Identification of 3',5'-cyclic adenosine monophosphate-inducible nuclear factors binding to the human urokinase promoter in mouse Sertoli cells

Modulation of Dhh signaling and altered Sertoli cell function in mice lacking the GPR37-prosaposin receptor

The mammalian G-protein-coupled receptor 37 (GPR37) is expressed in brain, in adult testis, and during the early phase of gonad differentiation. Somatic Sertoli cells (SCs) are located within the seminiferous tubules where they support the germinal epithelium. An adequate number of SCs is required for the complete prepubertal differentiation of germ cells and adult fertility. This study shows that Gpr37 and its ligand prosaposin are both postnatally expressed by SCs, whose proliferation and maturation are affected in Gpr37-null mutant mice during postnatal testicular development. Mutant pups show a delayed timing in sperm cell development, with a partial arrest of spermatocytes at the meiotic pachytene (e.g., 1.5-fold increase in Gpr37(-/-) P21 pups) and their increased apoptosis (e.g., 1.8-fold and 3.5-fold increase in Gpr37(-/-) P21 and adult mice, respectively). Mutant adults have reduced testis weight (wild type, 299 ± 5 mg; knockout, 258 ± 16 mg; P < 0.05) and epididymal sperm count and motility (e.g., 1.5-fold and 1.45-fold decrease in Gpr37(-/-) mice, respectively). Lack of Gpr37 results in the reduction in androgen receptor levels during prepubertal testis development, alongside the altered expression of SC maturation markers. It also affects the prepubertal testis expression of desert hedgehog (Dhh) mitogenic cascade components (Dhh, 1.3-fold increase in Gpr37(-/-) P10 and P21 pups; Gli2, 1.4-fold and 1.6-fold increase in Gpr37(-/-) P10 and P21 pups, respectively) including patched homolog 1 (1.3-fold increase in Gpr37(-/-) P10 and P21 pups), which is found localized in prepubertal SCs and is associated with Gpr37 in cultured primary SC samples. These results indicate that Gpr37 is a specific modulator of murine testis Dhh mitogenic signaling and SC proliferation and maturation.

The faah gene is the first direct target of estrogen in the testis: role of histone demethylase LSD1

Estrogen (E(2)) regulates spermatogenesis, yet its direct target genes have not been identified in the testis. Here, we cloned the proximal 5' flanking region of the mouse fatty acid amide hydrolase (faah) gene upstream of the luciferase reporter gene, and demonstrated its promoter activity and E(2) inducibility in primary mouse Sertoli cells. Specific mutations in the E(2) response elements (ERE) of the faah gene showed that two proximal ERE sequences (ERE2/3) are essential for E(2)-induced transcription, and chromatin immunoprecipitation experiments showed that E(2) induced estrogen receptor β binding at ERE2/3 sites in the faah promoter in vivo. Moreover, the histone demethylase LSD1 was found to be associated with ERE2/3 sites and to play a role in mediating E(2) induction of FAAH expression. E(2) induced epigenetic modifications at the faah proximal promoter compatible with transcriptional activation by remarkably decreasing methylation of both DNA at CpG site and histone H3 at lysine 9. Finally, FAAH silencing abolished E(2) protection against apoptosis induced by the FAAH substrate anandamide. Taken together, our results identify FAAH as the first direct target of E(2).

Analysis of the gene expression profile of mouse male meiotic germ cells

Wide genome analysis of difference in gene expression between spermatogonial populations from 7-day-old mice and pachytene spermatocytes from 18-day-old mice was performed using Affymetrix gene chips representing approximately 12,500 mouse known genes or EST sequences, spanning approximately 1/3rd of the mouse genome. To delineate differences in the profile of gene expression between mitotic and meiotic stages of male germ cell differentiation, expressed genes were grouped in functional clusters. The analysis confirmed the previously described pre-meiotic or meiotic expression for several genes, in particular for those involved in the regulation of the mitotic and meiotic cell cycle, and for those whose transcripts are accumulated during the meiotic stages to be translated later in post-meiotic stages. Differential expression of several additional genes was discovered. In few cases (pro-apoptotic factors Bak, Bad and Bax), data were in conflict with the previously published stage-dependent expression of genes already known to be expressed in male germ cells. Northern blot analysis of selected genes confirmed the results obtained with the microarray chips. Six of these were novel genes specifically expressed in pachytene spermatocytes: a chromatin remodeling factor (chrac1/YCL1), a homeobox gene (hmx1), a novel G-coupled receptor for an unknown ligand (Gpr19), a glycoprotein of the intestinal epithelium (mucin 3), a novel RAS activator (Ranbp9), and the A630056B21Rik gene (predicted to encode a novel zinc finger protein). These studies will help to delineate the global patterns of gene expression characterizing male germ cell differentiation for a better understanding of regulation of spermatogenesis in mammals.

The molecular biology of cyclic nucleotide phosphodiesterases

Recent progress in the field of cyclic nucleotides has shown that a large array of closely related proteins is involved in each step of the signal transduction cascade. Nine families of adenylyl cyclases catalyze the synthesis of the second messenger cAMP, and protein kinases A, the intracellular effectors of cAMP, are composed of four regulatory and three catalytic subunits. A comparable heterogeneity has been discovered for the enzymes involved in the inactivation of cyclic nucleotide signaling. In mammals, 19 different genes encode the cyclic nucleotide phosphodiesterases (PDEs), the enzymes that hydrolyze and inactivate cAMP and cGMP. This is only an initial level of complexity, because each PDE gene contains several distinct transcriptional units that give rise to proteins with subtle structural differences, bringing the number of the PDE proteins close to 50. The molecular biology of PDEs in Drosophila and Dictyostelium has shed some light on the role of PDE diversity in signaling and development. However, much needs to be done to understand the exact function of these enzymes, particularly during mammalian development and cell differentiation. With the identification and mapping of regulatory and targeting domains of the PDEs, modularity of the PDE structure is becoming an established tenet in the PDE field. The use of different transcriptional units and exon splicing of a single PDE gene generates proteins with different regulatory domains joined to a common catalytic domain, therefore expanding the array of isoforms with subtle differences in properties and sensitivities to different signals. The physiological context in which these different isoforms function is still largely unknown and undoubtedly will be a major area of expansion in the years to come.

The same sequence mediates activation of the human urokinase promoter by cAMP in mouse Sertoli cells and by SV40 large T antigen in COS cells

Cell-specific activation by follicle-stimulating hormone and its intracellular mediator, cAMP, of the human urokinase promoter in mouse Sertoli cells requires overlapping purine-rich and GC-rich sequences between -54 and -42 from the transcriptional start site. We have previously shown that binding of unidentified nuclear factors to these sequences is induced by cAMP stimulation, and that sequences from the enhancerless SV40 replication origin can interfere with the binding, whereas consensus Sp1 binding sites are ineffective. We now show that sequences within the SV40 origin able to compete for the formation of cAMP-induced DNA-protein complexes in Sertoli cell nuclear extracts are binding sites for the SV40 large T antigen. Large T antigen expressed in COS cells binds the cAMP-responsive sequences of the human urokinase gene and transactivates the proximal promoter, thus mimicking the effect of nuclear factors induced by cAMP in Sertoli cells. We show that Egr-1 is one of the factors present in cAMP-induced DNA-protein complexes formed between the human urokinase promoter and Sertoli cell nuclear extracts. However, Egr-1 levels are similar in unstimulated and cAMP-treated Sertoli cells, suggesting that this factor interacts with a different GC-box binding factor, that we have previously shown to be strongly induced by cAMP treatment of Sertoli cells. We propose that SV40 large T antigen in COS cells can mimick the action of heterodimers formed in cAMP stimulated Sertoli cells between Egr-1 and a cell specific cAMP-induced GC-box binding factor.