VLPs containing stalk domain and ectodomain of matrix protein 2 of influenza induce protection in mice

This study was approved by the Animal Care and Use Committee of Jiaxing University. All animal experiments were performed in accordance with the committee’s guidelines. Anesthesia was administered before immunization and sampling.

Sf9 insect cells were maintained in SF900II (Life Technologies, San Diego, CA, USA) at 27 °C in cell culture. Influenza virus A/Aichi/68 H3N2 and A/Anhui/2013 H7N9 provided by Dr. Zhu Hongwei (Ludong University) were propagated in the allantoic cavities of 10-day-old embryonated hen’s eggs at 37 °C for 2 d. Allantoic fluid was harvested and centrifuged at 2000 rpm for 10 min. The viruses were titrated by infection of mice with serial dilutions, and LD₅₀ (50% lethal dose) was calculated using the method of Reed and Muench.

The HA, M2e, and M1 genes were from the H3N2 sequence. The stalk domain (mHA) construction was based on our previous reports, the head domain of HA was replaced by peptide linker (GGGGGS)4 [14], and the M2e amino acid sequence is MSLLTEVETPIRNEWGCRCND. The coding sequence of glycosylphosphatidyl-inositol-GPI anchor was fused to the ends of the mHA and 4M2e coding gene to generate the gene encoding the membrane-anchored mHA and four individual repeat M2e (4M2e). The full-length construct encoding a GPI-anchored mHA and 4M2e was confirmed by DNA sequencing. The gene encoding M1, membrane anchored mHA, and M2e protein were cloned into pFastBac1. Recombinant baculovirus (rBVs) expressing mHA, 4M2e, and M1 were generated by using a Bac-to-Bac expression system (Invitrogen, Carlsbad, CA) according to the manufacturer’s protocol. Three different VLPs, namely, (mHA + 4M2e) VLP (chimeric VLPs, cVLPs) and standard VLP group (mHA VLP and 4M2e VLP), were produced by an insect cell expression system. For cVLPs, three rBVs expressing influenza M1, GPI-4M2e, and GPI-mHA co-infected sf9 cells at multiplicity of infection (MOI) of 1:4:2. Then, 4M2e VLP was produced by co-infection of sf9 cells with rBVs expressing 4M2e and M1 at MOI of 2:1. mHA VLP were produced by co-infection of sf9 cells with rBVs expressing mHA and M1 at MOI of 2:1. After 48–72 h infection, the culture supernatant was collected and VLPs were concentrated by experimental tangential flow concentration and purification dialysis system (Merck Millipore, Germany) followed by sucrose density gradient ultracentrifugation. The VLPs were characterized by Western blotting using antibodies against HA and M2e (R&D system). VLP protein concentration was determined by ELISA in which purified proteins were used to generate the quantitative standard curve. Bio-Rad protein assay (Bio-Rad Laboratories Inc., Hercules, USA) was used to quantify the yield of mHA and M2e protein in VLP.

Three- to four-week-old female BALB/c mice were separated into five groups, 15 mice for each group. The mice were immunized with soluble mHA, 4M2e, or influenza VLPs through intramuscular (IM) and intranasal (IN) administration, followed by two boosts at four-week intervals. Group 1 (G1) mice were immunized with soluble mHA protein. Group 2 (G2) mice were immunized with soluble 4M2e protein, Group 3 (G3) mice were given 4M2e-VLP. Group 4 (G4) mice received mHA-VLP. Group 5 (G5) mice were immunized with cVLPs. The mice were immunized with 50 μg target protein everytime and 25 μg for each administration route. Samples were collected two weeks after final immunization. Blood samples were collected by retro-orbital plexus puncture, and nasal washes were collected by lavaging the mice nostrils repetitively with 200 μl PBS containing 0.05% Tween 20 (PBST). Supernatants were collected after centrifugation (5,000 rpm for 10 min). Serum and mucosal samples were stored at − 80 °C for further assays. Lymphocytes from spleen samples collected from mice sacrificed two weeks after the final boost were used for cytokine assays. Four weeks after the final boost, the mice were challenged with 5 × LD₅₀ of mouse-adapted influenza virus. Body weight loss and survival rates were monitored daily for 14 d post infection, Fig. 1a.

mHA and 4M2e-specific antibody titers in serum were measured by ELISA. For mHA -specific antibody test, soluble mHA was used as coating antigens. For M2e-specific antibody test, soluble 4M2e was used as coating antigens. ELISA plates were coated with soluble protein (250 ng/well) overnight, plates were washed three times with PBS plus 0.05% Tween-20 and blocked with 1% BSA in PBS for 1 h at 37 °C. Following washes, serial dilutions of samples were added to the plate and incubated for 1 h at 37 °C. Second antibody, goat HRP-conjugated anti-mouse IgG antibody was added and incubated for 1 h at 37 °C. After wash, the tetramethylbenzidine (TMB, R&D Systems, USA) was added and the OD value was read in an ELISA plate reader using a test wavelength of 450 nm. The highest dilution factor that gives an OD 450 of twice that of the naive sample at the dilution was designated as the antibody end point titer. For the lung virus titers, the mouse lungs were collected in each group 4 d post challenge and ground into lung homogenates. Then, they were centrifuged at 1000 rpm for 10 min to remove tissue debris. The lung virus titers were determined by MDCK cell-based plaque assay as described by Wang et al. [15]. For the long-lasting antibody test, three mice in each group were kept without virus challenge. Samples were collected one, three, and six months after final immunization, Fig. 1.

The chromium-release assay were used to calculate the percentage specific immune lysis (SIL) of infected A549 cells [16, 17]. A549 cells were transfected with plasmid DNA encoding for HA stalk domain or M2 of H3N2 and H7N9 using Lipofectamine2000 (Invitrogen) two days before the experiment, for chimeric VLPs group, A549 cells were co-transfected with plasmid encoding HA stalk domain and M2. One day before the assay, transfected cells were harvested and washed, centrifuged, and labeled with ⁵¹Chromium for 1 h, the labeled cells were washed and added at 2000 cells/well in 96-well plates, fourfold serial dilution of heat-inactivated serum were added to wells in replicates of three, then plates were incubated at 37℃. NK cells were purified through MagnisortTM Mouse NK cell Enrichment Kit (Fisher Scientific Inc., Rockford, IL), enriched NK cells were added at an E:T (effector:target) ratio of 5:1/well. Plates were spun at 250 × g for 5 min and thenincubated at 37℃ in a culture incubator for 2 h. Supernatants were harvested and counted in a gamma counter. Calculation of %SIL at each serum dilution = %lysis of sample– %lysis of NK cells alone. Calculation of %lysis at each serum dilution = (average CPM of experimental release – average CPM ofminimum release)/(average CPM of maximum release – average CPM of minimum release) × 100. Calculation of %lysis of NK cells = (average CPM of NK cell release—average CPM of minimum release)/(average CPM of maximum release—average CPM of minimum release) × 100. CPM (count per minute), lysis of transfected A549 cells by NK cells alone without sera was against transfected cells, when calculating SIL values, data on lysis of NK cells alone were subtracted. The highest serum dilution showing ≥ 15% SIL was defined as ADCC antibody end point titer. For calculation of mean endpoint titers, an undetectable ADCC titer was assigned a value of 5.

Interferon gamma (INF-γ) and interleukin 4 (IL-4) secretions from immunized mouse splenocytes were evaluated using ELISA kits (R&D System, Minneapolis, USA) in accordance with the manufacturer’s instructions.

The analyses were performed by using GraphPad Prism version 5.00 for Windows (GraphPad Software, San Diego, CA). P values less than 0.05 (p < 0.05) were considered statistically significant. **p < 0.01, n.s.,p > 0.05.

The diagrams of 4M2e and HA stalk domain fusion gene constructs are as shown in Fig. 1b, c. VLP constructs were designed as shown in Fig. 2a–c, The VLPs were purified by sucrose density gradient ultracentrifugation. 4M2e-VLP was observed to have a molecular mass of approximately 13 kDa as shown in Fig. 2d. mHA-VLP was observed to have a molecular mass of approximately 40 kDa as shown in Fig. 2e. cVLP was observed to have two bands with molecular mass of 40 kDa and 13 kDa, indicated by arrow.

The immune sera and mucosal samples were evaluated for antigen-specific IgG and IgA titers using ELISA with soluble 4M2e and mHA as coating antigens. 4M2e-specific IgG titers in sera of 4M2e-VLP immunized group were approximately 30-fold higher than those of soluble 4M2e protein immunized group (p < 0.01) as demonstrated in Fig. 3a. mHA-specific IgG titers in sera of mHA-VLP immunized group were approximately tenfold higher than those of soluble mHA protein immunized group (p < 0.05) as shown in Fig. 3a. In the cVLP immunized groups, 4M2e-specific IgG titers(G5-2) in sera were approximately seven-fold higher than those of soluble 4M2e protein immunized group (p < 0.05). The mHA-specific IgG titers (G5-1) in sera were approximately seven- to eight-fold higher than that of soluble mHA protein immunized group (p < 0.05). In cVLP immunized group, 4M2e and mHA-specific IgG titers were lower than those of the 4M2e-VLP and mHA-VLP immunized groups.

Figure 3b shows that the VLP immunization induced enhanced mucosal responses as compared to the soluble protein immunization. mHA-specific IgA end-point titers in nasal washes from the mHA-VLP group were six fold higher than those from the mice immunized by soluble mHA protein. 4M2e-specific IgA end-point titers in nasal washes from the 4M2e-VLP group were six fold higher than those of mice immunized by soluble 4M2e protein. 4M2e and mHA-specific IgA titers were similar between the cVLP and standard VLP groups.

Furthermore, we determined whether VLP vaccination could induce long-lasting protective antibody responses. As shown in Fig. 3c, in 4M2e-VLP and mHA-VLP immunized groups, the IgG titers were maintained at comparable levels after three months, while an approximate 50% decrease in IgG titers was observed (P > 0.05) after six months. By contrast, the IgG titers dropped after three months in the two soluble protein immunization groups and the antibody titers decreased significantly after six months (P < 0.05). The results demonstrated that the VLPs induced long-lasting protective humoral immune responses in mice.

The evaluation of ADCC responses demonstrated that sera from 1 out of 5 mice in soluble mHA group and all mice in VLPs group had detectable ADCC antibodies. Compared to soluble immunized group, sera from VLPs immunized group exhibited increased ADCC activity in H3N2 antigen group, fig. For H7N9 antigen group, 1 out of 5 mice had detectable ADCC antibodies in soluble mHA group, compared to soluble immunized group, sera from VLPs immunized group got increased ADCC activity, shown in Fig. 4.

IFN-γ and IL-4 production in the spleen lymphocytes of the immunized mice were evaluated using a cytokine ELISA kit. In 4M2e-VLP, mHA-VLP, and cVLP group, the IFN-γ and IL-4 production levels in the spleen lymphocytes were significantly higher than that of the soluble group, as shown in Fig. 5. In the cVLP group, the IFN-γ and IL-4 production levels were 537 pg/ml and 471 pg/ml, respectively. In mHA-VLP group, the IFN-γ and IL-4 production levels were 662 pg/ml and 508 pg/ml, respectively. In 4M2e-VLP group, the production of IFN-γ and IL-4 levels were 560 pg/ml and 471 pg/ml, respectively. Compared to soluble protein group, the difference was statistically significant, and only background levels of cytokine-secreting cells were detected in the soluble protein group.

An effective influenza vaccine could limit the virus titers in the lungs post-challenge and conferred protection against viral infection. For the H7N9 (heterologous virus) challenge, as shown in Fig. 6, all the mice in the soluble protein immunized groups died 7–8d post- infection (p.i.) (Fig. 6c) with high titers of viruses in the lungs (Fig. 6a) and over 25% body weight loss (Fig. 6b). All mice in the 4M2e-VLP immunized group died with high titers of viruses in the lungs and over 20% body weight loss. Mice in the cVLP and mHA-VLP groups showed 100% and 50% survival rates, respectively, as well as effectiveness in reducing lung virus titers. cVLP effectively reduced virus titers from immunized mice compared to the soluble group with 1 × 10⁵ pfu/lung versus 2.5 × 10⁶ pfu/lung and 3.6 × 10⁶ pfu/lung after the H7N9 challenge. The mice immunized with cVLP had over 20-fold lower virus titers compared with those in the soluble group. Those in the mHA-VLP group showed an effective reduction of virus titers than those in the 4M2e-VLP group with 5 × 10⁵ pfu/lung versus 1 × 10⁶ pfu/lung. For the H3N2 challenge, the mice in the soluble 4M2e protein-immunized groups died 7–8 d p.i., and those in the soluble mHA protein-immunized group died 10 d p.i. (Fig. 6f). The cVLP and mHA-VLP immunization conferred full protection and significantly reduced the virus titers in the lungs (Fig. 6d). The mice in the mHA-VLP group showed 8%–10% bodyweight loss (Fig. 6e) and the 4M2e-VLP group had a 40% survival rate. The cVLP and mHA-VLP groups showed greater effectiveness in reducing virus titers from immunized mouse lungs post-challenge compared to the soluble group mice with 3.7 × 10⁴ pfu/lung and 1.1 × 10⁵ pfu/lung versus 2.1 × 10⁶ pfu/lung and 5.7 × 10⁶ pfu/lung after the H3N2 virus challenge. The mHA-VLP group showed a greater reduction in virus titers than the 4M2e-VLP group with 1.1 × 10⁵ pfu/lung versus 3.1 × 10⁵ pfu/lung. The mice that received three vaccinations with cVLP were fully protected and exhibited slight bodyweight loss after homologous or heterologous virus challenges. The mice in the mHA-VLP immunized group survived after the H3N2 challenge and showed partial protection after the H7N9 challenge, but exhibited some bodyweight loss. These results demonstrated that the immunization of mice with cVLP and mHA-VLP conferred more protection than those of the other groups after homologous or heterologous virus challenges and effectively reduced lung virus titers. Our findings also suggested that stem antibodies conferred better protection than M2e antibodies.