PhD Graduate: Keng Choong Tat (Collaborative Anti-Viral Research Laboratory)
Thesis Title: Characterization of viral-viral and viral-host interactions that contribute to the replication and pathogenesis of the severe acute respiratory syndrome coronavirus (SARS-CoV).
In the first part of the study, we identified neutralizing epitopes in the region of the severe acute respiratory coronavirus (SARS-CoV) spike (S) protein encoded by amino acid residues 1055 to 1192 (SΔ10 protein fragment). To understand the mechanism of SARS-CoV neutralization by the anti-SΔ10 antibodies, a panel of monoclonal antibodies (mAbs) was generated. Epitope mapping revealed that the immuno-dominant region within SΔ10 lies between residues 1091 to 1130 and 1150 to 1192, which are part of the loop region between the heptad repeats 1 (HR1) and heptad repeats 2 (HR2) domains and the N-terminal of the HR2 domain, respectively. The HR1 and HR2 domains have previously been shown to be important for the viral-cell fusion process. Using the cell-cell fusion inhibition and in vitro neutralizing assays, the mAbs were shown to neutralize SARS-CoV infection and inhibit cell-cell fusion. Taken together, the results suggest that the mechanism of neutralization of SARS-CoV infection by these mAbs is due to the inhibition of the viral-cell fusion step during viral entry. As these mAbs target highly conserved regions in the C terminal of S, which is known as the S2 domain, two representative mAbs, 1A9 and 1G10, were selected for evaluation of cross-neutralizing properties. A recent report suggests that the bat SL-CoV Rf1 and Rp3 strains could have undergone a recombination event to yield the civet SARS-CoV SZ3 strain which in turn spills over to human during the SARS epidemic. By use of lentiviral pseudotyped particles carrying S on the surface, both mAbs targeting the N-terminal HR2 domain (mAb 1G10) and the loop region between the HR1 and HR2 domains were shown to cross-neutralize the civet SARS-CoV (SZ3 strain) and bat SL-CoV (Rf1 and Rp3 strains). In order to gain a better understanding of the mechanism of virus neutralization of mAbs 1G10 and 1A9, escape viruses were generated. From the results, two mutations (N1056K and D1128A) in S of the escape viruses generated using mAb 1A9 showed a synergistic effect on the escape activity in pseudovirus assay. Taken together, the results suggest that a novel conserved region in the loop domain between the HR1 and HR2 domains is important for the SARS-CoV and SL-CoV viral-cell fusion process.
In the second part of the study, we investigated the most obvious genomic difference between the human epidemic SARS-CoV strains and the animal and early human SARS-CoV strains, which involves the acquisition of a 29-nt deletion in the ORF8, resulting in the coding of two proteins, 8a and 8b, instead of a single 8ab protein. Using over-expression systems and co-immunoprecipitation experiments, the viral-viral interaction profiles of all the three proteins (8a, 8b and 8ab) were found to be different, and more interestingly, 8b was found to down-regulate E in a post-translational manner. Infection studies further showed that 8b is expressed late during infection and has mutually exclusive localization with E in infected cells. Cells stably expressing the GFP-tagged 8b were also shown to down-regulate E and reduce virus replication by approximately 10-fold. In addition, it was demonstrated that the mechanism of down-regulation of E by 8b is via a ubiquitin-independent proteasome pathway. An oligonucleotide microarray analysis also revealed that 8b alters the mRNA levels of several host proteins, suggesting that they may be involved in virus replication. In summary, we have identified an important viral-viral interaction that could have evolved in the human epidemic strains.
Figure Legend: Cross-neutralization activity of mAb 1A9 and 1G10. (A) Pseudovirus packaged with S(HK39849), S(SZ3), S(Rp3) and S(Rf1) were tested for their efficiency of entry in the absence of mAb. (B, C) Pseudovirus packaged with S(HK39849), S(SZ3), S(Rp3) and S(Rf1) were measured for the efficiency of entry into CHO-ACE2 cells in the presence of different mAb 1A9 and 1G10 concentrations, respectively, using a luciferase-based assay.