total synthesis of (±)-gomerone c (±)-fluorodanicalipin a and

Research Collection

Doctoral Thesis

Total Synthesis of ()-Gomerone C ()-Fluorodanicalipin A andStudies towards (+) and ()-Merochlorin A

Author(s): Huwyler, Nikolas Caspar Sebastian

Publication Date: 2015

Permanent Link: https://doi.org/10.3929/ethz-a-010610941

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For moreinformation please consult the Terms of use.

ETH Library

https://doi.org/10.3929/ethz-a-010610941http://rightsstatements.org/page/InC-NC/1.0/https://www.research-collection.ethz.chhttps://www.research-collection.ethz.ch/terms-of-use

Diss. ETH No. 22789

Total Synthesis of ()-Gomerone C

()-Fluorodanicalipin A and Studies towards

(+) and ()-Merochlorin A

A dissertation submitted to

Eidgenssische Technische Hochschule Zrich

(ETH Zurich)

for the degree of

Doctor of Sciences ETH Zurich

presented by

Nikolas Caspar Sebastian Huwyler

MSc ETH in Chemistry

Born 11th

December 1984

Citizen of Beinwil im Freiamt, AG

Accepted on the recommendation of

Prof. Dr. Erick M. Carreira, examiner

Prof. Dr. KarlHeinz Altmann, co-examiner

2015

i

Acknowledgements

I would like to thank Prof. Dr. ERICK M. CARREIRA for giving me the opportunity to conduct my

doctoral studies under his supervision. I highly appreciated the outstanding research environment he

provided, the exceptional scientific freedom I enjoyed as well as his trust in me with many additional

responsibilities such as group safety representative, head of the Synfacts team or as a substitute

lecturer. His passionate and critical mentoring along with numerous insightful discussions and group

seminars allowed me to steadily improve my theoretical and practical skills in organic chemistry and

I feel to have obtained an excellent education during my time in his group.

I am very thankful to Prof. Dr. KARLHEINZ ALTMANN for accepting the co-examination of this

thesis and for his highly valuable comments and corrections to the manuscript.

CHRISTIAN EBNER, STEFAN FISCHER, MATHIAS JACOBSEN, MATTHIAS WESTPHAL, and JOHANNES

BOSHKOW are most gratefully acknowledged for thoroughly and critically proof-reading individual

parts of this thesis. I highly appreciated their valuable comments, suggestions and corrections which

stimulated numerous fruitful discussions and substantially improved the original manuscript.

I am further indebted to Dr. CHRISTIAN GAMPE, Dr. MATTHIAS WEISS, Dr. MARTIN MCLAUGHLIN,

and Dr. DAVID SARLAH which in their role as more experienced group members helped, supported and

inspired me especially in the initial phase of my doctorate. In this regard, I would also like to sincerely

thank Dr. OLIVER JEKER and Dr. JULIAN EGGER who started a few months before me and immediately

became invaluable companions on this shared endeavor, inside and outside the laboratory.

Teaming up with STEFAN FISCHER in the Chlorosulfolipid Halologs Project has been a highly

enjoyable and memorable episode in my doctorate, and I would like to thank him very much for our

great and rewarding collaboration. I wish him all the best for the remainder of his graduate studies and

his future life. Dr. SUSANNE WOLFRUM and YESHUA SEMPERE are gratefully acknowledged for their

help in the toxicological assessment of fluorodanicalipin A and the synthesis of the amphotericin B/

polyamine derivative, respectively, and I wish all the best to MARCO BRANDSTTTER in continuing the

merochlorin A project.

I would like to thank all the past and present members of the Carreira Group for the very good

working atmosphere and the stimulating scientific environment. In this respect, special thanks go to

all my labmates in G338, Dr. CHRISTIAN GAMPE, Dr. SBASTIEN GOUDREAU, SAMY BOULOS, SIMON

BREITLER, CHRISTIAN EBNER, STEFAN FISCHER, ADRIEN JOLITON, LEONARDO NANNINI, and MARKUS

ROGGEN, for the fantastic working atmosphere, their helpful and optimistic attitude as well as the

ii

many great moments in and especially also outside the laboratory. Furthermore, I would like to thank

my writing-room compagnions SIMON KRAUTWALD and HANNES ZIPFEL for the good times and the

many interesting discussions.

During the course of my PhD I had the pleasure of supervising four highly motivated and talented

semester and master students and I would like to express my gratitude to MARLENE ROTHE, HELENE

WOLLEB, MINH DAO, and PHILIPP TANNER for their good work and contributions. I am sure that I have

learned as much from them as they have learned from me, and I wish them all the best for their future.

The work described in this thesis was greatly facilitated by the continuous help and support of

various members of the administrative and technical staff, as well as the excellent infrastructure at

ETH. Many thanks go to our groups secretaries FRANZISKA PEYER and ANKE KLEINT, the members

of the NMR service (Dr. MARCOLIVIER EBERT, REN ARNOLD, RAINER FRANKENSTEIN, and PHILIPP

ZUMBRUNNEN), the MS service team (LOUIS BERTSCHI, OSWALD GRETER, and ROLF HFLIGER), the

X-ray crystallographers (Dr. BERND SCHWEIZER, Dr. NILS TRAPP, and MICHAEL SOLAR), as well as the

whole HCI-Schalter and glassware cleaning team.

Special thanks go to BARBARA CZARNIECKI, NORA HILD, LEO BETSCHART, BASTIAN BRAND, SAM

FUX, JONAS HALTER, OLIVER JEKER, and LOTHAR OPILIK from the Freitagsmittagstisch, as well as to

SIMON JERMANN, MARTIN IGGLAND, and REMO SENN for their longstanding and close friendship with

all the many great moments, activities, travels and discussions we have had together over these years.

Mein ganz besonderer Dank gebhrt meiner Familie, meinen Eltern RUTH und BALZ RUST sowie

meiner Schwester ANNE-SOPHIE RUST, welche mich stets in allem uneingeschrnkt und bedingungslos

untersttzt haben. Ohne ihre Hilfe und Rckhalt wre ich sicherlich nie so weit gekommen im Leben

und htte dieses Vorhaben keinesfalls so umsetzten knnen.

iii

Publications

N. Huwyler, E. M. Carreira:

Total Synthesis and Stereochemical Revision of the Chlorinated

Sesquiterpene ()-Gomerone C

Angewandte Chemie International Edition 2012, 51, 1306613069.

S. Fischer, N. Huwyler, S. Wolfrum, E. M. Carreira:

Synthesis and Biological Evaluation of Bromo- and Fluorodanicalipin A

Angewandte Chemie International Edition 2016, 55, 25552558.

Poster Presentations




Scholarship Fund of the Swiss Chemical Industry (SSCI) Symposium, Zurich, November 2012.




Novartis Day, Zurich, Switzerland, June 2013.

N. Huwyler, S. Fischer, E. M. Carreira:

Total Synthesis of Bromo- and Fluorodanicalipin A

Novartis Day, Zurich, Switzerland, August 2015.

iv

Table of Contents

Acknowledgements i

Publications and Poster Presentations iiii

Table of Contents iiv

Abstract vviii

Zusammenfassung ixi

List of Abbreviations, Acronyms, and Symbols xiv

1 Total Synthesis of Gomerone C

1.1 Introduction 2

1.1.1 Halogenated Small Molecules as Agents of Defense 2

1.1.2 Isolation and Structural Elucidation of Gomerones A-C 2

1.1.3 Conclusions and Project Outline 4

1.2 Synthetic Planning 5

1.2.1 General Considerations 5

1.2.2 Retrosynthetic Analysis and Strategy 6

1.3 Results and Discussion 7

1.3.1 First Generation Approach 7

1.3.2 Second Generation Approach 15

1.4 Conclusions and Outlook 25

2 Total Synthesis of Fluorodanicalipin A

2.1 Introduction 28

2.1.1 Isolation and Occurrence of Chlorosulfolipids 28

2.1.2 Biosynthesis and Biological Relevance of Chlorosulfolipids 31

2.1.3 Total Syntheses of Chlorosulfolipids 35

2.1.4 Conclusions and Project Outline 37

v


2.2.1 General Considerations 38



2.3.1 Synthesis of the Olefin Fragment 40

2.3.2 Synthesis of the Nitrile Oxide Fragment 44

2.3.3 Nitrile Oxide Cycloaddition and Synthesis of 46

()-Fluorodanicalipin A

2.3.4 Structure Determination and 19

F-Modified JBCA 52

2.3.5 Preliminary Toxicological Assessment 67


3 Studies towards (+) and ()-Merochlorin A

3.1 Introduction 70

3.1.1 Actinomycetes in Drug Discovery 70

3.1.2 Isolation of Merochlorins A-D 70

3.1.3 Biological Activity of Merochlorins A-D 73

3.1.4 Total Syntheses of Merochlorins A and B 74

3.1.5 Biosynthetic Origin of Merochlorins A and B 81


3.2.1 General Considerations and Project Outline 86



3.3.1 Model System Studies 94

3.3.2 First Generation Precursors: Evaluation of Arene Substitution 107

3.3.3 Second Generation Precursors: Alternative Annulation Strategies 118

3.3.4 Third Generation Precursor and Approach of the End Game 123


vi

4 Experimental Part

4.1 General Methods and Materials 134

4.2 Total Synthesis of Gomerone C 137

4.2.1 Experimental Procedures of the First Generation Approach 137

4.2.2 Experimental Procedures of the Second Generation Approach 150

4.3 Total Synthesis of Fluorodanicalipin A 170

4.3.1 Experimental Procedures towards Fluorodanicalipin A 170

4.3.2 Experimental Details for the Toxicological Assessment 194

4.4 Studies towards (+) and ()-Merochlorin A 196

4.4.1 Experimental Procedures of the Model System Studies 196

4.4.2 Experimental Procedures for the First Generation Precursors 211

4.4.3 Experimental Procedures for the Second Generation Precursors 231

4.4.4 Experimental Procedures for the Third Generation Precursor 241

5 Appendix

5.1 Pattern Correlation Diagrams for the 19

F-Modified JBCA A2

5.2 X-Ray Crystallography A5

5.2.1 X-Ray Crystallographic Data for Compound 118 A5

5.2.2 X-Ray Crystallographic Data for Compound 132 A10

5.2.3 X-Ray Crystallographic Data for Gomerone C (102) A17

5.3 Supercritical Fluid Chromatography A23

5.3.1 SFC Data for Compound 470 A23

5.3.2 SFC Data for Compound 472 A25

5.4 NMR Spectroscopic Data Gomerone C A27

5.5 NMR Spectroscopic Data Fluorodanicalipin A A56

5.6 NMR Spectroscopic Data Merochlorin A A85

viii

Abstract

The gomerones constitute a small class of chlorinated sesquiterpenes isolated from the red algae

Laurencia majuscula that was found to utilize similar halogenated secondary metabolites as agents of

chemical defense against potential enemies and competitors. The angular tricyclic carbon skeleton of

these natural products comprises two contiguous quaternary centers and two tertiary chlorides, one of

which is positioned at the bridgehead of the bicyclo[3.2.1]octane core (Scheme I). The unprecedented

and congested gomerane scaffold rendered these compounds interesting for synthetic studies. The first

part of this doctoral thesis describes the development of a route towards ()-gomerone C (I) which not

only resulted in the first total synthesis of a member of this class, but also entailed a revision of the

relative stereochemistry of this metabolite, and hence a reassignment of gomerone B (3-epi-I).

Scheme I. Revised structure of gomerone C (I), reassigned structure of gomerone B (3-epi-I) and the gomerane skeleton.

The synthetic strategy of the successful approach relied on a DIELSALDER reaction for the early

construction of the two contiguous quaternary centers, as well as a late-stage CONIA-ene cyclization of

a chlorinated enol silane with an alkyne for the formation of the gomerane scaffold with the attendant

bridgehead chloride (Scheme II). Hydrochlorination of the exocyclic olefin afforded gomerone C (I).

Scheme II. Synthetic route towards gomerone C with the strategic DIELSALDER and CONIA-ene cyclizations.

This outcome, however, was rather surprising since it would have suggested that the exocyclic

double bond had undergone selective hydrochlorination from the sterically more hindered endo-face.

Clarification of this issue was achieved by X-ray diffractometry, which revealed that the chloride and

not the methyl group occupied the axial position in gomerone C (I). Consequently, this led to the

ix

stereochemical revision of gomerone C (I) and the reassignment of gomerone B (3-epi-I). The

targeted natural product could be obtained in 15 steps and 4% overall yield from commercially

available 1-acetyl-1-cyclopentene (III).

The second part of this thesis is devoted to the synthesis of ()-fluorodanicalipin A (VII), an

analog of the natural chlorosulfolipid danicalipin A in which all chlorines were replaced by fluorine

(Scheme III). The strategy, which was deduced from the collective insights of earlier chlorosulfolipid

syntheses and subsequently adapted to the requirements of organofluorine chemistry, relied on a

pivotal nitrile oxide cycloaddition that enabled the convergent union of the two elaborated fragments

IV and V. The targeted compound, whose relative configuration and solution conformation could be

unambiguously determined by means of an especially developed 19

F-Modified J-Based Configuration

Analysis, was thereby obtained in 10% yield over the longest linear sequence of 12 steps (18 steps

total). With ample quantities of the fluoro-analog in hand, a preliminary comparative study became

possible which uncovered that the high broadband toxicity of the natural chlorosulfolipids towards

Artemia salina is most likely due to the lipophilicity of the introduced chlorins.

Scheme III. Synthesis of fluorodanicalipin A (VII), the hexafluorinated analog of the chlorosulfolipid danicalipin A.

The third and last part of this thesis describes the systematic development of a domino-process for

the one-step synthesis of pentalenone systems en route to the novel antibiotic Merochlorine A (XIII).

Thereby, starting from a reported gold-catalyzed cyclopentenone synthesis, the method was further

advanced to mediate a second cyclopentannulation step with concomitant formation of a quaternary

center at the often difficult to functionalize central methine (Scheme IV). The acyclic precursors (i.e.

VIII) could be synthesized by standard palladium-catalyzed cross-coupling and hydrometallation

reactions which allowed for their modular assembly. The dihydropentalenones (i.e. IX or X) were

x

usually obtained in more than 20% overall yield from commercially available starting materials. If

enantiomerically enriched propargyl acetates were subjected to the reaction conditions the chirality at

the carbinol center was transferred to the two carbon centers at the pentalenone ring junction with only

minor erosion of enantiomeric excess. Likewise, if -disubstituted enals were used as substrates, the

intermediately formed secondary allyl alcohol was prone to undergo an acid induced dehydration

reaction. The obtained allylic carbocation was re-hydrated at the tertiary position in the presence of

water to yield X, or if dry solvents are used, eliminated a proton to afford diene IX.

Scheme IV. Gold-catalyzed one-step pentalenone synthesis from an acyclic precursor with concomitant formation of a

quaternary center and chirality transfer from the propargylic carbinol to the ring junction of the pentalenone.

With this powerful transformation in hand, the elaboration of the obtained pentalenone system

towards merochlorine A (IX) was addressed next. To this end, reductive dehydration of X with an

alkyl silane afforded enone XI which could be employed in a conjugated SAKURAI allylation (Scheme

V). The obtained tetrahydropentalenone XII thus contained three of the four contiguous stereocenters

of the targeted secondary metabolite, including the two quaternary ones.

Scheme V. Synthesis of tetrahydropentalenone XII en route to the antibiotic merochlorine A (XIII).

The major synthetic problems that remained to be addressed in order to achieve the first

asymmetric synthesis of merochlorine A (XIII) were -chlorination of the cyclopentanone in XII and

acylation to effect the final ring-closure. Efforts directed towards the completion of this synthesis are

currently ongoing in our laboratories.

xi

Zusammenfassung

Als Gomerone wird eine kleine Klasse von chlorierten Sesquiterpenen bezeichnet, die aus der

Rotalge Laurencia majuscula isoliert wurden, wobei letztere dafr bekannt ist, solche halogenierten

Sekundrmetabolite als chemische Abwehrstoffe einzusetzten. Das verzweigte, tricyclische

Kohlenstoffskelett dieser Naturstoffe enthlt zwei benachbarte quaternre Zentren, sowie zwei tertire

Chloride, wovon eines sich am Brckenkopf des Bicyclo[3.2.1]oktan-Kerns befindet (Schema I). Das

einmalige und dicht gepackte Gomeran-Gerst stellt ein interessantes Problem fr synthetische

Studien dar. Der erste Teil dieser Dissertation beschreibt die Entwicklung einer Route fr

()-Gomeron C (I), was nicht nur in der ersten Totalsynthese eines Molekls dieser Terpenklasse

mndete, sondern auch zu einer Revision der relativen Stereochemie dieser Metaboliten fhrte.

Schema I. Strukturrevision fr Gomeron C (I), neu zugewiesene Struktur fr Gomeron B (3-epi-I) und Gomeran-Gerst.

Die Synthesestrategie basierte auf einer frhen DIELSALDER Reaktion zur Konstruktion der zwei

benachbarten, quaternren Zentren, sowie einer CONIA-En Cyclisierung zwischen einem chlorierten

Silylenolether und einem Alkin zur Bildung des Gomeran-Gersts, sowie des Chlorids am Brcken-

kopf (Schema II). Hydrochlorierung des exocyclischen Olefins ergab schliesslich Gomeron C (I).

Schema II. Syntheseroute zu Gomeron C (I) mit der strategischen DIELSALDER und CONIA-En Cyclisierung.

Letzteres erschien allerdings eher fragwrdig, da dies impliziert htte, dass die exocyclische

Doppelbindung selektiv von der sterisch mehr gehinderten endo-Seite her hydrochloriert wurde. Ein

definitiver Beweis konnte durch Rntgendiffraktometrie erbracht werden, welche deutlich zeigte, dass

in Gomeron C (I) der Chloridsubstituent und nicht die Methylgruppe die axiale Position einnimmt.

xii

Dies fhrte konsequenterweise zu einer Revision der relativen Stereochemie von Gomeron C (I) und

einer neuen stereochemischen Zuweisung von Gomeron B (3-epi-I). Das angestrebte Zielmolekl

wurde somit in 15 linearen Schritten und 4% kombinierter Ausbeute erhalten.

Der zweite Teil beschftigt sich mit der Synthese von ()-Fluorodanicalipin A (VII), welches ein

hexafluoriertes Analogon des natrlich vorkommenden Chlorosulfolipids Danicalipin A darstellt

(Schema III). Die Strategie wurde aus den kollektiven Erkentnissen frherer Chlorosulfolipid-

synthesen abgeleitet und den Bedrfnissen der Organofluorchemie angepasst. Im Zentrum steht dabei

eine Nitriloxid Cycloaddition, welche die konvergente Verknpfung der zwei fortgeschrittenen

Fragmente IV und V ermglicht. Das Zielmolekl, dessen relative Stereochemie und Konformation in

Lsung mittels einer speziell dafr entwickelten 19

F-Modifizierten JBCA Methode ermittelt wurde,

konnte dabei in 10% Ausbeute ber die lngste lineare Sequenz von 12 Schritten erhalten werden. Mit

dem so erhaltenen Material konnte eine erste toxikologische Vergleichsstudie durchgefhrt werden.

Schema III. Synthese von Fluorodanicalipin A (VII), dem hexafluorierten Analogon von natrlichem Danicalipin A.

Der dritte und letzte Teil dieser Dissertation beschreibt die systematische Entwicklung eines

Domino-Prozesses fr die einstufige Synthese von Pentalenon-Systemen hin zum neuen Antibiotikum

Merochlorin A (XIII). Die Methode wurde ausgehend von einer literaturbekannten, Gold-kataly-

sierten Cyclopentanonsynthese entwickelt und ermglicht die zweite Annelierung eines Cyclopentans

unter gleichzeitiger Ausbildung eines quaternren Zentrums am nur schwer zugnglichen, verbrckten

Kohlenstoff dieser Systeme (Schema IV). Die acyclischen Vorlufermolekle (i.e. VIII) konnten

dabei mittels palladiumkatalysierten Kreuzkupplungen und Hydrometallierungen in modularer Form

aufgebaut werden. Die entprechenden Dihydropentalenone (i.e. IX oder X) wurden blicherweise in

mehr als 20% kombinierter Ausbeute von kommerziell erhltlichen Ausgangsmaterialien erhalten. Die

xiii

Verwendung von enantiomerenreinen Propargylacetaten in der Reaktion fhrte zu einem Transfer der

Chiralitt vom Carbinol zu den zwei Kohlenstoffatomen an der Verzweigung der Fnfringe unter nur

geringfgiger Verminderung des Enantiomerenberschusses. Weiter wurde festgestellt, dass beim

Gebrauch von -disubstituierten Enalen der intermedir gebildete sekundre Alkohol eine

sureinduzierte Dehydratisierung durchluft. Das allylische Carbokation, welches dabei gebildet wird,

untergeht entweder eine Rehydratisierung, falls Waser in der Reaktionslsung vorhanden ist, oder eine

Eliminierung zum Dien, wenn trockene Lsungsmittel verwendet werden.

Schema IV. Gold-katalysierte, einstufige Pentalenonsynthese aus acyclischen Vorlufermoleklen mit gleichzeitiger

Ausbildung eines quaternren Stereozentrums und Chiralitts-Transfers vom propargylischen Carbinol.

Mit dieser leistungsfhigen Transformation im Repertoir wurde die Synthese von Merochlorin A

(XIII) fortgesetzt. Die reduktive Dehydratisierung von X mittels eines Alkylsilans ermglichte die

Synthese von Enon XI, welches sogleich in einer konjugierten SAKURAI Allylierung Verwendung

fand (Schema V). Das so gewonnene Tetrahydropentalenon XII enthlt bereits drei von gesamthaft

vier benachbarten Stereozentren, inklusive der beiden quaternren.

Schema V. Eine konjugierte SAKURAI Allylierung ermglicht die Synthese von XII, welches bereits drei der insgesamt

vier Stereozentren, inklusive der beiden quaternren Zentren, von Merochlorin A (XIII) enthlt.

Die zwei Hauptprobleme, welche noch bestehen bevor die erste asymmetrische Totalsynthese von

Merochlorin A (XIII) verwirklicht werden kann, sind die -Chlorierung am Cyclopentanon und eine

Acylierung um den letzten Ring des Molekls zu schliessen. Anstrengungen in diese Richtung werden

zur Zeit in unseren Laboratorien unternommen.

xiv

List of Abbreviations, Acronyms, and Symbols

ngstrm

Ac acetyl

aq aqueous

Ar aryl, substituted aromatic ring

atm standard atmosphere, 105 Pascal

9-BBN 9-borabicyclo[3.3.1]nonyl

Bn benzyl

b.p. boiling point

br broad

brsm based on recovered starting material

Bu butyl

Bz benzoyl

C degree Celsius

18-c-6 1,4,7,10,13,16-hexahexaoxacyclooctadecane

CAM ceric ammonium molybdate

cat. catalytic

CoA coenzyme A

COSY correlation spectroscopy

Cy cyclohexyl

NMR chemical shift in ppm downfield of a specified standard

d day, doublet

DBU diazabicyclo[5.4.0]undec-7-ene

DCC dicyclohexylcarbodiimide

DCE 1,2-dichloroethane

DEAD diethyl azodicarboxylate

DEPT distortionless enhancement polarization transfer spectroscopy

DMAP 4-(dimethylamino)pyridine

DME 1,2-dimethoxyethane

DMF N,N-dimethyl formamide

DMP DESSMARTIN periodinane

dmp 3,5-dimethylpyrazole

DMSO dimethylsulfoxide

xv

dppf 1,1-bis(diphenylphosphino)ferrocene

DQF double quantum filtered

dr diastereomeric ratio

ee enantiomeric excess

EI electron impact ionization

ent inversion of all stereogenic centers

epi inversion of one stereogenic center

equiv equivalent

er enantiomeric ratio

ESI electron spray ionization

Et ethyl

et al. et alii, and others

exs excess

FT FOURIER transformation

g gram

h hour

HMBC heteronuclear multiple bond correlation spectroscopy

HMDS 1,1,1,3,3,3-hexamethyldisilazane

HMPA hexamethylphosphoramide

HOE heteronuclear OVERHAUSER effect

HOESY heteronuclear OVERHAUSER effect spectroscopy

HPLC high pressure liquid chromatography

HRMS high resolution mass spectrometry

HSQC heteronuclear single quantum correlation spectroscopy

Hz Hertz

i iso

IBX 2-iodoxybenzoic acid

imid imidazole

IR infrared

J coupling constant

JBCA J-based configuration analysis

JohnPhos (2-biphenyl)di-tert-butylphosphine

k kilo

L liter

xvi

LDA lithium diisopropylamide

m multiplet, milli

m meta

M molar, mega

[(M)+] molecule ion

mCPBA meta-chloroperbenzoic acid

Me methyl

MED 2-methyl-2-ethyl-1,3-dioxolane

min minute

mp melting point

micro

mmol millimole

mol mole

Ms methanesulfonyl

MS molecular sieves, mass spectrometry

n normal (prefix for alkyl groups)

n.a. not applicable

NBS N-bromosuccinimide

NCS N-chlorosuccinimide

nd not determined

Nf nonafluorobutanesulfonyl

NMR nuclear magnetic resonance

NOE nuclear OVERHAUSER effect

NOESY nuclear OVERHAUSER effect spectroscopy

nr no reaction

o ortho

p para

pH negative logarithm of hydrogen ion concentration

Piv pivaloyl

ppm parts per million

PPTS pyridinium para-toluenesulfonic acid

Pr propyl

Py pyridine

q quartet

xvii

Rf retention factor

RT ambient temperature

s singlet, second

s sec

SFC supercritical fluid chromatography

t triplet

t tert

T temperature, Tesla

TASF tris(dimethylamino)sulfonium difluorotrimethylsilicate

TBS tert-butyldimethylsilyl

TES triethylsilyl

Tf trifluoromethanesulfonyl

TFA trifluoroacetic acid

THF tetrahydrofuran

TLC thin layer chromatography

TIPS triisopropylsilyl

TMEDA N,N,N,N-tetramethylene-1,2-diamine

TMS trimethylsilyl

Ts 4-methylsulfonyl

UATR universal attenuated total reflectance technique (IR)

UV ultraviolet

XPhos 2-dicyclohexylphosphino-2,4,6-triisopropylbiphenyl

1 Total Synthesis of

Gomerone C



1.1 Introduction

1.1.1 Halogenated Small Molecules as Agents of Defense

In the Darwinian struggle for survival, the immobile and therefore especially exposed

plants often resort to chemical means to secure their positions in the ecosystem against their

potential enemies and other competing organisms. Consequently, plants have been an

extremely rich source of interesting secondary metabolites with a wide variety of biological

activities. Red algae from the genus Laurencia (Ceramiales, Rhodomelaceae), for example,

are the source of over 750 halogenated terpenes, cyclic ethers and acetogenides which in

several cases were suggested to serve as chemical defence or antifouling agents due to their

activity against invading marine bacteria, herbivores or parasites.1 Interestingly, examinations

of the coral tissues revealed that these often chamigrene derived halogenated sesquiterpenes

are stored in special intracellular compartments, termed corps en cerise (cherry bodies) due to

their characteristic shape,2 which upon injury as well as potentially by vesicle transport release

these agents of defence into their closer environment.3

1.1.2 Isolation and Structural Elucidation of Gomerones A-C

In 2008, the group of M. CUETO reported the isolation of three novel chlorinated

sesquiterpenes, gomerones A-C (101103), from samples of Laurencia majuscula collected at

the southern coast of La Gomera, Canary Islands (Figure 1.1).4 The collected algal samples

were extracted with acetone and the combined extracts subsequently fractionated by vacuum

flash chromatography. Two promising fractions were further subjected to filtration, column

chromatography and HPLC purification to afford three novel compounds as colourless oils.

1 a) J.-M. Kornprobst in Encyclopedia of Marine Natural Products, 2nd Edition, pp. 342363, Wiley-VCH, Weinheim,

2014; b) H. Choi, A. R. Pereire, W. H. Gerwick in Handbook of Marine Natural Products (Eds. E. Fattorusso, W. H.

Gerwick, O. Taglialatela-Scafati), pp. 55152, Springer Science & Business Media B. V., Berlin, 2012. 2 L. T. Salgado, N. B. Viana, L. R. Andrade, R. N. Leal, B. A. P. da Gama, M. Attias, R. C. Pereira, G. M. Amado Filho, J.

Struct. Biol. 2008, 162, 345355. 3 C. S. Vairappan, S. P. Anangdan, K. L. Tan, S. Matsunga, J. Appl. Phycol. 2010, 22, 305311. 4 A. R. Daz-Marrero, I. Brito, J. M. de la Rosa, J. Darias, M. Cueto, Tetrahedron 2008, 64, 1082110824.

Introduction 3

Examination of the first compound, gomerone A (101), revealed two EIMS signals for the

molecular ion at 284 and 286 m/z, respectively, with the relative intensities being indicative

for the presence of a single chlorine substituent. The molecular formula was determined on the

basis of HRMS to be C15H21ClO3 (HRMS-EI: exact mass calculated for C15H21(35

Cl)O3+

[(M)+] 284.1179 m/z; found 284.1182 m/z) which would require the presence of five

equivalents of unsaturation in the molecule. IR spectroscopic measurements showed

Figure 1.1. A) Originally assigned structures of gomerones A-C possessing the unprecedented gomerane skeleton. B) The

producing organism Laurencia majuscula.

characteristic absorptions at 3433 cm1

and 1658 cm1

which were suggestive of hydroxyl

groups and a conjugated carbonyl group. The assignment of the chemical constitution was

achieved by comprehensive NMR studies involving 1H NMR,

13C NMR, DEPT-90/135,

1H,

1H-COSY and HMBC experiments which implicated three methyls, four methylenes, two

methines, of which one appeared to be olefinic and the other one a carbinol, and six fully

substituted carbon atoms with two of them being quaternary. Furthermore, the investigations

uncovered the presence of a single carbonyl group and one alkene double-bond which allowed

to ascribe the remaining three equivalents of unsaturation to a tricyclic scaffold. The 13

C NMR

chemical shift of 51.4 ppm for one of the two quaternary carbons indicated a spiro-carbon

characteristic of a chamigrene type substructure. HMBC correlations enabled for the assembly

of the previously identified fragments, suggesting that gomerone A (101) possessed a

bicyclo[3.2.1]octane core with a fused cyclohexenone at C6 and C7, a secondary and a tertiary


carbinol at C14 and C3, respectively, as well as a chloride at the C2 bridgehead of the bicyclic

system. The relative stereochemical configuration could be assigned on the basis of a NOESY

correlation between the proton at the C14 carbinol and the methyl protons at C15. The

absolute stereochemistry, on the other hand, remained unassigned because it was not possible

to prepare the required MOSHERs ester derivative of 101 or obtain single crystals suitable for

X-ray diffractometry.

The EIMS analysis of gomerone B (102) and gomerone C (103) provided identical

molecular ion signals at 300, 302 and 304 m/z, respectively, with the relative intensities being

typical for the presence of two chlorides. Closer inspection by means of HRMS revealed that

the two molecules are isomers of each other with a common empirical formula of C15H18O2Cl2

and consequently six equivalents of unsaturation in both cases. The IR spectra showed no

signs of hydroxyl groups anymore but, on the other hand, indicated the presence of two

,-unsaturated carbonyls due to two characteristic absorptions at 1682 cm1

and 1742 cm1

.

The chemical constitution was again established by comprehensive NMR studies (1H NMR,

13C NMR, DEPT-90/135,

1H,

1H-COSY and HMBC) and comparison with gomerone A (101)

which implicated that the carbinol formerly at C14 would be oxidized to a carbonyl group

while the one at C3 would be replaced by two epimeric chlorides. Attempts to elucidate the

relative stereochemistry at C3 by an NOESY experiment were inconclusive according to the

isolation group, however, they argued that the downfield shift of the C15 methyl by

0.20 ppm in gomerone B (103) would be due to the deshielding effect of the carbonyl

group and thus assigned this group to reside in the equatorial position.

1.1.3 Conclusions and Project Outline

The unprecedented and structurally challenging gomerane skeleton in combination with

the unexplored biological activity as well as the halogenated nature of these natural products

rendered them interesting targets for total synthesis. Consequently, it was decided to embark

on a synthetic campaign towards the dichlorinated congeners, gomerones B (102) and C (103)

with the primary intention to find a synthetic strategy that would elegantly resolve the problem

of installing the two vicinal tertiary chlorides as well as allowing for the rapid construction of

the unprecedented tricyclic angular scaffold, including the two adjacent quaternary centers.

Synthetic Planning 5

1.2 Synthetic Planning

1.2.1 General Considerations

Nominal gomerone B (3-epi-103) was chosen as the main target for the synthetic studies

while gomerone A (101) and C (3-epi-102) were considered optional targets that could

potentially be addressed from an advanced intermediate. The fact that none of the targeted

secondary metabolites had been obtained by total synthesis in combination with their

unprecedented carbon scaffold allowed for an unbiased approach to the problem.

Amongst the different techniques utilized by organic chemists to deduce and design

synthetic plans towards architecturally complex molecules, the retrosynthetic disconnection

approach formalized by E. J. COREY and coworkers5 is arguably the most prominent and

routinely applied.6 According to this formalism, target structures are iteratively disassembled

into simpler precursors until readily available starting materials result. The quintessential aim

of the analysis thereby is the identification of strategic disconnections that entail the most

efficient reduction in molecular complexity. In practice, this concept is most often employed

in combination with higher level strategies such as transformation based considerations,

stereochemical implications, potential starting materials or topological pattern recognition.

The synthetic challenges encountered in the architecturally congested gomerones A-C

(101103) are represented by an angular, tricyclic carbon skeleton, two contiguous quaternary

centres at C6 and C11, as well as two tertiary chlorines, one of which is positioned at the

bridgehead of a bicyclo[3.2.1]octane. Thus, according to BREDTs rule,7 the latter chlorine

atom would be precluded from being introduced by enolate chemistry once both rings have

been formed. Consequently, it appeared reasonable to prioritize the synthetic strategy on the

diastereoselective construction of the vicinal dichloride motif which should be addressed late

in the route in order to enable a potentially divergent approach.

5 a) E. J. Corey, W. J. Howe, H. W. Orf, D. A. Pensak, G. Petersson, J. Am. Chem. Soc. 1975, 97, 61166124; b) E. J.

Corey, X.-M. Chen in The Logic of Chemical Synthesis, John Wiley & Sons, New York, 1989. 6 For examples, see: a) K. C. Nicolaou, E. J. Sorensen in Classics in Total Synthesis: Targets, Strategies Methods VHC,

Weinheim, 1996; b) K. C. Nicolaou, S. A. Snyder in Classics in Total Synthesis II: More Targets, Strategies Methods

Wiley-VHC, Weinheim, 2003; c) K. C. Nicolaou, J. S. Chen in Classics in Total Synthesis III: Further Targets,

Strategies Methods Wiley-VHC, Weinheim, 2011; d) T. Hudlicky, J. W. Reed in The Way of Synthesis: Evolution of

Design and Methods for Natural Product Synthesis, Wiley-VHC, Weinheim, 2007. 7 a) J. Bredt, Liebigs Ann. Chem. 1924, 437, 113; b) For a recent review and further literature, see the following and

references therein: J. Y. W. Mak, R. H. Pouwer, C. M. Williams, Angew. Chem. Int. Ed. 2014, 53, 1366413688.


In the next chapter, the synthetic strategy towards nominal gomerone B (3-epi-103) will

be derived by means of the retrosynthetic disconnection analysis until the complexity of the

target structure is simplified enough for a pattern/methods recognition approach.

1.2.2 Retrosynthetic Analysis and Strategy

With the considerations and limitations from above in mind, the retrosynthetic analysis

was commenced by cleavage of the tertiary chloride at the C3 carbon. Besides the possibility

of introducing such a substituent by means of a nucleophilic displacement of an activated

alcohol or epoxide (i.e. an APPEL-type reaction), a MARKOVNIKOV-type addition of hydrogen

chloride to the corresponding exocyclic olefin was also taken into consideration (Scheme 1.1).

Scheme 1.1. Retrosynthetic analysis for nominal gomerone B (3-epi-103).

The retron of this latter option appeared to be particularly attractive since the ,-unsaturated

carbonyl 105 that one would arrive at could itself be derived via an intramolecular CONIA-ene

reaction8 between a chlorinated enol derivative and a terminal alkyne in the side-chain. Such a

transformation, if successful, would thereby concomitantly construct the bicyclo[3.2.1]octane

scaffold and install the tertiary chloride at the bridgehead position in a single transformation.

Precursor 106, in turn, could be deduced from endione 107 by a regioselective -chlorination,

which would further reduce the problem to the synthesis of a tetrahydroindendione. The

installation of vicinal quaternary centers,9 on the other hand, can only be achieved by a limited

set of chemical transformations and it was thus at that point decided to change to a method

identification approach, the results of which will be discussed in due course (Section 1.3).

8 a) J. M. Conia, P. Le Perchec, Synthesis, 1975, 119; b) J. Drouin, M. A. Boaventura, J. M. Conia, J. Am. Chem. Soc.

1985, 107, 17261729. 9 a) S. F. Martin, Tetrahedron 1980, 36, 419460; b) E. A. Peterson, L. E. Overman, Proc. Natl. Acad. Sci., USA 2004,

101, 1194311948; c) K. W. Quasdorf, L. E. Overman, Nature 2014, 516, 181191.

Results and Discussion 7

1.3 Results and Discussion

1.3.1 First Generation Approach

The initial synthetic plan towards tetrahydroindene dione 107 was inspired by chemistry

that had originally been reported by M. E. JUNG and J. P. HUDSPETH in the late 1970s and that

was later employed by L. A. PAQUETTE and coworkers in the total syntheses of several

terpenoid natural products.10

Thereby, dimethoxy norbornenone 110 was usually treated with

an appropriate vinylic organometal species which entailed an endo-selective addition to the

carbonyl group due to the obstruction of the exo-face by the dimethoxy ketal. The 3-hydroxyl-

1,5-hexadienes thus obtained were prone to undergo charge accelerated oxy-COPE

rearrangements11,12

upon treatment with a base affording the corresponding dimethoxy tetra-

hydoindenones which could be further elaborated into the targeted structures (Scheme 1.2).

Scheme 1.2. First generation synthetic planning towards tetrahydroindenone 108.

The envisioned synthesis of tetrahydroindeneone 108 by means of this key transformation

would require that the 3-hydroxyl-1,5-hexadiene precursor 109 contained a homopropargyl

appendage at C6.13

Although no examples of substitution at either olefin carbon could be

found for 14-dimethoxy norbornenone type starting materials, there was however precedence

for the corresponding 14-dimethyl norbornene derivatives and it thus appeared reasonable to

assume that an alkyl group at this position would not preclude the projected [3,3]-sigmatropic

10 a) M. E. Jung, J. P. Hudspeth, J. Am. Chem. Soc. 1978, 100, 43094311; b) M. E. Jung, J. P. Hudspeth, J. Am. Chem.

Soc. 1980, 102, 24632464; c) M. E. Jung, L. A. Light, J. Am. Chem. Soc. 1984, 106, 76147618; d) L. A. Paquette, K. S.

Learn, J. L. Romine, H.-S. Lin, J. Am. Chem. Soc. 1988, 110, 879890; e) L. A. Paquette, J. L. Romine, H.-S. Lin, J.

Wright, J. Am. Chem. Soc. 1990, 112, 92849292; f) L. A. Paquette, Z. Gao, Z. Ni, G. F. Smith, J. Am. Chem. Soc. 1998,

120, 25432552; g) L. A. Paquette, R. C. Thompson, J. Org. Chem. 1993, 58, 49524962. 11 a) J. A. Berson, M. Jones, J. Am. Chem. Soc. 1964, 86, 50195020; b) J. A. Berson, M. Jones, J. Am. Chem. Soc. 1964,

86, 50175018; c) D. A. Evans, A. M. Golob, J. Am. Chem. Soc. 1975, 97, 47654766. 12 For selected reviews, see: a) L. A. Paquette, Tetrahedron 1997, 53, 1397114020; b) S. R. Wilson, Org. React. 1993,

43, 93250; c) R. K. Hill in Comprehensive Organic Synthesis, Vol. 5 (Eds. B. M. Trost, I. Fleming), Pergamon Press,

Oxford, 1991; d) L. A. Paquette, Angew. Chem. Int. Ed. 1990, 29, 609626. 13 The numbering corresponds to the gomerane scaffold and is used throughout this part of the thesis.


rearrangement.14

The -disubstituted nature of the allylic alcohol in 109, on the other hand,

seemed more likely to be problematic and, to the best knowledge of the author of this thesis,

only one report included such a substitution pattern, albeit with sterically less demanding

sp2-hybridized groups instead of the planned methyl groups.

15 Nontheless, the perspective to

forge the two contiguous quaternary centers in a transformation that would simultaneously

provide a tetrahydroindenone system with two differentiated carbonyl functionalities at the

correct positions in a single step seemed attractive. Consequently, it was decided to devote the

initial efforts towards nominal gomerone B (3-epi-103) on the evaluation of this oxy-COPE

rearrangement. The required precursor 109, in turn, should be obtained by a regioselective

functionalization of the norbornene double bond in order to circumvent the tedious

construction of this system.

The laboratory work was commenced with the synthesis of endo-norbornenol 114 for

which PAQUETTE and coworkers had detailed an experimental protocol that was derived from

an earlier investigations of JUNG and HUDSPETH.10d

Accordingly, commercially available

cyclopentadiene 111 was dissolved in neat vinyl acetate and heated at reflux temperature for

5 d during which the desired DIELSALDER cycloaddition ensued in 84% (Scheme 1.3).

Scheme 1.3. Synthetic route towards endo-alcohol 114 according to M. E. JUNG and J. P. HUDSPETH. Reagents and

conditions: a) Vinyl acetate (solvent, 21 equiv), reflux, 5 d, 84%; b) K2CO3 (10 mol%), MeOH, RT, 20 min; c) Na

(10 equiv), EtOH (3 equiv), NH3/Et2O (3:1), 78 C, 20 min, 68% over 2 steps.

The obtained tetrachloronorbornyl acetate 112 could be crystallized from MeOH and was

further subjected to acetate methanolysis (10 mol% K2CO3 in MeOH) before reductive

cleavage of the four chlorine atoms could be effected under BIRCH-conditions, furnishing the

required endo-norbornenol 114 in 57% yield over 3 steps. With this starting material in hand,

14 a) L. A. Paquette, F.-T. Hong, J. Org. Chem. 2003, 68, 69056918; b) Y. Hu, R. L. Bishop, A. Luxenburger, S. Dong, L.

A. Paquette, Org. Lett. 2006, 8, 27352737; c) L. A. Paquette, Y. Hu, A. Luxenburger, R. L. Bishop, J. Org. Chem. 2007,

72, 209222; d) For a C14 unsubstituted example, see: G. Brub, A. G. Fallis, Tetrahedron Lett. 1989, 31, 40454048. 15 L. A. Paquette, D. T. DeRussy, Tetrahedron 1988, 44, 31393148.


the regioselective functionalization of the norbornene double bond could be examined next. In

this respect, a literature screening for viable methodology revealed, amongst others, an

interesting haloselenylation reaction16

that was pioneered by P. VOGEL and coworkers in the

early 1980s.17

The required endo-acetoxy norbornene 115 and norbornenone 110 could both

be accessed in a single step and high yield by means of a DMAP accelerated acylation18

and a

JONES oxidation,19

respectively (Scheme 1.4). The latter procedure generally proved to be a

good choice for the oxidation of sterically more encumbered alcohols since the formation of

the activated species, i.e. the chromate esters, are not rate-determining but rather very fast

pre-equilibria. However, if once formed, the chromate esters release greater tension in

congested substrates which thus often leads to the somewhat counterintuitive observation that

CrO3-mediated oxidations proceed faster with increasing steric hindrance.20

In a first experiment, treatment of norbornenone 110 with one equivalent of PhSeBr in

CHCl3 at RT entailed a relatively quick reaction and after 4 h the starting material was found

to be completely consumed. Closer inspection of the crude reaction mixture after work-up by

TLC and 1H NMR spectroscopy indicated the formation of a single product which was

purified by crystallization from a mixture of Et2O and hexane. Structural elucidation of this

compound could be achieved on the basis of comprehensive 1H,

13C, NOE and 2D NMR

spectroscopy (HSQC, HMBC and 1H,

1H-COSY). Importantly, the assignment of relative

stereochemical configuration was greatly facilitated by the distinct chemical shifts of the

protons geminal to the newly introduced groups as well as the fact that the bridgehead protons

of norbornene-type systems only display scalar couplings to protons in an exo-position. For

the compound in question, it was found that H1 resonanced as a doublet of doublet of doublet

at 4.67 ppm with scalar couplings to H2 (3J = 4.8 Hz), H6 (

3J = 3.8 Hz) and H7 (

4J = 0.9 Hz)

whereas H6 appeared as an isolated doublet at 3.45 ppm with a single scalar coupling to

16 a) H. J. Reich, J. Org. Chem. 1974, 39, 428429; b) K. B. Sharpless, R. F. Lauer, J. Org. Chem. 1974, 39, 429430; c)

G. Schmid, D. Garratt, Tetrahedron 1978, 34, 28692872; d) D. Liotta, G. Zima, Tetrahedron Lett. 1978, 19, 49774980. 17 a) P.-A. Carrupt, P. Vogel, Tetrahedron Lett. 1982, 23, 25632566; b) K. A. Black, P. Vogel, J. Org. Chem. 1986, 51,

53415348; c) K. Lal, R. G. Salomon, J. Org. Chem. 1989, 54, 26282632; d) P.-A. Carrupt, P. Vogel, Helv. Chim. Acta.

1989, 72, 10081028; e) O. Arjona, R. F. de la Pradilla, J. Plumet, A. Viso. J. Org. Chem. 1991, 56, 62276229; f) O.

Arjona, J. Plumet, 1992, 57, 772774. 18 a) L. M. Litvinenko, A. I. Kirichenko, Dokl. Akad. Nauk SSSR, Ser. Khim. 1967, 176, 97; b) W. Steglich, G. Hfle,

Angew. Chem. Int. Ed. 1969, 8, 981981; c) For a review see: G. Hfle, W. Steglich, H. Vorbrggen, Angew. Chem., Int.

Ed. Engl. 1978, 17, 569583. 19 a) K. Bowden, I. M. Heilbron, E. R. H. Jones, B. C. L. Weedon, J. Chem. Soc. 1946, 3945; b) G. Tojo, M. Fernndez in

"Oxidation of Alcohols to Aldehydes and Ketones", pp. 195, Springer, Berlin, 2006. 20 J. Schreiber, A. Eschenmoser, Helv. Chim. Acta 1955, 38, 15291536.


H1 (3J = 3.8 Hz). In accordance with the above, this was in full agreement with an endo-

bromide residing at C1 and a phenylselenyl ether in the exo-position at C6. Further

confirmation for the assignment was derived from HMBC correlations, i.e. between H6/C(Ph),

H6/C8 and H1/C9, and NOE spectroscopy which allowed to unambiguously assign structure

116 to this product.

Scheme 1.4. Mechanistic rational for the different reaction outcomes in the bromoselenylation of the norbornene double-

bond of ketone 110 and endo-acetoxy 115. Reagents and conditions: a) CrO3 (2.0 equiv), H2O/H2SO4 (4:1), acetone, 0 C,

10 min, 86%; b) Ac2O (1.25 equiv), pyridine (2.5 equiv), DMAP (20 mol%), CH2Cl2, RT, 20 h, 96%; c) H2O2 (10 equiv),

H2O/THF, RT, 20 h, 84%.

This, however, was a rather curious result, since it suggested that PhSeBr had added to the

exo-face of the olefin double bond, irrespective of the steric hindrance imposed by the

dimethoxy ketal at C14, under formation of a seleniranium cation that was subsequently

opened in the usual, stereospecific anti-fashion. A tentative explanation for this intriguing

result could eventually be deduced from the reversibility of such additions.16

Consequently, it

was hypothesized that the kinetically preferred addition of PhSeBr indeed proceeded from the


expected endo-face,21

however, due to the presence of the 14-dimethoxy ketal, the formed

seleniranium was precluded from undergoing the mandated back-side attack by the rather

bulky bromide anion (rion= 1.96 ).22

Instead, the only remaining option for the nucleophile is

re-addition to the selenium atom which leads to the reconstitution of the norbornene double

bond. The small fraction of PhSeBr that adds to the exo-face, on the other hand, engaged in

the desired seleniranium ring-opening at carbon which furnishes the desired addition product.

In order to account for the observed regioselectivity, P. VOGEL and coworkers originally

invoked a hyperconjugative interaction between the carbonyl group and the p-orbital of the

norbornene double bond (see TS-I in Scheme 1.4).17a,d

Thereby, it was suggested that the

lone-pairs at the carbonyl oxygen would donate electron density into the binding orbital of the

C9C2 -bond, which itself is aligned in an almost parallel fashion with the C1 p-orbital that

engages in the seleniranium formation. As a consequence of the electron-donating

n(CO)-(C9,C2)-p(C1) homoconjugation the C1Se bond is destabilized and thus undergoes

substitution more readily than the C6Se bond. Alternatively, an explanation on the basis of

steric factors, i.e. by the avoidance of 1,3-diaxial interactions between the incoming bromide

and the endo-hydrogen at C8, has also been offered17c

and could in principle be combined with

the stereoelectronic reasoning from above. Nevertheless, it is noteworthy that the steric bias

can only influence nucleophiles that attack in an endo-fashion which in turn would necessitate

that the seleniranium ion is formed on the exo-face. At the outset of this project and with no

examples of C14 substitution in the literature, it was originally presumed that the seleniranium

ion would form selectively on the endo-face which potentially had allowed to distinguish the

steric from the stereoelectronic influence. As this was found not to be the case (vide supra), no

clarification of this question could be achieved. Contrariwise, however, it hence appeared very

likely that endo-acetoxy norbornene 115 would also undergo an equally regioselective

addition. To this end, subjection of 115 to PhSeBr in CHCl3 entailed an extremely slow

reaction which required 5 days to reach completion but nonetheless afforded the desired

addition product in 72% yield. As expected, with the endo-acetoxy substituent at C9 the

regioselectivity completely switched (see TS-II in Scheme 1.4) towards bromination at C6,

providing 117 as a single regio- and diastereomer. Treatment of this substrate with H2O2 in a

21 H. C. Brown, J. H. Kawakami, S. Ikegami, J. Am. Chem. Soc. 1970, 92, 69146917. 22 R. D. Shannon, Acta Cryst. 1976, 32,751767.


mixture of THF and H2O induced the anticipated selenoxide elimination23

and furnished vinyl

bromide 118 in 84% yield as faintly yellowish crystals that were feasible for X-ray

diffractometry (see Appendix 5.2.1). An attempt to directly use the unprotected alcohol 114,

on the other hand, led to the rapid formation of a complex mixture.

Having thus successfully introduced a bromo substituent at C6 of the norbornene double

bond, the installation of the homopropargyl appendage was envisioned to be effected by a

palladium-catalyzed B-alkyl SUZUKIMIYAURA cross-coupling.24,25

Amongst the available

protocols, the conditions of C. R. JOHNSON are arguably the most widely used in the context

of natural product synthesis.26

Accordingly, the required borate was generated by alkylation of

9-MeO-9-BBN27

with the organolithium species derived from iodide 119 by lithium-halogen

exchange with tBuLi28

and subsequently subjected to a DMF solution containing vinyl

bromide 118, 5 mol% PdCl2(dppf), 15 mol% AsPh3, Cs2CO3 and 12 equiv H2O (Scheme 1.5).

Scheme 1.5. Reagents and conditions: a) 119 (1.5 equiv), (9-MeO)-9-BBN (4 equiv), tBuLi (3.5 equiv), Et2O, 78 C,

15 min; then THF, 78 C to RT, 3 h; then 118 (1.0 equiv), PdCl2(dppf) (5 mol%), AsPh3 (15 mol%), CsCO3 (4 equiv),

H2O (12 equiv), DMF, 55 C, 16 h, 87%; b) LiAlH4 (1.3 equiv), Et2O, 0 C, 45 min, 88%; c) CrO3 (3 equiv), H2SO4/H2O

(1:4), acetone, 0 C, 5 min, 90%; d) 2-methyl-1-propenylmagnesium bromide (2.2 equiv), THF, RT, 2 h, 81%.

Heating of this mixture at 55 C for 16 h afforded the desired coupling product in 87% yield.

Removal of the acetoxy group was effected by the action of LiAlH4 in Et2O to give endo-

alcohol 121 in 88% yield. From intermediate 121 the required 3-hydroxyl-1,5-hexadiene 109

could be accessed by a JONES oxidation followed by an endo-selective GRIGNARD addition.

23 a) K. B. Sharpless, R. F. Lauer, J. Am. Chem. Soc. 1973, 95, 26972699; b) H. J. Reich, I. L. Reich, J. M. Renga, J. Am.

Chem. Soc. 1973, 95, 58135815; c) H. J. Reich, I. L. Reich, J. M. Renga, J. Am. Chem. Soc. 1975, 97, 54345447; d) H.

J. Reich, S. Wollowitz, J. E. Trend, F. Chow, D. F. Wendelborn, J. Org. Chem. 1978, 43, 16971705. 24 a) N. Miyaura, T. Ishiyama, M. Ishikawa, A. Suzuki, Tetrahedron Lett. 1986, 27, 63696372; b) N. Miyaura, T.

Ishiyama, H. Sasaki, M. Ishikawa, M. Satoh, A. Suzuki, J. Am. Chem. Soc. 1989, 111, 314321. 25 For an excellent review on the B-alkyl Suzuki-Miyaura cross-coupling, see: S. R. Chemler, D. Trauner, S. J.

Danishefsky, Angew. Chem. Int. Ed. 2001, 40, 45444568. 26 C. R. Johnson, M. P. Brown, J. Am. Chem. Soc. 1993, 115, 1101411015. 27 H. C. Brown, G. W. Kramer, J. Organomet. Chem. 1974, 73, 115. 28 E. Negishi, D. R. Swanson, C. J. Rousset, J. Org. Chem. 1990, 55, 54065409.


With intermediate 109 in hand, it was thus possible to attempt the projected anionic

oxy-COPE rearrangement.10

However, treatment of this substrate with KH or KHMDS in the

presence of 18-c-6 in THF only resulted in the cleavage of the TMS group at the alkyne which

most likely occurred by nucleophilic attack of the potassium alcoholate under formation of the

corresponding TMS ether 124 (Scheme 1.6). The latter was then most likely hydrolyzed

during aqueous work-up and 126 was usually obtained in about 80% yield.

Scheme 1.6. Results of the attemted anionic oxy-COPE rearrangements.

In an attempt to examine if the more strongly coordinating Na+ cation could help

suppressing this undesired side-reaction, 109 was subjected to NaH in THF at 70 C.

Unfortunately, however, these conditions performed even worse and mainly led to

decomposition of the starting material and the formation of minor amounts of 126 (21% yield)

and 127 (12% yield). Since the transfer of the TMS group from the alkyne to the alkoxide

prohibited the desired [3,3]-sigmatropic rearrangement from occurring at reasonable

temperatures, it was decided to try to regenerate the alkoxide by adding a fluoride source.

TASF29

appeared to be an ideal choice in this regard because it possesses an almost

non-coordinating counter-cation30

and in contrast to other reagents such as nBu4NF31

can be

29 a) W. J. Middleton, US Patent 3940402, 1976; b) W. J. Middleton, Org. Synth. 1986, 64, 221221; c) K. A. Scheidt, H.

Chen, B. C. Follows, S. R. Chemler, D. S. Coffey, W. R. Roush, J. Org. Chem. 1998, 63, 64366437. 30 a) R. Noyori, I. Nishida, J. Sakata, M. Nishizawa, J. Am. Chem. Soc. 1980, 102, 12231225; b) R. Noyori, I. Nishida, J.

Sakata, Tetrahedron Lett. 1981, 22, 39933996; c) R. Noyori, I. Nishida, J. Sakata, J. Am. Chem. Soc. 1983, 105, 1598. 31 A. S. Pilcher, P. DeShong, J. Org. Chem. 1996, 61, 69016905.


obtained in rigorously anhydrous form. Consequently, 3-hydroxyl-1,5-hexadiene 109 was

subjected to a THF solution containing KH and 18-c-6 and stirred at ambient temperature until

the starting material was completely consumed under the presumed formation of 124.32

Thereafter, 1.2 equivalents of TASF were added and a second transformation indeed started to

ensue as judged by TLC analysis. After reaching a conversion of approximately 50% the

reaction was aborted for product identification. However, NMR analysis of the newly formed

compound revealed that the regenerated alkoxide underwent a GROB fragmentation33

instead

of the desired oxy-COPE rearrangement and cyclopentenone 127 was now isolated in 58%

yield. Although the [3,3]-sigmatropic rearrangement is normally significantly faster than the

corresponding elimination or fragmentation these undesired side-reactions often occur if the

system is precluded from adopting the mandatory 6-membered, cyclic transition state.12

Consequently, the reluctance of 109 to undergo the projected anion-accelerated oxy-COPE

rearrangement and the observation of GROB fragmentation product 127 under more forcing

condition suggested that the desired transformation might be intrinsically impossible and thus

made it unlikely that the problem could be solved by alteration of the reaction conditions.

32 TMS ether 124 is not visible on TLC due to the instantaneous cleavage of the silyl group on the slightly acidic silica gel

and therefore it is only possible to infer its formation by monitoring 126. 33 For selected reviews, see: a) C. A. Grob, P. W. Schiess, Angew. Chem. Int. Ed. 1967, 6, 115; b) C. A. Grob, Angew.

Chem. Int. Ed. 1969, 8, 535622; c) K. Prantz, J. Mulzer, Chem. Rev. 2010, 110, 37413766.


1.3.2 Second Generation Approach

Accordingly, after the failure of the anionic oxy-COPE rearrangement strategy a new

synthetic plan towards tetrahydroindene dione 107 had to be devised and with the experiences

from the previous chapter in mind it appeared advisable to address the construction of the two

adjacent quaternary centers as early as possible in the synthetic route. In agreement with this

consideration, the second generation approach commenced with a DIELSALDER reaction

between known silyloxydiene 130 and commercially available 1-acetyl-1-cyclopentene (131),

thereby allowing for the envisioned rapid formation of the fused 5,6-membered ring system as

well as the concomitant construction of the two adjacent quaternary centers (Scheme 1.7).34

Scheme 1.7. Reagents and conditions: a) Me2AlCl (20 mol%), PhMe/CH2Cl2 (2:1), 15 C, 24 h; then 0 C, 24 h, 69%;

b) TFA (1.1 equiv), CH2Cl2, RT, 2 h, 89%; c) 2-ethyl-2-methyl-1,3-dioxolane (20 equiv), TsOHH2O (25 mol%), ethylene

glycol (12 mol%), RT, 45 min, 95%.

Despite the hindered character of diene 130 and the sometimes limited reactivity of

cyclopentene dienophiles (i.e. 131) in cycloaddition reactions, this transformation was found

to proceed well upon exposure of the two components to 20 mol% Me2AlCl in toluene/CH2Cl2

and after 48 h the corresponding DIELSALDER adduct 132 could be obtained in 69% yield. In

addition to the usual characterization by comprehensive NMR, IR and HRMS spectroscopy

the structure and relative configuration of the obtained product could be further corroborated

X-ray diffractometry (see Appendix 5.2.2). Subsequent acid-promoted hydrolysis of the TBS-

enol ether by treatment with a stoichiometric amount of TFA in CH2Cl2 followed by the

regioselective mono-acetalization of the obtained diketone at C9 with 2-ethyl-2-methyl-1,3-

dioxolane35

furnished methyl ketone 134 in 58% yield over three steps.

Next, the attention was turned to the introduction of the but-3-ynyl side chain which was

projected to be achieved by the conversion of methyl ketone 134 into the corresponding

34 a) M. E. Jung, D. Ho, H. V. Chu, Org. Lett. 2005, 7, 16491651; b) M. E. Jung, D. G. Ho, Org. Lett. 2007, 9, 375378. 35 H. J. Dauben, Jr., B.Lken, H. J. Ringold, J. Am. Chem. Soc. 1954, 76, 13591363.


terminal acetylene 135, followed by alkylation with ethyl iodide and finally isomerization of

the internal alkyne to the terminus. To this end, both of the two established procedures were

attempted in which the ketone in question is first transformed into the corresponding enol

phosphate or enol triflate, respectively, before subjection to a lithium amide base should effect

the desired elimination.36

However, after several unsuccessful experiments with these

traditional methods, recourse was taken towards a novel one-step procedure that was reported

to allow for the direct conversion of methyl ketones into terminal alkynes (Scheme 1.8).

Scheme 1.8. Reagents and conditions: a) LDA, THF, 78 C; then (EtO)2POCl; b) LDA, THF, 78 C to RT, 135 not

formed; c) LDA, THF, 78 C; then Tf2NPh; d) MeLi (4 equiv), EtI (8 equiv), THF, 0 C to RT, 89%; e) KH (6-8 equiv),

1,3-diaminopropane, 0 C or RT, slow decomposition of starting material.

Accordingly, treatment of methyl ketone 134 with phosphazene base P2tBu and F9C4SO2F

(NfF) in dry DMF and subsequent stirring at ambient temperature for 24 hours indeed

provided the desired terminal alkyne 135 in an excellent yield of 92%.37

Deprotonation of 135

with MeLi followed by trapping of the generated lithium acetylide with EtI furnished the

desired internal alkyne 136 in 89% yield and the stage was thus set for subjection of this

intermediate to the standard conditions of the acetylene-zipper reaction.38

Unfortunately,

however, it was in this respect found that exposure of alkyne 136 to an excess KH/1,3-

diaminopropane does not entail the desired isomerization to the terminal acetylene but instead

36 See the following and references therein: a) E. Negishi, A. O. King, W. L. Klima, W. Patterson, A. Silveira, Jr., J. Org.

Chem. 1980, 45, 25262528. b) E. Negishi, A. O. King, J. M. Tour, Org. Synth. 1986, 64, 4449. 37 a) I. M. Lyapkalo, M. A. K. Vogel, E. V. Boltukhina, J. Vavrk, Synlett 2009, 558561. b) I. M. Lyapkalo, M. A. K.

Vogel, Angew. Chem. Int. Ed. 2006, 45, 40194023; Angew. Chem. 2006, 118, 41244127. 38 C. A. Brown, A. Yamashita, J. Am. Chem. Soc. 1954, 97, 891892.


led to the slow decomposition of the starting material. Consequently, it became at this point

clear that an alternative route for the introduction of the side-chain had to be devised.

Starting again from the terminal alkyne 135 LEWIS acid promoted formylation of the

corresponding lithium acetylide with DMF/BF3OEt2 and subsequent reduction of the

intermediate propiolaldehyde with 10 mol% Pd/C under one atmosphere of H2 gave aliphatic

aldehyde 139 (Scheme 1.9).39

Subsequent transformation of the aldehyde into the

Scheme 1.9. Reagents and conditions: a) nBuLi (1.3 equiv), 78 C to RT, 40 min; then BF3OEt2 (1.3 equiv), 78 C,

10 min; then DMF (3.3 equiv), 78 C, 60 min, 78%; b) Pd/C (10 mol%), H2 (1 atm), EtOAc, RT, 2 h, 84%; c) PPh3

(4.0 equiv), CBr4 (1.9 equiv), CH2Cl2, 0 C, 45 min, 50% 140, 34% 141; d) nBuLi (2.5 equiv), THF, 78 C to RT, 2 h;

then TMSCl (1.8 equiv), 78 C to RT; then RT, 20 h, 84%; e) 0.1M aq. HCl (1.1 equiv), THF, reflux, 2 h, 89%;

f) dimethyl-1-diazo-2-oxopropylphosphonate (1.5 equiv), K2CO3 (2.3 equiv), MeOH, RT, 2.5 h, 88%; g) nBuLi

(1.5 equiv), TMSCl (1.8 equiv), THF, RT, 20 h; then 0.1M aq. HCl (1.0 equiv), THF, reflux, 4 h, 92%; h) LiHMDS

(1.4 equiv), THF, 78 C; then PhSeBr (1.4 equiv),1 h, 81%; i) mCPBA (1.3 equiv), iPr2NH (2.5 equiv), CH2Cl2, 78 C

to RT, 75%; j) LiHMDS (1.3 equiv), THF, 78 C; then PhS(=N-tBu)Cl (1.3 equiv), 2 h, 94%.

corresponding acetylene 137 was originally performed by means of the COREYFUCHS

protocol.40

However, similar to earlier reports by W. J. KERR and co-workers it was found

that, especially on larger scales, the initial formation of dibromoalkene with Ph3P/CBr441

was

accompanied by significant cleavage of the ketal protecting group at C9 and re-acetalization

of the aldehyde in the side-chain.42

By-product 141 was thereby obtained in up to 34% yield

39 K. Iguchi, M. Kitade, T. Kashiwagi, Y. Yamada, J. Org. Chem. 1993, 58, 56905698. 40 E. J. Corey, P. L. Fuchs, Tetrahedron Lett. 1972, 13, 37693772. 41 a) F. Ramirez, N. B. Desai, N. McKelvie, J. Am. Chem. Soc. 1962, 84, 17451747 ; b) H. J. Bestmann, H. Frey, Liebigs

Ann. Chem. 1980, 20612071; c) H. J. Bestmann, K. Li, Chem. Ber. 1982, 115, 828831. 42 a) J. G. Donkervoort, A. R. Gordon, C. Johnstone, W. J. Kerr, U. Lange, Tetrahedron 1996, 52, 73917420; b) C.

Johnstone, W. J. Kerr, J. S. Scott, Chem. Commun. 1996, 341342.


while the targeted dibromoolefin 140 was delivered only in modest 50% yield. Even though

the subsequent reaction of 140 with 2.5 equivalents of nBuLi and trapping of the transient

lithium acetylide with TMSCl proceeded uneventfully and, after an additional 1,3-dioxolane

cleavage with 0.1M aqueous HCl, furnished the desired ketone 145 the overall yield for this

short sequence was relatively low (37% over 3 steps). In addition to this, there were serious

concerns with regard to reproducibility on even larger scales. As an alternative, it was

therefore decided to switch to the corresponding OHIRABESTMAN procedure instead.43

Thus,

treatment of a methanolic solution of aldehyde 139 with 1.5 equivalents of dimethyl-1-diazo-

2-oxopropylphosphonate in the presence of K2CO3 directly afforded alkyne 137 in a much

improved yield of 88%. From there, the sequence was completed by a one-pot reaction which

involved protection of the corresponding lithium acetylide of 137 with TMSCl followed by

cleavage of the 1,3-dioxolane moiety with 0.1M aqueous HCl in THF. Albeit a step longer

than originally planned, this sequence allowed for the introduction of the but-3-ynyl side chain

and reliably delivered ketone 145 in 28% yield over 8 steps from the starting materials. With a

viable route for the synthesis of ketone 145 in hand, the remaining tasks towards the synthesis

of the required tetrahydroindene dione 151 were associated with the introduction of the

endione motif. To this end, ketone 145 was initially converted into the corresponding

-phenylselenyl ether (LiHMDS, THF, 78 C; then PhSeBr) which upon oxidation with

mCPBA underwent a selenoxide elimination to afford enone 146 in 61% yield over two steps.

However, it was later found that the transformation could be conveniently performed in a

single step and 94% yield under the action of the MUKAIYAMA dehydration protocol

(LiHMDS, THF, 78 C; then N-tert-butylbenzenesulfinimidoyl chloride).44

With ,-unsaturated ketone 146 in hand, a selection of linear dienol ethers 147 could be

selectively accessed in order to investigate their performance in the gamma-hydroxylation

according to the KIRKWILES procedure and modifications thereof (Table 1.1, Entries 16).45

In the experiments, the best results were eventually obtained by subjecting the corresponding

43 a) S. Ohira, Synth. Commun. 1989, 19, 561-564. b) S. Mller, B. Liepold, G. J. Roth, H. J. Bestmann, Synlett 1996,

521522. c) G. J. Roth, B. Liepold, S. G. Mller, H. J. Bestmann, Synthesis 2004, 5962. 44 a) T. Mukaiyama, J. Matsuo, H. Kitagawa, Chem. Lett. 2000, 12501251; b) T. Mukaiyama, J. Matsuo, M. Yanagisawa,

Chem. Lett. 2000, 10721073; c) J. Matsuo, D. Iida, T. Mukaiyama, Bull. Chem. Soc. Jpn. 2002, 75, 223234. 45 a) D. M. Kirk, J. M. Wiles, Chemm. Comm. 1970, 10151016; b) P. M. Wege, R. D. Clark, C. H. Heathcock, J. Org.

Chem. 1976, 41, 31443148; c) S. N. Suryawanshi, P. L. Fuchs, Tetrahedron Lett. 1981, 22, 42014204; For a similar

substrate, see: d) R. Iyengar, K. Schildknegt, M. Morton, J. Aub, J. Org. Chem. 2005, 70, 1064510652.


TBS dieneol ether, which was obtained in quantitative yield upon treatment of enone 146 with

TBSOTf in the presence of 2,6-lutidine, to 1.3 equivalents of KHSO5 in a mixture of saturated

aqueous NaHCO3 and acetone. Subsequent oxidation of the produced allylic alcohol 148 with

DESSMARTIN periodinane46

then concluded the sequence and allowed to synthesize the

targeted enedione 151 in 49% yield over 3 steps (Scheme 1.10).

Table 1.1. Optimization of the gamma-hydroxylation.

Entry Conditions 1 R Conditions 2

Yield of 148

1 isoprenyl acetate (120 equiv)

TsOHH2O (2 equiv), reflux Ac

mCPBA (2.5 equiv)

EtOH, RT, 18 h 30%

2 TESOTf (1.3 equiv)

2,6-lutidine (2.0 equiv) CH2Cl2, 0 C TES

mCPBA (1.2 equiv)

NaHCO3 (3.0 equiv), PhMe, 0 C 36%

3 TBSOTf (1.3 equiv)

2,6-lutidine (2.0 equiv) CH2Cl2, 0 C TBS

mCPBA (1.5 equiv)

NaHCO3 (6.0 equiv), CH2Cl2, 0 C 48%

4 TIPSOTf (1.3 equiv)

2,6-lutidine (2.0 equiv) CH2Cl2, 0 C TIPS

mCPBA (1.5 equiv)

NaHCO3 (6.0 equiv), CH2Cl2, 0 C 15%

5 TBSOTf (1.3 equiv)

2,6-lutidine (2.0 equiv) CH2Cl2, 0 C TBS

KHSO5 (1.3 equiv), NaHCO3

Acetone/H2O (2:1), RT 61%

6 CH(OMe)3 (1.5 equiv)

TsOHH2O (10 mol%), MeOH, RT Me

KHSO5 (1.3 equiv), NaHCO3

Acetone/H2O (2:1), RT 42%

However, inspired by a single report in literature, it was later found that subjection of TBS

dieneol ether 150 to an excess of CrO33,5-dimethylpyrazole complex in CH2Cl2 directly

entailed the formation of the desired enedione 151 in a single step and 66% yield.47

The

46 a) D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 41554156; b) D. B. Dess, J. C. Martin, J. Am. Chem. Soc. 1991,

113, 72777287; For an excellent compilation on hypervalent iodine mediated oxidations, see: c) G. Tojo, M. Fernndez

in "Oxidation of Alcohols to Aldehydes and Ketones" (Ed. G. Tojo), pp. 181214, Springer, Berlin, 2006. 47 a) J.-L. Brevet, G. Fournet, J. Gor, Synth. Commun. 1996, 26, 41854193; This reagent was originally designed for the

oxidation of alcohols to carbonyl compounds and later extended to allylic oxidations: b) E. J. Corey, G. Fleet, Tetrahedron

Lett. 1973, 14, 44994501; c) W. G. Salmond, M. A. Barta, J. L. Havens, J. Org. Chem. 1978, 43, 20572059.


tentatively assigned mechanism for this interesting gamma-oxidation thereby suggests an

initial attack by the TBS dienol ether 147 onto the electrophilic CrO3(3,5-dmp) complex

under formation of -chromate 149. This reactive intermediate would then undergo a

[2,3]-sigmatropic rearrangement with the allylic double bond to furnish the corresponding

chromate ester that upon abstraction of the carbinol hydrogen in the usual fashion delivers

endione 15148

With regard to the more conventional application of this reagent combination in

allylic oxidations it is also noteworthy that, probably due to the low temperature and the short

reaction times, no oxidation of the propargylic methylene position could be observed.47c

Scheme 1.10. Reagents and conditions: a) TBSOTf (1.3 equiv), 2,6-lutidine (2.0 equiv), CH2Cl2, 0 C, 90 min; b) KHSO5

(1.3 equiv), NaHCO3, acetone/H2O (1:1), RT, 4 h; c) DMP (1.5 equiv), CH2Cl2, RT, 45 min, 49% over 3 steps.

The envisioned chlorination of enedione 151 at the position of the 5-membered ring

ketone turned out to be surprisingly difficult. Selective formation of the lithium enolate at the

sterically more accessible C2 position with LiHMDS and subsequent treatment with TsCl

preferentially afforded the corresponding tosyl enol ether 153 in 60% yield instead of the

chlorinated product which is normally observed under these conditions (Table 1.2, Entry 1).49

Treatment of the same enolate with TfCl or NCS, on the other hand, entailed the rapid

formation of complex product mixtures (Entries 2 and 3). Nonetheless, after some additional

experimentation with other chlorinating conditions (e.g. Entries 4 and 5), it was eventually

found that transformation of 151 to the corresponding TBS enol ether 154, upon exposure to

48 For a discussion of chromium-based reagents in the oxidation of alcohols, see: c) G. Tojo, M. Fernndez in "Oxidation

of Alcohols to Aldehydes and Ketones" (Ed. G. Tojo), pp. 195, Springer, Berlin, 2006. 49 K. M. Brummond, K. D. Gesenberg, Tetrahedron Lett. 1999, 40, 22312234.


Et4NCl3 (MIOSKOWSKIS reagent), was directly transformed into the desired -chlorosilylenol

ether 154 thus rendering a subsequent re-formation of the TBS enol ether obsolete (Entry 6).50

Table 1.2. Attempts on the selective -chlorination of endione 151 at C2.

Entry Conditions Result Isolated Compounds

(Yield)

1 LiHMDS (1.1 equiv), TsCl (1.0 equiv)

THF, 78 C to RT, 2 h

full conversion, only one

major product, 152 not observed 153 (60%)

2 LiHMDS (1.1 equiv), TfCl (1.1 equiv)

THF, 78 C, 1 h

full conversion

complex mixture unknown

3 LiHMDS (1.2 equiv), NCS (1.2 equiv)

THF, 78 C, 30 min

full conversion


4 CuCl2(H2O)2 (1.1 equiv)

LiCl, DMF, 90 C, 60 min

full conversion


5 KHMDS (1.2 equiv), NCS (1.3 equiv)

THF, 78 C, 40 min

full conversion


6 1. KHMDS, TBSCl, THF, 78 C to RT

2. Et4NCl3, CH2Cl2, 78 C to RT

full conversion, dominant product

154, probably 1418% C8-chlorination 154 (51%)

50 T. Schlama, K. Gavriel, V. Gouverneur, C. Mioskowski, Angew. Chem. Int. Ed. 1997, 36, 23422344.


Having thereby established a feasible route toward -chlorosilylenol ether 154, the

attention could be turned to the assessment of the crucial CONIA-ene reaction. Gratifyingly,

after a few preliminary experiments with related precursors (e.g. the -unchlorinated TBS-

enol ether of 151, not shown) and minor changes in the reaction conditions, it was found that

subjection of -chlorosilylenol ether 154 to 50 mol% Au(MeCN)(JohnPhos)SbF6 in dry

acetone at 45 C after 5 h not only entailed the desired cyclization but also concomitantly

removed the meanwhile needless TMS protecting group (Scheme 1.11).51

Tricyclic olefin 105

was obtained in 65% yield and the stage was set to address the final hydrochlorination step.

Scheme 1.11. Reagents and conditions: a) KHMDS (1.3 equiv), TBSCl (2.0 equiv), THF, 78 C to RT, 3 h; b) Et4NCl3

(2.4 equiv), CH2Cl2, 78 C to RT, 5 h, 51% over 2 steps; c) HCl (gas), SnCl4 (30 equiv), CH2Cl2, 78 C; then sealed

tube, 78 C to RT, 5 h, 67%.

Due to the nearby electron-withdrawing functional groups, the exocyclic double-bond was

rather unreactive towards the desired transformation and attempts with some milder protocols

such as the cobalt-catalyzed hydrochlorination52

or with AcCl/EtOH, a mixture that was

reported to slowly generate HCl in situ,53

did not deliver any of the desired addition product.

To this end, resorting to vigorous conditions, namely saturation of a CH2Cl2 solution of the

olefin 105 with gaseous HCl at 78 C in the presence of 30 equivalents of SnCl4 followed by

slow warming to ambient temperature in a sealed tube (pressure build-up), allowed to bring

about the desired reaction.54

The 1H and

13C NMR spectra of the product that was obtained

from this hydrochlorination reaction were in very good agreement with those assigned to

nominal gomerone C (3-epi-102).

51 a) F. Barab, G. Btournay, G. Bellavance, L. Barriault, Org. Lett. 2009, 11, 42364238; b) Nieto-Oberhuber, S. Lpez,

M. P. Muoz, D. J. Crdenas, E. Buuel, C. Nevado, A. M. Echavarren, Angew. Chem. Int. Ed. 2005, 44, 61466148; b)

C. Nieto-Oberhuber, S. Lpez, A. M. Echavarren, J. Am. Chem. Soc. 2005, 127, 61786179. 52 B. Gaspar, E. M. Carreira, Angew. Chem. Int. Ed. 2008, 47, 57585760. 53 V. K. Yaday, K. G. Babu, Eur. J. Org. Chem. 2005, 452456. 54 For a comprehensive overview, see the following and references therein: R. C. Larock, W. W. Leong in Comrehensive

Organic Synthesis, Vol. 4 (Eds: B. M. Trost, I. Fleming), Pergamon Press, Oxford, 1991, pp. 269327.


Table 1.3. 1H and 13C NMR spectroscopic data for natural gomerone B and C, and synthetic gomerone C.

Atom

No.

1H NMR Chemical Shifts ()

[a]

of Natural Gomerone B

[c] 1H NMR Chemical Shifts ()

[a]

of Natural Gomerone C

[c] 1H NMR Chemical Shifts ()

[b]

of Synthetic Gomerone C

1 2.96 (d, J = 12.4 Hz)

2.18 (dd, J = 12.4, 3.1 Hz)

0.00

0.03

2.95 (d, J = 12.3 Hz)

2.14 (dd, J = 12.3, 3.0 Hz)

0.01

0.01

2.96 (d, J = 12.4 Hz)

2.15 (dd, J = 12.4, 2.9 Hz)

4 2.46 (ddd, J = 15.9, 4.9, 1.8 Hz)

1.72 (ddd, 15.9, 12.8, 4.9 Hz)

0.12

0.01

2.32 (m)[d]

1.70 (ddd, 12.8, 12.8, 5.4 Hz)

0.02

0.01

2.34 (ddd, J = 15.4, 5.0, 1.9 Hz)

1.71 (ddd, 15.4, 13.1, 5.0 Hz)

5 2.31 (ddd, J = 13.3, 13.3, 4.9 Hz)

1.98 (dddd, J = 13.3, 4.9, 2.4, 2.4 Hz)

0.03

0.01

2.26 (m)[e]

1.97 (m)

0.02

0.00

2.28 (app. dt, J = 13.1, 5.0 Hz)

1.97 (dddd, J = 13.1, 5.0, 2.8, 2.0 Hz)

8 6.57 (br s) 0.01 6.57 (br s) 0.01 6.58 (d, J = 1.0 Hz)

10 2.67 (d, J = 16.8 Hz)

2.21 (d, J = 16.8 Hz)

0.01

0.01

2.67 (d, J = 17.4 Hz)

2.21 (d, J = 17.4 Hz)

0.01

0.01

2.68 (dd, J = 17.1, 0.8 Hz)

2.22 (d, J = 17.1, 1.0 Hz)

12 1.14 (s) 0.00 1.13 (s) 0.01 1.14 (s)

13 0.99 (s) 0.01 0.99 (s) 0.01 1.00 (s)

15 1.90 (s) 0.19 1.70 (s) 0.01 1.71 (s)

[a] Chemical shifts referenced to the residual solvent signal of CDCl3 ( 7.25 ppm). [b] Chemical shifts referenced to SiMe4

( 0.00 ppm) with residual solvent peak of CDCl3 measured at 7.26 ppm. [c] Shift difference of 0.01 ppm due to dissimilar referencing.

[d] Proton originally assigned to C(5); reassigned to C(4) by the author of this thesis. [e] Proton originally assigned to C(4); reassigned

to C(5) by the author of this thesis.

Atom

No.


[a]

of Natural Gomerone B


[a]

of Natural Gomerone C


[b]

of Synthetic Gomerone C

1 43.9 1.9 42.0 0.0 42.0

2 79.9 0.4 79.5 0.0 79.5

3 70.8 2.7 73.6 0.1 73.5

4 40.2 1.2 39.0 0.0 39.0

5 29.9 -0.5 29.4 0.0 29.4

6 48.0 0.1 47.9 0.0 47.9

7 155.0 0.6 154.3 0.1 154.4

8 125.0 0.1 125.1 0.0 125.1

9 204.6 6.1 199.8 1.3 198.5

10 48.9 0.1 49.0 0.0 49.0

11 38.4 0.0 38.4 0.0 38.4

12 24.3 0.0 24.3 0.0 24.3

13 24.0 0.0 23.9 0.1 24.0

14 196.3 1.0 198.2 0.9 197.3

15 29.7 1.9 27.8 0.0 27.8


This, however, was rather puzzling, because it suggested that the double-bond in 105 had

undergone selective hydrochlorination from the endo-face of the bicyclo[3.2.1]octane

substructure; which molecular modelling indicated is the more hindered face (Figure 1.2A).

Consistent with our hesitation, no nuclear Overhauser effect (NOE) was observed between the

supposedly axial C15 methyl group and the two protons in the 1,3-diaxial position at C1 and

C5, even though a chair-conformation was clearly indicated by all the 1H,

1H-coupling

constants for the 6-membered ring.

Figure 1.2. A) Stereochemical model for the diastereoselectice hydrochlorination of the exocyclic olefin. B) Crystal

structure of gomerone C (102, 3-epi-103) and structural revision. Thermal ellipsoids are set at 50% probability.

Definitive evidence was obtained when it was possible to grow single crystals that were

suitable for X-ray diffractrometry (Figure 1.2B). The crystallographic data clearly showed that

in gomerone C (3-epi- 103) the chloride, and not the methyl group, occupied the axial position

(see Appendix 5.2.3). Consequently, the structure of gomerone C (3-epi-103) necessitated

revision, which we proposed corresponds to nominal gomerone B (102). It would therefore

appear that in the original isolation and characterization work the structures of gomerone B

and C had been interchanged. In agreement with this, the structure of gomerone B (3-epi- 102)

was therefore reassigned to that of nominal gomerone C (103).

Conclusions and Outlook 25

1.4 Conclusion and Outlook

In conclusion, the first total synthesis of gomerone C, which was achieved in 15 linear

steps and 4% overall yield, resulted in the stereochemical revision of the structure of the

natural product, i.e. gomerone C (102), and hence also led to reassignment of the related

gomerone B (103). The synthesis relied on a key, late-stage CONIA-ene reaction betwen a

chlorinated enol silane and an alkyne, which provided efficient access to the novel gomerane

skeleton with the attendant bridgehead chloride. The formed bicyclo[3.2.1]octane core was set

up for the subsequent introduction of the second tertiary chloride. Additional salient features

of the synthesis included a DIELSALDER cycloaddition, which introduced two contiguous

quaternary centers under concomitant formation of the fused perhydroindene core. Moreover,

in the context of a total synthesis the use of Schwesingers P2tBu base/F9C4SO2F for the

convenient one-step conversion of a ketone into an alkyne as well as the selective oxygenation

of a 1-silyloxydiene mediated by CrO3DMP could be showcased in more complex settings.

2 Total Synthesis of

Fluorodanicalipin A



2.1 Introduction

2.1.1 Isolation and Occurrence of Chlorosulfolipids

In 1969, J. ELOVSON and P. R. VAGELOS described for the first time the isolation and

identification of several chlorinated docosane-1,14- and tetracosane-1,15-disulfates55

from

cell-free extracts of the freshwater chrysophyte Ochromonas danica.56

Derivatization to the

corresponding TMS-ethers and analysis by GLC-MS revealed that these unusual lipids

contained one to six chlorines, of which maximally two were in a distal (C15 to C22) and four

in a proximal (C1 to C13) position. In the same year, and only shortly thereafter, T. H. HAINES

and co-workers independently found that the hexachlorinated (43%) and the monochlorinated

(30%) docosane-1,14-disulfates were the most abundant chlorosulfolipids in this mixture,

whereas other homologs contributed less than 10% each to the total mass. In addition, these

authors succeeded with the first complete structural elucidation of a chlorosulfolipid which

was determined to be (13R,14R)-13-chlorodocosane-1,14-disulfate (201) (Figure 2.1A).57

The

chemical constitution of the major component, on the other hand, could at that time only be

established by means of arduous derivatization and degradation studies and was devoid of

stereochemical information. Nonetheless, one year after its initial discovery, ELOVSON and

VAGELOS correctly identified this structure as 2,2,11,13,15,16-hexachloro-n-docosane-1,14-

disulfate (202)58

and it was not until 2009 that the relative stereochemistry of this intriguing

metabolite could be assigned by the g

total synthesis of (±)-gomerone c (±)-fluorodanicalipin a and

Documents