Dehydroepiandrosterone and dehydroepiandrosterone sulfate production in the human adrenal during development and aging.

Dehydroepiandrosterone (DHEA) is produced in prodigious quantities by the human adrenal, principally as the 3-sulfoconjugate DHEA sulfate (DS) during intrauterine life. The fetal zone and neocortex cells of the fetal adrenal express large amounts of DHEA sulfotransferase and minimal amounts, at least until very near the end of gestation, of 3beta-hydroxysteroid dehydrogenase. This pattern of enzyme expression favors substantial secretion of DHEA/DS with minimal cortisol produced; the DHEA/DS serves as the major precursor for placental estrogen formation in human pregnancy. Aside from adrenocorticotropin, other physiologic regulators of growth and steroidogenesis in the fetal adrenal have been postulated to exist, but have yet to be identified. Whereas intrauterine stressors may activate adrenal cortisol secretion, the fetal adrenal responds to many pregnancy conditions by suppressing DHEA/DS formation. After birth, the human adrenal undergoes reorganization whereby the large, inner fetal zone regresses, and DHEA/DS production is diminished. Just prior to gonadal maturation, the human adrenal undergoes morphologic and functional changes (adrenarche) that give rise to a prominent zona reticularis that is characterized by the presence of DHEA sulfotransferase, the absence of 3beta-hydroxysteroid dehydrogenase, and an enhancement of DHEA/DS production. The adrenal of the adult responds to stress in many instances like that of the fetus: increased cortisol secretion and diminished DHEA/DS secretion. The mechanisms for this divergence in the adrenocortical pathway is unknown. With aging, there is suppression of DHEA/DS secretion, possibly as the consequence of an involution of the zona reticularis, but corticosteroid production continues unabated.     

Estrogen-mediated regulation of CYP7B1: a possible role for controlling DHEA levels in human tissues

The current study examines regulation of CYP7B1, a DHEA 7alpha-hydroxylase, by sex hormones. Transfection with estrogen receptor alpha and treatment with 17beta-estradiol in human embryonic kidney 293 cells significantly increased CYP7B1 catalytic activity and mRNA, and stimulated a human CYP7B1 reporter gene. Transfection with estrogen receptor beta showed similar but less significant effects. In the absence of receptors, 17beta-estradiol suppressed CYP7B1 activity, suggesting that estrogenic effects may be different in cells not expressing receptors. Quantitation of CYP7B1 mRNA in adult and fetal human tissues showed markedly higher CYP7B1 mRNA levels in fetal tissues compared with the corresponding adult ones, except in the liver. This indicates a tissue-specific, developmental regulation of CYP7B1 and suggests an important function for this enzyme in fetal life. DHEA secreted by fetal adrenals is an essential precursor for placental estrogen formation. Since CYP7B1 diverts DHEA from the sex hormone biosynthetic pathway, estrogen receptor-mediated up-regulation of CYP7B1 should lead to less DHEA available for sex hormone synthesis and may help to maintain normal levels of estrogens and androgens in human tissues, especially during fetal development. Regulation by estrogens may also be of importance in other processes where CYP7B1 is involved, including cholesterol homeostasis, cellular proliferation, and CNS function.     

Precursor role of 15alpha-hydroxyestradiol and 15alpha-hydroxyandrostenedione in the formation of estetrol

Studies were designed to elucidate the origin of estetrol (15alpha-hydroxyestriol (estra-1,3,5(10)triene-3,15alpha,17beta-tetrol) or E4) during late human pregnancy. 3H-Labelled 15alpha-hydroxyestradiol (3,15alpha-dihydroxyestra-1,3,5(10)-trien-17-one or 15E2) and 14C-labelled 17beta-estradiol (estra-1,3,5(10)-triene-3,17beta-diol or E2) were infused into the fetus during transfusion in utero for erythroblastosis fetalis, and in another study the same substrates were injected intravenously into the maternal circulation. In a third study, 3H-labelled 15alpha-hydroxyandrostenedion (15alpha-hydroxyandrost-4-ene-3,17-dione or 15delta4) and 14C-labelled E2 were infused into the fetus. Maternal urine was collected for 5--6 days, and after Glusulase hydrolysis, the following metabolites were isolated: estriol (estra-1,3,5(10)-triene-3,16alpha,17beta-triol or E3) containing 14C only and 15alpha-hydroxyestrone (3,15alpha-dihydroxyestra-1,3,5(10)-trien-17-one or 15E1), 15E2, and E4, all containing both labels. From the isotope content of these metabolites, it was concluded that E4 was derived from both fetal E2 and 15delta4 and only partially via 15E2. When administered to the fetus E2 and 15delta4 contributed approximately equal amounts to urinary E4. The yield of 15alpha-hydroxylated estrogens from E2 injected into the mother was very low indicating the predominantly fetal origin of the 15alpha-hydroxylase. 15delta4 was a better precursor than E2 for urinary 15E2.     

Syntheses of estetrol monoglucuronides.

Four possible monoglucuronides of estetrol (estra-1,3,5(10)-triene-3,15 alpha, 16 alpha, 17 beta-tetraol) have been prepared from appropriately protected estetrol by the Koenigs-Knorr reaction employing cadmium carbonate as a catalyst. Condensation of methyl acetobromoglucuronate with estetrol 15,16,17-triacetate provided the 3-glucuronide acetate-methyl ester in a satisfactory yield. Introduction of the glucuronyl residue into C-17 was similarly attained by the use of estetrol 3-benzoate 15,16-acetonide. When estetrol 3,17-diacetate and acetobromosugar were stirred in anhydrous toluene in the presence of cadmium salt, the reaction occurred at C-16 and C-15 yielding two isomeric monoglucuronide derivatives in a ratio of ca. 5 to 2. Removal of the protecting groups in the four glucuronide acetate-methyl esters gave the desired estetrol glucuronides, respectively. These synthetic substrates underwent readily enzymatic hydrolysis with beef-liver beta-glucuronidase to afford estetrol.     

Effect of ACTH and prolactin on dehydroepiandrosterone, its sulfate ester and cortisol production by normal and tumorous human adrenocortical cells

The effect of ACTH and prolactin on the synthesis of dehydroepiandrosterone (DHEA) and its sulfate ester (DHEAS) was studied in cell suspensions of "normal" and tumorous (adenoma) human adrenal cortex. A stimulation of DHEA and no response of DHEAS production by ACTH in "normal" adrenocortical cell suspension was observed. However ACTH stimulated both DHEA and DHEAS synthesis in tumorous adrenocortical cells. Prolactin did not influence either the basal or the ACTH stimulated DHEA and DHEAS production of adrenocortical cells irrespective of their origin. Our results are compatible with the concept that the biosynthesis of DHEA is under ACTH control, while other factor(s) regulate(s) the sulfate pathway of DHEA secretion under normal conditions. In tumorous adrenocortical cells DHEA may be regulated--at least partly--by ACTH. Prolactin seems to have no direct effect on DHEA and DHEAS synthesis. It is postulated that the relationship between serum prolactin and DHEAS (or DHEA) levels observed by several authors might be an extraadrenal effect of prolactin on adrenal androgens.     

Reduced plasma dehydroepiandrosterone sulfate levels are significantly correlated with fatigue severity in patients with primary biliary cirrhosis

Fatigue is a common debilitating complication of primary biliary cirrhosis (PBC), the pathophysiologic mechanism of which is poorly understood. Recently, the neuroactive steroid dehydroepinadrosterone sulfate (DHEAS) was reported to be implicated in Chronic Fatigue Syndrome in the absence of liver disease. The present study was undertaken to analyse fatigue scores and their relationship with disease severity and circulating levels of DHEAS as well as its precursors DHEA and pregnenolone in PBC patients with (n=15) or without fatigue (n=10) compared to control subjects (n=11). Fatigue was assessed using the fatigue impact scale (FIS) including cognitive, physical and psychosocial subclasses. Steroids were measured by radioimmunoassay or gas chromatography/mass spectrometry. Plasma concentrations of DHEAS were significantly reduced in PBC patients with fatigue as compared to controls, while those of its precursors DHEA and pregnenolone remained within the control range. Plasma levels of DHEAS in PBC patients were significantly correlated with fatigue severity as reflected by total FIS scores including total (rp=-0.42; p=0.018), as well as the cognitive (rp=-0.37; p=0.03), physical (rp=-0.48; p=0.006) and psychosocial (rp=-0.35; p=0.04) subclasses of fatigue scores. No correlation of fatigue scores was observed with indices of liver function. These findings suggest that reduced levels of the neurosteroid DHEAS may contribute to fatigue in patients with PBC; substitutive therapy using DHEAS or its precursor DHEA could be beneficial in the management of fatigue in patients with low levels of DHEAS.     

Specific estradiol biosynthetic pathway in choriocarcinoma (JEG-3) cell line

Estradiol (E2) plays a crucial role in all reproduction processes. In the placenta, it is well recognized that E2 is synthesized from fetal dehydroepiandrosterone sulfate (DHEAS). However, there is some controversy about the biosynthetic pathway involved, some authors suggest that E2 is produced by aromatization of testosterone (T), while others suggest that E2 is produced by the conversion of estrone (E1) into E2 by type 1 17β-HSD, subsequent to the aromatization of 4-androstenedione (4-dione) into E1. In the present report, using the precursor [¹⁴C]DHEA, inhibitors of steroidogenic enzymes (chemical inhibitors and siRNA) and a choriocarcinoma (JEG-3) cell line that expresses all the enzymes necessary to transform DHEA into E2, we could determine the sequential steps and the specific steroidogenic enzymes involved in the transformation of DHEA into E2. Quantification of mRNA expression levels using real-time PCR, strongly suggests that type 1 3β-hydroxysteroid dehydrogenase (3β-HSD1), aromatase and type 1 17β-HSD (17β-HSD1) that are highly expressed in JEG-3 cells are the enzymes responsible for the transformation of DHEA into E2. Analysis of the intermediates produced in the absence and presence of 3β-HSD, aromatase and 17β-HSD1 inhibitors permits to determine the following sequential steps: DHEA is transformed into 4-dione by 3β-HSD1, then 4-dione is aromatized into E1 by aromatase and E1 is finally transformed into E2 by 17β-HSD1. Our data are clearly in favor of the pathway in which the step of aromatization precedes the step of reduction by 17β-HSD.     

Effect of prolactin on the secretion of dehydroepiandrosterone (DHEA), its sulfate (DHEA-S), and cortisol by the human fetal adrenal in vitro

The effect of prolactin on the secretions of dehydroepiandrosterone (DHEA) and its sulfate (DHEA-S) as well as that of cortisol were studied in vitro in order to investigate the possible regulatory role of prolactin on steroidogenesis of the human fetal adrenal at mid-gestational age. The addition of 0.5 microgram/ml of human prolactin to the incubation medium produced a significant (P less than 0.05) increase in DHEA, DHEA-S, and cortisol secretion. These results indicate that prolactin has a regulatory role in steroidogenesis in the human fetal adrenal at mid-gestation.     

Characterization, in the guinea pig, of a hepatic cytosol protein binding dehydroepiandrosterone sulfate and estrone sulfate

A protein which binds dehydroepiandrosterone sulfate and estrone sulfate was detected in the cytosolic fraction of female Guinea-pig liver. It is characterized by a molecular mass of 14,400 Da, its affinity for DHEA sulfate (KD = 8.8 microM) and estrone sulfate (KD = 8.5 microM), and its lack of affinity for free steroids such as dehydroepiandrosterone or estrone. It is eluted by gel filtration on Sephadex G-50 simultaneously with the inhibitor of microsomal DHEA sulfatase recently described by some of us. This protein could be implicated in the intracellular transport or in the metabolism of sulfated steroids.     

Metabolism of dehydroepiandrosterone by rat hippocampal cells in culture: possible role of aromatization and 7-hydroxylation in neuroprotection.

The rate of metabolism of the multifunctional neurosteroid, dehydroepiandrosterone (DHEA), by embryonic rat hippocampal cells maintained in culture was compared to that of 4-androstenedione (AD), the immediate precursor of estrone (E1). The experiments were carried out to assess the relative contribution of DHEA, its 7-hydroxylated metabolites and estrogen on their reported effects on memory and neuroprotection. The 3H-labeled steroids of high specific radioactivity were incubated for 1, 8, 24 and 48 h and the putative metabolites extracted from the culture medium with acetone-ethyl acetate before separation by TLC for radioassay. [3H]DHEA (2.0 ng/5x10(5) cells) yielded primarily the 7alpha- and 7beta-hydroxylated steroids in an almost equal ratio under conditions that resembled those used by others to study the protection of neurons by hippocampal astrocytes against excitatory amino acid-induced toxicity. The rate of conversion of DHEA to AD, and particularly to E1, was much lower. With [3H]AD as substrate, significant aromatization to estrogen occurred only after 24 h when most of [3H]DHEA had already been converted to its 7-hydroxylated products and the hydroxylase and aromatase systems would no longer be competing for the same coenzyme (NADPH). The hippocampal cells were still viable after 48 h of incubation with the steroids and were able to oxidize estradiol (E2) to E1 and reduce E1 to E2 and AD to testosterone (T). It is suggested that 7alpha- and 7beta-OHDHEA, the main metabolites formed in the rat hippocampus, might be responsible for some of the functions previously ascribed to estrogens in the brain and the reasons for this proposal are discussed.     

Prevention of diabetes,hepatic injury,and colon cancer with dehydroepiandrosterone

The levels of dehydroepiandrosterone (DHEA) and its sulfate (DHEA-S) peak in human in their twenties, then decrease gradually with age. The physiological importance of DHEA was not clear until recent research reports showing that DHEA has beneficial effects on preventing diabetes, malignancy, inflammation, osteoporosis, and collagen disease. We summarize our results concerning diabetes, hepatitis, and colon cancer.

In 1982, Coleman et al. [Diabetes 31 (1982) 830] reported that DHEA decreased hyperglycemia in diabetic db/db mice, which become insulin resistant. We measured hepatic gluconeogenic enzymes in an attempt to elucidate the mechanical mechanism of DHEA action. The activity and gene expression of hepatic gluconeogenic enzyme such as glucose-6-phosphatase (G6Pase) was increased in db/db mice despite hyperinsulinemia compared to control db/+m mice. DHEA, like troglitazone, decreased these levels in db/db mice. We also showed that DHEA improved the insulin resistance caused by aging or obesity using the glucose clamp technique in another animal model. In humans, the serum DHEA concentration was shown to be associated with hyperinsulinemia in diabetes. It also became clear that DHEA increased insulin secretion in old-aged db/db mice. DHEA increases not only insulin sensitivity due to the effects in the liver and muscle, but also insulin secretion.

As an effect of DHEA on T-cell mediated hepatitis induced by concanavalin A (ConA), DHEA reduced hepatic injury by inhibiting several inflammatory mediators and apoptosis. As an effect of DHEA on carcinogenesis, DHEA would be a potential chemopreventative agent against colon cancer because it decreases the number of azoxymethane (AOM) induced aberrant crypt foci, which is a possible precursor to adenoma and cancer in a murine model.

Thus, since DHEA has many beneficial effects experimentally, we should consider administration of DHEA in the future, and common mechanisms among these actions of DHEA should be elucidated in further studies.     

A radioimmunoassay of plasma unconjugated and conjugated estetrol.

A radioimmunoassay for the measurement of both unconjugated and conjugaged estetrol in plasma has been developed. The antiserum obtained after 6 months of immunization with 6-oxoestetrol-6-(O-carboxy-methyl)oxim-BSA was used at a final dilution of 1:90,000 and showed almost no cross reaction with other steroids except for estriol at 1.24%. Esterol-glucosiduronate was synthesized by incubating with adrenalectomized rat liver homogenate and uridine diphosphoglucuronic acid. Then, plasma estetrol-glucosiduronate was measured in the same manner for unconjugated estetrol after hydrolysis with beta-glucuronidase. Sephadex LH-20 column chromatography (7X110 mm, benzene:methanol, 85:15) was employed for accurate assessment. The sensitivity was 10 pg and the smallest amount measurable was 40 pg/sample. The method bland was consistently negligible. The intra and inter assay precision was 11.8% and 14.2% for unconjugated estetrol and that for estetrol-glucosiduronate was 13.5% and 17.1%.     

The primary sequence and the subunit structure of mouse alpha-2-macroglobulin, deduced from protein sequencing of the isolated subunits and from molecular cloning of the cDNA.

Mouse plasma alpha-2-macroglobulin (m alpha 2M) was isolated and the N-terminal amino-acid sequences determined after separation of the 165-kDa and 35-kDa subunits. These sequences were compared to the protein sequence predicted by the cDNA, which was cloned from a mouse liver library and sequenced. From these data it is evident that both subunits are encoded by one mRNA of approximately 5 kb expressed predominantly in liver. The smaller subunit, with the N-terminal sequence DLSSSDLT, comprises the C-terminal 257 residues of m alpha 2M and is derived from a single-chain precursor probably by proteolytic processing at an arginine residue in the sequence PTRDLSS. Analysis of the predicted protein further showed all the salient features of a proteinase inhibitor of the macroglobulin family: a bait region that deviates from all known sequences in this family, a very conserved internal thiolester site and conserved cysteine residues and putative N-glycosylation sites. The synthesis of m alpha 2M in adult liver was demonstrated by Northern blotting and in fetal liver by in-situ hybridization. Transient transfection of COS cells with the cDNA under control of a viral promoter demonstrated the secretion and partial processing of m alpha 2M in the culture medium. In plasma the level of m alpha 2M was found to be stable as expected for the murine counterpart of human plasma alpha-2-macroglobulin. The possibilities of using the mouse as a genetic model to study this proteinase inhibitor in vivo are discussed.     

Effects on DHEA levels by estrogen in rat astrocytes and CNS co-cultures via the regulation of CYP7B1-mediated metabolism

The neurosteroid dehydroepiandrosterone (DHEA) is formed locally in the CNS and has been implicated in several processes essential for CNS function, including control of neuronal survival. An important metabolic pathway for DHEA in the CNS involves the steroid hydroxylase CYP7B1. In previous studies, CYP7B1 was identified as a target for estrogen regulation in cells of kidney and liver. In the current study, we examined effects of estrogens on CYP7B1-mediated metabolism of DHEA in primary cultures of rat astrocytes and co-cultures of rat CNS cells. Astrocytes, which interact with neurons in several ways, are important for brain neurosteroidogenesis. We found that estradiol significantly suppressed CYP7B1-mediated DHEA hydroxylation in primary mixed CNS cultures from fetal and newborn rats. Also, CYP7B1-mediated DHEA hydroxylation and CYP7B1 mRNA were markedly suppressed by estrogen in primary cultures of rat astrocytes. Interestingly, diarylpropionitrile, a well-known agonist of estrogen receptor β, also suppressed CYP7B1-mediated hydroxylation of DHEA. Several previous studies have reported neuroprotective effects of estrogens. The current data indicate that one of the mechanisms whereby estrogen can exert protective effects in the CNS may involve increase of the levels of DHEA by suppression of its metabolism.     

Cell free synthesis of human prothrombin: Immunological characterization of the translation product

mRNAs prepared from adult and fetal human liver were translated in a cell free reticulocyte lysate system. The early precursor of human prothrombin, so called preprothrombin, was purified by immunoaffinity microchromatography. The precursor was identified by immunological competition as a single band displaying in Sodium dodecyl sulfate gel electrophoresis a Mr slightly lower than the Mr of human prothrombin (72 000 instead of 76 000). Immunological studies showed that human preprothrombin reacted more efficiently with antibodies raised against denatured prothrombin than with anti plasmatic prothrombin antibodies. The rate of synthesis of the precursor obtained under the direction of adult liver RNA was ten fold higher than that obtained under the direction of fetal liver RNA.     

Purification and characterization of a binding protein related to the Z class of cytosolic proteins in guinea-pig liver cytosol (guinea-pig Z protein).

The purification and characterization of a low-molecular-mass binding protein from female guinea-pig liver cytosol is reported. Its molecular mass (14.4 kDa), amino acid composition, abundance and biological properties identify it as belonging to the Z class of liver cytosolic proteins [Levi, A.J., Gatmaitan, Z. & Arias, I.M. (1969) J. Clin. Invest. 48, 2956-2167]. Among the most important members of this class of proteins are the fatty-acid-binding proteins (FABPs) and the sterol carrier protein2 (SCP2). The guinea-pig Z protein (G-ZP) has some similarities in its amino acid composition and NH2-terminal sequence with those of the rat liver FABP, but its isoelectric point is basic (pI 8.85), like that of SCP2. We also examined its binding affinities for a number of ligands bound by these two proteins. The results show that the purified G-ZP binds dehydroepiandrosterone sulfate, estrone sulfate, oleic acid and cholesterol, but shows no affinity for free steroids such as estrone and DHEA. Thus it can be said that G-ZP has some characteristics of FABPs and some of SCP2 but seems, however, to be different from both these proteins. The purified G-ZP inhibits microsomal DHEA sulfate sulfatase activity in a mixed noncompetitive way. This protein could be involved in the transport and/or metabolism of sulfated steroids.     

Dehydroepiandrosterone (DHEA) increases territorial song and the size of an associated brain region in a male songbird

In many species, male territorial aggression is tightly coupled with gonadal secretion of testosterone (T). In contrast, in song sparrows (Melospiza melodia morphna), males are highly aggressive during the breeding (spring) and nonbreeding (autumn and early winter) seasons, but not during molt (late summer). In aggressive nonbreeding song sparrows, plasma T levels are basal (< or = 0.10 ng/ml), and castration has no effect on aggression. However, aromatase inhibitors reduce nonbreeding aggression, indicating a role for estrogen in wintering males. In the nonbreeding season, the substrate for brain aromatase is unclear, because plasma T and androstenedione levels are basal. Aromatizable androgen may be derived from plasma dehydroepiandrosterone (DHEA), an androgen precursor. DHEA circulates at elevated levels in wintering males (approximately 0.8 ng/ml) and might be locally converted to T in the brain. Moreover, plasma DHEA is reduced during molt, as is aggression. Here, we experimentally increased DHEA in wild nonbreeding male song sparrows and examined territorial behaviors (e.g., singing) and discrete neural regions controlling the production of song. A physiological dose of DHEA for 15 days increased singing in response to simulated territorial intrusions. In addition, DHEA treatment increased the volume of a telencephalic brain region (the HVc) controlling song, indicating that DHEA can have large-scale neuroanatomical effects in adult animals. The DHEA treatment also caused a slight increase in plasma T. Exogenous DHEA may have been metabolized to sex steroids within the brain to exert these behavioral and neural effects, and it is also possible that peripheral metabolism contributed to these effects. These are the first results to suggest that exogenous DHEA increases male-male aggression and the size of an entire brain region in adults. The data are consistent with the hypothesis that DHEA regulates territorial behavior, especially in the nonbreeding season, when plasma T is basal.     

Dehydroepiandrosterone inhibits intracellular calcium release in beta-cells by a plasma membrane-dependent mechanism

Both dehydroepiandrosterone (DHEA) and DHEA sulfate (DHEAS) affect glucose stimulated insulin secretion, though their cellular mechanisms of action are not well characterized. We tested the hypothesis that human physiological concentrations of DHEA alter insulin secretion by an action initiated at the plasma membrane of beta-cells. DHEA alone had no effect on intracellular calcium concentration ([Ca(2+)](i)) in a rat beta-cell line (INS-1). However, it caused an immediate and dose-dependent inhibition of carbachol-induced Ca(2+) release from intracellular stores, with a 25% inhibition at zero. One nanometer DHEA. DHEA also inhibited the Ca(2+) mobilizing effect of bombesin (29% decrease), but did not inhibit the influx of extracellular Ca(2+) evoked by glyburide (100 microM) or glucose (15 mM). The steroids (androstenedione, 17-alpha-hydroxypregnenolone, and DHEAS) had no inhibitory effect on carbachol-induced intracellular Ca(2+) release. The action of DHEA depended on a signal initiated at the plasma membrane, since membrane impermeant DHEA-BSA complexes also inhibited the carbachol effect on [Ca(2+)](i) (39% decrease). The inhibition of carbachol-induced Ca(2+) release by DHEA was blocked by pertussis toxin (PTX). DHEA also inhibited the carbachol induction of phosphoinositide generation, with a maximal inhibition at 0.1 nM DHEA. Furthermore, DHEA inhibited insulin secretion induced by carbachol in INS-1 cells by 25%, and in human pancreatic islets by 53%. Taken together, this is the first report showing that human physiological concentrations of DHEA decrease agonist-induced Ca(2+) release by a rapid, non-genomic mechanism in INS-1 cells. Furthermore, these data provide evidence consistent with the existence of a specific plasma membrane DHEA receptor, mediating this signal transduction pathway by pertussis toxin-sensitive G-proteins.     

Steroid hydroxylations with Botryodiplodia malorum and Colletotrichum lini

An improved procedure for the microbial hydroxylations of dehydroepiandrosterone (DHEA, 1) and 15 beta,16 beta-methylene-dehydroepiandrosterone (2) was studied using whole cells of Botryodiplodia malorum and Colletotrichum lini. C. lini catalyzed 7 alpha- and 15 alpha-hydroxylation of 1 and 7 alpha-hydroxylation of 2, while B. malorum gave 7 beta-hydroxylation of both the substrates. The stability of the enzymatic activity was higher in the presence of co-substrates (i.e., glucose or mannitol) allowing for repeated batches of the biotransformations. The yields of 7 alpha,15 alpha-dihydroxy-1 production were improved obtaining 5.8 gl(-1) (recovered product) from 7.0 gl(-1) of substrate. The structures of the hydroxylated products were assigned by a combination of two-dimensional NMR proton-proton and proton-carbon correlation techniques.