Immunohistochemical characterization of nonhuman primate ovarian sex cord-stromal tumors.

This study evaluates the immunoreactivity of 12 sex cord-stromal tumors of nonhuman primates (11 granulosa cell tumors and 1 luteoma). The markers selected are used in the characterization of gonadal tumors in dogs and other species, including cytokeratins AE1/AE3, GATA-4, inhibin-α, neuron-specific enolase, protein gene product 9.5, and vimentin. A normal nonhuman primate ovary was used as a control and to optimize immunolabeling. Staining was graded as follows: 0 (nonstaining), 1+ (< 10% positive cells), 2+ (10%-50% positive cells), and 3+ (> 50% positive cells). Calretinin, GATA-4, neuron-specific enolase, and vimentin were the most consistently expressed markers (12 of 12). Cytokeratins AE1/AE3 were also consistently expressed (11 of 12). Inhibin-α and protein gene product 9.5 were expressed in 8 and 10 sex cord-stromal tumors, respectively. Results indicate that immunoreactivity of nonhuman primate sex cord-stromal tumors is similar to that observed in other species and that calretinin, GATA-4, and neuron-specific enolase are the most consistently expressed markers in nonhuman primate sex cord-stromal tumors.

Hyperandrogenic states in pregnancy.

Hyperandrogenic states in pregnancy are almost always the result of a condition that arises during pregnancy. The onset of virilization symptoms is often very fast. The mother is protected against hyperandrogenism by a high level of SHBG, by placental aromatase and a high level of progesterone. The fetus is protected from the mother's hyperandrogenism partly by the placental aromatase, that transforms the androgens into estrogens, and partly by SHGB. Nevertheless there is a significant risk of virilization of the female fetus if the mother's hyperandrogenic state is serious. The most frequent cause of hyperandrogenic states during pregnancy are pregnancy luteoma and hyperreactio luteinalis. Hormonal production is evident in a third of all luteomas, which corresponds to virilization in 25-35 % of mothers with luteoma. The female fetus is afflicted with virilization with two thirds of virilized mothers. Hyperreactio luteinalis is created in connection with a high level of hCG, e.g. during multi-fetus pregnancies. This condition most frequently arises in the third trimester, virilization of the mother occurs in a third of cases. Virilization of the fetus has not yet been described. The most serious cause of hyperandrogenism is represented by ovarian tumors, which are fortunately rare.

Stromal luteoma: report of a rare case.

Steroid cell tumours (SCTs) account for less than 0.1% of all ovarian tumours. Three major categories ofSCTs include (1) stromal luteoma, (2) steroid cell tumour not otherwise specified and (3) Leydig cell tumours that do not have another component. Stromal luteomas constitute 20% of SCTs. They usually occur in postmenopausal women and about 60% present with estrogenic manifestations. We report a case of stromal luteoma in an elderly lady who presented with postmenopausal bleeding secondary to endometrial hyperplasia. An interesting finding in our case was the presence of many eosinophilic hyaline globules scattered throughout the tumour, the significance of which remains to be determined.

Differential gonadotropin releasing hormone (GnRH) expression, autoregulation and effects in two models of rat luteinized ovarian cells.

GnRH has been suggested to participate in corpus luteum function. Here we studied the expression of GnRH mRNA and peptide in two models of rat luteinized tissues: ovarian cells from PMSG-hCG treated prepubertal rats (SPO) and from intrasplenic ovarian tumors (Luteoma). A GnRH autoregulatory effect was evaluated as well as its action on cell proliferation and apoptosis. GnRH mRNA was present in SPO, isolated corpora lutea from SPO and Luteoma from 1 week to 7 months of development. In vitro cultures of Luteoma cells expressed 2-fold higher GnRH mRNA and 10-fold higher GnRH peptide than SPO cells. Buserelin (GnRH analog) increased GnRH mRNA and peptide expression in SPO but not in Luteoma cells. While basal proliferation was very low in Luteoma cells, SPO cells showed a significant increase in cell number by both the thymidine and the MTS methods after 72 h in culture. Buserelin induced a decrease in cell number in both cell types to a similar degree. Although basal apoptosis levels were higher in SPO than in Luteoma cells, Buserelin-induced apoptosis was only detected in Luteoma cells after 48 h treatment. These results show that the two types of rat, luteinized tissues, Luteoma and SPO, markedly differed in some intrinsic properties and in their local GnRH systems. Luteoma cells proliferate very weakly, express and secrete high amounts of GnRH, do not show an autoregulatory effect and respond to the decapeptide with apoptosis stimulation. In contrast SPO cells proliferate significantly, secrete low levels of GnRH but possess a positive, autoregulatory mechanism and respond to GnRH stimulation with impairment of proliferation.

Gonadotropin-releasing hormone signaling pathways in an experimental ovarian tumor.

Previous results showed that GnRH signaling is altered in cells from rat luteinized ovarian tumors (tumor group) because it did not activate the phospholipase C pathway, in contrast to control ovarian cells from superovulated prepubertal rats (SPO). In the present work, alternate GnRH-induced second messengers such as phospholipase A(2) and phospholipase D activation, cAMP production, ERK1/2 phosphorylation, and the presence of G proteins were evaluated to determine GnRH mechanism of action in tumor cells. G proteins examined were present in both cell types. Buserelin, a GnRH agonist, (1, 10, and 100 ng/ml) increased phosphatidylethanol in SPO, indicating phospholipase D activation. Only 100 ng/ml buserelin induced a significant response in the tumor group. Buserelin (100 ng/ml) increased (3)H-arachidonic acid in culture media in SPO, indicating phospholipase A(2) activation; no effect was observed in the tumor group. Buserelin (100 and 1000 ng/ml) induced pertussis toxin-insensitive cAMP increases in both cell types, with similar potencies. In the tumor group, buserelin (100 ng/ml) inhibited human chorionic gonadotropin-induced cAMP and progesterone; this effect was protein kinase C (PKC) dependent (inhibited by GF109203X, a PKC inhibitor). Buserelin (100 and 1000 ng/ml) induced ERK1/2 phosphorylation in both cell kinds. Buserelin-induced ERK1/2 activation was G(i/0) independent and PKC dependent. Only in the tumor group, buserelin-induced ERK1/2 activation was cAMP dependent (abolished by SQ 22536, the adenylyl cyclase inhibitor). Furthermore, dibutyryl cAMP-induced ERK1/2 activation in the tumor group was PKC dependent (inhibited by GF109203X). In conclusion, activation of phospholipases in tumor cells does not seem to mediate GnRH effects. GnRH signaling seems to involve adenylyl cyclase activation, PKC stimulation, and ERK1/2 phosphorylation.

Actions of immunosuppressor drugs on the development of an experimental ovarian tumor.

Immunosuppression has been related to the incidence of tumor apparition, including endocrine tumors. The intrasplenic ovarian tumor (luteoma) is a typical benign endocrine tumor that develops under high gonadotropin stimulation and, from the immunological perspective, is located in a critical organ involved in immune response. To establish if immunosuppression could alter the development of this experimental tumor, the effects of cyclosporin A (CsA) and dexamethasone (Dex) were evaluated. After surgery, tumor-bearing and sham animals were kept without treatment for 4 weeks; thereafter, they were distributed into CsA (25 mg/kg), Dex (0.1 mg/kg), or vehicle (75:25 castor oil:ethanol) groups and were injected on alternate days for 50 days. Body weight was evaluated weekly. Animals were sacrificed after a jugular vein blood sample was obtained. Thymi were weighed. Tumors were measured and placed in formaline for histological studies. Serum luteinizing hormone (LH), follicle-stimulating hormone (FSH), prolactin (PRL), and estradiol were measured by radioimmunoassay. Hematological parameters were determined. CsA induced a significant decrease in survival rates both in tumor-bearing and sham animals (P < 0.01). Dex significantly impaired weight increase in both groups of animals. CsA induced a significant weight loss in sham animals, not observed in tumor-bearing animals. Dex induced thymus weight loss in both groups, whereas CsA induced thymus weight loss only in sham animals. Only Dex induced a decrease in lymphocyte number in both groups. CsA induced an increase in monocyte number only in sham animals. Treatments did not alter LH, FSH, or estradiol, whereas PRL was increased by CsA only in sham rats. Neither Dex nor CsA induced any significant variations in tumor volume, nor did they alter tumor histology. In addition, no visible metastases or alterations in other organs were observed. We conclude that, though immunological parameters were altered by the treatments, immunosuppressor drugs did not condition tumor development. In addition, tumors secrete one or more factor/s that counteract CsA effect.

Alterations in intracellular messengers mobilized by gonadotropin-releasing hormone in an experimental ovarian tumor.

Cells derived from an experimental luteinized ovarian tumor are more sensitive to GnRH endocrine action than control luteal cells. In an attempt to understand the possible causes of the differential sensibility to GnRH action, we examined the number and affinity of GnRH receptors and the second messenger response to GnRH stimulation in both tissues. For GnRH receptor studies membranes were obtained from 4- to 6-week-old ovarian tumors (luteoma) and ovaries from prepubertal rats treated with 25 IU PMSG and 25 IU hCG (SPO) and were incubated with [125I]Buserelin. The number of GnRH receptors were increased in luteoma compared with that in SPO ovaries; dissociation constants were similar in both tissues. GnRH stimulation of second messenger release was assessed in cells obtained from luteoma and SPO ovaries by collagenase treatment. Buserelin (100 ng/ml) induced a significant 35% calcium increase in SPO cells, as determined by the fura-2 method; in luteoma cells no response was observed after buserelin stimulation, although a calcium transient was induced by thapsigargin (0.5 microM), an inhibitor of Ca2+-adenosine triphosphatase associated with the endoplasmic reticulum. The effect of buserelin on inositol phosphates was evaluated after incubation of luteoma and SPO cells with [3H]myoinositol for 48 h. Buserelin induced a 400% increase in inositol trisphosphate in SPO cells. Again, luteoma cells did not respond to buserelin stimulation, although NaF (10 mM), an activator of G proteins coupled to phospholipase C, induced an 800% increase in inositol trisphosphate. Although the number of GnRH receptors is augmented in luteoma cells, justifying an increased endocrine response, neither inositol phosphates nor intracellular calcium were released by a GnRH analog, indicating the uncoupling of GnRH receptors from phospholipase C. These data provide evidence that the transformation of the ovary into a luteoma implies the acquisition of novel characteristics in the GnRH receptor second messenger-generating system.

GnRH receptors and GnRH endocrine effects on luteoma cells.

An ovary implanted into the spleen of an ovariectomized rat develops into a luteinized tumor, growing in response to gonadotrophins. Previously, it was shown that in vivo Buserelin, a gonadotrophin-releasing hormone (GnRH) analog, inhibited tumor growth. To determine if GnRH had a direct effect on tumor cells, the presence of GnRH receptors as well as the endocrine effects of buserelin were studied on tumoral tissue. GnRH receptors were present in luteoma in similar concentrations and dissociation constant (Kd) to control estrous ovaries. In vivo treatment with buserelin did not modify luteoma GnRH receptors. In organ incubations, luteoma secreted significantly higher estradiol and lower progesterone than estrous ovaries; addition of buserelin did not modify steroid secretion. The same difference in basal steroid secretion between luteoma cells and luteal cells superovulated prepubertal ovaries was observed in cell cultures. Although luteinizing-hormone (LH)-stimulated progesterone in both kinds of cells, buserelin significantly inhibited LH-stimulated progesterone only in luteoma cells. These results describe clear differences in basal steroid secretion between tumoral and normal tissue. Furthermore, they show that luteoma possess GnRH receptors similar to those in normal ovarian tissue, and that GnRH analogs have endocrine effects on these cells. Therefore, a direct effect of buserelin on luteoma cells can be postulated.