Rare protein variants encoded in hidden genes that are not revealed by a conventional genomic analysis, and which are expressed at very low levels by complex genetic mechanisms, can be considered to be a type of biological dark matter. They are not readily apparent or easily detectable, and yet they may exert significant effects in biological systems, analogous to the gravitational role of dark matter in physics. Our group has focused on two types of events that can underlie such phenomena, known to be utilized by RNA viruses in particular: 1) overlapping genes, simultaneously encoded in a single strand of DNA or RNA, and 2) the extension of a protein by suppression of one or more stop codons, leading to an elongated isoform of the protein. We have identified various sites in viruses and in human genes, where both of these mechanisms may be occurring in a single hidden gene. A special case involves incorporation of the dietary trace element selenium into proteins; the rare amino acid selenocysteine is encoded in RNA by the UGA codon, which normally is a stop signal for protein synthesis. Our recent publication in Curr Top Med Chem (http://www.ncbi.nlm.nih.gov/pubmed/26369818) suggests a solution to a 20 year old mystery as to how RNA viruses might accomplish this: simply by hijacking the mRNA of a human selenoprotein via antisense interactions. We show that both HIV-1 and Ebola virus can engage in such interactions, and that in both viruses the sites of antisense complementarity to the cellular mRNAs are located in close proximity to highly conserved UGA “stop” codons. In vitro assays show that these viral UGA codons can be suppressed (i.e., translated as an amino acid) in a selenium-dependent manner. This extends a body of work which has revealed mechanisms by which HIV-1 and Ebola can compete with their human hosts for selenium, which may be decreasing in the food chain due to acid rain. Our predictions from the mid-1990s that selenium would be a particularly important dietary factor for survival in HIV-AIDS and Ebola infection have been borne out by numerous clinical studies in the case of HIV-1, and unofficial reports in the case of Ebola. The fact that these predictions were based upon the discovery of hidden viral genes shows that the identification of biological dark matter can lead to significant and useful biological insights. Computational approaches to identifying such proteinaceous dark matter regions in genomic sequence data will be discussed.

Computational approaches to identifying biological dark matter, exemplified by “hidden genes" in HIV-1 and Ebola

251st National Meeting (2016)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.