Do Viruses Exchange Genes across Superkingdoms of Life?

Malik, Shahana S.; Azem-e-Zahra, Syeda; Kim, Kyung Mo; Caetano-Anollés, Gustavo; Nasir, Arshan

A simple comparative genomic analysis calculating the spread of viral protein domain structure FSFs in reported host and non-host cellular proteomes revealed that proteomes of virus hosts harbored several viral hallmark proteins necessary for virion assembly and successful viral infection cycles ( Table 1) . These viral hallmark proteins however were absent from the majority of closely-related organisms within the same superkingdom indicating that their rare presence in some host cellular proteomes could be an outcome of virus-to-cell gene transfer. In turn, proteomes not presumed to serve as natural hosts for viruses also shared homologous FSFs with viral proteomes. These FSFs included both viral-and cell-like proteins. This was especially obvious for FSFs shared between bacterioviruses and eukaryotic proteomes indicating either direct or indirect cross-superkingdom genetic exchange. This sharing could have been driven by the endosymbiotic and pathogenic lifestyle of bacteria that sometimes associate with eukaryotic cells. Interestingly, despite sharing the same ecosystem with Bacteria (e.g., the human gastrointestinal tract, Lurie-Weinberger and Gophna, 2015), our results suggested that little or no FSF sharing (or genetic exchange) occurred between archaeoviruses and the proteomes of Bacteria and Eukarya (e.g., AB was 1 and AE 0 in archaeoviruses, Figure 2 ). Bacteria are established pathogens and (endo)-symbionts of eukaryotes but Archaea are not known to infect eukaryotic organisms (Aminov, 2013) . The membranes of Archaea also differ in lipid composition with the membranes of Bacteria and Eukarya (ether-linked vs. esterlinked, Jain et al., 2014) , along with other differences (Gill and Brinkman, 2011) . These differences could therefore pose a barrier for archaeoviruses to cross/traverse bacterial and eukaryal membranes and participate in horizontal genetic exchange. In contrast and thanks to the relatively similar lipid organization of Bacteria and Eukarya, bacterioviruses may either directly traverse eukaryotic membranes or alternatively transduce benign bacterial species into human pathogens by transferring virulence factors , which in turn infect Eukarya. As stated by Gill and Brinkman (2011) , "eukaryotic viruses infect eukaryotes, and bacteriophages transduce Bacteria, which allows them to infect Eukarya". Moreover, there are many more known examples of obligate and facultative intracellular bacteria (e.g., Chlamydia, Rickettsia, Mycoplasma) in eukaryotes. Therefore, viral infection of bacterial endosymbionts or prophage integration into their genomes will create more opportunities for genetic interactions with eukaryoviruses and eukaryotic proteomes explaining more "crosstalk" between bacterioviruses and Eukarya (and perhaps eukaryoviruses and Bacteria) than between archaeoviruses and Bacteria/Eukarya. However, it must be noted that both archaeoviruses and archaeal species are relatively underrepresented in sequence databases. Thus, a global picture of the true contribution of archaeoviruses and archaeal proteomes to protein structure space remains elusive despite increased metagenome sequencing efforts. Indeed, Archaea constitute an important part of the animal microbiota (Hoffmann et al., 2013; Lurie-Weinberger and Gophna, 2015) , an ecosystem that is considered a "hot spot" for genetic exchange (Shterzer and Mizrahi, 2015) . A large cohort of universal protein domains shared between archaeoviruses, bacterioviruses, eukar