Nonstructural Nipah virus C protein regulates both the early host proinflammatory response and viral virulence

Abstract

Nipah virus (NiV) is a highly pathogenic, negative-strand RNA paramyxovirus that has recently emerged from flying foxes to cause serious human disease. We have analyzed the role of the nonstructural NiV C protein in viral immunopathogenesis using recombinant virus lacking the expression of NiV C (NiVΔC). While wild-type NiV was highly pathogenic in the hamster animal model, NiVΔC was strongly attenuated. Replication of NiVΔC was followed by the production of NiV-specific antibodies and associated with higher recruitment of inflammatory cells and less intensive histopathological lesions in different organs than in wild-type-NiV-infected animals. To analyze the molecular basis of NiVΔC attenuation, we studied early changes in gene expression in infected primary human endothelial cells, a major cellular target of NiV infection. The transcriptomic approach revealed the striking difference between wild-type and mutant NiV in the expression of genes involved in immunity, with the particularly interesting differential patterns of proinflammatory cytokines. Compared to wild-type virus, NiVΔC induced increased expression of interleukin 1 beta (IL-1β), IL-8, CXCL2, CXCL3, CXCL6, CCL20, and beta interferon. Furthermore, the expression of NiV C in stably transfected cells decreased the production of the same panel of cytokines, revealing a role of the C protein in the regulation of cytokine balance. Together, these results suggest that NiV C regulates expression of proinflammatory cytokines, therefore providing a signal responsible for the coordination of leukocyte recruitment and the chemokine-induced immune response and controlling the lethal outcome of the infection.

Fig 1

Generation of NiVΔC recombinant viruses. (A) Schematic illustration of mutation introduced into the viral genome in order to eliminate expression of the C protein. Nucleotide changes designed to knock out the expression of the C ORF are put in squares, and introduced stop codons are indicated. (B) Kinetic analysis of the production of infectious particles in the supernatant fluids of HUVEC infected with NiVwt, NiVΔC1, or NiVΔC2 (MOI = 0.01). (C) Vero E6 cells infected with NiVwt, NiVΔC1, or NiVΔC2 were harvested at 48 h p.i. and analyzed by Western blotting, using antibodies specific for the NiV proteins G, M, and C and actin.

Fig 2

Survival curves of hamsters infected with different doses of recombinant NiV. Hamsters (5 or 6 per group) were inoculated i.p. with indicated doses of NiVwt (A), NiVΔC1 (B), or NiVΔC2 (C). Animals were followed for signs of infection during 3 weeks. The results are expressed as percentages of surviving animals in each group. Differences in the rate of survival are statistically significant between NiVwt and NiVΔC for all groups injected with the same dose of virus (P < 0.01, chi-square test), except for the group injected with 1,000 PFU of NiVΔC1, which was not statistically different from the NiVwt group. Differences in virulence between NiVΔC1 and NiVΔC2 were not statistically significant.

Fig 3

Immunopathogenesis of NiV replication in hamsters. (A) Kinetics of NiV replication in the organs of infected hamsters determined by RT-qPCR. Hamsters were infected i.p. with either 10⁵ PFU of NiVwt (white histograms) or NiVΔC (black histograms), and every day p.i., 1 to 3 animals were euthanized and organs were analyzed by RT-qPCR, as described in Materials and Methods. The average NiV N RNA expression in indicated organs is shown, and bars present standard deviations. (B to D) Generation of inflammatory reaction in infected animals. Lung tissues from hamsters were infected with either mock (B), 10⁵ PFU of NiVwt (C), or 10⁵ PFU of NiVΔC and taken 5 days p.i. (D). Occasional nodular inflammation, edema, and necrosis (arrow) are observed (C), and more intense inflammation and cell infiltration with no edema or necrosis are seen (D). (E) Production of NiV-specific antibodies in the serum of hamsters infected with different doses of NiVΔC or UV-irradiated NiVΔC. Twenty-nine days after infection, serum was taken and analyzed by ELISA for the presence of virus-specific antibodies. Sera taken from animals before infection were always negative by ELISA. Histograms correspond to the average titer obtained from 3 to 6 animals, and bars present standard deviations.

Fig 4

Analysis of cytokine expression in NiVwt- and NiVΔC-infected HUVEC. HUVEC were infected with either NiVwt or NiVΔC (MOI = 1) or mock infected; RNA was isolated 8 h (A) or 24 h (B) later and analyzed for cytokine expression by RT-qPCR or for the IFN-β mRNA expression 24 h and 48 h p.i. (C). (D) Induction of IL-8 production after NiVwt and NiVΔC infection of HUVEC in cell supernatants taken 8, 24, and 48 H after infection and analyzed by using an IL-8 ELISA kit. Results are expressed as a fold change in induction by virus infection, compared to results for mock infection, and correspond to the average for 6 individual experiments, and bars represent standard deviations. Statistically significant differences either between infected and noninfected cells or between NiVwt and NiVΔC infection are indicated (*, P < 0.05, * *, P < 0.01, Mann-Whitney U test).

Fig 5

NiV C inhibits production of proinflammatory cytokines. (A and B) 293 cells (A) or 293C cells (B), stably expressing NiV C, were stained with anti-NiV C antibodies (green) and DAPI (blue) and observed by immunofluorescence microscopy (×100). The insert in each figure corresponds to higher magnification (×200). (C and D) Both 293 cells (black histograms) and 293C cells (white histograms) were infected with either NiVwt or NiVΔC (C) or stimulated with poly(I·C) (D) and analyzed 24 h later for the production of CXCL2, CXCL3, IL-8, and IFN-β by RT-qPCR. Representative results of 3 individual experiments are shown, and bars present standard deviations. Statistically significant differences between analyzed samples are indicated (*, P < 0.05; **, P < 0.01; ***, P < 0.001; calculated using two-way ANOVA with Bonferroni's correction test.