A Single Intramuscular Dose of a Plant-Made Virus-Like Particle Vaccine Elicits a Balanced Humoral and Cellular Response and Protects Young and Aged Mice from Influenza H1N1 Virus Challenge despite a Modest/Absent Humoral Response

Antibody response.

Overall, the standard measures of influenza vaccine humoral response (hemagglutination inhibition [HAI] [Fig. 1A] and microneutralization [MN] [Fig. 1B]) were weak regardless of the vaccine used in both young and old animals. Only the young animals that received the VLP vaccine i.m. consistently mounted detectable HAI (geometric mean titer [GMT], 14.51; P < 0.0001) and MN (GMT, 17.5; P < 0.0001) responses compared to the other groups. Very low HAI (≤10) and/or MN (≤10) titers were observed in a small number of young and aged animals across the other groups. When serum HAI and MN titers were found to be very low, we decided to assess the humoral response by enzyme-linked immunosorbent assay (ELISA), as well (total pandemic H1N1 [pdmH1N1] HA-specific IgG). Antibodies measured by ELISA were readily detected in most groups (Fig. 1C) but were consistently higher in the young animals that had received the VLP vaccine i.m. (GMT, 1,771.1 ng/ml) than in both the VLP i.n. group and the split-virion group (GMT, 109.1 ng/ml and 265.1 ng/ml; P < 0.01 and P < 0.0001, respectively). Low ELISA titers were also detected in the aged animals that received either the H1-VLP (GMT, 526.9 ng/ml) or the split (GMT, 364.3 ng/ml) vaccine i.m. (Fig. 1C). HA-specific IgA antibodies in the bronchoalveolar lavage fluid (BALF) were undetectable in all the animals, including the VLP i.n. group (data not shown).

Cellular immune response to vaccines.

Cytokine/chemokine and lymphoproliferative responses of splenocytes restimulated with H1 antigen ex vivo were also most consistently detected in the H1-VLP i.m. group (Fig. 2; see Fig. 4) and were generally more robust in young mice than in aged mice. Since split-virion formulations are the most common commercial vaccines, we compared the cytokine/chemokine responses of the H1-VLP i.m. group to those of the i.m. split-virion group at the two ages. Splenocytes from the young H1-VLP recipients produced a greater (2- to 14-fold) amount of a broad range of cytokines/chemokines than splenocytes from the split-vaccine recipients, including interleukin 2 (IL-2), granulocyte-macrophage colony-stimulating factor (GM-CSF), IL-3, gamma interferon (IFN-γ), IL-4, IL-17, IL-5, IL-10, monocyte chemoattractant protein 1 (MCP-1), MIP-1a, and IL-6 (Fig. 2A). Differences between the H1-VLP and split-virion groups in the aged animals were far more modest, so that animals that received the H1-VLP vaccine had slightly (1.1- to 1.4-fold) higher IL-2, GM-CSF, IFN-γ, IL-3, IL-4, IL-5, tumor necrosis factor alpha (TNF-α), and IL-6 production than the split-virion group (Fig. 2B) but slightly (0.9- to 0.6-fold) lower production of IL-1α, IL-1β, IL-12, IFN-γ, and RANTES. None of these differences reached statistical significance. Antigen-specific cytokine/chemokine responses comparing young versus aged animals are presented in Fig. 3. Overall, cytokine/chemokine responses of the young and aged animals to the VLP and split vaccines were similar, with a few striking exceptions. For the H1-VLP vaccine given i.m., the young animals made much greater quantities of IFN-γ, while IL-1α production was much greater in the aged animals. In response to the split vaccine, the young mice produced significantly more IFN-γ, IL-5, and IL-1α than the aged group. In preliminary experiments with either H1-VLP or split-virion formulations, the antigen used for restimulation (i.e., H1-VLP, whole inactivated pdmH1N1, or recombinant H1) did not significantly influence splenocyte proliferation or cytokine production (data not shown).

Compared to the split-virion groups, splenocyte proliferation was consistently higher in the H1-VLP i.m. groups for both age groups, although significance was reached only in the young mice (stimulation index [SI], 7.38 ± 2.57 versus 3.51 ± 1.38 [P < 0.0001]) versus the aged mice (SI, 4.12 ± 0.87 versus 2.52 ± 0.95 [P = 0.3325]) (Fig. 4).

Protection from homologous challenge.

When challenged i.n. with 5 times the mouse 50% lethal dose (mLD₅₀) of A/California/07/2009 H1N1 virus, young naive mice rapidly lost up to 21.5% of their baseline weight, and most were euthanized when a humane endpoint was reached (when the mice lost more than 20% of their initial body weight.) Only 1/11 (9.1%) of the young naive animals survived (Fig. 5A and C). Young mice given the H1-VLP vaccine i.n. also lost substantial weight (15.0%), but 80% of them recovered and survived. The young split-vaccine and H1-VLP i.m. groups lost the least weight (10.8% and 9.8%, respectively), and all of the animals survived. Overall, the aged animals lost more weight following challenge than the younger mice. The aged naive group also lost the most weight (20.2%), with kinetics similar to what was observed in the younger animals (Fig. 5B and D). Most of these animals met humane endpoints or died (22.2% survival). Aged mice that received the split-virion vaccine i.m. or the H1-VLP vaccine i.n. lost similar amounts of weight (17.8% and 18.3%, respectively) (Fig. 5B) and had similar rates of survival (60% and 55.6%, respectively). The aged mice given H1-VLP i.m. lost the least weight (14.2%), and 80% survived challenge. Although more than half of the aged animals in the VLP i.n. and split-virion groups survived, most of the animals remained well below their baseline weights (15 to 18% loss) at 12 days after challenge. Only the H1-VLP-vaccinated animals recovered a substantial amount of weight, reaching statistical significance in the last days of the experiment (P < 0.0001 versus the split-vaccine group at day 12).

Lung viral load at 3 days postinfection.

Overall, vaccination had relatively little impact on day 3 viral loads in either young or aged animals. In the young mice, the highest viral loads were found in the split-vaccine group (mean of log values, 4.07, versus naïve, 4.42) (Fig. 6). The greatest decrease in the day 3 viral load and the only decrease that reached statistical significance was observed in the young H1-VLP i.m. group (3.68; P < 0.05) (Fig. 6). The H1-VLP i.n. group had a slight decrease from the naive and split-vaccine groups (4.02). Among the aged animals, the split-vaccine and H1-VLP i.n. mice had the highest viral loads (4.27 versus 4.29, respectively) compared to the naïve mice (4.50). The H1-VLP i.m. group had a small decrease in the day 3 viral loads (4.06) (Fig. 6).

Antibody responses after infection.

At 3 days postchallenge, HAI titers were largely unchanged from the day 21 titers (Fig. 5A). All the other groups, except H1-VLP i.m. young mice, had undetectable HAI titers (<10) (Fig. 7A). ELISA titers rose 2- to 3-fold in the H1-VLP i.m. groups in both the young (GMT, 2,612.63 ng/ml) and aged (GMT, 719.13 ng/ml) mice. The GMT among the vaccine-naive aged mice was 53.25 ng/ml, suggesting that this range represents background in this assay. HA-specific IgA antibodies were not detectable in the BALF at 3 days postinfection (data not shown).

Lung histopathology.

Lung histopathology total scores were highest in the young split-vaccine group (7.2 ± 5.1) and in both young and aged mice in the H1-VLP i.m. groups (7.0 ± 3.3 and 7.0 ± 4.6, respectively) from a total possible score of 20. It is interesting that the groups with the highest scores also had the best survival after lethal infection. The naive groups of both young and aged animals had lower scores (3.0 ± 3.7 and 2.2 ± 1.3, respectively) that were very similar to the scores in other groups that also had lower survival (scores between 3.4 and 3.6).

Virus, mice, and vaccines.

Young adult (5- to 8-week-old) and aged (16- to 20-month-old) female BALB/c mice (Charles River Laboratories, Montreal, QC, Canada) were divided into the following groups: naive, detergent-inactivated split-vaccine given i.m. (A/California/07/2009 H1N1 or pdmH1N1; BEI Resources, Manassas, VA), and H1-VLP vaccine given either i.m. or i.n. (Medicago Inc., Quebec, Quebec, Canada). The H1-VLP vaccine was produced as previously described (35) using the wild-type sequence of HA protein from pdmH1N1. The complete study consisted of 6 to 15 mice per group in 5 separate experiments (Table 1). The mice received a single dose of vaccine (3 μg based on HA content) on day 0 or were left naive. For i.m. injections, 50 μl of vaccine was administered into the quadriceps muscle (right leg for the VLP vaccine; bilaterally for the split vaccine) using a 28-gauge, 1/2-in. needle. Intranasal instillation (25 μl/nare) was performed in mice anesthetized with isoflurane.

Peripheral blood was collected from the lateral saphenous vein before immunization (day 0) and 21 days postvaccination (day 21). The blood was collected in microtainer serum separator tubes (BD Biosciences, Mississauga, ON, Canada). Cleared serum samples were obtained by following the manufacturer's instructions and stored at −20°C. In each experiment, 6 mice per group were sacrificed in a CO₂ chamber at day 21. Serum was collected by cardiac puncture, BALF was obtained in complete RPMI (cRPMI) medium, and then spleens were harvested from individual mice and splenocytes were isolated as previously described (36).

The remaining mice (10 or 11 animals/group) were challenged on day 21 with 5 times the LD₅₀ of wild-type A/California/07/2009 H1N1 virus (1,800 50% tissue culture infective doses [TCID₅₀] in 50 μl: National Microbiology Laboratory, Public Health Agency of Canada) by i.n. instillation, and weight loss was monitored daily for 12 days. In preliminary experiments, the mLD₅₀ was found to differ between young and aged animals (approximately 663 and 105 TCID₅₀, respectively). Three days postinfection, 7 to 14 mice/group were sacrificed to measure the lung viral load. For these mice, serum, BALF, and lungs were collected. All procedures were carried out in accordance with the guidelines of the Canadian Council on Animal Care, as approved by the Animal Care Committee of McGill University.

Antibody titer measurements.

An HAI assay was performed to detect pdmH1N1-specific antibody in the mouse sera at day 0 and day 21 postvaccination and 3 days postinfection, as previously described (37). Briefly, mouse sera and receptor-destroying enzyme (RDE) (Cedarlane, Burlington, ON, Canada) were mixed 1:4 and incubated for 18 h at 37°C. The RDE was inactivated at 56°C for 30 min, and the sera were serially 2-fold diluted in phosphate-buffered saline (PBS) (pH 7.4) to a starting dilution of 1:10 in 96-well V-bottom plates (Corning Inc. Costar, Corning, NY). The diluted sera (25 μl/well) were then incubated with 8 HA units of pdmH1N1 for 30 min at room temperature before 0.5% turkey erythrocytes diluted in PBS (50 μl/well; Lampire Biological Laboratories, Pipersville, PA) were added to the wells. The HAI titer was determined by visual inspection as the highest dilution that inhibited erythrocyte agglutination, using standard criteria.

MN titers were measured as previously described (38). Briefly, confluent monolayers of Madin-Darby canine kidney (MDCK) cells (British Colombia Center for Disease Control) were incubated in 96-well flat-bottom plates (Falcon Corning Life Science, Corning, NY) in MegaVir (VWR, Radnor, PA) supplemented with 10 μg/ml gentamicin (Gibco Life Technologies, Burlington, ON, Canada), 0.25 μg/ml amphotericin B (Gibco Life Technologies, Burlington, ON, Canada), 100,000 U/ml penicillin G (Sigma, St. Louis, MO), and 10 μg/ml glutamine (Wisent, St. Bruno, QC, Canada). Sera were heat inactivated at 56°C for 30 min and diluted 2-fold in MegaVir starting at 1:10 in duplicate wells (60 μl/well). Each well then received 100 infectious units of pdmH1N1 virus diluted in MegaVir (60 μl/well), and the plates were incubated at 37°C in 5% CO₂. The cytopathic effect (CPE) was assessed at 4 days, and the titer was defined as the reciprocal of the highest serum dilution to completely block CPE.

ELISAs were performed as previously described (36). Briefly, U-bottom, high-binding 96-well plates (Greiner Bio-One, Frickenhausen, Germany) were coated overnight at 4°C with recombinant HA from pdmH1N1 (Immune Technologies, New York, NY; 0.5 μg/ml) in 100 mM bicarbonate/carbonate buffer at pH 9.5 (50 μl/well). Each plate contained a standard curve with 2-fold dilutions of purified mouse IgG (Sigma, St. Louis, MO) starting at 2,000 ng/ml. The wells were blocked with 2% bovine serum albumin (BSA) (Sigma) in PBS-Tween 20 (0.05%; Fisher Scientific, Ottawa, ON, Canada). The sera were heat inactivated (as described above), diluted 1:50 in blocking buffer, and added to four wells (50 μl/well). The plates were incubated for 1 h at 37°C; blocking buffer was added to the standard curves at this time. Horseradish peroxidase (HRP)-conjugated anti-mouse total IgG antibodies (Jackson Immuno Research Laboratories Inc., West Grove, PA) diluted 1:50,000 in blocking buffer were added (75 μl/well; 1 h at 37°C). 3,3′,5,5′-Tetramethyl benzidine (TMB) substrate (100 μl/well; Millipore, Billerica, MA) was used for detection, followed by 0.5 M H₂SO₄ after 15 min (50 μl/well). The optical density (OD) was measured at 450 nm with an EL800 microplate reader (BioTek Instruments Inc., Winooski, VT). The concentration of HA-specific IgG was calculated using the mouse IgG standard curve.

Splenocyte isolation and stimulation and cytokine analysis.

Individual spleens were harvested 21 days before challenge in Hanks' balanced salt solution (HBSS) without calcium or magnesium (Wisent, St. Bruno, QC, Canada) at room temperature (RT) and processed as previously described (36).

Supernatant was collected and measured as previously described, except at 3 × 10⁵ cells/well in 200 μl under the following conditions: cRPMI alone (unstimulated) or with H1-VLP (2.5 μg/ml HA) in cRPMI for 72 h at 37°C in vitro. Spleens were isolated from a total of 13 mice per group from 2 experiments and tested in duplicate on multiplex ELISA by Quansys (Logan, UT). Cytokine/chemokine data are presented as radar graphs as previously described (39).

Splenocyte stimulation and cell proliferation assay (CPA).

Splenocytes were placed in duplicate in 96-well U-bottom plates (BD Falcon, Mississauga, ON, Canada) at 10⁵ cells in 200 μl with cRPMI alone (unstimulated), with H1-VLP vaccine (2.5 μg/ml HA), or with concanavalin A (ConA) (2.5 μg/ml) for a stimulation control in cRPMI. After 72 h at 37°C plus 5% CO₂, the plates were spun down (300 × g; 10 min at RT), and the supernatant was removed. The cells were pulsed with 1 μCi/well [³H]thymidine (MP Biomedical, Solon, OH) for an additional 18 h. After one freeze-thaw, the cells were harvested on glass fiber filters with a Tomtec harvester 96 (Tomtec Inc., Hamden, CT), and [³H]thymidine incorporation was measured with a scintillation counter (Wallac Microbeta Trilux 1450 beta-counter; Wallec, Turku, Finland). Cell proliferation values were expressed as the SI; for each mouse, the SI was equal to the average Ag-stimulated counts per minute divided by the average unstimulated counts per minute. I.n. administration was excluded from this analysis, since in preliminary experiments, there was no evidence of a cellular response in the spleen (data not shown).

Lung viral load.

Lungs were collected 3 days postinfection and individually homogenized in an equal amount (wt/wt) of MegaVir medium (VWR, Radnor, PA) using a tube homogenizer. The samples were centrifuged at 14,000 × g for 5 min at 4°C, and the supernatants were collected and stored at −80°C. Virus titers were measured by the TCID₅₀. MDCK cell monolayers were prepared in 96-well flat-bottom plates. On the day of inoculation, the MegaVir was removed and the lung homogenates were serially diluted 1:10 with MegaVir TPCK (tosylsulfonyl phenylalanyl chloromethyl ketone)-treated trypsin (Sigma; 2 mg/ml) (Megavir-trypsin). Each dilution was added to 6 wells (100 μl/well) and incubated at 37°C with 5% CO₂ for 1 hour. The lung-MegaVir mixtures were then removed, and the plates were replenished with fresh MegaVir-trypsin (200 μl/well). The plates were followed for 4 days for CPE. The TCID₅₀ was calculated using the Karber method (40). Lung viral loads were assessed 3 days postinfection (day 3) from mice infected with 5 times the LD₅₀.

Lung histopathology.

Lung samples were processed using the standard hematoxylin and eosin (H&E) stain. Briefly, lung samples were fixed in 10% formalin (Fisher Scientific, Ottawa, ON, Canada) and then embedded in paraffin (Leica, Concord, ON, Canada). Sections (4 μm) were applied to slides and then heated at 50 to 60°C. Samples were washed with xylene (Chaptec, Quebec, QC, Canada) and then immersed in ethanol (Commercial Alcohols, Boucherville, QC, Canada) for 10 min. The slides were rinsed with distilled water and then briefly submerged in Harris hematoxylin (HHS-32; Sigma, St. Louis, MO) one-half solution in distilled water. The slides were washed under running tap water and then washed 10 times with ethanol, followed by 1 min of eosin-phloxin B (100 ml of 1% eosin Y [Sigma, St. Louis, MO], 10 ml of 1% phloxin B [Sigma], 780 ml of ethanol, and 4 ml of glacial acetic acid [Fisher Scientific, Ottawa, ON, Canada]). Samples were immersed in ethanol for 10 min. The slides were dried and then submerged in xylene for 10 min. The slides were fixed with 2 drops of acrytol (Leica, Concord, ON, Canada) with a coverslip and scored at ×10 and ×100 magnification. The slides were scored blinded, and the scoring system evaluated the following 5 parameters: (i) airway epithelial necrosis, attenuation, or disruption; (ii) airway inflammation; (iii) peribronchiolar and perivascular lymphocytic cuffing; (iv) alveolar cellular exudate/edema and interlobular edema; and (v) alveolar septal inflammatory cells and cellularity (41). Each parameter was scored from 0 to 4 for a total possible score of 20 (Table 2).

Statistical analysis.

The geometric mean ratios between groups and their 95% confidence intervals (CI) were calculated using GraphPad Prism 6.0 software. For statistical analysis, one-way analysis of variance (ANOVA) was performed on HAI and stimulation index values. All other statistical analysis was by two-way ANOVA. All analyses were performed using GraphPad Prism 6.0 software.