Mice carrying a null mutation for either of the two cyclooxygenase (COX) isoenzymes, necessary for prostanoid production, exhibit several isotype-specific reproductive abnormalities. Mice deficient in COX-1 are fertile but have decreased pup viability, whereas mice deficient in COX-2 fail to ovulate and have abnormal implantation and decidualization responses. The present study identifies the specific contribution of each COX isoenzyme in hypothalamic, pituitary, and ovarian function and establishes the pathology and rescue of the anovulatory syndrome in the COX-2-deficient mouse. In both COX-1- and COX-2-deficient mice, pituitary gonadotropins were selectively increased, whereas hypothalamic LHRH and serum gonadotropin levels were similar to those in wild-type animals (+/+). No significant differences in serum estrogen or progesterone were noted among the three genotypes. Exogenous gonadotropin stimulation with PMSG and hCG produced a comparable 4-fold increase in ovarian PGE2 levels in wild-type and COX-1(-/-) mice. COX-2(-/-) mice had no increase in PGE2 over PMSG-stimulated levels. Wild-type and COX-1(-/-) mice ovulated in response to PMSG/hCG; very few COX-2(-/-) animals responded to this regimen. The defect in ovulation in COX-2 mutants was attributed to both an abnormal cumulus oophorum expansion and subsequent stigmata formation. Gonadotropin stimulation and concurrent treatment with PGE2 or interleukin-1beta resulted in ovulation of COX-2(-/-) mice comparable to that in COX-2(+/+), whereas treatment with PGF2alpha was less effective. Collectively, these data demonstrate that COX-2, but not COX-1, is required for the gonadotropin induction of ovarian PG levels; that COX-2-related prostanoids are required for stabilization of the cumulus oophorum during ovulation; and that ovulation can be restored in the COX-2(-/-) animals by simultaneous treatment with gonadotropins and PGE2 or interleukin-1beta.

We have recently reported that a subpopulation of galanin (GAL)-immunoreactive perikarya in the preoptic area near the organum vasculosum of the lamina terminalis (OVLT) has morphological characteristics similar to those of LHRH-containing neurons. In fact, both peptides are colocalized in those neurons in the male rat brain. In these studies we describe sexual differences in the incidence of neurons colocalizing GAL and LHRH in this region. In male rats, about 20% of LHRH-immunoreactive cells in the diagonal band of Broca and the medial preoptic area are also immunoreactive for GAL. In contrast, in female rats, about 65% of LHRH-containing perikarya near the OVLT are immunostained for GAL. In addition, GAL and LHRH levels were measured in tissue extracts containing either the OVLT and surrounding areas or the median eminence. The data indicate that the preoptic area, including the OVLT, contains a significantly higher amount of GAL in females killed in proestrus than in those killed in estrus. The amount of GAL measured in males is lower than that in the proestrous females, but somewhat greater than that present in the estrous females. There were no significant differences among the three groups in LHRH content in this region. The GAL content of the median eminence was the highest in proestrous females, followed by estrous females and males. The LHRH content in the median eminence was not significantly different among the three groups. In conclusion, these observations indicate that the coexpression of GAL with LHRH in a discrete subpopulation of LHRH-producing neurons is a sex-related phenomenon. The fact that changes in GAL levels in specific regions can be seen at different times during the estrous cycle suggests that gonadal steroids, possibly estrogens, may physiologically regulate the expression of GAL in this subpopulation of LHRH neurons.

The incidence of colocalization of galanin (GAL) in luteinizing hormone-releasing hormone (LHRH) neurons is 4-to 5-fold higher in female than male rats. This fact and the finding that the degree of colocalization parallels estradiol levels during the estrous cycle suggest that GAL is an estrogen-inducible product in a subset of LHRH neurons. To analyze further this paradigm we evaluated the effects of gonadectomy and steroid replacement therapy in male and female rats. Ovariectomy resulted in a significant decrease in the number of cells colocalizing LHRH and GAL, whereas estradiol replacement to such animals restored the incidence of colocalization to that observed in controls. In males, however, estradiol treatment failed to enhance the incidence of colocalization of GAL and LHRH, indicating, therefore, that the colocalization of these peptides is gender-determined. This possibility-i.e., gender-specific determination of LHRH neurons coexpressing GAL-was evaluated by neonatal manipulation of hypothalamic steroid imprinting. As mentioned above, male rats did not respond to estrogen or testosterone by increasing GAL/LHRH colocalization as females did. Neonatally orchidectomized rats, whose hypothalami have not been exposed to testosterone during the critical period, when treated with estrogen in adulthood showed an increase in colocalization of GAL and LHRH similar to that seen in female animals. These observations indicate that the colocalization of LHRH/GAL is neonatally determined by an epigenetic mechanism that involves the testis. In summary, this sex difference in the incidence of colocalization of GAL and LHRH represents a unique aspect of sexual differentiation in that only certain phenotypic characteristics of a certain cellular lineage are dimorphic. The subpopulation of LHRH neurons that also produces GAL represents a portion of the LHRH neuronal system that is sexually differentiated and programed to integrate, under steroidal control, a network of LHRH neurons that could synchronize their activity to control the estrous cycle in rats.Recent observations from our laboratory indicate that a subpopulation of luteinizing hormone-releasing hormone (LHRH) neurons in the rat brain also expresses galanin (GAL) (1). The incidence of colocalization exhibits a striking sex difference-i.e., in females =80% of the LHRH neurons contain GAL, whereas in males only 10-15% of these neurons synthesize this peptide (2). In situ hybridization studies have corroborated our earlier findings since GAL mRNA is colocalized with the decapeptide LHRH in LHRH neurons (3). It has been reported that estradiol has a profound effect on GAL gene expression in the pituitary and the hypothalamus (see ref. 4 for a review). In this respect, we have observed differential coexpression ofGAL in LHRH neuronsThe publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.during the estrous c...