Vaccination with CpG-Adjuvanted Avian Influenza Virosomes Promotes Antiviral Immune Responses and Reduces Virus Shedding in Chickens

Introduction

T ype A avian influenza viruses (AIV) can cause lethal infections in poultry, and some of the AIV subtypes, including the H5 and H7 viruses, may have the capacity of transmission to humans (10,27). Recently developed experimental and commercial vaccines for AIV in chickens include inactivated, killed, whole virus (28,30), vaccines developed using reverse genetics (19), viral-vectored vaccines (9,17), and DNA (16). Despite the availability of AIV vaccines for chickens, the major concern is the failure of these vaccines in curtailing virus shedding, thereby failing to prevent spread of infection among avian species, as well as transmission to humans.

Recent studies have revealed the usefulness of virosomes as a means of vaccination and protection against a wide range of viral infections, including influenza viruses (7,18). Reconstituted influenza virosomes consisting mainly of viral surface glycoproteins, hemagglutinin (HA), and neuraminidase (NA), can induce both antigen-specific antibody- and cell-mediated immune responses (4,11). Although virosome-based vaccines against influenza viruses have been developed experimentally and also commercially for mammalian species (7,26), the usefulness of these vaccines for conferring protection against AIV in chickens has not yet been addressed. In a recent investigation, we have shown that H4N6 avian influenza virosomes adjuvanted with either CpG-ODN or recombinant chicken interferon-γ (rIFN-γ) had immunogenic ability in chickens (21). In the present study, we tested the potential of AIV-H4N6 virosomes alone, as well as adjuvanted with CpG-ODN or rIFN-γ, in an experimental challenge infection with AIV in chickens. We chose AIV A/Duck/Czech/56/H4N6 as a prototype virus, to take advantage of its low pathogenicity, with the eventual aim of employing the same strategy for highly pathogenic AIV in chickens. We hypothesized that immunizing chickens with AIV-H4N6 virosomes alone or in combination with CpG-ODN or rIFN-γ will confer protection against homologous virus challenge, leading to a reduction in virus shedding from mucosal tissues of infected chickens. Furthermore, we sought to examine the elicitation of virus-specific systemic as well as cell-mediated immune responses to subsequent vaccination and challenge.

Methods and Materials

AIV A/Duck/Czech/56/H4N6 was propagated in embryonated chicken eggs (ECEs), and virus purification was performed as described by Reynolds and Maraqa (15), with minor modifications (21). Virosomes from inactivated influenza A/Duck/Czech/56/H4N6 viruses were prepared following the method originally described by Bagai and Sarkar (5), and modified by Markgraf et al. (23). In order to enhance the immunogenic potential of virosomes, rIFN-γ protein, expressed using the baculovirus expression system (22), or CpG-ODN (TCGTCGTTGTCGTTTTGTCGTT; the underlined sequences indicate the CpG dinucleotides), was incorporated into AIV virosomes as per the method described earlier (21). The sequence of CpG-ODN used in the study was based on the report by Gomis et al. (13), while the dose of virosomes (20 μg/bird), and rIFN-γ (50 μg rIFN-γ/bird) was determined based on our previous study (21). In order to carry out the challenge experiments, 70 one-day-old specific-pathogen-free (SPF) White Leghorn chickens were randomly allocated into five groups, with 12 chickens in each group. To obtain pre-immune sera, 10 newly-hatched chicks were euthanized and sampled. The primary immunization was given on day 7 post-hatch, while the secondary immunization was done at day 14 post-hatch. Chickens in different groups received an equal volume (100 μL) of different vaccine formulations. The groups were as follows: group 1: virosomes only (20 μg/bird virosomes); group 2: virosome-encapsulated rIFN-γ (20 μg virosomes encapsulating 50 μg rIFN-γ); group 3: virosomes + CpG-ODN (20 μg virosomes with 10 μg CpG-ODN); and group 4: negative control (treated with HNE buffer; 5.0 mM HEPES, 0.15 M NaCl, 0.01 M EDTA, pH 7.4, a virus re-suspending buffer). Chickens were challenged with 200 μL of A/Duck/Czech/56/H4N6 virus containing 1×10⁶ EID₅₀ dose on day 7 post-secondary immunization by the oculo-nasal route. To assess the effects of immunization with different virosome formulations on virus shedding, oropharyngeal swab samples collected at days 3 and 6 post-challenge were subjected to virus titration assays in MDCK cells (3). To examine correlates of immunity conferred by virosome formulations, antibody responses and expression of interferon genes in the spleen and lungs were measured. To this end, HI antibody titer, as well as systemic (serum IgG) and mucosal (IgA in lacrimal secretion) antibodies, were assessed by HI and indirect ELISA methods, respectively, as previously described (21). In addition, virus-neutralizing ability of immunized sera was assessed by a colorimetric virus neutralization (VN) assay, as described by Lehtoranta et al. (20), with some modifications. Finally, the relative expression of interferon (IFN-γ, α/β) genes in spleen and lung tissues collected on days 14 and 21 post-challenge was determined by real-time PCR, as previously described (1,2). Real-time PCR data were analyzed using RelQuant Software (Roche, Missisauga, Ontario, Canada) to calculate the relative gene expression. For cytokine gene expression, the data were normalized using the reference gene β-actin, and corrected for PCR efficiencies. Data were analyzed by one-way analysis of variance (ANOVA) using GraphPad Prism 5.01 software. Comparisons were considered significant at p≤0.05, and approaching significance when 0.05≤p≤0.1.

Results

At day 6 post-challenge, a significant (p≤0.05) reduction in virus shedding was observed in the swab samples collected from chickens that had been immunized with virosomes adjuvanted with CpG-ODN or rIFN-γ compared to control chickens. Although virus shedding was higher at day 3 compared to day 6 post-challenge, a significant reduction was also detected in chickens immunized with CpG-ODN or rIFN-γ compared to control chickens (Fig. 1A; p≤0.05). This was associated with significantly higher HI antibodies in chickens vaccinated with virosomes containing CpG compared to those receiving virosomes alone or in combination with rIFN-γ (Fig. 1B). Chickens that received virosomes containing CpG-ODN also developed significantly higher serum IgG or mucosal IgA antibodies at all post-challenge time points (days 7, 14, and 21 post-challenge), compared to the unimmunized challenged chickens or the chickens that received either virosomes alone or virosomes containing rIFN-γ (p≤0.05; Fig. 1C and D). This observation is in agreement with recent studies that suggested similar CpG-mediated effects in enhancing antibody responses against influenza viruses in mice (33), as well as in chickens (31). However, immunization with virosomes alone did not reduce viral shedding at any of the time points following virus challenge (Fig. 1A). In fact, compared to the unimmunized challenged group, there was significantly more virus shedding in the virosome-treated group. This finding raises the possibility that due to the presence of non-neutralizing antibodies, virus entry into host cells was facilitated. This phenomenon, which is called antibody-dependent enhancement, has been suggested to contribute to the pathogenesis of secondary dengue virus infections (14), West Nile virus infections (24), and some formalin-inactivated vaccines (6), including those for measles and respiratory syncytial viruses (25).

OPEN IN VIEWER

FIG. 1.

Virus shedding and elicitation of antibody responses following immunization with adjuvanted virosome formulations. (A) Assessment of virus shedding by in vitro titration assay in MDCK cells. The 50% tissue culture infective dose (TCID₅₀) was calculated from the cell cytopathic effect (CPE) induced by the virus present in the oropharyngeal samples in the MDCK cell culture using the Reed and Muench formula. The data represent the log₁₀ TCID₅₀ of each group, which consisted of 6 samples at each time point. (B) Induction of the HI antibody response in virus-challenged chickens immunized with different vaccine formulations. Sera of chickens immunized with virosomes, virosomes + CpG-ODN, virosomes + rIFN-γ, and unimmunized controls, were assessed for the presence of HI antibodies at days 7, 14, and 21 post-challenge. (C) Induction of serum antibody responses (IgG) in virus-challenged chickens immunized with different vaccine formulations. Serum antibodies (IgG) of chickens immunized with virosomes, virosomes + CpG-ODN, virosomes + rIFN-γ, and unimmunized controls were determined at days 7, 14, and 21 post-challenge using indirect ELISA. (D) Induction of mucosal antibody responses (IgA) in lacrimal secretions of virus-challenged chickens immunized with different vaccine formulations. The presence of IgA antibodies in lacrimal secretions of infected chickens immunized with virosomes, virosomes + CpG-ODN, virosomes + rIFN-γ, and unimmunized controls were determined at days 7, 14, and 21 post-challenge using indirect ELISA. The data in Fig. 1 represent the mean±standard error of 6 individual chickens per group. The difference between groups was assessed by ANOVA, and comparisons were considered significant at p≤0.05. Bars with different letters (a, b, and c) represent comparisons that were significantly different, whereas bars that share the same letters represent comparisons that were not significantly different.

In addition to HI and antigen-specific antibodies, we examined the presence of neutralizing antibodies in sera of chickens vaccinated with the virosome formulation containing CpG-ODN. A significantly higher virus-neutralizing activity was observed in serum of chickens that received virosomes + CpG-ODN compared to unimmunized control chickens (data not shown).

With respect to expression of interferon genes, we found a significant (p≤0.05) upregulation of the IFN-γ gene on day 14 post-challenge in spleens of chickens immunized with virosomes containing rIFN-γ compared to those that were immunized with virosomes alone or virosomes + CpG-ODN, as well as unimmunized challenged chickens (Fig. 2A). This finding indicates that, at least for the rIFN-γ-containing formulation, the presence of this cytokine might be a correlate of protection. However, IFN-γ did not appear to be correlated with immunity conferred by the CpG-containing formulation. The reason behind this finding is not clear, but it may be related to the time points we selected for sampling, or alternatively, immunity afforded by the CpG-ODN-adjuvanted formulation may be mediated through mechanisms that are IFN-γ-independent.

OPEN IN VIEWER

FIG. 2.

Relative expression of interferon genes in tissues at different post-challenge time points. Interferon gene expression for IFN-γ (A), IFN-β (B), and IFN-α (C), in spleen and lungs of AIV-challenged chickens immunized with virosomes, virosomes + CpG-ODN, virosomes + rIFN-γ, unimmunized challenged and unimmunized unchallenged (normal chickens) were determined at day 14 and day 21 post-challenge by real-time quantitative PCR using β-actin as reference gene. Data represent the mean±standard error of 6 individual chickens per group. The difference between groups was assessed by ANOVA, and comparisons were considered significant at p≤0.05. Bars with different letters (a, b, and c) represent comparisons that were significantly different, whereas bars that share the same letters represent comparisons that were not significantly different.

Evaluation of transcriptional changes in type I IFN gene expression in immunized chickens showed significantly (p≤0.05) higher expression of IFN-β in both spleen and lungs of chickens immunized with virosomes containing CpG-ODN or rIFN-γ at day 14 post-challenge, compared to chickens immunized with virosomes alone (Fig. 2B). In addition, significantly higher expression of IFN-α was also observed in the spleens of chickens immunized with virosomes alone, or virosomes containing CpG-ODN or rIFN-γ, at day 14 post-challenge, compared to unimmunized controls (Fig. 2C, p≤0.05). At this point, the role of type-I IFNs in adjuvanted virosome-mediated immunity is not clear. However, it is possible that the increased transcription of these interferon genes might have resulted from either the increased proliferation of IFN-α/β-producing cells such as lymphoid and myeloid cells, or other cell types. Alternatively, this may be due to increased migration of interferon-producing cells to these tissues. In the present study, we also noted a significant reduction in the transcription of IFN-α and IFN-γ in the unimmunized challenged birds compared to those that neither received the vaccine nor the virus challenge. This finding may suggest the influence of AIV non-structural 1 (NS1) gene on IFN expression as an immune-evasive mechanism (12).

Discussion

Altogether, the findings presented here demonstrated that vaccination of chickens with virosomes adjuvanted with CpG-ODN can induce high HI, as well as systemic and mucosal antibodies, and more importantly, can significantly lower virus shedding following virus challenge. It is known that CpG-ODN can lead to the induction of innate and adaptive antiviral responses (29). Importantly, CpG-ODN as a vaccine adjuvant can effectively enhance B-cell activation, and subsequent plasma cell proliferation and differentiation, leading to enhanced antigen-specific antibody production in mammals (11). In chickens, TLR21, a functional homologue of mammalian TLR9, has been demonstrated to play a key role in the recognition and response to CpG-ODN (8). In the present study, it is possible that the physical association of CpG-ODN with influenza virosomal surface proteins may have offered an advantage for delivery of these molecules to the host cell cytosol due to the membrane fusogenic activity of virosomes (32). This may lead to combined cell activation through CpG-ODN, as well as presentation of virosomal antigens by the same cell type.