H. Hirschmann

Prochiral and pseudoasymmetric centers: Implications of recent definitions

Abstract New definitions of prochiral and of pseudoasymmetric elements have been proposed by Prelog and Helmchen [Helv. Chim. Acta 55, 2581 (1972)]. The revised definition of prochiral centers would fail to identify many centers bearing like ligands that can be distinguished experimentally. The new definition of the pseudoasymmetric center would cause some that were previously so designated to be considered chiral. Whenever this occurs, concepts like retention and inversion of configuration would no longer have their customary meaning. Concomitant changes in the sequence subrules would not always provide reliable information about the possibilities of stereoisomerism. A modification of the rules is outlined which would avoid this, but secure similar benefits.

C-22 ketals related to the sapogenins☆

Abstract Reduction of 5,6-dihydrokryptogenin diacetate with Raney nickel in ethanol reduced the 16-ketone group and gave a mixture of a 22-ethyl ketal with a hemiketal. Both undergo elimination reactions very readily to yield the Δ22-bond isomer of pseudotigogenin diacetate. The conversions are reversible and are catalyzed by acetic acid and in neutral solution by calcium sulfate and other surfaces. Alkaline hydrolysis followed by cyclization with acetic acid converted both the ketal and the Δ22-olefin to tigogenin. The relative stabilities of ketal epimers at C-22 are discussed and related to the stereo-chemistry of the sapogenins and the cyclopseudosapogenins.

Carbon-oxygen cleavage in the acid hydrolysis of androsterone sulfate

Abstract Whereas all prior studies of the cleavage of any 5α-steroid 3-sulfate to the alcohol had given only the one which would result from the rupture of the sulfur-oxygen bond, the hydrolysis of androsterone sulfate with boiling hydrochloric acid gave in addition to androsterone, 3β-hydroxy-5α-androstan-17-one, 2α-hydroxy-5α-androstan17-one and 2β-hydroxy-5α-androstan-17-one. The two rearrangement products were identified by partial syntheses. The 5α-androsten-17-one which accompanied these hydrolysis products of androsterone sulfate was shown by infrared spectroscopy to be a 4 : 1 mixture of the Δ 2 and Δ 3 -olefin which could be freed of the latter by formation of the dibromide or the bromohydrin, and subsequent treatment with zinc.

Empirical valence-force calculations of steroids—I : Conformations and relative energies of the 20-epimers of 5α-pregnane-3β,20-diol

Abstract Quantitative valence-force calculations (energy minimization by the Westheimer-Allinger-Boyd method) have been carried out on side chain conformers of both 5α-pregnane-3β,20-diols. The calculations allowed full relaxation of all internal coordinates of these steroids (comprising 59 atoms), but the large number of non-bonded interactions had to be reduced by imposing an arbitrary limit of 5 A, bringing the total number of interactions (stretching, bending, torsion, non-bonded, 1,3, Coulomb) below the maximum load (1200) of the computer. Four rotamers of each diol, 20α and 20β, were examined differing with respect to the distribution of the groups about the bonds between C-17 and C-20 and between C-20 and oxygen. The results indicate that non-bonded strain between the side chain and the steroidal nucleus is largely accommodated by distortion of bond angles, but little by twisting of the groups about the C-17, C-20 bond. Two rotamers appear to predominate to a large extent in the conformational equilibrium of the 20β-epimer. Both forms have the hydrogen at C-20 anti to that at C-17 but differ in the OH orientation. In contrast the calculations for the 20α-epimer show that a substantial fraction occurs as an additional rotamer with a gauche distribution of the 17α- and 20-hydrogens. The published proton spin coupling constants agree well with the calculated results. A new interpretation, based on the existence of a conformational equilibrium is, therefore, offered to account for the published NMR data for the 20α-epimer. In addition “conformation instability contributions” of relevant side chain atoms have been calculated.

Formolysis of 3β-acetoxy-5α-pregnan-20α-yl p-toluenesulfonate

Abstract In contrast to its 20-epimer, the formolysis of 3β-acetoxy-5α-pregnan-20α-yl tosylate proceeds predominantly without enlargement of the D ring. The products were identified as 17β-methyl-18-nor-5α, 17α-pregn-13-en-3β-yl acetate (2b), 3β-acetoxy-5α-pregnan-20α-yl formate (13), 3β-acetoxy-17-α-methyl-D-homo-5α-androstan-17αβ-yl formate (14). 5α-pregn-20-en-3β-yl acetate (12b) and 3β-acetoxy-5α-pregnan-20β-yl formate (15). During the degradation of 2b, two cis-13,14 glycols were prepared one of which had an unusually large Δv for the OH stretching bands of the hydrogens that do and that do not participate in hydrogen bonding. This large value (99 cm−1) is attributed to molecular deformation of the 13α,14α-diol. This configuration and the relative yield of this glycol would indicate that the β face of the 13,14 double bond in 2b is more accessible to the addition of OsO4 and probably of other reagents. The catalytic hydrogenation of this double bond is reported. A possible mechanism is suggested for the unexpected formation of the uranediol derivative (14).

Reflection-concordant stereospecific numbering

Abstract If enantiomers are to receive names that agree except for differences in the configurational affixes, it is not sufficient that all descriptors of pheroehiral elements are inverted on reflection. It is equally essential that the numbering does not change when the enantiomeric pair is interconverted. A set of rules is suggested which would accomplish this and which would provide for the differentiation by fixed numbering of all the numbered atoms of a molecule that are sterically distinct. As our definition of steric elements that are only graphochiral is basic to the formulation of the numbering rules, we have reaffirmed its appropriateness after considering the recently introduced pseudochirality operations.