Possible contributor to the virulence of the 1918 flu pandemic discovered

This discovery identifies a new interaction that has the potential to be exploited as a target for antiviral treatment of virulent strains of influenza.

Petit's in vitro experiments show that this particular 1918 flu protein directly binds to the human protein RIG-I -- a key trigger for the human immune response against flu. Further research will test whether that binding has biological importance to help the flu virus disable RIG-I and augment suppression of the immune response.

The 1918 protein that binds to RIG-I is a portion of the flu virus's nonstructural protein 1, or NS1. NS1 is rapidly produced as the flu virus invades a host cell, allowing influenza to circumvent the host immune response to viral infection. Other research has shown that NS1 affects multiple host proteins, including RIG-I, to help the flu virus grow and spread.

Petit and his UAB colleagues are the first to show that NS1 has a direct interaction with RIG-I, the cell's main sensor to detect flu virus infection. Furthermore, the portion of the 1918 NS1 RNA binding domain that binds to RIG-I had no previously known function.

"NS1 is almost like the Swiss army knife of proteins because it has so many functions," Petit said. He says NS1 appears to interact with 20 to 30 host proteins. Compared with other flu proteins, NS1 also has remarkable genetic plasticity, meaning that its effect on virulence can vary among strains, Petit says.

In contrast to the 1918 NS1, Petit's lab has found that the NS1 from the influenza A strain 1972 Udorn is unable to bind to RIG-I. After determining the solution structure of the portion of the 1918 NS1 that does bind RIG-I, the UAB group was able to precisely identify the RIG-I binding site for 1918 NS1.

The team also showed how the structure of the 1918 NS1 differs from the already known structure of the Udorn NS1, and they showed which amino acid alterations in the 1918 NS1 protein appear to account for the different binding abilities and shapes of the two NS1 variants. Thus the binding of NS1 to RIG-I appears to be strain-dependent.

"One of NS1's main functions is to combat the immune system, and without NS1, the host immune response quickly takes out the virus," Petit said. "But I find that there's not a lot of data that compares NS1 from strain to strain in the same kind of test."

This research article, "Structural basis for a novel interaction between the NS1 protein derived from the 1918 influenza virus and RIG-I," was recently published online ahead of print in the journal Structure, with Petit as corresponding author. Other authors are Alexander Jureka and Alex Kleinpeter, UAB Department of Biochemistry and Molecular Biology, and Gabriel Cornilescu, Ph.D., and Claudia Cornilescu, Ph.D., National Magnetic Resonance Facility, University of Wisconsin-Madison.

Some details of the research

Solution structure and binding studies were done using NMR spectroscopy in UAB's Central Alabama High-Field Nuclear Magnetic Resonance Facility, Department of Chemistry, UAB College of Arts and Sciences. To facilitate the NMR analysis, relevant domains of NS1 and RIG-I proteins were cloned. For the 1918 and Udorn NS1 proteins, these were the independently folding, RNA-binding domains (NS1RBD), which are both homodimers with a six-helical RNA-binding fold. For the RIG-I protein, this was the CARD2 domain, one of two caspase activation and recruitment domains on RIG-I.

The novel RIG-I binding site found on the 1918 NS1RBD is opposite to that of the RNA binding interface, and it had no previously known function. Two potential salt bridges that are present in the 1918 NS1RBD but absent in the Udorn NS1RBD appear to alter the 1918 NS1RBD structure by significantly increasing the average intramolecular distance between two of the protein helices, as compared with the Udorn NS1RBD. This is a potential structural explanation for the strain-specific nature of the interaction of 1918 NS1RBD with RIG-I.

The binding affinity between the 1918 NS1RBD and the RIG-ICARD2 was estimated by non-denaturing gel electrophoresis, and it is consistent with other interactions between viral and cellular proteins that are known to be biologically relevant. A mutant of 1918 NS1RBD that changed one of the amino acids involved in the putative salt bridge severely inhibited the interaction between the 1918 NS1RBD and RIG-ICARD2.