m6A regulates ADAR1 mediated RNA editing during macrophage activation

Dynamic mRNA modifications have a crucial impact on the fate and metabolism of a transcript, including splicing, subcellular localization, stability, and translation. N6-methyladenosine (m6A) is one of the most prominent modifications found in mRNA. It is catalyzed by Methyltransferase 3 (METTL3), which forms part of the core (catalytic) m6A writer complex together with Methyltransferase 14 (METTL14) and Wilms' tumor 1-associating protein (WTAP). The m6A writer complex has local preference for A within the consensus DRACH (D = G/A/U, R = G/A, A, C, H = A/U/C) motif. Within the transcript, m6A sites are enriched after the last splice site of the last exon and in the 3' untranslated region (3'UTR) in proximity to the stop codon. Finally, m6A can also be removed from a transcript by eraser proteins fat mass and obesity-associated protein (FTO) and AlkB Homolog 5, RNA Demethylase (ALKBH5). A variety of m6A reader proteins in different subcellular localizations recognize m6A and mediate m6A's regulatory function on mRNA metabolism. Established functions of m6A include promoting transcript decay, transcript stability, translation, splicing, and nuclear export. Finally, recent studies note the role of m6A in preventing formation of anomalous double stranded RNA (dsRNA) structures and their recognition by innate immune sensors. Adenosine deamination is also a highly prominent mRNA modification. The majority of A-to-I deamination is catalyzed by ADAR1 protein which deaminates A to inosine (I) in the context of dsRNA. Inosine pairs preferentially with cytosine and therefore is decoded as guanosine (G) by the cell's translational machinery. Two different ADAR1 isoforms exist, the constitutively expressed (and nuclear) p110 isoform and the interferon-inducible (and mostly cytoplasmic) p150 isoform which plays an important role during interferon responses. A-to-I editing occurs most frequently in repetitive elements such as short interspersed nuclear elements (SINE), introns, and the 3'UTR, due to their propensity to form dsRNA structures. Importantly, these dsRNA structures can activate cytoplasmic dsRNA sensors like MDA5. ADAR1 binding and editing of such structures prevents activation of innate immune sensing pathways. Both m6A and A-to-I deamination target adenosine, suggesting a potential for interactions. While several studies have investigated each modification independently, questions regarding their interplay remain unresolved (and conflicting) in part because the assignment of m6A or A-to-I instances used different (and incompatible) methodologies. In our LPS/IFN-g;-stimulated RAW 264.7 macrophages, ~60% of expressed genes (n = ~12,800) are annotated as m6A-modified and ~12% as harboring A-to-I editing sites, with ~10% of genes classified as harboring both mark types. Recent advances in direct RNA sequencing have made it possible to directly assess the co-occurrence of these two adenosine modifications within individual RNA molecules. Here we revisit this problem focusing on the transcriptome of macrophages for three reasons: First, macrophages readily adapt their phenotype and function, in response to changes in tissue physiology and external stimuli. This means that we can match changes in modification events to specific (and well understood) functional outputs, from the upregulation of activation markers to the phagocytic ability of cells proficient or deficient in modification enzymes. Second, there is already a body of literature on m6A and A-to-I occurrences within macrophage transcriptomes (which can serve as a positive control, despite the fact that their interplay on individual RNA molecules cannot be assessed due to earlier identification methods). And third, macrophage cell lines (including the RAW 264.7 [RAW] cell line studied here) have been used in the past for studies on RNA processing, providing a system in which post-transcriptional regulatory mechanisms can be mechanistically assessed. Here, we generated RAW cells deficient for METTL3 or ADAR1, and used Nanopore direct RNA sequencing (validated with Illumina sequencing, where appropriate), to characterize the relationship between these two mRNA modifications and their impact on macrophage function.