Phosphorylation of GATA4 serine 105 but not serine 261 is required for testosterone production in the male mouse.

BACKGROUND: GATA4 is a transcription factor essential for male sex determination, testicular differentiation during fetal development, and male fertility in the adult. GATA4 exerts part of its function by regulating multiple genes in the steroidogenic enzyme pathway. In spite of these crucial roles, how the activity of this factor is regulated remains unclear. OBJECTIVES: Studies in gonadal cell lines have shown that GATA4 is phosphorylated on at least two serine residues-serine 105 (S105) and serine 261 (S261)-and that this phosphorylation is important for GATA4 activity. The objective of the present study is to characterize the endogenous role of GATA4 S105 and S261 phosphorylation in the mouse testis. MATERIALS AND METHODS: We examined both previously described GATA4 S105A mice and a novel GATA4 S261A knock-in mouse that we generated by CRISPR/Cas9 gene editing. The male phenotype of the mutants was characterized by assessing androgen-dependent organ weights, hormonal profiles, and expression of multiple testicular target genes using standard biochemical and molecular biology techniques. RESULTS: The fecundity of crosses between GATA4 S105A mice was reduced but without a change in sex ratio. The weight of androgen-dependent organs was smaller when compared to wild-type controls. Plasma testosterone levels showed a 70% decrease in adult GATA4 S105A males. This decrease was associated with a reduction in Cyp11a1, Cyp17a1, and Hsd17b3 expression. GATA4 S261A mice were viable and testis morphology appeared normal. Testosterone production and steroidogenic enzyme expression were not altered in GATA4 S261A males. DISCUSSION AND CONCLUSION: Our analysis showed that blocking GATA4 S105 phosphorylation is associated with decreased androgen production in males. In contrast, S261 phosphorylation by itself is dispensable for GATA4 function. These results confirm that endogenous GATA4 action is essential for normal steroid production in males and that this activity requires phosphorylation on at least one serine residue.

Modulation of endogenous GATA-4 activity reveals its dual contribution to Müllerian inhibiting substance gene transcription in Sertoli cells.

Secretion of Müllerian inhibiting substance by fetal Sertoli cells is essential for normal male sex differentiation since it induces regression of the Müllerian ducts in the developing male embryo. Proper spatiotemporal expression of the MIS gene requires a specific combination of transcription factors, including the zinc finger factor GATA-4 and the nuclear receptor steroidogenic factor-1, which both colocalize with Müllerian inhibiting substance in Sertoli cells. To establish the molecular mechanisms through which GATA-4 contributes to MIS transcription, we have generated and characterized novel GATA-4 dominant negative competitors. The first one, which consisted solely of the GATA-4 zinc finger DNA-binding domain, was an efficient competitor of GATA transcription mediated both by direct GATA binding to DNA and protein-protein interactions involving GATA factors. The second type of competitor consisted of the same GATA-4 zinc finger DNA-binding domain but harboring mutations that prevented DNA binding. This second class of competitors repressed GATA-dependent transactivation by specifically competing for GATA protein-protein interactions without affecting the DNA-binding activity of endogenous GATA factors. These competitors, along with the GATA-4 cofactor FOG-2 (friend of GATA-2), were used to specifically modulate endogenous GATA-4 activity in Sertoli cells. Our results indicate that GATA-4 contributes to MIS promoter activity through two distinct mechanisms. Moreover, the GATA competitors described here should provide invaluable in vitro and in vivo tools for the study of GATA- dependent transcription and the identification of new target genes.

Nuclear receptor Dax-1 represses the transcriptional cooperation between GATA-4 and SF-1 in Sertoli cells.

A crucial step in mammalian sex differentiation is the regression of the Müllerian ducts in males. This is achieved through the action of Müllerian inhibiting substance (MIS), a key hormone produced by fetal Sertoli cells. Proper spatiotemporal expression of the MIS gene requires the concerted action of several transcription factors that include Sox9, SF-1, WT-1, GATA-4, and Dax-1. Indeed, SF-1 contributes to MIS gene expression by transcriptionally cooperating with other factors such as GATA-4 and WT-1. Dax-1 is coexpressed with SF-1 in many tissues, including the gonads, where it acts as a negative modulator of SF-1-dependent transcription. We now report that Dax-1 can repress MIS transcription in Sertoli cells by disrupting transcriptional synergism between GATA-4 and SF-1. Dax-1-mediated repression of GATA-4/SF-1 synergism did not involve direct repression of GATA-dependent transactivation, but rather, it occurred through a direct protein-protein interaction with DNA-bound SF-1. It is interesting that SF-1, Dax-1, and GATA factors are coexpressed in several tissues such as the pituitary, the adrenals, and the gonads. Because we have shown that other GATA family members also have the ability to synergize with SF-1, Dax-1 repression of GATA/SF-1 synergism may represent an important mechanism for fine-tuning the regulation of SF-1-dependent genes in multiple target tissues.

GATA factors differentially activate multiple gonadal promoters through conserved GATA regulatory elements.

The GATA factors are a group of transcriptional regulators that play essential roles in cell differentiation, organ morphogenesis, and tissue-specific gene expression during development. The six vertebrate GATA factors are expressed in a broad spectrum of tissues, including the hemopoietic system, heart, gut, brain, placenta, pituitary, and gonads. Interestingly, GATA-like DNA-binding proteins are found in the gonads of several species, ranging from lower invertebrates to humans, thus supporting an evolutionary conserved and crucial role for these factors in gonadal development and function. Indeed, GATA factors are expressed from the onset of gonadal development and are later found in multiple cell lineages of both the testis and ovary. We now report that GATA-4 differentially activates transcription of several genes expressed in the gonads that encode either steroidogenic enzymes (steroidogenic acute regulatory protein and aromatase), hormones (inhibin alpha and Müllerian inhibiting substance) and a transcription factor (SF-1) known to be essential for gonadal development and function. Thus, our results identify GATA-4 as an important regulator of gonadal gene transcription where its specificity of action is mediated through synergistic interactions with other transcription factors such as SF-1.

Transcription factor GATA-4 enhances Müllerian inhibiting substance gene transcription through a direct interaction with the nuclear receptor SF-1.

Secretion of Müllerian-inhibiting substance (MIS) by Sertoli cells of the fetal testis and subsequent regression of the Müllerian ducts in the male embryo is a crucial event that contributes to proper sex differentiation. The zinc finger transcription factor GATA-4 and nuclear receptor SF-1 are early markers of Sertoli cells that have been shown to regulate MIS transcription. The fact that the GATA and SF-1 binding sites are adjacent to one another in the MIS promoter raised the possibility that both factors might transcriptionally cooperate to regulate MIS expression. Indeed, coexpression of both factors resulted in a strong synergistic activation of the MIS promoter. GATA-4/SF-1 synergism was the result of a direct protein-protein interaction mediated through the zinc finger region of GATA-4. Remarkably, synergy between GATA-4 and SF-1 on a variety of different SF-1 targets did not absolutely require GATA binding to DNA. Moreover, synergy with SF-1 was also observed with other GATA family members. Thus, these data not only provide a clearer understanding of the molecular mechanisms that control the sex-specific expression of the MIS gene but also reveal a potentially novel mechanism for the regulation of SF-1-dependent genes in tissues where SF-1 and GATA factors are coexpressed.

Transcription factor GATA-4 is expressed in a sexually dimorphic pattern during mouse gonadal development and is a potent activator of the Müllerian inhibiting substance promoter.

Mammalian gonadal development and sexual differentiation are complex processes that require the coordinated expression of a specific set of genes in a strict spatiotemporal manner. Although some of these genes have been identified, the molecular pathways, including transcription factors, that are critical for the early events of lineage commitment and sexual dimorphism, remain poorly understood. GATA-4, a member of the GATA family of transcription factors, is present in the gonads and may be a regulator of gonadal gene expression. We have analyzed the ontogeny of gonadal GATA-4 expression by immunohistochemistry. GATA-4 protein was detected as early as embryonic day 11.5 in the primitive gonads of both XX and XY mouse embryos. In both sexes, GATA-4 specifically marked the developing somatic cell lineages (Sertoli in testis and granulosa in ovary) but not primordial germ cells. Interestingly, abundant GATA-4 expression was maintained in Sertoli cells throughout embryonic development but was markedly down-regulated shortly after the histological differentiation of the ovary on embryonic day 13.5. This pattern of expression suggested that GATA-4 might be involved in early gonadal development and possibly sexual dimorphism. Consistent with this hypothesis, we found that the Müllerian inhibiting substance promoter which harbors a conserved GATA element is a downstream target for GATA-4. Thus, transcription factor GATA-4 may be a new factor in the cascade of regulators that control gonadal development and sex differentiation in mammals.

The mRNAs for the steroid 5 alpha-reductase isozymes, types 1 and 2, are differentially regulated in the rat epididymis.

The enzyme steroid 5 alpha-reductase (EC 1.3.1.22) catalyzes the conversion of testosterone to its more potent form, dihydrotestosterone (DHT), in many androgen-sensitive target tissues. In the epididymis, the 5 alpha-reduced metabolites of testosterone, DHT and 5 alpha-androstan-3 alpha, 17 beta-diol (3 alpha-diol), are considered the primary regulators of epididymal structure and function. Two rat 5 alpha-reductase transcripts, designated types 1 and 2, have been identified. Our laboratory has previously characterized the endocrine and developmental regulation of the 5 alpha-reductase type 1 mRNA in the rat epididymis. However, regulation of the type 2 mRNA has not been investigated. Thus, we undertook to characterize the longitudinal distribution of the steady state 5 alpha-reductase type 2 mRNA as well as the effects of development and of unilateral efferent duct ligation on its expression in the rat epididymis. To contrast the regulation of the type 2 5 alpha-reductase isozyme mRNAs in the rat epididymis, these data have been compared with those previously obtained for the type 1 mRNA and enzyme activity. An analysis of the longitudinal distribution of the type 2 transcript along the epididymis revealed that its mRNA was expressed predominantly in the proximal caput region of the tissue. This regional distribution pattern differed markedly from the patterns previously described for the type 1 mRNA and enzyme activity. Surprisingly, a segment-by-segment analysis showed that epididymal 5 alpha-reductase type 2 mRNA levels were not altered as a function of increasing postnatal age. Again, this result was in marked contrast to the important changes in 5 alpha-reductase type 1 mRNA and enzyme activity that have been reported to occur during puberty in the rat. In the final experiment, the effect of unilateral efferent duct ligation revealed that 5 alpha-reductase type 2 mRNA levels increased in the initial segment of the ligated side but remained unchanged in the rest of the tissue; this was in marked contrast to the dramatic decrease (>60%) in type 1 mRNA levels observed specifically in the initial segment of the epididymis. Taken together, these experiments demonstrate that 5 alpha-reductase type 1 and type 2 mRNAs are differentially regulated in the rat epididymis. Moreover, this is the first report of a tissue in which the mRNAs for the steroid 5 alpha-reductase isozymes are regulated differently.

Steady state steroid 5 alpha-reductase messenger ribonucleic acid levels and immunocytochemical localization of the type 1 protein in the rat testis during postnatal development.

Steroid 5 alpha-reductase is the rate-limiting enzyme in the production of 5 alpha-reduced steroids in many tissues. Developmental changes in 5 alpha-reductase activity play an important role in regulating the amount of testosterone that is secreted by the testis. To date, the regulation of testicular 5 alpha-reductase has been studied extensively at the level of enzyme activity. Regulation at the messenger RNA (mRNA) and protein levels, however, has not been investigated. The objectives of the present study were to determine the steady state mRNA levels for the 5 alpha-reductase isozymes, types 1 and 2, and to immunolocalize the 5 alpha-reductase type 1 protein in the developing rat testis (7-91 days postpartum). Consistent with previously reported enzyme activity studies, type 1 5 alpha-reductase mRNA levels were most abundant in the immature animal (days 21-28). Unlike 5 alpha-reductase activity, however, type 1 mRNA levels did not decline thereafter to reach nearly undetectable levels in the adult (day 91). In contrast, 5 alpha-reductase type 1 mRNA levels remained relatively constant between days 42-91. The 5 alpha-reductase type 1 transcript size did not remain constant during postnatal testicular development. The characteristic 2.5-kilobase type 1 transcript size was detected in immature rats (days 21-28), whereas in the adult (day 91), a slower migrating 2.7-kilobase type 1 mRNA species was observed. An antipeptide antiserum specific to rat 5 alpha-reductase type 1 was used to immunolocalize the 5 alpha-reductase type 1 protein. At all ages examined, the immunoperoxidase reaction was localized predominantly to the interstitial tissue of the testis. On postnatal day 7, clusters of interstitial cells resembling fetal Leydig cells were clearly immunoreactive. The staining intensity increased steadily from day 7 onwards, so that by days 21 and 28, interstitial cells with the appearance of immature Leydig cells were intensely immunoreactive (peak expression). This was followed by a progressive decrease in staining intensity between days 28-91, so that by day 91 (adult) Leydig cell immunoreactivity was barely detectable. Immunocytochemical staining revealed a predominantly cytoplasmic localization; significant nuclear staining was not evident. We conclude that the expression of the 5 alpha-reductase type 1 protein is primarily found in the cytoplasm of Leydig cells, is dependent on age, and that this expression closely parallels 5 alpha-reductase enzyme activity.

Gene expression in the aging brown Norway rat epididymis.

The mammalian epididymis is the site where spermatozoa are matured and then stored. Though many studies have described epididymal functions and their regulation, little is known about how aging affects this tissue. The Brown Norway rat, which does not show the many age-related pathologies common to other rat strains, was used as a model to study aging of the epididymis. The present study was designed to determine the effect of aging on the mRNA levels for selected markers of epididymal function. Brown Norway rats ranging in age from 6 to 30 months were examined at 6-month intervals; epididymides were sectioned into caput-corpus and cauda regions. Relative mRNA concentrations were assessed using Northern blot analysis and specific cDNAs for the rat 5 alpha-reductase isozymes, types 1 and 2; proenkephalin; the androgen receptor; epididymal proteins B/C and D/E; and sulfated glycoprotein-2 (SGP-2, clusterin). Northern blots were quantitated by densitometric scanning. In the caput-corpus epididymidis, 5 alpha-reductase type 1 and type 2 mRNA levels decreased significantly by 43% and 33%, respectively, between 6 and 12 months and by 64% and 40%, respectively, between 6 and 30 months. No significant change, however, was found in the expression of the 5 alpha-reductase mRNAs in the cauda epididymidis. Interestingly, proenkephalin mRNA was only detected in the caput-corpus epididymidis of 6-month-old rats. In marked contrast to the 5 alpha-reductase isozymes and proenkephalin, no significant age-related changes were observed in the mRNA levels for the androgen receptor, protein B/C, or protein D/E. No age-related changes in mRNA expression for SGP-2 occurred in the caput-corpus epididymidis. However, in the cauda epididymidis, SGP-2 mRNA levels rose by twofold between 6 and 18 months and then decreased sharply by 75% between 18 and 30 months. We conclude that as the epididymis ages, the expression of genes for certain specific markers of epididymal function is affected in a region-specific manner. Further, the decrease in the concentrations of the mRNAs for the 5 alpha-reductase isozymes and proenkephalin in the epididymis between 6 and 12 months is thus far the earliest marker for aging in the male reproductive tract of the Brown Norway rat.

Regulation of epididymal epithelial cell functions.

It is well established that the epididymis is the site where spermatozoa are matured and stored, but our understanding of the regulation of epididymal epithelium functions and their effects on spermatozoa is still fairly limited. The most active regulator of epididymal functions seems to be dihydrotestosterone, the 5 alpha-reduced metabolite of testosterone. Our laboratory has focused on the regulation of 5 alpha-reductase, with studies encompassing its messenger RNA, protein and enzyme activity. We have also investigated the hormonal regulation and distribution of other specific key proteins found in epididymal epithelial cells that play critical roles in the function of these cells. These proteins include clusterin or sulfated glycoprotein-2 and the glutathione S-transferases (GST). Using complementary experimental approaches, including orchidectomy and hormonal replacement, efferent duct ligation, and developmental studies, we have established that 5 alpha-reductase enzyme activity is present in both nuclear and microsomal fractions; the nuclear enzyme appears almost exclusively in the initial segment of the epididymis. In addition, 5 alpha-reductase activity and the mRNAs for both the type 1 and type 2 form of the enzyme are regulated differentially with respect to age and site within the epididymis. Immunolocalization of the protein has revealed that it is located in principal cells and that its subcellular location is dependent on the region of the epididymis. These results indicate that there is both transcriptional and post-transcriptional regulation of the expression of 5 alpha-reductase. Clusterin is a hydrophobic protein secreted by Sertoli cells and found in high concentration in the epididymis. This glycoprotein is expressed at its highest levels in the initial segment and caput epididymidis and at very low levels in the corpus and cauda epididymidis of the intact rat, and it exhibits a novel pattern of androgen regulation. In the areas of highest expression, there is no androgen dependence; however, orchidectomy causes a dramatic increase in the message for clusterin, which is suppressible by androgens in the segments where expression is normally lowest. The GSTs are a family of enzymes thought to play a key role in detoxification. Members of the GST family are expressed in a region-dependent manner along the rat epididymis. We have found that the localization of one member of this enzyme family, GST P, or subunit Yp, is selective for basal cells in the corpus and cauda epididymidis.(ABSTRACT TRUNCATED AT 400 WORDS)

Immunocytochemical localization of 4-ene steroid 5 alpha-reductase type 1 along the rat epididymis during postnatal development.

Dihydrotestosterone (DHT), the active androgen in many tissues, is synthesized from testosterone by the enzyme 4-ene steroid 5 alpha-reductase (5 alpha-reductase; EC 1.3.1.22). In the epididymis, the maturation and storage of spermatozoa are dependent on the presence of 5 alpha-reduced androgens. The regulation of epididymal 5 alpha-reductase is complex. To date, the regulation of this enzyme has been studied extensively at the level of enzymatic activity and more recently at the mRNA level. Regulation at the level of the protein, however, remains poorly understood. We have raised rabbit polyclonal antibodies to a 24-mer synthetic peptide whose sequence was determined from the predicted amino acid sequence for rat 5 alpha-reductase type 1 to immunolocalize the 5 alpha-reductase type 1 protein in the rat epididymis during postnatal development. Western blot analysis revealed a specific immunoreactive band of 26 kilodaltons in male rat liver, epididymis, and prostate; this apparent molecular size is identical to that obtained when the 5 alpha-reductase type 1 cDNA is expressed in mammalian cells. Furthermore, the relative protein levels, liver > epididymis > prostate, were consistent with the mRNA levels for type 1 rat 5 alpha-reductase. Perfusion-fixed paraffin-embedded epididymal tissue sections were used to immunolocalize type 1 5 alpha-reductase. In the adult rat epididymis, the most intense immunoperoxidase reaction was observed in a discrete lobule of the initial segment of the epididymis. A progressive decrease in staining intensity occurred distally along the tissue to the cauda epididymis. The staining reaction was specific to cytoplasmic elements of epithelial principal cells; no reaction was evident over nuclei. However, specifically in the initial segment, very intense staining was seen in the infranuclear region of the principal cells. In the proximal caput epididymidis, the staining was primarily confined to an oval region above the nuclei, whereas in the remaining epididymal regions, weak staining was seen throughout the cytoplasm. Thus, the intracellular localization of the 5 alpha-reductase type 1 protein changed as one moved down the epididymis. Finally, the pattern of immunolocalization of 5 alpha-reductase type 1 protein was different in the epididymis of rats of different postnatal ages. On day 7, no reactivity was noted; by day 28, a weak apical staining of principal cells was seen throughout the epididymis; by day 47, the adult pattern of staining had been established. Our results revealed that the expression and intracellular localization of the 5 alpha-reductase type 1 protein are both age dependent and epididymal segment specific.

Expression of 4-ene steroid 5 alpha-reductase messenger ribonucleic acid in the rat epididymis during postnatal development.

The regulation of epididymal 5 alpha-reductase mRNA is multifactorial and segment-specific. To further investigate the regulation of the message for the enzyme, the expression of 5 alpha-reductase mRNA in the rat epididymis was studied as a function of postnatal development. Developmental changes in 5 alpha-reductase mRNA concentrations were assessed by probing Northern blots with the full-length cDNA for rat steroid 5 alpha-reductase. In the first experiment the effect of postnatal age on 5 alpha-reductase mRNA concentrations in the caput-corpus and cauda epididymides was studied. Male rats, taken at 1-week intervals between the ages of 7-91 days, were used. In both epididymal regions, the mRNA for 5 alpha-reductase was present at all ages examined; it appeared in the immature animal at least 2 weeks before detectable 5 alpha-reductase enzyme activity. In the caput-corpus epididymidis, mRNA levels for 5 alpha-reductase decreased by half between postnatal days 7 and 21, rose 5-fold by day 56, and then remained constant through day 91. No change with postnatal age, however, was observed in the cauda epididymidis. In the second experiment, the longitudinal distribution of 5 alpha-reductase mRNA on postnatal days 21, 42, 49, 56, 77, and 91 was studied. The mRNA levels for 5 alpha-reductase increased remarkably, by 6- to 7-fold, in the initial segment of the caput epididymidis between postnatal days 21 and 42 and stayed constant thereafter. However, no significant change comparable to that found in the initial segment was observed in the adjacent proximal caput region or in any of the other epididymal segments. Thus, the 5-fold rise in 5 alpha-reductase mRNA concentrations that occurred in the caput-corpus epididymidis in the first experiment can be attributed solely to changes in the initial segment. We conclude that steady state concentrations of epididymal 5 alpha-reductase mRNA vary dramatically at different postnatal ages and are highly specific with respect to epididymal segment.

Differential regulation of steady state 4-ene steroid 5 alpha-reductase messenger ribonucleic acid levels along the rat epididymis.

Epididymal nuclear 5 alpha-reductase enzyme activity is regulated by a testosterone-dependent factor from the testis. Regulation at the mRNA level, however, has not been investigated. Endocrine manipulation experiments were designed to determine whether 5 alpha-reductase is regulated at the steady state mRNA level. Steady state mRNA concentrations were assessed using the full-length cDNA for female rat liver 5 alpha-reductase. Longitudinal distribution showed that the highest mRNA concentrations were present in the initial segment of the caput epididymidis and were 3- to 7-fold higher than in the other tissue segments. The androgen dependence of the mRNA levels for 5 alpha-reductase was assessed by bilateral orchidectomy and simultaneous testosterone replacement therapy. One week after surgery, mRNA concentrations in orchidectomized rats were decreased to 15% of control levels in the initial segment of the caput epididymidis and to 40-50% of control levels in the remaining epididymal segments. Administration of testosterone at a dose that mimics normal serum concentrations (2.5-cm Silastic implant) restored 5 alpha-reductase mRNA concentrations to control levels in the corpus and cauda epididymidis, but these were not significantly different from orchidectomized levels (P greater than or equal to 0.05) in the initial segment and caput epididymidis. Administration of testosterone at a dose designed to approximate 5- to 8-fold normal serum concentrations (18.6-cm implant) maintained 5 alpha-reductase mRNA concentrations at only 50% of control levels in the initial segment, while complete maintenance was observed in the rest of the tissue. The effects of unilateral orchidectomy revealed that 5 alpha-reductase mRNA concentrations decrease selectively in the initial segment of the orchidectomized side. This is the first report that epididymal 5 alpha-reductase is regulated at the mRNA level and that the regulation is different with respect to the segment being studied.