2.1. Animals and Antibiotic Treatment

Seven-week-old male and female C57BL/6 mice (Vital River Laboratory Animal Technology Co. Ltd, Beijing, China) were housed under pathogen-free conditions at the Laboratory Animal Resources Center, Tsinghua University, according to approved protocols (No. 20-LZ1-1).

A mouse model of ulcerative colitis was established by oral administration of 2% dextran sulfate sodium (DSS, molecular weight 36000–50000 Da, MP Biomedicals, Irvine, CA) in drinking water for 1 week after 1 week of antibiotic uptake. Antibiotic administration was continued during DSS treatment. For the present study, neomycin (Neo), metronidazole (Met), and vancomycin (Van) were supplied in drinking water at concentrations of 1 g/L, 1 g/L, and 0.5 g/L, respectively [15]. All antibiotics were purchased from Sigma-Aldrich (St. Louis, MO). The antibiotic cocktail group (Abx) was treated with 1 g/L ampicillin, 1 g/L neomycin, 1 g/L metronidazole, 0.5 g/L, and vancomycin. To maintain the concentration of antibiotics and DSS against degradation, these reagents were renewed every 2 days.

Each mouse received 5 × 10⁵ MC38 cells (subcutaneous (s.c.)) on the right flank to establish a colon cancer model after 1 week of adaptation. Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin was used to culture MC38 cells (Cell Bank of the Type Culture Collection, Chinese Academy of Sciences) at 37°C in a 5% CO₂ atmosphere [26]. Tumor volumes were measured twice a week using the following formula: (width² × length)/2. One week after tumor inoculation, the mice were randomly divided into different groups according to tumor size (n = 6) to ensure the same average volume of each group. For the groups treated with αPD-L1 (P), on days 7 and 10, each mouse was injected with 200 μg of anti-PD-L1 mAb (αPD-L1, BE0101; BioXCell, Lebanon, NH) (intraperitoneal (i.p.)). The control group (CT) mice were injected with IgG2b isotype (BE0090, BioXCell) at the same time. For the antibiotic-treated groups, neomycin (PN) and metronidazole (PM) were added to drinking water at a concentration of 1 g/L for 1 week on days 7 to 14. Experiments in which a combination of αPD-L1 and antibiotics was administered to mice were conducted according to a previous study with slight adjustment [27]. For the groups treated with αPD-L1 and DSS (PDS) (2.0%, w/v) DSS was dissolved in drinking water for 1 week at days 7 to 14. Mice in the antibiotic treatment group received αPD-L1 (2.0%, w/v) DSS, neomycin (PDSN), or metronidazole (PDSM) at the same time.

2.2. Evaluation of DSS Colitis

The body weight, stool consistency, and rectal bleeding state of the mice were determined daily. The disease activity index (DAI) was evaluated as an index of disease activity by the scores of body weight loss (scored as 0, no change; 1, 1–5% loss; 2, 5–10% loss; and 3, 10–20% loss), stool consistency (0, normal; 1, soft but firm; 2, soft; and 3, diarrhea), and fecal blood (0, none; 1–2, blood; and 3, gross bleeding) [28].

2.3. Hematoxylin and Eosin (HE) Staining and Histologic Analysis

For histological analysis, colon tissues were fixed in 4% formalin and embedded in paraffin. Paraffin was then cut into sections and stained with hematoxylin and eosin. Histological changes were evaluated as previously described by Rangan et al. [29]. In brief, the histological scoring system used four scores for inflammatory infiltration of mucosa (0, none; 1, infiltrate around crypt basis; 2, infiltrate reaching to submucosae; 3, extensive infiltration to submucosae; and 4, full-layer infiltration), damage of goblet cells (0, normal; 1, partial loss of goblet cells; 2, loss of multiple goblet cells but with normal epithelial structure; 3, loss of goblet cells in large areas; 4, all destroyed goblet cells), damage to crypts (0, none; 1, loss of a third of the crypts; 2,loss of two-thirds of the crypts; 3, only maintain intact epithelium; 4, loss of all crypts and epithelium), and the degree of injury (0, none; 1, mucosa; 2, mucosa and submucosa; 3, muscularis; 4, full layer).

2.4. Flow Cytometry Analysis of Immune Cells

Immune infiltration in the colon lamina propria and tumor was analyzed, as previously described [30]. In brief, the tissues (i.e., tumor and colon) were minced and digested with 1 mg/mL collagenase IV and 0.15 mg/mL DNase I in Hank’s balanced salt solution at 37°C for 30 min. Next, single cells were obtained by filtering the digested tissues with a 70 μm cell strainer (BD Biosciences, Franklin Lakes, NJ). For the spleen and mesenteric lymph nodes (MLNs), immune cells were obtained by gently grinding with a plunger of a sterile syringe in a cell strainer. Spleen cells were treated with red blood cell lysis buffer (Beijing Solarbio Science & Technology, Beijing, China) on ice for 15 min. Single cells were stained with anti-mouse antibodies (eBioscience, Thermo Fisher Scientific, Waltham, MA) against CD4 (30-F11)-PE/Cy7, CD8a (53-6.7)-BrilliantViolet605, CD45 (GK1.5)-APC/Cy7, CD25 (PC61.5)-APC, CD19 (1D3)-AlexaFluor700, NK1.1 (PK136)-APC, Ly6C (HK1.4)-PE, Ly6G (1A8)-FITC, F4/80 (BM8)-BrilliantViolet421, and CD11b (M1/70)-APC in the dark (4°C for 30 min). After surface staining, the cells were fixed and permeabilized with a Foxp3/Transcription Factor Staining Buffer set (eBioscience) for intracellular staining in the dark (4°C, 30 min) with anti-mouse antibodies (eBioscience), including Foxp3 (150D)-PE, IL-17A (eBio17B7)-PE/Cy5, and INFγ (XMG1.2)-APC. Cells were resuspended in the phosphate-buffered saline after washing twice and analyzed on an LSRFortessa flow cytometer (BD Biosciences). All data were analyzed using the FlowJo software (version 10.6.2, TreeStar,, Ashland, OR).

2.5. Enzyme-Linked Immunosorbent Analysis (ELISA)

The cytokines determined were IL-6 and TNF-α. After euthanasia, blood was collected from the eye socket, and serum was separated by centrifugation (3000 rpm, 10 min) and frozen at -80°C until assayed for cytokine content by commercial ELISA (eBiosciences). Plates were read at 450 nm using a plate reader following the procedure of the manufacturer’s instructions.

2.6. RNA Extraction and Quantitative Real-Time PCR

Total RNA was extracted from cecum tissues that had been preserved in RNAlater solution (Invitrogen, Carlsbad, CA) using 1 mL of TRIzol Reagent (Invitrogen) as previously described. RNA was further used for cDNA production using the EasyScript All-in-One First-Strand cDNA Synthesis SuperMix for qPCR (Takara, Tokyo, Japan) according to the manufacturer’s instructions. Quantitative real-time (RT) PCR assays were performed to detect the effect of treatments on the mRNA expression of the cytokines TNF-α, IL-1β, IL-17, and intestinal tight junction proteins claudin-2, ZO-1, and occludin using SYBR Green on a CFX96 Touch Real-Time PCR Detection System (Bio-Rad, Hercules, CA). Specific primers for genes are as follows: GAPDH (forward, AGGTCGGTGTGAACGGATTTG; reverse, GGGGTCGTTGATGGCAACA), TNF-α (forward, CCCTCACACTCAGATCATCTTCT; reverse, GCTACGACGTGGGCTACAG), and IL-1β (forward, GCAACTGTTCCTGAACTCAACT; reverse, ATCTTTTGGGGTCCGTCAACT). All primers were designed in http://www.ncbi.nlm.nih.gov/ and synthesized by Thermo Fisher Scientific (Waltham, MA). The average cellular mRNA expression levels were calculated as previously described [13].

2.7. Myeloperoxidase Activity Assay

The activity of intestinal myeloperoxidase (MPO) was determined using an MPO assay kit (Nanjing Jiancheng Bioengineering Institute) [31]. Briefly, after the intestinal tissue was weighed, the homogenate medium was added at a volume ratio of 1 : 19 to prepare a 5% tissue homogenate and kept at 37°C for 15 minutes. Then, 3 mL of the chromogenic agent was added to the sample tube in a water bath at 37°C and incubated for 30 minutes. Finally, the samples were incubated at 60°C for 10 minutes with the addition of 0.05 mL of 3,3′-dimethoxybenzidine (0.17 mg/mL) before absorbance measurement at 460 nm. MPO activity in the supernatant was determined by measuring the H₂O₂-dependent oxidation and expressed as units per gram of total protein (U/g wet weight of colon tissue) according to the following formula: MPO activity = (OD_sample − OD_control)/(11.3 × sample weight(g)).

2.8. Microbial Analysis by 16S rRNA Gene Sequencing

The colon contents of mice were collected at the end of the experiment to extract genomic DNA using a QIAamp DNA Stool Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. DNA amplification and sequencing were performed by Major (Shanghai Major Bio-Pharm Technology, Shanghai, China). The 16S rRNA gene amplification methods targeting the V3–V4 hypervariable region and sequencing analysis were consistent with our previous experiments [32]. Alpha diversity was indicated by different indexes including richness, Shannon, Simpson, Chao1, and ACE, using the “vegan” package in R (R Foundation, Vienna, Austria). Principal component analysis (PCA) was performed using the R package “ade4” for beta-diversity. Statistical analysis of the taxonomic profiles was performed using STAMP v2.1.3, and statistically significant differences were calculated using one-way analysis of variance (ANOVA). Redundancy analysis (RDA) was performed using the “vegan” R package, with normalized operational taxonomic unit (OUT) abundance and environmental factors [33]. Linear discriminant analysis (LDA) effect size (LEfSe, http://huttenhower.sph.harvard.edu/galaxy) was used to identify the discriminating OTUs. The LDA threshold value was set to 2.0. Differentially abundant OTUs were used to calculate the microbial dysbiosis index (MDI) [34], using the following formula: log (total abundance of OTUs increased after DSS treatment)/(total abundance of OTUs decreased after DSS treatment). BugBase analysis was performed online (http://bugbase.cs.umn.edu) to obtain phenotypic information based on 16S rRNA gene sequences.

2.9. Statistical Analysis

GraphPad Prism 8.0 (GraphPad Software, San Diego, CA) was used for all analyses and data are presented as mean ± standard error of the mean (SEM). Student’s t test was used to analyze pairwise differences and one-way ANOVA corrected for multiple comparison.

3.1. Alleviating DSS-Induced Colitis in Female Mice with Different Narrow-Spectrum Antibiotics

Before DSS treatment, mice were exposed to Neo, Met, Van, and an antibiotic cocktail consisting of ampicillin, Neo, Met, and Van, respectively, in drinking water for one week, followed by DSS treatment for another week, while antibiotics continued to be administered (Figure 1(a)). The degree of splenic enlargement was calculated by the spleen index with the following formula: splenic weight (mg)/total body weight (g). As shown in Figure 1, antibiotic administration exhibited a protective effect on the decline of body weight (Figure 1(b)). The DAI score in the antibiotic-treated groups was much lower than that in the DSS group (Figure 1(c)), and the narrow-spectrum antibiotics were more effective than Abx, especially in the Neo group. The application of Neo and Met relieved the shortening and hyperemia of the colon in the DSS group (Figures 1(d) and 1(e)), whereas the Van and Abx groups showed severe fecal occult blood, and the cecum of these mice increased markedly (Figure 1(d)). In addition, antibiotic administration significantly alleviated splenic enlargement (Figure 1(f)). We also determined the serum levels of IL-6 and TNF-α. As shown in Figure 1(g), there was no significant difference in the levels of TNF-α among the groups, except DSS-treated mice that had slightly higher levels of TNF-α. However, IL-6 expression increased remarkably in DSS-treated mice (Figure 1(h)). All antibiotic-treated mice showed significantly alleviated elevation of IL-6 induced by DSS. Histological analysis further revealed differences in morphological damage of the colon among the groups (Figure 1(i)). Variable lymphocytic infiltration and crypt destruction of the epithelium were observed in all groups. Notably, DSS-treated mice displayed more severe crypt epithelial cell dropout and destruction, while the Met and Abx groups showed the least damage to colon tissue. The variation in immune cells in the spleen and colonic lamina propria was detected by fluorescence-activated cell sorting using the gating strategy detailed in Supplementary Figure S1. The proportions of CD45⁺ and CD4⁺ T cells in the spleen decreased in the DSS group, whereas antibiotic treatment upregulated it to normal levels (Figure S2A). As shown in Figure 1(j), there was no significant difference in the proportion of Th17 and regulatory T cells (Tregs) in the spleen among all groups. However, the ratio of CD11b⁺ myeloid cells significantly increased in the spleen after DSS treatment and was restored to the normal level after antibiotic treatment. Other subtypes of myeloid cells, including granulocytes, monocytes, and macrophagocytes, showed similar patterns of improvement (Figure S2A). As shown in Figure 1(k), the infiltration of CD11b⁺ myeloid cells in the colonic lamina propria was much higher in the DSS group and was attenuated to the control group level after antibiotic treatment. Other subtypes of myeloid cells, including granulocytes, monocytes, and macrophagocytes, exhibited similar trends (Figure S2B). In addition, the proportion of Th17 cells was downregulated in the Neo group; however, there was no significant difference in the ratio of Tregs. In summary, all the antibiotics tested in the present study exhibited positive effects on alleviating susceptibility to DSS-induced colitis in female mice and reduced inflammatory cell infiltration in the colon.

3.2. Alleviating DSS-Induced Colitis in Male Mice with Different Narrow-Spectrum Antibiotics

An identical experimental design and protocol were applied to male mice, as shown in Figure 2(a). Again, the Neo and Met groups showed a slight decrease in body weight (Figure 2(b)), similar to that observed in female mice. The DAI scores in the Neo and Met groups were much lower than those in the DSS group (Figure 2(c)). Antibiotic treatment reversed the shortening of the colon (Figure 2(e)), splenic enlargement (Figure 2(f)), and intestinal congestion (Figure 2(d)), but not in the Van group. Moreover, downregulation of serum TNF-α and IL-6 levels was observed in all antibiotic treatment groups (Figures 2(g) and 2(h)). Histological analysis indicated that the DSS group exhibited the most severe morphological damage to the colon (Figure 2(i)). Apparent lymphocytic infiltration, crypt destruction, and epithelium appeared in the Van group, while intestinal epithelial structure remained intact in the Neo, Met, and Abx groups. Compared with the CTR group, DSS treatment did not affect the proportion of CD45⁺ and CD4⁺ T cells in the spleen, but antibiotic treatment upregulated these ratios (Figure S3A). As shown in Figure 2(j), there was a downward trend in Th17 cells in the spleen in the Van and Met groups. However, there was no significant difference in the ratio of Th17 to Tregs, except for the Van group. The proportion of myeloid cells increased in the spleen after DSS treatment, whereas antibiotics decreased this proportion to normal levels (Figure 2(j)), among which monocytes, a subtype of myeloid cells, decreased most significantly (Figure S3A). Here, antibiotics resulted in a marked increase in myeloid cell infiltration in the colonic lamina propria compared with that in the DSS group (Figure 2(k)), which was opposite that observed in female mice (Figure 1(k)). The same pattern was observed for several myeloid cell subtypes (Figure S3B). However, the proportion of Th17 cells and the ratio of Th17 cells to Tregs in the colonic lamina propria decreased significantly, which may reduce inflammation. In conclusion, antibiotics could also alleviate the development of colitis in male mice through mechanisms different from those in female mice, in which Neo and Met exhibited superior protection performance over Van and Abx.

3.3. Narrow-Spectrum Antibiotics Alleviated Colitis in the αPD-L1 Treatment of Murine Colon Cancer: Female Mice Group

The effects of narrow antibiotics on the treatment of colitis associated with immunotherapy of murine colon cancer with αPD-L1 were first studied in female mice. Based on the above-mentioned results, Neo and Met were chosen for the following experiments because of their superior anti-inflammatory effects. The experimental design is shown in Figure 3(a), in which the acronyms are as follows: control (CT), αPD-L1(P), αPD-L1+DSS (PDS), αPD-L1+neomycin (PN), αPD-L1+metronidazole (PM), αPD-L1+DSS+neomycin (PDSN), and αPD-L1+DSS+metronidazole (PDSM). As shown in Figure 3(c), no significant body weight loss was observed in the PDSN and PDSM groups. Spleen enlargement was significantly attenuated in the PDSM group (Figure 3(d)), and colon shortening was alleviated to a certain extent, despite no statistical significance (Figure 3(e)). The serum levels of IL-6 were restored to normal in the antibiotic-treated groups (Figure 3(f)). Compared to the P group, there was an observable increase in the level of TNF-α in the PN and PM groups, which decreased after DSS treatment (Figure 3(g)). MPO is the predominant enzyme in neutrophil granulocytes. Figure 3(h) shows that intestinal MPO activity markedly increased after DSS treatment and decreased in the antibiotic-treated groups. In addition, αPD-L1 reduced the mRNA level in the cecal mucosa compared with CTR, whereas the other groups showed no significant changes (Figure 3(i)). DSS treatment led to a drastic increase in the mRNA levels of IL-1β in the cecal mucosa, which was significantly decreased by Met treatment (Figure 3(j)). Two weeks after DSS modeling, the intestinal epithelial structure was not fully restored in the PDS group (Figure 3(k)). In contrast, a certain recovery in the intestinal histological injury was achieved in the PDSN group, although the difference was not significant. These results confirm the effectiveness of Neo and Met in alleviating colitis during αPD-L1 treatment in mouse colon cancer. However, it should be noted that the tumor volume in the late stages of Neo and Met treatment increased more rapidly than that in the αPD-L1 treated group (Figure 3(b)). This undesirable effect calls our attention to the effects of DSS on immunotherapy with αPD-L1.

Thus, we examined immune cell infiltration in the colon lamina propria, MLNs, and tumor microenvironment. MLNs are vital sites for maturation and differentiation of intestinal immune cells. As shown in Figure 4(a), although the proportions of CD45⁺ and CD4⁺ T cells in MLNs were reduced in the PDS group, the proportion of CD11b⁺ cells increased prominently, which was consistent with the results observed in Figures 1(j) and 1(k). In addition, antibiotics induced a significant reduction in CD11b⁺ myeloid cells in MLNs, such as eosinophils, macrophages, monocytes, and granulocytes. In contrast, the proportion of CD8⁺ T cells showed no significant changes. Notably, the ratio of CD19⁺ cells showed an apparent upward trend in the P and PDS groups, suggesting that B cells (CD19⁺) may also play a role in the tumor immune response and inflammation. No significant differences were observed in the infiltration of inflammatory cells, such as eosinophils, macrophages, monocytes, and granulocytes, in the colonic lamina propria (Figure 4(b)). Analysis of lymphocyte infiltration in the tumor microenvironment showed that both DSS and antibiotics reduced the proportion of CD8⁺ T cells and CD8⁺ INFγ⁺ T cells, which might account for the reduced antitumor activity in the other groups (Figure 4(c)). These results suggest that the palliative effect of Neo and Met on colitis is not affected when combined with αPD-L1 antibodies, but antibiotics slightly hindered the antitumor efficacy of immunotherapy in female mice at the late stage, as shown by the faster rate of tumor volume growth (Figure 4(b)) and reduced tumor infiltrating CD8⁺ T cells (Figure 4(c)).

3.4. Narrow-Spectrum Antibiotics Alleviated Colitis in the αPD-L1 Treatment of Murine Colon Cancer: Male Mice Group

The animal protocol, which was the same as described above, is shown in Figure 5(a). As shown in Figure 5(b), the growth rate of the tumor volume in the PDS group was faster than that in the P group, similar to that observed in the female mouse group. It was noteworthy that application of both Neo and Met assisted the suppression of tumor growth, as compared with the P group (Figure 5(b)). This was different from that observed in female mice (Figure 4(b)). Both Neo and Met demonstrated a strong ability to alleviate colitis, as evidenced by reduced body weight loss in the PDSN and PDSM groups compared to the PDS group (Figure 5(c)). Neo and Met also showed strong protection against splenic enlargement and colon shortening (Figures 5(d) and 5(e)). Moreover, both antibiotics significantly reduced serum IL-6 levels with or without DSS treatment (Figure 5(f)). As for TNF-α, there was an upward tendency in the P group but a downward tendency in the PDS group (Figure 5(g)). This is different from the response of the female group shown in Figure 3(g). MPO activity was markedly elevated after DSS treatment and decreased in antibiotic-treated groups (Figure 5(h)), which was similar to the response of the female groups (Figure 3(h)). Moreover, DSS treatment led to a drastic increase in the mRNA levels of TNF-α and IL-1β in the cecal mucosa, which were significantly decreased by Neo and Met treatment (Figures 5(i) and 5(j)). Intestinal histological analysis also confirmed the protective effect of Neo and Met on the intestinal mucosa (Figure 5(k)), which was better than that observed in the female group (Figure 3(i)).

We then examined the effects of these two antibiotics and αPD-L1 on immune cell infiltration in the colonic lamina propria, MLNs, and tumor microenvironment. Figure 6(a) shows that there were no significant differences in the ratio of CD45⁺ cells among the groups in the MLNs. αPD-L1 treatment led to an increase in CD4⁺ T cells, while DSS and antibiotics resulted in a decrease in CD4⁺ T cells, which is similar to that observed in the female group. The proportion of CD11b⁺ cells dramatically increased after DSS treatment, whereas the application of antibiotics significantly suppressed the increase in CD11b⁺ cells in MLNs, especially granulocytes. The proportion of CD8⁺ T cells showed no obvious variation. Remarkably, the ratio of CD19⁺ cells increased in the PDS group and was restored to normal levels in the PDSN and PDSM groups, which is slightly different from the female group (Figure 5(a)).

Figure 6(b) shows that inflammatory cell infiltration in the colonic lamina propria was significantly higher in the PDS group than in the P group. Moreover, the proportion of CD11b⁺ cells, including eosinophils, macrophages, and monocytes, showed a sharp decrease. As shown in Figure 6(c), the application of colitis reduces the proportion of CD8⁺ T cells as well as CD8⁺ INFγ⁺ T cells, which may account for the reduced antitumor activity in the PDS group (Figure 6(b)). The application of Neo and Met slightly increased the proportion of CD8⁺ INFγ⁺ T cells, which is conducive to a better antitumor immune response.

3.5. The Connection of Colitis with Microbial Dysbiosis

To understand the sexual dimorphism in response to the treatment of colitis with antibiotics during αPD-L1 immunotherapy, we examined the alteration of gut microbiota by 16S rRNA sequence analysis. We compared the discrepant OTUs in both male and female mice before and after DSS treatment, as shown in Figure 7(a). These discriminative OTUs are the dominant species in colitis and are used to calculate the microbial dysbiosis index (MDI) for each group as follows: log (total abundance of OTUs increased after DSS treatment/total abundance of OTUs decreased after DSS treatment). Compared with the αPD-L1 group (P), MDI increased significantly after DSS treatment in both male and female mice and decreased in the Neo and Met treatment groups (Figure 7(b)). Notably, MDI in male mice was better restored after antibiotic treatment than that in female mice. This may explain the better response to antibiotics found in the male group. Additionally, we performed phenotypic predictions using BugBase. As shown in Figure 7(c), DSS treatment resulted in a striking increase in potentially pathogenic bacteria, which was significantly reduced by Neo and Met. The aerobic bacteria showed a similar tendency (Figure 7(d)). No significant differences in anaerobic bacteria were observed in any of the groups (Figure 7(e)). However, DSS-induced inflammation led to a slight increase in Gram-negative bacteria, and Neo and Met strongly inhibited these bacteria (Figure 7(f)). Gram-positive bacteria showed the opposite trend (Figure 7(g)). Given these results, it can be concluded that colitis is strongly associated with microbial dysbiosis. Male mice appear to have a better therapeutic effect against microbial dysbiosis than female mice, which may partially account for the aforementioned sexual dimorphism.

3.6. The Effect of Gut Microbiota on Sexual Dimorphism in Colitis and Immunotherapy

As shown in Figure 8(a), antibiotics and DSS led to a dramatic decrease in alpha diversity, including richness, chao1, and ACE, but Neo had less effect on the Shannon and Simpson indices. RDA indicated that sex had the greatest influence on gut microbiota, followed by DSS treatment (Figure 8(b)). PCA showed that there were significant differences in the microbiota composition between male and female mice (Figure 8(c)). Next, the analysis of the differential microorganisms at the genus level revealed that the abundance of Lachnospiraceae_NK4A136_group was meaningfully lower in male mice, whereas it was higher in female mice, and Muribaculaceae was the opposite (Figure 8(d)). PCA analysis and differential species identification in male mice are shown in Figure 8(e), in which the gut microbiota fluctuated considerably after Neo, Met, and DSS treatment in the male mouse group. The abundance of Muribaculaceae increased after αPD-L1 and DSS treatment. However, this trend was reversed in female mice (Figure 8(f)). Muribaculaceae have been shown to play an important role in azoxymethane/DSS-induced colorectal cancer, which may lead to diverse effects of antibiotics on colitis in male and female mice. αPD-L1 treatment significantly reduced the abundance of Lachnospiraceae_NK4A136_group, while Neo and Met increased it. In contrast, the abundance of Lachnospiraceae_NK4A136_group showed no significant differences in the female groups (Figure 8(f)). Lachnospiraceae has been confirmed to be associated with response to ICIs [35], which explains the disparate antitumor efficiency of antibiotics in male and female mouse groups. These results indicate that antibiotics have different effects on the abundance of some microorganisms in the gut and thus have a consequential impact on colitis and the therapeutic efficacy of αPD-L1.