Substrate recognition by ADAR1 and ADAR2.

RNA editing catalyzed by ADAR1 and ADAR2 involves the site-specific conversion of adenosine to inosine within imperfectly duplexed RNA. ADAR1- and ADAR2-mediated editing occurs within transcripts of glutamate receptors (GluR) in the brain and in hepatitis delta virus (HDV) RNA in the liver. Although the Q/R site within the GluR-B premessage is edited more efficiently by ADAR2 than it is by ADAR1, the converse is true for the +60 site within this same transcript. ADAR1 and ADAR2 are homologs having two common functional regions, an N-terminal double-stranded RNA-binding domain and a C-terminal deaminase domain. It is neither understood why only certain adenosines within a substrate molecule serve as targets for ADARs, nor is it known which domain of an ADAR confers its specificity for particular editing sites. To assess the importance of several aspects of RNA sequence and structure on editing, we evaluated 20 different mutated substrates, derived from four editing sites, for their ability to be edited by either ADAR1 or ADAR2. We found that when these derivatives contained an A:C mismatch at the editing site, editing by both ADARs was enhanced compared to when A:A or A:G mismatches or A:U base pairs occurred at the same site. Hence substrate recognition and/or catalysis by ADARs could involve the base that opposes the edited adenosine. In addition, by using protein chimeras in which the deaminase domains were exchanged between ADAR1 and ADAR2, we found that this domain played a dominant role in defining the substrate specificity of the resulting enzyme.