当前位置: 首页 > 医学版 > 期刊论文 > 内科学 > 内分泌学杂志 > 2005年 > 第2期 > 正文
编号:11168197
Pubertal and Adult Leydig Cell Function in Mullerian Inhibiting Substance-Deficient Mice
     Departments of Pediatrics and Cell Biology (X.W., M.M.L.) and Academic Computing (S.P.B.), University of Massachusetts Medical School, Worcester, Massachusetts 01655; and Pediatric Endocrine Division (R.A.), Duke University Medical Center, Durham, North Carolina 27710

    Address all correspondence and requests for reprints to: Mary M. Lee, M.D., Pediatric Endocrinology, Department of Pediatrics and Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655. E-mail: mary.lee@umassmed.edu.

    Abstract

    Mullerian inhibiting substance (MIS) causes Mullerian duct regression during sexual differentiation and regulates postnatal Leydig cell development. MIS knockout (MIS-KO) mice with targeted deletions of MIS develop Leydig cell hyperplasia, but their circulating androgen concentrations are reportedly unaltered. We compared reproductive hormone profiles, androgen biosynthesis, and the expression of key steroidogenic and metabolic enzymes in MIS-KO and wild-type (WT) mice at puberty (36 d) and sexual maturity (60 d). In pubertal animals, basal testosterone and LH concentrations in plasma were lower in MIS-KO than WT mice, whereas human chorionic gonadotropin-stimulated testosterone concentrations were similar. In adults, basal LH, and both basal and human chorionic gonadotropin (hCG)-stimulated testosterone concentrations were similar. Purified Leydig cells from pubertal MIS-KO mice had lower testosterone but higher androstanediol and androstenedione production rates. In contrast, testosterone, androstanediol, and androstenedione production rates were all lower in adult MIS-KO Leydig cells. Steroidogenic acute regulatory protein expression was lower in pubertal MIS-KO mice compared with WT, whereas 17?-hydroxy-steroid dehydrogenase and 5-reductase were greater, and P450c17 and P450scc were similar. The expression of steroidogenic acute regulatory protein and 17?-hydroxysteroid dehydrogenase was lower in adult MIS-KO mice, whereas that of 5-reductase, P450c17, and P450scc was similar. Collectively, these results suggest that in the absence of MIS, Leydig cells remain less differentiated, causing an altered intratesticular androgen milieu that may contribute to the infertility of MIS-KO mice. In immature mice, this deficit in steroidogenic capacity appears to be mediated by a direct loss of MIS action in Leydig cells as well as by indirect effects via the hypothalamic-pituitary-gonadal axis.

    Introduction

    MULLERIAN INHIBITING SUBSTANCE (MIS), also known as anti-Mullerian hormone (AMH), is a gonad-specific member of the TGF-? family of growth and differentiation factors (1). During male sexual differentiation, MIS inhibits differentiation of the Mullerian ducts into the uterus, Fallopian tubes, and upper vagina. Affected males with mutations of the MIS or MIS receptor genes have the persistent Mullerian duct syndrome, in which the external genitalia virilize, but the Mullerian ducts differentiate into uterine and oviductal tissue instead of undergoing involution (2). The residual Mullerian structures interfere with testicular descent and are associated with transverse testicular ectopia. The undescended testes are hypermobile and at risk for torsion as well as for malignant transformation.

    MIS not only induces regression of the Mullerian ducts during male sexual differentiation (3, 4, 5) but also plays a critical paracrine role in the regulation of Leydig cell development and testosterone biosynthesis (6, 7, 8, 9, 10). MIS has also been shown to inhibit proliferation of prepubertal progenitor Leydig cells and prevent regeneration of Leydig cells after chemical ablation (11, 12). These actions of MIS in the testis are mediated directly through the MIS type II receptor, which is abundantly expressed in Leydig cells (7). Male transgenic mice overexpressing MIS have feminized genitalia secondary to fewer Leydig cell numbers and decreased serum testosterone concentrations (4, 13). Conversely, mice with targeted deletions of MIS [MIS knockout (MIS-KO)] and/or its receptor develop Mullerian structures but also manifest Leydig cell hyperplasia, focal Leydig cell tumors, and infertility (14, 15). Recently, MIS has been shown to induce FSH mRNA expression and enhance LH? promoter activity in a pituitary cell line, indicating that changes in MIS action may also affect the hypothalamic-pituitary-gonadal axis (16).

    In adult MIS-KO mice, although Leydig cell numbers are increased and the mRNA expression of a key steroidogenic enzyme, P450c17, is more abundant, peripheral testosterone and LH concentrations are reportedly unaffected (8). These observations were surprising in light of the findings of decreased serum testosterone in MIS transgenic mice and the inhibitory actions of MIS on testosterone biosynthesis in both in vitro (6, 8, 10) and in vivo systems (9, 17). To explore the role of MIS in Leydig cell development in depth and clarify these seemingly discrepant findings on the effects of MIS on steroid production, we compared reproductive hormone profiles, androgen biosynthetic capacity, and the expression of key enzymes in the steroidogenic and metabolic pathway in MIS-KO and wild-type (WT) mice at two discrete ages, at puberty and in the adult.

    Materials and Methods

    Animals

    MIS-KO mice on the C57BL/6J background were a generous gift from Richard Behringer (M.D. Anderson Cancer Center, Houston, TX) (14). Heterozygous mice were mated to generate homozygous affected MIS-KO mice, which were identified by genotyping using a common sense primer (5'-ggaacacaagcagagcttcc-3') and antisense primers specific to either the KO (5'-tcgtgctttacggtatcgc-3') or WT (5'-gagacagagtccatcacgtacc-3') mice. At 36 and 60 d of age, baseline blood samples were obtained, and then MIS-KO and WT mice were injected ip with 60 IU human chorionic gonadotropin (hCG) to maximally stimulate testosterone production. hCG-stimulated blood samples were obtained after 90 min, and then the mice were weighed and euthanized for collection of testes according to established guidelines. The testes were used either for isolation of Leydig cells or preparation of testicular extracts by homogenization in 900 μl TBSG buffer (1 g/liter gelatin, 28 mM Trizma HCl, 22 mM Trizma base, 100 mM NaCl, and 15 mM NaN3). Both plasma and testicular extract supernatants were assayed for hormonal determination. All experimental protocols were approved by the Institutional Animal Care and Use Committee of Duke University Medical Center (A166-02-06).

    Isolation of primary Leydig cells

    Freshly harvested testes were decapsulated, enzymatically dispersed with collagenase (0.5 mg/ml), and then subjected to Percoll gradient fractionation as described previously for rat Leydig cell isolation (18, 19), with the following minor modifications for collection of murine Leydig cells (20). The Leydig cells were collected from the Percoll gradient at a specific gravity of 1.03 (d 36) and 1.09 g/ml (d 60) and then purified through a 2.5–10% BSA (Sigma Chemical Co., St. Louis, MO) gradient to remove contaminating sperm rather than undergoing elutriation (20). The purity of the Leydig cell cultures was evaluated for each cell isolation by histochemical staining for 3?-hydroxysteroid dehydrogenase (3?HSD) activity with 0.4 mM etiocholanolone (21) and were consistently greater than 90% enriched for Leydig cells.

    In vitro cell culture

    Freshly harvested Leydig cells were cultured at 104 cells/1.5-ml tube in 1 ml serum-free medium (phenol-red-free DMEM/F12) supplemented with 1 ng/ml LH, 15 mM HEPES, 26 mM sodium bicarbonate, 0.1% BSA, and 12 μg/ml gentamycin. After gentle agitation for 3 h in a 34 C rotatory water bath at 85 rpm, the cells were pelleted by centrifugation at 14,000 rpm for 8 min for collection of conditioned media, which were stored at –20 C until hormone assay.

    Hormonal determination

    Testosterone and LH were measured in basal and hCG-stimulated plasma samples. Testosterone, androstenedione, and 5-androstane-3-17?-diol (androstanediol) rather than 5-androstane-3?-17?-diol were measured in conditioned media from in vitro cultures and supernatant of total testicular homogenates (22, 23, 24). All samples were assayed in duplicate. The total androgen biosynthetic capacity of the isolated Leydig cells was estimated by summing the mean concentrations of the three predominant androgens synthesized at these two developmental ages.

    The testosterone antibody was obtained from Dr. Gordon Niswender (22) and the 5-androstane-3-17?-diol (androstanediol) rabbit antiserum was kindly provided by Dr. Gerald Barbe, University of Western Ontario, London, Canada (25). Both of these antibodies are highly specific and have minimal cross-reactivity with other commercial steroids. The commercial androstenedione rabbit antiserum from Sigma is 100% specific for androstenedione and has cross-reactivity of 67% with 5-androstane-3,17-dione (a hydroxylated metabolite of androstenedione), 6% with dihydroepiandrostenedione, and 4.5% with testosterone (26). The androstenedione RIA was performed as recommended by the manufacturer. The limit of detection for all three androgen RIAs was similar at 3.3 pg/ml. Intra- and interassay coefficients of variation (CV) were 9.04 and 7.76%, respectively, for testosterone, 6.1 and 5.76% for androstenedione, and 6.98 and 3.97% for androstanediol. LH was assayed in plasma from littermates by mouse LH sandwich IRMA at the University of Virginia Center for Research in Reproduction Ligand Assay and Analysis Core funded by National Institute of Child Health and Human Development (Specialized Cooperative Centers Program in Reproduction Research) Grant U54-HD28934. The sensitivity of the LH assay is 0.07 ng/ml. The intraassay CV of 6.0% is determined by taking the average of the CVs for all controls (in the low, medium, and high range) in that particular assay. The interassay CV of 12.5% is the mean overall CV for the LH assay in the calendar year, derived by calculating the mean of all the CVs for the low-, medium-, and high-range controls.

    Northern analysis

    Total RNA from testis was isolated using Trizol (Invitrogen, Carlsbad, CA). Total RNA (10 μg) samples were electrophoresed in 1.0% denaturing agarose gels in 3[N-morpholino]propanesulfonic acid and then blotted onto nylon membranes (NEN Life Science Products Inc., Boston, MA) and hybridized overnight at 68 C with 1 x 106 cpm/ml antisense riboprobe against P450scc and P450c17. Blots were reprobed with an S16 DNA probe at 42 C to normalize the target signal. Northern analysis was performed three times with three different isolations of total RNA from whole testes of MIS-KO and WT littermates. Band intensities were quantified by NIH Image Software, and the results were expressed as the mean and SEM of these three analyses.

    Immunohistochemistry

    Testes from a minimum of three MIS-KO and WT littermate mice were fixed in Bouin’s fixative overnight and then embedded in paraffin for cutting of 5-μm sections. The slides for immunohistochemistry were blocked for 1 h in 1.5% normal goat serum and then incubated overnight at 4 C with the primary antibody at a concentration of 1:6000 for anti-P450c17 (Dr. Dale Buchanan Hales, University of Illinois, Chicago, IL) and 1:3000 for anti-P450scc (Research Diagnostics, Flanders, NJ). The sections were incubated with biotinylated antirabbit IgG secondary antibody followed with the avidin-biotin immunoperoxidase system (Vectastain Elite ABC kit, Vector Laboratories, Inc., Burlingame, CA) and 3,3'-diaminobenzidine tetrachloride (Roche Molecular Biochemicals, Indianapolis, IN). The sections were counterstained with hematoxylin. Alternate sections were processed with PBS instead of the primary antibody as a negative primary antibody control. P450scc and P450c17 staining of sections from MIS-KO and WT mice were compared by counting 500 cells in random sections.

    Quantitative real-time PCR (QPCR)

    Freshly isolated Leydig cells were counted in a hemocytometer and aliquotted into PCR tubes for direct QPCR. The cells were subjected to hypotonic lysis with distilled H2O at a concentration of 1000 cells per tube and then snap-frozen and stored at –70 C until QPCR was performed. For each reverse transcriptase (RT) reaction, the lysis of 1000 Leydig cells after DNase treatment to remove contaminating DNA were used for cDNA synthesis using the Ominiscript RT kit (QIAGEN, Valencia, CA). One twentieth of each RT reaction was used for QPCR amplification using the iCycler iQ Real-Time PCR detection system (Bio-Rad, Hercules, CA). For all experiments, murine ?-actin was amplified as an internal quality control. The QPCR was carried out in a 25-μl volume using the Quantitect SYBR Green PCR kit (QIAGEN). Each reaction contained 1x reaction buffer, 2.5 mM MgCl2, 200 μM dNTPs, 200–400 nM of the specific primer, and HotStarTaq DNA polymerase. After 45 cycles at 95 C for 30 sec, 52 C for 30 sec, and 72 C for 45 sec, the relative amount of initial mRNA copies was determined by calculating the difference () in threshold cycle (Ct) value between the target gene and actin, which is then interpreted as 2–Ct to represent the target gene concentration. The real-time PCR analysis was repeated at least four times for each gene with different preparations of primary Leydig cells harvested from MIS-KO and WT littermates. The data represent the mean ± SEM of at least four RT-PCR analyses.

    Primers were designed to produce an amplicon spanning at least one intron boundary (see Table 1).

    TABLE 1. Primer sequences

    Statistical analysis

    The distributional characteristics of all dependent variables were evaluated both graphically through the visual inspection of frequency histograms of model residuals and by the Kolmogorov-Smirnoff goodness-of-fit test for normality (27). Dependent variables found to have poor compliance with a normal distribution were transformed with logarithms and retested. Thus, all outcomes were found to have normally distributed residuals either in their original units or after log transformation.

    The effects of MIS deletion on in vitro androgen production and plasma and intratesticular hormone concentrations were evaluated by ANOVA for general linear mixed models (28) with experimental sets (litters) as a random effect and genotyping as a fixed effect. The data were expressed as within-group means ± SEM based on the linear mixed and estimated population marginal means and SEM (29).

    mRNA enzyme expression and total androgen data were analyzed by the paired t test and expressed as arithmetic means ± SEM (30). The Student t test was used for the analysis of testis and body weight.

    All computations were performed using SPSS for Windows version 12 (2003) and SAS Proc Mixed (SAS 1997) (31).

    Results

    Body and testicular weight

    The mean body weights were similar in MIS-KO and WT mice at both 36 and 60 d of age. The mean testis weight in MIS-KO mice was not significantly heavier at 36 d but was significantly heavier at 60 d of age (P < 0.01) (Table 2). When we normalized testicular size by body weight, the testis to body weight ratio for the MIS-KO mice was 17.6% greater than that for WT mice at 36 d of age (P < 0.05) and 23.7% greater at 60 d of age (P < 0.01). Whole testis intratesticular fluid weight and volume were similar in both MIS-KO and WT mice (data not shown).

    TABLE 2. Mean body and testis weights in MIS-KO and WT mice at 36 and 60 d of age

    Plasma hormone levels

    At 36 d of age, basal (pre-hCG) testosterone concentrations were 49.6% lower and LH was 75.6% lower in MIS-KO mice than that of WT mice, but hCG-stimulated testosterone concentrations were similar (Table 3). In contrast, by 60 d of age, the concentrations of LH and both basal and hCG-stimulated testosterone were similar in MIS-KO and WT mice.

    TABLE 3. Basal testosterone, LH, and hCG-stimulated testosterone concentrations in pubertal and adult mice

    Intratesticular androgens

    In hCG-stimulated pubertal (36 d) and adult (60 d) mice, the concentrations of androgens (testosterone, androstenedione, and androstanediol) in intratesticular fluid were no different between MIS-KO and WT mice (Table 4).

    TABLE 4. Concentrations of androgens in hCG-stimulated pubertal and adult mice

    In vitro androgen production by Leydig cells

    At 36 d, the testosterone production rate was 48.2% lower in Leydig cells from MIS-KO mice than WT littermates (P < 0.01; n = 5 sets of littermates) (Fig. 1A). At 60 d, the in vitro testosterone production rate was 41.8% lower (P < 0.01; n = 3 sets of littermates) (Fig 1B).

    FIG. 1. In vitro androgen production rate of Leydig cells isolated from MIS-KO and WT testes at 36 and 60 d of age. The purified Leydig cells were cultured in medium with 1 ng/ml LH. Values are expressed as ng/106 cells·3 h (mean ± SEM). *, P < 0.05; **, P < 0.01.

    At 36 d, the production rate of androstenedione was 63.8% greater (P < 0.01; n = 6 sets of littermates) and that of androstanediol was 2-fold greater (P < 0.05; n = 7 sets of littermates) in Leydig cells from MIS-KO than WT mice (Fig. 1A). In contrast, at 60 d of age, the production of both steroids was reduced; androstenedione was 53.5% lower (P < 0.01; n = 3 sets of littermates) and androstanediol was 45.2% lower (P < 0.05; n = 4 sets of littermates) (Fig. 1B).

    The total in vitro androgen production rate in Leydig cells from MIS-KO mice at 36 d of age was not significantly lower than WT (KO 367.06 ± 69.36 and WT 604.20 ± 80.88 pmol/106cells·3 h; not significant) (data not shown). Total androgen production was reduced in MIS-KO mice at 60 d (KO 175.10 ± 9.52 vs. WT 397.60 ± 29.40 pmol/106cell/3 h; P < 0.05).

    Expression of steroidogenic enzymes

    Image analysis of the three separate Northern analyses for each gene detected no difference in the mRNA expression of P450scc or P450c17 at either age (Fig. 2A). P450scc and P450c17 immunohistochemistry also did not reveal noticeable differences in protein expression of these two enzymes (Fig. 2B).

    FIG. 2. Expression of P450scc and p450c17 in testes of MIS-KO and WT mice at 36 and 60 d of age by Northern blot analysis (A) and immunochemistry (B) (magnification, x200). A, Representative Northern analysis with a table of the mean and SEM image analysis results for three different Northern hybridizations shown below.

    At 36 d of age, the mean initial mRNA copy number of steroidogenic acute regulatory protein (StAR) was 60.3% lower in Leydig cells of MIS-KO than that of WT mice (0.23 ± 0.02 vs. 0.58 ± 0.04; P < 0.05) (Fig. 3A), whereas that of 5-reductase type I and 17?-hydroxysteroid dehydrogenase (17?-HSD) were approximately 6-fold and 2-fold higher, respectively (Fig. 3, B and C). At 60 d of age, the mean initial mRNA copy number of StAR was 54.1% lower in Leydig cells of MIS-KO than that of WT (Fig. 3A); whereas 5a-reductase type I was not statistically lower (Fig. 3B), and 17?-HSD was conversely decreased by 36.8% (Fig. 3C).

    FIG. 3. mRNA expression of StAR (A), 5-reductase type I (B), and 17?-HSD (C) in Leydig cells of MIS-KO and WT mice by quantitative RT-PCR. The target gene concentration was normalized by calculating the ratio of the initial targeted gene concentration to actin. The data represent the mean and SEM of at least four RT-PCR analyses for each target gene from different preparations of primary Leydig cells. *, P < 0.05; **, P < 0.01.

    Discussion

    Our data demonstrate that mice with a targeted deletion of MIS have an altered developmental profile of androgen biosynthesis and metabolism. MIS transgenic mice have been reported to have fewer Leydig cells and decreased testosterone production (4), and exogenously administered MIS inhibits testosterone biosynthesis in adult rodents (9, 17) and differentiated Leydig cells (6, 7, 10). Mice bearing targeted deletions of the MIS gene or receptor are known to develop Leydig cell hyperplasia, thus we anticipated finding elevated concentrations of androgens in the MIS-KO mice. In contrast, our studies show that pubertal 36-d-old MIS-KO mice have lower basal plasma testosterone concentrations than WT mice. This difference is eliminated after hCG stimulation. Contrary to our expectations, 60-d adult MIS-KO and WT mice had similar concentrations of both basal and hCG-stimulated plasma testosterone. Moreover, the concentrations of intratesticular androgens in MIS-KO and WT mice were also comparable at both ages.

    The determination of plasma and intratesticular androgens, however, is essentially a steady-state measurement that reflects the net actions of androgen biosynthetic and metabolizing enzymes. During the transition from a prepubertal to a sexually mature animal, the quantity and relative activity of these enzymes change, resulting in developmental differences in the specific androgen produced at different maturational stages (32). Consequently, adult rodents largely synthesize testosterone, whereas immature animals produce predominantly androstenedione and 5-androstane-3-17?-diol (androstanediol) (33, 34, 35, 36, 37). An evaluation of androgen production in pubertal animals must therefore be undertaken in the context of the normal maturational changes in androgen production and metabolism. In pubertal 36-d-old MIS-KO mice, the in vitro production rate of testosterone per cell was reduced, whereas that of both androstenedione and androstanediol was increased. These alterations in steroid production resulted in a relative increase in immature steroids, whereas total androgens were unaffected because of the reduced testosterone production. This pattern of androgen production by the MIS-KO Leydig cells resembles a more immature profile than is typically observed at this age (8, 15, 32, 33, 35, 38). In the adult mice, the androgen production rate per Leydig cell was decreased overall, as well as for each specific androgen measured, because of a general inhibition of androgen production.

    The analysis of mRNA expression of the steroidogenic enzymes supports our finding of relative immaturity in the profile of androgen production by the 36-d-old MIS-KO Leydig cells. The expression of StAR was lower, whereas the expression of 5-reductase was greater, consistent with expression in younger peripubertal testes. This pattern of expression would limit the synthesis of testosterone and increase the synthesis of metabolized androgens such as androstanediol. In early puberty, the expression of the LH receptor is low and increases with pubertal maturation (21, 32, 39). Similarly, the MIS-KO Leydig cells from 36-d-old mice are presumably less differentiated and their decreased expression of the LH receptor is more consistent with an earlier stage of differentiation. The reduction in in vitro testosterone synthetic capacity and the lower expression of the steroidogenic enzymes in the MIS-KO mice compared with WT suggests that the absence of MIS action causes a delay in pubertal differentiation of the Leydig cells. This maturational delay persists at 60 d of age, when the production of testosterone, androstenedione, and androstanediol are all reduced in Leydig cells from MIS-KO mice compared with their WT littermates, and StAR and 17?-HSD mRNA are less abundant. The lack of MIS action therefore affects androgen biosynthetic capacity by altering the relative expression of the steroidogenic and metabolic enzymes.

    In addition to steroidogenic capacity, net androgen production also depends on the number of steroidogenically active Leydig cells in the testis at a given age. In the MIS-KO mice, the reported increase in Leydig cell numbers must be considered when evaluating the total testicular capacity for androgen production (8, 15). The reduced androgen biosynthetic capacity of individual cells, therefore, is compensated by the increased number of Leydig cell numbers in MIS-KO mice. In sexually immature mice, this compensation is incomplete. In adult mice, the increase in cell numbers restores the plasma testosterone concentrations despite a 42% reduction in the capacity per cell for testosterone production.

    We have shown that the loss of MIS expression results in lower plasma LH and testosterone concentrations in young pubertal animals. In contrast, in adult mice at 60 days of age, our data confirm previous observations that LH and testosterone concentrations are no different in MIS-KO mice (8). MIS has been reported to activate an LH?-luciferase promoter construct (16), thus the absence of MIS action in the KO mice might also result in decreased pituitary secretion of LH. Consequently, the decreased production of testosterone and immaturity of the steroidogenic enzymes could be caused by a direct action of MIS on Leydig cell maturation as well as an indirect action secondary to its effect on the hypothalamic-pituitary-gonadal axis. These data reveal that the roles of MIS in the regulation of Leydig cell proliferation, differentiation, and functional maturation are complex and not the simple result of a direct specific action on P450c17 expression as presumed from previous findings in MIS transgenic mice and adult Leydig cells and Leydig cell tumors (9, 10, 14, 17, 40).

    Sertoli cell secretion of MIS is high at birth and then declines gradually during the prepubertal years to a lower level in the sexually mature testis (41, 42, 43). In the adult, the expression of MIS is stage specific with the highest expression at stage VII of the spermatogenic cycle (44). It is possible that the loss of MIS action affects local expression of the steroidogenic enzymes and modulates androgen production within specific foci within the testis. These focal alterations in androgen production might be detrimental for spermatogenesis within adjacent tubules, thereby causing subtle impairment of reproductive fertility in the MIS-KO mice.

    We believe that in the prepubertal animal, MIS expression is high and restricts proliferation of the Leydig cell mesenchymal precursors. At the onset of puberty, a number of factors, including LH and IGF-I, stimulate proliferation of the progenitor cells and induce the expression of the steroidogenic enzymes. During the process of pubertal maturation, Leydig cell proliferative capacity declines while the expression of steroidogenic enzymes increases. We speculate that MIS plays a role in this developmental process by suppressing the proliferation of the progenitor cells that promote their differentiation. In adult animals with fully differentiated Leydig cells, MIS negatively regulates steroidogenesis by down-regulating P450c17 expression, but this action is spermatogenic stage specific because of the expression pattern of MIS and its receptor (44). Targeted deletion of the MIS gene promotes continued proliferation of Leydig cell progenitors and induces a maturational delay in their differentiation. Individual Leydig cells from MIS-KO mice thus have a reduced capacity for testosterone biosynthesis that may cause focal deficits in androgen production. The loss of this critical antiproliferative action of MIS causes hyperplasia of the Leydig cells that compensates systemically for the reduced steroidogenic capacity of individual cells. We conclude that MIS, as one of multiple factors that play essential roles in the developing testis, has critical actions on Leydig cell development that varies with their state of differentiation and maturation.

    Acknowledgments

    We thank the University of Virginia Center for Research in Reproduction Ligand Assay and Analysis Core for performing the plasma LH assay.

    References

    Massague J, Wotton D 2000 Transcriptional control by the TGF-?/Smad signaling system. EMBO J 19:1745–1754

    Belville C, Josso N, Picard JY 1999 Persistence of Mullerian derivatives in males. Am J Med Genet 89:218–223

    Arango NA, Lovell-Badge R, Behringer RR 1999 Targeted mutagenesis of the endogenous mouse MIS gene promoter: in vivo definition of genetic pathways of vertebrate sexual development. Cell 99:409–419

    Behringer RR, Cate RL, Froelick GJ, Palmiter RD, Brinster RL 1990 Abnormal sexual development in transgenic mice chronically expressing Mullerian inhibiting substance. Nature 345:167–170

    Lee MM, Donahoe PK 1993 Mullerian inhibiting substance: a gonadal hormone with multiple functions. Endocr Rev 14:152–164

    Lee MM 2000 MIS actions in the developing testis. In: Goldberg E, ed. The testis: from stem cell to sperm function. New York: Springer-Verlag; 30–42

    Lee MM, Seah CC, Masiakos PT, Sottas CM, Preffer FI, Donahoe PK, Maclaughlin DT, Hardy MP 1999 Mullerian-inhibiting substance type II receptor expression and function in purified rat Leydig cells. Endocrinology 140:2819–2827

    Racine C, Rey R, Forest MG, Louis F, Ferre A, Huhtaniemi I, Josso N, di Clemente N 1998 Receptors for anti-Mullerian hormone on Leydig cells are responsible for its effects on steroidogenesis and cell differentiation. Proc Natl Acad Sci USA 95:594–599

    Sriraman V, Niu E, Matias JR, Donahoe PK, MacLaughlin DT, Hardy MP, Lee MM 2001 Mullerian inhibiting substance inhibits testosterone synthesis in adult rats. J Androl 22:750–758

    Teixeira J, Fynn-Thompson E, Payne AH, Donahoe PK 1999 Mullerian-inhibiting substance regulates androgen synthesis at the transcriptional level. Endocrinology 140:4732–4738

    Rouiller-Fabre V, Carmona S, Abou Merhi R, Cate R, Habert R, Vigier B 1998 Effect of anti-Mullerian hormone on Sertoli and Leydig cell functions in fetal and immature rats. Endocrinology 139:1213–1220

    Salva A, Hardy MP, Wu XF, Sottas CM, Maclaughlin DT, Donahoe PK, Lee MM 2004 Mullerian-inhibiting substance inhibits rat Leydig cell regeneration after ethylene dimethanesulphonate (EDS) ablation. Biol Reprod 70:600–607

    Lyet L, Louis F, Forest MG, Josso N, Behringer RR, Vigier B 1995 Ontogeny of reproductive abnormalities induced by deregulation of anti-Mullerian hormone expression in transgenic mice. Biol Reprod 52:444–454

    Behringer RR, Finegold MJ, Cate RL 1994 Mullerian-inhibiting substance function during mammalian sexual development. Cell 79:415–425

    Mishina Y, Ray R, Finegold MJ, Matzuk MM, Josso N, Cate R, Behringer RR 1996 Genetic analysis of the Mullerian-inhibiting substance signal transduction pathway in mammalian sexual differentiation. Genes Dev 10:2577–2587

    Bédécarrats GY, O’Neill FH, Errol RN, Ursula BK, Teixeira J 2003 Regulation of gonadotropin gene expression by Müllerian inhibiting substance. Proc Natl Acad Sci USA 100:9348–9353

    Trbovich AM, Sluss PM, Laurich VM, O’Neill FH, MacLaughlin DT, Donahoe PK, Teixeira J 2001 Mullerian inhibiting substance lowers testosterone in luteinizing hormone-stimulated rodents. Proc Natl Acad Sci USA 98:3393–3397

    Klinefelter GR, Hall PF, Ewing LL 1987 Effect of luteinizing hormone deprivation in situ on steroidogenesis of rat Leydig cells purified by a multistep procedure. Biol Reprod 36:769–783

    Klinefelter GR, Kelce WR, Hardy MP 1993 Isolation and culture of Leydig cells from adult rats. Methods Toxicol 3:166–181

    Salva A, Klinefelter GR, Hardy MP 2001 Purification of rat Leydig cells: increased yields after unit-gravity sedimentation of collagenase-dispersed interstitial cells. J Androl 22:665–671

    Payne AH, Downing JR, Wong KL 1980 Luteinizing hormone receptors and testosterone synthesis in two distinct populations of Leydig cells. Endocrinology 106:1424–1429

    Cochran R, Ewing L, Niswender G 1981 Serum levels of follicle stimulating hormone, luteinizing hormone, prolactin, testosterone, 5-dihydrotestosterone, 5-androstane-3, 17?-diol, 5-androstane-3?, 17?-diol, and 17?-estradiol from male beagles with spontaneous or induced benign prostatic hyperplasia. Invest Urol 19:142–147

    Cochran RC, Schuetz AW, Ewing LL 1979 Age-related changes in conversion of 5-androstan-17?-ol-3-one to 5-androstane-3,17?-diol and 5-androstane-3?,17?-diol by rat testicular cells in vitro. J Reprod Fertil 57:143–147

    Coffey JC, French FS, Nayfeh SN 1971 Metabolism of progesterone by rat testicular homogenates. IV. Further studies of testosterone formation in immature testis in vitro. Endocrinology 89:865–872

    Zamecnik J, Barbe G, Moger WH, Armstrong DT 1977 Radioimmunoassays for androsterone, 5-androstane-3, 17?-diol and 5-androstane-3?,17?-diol. Steroids 30:679–689

    Weinstein A, Lindner HR, Friedlander A, Bauminger S 1972 Antigenic complexes of steroid hormones formed by coupling to protein through position 7: preparation from 4-3-oxosteroids and characterization of antibodies to testosterone and androstenedione. Steroids 20:789–812

    Siegel S 1956 Nonparametric statistics for the behavioral sciences. New York: McGraw-Hill

    McLean R, Sanders WL, Stroup WW 1991 A unified approach to mixed linear models. Am Statistician 45:54–64

    Searle S, Speed TP, Milliken GA 1980 Population marginal means in the linear model: an alternative to least squares means. Am Statistician 34:216–221

    Snedecor G, Cochran WG 1989 Statistical methods. Ames, IA: Iowa State University Press

    Littell R, Milliken GA, Stroup WW, Wolfinger RD 1996 The SAS system for mixed models. Cary, NC: SAS Institute

    Ge RS, Shan LX, Hardy MP 1996 Pubertal development of Leydig cells. In: Payne AH, Hardy MP, Russell LD, eds. The Leydig cell. Vienna, IL: Cache River Press; 159–74

    Benton L, Shan LX, Hardy MP 1995 Differentiation of adult Leydig cells. J Steroid Biochem Mol Biol 53:61–68

    Ewing LL, Keeney DS 1993 Leydig cells structure and function. In: Desjardins C, Ewing LL, eds. Cell and molecular biology of the testis. New York: Oxford University Press; 137–165

    Ge R, Hardy MP 1998 Variation in the end products of androgen biosynthesis and metabolism during postnatal differentiation of rat Leydig cells. Endocrinology 139:3787–3795

    Hardy MP, Zirkin BR, Ewing LL 1989 Kinetic studies on the development of the adult population of Leydig cells in testes of the pubertal rat. Endocrinology 124:762–770

    Murono EP, Washburn AL 1989 5-Reductase activity regulates testosterone accumulation in two bands of immature cultured Leydig cells isolated on Percoll density gradients. Acta Endocrinol (Copenh) 121:538–544

    Viger RS, Robaire B 1995 Steady state steroid 5-reductase messenger ribonucleic acid levels and immunocytochemical localization of the type 1 protein in the rat testis during postnatal development. Endocrinology 136:5409–5415

    Shan LX, Hardy MP 1992 Developmental changes in levels of luteinizing hormone receptor and androgen receptor in rat Leydig cells. Endocrinology 131:1107–1114

    Laurich VM, Trbovich AM, O’Neill FH, Houk CP, Sluss PM, Payne AH, Donahoe PK, Teixeira J 2002 Mullerian inhibiting substance blocks the protein kinase A-induced expression of cytochrome p450 17-hydroxylase/C(17–20) lyase mRNA in a mouse Leydig cell line independent of cAMP responsive element binding protein phosphorylation. Endocrinology 143:3351–3360

    Lee MM, Donahoe PK, Hasegawa T, Silverman B, Crist G, Best S, Noto R, Hasegawa Y, MacLaughlin DT 1996 Mullerian inhibiting substance (MIS) in humans: normal levels from infancy to adulthood. J Clin Endocrinol Metab 81:571–576

    Rey R, Lordereau-Richard I, Carel JC, Barbet P, Cate RL, Roger M, Chaussain JL, Josso N 1993 Anti-Mullerian hormone and testosterone serum levels are inversely related during normal and precocious pubertal development. J Clin Endocrinol Metab 77:1220–1226

    Tran D, Meusy DN, Josso N 1981 Waning of anti-Mullerian activity: an early sign of Sertoli cell maturation in the developing pig. Biol Reprod 24:923–931

    Baarends WM, Hoogerbrugge JW, Post M, Visser JA, De Rooij DG, Parvinen M, Themmen AP, Grootegoed JA 1995 Anti-Mullerian hormone and anti-Mullerian hormone type II receptor messenger ribonucleic acid expression during postnatal testis development and in the adult testis of the rat. Endocrinology 136:5614–5622(Xiufeng Wu, Ramamani Arum)