The method for constructing a plasmid in which Gcf A was encrypted was as described in a previously published report [9]. Features of Gcf A include a central conserved region and cysteine residues (Cys-172, Cys-176, Cys-182, and Cys-186) [9]. The plasmid was transformed into ClearColi BL21 (DE3) (Lucigen, Middleton, WI, USA), then spread on a Luria-Bertani (LB) agar plate containing ampicillin and cultured overnight at 37℃. The single colony of Gcf A-transformed ClearColi was taken, added to fresh LB (+ampicillin) media (13 mL), and placed in a 37℃ shaking incubator overnight. Fresh LB media (+ampicillin) (500 mL) was filled with 5 mL of ClearColi (+Gcf plasmid) that had been cultured the day before and grown up to 0.5 of OD 600 during shaking incubation. After the addition of 1 mM IPTG and shaking incubation at 37℃ for 4 more hours, cells were obtained by centrifugation at 8,750 rpm and 4℃ for 20 minutes. After cell resuspension using a binding buffer (20 mM KPO₄, 500 mM NaCl, 10 mM imidazole), the cells were disrupted with a sonicator. After adding a binding buffer to HisTrap HP (5 mL, GE Healthcare, Pittsburgh, PA, USA) containing Ni²⁺ Sepharose, the proteins that had not reacted to Ni²⁺ were removed. Elution buffer (20 mM KPO₄, 500 mM NaCl, 500 mM imidazole) was applied to obtain Gcf A reacting with Ni²⁺. Gcf A treated with 4× sodium dodecyl sulfate (SDS) (+DTT) was run on 15% polyacrylamide gel electrophoresis (PAGE) gel after staining with coma blue reagent, confirming the presence of a band of Gcf A. Using Sephacryl S-200 (GE Healthcare), pure Gcf A was separated, and the concentration was measured using a BCA assay kit (Thermo Scientific, Waltham, MA, USA). Gcf A treated with 4× SDS (+DTT, −DTT) was run on 15% PAGE gel and was confirmed as pure Gcf A via banding after staining with coma blue reagent. The Gcf A was stored at −70℃.

The method for production of a Bac M2 was as described in a previously published report [10]. Bac M2 was grown in Spodoptera frugiperda 9 (Sf-9) insect cells (Invitrogen, Carlsbad, CA, USA) at 28℃ using SF-900 II serum-free medium (Gibco BRL, Rockville, MD, USA). In order to obtain Bac M2, the final culture supernatants of infected Sf-9 cells were collected, treated with sucrose cushion (25% w/w sucrose in 5 mM NaCl, 10 mM EDTA in phosphate buffered saline [PBS]), and centrifuged at 24,000 rpm for 75 minutes at 4℃. The pellets were resuspended in 0.25 M sucrose and stored in LN₂. The viral titer was determined by plaque assay using Sf-9 cells according to the manufacturer's protocol (Invitrogen).

To obtain RSV A2, HEp-2 cells (ATCC, Manassas, VA, USA) were cultured in minimal essential medium (MEM; Welgene, Gyeongsan, Korea) containing 10% fetal bovine serum (FBS). RSV A2 was proliferated in MEM containing 3% heat-inactivated FBS, 2 mM glutamine, and 20 mM HEPES. After 4 days, syncytia formation was verified and the viruses were harvested by ultra-centrifugation (25,000 rpm for 1 hour at 4℃). The pellets were resuspended with a 25-gauge needle and serum-free MEM and then sonicated. Viral titers were performed by standard RSV plaque assay and the viruses were stored at −70℃.

Six-week-old female BALB/cAnNCrljOri mice (n=5/group) were purchased from Orient Bio Inc. (Seongnam, Korea) and kept in a specific pathogen free room for a week-long adaptation period. Three vaccines were used: Gcf A (20 µg/mouse) alone, Gcf A with added Bac control (live baculovirus; Gcf A/Bac control, 1×10⁷ plaque-forming unit [PFU]/mouse), and Bac M2 (baculovirus expressing M2 protein; Gcf A/Bac M2, 1×10⁷ PFU/mouse) in Gcf A (20 µg/mouse). Immunization was performed on days 0 and 14 through intramuscular injection of the right leg (volume, 100 µL). Viral challenge was performed 23 days after the second immunization with RSV at a dose of 1×10⁶ PFU/mouse. The mice were anesthetized with isoflurane (Ifran, Hana Pharm, Seongnam, Korea) inhalation and the RSV (volume, 50 µL) was administered via the left intranasal route. The condition of the mice was monitored through weight measurement for 5 days. The vaccines and RSV were diluted in PBS. All mice were monitored for the duration of the experiment and efforts were made to minimize pain. The research was carried out after approval by and in accordance with the guidelines of the Ewha Womans University Animal Experimental Ethics Committee (IACUC: 18-002).

Blood collection was performed two days before the first immunization and 21 days after the second immunization. Blood was drawn via the retro-orbital plexus using a capillary tube containing heparin and collected in an Eppendorf tube. The blood was centrifuged for 15 minutes at 5,600 rpm and 4℃, and the sera were separated and stored at −70℃. The Gcf A-specific antibody titer was measured by direct enzyme-linked immunosorbent assay (ELISA). The Gcf A was diluted in PBS, held overnight in aliquots of 100 µL/well (50 ng/well) in a 96-well plate, washed 3 times with PBS (+0.05% Tween-20), and blocked with PBS (+0.05% Tween-20) containing 1% skim milk for 2 hours at room temperature (RT). After 2 additional washes with PBS (+0.05% Tween-20), 2-fold serial diluted serum was added to the coating antigen and incubated for 2 hours at RT. After 4 washes with PBS (+0.05% Tween-20), horseradish peroxidase-conjugated affinity-purified rabbit anti-mouse total IgG secondary antibody (Zymed Laboratory, San Francisco, CA, USA) was applied. For the second immunization serum, rabbit anti-mouse IgG1, IgG2a secondary antibody, horseradish peroxidase (Invitrogen) was applied. Incubation at RT for 1 hour was followed by washing 6 times with PBS. After the addition of tetramethylbenzidine peroxidase substrate (1:1; TMB, KPL, Gaithersburg, MD, USA), the reaction was stopped with 1 M H₃PO₄, and the analysis was carried out at OD 450 nm using a Thermo ELISA plate reader. Cut-off values were calculated with serum-free wells.

On day 5 after the RSV challenge, the mice were sacrificed by CO₂ anesthesia. BAL fluids were collected with a catheter (BD Angiocath Plus, Becton-Dickinson, Franklin Lakes, NJ, USA) and 1 mL of PBS. Cells and supernatants were separated by centrifugation and leukocyte recruitment was measured in the BAL cell samples. Lung tissues of the immunized mice were collected, homogenized using a 70 µm cell strainer (SPL), and centrifuged to separate lung mononuclear cells and supernatants. Then, the proportions of immune cells were measured and the supernatants were used to measure viral replication in the lungs.

On day 5 after the RSV challenge, all mice were sacrificed by CO₂ euthanasia and single-cell suspensions of lung tissue were obtained using MEM (Welgene) and a 70 µm cell strainer (SPL). The cells and supernatants were separated by centrifugation at 1,600 rpm and 4℃ and the supernatants were incubated at 37℃ for 5 days with HEp-2 cells (ATCC). Viral titers were recorded as PFU/g of lung tissue, with a limit of detection of 100 PFU/g.

To measure the granulocyte population in the BAL specimens, the BAL cells were treated with rat anti-mouse CD16/CD32 blocking antibody (BD Biosciences, San Jose, CA, USA) and reacted at RT for 10 minutes. BAL cells were stained in the dark for 30 minutes at 4℃ with anti-Siglec-F antibody (E50-2440, BD Biosciences), anti-Gr-1 antibody (RB6-8C5, BioLegend, San Diego, CA, USA), anti-CD11c antibody (N418, BioLegend), and anti-CD45 antibody (30-F11, BioLegend). To analyze the RSV M2-tetramer binding activity of CD8⁺ T cells, lymphocytes isolated from lung tissue were treated with CD16/32 blocking antibody (BD Biosciences) and reacted at RT for 10 minutes. Staining was performed at 4℃ for 30 minutes in the dark using with anti-CD8 antibody (53-6.7, BioLegend), anti-CD44 antibody (IM7, BioLegend), and MHC class I H-2K^d tetramer complexes (H-2K^d/SYIGSINNI) presented with RSV M2_82–90 peptide. All stained cells were fixed with fluorescence-activated cell sorting (FACS) lysing solution (BD Biosciences) and analysis was performed with a BD FACSCalibur (BD Biosciences).

Th1 interferon γ (IFN-γ)⁺ cells and CD8⁺ IFN-γ⁺ T cells were analyzed by intracellular staining (ICS). First, lung lymphocytes were resuspended in IMDM (Welgene) containing 10% heat-inactivated FBS. The cells were stimulated by adding 50 µg/mL phorbol myristate acetate (1:1,000, Sigma-Aldrich, St. Louis, MO, USA), 500 µg/mL ionomycin (1:1,000, Sigma-Aldrich), and either 10 µM RSV M2 peptide (82–90: SYIGSINNI) or 10 µM RSV G peptide (183–195: WAICKRIPNKKPG) to IMDM (+10% FBS) (Welgene) containing β-mercaptoethanol (1:1,000, Sigma-Aldrich), recombinant human IL-2 (BioLegend), and Brefeldin A (1:1,000, Invitrogen), followed by incubation in the dark for 5 hours at 37℃. Stimulated cells were stained with anti-CD3 antibody (17A2, BioLegend), anti-CD44 antibody (IM7, BioLegend), and either anti-CD8 antibody (53-6.7, BioLegend) or anti-CD4 antibody (RM 4–5, BioLegend) at 4℃ for 30 minutes in the dark and fixed with FACS lysing solution (BD Biosciences). For the intracellular cytokine staining, first the fixed cells were treated with FACS buffer (0.5% FBS, 0.09% NaN₃ in PBS) containing 0.5% saponin (Sigma-Aldrich) and then they were permeabilized at RT for 15 minutes in the dark. The permeabilized cells were stained with anti-IFN-γ antibody (XMG 1.2, BioLegend) at RT for 30 minutes in the dark. The stained cells were analyzed using a BD FACSCalibur (BD Biosciences) and all flow cytometry data were analyzed using Flowjo software (TreeStar Inc., San Carlos, CA, USA).

ANOVA and Bonferroni multiple comparisons test were used to determine the statistical difference between each group. The p-value was considered to be less than 0.05 when there was more than 95% confidence interval. Statistically significant values are marked with an asterisk.

We conducted the study according to the experimental plan shown in Fig. 1A and investigated the immunogenicity of each vaccine by comparing the Gcf A-specific IgG titer in each group through ELISA using sera obtained before immunization and after two immunizations (Fig. 1A, B). We found that the mean titers of serum antibodies increased considerably after the second immunization, and confirmed that the Gcf A/Bac control and Gcf A/Bac M2 groups had higher titers than the Gcf A group (Fig. 1B). After the failure of the FI-RSV vaccine, it has generally been thought that the Th2 response must be avoided and inducing the Th1 response is important for preventing VED [14]. Therefore, before verifying the responses of Th1 IFN-γ⁺ cells, we compared the IgG subtype titers that can indicate the responses of Th1 or Th2 cells using the sera from each group. Doing so, we confirmed that the Gcf A/Bac control and Gcf A/Bac M2 groups (reciprocal log2 titers, 14 and 15, respectively) showed higher mean titers of serum IgG1 antibodies than the Gcf A group (reciprocal log2 titer, 11). Importantly, no IgG2a response was observed in the Gcf A group, while it was evident in the other two groups (reciprocal log2 titers, 14 and 13, respectively). These results indicated that total IgG response was upregulated by Bac control or Bac M2 when co-immunized with Gcf A, and importantly, significant IgG2a responses were induced only in Bac co-immunization groups (Fig. 1C).

IL-2 secretion from Th1 cells activates T cells and induces antibody production by promoting B cell maturation. In addition, Th1 cells secrete IFN-γ, which can be helpful for increasing CTL activity and suppressing virus replication [141516]. In order to identify the Th1 INF-γ⁺ cells specific for RSV G when the Bac M2 was used as an adjuvant, we prepared lung cells of immunized mice on day 5 after the RSV challenge, stimulated them with RSV G peptide (183–195: WAICKRIPNKKPG), and enumerated the percentages of Th1 cells secreting IFN-γ. Th1 INF-γ⁺ cells were significantly more strongly induced in the Gcf A/Bac M2 group (mean Th1 cells, 4.5%) compared with the Gcf A group (Fig. 2). Interestingly, the Gcf A/Bac control group also showed a significant Th1 INF-γ⁺ cell response compared with the response in the Gcf A group (Fig. 2B). Even, the percentage of cell population was slightly higher in comparison with the Gcf A/Bac M2 group, though the difference was not statistically significant (Fig. 2).

IFN-γ secreted by CD8⁺ T cells during RSV infection is important because it can block the pulmonary eosinophils; this has been observed in experiments with mice immunized with a recombinant vaccinia virus (vacv M2) capable of expressing M2 protein [17] . In order to investigate the CTL response specific to RSV M2, we confirmed the percentages of CD8⁺ IFN-γ⁺ T cells by stimulation with M2 peptide (82–90: SYIGSINNI) in lung cells obtained on day 5 after the RSV challenge. When Gcf A-Gcf A/Bac M2 and Gcf A/Bac control-Gcf A/Bac M2 groups were compared, the response of CD8⁺ IFN-γ⁺ T cells specific to M2 peptide was shown to be significantly much higher in the Gcf A/Bac M2 group (~6.2% vs. ~22.0%) (Fig. 3). There was no significant difference in the responses of CD8⁺ IFN-γ⁺ T cells between the Gcf A group (mean CD8⁺ T cells, 6.5%) and Gcf A/Bac control group (mean CD8⁺ T cells, 5.7%).

Following identification of CD8⁺ IFN-γ⁺ T-cell responses, lung cells obtained 5 days after RSV challenge were also checked to confirm whether they are also reactive to M2 peptide-loaded MHC class I H-2K^d tetramer (H-2K^d/SYIGSINNI). The percentage of CD8⁺ Tet⁺ cells in the Gcf A/Bac M2 group (~58.8%) appeared to be about 2- to 3-fold higher than in the other two groups, while the percentages of these cells in the Gcf A group (~23.8%) and Gcf A/Bac control group (~17.7%) in the same gated cell population were similar (Fig. 4A). The similar percentages of CD8⁺ Tet⁺ cells in the Gcf A and Gcf A/Bac control groups and the significantly higher mean percentage of CD8⁺ Tet⁺ cells in the Gcf A/Bac M2 group were confirmed (Fig. 4B).

The detection of VED in has been integral to RSV vaccine development since the failure of the FI-RSV vaccine, which was found to cause lung inflammation due to hypersecretion of type II cytokines. Therefore, our study included investigation of the extent of eosinophil, alveolar macrophage, and neutrophil response for each group in the BAL cells collected from the airway. First, we found that the numbers of eosinophils (CD11c⁻, Siglec-F⁺) in Gcf A and Gcf A/Bac M2 groups appeared to be lower, whereas the Gcf A/Bac control group had a slightly greater number (Fig. 5A, C). Next, we observed that the alveolar macrophage (CD11c⁺, Siglec-F⁺) responses followed the order Gcf A group>Gcf A/Bac control group>Gcf A/Bac M2 group. Statistical significance was confirmed for the Gcf A/Bac control and Gcf A/Bac M2 groups compared with the Gcf A group, and for the Gcf A/Bac M2 group compared with the Gcf A/Bac controls (Fig. 5A, D). Finally, we found that the neutrophil (CD11c⁺, Gr-1⁺) response was highest in the Gcf A/Bac M2 group compared to a level of approximately 4%–5% in the Gcf A and Gcf A/Bac control groups (Fig. 5B, E).

Immunized mice were challenged with RSV to determine if the immune responses produced by the vaccination were protective against virus infection. Five days after the infection challenge, we performed plaque assays on the lung supernatants. The Gcf A group was verified to have 537 PFU/g of the lung, while the Gcf A/Bac control and Gcf A/Bac M2 groups showed no sign of virus replication (results below the limit of detection), indicating that these vaccines provide protection against RSV infection (Fig. 6).

Body weight monitoring is important because the excessive inflammatory responses in the lung during RSV infection are directly linked to weight loss. So, weight loss was checked for 5 days after the RSV challenge as a marker of immunopathology. The Gcf A group showed decreased body weight until day 2 but body weight increased to 98% of baseline on day 5. In contrast, Gcf A/Bac control and Gcf A/Bac M2 groups showed decreased body weight until day 3 and then body weight recovered. However, on day 3, body weight appeared to be 81% in the Gcf A/Bac control group and 83% in the Gcf A/Bac M2 group, which differed significantly compared with the body weights of the Gcf A group (Fig. 7). Furthermore, clinical disease scores were also low for the Gcf A/Bac control and Gcf A/Bac M2 groups on days 2 to 3, but the mice later recovered from these symptoms (data not shown).