Autoimmune orchitis can be induced by thymic lymphocytes autosensitized against syngeneic sertoli cells in vitro

Thymic lymphocytes from normal inbred Lewis/Wistar rats were cocultured with syngeneic Sertoli cell-peritubular cell preparations in the presence of heterologous or allogeneic serum. Thymic cells cultured in this manner bound to Sertoli cells, became autosensitized , and markedly altered syngeneic Sertoli cell surface properties and remodeling functions in vitro. In contrast, control thymic cells incubated with Sertoli cells in autologous or syngeneic serum did not become sensitized. Coculture of autosensitized thymic cells with syngeneic seminiferous tubule segments, or local transfer of such lymphocytes into syngeneic rat testes, resulted in intratubular infiltration by "light cells." Intratesticular injection of autosensitized thymic cells was followed by derangement of the seminiferous epithelium, and by morphologic changes characteristic of experimental autoimmune orchitis. Thymic cells incubated with Sertoli cells in autologous or syngeneic serum did not elicit these changes. Thymic cells incubated with peritubular cells in heterologous or autologous serum behaved like control thymocytes, and were not sensitized. Data presented indicate that thymic cells are potentially capable of recognizing syngeneic Sertoli cell self-antigens. We speculate that factors normally present in serum may inhibit the recognition by thymic lymphocytes of antigenic determinants present on Sertoli cells. We discuss the possibility that the modulation of interactions between immature thymic lymphocytes and Sertoli cells is implicated in the prevention of autoimmune reactions against the testis.

Fibronectin synthesis is a marker for peritubular cell contaminants in Sertoli cell-enriched cultures

With indirect immunofluorescent microscopic techniques, we have shown that fibronectin is distributed primarily in or along the basal lamina of the seminiferous tubule boundary tissue in sections of testes from 20-day-old rats. Purified rat Sertoli cell-enriched aggregates, maintained in culture in the presence or absence of serum, exhibit no detectable immunofluorescence with fibronectin antibody, whereas purified peritubular cells in culture do have a positive reaction to fibronectin antibody. Peritubular cells in culture incorporate [35S] methionine into fibronectin which can be immunoprecipitated with a fibronectin antiserum, but Sertoli cells do not. We have used various criteria to estimate the degree of purity of Sertoli cell-enriched preparations. The presence of peritubular myoid cells in conventional Sertoli cell-enriched aggregates, cultured in the presence or absence of serum, can be detected with transmission electron microscopic examination, by the Feulgen staining procedure, and by the immunocytochemical identification of fibronectin. We describe a technique to purify Sertoli cells in conventional Sertoli cell-enriched preparations by treatment with hyaluronidase, resulting in a lesser number of peritubular cells by the above criteria, even in preparations cultured in the presence of serum. Data presented suggest that some of the products previously attributed exclusively to Sertoli cells in Sertoli cell-enriched preparations, particularly those cultured in the presence of serum, may have been contributed by peritubular cells.

Extracellular matrix promotes rat Sertoli cell histotypic expression in vitro

We describe procedures for the preparation of a cell-free seminiferous tubule biomatrix, and provide evidence demonstrating that this material constitutes a useful substratum for maintaining the normal architecture of Sertoli cells in primary culture. Seminiferous tubule biomatrix, which has the morphological appearance of a fibrillar network rich in filaments and amorphous substance, is shown to consist of about 50% protein, most of which is collagen and glycoproteins. Fibronectin and laminin are also present in the seminiferous tubule biomatrix, as judged by immunofluorescence microscopy. Sertoli cell aggregates plated on this substratum retain a cuboidal to columnar shape, spread very slowly to form a monolayer, and survive for at least 3 weeks when cultured in a hormone-free, serum-free, chemically defined medium. In contrast, Sertoli cells plated onto uncoated polystyrene readily spread to form a monolayer of flat squamous cells which do not survive as well. Other morphological and ultrastructural characteristics are described which indicate that cells cultured on the seminiferous tubule biomatrix more closely resemble those of Sertoli cells in vivo than do cells plated on uncoated plastic. These differences in cell structure, including the maintenance of normal polarity as indicated by the presence of basolateral tight junctional complexes, remain evident for periods of 10 to 14 days after plating Sertoli cells onto biomatrix substratum. Rates of DNA synthesis by immature Sertoli cells plated onto biomatrix are less than rates by cells plated onto uncoated plastic. The data are discussed in relation to the role of substratum in the preservation of normal functions and histotype of Sertoli cells.

Hormonal influences on formation of plasminogen activator by cultured testis tubule segments at defined stages of the cycle of the seminiferous epithelium

We confirm that basal rates of formation of plasminogen activator by cells in cultured tubule segments at stages of the cycle associated with spermiation (stages VII and VIII) are higher than those by cells in tubule segments at any other stages of the cycle of the seminiferous epithelium. We demonstrate that addition of cAMP derivatives or follicle-stimulating hormone in the presence of a phosphodiesterase inhibitor results in a large stimulation of plasminogen activator formation by tubule segments at all stages of the cycle. The greatest percentage increase (approximately 100-fold) is observed in cells in tubule segments having lowest basal rates of plasminogen activator formation (stages IX-VI). Even under stimulated conditions, however, the amounts of plasminogen activator produced by cultured tubule segments at stages VII and VIII remain greater than those produced by cultured tubule segments at other stages of the cycle, and these differences persist during organ culture for 48 h. Insulin and testosterone do not alter rates of formation of plasminogen activator. We conclude that Sertoli cells, the primary source of formation of plasminogen activator in the testis, are metabolically heterogeneous in the seminiferous tubule and that the germ cell association patterns in various stages of the cycle modulate Sertoli cell functions. We discuss the data in relation to the tissue restructuring within the seminiferous tubule which occurs during spermatogenesis and the possible role of plasminogen activator in these processes.

Purification and characterization of a cell-aggregating factor (clusterin), the major glycoprotein in ram rete testis fluid

Clusterin has been purified from ram rete testis fluid by conventional techniques and by immunoaffinity chromatography. The molecule is characterized as a glycoprotein having a molecular mass of approximately 80,000 Da and an isoelectric point of 3.6. The purified protein retains the capacity to elicit clustering of cells in an in vitro assay. Under reducing conditions in the presence of sodium dodecyl sulfate, clusterin dissociates into subunits of about 40,000 Da. Heterogeneities in apparent molecular mass were eliminated after treatment of clusterin with neuraminidase. Gel filtration chromatography revealed that clusterin exists in dimeric and tetrameric forms under conditions of neutral pH and low salt concentrations. In the presence of 6 M urea, only the monomeric form is evident, with an apparent molecular mass of approximately 85,000 Da. Clusterin, which was found to contain 4.5% glucosamine, binds to concanavalin A-Sepharose and also to wheat germ agglutinin Sepharose. The amino acid composition of clusterin is reported. The possible cellular source of clusterin in rete testis fluid is discussed. It is shown that Sertoli cells in the seminiferous tubule are one potential source, since primary cultures of rat Sertoli cells secrete a protein having the same immunochemical and physical properties as clusterin isolated from ram rete testis fluid. Possible functions of clusterin are discussed.

Ram rete testis fluid contains a protein (clusterin) which influences cell-cell interactions in vitro

Ram rete testis fluid is shown to elicit clustering of suspensions of Sertoli cells from testes of immature rats, TM-4 cells derived from mouse testis, and erythrocytes from several species. Details of bioassay procedures and characteristics of the phenomenon are reported. Concanavalin A and wheat germ agglutinin prevent aggregation elicited by rete testis fluid, and this inhibition is specifically prevented by alpha-methylmannoside and N-acetyl-glucosamine, respectively. Influences of rete testis fluid on cell aggregation are not dependent on exogenous calcium, but clustering is blocked by various metabolic inhibitors such as dinitrophenol. Rete testis fluid addition to mixed suspensions of erythrocytes and TM-4 cells is followed by separate aggregation of each cell types. Using aggregation of TM-4 cells suspended in simple medium at low density in rotation as a bioassy, we have determined which fractions in rete testis fluid retain activity. We have shown that a heat-stable, trypsin-sensitive protein, having an isoelectric point below pH 4.0, retains the capacity to aggregate cells. We discuss the possible functions of this protein, named clusterin, in cell interactions.

Evidence for a defective seminiferous tubule barrier in testes of Tfm and Sxr mice

The competence of the seminiferous tubule barrier was evaluated by determining the sucrose space in isolated testis preparations obtained from adult wild-type mice and from mutants having defects in spermatogenesis. The distribution of [3H]sucrose in testes from wild-type mice was in a space which constituted 18.5% of the [14C]urea space. The corresponding sucrose space in testes from Tfm/Y and Sxr/+ sterile mutants was elevated to 57.7% and 52.0%, respectively. Independent approaches to detect an impairment in the seminiferous tubule barrier of testes from Tfm/Y mice consisted of estimations of the osmotic barrier to hypertonic LiCl, using histological techniques, and determination of the testis sucrose space in vivo. With both approaches, results obtained supported the presence of an impaired barrier in testes of Tfm/Y mice. The sucrose space was within normal limits in testes of the other mutants examined, which were defective in spermatogenesis. The results do not provide any evidence for an impaired barrier in testes of mice having Movbr/Y and Gy/Y mutations, or the t6twl genotype.

The absence of specific interactions of Sertoli-cell-secreted proteins with antibodies directed against H-Y antigen

Radiolabeled proteins secreted into the medium by rat Sertoli cells in primary culture have been examined for specific interactions with polyclonal and monoclonal antibodies directed against serologically detectable H-Y antigen(s). None of the proteins secreted by Sertoli cells reacted specifically with H-Y antibodies, as determined with immunoprecipitation procedures and immunoabsorbent affinity chromatography, followed by SDS gel electrophoresis. Radioactivity profiles of proteins obtained after reaction with H-Y antibodies were similar to those observed after treatment with nonimmune sera or with irrelevant antibodies. We obtained comparable findings with proteins secreted by the mouse cell line TM4, which is of presumptive Sertoli cell origin, and with proteins present in ram rete testis fluid. These and other findings presented do not support the contention that Sertoli cells secrete a protein having the properties of serologically detectable H-Y antigen as previously described.

An objective sperm cytotoxicity assay for male-specific antisera based on ATP levels of unlysed cells: application to assay of H-Y antigen

We correlated the decrease in levels of ATP in spermatozoa with the extent of cytotoxicity elicited by antibodies against antigenic components on sperm. In the presence of concentrations of complement which did not cause cytolysis or influence the ATP content of epididymal sperm, addition of heat-in-activated sera from non-immunized mice, rats or rabbits did not result in sperm cytolysis or a fall in ATP content. In contrast, addition of rabbit anti-rat spermatocyte sera, which has previously been shown to react with rat spermatozoa (Tung, P.S. and Fritz, I.B. (1978) Dev. Biol. 64, 297-315), did cause sperm cytolysis and a decrease in ATP content. The titre of this antiserum for 50% cytolysis was between 1 : 128 and 1 : 256, as determined by the fall in ATP content. Using these criteria, we examined the cytotoxicity against sperm of different samples of anti H-Y sera. We examined the influence of monoclonal antibody against H-Y, mouse H-Y antisera and rat H-Y antisera raised in inbred females immunized with spleen cells from males of the same strains. In all cases, anti-H-Y lowered ATP levels and lysed sperm with a cytotoxic titre between 1 : 8 and 1 : 16. Measurements of the decrease in ATP content in sperm have been shown to provide an objective and reliable estimate of the percentage of spermatozoa lysed by H-Y antisera. Cytotoxic activity of H-Y antisera was removed by absorption with spleen cells from male mice but not by absorption with spleen cells from female mice.

The control of the synthesis and secretion of plasminogen activator by rat sertoli cells in culture

Sertoli cells in primary cultures produce plasminogen activator activity, and release it into the medium at rates greatly influenced by a variety of factors, including cell density, the presence of hormones, incubation temperature and duration of culture. In Sertoli cells maintained in culture in the presence of dibutyryl cAMP, the amounts of plasminogen activator activity secreted per cells were maximal at cell densities up to 2.5 microgram DNA/cm2 (350 units/microgram cell DNA), and declined to 40 units/microgran cell DNA at a density of 22 micrograms DNA/cm2. Concentrations of follicle-stimulating hormone (FSH) required to elicit half-maximal stimulation of the production of plasminogen activator activity were 0.37 micrograms/ml for oFSH-NIH S12 and 8 ng/ml for the more purified of SH-S1528C2. The ED50 for dibutyryl cAMP was found to be 0.08 mM. Addition of an inhibitor of phosphodiesterase (3-isobutyl-l-methylxanthine) enhanced the formation of plasminogen activator by cells cultured in the presence of FSH. Addition to the culture medium of testosterone, epidermal growth factor, insulin, human chorionic gonadotropin or prostaglandins (E1, E2 or F1 alpha) did not result in increased production of PA activity by Sertoli cells. Cells in culture for as long as 14 days remained responsive to FSH or dibutyryl cAMP. Increases in cellular levels of plasminogen activator became evident within 2-4 after addition of either FSH or dibutyryl cAMP to the medium. The stimulation of FSH or dibutyryl cAMP of the production of plasminogen activator activity was shown to be dependent upon de novo synthesis of RNA and protein. Levels of enzyme activity released by Sertoli cells maintained in culture for 48 h at 37 degrees C were approx. 50% higher than plasminogen activator released by cells cultured at 32 degrees C. The control of the production of plasminogen activator activity by Sertoli cells was discussed in relation to the control of plasminogen activator production by granulose cells, and the possible role of plasminogen activator in gonadal functions.

Changes in levels of plasminogen activator activity in normal and germ-cell-depleted testes during development

Levels of plasminogen activator activity were determined in testes obtained from normal and irradiated rats in various ages. During normal development, plasminogen activator activity per g testis increased rapidly between 40 and 60 days of age, but a comparable rise did not occur in germ-cell depleted testes of irradiated rats. Levels of enzyme in various populations of testicular cells were highest in Sertoli (varying between 1800 and 6300 units/mg protein in cell maintained under different culture conditions), and lowest in peritubular myoid cells (about 1 unit/mg protein), with intermediate levels in germinal cells (ranging between 147 and 560 units/Mg protein in residual bodies, spermatocytes and spermatids). No protease inhibitor could be detected in germ-cell extracts. The addition to the medium in which Sertoli cells were in culture of particles which can be phagocytosed (autoclaved E. coli) resulted in an increased formation of plasminogen activator activity by Sertoli cells. A synergistic enhancement of enzyme production resulted following the addition of submaximal quantities of dibutyryl cyclic AMP and autoclaved bacteria to sertoli cells in culture. On the basis of these data, we suggest that the presence of advanced germinal cells during gonadal development may stimulate the synthesis of plasminogen activator by Sertoli cells, mediated in part by the phagocytosis of residual bodies by sertoli cells which occurs prior to spermiation.

Myoinositol biosynthesis by Sertoli cells, and levels of myoinositol biosynthetic enzymes in testis and epididymis

Levels of glucose-6-phosphate cyclase (myoinsitol-1-phosphate synthase, EC 5.5.1.4) and myoinositol-1-phosphate phosphatase (myoinositol-1-phosphatase, EC 3.1.3.25) were determined in extracts of testes from10-, 20-, and 30-day-old rats, and in extracts of Sertoli cells, germinal cells, and epididymides. The specific activity of the cyclase was approximately 1/10th that of the phosphatase in all extracts found to contain either enzyme. Among cells in the testis examined, Sertoli cells had highest levels of enzymes required for inositol biosynthesis from glucose, while spermatocytes and round spermatids did not have detectable activity. Spermatozoa from the epididymis also had no detectable cyclase or phosphatase activity. In contrast, extracts of washed epididymides contained exceedingly high specific activities of these enzymes. Primary cultures of Sertoli cells, maintained in a chemically defined medium without added inositol, released inositol into the medium during three successive 24-h periods. The amounts released were greater in cells stimulated by dibutyryl cyclic AMP. Results were interpreted to indicate that inositol in the fluid of seminiferous tubules most probably originates from Sertoli cells, which synthesize inositol from glucose. Additional inositol in the fluid of epididymal tubules could readily be provided by metabolism of glucose by epididymal epithelial cells

Follicle-stimulating hormone and testosterone independently increase the production of androgen-binding protein by Sertoli cells in culture

Dose-response curves were obtained for the production of androgen-binding protein (ABP) by Sertoli cells prepared from testes of 20-day-old rats and treated in culture with either FSH or testosterone (T). FSH stimulated ABP production by up to 3.5 times control levels. For NIH-FSH-S11, the ED50 was 3 ng/ml, and for highly purified ovine FSH, the ED50 was 0.066 ng/ml. Addition of T produced a stimulation of up to 3 times control levels; half-maximal response was obtained at a dose of 4 nM. The presence of small numbers of contaminating Leydig cells in some preparations resulted in production of endogenous T, especially when high doses of NIH-FSH, which contains some LH, were employed. A modified preparation method involving exposure of the cells to distilled water reduced the endogenous T production to low levels. In cultures of cells prepared in this way, addition of the antiandrogen cyproterone acetate at a concentration high enough to reduce fractional occupancy of androgen receptors by endogenous T to 0.014 or less had no effect on the stimulation by FSH of ABP production in the cultures. In contrast, cyproterone acetate inhibited stimulation by T of ABP production. We conclude that FSH and T act independently on Sertoli cells from immature rats to increase the secretion of ABP.

Metabolic responses of Sertoli cells in culture to various concentrations of follicle stimulating hormone and cholera toxin

The concentration of cholera toxin required for half-maximal stimulation of cAMP production by Sertoli cell enriched cultures (4.48 X 10(2) microgram/ml) is greater than that required for half-maximal stimulation of 17beta-estradiol synthesis from testosterone (2.34 X 10(-4) microgram/ml), [3H]thymidine incorporation into DNA (1.48 X 10(-5) microgram/ml), or androgen binding protein production (2.43 X 10(-6) microgram/ml). The same relative dose response hierarchy was obtained with respect to stimulation of Sertoli cells with follicle stimulating hormone (FSH) preparations. Again, highest concentrations were required to elicit maximal cAMP production. The data are discussed in relation to an apparent paradox: If cAMP is the mediating 'second messenger' following stimulation by FSH or cholera toxin, why should highest concentrations of these agents be required to elicit 50% of maximal cAMP levels?

Paradox of the dose response to polypeptide hormones

Highest concentrations of follicle-stimulating hormone (FSH) are required to elicit maximal increases in cAMP formation by isolated Sertoli cells in culture, and progressively lower concentrations are required to increase three late responses maximally (estradiol synthesis from testosterone, DNA synthesis, and synthesis of androgen-binding protein, respectively). Identical dose-response curve relations are observed when cholera toxin is used as the stimulatory agent. Results do not rule out the possibility that cAMP could mediate the various responses of Sertoli cells to FSH.