Functional genomic approaches to study human adenoviruses and frontier research on miRNAs

Abstract

Viruses are self-organized, symmetrical and rather simple structures, which can cause devastating diseases in humans and animals. They are able to readily react to cellular cues by changing gene expressions and conformations, which in turn enables dynamic interactions with their hosts. A large body of research has shown that the full complexity of viruses is unfolded as soon as they interact with host cells. Although studies of virus-host interactions have emerged as a key driving force in the research of infectious diseases, our understanding of the systems properties of viral infections has remained incomplete. Human adenoviruses (HAdVs) are most commonly associated with respiratory and gastrointestinal tract infections. In addition, HAdVs are the most common viral vectors in clinical gene transfer trials. Although HAdVs have been well characterized, there are many unresolved aspects, such as receptor usage by species B viruses (HAdV-B), dynamic expression profiles of viral and cellular proteins in the infected cells, and how this can be influenced by signal transduction, or virally encoded miRNAs. To address some of these issues, we employed comprehensive functional genomics
and biochemistry approaches to obtain a global picture of the infection dynamics of HAdVs in cultured cells, from entry to gene regulations. In the stage of HAdV entry, we
studied the interaction of HAdV-B3/HAdV-B7 fiber knob (FK) to cellular receptor CD46 by using Biacore and quantitative microscopy. In the progress of dynamic interactions and regulations, we targeted three molecular levels including miRNA, mRNA and proteins by using quantitative transcriptomics and proteomics of both viruses and host
infected human epithelial cells. The transcriptomics analysis was carried out by using the Agilent microarray and deep sequencing while proteomics was performed by using
iTRAQ-8plex labeling of peptides or label-free method, followed by LC-MS/MS analyses at different time points of infection with different HAdV types. Subsequently, the networks were analyzed by using Metacore. For identifying novel miRNAs, we applied both in silico predictions and multi-detection methods including RT-PCR, miRNA microarray, deep sequencing and Northern blot analysis. Our results show that direct binding of HAdV-B3/HAdV-B7-FK to CD46 occurs by an avidity mechanism and resolves the controversy about CD46 being a receptor for HAdV-B. Based on quantitative transcriptomics, proteomics and miRNAs, we defined some common networks of HAdV serotypes of 3, 5, 11p and 35, such as cytoskeleton and cell cycle. We found new components such as galectin-1 (gal-1) and galectin-3 (gal-3), which were triggered to be secreted into the cell culture supernate. Preliminary data suggested that gal-1 and gal-3 interacted with CD46 as well as viral capsids
including hexon and penton. Concerning the miRNAs, we found four small viral miRNAs (vmiRNAs) derived from VA-RNAI and one small vmiRNA derived from a noncoding viral sequence. Based on in silico analyses of potential targets for these
vmiRNAs and transcriptomics analysis we found that there are many potential cellular and viral gene targets. In summary, this study addressed several systems aspects of
HAdV infections, including post-translational modifications, potential anti-viral factors of innate immunity, small vmiRNAs, and it clarified the role of CD46 for HAdV-B infections.