ABSTRACT: 5-hydroxymethylcytosine is linked with expression of brain-specific protein-coding genes but not brain-specific microRNA host genes (RNA-Seq)
Project description:5-hydroxymethylcytosine is linked with expression of brain-specific protein-coding genes but not brain-specific microRNA host genes
Project description:5-hydroxymethylcytosine is linked with expression of brain-specific protein-coding genes but not brain-specific microRNA host genes [hMeDIP-Seq]
Project description:Background: Small nucleolar RNAs (snoRNAs) are mid-size non-coding RNAs required for ribosomal RNA modification, implying a ubiquitous tissue distribution linked to ribosome synthesis. However, increasing numbers of studies identify extra-ribosomal roles of snoRNAs in modulating gene expression, suggesting more complex snoRNA abundance patterns. Therefore, there is a great need for mapping the snoRNome in different human tissues as the blueprint for snoRNA functions. Results: We used a low structure bias RNA-Seq approach to accurately quantify snoRNAs and compare them to the entire transcriptome in seven healthy human tissues (breast, ovary, prostate, testis, skeletal muscle, liver and brain). We identify 475 expressed snoRNAs categorized in two abundance classes that differ significantly in their function, conservation level and correlation with their host gene: 390 snoRNAs are uniformly expressed and 85 are enriched in the brain or reproductive tissues. Most tissue-enriched snoRNAs are embedded in lncRNAs and display strong correlation of abundance with them, whereas uniformly expressed snoRNAs are mostly embedded in protein-coding host genes and are mainly non- or anticorrelated with them. 59% of the non-correlated or anticorrelated protein-coding host gene/snoRNA pairs feature dual-initiation promoters, compared to only 16% of the correlated non-coding host gene/snoRNA pairs. Conclusions: Our results demonstrate that snoRNAs are not a single homogeneous group of housekeeping genes but include highly regulated tissue-enriched RNAs. Indeed, our work indicates that the architecture of snoRNA host genes varies to uncouple the host and snoRNA expressions in order to meet the different snoRNA abundance levels and functional needs of human tissues.
Project description:Over 40 % of microRNAs are located in introns of coding genes, and many intronic microRNAs are co-regulated with their host genes. In such cases of co-regulation, the products of host genes and their intronic microRNAs can cooperate to coordinately regulate biologically important pathways. Therefore, we screened intronic microRNAs dysregulated in liver of obese mouse models to identify previously uncharacterized coding host genes that may contribute to the pathogenesis of obesity-associated insulin resistance and type 2 diabetes mellitus. Our approach identified that expression of both Ectodysplasin A (Eda), the causal gene of X-linked hypohidrotic ectodermal dysplasia (XLHED; MIM 305100) and its intronic microRNA, miR-676, was strongly increased in liver of obese mouse models. Moreover, hepatic EDA expression is increased in obese human subjects, reduced upon weight loss, and its hepatic expression correlates with systemic insulin resistance. Eda expression in murine liver is controlled via PPARg activation, increases in circulation and promotes JNK activation and inhibitory serine phosphorylation of IRS1 in skeletal muscle. Consistently, bi-directional modulation of hepatic Eda expression in mouse models affects systemic glucose metabolism with alterations of muscle insulin signaling, revealing a novel role of EDA as an obesity-associated hepatokine, which impairs insulin sensitivity in skeletal muscle.
Project description:Strand-specific RNA-seq libraries from different sea lamprey tissues (brain, heart, liver, kidney, ovary, testis) were produced in order to enhance the genome annotation for protein coding genes.
Project description:Approximately half of all microRNAs reside within intronic regions and are often co-transcribed with their host genes. However, most studies on intronic microRNAs focus on individual microRNAs, and conversely most studies on protein-coding and non-coding genes frequently ignore any intron-derived microRNAs. We hypothesize that the individual components of such multi-genic loci may play cooperative or competing roles in driving disease progression, and that examining the combinatorial effect of these components would uncover deeper insights into their functional importance. To address this, we perform systematic analyses of intronic microRNA:host loci in colon cancer. We observe that the FTX locus, comprising of a long non-coding RNA FTX and multiple intronic microRNAs, is highly upregulated in cancer and demonstrate that cooperativity within this multi-component locus promotes cancer growth. In addition, we show that FTX interacts with DHX9 and DICER and delineate its novel roles in regulating A-to-I RNA editing and microRNA expression. These results show for the first time that a long non-coding RNA can regulate A-to-I RNA editing, further expanding the functional repertoire of long non-coding RNAs. We further demonstrate the inhibitory effects of intronic miR-374b and -545 on the tumor suppressors PTEN and RIG-I to enhance the proto-oncogenic PI3K-AKT signaling. Finally, we show that intronic miR-421 may exert an autoregulatory effect on miR-374b and -545. Taken together, our data unveil the intricate interplay between intronic microRNAs and their host transcripts in the modulation of key signaling pathways and disease progression, adding new perspectives to the functional landscape of multi-genic loci.
Project description:Evidence from a few genes of diverse species suggests that marsupial X-chromosome inactivation (XCI) is characterized by exclusive, but leaky, inactivation of the paternally derived X chromosome. To comprehensively study the mechanism of marsupial XCI, we profiled parent-of-origin-specific-allele expression, DNA methylation, and histone modifications in opossum fetal brain and extra-embryonic membranes. The majority (152/176) of X-linked genes exhibited paternally imprinted expression with nearly 100% maternal allele expression, whereas 24 loci (14%) escaped inactivation showing varying levels of biallelic expression. In addition to regulation by the non-coding RSX transcript, strong depletion of H3K27me3 at escaper gene loci indicates that histone states also influence opossum XCI. Notably, the opossum does not show an association between X-linked gene expression and promoter DNA methylation. Our study provides the first comprehensive catalogue of parent-of-origin expression status for X-linked genes in a marsupial and sheds light on the regulation and evolution of imprinted XCI in mammals. Profiling of expression level and allele-specific expression ratios in embryonic day 13 opossum (Monodelphis domestica) fetal brain and extra-embyonic membranes by Illumina RNA-seq
Project description:Evidence from a few genes of diverse species suggests that marsupial X-chromosome inactivation (XCI) is characterized by exclusive, but leaky, inactivation of the paternally derived X chromosome. To comprehensively study the mechanism of marsupial XCI, we profiled parent-of-origin-specific-allele expression, DNA methylation, and histone modifications in opossum fetal brain and extra-embryonic membranes. The majority (152/176) of X-linked genes exhibited paternally imprinted expression with nearly 100% maternal allele expression, whereas 24 loci (14%) escaped inactivation showing varying levels of biallelic expression. In addition to regulation by the non-coding RSX transcript, strong depletion of H3K27me3 at escaper gene loci indicates that histone states also influence opossum XCI. Notably, the opossum does not show an association between X-linked gene expression and promoter DNA methylation. Our study provides the first comprehensive catalogue of parent-of-origin expression status for X-linked genes in a marsupial and sheds light on the regulation and evolution of imprinted XCI in mammals. Profiling of four histone modifications in embryonic day 13 opossum (Monodelphis domestica) fetal brain by Illumina ChIP-seq
Project description:Endogenous retroviruses (ERVs), which make up 8% of the human genome, have been proposed to participate in the control of gene regulatory networks. In this study, we find a region- and developmental stage-specific expression pattern of ERVs in the developing human brain, which is linked to a transcriptional network based on ERVs. We demonstrate that almost ten thousand, primarily primate-specific ERVs, act as docking platforms for the epigenetic co-repressor protein TRIM28 in human neural progenitor cells, which results in the establishment of local heterochromatin. Thereby, TRIM28 represses ERVs and consequently regulates the expression of neighboring genes. These results uncover a gene regulatory network based on ERVs that participates in control of gene expression of protein-coding transcripts important for brain development.
Project description:Tet3 is an Fe2+-dependent enzyme that oxidizes genomic 5-methylcytosine to 5-hydroxymethylcytosine with the help of alpha-ketoglutarate and oxygen. It is the most abundant Tet enzyme in differentiated tissues including brain. Adult brain contains the highest 5-hydroxymethylcytosine levels. How alpha-ketoglutarate is made available for the oxidation of mC in brain cells and how the Tet activity is linked to neural activity are unsolved questions. Our experiments with full mouse brains show that Tet3 interacts in the nucleus directly with selected enzymes of the mitochondrial citric acid cycle. This leads to the formation of isocitrate. Tet3 also interacts with aspartate aminotransferase, which produces oxaloacetate. Although oxaloacetate and isocitrate are biosynthetic alpha-ketoglutarate precursors, they function as inhibitors of Tet3 and are needed to protect the reactive Fe2+ center from degrading DNA. The supply of Tet3 with alpha-ketoglutarate is established by a direct interaction of Tet3 with glutamate dehydrogenase (Glud1), which converts the neurotransmitter glutamate directly into alpha-ketoglutarate. This links Tet3 function to neural activity.