The environmental estrogenic compound bisphenol A exerts estrogenic effects on mouse hippocampal (HT-22) cells: neuroprotection against glutamate and amyloid beta protein toxicity

We have examined using immortalized clonal mouse hippocampal cell line (HT-22) whether the environmental estrogenic compound bisphenol A (BPA), like estrogen, has any neuroprotective effect against glutamate and amyloid beta protein-induced neurotoxicity. BPA protects HT-cells against both 5 mM glutamate and 2 microM amyloid beta protein-induced cell death in a dose dependent manner. Optimum protection was attained at 1 microM and 500 nM BPA against 5 mM glutamate and 2 microM amyloid beta protein-induced HT-22 cell death, respectively. Using confocal immunoflourescence microscopy technique, we observed that 20 h of treatment with 5 mM glutamate resulted in intense nuclear localization of the glucocorticoid receptors (GR) in HT-22 cells as compared to control untreated cells. Interestingly, 1 microM BPA treatment for 24 h, followed by 20-h treatment with 5 mM glutamate, resulted in dramatic reduction in GR nuclear localization. We conclude that: (i) BPA mimics estrogen and exerts neuroprotective effects against both neurotoxins used; (ii) BPA inhibits enhanced nuclear localization of GR induced by glutamate; and (iii) HT-22 cells provide a good in vitro model system for screening the potencies of various environmental compounds for their estrogenic activity.     

Pregnenolone Protects Mouse Hippocampal (HT-22) Cells against Glutamate and Amyloid Beta Protein Toxicity

In the present work we have examined whether the neurosteroid pregnenolone has any neuroprotective effects against glutamate and amyloid beta protein neurotoxicity using immortalized clonal mouse hippocampal cell line (HT-22). The neurosteroid pregnenolone protects HT-22 cells against both 5 mM glutamate and 2 M amyloid beta protein induced cell death in a concentration dependent manner. Optimum protection was attained at 500 nM pregnenolone, against both 5 mM glutamate as well as 2 M amyloid beta protein induced HT-22 cell death. Furthermore, using confocal immunoflourescence microscopy we observed that 20 hours of treatment with 5 mM glutamate resulted in intense nuclear localization of the glucocorticoid receptor (GR) in HT-22 cells as compared to control untreated cells. Interestingly, 500 nM pregnenolone treatment for 24 hours, followed by 20 hours treatment with 5 mM glutamate resulted in dramatic reduction in GR nuclear localization. These results show that (i) pregnenolone has neuroprotective effects against both glutamate and amyloid beta protein neuropathology and (ii) prevention of glucocorticoid receptor (GR) localization to the nucleus may be involved in the observed neuroprotective effects of pregnenolone against glutamate neurotoxicity.     

Tamoxifen protects clonal mouse hippocampal (HT-22) cells against neurotoxins-induced cell death

In the present work using an established clonal mouse hippocampal (HT-22) cell line, we have examined whether the estrogen antagonist tamoxifen antagonizes the observed neuroprotective effects of estrogen against glutamate and amyloid beta protein neurotoxicity. Results obtained suggest that like estrogen, tamoxifen protects HT-22 cells against both 5mM glutamate and 2 microM amyloid beta protein induced cell death in a concentration dependent manner. Optimum protection was obtained at 500 nM tamoxifen. Tamoxifen was found to offer more potent protection at this dose against amyloid beta protein induced neurotoxicity when compared with glutamate neurotoxicity. We were unable to detect either estrogen receptor (ER)--ER alpha or ER beta presence in HT-22 cells using western blot technique. However, amyloid beta protein treatment significantly increases total glucocorticoid receptors (GRs) as determined by western blot technique, while prior treatment with estrogen or tamoxifen followed by amyloid beta protein resulted in the reduction of total GRs to the levels comparable to that observed for the control untreated cells. In addition, using confocal immunoflourescence microscopy technique, we observed that 20 h of treatment with 2 microM amyloid beta protein resulted in enhanced nuclear localization of GRs in HT-22 cells as compared to control untreated cells or 500 nM tamoxifen alone treated cells. Interestingly, 500 nM tamoxifen treatments for 24h, followed by 20 h treatment with 2 microM amyloid beta protein resulted in dramatic reduction in GRs nuclear localization. In conclusion, tamoxifen (i) protects HT-22 cells against amyloid beta protein neurotoxicity and (ii) neuroprotective effect is independent of ERs.     

Heat shock preconditioning and pretreatment with glucocorticoid antagonist RU 486 protect rat myogenic cells H9c2 against glutamate-induced cell death

We have observed that the treatment of rat-heart derived H9c2 myoblasts for 20 h with the excitatory amino acid glutamate resulted in cell death in a dose dependent manner as determined by LDH release. The optimum cardiotoxicity was seen at 25 mM glutamate. Preconditioning with either sublethal heat shock (42°C for 30 min) or pretreatment with 500 nM of the glucocorticoid antagonist RU 486 for 24 h almost completely protected H9c2 cells against subsequent 20 h treatment with 25 mM lethal glutamate. In addition, we have observed that glutamate treatment resulted in intense nuclear localization of glucocorticoid receptors (GR) in H9c2 cells as judged by the confocal immunofluorescence microscopy. Furthermore, pretreatment with either heat shock or RU 486 followed by glutamate treatment resulted in dramatic decrease in GR nuclear localization which was almost comparable to that observed with control untreated cells. In conclusion, we have shown for the first time using H9c2 cells that (i) protection from glutamate cardiotoxicity occurs with prior treatment with sub lethal heat shock or RU 486 and (ii) these measures down regulate the intense nuclear localization of GR induced by glutamate. The block to GR nuclear localization is likely to be involved in cardioprotective effects offered against glutamate toxicity by pretreatment with heat shock or RU 486.     

GABA(A)-mediated toxicity of hippocampal neurons in vitro

In the present study, we examined whether the elevation of GABA by gamma-vinyl-GABA protects cultured rat fetal hippocampal neurons against toxicity induced by a 20-min incubation with 100 microM L-glutamate. Neither a 24-h pretreatment nor posttreatment with gamma-vinyl-GABA (100 microM) had any neuroprotective effects, as determined by counting microtubule-associated protein-2 positive cells and lactate dehydrogenase assay 24 h after the glutamate treatment. Unexpectedly, gamma-vinyl-GABA alone induced a 20% loss of microtubule-associated protein-2-positive cells in a culture that was grown in medium containing 25 mM KCl. The toxic effect of gamma-vinyl-GABA was mimicked by a 24-h treatment with GABA (100 microM) and the GABA(A) receptor agonist, muscimol (10 microM), but not the GABA(B) receptor agonist, baclofen (10 microM). The GABA(A) receptor antagonist, bicuculline (10 microM), protected against gamma-vinyl-GABA and GABA-evoked toxicity. Neither gamma-vinyl-GABA nor GABA was toxic in culture medium containing 15 mM KCl. These data indicate that, under depolarizing conditions, an increased GABA level is toxic for a subpopulation of developing hippocampal neurons in vitro. The effect is GABA(A) receptor-mediated. These data provide a new view for understanding neurodegenerative processes, and raise a question of the safety of therapies aimed at increasing GABA concentration following brain insults, especially in immature brains.     

Neuroprotective effects of arachidonic acid against oxidative stress on rat hippocampal slices

Arachidonic acid (AA), 5,8,11,14-eicosateraenoic acid is abundant, active and necessary in the human body. In the present study, we reported the neuroprotective effects and mechanism of arachidonic acid on hippocampal slices insulted by glutamate, NaN(3) or H(2)O(2)in vitro. Different types of models of brain injury in vitro were developed by 1mM glutamate, 10mM NaN(3) or 2mM H(2)O(2). After 30 min of preincubation with arachidonic acid or linoleic acid, hippocampal slices were subjected to glutamate, NaN(3) or H(2)O(2), then the tissue activities were evaluated by using the 2,3,5-triphenyltetrazolium chloride method. Endogenous antioxidant enzymes activities (SOD, GSH-PX and catalase) in hippocampal slices were evaluated during the course of incubation. MK886 (5 microM; a noncompetitive inhibitor of proliferator-activated receptor [PPAR]alpha), BADGE (bisphenol A diglycidyl ether; 100 microM; an antagonist of PPARgamma) and cycloheximide (CHX; 30 microM; an inhibitor of protein synthesis) were tested for their effects on the neuroprotection afforded by arachidonic acid. Population spikes were recorded in randomly selected hippocapal slices. Arachidonic acid (1-10 microM) dose dependently protected hippocampal slices from glutamate and H(2)O(2) injury (P<0.01), and arachidonic acid (10 microM) can significantly improve the activities of Cu/Zn-SOD in hippocampal slices after 1h incubation. In addition, 10 microM arachidonic acid significantly increased the activity of Mn-SOD and catalase, and decreased the activities of Cu/Zn-SOD to control value after 3h incubation. These secondary changes of SOD during incubation can be reversed by indomethacine (10 microM; a nonspecific cyclooxygenase inhibitor) or AA 861 (20 microM; a 5-lipoxygenase inhibitor). Its neuroprotective effect was completely abolished by BADGE and CHX. These observations reveal that arachidonic acid can defense against oxidative stress by boosting the internal antioxidant system of hippocampal slices. Its neuroprotective effect may be mainly mediated by the activation of PPARgamma and synthesis of new protein in tissue.     

Circulating neuroactive C21- and C19-steroids in young men before and after ejaculation

Twelve neuroactive and neuroprotective steroids, androgens and androgen precursors i.e. 3alpha,17beta-dihydroxy-5alpha-androstane, 3alpha-hydroxy-5alpha-androstan-17-one, 3alpha-hydroxy-5beta-androstan-17-one, androst-5-ene-3beta,17beta-diol, 3beta,17alpha-dihydroxy-pregn-5-en-20-one (17alpha-hydroxy-pregnenolone), 3beta-hydroxy-androst-5-en-17-one (dehydroepiandrosterone, DHEA), testosterone, androst-4-ene-3,17-dione (androstenedione), 3alpha-hydroxy-5alpha-pregnan-20-one (allopregnanolone), 3beta-hydroxy-pregn-5-en-20-one (pregnenolone), 7alpha-hydroxy-DHEA, and 7beta-hydroxy-DHEA were measured using the GC-MS system in young men before and after ejaculation provoked by masturbation. The circulating level of 17alpha-hydroxypregnenolone increased significantly, whereas the other circulating steroids were not changed at all. This fact speaks against the hypothesis that a drop in the level of neuroactive steroids, e.g. allopregnanolone may trigger the orgasm-related increase of oxytocin, reported by other authors.     

Dehydroepiandrosterone and its 7-hydroxylated metabolites do not interfere with the transactivation and cellular trafficking of the glucocorticoid receptor

The human brain is a target tissue for glucocorticoids (GC). Dehydroepiandrosterone (DHEA) is a neurosteroid produced in the brain where it is transformed into 7alpha-hydroxy-DHEA and 7beta-hydroxy-DHEA. The antiglucocorticoid effects of both 7-hydroxylated metabolites have been investigated with evidence in mice that neither form of DHEA interfered with the binding of GC to its glucocorticoid receptor (GR), but contributed to a decreased nuclear uptake of the activated GR. Our objective was to use COS-7 cell culture to research DHEA, 7alpha-hydroxy-DHEA and 7beta-hydroxy-DHEA interferences with GR trafficking. These cells did not carry out the 7alpha-hydroxylation of DHEA and the oxidation of cortisol into cortisone. The cDNA of the human GR was inserted into pcDNA3 for a transient transfection of COS-7 cells. Human GR transactivation activity was measured from a luciferase-MMTV reporter gene. The transfected COS-7 cells were cultured using 10(-12) to 10(-5) M dexamethasone (DEX) or cortisol, which triggered the reporter expression. Treatment with 10(-12) to 10(-5) M DHEA, 7alpha-hydroxy-DHEA and 7beta-hydroxy-DHEA caused no change in the GC-induced GR transactivation. A reconstruction of the process associated EGFP to the human GR cDNA. Confocal microscopic examination of COS-7 cells transiently expressing the fusion protein EGFP-GR showed nuclear fluorescence 60 min after incubation with 10(-8) M DEX or cortisol. The addition of 10(-5) M DHEA, 7alpha-hydroxy-DHEA or 7beta-hydroxy-DHEA did not change its kinesis and intensity. These results contribute to the knowledge of DHEA, 7alpha-hydroxy-DHEA and 7beta-hydroxy-DHEA, in relation to antiglucocorticoid activity. We conclude that direct interference with GR trafficking can be discounted in the case of these hormones, therefore proposing new possibilities of investigation.     

Effects of meiosis-inhibiting agents and equine chorionic gonadotropin on nuclear maturation of canine oocytes

Experiments were conducted to determine the effects of meiosis-inhibiting-agents and gonadotropins on nuclear maturation of canine oocytes. The culture medium was TCM199 + 10 ng/ml epidermal growth factor supplemented with 25 microM beta-mercaptoethanol, 0.25 mM pyruvate, and 1.0 mM L-glutamine (Basal TCM). Initially, oocytes were cultured in Basal TCM alone or in Basal TCM + dibutylryl cyclic adenosine monophosphate (0.5, 1, 5, or 10 mM dbcAMP) for 24 hr. Dibutylryl cAMP inhibited resumption of meiosis in a dose-dependent manner; 60% of oocytes remained at the germinal vesicle (GV) stage after being cultured for 24 hr in 5 mM dbcAMP. The meiosis-inhibitory effect of dbcAMP appeared to be reversible, as the oocytes resumed meiosis and completed nuclear maturation after being cultured for an additional 48 hr in its absence. Oocytes were then cultured in Basal TCM alone or in Basal TCM + roscovitine (12.5, 25, or 50 microM) for 24 hr. Although approximately 60% of oocytes cultured in 25 microM roscovitine remained at the GV stage, this percentage was not significantly different from the 48% that also remained at the GV stage when cultured in its absence. Oocytes were cultured in Basal TCM + 25 microM roscovitine for 17 hr, exposed briefly to equine chorionic gonadotropin (eCG), and then cultured in Basal TCM for 48 hr. Short exposure of oocytes to eCG was beneficial, as it significantly increased the proportion of oocytes developing beyond germinal vesicle breakdown (P < 0.05) with approximately 20-30% of these were metaphase I (MI) oocytes. Study of the kinetics of nuclear maturation demonstrated that large numbers of oocytes remained at MI even after being cultured for 52 hr following brief exposure to eCG. This study showed that in vitro maturation of canine oocytes can be somewhat improved by short exposure of oocytes to eCG. However, further studies are still required to derive effective methods to mature canine oocytes in vitro.     

Distribution of 3 beta-hydroxysteroid dehydrogenase during development of the rat adrenal cortex and capsule

Fibroblasts of the adult adrenal cortex are considered to be nonsteroidogenic connective-tissue cells. However, it has been reported that in response to regenerative stimuli, adrenocorticotropic hormone (ACTH), and transformation to malignancy, these cells acquire characteristics of parenchymal cells, which includes delta 5, 3 beta-hydroxysteroid-dehydrogenase (delta 5, 3 beta-HSD) activity. To determine whether such delta 5, 3 beta-HSD activity in adult adrenocortical fibroblasts was due to the activation or augmentation of gene expression normally occurring during embryogenesis, a histochemical study of adrenocortical development, with particular attention to the connective-tissue capsule, was undertaken. Cryostat sections of rat embryos, from 14-days postconception (PC) to birth, and of adrenal glands 1-8, 44 and 90 days after birth were tested histochemically for delta 5, 3 beta-HSD. The same or adjacent sections were stained for PAS-positive material and reticulin, and with hematoxylin and eosin. delta 5, 3 beta-HSD activity overlapped with fibroblast-like cells and with extracellular connective-tissue components in the periphery of the glands from day-17 PC onward. delta 5, 3 beta-HSD activity over the capsule diminished shortly after birth and was absent in the adult. Appropriate controls showed that the staining within the capsule was specific and not an artifact. 3 beta-HSD activity in the capsule was more intensive when dehydroepiandrosterone (DHEA) was replaced by etiocholan-3 beta-ol-17-one (ETIO) as the steroid substrate. Furthermore, the spatial distribution of 3 beta-HSD activity in the cortex differed depending on the substrate used, and the distribution patterns changed with developmental age. (ABSTRACT TRUNCATED AT 250 WORDS)     

Biotransformation XLVII: transformations of 5-ene steroids in Fusarium culmorum culture

The course of the transformation of six 5-ene steroids with varying substituents at C-17 or/and C-3: dehydroepiandrosterone (DHEA), 5-androsten-3beta,17beta-diol, 17alpha-methyl-5-androsten-3beta,17beta-diol, 5-androsten-17-one, 5-androsten-3beta-ol and pregnenolone by Fusarium culmorum was investigated. Three substrates with oxygen functions at C-3 and C-17 i.e. DHEA, 5-androsten-3beta,17beta-diol and 17alpha-methyl-5-androsten-3beta,17beta-diol were hydroxylated entirely at 7alpha-axial, allylic position. The mixture of 7alpha-hydroxy- and 7alpha,15alpha-dihydroxyderivatives was formed during the transformation of pregnenolone and 5-androsten-17-one, from the latter 2alpha,7alpha-dihydroxyderivative was also obtained. 7alpha,15alpha- Dihydroxyderivative was the only product isolated from the 5-androsten-3beta-ol post-transformation mixture. The time-course of the DHEA transformation by F. culmorum shows that the substrate induces 7alpha-hydroxylase activity. DHEA was transformed by androstenedione induced F. culmorum cultures to a larger extent than by a noninduced microorganism; the selectivity of the transformation remained unchanged.     

Synthesis of 3β-hydroxy-androsta-5,7-dien-17-one from 3β-hydroxyandrost-5-en-17-one via microbial 7α-hydroxylation

The synthesis of 3β-hydroxy-androsta-5,7-dien-17-one from 3β-hydroxy-androst-5-en-17-one (dehydroepiandrosterone, DHEA) via microbial 7α-hydroxylation has been accomplished. At the first stage, 3β,7α-dihydroxy-androst-5-en-17-one was obtained in high yield (71.2%) using a strain of Gibberella zeae VKM F-2600, which was first applied for DHEA conversion. The further route included the substitution of 7α-hydroxyl group with chlorine followed by a dehydrochlorination stage, and required minimal purifications of the intermediate products. The steroids obtained at every step were characterized by TLC_,¹H NMR, MS, UV- and IR-spectrometry.The combination of microbial and chemical steps ensured 54.6% yield of the target 3β-hydroxy-androsta-5,7-dien-17-one from DHEA and can be applied for obtaining novel vitamin D derivatives.     

Different mechanisms of protection against apoptosis by valproate and Li+.

Acute treatment with valproate and Li+ was found to protect cultured cerebellar granule cells against apoptosis induced by low K+ (5 mM). Because the protection was unaffected by MK801 (N-methyl-D-aspartate receptor inhibitor), an increase in glutamate release cannot be responsible for the observed neuroprotection. Insulin also protects against low-K+-induced apoptosis of cerebellar granule cells. This protection is totally dependent on LY294002 (a phosphatidylinositol 3-kinase inhibitor). These results suggest a role for phosphatidylinositol 3-kinase in the neuroprotection induced by insulin. Likewise, and in contrast with the results observed with Li+, the protection induced by valproate is also dependent on insulin and LY294002. Moreover, valproate (a branched-chain fatty acid) does not change the plasma membrane microviscosity under physiological conditions. These results suggest that valproate protects against low-K+-induced apoptosis by acting in the phosphatidylinositol 3-kinase/protein kinase B pathway. The protection by Li+ is independent of this transduction pathway.     

Glutamate Inhibits Adenylate Cyclase Activity in Dispersed Rat Hippocampal Cells Directly via an N-Methyl-d-Aspartate-Like Metabotropic Receptor

Three major subtypes of glutamate receptors that are coupled to cation channels--N-methyl-D-aspartate (NMDA), kainate, and alpha-amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA) receptors--are known as ionotropic receptors in the mammalian CNS. Recently, an additional subtype that is coupled to GTP binding proteins and stimulates (or inhibits) metabolism of phosphoinositides has been proposed as a metabotropic receptor. Incubation of dispersed hippocampal cells from adult rats with glutamate or NMDA decreased forskolin-stimulated cyclic AMP (cAMP) accumulation; half-maximal effects were obtained with 5.6 +/- 2.2 and 6.4 +/- 2.3 microM, respectively. Kainate and quisqualate were less potent. The effect of glutamate was antagonized by 2,3-diaminopropionate and 2-amino-5-phosphonovalerate, NMDA/glutamate receptor antagonists, but not by 0.5 microM Joro spider toxin, a specific blocker of the AMPA receptor. The inhibitory effect of glutamate on cAMP formation was not blocked by 2 microM tetrodotoxin or by the absence of Ca2+. In hippocampal membranes, glutamate, similar to carbachol, inhibited adenylate cyclase activity in a GTP-dependent manner. These findings suggest that the glutamate inhibition of adenylate cyclase is direct and is not due to a result of the release of other neurotransmitters. The effect of glutamate on cAMP accumulation was observed in an assay medium containing 0.7 mM MgCl2, which is known to inhibit both ionotropic NMDA receptor/channels in the hippocampus and metabotropic NMDA receptors in the cerebellum. The inhibitory effect of glutamate was abolished by pertussis toxin treatment.(ABSTRACT TRUNCATED AT 250 WORDS)     

Effect of 5-Hydroxytryptamine on Steroidogenesis and Oocyte Maturation in Pre-ovulatory Follicles of the Medaka Oryzias latipes

The effect of 5-hydroxytryptamine (5-HT) on steroidogenesis and oocyte maturation in pre-ovulatory follicles of the medaka Oryzias lalipes was examined using in vitro culture system. The earliest breakdown of the germinal vesicle of intrafollicular oocytes occurred about 17 hr after the beginning of incubation in the presence of 5-HT at concentration of 10 ng/ml or more. 5-HT induced oocyte maturation in a dose-dependent manner. Cyanoketone inhibited this stimulation. The concentration of 5-HT required to induce oocyte maturation corresponded to that required to enhance the production (secretion) of estradiol-17β and 17α,20β-dihydroxy-4-pregnen-3-one by pre-ovulatory follicle cells. At a concentration of 1 μg/ml, the follicle had to be exposed to 5-HT for at least 4 hr for oocyte maturation accompanied by ovulation to occur. These results indicate that 5-HT induces in vitro maturation of medaka oocytes by stimulating 17α,20β-dihydroxy-4-pregnen-3-one production by pre-ovulatory follicular cells.     

gamma-Hydroxybutyrate modulation of glutamate levels in the hippocampus: an in vivo and in vitro study

The effect of gamma-hydroxybutyric acid on extracellular glutamate levels in the hippocampus was studied by microdialysis in freely moving rats and in isolated hippocampal synaptosomes. Intra-hippocampal (CA1) perfusion with gamma-hydroxybutyric acid (10 nM-1 mM) concentration-dependently influenced glutamate levels: gamma-hydroxybutyric acid (100 and 500 nM) increased glutamate levels; 100 and 300 microM concentrations were ineffective; whereas the highest 1 mM concentration reduced local glutamate levels. The stimulant effect of gamma-hydroxybutyric acid (100 nM) was suppressed by the locally co-perfused gamma-hydroxybutyric acid receptor antagonist NCS-382 (10 microM) but not by the GABA(B) receptor antagonist CGP-35348 (500 microM). Furthermore, the gamma-hydroxybutyric acid (1 mM)-induced reduction in CA1 glutamate levels was counteracted by NCS-382 (10 microM), and it was also reversed into an increase by CGP-35348. Given alone, neither NCS-382 nor CGP-35348 modified glutamate levels. In hippocampal synaptosomes, gamma-hydroxybutyric acid (50 and 100 nM) enhanced both the spontaneous and K(+)-evoked glutamate efflux, respectively, both effects being counteracted by NCS-382 (100 nM), but not by CGP-35348 (100 microM). These findings indicate that gamma-hydroxybutyric acid exerts a concentration-dependent regulation of hippocampal glutamate transmission via two opposing mechanisms, whereby a direct gamma-hydroxybutyric acid receptor mediated facilitation is observed at nanomolar gamma-hydroxybutyric acid concentrations, and an indirect GABA(B) receptor mediated inhibition predominates at millimolar concentrations.     

First synthesis of 7alpha- and 7beta-amino-DHEA, dehydroepiandrosterone (DHEA) analogues and preliminary evaluation of their cytotoxicity on Leydig cells and TM4 Sertoli cells

Efficient syntheses of new DHEA analogues, and their apoptotic and necrotic effects on Leydig cells and TM4 Sertoli cells are described. The key step in the synthetic strategy of 7-amino-DHEA derivatives involves a bromination on C-7 position to give an epimeric mixture of bromides which were substituted by azides and reduced to give 7alpha- and 7beta-amino-3beta-hydroxyandrost-5-en-17-ones. No cytotoxic effect induced by apoptosis mechanism was observed on Leydig and TM4 Sertoli cells by treatment with these amino-DHEA analogues. A necrotic effect was induced only in TM4 Sertoli cells. The best activity was obtained with 7alpha,beta-amino-androst-5-en-3beta-ol and 7beta-amino-3beta-hydroxy-androst-5-en-17-one.