A single-amino-acid polymorphism in reovirus protein μ2 determines repression of interferon signaling and modulates myocarditis

Abstract

Myocarditis is indicated as the second leading cause of sudden death in young adults. Reovirus induces myocarditis in neonatal mice, providing a tractable model system for investigation of this important disease. Alpha/beta-interferon (IFN-α/β) treatment improves cardiac function and inhibits viral replication in patients with chronic myocarditis, and the host IFN-α/β response is a determinant of reovirus strain-specific differences in induction of myocarditis. Virus-induced IFN-β stimulates a signaling cascade that establishes an antiviral state and further induces IFN-α/β through an amplification loop. Reovirus strain-specific differences in induction of and sensitivity to IFN-α/β are associated with the viral M1, L2, and S2 genes. The reovirus M1 gene-encoded μ2 protein is a strain-specific repressor of IFN-β signaling, providing one possible mechanism for the variation in resistance to IFN and induction of myocarditis between different reovirus strains. We report here that μ2 amino acid 208 determines repression of IFN-β signaling and modulates reovirus induction of IFN-β in cardiac myocytes. Moreover, μ2 amino acid 208 determines reovirus replication, both in initially infected cardiac myocytes and after viral spread, by regulating the IFN-β response. Amino acid 208 of μ2 also influences the cytopathic effect in cardiac myocytes after spread. Finally, μ2 amino acid 208 modulates myocarditis in neonatal mice. Thus, repression of IFN-β signaling mediated by reovirus μ2 amino acid 208 is a determinant of the IFN-β response, viral replication and damage in cardiac myocytes, and myocarditis. These results demonstrate that a single amino acid difference between viruses can dictate virus strain-specific differences in suppression of the host IFN-β response and, consequently, damage to the heart.

Fig 1

μ2 aa 208 determines reovirus repression of IFN signaling. (A) Viruses with a chimeric M1 gene or viruses with a mutated amino acid at position 208 of μ2 were generated by reverse genetics (21, 38). (B) L929 cells were infected with the indicated viruses, cells were treated with IFN, and mRNA was harvested for qRT-PCR. The fold repression was calculated as ([copies IRF7/GAPDH in IFN-treated mock-infected cells]/[copies IRF7/GAPDH in IFN-treated virus-infected cells]). Viruses that induce high levels of IFN resulted in values <1, but the relative repression between viruses was the same (56). The results are means ± the standard deviation (SD) (average of replicate wells; representative of four experiments). *, Different from T3D (P < 0.05); ‡, different from T1L (P < 0.05).

Fig 2

μ2 aa 208 modulates reovirus induction of IFN in cardiac cells. Primary cardiac myocyte (A and B) or fibroblast (C and D) cultures were infected with the indicated viruses, and mRNA was harvested for qRT-PCR using primers specific for IFN-β (A and C) or IRF7 (B and D) and GAPDH. The fold induction was calculated as copies IFN-β or IRF7/GAPDH in infected cultures relative to uninfected cultures. (A and C) Results are means ± the standard errors of the mean (SEM) (average of two experiments); (B and D) results are means ± the SD (average of replicate wells; representative of two experiments). See the statistical analyses in the text.

Fig 3

μ2 aa 208 determines reovirus single-cycle and multicycle replication in cardiac myocytes. (A) Primary cardiac myocyte cultures were infected with the indicated viruses at an MOI of 3 PFU per cell and harvested for plaque assay after 24 h or infected at 10 PFU per cell and harvested for immunoblotting after 24 h. The leftmost lane contains mock-infected sample. (B) Primary cardiac myocyte cultures were infected at an MOI of 0.1 PFU per cell and harvested for plaque assay after 5 days. (C) Primary cardiac myocyte cultures were infected as in panel B, treated with anti-IFN-β antibody where indicated, and harvested for plaque assay. All results are means ± the SD (average of replicate wells; representative of two to four experiments). See the statistical analyses in the text.

Fig 4

μ2 aa 208 influences reovirus single-cycle and multicycle replication in cardiac fibroblasts. Primary cardiac fibroblast cultures were infected with the indicated viruses at an MOI of 3 PFU per cell and incubated for 24 h (A) or at an MOI of 0.1 PFU per cell and incubated for 5 days (B) and harvested for plaque assay. (C) Primary cardiac fibroblast cultures were infected as in panel B, treated with anti-IFN-β where indicated, and harvested for plaque assay. All results are means ± the SD (average of replicate wells; representative of two to four experiments). See the statistical analyses in the text.

Fig 5

μ2 aa 208 determines reovirus CPE in cardiac myocytes only after spread. (A) Primary cardiac myocyte cultures were infected with the indicated viruses at an MOI of 10 PFU per cell and incubated for 2 days (single cycle) or at an MOI 0.1 PFU per cell and incubated for 5 or 7 days (multicycle) and harvested for MTT assay. The percent viable cells was calculated relative to mock-infected cells. The results at 2 and 5 days are means ± the SEM (average of at least two experiments); results at 7 days are means ± the SD (average of replicate wells for a single experiment). See the statistical analyses in the text. (B) Primary cardiac myocyte cultures were infected at an MOI of 0.1 PFU per cell, incubated for 7 days, and photographed at a final magnification of ×100 (representative fields shown).

Fig 6

μ2 aa 208 modulates myocarditis. Neonatal mice were inoculated with the indicated viruses and euthanized 7 days postinjection. Hematoxylin-eosin-stained cardiac sections were scored for lesions. (A) Representative sections. (B) A minimum of eight hearts (minimum of 15 sections per heart) were scored for each virus. Lesions in adjacent sections were not scored as independent lesions. Hearts were considered “positive” if more than a single lesion was detected in the ≥ 15 sections examined. *, Different from T3D; ‡, different from T1L (chi square, P < 0.05). The results are representative of two independent experiments.