Research Article: Species-Specific Inhibition of RIG-I Ubiquitination and IFN Induction by the Influenza A Virus NS1 Protein

Date Published: November 29, 2012

Publisher: Public Library of Science

Author(s): Ricardo Rajsbaum, Randy A. Albrecht, May K. Wang, Natalya P. Maharaj, Gijs A. Versteeg, Estanislao Nistal-Villán, Adolfo García-Sastre, Michaela U. Gack, Andrew Pekosz.


Influenza A viruses can adapt to new host species, leading to the emergence of novel pathogenic strains. There is evidence that highly pathogenic viruses encode for non-structural 1 (NS1) proteins that are more efficient in suppressing the host immune response. The NS1 protein inhibits type-I interferon (IFN) production partly by blocking the TRIM25 ubiquitin E3 ligase-mediated Lys63-linked ubiquitination of the viral RNA sensor RIG-I, required for its optimal downstream signaling. In order to understand possible mechanisms of viral adaptation and host tropism, we examined the ability of NS1 encoded by human (Cal04), avian (HK156), swine (SwTx98) and mouse-adapted (PR8) influenza viruses to interact with TRIM25 orthologues from mammalian and avian species. Using co-immunoprecipitation assays we show that human TRIM25 binds to all tested NS1 proteins, whereas the chicken TRIM25 ortholog binds preferentially to the NS1 from the avian virus. Strikingly, none of the NS1 proteins were able to bind mouse TRIM25. Since NS1 can inhibit IFN production in mouse, we tested the impact of TRIM25 and NS1 on RIG-I ubiquitination in mouse cells. While NS1 efficiently suppressed human TRIM25-dependent ubiquitination of RIG-I 2CARD, NS1 inhibited the ubiquitination of full-length mouse RIG-I in a mouse TRIM25-independent manner. Therefore, we tested if the ubiquitin E3 ligase Riplet, which has also been shown to ubiquitinate RIG-I, interacts with NS1. We found that NS1 binds mouse Riplet and inhibits its activity to induce IFN-β in murine cells. Furthermore, NS1 proteins of human but not swine or avian viruses were able to interact with human Riplet, thereby suppressing RIG-I ubiquitination. In conclusion, our results indicate that influenza NS1 protein targets TRIM25 and Riplet ubiquitin E3 ligases in a species-specific manner for the inhibition of RIG-I ubiquitination and antiviral IFN production.

Partial Text

Influenza A viruses (IAVs) are highly infectious pathogens that have caused major pandemics and annual epidemics with serious economic and health consequences [1], [2]. IAVs are naturally maintained in avian species but they also circulate in humans, horses, dogs and pigs [3]. Although multigenic host range restrictions exist, a combination of viral determinants can ultimately allow a virus to establish infection in a specific host [4]. This is particularly important because, although the current highly pathogenic avian IAVs that have been transmitted to humans lack the ability to spread from human to human, there is current concern that these avian viruses may adapt and develop the ability to spread efficiently among humans. In this respect, recent studies have demonstrated that only a few mutations in the hemagglutinin (HA) allow for transmissibility of highly pathogenic H5N1 viruses in ferrets [5], [6]. Moreover, pigs can be infected with human and avian viruses and provide an environment for reassortment and the generation of new influenza virus strains capable of human transmission [7]. Therefore, it is essential to better understand the mechanisms that allow influenza viruses to adapt to a new host species, in order to predict and protect from future cross-species transmission.

In this study we investigated the mechanisms by which the NS1 protein of influenza A viruses affects the TRIM25/RIG-I-mediated type-I IFN response in different host species (Figure 9). We demonstrated that NS1 proteins from avian and mammalian isolates are capable of binding to human, but not mouse TRIM25. In mouse cells, NS1 protein inhibits RIG-I ubiquitination and downstream IFN promoter activation primarily by a mechanism that involves binding to and inhibiting the ubiquitin E3 ligase Riplet. This mechanism thus clearly differs from the previously described mechanism for NS1 antagonism of TRIM25 in human cells. As mice are not natural hosts of influenza virus infection, we reasoned that the ability of influenza virus NS1 to inhibit mouse Riplet most likely reflects its inherent ability to inhibit Riplet from other animal species. Indeed, our study showed that the NS1 protein of human viruses efficiently bound and suppressed human Riplet, indicating that specifically human viruses have evolved to inhibit both TRIM25 and Riplet for the suppression of RIG-I antiviral activity in human cells. Interestingly, Riplet is highly similar to TRIM25 sharing 60.8% sequence homology. Like TRIM25, Riplet contains an N-terminal RING domain and a C-terminal SPRY domain. However, in contrast to TRIM25, Riplet does not contain a B-box domain and thus it does not belong to the TRIM protein family [26]. Importantly, bioinformatic analysis also predicted the presence of a central CCD in Riplet, with structural similarities to the TRIM25 CCD (data not shown). Further studies are needed to identify the precise binding site of NS1 in Riplet, and whether this interaction requires the Riplet CCD.