Germ cell-Sertoli cell interactions: the effect of testicular maturation on the synthesis of cyclic protein-2 by rat Sertoli cells

Cyclic Protein-2 (CP-2) is synthesized in a stage-specific manner by mature rat Sertoli cells within stage VI and VII seminiferous tubules. To determine how testicular maturation affects CP-2 synthesis, we cultured 20 cm of tubules encompassing all stages of the cycle from rats 17, 35, 45, and 75 days old. The greatest increase in CP-2 synthesis was found to occur between 35 and 45 days and exceeded that observed for transferrin and sulfated glycoprotein (SGP)-2. Additionally, two-dimensional gel analysis indicated that secretion of CP-2 increased from 35 to 45 days to a greater extent than the secretion of SGP-1 and SGP-2 and transferrin. Biochemical analysis also demonstrated that CP-2 synthesis was stage-specific by 45 days. Immunocytochemistry expanded these observations; CP-2 was not detected in 7-35-day-old Sertoli cells. However, at 36 days, CP-2 was detected in Sertoli cells in stage VI and VII tubules but not at any other stage. CP-2 concentration in stage VI-VII tubules was increased by 38 days, but was unchanged thereafter. Finally, we immunocytochemically examined age-related changes in CP-2 concentration of the proximal convoluted kidney tubule. This analysis revealed that, at 1 wk, CP-2 was present in all proximal tubules except those in the subcapsular area; however, by 14 days, CP-2 was detected in all proximal tubules. This comparison of Sertoli cells and proximal tubule cells indicates that CP-2 content is determined by the maturity of a cell and not by the age of the animal.     

Germ cell-Sertoli cell interactions: analysis of the biosynthesis and secretion of cyclic protein-2

Cyclic protein-2 (CP-2) is secreted in vitro in substantial amounts by mature rat Sertoli cells in intact Stage VI and Stage VII seminiferous tubules. This stage-dependent secretion has led us to postulate that the biosynthesis of this molecule is stimulated by germ cells at a specific state of development. In order to explore this hypothesis and to examine the steps in CP-2's biosynthesis, we generated a polyclonal antisera against this protein and used it to analyze the biosynthesis and secretion of CP-2. Analysis of the steps in the biosynthesis of CP-2 indicated that its polypeptide core represented most if not all of the translation product of the CP-2 mRNA and that a single aspargine-linked oligosaccharide became attached to this core. Analysis of the rate of biosynthesis of CP-2 at specific stages of the cycle of the seminiferous epithelium was also conducted. Two-millimeter segments of tubules at Stage II, VI, VIIa, b, VIII, and XII were cultured for 1 hr in the presence of [35S]methionine and radiolabeled CP-2 immunoprecipitated from the tubules. Data (35S-CP-2 synthesized per hour) demonstrated that the rate of CP-2's biosynthesis increased 9-fold from Stage II to Stages VI and VIIa, b and then decreased 13-fold by Stage XII. To determine whether these rates of biosynthesis were identical to the rates of secretion, tubules were cultured for 17 hr with [35S]methionine, CP-2 was immunoprecipitated from the culture medium and data were expressed as 35S-CP-2 secreted per hour. This analysis demonstrated that the rate of secretion of CP-2 varied in the same stage-specific manner as its rate of synthesis. However, at each stage, the apparent rate of biosynthesis of the molecule exceeded its apparent rate of secretion. In order to explain this observation, we analyzed the rate of export of newly synthesized CP-2 out of the tubules. This demonstrated that quantitative export of the protein into culture medium required at least 17 hr. This period of time was most likely due to the retention of the protein within the tubular lumen, since primary cultures of Sertoli cells were shown to rapidly secrete newly synthesized CP-2. We, therefore, concluded that CP-2 was biosynthesized in a stage-dependent manner and that all CP-2 was secreted.     

Immunocytochemical localization of urate oxidase, fatty acyl-CoA oxidase, and catalase in bovine kidney peroxisomes

We investigated the localization of urate oxidase, peroxisomal fatty acyl-CoA oxidase, and catalase in bovine kidney by immunoblot analysis and protein A-gold immunocytochemistry, using the respective polyclonal monospecific antibodies raised against the enzymes purified from rat liver. By immunoblot analysis, these three proteins were detected in bovine kidney and bovine liver homogenates. Subcellular localization of these three enzymes in kidney was ascertained by protein A-gold immunocytochemical staining of Lowicryl K4M-embedded tissue. Peroxisomes in bovine kidney cortical epithelium possessed crystalloid cores or nucleoids, which were found to be the exclusive sites of urate oxidase localization. The limiting membrane, the marginal plate, and the matrix of renal peroxisomes were negative for urate oxidase staining. In contrast, catalase and fatty acyl-CoA oxidase were found in the peroxisome matrix. These results demonstrate that, unlike rat kidney peroxisomes which lack urate oxidase, peroxisomes of bovine kidney contain this enzyme as well as peroxisomal fatty acyl-CoA oxidase.     

Identification of stage-specific proteins synthesized by rat seminiferous tubules

Experiments were conducted to determine how the cycle of the seminiferous epithelium influenced synthesis and secretion of proteins by seminiferous tubules. Tubular segments were treated with collagenase and then cultured with [35S]methionine. These myoid cell-depleted tubules isolated from different stages of the epithelial cycle exhibited, at Stages VI and XII, two distinct peaks of secretion of total radiolabeled proteins. Two-dimensional gel electrophoresis indicated that the patterns of secreted proteins from these two stages were remarkably different, while those from other stages were intermediate between those at the peaks. At least 15 proteins were secreted cyclically, many of them previously unrecognized products of the seminiferous epithelium. One product, designated Cyclic Protein-2 (CP-2), exhibited a pronounced cycle of secretion, its peak at Stage VI being 30-fold greater than at its nadir at Stages XII-XIV. Further investigation indicated that CP-2 did not appear to originate from myoid cells or dispersed germ cells but could be recovered from Sertoli cell-enriched cultures prepared from Stage VI tubules. Protein secretion by tubular segments was also characterized by immunoprecipitation with two polyspecific antisera directed against Sertoli cell products. Five secretory proteins were identified which had cycles different from one another and from CP-2. In contrast to secreted products, the synthesis of most cellular proteins by tubular segments remained relatively constant throughout the cycle. It is concluded: 1) segments of the seminiferous epithelium secrete proteins into the culture medium which are distinct from cellular proteins; 2) the synthesis of many of these proteins varies with the epithelial cycle; and 3) several of the secreted proteins are of Sertoli cell origin, including a newly identified protein, CP-2. This indicates that the morphology and the protein synthetic capacity of the seminiferous epithelium are coordinated over space and time.     

Densin is a novel cell membrane protein of Sertoli cells in the testis

Cell-cell interactions between Sertoli cells and germ cells are crucial for the maturation of germ cells in spermatogenesis but the structural and functional aspects of the interactions remain to be fully elucidated. Densin is a junction protein suggested to play a role in establishment of specific cell-cell contacts in the post-synaptic densities of the brain and the slit diaphragm of the kidney podocyte. In the present study, densin was discovered to be expressed in the testis of the man and the mouse. Expression of densin at the gene and the protein level was studied by using RT-PCR and Western blotting analyses, and the localization of densin was explored with immunofluorescence staining. RT-PCR and Western blotting analyses showed that densin is expressed at the gene and the protein levels. Immunofluorescence staining localized the expression of densin to the cell membranes of Sertoli cells suggesting that densin may be an adherens junction protein between Sertoli cells and developing germ cells. Densin is a novel testicular protein expressed in the cell membranes of Sertoli cells. Its functional role remains to be assessed.     

Localization of a 11.5 kDa Zn(2+)-binding protein (parathymosin) in different rat tissues. Cell type-specific distribution between cytosolic and nuclear compartment

The distribution of a small Zn(2+)-binding protein (11.5 kDa ZnBP), which we have shown to be identical with parathymosin, was studied in various rat tissues by immunocytochemistry, immunoblot analysis and quantified by enzyme-linked immunosorbent assay (ELISA), using monospecific polyclonal antibodies. The content in liver was 105 micrograms/g wet weight. Similar amounts were found in brain, adrenal gland and smooth muscle, whereas in testis, spleen, lung, and kidney about half the amount was detected. Very low levels were found in skeletal muscle (2 micrograms/g) and adipose tissue, while erythrocytes did not contain measurable amounts. The specificity of the antibodies was established by immunoblotting. Purified 11.5 kDa ZnBP as well as 11.5 kDa ZnBP detected in crude soluble fractions from various tissues appeared always as a doublet of protein bands of about equal intensity, indicative for two isoforms of the 11.5 kDa ZnBP. By immunocytochemistry, in brain, high concentrations of 11.5 kDa ZnBP were found in the deep cerebellar nuclei, in soma and dendrites of Purkinje cells, and in the large neurons of the pons/medulla. In most cell types reacting with the antibody, exclusively the cytoplasm was stained. In contrast, in duodenal and jejunal crypt cells immunostaining was restricted to the nuclei, whereas the more mature cells at the top of the villi contained most of the antigen in the cytosol. Immunostaining of the nuclei was also observed in pancreatic duct cells, whereas in the duct cells of the parotid gland immunostaining was detected exclusively in the cytoplasm. In both tissues immunostaining of the acinar cells was negative.     

Regulation of the synthesis and secretion of transferrin and cyclic protein-2/cathepsin L by mature rat Sertoli cells in culture.

While numerous studies have examined the response of immature rat Sertoli cells to specific hormones and growth factors, the regulation of mature cells in vitro has not been well examined because highly purified cells have been difficult to isolate. We now describe a detailed method for isolating Sertoli cells from mature (> 60 days of age) rats and generating primary cultures of these cells greater than 90% in purity. We demonstrate that cell density, hormones, and growth factors regulate the synthesis or secretion of two Sertoli cell products, transferrin and Cyclic Protein-2 (CP-2)/cathepsin L. Cell density modulated the response of mature Sertoli cells to some hormones; insulin (at 10 micrograms/ml) and epidermal growth factor (EGF) acted synergistically to stimulate transferrin synthesis by 80% when cells were cultured at a density of 1.65 x 10(5) cells/cm2 but had no effect on transferrin synthesis by cells cultured at 1.46 x 10(5) cells/cm2. A mixture of FSH, retinol, and testosterone increased transferrin synthesis by 30% at both cell densities, and this stimulation was independent of the effect of EGF and insulin. CP-2/cathepsin L synthesis was significantly stimulated by increased cell density. FSH, retinol, and testosterone also stimulated CP-2/cathepsin L synthesis by 30%; however, this stimulation just missed being statistically significant. Finally, we demonstrated that secretion of transferrin and CP-2 was reduced when cells were cultured in the presence of interleukin-1 alpha, a cytokine synthesized by Sertoli cells.(ABSTRACT TRUNCATED AT 250 WORDS)     

Sertoli and Leydig cells of the human testis express neurofilament triplet proteins

 Using RT-PCR, western blot and enzyme and fluorescence immunocytochemical techniques, the three isoforms of neurofilament proteins (NFPs), namely NF-L (NFP-68 kDa), NF-M (NFP-160 kDa) and NF-H (NFP-200 kDa) were found in Sertoli and Leydig cells of human testes. RT-PCR showed specific for the three NFP fragments in testicular tissue, in isolated seminiferous tubules and in isolated Leydig cells. In protein preparations from the same testicular components, western blot analysis detected bands with molecular weights characteristic for NF-H, NF-M and NF-L. Application of immunofluorescence and immunoenzyme methods on cryostat and paraffin sections resulted in differences in the staining pattern in Sertoli cells and Leydig cells. In these cells, the NFPs showed predominantly a perinuclear location from which bundles emerge that were directed towards the basal, apical and lateral extensions of the Sertoli cells as well as the periphery of Leydig cells. NF-H coexists with vimentin-type filaments as seen by dual staining and staining of conseccutive serial sections of material embedded in paraffin. In Sertoli cells, vimentin and NF-H showed distinct dynamic changes depending on the stage of spermatogenesis and some structural variations of seminiferous tubules. Although in some tubules both vimentin and NF-H immunoreactivity was present at high levels, in the Sertoli cells from most individuals an inverse relationship in the staining intensity of vimentin and NF-H was observed. The strongest NF-H immunoreactivity was detected in Sertoli cells associated with stage 3 spermatids, whereas vimentin immunoreactivity was most abundant in association with stage 5 spermatids. The leydig cells did not show functional changes of the NFP immunoreactivity. The results obtained provide new evidence for the heterogeneous phenotype of human Sertoli cells and raise the question of their exact nature and origin. Accepted: 17 November 1998     

Sterol carrier protein 2 (non-specific lipid transfer protein) is localized in membranous fractions of Leydig cells and Sertoli cells but not in germ cells.

The cellular and subcellular distribution of sterol carrier protein 2 (SCP2; nsL-TP) was reinvestigated in rat testicular cells by Western blotting and immunocytochemistry, using the affinity purified antibody against rat liver SCP2. Western blot analysis revealed high levels of the protein in the somatic cells of the testis, e.g., Leydig and Sertoli cells whereas it could not be detected in germ cells. This cellular localization of SCP2 was confirmed by Northern blotting. Immunocytochemical techniques revealed that in Leydig cells, immunoreactive proteins were concentrated in peroxisomes. Although SCP2 was also detected in Sertoli cells, a specific subcellular localization could not be shown. SCP2 was absent from germ cells. Analysis of subcellular fractions of Leydig cells showed that SCP2 is membrane bound without detectable amounts in the cytosolic fraction. These results are at variance with data published previously which suggested that in Leydig cells a substantial amount of SCP2 was present in the cytosol and that the distribution between membranes and cytosol was regulated by luteinizing hormone. The present data raise the question in what way SCP2 is involved in cholesterol transport between membranes in steroidogenic cells but also in non-steroidogenic cells.     

Tricellulin expression in brain endothelial and neural cells

Tricellulin is a tight junction (TJ) protein, which is not only concentrated at tricellular contacts but also present at bicellular contacts between epithelial tissues. We scrutinized the brain for tricellulin expression in endothelial and neural cells by using real-time polymerase chain reaction, Western blot and immunohistochemical and immunocytochemical analysis of cultured brain cells and paraffin sections of brain. Tricellulin mRNA was detected in primary cultures and in a cell line of human brain microvascular endothelial cells. Protein expression was confirmed by Western blot and immunofluorescence analysis, which further highlighted the localization of tricellulin in the cell membrane at tricellular and along bicellular contacts, and in the nucleus and perinuclear region. Compared with the well-studied TJ protein, zonula occludens-1, tricellulin expression was less marked at the cell membrane but more evident in the nuclear and perinuclear regions. The presence of tricellulin in cultured endothelial cells was corroborated by immunohistochemical and immunofluorescence staining in brain blood vessels, where it was colocalized with another TJ protein, claudin-5. Tricellulin mRNA was detected in neurons and astrocytes, whereas protein expression was observed in astrocytes but not in neurons, as shown by immunofluorescence analysis. This study reveals the presence and subcellular distribution of tricellulin in brain endothelial cells, both in vitro and in situ and its colocalization with other relevant TJ proteins. Moreover, it demonstrates the expression of the protein in astrocytes opening new avenues for future research to establish the biological significance of tricellulin expression in glial cells.     

Isolation of cyclic protein-2 from rat seminiferous tubule fluid and Sertoli cell culture medium

Cyclic Protein-2 (CP-2), a stage-specific secretory product of the rat seminiferous epithelium, has been isolated from seminiferous tubule fluid (STF) and Sertoli cell culture medium. Isolation from STF was accomplished by mixing STF with radiolabeled proteins secreted by Stage VI-VII seminiferous tubules and sequential fractionation of these proteins by hydroxylapatite, DEAE-agarose, and quaternary amine ion-exchange chromatography. Radiolabeled proteins were used to identify the chromatographic fractions that contained CP-2. Through use of these procedures, a highly purified preparation of radioinert CP-2 was obtained from seminiferous tubule fluid. Cyclic Protein-2 was also isolated from Sertoli cell culture medium, indicating that the Sertoli cell is its most likely source. Preliminary characterization of CP-2 was conducted. First, CP-2 appeared to be highly enriched in methionine. Second, the molecular weight of CP-2 was found to be 20,000. Third, analysis by reverse-phase hydrophobic chromatography indicated that CP-2 was relatively hydrophobic. We conclude that CP-2 is a small hydrophobic glycoprotein secreted in vivo and in vitro in a stage-specific manner by Sertoli cells.     

Ankrd7, a novel gene specifically expressed in Sertoli cells and its potential roles in Sertoli cell maturation

The somatic Sertoli cells play an essential role in testis determination and spermatogenesis by providing nutrition and structural support. In the current study, we report on the novel Ankrd7 gene that contains five ankyrin repeat domains. This gene was specifically expressed in Sertoli cells and was regulated in a maturation-dependent manner. Its expression was restricted to testicular tissue, and its mRNA could be detected in testes at as early as 14 dpp (days post partum) using RT-PCR analysis. In both testicular tissue sections and in vitro cultured Sertoli cells, the Ankrd7 protein was localized to the nucleus of the Sertoli cell. Immuno-histochemistry and immunocytochemistry investigations showed that the protein was detectable in testicular tissues at 20 dpp, at which time Sertoli cells were gradually differentiating into their mature cellular form. These results suggest that Ankrd7 is probably involved in the process of Sertoli cell maturation and in spermatogenesis.     

Sertoli cell processes have axoplasmic features: an ordered microtubule distribution and an abundant high molecular weight microtubule- associated protein (cytoplasmic dynein)

Microtubules in the cytoplasm of rat Sertoli cell stage VI-VIII testicular seminiferous epithelium were studied morphometrically by electron microscopy. The Sertoli cell microtubules demonstrated axonal features, being largely parallel in orientation and predominantly spaced one to two microtubule diameters apart, suggesting the presence of microtubule-bound spacer molecules. Testis microtubule-associated proteins (MAPs) were isolated by a taxol, salt elution procedure. Testis MAPs promoted microtubule assembly, but to a lesser degree than brain MAPs. High molecular weight MAPs, similar in electrophoretic mobilities to brain MAP-1 and MAP-2, were prominent components of total testis MAPs, though no shared immunoreactivity was detected between testis and brain high molecular weight MAPs using both polyclonal and monoclonal antibodies. Unlike brain high molecular weight MAPs, testis high molecular weight MAPs were not heat stable. Testis MAP composition, studied on postnatal days 5, 10, 15, and 24 and in the adult, changed dramatically during ontogeny. However, the expression of the major testis high molecular weight MAP, called HMW-2, was constitutive and independent of the development of mature germ cells. The Sertoli cell origin of HMW-2 was confirmed by identifying this protein as the major MAP found in an enriched Sertoli cell preparation and in two rat models of testicular injury characterized by germ cell depletion. HMW-2 was selectively released from testis microtubules by ATP and co-purified by sucrose density gradient centrifugation with MAP-1C, a neuronal cytoplasmic dynein. The inhibition of the microtubule-activated ATPase activity of HMW-2 by vanadate and erythro-(2-hydroxy-3-nonyl)adenine and its proteolytic breakdown by vanadate-dependent UV photocleavage confirmed the dynein-like nature of HMW-2. As demonstrated by this study, the neuronal and Sertoli cell cytoskeletons share morphological, structural and functional properties.     

Asynchronous expression of the homeodomain protein CUX1 in Sertoli cells and spermatids during spermatogenesis in mice

The homeodomain CUX1 protein exists as multiple isoforms that arise from proteolytic processing of a 200-kDa protein or an alternate splicing or from the use of an alternate promoter. The 200-kDa CUX1 protein is highly expressed in the developing kidney, where it functions to regulate cell proliferation. Transgenic mice ectopically expressing the 200-kDa CUX1 protein develop renal hyperplasia associated with reduced expression of the cyclin kinase inhibitor p27. A 55-kDa CUX1 isoform is expressed exclusively in the testes. We determined the pattern and timing of CUX1 protein expression in developing testes. CUX1 expression was continuous in Sertoli cells from prepubertal testes but became cyclic when spermatids appeared. In testes from mature mice, CUX1 was highly expressed only in round spermatids at stages IV-V of spermatogenesis, in both spermatids and Sertoli cells at stages VI-X of spermatogenesis, and only in Sertoli cells at stage XI of spermatogenesis. While most of the seminiferous tubules in wild-type mice were between stages VI and X of spermatogenesis, there was a significant reduction in the percentage of seminiferous tubules between stages VI and X in Cux1 transgenic mice and a significant increase in the percentage of seminiferous tubules in stages IV-V and XI. Moreover, CUX1 was not expressed in proliferating cells in testes from either wild-type or transgenic mice. Thus, unlike the somatic form of CUX1, which has a role in cell proliferation, the testis-specific form of CUX1 is not involved in cell division and appears to play a role in signaling between Sertoli cells and spermatids.