Whole-cell lysates from bacterial strains were analyzed by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and lectin blotting with soybean agglutinin (SBA). A total of 5 mL overnight cultures of STM-1 and S. Typhimurium 82/6915ΔaroA::pgl were grown at 37°C / 200 rpm, and C. hepaticus HV10^T was grown for 48 h on HBA plates under microaerophilic conditions. A total of 1 mL of each S. Typhimurium strain was harvested, C. hepaticus HV10^T plates were flooded with 2 mL PBS, harvested by gently scraping, and pelleted at 5,000 x g for 5 min, and washed three times in 1 mL PBS. Cell pellets were lysed as McDonald et al. (46) described with the addition of boiling for 5 min at 100°C. Supernatants from whole cell lysates were collected, and protein concentrations were determined using a Qubit assay kit (Thermo Fisher Scientific). Approximately 25 μg of protein extracts were suspended in 4x Laemmli sample buffer (Bio-Rad) containing 50 mM dithiothreitol or the equivalent amount of whole cell lysates treated with proteinase K (100 μg/mL) at 55°C for 30 min. Samples were heated for 5 min at 95°C, loaded on 8–16% precast polyacrylamide gels (Bio-Rad), and separated at 200 V for 45 min. SDS-PAGE gels were either developed using SimplyBlue Safe stain (Invitrogen) or transferred to a polyvinylidene fluoride (PVDF) membrane using an iBlot™ Dry Blotting system (Invitrogen). Lectin blotting was performed as previously described by McDonald et al. (25), using SBA as the primary detection reagent and streptavidin-HRP as the secondary detection solution. Signals were detected using tetramethylbenzidine (TMB), and blots were imaged on a ChemiDoc imaging system (Bio-Rad).

To determine whether the Campylobacter heptasaccharide had integrated as a part of S. Typhimurium 82/6915ΔaroA::pgl endogenous LPS, the LPS from this strain was isolated and analyzed. LPS was extracted from 3 L of freshly grown overnight cultures of S. Typhimurium 82/6915ΔaroA::pgl and a 1 L overnight culture of S. Typhimurium STM-1 using a method adapted from (47, 48). Briefly, 3 × 1.0–1.5 g cell dry weight of S. Typhimurium 82/6915ΔaroA::pgl and 1.0 g cell dry weight of STM-1 were washed with 20 mL PBS containing 0.15 mM CaCl₂ and 0.5 mM MgCl₂. Cell suspensions were pelleted and resuspended in 20 mL of the same buffer. Cell suspensions were boiled at 100°C for 20 min and mixed every 5 min by inverting tubes. Cell suspensions were cooled, treated with 200 μg/mL proteinase K, and heated at 59°C for 3 h. DNase and RNase were added to samples at a final concentration of 40 μg/mL and 80 μg/mL, respectively, in the presence of 20 μL of 20% MgSO₄ and 80 μL of chloroform, and the mixture was heated at 37°C for 1 h. Phenol/TRIS solution (20 mL) was heated to 65°C, added to enzymatically treated cell lysates (20 mL), and incubated at 65°C for 15 min with occasional vortexing and shaking. Cell suspensions were cooled on ice and centrifuged at 8500 x g for 15 min. Supernatants were transferred to fresh tubes, and the phenol phase was re-extracted by adding 20 mL of Milli-Q water. Sodium acetate at a final concentration of 500 mM and 10 volumes of 95% ethanol were added to LPS extracts and stored at −20°C overnight to precipitate LPS. Samples were centrifuged at 2,000 x g, 4°C for 10 min; supernatants were removed, and pellets were resuspended in 3 mL of Milli-Q water. Samples were heated at 75°C with vortexing until LPS samples were solubilized. A total of 2 mL of the LPS solutions were dialysed against Milli-Q water with three changes at 1 h intervals to remove residual phenol and other contaminants. A total of 2 mL of the LPS extracts were lyophilised and stored at 4°C.

Lyophilised LPS samples were resuspended in 1 mL of Milli-Q water, and 5 μL of each sample was added to 5 μL of 4x Laemmli sample buffer and 10 μL of PBS and loaded onto 8–16% precast polyacrylamide gels and separated at 100 V for 80 min. LPS extracts were visualized by direct staining either with the Pierce Silver stain kit (Thermo) or Pro-Q Emerald 300 Lipopolysaccharide Gel Stain Kit (Thermo) following the manufacturer’s instructions. Lectin blotting of the LPS samples was also performed to determine whether LPS was modified with the Campylobacter heptasaccharide.

Smaller amounts of LPS were extracted from S. Typhimurium 82/6915ΔaroA::pgl and STM-1 as described by Davis and Goldberg (49). Briefly, 1.5 mL suspensions of bacterial strains at an OD₆₀₀ = 0.5 were pelleted at 10,600 x g for 10 min, the supernatant removed, and the pellets resuspended in 200 μL of 1x Laemmli sample buffer containing 175 mM dithiothreitol (DTT). Cell pellets were lysed by heating at 100°C for 15 min. Lysed pellets were treated with 500 μg/mL proteinase K at 59°C for 3 h, and 200 μL of Tris-saturated phenol was added to each sample, vortexed, and heated at 65°C for 15 min. Samples were cooled, and 1 mL of diethyl ether was added. Samples were vortexed and centrifuged at 20,600 x g for 10 min. The bottom layer containing LPS was collected and subjected to a second round of Tris-saturated phenol and diethyl ether treatment and stored at −20°C until analysis.

Mild acid hydrolysis of LPS was performed to cleave the Kdo bond between the lipid A and the core oligosaccharide region of LPS as described by Batley et al. (50) with some modifications. Briefly, 40 μL of extracted LPS samples were treated with 1 or 2% glacial acetic acid (80% w/v) and heated at 100°C for 2 h. Samples were neutralized by the addition of 1 or 2% 10 M NaOH, respectively. A total of 10 μL of each LPS sample was separated by SDS-PAGE and visualized by direct staining either with the Pierce Silver Stain Kit (Thermo) or the Pro-Q Emerald 300 Lipopolysaccharide Gel Stain Kit (Thermo). Lectin blotting of LPS samples was also performed, as McDonald et al. (25) described, to determine whether LPS was modified with the Campylobacter heptasaccharide.

S. Typhimurium 82/6915ΔaroA::pgl and STM-1 lyophilised LPS samples were reconstituted in ultrapure water to 10 mg/mL. In some experiments, low molecular weight LPS was further purified by passing hot phenol extracts through a 10 kDa molecular weight cut-off spin filter (Merck Millipore) according to the manufacturer’s instructions. Reconstituted LPS extracts were acidified with the addition of acetic acid to a concentration of 1% (v/v), after which samples were subjected to partial hydrolysis by heating at 95°C for 90 min. Samples were lyophilised in a SpeedVac (Savant), reduced by reconstitution in 0.5 M NaBH₄ and 50 mM KOH, and incubated overnight at 50°C.

Following reduction, samples were diluted to a volume of 100 μL with ultrapure water, acidified with 1% acetic acid, and then desalted using HyperCarb™ porous graphitic carbon stage-tips (Thermo Scientific). Following the equilibration of columns with 0.1% trifluoroacetic acid (TFA), samples were loaded onto the stage tips three times. The columns were then washed with three volumes of 0.1% TFA. Subsequently, glycans were eluted using a solution of 50% acetonitrile (MeCN) and 0.1% TFA and were lyophilised prior to analysis.

For the analysis of free/released glycans, samples were reconstituted in ultrapure water and loaded onto a 3 cm x 1 mm inner diameter (i.d.) x 3 μm Hypercarb™ porous graphitic carbon (PGC) column (Thermo Scientific) in 100% buffer A (10 mM NH₄HCO₃) at 20 μL/min using an Agilent 1290 Infinity LC system. Glycans were eluted by altering the mobile phase to 14% buffer B (70% MeCN, 10 mM NH₄HCO₃) over a 1 min linear gradient, after which the mobile phase was altered to 40% over a 36 min linear gradient directly into a linear triple quadrupole Velos™ Pro mass spectrometer (Thermo Scientific). The instrument was configured to perform one full scan MS experiment (scan range 570–2000 m/z, automatic gain control [AGC] of 3e4, and maximum IT of 100 msec) with the top 5 precursors in fulfilment of the selection criteria (dynamic exclusion window of 15 s) selected for MS/MS (scan range 150–2000 m/z, AGC of 1e4, maximum IT of 100 msec, isolation window 1.4 m/z, normalized collision energy [NCE] set as 33). Ions pertaining to glycans and/or glycoconjugates were identified by the presence of relevant oxonium ions (e.g., 204 m/z), after which glycan structural configuration was confirmed manually.

Overnight cultures of STM-1 and S. Typhimurium 82/6915ΔaroA::pgl were grown at 37°C with shaking at 200 rpm, and C. hepaticus HV10^T was grown for 48 h on HBA plates under microaerophilic conditions. A total of 1 mL suspensions of bacterial strains at OD₆₀₀ = 1.0 were pelleted at 8,000 x g for 10 min and washed twice with 1 mL PBS. Bacterial pellets were resuspended, fixed in 1 mL of 3% paraformaldehyde, and incubated at RT for 30 min. Fixed cells were pelleted and washed three times with 1 mL PBS. Cell pellets were resuspended in PBS to a final OD₆₀₀ of 0.2–0.3 and stored at 4°C. Fixed cells were pelleted at 8,000 x g for 10 min, and the supernatants were discarded. Cells were blocked in 5% BSA for 1 h at RT or overnight at 4°C. The blocking solution was discarded, and cells were incubated for 1 h in the dark with either 25 μg/mL SBA fluorescein (Vector Laboratories) or 25 μg/mL FITC anti-Salmonella antibody (Abcam) diluted in PBST (PBS plus 0.1% Tween 20) containing 1% BSA. The unbound antibody was removed by washing the cells three times in 1 mL PBS.

For flow cytometry, cells were resuspended in PBS to approximately 10⁶ cfu/mL, and a total of 10,000 events were acquired and analyzed for each sample using a Beckman Cytoflex and CytExpert 2.5 software. For fluorescence microscopy, samples were counterstained with DAPI at a final concentration of 10 μg/mL in PBS-1% BSA solution for 15 min in the dark at RT. Cells were pelleted and washed twice in 1 mL PBS. Finally, cells were resuspended in 50 μL PBS and diluted 1/100. A total of 10 μL of stained bacteria were added onto coverslips and allowed to air dry in the dark for 30 min. Coverslips were then mounted onto Superfrost microscopic slides with 15 μL of Mowiol DABCO mounting solution and incubated at RT in the dark for 30 min before examination using a Leica DM2500 epifluorescence microscope. Images were captured using LAS software (V4.1; Leica Microsystems).

An animal study followed the protocol approved by the RMIT University Biosafety Committee (NLRD 2002) and the CSIRO Animal Ethics Committee, AEC approval 2050. A total of 45 six-week-old Hy-Line Brown commercial layer hens from a Campylobacter and Salmonella-free flock were obtained from a commercial layer pullet farm. Animal experiments were conducted at the Werribee Animal Facility of the Australian Animal Health Laboratory (CSIRO). Birds were randomly allocated into three groups and reared on deep litter (softwood shavings) in floor pens, identified using wing tags, and acclimatized for one week before conducting the trial. Cloacal swabs were collected before arriving at the facility. DNA was extracted from the swabs using a Maxwell RSC Buccal swab kit (Promega) according to the manufacturer’s instructions to confirm the absence of C. jejuni by PCR as described by Eyers et al. (51), which was also used to confirm the absence of C. hepaticus. The absence of S. Typhimurium was confirmed by PCR, as described below. The study involved the use of the groups detailed in Table 1.

To evaluate vaccine strain colonisation and persistence at specific sites, host tissues and caecal content samples were processed and enriched in tryptone soy broth. These samples were then plated on modified semi-solid Rappaport-Vassiliadis agar (MSRV), Oxoid Salmonella Chromogenic medium (OSCM), and Xylose Deoxycholate (XLD) agar, both with and without chloramphenicol. This approach facilitated the selection of S. Typhimurium 82/6915ΔaroA::pgl (Cm^R) from STM-1, which does not carry Cm^R. Whole blood samples were taken weekly from the same five birds from each group for serological analysis. The experimental plan and study design are depicted in Figure 1.

To prepare the vaccine strains, S. Typhimurium 82/6915ΔaroA::pgl was revived from −80°C glycerol stocks, streaked onto LB agar, and grown overnight (16–18 h) at 37°C. A single colony of S. Typhimurium 82/6915ΔaroA::pgl was subcultured into 5 mL LB broth and also grown overnight (16–18 h) at 37°C with shaking at 200 rpm. The optical density at 600 nm (OD₆₀₀) of the S. Typhimurium culture was measured, and the culture was diluted/adjusted to approximately 1 × 10⁷ CFU/100 μL, assuming an OD₆₀₀ of 1 corresponds to approximately 1 × 10⁹ CFU/mL. For STM-1 preparation, vaccine vials were reconstituted in distilled water and diluted/adjusted to the same concentration. Cell counts were confirmed by serial dilution and overnight incubation on Salmonella XLD selective agar plates (Thermo Scientific). Chickens in groups 2 and 3 were vaccinated via intramuscular (IM) injection with 100 μL of cultures containing 10⁷ CFU. This route of administration was chosen because IM injection of STM-1 had previously been shown to significantly improve antibody responses to vaccination (18) and is routinely used for administering multiple vaccines to pullets between 8 and 14 weeks of age. Vaccines were co-administered with 500 μL of a commercial multivalent egg drop syndrome/Newcastle disease killed vaccine (Nobilis® EDS + ND, MSD Animal Health) to simulate in-field conditions. Co-administration involved first drawing 500 μL of Nobilis® EDS + ND into the syringe, followed by 100 μL of S. Typhimurium vaccines, and injecting the unmixed preparation into the breast muscle. Mock/unvaccinated birds received 100 μL of LB broth. Birds were vaccinated twice, at 7 and 11 weeks of age.

Cloacal swabs and whole blood from wing veins were collected from five birds per treatment group on days 7, 14, 21, 28, and 35 of the trial to assess Salmonella shedding and monitor the development of antibody responses to vaccination by ELISA. On study days 2, 3, and 42, five birds from each group were euthanised by CO₂ asphyxiation in accordance with approved procedures before whole blood collection by cardiac puncture and post-mortem examination. Whole blood samples were centrifuged at 1000 x g for 10 min, and the collected serum was stored at −20°C for subsequent IgY analysis by immunoblotting and ELISA. Breast muscle tissue at the injection site, caecal content, cloacal swabs, and liver and spleen samples were harvested to determine the localisation and persistence of the vaccine strains post-intramuscular injection. Localisation and persistence were assessed by culturing tissue samples and extracting DNA for PCR analysis.

Breast muscle at the injection site, liver, spleen, and caecal content samples were processed for S. Typhimurium 82/6915ΔaroA::pgl and STM-1 isolation by culturing as described in (18, 52, 53). Briefly, samples were pulverized using a scalpel and tweezers, and homogenized tissue samples were placed in 6-well culture plates containing 4 mL tryptone soy broth and incubated at 37°C for 18 h while gently shaking. Three 30 μL spots of enriched cultures were plated onto MSRV agar and grown at 37°C for 24 h. Cultures were subcultured onto OSCM and grown at 37°C for 18 h. Purple colonies were presumed to be Salmonella, and individual colonies were subcultured onto XLD and XLD Cm₃₀ plates and grown at 37°C for 18 h. STM-1 can be differentiated from wild-type Salmonella due to its inability to produce H₂S on XLD; thus, colonies appear white and lack the black centres characteristic of wild-type Salmonella grown on XLD (54). To differentiate STM-1 from S. Typhimurium 82/6915ΔaroA::pgl, colonies were selected on XLD Cm₃₀ antibiotic plates. In addition, DNA was extracted from cloacal swabs and tissue samples using the Maxwell RSC Buccal Swab & PureFood GMO DNA kit (Promega), respectively, as per the manufacturer’s instructions. As mentioned above, DNA extracts were used as templates for PCRs designed to detect amplicons specific to each vaccine strain.

Immunoblotting was performed on extracted LPS samples as previously prepared. Briefly, 2 μL of LPS extracts were loaded onto 8–16% precast polyacrylamide gels and subjected to SDS-PAGE at 100 V for 75 min. The LPS samples were then transferred to PVDF membranes, which were blocked with 5% BSA for 1 h at room temperature (RT) while rotating on a roller mixer. Blots were then probed with 1:500 of pooled sera from birds in group 1 (unvaccinated) and group 3 (vaccinated with S. Typhimurium 82/6915ΔaroA::pgl) taken at day 42. Sera were diluted in PBST (0.05% Tween20) containing 1% BSA, 100 μL of total STM-1 LPS extracts, and 100 μL of STM-1 OD600 = 1 WCL to pre-adsorb any antibodies with cross-reactive responses to S. Typhimurium LPS. The blots were incubated for 1 h at RT.

Following incubation, the blots were washed three times for 5 min each in PBST (0.05% Tween 20) and subsequently incubated with a 1:5000 dilution of rabbit anti-chicken IgY (H + L) secondary antibody conjugated to horseradish peroxidase (HRP), diluted in PBST (0.05% Tween 20) containing 1% BSA for 1 h at RT. The blots were again washed three times for 5 min each in PBST (0.05% Tween 20), followed by three 5-min washes in PBS and a final single 5-min wash in Milli-Q water. Detection was carried out using a chromogenic TMB substrate.

IgY antibody responses specific to the Campylobacter heptasaccharide region of the vaccine were quantified using indirect ELISA. The coating antigen used was a Campylobacter N-glycoprotein CjaA_ChuA (9), modified with nine Campylobacter N-glycans and a hexahistidine (His₆) tag. CjaA_ChuA is a fusion protein that combines the antigenic Campylobacter N-glycoprotein, CjaA (55), with the predicted antigenic epitope of the Campylobacter ChuA protein, incorporating eight additional sequons for N-glycan attachment. This glycoprotein and its unglycosylated equivalent were purified from S. Typhimurium 82/6915ΔaroA::pgl [pUC57CjaA_ChuA (9)] (glycosylated) and S. Typhimurium 82/6915 [pUC57CjaA_ChuA (9)] (non-glycosylated), respectively, using immobilized metal affinity chromatography (IMAC).

Briefly, S. Typhimurium 82/6915ΔaroA::pgl [pUC57CjaA_ChuA (9)] and S. Typhimurium 82/6915 [pUC57CjaA_ChuA (9)] cultures were grown overnight in LB broth supplemented with ampicillin (100 mg/mL) at 37°C with shaking at 200 rpm. Pellets were harvested by centrifugation at 10,000 × g for 10 min. The pellets were then sonicated for 4 min (20 s on, 10 s off at 40% frequency) in Ni-NTA lysis buffer (300 mM NaCl, 50 mM NaH₂PO₄, 10 mM imidazole, pH 8). After sonication, the lysates were centrifuged at 7,000 × g for 10 min to collect the supernatant.

For His-tag purifications, 1–2 mL of Ni-NTA beads (Qiagen) were used for every 2–4 litres of culture media equivalent supernatant. The beads were placed in an appropriately sized gravity column washed with 10 column volumes (CV) of Ni-NTA wash buffer (300 mM NaCl, 50 mM NaH₂PO₄, 20 mM imidazole, and pH 8). The beads were then added to the supernatant and incubated with rolling for 1 h at RT or overnight at 4°C.

After incubation, the beads were loaded back into the column and washed with wash buffer. For glycosylated samples, the beads were washed with 3 CV of 20 mM imidazole wash buffer, while unglycosylated samples were washed with 2 CV of 20 mM imidazole wash buffer, followed by 2 CV of 30 mM imidazole wash buffer. Glycosylated samples were then eluted with a buffer containing 300 mM NaCl, 50 mM NaH₂PO₄, and 250 mM imidazole, pH 8. Unglycosylated samples were eluted with a buffer containing 300 mM NaCl, 50 mM NaH₂PO₄, and 80 mM imidazole, pH 8. The eluates were then desalted using Cytiva G-25 MidiTrap columns with PBS.

The glycosylated protein underwent an additional Ni-NTA purification step using Ni-NTA beads, repeating the previous steps with a 0.5–1.0 mL Ni-NTA bead bed. Finally, electroelution was used to remove any contaminating S. Typhimurium proteins from the purified protein samples. A total of 800 μL of desalted Ni-NTA purified unglycosylated sample was loaded into two pre-cast 12-lane gels and run at 200 V for 32 min. The first lane of each gel was cut and stained/destained with Coomassie, while the remaining lanes were left in SDS-Tris-glycine buffer. The stained lane was paired with the unstained lanes from the same gels, and the known 37 kDa band was excised from the gel. The unstained excised gel strip was inserted into a Slide-A-Lyzer™ G3 Dialysis Cassette, 20 K MWCO. Cassettes were placed into one side of a DNA gel electrophoresis tank in chilled SDS-Tris-glycine buffer and electroeluted at 250 V for 2 h. Electroelution was paused every 30 min to replace the buffer with fresh, chilled buffer to prevent overheating. Cassettes were then dialysed against PBS, and samples were extracted, 0.22 μm filtered, and concentrated using an Amicon® Ultra Centrifugal Filter, 10 kDa MWCO.

1 mL of SBA-bound agarose beads (Vector Laboratories, AL-1013-2) were washed with 10 CV of SBA Buffer (150 mM NaCl, 50 mM HEPES, pH 8) to obtain highly purified glycoprotein. The beads were loaded into the desalted, twice Ni-NTA purified glycosylated sample (1–1.5 mL of sample for every 1 L of culture starting material) and incubated with rolling for 1 h at RT or overnight at 4°C. The beads were loaded into a 2 mL gravity column and washed with 3 CVs of SBA Buffer. The sample was pre-eluted with 0.3 CV of Glycoprotein Eluting Solution (Vector Laboratories, ES-2100) before eluting with 3 CV of Glycoprotein Eluting Solution. The eluate was added to a Slide-A-Lyzer™ G3 Dialysis Cassette, 20 K MWCO, and dialysed against PBS. Samples were extracted from the dialysis cassette, 0.22 μm filtered, and concentrated using an Amicon® Ultra Centrifugal Filter, 10 kDa MWCO.

Protein concentrations were determined using a Qubit assay kit and separated by 8–16% SDS-PAGE. Protein expression and glycosylation status were verified by western blotting and SBA lectin blotting, respectively. For western blotting, gels were transferred to PVDF membranes and blocked in 5% BSA at RT for 1 h. Blots were then probed with mouse anti-his [1:3000] in PBST (0.5% Tween₂₀) containing 1% BSA and incubated at RT for 2 h while rotating. Blots were washed three times in PBST (0.05% Tween₂₀) followed by incubation in [1: 5000] goat anti-mouse IgG in PBST containing 1% BSA for 1 h at RT. Blots were then washed three times for 5 min in PBST and PBS and 5 min in Milli-Q water. Membranes were incubated in TMB for chromogenic detection of his-tagged proteins.

Serum IgY vaccine-specific responses were quantified using ELISA. To measure antibody (IgY) responses against S. Typhimurium, a Salmonella group B antibody test kit (BioChek) was used following the manufacturer’s instructions. A 96-well plate ELISA assay was developed to quantify serum IgY responses specific to the Campylobacter heptasaccharide in which serum was tested against unglycosylated or glycosylated forms of CjaA_ChuA (9) as the capture antigen. Briefly, 96-well plates were coated with 50 μL of purified CjaA_ChuA (9) or unglycosylated CjaA_ChuA (9) and resuspended in a carbonate–bicarbonate buffer at a final concentration of 1 μg/mL and incubated at RT for 2 h. Wells were blocked with 200 μL of 5% BSA for 2 h at RT. While blocking, sera samples were diluted 1/100 in 1% BSA and pre-adsorbed with 4.5 μg/mL S. Typhimurium 82/6915 WCL and 7 μg/mL of purified unglycosylated CjaA_ChuA (9) for 2 h at RT to reduce any potential cross-reactivity to Salmonella and CjaA_ChuA (9), respectively. Plates were washed three times with 200 μL of PBST (0.05% Tween 20), and 100 μL of 1:100 pre-adsorbed sera were added to each well in triplicate, and plates were incubated at RT for 1.5 h. Control wells were maintained for both capture antigens to which no serum was added. Plates were washed as above, and 100 μL rabbit anti-chicken IgY (H + L)– HRP [1:10000] was added to each well. Plates were incubated at RT for 1.5 h, washed as above, and 50 μL of TMB substrate was added to the plates. Plates were incubated at RT for 10 min in the dark, and the reaction was stopped with the addition of 50 μL of 1 M hydrochloric acid to each well. The optical density at 450 nm was measured using a plate reader with background corrections to absorbance values using the control wells to which each respective antigen was added, but sera were excluded.

The distribution of data was evaluated using the Shapiro–Wilk normality tests. Statistical analyses of IgY titers and OD_{450 nm} readings between groups of S. Typhimurium 82/6915ΔaroA::pgl, STM-1, and mock-vaccinated chickens were performed in GraphPad Prism version 10.00 using a one-way ANOVA test with Dunnett’s multiple comparisons test. Data were represented as mean values with standard error of the mean. p values <0.05 were deemed to be statistically significant. Statistical analyses of IgY titres against S. Typhimurium between groups of S. Typhimurium 82/6915ΔaroA::pgl-vaccinated, STM-1-vaccinated, and mock-vaccinated chickens were performed in GraphPad Prism version 9.00 (GraphPad Software) using the Kruskal-Wallis test followed by Dunn’s multiple comparisons test. p values <0.05 were deemed to be statistically significant. Data were represented as median values with 95% confidence intervals.

To demonstrate that the Campylobacter heptasaccharide was expressed in S. Typhimurium 82/6915ΔaroA::pgl, lectin blotting was performed using the SBA lectin that specifically binds to the N-acetylgalactosamine (GalNAc) residues of the heptasaccharide. A functional pgl operon may permit the conjugation of the heptasaccharide to either protein at an asparagine residue within naturally occurring S. Typhimurium N-glycosylation sequons or to the core oligosaccharide of S. Typhimurium LPS. Incorporation as a part of the LPS can occur due to the relaxed specificity of S. Typhimurium O-antigen ligase, WaaL, toward saccharide substrates and the addition of these sugars to the lipid A core of LPS. Thus, the binding of SBA to S. Typhimurium 82/6915ΔaroA::pgl whole cell lysates and the equivalent extracts treated with proteinase K were investigated (Figure 2).

Samples treated with proteinase K resulted in the digestion of proteins (Figure 2A) as expected. Lectin blotting of proteinase K-treated and untreated samples of S. Typhimurium 82/6915ΔaroA::pgl produced a dominant signal at 15 kDa (Figure 2B), which was absent in the proteinase K-treated whole cell lysate of S. Typhimurium lacking the pgl locus (S. Typhimurium 82/6915). This indicated that a Campylobacter saccharide containing GalNAc residues was incorporated into a non-proteinaceous molecule within S. Typhimurium 82/6915ΔaroA::pgl, most probably LPS.

To confirm the 15 kDa signal corresponded to the modification of S. Typhimurium 82/6915ΔaroA::pgl LPS with a Campylobacter saccharide containing GalNAc residues, LPS was purified and analyzed (Figure 3).

S. Typhimurium 82/6915ΔaroA::pgl and STM-1 LPS were extracted from vaccine strains and visualized by silver staining (Figure 3A). However, silver staining did not indicate whether the Campylobacter saccharide had been linked to the S. Typhimurium 82/6915ΔaroA::pgl endogenous LPS and impacted the polysaccharide structure of the endogenous LPS, specifically the O-antigen region. Pro-Q Emerald 300-stained LPS extracts revealed that the addition of the Campylobacter saccharide had no impact on the polysaccharide region of S. Typhimurium 82/6915ΔaroA::pgl LPS, demonstrated by the presence of intact O-antigen repeating units attached to the core-oligosaccharide (Figure 3B). SBA lectin blotting produced a dominant signal at 15 kDa (Figure 3C) and confirmed the incorporation of a Campylobacter saccharide that contained at least one GalNAc residue as a part of S. Typhimurium 82/6915ΔaroA::pgl LPS. To better understand which domain of S. Typhimurium 82/6915ΔaroA::pgl LPS the Campylobacter saccharide was incorporated into, mild-acid hydrolysis of LPS was performed (Figure 3D). The 15 kDa SBA-positive band was sensitive to mild acid hydrolysis, which resulted in hydrolysis of the Kdo residue and the subsequent loss of the 15 kDa SBA signal. Loss of the SBA signal indicated that the Campylobacter saccharide was incorporated into S. Typhimurium 82/6915ΔaroA::pgl LPS.

To determine the structure of the S. Typhimurium 82/6915ΔaroA::pgl LPS-Campylobacter carbohydrate conjugate, the polysaccharide region of S. Typhimurium 82/6915ΔaroA::pgl LPS and unmodified STM-1 LPS were cleaved from the lipid-A core, filtered, and analyzed by LC–MS/MS. Analysis using LC–MS/MS detected a heptasaccharide species unique to S. Typhimurium 82/6915ΔaroA::pgl. Analysis of MS2 fragmentation spectra for this heptasaccharide confirmed that it is composed of a linear chain of six HexNAc residues with a single branching Hex attached to the 4th HexNAc from the reducing terminus (Figure 4). This structure is consistent with the well-characterized C. jejuni heptasaccharide. However, one sugar residue differed whereby the reducing end, which typically consists of diNAcBac, had been substituted with a HexNAc. Fragment series were annotated in concordance with the nomenclature established by Doman and Costello (56). This heptasaccharide structure was absent in STM-1 LPS (Supplementary Figure 1), which served as a negative control for the Campylobacter heptasaccharide modification.

These results indicated that the Campylobacter heptasaccharide was incorporated into S. Typhimurium 82/6915ΔaroA::pgl LPS onto the core-oligosaccharide region as depicted in Figure 5.

An animal trial was conducted to investigate the persistence/attenuation status and to identify and quantify the humoral immune responses induced following intramuscular injection of S. Typhimurium 82/6915ΔaroA::pgl. Groups (Table 1) of 15-layer hens were vaccinated via intramuscular injection at 7 (day 0) and 11 weeks (day 29) of age. To understand the capacity of vaccine strains to colonize specific sites of the host and the persistence at these sites, tissue and caecal content samples were taken, and vaccine strains were identified by culturing or DNA extraction followed by PCR analysis. At days two and three post-vaccination, S. Typhimurium 82/6915ΔaroA::pgl was re-isolated from the injection site (breast muscle) of all five birds by culturing and was detected by PCR in group 3 vaccinated birds (Table 3). On day two post-vaccination, STM-1 was detected in the breast muscle of four out of five birds in group two, but by day three, this had reduced to three out of five birds re-isolated by culturing; however, it was detected in all birds by PCR. On day two post-vaccination, neither vaccine strain was detected in necropsied birds’ caecal content, liver, or spleen (Table 4). On day three post-vaccination, STM-1 was detected by PCR in the caecal content of two birds from the STM-1 vaccinated group, whereas S. Typhimurium 82/6915ΔaroA::pgl was not detected in the caecal content of any birds (Table 4). Neither vaccine strain was detected in the liver or spleen of necropsied birds on days two and three post-vaccination (Table 5). At day 42, neither vaccine strain nor any tissue samples or caecal content were isolated. In addition, vaccines were not detected by PCR.

To determine if vaccination with S. Typhimurium 82/6915ΔaroA::pgl produced an antigen-specific IgY immune response to the Campylobacter heptasaccharide, pooled chicken sera taken at day 42 from birds vaccinated twice with S. Typhimurium 82/6915ΔaroA::pgl were analyzed by immunoblotting against purified S. Typhimurium 82/6915ΔaroA::pgl LPS altered at the core-oligosaccharide with the Campylobacter heptasaccharide. IgY antibodies in the pooled sera bound to the LPS-Campylobacter heptasaccharide glycoconjugate produced an immunodominant signal at 15 kDa (Figure 7). This signal was absent in pooled sera from day 42 unvaccinated birds, which indicated the immunodominant band corresponded to a Campylobacter heptasaccharide-specific IgY immune response.

To quantify whether the IgY responses specific to the Campylobacter heptasaccharide were significantly different from those of mock and STM-1 vaccinated chickens, a serological ELISA was performed using a purified Campylobacter glycoprotein, CjaA_ChuA (9), and its unglycosylated equivalent as capture antigens (Supplementary Figure 2) to compensate for any cross-reactivity directed against the protein. Campylobacter heptasaccharide-specific immune responses were quantified by ELISA in sera taken on day zero prior to vaccination and days 35 and 42 post-secondary vaccination. Primary vaccination produced no significant increases in IgY responses to the heptasaccharide. However, significantly elevated Campylobacter heptasaccharide-specific IgY antibodies were detected in S. Typhimurium 82/6915ΔaroA::pgl-vaccinated birds (Figure 8D) post-secondary vaccination at day 42 compared to mock-vaccinated and STM-1-vaccinated chickens (Figure 8D). Humoral immune responses were confirmed to be specific to the heptasaccharide, as low binding was detected in sera probed against the unglycosylated protein in all groups (Figure 8C). Prior to the first vaccination, elevated levels of IgY (absorbance greater than one) were detected in two birds (36 and 40) from the mock-vaccinated control group (Figure 8A) and suggested some of these birds were pre-exposed to Campylobacter, which is supported by the decreased levels of IgY titres at days 35 and 42. However, no significant differences were observed among the mock, STM-1, and S. Typhimurium 82/6915ΔaroA::pgl groups before vaccination. Longitudinal analysis of humoral responses in S. Typhimurium 82/6915ΔaroA::pgl-vaccinated birds (Figure 8B) demonstrated IgY-specific responses to the heptasaccharide that peaked at day 35, a week post-secondary vaccination.

While a statistically significant difference in anti-Campylobacter heptasaccharide IgY antibodies was observed between S. Typhimurium 82/6915ΔaroA::pgl and both the STM-1 and mock-vaccinated groups, inter-animal variability was observed where some S. Typhimurium 82/6915ΔaroA::pgl vaccinated birds responded weakly to vaccination (e.g., B81) compared to others (B84 and B85).

To determine if modification of S. Typhimurium 82/6915ΔaroA::pgl endogenous LPS core oligosaccharide with the addition of the Campylobacter heptasaccharide impacted immune responses against S. Typhimurium, IgY responses specific to S. Typhimurium were quantified using ELISA (Figure 9). The primary IgY antibody responses to S. Typhimurium were weak after the first vaccination and gradually increased in S. Typhimurium 82/6915ΔaroA::pgl, and STM-1 vaccinated birds at days seven to 29 relative to mock vaccinated birds. Two weeks (day 14) post-primary vaccination, IgY titres in STM-1 vaccinated birds were significantly higher than in the mock-vaccinated group. Post-secondary vaccination, individual bird antibody titres increased in both S. Typhimurium 82/6915ΔaroA::pgl and STM-1 vaccinated birds. No detectable anti-S. Typhimurium-specific IgY titres were present in mock vaccinated birds. All STM-1 vaccinated birds produced IgY antibody titres above the cutoff of 654, with any values below this deemed non-responsive. One bird from the S. Typhimurium 82/6915ΔaroA::pgl vaccinated groups had antibody titres below the cutoff value, indicating no production of anti-S. Typhimurium antibodies. However, antibody titres in chicken sera against S. Typhimurium showed no significant differences between the STM-1 and S. Typhimurium 82/6915ΔaroA::pgl groups on days 14, 29, 35, and 42 (Figure 9).