Abstract
Protein, DNA, and RNA methyltransferases have an ever-expanding list of novel substrates and catalytic activities. Even within families and between homologs, it is becoming clear the intricacies of methyltransferase specificity and regulation are far more diverse than originally thought. In addition to specific substrates and distinct methylation levels, methyltransferase activity can be altered by complex formation with close homologs. We work with the N-terminal methyltransferase homologs NRMT1 and NRMT2. NRMT1 is a ubiquitously expressed distributive trimethylase. NRMT2 is a monomethylase expressed at low levels in a tissue-specific manner. They are both nuclear methyltransferases with overlapping consensus sequences but have distinct enzymatic activities and tissue expression patterns. Co-expression with NRMT2 increases the trimethylation rate of NRMT1, and here we aim to understand how this occurs. We use analytical ultracentrifugation to show that while NRMT1 primarily exists as a dimer and NRMT2 as a monomer, when co-expressed they form a heterotrimer. We use co-immunoprecipitation and molecular modeling to demonstrate in vivo binding and map areas of interaction. While overexpression of NRMT2 increases the half-life of NRMT1, the converse is not true, indicating that NRMT2 may be increasing NRMT1 activity by stabilizing the enzyme. Accordingly, the catalytic activity of NRMT2 is not needed to increase NRMT1 activity or increase its affinity for less preferred substrates. Monomethylation can also not rescue phenotypes seen with loss of trimethylation. Taken together, these data support a model where NRMT2 expression activates NRMT1 activity, not through priming, but by increasing its stability and substrate affinity.