Development of a nano-emulsion and evaluation of its intradermal adjuvant function of Swine Influenza H3N2 Vaccine

Abstract

Background

Intradermal (ID) immunization is a potent vaccination strategy, but its potential is limited by the lack of adjuvants that can effectively enhance immune responses while minimizing local skin reactogenicity.

Objectives

To develop and evaluate a novel composite nano-emulsion adjuvant, designated VP, for enhancing the efficacy of intradermal vaccination against swine influenza virus (SIV).

Methods

A novel adjuvant VP, formulated with squalene, sea buckthorn oil, and the TLR3 agonist Poly(I: C), was designed. Its physicochemical properties (particle size, zeta potential, viscosity) were characterized. The immunogenicity and protective efficacy of the VP-adjuvanted inactivated H3N2 vaccine were systematically evaluated in mice, comparing intradermal (ID) and intramuscular (IM) routes. Assessments included humoral immunity (serum and bronchoalveolar lavage fluid (BALF) IgG/HAI titers, duration), cellular immunity (splenic T cell subsets, cytokine profiles), and protection against homologous viral challenge (viral load, lung pathology, weight loss). Local skin reactogenicity and inflammatory responses were also investigated.

Results

The VP adjuvant exhibited a potent immune-stimulatory effect, comparable to commercial adjuvants. The VP/H3N2 formulation formed a stable nano-emulsion with an average particle size of ~ 137 nm, a near-neutral surface charge, and low viscosity (~ 1.3cP), facilitating antigen transport to skin-draining lymph nodes. Immunization induced a robust, Th1-biased immune response characterized by high IgG2a/IgG1 ratios and elevated IFN-γ secretion. Notably, the VP-adjuvanted vaccine elicited the highest proportions of IFN-γ⁺ CD4⁺, IFN-γ⁺ CD8⁺and memory-type CD44⁺ CD4⁺ T cells among all experimental groups. ID administration was superior to IM, yielding significantly higher and more durable serum/BALF antibody titers and longer-lasting immunity. VP-adjuvanted vaccination provided protection against challenge, with minimal lung pathology and markedly reduced viral loads. Crucially, ID immunization with VP elicited only a moderate, transient local inflammatory response with rapid absorption and no significant adverse skin reactions. Mechanistic studies revealed that VP functioned by upregulating key chemokines (e.g., CX3CL1, CCL19) at the injection site, significantly enhancing antigen uptake by dendritic cells in vitro, and promoting targeted antigen delivery to draining lymph nodes in vivo.

Conclusions

This study successfully developed a novel composite nano-emulsion adjuvant, VP, which demonstrated high efficacy in intradermal SIV vaccination by potently enhancing both humoral and cellular immunity, providing durable protection, and maintaining an acceptable safety profile. The adjuvant mechanism involves enhanced local immune activation, promotion of Dendritic cells (DCs) phagocytosis, and improved lymphoid trafficking. These findings provide a theoretical and practical foundation for developing next-generation influenza vaccine adjuvants.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12985-026-03184-4.

Background

Swine influenza (SI), caused by influenza A viruses, is a significant respiratory disease characterized by clinical signs such as coughing, dyspnea, nasal discharge, sneezing, and fever, often accompanied by pneumonia-related lesions at slaughter [1, 2]. As enveloped viruses of the Orthomyxoviridae family with eight single-stranded RNA segments, influenza A viruses are classified into subtypes based on the surface glycoproteins hemagglutinin (HA) and neuraminidase (NA) [3, 4]. The predominant subtypes affecting swine are H1N1, H1N2, and H3N2, which pose a substantial economic burden to the global pig industry due to their high prevalence and synergistic morbidity with secondary infections [5]. H1N1 and H3N2 strains are widely prevalent in pig populations [6]. Notable events include the 2009 H1N1 swine flu pandemic and H3N2 spillover incidents during agricultural fairs in the US in 2011, 2012, and 2016 [5].

Current commercial swine influenza vaccines are predominantly inactivated, administered via the intramuscular (IM) route. While capable of inducing strain-specific IgG responses, these vaccines offer limited breadth of protection, exhibit poor cross-reactivity against drifted or reassorted strains, and fail to elicit robust cell-mediated immunity, which is crucial for viral clearance [7]. This shortcoming is largely attributed to inefficient engagement of the endogenous antigen-presenting pathway by vaccine antigens [8]. Consequently, there is a pressing need for next-generation vaccine platforms capable of inducing broader and more potent immune responses in swine.

The skin represents an attractive alternative immunization site, harboring a dense and diverse network of professional antigen-presenting cells (APCs), including Langerhans cells and dermal dendritic cells, which can orchestrate potent adaptive immunity [9]. Despite its immunological advantages, intradermal (ID) vaccination has not been widely adopted, primarily due to the scarcity of adjuvants specifically designed for safe and effective cutaneous use and the technical challenges associated with precise injection into thin skin tissue (< 2 mm) [9, 10]. An ideal cutaneous adjuvant should be composed of safe, metabolizable components and elicit a transient, localized immunostimulatory effect without causing persistent inflammation or tissue damage [9]. Many conventional adjuvants, such as Alum, oil emulsions (e.g., ISA 51, ISA 720), and certain TLR agonists, induce unacceptable local reactogenicity when administered intradermally, limiting their application to IM routes [11–13].

To address these limitations, we designed a novel intradermal adjuvant and delivery system termed VP, engineered to synergistically enhance cross-protective, cell-mediated immunity against influenza. Squalene was selected as the core lipid component for its proven safety and efficacy as an emulsion-based adjuvant. The squalene-based emulsion does not function by forming a classical antigen depot; rather, it induces a localized, transient inflammatory response at the injection site. This response recruits monocytes, macrophages, and dendritic cells, while stimulating the production of chemokines and pro-inflammatory cytokines to shape an immunologically favorable microenvironment. These recruited immune cells uptake the antigen and migrate via lymphatic vessels to draining lymph nodes, where they facilitate B and T cell activation. Emulsion droplets themselves do not migrate to the lymph nodes; instead, the adjuvant effect is mediated through the mobilization and activation of APCs at the injection site. Additionally, the emulsion may promote endogenous danger signal release from stressed cells, potentially contributing to innate immune pathway activation such as the NLRP3 inflammasome [14].Seabuckthorn oil was incorporated not merely as a carrier but as a potential immunomodulatory component based on its documented biological activities, including antioxidant and anti-inflammatory properties [15]. The squalene/seabuckthorn oil nanoemulsion core is intended to facilitate antigen uptake by antigen-presenting cells (APCs) and contribute to an immunocompetent microenvironment at the injection site. To further enhance innate immune activation, the formulation includes a pattern recognition receptor agonist—either the TLR3 ligand Poly(I: C) or the STING agonist cGAMP—which is known to induce type I interferons (IFN-α/β) and drive Th1/cytotoxic T lymphocyte (CTL)-polarized responses [10, 16, 17]. Inosine pranobex, an immunopotentiator that amplifies antigen-specific T-cell expansion and effector function, was also incorporated [18]. By combining these components, VP is designed as a multifunctional intradermal adjuvant that orchestrates coordinated immunity, aiming to achieve robust cross-protective cell-mediated responses against influenza with a favorable safety profile.

While intradermal immunization generally demonstrates superior immunogenicity over the IM route, inconsistent results in some studies highlight the critical importance of adjuvant design [19–21]. A key objective was to ensure that robust immunogenicity was achieved without compromising safety. Unlike many adjuvants that cause persistent local reactions [10, 24, 25], VP was formulated with natural, metabolizable components to elicit a transient and balanced local immune activation without observable tissue damage—a prerequisite for any adjuvant intended for cutaneous use [7, 25]. The immunogenicity and safety profile of the VP adjuvant were rigorously evaluated against a strategically selected panel of established commercial adjuvants. To benchmark its immunopotentiating strength, VP was directly compared with Montanide™ ISA 15 V—a classic water-in-oil emulsion renowned for its potent depot effect and capacity to elicit robust antibody responses. Gel02, a modern polymer-based aqueous adjuvant that operates via matrix retention and direct immune modulation. To specifically assess local reactogenicity, a critical parameter for intradermal formulations, VP’s inflammatory profile at the inoculation site was evaluated against Montanide™ ISA 201. This widely used water-in-oil-in-water emulsion is recognized in the field for achieving an optimal balance between immunogenicity and local tolerance. This comparative framework was designed to enable a comprehensive and stringent evaluation of VP’s efficacy, safety, and local tissue response.

In this study, we evaluated the immunogenicity and protective efficacy of an H3N2 influenza vaccine adjuvanted with VP, administered intradermally in a murine model. We hypothesized that VP would enhance both humoral and cell-mediated immunity—including cross-reactive responses—while maintaining an acceptable local safety profile. Our results confirm that VP elicits a robust, Th1-biased immune response with potent cytotoxic T-cell activity and partial cross-reactivity against heterologous strains, all accompanied by a self-limiting local inflammatory profile. These findings provide mechanistic and efficacy evidence supporting the further development of VP as a promising adjuvant platform for next-generation, broadly protective swine influenza vaccines.

Methods

Animals

Six-week-old female Balb/c mice were obtained from the Comparative Medicine Center of Yangzhou University. All animal experimental procedures were conducted in accordance with the guidelines approved by the Institutional Animal Care and Use Committee (IACUC) of JAAS (Approval No. IACUC-PA-2023-06-003). After one week of acclimatization, mice were immunized as scheduled. Euthanasia was carried out in accordance with the IACUC-approved protocol, using 100% CO₂ in a sealed chamber at a displacement rate of 50% VDR/min. To ensure irreversible euthanasia, mice were maintained in the CO₂ atmosphere for a minimum of 5 min [22].

Cells and virus

A/swine/heilongjiang/10/2007 (H3N2) (SIV-H3N2 HLJ strain), A/swine/ Shandong/12/2023 (H1N1) (SIV-H1N1 SD strain) was obtained from Sinovet (Jiangsu) Biopharmaceutical Co.,Ltd. The adsorption, purification, and inactivation of the SIV were performed as detailed by Magiri et al. [23]. Madin-Darby canine kidney (MDCK) cells were infected with the H3N2 HLJ strain of SIV at a multiplicity of infection (M.O.I.) of 0.01 for 1 h using minimal essential medium (MEM), which was subsequently enriched with 0.2% BSA and 0.01 µg/µL TPCK. The cultures were maintained in a humidified environment containing 5% CO2 at 37 °C until approximately 90% cytopathic effect (CPE) was noted. Afterward, the viral media were harvested and centrifuged at 216 g for 10 min, followed by ultracentrifugation at 112,700 g for 2.5 h. The viral pellet was resuspended in TSE buffer and placed on a sucrose gradient with 30% and 60% layers. Centrifugation was performed at 107,170 g for 2.5 h. Finally, the virus was inactivated using 0.1% formalin. HA concentration was quantified as previous study [24].

Preparation of different nano-emulsions

Poly(I: C) (HY-107202, MCE, USA), cGAMP (BD01471627, Bide, Shanghai, China), and inosine pranobex (HY-107801, MCE, USA) were individually dissolved in phosphate-buffered saline (PBS) to prepare aqueous solutions. The oil phase was prepared by mixing CVA (Squalene, B50723, Yuanye, Shanghai), CVB (Sea buckthorn oil, V64177, Yuanye), Caprylocaproyl Polyoxylglycerides (Labrasol^®, Guangzhou Tianrun Pharmaceutical Co., Ltd., CAS No. 85536-07-8), and Tween 80 (TWEEN™ 80-LQ-(RB), CAS No. 9005-65-6, Croda Chemicals (Shanghai) Co., Ltd.) at a volume ratio of 1:1:0.2:0.2, designated as CV20.

To screen for the most effective immune booster, each aqueous solution containing either Poly(I: C), cGAMP, or Inosine pranobex was combined with CV20 to produce three distinct nano-emulsions: VP (CV20 + Poly(I: C)), VG (CV20 + cGAMP), and VY (CV20 + Inosine pranobex). To further optimize the emulsion composition, Poly(I: C) was separately added to CVA or CVB, yielding formulations A + P (CVA+Poly(I: C)) and B + P (CVB+Poly(I: C)).

For each formulation, the CVA, CVB, or CV20 was combined with the respective aqueous phase containing the immune booster in a sterile glass vial. The final concentration of the immune booster in the complete emulsion was standardized at 4 µg/µL. The mixture was immediately emulsified by gentle magnetic stirring at 600 rpm for 30 min using a digital magnetic stirrer (IKA^® RCT basic, Germany) to facilitate spontaneous nano-emulsion formation. No additional high-shear homogenization or sonication steps were employed.

Characterization of emulsion

To evaluate the in vitro stability of the VP emulsion and monitor phase separation, samples were stored at 2–8 °C and analyzed after 15 min, 1 month, and 3 months. Physical stability was assessed by regular visual inspection for creaming or aggregation. The average particle size, polydispersity index (PDI), and zeta potential were determined at each time point using a Mastersizer 2000 system and a Zetasizer Nano ZS 90, respectively (both from Malvern Instruments, UK). The dynamic viscosity of the emulsion was measured at 25 °C with a digital viscometer (DV II + Pro, Brookfield Engineering Laboratories, USA) [25]. Antigen content of different vaccines were quantified in triplicate via Single Radial Immunodiffusion (SRID) assay as described previously [26].

Preparation and immunization of vaccines

All vaccines were administered at two-week intervals, with the second immunization occurring 14 days after the first. The details regarding the grouping were presented in Tables 1, 2, 3 and 4. To determine the optimal components of the compound emulsions, five mice in each group were subcutaneous immunized twice with a fixed dose of H3N2 antigen (10 µg HA), either alone or in combination with various adjuvants (20µL). This resulted in a total volume of 200µL administered via subcutaneous injection.The final concentration of immune boosters was maintained at 0.4 µg/µL. The sera from the mice were collected at 28 dpi for the detection of H3N2-specific IgG, IgG1, IgG2a, and HAI antibodies (Tables 1 and 2).

Table 1.

The identification of the most effective immune enhancer

Group	H3N2/HA (µg)	Adjuvant (µl)	Vaccines (µl)
H3N2	10	PBS (20 µl)		200
H3N2/VG	10	VG (20 µl)		200
H3N2/VY	10	VY (20 µl)		200
H3N2/VP	10	VP (20 µl)		200
H3N2/CV20	10	CV20 (20 µl)		200
H3N2/Gel02	10	Gel02 (20 µl)		200
H3N2/ISA15	10	ISA15 (20 µl)		200

Table 2.

The determination of the optimal components in compound emulsions

Group	H3N2/HA (µg)	Adjuvant (µl)	Vaccines (µl)
H3N2	10	PBS (20 µl)	200
H3N2/VP	10	VP (20 µl)	200
H3N2/A + P	10	CVA+poly(I: C) (20 µl)	200
H3N2/B + P	10	CVB+poly(I: C) (20 µl)	200
H3N2/P	10	poly(I: C) (20 µl)	200

Table 3.

The identification of the relationship between adjuvant doses and antigen doses

Group	H3N2/HA(µg)	VP (µl)	Vaccines (µl)
Uniform dose of VP	0.4	5	50
	0.7	5	50
	1.4	5	50
	2.8	5	50
	5.6	5	50
Uniform dose of antigen	2.8	0	50
	2.8	2	50
	2.8	5	50
	2.8	10	50

Table 4.

The comparison between ID and IM

Group	H3N2/HA(µg)	VP (µl)	Vaccines (µl)
H3N2(ID)	5.6	-	50
VP/H3N2(ID)	5.6	5	50
VP/H3N2(IM)	5.6	5	50
H1N1(ID)	5.6	-	50
VP/H1N1(ID)	5.6	5	50
VP/H1N1(IM)	5.6	5	50

To elucidate the relationship between adjuvant and antigen doses, five mice in each group were intradermally immunized twice with different doses of H3N2 antigen (5.6, 2.8, 1.4, 0.7, 0.4 µg HA ) with or without 5 µl of VP adjuvant, resulting in a total volume of 50µL; the final concentration of Poly(I: C) was maintained at 0.4 µg/µl. Alternatively, mice received combinations of 0, 2, 5, or 10 µl of VP adjuvant along with a fixed dose of H3N2 antigen (2.8 µg HA ) in a total volume of 50 µL; the final concentrations of Poly (I: C) were specifically set at (0, 0.2, 0.4, or 0.8 µg/µl) (Table 3).The sera of the mice were collected at 28dpi for the detection of H3N2-specific IgG and HAI antibodies.

To compare the differences between IM and ID, five mice in each group were administered IM or ID with H3N2 or H1N1 antigen (5.6 µg HA) with 5 µl of VP adjuvant, resulting in a total volume of 50µL (Table 4). Serum samples were collected from day 0 to day 240 dpi to detect HAI antibodies, while IgG antibodies were assessed at 70 dpi. Additionally, bronchoalveolar lavage fluid (BALF) was collected at 28 dpi to evaluate H3N2-specific IgG and HAI antibodies.

Hemagglutination inhibition titer (HAI)

To assess HAI titers, sera underwent a four-fold dilution in kaolin to remove non-specific inhibitors, followed by heating at 56 °C for 30 min to inactivate the enzyme. The sera were then serially diluted and combined with 4 HA units of influenza virus, incubating for 1 h at 37 °C before the addition of 1% red blood cells. HAI titers are defined as the highest dilution of serum that completely prevents hemagglutination [25].

Enzyme-linked immunosorbent assay (ELISA)

SIV-specific IgG, IgG1, IgG2a antibody titers were quantified using ELISA as previously described [35]. Briefly, 10 µg/ml of purified H3N2 or H1N1 influenza virus antigen was coated onto plates in NaHCO3 buffer ( pH 9.6 ) overnight. Following the removal of the coating solution, the plates were washed three times with 200µL of PBST (1 M PBS containing 0.1% Tween 20). The coated plates were then blocked with BSA (5%) for 1 h at 37℃. After washing three times with PBST, serially diluted serum samples or bronchoalveolar lavage fluid (BALF) were added to the wells. Unbound antibodies from mouse sera were eliminated by washing with 200µL of PBST. Subsequently, HRP-conjugated goat anti-mouse IgG, IgG1, IgG2a antibodies (1:20000, A90-116P,105P,107P, Bethyl, USA) were introduced to detect specific subtypes for 1 h at 37℃. The plates underwent an additional three washes with PBST before TMB substrate( P0211, Beyotime Biotech ) was added. Absorbance was measured at a wavelength of 370 nm. For each dilution of the test serum, calculate the P/N (Positive-to-Negative) ratio using the following formula: P/N Ratio=(OD of the test serum dilution)/NC mean. A serum dilution is considered positive if its P/N Ratio is ≥ 2.1.The antibody titer is defined as the reciprocal of the highest serum dilution that yields a P/N Ratio ≥ 2.1.

Assessment of cellular immunity

Splenocytes were isolated from mice 7 days post-second immunization. Spleens were processed through 40 μm cell strainers to obtain single-cell suspensions. Red blood cells were lysed using ACK lysis buffer (C3702, Beyotime Biotech) for 2 min. After washing three times with PBS, the cell concentration was adjusted to 1 × 10⁶ cells/mL.

For cytokine secretion analysis, splenocytes were re-stimulated with inactivated influenza antigens (H1N1 or H3N2, 1 µg/mL) in RPMI-1640 medium supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin for 24 h at 37 °C with 5% CO₂. Unstimulated cells served as negative controls. After culture, supernatants were collected by centrifugation (1,500 rpm, 5 min), and IFN-γ and IL-4 levels were measured using commercial ELISA kits (431107, 430807, BioLegend) according to the manufacturer’s instructions.

For cellular immune response assessment, splenocytes were then re-stimulated with inactivated influenza antigens (H1N1 or H3N2) at a final concentration of 1 µg/mL overnight. GolgiPlug (BD Pharmingen) was added to the culture and incubated for another 5 h [24]. Then the cells were first blocked with an anti-CD16/32 antibody (101301, BioLegend) to minimize nonspecific Fc receptor binding. Subsequently, cells were surface-stained with fluorochrome-conjugated antibodies against mouse CD3 (APC-Cy7, 57596, BD), CD4 (APC, 100516, BioLegend), CD8a (FITC, 553030, BD), and CD44 (PE, 103024, BioLegend). Following surface staining, cells were fixed and permeabilized using a commercial fixation/permeabilization kit (420801, 421002, BioLegend), and then stained intracellularly with an antibody against mouse IFN-γ (BV421, 505830, BioLegend). After a final wash, cells were resuspended in PBS and analyzed on a flow cytometer. The detailed gating strategy for the flow cytometry analysis is provided in the Supplementary Fig. 2). Data analysis was performed using CytExpert software [24, 27].

Virus challenge of immunized mice

To elucidate the protective efficacy of intradermal immunization with VP adjuvant against viral challenge, mice were randomly assigned to three groups, with ten mice in each group, two weeks following the second immunization, mice were intranasally exposed to 5.5 × 10⁵ TCID₅₀ of the H3N2 virus strain (SIV-H3N2 HLJ strain, 50µL). At 5dpc, three mice from each group were euthanized, and their lungs were harvested and homogenized in PBS. The resulting homogenates were subjected to centrifugation, and the supernatants from each sample were collected for TCID₅₀ detection [23]. The survival rate, clinical signs and bodyweight of the remaining mice were monitored for 12dpc.

Histological examination

Three mice of each group were euthanized at day 5dpc, the lungs of these mice were harvested and preserved in 4% polyoxymethylene (Beyotime Biotech, Nantong, China, P0099). Subsequently, the samples were embedded in paraffin wax, sliced into cross-sections, and stained using hematoxylin and eosin (H&E) [28].

Evaluation of cellular uptake of VP/H3N2 vaccine in murine bone marrow-derived Dendritic Cells by confocal microscopy

Following isolation and maturation, murine bone marrow-derived dendritic cells (BMDCs) were seeded on glass coverslips and incubated with either the VP/H3N2 vaccine formulation or the purified antigen alone for 24 h at 37 °C. To terminate endocytosis, cells were immediately placed on ice and washed twice with ice-cold PBS.To remove surface-bound antigen and specifically detect internalized antigen, cells were incubated with ice-cold acid wash buffer (0.2 M glycine-HCl, pH 2.5, 150 mM NaCl) for 5 min on ice. The acid wash was neutralized by adding ice-cold complete medium containing 10% FBS, followed by three washes with PBS (5 min each).Cells were then fixed with 4% paraformaldehyde for 15 min at room temperature. After fixation, cells were permeabilized with 0.1% Triton X-100 in PBS at room temperature, followed by three washes with PBS (5 min each). The non‑permeabilized control group was processed in parallel up to this point, omitting the subsequent permeabilization step.Cells were blocked with 3% BSA in PBS for 30 min, and then stained with the lipophilic dye DiO for 45 min at 37 °C to visualize the plasma membrane. Subsequently, cells were incubated with a primary antibody against H3N2 (1:1000) for 1 h at 37 °C, followed by a Cy3-conjugated secondary antibody (1:500; A0521, Beyotime Biotech) for 45 min at 37 °C. Nuclei were counterstained with DAPI (C1002, Beyotime Biotech) for 5 min.

Coverslips were mounted and examined using a confocal laser scanning microscope (LSM 800, Carl Zeiss). For each sample, five random fields were captured under identical acquisition settings. The fluorescence signal observed in acid-washed samples represents truly internalized antigen, as surface-bound antigen was stripped by the acid treatment. For quantitative analysis, the mean fluorescence intensity (MFI) of Cy3 within individual cells was measured using ImageJ software. The cell boundary was outlined based on DiO membrane staining, and at least 20 cells per field (five fields per sample) were analyzed. Background fluorescence was subtracted by measuring the signal in cell-free areas. Data from three independent experiments were analyzed and expressed as mean ± SD.

Luminex Assay Procedure

Three mice in each group were intradermally immunized with H3N2 antigen (2.8 µg HA ) with or without 5 µl of VP adjuvant for 24 h. A 4–6 mm punch biopsy was obtained from the target skin site under local anesthesia.The tissue was immediately placed in a pre-weighed tube containing chilled PBS and kept on ice.The tissue was homogenized using a mechanical homogenizer (bead mill) in cold lysis buffer compatible with the subsequent immunoassay. The homogenate was then centrifuged at 10,000×g for 10 min at 4 °C to remove cellular debris.The clarified supernatant was carefully collected, aliquoted, and stored at -80 °C until analysis. Total protein concentration of the supernatant was determined using a standardized assay (BCA, P0009, Beyotime Biotech ) for subsequent data normalization. The multiplex cytokine analysis was performed using a commercially available Luminex magnetic bead-based assay kit, with the experimental service provided by UniBio Biotechnology Co., Ltd (Shanghai, China).

Tissue immunofluorescence for analysis of antigen distribution post intradermal immunization

To evaluate the spatiotemporal distribution of the vaccine antigen following intradermal immunization, mice were administered the VP/H3N2 or H3N2 vaccine formulation via intradermal injection. At the predetermined 72-hour time point post-immunization, draining lymph nodes (dLNs) were harvested. Tissues were immediately fixed overnight at 4 °C in 4% paraformaldehyde, followed by cryopreservation in 30% sucrose solution. Subsequently, tissues were embedded in OCT compound and sectioned into 10 μm slices using a cryostat.

For immunofluorescence staining, tissue sections were permeabilized with 0.3% Triton X-100 (R21411, Yuanye) in PBS and blocked with 1% bovine serum albumin (BSA, BS114, Biosharp). To detect the H3N2 antigen, sections were incubated overnight at 4 °C with a primary antibody specific for SIV H3N2(1:1000). After thorough washing, sections were incubated for 1 h at room temperature with an FITC-conjugated secondary antibody (Beyotime Biotech, A0568); incubated at 37 °C for 45 min as specified. Cell nuclei were counterstained with DAPI. Finally, stained sections were imaged using a confocal laser scanning microscope (LSM 800, Carl Zeiss).

Skin Reactogenicity Assessment

To evaluate the local safety profile of VP-adjuvanted vaccines following intradermal (ID) administration, a skin reactogenicity study was performed. Mice received a single intradermal injection in the dorsal skin with either H3N2 antigen alone, VP/H3N2, VP alone, or 201/H3N2. To assess safety under exaggerated conditions, a supra-therapeutic dose was administered, corresponding to twice the standard immunization dose. Each group contained five mice. Immediately after injection, the injection sites were examined for local reactions, including the presence of round protrusions, redness, and swelling. The absorption status of the protrusions was monitored at 5 min, 6, and 24 h post-injection. For histopathological evaluation, skin tissue samples (including the injection site) were collected on days 1, 2, 5, and 8 post-immunization.

Statistical analyses

Statistical analyses were performed using GraphPad Prism for Windows (version 5.02). For comparisons between two groups, an unpaired two-tailed Student’s t-test was applied. For comparisons involving more than two groups, one-way analysis of variance (ANOVA) was conducted, followed by Tukey’s post hoc test for multiple comparisons. Data are presented as mean ± SEM, with statistical significance denoted as *P < 0.05 and **P < 0.01.

Results

Poly(I: C) demonstrated the most potent immune-boosting effect

Mice were immunized with H3N2 antigen, or H3N2 antigen combined with VP(CV20 + Poly(I: C)), VG(CV20 + cGAMP), VY (CV20 + Inosine pranobex), CV20 (CVA + CVB), ISA15 or Gel02. The antibody titers of mice in different groups were measured using Enzyme-linked immunosorbent assay (ELISA) and Hemagglutination inhibition titer (HAI) assays. Comparison to the H3N2 group, VP and Gel02 groups both significantly enhanced the HAI and IgG titers in the mice. Poly(I: C) combined with CV20, displayed the most potent immune-boosting effect, comparable to that of the commercial adjuvant Gel02(Fig. 1A-B). The IgG and HAI titers in the VP group were significantly higher than those in the B + P and P groups (P < 0.05). Although the titers in the VP group were also higher than those in the A + P group, this difference was not statistically significant (P > 0.05)(Fig. 1C-D).Therefore, the combination of Poly(I: C), CVA and CVB showed the most effective immune effect for H3N2 antigen.

Fig. 1.

Poly(I: C) demonstrated the most potent immune-boosting effect. A-B:Mice were immunized twice at a 2-week interval with inactivated H3N2 vaccine adjuvanted with VG, VP, VY, CV20, ISA15, or Gel02. Sera were collected two weeks after the final immunization for measurement of H3N2-specific IgG and hemagglutination inhibition (HAI) antibody levels; C-D: Mice were immunized twice at a 2-week interval with inactivated H3N2 vaccine adjuvanted with P, A + P, B + P, or VP. Sera were collected two weeks after the final immunization for measurement of H3N2-specific IgG and HAI antibody levels.All data are presented as mean ± SEM from three independent experiments, each performed in triplicate (n = 5 mice per group). The results were analysed for significance using one-way ANOVA using GraphPad Prism for Windows version 5.02 (*P < 0.05,**P < 0.01). Data are presented as mean ± SEM (n = 5)

The VP adjuvant tended to Th1-type immune responses

Mice were immunized with H3N2 antigen alone or formulated with one of the adjuvants: VP, CV20, ISA15, or Gel02. H3N2-specific IgG1 and IgG2a antibody titers were quantified by ELISA. As shown in Fig. 2A, the IgG2a titers in the VP and CV20 groups were significantly higher than the corresponding IgG1 titers (P < 0.05), whereas no significant difference between IgG1 and IgG2a was observed in the Gel02 and ISA15 groups.

Fig. 2.

The VP adjuvant tended to Th1-type immune responses. Mice were immunized twice at a 2-week interval with inactivated H3N2 vaccine adjuvanted with VP, CV20, ISA15, or Gel02.A: H3N2-specific IgG1, IgG2a antibody titers in the mice immunized with different H3N2 vaccines; B: The concentrations of IFN-γ and IL-4 in the splenocyte supernatants following stimulation with inactivated H3N2 antigen of the mice immunized with different H3N2 vaccines. All data are expressed as mean ± SEM of three independent experiments, each of which was performed in triplicate (n = 5).The results were analysed for significance using one-way ANOVA using GraphPad Prism for Windows version 5.02 (*P < 0.05,**P < 0.01)

To further characterize the immune response induced by VP, cytokine secretion was assessed in splenocyte supernatants after stimulation with inactivated H3N2 antigen. Levels of the Th1-type cytokine IFN-γ and the Th2-type cytokine IL-4 were measured. As shown in Fig. 2B, the VP/H3N2 and Gel02/H3N2 groups produced approximately 3-fold higher concentrations of IFN-γ than the H3N2-alone group (P < 0.05), while no significant differences in IFN-γ were observed among the remaining groups. In contrast, the Gel02/H3N2 and ISA15/H3N2 groups generated markedly elevated levels of IL-4, approximately 6-fold and 3-fold higher, respectively, compared to the H3N2-alone group. No significant differences in IL-4 were detected in the other groups.

Collectively, these results indicate that VP preferentially promotes a Th1-biased immune response, characterized by enhanced IFN-γ secretion and a higher IgG2a/IgG1 ratio. In comparison, Gel02 and ISA15 induced more balanced secretion of both IFN-γ and IL-4, suggesting a mixed Th1/Th2 response profile.

Characterization of the nano-emulsion VP

The physicochemical properties and stability of the VP-adjuvanted vaccine (VP/H3N2), including particle size, zeta potential, polydispersity index (PDI), viscosity, and pH, were systematically evaluated. The formulation presented a uniform milky white appearance (Fig. 3A). Dynamic light scattering analysis revealed that VP/H3N2 had an average particle size of approximately 136.9 nm (Fig. 3C), accompanied by a consistently moderate PDI value of around 0.28 throughout the stability study (Table 5), indicating an acceptable particle size distribution. Meanwhile, the viscosity of VP/H3N2 was approximately 1.2 cP, showing no statistically significant difference compared to the Gel02/H3N2 and ISA15/H3N2 formulations (P > 0.05) (Fig. 3B).

Fig. 3.

Characterization of the nano-emulsion VP. A:Appearance; B: Viscosity; C: size. All data are expressed as mean ± SEM of three independent experiments, each of which was performed in triplicate (n = 3) .The results were analysed for significance using one-way ANOVA using GraphPad Prism for Windows version 5.02 (*P < 0.05, **P < 0.01)

Table 5.

Zav, PdI, and ZP values of VP-adjuvanted vaccine at different times

Time	Zav(d.nm)	PdI	Mean Zeta Potential values(mv)	PH	HA concentration (µg/µl)
15 min	136.40 ± 0.80	0.29 ± 0.02	-0.04 ± 0.007	7.80 ± 0.12	0.048 ± 0.002
1month	131.90 ± 0.50	0.28 ± 0.03	-0.05 ± 0.001	7.70 ± 0.11	0.047 ± 0.006
3months	134.40 ± 0.40	0.27 ± 0.01	-0.05 ± 0.001	7.70 ± 0.10	0.049 ± 0.003

The surface charge characteristics were also investigated. While the H3N2 antigen alone exhibited a zeta potential of about − 14.9 mV, formulation with adjuvants markedly altered this parameter: VP/H3N2 showed a near-neutral charge of − 0.04 mV, whereas Gel02/H3N2 and ISA15/H3N2 registered values of − 16.2 mV and − 6.02 mV, respectively (Table 5). Furthermore, the pH of the VP/H3N2 formulation remained stable at around 7.8 over the entire three-month monitoring period (Table 5).Antigen integrity was assessed via single radial immunodiffusion (SRID), which confirmed that the hemagglutinin (HA) concentration remained stable within the range of 0.047–0.049 µg/µL during the three-month storage(Table 5). In summary, all key physicochemical parameters-particle size, PDI, zeta potential, viscosity, pH, and HA content-of the VP/H3N2 vaccine were effectively maintained over the three-month stability study, underscoring its robust formulation stability.

The VP-adjuvanted vaccine led to increased levels of serum IgG and HAI antibodies

Mice were immunized with varying doses of H3N2 antigen coupled with a consistent dose of VP adjuvant, or with a uniform dose of H3N2 antigen combined with varying doses of VP adjuvant. Serum samples were collected 28dpi. The VP-adjuvanted vaccine elicited significantly higher HAI and IgG titers in the groups receiving HA doses of 5.6 and 2.8 µg (Fig. 4A-B) compared to the non-adjuvanted H3N2 antigen groups. Both HAI and IgG titers exhibited a dose-dependent relationship concerning the H3N2 antigen doses; however, this was not observed for the VP adjuvant doses. Notably, the group receiving 5µL of VP demonstrated higher titers than all other groups(Fig. 4C-D).

Fig. 4.

The VP-adjuvanted vaccine resulted in elevated levels of serum IgG and HAI antibodies. Mice were immunized with different H3N2 vaccines.Serum samples were collected 28dpi. A-B:Levels of H3N2-specific IgG and HAI antibodies in mice immunized with varying doses of H3N2 antigen (containing 5.6, 2.8, 1.4, 0.7, and 0.4 µg HA) combined with 5 µl of VP adjuvant; C-D:Levels of H3N2-specific IgG and HAI antibodies in mice immunized with an H3N2 antigen (containing 2.8 µg HA) in conjunction with different volumes of VP adjuvant (0, 2, 5, and 10 µl). All data are expressed as mean ± SEM of three independent experiments, each of which was performed in triplicate (n = 5) .The results were analysed for significance using one-way ANOVA using GraphPad Prism for Windows version 5.02 (*P < 0.05,**P < 0.01)

The VP-adjuvanted vaccine administered via ID immunization elicited higher titers and longer duration compared to IM immunization

Mice were immunized with a VP-adjuvanted vaccine via intradermal (ID) or intramuscular (IM) administration. Serum or bronchoalveolar lavage fluid (BALF) samples were collected at various days post the first immunization, and neutralizing antibodies were analyzed using the hemagglutination inhibition (HAI) assay or the virus microneutralization assay (VNA) [29].

When the vaccine was formulated with the H3N2 antigen, the duration of immunity in the VP/H3N2 (ID) group was found to be at least 240 days, which was longer than that observed in the VP/H3N2 (IM) group (150 days) (Fig. 5A). Consistently, the H3N2-specific IgG antibody titers at 70 days post immunization (dpi) were significantly higher (P < 0.01) in the ID group compared to the IM group (Fig. 5B). In contrast, when serum was tested against the heterologous H1N1 antigen, cross-reactive IgG antibodies were also significantly higher in the ID group than in the IM group (Fig. 5C).In BALF collected at 28 dpi, H3N2-specific IgG and HAI antibody titers were significantly higher (P < 0.05) in the VP/H3N2 (ID) group than in the IM group (Fig. 5D–E). However, when BALF was tested against the heterologous H1N1 antigen, no significant difference in cross-reactive HAI antibody titers was observed between the ID and IM groups (Fig. 5F).

Fig. 5.

The immune responses induced by ID and IM immunization with the VP-adjuvanted vaccine were compared. Mice received the VP adjuvanted vaccine via either intradermal (ID) or intramuscular (IM) administration. Serum and BALF samples were collected at the indicated days post immunization. A:HAI antibody titers in serum from day 0-240 dpi. B:H3N2 specific IgG antibody titers in serum collected at 70 dpi; C:Cross reactive IgG antibody titers against heterologous H1N1 in serum at 70 dpi; D-E:H3N2 specific IgG antibody titers in BALF at 28 dpi; F:Cross reactive HAI antibody titers against H1N1 in BALF at 28 dpi. All data are presented as mean ± SEM from three independent experiments (n = 5 mice per group). Statistical significance was determined by t test (ID vs. IM) using GraphPad Prism for Windows version 5.02 (*P < 0.05, **P < 0.01)

In parallel, the immune responses induced by the VP-adjuvanted vaccine containing the H1N1 antigen were also evaluated following ID or IM administration (Supplementary Fig. 1). As shown in Supplementary Material-Figure 1 A, serum antibody analysis revealed that the ID route induced significantly higher levels of H1N1-specific IgG and HAI antibodies compared to the IM route. In contrast, only weak antibody responses against H3N2 were detected, with no significant difference between the two immunization routes. Similarly, in BALF (Supplementary Fig. 1B), the ID group exhibited markedly higher H1N1-specific IgG and HAI antibody titers than the IM group. However, antibody levels against H3N2 remained low, and again no significant difference was observed between the ID and IM groups.

The VP-adjuvanted vaccine induced T-cell response

Analysis of splenocytes from vaccinated mice demonstrated that VP not only augmented humoral immunity but also substantially enhanced cellular immune responses. The VP-adjuvanted H3N2 vaccine elicited a 12-fold and 7-fold increase in IFNγ⁺CD4⁺ T cell frequency compared to H3N2 antigen alone or VP alone, respectively (P < 0.01,Fig. 6A). IFN-γ is a key effector cytokine for Th1-type responses and is critical for viral clearance. Furthermore, this vaccine combination induced the highest proportions of IFNγ⁺CD8⁺ T cells (Fig. 6B) and memory-type CD44⁺CD4⁺ T cells(Fig. 6C) among all experimental groups. Notably, when splenocytes were stimulated with the heterologous H1N1 antigen, detectable cellular responses were also observed, albeit at lower magnitudes compared to H3N2 stimulation, suggesting that VP-adjuvanted vaccination elicits T cell responses with partial cross-reactivity (Fig. 6D-F).

Fig. 6.

Synergistic enhancement of cellular immunity by the VP adjuvant. Mice were immunized intradermally with PBS (control), H3N2 antigen alone, VP alone, or VP/H3N2. Splenocytes were isolated 7 days after the second immunization and restimulated with either homologous H3N2 antigen (A–C) or heterologous H1N1 antigen (D–F). The proportions of IFN-γ⁺ CD4⁺ T cells (A, D) and IFN-γ⁺ CD8⁺ T cells (B, E), as well as memory CD4⁺ T cells (CD44⁺CD4⁺, C, F), were determined by flow cytometry. All data are presented as mean ± SEM from three independent experiments (n = 5 mice per group). Statistical significance was determined by one-way ANOVA using GraphPad Prism for Windows version 5.02 (*P < 0.05, **P < 0.01 compared to the H3N2 antigen alone group)

Collectively, these findings indicate that VP acts as an adjuvant that primarily potentiates Th1-type immune responses and cytotoxic T cell activity, while also promoting the formation of immunological memory. This combination strategy significantly improves comprehensive protective immunity against H3N2 and may confer broader cross‑recognition of related viral strains at the T cell level.

The VP-adjuvanted vaccine conferred effective protection against homologous viral challenges

All mice survived the challenge without exhibiting any clinical signs. At 5 days post-challenge(dpc), lung tissues were collected and fixed for histopathological analysis. Both the H3N2 and PBS groups exhibited alveolar wall thickening, infiltration of lymphocytes and neutrophils, and localized bleeding in the lungs. These findings were significantly more severe than those observed in the VP/H3N2 group (Fig. 7A). In addition, the viral titer in the VP/H3N2 group was significantly lower than that observed in the other groups, exhibiting a 100-fold reduction compared to the control group (Fig. 7B). The VP/H3N2 group exhibited a slight body weight loss of 1% to 2% from 2 to 7dpc but recovered by 8dpc. In contrast, the H3N2 group experienced weight loss of 2.5% to 3.5%, also recovering by 8dpc. PBS immunized mice displayed significant and progressive weight loss, with the most severe decline of 5% occurring on 6dpc (Fig. 7C).

Fig. 7.

Virus challenge in immunized mice. Two weeks after the second immunization, mice were challenged intranasally with the H3N2 virus strain. Lung tissues were collected at 5 days post-challenge for further analysis;A: Histopathological examination of lung tissue sections; B: Viral load in lung homogenates assessed by TCID₅₀ assay;C:Changes in body weight following viral challenge.All data are presented as mean ± SEM from three independent experiments, each performed in triplicate (n = 5 mice per group). Statistical analysis was conducted using one-way ANOVA in GraphPad Prism for Windows version 5.02 (*P < 0.05, **P < 0.01)

The nano-emulsion VP evoked moderate inflammatory response

Given the superior adjuvant effect of VP in skin vaccination, we subsequently examined its local reactogenicity following intradermal injection. Mice received a single intradermal injection of H3N2 antigen alone, VP/H3N2, VP alone, or 201/H3N2 (Montanide™ ISA 201, a commercial adjuvant). In addition, to evaluate safety under exaggerated conditions, a supra-therapeutic dose of VP/H3N2 was administered.

The alterations in skin absorption for each group are illustrated in Fig. 8A. Round protrusions (wheals) were observed on the skin surface across all groups following intradermal immunization. In the H3N2, VP/H3N2, and VP groups, these protrusions were gradually absorbed within 6 h and completely disappeared by 24 h. In contrast, the protrusions in the 201/H3N2 group were absorbed more slowly, persisting for up to 24 h. No redness or swelling was noted on the skin in any group. Notably, even when administered at the supra-therapeutic dose (VP/H3N2 2×), the protrusions were absorbed within 6 h, and no local or systemic adverse reactions were observed, indicating that the VP-adjuvanted vaccine remains safe under exaggerated dose conditions.

Fig. 8.

The nano-emulsion VP evoked moderate inflammatory response. Mice were immunized intradermally with H3N2, VP/H3N2, 2×VP/H3N2,VP or 201/H3N2 vaccines, and skin samples were collected at specified time points post-immunization for analysis. A: The alterations in skin absorption for each group; B: The mRNA expression levels of IL-1β and TNF-α; C: Histological examination. All data are expressed as mean ± SEM of three independent experiments, each of which was performed in triplicate (n = 5).The results were analysed for significance using one-way ANOVA using GraphPad Prism for Windows version 5.02 (*P < 0.05,**P < 0.01)

The mRNA expression levels of IL-1β and TNF-α were assessed at various time points using RT-qPCR(Fig. 8B). On day 2, TNF-α expression was significantly upregulated in the VP/H3N2, VP, H3N2 and 201/H3N2 groups. While TNF-α expression decreased in the H3N2 group from days 2 to 5, it significantly increased in both the VP/H3N2 and VP groups. Furthermore, IL-1β expression in the VP/H3N2, VP, and H3N2 groups was elevated as early as day 1 and significantly rose from days 2 to 5, with the exception of the H3N2 group (Fig. 8B).

Histological examination was performed to assess pathological changes in the skin at the inoculation site. On Day 1, no significant inflammatory reactogenicity was observed in any group. By Day 2, inflammatory cell infiltration was evident in all intradermally injected groups, with the H3N2 and 201/H3N2 groups showing more pronounced changes. The inflammatory response resolved in the H3N2 group by Day 5 but persisted in the other groups (Fig. 8C). By Day 8, inflammation had subsided in all groups except the 201/H3N2 group.

These findings indicate that the VP adjuvant elicits a moderate, self-limiting local immune activation upon intradermal delivery, as evidenced by transient microscopic inflammation that resolved completely by day 8. Notably, this process was not accompanied by clinically observable adverse reactions, even at a supra-therapeutic dose. Moreover, the inflammatory response induced by VP resolved more rapidly than that triggered by the benchmark commercial adjuvant ISA 201, which persisted through day 8. Together, these results demonstrate that the VP adjuvant establishes a favorable activation–resolution kinetic profile in the skin, effectively stimulating the immune system while maintaining excellent local tolerance.

VP adjuvant enhances antigen uptake by dendritic cells in vitro and promotes targeted delivery to draining lymph nodes in vivo

This study systematically evaluated the effect of the VP adjuvant on the uptake and transport of H3N2 influenza antigen through in vitro and in vivo experiments. In vitro DCs phagocytosis assay. Compared to H3N2 antigen alone, co‑incubation with the VP adjuvant significantly enhanced the antigen uptake efficiency by DCs, as demonstrated by a higher fluorescence intensity (P < 0.01, Fig. 9A–B). Notably, in the non-permeabilized control group (acid-washed but not permeabilized), no specific fluorescence signal was detected (data not shown), confirming that the acid wash effectively removed surface-bound antigen and that the observed signal in permeabilized samples represents truly internalized antigen.

Fig. 9.

In vitro and in vivo analyses of antigen uptake by DCs and targeted delivery to draining lymph nodes. A–B: In vitro antigen uptake by DCs. DCs were incubated with H3N2 antigen alone or with H3N2 antigen plus VP adjuvant for 24 h. Cell membranes were stained with DiO (green), H3N2 antigen is shown in red (anti‑H3N2/Cy3), and nuclei were stained with DAPI (blue). Representative immunofluorescence images are shown in (A), and quantitative fluorescence intensity is presented in (B) (scale bar: 20 μm);C–D:In vivo antigen trafficking after intradermal immunization. Skin‑draining lymph nodes (sdLNs) were collected 72 hpi. Tissue sections were stained for H3N2 antigen (green) and with DAPI (blue). Representative images are shown in (C), and quantitative fluorescence intensity is presented in (D) (scale bar:200 μm).All data are presented as mean ± SEM from three independent experiments (n = 5 mice per group). Statistical significance was determined by one‑way ANOVA (**P < 0.01 compared to the H3N2 alone group)

In vivo intradermal immunization and trafficking analysis. We further tracked antigen transport following intradermal co-administration of VP adjuvant with H3N2 antigen. Immunofluorescence staining of skin‑draining lymph nodes (sdLNs) revealed that, at 72 h post‑immunization, the VP/H3N2 group exhibited a markedly stronger H3N2‑specific signal compared to the H3N2 alone group (Fig. 9C–D). This finding demonstrates that the VP adjuvant effectively enhances antigen delivery to draining lymph nodes in vivo.

Liquid chip microarray analysis revealed enhanced expression of chemokines in skin tissue following VP-adjuvanted H3N2 immunization

The expression of cytokines in skin tissue was analyzed using liquid chip microarray technology. Compared with the VP alone, H3N2 alone, and PBS‑treated control groups, the VP/H3N2 group showed significantly upregulated expression of several chemokines in the skin, including CX3CL1, CCL19, CXCL16, CCL3, CCL4, and IL‑16 (P < 0.01). Additionally, elevated levels of BCA‑1(CXCL13), CCL2, IL‑1β, CCL22, and CCL5 were observed, with the latter showing a notable increase, although these differences did not reach statistical significance (P > 0.05) (Fig. 10). These findings suggest that the VP‑adjuvanted H3N2 vaccine elicits a pronounced local chemokine response, which may contribute to enhanced leukocyte recruitment and immune activation at the vaccination site.

Fig. 10.

Chemokine expression profile in skin tissue following intradermal immunization analyzed by liquid chip microarray. Skin tissue samples were collected at the injection site 24 hpi. Cytokine expression was quantified using liquid chip microarray technology. The concentration of upregulated chemokines in the skin tissue of VP/H3N2 group compared to the VP, H3N2, and control groups (**P < 0.01). All data are expressed as mean ± SEM of three independent experiments, each of which was performed in triplicate (n = 3) .The results were analysed for significance using one-way ANOVA using GraphPad Prism for Windows version 5.02 (*P < 0.05, **P < 0.01)

In addition to the chemokines mentioned above, the expression levels of CCL20, CXCL5, TNF‑α, GM‑CSF, CCL1, IFN‑γ, CXCL10, IL‑2, IL‑6, IL‑4, IL‑10, CCL27, CCL17, CXCL12, CCL24, and CCL3 were not significantly altered in the VP/H3N2 group compared to the VP alone, H3N2 alone, or control groups (data not shown).

Discussion

Adjuvants enhance the host immune system’s ability to recognize administered antigens as foreign entities and mount an effective response [30]. This study aimed to develop an advanced intradermal adjuvant that balances potent immunogenicity with favorable local tolerance. We demonstrate that the VP adjuvant—formulated with sea buckthorn oil—met this objective by eliciting a robust and durable protective immune response against influenza H3N2. The response was characterized by enhanced systemic antibodies, pronounced Th1-type and cytotoxic T-cell immunity, and the establishment of long-term immunological memory. We propose that these outcomes stem from the adjuvant’s rationally designed physicochemical properties and, critically, from the bioactive components of sea buckthorn oil, which appear to orchestrate a self-limiting “activate‑then‑resolve” immune microenvironment within the skin.

Among the critical properties of vaccine carriers, surface charge is one of several factors that can influence interstitial mobility and, consequently, lymphatic drainage [31]. The dermal interstitium is composed of a negatively charged meshwork of collagen fibers and glycosaminoglycans, such as hyaluronic acid [32]. While positively charged carriers are susceptible to entrapment within this matrix [33], the relationship between zeta potential and in vivo trafficking is complex. Although near-neutral or negative surface charges have been suggested to facilitate passive interstitial transport [33], such a correlation does not inherently predict enhanced lymphatic drainage following intradermal administration, as active cellular transport predominates. However, it is well-established that active, cell-mediated transport—particularly the migration of antigen-presenting cells (APCs) like dendritic cells—plays a dominant role in the delivery of antigens to skin-draining lymph nodes following intradermal administration.We acknowledge that a zeta potential approaching zero typically warrants consideration of colloidal stability. However, our in vitro stability assays confirmed that the VP/H3N2 formulation remained well-dispersed and functionally intact, indicating that its stability was maintained under the tested conditions.

Particle size represents another fundamental parameter that governs the fate of vaccine carriers. Larger particles (> 500 nm) are predominantly cleared through phagocytosis by local macrophages [31]. Carriers that cannot directly access lymphatic capillaries must rely on uptake and subsequent shuttling by resident APCs [31]. Optimal lymph node drainage is generally achieved with particles ranging from 10 to 100 nm in size, as these dimensions enable efficient transport through the interstitial fluid and facilitate entry into initial lymphatic capillaries [31, 34, 35]. In the present study, the VP/H3N2 formulation exhibited a heterogeneous size distribution with an average diameter of approximately 136.9 nm and a predominant population within the 10–100 nm range. This size profile likely enabled a dual-mode delivery strategy: smaller particles within the optimal range could drain directly to the lymph nodes, while larger particles were effectively internalized by epidermal Langerhans cells and dermal dendritic cells for cell-mediated transport. This combination enabled sustained antigen presentation and robust activation of APCs, thereby contributing to the potent immune response observed.

This observed cytokine signature—marked by the selective upregulation of key chemokines without a concurrent rise in classic pro-inflammatory cytokines (TNF-α, IL-6)-suggested a coordinated and potentially tailored immune activation rather than a generalized inflammatory response. The specific chemokine profile is highly indicative of efficient antigen trafficking and T-cell priming within lymphoid tissues. CX3CL1 promotes the adhesion and migration of CX3CR1-expressing dendritic cells and T cells, whereas CCL19 plays a critical role in directing CCR7-positive naïve T cells and mature dendritic cells to the T-cell zones of lymph node [36, 37]. The upregulation of CXCL16 and CCL4 further enhances the recruitment of effector T cells as well as monocytes and macrophages [38, 39]. Notably, the absence of a significant increase in TNF-α and IL-6 suggests that the vaccine formulation did not induce a robust, potentially reactogenic, innate inflammatory response, which may confer a favorable safety profile. Critically, this restrained innate profile was followed by a rapid and complete resolution of local inflammation at the injection site by Day 8, as evidenced by histology. This ‘activate‑then‑resolve’ kinetic pattern underscores the self-limiting and precisely orchestrated nature of the immune response elicited by VP. Notably, even at elevated antigen and adjuvant doses, no significant systemic adverse reactions or exacerbated local reactogenicity were observed. This indicated that the VP-adjuvanted platform maintains a favorable safety margin, supporting its potential for flexible dosing strategies in future translational studies. This overall pattern was consistent with the characteristics of the VP/H3N2 vector, whose near-neutral surface charge and optimal particle size likely facilitate efficient drainage to and activation within lymph nodes, thereby promoting a chemokine environment specialized for immune cell recruitment and spatial organization, rather than inducing systemic inflammation.

To specifically evaluate T-helper cell polarization, we focused on IFN-γ and IL-4 as definitive surrogate markers for Th1 and Th2 immunity, respectively [40]. A dominant IFN-γ response indicates Th1-biased immunity, which supports cytotoxic T-cell activity, whereas a dominant IL-4 response reflects a Th2 bias that promotes antibody production [40]. The relative levels of these cytokines offer a clear, interpretable snapshot of the vaccine’s immunologic profile. Although other cytokines (e.g., IL-2, TNF-α, IL-5, IL-13) participate in immune responses, IFN-γ and IL-4 represent the most specific and defining markers for the primary Th1/Th2 dichotomy in this proof-of-concept stage [41].

Further analysis revealed that the VP adjuvant effectively induced a Th1-polarized response, characterized by dominant IFN-γ production, and promoted the activation and proliferation of cytotoxic T cells (CTL). This immunological profile has important implications for influenza vaccine efficacy. Th1-biased immunity enhances viral clearance and suppresses intracellular replication, while activated CTL directly lyse infected cells, thereby helping to control viral loads in tissues such as the lungs [42, 43]. Together, this cellular immune axis provides a critical layer of protection against antigenic drift or heterologous viral challenge, complementing humoral immunity. Consistent with this broad protective potential, VP-adjuvanted vaccination not only strengthened homologous immunity against H3N2 but also elicited partial cross-reactive cellular and humoral responses against the heterologous H1N1 subtype. Notably, in mice immunized with the H3N2 vaccine, cross‑reactive IgG antibodies against H1N1 were significantly higher in serum following ID immunization than in the IM group, whereas no such difference was observed in bronchoalveolar lavage fluid, suggesting that the cross-reactive humoral response is primarily systemic rather than mucosal. This cross-reactivity suggests that VP broadens immune recognition, likely by promoting dendritic-cell-mediated cross-presentation and facilitating the activation of T cells targeting conserved epitopes, while potentially steering B-cell responses toward more conserved regions of viral antigens. Interestingly, despite inducing potent local and systemic immunity, the VP-adjuvanted vaccine did not significantly enhance mucosal IgA responses, indicating that protection is mediated predominantly through systemic rather than mucosal mechanisms. Collectively, the Th1/CTL polarization elicited by VP, supported by its cross-reactive capacity, establishes a mechanistic basis for broader and more durable anti-influenza immunity [24].

Our findings position the VP adjuvant within the evolving landscape of intradermal vaccine strategies. Unlike molecular adjuvants such as cGAMP, which primarily activate a single innate sensor (STING) [10], or synthetic polymers like PCEP that rely on physicochemical retention [23], VP employs an emulsion platform analogous to MF59. This platform facilitates the generation of a local inflammatory milieu, characterized by robust cellular infiltration at the injection site. While classical depot formation is not a defining feature of MF59-like emulsions, the observed cellular infiltration likely increases the local availability of antigen-presenting cells (APCs) for subsequent antigen uptake. We note that direct assessment of APC recruitment by VP was not performed in this study; therefore, this interpretation remains inferential based on the observed cellular infiltrate. The formulation incorporates Poly(I: C) as an immunomodulator, which is known to activate TLR3 pathways and can induce type I interferons (IFN-α/β). IFN-α/β have been reported to play a role in bridging innate and adaptive immunity by enhancing APC licensing and cross-presentation, promoting T cell activation and Th1 polarization, and supporting the development of virus-specific CD8⁺ cytotoxic T lymphocyte (CTL) responses. However, the present study did not directly measure IFN-α/β production, and we did not obtain data demonstrating an IFN-α/β-dependent mechanism. The observed Th1-biased immune responses are consistent with the activation of pathways associated with this class of immunomodulators, but we acknowledge that causal links between Poly(I: C) engagement and the adaptive outcomes reported here remain to be experimentally established. Further investigation, including direct measurement of IFN-α/β and related innate signaling, would be required to delineate the precise mechanisms underlying VP-adjuvanted vaccination.A key differentiating feature of VP, however, is its kinetic profile: the inflammatory response resolved completely by Day 8(Fig. 8C), contrasting with the more persistent reactions typically associated with standard emulsion adjuvants. This suggests that VP achieves a more favorable balance between immunogenicity and local tolerability—a critical parameter for intradermal formulations.

The distinctive properties of sea-buckthorn oil likely underlie this balanced profile. Beyond serving as a biodegradable lipid core, sea-buckthorn oil is rich in bioactive compounds, including palmitoleic acid and tocopherols, which are known for their immunomodulatory and tissue-repair functions [15]. We hypothesize that these components do not act merely as inert carriers but actively orchestrate the local immune milieu. They may initially promote the acute recruitment and activation of innate cells, yet subsequently facilitate the timely resolution of inflammation. This “activate‑then‑ resolve”mechanism, distinct from purely pro-inflammatory adjuvants, could explain the observed combination of strong humoral/cellular responses with rapid inflammatory clearance.

Before proceeding to biologically variable large animal studies, the mouse model offers a well-controlled, reproducible, and high-throughput platform for screening multiple novel adjuvant formulations [29, 44]. This enables efficient assessment of fundamental vaccine characteristics, including the capacity to induce systemic and cellular immune responses—key immunological correlates of protection that are conserved across mammalian species.The well-defined murine immune system and the abundance of species-specific reagents enable detailed mechanistic investigations that are often not feasible in pigs at an early stage. Mice are the most commonly used species in influenza research due to their accessibility, cost-effectiveness, and the extensive availability of immunological tools [44]. For instance, our flow cytometric analysis of CD4⁺CD44⁺ memory T-cell induction and the characterization of IgG subclass profiles (IgG2a vs. IgG1) provide crucial insights into the quality of the vaccine-induced immune response—information essential for the rational selection of adjuvants. Strains such as C57BL/6 and Balb/c are well-established models for studying vaccine efficacy, pathogenesis, and antiviral development [29, 45]. Although some clinical signs (e.g., fever, dyspnea) are less readily observed in mice [44], the weight loss observed in our study—coupled with significant lung pathology indicative of viral replication—provided a stringent and objective basis for assessing vaccine protection. The positive immunogenicity and protection observed in mice establish a strong rationale and a necessary predicate for advancing to target species testing. Therefore, a direct and logical next step is to evaluate the selected VP-adjuvanted vaccine in swine.

Future translation of findings from murine models to target species requires careful consideration of interspecies differences in anatomy, physiology, and immunology. In particular, variations in skin thickness, lymphatic drainage, innate immune cell populations, and cytokine networks may significantly influence the safety and efficacy profile of intradermally administered, adjuvanted vaccines [46]. To bridge this translational gap, subsequent studies should employ a dedicated swine model to directly evaluate candidate vaccines under conditions better reflecting intended use. For dose translation, scaling the immunization dose based on body surface area (BSA) represents a scientifically established approach that correlates more closely with physiological processes such as metabolic rate and immune cell turnover than body weight alone. Accordingly, effective doses identified in mice can be converted for use in piglets following U.S. FDA guidelines for dose conversion by BSA normalization [46]. This provides a rational starting point for dose-ranging studies in swine, which may then be refined based on direct observations of local reactogenicity and systemic immune responses. Ultimately, such studies systematically assess vaccine safety, immunogenicity, and protective efficacy against both homologous and heterologous swine influenza virus challenges in pigs, thereby facilitating the translation of preclinical findings into practical applications.

Conclusion

In summary, we propose that the VP adjuvant functions through a sequential, spatially coordinated mechanism involving agonist‑induced and bioactive‑lipid‑ mediated modulation of the local cytokine/chemokine milieu, which promotes a Th1/ CTL‑polarized adaptive response while actively facilitating inflammatory resolution, thereby minimizing local tissue stress. This model positions VP as a novel example of an immunomodulatory emulsion. Future work will focus on delineating the specific contributions of key sea buckthorn constituents, benchmarking VP against a broader panel of licensed adjuvants, and evaluating its efficacy in challenge models to fully validate its translational potential.We acknowledge that the present study was conducted in a murine model, and further evaluation in the target species (swine) will be essential to confirm the safety and efficacy profile of VP‑adjuvanted vaccines prior to clinical application.

Abbreviations

SIV

Swine influenza virus

ID

Intradermal

IM

Intramuscular

MDCK

Madin Darby canine kidney cells

BALF

Bronchoalveolar lavage fluid

APCs

Antigen presenting cells

ELISA

Enzyme-linked immunosorbent assay

dpi

Days post immunization

dpc

Days post challenge

HAI

Hemagglutination inhibition titer

TLR

Toll-like receptor

STING

Stimulator of interferon genes

Author contributions

LY and WHY conceived and directed the study. RZY and WHY analyzed the data and wrote the manuscript. DBH conducted experiments. YWZ and MF contributed to manuscript writing and proofreading. All authors participated in reading and approving the final manuscript.

Funding

This work was financially supported by State Key Laboratory for Animal Disease Control and Prevention Foundation(SKLADCPKFKT202501).The Jiangsu Agriculture Science and Technology Innovation Fund (CX(22)2018).The Natural Science Foundation of Jiangsu Province (BK20221431, BK20241181). Jiangsu Association for Science and Technology Youth Science and Technology Talents Support Project (JSTJ-2024-114).

Data availability

The datasets used and analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

All experimental procedures involving animals adhered to the guidelines set by the Institutional Animal Care and Use Committee (IACUC) at JAAS (NO. IACUC-PA-2023-06-003).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.