Aptamer-based Salmonella enteritidis-inactivated vaccine injected intradermally elicits local leukocyte recruitment and mucosal immunity in broiler chickens Authors: Santiago Uribe-Diaz, Abigeal I Omolewu, Jossie M Santamaria, Chrysta N Beck, Andressa Stein, Raul Marcon, Christine N Vuong, Gisela F Erf, Billy M Hargis, Young M Kwon Published: 2026 ABSTRACT AbstractOn-farm non-adapted Salmonella enterica ser. Enteritidis (S. Enteritidis) control in poultry requires vaccines that elicit rapid local immune responses maturing into long-lasting systemic and mucosal protective immunity. To assess the responses to primary and secondary intradermal (i.d.) immunization with CD40-agonistic DNA aptamer-based S. Enteritidis-inactivated vaccine (CD40-AP), thirty chicks were randomly assigned to six treatment groups (n = 5 per group) at hatch. Groups g1 - g3 received only a primary i.d. immunization on day 32, while groups g4 - g6 received a subcutaneous primary immunization on day 14 and an i.d. booster immunization on day 32. Vaccine treatment groups were: phosphate-buffered saline (PBS) control (g1, g4), mannosylated chitosan-based vaccine (MCA) (g2, g5), or CD40-AP (g3, g6). Blood and cloacal swabs were collected before immunization and 7 days after each immunization to determine relative levels of S. Enteritidis-specific plasma IgY and mucosal IgA. Growing feather (GF)-pulps were collected before (0 h) and at 0.25 (6 h), 1, 2, 3, 5, and 7 days after i.d. primary and booster immunization to assess the dynamics of the local leukocyte response. Both MCA and CD40-AP-inactivated vaccines induced macrophage and heterophil infiltration after primary and booster immunization (P < 0.05), while the secondary CD40-AP immunization promoted higher and prolonged presence of T cells, including both CD4+ and CD8+ subsets, and B cells compared to MCA (P < 0.05). The CD40-AP primary immunization induced higher levels of S. Enteritidis-specific mucosal IgA on day 7 compared to MCA (P < 0.05); whereas the MCA primary and booster immunization induced higher plasma levels of S. Enteritidis IgY than CD40-AP (P < 0.05). The more pronounced local T and B cell- and S. Enteritidis-specific mucosal-immune responses with CD40-AP, support the potential of the CD40-agonistic DNA aptamer-based vaccine platform to promote cell-mediated and mucosal immune responses while preserving the safety of inactivated vaccines in chickens.Keywords: Broilers, Salmonella, Aptamer-based vaccine, Immune responses Introduction With the global population projected to reach 9.7 billion by 2050, the demand for high-quality protein is steadily increasing (Henchion et al., 2021). Poultry production aids in meeting the rising global protein demand (Karabasanavar et al., 2020). However, intensive poultry production practices are associated with higher microbiological risk, particularly the spread of foodborne pathogens such as Salmonella (Nazeer et al., 2021). In chickens, non-adapted Salmonella serovars, such as Salmonella Enteritidis (S. Enteritidis), Typhimurium, and Infantis, colonize the gastrointestinal tract (GIT), causing minimal or no disease, while facilitating contamination of poultry products (Crouch et al., 2020; Rabie and Amin Girh, 2020). Therefore, on-farm strategies capable of reducing poultry GIT colonization with non-adapted Salmonella serovars remain essential to reduce public health risk. Salmonella vaccines are a key tool for controlling Salmonella in poultry; nevertheless, current commercially available vaccines are not 100% effective (Sáenz et al., 2022). Inactivated vaccines (killed vaccines or bacterins) offer superior safety than live vaccines but predominantly elicit systemic humoral responses biased toward IgY production, while mucosal and CMI responses are often suboptimal (Savelkoul et al., 2015; Tran et al., 2010; Pan et al., 2024). Improving inactivated Salmonella vaccine performance requires adjuvant strategies that preserve safety while selectively enhancing mucosal and cell-mediated immune responses. Innovation in biomaterials research and molecular biotechnology plays an important role in the development of next-generation vaccine platforms and adjuvants (Gong et al., 2022; Uribe-Diaz et al., 2025). In this context, chitosan, the product obtained by deacetylating the natural polysaccharide chitin, has been widely studied (Dolatyabi et al., 2024; Han et al., 2020a, 2020b). Chitosan molecules allow chemical modification, such as mannose binding, which can enhance antigen uptake by antigen-presenting cells (APCs). In addition, encapsulation of whole-cell or Salmonella antigenic units within chitosan has been shown to enhance antigen stability, mucosal interaction, and antigen presentation, resulting in enhanced innate, and humoral mucosal and systemic immune responses (Acevedo-Villanueva et al., 2022; Acevedo-Villanueva, et al., 2021b). Remarkably, a mannosylated chitosan-based avian influenza vaccine using a Bacillus vector demonstrated protective efficacy against challenge when administered subcutaneously (Vuong et al., 2020). In parallel, a promising approach for adjuvant development is to couple inactivated antigens together with agonistic immune receptor-targeting molecules that selectively activate the immune response (Al-Ogaili et al., 2020; Senevirathne et al., 2021; Takehara et al., 2003). It has been established that the engagement of CD154 (CD40 ligand) with CD40 is an essential co-stimulatory signal for optimal APC activation; an event required for efficient priming and expansion of antigen-specific effector and memory T cells, and robust B cell responses (Chen et al., 2010, 2012). Besides, CD40 receptor engagement may provide a route to B cell class switching to IgA (Grasset et al., 2020; Zan et al., 1998). Chou et al. (2016) demonstrated that the use of previously developed agonistic anti-CD40 monoclonal antibodies (mAb) delivered together with an oligopeptide through mucosal and parenteral routes triggered the production of peptide-specific mucosal IgA responses (Chen et al., 2010; Chou et al., 2016). Nevertheless, the use of agonistic anti-CD40 mAbs as vaccine adjuvants is unlikely due to their high production costs. DNA aptamers are short, linear, single-stranded oligonucleotides capable of folding into three-dimensional structures with high binding affinity to specific targets (Marimuthu et al., 2012), like antibodies. Our research group developed agonistic CD40-targeting DNA aptamers using systematic evolution of ligands by exponential enrichment (SELEX) (Al-Ogaili et al., 2020). Other research groups have developed Salmonella-targeting DNA aptamers with high affinity to a variety of serovars, including S. Enteritidis (Moon et al., 2013; Omolewu et al., 2026). To enhance CD40 engagement the rolling circle amplification technique (RCA) was used to design a DNA backbone that assembles multiple agonistic CD40-targeting DNA aptamers and Salmonella-targeting DNA aptamers carrying whole-cell inactivated bacteria (Patent: pending US application 17/602,951). Accordingly, this study aimed to evaluate the cellular and humoral immune dynamics of primary and booster intradermal (i.d.) immunizations with an experimental next-generation DNA aptamer-based S. Enteritidis inactivated vaccine (CD40-AP), including local leukocyte responses and the systemic IgY and mucosal IgA S. Enteritidis-specific antibody responses in broiler chickens. The growing feather pulp cutaneous bioassay was used to evaluate local leukocyte profile dynamics at the site of i.d. immunization, and plasma samples and cloacal swabs to determine the relative levels of S. Enteritidis-specific plasma-IgY and mucosal-IgA by ELISA. This three-window approach provides comprehensive insights into local and systemic innate and adaptive immune responses and how individual inactivated vaccine platforms differentially shaped local, systemic and mucosal immune responses in broiler chickens. Materials and methods Experimental design Animal handling and experimental procedures were approved by the University of Arkansas Systems Division of Agriculture Animal Care and Use Committee (protocol #23091). On day-of-hatch, thirty Cobb 500 male broiler chicks were obtained from a local hatchery. Chicks were randomly assigned to six treatment groups (n = 5 per group) and placed in separate battery cages with ad libitum access to feed and water. Diets and lighting programs followed industry standards. Birds were assigned to either a primary-immunization-only cohort (single i.d. dose at 32 d of age) or a recall cohort (primary subcutaneous (s.c.) immunization at 14 d, i.d. booster at 32 d). Treatments were assigned as follows: g1, primary i.d. administration of endotoxin-free phosphate-buffered saline (PBS) (Sigma-Aldrich, St. Louis, MO, USA); g2, primary i.d. immunization with mannosylated chitosan S. Enteritidis (MCA); g3, primary i.d. immunization with CD40-AP; g4, primary s.c. and i.d. secondary administration of PBS; g5, primary s.c. and i.d. booster immunizations with MCA; and g6, primary s.c. and i.d. booster immunizations with CD40-AP. CD40-AP vaccine production CD40-AP vaccine production was done following procedures previously described (Omolewu et al., 2026). Template circularization and primary annealing A phosphorylated linear DNA template (10 µM, 2.5 µL) (Integrated DNA Technologies, Coralville, IA, USA) and RCA DNA primer (10 µM, 2.5 µL) (Integrated DNA Technologies, Coralville, IA, USA) were diluted in 45 µL of Tris–EDTA (TE) buffer, pH 8.0 (Thermo Fisher Scientific, Waltham, MA, USA). The mixture was denatured at 90°C for 10 minutes and annealed at 56°C for 5 minutes using a Bio-Rad T100 Thermal Cycler (Bio-Rad Laboratories, Hercules, CA, USA). The ligation step was pursued by adding 6 µL 10x Hi T4 DNA ligase buffer (New England Biolabs, Ipswich, MA, USA), 1 µL T4 DNA ligase (New England Biolabs, Ipswich, MA, USA), and 3 µL of Milli-Q ultrapure water (MilliporeSigma, Burlington, MA, USA) to the 50 µL product obtained after denaturation. Then, the mixture was incubated at 16°C for 8 hours, followed by 10 minutes at 65°C to inactivate the enzyme. The resulting circular template hybridized with the primer (60 µL) was used as the substrate for the rolling circle amplification enzymatic reaction. RCA amplification The RCA backbone amplification reactions were performed by adding a mixture of 2 µL of phi29 DNA polymerase (10 Units/µL) (New England Biolabs, Ipswich, MA, USA), 8 µL 10x phi29 DNA polymerase reaction buffer (New England Biolabs, Ipswich, MA, USA), 0.8 µL of 100x BSA (Sigma-Aldrich, St. Louis, MO, USA), and 10 µL dNTPs (10 mM) (New England Biolabs, Ipswich, MA, USA). The mixture was incubated at 30°C for 16 hours. The size of the RCA product was analyzed with a 0.8% agarose electrophoresis gel stained with GelRed (Sigma-Aldrich). The concentration of RCA products was quantified using a Qubit fluorometer (Thermo Fisher Scientific, Waltham, MA, USA). Then, 50 µg of RCA product was mixed with 5 µL of a mix of agonistic CD40 and Salmonella-targeting DNA aptamers (10 µM) (New England Biolabs) per vaccine dose. The mixture was heated to 95°C for 10 minutes, then slowly cooled to room temperature (RT) for aptamer binding to the RCA backbone product. Antigen coupling Formalin-inactivated (0.01%) S. Enteritidis were attached to the Salmonella-specific DNA aptamer-RCA component, creating the complete vaccine construct. Squalene (0.4% v/v) and Tween 80 (1% v/v) were added. The final mixture contained 1 × 108 inactivated cells/mL. MCA vaccine production The MCA was prepared by combining two premixes (described below) in a final step. The final step for MCA vaccine preparation included the addition of the formalin-inactivated (0.01%) S. Enteritidis at a ratio of 2:1 v/v, respectively, and storage at 4°C. The final mixture contained 1 × 108 inactivated cells/mL. Mannose premix: using deionized water, glacial acetic acid (J.T. Baker, Phillipsburg, NJ, USA) was added until reaching 0.5%. Sodium acetate anhydrous (EM Science, Gibbstown, NJ, USA) was dissolved to 0.2% w/v, and pH of the solution was adjusted to 4.0. d-mannose (0.12 g/mL) (PureBulk, Roseburg, OR, USA) was added to the solution and stirred until fully dissolved. Chitosan premix: a food-grade chitosan (deacetylated 95%) with a molecular weight of 250 kDa (Paragon Specialty Products, LLC, Rainsville, AL) was diluted into deionized water (2% w/v). The solution was adjusted with glacial acetic acid 1.48% v/v and stirred until homogeneous. Immunizations At 14 days of age, 26 growing feathers (GFs) were plucked from the breast tracts of every broiler to yield uniform 18-day-old, regenerated GFs for i.d. immunization. Primary immunization for recall cohorts: On 14 d of age, broiler chickens in g4, g5, and g6 were subcutaneously primed with a 250 µL dose of PBS, MCA, or CD40-AP, respectively. Intradermal GF-pulp immunization: On 32 d of age (18 d post-feather plucking), the pulp-dermis of 25 regenerated GFs was injected with 10 µL of PBS (g1 and g4), MCA (g2 and g5), or CD40-AP (g3 and g6) using a 0.3 mL syringe and 31-gauge x 8 mm needle. The 18 d regenerated GFs were prepared for injection by cutting off the emerging barbs and the sheath portion above the pulp epidermis as previously described (Erf and Ramachandran, 2016; Santamaria et al., 2024). Blood, Cloacal Swab and GF-Pulp collection Blood sampling One mL of blood was collected from the wing vein of each broiler chicken using heparinized 3 mL syringes with 25-gauge x 1-inch needles, before the primary s.c. immunization at 14 days of age from all chickens, and 7 d post-primary s.c. immunization from chickens in g4, g5 and g6. Blood was also collected at 7 d post-i.d., primary and booster immunization. Plasma was separated from the blood samples by centrifugation (2000 x g for 3 minutes) at RT and placed into labeled 1.5 mL microcentrifuge tubes. All samples were stored at −80°C until analyzed. Cloacal swab sampling Cloacal swabs (CS) were collected at the same sampling time points as blood samples using sterile swabs, which were stored in labeled 1.5 mL microcentrifuge tubes containing 1 mL of PBS. Tubes were kept on ice and stored at −80°C until use. GF-pulp sampling From each bird, one 18-day-old regenerated GF-pulp was collected before (0 h) and at 0.25 (6 h), 1, 2, 3, 5, and 7 days post-i.d. immunization. Samples were placed into labeled 1.5 mL microcentrifuge tubes containing ice-cold Dulbecco’s PBS (DPBS). The microcentrifuge tubes containing the samples were kept on ice until pulp cell suspensions were prepared. Preparation of pulp cell suspensions Pulp cell suspensions of the collected GFs were prepared as previously described in Santamaria et al. (2024). Briefly, the sheath of each GF was cut longitudinally using a razor blade, and the entire pulp was extracted and placed into labeled 1.5 mL microcentrifuge tubes containing 0.5 mL of 0.1% collagenase (type IV, Life Technologies, Carlsbad, CA)-dispase (Boehringer Mannheim, Germany). The 1.5 mL microcentrifuge tubes were incubated in a water bath at 37 °C for 15 minutes. After incubation, the pulp tissue and solution were gently pushed through a 60 µm nylon mesh while generously rinsing the mesh with ice-cold DPBS. Cells were washed twice by centrifugation at 250 x g at 4 °C, and the pellet was resuspended in 0.4 mL DPBS containing the staining buffer (DPBS, 1% BSA, and 0.1% sodium azide). Cell suspensions were kept on ice until immunofluorescent staining. Immunofluorescent staining and flow cytometry analyses Cell populations in the GF-pulp suspensions were stained using a panel of fluorescently labeled mouse mAbs specific for chicken leukocyte markers (CD45-SPRD, KUL-01-PE, TCR1-F, CD4-PE, CD8α-SPRD, TCRαβ1-F, and Bu1-PE) (SouthernBiotech, Birmingham, AL, USA) as previously described by French et al., 2020. Three-color cell population analyses were performed for data acquisition on a BD Accuri™ C6 Plus flow cytometer (Becton, Dickinson, San Jose, CA, USA). Events were identified based on size (forward scatter), internal complexity (side scatter), and fluorescence against the targeted leukocyte marker using FlowJo software (Rocchi et al., 2023). In the absence of a specific marker for chicken heterophils, this population was identified based on forward- and side-scatter characteristics within the CD45+ leukocyte gate as previously described (Seliger et al., 2012). Controls were included to detect non-specific binding of labeled mAbs, to determine positive and negative fluorescence thresholds, and to set compensations as previously described (Erf and Ramachandran, 2016). The resulting data were expressed as the percentage of stained cells within the total pulp cell suspension. Enzyme-linked immunosorbent assay (ELISA) The relative levels of S. Enteritidis-specific plasma IgY and mucosal IgA were quantified using the ELISA protocol described in Santamaria et al., 2024. Briefly, 0.1% formalin-inactivated S. Enteritidis were suspended in PBS (1 × 107 CFU/mL). Flat-bottomed 96-well plates (Thermo Scientific, Waltham, MA, USA) were coated with 100 µL/well of S. Enteritidis suspension and incubated at 37°C for 2 hours, then stored overnight at 4°C. The plates were washed 3x with a wash buffer (50 mM Tris, 0.14 M NaCl, 0.05% Tween 20; pH 8.0), and 200 µL blocking buffer (50 mM Tris, 0.14 M NaCl, 1% BSA, pH 8.0) was applied to each of the 96 wells and plates incubated for 30 minutes at RT. After blocking, the washing step was repeated. Frozen plasma samples were thawed and diluted in diluent buffer (50 mM Tris, 0.14 M NaCl, 1% BSA, 0.05% Tween 20, pH 8.0) to 1:300 dilution for IgY ELISAs. Cloacal swab samples for relative mucosal IgA level determination were used without dilution. Then, 100 µL of the sample were run in triplicate wells. In addition, six dilutions of a freshly thawed aliquot from the same pooled plasma sample were included in every plate to establish the standard curve: 1:300 to 1:9600 (IgY), and 1:25 to 1:800 (IgA). Each plate also included a blank and a non-specific binding control. After adding the samples and controls, the plates were incubated for 2 hours at 37°C. Then, the washing step was repeated. The secondary antibodies used were HRP-conjugated affinity-purified goat anti-chicken IgG (Ig γ-heavy chain) and IgA (IgA (Ig α-heavy chain) (Bethyl Laboratories, Montgomery, TX, USA), diluted 1:20,000 (IgY) or 1:10,000 (IgA), respectively, in dilution buffer. The plates were incubated for 1 hour at 37°C, washed three times, and the TMB substrate solution (Bethyl Laboratories, Montgomery, TX, USA) was added (100 µL per well). After half an hour incubation at 37°C, each well received 100 µL of 2 M sulfuric acid to stop the enzymatic reaction. Absorbance was read at 450 nm using a microplate reader (ELx800, BioTek Instruments, Winooski, VT, USA). Results were noted in absorbance units (a.u.). The pooled standard a.u. values from the serial dilutions were used to generate a standard curve per plate, and sample readings were expressed as relative a.u. for S. Enteritidis-specific IgY and IgA. Statistical analyses Statistical analyses were performed using SigmaPlot 15 (Systat Software, Inc., San Jose, CA). For flow cytometry, a two-way ANOVA was used to assess the effects of treatments (PBS, MCA, and CD40-AP), time points (0, 6 h, 1 d, 2 d, 3 d, 5 d, and 7 d), and treatment-by-time points interactions on various leukocyte populations (% pulp cells). For ELISA, a one-way ANOVA was used to assess the effect of treatments for each time point. When required for flow cytometry or ELISA data, pairwise multiple comparison procedures were carried out using Tukey’s HSD multiple mean comparison test. Statistical significance was set at P < 0.05 (Fig. 1). Broiler chickens (n = 5 per group) were assigned to a primary-immunization-only cohort (Groups 1-3) or a recall cohort (Groups 4-6). At 32 d of age, all groups received an intradermal (i.d.) growing feather (GF) immunization consisting of 10 µL injected into each of 25 regenerated GFs (total i.d. dose = 250 µL) with PBS, MCA, or CD40-AP. Recall-cohort birds (Groups 4-6) also received 250 µL subcutaneous (s.c.) primary dose at 14 d with the same respective treatments. Blood and cloacal swabs were collected at baseline, 7 d after s.c. priming (recall cohort only), and 7 d after i.d. primary and booster immunizations to quantify anti-S. Enteritidis IgY and IgA. GF-pulps were sampled at 0 h and 6 h, 1, 2, 3, 5, and 7 d post-i.d. immunization to assess local leukocyte dynamics. Results Leukocyte dynamics in GF-Pulps to intradermal administration of PBS, MCA, and CD40-AP Innate immune cells Heterophils Baseline heterophil levels (% of total pulp cells) did not differ among treatment groups before either the primary or booster intradermal (i.d.) immunization (0 d; P > 0.05), and i.d. PBS administration did not alter heterophil levels over time (P > 0.05) (Fig. 2). Heterophil and macrophage population dynamics following intradermal (i.d.) immunization with PBS, mannosylated chitosan-based (MCA), or CD40-agonistic DNA aptamer-based Salmonella Enteritidis-inactivated vaccine (CD40-AP) in broiler chickens. For the primary-only cohort, birds (32 d) were i.d. injected into GF-pulps with PBS, MCA, or CD40-AP (10 µL per GF; 25 GF per bird; total 250 µL). For the recall cohort, birds were primed subcutaneously (s.c.) at 14 d with PBS, MCA, or CD40-AP (250 µL) and boosted i.d. at 32 d into GF-pulps (10 µL per GF; 25 GF per bird; total 250 µL). GF-pulps were collected before immunization (0 h) and at 6 h, 1, 2, 3, 5, and 7 days post-i.d. immunization. Pulp cell suspensions were stained with fluorescence-conjugated, chicken-specific mouse mAbs CD45-SPRD (all leukocytes) and KUL-01-PE (macrophages). Heterophils were gated as CD45⁺ cells with characteristic FSC/SSC (size/granularity). The population’s proportion expressed as % of total pulp cells was quantified by flow cytometry. Data are mean ± SEM. a-c: Different letters indicate differences between treatments within a time point; x-z: different letters indicate differences among time points within a given treatment (two-way ANOVA, Tukey’s HSD, P < 0.05). Following primary i.d. immunization, heterophil levels increased at 6 h in both MCA and CD40-AP groups relative to pre-injection levels (P < 0.05). Heterophil responses in the MCA group persisted through 5 d, whereas CD40-AP–associated heterophil elevations persisted through 3 d. At 6 h and 1 d, both vaccine groups exhibited higher heterophil levels than PBS (P < 0.05). MCA induced higher heterophil levels than CD40-AP at 6 h, 1 d, and 3 d (P < 0.05) (Fig. 2). Following booster i.d. immunization, heterophil levels increased at 6 h in both MCA and CD40-AP groups relative to pre-injection levels (P < 0.05), persisting through 2 d and 1 d, respectively. At 6 h, both vaccine groups exceeded PBS (P < 0.05), whereas at 1 d only MCA remained higher than PBS (P < 0.05). At 1 d post-booster, heterophil levels were higher in the MCA group than in the CD40-AP group (P < 0.05) (Fig. 2). Macrophages Baseline macrophage levels (% of total pulp cells) did not differ among treatment groups prior to either the primary or booster i.d. immunization (0 d; P > 0.05), and PBS administration did not alter macrophage levels over time (P > 0.05) (Fig. 2). Following primary i.d. immunization, macrophage levels increased at 6 h in both MCA and CD40-AP groups relative to pre-injection levels, with responses persisting through 2 d (P < 0.05). At 6 h and 1 d, macrophage levels in both vaccine groups were higher than PBS (P < 0.05). Macrophage levels were higher in MCA than CD40-AP at 1 d and 5 d (P < 0.05) (Fig. 2). Following booster i.d. immunization, macrophage levels increased at 6 h in the CD40-AP group and at 1 d in the MCA group, with responses persisting through 2 d (P < 0.05). Compared with PBS, CD40-AP showed higher macrophage levels at 6 h and 1 d (P < 0.05), whereas MCA showed higher levels at 1 d and 5 d (P < 0.05). At 6 h, macrophage levels were higher in CD40-AP than MCA (P < 0.05), whereas at 3 d and 5 d macrophage levels were higher in MCA than CD40-AP (P < 0.05) (Fig. 2). Lymphocytes B Cells Baseline B-cell levels did not differ among treatment groups before either primary or booster i.d. immunization (0 d; P > 0.05), and PBS did not alter B-cell levels over time (P > 0.05) (Fig. 3). B and T cell population dynamics following intradermal (i.d.) immunization with PBS, mannosylated chitosan-based (MCA), or CD40-agonistic DNA aptamer-based Salmonella Enteritidis-inactivated vaccine (CD40-AP) in broiler chickens. For the primary-only cohort, birds (32 d) were i.d. injected into GF-pulps with PBS, MCA, or CD40-AP (10 µL per GF; 25 GF per bird; total 250 µL). For the recall cohort, birds were primed subcutaneously (s.c.) at 14 d with PBS, MCA, or CD40-AP (250 µL) and boosted i.d. at 32 d into GF-pulps (10 µL per GF; 25 GF per bird; total 250 µL). GF-pulps were collected before immunization (0 h) and at 6 h, 1, 2, 3, 5, and 7 days post-i.d. immunization. Pulp cell suspensions were immunofluorescent stained for B cells and T cells. Population’s proportion expressed as % of total pulp cells were quantified by flow cytometry. Different lowercase letters (a-b) indicate statistical differences among treatments within a given time point, and different lowercase letters (w-z) indicate statistical differences among time points within a given treatment (two-way ANOVA, Tukey’s HSD, P < 0.05). After primary i.d. immunization, no changes in B-cell levels were detected in any treatment group across sampling time points (P > 0.05) (Fig. 3). Following booster i.d. immunization, B-cell levels in MCA-treated birds did not change over time (P > 0.05). In contrast, CD40-AP increased B-cell levels at 1 d post-booster (P < 0.05), with elevated levels persisting through 5 d and declining toward baseline by 7 d. At 1 d, 3 d, and 5 d post-booster, B-cell levels in the CD40-AP group were higher than those in both PBS and MCA groups (P < 0.05), whereas MCA did not differ from PBS at any time point (Fig. 3). T Cells Baseline total T-cell levels did not differ among treatment groups before either primary or booster i.d. immunization (0 d; P > 0.05), and PBS did not alter T-cell levels over time (P > 0.05) (Fig. 3). After primary i.d. immunization, no changes in total T-cell levels were observed in PBS or MCA groups (P > 0.05). In the CD40-AP group, total T-cell levels increased at 6 h (P < 0.05), followed by intermediate levels from 1 to 5 d and return to baseline at 7 d. Total T-cell levels in CD40-AP did not differ from MCA at any time after primary immunization (Fig. 3). Following booster i.d. immunization, total T-cell levels in MCA did not change over time (P > 0.05). In the CD40-AP group, total T-cell levels increased at 1 d (P < 0.05) and remained elevated through 2 d; a second increase was observed at 5 d (P < 0.05), with baseline levels restored by 7 d. At 6 h, 1 d, and 5 d post-booster, total T-cell levels in CD40-AP were higher than those in MCA (P < 0.05), while MCA did not differ from PBS at any time point (Fig. 3). CD4+ T Cells Baseline CD4⁺ T-cell levels did not differ among treatment groups before either primary or booster i.d. immunization (0 d; P > 0.05), and PBS did not alter CD4⁺ T-cell levels over time (P > 0.05) (Fig. 4). No changes were detected in any group following primary immunization (P > 0.05). Following booster immunization, CD40-AP increased CD4⁺ T-cell levels at 1 d (P < 0.05) relative to pre-injection values, followed by intermediate levels at 2 d and return to baseline by 7 d. A second increase occurred at 5 d (P < 0.05). At 1 d and 5 d, CD4⁺ T-cell levels in CD40-AP were higher than those in MCA (P < 0.05). MCA did not differ from PBS at any time point (Fig. 4). CD4⁺ T cells, CD8⁺ T cells, and TCR1+ cells population dynamics following intradermal (i.d.) immunization with PBS, mannosylated chitosan-based (MCA), or CD40-agonistic DNA aptamer-based Salmonella Enteritidis-inactivated vaccine (CD40-AP) in broiler chickens. For the primary-only cohort, birds (32 d) were i.d. injected into GF-pulps with PBS, MCA, or CD40-AP (10 µL per GF; 25 GF per bird; total 250 µL). For the recall cohort, birds were primed subcutaneously (s.c.) at 14 d with PBS, MCA, or CD40-AP (250 µL) and boosted i.d. at 32 d into GF-pulps (10 µL per GF; 25 GF per bird; total 250 µL). GF-pulps were collected before immunization (0 h) and at 6 h, 1, 2, 3, 5, and 7 days post-i.d. immunization. Pulp cell suspensions were immunofluorescent stained for CD4⁺ T cells, CD8α⁺ T cells), and γδ+ T cell. The population’s proportion expressed as % of total pulp cells was quantified by flow cytometry. Different lowercase letters (a–b) indicate statistical differences among treatments within a given time point, and different lowercase letters (x-z) indicate statistical differences among time points within a given treatment (two-way ANOVA, Tukey’s HSD, P < 0.05). CD8+ T Cells Baseline CD8⁺ T-cell levels did not differ among treatment groups before either primary or booster i.d. immunization (0 d; P > 0.05), and PBS did not alter CD8⁺ T-cell levels over time (P > 0.05) (Fig. 4). Following primary immunization, CD8⁺ T-cell levels remained unchanged in PBS and MCA (P > 0.05). CD40-AP increased CD8⁺ T-cell levels at 6 h (P < 0.05), with elevations maintained through 3 d. CD40-AP–treated birds had higher CD8⁺ T-cell levels than PBS at 6 h, 1 d, 2 d, and 5 d (P < 0.05), and higher levels than MCA at 6 h (P < 0.05) (Fig. 4). Following booster immunization, CD8⁺ T-cell levels in MCA did not change over time (P > 0.05). CD40-AP increased CD8⁺ T-cell levels at 1 d (P < 0.05), with elevations persisting through 2 d. CD8⁺ T-cell levels in CD40-AP exceeded MCA at 1 d and 3 d (P < 0.05). MCA did not differ from PBS at any time point (Fig. 4). TCR1+ γδ T cells Baseline γδ T-cell levels did not differ among treatment groups (Fig. 4), and PBS did not alter γδ T-cell levels over time (P > 0.05). After primary immunization, γδ T-cell levels remained unchanged in PBS and MCA (P > 0.05). CD40-AP increased γδ T-cell levels at 6 h (P < 0.05); at this time point, CD40-AP exceeded both PBS and MCA (P < 0.05). γδ T-cell levels in CD40-AP returned to baseline by 1 d and did not differ from other groups at later time points (P > 0.05) (Fig. 4). After booster immunization, CD40-AP increased γδ T-cell levels at 6 h (P < 0.05), with elevated levels maintained through 3 d and return to baseline by 7 d. In MCA-treated birds, γδ T-cell levels increased to intermediate levels at 6 h, peaked at 2 d, and returned to baseline by 7 d. Despite these within-group temporal patterns, γδ T-cell levels did not differ among PBS, MCA, and CD40-AP groups at any post-booster time point (P > 0.05) (Fig. 4). Humoral immune response after s.c. and i.d. GF-Pulp PBS, MCA, and CD40-AP administration Enteritidis Plasma IgY Anti-S. Enteritidis IgY levels did not differ among PBS, MCA, and CD40-AP groups at baseline (0 d s.c.) or 7 d after s.c. immunization (P > 0.05) (Fig. 5A). At 7 d after primary i.d. immunization, IgY levels were higher in MCA-treated birds than in CD40-AP–treated birds (P < 0.05). At 7 d after booster i.d. immunization, MCA again exhibited higher IgY levels than CD40-AP (P < 0.05) (Fig. 5A). Salmonella Enteritidis-specific antibody responses following subcutaneous (s.c.) and intradermal (i.d.) immunization with PBS, mannosylated chitosan–based (MCA), or CD40-agonistic DNA aptamer–based S. Enteritidis inactivated vaccine (CD40-AP) in broiler chickens. (A) Plasma IgY and (B) mucosal IgA relative levels (cloacal swabs) were measured by ELISA at baseline (0 s.c.), 7 d after the s.c. primary, 7 d after the i.d. primary (32 d), and 7 d after the i.d. booster. Anti-S. Enteritidis IgY and IgA were quantified by ELISA and are expressed as relative absorbance units (a.u.). Data are mean ± SEM. a,b: Within a time-point, different letters indicate significant differences between treatments (P < 0.05). Anti-S. Enteritidis mucosal IgA Anti-S. Enteritidis IgA levels did not differ among treatment groups at baseline (0 d s.c.) and did not change 7 d after s.c. administration (Fig. 5B). At 7 d after primary i.d. immunization, mucosal IgA levels were higher in CD40-AP than in MCA (P < 0.05). At 7 d after booster i.d. immunization, both CD40-AP and MCA were elevated, with no difference between vaccine groups (P > 0.05) (Fig. 5B). Discussion Unlike live-attenuated vaccines, inactivated Salmonella vaccines eliminate the risk of vaccine strain persistence, reversion to virulence, and vertical transmission. Inactivated Salmonella vaccines in poultry have been shown to consistently elicit strong and long-lasting systemic IgY responses (Shaji et al., 2025; Tran et al., 2010), thus aiding in bacterial opsonization and complement activation, which support systemic bacterial clearance and challenge resistance (Jawale and Lee, 2016; Santamaria et al., 2024). Nevertheless, inactivated vaccines lead to under-stimulation of the mucosal and cell-mediated immunity (Gayet et al., 2017). This has become the main limiting factor for their use in commercial settings. Several research groups have been working on developing next-generation adjuvants and delivery platforms to enhance the inactivated vaccine's ability to trigger protective mucosal and cell-mediated immunity (Acevedo-Villanueva et al., 2021a; Ishida et al., 2018; Renu et al., 2018). As an example, the use of mannose-binding adjuvants was demonstrated to enhance CMI to S. Enteritidis subunit proteins in broilers (Han et al., 2020a). Recently, Dolatyabi et al. (2024) established the use of Salmonella outer membrane and flagellin proteins entrapped in mannose chitosan nanoparticles as an inducer of protective mucosal IgA responses (Dolatyabi et al., 2024). On the other hand, conceptually, combining signals 1 (antigen: S. Enteritidis), 2 (co-stimulation via the CD40 receptor downstream pathway), and 3 (cytokines) can initiate a synergetic effect capable of promoting humoral mucosal and CMI. Therefore, studies have characterized immune responses induced by CD40 agonistic molecules as vaccine adjuvants (Chen et al., 2010, 2012; Layton et al., 2011). In chicken HD11 macrophage-like cells, engagement of CD40 with a CD40-agonistic mAb induced nitric oxide production, indicating direct APC activation through CD40 signaling without the need for microbe-associated or damage-associated molecular patterns (MAMPs; DAMPs) (Chen et al., 2010, 2012). Thus, building on the effectiveness of agonistic CD40 mAbs and peptides, our research group engineered a set of CD40-agonistic DNA aptamers multimerized via rolling-circle amplification (Al-Ogaili et al., 2020; Omolewu et al., 2026). During in vitro tests, this multimeric CD40-agonistic RCA platform was demonstrated to activate the HD11 chicken macrophage-like cell line, inducing nitric oxide production (Al-Ogaili et al., 2020; Omolewu et al., 2026). To confirm CD40-agonistic aptamers interaction with CD40 receptor, the aptamer’s secondary and tertiary structures and docking interactions were determined (Aceves-Hernández et al., 2026). This study aimed to evaluate local leukocyte dynamics and humoral mucosal and systemic immune responses in broiler chickens intradermally immunized with an experimental CD40-agonistic DNA aptamer-based S. Enteritidis inactivated vaccine. The GF-pulp bioassay enabled minimal-invasive, longitudinal determination of local leukocyte dynamics following i.d. GF-pulp immunization. Leukocyte dynamics observed in the GF-pulp following intradermal immunization reflects localized immune activation at the site of vaccine delivery, including early innate cell infiltration, antigen sensing, and antigen processing that promote adaptive immune engagement. Accordingly, these responses are interpreted as functional indicators of antigen uptake and immune activation, rather than nonspecific transient inflammation, and provide insight into the response profiles induced by different inactivated vaccine platforms. The determination of plasma Salmonella-specific IgY and mucosal IgA relative levels revealed the humoral outcomes of MCA and CD40-AP immunization. Thus, this experiment used a three-window approach (local, mucosal, and systemic plasma). Previously, a two-window approach using the GF-pulp bioassay for leukocyte dynamics evaluation and blood sampling for systemic humoral responses was effectively used in chickens and turkeys (Byrne and Erf, 2024; Erf et al., 2023; Abe et al., 2025). Local innate immune leukocytes dynamics Following primary and booster i.d. immunization, both MCA and CD40-AP S. Enteritidis vaccine platforms elicited rapid and transient recruitment of heterophils and macrophages in GF-pulps, whereas PBS produced a slight but not significant infiltration over time of any of these populations. The primary exposure of the GFs was phagocyte-dominated, with very low lymphocyte recruitment, consistent with the inflammatory profile after a primary immunization mentioned by Santamaria et al. (2024). The phagocyte recruitment triggered by primary immunization is intended to aid in early innate protection at the tissue level (Beck et al., 2025; Santamaria et al., 2024), in addition to promoting antigen sensing, processing and presentation. Interestingly, MCA immunizations induced significantly more robust and longer-lasting heterophil and macrophage infiltration than the CD40-AP vaccine platform, both following primary and early after the booster immunization. However, macrophage infiltration was observed more rapidly in the CD40-AP than in the MCA-treated group following both primary and booster immunization. The heterophil and macrophage i.d. cellular dynamics observed in this study appeared consistent with a differential, adjuvant-driven, tissue pro-inflammatory response following initiation of the canonical innate immune signaling pathway, typically observed after recognition of MAMPs and DAMPs by sentinel cells (Ciraci et al., 2010). Notably, these vaccine-initiated responses are the opposite of the events occurring in the ceca during S. Enteritidis persistent infection in chickens (non-canonical signaling pathway activation) (Kogut and Arsenault, 2015). Quicker macrophage infiltration seen after primary and booster CD40-AP immunization was likely triggered by local sentinel cell’s cytokines and chemokines release following MAMPs and DAMPs sensing, aptamers-CD40 receptor engagement, and stimulation of its downstream signaling pathways. It has previously been determined that an inactivated platform displaying CD40L on a bacterial envelope can increase pro-inflammatory cytokines and chemokines, reinforcing the role of CD40 co-stimulation in shaping early tissue innate responses (Senevirathne et al., 2022). Local and systemic cytokine and chemokine responses were not evaluated in this study; further studies are required to confirm these. However, it is well known that heterophils, macrophages, endothelial, and epithelial cells release cytokines and chemokines such as IL-8 to initiate and maintain tissue cell recruitment and infiltration. The key role of IL-8 in the recruitment of heterophils in chickens during S. Enteritidis infection has been previously evaluated (Kogut, 2002). On the other hand, i.d. heterophil and macrophage infiltration dynamics observed in MCA-immunized chickens are consistent with a depot effect, as observed with oil-based bacterins. In addition, the inclusion of the mannose modification of the chitosan has been reported to possess a highly pro-inflammatory effect mediated by APCs stimulation to release cytokines such as IL-1β, and TNF-α, which are mediators for the arrival of innate immune cells to the tissue (Han et al., 2020b). Therefore, we hypothesized that these factors could account for a more robust heterophil and macrophage response in MCA-treated chickens compared with CD40-AP-treated. Importantly, inactivated vaccines with depot effect tend to cause pain, excessive swelling, and tissue necrosis (Siddique et al., 2024; Tran et al., 2010). These tissue effects agree with findings during GF-pulp collection and processing at 5 and 7 d following MCA immunization, where a low number of the GF pulps collected were found to be necrotic. Overall, the results in this manuscript match with reports that have shown that a primary i.d. immunization with Salmonella bacterin drives heterophil and macrophage early infiltration (6 h) and that booster i.d. immunization extends local activity to 3 d, likely mediated by IL-8 expression (Beck et al., 2025; French et al., 2020), validating a conserved early inflammatory response (Santamaria et al., 2024). However, as our results imply, it seems that the conserved inflammatory response can be modulated by vaccine adjuvants. Local lymphocyte dynamics In contrast to the i.d. heterophil and macrophage dynamics triggered by CD40-AP, this vaccine platform preferentially increased T (CD4+, CD8+, and γδ) and B cell infiltration, especially following the booster immunization, compared with PBS and MCA. The higher and prolonged CD4+, CD8+, and γδ T cell levels exhibited by CD40-AP-treated birds, suggest that CD40 receptor engagement contributes to sustaining the local presence of different T cell subsets. The local T and B cell dynamics following i.d. CD40-AP immunization suggest that CD40-AP immunization promoted sustained local cellular engagement, with rapid and robust T and B cell recruitment, as reported with emulsified vehicles alone (Beck et al., 2025). Consistent with this, CD40-CD40L interactions can upregulate the expression of major histocompatibility complex (MHC), promote cytokine secretion, and thereby improve subsequent interactions between T cells and macrophages (Chen et al., 2010, 2012). Likewise, in vitro tests on avian B cells have shown that CD40 stimulation upregulated MHC class II within 24 to 48 h (Kothlow et al., 2008), then promoting B cell survival and engagement. The elevated GF pulp B cell levels following CD40-AP i.d. booster immunization may suggest local B cell amplification in response to secondary stimuli; this proposes local enhanced B cell collaboration with T cells following improved T cell tissue availability. In the avian immune system, CD40 stimulation drives B cell proliferation and survival, providing a sustained environment for T cell priming and germinal center development (Kothlow et al., 2008). The increase in B cell, CD4⁺ T cell, and CD8⁺ T cell infiltration following CD40-AP booster immunization may suggest enhanced systemic and local adaptive immune priming mediated by CD40 receptor engagement. However, to our knowledge, there is no report of dermal germinal center formation, and further experiments are needed to determine whether local T cell licensing and germinal center formation may have been possible following CD40-AP i.d. immunization. In addition, to determine whether local immune cellular are translated into CMI and systemic and mucosal humoral protective outcomes, further experiments should evaluate whether local T and B cell levels translate into differential blood and mucosa cellular dynamics in chickens vaccinated with CD40-AP following a challenge. In agreement with our findings, the composition of vaccine adjuvants can strongly influence local lymphocyte recruitment dynamics in the dermis. For example, Beck et al. (2025) and Santamaria et al. (2024) demonstrated that a water–oil–water emulsion vehicle, when administered intradermally into the growing feather pulp rapidly recruited T and B cells; however, this lymphocyte infiltration was markedly reduced when the vehicle was combined with LPS or whole killed Salmonella bacteria (Beck et al., 2025; Santamaria et al., 2024). In contrast, in the present study, the CD40-AP vaccine platform promoted sustained recruitment of both T and B cells following booster immunization despite the presence of whole-cell inactivated S. Enteritidis. These findings suggest that CD40 receptor engagement by the CD40-AP vaccine platform provide a strong co-stimulatory signal capable of maintaining local adaptive immune cells recruitment even in the context of bacterial antigen exposure. Humoral immune responses The data on anti-S. Enteritidis plasma IgY and mucosal IgA showed distinct outcomes between vaccine platforms, suggesting an adjuvant-driven effect like that observed when local leukocyte recruitment dynamics were evaluated. Antibodies containing the γ-heavy chain (IgY; the avian functional counterpart of mammalian IgG) are known to bind extracellular Salmonella located in tissue or blood to drive complement activation and opsonization, thus favoring phagocytosis (Jawale and Lee, 2014). Interestingly, IgY can also be detected in fluid collected from the chicken’s GIT, and its mucosal presence has been correlated with vaccine protective outcomes (Ishida et al., 2018). Several studies using non-adapted Salmonella inactivated vaccines have shown a link between blood anti-S. Enteritidis IgY levels and challenge protection, reduced bacterial cecal loads, and Salmonella clearance from poultry (Lee et al., 2002; Sáenz et al., 2022). Therefore, anti-Salmonella systemic IgY levels have been widely used as a measure of protective immunity. In the present study, the primary and booster i.d. administration of MCA in broiler chickens significantly increased systemic relative levels of anti-S. Enteritidis IgY 7 d after i.d. immunizations. This IgY pattern aligns with classical T cell-dependent responses, which are characterized by an IgM to IgY class switch and the emergence of a memory B cell population following secondary exposure (Selander et al., 1999). Similar results were previously found, showing increased IgY plasma levels after the first week post-inactivated vaccine administration (Beck et al., 2025; Santamaria et al., 2024). Although anti-Salmonella IgY levels in plasma or serum have been evaluated as a measure of protective immunity, different authors have disagreed with this idea, reporting results that do not imply IgY titers as a reliable indicator of protection against Salmonella. In fact, it has been shown that the level of serum IgY Salmonella-specific antibody responses in chickens of different ages and lines did not correlate with disease resistance (Beal et al., 2006). In contrast to plasma IgY’s suggested lack of protection against Salmonella infection in chickens, the dimeric secretory form of IgA, was shown to markedly limit Salmonella GIT colonization by blocking adherence and neutralizing Salmonella (Mastroeni and Rossi, 2020; Rossi et al., 2019). Therefore, it is highly relevant to design inactivated vaccine platforms that elicit robust production of anti-Salmonella IgA. Mechanistically, CD40 signaling is central to T cell-dependent class switching, affinity maturation, germinal-center formation, and memory B cell development (Al-Ogaili et al., 2020; Chen et al., 2010, 2012; von La Roche et al., 2024). The CD40-AP-linked mucosal IgA levels found 7 d after i.d. primary immunization are consistent with platforms that promote mucosa-relevant T cell help, potentially via cytokine milieus associated with T helper subsets (e.g., TGF-β-dependent pathways). While cytokines or homing markers were not quantified here, the pattern of the responses to the CD40-AP i.d. booster immunization (local T and B cell participation and early mucosal S. Enteritidis-specific IgA), may suggest that CD40-AP preferentially conditioned local environments conducive to mucosal IgA isotype responses. However, in contrast to our results, another CD40L-bearing inactivated construct elicited robust antigen-specific IgY following primary and booster vaccine administration (Senevirathne et al., 2022). In summary, novel next-generation inactivated vaccines will depend on adjuvants to trigger differentiated local, systemic, mucosal, and cell-mediated immunity. Overall, these results highlight MCA’s potential to favor systemic IgY and depot-like persistence of vaccine antigens. In contrast, the CD40-AP vaccine platform responses were shifted toward mucosal IgA and CMI immunity through CD40-targeted co-stimulation. These complementary profiles outline the potential of both MCA and CD40-AP vaccine platforms and suggest synergy in sequential or combined immunization regimens. Future studies should evaluate blood and mucosal leukocyte dynamics and local and systemic cytokine and chemokine profiles triggered by the vaccine platforms with and without challenge to resolve trafficking pathways and helper T cell polarization, as well as protective efficacy. In addition, challenges, doses, route of administration, and safety studies must be conducted. Importantly, increasing the sample size in future studies may improve the clarity of findings. CRediT authorship contribution statement Santiago Uribe-Diaz: Writing – review & editing, Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Abigeal I. Omolewu: Writing – review & editing, Resources, Investigation. Jossie M. Santamaria: Writing – review & editing, Methodology, Investigation. Chrysta N. Beck: Writing – review & editing, Methodology, Investigation, Formal analysis. Andressa Stein: Writing – review & editing, Resources, Methodology. Raul Marcon: Writing – review & editing, Resources, Methodology. Christine N. Vuong: Writing – review & editing, Supervision, Methodology, Conceptualization. Gisela F. Erf: Writing – review & editing, Supervision, Resources, Investigation, Conceptualization. Billy M. Hargis: Writing – review & editing, Supervision, Methodology, Conceptualization. Young M. Kwon: Writing – review & editing, Supervision, Project administration, Funding acquisition, Conceptualization. Disclosures The authors declare the following financial interests which may be considered as potential competing interests: Christine N. Vuong, Adil Al-Ogali, Billy M. Hargis, and Young Min Kwon, co-inventors on pending US application 17/602,951, which is owned by the Board of Trustees of the University of Arkansas and The Texas A&M University System, may have certain rights to revenues which may result from its commercialization. Acknowledgements This work was supported by the USDA National Institute of Food and Agriculture, Agricultural and Food Research Initiative Competitive Program, grant number: 2021-08164. Title: Controlling Salmonella through enhanced understanding of horizontal transmission and a novel and scalable vaccination strategy in broilers. References