A novel immunity system for bacterial nucleic acid degrading toxins and its recruitment in various eukaryotic and DNA viral systems

Zhang, Dapeng; Iyer, Lakshminarayan M.; Aravind, L.

Other domains with a possible role in nucleic acid modifications. We found three other families of domains in proteins that were encoded by genes which occupied positions adjacent to SUKH superfamily genes in certain predicted operons, equivalent to positions of the genes encoding the above nucleases. Additionally, these families of domains are also found as representatives of the polymorphic C-terminal module of the proteobacterial CDIs. Together these observations hinted that they are potentially uncharacterized enzymatic domains operating on nucleic acids. PSI-BLAST and JACKHMMER searches showed that the first of these families belonged to the nucleotide deaminase superfamily that includes RNA-editing enzymes, such as the APOBECs and DNA-modifying enzyme AID of vertebrates. Hence, like the nucleases, these enzymes are likely to function as toxins that mutate nucleic acids in the target cells. We discuss the natural history of these enzymes in a separate article (Iyer LM, Zhang D, Aravind L, manuscript in preparation). The second of these families prototyped by the B. cereus protein bcere0017_55840 (gi: 229119351) is characterized by a conserved signature [NS]HH followed by another conserved histidine (Supplementary Data). Although we were unable to unify this family with any of the other nuclease folds, the presence of the HH motif typical of many of the above families of HNH/EndoVII fold nucleases might point to a divergent relationship with those proteins. The third of these families, typified by the CDI system from P. luminiscens (gi: 37524545) includes a globular domain of 170-200 amino acids that might define yet another uncharacterized nucleic acid-modifying domain (CdiAC in Figure 5 , Supplementary Data).