Luteinizing hormone inhibits Fas-induced apoptosis in ovarian surface epithelial cell lines.

Gonadotrophins including LH have been suggested to play an important role in the etiology of epithelial ovarian cancers. The goal of the present study was to obtain more insight in the mechanism of gonadotrophin action on ovarian surface epithelium (OSE) cells. As the Fas system is known to be a major player in the regulation of the process of apoptosis in the ovary, we investigated whether LH interfered with Fas-induced apoptosis in the human OSE cancer cell lines HEY and Caov-3. Activation of Fas receptor by an agonistic anti-Fas receptor antibody induced apoptosis, as was evaluated by caspase-3 activation, poly(ADP-ribose) polymerase fragmentation, phosphatidylserine externalization and morphological changes characteristic of apoptosis. Co-treatment with LH reduced the number of apoptotic cells following activation of Fas in a transient manner, while LH by itself did not affect apoptosis or cell proliferation. The anti-apoptotic effect of LH could be mimicked by the membrane-permeable cAMP analog 8-(4-chlorophenylthio) cAMP (8-CPT-cAMP), and blocked by H89, a specific inhibitor of protein kinase A (PKA). In conclusion, these findings suggest that LH protects HEY cells against Fas-induced apoptosis through a signaling cascade involving PKA. Although it is plausible that in vivo LH might also enhance OSE tumor growth through inhibition of apoptosis, further research is necessary to confirm this hypothesis.

Effect of fibroblast growth factor-2 on Sertoli cells and gonocytes in coculture during the perinatal period.

Sertoli cell-gonocyte cocultures obtained from rat testes 20 days postcoitum, 1 day postpartum, and 3 days postpartum were used to investigate the effect of FGF-2 on both somatic and germ cells in vitro during the perinatal period. With cells isolated from fetal, newborn, or 3-day-old animals, FGF-2 was found to significantly increase the number of Sertoli cells after 3 or 6 days of cultures, starting at a concentration of 1 ng/ml. FGF-2 did not increase the [3H]thymidine labeling index of Sertoli cells, indicating that FGF-2 is a survival factor for these cells in vitro. FGF-2 (1, 5, or 10 ng/ml) also significantly increased the number of gonocytes after 6 days of culture with cells from either newborn or 3-day-old animals. About twice as many germ cells were found in those cultures compared to the control cultures. Addition of a neutralizing antibody against FGF-2 to control cultures caused a significant decrease in the number of gonocytes compared to that in untreated cultures after 6 days, whereas with FGF-2, the antibody decreased the number of germ cells to control levels. FGF-2 significantly stimulated the proliferative activity of the gonocytes after 3 or 5 days, indicating that FGF-2 is a survival as well as a mitogenic factor for these cells. Taken together, these data suggest that FGF-2 is an important factor around the start of spermatogenesis, at least in vitro.

Purified gonocytes from the neonatal rat form foci of proliferating germ cells in vitro.

Due to the lack of a specific marker for gonocytes from newborn rats, isolation of these cells has proven difficult and laborious. We have found a specific cell membrane marker, the Epithelial Cellular Adhesion Molecular (Ep-CAM) that can be used to isolate these cells using antibody directed cell sorting. 4 days post partum (dpp) rat testes were enzyme treated to attain a cell suspension, which was labelled with an antibody (GZ1) against Ep-CAM and tagged with a fluorescent probe. The labelled cell suspension was run through a FACS cell sorter, from which a gonocyte suspension of >85% purity was attained. The cells remained viable in culture and proliferated actively as determined by double labelling the cells with anti-HSP90alpha (a specific germ cell marker) and anti-BrdU antibodies (after BrdU incorporation). During culture, these cells formed chains of 2 to 4 cells and aggregates of proliferating germ cells were found after 8 days of culture.

Leydig cell apoptosis after the administration of ethane dimethanesulfonate to the adult male rat is a Fas-mediated process.

Leydig cells undergo apoptosis in response to the cytotoxin ethane dimethanesulfonate (EDS), with numbers declining at 12-18 h and maximal apoptosis at 24 h postinjection. The Bcl-2 family members, Bcl-2, Bcl-xl, and Bax, appear not to be involved in this process. To further investigate this phenomena, a single dose of EDS was administered to adult rats to induce the killing of Leydig cells. The interstitial cells were examined up to 3 days after EDS administration by Western blot analysis for the Bcl-2 family members (Bak and Bcl-w). Western blotting showed that Bak expression in the interstitial cell preparations was unchanged after EDS, and immunohistochemistry showed that it was not up-regulated in Leydig cells in response to EDS. Bcl-w expression in the Leydig cells and interstitial cell preparations was unchanged until 48 h when it became undetectable, suggesting that Leydig cell-associated Bcl-w is not involved in initiating apoptosis. We also investigated the role of the Fas system in Leydig cell apoptosis. Both Fas receptor and Fas ligand protein levels increased after EDS, peaking at 12-18 h and declining thereafter. Fas receptor and ligand were shown by immunohistochemistry to be present in Leydig cells, and after EDS all Leydig cells became strongly positive for both proteins. The intensity of staining increased in the early stages of apoptosis and decreased as the nuclear morphology became more fragmented. These data suggest that Bcl-2 family members are not involved in Leydig cell apoptosis after EDS administration. However, up-regulation of the Fas system does occur, implicating activation of Fas receptor in the induction of Leydig cell apoptosis.

Leukemia inhibitory factor and ciliary neurotropic factor promote the survival of Sertoli cells and gonocytes in coculture system.

Leukemia inhibitory factor (LIF) and ciliary neurotropic factor (CNTF) were found to be pleiotropic modulators of Sertoli cell and gonocyte development (both isolated from the neonatal rat testis) in a coculture system, whereas IL-6, another member of this cytokine family, had no effect on these cells. LIF and CNTF significantly enhanced the survival of the Sertoli cells in a dose- and time-dependent manner. The effect of LIF on the Sertoli cells was significant at a concentration of 1 ng/ml after 3 or 6 days of culture, whereas CNTF had a significant effect at 10 ng/ml. Neither LIF nor CNTF had an effect on Sertoli cell proliferation. The survival of proliferating gonocytes (isolated from 3-day-old rats testes) was also significantly higher in cultures to which LIF (7.5 ng/ml) or CNTF (10 ng/ml) was added. No effect of these cytokines was found on the mitotic activity of proliferating gonocytes. However, LIF (7.5 ng/ml) stimulated the proliferation of quiescent gonocytes (isolated from day 1 testes) after 3 days of culture. Combinations of LIF (or CNTF) with fibroblast growth factor 2 (10 ng/ml) and steel factor (50 ng/ml) did not further improve the long term culture of the gonocytes. LIf- and CNTF-like proteins of the expected molecular masses (32,000 and 22,000 daltons, respectively, under reducing conditions) were found by Western blotting in testicular extracts of 3-day-old rats. Taken together, these results indicate that LIF or CNTF may play a role at the start of the spermatogenesis. The characterization of receptors for LIF or CNTF on the gonocytes and/or neonatal Sertoli cells will aid in a better understanding of the physiological role of these cytokines in the reproductive system.

A monoclonal antibody recognizing a differentiation marker on rat gonocytes.

Monoclonal antibodies (MAb) were raised against a testicular membrane fraction from 18-day post coitum (p.c.) rat testes. One antibody, designated 4B6.3E10 (mu, kappa), was obtained which specifically reacted with gonocytes in the fetal testis. No significant cross-reactivity with other tissues from the 18-day p.c. embryo was found. MAb 4B6.3E10 was reactive with rat gonocytes from 17-day p.c. until the day of birth. Germ cells at later stages of testis development did not show any labelling. The epitope recognized by 4B6.3E10 is a carbohydrate as periodate treatment leads to a loss of reactivity of the antibody. By SDS-PAGE and Western blotting of proteins extracted from a testicular membrane fraction from 18-day p.c. testes, MAb 4B6.3E10 was found to recognize at least 3 protein moieties with apparent molecular weights in the ranges of 80-100, 120, 160-180 kDa (either under reducing- or non-reducing conditions). The results suggest that MAb 4B6.3E10 recognizes a specific differentiation marker for fetal rat gonocytes.

Immunohistochemical distribution of oestrogen and progesterone receptors and tissue concentrations of oestrogens in the cervix of non-pregnant cows.

An immunohistochemical study of the expression of oestrogen (ER) and progesterone receptors (PR) in different regions along the longitudinal and vertical axes of the cervix of non-pregnant cows was performed. Animals were separated into two groups depending on the presence or absence of a functional corpus luteum in their ovaries, as indicated by blood progesterone concentrations. The high progesterone group (HP4) had serum progesterone concentrations > 2.0 ng mL(-1) (n = 6) and the low progesterone group (LP4) had serum progesterone concentrations < or = 0.5 ng mL(-1) (n = 4). Significantly higher concentrations of oestrogen were found in the cervical tissue of animals in the LP4 group than those in the HP4 group (473 +/- 53 v.149 +/- 46 pg g(-1) wet weight; P < 0.01). Furthermore, there was a significant effect of tissue layer (epithelium to deep stroma) on the number of ER (P < 0.01) and PR (P < 0.05) immunoreactive nuclei per 1000 cells. For both ER and PR the proportion of cells expressing the receptor increased from epithelium to subepithelial stroma (P < 0.01) and from subepithelium to deep stroma (ER P < 0.05; PR P =0.061). When the number of receptor-positive cells were expressed per mm2 tissue, differences between the subepithelial stroma and the deep stroma became even more marked. In addition, the vaginal part of the cervix had significantly more (P < 0.01) ER and PR immunoreactive nuclei per 1000 cells than the uterine part, but these differences were no longer apparent when a correction was made for cell density. There was no relationship between progesterone status of the animals, nor local tissue oestrogen concentrations and ER or PR immunoreactivity in the cervix of these non-pregnant cows. Instead, a strong relationship between both longitudinal and vertical positioning of tissue in the cervix and expression of both receptor types was shown. In addition, a strong correlation between ER and PR expression in the subepithelial stroma (R = 0.85, P < 0.01) and the deep stroma (R = 0.83 P < 0.01) was evident. In conclusion, these results demonstrate that in studies of steroid hormone receptor expression in the cervix, careful description of sampling site and depth are necessary if the results are to be interpreted meaningfully.

Irregularly shaped inclusion cysts display increased expression of Ki67, Fas, Fas ligand, and procaspase-3 but relatively little active caspase-3.

Human ovarian cancers are thought to arise from sequestered ovarian surface epithelial (OSE) cells that line the wall of inclusion cysts. Nevertheless, the early events toward neoplasia are not well understood. In this study, immunoreactivity for apoptotic proteins in human OSE of control and tumor ovarian sections was examined. Ki67, a marker for cell proliferation, was generally absent in the flat-to-cuboidal OSE cells on the ovarian surface and in regularly shaped inclusion cysts. Fas, Fas ligand, and caspase-3, components of the apoptotic pathway, were also largely absent. Ki67, Fas, Fas ligand, and procaspase-3 expression, though not active caspase-3 expression, was more frequently observed in epithelial cells lining irregularly shaped inclusion cysts, particularly in the columnar and Müllerian-like OSE cell types that resembled ovarian tumor OSE cells. Immunoreactivity for these factors as well as active caspase-3 was found frequently in ovarian tumors. We postulate that the appearance of the Fas system and its related proteins in sequestered columnar OSE cells of irregularly shaped inclusion cysts may contribute to balance cell growth with cell death, although little active caspase-3 expression was observed. Further studies are required to identify whether inhibition of apoptosis in inclusion cysts is an early event in ovarian carcinogenesis.

Identification of markers for precursor and leydig cell differentiation in the adult rat testis following ethane dimethyl sulphonate administration.

Administration of ethane dimethane sulphonate (EDS) to adult rats results in the destruction of all Leydig cells, followed by a complete regeneration. We investigated this regeneration process in more detail, using different markers for precursor and developing Leydig cells: the LH receptor, 3beta-hydroxysteroid dehydrogenase (3beta-HSD), transforming growth factor alpha (TGFalpha), and a new marker for Leydig cell maturation, relaxin-like factor (RLF). LH receptor immunoreactivity was found in Leydig cell-depleted testes at 3 and 8 days after EDS administration. The positive (precursor) cells had a mesenchymal-like morphology. The number of LH receptor-positive cells 8 days after EDS administration was 15 +/- 4 per 500 Sertoli cell nuclei. Fifteen days after EDS administration, the first new Leydig cells could be observed. These cells stained positively with both the antibodies against the LH receptor and 3beta-HSD, while some cells also stained positively for TGFalpha. After EDS administration, RLF mRNA disappeared from the testis and reappeared again at the time of the appearance of the first Leydig cells. Concomitant with the increase in the number of Leydig cells, the number of RLF-expressing cells increased. The observations of the present study give further support to the hypothesis that Leydig cell development in the prepubertal testis, and in the adult testis following EDS administration, takes place along the same cell lineage and suggest, therefore, that the adult EDS-treated rat can serve as a model for studying the adult-type Leydig cell development that normally occurs in the prepubertal rat testis.

Apoptosis in the rat spermatogenic epithelium following androgen withdrawal: changes in apoptosis-related genes.

Programmed cell death is an important regulatory event in spermatogenesis. However, the molecular events governing apoptosis have not been characterized. Using the Leydig cell-specific toxin ethane dimethanesulfonate (EDS) to withdraw androgen support, we have investigated the relationship between apoptosis and apoptosis-related genes. Adult male Sprague-Dawley rats were injected (i.p.) with 100 mg/kg EDS and killed at times of androgen depletion 2, 5, and 8 days postinjection. A 24-fold increase in the apoptotic index 8 days after EDS administration was demonstrated in tissue sections by in situ end-labeling of fragmented DNA. Leydig cell death and androgen withdrawal were confirmed by the absence of 3beta-hydroxysteroid dehydrogenase in testes from animals treated with EDS for 2 days. After androgen withdrawal, there were no significant changes in the levels of clusterin, Bcl-xl, Bak, and Bad. However, the expression of Bcl-2 and Bax was up-regulated at 8 days after EDS administration. The induction of Bax at this time suggests that it may play a role in germ cell apoptosis following androgen withdrawal. The concomitant elevation in Bcl-2 expression may represent a survival mechanism for the remaining germ cells. There was also a decline in the expression of Fas-L and Fas-R in the pachytene spermatocytes and spermatids. Fas-R was also present in Sertoli cells, although Fas-L staining was minimal. As the colocalization of Fas-L and Fas-R correlates with the germ cell types that die in response to androgen withdrawal, the potential exists for apoptosis in the rat spermatogenic epithelium to be regulated by the Fas pathway.

Ontogeny and localization of an oncostatin M-like protein in the rat testis: its possible role at the start of spermatogenesis.

Oncostatin M (OSM), a member of the interleukin 6 family of cytokines, was found to be highly expressed in the late fetal and early neonatal rat testis, as well as in the maturing and adult testis. Two different forms of OSM were observed, one of M(r) 22,000 and the other of M(r) 36,000. In the prepubertal rat testis [19 days post coitum, 8 days post partum (dpp), and 15 dpp], the form with the higher molecular weight prevailed, whereas in the maturing testis (30 dpp, 45 dpp, and 12 weeks post partum), a shift toward the lower molecular weight form was observed, as well as a decrease in its relative amount. By immunohistochemistry on testicular sections, OSM-specific immunostaining was observed in the interstitial tissue at every age studied. In contrast, OSM immunoreaction was localized in the Sertoli cells exclusively around the start of spermatogenesis, being strongest at 3 dpp. In vitro studies revealed that neonatal Sertoli cells produce OSM. The possible role of OSM at the start of spermatogenesis was investigated by using a coculture of Sertoli cells and gonocytes isolated from newborn rats. OSM significantly increased the survival of both Sertoli cells and gonocytes in a dose-dependent manner. The proliferative activity of the Sertoli cells was not affected by OSM, whereas that of gonocytes was increased by almost 60% after 6 days of culture. Comparison of the effect of OSM on these cocultures with other members of the interleukin 6 family of cytokines demonstrated that this factor is more potent than leukemia inhibitory factor or ciliary neurotrophic factor. On the basis of these findings, it can be concluded that OSM is present in the rat testis, and it is likely to play an important role at the start of spermatogenesis.

Survival and proliferation of rat gonocytes in vitro.

Quiescent gonocytes were isolated from fetal testes of rat 18-day post coitum and cultured alone or on monolayers of somatic cells from different origins. The gonocytes specifically adhered to Sertoli cells, isolated from 21 to 23-day-old rat testes; this adherence was necessary for their survival in vitro. Addition of follicle-stimulating hormone and testosterone to these cultures did not increase the viability of the gonocytes. Serum was found to be deleterious to the germ cells. Electron-microscopic examination of Sertoli-cell-gonocyte co-cultures revealed the presence of numerous adhesion plaques between these cells, indicating that Sertoli cells and gonocytes are able to communicate in vitro. Gonocytes, in co-culture with Sertoli cells, were viable for at least 9 days. The gonocytes did not spontaneously resume proliferation. The simple culture system described in the present paper should be useful in studying the nature of the factors that are responsible for sending the quiescent gonocytes into the cell cycle and for stimulating the formation of A spermatogonia, a process characterizing the start of spermatogenesis.

Regional differences in water content, collagen content, and collagen degradation in the cervix of nonpregnant cows.

The cow could be a suitable model for studies concerning functional changes of the cervix. However, as in many species, the bovine cervix becomes softer in texture during the follicular phase of the estrous cycle compared to the luteal phase. In the present study, we explored if changes in the collagen network take place that could be responsible for this phenomenon and if regional differences in water content, collagen content, and collagen degradation along the cross-sectional and longitudinal axes of the cervix were present. Two groups of nonpregnant animals with different progesterone status were studied. One group (n = 11) was under high progesterone influence, and the other group (n = 12) was under low progesterone influence. The water content was derived from the weight of the samples before and after lyophilization. The collagen content and the ratio of collagenous to noncollagenous proteins (hydroxyproline:proline ratio) were determined by performing amino acid analysis on hydrolyzed samples using high-performance liquid chromatography. Collagen denaturation was quantified with a colorimetric assay by determining the amount of hydroxyproline released from samples treated with alpha-chymotrypsine. The water content of the superficial layer of the submucosa was always significantly (P < 0.01) higher than the water content of the deep layer in the vaginal, mid, and uterine segments, but this was unrelated to the progesterone status of the animals. No effect of the tissue layers or of the progesterone status of the animals on the collagen content was observed, but an effect of segment was noted. The collagen content (mug/mg dry wt) in the vaginal segment of the cervix was significantly higher than in the mid (P < 0.05) and the uterine (P < 0.01) segments. The hydroxyproline:proline ratio showed the same pattern as the collagen content. The percentage of collagen denaturation in the superficial layer was always significantly (P < 0.01) higher than that in the deep layer, but no effect of the progesterone status or of the segment along the longitudinal axis was seen. It is concluded that regional differences in collagen biochemistry are present in the cervix of nonpregnant cows, which may account for the difference in firmness of different parts along the circular or the longitudinal axis of the cervix. However, differences in texture of the cervix between the two groups of cows could not be explained by differences in the collagen content, percentage of collagen denaturation, or water content.