All animal experiments were carried out in the Animal Biosafety Level 3 facility of Seoul National University Hospital and the Korea Zoonosis Research Institute. The studies received approval from the Institutional Animal Care and Use Committees (IACUC) of Seoul National University Hospital and Jeonbuk National University (SNUH IACUC No. 21-0288-S1A1 and JBNU IACUC No. 2020-0142), and were conducted in strict adherence to the recommended guidelines for the care and use of laboratory animals.

The genes encoding Gn and Gc from the KASJH strain (GenBank Accession No. KP663732, Supplementary Table S1) were cloned into a pcDNA3 vector (Addgene, Watertown, MA, United States). These vectors were transfected into HEK293F cells (Thermo Fisher Scientific, Waltham, MA, United States) using polyethylenimine. Following transfection, the cells were cultured in FreeStyle 293 expression medium (Gibco, Waltham, MA, United States) for 5 days. Recombinant proteins collected from the supernatants were purified using HisTrap HP histidine-tagged protein columns (GE Healthcare, Chicago, IL, United States) with the AKTA start^™ system (GE Healthcare), following the manufacturer’s instructions. The SFTSV NP recombinant protein purification process was conducted as previously outlined (Kang et al., 2020). Briefly, the pET28a(+) vector cloned with the gene encoding SFTSV NP was transfected into the Escherichia coli strain BL21 (DE3). The NP protein, overexpressed after 18 h of induction with isopropyl β-D-thiogalactoside at 16°C, was purified using HisTrap HP histidine-tagged protein columns (GE Healthcare) as per the manufacturer’s guidelines. All proteins were confirmed to contain less than 0.05 EU/mg of endotoxin. The identity and purity of the proteins, verified to be greater than 90%, were determined through western blotting and coomassie blue staining, respectively (Supplementary Figure S1). The purity of the purified antigens was assessed by measuring their optical intensity relative to the background intensity, utilizing ImageJ software (Schneider et al., 2012).

To assess the antibody titers against recombinant SFTSV proteins in the sera of immunized and non-immunized (mock) mice, immunoassay plates (96-well plates; Nunc, Rochester, NY, United) were prepared by coating each well with 100 ng of purified antigens and incubating overnight at 4°C. Following the antigen coating, the plates were washed with PBS containing 0.05% Tween 20 (PBST; Sigma-Aldrich, St. Louis, MO, United States) and subsequently blocked for 2 h at room temperature using PBST mixed with 5% skim milk (BD Difco, Sparks, MD, United States). Serum samples from each mouse were serially diluted in a 4-fold increment across 8–10 points starting at a 1:100 dilution, with 100 μL of these diluted samples then incubated for 1 h at room temperature in the plates. Sera from pre-immunized mice served as negative controls to establish the cut-off titer (mean O.D. ± 3 × S.D. at 1:100 diluents). For detection, horseradish peroxidase-conjugated goat anti-mouse IgG antibodies (diluted 1:10,000; Thermo Fisher Scientific) were employed as secondary antibodies. Plates were then washed with 0.05% PBST before adding 3,3′,5,5′-tetramethylbenzidine peroxidase substrate solution (BioLegend, San Diego, CA, United States) to develop the color for 7 min. The reaction was stopped with a 1 M H₃PO₄ solution, and the absorbance was measured at 450 nm using a microplate reader (TECAN, Mannedorf, Switzerland).

Spleens collected from mice were processed by grinding and filtering through a 70 μm cell strainer (BD Biosciences, San Jose, CA, United States), and the resulting splenocytes were suspended in RPMI 1640 media (Gibco) supplemented with 10% FBS (Gibco) and 1% penicillin/streptomycin (Gibco). Following lysis of red blood cells with Red Blood Cell lysing buffer Hybri-Max^™ (Sigma-Aldrich), the splenocytes were cultured in complete RPMI medium (enhanced with 10% FBS and 1% penicillin/streptomycin solution) at a density of 10 μg of purified NP, Gn, and Gc proteins per well for 18 h at 37°C with 5% CO₂ on a 96-well cell culture plate, with each experiment duplicated for individual mice. For intracellular IFN-γ detection, splenocytes (2 × 10⁶ cells/well) were treated with 1 μg of Golgi-plug (BD Bioscience) for the last 5 h of incubation. Cells were first stained with Zombie aqua fixable dye (BioLegend) and then blocked with ultra-block solution containing 10% rat sera, 10% hamster sera, and 10% mouse sera (Sigma-Aldrich). Following blocking, cells were incubated with fluorescent dye-conjugated antibodies against CD3 (145-2c11; BD Biosciences), CD4 (RM4-5; BD Biosciences), and CD8 (53-6.7; BioLegend) for 30 min at room temperature. After surface staining, cells were fixed and permeabilized using the Cytofix/Cytoperm kit (BD Biosciences), allowing for the intracellular staining with anti-IFN-γ antibody (XMG1.2; BD Pharmingen, Franklin Lakes, NJ, United States). The stained cells were analyzed using a CytoFLEX S flow cytometer (Beckman Coulter, Brea, CA, United States), and the flow cytometry data were processed using FlowJo software version 10.8 (Tree Star, Ashland, OR, United States). The gating strategy is presented in Supplementary Figure S2.

The SFTSV strain 2015-JJ01 (NCBI Genbank accession numbers MN329148-MN329150) was cultured in VeroE6 cells (ATCC No. CRL-1586), as described previously (Kang et al., 2020). The focus-forming units (FFU) of SFTSV were quantified using a focus forming assay with methylcellulose media. In brief, viral supernatants were filtered, serially diluted in 10-fold increments, and then applied to a monolayer of VeroE6 cells. After incubation for 1 h at 37°C, the supernatants were discarded, and the cells were overlaid with a medium consisting of DMEM supplemented with 2% FBS and 1% methylcellulose, followed by incubation at 37°C for 6 days. Subsequently, the cells were fixed and permeabilized using 100% methanol for 20 min at room temperature. SFTSV foci were identified with a rabbit anti-SFTS NP antibody (AbClon, Seoul, Republic of Korea) and a goat anti-rabbit IgG secondary antibody conjugated to horseradish peroxidase (Invitrogen, Waltham, MA, United States). The presence of viral foci was visualized through incubation with 3,3′-diaminobenzidine (DAB) Substrate (Merck, Darmstadt, Germany).

To assess the neutralizing efficacy of sera from immunized mice, a 50% focus reduction neutralization titer (FRNT50) assay was utilized. Sera were first inactivated by heating at 56°C for 30 min and then diluted four-fold across four points, ranging from 1:40 to 1:2560. Each serum dilution was mixed and incubated with an equal volume of SFTSV solution (containing 100 FFU) at 4°C for 1 h. This inoculated mixture was then applied to a monolayer of VeroE6 cells housed in a 24-well plate (SPL Life Sciences, Pocheon, Republic of Korea) and incubated for 2 h at 37°C. After this period, supernatants were discarded, and cells were cultured in an overlay medium (DMEM supplemented with 2% FBS, 1% penicillin/streptomycin, and 1% methylcellulose) at 37°C for 6 days. The process for visualizing viral foci is as described above. Percentage focus reduction was calculated as [(No. of plaques without antibodies) − (No. of plaques with antibodies)]/(No. of plaques without antibodies) × 100. All the experiments were duplicated and the FRNT₅₀ was calculated using nonlinear regression analysis (log[inhibitor] versus normalized response method) embedded in GraphPad Prism Software v8.0 (GraphPad Software; https://www.graphpad.com). All FRNT50 assays were conducted alongside negative control sera to establish the limit of detection under our experimental conditions.

Wild type C57BL/6 female mice aged between 6 to 8 weeks (Orient Bio Inc., Seongnam, Republic of Korea), as well as IFN α/β receptor-knockout (IFNAR-KO) mice on a C57BL/6 background (Min et al., 2018), were utilized for immunization and subsequent challenge experiments. These mice were kept in a specific pathogen-free environment at either the Seoul National University College of Medicine or the Korea Zoonosis Research Institute of Jeonbuk National University. For immunization, mice received subcutaneous injections of 10 μg of recombinant SFTSV NP-His, Gn-His, and Gc-His proteins, combined with aluminum hydroxy chloride (alum; Alhydrogel^® adjuvant 2%; InvivoGen, Hong Kong), administered three times at two-week intervals. Serum samples were collected from the mice weekly following each immunization and then every months for a year after the last immunization. Two weeks after receiving the final dose of vaccine, the mice were exposed to a challenge dose of 10³ FFU of SFTSV (strain 2015-JJ01). Blood samples were taken 4 days post-infection (dpi), and both survival rates and body weight changes were monitored and recorded up to 14 dpi.

Total RNA was isolated from the blood of SFTSV-infected mice utilizing TRIzol LS reagent (Life Technologies, Carlsbad, CA, United States), according to the manufacturer’s guidelines. Blood samples underwent lysis employing a TissueLyser II (Qiagen, Hilden, Germany), operating at a frequency of 30 Hz for a duration of 5 min, to facilitate the RNA extraction process. Subsequently, the harvested RNA was converted into cDNA using the HiSenScript RH (−) RT Premix kit (Intron, Seongnam, Republic of Korea). Quantification of the resultant cDNA was achieved through the application of TaqMan Universal Master Mix II (Applied Biosystems, Waltham, MA, United States). Quantitative reverse transcript-polymerase chain reaction (qRT-PCR) analysis was conducted using a BioRad CFX connect real-time PCR system (Bio-Rad, Hercules, CA, United States), adhering to the following thermal cycling conditions: an initial uracil-N-glycosylase (UNG) incubation step at 50°C for 2 min, followed by polymerase activation at 95°C for 10 min, denaturation at 95°C for 15 s, and then annealing/extension at 53°C for 1 min, across a total of 40 cycles. Fluorogenic signals were captured during the annealing/extension phase. The specific primer set and probe utilized for detecting the NP gene of SFTSV in the qRT-PCR were as follows: NP forward primer (5′-CCTTCAG GTCATGACAGCTGG-3′), NP reverse primer (5′-ACCAGGCTCTCAATCACTCCTGT-3′), and the fluorogenic detecting probe (5′-6FAM-AGCACATGTCCAAGTGGGAAGGCTCTG-BHQ1-3′). The quantification of viral RNA copy numbers was determined relative to a standard control, enabling precise measurement of the viral load in the samples.

For the purpose of hematological analysis, animals were euthanized and subjected to blood collection via the eye-bleeding technique. The collected blood samples were immediately placed in tubes pre-coated with 0.5 M ethylenediaminetetraacetic acid (EDTA) (Enzynomics, Daejeon, Republic of Korea) to prevent coagulation. To ensure the inactivation of any viruses present, a solution of 4% paraformaldehyde was added to the EDTA-anticoagulated whole blood samples at a 1:1 volume ratio prior to the assessment of platelet counts. The analysis of platelet counts was conducted using the BC-2800 Vet Auto Hematology Analyzer (Mindray Bio-Medical Electronics Co., Ltd., China).

Data analysis was conducted utilizing GraphPad Prism version 8.0 (GraphPad Software). For statistical evaluations, a two-tailed Mann–Whitney U test or a Kruskal–Wallis test followed by Dunn’s multiple comparisons test was employed to assess differences among various groups. The results are presented as the mean ± standard deviation (S.D.). The analysis of survival rates was carried out using the Mantel–Cox log rank test. A p-value of less than 0.05 was deemed to indicate statistical significance.

To evaluate the immune responses elicited by viral antigens, wild type C57BL/6 mice and IFNAR-KO mice were immunized subcutaneously with 10 μg of recombinant NP, Gn, or Gc proteins, each formulated with alum, at two-week intervals for a total of three doses (Figure 1A). Following immunization, IFNAR-KO mice exhibited a significant increase in antibody responses to the NP antigen after each of the three doses (Figure 1B). In contrast, antibody responses to Gn or Gc antigens in IFNAR-KO mice immunized with these proteins did not show an increase after the first dose but demonstrated significant elevation after the second and third doses (Figure 1B). Neutralizing antibody levels against SFTSV were measured in sera 1 week post-final immunization. The neutralizing antibody titers (FRNT50) in IFNAR-KO mice immunized with Gn were modestly higher than those in the control group (mean = 232.8 vs. 151.9), and titers were significantly higher in the Gc-immunized group (mean = 498.9) compared to the controls (Figure 1C). However, neutralizing antibody titers in mice immunized with NP (mean = 96.9) were comparable to those in the control group. These findings indicate that while the NP antigen is highly immunogenic for inducing an antibody response, it is barely effective at inducing neutralizing activity against SFTSV.

The analysis of antigen-specific T cell responses in IFNAR-KO mice immunized with individual recombinant proteins was conducted by evaluating IFN-γ secretion. Splenocytes from immunized mice were stimulated with the corresponding immunizing antigens (NP, Gn, or Gc) 1 week after the final immunization. The frequency of IFN-γ-producing CD4⁺ or CD8⁺ T cells in response to NP or Gc antigens showed a slight increase in the groups of mice immunized with the corresponding antigens compared to the mock-immunized mice; however, these increases were not statistically significant (Figures 1D,E). In addition, the frequencies of IFN-γ-producing CD4⁺ or CD8⁺ T cells in response to Gn did not show significant differences between the immunized mice and control group. This suggests that the vaccine formulation, comprising a viral subunit protein and alum hydroxide adjuvant, may not effectively generate cellular immunity against the viral antigens.

Investigating the potential cause of the weak cellular immunity in IFNAR-KO mice, we conducted parallel experiments in wild type C57BL/6 mice (Figures 1F–H). The patterns of humoral and cellular immune responses observed in the wild type mice, following immunization with the subunit antigens, were similar to those seen in IFNAR-KO mice. This confirmed the superior immunogenicity of the NP antigen for antibody responses and the significant induction of neutralizing activity by Gc immunization. The frequencies of NP-specific CD4⁺ or CD8⁺ T cell responses were substantially higher in NP-immunized mice compared to those in mock groups, yet these increases did not achieve statistical significance (Figure 1H). This consistently underscores the subunit antigen’s limited capacity to elicit cellular immunity, even in wild-type mice, when paired with the alum hydroxide adjuvant.

To assess the protective efficacy of vaccination with recombinant protein subunits, IFNAR-KO mice immunized with NP, Gn, or Gc proteins formulated with alum hydroxide received subcutaneous challenges with 10³ FFU of SFTSV. Following infection, all mice in the mock and Gc-immunized groups succumbed by days 6 and 5, respectively (Figure 2A). In contrast, four out of six mice (66.7%) vaccinated with NP and one out of six mice (16.7%) immunized with Gn survived the challenge. These surviving mice experienced a gradual loss of body weight until 6dpi before regaining their weight thereafter (Figure 2B). At 4 dpi, a substantial reduction in platelet counts was observed across all infected mice, with Gc-immunized mice showing a notably lower platelet count compared to uninfected control mice (Figure 2C). Although viral loads in blood of immunized mice generally decreased by 4 dpi, these reductions were not statistically significant when compared to mock-immunized controls (Figure 2D). Consequently, the strong antibody response and moderate T cell reactivity elicited against the NP antigen by the subunit vaccine may contribute to significant improvement in the survival rate of mice lethally challenged with SFTSV. However, such protective effects were not achieved through immunization with either the Gn or Gc antigens in our vaccine formulation.

To investigate the potential synergistic effects of combined vaccination with recombinant SFTSV proteins on the morbidity and mortality of mice challenged with a lethal SFTSV infection, we evaluated the humoral and cellular immune responses in IFNAR-KO mice immunized with various combinations of recombinant SFTSV proteins. Mice were vaccinated three times at two-week intervals with the specified mixtures of recombinant proteins, after which antibody and T-cell responses were analyzed in sera and spleens, respectively, 1 week post the third vaccination. Antibody responses against the specified antigens were significantly heightened in all immunized groups, with a notable increase in anti-NP IgG levels observed in some mice (three out of seven) immunized with NP/Gc, compared to the mock group (Figure 3A). However, no significant differences were found in NP, Gn, or Gc-specific antibodies among the different immunization groups. The FRNT50 values in sera from mice immunized with the NP/Gn/Gc combination showed the highest mean (678.4 ± 1268.3, n = 7) compared to the NP/Gn (143.7 ± 185.4), NP/Gc (143.5 ± 118.2), and mock (25.7 ± 11.1) groups, although there were individual variations within each group and the group differences were not statistically significant (Figure 3B). The frequency of IFN-γ-secreting CD4⁺ or CD8⁺ T cells in response to NP was substantially increased in all immunized groups compared to non-immunized controls (Figures 3C,D). Notably, the frequency of NP-specific IFN-γ-producing CD4⁺ cells in the NP/Gc group was significantly higher than in the mock group. However, the frequency of Gn-or Gc-specific IFN-γ-producing T cells in all immunized groups did not show a significant increase compared to non-immunized controls.

The protective efficacy of the combined recombinant proteins against lethal SFTSV infection was evaluated by monitoring survival rates and body weight changes (Figures 3E,F). Survival rates significantly improved in all immunized groups compared to non-immunized controls, where all mice succumbed to the virus by day 6 post-infection. The NP/Gc group exhibited the highest survival rate (85.7%), followed by the NP/Gn group (71.4%) and the NP/Gn/Gc group (57.1%). Furthermore, the NP/Gc group experienced relatively milder disease morbidity, as evidenced by less severe body weight loss, compared to other immunized and mock groups, although the differences were slight (Figure 3F).

To investigate the longevity of the humoral immune response in mice immunized with the recombinant SFTSV proteins, we analyzed antibody titers against NP, Gn, and Gc antigens, as well as neutralizing antibodies against SFTSV in immune sera from wild type C57BL/6 and IFNAR-KO mice over a period of 1 year. To confirm whether C57BL/6 mice exhibited similar patterns of antigen-specific antibody titers and neutralizing titers, they were immunized with each single subunit recombinant protein (NP, Gn, or Gc) in the same manner. Both immunized C57BL/6 and IFNAR-KO mice displayed parallel trends in antibody titers, peaking 2–3 months after the final immunization and gradually declined (Figures 4A,B). Notably, in both mouse strains, the NP group showed a higher antigen-specific antibody titer than the Gn and Gc groups. The FRNT₅₀ titers in sera from C57BL/6 mice immunized with Gn and Gc were significantly higher than those in the mock group, peaking at 1 to 3 months after the final immunization (Figure 4C). Similarly, IFNAR-KO mice groups showed a comparable pattern of neutralizing antibody levels, with the FRNT₅₀ titers in Gc immunized IFNAR-KO mice being significantly higher than those in mock group at one to 3 months after immunization (Figure 4D). Across the observation period, groups vaccinated with Gc consistently showed higher neutralizing titers than those vaccinated with Gn. Nevertheless, this neutralizing activity began to decline 3 months post-immunization and became insignificant between 6 to 12 months after immunization in both immunized C57BL/6 and IFNAR-KO mice. When we assessed long-term T cell responses in NP-immunized C57BL/6 mice, the frequency of NP-specific IFN-γ-producing CD4⁺ T cells persistently elevated up to 12 months after immunization, but the levels were not significantly higher than those of non-immunized controls (Supplementary Figure S3). In case of IFN-γ-positive CD8⁺ T cells specific to NP antigen, they peaked at 3 months after the final immunization and sharply declined to baseline levels at 6 to 12 months.

Finally, we evaluated the enduring protective capacity of the recombinant protein vaccines in IFNAR-KO mice by exposing them to a lethal challenge dose of 1 × 10³ FFU of SFTSV at 12 months after their last immunization. Although the group vaccinated with NP exhibited a slightly prolonged survival compared to the other vaccinated groups and non-immunized controls, all the immunized mice ultimately succumbed to the viral challenge (Figure 4E), indicating that the subunit vaccines confer limited durability in terms of long-term immunity against SFTS.