Dual role of histone variant H3.3B in spermatogenesis: positive regulation of piRNA transcription and implication in X-chromosome inactivation.

The histone variant H3.3 is encoded by two distinct genes, H3f3a and H3f3b, exhibiting identical amino-acid sequence. H3.3 is required for spermatogenesis, but the molecular mechanism of its spermatogenic function remains obscure. Here, we have studied the role of each one of H3.3A and H3.3B proteins in spermatogenesis. We have generated transgenic conditional knock-out/knock-in (cKO/KI) epitope-tagged FLAG-FLAG-HA-H3.3B (H3.3BHA) and FLAG-FLAG-HA-H3.3A (H3.3AHA) mouse lines. We show that H3.3B, but not H3.3A, is required for spermatogenesis and male fertility. Analysis of the molecular mechanism unveils that the absence of H3.3B led to alterations in the meiotic/post-meiotic transition. Genome-wide RNA-seq reveals that the depletion of H3.3B in meiotic cells is associated with increased expression of the whole sex X and Y chromosomes as well as of both RLTR10B and RLTR10B2 retrotransposons. In contrast, the absence of H3.3B resulted in down-regulation of the expression of piRNA clusters. ChIP-seq experiments uncover that RLTR10B and RLTR10B2 retrotransposons, the whole sex chromosomes and the piRNA clusters are markedly enriched of H3.3. Taken together, our data dissect the molecular mechanism of H3.3B functions during spermatogenesis and demonstrate that H3.3B, depending on its chromatin localization, is involved in either up-regulation or down-regulation of expression of defined large chromatin regions.

Acetyl-CoA synthetase (ACSS2) does not generate butyryl- and crotonyl-CoA.

Acetyl and other acyl groups from different short-chain fatty acids (SCFA) competitively modify histones at various lysine sites. To fully understand the functional significance of such histone acylation, a key epigenetic mechanism, it is crucial to characterize the cellular sources of the corresponding acyl-CoA molecules required for the lysine modification. Like acetate, SCFAs such as propionate, butyrate and crotonate are thought to be the substrates used to generate the corresponding acyl-CoAs by enzymes known as acyl-CoA synthetases. The acetyl-CoA synthetase, ACSS2, which produces acetyl-CoA from acetate in the nucleocytoplasmic compartment, has been proposed to also mediate the synthesis of acyl-CoAs such as butyryl- and crotonyl-CoA from the corresponding SCFAs. This idea is now widely accepted and is sparking new research projects. However, based on our direct in vitro experiments with purified or recombinant enzymes and structural considerations, we demonstrate that ACSS2 is unable to mediate the generation of non-acetyl acyl-CoAs like butyryl- and crotonyl-CoA. It is therefore essential to re-examine published data and corresponding discussions in the light of this new finding.

Abnormal steroidogenesis and aromatase activity in preeclampsia.

INTRODUCTION: Estrogens and progesterone play critical roles in angiogenesis and vasodilation. Moreover, placental aromatase deficiency is detected in women with preeclampsia (PE) at delivery. We hypothesized that abnormal steroidogenesis occurs much earlier than typical PE diagnosis. Thus, we investigated whether the circulating steroid profile was already disturbed at 24-29 weeks of gestation in women with subsequent PE, and compared the profile with that of women with "placental" small gestational age (SGA) without PE. METHODS: We selected nulliparous women (n = 90) from the MOMA trial, including women with PE (n = 25), SGA (n = 25), and controls (NP; n = 40), for plasma steroid profiling by gas chromatography/mass spectrometry and to measure placental growth factor and soluble fms-like tyrosine kinase-1. Placental aromatase expression was evaluated in a new set of women. RESULTS: Compared with that of controls, the women with PE had a significantly lower estrone/androstenedione ratio, and exhibited a decreasing trend for estradiol and estrone levels. Lower estriol levels were observed in the SGA group compared to the NP group. Compared with that of controls, the women with PE and SGA had significantly higher levels of 20α-dihydroprogesterone (20α-DHP) and 20α-DHP/progesterone ratios. Pregnenolone sulfate levels were lower in the PE group than in the NP and SGA groups. Decreased expression of aromatase was observed in the PE group compared to the control group. DISCUSSION: Preeclampsia appears to be characterized by specific steroidogenesis dysregulation long before PE diagnosis, highlighting potential new biomarkers of PE.

Dynamics of Sun5 localization during spermatogenesis in wild type and Dpy19l2 knock-out mice indicates that Sun5 is not involved in acrosome attachment to the nuclear envelope.

The acrosome is an organelle that is central to sperm physiology and a defective acrosome biogenesis leads to globozoospermia, a severe male infertility. The identification of the actors involved in acrosome biogenesis is therefore particularly important to decipher the molecular pathogeny of globozoospermia. We recently showed that a defect in the DPY19L2 gene is present in more than 70% of globozoospermic men and demonstrated that Dpy19l2, located in the inner nuclear membrane, is the first protein involved in the attachment of the acrosome to the nuclear envelope (NE). SUN proteins serve to link the nuclear envelope to the cytoskeleton and are therefore good candidates to participate in acrosome-nucleus attachment, potentially by interacting with DPY19L2. In order to characterize new actors of acrosomal attachment, we focused on Sun5 (also called Spag4l), which is highly expressed in male germ cells, and investigated its localization during spermatogenesis. Using immunohistochemistry and Western blot experiments in mice, we showed that Sun5 transits through different cellular compartments during meiosis. In pachytene spermatocytes, it is located in a membranous compartment different to the reticulum. In round spermatids, it progresses to the Golgi and the NE before to be located to the tail/head junction in epididymal sperm. Interestingly, we demonstrate that Sun5 is not, as initially reported, facing the acrosome but is in fact excluded from this zone. Moreover, we show that in Dpy19l2 KO spermatids, upon the detachment of the acrosome, Sun5 relocalizes to the totality of the NE suggesting that the acrosome attachment excludes Sun5 from the NE facing the acrosome. Finally, Western-blot experiments demonstrate that Sun5 is glycosylated. Overall, our work, associated with other publications, strongly suggests that the attachment of the acrosome to the nucleus does not likely depend on the formation of SUN complexes.