All animal experiments were approved by the Inner Mongolia Agricultural University Animal Care and Use Committee, Hohhot, P. R. China (approval number: NND2021090).

Artemisia annua L. (A. annua) plants were collected from Hohhot, Inner Mongolia, China. The collected plants were washed with distilled water and air-dried in the shade at room temperature. A quantity of 500 g of A. annua powder was taken and subjected to ultrasound treatment using petroleum ether for 30 min to remove fat. The resulting powder was then passed through a 60-mesh sieve. After the natural volatilization of petroleum ether, A. annua polysaccharide extraction was performed using a complex enzyme with a solid–liquid ratio of 1:30. The enzyme mixture consisted of cellulase, pectinase, and papain, added at proportions of 23.2, 19.7, and 15.6%, respectively. The mixture was incubated in a constant temperature shaker at 50°C and 150 rpm for 3 h. Subsequently, it was transferred to a 70°C water bath for 1 h to inactivate the enzymes. The resulting mixture was filtered and concentrated, followed by the addition of anhydrous ethanol into the concentrate at a 4:1 volume ratio. The solution was refrigerated in a 4°C refrigerator for 48 h, then the solution was centrifuged at 3000 × g for 5 min, and the precipitate was collected. The collected precipitate was washed with anhydrous ethanol and acetone, each three times. Subsequent to the washing, the precipitate was freeze-dried to obtain the A. annua polysaccharide. The total soluble sugar content of AAP was measured at 397.42 mg/g, while its molecular weight was identified as 14.639 kDa. AAP was consisted of rhamnose, arabinose, galactose, glucose, mannose, galacturonic acid, and glucuronic acid, with a molar ratio of 2.93:3.75:7.44:58.47:20.87:5.49:1.07. According to the previous study, the appropriate dosage of AAP is 750 mg/kg (Zhang, 2023).

Escherichia. coli O78 (CVCC1490) was obtained from CVCC (China Veterinary Culture Collection Center). The strain was cultured at 37°C in Luria-Bertani (LB) broth (Guangdong Huankai Microbial Sci. & Tech. Co., Ltd., Guangzhou, China) for 24 h to reach a final concentration of 1.0 × 10¹² CFU/mL in sterile LB liquid medium.

A total of 256 Arbor Acres broilers were purchased from a commercial hatchery (Hohhot, China). These birds were randomly divided into four treatments according to their initial body weight including the control group (fed basal diet), the AAP group (fed basal diet supplemented with 750 mg/kg AAP), E. coli group (fed basal diet and orally administered E. coli (3.2 × 10¹¹ CFU/kg body weight)), and AAP + E. coli group (fed basal diet supplemented with 750 mg/kg AAP and orally administered E. coli (3.2 × 10¹¹ CFU/kg body weight)), respectively. Each treatment group had eight replicates with eight birds per replicate, half male and half female. The trial included the pre-feeding period (d 1 to 14), and the formal trial period (d 15 to 42). During d 15 to 20 (E. coli challenge phase I), the E. coli treatment groups were given oral administration of 0.5 mL E. coli saline suspension (2.5 × 10¹¹ CFU/mL) in the morning, the control group was given the same amount of normal saline. During the earlier period (d 1 to 21), the chicks are characterized by an underdeveloped immune system and limited resistance, rendering them vulnerable to external pathogens. Additionally, their intestinal microflora is in the establishment stage, and is susceptible to environmental influence. This phase is provided an opportune time for the observation and analysis of the infection process, pathogenic mechanisms, and the broiler’s resistance to E. coli. As broilers progress into the growth period (d 22 to 42), their physiological and immune function gradually improve. This developmental stage presents an opportunity for deeper exploration of the infection characteristics and pathogenicity of E. coli at different growth stages. By conducting E. coli challenge experiments during this phase, valuable insights into the interplay between E. coli infection and broiler physiology can be gained, contributing to a more nuanced understanding of host-pathogen interactions in poultry production systems. Therefore, during d 36 to 42 (E. coli challenge phase II), the E. coli treatment groups were given oral administration of 2 mL E. coli saline suspension (3.2 × 10¹¹ CFU/mL) in the morning, the control group was given the same amount of normal saline. By conducting E. coli challenges in two stages, the dynamics of E. coli infection in broilers can be comprehensively investigated, thus enhancing the understanding of its pathogenicity. The workflow is shown in Figure 1. Diets were formulated to meet the nutritional recommendations of the Feeding Standard of Chicken, China (NY/T 33-2004) (Table 1). The birds had ad libitum access to food and water. At the end of the experiment, eight birds were randomly selected from each treatment (one chicken per replicate) and slaughtered on d 21 and d 42, respectively. Blood was collected from wing veins, and the serum was separated and preserved at −20°C. Additionally, the jejunal tissues and chyme were promptly collected and stored at −80°C for further analysis.

On d 1, 14, 21, 35, and 42 of the experiment, the body weight (BW) and feed intake of experiment birds in each replicate were meticulously measured and recorded. Subsequently, the average daily gain (ADG), average daily feed intake (ADFI), and feed conversion ratio (FCR) were calculated for each stage of the experiment. During d 19 to 21 and d 40 to 42 of the experiment, feces was methodically collected from each replicate, and the fecal weight and feed intake of each replicate were meticulously recorded after continuous fecal collection for three consecutive days. The apparent nutrient retention was measured via the total feces collection method, and then calculating the apparent metabolic rate of feed dry matter (DM), crude protein (CP), crude fat (ether extract, EE), calcium (Ca), and phosphorus (P). To provide a more accurate representation of growth, the values were corrected for mortality rate.

On d 21 and 42, one bird was randomly selected from each replicate pen and weighed accurately. Blood samples were then collected from the wing vein and centrifuged (3,000 × g, 10 min) at 4°C. The serum was then carefully harvested and stored at −20°C until further analysis. The birds were humanely euthanized by cervical dislocation. The jejunum’s middle segment was carefully excised, rinsed with sterile cold saline, placed in a sterile tube, and then flash-frozen in liquid nitrogen. The samples were stored at −80°C for mRNA expression analysis. Additionally, 1 cm section of the jejunum was gently washed with 0.9% (w/vol) physiological saline and fixed in a 10% formalin solution for further morphological examination.

The serum diamine oxidase (DAO, product code: A088-1, assay range: 0 U/L–100 U/L) was tested using commercial assay kits, read absorbance OD at 340 nm, the difference in inter-assay is less 10% (Jiancheng Bioengineering Institute, Co. Ltd., Nanjing, China). The serum endotoxins (ET, catalogue number: JYM0109Ch, assay range: 1.5 ng/mL–100 ng/mL) and D-lactate (D-LA, catalogue number: JYM0160Ch, assay range: 0.8 ng/mL–50 ng/mL) were tested using ELISA kits, read absorbance OD at 450 nm, the difference in intra-assay and inter-assay is less than 9 and 15%, respectively, (Wuhan Gene Beauty Biotechnology Co. Ltd., China).

The jejunal tissues were processed using a hand-held homogenizer (FA6/10, FLUKO, Shanghai, China) at 4°C in ice-cold 0.9% NaCl solution (wt/vol, 1:9) and then centrifuged at 4000 × g for 15 min at 4°C. The resulting supernatant was collected for further analysis. The protein content of the homogenate was determined using the Coomassie Brilliant Blue assay (product code: A045-2, absorbance OD at 595 nm) according to the manufacturer’s instructions for the commercial kits (Nanjing Jiancheng Institute of Bioengineering, Nanjing, China).

The concentration of interleukin-1 beta (IL-1β, catalogue number: JYM0041Ch, assay range: 1.5 pg/mL–100 pg/mL), interleukin-6 (IL-6, catalogue number: JYM0028Ch, assay range: 1 pg/mL–100 pg/mL), tumor necrosis factor-α (TNF-α, catalogue number: JYM0033Ch, assay range: 1.2 pg/mL–100 pg/mL), immunoglobulin A (IgA, catalogue number: JYM0012Ch, assay range: 1 μg/mL–70 μg/mL), immunoglobulin G (IgG, catalogue number: JYM0001Ch, assay range: 0.3 μg/mL–20 μg/mL), immunoglobulin M (IgM, catalogue number: JYM0060Ch, assay range: 8 ng/mL–450 ng/mL), and secretory immunoglobulin A (sIgA, catalogue number: JYM0036Ch, assay range: 15 pg/mL–1000 pg/mL) in the serum and intestinal tissue homogenate supernatant was analyzed using ELISA kits (Wuhan Gene Beauty Biotechnology Co. Ltd., China) following the manufacturer’s instructions, read absorbance OD at 450 nm, the difference in intra-assay and inter-assay is less than 9 and 15%, respectively.

The total antioxidant capacity (TAC, product code: A015-2-1, assay range: 0.5 mM–2 mM, absorbance OD at 405 nm), the activity of total superoxide dismutase (SOD, product code: A001-1, absorbance OD at 550 nm), glutathione peroxidase (GPx, product code: A005, absorbance OD at 412 nm), and catalase (CAT, product code: A007-1-1, absorbance OD at 405 nm), and the concentration of glutathione (GSH, product code: A006-2-1, absorbance OD at 405 nm) and malondialdehyde (MDA, product code: A003-1, absorbance OD at 532 nm) in the serum and intestinal tissue were determined by a spectrophotometric method according to the instructions of the commercial kits (Nanjing Jiancheng Institute of Bioengineering, Nanjing, China).

A small portion of the jejunum tissues were fixed in 10% formalin and then embedded in paraffin, then sliced into thin sections with a thickness of 7 μm using a rotary microtome (YD-1508R Rotary Slicer, Yidi Medical Equipment Factory, Jinhua, Zhejiang, China), and stained with hematoxylin and eosin. The 10 intact villi of each tissue were precisely measured for villus height (VH) and crypt depth (CD) by high-resolution photography under 100× magnification using a light microscope (Olympus SZX10, Tokyo, Japan), and the average values for each tissue for each tissue were calculated.

Total RNA from jejunal tissue samples was obtained using Trizol reagent (TaKaRa Biotechnology Co. Ltd., Dalian, China). The purity and quantity of the total RNA were assessed with a spectrophotometer (Pultton P200CM, San Jose, CA, United States). Subsequently, the DNA of total RNA was removed by incubation for 2 min at 42°C with a gDNA digester (Yeasen Biotechnology Co., Ltd. Shanghai, China). Total RNA was reverse transcribed to cDNA on Labcycler (SensoQuest GmbH, Göttingen, Germany) using Hifair^® II SuperMix plus (Yeasen Biotechnology Co., Ltd. Shanghai, China). The reactions were incubated for 5 min at 85°C, 30 min at 42°C, and 5 min at 85°C.

Real-time PCR was performed using LightCycler^® 96 instrument and application software analysis system (LightCycler^® 96 Instrument, Roche Diagnostics, Indiana, United States) with a Hieff^® qPCR SYBR^® Green Master Mix (No Rox) Kit (Yeasen Biotechnology Co., Ltd. Shanghai, China). The reactions were: 95°C for 30 s (hold stage), followed by 40 cycles of 95°C for 5 s, 60°C for 30 s, and 72°C for 20 s (PCR stage), then 95°C for 15 s, 60°C for 1 min, 95°C for 15 s (melt-curve stage). All samples were run in duplicate in 10 μL reaction volume and melt curve analysis was performed to ensure the specificity of the PCR-amplified product. The mRNA expression of each gene was normalized to that of 𝛽-actin. The fold change relative to the control group was analyzed according to the 2^−ΔΔCT method. The specific sequences of primers are listed in Table 2.

Microbial community genomic DNA was extracted from jejunum content samples using the E.Z.N.A.^® soil DNA Kit (Omega Bio-tek, Norcross, GA, United States) according to manufacturer’s instructions. The DNA extract was checked on 1% agarose gel, and DNA concentration and purity were determined with NanoDrop 2000 UV–vis spectrophotometer (Thermo Scientific, Wilmington, United States). The hypervariable region V3-V4 of the bacterial 16S rRNA gene were amplified with primer pairs 338F (5’-ACTCCTACGGGAGGCAGCAG-3′) and 806R (5’-GGACTACHVGGGTWTCTAAT-3′) by an ABI GeneAmp^® 9,700 PCR thermocycler (ABI, CA, United States). The PCR amplification of 16S rRNA gene was performed as follows: initial denaturation at 95°C for 3 min, followed by 27 cycles of denaturing at 95°C for 30 s, annealing at 55°C for 30 s and extension at 72°Cfor 45 s, and single extension at 72°C for 10 min, and end at 10°C. The PCR mixtures contain 5 × Pro Taq 10 μL, forward primer (5 μM) 0.8 μL, reverse primer (5 μM) 0.8 μL, template DNA 10 ng/μL, and finally ddH₂O up to 20 μL. PCR reactions were performed in triplicate. The PCR product was extracted from 2% agarose gel and purified using the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, United States) according to manufacturer’s instructions and quantified using Quantus™ Fluorometer (Promega, United States). Purified amplicons were pooled in equimolar and paired-end sequenced on an Illumina MiSeq PE300 platform/NovaSeq PE250 platform (Illumina, San Diego, United States) according to the standard protocols by Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China). The raw 16S rRNA gene sequencing reads were demultiplexed, quality-filtered by fastp version 0.20.0 and merged by Flash version 1.2.7. Operational taxonomic units (OTUs) with 97% similarity cutoff were clustered using Uparse version 7.1, and chimeric sequences were identified and removed. The taxonomy of each OTU representative sequence was analyzed by RDP Classifier version 2.2 against the 16S rRNA database (Silva v138) using confidence threshold of 0.7. α diversity, β diversity, community composition, and analysis of different communities are carried out on the I-Sanger Cloud Platform provided by Majorbio Bio-Pharm Technology Co., Ltd. (Qiao et al., 2022b).

Data were analyzed by one-way ANOVA with the general linear model procedure of SAS version 9.2 (SAS Institute Inc., Cary, NC), and a pen of broilers (a replicate) served as the experimental unit for all data. The differences among treatments were tested by Duncan’s multiple comparison analysis and were considered significant at p < 0.05. The results were expressed as the mean and standard error of the mean (SEM).

As shown in Table 3, compared with the control group, dietary AAP significantly increased BW of broilers on d 42 (p < 0.01). E. coli-challenged broilers had significantly decreased BW on d 21, d 35, and d 42, ADG and ADFI on d 36–42 (p < 0.05). And there was no difference between AAP + E. coli group and control group. In addition, dietary AAP supplementation significantly enhanced ADG compared to the control group during d 36–42 (p < 0.01). However, there was no difference in FCR among treatment groups.

As described in Figure 2, compared with the control group, dietary AAP supplementation significantly increased the apparent metabolic rate of CP on d 21 and 42 (p < 0.05). The apparent metabolic rate of P on d 21, EE and Ca on d 42 in E. coli-challenged broilers was significantly lower than that of control group (p < 0.05). And there was no difference between AAP + E. coli group and the control group.

As summarized in Table 4, compared with the control group, E. coli-challenged broilers had significantly increased DAO activity on d 21 and 42, and ET content on d 42 (p < 0.05), and there was no difference between the AAP + E. coli group and the control group.

As illustrated in Figure 3, compared with the control group, E. coli-challenged broilers had reduced VH on d 21 (p = 0.055) and 42 (p < 0.05), and VH/CD d 21 and 42 (p < 0.01), but there was no difference between the AAP + E. coli group and the control group. Moreover, dietary AAP supplementation significantly enhanced VH/CD on d 42 (p < 0.01).

As shown in Table 5, dietary AAP supplementation significantly increased serum IgA content on d 21 (p < 0.01). Compared with the control group, E. coli-challenged broilers significantly reduced serum IgM content on d 21 (p < 0.05) and increased IL-6 content on d 42 (p < 0.01), and there was no difference between the AAP + E. coli group and the control group.

As presented in Table 6, compared with the control group, E. coli-challenged broilers had significantly decreased jejunum IgG content (p < 0.05), but there was no difference between the AAP + E. coli group and the control group on d 21. The jejunum IgM content of E. coli group and AAP + E. coli group was significantly lower than that of the control group on d 21 (p < 0.01). Compared with the control group, E. coli-challenged broilers markedly increased jejunum IL-1β and IL-6 content (p < 0.05), but there was no difference between the AAP + E. coli group and the control group on d 42.

As shown in Table 7, compared with the control group, E. coli-challenged broilers had significantly reduced serum GPx activity on d 21 (p < 0.05), but there was no difference between the AAP + E. coli group and the control group. Compared with the control group, dietary AAP inclusion had notably decreased serum MDA concentration on d 21 (p < 0.01).

As presented in Table 8, compared with the control group, E. coli-challenged broilers had significantly reduced jejunum TAC, CAT, SOD activity (p < 0.05), but there was no difference between the AAP + E. coli group and the control group on d 42. Moreover, the jejunum MDA content of E. coli-challenged broilers trended to be higher than that of the control group on d 21 (p = 0.098) and 42 (p = 0.052).

As shown in Figure 4, compared with the control group, dietary AAP supplementation extremely significantly increased jejunum mRNA expression levels of Occludin (d 21), ZO-1 (d 42) and Mucin-2 (d 42) (p < 0.01). Besides, compared with the control group, E. coli-challenged broilers noticeably reduced jejunum mRNA expression level of Claudin-1 (d 21), Occludin (d 21 and 42), ZO-1 (d 42) and Mucin-2 (d 21 and 42) (p < 0.05), but there was no difference between AAP + E. coli group and control group.

As indicated in Figure 5, compared with the control group, E. coli-challenged broilers had extremely significantly increased jejunum mRNA expression level of IL-1β and TLR4 (p < 0.01), while there was no difference between the AAP + E. coli group and the control group on d 21 and 42. Compared with the control group, E. coli group and AAP + E. coli group significantly increased jejunum mRNA expression level of IL-6, but which was markedly lower in AAP + E. coli group than that in the E. coli group on d 21 and 42 (p < 0.01). Compared with the control group, E. coli-challenged broilers extremely significantly increased jejunum mRNA expression level of TNF-α (d 21) and MyD88 (d 42) (p < 0.01), but there was no difference between the AAP + E. coli group and the control group.

As shown in Figure 6, compared with the control group, E. coli-challenged broilers dramatically decreased jejunum mRNA expression level of CAT and Nrf2 (p < 0.01), but there was no difference between the AAP + E. coli group and the control group on d 21. Dietary AAP supplementation extremely significantly increased jejunum mRNA expression level of CAT (d 42), SOD (d 42; p = 0.079) and Nrf2 (d 21 and 42) (p < 0.01). Besides, compared with the control and AAP group, E. coli group and AAP + E. coli group markedly significantly increased the jejunum mRNA expression level of Keap1 on d 21 and 42 (p < 0.01), however, AAP + E. coli group had significantly lower jejunum mRNA expression level of Keap1 than E. coli group on d 42 (p < 0.01).

The bacterial α diversity indices are presented in Figures 7, 8. There was no significant difference in α diversity indices (Sobs, Chao, Simpson, Shannon, Ace, and Coverage) on d 21 (Figures 7A–F; p > 0.05). However, the Simpson indexes (Figure 7C; p = 0.075) and Ace indexes (Figure 7E; p = 0.074) of the E. coli group tended to be significantly higher and lower, respectively, than those of the other groups. On d 42, E. coli-challenged broilers significantly increased the indexes of Sobs, Chao, Shannon and Ace (Figures 8A,B,D,E; p < 0.05), and significantly decreased the indexes of Simpon and Coverage (Figures 8C,F; p < 0.05). However, there was no difference between the AAP + E. coli group and the control group.

On d 21, the Venn diagram (Figure 9A) showed that a total of 322 OTU is shared among the 4 treatment groups. In addition, the unique OUT numbers corresponding to the control group, AAP group, E. coli group, and AAP + E. coli group were, respectively, 71, 248, 39, and 105 on d 21. Principal co-ordinates analysis (PCoA) (Figure 9B) showed that the microbial community composition changed among the four treatment groups on d 21. The composition of the jejunal microbiota is shown in Figure 9 on d 21. At the phylum level, the dominant bacteria were Firmicutes, Proteobacteria, Cyanobacteria, Patescibacteria, and Actinobacteria (Figure 9C). The abundance of Desulfobacterota in phylum level in AAP group and AAP + E. coli group was significantly higher than that in control group and E. coli group (Figure 9D; p < 0.05). At the genus level, the dominant bacteria were Lactobacillus, Streptococcus, Enterococcus, Novosphingobium, and Romboutsia (Figure 9E). Compared with the control group and E. coli group, the AAP group and AAP + E. coli group significantly decreased the abundance of norank_f_Obscuribacteraceae, Ralstonia, Mitsuokella, Megasphaera, Megamonas and Bifidobacterium, and significantly increased the abundance of Aerococcus, Desulfovibrio and Candidatus_Saccharimonas at the genus level (Figure 9F; p < 0.05). Besides, the abundance of Enterorhabdus in the AAP + E. coli group was significantly higher than that in the other groups (Figure 9F; p < 0.05).

On d 42, the Venn diagram (Figure 10A) showed that a total of 227 OTU is shared among the 4 treatment groups. In addition, the unique OUT numbers corresponding to the control group, AAP group, E. coli group, and AAP + E. coli group were, respectively, 183, 24, 111, and 81 on d 42. Principal co-ordinates analysis (PCoA) (Figure 10B) showed that the microbial community composition changed dramatically among the four treatment groups on d 42. The composition of the jejunal microbiota is shown in Figure 10 on d 42. At the phylum level, the dominant bacteria were Firmicutes, Proteobacteria, Actinobacteria, Cyanobacteria, and Bacteroidota (Figure 10C). The abundance of Bacteroidota in AAP + E. coli group was significantly higher than the other treatment groups at the phylum level (Figure 10D; p < 0.05). At the genus level, the dominant bacteria were Lactobacillus, Streptococcus, Enterococcus, Lactococcus, and Turicibacter (Figure 10E). Compared with the control group and AAP group, the E. coli group and AAP + E. coli group significantly decreased the abundance of Lactobacillus, and significantly increased the abundance of Romboutsia, Turicibacter, Christensenellaceae_R-7_group, UCG-005, Eisenbergiella, unclassified_f_Lachnospiraceae, norank_f_norank_o_Clostridia_UCG-014, Ruminococcus_torques_group and unclassified_f_Peptostreptococcaceae at the genus level (Figure 10F; p < 0.05). Among them, the abundance of Christensenellaceae_R-7_group, UCG-005, Eisenbergiella, unclassified_f_Lachnospiraceae, norank_f_norank_o_Clostridia_UCG-014, Ruminococcus_torques_group and unclassified_f_Peptostreptococcaceae in AAP + E. coli group was significantly higher than that in E. coli group (Figure 10F; p < 0.05).

The linear discriminant analysis (LDA = 3) effect size (LEfSe) algorithm was used to analyze the taxonomic abundance of microbiota. The results for d 21 are shown in Figures 11A,B. c_Vampirivibrionia, f_Obscuribacteraceae, g_norank_f_Obscuribacteraceae, o_Obscuribacterales, c_Negativicutes, o_Veillonellales-Selenomonadales, f_Veillonellaceae, o_Streptosporangiales, g_Ureibacillus, and f_Selenomonadaceae were enhanced in the control group. f_Streptococcaceae, f_Aerococcaceae, g_Aerococcus, and g_Harryflintia were enhanced in AAP group. g_Candidatus_Saccharimonas, g_Aureimonas, and g_norank_f_Lachnospiraceae were enhanced in AAP + E. coli group. The results for d 42 are shown in Figures 12A,B. g_Lactobacillus and f_Lactobacillaceae were enhanced in the control group. o_Lactobacillales and c_Bacilli were enhanced in AAP group. c_Clostridia, g_Ruminococcus_gauvreauii_group, o_Oscillospirales, o_Lachnospirales, f_Lachnospiraceae, o_Christensenellales, f_Christensenellaceae, g_Christensenellaceae_R-7_group, g_Dubosiella, f_Ruminococcaceae, f_Oscillospiraceae, g_Sellimonas, g_UCG-005, and g_Eisenbergiella were enhanced in E. coli group (LDA ≥ 4). o_Peptostreptococcales-Tissierellales, f_Peptostreptococcaceae, g_Romboutsia, o_Erysipelotrichales, f_Erysipelotrichaceae, g_Turicibacter, g_Brevibacterium, f_Brevibacteriaceae, and f_Geodermatophilaceae were enhanced in AAP + E. coli group (LDA ≥ 4).

The co-occurrence network diagram of intestinal microbial at phylum and genus levels is depicted in Figure 13. The results for d 21 are shown in Figures 13A,B. At the phylum level, p_Acidobacteriota, p_Deinococcota, and p_WPS-2 were only enriched in the AAP group (Figure 13A). p_unclassified_k_norank_d_Bacteria were only enriched in the AAP + E. coli group (Figure 13A). At the genus level, g_norank_f_Caulobacteraceae, g_unclassified_f_Lachnospiraceae, g_Lachnoclostridium, g_Eisenbergiella, and g_UCG-005 were only enriched in the control group (Figure 13B). g_Christensenellaceae_R-7_group, g_Clostridium_sensu_stricto_1, g_Gemmobacter, g_norank_f_norank_o_Clostridia_UCG-014, and g_unclassified_o_Saccharimonadales were only enriched in the AAP group (Figure 13B). g_Subdoilgranulum were only enriched in the E. coli group (Figure 13B). The results for d 42 are shown in Figures 13C,D. At the phylum level, p_Fibrobacterota and p_Spirochaetota were only enriched in the control group (Figure 13C). p_Acidobacteriota and p_Halanaerobiaeota were only enriched in the AAP + E. coli group (Figure 13C). At the genus level, g_bacteroides and g_Candidatus_Arthromitus were only enriched in the AAP + E. coli group (Figure 13D). g_Blautia, g_Bradyrhizobium, g_Lachnoclostridium, g_UCG-005, g_NK4A214_group, g_Subdoligranulum, g_Veillonella, g_Eienbergiella, g_Ruminococcus_torques_group, g_unclassified_f_Ruminococcaceae, g_unclassified_f_Lachnospiraceae, g_unclassified_f_Peptostreptococcaceae, and g_norank_f_norank_o_Clostridia_UCG-014 were only enriched in the E. coli group (Figure 13D).

The correlation between intestinal microbial proportions at the genus level and intestinal oxidative status, as well as pro/anti-inflammatory markers among four groups, is illustrated in Figure 14. The results for d 21 are shown in Figure 14A. The proportion of Lactobacillus had a positive correlation with Keap1 mRNA expression level (p < 0.05), and had a negative correlation with CAT mRNA expression level (p < 0.05). The proportion of norank_f_Obscuribacteraceae had a positive correlation with IL-1β content and MyD88 mRNA expression level (p < 0.01), and had a negative correlation with GPx activity (p < 0.05). The proportion of Lachnoclostridium had a positive correlation with IL-6 content (p < 0.01). The proportion of Eisenbergiella had a negative correlation with GSH content (p < 0.01). The proportion of Streptococcus had a positive correlation with CAT activity (p < 0.05), and had a negative correlation with IL-6 and Keap1 mRNA expression level (p < 0.05). The proportion of Romboutsia had a positive correlation with NF-κB p65 mRNA expression level (p < 0.05), and had a negative correlation with TNF-α content (p < 0.05). The proportion of Enterococcus had a positive correlation with CAT activity (p < 0.05). The proportion of Bradyrhizobium had a positive correlation with MyD88 mRNA expression level (p < 0.01). The proportion of norank_f_Caulobacteraceae had a positive correlation with IL-1β and IL-6 content, MyD88 mRNA expression level (p < 0.01), and had a negative correlation with Keap1 mRNA expression level (p < 0.05). The proportion of Novosphingobium and Lactococcus had a negative correlation with Keap1 mRNA expression level (p < 0.05). The proportion of Christensenellaceae_R-7_group had a negative correlation with GPx activity (p < 0.05). The proportion of unclassified_f_Lachnospiraceae had a positive correlation with IL-6 content (p < 0.05).

The results for d 42 are shown in Figure 14B. The proportion of Lactobacillus had a positive correlation with TAC, CAT, SOD, and GPx activity, GSH content, CAT, SOD, GPx, and Nrf2 mRNA expression level (p < 0.05), and had a negative correlation with IL-1β, IL-6 and MDA content, IL-1β, IL-6, TLR4, MyD88, and Keap1 mRNA expression level (p < 0.05). The proportion of Romboutsia had a positive correlation with TNF-α and MDA content, and IL-1β, IL-6, TLR4, MyD88, and Keap1 mRNA expression level (p < 0.05), and had a negative correlation with TAC, CAT, and SOD activity, GSH content, CAT, SOD, GPx, and Nrf2 mRNA expression level (p < 0.05). The proportion of unclassified_f_Lachnospiraceae had a positive correlation with IL-6 and TNF-α content, and IL-1β, IL-6, TLR4, MyD88, and Keap1 mRNA expression level (p < 0.05), and had a negative correlation with TAC, CAT, and SOD activity, GSH content, CAT, GPx, and Nrf2 mRNA expression level (p < 0.05). The proportion of norank_f_norank_o_Clostridia_UCG-014 had a positive correlation with IL-1β and TNF-α content, and IL-1β, IL-6, TLR4, MyD88, and Keap1 mRNA expression level (p < 0.05), and had a negative correlation with TAC, CAT, and SOD activity, CAT, GPx, and Nrf2 mRNA expression level (p < 0.05). The proportion of Eisenbergiella had a positive correlation with IL-1β, IL-6 and TNF-α content, and IL-1β, IL-6, TLR4, MyD88, and Keap1 mRNA expression level (p < 0.05), and had a negative correlation with TAC, CAT, and SOD activity, GSH content, CAT, SOD, GPx, and Nrf2 mRNA expression level (p < 0.05). The proportion of Christensenellaceae_R-7_group had a positive correlation with IL-1β, IL-6, TLR4, and Keap1 mRNA expression level (p < 0.05), and had a negative correlation with TAC, CAT, and SOD activity, CAT, SOD, GPx, and Nrf2 mRNA expression level (p < 0.05). The proportion of UCG-005 had a positive correlation with MDA content, and IL-1β, IL-6, TLR4, MyD88, and Keap1 mRNA expression level (p < 0.05), and had a negative correlation with TAC, CAT, and SOD activity, CAT, SOD, GPx, and Nrf2 mRNA expression level (p < 0.05). The proportion of Turicibacter had a positive correlation with IL-1β, IL-6, TLR4, and Keap1 mRNA expression level (p < 0.05), and had a negative correlation with TAC, CAT, and SOD activity, GSH content, CAT, SOD, GPx, and Nrf2 mRNA expression level (p < 0.05). The proportion of Subdoligranulum had a positive correlation with IL-1β and Keap1 mRNA expression level (p < 0.05), and had a negative correlation with CAT and SOD activity, CAT, GPx, and Nrf2 mRNA expression level (p < 0.05). The proportion of Ruminococcus_torques_group had a positive correlation with TNF-α content, IL-1β, IL-6, and Keap1 mRNA expression level (p < 0.05), and had a negative correlation with TAC and CAT activity, CAT, GPx, and Nrf2 mRNA expression level (p < 0.05). The proportion of Lactococcus had a positive correlation with NF-κB p65 mRNA expression level (p < 0.05). The proportion of Veillonella had a positive correlation with IL-6 content and NF-κB p65 mRNA expression level (p < 0.05). The proportion of Enterococcus had a positive correlation with TNF-α content, and had a negative correlation with GPx mRNA expression level (p < 0.05). The proportion of Rothia had a negative correlation with GPx mRNA expression level (p < 0.01). The proportion of Macrococcus had a positive correlation with IL-6 mRNA expression level (p < 0.05), and had a negative correlation with CAT activity, GPx and Nrf2 mRNA expression level (p < 0.05). The proportion of Corynebacterium had a negative correlation with Nrf2 mRNA expression level (p < 0.05). The proportion of Candidatus_Arthromitus had a negative correlation with TAC ability, NF-κB p65 mRNA expression level (p < 0.05).