Study of the host specificity of PB1-F2-associated virulence

Abstract

Influenza A viruses cause important diseases in both human and animal. The PB1-F2 protein is a virulence factor expressed by some influenza viruses. Its deleterious action for the infected host is mostly described in mammals, while the available information is scarce in avian hosts. In this work, we compared the effects of PB1-F2 in avian and mammalian hosts by taking advantage of the zoonotic capabilities of an avian H7N1 virus. In vitro, the H7N1 virus did not behave differently when PB1-F2 was deficient while a H3N2 virus devoid of PB1-F2 was clearly less inflammatory. Likewise, when performing in vivo challenges of either chickens or embryonated eggs, with the wild-type or the PB1-F2 deficient virus, no difference could be observed in terms of mortality, host response or tropism. PB1-F2 therefore does not appear to play a major role as a virulence factor in the avian host. However, when infecting NF-κB-luciferase reporter mice with the H7N1 viruses, a massive PB1-F2-dependent inflammation was quantified, highlighting the host specificity of PB1-F2 virulence. Surprisingly, a chimeric 7:1 H3N2 virus harboring an H7N1-origin segment 2 (i.e. expressing the avian PB1-F2) induced a milder inflammatory response than its PB1-F2-deficient counterpart. This result shows that the pro-inflammatory activity of PB1-F2 is governed by complex mechanisms involving components from both the virus and its infected host. Thus, a mere exchange of segment 2 between strains is not sufficient to transmit the deleterious character of PB1-F2.

Comparison of PB1-F2 Proximity Interactomes Reveals Functional Differences between a Human and an Avian Influenza Virus

“…Among the PB1-F2 interactors, after CRAPome filtering, only 10 of them interact with both proteins ( Figure 4 A). This wide diversity of interactors between the two proteins is consistent with their opposite functions observed in the mouse model [ 26 ]. Using Gene Ontology (GO), we analyzed the functions associated with these 10 proteins.…”

“…In mouse infections, the avian PB1-F2 was much more inflammatory than the human PB1-F2. However, a chimeric H3N2 virus expressing H7N1 PB1-F2 had reduced pro-inflammatory activity compared to wild-type H3N2 virus [ 26 ]. These observed differences could be explained by more or less pronounced fiber-forming abilities.…”

“…In this paper, we compared the interactomes of two PB1-F2 endowed with opposite behavior in vivo. As a matter of fact, in the mouse model and using a chimeric 7:1 H3N2 virus harboring an H7N1-origin segment 2 and therefore expressing the H7N1 PB1-F2, we have shown that the chimeric virus induced a lower inflammatory response compared to the wild-type H3N2 [ 26 ]. The fibrillation capacities of these two proteins are very comparable and are therefore probably not the cause of the differences observed in the infected mice.…”

“…We recently investigated the contribution of the accessory protein PB1-F2 to species specificity. For this purpose, we produced human chimeric viruses in which segment #2, encoding PB1-F2, was of avian origin [ 26 ]. Unexpectedly, by knocking-out this protein within the chimeric virus, we made a virus more inflammatory than the virus able to express PB1-F2.…”

Antiviral activity of intracellular nanobodies targeting the influenza virus RNA-polymerase core

“…Direct repeats were removed from the reporter gene by introducing silent mutations at the 3’ end of the PA ORF. Reverse genetic systems for avian influenza A/Turkey/Italy/977/1999 [H7N1] and human influenza A/Scotland/20/1974 [H3N2] viruses have been previously elaborated and used to generate the recombinant reporter viruses H7N1-Nluc and H3N2-Nluc [ 37 ], [ 38 ]. Both these Nluc reporter viruses were designed using the same strategy than H1N1-Nluc, with their PA segments encoding a PA-2A-Nluc polyprotein, and were kindly provided by Ronan Le Goffic.…”