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.

Interestingly, in addition to gene-neighborhoods with multiple tandem divergent SUKH superfamily genes, in several bacteria, we also observed notable lineage-specific expansions of SUKH domain proteins (e.g. 21 paralogs in Gemmata obscuriglobus, 20 paralogs in C. gingivalis and 15 in S. albus). These observations also make sense in light of the above toxin-immunity protein hypothesis: while the SUKH superfamily gene adjacent to a nuclease or CDI toxin gene is likely to provide immunity to the 'self' toxin, the supernumerary SUKH superfamily genes, which occur as tandem arrays or as isolated versions, might provide immunity against other 'non-self' toxins delivered by competing bacteria in the environment. Such associations of multiple distinct immunity genes have also been observed in the case of plasmid-borne colicin gene operons (8) . Other features of the genomics of the SUKH superfamily also support this proposal. Gene neighborhoods encoding SUKH proteins and linked nucleases or CDI toxin are highly variable in terms of being present or absent between different strains of the same species or between different closely related species which share an otherwise similar genomic organization. Secondly, there appear to have been recent duplications of entire loci encompassing these gene-neighborhoods within the same genome in several bacteria (Supplementary Data). This kind of phyletic and genomic polymorphism is also typical of loci involved in inter-and intra-genomic competition such as toxin-antitoxin, restriction-modification and virulence toxin systems (6, 9, 10, 15) , suggesting that even systems with SUKH superfamily proteins might have comparable roles. To test this proposal further, as the first line of investigation, we aimed at exploring further the link between nucleases and the SUKH domain proteins. While the polymorphic C-terminal domains of two CDI toxins have been characterized as nucleases, the C-terminal domains of those CDI toxins which are found linked to the SUKH superfamily immunity proteins have not be characterized. We speculated that these domains, along with some of the other uncharacterized domains in proteins encoded by conserved gene-neighborhoods containing a SUKH superfamily gene, might be as yet uncharacterized nuclease domains. As a second line of investigation we sought to uncover those among the associated uncharacterized domains, which might have a role in distinct toxintrafficking mechanisms, comparable to the two-partner system used by the proteobacterial CDIs. Therefore, to accomplish these two objectives and identify other components of these systems we resorted to systematic sequence analysis of the uncharacterized proteins recovered in the above gene-neighborhood analysis.