A BcfA-containing intranasal vaccine generated TH17 immunity and reduced B. bronchiseptica colonization and disease in mice Authors: Alexa R Aubrey, Komal K Patel, Jesse M Hall, Mohamed M Shamseldin, Myra Guo, Gianni B DeLucia, Siddhi Shah, Kara N Corps, Rajendar Deora, Purnima Dubey Published: 2026 ABSTRACT AbstractBordetella bronchiseptica (Bb) is a significant etiologic agent of respiratory disease in farm and companion animals. Current vaccines provide short-lived protection against disease, but do not elicit mucosal immunity that is critical for sustained protection. Here we tested a trivalent protein subunit vaccine composed of Bordetella antigens filamentous hemagglutinin (FHA) pertactin (Prn), and the dual antigen-adjuvant Bordetella colonization factor A (BcfA). Intranasal immunization of mice generated TH17 polarized systemic and tissue-resident CD4 + T cells and systemic and mucosal IgG and IgA antibodies. Immunized mice were protected against weight loss, bronchoconstriction and cough and displayed reduced lung inflammation. Following challenge, Bb was cleared from the lungs and reduced in the nose of immunized mice. This subunit vaccine has the potential to provide sustained protection against Bb infection and disease in companion and farm animals.Subject terms: Adaptive immunity, Infection, Infectious diseases, Vaccines Introduction Bordetella bronchiseptica (Bb) is a Gram-negative coccobacillus that causes canine infectious respiratory disease complex (CIRDC)1, or “kennel cough”. Bb also causes disease in cats, rabbits and in farm animals including pigs. Infection spreads through aerosolized particles released by coughing or by close contact with the infected host2 resulting in colonization of the respiratory mucosa. Bb has also been isolated from immunocompromised humans3, including individuals with HIV/AIDS4 or cystic fibrosis5–7. Cases of infection in immunocompromised humans have been associated with contact with pets that are colonized with Bb6,8 suggesting the potential for zoonotic transmission of this pathogen. Thus, effective vaccines against Bb should reduce disease severity while providing long-lasting immunity that limits or prevents transmission. Currently licensed live and live-attenuated veterinary Bb vaccines are administered intranasally (IN), orally or by intramuscular (IM) injection9,10. However, little information is available on the quantity and quality of the immune responses generated by these vaccines and is limited to analysis of serum antibody responses and bacterial colonization detected in nasal lavage11. Furthermore, nasal lavage vastly underrepresents both the bacterial burden and immune responses in the nasal cavity as these are predominantly sequestered within the nasal associated tissue12. Thus, characterization of Bb vaccine responses with higher resolution techniques is necessary. Previous studies showed that TH1/TH17 immune responses elicited by natural infection or vaccination with a formulation containing a TH1/TH17 polarizing adjuvant clear Bordetella pertussis (Bp) infections from the respiratory tract13–15. However, the immune phenotype elicited by Bb infection or by approved Bb vaccines is not well defined. Additionally, although these vaccines reduce disease severity after infection, they do not provide long-lasting immunity1 and their ability to generate mucosal immunity remains unclear. Bordetella colonization factor A (BcfA) is an outer membrane protein expressed by Bb16. BcfA has dual function as an antigen and an adjuvant17 that elicits TH1/TH17 polarized systemic and mucosal T-cell responses to a variety of protein antigens18,19. We showed previously that mice immunized with a trivalent vaccine composed of BcfA, and the Bordetella antigens filamentous hemagglutinin (FHA) and pertactin (Prn) (BcfA/FHA/Prn), delivered IM, elicited systemic TH1 immune responses and cleared Bb from the lungs and trachea17 while nasal colonization was unaffected. Disease symptoms including cough and respiratory distress were not evaluated. While IM administration primes the systemic immune response, an IN immunization generates immune responses at the site of infection18,19. In this study, we tested the effect of immunizing mice IN with BcfA/FHA/Prn. Immunization elicited TH17 polarized tissue-resident and systemic CD4 + T cells along with mucosal and systemic IgG and IgA antibodies. Immunized mice challenged with Bb were protected from weight loss, cough, and broncho-constriction. Bacterial burden was cleared from the lungs and reduced in the nose, and lung inflammation was attenuated. Together, these studies showed that an IN trivalent protein subunit vaccine can prevent respiratory disease and pathology mediated by Bb. This vaccine has the potential to provide sustained protection against Bb infection and disease in companion and farm animals decreasing the risk of associated zoonotic infections. Materials and methods Bacterial strains, media and growth conditions Wild-type Bb strain RB5020 was maintained on Bordet-Gengou (BG) agar (BD, cat. # 248200), containing 10% defibrinated sheep’s blood (Hemostat Labs) supplemented with 100 µg/mL streptomycin (Sigma, cat. # 3310-74-0). Bacterial plates were incubated for 36 h at 37 °C. For animal infections, liquid cultures from single colonies were grown at 37 °C in a roller drum to an optical density of 1 (OD600) in Stainer-Scholte medium with 100 µg/ml streptomycin. Mice All experiments were reviewed and approved by The Ohio State University Institutional Animal Care and Use Committee (2017A00000090-R2). C57BL/6 J mice were obtained from Jackson Labs and bred in our vivarium. Male and female mice, 6-10 weeks old, were used for immunizations. Reagents BcfA was produced and purified as described previously with endotoxin removal16. The endotoxin level was <5 endotoxin units/mg protein. FHA and Prn derived from Bp were produced by Dr. Jennifer Maynard (The University of Texas at Austin) and List Biologicals Inc., respectively. Fetal bovine serum (FBS) (cat. # F2442) was from Sigma-Aldrich (St. Louis, MO). ELISA kits were from eBioscience (Thermo Fisher Scientific). Flow cytometry antibodies were from eBioscience, BD Biosciences, or R&D Systems as indicated below. Immunizations Mice were anesthetized with inhaled 2.5% isoflurane/O2, followed by IN immunization with a mixture of 3 µg BcfA, 1 µg FHA, and 1 µg Prn in sterile 1x PBS on day 0 and 28. The vaccine was divided between both nares with a total volume of 25 µl/nare. Intranasal challenge Mice were anesthetized with inhaled 2.5% isoflurane/O2 and IN challenged with 5 × 105 CFUs of RB50 in 50 μL of sterile PBS, divided between both nares. The bacterial challenge dose was confirmed by plating 100 μL of the inoculum. Colony enumeration Mice were euthanized on days 0, 4, 7, 14, and 28 post-bacterial challenge. Lung and nasal tissue were collected to determine bacterial burden as previously demonstrated21. Tissues were either enzymatically digested for flow cytometry analysis or mechanically homogenized when assessing bacterial burden alone. Tissue collections were serially diluted and plated on BG plates containing 10% defibrinated sheep’s blood and supplemented with 100 μg/mL streptomycin. CFUs were counted after 36 h of incubation at 37 °C in a humidified incubator. Splenocyte stimulation and cytokine ELISA assays The spleens were dissociated by pushing the spleen through a 40 µm strainer (Fisher, cat. # 22363547), and red blood cells were lysed using ACK buffer (Thermo Fisher Scientific, cat. # A1049201). Single-cell suspension was plated at 2.5 × 106 cells/well of complete T-cell medium (RPMI (Thermo Fisher Scientific, cat. # 11875093), 10% FBS (Sigma-Aldrich, cat. # F2442), 10 µg/ml gentamycin (Thermo Fisher Scientific, cat. # 15710064), 5 × 10-5M 2-mercaptoethanol (Sigma-Aldrich, cat. # M3148)) and stimulated with 1 µg/mL BcfA, FHA, Prn or medium alone as a negative control. Supernatants were collected on day 3 post-stimulation. Sandwich ELISA was used to quantify IFNγ (R&D Systems, cat. # DY485-05), IL-5 (Invitrogen, cat. # 88-7054-88), and IL-17 (Invitrogen, cat. # 88-7371-88) per the manufacturer’s instructions. Plates were read at absorbance 450 nm (A450) on a SpectraMax i3x plate reader (Molecular Devices, Inc.), and the concentration was calculated based on the standard curve. Antibody ELISAs High-binding 96-well plates (Corning, cat. # 9018) were coated with 0.5 μg/mL BcfA, FHA, or Prn at 4°C overnight. Plates were washed three times with PBST (PBS with Tween 20 (Sigma-Aldrich, cat. # P2287)), blocked for 2 h at room temperature with ELISA diluent (1% BSA in PBS), and then washed twice with PBST. Dilutions of lung and nasal supernatants were added to the coated plates and incubated at room temperature for 2 h. Plates were then washed with PBST and incubated with secondary anti-mouse IgG HRP (Abcam, cat. # abc6789) or anti-mouse IgA HRP (Invitrogen cat. # 292318) antibodies for 1 h at room temperature. To detect color change, 50 μg of tetramethylbenzidine (Biolegend, cat. # 421501) was added to plates for 5 to 15 min, then H2SO4 was added to stop the reaction. Plates were read at 450 nm (A450) on a SpectraMax i3x plate reader. Intravascular staining to discriminate between circulating and resident immune cells Anti-CD45-PE (BD Biosciences, clone 30-F11) (3 µg in 50 µl of sterile PBS) was injected retro-orbitally 10 min prior to sacrifice to label circulating lymphocytes22,23. Peripheral blood was collected at the time of sacrifice and checked by flow cytometry to confirm that >90% of circulating lymphocytes were CD45-PE+ (data not shown). Tissue dissociation and flow cytometry Lungs and nasal septum were isolated and enzymatically digested into a single-cell suspension (Miltenyi Biotec, cat. # 130-095-927), followed by filtration through a 40-µm filter and RBC lysis with ACK lysis buffer. Cells were resuspended in T-cell media (RPMI, 10% FBS, 10 µg/ml gentamycin, 5 × 10−5M 2-mercaptoethanol) and stimulated with PMA (500 ng/ml)/ionomycin (500 ng/ml) plus protein transport inhibitor (eBioscience, cat. # 00-4975-93) or only T-cell media as a negative control for 5–6 h at 37 °C in a 5% CO2 incubator. Following stimulation, cells were washed with cold PBS prior to staining with Live/Dead Zombie Nir Fixable Viability dye (BioLegend, cat. # 423105) for 30 minutes at 4 °C. Cells were then washed twice with FACS Buffer (PBS + 1% FBS) and resuspended in Fc block (CD16/CD32 monoclonal antibody, clone 93, eBioscience, cat. # 14-0161-86) at 4 °C for 5 minutes. After incubation, cells were stained with the following antibody cocktail for 20 minutes at 4 °C: CD3 V450 (clone 17A2; BD Biosciences), CD4 BV750 (clone H129.19; BD Biosciences), CD8a Alexa Fluor-700 (clone 53-6.7; BioLegend), CD44 PerCP-Cy5.5 (clone IM7; BD Biosciences), CD62L BV605 (clone MEL-14; BD Biosciences), CD69 BV711 (clone HI.2F3; BD Biosciences) and CD103 PE-CF594 (clone M290; BD Biosciences). Cells were washed twice in FACS buffer and then resuspended in intracellular fixation buffer (eBioscience, cat. # 88-8824-00) for 30 min or O/N at 4 °C. Cells were washed and stained for 30 min at 4 °C with IFNγ FITC (clone XMG1.2; 29 eBioscience), IL17 PE-Cy7 (clone eBio17B7; eBioscience), and IL5 APC (clone TRFK5; BD Biosciences). Fluorescence minus one (FMO) of each antibody was used as a negative control. Cells were washed with a permeabilization buffer and resuspended in FACS buffer. Samples were collected on a Cytek Aurora 4-laser flow cytometer (Cytek Biosciences). Analysis was performed using FlowJo software version 10.10.0. To quantify neutrophils and macrophages, lung and nasal tissues were collected on days 4, 7, and 14 post-challenge and dissociated as described above. Single-cell suspensions were washed with cold PBS and then stained with Live/Dead Fixable Aqua (Thermo Fisher Scientific, cat. # L34966) for 30 minutes at 4 °C. After two washes in FACS buffer, cells were resuspended in CD16/CD32 monoclonal antibody at 4 °C for 5 min. Cells were then stained with the following antibody cocktail for 20 minutes at 4 °C: Ly6G BV711 (clone 1AB; BD Horizon), CD11b APC (clone M1/70; Invitrogen), CD11c APC-Cy7 (clone HL3; BD Pharmingen), Ly6C FITC (clone AL-21; BDPharmingen), SiglecF PerCP-Cy 5.5 (clone E50-2440; BD Pharmingen), CD3 V450 (clone 17A2; BD Horizon), CD4 BV750 (clone H129.19; BD OptiBuild), and F4/80 PE (clone BM8.1; TONBO Biosciences). Finally, cells were resuspended in FACS Buffer and collected on a Cytek Aurora flow cytometer. Data was analyzed using FlowJo. Histology The left lung lobe was harvested from Bb-challenged mice and fixed in 10% neutral buffered formalin for at least 24 h. Tissues were processed and embedded in paraffin. Five-micron sections (3 per tissue) were stained with hematoxylin and eosin by the Comparative Pathology and Digital Imaging Shared Resource at The Ohio State University. A board-certified veterinary pathologist (K.N.C.) was blinded to experimental groups, and sections were scored semi-quantitatively on a scale ranging from 0 to 5 for the degree of cellularity and consolidation, thickness of alveolar walls, degeneration and necrosis, edema, hemorrhage, infiltrating alveolar/interstitial polymorphonuclear cells (PMNs), intrabronchial PMNs, perivascular and peribronchial lymphocytes and plasma cells, and alveolar macrophages. The inflammation score was calculated by adding the scores in each category. Whole body plethysmography Respiratory symptoms following infection were measured using a Buxco® FinePointeTM Whole Body Plethysmography (WBP) (Data Sciences International, Inc, St; Paul, MN). Briefly, mice were acclimated to the chambers for 15 min, two days prior to challenge. Each day post challenge, the mice were acclimated for 5 min, after which their respiratory profiles were recorded for 10 minutes. Respiratory parameters including PenH and cough count, were derived from box flow and airflow of the subjects. FinePointe® software was used for data collection and analysis. Statistical analysis All data were analyzed using GraphPad Prism v10. The nonparametric Mann Whitney U tests were used to determine significance between two groups at each timepoint for all figures. P < 0.05 was considered significant. Outliers were identified and removed by ROUT test with a Q = 1% for all figures. Results Intranasal prime-boost with BcfA/FHA/Prn elicited TH1/TH17-polarized T resident memory cells (TRM) Cellular and humoral immune responses at the site of infection are correlated with pathogen clearance from the mucosal site and sustained immunity10. Current Bb vaccines do not elicit mucosal immunity, and the cellular immune response generated by these vaccines is not well described1,10,11. CD4 + T cells are required for protection against Bordetellae14,24. To generate mucosal responses, we immunized mice IN with a vaccine containing three Bordetella proteins, BcfA, FHA, and Prn. BcfA served as the adjuvant and as an antigen. First, we evaluated CD4 + TRM cells in the lungs and nose of unimmunized (naïve) and immunized mice on day 42 (2 weeks post-boost). Anti-CD45-PE antibody was injected retro-orbitally 10 minutes before sacrifice to distinguish the circulatory cells from tissue-resident memory cells22,23. The gating strategy is shown in (Figure S1). The percentage (Fig. 1A) and number (Figure S2A) of CD4 + CD45-CD44 + CD62L-CD69+ cells in the lungs of immunized and naïve mice increased following immunization. The percentage of IL-17+ (Fig. 1B), IFNγ+ (Fig. 1C) and IL-5+ (Fig. 1D) TRMs increased in the lungs of immunized mice compared to unimmunized (naïve) mice, similarly the number of cells were also increased in immunized mice (Figure S2B-D). IL-17+ producing CD4 + TRM were the predominant population. The percentage of CD4 + CD45-CD44 + CD62L-CD69+ cells (Fig. 1E) and the percentage of IL-17+ (Fig. 1F) and IFNγ+ (Fig. 1G) cells increased in the nose. Thus, IN immunization elicited a primarily TH17 polarized response in the respiratory tract. We also quantified CD8 + T cells in both the lung and nose and observed higher proportions and numbers in immunized mice compared to naïve controls. Lung CD8 + T cells produced both IFNγ+ and IL-17 + , where only IFNγ + CD8+ TRMs were found in the nose (Figure S3). Intranasal prime-boost immunization with BcfA/FHA/Prn drives a mucosal memory TH1/TH17 T-cell response. A The percentage of CD4 + CD45-CD44 + CD62L-CD69+ antigen-experienced CD4 + TRM in the lungs. The percentage of (B) IL-17 + , (C) IFNγ+ and (D) IL-5 + CD4 + TRM in the lungs. E The percentage of CD4 + CD45-CD44 + CD62L-CD69+ antigen-experienced CD4 + TRM in the nose. The percentage of (F) IL-17 + , (G) IFNγ + , and (H) IL-5 + CD4 + TRM in the nose are shown. Gating strategy in Figure S1. Mean ± SD is shown *p < 0.05, **p < 0.01. N = 10–15 mice/group pooled from two independent experiments. IN immunization generated mucosal IgG and IgA responses Mucosal antibodies contribute to clearance of Bordetella species from the respiratory tract25,26. We evaluated the antigen-specific IgG and IgA in the lungs and the nose of mice immunized with BcfA/FHA/Prn. Anti-BcfA (Fig. 2A), anti-FHA (Fig. 2B) and anti-Prn (Fig. 2C) specific IgG antibodies were detected in the lungs, and anti-BcfA (Fig. 2D) anti-FHA (Fig. 2E), and anti-Prn (Fig. 2F) specific IgG antibodies were detected in the nose following immunization. IgA antibodies specific to BcfA (Fig. 2G), FHA (Fig. 2H) and Prn (Fig. 2I) were detected in the lungs, and anti-BcfA (Fig. 2J), anti-FHA (Fig. 2K) and anti-Prn (Fig. 2L) specific IgA antibodies were detected in the nose. Thus, the trivalent vaccine elicits antigen-specific mucosal antibodies. Immunization with BcfA/FHA/Prn elicited mucosal antigen-specific IgG and IgA antibody responses. A anti-BcfA specific, (B) anti-FHA specific and (C) anti-Prn specific IgG antibody titers in the lung. D anti-BcfA specific, (E) anti-FHA specific and (F) anti-Prn specific IgG antibody titers in the nasal tissue. G anti-BcfA specific, (H) anti-FHA specific, and I. anti-Prn specific IgA antibody titers in the lung. J anti-BcfA specific, (K) anti-FHA specific, and (L) anti-Prn specific IgA antibody titers in the nose. Mean ± SD with the dotted line showing the limit of detection, corresponding to naïve mice. *p < 0.05, **p < 0.01. N = 10–15 mice/group pooled from two independent experiments. IN immunization with BcfA/FHA/Prn generates systemic T-cell and antibody responses We then tested whether IN immunization also elicited systemic immune responses. Splenocytes harvested from immunized mice were stimulated with medium alone (NS) or with BcfA, FHA or Prn for 3 days. ELISA analysis did not detect a significant increase in IFNy (Fig. 3A) produced by immunized splenocytes stimulated with BcfA, FHA or Prn. However, splenocytes from immunized mice stimulated with BcfA or FHA produced IL-17 (Fig. 3B) and IL-5 (Fig. 3C). Serum IgG responses specific for BcfA (Fig. 3D), FHA (Fig. 3E) and Prn (Fig. 3F) were detected in immunized but not naïve mice. Thus, systemic cellular and humoral responses were generated by IN immunization. BcfA/FHA/Prn immunization elicited systemic T-cell cytokines and serum antigen-specific IgG antibodies. A IFNγ, (B) IL-17 and (C) IL-5 in the supernatant of splenocytes stimulated with each antigen. Serum IgG specific to (D) BcfA, (E) FHA and (F) Prn. Mean ± SD with the dotted line showing the limit of detection, corresponding to naïve mice. *p < 0.05, **p < 0.01. N = 10–15 mice/group pooled from two independent experiments. Immunization reduced Bb colonization of the respiratory tract To test whether the vaccine protected mice against Bb colonization of the respiratory tract, immunized and unimmunized control mice were challenged IN with 5 × 105 CFU of RB50 at 2 weeks post-booster immunization (Fig. 4A). Lungs and nasal tissue were harvested on days 0, 4, 7, 14, and 28 post challenge for enumeration of CFUs. The bacterial burden in the lungs of immunized mice was reduced by 2 logs compared to the naïve mice by day 4 post infection and was at the limit of detection by day 28 post infection (Fig. 4B), while bacterial burden remained high in the lungs of unimmunized mice at all time-points tested. Bacterial burden in the nasal tissue of immunized mice was ~1 log lower than unimmunized mice on days 4, 7, 14, and 28 post infection (Fig. 4C). Thus BcfA/FHA/Prn immunization of mice effectively cleared a Bb challenge from the lungs and reduced nasal colonization. All three vaccine components were necessary for this effect, as bacterial load remained high in the lungs of mice immunized with BcfA/FHA or BcfA/Prn alone (Figure S4). IN immunization with BcfA/FHA/Prn reduced Bb colonization and disease symptoms. A Timeline of immunization, challenge and analysis. Mice were infected with RB50 and CFUs were enumerated at various times post-challenge in (B) lung and (C) nasal tissue (N = 5 mice/group). Dotted line annotates the limit of detection. D Percentage of original body weight post-challenge (N = 5 mice/group). E Average PenH per 10 minutes for each mouse. F Area under the curve analysis for PenH for each mouse. G Average coughs per 10 minutes for each mouse. H Area under the curve analysis for cough count. (Data are pooled from two independent experiments with 5 mice/group each). Mean ± SD is shown. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. BcfA/FHA/Prn immunization reduced disease symptoms in mice Bb infection causes weight loss and respiratory distress in infected animals27. To test whether IN immunization reduced Bb disease symptoms, we weighed mice daily for 13 days post-challenge (Fig. 4D). Unimmunized mice lost ~8% of their body weight on day 1 and nearly 12% of their weight by day 3 post-challenge. Weight loss gradually recovered to 95% of original weight by day 8 post-challenge. In contrast, immunized mice displayed less than 5% fluctuation in body weight during the infection. We used whole body plethysmography to evaluate lung function in Bb infected mice. This technique was previously used on rats infected with Bp and showed that PenH and the number of coughs increased following challenge28–30. PenH or enhanced pause is a measure of pulmonary distress. An increase in PenH correlates with a restricted airway caused by inflammation because of infection. Challenged unimmunized mice showed an early spike in PenH on days 1 and 2 post infection (Fig. 4E). Area under the curve (AUC) analysis showed that PenH was not elevated in immunized mice (Fig. 4F). Cough counts increased in unimmunized mice (Fig. 4G), while AUC analysis showed that cough was reduced in immunized animals (Fig. 4H). Macrophages and neutrophils are recruited to the respiratory tract following Bb challenge Phagocytes contribute to clearance of Bordetella from the respiratory tract31,32. We tested whether immunization altered the kinetics or magnitude of phagocyte recruitment to the lungs and nose after Bb challenge. The percentage of CD11b + F4/80+ cells in the lungs (Figure S5A) and nose (Figure S5B) of immunized challenged mice did not increase after infection compared to unimmunized mice. Ly6G+ neutrophils, particularly the Siglec-F+ subset of neutrophils is important for reducing nasal colonization in convalescent mice re-challenged with Bp32. To test whether neutrophils were recruited to the respiratory tract following Bb challenge of mice, we quantified the CD11b + Ly6G+ cells in the lungs and nose. The percentage of CD11b + Ly6G+ neutrophils (Fig. 5A) increased in the lungs of unimmunized mice compared to immunized challenged mice on day 7 post infection. The CD11b + Ly6G+Siglec-F+ (Fig. 5B) subset was similar in the lungs. Similarly, in the nose, the percentage of CD11b + Ly6G+ cells (Fig. 5C) was lower in immunized and challenged mice compared to unimmunized challenged mice. However, the percentage of CD11b + Ly6G+Siglec-F+ (Fig. 5D) cells increased in immunized challenged mice on day 7 post-challenge. Thus, Bb challenge recruited Siglec-F+ neutrophils to the nose, with accelerated kinetics in immunized mice. Transient increase in Siglec-F+ neutrophils in the nose following Bb challenge. A The percentage of total CD11b + Ly6G+ neutrophils and (B) the percentage of CD11b + Ly6G+ cells that are Siglec-F+ in the lungs. C The percentage of CD11b + Ly6G+ neutrophils and (D) the percentage of CD11b + Ly6G+ cells that are Siglec-F+ in the nose. Mean ± SD is shown. **p < 0.01. N = 10–15 mice/group pooled from two independent experiments. BcfA/FHA/Prn immunization reduced Bb induced lung inflammation Lung tissues from Bb challenged mice were harvested on days 4, 7 and 14 post-challenge. On day 4 post-challenge, H&E staining of lung tissue sections detected inflammation and necrosis affecting the interstitium and airways in naïve infected mice with minimal perivascular expansion of inducible bronchus-associated lymphoid tissue (iBALT) (Fig. 6A). In contrast, the immunized group had markedly reduced pathology and organized perivascular and peribronchiolar expansion of iBALT (Fig. 6B, arrows denoting iBALTs). Semi-quantitative evaluation of histopathological changes detected thickened alveolar walls in unimmunized challenged mice (Fig. 6C), and increased cellularity/consolidation (Fig. 6D) at day 4 post-challenge. There was increased infiltration of polymorphonuclear cells and macrophages in the lungs of unimmunized challenged mice (Fig. 6E), with an increase at day 7 and 14 post-challenge in total histopathological score (Fig. 6F). Thus, BcfA/FHA/Prn immunization protected mice from lung damage. BcfA/FHA/Prn immunization reduced inflammation and prevented lung damage following Bb infection. H&E-stained representative lung samples on day 4 post-challenge in (A) unimmunized challenged mice and (B) immunized challenged mice. Arrows denote iBALTS. Scale bar is 500 µm. Semi-quantitative evaluation of (C) alveolar wall thickening, (D) cellularity/consolidation, (E) airway PMNs/Macs and (F) total pathologic score. Mean ± SD is shown. *p < 0.05, **p < 0.01. N = 10–15 mice/group pooled from two independent experiments. Discussion Improved vaccines against Bb are urgently needed, as respiratory diseases caused by Bb pose significant health risks to both animals and humans2. In this work, we tested whether an IN administered protein subunit vaccine could elicit mucosal immune responses and confer protection against Bb colonization and disease. Studies in murine models of Bp showed that sustained protection against infection is associated with TH1/TH17 polarized systemic and mucosal T-cell responses24,33,34. Generation of mucosal immunity in the respiratory tract requires at least one vaccine dose to be administered IN35,36. Inactivated whole cell vaccines or cellular extracts, while immunogenic, may cause adverse reactions. Current Bb vaccines are composed of either attenuated live vaccines or cellular extracts, licensed as injectable or oral vaccines9,10. Their ability to generate mucosal immunity is unclear24. In contrast, adjuvanted protein subunit vaccines are versatile and amenable to IN administration and elicit mucosal immunity when an appropriate adjuvant is included 37,38. Here we tested the ability of BcfA to elicit T-cell responses in the upper and lower respiratory tract to itself and to two Bordetella antigens, FHA and Prn, that are expressed in all Bordetella species39. These proteins are also included in all current pertussis vaccines40. BcfA/FHA/Prn immunization generated TH17 polarized CD4 + TRM in the nose and the lungs demonstrating that this vaccine formulation generated the immune phenotype correlated with long-term protection against infection with Bordetellae41. CD8 + T cells that produced IFNγ and IL-17 were found in the respiratory tract. Furthermore, mucosal and systemic IgG antibodies specific to all three antigens were detected. TH17 polarized T-cell responses elicit IgA production in the mucosa42,43. Mucosal IgA antibodies prevent pathogen entry at barrier sites and contribute to bacterial clearance through multiple innate immune effector mechanisms44, including antibody-dependent cellular cytotoxicity (ADCC)45. In this study, IgA and IgG antibodies recognizing all three antigens were detected in the lungs. CD11b + F4/80+ macrophages were recruited to the lungs following Bb challenge of unimmunized and immunized mice, but did not increase as a result of immunization. These antibodies may have facilitated bacterial clearance by the macrophages through a variety of mechanisms including ADCC46–48. Follow-up studies will be important to dissect the mechanism of Bb clearance from the lungs. Bb nasal colonization was significantly reduced but not eliminated in BcfA/FHA/Prn immunized mice. Studies in animal models of the human pathogen Bp show that TH17 polarized T-cell responses in the nose recruit the Siglec-F+ subset of neutrophils that contribute to reduction of Bp nasal colonization32. In this study, we observed a transient increase in localization of Siglec-F+ neutrophils to the nose. The limited recruitment of this neutrophil subset, along with limited mucosal IgA responses to the antigens, may explain the continued colonization of the nose. Additional optimization of the vaccine formulation by modification of the antigen dose and inclusion of additional novel Bordetella antigens (manuscript under revision) that are recognized by nasal CD4 + TRM may improve nasal clearance. Furthermore, additional unknown effector mechanisms may contribute to elimination of the bacteria from the upper respiratory tract. Alternatively, the nasal microenvironment may actively inhibit effector responses. These possibilities are important areas for further investigation. Non-invasive tools to measure lung function are a powerful means for longitudinal analysis of the effects of infection on the respiratory tract. While paroxysmal cough is a classic symptom of pertussis disease in humans2, and baboons infected with Bp also cough49,50, coughing by mice infected with Bb or Bp was not previously observed. Here we showed that mice infected with Bb also cough, and that coughing can be quantified by plethysmography. Importantly, IN BcfA/FHA/Prn immunization prevented coughing. Elevated lung PenH was detected in unimmunized challenged mice, but not in immunized challenged mice. Together, these non-invasive measurements showed that Bb infection causes respiratory distress that is prevented by our vaccine. To our knowledge, this is the first demonstration that clinical disease symptoms caused by Bordetella infections can be observed and measured in mice. Immunized animals maintained their body weight while unimmunized challenged animals lost weight in the first week after infection, also indicating that immunization prevented disease. Histopathologic analysis of lung tissues showed that inflammation was markedly reduced in immunized and challenged mice compared to unimmunized controls, suggesting that the IN vaccine prevented damage to respiratory epithelium. Together, these data show that IN immunization with a TH17 polarizing vaccine protects mice against Bb disease and respiratory pathology. This vaccine has the potential to prevent disease in farm and companion animals. Further studies will be important to determine the longevity of protection provided by this vaccine and immunization strategy, and whether transmission of this pathogen may be prevented. Supplementary information Supplementary Information Supplementary Information Acknowledgements We thank the Comparative Pathology & Digital Imaging Shared Resource at The Ohio State University Comprehensive Cancer Center, Columbus, OH for tissue processing, staining and sectioning of mouse tissues. Author contributions A.R.A., K.K.P., J.M.H., M.M.S., M.G., G.D., and S.S. conducted experiments. A.R.A., K.K.P., and J.M.H. analyzed data and prepared figures. K.N.C conducted histopathology analysis. A.R.A., J.M.H., P.D., K.N.C. and R.D. wrote the manuscript. All authors read the manuscript and provided editorial comments. Data availability Primary data will be made available upon request to the corresponding author. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Alexa R. Aubrey, Komal K. Patel, Jesse M. Hall. Supplementary information The online version contains supplementary material available at 10.1038/s41541-026-01472-2. References Associated Data Supplementary Materials Supplementary Information Supplementary Information Data Availability Statement Primary data will be made available upon request to the corresponding author.