Essay: Avian influenza

Avian influenza (AI) remains an increasingly global threat to economic and social well-being [1]. AI causes morbidity and mortality in humans and domestic animals. In many developing countries, H5N1 viruses have a serious effect on the poultry industry. Since 1997, direct avian-to-human transmission of the H5 subtype lethal avian influenza viruses that caused fatal human disease has been reported [2], which elevated the need to control AI beyond economic considerations. Vaccination has been a cost-effective means to control influenza. The approved influenza vaccines are primarily chemically inactivated detergent-solubilized virions that are incapable of inducing strong cytotoxic lymphocyte responses [3, 4]. Hemagglutinin (HA) and neuraminidase (NA), known as antigenic drift and shift, undergo progressive amino acid substitutions that can result in evasion of previously acquired immunity. Thus, influenza vaccines must be updated annually to be effective against circulating viral strains. The attenuated influenza virus vaccine licensed only for seasonal influenza could lead to the emergence of more aggressive revertants in cases of a limited number of amino acid changes in vaccine strains. There is a constant threat of an emergence of pandemic influenza A strains by a reassortment of segmented viral genomes [5, 6]. Both types of the licensed influenza vaccines described above primarily rely on fertilized chicken egg systems for production, which is time-consuming process susceptible to an influenza pandemic outbreak that could threaten global egg availability for vaccine manufacturing. Because current influenza vaccines have limited efficacy [7], there is an urgent need to develop novel vaccines targeting conserved viral antigens to combat the H5N1 influenza virus effectively.
As an alternative to conventional egg-based influenza vaccine processes, a non-infectious and self-assembled influenza virus-like particles (VLPs) platform has been developed. Lacking the viral genome, VLPs are safer than virion-derived vaccines. They elicit a robust and broad reactive immune response because VLPs accurately mimic the overall structure of virions and present conformational epitopes of surface proteins that are readily recognized and processed by antigen presenting cells such as dendritic cells, capturing the antigens for presentation to the B and T lymphocytes [8-12]. Recombinant VLPs are not involved in the use of live infectious viruses and do not require exceptional biosafety containment or pose a threat to vaccine production workers. Recombinant VLPs can be quickly manufactured for an emergency influenza pandemic, particularly because of their ability to differentiate vaccinated birds from infected ones easily. High-yielding influenza VLPs can be achieved in insect cells using the Bac-to-Bac expression system; for example, 5’10 mg of purified VLPs can be produced in 1 L of insect-cell culture that is similar in range to the amount of purified influenza virus obtained from 1 L of egg allantoic fluid [13]. Because insect cells are cultured in suspension, it is relatively easy to expand them to large bioreactors with competitive production costs. Because of these advantages over the licensed influenza vaccines, the VLPs have received significant attention in recent years.
In previous studies, VLPs (containing the HA, NA, and M1/M2 proteins) have been shown to be efficient inductors of an immune response in a mice/ferret model [14-20]. The nucleoprotein (NP) of the influenza virus is highly conserved, and the CD8+ T cells against the conserved epitopes of the NP could contribute to protection against morbidity and mortality from influenza [4, 16, 21-23]. The NP is associated with RNA segments and then packaged into virions. There are no studies on whether the NP can be packaged in VLPs without RNA segments.
If the NP is incorporated into virions, then the VLPs should have an enhanced immune response. To investigate the effects of the NP on the immunity protection of the VLPs, we engineered two influenza H5N1 VLPs; one contains HA, NA, and M1 (VLP3), and the other contains HA, NA, M1, and NP (VLP4). Our work demonstrated that VLP3 and VLP4 elicited vigorous humoral and cellular immune responses. Furthermore, these findings indicated that VLP4, but not VLP3, provided full protection against a heterogeneous H5N1 challenge in chickens.
Materials and Methods
Viruses and cell lines
Spodoptera frugiperda Sf9 cells were maintained in a serum-free SF900II medium (GIBCO, Grand Island, NY, USA) at 28??C in spinner flasks at a speed of 100 rpm. The wild-type strain of influenza A/Duck/Fujian/31/2007(H5N1) was provided by the Harbin Veterinary Research Institute (Harbin, China) for virus challenges in chicken models.
Generation of recombinant baculoviruses
A Bac-to-Bac baculovirus expression system was used for the generation of recombinant baculovirus vectors expressing influenza virus genes. The nucleotide sequence of segments 4, 6, 7, and 5 encoding the HA, NA, M1, and NP proteins, respectively, of A/goose/GD/1996(H5N1), whose accession numbers were NC_007362, NC_007361, NC_007363 and NC_007360, respectively, were de novo-synthesized by Invitrogen (Guangzhou, China). The fragments containing the influenza HA, NA, M1, and NP genes were cloned into the pFastBac Dual vector (Invitrogen, Carlsbad, CA, USA), followed by PCR using specific primers annealing to the 3??and 5?? terminus of each gene. The nucleotide sequences of the HA, NA, M1, and NP genes were confirmed by the DNA sequencing. The recombinant bacmids were generated by site-specific homologous recombination and transformation of the influenza genes-containing plasmid into E. coli DH10-Bac competent cells that contained the AcMNPV baculovirus genome (Invitrogen). Then, 1??g of purified recombinant bacmid DNA was transfected into Sf9 insect cells seeded in 6-well plates at 5??105 cells/ml using CellFectin reagent (Invitrogen). The cells were incubated for 3 days, and the virus harvested from the supernatant was subjected to three rounds of plaque purification.
Formation and purification of influenza VLPs
The influenza VLPs were attained by co-infection of the Sf9 insect cells with the following baculovirus recombinants: rBacHA-NA, rBacM1-NP, or rBacM1. The Sf9 cells were seeded at a density of 2??106 per flask and allowed to settle at room temperature for 30 min. Subsequently, the Sf9 insect cells were co-infected with the rBVs expressing HA-NA, M1-NP, or M1 at multiplicities of infection (MOI) of 3-5 and incubated for 72 h at 28??C. The culture supernatant (200 ml) from the Sf9 cells containing VLP3 and VLP4 were harvested and clarified by centrifugation for 30 min at 2000 ?? g at 4??C. The VLPs in the supernatant were pelleted by ultracentrifugation for 60 min at 100,000??g at 4??C. The sedimented particles resuspended in 1 ml of a phosphate buffered saline (PBS) solution (pH 7.2) were loaded onto a 20%-30%-60%(w/v) discontinuous sucrose step density gradient and sedimented by ultracentrifugation for 60 min at 100,000??g at 4??C. The VLP bands were collected and analyzed by SDS-PAGE and western blot. The functionality of the HA incorporated into the VLPs was assessed by the hemagglutination activity performed, as described, using 1% red blood cells (RBCs) from chickens [24].
SDS-PAGE and the western blot analysis of the purified VLPs
The protein content and identity of the VLPs was evaluated by SDS’PAGE using 5%’10% gradient polyacrylamide gels (Invitrogen) and western blot as described by Pushko et al. [18]. The expressed influenza proteins, HA, NA, M1, and NP, were detected with chicken polyclonal sera (Harbin Veterinary Institute) raised against the H5N1 influenza A/virus and the HRP-conjugated donkey anti-chicken secondary antibody (PTGLAB, USA). The amount of the HA protein present in the VLPs was estimated by densitometry of the coomassie blue stained SDS-PAGE gels [25].
Indirect immunofluorescence examination of infected Sf9 cells
The Sf9 cells were seeded in 24-microwell plates (Greiner Bio-One, Germany) at a density of 1??105 cells/well and infected with the triple/quadruple recombinant baculovirus at MOI of 3-5. After 72 h of infection, the Sf9 cells were washed three times for 5 min with PBS containing 0.05% Tween-20 and fixed in 100% pre-cooled methanol at 4??C for 10 min, then incubated with 0.5% Triton X-100 (USB, USA) at room temperature for 10 min. The RBITC/FITC-conjugated mouse monoclonal antibody of the influenza A virus M1/NP (ABCAM, England)(1:200 dilution) was used to detect the expression of the corresponding proteins. The plate was incubated at 37??C for 1 h. At the end of the incubation, the wells were washed extensively with PBS containing 0.05% Tween-20. The wells were overlaid with 200??l of 90% (v/v) glycerol in PBS and read under an UV microscope (Observer Z1, Carl Zeiss’Germany).
Electron microscopy
For the negative staining of the VLPs, the sucrose gradient-purified VLPs were applied to a carbon-coated Formvar grid for 2 min. The excess VLP suspension was removed by blotting with filter paper, and the grid was immediately stained with 1% phosphotungstic acid (pH 6.5) for 60 s. The excess stain was removed by filter paper, and the stained VLPs were observed with a transmission electron microscope (JEM -100CX-‘, JEOLLTD, Japan) at magnifications ranging from 6,000?? to 100,000??.
Preparation of the oil emulsion VLPs vaccine
For preparation of the oil emulsion VLPs vaccine, the purified VLPs were incubated with 0.18% formaldehyde (BBI, Canada) at 37??C for 24 h to inactivate the recombinant baculoviruses. After the virus activity testing in the sf9 insect cells, one part of the inactivated viruses containing 2% Tween-80 (BBI, Canada) was homogenized thoroughly with two parts of mineral oil for 5 min. Subsequently, one drop of the oil emulsion was dropped into a petri dish containing clear tap water in which the drop should retain its physical integrity without dilution. The oil emulsion VLPs vaccine was stored at 4??C without repeated heating and freezing[26]. The oil emulsion vaccine should be removed from the refrigerator the night before use and warmed to room temperature slowly.
Animal immunization and sample collection
Female inbred SPF BALB/c mice (Mus musculus) 4 weeks of age were housed in micro isolator units and allowed free access to food and water. For vaccination, all the mice were anesthetized with a 0.03’0.04 ml mixture of ketamine HCl (100 mg/ml) and xylazine (20 mg/ml) by a 5:1 volume ratio. Fifteen mice per group were intramuscularly (i.m.) immunized with VLP3 or VLP4 containing 1??g of HA three times at two-week intervals (at weeks 0, 2 and 4). Another group of mice were inoculated intramuscularly with the oil emulsion whole inactivated H5N1 virus vaccine (WIV) (Harbin Veterinary Research Institute). In the negative control group, the BALB/c mice were injected with PBS instead of the VLPs and H5N1 inactivated vaccine. The blood samples were collected from the anesthetized mice via the retro-orbital sinus before immunization at 2 weeks after the primary immunization and at 2 weeks after the final immunization. After the blood samples were allowed to clot and were centrifuged, the serum samples were collected and stored at -80??C prior to the antibody titration. Lymphocytes from spleen samples were collected from the sacrificed mice and used for the enzyme-linked immunospot (ELISPOT) analysis.
Eighty 1-day old SPF chickens were allotted randomly into four groups of twenty chickens each. All the chickens were housed in micro isolator units with filtered air. Two groups of chickens were intramuscularly inoculated with VLP3 and VLP4 containing 2.5 ??g of HA at 15 days of age. The positive control group of chickens was vaccinated with commercial WIV (Harbin Veterinary Research Institute). One group remained unvaccinated for the control. Blood samples were collected 3 weeks post-immunization to test the IgG antibody and the HI titer.
In the animals in this study, no anti-H5N1 serum antibody could be detected prior to vaccination. The experiments were conducted in accordance with the ethical guidelines for animal protection rights in China.
Hemagglutination inhibition activity in sera
A hemagglutination inhibition (HI) assay was used to assess the functional HA antibodies able to inhibit agglutination of the chicken erythrocytes, conducted as previous described[27]. To inactivate the non-specific inhibitors, the sera were treated with receptor destroying enzymes (RDE) prior to being tested[28]. The RDE-treated sera were two-fold serially diluted in v-bottom 96-well microtiter plates. An equal volume of adjusted 4 hemagglutination units of inactivated influenza virus Re-1 strain provided by the Harbin Veterinary Research Institute (Harbin, China) was added to each well. The plates were covered and incubated at room temperature for 20 min followed by the addition of 1% RBCs in PBS. The RBCs were stored at 4??C and used within 72 h of preparation. The HI titer was determined by the reciprocal dilution of the last row, which contained the non-agglutinated RBCs. Positive and negative serum controls were included for each plate. Geometric mean HI titers and the standard errors were calculated within the groups.
Evaluation of humoral immune responses
The influenza virus-specific antibodies of the different subtypes (IgG, IgG1, IgG2a, IgG2b, and IgG3) were determined in the mice sera samples by the Enzyme-Linked Immunosorbent Assay (ELISA), as described previously [19, 29]. Briefly, 96-well plates (Greiner Bio-One, Germany) were coated with 100 ??l of the inactivated influenza virus Re-1 strain at a concentration of 3 ??g/ml in a coating buffer (0.1 M of sodium carbonate, pH 9.6) at 4??C overnight. The plates were blocked with PBS-containing 0.05% Tween-20 and 1% BSA at 37??C for 1 h and incubated with serial dilutions of each sample at 37??C for 1 h. Following thorough washing in PBS containing 0.05% Tween-20, all of the samples were incubated with horseradish peroxidase (HRP)-labeled goat anti-mouse IgG, IgG1, IgG2a, IgG2b, or IgG3 (Bethyl, USA) 1:10,000 diluted in PBS containing 0.05% Tween-20 and 0.1% BSA at 37??C for 1 h. The unbound antibodies were removed and the wells were thoroughly washed. The substrate 3,3,5,5-Tetramethyl ben (TMB, Sangon, China) in citrate-phosphate buffer (pH 5.0) containing 0.01% H2O2 (BBI, Canada) was used for the color development. The reaction was terminated with 0.2 M H2S04, and the absorbance was determined at 450 nm using a spectrophotometer (Bio-Tek ELx800UV, USA).
The virus-specific IgG antibodies of chicken influenza were examined by ELISA, as described above, except that all of the chicken serum samples were incubated with horseradish peroxidase (HRP)-labeled donkey anti-chicken IgG (PTGLAB, USA) at 1:10000 dilution.
Cytokine assays
The spleens were harvested from the vaccinated mice, and the lymphocytes were isolated for the cytokine ELISPOT assays as described previously [16, 19]. Briefly, the spleens were carefully rinsed with sterile PBS and depleted of erythrocytes by treatment with ammonium chloride (0.1 M, pH 7.4). Following thorough washing with PBS, the cells were isolated from spleens using Mouse 1??Lymphocyte Separation Medium (DAKEWE, China). The collected cells were centrifuged at 800??g for 30 min at room temperature and resuspended in Lympho-Spot Serum-free Medium Rodent (DAKEWE). Cell viability was determined by 0.4% trypan blue (Sigma) staining. The number of specific mouse IFN-?? (mIFN-??) or mouse IL-4 (mIL-4) secreting cells was determined by an ELISPOT assay (R&D Systems, Minneapolis, MN, USA). Briefly, pre-coated anti-mIFN-?? (3 ??g/ml in coating buffer, BD/PharMingen) 96-well plates (Millipore) were incubated with 200 ml of Lympho-Spot Serum-free Medium Rodent at 25??C for 10 min and then with splenocytes isolated from vaccinated mice at 5??106/well. The splenocytes were stimulated with a mixture of NP-specific peptides (FWRGENGRKTRSAYERMCNILKGK, RLIQNSLTIERMVLSAFDERNK, and AVKGVGTMVMELIRMIKRGINDRN) at a concentration of 10 ??g/ml [14, 16, 19]. Additional wells of cells were stimulated with PMA (10 ng/ml) and ionomycin (500 ng/ml) or were mock stimulated. The plates were incubated for 20 h at 37??C with 5% CO2. The plates were thoroughly washed with PBS containing 0.05% Tween-20 and incubated with biotinylated anti-mIFN-?? at 37??C for 1.5 h. The plates were washed and incubated with streptavidin conjugated to alkaline phosphatase at 37??C for 1.5 h. Following extensive washing, the antibody-cytokine-antibody complexes were incubated with stable BCIP/NBT chromogen at 25??C for 45 min. The plates were rinsed with ddH2O and air dried at 25??C for 2 h. The spots were counted by an Immunospot ELISPOT reader (Bioreader4000, BIO-Sys, Germany).
Challenge with influenza virus
To test the protective efficacy of the VLPs, the vaccinated SPF chickens were challenged intranasally with a median lethal dose (50 LD50) of the clade 2.3.4 A/duck/Fujian/31/2007 (H5N1) virus 3 weeks after the immunization at the Harbin Veterinary Institute. The chickens were observed daily to monitor the morbidity and mortality during the viral challenge.
Statistical analysis
All the parameters were recorded for individual mice within all the groups. The statistical comparisons of data between the groups were conducted using the Analysis of Variance test (ANOVA). The statistical analyses were performed using Student’s two-tailed test with equal variance, and p values (p) less than 0.05 were considered statistically significant.
Results
Production and analysis of influenza VLPs
Influenza virus H5N1 VLPs were produced in insect cells co-infected with rBVs expressing the A/goose/GD/96 influenza virus (H5N1) HA, NA, M1, and NP (Fig. 1). The VLPs were purified from the culture supernatant as described above, and the relative HA content in the VLPs was estimated to be 20%. Two major bands could be observed between 30% and 60% of sucrose density, and the presence of the HA, NA, M1, and NP proteins in the purified VLPs was evaluated by western blot (Fig. 2A) and an indirect immunofluorescence assay (Fig. 2B). To localize the subcellular distribution of the expressed proteins, an indirect immunofluorescence assay of the Sf9 cells infected with triple/quadruple recombinant baculovirus constructs was performed. In the result, two conserved proteins, M1 and NP of VLP4, were efficiently expressed in the Sf9 cells (Fig. 2B).
Examination by electron microscopy of the negatively stained samples revealed the presence of the A/goose/GD/1996 (H5N1) influenza virus VLPs, with a diameter that ranged from approximately 80 to 120 nm, reminiscent of the authentic influenza virus and appeared as groups of bead-like structures (Fig. 2C).
An intrinsic property of all influenza viruses is the capacity to agglutinate RBCs of different species. It was important to determine whether the purified VLPs, which morphologically resembled the influenza virus, were able to agglutinate RBCs. The agglutination assay showed that the HA titer of the VLPs was 128 per 10 ??g of VLPs.
Vaccine induced HI levels in BALB/C mice
Following three immunizations, the VLP vaccine induced HI antibodies against the homologous strain in mice. After the immunization, all the mice developed functional HI titers (expressed as log2) (Table 1). No detectable HI activity was observed against any of the H5N1 HA antigens in the tested mice that received a mock vaccine.
Humoral immune responses elicited by influenza VLPs vaccine in BALB/C mice
The H5N1 VLP vaccines were administered by the intramuscular route in the presence of oil emulsion adjuvants in the mice model. The serum antibody response was determined by indirect ELISA. Similar assays were performed to measure the serum antibody response elicited by three intramuscular immunizations with WIV and a placebo control. All the mice remained healthy and showed no signs of abnormal behavior after vaccination with the pandemic influenza H5N1 VLPs. The prime immunization of the mice with H5N1 VLPs or WIV induced detectable levels of IgG antibodies as measured by ELISA (Fig. 3A and Fig. 3B). The IgG titers were greatly increased after the boost immunization in all the immune groups. The group of intramuscular application of VLP4 showed significantly enhanced levels of IgG titers in contrast to that of VLP3 (p < 0.05, Fig. 3A). These results indicated that the VLPs administered by the intramuscular route could effectively induce robust humoral antibody responses. The dominant serum IgG isotypes elicited at week 6 in the VLP-vaccinated mice were IgG2a, IgG2b, and IgG1, which was indicative of a T helper -type 1(Th1) and -type 2 (Th2) response (Fig. 3B); this finding is in agreement with previous reports [14, 15, 19]. No anti-H5N1 serum antibody was detected in the mice before vaccination.
VLPs promoted cell-mediated immunity in the BALB/C mice model
To investigate the specific T cell-mediated immunity elicited by the purified influenza VLPs, splenocytes from the mice treated with intramuscular injection of VLPs were harvested 3 weeks after the final immunization and were stimulated in vitro with NP-specific peptides to analyze the cytokine-secreting cells secreting IFN-?? (Fig. 4A) and IL-4 (Fig. 4B) cytokines. The highly specific and sensitive ELISPOT assays were performed. In this study, the use of the lymphocyte separation medium specific for mice and culturing the splenocytes in a serum-free-medium in an ELISPOT assay greatly enhanced spot detection and achieved perfectible results. In this case, significant levels of IFN-?? (Th1-type) and IL-4 (Th2-type) were detected in the VLP4-immunized mice, but not in the VLP3-immunized and negative control mice. The induction of the cellular immune responses could play a significant role in controlling or suppressing viral replication and infection.
VLPs induced HI and IgG antibodies in chickens
The HI results from the research with chickens showed that a single intramuscular immunization with the VLPs vaccine in the presence of an adjuvant elicited a high level HI titer against the inactivated H5N1 influenza virus Re-1 strain, indicating the strong immunogenicity provided by the VLPs (Table 1).
After one intramuscular immunization with the VLPs vaccine, the chickens exhibited a high level of IgG antibody (Fig. 5), and VLP4 elicited a strikingly enhanced level of IgG titers in contrast to VLP3 (p < 0.01), as in the results in the mice model.
VLPs’ protective efficiency against the heterologous lethal H5N1 viral challenge in chickens
Providing protective immunity to laboratory animals against a lethal virus challenge is a most important goal for preclinical vaccine research. To evaluate the level of protection afforded by the VLPs vaccine in the presence of an adjuvant administered via the intramuscular route, the chickens in the four groups were challenged intranasally with 50 LD50 of the clade 2.3.4 of the A/duck/Fujian/31/2007 (H5N1) virus 3 weeks after the single immunization. Our preliminary experiments in non-vaccinated chickens showed that the virus at this dose used as challenge caused severe influenza illness with typical clinical signs such as progressive inactivity; depression or poor spirit; sharp and labored breathing; sneezing; diarrhea; and reduced water and food intake, eventually leading to severe weight loss and death. These parameters were subsequently monitored each day during the viral challenge.
After the viral challenge, two measurements were used to assess the vaccine protective efficacy afforded by the VLPs vaccine as follows: clinical signs of illness and death. The chickens intramuscularly immunized with the VLP4 (the NP-containing VLPs) and the WIV vaccine in combination with oil emulsion as an adjuvant manifested no clinical signs; they showed no clinical signs of influenza infection and experienced a slight decrease in body weight at day 3 post challenge. The placebo control group showed quite clear and strikingly severe illness symptoms and succumbed to the lethal H5N1 virus infection at day 2, 3, and 4 post challenge (Fig. 6). The chickens immunized with VLP4 (the NP-containing VLPs) and WIV showed 100% survival after the lethal H5N1 influenza virus challenge (Fig. 6). In contrast, the survival rate of the chickens immunized with VLP3 was approximately 50%. Rapid clearance of the virus from the body and a low virus titer after the challenge are important signs of death and decreased morbidity and mortality post virus infection, and the VLP4 vaccination afforded a more significant decrease of cloacal shedding and tracheal shedding after the viral infection than was shown by VLP3 (data not shown). The incorporation of the conserved NP protein into the VLPs, and an effective adjuvant might have augmenting effects on viral clearance and protective efficacy, which to some extent broaden the spectrum of immunity to confer protection against a heterologous lethal H5N1 viral challenge.
Discussion
In this study, we attempted to improve the immune characteristics and protections of influenza H5N1 VLPs against heterogeneous strains. The results showed that the incorporation of the conserved NP of influenza H5N1 virus into VLPs (VLP4) enhanced their immune responses and protection against heterogeneous H5N1 strains.
In spite of the abundant experimental data demonstrating that VLP3, such as containing HA, NA, and M1/M2 proteins, could efficiently interact with the immune system, leading to the activation of lymphocytes and eventually providing solid protection from homologous or heterologous lethal viral challenge in mice or ferret models [14-20, 30], few efforts have been devoted to investigating the character of the conserved NP proteins of the influenza virus that play essential roles and confer protection in influenza vaccines.
Many research groups have repeatedly demonstrated that NP as well as the internal conserved M1 protein are effective at inducing a broad spectrum T-cell immune response such as stimulation of CD8+ T cells and generation of CTLs, and it is hypothesized that CTLs play important effector roles in the protection against a lethal influenza virus challenge. Incorporation of conserved NP protein into influenza VLPs might increase to some extent the efficiencies of the particle formation, assembly, budding, release and morphology, eventually stimulating strong cell immune responses and providing sound protection from homologous or heterologous lethal viral challenges. This finding is consistent with our findings in the chicken model virus challenge, that NP-containing VLPs (VLP4, derived from A/goose/GD/1996(H5N1), clade 0) afforded 100% protection against a heterologous lethal influenza virus challenge (clade 2.3.4) whereas the VLP3 vaccine conferred 50% protection; thus, it is reasonable to speculate that VLP4 is superior to VLP3 in protective efficacy, and VLP4 is hypothesized to have the greatest promise of generating a broad-spectrum anti-influenza vaccine.
We investigated the detailed immune responses induced by the insect cell-produced influenza VLPs, including the hemagglutination-inhibition activity, serum antibody isotypes, and cellular immune responses in a mice model. To test whether such a protective effect could be observed in different hosts, we conducted experiments in SPF chickens. The virus challenge in chickens results demonstrated that the influenza VLPs containing HA, NA, M1, and NP (VLP4) could induce protective immunity and provide complete protection against a lethal virus challenge with the heterologous virus strains of A/duck/Fujian/31/2007 (H5N1), which was consistent with the present finding that a clade 0 virus (A/goose/GD/1996(H5N1))-based vaccine could provide solid cross-protection against the heterologous virus in clades 1, 2.2, and 2.3.4 [30]. We could not rule out the possibility the NP-specific responses could have aided in the survival of the VLP4-vaccinated chickens. It is assumed that the inclusion of the highly conserved NP protein is advantageous because influenza-specific CTLs primarily target internal conserved proteins such as NP [31, 32], which could provide partial protection across the heterologous strains by promoting viral clearance and reducing the severity of symptoms. Incorporation of conserved NP into the VLPs (VLP4) might contribute to protection against morbidity and mortality from influenza [21, 33]. Mice immunized with H5N1 VLPs elicited high-level serum antibodies as well as IFN-?? and IL-4 secreting cells, which could be important for suppressing viral infection. This finding is consistent with our previous studies that influenza VLPs vaccines elicit predominately IgG2a, IgG2b, and IgG1 antibodies, indicative of a T-helper type 1 biased CD4+T cell response. In this study, all the mice vaccinated intramuscularly with the VLPs vaccines exhibited robust humoral responses and cellular responses. According to the results of HI and ELISA, the chickens administered with the identical dosage of the VLPs vaccines exerted results similar to those with mice. With regard to the antibody responses, the VLPs from various influenza virus strains have been shown to induce high IgG titers in species ranging from mice[14, 16, 18, 19] and ferrets[15] to chickens and ducks[14-16, 18, 19, 25]. In mice, these antibodies are dominated by IgG2a, IgG2b, and IgG1. The isotypes of IgG2a and IgG2b antibodies are likely to take part in effective viral infection and homologous as well as heterologous protection by the VLPs vaccination. The IgG2a isotype antibody is more effective than the other isotypes in viral clearance via multiple mechanisms including complement activation, the stimulation of antibody-dependent cellular cytotoxicity that aids in the clearance of viral infections[34-36]. HI is a widely used serological assay for measuring functional influenza-specific serum antibodies to neutralize viral particles following immunization with influenza vaccines. The HI titer is likely a major contributor to protection and effective clearance of the virus. Previous studies found less correlation between the HI titer and protection against the H5N1 viral infection[14, 37]. Our results confirm previous observations[15] that the antibody responses induced by immunization with the influenza VLPs in mice exhibited broader and higher levels of hemagglutination inhibition activity than immunization with the WIV vaccines containing the identical amount of HA. In contrast, the chickens immunized with a single intramuscular dose of H5N1 WIV showed a higher HI titer than the chickens immunized with the H5N1 VLPs. Most previous studies in mice, ferrets or chickens have focused on the immune responses and protective efficacy induced after a prime-boost immunization regimen. It is significant, as demonstrated in this study, that a single intramuscular dose of VLP4 could provide complete protection against the 50LD50 of the wild type A/duck/Fujian/31/2007(H5N1) in chickens. With regard to the cellular responses, our studies, with previous reports in mice[14, 16, 19], speculate that induction of the IFN-?? and IL-4 secreting T cell responses might correlate with reduced viral replication in the lungs and host protective immunity. It is conceivable that these memory T cells could be rapidly recruited to the respiratory tract; these cells could possibly contribute to lowering the lung viral titers and their survival. The enhanced protection provided by the VLP4 vaccines in chickens suggests that the conserved protein NP might have stimulated a specific cell-mediated response that hastened virus clearance in the airways or augmented the levels of the cross-reacting anti-HA antibodies, eventually improving the cross-protection against the A/duck/Fujian/31/2007 (H5N1) challenge. The VLP4 (containing NP protein) described as providing enhanced protection could potentially contribute in some degree to drift or heterologous protection owing to the greater degree of conserved NP protein compared to HA and NA. The VLPs vaccines consisting of the conserved M1 and NP components could provide significant protection, which illustrated that the immune responses against the conserved internal proteins might play an important defensive role against the influenza virus. Studies on the possible role of these proteins in VLP-induced protective immunity are under way.
This study is the first report to demonstrate that incorporation of the conserved NP protein into the VLPs (VLP4) can stimulate robust and vigorous humoral immune responses and cellular immune responses in a mice model. In chickens with one immunization, VLP4 can provide protection from a lethal virus infection by heterologous strains within the same subtype. Our studies provide insight for developing an efficient, safe and promising replication-defective VLP vaccine, especially a conserved NP protein containing VLP vaccine, to combat pandemic and pathogenic H5N1 influenza viruses.