroacetone (a 1,2-difunctional compound and bis-electrophile)
O
12/21/2010
O 128
O O
NaIO4
Organic Synthesis via Examination of Selected Natural Products
OH 127
KMnO 4 (cat)
O
OMe
O
OMe
b1026_Chapter-10.qxd
OMe
OMe
438
OMe
OMe
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 439
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
439
Morphine-18 Oxidation of 127 with PCC, buffered to discourage acetal hydrolysis, gave 128. Alkylation of 128 with methallyl chloride, followed by oxidative cleavage of the olefin, gave 130. The transformation of 128 to 130 involves construction of a 1,4-difunctional relationship. The use of an allylic halide in the alkylation, rather than an α-haloacetone (a 1,2-difunctional compound), avoids the use of a bis-electrophile that might suffer from regiochemical problems. Treatment of 130 with base, promoted an aldol-dehydration to provide 131. The ketone was then converted to nitrone 132, and a Beckman-type rearrangement provided 133. Lithium aluminum hydride reduction of the lactam gave enamine 134. This intermediate is similar to enamine 113 (Morphine-16), but there is a difference in the placement of the methoxy groups on the aromatic ring. It turned out that reaction of 134 with perchloric acid, followed by diazomethane, directly gave “methylene insertion product” 137, via iminium ion 135 and presumably alkyldiazonium salt 136 as intermediates. The rate of capture of the intermediate 136 by the aromatic ring was apparently faster in 136 than in the comparable ion derived from 112. The synthesis of O-methylpallidinine was completed by a series of oxidation state adjustments in the cyclohexanone ring. This concludes our look at morphine and indeed, at alkaloids. In closing I would like to make a suggestion, first made to me by one of my former teachers, Henry Rapoport. Rap was trying to convince me (and other students in a class) that heterocyclic chemistry is actually much easier that hydrocarbon chemistry (by this I think he meant terpenoid chemistry). To paraphrase what he said, “Making carbon-heteroatom bonds is easy relative to making carboncarbon bonds. There are electronegativity differences between the atoms, but carbon is carbon and you have to mess with it to be able to make a carbon-carbon bond. I am surprised more of you don’t want to do heterocyclic chemistry”. Of course he was trying to sell us on his area of research interest, but that simple statement helped me be less afraid of nitrogen (for example) as a young student of synthesis. It made a difference to me and I hope what we have looked at here may do the same for you.
b1026_Chapter-10.qxd
12/21/2010 b1026
440
10:57 AM
Page 440
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
References 1. Rice, K. C. “Synthetic Opium Alkaloids and Derivatives. A Short Total Synthesis of (±)-Dihydrothebainone, (±)-Dihydrododeinone, and (±)-Nordihydrocodeinone as an Approach to a Practical Synthesis of Morphine, Codeine and Congeners” J. Org. Chem. 1980, 45, 3135–3137. Morrison, G. C.; Waite, R. O.; Shavel, J. Jr. “Alternate Route in the Synthesis of Morphine” Tetrahedron Lett. 1967, 9, 4055–4056. 2. Mitsunobu, O. “The Use of Diethyl Azodicarboxylated and Triphenylphosphine in Synthesis and Transformation of Natural Products” Synthesis 1981, 1–28. 3. Mancuso, A. J.; Swern, D. “Activated Dimethyl Sulfoxide: Useful Reagents for Synthesis” Synthesis 1981, 165–185. Mancuso, A. J.; Brownfain, D. S.; Swern, D. “Structure of the Dimethyl Sulfoxide-Oxalyl Chloride Reaction Product. Oxidation of Heteroaromatic and Diverse Alcohols to Carbonyl Compounds” J. Org. Chem. 1979, 44, 4148–4150. Mancuso, A.; J.; Huang, S-L.; Swern, E. “Oxidation of Long-Chain and Related Alcohols to Carbonyls by Dimethyl Sulfoxide “Activated” by Oxalyl Chloride” J. Org. Chem. 1978, 43, 2480–2482. Omura, K.; Swern, D. “Oxidation of Alcohols by “Activated” Dimethyl Sulfoxide. A Preparative Steric and Mechanistic Study” Tetrahedron 1978, 34, 1651–1660. 4. Luche, J. L. “Lanthanides in Organic Chemistry. 1. Selective 1,2-Reductions of Conjugated Ketones” J. Am. Chem. Soc. 1978, 100, 2226–2227. 5. Morgans, D. J. Jr.; Sharpless, K. B.; Traynor, S. G. “Epoxy Alcohol Rearrangements: Hydroxyl-Mediated Delivery of Lewis Acid Promoters” J. Am. Chem. Soc. 1981, 103, 462–464. 6. For reviews and books see: Julia, M. “Cyclizations by Radical Reactions” Rec. Chem. Progr. 1964, 25, 3–29. Julia, M. “Free Radical Cyclizations” Pure Appl. Chem. 1967, 15, 167–183. Julia, M. “Free-Radical Cyclizations” Acc. Chem. Res. 1971, 4, 386–392. Beckwith, A. L. J. “Some Guidelines for Radical Reactions” J. Chem. Soc., Chem. Commun. 1980, 482–483. Beckwith, A. L. J. “Regioselectivity and Stereoselectivity in Radical Reactions” Tetrahedron 1981, 37, 3073–3100. Hart, D. J.; “Free-Radical Carbon-Carbon Bond Formation in Organic Synthesis” Science 1984, 223, 883–887. Beckwith, A. L. J. “Mechanism and Applications of Free Radical Cyclization” Rev. Chem. Int. 1986, 7, 143–154. Giese, B. “Radicals in Organic Synthesis: Formation of CarbonCarbon Bonds” Pergamon Press, 1986 (294 pages). Stork, G. “Radical Cyclization in the Control of Regio- and Stereochemistry” Bull. Chem. Soc. Jap. 1988, 61, 149–151. Giese, B.; Kopping, B.; Gobel, T.; Dickhout, J.; Thoma, G.; Kulicke, K. J.; Trach, F. “Radical Cyclization Reactions” Organic Reactions 1996, 48, 301–856.
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 441
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
441
7. Link, J. T. “The Intramolecular Heck Reaction” Organic Reactions 2002, 60, 157–534. Shibasaki, M.; Vogl, E. M. “Heck Reaction” Comprehensive Asymmetric Catalysis I-III 1999, 1, 457–487. 8. Corey, E. J.; Fleet, G. W. J. “Chromium Trioxide-3,5-Dimethylpyrazole complex as a Reagent for Oxidation of Alcohols to Carbonyl Compounds” Tetrahedron Lett. 1973, 15, 4499–4501. 9. Krapcho, A. P. “Recent Synthetic Applications of the Dealkoxycarbonylation Reaction. Part 1. Dealkoxycarbonylations of Malonate Esters. ARKIVOC (Gainesville, FL, USA) 2007, 1–53. Krapcho, A. P. “Recent Synthetic Applications of the Dealkoxycarbonylation Reaction. Part 2. Dealkoxycarbonylations of β-Keto Esters, α-Cyano Esters and Related Analogues” ARKIVOC (Gainesville, FL, USA) 2007, 54–120. Krapcho, A. P.; Jahngen, E. G. E. Jr.; Lovey, A. J.; Short, F. W. “Decarbalkoxylations of Geminal Diesters and β-Keto Esters in Wet Dimethyl Sulfoxide. Effect of Added Sodium Chloride on the Decarbalkoxylation Rates of Mono- and Disubstituted Malonate Esters” Tetrahedron Lett. 1974, 16, 1091–1094. Krapcho, A. P.; Lovey, A. J. “Decarbalkoxylations of Geminal Diesters, β-Keto Esters, and α-Cyano Esters Effected by Sodium Chloride in Dimethyl Sulfoxide” Tetrahedron Lett. 1973, 15, 957–960. 10. Chen, M. S.; White, M. C. “A Predictably Selective Aliphatic C-H Oxidation Reaction for Complex Molecule Synthesis” Science 2007, 318, 783–787. Davies, H. M. L. “Recent Advances in Catalytic Enantioselective Intermolecular C-H Functionalization” Angew. Chem. Int. Ed. 2006, 45, 6422–6425. Tambar, U. K.; Ebner, D. C.; Stoltz, B. M. “A Convergent and Enantioselective Synthesis of (+)Amurensinine via Selective C-H and C-C Bond Insertion Reactions” J. Am. Chem. Soc. 2006, 128, 11752–11753. Feldman, K. S. “Alkynyliodonium Salts in Organic Synthesis” in Strategies and Tactics in Organic Synthesis, Harmata, M. Ed.; 2004, 4, 133–170. Doyle, M. P. “Synthetic Carbene and Nitrene Chemistry” Reactive Intermediate Chemistry 2004, 561–592. Hinman, A.; Du Bois, J. “A Stereoselective of (−)-Tetrodotoxin” J. Am. Chem. Soc. 2003, 125, 11510–11511. Salzmann, T. N.; Ratcliffe, R. W.; Christensen, B. G.; Bouffard, F. A. “A Stereocontrolled Synthesis of (+)-Thienamycin” J. Am. Chem. Soc. 1980, 102, 6161–6163. Cory, R. M.; McLaren, F. R. “Bicycloannulation. Carbon Atom Insertion: An Efficient Synthesis of Ishwarane” J. Chem. Soc., Chem Commun. 1977, 587–488. 11. Kornblum, N.; Powers, J. W.; Anderson, G. J.; Jones, W. J.; Larson, H. O.; Levand, O.; Weaver, W. M. “New and Selective Method of Oxidation” J. Am. Chem. Soc. 1957, 79, 6562.
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 442
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
442
Problems 1. Discuss the regioselectivity in the nitrosation of 10. (Morphine-2) 2. Describe the sequence of events leading from 16 → 17. (Morphine-2) 3. Suggest tactics for the conversion of morphine (1) to β-thebainone (20). (Morphine-3) 4. Provide mechanisms for the conversion of 21 → 22 and the oxidation of 22 → 23. (Morphine-3) 5. Suggest a structure for the tribromide derived from 25, and a mechanism for its transformation to 26. (Morphine-4) 6. The 2,4-DNP chemistry in the Gates synthesis is quite interesting. Is the following transformation mechanistically related? (Morphine-4) NNHSO 2Ph Br
NNHSO 2Ph PhCu (2.5-3.0 equiv)
Ph
aq. acetone
THF-Et2 O -60 °C
O Ph
BF3 -Et 2 O
75%
95%
Sacks. C. E.; Fuchs, P. L. “α-Arylation of Carbonyl Groups. Utilization of the p-Toluenesulfonylazo Olefin Functional Group as an Enolonium Synthon” J. Am. Chem. Soc. 1975, 97, 7273–7274. See also Stork, G.; Ponaras, A. A. “α-Alkylation and Arylation of α,β-Unsaturated Ketones” J. Org. Chem. 1976, 41, 2937–2939. 7. Show the structural relationship between the following natural products and oxidative phenolic coupling precursors. (Morphine-5) OH O
O O
O
O O
O
Steganone MeO MeO
HO2C
OMe HO
N H
H N
NHAc O
K-13
8. What is the basis for the comment that the carbamate (in 40) forces the “benzylic group” to be pseudo-axial on the tetrahydroisoquinoline core? (Morphine-7) 9. Provide tactics for the conversion of 48 → 49. (Morphine-8)
b1026_Chapter-10.qxd
12/21/2010 b1026
10:57 AM
Page 443
Organic Synthesis via Examination of Selected Products
Morphine and Oxidative Phenolic Coupling
443
10. Outline several syntheses of 50. Use acyclic starting materials. Use catechol dimethyl ether as a starting material. (Morphine-8) 11. Provide the structures of intermediates en route from 59 → 60. Suggest reagents for accomplishing transformations where they are missing from the text. (Morphine-9) 12. Provide a mechanistic interpretation of the conversion of 74 → 75. Note that no reaction is expected in the absence of the free hydroxyl group. (Morphine-10) 13. Outline a reaction sequence that will accomplish the following transformation. (Morphine-10) Me Me3Si
O
Me
Me H
?
O
OH
Br
Me3Si OH
OH
14. Speculate regarding the success of the second free radical cyclization in the conversion of 78 → 79 (why 6-endo over 5-exo). (Morphine-11) 15. Suggest a mechanism for the conversion of 92 → 93. (Morphine-13) 16. Outline a reaction sequence that would convert 102 → 103. (Morphine-15) 17. Mesembrine is an angularly arylated perhydroindole alkaloid that is structurally similar to morphine. Design several syntheses of mesembrine that use some of the strategies described in Chapters 8–10. Compare your approaches to those reported in the literature. (Morphine-16) OMe MeO
N O Me H mesembrine
18. Draw structures of 112-β and 112-α that show their respective conformations. (Morphine-16) 19. Provide mechanisms for the conversion of 112-α → 120 and of 120 → 121. (Morphine-17)
b1026_Chapter-10.qxd
12/21/2010 b1026
444
10:57 AM
Page 444
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
20. Outline a synthesis of 126 from commercially available materials. (Morphine-17) 21. Provide a mechanism for the conversion of 132 → 133. (Morphine-18) 22. Provide the structure of intermediates after each step in the conversion of 137 → 125. (Morphine-18)
b1026_Chapter-11.qxd
446
Cecropia Juvenile Hormone Trost, B. M. "The Juvenile Hormone of Hyalophora Cecr opia" Acc. Chem. Res. 1970, 3, 120.
3
Cecropia Juvenile Hormone (1)
OMe HWE
CO2 Me
O O
2
P
(OMe)2
5 O CO2 Et 2.14
2.63
CO2 Me
CO2 Me
1.87 2.15
cis (Z)
trans (E)
In β−substituted α, β−unsaturated esters, the β−group cis to the CO2R is more deshielded than the β−group trans to the CO2R.
Cecropia Juvenile Hormone-1
Page 446
O
4
O OMe
10:57 AM
O
12/21/2010
acetoacetic ester
Organic Synthesis via Examination of Selected Products
Dahm, K. H.; Trost, B. M.; Roller, H. "The Juvenile Hormone. V. Synthesis of the Racemic Juvenile Hormone" J. Am. Chem. Soc. 1967, 89, 5292-5294.
b1026
Once the structure was known, many syntheses were reported and all of a sudden the literature was loaded with stereoselective methods f or olefin synthesis. Although it is only a guess, it is probable that the discovery of Cecropia juvenile hormone (CJH) was responsible (in part) for this burst of activity. Another reason would be the interest in carbocycle synthesis via polyolef in cyclizations (the development of steroid and terpene chemistry) which slightly preceeded the discovery of CJH. We will start by looking at the Trost group synthesis (the first synthesis) of CJH and then examine several other syntheses wherein the focus is stereoselective olefin synthesis.
Organic Synthesis via Examination of Selected Natural Products
One field to which organic synthesis has contributed is the identification and production of insect chemical signaling agents (growth hormones, sex hormones). These are of interest (in part) as a means of contolling insect populations. One such hormone is the Cecr opia juvenile hormone. This compound is involved in the molting process of the moth. The developmental sequence of many insects involves three stages: larval, pupal and adult. In some insects a "molt" occurs that results in the insect proceeding directly from the larval to the adult stage. These molts are controlled by hormones. In Hyalophora cecr opia, the "juvenile hormone" results in retention of juvenile characteristics as each molt occurs. The more "juvenile hormone" is present, the more juvenile characteristics are retained after the molt. The Cecropia juvenile hormone was isolated in 1965 by the Roeller group in the Department of Zoology at the University of Wisconsin from the abdomen of the male adult (0.5 µg/abdomen).Only 200 µg of material was available at any point in time. Based on NMR and mass spectral analyses, a structure (without stereochemical detail) was proposed. Due to the stereochemical ambiguities and lack of material, synthesis was used to ultimately determine its structure as shown below. The first synthesis was stereorandom (no doubt on purpose) and provided a number of isomers. Bioassays were used to determine which isomer was the true hormone.
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 447
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
447
Cecropia Juvenile Hormone-1 This chapter focuses on selected aspects of stereoselective olefin synthesis. This is an important topic for a number of reasons: (1) When olefins are targets it is simply necessary to have a litany of methods available for their synthesis. Olefins are ubiquitous in nature as we have already seen (for examples see the prostaglandin and poison-dart alkaloid sidechains). (2) Olefin stereochemistry can be used to control non-olefinic stereochemical relationships that must be established during the course of a target-oriented synthesis. For examples, see the syntheses of pyrrolizidine alkaloids where olefin addition reactions were used to control vicinal stereochemistry (also see Prostaglandins-23), or the Johnson approach to steroids where olefin stereochemistry is translated into vicinal stereochemistry (see Steroids-14 and 17), or the importance of olefin geometry in transfer-of-chirality reactions (see Prostaglandins-24 and 25). We will explore the topic of stereoselective olefin synthesis within the context of Cecropia Juvenile Hormone (1) in this chapter and explore additional methods in Chapter 12. So what is Cecropia juvenile hormone and why was it of interest as a target for synthesis? For a discussion see page 446 (CJH-1).1 The synthetic plan was to prepare 1 by regioselective epoxidation of triene 2. This is a reasonable idea because one of the olefins is electron-deficient, and of the other two, the “internal” olefin might be expected to be sterically more conjested if 2 coils into a ball-like structure. The plan was to build “from left-to-right” using an iterative (repetitive) reaction sequence. Thus, it was imagined that 2-butanone (3) would undergo a HornerWadsworth-Emmons (HWE) reaction with β-ketophosphonate 4 to provide a mixture of α,β-unsaturated ester geometrical isomers. This reaction would establish what was to become the “left-hand” olefin of 2. Little stereoselectivity was expected in this reaction because the difference in size between methyl and ethyl groups is minimal. It was planned to convert the ester to a halide (bromide or iodide) and then perform an acetoacetic ester synthesis using β-ketoester 5 as the nucleophile.2 This would provide a ketone that could be subjected to another HWE reaction-separation-conversion to halide-alkylation sequence to introduce the “middle” olefin of 2. A partial repeat of this sequence (stopping after the HWE reaction) was to complete the synthesis of 2. Let’s move to the synthesis.
b1026_Chapter-11.qxd
448
CN
EtMgBr
CO2 Et
1.87 ( 2.14)
1. alkylate
Br
7
17% (37% trans)
O
2. NaOH 3. H3 O+
2. PBr 3 (45%)
2. chromatography
73%
ethyl acetoacetate
Br
1. LiAlH4 (93%) 2. PBr 3 (52%) 39% (30% isomer)
12
O
2. NaOH 3. H3 O+
11
CO 2Me
m-CPBA
CO 2Me O
30% (18% isomer)
2
2. chromatography
47% 13
+
CO 2Me O
40%
10% 14
r ac-CJH
1
Although this synthesis was stereorandom, it was necessary because the goal was structure determination. This is the kind of synthesis that is sometimes pursued in the early stages of a medicinal chemistry effort where the goal is activity and structure is secondary. In otherwords, the structure of the specific target f ollows f rom biological activity.
Cecropia Juvenile Hormone-2
Page 448
CO2 Me
1. (MeO)2 POCH2 CO 2Me NaH
1. alkylate
10:57 AM
9
6
Organic Synthesis via Examination of Selected Products
3
10 1. (MeO)2 POCH2 CO 2Me NaH
12/21/2010
2. distillation
52%
b1026
O
1. LiAlH4 (86%) CO2 Me
O CO2 Et
Organic Synthesis via Examination of Selected Natural Products
1. (MeO)2 POCH2 CO 2Me NaH
8
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 449
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
449
Cecropia Juvenile Hormone-2 The initial HWE reaction did provide a mixture of stereoisomeric olefins, from which 6 was isolated by a spinning band distillation. Reduction to the allylic alcohol was followed by conversion to bromide 7. β-Ketoester 8 was prepared from ethyl cyanoacetate and used to convert 7 to ketone 9 using a classical “acetoacetic ester synthesis”.2 Note that the ester, which is eventually sacrificed, controls the regiochemical course of the alkylation reaction. A repeat of the HWE reaction gave 11, which was converted to ketone 13 via bromide 12 in the standard manner. A final HWE reaction provided 2. Epoxidation of 2 proceeded with modest regioselectivity to give 1 along with a small amount of the regioisomeric epoxide 14. Although this synthesis was stereorandom, it was necessary because the goal of the synthesis was structure determination. This is the kind of synthesis that is sometimes pursued in the early stages of a medicinal chemistry effort where the goal is activity and structure is secondary. In other words, the structure of the specific target follows from biological activity. We will now move to some stereoselective syntheses.
1
O
R2 a
O
20
28
O
O
MeO OMe
CO2Me
O
O
CO2Me O
24 31
E:Z = 99:1
140 °C
110 oC
29
CO2Me
a
23
30
30
NMe2 a
O
O
OH
22
a
NMe2
17
85-98% E-geometry
a
21
110 °C
CO2Me
OH
E:Z = 9:1
E:Z = 99:1
O
CN
CHO 27
19 110 °C
O
26
Results are consistent with Perrin-Faulkner analysis: Perrin, C. L.; Faulkner, D. J.; Tetrahedron Lett. 1969, 2873.
NaBH4
O
25
O OMe
OMe O
32
Cecropia Juvenile Hormone-3
OH
33
Page 450
O
10:57 AM
R2
Organic Synthesis via Examination of Selected Products
Faulkner, D. J.; Petersen, M. R. “Application of the Claisen Rearrangement to the Synthesis of Trans Trisubstituted Olefinic Bonds. Synthesis of Squalene and Insect Juvenile Hormone” J. Am. Chem. Soc. 1973, 95, 553.
R2
18
epoxide
R1
R1
OH
OH 17
16
15
Claisen Rearrangement as an “enforced” SN2' reaction.
R1
O
b1026
Cecropia Juvenile Hormone
HO
12/21/2010
OMe
OMe
O
Organic Synthesis via Examination of Selected Natural Products
a
1. SN2' 2. Reduce 3. SN2' 4. Reduce 5. diol
O
O a
b1026_Chapter-11.qxd
450
Claisen Rearrangements: The Faulkner Approach
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 451
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
451
Cecropia Juvenile Hormone-3 Johnson and Faulkner reported a synthesis of 1 that revolved around stereoselective synthesis of trisubstituted olefins via the Claisen rearrangement. The plan was to build CJH (1) from 15–17 (or derivatives thereof) starting with allylic alcohol 17. How might one disconnect 1 into these pieces? Imagine the enol or enolate of 16 reacting with an appropriate derivative of 17 via an SN2′ reaction. This would establish the right-hand “a”-bond shown in structure 1 (the olefin conjugated to the carbomethoxy group). As we will see, the Claisen rearrangement is a reaction that accomplishes this transformation with good control of olefin geometry.3 Reduction of the resulting ketone to the corresponding allylic alcohol, and repetition of this process using the enolate of 15, would provide an α-hydroxyketone that could be carried forward to CJH via reduction of the ketone, and conversion of the resulting vicinal diol to the epoxide. Let’s first examine some stereochemical aspects of the Claisen rearrangement. This transformation involves conversion of an allylic alcohol (18) to an allyl vinyl ether (19), followed by thermal rearrangement to a γ,δ-unsaturated carbonyl compound (20). Notice that this reaction looks like a reaction between an enol of a carbonyl compound, acetaldehyde for the specific reaction shown, with an allylic alcohol derivative with SN2′ regiochemistry. There is ample evidence that this rearrangement takes place via a transition state in which the allyl vinyl ether is chair-like. The terminal olefin substituents occupy axial or equatorial sites, the internal olefin substituents occupy axial sites, and substituents at the original carbinol center occupy equatorial or axial sites that minimize 1,3-diaxial interactions in the rearrangement transition state. Thus, if one of the carbinol subsitutents is hydrogen, it occupies an axial site. This transition state analysis predicts (in accord with fact) that 19 will provide the trisubstituted olefin geometry indicated in structure 20. Specific examples are the conversion of 21, 23 and 25 → 22, 24 and 26, respectively. Note that the rearrangement of 21 is less stereoselective than the rearrangements of 23 or 25. This is because one of the internal olefin substitutents in 21 is hydrogen and neither of the internal olefin substituents in 23 and 25 are hydrogens. You should work the problem at the end of this chapter to see if you understand, in depth, the reason for this difference. Application of this olefin synthesis to the CJH problem began with conversion of methacrolein (27) to allylic alcohol 17 via cyanohydrin 28. Reaction of 17 with ketal 29 provided 31/32 with good olefin stereoselectivity, presumably via allyl vinyl ether intermediate 30. Reduction of 32 with sodium borohydride gave allylic alcohol 33 and set the stage for the second “forced” SN2′ reaction.
b1026_Chapter-11.qxd
452
O
MeO OMe
HO
OMe OH
34
HO
O
110 °C
O
33
O
NaBH4 OMe
HO
0 °C
OMe OH
35
36
1. TsCl, pyridine
O
O HO
38
O OMe
OH
HgO, BF3-Et2 O HO
CF3CO2 H, MeOH
1. 2. 3. 4.
HO MeO OMe
1
phthalic anhydride resolve as brucine salt NaOH, H2O Distill
Julia Olefin Synthesis Approach Brady, S. F.; Ilton, M. A.; Johnson, W. S. J. Am. Chem. Soc. 1968, 90, 2882. Diastereselective
O OMe
O
O
Addition
Alkylation
O OMe X
1
O Br
OMe 40
39 Julia Olefin Synthesis
CO2Me
1. (MeO)2 C=O, NaH
O 42
1. Ba(OH) 2, MeOH OMe
O
2. Br
O
O OMe
44a
O
2. H3 O+ 3. CH 2N 2
OMe
Br X
43
Cecropia Juvenile Hormone-4
41
Page 452
Synthesis of Tertiary Alcohol
OMe
O
37
10:57 AM
OMe 2. NaOMe, MeOH
Organic Synthesis via Examination of Selected Products
O
12/21/2010
2. NaOMe, MeOH
b1026
1. TsCl, pyridine
Organic Synthesis via Examination of Selected Natural Products
Reduction gives mixture of diastereomers
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 453
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
453
Cecropia Juvenile Hormone-4 Reaction of 33 with 34 gave 35 with excellent olefin stereoselectivity. When racemic 33 was used, racemic 35 was obtained. When a single enantiomer of 34 was used (synthesis outlined without comment), a single enantiomer of 35 was obtained. Sodium borohydride reduction of 35 gave a mixture of diols 36 and 37. Conversion of the secondary alcohol of 36 to a tosylate, followed by treatment with base, gave CJH (1) via a Williamson ether synthesis. Diastereomeric diol 37 was converted to CJH diastereomer 38 in a similar manner. This synthesis not only demonstrated the importance of the Claisen rearrangment as methodology for stereoselective olefin synthesis, but established the absolute configuration of Cecropia Juvenile Hormone. The Johnson group also examined a synthesis that relied on the Julia olefin synthesis for construction of the central trisubstituted olefin.4 In this plan, 1 was to be prepared from an α-haloketone of type 39 via diastereoselective addition of a methyl group to the ketone, followed by a Williamson ether synthesis. Ketone 39 was to be prepared from 40 using an acetoacetic ester synthesis. Compound 40 was to be prepared from 41 using the Julia olefin synthesis. A make-or-break aspect of this plan was the stereochemical course of the Julia synthesis. Of course it was anticipated that the proper stereochemistry would result as will be seen shortly. Julia substrate 41 was prepared as follows. Cyclopropylketone 42 was converted to the corresponding β-ketoester using sodium hydride and dimethyl carbonate. The anion derived from the β-ketoester was alkylated with allylic bromide 43 to provide 44a. Ester hydrolysis, decarboxylation of the intermediate β-ketoacid, and esterification of the terminal carboxylic acid using diazomethane, gave ketone 44b (CJH-5).
O 44b
OMe
Br 45
45%
O
2. Ba(OH)2, EtOH O
O
1
Br Br
Et Cl
H
Cl
Cl O
51
49 8% of diastereomer
Br
1. minimize steric interactions 2. anti-elimination
Cl
Stereochemistry in the α-Chloroketone Addition
Stereochemistry in the Julia Olefin Synthesis
50
OMe
HO
53
The Felkin-Ahn model (shown here) and the Cornforth model (not shown here) both predict the stereochemical course of the epoxide formation. This is an “early example” of an application of principles in acyclic diastereoselection.
O
H
H
OH
Et Me
Cl Me Et
52
Cecropia Juvenile Hormone-5
H O 55
Et
Me H
HO
54
Page 454
MeOH 10 min
8% of isomeric epoxide
O
K2CO3 OMe
O
48
-78 oC
MeMgCl, THF
All yields (except those indicated) at least 90%
Br
45%
10:57 AM
dehydrohalogenation competes with alkylation
Ha
OMe Cl
47
Organic Synthesis via Examination of Selected Products
THF-5% HMPA, ∆ 4 days
O
1. LiCl, CuCl OMe
b1026
O O
40
12/21/2010
96% E, E 4% E, Z
Organic Synthesis via Examination of Selected Natural Products
46
NaI, HMPA
Br
Li enolate OMe
O
ZnBr2, Et2O OMe
LiBr, collidine Et2O
O I
O
b1026_Chapter-11.qxd
OMe
454
1. NaBH4 2. PBr3
O
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 455
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
455
Cecropia Juvenile Hormone-5 Reduction of 44b gave the corresponding alcohol, which was converted to bromide 45. Treatment of 45 with ZnBr2 provided the desired olefin (40) with excellent control of stereochemistry. The observed stereoselectivity was rationalized as follows. It was reasoned that 45 (see structure 50 for a truncated version of this substrate) would react from conformation 51 in which (1) steric interactions between the carbon chain and cyclopropane are minimized (notice that the C–H bond, rather than the C–C bond, bisects the cyclopropane “a” bond) and (2) the C–Br and cyclopropane C–C bonds are disposed such that the reaction occurs via an anti-elimination (this is an important point that we will revisit shortly). Continuing with the synthesis, 40 was converted to iodide 46 using a Finkelstein reaction. Alkylation of heptan-3,5-dione with 46 provided 47. Chlorination of 47 followed by deacylation of the resulting non-enolizable 1,3-diketone gave 48. Treatment of 48 with methylmagnesium chloride gave 49 contaminated with a small amount of the diastereomeric chlorohydrin. A Williamson ether synthesis gave 1 as a racemic mixture. The stereochemical course of the Grignard addition reaction could be rationalized using either the Cornforth or Felkin-Ahn models for asymmetric induction.5,6 The FelkinAhn model is shown here using a truncated version of 48 (see conversion of 53 → 54). This is an “early application” of principles in acyclic diastereoselection to a target-oriented synthesis problem.
Nonselective Trisubstituted Olefin Synthesis
Disubstituted Olefin Synthesis
R
R
Br
R’
57
56
48% HBr 58
59
Me favored
61
60 OH
OH Ha Hb
R
64
OH Me
ZnBr 2
Me Br
O
48% HBr
67
R
Br
O 68 81% (96:4)
Me Me
(97)
Br
OH
66 73% (>97% pure)
Me Br
R
OH
H
65 Me
(3)
H
Me R
Br
63
62
Me
Stereoselective Trisubstituted Olefin Synthesis
Me
ZnBr 2
Bu
Br
Bu
Me
Stereoselective Trisubstituted Olefin Synthesis Me OH
For another CJH synthesis see: Mori, K. "Synthesis of Compounds with Juvenile Hormone Activity - XII. A Stereoselective Synthesis of 6-Ethyl-10-methyldodeca5-tr ans,9-cis-dien-2-one, A Key Intermediate in the Synthesis of C18-Cecropia Juvenile Hormone" Tetr ahedr on, 1972, 28, 3747.
R
Bu
R
R
Cecropia Juvenile Hormone-6
Nakamura, H.; Yamamoto, H.; Nozaki, H. Tetr ahedr on Lett . 1973, 111
OH
Page 456
Brady, S. F.; Ilton, M. A.; Johnson, W. S. J. Am. Chem. Soc. 1968, 90, 2882.
(3)
Ha
Br
Me
Br
Bu
10:57 AM
Hb
(1)
R
Br
Organic Synthesis via Examination of Selected Products
Ha
Ha
R
R’
OH
OH
R
R
Br
OH
12/21/2010
48% HBr
b1026
OH R
Organic Synthesis via Examination of Selected Natural Products
Julia, M. Bull. Soc. Chim. Fr. 1960, 1072; 1961, 1849.
b1026_Chapter-11.qxd
456
Julia Olefin Synthesis Substituent Effects
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 457
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
457
Cecropia Juvenile Hormone-6 A number of variations of the Julia olefin synthesis are summarized in CJH-6. Substrates of type 56 provide E-olefins (57) with high selectivity. The rationale is the same as that used to explain selectivity in the Johnson synthesis of Cecropia Juvenile Hormone (Hb bisects cyclopropane rather than R). We have already seen that substrates of type 63 give trisubstituted olefins of type 64 with excellent stereoselectivity. Substrates of type 58, however, do not give good stereoselectivity because there is little size differential between R and R′. Thus, substrate 60 gives a 1:3 mixture of 61 and 62, respectively. Finally, epoxides (rather than bromides or alcohols) can serve as the initiating group in this process to provide interesting trifunctional products (halide, olefin, alcohol) with good control of olefin stereochemistry. For example, 65 and 67 can be converted to 66 and 68, respectively, under typical Julia olefin synthesis conditions.
TsO
OMe
6
O H R2
2
Cecropia Juvenile Hormone (1)
69
fragmentation
anti-elimination
OTs
OH
O
O
O
KOH, MeOH, ∆
OH
p-TsOH, PhH, ∆
NaBH4 , EtOH, 5 °C O
PVK 75
88% alkylation from least hindered face with 95% selectivity
77
OH
DHP, H +
OTHP 1. Hydrolysis
OH
78 O
67%
O 76
OTHP 1. KO-t -Bu (5 eq) 79
HO 81
82
70
74
73
72
O
HO
O
O
71
O
6
R2
O
OH
OH
1
2
74%
2. Li(O-t -Bu) 3AlH ∆, THF
O
2. MeI 80
62% from enedione
Cecropia Juvenile Hormone-7
O
Page 458
OH
TsO
5 4
5
4
10:57 AM
2
R1
12/21/2010
3
1
3
Grob
Organic Synthesis via Examination of Selected Products
OMe
O
R1
O
b1026
The Syntex approach relies on a series of fragmentation reactions (Grob fragmentations) that unravel a bicyclic ring system (perhydroindan) into the target acyclic structure. The "retrosynthetic analysis" leading to this route requires recognition that fragmentation of 2,2-disubstituted cycloalkane-1,3-diols can afford ketonic trisubstituted olefins with control of olefin geometry, a substructure of use as an intermediate in any projected synthesis of CJH. The Syntex synthesis is follows an ingenious design.
Organic Synthesis via Examination of Selected Natural Products
Zurfluh, R.; Wall, E. N.; Sidall, J. B.; Edwards, J. A. "Synthetic Studies on Insect Hormones. VII. An Approach to Stereospecific Synthesis of Juvenile Hormones" J. Am. Chem. Soc. 1968, 90, 6224. For full papers see: Henrick, C. A.; Schaub, F.; Siddall, J. B. "Stereoselective Synthesis of the C-18 Cecropia Juvenile Hormone" J. Am. Chem. Soc. 1972, 94, 5374-5378. Anderson, R. J.; Henrick, C. A.; Sidall, J. B.; Zurfluh, R. "Stereoselective Synthesis of the Racemic C-17 Juvenile Hormone of Cecropia" J . Am. Chem. Soc. 1972, 94, 5379-5386.
O
b1026_Chapter-11.qxd
458
The Syntex-Zoecon Route to CJH
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 459
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
459
Cecropia Juvenile Hormone-7 When the mechanisms by which insects develop and communicate (via pheremones) were first being elucidated, the economic potential of controlling insect populations through manipulation of these processes was recognized. In part for these reasons, Zoecon (the equivalent of a start-up company nowadays) and Syntex (a non-profit research institute at the time) developed a synthesis of CJH (1). This synthesis relied on a series of fragmentation reactions (Grob fragmentations)7 that unravelled a bicyclic ring system (perhydroindan) into the target acyclic structure. The “retrosynthetic analysis” leading to this route required recognition that fragmentation of 2,2-disubstituted cycloalkane-1,3-diols would afford ketonic trisubstituted olefins with control of olefin geometry. For example, tosylate 69 would be expected to provide 70. This ingenious design led to development of the famous Syntex-Zoecon route to CJH. The plan for CJH was to prepare fragmentation substrate 71. It was expected that this would fragment to provide 72, which was to be manipulated to provide 73. Fragmentation of 73 was to then provide 74, which would be converted to 1 following the path developed by Trost (CJH-2). Before we look at the details of the synthesis, let’s relate this fragmentation to something simple. Consider the first olefin synthesis you are likely to have learned, dehydrohalogenation of an alkyl halide. Dehydrohalogenations are anti-eliminations (with rare exceptions). The stereochemistry of the resulting olefin is a function of the vicinal stereochemistry of the dehydrohalogenation substrate, and the conformation(s) in which the leaving group and β-hydrogen can adopt an anti-relationship. Work some of the problems presented below to test your understanding of this reaction. The fragmentation of 69 to 70 can also be regarded as an anti-elimination. The leaving group is a tosylate and the βsubstitutent is a carbon atom. But this carbon atom has electronic properties that are not unlike the properties of the β-hydrogen in a dehydrohalogenation (it becomes electron deficient in the elimination). Carrying this analogy further, the Julia olefin synthesis can be viewed as a “fragmentation” reaction … the C–X bond ionizes, the adjacent σ-bond is cleaved with formation of an olefin, and the remaining cation (or partially charged carbon) is captured by a nucleophile. It is mechanism that unifies these approaches to stereoselective olefin synthesis. The synthesis began with cyclopentan-1,3-dione 75. A Robinson annelation gave 77 via intermediate trione 76. Reduction of the more electrophilic of the two ketones gave 78, which was protected as tetrahydropyranyl ether 79. Deconjugative alkylation of 79 gave 80. Hydrolysis of the THP protecting group and reduction of the ketone (pseudo-axial delivery of hydride) provided 81/82.
m-CPBA, CH2 Cl2
LiAlH 4
HO
O
OH p-TsCl (5 eq)
HO
TsO
OH
pyridine
OH
NaH, THF, rt
2. MeLi, Et2 O
pyridine
O
HO
HO
86
73
85
95%
OH
1. DHP, H +
3. Deprotection
57%
O 72 100%
OH O TsO
OH
Several
O CO 2Me
Steps
87
1
74 80%
NaH, THF, rt OH OH +
O
Sometimes the best laid plans are undercut by nature
O 72 15%
88
85%
Cecropia Juvenile Hormone-8
CJH
Page 460
NaH, THF, rt
10:57 AM
HO
OH
p-TsCl
Organic Synthesis via Examination of Selected Products
OH
b1026
m-CPBA, Et2O
12/21/2010
71 89%
84 65%
83 50%
Organic Synthesis via Examination of Selected Natural Products
82
OTs
b1026_Chapter-11.qxd
OH
HO
460
OH
OH
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 461
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
461
Cecropia Juvenile Hormone-8 Oxidation of 82 with m-CPBA in dichloromethane provided 83. Reduction of the epoxide with lithium aluminium hydride gave 84, which was converted to fragmentation substrate 71. The fragmentation occurred in high yield to provide the desired geometrical isomer 72. Protection of the secondary alcohol, addition of methyllithium to the ketone, and removal of the THP protecting group gave diol 85. Conversion of the secondary alcohol to tosylate 73 was followed by the second fragmentation reaction to give 74, an intermediate in the Trost synthesis of CJH (1). It is notable that tosylate 87, derived from isomeric epoxide 86, available in turn by epoxidation of 82 in diethyl ether rather than dichloromethane, is also disposed to undergo fragmentation to 72. Unfortunately, when 87 was subjected to sodium hydride, a Williamson ether synthesis (87 → 88) became competitive with the desired fragmentation. Sometimes the best laid plans are undercut by nature. Regardless, the Syntex-Zoecon synthesis remains one of the least obvious, yet creative and successful, routes to Cecropia Juvenile Hormone.
89
CO 2Me
2. NaBH4 , EtOH -78 °C
90
TsO
pyridine, 0 °C
91 52% overall
CO 2Me 92
4h
LiAlH 4 LiAlH 4 (2 eq) NaOMe (4 eq), ∆
1.
Li
CH 2OTHP
HMPA, 0 °C, 3h
O
p-TsCl
OH 95
96
2. MeOH, H +
30%
pyridine
OTs 94
OH 93 65% overall
I2
I
1. PBr 3 (1.2 eq), Et2O
Et2CuLi OH
97
Et2O -30 °C
65% [6% of regioisomer (Z)]
OH 78%
98
2. LiH 2C
R
2. I2
SiMe3 99
AgNO3 ; KCN n-BuLi; CH 2O
R = SiMe3 R=H R = CH 2OH
80% overall For olefin methodology see J. Am. Chem. Soc. 1967, 89, 4245
Cecropia Juvenile Hormone-9
1. LiAlH4 -NaOMe THF
100
3. Me2 CuLi
Page 462
R
Al
10:57 AM
-33 °C
p-TsCl (1.1 eq)
HO
Organic Synthesis via Examination of Selected Products
NH3 -THF
1. O3 (1 eq), -78 °C MeOH-Me2S (10:1)
12/21/2010
OMe
b1026
Li (5 eq) t-amyl alcohol (1 eq)
Organic Synthesis via Examination of Selected Natural Products
Corey, E. J.; Katzenellenbogen, J. A.; Gilman, N. W.; Roman, S. A.; Erickson, B. W. "Stereospecif ic Total Synthesis of the dl-C18 Cecropia Juvenile Hormone" J . Am. Chem. Soc. 1968 , 90, 5618. The purpose of this work was to prepare CJH, to develop a new iterative approach to stereoselective trisubstituted olefins, and to illustrate the use of other new synthetic methods. We will skip the "analysis" and simply proceed through the synthesis.
OMe
b1026_Chapter-11.qxd
462
Synthesis-Driven Methodology Development
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 463
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
463
Cecropia Juvenile Hormone-9 The purpose of the next synthesis was to develop a new, iterative, and stereoselective approach to trisubstituted olefins, and also to illustrate the use of other new synthetic methods. We will skip the “analysis” and go directly to the synthesis, which runs from “left-to-right”. The first stage of the synthesis called for preparation of alcohol 93. The strategy used to prepare 93 involved starting with all carbons intact in the form of p-methoxytoluene (89). Birch reduction of 89 provided 90. Regioselective ozonolysis of the more electron rich olefin, followed by a reductive work-up, gave 91. Appropriate adjustment of oxidation states at the terminal carbons provided 93. This portion of the synthesis uses a ring to set trisubstituted olefin geometry. This approach to controlling olefin stereochemistry is less common nowadays, but is still occasionally used. Conversion of 93 to tosylate 94, followed by an acetylide displacement to give 95, set the stage for application of the new trisubstituted olefin synthesis. It had been known for some time that reduction of propargylic alcohols with lithium aluminum hydride, followed by hydrolysis of the reaction mixture, gave E-allylic alcohols with excellent stereoselectivity. Corey and Katzenellenbogen found that, under appropriate conditions, a presumed intermediate vinylalanate (96) could be captured by iodine to give trisubstituted iodoalkene 97 with high stereoselectivity. Furthermore, they were able to couple iodoalkenes with lithium dialkylcuprates with retention of stereochemistry to provide trisubstituted alkenes. In the CJH synthesis, 97 was converted to 98 using Et2CuLi. We will see variations of this coupling strategy again in the next chapter. Chain extension of 97 to 100 (R=CH2OH), using organolithium 99 as a key reagent, set the stage for repetition of this reaction sequence. Repetition of the reduction-iodination-coupling sequence provided allylic alcohol 101 (see CJH-10).
101 53%
CJH (1)
CO 2Me
2. NaCN, MeOH AcOH, MnO2
2. i-PrONa, i-PrOH 30 min, 0 oC
2
52%
2. RCHO, - 78 °C
-25 °C
OHC
HO
R
103
2. (CH2 O) n
25 °C
OTHP
R
R CO 2Me
OTHP 50%
1. s-BuLi
104
102
1
105
Ph 3P O
CO 2Me O
106
R = Me, H
107
R = Me, H
This is a variation of the Schlosser modification of the Wittig reaction [Schlosser, M.; Coffinet, D. Synthesis 1971, 380 and Sy nthesis 1972, 575]. Vedejs has published interesting work that provides a rational for the stereochemical and regiochemical course of the key reaction: Vedejs, E.; Snoble, K. A. J. "Direct Observation of Oxaphosphatanes from Typical Wittig Reactions" J. Am. Chem. Soc. 1973, 95, 5778.
Cecropia Juvenile Hormone-10
Page 464
0 °C
10:57 AM
O
1. n-BuLi, -78 °C
12/21/2010
PPh3 I
Organic Synthesis via Examination of Selected Products
Corey, D. J.; Yamamoto, H.; "Simple Stereospecific Synthese of C 17 - and C18-Cecropia Juvenile Hormones (Racemic) from a Common Intermediate" J. Am. Chem. Soc. 1970, 92, 6636.
b1026
Modification of the Schlosser Modification of the Wittig Reaction
Organic Synthesis via Examination of Selected Natural Products
For oxidation methodology see J. Am. Chem. Soc. 1968, 90, 5616
CO 2Me
b1026_Chapter-11.qxd
OH
464
1. NBS (1.1 eq), DME-H2 O
1. MnO 2, hexane
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 465
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
465
Cecropia Juvenile Hormone-10 Oxidation of 101 to ester 2 and epoxidation of the sterically most accessible olefin, completed the synthesis of CJH. The conversion of 101 to 2 is notable, and relies on the selective oxidation of allylic alcohols to aldehydes by manganese dioxide. The sequence of events is: (1) oxidation of 101 to the corresponding aldehyde, (2) cyanohydrin formation, (3) oxidation of the cyanohydrin to an acyl cyanide and (4) methanolysis of the acyl cyanide to provide the ester. Another synthesis of CJH eminating from the Corey labs involved a modification of the Schlosser modification of the Wittig Reaction. Whereas Wittig reactions between non-stablilized phosphoranes and aldehydes usually provide Z-olefins, Schlosser reported a variation of this reaction that provided E-olefins. His group showed that when non-stabilized phosphoranes are reacted with aldehydes in the presence of lithium halides (as opposed to the salt-free conditions that favor Z-olefins), followed by deprotonation of the intermediate adduct, and reprotonation of the resulting β-oxido ylid, E-olefins are obtained. Corey and Yamamoto reported a variation of this reaction in which the intermediate β-oxido ylid was treated with formaldehyde, resulting in formation of trisubstituted allylic alcohols. For example, 102 reacted sequentially with aldehyde 103 and formaldehyde to provide 105. This material was then carried on to CJH intermediate 106, and the CJH derivative where R=H.
b1026_Chapter-11.qxd
466
Still Synthesis of CJH Still, W. C.; McDonald, J. H. III; Collum, D. B.; Mitra, A. "A Highly Stereoselective Synthesis of the C18 Cecropia Juvenile Hormone" Tetr ahedr on Lett. 1979, 593-594.
CO 2Me
2
1
X
X
R
R
n-BuLi
KH
H
112 Bu3 SnCH2 I
OH
O
CH 2SnBu 3
O
When the ball is an n-butyl group and R = Me, the stereoselectivity is greater than 95%. When R = H the selectivity nearly vanishes (Z:E = 60:40). Similar reduction in selectivity is observed when the Me group is moved to the olef in terminus. An early transition state is proposed.
113
OH 114
H
R O
H 115
R
H
1. t-BuLi Br
Br 116
H
H
Steric Interaction
2-propenyllithium OHC
O
R
111
110
109
CH 2
2. RCHO
OH 117 93%
OH
1. KH, THF OEE OH
(f rom geraniol)
Cecropia Juvenile Hormone-11
OH 118
2. Bu3 SnCH 2 I
Page 466
R R
10:57 AM
Still, W. C.; Mitra, A. "A Highly Stereoselective Synthesis of Z-Trisubstituted Olefins via [2,3]-Sigmatropic Rearrangement. Pref erence for a Pseudoaxially Substituted Transition State" J. Am. Chem. Soc. 1978, 100, 1927. For additional reading see: Still, W. C.; Sreekumar, C. " α-Alkoxyorganolithium Reagents. A New Class of Configurationally Stable Carbanions for Organic Synthesis" J . Am. Chem. Soc. 1980, 102, 1201-1202.
Organic Synthesis via Examination of Selected Products
108
Methodology
b1026
O
12/21/2010
Me
CO 2Me
Organic Synthesis via Examination of Selected Natural Products
Me CO 2Me
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 467
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
467
Cecropia Juvenile Hormone-11 A conceptually interesting approach to the synthesis of CJH involves the reaction of allylic substrates of type 108 with methyl carbanion equivalents in an SN2′ manner. Several syntheses have followed this approach. The StillCollum-Mitra synthesis used enforced SN2′ methodology, a 2,3-sigmatropic rearrangement of allylic alcohols (108 where X=OH). The “big questions” in this approach were (1) what would be used as a “CH3–” equivalent and (2) would the desired olefin stereocontrol be achieved? The basic methodology is illustrated with structures 109–115. The notion was that an allylic alcohol of type 109 would be converted to 110, a precursor of carbanion 111. Carbanion 111 would undergo 2,3-sigmatropic rearrangement from either conformation 112 or 113 to provide homoallylic alcohol 114 or 115, respectively. The hydroxyl group would later be converted to a hydrogen to establish the necessary equivalency. The plan worked remarkably well with substrates of type 109. When R and the “ball” were methyl and n-butyl groups respectively, the ratio of 114:115 was greater than 95:5. It was suggested that the rearrangment occurred via an early transition state in which steric interactions between the methyl and butyl substitutents were minimized (112 lower energy than 113). In accord with this suggestion, substrates where R=H gave nearly equal mixtures of Z- and E-olefins. In the application to CJH, aldehyde 116 (prepared as described on CJH12) was converted to 117. Metallation of 117 followed by addition of the vinyllithium reagent to the appropriate aldehyde (prepared as described on CJH-12) gave bis-allylic alcohol 118. Derivatization of 118 gave bis-sigmatropic rearrangement substrate 119.
O SnBu3
OH
n-BuLi
1. TsCl, pyridine (93%) OEE
SnBu 3
OH
2. LiAlH4 , Et2O (98%)
121
3. AcOH, H2 O, 45 °C (92%)
1. LiAlH4 (95%)
t -Bu-O 2.
122
Br
Br
2. CrO3 -2 pyr (45%)
O
(At least 95% pure by GC and 13C NMR)
Br 116
123
Br
OHC
O3 ; Me 2S OEE
OHC
125
geraniol
OH
54% yield at 77% conversion 126
124
Strategically Related Syntheses van Tamelen, E. E.; McCormick, J. P. "Synthesis of Cecropia Juvenile Hormone from tr ans, tr ans-Farnesol" J. Am. Chem. Soc. 1970, 92, 737-738.
1. LiNEt2 (5 eq) PhH, ∆
1. m-CPBA OAc farnesyl acetate 127
2. K2CO3 , EtOH
O
O 128
OH
OTr
2. Ph3 CCl, pyr OH Rickborn Reaction syn-elimination
Cecropia Juvenile Hormone-12
OH 129
Page 468
CH2 =CHOEt, H+ OH
10:57 AM
O
1. LDA
Organic Synthesis via Examination of Selected Products
CH3
b1026
CJH
12/21/2010
120 79%
Organic Synthesis via Examination of Selected Natural Products
119
Synthesis of Components
t -Bu-O
b1026_Chapter-11.qxd
O
468
OH OEE
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 469
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
469
Cecropia Juvenile Hormone-12 Metallation of 119 followed by the double 2,3-sigmatropic rearrangement provided bis-homoallylic alcohol 120. The aforementioned “CH3–” equivalency was established by converting the hydroxymethyl groups to methyl groups (via reduction of the derived tosylates). Hydrolysis of the ethoxyethyl protecting group gave 121 which was converted to CJH following established procedures. A strategically related, although tactically quite different, synthesis of CJH had been reported by the VanTamelen group (Stanford) some years earlier. In this approach farnesyl acetate (127), a commercially available sesquiterpene, was converted to bis-epoxide 128. The inductive electron-withdrawing effect of the acetoxy group presumably helped with regioselectivity in this transformation. The bis-epoxide was then converted to bis-allylic alcohol 129 (compare with 118 on CJH-11) using a reaction developed by Rickborn, followed by selective protection of the primary alcohol as a trityl ether.8
OTr
OTr Cl
Cl
130
CJH 3. NBS, H 2 O 4. NaOMe
131
132
O
OH
H HO 134
73%
syn-elimination
- Me 2NH
H OTr OH
OH
O
O
O
138
O OH 139
H O 137
136 CJH
OH
OH
OH
OH 140
OTr OH
- CO2
O
Me2 N
Diastereomeric diols give diastereomeric epoxides, but this is of no consequence to the stereochemical course of the overall transformation.
Choice of one diastereomer is arbitrary.
135 80%
133
Minimize A1,2 Strain
∆
OTr
OTr 58%
Cecropia Juvenile Hormone-13
141 33%
Page 470
PhH
H HO
10:57 AM
t-BuOOH
Me2 NCH(OMe)2
Bu2 CuLi
VO(acac)2
12/21/2010
The Methodology
Organic Synthesis via Examination of Selected Products
Tanaka, S.; Yamamoto, H.; Nozaki, H.; Sharpless, K. B.; Michaelson, R. C.; Cutting J. D. "Stereoselective Epoxidations of Acyclic Allylic Alcohols by Transition Metal-Hydroperoxide Reagents. Synthesis of dl-C 18 Cecropia Juvenile Hormone f rom Farnesol" J. Am. Chem. Soc. 1974, 96, 5254.
b1026
Organic Synthesis via Examination of Selected Natural Products
The sequence produces a mixture of f our olef in geometrical isomers in nearly equal amounts (one of which is CJH). Thus, the cuprate chemistry is not stereoselective.
H HO
b1026_Chapter-11.qxd
1. HCl, THF 2. MnO 2, MeOH KCN
Me2 CuLi
2. LiCl
470
1. TsCl, pyridine
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 471
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
471
Cecropia Juvenile Hormone-13 Diol 129 was converted to the corresponding bis-chloride (130) via an intermediate bis-tosylate. Treatment of 130 with lithium dimethylcuprate gave 131 via a double SN2′ reaction. This reaction, however, gave a nearly statistical mixture of the four possible geometrical isomers and thus, upon completion of the synthesis, a mixture of products was obtained that included some CJH. Therefore, the tactics used by the Still group for accomplishing this double SN2′ strategy were better than those employed by the VanTamelen group. Another methodology-driven synthesis of CJH that is strategically related to the Still and VanTamelen syntheses was reported by the Sharpless group. This method relies on a trisubstitued olefin synthesis that revolves around a diastereoselective epoxidation reaction developed in the Sharpless laboratories. Sharpless had shown that allylic alcohol 132 underwent a highly diastereoselective epoxidation to give 133 upon exposure to tert-butyl hydroperoxide in the presence of an appropriate vanadium catalyst. This reaction presumably results from the conformation of 132 that minimizes 1,2-allylic strain (between the methyl and butyl groups for example). Covalent binding of both the alcohol and tert-butyl hydroperoxide to the vanadium result in directed delivery of “oxygen” to the expected face of the olefin. Treatment of 133 with lithium dibutylcuprate gave 134 (epoxide opening at primary carbon, rather than tertiary carbon). Treatment of 134 with N,N-dimethylformamide dimethyl acetal gave 135. This overall synelimination of vicinal hydroxyl groups is imagined to occur by conversion of 134 to amide acetal 136, α-elimination of dimethylamine to provide carbene 137, and chelatropic extrusion of carbon dioxide to give 135. This method was applied to a CJH synthesis by first preparing 139 (from 129). Reaction of this bis-epoxide with lithium dimethylcuprate gave 140 and a bis-elimination reaction gave 141 (same as 131), an intermediate in the VanTamelen synthesis of CJH.
O
O
OMe
2. dehydrate
O
1. n-BuLi, DABCO
143
S
S
2. epoxide
146
143
1. n-BuLi, DABCO
3. SOCl2 pyridine
145
S OTHP
OTHP
147 72% Br
148 60%
145
1. Hydrolysis 2. Li, EtNH2 , -70 °C SH
SH
CJH OAc
144
OH
2. W-2 Raney-Ni
150 55%
149
Another Iterative Approach
EtCu H 3C
Cu
H 151
152
CO2
2 Steps
again CO 2H
X X = CO2H 153 X = CH2 Cl
154
and so on
155
Chuit, C.; Cahiez, G.; Normant, J.; Villieras, J. "Synthese Stereospecifique Recurrente de Structures Apparentees a Celle de l’Hormone Juvenile de Hyalphora Cecropia" T etr ahedr on 1976, 32, 1675-1680.
Cecropia Juvenile Hormone-14
Page 472
O
S
1. Ac2 O 65%
144
2.
Me2 S=CH 2
S
Br
10:57 AM
S
OR
O
142
1. MeMgI
S
Organic Synthesis via Examination of Selected Products
1
S
S
12/21/2010
S
b1026
S OMe
Organic Synthesis via Examination of Selected Natural Products
Kondo, K.; Negishi, A.; Matsui, K.; Tunemoto, D.; Masamune, S. "A New Approach to the Stereospecific Total Synthesis of Racemic Cecropia Juvenile Hormone" J. Chem. Soc., Chem. Commun. 1972, 1311-1312. For a closely related approach see: Demonte, J.-P.; Hainaut, D.; Toromanoff, E. "Sur une nouvelle synthese, stereospecifique, de l’hormone juvenile en C18 Hyalophora cecropia" Comptes Rendus Acad. Sci. Paris 1973, 277, 49-51. O
b1026_Chapter-11.qxd
472
Set Olefin Stereochemistry in a Ring
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 473
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
473
Cecropia Juvenile Hormone-14 Two more syntheses of CJH are presented here. The first synthesis relies on setting olefin stereochemistry by incorporating it into a ring (recall the first Corey synthesis of CJH shown on CJH-9). The idea was to use a sulfurcontaining heterocycle to establish double bond geometry and then reduce the two carbon-sulfur bonds to liberate the acyclic olefin. Thus, it was hoped that 142 would serve as a precursor of 1 that could be stitched together from 143–145. In the forward direction, tetrahydrothiapyrone 146 served as the precursor of both 143 and 144. Metallation of 143 (an allylic sulfide) using n-butyllithium, followed by a reaction of the derived carbanion with 144, gave 147 after dehydration of the intermediate tertiary alcohol. Metallation of 147 followed by alkylation of the resulting anion with bromide 145, provided 148. Hydrolysis of the THP ether followed by reduction of the allylic C–S bonds with lithium in diethylamine, gave bis-sulfide 149. This reaction would have been expected to proceed via allyic radical and/or anion intermediates and thus, loss of olefin regiochemistry might have been anticipated, but the reaction seems to have proceeded without a major problem. Esterification of the alcohol (and most likely the thiols) and reduction of the homoallyic C–S bonds with Raney-Ni, gave 150. The synthesis of CJH (1) was completed in the usual manner. The final CJH synthesis we will examine, albeit briefly, is outlined using structures 151 to 155. This is another iterative approach (from the Normant group in France) that uses syn-additions of organometallics to alkynes to establish olefin stereochemistry. This is an approach to olefin synthesis that has gained popularity in recent years, with a variety of reagents having been developed to “carbometallate” alkynes.9 The Normant synthesis began with a reaction between propyne (151) and ethyl copper to provide vinylic organometallic 152. Carboxylation of 152 took place with retention of olefin geometry to provide carboxylic acid 153 (X=CO2H). The acid was converted to the corresponding allylic chloride, which was converted to alkyne 154 using methodology that resembles that used by Corey in his CJH synthesis (CJH-9). Repetition of the sequence provided 155 and so on to CJH. In closing this chapter I point out that whereas CJH (1) has been used as the context for our discussion of olefin synthesis, the lasting value of the research presented here is surely the methodology developed in pursuit of CJH (or illustrated by application to CJH syntheses). Furthermore I have not been anywhere near exhaustive in presenting olefin syntheses. We will see more in the next chapter. I also hope you recognize that the olefin syn-
b1026_Chapter-11.qxd
12/21/2010 b1026
474
10:57 AM
Page 474
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
theses we examined in this chapter are greater in number than they are in terms of strategy. To summarize, we have seen methods that set stereochemistry in β-elimination reactions (Julia synthesis, Sharpless synthesis, use of Wittig reactions), by converting cyclic olefins to acyclic olefins (ringopening reactions such as those used by Corey and Masamune), by stereocontrolled SN2′ reactions (Johnson-Faulkner, Still), and by stereocontrolled addition reactions of alkynes (Corey and Normant). As you learn more about olefin synthesis, you will surely come across other strategies, or other tactics, that accomplish olefin synthesis using the broad strategies presented in this chapter.
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 475
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
475
References 1. Eisner, T.; Meinwald, J. “The Chemistry of Sexual Selection” Proc. Nat. Acad. Sci. (USA) 1995, 92, 50–55. Karlson, P. “The Chemistry of Insect Hormones and Insect Pheromones” Pure Appl. Chem. 1967, 14, 75–87. Karlson, P. “Chemie und Biochemie Der Insektenhormone” Angew. Chem. 1963, 75, 257–265. 2. Houser, C. R.; Hudson, B. E. Jr. “The Acetoacetic Ester Condensation and Certain Related Reactions” Organic Reactions 1944, 1, 266–302. 3. Castro, A. M. M. “Claisen Rearrangement over the Past Nine Decades” Chem. Rev. 2004, 104, 2939–3002. Rhoads, S. J.; Raulins, N. R. “Claisen and Cope Rearrangements” Organic Reactions 1975, 22, 1–252. 4. Julia, M.; Julia, S.; Yu, T. S. “Stéréochemie des Alcool β,γ-Éthyléniques Issus de la Tranposition Homoalliques” Bull. Soc. Chim. Fr. 1961, 1849–1853. Julia, M.; Julia, S.; Guégan, R. “Préparation de Composés Terpéniques et Apparentés, à Partir de Méthyl Cyclopropyl Cétone” Bull. Soc. Chim. Fr. 1960, 1072–1079. 5. Cornforth, J. W.; Cornforth, R. H.; Mathew, K. K. “A General Stereoselective Synthesis of Olefins” J. Chem. Soc. 1959, 112–127. Evans, D. A.; Siska, S. J.; Cee, V. J. “Resurrecting the Cornforth Model for Carbonyl Addition: Studies on the Origin of 1,2-Asymmetric Induction in Enolate Additions to HeteroatomSubstituted Aldehydes” Angew. Chem. Int. Ed. 2003, 42, 1761–1765. 6. Ahn, N. T.; Eisenstein, O. “Theoretical Interpretation of 1,2-Asymmetric Induction. The Importance of Antiperiplanarity” Nouv. J. Chim. 1977, 1, 61–70. Cherest, M.; Felkin, H.; Prudent, N. “Tortional Strain Involving Partial Bonds. The Stereochemistry of the Lithium Aluminum Hydride Reduction of Some Simple Open-Chain Ketones” Tetrahedron Lett. 1968, 10, 2199–2204. 7. Grob, C. A.; Kiefer, H. R.; Lutz, H.; Wilkens, H. “The Stereochemistry of Synchronous Fragmentation” Tetrahedron Lett. 1964, 6, 2901–2904. 8. Rickborn, B.; Thummel, R. P. “Stereoselectivity of the Base-Induced Conversion of Epoxides to Allylic Alcohols” J. Org. Chem. 1969, 34, 3583–3586. 9. For two representative reviews see: Negishi, E. “Bimetallic Catalytic Systems Containing Titanium, Zirconium, Nickel and Palladium. Their Applications to Selective Organic Syntheses” Pure Appl. Chem. 1981, 53, 2333–2356. Fallis, A. G.; Forgione, P. “Metal Mediated Carbometallation of Alkynes and Alkenes Containing Adjacent Heteroatoms” Tetrahedron 2001, 57, 5899–5913.
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 476
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
476
Problems 1. Propose a mechanism for the Horner-Wadsworth-Emmons reaction. The reaction of aldehydes with 10 provides 1,2-disubstituted olefins with high E-selectivity. Provide (or look up) a rationale for this observation. Then explain why the reaction of 3 with 10 shows little stereoselectivity. (CJH-2) 2. Rearrangement from the other chair-like conformation available to 21, 23 and 25 (not shown in CJH-3) results in formation of the minor products (Z-geometrical isomers). Provide 3-dimensional structures of these transition geometries (as in CJH-3) and indicate the steric interaction that is responsible for the increased stereoselectivity in the reactions of 23 and 25 relative to 21. (CJH-3) 3. Provide a mechanism for the conversion of 17 + 29 → 30. (CJH-3) 4. Provide the transition state structure of the allyl vinyl ether that results in the conversion of 33 + 34 → 35. (CJH-4) 5. Propose a practical synthesis of 42 and 43. (CJH-4) 6. Provide the structure of the minor epoxide obtained from the conversion of 48 → 1. (CJH-5) 7. Discuss the factors behind selection of the “reactive conformation” in the Felkin-Ahn model for conversion of 53 to 54. How would Cornforth have rationalized this stereochemical result? [For an interesting system where the Cornforth and Felkin-Ahn models predict different stereochemical results see Evans, D. A.; Siska, S. J.; Cee, V. J. “Resurrecting the Cornforth Model for Carbonyl Addition: Studies on the Origin of 1,2-Asymmetric Induction in Enolate Additions to Heteroatom-Substituted Aldehydes” Angew. Chem. Int. Ed. 2003, 42, 1761–1765] (CJH-5) 8. Provide an explanation for the regiochemical course of the following reactions (indicate transition state structures). (CJH-7) X
KOEt, EtOH, ∆ + 1
+ 2c
2t
Ratio of Olefins (1 : 2t : 2c)
X
Yield of Olefins
Br
73%
31 : 51 : 18
OTs
50%
48 : 34 : 18
SMe 2
60%
87 : 8 : 5
NMe 3
71%
98 : 1 : 1
b1026_Chapter-11.qxd
12/21/2010 b1026
10:57 AM
Page 477
Organic Synthesis via Examination of Selected Products
Olefin Synthesis and Cecropia Juvenile Hormone
477
Brown, H. C.; Wheeler, O. H. “Steric Effects in Elimination Reactions. IX. The Effect of the Steric Requirements of the Leaving Group on the Direction of Bimolecular Elimination in 2-Pentyl Derivatives” J. Am. Chem. Soc. 1956, 78, 2199–2202. Also see Bartsch, R. A.; Bunnett, J. F. “Orientation of Olefin-Forming Elimination in Reactions of 2-Substituted Hexanes with Potassium tert-Butoxide-tert-Butyl Alcohol and Sodium Methoxide-Methanol” J. Am. Chem. Soc. 1969, 91, 1376–1382. 9. Provide the products expected from the following olefin-forming reactions. (CJH-7) Cl
EtONa
EtONa ?
? Cl
Huckel, W.; Tappe, W.; Legutke, G. “Elimination Reactions and Their Steric Course” Ann. Chem. 1940, 543, 191–230. OMs 1. BH 3-THF ? 2. NaOH, H 2 O, ∆
Marshall, J. A.; Bundy, G. L. “A New Fragmentation Reaction. The Synthesis of 1-Methyl-trans,trans-1,6-cyclodecadiene” J. Am. Chem. Soc. 1966, 88, 4291–4292. 10. Draw the conformation of 87 that is disposed to fragment to 72. (CJH-8) 11. Outline a synthesis of 93 using the Corey-Katzenellenbogen synthesis to set stereochemistry. How might 93 be prepared using a Julia olefin synthesis? Comment on stereochemical aspects of the Julia route to 93. (CJH-9) 12. Provide the structures of intermediates en route from 101 to 2. Explain the sequence of events that leads from I to II. (CJH-10) SO 3N a OH
M eO I
1 . D M S O, A c 2 O CO 2H 2 . N aO H, H 2 O 3 . n eutraliz e
M eO II
7 6%
Wuts, P. G. M.; Bergh, C. L. “The Oxidation of Aldehyde Bisulfite Adducts to Carboxylic Acids and their Derivatives with Dimethyl
b1026_Chapter-11.qxd
12/21/2010 b1026
478
10:57 AM
Page 478
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
Sulfoxide and Acetic Anhydride” Tetrahedron Lett. 1986, 27, 3995–3998. 13. Provide the structures of the four products expected from the reaction of 130 with lithium dimethylcuprate. (CJH-13) 14. An interesting olefin synthesis is presented below. To which of the aforementioned strategies does this belong? (CJH-14) C6H13CHO
Ni(COD)2 (10 mol%)
+ H3C
Ph + Et3SiH
Mes N
N Mes
(10 mol%)
THF (25-45 °C)
Et3 SiO
CH 3
C 6H 13
Ph H
82% (>98:2 regioselectivity)
Mahandru, G. M.; Liu, G.; Montgomery, J. “Ligand-Dependent Scope and Divergent Mechanistic Behavior in Nickel-Catalyzed Reductive Couplings of Aldehydes and Alkynes” J. Am. Chem. Soc. 2004, 126, 3698–3699.
O
17
28
O
18
26
28
25
18 t
17
c = cis t = trans
21
O
1
1
HO
O 1
O
O Original Structure
19
22
21
HO
O
25
O
22
23
c
11
11
O 7
(-)-Lasonolide A
10 9
OH
O 7
The enantiomer is 103 times less active in assays against several tumor cell lines.
Lasonolide-1
10 9
OH
Page 480
19
10:57 AM
O
12/21/2010
23
b1026_Chapter-12.qxd
OH 26
Organic Synthesis via Examination of Selected Products
OH O
b1026
It turns out that there were some mistakes in the original structure assignment. The actual structure is shown below. Thus, synthesis determined the stereochemistry at C28 (relative to other centers) and corrected the assignments of olefin geometry at C17-18 and C 25-26. It is interesting that the natural product was reported to be dextrorotatory (+), but by synthesis it was shown that biological activity actually resides in the levorotatory (-) enantiomer. We will not go through all of the syntheses of the many stereoisomers that were prepared en r oute to the first synthesis of lasonolide A, but will jump right into the first total synthesis performed by the Lee group in Korea: Lee, E.; Song, H. Y.; Kang, J. W.; Kim, E.-S.; Jung, C.-K.; Joo, J. M. "Lasonolide A: Structural Revision and Synthesis of the Unnatural (-)-Enantiomer" J . Am. Chem. Soc. 2002 , 124, 384-385. For a f ull paper see: Song, H. Y.; Joo, J. M.; Kang. J. W.; Kim, D.-S.; Jung, C.-K.;Kwak, H. S.; Park, J. H.; Lee, E.; Hong, C. Y.; Jeong, S. W.; Jeon, K.; Park, J. H. "Lasonolide A: Structural Revision and Total Synthesis" J. Org. Chem. 2003 , 68, 8080-8087. For work that indicates biological activity to reside in the (-)-enantiomer see: Lee, E.; Song, H. Y.; Joo, J. M.; Kang, J. W.; Kim, D. S.; Jung, C. K.; Hong, C. Y.; Jeong, S. W.; Jeon, K. "Synthesis of (+)-Lasonolide A: (-)-Lasonolide A is the Biologically Active Enantiomer" Bioor g. Med. Chem. Lett. 2002, 12, 3519-3520.
Organic Synthesis via Examination of Selected Natural Products
Lasonolide A provides a modern example of synthesis playing a role in natural product structure determination. There are other examples f or sure, but I have selected this one because of a personal interest in the problem, and because it also is a polyolef in that will let us continue our discussion of diastereoselective olefin synthesis. Lasonolide A is a marine natural product produced by a sponge (or microorganisms that live on the sponge ... these situations can be difficult to differentiate). The structure was assigned as shown below [Horton, P. A.; Koehn, F. E.; Longley, R. E.; McConnell, O. J. J. Am. Chem. Soc. 1994, 116, 60156016] based largely on NMR studies. The stereochemistry at C28 was not determined, nor was its absolute configuration. The natural product showed activity against several tumor cell lines with activity (IC 50 values) in the ng/mL region. It also inhibited cell adhesion in several cell lines. Thus, lasonolide A promptly became an attractive target for synthesis: (1) to establish the structure and (2) to provide more material f or biological evaluation. Since the isolation of lasonolide A, the lasonolides have grown into a small family of natural products [Wright, A. E.; Chen, Y.; Winder, P. L.; Pitts, T. P.; Pomponi, S. A.; Longley, R. E. "Lasonolides C-G, Five New Lasonolide Compounds from the Sponge For cepia sp." J . Nat . Pr od. 2004, 67, 1351-1355. The newer lasonolides dif fer in the C 23 sidechain and esterification of the C 22 hydroxymethyl group (laurate esters)
480
Total Synthesis and Structure Determination: A Modern Example
b1026_Chapter-12.qxd
12/21/2010 b1026
10:57 AM
Page 481
Organic Synthesis via Examination of Selected Products
A Recent Example of Structure Determination
481
Lasonolide-1 The next molecule we will consider is lasonolide A (1). Lasonolide A provides a modern example of synthesis playing a role in natural product structure determination. There are other examples for sure, but I have selected this one because of a personal interest in the problem, and because it also is a polyolefin that will let us continue our discussion of diastereoselective olefin synthesis. Lasonolide A (1) is a marine natural product produced by a sponge (or microorganisms that live on the sponge ... these situations can be difficult to differentiate). The structure was assigned based largely on NMR studies. The stereochemistry at C28 was not determined, nor was the absolute configuration of lasonolide A determined. The natural product showed activity against several tumor cell lines with activity (IC50 values) in the ng/mL region. It also inhibited cell adhesion in several cell lines. Thus, lasonolide A promptly became an attractive target for synthesis: (1) to establish the structure and (2) to provide more material for biological evaluation. Since the isolation of lasonolide A, the “lasonolides” have grown into a small family of natural products. The newer lasonolides differ in the C23 sidechain, and by virtue of esterification of the C22 hydroxymethyl group (laurate esters). It turns out that there were some mistakes in the original structure assignment. The actual structure is shown in Lasonolide-1 along with the original structure. Thus, synthesis determined the stereochemistry at C28 (relative to other centers) and corrected the assignments of olefin geometry at C17-C18 and C25-C26. It is interesting that the natural product was reported to be dextrorotatory (+), but by synthesis it was shown that biological activity actually resides in the levorotatory (−) enantiomer. Thus there is still some confusion regarding the absolute stereochemical details of the natural product (see Gephyrotoxin-2 in Chapter 9 for another example of this problem). We will not go through all of the syntheses of the many stereoisomers that were prepared en route to the first synthesis of lasonolide A, but will jump right into the first total synthesis performed by the Lee group at Seoul National University in Korea.
b1026_Chapter-12.qxd
482
First Synthesis of (-)-Lasonolide A from malic acid Outline of Plan
Wittig 23
c
19
18
t
A
25
O
O
1. Convergent approach calling for three pieces: the A-ring THP, the B-ring THP, and the hydroxyester side chain.
Julia-Lythgoe
17 15
22
2. Connect the two THP pieces by (a) Julia-Lythgoe olefin synthesis followed by (b) intramolecular Stille coupling.
21
O 1
11
O
10
B
Stille coupling 7
9
OH
Synthesis of Upper THP
1. BH3-Me2S (1.03 eq) NaBH4 (0.05 eq), THF
CO2Et
HO
2
EtO2C
HO
2. Bu2SnO (1.0 eq) PhH, ∆ (-H2O)
THF MgBr (3 eq)
1.
O
Me3Al/ MeNHOMe-HCl THF
MeO
72%
N
THF-MeOH
PMP = p-methoxyphenyl PMB = p-methoxybenzyl OBn EtO2C Br
OBn
O Si
BrCH2SiMe2Br
O 9
O
EtO2C
5
8
57% OMe
MeO OMe
CSA CH2Cl2
OBn 1. ethyl propiolate Et3N
HO
Et3N, 4-DMAP
HO
2. Et3B; NaBH4
Me
3. BnBr (2 eq), TBAI (2 eq), ∆
diethyl malate
HO
4
3
EtO2C
OBn
HO
OBn
HO
i-Bu2AlH
PMBO 7
2. DDQ, CH2Cl2-H2O
CH2Cl2
88%
97%
ROBn directs ionization?
Lasonolide-2
O*
PMP O
6
Page 482
OBn OBn
10:57 AM
3. Add the hydroxyester side chain by a Wittig reaction.
O
1
Still-Gennari
14
Organic Synthesis via Examination of Selected Products
HO
12/21/2010
26
b1026
28
Organic Synthesis via Examination of Selected Natural Products
OH O
b1026_Chapter-12.qxd
12/21/2010 b1026
10:57 AM
Page 483
Organic Synthesis via Examination of Selected Products
A Recent Example of Structure Determination
483
Lasonolide-2 Lasonolide A (1) has 9 stereogenic centers. Eight of these appear in two “clusters”, the tetrahydropyran (THP) rings labelled A and B. The ninth stereogenic center is isolated from the THP rings and appears at C28. Thus, lasonolide A demands a convergent strategy in which the absolute stereochemistry is set in pieces, and then the pieces are assembled. The three pieces selected by Lee appear on the next few pages. They include the A-ring (18), the B-ring (31) and the C28-containing side chain (37). You will notice that these pieces lack C1, C2 and C3. The selection of these pieces dictated the manner in which they were to be assembled. A Julia-Lythgoe-Kocienski olefin synthesis was to be used to connect the A- and B-ring pieces (construction of the C14-C15 olefin), and an intramolecular Stille coupling was to be used to close the macrocyclic lactone, after introduction of C1-C3.1,2 Finally, the hydroxyester side chain was to be introduced using a Wittig reaction to establish the C25-C26 olefin. A few other points are of interest before looking at the synthesis. Lasonolide A is an example of a family of natural products known as macrolides (macrocyclic lactones). We will see another example of such a natural product in Chapter 14. Whereas macrolides can often be assembled using a macrolactonization (intramolecular esterification) to close the ring, this approach was not followed by the Lee group, perhaps because this portion of the target looks sterically congested. What is the biosynthetic origin of lasonolide A? One can speculate that it is derived largely from “acetate”. There are a large number of natural products that are “polyacetates” in origin. They are derived from iterative condensation reactions (Claisen condensations) between acetic acid derivatives (for example acetyl coenzyme-A units) followed by adjustment of oxidation states along the chain. Fatty acids are biosynthesized in this manner. A “signature” of natural products that arise from this biosynthetic pathway is the appearance of odd (1,3 or 1,5 or 1,7) difunctional relationships within the molecule (usually between oxygens). In the case of lasonolide A, oxygens appear at C1, C7, C9, C11, C19, C21 and C23 (with a few extras added). We already have seen that such relationships can be constructed using carbonyl addition chemistry (Claisen and aldol condensations) (Chapter 6) and we will see that this bond construction plays a minor role in Lee’s synthesis of the B-ring. The A- and B-rings of lasonolide A have similar substitution patterns (2,3,4,6-tetrasubstituted THPs). If you remove the C10 substitutent from the B-ring and the C22 substitutents from the A-ring, they have the same substitution pattern, including relative stereochemistry. Therefore, it is not surprising that the same strategy was adopted to prepare both THP rings. Finally, the strategy used to prepare the THP rings involved free radical cyclization chemistry (look ahead to 9 → 11 and 25 to 26) originally
CO 2Et O
OBn O
EtO2C
Si
O
EtO2C
OBn
PhH, ∆, AIBN
3. 30% H 2O2 , KF KHCO3 , MeOH, THF HO
O
10
80%
Tamao-Fleming Oxidation
11
2. CSA, MeOH
O
CHO
78% N
LiN(SiMe 3) 2
S
PivO
OBn
O
O
1. (o-NO2)PhSeCN Bu3P, THF
OTBS
2. 30% H 2O2
1. OsO4 (cat), NMO
HO
3. H2 , Pd/C MeOH 4. Pyr-SO3 , Et3N DMSO
O
16
2. Me2 C(OMe)2 TsOH, acetone
PivO
OBn
1. HCl, MeOH
15
OTBS 83%
2. NaIO 4 3. NaBH4
14
A-ring unit ready f or macrocycle construction
E:Z = 12:1 1. TBAF OTBDPS
PivO
2. Ph3 P, DIAD S
17 O
O
O O
N
O S
O
SH
18 3. H2 O2 , (NH 4) 6Mo 7O24
O
67%
O 89%
Lasonolide-3
OTBS 13
NMO = N-methylmorpholine N-oxide
THF-HMPA (5:1)
O
HO
96%
O S (CH2) 3OTBDPS O
PivO
OBn O
N
Page 484
PivO
PivO
10:57 AM
1. TBSOTf (5 eq) 2,6-lutidine VanRheenan Oxidation
O
12
Organic Synthesis via Examination of Selected Products
6-endo competes with 5-exo radical cyclization because (in part) of long Si-O bond
OH 76%
b1026
Si
OBn O
12/21/2010
O
Br 9
1. LiBH 4, ether 2. PivCl, 4-DMAP CH 2Cl2, pyridine
Organic Synthesis via Examination of Selected Natural Products
Me 2Si
n-Bu3 SnH
O
PivO
OBn
b1026_Chapter-12.qxd
484
Piv = Me3 CC=O
b1026_Chapter-12.qxd
12/21/2010 b1026
10:57 AM
Page 485
Organic Synthesis via Examination of Selected Products
A Recent Example of Structure Determination
485
developed for other purposes by the Lee group. Thus, this approach also was dictated by the practitioner’s interest in developing and testing methodology. Let’s move to the synthesis of the A-ring (18 via 9). Recall that this strategy (and molecule) demands an enantioselective approach. The “chiral pool” approach was taken and the synthesis began with diethyl malate (2) (malic acid is an inexpensive dicarboxylic acid that can be isolated from apples, among other sources). Reduction of the ester proximal to the hydroxyl group gave a diol. Conversion of the diol to the corresponding bis-tin alkoxide, and alkylation of the less hindered primary alkoxide, gave 3. Treatment of 3 with the dimethylaluminum amide, derived from N,O-bis(dimethyl) hydroxylamine, gave 4, which reacted with isopropenylmagnesium bromide to afford a β-hydroxy ketone. These two reactions constitute a nice way of converting an ester to a ketone. Both transformations were developed by the Weinreb group (now at Pennsylvania State University).3,4 Chelation controlled reduction of the β-hydroxy ketone provided diol 5 with good stereoselectivity.5 Acetal formation, followed by reductive cleavage of 6 yielded 7. The regioselectivity of this reaction might be controlled by direction, by the benzyloxy group, of the Lewis acidic aluminum to the oxygen of the acetal marked with an asterisk. Treatment of 7 with ethyl propiolate gave vinylogous carbonate 8, and derivatization of the allylic alcohol provided 9. This set the stage for the featured free radical cyclization.
Lasonolide-3 Treatment of 9 with tri-n-butyltin hydride gave THP 11, presumably via sequential free radical cyclizations. The first cyclization follows a 6-endo course to provide 10, which undergoes a 6-exo cyclization via a chair-like conformation to ultimately provide 11 with the proper stereochemistry at the two newly formed stereogenic centers. Reduction of the ester, protection of the resulting primary alcohol, and a Tamao-Fleming oxidation, converted 11 to 12.6,7 The hydroxyethyl group was then degraded to a hydroxymethyl. This was accomplished using the Grieco-Sharpless procedure for formal dehydration of alcohols (13 → 14) followed by oxidative degradation of the vinyl group to an aldehyde and subsequent reduction to alcohol 15. Removal of the TBS protecting group was followed by formation of an acetonide, hydrogenolysis of the benzyl group, and oxidation of the resulting primary alcohol to provide aldehyde 16. Application of a variation of the Julia-Lythgoe-Kocienski olefin synthesis gave 17 with good control of olefin geometry.8 The synthesis of 18 was completed by removal of the silicon protecting group, displacement of the primary alcohol, and oxidation of the resulting sulfide to the sulfone oxidation state.
b1026_Chapter-12.qxd
486
Synthesis of B-Ring The substitution pattern of the B-ring is similar to that of the A-ring and thus, it is not surprising that the syntheses have a common f lavor. Ph
Ph
OH OBn
O
O OBn
2. PhCH(OMe) 2 CSA
2. CH 2=CHMgBr 21 69%
20
THF-MeOH
22 i-Bu2 AlH
n-Bu3SnH
O
EtO2C
EtO 2C
96%
1. H2 , Pd/C 2. TBSOTf, 2,6-lutidine 3. LiBH 4, Et2O
OBn
2. HCl, MeOH
Br 25 88%
26
OBn OH
83%
3. CBr 4, Ph3P
OBn
96%
OTBS 27
1. Pyr-SO3 , Et3N DMSO-CH2 Cl2 2. CrCl2 , CHI3 dioxane, THF
OBn
3. NaBH4 , MeOH 4. TBSCl, imidazole
80% 23
Z:E=15:1
OTBS O
OBn OH
E:Z = 8:1 I
HO
24
1. OsO4 , NMO acetone-H 2 O 2. NaIO 4
I
I
OTBS
OHC
MeO2C
O
1. CSA, MeOH
OTBS
2. Pyr-SO3 , Et3N DMSO-CH2 Cl2
51%
3. KN(SiMe 3) 2
O
(CF3CH2O) 2
P
O
2. MnO 2, CH 2Cl2 30
OTBS
91% OTBS
31
85%
O
28
1. i-Bu2AlH
CO2 Me
29
Lasonolide-4
Gennari-Still Olefination
B-ring unit ready f or macrocycle construction
Page 486
AIBN, PhH, ∆ OBn
O
OBn 1. HC C CO2 Et TBSO NMM
10:57 AM
OBn
Organic Synthesis via Examination of Selected Products
19
1. Et3 B; NaBH4
12/21/2010
Ph
O
OBn 1. Me Al, MeONHMe-HCl 3
N
O 2. BnOCH 2CHO
OH
b1026
N
O
O
O
1. Bu2 BOTf, Et3N THF, CH 2 Cl2
Organic Synthesis via Examination of Selected Natural Products
O
O
b1026_Chapter-12.qxd
12/21/2010 b1026
10:57 AM
Page 487
Organic Synthesis via Examination of Selected Products
A Recent Example of Structure Determination
487
Lasonolide-4 The substitution pattern of the B-ring is similar to that of the A-ring and thus, it is not surprising that the syntheses have a common flavor. The synthesis of the B-ring via a free radical cyclization called for the preparation of vinylogous carbonate 25. The starting point was a diastereoselective aldoltype condensation reaction between the boron enolate of imide 19 and 2-benzyloxyacetaldehyde. The conversion of 20 to 24 followed tactics closely related to the conversion of 3 to 7 (Lasonolide-2). Alcohol 24 was transformed to the corresponding vinylogous carbonate. Hydrolysis of the TBS protecting group and treatment of the resulting alcohol with triphenylphosphinebromine gave primary bromide 25. The key free radical cyclization proceeded smoothly to give 26 with surprisingly good stereoselectivity. The benzyl protecting group was removed by hydrogenolysis and replaced with a TBS group. Lithium borohydride was then used to reduce the ester to primary alcohol 27. Oxidation of the alcohol gave an aldehyde, which reacted with chromyl chloride and iodoform to provide vinyl iodide 28 along with a small amount of the Z-isomer. The TBS protecting group was removed with acid, and another Moffatt-type oxidation gave the corresponding aldehyde.9 The next stage of the synthesis called for introduction of the C12-C13 olefin. This was accomplished using the Gennari-Still modification of the HornerWadworth-Emmons (HWE) reaction.10 The normal HWE reaction of phosphonoacetates with aldehydes employs dimethyl or diethyl phosphonate anions.11 The resulting olefins have largely E-geometry. This is presumably a consequence of the greater thermodynamic stability of such olefins relative to their Z-counterparts (regardless of whether the olefin is di- or tri-substituted. The Gennari-Still modification employs 2,2,2-trifluoroethoxy groups (for example 29) and provides largely Z-olefins (for example 30). Although the reasons for this change in stereoselectivity are not well established, it is possible that the rate determining-step of this process is changed such that olefin geometry is set in the C–C bond-forming step. In other words, the carbonyl addition may normally be reversible and elimination rate-determining, while in the Gennari-Still method, the addition may be rate-determining due to the electron-withdrawing nature of the trifluoroethoxy groups. Regardless, this HWE modification is a widely used method for the synthesis of Z-α,βunsaturated esters. Moving forward, a reduction-oxidation sequence converted 30 to the desired B-ring unit 31.
PivO O O
PivO
O S
t
O 34
3. DCC, 4-DMAP
I
O
O
70%
OTBS
I
Bu 3Sn
O
OH
OTBS
40%
Pd2 (dba) 3 (10 mol%) i-Pr2 EtN (10 eq) N-methylpyrrolidinone (4mM)
OH
26
O
c
O
HO
O
OHC
t
25
1. Ph3 P=CHR (KHMDS) TBSO
O
O
PivO t
t
O
1. LiEt3BH (62%)
TBSO
O
1 O O
(-)-Lasonolide A
2. HF-pyridine THF
O
2. Pyr-SO3 , Et3N DMSO-CH2 Cl2
O
36
O 35
(88%)
4
O
62%
OTBS
OH
3
OTBS
60%
Synthesis of Side Chain Ylid
1. cyclohexanone BF3-Et2 O
OH HO2 C
CO2 Et 38
from malic acid
2. BH3 -Me 2S, THF (MeO3)B 3. TBSCl, imidazole CH 2Cl2
1. ROH, NaH, THF 2. TBSCl 3. HF-pyridine (42%)
O OTBS 39
PPh3 I O
O O
OH O
4. Ph3 P, I2 , imidazole 5. Ph3 P, CH 3CN (86%)
65%
Lasonolide-5
37
Page 488
O
10:57 AM
33
32
Organic Synthesis via Examination of Selected Products
Bu3Sn
12/21/2010
1. LDA 2. Aldehyde 31
O
2. TBSCl, imidazole
b1026
14
Organic Synthesis via Examination of Selected Natural Products
1. CSA, (CH 2 OH) 2
O
18
O
O
TBSO
15 21
O
PivO
E:Z = 20:1
t 22
N O
O
O
b1026_Chapter-12.qxd
488
Assembling the Macrocycle
b1026_Chapter-12.qxd
12/21/2010 b1026
10:57 AM
Page 489
Organic Synthesis via Examination of Selected Products
A Recent Example of Structure Determination
489
Lasonolide-5 Sulfone 18 and aldehyde 31 were coupled using a Julia-Lythgoe-Kocienski reaction to provide 32 with good stereoselectivity. The acetal was removed in an exchange reaction with ethylene glycol, the primary C22 hydroxymethyl group was protected, and the secondary C21 alcohol was acylated using acrylic acid derivative 33. An intramolecular Stille reaction of the resulting iodostannane (34) provided 20-membered ring lactone 35.2 The pivalate was reductively removed at C25 using lithium triethylborohydride. It is notable that the lactone was not disturbed, perhaps an indication of the aforementioned steric congestion. Oxidation of the primary alcohol then gave aldehyde 36. A Wittig reaction with the phosphorane derived from 37 gave the Z-olefin and removal of the C9 protecting group completed the synthesis of lasonolide-A (1). The starting material for the synthesis of 37 was 38, prepared in turn from malic acid, both enantiomers of which are commercially available. The choice of this starting material facilitated determination of C28 stereochemistry as both enantiomers could be used in this coupling reaction. Note that this Wittig reaction provided largely the Z-olefin geometry as expected (recall introduction of the “upper” prostaglandin side chain via a similar reaction).
t 15
OH
O O
(+)-Lasonolide A
3
2.
40
O
O
O
OH
O
CO2 iPr
O
42
2. Separate (72:28) 3. Recycle once
27:1
A
2. NaBH4 (96%) 3. H2 , Pd/C AcOH (89%)
1. TBSCl, imidazole 2. TESOTf, 2,6-lutidine 3. K2CO3 , MeOH
TBSO
O
O
O 46 Ph
1. Swern oxidation
2. BzONa, NMP 100 oC (97%)
O
2.
O 45
Ph 77% (91%ee)
O B O
CO2 iPr
48 TES = SiEt3
O Ph
O
TBSO
O N S ( CH 2)3S O Ph
N N
N
N
S
3. (NH4 )6 Mo7 O24-4H 2O H2 O2 , phosphate buffer at pH 7.6
49
43a
OH
CO2 iPr
vs
44
Bis-sulfone prepared from 1,3-propanediol in 3 steps
1. Dess-Martin Periodinane Pyridine-CH2 Cl2 2. KHMDS
OTES
82%
OH
1. I2 , K2 CO 3 CH 3CN (95%)
OH
O A
TESO
43
O Ph
O Ph
31:1
OH
O
O O S
O
43b
H-bonding ?
N
75% TESO OTES 50 1. TsOH-monohydrate MeOH
Lasonolide-6
O
2. TBSCl
Can you calculate the free energy dif ference between the diastereomeric dioxolanes?
Page 490
O
47
CO2 iPr
86% (78%ee)
1. O3, MeOH OH
O B O
OBz
O
OH
OH
1. PhCHO, CF 3CO2 H
21
41
OBz HO
OH
1. Swern oxidation
14 2
10:57 AM
HO 1
Organic Synthesis via Examination of Selected Products
O
12/21/2010
Synthesis of the A-Ring
O c
b1026
OH
Organic Synthesis via Examination of Selected Natural Products
Kang, S. H.; Kang. S. Y.; Kim, C. M.; Choi, H.-W.; Jun, H.-S.; Lee, B. M.; Park, C. M.; Jeong, J. W. "Total Synthesis of Natural (+)-Lasonolide A" A ngew. Chem. Int . Ed. 2003, 42, 4779. This synthesis f ollows the same bond disconnections (for assembling the macrocycle) with the exception that the f inal ring closure was to be accomplished by an intramolecular HWE reaction for construction of the C2-3 bond. The details for construction of the A-ring and B-ring also dif fer and the synthesis is shorter. For the full paper see: Kang, S. H.; Kang. S. Y.; Choi, H.-W.; Kim, C. M.; Jun, H.-S.; Youn, J.-H.; "Stereoselective Total Synthesis of the Natural (+)-Lasonolide A" Synthesis 2004, 1102-1114. O
b1026_Chapter-12.qxd
490
Second Synthesis of Lasonolide A
b1026_Chapter-12.qxd
12/21/2010 b1026
10:57 AM
Page 491
Organic Synthesis via Examination of Selected Products
A Recent Example of Structure Determination
491
Lasonolide-6 We will examine two more syntheses of lasonolide A (1). The Kang group (KAIST in Korea) decided to assemble the macrocycle by a route that used pieces 51 (A-ring), 59 (B-ring) and the C28 triethylsilyl ether of phosphonium salt 37. The C14-C15 bond was constructed using Julia-Lythgoe chemistry and the macrocycle was closed using an intramolecular HWE reaction to set the C2-C3 olefin. The synthesis uses an effective reaction sequence for introducing β-hydroxyaldehyde units on three occasions: (1) addition of an allyl group to an aldehyde followed by (2) oxidative cleavage of the terminal olefin to reveal the aldehyde carbonyl group. The construction of such units is important in the broad field of polyacetate-derived natural products synthesis (and a modification is important to the related field of polypropionate synthesis). Synthesis of the A-ring THP (51) began with acetal 40. The primary alcohol was oxidized using the Swern conditions (another Moffat-type oxidation)12 and the resulting aldehyde was reacted with chiral allylic boronate 41. This type of reagent was first developed in the laboratories of Bill Roush for the asymmetric allylation of aldehydes.13 In the current case, the yield was excellent and an 89:11 mixture of diastereomers (at the incipient C21) was obtained. Treatment of 42 with benzaldehyde in the presence of acid gave 43a along with diastereomeric acetal 43b. Separation of the mixture, and recycle of the minor acetal (43b), provided an 82% overall yield of 43. It is possible that intramolecular hydrogen bonding renders 43a more stable than 43b. Oxidation of the primary hydroxyl group of 43a, followed by another aldehyde allylation, gave 45. The allylation reagent (44) is the enantiomer of the reagent used in the conversion of 40 to 42. The two reagents (41 and 44) provide opposite absolute stereochemistry at the newly formed stereogenic center, a nice example of reagent-controlled asymmetric synthesis (see Prostaglandins-12). Treatment of 45 with iodine (iodoetherification) was followed by a nucleophilic substitution reaction to provide benzoate 46.14 The double bond was cleaved, the resulting aldehyde reduced, and the acetal was removed by hydrogenolysis to provide 47. The least hindered primary alcohol was protected as a TBS ether, the remaining alcohols were protected as TES ethers, and the benzoate was hydrolyzed to give 48. Oxidation of the alcohol was followed by a Julia-Lythgoe-Kocienski type reaction using sulfone 49, and the 2-thiobenzimidaxole was “armed” for the next Julia-Lythgoe-Kocienski by oxidation to the corresponding sulfone 50. Removal of all of the protecting groups and reprotection of the two primary alcohols gave A-ring substrate 51 (Lasonolide-7).
O TBSO
O
Synthesis of B-Ring
O S
OBn
88%
HO
2. OsO4 , NaIO4 H2 O, THF, rt
52
1. OsO4 (cat) NaIO 4
3. (-)-Ipc 2 BCH 2CH=CH2
3. TIPSOTf 2,6-lutidine
74%
CN
Cl
CN O
6.6:1
TBSO OHC
TBSO
EtO2 C
OH
O
58
88%
2. Gennari-Still KHMDS, 18-C-6
OTIPS
OTIPS 57
89%
Couple with A-ring using LiN(SiMe3 )2 as base
TBSO
20:1
1. (EtO)2 POCH2 CO 2H DCC, 4-DMAP, CH 2Cl2 2. PPTS, MeOH
TBSO 82%
HO
O
OBn
3. i-Bu2AlH (91%) 4. Li, NH 3 (87%) 5. NaH, TBSCl (94%)
OTIPS 56
3. MnO 2 4. K2CO3 , 18-C-6 toluene, ∆
O I OAc AcO OAc
O t
TBSO
OH 60
CO 2Et O
O
O t
1. i-Bu2AlH 2. (EtO)2 POCH2 CO 2Et NaH (89%)
B
3. i-Bu2AlH OTBS
O
1. Swern
1. Dess-Martin Periodinane pyridine
O
3
4
3. HF-pyridine, THF, rt (87%) O
B OTBS
65% overall
Lasonolide-7
(+)-Lasonolide A
2. C28-OTES Ph3P=CHR (87%)
61
O
OH
Longest linear sequence = 26 steps Overall yield = 7.4%
Page 492
1. TBAF 2. TBSOTf, 2,6-lutidine
1. DDQ, H2 O CH 2Cl2 (87%) 2. NaH (81%)
10:57 AM
TBSO
A
OTIPS 86% overall
Organic Synthesis via Examination of Selected Products
Cl
(DDQ)
TBSO
55
O
B-ring ready f or coupling
59
PMBO
2. Ph3 P=CHCO2 Et EtO C 2 54
OH
53
O
OBn
22:1
b1026
A-ring ready f or coupling
OBn PMBO
Organic Synthesis via Examination of Selected Natural Products
51
1. NaH, PMBCl TBAI, DMF THF, rt (97%)
12/21/2010
OH
TIPS = Si(i-Pr) 3
O N
TBSO
b1026_Chapter-12.qxd
492
31:1
b1026_Chapter-12.qxd
12/21/2010 b1026
10:57 AM
Page 493
Organic Synthesis via Examination of Selected Products
A Recent Example of Structure Determination
493
Lasonolide-7 The B-ring THP (59) was prepared starting with known alcohol 52. Protection of the secondary alcohol, cleavage of the olefin, and allylation of the resulting aldehyde with a reagent developed by the Brown group at Purdue (53) gave 54.15 Ozonolysis of the double bond was followed by a Wittig reaction and protection of the secondary alcohol to provide 55. Oxidative removal of the p-methoxybenzyl protecting group, and an intramolecular conjugate addition formed B-ring THP 56, which was converted to 59 using a standard series of reactions. The A-ring and B-ring THPs were coupled to provide 60, using a JuliaLythgoe-Kocienski reaction. The C21 alcohol was esterfied, and three of the four TBS groups (presumably the least hindered groups) were removed. Oxidation of the allylic alcohol using activated manganese dioxide was followed by the intramolecular HWE reaction to give macrocycle 61. Oxidation of the primary alcohol using the Dess-Martin periodinane gave the expected aldehyde. A Wittig reaction and deprotection of the silicon protecting groups (TBS and TES) completed the synthesis of lasonolide-A (1). This was quite an effective synthesis which illustrates, once again, the importance of convergence in planning a synthesis.
b1026_Chapter-12.qxd
494
Third Synthesis of Lasonolide A Yoshimura, T.; Yakushiji, F.; Kondo, S.; Wu, X.; Shindo, M.; Shishido, K. "Total Synthesis of (+)-Lasonolide A" Or g. Lett. 2006, 8, 475-478.
O
23
O
19
HO
21
O O
O
Wittig 1
1
11
O
Lactonization
10 9
OH
OH
O
64 OH
TBDPSO O
O
O 16 Steps
13 Steps 64
65
BnO
Deba, T.; Yakushiji, F.; Shindo, M.; Shishido, K. Sy nlet t 2003, 1500
66
H
OH
We have seen 62
63
O O
Takeuchi, M.; Taniguchi, T.; Ogasawara, K. Synthesis, 1999, 341 OBn
There have been a number of other partial syntheses (and one formal synthesis) published. For a very interesting approach to the A-ring see: Dalgard, J. E.; Rychnovsky, S. D. "Enantioselective Synthesis of the C18 -C 25 Segment of Lasonolide A by an Oxonia-Cope Prins Cascade" Org. Let t. 2005, 7, 1589-1592.
O
O O
BnO H
Lasonolide-8
CO 2Et OBn
O 67
68
before.
Page 494
7
Gennari-Still
Et3SiH-BF3 etherate
10:57 AM
22
63 O
Stille
17
Organic Synthesis via Examination of Selected Products
25
62
18 t
12/21/2010
26 c
28
O
TBDPSO
PPh3 I O
b1026
OH O
O
Metathesis
Organic Synthesis via Examination of Selected Natural Products
OTBS Wittig
b1026_Chapter-12.qxd
12/21/2010 b1026
10:57 AM
Page 495
Organic Synthesis via Examination of Selected Products
A Recent Example of Structure Determination
495
Lasonolide-8 We will look at a third synthesis of lasonolide-A in abbreviated form. The Shishido plan (University of Tokushima in Japan) was to assemble subunits 62–64. The side chain was to be introduced via a Wittig reaction in the usual manner. The A-ring THP (63) and B-ring THP (64) were to be joined using a crossed-olefin metathesis reaction to construct the C17-C18 double bond, and a lactonization to close the macrocycle. The olefins in 64 were to be installed sequentially using Gennari-Still, Wittig (stabilized phosphorane) and Stille coupling reactions. The C11 stereochemistry in 64 was to be set in an ionic reduction of a C11 hemiacetal. Briefly, B-ring THP 64 was prepared from 65 (which took several steps to prepare) using a 13 reaction sequence. A-ring THP 63 was prepared in 16 steps from known compound 66 via key intermediates 67 and 68.
+
(MeO) 2
69
O
O
64
(10 mol%)
Ru CHPh
Cl
PCy3
3. HOCH 2CH 2OH, TBAI, CSA CH 2Cl2, THF (70%)
O HO
4. LiOH, H 2O, THF, MeOH (84%)
OTBS
5. TBSCl, imidazole (94%)
70
t
O
TBDPSO O Cl
Cl
1.
OH
t
Et3N TBSO
Cl
73
1. TBAF, AcOH (45%)
O 74 O
2. 4-DMAP, PhH
2. Pyr-SO3 , DMSO, Et3 N, CH 2Cl2 (86%)
O
O
HO2 C
58%
OTBS
OH
O
OHC
OTBS
1. KHMDS
t
62
(45%)
c
O
t
O
O TBSO
2. HF-pyridine, THF (70%)
O 75
(+)-Lasonolide A HO
O 1
O
O O
O OTBS
OH
Lasonolide-9
Page 496
TBSO
O
Cl
10:57 AM
TBDPSO
Organic Synthesis via Examination of Selected Products
(19% of homodimer of A as a mixture of geometrical isomers)
b1026
Cl
70%
72
Organic Synthesis via Examination of Selected Natural Products
2.
NMes
CO2 Me
12/21/2010
71 MesN
P O
b1026_Chapter-12.qxd
63
PCy3
2. LiOH, H 2O, THF (86%)
(20 mol%)
Ru CHPh
Cl
1. Pyr-SO3 , DMSO, Et3 N (88%)
t
O
496
1.
TBDPSO
PCy3
Cl
b1026_Chapter-12.qxd
12/21/2010 b1026
10:57 AM
Page 497
Organic Synthesis via Examination of Selected Products
A Recent Example of Structure Determination
497
Lasonolide-9 Sequential treatment of what is presumed to be an equimolar mixture of 63 and 64 with Grubb’s olefin metathesis catalysts 69 and 70, gave a good yield of 71 along with lesser amounts of the dimer of 64 (derived from the terminal olefin). The olefin geometry of the newly formed double bond was clearly E or trans. The primary alcohol was oxidized to the aldehyde and a vinylogous HWE reaction using phosphonate 72 introduced the C5-C6 double bond. An acetal exchange reaction, hydrolysis of the methyl ester, and protection of the C22 hydroxymethyl group gave hydroxy acid 73. Activation of the acid by formation of a mixed anhydride, and cyclization using 4dimethylaminopyridine as a catalyst (the Yamaguchi procedure) gave 74 in 58% overall yield from 71.16 Removal of the sterically most accessible silyl ether protecting group and subsequent oxidation of the primary alcohol gave aldehyde 75. The now familiar Wittig reaction and global deprotection provided lasonolide A (1). In this chapter you have been introduced to lasonolide A (1) as both a polyacetate-derived natural product and as a macrolide. The next chapter will focus on a biosynthetically related polypropionate-derived natural product, and introduce a family of natural products known as ionophores. The final chapter will deal with a famous polypropionate-derived macrolide. The targets selected to represent ionophores and polypropionate-derived macrolides have both been selected for historical reasons. Let’s move on.
b1026_Chapter-12.qxd
12/21/2010 b1026
498
10:57 AM
Page 498
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
References 1. Julia, M.; Paris, J. M. “Syntheses a L’aide de Sulfones. V. Methode de Syntheses Generale de Doubles Liasons” Tetrahedron Lett. 1973, 15, 4833–4836. Kocienski, P. J.; Lythgoe, B.; Ruston, S. “Scope and Stereochemistry of an Olefin Synthesis from β-Hydroxy-Sulfones” J. Chem. Soc., Perkin Trans. 1 1978, 829–834. 2. Farina, V.; Krishnamurthy, V.; Scott, W. J. “The Stille Reaction” Organic Reactions 1997, 50, 1–652. Pattenden, G.; Sinclair, D. J. “The Intramolecular Stille Reaction in Some Target Natural Product Syntheses” J. Organomet. Chem. 2002, 653, 261–268. 3. Lipton, M. F.; Basha, A.; Weinreb, S. M. “Conversion of Esters to Amides with Dimethylaluminum Amides: N,N-Dimethylcyclohexanecarboxamide” Organic Syntheses 1980, 59, 49–53. Basha, A.; Lipton, M.; Weinreb, S. M. “A Mild, General Method for Conversion of Esters to Amides” Tetrahedron Lett. 1977, 18, 4171–4174. 4. Nahm, S.; Weinreb, S. M. “N-Methoxy-N-methylamides as Effective Acylating Agents” Tetrahedron Lett. 1981, 22, 3815–3818. 5. Still, W. C.; McDonald, J. H., III “Chelation-Controlled Nucleophilic Additions. 1. A Highly Effective System for Asymmetric Induction in the Reaction of Organometallics with α-Alkoxy Ketones” Tetrahedron Lett. 1980, 21, 1031–1034. Still, W. C.; Schneider, J. A. “Chelation-Controlled Nucleophilic Additions. 2. A Highly Effective System for Asymmetric Induction in the Reaction of Organometallics with β-Alkoxy Aldehydes” Tetrahedron Lett. 1980, 21, 1035–1038. 6. Tamao, K.; Ishida, N.; Kumada, M. “(Diisopropoxymethylsilyl)methyl Grignard Reagent: A New, Practically Useful Nucleophilic Hydroxymethylating Agent” J. Org. Chem. 1983, 48, 2120–2122. Tamao, K.; Kakui, T.; Akita, M.; Iwahara, T.; Kanatani, R.; Yoshida, J.; Kumada, M. “Oxidative Cleavage of Silicon-Carbon Bonds in Organosilicon Fluorides to Alcohols” Tetrahedron 1983, 39, 983–990. 7. Fleming, I. “Silyl-to-Hydroxy Conversion in Organic Synthesis” Chemtracts: Organic Chemistry 1996, 9, 1–64. 8. Blakemore, P. R.; Cole, W. J.; Kocienski, P. J.; Morley, A. “A Stereoselective Synthesis of trans-1,2-Disubstituted Alkenes Based on the Condensation of Aldehydes with Metalated 1-Phenyl-1H-tetrazol-5-yl Sulfones” Synlett 1998, 26–28. Blakemore, P. R. “The Modified Julia Olefination: Alkene Synthesis via the Condensation of Metallated Heteroarylalkylsulfones with Carbonyl Compounds” J. Chem. Soc., Perkin Trans. 1 2002, 2563–2585. 9. Parikh, J. R.; Doering, W. v. E. “Sulfur Trioxide in the Oxidation of Alcohols by Dimethyl Sulfoxide” J. Am. Chem. Soc. 1967, 89, 5505–5507. Pfitzner, K. E.;
b1026_Chapter-12.qxd
12/21/2010 b1026
10:57 AM
Page 499
Organic Synthesis via Examination of Selected Products
A Recent Example of Structure Determination
10.
11.
12. 13.
14.
15.
16.
499
Moffatt, J. G. “Sulfoxide-Carbodiimide Reactions. I. A Facile Oxidation of Alcohols” J. Am. Chem. Soc. 1965, 87, 5661–5670. Pfitzner, K. E.; Moffatt, J. G.; “Sulfoxide-Carbodiimide Reactions. II. Scope of the Oxidation Reaction” J. Am. Chem. Soc. 1965, 87, 5670–5678. Still, W. C.; Gennari, C. “Direct Synthesis of Z-Unsaturated Esters. A Useful Modification of the Horner-Emmons Olefination” Tetrahedron Lett. 1983, 24, 4405–4408. Maercker, A. “The Wittig Reaction” Organic Reactions 1965, 14, 270–490. Maryanoff, B. E.; Reitz, A. B. “The Wittig Olefination Reaction and Modifications Involving Phosphoryl-Stabilized Carbanions. Stereochemistry, Mechanism, and Selected Synthetic Aspects” Chemical Reviews 1989, 89, 863–927. Blanchette, M. A.; Choy, W.; Davis, J. T.; Essenfeld, A. P.; Masamune, S. Roush, W. R.; Sakai, T. “Horner-Wadsworth-Emmons Reaction: Use of Lithium Chloride and an Amine for Base-Sensitive Compounds” Tetrahedron Lett. 1984, 25, 2183–2186. Mancuso, A. J.; Swern, D. “Activated Dimethyl Sulfoxide: Useful Reagents for Synthesis” Synthesis 1981, 165–185. Roush, W. R.; Halterman, R. L. “Diisopropyl Tartrate Modified (E)Crotylboronates: Highly Enantioselective Propionate (E)-Enolate Equivalents” J. Am. Chem. Soc. 1986, 108, 294–296. Roush, W. R.; Ando, K.; Powers, D. B.; Palkowitz, A. D.; Halterman, R. L. “Asymmetric Synthesis Using Diisopropyl Tartrate Modified (E)- and (Z)-Crotylboronates: Preparation of the Chiral Crotylboronates and Reactions with Achiral Aldehydes” J. Am. Chem. Soc. 1990, 112, 6339–6348. Bartlett, P. A. “Olefin Cyclization Processes that Form Carbon-Heteroatom Bonds” in Asymmetric Synthesis 1984, 3, 411–454. Bartlett, P. A. “Olefin Cyclization Processes that Form Carbon-Carbon Bonds” in Asymmetric Synthesis 1984, 3, 341–409. Brown, H. C.; Bhat, K. S.; Randad, R. S. “Chiral Synthesis via Organoboranes. 21. Allyl- and Crotylboration of α-Chiral Aldehydes with Diisopinocampheylboron as the Chiral Auxiliary” J. Org. Chem. 1989, 54, 1570–1576. Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M. “A Rapid Esterification by Mixed Anhydride and its Application to Large-Ring Lactonization” Bull. Chem. Soc. Jpn. 1979, 52, 1989–1993.
b1026_Chapter-12.qxd
12/21/2010 b1026
10:57 AM
Page 500
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
500
Problems 1. Reactions of esters with an excess of a Grignard reagent, followed by a protic work-up, give tertiary alcohols. However, the reaction of 4 with excess Grignard, followed by protic work-up, affords a ketone. Provide a mechanistic explanation for this observation. Provide a mechanistic explanation for the following related transformation. (Lasonolide-2) O
MgBr
N S
O
1. 2. H 2O
83%
Mukaiyama, T.; Araki, M.; Takei, H. “Reactions of S-(2-Pyridyl)thioates with Grignard Reagents. Convenient method for the Preparation of Ketones” J. Am. Chem. Soc. 1973, 95, 4763–4765. 2. Cyclization of the 5-methyl-5-hexenyl radical (1) provides a 24:1 mixture of II and III. Explain the reversed regioselectivity in the cyclization of 9 → 10. (Lasonolide-3) cyclization + I
II
III II:III = 24:1
3. 4. 5.
6.
Walling, C.; Cioffari, A. “Cyclization of 5-Hexenyl Radicals” J. Am. Chem. Soc. 1972, 94, 6059–6064. Provide a mechanism for the oxidation of the C-Si bond (to a C-O bond) in Tamao-Fleming reaction. (Lasonolide-3) Provide the structures of the product after each step en route from 13 → 16. (Lasonolide-3) The conversion of 16 → 17 involves metallation of the sulfone, addition of the resulting anion to the aldehyde, intramolecular transfer of the benzothiazole to the resulting alkoxide, and an elimination reaction (or fragmentation reaction) to provide the product. Provide the mechanistic details for this transformation and speculate on the origin of stereoselectivity in the process. (Lasonolide-3) Propose a transition state geometry for the cyclization of 25 → 26. Compare this with alternative cyclization geometries and suggest why the reaction was stereoselective. (Lasonolide-4)
b1026_Chapter-12.qxd
12/21/2010 b1026
10:57 AM
Page 501
Organic Synthesis via Examination of Selected Products
A Recent Example of Structure Determination
501
7. Propose a mechanism for the oxidation of 27 to the corresponding aldehyde. (Lasonolide-4) 8. Provide equations that describe the mechanistic discussion of the HWE and Gennari-Still reactions presented here. (Lasonolide-4) 9. Provide the structure of each product in the reaction sequence leading from 38 → 37. (Lasonolide-5) 10. Assume that the partitioning between 43a and 43b is controlled by thermodynamics (all reactions reversible). Assume that the reactions (42 or 43b → 43a + 43b) are run at room temperature (25 oC). Estimate the free energy difference between 43a and 43b. 11. Provide an explanation for the diastereoselectivity (27:1) observed in the conversion of 45 → 46. (Lasonolide-6) 12. Alcohol 52 was prepared by asymmetric “crotylation” of 2-benzyloxyacetaldehde with the “crotyl derivative” of 53. What is the structure of this reagent? Provide a mechanistic rationale for the diastereoselectivity of the following reaction (the product is a racemic mixture). (Lasonolide-7) B O
O
1. BnOCH 2 CHO
HO
OBn
2. hydrolysis
13. Discuss the stereochemical issues involved in the stereoselective closure of 55 to B-ring THP 56. Discuss the relationship of this reaction to the conversion of 25 → 26 in the Lee synthesis (Lasonolide-7) 14. Provide the structure of the product obtained in each reaction in the conversion of 56 → 59. (Lasonolide-7) 15. Develop a plan for the synthesis of 65 from simple starting materials. Do not worry about stereochemistry at first, but eventually discuss the stereochemical issues you will have to face given your synthetic strategy. (Lasonolide-8) 16. Outline tactics that would accomplish the following transformation. (Lasonolide-8)
TBDPSO
TBDPSO O
CHO
OTBS
?
O
OTBS
11
N
12
18
O 15
HO
10
O 14
13
CO 2H
O 20
O O H
H N
19
HO
HO H H
OCH3
O H O
N H OH
OH
B-ring
O
O
CO2 H HO
16 17
N
O
O
H
O
O
O O
N O
H
OH OH
OH Calcimycin (A-23187) (complexes calcium) (1)
Monensin (complexes sodium) (2) Enterobactin (complexes Fe +3 ) (3)
Oa exocyclic and axial to B-ring
Ob exocyclic and axial to A-ring
The Kishi synthesis of monensin features allylic conformational analysis to predict stereochemistry of hydroborationoxidations in acyclic systems. The Still synthesis features acyclic diastereoselection in carbonyl addition reactions (chelation control and Felkin-Ahn control).
Calcimycin-1
Page 504
8 9
10:58 AM
A-ring
12/21/2010
HO NHMe
Organic Synthesis via Examination of Selected Products
Just as with lasonolide A, these targets are amenable to convergent syntheses. It is not surprising that there are features common to all of the calcimycin syntheses. There are some strategic diff erences, however, and these will emerge as we consider each approach. We will begin with the first synthesis of calcimycin.
b1026
Both calcimycin and monensin contain spiroacetal substructures, a common structural motif in ionophore natural products. In both cases the acetal carbon is a stereogenic center. The problem of stereocontrol at this acetal center is usually handled by thermodynamics and follows a stereochemical analysis with its foundation in the anomeric ef fect. The anomeric effect suggests that spiroacetals should prefer configurations that allow conformations that place both "exocyclic oxygens" on sites that are axial on the ring to which they are attached (see structure of calcimycin). The situation is similar for monensin.
Organic Synthesis via Examination of Selected Natural Products
Ionophores constitute a large f amily of natural products. These compounds complex ions and thus, can have interesting pharmacological properties. Three examples of ionophores are shown below. Calcimycin (originally known as A-23187) and monensin are examples of polyether ionophores. They complex metal ions much as do crown ethers. The ligating atoms in both ionophores are largely oxygen atoms. Enterobactin is a siderophore (iron complexing agent) in which the three catechols provide six oxygen ligands for metal complexation. We will focus on the synthesis of calcymicin.
O
b1026_Chapter-13.qxd
504
Ionophores: Calcimycin (A-23187)
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 505
Organic Synthesis via Examination of Selected Products
Ionophores: Calcimycin
505
Calcimycin-1 This chapter will deal with calcimycin (1), originally known as A-23187. Calcimycin is an example of an ionophore, a large family of natural products known to complex ions in water. Other “famous” ionophores include monensin (2) and enterobactin (3), although I will not dwell on the synthesis of these natural products.1,2 I have selected calcimycin as a target (1) for personal reasons (I wrote an NIH postdoctoral fellowship on this topic after the molecule was suggested to me by my postdoctoral mentor as an interesting target), (2) because a large number of syntheses have been completed and thus, comparison is possible, (3) because calcimycin is one of the first ionophores to attract the attention of synthetic organic chemists (many others have followed), (4) calcimycin is a polypropionate-derived natural product (at least in part) and (5) calcimycin is one of the first spiroketal-containing natural products to be prepared by total synthesis. Any synthesis of calcimycin must deal with issues of stereochemistry along the chain labelled C8-C20. Six of the seven stereogenic centers are contained in rings and one (C19) is not. The acetal carbon is particularly interesting. It is a common structural motif in many natural products (see monensin for another example). The problem of stereocontrol at this acetal center is usually handled by thermodynamics and follows from a stereochemical analysis with its foundation in the anomeric effect.3 The anomeric effect suggests that spiroacetals should prefer configurations that allow placement of both “exocyclic oxygens” on sites that are axial on the ring to which they are attached (see structure of calcimycin). Just as with lasonolide A, most ionophores are amenable to convergent syntheses. It is not surprising that there are features common to all of the calcimycin syntheses. There are strategic differences, however, and these will emerge as we consider each approach. We will begin with the first synthesis of calcimycin, described by the Evans group (Caltech at the time).
b1026_Chapter-13.qxd
506
First Synthesis of Calcimycin Evans, D. A.; Sacks, C. E.; Kleschick, W. A.; Taber, T. R. "Polyether Antibiotics Synthesis. Total Synthesis and Absolute Configuration of the Ionophore A23187" J . Am. Chem. Soc. 1979, 101, 6789-6791.
11
O 18
O 20
OH O HO
H N
19
O
OR 2 O R 1O
H N
18
18
14
OR1
10
11
15
17
O
OR2
16 17
5
4
1
Latent C2 axis of symmetry
1. Set C 11 and C17 stereochemistry using latent C 2 axis of symmetry to advantage. 2. Use themodynamics to set stereochemistry at C15 (see note 3). 3. C13 and C 15 exchange hydrogen without epimerization upon exposing calcimycin to DCl, dioxane, ∆, 18h.
13
O H 3C N
Synthesis of Benzoxazole
60%
NO 2 HO
NMeCOCF3 CO 2Me
11 CH3 I, K2CO3
HO
(CF3 CO)2 O NH2 CO 2Me 8
Et3N
HO
NHCOCF3 CO 2Me
HNO 3, Et2O NHCOCF3 CO 2Me 9 92%
1. H2 , Pd/C
+ HO O2 N
10
2. AcCl, xylene 140 oC
NHCOCF3 CO 2Me
2:1 in favor of 1,2,3,4-tetrasubstitution
Calcimycin-2
acetone
O H 3C N
NHCOCF3 CO 2Me 12
Page 506
Notes on Strategy
10:58 AM
O
13
7
CO 2R
10
10
14
15
N
CO 2R
H N
Organic Synthesis via Examination of Selected Products
12
N
CO 2H
O
NMeX
O
12/21/2010
N
6
b1026
8 9
NMeX
O
Organic Synthesis via Examination of Selected Natural Products
NHMe
O
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 507
Organic Synthesis via Examination of Selected Products
Ionophores: Calcimycin
507
Calcimycin-2 The plan revolved around the aforementioned notion that thermodynamics would control spiroacetal stereochemistry (C14). This is a reasonable assumption given that acetal formation is a reversible process. Thus a keto-diol of type 4 was selected as a late intermediate in the projected synthesis. It was hoped that thermodynamics might also control stereochemistry at C15. In principle, this center is epimerizable. The presence of a 1,3-diaxial methylmethyl interaction in the C15-epimer of 1 would be expected to render it much higher energy that 1 itself. In practice, treatment of calcimycin (1) with DCl in dioxane led to exchange of the C13 and C15 hydrogens without erosion of stereochemistry. This provided experimental evidence to support the plan for setting these stereogenic centers. Compound 4 was to be prepared from a protected keto-diol of type 5 (the C10-C18 fragment of calcimycin). Addition of the anion derived from a benzimidazole of type 6 to a C10aldehyde, and an aldol reaction between 2-acylpyrrole 7 and a C18-aldehyde, were to be used to assemble 4 from pieces 5–7. There were stereochemical issues associated with both of the carbonyl additions, and also some protecting group issues associated with the “timing” of these steps, that had to be addressed during execution of the plan. It was noticed that 5 has a latent C2-axis of symmetry (consider 5 without the C15 methyl group). It was hoped that this would simplify the preparation of this intermediate. The benzimidazole was prepared from 8. Acylation of the amino group and nitration of the resulting trifluoroacetamide (9) gave a mixture of regioisomers 10 and 11. The nitro group of the 10 was reduced and the resulting o-aminophenol was converted to benzimidazole 12 upon reaction with acetyl chloride. N-Alkylation of the amide provided 13, which ultimately played the role of 6 in the synthetic plan.
OH Li, NH 3
1. KH, KO-t-Bu (cat), ∆ TBDPSO
NNMe 2
TBDPSO
2. R-I
NNMe 2
1. CuCl2 , THF-H 2 O 2. Me2 C(CH 2OH) 2 TsOH
O
TBDPSO
O
2. (MeO)3 B TBDPSO 3. H2 O2
OH
NMeCOCF3 CO2Me
22
23
Trifluoroacetamide removed by TBAF
1. (CO2 H) 2, MeOH, 25 °C (87%) 88:12
O
TBDPSO
O
OH
25
O N
3. CrO3 -pyridine (80%)
MeO2C
NMeCOCF 3
27 BioRad AG 50W-X8 PhMe, 100 °C, 10 h
Model with PhCHO gives 7:3 of desired isomer (Cram)
NHMe
NHMe
O N
O
one isomer
O
O
2. ZnCl2, DMEH2 O (1:1)
N MeO2C
OHC
80%
26
33% (separate isomers by chromatography; assign stereochemistry as predicted by Cram’s Rule)
23% from aldehyde
O
tBuO
1. LDA
O
2. TBAF (60%)
OBn
21 N
(metallate with LDA)
O
CO 2CH3
O
H N
NHMe
O N n-PrSLi, HMPA
O
O
CO 2H
O O
28
1
Calcimycin-3
H N
Calcimycin or A-23187
Page 508
N
O
10:58 AM
1. s-BuLi
CrO 3, pyridine
O Li
OBn
20
Organic Synthesis via Examination of Selected Products
O
NNMe 2
b1026
O
TBDPSO
24
TBDPSO
91% 19
CHO
11 1:1
18
17
1. LDA
15
17
OBn
12/21/2010
PhS
15
Organic Synthesis via Examination of Selected Natural Products
PhS 2. R-I
I
14
b1026_Chapter-13.qxd
TBDPSO I 16
NNMe 2
Inexpensive starting material from Roche fermentation process
CO2 H
508
Synthesis of C10- C18 Substructure
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 509
Organic Synthesis via Examination of Selected Products
Ionophores: Calcimycin
509
Calcimycin-3 The C10-C18 substructure of calcimycin was prepared using 14 as the source of absolute stereochemistry at both C11 and C17. This commercially available starting material was converted to both 15 and 16 through a standard series of manipulations. The anion derived from hydrazone 17 was then alkylated with 16 to provide 18. The activating group (PhS) was removed using a dissolving metal reduction. The resulting mixture of diastereomers (19) was again metallated at the least hindered site, and alkylation of the resulting anion with 15 gave hydrazone 20. The hydrazone was hydrolyzed and the resulting ketone was converted to ketal 21. The purpose of differential protection of the C10 and C18 alcohols was to help with the aforementioned “timing” issues. It was planned to introduce the benzimidazole unit first and the pyrrole unit last. This called for initial removal of the C10 benzyl protecting group. This turned out to be a big problem using standard conditions such as hydrogenolysis. Eventually metallation of 21, followed by trapping the resulting carbanion with trimethyl borate and oxidation of the resulting C–B bond, gave 22. Collins oxidation of 22 gave aldehyde 23. Treatment of 23 with 24 (from benzimidazole 13) provided a separable mixture of diastereomeric alcohols. It was anticipated that 25 would be the major isomer based on expected Cram or Felkin-Ahn selectivity.4,5 Hydrolysis of the acetal using methanolic oxalic acid provided a mixture of dihydropyrans which were separated by chromatography. Removal of the C18 protecting group and another Collins oxidation provided aldehyde 26. The presumed zinc enolate of 27 reacted with aldehyde 26 to give a mixture of materials that were treated with an acidic resin to convert the aldol adduct to spiroacetal 28. Model reactions using benzaldehyde as the electrophile suggested the aldol should occur with only modest stereoselectivity. The overall yield from 26 to 28 was low, but 28 was the only stereoisomer obtained. This suggests that the plan for controlling stereochemistry at C14 and C15 had merit, although it was possible that stereoisomers at C15 simply did not undergo cyclization. The synthesis was completed by deprotection of the methyl ester using SN2 chemistry. This synthesis of calcimycin was one of the first total syntheses of an ionophore. It was preceeded by syntheses of nonactin (Gerlach and Schmidt),6 followed shortly by syntheses of lasalocid A (X537A) (Ireland),7 and then by the landmark syntheses of monensin (2) developed by Kishi and Still.1 The Evans synthesis illustrates the importance of symmetry in synthesis design, and documented observations that were clearly useful to those that followed. The synthesis suffered somewhat from stereocontrol, particularly
b1026_Chapter-13.qxd
510
Grieco Synthesis of Calcimycin Martinez, G. R.; Grieco, P. A.; Williams, E.; Kanai, K., Srinivasan, C. V. "Stereocontrolled Synthesis of Antibiotic A-23187 (Calcimycin)" J. Am. Chem. Soc. 1982, 104, 1436-1438.
O
O
15
addition-oxidation
O
29
20
1
20
16
H CH2 Cl2 , 0 °C, 20 min
35
m-CPBA NaHCO 3 CH2 Cl2
O
TBSO HO
H 17 18
O
PhH
H 31
H
H 19
1. LiAlH4
HO
O
2. H2 , PtO2 , EtOAc
HO H 33
O H 32 65%
15
TBSO
Me O
O 36
15
O
O
Me
Calcimycin-4
TBSO O
36a
H
1. LDA, HMPA
O
15
36e
15
2. MeI
CH2 Cl2
H
16
14
1. LDA, HMPA
H alkylation of half-chair enolate of this conformation gives "axial entry" of Me without 1,3diaxial interactions
34 90%
O
20
30
t-BuMe2 SiCl 4-DMAP, Et3 N
CrO 3-2 pyr O
O
MeOH
Grieco, P. A.; Williams, E.; Tanaka, H.; Gilman, S. J. Or g. Chem. 1980, 45, 3537.
H TBSO
BF3 -Et2 O
H 2O2 , NaOH
O
2. MeI
15
90%
O 37
Page 510
19
O
H N
19
10:58 AM
10
19
18
14
12/21/2010
17
CO 2H
Organic Synthesis via Examination of Selected Products
N
9
b1026
NHMe
O
carbonyl addition
Organic Synthesis via Examination of Selected Natural Products
One of the low points of the f irst calcimycin synthesis is introduction of the pyrrole unit via an aldol condensation. The yields are low and the stereocontrol at C 19 is probably marginal. The Grieco synthesis disconnects calcimycin between C 20 and the 2-position of the pyrrole. Therefore a significant diff erence between this approach and the Evans approach is an attempt to achieve better control of stereochemistry at C 19 . A secondary difference is that C15 was to be introduced with complete control of stereochemistry rather than relying on thermodynamics for stereocontrol. It will be seen that a consequence of this plan is an increase in the length of the synthesis.
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 511
Organic Synthesis via Examination of Selected Products
Ionophores: Calcimycin
511
over the C10 and C18 stereogenic centers. We will next move to a synthesis that addresses this deficiency.
Calcimycin-4 One of the low points of the Evans calcimycin synthesis was introduction of the pyrrole unit via an aldol condensation. The yields were low and the stereocontrol at C19 was probably marginal. The Grieco group (Indiana at the time) described a synthesis that disconnected calcimycin between C20 and the 2-position of the pyrrole. Therefore a significant difference between this approach and the Evans approach was an attempt to achieve better control of stereochemistry at C19. A secondary difference was that C15 was to be introduced with complete control of stereochemistry, rather than relying on thermodynamics for stereocontrol. It will be seen that a consequence of this plan is an increase in the length of the synthesis. As a personal note, it is fun for me to discuss this synthesis because of my connection with two of the coauthors. Greg Martinez was an undergraduate student at UC Berkeley when I was a graduate student, where we worked on a project together. Kenichi Kanai was my first postdoctoral student at OSU, moving to my group from the Grieco labs at about the time this synthesis was published. The starting point for the synthesis was norbornenone derivative 29, prepared (in part) as described at the bottom of Calcimycin-6. This starting material contains carbons C14-C20 (and the C17 methyl group) of calcimycin and thus, the initial stages of the synthesis were largely an exercise in oxidation chemistry. Given that 29 has a bicyclo[2.2.1]heptane substructure, two C–C bonds (C16-C20 and C14-C18) had to be broken to arrive at an acyclic compound of type 37. In addition, two methyl groups had to be introduced at the incipient C19 and C15 positions. Both C–C bonds were to be cleaved using Baeyer-Villiger oxidations, and the methyl groups were to be introduced by alkylation of cyclic intermediates, providing stereocontrol that was predictable. The synthesis began with Baeyer-Villiger oxidation of 29 to provide lactone 30. Acid was then used to rearrange 30 to lactone 31. There is a clear relationship between these transformations and chemistry developed for the synthesis of prostaglandins. Oxabicyclo[3.3.0]octane 31 provided a rigid template for introducing the C19 methyl group. Thus, alkylation of the enolate derived from 31 took place from the convex face to provide 32. A series of reductions provided 33. Protection of the primary alcohol and oxidation of the secondary alcohol gave 35, setting the stage for the second C–C bond oxidation. In accord with the plan, Baeyer-Villiger oxidation of 35 gave lactone 36. Alkylation of the enolate derived from 36 gave 37 with the desired
20
O
19
OH O
17
15
OH
acetone
OH
CuSO4
38
CH2 Cl2
1. CrO3 -2 pyr
OH O
2. H2 C=CHMgBr -78 °C
O 39
43
2. Me2 SO, (COCl) 2; Et3N
18 20
19
O
17
15
10 11
CO2 Me
14
O
O 4. TBAF 5. CH 2N 2
42
13
20
OMe CO2 Me
14
MeOH, TsOH
10
O
CO2 Me 10
O 77%
minor
45
1. ClCH2 O(CH 2) 2OMe i-Pr 2 NEt, CH2 Cl2
TBS 1.
Many oxidants f ailed: Swern, Collins, AgCO3 -Celite, MnO2 , NiO2 OMe
1. DDQ, dioxane
OTBS
2. TBAF, THF
O 20
49
46 20
95%
OH
OH
10
Li N NMe 2 THF, -78 o C
48
Calcimycin-5
N
CH2 Cl2
OMEM
95%
N 2. n-BuLi, heptane; Hg(OAc)2 , H 2O, THF
OMe OTBS 14
2. 78%
N NMe2
1. CrO3 -2 pyridine CH 2Cl2 (Collins)
10
O
2. LiAlH4
20
44
OH 14
10
O
93% 20
OH
47
41
Page 512
OMe
O
OH
Transfer of Chirality: Enolate Geometry Plays a Crucial Role
1. Al(Hg), THF H2 O, EtOH NaHCO3 2. CH 2N 2
14
O
88%
10:58 AM
1. CH 3CH 2COCl pyridine 2. LDA, THF, HMPA 3. TBSCl, ∆
Organic Synthesis via Examination of Selected Products
13
O
13
+ 75% (2:1; separated)
12/21/2010
90%
major
b1026
1. OsO4 , pyridine; NaHSO3 ; CSA, PhH
O 14
O
OH
Organic Synthesis via Examination of Selected Natural Products
3. TBSCl, ∆ 4. TBAF 5. CH 2N 2
O
O
O 40
1. CH 3CH 2COCl pyridine 2. LDA, THF
Need directed hydration
O
O
75%
37
b1026_Chapter-13.qxd
18
LiAlH 4
512
H TBSO
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 513
Organic Synthesis via Examination of Selected Products
Ionophores: Calcimycin
513
stereochemistry at C15. How can the stereoselectivity of this alkylation be explained? Alkylation of the half-chair conformation of the enolate derived from 36a, with axial entry of the electrophile (iodomethane), would give the observed product. The same stereoelectronic course of alkylation of the enolate derived from conformation 36e, would experience a 1,3-diaxial Me-Me interaction. This type of analysis can be useful in general in predicted the stereochemical course of many 6-membered ring enolate alkylations. It certainly works here.
Calcimycin-5 Continuing with the synthesis, reduction of lactone 37 was accompanied by renewal of the alcohol protecting group to give triol 38. The 1,3-diol unit was converted to an acetonide (39). Oxidation of the residual alcohol followed by reaction of the resulting aldehyde with vinylmagnesium bromide gave a separable mixture of 40 (major) and 41 (minor). These compounds differ by virtue of relative stereochemistry at the C14 and C15 stereogenic centers. Both diastereomers were moved forward to 42 using the ploy of transfer of chirality (Prostaglandins-24). Thus, conversion of 40 to the corresponding propionate ester, enolate formation in THF, trapping the enolate to provide the silyl ketene acetal, Claisen rearrangement of this intermediate (presumably via the most stable chair conformation), and conversion of the resulting silyl ester to a methyl ester, gave 42. When 41 was subjected to the same reaction sequence, only using THF-HMPA as solvent during enolate formation, 42 was also obtained. This sequence revolves around the “Ireland-Claisen” or “Enolate-Claisen” rearrangement.8 It is enolate geometry (E from 40 and Z from 41) that dictates the course of transfer of chirality in this reaction sequence. Try to work your way through the details by working the problem provided at the end of this chapter. It is notable that the reaction sequence translates 1,2-diastereoselectivity (although 2:1 is not very good) into 1,4-diastereoselectivity, much as we saw in the Taber approach to prostaglandins (Prostaglandins-25). The next task was to convert 42 to ketone 44. This transformation called for a regioselective hydration of the C13-C14 olefin. This was accomplished by vicinal dihydroxylation of the olefin, which was accompanied by formation of the γ-lactone (rather that the δ-lactone). The preference for γ-lactonization (which is most often the case in such situations) left the C14 hydroxyl group free for oxidation to ketone 43. Reductive cleavage of the α-C–O bond, and esterification of the resulting acid, gave 44 (see 99 to 100 in Alkaloids-12 for comparison). Ketoester 44 contains five of the seven stereogenic centers of calcimycin. Completion of the synthesis called for introduction of the pyrrole and
O
60% O
24
50
51
N NMe2
N
NMeCOCF3
O
CO 2Me
N
CO 2H
O
O
NMe2 N
2. K2CO3 , MeOH-H2 O (1:1)
O
O
O
H N
O 30%
52
50%
1
Synthesis of Starting Material
CO 2Me
MeO2 C Br
DBU, DMF O
53
O
∆
O O
Me
1. LiAlH4 2. TsCl, pyridine
J. Am. Chem. Soc. 1977, 99, 4111
3. LiEt3BH 4. HCl, THF 29
54
Calcimycin-6
O
Page 514
Synthesis is 35 steps from starting norbornenone. The Evans synthesis is about 20 steps shorter.
1. Cr 2(OAc)4 -2 H 2O, EtOH
10:58 AM
NMeCOCF3
O
Organic Synthesis via Examination of Selected Products
inseparable mixture of C 10 diastereomers
12/21/2010
NMeCOCF3 CO2Me
b1026
O
N
Organic Synthesis via Examination of Selected Natural Products
2. Li
10-CSA, CH 2 Cl2 , -15 oC, 10 h
OH
O
80% N NMe2
CO2Me
b1026_Chapter-13.qxd
N
1. Collins Oxidation O
514
10
OH
NMeCOCF3
O
OMe
OMe
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 515
Organic Synthesis via Examination of Selected Products
Ionophores: Calcimycin
515
benzimidazole fragments. Since these were to be introduced by carbonyl addition reactions, the C14 ketone had to be protected. This was accomplished by conversion of 44 to THP 45. This liberated C20 at the alcohol oxidation state. Some protecting group manipulations provided 47 which was poised for introduction of the pyrrole group. Oxidation of the primary alcohol followed by addition of 48 to the intermediate aldehyde, gave 49 as a mixture of diastereomers. Oxidation of the pseudo-benzylic C20-alcohol with DDQ (lots of the search part of research here) and removal of the C10 protecting group, provided 50, ready for introduction of the benzimidazole unit (Calcimycin-6).
Calcimycin-6 The synthesis of calcimycin continued with oxidation of 50, reaction of the aldehyde with 24 to give 51 as a mixture of C10-diastereomers. Treatment of this mixture with 10-camphorsulfonic acid gave 52. Presumably only the desired C10 diastereomer cyclized to the spiroacetal. The synthesis was completed by reductive cleavage of the N–N bond (pyrrole nitrogen protecting group) and hydrolysis of the trifluoroacetamide and methyl ester. This synthesis illustrates one strategy for the preparation of acyclic molecules containing multiple stereogenic centers — use cyclic structures to control stereochemistry and then liberate the acyclic structure. The strategy is not unlike several of the Cecropia juvenile hormone syntheses we examined, where stereoselective olefin synthesis was the goal. Whereas three pieces are ultimately assembled, the synthesis of the “central fragment” is linear and there was a price to pay for this approach. It is long.
b1026_Chapter-13.qxd
516
Synthesis of Calcimycin from D-Glucose Nakahara, Y.; Fujita, A.; Beppu, K.; Ogawa, T. "Total Synthesis of Antibiotic A23187 (Calcimycin) from D-Glucose" T etr ahedr on 1986, 42, 6465-6476
20
O
OH
OR
O
OH
8
O
H N
O
O
3
X
O
18
RO 20
19
55
Calcimycin
OH
17
15
OR
O
OH
D-Glucose
14
16
58
1
2
OH 57E
57N
20
OTr O
59 O
OMe
OTr O
1. Ph3 P=CH2 (94%)
1. Na, NH3 -THF (83%)
17
2. TsCl, pyridine (98%)
15
2. H2 , Pd/C (73%)
15
OMe
61
60 Yunker, M. B.; Planmann, D. E.; Fraser-Reid, B. Can. J. Chem. 1977 , 55, 4002. Holder, N. L.; Fraser-Reid, B. Can. J. Chem. 1973 , 51, 3357. 17
HO HO 16
58
18
15
HO
14
OH
D-Glucose
OMe
OEE O
17
16
O
Me2 CuLi (88%)
OMe
17
OEE O O OMe 65
64
62
17
18 15
63
CONH 2 O PhMe, TsOH 14
OMe
57% from tosylate
Calcimycin-7
OMe
30% H2 O2 aq. NaHCO 3 acetone
4 Steps
OH O
CN O
OTs O NaCN, TBAB, DMF, 85 °C
∆
Page 516
Synthesis of Acyl Anion Equivalent
10:58 AM
O
HO
X
14
5
Organic Synthesis via Examination of Selected Products
O
OH O
6 4
HO HO
56
12/21/2010
CO 2H
14
RO
b1026
N
X
Organic Synthesis via Examination of Selected Natural Products
NHMe
O
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 517
Organic Synthesis via Examination of Selected Products
Ionophores: Calcimycin
517
Calcimycin-7 The Evans synthesis gave a single enantiomer of calcimycin, whereas the Grieco synthesis was performed with racemic starting material and thus, gave racemic calcimycin. Note that Evans’ convergent strategy called for starting with enantiopure material whereas Grieco’s linear strategy was amenable to a racemic synthesis. The next synthesis we will examine returns to preparation of a single enantiomer. Carbohydrates are a rich source of chirality. They are popular members of the “chiral pool” as a starting point for enantioselective synthesis (see Prostaglandins-15).9 The Nakahara group (RIKEN in Japan) used D-glucose as the point of departure in their synthesis of calcimycin. The plan was to prepare a compound of type 55 from a keto-diol of type 56, and then introduce the heterocyclic end-groups. The spiroacetal precursor (56) was to be assembled by reaction of an acyl anion equivalent of type 57N with an electrophile of type 57E. Both the nucleophile and electrophile were to be prepared from D-glucose. Whereas this approach did lead to calcimycin, a glance at D-glucose reveals that a lot of work would have to be done to arrive at 57N and 57E. This is the down-side of the approach. Dithiane 71 (see Calcimycin-8) was selected as the acyl anion equivalent (57N). The plan was to use C1-C6 of D-glucose as C14-C19 of the target. This called for a one-carbon homologation at C6 of the glucose (to install C20) and a myriad of oxidation state changes and “methylations”. The known ketone 59 was chosen as the point of departure. Wittig olefination of the ketone followed by catalytic hydrogenation from the sterically most accessible face gave 60. The trityl group was then exchanged for a tosylate (60 → 61). An alternative route to 61 was developed from enone 64. The C17 methyl group was introduced via a cuprate addition (pseudo-axial delivery of methyl), and 65 was converted to 61 in a manner resembling its preparation from 59. It turned out the ethoxyethyl protecting group had operational advantages to the trityl protecting group upon scale-up. Continuing with the synthesis, the incipient C20 was introduced as a nitrile (61 → 62) and the nitrile was hydrolyzed to afford amide 63. Treatment of 63 with p-toluenesulfonic acid provided lactam 66 in excellent yield (Calcimycin-8).
17
(EtO) 3O BF4
14
O
OEt N
N
19
1. LDA, THF
O
15
OMe
2. LiAlH4
*
69
discard O
65
OEE
OMe
1. NaBH4
CN O
1. TsOH, MeOH (86%)
72
2. LiAlH4
BnO
3. NaCN, DMF (69% f or 2 steps)
CH2 OH 1. H2 , Pd/C, EtOH 2. Ac2 O, BF3 -Et2O O
1. MeOH, HCl (84%)
OMe
BnO
73
OH 11
74
OH
13
3. NaOMe, MeOH 4. NaIO 4 5. NaBH4
OMe
10
75
HO
MeO OMe
CSA, DMF Dithiane 71 I 14
TBSO 20
OEE 70%
S
8
S
O
O
1. t-BuLi, hexane-pentane O
O
77
OH
N
I
Ph3 P, toluene ∆ electrophile
Calcimycin-8
I
I
2. R-I (electrophile), HMPA
78
H N
O
O 76
43% overall
Page 518
OMe
70 55% overall
acetylate alcohols f or operational reasons
15 Steps from Enone
2. Tf2 O, Hunig’s base BnO
S
acyl anion equivalent
2. NaH, BnBr, DMF
OEE O
OH
2. ethyl vinyl ether PPTS, CH 2Cl2
71
13
S
HO 3:1
S
10:58 AM
20
1. TBDMSCl, DMF imidazole
S
Organic Synthesis via Examination of Selected Products
10
11
14
TBSO
OEE O
b1026
Synthesis of Electrophile
Organic Synthesis via Examination of Selected Natural Products
1,3-propanedithiol BF3 -Et2 O
12/21/2010
68
75% 67
66
1. H2 SO4 , MeOH 3:1
2. MeI
15
81%
CH2 OH O
O
b1026_Chapter-13.qxd
18
OEt
H N
20
518
O 19
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 519
Organic Synthesis via Examination of Selected Products
Ionophores: Calcimycin
519
Calcimycin-8 The rigid bicyclic framework of 66 was used to introduce the C19 methyl group with control of stereochemistry. Thus, conversion of 66 to imidate 67 was followed by deprotonation and alkylation to provide 68. If you compare C15 of 68 with the target (57N) you will see that it has the incorrect configuration. This was adjusted (in part) by treating 68 with acidic methanol. This gave methyl glycoside 69, in which partial epimerization had occurred at C15. Recall that it was already known that this stereogenic center could be equilibrated to provide the needed isomer in calcimycin itself (Calcimycin-2). The reaction of 69 with propan-1,3-dithiol gave 1,3-dithiane 70. This is a standard move in carbohydrate chemistry, used to open a pyranose or furanose to a derivative of the acylic polyhydroxyaldehyde. Sequential protection of the primary and secondary alcohols completed the synthesis of 71 (as a 3:1 mixture of diastereomers at C15). Iodide 77 was chosen to play the role of the electrophile (57E) in the projected coupling with dithiane 71. The plan was to use 65 as a point of departure. The anomeric carbon was to be discarded, the ketone was to become the iodide (C13 of calcimycin), and C4 and C5 of this monosaccaride derivative were to become the two stereogenic centers in 77. Reduction of 65 with sodium borohydride and protection of the resulting alcohols as benzyl ethers gave 72. Homologation of 72 to alcohol 74, via nitrile 73, was accomplished using chemistry similar to that used in the preparation of 71. The anomeric carbon was then excised using a five-reaction sequence. The resulting triol (75) was converted to acetal 76 and the free hydroxyl group was converted to iodide 77. The coupling of 71 and 77 proceeded well. Metallation of 71 was followed by alkylation with 77 to give 78 in 70% yield.
OH 14
TBSO 20
S
OEE
8
1. TBAF, THF
BzO 20
S
O
O
1. HgCl2 , CaCO3 , rt H2 O-CH 3CN
S
OEE
S
O
O
O
14
70%
80
15
67%
1 isomer ... validates Evans’ approach
8
8
8
OTBS
O O
83
N MgBr
H N
O
THF-Et2O, CuI
20
O
20
2. Ph3 P
OH
O N
S 2
70%
82
72%
84
20
S N
80%
O
1. Jones oxidation
O
81
1. TBAF (92%) 2. Jones oxidation (75%) NHMe
O 8
1. MeOCOCl
CO 2H
HO
2. O
O O
20
H N
H 2N
N
NMeCOCF3 CO2Me 10
N
CO2Me
THF
CO2H
n-PrSNa O
O
H N
DMF
O
O O
O
3. PPTS, ClCH 2CH2 Cl 4. NaHCO3 , H 2O
85
NHMe
O
86
Calcimycin-9
24%
1
H N
Page 520
O
OTBS
OTBS
10:58 AM
2. K2CO3 , MeOH
Organic Synthesis via Examination of Selected Products
1. TBDMSCl, DMF imidazole (85%)
b1026
78
OBz
O
79
Organic Synthesis via Examination of Selected Natural Products
2. H3 PO4 , THF, ∆
12/21/2010
2. PhCOCl, Et3 N
b1026_Chapter-13.qxd
520
8
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 521
Organic Synthesis via Examination of Selected Products
Ionophores: Calcimycin
521
Calcimycin-9 The C20 protecting group was exchanged to provide 79. Hydrolysis of the thioketal and the acetal protecting groups provided 80 in good yield as a single stereoisomer (recall Evans’ plan for thermodynamic control of C15 stereochemistry). The C8 alcohol was protected as a TBS ether and the C20 ester was hydrolyzed to set the stage for introduction of the pyrrole. Alcohol 81 was oxidized to the corresponding acid and then converted to thioester 82. Copper promoted coupling of 82 with pyrrole derivative 83 gave 2acylpyrrole 84. The C8 protecting group was removed and a CrVI oxidation gave acid 85. Acylation of o-aminophenol 10, followed by acid promoted ring closure and hydrolysis of the trifluoroacetamide, gave 86. The synthesis of calcimycin (1) was completed in the usual manner. Looking back at the Nakahara synthesis, it relies heavily on the use of cyclic compounds to set relative stereochemistry. The absolute stereochemistry at C10 and C18 came from D-glucose. The stereochemistry at C11, C16 and C19 was controlled by constraints imposed by cyclic compounds. And the stereochemistry at C14 and C15 was controlled by thermodynamics. We will now look at a synthesis that uses a different strategy for construction of the spiroacetal substructure of calcimycin.
b1026_Chapter-13.qxd
522
An Alternative Approach to Formation of the Spiroketal Negri, D. P.; Kishi, Y. "A Total Synthesis of Polyether Antibiotic (-)-A23187 (Calcimycin)" T et rahedron Let t. 1987, 28, 1063-1067
O
H N
14
18
O
OH
O
O
14
10
O
N
89
O
R’O 2C
NMeCOR
88 This plan has precedence in some degradation studies perf ormed on the oligomycin family of macrolide antibiotics.
EWG O
O
17
15
15
m-CPBA
Me2 CuLi
PhMgBr (XS)
PCC
14
Et2O O 90
-20 °C
CH2 Cl2 O 91 91%
0 °C
HO
O
18
Ph
14
O
Ph
92 93%
93 93%
Calcimycin-10
CH2 Cl2
HO(CH2 )3 OH CHO
Ph TsOH, PhH Ph
94 93%
∆
Page 522
Anticipates a late transition state such that double anomeric effect dictates cyclization stereochemistry
Calcimycin
10:58 AM
O
9
13
19
N H
12/21/2010
87 O
Organic Synthesis via Examination of Selected Products
CO 2H
HWE
b1026
N
C=O addition
EWG
Organic Synthesis via Examination of Selected Natural Products
aldol NHMe
O
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 523
Organic Synthesis via Examination of Selected Products
Ionophores: Calcimycin
523
Calcimycin-10 Kishi and Negri reported a synthesis of calcimycin that revolved around a base-initiated cyclization of 89. The hope was that the C18 hydroxyl group of 89 would add to the C14 carbonyl group to provide an equilibrium mixture of 87 and 88. It was anticipated that these might undergo intramolecular conjugate addition to provide spiroketals related to calcimycin. It was hoped that the rate of cyclization from 87 might exceed the cyclization rate from 88 because the double anomeric effect would be felt in that cyclization transition state (and not in the transition state derived from 88). This zipper reaction has some precedence in degradation studies performed on the oligomycin family of macrolide antibiotics.10 The cyclization substrate was to be prepared using an aldol condensation to set the C18-C19 bond, a carbonyl addition reaction to prepare the C13-C14 σ-bond, and a Horner-Wadsworth-Emmons to prepare the C9-C10 double bond. Cyclohexenone 90, which can be prepared from the monoterpene pulegone, was to supply the C11 and C15 stereogenic centers, and provide a handle for introduction of the C17 stereogenic center. An early goal was preparation of ketone 100 (Calcimycin-11). The olefin was to be a handle for introduction of the benzimidazole. An aldehyde was to be generated from the acetal and used to introduce the pyrrole substructure via an aldol condensation. Ketone 100 was to be prepared from aldehyde 93 and bromide 99, both of which were to be prepared from cyclohexenone 90. Moving forward, a cuprate addition gave C2-symmetric ketone 91. The plan called for the carbonyl carbon of 91 to be oxidatively excised, leaving C14 and C18 at the aldehyde oxidation state, but differentially protected. With this in mind, Baeyer-Villiger oxidation of 91 provided lactone 92. Reaction of 92 with excess phenylmagnesium bromide gave alcohol 93 after facile dehydration of an intermediate tertiary alcohol. Oxidation of the alcohol was followed by protection of 94 to give acetal 95. Ozonolysis of the olefin liberated the needed C14 aldehyde (Calcimycin-11).
Ph
O
O
CHO *
14
56% overall
Ph HO Ph
3. NaBH4
O 98
97
90
93%
HO
NaIO4
HO
2. O3 O
96
3. LiBr
2. Me2 S
18
O
Ph
N 102 MeO 2C
OHC
N B oc
2. CuO, CuCl2 H2 O, acetone NMeTroc
11
15
O
N
103 MeO 2C
NMeTroc
Some complications with retroaldol NHMe
O N 18
N H
19
O
O 17
11
15
OH
O
N
1. CH 2Cl2 , MeOH, NaOMe (cat) 2. Zn, THF, AcOH
61% + 16% of "stereoisomer"
O
MeO 2C
LiSPr, HMPT, rt O
O
NMeTroc
105
Calcimycin-11
Calcimycin
CO 2Me
10
14
104
O
Cy2 NMgBr THF, -50 o C
O 17
99
42%
H N
Page 524
1. 1,3-propanedithiol BF3-Et2 O, CH 2Cl2 0 °C
O
Ph
3. PCC (88%)
2 Steps needed to avoid C 15 epimerization
O O
2. RCHO (71% overall)
10
O
O
100 NaH, Et2 O
Br
Ph
101 HWE
1. Mg, Et2 O, ∆
14
10:58 AM
CHO O
O
O
Organic Synthesis via Examination of Selected Products
Ph 1. O3 , EtOAc
O
b1026
2. MsCl, Et 3N
Organic Synthesis via Examination of Selected Natural Products
1. PhMgBr
12/21/2010
93%
11
HO 1. PhLi
2. Me2 S
Ph 95
O
11
b1026_Chapter-13.qxd
O
1. O3 , EtOAc
524
15
17
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 525
Organic Synthesis via Examination of Selected Products
Ionophores: Calcimycin
525
Calcimycin-11 Conversion of 90 to 99 called for excision of the carbon marked with an asterisk. This was set up by converting 90 to triol 97 using a straightforward reaction sequence. Periodate cleavage of the vicinal diol accomplished the required one-carbon excision. The resulting ketone (98) was converted to 99 in a straightforward manner. Bromide 99 was converted to the corresponding Grignard reagent, reacted with aldehyde 96, and an oxidation completed the synthesis of 100. Next the benzimidazole was introduced. Thus, cleavage of the olefin liberated C10 as an aldehyde. The HWE reaction proceeded without incident. The C15 ketone was protected and the C18 acetal hydrolyzed to give aldehyde 103. An aldol condensation then provided the Felkin-Ahn (or Cram) product 104, in good overall yield. The key cyclization occurred as anticipated using sodium methoxide as a catalyst to trigger the process (104 to 105). The Troc protecting group was removed and the methyl ester was converted to the acid to provide calcimycin (1).
b1026_Chapter-13.qxd
526
Another Synthesis with All But C15 Stereochemistry Set Boeckman, R. K. Jr.; Charette, A. B.; Asberom, T.; Johnston, B. H. "A Convergent General Synthetic Protocol f or Construction of Spirocyclic Ketal Ionophore: An Application to the Total Synthesis of (-)-A-23187 (Calcimycin)" J. Am. Chem. Soc. 1987, 109, 7553-7555.
O
O
O
MOMO
OH
O
O
107
108
19
CH2 Cl2
O
3. LiCH(SPh)(OMe) H 4. HgCl2 MOMO 5. H2 O2 , pH 7 111 52%
17
OH
-23 °C, 3 h 110
1.4:1
18
MOMO
88%
Brown, H. C.; Bhat, K. S. J. Am. Chem. Soc. 1986, 108, 5919
Synthesis of R-Br 1. TBDPSCl (2 eq), 6h imidazole, DMF, rt
11
TBDPSO 2. O3 , MeOH, CH 2Cl2; Me2 S
CHO
MsCl, Et 3 N
OH
H
115
Calcimycin-12
85%
112 1. n-BuLi (3 eq) KO-t -Bu (3 eq) -78 °C , THF
1. Z-crotyldiisopinocampheylborane THF, -78 °C 2. H2 O2 , NaOH, H 2O, THF
O
MOMO
(next)
2. n-Bu 3SnCl
H
O
MOMO 113 100%
SnBu 3
Page 526
1. BH3 -THF 2. MeOH
6.7:1 mixture of diastereomers with 110 major
114
OTBDPS
OTBDMS
chelation control
MgBr 2-Et2 O
HO
8
11
Metallation with t-BuLi fails SnBu 3
109
Br
Synthesis of Dihydropyran
Keck, G. E.; Abbott, D. E. Tetrahedron Lett. 1984, 25, 1883
CHO
10
+
No Stereogenic Centers Left to Chance
106
MOMO
20
Li
10:58 AM
H N
H
O
Organic Synthesis via Examination of Selected Products
1
13
14
18
12/21/2010
CO 2H
b1026
17
N
Organic Synthesis via Examination of Selected Natural Products
OTBDPS
NHMe
O
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 527
Organic Synthesis via Examination of Selected Products
Ionophores: Calcimycin
527
Calcimycin-12 The last synthesis of calcimycin we will examine was reported by the Boeckman group (Rochester) at about the same time as the Kishi and Nakahara syntheses. The synthesis was (1) convergent, (2) featured metallated dihydropyran chemistry developed by the Boeckman group, (3) set all of the stereogenic centers with the exception of C14 and C15 prior to spiroacetal synthesis and (4) used crotylation-oxidative cleavage as a strategy for introduction of propionate units. The plan was to prepare calcimycin from spiroacetal 106, which was to be assembled from metallated dihydropyran 107 and bromide 108. The C17-C18 and C10-C11 bonds of these respective targets were to be prepared using the aforementioned crotylation chemistry. Let’s begin with the preparation of a 113, the precursor of 107. Known aldehyde 109 was treated with trans-crotyltributylstannane in the presence of a Lewis acid to provide 110. The relative stereochemistry between C18 and C19 was chelation controlled. The relative stereochemistry between C17 and C18 resulted from presumed addition of the stannane to the aldehyde via a chair-like transition state with Sn coordinated to the carbonyl oxygen. An interesting procedure was used to convert 110 to the homologous aldehyde, which cyclized to a mixture of anomeric hemiacetals 111. Dehydration of the mixture provided 112. Metallation of 112 using Schlosser’s base, followed by stannylation of the resulting vinyl anion, provided 113.11 The synthesis of 108 began with homoallylic alcohol 114. Protection of the alcohol was followed by ozonolysis to give aldehyde 115. Reaction of the aldehyde with Z-crotyldiisopinocampheylborane, a chiral crotylborane, gave 116 in high yield with excellent diastereo- and enantioselectivity (see Calcimycin-13).12 Note that oxidative cleavage of the olefins in 110 and 116 would provide α-methyl-β-hydroxy carbonyl compounds. This is the “repeating unit” found in polypropionate natural products. This methodology has been widely used in syntheses of such natural products. The same unit can be produced by aldol methodology (for example see 19 to 20 on Lasonolide-4). This methodology has also been widely used.
Et3N (3 eq)
Ph3 P (2 eq), CBr4 (2 eq)
OH
TBDPSO
Et2O, rt
2. H2 O2 , NaOH
CH2 Cl2 , rt
116 80%
OTBDMS
1. BH3 -THF
TBDPSO 117
118
H
108
MOMO
70%
119
H
Et2Zn
OTBDPS
HO
rt, 5 h
OTBDMS
O 120 TsOH-monohydrate PhH, 55 °C OTBDPS
1.
O
O O
123
H N
11
(18 eq) N MgCl
O
O O
2. TBAF 3. Jones oxidation
N S
122
80%
N
THF, Et3N NMeCOCF3 CO2Me
2. PPTS, ClCH 2CH2 Cl
OH
O 2. Ph3 P
N
S 2
15
121
Note stereoselectivity at C 14 and C15
NMeCOCF3
O
(BOP )
H 2N
14
55% from DHP
N PF 6 N O P(NMe ) 23 HO
30% of diol from loss of TBDPS
O
1. Jones oxidation
N
1.
10
CO2Me 1. TBAF, THF
O
O O
73%
Calcimycin H N
2. LiSPr (10 eq) HMPA, rt, 3.5 h
124
Calcimycin-13
98%
Page 528
OTBDPS CO 2H
10:58 AM
2. R-Br, THF-HMPA 0 °C rt
OTBDPS
14
15
12/21/2010
TBDPSO
O
Et2O hexanes CH2 I2
OTBDMS
Organic Synthesis via Examination of Selected Products
Br
15
b1026
1. Metallate vinylstannane w/ n-BuLi at -78 °C (107)
Organic Synthesis via Examination of Selected Natural Products
Coupling of Fragments
OTBDMS
b1026_Chapter-13.qxd
OTBDMS
TBDMSOTf (1.3 eq)
11 10
528
OH
8
TBDPSO
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 529
Organic Synthesis via Examination of Selected Products
Ionophores: Calcimycin
529
Calcimycin-13 The conversion of 116 → 108 was straightforward. The coupling of fragments was accomplished by converting stannane 113 → 107, and alkylating 107 with bromide 108. This provided dihydropyran 119. The C15 methyl group was installed by cyclopropanation of 119 to provide 120 as a mixture of diastereomers. Treatment of 120 with p-toluenesulfonic acid accomplished protonolysis of the cyclopropane, and cyclization of the resulting oxocarbenium ion to give 121 as a single diastereomer at C14 and C15. Epimerization at C15 with thermodynamic control of stereochemistry was clearly at work once again. The rest of the synthesis followed chemistry we have largely seen before, and is outlined in Calcimycin-13 without comment.13
b1026_Chapter-13.qxd
12/21/2010 b1026
530
10:58 AM
Page 530
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
References 1. For two syntheses see the following articles and references cited therein: Fukuyama, T.; Akasaka, I.; Karanewsky, D. S.; Wang, C. L. J.; Schmid, G.; Kishi, Y. “Synthetic Studies on Polyether Antibiotics. 6. Total Synthesis of Monensin. 3. Stereocontrolled Total Synthesis of Monensin” J. Am. Chem. Soc. 1979, 101, 262–263. Collum, D. B.; McDonald, J. H., III; Still, W. C. “Synthesis of the Polyether Antibiotic Monensin. 3. Coupling of Precursors and Transformation to Monensin” J. Am. Chem. Soc. 1980, 102, 2120–2121. 2. Shanzer, A.; Libman, J.; “Total Synthesis of Enterobactin via an Organotin Template” J. Chem. Soc., Chem. Commun. 1983, 846–847. Rastetter, W. H.; Erickson, T. J.; Venuti, M. C. “Synthesis of Iron Chelators, Enterobactin, Enantioenterobactin, and a Chiral Analog” J. Org. Chem. 1981, 46, 3579–3590. Corey, E. J.; Bhattacharyya, S. “Total Synthesis of Enterobactin, a Macrocyclic Iron Transporting Agent of Bacteria” Tetrahedron Lett. 1977, 19, 3919–3922. 3. Gelin, M.; Bahurel, Y.; Descotes, G. “Dipole Moment Studies of Substituted 2-Alkoxytetrahydropyrans” Bull. Soc. Chim. Fr. 1970, 3723–3729. 4. Cram, D. J.; Greene, F. D. “Stereochemistry. XX. Steric Control of Asymmetric Induction in the Preparation of the 3-Cyclohexyl-2-butanol System” J. Am. Chem. Soc. 1953, 75, 6005–6010. Eliel, E. L. “Application of Cram’s Rule: Addition of Achiral Nucleophiles to Chiral Substrates” in Asymmetric Synthesis, Morrison, J. D., Ed.; Academic Press, New York, New York, 1983, 2, 125–155. 5. Nguyen, T. A.; Eisenstein, O. “Theoretical Interpretation of 1,2-Asymmetric Induction. The Importance of Antiperiplanarity” Nouv. J. Chim. 1977, 1, 61–70. Cherest, M.; Felkin, H.; Prudent, N. “Tortional Strain Involving Partial Bonds. The Stereochemistry of the Lithium Aluminum Hydride Reduction of Some Simple Open-Chain Ketones” Tetrahedron Lett. 1968, 10, 2199–2204. 6. Gerlach, H.; Oertle, K.; Thalmann, A.; Stefano, S. “Synthesis of Nonactin” Helv. Chim. Acta 1975, 58, 2036–2043. 7. Ireland, R. E.; Anderson, R. C.; Badoud, R.; Fitzsimmons, B. J.; McGarvey, G. J.; Thaisrivongs, S.; Wilcox, C. S. “Total Synthesis of Ionophore Antibiotics. A Convergent Synthesis of Lasalocid A (X537A)” J. Am. Chem. Soc. 1983, 105, 1988–2006. 8. Ireland, R. E.; Willard, A. K. “Stereoselective Generation of Ester Enolates” Tetrahedron Lett. 1975, 16, 3975–3978. Ireland, R. E.; Mueller, R. H.; Willard, A. K. “Ester Enolate Claisen Rearrangement. Construction of the Prostanoid Skeleton” J. Org. Chem. 1976, 41, 986–996. 9. S. Hanessian “Total Synthesis of Natural Products: The Chiron Approach”, Pergamon Press, 1983.
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 531
Organic Synthesis via Examination of Selected Products
Ionophores: Calcimycin
531
10. Prouty, W. F.; Thompson, R. M.; Schnoes, H. K.; Strong, F. M. “Oligomycin: Degradation Products and Part Structure of Oligomycin B1” Biochem. Biophys. Res. Commun. 1971, 44, 619–627. 11. Rauchschwalbe, G.; Schlosser, M. “Selective Synthesis with Organometallics. IV. Hydroxylation of Allyl Positions” Helv. Chim. Acta 1975, 58, 1094–1099. 12. Brown, H. C.; Bhat, K. S.; Randad, R. S. “Chiral Synthesis via Organoboranes. 21. Allyl- and Crotylboration of α-Chiral Aldehydes with Diisopinocampheylboron as the Chiral Auxiliary” J. Org. Chem. 1989, 54, 1570–1576. Brown, H. C.; Bhat, K. S. “Chiral Synthesis via Organoboranes. 7. Diastereoselective and Enantioselective Synthesis of Erythro- and Threo-β-methylhomoallyl alcohols via Enantiomeric (Z)- and (E)-Crotylboranes” J. Am. Chem. Soc. 1986, 108, 5919–5923. 13. For other syntheses of calcimycin see: Ziegler, F. E.; Cain, W. T. “Formal Synthesis of (−)-Calcimycin (A-23187) via the 3-Methyl-γ-butyrolactone Approach” J. Org. Chem. 1989, 54, 3347–3353.
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 532
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
532
Problems 1. Propose a mechanism for the exchange of the hydrogens at C13 and C15 of 1 upon treatment with DCl in dioxane. (Calcimycin-2) 2. Provide the structure of intermediates and a mechanism that explains how 21 was converted to 22. (Calcimycin-3) 3. Show your understanding of the Cram and Felkin-Ahn models for asymmetric induction in carbonyl addition reactions by predicting (or rationalizing) the stereochemical course of the following reactions. (Calcimycin-3) O
O H LiAlH4
CH 3MgBr or CH 3Li ?
?
4. Examine all of the difunctional relationships from C8 → C20 in compound 28. Classify them as “odd” or “even”. Go through the Evans synthesis and determine how each difunctional relationship was constructed, or if they were purchased. (Calcimycin-3) 5. Predict the stereochemical course of the following enolate alkylations. (Calcimycin-4) O
O EtMe2 C O Na
O
? MeI
1. LDA, THF ? 2. MeI
O
O 1. Li, NH 3 2. CH2 =CHCH2 Br
?
O
1. LDA, THF ? 2. Me3 SiC CCH 2Br
6. Provide the structure of intermediates along the route from 40 → 42; from 41 → 42. How would you convert 40 to the C11 epimer of 42? 41 to the C11 epimer of 42? (Calcimycin-5) 7. Provide a mechanistic explanation for the conversion of 44 → 45 including stereochemistry at C14. (Calcimycin-5) 8. Provide the structure of the spiroacetal derived from the “undesired” C10 diastereomer of 51. Suggest why this diastereomer might not have cyclized to this spiroacetal. (Calcimycin-6)
b1026_Chapter-13.qxd
12/21/2010 b1026
10:58 AM
Page 533
Organic Synthesis via Examination of Selected Products
Ionophores: Calcimycin
533
9. Outline a synthesis of 64 from D-glucose. (Calcimycin-7) 10. Provide the structure of all intermediates in the conversion of 74 → 75. (Calcimycin-8) 11. Propose a mechanism for the conversion of 76 → 77. (Calcimycin-8) 12. The analysis provided for the conversion of 89 to a spiroketal is abbreviated. What other chair conformations are available to 87 and 88? What products would result from cyclization through 88 and/or these additional conformations? What would be the expected product if thermodynamics controlled the entire cyclization process? (Calcimycin-10) 13. Suggest tactics that would accomplish the following transformation. (Calcimycin-10) O
O ?
14. Provide an explanation for the stereochemical course of the conversion of 90 → 91. (Calcimycin-10) 15. Provide the structure of the HWE reagent needed to convert 101 → 102. (Calcimycin-11) 16. Provide a mechanism for removal of the Troc [(Cl3CCH2OC(=O)] protecting group in the penultimate step of the Kishi synthesis. (Calcimycin-11) 17. Examine all of the difunctional relationships from C8 to C20 in compound 104. Classify them as “odd” or “even”. Go through the Kishi synthesis and determine how each difunctional relationship was constructed. Indicate where A-functions were used (see Chapter 6). (Calcimycin-11) 18. Propose (with structures) a transition state model that explains the stereochemical course of the conversion of 109 → 110. (Calcimycin-12) 19. Provide the structure of the product expected from the following reaction sequence. (Calcimycin-12) 1. CH3OCH2Cl, i-Pr2NEt 2. O3, MeOH; Me2S MOMO
? OH
3. E-CH3CH=CHCH2SnBu3, MgBr2-etherate CH2Cl2
20. Propose a mechanism for the conversion of 110 → 111. (Calcimycin-12) 21. Propose a mechanism for the conversion of 120 → 121. (Calcimycin-13)
H
OH
12
11
14
O
1
11
OH OH
O
14
MTMO 13
12
3
O
OBz
BzO O
6
OH
CO2 H
C 1-C9
X = OH Erythronolide A (2) X = H Erythronolide B (3) syn-dihydroxylation
4
BzO 10
6
OBz 2
1
O
13
9 CO2 H
6
7
Thermodynamic Control Phenol Alkylation
Olefin Hydration
"Enone Hydration"
O
OH
O
O 3
CO 2H 11
HO
10
9
Erythronolide-1
O
8
O
8
Page 536
Corey, E. J.; Hopkins, P. B.; Kim, S.; Yoo, S.; Nambiar, K. P.; Falck, J. R. "Total Synthesis of Erythromycins. 5. Total Synthesis of Erythronolide A" J. Am. Chem. Soc. 1979, 101, 7131-7134.
Baeyer-Villiger
10:58 AM
C 10 -C 13
5
9
Organic Synthesis via Examination of Selected Products
8
4
2
7
H
2
1
12/21/2010
X
8
5
I 10
5
b1026
9
10
6
Organic Synthesis via Examination of Selected Natural Products
O HO
b1026_Chapter-14.qxd
536
Corey Synthesis of Erythronolide A Aglycone
b1026_Chapter-14.qxd
12/21/2010 b1026
10:58 AM
Page 537
Organic Synthesis via Examination of Selected Products
Erythromycin A Aglycone
537
Erythronolide-1 Erythromycin A (1) is a member of an important family of therapeutically useful antibiotics. The structure of 1 is shown on Erythronolide-5. This natural product consists of an “aglycone” (see 2 in Erythronolide-1) linked to two unusual monosaccharides (L-cladinose and D-desosamine) via glycosidic bonds to the C3 and C5 hydroxyl groups, respectively. We will conclude by examining two syntheses, one of erythronolide A (2) and the other of erythromycin A. Before jumping into the synthesis, I will make a few comments. Erythronolide A (2) contains seven propionic acid units linked from the α-carbon of one to the carbonyl carbon of the next unit. Using polymer terminology, 2 is an oligomer derived from propionic acid (the monomer). The hydroxy acid derived from a macrolide is refered to as the seco-acid (see Erythronolide-5 for the seco-acid of 2). Lactonization of the seco-acid is the most obvious way to approach any macrolide, although not the only way as we saw with lasonolide A (Chapter 12). To cyclize the seco-acid of 2, most of the hydroxyl groups spattered along the C1-C13 chain would have to be protected. For a long time it was felt that such lactonizations would not be feasible, but a number of macrolactonization methods have been developed over the years (see Lasonolide-9 for one method) and macrolactonization is now considered a viable step in a synthetic strategy. Erythronolide A (2) has 10 stereogenic centers distributed along the 13-carbon chain. Setting these centers with the proper relative and absolute stereochemistry is the major challenge in assembling 2 (and a myriad of other polypropionate and polyacetate derived natural products). With lasonolide A and calcimycin we saw several broad approaches to this challenge: (1) set stereochemistry in a ring and open the ring (2) set stereochemistry in intermolecular reactions (3) use convergence to advantage where possible (make small pieces and couple them using reactions that either do or do not generate stereogenic centers). We will examine two historically significant syntheses of erythronolide A that employ all three of these strategies at one point or another. The Corey group was the first to report a synthesis of 2. The plan was to cyclize an appropriately protected seco-acid using macrolactonization methodology developed in house.1 The seco-acid was to be assembled from C10-C13 fragment 4 and C1-C9 fragment 5. The smaller fragment (4) was to be prepared from enyne 6 via syn-dihydroxylation of the olefin and formal regioselective syn-addition of HI to the alkyne. The more complex fragment (5) was to be prepared by Baeyer-Villiger oxidation of 7 (migration of the more highly substituted carbon to control regiochemistry). Cyclohexanone 7
O
1. NMO, OsO4 THF-H 2O
10
1. Me
O
Li
HO
O
separated (resolution from diastereomer = 84% of theory (42%)
65% pure
93% (80% pure; 10% E -isomer)
(crystallization)
(10% 1,1-disubstituted alkene)
11
12
1. Cy 2 BH 2. Me3 NO
14
MTMO 13
OH
OTBS
B(OCy) 2
CH 3SCH2 O
MeO
2. TBSCl, 4-DMAP DMF (99%)
16 87%
15 87%
MTM = CH 3SCH2
HgCl 10
OH
O
and
via
O
NaOMe
Synthesis of C1 -C 9 17
allyl bromide, PhH, D
18 11
See J . Am. Chem. Soc. 1978, 100, 4618
H 2N
O
O 4
3
O 1
23
5
n-Bu3 SnH
H O 9
O
PhH, ∆ AIBN
10 1. BH3 -THF
Resolved with
O
Me
21
2. H2 O2 , NaOH
O
Br Br2 , KBr
H O 22
H
H 2O O
Br KOH, THF
O
20
Erythronolide-2
O
O
O
Br2 , KBr
H O
H 2O HO2 C
3. CrO3 , H 2SO4 acetone
H 2O 19
HO2 C
9
Page 538
3. Hg(OAc)2 4. NaCl 5. I2 , pyridine
4
10
O 1. KOH, MeOH (87%)
MTMO
10:58 AM
10
Organic Synthesis via Examination of Selected Products
Ac 2O, DMSO, AcOH I
b1026
14
13
12/21/2010
MeO
2. Na 2S2 O5 H2 O
6
12
HO
Organic Synthesis via Examination of Selected Natural Products
13
Acid Chloride
OH
13
12
2. TsOH, 100 oC
C 10 -C 13
b1026_Chapter-14.qxd
538
Synthesis of C10-C13
b1026_Chapter-14.qxd
12/21/2010 b1026
10:58 AM
Page 539
Organic Synthesis via Examination of Selected Products
Erythromycin A Aglycone
539
was to come from cis-oxadecalin 8, using the ring system to control introduction of stereogenic centers at C3 and C8. A very creative aspect of this approach was recognition that 8 was to be prepared from symmetrical cyclohexadienone 9. The transformation of 9 to 8 requires directed syn-hydration of the two olefins and desymmetrization at some point via lactone formation. Dienone 9 was to come from 10 which was to be prepared using a ClaisenCope rearrangement to dearomatize 11. The plans for both pieces (4 and 5) require that either asymmetric induction be used to establish absolute stereochemistry, or classical resolutions be performed along the way. In addition, a novel feature of the plan was to install the C10 and C11 stereogenic centers after macrolactonization by “hydration” of the C10-C11 olefin present in fragment 4. Let’s move to the synthesis of 4.
Erythronolide-2 Addition of 1-lithiopropyne to 2-pentanone (12) was followed by dehydration to give 6. Although this material was contaminated with some of the E-isomer and the regioisomeric dehydration product, it was possible to move forward and separate at the next step. Vicinal dihydroxylation of 6 was accomplished using a VanRheenan oxidation (developed at UpJohn in Kalamazoo) — osmium tetraoxide was used in catalytic amounts and N-methylmorpholine-N-oxide (NMO) was used as a stoichiometric oxidant. This provided racemic diol 13 which was purified by crystallization. Esterification of the least hindered alcohol with the acid chloride derived from O-methylmandelic acid, afforded 14 as a single diastereomer. Thus, a classical resolution was effective in handling the absolute stereochemistry issue in the synthesis of 4. The tertiary alcohol of 14 was protected as a methylthiomethyl (MTM) ether. Ester 15 was hydrolyzed and the liberated hydroxyl group was protected as a TBS ether, setting the stage for the formal addition of HI across the alkyne. Regioselective hydroboration of 16 (controlled by steric effects) was followed by oxidation with trimethylamine oxide to give vinylboronic acid derivative 17. Treatment of 17 with mercuric acetate and then sodium chloride gave vinyl mercurial 18, which reacted with iodine to provide the desired vinyl iodide 4. The synthesis of the C1-C9 fragment began with phenol 11 [the oxidation product derived from 1,3,5-trimethylbenzene (mesitylene)]. Reaction of 11 with allyl bromide and base in benzene at reflux gave cyclohexadienone 10. Hydroboration-oxidation of the least hindered olefin gave 9, setting the stage for the aforementioned directed hydrations. Halolactonization of 9 gave 19, and lactone hydrolysis gave epoxide 20. The sequence moving from 9 to 20
3
O 1
H 2, Pd(OH)2
H O
5
H O
HO
THF-AcOH (40:1)
BzO
3. LiOH, H 2O 4. CrO3 , H 2SO4
O
8
BzO via
O 25
O
9
24
O
CO2 H
5
O
7 S 2
Bu3 P 10
1.
BzO
Li
23 O
MTMO
13 O TBS
O O
1
(via R-I and t -BuLi)
O BzO
O O
58% + 14% C9 epimeric alcohol
BzO
OH 1 O H
9
OTBS 13
H MTMO 29
13
H
Zn(BH4 )2 DME
MTMO 78%
1. LiOH (3 eq), H2 O2 (1.1 eq) HO(CH2 )2 S(CH2 )2 OH (30 eq) 2. Pt (to decompose H2 O2 ) 3. KOH, DME (benzoate hydrolysis) 4. CH 2N 2
OTBS
9
S
27
OBz
O
1 10
H
N 2. CuI
9
OBz
BzO
HO
H
OH 5
H MTMO
O
OH OH 3
OR MeO 30
28
R = TBS (82%)
Erythronolide-3
1. pyridinium p-toluenesulfonate (PPTS) H2 C=C(OMe)CH 3 2. Ac2 O, 4-DMAP
Page 540
N 26
10:58 AM
EtOAc, 55 °C
O
BzO
OBz
Organic Synthesis via Examination of Selected Products
BzO C 1-C9
O
1
OBz
b1026
O CO2 H
O CO2 H
7
65% overall
BzO CH3 CO 3H
BzO
24
8
OBz
H O
BzO
2. (PhCO) 2O 4-DMAP O pyridine
12/21/2010
O
23
OBz 1. LDA, THF 2. MeI, HMPA
Organic Synthesis via Examination of Selected Natural Products
9
OBz 1. Zn(BH4 )2 , DMF
b1026_Chapter-14.qxd
O
540
O
b1026_Chapter-14.qxd
12/21/2010 b1026
10:58 AM
Page 541
Organic Synthesis via Examination of Selected Products
Erythromycin A Aglycone
541
destroys symmetry and provides racemic 20. The acid, however, was resolved using amine 21 to provide a single enantiomer of 20. Thus, once again a classical resolution was used to establish absolute stereochemistry. Halolactonization onto the remaining olefin provided 22, and reduction of the halide with tri-n-butyltin hydride gave 23. Whereas it is debatable whether or not the stereochemistry of the reduction of 22 was controlled by kinetics, 23 is certainly the thermodynamically most stable C4 epimer.
Erythronolide-3 Hydrogenolysis of epoxide 23 using a Pd catalyst in the presence of acid provided 8. Axial delivery of hydride to the C3 ketone (also from the convex face of the cis-oxadecalin) and conversion of the resulting alcohol to a benzoate ester gave 24. The C8 methyl group was next introduced by alkylation of the enolate derived from 24. The product was the thermodynamically more stable 25. I presume that either the C6 methyl group encouraged alkylation of the enolate via a boat-like transition state, or epimerization occurred under the basic reaction conditions to provide the product expected on the basis of thermodynamics. Hydrolysis of the lactone and oxidation of the alcohol gave ketone 7. A Baeyer-Villiger oxidation completed the synthesis of the C1-C9 fragment (5). The two fragments were coupled via a Cu mediated coupling of thioester 27 (derived from 5 and disulfide 26) and the lithium reagent derived from vinyl iodide 4. Reduction of the resulting ketone (28) provided a mixture of C9 diastereomers from which lactone 29 was isolated in 58% yield. The lactone was hydrolyzed using basic hydrogen peroxide. Excess peroxides were destroyed using Pt wire. The benzoates were hydrolyzed and the resulting acid was converted to methyl ester 30 with diazomethane. The hydroxyl groups at C3 and C5 were converted to the corresponding acetonide and the remaining secondary hydroxyl group was acetylated to provide 31a.
OH O
5
H MTMO
H
MTMO
O
O
OH 1. m-CPBA (4 eq) (β-face epoxidation)
10
HO
O
O
3. H2 (1 atm), 10% Pd/C MeOH, NaHCO3, 25 °C OH
1. 2-methoxypropene POCl3 (protect)
OH 11
2. PDC (CH 2 Cl2 ) (oxidize 2° ROH)
34
80%
OH O
O
O
30%
9
O
O
11
CH2 Cl2
O 10
O
2. Triton B (cat) MeOH (epimerize)
O
9
OH O
11
HO
1. hydroxylamine-HCl
O O
O
3. PPTS (deprotect) 92%
2. toluene, ∆
36
35
25% overall from epoxide 10-epi-erythronolide A OH O 10
HO
9
OH O
11
OH OH
erythronolide-A
O 2
Erythronolide-4
2. 3% HCl, MeOH (54%) 3. NaNO 2, MeOH, H 2O, HCl (76%)
Page 542
33
K2CO3 , MeI
2
10:58 AM
O
O
H HO
N
12/21/2010
O
32
Organic Synthesis via Examination of Selected Products
H
HO 9
O
S
R = TBS (83%)
OMTM
H MTMO
N
31b
b1026
H
O
Organic Synthesis via Examination of Selected Natural Products
OH HO
12
R = TBS (83%)
MTMO
1. Bu3 P, toluene, ∆
O
H MTMO
3. Ac2O, DMSO, NaOAc (80%) 4. NaOH, MeOH, 55 °C 5. TBAF (70%)
31a
OMTM 6
9
O
3
OR MeO
1. Ac2 O, DMSO, NaOAc (92%) 2. K2CO3 , MeOH
b1026_Chapter-14.qxd
H
542
AcO
b1026_Chapter-14.qxd
12/21/2010 b1026
10:58 AM
Page 543
Organic Synthesis via Examination of Selected Products
Erythromycin A Aglycone
543
Erythronolide-4 The next task was to modify 31a such that the macrolactonization could be accomplished. The ultimate cyclization substrate was hydroxy acid 31b. An MTM protecting group was first introduced at C6. The C9 acetate was hydrolyzed and replaced with an MTM group. The methyl ester was hydrolyzed and then the silicon protecting group at C13 was removed. Treatment of 31b with disulfide 32 and tri-n-butylphosphine provided the hydroxy thioester, and heating in toluene under high dilution conditions gave the macrocyclic lactone 33 albeit in modest yield. We later see that this is a very demanding macrolactonization! Lactone 33 is not conformationally mobile because many conformational changes introduce transannular and torsional strain. Thus, it was possible to functionalize the C10-C11 olefin with predictable control over stereochemistry. In preparation for this chemistry, the MTM groups were first removed by S-alkylation, followed by hydrolysis using potassium carbonate in aqueous acetone. The choice of the MTM protecting group was most likely because these deprotections had to be accomplished in the presence of the lactone and acetal groups. Triol 34 was epoxidized with m-CPBA (hydroxyl-directed epoxidation), the secondary alcohol at C9 was oxidized to a ketone, and the epoxide was hydrogenolyzed to give 10-epi-erythronolide A (35). The C10C11 vicinal diol group was protected as an acetonide, C10 was epimerized to the thermodynamically more stable pseudo-equatorial isomer, and the acetonide was removed to provide 36. The keto-acetonide was unstable to acid and thus, the 9-keto group was protected as an oxime prior to acetonide hydrolysis. Finally, oxidative removal of the oxime provided erythronolide A (2). Overall this was a superb plan. I imagine this summary does not do justice to the trial and error and effort that must have gone into translating this plan into reality.
b1026_Chapter-14.qxd
544
Woodward Synthesis of Erythronolide A D-desosamine
10
8 11
HO
O 1
12
O
2
9 10
5
7 8
6
OH OH O
L-cladinose
S
MeO
Erythromycin A
RO
O
OH
seco-acid
OH OH O 37
S
S 10
12
13
1 2
11
+ 9
O
7 6
RO degrade CH2OBn
38
From Streptomyces erythreus
8
OBn CHO
H
O
S 5 4
O
3
OBn
38
configurationally stable due to instability of thioaldehyde S 1. (CH2OH)2 TsOH, PhH, ∆
S
2. NCS, CCl4 O 39
Cl H N 2
S
acetone, 25 oC O
O 40
S
S
NH2
NH NH2
O
O 41
S
1. NaOH, 25 oC, H2O 2. HCl, THF, H2O
MeO
3. (MeO)3CH, TsOH, MeOH
SH 1. (-)-camphanyl chloride OMe
42
2. crystallization*
SH
S
MeO
OMe
3. NaOMe, MeOH
43
*configuration by crystallography OMs
1. MeOH, H2SO4
CO2H OBn 44
2. LDA; HCO2Me
1. (MeO)3CH, TsOH, MeOH
CO2Me OHC
OBn 45
2. LiAlH4 3. MsCl, pyridine, 0 oC
Erythronolide-5
MeO
OBn
1. ROMs, NaH, DMSO, rt 2. AcOH, H2O, rt
OMe 46
next
Page 544
C8 Enolate from C7 Ketone
C9 Aldehyde
10:58 AM
HO
H
3 4
Organic Synthesis via Examination of Selected Products
1
HO 11
13
O
14
O
12
HO
O
5
b1026
9
6
NMe2 OH
O
12/21/2010
O HO
OH
Organic Synthesis via Examination of Selected Natural Products
H
b1026_Chapter-14.qxd
12/21/2010 b1026
10:58 AM
Page 545
Organic Synthesis via Examination of Selected Products
Erythromycin A Aglycone
545
Erythronolide-5 I will return to Woodward for the final synthesis of this book. This synthesis of Erythromycin A was published after his death in 1978. It is a synthesis in the true Woodwardian style (after all it is a Woodward synthesis), making elegant use of symmetry, tying a complex molecule into knots and then springing loose the target as an intimate part of the endgame. I remember hearing Woodward lecture on this topic at the 1977 National Organic Symposium held at the University of West Virginia in Morgantown, the only time I heard the master give a lecture. He used slides, an unusual practice for Woodward, who was famous for chalk-talks where he never had to erase anything. The overall strategy was to cyclize an appropriate seco-acid derivative prior to installation of the monosaccharides. The seco-acid derivative was to be prepared from 38. It was hoped that 38 would provide both the C3-C8 and C9-C13 portions of the seco-acid. The marriage of 38 with itself was to involve construction of the C8-C9 bond via an aldol condensation (the original plan involved nitrile oxide cycloaddition chemistry to join the two pieces, but the aldol is what ultimately worked). Detection of identical fragments (C4-C6 and C10-C12) within the seco-acid was critical to development of this plan. It was also planned to set most of the stereogenic centers in conformationally welldefined molecules. The methyl groups at C4, C6, and C8 as well as C10, C12, and the terminal methyl of the C13 ethyl group, were joined to accomplish this task. Because of the convergent nature of the synthesis, the synthesis of 38 demanded enantioselectivity. The synthesis of 38 began with 39. Acetal formation and chlorination adjacent to sulfur provided 40. Thiourea was used to introduce sulfur. Hydrolysis of 41 provided the free thiol and a ketal exchange (hydrolysis-protection) gave 42. This compound was configurationally stable at the “anomeric center” and thus, was resolved via the thioester derived from reaction with (−)-camphanyl chloride. The absolute configuration of the proper enantiomer was established by X-ray crystallography of this thioester. SAlkylation of 43 with racemic mesylate 46 provided a mixture of diastereomers 47 (Erythronolide-6).
H
S
S
alumina, EtOAc, rt OBn O
(OMe)2
HO
52B
B
70% combined
S
S
HO
11
13
OBn
H
S 10
12
OMOM
1. NaBH4
S
2. KH, ICH2 OMe
13
11
O
38 where R = MOM
OBn
OBn
H
S
OBn 10
OMOM 51
50
1. OsO4 , Et2 O, rt
S
H
52A
S
OBn 10
13
2. NaHSO 3 pyridine, rt
MOMO HO OH 52
TsOH CH2 Cl2
1. TFA, CH2 Cl2 , rt 2. TFAA, DMSO, CH2 Cl2 3. i-Pr2 NEt (Hunig's base)
S
H
7 8
O
1. o-NO2 PhSeCN Bu3P, THF, rt 2. H2 O2 , THF
S 11
OBn
4
O
13
3
5
O
0 oC
MOMO
55
O
10
13
CHO 9
O
3. O3 , MeOH
MOMO
W-2 Raney Ni
11
O
10
O
H
13
OH EtOH, ∆
MOMO
S 11
O
OBn
10
O
4. Me2 S
Li
1. THF, -50 °C 2. RCHO 3. TFAA, DMSO; i-Pr 2 NEt, -60 0 °C
S
54 S 13
MOMO
9
11
O
10
O
38 where R = MOM
53
7 8
O 56
H
O
S 3
5
O
OBn
4
O 76%
1. KH, HMPA, THF 2. AcCl
S 13
3. NaBH4 , MeOH, CH 2 Cl2 MOMO 4. MsCl, pyridine, 0 °C; 4-DMAP, pyridine, MeOH, 30 °C
Erythronolide-6
9
11
O
O
7 8
10
H
O 57
S 3
5
O
4
O
OBn
Page 546
Me2 C(OMe)2
10:58 AM
S
O
S
49
12/21/2010
O
OH
Organic Synthesis via Examination of Selected Products
MOMO
OH OMOM
OBn
1. MsCl, pyridine, rt 2. alumina, EtOAc
S 10
O
48
H
b1026
13
OH
A
There is some interesting organocatalysis of this aldol condensation presented in the paper, but we will just continue with isomer A. H
S
Organic Synthesis via Examination of Selected Natural Products
47
S
OBn
H
OBn
OH
S
+
1:1
O
H
b1026_Chapter-14.qxd
S
S
546
H
S
b1026_Chapter-14.qxd
12/21/2010 b1026
10:58 AM
Page 547
Organic Synthesis via Examination of Selected Products
Erythromycin A Aglycone
547
Erythronolide-6 Treatment of 47 with alumina gave a separable 1:1 mixture of intramolecular aldol condensation products 48 and 49. The required isomer (48) was converted to the mesylate and dehydration afforded 50. Reduction of 50 from the convex face of the dithiaoctalone gave 51 after protection of the alcohol with a MOM group. Vicinal dihydroxylation of the olefin from the most accessible face, and acetonide formation, completed the synthesis of 38 (R=MOM). The next portion of the synthesis involved conversion of 38 into derivatives that would be suitable for joining via the aforementioned aldol condensation. Thus 38 was desulfurized and debenzylated using Raney-Ni to provide alcohol 53. The primary alcohol was subjected to the GriecoSharpless dehydration procedure (via the selenide and selenoxide) and resulting olefin was cleaved with ozone to provide aldehyde 54 (the C9-C13 fragment).2 The C3-C8 fragment was prepared from 38 by removal of the MOM group, and oxidation of the secondary alcohol to a ketone using a Moffat-type oxidation. Ketone 55 and aldehyde 54 were then joined via a crossed aldol condensation. The resulting alcohol was oxidized to give 56. The enolate derived from β-dicarbonyl 56 was O-acylated at C9. The C7 carbonyl group was reduced to the alcohol oxidation state with concomitant reductive cleavage of the enol acetate. Treatment of the resulting β-hydroxy ketone with mesyl chloride gave 57.
MOMO
9
11
O
O
7 8
10
S 3
5
O
O
RSH, n-BuLi OBn
4
THF, -50 °C
O
8
MOMO
O
O
O H
H S S SR 7
O
1. LiAlH4 , Et2O, -20 °C OBn 2. Ac2 O, 4-DMAP, CH2 Cl2
O
8
O
O AcO H
7
O
1. W-2 RaNi, EtOH, ∆ 2. o-NO2 PhSeCN, Bu3 P, rt OBn
O
92%
3. H2 O2 , THF, rt 4. O3 , MeOH, CH 2Cl2 5. Me2 S, NaHCO3 , -78 °C
CH3 CH(Li)COStBu
13
MOMO
O
O
O AcO
CHO O
rt
59
3
(with LDA in THF)
60
66%
wrong stereochemistry
3
MOMO
O
O AcO
O
2
O
COStBu
OH
1. t-BuLi, TMEDA, THF, -110 °C 2. AcOH, -110 °C
61
13
MOMO 62
85% (Cram product)
O
O AcO
O
2
O
COStBu
OH
90% (X-ray) + 8% C 2 epimer
Woodward, R. B.; et al (including T. V. Rajanbabu) "Asymmetric Total Synthesis of Erythromycin. 1. Synthesis of Erythronolide A Seco Acid Derivative via Asymmetric Induction" J. Am. Chem. Soc. 1981, 103, 3210.
Erythronolide-7
Page 548
MOMO
H S S SR
10:58 AM
must excise
Organic Synthesis via Examination of Selected Products
C 8 stereochemistry known; C 7 stereochemistry unknown; C 8 stereochemistry set in kinetic protonation of conjugate adduct.
b1026
58
12/21/2010
83%
Organic Synthesis via Examination of Selected Natural Products
R = Bn 57
b1026_Chapter-14.qxd
13
H
548
S
b1026_Chapter-14.qxd
12/21/2010 b1026
10:58 AM
Page 549
Organic Synthesis via Examination of Selected Products
Erythromycin A Aglycone
549
Erythronolide-7 The next stage of the synthesis required reduction of the C7-C8 double bond with control over stereochemistry at C8. The tactics ultimately used to accomplish this transformation involved conjugate addition of thiophenoxide to the enone to provide 58 with C7 stereochemistry that was never established. The critical stereochemistry (C8), however, was clean and presumably controlled by kinetic protonation of the intermediate enolate. Reduction of the C9 ketone was followed by esterification to provide acetate 59 as a single stereoisomer (C7 stereochemistry still not defined). Reduction of the C7 thiol was followed by excision of the extra carbon in the usual manner to provide aldehyde 60. The final carbons of the seco-acid were introduced via crossed condensation of the enolate derived from a thioester of propionic acid, with aldehyde 60. This reaction provided the proper stereochemistry at C3, but the undesired stereoisomer at C2. The C2 stereochemistry was corrected by kinetic protonation of the enolate derived from 61 with acetic acid. The structure of the resulting seco-acid derivative (62) was established by X-ray crystallography.
b1026_Chapter-14.qxd
550
Woodward, R. B. et al "Asymmetric Total Synthesis of Erythromycin. 2. Synthesis of an Erythronolide A Lactone System" J . Am. Chem. Soc. 1981, 103, 3213
degradation
O
O
O
O
O
70%
9
OH
10
63
5
O 11
HO
O
3
14
O
1
2
64 Important Features: (1) S-configuration at C 9 (2) C3,5 and C9,11 must be cyclic acetals. Eventually the Woodward group settled on the f ollowing compound which was prepared f rom the natural product and cyclized in 70% yield.
O 11
HO
OH
9 10
O
NH
HO
5
3
O
O
S
N
Toluene, 110 °C
O 70%
O
HN 9 10
HO 12
65
OH 5
O 11
3
14
O
1
66 The next task was to convert the "bis-acetonide" into the cyclization substrate.
Erythronolide-8
O
O 2
O
Page 550
12
O
O
10:58 AM
O
Toluene, 110 °C
N
12/21/2010
S HO
Organic Synthesis via Examination of Selected Products
HO
9 10
b1026
11
OH
Organic Synthesis via Examination of Selected Natural Products
17 Substrates (all thiopyridyl esters) of which only 3 could be lactonized. Two gave low yields but ...
Erythromycin
b1026_Chapter-14.qxd
12/21/2010 b1026
10:58 AM
Page 551
Organic Synthesis via Examination of Selected Products
Erythromycin A Aglycone
551
Erythronolide-8 In principle, all that remained to do to reach erythronolide (2) was to deprotect the C13 hydroxyl group of 62, conduct the macrolactonization, remove the acetonides, and carry out an oxidation state adjustment at C9. Of course this was easier said than done. The macrolactonization turned out to be very difficult. The Woodward group degraded Erythromycin A (1) to 17 thiopyridyl ester substrates, only three of which underwent lactonization under the Corey-Nicolaou conditions. Two of these derivatives gave low yields, but 63 cyclized to 64 in good yield. This effort established that (1) the S-configuration was essential at C9 and (2) the C3-C5 and C9-C11 diol units had to be protected as cyclic acetals. Based on this information, carbamate-acetal 65 was eventually prepared from the natural product and found to cyclize to 66 in 70% yield. Thus, to complete a total synthesis, it was necessary to convert bis-acetonide 62 to 65, and move 66 forward to erythromycin A (1).
62
O
O
O
OMs 67
CO2 Me O
OH
75%
11
HO
9
5
HO
OH HN O
CO2 Me OH
O
HO
Et3N, CH2 Cl2 , rt
OH
11
HO
68
9
O
NH O
69
5
HO
3
CO2 Me OH
OH
70% from azide
NO 2
HO
HO 11
9
5
ArCH(OMe)2 , TFA HO
O
NH O
85%
HO
3
CO2 Me O
O
11
1. EtSLi, HMPA 2. Et3 N, CH 2Cl2 N
S
Cl O
70
HO
9
O
NH
HO
5
3
O
O
S
N
O
O 65
For lactonization methodology see Corey, E. J.; Nicolaou, K. C. "An Efficient and Mild Lactonization Method for the Synthesis of Macrolides" J. Am. Chem. Soc. 1974, 96, 5614.
Erythronolide-9
Page 552
3. p-NO2 PhOCOCl, CH2 Cl2 NaHCO3 (amine acylation) 4. NH 2OH-HCl, KH 2PO4 H2 O, MeOH (MOM removal; NH2 OH traps CH2 =O)
3
10:58 AM
HO
Organic Synthesis via Examination of Selected Products
1. LiN 3, HMPA, H 2O, 60 °C (mesylate azide) (75%) 2. H2 , PtO2 , THF, rt (azide amine)
MOMO
b1026
3. MsCl, pyridine, 0 °C (C 9 mesylation) 4. LiOH, 30% H2 O2 , THF, rt (C 3 deprotection)
OH
12/21/2010
O
b1026_Chapter-14.qxd
O AcO
COStBu
Organic Synthesis via Examination of Selected Natural Products
MOMO
1
3
O
552
9
O
1. Na 2CO3 , MeOH, rt (transesterif ications) 2. (PhOCH 2CO) 2O, pyridine 4-DMAP, CH2 Cl2 (C 3 protection)
b1026_Chapter-14.qxd
12/21/2010 b1026
10:58 AM
Page 553
Organic Synthesis via Examination of Selected Products
Erythromycin A Aglycone
553
Erythronolide-9 The 12-step conversion of 62 to 65 began with methanolysis of the C1 thioester and C9 acetate. The C3 hydroxyl group was then selectively protected as the α-phenoxyacetate. The C9 hydroxyl group was converted to a mesylate and the C3 ester was removed to provide 67. The mesylate was displaced with azide, the azido group was reduced to the amine, the amine was acylated, and the acetonide and MOM protecting groups (all acetals) were removed under mild acidic conditions. The resulting polyol 68 cyclized to carbamate 69 upon treatment with triethylamine. The C3-C5 acetal was installed to give 70. The methyl ester was then converted to the corresponding acid and then thioester 65.
HO 12
3
14
O
1
3 Steps
O
O
O
70%
HO
6 Steps
OH OH
O O
2
D-desosamine H 2N
H OH
OH
H
O
O
1. NCS, pyridine (N-chloroamine)
3. H2 O, 5 °C (imine hydrolysis) MeO HO
10
X 12
NMe2 OH
O 5
9 11
2. AgF, HMPA (imine)
O
72
OH
O HO
O O
HO
NMe2 OH
O
O
O O
1
2
L-cladinose MeO
40% 1
Erythromycin
Erythronolide-10
O
O
14
HO
Page 554
71
66
10:58 AM
11
12/21/2010
O
b1026_Chapter-14.qxd
9 10
OH
NH OH
5
Organic Synthesis via Examination of Selected Products
HN
OH
b1026
O
O
Organic Synthesis via Examination of Selected Natural Products
We will not discuss the completion of the synthesis in detail. The fundamental operations, however, can be summarized as follows: (1) removal of the cyclic carbamate and acetal groups maintaining carbamate protection of the C 9 amino group (2) sequential introduction of the desosamine and cladinose units (3) oxidation of the C 9 amine to the required C9 ketone via N-chloroamine and imine intermediates.
554
Woodward, R. B. et al "Asymmetric Total Synthesis of Erythromycin. 3. Total Synthesis of Erythromycin" J. Am. Chem. Soc. 1981, 103, 3215-3217.
b1026_Chapter-14.qxd
12/21/2010 b1026
10:58 AM
Page 555
Organic Synthesis via Examination of Selected Products
Erythromycin A Aglycone
555
Erythronolide-10 I will not discuss the completion of the erythromycin A synthesis in detail. The fundamental operations, however, can be summarized as follows. The cyclic carbamate and acetal protecting groups of 66 were removed while maintaining protection of the C9 amino group. The C5 and C3 hydroxyl groups of 71 were sequentially glycosylated with suitable desosamine and cladinose derivatives, respectively. The “sugars” were deprotected and the C9 amide was hydrolyzed to provide 72. Finally the C9 amino group was converted to the C9 ketone via dehydrohalogenation of an intermediate chloramine and hydrolysis of the intermediate imine (recall Corey prostaglandin synthesis). Several other syntheses of erythromycins have been reported. These all contain unique features and good lessons for students of organic synthesis. I encourage you to look some of these up and perhaps prepare your own flow sheets and commentary to add to this chapter.3
b1026_Chapter-14.qxd
12/21/2010 b1026
556
10:58 AM
Page 556
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
References 1. Corey, E. J.; Nicolaou, K. C. “An Efficient and Mild Lactonization Method for the Synthesis of Macrolides” J. Am. Chem. Soc. 1974, 96, 5614–5616. 2. Grieco, P. A.; Gilman, S.; Nishizawa, M. “Organoselenium Chemistry. A Facile One-Step Synthesis of Alkyl Aryl Selenides from Alcohols” J. Org. Chem. 1976, 41, 1485–1486. Sharpless, K. B.; Young, M. W. “Olefin Synthesis. Rate Enhancement of the Elimination of Alkyl Aryl Selenoxides by ElectronWithdrawing Substituents” J. Org. Chem. 1975, 40, 947–949. 3. Bernet, B.; Bishop, P. M.; Caron, M.; Kawamata, T.; Roy, B. L.; Ruest, L.; Sauve, G.; Soucy, P.; Deslongchamps, P. “Formal Total Synthesis of Erythromycin A. Part I. Total Synthesis of a 1,7-Dioxaspiro[5.5]undecane Derivative of Erythronolide A” Can. J. Chem. 1985, 63, 2810–2814 and following two papers in CJC. Kinoshita, M.; Arai, M.; Ohwawa, N.; Nakata, M. “Synthetic Studies of Erythromycins. III. Total Synthesis of Erythronolide A Through (9S)Dihydroerythronolide A” Tetrahedron Lett. 1986, 27, 1815–1818 preceeding two articles in this series. Stork, G.; Rychnovsky, S. D. “Concise Total Synthesis of (+)-(9S)-Dihydroerythronolide A” J. Am. Chem. Soc. 1987, 109, 1565–1567. Kochetkov, N. K.; Sviridov, A. F.; Ermolenko, M. S.; Yashunsky, D. V.; Borodkin, V. S. “Stereocontrolled Synthesis of Erythronolides A and B from 1,6-Anhydroβ-D-glucopyranose (levoglucosan). Skeleton assembly in (C9-C13) + (C7-C8) + (C1-C6) Sequence” Tetrahedron 1989, 45, 5109–5136. Hikota, M.; Tone, H.; Horita, K.; Yonemitsu, O. “Stereoselective Synthesis of Erythronolide A via an Extremely Efficient Macrolactonization by the Modified Yamaguchi Method” J. Org. Chem. 1990, 55, 7–9. Hikota, M.; Tone, H.; Horita, K.; Yonemitsu, O. “Chiral Synthesis of Polyketide-Derived Natural Products. 31. Stereoselective Synthesis of Erythronolide A by Extremely Efficient Lactonization Based on Conformational Adjustment and High Activation of Seco-Acid” Tetrahedron 1990, 46, 4613–4628. Stürmer, R.; Hoffmann, R. W. “Stereoselective Synthesis of Alcohols. XLVIII. Linear Synthesis of (9S)-Dihydroerythronolide A” Chem. Ber. 1994, 127, 2519–2526. Muri, D.; Lohse-Fraefel, N.; Carriera, E. M. “Total Synthesis of Erythronolide A by Mg(II)-mediated Cycloadditions of Nitrile Oxides” Angew. Chem., Int. Ed. 2005, 44, 4036–4038.
b1026_Chapter-14.qxd
12/21/2010 b1026
10:58 AM
Page 557
Organic Synthesis via Examination of Selected Products
Erythromycin A Aglycone
557
Problems 1. Suggest why 6 is the major product from the dehydration of the adduct of 12 and 1-lithiopropyne. (Erythronolide-2) 2. The conversion of 14 → 15 is an example of a Pummerer reaction. Provide a mechanism for this transformation. (Erythronolide-2) 3. Provide a mechanistic explanation for the conversion of 16 → 17. (Erythronolide-2) 4. Provide a mechanism for the conversion of 11 → 10 that begins with the expected O-alkylation. (Erythronolide-2) 5. Propose a mechanism for the removal of the MTM groups en route from 33 → 34. (Erythromycin-4) 6. Propose a mechanism for the conversion of 39 → 40. (Erythronolide-5) 7. What is the purpose of the trimethyl orthoformate in the conversions of 41 → 42 and 45 → 46? (Erythronolide-5) 8. Provide a mechanistic explanation for the stereochemistry of the crossed condensation reactions that converted 60 → 61 → 62. (Erythronolide-7)
12/21/2010
10:58 AM
Page 560
Organic Synthesis via Examination of Selected Products
MeO
O
Concluding Remarks
3 colchicine
Monensin
Discodermolide
2
OH 1
OH CO2H HO OMe
O OH O O O H
NH2 OH MeO
OCH3
O H O H NHAc
O HO MeO
O H
HO
HO
Organic Synthesis via Examination of Selected Natural Products
b1026
b1026_Concluding_Remarks.qxd
560
Additional Targets
b1026_Concluding_Remarks.qxd b1026
12/21/2010
10:58 AM
Page 561
Organic Synthesis via Examination of Selected Products
Concluding Remarks
561
Concluding Remarks This ends our look at some selected natural product syntheses. At the onset of this course I had intended to cover three additional natural products: the alkaloid colchicine (1), the ionophore monensin (2) and disocodermolide (3). Time did not allow me to cover these topics. It turns out that an analytical and up-to-date compilation of colchicine syntheses has been published, and monensin has also been nicely covered elsewhere.1,2 Discodermolide is a modern example where synthesis met the demand for supply of a complex natural product with medicinal promise. I encourage students and teachers alike to write an additional chapter that discusses syntheses of discodermolide within the context of what appears in Chapters 1–14. This is only a suggestion, and I am sure that you will all eventually have your favorites that you can add to the collection of work described in this book. For those of you beginning careers in organic synthesis, I think it is important to learn from others (see how practitioners of synthesis have solved problems), and to develop and improve your skills through practice. In this regard, I asked students to work a few problems during the course of the quarter (about 2% of what appears in this book) and also to propose a synthesis of a natural product. For teachers, I think it is important to provide students with guidance. For this class, I provided students with one way (there are many) to identify a target for synthesis. I chose a method that was well-defined and was manageable within the context of a course (I directed students to the website of the journal Heterocycles for a searchable compilation of natural product structures). I provided students with help navigating the website, feedback on their target selection, and also provided feedback on their synthesis plan at an intermediate point in the course. I hope you find this information useful.
References 1. Graening, T.; Schmalz, H.-G. “Total Synthesis of Colchicine in Comparison: A Journey Through 50 Years of Synthetic Organic Chemistry” Angew. Chem. Int. Ed. 2004, 43, 3230–3256. 2. Nicolaou, K. C.; Sorensen, E. J. “Classics in Total Synthesis: Targets Strategies, Methods”, VCH, 1996.
b1026_Concluding_Remarks.qxd b1026
12/21/2010
10:58 AM
Page 562
Organic Synthesis via Examination of Selected Products
This page intentionally left blank
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
Page 563
Organic Synthesis via Examination of Selected Products
Index
The following index can be used to find topical information within the text. A comprehensive index of compounds, reagents, reactions and topics covered in graphical form on even-numbered pages is available at the following website: http:// www.worldscibooks.com/chemistry/7815.html relationship between prostaglandins and pyrrolizidines, 125 Acyclic stereogenic centers, control of, 29 Acyl anion equivalents, 83, 185 metallated, dithiane as, 77, 517 metallated cyanohydrins as, 103 metallated dihydropyran as, 527 metallated furans as, 55 Acylnitroso compounds, in approach to pumiliotoxin-C, 369 Aglycone, 537 Aldol condensation, 269 and Cram’s rule, 525 and the Felkin-Ahn rule, 525 intramolecular, vinylogous, 81 intramolecular, 77, 79 minimization of β-elimination during, 77
Absolute stereochemistry, of Cecropia juvenile hormone, 453 Absolute stereochemistry, reagent based control of, 185 Acetoacetic ester synthesis, of substituted acetones, 53, 451 Acetylide addition, to ketone, 39 Activated cyclopropanes, opening with thiolate anion, 123 Activating groups (SPh), 509 Acyclic diastereoselection, 3, 5, 13 and the prostaglandins, 123 early application to synthesis, 455 and juvabione, 5 and prostaglandins, 5 and pyrrolizidine alkaloids, 5 and steroids, 5 relationship between prostaglandins and steroids, 125 563
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
564
Page 564
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
versus alkylation for carbon-carbon bond formation, 107 Aldol-dehydration, double, intramolecular, 51 Aldol-dehydration, intramolecular, 27, 39 Aldols, synthesis of using crotylation reactions, 527 Alkaloids, total synthesis of, 281 Alkenes, as latent carbonyl groups, 233 Alkylation deconjugative, 35, 37 diastereoselective of ester enolates with arene-chromium tricarbonyls, 179 intramolecular of enamine, 376 intramolecular, of enolate, 253, 261 of enolate, stereoelectronics of, 513 of enolates, 1,2-asymmetric induction model, 157 of imidate anion, 519 versus aldol for carbon-carbon bond formation, 107 Alkyne, semi-hydrogenation to alkene, 105 Allenol ethers, synthesis and hydrolysis to cis-unsaturated aldehyde, 377 Allylation, Pd-mediated, 189 Allylic oxidation, using Corey-Fleet reagent, 427 Allylic rearrangement, promoted by HBr, 39 Allylic strain as a conformational control element, 385 importance in 1,2-asymmetric induction model for enolate alkylation, 157
in planning synthesis of gephyrotoxin, 379 importance of in route to pumiliotoxin-C, 369 role in determining stereochemistry of Diels-Alder, 361 Allylsilanes, model for asymmetric addition to enones, 193 Alternate or ambident functional groups (A-functions), examples of, 209, 211 Aminoketone, β-, construction of via Mannich reaction, 293 Analgesics (painkillers), 405 Androsterone, 29 Annulation reactions (annelation reactions), 23 Anodic oxidation, of amides, 289 Anomeric effect, 505 Anomeric effect, double as stereocontrol element in N,O-acetal formation, 289 in spiroacetals, 523 Anthranilic acid, 371 Arachidonic acid, 73 Arene-chromium tricarbonyl complex, reaction with ester enolates, 179 Aspirin, 75 Asymmetric conjugate addition catalytic, of methyl group to enone, 373 of amide nucleophile to unsaturated ester, 373 Asymmetric Diels-Alder, predictive model for, 93 Asymmetric induction, 1,2-, 159 Asymmetric induction, Cornforth model, 455 Felkin-Ahn model, 455
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
Page 565
Organic Synthesis via Examination of Selected Products
Index
in conjugate addition reactions, 191 in intramolecular cyclopropanations, 139 relative, 95 Asymmetric reduction, reagent controlled, 61 Asymmetric synthesis, of aldols using crotylation reactions, 527 of prostaglandins, 101 the chiral pool approach, 181 Aziridinium ions, preparation and chemistry of, 435, 437, 439 Azomethine ylids, method of generation for use in cycloadditions, 149 Baeyer-Villiger oxidation, 187, 235, 309, 327, 511, 523, 539, 541 Baeyer-Villiger oxidation, regioselective, 85 Baker’s yeast, 187, 195 Bamford-Stevens reaction, 287 Barbier-Wieland degradation, 29 Barton reaction, 105, 339, 381 Batrachotoxin, 337 Beckman rearrangement, 235, 339 Biomimetic synthesis, 41, 291 of morphine, 7, 413 of pumiliotoxin-C, 367, 369 of steroids, 3 Biosynthesis of morphine, 411, 413 proposal for Dentrobatid alkaloids, 367 Biosynthetic pathway, 3 Bischler-Napieralski reaction, 299, 305 Blocking groups, 25 Brown allylation, 351
565
Cahn-Ingold-Prelog (CIP) convention, 167 Calcimycin (A23187), 11, 505-529 Carbene generation, by α-elimination reaction, 433 Carbocation, vinylic, nucleophilic capture of, 57 Carbocycle synthesis, 3 Carbohydrates, as source of chirality, 517 Carbon-hydrogen activation, via Barton reaction, 339 Carbon-hydrogen insertion, of carbene (vinylidene) in approach to morphine, 433 Carbopalladation, approach to prostaglandins, 111 Cecropia juvenile hormone, 11, 445474 Celebrex, as COX inhibitor, 75 Charge affinity inversion (charge reversal), 211 Chiral auxiliary, for Diels-Alder reaction, 91 phenylglycinol as in synthesis of pumiliotoxin-C, 367, 369 proline as in synthesis of pumiliotoxin-C, 371 Chiral pool approach to asymmetric synthesis, 181, 183, 185, 187 definition, 103 in approach to lasonolide A, 485 in calcimycin synthesis, 517 Chiral reducing agent, design of, 97 Chiral, crotylborane, 527 Chirality transfer, issues of geometry in Ireland-Claisen, 513 Chirality, transfer of, 121
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
566
Page 566
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
Chloroacrylonitrile, α-, as ketene equivalent in Diels-Alder, 85 Chloronitrile, α-, conversion to ketone, 85 Cholesterol, 3, 17, 19 Cholesterol, biosynthesis of, 41 Cholesterol, synthesis of, 29 Cholic acid, 17 Cladinose, L-, in erythromycin A, 537 Claisen condensation, crossed, 311 Claisen condensations, iterative in biosynthesis of polyacetates, 483 Claisen rearrangement, 197, 343 and transfer of chirality, 121, 513 as enforced SN2′ reaction, 55, 451 enolate, 513 how to see potential for use in synthesis, 451 Johnson-Faulkner, 55 stereochemical course of, 171, 451 Codeine, as cough suppressant, 405 Codeine, conversion to morphine, 411, 413, 421 Colchicine, 561 Collins oxidation, 509 Computer-assisted synthesis design, 282, 283 Configurational stability, of S,Shemiacetal, 543 Conformational analysis in allylic systems within context of ester enolate alkylation, 57 of allylic alcohols, 471 of N-acyltetrahydroisoquinolines, 417 of substituted cyclohexane, 379, 381 Conformational mobility, 543
Conformations, of allylic alcohols and effect on epoxidation stereochemistry, 471 Conjugate addition, to enone for introduction of PG sidechain, 101 Conjugate addition-enolate trapping, 105, 373 Convergence, importance in planning a synthesis, 493 Convergent synthesis, 517, 527, 537 comparison with linear synthesis, 99 importance in synthesis design, 204, 205 of prostaglandins, 105 Cope elimination, 143, 267 Cope rearrangement, 311 in plan for synthesis of reserpine, 309 stereochemistry of, 171 Corey lactone, for prostaglandin synthesis, 89 Corey-Fuchs reaction, 349 Corey-Nicolaou lactonization, 549 Corey-Rucker reagent, in Peterson olefination, 389 Corey-Suggs reduction, 115 Corey-Winter reaction, 263 Cornforth model, for 1,2diastereoselection, 455 Corticosteroids, introduction of C11 oxygen, 31, 35 Corticosteroids, preparation by polyolefin cyclization, 59 Cortisone, 17, 19, 31, 39 Crabtree’s catalyst, for directed hydrogenation, 371 Cram selectivity, 509, 525 Crotylation-oxidative cleavage, strategy for polypropionate synthesis, 527
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
Page 567
Organic Synthesis via Examination of Selected Products
Index
Crotylborane, chiral, 527 Cuprate addition, stereoelectronics of, 517 Curtin-Hammett Principle, 295 Curtius rearrangement, 267, 353 to establish ketene equivalency, 91 Cyanohydrin, metallated as acyl anion equivalent, 103 Cyclic structures, for controlling multiple acyclic stereogenic centers, 515, 521 Cyclization, base-initiated, 523 Cycloaddition, [2+2] of ynamine with enone, 169 Cycloalkene synthesis, general strategy, 103 Cycloalkenes, as latent dicarbonyl compounds, 233 Cyclohexenes, conversion to acylcyclopentanes, 20 Cyclohexenone, from cyclopentene, 51 Cyclooxygenase (COX) in biosynthesis of prostaglandins, 75 inhibitors of, 75 Cyclopentane, from cyclohexene, 77 Cyclopropanation, of alkenes, 123 Cyclopropanation, stereochemistry of intramolecular carbene-alkene reactions, 139 Cyclopropane, geminally activated, 127, 139, 141 Dart-poison frogs, 337 Decahydroquinoline alkaloids, 359 Deconjugation, of enone, 29, 107 Demerol, as analgesic, 405 Desosamine, D-, in erythromycin A 537
567
Dess-Martin periodinane oxidation, 493 Desulfurization, 253 Desymmetrization in enzyme-mediated reduction, 35 of cyclohexanone, 165 in synthesis of erythronolide A, 539 Deuterium labeling experiments, in design of calcimycin synthesis, 507 Diastereoselection, in carbene-alkene addition reactions, 139 reagent control of, 95 relative within context of prostaglandin synthesis, 95 substrate control of, 95 Diastereoselective hydrogenation, in cyclic system to control stereochemistry, 177 Diastereoselectivity, 1,2-, 513 Diastereoselectivity, 1,4-, 513 Diastereoselectivity, in crotylation reaction, 527 Diastereoselectivity, in prostaglandin synthesis, 119 Dicarbonyl compounds, 1,4-, synthesis of, 55 Diels-Alder reaction, 21, 165, 197 aromaticity as a driving force, 393 in plan for synthesis of reserpine, 309 ketene equivalents, 83 Lewis acid promoted, 83, 91 lowering temperature through use of more reactive dienophile, 91 of nitroalkene and butadiene, 77 of acylnitroso compounds, 145 rate as a function of dienophile, 91
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
568
Page 568
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
retro, for generation of acylnitroso compound, 147 stereocontrol in, 85 Diels-Alder-retro-aldol, strategy for controlling vicinal stereochemistry, 165 Diels-Alder-retro-Diels-Alder, in plan for synthesis of reserpine, 315 Dienamide synthesis, 361, 363 Difunctional relationships, 5, 7, 205 1,2-, construction of, 219 1,3-, construction of, 217 1,3-, relevance to Mannich reaction, 281 1,4-, and construction of 5membered rings, 227 1,4-, construction of, 221, 223 1,5-, and construction of 6membered rings, 225, 227 1,5-, construction of, 223 and construction of cyclohexanones, 225 and construction of cyclopentanones, 227 classification of, 213 generation by insertion reactions, 235 importance in retrosynthetic analysis, 247 importance in synthetic plan for porantherine, 281 importance in synthetic plan for porantheridine, 287 in analysis of lasonolide A substructures, 483 in approach to pumiliotoxin-C, 371 in Diels-Alder route to pumiliotoxin-C, 365
in retrosynthetic analysis of twistane, 247 in synthesis of triquinacene, 271 reinforcing and interfering, 215 Dipolar cycloaddition, 1,3importance of reversibility in nitrone approach to HTX, 357 azomethine ylid in approach to pyrrolizidine alkaloids, 149 of nitrones in approach to pyrrolizidine alkaloids, 143 Dipole-dipole repulsion, importance in asymmetric Diels-Alder, 93 Directed reactions, 307 addition of free radicals, 116 addition to unactivated alkenes, 111 hydrogenation, heterogenous, 376 hydrogenation, in synthesis of pumiliotoxin-C, 371 Dodecahedrane, relationship to triquinacene, 265 Electrochemistry, anodic oxidation of amides, 289 Electrophilic addition, iodolactonization, 87 Electrophilic addition, selectivity between alkenes, 39 Electrophilic functional groups (E-functions), examples of, 209 Elimination, β-, avoidance of, 87 Enantiomer production, of natural products depending upon plant source, 377, 379 Enantiomers, separation by chromatography of chiral support, 429
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
Page 569
Organic Synthesis via Examination of Selected Products
Index
Enantioselective reduction, using enzymes, 35 Enantioselective synthesis, of twistane using enzymatic reduction, 255 Enantioselectivity, in crotylation reaction, 527 Ene-reactions, intramolecular of Nacylnitroso compounds, 345 Enolate alkylation, stereochemical course of, 513 Enolate protonation, kinetic control of stereochemistry in, 549 Enolate-Claisen, importance of enolate geometry, 513 Enyne synthesis, via Peterson-type olefination, 383 Enzymatic reduction, 35 of ketone to alcohol using Baker’s yeast, 187, 195 using horse liver alcohol dehydrogenase (HLAD), 255 Epimerization and thermodynamics, to control stereochemistry in spiroacetals, 529 Epoxidation, stereochemistry in corticosteroid synthesis, 13 Epoxide opening, regioselectivity in, 351 Equilenin, 3, 19 Equivalency concept iodonium ion as proton equivalent, 87 ketene equivalents, 83 1,2-bis-sulfonylethylene as ethylene equivalent, 391 of terminal alkene to aldehyde, 283 Erythromycin, 13, 537-555 Eschenmoser sulfide contraction, 383 Eschweiler-Clark methylation, 267 Estrone, 3, 19
569
Felkin-Ahn model, for 1,2diastereoselection, 455, 509, 525 (in aldol) Finkelstein reaction, 321, 455 Formyl anion equivalent, lithiated dithiane as, 219 Free radical addition approach to prostaglandins, 115 intermolecular, 115, 117 Free radical allylation, in approach to perhydrohistrionicotoxin, 345 Free radical cyclization conformational analysis of, 485 of N-centered radical in morphine synthesis, 425 stereochemistry of, 321 Free radical intermediate, to avoid βelimination, 87 Functional group relationships, and appearance of synthetic approaches, 395 Functional group transformation (FGT), to establish equivalency, 85, 219 Functional groups, importance of compatability in synthesis design, 205 Functionality, level of, 20 Furans, metalated, as acyl anion equivalents, 55 Furst-Plattner rule in epoxide openings, 295 variation of for nucleophilic opening of bromonium ion, 303 Fusidic acid, 17 Geminally activated cyclopropanes, reactions with amines, 139, 141 Gennari-Still reaction 487, 495
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
570
Page 570
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
Gephyrotoxin, 9, 374-395, 337 difunctional relationships in, 375 enantioselective synthesis, debate over absolute stereochemistry, 377 structure and determination of absolute configuration, 377 Grieco-Sharpless method, formal dehydration of terminal alcohol, 485, 547 Grob fragmentation, relationship to Julia olefin synthesis, 459 Hastanecine, 141 Heck reaction, intramolecular, as oxidative phenolic coupling alternative, 427 Henry reaction, 77 example of nitro group as an Nfunction, 209 Heroin, as analgesic, 405 Heterogenous catalytic hydrogenation, stereocontrol of, 376, 377 Histrionicotoxin, 7, 335-357, 337 enantioselective synthesis via double-alkylation strategy, 351, 353 of ene-yne substructure via Sonogashira reaction, 351 synthesis of ene-yne substructure via Wittig reaction, 349 synthesis via intramolecular nitrone cycloaddition, 355, 357 Histrionicotoxins, 337 Homosteroid, D-, 20, 33 Horner-Wadsworth-Emmons reaction, 81, 363, 447, 497, 523, 525 Gennari-Still modification of, 487, 495 Hydration of nitriles, catalytic, 371
Hydrazones, as A-functional groups, 211 Hydrogenation, semi, of alkyne, 39 Hydrogenation, stereochemistry of, 25 Imide, size as a cyclohexane substituent, 381 Imides, ketone-like reactivity of, 375 Iminium ion, stereoelectronic model for addition of nucleophile to, 285, 289 Insect sex hormones, 447 Insertion reactions, for generation of difunctional relationships, 235 Intramolecular alkylation, of enamine, 393 conjugate addition, to construct pyrrolidine, 387 delivery of nucleophile in epoxide opening, 301 delivery, of nucleophile via N,Nacetal formation, 295 Ionophores, as a family of natural products, 505 Ionophores, introduction to, 497 Ireland-Claisen, importance of enolate geometry, 513 Isomerization, of cyclopentadienes, 83 Isomerization, Pd-mediated, 37 Isonitriles, as free radical traps, 117 Isotope dilution method, application to morphine synthesis, 413 Iterative reactions, in Cecropia juvenile hormone syntheses, 473 Iterative synthesis, 449 of trisubstituted olefins, 463 Johnson-Faulker Claisen, 451 Johnson-Lemieux oxidation, 283, 437 Jones oxidation, 297
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
Page 571
Organic Synthesis via Examination of Selected Products
Index
Julia olefin synthesis, 453, 457 relationship to Grob fragmentation, 459 variations of, 457 Julia-Lythgoe-Kocienski olefin synthesis, 485, 489, 491, 493 Juvabione, 5, 29, 159-197 determination of absolute configuration via synthesis, 181 Juvenile hormones, 159, 447 Ketene equivalent chiral, 93 unsaturated ester as, 91 α-chloroacrylonitrile as in DielsAlder reaction, 293 α-chloroacryloyl chloride as, 91 for Diels-Alder reaction, 83 Ketone reduction, axial delivery of hydride, 29 Kinetic control, of stereochemistry in enolate protonation, 311, 549 Kornblum oxidation, 219, 437 Lanosterol, 41 Lasolocid A (X537A), as an ionophore, 509 Lasonolide A, 11, 479-497, 537 biological activity of enantiomers, 481 Latent C2-axis of symmetry, in design of calcimycin synthesis, 507 Latent carbonyl groups, alkenes as, 233 Laudanosine, 411 LeChatelier’s Principle, 21 Limonene, R-, 181, 183 Linear strategy, 517 Linear synthesis, comparison with convergent synthesis, 99
571
Luciduline intermediate, structural relationship to reserpine intermediate, 309 Lycopodium alkaloids, examples of, 293 Lysine, as chiral pool starting material for porantheridine, 287 Macrolactonization, 497, 537, 543, 549, 551 Macrolides, 11 Mannich reaction and 1,3-difunctional relationships, 281 double intramolecular in sparteine synthesis, 291 in alkaloid synthesis and biosynthesis, 281 Mass spectrometry, in structure determination of dart-poison alkaloids, 337 McGarvey-Fleming model, for asymmetric alkylation of enolates, 157 Mechanisms, importance in synthesis design, 204, 205 Meerwein-Pondorf-Verley reduction, 301 Methadone, as analgesic, 405 Michael addition, deconjugative, 33 Michael reaction, 23 Michael-Michael reaction, of dienolate with acrylate as Diels-Alder equivalent, 327 Midland reduction, 61 Mislow-Evans rearrangement, and transfer of chirality, 121 Mitsunobu reaction, 107, 419, 423 Moffatt-type oxidation, 487
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
572
Page 572
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
Monensin, as an ionophore 505, 509, 561 Morphine, 7 Morphine, 403-439 and analgesics, 405 biosynthesis of, 411, 413 synthesis via intramolecular electrophilic aromatic substitution, 435 synthesis via perhydrophenanthrene-based approach, 429 N,O-relationships, 1,3-, in alkaloids, 297 N-acyliminium ion cyclizations in Mannich-type reactions with alkenes, 341 in synthesis of gephyrotoxin, 379 N-acylnitroso compounds, intramolecular ene reactions of, 345 Nef reaction, for conversion of N-function to E-function, 209, 221 Neighboring group participation, 301 Neuroscience, importance of dartpoison frogs to, 337 Nitrile chemistry, comparison of pumiliotoxin-C and reserpine syntheses, 367 Nitroalkanes as A-functions, 209 importance of tautomers, 209 Nitrone cycloaddition intramolecular, 297 in pyrrolizidine alkaloid synthesis, 143 to establish 1,3-N,O-relationship, 297
NMR spectroscopy, in structure determination of dart-poison alkaloids, 337 Nonactin, as an ionophore 509 Nucleophilic functional groups (N-functions), examples of, 209 Olefin geometry, importance in establishing vicinal stereochemical relationships, 447 Olefin metathesis, 497 Olefin stereochemistry, as function of boat versus chair transition states in oxy-Cope, 175 Olefin synthesis classification of according to method of construction, 474 trisubstituted, 11 Organopalladium chemistry, 111, 113 Oxetane formation, in competition with Grob fragmentation, 461 Oxidation state adjustment of in PG synthesis, 105 adjustments, in synthesis of reserpine, 317 changes at carbon, 21, 23 importance of in synthesis of reserpine, 311 role in selection of starting materials for synthesis, 185 Oxidation, 511 Oxidation, conversion of amine to ketone, 79 Oxidation, Saegusa, 117 Oxidative cleavage of alkene to ketoaldehyde, 79 Oxidative phenolic coupling a mechanistic interpretation, 413 alternatives to, 419
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
Page 573
Organic Synthesis via Examination of Selected Products
Index
regiochemical problems in synthesis of morphine, 415, 417 site selective equivalent via aryl diazonium chemistry, 415 Oxime derivatives, as A-functional groups, 211 Oxy-Cope rearrangement in synthesis of luciduline, 293 stereochemistry of, 171 Ozonolysis, 523, 527 Painkillers (analgesics), 405 Perhydrohistionicotoxin retrosynthetic analyses, 339 via N-acyliminium ions and Nacylnitroso compounds, 339 Perhydroindan, 20 Perhydronaphthalene (decalin), 20 Perillaldehyde, 185 Periodate cleavage, of vicinal diol, 525 Peterson olefination, 357 PGA, structure of, 77 PGA1, synthesis of, 79 PGA2, the problem of acyclic diastereoselection, 123 PGE1, structure of, 77 PGE1, synthesis of, 79 PGE2, structure of, 75 PGE2, synthesis of, 109 PGF1α, synthesis of, 79 PGF2α, synthesis of, 117 PGF2α, synthesis of, 87 PGF2β, synthesis of, 79 PGG2, structure of, 75 PGH2, structure of, 75 Photochemistry, in triquinacene synthesis, 269, 273 Photocycloadditions, directed and intramolecular, 307 Pictet-Spengler reaction, 299, 329
573
Pictet-Spengler reaction in biosynthesis of morphine, 411 Polyacetates, natural products derived from acetic acid, 483 Polyketides, 13 Polymers, polypropionates as, 537 Polyolefin cyclizations, 43 acetal as initiator, 51 alkynes as terminators, 53 allylic alcohol as initiator, 49 effect of C11 substitutents in approach to corticosteroids, 59 rates as a function of substituents, 63 termination with an allylic silane, 63 termination with ethylene carbonate, 57 termination with nitroethane, 57 Polypropionate natural products, 13, 505, 537 calcimycin as, 505 introduction to, 497 repeating unit in, 527 synthesis via crotylation-oxidative cleavage strategy, 527 Porantherilidene, 285 Porantherine, 281 Practical synthesis, of juvabione, 63 Principle of vinylogy, 375 Progesterone synthesis, analysis in terms of difunctional relationships, 229 Progesterone, 17 Propargylic alcohols, use in trisubstituted olefin synthesis, 463 Prostaglandin synthesis, analysis in terms of difunctional relationships, 231
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
574
Page 574
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
Prostaglandins, 3, 71-127 from arachidonic acid, 73 strategies for synthesis of, 99 Protecting group selection, compatability in complex synthesis, 545 Pumiliotoxin, comparison of DielsAlder routes to, 359, 365 Pumiliotoxin-C, 7, 358-373 Pummerer rearrangement, 219 Pyrrolizidine alkaloids, 5, 137-151 Quinic acid, as starting material for reserpine, 321 Reaction classification, 207 Reaction conditions, as tool for achieving thermodynamic or kinetic control, 329 Reaction sequencing, importance of in reserpine synthesis, 327, 329 Reactive intermediates, tactics for generation of, 145, 147 Reagent-controlled asymmetric synthesis, in allylation reactions, 491 Recycling, of stereoisomeric alcohols via oxidation-reduction, 87 Reduction of alkynes, control of stereochemistry, 171 of iodide in presence of ester and lactone, 87 selective, of nitroalkane, 79 Reductive homologation, of ketone, 337 Regiochemistry, of N-acyliminium ion cyclizations, 343 Regioselective hydration of alkene, directed, 87 Relative asymmetric induction, 191
Relative stereochemistry, use of cyclic compounds to control, 521 Relay synthesis, 27 of erythromycin A, 551 of morphine, 409 Research, the search part of, 387 Reserpine intermediate, structural relationship to luciduline intermediate, 309 Reserpine, 9, 299-329 as substituted cyclohexane, 299 introduction to, 299 synthetic plan, 299 Reserpine-isoreserpine, stereochemical relationship between, 313, 319, 325 Resolution in enantioselective approach to twistane, 257 of acid as ephedrine salt, 89 of ketone via chiral acetal formation and diastereomer separation, 433 of S,S-hemiacetal in synthesis of erythronolide A, 545 Reticuline, 411 Retroaldol condensation, 165, 167 Retro-Aldol condensation, in plan for synthesis of reserpine, 307 Retro-Claisen condensation, 169 Retro-Diels-Alder, to generate Nacylnitroso reactive intermediate, 345 Retronecine, 143, 145 Retrosynthetic analysis, 20, 205 Ring expansion, cyclopentanone to lactam via Beckmann rearrangement of nitrone, 439 Ring opening reaction, of cyclobutane with relief of strain, 169
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
Page 575
Organic Synthesis via Examination of Selected Products
Index
Ring synthesis five-membered, 3 six-membered, 3 Robinson annulation, application to perhydrophenanthrene approach to morphine, 429 Saegusa oxidation, 117 Sakurai, reaction, 193 Salutaridine, 413 Schlosser modification of Wittig reaction, 55 Schlosser’s base, 527 Seco-acid cyclization of, 545 definition of, 537 Semi-hydrogenation, of alkyne, 141 Sex hormones, 17 Sharpless epoxidation, 351 Sigmatropic rearrangements 2,3- as formal SN2′ reactions, 467471 and transfer of chirality, 121 Singlet oxygen, use in prostaglandin synthesis, 101 SN2′ reactions Claisen rearrangements as equivalent of, 343 in approaches to Cecropia juvenile hormone, 467 Solvolysis of allylic alcohol, 49 of unsaturated nosylate, 45 Sonogashira reaction, 351 Sparteine, biomimetic synthesis of, 291 Spiroacetals, 505 thermodynamic control over stereochemistry of, 507 and the anomeric effect, 505 Squalene oxide, 43
575
Squalene, 41 Stacking, π-π, importance in asymmetric Diels-Alder, 93 Starting materials, importance of recognition in synthesis design, 204, 205 Stereochemical control, importance of kinetics and thermodynamic considerations, 325 Stereochemical relationship, control of 1,2 by addition reactions, 123, 139 1,4 by transfer of chirality, 123 by thermodynamics or kinetics, 37 vicinal relationships through DielsAlder reaction, 167 vicinal relationships, 139 through bowl-shaped nature of reactant, 189 through molecular shape, 301, 321 starting material vs thermodynamics, 521 Stereochemistry importance of controlling relative, 29 in formation of bromohydrin, 39 kinetic vs thermodynamic control of, 435 of addition of acetylide to steroidal ketone, 39 of enolate alkylation, 513 of iminium ion reduction in reserpine synthesis, 305 reagent control of, 97 remote control of in hydrogenation, 383 thermodynamic control in spiroacetal formation, 507 thermodynamic control of epimerization, 527 thermodynamic control over, 505
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
576
Page 576
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
vicinal, control through kinetic protonation, 169 vicinal, the problem of control in acyclic systems, 161 Stereocontrol acyclic using ring-opening strategy, 537 acyclic and cyclic, 3 importance of order of operations, 45 importance of, 25 of cyclic stereogenic centers by olefin geometry, 53 Stereoelectonics analysis of imine addition reactions, 385 considerations, 329 control in nucleophilic opening of epoxide, 295, 301 control, of stereochemistry in ketone reduction, 541 model for addition of nucleophile to imine, 285 of N-acyliminium ion cyclizations, 341, 343 of cuprate addition, 517 Stereoelectronic effects, on reserpineisoreserpine oxidation-reduction chemistry, 319 Stereorandom synthesis purpose of, 447, 451 value to medicinal chemistry, 81 Stereoselective olefin synthesis, 447 Steric effects, in cis-decalin conformational equilibria, 295 Steroid side chain problem, 157, 159 Steroid synthesis annulation strategy, 35 industrial scale, 35 Steroids, 3
Steroids, ring juncture stereochemistry, 17 Stitching methodology, using boranes, 185 Stork-Danheiser synthesis, 327 Stork-Eschenmoser hypothesis, 43 Strategy, 13, 205 3-component coupling for PG synthesis, 105 allylation-oxidation for use in polyacetate synthesis, 491 for control of vicinal stereochemistry, 119 for molecules containing multiple acyclic stereogenic centers, 515 for synthesis of lasonolide A via intramolecular Stille coupling, 483 for synthesis of lasonolide via macrolactonization reaction, 495 Structure determination, by synthesis, 11, 447, 449, 481 Sulfonium salts, as A-functional groups, 211 Sulfoxides, as A-functional groups, 211 Sulfur-containing heterocycles, use to set stereochemistry in acyclic systems, 473 Swern oxidation, 425, 491 Symmetry as a consideration in retrosynthetic analysis, 375 in approach to porantherine, 281 in synthetic approaches to triquinacene, 265, 269, 273 use of in erythronolide A strategy, 539 Synthesis and structure determination, 11
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
Page 577
Organic Synthesis via Examination of Selected Products
Index
as template for reaction development, 179, 185 as tool for understanding reaction stereochemistry, 193 biomimetic, 3, 41 importance of mechanistic, stereochemical principles during design, 204, 205 selection of key intermediates, 19 to meet demand for supply of compound, 561 Tactics, 13, 205 improvement of for problems in prostaglandin synthesis, 89 Tamao-Fleming oxidation, 327, 485 Tandem reactions, 197 aldol-dehydration-alkylation approach to morphine, 431 radical cyclization-additionelimination approach to morphine, 423 reduction-hydrogenolysis-imine formation-reduction, 363 stereochemistry controlled by thermodynamics versus kinetics, 523 sulfone addition-alkylation approach to morphine, 419, 421 Target-oriented synthesis, necessary considerations, 205 Tautomerization, importance of in determining functional group reactivity, 213 Terpenoids, side chain stereochemistry problem, 5 Testosterone, 29 Thermodynamic control of epimerization stereochemistry, 527
577
of stereochemistry in ketone alkylation, 541 of stereochemistry, 521 as a stereochemical control element, 305 control of intramolecular aldol regiochemistry, 79 control over stereochemistry, 505 for controlling spiroacetal stereochemistry, 507 Thiele’s acid, 269 Three-component coupling, strategy for PG synthesis, 99, 107, 109 Thromboxanes, 93 Tin hydride reductions, catalytic in tin, 93 Todomatuic acid, 163 Torsional strain importance in enolate alkylations, 157 in macrocycle, 543 Total synthesis, as a tool for methodology development, 151 Transannular strain, in macrocycle, 543 Transfer of chirality, 1,2- to 1,4, 513 Transfer of chirality, 121, 513 in Claisen rearrangement, 513 Triquinacene, 7 Triquinacene origin of 5-membered rings in synthesis of, 267, 269 relationship to dodecahedrane, 265 synthesis of, 265-273 Trisubstituted olefin synthesis, GennariStill compared with normal HWE reactions, 487 Twistane, 7 Diels-Alder route to, 249, 257 retrosynthetic analysis of, 247 synthesis of, 249-263
b1026_Index.qxd
12/21/2010
10:58 AM
b1026
578
Page 578
Organic Synthesis via Examination of Selected Products
Organic Synthesis via Examination of Selected Natural Products
Umpolung, 211 VanRheenan oxidation, 541 Vicinal diastereoselection, in addition of metallated vinyl sulfoxide to enone, 195 Vicinal dihydroxylation, with OsO4, 39 Vicinal stereochemistry control by ring-opening strategy using Baeyer-Villiger, 177, 187 control of by addition to alkenes, 123 control of in oxy-Cope rearrangement, 175 control through alkene addition reactions, 119 control through cycloaddition-ring opening strategy, 169 Vinylogy, the principle of, 215 Vioxx, as COX inhibitor, 75 Water scavenger, use to improve yield in sensitive Diels-Alder reaction, 361
Weinreb amide synthesis, 485 Weiss reaction, in triquinacene synthesis, 271 Williamson ether synthesis, 317, 453, 455 Wittig reaction, 187, 261, 289, 349, 353, 489, 495, 517, for introduction of prostaglandin sidechain, 87 modifications of in stereoselective olefin synthesis, 465 Schlosser modification of, 55, 465 stereochemistry of, 55 Wolf-Kishner reduction, 249, 257, 291, 409 Yamamoto-Peterson olefination (ene-yne synthesis), 383 Ynamine cycloaddition, 169 Z-Olefin synthesis, via Wittig reaction of unstabilized phosphoranes, 55
