Rat testicular interstitial fluid contains mediators of vasopermeability

An intradermal injection of testicular interstitial fluid (IF) produced a marked increase in vasopermeability in a dose-dependent manner. Likewise bovine follicular fluid caused a smaller but significant response. The effect of IF was associated with accumulation of polymorphonuclear leucocytes (PMNs) inside the dermal venules and with their adherence to the venular endothelium. A minor but significant response was noticed after injecting anterior chamber fluid, but there was no response after an injection of amniotic fluid or serum intracutaneously. Destroying the Leydig cells with ethane dimethanesulphonate did not change the vasopermeability-increasing effect of IF, but after denaturation of IF proteins the effect was diminished by about 50%. Intravenous administration of hCG did not increase the ability of IF to cause the effect. These results suggest that rat testicular interstitial fluid contains mediators of vasopermeability, probably specific for the testis and also follicular fluid. The vasopermeability effect of IF does not seem to depend on the collecting time or on Leydig cells and is at least partly mediated by PMNs which are seen in the dermal venules shortly after an injection of IF.     

Comparison of testosterone and insulin-like peptide 3 secretions in response to human chorionic gonadotropin in cultured interstitial cells from scrotal and retained testes in dogs

Levels of testosterone and insulin-like peptide 3 (INSL3) secretions in response to different doses of human chorionic gonadotropin (hCG) in cultured interstitial cells were compared between retained and scrotal testes in dogs. Retained (n=10) and scrotal (n=9) testes were obtained from small-breed dogs. The testicular tissues were dispersed in Dulbecco's Modified Eagle Medium with Ham's nutrient mixture containing 2000 PU/ml dispase II and 10% fetal bovine serum. The cells were plated with differing concentrations (0-10 IU/ml) of hCG for 18 h in multiwell-plates. Testosterone and INSL3 in the same spent medium were measured by enzyme-immunoassays (EIA). A new EIA with a reliable detection range of 0.025-5 ng/ml was developed in order to measure canine INSL3 in culture medium. Dose-dependent stimulation of testosterone by hCG was observed in the cells of both retained and scrotal testes. The incremental rate of testosterone secretion was significantly lower at 0.1, 1 and 10 IU/ml hCG in the cells of retained testes than in scrotal testes, however. INSL3 secretion was significantly stimulated at 10 IU/ml hCG relative to unstimulated controls comprising cells of scrotal testes; no such stimulation was observed in the cells of retained testes. At 10 IU/ml hCG, the incremental rate of INSL3 was significantly lower in the cells of retained testes than scrotal testes. These results suggest that LH-induced secretory testosterone and INSL3 responses are lower in the interstitial cells of retained testes than of scrotal testes. Furthermore, the high concentrations of LH may acutely stimulate INSL3 release in scrotal testes of dogs, but not in retained testes.     

Impaired gonadotrophin uptake in vivo by the cryptorchid rat testis

The specific testicular uptake in vivo of 125I-labelled hCG was compared in control adult rats and adult rats made bilaterally cryptorchid 5 weeks previously. Although a similar temporal pattern of uptake was observed in both groups, uptake of hCG by cryptorchid testes was reduced at all times after injection by up to 70%. The possible causes of this impairment were investigated. It could not be accounted for by differences in the rate of absorption or clearance of 125I-labelled hCG in the two groups. Therefore, because hCG-induced increase in the permeability of testicular capillaries is a crucial factor in determining hCG uptake by the testis, this change was compared in control and cryptorchid testes. Although hCG induced a characteristic increase in testicular capillary wall permeability in both groups, this change was temporally delayed in cryptorchid testes, and occurred after hCG values in the blood had fallen. Even when hCG had crossed the capillary wall into testicular interstitial fluid, its uptake into the testicular tissue was significantly lower in cryptorchid than in control testes. These changes probably account for the impairment of gonadotrophin uptake by the cryptorchid testis and have important implications with respect to the aetiology of Leydig cell changes in cryptorchidism.     

An in-vivo method for estimating inhibin production by adult rat testes

The concentrations of inhibin in samples of rat testicular venous and arterial blood and interstitial fluid were measured by an in-vitro bioassay using pituitary cells in culture in which the standard was an ovine testicular lymph preparation (assigned potency 1 unit/mg). Inhibin levels were undetectable (less than 2 U/ml) in both blood samples but reached a mean concentration of 120 +/- 7 U/ml in testicular interstitial fluid. After unilateral efferent duct ligation the rate of inhibin accumulation in seminiferous tubules was determined by the difference in the inhibin content of the ligated and unligated testes. Additionally, the rate of seminiferous tubule fluid production was obtained from the difference in weight between the ligated and non-ligated testes. In the 24 h after efferent duct ligation there were linear increases in inhibin (18.5 +/- 1.0 U/h) and in seminiferous tubule fluid production (26 +/- 1 microliter/h), but there were no changes in serum FSH and LH levels. Experimental induction of bilateral cryptorchidism led to a decrease in the inhibin content of the testis after 10 days. The rate of inhibin accumulation after efferent duct ligation declined more rapidly than the inhibin content, being significantly depressed in cryptorchid testes after 3 days, suggesting that this measurement is a more sensitive index of inhibin production than the determination of testicular inhibin content.     

Gonadotrophin-induced accumulation of 'interstitial fluid' in the rat testis.

'Interstitial fluid' containing high levels of testosterone (60-250 ng/ml) was recovered from the testes of rats, the amounts increasing with increase in age and testis weight. Injection of 170 i.u. hCG/kg resulted 20 h later in significant increases in interstitial fluid and its testosterone content (300-800 ng/ml). In immature rats this effect of hCG was dose-dependent and time-related and the accumulated fluid contained high levels of potassium and phosphate; levels of sodium, calcium and protein were similar to those in serum. At 20 h after injection of hCG, other testicular changes were (1) increased 'adhesiveness', (2) reduced in-vitro binding of 125I-labelled hCG, and (3) an hCG-induced increase in the testis:blood ratio of hCG in vivo.     

Stimulation of Leydig cell function by a polypeptide present in testicular interstitial fluid

Rat testicular interstitial fluid (IF) contains a polypeptide factor(s) which when added to Percoll-purified rat Leydig cells in vitro, enhances both basal and hCG-stimulated production of testosterone in a dose-dependent manner. To assess the possible physiological significance of this factor(s), its levels in testicular IF from individual rats were assessed following the suppression of intratesticular levels of testosterone by either (a) treatment with anti-LH, (b) induction of unilateral cryptorchidism, or (c) treatment with ethane-1,2-dimethane sulphonate (EDS). These treatments suppressed testosterone levels in IF by 80 to 99% and in each instance levels of the IF-factor were increased. This increase occurred as early as 5h after anti-LH treatment, but restoration of normal intratesticular levels of testosterone at 20-40 h after anti-LH treatment by testosterone injection failed to normalize levels of the IF-factor. In contrast, injection of LH into normal rats raised IF testosterone to supranormal levels and this was associated with a reduction in levels of the IF-factor, and similar effects of hCG were observed in rats pretreated with anti-LH. Therefore, whilst these studies suggest that the IF-factor(s) may be of physiological importance, its local regulation does not appear to involve a straightforward negative feedback control by testosterone.     

Studies with purified immature porcine Leydig cells in primary culture

The steroidogenic capacity of purified immature porcine Leydig cells in culture was studied over several days. The cells were obtained by fractionating crude testicular interstitial cell suspensions on a discontinuous Percoll gradient (d = 1.037, 1.042, 1.052, 1.098 g/ml), and characterized by specific binding of 125I-human chorionic gonadotropin (hCG), testosterone (T) and cyclic adenosine 3':5'-monophosphate (cAMP) production in response to hCG, and the enzymatic determination of delta 5-3 beta-hydroxysteroid dehydrogenase (3 beta-HSD) activity. The Leydig cells were recovered in a density band between 1.052-1.068 g/ml and grown in a chemically defined medium (Mather et al., 1981). In the absence of hCG, T production was low throughout the 6 days of culture. However, in response to hCG (10 mIU/ml), the cultured Leydig cells showed a progressive increase in T synthesis, which reached a maximum at Days 3-4. 8-Br-cAMP (1 mM) induced a comparable rise in T production to that obtained with hCG throughout the culture period. In contrast, 8-Br-cAMP induced a near maximal increase in dehydroepiandrosterone (DHEA) production from Day 1. This paper demonstrates that purified immature porcine Leydig cells in primary culture are a valuable model to study the ontogeny of Leydig cell function.     

Effects of photoperiod, beta-endorphin, and naloxone on in vitro secretion of testosterone in white-footed mouse (Peromyscus leucopus) testes

The effects of the pro-opiomelanocortin-derived beta-endorphin (B-EP) and the opioid antagonist naloxone on in vitro secretion (accumulation of testosterone (T) in the medium) of T by testicular cells were assessed in adult white-footed mice (Peromyscus leucopus). Animals were housed under long days (16L:8D) to maintain testicular function or under short days (8L:16D) to induce gonadal regression. In vitro treatment with B-EP or naloxone did not affect basal secretion of T in dispersed cells from active or regressed testes. However, B-EP caused a dose-dependent reduction in secretion of T from cells stimulated maximally with human chorionic gonadotropin (hCG) or dibutyryl cyclic adenosine 3', 5'-monophosphate (dbcAMP). Conversely, naloxone enhanced maximal hCG- and dbcAMP-stimulated secretion of T in testicular incubates from both long- (1.5-fold) and short-day (3.5-fold)-exposed mice. The finding that the addition of naloxone to maximally stimulated cells increased further the secretion of T is evidence that B-EP may act to inhibit gonadotropin-stimulated secretion of T. Also, the stimulatory effect of naloxone on cells from regressed testes indicates that B-EP may be involved in suppressing production of T during the gonadally regressed state. Testicular B-EP-like immunostaining is present within the cytoplasm of interstitial cells and is not apparent in the seminiferous tubules. Together, these results support the idea that in P. leucopus endogenous opioid peptides in the testes may aid in the regulation of testicular function throughout the yearly breeding cycle.     

Low aromatase activity and gene expression in human fetal testes

Because of previous indications that estradiol (E2) plays a role in the regulation of testicular testosterone (T) production in some species, the production of E2 and aromatase gene expression in human fetal testes were investigated. Testicular minces from 14 fetuses (fetal age 15-23 weeks) were incubated with and without 200 ng/ml highly purified hCG, and the production of E2 and T was measured by RIA. Basal T production was high at 15-18 weeks of gestation and decreased thereafter. Estradiol production was low in all testes. Aromatase mRNA (P-450 arom messenger ribonucleic acid) was not detectable in fetal testicular tissues when studied by Northern and dot blot techniques. Placenta and fetal liver expressed aromatase mRNA, but fetal ovary contained only miniscule amounts. HCG significantly stimulated the production of both T and E2 in the testes of older fetuses (19-23 weeks), but the testicular E2 production of the youngest fetuses (15-18 weeks) did not increase significantly after hCG stimulation. These results indicate that aromatase activity and gene expression are very low in human fetal testes. These findings suggest that E2 may not play a major role in testicular T production in the human fetus.     

Testicular responsiveness to hCG of the ovine fetus in the last third of gestation

The in vivo and in vitro testicular responsiveness to hCG of hemicastrated lamb fetuses 95-99, 110-118 and 130-141 days of gestational age was studied. Basal plasma testosterone (T) levels were similar at all ages (less than 0.25 ng/ml), while the mean testicular concentrations of dehydroepiandrosterone sulfate (DHA-S), 17 alpha-hydroxyprogesterone (17-OHP) and T were higher in 95- to 99-day-fold fetuses. Plasma T levels and the concentration of T, DHA-S, 17-OHP, androstenedione (A) and cyclic adenosine 3'5'-monophosphate (cAMP) were increased by hCG in the hemicastrated animal at all ages. cAMP and T production by enriched preparations of dispersed interstitial cells from control testes was increased by hCG in all groups. In fetuses pretreated with hCG in vivo the addition of hCG in vitro failed to modify cAMP and T production. 100 micrograms of LHRH to a 130-day-old fetus increased plasma LH and T levels. From these experiments, it is suggested that the low plasma LH and T levels found throughout the last trimester of fetal life reflect a relative lack of endogenous LHRH synthesis and/or release, rather than reduced testicular steroidogenic capacity.     

Inhibition of testosterone secretion by digitoxin in rat testicular interstitial cells.

Both in vivo and in vitro experiments were conducted to determined the effects of digitoxin on the secretion of testosterone, and its underlying mechanisms including testicular adenosine 3':5'-cyclic monophosphate (cAMP), and the activities of steroidogenic enzymes. Male rats were injected with digitoxin, human chorionic gonadotropin (hCG), or hCG plus digitoxin via a jugular catheter. Blood samples were collected immediately before and at 30 and 60 min after the challenge, and analyzed for testosterone by radioimmunoassay. In an in vitro study, rat testicular interstitial cells were isolated and incubated with digitoxin, hCG, 8-bromo-cAMP (8-Br-cAMP), digitoxin plus hCG, or digitoxin plus 8-Br-cAMP at 34 degrees C for 1 h. The media were collected and analyzed for testosterone. For studying cAMP accumulation, testicular interstitial cells were incubated for 1 h in the medium containing isobutyl-1-methylxanthine (IBMX) and different doses of digitoxin with the absence or presence of hCG. After incubation, cells were processed for determining cAMP content. Intravenous injection of digitoxin decreased hCG-stimulated, but not basal, plasma testosterone levels. Administration of digitoxin in vitro resulted in an inhibition of both basal and hCG- as well as 8-Br-cAMP-stimulated release of testosterone. In addition, digitoxin diminished hCG-stimulated cAMP accumulation in rat testicular interstitial cells. Furthermore, digitoxin inhibited the activity of cytochrome P450 side chain cleavage enzyme (P450scc) but failed to affect the activities of other steroidogenic enzymes. Taken together, these results suggest that the acute inhibitory effect of digitoxin on the testosterone production in testicular interstitial cells involves, at least partly, an inefficiency of post-cAMP events, and a decrease of P450scc activity.     

Androgen secretion and characteristics of testicular HCG binding in cryptorchid rats

Response of the cryptorchid testis to gonadotrophic stimulation was assessed by comparison of the androgen production capability in vivo and in vitro with that of the normal scrotal testis. Serum androgen concentrations in cryptorchid rats were similar to those in normal rats, and the incremental increase 60 min after 50 i.u. hCG (i.v.) was about 7-fold for both groups. Basal and hCG-stimulated androgen production in vitro was higher for abdominal testes (557 and 3286 ng/pair) than for scrotal tests (157 and 504 ng/pair). Specific binding of hCG by testicular homogenates was slightly higher (P < 0.05) for cryptorchid testes when expressed per unit weight, but Scatchard analysis indicated that although hCG binding affinities did not differ (Ka = 2 x 10(10) M-1), hCG binding capacity of cryptorchid testes was only 75 ng, compared to 219 ng for scrotal testes. These data indicate that a discrepancy exists between androgen production in vivo and in vitro by cryptorchid testes and that normal serum androgen concentrations are maintained in the presence of decreased numbers of testicular LH/hCG receptors.     

The influence of short photoperiod on testicular and circulating levels of testosterone precursors in the adult golden hamster

The in vivo effects of short photoperiod (SPP, 6L:18D) for 8 and 12 wk on plasma and testicular levels of testosterone (T) precursors in adult golden hamsters were evaluated. Plasma and testicular progesterone (P), 17 alpha-hydroxyprogesterone (17 alpha-OHP), androstenedione (A-dione), and T were measured after 5 injections of saline or human chorionic gonadotropin (hCG) (5 or 25 IU/day). The basal levels of follicle-stimulating hormone (FSH), luteinizing hormone (LH), and prolactin (PRL) in circulation were also determined. There were significant reductions in the weight of the testes in animals exposed to SPP. After 12 wk in SPP, circulating levels and testicular content of 17 alpha-OHP, A-dione, and T were significantly reduced, suggesting that the decrease in T secretion may be associated with the impairment of synthesis and/or action of 17 alpha-steroid hydroxylase, C17-20 steroid lyase, and 17 beta-hydroxysteroid dehydrogenase enzymes in the testes. Exposure to SPP for 8 wk resulted in decreased plasma and testicular content of T. Although there were reductions in testicular content of 17 alpha-OHP and A-dione, this was not reflected in plasma levels of these steroids. After 8 and 12 wk of exposure to SPP, hCG treatment increased the total amounts of T precursors (except P at 8 wk) in the testes, but the values attained in animals exposed to 12 wk of SPP remained below those observed in hamsters kept in a long photoperiod (14L:10D), suggesting that gonadotropin replacement alone may be insufficient to normalize testicular steroidogenesis.(ABSTRACT TRUNCATED AT 250 WORDS)     

Reduction of testicular human chorionic gonadotropin receptors by human chorionic gonadotropin in vivo and in vitro

Changes in rat and human testicular human chorionic gonadotropin (hCG) binding sites induced by hCG were estimated in vivo and in vitro. After a single administration of hCG, the specific 125I-hCG bindings were significantly reduced for 7 and 5 days in rat and human testes, respectively. Thereafter, 125I-hCG bindings had recovered to pretreatment values by the 14th day after the administration. Occupied hCG bindings accounted for about half of the reduced bindings on the day after administration of hCG. After this time, however, the occupancy did not contribute so much to the reduction of the bindings. In experiments in vitro using the organ culture technique, an exposure to hCG for 24 h induced a dose-related significant loss of the specific 125I-hCG bindings for 7 and 5 days in rat and human testes, respectively. Thereafter, the loss was gradually recovered. These patterns of changes in 125I-hCG bindings in vitro were similar to those in vivo. These findings suggest that the reduction in hCG binding sites by hCG is due to not only occupancy but also downregulation of the binding sites and that the testicular organ culture method used in the present study is useful to study hormonal regulation of testicular function, especially in human testes.     

Stimulation of leydig cell testoterone secretion in vitro and in vivo in hypophysectomized rats by an agonist of luteinizing hormone releasing hormone

Injection of a luteinizing hormone-releasing hormone (LHRH) agonist into 55-day-old male rats which had been hypophysectomized 3 days earlier resulted in a 10- to 30-fold increase in the levels of testosterone in serum and testicular interstitial fluid (IF) in the 4h following injection. The levels achieved were within or above the normal range for intact untreated rats of this age. In similar animals, injection of LHRH agonist also enhanced the serum testosterone response to injected hCG at $\text{1}\text{1}\text{2}\text{h}$, but not at later times after injection, and by 24h reduced IF levels of testosterone suggested that LHRH agonist had begun to inhibit stimulation by hCG. $\text{In vitro}$, dispersed Leydig cells from untreated hypophysectomized rats showed a 2-fold increase in testosterone responsiveness to LHRH agonist when compared to cells from intact rats, and this change was associated with an 80% increase in the number of Leydig cell LHRH-receptors.     

Testosterone, androstenedione, progesterone and 17 alpha-hydroxyprogesterone in plasma and testes of immature rats under basal conditions and after hCG stimulation. Effect of bilateral cryptorchidism

Rats were made bilaterally cryptorchid at 21 days of age; sham-operated rats were used as controls. At 35 days, the animals were injected i.m. with saline or with 10 IU hCG. Progesterone, 17-hydroxyprogesterone, androstenedione and testosterone were measured in both testes and plasma under basal conditions and 2, 4, 8, 12, 24 and 72 h respectively after injection. The plasma levels and intratesticular contents of the steroids were generally lower in cryptorchid rats. The patterns of the steroid response to hCG were similar in both groups: in the testes and in the plasma, they increased acutely following hCG injection (except testicular androstenedione), then, after 72 h, returned to normal values in the plasma but remained higher than the basal values in the testes. These results suggest that there are no gross abnormalities in the testicular steroidogenic pathways and that the mechanism of action of hCG on the Leydig cells is unaltered in bilaterally cryptorchid immature rats.     

Effect of serum and serum lipoproteins on testosterone production by adult rat Leydig cells in vitro.

The effect of serum factors other than luteinizing hormone on Leydig cell testosterone secretion was examined using an in vitro bioassay system based on the stimulation of purified adult rat Leydig cells during a 20 h incubation in the presence of a maximal dose of human chorionic gonadotrophin (hCG). Charcoal-extracted serum and testicular interstitial fluid (IF) from normal adult male rats were separated into lipoprotein and lipoprotein-deficient fractions by density ultracentrifugation. Stimulatory bioactivity was found in the lipoprotein fraction of both serum and IF, although the levels of lipoprotein and corresponding bioactivity recovered from IF were significantly lower (25%) than those of serum. There was no difference between the effects of serum lipoproteins on Leydig cell testosterone production stimulated by either hCG or dibutyryl cAMP. In time-course studies, the serum lipoprotein fraction had no effect on hCG-stimulated testosterone production in vitro at 3.0 or 6.0 h, but partially prevented the normal decline in hCG-stimulated testosterone production after 6.0 h. In contrast, unfractionated serum was stimulatory at all time-points. In the absence of hCG, the lipoprotein fraction was stimulatory at both 6.0 and 20 h, although not at 3.0 h. The lipoprotein-deficient protein fraction of serum had no effect on hCG-stimulated testosterone production alone, but significantly enhanced the bioactivity of the lipoprotein fraction, and caused a dose-dependent stimulation of testosterone production in the presence of a constant concentration of serum lipoproteins. Both a stimulatory peak of activity (apparent MW 40-80 kDa), and a large MW (> 100 kDa) inhibitor of testosterone production were identified in serum after fractionation by gel filtration (Sephadex G-100). The data indicate that (i) the stimulatory effect of serum on short-term hCG-stimulated Leydig cell testosterone production in vitro is predominantly due to the serum lipoprotein fraction, possibly by providing additional precursors for testosterone synthesis, (ii) the biological activity of the lipoproteins is influenced by both stimulatory and inhibitory serum proteins in addition to luteinizing hormone, and (iii) that serum lipoproteins may be involved in supporting Leydig cell steroidogenesis in vivo.     

Effect of GABA and benzodiazepines on testicular androgen production

We have evaluated the effect of Ro5-4864, a selective probe to label peripheral type benzodiazepine receptor, on "in vitro" testicular androgen production. Decapsulated testes from adult rats showed a significant increase in the basal and hCG-stimulated testosterone secretion into the medium in response to 10(-5) M, 10(-6) M, and 10(-7) M Ro5-4864. In addition, we have studied the changes in testicular GABA content at three different ages and we found its highest concentration at 31 days of age. When we evaluated the effect of GABA on "in vitro" androgen production at different stages of gonadal maturation we observed that the highest concentration of GABA (10(-6) M) was able to modify the basal and hCG-stimulated androgen production from adult (60 days) and pubertal (45 days) testes. In addition, when prepubertal testes (31 days) were incubated under basal conditions, 10(-6) M GABA induced a significant increment of androstanediol production, while the stimulatory effect of hCG was reduced in the presence of the same GABA concentration. The present results suggest that GABA plays a physiological role in the regulation of rat testicular androgen production depending on the stage of sexual maturation.     

In vitro effects of estradiol-17β, monobutyl phthalate and mono-(2-ethylhexyl) phthalate on the secretion of testosterone and insulin-like peptide 3 by interstitial cells of scrotal and retained testes in dogs

The objective was to determine the effects of estradiol-17β, monobutyl phthalate (MBP) and mono-(2-ethylhexyl) phthalate (MEHP) on testosterone and insulin-like peptide 3 (INSL3) secretions in cultured testicular interstitial cells isolated (enzymatic dispersion) from scrotal and retained testes of small-breed dogs. Suspension cultures were treated with estradiol-17β (0, 10, and 100 ng/mL), MBP (0, 0.8, and 8 mmol/L) or MEHP (0, 0.2, and 0.8 mmol/L) for 18 h, in the presence or absence of 0.1 IU/mL hCG. Testosterone (both basal and hCG-induced) and INSL3 (basal) concentrations were measured in spent medium. Effects of estradiol-17β, MBP, and MEHP on testosterone and INSL3 secretions were not affected (P > 0.15) by cell source (scrotal versus retained testis); therefore, data were combined and analyzed, and outcomes reported as percentage relative to the control. In testicular interstitial cells, basal testosterone secretion was increased (P < 0.01) by 100 ng/mL estradiol-17β (130.2 ± 10.6% of control). Among phthalates, 0.2 and 0.8 mmol/L MEHP stimulated (P < 0.01) basal testosterone secretion (135.5 ± 8.3% and 154.6 ± 12.9%, respectively). However, hCG-induced testosterone secretion was inhibited (P < 0.01) by 8 mmol/L MBP (67.7 ± 6.0%), and tended to be inhibited (P = 0.056) by 0.8 mmol/L MEHP (84.5 ± 5.6%). Basal INSL3 secretion was inhibited (P < 0.01) by 8 mmol/L MBP (73.6 ± 6.8%) and 0.8 mmol/L MEHP (76.9 ± 11.3%). In conclusion, we inferred that estradiol-17β and certain phthalate monoesters had direct effects on secretions of testosterone and INSL3 in canine testicular interstitial cells, with no significant difference between scrotal and retained testes.