Effect of synthetic steroids on GABAA receptor binding in rat brain.

Neuroactive steroids, like allopregnanolone (A) and pregnanolone (P), bind to specifics sites on the GABAA receptor complex and modulate receptor function. They are capable to inhibit or stimulate the binding of GABAA receptor-specific ligands, like t-butyl-bicyclophosphorothionate, flunitrazepam and muscimol. We have previously characterized a set of oxygen-bridged synthetic steroids (SS) analogs to A or P using synaptosomes. Considering that the subunit composition of the GABAA receptor throughout the central nervous system affects the magnitude of the modulation of the GABAA receptor by NAS, we evaluated the action of two selected SS, in brain sections containing the cerebral cortex (CC) and hippocampus (HC) using quantitative receptor autoradiography. Both SS affected the binding of the three ligands in a similar way to A and P, with some differences on certain CC layers according to the ligand used. One of the SS, the 3α-hydroxy-6,19-epoxypregn-4-ene-20-one (compound 5), behaved similarly to the natural neuroactive steroids. However, significant differences with compound 5 were observed on the HC CA2 region, making it steroid suitable for a specific action. Those differences may be related to structural conformation of the SS and the subunits' composition present on the receptor complex.

Synthesis and GABA(A) receptor activity of 6-oxa-analogs of neurosteroids.

The 6-oxasteroids 3alpha-hydroxy-6-oxa-5alpha-pregnan-20-one (3) and 3alpha-hydroxy-6-oxa-5beta-pregnan-20-one (4) were obtained from pregnenolone acetate via the corresponding (5alpha or 5beta) 3beta, 20beta-diacetoxy-6-oxa-pregnane. Both steroids showed ca. 100-fold reduced potency for modulating [(3)H]flunitrazepam, [(3)H]muscimol or [(35)S]TBPS binding to the GABA(A) receptor when compared to their natural carbon analogs 3alpha-hydroxy-5alpha-pregnan-20-one (1) and 3alpha-hydroxy-5beta-pregnan-20-one (2).

The glucocorticoid properties of the synthetic steroid pregna-1,4-diene-11beta-ol-3,20-dione (deltaHOP) are not entirely correlated with the steroid binding to the glucocorticoid receptor.

The natural steroid 11beta-hydroxyprogesterone is not only a modulator of 11beta-hydroxy-steroid dehydrogenase activity, but also an efficient inducer of tyrosine aminotransferase activity in hepatocytes. In contrast with the low affinity for the mineralocorticoid receptor. 11beta-hydroxyprogesterone binds well to both the glucocorticoid receptor and the carrier protein transcortin. It is accepted that the introduction of a 1:ene double bond into 3-keto 4:ene steroids increases the glucocorticoid potency, so that 3-keto-1,4:diene steroids show improved chemical stability and are more potent glucocorticoids than their respective 4:ene analogs. The steroid pregna-1,4-diene-11beta-ol-3,20-dione (deltaHOP) had previously been described as an anti-inflamatory compound and an inhibitor of macromolecular biosynthesis in thymocytes and lymphocytes. In such studies, deltaHOP also exhibited some particular glucocorticoid properties which made it attractive as a tool for the study of the mechanism of action of glucocorticoids. In the present paper we show that deltaHOP possesses some classical biological actions of glucocorticoids such as deposition of glycogen in rat liver, induction of TAT activity in hepatocytes, and inhibition of the uptake of leucine and thymidine by thymocytes. It also exhibits minimal sodium-retaining properties. Consistent with these biological effects, deltaHOP shows a 70 times lower relative binding affinity for the mineralocortioid receptor than aldosterone, but a reasonable affinity for the glucocorticoid receptor, and is as efficient as dexamethasone in dissociating the 90 kDa heat shock protein from the glucocorticoid receptor heterocomplex. However, the inhibition of the uptake of amino acids and nucleotides observed in the presence of deltaHOP is not efficiently blocked when thymocytes are coincubated in the presence of steroids with known antiglucocorticoid activity. deltaHOP is similarly inefficient in inducing chloramphenicol-acetyl transferase activity in cells transfected with a plasmid that possesses two canonical glucocorticoid-responsive elements. Unlike most glucocorticoids, deltaHOP does not induce the fragmentation of DNA in a regular pattern characteristic of apoptosis and it does not reduce thymus weight. This unusual dissociation of glucocorticoid parameters makes deltaHOP a useful tool to discriminate between mechanisms of action by which steroids can exert their biological effects.

Synthesis of oxido-bridged analogs of 18-hydroxyprogesterone.

18-Hydroxy-6,19-oxidoprogesterone and 18-hydroxy-11, 19-oxidoprogesterone were synthesized from readily available materials. The functionalization of C-18 was accomplished with phenyliodosodiacetate/iodine, whereas that of C-19 was carried out with the mercuric oxide/iodine system, both under irradiation with visible light. For 18-hydroxy-11,19-oxidoprogesterone, C-19 was functionalized before C-18, whereas the reverse order had to be used for the 6,19-oxido derivative.

Sodium-retaining activity of some natural and synthetic 21-deoxysteroids.

The effect of progesterone and six other C21-deoxysteroids on renal sodium retention by male adrenalectomized rats was compared with the effect exerted by the natural corticoids aldosterone, 11-deoxycorticosterone, and corticosterone. Steroids were active in the following order: aldosterone > 11,19-oxidoprogesterone > 5 alpha H-3,20-pregnanedione > or = 5 beta H-3,20-pregnanedione > progesterone = 11-ketoprogesterone > 6,19-oxidoprogesterone = 11-keto-6,19-oxidoprogesterone > or = corticosterone. All C21-deoxysteroids, except 11,19-oxidoprogesterone, exhibited parabolic log dose-response functions, indicating an effect that opposes renal sodium retention at high doses. 11,19-Oxidoprogesterone and the natural corticoids exhibited normal, exponential, log dose-response curves. Diverse geometric parameters related to molecular planarity were calculated and their correlation with biopharmacological properties was attempted. The best linear regression was obtained for correlation of the concavity of log dose-response parabolas (second-order coefficients) of C21-deoxysteroids with the C3 = O/ring D angle of these molecules. A good linear regression could also be obtained for correlation of the affinity of C21-deoxysteroids, except 11,19-oxidoprogesterone, for purified type I mineralocorticoid receptors with those angles. The latter correlation deteriorated upon incorporation of the affinity data for the three natural corticoids, due to similar affinities of these hormones for type I mineralocorticoid receptors, but could be restored when the binding data for the unpurified, corticosterone-binding globulin-containing stage of the receptors were considered. In vivo binding data followed the same trend as that for unpurified receptors.