Project description:Active DNA demethylation, the enzymatic removal of methyl groups from DNA initially triggered by their successive oxidation, is a fundamental process critical to regulating cellular identity. The major enzymes involved in this process have been recently elucidated, but other important components in this pathway and the independent contributions of oxidized 5-methylcytosine derivatives to the regulation of gene expression remain unclear. A major obstacle to the elucidation of this pathway is the low abundance of these derivatives and limited power of current detection technologies. Here, we first develop a technique for their detection with APOBEC3A conversion and sequencing (APOBEC-seq) to directly discriminate deamination-insensitive oxidized cytosines from methylated or unmethylated counterparts, which now facilitates robust DNA demethylation analysis suitable to interrogate the active DNA demethylation pathway by a comprehensive set of experiments. Our results demonstrate that APOBEC-seq is a powerful tool for the detection of oxidized cytosines, revealing insights about the relationship between oxidized gene re-activation and demethylation as a function of TDG, APEX1, and the MBD family of proteins. We also report a ubiquitous binding of TDG to all active unmethylated and unoxidized promoters – regardless of RNA polymerase subtype – suggesting that, together with TET family of enzymes, the presence of TDG at these regions may safeguard active genes from DNA hypermethylation-induced silencing and may explain the failures of CRISPR/dCas9-based DNA methylation editing tools to introduce persistent DNA hypermethylation at targeted promoters. We also report an interaction between MBD3/NuRD and TDG which recruits TDG to its targets and may shed light on the critical role of MBD3 in development. Finally, we apply APOBEC-seq to profile oxidation in vivo in the mouse cortex and report a dramatic tissue-specific pattern of oxidation in genes and enhancers.

Project description:Purpose: RNA editing by ADAR1 is essential for hematopoietic development. The goals of this study were firstly, to identify ADAR1-specific RNA-editing sites by indentifying A-to-I (G) RNA editing sites in wild type mice that were not edited or reduced in editing frequency in ADAR1 deficient murine erythroid cells. Secondly, to determine the transcription consequence of an absence of ADAR1-mediated A-to-I editing. Methods: Total RNA from E14.5 fetal liver of embryos with an erythroid restricted deletion of ADAR1 (KO) and littermate controls (WT), in duplicate. cDNA libraries were prepared and RNA sequenced using Illumina HiSeq2000. The sequence reads that passed quality filters were analyzed at the transcript level with TopHat followed by Cufflinks. qRTâPCR validation was performed using SYBR Green assays. A-to-I (G) RNA editing sites were identified as previously described by Ramaswami G. et al., Nature Methods, 2012 using BurrowsâWheeler Aligner (BWA) followed by ANOVA (ANOVA). RNA editing sites were confirmed by Sanger sequencing. Results: Using an optimized data analysis workflow, we mapped about 30 million sequence reads per sample to the mouse genome (build mm9) and identified 14,484 transcripts in the fetal livers of WT and ADAR1E861A mice with BWA. RNA-seq data had a goodness of fit (R2) of >0.7, p<0.0001 between biological duplicates per genotype. Clusters of hyper-editing were onserved in long, unannotated 3'UTRs of erythroid specific transcripts. A profound upregulation of interferon stimulated genes were found to be massively upregulated (up to 5 log2FC) in KO fetal liver compared to WT. 11.332 (6,894 novel) A-to-I RNA editing sites were identified when assessing mismatches in RNA-seq data. Conclusions: Our study represents the first detailed analysis of erythroid transcriptomes and A-to-I RNA editing sites, with biologic replicates, generated by RNA-seq technology. A-to-I RNA editing is the essential function of ADAR1 and is required to prevent sensing of endogenous transcripts, likely via a RIG-I like receptor mediated axis. Fetal liver mRNA profiles of E14.5 wild type (WT) and ADAR Epor-Cre knock out mice were generated by deep sequencing, in duplicate using Illumina HiSeq 2000.

Dataset Information

Genome-wide identification and functional analysis of Apobec-1 mediated C-to-U RNA editing in mouse small intestine and liver

Dataset's files

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets