The experiments were approved by the Institutional Animal Care and Use Committee of Yangtze University (Approval No. 202401003).

Bacillus coagulans strain MF-06 was previously isolated and identified in our laboratory. Briefly, the strain was isolated from the intestinal contents of chicks and then identified through physiological, biochemical, and molecular biological analyses. The dominant B. coagulans strains MF-06 were selected based on the preliminary results of physiological, biochemical, and in vitro antibacterial experiments (Lee et al., 2014). The isolated probiotics were prepared into freeze-dried powder, which was modified according to Chen et al. (2020) the research method. The glycerin storage tubes containing strains were inoculated with 2% inoculate into YPD medium and activated for three generations. Strains were activated with 2% inoculum to 100 mL YPD medium at 37°C for 16 h. Then, 2% of the activated strains were inoculated into 2.5 L YPD medium and incubated in a 5 L bioreactor (FR-P series; Shanghai Jichen Biotechnology Co., LTD., Shanghai) in the high-density fermentation, 37°C for 16 h. The bacterial sludge was centrifuged at 4°C for 20 min at 8000 rpm and collected. The bacteria pellet was resuspended in 250 mL of sterilized degreased milk powder protectant, mixed, and frozen at −80°C for 12—14 h. The frozen samples were then freeze-dried under vacuum (ZLGJ-18 multi-manifold model, Huachen Instrument Co., LTD., Zhengzhou, China) at −50°C for 48 h. This freeze-dried powder was the final probiotic. Lyophilized powder was prepared and viable counts per gram of B. coagulans were determined by the plate coating method. The samples were then stored at 4°C until use. Neomycin sulfate-soluble powder was purchased from Inner Mongolia Huatian Pharmaceutical Co. Ltd. (Inner Mongolia, China).

The S. pullorum strain (CVCC No. 534) was obtained from the China Veterinary Microbial Strain Preservation and Management Center in Beijing, China. The species of bacteria were preserved in 20% glycerol at −80°C. Bacterial viability was measured by the plate smear method, and the corresponding Optical density (OD) values were measured to plot the growth curve of S. pullorum. Bacterial suspensions were prepared after determined the concentrations.

A total of 150 one-day-old SPF chicks (Jinghong No.1) were Hubei Yukou Poultry Industry Co., Ltd., Hubei. They selected and randomly divided into five groups: CK group: control group, which was fed with basic diet with no administration of antibiotics or probiotics. EM group: probiotics group, whose basal diet was added with 1.0 × 10⁸ CFU/g B. coagulans MF-06; PT group: probiotics treatment group, whose basal diet was added with 1.0 × 10⁸ CFU/g B. coagulans MF-06 from days 1 to 14, and when chicks were seven-day-old, 1.0 × 10⁹ CFU/0.5 mL of S. pullorum was administered by gavage; AT group: antibiotic treatment group, for whom basal diet was provided for the first 7 days, and when chicks were seven-day-old, they administered with 1.0 × 10⁹ CFU/0.5 mL of S. pullorum by gavage, and the basal diet supplemented with 0.375 g/kg neomycin sulfate from days 7 to 14. SI group: S. pullorum infection group, which was fed with basic diet, and when chicks were seven-day-old, they were administered with 1.0 × 10⁹ CFU/0.5 mL of S. pullorum by gavage (Deng et al., 2021); Each group comprised six replicates, with each replicate consisting of five chicks. The added dose of B. coagulans was obtained from a preliminary dose exploration experiment. The trial design is shown in Supplementary Figure S1, the chicks in the CK and EM groups were accommodated in an independent house in case of cross-contamination and humidity, and chicks in each group were housed in wire cages (110 × 80 × 45 cm). All chicks tested negative for pathogens before the experiment and the chicken house was thoroughly sterilized before the experiment. During the 14-day experiment, all chicks were allowed unrestricted access to water and fed with a starter feed devoid of antibiotics. All feeds were ground into powders and stored in a low-temperature dry environment, the basal diet composition and basic feed composition are presented in Table 1. At 9 am, the time of the first feeding and water change, chickens were observed for any abnormalities. The second feeding and water change was at 3:00 pm, when the chickens were observed again. The amount of feed remaining was recorded daily and the enclosure was cleaned regularly. The same rearing conditions were used, and the chick chambers were maintained at 34–36°C and a 60% relative humidity from day 1 to day 3. The temperature was decreased by 1°C each day from day 4 to day 14 and maintained at 31–33°C. The chambers were illuminated for 24 h on days 1 to day 3, for 22 h on days 4 to day 7, and for 20 h on days 8 to day 14. The growth status, body weight, survival rate, and clinical symptoms were recorded daily.

On the 14th day, two chicks were randomly selected from each replicate group for a total of six chicks per group. Two mL of blood sample was collected from the vein under the wing and stored in procoagulant tubes, followed by incubation at room temperature for 2 h and centrifugation at 3,000 g and 4°C for 15 min for separation of the serum, which was then collected and stored at −80°C for subsequent analysis. The chicks were euthanized under anesthesia by cervical dislocation, and then sample was aseptically collected from the duodenum, jejunum, ileum, and cecum and stored at 4°C for quantifying the bacterial load. To perform 16S rRNA sequencing analysis, 0.5 g of the remaining cecal content was collected, frozen in liquid nitrogen, and stored at −80°C. The jejunum (50 mg) was aseptically collected and immersed in an RNA sample preservation solution, flash-frozen with liquid nitrogen, and stored at −80°C for total RNA isolation. Additionally, Midsections (3—5 cm) of the jejunum were collected and fixed in a 4% paraformaldehyde solution for subsequent immunohistochemical analysis.

The D-lactic acid (D-LA) level in the serum was assessed with a commercially available ELISA kit (Shanghai Enzyme Linked Biology Co., Ltd., Shanghai, China); The diamine oxidase (DAO) level in the serum was measured with a commercially available kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China).

The bacterial load in tissues was determined according to the previous method with some modifications (Li et al., 2024), the tissue specimens of duodenum, jejunum, ileum, and cecum (100 mg) after sterile collection were placed in 2 mL sterile centrifuge tube. Subsequently, 900 μL of PBS buffer was added, and the mixture was thoroughly homogenized. Following gradient dilution (10⁻²–10⁻⁸), all samples were inoculated on bismuth sulfite agar; the process six experiments were repeated, with each replicate conducted three times, and then the samples were cultured at 37°C for 18–24 h. The morphological characteristics of the colonies were used to quantify S. pullorum in the intestinal tissue. S. pullorum counts, expressed as logarithmic transformation (log₁₀), were reported as colony-forming units per 0.1 g (CFU/0.1 g).

Before the chicks were dissected, body weight was recorded, duodenum, jejunum, and ileum were isolated, and the lengths of the corresponding parts were measured. Three parallel biological replicates were used for all procedures and measurements.

Small segments of mid-jejunum were fixed with 4% paraformaldehyde solution and dehydrated through an ascending ethanol gradient. The segments were cleaned with xylene and embedded in paraffin. Next, the samples were sectioned at 5 μm for immunohistochemical analysis of Proliferating Cell Nuclear Antigen (PCNA) and v-myc avian myelocytomatosis viral oncogene homolog (C-MYC), the immunohistochemistry was conducted following the method by previous research (Shen et al., 2024). The sections were dewaxed with xylene, anhydrous ethanol, distilled water and incubated with 3% H₂O₂ for 20 min at room temperature (25 ± 2°C) to restrict the endogenous peroxidase activity. Then, the citric acid buffer (G1202 Servicbio Co., Ltd., Wuhan, China) was used to restore antigen activity for 10 min. Next, the sections were immersed in 3% Bovine serum BSA (GC305010, Servicbio Co., Ltd., Wuhan, China) for 30 min at room temperature. Gently shake off the sealing solution, add PBS to the section in 1:500 proportion of primary antibody (Anti -PCNA Rabbit pAb and Anti -C-MYC Rabbit pAb), and the section is placed flat in a wet box at 4°C for overnight incubation. The slide was placed in PBS (pH 7.4) and washed by shaking on the decolorizing shaker for three times, 5 min each time. Next, the sections were incubated with biotinylated goat anti-rabbit antibody (1:200, GB23303 Servicbio Co., Ltd., Wuhan, China) at room temperature for 50 min, and washed with PBS (pH 7.4) for three times. After the sections were slightly dried, the freshly prepared DAB color developing solution (G1212, Servicbio Co., Ltd., Wuhan, China) was added into the circle, the color developing time was controlled under the microscope. The positive color was brown and yellow, and the sections were rinsed with tap water to terminate the color development. Next, hematoxylin restaining for about 3 min, washing with tap water, hematoxylin differentiation solution for a few seconds, rinse with tap water, hematoxylin return to blue solution, and rinse with running water. Then were conventionally dehydrated to transparency and sealed using neutral balsam. Four sections were randomly selected for each group and chose three different regions of each section. The average optical density of immune positive substances in the intestine was measured using ImageJ software (version 1.8).

Total RNA was extracted from jejunum samples using the TRIzol™ Reagent Kit (Thermo Fisher Scientific, Waltham, Massachusetts, USA) according to the manufacturer’s protocol, and the RNA concentration was measured using a microspectrophotometer (nanodrop2000, Thermo Fisher Scientific, Massachusetts). cDNA was synthesized from the RNA using the Color Reverse Transcription Kit with gDNA Remover (EZBioscience, USA). All gene expression levels were normalized with β-actin as the internal standard. PCR primer sequences (Table 2) were designed and synthesized by Shanghai SANGON bio (Shanghai, China). SYBR Green fast qPCR mix (RK21203; Abclonal, Wuhan, China) was used with the following program: 95°C for 3 min, 30 cycles of 95°C for 5 s, 60°C for 30 s. This experiment was performed in six biological replicates with three replicates each. The quantitative RT-PCR (RT-qPCR) assay was performed to analyze the relative mRNA levels of the targeted genes using the 2^-△△CT method as previously described (Livak and Schmittgen, 2001).

For each group, samples of cecal content were randomly selected for full-length 16S rRNA sequencing. Total genome DNA from samples was extracted using CTAB method. DNA concentration and purity was monitored on 1% agarose gels. According to the concentration, DNA was diluted to 1 ng/L using sterile water (Jiali et al., 2024). 16S rRNA genes of regions (V3-V4/16S) were amplified used specific primer (F: 5′-CCTAYGGGRBGCASCAG-3′, R: 5’-GGACTACNNGGGTATCTAAT-3′). All PCR reactions were carried out with 15 L of Phusion® High-Fidelity PCR Master Mix (New England Biolabs); 2 M of forward and reverse primers, and about 10 ng template DNA. Thermal cycling consisted of initial denaturation at 98°C for 1 min, followed by 30 cycles of denaturation at 98°C for 10 s, annealing at 50°C for 30 s, and elongation at 72°C for 30 s. Finally, 72°C for 5 min. Mix same volume of 1X loading buffer (contained SYB green) with PCR products and operate electrophoresis on 2% agarose gel for detection. PCR products was mixed in equidensity ratios. Then, mixture PCR products was purified with Qiagen Gel Extraction Kit (Qiagen, Germany). Then, Sequencing libraries were generated using TruSeq® DNA PCR-Free Sample Preparation Kit (Illumina, USA) following manufacturer’s recommendations and index codes were added. The library quality was assessed on the Qubit@ 2.0 Fluorometer (Thermo Scientific) and Agilent Bioanalyzer 2,100 system. At last, the library was sequenced on an Illumina NovaSeq platform and 250 bp paired-end reads were generated.

Paired-end reads was assigned to samples based on their unique barcode and truncated by cutting off the barcode and primer sequence. Quality filtering on the raw tags were performed under specific filtering conditions to obtain the high-quality clean tags according to the fast (v0.22.0).¹ Paired-end reads were merged using FLASH (v1.2.11),² The tags were compared with the reference database³ using UCHIME Algorithm⁴ to detect chimera sequences, and then the chimera sequences were removed. Then the Effective Tags finally obtained. Sequences analysis was performed by Uparse software (Uparse v7.0.1001).⁵ Sequences with 97% similarity were assigned to the same OTUs. Representative sequence for each OTU was screened for further annotation. Amplicon sequence variant (ASV) was analyzed by Deblur, which uses error profiles to obtain putative error-free sequences from Illumina MiSeq and HiSeq sequencing platforms. For each representative sequence, the Silva Database⁶ was used based on Mothur algorithm to annotate taxonomic information. OTUs abundance information were normalized using a standard of sequence number corresponding to the sample with the least sequences. Alpha diversity is applied in analyzing complexity of species diversity for a sample through 6 indices, including Observed-species, Chao1, Shannon, Simpson, ACE, Good-coverage. All these indices in our samples were calculated with QIIME and displayed with R software (Version 4.1.2). To further investigate the differences in community structure and variations in species composition among grouped samples, we performed Student t-tests and LEfSe analysis.

In this study, Kruskal-Wallis test was used to compare the differences of different microbial communities in each group, Because the data did not conform to the assumption of normal distribution, we chose to use nonparametric test methods to analyze differences in microbial diversity or abundance between groups. If the test results showed a significant difference (p < 0.05), a further Dunn’s multiple comparison test was performed to determine which groups were significantly different. Growth performance was analyzed by a Mixed Effects Model using SPSS software (version 25.0, SPSS Inc., Chicago, IL, USA), and other data, such as bacterial load, serum indicators and gene expression, were analyzed by one-way ANOVA significant differences between means were compared using Duncan’s multiple range test. Statistical significance of differences was expressed by the presence of a-d letters within a column; different letters within a column signify a significant difference between means (p < 0.05). Data visualization was achieved using GraphPad Prism 9.0 (GraphPad Inc., La Jolla, CA), and the ImageJ software (version 1.8) was employed to measure the data. All experimental data have at least three independent replicates, and the results are expressed as mean ± standard deviation (SD).

Table 3 presents the impact of B. coagulans MF-06 on the growth performance of chickens infected with S. pullorum. On day 7, the body weight of the EM group (62.85 g) was significantly greater than that of the other groups (p < 0.05), while no significant differences were observed among the remaining four groups (p > 0.05). By day 14, the body weight of the SI group had decreased by 5.39% in comparison to the CK group (p < 0.05), the body weights of the PT and AT groups increased by 4.75 and 5.68%, respectively, when compared to the SI group (p < 0.05), the EM group exhibited the highest body weight at 126.50 g on day 14. During the experimental period (1–14 days), the SI group experienced reductions in ADG and ADFI of 7.30 and 2.6%, respectively, compared to the CK group, with an increase in FCR of 4.73% (p < 0.05). When compared to the SI group, the ADG in the PT and AT groups increased by 6.50 and 7.30%, respectively (p < 0.05), while ADFI increased by 2.54 and 3.20% (p < 0.05), additionally, FCR decreased by 3.51 and 4.02% in the PT and AT groups, respectively (p < 0.05). The average daily gain (ADG) in the EM group (6.70 g) was significantly higher than that of the other groups (p < 0.05), while the feed conversion ratio (FCR) was lower than that of the other groups (p < 0.05).

The number of S. pullorum colonizing the duodenum, jejunum, ileum, and cecum in each group is shown in Figure 1. S. pullorum was not detected in the tissues from the CK and EM groups. The colonization levels in the duodenum, jejunum, ileum, and cecum of the SI group were 4.76 × 10⁶, 1.33 × 10⁶, 1.18 × 10⁷, and 6.73 × 10⁶ CFU/0.1 g, respectively, which were higher than those recorded in other groups (p < 0.05). For the PT and AT groups, the colonization number of S. pullorum was 1.62 × 10⁶ and 1.40 × 10⁶ CFU/0.1 g in duodenum, 6.41 × 10⁴ and 5.25 × 10⁵ CFU/0.1 g in jejunum, 6.31 × 10⁵ and 2.76 × 10⁶ CFU/0.1 g in ileum, and 2.72 × 10⁶ and 2.28 × 10⁶ CFU/0.1 g in cecum, respectively. These values were lower than those found in the corresponding parts of the SI group (p < 0.05).

The effects of B. coagulans MF-06 on intestinal index of chicks infected with S. pullorum are shown in Table 4. There was no significant difference in duodenal length index among the groups (p > 0.05). The jejunum length index was similar among the SI, PT, and AT groups (21.9, 20.89, 20.87%), and the values were significantly lower than that of the CK group (23.9%) (p < 0.05). The ileum length index in the SI group (17.29%) was lower than that in the CK group (22.51%) and PT group (20.54%) (p < 0.05), whereas showed no significant difference from that in the AT group (17.82%) (p > 0.05).

To assess the intestinal barrier function across the five groups, the mRNA expression levels of TJP1, CLDN1, CLDN2, Occludin, and MUC2 genes in the jejunum were analyzed by RT-qPCR. As shown in Figure 2, the expression levels of these five genes were up-regulated in the EM group relative to that in the CK group (p < 0.05). In contrast, the TJP1, CLDN1, Occludin, and MUC2 genes were down-regulated in the SI group compared with those in the CK group (p < 0.05). The CLDN1, Occludin, and MUC2 genes were up-regulated in the PT group relative to those in the SI group (p < 0.05). Additionally, all these five genes were up-regulated in the AT group compared with those in the SI group (p < 0.05).

The levels of D-LA and DAO in serum samples were determined to assess the intestinal permeability of the five groups. As shown in Figure 3, there were no significant differences in D-LA and DAO levels between the CK and EM groups (p > 0.05). However, the SI group had higher DAO and D-LA levels than the CK group, which reached 20.48 U/L and 76.40 nmol/L, respectively. Additionally, the PT and AT groups showed lower DAO (17.85 U/L and 17.38 U/L) and D-LA (62.82 nmol/L and 63.34 nmol/L) than the SI group (p < 0.05).

The expression levels of apoptosis-related genes (BAX, BCL2, CASP8, and CASP3) in the jejunum are shown in Figure 4. The pro-apoptotic gene CASP8 was down-regulated and the anti-apoptotic gene BCL2 was up-regulated in the EM group compared with those in the CK group (p < 0.05). The expression of BAX, CASP8, and CASP3 was up-regulated (p < 0.05), and that of BCL2 was down-regulated in the SI group (p < 0.05). Moreover, BAX, CASP8, and CASP3 were down-regulated (p < 0.05) and BCL2 was up-regulated in the PT group compared with those in the SI group (p < 0.05). Compared with the SI group, the AT group showed significant down-regulation of BAX and CASP8 (p < 0.05), while significant up-regulation of BCL2 (p < 0.05).

The effect of B. coagulans MF-06 on expression of proliferating cell nuclear antigen (PCNA) in jejunum is shown in Figure 5. The expression of PCNA showed no significant difference between the EM group and the CK group (p > 0.05). In contrast, the PCNA expression in the SI group decreased (p < 0.05). Moreover, the PT and AT groups had higher expression of PCNA than the SI group (p < 0.05).

As shown in Figure 6, The EM group had higher expression of Wnt3, C-MYC, β-catenin, and LGR5 than the CK group (p < 0.05). In contrast, the SI group had lower expression of all the genes related to the Wnt/β-catenin signaling pathway (p < 0.05) compared with the CK and PT group, and lower expression of Wnt3, TCF4, C-MYC, and LGR5 (p < 0.05), but non-significant increase in the expression of β-catenin and BMI1 (p > 0.05) compared with the AT group. Immunohistochemistry (Figures 7A,B) showed that compared with that of the CK group, the expression of C-MYC protein in jejunum was higher (p < 0.05) in the EM group, but lower in the SI group (p < 0.05). Moreover, the PT and AT groups showed higher expression of the C-MYC protein than the SI group (p < 0.05).

We evaluated the cecal microbial richness through 16S rRNA sequencing. The results are presented in Figure 8. The petal diagram for the structure of the gut microbiota demonstrates the common and unique ASVs across different groups: there were 99 common ASVs, and the unique ASVs included 649 ASVs in CK, 531 ASVs in EM, 624 ASVs in PT, 986 ASVs in AT, and 687 ASVs in SI (Figure 8J). The α-diversity, which reflects species abundance and diversity in a single sample, showed no significant difference in the Chao, Ace, and Simpson indices (p > 0.05), but Shannon showed a distinction trend among 5 groups (p = 0.06), with a significant difference between the AT group and the CK group (p < 0.05) (Table 5). According to β-diversity results, principal component analysis (PCA) showed a distinction trend (p < 0.05), which implied that the structure of intestinal microbiota was pronounced altered by B. coagulans MF-06 supplementation.

We then conducted a taxonomic analysis to compare the relative abundance at the phylum, genus, and species levels. In terms of different phyla, Actinobacteriota, Proteobacteria, and Firmicutes were predominant (Figure 8A). The EM group showed no significant differences in the relative abundance of these three phyla from the CK group (p > 0.05). The SI group exhibited 4.46 times more abundant Actinobacteriota than the CK group (p < 0.05) (Figure 8B), but no significant difference in the relative abundance of Proteobacteria and Firmicutes (p > 0.05) (Figures 8C,D). Compared with the SI group, the PT group exhibited a 0.82-fold decrease in the relative abundance of Actinobacteriota (p < 0.05) (Figure 8B), but no significant difference in Proteobacteria and Firmicutes (p > 0.05) (Figures 8C,D), while the AT group showed no significant differences in the relative abundance of all the three phyla (p > 0.05).

At the genus level, there were statistically significant differences in the relative abundance of Lachnoclostridium, Shuttleworthia, and Unidentified_Eggerthellaceae among groups (p < 0.05) (Figure 8E). The SI group had higher relative abundance of Shuttleworthia and Unidentified_Eggerthellaceae than the CK group and the PT group (p < 0.05), but no significant differences in the relative abundance of these genera from the AT groups (p > 0.05) (Figures 8F–H). At the species level, the SI group showed higher relative abundance of Coriobacteriaceae bacterium CHKCI002 than the CK and PT groups (p < 0.05), but no significant difference from the AT group (p > 0.05) (Figure 8I). Based on the linear discriminant analysis (LDA), we identified 18 species with significant differences in relative abundance across all taxonomic levels (Figure 8K). The results indicated that B. coagulans MF-06 can maintain intestinal health and mitigate S. pullorum-induced inflammation by modulating cecal microbial composition.

The results of Spearman correlation analysis for microbiome and production performance, Wnt/β-catenin signaling pathway-related genes, and intestinal barrier indices are presented in Figures 9A–C. The relative abundance of Actinobacteriota, Unidentified_Eggerthellaceae, and Coriobacteriaceae bacterium CHKCI002 was negatively correlated with ADG and the mRNA expression of Wnt3, TCF4, C-MYC, β-catenin, BMI1, LGR5, BCL2, TJP1, CLDN1 and MUC2, but positively correlated with FCR, the mRNA expression of BAX, CASP, CASP8, and DAO content.