Terpenoids and Steroids: Volume 2 (SPR Terpenoids and Steroids (RSC))

A Specialist Periodical Report Terpenoids and Steroids Volume 2 A Review of the Literature Published between Septembe...

Author: K H Overton

42 downloads 1417 Views 21MB Size Report

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

DOWNLOAD PDF

A Specialist Periodical Report

Terpenoids and Steroids Volume 2

A Review of the Literature Published between September 1970 and August 1971

Senior Reporter

K. H. Overton, Department of Chemistry, University of Glasgo w Reporters J. D. Connoliy, University of Glasgow

P. Crabbe, National University of Mexico J. R . Hanson, University of Sussex D. N. Kirk, Westfield College, University of London

G. P. MOSS, Queen Mary College, University of London J. S. Roberts, University of Stirling

A. F. Thomas, Firmenich et Cie., Geneva, Switzerland

0 Copyright 1972

The Chemical Society Burlington House, London, W I V OBN

I S B N : 0 85186 266 7 Library of Congress Catalog Card No. 74-61 5720

Set in Times on Monophoto Filmsetter and printed offset by J. W. Arrowsmith Ltd., Bristol, England Made in Great Britain

General Introduction

The period covered by this Report is September 1970 to August 1971. The aims of our survey and our presentation of it remain as set out in the General Introduction to last year’s Report. J.D.C. P.C. J.R.H. D.N.K.

G.P.M. K.H.O. J.S.R. A.F.T.

Contents Part I Terpenoids Introduction

3

Chapter 1 Monoterpenoids By A. F. Thomas 1 Analytical Methods and General Chemistry

5

2 Biogenesis and Biological Activity

6

3 Acyclic Monoterpenoids Telomerization of Isoprene 2,6-Dimethyloctanes Artemisyl, Santolinyl, Lavandulyl, and Chrysanthemyl Derivatives

13

4 Monocyclic Monoterpenoids Cyclopentanes, including Iridoids p-Menthanes General Chemistry and Hydrocarbons Oxygenated p-Menthanes o-Menthanes Tetramethylcyclohexanes Cycloheptanes

16 16 21 21 23 34 35 35

5 Bicyclic Monoterpenoids Bicyclo[3,1,O]hexanes Bicyclo[2,2,llheptanes Bicyclo[3,1,llheptanes Bicyclo[4,1,O]heptanes

36 36 38 48 56

6 Furanoid Monoterpenoids

58

7 Cannabinoidsand other Phenolic Monoterpenoids

60

Chapter 2 Sesquiterpenoids By J. S. Roberts 1 Farnesane

65

Terpenoids and Steroids

vi 2 Monocyclo- and Bicyclo-Farnesanes

71

3 Bisabolane and Sesquicarane

73

4 Daucane

75

5 Cadinane and Related Tricyclic Sesquiterpenoids

76

6 Campherane and Santalane

79

7 Thujopsane, Acorane, Chamigrane, Bazzanane, and Trichothecane

82

8 Longifolane

86

9 Caryophyllane, Humulane, and Related Compounds

88

10 Germacrane

92

11 Elemane

98

12 Eudesmane

100

13 Eremophilane, Valencane, Vetispirane, Trkyclovetivane, etc.

103

14 Guaiane

113

15 Aristolane

121

16 General

123

Chapter 3 Diterpenoids By J. R. Hanson 1 Introduction

124

2 Physical Methods

124

3 Bicyclic Diterpenoids The Labdane Series The Clerodane Series

126 126 128

4 Tricyclic Diterpenoids Pimaranes Abietanes Rosanes Cassane and Miscellaneous Tricyclic Diterpenoids The Chemistry of Ring A The Chemistry of Ring B The Chemistry of Ring c

129 129 130 133 134 135 136 137

vii

Contents

5 Tetracyclic Diterpenoids The Kaurene Series Trachylobanes Gibberellins Grayanotoxins

140 140 145 145 147

6 Diterpene Alkaloids

148

7 Macrocyclic Diterpenoids and their Cyclization Products Phorbol and its Relatives Taxane Diterpenes

149 149 151

8 Diterpenoid Synthesis

152

Chapter 4 Triterpenoids By J. 0.Connolly 1 Squalene Group

155

2 Fusidane-Lanostane Group

159

3 Dammarane-Euphane Group Quassinoids Baccharis Oxide

163 167 168

4 Lupane Group

169

5 Oleanane Group

170

6 Hopane Group

176

7 Onocerane Group

179

Chapter 5 Carotenoids and Polyterpenoids B y G. P. Moss 1 Introduction

180

2 Physical Methods

180

3 Carotenoids Acyclic Carotenoids Cyclic Carotenoids Allenic and Acetylenic Carotenoids Isoprenylated Carotenoids Carotenoid Reactions

183 183 184 188 190 191

4 Degraded Carotenoids

192

5 Polyterpenoids

195

viii


Chapter 6 Biosynthesis of Terpenoids and Steroids By G. P. Moss 1 Introduction

197

2 Acyclic Precursors

198

3 Hemiterpenoids

20 1

4 Monoterpenoids Cyclopentanoid Monoterpenoids

202 203

5 Sesquiterpenoids

204

6 Diterpenoids

208

7 Steroidal Triterpenoids Cyclization of Squalene Steroidal Trisnortriterpenoids Loss of the 4,4-Dimethyl Groups Loss of the 14a-Methyl Group and Isomerization of the Double Bond Side-chain Alkylation A22-Double Bond

210 2 10 21 1 213

8 Cholesterol Metabolism Spirostanols, Cardenolides, and Related Compounds Side-chain Cleavage Metabolism of the Steroid Nucleus

216 217 218 219

9 Triterpenoids

219

213 214 215

10 Carotenoids

22 1

11 Taxonomy

223 224

Arthropod Sterols

Part I/ Steroids Introd uct ion

227

Chapter 1 Steroid Properties and Reactions B y D. N. Kirk 1 Structure, Stereochemistry,and Conformational Analysis Spectroscopic Methods Chiroptical Properties (O.R.D. and C.D.) N.M.R. Spectroscopy Mass Spectrometry

229 23 1 232 237 240

ix

Contents 2 Alcohols, their Derivatives, Halides, and Epoxides Nucleophilic Substitution Solvolytic and Elimination Reactions Ring-opening of Epoxides Esters, Ethers, and Related Derivatives of Alcohols Oxidation Reduction Miscellaneous

242 242 243 245 246 247 249 25 1

3 Unsaturated Compounds Electrophilic Addition Other Addition Reactions Reduction of Unsaturated Steroids Oxidation and Dehydrogenation Miscellaneous Reactions

253 253 258 264 265 268

4 Carbonyl Compounds Reduction of Ketones Other Reactions at the Carbonyl Carbon Atom Oxidation and Dehydrogenation Enolization and Related Reactions Reactions of Enolate Anions Reactions of Enol Derivatives and Enaminzs Oximes Other Nitrogen-containing Derivatives of Ketones Sapogenins : Reactions of the Spiro-acetal System Reactions of Aldehydes, Carboxylic Acids, and their Derivatives Miscellaneous

269 269 272 273 274 276 278 280 28 1 283

5 Compounds of Nitrogen and Sulphur

290

6 Molecular Rearrangements Contraction and Expansion of Steroid Rings The ‘Westphalen’ and ‘Backbone’ Rearrangements Epoxide Rearrangements Aromatization Miscellaneous Rearrangements

298 298 301 306 309 313

7 Functionalization of Non-activated Positions

316

8 Photochemical Reactions Unsaturated Steroids Carbonyl Compounds Miscellaneous Photochemical Reactions

3 19 319 321 323

9 Miscellaneous Reactions Analytical Methods

326 326

286 289

Terpnoids and Steroids

X

Chapter 2 Steroid Synthesis By P. Crabbe In collaboration with G. A. Garcia, J. Haro, L. A. Maldonado, C. Rius, and E. Santos 1 Introduction

329

2 Total Synthesis

329

3 Photochemical Reactions

338

4 Halogeno-steroids

343

5 Oestranes

348

6 Androstanes

363

7 Pregnanes and Corticoids

372

8 Seco-steroids

388

9 Cholestane and Vitamin D, and its Analogues

398

10 Steroidal Insect and Plant Hormones

406

11 Steroidal Alkaloids

412

12 Sapogenins

424

13 Bufadienolides

427

14 Cardenolides

430

Errata

435

Author Index

436

Part I TERPENOIDS

Introduction"

Unexpected results have come to light bearing on monoterpenoid biosynthesis (Chapter 1). Banthorpe's group have shown",'3 that in the formation of the thujane and camphor skeletons, activity from labelled mevalonic acid can appear predominantly in the C, unit supposedly derived from isopentenyl pyrophosphate and only to a minor extent in the dimethylallyl pyrophosphate-derived portion. Banthorpe has also presented52 evidence for a chrysanthemyl intermediate, analogous to presqualene alcohol, in the biosynthesis of artemesia ketone. Laboratory synthesis again dominates the year's activity in the sesquiterpenoid field (Chapter 2) and continues to elicit much ingenuity. Notable are the routes developed5 by Corey's group to the C,, and C,, Cecropia juvenile hormones, the synthesis56 of trichodermin, the first member of the trichothecane group to be synthesized, syntheses4' of copacamphor, copacamphene, and cyclocopacamphene, and extension42 of Money's camphor synthesis to campherenone and campherenol with potential for further elaboration to e.g. longifoline and sativine. Routes to nootkatone and a-vetivone,' 1 9 , 1 2 0 zizanoic acid,lZ6 and p a t c h ~ u l e n o n e also ' ~ ~ merit mention among a long list of synthetic achievements. The structure22 of bilobalide, a highly oxygenated sesquiterpenoid containing the unusual t-butyl group, is of special interest. It could be derived from the structurally related C,, gingkolides, whose biosynthesis has been clarified.23 X-Ray analysis, increasingly by the direct method, is coming into routine use for structure determination. A notable concentration of effort is evident in the l 7 and t a ~ a n e ' series ~ ~ of diterpenoids p h ~ r b o l , ' ~ ~ - - 'grayanotoxin' ~~,'~~ (Chapter 3), where novel skeletons and complex functionality make pre-X-ray methods quite unsuitable. But the relatively ready access to X-ray facilities is ~) to underlined by analyses (e.g. grayanotoxin-1,117and t a ~ i n i n e ' ~undertaken establish doubtful points of stereochemistry. The structure of presqualene alcohol has been established beyond reasonable doubt by three independent rational syntheses (Chapter As the last isolable intermediate between acetate and squalene to be formulated, its structure has been a subject of controversy since its isolation in 1966, Its formulation therefore represents a major advance which makes it possible to consider its mode of formation from farnesol and its transformation into squalene. Enzymic '9,

4).334,5

* Reference numbers are those of the relevant chapter.

4


and non-enzymic cyclizations of oxidosqualene and related substances continue with vigour. (+)-Malabaricanediol is the first natural product to be formed' in vitro by cyclization of a squalene derivative. On the basis of numerous in uivo and in vitro experiments, van Tamelen has delineated12 the minimum substrate requirements of the enzyme 2,3-oxidosqualene sterol cyclase. New skeletal types of triterpenoids now appear only rarely. Baccharis oxides6 is such a type, but its structure is readily derivable from an intermediate cation, the result of squalene cyclization, which is assumed to lead to the lupane, oleanane, and ursane families of triterpenoids. The total synthesis of unsymmetrical triterpenoids has represented a major challenge for many years ; this year has seen the completion of total syntheses of germanic01~~ and alnusenone.' The unusual structure of the carotenoid pigment peridinin required for its solution' the collaboration of four laboratories and a combination of all available physical techniques (Chapter 5). The problem of whether cis- or trans-olefinic double bonds are involved in any particular polyisoprenoid biosynthesis has been brought into prominence this year (Chapter 6). Thus the sesquiterpenoid dimer gossypol is bio~ynthesized~ from cis, cis-farnesyl pyrophosphate. However, it is not clear whether the central cis-unit is incorporated as such (as in the case of rubber) or whether geranyl pyrophosphate is isomerized to neryl pyrophosphate before the third C, unit is added. Nerol itself is formed, like geraniol, initially from all-trans units and must therefore include an isomerization step in its genesis. These results raise the interesting possibility that any of the appropriate geometrically isomeric openchain polyenes may be involved in a particular polyisoprenoid biosynthesis. The long-postulated 1,2-hydrogen shift from C-13 to C-17 in the biosynthesis of lanosterol and P-amyrin has been demonstrated by incorporation of the appropriately tritiated oxidosqualene. Euphol is excluded142 as a biosynthetic precursor of the quassinoid bitter principle glaucaroubolone by incorporation experiments with the appropriately tritiated mevalonic acid, and lanosterol has been similarly excluded7' from curcurbitacin biosynthesis. An interesting result to emerge from biosynthetic studies59 with mycophenolic acid is that the side chain represents a degraded farnesyl rather than geranyl unit. Nakanishi and his colleagues have proposed a mast ingenious biogenetic derivation67 for gingkolide B from a pimarane; the unusual t-butyl group is formed from an isopropylidine group (ex C-4) and methionine.

I Monote r peno ids BY A. F. THOMAS

Although this report covers the period from September 1970 to August 1971, certain earlier publications that came too late for inclusion in the previous Specialist Report in this series will be mentioned. It is depressing to find, among the papers reviewed, several reporting works that had been previously published. 1 Analytical Methods and General Chemistry The problems associated with lability of double bonds during the mass spectrometric examination of monoterpenes have been discussed.' The mass spectra of ketones are not as easy to interpret as those of thioketones, the latter having a higher proportion of heteroatom-containing fragments. They are readily available by reaction of the ketones either with phosphorus pentasulphide, or with hydrogen sulphide and dry hydrogen chloride, and are recommended for the study of bicyclic ketones in the norbornane series2 The mass spectra of many monoterpenoids have been publi~hed.~ Analysis by gas chromatography of the mixture which constitutes the sex pheromone of the boll weevil (Anthonornus grandis Boheman) has been d e ~ c r i b e d .It~ consists of a cyclobutane monoterpenoid (Vol. 1, p. 18), and three 3,3-dimethyl-A'."cyclohexane-ethanols and -acetaldehydes. Scott and Wrixon have developed a quadrant rule for the c.d. of platinum(1rF olefin complexes that depends on d d orbital transitions. Application of the rule to rnonoterpenes was considered, and generally conformed to expectations based on known absolute configurations, but in some cases (notably #?-pinene) the results were not satisfactory.' The complex measured may be that of a-pinene, for which a Cotton curve of the opposite sign is predicted. Further work on the use H. Rapoport and U. T. Bhalerao, J . Amer. Chem. SOC.,1971,93, 105. M. M . Campbell, G. M. Anthony, and C. J. W. Brooks, Org. Mass Spectrometry, 1971, 5 , 297. E. VOASydow, K. Anjou, and G. Karlsson, Arch. Mass Spectral Data, 1970,1,392, and subsequent papers. D. L. Bull, R. A. Stokes, D. D. Hardee, and R. C. Gueldner, J . Agric. Food Chem., 1971, 19, 202. A. I . Scott and A. D. Wrixon, Tetrahedron, 1971, 21, 2339.

6


of I9F n.m.r. spectra of terpene alcohol derivatives has appeared.6 The interaction of epoxide with the hydroxy-group in the epoxypulegols has been examined by following the i.r. frequency of the OH band.' In the course of an examination of the autoxidation of terpene hydrocarbons, Bardyshev and Shavyrin have found, predictably, that those containing conjugated double bonds (e.g. allo-ocimene, myrcene) are oxidized most rapidly, those with isolated double bonds or cyclopropane rings more slowly (e.g. limonene, carene), and those with a single double bond slowest (e.g. pinene). The effect of light, heat, and inhibitors was studied.8 The rearrangement of monoterpenoid epoxides on alumina' and silica gel' surfaces has been studied. On the latter support, the rearrangements are typical of carbonium ions.

2 Biogenesis and Biological Activity The main advances in monoterpenoid biogenesis have been achieved by Banthorpe's group, who have extended their work (published earlier in note form) on the thujane derivatives obtained from Thuja, Tanacetum, and Juniperus species. More than 90 % of the label from [2-'4C]mevalonic acid is incorporated in that part of the skeleton derived from isopentenyl pyrophosphate, the part supposedly derived from 3,3-dimethylallyl pyrophosphate being essentially unlabelled. These results are not consistent with the accepted view that both isopentenyl and 3,3-dimethylallyl pyrophosphates are directly derived from mevalonic acid. However, in a second experiment concerned with the incorporation of [2-14C]mevalonic acid into the petals of rose flower heads, the results accorded with the accepted pattern, geraniol being labelled as in (I), with a similar distribution being found in nero1.l' The anomaly in the thujane experiments could be explained by the existence of a metabolic pool of dimethylallyl pyrophosphate, by compartmentation effects, or by a non-mevaloid source for the compound. In this connection it is possibly significant that the leaf and stem tissues employed in the thujane work contain discrete oil glands not seen in petal tissue. In the biosynthesis of (+)- and (-)-camphor in Artemisia, Saluia, and Chrysanthemum species, 73--83% of the label is incorporated from [2-'4C]mevalonic acid at C(6) as shown in (2); again, that part of the skeleton supposedly derived from 3,3-dimethylallylpyrophosphate was not equivalently labelled. The biogenesis

'

''

W. Ebbinghausen, E. Breitmaier, Ci. Jung, and W. Voelter, Z . Naturjorsch, 1970, 25b, 1239; H.-J. Schneider, G . Jung, E. Breitmaier, and W. Voelter, Tetrahedron, 1970, 26, 5369. ' T. Suga, S. Watanabe, T. Shishibori, and T. Matsuura, Bull. Chem. Soc. Jupun, 197 1 , 4 4 , 204. I . I . Bardyshev and V. S. Shavyrin, Sbornik. Trudy., Tsent. Nauch., Issled. Proekt. I n s t . Lesokhim. Prom., 1969, 15, 23 (Chem. Abs., 1 9 7 1 , 7 5 , 2 0 6 4 7 , 2 0 639). V. S. Joshi, N. P. Damodaran, and Sukh Dev, Tetrahedron, 1971, 27, 459. l o V. S. Joshi, N. P. Damodaran, and Sukh Dev, Tetrahedron, 1 9 7 1 , 2 7 , 4 7 5 . l 1 D. V. Banthorpe, J . Mann, a n d K. W. Turnbull, J . Chern. Soc. ( C ) , 1970, 2689. ' * M. J . 0. Francis, D . V. Banthorpe, and G. N. J . Le Patourel, Nufure,'1970,228,1005. l 3 D . V. Banthorpe and D. Baxendale, J . Chem. SOC.(0,1970, 2694.

Mono terpenoids

7

of the artemisia monoterpenoids is mentioned later.

Zavarin has continued his chemotaxonomic approach to biogenetic problems with a study of the leaf monoterpenes of some Cupressus species.14 Tidd has clarified the role of pyrophosphates in terpene biogenesis by measuring the hydrolysis rate of isopentenyl pyrophgsphate and related pyrophosphates over the physiological pH range. Potty and Bruemmer, continuing their search for enzymes causing transformations of terpenes in citrus fruits, have discovered a system that reduces (+)-limonene [but not (-)-limonene] in the orange.16 Because of their ready availability, there is a constant search for possible uses for the more common naturally occurring terpenes and their simple derivatives. This year has seen the claim of insecticidal 1 7 * and juvenile hormone l 8 activity for esters of geraniol and its epoxide (see below). Pharmacological (hypoglycaemic) activity was found in the piperidinesulphonamide of D-camphor endo-3-carbonic acid,' but less successful were the esters of guaiacol, thymol, and carvacrol, which were almost non-toxic.20 Some of the 1-(1'-hydroxyethy1)2,2-dimethyl-3-(2'-dialkylaminoethyl)cyclobutanes(3), obtained from the reduction of pinonic acid amides, are reported to show antiparkinson activity.21 1-01s (4) are claimed to be growth Quaternized 2-dimethylaminomenth-8-enregulants, nematicides, and fungicides,22and P-pinene resins are said to potentiate a herbicide.23

' l6 l7

'' 2o 21 22

23

E. Zavarin, L. Lawrence, and M. C. Thomas, Phytochernistry 1971, 10, 379. B. K. Tidd, J. Chem. SOC.( B ) , 1971, 1168. V. H. Potty and J. H . Bruemmer, Phytochemistry, 1970, 9 , 2319. H. Lee, J . J. Menn, and F. M. Pallos, Ger. Offen. 2 023 791 (Chem. A h . , 1971, 74, 31 868); Ger. Offen, 1 9 3 2 062 IChem. A h . , 1971,74, 22 682). J. Ratusky and F. Sorm, Ger. Offen. 2 022 363 (Nov. 19, 1970). H. Bretschneider, K. Hohenlohe-Oehringen, A . Grussner, and K. zur Nedden, Ger. Offen. 2 004 327 (Chem. A h . , 1971,74, 13 301). F. De Marchi, M. V. Torrielli, and G . Tamagone, Chim. Ther., 1968, 3,433. P. Schenone, G . Minardi, and M . Longobardi, Farmaco, Ed. Sci., 1970, 25, 533. W. F. Newhall, U.S. P. 3 564 046 (Chem. A h . , 1971,74, 100 237). W. Hurtt and A. R. Templeton, Chem. and Eng. News, 1971,49, No. 2, 25.


8 3 Acyclic Monoterpenoids

Telornerizatisn of Isoprene.-Reviews have appeared on isoprene24 and chloro~ r e n e and , ~ ~on the complex reactions of isoprene to form terpenoids26 (in Japanese). Isoprene reacts with magnesium, especially in the presence of Lewis acids, and the resulting complex gives adducts with aldehydes. As usual in this type of reaction, a very complex mixture is ~ b t a i n e d . ~The ' palladiwn-chloridecatalysed reaction of isoprene with acetic acid gives different products in different solvents. Monomers predominate in benzene [2-methylbut-2-enyl acetate (5) and 3-methylbut-2-enyl acetate (6)]while dimers [(7),(8),neryl(9), and geranyl(10) acetates] tend to be formed in tetrahydrofuran.28 Further details of the synthesis of C,, alcohols from isoprene and naphthyl-lithium are available,29 as well as of the in situ ~xidation,~' but there is little of novelty (see Vol. 1, p. 17).

(9)

2,6-Dimethyloctanes.-The full account of the synthetic work on achillene (see Vol. 1, p. 9) includes a technique for improvement of the yield of natural cis-achillene (12) by irradiation of the trans-compound (1I), in the presence of benzophenone, the equilibrium mixture containing 45 % cis-a~hillene.~Thermal isomerizatiori of cis-p-ocimene (13) [ = (16)]yields 6-cis-allo-ocimene (14)without any trans-isomer (15 ) ; this is presumably because the preferred conformer (16) has the bulky isobutenyl group in a pseudo-equatorial position (the 6-trans-

'

24 25

26 27

2n 29

30 31

W. J . Bailey, High Polymers, 1971, 24, part 2, 997. P. S. Bauchwitz, J . B. Finlay, and C . A. Stewart, jun., High Polymers, 1971, 24, part 2, 1149. K. Suga and S. Watanabe, Yukcgaku, 1970,19, 1061. M. Yang, K. Yamamoto, N. Otake, M. Ando, and K. Takase, Tetrahedron Letters, 1970, 3843. K. Suga, S. Watanabe, and K. Hijikata, Austral. J. Chem., 1971,24, 197. S. Watanabe and K. Suga, Austral. J. Chem. 1971, 24, 1301. K. Suga, S. Watanabe, T. Watanals, and M. Yonemitsu, Yukagaku, 1971,20,82 (Chem. A h . , 1971, 75, 20 652). K. H. Schulte-Elte and M. Gadola, Helu. Chim. Actu, 1971, 54, 1095.

Monoterpenoids

9

product would require the pseudo-axial position for this group). The 6-transisomer (15) can be made by treating cis-p-ocimene(13) with potassium t-butoxide, both isomers then being formed.32 Sasaki et al. have examined the 1,4-cycloadditions of nitrosobenzene with isoprene, chloroprene, and myrcene (17). The proportions of the two products (18) and (19)from the latter compound vary with

3 '0 hv-sens.

7 c

t e m p e r a t ~ r e .The ~ ~ same authors have found that the 1,4-cycloadditionreactions of allo-ocimene [a mixture of (14)and (15)] with a variety of dienophiles all appear to occur via the trans-compound (15).34 Cyclization of allo-ocimene in sodium and isopropylamine leads to a mixture of alkylated cycloheptadienes (20)--(23), only 5 % possessing the eucarvone skeleton (23).35

45 %

5%

'' T. Sasaki, S. Eguchi, and H. Yamada, Tetrahedron Letters, 1971, 99. 33

34 35

T. Sasaki, S. Eguchi, T. Ishii, and H. Yamada, J . Org. Chem., 1970,35,4273. T. Sasaki, S . Eguchi, and H. Yamada, J . Org. Chem., 1971,36, 1584. L. David and A. Kergomard, Tetrahedron, 1971, 27, 653.

c o/

10


Further investqghcw with palladium-myrcene complexes show that, in acetic acid, a mixture of acetates [linalyl (24), neryl (9), geranyl (lo), (25), and (26)] is formed,36 while a cyclic complex (27) is formed in methanol.37 Analogous complexes of ocimene have been studied, and found to yield geranyl methyl ether with methanol, and a dimer of limonene with acetone.37 An interesting method proposed for the removal of myrcene from a terpene mixture is selective clathration with nickel tetra-(4-methylpyridine)dithio~yanate.~

r+p

CH,OAc

HOAc + Pd or PdC1,

(17)

(9) + (10)

+

+

A

According to de Haan, confirmation of the stereochemistries of geraniol and nerol was obviously needed, and it is gratifying that his n.m.r. spectral measurements using tris(dipivalomethanato)europium(rrr) as a shift reagent support the commonly accepted assignments (nerol = cis, geraniol = trans).39 When geraniol and phenol react together in the presence of 85 phosphoric acid, there are formed, in addition to 0-and p-geranylphenols, the singly cyclized chroman (28) and the doubly cyclized hexahydroxanthene (29).40

Several simple p-alkylphenyl ethers of geraniol epoxide have been found to possess juvenile hormone activity. The p-ethylphenyl ether, tested on the yellow 3h

37 3n

39 40

K . Suga, S. Watanabe, and K. Hijikata, Chem. und Ind., 1971, 33. K. Dunne and F. J . McQuillin, J . Chem. SOC.(0, 1970, 2196. F. P. McCandles, Fluuour Znd., 1971, 2, 33. J. W. de Haan and L. J . M . van de Ven, Tetrahedron Letters, 1971,2703. S. Yamada, T. Katagiri, and J. Tanaka, Yuki Gosei Kagaku Kyokai Shi, 1971, 29, 81 (Chem. A h . , 1971,75, 6119).

Mono te rpenoids

11

mealworm (Tenebrio molitor) was 1000 times more active than the Cecropia hormone, although it seems that it acts on many other insects ~ o o . ~ ~ , ~ ~ Epoxidation of the appropriate geranyl ether and acid hydrolysis to the corresponding glycol affords the natural products marmin (30)42and severin

(30; R

=

(31 ; R

=

Q

o

j

a

CH,CH,NHCOPh)

(31)43;leading references to this type of substance are given by D r e ~ e r Citron.~~ ellyl acetate glycol (32) reacts with toluene-p-sulphonic acid in benzene ; the ketone (33) which is first produced, on further acid treatment, cyclizes to the cyclopentyl ketone (34).44 The reaction of geraniol with boron trifluoride etherate has been reported to give, after seven days, digeranyl ether, linalyl geranyl ether, various hydrocarbons, a-terpineol, and much unchanged gerani 0 1 . ~ The ~ complexing of linalyl acetate with Pd" has been examined.46

TsOH PhH ____)

(33)

(34)

Yet another method for the preparation of hydroxycitronellal (35) has been developed; it depends on the fact that the immonium salt (36) is hydrated by aqueous sulphuric acid, hydrolysis of the imine group taking place with sodium hydroxide.47 The effect of catalysts supported on silica gel on the well-known thermal conversion of citronella1 to isopulegol has been studied.48 The abstract 41

42

" 44

45

4b 47 48

F. M. Pallos, J. J. Menn, P. E. Letchworth, and J. B . Miaullis, Nature, 1971, 232, 486. R. M . Coates and L. S. Melvin, jun., Tetrahedron, 1970, 26, 5699. D. L. Dreyer, Tetrahedron, 1970, 26, 5745. L. Lizzani and R. Luft, Bull. SOC.chim. Frunce, 1971, 198. K. Nagai, Chem. and Phurm. Bull. (Japan), 1970, 18,2123. K. Dunne and F. J. McQuillin, J . Chem. SOC.(C), 1970,2200. M. Vilkas and G. Senechal, Ger. Offen. 2 045 888 (Chem. Ahs., 1971, 74, 125 873). T.-C. Chang, S. Washio, and H. Ueda, Agric. and B i d . Chem. (Jupan), 1970, 34, 1734.


12

CHo

l.MeZNH: ii, aq. H,SO,

of a Russian Patent for the preparation of citral (following a well-known route from isoprene)49 appears to be incorrect, so the novelty of the process cannot be assessed. In the presence of magnesium oxide, citral reacts with unsaturated ketones to give substituted acetophenones. The two geometric isomers of citral react differently, neral (cis-citral) giving mixtures of an acetophenone (37) and its dihydro-analogue (38), geranial (trans-citral) giving an isomeric acetophenone (39).5 0 a

jo"' -

PHO +

R

R 4y

O''

MgO, 250 "C. 25 mmHg

=

Ph, a-furyl, or a-thienyl

V. I. Artem'ev et al., U.S.S.R. P. 268 404 (Chem. Abs., 1970, 73, 87 432). N. Ronzani, Tetrahedron Letters, I97 1, 245 1.

Mono terpenoids

13

Solvolysis of (S)-2,6-dimethyloct-5-yl toluene-p-sulphonate gives a tetrahydrolinalool, (R)-2,6-dimethyloctan-6-01,with about 60 % retention of asymmetry. Kirmse and Arold have described several other similar reactions and suggest that a hydrophobic anchimeric interference of alkyl residues persists during the rearrangement, giving chirality to the carbonium ion.50b Analysis of Tugetes minuta. L. (Compositae) revealed the presence of cis- (40a) and trans- (40b) -0cimenones (= tagetenones), in addition to the previously known tagetones and dihydrotagetones. The new compounds were synthesized by treatment of a mixture of 3-methylbut-2-enoyl chloride and isoprene with a Lewis acid, e.g. SbCl, . 5 1

(404

Artemisyl, Santolinyl, Lavandulyl, and Chrysanthemyl Derivatives.-Probably the most significant work of the year in this group of monoterpenes is the evidence presented by Banthorpe and Charlwood for a chrysanthemyl intermediate in the biogenesis of artemisia ketone (41). The potassium salt of [2-'4C]-mevalonic acid was fed to cut stems of Artemisiu annua; there was low incorporation into the artemisia ketone isolated, but almost all the activity (ca. 90 %) was at the marked positions in formula (41), the remaining methyl groups carrying ca. 10% of the activity.52 This accords with the notion that the intermediate to artemisia and santolina monoterpenoids is chiral (i.e. formed from 3,3-dimethylallyl pyrophosphate + isopentenyl pyrophosphate). Banthorpe points that the 'sigmatropic' pathway (Vol. 1, p. 14) does not fit the facts. However, the chrysanthemyl intermediate (42) which he favours, if formed analogously to the postulated homologue in squalene biogenesis (Scheme 1),53does not fit the observed labelling either, since only 3,3-dimethylallylpyrophosphate is employed in its formation. Banthorpe is very cautious about drawing deductions too firmly, and there is a pressing need for further results. An interesting facet of the 'squalene' route is that there is no need for lavandulol (43) to be derived from a chrysanthemyl skeleton (Scheme 1).

5"b

52

53

W. Kirmse and H . Arold, Chem. Ber., 1971, 104, 1800. D. J. J. de Villiers, C. F. Garbers, and R. N. Laurie, Phytochemisrry, 1971, 10, 1359. D. V. Banthorpe and B. V. Charlwood, Nature New Biol., 1971, 231, 285. R. V. M. Campbell, L. Crombie, and G . Pattenden, Chem. Comm., 1971, 218; E. E. van Tamelen and M. A. Schwartz, J . Amer. Chem. Soc., 1971,93, 1780; L. J. Altman, R. C . Kowerski, and H . C. Rilling, ibid., p. 1782; H . C. Rilling, C . D. Poulter, W. W. Epstein, and B. Larsen, ibid., p. 1783; R. M. Coates and W. H. Robinson, ihid., p. 1785.


14

HOCH,/\f (43)

I

R

= Me

=

+/ *

cH2w R

(for monoterpenes)

A

0

(for squaiene)

(42) Scheme 1

In the laboratory, however, the route from prenyl ethers or thioethers to artemisia compounds and thence to the santolina skeleton54 is possible, and an interesting conversion to the lavandulyl skeleton of prenyl compounds has been de~cribed.~A mixture of 3-methylbut-2-enyl acetate (44) (or chloride) and 3-methylbut-2-enylthiol acetate (45) is treated with aluminium chloride or zinc chloride, when, in addition to the thioether (46), 19% of the lavandulylthiol acetate (47) and 604 of the corresponding isolavandulyl compound (48) are formed. An ingenious two-step synthesis of trans-chrysanthemyl alcohol (49) has also been achieved from prenyl alcohoi (50) and 3-methylbut-1-yn-3-yl

54

55

A . F. Thomas, Chimiu (Switz.), 1970,25,452;A. F. Thomas and W. Pavlak, Helu. Chtm. Acta, 197 1 , 54, 1822. K. Takabe, T . Katagiri, and J . Tanaka, Tc>truhedrotzLetters, 1971, 1503; see also J. Tanaka, T. Katagiri, and K . Takabe, Nippon Kuguku Zusshi, l968,89, 872; Haarman and Reimer, GmbH Belg. P. 615 962 (Chern. A h . , 1963, 58. I 1 223).

15

Monoterpenoids

chloride (51), presumably involving the dimethylallene carbene (52) as an intermediate.56 The photochemical trans-cis isomerizations of t-butyl trans-chrysanthemate (53) and -pyrethrate (54) have been reported. In hexane, there is 64% trans- and 36 % cis-chrysanthemate at equilibrium (with some racemization) only when a sensitizer (acetophenone) is present. The pyrethrate forms all four isomers upon unsensitized irradiation but the process is accelerated in the presence of a ~ensitizer.~’The addition of (I2(’)to chrysanthemic acids proceeds

I

R

(49)

Me) (54; R = C 0 2 H )

(53;R

=

differently from the reaction with the alcohols (see Vol. 1, p. 14). cis-Chrysanthemic acid reacted very slowly, and after 10 days the only product identified was the ketone (55). When the reaction was interrupted after 3 days, and the hydroperoxides were reduced with dimethyl sulphide, this furnished, in addition to the ketone (55) and the corresponding alcohols (56), the lactone (57) previously

P

O -

2

H

_____) %,($:;-sens.;

H & 2:

H OH (58)

56 57

R. W . Mills, R. D. H. Murray, and R. A. Raphael, Chem. Cornrn., 1971,555. K . Ueda and M. Matsui, Tetrahedron, 1971, 27, 2771.

16


obtained from lead tetra-acetate oxidation of the cis-acid. Photo-oxygenation of the trarzs-acid was more rapid, and after reduction the main products were the two diastereoisomeric trans-alcohols (58).58 Full details of Sucrow’s synthesis of yomogi alcohol have now been published.s9 Although pyrolysis of chrysanthemyl oxalate gives only artemisiatriene (59), deamination of chrysanthemylamine (60) with isoamyl nitrite in acetic acid furnishes two other products, crysanthemyl acetate (6 l) and artemisia acetate (62). The selectivering cleavage between C(1)and C(3)suggests that isobutenyl is better able to stabilize a carbonium ion than is gern-dimethyI.60“The thermoiysis of pyrethric acid [the diacid corresponding to (54)] is important in biogenetic studies, since this has been used for the degradation of labelled compounds. The reaction has been studied by Crombie et al., who found that 65 % of the total acidic products were isomers of 2,6-dimethylheptadienoic acids, resulting from ring cleavage and decarboxylation, the carbon dioxide loss being from the carboxy-group attached to the cyclopropane ring.60h

4 Monocyclic Monoterpenoids Cyclopentanes, including 1ridoids.-A summary of the botanical distribution, structure, and properties of the iridoids and seco-iridoids has been compiled, including about 80 natural products.60‘ The full paper concerning galiridoside, isolated from Galeopsis tetrahit (see Vol. 1, p . IS), has appeared.61 A thorough study of ca. 40 species of various Valerianaceae has shown that the valepotriates, the main component of which is T. Sasaki, S. Eguchi, and M . Ohno, Synth. Comm., 1971, 1, 75. W. Sucrow and W. Richter, Chem. Ber., 1970, 103, 3771. ‘ O a T. Sasaki, S. Eguchi, M. Ohno, and T. Umemura, J . Org. Chern., 1971, 36, 1968. ‘ O h L. Crombie, C. F. Doherty, G . Pattenden, and D. K . Woods, J . Chem. SOC.(C), 1971, 2739. V. Plouvier and J . Favre-Bonvin, Phytochemistry, 1971, 10, 1697. ‘’‘ O C 0. Sticher, Helu. Chim. Acta, 1970, 53, 2010. ”

”

Monoterpenoids

17

valtrate (63), are characteristic for the tribe Valerianeae.62 Surfher work on the structure of loganin has established the absolute stereochergistry by chemical transformation of asperuloside (64), of known absolute structure, into loganin penta-acetate (65).63 Derived from secologanin (66) are another group of mono-

GluAc,

=

P-glucose tetra-acetate

(44)

(45)

terpenoid glucosides, the seco-iridoids, those from Swertia japonica being related to sweroside (67; R2 = p-glucosyl) or to swertiamarin (68; R2 = p-glucosyl) (cf: ref. 64). The most complex glucosides of this type are amarogentin [67; R2 = (69)] and amaroswerin [68; R2 = (69)].65 Compounds of this nature OHC

*H...L

COzMe

iOYO

-0,

HOC

(67; R' = H) (68; R' = O H )

(49)

illustrate the progressive blurring of the classical divisions of organic chemistry. Should, for example, the biogenesis of the Ipecac alkaloids from secologanin (66) be considered as the formation of an alkaloid66 or of a monoterpenoid? Most alkaloids are derived more or less directly from terpenes, but alangoside (70) is a monoterpenoid l a ~ t a mand ,~~ the relation of strictosidine (71)68ato secologanin (66) is evident. Other alkaloids are known which derive even more closely from "

E. Stahl and W. Schild, Phytochemistry, 1971, 10;147. H. Inouye, T. Yoshida, S. Tobita, and M . Okigawa, Tetrahedron, 1970, 26, 3905. h 4 H. Inouye, S. Ueda, and Y . Nakamura, Chem. and Pharm. Bull. (Japun), 1970, 18, 1856. 6 5 H. Inouye and Y. Nakamura, Tetrahedron, 1971,27, 195 I . '"A. R. Battersby and R. J . Parry, Chem. Comm., 1971,901. 6 7 R. S. Kapil, A. Shoeb, S. P. Popli, A . R. Burnett, G. D. Knowles, and A. R. Battersby, Chem. Comm., 1971,904. 6 8 a K . T. D. De Silva, G. N. Smith, and K. E. H . Warren, Chem. Comm., 1971,905; K . T. D. De Silva, D. King, and G . N. Smith, Chem. Comm., 1971,908.

63

18

(70: R ' = Me, R 2 = H, or R' = H, R 2


=

Me)

(71)

loganin, and some arise as artefacts generated from it in the course of extraction ( 310nm or temperature of - 1 5 "C; iii, HFSO,, -78 "C; iv, K,CO,-MeOH.

Scheme 5

5 Bicyclic Monoterpenoids Bicyclo [3,1,O 1hexanes.-Sabinene is the starting material for a synthesis of nootkatone (see Chapter 2, p. 109). It is first converted to sabinaketone (193), ~~ has given a very which is methylated to the isomer (194) of t h ~ j 0 n e . IDoering 142 143

K. E. Hine and R. F. Childs, J. Amer. Chem. SOC.,1971, 93, 2323. F. Rijkens, H. Boelens, H. G. Haring, and A. van der Gen, A.C.S. Meeting, Los Angeles, 1971, Abstracts.

37

Monoterpenoids

detailed account of his work on the racemization of thuj-3-ene (195), using labelled material. The concerted and symmetry-allowed path is shown not to

A

A

A

(193)

(194)

(195)

be followed, and the mechanism is the combination of two processes involving opening of the cyclopropane ring to give an intermediate consisting of two independent radicals. In the favoured process (Scheme 6), the original conformation is preserved ;the competing process requires conformational inversion (at a cost of 1.5 kcal mol-’) of the intermediate. The barrier to inversion is sufficient to explain the high steresselectivity of this non-concerted rearrangement.’44 Similar systems are discussed by Swenton and Wexler, who also oppose the suggestion of a concerted reaction and favour a biradical type.’45

Scheme 6

In the oil of the Western Red Cedar (Thuja plicata Donn) there is a mixture of 5-10% (+)-thuj-3-one (196) and 70-80% (-)-isothuj-3-one (197).* If it is

W. von E. Doering and E. L. G. Schmic t, Tetrahedron, 1971, 27, 2005. J. S. Swenton and A. Wexler, J. Amer. Chem. SOC.,1971, 93, 3066. 1 4 6 S. P. Achakya, H. C. Brown, A. Suzuki, S. Nozawa, and M. Itoh, J. Org. Chem., 1969,34, 3015. * The nomenclature suggested by Brown et ~ 1 . ’ ~is’ employed here, as it is in ref. 147, namely that the prefix “iso” is reserved (as in the menthone and carvomenthone cases) for methyl and isopropyl groups cis to each other. Unfortunately, ref. 148 does not follow this very reasonable and well-established precedent. 144

145

38


required to raise the ( + bthujone content, epimerization with base gives a 65 : 35 mixture (196):(197),and (+)-3-thujone can be removed as the crystalline bisulphite c~mpound.’~’On irradiation with 305 nm light, both isomers [(196) and (197)] give the same ratio of the decarbonylated and ring-opened olefins [(198): (199) = 87 : 131, some photochemical isomerization of the ketones occurring beforehand.14* Bicyclo[2,2,1 Iheptanes. -The crystal structure of (+)-3-bromocamphor has been redetermined (a preliminary publication appeared some time ago), making definite the absolute configuration (200) of this and related m o n ~ t e r p e n o i d s . ’ ~ ~ The c.d. and U.V. spectra for various substituted camphors have clarified the chiroptic properties of a-ketols and related substances (see Vol. 1, p. 273). ‘Axial’ (i.e. endo-) 3-hydroxy- and 3-acetoxy-derivatives of ( + )-camphor, and 2-hydroxyor acetoxy-derivatives of ( - )-epicamphor (201), produce an ‘anti-octant’ contribution,’” and although 3-endo-aminobornan-2-one (202) exhibits ‘octant’ behaviour, increasing alkylation of the nitrogen atom produces increasing conformational restraint, resulting in incorrect orientation of the lone pair for coupling and hence an increasing ‘anti-octant’ trend.’” The c.d. spectra of anti-n-substituted camphors (203) show that there is an interaction between the substituent and the carbonyl group.lS2 The n.m.r. spectra of the epimers of 2-acetoxy-3-endo-dimethylaminobornane and 3-acetoxy-2-endo-dimethylaminobornane and their quaternized derivatives (acetylcholine analogues) have been d i s c ~ s s e d , ’and ~ ~ all the 13C signals of isoborneol have been identified using alternately pulsed n.m.r. and lanthanide-induced chemical shifts.154

A synthesis of chiral ( -)-(7R)-8-deuteriobornadiene (204) follows the route shown in Scheme 7. The overall yield of (+)-9-deuteriobornylene (205) from (+)-camphor via the known (+)-9-bromocamphor (206) was 35%, and the 147

148 149 1so

151 152

153

1s4

V. Hach, R. W. Lockhart, E. C. McDonald, and D. M. Cartlidge, Canad. J. Chem., 1971, 49, 1762. R. S. Cooke and G. D. Lyon, J. Amer. Chem. Soc., 1971, 93, 3840. F. H . Allen and D. Rogers, J. Chem. Sac. ( B ) , 1971, 632. L. Bartlett, D. N . Kirk, W. Klyne, S. R. Wallis, H. Erdtman, and S. Thoren, J . Chem. SOL..( C ) , 1970, 2678. A. H. Beckett, A . Q . Khokhar, G . P. Powell, and J. Hudec, Chem. Comm., 1971, 326. M. T. Hughes and J. Hudec, Chem. Com m . , 1971, 805. T. Ahmad, M . N . Anwar, M . Martin-Smith, R . T. Parfitt, and G. A . Smail, J. Chem. SOL..( C ) , 1971, 847. 0. A . Gansow, M. R. Wilcott, and R. E. Lenkinski, J. Amer. Chem. Soc., 1971, 93, 4295.

Monoterpenoids

39

BrCH2Y

DCH,

x -DCH

DCH,

+ CO2Et

EtO,C

(20

I

yield)

Reagents: i, HOCH,CH,OH-Ht; ii, Na-MeOD; iii, H , O + ; iv, TsNHNH,; v, B u L i ; vi, A ; vii, HC-C -CO,Et; viii, KOH-H,O; ix, copper chromite-quinoline.

Scheme 7

optical purity at this stage was 98 %. Pyrolysis of the chiral bornylene (207) gave only 20 % of the possible yield of the desired product, and the isotope distribution at the end of the series was 72 : 14 : 14 in, respectively, the methyl groups C(8), C(9), and C(10).1s5Specifically deuteriated camphor can be made by treating 3-bromocamphor with zinc and [carbo~y-~HIacetic acid. An inversion occurs, and exo-deuteriated camphor is obtained (88 % 2Hl)from 3-endo-bromocamphor, the exo-bromocamphor yielding 3-endo-deuteriated camphor. 5 6 The preparation of camphor labelled in any of the methyl groups with 14C has been described. l S The conversion of pinene to camphene (and tricyclene) is a large-scale commercial process. A recent series of papers describes a three-phase system comprising a vertical column in which a liquid suspension of a titanium catalyst in camphene flows downwards, and pinene vapour passes upwards : a mathematical 155 156 157

M. R. Wilcott, tert., and C. J. Boriack, J . Amer. Chern. Soc., 1971, 93, 2354. R. R. Sauers and C. K. Hu, J. Org. Chern., 1971, 36, 1153. 0. R. Rodig and R. J. Sysko, J . Org. Chern., 1971,36, 2324.

40


analysis of the underlying kinetic theory is i n ~ 1 u d e d . lA~ ~simple and probably general approach to the bicyclo[2,2,l]heptane system is illustrated by cyclization of 3-(~-toluene-p-sulphonyloxyethyl)cyclopentanone (208) to norcamphor (209). In acetic acid the yield is only 3%, but if urea is added it rises to 74%.lS9 A

-&

-

LOTS

(208)

(20%

synthesis of fenchone (210) based on this approach is illustrated in Scheme 8 and involves the preparation of a cyclopentanone derivative (211). To obtain

I

CHO

Reagents: i, HBr; ii, AgOAc-HOAc; i i i , LiAIH,; iv, M e C 0 , H ; v, BF,--Et,O; vi, SOCl, n.. vii, EtO ; viii, Arndt-Eistert procedure.

Scheme 8 158

lS9

V. A. Vyrodov, E. V . Afanas'eva, and b. Ya. Korotov, Nauch. Tr .,Leningrad. Lesotekh. Akad., 1970, No. 135 (part 11, p. 3, and subsequent papers. (Chem. Abs., 1971, 74, 112 076-1 42 079) J. L. Marshall, Tetrahedron Letters, 1971, 753.

Monoterpenoids

41

the racemate, the cyclopentanonecarboxylic acid (212) was homologated by the Arndt-Eistert procedure, but for preparation of the optically active ( + )-fenchone, the cyclopentanone (211)was synthesized from a bromo-isofenchol(213) obtained from the reaction of (+)-a-pinene epoxide (214) or (-)-trans-pinocarveol (215) with hydrogen bromide. Treatment of the chloride (211)with ethoxide ion gave (+)-fenchone (210). (f)-Fenchone was similarly obtained from (212) via racemic (211).160 The ‘traditional’ way of making bicyclo[2,2,l]heptanes via a DielsAlder reaction was used in a synthesis of teresantalol (216). A departure from the usual routine was the use of the allene (217) as dienophile. The required endo-acid (218)’ separated from the products by iodolactone formation and regeneration with zinc, was treated with 97% formic acid to give the tricyclene lactone (219). The remainder of the route is shown in Scheme 9.16’ The exo-acid

li

Reagents: i, H C 0 , H ; ii, PhSNa; iii, esterify; iv, LiAIH,; v, Ni-H,; vi, Br,-CH,Cl, sodium salt.

on

Scheme 9

(220) obtained in the first stage of this synthesis does not form an iodolactone, but it reacts (as the salt) with bromine in methylene chloride to give the unusual bromo-P-lactone (221).1 6 2 I6O

162

P. H. Boyle, W. Cocker, D. H. Grayson, and P. V. R. Shannon, Chem. Comm., 1971, 395; J. Chem. SOC.(C), 1971, 2136. W. E. Barnett and J. C. McKenna, Tetrahedron Letters, 1971, 227. W. E. Barnett and J. C. McKenna, Chem. Cornm., 1971, 551.

42


A novel synthesis of substituted fenchanes has been effected from the readily available tricyclic ether (222). With acetic anhydride in boron trifluoride etherate, this furnished a 90 "/d yield of 9-acetoxyfenchyl acetate (223). Acetyl toluene-psulphonate in acetonitrile gave 9-acetoxyfenchyl toluene-p-sulphonate.'63"

(222)

OAc (223)

The configurations o f the exo- and endo-fenchane hydrates (2,7,7-trimethylnorbornan-2-01s) conventionally prepared, have been confirmed. 6 3 b Nickon et ul. have found the apparent first-order rate constant for the isomerization of 1-hydroxycamphenilone (224) to 1-hydroxy-apocamphor (225) in aqueous solution buffered at pH 10 to be 5.0 x l O P 4 s p ' , that of the reverse reaction being 2.5 x 10-4s-'. They used three routes to make the desired material, two of which are shown in Scheme 10; the third route also starts from the nitrocamphene (226), but involves the use of base in the last stage, and so necessitates the separation of the two isomers (224) and (225),Ih4

(226)

(224)

(225)

Reagents: i, L i ; ii, C O , ; iii, KMnO,-NaIO,; i v , Curtius degradation; v, N a N O , - H + ; vi, O H ; vii, O 3; viii, H ,-Pd/C. Scheme 10 ~

"'"N. Bosworth a n d P. D. Magnus, Chem. Cornnz., 1971, 618. ' b 3 b L Pirila, . A n n . Acad. Sci. Fennicae, Ser. A2, 1971, No. 157, 52. l h 4 A . Nickon, T. Nishida, J. F r a n k , a n d R. Muneyuki, J . O r g . Chem., 1971, 36, 1075.

Monoterpenoids

43

A number of ring-opening reactions of bicyclo[2,2,l]heptane terpenoids have been described. The green oil obtained when camphor nitrimine (227) [from the action of sodium nitrite on camphoroxime (228)]is irradiated at 253.7 nm under nitrogen has been shown to consist of a complex mixture of substances ( 2 2 8 t (233),four of which [(230H232) and camphor] are also obtained on thermolysis of the nitrimine. 6 s Detosylation of 1,3,3-trimethyl-2-e~1zdo-hydroxy-6-toluene-

(227)

(229; R (230; R

= =

CHO) CN)

(231)

p-sulphonyloxynorbornane (234) (a derivative of fenchol) by a base results in a single product (235).' 6 6 Solvolysis of fenchyl toluene-p-sulphonate (236) in acetic acid has been examined under various conditions ; the main products are exo-isofenchyl acetate (237) and exo-4-methylsantenyl acetate (238) via a series of Nametkin and Wagner-Meerwein rearrangement^.'^^ Acetolysis of exo- and endo-camphenilol toluene-p-sulphonates (239) to apocycline (240) has also been examined.' h 8 Solvolysis of 2-endo-phenyl-2-exo,3-exo-dihydroxybornane h

'65 166 167

16'

I

L. J. Winters, J. F. Fischer, and E. R . Ryan, Tetrahedron Letters, 1971, 129. Z . Chabudzinski, U . Lipnicka, and Z . Rykowski, Roczniki Chem., 1970, 44, 2181 A. Coulombeau and A . Rassat, Bull. Soc. chim. France, 1970, 4389. W. Hiickel and H. Eben, Suomen Kern., 1971,44, B, 61.


44

(239)

(240)

3-toluene-p-sulphonate (241)leads to the exo-epoxide (242),and similar treatment of the endo,endo-dihydroxy-compound leads to the endo-epoxide (243). Acid treatment of the epoxide leads, in the case of the endo-epoxide, to the bicyclic ketone (244), but the exo-epoxide (242) gives a ring-opened aldehyde (245). 6 9

The deamination of the bornylamines by various methods has been examined, including the use of nitrous acid in acetic acid.17' Whereas ring-opened compounds are obtained to some extent from endo-bornylamine, practically none are obtained from the exo-isomer. This work should be taken in conjunction with an earlier paper dealing with the same reaction in water,'71 conditions which lead to less camphene and more hydroxylated compounds with both endo- and exo-bornylamines. It is claimed that the camphor obtained in the deamination with nitrous acid is formed via a 2,3of 2-exo-hydroxy-3-exo-aminobornane endo-endo hydrogen shift. 7 2 The ring-opening oxidation with nitric acid of 3,n-dibromocamphor (246) has been re-examined by Brienne and Jacques, who have shown that treatment of the n-bromocamphoric acid (247 ; R = H) thus obtained with one equivalent of alkali gives the unusual trans-fused y-lactone [(248 ; R = H), the trans-n-camphanic acid of the 19th century]. The structure already derives from the fact that J . M . Coxon, M. P. Hartshorne, and A. J. Lewis, Chem. C o m m . , 1970, 1607; Austral. J . Chem., 1971, 24, 1009. 1 7 0 D. V. Banthorpe, D. G. Morris, and C. A. Bunton, J . Chem. SOC.(B), 1971, 687. 17' W. Hiickel and H.-J. Kern, Annalen, 1969, 728, 49. "' P. Wilder,jun., and W.-C. Hsieh, J. Org. Chem., 1971, 36, 2552. lh9

Monoterpenoidr

45

the monoester (247 ; R = Me) yields the ester of the lactone (248 ; R = Me) with base, but the authors give much other evidence to support the structure. 1 7 3

(246)

(247)

(248)

Nitration of bornane gives a 53 % yield of 3-nitrobornane, with an exo : endo ratio of 55 : 45, and some 2-nitrobornane. When equilibrated with alkali, the 2-nitro-compound consists of 95 % endo- and 5 % exo-isomer, but 3-nitrobornane contains 52 % endo- and 48 % exa-isomer, the figures illustrating the disadvantage of eclipsing the substituent with the 1-methyl group.'74 2-Nitrobornane reacts with 20 % potassium hydroxide in glycerol or diethylene glycol at 20&220 "C to give up to 60 % of bornane and ammonia. Small amounts of camphor and its oxime are also obtained but not amines, and the mechanism of the reaction is said to be as shown in Scheme 11.175Epiborneol (endo-hydroxy-group) and epi-

solvent +

i ! C-0-K'

\I

/

-+

\ /

C=O

Scheme 11

isoborneol (exo-hydroxy-group) react differently in oxidation with nitric acid, the endo-compound yielding epicamphor while the exo-compound is esterified to the nitrate ester.176 Metal-ammonia reduction of bicyclo[2,2,l]heptanones has been examined by Coulombeau and Rassat,' 7 7 who point out that the conformationally more stable forms of substituted alcohols are the exo-isomers in the case of norbornan2-01 and the 1-methyl homologue, but the endo-isomers in the case of borneol (1,7,7-trimethyl) and fenchol (1,3,3-trimethyl).'7 8 The same authors have also examined the reduction of 'camphorquinone' [2,3-dioxo-bornane,* (249)] with J.-M. Brienne and J. Jacques, Tetrahedron, 1970, 26, 5087. H. Toivonen, Suomen Kem., 1971,44, B , 54. H. Toivonen, S. A . Laurema, and P. J. Ilvonen, Tetrahedron Letters, 1971, 3203. 17' E. Heinanen, Suomen Kem., 1971, 44, B , 114. 1 7 7 A. Coulombeau and A . Rassat, Bull. SOC.chim. France, 1971, 4399, and subsequent papers. 1 7 * A. Coulombeau and A. Rassat, Bull. Soc. chim. France, 1971, 4393. * Chemical Abstracts includes the older name camphane' for this paper. 173 174

46


zinc and acetic acid, when the two products are 2-endo-hydroxy-3-oxobornane (250) (60 7J and 3-erzdo-hydroxy-2-oxobornane (251) (40 %). Equilibration of the products in base gives all four of the possible isomers in the proportions shown in Scheme 12. Lithium aluminium hydride reduction gives only exo-hydroxygroups.'79 Lithium aluminium hydride reduction of oximes in this series also

-I-----

111

A00-

.c.

h

I

10

:

3

1

\

:

62

:

24

Reagents: i, Zn-HOAc; ii, base; iii, LiAlH,.

Scheme 12

leads only to exo-amines, while reduction with sodium in alcohol gives endoamines. 8 o 'Camphorquinone' (249)is responsible for the yellow colour of freshly prepared 3-hydroxymethylenecamphor ; it is formed by oxidation,' and can be made From the hydroxymethylene compound by treatment with a diamine, the resulting bis-aminomethylenecamphor being even more readily oxidized in air.'82 In'

"'

''I

A. Coulombeau and A . Rassat, Bull. SOC.chint. France, 1971, 505. A. Daniel and A . A . Pavia, B i d . Soc. chim. Frunce, 1971, 1060. N . B. Kupletskaya, G. V. Panova, Le Quang Lien, a n d A . P. Terent'ev, Zhur.fiz. Khim., 197 1. 45, 707. A . P. Terent'ev, G. V . Panova, and 0. V . Toptygina, Zhur obshchei Khirn., 1971, 41, 476.

Moizoterpenoids

47

The ionic addition of hydrogen bromide to born-2-ene in ether at -10°C leads to a mixture containing 50 % of epi-isobornyl bromide (252) and 50 % of the camphyl bromides (253).’ 83

(252)

(253)

Chromatography (especially on silica gel) of the aryl alcohols derived from Grignard reactions on camphor and other bicyclic ketones can give products formed by bond migration (254) or by dehydration (255), and since the reactions are stereoscopic and selective, Ramage et al. have suggested that this is a good method for effecting such steps, since the complex mixtures formed by normal acid-catalysed rearrangements arc avoided. 84 The endo-hydroxylated compounds are more stable, and corresponding products from isofenchone or isocamphone [e.g. (256) or (257)] are not so easily dehydrated.lss Coxon et al. have found that epoxidation of 2-phenylborn-2-ene (255) using m-chloroperbenzoic acid leads to 41 of the ketone (258) and 1 of the epoxide (259) of the hydrocarbon (254). Although rearrangement of the phenylbornene before epoxidation is only one of the possibilities considered by Coxon et al., it seems, in view of Ramage’s work, to be the more likely. Other products of this reaction are also discussed. In the same paper, the reactions of 2-phenylbornene with diborane and peroxide (leading to hydroxylation in the 3-position), and with N-bromosuccinimide, leading mainly to the rearranged bromo-olefin (260), are described. 86 Reaction of the hydrocarbons in this series with phenols or phenol ethers in the presence of suitable catalysts is well known, a-fenchene and phenol yielding a mixture of 0- and p-isofenchylphenols in the presence of aluminium phenoxide ; 87 practically the same reaction was described seven years ago. Reaction of camphene with a phenol ester in the presence of stannic chloride is reported to give a bornyl but terpenoid rearrangements are notorious under such conditions, and it is doubtful whether this is the only product. 3,3-Dibromocamphor (261) has been converted into substituted cyclohexanecarboxylic acids by the route shown in Scheme 13.190 L. Borowiecki, J. Glowinska, and W. Zacharewiw, Roczniki Chem., 1970, 44, 531. C. R. Hughes, D. F. MacSweeney, and R . Ramage, Trtruhedron, 1971, 27, 2247. l X 5 A. F. Thomas, unpublished work. l X 6 J. M. Coxon, M. P. Hartshorn, and A. J. Lewis, Austral. J . Chem., 1971, 24, 1017. T. F. Gavrilova, I. S. Aul’chenko, and L. A . Kheifits, Z h r . org. Khirn., 1971, 7, 94. 1 8 8 E. Demole, H e h . Chim. Acta, 1964, 47, 1766. l X 9 J. Mardiguian and P. Fournier, Ger. Offen. 2 032 170 (C/w.m.ADS., 1971, 74, 88 171). A. J. Reuvers, J. A. Jongejan, J . Klomp, and H. Van Bekkum, Org. Prcp. Proc. Int., 1971, 3, 83. lX3

IX4

48


(258)

(259)

m-chloroper benzoic acid

Br

Br Br

Reagents: i, AgNO,; i i , H,-PdlC

Scheme 13

Bicyclo [3,1,1]heptanes.-The discussion of the conformation of substituted pinanes has led to general agreement that the isonopinone structure is that of a flattened boat (262). This conclusion was reached by Baretta, Jefford, and

Monoterpenoids

49

Waegell by considering the n.m.r. spectra of the ketones and their reduction product^,^^'^'^^ and by Bessiere-Chretien and Grison also from n.m.r. studies and from calculations, based on models, which simplified the problem by supposing the cyclobutane ring to be ~ymmetrica1.l~~ Unfortunately, the two groups have used opposite conventions, and in this Report the Bessiere-Chretien system is followed; the shape of the cyclohexane refers to the ring carrying the gemdimethyl groups (as was previously employed by Abraham et ~ 1 . l ~A~ useful ). discussion of the strains involved in ring-inversions of these and other bridged cyclohexanes has been given by Jefford and Burger.19' There is a discrepancy in the conformation found in nopinone (263), and it would be desirable to have this clarified because it is used by Bessiere-Chretien as a model for the P-pinene conformation as a first a p p r o ~ i m a t i o n . ' ~Abraham ~ et ul. described it as a flattened chair (263), an attribution quoted, apparently with approval, at the beginning of the second paper by Baretta et ~ l . " ~At the end, owing to a misprint,lg7it is suggested that the opposite (boat) conformation would be closer to the truth. Bessiere-Chrittien has recently repeated her belief that nopinone is in the quasi-chair conformation, pointing out that this would favour approach of borohydride from the side of the molecule away from the gem-dimethyl groups to give cis-nopinol (203),Ig6as, indeed, Baretta et al. also point out in the early part of their paper. 9 2 Bessikre-Chretien and Meklati have explained some interesting

(262)

(263)

(264)

differences between the reactions of P-pinene and 'orthodene' (265) derivatives in terms of conformations proposed. The latter, for instance, is hydroborated to a mixture of cis- and trans- primary alcohols (266) while P-pinene gives almost exclusively the cis-alcohol (267).19' CH,OH

w - uCH2* 6b i. BH, : ii, H,O

19* 193 194

195

19'

A. J. Baretta, C. W. Jefford, and B. Waegell, Bull. SOC.chim. France, 1970, 3899. A. J. Baretta, C. W. Jefford, and B. Waegell, Bull SOC.chim. France, 1970, 3985. Y. Bessiere-Chretien and C. Grison, Bull. SOC.chim. France, 1971, 1454. R. J. Abraham, F. H . Bottom, M. A. Cooper, J. R. Salmon, and D. Whittaker, Org. Magn. Resonance, 1969, 1, 51. C. W. Jefford and U. Burger, Chimia (Switz.),1970, 24, 385. Y. Bessiere-Chretien and B. Meklati, Bull. SOC.chim. France, 1971, 2591. C. W. Jefford, personal communication.

Trrperioids and Steroids

50

Carbene addition to the pinene system gives a tricyclic compound (268)(whose conformation has also been discussed by Vereshchagin et ul.' 98), and Hatem and Waegell have shown how electrocyclic ring-opening occurs to give the two bicyclo[4,l,l]octadienes (269)and (270).'99 The bicyclo[4.l,l]octane ring system has also been synthesized by Joulain and Rouessac, following a similar path to that used by Hatem and Waegell; their note described a number of other unsuccessful routes. O 0 Acetylation of pinane (77) (90 '/" cis) with acetyl chloride and aluminium chloride under Friedel-Crafts conditions leads to the same mixture of l-chloro2-acetylmenthanes (27 1) as is obtained in a similar reaction using menth-1-ene (103), through which the reaction is assumed to pass. The chloroketones are unstable, and yield unsaturated ketones under a variety o f conditions* (Scheme

14)." '

Conversion of (-)-b-pinene to (-)-a-pinene is said to occur in better chemical yield and good optical yield when refluxed with 15mole benzoic acid for 48 h.2"2 The mechanism of the rearrangement of the pinenes with acid has been discussed again by Williams and Whittaker, who postulate the existence of two interconvertible carbonium ions (272) and (273), which can either revert to CIpinene or react further, the one (272) leading to the fenchane skeleton and the other to the bornane skeleton, camphene, or ring-opened compounds. Using ;(:

I"'

'"" 200 '01 '("

A . N . Vereshchagin, S. G . Vul'fson, N. 1. Gubkina, and B. A . Arbuzov, Izvest. Akad. h'uuk S.S.S. R . , Ser. K h i m . , 1970,2467. J . Hatem and B. Waegell, Tetralzedrorz Letters, 1971, 2069. D. Joulain and F. Rouessac, Conipt. rend., 1971, 273, C , 561. R. F. Tavares, J. Dorsky, and W . M . Easter, J . Org. Chew!., 1971, 36, 2434. R . L. Settine, J. Org. Clirm., 1970, 35, 4266.

* Only the two isomers shown [(271a) and (271b)] are mentioned in this paper, justification for only axial chlorine being given, but the products from the HCI eliminations are not easily understood.

Monoterpenoids

51

J

r

h

Reagents: i, AlC1,; ii, AcCl-AlCI, .

Scheme 14

acetic acid as solvent, olefins are formed rather than acetates.203 The rearrangement of a-pinene in a mixture of deuteriated acetic acid and acetic anhydride has been shown by Tori et al. to result in a single bornyl acetate with the label in the 6-endo-position (274), while the isobornyl acetate (275) has the label distributed among four positions in the ratio C(8):C(10): C(9):C(6) = 37 : 29 : 24 : To account for this, further olefinic intermediates appear to be necessary (since deuterium is not incorporated directly into a carbonium ion*), in addition to those proposed by Whittaker.

(273) 203

C. M. Williams and D. Whittaker, J . Chem. Soc. ( B ) , 1971, 668, 672. * 0 4 K . Tori, Y . Yoshimura, and R. Muneyuki, Tetruhedron Letters, 1971, 333. * Leading refs. are given in ref, 205.


52 9

8

The reaction of a-pinene with t-butyl perbenzoate has been re-examined by Lalande206with addition of various catalysts, and the yields of pinocarveols and myrtenol are somewhat different from those found in the earlier Lalande has also published the full account of the work begun in 1966 on the ring-opening of b-pinene with aliphatic nitriles, leading to 7-substituted menth-1-ene derivatives.208 A useful reaction in this respect is that of the pinenes with dry hydrogen chloride, leading, in the case of a-pinene, to 6-bromomenth-l-ene, and in the case of P-pinene to 7-brom0rnenth-l-ene.~~~ In the presence of di-t-butyl peroxide as initiator, the pinenes react with thiols, to yield p-menth-1-ene 2-sulphides from a-pinene and 7-sulphides from P-pinene.210 (cf: Vol. 1, p. 43.)

New reactions of the pinane ring oxygenated at C(2) include much work by Coxon, Garland, and Hartshorn, who have investigated the pyrolysis of nopinol and nopinone enol acetate. Nopinol (264) yields (among other normal cleavage products) 10 2-methyIocta-2,7-dien-4-01 (276) and 6 % of 4,6-dimethylhept5-enal(277). The mechanism proposed (Scheme 15) was supported by deuterium labelling studies.2l 1 Thermolysis of nopinone enol acetate (278) at 465 "C gives a mixture containing much 2-methylocta-2,4,6-trien-6-y1acetate (279), but at higher temperatures this reacts further in the Cope manner to yield 1,1,2-trimethylcyclohexa-2,4-dien-3-yl acetate (280), together with (28 l),(282), and other

'" A. 207 208 209 *lo

111

F. Thomas, 'Deuterium Labeling in Organic Chemistry', Appleton-CenturyCrofts, New York, 1971, Ch. 3. R. Lalande and J.-J. Villenave, Compt. rend., 1971, 272, C , 1825. H. Heikman, P. Baekstrom, and K . Torssell, Acta Chem. Scand., 1968, 22, 2034. M. Cazaux and R. Lalande, Bull. Soc. chim. France, 1971,461. A . Gaiffe and J . Castanet, Compt. rend., 1971, 272, C, 96. A. Gaiffe and J . Castanet, Compt. rend., 1970, 271, C, 1012. J. M . Coxon, R. P. Garland, and M. P. Hartshorn, Chem. Comm., 1970, 1709 (Chem. Abs. 1971, 74, 142 073, summarizing this work, is incorrect).

Monoterpenoids

53

products.2' Further pyrolyses reported are of cis- and trans-pinocarveol ; the products from the cis-isomer (283) are given in Scheme 16." Pyrolysis of the trans-acetate gives more than 30 products, some of which are presumed to arise through verbenene.2l 3

(284)

Scheme 16

The methanolysis of pinanyl p-nitrobenzoate (285) gives a mixture including both cis- and trans- products (286); the results are contrasted with those from the nopinol system, and a unified mechanism is proposed.214 The preparation and properties of a-ethylnopinone (287) and a-ethylisonopinone (288) have been described.2

A full paper describing the synthesis and stereochemistry of the four pinane2,3-diols, summarizing all the earlier work, has appeared.'16 A reagent for the asymmetric reduction of ketones has been prepared from lithium aluminium hydride and two of these diols, (289a) and (289b).217 J. M. Coxon, R. P. Garland, and M. P. Hartshorn, Austral. J . Chem. 1970, 23, 2531. J. M. Coxon, R. P. Garland, and M. P. Hartshorn, Austral. J . Chem., 1971, 24, 1481. 2 1 4 J. R. Salmon and D. Whittaker, J . Chem. Suc. ( B ) , 1971, 1249. 2 1 5 Y. Bessiere-Chretien and M. M. ElGaied, Bull. SOC.chim. France, 1971, 2189. 2 1 6 R. G. Carlson and J. K. Pierce, J . Org. Chem., 1971, 36, 2319. 2 1 7 H.-J. Schneider and R. Haller, Annalen, 1971, 743, 187. * The diene structure (284) is misprinted in this paper. 212

213


54

6HoH go" (289b)

(2894

The evidence for the base-catalysed rearrangement of the two epoxypinan-3-01s (290) to the unsaturated ketone (291) riu the carbanion (292) has been published.2 '' The ring contraction with base of cis-pinane-cis-2,3-diol 3-tosylate (294) Lgiving 50":, of (293)], reported in 1968 by Suga, has been described in detail ; * I u the metal hydride reduction of the same compound (294) also gives some ringcontracted product.**' 2-Hydroxypinocamphone (295)is known to give campholenic acid with acid. If oxalic acid is used and the reaction time prolonged, the main product is the cyclopentane 7-lactone (296).22*

A cyclopentane aldehyde (297) is obtained when verbenone epoxide (298) is treated with zinc bromide. The presence of pinene in the products is difficult to explain, and the difference in products oblained with aluminium chloride (Vol. 1, p. 45) is remarkable.z22 When the toluene-p-sulphonylhydrazoneof the epoxide (298) is treated with potassium t-butoxide, both isomers of the cyclobutylacetylene (299) are obtained2**in an Eschenmoser fragmentation.223 Another synthesis of nopadiene (300) by a conventional route from a-pinene has been described.22s 'Ix

'I4

220

221 ?22

*13

'''

J . M. Coxon, E. Dansted, R. P. Garland, M . P. Hartshorn. and W . B. Joss, Tctrcihedron, 1971, 27. 1287. T. Hirata and T. Suga, J . O r g . Chem., 1971, 36, 412. %. Chabudzinski. Z . Rykowski. and IJ. Lipnicka, Koczniki Chpwi., 1971, 45, 27. T. Suga, T. Hirata, M. Noda, and T. M a t si ~ u r a E,xperientia, , 1970, 26, 1192. Y . Bessiere-Chreticn, J.-P. Montheard. M . M . El Gai'ed, a n d J.-P. Bras. Compt. rend., 197 1, 273, C, 272. A . Eschenmoser, D . Felix, and G . OhloK, f f e / i . .Cfqirn. A c / a . , 1967, S O , 708. B. Bochwic and S. Markowicz, Roczt7i!=/

H (15 )

(14)

(13)

lA

methylenetriphenylphosphorane on the resultant aldehyde gave the diene (20) which, on selective reduction with di-imide and hydrolysis, yielded the alcohol (17). Alternatively, reaction of the allylic bromide derived from (19) with trimethylironlithium and subsequent hydrolysis also afforded the alcohol (17). The stage was then set for the completion of the juvenile hormone syntheses. The phosphonium ylide (21), derived from (17) via the corresponding iodide, was reacted with the aldehyde (22) and, subsequently, the P-oxido-ylide was treated with paraformaldehyde to yield the alcohol (23). This important intermediate was e l a b ~ r a t e d to~both ~ the C,, and C,, juvenile hormones (24; R = Me) and CH,OTHP

TPPh3 + +

CH,O

CHO

(21)

(22)

HO, C €

P

-w e

*

0

(24)


68

(24; R = Et) by a sequence of standard reactions. Using precisely the same methodology, Corey and Yamamotos‘ have accomplished short and stereospecific syntheses of farnesol(25) and a positional isomer of C,, Cecropia juvenile hormone (26) as outlined in Scheme 1. A preliminary examination of the biological activity of (26) indicates that it is very active and species specific. CH,OTHP

TPPh3 + +

CH20

CHO

(22)

(25)

Scheme 1

Although the past three years have witnessed many ingenious syntheses of the two insect juvenile hormones, the preparation of the naturally-occurring dextrorotatory C,, hormone (cis-epoxide)has only recently been achieved. Indeed, in two independent both enantiomeric pairs of the cis- and trans-epoxides have been prepared and it has been firmly established that the two chiral centres have the 10R,11S configurations in the naturally-occurring material. Findlay et a[.* have now published full details of their previously announced synthesis of the two juvenile hormones and other double-bond isomers. In an earlier synthesis of the C,, hormone, Corey et ul. used the dienol (30; R = Et) as a key intermediate and now they’ have described two new stereospecific routes to this compound (Scheme 2). In the first synthesis the lactone (27) was converted into the hydroxy-olefin (28) by hydrolysis, esterification, tosylation, and lithium P. Loew and W. S. Johnson, J . Amer. Chem. SOC.,1971, 93, 3765. ’ D. J. Faulkner and M. R. Petersen, J . Amer. Chem. SOC.,1971, 93, 3766.

*

J. A . Findlay, W. D. MacKay, and W. S. Bowers, J . Chem. SOC.(0,1970,263 1. E. J. Corey, J. A . Katzenellenbogen, S. A. Roman, and N. W. Gilman, Tetrahedron Letters, 1971, 1821.

69

Sesquiterpenoids

aluminium hydride reduction. Treatment of (28) with phosphorus tribromide, followed by alkylation with lithio-1-trimethylsilylpropyneand subsequent desilylation, yielded the acetylenic olefin (29) which was converted into (30; R = Et) by a previously designed synthetic sequence. The alternative synthesis of (30 ; R = Et) involved the preparation of the diyne-ol (31) by reaction of the tosylate of pent-3-yn-1-01 with 3-lithiopropargyl tetrahydropyranyl ether followed by hydrolysis. Reaction of (31) with LiAlH,-NaOMe and iodination of the resultant organoaluminium intermediate (32) produced the di-iodo derivative (30 ; R = I) in 45 % yield which was, in turn, converted into (30 ; R = Et) by reaction with diethylcopperlithium. Corey et al. have converted the aldehyde derived from (30; R = Et) into the dehydro-analogue of c18 juvenile hormone (34) by reaction with the lithio-derivative of (33) followed by epoxidation. With certain insect species, this dehydro-derivative is more active than the c18 hormone itself. In view of the known inhibition of 2,3-iminosqualene towards the enzymatic conversion of 2,3-oxidosqualene into lanosterol, Corey et al. have synthesized the

''

JP

C0,Me

s'-.-...; \

0

Reagents: i, O H - ; ii, H'; iii, CH,N,; iv, p-MeC,H,SO2C1; v, LiAlH,; vi, PBr,; vii, LiCH,C=CSiMe,; viii, A g + , C N - ; ix, BuLi-CH,O; x, LiAlH,-NaOMe; xi, I,; xii, LiEt2Cu.

Scheme 2 lo

L. M. Riddiford, A. M. Ajami, E. J. Corey, H. Yamamoto, and J. E. Anderson,

J. Amer. Chem. SOC.,1971, 93, 1815.

70


two aziridine juvenile hormone analogues (35; R = Me) and (35; R = Et). Bioassays with these two compounds indicate that they have a synergistic effect upon the action of C,, juvenile hormone, and this is rationalized by the reasonable assumption that the aziridine analogues will bind strongly to the sites involved in metabolism - deactivation of juvenile hormone.

Two new toxic sesquiterpenoids, myodesmone (36) and isomyodesmone (37), have been isolated from the essential oils of certain Myoporum species." It has been suggested that these two compounds may be derived from myoporone (38), also found in some specimens of Myoporurn deserti, by an in vivo aldolizationdehydration process via the ketol(39). In addition to ipomeamarone (40 ; R = H), it is reportedl2'l3 that a new sesquiterpenoid, ipomeamaronol (40; R = OH) also occurs in diseased sweet potato root tissue.

l3

I. D. Blackburne, R. J. Park, and M. D. Sutherland, Austral. J . Chem., 1971, 24, 995. N.Kato, H . Imaseki, N . Nakashima, and I. Uritani, Tetrahedron Letters, 1971, 843. D. T. C. Yang, B. J . Wilson, andT. M. Harris, Phytochemistry, 1971, 10, 1653.

71

Sesquiterpenoids

Another non-stereospecific synthesis of davanone (41) has been reported by Birch et ~ i . , who ' ~ converted 2-methylhept-2-en-6-one into (42) by ozonolysis and treatment with a-ethoxycarbonylethylidenetriphenylphosphorane.Ethynylation of (42) followed by base-induced cyclization and partial reduction produced the ester (43; R = C0,Et). Treatment of the corresponding acid with dimethylallyl-lithium gave, in low yield, davanone (41) together with three other diastereoisomers. Ohloff and Giersch15 have also synthesized davanone and, from a combination of their work with the results of Birch et al.,14 the stereostructure (44) is suggested for ( )-davanone. Another sesquiterpenoid, artemone (45), has been foundI6 to co-occur with davanone, and its synthesis has been achieved by treatment of the aldehyde (43 ; R = CHO) (previously used in the first synthesis of davanone) with 3-methylbut-2-enylmagnesium bromide followed by Jones oxidation.

+

2 Monocyclo- and Bicyclo-farnesanes In the course of biosynthetic studies on the antibiotic, siccanin (46),Nozoe et ~ 1 . ' ~ have not only isolated trans-y-monocyclofarnesol (47) from the mycelia of the fungus, Helrninthosporium siccay1s, but they have also demonstrated that a cell-free enzymatic preparation from the fungi converts mevalonic acid lactone into (47) in high yield. A key intermediate in the synthesis of abscisic acid is (48), which l4 l5 l6

l7

A. J. Birch, J. E. T. Corrie, and G. S. R. Subra Rao, Austral. J . Chem., 1970,23, 18 11. G. Ohloff and W. Giersch, Helu. Chim. A m , 1970, 53, 841. P. Naegeli, J. KlimeS, and G. Weber, Tetrahedron Letters, 1970, 5021. K. T. Suzuki, N . Suzuki, and S. Nozoe, Chem. Comm., 1971, 527.

72


can now be prepared in 43% yield starting from p-ionone.'8 Thus, allylic bromination-dehydrobromination of p-ionone yields the dehydro-derivative (49) which is selectively epoxidized at the tetrasubstituted double bond before Jones oxidation to (48). An X-ray crystallographic study" has amended the structure of the interesting fungal metabolite, isocollybolide, to (50). Cyclonerodiol ( 5 1) has been isolated2' from a second fungal source, Gibberella fujikuroi.

0

'

OCO-Ph (50)

D OH

(51)

Following the successful synthesis of cinnamolide (52) from monocyclofarnesic acid, Kitahara et aL2' have now converted the former compound into polygodial (53). This was effected by oxidative modification of cinnamolide (52) to the aldehyde-ester (54), which was converted into polygodial by standard protection, reduction, and oxidation methods. As the result of a detailed n.m.r. analysis of bilobalide and its derivatives, Nakanishi et ~ 1have . deduced ~ ~ the structure (55) for this unique sesquiterpenoid. In the light of the biosynthetic studies on the diterpenoid, ginkgolide B (56),23

' * J. A. Findlay and W. D. MacKay, Canad. J. Chem., 1971,49, 2369. l9

2o 21

22

23

C. Pascard-Billy, Chem. Comm., 1970, 1722. B. E. Cross, R. E. Markwell, and J. C . Stewart, Tetrahedron, 1971, 27, 1663. T. Kato, T. Suzuki, M. Tanemura, A. S . Kumanireng, N. Ototani, and Y . Kitahara, Tetrahedron Letters, 1971, 1961. K. Nakanishi, K. Habaguchi,Y. Nakadaira, M. C. Woods, M. Maruyama, R.T. Major, M. Alauddin, A. R. Patel, K. Weinges, and W. Bahr, J . Amer. Chem. SOC.,1971, 93, 3544. K . Nakanishi and K. Habaguchi, J . Amer. Chem. SOC.,1971, 93, 3546.

Sesquiterpeno ids

73

(53)

(52)

(54)

the biogenesis of bilobalide can be rationalized in terms of either a degraded ginkgolide or as a genuine sesquiterpenoid derivable from a bicyclofarnesyl pyrophosphate, e.g. (57), and subsequent oxidative modification t ~ i athe rearranged carbon skeleton (58). 0

0

H (55)

PPO,

(57)

3 Bisabolane and Sesquicarane Vig et al. have reported the syntheses of y-bisabolene (59)24 and dehydro-acurcumene (60).25Both isomeric atlantones (61) and (62) have been isolated from Cedrus deodara26and a-bisabolol (63) has been isolated from the essential oil of the cotton plant.27 Two syntheses of nuciferal (64) have been reported by Gast and Naves28starting from 2-(ptolyl)propanal. Russian workers29have described the synthesis of Ar-juvabione (67) by alkylation of the P-keto-ester (65) with (66) followed by hydrolysis, decarboxylation and re-esterification. The isolation and 24

25 26

2’

28 29

0. P. Vig, B. Ram, C. P. Khera, and J . Chander, Indian J . Chem., 1970, 8, 955. 0. P. Vig, R. C. Anand, A. Singh, and J . P. Salota, Indian J . Chem., 1970, 8, 953. B. S . Pande, S. Krishnappa, S. C. Bisarya, and S. Dev, Tetrahedron, 1971, 27, 841. P. A. Hedin, A. C. Thompson, R. C . Gueldner, and J . P. Minyard, Phytochemistr:,, 1971,10, 1693. G. Gast and Y.-R. Naves, Helv. Chim. Acta, 1971, 54, 1369. 0. V. Efimova, A. A. Drabkina, and Yu. S . Tsizin, J . Gen. Chem. ( U . S . S . R . ) , 1970, 2497.

74


structural elucidation o f xanthorrhizol (68)30 and angelikoreanol (69)3 have been reported.

pp CHO

'"H. Rimpler, R. Haensel, and L. Kochendoerfer, Z . Narurfursch., 1970, 25b, 995. 3L

K . Hata, M. Kozawa, K. Baba, M . Konoshima, and H.-J. Chi, Tetrahedron Letters, 1970. 4379.

75

Sesquiterpeno ids

As an alternative route to the sesquicarane-type sesquiterpenoids, Hortmann and Ong32 have examined the carbanionic opening of the epoxy-ester (70) derived from perillaldehyde (71). The two products of this reaction are (72) and (73) in the ratio 3 : I . Further elaborations of these two compounds have still to be carried out. Plattner and R a p ~ p o r have t ~ ~ now reported the preparation of both (+)- and (-)-sirenin. This was accomplished by preparative g.1.c. separation of the two diastereoisomeric pairs of ketals derived from the synthetic ketone (74) and D-( -)- and L-( +)-butane-2,3-diols. Synthetic procedures for the conversion of racemic (74)into both racemic sirenin and racemic sesquicarene had already been developed. Furthermore, these authors have firmly established by chemical correlation and c.d. studies that naturally-occurring sirenin and sesquicarene have the same absolute stereochemistry, i.e. (75) and (76) respectively. 70,Me

CHO

B 9

C0,Me

4 Daucane

As a result of extensive chemical degradation and correlation with other members in this group, e.g. carotol (77), the absolute configuration of laserpitin (78) has 32 33

A. G . Hortmann and A . Q. Ong, J . Org. Chem., 1970,35,4290. J. J. Plattner and H. Rapoport, J . Amer. Chem. SOC.,1971, 93, 1758.

76


been deduced.34 Korthals et aL3 claim to have identified the two daucane-type sesquiterpenoids (79) and (80) in the essential oil of Fokienia hodginsii L.

DH

H O ,

H :

(78) Ang

(77)

=

angeloyl

5 Cadinane and Related Tricyclic Sesquiterpenoids Gerber has isolated (+)-epicubenol (81) from a Streptomyces species.35 This compound is the enantiomer of that found in the essential oil of Cedrefa toona Roxb. The syntheses of four naturally-occurring phenolic sesquiterpenoids obtained from the essential oil of elm wood have been reported,36 viz. (82; R = Me), (82; R = CHO), (83; R = Me), and (83; R = CHO). The known o-quinone, mansonone C (84), has also been isolated from elm wood.37 The structure (85) of sesquichamaenol, a minor component of the essential oil of Chamaecyparis jorrrzosensis, has been deduced on the basis of spectroscopic evidence and synthesis.38 Piers et ~ 1 have . published ~ ~ complete details of their syntheses of a- and b-cubebenes (86).

(81) 34

35

36

37 38 39

(82)

(83)

M . Holub, J . Tax, P. Sedmera, and F. Sorm, Coll. Czech. Chem. Comm., 1970, 35, 3597. N . N . Gerber, Phytochemistry, 1971, 10, 185. J. Alexander and G. S. K. Rao, Tetrahedron, 1971, 27, 645. V. Krishnamoorthy and R. H. Thomson, Phytochemistry, 1971, 10, 1669. M . Ando, S. Ibe, S. Kagabu, T. Nakagawa, T . Asao, and K. Takase, Chem. Comm., 1970, 1538. E. Piers, R. W. Britton, and W. de Waal, Canad. J. Chem., 1971, 49, 12.

Sesquiterpeno ids

77

In continuation of their programme of sesquiterpenoid synthesis, Piers et ~21.~’ have accomplished very elegant syntheses of ( + )-copacamphor (87), ( - )-copacamphene (88), and (-)-cyclocopacamphene (89). Not only do these syntheses completely confirm the structures of these compounds but they also corroborate the absolute stereochemistries assigned to these and related compounds. The key compound in the synthetic sequence is the diketone (92) which was obtained from (+)-carvomenthone (90). This was achieved by conversion of (90) into the corresponding n-butylthiomethylene derivative followed by stereoselective alkylation with methyl 2-iodopropionate, removal of the blocking group, and esterification to yield the keto-ester (91). Treatment of (91) with sodium bis(trimethylsily1)amideafforded (92) in 90 %’ yield. Hydrogenation of the corresponding enol acetate yielded the keto-acetate (93) which was converted into the homologated hydroxy-aldehyde (94)by a Wittig reaction with methoxymethylenetriphenylphosphorane. A further Wittig reaction with methylenetriphenylphosphorane followed by hydroboration yielded the diol (95) which was converted into the keto-tosylate (96) by selective tosylation and Collins oxidation. Treatment of (96) with dimsyl sodium yielded (+)-copacamphor (87) which had

40

E. Piers, R. W. Britton, R. J. Keziere, and R. D. Smillie, Canad. J. Chern., 1971, 49, 2620,2623.


78

v

v

v

I

I

1

(93)

(94)

(95)

v

TsO

v

/--

(97)

previously been transformed into ( + )-copaborneol(97)and ( + )-copaisoborneol (98). For the syntheses of ( -)-copacamphene and ( -)-cyclocopacamphene, Piers et aL4' converted the diketone (92) into the keto-alcohol (99), the tosylhydrazone of which yielded the ene-ol (100; R = H, OH) by treatment with methyl-lithium. Collins oxidation of (100; R = H, OH) gave the enone (100; R = 0) which was converted into the homologated ene-ol (101) by the same method as was employed in the synthesis of (95) (disamylborane was used for the selective hydroboration stage). The tosylate of (101) rapidly cyclized to (-)copacamphene (88) on standing. It is reported that the sample of copacamphene thus obtained is laevorotatory, which is at variance with a previous report. Finally,

v

v

I (99)

p@ v

v

HO-

N=N

79

Sesquiterpenoids

the tosylhydrazone of the olefinic aldehyde corresponding to (101) was thermolysed to give the pyrazoline (109 which afforded, on photolysis, (-)-cyclocopacamphene (89), the enantiomer of the tetracyclic hydrocarbon derived from cyclocopacamphenic acid, a constituent of vetiver oil. A synthetic approach to the sesquiterpenoid alkaloid, dendrobine (103) has been a n n ~ u n c e d . ~This ' involved the formation of the bicyclic ketone (105) by a Diels-Alder reaction of carvotanacetone (104) with butadiene. cis-Hydroxylation of (105), followed by periodate cleavage and aldolization, led to (106) which was converted into the ketal-ester (107) by standard procedures. Reaction of (107) with methylamine and subsequent reduction and deketalization yielded (108) which was converted into (109) by hydrogenolysis under forcing conditions (PtO-200 "C-50 atm).

MeHN-)T

6 Campherane and Santalane

As a sequel to the elegant and highly efficient synthesis of camphor by an intramolecular cyclization, Money et have successfully applied this method to the synthesis of the naturally-occurring sesquiterpenoids campherenone (110) and campherenol (111) whose structures and absolute stereochemistries have recently been deduced.43 In this sequence, dihydrocarvone (112) was readily converted 'into the keto-ketal (113) from which the homologated chloro-ketone (114) was obtained. Treatment of the corresponding enol acetate (115) with boron trifluoride in wet methylene chloride afforded the two bicyclic chloro-ketones (116) and (117) in 55-60 % yield. Conversion of the corresponding iodo-ketals 41 42 43

K. Yamamoto, I , Kawasaki, and T. Kaneko, Tetrahedron Letters, 1970, 4859. G. L. Hodgson, D. F. MacSweeney, and T. Money, Chem. Comm., 1971, 766. H . Hikino, N. Suzuki, and T. Takemoto, Chem. and Pharm. BuII. (Japan), 1971,19,87.

80


into the Wittig salts and reaction with acetone yielded campherenone (110) and epicampherenone (118) respectively," each of which afforded the four possible alcohols on reduction with sodium in n-propanol (endo-OH)or with lithium aluminium hydride (exo-OH). Not only does this synthetic route embody a very plausible biogenetic pathway to these compounds, but further cyclizations of campherenone-type precursors could lead directly (both in the synthetic and biogenetic sense) to other known sesquiterpenoids, e.g. longiborneol, copaborneol, longifolene, sativene, etc. Within this area it is now abundantly clear that there is a close parallel between certain monoterpenoids and their sesquiterpenoid counterparts. It should, however, be pointed out that from the absolute stereochemical point of view ( - )-campheren0r.e correlates with the known sesquiterpenoids associated with the longi series [including ( + )-sativene but excluding culmorin] but not with those in the copa series [except (+)-cyclocopacamphene]. Hikino et al. have also shown that treatment of epicampherenol (119) with p-toluenesulphonyl chloride in pyridine gives rise to p-santalene (120). In view of the fact that a number of sesquiterpenes are isoprenologues of tricyclene (121), Barnett and M ~ K e n n have a ~ ~ deveioped an efficient route to the

0

R

OH

3"' 44

W. E. Barnett and J. C. McKenna, Tetrahedron Letters, 1971, 227.

* Racemic campherenone has been converted into both

OL- and fi-santalene in good yield whereas racemic epicampherenone afforded epi-B-santalcne. Personal communication from Professor T. Money.

Sesquit erpenoids

81

tricyclic lactone (125). This was achieved by a Diels-Alder reaction of cyclopentadiene with the allenic acid (122) which gave the two acids (123) and (124), the former of which underwent a smooth acid-catalysed cyclization to (125). The full synthetic potential of this lactone has still to be realized in terms of such compounds as a-santalene (126), cyclosativene (127), and longicyclene (128). To date Barnett and McKenna have transformed (125) into teresantalol (131) via the thioethers (129) and (130).

In the course of developing stereospecificroutes to trisubstituted double bonds, Corey et ~ 1 have . completed ~ ~ an efficient synthesis of a-santalol (132), many stages of which involve recent synthetic techniques from Corey's laboratory. Thus (-)-71-bromotricyclene (133) was converted into (134) by successive treatment with lithio-1-trimethylsilylpropyne,silver nitrate (desilylation), and potassium 45

E. J. Corey, H. A. Kirst, and J. A. Katzenellenbogen, J. Amer. Chem. Soc., 1970, 92, 63 14.


82

cyanide. The substituted propargyl alcohol (135) was then prepared from the lithio-derivative of (134) by treatment with paraformaldehyde. Reaction of (1 35) with butyl-lithium followed by treatment with di-isobutylaluminium hydride and then iodine produced the iodo-alcohol (136; R = OH) which was converted into the vinylic iodide (136; R = H) by successive treatment with mesyl chloride, lithium bromide, and sodium borohydride. Finally, reaction of (136; R = H) with nickel carbonyl-sodium methoxide and aluminium hydride reduction of the resultant methyl ester yielded a-santalol (132).

(136)

(135)

7 Thujopsane, Acorane, Chamigrane, Bazzanane, and Trichothecane It has been suggested46that the formation of the keto-aldehyde (137)in relatively high yield from the sensitized photo-oxidation of thujopsene (138) can best be explained in terms of a dioxetan intermediate (139), similar examples of which have recently been found in singlet oxygen addition to electron-rich double bonds. An extensive analysis of the products of acid-catalysed rearrangement of thujopsene (138) has been carried Under different acid conditions ten products have been isolated and identified ; these include the known compounds, chamigrene (140),cuparene (141j, and widdrol(l42; R = H) together with the previously unknown compounds (142; R = Et) and (143)--(148). The authors have put forward a mechanistic scheme to explain the formation of all these compounds based on interconversions of cyclopropylcarbinyl and homoallyl cations.

(137)

(138)

(1 39)

'' S. It6, H. Takeshita, and M . Hirama, Tetrahedron Letters, 1971, 1181. '' S. It6, M. Yatagai, and K . Endo, Temdzedrun Letters, 1971, 1149; S . It6, M. Yatagai, K . kndo, and M . Kodama, Tetrahedrun Letters, 197 I , 1153.

Sesquiterpenoids

83

P-9

Minato et aL4*have reported the isolation of acorenone (149) from Acovus calumus L. The physical properties (m.p., [a],) which they ascribe to this compound are markedly different from two previous sets of Hydrogenation of acorenone is reported to give isoacorone (150 ;R1 = Me, R2 = H) and acorone (150; R' = H, R2 = Me). Recently, Conia et al.'l have demonstrated that thermal cyclization (220 "C) of the appropriately substituted cyclohexanone (151) [derived from ( + )-3-methylcyclohexanone]yielded four isomeric spiro-diketones closely related to the acorane skeleton. The intermediacy of the isopropylidene isomer (152) was indicated and from a detailed study of n.m.r. solvent shifts and c.d. spectra it was concluded that these four spiro-diketones can be represented as (1 53H156).

q 0

* y 2

-_

R'

0

(149) 48 49 50

( 150)

H. Minato, R. Fujioka, and K. Takeda, Chem. and Pharm. Bull. (Japan), 1971, 19,638.

J. VrkoE, V . Herout, and F. Sorm, Coll. Czech. Chem. Comm., 1961, 26, 3183. G . V. Pigulevski and V. Kovaleva. Rustitel'nye Resuray. Akad. Nauk U.S.S.R., 1966. 2(4), 527. J. M . Conia, J . P. Drouet, and J. Gore, Tetrahedron, 1971, 27, 2481.


84

Two new halogenated sesquiterpenoids have been isolated from different species of the marine alga Laurencia. Both compounds, spirolaurenone (157)” from L. glandulifera and pacifenol (158)53from L. pacijica, can be considered to belong to the chamigrane type of sesquiterpenoid. From a biogenetic standpoint it is interesting to note that the previously known laurinterol (159) co-occurs with pacifenol. Further evidence in support of the structural assignment of bazzanene (160)has now been presented.54

The trichothecane group of sesquiterpenoids are now well recognized metabolites from such fungal sources as Trichotheciwm, Myrothecium, and Fusurium. To date the main interest in this group has centred around isolation, structural 53

M. Suzuki, E. Kurosawa, and T. hie, Tetrahedron Letters, 1970, 4995. J. J. Sims, W. Fenical, R. M. Wing, and P. Radlick, J . Amer. Chem. SOC.,1971, 93,

54

A. Matsuo, Tetrahedron, 1971, 27, 2757.

52

3774.

85

Sesquiterpenoids

elucidation, and biosynthesis." A recent example is the antibiotic roridin E ( 161).55The challenging task of synthesis of a member of this group has now been

accomplished in the form of trichodermin (162) by Raphael et a1.56 The key compound on which this route hinged was the bicyclic y-lactone (164). This compound was obtained in good yield from the known enone-ester (163) by selective Grignard addition to the keto-group with methylmagnesium chloride followed by acid-catalysed lactonization (an anionotropic rearrangement is involved) and subsequent methylation with methyl iodide in the presence of lithium di-isopropylamide. The construction of the requisite 2-oxabicyclo[3,2,1]octan-6-01-8-one was achieved in the following manner. Reaction of (164) with 3,3-diethoxypropyne followed by borohydride reduction of the derived hemiketal (165) and subsequent partial reduction of the triple bond with sodium in liquid ammonia yielded (166). Mild acid treatment of (166) effected both the desired deacetalization and the intramolecular conjugate cyclization to afford the hydroxy-aldehyde (167). The corresponding keto-aldehyde could not be induced to undergo an intramolecular aldolization, and thus the hydroxy-aldehyde (167) was converted into the keto-acid (168) in two steps. Lactonization of (168) and reaction of the resultant enol-lactone (169) with lithium tri-t-butoxyaluminium hydride afforded (170) in moderate yield. The concluding steps in the synthesis were achieved by treatment of the acetoxy-derivative of (170) with methylenetriphenylphosphorane to give (171) and then epoxidation of the corresponding alcohol followed by a final acetylation.

H

(163)

(164)

P. Traxler, W. Zurcher, and Ch. Tamm, Helv. Chim. Acta, 1970, 53, 2071. 5 6 E. W. Colvin, R. A . Raphael, and (in part) J. S. Roberts, Chem. Comm., 1971, 858. * Two recent biosypthetic papers-are discussed in the Biosynthesis chapter. 55


86

8 Longifolane Compou,ids belonging to this group are well-known participants in the field of molecular acrobatics. Ourisson et aL5’ have now carried out an extensive examination of the 1,5-hydride shift which occurs in the course of solvolysis of the bromides (172).Ten R variants with differing electronic demands (e.g.R = CN, CH,OMe, CH,Me) were synthesized ; rates of ethanolysis were determined and product analyses were performed. The results of this very interesting study show that : ( a ) there is a 200-fold rate increase in going from R = CN to R = Et ; (b) when R is an electron-withdrawing group the major solvolysis products are ring-contracted, e.g. (173); (c) when R is an electron-releasing group the major product is the result of a 1,5-transannular hydride shift, e.g. (172; R = Me) gives

57

L. Stehelin, J . Lhomme, and G. Ourisson, J . A m e r . Chern. Soc., 1971, 93, 1650.

87

Sesquiterpenoids

Iongifolene(174). Thus it is seen that the transannular hydride shift is an accelerating factor when compared with the neopentylic rearrangement, and this can be interpreted in terms of a(C-H) anchimeric assistance in the rate-determining step. Another rearrangement of the longifolane skeleton has been recorded by Lhomme and O u r i ~ s o n . ~They * have found that the keto-ester (175) is transformed into the isomeric compound (176) on treatment with boron trifluoride in benzene and they have suggested the mechanism outlined in Scheme 3. Coates

Me02C \

Scheme 3

and ChenS9 have shown that longicamphenilyl tosylate (177), on acetolysis, gives rise to a number of compounds of which (178H181) have been identified. With formic acid, (179) is isomerized to (180).

(179) 59

(180)

(181)

J . Lhomme and G. Ourisson, Bull. SOC.chim.France, 1970, 3935. R. M. Coates and J. P. Chen, Chem. Comm., 1970, 1481.


88

9 Caryophyllane, Humulane, and Related Compounds

Direct photolysis of caryophyllene (182) and isocaryophyllene (183) gives rise to an extremely complex mixture of isomeric hydrocarbons. The majority of these have now been identified and rationalizations for their formation have been presented.60 In addition to the photoisomerization between caryophyllene and isocaryophyllene, caryophyllene produces (184)-(188), whereas isocaryophyllene yields (187H192).

q-J q-J H'

H'

6o

K. H . Schulte-Elte and G . Ohloff, Helv. Chim. Acta, 1971, 54, 370.

Sesquiterpenoids

89

From the essential oil of Copaifera multijuga Hayne a new caryophyllene derivative, a-multijugenol(193), has been isolated.61 Neither the stereochemistry of the endocyclic double bond nor the configuration of the hydroxy-group has been deduced, although the latter should be o! from the fact that a-muitijugenol can be derived from caryophyllene epoxide (194). Recently, Gupta and Dev62 have reported the isolation of (195H197) from the oleoresin of Dipterocarpus pilosus [again, the stereochemistry of the endocyclic double bond in (195) and (196) is not known with certainty]. From the physical constants quoted neither (195) nor (196) appears to be identical with a-multijugenol. Certain discrepancies also exist with respect to the apparently identical dienols recently obtained by

(193)

sensitized photo-oxidation of caryophyllene. The structure (201) of the major hydrocarbon derived by dehydrochlorination of caryophyllene dihydrochloride

(201) 61

62

G . D. Monache, I . L. D'Albuquerque, F. D. Monache, G . B. M. Bettolo, and G . M. Nano, Tetrahedron Letters, 1971, 659. A . S. Gupta and S . Dev, Tetrahedron, 1971, 27, 6 3 5 .

90

Terpenoids und Steroids

(198)has been deduced from the X-ray analysis of the corresponding d i b r ~ m i d e . ~ ~ Its genesis probably involves the intermediates (199) and (200). A tentative structure (202) has been assigned3 to siamol, another constituent of Fokieiziu hodginsii L.

(202)

X-Ray analyses of two heavy-atom derivatives of illudol (203)64and marasmic the cis-fused hydrindane skeletons in each and this acid (204) have fact is in accord with the stereochemistry of the hypothetical precursor (205) derivable from humulene.* Matsumoto and co-workers66have now completed

OH

O-0 (204)

(203)

0

o

q

(207) h3

h4

p

o OH (208)

~ 0 (209)

K . Gollnick, G . Schade, A. F. Cameron, C. Hannaway, and J . M. Robertson, Chem. Comm., 197 I , 46. T. C . McMorris, M . S. R . Nair, P. Singh, and M . Anchel, Phytochernistry, 1971, 10, 1611.

65 bh

P. D. Cradwick and G. A. Sim, Chem. Comm., 1971, 431. T. Matsumoto, H. Shirahama, A. Ichihara, H. Shin, S. Kagawa, F. Sakan, and K. Miyano: letrahedron Letters, 1971, 2049.

* A recent paper on the biosynthesis of illudins S and M is discussed in the Biosynthesis chapter.

Sesqu iterpeno ids

91

the synthesis of illudin S (206 ;R = OH) along lines very similar to their successful route to illudin M (206; R = H). Two new metabolites (207) and (208) cooccurring with fomannosin (209) have been identified.67 Last year three new compounds, coriolin and coriolin B and C, were identified as illudol-type sesquiterpenoids but, as a result of further chemical and spectroscopic data, these three compounds are now shown to be (210; R = H), (211 ; R = COC7HI5)and [210; R = C(O)CH(OH)-C,H,,] respectively and as such are more closely related to hirsutic acid (212).68 Two synthetic approaches to this latter compound have been initiated by Lansbury et a1.69 In the first of these the pyrrolidine enamine of 3-methoxycarbonyl-3-methylcyclopentanonewas

alkylated with 3-bromobutan-2-one yielding (213) as the major product. Aldolization of (213) and subsequent reduction afforded the substituted bicyclo[3,3,0]octan-3-one (214) which was stereoselectively transformed into the chloro-olefin (216) via a Claisen rearrangement of the trans-P-chlorocrotyl enol ether (215). Acid hydrolysis of (216) and aldolization with potassium t-butoxide produced (217), one racemate of which corresponded to a known degradation product of hirsutic acid. As an alternative entry into the basic cis,anti,cistricyclo[6,2,0,02~6]undecane framework of hirsutic acid, model studies have been carried out on the bicyclic ketone (218). Reaction of (218) with the Grignard reagent derived from 2-chloro-5-bromopent-2-ene produced (219), which cyclized in formic acid to give (220). This latter compound could also be derived by formation of (221) from (218) and subsequent formolytic cyclization and hydrolysis. 6’ 6g

S. Nozoe, H. Matsumoto, and S. Urano, Tetrahedron Letters, 1971, 3125. S. Takahashi, H. Naganawa, €3. Iinuma, T. Takita, K. Maeda, and H. Umezawa, Tetrahedron Letters, 1971, 1955.

69

P. T. Lansbury, N . Y. Wang, and J . E. Rhodes, Tetrahedron Letters, 1971, 1829; P. T. Lansbury and N . Nazarenko, ibid., p. 1833.


92 H

C0,Me \

Cl

Hp0 c1

H

(219)

10 Germacrane

Further work on the conformation of certain germacrane sesquiterpenoids has been performed using both X-ray analyses and the intramolecular nuclear Overhauser effect (n.0.e.). Complete details of the X-ray analysis of the silver nitrate adduct of germacratriene have been p~blished.~'In addition to demonstrating its conformation as (222), this analysis correlates the known reactivity (to epoxidation) of the three double bonds (4=5 > 1=10 > 7=11) with their torsional strains. The X-ray analysis7' of a heavy-atom derivative of pyrethrosin (223) has revealed that some stereochemical adjustments are required with respect 'O

"

F. H. Allen and D. Rogers, J . Chem. SOC.( B ) , 1971, 257. E. J . Gabe, S. Neidle, D. Rogers, and C . E . Nordman, Chem. Comm., 1971, 559.

Sesqu iterpen o ids

93

to the original structure deduced some fourteen years ago by Barton and de Mayo. On the bases of n.m.r., u.v., and c.d. spectra, the conformation and absolute stereochemistry of zederone is as depicted in (224).72 Complete details of the conformational assignment of isabelin (225)have been published7 and it has also been shown from n.0.e. experiments that isoisabelin, derived from isabelin by boron trifluoride etherate treatment, exists in the conformer (226).

Germacrone (227) can be photoisomerized into (228) and (229), of which the former can be converted into the previously unknown trienone (230).74 Hirose et ~ 1 . ~have ' demonstrated that a cell-free extract from immature seeds of Kadsura japonica Dunal can convert both [2-'4C]mevalonic acid lactone and trans,transfarnesyl pyrophosphate into germacrene C (231). In view of the fact that germacratriene epoxides may be intimately involved in biosynthetic pathways to certain bicyclic sesquiterpenoids (e.g. eudesmanes and guaianes), Hikino et have examined the microbial oxidation of germacrone (227) with CunningharnelEa blakesleeana. This study has revealed the formation of three major oxidation products. As a result of n.m.r. and 0.r.d. spectral data, together with an X-ray

l2 l3

74

l6

H. Hikino, K. Tori, I. Horibe, and K. Kuriyama, J . Chem. SOC.( C ) , 1971, 688. K. Tori, I. Horibe, H. Yoshioka, and T. J. Mabry, J. Chem. Soc. ( B ) , 1971, 1084. K. Takeda, I. Horibe, and H. Minato, Chem. Cumm., 1971, 87. K. Morikawa, Y. Hirose, and S . Nozoe, Tetrahedron Letters, 1971, 1131. H. Hikino, C . Konno, T. Nagashima, T. Kohama, and T. Takemoto, Tetrahedron Letters, 1971, 337.

94

Terpenoids and Steroih

analysis of the silver nitrate adduct of germacrone itself, the stereostructures of the three products are suggested to be as shown (232H234). Furthermore, treatment of epoxide (232) with p-toluenesulphocic acid afforded procuscumenol (235)of about 25 yi optical purity. Sam and S ~ t h e r l a n d ’in~ a continuation of their studies on the cyclization of germacratriene (222) have shown that both radical- and cation-induced cyclizations follow similar directional and stereoselective pathways. Thus irradiation of germacratriene in the presence of carbon tetrachloride or benzenethiol leads to (236; R 1 = CCI,, R2 = C1) and (236; R’ = SPh, R2 = H) respectively. These results, taken in conjunction with the cation-induced cyclizations, indicate that C-1-X (X = H or CCI,) and C-54-10 bonding are synchronous. Jain and McCloskey’’ have continued their studies on the cyclization of costunolide (237). On the expectation that the hydroxyeudesmanolide (238), the product of stereospecific hydration of costunolide, might be found in Nature they have prepared

R’

R2’

0 (236)

77

(237)

T. W. Sam and J. K . Sutherland, Chem. Comm., 1971, 970.

’’ T. C. Jain and J . E. McCloskey. Tetrahedron Letters, 1971, 1415.

Sesqu iterpeno ids

95 Br 1

H 0-

HO’ 0

0

this compound by two methods : (a) in very low yield by boron trifluoride etherate cyclization of costunolide and (b)by hydrogenolysis and retroaminolysis of (239), the product derived by N-bromosuccinimide cyclization of the amino-adduct of costunolide. There is an unconfirmed report that (238) is indeed a natural product. The dihydroxy-ketal (240), previously prepared from ( - )-santonin, has been used to synthesize a number of related sesq~iterpenoids.’~Thus the diacetate of (240) was converted in six steps into (241), which was then treated with isopropenyl acetate-sulphuric acid ; the derived enol-acetate was cleaved to the triol (242) by ozonolysis and lithium aluminium hydride reduction. The triol (242) was then converted into the di-iodo acetate (243) in a number of steps and thence to shyobunone (244) by dehydroiodination, reduction, and oxidation. Thermolysis of shyobunone at 1 6 G 1 8 0 “Cgave preisocalamendiol(245) in about 30% yield. More recently, Iguchi et have shown that preisocalamendiol (245) can be cyclized to isocalamendiol (246) in aqueous acetic acid ; no trace of calamendiol (247)* was found. A number of other interesting acid-catalysed cyclizations have been observed in this area, e.g. the formation of (248; R = OH) and (248; R = OAc) from (249) and the formation of (250) from (251). Finally, E-cadinene (252) has been obtained from (253), the lithium aluminium hydride product of preisocalamendiol (245).

HO

OH (242)

OAc (243)

’’ K. Kato, Y. Hirata, and S . Yamamura, Chem. Comm., 1970, 1324. M . Iguchi, M. Niwa, and S. Yamamura, Chem. Comm., 1971,974. M . Iguchi, A. Nishiyama, M. Niwa, S. Yamamura, and Y . Hirata, Chem. Comm., 1970, 1323.


96

(244)

(245)

OH

(249)

97

Sesqu it erpeno ids New germacranolides are listed in Table 1.

Name

Position(s) of double bond@)”

-OH

Other

la

-

-

1-keto

la;2u 3c!

-

Artemorin Anhydroartemorin Verlotorin Hydroxypelenolide Lanuginolide

4 3 ; 11,13; 10,15 4 3 ; 11,13; 10,15 4 3 ; 11,13; 10,151 1,lO 1,lO

Dihydroparthenolide

1,lO

-

Nobilinb Chihuahuin

4 3 ; 11,13; 1,lO 4 3 ; 11,13; 1,lO

3P 3a

-

1la-H ; 4P-H 4a,5/?-oxido; 11P-H 4a,5P-oxido ; 11P-H ; 8a-OAc 8a-OAng‘ 8a-OAc

Ref.

82 82 82 83 84 84 85 86

“ T h e 1,lO- and 4,s-double bonds are trans (or assumed to be). *Possibility of either 6/?,12-olide or the configurations at 3, 6, 7, and 8 are reversed, i.e. a, a, a, and a respectively. ‘ Ang angeloyl. =I

Enhydrin (254) has been shown8’ to be the 9,lO-epoxide of uvedalin. The structure of vernolide (255) has been unambiguously settled by X-ray analysis.88 Neosericenyl acetate (256) co-occurs with dehydrolindestrenolide (257) in Lindera strychinfolia Vill.89 Yet another constituent, pemoulone (258), has been isolated from Fokienia hodginsii L.3

82

83 84

” 86

” 89

T. A. Geissman and K . H. Lee, Phytochemistry, 1971,10,419. R. B. Bates, C. J. Cheer, and T. C. Sneath, J. Org. Chem., 1970,35, 3960. S. K. Talapatra, A. Patra, and B. Talapatra, Chem. Comm., 1970, 1534. V. BeneSova, Z. Samek, V. Herout, and F. Sorm, Tetrahedron Letters, 1970, 5017. W. Renold, H. Yoshioka, and T . J. Mabry, J. Org. Chem., 1970,35,4264. B. S. Joshi, V. N. Kamat, and H. Fuhrer, Tetrahedron Letters, 1971, 2373. C. Pascard, Tetrahedron Letters, 1970, 41 3 1. H. Tada, H. Minato, and K. Takeda, J. Chem. SOC. (0,1971, 1070.


98

CH,OAc

(257)

(256)

11 Elemane

A detailed n.0.e. study of the conformations of linderalactone (259; R = H), litsealactone (259 ; R = OAc), and the related furanogermacradienes (260) and (261) has shown conclusivelyg0 that the preferred conformer of each dictates the stereochemical outcome of the Cope rearrangement (via the preferred chair-like transition state). It is considered that the use of the word 'antipodal' to describe those elemadienes with a 10a-methyl group and a 5p-hydrogen atom is misleading since, in many cases, there are more than two chiral centres in the molecule. A number of furanogermacradienes are known with a cis-1,lO-double bond and a trcirzs-4,5-double bond and ziice versa. In the light of previous work, these might have been expected to give rise to cis-1,2-divinyl derivatives, but Takeda et have now shown that this is not the case. Thus neolinderalactone (262), sericenine (263 ; R = CO,Me), and (263 ; R = Me) give rise to the trans-compounds (264), (265 ; R = CO,Me), and (265 ; R = Me) respectively [the yield of (264) at 300 "C for 1-2 min is 5 that of (265; R = C0,Me) at 170 "C for 15 min is 'good' and (265 ; R = Me) i s obtained in 27 "/, yield at 200 "C for 4 h]. This result is taken to mean that the energy required for isomerization of the cis-double bond is less

0 (259)

"" ''

K . Takeda, K . Tori, I. Horibe, M. Ohtsuru, and H. Minato, J. Chem. SOC.(C), 1970, 2697. K . Takeda, I. Horibe, and H. Minato, J . ChPm. SOC.(C), 1970, 2704.

Sesquiterpenoids

99

1

I H

R

than that required for the Cope rearrangement. As a result of further work by Takeda et it is now possible to conclude that the furano-ring is responsible for the ‘abnormality’ of the foregoing Cope rearrangements involving cis,trarzsfuranocyclodeca-1,5-dienes.It has been shown that both (266; R = H2) and (266; R = 0),on thermolysis, yield the cis-isomers (267; R = H,) and (267; R = 0)respectively. More importantly, the vinyl ether (268)underwent a Claisen rearrangement to the aldehyde (269) which, on further heating, rearranged to the cis-isomer (270). By contrast, the isomeric vinyl ether (271) underwent the double rearrangement to give (272). It is thus concluded that although (269) has a cis-7,8-double bond, torsional strain about this double bond will be possible thus permitting attainment of the preferred transition state for the Cope process. Torsional strain should not be possible when a furano-ring is joined at positions 7 and 8.

(271) 92

(272)

K . Takeda, I. Horibe, and H . Minato, Chern. Cornm., 1971, 88.

Terpenoidr and Steroids

100

A plethora of products is obtained when either elemol(273), in the presence of p-nitrobenzoic acid (or benzoic acid), or elemyl-p-nitrobenzoate is pyrolysed at about 200 "C. These include the elemenes (274H276) together with the selinenes (277H28 1). The sequence of formation and the factors affecting the thermolyses are discussedg3 in depth, together with the data for the thermolysis of dihydrogeijerene (282). This latter compound has been synthesized from germacrone (227)by a Cope rearrangement to p-elemenone (283)followed by a retro-aldolization to (284) and subsequent Wolff-Kishner reduction.

OH

12 Eudesmane

In the past, synthetic routes to the eudesmane skeleton have involved a Robinson annelation sequence in the construction of the bicyclic framework. Although this procedure has enjoyed moderate success in the synthesis of certain members of this class, the associated small yields and stereochemical problems have been major drawbacks. Huffman and Mole94have developed an alternative approach 93 94

C. Ganter and B. Keller-Wojtkiewicz, Helv. Chim. A m , 1971, 54, 183. J. W. Huffman and M. L. Mole, Tetrahedron Letters, 1971, 501.

Sesquiterpenoids

101

to overcome these problems which is both short and stereoselective. Thus, Clemmensen reduction of the known tetralone (285) followed by Birch reduction with subsequent acid treatment gave the enone-acid (286) in about 30% yield, which was converted into an isomeric mixture of keto-acids (287) by conjugate methylation. A Wittig reaction with methylenetriphenylphosphorane on (287) followed by esterification, equilibration and hydrolysis gave (288) in reasonable yield. This acid has previously been used in the synthesis of P-eudesmol. Fringuelli and T a t i ~ c h i in , ~ a~ series of papers, have examined various aspects (reduction, hydroboration, etc.) of the chemistry of such compounds as (289)(291). McMurry et have re-investigated some stereochemical aspects of

(289)

(290)

(291)

certain compounds in the nordesmotroposantonin series. In the first place, the phenolic lactone (292), obtained from santonin (293) either by treatment with zinc dust in dimethylformamide followed by acid-catalysed isomerization or by pyrolysis,97 has a negative specific rotation in line with other such derivatives with a 6whydrogen. Secondly, vigorous acid treatment of (294) has been shown to lead to (295).

95

96

”

F . Fringuelli and A. Taticchi, J . Chem. SOC.(C), 1971, 756, 1809, 201 1. T. B. H. McMurry, D. F. Rane, and S. G. Traynor, Chem. and Ind., 1971, 658. T. B. H. McMurry and D. F. Rane, J . Chem. SOC.( C ) ,1971, 1389.


102

(294)

(295)

Recently, Kupchan et d9* have isolated two novel sesquiterpenoids of the eudesmane type, maytoline (296; R = OH) and maytine (296; R = H). These compounds not only contain a highly oxygenated sesquiterpenoid skeleton and a nicotinoyl ester grouping, but the stereochemistry of the isopropoxy-bridge is most unusual when taken in conjunction with the P-methyl and P-acetoxymethylene groups. The absolute stereochemistry of these two compounds is awaited with interest. A number of new eudesmanolides have been isolated and characterized. These include hybrifarin (297),99 arbusculin-C (298),' O0 rothin-A (299),'0° rothin-B (300),'00 dihydro-P-cyclopyrethrosin (301),101and chrysanin (302).'01 The last two co-occur with the germacranolides pyrethrosin (223) and chrysanolide (303).

A

c

O OAC; a :

R-

I

o+
7a > 12a) have been attributed to conformational and steric effects [3a-equatorial: 7a-and 12a-axial, and seemingly differentiated by greater steric hindrance at C(12), due to the side-chain]. A new investigation, however, shows that the reactivities of the monohydroxycholanic acids are in ratio : 3a : 7a : !2a = 97 : 1.0 : 1.5.1°2 A novel radioisotope method was used for kinetic studies. The enhancement of the relative rate of acetylation of 7cx-OH in cholic acid is apparently due to intramolecular catalysis by 3a- and 12aacetoxy-groups. Since catalysis is observed also when a 3p-acetoxy-group is present, an inductive mechanism is postulated. Slight enhancement of the rate of acetylation of a 6P-OH group by a 17-0x0- or 17~-benzoyloxy-group demonstrates a similar effect.'03

H (43)

OAc

OAc

H

H (44)

(45)

Acetyl migration between neighbouring hydroxy-groups affords a mixture (ca. 1 : 3) of the 16- (43) and 17-acetates (45) of a 16[j,17P-diol,when either of the acetoxy-ketones (42) or (44) is reduced with borohydride.' O4 Cyclic sulphites have been prepared from 3a,5-dihydroxy-5a-cholestan-6-one and its 3P,SP-isomer.l o 5 The compounds exist with the six-membered sulphite ring in a boat conformation. Reduction of the 6-0x0-group in the 3,$5P-compound I"'

I"'

A . Sattar and R . T. Blickenstaff, Steroids, 1971, 17, 357.

K. T. Blickenstaff, K . Atkinson, D. Breaux, E. Foster, Y . Kim, and G. C. Wolf, J . O r g . Chern., 1971, 36, 1271. T. Nambara, Y . Matsuki, T. Kudo, and T. Iwata, Chem. and Pharnz. Bull. (Japan), 1970, 18, 626. A. T. Rowland, T. B. Adams, H . W. Altland, W . S. Creasy, S . A. Dressner, and T. M . Dyott, Tetrahedron Letters, 1970, 4405.

Steroid Properties and Reactions

247

is followed by partial migration of the cyclic ester to give a mixture of 33- and 5,6-sulphites of the 3P,5P,6p-triol.' O6 The formation of hydrogen sulphates has been employed to separate alcoholic from non-alcoholic steroids.' O7 The preparation of cholesteryl pyrophosphate'" and phosphorodichloridate, and some related derivatives, is reported.'O' 4a-Methyl-5a-cholestane-4~,6a-diol (46) reacts with acetone znd perchloric acid, but instead of giving an acetonide, the reaction affords the novel cyclized products (48). l o Acetone apparently condenses with the intermediate 4-

'

H,O ___, -

1

acetone

C Me' (48) (A3

1 s ' h

Me

+ A4)

methylene-6a-alcohol (47). The 6a-methyl-4a,6/3-diol reacts similarly, but the 4a-methyl-4P,6P-diol affords only a mixture of dienes. A general survey of the scope of the reaction between alcohols and dihydropyran to form tetrahydropyranyl ethers"' may be of value in devising syntheses requiring the protecting of hydroxy-groups. The glycosylation of 3P-hydroxysteroids by potato-tuber slices has been systematically investigated. l 2 Oxidution. The chromium trioxide-pyridine complex is conveniently and safely prepared from the components in dichloromethane : this solution readily oxidizes alcohols to give aldehydes or ketones. l 1 The complex has more vigorous oxidizing properties in acetic acid.' l 4 Primary and secondary alcohols are lo6

lo'

Io8 Io9 'lo

'Iz IL3 'I4

A. T. Rowland, H. W. Altland, W. S. Creasy, and T. M. Dyott, Tetrahedron Letters, 1970,4409. J. K. Norymberski and A. Riondel, Biochem. J . , 1970, 119, 795. R . J. W. Cremlyn and N . A. Olsson, J . Chern. Soc. (C), 1970, 1889. R. J. W. Cremlyn and N. A, Olsson, J . Chem. Soc. (0, 1971,2023. J . R. Bull and A. Tuinman, Chem. Cornm., 1971, 717. H. Auterhoff and D. Egle, Arch. Pharm., 1970, 303, 688. 2. Prochazka, Coll. Czech. Chem. Cornm., 1971,36, 132. R . Ratcliffe and R. Rodehorst, J . Org. Chem., 1970, 35, 4000. K.-E. Stensio, Acta Chem. Scand., 1971, 25, 1125.

248


cleanly oxidized to give aldehydes and ketones, respectively, in a rapid reaction (10 min) and the solution generally remains homogeneous, facilitating isolation of the product. Cholesterol affords cholest-4-ene-3,6-dione in high yield, in contrast to the formation of cholest-5-en-3-one when the solvent is dichloromethane. Oxidation of alcohols in a two-phase system of aqueous chromic acid and ether is said to be highly efficient, and to avoid isomerization or isotope exchange in labile 0x0-products.' Ruthenium tetroxide, generated in situ from a suspension of the dioxide in CCl,, by adding aqueous sodium metaperiodate, appears to be an excellent reagent for the oxidation of secondary alcohols in neutral or basic media.'16 t-Amy1 or cumyl hydroperoxide, with molybdenum pentachloride, readily oxidizes steroidal alcohols ; cholesterol affords the 5a-hydroxy-3,6-dione in good yield.' Although some alcohols can be oxidized with lead tetra-acetate to give ketones, the 19-hydroxy-A5-system (49) undergoes C( lOkC(19) bond cleavage, giving

'

'

(49)

L

1

the 6/3-acetoxyoestr-5(10)-ene (51).' The resonance-stabilized ally1 radical (50) seems a probable intermediate.' A somewhat similar bond rupture occurred when the 2a,5a-epoxy- and 2a,5a-epithio-alcohols (52) were oxidized with lead tetra-acetate.' l9 The products were the ~-nor-3-oxa-and -3-thia-steroidal aldehydes (53). Ketone acetals may be oxidized by hydride transfer, using trityl fluoroborate (see also Part 11, Chap. 2, p. 388j.'" Ethylene acetals of steroidal ketones afforded the parent ketones in good yield, under non-polar conditions (in CH,Cl, at room temperature). The proposed mechanism (Scheme 3) results in oxidation ' I 5 'I6

11'

' IZo

H . C. Brown, C. P. Garg, and K.-T. Liu, J . Org. Chem., 1971,36, 387. R. M. Moriarty, H. Gopal, and T. Adams, Tetrahedron Letters, 1970, 4003. U . M. Dzhemilev, V. P. Yurev, and G. A. Tolstikov, Zhur. obshchei. Khim., 1970, 40, 2518.

A. Guida and M. Mousseron-Canet, Bull. Soc. chim. France, 1971, 1098. M. Kishi and T. Komeno, Tetrahedron, 1971, 27, 1527. D. H. R. Barton, P. D. Magnus, G. Smith, and D. Zurr, Chem. Cotnm., 1971, 861.


(52) X

=

249

i

0 or S

CHO (53)

of the diol component of the acetal to give a hydroxy-ketone, demonstrated by cleavage of Sa-cholestane-2/?,3P-diol acetonide (54) to give 3P-hydroxy-5acholestan-2-one (55). The spirostan spiro-acetal system was cleaved to give the 16,22-dione (56).

0

Scheme 3

The spiro-ether (tetrahydrofuran) (57) is oxidized by t-butyl chromate to give the spiro-lactone (58).12'

Reduction. Catalytic hydrogenation of 3P-acetoxy-5-enes may give as much as 10% hydrogenolysis of the acetoxy-group.' 2 2 Zinc-copper couple in refluxing lZ1

lz2

G . F. Reynolds, G . H . Rasmusson, L. Birladeanu, and G. E. Arth, Tetrahedron Letters, 1970, 5057. G . R. Pettit and B. Green, Cunad. J . Chem., 1970,48,2635.

250


ethanol is reported to reduce some epoxy-steroids slowly to give the corresponding olefins (e.g. 2,!3,3,!3- or 5a,6a-epo~ides).'~~ 'a'-Halogeno-ketones are dehalogenated by lithium iodide and boron trifluoride in ether, the mechanism probably

+: Lir

q "BF,

"'.a

G+

H

being of the type illustrated [(59)-+(60)].12" Steroidal 4,6-dichloro-4,6-dien-3ones (61) lose the 4-chloro-substituent during in vitvo incubation with rat liver. 12' Enzymic thiol groups are thought to be responsible, for the dechlorination may also be affected by simple thiols. The most effective thiols, however, are those providing a second (intramolecular) nucleophilic centre, such as an amido-group, to permit formation of a heterocyclic by-product. The postulated mechanism is illustrated in Scheme 4. Other 4-chloro-4,6-dienones react similarly, though giving lower yields.

R'

NH

I

(61)

R

J

R

Tributyltin hydride, with a free-radical initiator, reduces secondary and tertiary halides via the corresponding alkyl radicals.' 2 6 The Sa-chloro-3p,4/?-diol l 2

'

125

lZb

S. M . Kupchaii and M . Maruyama, J . 01-g. Chrm., 1971, 36, 1187. J . M . Townsend and T. A. Spencer, 7etrahedron Letters, 1971, 137. R. A. LeMahieu, M. Carson, D. E. Maynard, P. Kosen, and R. W Kierstead, J . Amer. Chein. SOC.,1971, 93, 1664. S. Julia and R. Lorne, Conzpf. rend., 1971, 273, C, 174.


(62) x = CI (63) X = H

25 1

Me (64)

1

c=o I

diesters (62) afford the 5a-H compound (63), but 6P-halogen0-3P,Sa-diol diacetates (64) react with participation of the 5a-acetoxy-group to give the mesomeric acetoxonium radical (65), resulting in final formation of the 3P,6a-diol diacetate (66). A 4P-halogen0-3P,Sa-diol diacetate similarly gives the 3P,4a-diol diacetate. Miscellaneous. The 6P,7P-dibromomethylene steroid (67) reacted with methyllithium to give the very strained bicyclobutane (68).'27 The reaction presumably involves insertion of the bridging carbon atom [C(7a)] in carbenoid form, into the C(8)-H bond (see also p. 3 15).

The 7,!l-fluoro-substituent in the s-homo-oestrane derivatives (69) is unusually stable, surviving aromatization of ring A, or reduction of the 3-0x0-group by lithium aluminium hydride, except in refluxing diglyme. 28 I*'

E. Galantay, N. Paolella, S. Barcza, R. V . Coombs, and H. P. Weber, J . Amer. Chem. SOC.,1970, 92, 5771. E. Verlarde, L. H . Knox, A. D. Cross, and P. Crabbe, Annulen., 1971, 748, 123.


252

6

r-

5; r-

Y

v

go 0

z

ir3

v

'0 5:

0 3:

253

Steroid Properties and Reactions T-----

Various reactibs of the 5a,8a-epidioxy-A6-olefinicsystem have been described (Scheme 5).'29 Rearrangement of compound (70) in refluxing n-decane afforded the epoxy-ketone (71). Palladium in ethanol (as a hydrogen source) reduced the peroxy-group, giving the 5a,Sa-diol (72) but prolonged contact with the catalyst caused slow dehydrogenation and dehydration to give the 4,6,8(14)-triene-3,17dione (73). The corresponding 3-oxo-5a,8a-epidioxy-compoundrearranged in pyridine solution, probably via the A3-enol, to give the epoxy-ketone (74), which rearranged further in acidic or basic solution to give the respective products (75) and (76).

3 Unsaturated Compounds Electrophilic Addition.-The olefinic bond in an ergost-22-ene undergoes stereospecificand regiospecificaddition reactions, controlled by its dissymmetric environment.' The preferred conformation, revealed by X-ray analysis, is represented by (77). Frontal hindrance by the steroid nucleus, and the 24-methyl H

IZ-AgOAc

Me**+ steroid

Me

Me'H

MeH Me

'H

group, enforces rear attack by electrophiles, so that iodine-silver acetate in acetic acid, for example, leads through the iodonium ion (78) to the iodohydrin acetate (79): nucleophilic attack on the iodonium ion occurs at the less-hindered C(23). The 22,23-epoxide (SO), derived via the iodohydrin acetate, corresponded, as expected, to the minor product of dirzct epoxidation of the AL2-olefinicbond. 13"

W. F. Johns, J . O r g . Chem., 1971, 36, 2391. D. H . R. Barton, J. P. Poyser, and P.G. Sammes, Chem. Conzrn., 1971, 715.

Terpeizoids and Steroids

254

Bromine in DMF, with silver perchlorate to precipitate bromide ions, reacts with a steroidal 2-ene (81) to give the 3a-bromo-2~-formyloxy-derivative(83) in high yield ;l the reaction provides a smooth synthesis of the 2fi,3b-epoxide by alkaline hydrolysis. The 2a,3a-bromonium intermediate (82), formed stereospecifically, similarly affords the 2b-01 nitrate (84)with silver nitrate in ~ y r i d i n e , ' ~

'

Me

Me

Me

(83) X = OCHO (84) X = ONO, ( 8 5 ) X = N,

or the 2~-azido-3a-bromo-derivative(85) when the reagent is bromine azide (generated from NaN, + Br, + HC1) in dichloromethane-nitr~rnethane.'~~ Further details have been published of the reaction between olefins and lead tetra-acetate-trimethylsilyl azide which, at - 20 "C affords 'a'-(axial)-azidoketones.'33 Fuming nitric acid at - 5 "C converted cholest-5-ene into a mixture of 6-nitro(87).' 34 The cholest-Sene (86) and, surprisingly, 5-hydroxy-5a-cholestan-6-one

latter product appears to arise z)ia attack by nitrosocium ion, derived from dissolved oxides of nitrogen. Deliberate nitrosation of cholesteryl acetate with

'" 132 133 134

J . Klinot, K . Waisser, L. Streinz, and A. Vystreil, Coil. Czech. Chern. Cornm., 1970, 35,3610. A. Hassner, F. P. Boerwinkle, and A. B. Levy, J . Amer. Chern. Soc., 1970, 92, 4879. E. Zbiral and G. Nestler, Tetrahedron, 1971, 27, 2293; see also ref. 96, p. 302. C . R. Narayanan, M. S. Parker, and M. S. Wadia, Tetrahedron Letters, 1970, 4703.

25 5


nitrous acid in acetic acid gave the 5a-nitrito-6-acetoximino-derivative(88), together with some 5a-hydroxy-6-ketone (89).’ Some minor products from the chlorination of cholesterol (90) in aqueous t-butanol are represented in Scheme 6 they presumably result from intermediate chloronium ions, as illustrated. Reaction of chlorine with cholest-4-ene-3P,bP-diol (91) gave the epoxyderivative (92) in a somewhat similar reaction involving 6,%hydroxy attack upon a 4a,5a-chloronium ion. P-Chlorocarbamates [e.g.(93)] result when olefins react

H

HO

C1

I

‘CI”

@ (91)

oH

TI+,1

J33

HO

(92)

Scheme 6

with N-chlorocarbamates in the presence of chromium(1r)ch10ride.I~ Addition of chlorine on to a 5-en-7-one7followed by dehydrochlorination with pyridine, affords the 6-chloro-5-en-7-one (94).’3 8

AcO NH-CO,R (93) 13’

136 13’

138

CI (94)

M. Onda and A. Azuma, Chem. Pharm. Bull. Tokyo, 1971, 19, 859. B. 0 . Lindgren and C . M . Svahn, Acta Chem. Scand., 1970,24,2699. J. Lessard and J. M . Paton, Tetrahedron Letters, 1970, 4883. R. A. LeMahieu, A. Boris, M. Carson, and R. W. Kierstead, J. Medicin. ChPm., 1971, 14. 291.


256

Unlike simple steroidal 5-enes, the 6-methyl derivative (95) reacted with chlorine to give a mixture which afforded the labile allylic chlorides (96) and (97).

+

AcO

Me

CH2

Me

An intramolecular addition occurred 40 when the bisnor-chol-7-en-22-01(98) reacted with PBr, in CHCl, . The product was the 14P,22-ether(loo),apparently (99) followed by resulting from acid-catalysed olefinic bond migration to addition of the alcoholic group.

YCH20H

3P-Acetoxylanost-9(11)-ene reacted in unexpected fashion with hypobromous acid. Instead of a bromohydrin, the product was the 7cx-brorn0-8-en-ll-one.~~~ The detailed mechanism is unknown, although there is a clear resemblance to the conversion of 7-enes into 7,1l-disubstituted-8-enes. Attempted preparation of acetonides from a 4a-methyl-4~,6a-diol, or the analogous ba-methyl-4a,6P-diol, also gave cyclic ethers resultipg from intramolecular electrophilic addition to an olefinic bond (see p. 247). Comparable cyclizations occurred on solvolysis of the 5,lO-secosteroid (101)-+(102) (see also Part 11, Chap. 2, p. 404),142and in the synthetically useful formation of

"' R. A . LeMahieu,

A. Boris, M. Carson, and R. W. Kierstead, J . Medicin. Chem., 1971,

14, 629. 140

'"

D. J. Aberhart and E. Caspi, J . Chem. SOC.(C), 1971, 2069. I . G. Guest and B. A. Marples, J . Chem. SOC.( C ) , 1971, 1468. M . Lj. Mihailovic, M . Davobic, Lj. Lorenc, and M . GaSic, Tetrahedron Letters, 1970, 4245.


257

AcO

Ac 0

I

co

(104)

(four isomers)

isomeric pregnan-20-ones (104) from the 13,17-seco-intermediate (103).143 Neighbouring-group participation by suitab!e 3P-amido-substituents, when the

MeN

I

co

OH

I

CF, 143

(106)

P. T. Lansbury, P. C . Briggs, T. R. Demmin, and G . E. DuBois, J . Amer. Chem. Soc., 1971,93, 1311.

258


As -unsaturated compound [e.g. (lOS)] (or its A4-isomer) was treated with trifluoroacetic acid, resulted in stereospecific introduction of a 5P-hydroxy-group ( 106).'44 The 3a-amido-derivatives similarly afforded 5a-hydroxy-products.

Other Addition Reactions.-A quantitative study of the epoxidation of 3substituted cholest-5-enes with peroxy-acid shows that both the rate and the epimer ratio vary according to the C(3)-s~bstituent.'~' The epoxidation clearly has some electrophilic character. o-Sulpho-perbenzoic acid, which may be used in aqueous-organic solvents, converted cholesterol efficiently into the a-epoxide (89 %). 1 4 6 The A' 6-olefinic bond in cholestan-5,16-dien-3P-o1 is sufficiently reactive, perhaps as a consequence of ring strain, to permit selective 16a,17aepoxidation.14' A mixture of lead tetra-acetate and hydrogen fluoride, probably providing PbF,(OAc), and PbF,, reacted with cholesterol (107) to give a complex mixture of products including compounds (108)--(111) (Scheme 7).14' The preferred though admittedly speculative mechanism involves an initial electrophilic plumbation as illustrated, followed by a variety of reactions, including skeletal rearrangement accompanying C-Pb bond cleavage. Cycloaddition of dichloroketen (from CC1,COCl and zinc) with ring-A olefins [e.g. (I 12)] afforded a regiospecific synthesis of fused cyclobutanones As dichloroketen generated from dichloroacetyl chloride-triethyl(113).'49-'5 amine failed to react with cholest-2-ene,15' it is suggested that a dichloroketenzinc complex has enhanced reactivity (electrophilic). Stereoelectronic effects, discussed in detail, are considered to control the specific formation of that cyclobutanone which possesses an axial C-CO bond, in preference to its isomer.'" Transposition of the 0x0-group to give the cyclobutanone (114) was achieved in high overall yield by the sequence illustrated (Scheme 8).l5' Photochemical addition of trifluoroiodomethane on to a 5a-steroidal-3-ene affords the novel 3a-trifluoromethyl-4[~-iodo-derivative (115).'5 2 The Diels-Alder reaction of diethyl azodicarboxylate with a 20-acetoxypregna16,20-diene(116)gave the heterocyclic derivatives (1 17 ;predominantly 16a),which were hydrolysed to the corresponding ketones (118)' 53 Tetrafluorobenzyne added across either C(lkC(4) or C(2bC(5)in the phenolic ether (1 19), giving( 120)and(121)respectively. Benzynes add in the normal manner to the 1(10),9(11)-diene(122) to give the la,lIa-substituted 9(10)-ene(123).lS4 '.IJ

"'

A . Ahond, A . Cave, C. K a n - t a n , and P. Poticr, Bull. SOC.chim. France, 1970, 3624.

K. D. Bingham, T. M . Blaiklock, R. C. B. Coleman, and G. D. Meakins, J . Chem. Soc. (C), 1970, 2330. '"' J . M . Bachhawat and N. K. Mathur, Tetrahedron Letters, 1971, 691. N . K. Chaudhuri. R. C . Nickolson, and M . Gu t , Steroids, 1970, 16, 495. J . Levisalles and J. Molimard, Bull. Soc. chiin. France, 1971, 2037. I J 4C. M . L. Cragg, J . Chenz. Soc. (C), 1970, 1829. 15' A . Hassner and V. R. Fletcher, Tetrahedron Letters, 1970, 5053. 15' A. Hassner, V. R. Fletcher, a n d D. P. G. Hamon, J . Atner. Chem. Soc., 1971,93, 264. '' A. F. Pascual and M. E. Wolff, J . Medzcin. Cheni., 1971, 14, 164. J . Yoshizawa and M. Tomoeda, J . Chem. SOC.(C), 1971, 1741. Is' I. F. Eckhard, H . Hcaney, and B. A. Marples, J . Chem. SOC.(C), 1970, 2493.

259


h

3 3

0

z

I

0

z

I

0

z ' f P

m

0

z

0

z


260

1

(i) reduction (ii) MeS0,CI

Scheme 8

,CO,Et

*(XI H

CF3-

I

NCO,Et NC0,Et

26 1


'8 F

Me0

a 1

F\

F ___+

OMe

RO

RO

The reaction between 5,7-dienes [or 5,7,9(11)-trienes] and tetracyanoethylene afforded novel dehydrogenated ene-adducts (124).' 5 5 Diels-Alder addition to a 7,14-diene afforded the 7a,15a-cyclic adduct (125).

Difluorocarbene reacts with 2-benzylidene-5a-cholestan-3-one(126) to give the furan (127).156A rearrangement step is clearly required, although its nature is somewhat uncertain.

(126) ' 5 5 156

(127)

A. L. Andrcws, R. C . Fort, and P. W. Le Quesne, J . Org. Chem., 1971, 36, 8 3 . M . Derenberg and P. Hodge, Chem. Comm., 1971, 2 3 3 .

262


Conjugate hydrocyanation of '@-unsaturated carboxylic esters, acid chlorides, and acyl cyanides, is effected by reaction with diethylaluminium cyanide [e.g. (1 28) -P (1 29)]. 5 7 The acid chloride and cyanide are the most reactive. A double conjugate addition of nitromethane with a 2-methylene-4-en-3-one (130) gave a

H2ca: 0

2,5-bridged adduct, considered on the basis of its rotatory dispersion to have the configuration illustrated (132).' 5 8 1,4-Addition of hydrogen chloride or bromide on to a 6-nitro-5-ene (133) apparently forms the nitronic acid (134), which is reduced to give the 5-halogeno-6-oximino-derivative (135).' 5 9 Intramolecular

I

I

-

R NOH

(133)

(135)

X = C1 or Br 15'

L58 Is'

W. Nagata, T. Okumura, and M. Yoshioka, J . Chern. SOC.( C ) , 1970, 2347. M . Kocor and W. Kroszczyhski, Tetrahedron Letters, 1970, 5143.

Y . Komeichi, S. Tomioka, T. Iwasaki, and K . Watanabe, Tetrahedron Letters, 1970, 4677.

Steroid PrGperties and Reactions

263

conjugate addition of a 9a-acetamido-group on to a 4-en-3-one (136) occurs very readily in acidic media [e.g.during acetal formation at C(3), or oxidation of 1lflOH by chromium trioxide in acetic acid] to give derivatives of the type (137).99 Under Reformatsky conditions (Zn or Mg), a pregn-16-en-2O-one (138) reacts with a-bromo-isobutyric ester to give the enol lactone (139):160 the key step must be nucleophilic attack of an incipient carbanion on the steroidal C(16) position,

Me

I

is

0

II

+

C-OEt -Me

'

Br

/c\

Me

H (138)

H (139)

followed by lactonization. Alkaline hydrolysis of the lactone afforded the 16aalkylated 20-ketone (140). The pregn-16-en-20-one (138) ('pregnadienolone' acetate) also reacts with potassium thiocyanate in acetic acid at 100°C to give the 16a-thiol acetate, the product of conjugate addition of thiolacetic acid.'61 Earlier work had demonstrated the formation of a mixture of 16-isothiocyanates, epimeric at C(16)and C(17),when a similar reaction was performed at 60 "C. The formation of 16-isothiocyanates from the 16-en-20-one appears to be reversible, and it is now suggested that the 16-en-20-oneis regenerated and subsequently adds thiolacetic acid, derived from potassium thiocyanate at 100 "C by acetolysis oia thioacetamide.

'" 16'

C. Gandolfi, G. Doria, M. Amendola, and E. Dradi, Tetrahedron Letters, 1970, 3923. A. A. Akhrem, Z . I. Istomina, A. I. Kuznetsova, and A. M. Turuta, BuU. A c a d Sci. U . S . S . K . , 1970, 1941.

264


Reduction of Unsaturated Steroids.-Twenty-five steroidal 4-en-3-ones, variously substituted at C(l l), C(17), and C(20),have been hydrogenated over a pre-reduced palladium catalyst. ' 6 2 Wide variations in the 5a : 5P ratio, depending upon substitution and solvent, augment the large collection of data already available in this field.'63 It is suggested'" that hydroxy-groups exert their directive influence largely by electronic rather than by steric effects. Catalytic hydrogenation of the 4,6,8(14)-trien-3-one(141)reduces the 4,6-diene system selectively to give the 5p-8( 14)-en-3-one (142).' 64 Palladium-calcium R

R

H (142)

(143)

carbonate, deactivated by pyridine, has been used to reduce the la,2a-epoxy-4,6dien-3-one (143) selectively at the A6-olefinic bond.'65 Prolonged reaction opened the epoxide ring to give the la-hydroxy-4-en-3-one (see also Part 11, Chap. 2, p. 398). Bis(pyridine)dimethylformamidedichlororhodium borohydride [from (py),RhC1,-NaBH, in DMF] is an effective catalyst for homogeneous hydrogenation of steroidal 4-en-3-ones.' 66 Cholest-4-en-3-one and 17a-methyltestosterone gave predominantly the SP-3-ketones, whereas testosterone and progesterone afforded the Sa-isomers (ca. 78 %). Data for solid rhodium catalysts are given for comparison. A polymer-supported rhodium catalyst reduced cholest-2-ene only slowly.' 67 A mixture of trifluoroacetic acid and triethylsilane (or triphenylsilane) reduces olefinic bonds.'68 The silane transfers hydride ion to a carbonium ion, resulting from protonation of the olefin. 3-Methyl-5a-cholest-2-ene gave 3P-methyl-%cholestane. Carbonium ions prone to rearrangement may isomerize before being reduced, suggesting that hydride transfer is the slow step. Tertiary alcohols are also reduced to hydrocarbons via the corresponding tertiary carbonium ions. Conditions normally suitable for Leuckart-Wallach reductive amination (90 formic acid and N-methylformamide at 170 "C)converted 4-azacholest-5-en-3-one (144) into the 5a-saturated lactam (145).'6 9 A 17-0x0-group was simultaneously converted into the 17~-N-methylformamido-derivative. 162

163 Ib4 165

'66

lh4

K. Mori, K. Abe, M . Washida, S. Nishimura, and M. Shioto, J . O r g . Chem., 1971, 36, 231. Ref. 87, pp. 83-86;

Ref. 96, p. 308. B. Pelc and E. Kodicek, J . Chem. Sac. (C), 1971, 859. B. Pelc and E. Kodicek, J . Chern. SOC.(C), 1971, 1568. P. Abley, I . Jardine, and F. J. McQuillin, J. Chem. SOC.(C), 1971, 840. R. H. Grubbs and L. C . Kroll, J. Amer. Chern. Sac., 1971,93, 3062. F. A. Carey and H. S. Tremper, J. Org. Chenz., 1971, 36, 758. N . J. Doorenbos and W. E. Solomons, Chem. andZnd., 1970, 1322.


265

H (144) A5 (145) 5a - H

The 11-0x0-oestrapentaene (146) is selectively reduced by sodium-ethanol to give the 6,7-dihydro-derivative (147)."' Oxidation and Dehydrogenation.-Some novel oxidations of A5-olefinic steroids are reported. Silver oxide, in refluxing benzene or toluene, converted cholesterol into the 6,6'-dimeric compound (148),"' probably through a oneelectron process. The dimer decomposed above its m.p. to give an equimolar

mixture of cholest-4-en-3-one and the 4,6-dien-3-one. Potassium permanganate and periodate in aqueous pyridine gave a complex mixture of products, variously oxidized in ring B.172 t-Amy1 or cumyl hydroperoxide, with molybdenum (149) pentachloride, oxidizes cholesterol to give Sa-hydroxycholestane-3,6-dione in high yield. l 7 Cholest-4-en-3P-01similarly gives the 4-en-3-one. 17'

17'

D. K. Banerjee, E. J. Jacob, and N . Mahishi, Steroids, 1970, 16, 733. G. Stohrer, Steroids, 1971, 17, 587. H. R. Nace and A. L. Rieger, J. Org. Chem., 1970, 35, 3846.

266


A novel and efficient process for the allylic oxidation of A5-olefinsto give 5-en-7ones involves U.V. irradiation of a solution in cyclohexane or t-butanol, containing 1-2 molar proportions of mercuric bromide, in an open quartz v e ~ s e 1 . ICyclo~~ hexene and tetralin are oxidized to give cyclohexenone and a-tetralone, respectively. Since 7a-hydroperoxy- and 7-hydroxy-cholesteryl acetate gave 7-0x0cholesteryl acetate in high yield, it is thought likely that the reaction proceeds by attack of oxygen at C(7) on an allylic free radical, derived by initial hydrogen abstraction from C(7). The dehydrogenation of ring D in neoergosterol with dichlorodicyanoquinone (DDQ), mentioned last year,'74 has been reported in full.'75 Oestrone methyl ether, and similar compounds, give first the 9( 11)-dehydro-derivatives (150) on reaction with DDQ.'76 Continued reaction with an excess of DDQ in dioxan afforded the 12a-hydroxy-derivative (152),which was further oxidized to give the 12-ketone in low yield. ' 7 6 A similar reaction occurred in wet benzene as solvent, although the 12a-hydroxy-intermediate was then trapped by a second steroid molecule to give the 12,12'-anhydro-dimer (154).'77 Benzene, containing 3 %

R'O

(152) R' (153) R' (154) R'

= = =

H Me another steroid 12a-yl group

methanol, gave the 12a-methoxy-derivative (1 53). These reactions provide evidence for hydrjde abstraction from C(12), presumably with electromeric assistance from the 3-methoxy-substituent. The resulting mesomeric cation (151) 17' 174 176

N . Friedman, M. Gorodetsky, and Y. Mazur, ChPrn. Comm., 1471, 874. Ref. 96, p. 315. W. Brown and A. B. Turner, J . Chern. SOC.(C), 1971, 2057. H. Dannenberg and A . Bodenberger, Naturwiss., 1971, 58, 96. J. Ackrell and J. A. Edwards, Chem. and Inn., 1970, 1202.

267


is stable enough to survive until attacked by traces of water or methanol, or by the 12a-hydroxy-steroid, once this has been formed in significant amount. DDQ dehydrogenation of '9( 11)-dehydro-methyltestosterone', with acidic catalysts, afforded a phenanthrene derivative in a stepwise process (see p. 312). Selenium dioxide dehydrogenated the 3P-fluoro-oestra-5(10)-ene(155) to give the 5(10),9(11)-diene(1 56).' l8 Allylic oxidation of a 3P-fluoro-A5-steroid (157) with selenium dioxide gives the 4P-hydroxy-derivative ( 160).'78 A similar oxidation is well known in 3P-hydroxy-A5-steroids, where the 3p-hydroxy-group

(157)

(155) 9a-H (156) A9(11)

obviously hasno special role in deciding the site of attack. The oxidation mechanism discussed last year' 7 9 is capable of explaining this regiospecificity: electrophilic addition of a selenium species in the Markovnikoff sense to the olefinic bond (158), and proton loss from C(4), would afford the derivative (159), transformed by allylic substitution into the 4P-hydroxy-5-ene. Such a mechanism would be quite distinct from the free-radical process involved in allylic oxidations at C(7) (see p. 266). Neoergosterol is available in a single step by treating ergosterol with dibutyl peroxide in n-decane under reflux.' A 2-methoxy-4-methyloestra- 1,3,5(10)triene (161)is oxidized by peracetic acid to give the 5f3-hydroxy-1(10),3-dien-2-one 17'

M. Mousseron-Canet, C. Chavis, and A. Guida, Bull. SOC.chim. France, 1971, 627. Ref. 96, p. 312. W. H . Schuller and R. V. Lawrence, J . Medicin. Chem., 1971, 14, 466.


268

\ Me0

Me (163) R (164) R

= =

Me CHO

(165) X = H, OAc, or C1

(166)

(162),18' in a reaction with obvious mechanistic similarity to the cxidation of enol ethers with peroxy-acids.18' Ceric ammonium nitrate in acetic acid, which converts a 1-methyl-oestrogen methyl ether (163) into the 1-aldehyde (164), had a different effect upon 4-methyloestra-l,3,5( lO)-trienes, and their 1-acetoxy- or 1chloro-derivatives (145). Preferential oxidation at C(6) gave the 6P-acetoxyderivative (166), although the I-methoxy-analogue underwent some attack (25 %) on the 4-methyl group to give the a l d e h ~ d e . " ~ Miscellaneous Reactions.-Friedel-Crafts acetylation of an oestra- 1,3,5( 10)triene (167)occurs regioselectively at C(2)( 168),Ig4suggesting that para-activation

Me0

& '

(167) R = H (168) R = AC

I

F (1 72) "I

'" IM3 184

R (173) R = H (174) R = CH,NEt,

G. F. Burkinshaw, B. R. Davis, and P. D. Woodgate, J . Chem. SOC.( C ) , 1970, 1607. D. N. Kirk and J. M. Wiles, Chem. Comm., 1970, 1015; see also ref. 96, p. 336. D . M . Piatak and L. S. Eichmeier, Chem. Cornm., 1971, 772. T. Nambara, S. Honma, and S. Akiyama, Chern. and Pharm. Bull. (Japan), 1970, 18, 474.


269

(hyperconjugation?) by the secondary C(6) is more effective than by the tertiary C(9), which would have activated C(3). The analogous 3-methoxy-compounds are also acetylated selectively at C(2). Reaction of the 3-methoxy-compound (169) with fluoroxytrifluorornethane (CF,OF ; a source of electrophilic fluorine), (170)and the 4-fluoro-aromatic ether however, gave the 10P-fluoro-1,4-dien-3-one (171),further transformed into the 4,lOP-difluorodienone (172).’” It is not clear why C(2) is not attacked by this reagent. Condensation of the acetylenic compound (173) with formaldehyde and diethylamine, catalysed by copper(I1)acetate, gave the diethylaminomethylacetylene (174).ls6 The derived quaternary salt undergoes an interesting cyclization (see p. 290). 4 Carbonyl Compounds

Reduction of Ketones.-Selective reduction of the 20-0x0-group in 17a,21 dihydroxy-pregn-4-ene-3,20-dione (Reichstein’s S) with sodium borohydride in methanol at 0 “C gave the 208- and 20~-01sin ratio 2 : 1,187contrary to an earlier report that the 2OP-01 is formed almost exclusively. 5-Cyano-substituents exert a powerful influence on the reduction of 3-0x0steroids by hydride donors. An unambiguous assignment of stereochemistry at C(3) in the resulting alcohols corrects earlier reports. It is now clear that the 5cyano-group [either 5a- (175)or 5p- (178)]exerts both a polar and a steric influence on the approach of the reagent towards the carbonyl group, resulting in preferred attack on the side opposite the cyano-substituent.lS8 Sodium borohydride, therefore, generally favours production of the axial alcohols (176)and (179),but the

H

0-

c

\\ NH

(177)

‘” ’**

C. Chavis and M. Mousseron-Canet, Bull. SOC.chin]. France, 1971, 632. H. Kaufmann and J. Kalvoda, Chimia (Swifz.), 1971, 25, 248. A. A . Akhrem and T. K. Ustynyuk, Bull. Acud. Sci. U . S . S . R . , 1970, 1718. W. Nagata, T. Wakabayashi, M. Narisada, and Y . Hayase, J . Chern. SOC.( C ) , 1971, 2415.


4

NeK HO

H

(179) (major product)

more bulky lithium tributoxyaluminium hydride shows even greater selecrlvity, giving ca. 800/, of axial alcohols. The two axial alcohols were characterized by their reactions with toluene-p-sulphonic acid, giving the water-soluble salts of the imino-lactones (177) and (180),from which the cyaro-alcohols were regenerated by the action of base. Attempts to reduce the 3-0x0-group in the ketol pyridinium sulphate (181), using sodium borohydride in anhydrous methanol, afforded instead the stable 3,3dimethoxy-derivative (182),'s9 The pyridinium ion is sufficiently acidic to promote this acetalization, although alkali-metal salts of the ketol sulphate were

reduced normally. The 3a : 38 ratio of the resulting 3-hydroxy-derivatives varied markedly, however, according to the cations in solution. Sodium borohydride reduces 4-hydroxycholest-4-en-3-one(183) to give 5acholestane-3fi,4/3-diol(184).I9O Lithium aluminium hydride, in contrast, reduces only the 3-oxo-group, the 4-en-4-01 system being protected as its aluminate derivative (185). Hydrolysis during working-up affords the 3P-hydroxy-4-ketone (186). The enamino-ketone (187) is almost inert to sodium borohydride, but is reduced by lithium aluminium hydride to give a mixture of ketols (186) and (190). 189

D. Baxendale, D. N. Kirk, M. S. Rajagopalan, and A. B. Turner, J . Chem. SOC.(C), 1971, 2563. C. H. Robinson, L. Milewich, and K. Huber, J . Org. Chem., 1971,36, 211.


27 1

LIAIH,

7 -

r

1

a 1

0

‘

H

OH

It appears possible that initial formation of the reduced complex (188) could lead, during hydrolysis, to the ene-diol(l89) and finally to the ketol mixture. Ethyleneacetals of some oxo-steroids are reduced by lithium aluminium hydride-aluminium chloride to give 2’-hydroxyethyl ethers.’ 91 Polarographic reduction of unsaturated 3-, 7-, and 20-oxo-steroids in aqueous D M F has been studied.192The ease ofreduction increases with extended conjugation, as in a 4,6-dien-3-one. I” 192

M . S. Ahmad a n d S . C . Logani, Austral. J . Chem., 1971, 24, 143. N. Shinriki and T. Nambara, J . Pharm. SOC.Japan, 1971, 91, 5.

272


Other Reactions at the Carbonyl Carbon Atom.-Acetal formation from 3-0x0steroids in anhydrous alcohols is far more effeciive than was previously realised. C.d. studies revealed almost quantitative conversion of 5a- or 5P-3-0x0-steroids in methanol into their 3,3-dimethoxy-derivatives, and ethanol or propan-2-01 also gave considerable proportions of the corresponding acetals.' 9 3 Contrary to earlier belief, even the hindered 2-0x0-group gave the 2,2-dimethoxy-derivative (73% at equilibrium). Acetal formation was drastically reduced by traces of water, however, or by alkyl substitution adjacent to the 0x0-group.

CH,'S'H

1

CH,OH

(193)

1

I

CH,OH

(195)

IH+

CH,~H

(196)

Ethylene acetal formation from a 4-hydroxy-4-en-3-one (191) gave the 3 3 dieno[3,4-b]-dioxan (192), as well as the expected 3,3-ethylenedioxy-4-ketone (193).17 The abnormal product (192) was also obtained from the 4P,Sfi-epoxy-3ketone (194). Similar reactions, using 2-hydroxyethanethiol instead of ethanediol,

resulted in isomeric oxathians (195) and (196)from the two steroidal ketones. Two distinct mechanisms are proposed, each involving an initial nucleophilic attack by sulphur as illustrated. 3-0x0-steroids give their epimeric hemithioacetals in equal proportion^,^^ contrary to an earlier report. The epimers were distinguished by their n.m.r. properties (see p. 238). Thioacetals (197) are smoothly converted 193

L. H. Zalkow, R. Hale, K. French, and P. Crabbe, Tetrahedron, 1970, 26, 4947.


273

into their parent ketones (199)by treatment with l-chlorobenztriazole at to give the di-sulphoxide (198), followed by alkaline h y d r o l y ~ i s . ' ~ ~

- SO "C

Oxidation and Dehydrogenation.-A stirred solution of a 14P-androstan- 17-one (200) and potassium t-butoxide is oxidized to give the 16-hydroxy-15-en-17-one (201), presumably via attack of atmospheric oxygen upon the enolate anion.' 9 s The neutral A' 7(20)-enol(203),which may be obtained by careful '1,4'-hydrogenation of a pregn-16-en-20-one (202) over a neutral palladium catalyst, also absorbs I

Me

I

co

Me Me -OH

I

H

oxygen when stirred in benzene solution.196 The product is the 17a-hydroperoxy20-ketone (204), although yields are good only in the 16-methyl derivative (R = Me). New applications of dichlorodicyanobenzoquinone (DDQ) for the dehydrogenation of ketones include the conversion of a 4,7-dien-3-one into the 4,6,8(14)trienone, which is further dehydrogenated to the 1,4,6,8(14)-tetraenone with acidic c a t a l y s i ~ . ' The ~ ~ dehydrogenation of a 4-en-6-one with DDQ affords the 2,4dien-6-one.' 97 Selective 1,2-dehydrogenation of 5a-cholestan-3-one has been achieved with palladium acetylacetonate and oxygen. 98 Possible alternative mechanisms are discussed. Baeyer-Villiger oxidation (perbenzoic acid) of some 5a-cholestan-6-ones, including the 3a,5-cyclo-analogue, gave the 6-oxa-~-homocholestan-7-ones,with exclusive migration of C(5) rather than C(7) in each case'99 (cf Beckmann Rearrangements, p. 281). P. R. Heaton, J. M. Midgley, and W. B. Whalley, Chem. Comm., 1971, 750. T. Nambara, H. Hosoda, M. Usui, and T. Anjyo, Chem. and Phurm. Bull. (Japan), 1971, 19, 612. 1 9 6 P. R. Enslin, Tetrahedron, 1971, 27, 1909. I g 7 J. R. Hanson and T. D. Organ, J . Chem. Soc. (C), 1970,2473. I g n R. J. Theissen, J . O r g . Chem., 1971, 36, 752. 19' M. S. Ahmad, Shafiullah, M. Mushfiq, and M. Asif, Zndiun J . Chem., 1970,8, 1062. ly4

274


‘Dimethisterone’ (205) suffers auto-oxidation to give 6r- and 60-hydroxy(206) and hydroperoxy-derivatives, as well as in the side-chain to produce a terminal carboxylic acid group (207).200 Enolization and Related Reactions.-The influence of various substituents [Me, Ph, or CN at C(2); alkyl at C(6)] on the enol acetylation of steroidal 4-en-3ones has been further investigated.201 Under thermodynamic control (Ac20HBr), a conjugating substituent at C(2) causes formation of quite appreciable proportions of the A2,4-dienol acetate, although A3*5-dienolderivatives are normally favoured. A A9(l ‘)-olefinicbond enhances formation of the A2,4-dienol, apparently through coiiformational transmission.202 The enol acetylation of a 3-oxo-SP-steroid, normally favouring the A3-enol,shifts strongly in favour of the A2-enol when a A7-olefinicbond is present. Conversely, a 5a-7-en-3-one gives a 1 : 1 mixture of the A2- and A3-en01 acetates with isopropenyl acetate (kinetic control), instead of showing the normal very marked preference for A2-enolization in the 4-Chloro-4-en-3-ones (208) form dienolic ethers (209) normally, but steric strain apparently inhibits enol-ether formation from the 4,6~-dichloro-compound(2lO).’

(208) X = H (210) x = c1

Partial deconjugation of a 19-hydroxy-4,6-dien-3-one (211) was effected with sodium methoxide in DMSO, followed by careful a~idification.~’~ The product, the 4,7-dien-3-one (21 2), was readily transformed into the 8@,19-oxido-4,6-dienone (213) on mild oxidation. Attempted deconjugation of the 4-chloro-19-hydroxydienone (214), however, gave the free A3?5,7-trienol (215), which could not be ’(”

”” lo3 2 04

R. F. MaJewski, J . M . Berdahl, L. D. Lost, T. A. Martin, J . C. Simms, J . G. Schmidt, and J. R. Corrigan, Steroids, 1970, 16, 15. P. Toft and A . J . Liston, J . Cherrr. Soc. (C), 1971, 439. P . Toft and A. J . Liston, Tetrahedron, 1371, 27, 969. A. J. Liston, P. Toft, P. Morand, and H. Stollar, J. Chetn. SOC.( C ) , 1970, 2121. G. Kruger, J . O r g . Chenr., 1971, 36, 2129.

275


*B

0’

X (211) x (214) X

= =

H C1

-&10

transformed efficiently into the 4,7-dienone, but was dehydrochlorinated in pyridine to give the 8P,19-oxido-dienone (213).204 A study of rates of amine-catalysed epimerization of 2P-methyl-Sa-cholestan-3one showed that deprotonation at C(2) is rate controlling.205 The kinetics and direction of enol acetylation at C(3) in the ketone (216) have been studied.206

COMe

H2C

COMe

(218) *05

’06

T. A. Spencer and L. D. Eisenhauer, J. Org. Chem., 1970, 35, 2632 G. Langbein and E. Stutter, J. prukt. Chem., 1970, 312, 221.

276

Terpenoidsand Steroids

18-Benzoylaminoprogesterone (217)reacts with benzoyl chloride in pyridine to give the internal en-amide (218) (cf. p. 295).207 Reactions of Enolate Anions.-Base-catalysed methylation of 5fl-cholest-7-en3-one gave the 2P-methyl and 2,2-dimethyl derivatives,208in accordance with the preferred enolization towards C(2), resulting from conformational transmission from the A7-olefinic bond (see p. 243). Intramolecular alkylation occurs when the tosylate of a 4-hydroxymethyl-4-methyl-3-oxo-steroid (219) or (224) is treated with base.209 Expulsion of the tosylate group by a A2-enolate ion (220).is followed by skeletal rearrangement to give a bicyclo[3,2,0]heptanone analogue [(222) or (225), respectively] with stereochemistry controlled by the original configuration at C(4). To explain the stereospecific rearrangement, it is suggested that a non-classical intermediate (221) undergoes the rearrangement indicated (Scheme 9 ;path fi). The reaction differs significantly from the known behaviour of

fl

base

,

a

0

0 H,C

Me

I

j H Me

n

-0

: H Me

(223)

TsO (224)

I

-

H

O

q

p

/

Me

Me H

(225)

Scheme 9 207 208

A . Kasal, COD. Czech. Chem. Comm., 1970, 35, 3821. P. Morand, J . M . Lyall, and H. Stollar, J . Chem. SOC.( C ) , 1970, 2117. Y . Tseuda, T. Tanno, A. Ukai, and K. Isobe, Tetrahedron Letters, 1971, 2009.

1


277

the monocyclic (cyclohexanone) analogue, which is partially transformed via path b into the more normal bicyclo[3,l,l]heptanone (223). Steric hindrance from the lop-methyl group is thought to prevent reaction b in the steroid. A novel reductive alkylation of 'a,B'-epoxyketones, so far applied only to epoxycyclohexanone derivatives,210offers promise as a route to 'a'-alkylated steroidal enones (e.g. 4-methyl-4-en-3-ones). The preparation of 4-methyl-4-en-3-ones by thiomethylation, followed by desulphuration, has now been extended to a 5-en-7-one7 as a route to its 6-methyl d e r i ~ a t i v e . ' ~ ~ A further paper elaborates upon the internal aldol condensation of an 8,9-seco8,9,1l-trione to give so-called 'linear' steroid analogues.21 Acrylonitrile alkylates the enolate anion derived from a pregn-l7(20)-en-21-al (226), or similar compounds, to give 16-cyanoethyl derivatives (227), which CHO

CHO

I

yJ

1

CH

+

c/cy

H

+

(228)

H (229)

undergo aldol cyclization under the reaction conditions to give pentacyclic compounds of the types (228) and (229).213-216 A novel procedure for the selective mono-alkylation of ketones is described on p. 291.217 J . D. McChesney and A . F. Wycpalek, Chem. Comm., 1971, 542. Ref. 96, p. 358. S. Aoyama, Chem. and Pharm. Bull. (Japan), 1971, 19,896. 2 1 3 S. Bory and C. R. Engel, Bull. SOC.chim. France, 1970, 3043. 2 1 4 C. R . Engel and J. Lessard, Canad. J. Chem., 1970, 48, 2819. 'I5C. R. Engel, V. S. Salvi, and L. Ruest, Canad. J . Chern., 1970, 48, 3425. 2 1 6 C. R. Engel and L. Ruest, Canad. J . Chem., 1970,48, 3136. 'I7 J . Hooz, D. M . Gunn, and H . Kono, Canad. J. Chem., 1971,49, 2371. 'lo

'I1

278


Reactions of Enol Derivatives and Enamines-Enolic acetates are transformed into fluoro-ketones by the action of fluoroxytrifluoromethane, a source of electrophilic fluorine (Scheme lo)." Peroxylauric acid reacted with the enol

H

(233) (234) (235) (236)

X X

--' XX

= = = =

F OAC 0-Lauryl OH

Scheme 10

acetate (232) of a 17-0x0-steroid to give the 16a-acetoxy-17-ketone (234), in a reaction involving acetyl migration.218 A minor product, however, was the 16alaurate (235). The precise mode of incorporation of the lauryl group is not clear. rn-Chloroperbenzoic acid afforded the 16a-hydroxy-17-ketone (236) in a single step. Enol ethers (237) of androstan-17-ones add dichloro- or dibromo-carbene, but the initially formed cyclopropane derivatives (238)are too unstable to isolate, being rapidly rearranged to give the 17-halogeno-~-homoandrost-16-en-17a-ones (239)(see also Part 11, Chap. 2, p. 346).2'9 Carbene addition, using diethylzinc and di-iodomethane, gave the cyclopropane (240) which is reasonably stable, although strong acids rearranged it to give the ~-homo-17a-ketone(241). With bromine or iodine, the o-homo-16-en-17a-one (242) was formed, but an excess of bromine gave the 17-bromo-enone (239 ;X = Br). The enol acetate (243)similarly formed a methylene adduct, which afforded the cyclopropanol(244) with lithium aluminium hydride. The cyclopropanol rearranged with base to give the D-homoketone (241), or with acid to give the 16a-methylandrostan-i7-one (245; 60%) as well as the D-homo-ketone (241 ;40 The cyclic enolic ether, furostene (246), derived from diosgenin. was opened with substitution at C(16) by aqueous hydrobromic acid, to give the 16-bromocholestan-22-one (247).220

x).

ZIR

219 "O

M. G . Combe, W. A. Denny, G. D. Meakins, Y . Morisawa, and E. E. Richards,

J. Chem. Soc. ( C ) ,1971, 2300. W. F. Johns and IS.W. Salamon, J .

O r g . Chrtn., 1971, 36, 1952. G. A. Smith and D. H. Williams, Chzni. Comm., 1971, 402.

279

Steroid Properties and Reactions OR

H (237) R (243) R

= =

1

,:OR

alkyl AC

y3

-

(238)

Br, _I_,

or I,

H (244)

The Reformatksy adduct (248), obtained from a pregn-16-en-20-one and abromoisobutyric ester (see p. 263),has the properties of an enolic ester."' Electrophilic bromination affords 17-bromo-20-ketones (249),with C(17) stereochemistry surprisingly depending upon the reagent (CuBr, in methanol -+ 17ct-Br; Br, --+ 17P-Br),as revealed by subsequent transformations. 221

C . Gandolfi, G. Doria, M. Amendola, and E. Dradi, Tetrahedron Letters, 1970, 3927.

280

Terpenoids and Steroids MeCO

H (249)

(248)

Enamines (250), or the corresponding ‘a’-hydroxymethylene ketones (25l), react with trimet hy lene dit hio tos ylate to give the trime t hy lenedit hioke tones (252), providing a convenient means of protecting reactive methylene groups from attack.222 The protecting group is readily removed by Raney nickel. The Vilsmeier reagent (Me,&=CHCl OPOCI,), which converts many ketones into vinylic chlorides, may react further to give ‘a’-formyl derivatives. The 3,5-dien-7one (253) affords the 7-chloro-2-formyl-2,4,6-triene(254). A 4,6-dien-3-one gave

H

the 3-chloro-2-formyl-2,4,6-triene (255); 2 2 3 although this represents the less favourable mode of enolization of the dienone (A3*’,’ is usual), unsaturated systems are known to isomerize under Vilsmeier conditions to permit formylation at an unhindered site.223“

0ximes.-Oximes of some saturated ketones are reduced by aqueous alkaline sodium borohydride under reflux to give the corresponding alcohols.224 Selective reduction of a 3,17-dioxime is possible, at C(3). ‘a’-Oximino-ketones afford diols. Diborane, in contrast, reduces oximes to give alkyl hydroxylamines ; a recent variant using sodium borohydride on silica gel in benzene gave the R. €3. Woodward, I. J. Pachter, and M. L. Scheinbaum, J . Org. Chenz., 1971,36, 1137. A. Consonni, F. Mancini, U. Pallini, B. Patelli, and R. Sciaky, Gazzetta, 1970, 100, 244. 2 2 3 a Ref. 96, p. 303. 2z4 K. H. Bell, Auslral. J . Chern., 1970, 23, 1415.

22-’


28 1

hydroxylamine N-borane derivative^.^^ Polarographic reduction of the 0methyloximes of testosterone and other steroids, proceeds in two distinct steps, passing via the imino-derivatives to the amines.226 The Beckmann rearrangement of oximes may be induced by means of dicyclohexylcarbodi-imide and trifluoroacetic acid in dimethyls~lphoxide.~~~ The oxime (256) of a 17-oxo-steroid gave the usual lactam (257) and seco-nitrile (258). The oxime of a 3a,5a-cyclo-6-ketone gives the ring-B lactam resulting from migration of C(5)227" (cf. p. 273). A kinetic study of the Beckmann rearrangement H

ON

C1

H

of 6cr-methyl-l7a-acetoxy-progesterone3-oxime employed cathode-ray polarography to estimate unreacted oxime.228 The photochemically induced Beckmann rearrangement has been studied (see p. 323). Reaction between 5a-cholestan-3-one oxime and chlorine affords the blue 3-chloro-3-nitroso compounds (259),the 3a-chloro-3~-nitroso-isomerpredominating.22 Other Nitrogen-containing Derivatives of Ketones-The Schmidt reaction of 6- and 7-oxo-steroids (with NaN,-polyphosphoric acid) gave the same lactams as the Beckmann rearrangements of the corresponding o x i m e ~ . ~ ~ ' Tosylhydrazones (260) of 3-oxo-steroids (and presumably of other suitable carbonyl derivatives) are reduced to saturated hydrocarbons (261) by sodium cyanoborohydride and toluene-p-sulphonic acid in dimethylformamide-sulpho-

225

F. Hodosan and V. Ciurdaru, Tetrahedron Letters, 1971, 1997.

"' N. Shinriki and T . Nambara, J . Phurrn. SOC.Japan., 1971, 91, 1 5 1 . A. H. Fenselau, E. H. Hamamura, and J . G. Moffatt, J . Org. Chem., 1970, 35, 3546. 2 2 7 a M. S. Ahmad, Shafiullah, and M. Mushfiq, Austral. J . Chem., 1971, 24, 213. 2 2 R A. P. Shroff and C. J . Shaw, Analyt. Cheiti., 1971, 43, 455. 2 2 9 G, Kresze, N. M. Mayer, and J. Winkler, Annalen, 1971, 747, 172. 2 3 0 B. Matkovics, Z. Tegyey, M. Resch, F. Sirokman, and E. Boga, Acta Chim. Acad. Sci. Hung., 1970, 66, 333.

227


282

N

r-r

I

+OP

L5

H-N H

HN

I

H

Tb-NH

Ts-NH

TS -NH

a H l

\I

(260)

I Scheme 11

lane.23 The cyanoborohydride has the advantage of being stable in moderately acidic media, allowing protonation of the hydrazone. Hydride transfer from the reagent is then facilitated, and is followed by decomposition of the primary product with expulsion of nitrogen (Scheme 11). Alkylation during the reaction of tosylhydrazones with an excess of alkyl-lithium has now been reported in full.232 Although limited amounts of alkyl-lithium favour olefin formation, it seems that a high concentration of alkyl-lithium favours the alternative attack upon the

(262)

23'

R. 0. Hutchins, R. E. Maryanoff, and C . A. Milewski, J . Amer. Chem. SOC.,1971,93,

13'

J . E. Herz and C. V. Ortiz, J . Chern. Soc. ( C ) , 1971, 2294.

1793.

283


hydrazone carbon atom of the anion (262), with expulsion of toluene-p-sulphonate ion. Loss of nitrogen and protonation then gives the alkyl-steroid (263) (see also Part 11, Chap. 2, p. 388). The (2’-hydroxyethyl)imino-derivative of a methyl ketone [e.g. of a pregnan20-one (264)]is halogenated selectively in the methyl group by hypohalous The absence of C=N absorption in the primary halogenated product suggests reaction uia the enamine (265), with hydroxy-group participation to give the heterocycle (266). Subsequent hydrolysis liberates the halogenated ketone (267).

CH,CI

I

CO

H

Nitrones (268),derived from ketones, undergo a Beckmann-like rearrangement when treated with tosyl chloride in ~ y r i d i n e Unlike . ~ ~ ~ the Beckmann rearrangement, however, the reaction is independent of the nitrone configuration, and shows a preference for vinyl migration. The consequence, for a 4-en-3-one derivative, is the formation of the unusual enamine-lactam (270). The hydroxylamino-0-tosylate derivative (269) is considered a likely intermediate (see also Part 11, Chap. 2, p. 353). Sapogenins: Reactions of the Spiro-acetal System.-Nitrous acid, provided by sodium nitrite and boron trifluoride in acetic acid, readily nitrosates the C(23) position in a spirostan (271), giving the oxime (272),from which the 23-0x0-derivative is available in high yield.235 Paraformaldehyde and acetic acid gave a 233

234

235

J . F. W. Keana and R. R. Schumaker, Tetrahedron, 1970,26, 5191. D. H. R. Barton, M. J . Day, R. H. Hesse, and M . M . Pechet, Chem. Comm., 1971,945. D. H. R. Barton, P. G. Sammes, M. V. Taylor, and E. Werstiuk, J . Chem. Soe. ( C ) , 1970, 1977.


284

mixture of somewhat complex products with modified side-chain structures of the type (273). Probable mechanisms have been suggested. Oxidation of the spiro-acetal system occurs more easily than has generally been realized. Chromium trioxide in aqueous acetic acid for six hours gave the diketoacid (274) in excellent yield.236 Alkylation of the methyl ester by Grignard

@o

.jJy-d O%O+

'

I

H

H

(273)

,SOP sop NOH

(272)

'."

NO

H

D. H. R . Barton, Y . D. Kulkarni, and P. G. Sammes, J . Chen?. SOC.( C ) , 1971, 1149.

285

Steroid Properties and Reactions 0

'R H (275)

/

reagents afforded a 16a,26,26-trialkylated spirostan analogue (276), derived by acetal formation from the dihydroxy-ketone (275). Resistance of the 22-0x0group to Grignard attack is noteworthy. Degradation of the alkylated compound (276) gives a new route to 16-alkyl-pregn-16-en-20-ones. Bromination of sapogenins at C(23) has been known for many years. The 23R-bromo (axial) isomer (277) may be separated from the resulting mixture, and has been found to dehydrobrominate smoothly with base to give the spirost23-ene derivative (278).237Base-catalysed isomerization of this olefin at 100 "C

237

W. H . Faul, A. Failli, and C. Djerassi, J . O r g . Chem., 1970, 35, 2571.

286


gives the more stable 24-ene (279). The derived epoxides, alcohols, and ketones have been prepared, as well as specifically deuteriated derivatives. Some by-products from the degradation of diosgenin to give 3fi-hydroxypregna-5,16-dien-2O-one have been identified.238

Reactions of Aldehydes, Carboxylic Acids, and their Derivatives.-1 7aHydroxy-3,20-dioxopregn-4-en-21-a1 isomerized in alkaline solution to give a mixture from which the C(20)-isomeric 17a,20-dihydroxypregnan-21-oicacids and the corresponding 17a-hydroxyetianic acid were isolated (cf‘. p. 327).23 Some reactions of 3,17-dioxoandrost-4-en-19-a1 (280) are described. Autooxidation eliminated the aldehyde group, leaving the lO~-hydroperoxy-oestr-4en-3-one (281), which was reduced and aromatized to give ~ e s t r o n e . * ~ A ’

J

retro-aldol reaction of the 19-aldehyde with alkali resulted in rupture of the C( lOkC( 19) bond to give the mesomeric anion (282), which may be protonated to give either the non-conjugated (283) or the conjugated enone (284). Another product obtained under alkaline conditions was the 2-hydroxymethylene-oestr-4en-3-one (286), resulting from intramolecular transfer of the formyl group, L. G. Gatsenko, V. I. Maksimov, and L. M. Alekseeva, Khirri.-Farmatseut. Zhur., 1971, 20. A. A. Akhrem and T. K. Ustynyuk, Bull. Acad. Sci. U.S.S.R., 1970, 1722. C. M. Siegmann and M. S. de Winter, Rcc. Trai:. chim., 1970, 89, 442.

lJ8

23y

240


287

apparently via an aldol-type reaction involving attack by a A2-enolate anion upon the aldehyde group as illustrated. Hydrocyanation of the unsaturated aldehyde (287) under mild conditions afforded the cyanohydrin (288), the product of kinetic control. More forcing conditions led to the 16a-cyano-aldehyde (289).241 A cyanohydrin is unstable only in a relatively hindered site such as C(20); attempted conjugate addition of cyanide to the unsaturated B-nor-aldehyde (290) was unsuccessful.

A 23-bromocholan-24-a1(291) affords the rearranged ketol(292) on hydrolysis with bicarbonate, but stronger alkali causes absorption of oxygen to give the ‘a’hydroxy-acids (293),isolated as their esters.242 A remarkable feature is that 90 % of this product mixture possesses the 23R configuration, the reaction presumably being influenced asymmetrically by hindrance from the remainder of the steroid molecule.

H

(293)

The 17a-hydroxy-etianic acid (294) forms an acetonide [‘dioxolone’ (295)] under catalysis by gerchloric acid.243 The same dioxoione was formed by oxidation of the triol-acetonide (296) with chromium trioxide-pyridine. The hydroxyacid (297)also forms a dioxolone. These derivatives are hydrolysed by alkali, but 241

242

”-’

W . Nagata, M. Yoshioka, T. Okumura, and M . Murakami, J . Chrin. Soc. ( C ) , 1970, 2355. Y . Yanuka, R. Katz, and S . Sarel, Tetrahedron Letters, 1970, 5229. M. L. Lewbart, J . Org. Chert?., 1971, 36, 586.


288

CO,H

I

H : Br

Br

are fairly stable in 60 % acetic acid. Some steroid lactones [e.g. (298)] can be converted into ortho-esters (299) with ethane diol under acetalization Cholan-24-oic acids ( 3 0 0 ) are oxidatively decarboxylated by lead tetra-acetate to give norchol-22-enes (301), which are further degraded by permanganateperiodate to give bisnorcholan-22-oic acids (302) in high overall yield.245

The reduction of the maleic anhydride adduct (303) with lithium aluminium hydride was previously reported to occur selectively to give the lactone (304).246 The lower selectivity now observed with sodium aluminium hydride (none at all with sodium borohydride) is interpreted as evidence for a complex (305) of the ester and anhydride carbonyl groups with a solvated lithium ion when lithium aluminium hydride is used, leading to selective reduction of the free carbonyl Sodium ions are considered not to form so stable a complex. The 5,7-dienol-lactone (306) reacts with Grignard reagents to give 4(5 -+6) abeo structures of the type (307), probably via an aldol condensation as illustrated.248 244 245

"'

24'

2'8

R. A . LeMahieu and R. W. Kierstead, Tetrahedron Letters, 1970, 5 1 1 I . J . W. Huffman and R. R. Sobti, Sferoids, 1970, 16, 755. Ref. 96, p. 348. D. E. Burke and P. W. Le Quesne, J . Org. Chem., 1971,36, 2397. J. Overnell and J. S . Whitehurst, J . Chem. Soc. ( C ) , 1971, 378.


289

(303) X = 0 (304)X = H,

(305)

The Dieckmann cyclization of the 2,3-seco-diester (308) gives the 3a-carbomethoxy-~-nor-2-one(309).249 Earlier confusion over the configuration at C(3) has been resolved by n.m.r. and 0.r.d. studies. The cyclization was effected with potassium t -butoxide in b e n ~ e n e - D M S 0 . ~ ~ '

R

1

Meo2 o& He

MeO,C

H

C0,Me

Miscellaneous.-The 2-hydroxymethylene-ketone (3 10) forms a reasonably stable crystalline mesomeric complex (311) by reaction with boron t r i f l ~ o r i d e . ~ ~ ' Reaction of the complex with methyl-lithium, followed by acid, gave the 2ethylidene-ketone (312), though in low yield. 249

250

25L

B. V. Paranjape and J. L. Pyle, J . Org. Chem., 1971, 36, 1009. H . R. Nace and J. L. Pyle, J . Org. Chem., 1971,36, 81. R. A. J . Smith and T. A. Spencer, J . Org. Chem., 1970,35,3220.

290


The substituted acetylenic ketone (313), formed by quaternization of the free amine (174) (see p. 268), undergoes internal condensation on reaction with potassium t-butoxide, to give the 3cr-ethynyl-~-nor-3P,5~-epoxide (314).‘86 A possible mechanism is illustrated.

CI --CH

It

MeNEt,

I cI+

(3 14)

MeNEtb

5 Compounds of Nitrogen and Sulphur Reactions of oximes and other ketonic derivatives are discussed in Section 4 (see p. 280). A 2b,3P-imino-5a-steroid (3 15) undergoes a variety of reactions, including those summarized in Scheme 12.2 The amine-oxide ( 3 16) reacts with trifluoroacetic anhydride in the manner depicted in Scheme 13.253The iminium ion (317) has the properties of a Mannich reagent, and alkylates a second steroid molecule in the enamine form (318) to give the 2-methylene-3-ketone (31.9) after hydrolysis. The demethylated amine was formed at the same time. Different behaviour of the A5-unsaturated amine oxide (320) results from participation of the olefinic bond in a novel type of Grob

‘” K . Ponsold a n d W. Preibsch, Cheiii. Ber., 1971, 104, 1752. 2 5 3

A. A h o n d , A . Cave. C. K a n Fan, a n d P. Potier, B d l . Soc.

chiiir.

Fruiicr,

1970, 2707.


29 1

Scheme 12

f r a g m e n t a t i ~ n Cleavage .~~~ of the C(3)-C(4) bond presumably leads to an allylic cation (321) which can accept trifluoroacetate to give the 3,4-seco-iminium ion (322). Products isolated after either hydrolysis, methanolysis, or hydrogenation of the 3,4-seco-intermediate are compatible with this mechanism (Scheme 14). Tosylation of the 16/?-dimethylaminopregnan-2O/?-ol (323) led to internal nucleophilic substitution at C(20)to give the azetidinium tosylate (3241,degraded by base to give the pregn-20-ene (325) (Hoffmann Reduction of the strained azetidinium ring with lithium aluminium hydride gave the 16gdimethylaminopregnane (326). Although the C(17)-oxiran (327) failed to react with most nitrogenous bases (NH, . NH,, phthalimide ion, urea, etc.), guanidine effected substitution at C(20), which was followed by expulsion of ammonia to give the oxazoline (328).*9 This compound could not be hydrolysed, but was deaminated by nitrous acid to give the 17P,2O-cyclic carbonate (329). A new and promising procedure for specific mono-alkylation of a cyclic ketone comprises treating the ‘a’-diazoketone [e.g. (330)] with a trialkylborane in the presence of water.* The ‘a’-monoalkyl ketone is obtained in yields exceeding 80 %. The preparation of 16p-ethyloestrone methyl ether (33I), illustrates the process, which appears to be limited only by the accessibility of the diazoketone, and the trialkylborane. The authors are cautious in commenting upon the

’’

254 255

A . Ahond, A. Cave, C. Kan Fan, and P. Potier, B u l l . SOC.chiivz. Frunte, 1970, 391 1. M. Heller and S. Bernstein, J . O r g . Chem., 1971, 36, 1386.

292


Me

I

Me

Me

I*y

Me-N

I

Me

I

Me

H

Scheme 13

mechanism : one possible sequence, involving an enol borinate, is illustrated here (Scheme 15). An ingenious application of the Bredt rule permitted the selective deuteriation of Sa-con-20(N)-ene(332)at the 17a- or the C(21)-positions(Scheme 16).256Direct base-catalysed exchange with solvent (MeOD + D,O) gave the 21-trideuterioderivative (333), a A’ 7(20)-olefinicbond being forbidden. Schotten-Baumann benzoylation of the 20(N)-ene opened the heterocyclic ring to give the 18-benzamidopregnan-20-one (334), which readily exchanged deuterium for hydrogen at both C(17) and C(21) (335). Acidic conditions re-closed the heterocyclic ring, 15’

G . Lukacs, A. Picot, L. Cloarec, A. Kornprobst, L. Alais, and X . Lusinchi, Tetrahedron, 197 I , 27, 32 15.

293


d h

P m 4 v

hZ +z-5

\

I

5

'0 g

0--u

T

-7

kz +z-2 0

/

5

$2; 8 8 E

E

f

5

\

% +z-2

+Z -


294 H

II

C

\\

N

35' ---CH2

H

o/c'

---CH, 'O

il

and the C(21)-deuterium was selectively removed by base to give the 17a-D compound (336). The site of dehydrogenation of the steroidal N-phenyl[3,2-c]pyrazoles (337) and (339) with DDQ appears to be controlled by the location of the phenyl group, and the attendant distribution of unsaturation in the heterocycle.257The indazole (338) was an expected product, but the isomeric 4,6-diene (340) resulted from the second pyrazole. The reviewer suggests that selective removal of a hydride ion from C(6) is favoured by extending conjugation in the intermediate (341). The spiro-pyrazoline-ketone (342) expels nitrogen when treated with either boron trifluoride etherate or acetic anhydride-pyridine, giving the cyclopropyl ketone (343).49 2a,Sa-Epithio-compounds have received considerable attention. Some solvolytic reactions are mentioned on p. 243, the oxidation of the 3-hydroxy-2a,5aepithio-compounds with lead tetra-acetate on p. 248, and the photolysis of a 30x0-derivative on p. 321. Alkalis cause elimination from the Scc-position in the 151

L. J . Chinn, J . O r g . Chem., 1971, 36, 1597

295


Scheme 15

D I

(332)

COPh I

(333)

COPh

I

J-$---y \

\ iii ; I (334)

(336) Reagents: i, D,O-MeOD-MeONa;

Scheme 16

ii, PhCOCl-base; iii, H

+


296

r

~

N,

DDQ,

Iili-Jy ' I

y9.J Ph

\

Ph

(337)

(338)

H"

(339)

(340)

\Q

0

0

H

H (342)

(343)

epithio-ketone (344), via the enolate anion (349, to give the 2a-thiol(346), readily oxidized by air to give the bis-steroidal d i ~ u l p h i d e . ~ ~ ' 2a,S-Epithio-5a-cholestane(347) is oxidized selectively by rn-chloroperbenzoic acid to give the anti-S-sulphoxide (348). The syn-R-isomer (349)is formed only in traces. t-Butyl hypochlorite, at - 78 "C, gave mainly the syn-R-isomer. Both sulphoxides were stable at 250 "C, and to acids.259 The A2-unsaturated-5a-thiol (350) reacted with bromine or chlorine to give the 3fl-halogeno-2a,5a-epithio-derivative (353).90 It is suggested that a 5a-sulphenyl halide [e.g. (351)] is first formed, and then attacks the olefinic bond to give the sulphonium ion (352), which suffers nucleophilic attack upon C(3) by halide ion. Lead tetra-acetate similarly afforded the 3~-acetoxy-2a,5a-epithio-compound. 258

259

T. Komeno and M. Kishi, Tetrahedron, 1971, 27, 1517. M. Kishi and T. Komeno, Tetrahedron Lettcrs, 1971, 2641

297


The la,Sa-bridged compound (354) (a dithiolan) is converted by tris(diethy1amino)phosphine into the derivative (355).260 Selenourea, like thiourea, is a powerful nucleophile, replacing a 3~-tosyloxysubstituent (356) to give the 3a-selenouronium salt (357), which was hydrolysed

(350)

(351)

J (353)

EtzN -P -NEt,

I

S

(354) 260

(355)

D. N. Harpp and J. G. Gleason, J . Org. Chem., 1970,35, 3259.


29 8

to give the 3a-selenol (358).261 Because of high susceptibility to oxidation, the selenol was isolated as its Se-benzoyl derivative. A similar reaction sequence with cholesteryl tosylate gave the 3P-benzoylseleno-derivative, with the usual retention of configuration resulting from participation of the A5-olefinicbond.

TsO

H (356)

(358)

(357)

6 Molecular Rearrangements

Contraction and Expansion of Steroid Rings.-The 2-mesylate 4359) of a 3pmethyl-2a,3a-diol undergoes a normal pinacol rearrangement in alkaline methanol, although the product, the 2b-acetyl-~-nor-5a-steroid(361), implies epimerization to the more stable configuration at C(2); a concerted rearrangement would Me 4

Hk$/

CO

+ H H

':OH (359)

afford the 2a-epimer (360) initially. A comparable reaction sequence, using the 3-methyl-3P,4fl-diol in the 5p-series, afforded a 3-acetyl-~-nor-5[hteroid, of uncertain configuration at ~ ( 3 ) . ~ ~ ~ The solvolysis of 3a-mesyloxy-2a,5cr-epoxy-(362)or 2a,5~-epithio-(363)steroids (see p. 343) resulted in migration'of the anti-periplanar C(lkC(2) bond, probably with participation of electrons from the bridging atom, especially where this is sulphur.92 Attack upon the intermediate (364) by hydroxide ion affords the thio-hemiacetal (365) or the hydroxy-aldehyde (366), according to the nature of the bridging atom. The tosylhydrazone (367) of the 2a,5a-epoxy-3-oxo-derivative undergoes an essentially similar rearrangement on heating with a solution of sodium in e t h a n e d i 0 1 . ~In ~ ~this solvent the rearranged onium ion [cf. (364)] gives the (2'-hydroxyethy1)ether (368). A minor product from this reaction, which became the major product when the tosylhydrazone reacted with lithium hydride in toluene or xylene, was the lp,2~-cyclo-2a,5a-epoxide(370). The two reaction ''I

'" 263

S. M . Hiscock, L). A . Swann, and J. H. Turnbull, Cizem. C o ~ n m .1970, , 1310. A. K . Bose and N. C . Steinberg, J . Org. Chern., 197 I , 36, 2400. T. Komeno and H. Itani, Cherii. und Phurm. Bull. ( J a p a n ) , 1971, 19, 1123.


299

(362) X = 0 (363) X = S

CHO OH (366)

paths are typical of tosylhydrazone decomposition in polar and non-polar media, which afford reactive intermediates having carbonium-ion and carbenoid


300

character (369), respectively. The lP,3P-cyclo structure (a 3-oxatricyclo[2,2,1, 02,6]heptane)was deduced from spectral data, and from catalytic reduction to give the products (371H373). Rearrangement during solvolysisof the tosylate of a 4-hydroxymethyl-Cmethyl3-0x0-steroid is discussed on p. 276. Simultaneous contraction and expansion of adjoining rings affords some curious products when cholesterol reacts with lead tetra-acetate and hydrogen fluoride (p. 258). The 12P-hydroxyconanine mesylate Me

-

M #;e

Me

'H-'

H (374)

(375)

(376)

(374) rearranges into the c-nor-D-homo-structure (376) on reduction.264 Specific labelling by deuterium at the 18a- and 20a-positions afforded the rearranged Me

I

KOBu'

conformation)

',,+ , - , o d -

K'

H

(378)

(377)

p~oz

solventin 1lconf:ation

(379) 0'- Me

M *e

0

0'Scheme 17

G. Lukacs, P. Longevialle, and X. Lusinchi, Tetrahedron, 1971, 27, 1891.

30 1


product (376), the 18a-deuterium atom having migrated to the 13a- (now 17aa-) position, in accordance with the mechanism proposed. The base-catalysed D-homoannulation of ‘Reichstein’s L’ (377) has been investigated in various media.265 Earlier views on the mechanism of rearrangement have been slightly modified, to account for differences in the rates of formation and subsequent equilibration of the two products (378)and (379),depending upon the reaction medium. The main conclusions are summarized in Section 17. The preparation of some 16cr,l7a-methylene androstane (cyclopropane) derivatives, and their rearrangement to give ~-homoandrostan-l7a-ones,is described on p. 278. The ‘Westphalen’ and ‘Backbone’ Rearrangements.-Studies of rearrangements of C(5)-carbonium ions (or related species) continue. The presence of either 4Pmethyl (380) or 4,4-dimethyl substitution (381) in 5cr-cholestane-3/r’,5cr,6/3-triol 3,6-diacetate does not interfere with the Westphalen rearrangement, giving 5Pmethyl-19-nor-A’-derivatives (382).266,267Even the Ga-acetoxy- and 6-0x0analogues, with 4P-methyl or 4,4-dimethyl substituents, undergo the rearrangement, in contrast to the corresponding 6-substituted compounds without C(4)substitution. The &-methyl isomers, however, behave like the respective unsubstituted compounds, only the 6~-acetoxy-compoundin this series undergoing rearrangement.267 Elimination of 5a-OH to give the A4-unsaturated compounds (385) occurs in the 4a-methyl-6a-acetoxy- (383) or 6-oxo- (384) compounds, presumably because an anti-periplanar 4P-hydrogen is available for elimination. The other results can be rationalized according to the earlier view that 10P-methyl migration is favoured whenever an electronegative 6P-substituent is present, with

AcO :HO R’ R2 (380) R’ = H ; R2 = P-OAC (381) R’ = Me; R 2 = P-OAc

I

R’

R2

(382)

Me

(383) R (384) R z65

266 267

=~=0

0Ac

(385)

D. N. Kirk and A . Mudd, J . Chem. Soc. (C), 1970, 2045. J. G. L1. Jones and B. A. Marples, J . Chewr. Sor. (C), 1970, 2273. J. G. L1. Jones and B. A. Marples, J . Chem. Soc. ( C ) , 1971, 572.

OAc

302


the added postulate that a 4p-methyl substituent, implying the absence of a 40hydrogen, precludes direct elimination to give the A,"-unsaturated product : lopmethyl migration is then the best alternative pathway for removal of positive charge from C(5). Clearly a 6B-directed dipole is not mandatory for the rearrangement to occur. Reaction rates, relative to the C(4)-unsubstituted 3p,5a,6p-triol diacetate taken as unity (4P-Me, 158 ;4a-Me, 2.0 ;4,4-Me, 1.8),imply considerable relief of strain during ionizatioii at C(5)in the 4P-methyl compound, probably opposed in the 4,4-dimethyl compound by developing compression between the 4a-Me and 6 ~ - H as C(5)assumes trigonal geometry (386). A normal Westphalen rearrangement occurred in the 19-methylene derivative (387), giving the SP-ethenyl (vinyl) product (388).268 Migration of the ethenyl

AcoWc

AcO

Ho O A c

H

H

group may occur via one of several possible intermediates, for example a bicyclobutonium ion (389), a cyclobutonium ion (390), or a cyclopropylcarbinyl cation (391). Partial isotopic scrambling observed in a deuteriated ethenyl group (-CH=CD,) is compatible with any of these possibilities. 5a,9[j-Cholestane-3p,5,6P-triol 3,6-diacetate (392), under the usual conditions of the Westphalen reaction, reacts rapidly (150 x rate for 9a-isomer) to give the lh8

1. G . Gucst, J. G. L1. Jones, and B. A. Marples, Tetruhrdron Letters, 1971, 1979.


$

* OAc

Ac 0

OAc (392)

H2cT-,-i?

irx OAc

303

(393)

AGO

Me

(394)

OAc

(395)

ring-A contracted C(5)-spiran products (393)and (394).26’ The rate acceleration is attributed to strain relief, but the precise reason for specific migration of C(1) rather than the usual C(19) is not clear. Since cis migrations of this type are usually considered to require a fully developed carbonium ion, the high rate of reaction may constitute evidence against 10/3-methylparticipation in the ordinary Westphalen rearrangement. Further study is needed. When the 9P-compound (392) was treated with thionyl chloride in pyridine, the C(5)-spiran (393) was the major product, but the A7-unsaturated product (395) of lop-methyl migration was also obtained.270 The reason for the different reaction path is unknown: formation of the 7-ene appears to involve three migraticlns of syn-related groups. A ‘retro-Westphalen’ rearrangement occurred when the 10P-fluoro-SPmethyl-7/3-01 (396) was treated with boron trifluoride, to abstract fluoride ion. The product was the 4-en-7/3-01 (397).271 Contrasting behaviour when a lophydroxy-group was eliminated by acid treatment, reported last year to give a ‘backbone-rearranged’ ’)-olefinic l a requires an explanation. The most obvious difference in the reactants, apart from the different 10P-leaving

(396)

(397)

J. M . Coxon, M. P. Hartshorn, and C . N. Muir, Chem. Conznz., 1970, 1591. J . M. Coxon, M. P. Hartshorn, and C . N. Muir, Chern. C‘or~mr.,1971, 659. 2” J.-C. Jacquesy, R. Jacquesy, and S. Moreau, Bull. SOC.chirri. France, 1970, 4513. 2 7 1 a Ref. 96, p. 363. 264

270

304

Terperzoidsand Steroids

group, is the presence of the polar 7P-hydroxy-group in the lop-fluoro-compound (396). The 7P-substituent may oppose carbonium-ion migration towards C(8), whereas a 6P-acetoxy-group present in the 1OP-hydroxy-compound would similarly inhibit migration of the centre of positive charge towards C(5). ‘Backbone rearrangements’, initiated by development of a carbonium ion at C(5), appear to depend critically upon the pattern of substitution. Although A’ 3(1’)-olefins (398) normally result in the cholestane series, the acid-catalysed rearrangements of androst-5-ene and ~-homoandrost-5-enegive only A8(9)olefinic products [not a A8(l4)-olefin,as reported last year].272 In the androstane series, isomerization to the 14P-configuration relieves strain originally present in

Hz)n

l7

Me Me

H Me

(398)

\

H2)n

H+

r ~

H (401) IZ

Me

(402) n

= =

(399) n

=

1

&

1 2

Me

H2)n

Me

(400) n = 1 (403) n = 2

Me

Scheme 18

the trans-junction of rings c and D. Prolonged reaction of androst-5-ene with acids, however, also inverted the configuration at C(10)’and finally at C(5),to give an equilibrated mixture of olefins (399x401). Comparable reactions in the Dhomo-analogue equilibrated all four .ring-junction configurations to give the

’’’ D. N. Kirk and P. M. Shaw, Chern. Cornm., 1971, 948,


305

meso (402) and racemic Asc9)-olefins[(403) and its enantiomer]. Probable mechanisms for inversion at C(5) and at C(10)are depicted in Scheme 18.272 The occurrence of rearrangement in the D-homo series shows that strain at the C/D ring-junction is not a prerequisite. We can now see 'backbone' and related rearrangements as a search for the most stable olefinic structure or structures accessible by multiple Wagner-Meerwein rearrangements of tertiary carbonium ions ; a high degree of strain in the initial reactant is not essential. A similar result was observed, also in the androstane series, when the 3a-amine (404) rearranged in sulphuric acid to give an equilibrated mixture of the 10a- and 10P-A8(9)-unsaturatedcompounds (405).27 3 0

MeNH'

OH

@ J'

Ho-

(408)

1

HF

HO

(407)

HO

(409)

Treatment of the des-~-A~('l)-unsaturated compound (406) with hydrogen fluoride gave the 9a-fluoro-product (407) and the 'backbone-rearranged' ketone (409).274Migration of a deuterium atom from the 17a- to the 13a-position, and failure to incorporate any deuterium other than at C(11) when the reagent was deuterium fluoride, established a mechanism of the type illustrated (408),presumably with rapid sequential hydride and methyl shifts,not involving any oIefinic 273

F. Frappier, J. Thierry, and F.-X. Jarreau, Tetrahedron Letters, 1971, 1887.

274

J . P. Berthelot and J. Levisalles, Chem. Comm., 1970, 1162.


306

path a

F

Y

F Ft

Scheme 19

intermediate^."^ The addition of hydrogen fluoride on to the A5-olefinicbond of ‘pregnenolone’ was accompanied by both partial and complete backbone rearrangements as minor side-rea~tions.’~~ Full details have appeared of the reaction between cholesterol (410)and hydrogen fluoride, which gave, as minor products, the 25-fluoro-~-homo-compounds (41 1)’’’ Some other unusual rearranged products (412)and (413) have now been identified, and possible mechanisms of the type illustrated in Scheme 19 are discussed. Epoxide Rearrangements.-The rearrangements of 5a,6a-epoxy-steroids (414) with boron trifluoride have been extended to 19-methyl”’ and 19-methylene Berthelot and J. Levisalles. HIilI. SOL.rh1in. Fruntr, 1971, 1888. ’’‘ JP.. P.Bourguignon, J.-C. Jacquesy, K. Jacquesy, J. Levisalles, and J . Wagnon, Bull. SOC. ”’

( h i i n . Fraritr. 1971, 269 27’

1. G. Guest and B. A . Marples, J. Cheirz. SOL.(C), 1971, 576.

,

Steroid Properties and Reacticns

307

whenRisCH2=CH-

Ac 0

~

HO

HO

when = Et

1R

+ AcO

+

m‘

Ac 0

CHO

Ac 0

0

+

backbone-rearranged traces of products

Et O H Scheme 20

derivatives:68 with further diversity of results (Scheme 20). Apart from fluorohydrins (415), the various products arise from a C(5)-carbonium ion in each case, but there seems to be no simple explanation of the subtle effects of substituents, including those at C(19), upon the subsequent rearrangements of the carbonium ion. The postulated fragmentation of some epoxides to give unsaturated aldehydes, discussed last year,278has been confirmed in the bicyclic 5a,6a-epoxide

(416), which reacted with trifluoroacetic acid to give the seco-unsaturated aldehyde (417)as major product.279 Steroidal analogues undergo re-cyclization with BF, to give unsaturated alcohols, which may have the inverted configuration of the hydroxy-oxygen atom.278

”* 279

Ref. 96, p. 368. H. W. Whitlock, jun. and A. H. Olson, J . Amer. Chem. SOC.,1970,92, 5 3 8 3 .


308

The l-methylepoxy-ketone (418) rearranged in formic acid to give the A-noraldehydes (419) and (420) in the ratio 7 : 3.280 Although the major product could arise by a concerted mechanism (421), a non-concerted reaction via a C(1)carbonium ion seems to be required to explain the formation of the minor isomer.

“‘m

Me

H (420)

H+

In the lanostane series, Lewis acids converted the 8a,9u-epoxide (422) into the 7,9(11)-diene (423), and the 9a,lla-epoxide (424) into a mixture of the 9p,11ketone (425)and the diene (426),resulting from a partial backbone rearrangement of a C(9)-carbonium ion.’ 4 1 Although 9a,1la-epoxyandrost-4-ene-3,17-dione reacted with boron trifluoride to give only a modest yield of a 19(10 -+ 9)abeo

I@/

* H

H

H

(424) 2xo

+

(425)

(426)

V. Tortorella, L. Toscano, C . Vetuschi, and A. Romeo, J, Chetn. Sor. (C),1971, 2422.


309 0

0

product, allylic activation of the C(lO)-C(19) bond by A5-unsaturation in the compound (427)led to a high yield of the 9~-methyl-19-nor-l(l0),5-diene (428).281 Aromathation.-The influence of substituents upon the aromatization of 1,4d i e n - 3 - 0 n e s ~has ~ ~ been further investigated. The 6a-phenyldienone (429) rearranges in trifluoroacetic anhydride to give the normal l-hydroxy-4-methyl aromatic product (43l), accompanied by lesser amounts of the 4-hydroxy-lmethyl isomer (432)and other products.283 Assuming a spiro-cation intermediate

dK

f6 -’

(430)

(431) R’ = O H ; R2 = Me (432) R’ = M e ; R2 = OH

(430), the Ga-phenyl substituent apparently enhances the migratory aptitude of C(6) to the point where it can compete, though still relatively weakly, with the usual C(9) migration. The 3-(2’,4’-dinitropheny1hydrazone)and 3-(2’-hydroxyethy1imino)-derivativesof a 1,4-dien-3-onerearranged normally in trifluoroacetic anhydride, giving 1-substituted-4-methyl-derivatives (433) and (434), respectively, uia a C(5)~ p i r o - c a t i o n .The ~~~ Grignard reaction introduced alkyl or aryl groups into the 1,4-dien-3-oneto give, after dehydration, the 3-alkyl (or aryl)-1,3,5-triene (435). Aromatization afforded l-alkyl (or aryl)-substituted-4-methyl comp o u n d ~ .Detailed ~ ~ ~ product data for various C(3)-substituted spiro-cations (436) are tabulated, and are discussed in terms of the electronic features of the C(3)-substi tuen t, R. Steroidal compounds with three potential sites of unsaturation in rings A and B are aromatized on heating at 85 “C with acetyl bromide containing hydrogen bromide, giving the 4-rnethyloestra-1,3,5(10)-triene (440).285 The detailed

”’ 2g2 2M3

284 285

J. W. ApSimonand .I.M. Rosenfeld, Chem. Comm., 1970, 1271. Ref. 87, p. 277; ref. 96, p. 376. H. Dannenberg and T. Wolff, 2. Naturforsch., 1970, 25b, 823. T. Wolff and H. Dannenberg, Terrahedron, 1971, 27, 3417. J . Libman and Y . Mazur, Chem. Comm., 1971,729.


310 R-NH

\

\ +

R Me

(433) R

=

-NH

(434) R

=

-CH,--CH,OH

(or aryl)

(435 )

,

.*

(436)

mechanism must depend upon the particular reactant, but in general terms, comprises sequential elimination (or addition) and rearrangement steps, leading to the C'(5)-spiro-cation (439). Examples include the 3/3-acetoxy-Sa,6/3-dibromocompound (437) and the 4-en-3-one system (438). Probable first steps in ihe latter case are illustrated in Scheme 21. The 4-bromo-3P-hydroxy-5-ene (441) affords the same aromatic product (440) under much milder conditions. 296

(441)

(440)

Scheme 21 286

J . R . Hanson and T. D. Organ, J . Chr.m. SOC.(C), 1971, 1313.


31 1

A somewhat different reaction occurred in compounds with two potential sites of unsaturation in rings A and B, together with a 17/~-acetoxy-group[e.g. (442)]. With acetyl bromide-hydrogen bromide, a mixture of isomeric anthrasteroids (444) res~lted.''~ Reaction in this case proceeds through the spiro-cation (443),

R' HBr ___)

(several steps)

AcO'

(442)

04c

Br

which derives part of its unsaturation from elimination of the C(17)-substituent, and migration of the resulting oiefinic bond into ring B. Contrasting behaviour in the 17/?-acetoxy-14/3-androstaneseries (445), however, afforded a 17P-bromoderivative (446) without rearrangement. Similar principles, depending upon substituents representing potential unsaturation, explain the conversion of the cholic acid derivative (447) into the

(447) 'y7

J . Libman and Y . Mazur, C'hcni. Coinnz., 1971, 730.


312

18(13 + 12)-uheo-product (448) with an aromatic ring c.288Oestrone methyl ether (449), where the 17-0x0-group provides the oxidation equivalent of two olefinic bonds, rearranges with dehydration to give racemic 17P-methyl-14pgona-1,3,5( 10),6,8-pentaen-3-01 (450) on treatment with pyridine hydrochloride, or, better, with hydrogen bromide in DMF.289

(449)

(450)

Dehydrogenation of 9( 1l)-dehydro-17a-methyl testosterone (451) with D D Q and an acidic catalyst gave the cyclopentenophenanthrene (454).290 Examination of the reacting mixture showed that the 1,4,6,9[11),16-pentaen-3-one (452) is a key intermediate. Its formation by acid-catalysed dehydration at C( 17) is most unusual, but it is suggested that the presence of A9(")-unsaturation avoids the usual compression of the 13P-methyl group. Such compression certainly exists when an 11'P-hydrogen is present, and is suggested as providing the driving force for 13P-methyl migration to C(17),normally observed when tertiary C(17)alcohols are dehydrated (see also below). The aromatization steps presumably comprise a

Hi

H'

(454)

lHH 28') 24''

(453)

.I. Meney, Y . - H . Kim, a n d R. Stevenson, C'hern. C ' o r ~ r i r . ,1970, 1706 J . C. Hilscher, Chenz. Ber., 1971, 104, 2341. W. Brown and A . 5. T u r n e r , f. CIwin. SOC.(0,1971, 2566.


313

trienone-phenol rearrengement in rings A and B, and acid-catalysed migration of the 13p-methyl group as illustrated. A final dehydrogenation step in ring c completes the generation of the phenanthrene derivative (454). Miscellaneous Rearrangements.-The dehydrations of a 20-methylpregnan-20-01 (455) tc give the rearranged olefin (456), and of the D-homo-alcohol(457) to give the olefins (458) and (459), are now reported in The formolysis of a pregnan-20a-ol tosylate (460)also led to migration of the 13b-methylgroup, giving the

H

(455)

(456)

yJoH 1 0 I5 --.+

H

H

+

H

(459)

olefin (461).292One of several possible mechanisms is illustrated here. Lesser amounts of several other products were obtained (462)-(464)but only one of these (464)was a D-homoandrostane derivative, in contrast to the solvolysis of the 20B01 tosylate, which leads smoothly to the D-homo-system because of favourable conformational features.

Me

I

HC

-OCHO

H

H -

(463) "I 2'2

J

(464)

F. Kohen, K. A. Mallory, a n d I. Scheer, J . 0r.g. Chem., 1971, 36, 716. F . €3. Hirschmann, D. M . Kautz, S. S. Deshmane, and H. Hirschmann, rerrakeriron, 1971, 27, 2041.

314


The 17[&hydroxy-4,0-dienone (365), and similar ctsmpounds, rearranged with dehydration in 62 7; sulphuric acid to give a fluorescent solution, which afforded the 4,6,8(14)-trien-3-one (466).293Fluorescence is attributed to the protonated trienone.

The dehydration of the D-homo-diol 17-acetate (467) by various acids was reported last year to give several different full details have now been published.295

The 17,17-dirnethyl-lS-nor-13-ene (468), in sulphuric acid at 0 “C,afforded the rearranged and reduced product (469) in 60 ‘?.A The D-homo-analogue reacted similarly. N o detailed mechanism for the reduction (hydride donation) has been ofrered. The 3,5-cyclocholestanyl radicals (471) and (474), generated from the chlorocompounds (470) and (473), respectively, in an irradiated solution containing azobis-isobutyronitrile and triphenylstannane, each fragmented specifically at the Sa-bond in the cyclopropane moiety; hydrogen transfer then gave the hydrocarbons (472) and (475),re~pectively.~~’ The fragmentation is stereoelectronically controlled by the spatial direction of the half-occupied p-orbital at C(6), which approximates to the Sa-bond direction. 24-Methylene-sterols (476) equilibrate with their A24(25’-isomers(477) during chromatography in silica Iodine in boiling benzene isomerizes the A24(28)-oIefinicfucosterol acetate (478) to give the 2L’3

””

”’’ ””

2‘J7

””

W. Sadee, S. Riegelman, a n d L. F. Johnson, Steroids, 1971, 17, 595. Kcf. 96. p. 384. c‘. Monnerct a n d 0. K h u o n g - H u u , Bir//. SOC.c.hipn. Fruiircl, 1971, 6 2 3 . c‘. Monneret, P. Choay, Q. K h u o n g - H u u , a n d K. Goutarel, Trtruhedrvon Letters, 1971, 3123 ; see also ref. 5. A . L. J. Beckwith a n d (3. Phillipou, C’heiii. C‘omui., 1971, 658. M . G . N a i r a n d F. C. C h a n g , TcJtrahedronLcttrrs, 1971, 2513.


315

g 24

1; H

H

(476) A24(28) (477) A24(25)

(478) A24(28' (479) A24(25)

A24-isomer (479); a 7-ene is similarly isomerized uia the S(14)-ene to give finally the 14-ene.299 The highly-strained bicyclobutane (480) (cf. p. 25 1) rearranged on protonation in aqueous acetic acid to give the 6P,7/?-methylene-As('4)-unsaturatedcompound (481).12' The C(7a)-C(8) bond is presumably selected for cleavage by virtue of being the most strained. Hydrogenolysis of the bicyclobutane gave the 8a-methylA6-olefin (482), as a consequence of attack upon the more exposed bonds of the bicycl obutane.

ink Me0

Me0 H+

(482)

(480;

29')

N . Ikekawa, Y . Honrna, N. Morisaki, and K . Sakai, J . Org. Chem., 1970, 35, 4145.

Terpenods and Steroids

316

dH

o

/

(483)

17a-Hydroxy-17P-methylandrosta1,4-dien-3-one (483) has been identified as a metabolite in man of the anabolic 17P-hydroxy-17cc-methyl isomer."' The mechanism of inversion at C(17) is unknown. 7 Functionalization of Non-activated Positions Lead tetra-acetate converted the 1 OP-ethyl compound (484) into a mixture of 6P,19-epoxides (485) and (486), the 19R-isomer (4853 predominating as a consequence of the conformational preference of the ethyl g r o ~ p . ~ The " formation of

A

c

O Br

Y

Br

OH (485) 19R

(484)

(486) 19s

H O Br

(487)

a 2p, 19-epoxy-compound from a 2B-alcohol is equally eRective in the presence of a lr-bromo-substituent (487). Reduction of the bromo-epoxide (488) with zinc provided access to 19-substi tuted Sa-cholest-l-enes [e.g. (489)] and 19-nor analogues (see p. 321).302 The conversion of 5a-cholestan-6P-01 into the 6B,19-epoxide, by reaction with bromine and silver oxide, has been improved by the use of certain silver salts with bromine in pentane, in the dark.303 Silver acetate, carbonate, and joo

"' 302

'("

B. S. Macdonald, P. .I.Sykes, P. M . Adhikary, a n d R. A. Harkness, Biochern. J . , 1971, 122, 26P. Y . W a t a n a b e a n d Y . Mizuhara, .J. Org. C h i t i . , 1971, 36, 2558. C . W. Shoppee, J. C . Coll, and R. E. Lack, J . Clwin. SOC.( C ) , 1970, 1893. N . M . Roscher, Cliein. Coinin., 1971, 474.


317

sulphate favoured epoxide formation, but silver oxide or trifluoroacetate gave mainly 5a-cholestan-6-one under similar conditions. The Barton reaction (photolysis of a nitrite) has been applied to the C(20)alcoholic derivatives (490) and (491) in the c-nor-D-homo-steroid series. with the results indicated (Scheme 22)."04 In the 20,6-series, attack upon C(15) also

oz

+

-f-H

AcO#.-

\

H

fiNoYHH Me

H

(490) 20a(20S)

H 1

ON0

OH .-{-Me

I

0,. H

H

H

(491) 20fl(20R)

NOH

Scheme 22

occurred when the 20b-alcohol reacted with lead tetra-acetate ; Dreidilrg models show that the necessary boat-conformation of ring D in compound (492)is readily accessible. Photolysis of the lla-nitriie (493) gave a mixture of the l-oximinoderivative (495) and the A-nor-B-homo-dienone (497).3O 5 Both products are considered to arise from the C(1)-radical (494). 1,5-Cyclization would give the resonance-stabilized C(4)-radical (496), which can rearrange as indicated to give the dienone. 'Remote oxidation' by a photochemically generated benzophenone triplet, linked via a suitable ester group to the steroid molecule, was reported last year.306 An interesting variation uses a hydrogen-bonded comp!ex (498) between the carboxyl group in p-benzoylbenzoic acid and the free carboxyl group in 5aandrostan-3a-yl hydrogen succinate. The complex is sufficiently stable to direct 30J

'05

306

H. Suginorne, T. Kojirna, K. Orito, and T. Masamune, Terrahedron, 1971, 27, 291. H. Reirnann and 0. Z . Same, Canad. J . Chem., 1971, 49, 344. Ref. 96, p. 390.

318

Terpnoids and Steroids

(499)


319

attack of the excited benzophenone moiety to C(16),giving the 16-0x0-derivative (499) in yields up to 38 %.307 Suitable a-(p-benzoylpheny1)alkanoicesters of 5aandrostan-l7p-01 and 5,0-cholan-24-01 provided low yields of products with functional groups located in rings c and D . ~ ~ ~

The oxidation of the androstan-17-one derivative (500) by chromic acid in acetic acid introduces a 14a-hydroxy-group (501). Yields are now shown to be critically dependent upon traces of water in the solution, reaching the maximum (ca. 31 %) in the presence of 2-3 % of water, but falling off rapidly above or below this

8 Photochemical Reactions Unsaturated Steroids.-Cycloaddition of 1,l -dichloroethylene or vinyl acetate to a 1-en-3-one (502) afforded the cyclsbutano-compounds (503),although in low yield^.^ Ethylene, acetylene, 1,l-difluoroethylene, and maleic anhydride failed to react.

(502)

(503) (X,Y

= C1,

or H, OAc)

When the 3,5-dienes (504) or (505)are irradiated in pentane, under helium, the reactive bicyclobutanes (506) are formed, the 3,0,5p- and 4,6a-bonds arising by an internal cycloaddition between the two olefinic bonds. Addition of hexafluoroacetone to the pentane solution of the bicyclobutane opened the 4,6a-bond to give the adducts (507) or (508),according to the substitution type."1 307

309

310 311

R. Breslow and P. C . Scholl, J . Amer. Clzem. S O C . ,1971, 93, 2331. R. Breslow and P. Kalicky, J . Amer. Chem. Soc., 1971, 93, 3540. H. J. C . Jacobs, M. G. J. Bos, a n d C . M. Hol, K r c . Trav. chim., 1971,90, 549. P. Boyle, J. A. Edwards, a n d J. H. Fried, J. Org. Chem., 1970, 35,2560. E. Lee-Ruff and G . Just, Tetrahedron Letters, 1970, 4017.


320

(504) R' = H ; R 2 = Me (505) R' = M e ; RZ = H

Photoadditions ef ethylene and of cyclopentene to the A4-olefinic bond in steroidal 4-en-3-ones3or 4,6-dien-3-one$(Scheme 23), is reported to give cis-4~,5al 2 Configurations are assigned from consideration and/or trun~-4cr,5/~-adducts.~

(4-en-3-one or 4,6-dien-3-one)

+

@

0

OHt

0

Scheme 23

of strain in the cyclobutane ring, base-catalysed isomerization of truns-4~,5p-into cis-4~~.5/~-adducts, and 0.r.d. data. Considerable distortion of rings A and B is M. B. Rubiii, T. Maymon, and D. Glover, Israel J . C h n i . , 1970, 8, 717.


32 1

required to accommodate the strain involved in the trans-4a,5P-fusion, but the alternative trans-4~,5a-configuration is virtually impossible to ccmstruct from Dreiding models even with gross distortion. 17-Bromoandrost-16-ene adds hydrogen bromide under irradiation to give 16p,17P-dibrornoandro~tane.~A photochemical addition of trifluoroiodomethane on to a A3-olefin is mentioned on p. 258. Carbonyl Compounds.-Photolysis of the 1-en-19-a1 (509) afforded 19-nor-5achoiest-1-ene (510).302The 2a,5a-epoxy- and 2a75a-epithio-3-ketones(511) were

transformed upon irradiation into the furan (512a) and the thiophen (512b) respectively, with loss of carbon atoms 3 and 4.119

(511) (a) X (b) X

= =

0

S

The photo-isomerization of 1,4-dien-3-ones( 5 13)into 1,5-cyclo-l0a-compounds (514) has been carried out in the presence of a variety of alkyl groups (R) at C(4). Quenching by cyclohexa-1,3-dienesuggests that a triplet excitation is i n ~ o l v e dl .3~

@ \ 0

’ R

2537A

___)

@ 0 R

(513)

(514)

Irradiation of the 1,5-diene-3,7-dione( 5 15) afforded the diketone (516), and later the isomer (517).314 The energetics of the excitation (triplet) of the diene-dione (515) have been investigated in detail, and compared with data for the corresponding 1,5-dien-3-oneand 4,4-dimethyl-5-ene-3,7-dione.

”’ 314

D. I. Schuster and W. C . Barringer, J . Amer. Chem. Soc., 1971, 93, 731. S. Domb and K. Schaffner, Helv. Chirn. A c t a , 1970, 53, 1765.

322


Irradiation of the 5-hydroxyoestra- 1(10),3-dien-2-one(518)in benzene afforded the spiro-lactone (519).lg1

The photolysis of the 6-oxo-3a,5a-cyclo-19-oic acid (520) gives initially the 4-en-6-one (521), but in t-butanol a rapid photo-addition then affords the 4a-tbutoxy-6-hydroxy-lactone (522).31 Similar reactions transformed the methyl ester (523) into the ketonic derivative (525). Photolysis of the saturated keto-acid (526),in an alcohol as solvent, gave first the corresponding 6-monoester (527) of the 5,6-seco-6,19-dioic acid, and finally the 6,19-anhydride (528). Formation of the 6-ester (527) probably involves addition of the solvent alcohol to a keten intermediate.315 u The properties of the low-lying excited singlet and triplet states of 19 different steroidal enones have been investigated by phosphorescence excitation spectroscopy. at 77 and 4.2 K 3 I 6 Information was obtained on the ordering of excited states ['(n,n) > 3(12,7c) > 3(7c,7c)], on the polarization of the transitions, and on changes in geometry in excited states. ' I 5a

K . Kojima, Chrnz. arid Pharni. Bull. ( J a p a n ) , 1971, 19, 737. Ref. 87, p. 422. G. Marsh, D. R . Kearns, and K . Schaffner, J . Atner. Chcin. Soc., 1971,93, 3129.


323

(K = H)

hx, + Bu'OH

0 (520) R (523) R

= =

H Me

-+

0

: H : Bu'O OH

(521) R = H (524) R = Me

Bu'O

0

0

Miscellaneous Photochemical Reactions.-A re-investigation of the photolysis of a 20a-azido-pregnane (529) under various conditions317 has confirmed the absence of any products of insertion reaction^.^ 1 7 a Initial expulsion of molecular nitrogen is thought to afford the nitrene (530),which may dimerize : rearragement of the dimer (53 l), followed by a reorganization involving extrusion of acetaldimine, afforded the di-steroidal imine (532). In the presence of triplet quenchers the 20-iminopregnane (533)was the only product, resulting from hydrogen migration in the nitrene, whereas triplet sensitizers led to hydrogen abstraction from the solvent to give the 20-aminopregnane (534). GP-Azido-3a,5-cyclo-5a-pregnane (535) afforded the 6-imino-derivative (536) with traces of the isomeric 7-aza-~homo-enes (537).318 The photo-Beckmann rearrangements of the oximes of 5a- and 5p-cholestan-6ones preserve the stereochemical integrity at C(5),giving in each case an isomeric pair of lactams : the rearrangement must occur without dissociation of the C(5)C(6)bond.3

'

A . Pancrazi, Q. Khuong-Huu and R. Goutarel, Bull. Snc. chim. France, 1970, 4446. 3 1 7 a DH. . R. Barton and A. N. Starratt, J . Chem. SOC.,1965, 2444; cJ. ref. 96, p. 399. l 8 Q. Khuong-Huu and A. Pancrazi, Tetrahedron Letters, 1971, 37. 3 1 9 H. Suginome and H . Takahashi, Tetrahedron Letters, 1970, 51 19.

3 1 7


324

H (533)

J

triplet sensitizer

z2 H

(534)

L

(531)

L

H

(532)

Photochemical extrusion of nitmgen from the dimethylpyrazole (538) affords the substituted bicycio[4,l,O]heptanone (539), which then suffers photo-decarbonylation, probably via the cyclopropanone (540),to give a I a-isogropenyl-~-nor-2 ene (541)"' '?()

C. Berger, M . Franck-Neumann, and G . Ourisson, rerrahedrun Letters, 1970, 353 1 .

325


NH (536)

(537) (A6

+ A’)

(538)

& ‘

-co c-

H (541)

Sunlight, with methylene blue as sensitizer, partially demethylates dimethylamino-compounds. A 20a-dimethylaminopregnane, for example, gave the 20amethylamino-derivative in excellent yield.32 Irradiation of the nitrosoamine (542)or the chloramine (543) in acidic solution proceeds with a concerted ‘Grob fragmentation’, giving the corresponding pregnan-20-one (544).322 A novel and versatile synthesis of dialkyl steroid phosphates employs the steroidal alkoxyl radical (5451,generated by photolysis of the corresponding nitrite.323In the presence of a trialkyl phosphite, the radical attacks at phosphorus to give a phosphoranyl radical (546), which loses one of its four alkyl groups to give the desired phosphate (547). The preference for expulsion of one of the small alkyl groups, rather than the steroidal moiety, is thought to be determined by jil

322 323

F. Khuong-Huu and D. Herlem, Tetrahedron Letters, 1970, 3649. G. Adam, D. Voigt, and K. Schreiber, Tetrahedron, 1971, 27, 2181. D . H. R. Barton, T. J. Bentley, R. H. Hesse, F. Mutterer, and M. M . Pechet, Cherii. Comm., 1971,912.

326

Terpenoids and Steroids Me

I

H

(542) X (543)

x

= =

NO

(544)

c1

J3

‘0

H

0 (545)

(546)

(547)

ponderal rather than structural or statistical factors. Even a steroidal 1lp-nitrite gave the derived di-isopropyl phosphate. Photosensitized oxygenation of a 6-methyl-Sene (548), followed by reduction of the unstable hydroperoxide, afforded the 5a-hydroxy-6-methylene steroid

(549),’39 in contrast to the reaction of unsubstituted 5-enes to give 5a-hydroxy-6enes. The 6-methyl group presumably provides a more accessible hydrogen atom for removal than C(7). 9 Miscellaneous Reactions

Analytical Methods.-Convenient new methods are described for the estimation of steroidal alcohols and ketones as fluorescent derivatives. l-Dimethylaminonaphthalene-5-sulphonyl chloride [‘dansyl chloride’ 550)] reacts with steroidal alcohols to provide fluorescent esters which are suitable for thin-layer chromatog r a p h ~ The . ~ ~corresponding ~ sulphonyl hydrazine [‘dansyl hydrazine’ (551)] provides similar fluorescent derivatives of steroidal ketones. A fluorescent 324

R. Chayen, R. Dvir, S. Gould, and A. Harell, Israel J . Chem., 1970, 8, 157p; R. Dvir and R. Chayen, J . Chromntog., 1970, 52. 5 0 5 ; R. Dvir and R. Chayen, J . Chromatog., 1969, 45, 7 6 ; R. Chayen, S. Gould, and A. Harell, Analyt. Biochem., 1971, 39, 533.

Steroid Properties und Reactions

3 27 OEt

so1 I

X

x = c1

(550) X = C1 (551) X = NHNH,

(552) (553) X

=

RO, where R represents C(21)of a corticosteroid

reagent which reacts specifically with the primary alcoholic group at C(21)in [EDTN (552)], corticosteroids is 1-ethoxy-4-(dichloro-syrn-triazinyl)naphthalene which reacts by substitution of a chlorine atom to give the ether (553). Fluorescence spectra are also reporled for 6-substituted oestrogens under the conditions of the Kober-Ittrich-Brown reaction (heating with H2S0,, followed by extraction with p-nitrophenol in an organic solvent).3" In an attempt to identify the chromophoric systems resulting from reactions of steroids with strong acids ('colour reactions' ; Kober, Allen, Oertel. Talbot, Salkowski, and Liebermann-Burchard reactions) a detailed study has been made, with use of all the usual spectroscopic techniques.326 Visible colours are attributed generally io charge-transfer between the steroid, as donor, and a aelocalized carbonium icn, generated from the steroid in acidic media, acting as acceptor. The structures of the steroidal earbonium ions are discussed. The dihydroxyacetone side-chain (554) in 'Reichstein's S' is oxidized by alkaline triphenyltetrazohm chloride to the 21-aldehyde (555); further reaction under the influence ol' alkali affords a salt of the 17a,20-dihydroxypregnan-21-01~ acid (556).3

lB-o" CH,OH

CHO

I

I

~

[ a -coO H OH-,

'

(554)

I

lj/Jo"CHOH

H

H

COzH

H (555)

(556)

Manganous chloride-sulphuric acid is a useful spray reagent for t.1.c. of sterols and bile acids, giving characteristic c o l o u r ~2 8. ~A simple technique for thin-layer 325 326 327 328

J. Stastny and L. Dubey, Z . ancniyt. Chern., 1970, 252, 309. H. A. Jones, Canad. Spectroscopy, 1971, 16, 10. H. Mohrle and D. Schittenhelm, Arch. Pliarm., 1970, 303, 771 S. K. Goswami and C . F. Frey, J . Chromatog., 1970, 53, 389.


328

partition chrDmatography of steroids has been described ;329 Bush-type solvent systems are used, and good separations are claimed. The rrimethylsilylation of relatively unreactive hydroxy- and oxo-steroids with hexamethyldisilazane, for gas chromatographic-mass spectrometric analysis, is catalysed by the addition of brornotrimethyl~ilane.~~~ Oxo-steroids readily aff‘ord trimethylsilyl en01 ethers. Relative retention times (g.1.c.) have been obtained for about 90 sterols and their acetates, on a variety of columns.331 CH,-0,

I

,B

-R

CH-0

)fI

H (557)

The gas-chromatographic separation of corticosteroids as cyclic boronates offers considerable promise. Alkyl (or phenyl) boronic acids condense readily with 17a,20-, or 20,21-diols, and a!so with 17a,21-hydroxy-20-oxo steroids to form cyclic esters [e.g. (557) and (558)] with excellent g.1.c. properties.332 The mass spectra of these boronates exhibit prominent molecular ions. Under the title : ‘Cholesterol is Stable’, a sample purified via the dibromide in 1937, and stored in the dark, is reported to contain no detectable trace of impurity when examined recently.333 Ordinary cornr,;crcial samples, however, are highly susceptible to auto-oxidation during storage.334 Compounds isolated from aged cholesterol include the 20a-, 24R-, 24S-, and 25-hydroperoxy-derivati~es.~ Nucleation studies on supercooled cholesteric liquid crystalline materials (cholesteryl caproate and nonanoate) are r e p ~ r t e d36. ~ The presence of sodium deoxycholate induces optical activity in bilirubic, with significant Cotton effects near 410 and 460 nm.-337

324 33u ‘’I ’j2

333 334 335 j3’

337

D. J. Watson and D. Bartosik, J . Chrornatog., 1971, 54, 91. L. Aringer, P. Eneroth, and J.-A. Gustafsson, Steroids, 1971, 17, 377. G. W. Patterson, Analyr. Chem., 1971, 43, 1165. C. J. W. Brooks and D. J . Harvey, J. Chromutog., 1971, 54, 193. L. L. Engel and P. Brooks, Steroids, 1971, 17, 531. J . E. van Lier and L. L. Smith, J. O r g . Chem., 1971,36, 1007. J. E. van Lier and L. L. Smith, J. Org. Chem., 1970,35, 2627. J. M. Pochan and H. W. Gibson, J . Amer. Chern. SOC.,1971,93, 1280. J H p . . errin and M. Wilsey, Chem. Comm., 1971, 769.

2 Steroid Synthesis BY P. CRABBE, in collaboration with G. A. GARCIA, J. HARO, L. A. MALDONADO, C. RIUS, A N D E. SANTOS

1 Introduction Synthetic work in the steroid field remains an active area of research. A number of useful approaches to the steroid system, by either partial or total synthesis, have been reported during the year under review. The purpose of the present chapter is to put emphasis on original synthetic processes and sequences of general applicability. No mention will be made of particular reactions, the application of physical methods, biosynthetic studies, or microbiological reactions. Because of the general applicability of synthetic sequences, synthetic routes reported in one series can often be applied in other series. The selection of particular examples is therefore illustrative and to some extent arbitrary. The English edition of a book dealing with the total synthesis of steroids has been published. A review article has appeared on certain groups of biologically active steroids and raw materials for their production.2 Several accounts of different aspects of steroid synthesis have also a ~ p e a r e d . ~ - ~

2 Total Synthesis Several significant stereospecific cyclizations of olefins have been described. These include cyclizations of olefins devoid of asymmetric centres, but yielding racemic tetracyclic compounds containing five or six asymmetric carbon atoms, each having the configuration of the naturally occurring steroids. One of these sequences, first applied to the synthesis of 16,17-dehydroprogesterone, involved acid-catalysed cyclization of the acyclic tetraene ( I ) to the tetracyclic compound (2).* Similarly, when the substituted cyclopentenone (3) was A. A . Akhrem and Yu. A. Titov, ‘Total Steroid Synthesis,’ Plenum Press, New York,

1970. ’ R. Wiechert, Angew. Chem. Internat. Edn., 1970, 9, 321.

*

A. B. Turner, A n n . Reports ( B ) , 1969, 66, 389. S. Bernstein, ‘Chemical and Biological Aspects of Steroid Conjugation,’ Springer, New York, 1970. K. Wiedhaup, Chem. Weekblad, 1971,67, S3, S8. W. N. Speckamp, Chern. Weekblud, 1971,67, S20. M. P. Rappoldt, Chem. Weekblad, 1971,67, S18. W. S. Johnson, M. F. Semmelhack, M. U. S. Sultanbawa, and L. A. Dolak, J . Amer. 1968,90,2994; W. S. Johnson, K. Wiedhaup, S. F. Brady, and G. L. Olson, Chem. SOC., J . Amer. Chem. SOC.,1968, 90, 5277.

330


(3)

(4)

reduced and it reacted with trifluoroacetic acid, the D-homo-A-nor-steroid (4) was ~ b t a i n e d .Improvement ~ of the cyclization conditions afforded high yields of tetracyclic compounds." A fascinating total synthesis (Scheme 1) of (+)-progesterone (16) involving acetylenic participation in a polyolefinic cyclization has been disclosed.' 2Methylfuran ( 5 ) was alkylated with 1,4-dibromobutaneto afford the disubstituted furan (6). Ketalization with ethylene glycol in the presence of acid and a trace of hydroquinone gave the diketal bromide. This was converted into the corresponding iodide and treated with triphenylphosphine, thus providing the phosphonium salt (7). This, with phenyl-lithium, gave the ylide (S), which reacted with the enyne aldehyde (9), prepared in four steps from methacrolein, under special conditions'2 to yield the intermediate (10). Acid hydrolysis of the diketal (10) afforded the dienyne diketone (11) which was treated with base,

n

u

(7)

S. J. Daum, R . L. Clarke, S. Archer, and W . S. Johnson, Proc. Nut. Acad. Sci. U.S.A., 1969, 62, 333. W. S. Johnson, L. Li, C. A. Harbert, W. R. Bartlett, T. R. Herrin, B. Staskun, and D. H. Rich, J . Amer. Chem. SOC.,1970, 92, 4461. W. S. Johnson, M. B. Gravestock, R. J. Parry, R. F. Myers, Th. A. Bryson, and D . H. Miles, J . Amer. Chem. SOC.,1971,93,4330; W. S . Johnson, M. B. Gravestock, and B. E. McCarry, J . Amer. Chem. SOC.,1971, 93,4332. M. Schlosser and K. F. Christmann, Angew. Chem. Internat. Edn., 1966, 5 , 126.

om/'dj, w,

Steroid Synthesis

33 1

r-7

t V I I , Vlll

+

C0,Et

CHO

P

0

(9)

Br

Y += CHO Me

xii, xiii

I

Mg

XI

+--

c---

Reagents: i, Bu"Li; ii, Br(CH ) Br; iii, (CH,OH), , H ; iv, N a I ; v, Ph,P; vi, MeC(OEt), , H + ; vii, LiAlH,; [O]; ix, H , O t ; x, 2 % N a O H , EtOH-H,O; xi, MeLi; +

%,

xii, CF,CO,H, [):=O

; xiii, K,CO,, MeOH-H,O; xiv, 0,; xv, KOH, H,O.

Scheme 1

332


thus providing the substituted cyclopentenone (12) in 40 7; overall yield from the enyne aldehyde (9). Reaction of ketone (12) with an excess of methyl-lithium gave the trienyrrol (13),-which was cyclized with trifluoroacetic acid in dichloroethane containing ethylene carbonate. The tetracyclic compound (14),obtained as a 5 : 1 mixture of 17p- and 17a-isomeric ketones, was isolated in 71 yield. Ozonolysis of the tetracyclic ketone (14) gave the triketone (15),which was treated so as to induce intramolecular aldol condensation, thus affording ($)-progesterone (16) in 45 overall yield from (14), with the correct configuration at all asymmetric centres.' ' Several studies have been reported concerning the biogenetic-type cyclization of squalene derivatives to the tetracyclic ring-systems of steroids and triterpenes. The monocyclic epoxide (23) was chosen as the key intermediate for the total synthesis (Scheme 2) of the isoeuphenol system (24).13 Reaction of 8,9-dihydro(S)-(-)-limonene (17) with peracetic acid furnished the epoxide, which was converted into the crystalline diol (18) on hydrolysis with acid. Cleavage of diol (18) with sodium periodate gave the keto-aldehyde, which was condensed in the presence of piperidine to the ap-unsaturated aldehyde (19a). Reduction to the allylic alcohol (19b) was followed by conversion into the corresponding chloride (1%) and then the phosphonium chloride (20). Coupling of the tencarbon unit (20) with farnesyl bromide trisnoracetal (21) gave, after reduction with lithium-ethylamine, the monocyclic acetal (22). The aldehyde formed by acid hydrolysis of (22) was converted into the desired epoxide (23) by reaction with diphenylsulphonium isopropylide. Treatment of the epoxide (23)with boron trifluoride etherate or stannic chloride in different solvents afforded the sterol (24), structurally and stereochemically related to isoeuphol.' Non-enzymatic chair-boat cyclization of the diene terminal epoxide (29) produces a protolanosterolic intermediate (30a), which has been converted into the A7,9(11J-lanosterol derivative (31).l 4 The epoxide (29) was obtained through a coupling reaction, the key component for coupling being prepared from lanosterol (see Scheme 3). Dehydrobromination of 2rx-bromolanost-7-en-3-one furnished the A'-ketone (25). Pyrolytic cleavage of (25) gave the diene (26a). Treatment with one equivalent of osmium tetroxide, followed by sodium periodate oxidation, afforded the conjugated aldehyde (26b). Reduction of this provided the allylic alcohol (26c), which was then converted into the bromide (26d) by reaction with carbon tetrabromide and triphenylphosphine in methylene chloride. The bromide (26d) was allowed to react with the ylide derived from the tri-nbutylphosphonium salt (27), affording the coupled phosphonium salt. This salt was reduced with lithium in ethylamine to yield the acetal(28). Acid hydrolysis of (28) generated the corresponding aldehyde, which was converted into the epoxide (29) by means of diphenylsulphonium isopropylide. The reaction of (29) with stannic chloride afforded the A7-sterol (30a). Its acetate (30b), with hydrogen chloride in acetic acid, gave rise to an equilibrium

:;

l3

l4

E. E. van Tamelen, G. M. Milne, M. I. Suffness, M. C. Rudler Chauvin, R.J. Anderson, and R. S. Achini, J . Amer. Chrm. Soc., 1970, 92, 7202. E. E. van Tamelen and J. W. Murphy, J. Amer. Chem. SOC.,1970, 92, 7204.

9 TH

Steroid Synthesis

+ i,ii

333

--+

'H (17)

(18)

(19) a ; R b; R C: R

= = =

CHO CH,OH CH,Cl

c1-

+

0

U

I

iii, iv

(24)

(23) f -

Reagents: i, MeCOSH; ii, HClO,; iii, H , O + ; iv, Ph,SCMe,; v, BF,,Et,O or SnC1,MeNO,.

Scheme 2

mixture of the (9aH)-A7-and -A'-dihydrolanosteryl acetates. Treatment of the acetate (30b) with mercuric acetate furnished the A7y9(l')-lanosterol derivative (31).


334

(26) a ; R

C:CH, H b ; R = CHO C ; R = CH,OH d ; R = CH,Br =

mPBun,

0

u

Br-

(27)

0

u

(28)

1

11, 111

(29) (30) a ; R = H b ; R = AC

t -

Reagents:

i,

A;

i ~ EI,Ot: , 111.

Ph,SCMe,;

i ~ SnCl,, .

P h H ; v, Hg(OAc),.

Scheme 3

Since A7- or As-dihydrolanosterol and tirucallenol are not directly obtained during the above cyclization, reaction via a chair-chair conformation is apparently prevented by the severe steric interaction existing between vinyl methyl and either angular methyl group in the bicyclic moiety.I4 Consequently

335

Steroid Sjwthesis

the more favourable chair-boat conformation (29) initially leads to the 9,lO-cis isomer. This is in agreement with previous biogenetic proposals,' which suggested a boat conformation of ring B in squalene prior to cyclization srid predicted cyclization to a protolanosterol. In related work, a series of important studies have been devoted to enzymatic cyclization of unnatural ianosterol precursors. Thus both epoxides (23) and (29), despite being notably different in structure from squalene oxide (32),' were transformed enzymatically into the pentanorlanosterol (33a) and dihydrolanosterol (33b), re~pectively.'~These results lend support to the suggestion that the methyl-hydrogen migration sequence rests solidly cn physico-chemical

(33) a ; R

=

b;R

=

H C5H,,

principles. 5,1 In addition, this indicates that although the trisubstituted epoxide moiety is mandatory, methyl groups at positions 6, 10, and 15 and n-bonds at positions 14 and 18 in squalene oxide (32) are not essential for enzymatic cyclization. This reactivity pattern suggests that the epoxide-tetra-nbond sequence [(a, p, y ) in (32)] constitutes the essential structural requirement of the substrate for sterol formation." Such a hypothesis is supported by conversion of de-6-methyl-2,3-oxidosqualene into 19-norlanosterol' ' s 2 0 and into AIB-cI's1 9 - n o r l a n o ~ t e r oby l~~ 2,3-oxidosqualene-sterol cyclase. These studies firmly establish the role of 2,3-epoxysqualene (32) as progenitor of the polycyclic triterpenoids and steroids. However, a number of triterpenoids lack an oxygen function at C(3). In these cases, it has been suggested that a plausible biogenesis mechanism would involve the proton-initiated cyclization of squalene rather than of the terminal epoxide. An alternative hypothesis 15

16

17

18

19

20

A. Eschenmoser, L. Ruzicka, 0. Jeger, and D. Arigoni, Helc. Chim. Actu, 1955, 38, 1890. E. J. Corey, W . E. Russey, and P. R. Ortiz de Montellano, J . Amer. Chem. Soc., 1966, 88, 4750; E. E. van Tamelen, J. D. Willet, R. B. Clayton, and K. E. Lord, J . Amer. Chem. Soc., 1966, 88, 4752. E. E. van Tamelen and J. H. Freed, J . Artier. Chem. Soc., 1970, 92, 7206. E. J. Corey, K. Lin, and H. Yamamoto, J . Amer. Chem. Soc., 1969, 91, 2132; E. E. van Tamelen, R. P. Hanzlick, R. B. Clayton, and A. L. Burlingame, J . Amer. Chem. Soc., 1970, 92, 2137. E. J. Corey, A. Krief, and H. Yamamoto, J . Amer. C,hem. SOC.,1971, 93, 1493, and references therein. E. E. van Tamelen, J . A . Smaal, and R. B. Clayton, J . Amcr. Chem. Soc.. 1971, 93, 5279

336


postulates a reductive cyclization of the epoxide and resent experiments have shown that certain primitive organisms can generate polycyclic triterpenoids by both oxidative and non-oxidative cyclization of squalene, via a protoninitiated cyclization process.2' The hydride shifts accompanying cyclization of 2,3-epoxysqualene (32) to lanosterol in yeast and to the pentacyclic triterpenoid P-amyrin in peas have been confirmed,22thus supporting the previous proposal^.'^ A survey has appeared of the differences between in v i m and in uitro backbone rearrangements in the poiycyclic systems related to steroids and t r i t e r p e n ~ i d s . ~ ~ An ingenious synthesis (Scheme 4) of the steroidal tetracyclic system, based on a trisannellation reaction, has been reported.24 Condensation of the dichloro-ketone (34) with methylcyclopentanedione (35) afforded the trione (36),

J

0

(40) Reagents: i. Bu'OK; ii, TsOH: iii, H,SO,- C H z C l 2 ;iv, K O H , dioxan-H20.

Scheme 4 21

22

'' 24

D. H. R. Barton, G. Mellows, and D. A. Widdowson, J . Chem. SOC.( C ) , 1971, 110. D. H. R. Barton, G . Mellows, D. A. Widdowson. and J. J. Wright, J . Chem. Sor. (0, 1971, 1142. J. Bascoul and A. Castres de Paulet, Steroidologia, 1970, 1, 321. S. Danishefsky, L. S. Crawley, D. M. Solomon, and P. Heggs, J . Amer. Chem. S O C . , 1971, 93, 2356.

337

Steroid Synthesis

which was allowed to react with the di-t-butyl ester (37). The triene-dione (38) which was formed was then treated with acid, affording the diene-trione (39), which was readily converted with base into the tetracyclic steroid (40). A short and efficient route (Scheme 5 ) to the D-ring of 20-keto-steroids uia intermolecular alkylation of internal chloro-olefins or acetylenes has been reported.2s The key step of this approach involves the reaction of a chloroolefin, such as (42), with methylmagnesium iodide, followed by acid treatment, to produce a mixture of isomeric methyl ketones (43) in quantitative yield.

c1

iv, v

I

R2 (43) a ; R' b ; R' C ; R' d ; R'

R2 = a-H R2 = B-H = X-H;R2 = P-H = b-H; R 2 = a-H

= =

Reagents: i, C , H , , N H , ; ii, MeMgBr; iii, C l C H 2 ( C H 2 ) 2 C H : C M e C l ;i \ , v, H C 0 , H .

MeMgI;

Scheme 5

A number of other synthetic approaches to the steroid skeleton are discussed in the following sections, which are devoted more specifically to the synthesis of oestrane, androstane, and pregnane derivatives, as well as to seco-compounds and steroidal alkaloids. Until recently, the preparation of the bicyclic ene-diones (47a,b), which are important intermediates in steroid total synthesis, has led only to racemic mixtures. This deficiency has now been met as follows. Michael addition of the vinyl ketme (44) to the cyclic diketones (45a) and (45b) afiorded the triketointermediates (46a) and (46b) in high yield, each containing a prochiral centre. Optically active amines and amino-acids were used26 as chiral reagents to 25

26

P. T. Lansbury, P. C. Briggs, T. R. Demmin, and G. E. Dubois, J. Amer. Chern. S O C . , 1971, 93, 13 11 ; P. T. Lansbury, E. J. Nienhouse, D. J. Scharf, and F. R . Hilfiker, J . Amer. Chem. SOC.,1970, 92, 5649; P. T. Lansbury, P. C. Briggs, T. R. Demmin, and G. E. Dubois, Chem. Comm., 1971, 1107. U. Eder, G . Sauer, and R. Wiechert, Angew. Chem. Internut. Edn., 1971, 10, 496.

m~, Terpenoids and Steroids

338

0

+ o f i e l j 2 j n

R' (44)

-+

0

R' (46) a ; n = 1 b;n=2

(45) a ; ?I = I b;n=2

1 I

R' (47) a; IT = 1 b:n=2

effect the dissymetric cyclization of (46a,b) to (47a,b). The chiral induction was strongly dependent on reaction conditions, such as the substrate, amine component, solvent, time, and acid used.26

3 Photochemical Reactions Several surveys devoted to photochemical reactions report useful applications to steroid synthetic problems.27 -30 In addition, various publications have indicated that the steroid molecule is still frequently used as a probe in the study of photochemical reaction mechani~ms.~' The photochemical rearrangement of ap-epoxyketones to /?-diketones has been studied further. Results with 3-oxo-4,5-oxido-steroids indicate p + a alkyl shifts to be subject to stereoelectronic control, providing selectivity of the migrating group and stereospecificity of the rearrangement. The selective isomerizations of the epoxy-enone (48a)- (49) at -65 "C and of (48b) -+ (50) plug (51I at 20 "C support such a hypothesis. Indeed, at higher temperatures even (48a) can be converted into (51), presumably cia an additional radical intermediary step. The photorearrangement of the steroidal I ,5-diene-3,7-dione (52) afforded 3,7-ct:oxo-4,4-dimethyl-~17P-acetoxy-1( 10 --+ 5a)0beo-6/+,1 OP-cydoandrost-1-ene (53). An excited triplet state is involved in this reaction, and E , is 61--65 kcal mol- (sensitira tion) axad 54-33 kcal mol- (quenching).31 Various A'-4-

'

''

P G Bauslaugh, Synthrsis, 1970, 287.

'' P. G. Sammes, Quart. Kec., 1970, 24, 37. 'O

''

''

P. G. Samrnes, Svnthesis, 1970, 636. 0. Jeger and K. Schaffner, Pure A p y l . Ch(.m., 19?0, 21, 247. S. Domb and K. Schaffner, Helc. Chirn. A r f a , 1970, 53, 677, 1765. M. Debono, R. M. Molloy, D. Bauer, T. lizuka. K. Schaffner, and 0. Jeger, J . Atner. Chrm. SOC.,1970, 92, 420; K . Schaffner, A ~ ~ c Chern. J M ~ Internat. . Edn., 1971, 10, 201.

Steroid Synthesis

339 OAc

(48) a ; 9a, lOa b;

9/39

lop

OH

alkyl-testosterone analogues (54) have also been submitted to irradiation to provide steroidal lumisantonines (55).3 Irradiation of the steroidal nitroxide (56) in toluene solution afforded 39 of (57) and 57% of the unstable N-hydroxy-derivative (58). This shows the N-hydroxy-group to be a potential tool for remote functionalization in appropriate molecules. These results, along with similar ones obtained in other series,

Me > C1 > SR >> 0 ~ ~ . 1 0 3 Several steroidal N-oxides have been reported and in some cases the binding to a hydrophobic region of certain membranes has been investigated in connection with the paramagnetic resonance spectra. O4 Various studies have been devoted to the synthesis of amino-substituted steroids. These compounds can be prepared either from an alcohol, by displacement of a toluene-p-sulphonyl group by an amine,"' from a ketone, either by reduction of the corresponding oxime'0s,106or by Beckmann rearrangement of 20-keto-oximes,'07 or by catalytic reduction of a nitro-group."' Another lo' lo'

lo4

lo' lo6

lo'

lo'

P. Crabbe and A. Guzman, Chem. and Znd., 1971, 851. A. Moscowitz, E. Charney, U. Weiss, and H. Ziffer, J . Amer. Chem. SOC.,1961, 83, 4661. W. Nagata, T. Okumura, and M. Yoshioka, J . Chem. SOC.(C), 1470,2347; W. Nagata, M. Yoshioka, T. Okumura, and M. Murakami, ibid., 1970: 2355; W. Nagata, M. Yoshioka, and T. Okumura, ibid., 1970, 2365; W. Nagata, T. Wakabayashi, M. Narisada, and Y . Hayase, ibid., 1971, 2415. W. L. Hubbell and H. M. McConnell, J . Amer. Chem. SOC.,1971,93,314;E. C h m b a z , G. Defaye, A. Hadjian, P. Martin, R. Ramasseul, and A. Rassat, Compt. rend., in the press; R. Ramasseul and A. Rassat, Tetrahedron Letters, 1971, 4623. R. Glaser and E. J. Gabbay, J . Org. Chem., 1970, 35, 2907. P. Crabbe, M. J. Durazo, R. M. Saloma, and P. G. Holton, Bull. SOC. chim. belges, 1962, 71, 203; G. Defaye and P. Jandon, Compt. rend., 1971, 272, C,702. G . E. Arth, A. A. Patchett, T. Jefopoulus, R. L. Bugianesi, L. H. Peterson, E. A . Ham, F. A . Kuehl, and N. G . Brink, J. Medicin. Chern., 1971, 14, 675. G. Defaye, Bull. Acad. polon. Sci., Ser. Sci. chim., 1971, 19, 1 .

365

Steroid Synthesis

method involves either the reduction of a cyanohydrin or the opening of an epoxide ring with the appropriate amine."' Stereoselectivehydroxylation at C(5)has been achieved from the corresponding 3-amino-derivatives by neighbouring-group participation. A possible mechanism involving a dihydro-oxazinium cyclic intermediate has been proposed. The acid-catalysed cleavage of the 9p,1 lp-epoxide (188) by acetonitrile in the presence of an acid afforded the corresponding 9a-acetamido-ll,&hydroxyderivative (189). On oxidation with chromic acid in the presence of acetic acid, the isoxazoline (190) was formed.'"

MeCN,

HClO

---4

The application of the method used for the preparation of cholestane 2a,5aepisulphide for the obtention of 3p,17P-dihydroxy-5a-androstane2a,5a-episulphide (198a) and its 19-nor-derivative (198b) has been mentioned in recent reports.'12 The sequence is shown in the case of the conversion of (191) into the thia-steroid (197) (see Scheme 15). After performing such a synthesis, the norsteroid (198b) was converted into the ~-nor-3-thia-steroid(199a) by lead tetraacetate fragmentation to (200) followed by pyrolysis and photochemical elimination.'I3 The ~-nor-3-oxa-analogue(199b) was obtained directly from the 2a,5a-epoxide (201) by irradiation, although the yield was lower.' l 3 The four possible 5a-androstane-16,17-diols, both 17-oxo-5a-androstan-16-yl acetates, and 16-0~0-5a-androstan-17P-ylacetate have been prepared by classical log

'lo ''I

'

l3

J. B. Jones and J. D. Leman, Canad. J . Chem., 1971,49, 2420; D. N. Kirk and M. A. Wilson, J. Chem. SOC.( C ) , 1971, 414. A . Ahond, A. Cavt, C. Kan-Fan, and P. Potier, Bull. SOC.chim. France, 1970, 3624. J. M. Teulon, T . T. Thang, and F. Winternitz, Compt. rend., 1971, 272, C , 1254. T. Komeno, M. Kishi, and K. Nabeyama, Tetrahedron, 1971, 27, 1503; T. Komeno and M. Kishi, ibid., 1971, 27, 1517. M. Kishi and T. Komeno, Tetrahedron, 197 I , 27, 1527.


366

7

O=O

I

0

0 3:

Ik 3: 0

O= :i,

I

h 3

2

v

O

q

j

367

Steroid Synthesis

(198) a ; R' b;R1

= =

Me; R 2 = O H ; X H ; R2 = O H ; X

=

=

S S

(199) a ; X b;X

= =

S 0

methods, either from the 17-ketone enol acetate or from the 16a,l7a-epo~ide."~ The syntheses of 17a-chloroethynyl- and 17a-ethynyl-3P717P-dihydroxyandrost-5-en-7-ones have been reported,' l 5 as well as of 17-bromo-l4~-androstan16-ones.' I 6 In addition, 3~-acetoxy-14P-androst-5-en-17-one has been prepared from its 14a-isomer, through protection of the As-double-bond as its 5a,6Pdichloro-derivative and chromic anhydride oxidation in aqueous acetic acid, yielding the 14a-hydroxy-deri~ative~'~ in ca. 24 % yield. Dehydration at C(14), followed by catalytic hydrogenation, afforded the 14P-isomer.' An elegant

(202) '14

'15 li6

'"

(203)

M. G. Combe, W. A. Denny, G. D. Meakins, Y . Morisawa, and E. E. Richards, J. Chem. SOC.(0,1971, 2300. J. H. Siemann, Z . Chem., 1971, 11, 15. T. Nambara, H. Hosoda, M. Usui, and T. Anjyo, Chem. and Pharm. Bull. (Japan), 1971, 19, 612. A. F. St. Andre, H. B. MacPhillamy, J. A. Nelson, A. C . Shabica, and C. R. Scholz, J. Amer. Chem. SOC.,1952, 74, 5506; P. J. Sykes and R. W. Kelly, J. Chem. SOC.( C ) , 1968, 2346; C. M. Hol, M. G. J. Bos, and H. J . C . Jacobs, Tetrahedron Letters, 1969, 1157. H. J. C. Jacobs, M. G. J. Bos, and C. M. Hol. Rec. Trav. chim., 1971,90, 549.


368

means of introducing a 14P-hydroxy-group has been described. This consists of treatment of 3a-acetoxy-23,24-dinor-5~-chol-7-en-22-01 (202) with phosphorus tribromide, affording 14,22-epoxy-23,24-dinor-5P-cholan-3~-yl acetate (203), which may readily be converted into 14P-hydroxy-steroids.' The 12a,l3P-etiojervane analogue (204) of testosterone has been prepared from hecogenin.' 2 o It was shown that reaction of the 2-benzylideneandrostane (205) with phenylhydrazine in acetic acid afforded three isomeric products

'

Ph

(209) lZo

D. J. Aberhart and E. Caspi. J . Chem. SOC.( C ) , 1971,2069. W. F. Johns, J. Org. Chem., 1970,35, 3524; 1971,36, 711.

Steroid Synthesis

369

(206), (207), and (208). These were all converted into the phenylpyrazole (209) in refluxing chloroform in an oxygen atmosphere.' 21 The reaction of conjugated enamines with a-bromoketones leads to the formation of substituted furans.122 This sequence has been applied to the synthesis of the steroido[3,4-b]furan (211) from the conjugated enamine (210).12'

0

II

Et-C-CH,Br DMF

,

Several steroidal spiro-lactones have been prepared by treatment of a 17(212), in a Reformatsky keto-steroid, such as 3/3-acetoxyandrost-5-en-17-one reaction with a bromo-ester, which gave an isomeric pair of 17-spiro-lactones, such as (213).'23 Similar results have been obtained in the oestratriene series.

OEt I BrCH,-C=CH-CO,Et b

OEt

Qc0 (2 12)

Ac 0

Various accounts report the synthesis of heterocyclic androstane derivatives. 17/3-Acetoxy-4-oxa-androst-2-ene (21 5 ) has been prepared in three steps from the hemiacetal (214).124 Since various 4-aza-steroids exhibit antimicrobial activity, several new 4-aza-androstanes have been prepared. The reduction of ring A enamine lactams, such as 4-aza-androst-5-ene-3,17-dione (216a), under conditions of'the Leuckart-Wallace reductive amination, provided 17fL(N-methyl12'

lZ4

J. B. Cazaux, R. Jacquier, and G. Maury, Tetrahedron Letters, 1971, 41. U. K. Pandit, H. R. Reus, and K. De Jonge, Rec. Trao. chim., 1970, 89, 956. 0. A. De Bruin, J. Botterman, and P. Westerhof, Rec. Tratl. chim., 1970, 89, 961. G. R. Pettit and T. Kasturi, J . Medicin. Chem., 1970, 13, 1244.

370


111,

PTSA

HO'.

formamido)-4-aza-5a-androstan-3-one (2i 7a) in high yield.' Catalytic hydro(216b) provided genation of 4,17a-dimethy1-4-aza-androst-5-en-17~-01-3-one the saturated aza-androstane (217b). Reduction of (217b) with lithium aluminium hydride gave the 3-deoxy-aza-steroid (218) in good yield.' *' Similarly, several types of N-substituted 17a-aza-~-homo-androst-5-en-3fi-ols(219) have been synthesized by classical techniques. ' 2 7 0

I1

a ; HC-NMc,

b ; HZ-Pt, AcOH

& H

0 (216) a ; R '

=

H ; RZ = =O

(217) a ; R '

=

NMeCHO

H ; R2

=

.,.

U

s

.->

m

2

d

+

d h

m Ic)

2

395

Steroid Synthesis 0

0

(339)

H20:&

Me0

(340)

'C2H5

: PVi : : : H &

/

Me0

/

(342)

(341)

Reagents: i, CH,:C(Me)OAc; ii, OsO,; iii, HIO,; iv, CH,N,; v, H , , e - , H,SO,-H,Odioxan; vi, NaOH, MeOH.

Scheme 23

(344; n

(345)

=

0, 1, or 2)

(344)


396

0

-

M e 0QCHO (348)

M e 0a

'\

M

H e

(349)

1

0

vii

H

Me

lix

0

0

Me0

*k

H Me

\ '

(353)

&i*Et

xiii, xiv

Me0

Me0

Me (355)

(354) a ; R = p-Me b ; R = a-Me

Me0 (356) Reagents: i, (EtCO),O; ii, H,-5% Pd/C; iii, (COCI),; iv, C H 2 N Z ;V , Ag,O; vi, HF-THF; , EtOH; ix, TsOH, PhH ;

vii, C H I . CHMgBr, THF; viii,

O

a

0

x, NaOH, E t O H ; xi, H', 20 " C ;xii, Mel, D M A ; xiii, e - , H,SO,-H,O-dioxan; xiv, NaOH, (CH,OH),; xv, Pd/C, Me,C,H,.

Scheme 24

397

Steroid Synthesis

ccydimethyltetronic acid afforded the tricyclic intermediate (351). Ring closure under acidic conditions provided the tetracyclic lactone (352),which was converted with base into the lzctol (353). Treatment of (353) witn methyl iodide afforded the expected keto-ester D-seco-steroid as a mixture of 7a-and 7B-methyl isomers (354), separated by preparative thin-layer chromatography. Electrochemical reduction of the ketone was followed by alkaline hydrolysis of the acid (355). ester group, thus yielding the 7-methyl-8-dehydro-cis-doisynolic Dehydrogenation of (355) either with DDQ or catalytically with palladium provided the 7-methyl-bis-dehydro-cis-doisynolic acid anzlogue (356). 8 9 A number of A4-3-keto-~-seco-17-alcohols have also been prepared.lg0 Base-promoted cleavage of the D ring of oestrone (114a), followed by esterification, lithium aluminium hydride reduction, and Birch reaction, gave, after mild

r V C H 2 O H

S 0'

(357)

z

'

:

O

H

(358)

dEroH -,J,

0

CH ,OAC

' (359)

-0 (362; R = H or Me)

acid hydrolysis, the By-unsaturated ketone (357), whilst strong acid provided the conjugated ketone (358). Treatment of (357) with pyridinium bromide perbromide afforded the A4,9('0)-dien-3-one(359). Acetylation at C(17) followed by reaction with methanolic hydrogen chloride gave the ')-isomeric diene (360). Treatment of (360) with DDQ and subsequent base hydrolysis Following similar sequences of reaction, afforded the 4,9,1 l-trien-3-one (361). the corresponding ring-A-alkylated D-seco-steroids (362) have been prepared. A5(10)79(1

'

19"

A. Cervantes, J. Iriarte, H . Ponce, P. Crabbe, and J. H. Fried, unpublished results.

398


B-Seco- as well as 4-substituted-A4-3-keto-analogues have also been obtained by similar r o ~ t e s . ' ' ~

9 Cholestane and Vitamin D, and its Analogues Syntheses in the cholestane series have featured in previous sections (see Part 11, Chapter 1, ref. 190). Some further interesting synthetic sequences now follow. Treatment of 3~-tosyloxy-5a-cholestan-6~-ol (363) with sodium azide gave the 3a-azido-derivative (364), which was oxidized to the corresponding 6-ketone (365). The 3P-azido-5b-cholestan-6-one(366)was prepared by a similar sequence. Equilibration experiments of (365) and (366) showed the influence of intramolecular electrostatic interaction between the azido- and keto-groups.

''

(366)

(365)

Lead tetra-acetate oxidation of 3~-acetoxy-5a-bromo-6~-hydroxy-19a-methylcholestane gave predominantly the 6PJ9-0xide with the (19R)-configuration. 92 Cholesta-l,4,6-trien-3-one (367) was converted into la,2a-epoxycholesta-4,6dien-3-one (368) with hydrogen peroxide in basic medium. Catalytic hydrogenation over palladium on calcium carbonate in pyridine solution gave exclusively la,2a-epoxycholest-4-en-3-0ne(369). Dehydrogenation of ergosterone (370) with DDQ gave ergosta-4,6,8(14)-trien-3-one (371). Treatment of (370) with hydrochloric acid afforded ergosta-4,6-dien-3-one, which when dehydrogenated with DDQ in the presence of acid yielded ergosta-1,4,6-trien3-one (372).194Various phosphorodichloridates have been prepared by reaction 191

D. N. Jones, K. J. Wyse, and D. E. Kime, J. Chem. SOC.(0, 1971,2763. Y.Watanabe and Y . Mizuhara, J. Org. Chem., 1971, 36,2558. B. Pelc and E. Kodicek, J. Chem. SOC.(0,1971, 1568. B. Pelc and E. Kodicek, J . Chem. SOC.(0, 1971, 859.

lYz

ly3 194

Stcroid Synthesis

399

(367)

(369)

of the corresponding alcohols with phosphoryl ch10ride.l'~ When the same reaction conditions were used with 6P-hydroxy-i-cholestane and 6P-hydroxyA4-3-ketocholestane, cholesteryl 3P-phosphorodichloridate and cholestane-3,6dione were formed, respectively. 9 5

(370)

A total synthesis of precalciferol (377) has been reported, which involved nucleophilic addition of the lithium salt (374) to the ketone (373), giving the acetylenic tricyclic intermediate (375). Elimination of HOCl from the chlorohydrin (375) with bis(ethylenediamine)hromium(rr) afforded the en-yn-ene (376), which was reduced to precalciferol (377) with Lindlar's catalyst. Thermal isomerization of (377) then afforded vitamin D, (378).Ig6 R. J . W. Cremlyn and N. A. Olsson, J . Chem. SOC.( C ) , 1971,2023. J. Dixon, P. S. Littlewood, B. Lythgoe, and A . K. Saksena, Chem. Comm., 1970, 993.


400

0

C

(373)

(374)

Ill

C

(en),Cr, DMF

1

(377) PhH. A

I

(376)

Several hydroxylated analogues of vitamin D have been prepared by irradiation of a steroidal A5,7-dienep r e c u r ~ o r , following '~~ a sequence similar to that used in the synthesis of retroprogesterone. 44 Photobromination of cholesteryl benzoate also gave a triene derivative analogous to vitamin D, (378).19* 19'

J. S. Bontekoe, A . Wignall, M. P. Raffold, and J. R . Roborgh, Znternat. Z . Vitaminforsch., 1970, 40, 589. R . Ikan, A. Markus, and E. D. Bergmann, Israel J . Chem., 1970, 8, 819.

Steroid Synthesis

40 1

(379)

It has been suggested that vitamin D, (378) is metabolized into a more polar substance before stimulating calcium transport to the intestine. The principal metabolite from the blood, produced by the liver, has been found to be 25hydroxycholecalciferol(379), whereas the trihydroxy-derivative (380) is the principal metabolite from the intestine.'99 Autoxidation of cholesterol via hydroperoxide intermediates afforded a variety of hydroxylated cholesterol derivatives and products of side-chain degradation.200 The synthesis (Scheme 25) of ergosta-5,7,22,24(28)-tetraen-3B-o1(386), a biogenetic precursor of ergosterol (38la), has been reported.201 Diels-Alder addition between 4-phenyl-l,2,4-triazoline-3,5-dione (382) and ergosterol acetate (381b) afforded the adduct (383). Reduction of (383) with lithium aluminium hydride regenerates ergosterol (38la). On the other hand, ozonolysis of (383) selectively cleaves the side chain, thus providing the aldehyde (384). Wittig reaction of the aldehyde (384) with (3-methyl-2-methylenebutyljtriphenylphosphonium bromide gave, after reacetylation, the protected tetraene (385). Reduction with lithium aluminium hydride gave the tetraenol (386), identical with the natural material.201 Similar types of alkylation have been applied in syntheses of other steroidal ~ i d e - c h a i n s , ~ ' ~including .~'~ that of 22-trans-26,27-dinorergosta-5,22-dien-3~-01 (387), a novel marine In addition, the Wittig reaction has been used to prepare various possible polyene intermediates in phytosterol biosynthesis.2 0 5 The aldehydes (389) and (390) were prepared (Scheme 26) from stigmasterol acetate (388b) by modification of a known procedure. These aldehydes were then alkylated with a variety of ylides derived from phosphonium salts, leading to a series of polyenes (391) and (392).205 199

zoo 201

202

*03 204 205

A. W. Norman, J. F. Myrtle, R. J. Midgett, H. G. Nowicki, V. Williams, and G. Popjak, Science, 1971, 173, 51. J. E. Van Lier and L. L. Smith, J. Org. Chem., 1970,35,2627. D. H. R. Barton, T. Shioiri, and D. A. Widdowson, Chem. Comm., 1970,939; J . Chern. SOC.(0, 1971, 1968. S. Bory, D. Jung Lin, and M. Fetizon, Bull. SOC.chim. France, 1971, 1298. R. Ikan, A. Markus, and E. D. Bergmann, Steroids, 1970, 16, 517. M. Fryberg, A. C . Oehlschlager, and A. M. Unrau, Chem. Comrn., 1971, 1194. M. Fryberg, A. C . Oehlschlager, and A. M. Unrau, Tetrahedron, 1971, 27, 1261.


402

+ RO (381) a ; R

=

H

+

Reagents: i, LiAlH,; il,O , , CH,CI,, MeOH, -70 "C; iii, Me,CHC(:CH,)CH,PPh,

Scheme 25

Br

Steroid Synthesis

403

(388) a ; R = H

b; R i-iii

\

=

AC

LCH

k C H 0

HOW

1

(390)

Reagents: i, PhIBr,; ii, 0,; iii, Zn-AcOH; iv, (CH,OH),, H'; v, NBS; vi, P(OMe),; vii, H 3 0 + .

Scheme 26

A number of modified steroid side-chains have also been obtained by Grignard

reaction^.^^^.^^^ Several different studies have been devoted to the synthesis of modified side-chains, sometimes containing heteroatoms.208-2 l4 206 207

208 209

210 211

2 12

213

2 14

B. M. Kapur, A. Mannan, and G. R. Duncan, Chem. Comm., 1971, 775. W. Sucrow and P. Polyzou, Tetrahedron Letters, 1971, 1883. Y. Yanuka, R. Katz, and S. Sarel, Tetrahedron Letters, 1970, 5229. S . Sarel, Y. Yanuka, R. Katz, B. A. Weissman, and Y. Stein, Tetrahedron Letters, 1971, 369. S. Sarel, B. A. Weissman, and Y. Stein, Tetrahedron Letters, 1971, 373. J. E. Herz and S. Cruz Montalvo, Steroids, 1971, 17, 649. W. Sucrow and B. Raduchel, Chem. Ber., 1970, 103, 271 1. N. K . Chaudhuri, R. C. Nickolson, and M. Gut, Steroids, 1970, 16, 495. V. V. Ranade, F. Kohen, and R. E. Counsell, J. Medicin. Chem., 1971, 14, 3 8 .


404

A-Hornocholestane derivatives have been obtained by the Demjanov ringenlargement r e a ~ t i o n . ~ Transannular solvolysis reactions have been observed in seco-steroids of type (393) containing a ten-membered ring. Thus, treatment of (393) in aqueous acetone solution produced 5(10-+ lPH)abeo-5~-cholestlO(19)-ene 3P-acetate (394) in high yield.216 The synthesis of various 2- and 4-oxa, -thia-, and -aza-5a-cholestanes has been r e p ~ r t e d . ~ Attempts ” have also been made to introduce a phosphorus atom

’

/

(394)

o=c

NO, 215 z1

217

(393)

H. Velgowa and V. Cerny, Coll. Czech. Chem. Comm., 1970, 35, 2408. M . Lj. Mihailovic, M. Dabovic, Lj. Lorenc, and M. Gasic, Tetrahedron Letters, 1970, 4245. Y . Kashman and M. Sprecher, Tetrahedron, 1971, 27, 1331; Y. Kashman and E. D. Kaufman, ibid., 1971, 27, 3437.

405

Steroid Synthesis

&-.!+ C&,,

0

Ho2c& (396)

@ 1M c7210e

(395)

L o

+ (397)

'V,V

1

C8H 1 7

0&17 .O=P-OR

AcO

I

OR (401) a ; R = H b ; R = Me

(398)

+

OzP-OMe \ OMe

O=P-o~'

I

OR2 (399) a ; R ' = R2 = H b;R' = H;R2 =Me c ; R 1 = R2 = Me

(400)

Reagents: i, KMnO,, NaIO,; ii, CH,N,; iii, (CH,OH), , H'; iv, LiAlH,; v , AcOH-H,O; vi, (MeO),POH, TsOH; vii, (MeO),POH.

Scheme 27

into the cholestane nucleus. These heterocyclic steroids were prepared by cleavage of ring A, followed by ring closure. One of the approaches used is illustrated (Scheme 27) in the conversion of cholestenone (395) into the heterocyclic derivatives (399) and (401).217


406

10 Steroidal Insect and Plant Hormones A detailed survey of the isolation and chemistry of the ecdysones has appeared.21* The isolation of deoxycrustecdysone (402a), deoxyecdysone (402b), and a-ecdysone (402c) has been reported.219 The structures of stachysterone A

(402) a ; R' b ; R' C; R '

=

H ; R2

=

RZ = H

=

OH; R2

=

OH =

H

(403), the first naturally occurring 27-carbon steroid with a rearranged methyl group, and of stachysterone B (404)have been established.220 The stereochemistry of the latter has been correlated with that of ponasterone A (405) by selective dehydration at C(14). The configuration of cyasterone (406), an insectmetamorphizing substance possessing the stigmastane skeleton, has been established by chemical correlation with ponasterone and ecdysone.22 The structures of isocyasterone (407) and epicyasterone (408), which are newly discovered insect moulting substances, have also been elucidated.222 The OH

*I8 '19 220

"' '*'

K . Nakanishi, Pitre Appl. Chem., 1971, 25, 167. Y. K. Chong, M. N. Galbraith, and D. J3.S . Horn, Chem. Comm., 1970, 1217. S. Imai, E. Murata, S. Fujioka, T. Matsuoka, M. Koreeda, and K. Nakanishi, J. Amer. Chem. SOC.,1970, 92, 7510. H. Hikino, K . Nomoto, and T. Takemoto, Tetrahedron, 1971, 27, 315; Chem. and Pharm. Bull. (Japan), 1970, 18, 2132. H . Hikino, K. Nomoto, and T. Takemoto, Chem. and Pharm. Bull. (Japan). 1971, 19, 433.

407


408

HO 0 (408)

absolute stereochemistry at C(20) and C(22) in ecdysones has been defined as 20R,23R.223 Moreover, ajugalactone (409), a novel insect moulting inhibitor, has been isolated and its structure fully established.224

OH

(410) a ; R 1 = R' = O H b ; R L = H ;R' = O€I c ; R ' = OH; R' = H ?'

M . Koreeda, D. A . Schooley, K. Nakanishi, and H. Hagiwara, J . Anzrr. Clirrn. Soc., 197 1 , 93. 4084. M . Koreeda, K . Nakanishi. and M . Goto. J . A t w r . C'hrini. Soc., 1970, 92, 7512.

'lJ

409

Steroid Synthesis

Several accounts have also been devoted to the synthesis of ecdysones and related substances. The analogues (410) of a-ecdysone (411) were prepared by a reaction sequence (Scheme 28) which involved bromination at position 5 of

(414)

(410a)

Reagents: i, Br,, HBr, AcOH, THF; ii, KOH, MeOH, 20 "C; iii, Br,, HBr, AcOH; iv, Li,CO,, D M F ; v, SeO,, dioxan; vi, KHCO,, MeOH, 50 "C.

Scheme 28

the 6-keto-steroid (412) and solvolysis with base. The 5b-hydroxy-6-ketoderivative (413) so formed was then brominated at C(7), and dehydrobrominated to give the A7-6-ketone (414). Introduction of the 14 a-hydroxy-group was performed by selenium dioxide oxidation. Base hydrolysis then provided the bis-deoxyecdysone analogue (4 10a).225 A new synthesis (Scheme 29) of a-ecdysone (411) from stigmasterol (388a) has been reported.226 By a nine-step process, (388a) was converted into the diketo-lactone (415). Selective reduction of the 3-carbonyl group and protection of the 6-ketone gave (416), which was converted into the 2,3-diol-6-ketone (417). After acetylation, bromination at C(7), and dehydrobromination, the enol acetate (418) was treated with peracid, and epoxide opening afforded the trihydroxy-ketone (419). Selective Grignard reaction with methylmagnesium bromide on the y-lactone group then provided a-ecdysone (411).226 225

M. J. Thompson, W . E. Robbins, C. F. Cohen, J. N. Kaplanis, S. R. Dutky, and R. F. Hutchins, Steroids, 1971, 17, 399. H. Mori, K. Shibata, K. Tsuneda, and M. Sawai, Tetrahedron, 1971, 27, 1157.


410

9 steps

( 3 8 t h ) --+

i, ii

1

(415)

111, I V , I , v. v1 t------

HO HO

HO 0

(417)

\ll-X

04

0

//

0

...+

xi,

xii

xiii

Reagents: i, NaBH,; ii, (CH,OH),, H + ; iii, CrO,, p y : iv, Bu'OK, 0,; v. MeCOMe, H'; vi, H,O+;vii,Ac,O,py;viii, Br,,AcOH,HBr;ix, L i ,CO , , D MF;x, A c , O , H'; xi, P h C 0 , H ; xii, K,CO,, MeOH-H,O; xiii, MeMgBr.

Scheme 29

41 1

Steroid Synthesis

Rubrosterone (424) has also been synthesized by an analogous route (Scheme 30).227 The 6-keto-diol (420) was converted into the trio1 (421). This, in turn, was transformed into the enol acetate (422), as indicated in Scheme 29. Reaction OH

OH

(420)

(423)

1

vii,

viii, i i ,

ix

0

(424) Reagents: i, (CH,OH), , H ;ii, CrO,, p y ; iii, Bu'OK, 0,; iv, NaBH,; v, +

vi, OH -; vii, MeCOMe, H ; viii, separation of isomers; ix, H,O +. +

Scheme 30

of (422) with monoperphthalic acid was followed by base treatment, yielding the tetrol (423) as a mixture of isomers at C(5). The isomers were separated as their 2,3-acetonides. Oxidation of the 17-hydroxy-group and acid-catalysed opening of the acetonide then afforded rubrosterone (424).22 ' l i

K. Shibata and H. Mori, Tetrahedron, 1971, 27, 1149.


41 2

THPO

HO

Finally, 23-deoxyantheridiol (425) has been isolated and its structure confirmed.228 Partial synthesis from the intermediate (426), led to an inactive isomer of (425).222 11 Steroidal Alkaloids A review of steroidal alkaloids derived from 3-aminopregnanes, 20-aminopregnanes, 3,20-diaminopregnanes, 3-aminoconanines, and 20-piperidylpregnanes has appeared.229 A partial synthesis of 12-substituted derivatives of N-demethyI-Sa-con-20(N)en-3-one from holarrhenine has been reported.230 Furthermore, it has been shown that treatment of the methoxylated 20-imino-steroid (427) with methylmagnesium iodide afforded the steroid A’-pyrroline derivative (428).231 In Me Me MeMgl

connection with studies of steroids possessing a heteroatom between C(18) and C(20), introduction of a hydroxy-group at C(14p) in 3p,lS-diacetoxy-5~pregn- 14-en-20-one (429) led to a partial synthesis of holantogenin (430).232 Sublimation of (430) provided anhydroholantogenin (431).232

’” 23” 231

’”

D. M. Green, J. A. Edwards, A. W. Barksdale, and T. C. McMorris, Tetrahedron, 1971, 27, 1199. Y. Sato, Chem. Alkaloids, 1970, 591. G. Lukacs, G. Roblot, A. Picot, and X. Lusinchi, Ann. pharm.frang. 1970, 28, 3 6 3 . J. P. Alazard and X. Lusinchi, Compt. rend. 1970, 271, C, 1386. P. Choay, C . Monneret, and Q. Khuong-Huu, Compt. rend., 1971, 272, C , 782,

Steroid Synthesis

413 Me

I

&=

Ac 0

Me

0

OH ___--

H (429)

HO (430)

A synthesis (Scheme 31) of solanidine (435a) from the 16-ketopregnane (432) has been reported.233 Alkylation of (432) with the nitro-ester salt (437) (see below) furnished the nitro-intermediate (433), which was cyclized by treatment with zinc in acetic acid and then hydrolysed with base to afford (434). Lithium aluminium hydride reduction then yielded solanidine (435a).23322-Isosolanidine (435b) was also prepared by an analogous route. In order to establish the configuration with certainty, (435b) was catalytically hydrogenated in the presence of platinum, conditions known to cause epimerization at C(22). Demissidine (436a) was' obtained, an alkaloid of defined structure and stereochemistry, thus establishing the configuration of (435b).233Tomatid-5-en-3/3-01(440) and solasodine (441) were also prepared, following an analogous synthetic pathway as shown in Scheme 32.234Alkylation of (432)as above, with the (S)-nitro-ester (437),gave the intermediate (438). Reduction of the 16-ketone, followed by zinc treatment, afforded the lactam (439). Lithium aluminium hydride reduction was followed by reaction with N-chlorosuccinimide and strong base to provide tomatid-5-en-3P-01 (440). Similarly, Michael addition of the (R)-isomer of (437)to (432),followed by the same sequence of reactions, completed the synthesis ofsolasodine (441).234 Acetyldemissidine(436b)reacted smoothly with cyanogen bromide in refluxing chloroform to give the bromocyanamide (442).23 s When this degradation product (442) was treated with lithium aluminium hydride in refluxing tetrahydrofuran, 233

234 235

S. V. Kessar, A . L. Rampal, S. S. Gandhi, and R . K. Mahajan, Terrahedroron, 1971, 27, 2153. S. V. Kessar, Y . P. Gupta, M. Singh, and R. K. Mahajan, Tetrahedron, 1971,27,2869. J. A. Beisler and Y. Sato, J. Chem. SOC.(0,1971, 149.

414


(432)

CO,Me

H0

(433)

HO

(434) iii

1

HO

H

(435) a ; 221~-H b ; 22B-H

R (436) a ; R b; R

= =

H AC

Reagents: i. Zn-AcOH; ii, OH ; iii, LiAIH,; iv, H,. PtO,, MeOH. ~

Scheme 31

Steroid Synthesis

415 MeH

(432)

+

O'N+co2Me

K+

-

(437)

HO

1

i, ii

Reagents: i, NaBH,, AcOH; ii, Zn, AcOH; iii, LiAIH,; iv, N C S - ; v, MeO- N a + .

Scheme 32

demissidine (solanidanol) (436a) was isolated from the reaction mixture. This unusual von Braun-retro-von Braun sequence was also observed in the solanidine series.235 Solasodine (441) has been converted into solafloridine (446) by ring E opening (see Scheme 33) giving (443). Reduction at C(5) and nitrogen protection afforded compound (444). Inversion of the configuration at C(16) provided the 16ahydroxy-derivative (445). Finally, introduction of the double bond into the heterocyclic ring gave (446).236 Similarly, solasodine (441) has been transformed 23h

G. Kusano, N. Aimi, and Y. Sato, J . Org. Chem., 1970, 35,2624.


416

(441)

I , I1

0

II

H0

AcO

H (445)

(444)

Reagents: i, AcOH-TsOH; ii, NaBH,; iii, H,-Pd/C; iv, PhCH,OCOCI; v, CrO,, H,SO,; vi, Na, Pr'OH; vii, NCS , CH,Cl,; viii, M e O - N a + . Scheme 33

into solacongestidine (447).236 In turn, solafloridine (446) has been converted into solanocapsine (45 1).237 The synthetic sequence (Scheme 34) involves allylic oxidation in the heterocyclic ring, giving the vinylogous amide (448). 237

H. Ripperger, F. J. Sych, and K. Schreiber, Tetrahedron Letters, 1970, 5 2 5 1 .

Steroid Synthesis

417

(446) i , ri

1

PhCH2-0

r I

c=o

iiikv 4

I

.OAc

--

(448)

OAC

(449)

0

PhCH,-0-C-N

u HO HOW (451)

H (450)

Reagents: i, Ac,O, ZnCI,; ii, MnO,, CHC1,; iii, NaBH,; iv, PhCH,OCOCI; v, Jones reagent; vi, KOH, MeOH; vii. CrO,, p y ; viii, HBr, A c O H ; ix, NH,OH; X, H,-Pt, AcOH.

Scheme 34


418

Ac

I

1

i, ii

Ac

J/

iii, 11, IV,

v,

II

Ac

Reagents: i. dioxan, A ; i i , O H - : iii, N a B H , ; iv. separation of isomers; v, A c , O ; vi, Jones reagent; vii, Br,; viii, NaI; ix, CrCI2; x, MeC(OAc):CH,, H ‘ ; xi, NaBH,, E t O H ; xii, KOH, (CH,OH),, N,H,.

Scheme 35

Reduction, nitrogen protection, and oxidation then gave the keto-derivative (449), which was cyclized to (450). Finally, the 3p-amino-group in (451) was introduced by catalytic reduction of the ~ x i m e . ~ ~ ’

Steroid Synthesis

419

Various partial synthetic approaches to veratramine (456) have been reported.238*239 The full paper on the synthesis of (456) has now appeared. The route which was followed (Scheme 35) involves condensation of the D-homoc-nor-steroid fragment (452) with the enamine (453), yielding, among others, the desired intermediate (454).240 Reduction of the carbonyl group in (454), mild hydrolysis, and separation of the correct isomer gave, after acetylation and

K-" I

:

H

I

-0

Reagents: i, Ac,O, py; ii, Jones reagent; iii, (CH,SH),, H ' ; iv, Raney nickel; v, KOH, (CH,OH), , N2H,.

Scheme 36

selective hydrolysis at C(3),the pentacyclic derivative (455). Finally, the introduction of the A5-double-bond which is present in veratramine (456) was performed by a classical reaction sequence.240 Veratramine (456) was converted into veranine (458) as shown in Scheme 36.241 This was achieved by partial acetylation of (456), followed by oxidation of the heterocyclic hydroxy-group to the keto-intermediate (457). Formation of the thioketal, desulphurization, and hydrolysis in strong base thus afforded veranine (458).241 Batrachotoxin (459), the steroidal alkaloid from the poison arrow frog Phyllobates aurotaenia, continues to engage the attention of chemists and pharmac o l o g i s t ~ .This ~ ~ ~substance exerts novel, selective effects on electrogenic membranes. In many cases this activity can be explained in terms of an irreversible increase in permeability to sodium ions. The subsequent reactions promoted by (459)can be blocked reversibly by t e t r o d o t o ~ i n . ~ ~ ~ , ~ ~ ~ 238

239

241

242

243 244

E. Brown, M. Ragault, and J. Touet, Bull. SOC.chim. France, 1971, 2195. J. W. Huffman and R. R. Sobti, Steroids, 1970, 16, 7 5 5 . T . Masamune, M. Takasugi, and A. Murai, Tetrahedron, 1971, 27, 3369. T. Masamune, I. Yamazaki, K. Orito, and M. Takasugi, Tetrahedron, 1971, 27, 3387. T. Tokuyama, J. Daly, and B. Witkop, J. Amer. Chem. Soc., 1969, 91, 3931. E. X. Albuquerque, J. W. Daly, and B. Witkop, Science, 1971, 172, 995. B. Witkop, Experientia, 1971, 27, 1121.


420

The synthesis of the steroidal moiety of batrachotoxin (459) has been attempted. In a first pilot-reaction sequence, 5~0,19N-[ep(oxyethano-N-methylimino)landrostsn-17-01 (461) was prepared from 17)~9-diacetoxyandrost-4-en-3-one (460) by a multi-step sequence.245 The technique developed in this model synthesis was then applied to build the C(14kC(18) ring of 3j3,20

Terpenoids and Steroids: Volume 2 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 4 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 6 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 9 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 11 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 5 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 10 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 7 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 1 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 8 (Terpenoids & Steroids)

Terpenoids and Steroids: Volume 12 (Specialist Periodical Reports)

Terpenoids and Steroids: Volume 3 (Specialist Periodical Reports)

Steroids

Steroids

Anabolic Steroids

Laymans Guide to Steroids Part 2

Layman s Guide to STEROIDS Part I

CHROMATOGRAPHY OF STEROIDS, Volume 8 (Journal of Chromatography Library)

Laymans Guide to Steroids Part 3

Organophosphorus Chemistry - Volume 22 (SPR Organophosphorus Chemistry (RSC))

Chemical Wizardry - Anabolic Steroids for Bodybuilders

Photochemistry (SPR Photochemistry (RSC)) (Vol 30)

Photochemistry (SPR Photochemistry (RSC)) (Vol 31)

Organophosphorus Chemistry - Volume 15 (SPR Organophosphorus Chemistry (RSC))

Photochemistry (SPR Photochemistry (RSC)) (Vol 29)

Inhaled Steroids in Asthma: Optimizing Effects in the Airways

Organophosphorus Chemistry (SPR Organophosphorus Chemistry (RSC)) (v. 35)

Organophosphorus Chemistry (SPR Organophosphorus Chemistry (RSC)) (v. 6)

Organophosphorus Chemistry (SPR Organophosphorus Chemistry (RSC)) (Vol 26)

Organophosphorus Chemistry (SPR Organophosphorus Chemistry (RSC)) (v. 8)

Theoretical Chemistry (SPR Theoretical Chemistry (RSC)) (v. 3)

Terpenoids and Steroids: Volume 2 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 4 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 6 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 9 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 11 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 5 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 10 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 7 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 1 (SPR Terpenoids and Steroids (RSC))

Terpenoids and Steroids: Volume 8 (Terpenoids & Steroids)

Terpenoids and Steroids: Volume 12 (Specialist Periodical Reports)

Terpenoids and Steroids: Volume 3 (Specialist Periodical Reports)

Steroids

Steroids

Anabolic Steroids

Laymans Guide to Steroids Part 2

Layman s Guide to STEROIDS Part I

CHROMATOGRAPHY OF STEROIDS, Volume 8 (Journal of Chromatography Library)

Laymans Guide to Steroids Part 3

Organophosphorus Chemistry - Volume 22 (SPR Organophosphorus Chemistry (RSC))

Chemical Wizardry - Anabolic Steroids for Bodybuilders

Photochemistry (SPR Photochemistry (RSC)) (Vol 30)

Photochemistry (SPR Photochemistry (RSC)) (Vol 31)

Organophosphorus Chemistry - Volume 15 (SPR Organophosphorus Chemistry (RSC))

Photochemistry (SPR Photochemistry (RSC)) (Vol 29)

Inhaled Steroids in Asthma: Optimizing Effects in the Airways

Organophosphorus Chemistry (SPR Organophosphorus Chemistry (RSC)) (v. 35)

Organophosphorus Chemistry (SPR Organophosphorus Chemistry (RSC)) (v. 6)

Organophosphorus Chemistry (SPR Organophosphorus Chemistry (RSC)) (Vol 26)

Organophosphorus Chemistry (SPR Organophosphorus Chemistry (RSC)) (v. 8)

Theoretical Chemistry (SPR Theoretical Chemistry (RSC)) (v. 3)

Recommend Documents