A polyvalent virosomal influenza vaccine induces broad cellular and humoral immunity in pigs

Background Influenza A virus (IAV) is endemic in pigs globally and co-circulation of genetically and antigenically diverse virus lineages of subtypes H1N1, H1N2 and H3N2 is a challenge for the development of effective vaccines. Virosomes are virus-like particles that mimic virus infection and have proven to be a successful vaccine platform against several animal and human viruses. Methods This study evaluated the immunogenicity of a virosome-based influenza vaccine containing the surface glycoproteins of H1N1 pandemic, H1N2 and H3N2 in pigs. Results A robust humoral and cellular immune response was induced against the three IAV subtypes in pigs after two vaccine doses. The influenza virosome vaccine elicited hemagglutinin-specific antibodies and virus-neutralizing activity. Furthermore, it induced a significant maturation of macrophages, and proliferation of B lymphocytes, effector and central memory CD4+ and CD8+ T cells, and CD8+ T lymphocytes producing interferon-γ. Also, the vaccine demonstrated potential to confer long-lasting immunity until the market age of pigs and proved to be safe and non-cytotoxic to pigs. Conclusions This virosome platform allows flexibility to adjust the vaccine content to reflect the diversity of circulating IAVs in swine in Brazil. The vaccination of pigs may reduce the impact of the disease on swine production and the risk of swine-to-human transmission. Supplementary Information The online version contains supplementary material available at 10.1186/s12985-023-02153-5.


Background
Influenza A virus (IAV) subtypes H1N1, H1N2 and H3N2 are endemic in swine herds globally, causing economic losses for the swine industry and public health concerns.The control of influenza in pigs in Brazil is a challenge due to co-circulation of multiple genetically distinct viruses of subtypes H1N1, H1N2 and H3N2.Human seasonal IAVs of subtypes H1N1 and H1N2 were introduced in swine in the middle of 1980s, and in the early 2000s, respectively, and formed three Brazilian genetic clades within the lineage H1-1B (1B.2.3, 1B.2.4 and 1B.2.6) [1][2][3].In the middle 1990s, a human seasonal H3N2 IAV was introduced in swine and diversified into three genetic clades (1990.5. 1, 1990.5.2 and 1990.5.3) [3].Regarding to H1N1 pandemic (pdm) IAV, several human-to-swine spillover events have occurred since 2009, however, only four of these introductions resulted in sustained onward transmission in swine, giving rise to four distinct genetic clusters within the lineage 1A.3.3.2 [4].A substantial antigenic diversity between distinct subtypes and lineages of Brazilian swine IAVs (swIAVs) has been found, and may impact vaccination [5].In recent years, the genetic and antigenic diversity of swIAVs has expanded through reassortment events among viruses co-circulating in pigs and accumulation of amino acid changes in genes encoding the viral glycoproteins, hemagglutinin (HA) and neuraminidase (NA) [1,6].The HA binds to cell receptors and mediates virus entry into cells, and is the main antigen against which neutralizing antibodies are induced during infection or vaccination [7].The HA gene is highly variable, harboring mutations that can lead to antigenic variation and, consequently an antigenic mismatch between the vaccine and infection strain that can lead to vaccine failure, and thus immune escape [8].
Vaccination is the most effective measure to mitigate and control influenza-associated morbidity and mortality in swine populations.Additionally, IAV vaccination in pigs contributes to human health by reducing zoonotic transmission and the appearance of "variants" in humans, as well as the emergence of pandemics.However, the rapid viral evolution and co-circulation of multiple distinct IAV lineages pose a challenge for the development of cross-protective vaccines [9].Whole inactivated influenza virus (WIV) vaccines are commercially available in many countries around the world [10] and induce humoral immune response and protection in pigs against challenge with homologous virus [11].Liveattenuated influenza vaccine (LAIV) induces a cell-mediated immune response and improved protection against challenge with heterologous virus but may reassort with wild-type IAV or revert virulence [12,13].Other vaccine platforms using viral vectors, nucleic acid-based particles and virus-like particles have also been tested in pigs [14][15][16][17].These vaccines stimulate antibody and cell-mediated responses, are safe and can be used to construct polyvalent vaccines that can be updated [18].In Brazil, a WIV containing the H1N1pdm virus as well as autogenous vaccines have been commercialized since 2014 and 2017, respectively.However, IAV vaccination in pigs is not a common practice in Brazilian farms.
Therefore, novel vaccine strategies that induce wide cross protection, are safe and can be rapidly updated are required.Virosomes consist of reconstituted viral envelopes, but without virus genetic material [19,20].
Their use is an alternative for the control of influenza in pigs since they mimic virus infection, eliciting a broad immune response [20].Influenza virosomes preserve the receptor-binding and membrane fusion activities of the HA, allowing the presentation to the major histocompatibility complex (MHC) class I and II, interacting with the immune system through pathways similar to IAVs, and resulting in high immunogenicity [19].Virosomal influenza vaccines have already been developed for humans and poultry [21,22].However, its use in pigs has been poorly studied.
The aim of this study was to assess the immune response kinetics of a virosomal-based influenza vaccine containing the viral envelope proteins of H1N1pdm, H1N2, and H3N2 IAVs in pigs.
The virosomal influenza vaccine was prepared as previously described by Fonseca et al. [24].Briefly, the three concentrated viruses were mixed with 200 mM of 1,2-dicaproyl-sn-glycero-3-phosphocholine (DCPC), then diluted 1:2 (v/v) in TNE buffer, and the final mixture was incubated in an ice bath for 30 min to ensure viral dissolution.Viral nucleocapsids were removed by ultracentrifugation (100,000 × g for 30 min at 4 °C).The supernatant was extensively dialyzed against TNE buffer for 48 h at 4 °C to remove the DCPC, which led to the self-assembling of virosomes.The virosomal particle was characterized previously by Fonseca et al. [24].The influenza vaccine contained 128 µg of total HA and 20% (v/v) of Emulsigen ® -D (MVP Laboratories) per mL.The HA concentration corresponded to 6, 21 and 73% for H1N1, H1N2 and H3N2, respectively, as evaluated by SDS-PAGE [24].

Animal study design
Forty-three (43) four-week-old pigs were obtained from a SPF herd composed of crossbreed pigs (MS 115-composite terminal sire developed by Embrapa Swine and Poultry with Landrace × Large White sows).All pigs were previously tested negative for IAV antibodies (Multispecies Influenza A Antibody Test kit, BioChek) and IAV RNA [25].All pigs were transferred to the biosafety level 1 (BSL1) experimental facility one week before the beginning of the experiment and cared for in compliance with the Animal Use Ethics Committee of Embrapa Swine and Poultry (protocol number 001/2017).The pigs were randomly divided into two groups: G1: 10 non-vaccinated pigs; and G2: 33 vaccinated pigs.Pigs from G2 were vaccinated on days (D) 0 and 14 with 1 mL of the adjuvanted virosomal IAV vaccine, by the intramuscular route in the neck.Pigs were daily monitored for clinical signs, behavior, appetite and temperature or any adverse effects related to vaccination.Blood and nasal swab samples were collected from all pigs on D0, D14 and D28.Three pigs from the G2 group were kept in the experimental facility until D90 to assess the long-term immunity induced by vaccination, and additional blood and nasal swab samples were collected on D60 and D90.The pigs were anesthetized with 6 mg/kg of Zoletil ® (Zolazepam + Tiletamine; 100 mg/mL, Virbac) by the intramuscular route and euthanized with one step electrocution followed by bleeding.Necropsy was performed on D28 for the G1 and G2 groups, and on D90 for the three remaining pigs from the G2 group.Spleen from all pigs were excised and kept in RPMI 1640 medium supplemented with 1 × penicillin, streptomycin and fungizone for the in vitro cell proliferation assay.Bronchoalveolar lavage fluid (BALF) [26] and blood with anticoagulant (BD Vacutainer ® EDTA K2) were collected for cell profile analysis by flow cytometry.Lung, mediastinal lymph node, spleen, liver and kidney fragments were collected and preserved in 4% paraformaldehyde for histopathological assessment (Fig. 1A, B).

Diagnostic microbiology
The presence of other respiratory pathogens was investigated as follows: RNA was extracted from nasal swab samples using MagMAX ™ Viral RNA Isolation kit (Thermo Fisher Scientific) and tested by RT-qPCR targeting IAV/matrix gene [25].Serum and BALF samples were submitted for DNA extraction using the DNeasy ® Blood & Tissue kit (Qiagen).Serum samples were evaluated by qPCR for detection of porcine circovirus type 2 (PCV2) [27].BALF samples were evaluated by qPCRs for detection of Actinobacillus pleuropneumoniae [28], Glaesserella parasuis [29], Mycoplasma hyopneumoniae [30] and Pasteurella multocida [31].Samples were not tested for porcine reproductive and respiratory syndrome virus (PRRSV), since Brazil is considered free of PRRS [32].

Histopathological evaluation and immunohistochemistry
Lung, mediastinal lymph node, spleen, liver and kidney tissue samples fixed in 4% paraformaldehyde, were routinely processed and stained with hematoxylin and eosin (H&E) [33].The mediastinal lymph node samples were evaluated for PCV2 detection by immunohistochemistry [34].Additionally, to confirm that the vaccine was not cytotoxic to the pigs, the qualitative and quantitative analysis of apoptotic cells was performed in lung, mediastinal lymph node, spleen, liver and kidney tissues by using the DeadEnd ™ Colorimetric TUNEL (Terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling) System kit (Promega) in accordance with the manufacturer's recommendations but using the 3-amino-9-ethylcarbazole (AEC) chromogen.Microscopic analysis was carried out using optical microscopy (Axio Scope.A1, Zeiss) at 400 × magnification and consisted of blind quantification of positive and total cells in ten different fields, each field measuring 37,000 µm 2 .

Determination of cellular profile
The cellular profile from the blood and BALF collected from the pigs was assessed by flow cytometry.Heparinized whole blood samples were diluted 1:3 (v/v) in PBS and the peripheral blood mononuclear cells (PBMCs) were isolated by density gradient centrifugation (Ficoll-Paque ™ PLUS, GE Healthcare), following the manufacturer's recommendations.PBMCs were resuspended in flow cytometry buffer (PBS supplemented with fetal bovine serum (FBS, 2% v/v, Gibco), bovine serum albumin (BSA, 2% w/v, Sigma-Aldrich) and sodium azide (0.01% w/v, Sigma-Aldrich)).BALF samples were centrifuged at 800 × g for 10 min at 4 °C and the pellet was resuspended in flow cytometry buffer.The PBMC and BALF cell concentration was adjusted and distributed to approximately 1 × 10 6 cells/well (96-well plate).The cells were stained with specific monoclonal antibodies (mAbs) [37] to assess major lymphocyte, monocyte/macrophage, granulocyte and natural killer (NK) cell populations by four-color flow cytometry (Table 1).The PBMCs and BALF cells resuspended in flow cytometry buffer were incubated for 30 min at room temperature with a cocktail of specific mAbs.The fluorochrome-labeled mAbs used for the PBMCs and BALF cells are described in Table 2.For intracellular staining, the cells were treated with the Cytofix/Cytoperm ™ Fixation/Permeabilization kit (BD), according to the manufacturer's instructions, and then the cells were stained with CD79a (the epitope recognized by the mAb is located in the cytoplasmic domain) and CD3 (the PPT3 clone recognizes an extracellular and intracellular epitope on CD3) mAbs.All antibodies used in the staining were previously titrated for their optimum concentrations.To evaluate fluorochrome unspecific staining, respective isotype controls for anti-IgG1, anti-IgG2a and anti-IgG2b were analyzed in the preliminary procedure to set up technical parameters.
The stained cells were acquired using an Accuri ™ C6 Plus flow cytometer (BD).Fifty thousand events were analyzed based on forward scatter (FSC) and side scatter (SSC), using Accuri ™ C6 Plus (BD) and FlowJo ™ (Tree Star Inc.) software.Before sample analysis, flow cytometer settings were verified using Cytometer Setup and Tracking beads (CS&T beads, BD) as described in the manufacturer's instructions.Compensation beads were used with single stains of each antibody to establish the compensation settings.The SSC threshold was set at 8,000 units to eliminate debris.Gates considered to indicate positive and negative staining cells were set based on fluorescence minus one (FMO) tests of samples, and these gates were performed systematically on each sample, allowing minor adjustments for SSC variability.Dead cells were excluded by discrimination with 7-AAD dye, according to our protocol previously described [38].

In vitro cell proliferation assay
Due to the lack of an adequate biosafety structure for the challenge of pigs with influenza virus, the in vitro challenge by culturing swine splenocytes and stimulating them with the three vaccine virus strains was performed.For this, spleen fragments were mechanically dissociated under aseptic conditions and filtered through a 70 μm Nylon Cell Strainer (Corning).After this, red blood cells were depleted using Pharm Lyse ™ Buffer (BD Biosciences) for five minutes at room temperature.The splenocytes were obtained after the addition of RPMI 1640 medium to stop the lysis reaction, followed by washing the cells twice with RPMI 1640 medium.Splenocytes were labeled with 2.5 µM carboxyfluorescein succinimidyl ester (CFSE) by applying the CellTrace ™ CFSE Cell Proliferation kit (Invitrogen), for 15 min at 37 °C in the dark.The reaction was stopped by adding six volumes of RPMI 1640, supplemented with 10% FBS, followed by incubation for 5 min in an ice bath, in the dark.Finally, the cells were washed three times with RPMI 1640-10% FBS, and further resuspended at a concentration of 5 × 10 6 cells/mL in the same medium.Viable cells were cultured in RPMI 1640 medium supplemented with 10% FBS (Gibco), 1 mM GlutaMAX (Gibco), 25 mM HEPES (Sigma-Aldrich), 1 mM sodium pyruvate (Sigma-Aldrich), 50 M 2-mercaptoethanol (Gibco) and 100 U/mL penicillin-streptomycin (Sigma-Aldrich) at 37 °C under 5% CO 2 .During 96 h in the dark, cells were stimulated in vitro by adding 8000 TCID 50 /mL of each of the three vaccine viruses separately (H1N1, H1N2 and H3N2).For the negative control, culture medium was added to one well of the microplate (non-virus-stimulated cells).For the positive control, two wells of the microplate were prepared, one well with 3 µg/mL of Concanavalin A from Canavalia ensiformis (Sigma-Aldrich) and another well with 10 µg/mL of lipopolysaccharide from Escherichia coli (Sigma-Aldrich).CFSE in combination with mAbs, enabled the concomitant access to cell proliferation and activation status of cell subpopulations.Flow cytometry analysis was performed to identify and quantify lymphocyte subpopulations (CD3, CD4, CD8, CD79a mAbs), to measure the levels of cellular activation marker expression (CD25, CD19 mAbs), cellular memory marker expression (CD27 mAb), interferon-γ (IFN-γ) cytokine and also to quantify macrophages (BioRad Serotec) (Table 1).Proliferation was detected by loss of CFSE fluorescence [38].
Cells obtained after 96 h of stimulation were resuspended in flow cytometry buffer, and the cell concentration was adjusted and distributed to approximately 1 × 10 5 cells/well (96-well plate).To assess major lymphocyte and macrophage populations, cells were stained with the fluorochrome-labeled mAbs described in Table 2.
For intracellular staining, the cells were treated with the Cytofix/Cytoperm ™ Fixation/Permeabilization kit (BD), according to the manufacturer's instructions, and stained with CD79a, CD3 and IFN-γ mAbs.The respective isotype controls and the acquiring stained cells were performed as described in the previous section.
The gate was performed as described in a previous study conducted by our research group [38].Gates were set using the non-virus-stimulated sample for each individual pig.In summary, the gate was based on forward scatter (FSC) and side scatter (SSC) properties to estimate lymphocyte population and debris exclusion.The doublet cells were subjected to doublet plotting showing forward scatter height (FSC-H) against forward scatter area (FSC-A).Dead cells were excluded from the analysis using 7-AAD staining.Counterstaining with CD3e/ CD79a allowed us to gate on T and B cells, respectively (Fig. 2).Among T lymphocytes, subsets of CD4 + and CD8 + T cells were captured using panels.The gate to identify monocyte/macrophage cell population was based on which T lymphocytes could be recovered from CD4 bright cells.Among B-lymphocytes cells were defined as the sum of CD79a + , the gate is similar to what has been described for T lymphocytes.The gate for positive cells to CD79a + was determined, which was differentiated in SWC7 + (conventional B cells).

Statistical analysis
Differences between vaccinated (G2) and non-vaccinated (G1) groups were evaluated using the two-sided Student's t test in the statistical analysis system (SAS) [39].P values ≤ 0.05 were considered statistically significant.

Absence of clinical signs and respiratory pathogens
No respiratory clinical signs, changes in behavior, appetite or hyperthermia were observed in the pigs during the experiment.Also, no local inflammatory reactions or any adverse effects related to the vaccination were identified.During the experiment, two pigs died, one on D14 (from G2 group) and another on D22 (from G1 group).Both deaths were not related to the experiment.The pig from G2 group had unintended lesion on vagus nerve during blood collection on jugular groove which led to respiratory and cardiac depression followed by death.For the pig from G1 group, a necropsy was performed and no macroscopic or microscopic lesions were observed.Thus, the G1 and G2 groups remained with 9 pigs and 32 pigs, respectively.
All nasal swab samples collected on D0, D14, D28, D60 and D90 were negative for IAV by RT-qPCR.Moreover, serum samples collected on D0, D28 and D90 were negative for PCV2 and BALF samples were negative for Actinobacillus pleuropneumoniae, Glaesserella parasuis, Mycoplasma hyopneumoniae and Pasteurella multocida by qPCR.

Safety of the virosomal influenza vaccine in pigs
In the histopathological analysis, no significant lesions were detected in lung, spleen, liver and kidney tissue samples collected from all pigs.However, in mediastinal lymph node samples, follicular lymphoid hyperplasia was observed in 71.87% (23/32) of pigs from the G2 group, and mild histiocytic infiltration was observed in 11.11% (1/9) of pigs from the G1 group and 15.62% (5/32) from the G2 group.All mediastinal lymph node samples tested negative for PCV2 by immunohistochemistry.
In the TUNEL assay, there was no difference in the number of apoptotic cells observed in lung, mediastinal lymph node, spleen, liver and kidney tissues between non-vaccinated pigs (G1 group) and vaccinated pigs (G2 group) (Additional file 1: Fig. S1).
The polyvalent influenza virosome vaccine proved to be safe and non-cytotoxic to pigs, as demonstrated by the results obtained in the TUNEL assay and by the lack of adverse reactions after vaccination.The follicular lymphoid hyperplasia observed in the mediastinal lymph nodes from vaccinated pigs remains to be investigated, since immunohistochemistry of mediastinal lymph nodes and qPCR from serum samples were negative to PCV2.

Virosomal influenza vaccine elicited humoral immune response in pigs
All serum samples from the non-vaccinated pigs (G1 group) were negative for IAV by HI and SVN assays.
Virus-neutralizing-antibodies for H1N2 were detected in 3.13% (1/32) of pigs (titer of 10) and for H3N2 in 100% (32/32) of pigs (titers of 10-160) 14 days after the first vaccine dose.On D28, virus-neutralizing antibodies were detected in all pigs from the G2 group as follows: Fig. 2 Hierarchical gating strategy applied to samples from the experiment.A Flow cytometry dot plot gate shows lymphocytes (left).B The forward scatter (FSC) A × FSC-H dispersion was used to gate singlet cells.C Viable cells were defined by a gate encompassing the 7-AAD negative cells in a FSC-A vs. 7-AAD dot-plot.D Flow cytometry histograms gated into CFSE + events displaying the CFSE low region (the cell proliferation region).Results of cell proliferation with H1N1, H1N2 and H3N2 challenges.E From the viable cell gate (7-AAD negative), the CD3 vs. CD79a negative dot-plot was used to define T lymphocytes, and F CD79a positive vs. CD3 negative dot-plot was used to define B lymphocytes.G SWC7 vs. CD79a from a gate on cell population, the frequencies of conventional B cells.H, K From a gate on each cell population (exemplified here by CD4 + T cells and CD8 + T cells, respectively), the frequencies of I effector CD4 + T cells (CD4 + CD25 + ), J central memory CD4 + T cells (CD4 + CD27 + ), L effector CD8 + T cells (CD8α + CD25 + ), M central memory CD8 + T cells (CD8α + CD27 + ), and N cytotoxic CD8 + T cells (CD8α + IFNγ + ) were determined in the respective dot-plots antibodies titers ranged from 10 to 1280 for H1N1 and H1N2 viruses, and ranged from 320 to 10,240 for H3N2 virus (Fig. 3B).On D90, one out of three pigs from G2 group had antibodies for H1N1 (titer of 320) and H1N2 (titer of 20), and all three pigs had antibodies for H3N2 (titers ranging from 40 to 320) (Additional file 2: Fig. S2B).

Discussion
The genetic and antigenic diversity of swIAVs in Brazil has increased since the emergence of H1N1pdm in 2009 [2,[40][41][42][43].Currently, distinct lineages of H1N1, H1N2 and H3N2 IAVs circulate in swine herds across several Brazilian states [2,3].In addition, the Brazilian swIAVs are genetically distinct from the viruses circulating in pigs in other countries [1][2][3].Consequently, the development of influenza vaccines that incorporate this genetic and antigenic diversity and can rapidly be updated, including locally adapted swIAVs, is relevant.In our study the vaccine strains represent the most frequently detected influenza virus lineages (H1-1A.3.3.2;H1-1B.2.4; H3-1990.5.2) in pigs in Brazil [1,3,4].As a significant antigenic variation exists between them, no Fig. 4 In vitro cell proliferation assay.Immune cells in the splenocyte proliferation assay stimulated with the vaccine viruses (H1N1, H1N2 and H3N2) were compared as a fold change from the vaccinated group (G2) to the non-vaccinated group (G1) on D28 post-vaccination.Data are shown for each pig per virus subtype, and the black lines represent the mean ± standard error.Asterisks (*) denote significant differences between non-vaccinated (G1) and vaccinated (G2) groups (P ≤ 0.05) Fig. 5 Cell profile in blood.Immune cells in the peripheral blood mononuclear cells (PBMCs), as a fold change from the vaccinated group (G2) over the non-vaccinated group (G1) on D28 post-vaccination.Data are shown for each pig and the black lines represent the mean ± standard error.Asterisks (*) denote significant differences between non-vaccinated (G1) and vaccinated (G2) groups (P ≤ 0.05) Fig. 6 Cell profile in bronchoalveolar lavage fluid (BALF).Immune cells in the BALF cells, as a fold change from the vaccinated group (G2) over the non-vaccinated group (G1) on D28 post-vaccination.Data are shown for each pig and the black lines represent the mean ± standard error.Asterisks (*) denote significant differences between non-vaccinated (G1) and vaccinated (G2) groups (P ≤ 0.05) cross-reactive antibody response in vaccinated pigs was expected.
The virosomal vaccines have been explored as an alternative to the conventional vaccine platforms, such as WIV vaccines, since they elicit both humoral and cellular immune responses while maintaining a favorable safety profile [19,20].Research on virosome vaccines, specifically in the context of targeting influenza or swine-related pathogens, remains relatively scarce.Virosome-based influenza vaccines have been reported for use in chickens [22] and humans [21].Additionally, they have been investigated in mice [44] and ferrets [45] as an experimental model.This vaccine technology has also been successfully applied for other viruses infecting both animals, such as Newcastle disease virus in poultry [46] and avian metapneumovirus in turkeys [47], or humans, as seen with SARS-CoV-2 [48] and hepatitis A [49].In this study, we demonstrated the immunogenicity in swine of a virosomal influenza vaccine containing the glycoproteins of the most prevalent virus subtypes circulating in Brazilian pig herds.Previously, the vaccine safety was evaluated in mice through the analysis of liver and kidney function, histopathology, and in vitro and in vivo cytotoxicity assays.It was demonstrated that the vaccine formulation was safe and non-cytotoxic for mice [24].
In swine, a robust humoral and cellular immune response was induced after two doses of the vaccine.HI and virus-neutralizing antibodies were detected against H1N1pdm, H1N2 and H3N2 viruses after booster immunization.In humans, the intramuscular administration of a virosomal influenza vaccine induced HI titers like those induced by WIV or subunit vaccines [50].In poultry vaccinated by a virosomal influenza vaccine, low HI titers were detected seven days post vaccination, showing a significant rise in HI titers 14 days post vaccination, that were maintained at day 28 post vaccination [22].Antibodies raised against HA are correlated with protection from clinical disease and are important to block the virus attachment and entry, preventing the virus infection [7].Virus-neutralizing antibodies also have a vital function in preventing the binding of IAV to sialic acid receptors, the fusion process, and the release of newly formed viral particles [51].A commercial virosomal vaccine for influenza in humans is available in several countries.This vaccine was considered highly effective by mimicking natural infection, and was immunogenic in both healthy and immunocompromised elderly, adults, and children [21].
The induction of influenza HA-specific antibody titers is affected by virus dose [52], and possibly by the virus subtype.Some studies have shown that pigs infected with H3N2 had higher antibody titers than those infected with H1N1 and H1N2 [53,54].Moreover, some viruses are more immunogenic than others, have dominant epitopes, and induce higher antibody titers [9].The process of virus glycoprotein incorporation into the virosome is random, and as this formulation contained the glycoproteins from three different viruses, it is possible that a different concentration of each glycoprotein has been incorporated.According to a previously conducted SDS-PAGE, the virosome formulation used here contained more HA from the H3N2 virus, followed by H1N2 and H1N1 viruses [24].This could explain high antibody titer for H3N2 virus compared to the antibody titers for H1N1 and H1N2 viruses by HI and SVN assays.Although antibody titers were not high for H1N1 and H1N2 subtypes, a robust cellular immune response was observed for these viruses.
In our study, immunization by a virosome-formulated influenza vaccine primed peripheral blood T and B cell subsets for early recall responses to H1N1pdm, H1N2 and H3N2 strains, including central memory CD4 + T cells (CD4 + CD27 + CD45RA − ), cytotoxic T lymphocytes (CD3e + CD8 + ), and B lymphocytes (CD79a + , CD79a + SWC7 + CD5 + and CD79a + SWC7 + CD5 − ).As indicated, the virosome vaccine was optimal for stimulation of cell-mediated immunity, cytotoxic T lymphocyte (CTL) activity and in particular, central memory CD4 + T cells.CTL is responsible for elimination of virus-infected cells and clearance of influenza virus infection [55].The central memory CD4 + T cells and B cells subsets (CD79a + , CD79a + SWC7 + CD5 + and CD79a + SWC7 + CD5 − ) are crucial for the development of memory B cells, antibody production, and antibody class switching [19].Only the B cell subset CD79a + CD5 + and macrophages differed in BALF samples from vaccinated pigs, revealing a differentiated cellular response in the mucosa even though the vaccine has been administered by the intramuscular route.
After the in vitro stimulation of splenocytes from the vaccinated group, evidence of cellular immune response was marked by high maturation of macrophages, and high proliferation of B lymphocyte subsets (conventional B cells), T lymphocyte subsets (effector and central memory CD4 + and effector and central memory CD8 + ), and CD8 + IFNγ + T cells (CTL).The T cell subsets with the most significant responses to viral stimulation were the central memory helper (CD3e + CD4 + CD27 + ) and cytotoxic (CD3e + CD8α + CD27 + ) T cells.These cell populations can mediate T helper function and express perforin to mediate cytolytic activity against virus infected cells [56,57].Taken together, these populations are required for enduring CD8 + T cell memory [58].In addition, the virosome influenza vaccine induced significant proliferation of CD3e + CD8α + IFNγ + T cells in swine, similar to previous studies in vaccinated mice [44], and pigs inoculated with a influenza vaccine [59].The whole inactivated virus (WIV) vaccine does not induce significant CTL activity, but the virosome is optimal for delivery of the antigen to the cytosol of antigen-presenting cells contributing to the clearance of influenza virus [55].Nonetheless, higher levels of B cells (CD79a + SWC7 + ) proliferated in the splenocytes from the vaccinated pigs.These findings suggest that long-term memory cells preferentially traffic in lymphoid tissues like the spleen [60].This could explain their low frequency in peripheral blood and BALF after immunization of pigs, in contrast to what was observed in the spleen following the in vitro stimulation.The characteristic recall responses of the two CD27 + populations, following the in vitro stimulation, became noticeably different, indicating that they were likely biased to a T helper 1 (Th1), T helper 2 (Th2) or T follicular helper (Tfh) phenotype.The commitment of memory CD4 + T cells to Th1 or Tfh lineages and memory CD8 + T cells to Th2 lineages provide cells that are poised for the lineagespecific expression of effector molecules upon preexposure to antigen.Upon recognizing virus-infected cells, CD8 + T cells readily respond by killing infected cells by producing antiviral cytokines and promoting the recruitment of immune cells [61].These findings have important implications for vaccine design, as adjuvanted-virosomal vaccines could promote a higher quantity and quality of memory Tfh cells, potentially allowing for enhanced humoral immunity after prime and boost vaccination.In this way, specific cells generate responses against IAV that can help eliminate the virus from the infected cells.Moreover, they provide protection against future infections with the presence of memory CD4 + and CD8 + T cells that can respond quickly to new virus infections.This broad proliferation of different immune cells, acting in association as observed here, locally, and systemically, is important for the mitigation of influenza infection in swine herds.
Another relevant T-cell response data, after the in vitro stimulation, was that both CD3 + CD4 + and CD3 + CD8 + cells induced an upregulation in CD25 expression.It indicates cell activation [12], which might be associated with a protection response against IAV as observed in clinical challenge studies [62,63].These findings indicated that immunization of pigs by an influenza-virosomal vaccine efficiently primes and activates CD8 + T cells that are important for the elimination of virus-infected cells, and reduction of virus shedding [64].It is not clear why the CD3e + CD4 + CD25 + and CD3e + CD8α + CD25 + responses of pigs vaccinated with H1N1, H1N2 and H3N2 remained lower than CDe3 + CD4 + CD27 + and CD3e + CD8α + CD27 + after in vitro stimulation.One possibility is that the IAV antigens activated in vitro the quiescent memory cells, and the analysis time after the activation of memory cells was insufficient to activate and detect the effector cells.
For the design of IAV vaccines, both humoral and cellular immunity should be considered, along with the capability to elicit strong and long-lasting immunity.In the production system in Brazil, the pig market age is around 130-140 days.Here, we assessed the immune response in pigs 90 days after the first vaccine dose, when pigs reached 130 days old.Despite the antibody titers were low at 90 days after vaccination, high proliferation of B lymphocytes, as well as central memory CD4 + and CD8 + T lymphocytes, and CTL (CD3e + CD8α + IFNγ + ) were demonstrated for the three vaccine viruses (H1N1pdm, H1N2 and H3N2) on day 90.Moreover, the specific memory cells detected 90 days' post-vaccination allow rapid clonal expansion in a future exposure to IAV.Although a low number of pigs (n = 3) was evaluated, our results are encouraging since the immune response for influenza after vaccination persisted during the production phase when pigs are more at risk of influenza infection, during the nursery and the finishing phase.Further studies are needed to assess the duration of immunity induced by the virosome vaccine in swine.

Conclusions
In conclusion, the virosomal-based influenza vaccine developed here showed a robust antibody-and cellmediated immune responses in pigs, with the potential to confer long-lasting immune memory to pigs (until the market age), and proved to be safe.It also allows for the rapid update of vaccine virus components.Additional studies are required to assess whether the vaccine also induced NA-specific antibodies and to ascertain whether this vaccine formulation is protective against the in vivo challenge by swIAV.For vaccine composition, selection of swIAV strains, by antigenic cartography, that better match with circulating viruses in swine in Brazil may contribute to the control of influenza in swine herds, reducing virus transmission among pigs, and the potential likelihood of generation of novel viruses.

Fig. 1
Fig. 1 Experimental design.A Timeline of G1 (non-vaccinated) and G2 (vaccinated) groups, highlighting the time points (D0, D14, D28, D60 and D90) of blood and nasal swab collection, administration of virosomal influenza vaccine (first and second dose) and necropsy of pigs.B Main laboratory assays carried out with biological samples collected from pigs of G1 and G2 groups.BALF = Bronchoalveolar lavage fluid; D = day; H&E = hematoxylin and eosin; HI = hemagglutination inhibition; PBMCs = peripheral blood mononuclear cells; SVN = serum virus neutralization.Illustration created with BioRender.com

Fig. 3
Fig. 3 Hemagglutination inhibition (HI) and serum virus neutralization (SVN) assays.Antibody titers by A HI and B SVN assays for H1N1, H1N2 and H3N2 subtypes of serum samples collected from pigs in the non-vaccinated (G1) and vaccinated (G2) groups on D14 and D28 post-vaccination.Data are shown for each pig per group and the black lines represent the geometric mean titers ± standard deviation