Sequence-Based Classification of Select Agents: A Brighter Line (2010)
Chapter: Appendix G: Influenza A and SARS-CoV
Appendix G
Influenza A and SARS-CoV
INFLUENZA A
The influenza A virus proapoptotic PB1-F2 protein has been clearly implicated as a major virulence factor in some highly pathogenic influenza virus strains, but the H1N1 swine-origin influenza virus pandemic strain codes for a truncated PB1-F2 protein that terminates after 11 amino acids. It was predicted that the truncation would attenuate swine influenza pathogenesis, identifying a possible key mutation that could emerge to enhance H1N1 virulence and contribute to an expanding epidemic. However, disruption of PB1-F2 expression in several other influenza virus backgrounds or by intermixing functional PB1-F2 between strains had little effect on viral lung load in mice. The data suggests that the PB1-F2 virulence determinant may be context- or host-dependent, perhaps by enhancing virulence by other mechanisms that are independent of replication. The effects of restoring a full length functional PB1-F2 protein on 2009 swine H1N1 in vivo pathogenesis are difficult to predict because its virulence-enhancing activities may depend on co-evolutionary changes elsewhere in the genome. As a working model for predicting virulence from sequence information, the preponderance of influenza data suggest that restoration of a full length PB1-F2 protein will enhance the virulence of swine H1N1—a hypothesis that will probably be tested using reverse genetics in the near future (McAuley, Zhang et al.)
SARS-COV
It is also clear that distantly related viral proteins can interact with a conserved cellular protein target and thereby augment their pathogenic potential. Among coronaviruses as with many other viruses, receptor interactions are
an important determinant of species specificity, tissue tropism, virulence, and pathogenesis. Pathogenesis depends upon the ability of a virus to dock and enter into a suitable human host cell. For example, the highly pathogenic emerging group 2 coronavirus that causes severe acute respiratory syndrome, coronavirus (SARS-CoV) and a distantly related less pathogenic group 1 human coronavirus, NL63-CoV, both encode a large 180/90kDa spike glycoprotein (S) that engages a host cellular receptor(s) to mediate docking and entry into cells. The SARS-CoV and NL63-CoV S glycoproteins are about 40 percent identical and encode novel, yet unrelated receptor binding domains (RBD) in S that engage the same cellular receptor, angiotensin-converting enzyme 2 (ACE2) to mediate virus docking and entry into cells. Despite the absence of structural homology in the RBD cores of NL63-CoV and SARS-CoV, the two viruses recognize common ACE2 regions by using novel protein-protein folds and interaction networks. On the basis of sequence, it was not possible to predict that the two highly divergent coronavirus RBDs would engage a similar “hot spot” on the surface of the ACE2 receptor and thus mediate docking and entry into cells. Moreover, the pathogenic potential of the two human coronaviruses are distinct: SARS-CoV causes an atypical pneumonia that results in acute respiratory distress syndrome with mortality exceeding 50 percent in people over 60 years old, whereas NL63-CoV causes a self-limiting denuding bronchiolitis and croup, primarily in infants and children. Clearly, other factors besides virus-receptor interaction and entry are regulating severe acute end-stage lung-disease outcomes during SARS-CoV infection, and this complicates sequence-based predictions of virus-receptor interaction networks and virulence outcomes (Proc Natl Acad Sci U S A. 2009 Nov 24;106(47):19970-19974. Epub 2009 Nov 9. Crystal structure of NL63 respiratory coronavirus receptor-binding domain complexed with its human receptor (Wu, Li et al. 2009).
