The bacterial strains and plasmids used are listed in Table 1. Y. enterocolitica ATCC 23715 Amp (wild-type, WT) was cultivated in Luria-Bertani (LB) medium. Escherichia coli (E. coli) strains were cultured in LB medium. Y. enterocolitica strains were cultured at 26 °C. E. coli strains were cultured at 37 °C. The concentrations of antibiotics used in certain experiments were as follows: 20 μg/mL of chloramphenicol (Cm) and 100 μg/mL of ampicillin (Amp) in the experiment.

SacB-based allelic exchange was used to create the in-frame deletion mutant as previously described (Ma et al., 2021). The primers sodA-UP-F/sodA-UP-R and sodA-DOWN-F/sodA-DOWN-R were used to amplify the upstream and downstream regions around the sodA gene (Table 1). The primers sodA-UP-F/sodA-DOWN-R were used to splice the PCR products using overlap extension-PCR. Linearized pDM4 plasmid and the purified sodA-UD fragment were ligated using the Gibson assembly method. After preparing the reaction system according to the protocol, the mixture was incubated at 50 °C for 1 h to facilitate ligation and fusion. The resulting fusion product was cloned into pDM4 and digested with Xho I to obtain the recombinant expression vector pDM4-sodA. Following purification, the plasmid was transformed into E. coli S17-1 (λpir) by heat shock and transferred into the WT. Single crossover recombination occurred on LB agar plates supplemented with Cm, and double crossover recombination events occurred on LB agar plates supplemented with 10% sucrose. Ultimately, the ΔsodA mutant was verified using the primers sodA-IN-F/ sodA-IN-R and sodA-UP-F/sodA-DOWN-R. The primers sodA-Hind III-F and sodA-BamH I-R were used to amplify the sodA gene and its promoter region by PCR (Supplementary Table S1). The PCR product was inserted into Hind III and BamH I double-digested pACYC184, resulting in the complemented plasmid pACYC184-sodA. This plasmid was transferred into ΔsodA, and the complemented isolates were selected on LB agar plate containing Cm and Amp. Eventually, the complemented strain ΔsodA^C was confirmed using the primers sodA-IN-F and sodA-IN-R (Supplementary Figure S1).

The effect of sodA deletion on acid tolerance and oxidative stress resistance were evaluated based on the method of Ma et al. (2021) with minor modifications. Overnight bacterial cultures of WT, ΔsodA, and ΔsodA^C strains were diluted to 1 × 10⁷ CFU/mL using LB broth. 300 μL bacterial suspension was inoculated into 30 mL LB broth (pH 4.0) and incubated at 26 °C for 1 h. Serial dilutions were plated on LB agar, with viable counts enumerated after 48 h incubation at 26 °C. Cell suspensions were adjusted to OD₆₀₀ = 0.5 in LB broth. Aliquots (100 μL) were lawn-cultured on LB agar. Sterile Oxford cups were placed on agar, filled with 200 μL 10% (v/v) H₂O₂, and incubated 24 h at 26 °C. Inhibition zone diameters were measured with vernier calipers. Bacterial suspensions (OD₆₀₀ = 0.5) were diluted 1:100 in LB broth containing 5 mM H₂O₂. After 1 h incubation at 26 °C, cells were serially diluted in PBS and viable counts determined by spread plating.

This study aimed to comprehensively assess sodA-dependent regulation of biofilm formation through quantitative biomass, metabolic activity, and architectural analyses as previously described (Bai et al., 2019). Overnight cultures of WT, ΔsodA, and ΔsodA^C strains were adjusted to OD₆₀₀ = 0.5. Aliquots (200 μL) were dispensed into 96-well polystyrene plates (triplicate wells per strain), with sterile LB as blank control. After 48 h static incubation at 26 °C, planktonic growth was measured at OD₆₃₀. Biofilm biomass was quantified via crystal violet staining (OD₅₇₀) to calculate Specific Biofilm Formation (SBF) index using the formula: SBF = (OD_{570 nm}−OD_control)/(OD_{630 nm}−OD_control). Biofilms grown as above were incubated with 0.5 mg/mL MTT (200 μL/well) for 4 h at 26 °C. Formazan crystals were dissolved in DMSO (150 μL/well) with gentle agitation. Metabolic activity was measured at OD₄₉₀. Biofilms were formed on glass coverslips (Ø10 mm) in 24-well plates containing 2 mL bacterial suspension (OD₆₀₀ = 0.5). After 48 h at 26 °C, coverslips were crystal violet-stained (0.4%, 20 min), washed, air-dried, and examined at 400 × magnification. Then fixed in 2.5% glutaraldehyde (4 °C overnight), dehydrated in ethanol gradient (30–100%), critical-point dried, gold-sputtered, and imaged by FESEM at 4,000×.

Caco-2 cells were cultured in DMEM supplemented with 10% FBS, 1% non-essential amino acids, and 1% antibiotic-antimycotic solution at 37 °C with 5% CO₂. Once 85% confluent, cells were washed with PBS, digested with 0.25% trypsin–EDTA, and passaged at a 1:5 ratio into new flasks. The adhesion and invasion assays for Caco-2 cells were performed with reference to the method of Medeiros et al. (2021) with minor modifications. Briefly, Caco-2 cells were digested and diluted to 1 × 10⁵ cells/mL, then seeded into 24-well plates and incubated overnight. Overnight bacterial cultures of WT, ΔsodA, and ΔsodA^C strains were diluted to 1 × 10⁷ CFU/mL using DMEM. After gently washing the Caco-2 cells, 1 mL of bacterial suspension was added to each well, followed by centrifugation (600 × g, 5 min). Samples were incubated at 37 °C with 5% CO₂ for 2 h, after which the supernatant was discarded, and the cells were washed with sterile PBS.

For the adhesion assay, 0.1% (v/v) Triton X-100 was added to each well to lyse the cells at 4 °C for 20 min. The bacterial suspension was pipetted to homogenize, serially diluted, and plated. Colony counts were recorded after incubation at 26 °C for 48 h. For the invasion assay, 1 mL of medium containing gentamicin (100 μg/mL) was added to each well and incubated at 37 °C with 5% CO₂ for 45 min. After incubation, the cells were washed, lysed, and plated to quantify the number of invasive bacteria.

The survival and proliferation abilities of Y. enterocolitica in RAW264.7 macrophages after sodA gene deletion were evaluated based on the method of Jin et al. (2024) with minor modifications. The RAW264.7 cell culture process was similar to that described in Section 2.3, except the digestion time was 3.5 min. After digestion, complete medium was added to halt digestion, and the cell suspension was passaged at a ratio of 1:5 into new flasks. RAW264.7 cells were cultured similarly to Caco-2 cells. RAW264.7 cells were seeded in 24-well plates at 1 × 10⁵ cells/mL. Overnight bacterial cultures of WT, ΔsodA, and ΔsodA^C strains (~1 × 10⁷ CFU/mL) were added to each well, centrifuged (600 × g, 5 min), and incubated at 37 °C with 5% CO₂ for 45 min. The supernatant was then discarded, and the cells were washed once with PBS.

For the survival assay, 1 mL of medium containing gentamicin (100 μg/mL) was added to each well, and the samples were incubated at 37 °C with 5% CO₂ for 30 min. After incubation, the cells were lysed using 0.1% Triton X-100, and bacterial colonies were quantified by plating. For the proliferation assay, 1 mL of medium containing gentamicin (10 μg/mL) was added to each well and incubated for 24, 48, and 72 h. Samples were collected every 24 h, washed, lysed, and plated to quantify bacterial proliferation.

Sixty specific pathogen-free (SPF) female BALB/c mice (6 weeks old) were used in this study. The animal experiments in this study were conducted in strict accordance with the Guide for the Care and Use of Laboratory Animals (8th edition, ISBN-10: 0-309-15396-4) and the Laboratory Animal Management Regulations of Northwest A&F University. The mice were housed in a controlled environment at 25 °C with free access to food and water. The housing conditions were kept dry and clean, with bedding changed every 2 days, and five mice were housed per cage.

The mice were randomly divided into four groups: control (Control), wild-type strain infection (WT), sodA deletion strain infection (ΔsodA), and sodA complemented strain infection (ΔsodA^C). Based on preliminary experiments, the minimum infectious dose of Y. enterocolitica ATCC 23715 causing clinical symptoms in BALB/c mice was determined to be approximately 1 × 10⁹ CFU. For this study, the infection dose was set at 4 × 10⁹ CFU. WT, ΔsodA, and ΔsodA^C bacterial suspensions were prepared as described in Section 2.1, and the concentration was adjusted to 4 × 10¹⁰ CFU/mL. After a 7-day acclimation period, the mice were orally gavaged with 100 μL of the bacterial suspensions (WT, ΔsodA, or ΔsodA^C). The control group received 100 μL of sterile PBS. Mice were weighed daily after infection. On day 5 post-infection, mice were anesthetized with 3.5% chloral hydrate via intraperitoneal injection, euthanized by cervical dislocation, and dissected using sterile instruments. Liver, spleen, kidneys, and intestinal tissues (from duodenum to rectum) were collected. The tissues were gently washed with pre-cooled sterile saline and fixed in 4% paraformaldehyde for at least 24 h. Remaining tissues were snap-frozen in liquid nitrogen and stored at −80 °C for further analysis.

Fecal samples were collected at 0, 1, 2, 3, 4 and 5 days post-infection, weighed, and subjected to tenfold serial dilution in PBS. A 100 μL aliquot of each dilution was plated on CIN-1 agar and incubated at 26 °C for 24 h. On day 5 post-infection, mice were euthanized as described above. The kidneys, liver, spleen, ileum, and colon were harvested, washed with pre-cooled PBS, and blotted dry with filter paper. Samples were weighed and placed into pre-cooled microcentrifuge tubes. Tissue homogenates were prepared using a tissue homogenizer with PBS based on the weight of the sample. Homogenates were serially diluted in PBS, and 100 μL of each dilution was plated on CIN-1 agar and incubated at 26 °C for 24 h. The number of characteristic colonies on CIN-1 agar was recorded and subjected to statistical analysis.

The tissues for histological examinations were fixed overnight with 4% paraformaldehyde overnight and paraffin- embedded. After sliced into 3 μm sections, the tissue sections were then dewaxed and stained with hematoxylin and eosin (H&E; Solarbio) and analyzed by a light microscope (Leica, Wetzlar, Germany).

The level of NF-κB p65 in the ileum was assessed using immunohistochemical staining (Jin et al., 2022). Briefly, paraffin sections were treated with dewaxing, antigen retrieval, and blocking endogenous peroxidase, followed by a 30 min incubation with 10% rabbit serum for blocking. The sections were incubated with the primary antibody at 4 °C overnight followed by a 50 min incubation with the secondary antibody, and they were finally stained with DAB and counterstained with hematoxylin. The results were observed by fluorescence microscopy (Leica, Wetzlar, Germany), and the mean optical density was evaluated by Image J (Version 1.8.0.112).

Total RNA was extracted by a Steady Pure RNA extraction kit (AG, Changsha, China), followed by reverse transcription into cDNA by an Evo M-MLV reverse transcription kit. The qRT-PCR analysis was conducted using an IQ5 system (Bio-Rad). The primer sequence, along with GAPDH as the reference gene, is presented in Supplementary Table S2.

The data are presented as means ± SD. SPSS 19.0 was used for one-way analysis of variance. The data were analyzed using one-way ANOVA. *: p < 0.05; **: p < 0.01. #: p < 0.05; ##: p < 0.01; §: p < 0.05; §§: p < 0.01. Primer design was performed using Primer 5.0 software.

Genomic DNA from Y. enterocolitica ATCC 23715 was used to amplify the upstream and downstream homologous arms of the sodA gene, yielding 947 bp and 945 bp fragments, respectively, confirmed by agarose gel electrophoresis. The fused homologous arm fragment (1892 bp) was successfully assembled into the XhoI-digested pDM4 vector using Gibson assembly and transformed into E. coli S17-1 (λpir). Positive clones were identified by PCR, confirming the construction of the recombinant plasmid pDM4-sodA. To verify single colonies on sucrose plates, sodA-UP-F/sodA-DOWN-R primers and sodA-IN-F/R primers were utilized. The WT U + sodA+D fragment measured 2,407 bp, while the deletion strain U + D fragment was 1892 bp. PCR amplification was detected for the deletion strain, confirming the successful knockout of the sodA gene using the sucrose-sensitive suicide vector pDM4. The sodA gene fragment, including its promoter, was subsequently cloned into the HindIII and BamHI-digested pACYC184 vector through double digestion. PCR amplification with sodA-HindIII-F/sodA-BamHI-R primers yielded a 2,409 bp fragment. The positive recombinant strain successfully amplified a 1734 bp band, consistent with the expected result, indicating successful construction of the recombinant plasmid pACYC184-sodA. Complementation was verified by PCR screening using sodA-IN-F/R primers. Both the WT and complementation strains produced a 601 bp sodA fragment, confirming the successful construction of the sodA complementation strain.

In agar well diffusion assays (Figure 1A), ΔsodA displayed a 23.3% larger inhibition zone (2.33 ± 0.07 mm vs. WT: 1.89 ± 0.05 mm; p < 0.01) after 24 h incubation at 26 °C, indicating heightened H₂O₂ sensitivity. The complemented strain (ΔsodA^C) showed no significant difference from WT (p > 0.05). As shown in Figure 1B, all strains exhibited significantly reduced viability under 5 mM H₂O₂ exposure. While ΔsodA showed comparable survival to WT and complemented ΔsodA^C at 5 min (p > 0.05), it demonstrated progressively diminished viability at extended timepoints (10, 15, 20, and 30 min), with statistically significant reductions versus controls (p < 0.05). These data establish that sodA deletion compromises oxidative stress tolerance in Y. enterocolitica, confirming its essential role in ROS detoxification.

Figure 2A demonstrates that sodA deletion significantly impaired biofilm biomass formation (p < 0.01). The ΔsodA mutant exhibited 29.37% reduction in crystal violet-stained biomass, while ΔsodA^C partially restored this phenotype. MTT metabolic assays (Figure 2B) revealed severely compromised biofilm viability in ΔsodA (48.09% reduction vs. WT; p < 0.01). ΔsodA^C showed significantly higher metabolic activity than ΔsodA (p < 0.01), indicating sodA-mediated regulation of biofilm metabolic processes. Morphological analyses corroborated the quantitative biofilm defects: wild-type (WT) biofilms developed dense, stratified architectures with robust surface adherence (Figures 2C,F), whereas the ΔsodA mutant formed sparse monolayers exhibiting compromised structural cohesion (Figures 2D,G). ΔsodA^C restored the biofilm morphology, yielding consolidated multilayered structures (Figures 2E,H). This visual evidence from both light and field-emission scanning electron microscopy consistently demonstrates sodA essential role in maintaining biofilm architectural integrity.

Y. enterocolitica initiates infection through adhesion to and invasion of host cells. As shown in Figure 3A, the adhesion rates of the ΔsodA and ΔsodA^C strains to Caco-2 cells were reduced to 55.14 and 88.29%, respectively, compared to the WT strain. Upon complementation of the sodA gene, the adhesion ability of Y. enterocolitica was significantly restored (p < 0.01). Additionally, sodA deletion suppressed the bacterial invasion of Caco-2 cells, with invasion rates decreasing to 75.84 and 87.57% (Figure 3B). These results indicate that the sodA gene plays a critical role in the adhesion and invasion of host cells by Y. enterocolitica.

RAW264.7 cells, a mouse monocyte–macrophage leukemia cell line derived from BALB/c mice, were used to evaluate intracellular survival. As shown in Figure 3C, the ΔsodA strain exhibited higher phagocytosis sensitivity in RAW264.7 macrophages compared to the WT and ΔsodA^C strains.

Figures 4A,B illustrate body weight changes in mice over 5 days of Y. enterocolitica infection. Control mice administered sterile PBS showed continuous weight gain, whereas significant weight loss was observed in the WT-infected group on day 2 (p < 0.01). Similarly, mice infected with ΔsodA and ΔsodA^C strains also showed notable weight reduction (p < 0.05). Over the five-day period, weight loss was −2.84 ± 0.29 g, −1.99 ± 0.43 g, and −2.63 ± 0.22 g in the WT, ΔsodA, and ΔsodA^C groups, respectively, with significantly greater weight loss in the WT group compared to the ΔsodA group (p < 0.01).

Using CIN-1 selective media, Y. enterocolitica colonies were quantified in various tissues of mice (Figures 4C,D). No colonies were detected in the tissues of control mice. Infected groups showed low bacterial colonization in the liver and spleen. The WT and ΔsodA^C groups exhibited significantly higher colonization in ileum and colon tissues (~400 CFU/mg, p < 0.01) compared to ΔsodA (179 CFU/mg and 258 CFU/mg, respectively). Fecal bacterial counts after 5 days of infection (Figure 4E) were significantly lower in the ΔsodA group compared to the WT and ΔsodA^C groups (p < 0.01). These findings suggest that sodA gene deletion reduces the number of Y. enterocolitica colonization in the mouse gut.

H&E staining results of mouse tissues are shown in Figure 5. Control group liver tissue displayed normal cell morphology, while the WT group exhibited severe shrinkage and nuclear chromatin condensation. Mice infected with ΔsodA strains showed slight shrinkage with otherwise normal morphology, while those infected with ΔsodA^C strains displayed moderate shrinkage. Similar trends were observed in kidney tissue, with significant tubular swelling and glomerular damage in WT and ΔsodA^C groups. Splenic tissue from infected groups showed increased immune cells in red pulp and macrophage proliferation. At 200 × magnification, in the Control group, the intestinal villi were neatly arranged, with intact glandular structures, and no significant pathological changes such as villus shedding or lamina propria edema were observed. The WT group exhibited the most severe lesions, characterized by sparse and fractured villi in the ileum, nearly absent crypts, a thinned intestinal wall, pronounced congestion, and inflammatory cell infiltration. In the colon, there was separation of the mucosal and lamina propria layers, severe damage to villus and crypt structures, and significant glandular injury accompanied by inflammatory cell infiltration. The ΔsodA group showed relatively intact intestinal villi, although they were sparse; the overall tissue morphology was more preserved compared to the WT group. The ΔsodA^C group exhibited lesions more severe than the ΔsodA group, with partial destruction of villus structures, but less severe than the WT group.

Immunohistochemical staining assessed NF-κB p65 expression in colon tissues (Figure 6A), with quantitative analysis using ImageJ (Figure 6B). Positive staining appeared as yellow or brown spots. WT, ΔsodA, and ΔsodA^C groups showed significantly more brown spots than the control group, indicating enhanced inflammatory responses. Relative gray value analysis confirmed increased NF-κB p65 levels in infected groups (p < 0.01). The ΔsodA group exhibited lower NF-κB p65 expression compared to WT (p < 0.01). This suggests that the sodA gene deletion attenuates NF-κB pathway activation during Y. enterocolitica infection.

The transcriptional levels of immune-related genes in mouse colons are shown in Figure 6C. The control group displayed low expression of inflammatory cytokines, whereas infection significantly upregulated genes such as IL-1β, TNF-α, NF-κB p65, INF-γ, and IL-10 (p < 0.01). The ΔsodA group showed reduced transcriptional levels of IL-1β, TNF-α, IL-6, TLR4, INF-γ, and IL-10 compared to WT (p < 0.01). No significant differences were observed in IL-6, iNOS, or TLR4 expression between ΔsodA and the control (p > 0.05). The ΔsodA^C group displayed higher transcriptional levels than ΔsodA, with TLR4 and INF-γ being particularly elevated, reaching 4.74 and 5.24 times the control, respectively.