| dc.description.abstract | The thesis entitled “Synthesis of Steroids: Synthetic Investigations on Oestrone” is divided into two chapters.
Chapter I consists of a comprehensive review of earlier work on the chemistry of oestrone, dealing with its isolation, structural elucidation, biosynthesis, and various total syntheses.
Chapter II is divided into two sections, A and B.
Section A
Section A deals with a discussion of the present investigation on the total synthesis of oestrone, the main objective being the development of a short and stereospecific route towards its synthesis starting from the easily available 6-methoxytetralone.
Condensation of 6-methoxytetralone under aqueous alkaline conditions with glyoxylic acid, liberated in situ from tartaric acid by metaperiodic acid oxidation, afforded -keto-6-methoxy--tetralidene acetic acid (IA) in excellent yield. This acid (IA), on treatment with dimethyl sulphate and potassium carbonate, gave the corresponding methyl keto ester (IB) in very good yield.
Michael condensation of the unsaturated keto acid (IA) with -methylacetoacetonitrile in the presence of aqueous sodium hydroxide at 110° yielded two isomeric tricyclic cyano keto acids (IIA and IIB) in an approximate ratio of 10:1 respectively. The preponderant isomer (IIA) could also be obtained in one step, though in very poor yield, when the bicyclic keto acid (M), formed by the aforementioned Claisen condensation of methoxytetralone with glyoxylic acid, was treated in situ with -methylacetoacetonitrile at 110°.
Michael condensation of the bicyclic keto ester (IB) with -methylacetoacetonitrile in the presence of sodium ethoxide in ethanol gave two isomeric cyano keto esters (IIIA and IIIB) in an approximate ratio of 4:1 respectively. Esterification of the preponderant cyano keto acid (IIA) with diazomethane gave the corresponding cyano keto ester (IIIA). Similarly, the less preponderant cyano keto acid (IIB), which could not be isolated in the pure state, on treatment with diazomethane followed by purification afforded the corresponding less preponderant cyano keto ester (IIIB).
Chemical proof in favour of the phenanthrene nucleus of the key intermediate, the tricyclic cyano keto acid (IIA), was provided by converting it into 1,2-dimethyl-7-methoxyphenanthrene. To this end, (IIA) was refluxed with hydrobromic acid and acetic acid to furnish the phenolic acid (IV) in good yield. Treatment of (IV) with dimethyl sulphate and potassium carbonate gave the methylated keto ester (V) in good yield. The diol (VI), obtained by lithium aluminium hydride reduction of (V), was dehydrogenated with selenium in a sealed tube at 330° to afford 1,2-dimethyl-7-methoxyphenanthrene, the identity of which was proved by comparison with an authentic sample.
That the two isomers (IIIA and IIIB) differed only in the stereochemical configuration at the asymmetric centre was proved by the fact that (IIIB) also, on hydrolysis, demethylation and decarboxylation by treatment with hydrobromic acid–acetic acid mixture, followed by methylation and esterification with dimethyl sulphate and potassium carbonate, gave the same unsaturated keto ester (V) as that obtained from the cyano keto acid (IIA) by similar treatment.
Considering that the operation of electrical repulsion between the ester carbonyl and the cyano dipoles was likely to be greater in the cis-relationship of the two groups than in the trans-relationship, and also the fact that the electrical factor might dominate the normal steric factor, the conformations of the CH and CN groups were assumed to be axial and equatorial in the preponderant isomer. It therefore followed that the less preponderant isomer would have the CH and CN equatorial and axial respectively.
Catalytic hydrogenation of the aforementioned unsaturated keto ester (V) gave two isomeric saturated hydroxy esters (VIIA and VIIB) in almost equal proportions. These two alcohols on oxidation gave two different keto esters (VIIIA and VIIIB), thus indicating the difference in the stereochemistry of the B/C-ring junction in alcohols (VIIA and VIIB).
Considering that adsorption on the catalyst surface and the concomitant addition of hydrogen are possible from both sides of the unsaturated keto ester (V), a study of molecular models showed that simultaneous hydrogenation of the 9,11-olefinic bond and the carbonyl at C-17 from the side containing the equatorial CH group should lead to an equatorial hydroxyl at C-17 and a trans-fusion of the B/C rings, while hydrogenation from the opposite side should result in an axial hydroxyl and a cis-fusion of the B/C rings.
The axial and equatorial conformations of the hydroxyl groups in (VIIA and VIIB) respectively were determined by oxidation and esterification reactions. Accordingly, the B/C-ring fusion in (VIIA and VIIB) was assigned cis and trans configurations respectively. Additional proof in favour of the equatorial conformation of the hydroxyl group and the trans-fusion of the B/C rings in (VIIB) was obtained when the unsaturated keto acid (IX), on reduction with sodium in liquid ammonia followed by esterification with diazomethane, gave the hydroxy ester (VIIB).
A study of the molecular model of the unsaturated cyano keto ester (IIIA) indicated that the side containing the bulky axial CH group would be sterically hindered for adsorption on the catalyst surface; thus hydrogenation might lead mainly to a single product with trans B/C-ring fusion and an equatorial OH group. Catalytic hydrogenation of the preponderant cyano keto ester (IIIA) gave, as anticipated, a single crystalline saturated hydroxy cyano ester (X), m.p. 169–170°, in quantitative yield. The equatorial conformation of the hydroxyl group in (X) was determined by oxidation and esterification reactions, and hence the B/C-ring fusion was trans.
The saturated hydroxy cyano ester (X), m.p. 169–170°, on refluxing with phosphorus oxychloride and pyridine gave the crystalline chloro cyano ester (XI) in 60% yield. Dechlorination of (XI) gave a mixture of 1-methoxycarbonyl-2-methyl-7-methoxy-1,4,9,10,11,12-hexahydrophenanthrene (XII) and the desired methyl cyano ester (XIII) in the ratio of 3:2 respectively. Later, the yield of the methyl cyano ester could be considerably increased by effecting the dechlorination of (XI) with zinc and formic acid instead of acetic acid.
Selective saponification of (XIII) afforded the expected methyl cyano acid (XIV), which on Arndt–Eistert homologation via its acid chloride failed to furnish the homologous cyano ester (XV). Other attempts to convert (XIV) into the desired homologous cyano ester (XV) were also unsuccessful. | |
