Evasion of antiviral innate immunity by Theiler’s virus L* protein

(eng) Theiler’s virus is a natural pathogen of mice. Persistent strains of this virus are responsible for a chronic infection of the spinal cord characterized by the presence of inflammation foci as well as viral and immune-mediated demyelination. These strains of TMEV encode an accessory protein translated from an alternative open reading frame overlapping the coding region of the viral polyprotein. This protein, called L*, is crucial for viral persistence in susceptible mice. Little is known about the mechanisms that allow L* protein to stimulate the persistence of TMEV in the central nervous system. Previous work indicated that the expression of L* protein increased viral replication in macrophage cell lines, likely by specifically inhibiting apoptosis in these cells. It was also suggested that L* is associated with microtubules in infected cells. However, this intracellular localization was in contradiction with preliminary observations made in our lab. Therefore, the first part of this work was to investigate the subcellular localization of the L* protein as the knowledge of its intracellular localization might give some hint on its function during infection. We observed that L* protein is expressed as both a cytoplasmic and mitochondrial protein during infection. Biochemical analysis showed that L* is anchored in the mitochondrial outer membrane and facing the cytosol. This mitochondrial targeting is independent of viral replication and involves a non-canonical internal targeting signal. In the course of these experiments, we observed that RNA extracted from macrophages infected with L* mutant viruses was strongly degraded as compared to RNA extracted from wild-type or mock-infected macrophages. This led us to hypothesize that L* protein could inhibit a cellular RNase, which is activated during viral infection. Infection of RNase L knock-out peritoneal macrophages demonstrated that RNA degradation was RNase L-dependent and completely inhibited by L* protein. This observation prompted us to investigate the mechanism of RNase L inhibition. Transfection of poly(I:C) or oligoadenylates in L*-expressing cells demonstrated that the inhibition of RNase L involved an effect of L* protein on either the oligoadenylates or the RNase L itself. As we observed that L* translation was required for RNase L inhibition, we cloned the murine RNase L and performed immunoprecipitation experiments that allowed us to detect an interaction between L* and RNase L in the course of infection. This interaction is species-specific as L* is not able to inhibit RNase L originating from other species, including humans. Finally, using a human/murine hybrid system the interaction between L* and the murine RNase L was shown to be direct.