dc.description.abstract | Rotavirus is one of the leading etiological agents of gastroenteritis in young of many species including humans worldwide and is responsible for about 600,000 infant deaths per annum. Rotavirus belongs to the Reoviridae family, and its genome is composed of 11 double-stranded RNA segments that encode six structural proteins and six nonstructural proteins. Rotavirus replication is fully cytoplasmic and occurs within highly specialized regions called viroplasms. NSP2 and NSP5 have been shown to be essential for viroplasm formation and, when co-expressed in uninfected cells, to form viroplasm¬like structures. A recent study suggest a key role for NSP5 in architectural assembly of viroplasms and in recruitment of viroplasmic proteins, containing four structural (VP1, VP2, VP3 and VP6) and two nonstructural (NSP2 and NSP5) proteins. NSP5, the translation product of gene segment 11 has a predicted molecular eight of 21 kDa. NSP5 has been reported to exist in multiple isoforms ranging in size from 28-and 32-35 kDa from a 26-kDa precursor has been attributed to O-glycosylation and hyperphosphorylation. To study different properties of the protein, recombinant NSP5 containing an N-terminal hisidine tag was expressed in bacteria and purified by affinity chromatography. A significant observation was the similarity in phosphorylation property of the bacterially expressed and that expressed in mammalian cells. While the untagged recombinant protein failed to undergo phosphorylation in vitro, addition of His tag or deletions at the N-terminus promoted phosphorylation of the protein in vitro, which is very similar to the reported properties exhibited by the corresponding proteins expressed in mammalian cells. Phosphorylation of NSP5 in vitro is independent of the cell type from which the extract is derived suggesting that the kinases that phosphorylate NSP5 are distributed in all cell types. Among the C-terminal deletion mutants studied, NH-∆C5 and NH-∆C10 were phosphorylated better than full-length NSP5, but NH-∆C25 and NH¬∆C35 showed substantial reduction in the level of phosphorylation compared to full-length NSP5. These results indicate that the C-terminal 30 residues spanning the predicted α-helical domain of NSP5 are critical for its phosphorylation in vitro which is in correspondence with previous findings that C-terminal 21 amino acids of NSP5 direct its insolubility, hyperphosphorylation, and VLS formation. The results revealed that though the tagged full-length and some of the mutants could be phosphorylated in vitro, they are not suitable substrates for hyperphosphorylation unlike the similar proteins expressed in mammalian cells or infected cells. Analysis by western blot and mass spectrometry revealed that the bacterially expressed NH-NSP5 is indeed phosphorylated. It appears that prior phosphorylation in bacteria renders the protein conformationally not amendable for hyperphosphorylation by cellular kinases in vitro. Mutation of the highly conserved proline marginally enhanced its phosphorylation in vitro but the stability of protein is affected. Notably, mutation of S67A, identified as a critical residue for the putative caesin kinase-I and-II pathways of NSP5 phosphorylation, affected neither the phosphorylation nor the ATPase activity of NSP5. These results suggest that bacterially expressed NSP5 by itself has undectable auto-kinase activity and it is hypophosphorylated. Purified recombinant NSP5 has been reported to possess an Mg¬ 2+-dependent ATP-specific triphosphatase activity. The results indicated that deletion of either C-terminal 48 amino acids or N-terminal 33 residues severely affected the ATPase activity of recombinant NSP5, underlying the importance of both N-and C-terminal domains for NSP5 ATP hydrolysis function.
NSP5 expressed in rotavirus infected cells exists as inter-molecular disulfide-linked dimeric forms and it appears that the 46 kDa isoforms, that are phosphorylated, corresponds to dimer as revealed by western blotting. Analytical gel filtration analysis of NH-NSP5, NH-ΔN43 and NH-ΔN33-ΔC25 showed most of the proteins in void volume, but an additional peak corresponding to the mass of dimeric species further supports that NSP5 is basically a dimer that undergoes oligomerization. Analysis by sucrose-gradient fractionation revealed that NH-NSP5 and NH-ΔN43 proteins were mainly distributed in the lower fraction of the gradient suggesting the existence of high molecular weight complexes or higher oligomers. The multimeric nature of NSP5 and its mutants was further confirmed by dynamic light scattering which suggests that high molecular weight complexes are of homogenous species. The correlation curves showed a low polydispersity distribution and a globular nature of recombinant NH-NSP5 proteins. The present results clearly demonstrate that dimer is the basic structural unit of NSP5 which undergoes oligomerization to form a complex consisting of about 20-21 dimers.
The nonstructural protein 5 is hyperphosphorylated in infected cells and cellular kinases have been implicated to be involved in its phosphorylation. NSP5 contains multiple consensus sites for phosphorylation by several kinases, but the cellular kinases that specifically phosphorylate NSP5 in infected cells are yet to be identified. Previous studies from our laboratory using signaling pathway inhibitors revealed that recombinant NH¬NSP5 and its deletion mutants can be phosphorylated in vitro by purified cellular kinases and by mammalian cell extracts. These studies also showed the involvement of PI3K-Akt and MAPK signaling pathways in NSP5 phosphorylation and a negative role for GSK3β in the phosphorylation of bacterially expressed recombinant NSP5 in vitro. In the present work, using phospho-specific anti-Ser9 GSK3β antibody, we observed that GSK3β is inactivated in a rotavirus infected MA104 cells in a strain-independent manner. GSK3β¬specific small interfering RNA (siRNA-GSK3β) reduced GSK3β levels leading to increased level of synthesis of the structural rotavirus protein VP6 and NSP5 hyperphosphorylation compared to control siRNA. The pharmacological kinase inhibitors (LY294002, Genistein, PD98059, and Rapamycin) studies at the concentrations tested did not significantly affect rotavirus infection as seen from the number foci, while U0126 severely affected rotavirus replication. The results clearly demonstrated the importance of the MEK1/2 signaling pathway in the successful replication of rotavirus and NSP5 hyperphosphorylation in rotavirus-infected cells. In contrast inhibition of GSK3β activity by LiCl, increased in general, the number of foci by greater than 2-fold for all viral strains studied. These results suggest that MEK1/2 pathway majorly contributes to GSK3β inactivation in rotavirus infected cells. Thus, our results reveal that rotavirus activates both the PI3K/Akt and FAK/ERK1/2 MAPK pathways and appears to utilize them as a strategy to activate mTOR, and inhibit GSK3β through phosphorylation on serine 9, the negative regulator of rotavirus NSP5 phosphorylation, and thus facilitate translational competence of rotaviral mRNAs during virus replication cycle.
It was shown previously in the laboratory by co-immunoprecipitation assay that Hsp70 interacts with rotaviral proteins VP1 and VP4 in rotavirus-infected mammalian cells. In this study, the interactions between Hsp70 with VP1 and VP4 were further evaluated in vitro by GST-pull down assay. It was observed that the N-terminal ATPase and C-terminal peptide-binding domain of Hsp70 is necessary for its direct interaction with VP1 and VP4. The presence of Hsp70 in purified double-and triple-layered virus particles further supported the observed interactions of rotaviral proteins VP1 and VP4 with Hsp70. However, the specific interaction observed between Hsp70 and rotaviral capsid proteins, VP1 and VP4 in viral particles suggests that Hsp70 has an important role during rotavirus assembly which requires further investigation. | en_US |