Search

Additional file 1: Figure S1. Schematic representation of HIV-1 RNAs detected by RT-qPCR in this study. Organization of the NL4-3 genome and positions of SD and SA sites are indicated. Introns and exons nomenclature are according to [34]. Thick lines represent retained exon and thin lines excised introns. RNA species were named as indicated on the left side according to [34]. Forward (-f) and reverse (-r) primers used in this study are indicated by arrows. Of note, Vpr3 primers designed to detect the Vrp3 IS isoform also amplifies the Vrp1 MS isoform (not shown). However, Vpr3 represents more than 80% of its Vpr transcripts family compare to less than 7% for Vpr1 [34]. Figure S2. YTHDF2 depletion increases HIV-1 mRNA transcripts abundance in HeLa infected cells (A) HeLa cells were transfected with a control (Ctrl) or YTHDF1, YTHDF2 and YTHDF3 targeting siRNAs, as indicated. 3 days later, the respective mRNA abundance of each of the 3 readers was measured by RT-qPCR. Data are presented as mean �� S.D. (n = 5 for siYTHDF1 and n = 3 for siYTHDF2 and siYTHDF3). (B) Cellular proteins knockdown was confirmed by western blot analysis of cell lysates transfected with indicated siRNAs. Note that YTHDF1 was detected on a membrane already probed with the YTHDF2 antibody (��, YTHDF2; *, nonspecific band). (C) 3 days after siRNA transfection, cells were infected with a single round VSVg-pseudotyped HIV-1 virus. 24 h.p.i., the relative abundance of US, IS and MS viral RNAs was monitored by RT-qPCR. Data are presented as mean �� S.D. (n = 3). Results are expressed in fold change over the control siRNA (Ctrl). Data are presented as mean �� S.D. (n = 3). P values were calculated using one-sample t-test (*, p

The histone variant macroH2A1 (mH2A1) is involved in cellular growth, differentiation and reprogramming, but the underlying molecular mechanisms are a matter of debate. Different roles of mH2A1 in gene expression may relate to functional differences of its two splicing isoforms, mH2A1.1 and mH2A1.2. Here, we map for the first time genome-wide localization of endogenous mH2A1.1 and link the distribution of mH2A1.1 to control of gene expression in human breast cancer cells. In addition to localization shared with mH2A1.2 to facultative heterochromatin, mH2A1.1 specifically associates with regulatory elements required for gene activation, super-enhancers and promoters of highly expressed genes. Depending on the recruitment profile of mH2A1.1 to these elements, selective depletion of mH2A1.1 up- or downregulates its target genes. mH2A1.1 represses transcription when its binding is spread over the entire gene and promoter, and activates transcription when its binding is strictly confined to the transcription start site (TSS). Notably, RNA Polymerase II was frequently in pause at mH2A1.1-activated genes. Functionally, mH2A1.1-dependent regulation of a subset of paused genes impedes mammary tumor cell migration. Molecular mechanisms of mH2A1.1 function at the TSS uncovered by our study define an intriguing new mode of transcription regulation in cancer cells. Author Summary Control of gene expression driving cellular functions from differentiation to epistasis and causing, when dysfunctional, uncountable diseases, relies on modifications of chromatin structure. One key element enabling chromatin plasticity is the replacement of canonical histones by histone variants. Among histone variants macroH2A1 (mH2A) is an extraordinary H2A variant possessing a large non-histone domain placed outside of the nucleosome. Two splicing isoforms, mH2A1.1 and mH2A1.2, are produced, but these are rarely studied separately because they only differ in a 30 amino acid region and are difficult to distinguish experimentally, which likely explains contradictory functions reported in the literature. Here, we take advantage of a mH2A1.1 specific antibody to generate the first genome-wide chromatin-associated map of this histone variant in the invasive breast cancer cells line MDA-MB231. We confirm that mH2A1.1, like mH2A1.2, is enriched at facultative heterochromatin in agreement with its reported role as a repressor. However, we discovered that unlike its splicing isoform, mH2A1.1 specifically binds to super-enhancers and the transcription start site of highly transcribed genes. mH2A1.1 is necessary for regulating transcription of these genes. At the cellular level, we demonstrate that mH2A1.1 inhibits migration capacity of highly metastatic breast cancer cells. Our study characterizes for the first time binding profiles of mH2A1.1 that are linked to regulation of gene expression, thereby providing a new molecular mechanisms which govern the plasticity of human tumor cells.

Controlling cholesterol levels is a major challenge in human health, since hypercholesterolemia can lead to serious cardiovascular disease. Drugs that target carbohydrate metabolism can also modify lipid metabolism and hence cholesterol plasma levels. In this sense, dichloroacetate (DCA), a pyruvate dehydrogenase kinase (PDK) inhibitor, augments usage of the glycolysis-produced pyruvate in the mitochondria increasing oxidative phosphorylation (OXPHOS). In several animal models, DCA decreases plasma cholesterol and triglycerides. Thus, DCA was used in the 70 s to treat diabetes mellitus, hyperlipoproteinemia and hypercholesterolemia with satisfactory results. However, the mechanism of action remained unknown and we describe it here. DCA increases LDLR mRNA and protein levels as well as LDL intake in several cell lines, primary human hepatocytes and two different mouse models. This effect is mediated by transcriptional activation as evidenced by H3 acetylation on lysine 27 on the LDLR promoter. DCA induces expression of the MAPK ERK5 that turns on the transcription factor MEF2. Inhibition of this ERK5/MEF2 pathway by genetic or pharmacological means decreases LDLR expression and LDL intake. In summary, our results indicate that DCA, by inducing OXPHOS, promotes ERK5/MEF2 activation leading to LDLR expression. The ERK5/MEF2 pathway offers an interesting pharmacological target for drug development. This work was supported by a grant from Fondation de France (0057921), fellowship from the Higher Education Commission, Pakistan (AK) and fellowships from the Ministère de l’Enseignement Supérieur et de la Recherche (MESR) (DNV).