Effects of aging on gene expression and mitochondrial DNA in the equine oocyte and follicle cells.

We hypothesised that advanced mare age is associated with follicle and oocyte gene alterations. The aims of the study were to examine quantitative and temporal differences in mRNA for LH receptor (LHR), amphiregulin (AREG) and epiregulin (EREG) in granulosa cells, phosphodiesterase (PDE) 4D in cumulus cells and PDE3A, G-protein-coupled receptor 3 (GPR3), growth differentiation factor 9 (GDF9), bone morphogenetic protein 15 (BMP15) and mitochondrial (mt) DNA in oocytes. Samples were collected from dominant follicles of Young (3-12 years) and Old (≥20 years) mares at 0, 6, 9 and 12h after administration of equine recombinant LH. LHR mRNA declined after 0h in Young mares, with no time effect in Old mares. For both ages, gene expression of AREG was elevated at 6 and 9h and EREG was expression was elevated at 9h, with higher expression in Old than Young mares. Cumulus cell PDE4D expression increased by 6h (Old) and 12h (Young). Oocyte GPR3 expression peaked at 9 and 12h in Young and Old mares, respectively. Expression of PDE3A increased at 6h, with the increase greater in oocytes from Old than Young mares at 6 and 9h. Mean GDF9 and BMP15 transcripts were higher in Young than Old, with a peak at 6h. Copy numbers of mtDNA did not vary over time in oocytes from Young mares, but a temporal decrease was observed in oocytes from Old mares. The results support an age-associated asynchrony in the expression of genes that are essential for follicular and oocyte maturation before ovulation.

KiSS-1 peptide induces release of LH by a direct effect on the hypothalamus of ovariectomized ewes.

The objectives of this study were: to determine if peripheral administration of the ovine C-terminal decapeptide (oKiSS-1) increases secretion of LH and FSH in ovariectomized (OVX) ewes; to evaluate the effect of ovarian steroid hormones on the predicted increase in gonadotropin secretion induced by oKiSS-1; to establish if the stimulatory effect of oKiSS-1 is due to a direct action on the hypothalamus or was mediated directly at the anterior pituitary gland. Intravenous injection of 3mg oKiSS-1 to OVX ewes (n=6) increased (P<0.01) basal and mean serum concentrations of LH. The amplitude and the area under the curve (AUC) of LH induced by oKiSS-1 was greater (P<0.01) than the highest pulse of LH detected before oKiSS-1. Pre-treatment with steroid hormones (1mg progesterone plus 50 microg estradiol-17beta for 3 days) in OVX ewes (n=6) reduced (P<0.1) the magnitude of the stimulatory effect of oKiSS-1 on LH secretion. In these animals, oKiSS-1 increased (P<0.1) basal and mean concentrations of LH, and the amplitude of the pulse of LH induced by oKiSS-1, but not the AUC (P>0.1). In our hands, oKiSS-1 failed to increase serum concentration of FSH at the dose used in this study. Administration of GnRH antiserum completely suppressed the secretory patterns of LH and prevented the increase in secretion of LH induced by oKiSS-1. We conclude that the rapid increase in secretion of LH induced by peripheral administration of oKiSS-1 is the result of a direct action on the hypothalamus.

Gonadotropin-releasing hormone induction of extracellular-signal regulated kinase is blocked by inhibition of calmodulin.

Our previous studies demonstrate that GnRH-induced ERK activation required influx of extracellular Ca2+ in alphaT3-1 and rat pituitary cells. In the present studies, we examined the hypothesis that calmodulin (Cam) plays a fundamental role in mediating the effects of Ca2+ on ERK activation. Cam inhibition using W7 was sufficient to block GnRH-induced reporter gene activity for the c-Fos, murine glycoprotein hormone alpha-subunit, and MAPK phosphatase (MKP)-2 promoters, all shown to require ERK activation. Inhibition of Cam (using a dominant negative) was sufficient to block GnRH-induced ERK but not c-Jun N-terminal kinase activity activation. The Cam-dependent protein kinase (CamK) II inhibitor KN62 did not recapitulate these findings. GnRH-induced phosphorylation of MAPK/ERK kinase 1 and c-Raf kinase was blocked by Cam inhibition, whereas activity of phospholipase C was unaffected, suggesting that Ca2+/Cam modulation of the ERK cascade potentially at the level of c-Raf kinase. Enrichment of Cam-interacting proteins using a Cam agarose column revealed that c-Raf kinase forms a complex with Cam. Reconstitution studies reveal that recombinant c-Raf kinase can associate directly with Cam in a Ca2+-dependent manner and this interaction is reduced in vitro by addition of W7. Cam was localized in lipid rafts consistent with the formation of a Ca2+-sensitive signaling platform including the GnRH receptor and c-Raf kinase. These data support the conclusion that Cam may have a critical role as a Ca2+ sensor in specifically linking Ca2+ flux with ERK activation within the GnRH signaling pathway.

Activin responsiveness of the murine gonadotropin-releasing hormone receptor gene is mediated by a composite enhancer containing spatially distinct regulatory elements.

The promoters of mouse and rat GnRH receptor (GnRHR) genes differ markedly in regard to activin regulation. Activin stimulates the mouse GnRHR promoter, although it has no impact on that of the rat. To test whether this difference was due to a single nucleotide change in the rat GnRHR activating sequence (GRAS) homolog, we tested a mouse promoter with the rat GRAS homolog and a rat promoter with intact mouse GRAS. The single change in GRAS eliminated activin responsiveness of the mouse GnRHR promoter; however, intact mouse GRAS did not confer activin responsiveness to the rat promoter. Thus, although necessary, GRAS is not sufficient for activin responsiveness of the murine GnRHR promoter. Use of chimeric rat and mouse promoters led to the identification of a 36-bp region residing immediately downstream of GRAS that is necessary for activin responsiveness of the mouse GnRHR gene promoter. Scanning mutagenesis of the 36-bp region localized the functional boundaries of the key regulatory element to adjacent TAAT motifs. The presence of tandem TAAT motifs, the core binding site for multiple members of the homeodomain family of binding proteins, raised the possibility that this region represented a binding site for a homeodomain protein. This region displayed specific binding to a recombinant homeodomain of LHX2. We suggest that GRAS and the downstream activin regulatory element together define a unique and complex activin/TGFbeta-responsive "enhanceosome" whose functional attributes depend on the binding of multiple classes of transcription factors at spatially distinct sites in the promoter of the murine GnRHR gene.

Multiple core homeodomain binding motifs differentially contribute to transcriptional activity of the murine gonadotropin-releasing hormone receptor gene promoter.

Multiple homeodomain (Hbox) proteins have been shown to organize expression of key markers of gonadotropes. Nine putative Hbox-binding sites, characterized by the homeospecific TAAT motif, are located within the proximal 600 bp of the murine GnRHR promoter. Homeoproteins bind separate Hbox sites within this promoter, supporting basal- and endocrine-directed transcription. The function of the most proximal sites (Hbox1 and Hbox2) in the murine GnRHR is unknown; thus, understanding of the global contribution of homeospecific TAAT sites to promoter function is incomplete. Site-directed mutagenesis revealed that loss of Hbox2 reduced promoter activity in a cell-specific manner, having no effect in alphaT3-1 cells but reducing promoter function in LbetaT2 cells, another gonadotrope-derived cell line representing a later developmental stage. In contrast, eliminating Hbox1 reduced basal activity in both lines. This region displayed specific binding to homeoprotein Oct-1. Mutagenesis of a previously identified Oct-1-binding site in concert with Hbox1 led to further reduction in activity. We suggest that the two most proximal homeodomain-binding sites in the murine GnRHR promoter may regulate the promoter in a developmentally dependent fashion and that Oct-1 acts at multiple but distinct TAAT sites to support basal transcription.

A specific helical orientation underlies the functional contribution of the activin responsive unit to transcriptional activity of the murine gonadotropin-releasing hormone receptor gene promoter.

Activin responsiveness of the murine GnRH receptor (GnRHR) gene promoter requires two spatially distinct regulatory elements termed the GnRH receptor activating sequence or GRAS and the downstream activin regulatory element or DARE. While GRAS interacts with multiple transcription factors, DARE activity requires tandem homeodomain binding motifs (TAAT) and displays specific binding to the LIM homeodomain protein LHX3. Herein, we find that both the murine GnRHR gene promoter and DARE fused to a minimal heterologous promoter are responsive to LHX3 overexpression. A dominant-repressor of LHX3 attenuates transcriptional activity of the murine GnRHR gene promoter but had no impact on activin responsiveness. Thus, LHX3 would not appear to be the protein mediating activin responsiveness of this promoter. Within DARE itself, the tandem TAAT motifs are separated by 4 bp. Although this arrangement differs from the prototypical P2 or P3 binding sites characterized for paired-like homeodomain proteins and from the directly abutting TAAT motifs found for LHX3, a LIM-class homeodomain protein, we find that separation of the TAAT sites by 5 and 10 bp decreases GnRHR promoter activity to a level similar to promoters containing loss of function mutations in either the proximal or distal TAAT motif. Thus, the juxtaposition of the TAAT sites is critical for the functional activity of DARE. That activin responsiveness of the GnRHR promoter requires both GRAS and DARE suggests that these elements may be both functionally and structurally coupled. As to the latter, GRAS and DARE are separated by 20 bp, thus placing the elements on the same side of the helical backbone. To determine if this spatial organization is functionally relevant, multiples of 5 bp were inserted or deleted between GRAS and DARE. Any insertion or deletion that resulted in a half-turn alteration in the helical positioning between the two elements reduced promoter activity. Thus, an important spatial relationship underlies functional cooperation between GRAS and DARE and the emergence of a complex activin responsive unit (ARU) within the mouse GnRHR promoter.

Differential impact of intracellular carboxyl terminal domains on lipid raft localization of the murine gonadotropin-releasing hormone receptor.

The mammalian type I GNRH receptor (GNRHR) is unique among G protein-coupled receptors (GPCRs) because of the absence of an intracellular C-terminus. Previously, we have found that the murine GNRHR is constitutively localized to low-density membrane microdomains termed lipid rafts. As such, association of the GNRHR with lipid rafts may reflect both a loss (C-terminus) and a gain (raft association address) of structural characteristics. To address this, we fused either the full-length C-terminus from the nonraft-associated LH receptor (LHCGR; GNRHR-LF) or a truncated (t631) LHCGR C-terminus to the GNRHR. These chimeric receptors are trafficked to the plasma membrane, bind ligand, and display increased agonist-induced receptor internalization, but they do not partition into lipid rafts. Thus, a heterologous C-terminus from a nonraft-associated GPCR redirects localization of the GNRHR to nonraft domains. In contrast to the murine GNRHR, the catfish GNRHR (cfGNRHR) possesses an intracellular C-terminus. We found that the cfGNRHR was localized to lipid rafts and that the cfGNRHR C-terminus did not alter raft localization of the mammalian receptor. Consistent with placement in different lipid microenvironments within the plasma membrane, fluorescence recovery after photobleaching revealed different lateral diffusion phenotypes of the raft-associated GNRHR and cfGNRHR versus the nonraft-associated GNRHR-LF fusion protein. We conclude that whereas an intracellular C-terminus is capable of redirecting the GNRHR to nonraft compartments, this is not a generalized feature of GPCR C-terminal tails. Thus, constitutive raft localization of the GNRHR is not simply a result of the loss of an intracellular C-terminus.

Constitutive localization of the gonadotropin-releasing hormone (GnRH) receptor to low density membrane microdomains is necessary for GnRH signaling to ERK.

Specialized membrane microdomains known as lipid rafts are thought to contribute to G-protein coupled receptor (GPCR) signaling by organizing receptors and their cognate signaling molecules into discrete membrane domains. To determine if the GnRHR, an unusual member of the GPCR superfamily, partitions into lipid rafts, homogenates of alpha T3-1 cells expressing endogenous GnRHR or Chinese hamster ovary cells expressing an epitope-tagged GnRHR were fractionated through a sucrose gradient. We found the GnRHR and c-raf kinase constitutively localized to low density fractions independent of hormone treatment. Partitioning of c-raf kinase into lipid rafts was also observed in whole mouse pituitary glands. Consistent with GnRH induced phosphorylation and activation of c-raf kinase, GnRH treatment led to a decrease in the apparent electrophoretic mobility of c-raf kinase that partitioned into lipid rafts compared with unstimulated cells. Cholesterol depletion of alpha T3-1 cells using methyl-beta-cyclodextrin disrupted GnRHR but not c-raf kinase association with rafts and shifted the receptor into higher density fractions. Cholesterol depletion also significantly attenuated GnRH but not phorbol ester-mediated activation of extracellular signal-related kinase (ERK) and c-fos gene induction. Raft localization and GnRHR signaling to ERK and c-Fos were rescued upon repletion of membrane cholesterol. Thus, the organization of the GnRHR into low density membrane microdomains appears critical in mediating GnRH induced intracellular signaling.