GATA depletion impacts insulin-like growth factor 1 mRNA and protein levels in luteinizing porcine granulosa cells.

GATA4 and GATA6 are zinc-finger transcription factors that regulate specific genes involved in steroidogenesis. Using RNA interference (RNAi)-mediated reduction of GATA4 and/or GATA6 with microarray analysis, we aimed to identify novel GATA target genes in luteinizing porcine granulosa cells under vehicle- and cAMP-treated conditions. Microarray analysis identified IGF1 mRNA to be cAMP- and GATA-responsive, and real-time PCR demonstrated that the cAMP-induced increase in IGF1 mRNA was reduced under conditions of GATA6 depletion and GATA4 plus GATA6 depletion, but not GATA4 depletion. Insulin-like growth factor 1 protein levels in media were also decreased by GATA6 or GATA4 plus GATA6 reduction. IGFBP2 and IGFBP4 mRNAs were increased and IGFBP5 mRNA decreased with vehicle and cAMP treatment under GATA4 plus GATA6 RNAi conditions. GATA6 reduction alone increased basal IGFBP4 and decreased IGFBP5 with both vehicle and cAMP, and GATA4 reduction alone lowered cAMP IGFBP5 levels with cAMP. No changes in IGFBP3 mRNA were observed with GATA reduction relative to the control RNAi condition. Levels of insulin-like growth factor binding proteins 2-5 in media as assessed by Western ligand blotting were not altered by GATA reduction. Electromobility gel shift assays with two GATA-containing oligonucleotides of the IGF1 5'-regulatory region showed GATA4 and GATA6 could bind the more proximal GATA-B site. These studies indicate that although GATA4 and GATA6 can bind the porcine IGF1 5'-region, GATA6 is functionally most important for cAMP-stimulated mRNA levels. Using microarray analysis, we identified other mRNAs that were altered by GATA-reduced conditions, including ALDH1, DIO2, and EDNRB. Our findings further support GATA as a coordinator of endocrine/paracrine/autocrine signals in the ovary.

GATA4 reduction enhances 3',5'-cyclic adenosine 5'-monophosphate-stimulated steroidogenic acute regulatory protein messenger ribonucleic acid and progesterone production in luteinized porcine granulosa cells.

Previous studies with cultured granulosa cells implicated GATA4 in gonadotropin regulation of the steroidogenic acute regulatory protein (STAR) gene. Caveats to these prior studies exist. First, GATA4 levels are reduced in granulosa-luteal cells after the LH surge when GATA6 expression is relatively high. Second, STAR mRNA expression is negligible in granulosa cells until after the LH surge. Both exogenous GATA4 and GATA6 can transactivate STAR gene promoter constructs. We used an RNA interference (RNAi) approach to determine the contributions of GATA4 and GATA6 to cAMP analog regulation of the endogenous STAR gene in luteinizing granulosa cells. STAR mRNA was stimulated by cAMP under control RNAi conditions. Surprisingly, GATA4 reduction by its respective RNAi approximately doubled the cAMP induction of STAR mRNA. At 24 h cAMP treatment, this augmentation was abolished by co-down-regulation of GATA4+GATA6. GATA6 down-regulation by itself did not alter STAR mRNA levels. GATA4+GATA6 co-down-regulation elevated basal CYP11A mRNA at 24 h treatment but did not affect its induction by cAMP. Basal levels of HSD3B mRNA were reduced by GATA4 RNAi conditions leading to a greater fold induction of its mRNA by cAMP. Fold cAMP-stimulated progesterone production was enhanced by GATA4 down-regulation but not by GATA4+GATA6 co-down-regulation. These data implicate GATA6 as the facilitator in cAMP-stimulated STAR mRNA and downstream progesterone accumulation under reduced GATA4 conditions. Data also demonstrate that basal levels of GATA4/6 are not required for cAMP induction of the STAR gene. The altered ratio of GATA4 to GATA6 after ovulation may allow GATA6 to enhance STAR mRNA accumulation.

Concerted regulation of the porcine steroidogenic acute regulatory protein gene promoter activity by follicle-stimulating hormone and insulin-like growth factor I in granulosa cells involves GATA-4 and CCAAT/enhancer binding protein beta.

We previously demonstrated that FSH alone or in combination with IGF-I activated the porcine steroidogenic acute regulatory protein gene promoter in a concerted manner in primary cultures of granulosa cells. Studies were undertaken to further delineate cis- and trans-acting elements of the porcine promoter and mechanisms mediating FSH stimulation and its augmentation by IGF-I. Mutation of several putative regulatory elements localized hormone-stimulated activity to the highly conserved GATA site and identified novel nucleotides downstream as a functional CCAAT/enhancer binding protein (C/EBP)beta site. In granulosa cell nuclear extracts, GATA-4 and C/EBPbeta formed a high-molecular-weight complex with an oligonucleotide spanning -76 to -32 bp of the porcine promoter. The intensity of this high-molecular-weight complex was increased in granulosa cell nuclear extracts by treatment with FSH alone and was enhanced with the combination of FSH and IGF-I at 2-3 h of treatment. GATA-4 and C/EBPbeta proteins were uniformly expressed with all treatments at time points associated with increased DNA binding. Treatment (2 h) with FSH alone or FSH + IGF-I increased phosphorylation of GATA-4 on a protein kinase A consensus site. The 38-kDa isoform of C/EBPbeta exhibited greater phosphorylation with FSH + IGF-I treatment. Porcine luteal cell nuclear extracts also demonstrated GATA-4 and C/EBPbeta binding to the -76 to -32 bp region of the promoter providing evidence for their cooperation in vivo. These data indicate that GATA-4 and C/EBPbeta are both required for FSH +/- IGF-I stimulation of the porcine steroidogenic acute regulatory protein gene promoter in homologous granulosa cell cultures.

STARD6 is expressed in steroidogenic cells of the ovary and can enhance de novo steroidogenesis.

STARD6 is a member of the StAR-related lipid transfer (START) domain family of proteins whose function thus far remains obscure. While it recently was shown to facilitate steroidogenesis in a cell-free setting, it has not been localized to steroidogenic cells of normal reproductive tissues. In a recent microarray study, we detected STARD6 mRNA in cultured porcine ovarian granulosa cells which are steroidogenic. In the present study, we examined regulation of STARD6 mRNA in porcine granulosa cultures, and found that it was not regulated by cyclic AMP, but it was reduced by combined knockdown of the transcription factors GATA4 and GATA6. We detected both STARD6 mRNA and protein in fresh granulosa cells and whole antral follicles and different stage corpora lutea of pig. The highest levels were discovered in the mid-luteal phase corpus luteum. Immunolocalization within ovarian tissues indicated robust STARD6 immunoreactivity in steroidogenic cells of the corpus luteum. Relatively lesser amounts of STARD6 signal were found in granulosa cells, theca cells, and oocytes. To test the ability of STARD6 to facilitate de novo steroidogenesis, non-steroidogenic COS-1 cells were co-transfected with components of the P450 cholesterol side-chain cleavage system, enabling them to make pregnenolone, and STARD6. STARD6 increased pregnenolone production by two- to three-fold over the empty vector control. In summary, STARD6 is found in the pig ovary, exhibits the strongest expression in highly steroidogenic luteal cells, and significantly enhances pregnenolone production in transfected COS cells independent of cyclic AMP treatment. Collectively, these findings indicate that STARD6 may contribute to steroidogenesis in ovarian cells, but also suggests other cellular functions that require cholesterol trafficking.

Perinatal exposure to low-dose DE-71 increases serum thyroid hormones and gonadal osteopontin gene expression.

Polybrominated diphenyl ethers (PBDEs) are flame retardants that have been widely used in manufacturing. They are major household and environmental contaminants that bioaccumulate. Humans are exposed primarily through dust inhalation and dietary ingestion of animal products. In animal studies, high doses of penta-brominated diphenyl ethers (penta-BDEs) in the mg/kg body weight (BW) range negatively impact brain development, behavior, memory, circulating thyroid hormone concentrations, the reproductive system and bone development. We investigated the effects of ingestion of a relatively low dose of the penta-BDE mixture DE-71 by pregnant and lactating rats on reproductive and thyroid parameters of the F1 offspring. F0 mothers received 60 µg/kg BW of DE-71 or vehicle daily by gavage from Day 1.5 of pregnancy through lactation (except the day of parturition). F1 pups were sacrificed at 21 d of age or outbred at approximately 80 d of age. Bred F1 females were sacrificed at Day 14.5 of pregnancy or at five months of age. Bred F1 males were sacrificed at five months of age. DE-71 treatment of the mothers affected the F1 females as evidenced by lower body weights at 80 d and five months of age, elevated serum T3 and T4 concentrations at Day 14.5 of pregnancy and increased thyroid gland weight and ovarian osteopontin mRNA at five months of age. Perinatal DE-71 exposure also increased testicular osteopontin mRNA in 21-day-old F1 males. Utilizing a granulosa cell in vitro model, we demonstrated that DE-71 activated the rat osteopontin gene promoter. Our results are the first to demonstrate that PBDEs increase rodent circulating T3 and T4 concentrations and gonadal osteopontin mRNA, and activate the osteopontin gene promoter. These changes may have clinical implications as others have shown associations between human exposure to PBDEs and subclinical hyperthyroidism, and overexpression of ovarian osteopontin has been associated with ovarian cancer.

Expression of CCAAT/enhancer binding proteins alpha and beta in the porcine ovary and regulation in primary cultures of granulosa cells.

CCAAT/enhancer binding proteins alpha and beta (CEBPA/ CEBPB) were evaluated in the porcine ovary during the estrous cycle. CEBPB mRNA was present in antral follicles and was significantly increased in healthy corpora lutea (CL), whereas CEBPA mRNA was constitutively expressed in these structures. Both isoforms of CEBPA (42 and 30 kDa) exhibited greater expression in preovulatory follicles, and the 42-kDa isoform increased in CL, whereas the 30-kDa isoform decreased. All major isoforms of CEBPB (38, 34, and 20 kDa) were expressed, with the 34- and 20-kDa isoforms being more abundant in preovulatory follicles and further increased in CL. The effects of FSH and cAMP analogue on the distribution of CEBP isoforms were evaluated in primary cultures of porcine granulosa cells. FSH and 8-Br-cAMP had little stimulatory effect on isoform distribution, but cAMP treatment for 24 h tended to decrease the 30-kDa form of CEBPA and the 34-kDa form of CEBPB. The 34-kDa form of CEBPB was decreased by the protein kinase A inhibitor H89 at 4 h (with FSH treatment), and by both protein kinase A and phosphatidylinositol 3-kinase inhibitors at 24 h of treatment. In transfected granulosa cells, FSH and cAMP analogue stimulated a CEBP consensus sequence-reporter construct that was blocked by H89. These data implicate protein kinase A as the major regulator of CEBPB isoform distribution and CEBP-mediated transactivation in granulosa cells. The differential expression of specific CEBPA/B isoforms observed in maturing follicles and CL may contribute to changes in follicular cell differentiation and increasing steroidogenic capacity.

Epidermal growth factor-mediated inhibition of follicle-stimulating hormone-stimulated StAR gene expression in porcine granulosa cells is associated with reduced histone H3 acetylation.

Steroidogenic acute regulatory protein (StAR) mediates cholesterol transport into the mitochondria and is essential for ovarian steroidogenesis. Epidermal growth factor (EGF) has been reported to inhibit FSH-stimulated differentiation in porcine granulosa cells. Previous studies have demonstrated FSH stimulates StAR mRNA accumulation and gene promoter activation in granulosa cells. Treatment of granulosa cells with FSH (5 ng/ml, 6 h) increased StAR mRNA, whereas coaddition of EGF (10 ng/ ml) significantly reduced (P < 0.05) FSH-stimulated mRNA accumulation by 62.7% +/- 13.9%. Under these same conditions, FSH-stimulated cAMP accumulation in cultures was unaltered by coincubation with EGF. RNA stability studies showed that cotreatment with FSH and EGF did not alter the StAR mRNA half-life compared with FSH alone, t(1/2) = 1.9-3.8 and 2.7-4.1 h, respectively. EGF significantly inhibited (P < 0.05) FSH-stimulated StAR heterogeneous nuclear RNA levels by 47.6% +/- 6.8 %, implicating a repressive effect on transcription. Surprisingly, EGF (1-50 ng/ml) did not affect FSH stimulation of a 1423-base pair StAR gene promoter-luciferase construct in transient transfection assays in porcine granulosa cells. To evaluate FSH and EGF effects on the endogenous StAR gene, chromatin immunoprecipitation assays were performed in combination with real-time polymerase chain reaction. FSH increased histone H3 acetylation (lysines 9, 14) within the proximal region of the StAR gene promoter and coincubation with EGF blocked this effect. Dimethylation (lysine 9) of histone H3 was not influenced by treatments. In conclusion, EGF repression of FSH-stimulated StAR transcription in porcine granulosa cells is accompanied by reductions in histone H3 acetylation associated with the StAR gene promoter.

GATA-4 and GATA-6 transcription factors: expression, immunohistochemical localization, and possible function in the porcine ovary.

The expression and localization of GATA-4 and GATA-6 mRNAs and proteins were assessed in porcine ovaries at different stages of the estrous cycle. Reverse transcription polymerase chain reaction and Western blot analyses revealed that GATA-4 and GATA-6 transcripts and proteins were strongly expressed in granulosa cells isolated from antral follicles, intact antral follicles, corpora hemorrhagica (CH), and midluteal phase corpora lutea (CL). Immunoblot analyses showed two predominant proteins with molecular masses of approximately 53 and 55 kDa for GATA-4 and one 55-kDa protein for GATA-6. Immunohistochemical studies revealed GATA-4 and GATA-6 nuclear staining in granulosa cells of healthy primordial and primary antral follicles and antral follicle of various sizes. The percentage of immunopositive thecal cell nuclei increased with follicular development. In CH and CL, luteal cells displayed nuclear immunoreactivity for both transcription factors. Regressing CL showed a decrease in GATA-immunopositive cells. Immunoreactivity for GATA-4 and GATA-6 was present in most blood vessels. In electrophoretic mobility shift assays, nuclear protein extracts isolated from granulosa cells and CL exhibited both GATA-4 and GATA-6 binding to a GATA consensus oligonucleotide, with GATA-4 the predominant binding protein. GATA-4 and GATA-6 DNA binding was elevated in granulosa cell nuclear extracts from preovulatory (8-10 mm) follicles. Cotransfection of primary cultures of luteinizing granulosa cells with GATA-4 or GATA-6 expression vectors increased the activity of the porcine steroidogenic acute regulatory protein gene promoter significantly but did not significantly activate the inhibin alpha gene promoter. The detection of GATA-4 and GATA-6 mRNAs and proteins in porcine ovaries and the identification of at least one possible target gene may help to establish roles for these GATA factors in follicular development and luteal function.

Cloning and characterization of porcine ovarian estrogen receptor beta isoforms.

The cDNA for the full-length porcine estrogen receptor beta (ER beta) and an alternatively spliced transcript with a deletion of exon 5 (ER beta delta 5) was cloned from pig ovary. RNase protection assays revealed that ER beta mRNA was expressed in the preovulatory follicles and early, midluteal, and regressing corpora lutea (CL) of eCG +/- hCG-primed gilts. ER beta and ER beta delta 5 transcripts were shown by semiquantitative reverse transcription polymerase chain reaction to be expressed at a ratio of approximately 2:1 in granulosa cells, small, medium, and large antral follicles, and midluteal phase corpora lutea of unprimed animals. Immunoreactive ER beta proteins corresponding to the size of in vitro translated ER beta and ER beta delta 5 were detected by immunoblot. Full-length ER beta was detected in granulosa, small, medium, and large antral follicles, and midluteal phase CL of unprimed animals. Putative ER beta delta 5 immunoreactive bands were abundant only in granulosa cell extracts. In COS-1 cells, transfected ER beta delta 5 had no effect on basal transcription of an estrogen-responsive reporter construct but did repress wild-type ER beta transactivation when cotransfected at 10-fold excess plasmid. No repression of ER alpha transactivation was observed. In primary granulosa cell cultures, transfected ER beta delta 5 plasmid did not inhibit basal reporter activation. ER beta delta 5 was shown by immunofluorescence to localize to the nucleus in transfected COS-1 cells. In vitro translated ER beta delta 5 proteins bound estrogen response elements in DNA in electrophoretic mobility shift assays, as indicated by supershift analysis. ER beta is abundant in porcine ovary, and a naturally occurring splice variant missing exon 5 may have biological function.