Thyroid Hormone and Follicle-Stimulating Hormone Regulate ...

0013-7227/98/$03.00/0 Endocrinology Copyright © 1998 by The Endocrine Society

Vol. 139, No. 11 Printed in U.S.A.

Thyroid Hormone and Follicle-Stimulating Hormone Regulate Müllerian-Inhibiting Substance Messenger Ribonucleic Acid Expression in Cultured Neonatal Rat Sertoli Cells* NIROMI K. ARAMBEPOLA, DAVID BUNICK,

PAUL S. COOKE

AND

Department of Veterinary Biosciences, University of Illinois, Urbana, Illinois 61802 ABSTRACT Thyroid hormone is a major regulator of Sertoli cell development, and the present study sought to determine the role of T3 in Müllerianinhibiting substance (MIS) messenger RNA (mRNA) expression. MIS, a Sertoli cell secretory protein that induces Mu¨llerian duct regression and also may be critical for germ and Leydig cell development, is maximal perinatally, then decreases as Sertoli cells mature. The fall in MIS mRNA expression is delayed by hypothyroidism in vivo, indicating that T3 could regulate MIS mRNA. However, understanding of the hormonal regulation of MIS has been limited due partly to the lack of a primary Sertoli cell culture system in which sustained expression of MIS or its mRNA can be obtained. We have developed a Sertoli cell culture system for examining hormonal regulation of MIS mRNA. We then tested the effects of T3 and/or FSH treatment on MIS mRNA levels in this new system. Initial studies indicated that MIS mRNA production by 5-day-old rat Sertoli cells was minimal in vitro. Therefore, Sertoli cells from 2-day-old rats were cultured for 2 or 4 days. After 2 days in vitro, steady state MIS mRNA levels were decreased to 36% of the levels seen in freshly isolated Sertoli cells from

M

ÜLLERIAN-inhibiting substance (MIS), also called anti-Müllerian hormone, is a member of the transforming growth factor-b family of cytokines that includes transforming growth factors-b, activins, inhibins, and the bone morphogenetic proteins. MIS induces regression of the Mu¨llerian ducts in the developing male fetus (reviewed in Refs. 1 and 2), but MIS has important effects within the testis as well. The effects of MIS on steroidogenesis by developing Sertoli cells have been reported (3). MIS may also have a role in the maturation of gonocytes to type A spermatogonia in the neonatal mouse testis, presumably as a result of effects on Sertoli cell activity (4). A possible role for MIS in spermatogenesis has been suggested by the finding that the level of MIS receptor expression shows dramatic variation in the seminiferous tubule during different stages of the spermatogenic cycle (5). MIS is involved in Leydig cell development, because male MIS knockout mice exhibit pronounced Leydig Received January 30, 1998. Address all correspondence and requests for reprints to: Dr. Paul Cooke, Department of Veterinary Biosciences, University of Illinois, 2001 South Lincoln Avenue, Urbana, Illinois 61802. E-mail: p-cooke@ uiuc.edu. * This work was supported by NIH Grant HD-29376 (to P.S.C.). Portions of this work were presented in abstract form at the 23rd Annual Meeting of the American Society of Andrology, Long Beach, CA, March 1998.

2-day-old rats. However, by day 4 of culture, steady state MIS mRNA production had recovered to 67% of that seen in freshly isolated 2-day-old Sertoli cells, which closely paralleled the decrease seen in MIS production in vivo from days 2– 6. MIS mRNA levels were decreased 53%, 64%, and 86% in cultures treated with 0.01, 0.1, and 1.0 nM T3 (P , 0.05), respectively. This decrease in Sertoli cell MIS mRNA did not reflect a nonspecific effect on cell viability and/or activity, as shown by a dose-responsive increase in inhibin-a mRNA in these same cultures. FSH (2.5–100 ng/ml) also produced a dose-responsive decrease in MIS mRNA levels, and FSH and T3 together had an additive inhibitory effect on MIS mRNA levels, indicating that these hormones may act through distinct mechanisms. In summary, this is the first primary culture system in which sustained MIS mRNA production can be demonstrated, and it should prove useful for understanding the regulation of MIS in developing Sertoli cells. In addition, T3 and FSH are major regulators of the postnatal decrease in MIS production by the rat Sertoli cell, and these hormones may act through separate pathways. (Endocrinology 139: 4489 – 4495, 1998)

cell hyperplasia and show increased expression of some steroidogenic enzymes (6, 7). Conversely, mice overexpressing MIS show an impairment in the differentiation of Leydig cell precursors (7) and reduced adult Leydig cell numbers and volume (7, 8). MIS has also been shown to inhibit LH-stimulated testosterone production by fetal Leydig cells (3). In the male, MIS is produced only by Sertoli cells (5). In the testis, MIS receptor messenger RNA (mRNA) expression was originally reported to occur only in Sertoli cells (5, 9), but a recent report has indicated that Leydig cells also express MIS receptor mRNA (7). Thus, the testicular actions of MIS may result from both autocrine actions on the Sertoli cells as well as paracrine actions on adjacent Leydig cells. The expression of MIS protein and mRNA is high before birth in the rat, then drops sharply during the postnatal period (10, 11). The apparent importance of MIS for various aspects of testicular development makes it critical to understand the factors that regulate the decline in the production of this hormone during neonatal life. One potential endocrine regulator of MIS expression during Sertoli cell development is thyroid hormone. Extensive work in the past few years has indicated that thyroid hormones are major regulators of Sertoli cell development. Transient neonatal hypothyroidism in rats produces unprecedented increases of 80% and 140% in adult testis weight and sperm production, respectively (12, 13). This

4489

4490

THYROID HORMONE EFFECTS ON MIS

increase in testis size results primarily from increased Sertoli cell proliferation during neonatal and juvenile life (14) and the consequently larger populations of adult Sertoli and germ cells (14, 15). Sertoli cells express thyroid hormone receptor and its mRNA during the neonatal period (16 –18), and T3, the biologically active thyroid hormone, inhibits Sertoli cell proliferation in vitro (19). T3 stimulates maturation of Sertoli cells (20) and has been shown to increase overall protein synthesis and production of certain Sertoli cell proteins (21) and decrease aromatase activity (22, 23). T3 treatment of cultured neonatal Sertoli cells also increases their levels of mRNA for inhibin-a, a marker of Sertoli cell maturation (19).1 Knowledge of the factors that regulate the postnatal decline in MIS is limited, due in large part to the inability of primary cell cultures to maintain MIS production in vitro (reviewed in Ref. 1). The aim of the present study was to develop a Sertoli cell culture system for examining hormonal regulation of MIS mRNA, then to test the effects of T3 and FSH, alone and in combination, on MIS mRNA production. Materials and Methods Animal purchase, care, and breeding were described previously (12). All experiments described here involving animals were approved by the laboratory animal care advisory committee of the University of Illinois and were conducted in accordance with the Guiding Principles for the Care and Use of Research Animals.

Culture of Sertoli cells from 2- and 5-day-old rats Sertoli cells from 2- and 5-day-old Sprague-Dawley rats (day of birth 5 day 0) were isolated using a sequential enzymatic procedure that has been previously described (24, 25) and used extensively in our laboratory to examine the effects of T3 on neonatal Sertoli cells (19, 26). Briefly, for each culture, pools of Sertoli cells were obtained from 8 –10 males from 2 litters. Sertoli cells were grown in 24-well plates coated with Matrigel (Collaborative Research, Waltham, MA) diluted 1:5 with HBSS. Cells were plated at a density of 4 3 105 cells/well. The nutritive medium was DMEM supplemented with sodium pyruvate (1 mm), nonessential amino acids (0.1 mm), and an antimicrobial solution (24, 25). Cells were normally grown for 4 days in a humidified atmosphere of 95% air-5% CO2 at 34 C. Medium was changed every 24 h. Some cultures received T3 (Sigma Chemical Co., Inc., St. Louis, MO) at 0.01–1 nm or FSH (ovine FSH-18, USDA Animal Hormone Program) at 2.5–100 ng/ml. To determine possible additive effects of these hormones, some cultures received both T3 (0.1 nm) and FSH (100 ng/ml). T3 was dissolved in 0.025 n NaOH, then diluted with physiological saline to make a stock solution; FSH was dissolved in physiological saline. Other cultures were grown without hormonal supplementation (controls) and received only vehicle. In all cases, T3 was added at the initiation of the culture, whereas FSH was added at the beginning of the third day, and fresh hormone was added when medium was changed. After 4 days in vitro, some wells derived from the testes of 2-day-old rats were used to determine the purity of the cultured cells, as described below; the remainder were used to obtain RNA for Northern analysis.

Viability and purity of cultured cells For all cell cultures described, the percentage of viable cells was determined by trypan blue exclusion before plating. At the end of the 1 We inadvertently reversed the identification of inhibin-a and inhibin-bB mRNA in our original paper on this topic (18). Therefore, the transcript identified as inhibin-bB mRNA in that and a subsequent paper (18, 19) is actually inhibin-a mRNA, whereas the transcript identified as inhibin-a mRNA in the former paper is actually inhibin-bB mRNA (see erratum, Biol Reprod 59:216, 1998).

Endo • 1998 Vol 139 • No 11

culture, wells derived from 2-day testes were stained for alkaline phosphatase and 3b-hydroxysteroid dehydrogenase (3bHSD) to determine peritubular and Leydig cell contamination, respectively, as previously described (26). Leydig cells differentiate during early postnatal life and may not stain as intensely for 3bHSD during this period as fully mature adult Leydig cells. In addition, recent results have indicated that MIS may suppress 3bHSD in Leydig cells (7). Therefore, both intensely and weakly stained cells were considered positive.

Northern analysis Northern analyses were performed as described previously (26). Briefly, total RNA was prepared from all cell cultures using the RNeasy Mini Kit (Qiagen, Chatsworth, CA). Total RNA was also isolated from 2-day-old testis and from freshly isolated preparations of Sertoli cells from 2-day-old testis, which served as positive controls for the MIS and inhibin-a probes, and from adult spleen, which served as a negative control for these probes. Purified total RNA was dissolved in diethylpyrocarbonate-treated water. The purity and concentration of the RNA were determined by UV (260/280 nm) absorbance in a spectrophotometer. Equal amounts of total RNA (8 mg) from the various treatment groups were electrophoresed on a 1.5% agarose formaldehyde gel. Gels were blotted to nylon membrane, and the RNA was fixed onto the membrane by UV cross-linking. The following probes were used in this study: 1) rat MIS (11); 2) rat inhibin-a (27), and 3) human 28S ribosomal RNA (rRNA) (28). The complementary DNA (cDNA) inserts were isolated from the plasmid vector by restriction digestion and gel purification. The insert was labeled with [32P]deoxy-CTP using the Multiprime DNA labeling system (Amersham, Arlington Heights, IL) and used to probe the membrane. All hybridizations were carried out in QuikHyb (Stratagene, La Jolla, CA) according to the manufacturer’s recommendations at 68 C in a Robbins Scientific hybridization oven (Sunnyvale, CA). The hybridized membrane was washed, covered with plastic wrap, and exposed to Kodak X-Omat x-ray film (Eastman Kodak Co., Rochester, NY) with intensifying screens. After hybridization with the MIS cDNA probe, some membranes were stripped of probe by incubation in 50% formamide at 65 C for 1 h and then rehybridized with inhibin-a cDNA probe. For normalization of RNA load levels between lanes, the membrane was reprobed a final time with labeled 28S rRNA cDNA probe. The mRNA bands on the autoradiograms were scanned and quantitated using a computer-linked scanning laser densitometer and RFLPrint software (Pdi, Huntington Station, NY). Relative levels of mRNA transcripts were adjusted to compensate for differences in total RNA loaded per gel lane as determined by densitometry of 28S rRNA hybridization signals (18). All statistical analyses were performed using the SYSTAT statistical package (29). The optical densities of the bands in the various treatment groups were compared by two-way ANOVA, and differences between various treatment groups were compared using Tukey’s honest significant difference test. Differences were considered significant when P , 0.05.

Results Characterization of Sertoli cell cultures

Trypan blue exclusion experiments indicated that more than 95% of the isolated Sertoli cells were viable. The Sertoli cell cultures from the 2-day-old rats had levels of peritubular cell contamination (;5%) similar to those we have previously observed in Sertoli cells from 5-day-old rats grown under the same culture conditions (26). A small number of Leydig cells (,1% of the overall cell population) could be identified; although these cells stained relatively lightly for 3bHSD, possibly as a result of the suppressive effect of MIS on the expression of this steroidogenic enzyme (7), they could be clearly identified among the surrounding unstained Sertoli cells.

THYROID HORMONE EFFECTS ON MIS Expression of MIS mRNA in Sertoli cell cultures from 2and 5-day-old rats

4491

Preliminary experiments detected only a faint 1.8-kb band, previously shown to correspond to MIS mRNA (11), in 5-day Sertoli cells cultured for 4 days. T3 treatment of these cultures resulted in an undetectable MIS mRNA signal (data not shown). As expression of MIS and its mRNA is high in the rat until birth, then declines sharply during the neonatal period (10, 11), we reasoned that utilization of Sertoli cells from younger rats might result in greater MIS mRNA expression in vitro. MIS mRNA expression in freshly isolated Sertoli cells from 2-day rats and in Sertoli cells from 2-day-old rats that had been cultured for 2 and 4 days is shown in Fig. 1. High levels of a 1.8-kb band corresponding to MIS mRNA were detected in freshly isolated Sertoli cells from 2-day-old rats. The level of MIS mRNA in the cultured Sertoli cells after 2 days in vitro was only 36% of that measured in freshly isolated 2-day Sertoli cells, but rebounded to 67% of the level seen in freshly isolated 2-day Sertoli cells after 4 days in vitro. Additional work indicated that MIS mRNA declined between days 4 and 6 in culture (data not shown), so the 4-day culture period was used for subsequent experiments. Initial experiments to examine the effects of T3 on MIS mRNA levels used doses of 1, 10, and 100 nm derived from our earlier studies of the effects of T3 on Sertoli cell proliferation in vitro (19). Unexpectedly, all of these doses initially

produced similar large decreases in MIS mRNA expression (data not shown). Subsequent experiments, therefore, used a lower range of T3 doses. These lower doses produced a clear dose-response effect, with the 0.01, 0.1, and 1.0 nm T3 doses suppressing MIS mRNA by 53%, 64%, and 86%, respectively, compared with the control value (Fig. 2). FSH also produced a dose-responsive suppression of MIS mRNA expression by cultured Sertoli cells (Fig. 3). However, the highest FSH dose (100 ng/ml), previously shown to produce maximal effects on Sertoli cell proliferation and inhibin-a mRNA levels (19), produced a decrease in MIS mRNA expression of only 43%, compared with the 86% decrease observed with T3 doses of 1 nm or higher. To examine the potential additive effects of T3 and FSH, Sertoli cells were cultured with T3 (0.1 nm) and a maximally effective dose of FSH 100 ng/ml (Fig. 4). T3 was given at a 0.1-nm dose in this experiment because earlier studies indicated that the maximally effective T3 dose (1 nm) produced 86% suppression of MIS mRNA levels, and the intensity of the Northern bands in this treatment group were near the limit of detectability. Therefore, any additive effects of FSH with this high dose of T3 would have been obscured by difficulties in attempting to measure the MIS mRNA signal in this low range, so a lower 0.1 nm T3 dose was chosen for the studies involving treatment with both T3 and FSH. In combination, T3 and FSH treatment resulted in an inhibition

FIG. 1. MIS mRNA expression in freshly isolated and cultured Sertoli cells. a, Northern blot detection of steady state MIS mRNA in freshly isolated Sertoli cells from 2-day-old rats (control) and Sertoli cells from 2-day-old rats cultured for 2 or 4 days. A 1.8-kb transcript corresponding to MIS was seen in all samples. b, Normalized densitometric data of MIS mRNA expression. Signals were normalized against a 28S rRNA cDNA probe to compensate for interlane differences in RNA loading. All values are shown as a percentage of the values for the freshly isolated Sertoli cells, which served as the control. The data are presented as the mean 6 SEM of values from four separate experiments. All values shown were significantly different from each other (P , 0.05).

FIG. 2. Dose-response of T3 on steady state MIS mRNA expression in cultured Sertoli cells. a, Northern blot detection of steady state mRNA in 2-day-old testis, spleen, and Sertoli cells from 2-day-old rats cultured for 4 days without hormonal supplementation (control) or with increasing concentrations of T3 (0.01–1 nM). MIS mRNA was detected in 2-day testis, a positive control to ensure the assay conditions would detect the 1.8-kb MIS transcript. MIS mRNA expression was not detected in spleen, a negative control. T3 treatment produced a doseresponsive decrease in steady state MIS mRNA levels. b, Normalized densitometric data of MIS mRNA expression. As in Fig. 1, signals were normalized against a 28S rRNA cDNA probe. All values are shown as a percentage of the control values. The data are presented as the mean 6 SEM of values from four separate experiments. All values shown were significantly different from each other (P , 0.05).

4492


FIG. 3. Dose-responsive effects of FSH on steady state MIS mRNA expression in cultured Sertoli cells. a, Northern blot detection of steady state mRNA in testis, spleen, and Sertoli cells from 2-day-old rats cultured for 4 days without hormonal supplementation (control) or with increasing concentrations of FSH (2.5–100 ng/ml). b, Normalized densitometric data of MIS mRNA expression. Signals were normalized against a 28S rRNA cDNA probe to compensate for RNA loading differences. Values are shown as a percentage of the control value. Data are presented as the mean 6 SEM of values from four separate experiments. All values shown were significantly different from each other (P , 0.05), except for the 2.5 and 10 ng/ml FSH groups.

of MIS mRNA (76%) that was significantly greater than that with either hormone alone (Fig. 4). Despite the significant decreases in MIS mRNA seen in Sertoli cell cultures treated with T3 and/or FSH, Northern blotting indicated that both T3 and FSH treatment resulted in an increase in steady state levels of inhibin-a mRNA (Fig. 5). Therefore, the decreased MIS mRNA levels in response to increasing T3 or FSH treatment do not reflect decreased viability or activity of the cultured Sertoli cells, but, rather, show an effect of the hormonal treatments. Discussion

Previous results have shown that neonatal hypothyroidism in the rat alters the pattern of testicular MIS mRNA expression by slowing its normal postnatal decline, suggesting that T3 could be involved in the regulation of MIS expression (18). However, hypothyroidism also produces decreases in hormones such as FSH and LH (30), which may be involved in the regulation of MIS mRNA. Therefore, it was unclear whether the altered testicular MIS mRNA expression following hypothyroidism in vivo reflects the effects of thyroid hormone deficiency on the Sertoli cells themselves or was secondary to other hormonal and/or metabolic changes induced by hypothyroidism.

Endo • 1998 Vol 139 • No 11

FIG. 4. Additive effects of T3 and FSH on steady state MIS mRNA expression in cultured Sertoli cells. a, Northern blot detection of steady state mRNA in testis, spleen, and Sertoli cells from 2-day-old rats cultured for 4 days without hormonal supplementation (control), with 0.1 nM T3, 100 ng/ml FSH, or with both T3 and FSH. b, Normalized densitometric data of MIS mRNA expression. Signals were normalized against a 28S rRNA cDNA probe. Values are shown as a percentage of the control value. Data are presented as the mean 6 SEM of values from three separate experiments. All values shown were significantly different from each other (P , 0.05). Due to the low SEM for the T3 group, the SE bar is not discernible in the graph.

The present results are the first to show that T3 can regulate MIS mRNA production in Sertoli cells. MIS mRNA levels in cultured Sertoli cells are very sensitive to T3, in that decreases of more than 50% are produced by as little as 0.01 nm T3, and maximal decreases are produced by 1 nm. The effect is also extremely robust, in that 1 nm T3 decreases MIS mRNA by over 85% compared with the control value. The genes for MIS have been cloned in the rat and mouse (31, 32) as well as in other species, such as cattle and humans (33). Although full promoter sequences are not available for all of these genes, analysis of the promoter sequences available in GenBank did not reveal the presence of any of the various thyroid hormone response elements that have been previously described (34). In addition, T3 induces changes associated with increasing maturation in the neonatal Sertoli cell, such as decreased proliferation, altered production of various basement membrane components, increased production of secretory proteins typical of the mature Sertoli cell, and changes in the production of steroid hormones and their receptors (14, 19, 20, 22, 23, 26, 35–37). The lack of known thyroid hormone response elements in the available MIS promoter sequences in conjunction with the known stimulation of Sertoli cell development by T3 indicate that it is likely that the decreases in steady state MIS mRNA levels in response to T3 reflect general stimulatory effects of thyroid hormone on Sertoli cell maturation rather than direct effects on MIS mRNA production.


FIG. 5. Effects of T3 and FSH on steady state inhibin-a mRNA levels in cultured Sertoli cells. The membranes shown in Figs. 2a and 3a were stripped and reprobed for inhibin-a mRNA. Treatment with increasing T3 concentrations (top) resulted in a dose-dependent increase in inhibin-a mRNA levels; after normalization with a 28S rRNA cDNA probe, the steady state level of inhibin-a mRNA for the 1-nM dose of T3 was approximately 100% greater than the control value. Similarly, FSH (bottom) increased inhibin-a mRNA levels in a dose-responsive manner, with the highest FSH dose (100 ng/ml) producing a normalized densitometric inhibin-a mRNA signal that was 66% greater than the control value. This effect was seen in all replicates of this experiment.

Thyroid hormone levels are low in the rat before and at birth, then rise sharply (4- to 5-fold) during the immediate postnatal period (30). Conversely, MIS production is high until birth, then falls dramatically shortly after birth (10). The temporal correlation between the increase in thyroid hormones and the decrease in MIS production by Sertoli cells during early neonatal life coupled with the ability of physiological levels of thyroid hormones to produce marked decreases in MIS mRNA production in cell culture suggest that T3 is a major regulator of the postnatal decrease in MIS production in vivo. T3 appears to act in concert with FSH, which also produces significant declines in MIS mRNA at physiological doses, to induce the decrease in MIS mRNA and protein production during the neonatal period. However, as low T3 doses produce significant declines in MIS mRNA levels, and physiological doses induce substantially greater suppression of MIS mRNA than even supraphysiological FSH doses, T3, rather than FSH, may be the main regulator of the progressive postnatal MIS decline. Understanding of the factors that regulate MIS production in the Sertoli cell has been limited due to the lack of cell culture systems in which MIS production is maintained (reviewed in Ref. 1). Dutertre et al. (38) recently used targeted oncogenesis to develop a transformed Sertoli cell line that stably expresses MIS and its receptor, but previous attempts to examine hormonal regulation of MIS by primary Sertoli cell cultures were hampered by rapid declines in MIS mRNA and protein when the cells were placed in vitro. Vigier et al. (39), using Sertoli cells isolated from 15- to 25-day calves, reported that MIS in both the culture medium and the cells themselves fell approximately 95% between days 1 and 4 in vitro. Similarly, MIS production by cultured Sertoli cells taken from calves between birth and 1 week of age showed

4493

an approximately 95% decline in MIS production between days 1 and 3 of culture, and MIS production by these cells became undetectable by day 4 of culture (40). The one report looking at MIS mRNA production from human tissue in vitro reported similar large decreases. MIS mRNA levels in testis cells from 20-week-old human fetuses were decreased almost to the limit of detectability after 3 days of culture compared with the relatively high expression in intact human fetal testis (41). In contrast, although MIS mRNA levels in 2-day-old Sertoli cells following 2 days of culture are only one third of those seen in freshly isolated Sertoli cells from 2-day-old rats, between days 2 and 4 MIS mRNA levels recover, and by 4 days of culture they are approximately two thirds of the level seen in the freshly isolated Sertoli cells from 2-day-old rats. This 33% fall over the 4-day culture period is very similar in magnitude to the normal fall in MIS expression that occurs between days 2 and 6 in neonatal rats (10). MIS mRNA expression in our cultured cells is therefore present at levels similar to what would be seen in Sertoli cells at a similar age, and this MIS mRNA expression is also relatively stable throughout the culture period, as opposed to the precipitous and progressive declines in either MIS protein or mRNA reported in other culture systems (39 – 41). Thus, this is the first primary culture system in which sustained MIS mRNA production can be demonstrated, and it should prove useful for studying the regulation of MIS in developing Sertoli cells. Our cultures were grown on Matrigel, a reconstituted basement membrane derived from the EHS tumor cell line, whereas Sertoli cells in the earlier reports were grown on plastic (39 – 41). Sertoli cells grown on Matrigel are morphologically and functionally more similar to these cells in vivo than those grown on plastic (42), which become flattened and show pronounced changes in growth factor responsiveness and vectorial protein production. Thus, the Matrigel used in these cultures may be a critical factor in the maintenance of MIS mRNA production in our culture system. Previous work in vivo has indicated that FSH treatment of newborn rat pups decreased MIS bioactivity compared with that in testes that did not receive the hormonal treatment, indicating that FSH may normally inhibit MIS expression (43). This conclusion has been confirmed by Kuroda et al. (44), who reported that FSH injection into neonatal rats reduced both MIS protein and mRNA. Similar results were later reported for rat fetuses injected with FSH during the last 2 days of gestation (45). In contrast, the previous studies with cultured neonatal calf Sertoli cells (39, 40) or human fetal testis cells (41) did not show an effect of FSH on MIS protein or mRNA production. The present results are therefore the first demonstration that the inhibitory effects of FSH on MIS seen in vivo can also be obtained in cell culture. Furthermore, the magnitude of the decrease in MIS immunostaining obtained by Kuroda et al. (44) in neonatal rats following 4 days of FSH injection (40%) is in good agreement with the 35–55% decreases in MIS mRNA levels reported here for neonatal rat Sertoli cells exposed to various doses of FSH during a 4-day culture. More significantly, our present results indicate that FSH treatment of Sertoli cells results in decreased MIS mRNA levels, and therefore, the FSH effects seen in previous in vivo studies at

4494

Endo • 1998 Vol 139 • No 11


least predominately reflect FSH actions on Sertoli cells, rather than a secondary effect(s) of FSH at a site other than the Sertoli cell, which then results in decreased MIS expression. The reasons for the discrepancy between our results and those of the previous in vitro studies may involve the difficulties in assessing the effects of a hormonal treatment such as FSH when the overall production of MIS protein or mRNA is declining sharply, which would have made it difficult to resolve an effect of FSH in the earlier studies. The maintenance of a more normal morphology in the Sertoli cells cultured on Matrigel may also be a factor in allowing them to respond to FSH in vitro as they do in vivo, as discussed above. The ability of T3 plus FSH to cause greater suppression of MIS mRNA than a maximal dose of FSH alone suggests that these hormones may be exerting their effects through different pathways. The ability of the T3 and FSH combination to produce greater suppression than only 0.1 nm T3 is clearly consistent with this conclusion. However, as the administered dose of T3 used was less than the maximally effective T3 dose, it could not be concluded from comparing the T3 and the T3 plus FSH groups alone that FSH was acting through a separate pathway than T3, as a T3 dose higher than the 0.1-nm dose used in this experiment would itself produce greater suppression. Similar results showing the additive effects of T3 and FSH on other Sertoli cell parameters, such as production of inhibin-a mRNA (19) and androgen receptor mRNA (26), have been observed, although T3 and FSH have opposite effects on Sertoli cell proliferation (19). The developmental changes seen in the expression of MIS closely parallel those reported for MIS mRNA (10, 11), and FSH treatment of rats decreases both MIS and its mRNA in vivo (44). These results suggest that the normal developmental changes in MIS production as well as hormonal effects on this process primarily reflect transcriptional changes. Thus, the effects of T3 and FSH on MIS mRNA levels reported here are presumably accompanied by similar changes in MIS production, although that cannot be established from the present data. In conclusion, our understanding of the role of T3 in Sertoli cell development is still incomplete, but T3 appears to exert pleiotropic effects on Sertoli cell maturation. T3 induces production of secretory proteins, but also has effects on estrogen receptor expression and the production of aromatase, the enzyme that converts androgens to estrogens (22, 23, 35). T3 also stimulates the production of androgen receptor and its mRNA in vitro (26, 35), another process believed to be important for the normal maturation of Sertoli cells. The present results indicate that T3 also regulates MIS, most likely by an indirect mechanism related to its overall stimulatory effects on Sertoli cell maturation. Therefore, in addition to the effects of T3 on critical maturational processes, T3 appears to induce complex changes in the production and/or receptor expression for other hormones that themselves may be important modulators of Sertoli cell activity. Acknowledgments The authors thank the National Hormone and Pituitary Program and the USDA Animal Hormone Program for providing the FSH used in these studies. The authors thank Drs. P. Donahoe (Harvard Medical School) and K. Mayo (Northwestern University) for the MIS and in-

hibin-a cDNA probes, respectively, and David Buchanan for proofreading the manuscript.

References 1. Lee MM, Donahoe PK 1993 Mu¨llerian inhibiting substance: a gonadal hormone with multiple functions. Endocr Rev 14:152–164 2. Josso N 1986 Anti-mullerian hormone: new perspectives for a sexist molecule. Endocr Rev 7:421– 433 3. 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 immmature rats. Endocrinology 139:1213–1220 4. Zhou B, Satts LM, Hutson JM 1993 Germ cell development in neonatal mouse testes in vitro requires mullerian inhibiting substance. J Urol 150:613– 616 5. Baarends WM, Hoogerbrugge JW, Post M, Visser JA, de Rooij DG, Parvinen M, Themmen APN, Grootegoed JA 1995 Anti-Mullerian hormone and antiMullerian hormone type II receptor messenger ribonucleic acid expression during postnatal testis development and in the adult testis of the rat. Endocrinology 136:5614 –5622 6. Behringer RR, Finegold MJ, Cate RL 1994 Mullerian-inhibiting substance function during mammalian sexual differentiation. Cell 79:415– 425 7. Racine C, Rey R, Forest M, Louis F, Ferre A, Huhtaniemi I, Josso N, di Climente 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 8. Behringer R, Cate R, Frelick G, Palmiter R, Brinster R 1990 Abnormal sexual development in transgenic mice chronically expressing Mullerian-inhibiting substance. Nature 345:167–170 9. Baarends WM, Helmond MJL, Post M, van der Schoot P, Hoogerbrugge JW, de Winter JP, Uilenbroek J-ThJ, Karels, B, Wilming LG, Meijers JHC, Themmen APN, Grootegoed JA 1994 A novel member of the transmembrane serine/threonine kinase receptor family is specifically expressed in the gonads and in the mesenchymal cells adjacent to the mullerian duct. Development 120:189 –197 10. Ueno S, Takahashi M, Manganaro TF, Ragin RC, Donahoe PK 1989 Cellular localization of Mullerian inhibiting substance in the developing rat ovary. Endocrinology 124:1000 –1006 11. Hirobe S, He W-W, Lee MM, Donahoe PK 1992 Mullerian inhibiting substance messenger ribonucleic acid expression in granulosa and Sertoli cells coincides with their mitotic activity. Endocrinology 131:854 – 862 12. Cooke PS, Meisami E 1991 Early postnatal hypothyroidism causes increased adult size of testis and other reproductive organs but does not increase testosterone levels. Endocrinology 129:237–243 13. Cooke PS, Hess RA, Porcelli J, Meisami E 1991 Increased sperm production in adult rats following transient neonatal hypothyroidism. Endocrinology 129:244 –248 14. van Haaster LH, de Jong FH, Docter R, de Rooij DG 1992 The effect of hypothyroidism on Sertoli cell proliferation and differentiation and hormone levels during testicular development in the rat. Endocrinology 131:1574 –1576 15. Hess RA, Cooke PS, Bunick D, Kirby JD 1993 Adult testicular enlargement induced by neonatal hypothyroidism is accompanied by increased Sertoli and germ cell numbers. Endocrinology 132:2607–2613 16. Jannini EA, Olivieri M, Francavilla S, Gulino A, Ziparo E, D’Armiento M 1990 Ontogenesis of nuclear 3,5,39-triiodothyronine receptor in the rat testis. Endocrinology 126:2521–2526 17. Jannini EA, Dolci S, Ulisse S, Nikodem VM 1994 Developmental regulation of the thyroid hormone receptor a1 mRNA expression in rat testis. Mol Endocrinol 8:89 –96 18. Bunick D, Kirby JD, Hess RA, Cooke PS 1994 Developmental expression of testis mRNAs in the rat following propylthiouracil-induced neonatal hypothyroidism. Biol Reprod 51:706 –713 19. Cooke PS, Zhao Y-D, Bunick D 1994 Triiodothyronine inhibits proliferation and stimulates differentiation of cultured neonatal Sertoli cells: possible mechanism for increased adult testis weight and sperm production induced by neonatal goitrogen treatment. Biol Reprod 51:1000 –1005 20. van Haaster LH, de Jong FH, Docter R, de Rooij DG 1993 High neonatal triiodothyronine levels reduce the period of Sertoli cell proliferation and accelerate tubular lumen formation in the rat testis, and increase serum inhibin levels. Endocrinology 133:755–760 21. Jannini EA, Ulisse S, D’Armiento M 1995 Thyroid hormone and male gonadal function. Endocr Rev 16:443– 459 22. Panno ML, Beraldi E, Pezzi V, Salerno M, De Luca G, Lanzino M, Lepera M, Sisci D, Prati M, Palmero S, Bolla E, Fugassa E, Ando S 1994 Influence of thyroid hormone on androgen metabolism in peripuberal rat Sertoli cells. J Endocrinol 140:349 –355 23. Ulisse S, Jannini EA, Carosa E, Piersanti D, Graziano FM, D’Armiento M 1994 Inhibition of aromatase activity in rat Sertoli cells by thyroid hormone. J Endocrinol 140:431– 436 24. Rong-Xi D, Djakiew D, Dym M 1987 Endocytic activity of Sertoli cells grown in bicameral culture chambers. Anat Rec 218:306 –312 25. Orth JM, Boehm R 1990 Functional coupling of neonatal rat Sertoli cells and gonocytes in coculture. Endocrinology 127:2812–2820

THYROID HORMONE EFFECTS ON MIS 26. Arambepola NK, Bunick D, Cooke PS 1998 Thyroid hormone effects on androgen receptor (AR) messenger RNA expression in rat Sertoli and peritubular cells. J Endocrinol 156:43–50 27. Woodruff TK, Meunier H, Jones PBC, Hsueh AJW, Mayo KE 1987 Rat inhibin: molecular cloning of a and b subunit complementary deoxyribonucleic acids and expression in the ovary. Mol Endocrinol 1:561–568 28. Erickson JM, Schimickel RD 1985 A molecular basis for discrete size variation in human ribosomal DNA. Am J Hum Genet 37:311–325 29. Wilkinson L 1990 SYSTAT: The System for Statistics. Systat, Evanston 30. Kirby JD, Jetton AE, Cooke PS, Hess RA, Bunick D, Ackland JF, Turek FW Schwartz NB 1992 Developmental hormonal profiles accompanying the neonatal hypothyroidism induced increases in adult testis size and sperm production in the rat. Endocrinology 131:559 –565 pins and steroids. Endocrinology 129:2985–2992 31. Shen WH, Moore CC, Ikeda Y, Parker KL, Ingraham HA 1994 Nuclear receptor steroidogenic factor 1 regulates the Mullerian inhibiting substance gene: a link to the sex determination cascade. Cell 77:651– 661 32. Haqq C, Lee MM, Tizard R, Wysk M, DeMarinis J, Donahoe PK, Cate RL 1992 Isolation of the rat gene for Mullerian inhibiting substance. Genomics 12:665– 669 33. Cate RL, Mattolino RJ, Hession C, Tizard R, Farber NM, Cheung A, Ninfa EG, Frey AZ, Gash DJ, Chow EP, Risher RA, Bertunis JM, MacLaughlin DT, Donahoe PK 1986 Isolation of the bovine and human genes for Mullerian inhibiting substance and expression of human gene in animal cells. Cell 45:685– 698 34. Williams GR 1994 Thyroid Hormone Regulation of Gene Expression. Landes, Austin, pp 1–143 35. Panno ML, Sisci D, Salerno M, Lanzino M, Pezzi V, Morrone EG, Mauro L, Palmero S, Fugasso E, Ando S 1996 Thyroid hormone modulates androgen and oestrogen receptor content in the Sertoli cells of peripubertal rats. J Endocrinol 148:43–50

4495

36. Ulisse S, Rucci N, Piersanti D, Carosa E, Graziano FM, Pavan A, Ceddia P, Arizzi M, Muzi P, Cironi L, Gnessi L, D’Armiento M, Jannini EA 1998 Regulation by thyroid hormone of the expression of basement membrane components in rat prepubertal Sertoli cells. Endocrinology 139:741–747 37. Loveland K, Schlatt S, Sasaki T, Chu M-L, Timpl R, Dziadek M 1998 Developmental changes in the basement membrane of the normal and hypothyroid postnatal rat testis: segmental localization of fibulin-2 and fibronectin. Biol Reprod 58:1123–1130 38. Dutertre M, Rey R, Porteu A, Jonno N, Picard JY 1997 A mouse Sertoli cell line expressing anti-Mullerian hormone and its type II receptor. Mol Cell Endocrinol 1997 136:57– 65 39. Vigier B, Picard P-Y, Campargue J, Forest MG, Heyman Y, Josso N 1985 Secretion of anti-Mullerian hormone by immature bovine Sertoli cells in primary culture, studies by a competition-type radioimmunoassay: lack of modulation by either FSH or testosterone. Mol Cell Endocrinol 43:141–150 40. LaQuaglia M, Shima H, Hudson P, Takahashi M, Donahoe PK 1986 Sertoli cell production of Mullerian inhibiting substance in vitro. J Urol 136:219 –224 41. Voutilainen R, Miller WL 1987 Human Mullerian inhibitory factor messenger ribonucleic acid is hormonally regulated in the fetal testis and in adult granulosa cells. Mol Endocrinol 1:604 – 608 42. Hadley MA, Byers SW, Suarez-Quian CA, Kleinman HK, Dym M 1985 Extracellular matrix regulates Sertoli cell differentiation, testicular cord formation, and germ cell development in vitro. J Cell Biol 101:1511–1522 43. Bercu BB, Morikawa Y, Jackson IMD, Donahoe PK 1979 Inhibition of Mullerian inhibiting substance secretion by FSH. Pediatr Res 13:246 –249 44. Kuroda T, Lee MM, Haqq CM, Powell DM, Manganaro TF Donahoe PK 1990 Mullerian inhibiting substance ontogeny and its modulation by follicle-stimulating hormone in the rat testes. Endocrinology 127:1825–1832 45. Kuroda T, Lee MM, Hirobe S, Ragin RC, Donahoe PK 1991 Mullerian inhibiting substance production and cleavage is modulated by gonadotropins and steroids. Endocrinology 129:2985–2992