Endorphin suppresses FSH-stimulated proliferation of isolated neonatal sertoli cells by a pertussis toxin-sensitive mechanism.код для вставкиСкачать
THE ANATOMICAL RECORD 226:320-327 (1990) Endorphin Suppresses FSH-Stimulated Proliferation of Isolated Neonatal Sertoli Cells by a Pertussis Tox in- Sensit ive Mechanism JOANNE M. ORTH AND ROSEMARIE BOEHM Department of Anatomy, Temple University School of Medicine, Philadelphia, Pennsylvania 19140 ABSTRACT During perinatal development, when the size of the Sertoli cell population is determined, Leydig cells produce p-endorphin, a peptide which may interact with Sertoli cells to modify their FSH-responsiveness, as suggested by our previous work. The goal of the present study was first, to test directly the possibility that p-endorphin modifies the proliferative response of neonatal Sertoli cells to FSH, and second, to gain information on a mechanismb) involved in any observed effect. We treated isolated 6-day-old Sertoli cells with FSH or vehicle in vitro and measured their incorporation of exogenous, radiolabeled thymidine with quantitative autoradiography. After 2 days in culture with FSH, we detected a 10-fold increase in the rate of Sertoli cell proliferation. The level of cell division in these FSH-treated cultures was identical to that in other cultures exposed to cAMP under similar conditions. In addition, inclusion of P-endorphin 3 h r prior to FSH or cAMP decreased the effect of the hormone by 50% but left the cAMP response unchanged. Thus, p-endorphin acts on isolated, neonatal Sertoli cells at a point prior to intracellular production of cAMP to suppress their response to FSH. When other cultures were treated with pertussis toxin, a blocker of intracellular GTPbinding proteins such as Gi, before sequential addition of endorphin and FSH, the effect of 6-endorphin on FSH-responsiveness was abolished. Moreover, when other cultures were exposed to pertussis toxin in the absence of endorphin, followed by FSH, their response to the hormone was unchanged. Thus, P-endorphin apparently modifies the proliferative response of neonatal Sertoli cells to FSH via a mechanism involving one or more G proteins. These observations, along with our previous data showing enhanced Sertoli cell division in vivo in the presence of a n opiate blocker, point to the existence of endorphin-mediated communication between Leydig and Sertoli cells during perinatal development and provide new evidence suggesting that paracrine mechanisms modify Sertoli cell function during perinatal development, when the size of this population is established. The size of the Sertoli cell population in rats is of critical importance in maintaining normal output of sperm in adults, as evidenced by our findings in Sertoli cell-depleted rats (Orth et al., 1988). Following neonatal treatment with a n anti-mitotic drug, these rats possess lowered numbers of Sertoli cells at maturity. Moreover, although germ cells in neonates are nonproliferative and hence unaffected by the anti-mitotic drug, testes of these rats a s adults display a drop in numbers of spermatids that parallels the decrease in number of Sertoli cells. Hence, a n important quantitative relationship exists between Sertoli and germ cells in the adult testis. Data from our laboratory (Orth, 1982) and that of others (Steinberger and Steinberger, 1971) indicate that Sertoli cells of rats proliferate only during prenatal and early postnatal life, with cessation of division occurring between 10 and 21 days of age. Thus, the perinatal period of testicular development when these cells divide is critical in establishing a Sertoli cell population of a size adequate to insure produc0 1990 WILEY-LISS. INC tion of normal numbers of sperm at adulthood. As a result, factors that modify Sertoli cell proliferation during perinatal life are likely to be of great importance in insuring fertility of the adult. In previous studies both in vivo and in vitro, we have shown that FSH stimulates Sertoli cells of perinatal rats to divide via a CAMP-mediated process (Orth, 1984) and that the level of Sertoli cell proliferation can be modified during development by altering pituitarytesticular feedback mechanisms. For example, when neonates are hemi-castrated the rate of Sertoli cell division in the remaining testis is elevated in response to a rise in circulating FSH (Orth et al., 1984). Substan- Received March 8, 1989; accepted May 25, 1989. Data included in this report were presented at the joint meeting of the American Society for Cell Biology and the American Society for Biochemistry and Molecular Biology, San Francisco, CA, January 1989. ENDORPHIN, FSH, AND SERTOLI CELL PROLIFERATION tial evidence from other laboratories also indicates that other, intratesticular mechanisms exist, at least in adults, through which the function of testicular cells, including Sertoli cells, can be modified (Sharpe, 1984; Bardin e t al., 1988). One possible route of paracrine regulation of Sertoli cells involves Leydig cells, shown conclusively to be a n intratesticular source of POMCderived peptides such as 0-endorphin throughout fetal and postnatal life (Pintar et al., 1984; Shaha et al., 1984; Bardin e t al., 1987). Several lines of evidence suggest that the Sertoli cell may be a target of endorphin. These include our previous demonstration that interfering with the action of endogenous P-endorphin in testes of fetal or newborn rats, either in vivo or in organ culture, results in suppression of FSHstimulated proliferation of Sertoli cells (Orth, 1986). This observation, along with those pointing to the Leydig cell as a source of endorphin, suggests that endorphin-mediated interaction may occur between Leydig and Sertoli cells during development and that such interaction may modify the responsiveness of Sertoli cells to FSH. Thus, communication between the interstitial and seminiferous compartments of the maturing testis could play a n important role in determining the size of the Sertoli cell population, and therefore the eventual output of sperm, in adult rats. The aim of the present study was to probe further the relationship between FSH-stimulated division of Sertoli cells and p-endorphin. Sertoli cells were isolated from newborns and their rate of proliferation in vitro was studied in the presence or absence of FSH, with or without exogenously added endorphin. Since cells were isolated from neonates, prior to development of SertoliSertoli junctional complexes, it was first important to characterize the makeup of the cultures prior to their use in these studies in order to verify that Sertoli cells were indeed the major component. Therefore, incorporation of 3H-thymidine into cell nuclei and quantitative autoradiography were used as a measure of Sertoli cell proliferation under different conditions. In addition, to test the possible involvement of GTP-binding proteins in any observed effect of endorphin on FSHstimulated proliferation, pertussis toxin was included in some experiments. The resulting data, coupled with our previous findings from studies in vivo, strongly suggest t h a t p-endorphin of Leydig cell origin acts directly on neonatal Sertoli cells to suppress their proliferative response to FSH and that this effect involves interaction of endorphin with GTP-binding proteins in these cells. Thus, p-endorphin-mediated communication between Leydig and Sertoli cells apparently provides a paracrine route through which Leydig cells may modify the hormonal response of Sertoli cells during perinatal development. MATERIALS AND METHODS Sertoli Cell Cultures For each set of cultures, 20 male pups 6 days old (day of birth = day 1) were killed by decapitation; their testes were removed under aseptic conditions and decapsulated in sterile, ice-cold Dulbecco's Minimum Essential Medium (MEM; Gibco). To isolate Sertoli cells, the testicular tissue was minced and treated further according to the method of Rong-Xi et al. (19871, with 32 1 minor modifications. In brief, the minced tissue was incubated with gentle shaking for 20 min in 30 ml of MEM containing a mixture of 0.1% hyaluronidase (Sigma), 0.1% collagenase (Gibco), and DNAase (Sigma; 0.01 mg/30 ml) in MEM at 37"C, followed by two washes in fresh MEM. After the enzyme treatment and each wash, fragments of seminiferous cords were allowed to settle on ice a t unit gravity for 15 min. Following the last wash, the resulting fragments were incubated at 37°C for 30 min each in 20 ml collagenase (0.1% in MEM) + DNAase twice more, followed by a final rinse in fresh MEM. The resulting fragments were then dispersed into single cells by a brief treatment (approximately 2 min) in Ca-Mg-free Hank's buffer containing 0.05% trypsin (Gibco) and 6 x 10-4 M EDTA. After sedimenting at approximately 200g and washing with BSA-trypsin inhibitor (Sigma; 0.65% and 0.05%, respectively), the resulting cells were resuspended in a small volume of plating medium (see below). For each set of cultures, cell viability was evaluated by exposing a n aliquot of the single cell suspension to trypan blue and determining the yield of dyenegative cells in a hemocytometer. Cells were cultured a t 37°C in a 5% COz atmosphere in four-chamber tissue culture slides (Lab-Tek) previously coated with Matrigel (Collaborative Research); the latter was diluted 1 : l with MEM, applied as a thin coat to the slides and allowed to gel for 2-3 h r a t 32°C prior to use. Two million cells were plated in each chamber (1 x 104/mm2)in serum-free Eagle's D-val MEM (Gibco) supplemented with Na Pyruvate (1mM) and non-essential amino acids (0.1 mM), with fungizone (2.5 pg/ml) and penicillin (100 U/ml)-streptomycin (100 pg/ml) also included. The next morning, approximately 18 h r after plating, all chambers were rinsed in warm medium and fresh medium was added. To determine the proportion of Sertoli cells to other testicular cells (e.g., peritubular and Leydig cells) in these preparations, several approaches were applied to representative chambers. First, some cultures were evaluated morphologically by processing them for electron microscopy, a s described below. Second, the ability of cells in other chambers to respond to cAMP with a change in shape was determined by adding db cAMP (0.5 pM) or vehicle for 3-18 hr, followed by fixation and inspection of the cells with phase contrast microscopy. Finally, several chambers were evaluated with cytochemistry as previously described (Orth and Weisz, 1980) for the presence of cells displaying activity of either 3P-hydroxysteroid dehydrogenase, a Leydig cell marker, or alkaline phosphatase, a n indicator of peritubular cells (Chapin et al., 1987). Studies of Sertoli Cell Proliferation In Vitro To evaluate the effect of P-endorphin on Sertoli cell division following exposure of cells to FSH or CAMP, the following treatment grdups were established: first, some cultures received FSH (oFSH-17; 1 pg/ml) from either day 2 of the experiment onward or for only the final 24 h r of culture, with and without addition of endorphin (1 pg/ml) 3 h r before FSH. Other cultures received db cAMP (0.5 pM), again with and without prior addition of endorphin, either from day 2 onward or for only the last day of culture. To probe the mechanism involved in any observed 322 J.M. ORTH AND R. BOEHM effect of p-endorphin, additional chambers first received pertussis toxin (1 pg/ml), a blocker of GTPbinding proteins, including Gi (Murayama and Ui, 1983);either p-endorphin or vehicle was added to these chambers 3 h r later, followed by FSH, CAMP,or vehicle after another 3 h r period. Additional cultures received P-endorphin or pertussis toxin alone or in combination, but no FSH or CAMP.Controls received only vehicle a t appropriate times. All cultures were maintained for a total of 4 days, with the various agents present from the time of addition onward. For chambers that received hormone andlor peptide from days 2 through 4, a n agentb) was freshly added on day 3 when medium was changed. In all chambers, 3H-thymidine (New England Nuclear; 6.7 Ci/mmol; 1 pCi/ml final concentration) was included for the final 24 h r in vitro. At the end of the 4th day, cultures were fixed a s described below and subjected to autoradiography as detailed previously (Orth, 1982). In brief, slides were coated after fixation with undiluted Kodak NTB-3 emulsion and allowed to expose in darkness for 7 days. At the end of this period, the cultures were photographically developed, stained with methylene blue, and viewed with conventional light microscopy for quantitation. Proliferative nuclei were heavily labeled by 3H-thymidine and, after this length of photographic exposure, were overlain by silver grains and easily distinguished from those surrounding nuclei that had not incorporated label. Quantitation and Statistical Analysis These experiments were carried out on three separate occasions, with a different cell isolation providing the cultures each time. For each trial, cell proliferation was measured in all treatment groups a s follows: cultures were examined a t 100 x magnification; 2,0003,000 cell nuclei were chosen in each chamber in a non-random manner to insure against viewer bias and were scored a s labeled or unlabeled by 3H-thymidine. The data for each chamber were expressed as the percentage of cells counted that had incorporated label into their nuclei. Thus, three or four chambers were studied and a total of 6,000 or more cells were quantified for each treatment. The final data were expressed as the mean percentage of cells labeled ( & SEM) for the various groups. A one-way analysis of variance was then used to determine whether differences existed among the groups, and these differences were subsequently located with a Newman-Keuls test. Electron Microscopy Plastic chambers were removed from the culture slides and cells were processed in situ for electron microscopy a s follows. After fixing for 30 min a t 4°C in 2.5% glutaraldehyde-0.1M Na cacodylate, pH 7.4, containing 5% sucrose, the cultures were post-fixed in 1% osmium tetroxide reduced with 1.5%K ferrocyanide for 30 min and then mordanted in 1% tannic acid in O.1M Na cacodylate for 30 min at 4°C. Following subsequent dehydration through 100% ethanol, the cultures were infiltrated with Epon-Araldite. Beem capsules containing unpolymerized plastic were then inverted over the cells and the slides were placed overnight at 60°C. The next day, capsules containing polymerized plastic with embedded cells were removed from the slides by brief Fig. 1. An electron micrograph of cells isolated from 6-day-old pups, fixed the morning after plating, and sectioned parallel to the surface of the chamber. The great majority of cells in these cultures possessed irregular, indented nuclei with single nucleoli, elongated mitochondria, and occasional lipid droplets, characteristics consistent with those of Sertoli cells. x 6,630. immersion of the entire slide in liquid N P . Thin sections were cut parallel to the surface of the cultures with a Reichert Ultracut E microtome, post-stained in uranyl acetate and Reynold's lead citrate, and viewed and photographed with a Philips 300 electron microscope. RESULTS Characterization of Cultures Isolated cells adhered to and spread upon the underlying Matrigel substrate within 30-60 min after plating to form confluent cultures. Figure 1is a representative ultrastructural view of a culture that was rinsed and fixed the morning after plating; the plane of section was parallel to the surface of the chamber. The great majority of the cells in these cultures had a morphology consistent with that of Sertoli cells, with irregular, often indented nuclei containing some peripheral heterochromatin. Nucleoli, where present in the section, were single and displayed a tripartite morphology characteristic of Sertoli cells. In addition, mitochondria were typically elongated and cytoplasmic liquid droplets were occasionally seen. Peritubular cells, easily identifiable by their rough endoplasmic reticulum con- ENDORPHIN, FSH, AND SERTOLI C E L L PROLIFERATION 323 Fig. 2. Cultures incubated on the morning after isolation with either dibutyryl cAMP (b)or vehicle (a) and fixed 3 hr later. Nearly all cells responded to the presence of the cyclic nucleotide with a rapid and dramatic change in shape which was maintained as long as cAMP was present. x 400. taining copious flocculent material, were encountered on rare occasion. When similar cultures were exposed to dibutyrl cAMP and viewed with phase contrast microscopy, the vast majority of the cells became rounded centrally and displayed elongated peripheral processes, a s shown in Figure 2. This change in cell shape was apparent in most cells by 1 h r and was maintained so long as cAMP was present in the cultures. In addition, less than 5%of the cells in representative cultures displayed reaction product following cytochemical incubation for alkaline phosphatase, a n enzyme characteristic of myoid cells. Finally, to probe the possibility of contamination by Leydig cells, the activity of the steroidogenic enzyme 3p-hydroxysteroid dehydrogenase was also visualized with LM cytochemistry. Virtually no cells reactive for this enzyme were encountered in any of the chambers examined. FSH, p-Endorphin, and Sertoli Cell Proliferation In Vitro Figure 3 provides typical autoradiographs of cultures that were either untreated or exposed to FSH from day 2 of culture onward, with and without inclusion of pendorphin. Labeled thymidine was present during the final 24 hours of culture for all treatments. While few labeled nuclei were found in unstimulated controls (Fig. 3a), FSH caused a dramatic and obvious increase in the proportion of Sertoli cells t h a t incorporated 3H-thymidine into their nuclei (Fig. 3b). However, when p-endorphin was added 3 h r prior to the hormone, the proliferative response of the cells to FSH was apparently suppressed (Fig. 3c). This qualitative obser- vation was confirmed when dividing Sertoli cells were quantified in autoradiographs of similarly treated cultures; the final data from all groups studied, expressed as the mean of three trials for each treatment, are given in Figure 4. The level of proliferation in unstimulated control cultures was low, below 2% in all dishes, and unaffected by pertussis toxin alone or pertussis toxin plus endorphin. However, addition of either FSH or cAMP on day 2 resulted in a n approximately 10-fold increase (P<.Ol)in the percent of cells incorporating labeled thymidine. When p-endorphin was added to cultures prior to FSH, the response of the cells to the hormone was substantially diminished (P<.Ol),by approximately 50%. In addition, exposure of cells to pertussis toxin before addition of 6-endorphin completely abolished the ability of endorphin to suppress their proliferative response to FSH; the level of proliferation in FSH-stimulated cultures treated with pertussis toxin followed by endorphin was essentially identical to that in other cultures given FSH alone. Finally, when cells were treated with endorphin prior to addition of CAMP, the percent of cells labeled was not different than that measured in cultures exposed to cAMP alone from day 2 of the experiment onward. To gain information on the responsiveness of neonatal Sertoli cells cultured without constant exposure to hormone or cyclic nucleotide, other cultures were maintained in vitro for 3 days without any additions followed by inclusion of FSH or cAMP for the 4th day; 3H-thymidine was added to these cultures for the final 24 h r of culture, and the percentage of cells dividing was compared to t h a t in unstimulated controls. Under 324 J.M. ORTH AND R. BOEHM these conditions, FSH had no significant effect on the level of Sertoli cell proliferation (1.61% vs. 1.43%), while cAMP dramatically increased the percent of Sertoli cells labeled (23.1% vs. 1.43% for controls) to a level similar to that seen in the earlier incubations for which cAMP was present from day 2 onward. Moreover, in these incubations endorphin again had no effect on the proliferative response of Sertoli cells to the cyclic nucleotide. DISCUSSION Fig. 3. Low-power views of autoradiographs showing representative areas of cultures maintained for 4 days, with 3H-thymidine present for the last 24 hr. Individual nuclei of proliferating cells are heavily labeled with 3H-thymidine and appear black. Treatments were: a) vehicle alone, b) FSH from days 2 through 4, c ) P-endorphin and FSH from days 2 through 4, with the peptide added 3 hr prior to FSH. x 360. Our observation of lowered FSH-induced proliferation in isolated, neonatal Sertoli cells exposed first to endorphin indicates that the latter interacts with these cells to depress their responsiveness to FSH, a hormone shown previously by us to be a major factor controlling growth of the Sertoli cell population during testicular development (Orth, 1984). Since we also observed that endorphin does not interfere with stimulation of Sertoli cell division by CAMP, the peptide apparently suppresses the responsiveness of Sertoli cells to FSH by acting at a point prior to production of cyclic nucleotide within the cells. Moreover, we also found that prior inclusion of pertussis toxin with Sertoli cells blocked the negative effect of endorphin on their response to FSH but had no effect by itself on FSH-stimulated division. From this, we conclude that pertussis toxin interferes with the action of endorphin on Sertoli cells exposed to the hormone. Pertussis toxin has been shown by several investigators to ADP-ribosylate and hence inactivate several membrane-associated GTPbinding proteins, including Gi (Murayama and Ui, 19831, a regulatory component of adenyl cyclase that inhibits its activity (Birnbaumer et al., 1985). Thus, our findings with pertussis toxin suggest that endorphin blunts the enhancement of adenyl cyclase activity in Sertoli cells following their exposure to FSH by potentiating the interaction of GTP with one or more G proteins in these cells. Our current observations are generally consistent with findings of others concerning the effects of inhibitory ligands on Sertoli cells, a t least from older rats. Opiate receptors have been detected on mature Sertoli cells in vitro and shown to mediate suppression of FSHstimulated production of both ABP (Fabbri e t al., 1985) and inhibin (Morris et al., 1987) by these cells in culture. Moreover, Monaco et al. (1988) recently demonstrated that another inhibitory factor, adenosine, substantially depresses FSH-stimulated cAMP accumulation in Sertoli cells isolated from prepuberal rats. In addition, findings arising from use of pertussis toxin in that study suggest that adenosine receptors are coupled to the inhibitory component of adenyl cyclase, Gi. These observations on adenosine and prepuberal Sertoli cells largely parallel ours on endorphin and cells from neonates, suggesting that other factors may act to regulate hormonal responsiveness of these cells during testicular maturation. Our current data, along with that of our previous study (Orth, 19861, suggest that the physiological role of testicular endorphin in the fetal and neonatal rat may be to modify growth Of the Sertoli population by regulating the response of these cells to FSH. In our earlier study in viva, we found that local injection of testes of neonateswith antiserumto endorphin caused ENDORPHIN, FSH, AND SERTOLI CELL PROLIFERATION 325 I 1 I C W X + X C I v) v) U U +, C I- + W - 0 n I Fig. 4. The percentage of Sertoli cell nuclei labeled by 3H-thymidine in autoradiographs of cultures exposed to FSH, CAMP,or vehicle from days 2 through 4 in vitro, with or without prior inclusion of p-endorphin. In some chambers, pertussis toxin was also added 3 hr before f-endorphin andior FSH. Controls received vehicle alone, and H-thymidine was added to all cultures for the final 24 hr. In each group, 2,000-3,000 cells were examined in each of three or four chambers and labeled nuclei were quantified. Both FSH and cAMP dra- matically increased the percentage of cells dividing compared to controls (P<.Ol).Although prior addition of P-endorphin left the response to cAMP unaffected, it greatly reduced that of the cells to FSH (P<.Ol).However, when pertussis toxin was added before P-endorphin, the proliferative response of the cells to FSH was identical to that seen with FSH alone. Neither P-endorphin nor pertussis toxin had any effect on cellular proliferation in controls. a dramatic drop in the rate of Sertoli cell proliferation in those testes 6-17 hr later (Orth et al., 1984).In that study, we also examined intact fetal testes in organ culture and found that FSH caused a major elevation of Sertoli cell division under these conditions. However, when naloxone was added to the cultures prior to FSH, the response of the cells to the hormone was substantially increased. This suggested that, in the absence of naloxone, endogenous opiate partially suppressed the response of the cells to FSH, a conclusion borne out by the findings of the present study. It is interesting to note that, when we subsequently tested the effect of exogenously added endorphin on Sertoli cell division in other experiments in vivo and in organ culture, we were unable to demonstrate any direct effect of the peptide (unpublished observations). This presumably reflected the fact that Sertoli cells in vivo or in situ in organ cultures were already maximally affected by endogenous endorphin. Our present findings obtained with isolated Sertoli cells support this conclusion and confirm that endorphin interacts directly with Sertoli cells, in the virtual absence of other testicular cells, to regulate their FSH-responsiveness. A major source of testicular endorphin in vivo is no doubt the Leydig cell. This conclusion is supported by substantial evidence from several studies showing that Leydig cells of rats express the POMC-gene during development and again at maturity (Pintar et al., 1984; Shaha et al., 1984; Bardin et al., 1987) and release P-endorphin into testicular interstitial fluid in vivo (Valenca and Negro-Vilar, 1986). Indeed, since levels of endorphin in interstitial fluid greatly surpass those measured in peripheral blood of adult rats, Leydig cells apparently release considerable amounts of endorphin in vivo. When these observations are considered along with our data on neonatal Sertoli cells and those of others on cells from adults (Fabbri et al., 1985), it seems likely that endorphin of Leydig cell origin is a component of a local communication system functioning between these cells and Sertoli cells both during development and at maturity. This notion is further supported by the observation of a bi-phasic secretion of endorphin by Leydig cells. Both the number of Leydig cells immunostainable for p-endorphin (Shaha et al., 1984) and the amount of POMC transcript detectable in these cells (Gizang-Ginsberg and Wolgemuth, 1987) are high in perinatal testes when Sertoli cells are dividing; subsequently, testicular endorphin becomes low or undetectable in prepuberal rats after Sertoli cells cease mitosis. After puberty, endorphin production resumes and continues during adult life (GizangGinsberg and Wolgemuth, 1987). Although our data suggest that, by virtue of their secretion of endorphin, Leydig cells of perinatal rats play a paracrine role as intratesticular regulators of Sertoli cell population size, the function of endorphin in the adult testis is presently unknown. It is interesting that, in the presence of the relatively 326 J.M. ORTH AND R. BOEHM large dose of P-endorphin used in the current study, the percentage of proliferative Sertoli cells in FSH-treated cultures was reduced by about half compared to other chambers containing hormone but lacking p-endorphin. Thus, the level of cell division in cultures exposed to endorphin and FSH was still greater than that in untreated controls. Although correlating Sertoli cell responsiveness with various levels of endorphin dosage was beyond the scope of this initial study, it is unlikely that insufficient peptide was present to fully suppress the FSH-responsiveness of the cells. Other explanations for the persistence of some proliferative Sertoli cells in endorphin-treated,FSH-stimulated cultures include the possibility that not all cells react equally to p-endorphin, perhaps due to the presence of a sub-population in these cultures that lacks endorphin receptors. This and other alternatives will be explored further in future studies of endorphin-FSH interaction with Sertoli cells of newborn rats. We also found that the responsiveness of neonatal Sertoli cells to FSH after several days in vitro apparently depends on the conditions under which the cells are initially cultured. FSH elicited a dramatic increase in the percent of Sertoli cells labeled by 3H-thymidine when the hormone was included from day 2 through day 4 in vitro and the isotope was added for the 4th day of culture. In contrast, if FSH was added at the same time as 3H-thymidine, for only the final 24 h r in vitro, no significant increase in the rate of proliferation of these cells was detectable. Failure of the cells to respond to FSH under these conditions was apparently not due to insufficient time of exposure to the hormone since, in a previous study, we observed a proliferative response of fetal Sertoli cells after a considerably shorter incubation with FSH (Orth, 1984). Moreover, since we found that the cells still responded to CAMP after 3 days of culture without added factors, they apparently did not lose their inherent ability to initiate division during this period in vitro. These observations suggest that isolated neonatal Sertoli cells require sustained exposure to FSH to retain their responsiveness to this hormone. The most likely explanation for this finding is that the presence of FSH throughout the culture period insures the ability of Sertoli cells to respond by maintaining availability of FSH receptors on these cells. This apparent up-regulation of neonatal Sertoli cells by FSH may involve either stimulation of receptor synthesis de novo or enhanced re-cycling of receptor to the cell surface. Similar observations for LH (Huhtaniemi et al., 1981)suggest that this hormone also upregulates its receptor on immature Leydig cells, in contrast to its down-regulation of mature Leydig cells. Thus, both FSH and LH may interact differently with their target cells in developing animals than in adults. In summary, p-endorphin acts directly on Sertoli cells to blunt their proliferative response to FSH via a mechanism likely to involve a n intracellular G protein that modifies the level of adenyl cyclase activity in the presence of the hormone. The current data, coupled with previous findings from our laboratory, provide evidence supporting the concept of communication between Leydig cells and Sertoli cells of perinatal rats via a paracrine system involving P-endorphin produced by Leydig cells. Our previous reports have documented the role of FSH as a physiological stimulus of Sertoli cell division during testicular development that regulates the ultimate size of the Sertoli cell population and thus critically affects the output of spermatogenic cells at maturity. The existence of a n endorphin-based mechanism whereby the interstitial compartment modifies the response of developing Sertoli cells to FSH, as supported by our current data, suggests that Leydig cells also play a vital role in defining the size of the Sertoli cell population. Thus, functional interaction between Leydig and Sertoli cells is apparently established during development, underscoring further the critical nature of the perinatal period for future testicular function in the adult. ACKNOWLEDGMENTS The authors acknowledge with gratitude a gift of ovine FSH from the National Pituitary Agency. This work was supported by NIH Grant HD-15563 (to J.O.) and Research Career Development Award HD00591 (to J.O.). LITERATURE CITED Bardin, C.W., C.-L.C. Chen, P.L. Morris, I. Gerendai, C. Boitani, A.S. Liotta, A. Margioris, and D.T. Krieger 1987 Proopiomelanocortinderived peptides in testis, ovary and tissues of reproduction. Recent Prog. Horm. Res., 43:l-25. Bardin, C.W., C.Y. Cheng, N.A. Musto, and G.L. Gunsalus 1988 The Sertoli cell. In: The Physiology of Reproduction. E. Knobil and J . Neill, eds. Raven Press Ltd, New York, Vol. 1, p. 952. Birnbaumer, L., J . Codina, R. Mattera, R.A. Cerione, J. Hildebrandt, T. Sunyer, F.J. Rojas, M.G. Caron, R.J. Lefiowitz, andR. Iyengar 1985 Regulation of hormone receptors and adenylyl cyclases by guanine nucleotide binding N proteins. Recent Prog. Horm. Res., 41 t41-66. Chapin, R., J . Phelps, B. Miller, and T. Gray 1987 Alkaline phosphatase histochemistry discriminates peritubular cells in primary rat testicular cell culture. J . Androl., 8:155-161. Fabbri, A., C. Tsai-Morris, S. Luna, F. Fraioli, and M. Dufau 1985 Opiate receptors in the rat testis. Identification and localization in Sertoli cells. Endocrinology, 11 7:2544-2546. Gizang-Ginsberg, E., and D.J. Wolgemuth 1987 Expression of the proopiomelanocortin gene is developmentally regulated and affected by germ cells in the male mouse reproductive system. Proc. Natl. Acad. Sci. USA, 84:1600-1604. Huhtaniemi, I.T., M. Katikineni, and K.J. Catt 1981 Regulation of luteinizing hormone receptors and steroidogenesis in the neonatal rat testis. Endocrinology, 109:588-595. Monaco, L., D. DeManno, M.W. Martin, and M. Conti 1988 Adenosine inhibition of the hormonal response in the Sertoli cell is reversed by pertussis toxin. Endocrinology, 122:2692-2698. Morris, P., W. Vale, and C.W. Bardin 1987 p-endorphin regulation of FSH-stimulated inhibin production is a component of a short loop system in testis. Biochem. Biophys. Res. Commun., 148:15131519. Murayama, T., and M. Ui 1983 Loss of the inhibitory function of the guanine nucleotide regulatory component of adenylate cyclase due to its ADP ribosylation by islet-activating protein, pertussis toxin, in adipocyte membranes. J . Biol. Chem., 258:3319-3328. Orth, J.M. 1982 Proliferation of Sertoli cells in fetal and postnatal rats: A quantitative autoradiographic study. Anat. Rec., 203: 485-492. Orth, J.M. 1984 The role of follicle stimulating hormone in controlling Sertoli cell proliferation in testes of fetal rats. Endocrinology, 115:1248-1255. Orth, J.M. 1986 FSH-induced Sertoli cell proliferation in the developing rat is modified by P-endorphin produced in the testis. Endocrinology, 119:1876-1878. Orth, J.M., G.L. Gunsalus, and A.A. Lamperti 1988 Evidence from Sertoli cell-depleted rats indicates that spermatid number in adults depends on numbers of Sertoli cells produced during perinatal development. Endocrinology, 122:787-794. Orth, J.M., C. Higginbotham, and R. Salisbury 1984 Hemi-castration causes and testosterone prevents enhanced uptake of [3H]thymidine by Sertoli cells of immature rats. Biol. Reprod., 30: 263-270. ENDORPHIN, FSH, AND SERTOLI CELL PROLIFERATION Orth, J.M., and J. Weisz 1980 Development of A5-3phydroxysteroid dehydrogenase and glucose-6-phosphate dehydrogenase activity in Leydig cells of the fetal rat testis: A quantitative cytochemical study. Biol. Reprod., 22:1201-1209. Pintar, J.E., B.S. Schachter, A.B. Herman, S. Durgerian, and D.T. Krieger 1984 Characterization and localization of proopiomelanocortin messenger RNA in the adult rat testis. Science, 225r632634. Rong-Xi, D., D. Djakiew, and M. Dym 1987 Endocytic activity of Sertoli cells grown in bicameral culture chambers. Anat. Rec., 218: 306-312. Shaha, C., A. Liotta, D. Krieger, and C.W. Bardin 1984 The ontogeny 327 of immunoreactive P-endorphin in fetal, neonatal and pubertal testes from mouse and hamster. Endocrinology, 114:1584-1591. Sharpe, R.M. 1984 Intratesticular factors controlling testicular function. Biol. Reprod., 30:29-49. Steinberger, A., and E. Steinberger 1971 Replication pattern of Sertoli cells in maturing rat testis in vivo and in organ culture. Biol. Reprod., 4:84-87. Valenca, M. and A. Negro-Vilar 1986 Proopiomelanocortin-derived peptides in testicular interstitial fluid: characterization and changes in secretion after human chorionic gonadotropin or luteinizing hormone-releasing hormone analog treatment. Endocrinology, 118:32-37.