The BABL/c animals used in this study were maintained in a specific-pathogen-free animal facility in the Basic Medical Science Building (BSMB). They were all used according to the protocols approved by the Central Animal Care Facility (CACF), University of Manitoba (Protocol Approval Number 20-017).

The construction of pCAGGS-E$\mathrm{\Delta }$M-tM2e has been previously described [21, 22, 23, 24, 25]. The tM2e consists of two copies of influenza M2e from the human strain (SLLTEVETPIRNEWGCRCNDSSD), one from avian (SLLTEVETPTRNGWECKCSDSSD), and one from swine (SLLTEVETPIRNGWECRCNDSSD). The pCAGGS-HuM2 was described previously [26]. For constructing the plasmids expressing the chimeric protein AvM2c or SwM2c that contain the ectodomain of AvM2 or SwM2 with the HuM2 transmembrane and cytoplasmic domains, a polymerase chain reaction (PCR)-based technique was used to amplify cDNA encoding AvM2c or SwM2c. The 5${}^{\prime }$-primers encoding for AvM2c (SLLTEVETPTRTGWECNCSGSSD) and SwM2c (SLLTEVETPTRNEWECRCSDSSD) are 5-TCCGAATTCGTGCCGCCATGTCCCTGCTGACTGAAGTGGAAACACCAACACGGACTGGATGGGAATGCAACTGCAGCGGTAGCAGCGAC and 5-TCCGAATTCGTGCCGCCATGTCCCTGCTGACTGAAGTGGAAACACCAACACGGAGTGAATGGGAA, respectively. The 3${}^{\prime }$-primer is 5-AAAAGATCTGCTAGCTCATTC. Then, each PCR-amplified cDNA encoding for AvM2c or SwM2c was cloned into pCAGGS plasmid with EcoRI and BglII sites. The sequence of AvM2c and SwM2c was confirmed by sequencing analysis.

The influenza M2e monoclonal antibody (Santa Cruz Biotechnology, Dallas, TX, USA; 14C2: sc-32238) was used. The peptides used are; the human M2e peptide (SLLTEVETPIRNEWGCRCNDSSD; GenScript; RP20206), the swine influenza H1N1 M2e peptide (SLLTEVETPTRSEWECRCSDSSD) and the avian influenza H7N9 M2e peptide (SLLTEVETPTRTGWECNCSGSSD) and were synthesized by GenScript (Piscataway, NJ, USA) using the methods described previously [27]. The Western Blot (WB) detection enhanced chemiluminescence (ECL) kit was obtained from PerkinElmer Life Science (Waltham, MA, USA) and the QUANTI-Luc™ 4 Lucia/Gaussia was purchased from Invivogen (San Diego, CA, USA).

Human Embryonic Kidney 293T and Vero E6 purchased from ATCC were cultured in Dulbecco’s Modified Eagle Medium (DMEM), supplemented with 10% heat-inactivated fetal bovine serum (FBS) in a humidified incubator at 37 °C with 5% CO${}_2$. The Jurkat-Lucia NFAT-CD16 Human T Lymphocytes (NFAT-Jurkat) purchased from Invivogen (jktl-nfat-cd16) were grown and maintained in Iscove’s Modified Dulbecco’s Media (IMDM) containing 2 mM L-glutamine, 25 mM HEPES and 10% FBS in a humified incubator at 37 °C with 5% CO${}_2$. The 293T, Vero-E6, and NFAT-Jurkat cells were validated by STR profiling and underwent meticulous mycoplasma testing to ensure that they were free from contamination using Myco-Sniff-Rapid™ Mycoplasma Luciferase Detection Kit (ThermoFisher, Waltham, MA, USA).

The production of rVSV-E$\mathrm{\Delta }$M-tM2e which expresses an Ebola Glycoprotein (E$\mathrm{\Delta }$M) fused with four copies of conserved M2e consisting of 2 copies from influenza human strain, one from influenza avian and swine strains has been described previously [22]. The mouse-adapted strains of influenza viruses used in the animal challenge studies were A/Puerto Rico/8/1934 (H1N1) that were generated using reverse genetics which allows for manipulating and generating specific viral strains [28]. The titration of the rVSV-E$\mathrm{\Delta }$M-tM2e virus was quantified in Vero E6 cells using the TCID${}_{50}$ method [29]. The UV-inactivation of rVSV-E$\mathrm{\Delta }$M-tM2e was performed using a self-contained UV-C (CleanSlateUV, Limestone Labs, Toronto, Canada) consisting of six GLP/4P 18W germicidal lamps, three on top and three on the bottom, that emit UV-C light at a wavelength of 254 nm (Fig. 1A). A 6-well plate of 35 mm in diameter with 1 mL of rVSV-E$\mathrm{\Delta }$M-tM2e viral material (titer, 10${}^8$ TCID${}_{50}$/100 µL) was placed under the upper lamp at 128 mm, with plastic lid removed, and irradiated for 30 seconds eight times with triple shaking of the liquid at 3-minute intervals. The efficiency of virus inactivation was monitored by viral titration and the observation of the cytopathic effect on Vero cells.

Fig. 1.

The production and characterization of the UV-inactivated-rVSV-E$\mathrm{\Delta }$M-tM2e. (A) The representative diagram of the UV-C self-contained chamber light irradiation system of wavelength of 254 nm. rVSV-E$\mathrm{\Delta }$M-tM2e with a titer of 10${}^8$ TCID${}_{50}$/mL was placed in a 35 mm well in a 6-well plate dish, 128 mm below the UV-C light tube. (B) Viral titer of the UV- rVSV-E$\mathrm{\Delta }$M-tM2e and Live- rVSV-E$\mathrm{\Delta }$M-tM2e in a Vero E6 cells. (C) The expressions of the E$\mathrm{\Delta }$M-tM2e proteins in the UV-inactivated rVSV-E$\mathrm{\Delta }$M-tM2e and Live- rVSV-E$\mathrm{\Delta }$M-tM2e by WB. (D) Observation of cytopathic effect by UV-rVSV-E$\mathrm{\Delta }$M-tM2e or live-rVSV-E$\mathrm{\Delta }$M-tM2e, with MOI 0.01 at 48 h after infection, Scale bars: 100 µm. UV, ultraviolet light; rVSV, recombinant vesicular stomatitis virus; E$\mathrm{\Delta }$M, DC-targeting domain of Ebola Glycoprotein (with deleted mucin-like domain); tM2e, four copies of influenza matrix 2 protein ectodomain (two from human strains, one from avian and swine strains); WB, Western Blot.

To detect protein expression in cells, the HuM2, AvM2c or SwM2c transfected cell lysates or lysed rVSV E$\mathrm{\Delta }$M-tM2e samples were loaded onto a 10% or 12% SDS-PAGE and analyzed by using corresponding antibodies, as described previously [22].

A 6–8 weeks old of female BALB/c mice were immunized intranasally (IN) with live rVSV E$\mathrm{\Delta }$M-tM2e vaccine (1 $\times$ 10${}^5$ TCID${}_{50}$) or 1 $\times$ 10${}^8$ TCID${}_{50}$ UV-inactivated rVSV E$\mathrm{\Delta }$M-tM2e vaccine candidate and boosted with the same dosage on day 21 in a group of four. The immunized and the PBS-treated mice were sacrificed on day 35, and blood samples were collected for analyses on days 20 and 35. The influenza viral challenge in mice was conducted two weeks after the final immunization by infecting the mice intranasally with 2.1 $\times$ 10${}^3$ PFU of H1N1, while mice injected with PBS were challenged as a negative control. Weight loss or gain of the mice was monitored daily for two weeks after the challenge.

To detect M2-specific antibodies in mouse sera, ELISA plates (NUNC Maxisorp, Thermo Scientific) were prepared by coating them with 100 µL of human, avian influenza H7N9, or swine influenza H1N1 M2e antigens. Subsequently, 100 µL of diluted mouse serum samples were added to each well and incubated for two hours at 37 °C. Following washing steps, 100 µL of peroxidase-conjugated goat anti-mouse immunoglobulin G (IgG), IgG1, or IgA secondary antibodies were introduced and incubated for one hour at 37 °C. Finally, TMB (3${}^{\prime }$,3${}^{\prime }$,5${}^{\prime }$,5${}^{\prime }$-tetramethylbenzidine; Mandel Scientific, Guelph, Ontario, Canada) was added, and the absorbance at 450 nm (OD450) was measured.

The target cells were prepared to study influenza M2-induced ADCC by transfecting HEK 293 T cells with HuM2-, AvM2c- or SwM2c-expressing plasmids. After 24 hours of transfection, the cells were detached and washed once with DMEM and then adjusted to 5 $\times$ 10${}^4$ cells/well and added to the 96 well plates (50 µL/well). Meanwhile, 1.5 $\times$ 10${}^4$ cells/well (50 µL/well) of effector cells (NFAT-Jurkat cells) were mixed with serially diluted sera (50 µL/well) from the immunized mice and mixed with the target cell, resulting in a final volume of 150 µL. The cells were then incubated overnight. 30 µL supernatant from each well was applied to 25 µL QUANTI-Luc™ 4 Lucia/Gaussia in a 96-well flexible plate for evaluation in a POLARstar OPTIMA microplate reader (BMG Labtech, Ortenberg, Germany) [30, 31].

The M2e expression on the surfaces of transfected 293T cells was evaluated using anti-M2 (from the serum of rVSV E$\mathrm{\Delta }$M-tM2e immunized mice) and analyzed by fluorescence-activated cell sorting (FACS) [32]. Briefly, 293T cells were transfected with HuM2-, AvM2c- or SwM2c-expressing plasmids in a 6-well plate. After 24 hours of transfection, the cells were detached from the plate, incubated with mouse serum at 4 °C for 2 hours, washed and stained with anti-mouse IgG-FITC antibody in the dark for 30 min at 4 °C. After three washes with PBS, cells were acquired immediately on BD LSR-Fortessa (Becton, Dickinson, and Company (BD) Bioscience, Franklin Lakes, NJ, USA) and subsequently analyzed on FlowJo version 10.6.1 (FlowJo, LLC, Ashland, OR, USA).

Statistical analysis of antibodies was evaluated using the unpaired t-test. Significance denoted by p $\ge$ 0.05, was determined using GraphPad Prism 9.3.1 software (GraphPad Software, San Diego, CA, USA). With the use of a one-way ANOVA, multiple comparison test and Tukey’s test with Prism, the statistical analysis for the antibodies absorbance, endpoint titers, and mouse weight loss was performed. The two-way ANOVA multiple comparison test was used for the ADCC activity assessment. And the Log-rank (Mantel-Cox) test was used for the survival rate.

The rVSV-E$\mathrm{\Delta }$M-tM2e vaccine candidate has been previously described and characterized [21, 22]. Briefly, the vaccine candidate was constructed by fusing the conserved four copies of influenza M2e from human strains (2 copies), bird (1 copy) and swine (1 copy) (tM2e) to the E$\mathrm{\Delta }$M to form E$\mathrm{\Delta }$M-tM2e and expressed in the vesicular stomatitis virus (VSV) vector to generate the rVSV-E$\mathrm{\Delta }$M-tM2e as described [22]. We included M2e from different host strains because the M2e of avian or swine-origin influenza viruses have some amino acid substitutions. The rVSV-E$\mathrm{\Delta }$M-tM2e was inactivated in this study using UV-C light at the wavelength of 253.7 nm. Briefly, the rVSV-E$\mathrm{\Delta }$M-tM2e with a titer of 10${}^8$ TCID${}_{50}$/mL was placed in a 35 mm well in a 6-well plate dish, 12.8 cm below the UV-C light tube, and irradiated for 30 seconds, 8 times at 3-minute intervals (Fig. 1A).

To ensure the complete deactivation of the virus, both live and UV-inactivated rVSV-E$\mathrm{\Delta }$M-tM2e (UV-rVSV-E$\mathrm{\Delta }$M-tM2e) were subjected to titration on Vero E6 cells. It was observed that the live rVSV-E$\mathrm{\Delta }$M-tM2e virus efficiently infected Vero E6 cells while the UV- rVSV-E$\mathrm{\Delta }$M-tM2e had no measurable titre detected after 48 hours of infection (Fig. 1B). The expression of influenza M2e was detected in the UV-rVSV-E$\mathrm{\Delta }$M-tM2e at an equivalent level to the rVSV-E$\mathrm{\Delta }$M-tM2e (Fig. 1C). The cytopathic effect was observed on Vero E6 cells infected with the live rVSV-E$\mathrm{\Delta }$M-tM2e virus, but not the UV- rVSV-E$\mathrm{\Delta }$M-tM2e (Fig. 1D). The collective data indicates that the UV-inactivation of the rVSV-E$\mathrm{\Delta }$M-tM2e was effectively achieved and the inactivated rVSV-E$\mathrm{\Delta }$M-tM2e lost its infection ability.

Given the established evidence that the UV-rVSV-E$\mathrm{\Delta }$M-tM2e retains the presence of influenza M2e fused with E$\mathrm{\Delta }$M (E$\mathrm{\Delta }$M-tM2e) while lacking replication capabilities, we proceeded to examine the capacity of the UV-rVSV-E$\mathrm{\Delta }$M-tM2e in eliciting specific immune responses against influenza. Briefly, via IN immunization, the BALB/c mice received 1.0 $\times$ 10${}^5$ TCID${}_{50}$ of Live-rVSV-E$\mathrm{\Delta }$M-tM2e or UV-rVSV-E$\mathrm{\Delta }$M-tM2e (1.0 $\times$ 10${}^8$ TCID${}_{50}$) on day 0 and day 21. On days 20 and 35, mice sera were collected and measured for anti-M2e immune responses (Fig. 2A). Results showed that both the Live or UV-rVSV-E$\mathrm{\Delta }$M-tM2e immunization in mice yielded a notable robust anti-M2e IgG against human, swine and avian influenza M2e (Fig. 2B,C). Subsequent booster administration significantly heightened the immune response titre. Notably, no statistically significant difference was observed between the induced humoral immune responses by Live and UV-rVSV-E$\mathrm{\Delta }$M-tM2e. However, it is crucial to highlight that the UV-rVSV-E$\mathrm{\Delta }$M-tM2e dosage is 10${}^3$ higher than the Live-rVSV-E$\mathrm{\Delta }$M-tM2e. We further examine their ability to induce M2e IgG, IgG1 and IgA immune responses against human, swine, and avian influenza strains (Fig. 2D–F). While the rVSV-E$\mathrm{\Delta }$M-M2e vaccine demonstrated the ability to generate substantial immune responses targeting various influenza strains, our findings indicate notable differences in the immune responses elicited against different strains (Fig. 2B–F). Notably, the immune responses triggered against human strains were the most robust, whereas the responses against avian strains were comparatively higher but not significantly higher than those against swine strains (Fig. 2D,E). Interestingly, the IgA antibodies produced by either Live-rVSV-E$\mathrm{\Delta }$M-tM2e or UV-rVSV-E$\mathrm{\Delta }$M-tM2e were not significantly different (Fig. 2F). Overall, our data revealed that the inactivation of the viral vector vaccine candidate rVSV-E$\mathrm{\Delta }$M-tM2e with UV could still induce robust, broad immune responses against influenza M2e from different strains.

Fig. 2.

The specific anti-M2e elicited by UV-rVSV-E$\mathrm{\Delta }$M-tM2e or Live-rVSV-E$\mathrm{\Delta }$M-HM2e in sera from immunized mice. (A) Following IN immunization of the BALB/c mice with 1.0 $\times$ 10${}^8$ TCID${}_{50}$ of UV-rVSV-E$\mathrm{\Delta }$M-tM2e or 1.0 $\times$ 10${}^5$ TCID${}_{50}$ of Live-rVSV-E$\mathrm{\Delta }$M-tM2e in a prime-boost regimen (Day 0 and Day 21), the mice were bled on days 20 and 35 (n = 4/5 mice per group). The enzyme-linked immunosorbent assay (ELISA) assay detected (B) Anti-M2e IgG against M2e from human influenza strains after prime and boost immunizations respectively. (C) Anti-M2e IgG levels against M2e from swine and avian influenza strains after boost immunization. (D) Anti-M2e IgG levels present in the serum of the Live-rVSV-E$\mathrm{\Delta }$M-tM2e immunized mice against human (HuM2e), swine (SwM2e) or avian (AvM2e) influenza M2e strains. (E) Anti-M2e IgG1 levels in the serum of the mice immunized with Live-rVSV-E$\mathrm{\Delta }$M-tM2e against human (HuM2e), swine (SwM2e) or avian (AvM2e) influenza M2e strains. (F) Anti-M2e IgA levels against M2e from human, swine and avian influenza strains. Using an unpaired t-test, and significant p-values are represented with asterisks, *p $\le$ 0.01, the statistical significance between the groups was determined. ns, no significance.

In addition, we tested whether the influenza M2e antibodies in the immunized mice sera could recognize and react with the influenza M2e present on the cell surface. To achieve this, we constructed plasmids expressing human M2 (HuM2) and chimeric proteins (AvM2c and SwM2c) in which the HuM2e was replaced by the AvM2e or SwM2e (Fig. 3A), and transfected each of them into 293T cells. The expression of the HuM2, AvM2c, and SwM2c was detected by WB using Live-rVSV-E$\mathrm{\Delta }$M-tM2e immunized mice sera (Fig. 3B). Similarly, fluorescence-activated cell sorting (FACS) analysis was conducted on HuM2, AvM2c, or SwM2c transfected 293T cells that were probed with the mice sera. Meanwhile, E$\mathrm{\Delta }$M-tM2e or influenza HA transfected cells, were used as control. Interestingly, our results showed that the sera effectively recognized and bound with huM2, SwM2c, AvM2c and E$\mathrm{\Delta }$M-tM2e, but not HA (Fig. 3C). These findings strongly indicate that the vaccine candidates possess the potential for cross-protection against influenza strains affecting humans, swine, and birds.

Fig. 3.

Characterization of anti-M2e induced immune responses by Live-rVSV-E$\mathrm{\Delta }$M-tM2e. (A) Schematic representation of the influenza HuM2 expressing M2e from human (HuM2e), avian (AvM2e) or swine (SwM2e) and are named as HuM2, AvM2c or SwM2c. The ectodomain of the pCAGGS M2H5N1 was deleted and replaced with the ectodomain of avian (SLLTEVETPTRTGWECNCSGSSD) or swine (SLLTEVETPTRSEWECRCSDSSD). The M2e expression on the surfaces of transfected 293T cells was evaluated by anti-M2 staining (from the serum of immunized mice) and analyzed by WB (B) and fluorescence-activated cell sorter (FACS) and (C) analyzed on FlowJo version 10.6.1.

Considering that the UV-rVSV-E$\mathrm{\Delta }$M-tM2e exhibited the capacity to elicit an immune response like that of the Live-rVSV-E$\mathrm{\Delta }$M-tM2e, we proceeded to subject the mice that had been immunized with the UV-rVSV-E$\mathrm{\Delta }$M-tM2e to influenza H1N1 challenge. Groups of 4 mice were immunized with 10${}^5$ TCID${}_{50}$ of rVSV-E$\mathrm{\Delta }$M-tM2e, 10${}^8$ TCID${}_{50}$ of UV-rVSV-E$\mathrm{\Delta }$M-tM2e or PBS through the IN route on Day 0 and boosted with the same dosage on Day 21. After two weeks following the booster dose administration, the mice that had received immunization were challenged with the influenza virus, H1N1, with a viral titer of 2.1 $\times$ 10${}^3$ TCID${}_{50}$, employing the same methodology described earlier [22] (Fig. 4A). The mice were observed for 14 days for weight loss and survival. Both the Live rVSV-E$\mathrm{\Delta }$M-tM2e and the UV-rVSV-E$\mathrm{\Delta }$M-tM2e displayed a gradual decrease in weight of approximately 10% until Day 4, followed by a progressive weight gain without any instances of mortality. In contrast, mice in the PBS group deteriorated and lost weight sporadically within the four days of the viral challenge without survival (Fig. 4B,C). Interestingly, the anti-M2e immune responses increased after the challenge in both live and UV- rVSV-E$\mathrm{\Delta }$M-tM2e immunized mice, but not in the group of mice that received PBS (Fig. 4D). These results revealed the potency of the immune response induced by the UV-rVSV-E$\mathrm{\Delta }$M-tM2e to protect against influenza challenge.

Fig. 4.

The UV-rVSV-E$\mathrm{\Delta }$M-tM2e vaccination protected against the influenza H1N1. (A) The female BALB/c mice immunized IN with 1.0 $\times$ 10${}^8$ TCID${}_{50}$ of UV-rVSV-E$\mathrm{\Delta }$M-tM2e or 1.0 $\times$ 10${}^5$ TCID${}_{50}$ Live-rVSV-E$\mathrm{\Delta }$M-tM2e on days 0 and 21 were challenged with 2.1 $\times$ 10${}^3$ PFU of H1N1 on Day 14 post-immunization. (B) The percentage increase or decrease in weight loss or gain of the mice. (C) The survival rates of mice after infection with H1N1. (D) Levels of anti-M2e IgG in the serum after the challenge detected by ELISA. Using an unpaired t-test, the statistical significance between the groups was determined. ns, no significance.

Studies have suggested that the anti-influenza M2e that confers protection against influenza challenge could be through the ADCC [33, 34, 35, 36]. We, therefore, expanded our investigation to determine whether the Live rVSV-E$\mathrm{\Delta }$M-tM2e or UV-rVSV-E$\mathrm{\Delta }$M-tM2e vaccine can also elicit ADCC activity against influenza M2e proteins from avian and swine influenza strains. This is particularly significant because the vaccine formulation includes M2e regions from human, swine, and avian sources, making it crucial to evaluate its potential effectiveness against these diverse strains. ADCC is a key immune mechanism where effector cells with Fc receptors eliminate target cells coated with antibodies expressing pathogen-derived antigens [37]. Associations between ADCC activity and Fc receptor polymorphisms are observed in vaccination [37, 38]. The NFAT signalling pathway is crucial for Fc$\gamma$ receptor (Fc$\gamma$R) activation, playing a critical role in ADCC [39]. In this study, we employed NFAT-reporter systems, utilizing human Fc$\gamma$RIIa (CD16a) expressed in Jurkat cells as effector cells, and 293T cells transfected with influenza human M2 as target cells. We used Influenza huM2, SwM2c, or AvM2c transfected 293 T cells as targets, mixed with effector cells (NFAT-Jurkat cells) and serum from immunized mice. After overnight incubation, we evaluated the supernatant using QUANTI-LucTM 4 Lucia/Gaussia and a microplate reader (BMG Labtech, Germany) [30, 31]. The Live rVSV-E$\mathrm{\Delta }$M-tM2e vaccine could induce immune responses that elicited ADCC activity against the HuM2e, SwM2c and AvM2c. Interestingly, the UV-rVSV-E$\mathrm{\Delta }$M-tM2e vaccine also demonstrated ADCC activity against the HuM2, SwM2c and AvM2c, even though the level of ADCC activity was notably lower compared to that induced by the Live rVSV-E$\mathrm{\Delta }$M-tM2e vaccine (Fig. 5A–C). However, even with this lower ADCC activity, the antibody induced by UV-rVSV-E$\mathrm{\Delta }$M-tM2e was sufficient to protect mice against the highly virulent H1N1 influenza strain. Moreover, to further prove the specificity of the anti-M2e-mediated ADCC activity, the M2e-antibody-induced ADCC assay was performed in the presence of different concentrations of HuM2e peptides, and results showed that the anti-HuM2e ADCC were blocked using HuM2e peptides at different levels with various concentrations. Our results revealed that at the concentration of HuM2e peptide (10 µg/mL), the ADCC activity against HuM2 was significantly reduced (Fig. 5D). These data support the specificity and immunogenicity of the ADCC activity induced by the rVSV-E$\mathrm{\Delta }$M-tM2e.

Fig. 5.

The Live or UV-rVSV-E$\mathrm{\Delta }$M-tM2e immunized serum-mediated anti-M2e ADCC activity against influenza M2e from different host strains. ADCC activity induced by UV-rVSV-E$\mathrm{\Delta }$M-tM2e or Live rVSV-E$\mathrm{\Delta }$M-tM2e in the serum against (A) HuM2 (B) SwM2c or (C) AvM2c. (D) The Antibody-Dependent Cellular Cytotoxicity (ADCC) activity levels of the anti-M2e antibodies in the presence of HuM2e peptide. Using an unpaired t-test, and significant p-values are represented with asterisks, *p $\le$ 0.01, the statistical significance between the groups was determined.