Current Approaches on Viral Infection: Proteomics and Functional Validations

Zheng, Jie; Tan, Boon Huan; Sugrue, Richard; Tang, Kai

In addition to MS based proteomics, protein microarrays have also been considered as a set of rapidly evolving technologies capable of identifying PPI networks, quantitatively profiling protein expression levels, and complementing the high throughput proteomic data (Sobek et al., 2006; Pollard et al., 2007) . In the recent decade, much effort has been devoted to systematically study biochemical activities of proteins in a high throughput manner. Yet the major difficulties are screening an entire proteome by expressing clones, accommodating low volume proteins, and meanwhile retaining biochemical activities. Zhu et al. (2001) cloned 5800 open reading frames and purified corresponding proteins in order to construct a yeast proteome chip. GST-labeled proteins were immobilized onto the glass microscope slides through covalent attachment by using aldehyde-amine or nickel-HisX6 tags. In a parallel study, microarrays containing thousands of Cy3 or Cy5 labeled proteins were fabricated by a high-precision contact-printing robot to generate nanoliter protein spots on glass slide. Its applications, such as protein-small molecule and kinase-substrate interactions were successfully verified (MacBeath and Schreiber, 2000) . In recent years, both proteomics and protein microarrays are predominantly used as reliable tools in glycomic profiling with a screening effect (Mahal, 2008) . Characterization of protein-glycan interaction related to virus infection is another important task in the field of glycoproteomics and glycan microarrays. For instance, the binding specificity of influenza A virus with sialic acids is the main determinant for virus entry and species-dependent infection. To discover the detailed glycan-binding preference for the pandemic "triple reassortant"influenza A virus and a low infectious influenza A virus, Bateman et al. (2010) utilized MALDI TOF MS and GC-MS/MS to extensively map the linear or branched, N-or O-linked glycans expressed on the surface of primary swine respiratory epithelial cells (SRECs). Different sialic acid linkages, such as α-2, 3 or α-2, 6 linkage, were also determined by MS after corresponding sialidase digestions. Both NeuAc and NeuGc were identified on the SRECs surface with NeuAc occupying a much higher abundance. By applying glycomic microarray analysis, it revealed that both virus strains were ready to bind to NeuAcα2-6 glycans, sialylated polylactosamine and sialylated N -glycans. Therefore, the structural characterization of a wide variety of glycans by proteomics and functional microarray based on HA-sialic acid interactions could expand our knowledge on the molecular basis of virus-host interactions. In a similar study, Song et al. (2011) generated a total of 77 α-2, 3 or α-2, 6-sialylated structures incorporating different types of modified or nature sialic acids. All these sialylated glycans were examined by MALDI TOF MS and fabricated on a NHS-activated microarray glass slide. Several human influenza viruses, such as H1N1, H3N2 were tested on this glycan microarray, and both viruses were able to bind specifically to α-2, 6 linked sialic acid derivatives, α-2, 6 linked Neu5Ac, and α-2, 6 linked Neu5Ac9Lt. H1N1 displayed a broader range of binding specificities than H3N2 with an additional binding to α-2, 6-sialylatedNA2 structures. In this way integrated with proteomics, microarray screening of glycan-pathogen interactions with a wide variety of sialic acids provides an efficient way to explore glyca