MafA is a transcription factor that binds to the promoter in the insulin gene and has been postulated to regulate insulin transcription in response to serum glucose levels, but there is no current in vivo evidence to support this hypothesis. To analyze the role of MafA in insulin transcription and glucose homeostasis in vivo, we generated MafA-deficient mice. Here we report that MafA mutant mice display intolerance to glucose and develop diabetes mellitus. Detailed analyses revealed that glucose-, arginine-, or KCl-stimulated insulin secretion from pancreatic ␤ cells is severely impaired, although insulin content per se is not significantly affected. MafA-deficient mice also display age-dependent pancreatic islet abnormalities. Further analysis revealed that insulin 1, insulin 2, Pdx1, Beta2, and Glut-2 transcripts are diminished in MafA-deficient mice. These results show that MafA is a key regulator of glucose-stimulated insulin secretion in vivo.Insulin is the only polypeptide hormone that is essential for the regulation of blood glucose levels and is synthesized exclusively in ␤ cells of the islets of Langerhans in the pancreas. The molecular mechanisms that control ␤-cell-specific insulin gene transcription are well characterized. Three conserved cis-regulatory elements within the promoter, E1, A3, and RIPE3b/ C1, respectively, appear to be indispensable for proper insulin gene regulation (22,25). Islet-restricted transcription factors Beta2/NeuroD and Pdx1 bind to the E1 and A3 elements in vitro. Gene disruption experiments in mice have revealed that both Beta2 and Pdx1 play critical roles in insulin gene regulation as well as in islet development and function (1,8,21). Furthermore, mutations in both the Beta2 and Pdx1 genes have been identified within populations of patients with type II diabetes (18,29,30).The third regulatory element, RIPE3b/C1, has also been shown to play a critical role in ␤-cell-specific insulin gene transcription as well as in glucose-regulated expression. Previous studies identified a pancreatic ␤-cell-restricted factor, called the RIPE3b1 activator, that is enriched in response to glucose in pancreatic ␤-cell nuclear extracts. Very recently, four groups reported that the RIPE3b1 activator is a member of the Maf family of transcription factors, MafA (10,12,20,26). The large Maf proteins, MafA/L-Maf/SMaf1 (2, 9, 24), MafB (11), c-Maf (23), and Nrl (31), each contain a basic motif followed by a leucine zipper, and all four family members harbor acidic domains that act as transcriptional activation domains. Although a role for MafA in insulin gene regulation was hypothesized, in vivo tests of the hypothesis have not been reported. To elucidate MafA function in insulin gene regulation, we generated MafA-deficient mice. MATERIALS AND METHODSTargeted disruption of the mafA gene. mafA genomic clones were isolated from a 129/SvJ genomic library (Stratagene) using a partial mouse MafA cDNA as a probe. The targeting vector was constructed with the bacterial lacZ gene containing a nuclear loca...

MafB is a member of the large Maf family of transcription factors that share similar basic region/leucine zipper DNA binding motifs and N-terminal activation domains. Although it is well known that MafB is specifically expressed in glomerular epithelial cells (podocytes) and macrophages, characterization of the null mutant phenotype in these tissues has not been previously reported. To investigate suspected MafB functions in the kidney and in macrophages, we generated mafB/green fluorescent protein (GFP) knock-in null mutant mice. mafB homozygous mutants displayed renal dysgenesis with abnormal podocyte differentiation as well as tubular apoptosis. Interestingly, these kidney phenotypes were associated with diminished expression of several kidney disease-related genes. In hematopoietic cells, GFP fluorescence was observed in both Mac-1-and F4/80-expressing macrophages in the fetal liver. Interestingly, F4/80 expression in macrophages was suppressed in the homozygous mutant, although development of the Mac-1-positive macrophage population was unaffected. In primary cultures of fetal liver hematopoietic cells, MafB deficiency was found to dramatically suppress F4/80 expression in nonadherent macrophages, whereas the Mac-1-positive macrophage population developed normally. These results demonstrate that MafB is essential for podocyte differentiation, renal tubule survival, and F4/80 maturation in a distinct subpopulation of nonadherent mature macrophages.

SUMMARY Gametogenesis is dependent on the expression of germline-specific genes. However, it remains unknown how the germline epigenome is distinctly established from that of somatic lineages. Here we show that genes commonly expressed in somatic lineages and spermatogenesis-progenitor cells undergo repression in a genome-wide manner in late stages of the male germline and identify underlying mechanisms. SCML2, a germline-specific subunit of a Polycomb repressive complex 1 (PRC1), establishes the unique epigenome of the male germline through two distinct antithetical mechanisms. SCML2 works with PRC1 and promotes RNF2-dependent ubiquitination of H2A, thereby marking somatic/progenitor genes on autosomes for repression. Paradoxically, SCML2 also prevents RNF2-dependent ubiquitination of H2A on sex chromosomes during meiosis, thereby enabling unique epigenetic programming of sex chromosomes for male reproduction. Our results reveal divergent mechanisms involving a shared regulator by which the male germline epigenome is distinguished from that of the soma and progenitor cells.

Repressive H3K27me3 and active H3K4me2/3 together form bivalent chromatin domains, molecular hallmarks of developmental potential. In the male germline, these domains are thought to persist into sperm to establish totipotency in the next generation. However, it remains unknown how H3K27me3 is established on specific targets in the male germline. Here, we demonstrate that a germline-specific Polycomb protein, SCML2, binds to H3K4me2/3-rich hypomethylated promoters in undifferentiated spermatogonia to facilitate H3K27me3. Thus, SCML2 establishes bivalent domains in the male germline of mice. SCML2 regulates two major classes of bivalent domains: Class I domains are established on developmental regulator genes that are silent throughout spermatogenesis, while class II domains are established on somatic genes silenced during late spermatogenesis. We propose that SCML2-dependent H3K27me3 in the male germline prepares the expression of developmental regulator and somatic genes in embryonic development.