ADAR2 induces reproducible changes in sequence and abundance of mature microRNAs in the mouse brain

Abstract

Adenosine deaminases that act on RNA (ADARs) deaminate adenosines to inosines in double-stranded RNAs including miRNA precursors. A to I editing is widespread and required for normal life. By comparing deep sequencing data of brain miRNAs from wild-type and ADAR2 deficient mouse strains, we detect editing sites and altered miRNA processing at high sensitivity. We detect 48 novel editing events in miRNAs. Some editing events reach frequencies of up to 80%. About half of all editing events depend on ADAR2 while some miRNAs are preferentially edited by ADAR1. Sixty-four percent of all editing events are located within the seed region of mature miRNAs. For the highly edited miR-3099, we experimentally prove retargeting of the edited miRNA to novel 3′ UTRs. We show further that an abundant editing event in miR-497 promotes processing by Drosha of the corresponding pri-miRNA. We also detect reproducible changes in the abundance of specific miRNAs in ADAR2-deficient mice that occur independent of adjacent A to I editing events. This indicates that ADAR2 binding but not editing of miRNA precursors may influence their processing. Correlating with changes in miRNA abundance we find misregulation of putative targets of these miRNAs in the presence or absence of ADAR2.

Post-transcriptional regulation of LINE-1 retrotransposition by AID/APOBEC and ADAR deaminases

“…Inosine is interpreted as guanosine by several cellular mechanisms, such as splicing, translation and reverse transcription (Bass et al 1997(Bass et al , 2002(Bass et al , 2002Goodman et al 2012;Nishikura 2016). As a consequence, A-to-I editing can potentially change the meaning of a specific codon, alter the splicing process, impair the microRNA processing and function, and affect several other aspects of RNA metabolism (Mannion et al 2015;Nishikura 2016;Chen and Carmichael 2012;Wang et al 2013;Ivanov et al 2015;Hogg et al 2011;Rueter et al 1999;Chen et al 2008b;Prasanth et al 2005;Capshew et al 2012;Vesely et al 2012Vesely et al , 2014. A-to-I editing occurs mostly in the non-coding regions (3' UTRs and introns) containing inversely oriented repeat elements, such as Alu and LINE-1 (Bazak et al 2014;Ramaswami et al 2012).…”

Accumulation of nuclear ADAR2 regulates adenosine-to-inosine RNA editing during neuronal development

“…Although most A-to-I editing sites within these repetitive elements have not been assigned specific functions, there are examples where editing leads to alternative splicing and exonization of introns (Lev-Maor et al, 2007). The non-coding microRNAs (miRNAs) are also edited, particularly in the targeting seed sequence (Alon et al, 2012;Ekdahl et al, 2012;Kawahara et al, 2007;Vesely et al, 2014). The ADAR enzymes target the nuclear miRNA precursors and can affect their maturation or mRNA-targeting capacity.…”

A comparative analysis among ADAR mutant mice reveals site-specific regulation of RNA editing

“…This method was laborious, but has the advantage of quantifying the editing ratio at all sites to be investigated with high precision, regardless of the expression level of each target RNA. To list the conserved editing sites in CDS and miRNAs, we initially selected A-to-I RNA editing sites reported to be conserved in humans and mice (Kawahara et al 2008b;Chiang et al 2010;Maas et al 2011;Alon et al 2012;Danecek et al 2012;Ekdahl et al 2012;Gu et al 2012;Daniel et al 2014;Pinto et al 2014;Ramaswami and Li 2014;Vesely et al 2014;Nishikura 2016;Terajima et al 2016). Next, by Sanger sequencing we preliminarily examined RNA editing in the cerebral cortex and spleen for sites reported to be possibly edited in mice and that may be conserved in humans (HIST2H2AB L/L, HIST2H2AC N/S, ZNF397 N/D, miR-542-3p, miR-574-5p and miR-708-3p) (Cattenoz et al 2013;Vesely et al 2014;Hosaka et al 2018).…”

“…In contrast, ADAR1 has two isoforms: a short p110 isoform that is mainly localized in the nucleus and is highly expressed in the brain, and a long p150 isoform that is mainly localized in the cytoplasm and is highly expressed in the thymus and spleen, where ADAR2 is expressed at low level (Huntley et al 2016;Nakahama et al 2018). In addition to the different expression patterns of ADARs, it is known that ADAR1 and ADAR2 regulate RNA editing in a competitive manner in some cases, which has made it difficult to determine the contribution of each ADAR to the RNA editing of each conserved site in vivo (Kawahara et al 2007b;Riedmann et al 2008;Wahlstedt et al 2009;Vesely et al 2014;Picardi et al 2015;Huntley et al 2016;Tan et al 2017). One simple solution to this problem may be by comparing the editing ratio of each site among wild-type (WT), Adar1-deficient (Adar1 -/-) and Adar2-deficient (Adar2 -/-) mice.…”