cover_image

Table of Contents

Related Titles

Title Page

Copyright

Dedication

Foreword

Acknowledgment

Preface

Abbreviations and Acronyms

1: Introduction

1.1 Intermediate in the Total Synthesis of Brevetoxin A [16]

References

2: The Inertia of Conventional Functional Groups

2.1 Welwitindolinone A Isonitrile [1]

2.2 “Overprotecting-Maverick” Protecting Groups

3: The Diels–Alder Reaction

3.1 (−)-Platensimycin [2]

3.2 Stephacidins A and B [13]

: The Aldol Condensation

4.1 The ABC Ring Moiety of (−)-Norzoanthamine [1]

4.2 (+)-TMC-151C [15]

5: Cyclizations: Concerted, Radical, or Polar?

5.1 (±)-Merrilactone A [3]

5.2 (−)-Pseudolaric Acid B [14]

5.3 Formal Synthesis of Platensimycin [28]

6: Macrocycles: from Reluctant Ring Closure to Reluctant Ring Opening

6.1 Ustiloxin D [2]

6.2 Largazole [10]

7: Stereochemistry, Controlled or Uncontrolled?

7.1 (−)-Samaderin Y [3]

7.2 (±)-Lyconadin A [10]

7.3 Vannusal B [21]

8: Transition Metal-Mediated Transformations vs “Conventional” Reactions

8.1 Blumiolide C [5]

8.2 Arylomycin A2 [17]

8.3 (−)-Strychnine [30]

9: When Metathesis Fails

9.1 (+)-Saxitoxin [6]

9.2 Laulimalide Family [19]

10: Oxidations in C–C Bond Building

10.1 Cortistatins A and J [2]

10.2 Gambierol ABC Moiety [9]

References

11: The Failure of Epoxide Ring Opening and the Limits of Cascade Reactions

11.1 C′D′E′F′ Domain of Maitotoxin

References

Index

Related Titles

Sierra, M.A., de la Torre, M. C.

Dead Ends and Detours

Direct Ways to Successful Total Synthesis

2004

ISBN: 978-3-527-30644-7

Frenking, G., Shaik, S. (eds.)

Chemical Bonding – Fundamentals and Models

2013

ISBN: 978-3-527-33314-1

Christmann, M., Bräse, S. (eds.)

Asymmetric Synthesis II

More Methods and Applications

2012

ISBN: 978-3-527-32921-2

Ding, K., Dai, L. (eds.)

Organic Chemistry – Breakthroughs and Perspectives

2012

ISBN: 978-3-527-33377-6

Nicolaou, K. C., Chen, J. S.

Classics in Total Synthesis III

Further Targets, Strategies, Methods

2011

ISBN: 978-3-527-32958-8

Warren, S., Wyatt, P.

Organic Synthesis

The Disconnection Approach Second Edition

2009

ISBN: 978-0-470-71236-8

Title Page

Para Virginia y Javier. Sois nuestra mayor alegría”

Ana, Unai eta Annierentzat, maite zaituzten belarrimotxaren partez”.

Foreword

Not just Don Quixote, but we also live by our stories; narrative is the human condition. When the hunters of yore returned from their day's work, yes, they did tell their brethren how to reach the bison grounds – across two creeks, then left at the red rocks, climb a jagged outcrop, beyond is the valley where the bison graze. So the facts that allowed the next hunt were conveyed. That night, by the fire, the hunters told of the slippery rocks along the way, the cold, the water ford that turned so much deeper on the way back, the wrong turn taken, and a sheer, unscalable cliff. And the next week, one of them went into the secret cave at Altamira, drew that bison, and blew ochre over his hand.

We, the hunters of molecules, builders of astounding structures, are not as different from our ancestors as we imagine. True, in much of what we do, our proclivity to tell stories may have been beaten out of us. “That's not science, can't you convert it into a mathematical expression?” Yet storytelling, plain and scientific, is natural. And in synthetic organic chemistry, it is innate.

Synthesis, especially organic synthesis, is the most rational of chemical activities. Yet its logic is not mathematical – in fact, synthesis vanquishes the mathematicizing fashion of most of science and restores story telling to its proper role. Oh, there is essential mathematics, in the weighing out of reagents, and in the theory of the physical tools, such as NMR and X-ray diffraction, that are absolutely necessary to assign structures. And there is thermodynamics, often qualitative, behind synthetic methodology.

But look at the language of syntheses, and you see structures and stereochemistry. And reagents and reactions. Algebra may have triumphed elsewhere, but the making of molecules is a very geometrical and descriptive undertaking. In what is above and below the arrow is the Heraclitean essence of chemistry – substances and their transformations, change that is fundamental. Note incidentally the simple, unidirectional arrows, the arrows of Paul Klee; chemists well aware of ubiquitous equilibrium, nevertheless desire a product. They know how to perturb any equilibrium so they get what they want.

The story telling – well, it's just built into the enterprise. To every making of a molecule there is a beginning, a journey, and an end. The curious feature of synthetic story telling, one that sets it apart from a story by, say, Nikolai Leskov, is that time and time again the end is set out in plain view at the beginning of the story. There stands the molecule to be made, often a natural product of some real or suspected utility. An unreligious modern-day Grail, to be sure, but indeed a grail. The achievement is the path, the way of making that molecule. Nowhere else in the world of stories or science is there such a repeated, direct emphasis on the path taken, the journey, as in synthetic organic chemistry.

Once in a blue moon there is a plot twist – the molecule carefully synthesized, its structure identical to the putative structure of the natural product, turns out to be different. The original structural assignment was wrong.

Any path is a story. The way is difficult; were it not so, others would have surely trodden it earlier. This is not El Camino de Santiago, some centuries down the line, but the first ascent, by Maurice Herzog and Louis Lachenal, of Annapurna. A misstep, and there goes that asymmetric center, hopelessly racemized. Why the heck did that Michael addition fail? In seminars, more than in their papers, chemists other than synthetic chemists relax and tell a story. But organic chemists engaged in synthesis, by virtue of their métier, quite naturally trace the steps of a synthesis, tell a story. Every time they write.

The universe is made of stories,

not of atoms.

Muriel Rukeyser (1913–1980), The Speed of Darkness.

Atoms too, like human beings, sometimes refuse to cooperate. This is a book of dead ends and obstacles in the course of syntheses. Not the authors' syntheses, but those of others. And not just the journeymen and women work of organic chemistry, but the oeuvre of the master craftspeople of the profession. How come we know of the obstacles they encountered? Because the makers have recounted them. Why did they tell them, wouldn't it be better if they were perfect, if the plan was infallible? No, because the masters are people. And it is the nature of journeys, all journeys, that wrong turns will be made, even by the cleverest of researchers. Not everything is foreseen.

What follows can be described in a number of ways. The overcoming of obstacles, finding a different way to make that CC bond surrounded by the most interfering of functional groups, enhances the story. Mind you, no one will tell a synthetic tale with 10 detours. You would begin to doubt the intelligence of the designer, and in a profession where reason and design have been deified, that won't do. But a couple of sheer cliffs climbed is good for the story. Overcoming the obstacles adds to the image one wants to convey, of the one whose mind was flexible, prepared to recognize chance. Never mind the despair of the graduate student who ran into the cliff, commanded to run that recalcitrant reaction a hundred times under different conditions. And how often do you see in these papers a postdoc given by name explicit credit for suggesting the solution?

If everything turned out the way it was planned, the synthesis would get done more expeditiously. But the story would be all the poorer. And under what circumstances does a one-pot synthesis become praiseworthy?

I love this book just because it focuses on the dead ends, because it excavates and brushes off affectionately the things that went wrong in the “best-laid schemes o' mice an' men.” It also undertakes to comprehend why those schemes “gang aft agley,” as Robert Burns wrote. To do this, “More Dead Ends and Detours” brings to bear on the detours that it reveals original quantum mechanical (or molecular mechanics) calculations by the authors. In this way, it makes a gentle case for the necessity of such calculations to enter the toolkit of the synthetic chemist. I worry a little about the replacement of one kind of certainty (that the reaction will form that bond, say) by another (this calculation really explains why it did not happen). But I concur, theory can help.

Que quando una puerta se cierra otra suele abrir la fortuna.

When one door closes, fortune will usually open another.

Fernando de Rojas (1465–1538), La Celestina, Act 15.

A hundred years later, when Cervantes wrote Don Quixote, Rojas's observation was familiar enough to be described as a proverb (Chapter 21 of the First Part). Or maybe it was a proverb all along. With the truth-telling power characteristic of proverbs.

Failings are human, but they are overridden by what makes us optimistic about humanity, the great truth of human creation. Especially the improvization that follows confrontation with adversity. The making of molecules is a very special kind of building – the construction of complex microscopic structures effected on a macroscopic scale. It is different from other kinds of building – of a museum or a castle, or a relationship between two people – even as it shares an aesthetic dimension with all constructive acts. In surmounting obstacles, the organic chemist is a superb creator. “We shall overcome.” And we do, wonderfully crafting, no hype needed, the most beautiful stories of our time.

Roald Hoffmann

May 2013

Acknowledgment

I am grateful to Georgios Markopoulos for giving me the line quoted from Muriel Rukeyser's poem and to Jeff Seeman for reminding me of the rare structure misassignment revealed by synthesis.

Preface

The proverb “segundas partes nunca fueron buenas” (sequels were never good) written by Miguel de Cervantes in Chapter 4 of the second part of Don Quixote, expresses the distrust of we, Spaniards, in abounding in a topic. However, 10 years after publication of the book dead ends and detours: direct ways to successful total synthesis (Wiley 2004), we consider that subsequent emphasis on this topic would be useful for the chemical community. The reason is that 10 years in Science is an eternity. In particular, the Organic Synthesis of this second decade of the XXI century is evolving by giant leaps in terms of creativity and intellectual innovation. These days we are facing a tremendous crisis, which began as an economic fall down and probably will end as a worldwide economic and social restructuration. Within this demanding context, many solutions to technical problems that were previously acceptable will not be useful anymore. Chemistry has to keep pace and contribute to what the society will require, but that must not occur at just any price. It has to be done efficiently.

How can Organic Synthesis fulfill its part in this overambitious and broad target? There are probably several answers to this question, but the simplest way to envisage most of them is by integrating the many tools we now have in hand, into more efficient ways of working. Chemical synthesis of new or known molecules will require to meet more and more demanding efficiency criteria: better yields, larger scales, improved selectivities, atom and solvent economy, better biological and environmental profiles, etc. These requirements will in turn result in additional hurdles that have to be overcome. Therefore, predictability of chemical transformations will be an increasingly relevant aspect of Organic Chemistry in a wider sense. And, once again, total synthesis of complex molecules constitutes a formidable benchmark to put new or known methodologies to their limits.

The aim of this book is to show how, 10 years later, well-designed synthetic routes can still find unexpected issues and dead ends that in turn force detours or even complete new designs. As an additional feature, in this new book, we will try to convince the reader that the fusion of experimental and computational chemistry can provide realistic and accurate answers to both general and particular problems in multistep Organic Synthesis. We are not talking about future possibilities or desirable trends: we think that in the years to come we will witness more and more examples of the synergy between theory and experiment in Organic Chemistry.

Therefore, this book is not a mere sequel of our original text of 2004, showing just dead ends and detours in multistep organic synthesis. This book has a more ambitious goal, well beyond the description of a casuist, which is intrinsic to the contemporary Organic Synthesis. In fact, we have selected several examples of textbook fundamental organic reactions that have failed when carried out in complex molecules. These failures elicit substantial modifications or even truncate a superb synthetic planning. To reinforce the underlying argument behind this book, namely that organic synthesis can benefit from this computational experimental approach, we have carried out specific calculations to try to unveil the reasons behind every pitfall. The results of our calculations show that, in many cases, a few hours or days of computer time would have foreseen the experimental difficulty behind a problematic synthetic step. In addition, the knowledge thus obtained can serve to anticipate similar problems in other synthetic endeavors. For this reason, we have included a take-home messages section at the end of each chapter. It is our feeling that the full implementation of computational chemistry in daily experimental work would downsize the risk of unexpected surprises in multistep organic synthesis. The result will be just efficiency. As a corollary to the above argumentation, the synergy between the experiment and the computation will end, beyond any doubt, in the design of innovative or even counterintuitive strategies for the synthesis of complex molecules.

We would like to thank Prof. Roald Hoffmann for his beautiful and inspiring foreword. In his text, he writes about Chemistry and Organic Synthesis and its resonance with several aspects of human nature such as the fascination for pathfinding, art, and storytelling. Please read the book as you would a story and decide if you like it.

June 2013

Miguel A. Sierra

María C. de la Torre

Fernando P. Cossío

Abbreviations and Acronyms

Ac acetyl
AdNu nucleophilic addition
AIBN 2,2′-azobisisobutyronitrile
AZADO 2-azaadamantane-N-oxyl
9-BBN 9-borabicyclo[3.3.1]nonane
BHT 2,6-di-t-butyl-4-methylphenol
BINAP 2,2′-bis(diphenylphosphino)-1,1′-binaphthyl
BINAPO 2-diphenylphosphino-2′-diphenylphosphinyl-1,1′-binapthalene
Bn benzyl
BOC or Boc tert-butoxycarbonyl
BOM benzyloxymethyl
BOP benzotriazole-1-yl-oxy-tris-(dimethylamino)-phosphonium hexafluorophosphate
BOPCl bis(2-oxo-3-oxazolidinyl)phosphonic chloride
BtOH N-hydroxybenzotriazole
s-Bu sec-butyl
t-Bu tert-butyl
CAN ceric ammonium nitrate
Cbz benzyloxycarbonyl
CDI carbonyldiimidazole
m-CPBA or MCPBA 3-chloroperbenzoic acid
CSA 10-camphorsulfonic acid
Cy cyclohexyl
dba tran,trans-dibenzylideneacetone
DBU 1,8-diazabicyclo[5.4.0]undec-7-ene
DCC 1,3-dicyclohexylcarbodiimide
DCE 1,2-dichloroethane
DCM dichloromethane
DDQ 2,3-dichloro-5,6-dicyano-1,4-benzoquinone
DEAD diethyl azodicarboxylate
DET diethyl tartrate
DIAD diisopropyl azodicarboxylate
DHP 3,4-dihydro-2H-pyran
DIBAL or DIBALH diisobutylaluminium hydride
DIC N,N′-diisopropylcarbodiimide
DIPA diisopropylamine
DIPEA N,N-diisopropylethylamine
DMAP 4-dimethylaminopyridine
DMDO dimethyldioxirane
DMF dimethylformamide
DMM dimetoxymethane
DMP Dess-Martin periodinane
2,2-DMP 2,2-dimethoxypropane
3,5-DMP 3,5-dimethylpyrazole
DMS dimethyl sulfide
DMSO dimethyl sulfoxide
dppf 1,1′-bis(diphenylphosphino)ferrocene
d.r. diastereomeric ratio
EDC 1-ethyl-3(3-dimethylaminopropyl)-carbodiimide
esp α,α,α′,α′-tetramethyl-1,3-benzenedipropionic acid
Et ethyl
Fm fluorenylmethyl
Fmoc 9-fluorenylmethyloxycarbonyl
GbII Grubbs catalyst second generation
HATU 2-(7-aza-1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate
hfacac hexafluoroacetylacetonate
HMDS hexamethyldisilazane
HMPA hexamethylphosphoramide
HOBt or HOBT N-hydroxybenzotriazole
IBDA iodosobenzene diacetate
IBX 2-iodoxybenzoic acid
IMDA intramolecular Diels-Alder
imid imidazole
KHMDS potassium bis(trimethylsilyl)amide
LHMDS lithium bis(trimethylsilyl)amide
LiDBB lithium di-tert-butylbipheny
LDA lithium diisopropylamide
Mbs p-methoxybenzenesulfonyl
Me methyl
Mes 2,4,6-trimethylphenyl
MOM methoxymethyl
MW or Mw microwave irradiation
NBS N-bromosuccinimide
NCS N-chlorosuccinimide
NIS N-iodosuccinimide
NMO N-methylmorpholine-N-oxide
NMP N-methyl-2-pyrrolidinone
n.O.e. nuclear Overhauser effect
Ns nitrobenzenesulfonyl (nosyl)
OPCW Organization for the Prohibition of Chemical Weapons
PDC pyridinium dichromate
Ph phenyl
Pin pinacolate
Piv pivaloyl
PMB p-methoxybenzyl
PMP 1-phenyl-3-methyl-5-pyrazolone
PPTS pyridinium p-toluenesulfonate
i-Pr isopropil
PyAOP (7-azabenzotriazol-1-yloxy)tripyrrolidinophosphonium hexafluorophosphate
Py pyridine
RCAM ring closing alkyne metathesis
RCM ring closing meyathesis
rt room temperature
SEM 2-(trimethylsilyl)ethoxymethyl
TBAF tetra-n-butylammonium fluoride
TBAI tetra-n-butylammonium iodide
TBDPS tert-butyldimethylphenylsilyl
TBHP tert-butylhydroperoxide
TBS tert-butyldimethylsilyl
TBTH tributyltin hydryde
TEA triethyl amine
TEMPO (2,2,6,6-tetramethyl-piperidin-1-yl)oxyl
TES triethylsilyl
THP tetrahydropyranyl
Tf trifluoromethylsulfonyl (triflyl)
TFA trifluoroacetic acid
Thex thexyl
THF tetrahydrofuran
TIPS triisopropylsilyl
TMS trimethylsilyl
TPAP tetra-N-propylammonium perruthenate
Tr trityl
Ts tosyl
p-TsOH p-toluenesulfonic acid

1 Introduction

It's not so funny when you're sunk

And there ain't nothing you can do

When your options are all dead ends

Electric President (Safe and Sound)

The book entitled Dead Ends and Detours by which this work is directly inspired appeared in 2004 [1]. Many things have changed since then in Organic Synthesis. The first, perhaps more appealing and somewhat weird, is that Total Synthesis is not a priority area of research anymore. In fact, while the book Beyond the Molecular Frontier, Challenges for Chemistry and Chemical Engineering [2] that defines the future of Chemistry in the twenty-first century does indeed dedicate an entire chapter to Synthesis and Manufacturing (Chapter 3, page 22) and explicitly includes as challenges those essential and inherent to Organic Synthesis, from developing new methods that will enable synthesis of all important molecules in reasonable yields using compact synthetic schemes, so that no useful compound is inaccessible to practical synthesis to intensify manufacturing processes… the feeling across the funding agencies and the different journals seems to be very different. To avoid biasing, the ISI web of knowledge listed under the searching string ‘total synth’ 145 entries in 2004 in two journals of reference for this field (namely Angewandte Chemie International Edition and the Journal of the American Chemical Society), and 183 entries in 2011. The increase in the number of entries (38)1 does not correlate with the increase in the number of pages (6827 vs 12659 and 17086 vs 21028, respectively) and articles (1224 vs 2002 and 3167 vs 3176, respectively).

On the other hand, two Nobel Prizes (2005 and 2010) have been awarded to fundamental breakthroughs in synthetic methodology. Coupled to the 2001 Nobel Prize a good share of the Nobel Prizes awarded in the twenty-first century recognized the importance of Organic Synthesis. What are the reasons for this (apparent) lack of interest by the chemical community in Total Synthesis? Is Total Synthesis out of fashion, or does the chemical community simply consider that it is better to develop synthetic methodology without applying it to formidable synthetic problems like the ones represented by complex natural products? Evidently, in these times of economic restrictions, developing new (what should be considered new is also an arguable question but out of the scope of this book) methodologies, stockpiling them in the arsenal of synthetic methods (with the idea in mind that should a problem arise they will be applicable to solve it, even if their real scope is unknown) is cheaper and more productive in terms of articles and patents than expending years of work and money in stressing the stockpiled reagents and protocols in the synthesis of densely functionalized molecules. Knowledge in Chemistry seems in these times of crisis secondary.

Paralleling the synthesis of a complex natural product or a complex synthetic molecule with a good planned battle, we should remember that in the words of German military strategist Helmuth von Moltke ‘No battle plan survives contact with the enemy.’ Translating this phrase to Total Synthesis it is clear that, according to the literature, still too many times when a good synthetic planning meets the real world of Total Synthesis (where common or sophisticated reagents and protocols tested in simple molecules go wild), the real world wins and the planning fails. Key steps based on insufficiently tested reagents, methods, or protocols do not work. Mistaken suppositions come back to bite you and the most brilliant plan may fail by the failure of a simple, seemingly innocuous, step. Therefore, Total Synthesis is the best way to know the real power and the actual and future knowledge of the chemical synthesis of the twenty-first century. Abandoning the testing of the synthetic methodology in Total Synthesis may lead to the risk of having useless, untested weapons to fight a real battle. Moreover, the development of synthetic methodology is intimately joined to Total Synthesis. When the preparation of a molecule is of high priority, the answers to real problems usually lead to better solutions than the answers to hypothetical situations.

The examples in this book show some of these situations. Well-established and, in some cases, venerable synthetic methods fail when applied to Total Synthesis. Moreover, a computational interpretation of the reason behind the failure will be presented when possible, to exemplify how expending a few hours of computational time may in some cases save months or years of hard experimental work. Computational chemistry is a potentially useful tool that can be added to the arsenal of Organic Synthesis. It is foreseeable that with the appearance of new user-friendly software and the increasingly strong interaction between theoreticians and experimentalists, the computational arsenal will be routinely used in Synthesis in the near future. This alliance will be adequate not only to explain a posteriori why any given process has produced a specific outcome, but also to predict if an advanced critical intermediate will behave properly in the next step of the total synthesis. If it is able to do this, computational chemistry will serve as a first-class scientific tool, following the statement of Prof. Kitaigorodskii: ‘A first-class theory predicts; a second-rate theory forbids; and a third-class theory explains after the event.’

All the computational analyses presented in this book have been carried out by means of well-established accurate enough theoretical tools [3]. Molecular Mechanics computations [4] were performed using the MM3 [5] and Optimized Potentials for Liquid Simulations (OPLS) 2005 [6] force fields as implemented in the MacroModel package [7]. Density Functional Theory (DFT)-based calculations [8] were performed by means of the hybrid three-parameter Becke, three-parameter, Lee-Yang-Parr exchange-correlation functional (B3LYP) functional [9] as implemented in the Gaussian 09 suite of programs [10]. Split valence partially polarized 6-31G basis set [11] and the Los Alamos National Laboratory 2-double-z (LANL2DZ) basis set and effective core potential (ECP) [12] were used for main group elements and transition metals, respectively. All the structures presented in the figures of this book were prepared by means of CYLview, [13] Maestro [14], and Spartan'10 [15] programs.

To begin this book, the failure of one of the most robust and powerful reaction available today, the olefin metathesis, will be discussed (Chapter 9 is dedicated to these reactions), together with a computational explanation of the reasons behind the failure.

Section 1.1 Synthesis of an intermediate in the total synthesis of brevetoxin A

Failure of a ring closing metathesis step.

The synthesis of organic molecules requires a careful study of the sequence in which the reactions leading to the target molecule are effected. This may be a self-evident statement for organic chemists. However, the placing of a substituent or a side chain at one step of the synthetic sequence instead of another may truncate the synthesis. This situation usually obliges one to revisit the synthetic tactics in the best of the cases. In the worst case, this is a success-compromising, not easily resoluble scenario, which leads to the total redesign of the synthetic sequence from the very beginning.

image

1.1 Intermediate in the Total Synthesis of Brevetoxin A [16]

image

1.1.1 Target Relevance

Brevetoxin A (1.1) was isolated from the blooms of the marine dinoflagellate Karenia brevis by Shimizu et al. [17] and Nakanishi [18], who independently established its structure. It is a potent neurotoxic agent whose activity is due to its capability of interacting with sodium ion channels, inhibiting their inactivation [19]. Brevetoxin A belongs to the so-called ladder toxins since it possesses a decacyclic skeleton built of 5-, 6-, 7-, 8-, and 9-membered oxacycles, fitting 22 tetrahedral stereocenters. Brevetoxin A was first synthesized by Nicolaou in 1998 [20] and recently by Crimmins [16, 21].

1.1.2 Synthetic Planning for Brevetoxin A (1.1)

The synthetic approach to brevetoxin A (1.1) is based on its disconnection into two tetracyclic halves 1.2 and 1.3, containing rings B–E and G–J, respectively (Scheme 1.1). These advanced synthetic intermediates will be accessed through the novel convergent [X + 2 + X] strategy [22] from the simple rings B and E (for 1.2) and G and J (for 1.3). Therefore, development of a common methodology for the preparation of rings B, E, G, and J will make the synthesis highly convergent, and it may be considered one of the key steps of the synthetic planning. This example will discuss the tactical changes that were needed to prepare aldehyde 1.4 (a key intermediate in the synthesis of 1.1), which resulted in a complete redesign of the synthetic route to this intermediate.

Scheme 1.1

image

1.1.3 Synthetic Planning of Aldehyde 1.4

The common strategy to prepare the building blocks that will lead to brevetoxin A requires the building of the subunits using a ring-closing metathesis reaction (RCM) [23]. This ring closure will be applied to an open-chain diene (1.5). The catalytic hydrogenation of the unsaturated ring obtained in this closure would yield the desired aldehyde 1.4 (Scheme 1.2). The key diene 1.5 would be available from triol 1.6 using conventional synthetic methodology.

Scheme 1.2

image

1.1.4 Predictable Problems

Both the RCM and the homogeneous hydrogenation reactions are well-established synthetic methodologies. A wide range of catalysts and reaction conditions are available. Therefore, there are no problematic steps foreseeable for this synthesis.

1.1.5 Synthesis of 1.4

The working hypothesis was that some level of conformational restriction imposed by the tetrahydrofuran (THF) moiety of 1.7 should expedite the RCM reaction [23]. First, the RCM cyclization was tested using the model compound 1.8, which lacks the methyl group in the C8 carbon present in the target aldehyde 1.4. Diene 1.8, available from triol 1.6 in 13 linear steps [21], was heated in dichloromethane (DCM) using a second-generation Grubbs catalyst (Grubbs second) (Scheme 1.3). Oxocene 1.9 was obtained in 70% yield. Then, diene 1.7 having the required methyl group at C8 was prepared also from 1.6 in 13 steps and submitted to the RCM reaction under conditions analogous to those used for the structurally close model diene 1.8. Neither these conditions nor the many different RCM conditions tested produced the desired cyclization product. The unreacted starting material 1.7 was obtained in all cases. The obvious solution, namely the introduction of the methyl group at C8 in oxocene 1.9, is not an easy task. Therefore, the lack of reactivity of diene 1.7 thwarted this approach.

Scheme 1.3

image

The reluctance of intermediate 1.7 to experience RCM required redesigning the tactical approach to the cyclic building blocks for brevetoxin A. In this approach the troublesome methyl group will be placed in the C8 after ring closure. Diene 1.11 bearing the required functional groups to introduce the methyl group after the RCM step was synthesized from the beginning (six linear steps from triol 1.6). Compound 1.11 has also the cyclic 1,3-dioxane system to favor the cyclization step.

Diene 1.11 was heated in the presence of Grubbs second leading to the expected oxocene 1.12 in an 83% yield. To effect hydrogenation of the double bond the benzyl group was replaced by a TBS-group. Unexpectedly, olefin 1.13 was recovered unaltered in all the conditions tested (Scheme 1.4). The inertia of 1.13 toward catalytic hydrogenation was attributed to the steric hindrance exerted by the groups attached at C3 and C4. Therefore, the Triisopropylsilyl (TIPS) and tert-Butyldimethylsilyl (TBS) silyl protecting groups were removed and substituted by a cyclohexylidene ketal. While this drawback did not truncate the synthesis it still requires the introduction of six additional synthetic steps. Hydrogenation of 1.14 occurred in the presence of Pearlman catalyst (Pd(OH)2/C). Unfortunately, in spite of the excellent yield of this reaction (90%) the hydrogenation occurred with total absence of selectivity. Furthermore, the ketal protecting group broke under the hydrogenation conditions and the final product was 1.15 (Scheme 1.5). Lack of selectivity in the hydrogenation of 1.14 prompted the authors to abandon this synthetic plan. However, it should be noted that the route was discarded once the drawback resulting from the premature introduction of the methyl-group was solved.

Scheme 1.4

image

Scheme 1.5

image

1.1.5.1 Revised Approach to Aldehyde 1.4

Since the placement of the double bond between the C5 and C6 carbons of the intermediate 1.4 resulted in an unviable synthetic route, using the same RCM, catalytic hydrogenation and introduction of the troublesome C8-methyl group sequence redesigned the building of this aldehyde. However, building a double bond between C6 and C7 now closes the cyclooctane ring to form an intermediate (1.16). The directing ability of the allylic alcohol at C8 should control the stereochemical outcome of the catalytic hydrogenation, ensuring the right stereochemistry at the C6-methyl group [13]. The apparently simple modification of the synthetic planning required a complete different approach to a new diene intermediate (1.17), which was accessed in 10 linear steps from Evans' oxazolidinone derivative 1.18 (Scheme 1.6).

Scheme 1.6

image

Reaction of diene 1.19 (obtained as a mixture of epimers at C8 in 10 steps from 1.18) with second generation Grubbs catalyst resulted in ring-closure to form the mixture of epimers 1.20/1.21. The minor alcohol was recycled following a standard ox-red sequence to form the desired alcohol 1.21. Finally, hydrogenation of compound 1.21 using the Crabtree catalyst ([Ir(cod)(PCy3)(pyr)]·PF6) at low temperature yielded oxocane 1.22 as a >19 : 1 mixture of epimers at C6. Introduction of the methylgroup at C8 and oxidation of the hydroxyl group complete the synthesis of aldehyde 1.4 (Scheme 1.7).

Scheme 1.7

image

1.1.6 Discussion

The early placement of a methyl group in 1.7, the required substrate for the RCM approach leading to the building block of brevetoxin A aldehyde 1.4, thwarted the ring closure and made it necessary to redesign a new intermediate (1.11) having the necessary functionality to introduce this methyl group after ring closure. The absence of selectivity during the hydrogenation of the double bond of 1.14 resulted in abandoning this synthetic approach.

Finally, the target molecule 1.4 was obtained through the Δ6 cyclooctene 1.21, using a newly designed synthetic approach, which used a hydroxyl group to improve selectivity in the critical catalytic hydrogenation, ensuring the right stereochemistry at C6 position (Scheme 1.8). In brief, four dienes (1.7, 1.8, 1.11, and 1.19) have been synthesized to establish the right sequence of events, namely metathesis, catalytic hydrogenation, and placing of a methyl group at C8. The final sequence leading to 1.4 established oxocene 1.21 as the actual intermediate in the synthesis of this aldehyde.

Scheme 1.8

image

The structures and energetically accessible conformations of dialkenes 1.7 and 1.8 were examined after Molecular Mechanics Monte Carlo simulations2 [24]. In the case of compound 1.7, a quite flexible conformational distribution was found (Figure 1.1). In particular, the two alkene moieties (highlighted by two hollow arrows in this figure) are quite close to each other, the distances between the two terminal =CH2 groups being about 5.4–6.8 Å. Moreover, these groups are in a favorable orientation for the head-to-head coordination required for the ring-closing metathesis reaction. In contrast, the methyl group present in compound 1.8 and the configuration of the C2 and C5 atoms of the tetrahydrofuran moiety result in a quite rigid structure in which the distances between the two =CH2 groups lie in the range 7.9–8.0 Å (Figure 1.9). Therefore, the geometries of the energetically accessible conformations for this latter compound are not adequate for coordination with the catalyst, and the ring-closing metathesis reaction will not proceed.

Figure 1.1 Fully optimized conformations of dialkenes 1.7 and 1.8. The most stable conformations are shown in ball and stick representations. The remaining three conformations closest in energy are displayed in stick representations. The carbon atoms required for the ring-closing metathesis reactions are highlighted by hollow arrows and ellipses. An asterisk highlights the methyl group that dictates the main conformational changes in 1.8.

image

A similar computational treatment can help to understand the reluctance of compound 1.13 to hydrogenation under a variety of conditions. The main energetically accessible conformations of alkene 1.13 and its reactive (albeit in a non stereoselective reaction) derivative 1.14 are gathered in Figure 1.2. Inspection of the structure of 1.13 (Figure 1.2) reveals that the two bulky TBS and TIPS protecting groups, located in opposite faces of the double bond, hinder both si and re faces of the cyclic alkene, thus making this molecule unreactive. In addition, the optimized geometries indicate that the acetal moiety does not participate significantly in this blockade. In contrast, compound 1.14 (Figure 1.2) is much less flexible than 1.13 but the steric hindrance in both faces is much lower and occurs to a similar extent. Therefore, it is not surprising that this compound is reactive but not stereoselective toward catalytic hydrogenation.

Figure 1.2 Fully optimized conformations of alkenes 1.13 and 1.14. The most stable conformations are shown in ball and stick representations. The remaining three conformations closest in energy are displayed in stick representations. The circles highlight the carbon atoms involved in the hydrogenation reactions.

image
Take-home messages
Macrocyclization reactions: If possible, make sure that the acyclic precursor has energetically favorable conformations that permit the proper interaction of the active sites with the catalyst or the activation reagent.
Addition reactions on cyclic alkenes: Try to avoid using bulky protecting groups on both sides of the cyclic unsaturated system. If stereocontrol is required, try to use protecting groups of different sizes.
Reagent Highlight: Crabtree's Catalyst

Crabtree's catalyst (1.24, Figure 1.3) is a highly efficient catalyst to hydrogenate sterically hindered, neutral (non-functionalized) double bonds. Crabtree [25] reported it in 1981 and it has been prolifically used (in its different variants) in the asymmetric hydrogenation of alkenes [26].

Figure 1.3 Crabtree's catalyst.

image