Virus replicon particle vaccines expressing nucleoprotein of influenza A virus mediate enhanced inflammatory responses in pigs

Abstract

Studies in the mouse model indicate that the nucleoprotein of influenza A virus represents an interesting vaccine antigen being well conserved across subtypes of influenza virus but still able to induce protective immune responses. Here we show that immunizations of pigs with vesicular stomatitis virus- and classical swine fever virus-derived replicon (VRP) particles expressing the nucleoprotein (NP) of H1N1 A/swine/Belzig/2/01 induced potent antibody and T-cell responses against influenza A virus. In contrast to a conventional whole inactivated virus vaccine, the VRP vaccines induced both NP-specific CD4 and CD8 T cells responses, including interferon-γ and tumor-necrosis-factor dual-secreting cell. Although T-cells and antibody responses were cross-reactive with the heterologous H1N2 A/swine/Bakum/R757/2010 challenge virus, they did not provide protection against infection. Surprisingly, vaccinated pigs showed enhanced virus shedding, lung inflammation and increased levels of systemic and lung interferon-α as well as elevated lung interleukin-6. In conclusion, our study shows that NP, although efficacious in the mouse model, appears not to be a promising stand-alone vaccine antigen for pigs.

Figure 1

VRP vector mediated NP expression and induce NP-specific antibodies in pigs. (a) BHK-21 cells infected by VSV*ΔG-NP express NP. The left plot demonstrates GFP expression only after infection with VSV*ΔG, and the right plot GFP/NP double positive cells after infection with VSV*ΔG-NP. (b) Overlay histogram of SK-6 cells infected with CSFVΔE^rns (blue histogram) and CSFVΔE_rns-NP (red histogram). (c) Kinetics of NP antibody responses induced by VRP and WIV vaccines determined by ELISA. Pigs were vaccinated with either VSV*ΔG-NP, CSFVΔE_rns-NP, empty vectors or WIV vaccines, and immune sera collected at different time points and tested by ELISA. Titers were determined as the highest serum dilution leading to photometric values above the detection threshold. The pigs received a booster immunization at day 28. (d) Reactivity of sera collected at day 48 after VRP vaccination with MDBK cells infected with Belzig (H1N1, homologous virus), R757/10 (H1N2, heterologous challenge virus) or mock. In the upper row, representative results for sera from the VSV VRP vector-immunized pigs, and in the lower row results from the CSFV VRP-immunized animals are shown. The blue histograms represent mock-infected cells, and is overlaid with the red histograms representing the reactivity with virus-infected cells. (e) Reactivity of all sera from VSV*ΔG-NP, CSFVΔE^rns-NP and empty vectors injected pigs with Belzig (H1N1)- or R757/10 (H1N2)-infected MDCK cells, determined as described in Fig. 1d. Statistical significant differences between two groups were calculated using unpaired one-way ANOVA followed by Tukey’s multiple comparisons test (***P < 0.0001; **P < 0.005; *P < 0.05).

Figure 2

Peripheral blood CD4 and CD8 T-cell responses induced by VRP and WIV vaccines. PBMC were isolated at day 48 post vaccination (20 days after the booster injection) and re-stimulated in vitro with CAF or R757/10 (H1N2) virus to determine the percentage of virus-specific IFNγ (a,d), TNF (b,e) and dual cytokine (c,f) producing T cells in both the CD4 (A–C) and the CD8 (d–f) T cell subset. Asterisks indicate significant differences between two groups as calculated using unpaired one-way ANOVA followed by Tukey’s multiple comparisons test (***P < 0.0001; **P < 0.005; *P < 0.05).

Figure 3

Body temperature following heterologous challenge. VSV*ΔG-NP, CSFVΔE_rns-NP, empty vectors or WIV-vaccinated animals were challenged with R757/10 (H1N2). In (a), mean values and standard deviations of body temperatures are shown for each group. In (b), values for individual animals are shown for day 2 p.i. Asterisks indicate significant differences between two groups as calculated using unpaired one-way ANOVA followed by Tukey’s multiple comparisons test (***P < 0.0001; **P < 0.005; *P < 0.05).

Figure 4

Lung lesions after challenge infection. Lung lesions were assessed microscopically from each lobe at three days p.i. with R757/10 (H1N2). Scores were determined in a blinded manner by an even-handed pathologist for each lobe, and pooled to one value per animal. Asterisks indicate significant differences between two groups as calculated using unpaired one-way ANOVA followed by Tukey’s multiple comparisons test (***P < 0.0001; **P < 0.005; *P < 0.05).

Figure 5

Viral RNA loads after challenge infection. Viral RNA copies/ml were quantified by RT-qPCR in swabs on days one (a), two (b) and three (c), in BAL fluid on day one (d) and day three (e), and in lung tissue on day three (f) p.i. with R757/10 (H1N2). Asterisks indicate significant differences between two groups as calculated using unpaired one-way ANOVA followed by Tukey’s multiple comparisons test (***P < 0.0001; **P < 0.005; *P < 0.05).

Figure 6

In vivo IFN-α responses after challenge infection. IFN-α levels were determined by ELISA in serum of infected animals on days one (a), two (b) and three (a), in BAL fluid on day one (d) and day three (e), and in lung tissue on day three (f) p.i. with R757/10 (H1N2). Asterisks indicate significant differences between two groups as calculated using unpaired one-way ANOVA followed by Tukey’s multiple comparisons test (***P < 0.0001; **P < 0.005; *P < 0.05).

Figure 7

In vivo IL-6 and IL-1β responses in the lung after heterologous challenge. Levels of the inflammatory cytokines IL-6 (a) and Il-1β (b) were determined by ELISA in lung homogenates prepared on day three p.i. with R757/10 (H1N2). Asterisks indicate significant differences between two groups as calculated using unpaired one-way ANOVA followed by Tukey’s multiple comparisons test (***P < 0.0001; **P < 0.005; *P < 0.05).