TfM mutation and masculinization versus feminization of the mouse central nervous system

The sexual behavior of mice of various genotypes has been studied in our stock in which X-linked genes, Tfm and (O^hv), as well as an autosomal dominant gene, Sxr, are maintained. Since the absence of neonatal imprinting in Tfm (O^hv)/Y^ leads to no sexual behavior, we conclude that neonatal imprinting is the prerequisite for feminization as well as masculinization of the central nervous system. Since individual components of sexual behavior may become feminine or absent instead of being masculine in sex-reversed Tfm (O^hv)/+(O⁺, Sxr/+^♂ with mosaic brains, we conclude that neonatal imprinting directly involves individual neurons, and that different degrees of imprinting by the same agent lead to masculinization or feminization. In accordance with recent views, we believe estradiol to be this imprinting agent.We envisage the role of the Tfm locus in the central nervous system as follows: within individual neurons for sexual behavior, the synthesis of aromatizing enzymes is normally inducible by androgens; these enzymes are therefore noninducible in Tfm (O^hv)/Y^. In normal neonates, exposure to testosterone leads to the induced intracellular conversion of testosterone to estradiol, self imprinting by estradiol causing masculinization. Feminization may normally be due to the direct exposure of these neurons to a low circulating concentration of estradiol. An alternative explanation might be that the estradiol-receptor protein in these neurons also is normally inducible by testosterone. In this case, neonatally testosterone-exposed neurons would become inherently more responsive to estradiol than neonatally estradiol-exposed neurons.     

Metabolic cooperation between Tfm and wild-type cells in mosaic mice after induction of DNA synthesis

Mosaic mice composed of androgen-insensitive Tfm and androgen-sensitive wild-type cells are constructed by virtue of the natural X inactivation: XX mice heterozygous for X-linked testicular feminization (Tfm) are reverted to males by the sex reversal (Sxr) mutation. After stimulation with testosterone, in the epididymis of the mosaic mice, the incorporation of ³H-thymidine is compared in both cell fractions. The labeling index of Tfm and wild-type cells is in the same order of magnitude. The result indicates that the stimulus for DNA synthesis exerted by testosterone is conveyed from the androgen-sensitive wild-type to the androgen-insensitive Tfm cells by metabolic cooperation on tissue level.On the other hand, the proportion of Tfm cells in the epididymal mosaic is far less than expected from X inactivation. The reason is that in the mosaic, in spite of normal proliferation, the undifferentiated Tfm cells die off steadily. Typical dense bodies are observed as signs of physiological cell death.     

The analysis of Lyon's hypothesis through preferential X-activation

Lyon's hypothesis of random X-inactivation or activation can be represented as a problem of compound binomials: (p + q)ⁿ, where p = q = 0.5 and n is the number of early cells representing a future embryo, expresses initial clonal compositions of XX-embryos immediately after the random but irreversible decision on the fate of individual X-chromosomes. Clonal compositions of a given somatic cell type in the adult body, on the other hand, can be expressed as (p′ + q′)^n′ where p′ and q′ are now variables determined by the initial clonal composition, while n′ is the number of embryonic cells which serve as the immediate progenitors of that somatic cell type. Since n is considerably larger than 10 even in the mouse, a paradox is created because so long as the initial decision process remains random, experimental tests of the hypothesis also remain impractical.We have determined the frequency with which the entire somatic cell type becomes Tfm + (0^hv) Blo monoclonal on a number of divergent organs of Tfm + (0^hv) Blo/ + + (0⁺) + heterozygotes. The results are compatible with parameters of (0.8 + 0.2)⁶⁰, determining the initial clonal compositions of these heterozygotes. Alleles, such as (0^hv) and (0⁺) of Cattanach's “controlling element” site on the X-chromosome, apparently determine the values which p and q take in the initial decision process.     

Biosynthesis and metabolism of testosterone by sertoli cell-enriched seminiferous tubules

Sertoli cell-enriched tubules isolated from rats which had been treated with 1,4-dimethyl sulfonyloxybutane were incubated with either [¹⁴C] progesterone or [¹⁴C] testosterone for 2 hours. Tubules of normal rats and fragments of Sertoli cell-enriched testes were incubated under the same conditions. Sertoli cell-enriched tubules converted progesterone to 20α-dihydroprogesterone, 17α-hydroxyprogesterone, androstenedione and testosterone. The major metabolite was 20α-dihydroprogesterone. The percentage conversion of progesterone into testosterone corresponded to a production of 10–20 ng testosterone. Sertoli cell-enriched tubules converted testosterone to dihydrotestosterone, androstenedione, 3α-androstanediol and 3β-androstanediol. Under our experimental conditions, dihydrotestosterone was the major 5α-reduced metabolite. Normal tubules converted progesterone and testosterone to the same metabolites as Sertoli cell-enriched tubules. Fragments of Sertoli cell-enriched testes were much more active than isolated tubules in metabolizing progesterone. They produced the same amounts of 5α-reduced metabolites of testosterone.     

Cytosol androgen receptor from kidney of normal and testicular feminized (Tfm) mice

Steroid binding has been studied in cytoplasmic extracts of normal mouse kidney, an androgen sensitive organ, and of kidney from mice affected with testicular feminization (Tfm mutant) that have inherited androgen resistance. Macromolecules that bind ³H-5α-dihydrotestosterone (DHT, the presumed active androgen in most testosterone target organs) and sediment in glycerol gradient at 8–9S can be observed in cytosol from kidney of mice of different sex, age, and hormonal history. The 8–9S component from normal females is heat labile, pronase sensitive, and dissociated by high salt to a lower molecular weight entity. The apparent equilibrium dissociation constant (K_d) for the DHT-receptor complex is 1.4 × 10^?9M, and there are about 1500 binding sites per testosterone-sensitive kidney proximal tubule cell. Cytosol from Tfm/Y ^ animals also shows a sharp peak of ³H-DHT-binding activity at 8–9S. The Tfm protein, however, has reduced affinity for DHT and binds only 10–25% as much ³H-DHT as wild-type receptor at ³H-DHT concentrations from 5 × 10^?11M to 1.2 × 10^?8M. Scatchard analysis, and studies involving competition with unlabeled steroids, relative binding of various androgens, and dissociation of the ³H-DHT-binding protein complex after extensive dialysis have led to the conclusion that Tfm kidney contains very little, if any, androgen receptor with properties similar to that found in normal kidney.     

Effects of ethanol and acetaldehyde on the biosynthesis of testosterone in the rodent testes

The effects of ethanol and acetaldehyde on testicular steroidogenesis were examined in enzymatically dispersed cells of the rodent testes. Both drugs significantly inhibited gonadotropin-stimulated steroidogenesis, but acetaldehyde was considerably more potent (>1000 times) than ethanol. To determine the step in testosterone's biosynthetic pathway which was inhibited by the two drugs, cells were incubated in the presence of [³H]pregnenolone and [³H]progesterone, and the amount of label incorporated into testosterone and its precursors was determined. Ethanol and acetaldehyde inhibited only the conversion of androstenedione to testosterone; none of the other precursors of testosterone was affected.     

The in vivo metabolism of androgens by muscle and adipose tissue of normal men

Androstenedione and testosterone labeled with ³H and ¹⁴C were infused simultaneously at constant rates into the brachial arm vein of 10 normal men. During the infusions blood samples were obtained from the brachial artery, a deep vein draining primarily muscle and a superficial vein draining primarily adipose tissue of the arm contralateral to the infusion. In the 10 men the mean ± SE value for the fractional metabolism of androstenedione by muscle is 0.20 ± 0.30 which is not different from the mean value for the fractional metabolism of androstenedione by adipose tissue, 0.29 ± 0.04. The mean value for the metabolism of testosterone by muscle, 0.04 ± 0.01, is significantly less than the metabolism by adipose tissue, 0.11 ± 0.01. Intercpnversion between androstenedione and testosterone occurs in both tissues. The mean value for ρ^A,T_A,M is 0.024 ± 0.005 and for ρ^A,T_A,AT is 0.024 ± 0.005. The mean value for ρ^T,A_A,M is 0.005 ± 0.003 and for ρ^T,A_A,AT is 0.008 ± 0.003. The fractional metabolism of these androgens by these tissues is similar to the fractional metabolism of estrone and estradiol by these same tissues. Muscle appears to contribute about 5–12% of the overall metabolism of androstenedione and testosterone and 10–15% to the overall conversion of androstenedione to testosterone. Adipose tissue contributes about 2–7% of the overall metabolism of these androgens and 5–10% of the overall conversion of androstenedione to testosterone, but < 2% to the overall conversion of testosterone to androstenedione. In normal men, muscle appears to be more important to the metabolism of androstenedione and testosterone than is adipose tissue.     

A change in the steroid metabolic pathway in human testes showing deteriorated spermatogenesis

During androgen biosynthesis, the human testes normally produce only small quantities of Δ⁴-C₂₁ steroids as these are products of the Δ⁴-pathway and healthy human testes preferentially use the Δ⁵-pathway. However, the Δ⁴-C₂₁ steroid progesterone accumulates in the thickened lamina propria of the seminiferous tubules in testes with deteriorated spermatogenesis. The objectives of this study were to analyse the pregnenolone metabolites in testes with deteriorated spermatogenesis and to establish whether the androgen biosynthesis pathway changes in this condition. Biopsied or orchiectomised testicular samples were obtained from patients with varicocele, non-obstructive azoospermia, obstructive azoospermia, testicular cancer, and cryptorchidism. The samples were segregated into spermatogenesis related Johnsen’s score groups: Low-JS (< 5.0) and High-JS (> 7.8). Higher levels of progesterone and 17α-hydroxyprogesterone were metabolised under in vitro conversion in the Low-JS testes than the High-JS testes when cell-free homogenates from each group were separately incubated with ¹⁴C-labelled pregnenolone. Nevertheless, the serum hormone levels did not differ between groups. Two novel pregnenolone metabolites 5β-pregnan-3β-ol-20-one and 5α-pregnan-3α, 21diol-20-one were identified from in vitro conversion in Low-JS testes and by recrystallisation. Immunohistochemistry revealed the higher βHSD expression in the Low-JS than the High-JS testes. However, the CYP17A1 expression levels did not differ between groups. Infertile testes increase the relative βHSD levels in their Leydig cells and synthesised testosterone from pregnenolone via the Δ⁴- rather than the Δ⁵-pathway. A new insight into a change of metabolites in Low-JS testes will be relevant to understand the mechanism of the deteriorated spermatogenesis under the normal range of testosterone level.     

Testicular in vitro conversion of progesterone to testosterone and androstenedione in 17α-hydroxylase deficiency

Incubations of [³H]-progesterone with testicular tissue obtained from a new case of male with 17α-hydroxylase deficiency were performed. The per cent conversion to androstenedione and testosterone was virtually absent when compared to that obtained from an identical incubation performed using testicular tissue from a normal male with cryptochordism. The findings provide an in vitro evidence in support of the existence of 17α-hydroxylase testicular defect in this disorder.     

Effect of hypophysectomy and gonadotropin treatment on metabolism of 3H-progesterone by rat testicular tissue

This study was conducted to define the pattern of $\text{in}$$\text{vitro}$ metabolism of ³H-progesterone in incubates of rat testicular tissue at various time intervals after hypophysectomy and to determine the effect of $\text{in}$$\text{vivo}$ gonadotropin treatment on the metabolism of ³H-progesterone in posthypophysectomy regressed testes. Formation of tritium labeled testosterone, androstenedione, 5α-androstanediol and androsterone was markedly diminished within two weeks and only traces of these substances were formed between the 23rd and 54th day after hypophysectomy. The major metabolite throughout this time period was ³H-20α-dihydroprogesterone. These data demonstrate that in posthypophysectomy-regressed testes ³H-progesterone metabolism does not revert to that observed in fetal testes or testes from immature animals. Treatment with HCG, commencing on the 33rd day after hypophysectomy resulted first in formation of 5α-reduced androgens and marked decrease in 20α-dihydroprogesterone. Additional treatment produced increased formation of radiolabeled testosterone and androstenedione and diminution of 5α-reduced androgens. This metabolic pattern is reminiscent of that observed in normally developing testes. Treatment with PMS commencing on the 33rd day after hypophysectomy resulted in formation of large amounts of androstenedione and testosterone and decrease of 20α-dihydroprogesterone to trace amounts within 10 days of initiation of treatment. After additional 10 days of treatment the formation of androstenedione diminished, testosterone remained unchanged. The possibility is suggested that FSH activity in PMS may be responsible for the different pattern of progesterone metabolism. The data of an three experiments suggest that the 20α-hydroxysteroid oxidoreductase activity may be influenced by gonadotropins.     

Effects of Alleles at the W Locus on Testicular Luteinizing Hormone Receptors in Adult Mice

Abstract

The autoregulation of testicular LH receptors was studied in W^x/W^v mice with germ cell aplasia and in normal (±/±) mice. To assess the effects of each individual allele, W^x/± and W^v/± mice were also examined. Basal testicular LH receptor concentration was higher in W^x/W^v mice than in all other genotypes, and higher in W^x/± than in ±/± mice. Twenty-four h after injection of 0.3 IU hCG/g bw, LH receptor concentration was decreased in ±/± and W^v/± mice, but not in W^x/W^v or W^x/± animals. Administration of hCG caused a significant increase in plasma testosterone levels in all genotypes. However, injection of the highest dose of hCG used (0.9 IU/g bw) caused a significantly greater elevation in plasma testosterone in W^x/W^v than in ±/± mice. Plasma gonadotropin levels were significantly higher in W^x/W^v mice than in all other genotypes. The present results indicate that the W^x allele is responsible for the changes in testicular function observed in W^x/W^v mice, and suggest that this allele may be involved in the genetic regulation of testicular LH receptors in the mouse.     

Genetically directed preferential X-activation seen in mice

The nature of the O^hv mutation of the X chromosome of the mouse is defined. This locus exerts a cis position effect upon the expression of genes Tfm and Blo mapping in the same region; genes on the same chromosome as the O^hv mutation are preferentially activated and genes on the other X chromosome are usually inactive. Following the proportion of Tfm cells in the kidneys of heterozygotes confirms that the variegation seen of the locus Blo in the coat is matched in inner tissues. By introducing the sex reversal gene Sxr into these stocks, a situation can be created in which wildtype kidney cells have a selective advantage over Tfm cells in the embryonic Wolfflan duct and urogenital sinus. In spite of this difference in cell advantages, Blo coat variegation and Tfm Wolffian duct cell preponderance continue to exhibit a good correlation. This excludes the possibility that the variegation depends upon a selective advantage of cells carrying Tfm alleles after random X-inactivation and therefore reinforces the conclusion that the O^hv mutation is directly concerned in the X-activation process. A model is presented in which this locus acts as a receptor site recognized by molecules which activate one X chromosome.     

Testicular dysfunction and sexual impotence in the alcoholic rat

Evidence suggests that male alcoholics show a decrease in plasma testosterone levels which may lead to sexual impotence. Testosterone is mainly synthesized in the testes, and the object of this study is to investigate how chronic alcohol ingestion by male rats affects testosterone synthesis in the testis and sexual behavior. Male rats were isolated at 25 days of age and randomly assigned to one of the following diet groups: (1) Alcohol Group, 20% alcohol in water and rat chow ad lib; (2) Sucrose Group, matched to Alcohol Group in rat chow intake and receiving a sucrose solution calorically equivalent to the alcohol intake; (3) Control Group, water and rat chow ad lib. Four rats from each group were sacrificed at 52 and at 154 days old. Testicular tissue was incubated with [³H]-progesterone and the percentage of conversion to testosterone, androstenedione, estradiol and 17α-hydroxyprogesterone was measured. In the older Alcohol subgroup, there was a decrease in the conversion of progesterone to testosterone and androstenedione, accompanied by an increase in the conversion to 17α-hydroxyprogesterone, suggesting an inhibitory effect of alcohol on 17–20 desmolase. Measures of sexual behavior indicated that Alcohol rats tended to show less sexual interest behaviors in the presence of females in heat than the Sucrose and Control rats.     

Inhibition of testosterone biosynthesis by ethanol: relation to the pregnenolone-to-testosterone pathway

The concentrations of metabolites in the pregnenolone → testosterone pathway were determined in freezestopped testes in control rats and during ethanol intoxication (2 h after injection of 1.5 $\text{g ethanol}\text{kg}$ body wt). Ethanol lowered the mean testicular concentrations of testosterone (by 63–74%), androstenedione (49–81%), 17-hydroxyprogesterone (60–76%), progesterone (29–67%) and pregnenolone (12–25%). 4-Methylpyrazole had no effect on the ethanol-induced changes. The present results reveal no inhibition at the 17-hydroxyprogesterone → androstenedione → testosterone steps, but do not exclude inhibition before the step yielding pregnenolone and at the pregnenolone → progesterone → 17-hydroxyprogesterone steps.     

The continuous presence of ewes in estrus in spring influences testicular volume,testicular echogenicity and testosterone concentration,but not LH pulsatility in rams

The continuous presence of active male small ruminants prevents seasonal anestrus in females, but evidence of the same mechanism operating from the females to the males is scarce. This study assessed the effects of the continuous presence of ewes in estrus in spring on ram sexual activity, testicular size and echogenicity, and LH and testosterone concentrations. On 1 March, 20 rams were assigned to two groups (n = 10 each): isolated (ISO) from other sheep, or stimulated (STI) by 12 ewes, which were separated from the rams by an openwork metal barrier, allowing contact between sexes. Each week, four ewes were induced into estrus by intravaginal sponges. Live weight, scrotal circumference, testicular width (TW) and length (TL) were recorded at the beginning and at the end of the experiment, and testicular volume (TV) was calculated; at the same time, testicular ultrasonography and color Doppler scanning were performed. Blood samples (March to May) were collected once per week for testosterone determinations, and at the end of the experiment, blood samples were collected for 6 h at 20-min intervals for LH analysis. Rams were exposed to four estrous ewes in a serving-capacity test. Scrotal circumference, TW and TL were higher in the STI than in the ISO rams (P < 0.05) in May, and TV was higher (P < 0.05) in the STI (391 ± 17 cm³) than in the ISO rams (354 ± 24 cm³). In ISO rams, the number of white pixels was higher (P < 0.01) in May (348 ± 74) than in March (94 ± 21) and differed significantly (P < 0.01) from that of the STI rams in May (160 ± 33). In ISO rams, the number of grey pixels was higher (P < 0.05) in May (107 ± 3) than it was in March (99 ± 1). Stimulated and ISO rams did not differ significantly in mean LH plasma concentrations (0.8 ± 0.5 v. 0.9 ± 0.4 ng/ml), LH pulses (2.1 ± 0.5 v. 2.2 ± 0.2) and amplitude (2.0 ± 0.4 v. 3.2 ± 0.7 ng/ml, respectively). Stimulated rams had significantly higher testosterone concentrations than ISO rams from April to the end of the experiment. Stimulated rams performed more (P < 0.05) mountings with intromission (3.0 ± 0.4) than did ISO rams (1.5 ± 0.5). In conclusion, after 3 months in the continuous presence of ewes in estrus in spring, rams had higher TV and some testicular echogenic parameters were modified than isolated rams. Although exposed rams also had higher levels of testosterone after 2 months in the presence of estrous ewes, their LH pulsatility at the end of the study was not modified.     

Stimulation of growth by testosterone via the mesenchyme

Summary To establish testosterone-dependent growth in organ-culture, anlagen of preputial glands from normal wild-type and from androgeninsensitive mouse embryos carrying the testicular feminization mutation (Tfm) were explanted and cultured in the presence of testosterone. Within six days a size difference developed between Tfm and wild-type explants involving length of hair follicle, amount of preputial gland tissue, and overall size.Anlagen from Tfm and wild-type preputial glands were then separated into epithelial bud and mesenchyme. Reciprocal recombinants were prepared and cultured with testosterone. In the recombinants development of hair follicles and gland tissue was inconsistent. Nevertheless, the effect of testosterone was expressed in the overall size of the explants. The size correlated with the type of mesenchyme used, but not with the type of epithelium: Androgen-insensitive Tfm epithelium combined with wild-type mesenchyme reached the same size as whole wild-type glands and wild-type/wild-type recombinants. Wild-type epithelium with Tfm mesenchyme resulted in small explants, which were in the range of the whole androgen-insensitive Tfm glands. Tfm/Tfm recombinants showed very poor growth, probably related to the fact that in this group no hair or gland structures developed.     

Nuclei in testicular feminization (Tfm) are not activated by intact testosterone receptor complexes: A morphometric study in striated urethral muscle of mosaic mice

Summary Sex reversed mice heterozygous for the X-linked Tfm mutation are mosaics with respect to the Tfm locus. In the androgen-dependent striated urethral muscle, nuclei coding for the intact testosterone receptor protein and nuclei coding for the defective Tfm receptor protein are incorporated in the same multinucleate muscle fibres. The intact testosterone receptor complex can thus be expected to enter the Tfm nuclei. Our measurements show that the fibre diameters of the mosaic muscle form a homogeneous population, intermediate in size between induced male and non-inducible Tfm phenotypes. By contrast, the nuclear size shows a bimodal distribution, the subpopulations corresponding to Tfm and wild type nuclei. The results indicate that the Tfm nuclei are not activated by the intact testosterone receptor complex.     

Hormonal changes during the perinatal period: Serum testosterone,some of its precursors,and FSH and prolactin in preterm and fullterm male infant cord blood and during the first week of life

To investigate fetal regulation of the endocrine testis during the third trimester of gestation, pregnenolone, progesterone, 17-hydroxyprogesterone, androstenedione, testosterone, FSH and prolactin concentrations were measured in the umbilical circulation of 31–35 preterm (27–37 weeks) and 18–19 fullterm (39–42 weeks) male infants, and postnatally until 5 days of age in 27–39 fullterm male infants. 17-hydroxyprogesterone and prolactin concentrations increased significantly (P < 0.001) between 27–37 weeks of gestation; the other hormones measured were unchanged. The levels of progesterone in preterm infants, and prenenolone, progesterone and 17-hydroxyprogesterone in the cord vein of fullterm infants were significantly (P < 0.001–0.05) higher than those in the cord artery. Androstenedione concentrations were similar in the cord artery and vein, and decreased less than pregnenolone, progesterone and 17-hydroxyprogesterone after birth, reflecting major androstenedione production in the fetus. Testosterone concentrations were higher (P < 0.01–0.05) in the cord artery than in the vein, both in preterm and fullterm infants, showing the main site of testosterone production to be the fetal compartment. Postnatally, testosterone increased clearly from concentrations of 0.25 ± 0.05 (SE) mg/ml in the cord artery and 0.10 ± 0.01 in the cord vein to 0.94 ± 0.14 ng/ml in the peripheral vein on the first postnatal day, and decreased thereafter clearly between 3–5 days. FSH did not change during the first 5 postnatal days. Concentrations of all the other hormones measured decreased significnatly after birth.It is concluded from the cord blood hormone levels of infants born between 27–42 weeks of gestation that: (1) The third trimester of gestation represents a stable phase of endocrine development with relatively small changes in circulating hormone levels; (2) Both the placenta and the fetus seem to be able to produce androstenedione in the perinatal period; and (3) The initial increase in testosterone after birth is indicative of the inhibitory effect of placental steriods on testicular endocrine function during the last trimester of gestation.     

Kinetic study of HCG induced decrease of microsomal 7α-hydroxylase activity in rat testes

Microsomes from rat testes were incubated with varying concentrations of ¹⁴C labelled testosterone and androstenedione. The production of 7α(-hydroxytestosterone and 7α-hydroxyandrostenedione was followed; K_m and V_m values were calculated from Lineweaver-Burk curves.A sustained treatment of rats with HCG resulted in a considerable decrease of the maximal 7α-hydroxylation rate (V_m) whereas the K_m value was not changed. V_m of microsomes from normal rats, when incubated with microsomes from HCG-treated animals, was also decreased substantially. It is concluded that HCG-induced depression of 7α-hydroxylation capacity of testicular microsomes is at least in part due to non-competitive inhibition of the enzyme.     

Production of testosterone from progesterone by rat testicular microsomes without release of the intermediates 17 alpha-hydroxyprogesterone and androstenedione

It has been shown that during the in vitro conversion of progesterone to androstenedione, 17 alpha-hydroxyprogesterone is not an obligatory intermediate which equilibrates with freely diffusible steroids in the incubation medium. Recently a cytochrome P-450 was purified that catalyzed, in addition to hydroxylase/lyase activities, reduction of androstenedione to testosterone. In order to determine whether progesterone could be transformed to testosterone without both intermediates (17 alpha-hydroxyprogesterone and androstenedione) being equilibrated with steroids in the medium, several double-label double-substrate experiments were performed. When rat microsomes were incubated with an equimolar mixture of [14C]progesterone and 17 alpha-hydroxy[3H]progesterone, androstenedione was isolated with a 11-fold higher 14C/3H ratio than 17 alpha-hydroxyprogesterone, indicating that androstenedione could not be produced from free, diffusible 17 alpha-hydroxyprogesterone. Incubation of an equimolar mixture of 17 alpha-hydroxy[3H]progesterone and [14C]androstenedione with testicular microsomes resulted in the incorporation of 3-4-fold more 17 alpha-hydroxyprogesterone into testosterone than of androstenedione, although the latter is the immediate precursor of testosterone. In an experiment in which equimolar concentrations of [3H]progesterone and [14C]androstenedione were incubated with testicular microsomes, the large pool of progesterone inhibited competitively lyase activity, but still the label of progesterone was incorporated into testosterone to the same extent as that of androstenedione. These results indicate that testosterone can be produced by immature rat testicular microsomes from added progesterone on an organized unit without the intermediates equilibrating with the incubation medium.