South Parks Road, Oxford OX1 3QYStructural investigation of the products obtained by microbiological hydroxylation of substrates in the androstane, pregnane, and mstrane series has involved examining the l H n.m.r. spectra of 344 steroids, 243 of them being new compounds. Mild oxidation of the products gives polyketones in which the positions of the oxo-groups are characterised by the chemical and solvent shifts of the 18-H and 19-H signals. This information is supplemented by the chemical shifts of the products' 18-H and 19-H signals, and by the positions and form of their )CH-OH resonances : the latter are especially useful in establishing the configurations of the hydroxy-groups.FOR the past five years we have been studying the hydroxylation of mono-and di-oxygenated steroids with a range of micro-0rganisrns.l The intention was to vary the positions of the oxygen substituents around the steroid nucleus in a systematic manner: it was -gate, ibid., 1969, 463. hoped that the use of relatively simple substrates would facilitate interpret at ion of the microbiological behaviour. A considerable body of results has accrued from examining monoketones, keto-alcohols, and diketones derived from handrostane, 5a-pregnane, and 5a-cestrane ; a series of papers describing the work will be submitted shortly to this Journal. The preparation of substrates and related chemical studies are being recorded in a separate series2

No abstract

A GABA(A) receptor study of several B-nor analogues of allopregnanolone and pregnanolone has been carried out. B-norallopregnanolone (i.e., 3alpha-hydroxy-7-nor-5alpha-pregnan-20-one) was found comparable to allopregnanolone when measured with labeled TBPS. Analogous results were obtained from their effect on neurons in culture: this time, both 3alpha-hydroxy-7-nor-5xi-pregnan-20-ones (5 and 6) were found to stimulate [3H]flunitrazepam binding and GABA-induced 36Cl- influx. These effects were inhibited by GABA(A) receptor antagonists. Other analogues carrying electronegative substituents (epoxides 9 and 10 and ketone 12) in the B ring were inactive. Similarly, B-normal ketones 17, and 18 and 6-azasteroids 20 and 21 were also inactive. B-Nor analogues 5 and 6 did not induce neurotoxicity at relevant concentrations. A computational analysis of active and inactive neurosteroid analogues allowed the proposal of a 3D pharmacophoric hypothesis of their interaction with the GABA(A) receptor.