•One of the largest miRNA clusters in the Sfmbt2 gene is paternally expressed in mice•Loss of Sfmbt2 miRNA results in severely impaired development of the placenta•Impaired placental development is associated with developmental delay of fetuses•Sfmbt2 miRNAs may promote proliferation by silencing tumor suppressorsMicroRNAs (miRNAs) represent small noncoding RNAs that are involved in physiological and developmental processes by posttranscriptionally inhibiting gene expression. One of the largest miRNA clusters in mice is located in intron 10 of the Sfmbt2 gene, containing 72 miRNA precursor sequences. In this study, we generated mice lacking the entire Sfmbt2 miRNA cluster to elucidate its functions during development. The Sfmbt2 miRNAs were expressed predominantly from the paternal allele in the placenta, as is the host Sfmbt2 gene. Loss of the paternal allele resulted in severely impaired development of the placenta, especially the spongiotrophoblast layer, and frequent lethality or defects of fetuses. The predicted target sequences of the miRNAs and gene expression analysis defined at least nine putative target genes, which function as tumor suppressors or apoptosis inducers. Our study has provided experimental evidence for the indispensable roles of placental miRNAs in trophoblast proliferation and thus fetal development.Download high-res image (163KB)Download full-size image

•Mettl23 is an arginine methyltransferase that catalyzes H3R17me2a in mouse oocytes•H3R17me2a is responsible for H3.3 incorporation and active DNA demethylation•Mettl23 interacts with Tet3 via gonad-specific expression (GSE) protein•GSE and Mettl23 are indispensable for active DNA demethylationAt fertilization, the paternal genome undergoes extensive reprogramming through protamine-histone exchange and active DNA demethylation, but only a few maternal factors have been defined in these processes. We identified maternal Mettl23 as a protein arginine methyltransferase (PRMT), which most likely catalyzes the asymmetric dimethylation of histone H3R17 (H3R17me2a), as indicated by in vitro assays and treatment with TBBD, an H3R17 PRMT inhibitor. Maternal histone H3.3, which is essential for paternal nucleosomal assembly, is unable to be incorporated into the male pronucleus when it lacks R17me2a. Mettl23 interacts with Tet3, a 5mC-oxidizing enzyme responsible for active DNA demethylation, by binding to another maternal factor, GSE (gonad-specific expression). Depletion of Mettl23 from oocytes resulted in impaired accumulation of GSE, Tet3, and 5hmC in the male pronucleus, suggesting that Mettl23 may recruit GSE-Tet3 to chromatin. Our findings establish H3R17me2a and its catalyzing enzyme Mettl23 as key regulators of paternal genome reprogramming.Download high-res image (217KB)Download full-size image

Substantial proportions of mammalian genomes comprise repetitive elements including endogenous retrotransposons. Although these play diverse roles during development, their appropriate silencing is critically important in maintaining genomic integrity in the host cells. The major mechanism for retrotransposon silencing is DNA methylation, but the wave of global DNA demethylation that occurs after fertilization renders preimplantation embryos exceptionally hypomethylated. Here, we show that hypomethylated preimplantation mouse embryos are protected from retrotransposons by repressive histone modifications mediated by the histone chaperone chromatin assembly factor 1 (CAF-1). We found that knockdown of CAF-1 with specific siRNA injections resulted in significant up-regulation of the retrotransposons long interspersed nuclear element 1, short interspersed nuclear element B2, and intracisternal A particle at the morula stage. Concomitantly, increased histone H2AX phosphorylation and developmental arrest of the majority (>95%) of embryos were observed. The latter was caused at least in part by derepression of retrotransposons, as treatment with reverse transcriptase inhibitors rescued some embryos. Importantly, ChIP analysis revealed that CAF-1 mediated the replacement of H3.3 with H3.1/3.2 at the retrotransposon regions. This replacement was associated with deposition of repressive histone marks, including trimethylation of histone H3 on lysine 9 (H3K9me3), H3K9me2, H3K27me3, and H4K20me3. Among them, H4K20me3 and H3K9me3 seemed to play predominant roles in retrotransposon silencing, as assessed by knockdown of specific histone methyltransferases and forced expression of unmethylatable mutants of H3.1K9 and H4K20. Our data thus indicate that CAF-1 is an essential guardian of the genome in preimplantation mouse embryos by deposition of repressive histone modifications via histone variant replacement.

Although ovarian theca cells play an indispensable role in folliculogenesis by providing follicular structural integrity and steroid substrates for estrogen production, little information is available about their recruitment, growth, and differentiation because their immature forms have not been identified. We have isolated putative thecal stem cells with the ability to self-renew and differentiate in vivo and in vitro. They are similar to fibroblasts in morphology and proliferate in vitro as round colonies with a homogenous cell population. They were induced to differentiate into early precursors and steroidogenic cells in a stepwise manner after treatment with serum, luteinizing hormone, and paracrine factors from granulosa cells. At each differentiation step, these cells displayed appropriate gene expression and morphological markers and later secreted androstenedione. The fully mature morphology was achieved by coculture with isolated granulosa cells. When transplanted into the ovaries, the putative thecal stem cells colonized exclusively in the ovarian interstitium and the thecal layer of follicles as differentiated cells. Thus, thecal stem cells appear to be present in neonatal ovaries and can be isolated, purified, and induced to differentiate in vitro. Thecal stem cells could provide an invaluable in vitro experimental system to study interactions among the oocytes, granulosa cells, and theca cells during normal folliculogenesis and to study ovarian pathology caused by theca cell dysfunction.

Cryopreservation of male germ cells is a strategy to conserve animal species and strains of animals valuable to biomedical research. We tested whether mouse male germ cells could be cryopreserved without cryoprotection by simply freezing epididymides, testes, or whole bodies. The reproductive organs were isolated from killed mice and frozen for 1 week to 1 year at −80°C before spermatozoa and spermatids were collected and injected into mature oocytes. Normal pups were born irrespective of strains tested (ICR and C57BL/6). Epididymides and testes frozen and transported internationally to another laboratory by air could produce pups of inbred C57BL/6 mice. Testicular spermatozoa retrieved from the bodies of male mice (BALB/c nude and C3H/He strains) that had been kept frozen (−20°C) for 15 years could also produce normal offspring by microinsemination. Thus, freezing of either male reproductive organs or whole bodies is the simplest way to preserve male germ cells. Restoration of extinct species could be possible if male individuals are found in permafrost.