Enhanced dehydroepiandrosterone synthesis by amnion compared to chorion: a comparative study using the reverse-isotope dilution technique

With a view to establishing whether the term human fetal membranes possess the enzymic ability to synthesize dehydroepiandrosterone (DHEA) from pregnenolone, homogenates of amnion and chorion obtained from women (n = 5, age 27-34 years) after spontaneous labor at term (37-42 weeks gestation) from uncomplicated pregnancies were incubated with [7n-3H]pregnenolone as substrate. Reverse-isotope dilution analysis gave positive identification of [3H]DHEA acetate in all incubations of viable tissues. No such metabolite was evident in control incubations with heat-denatured tissues. Virtually radiochemically pure esters under three recrystallizations were obtained with mean concentrations of between 15787 and 30137 dpm mol(-1) for amnion which was considerably higher than that of chorionic tissues at 4316-5528 dpm mol(-1). The magnitude of elevation in DHEA production by amnion was noted to be between 3.6- and 5.5-fold higher than the corresponding chorion. This study provides evidence that the fetal membranes possess 17-alpha hydroxylase and C-17, 20 lyase activities capable of synthesis of DHEA, an important androgen necessary for aromatization to estrogens in need by the developing fetus.     

Expression of C-20, 22 desmolase activity by the human fallopian tube in vitro: evidence for steroidogenesis.

With a view to establishing whether cells of the human Fallopian tubes possess the cholesterol side-chain cleavage activity, homogenates of the tubes, obtained from 6 women (39-45 years) following abdominal hysterectomy for benign conditions, were incubated with (7n-3H)-cholesterol as substrate. Controls (n=6, age 40-44 years) were homogenates heated in a boiling water bath for 10 min. Using the reverse-isotope dilution technique, (3H)-pregnenolone was isolated and characterized. No such metabolite was evident in control incubations of heat-denatured enzymes. The extent of enzymic conversion varied from 1.9 x 10(-3) to 1.3 x 10(-2)%. The results reveal for the first time the existence of an active cholesterol-specific C-20, 22 desmolase system in the viable tissues. It is suggested that there exists a potential for substantial pregnenolone synthesis in vivo. This rate-limiting steroid biosynthetic conversion provides a new dimension to the functional capacity of the Fallopian tubes in the synthesis of steroids, which may be necessary for modulating ciliary beat frequency and in maintenance of hormonal milieu essential for embryogenesis.     

Steroidogenesis in interstitial cells and microsomal fraction of immature pig testes.

Testicular interstitial cells (greater than 90% viable) obtained from 6-day-old and 3-6-week-old piglets were capable of producing dehydroepiandrosterone (DHEA, 5-10 ng/500,000 cells) and responded to hCG (60 mi.u./ml), dibutyryl-cAMP (1 mmol/l) and cholera toxin (5 ng/ml) with a 2-3-fold increase in DHEA. Aminoglutethimide (100 mumol/l) abolished the response. Testosterone was produced in comparatively minor quantities (less than 0.3 ng/500,000 cells) and was unaffected by stimulation or inhibition. When cells from both age groups were incubated with [14C]- or [3H]-pregnenolone (360 and 3.0 nmol/l), 17-hydroxypregnenolone (15%) and DHEA (5-10%) were the major metabolites on the androgen pathway and 5,16-androstadien-3 beta-ol (andien-beta, 5-10%) and 4,16-androstadien-3-one (dienone, 5-10%) on the 16-androstene pathway. Stimulation and inhibition of endogenous steroidogenesis did not alter the metabolism of exogenous pregnenolone, the same metabolites being found in the same proportions at similar times. Microsomal enzyme activities accurately reflected the metabolic profile of pregnenolone metabolism seen in intact cells, with low activities for 17 beta-HSD, 3 beta-HSD-isomerase, and 16-ene-5 alpha-reductase being observed. Since steroidogenic capacity, enzyme complement and pregnenolone metabolism were the same in testes from both age groups, the differences in Leydig cell activity observed in vivo would not appear to be consequences of changes in steroidogenic enzymes or responsiveness to gonadotrophin stimulation. The lack of effect of stimulation and inhibition of steroidogenesis on the cellular metabolism of exogenous pregnenolone suggests that the endogenous and exogenous supplies of pregnenolone are metabolized by different populations of enzymes. The relative magnitudes of these populations indicate that most of the steroidogenic enzymes in the interstitial cells are not involved in the normal response to trophic stimulation.     

Biosynthesis of [7-3H]16 alpha-hydroxy-dehydroepiandrosterone having high specific activity.

Biosynthesis of [7-3H]16alpha-hydroxy-dehydroepiandrosterone in high specific activity has been studied. [7-3H] dehydroepiandrosterone (13.9 C/mM) in trace quantity was oxidized by Streptomyces roseochromogenes (NRRLB-1233) for 5 min at 27 degrees C. The radioactive products were chromatographically separated, identified and their radiochemical purity established by isotopic dilution analysis. [7-3H]16alpha-hydroxy-dehydroepiandrosterone (2.5 x 10(7) dpm) was obtained by microbial hydroxylation of substrate (1.9 X 10(9) dpm). In some cases [7-3H])5-androstene-3beta, 16alpha, 17beta-triol in a small amount of radioactivity could be found at the prolonged reaction for 30 hr.     

Metabolism of C19-steroids by homogenates of normal rat and mouse adrenal tissue and of the Snell transplantable rat adrenocortical tumour 494

C(19)-steroid metabolism in homogenates of adrenal tissue from rats and mice has been studied. Production of these compounds from [7alpha-(3)H]cholesterol by rat adrenal tissue appeared to follow a route independent of pregnenolone. The major products of [7alpha-(3)H]-dehydroepiandrosterone metabolism by rat adrenal tissue were 5alpha-reduced steroids, principally androsterone, epiandrosterone and 5alpha-androstanedione. No differences in metabolism of [7alpha-(3)H]dehydroepiandrosterone or [4-(14)C]pregnenolone were detected between adrenal tissue from Sprague-Dawley, Wistar and Osborne-Mendel rats, but experiments with the Snell rat adrenocortical tumour 494 showed that this tissue had low 5alpha-reductase activity. In contrast, the major products of [7alpha-(3)H]dehydroepiandrosterone metabolism by mouse adrenal tissue were 5beta-reduced steroids. Differences were observed between LACA and NH strains of mice in that there was a lower metabolism of androstenedione by NH mouse adrenal and a considerable difference in the proportions of aetiocholanolone and epiaetiocholanolone produced.     

Evidence for de novo cholesterol synthesis by term human fetal amnion and chorion: a comparative study using the reverse-isotope dilution technique

The cholesterol biosynthetic activity was assessed using [2-(14)C]-acetate as substrate in the homogenates of amnion and chorion obtained from women (n = 6, age 26-39 years) after spontaneous labour at term (37-40 weeks of gestation) having uncomplicated pregnancies. Reverse-isotope dilution analysis gave positive identification of [(14)C]-cholesterol acetate in all incubations of viable tissues. This metabolite was not evident in heat-denatured homogenates which served as controls. The extent of enzymic conversion for amnion at 2.6 x 10(-3) to 0.19% was persistently higher than that of the chorion at 1.7 x 10(-3) to 9.0 x 10(-3)%. The results indicate that human term fetal membranes possess the full complement of enzymes to catalyze the transformation of acetate to cholesterol. This study provides evidence that fetal membranes possess the capacity for de novo cholesterol biosynthesis, the sterol being essential for steroidogenesis as well as in embryo viability during pregnancy.     

Inhibition of the conversion of cholesterol to pregnenolone in bovine adrenocortical preparations.

The inhibition of the conversion of [4-14C] cholesterol to [4-14C] pregnenolone by a number of steroids has been studied in bovine adrenocortical mitochondrial acetonedried preparations. At equimolar substrate and inhibitor concentrations (3.3 muM) the most potent inhibitors were cholesterol derivatives containing a nitrogen function at c-22, followed by derivatives containing oxygen functions at c-22 or c-20 or both. The presence of a hydroxyl group at c-17 or the replacement of the 3beta-hydroxyl group by fluorine reduced the inhibitory efficacy. In the presence of inhibitors that were also relatively good substrates of the cholesterol side-chain cleavage system, such as some cholesterol derivatives hydroxylated in the side-chain,the rate of [4-14C] pregnenolone formation increased with time as the inhibitor was consumed. (20S)-20,21-Dihydroxycholesterol exerted such an effect on the kinetics of [4-14C]pregnenolone formation, and yielded 21-hydroxypregnenolone which was identified by gas chromatography-mass spectrometry. The synthesis of (20R)-22-ketocholesterol, of (20R,22R)-22hydroxycholesterol, (20R,22S)-hydroxycholesterol, and of (20S)-desmosterol is described.     

In vitro formation of testosterone via the delta 5-pathway in the human umbilical cord.

Homogenates of two groups of term umbilical cord (n = 6, 37-40 weeks; n = 6, 38-40 weeks) were separately incubated with [7n-3H]pregnenolone and [1,2,6,7-3H]dehydroepiandrosterone. Using the reverse-isotope dilution technique, [3H]dehydroepiandrosterone and [3H]testosterone formed from the respective substrates were isolated and characterized. The extent of enzymic conversions were 0.015-0.28% and 0.044-2.2%. These results provide evidence for the metabolic transformation of pregnenolone to testosterone via the delta 5-3 beta-hydroxy route.     

Incorporation of [3H]Arachidonic and [14C]Eicosapentaenoic Acids into Glycerophospholipids and Their Metabolism via Lipoxygenases in Isolated Brain Cells from Rainbow Trout Oncorhaynchus mykaiss

The incorporation of [3H]arachidonate [( 3H]AA) and [14C]eicosapentaenoate [( 14C]EPA) into glycerophospholipids was studied in isolated brain cells from rainbow trout, a teleost fish whose lipids are rich in (n-3) polyunsaturated fatty acids (PUFAs). EPA was incorporated into total lipid to a greater extent than AA, but the incorporation of both PUFAs into total glycerophospholipids was almost identical. The incorporation of both AA and EPA was greatest into phosphatidylethanolamine (PE). However, when expressed per milligram of individual phosphoglycerides, both AA and EPA were preferentially incorporated into phosphatidylinositol (PI), the preference being significantly greater with AA. On the same basis, significantly more EPA than AA was incorporated into phosphatidylcholine (PC). When double-labelled cells were challenged with calcium ionophore A23187, the 3H and 14C released from the cells closely paralleled each other, peaking at 10 min after addition of ionophore. The 12-monohydroxylated derivative was the pre-dominant lipoxygenase product from both AA and EPA with a rank order of 12-hydroxyeicosatetraenoic acid (12-HETE) greater than leukotriene B4 (LTB4) greater than 5-HETE greater than 15-HETE for the AA products and 12-hydroxyeicosapentaenoic acid (12-HEPE) greater than 5-HEPE greater than LTB5 greater than 15 HEPE for EPA products. The 3H/14C (dpm/dpm) ratios in the glycerophospholipids, total released radioactivity, and the lipoxygenase products suggested that PC rather than PI was the likely source of eicosanoid precursors in trout brain cells.     

Discrimination between cholesterol and sitosterol for absorption in rats

The intestinal absorption of cholesterol and sitosterol was compared in rats. The intragastric administration of a single emulsified lipid meal containing either 50 mg of [4-14C]cholesterol or [4-14C]sitosterol resulted in the lymphatic absorption of 18.2% and 0.42% of each sterol, respectively, in 6 hr. This difference was unaltered when the mucosal sterol load was equalized by reducing the cholesterol to 1 mg in the emulsified lipid meal while maintaining the same sitosterol load or when the physical state in the lumen was equalized by infusion of a micellar solution containing both sterols into bile-diverted intestine. Lymphatic cholesterol was 90% esterified compared to 12% for sitosterol. Both sterols were associated predominantly (greater than 70%) with the chylomicron fraction. Eighty percent of the chylomicron cholesterol was recovered as ester with the core lipids, while 77% of the sitosterol was recovered as free sterol with the chylomicron coat. In mucosal homogenates at 6 hr, sitosterol recovery was one-eleventh that of cholesterol. When [3H]cholesterol (10 mg) and [14C]sitosterol (10 mg) were co-administered in an emulsified intragastric lipid meal, sitosterol associated with the brush border isolated 2 hr later was one-fifth that of cholesterol. Similar differences were seen when brush border membranes were incubated in vitro with micellar solutions containing either 50 microM [3H]cholesterol or [14C]sitosterol and the relative uptake of each sterol was unaffected by micellar phospholipid type (egg yolk phospholipids, phosphatidylcholine, or phosphatidylethanolamine).(ABSTRACT TRUNCATED AT 250 WORDS)     

Effect of activin A on dehydroepiandrosterone and testosterone secretion by primary immature porcine Leydig cells.

In the present study, we evaluated the effect of the homodimer activin A on immature porcine Leydig cell functions in primary culture. Activin A (0.5-100 ng/ml) reduced hCG-stimulated dehydroepiandrosterone (DHEA) accumulation in a dose- and time-dependent manner, with a maximal inhibitory effect (58% decrease) at 20 ng/ml (8 x 10(-10) M). Activin A was found not to control steroidogenesis, either through a modulation of the gonadotropin LH/hCG binding or low-density lipoprotein cholesterol binding and internalization. However, activin A significantly decreased pregnenolone (p less than 0.002) and DHEA (p less than 0.001) formation (evaluated in the presence of 10(-5) M of WIN 24540, an inhibitor of 3 beta-hydroxysteroid dehydrogenase/isomerase [3 beta-HSDI]activity) in Leydig cells maximally stimulated with hCG (3 ng/ml, 3 h) or incubated in the presence of 22R-hydroxycholesterol (5 micrograms/ml, 2 h). These findings indicate that activin A probably exerts a partial inhibitory effect on cholesterol side-chain cleavage cytochrome P450 (P450scc) activity. On the other hand, activin A significantly (p less than 0.001) enhanced the conversion of exogenous pregnenolone and DHEA (500 ng/ml) but not of progesterone and androstenedione (500 ng/ml) into testosterone, suggesting that activin A potentially enhances 3 beta-HSDI activity in Leydig cells. Activin A action on 3 beta-HSDI activity was found to be closely related to that of transforming growth factor-beta 1 (TGF beta 1), since both activin A (20 ng/ml) and TGF beta 1 (2 ng/ml) induced a comparable and non-additive increase in 3 beta-HSDI activity.(ABSTRACT TRUNCATED AT 250 WORDS)     

The effects of steroid hormones and gonadotropins on in vitro placental conversion of pregnenolone to progesterone

Using human term placental mitochondrial preparations, optimal conversion of [3H]pregnenolone to [3H]progesterone was obtained at 30 min incubation and with a mitochondrial protein content of 2.5-3.5 mg/ml. Estradiol, estrone, progesterone and testosterone in a dose range of 0.03-8.66 mumol inhibited the in vitro conversion of [3H]pregnenolone to [3H]progesterone by placental homogenates. All four steroids inhibited the pregnenolone to progesterone conversion in a dose-dependent manner. The ID50 (dose required to inhibit conversion of pregnenolone to progesterone by 50%) was 0.04 mumol for estradiol, 0.13 mumol for testosterone, 0.3 mumol for progesterone and 1.0 mumol for estriol. Neither gonadotropin releasing hormone (50-1000 ng) nor human chorionic gonadotropin (5-500 IU) affected the placental basal conversion rate of pregnenolone to progesterone in vitro. Our findings indicate that steroid hormones such as estradiol, estrone, testosterone and progesterone can inhibit local placental progesterone biosynthesis through inhibition of the enzyme complex 5-ene-3 beta-hydroxysteroid dehydrogenase.     

Lipogenesis and cholesterogenesis in the liver of rats after cholesterol loading

Intensity of fatty acids and separate classes of lipids synthesis was studied in vitro in the liver of white rats at loading by cholesterol in the dose of 300 mg/kg once a day during 30 days by incubation of organ homogenate with [6-(14)C] glucose, [2-(14)C] lysine, [1-(14)C] palmitic acid with following determination of radioactivity of fatty acids, phospholipids, cholesterol, acylglycerols radioactivity was investigated. The inhibition of fatty acids and separate classes of lipids synthesis in vitro in the liver of white rats at loading by cholesterol at the use of [6-(14)C] of glucose and [2-(14)C] lysine, as predecessors of fatty acids and lipids and stimulation of lipids synthesis at the use of [1-(14)C] palmitic acid as the predecessor was established. The loading of white rats by cholesterol results in its synthesis inhibition in the liver during incubation of its homogenates with [6-(14)C] glucose and does not influence the cholesterol synthesis during incubation of homogenates with [2-(14)C] lysine and [1-(14)C] palmitic acid. Thus synthesis of fatty acids and their use in the phospholipids and acylglycerols synthesis in the liver of white rats with hypercholesterolemia sharply decreases during incubation of their homogenates with [6-(14)C] glucose and [2-(14)C] lysine, and the synthesis of cholesterol, phospholipids and acylglycerols - increases during incubation with [1-(14)C] palmitic acid.     

Effects of different steroid-biosynthesis inhibitors on the testicular steroidogenesis of the toad Bufo arenarum

Testis fragments from Bufo arenarum were incubated with [7(n)-(3)H]pregnenolone (P5), [1,2-(3)H]dehydroepiandrosterone (DHEA) and [1,2,6.7-(3)H]testosterone (T), and different steroid-biosynthesis inhibitors. The inhibitors used were: cyanoketone (CNK), spironolactone (SPNL) and finasteride (FIN). CNK significantly increased the recovery of 3beta-hydroxy-5-ene steroids while SPNL reduced the metabolism of P5 and the production of C19-steroids. The metabolism of C19-substrates was only modified by CNK, which reduced the transformation of DHEA without modifying the metabolism of T. To determine the degree of inhibition exerted by the inhibitors used, the activities of the enzymes were estimated as the percentage of their contribution to the total steroid metabolism. CNK strongly inhibited the activity of hydroxysteroid dehydrogenase/isomerase if its contribution was estimated using both P5 and DHEA. If the analysis was made considering both activities associated to cytochrome P450 17chi-hydroxylase, C17-20 lyase (P450c17), it became evident that SPNL inhibited both of them. The percent contribution of 17beta-hydroxysteroid dehydrogenase (17betaHSD) activity diminished in the presence of CNK only if it was estimated considering P5 and DHEA metabolism. SPNL produced a significant inhibition of 17betaHSD when its contribution was estimated considering P5 metabolism. However, SPNL was insufficient if DHEA or T were considered. The effect of SPNL on the contribution of 17betaHSD could be due to the reduction of C19-substrates. The activity of 5chi-reductase was inhibited by CNK only if results from P5 and DHEA were considered.     

Ontogeny of mitochondrial steroidogenesis in the guinea pig gonad

To determine whether steroidogenesis in the developing guinea pig may be limited by the formation of pregnenolone, cholesterol side chain cleavage activity was ascertained at various stages of development. The conversion of [1,2-³H]cholesterol to [1,2-³H]pregnenolone was detected in mitochondria isolated from fetal guinea pig ovaries and testes as early as day 35 of gestation, while no metabolism was noted in day 30 animals. Moreover, no [l,2-³H]progesterone was formed during the 60 minute incubation. From day 35 of gestation to the day of birth, the percentage of pregnenolone formed per testis (total activity) increased, while total activity in the ovary declined. In contrast, gonadal mitochondria from adult guinea pigs converted cholesterol to both pregnenolone and progesterone and total activity in these animals was substantially higher than in their fetal counterparts. In the three females examined, the rate of pregnenolone and progesterone synthesis varied according to the stage of the estrous cycle during which these animals were sacrificed. Conversion of pregnenolone to progesterone was most rapid in the early luteal phase animal, while conversion of cholesterol to pregnenolone occurred more rapidly in the periovulatory animals than in ovarian mitochondria from the late luteal phase of the cycle. The results indicate that during prenatal and postnatal development of the gonad, cholesterol side chain cleavage activity changes and that mitochondria may acquire a Δ⁵-3β-hydroxysteroid dehydrogenase.     

Influence of experimental cryptorchidism on cholesterol side-chain cleavage enzyme and delta5-3beta-hydroxysteroid dehydrogenase activities in rat testes.

Testicular cholesterol side-chain cleavage enzyme (CSCCE) and delta5-3beta-hydroxysteroid dehydrogenase (delta5-3beta-HSD) activities were assessed 12 hours and 2, 4, 8, 16, and 32 days after surgical induction of bilateral cryptorchidism in adult rats. Within 12 hours after surgery CSCCE activity (expressed as dpm of isocaproic acid-14C formed from cholesterol-26-14C/3 hours/testis) was significantly reduced (P less than 0.01) in cryptorchid testes to approximately 55% of sham-operated control values and remained depressed at less than 50% of control activities 2, 4, 16, and 32 days after surgery. Cryptorchid testis delta5-3beta-HSD activity (measured by a pregnenolone substrate-depletion assay and expressed as mumoles of products/30 minutes/testis) did not differ from controls (P greater than 0.05) 1/2, 2, or 4 days after translocation of testes to the abdominal cavity. By day 8 of cryptorchidism, however, delta5-3beta-HSD activity was reduced to 60% of control values (P less than 0.05) and continued to decline to approximately 30% of controls during the remainder of the experimental period. These observed alterations in enzyme activities suggest an impairment in the ability of cryptorchid rat testes to synthesize androgens and further indicate that testicular CSCCE is more acutely sensitive to the cryptorchid milieu than delta5-3beta-HSD.     

Loss of tritium from coprostanone derived from [1,2(n)-3H]cholesterol or [7(n)-3H]cholesterol

After oral administration of a mixture of [1,2(n)-3H]cholesterol and [4-14C]cholesterol to a baboon, fecal coprostanone had a 46% lower 3H/14C ratio than the dose administered. Loss of 3H by enolization of the 3-ketone could account for the decrease in 3H/14C. If [7(n)-3H]cholesterol was administered instead of [1,2(n)-3H]cholesterol a 23% loss of 3H from coprostanone was found. Procedures requiring measurement of 3H-coprostanone derived from [1,2(n)-3H]- or [7(n)-3H]cholesterol could be seriously in error unless an appropriate correction for loss of 3H is made.     

Steroid metabolism in corpora lutea of the western spotted skunk (Spilogale putorius latifrons)

The present study reports steroid metabolism by corpora lutea (CL) obtained from skunks with diapausing embryos ('delay' CL) and with activated embryos (activated CL). CL from both reproductive periods were incubated with various radioactive precursors. Control incubations without any tissue or with 50 microliter of packed skunk blood cells were also conducted simultaneously. Incubation of skunk CL with [3H]-pregnenolone for 3 h resulted in 36% of the precursor accumulating as progesterone. Metabolism of [3H]dehydroepiandrosterone (DHEA) to androstenedione proceeded with approximately the same amount of product accumulating (34-46%) as was observed in the conversion of pregnenolone to progesterone. These results suggest that delta 5 isomerase, 3 beta-hydroxysteroid dehydrogenase, is the most prominent enzyme in skunk CL. Metabolism of [3H]pregnenolone to 17 alpha-hydroxypregnenolone and [3H]progesterone to 17 alpha-hydroxyprogesterone occurred at low rates (1-7%), suggesting the presence of C21 steroid 17 alpha-hydroxylase in skunk CL. Aromatase activity, as estimated by measuring accumulation of oestradiol-17 beta from [3H]testosterone, was demonstrated in activated CL. These results suggest that skunk CL appear to metabolize steroids in a manner similar to CL of other mustelids such as the ferret and American badger.     

Ontogeny of steroid metabolizing enzymes in rat oligodendrocytes

Rat oligodendroglial cells were isolated from newborn and developing brains and used immediately after, for quantification of steroid metabolizing activities. Oligodendrocytes (Ol) and their progenitor cells were incubated with [(14)C] testosterone, [(14)C] progesterone, [(14)C] pregnenolone or [(14)C] dehydroepiandrosterone (DHEA). Oligodendrocytes and their progenitor cells expressed different steroid metabolizing enzymes. The main activities were 5 alpha reduction of testosterone and progesterone and 3 beta hydroxy steroid dehydrogenase-isomerase which transformed pregnenolone into progesterone and DHEA into Delta 4 androstenedione. 5 alpha reductase activity increased in male and female rats in parallel with testosterone or progesterone. Contrary to this, 3 beta hydroxysteroid dehydrogenase-isomerase activity was found to be high in the young rat and to decrease when testosterone and progesterone plasma concentration increased.     

The biosynthesis of some androst-16-enes from C21 and C19 steroids in boar testicular and adrenal tissue

1. The formation of androst-16-enes from [4-(14)C]progesterone has been investigated with long-term incubations and short-term kinetic studies. After 4hr., 1.7 and 10.3% respectively of 3alpha- and 3beta-hydroxy-5alpha-androst-16-enes were formed in boar testis minces, but much smaller yields were obtained in boar adrenal. Both tissues formed small quantities of androsta-4,16-dien-3-one. 2. The amounts of androst-4-ene-3,17-dione and testosterone isolated were small, suggesting that androst-16-ene formation may occur preferentially in the boar testis. 3. In the absence of tissue no radioactive androst-16-enes were formed. 4. Incubation of both [4-(14)C]pregnenolone and [7alpha-(3)H]progesterone resulted in 3alpha- and 3beta-hydroxy-5alpha-androst-16-enes containing (3)H/(14)C ratios of near unity and confirmed that both C(21) steroids were precursors. A similar incubation with 17alpha-hydroxy[4-(14)C]-progesterone and [7alpha-(3)H]progesterone gave the same Delta(16)-alcohols, but they contained only (3)H, indicating that side-chain cleavage of pregnenolone and progesterone occurred before 17alpha-hydroxylation. 5. Dehydroepiandrosterone, testosterone, testosterone acetate and 16-dehydroprogesterone were not found to be precursors of Delta(16)-steroids. 6. A pathway is proposed for the biosynthesis of 3alpha- and 3beta-hydroxy-5alpha-androst-16-enes from pregnenolone and progesterone; this may involve androsta-4,16-dien-3-one as an intermediate, but excludes 17alpha-hydroxyprogesterone, testosterone and dehydroepiandrosterone.