MGQD is a Chinese herbal mixture composed of Pueraria montana var. lobata (Willd.) Maesen and S.M.Almeida ex Sanjappa and Predeep [Fabaceae; Puerarie Lobatae Radix] (24 g), Scutellaria baicalensis Georgi [Lamiaceae; Scutellariae Radix] (9 g), Coptis chinensis Franch. [Ranunculaceae; Coptidis Rhizoma] (9 g), Glycyrrhiza uralensis Fisch. ex DC. [Fabaceae; Glycyrrhiza Radix Et Rhizoma] (6 g), Zingiber officinale Roscoe [Zingiberaceae; Zingiberis rhizoma praeparatum] (9 g) and Talcum (9 g) at a ratio of 8:3:3:2:3:3 (w/w/w/w/w/w), which are the daily clinical doses for humans. Therefore, the total dosage of MGQD is 66 g. The equivalent dose of 60 kg for adults and mice was calculated according to the conversion coefficient of body surface area. Therefore, the MGQD was administered to mice at 5 g crude herbs/kg (low dose, equivalent to half of the dose of MGQD commonly used in clinical patients), 10 g crude herbs/kg (medium dose, equivalent to the dosage of MGQD commonly used in clinical patients) and 20 g crude herbs/kg (high dose, equivalent to twice the dosage of MGQD commonly used in clinical patients). According to the volume of intragastric administration of mice was 0.1 ml/10 g body weight, low, medium and high concentrations of MGQD extract were concentrated to 0.5 g/ml (0.5 g crude herbs per milliliter water), 1 g/ml (1 g crude herbs per milliliter water) and 2 g/ml (2 g crude herbs per milliliter water), respectively. All crude herbal medicines were purchased from Xiyuan Hospital, China Academy of Chinese Medical Sciences, and prepared by the Department of Pharmaceutical Preparation of Xiyuan Hospital in accordance with good manufacturing practices. The MGQD water extract was obtained according to the standard of TCM decoction method. 240 g Pueraria montana var. lobata (Willd.) Maesen and S.M.Almeida ex Sanjappa & Predeep, 90 g Scutellaria baicalensis Georgi, 90 g Coptis chinensis Franch., 60 g Glycyrrhiza uralensis Fisch. ex DC., 90 g Zingiber officinale Roscoe and 90 g Talcum were mixed and immersed in distilled water (1:8, w/v) for 1 h at room temperature, then Talcum was first decocted for 15 min, then all the herbs were decocted for 40 min, then filtered. After filtration, the residue was then boiled again for 30 min with an addition of 6 times volume of distilled water (1:6, w/v). After that, the extracted filtrates were combined and concentrated to 0.5, 1 and 2 g crude drug/mL, respectively. And the concentrated decoction was sealed and placed at 4°C for later use. The extraction rate of 21.261% was obtained by evaporation of the concentrate at 100°C. These doses used in this study were within the range of in vivo studies reported by others (Luo et al., 2019; Lin et al., 2020; Ma et al., 2021; Song et al., 2021).

The high concentration of MGQD extract (2 g crude herbs/mL) was determined using an Agilent 1,200 series HPLC system (Agilent Technologies, Palo Alto, CA, United States) equipped with a G1315D DAD detector and an Agilent 5 TC-C18 column (250 × 4.6 mm, 5 μm). The mobile phase consisted of acetonitrile (A) and 0.1% phosphoric acid solution (B) with a gradient elution as follows: 0–3 min for 15% A, 85% B; 3–45 min for 15–35% A, 85%–65% B. The detection wavelength was set at 275 nm, the flow rate was 1.0 ml/min, the loading volume was 10 μL and the column temperature was maintained at 30°C. According to the retention time and peak area of the reference standard, a total of 5 components of MGQD were detected, namely, puerarin, baicalin, wogonin, berberine chloride, and palmatine chloride. HPLC results are shown in Supplementary Figure S1.

Male C57BL/6J mice (6–8 weeks old; weighing 18–22 g) were purchased from SPF Biotechnology Co., Ltd (Beijing, China). The animals were housed under specific pathogen-free (SPF) conditions (12 h light/dark cycle, 21 ± 2°C with a relative humidity of 45 ± 10%) with free access to autoclaved food and water in the experimental animal center of Xiyuan Hospital, China Academy of Chinese Medical Sciences. All procedures and experiments carried out here were approved by the Ethics Review Committee for Animal Experimentation of Xiyuan Hospital, China Academy of Chinese Medical Sciences (Approval NO. 2019XLC003-2).

The three experimental procedures were as follows:

Ⅰ To investigate the protective effect of MGQD against UC, 60 mice were stochastically divided into six groups (n = 10/group): a control group, DSS group, low-dose MGQD group (GL), medium-dose MGQD group (GM), high-dose MGQD group (GH), and 5-ASA group. Except for the normal group, all other five groups received 3% (w/v) DSS diluted in drinking water to induce acute experimental UC for 7 days (Wu et al., 2021). At the same time, each treatment group was given the corresponding dose of drugs once per day orally for seven consecutive days, namely, GL (5 g/kg), GM (10 g/kg), GH (20 g/kg) and 5-ASA (100 mg/kg) (Jing et al., 2019). Each mouse was given an intragastric dose of 10 μL/g body weight once per day. The MGQD dose used here was calculated based on the clinical dose of raw materials. Meanwhile, the normal group and DSS group were intragastrically given the same dosages of distilled water. On day 8, all mice were sacrificed under ether anesthesia, and the length of the colon was measured. The cecal contents, feces, colon tissues, spleens and mesenteric lymph nodes (MLNs) were collected for further analysis. The detailed experimental procedure is presented in Figure 1A.

Ⅱ To investigate the role of intestinal flora in the improvement of MGQD in colitis, 30 mice were randomly divided into the ABX⁺DSS⁻GH^- group (ABX), ABX⁺DSS⁺GH⁻ group [ABX (DSS)], and ABX⁺DSS⁺GH⁺ group [ABX (DSS + GH)] with 10 mice in each group. All mice were given broad-spectrum antibiotic cocktails (1 g/L ampicillin, 1 g/L metronidazole, 0.5 g/L vancomycin and 0.5 g/L neomycin) for 5 days in advance to deplete the gut microbiota (Wu et al., 2020a). After 1 day of recovery, GH administration and UC induction were the same as those in Part Ⅰ. The mice were sacrificed under anesthesia on the 8th day. The experimental timeline is shown in Figure 2A.

Ⅲ FMT was carried out according to a protocol described previously with minor modifications (Sun et al., 2020; Li et al., 2021). In brief, 24 donor mice were randomly assigned to the control group, DSS group, and DSS + GH group (n = 8/group). GH administration and induction of UC were described in the Part Ⅰ experiment. Fresh feces from each donor group were collected every day under aseptic conditions and sealed in a sterile container on ice. Weighing all collected feces and 100 mg of stools were resuspended in 1 ml of sterile PBS. After stirring, four layers of sterile gauze were used to filter the suspension. The solution was then violently mixed using a table vortex for 10 s, followed by centrifugation at 800 g for 3 min, and the supernatant was pooled and used as graft material within 10 min prior to gavage. Thirty recipient mice with depleted gut microbiota were divided into the FMTCON group, FMTDSS group and FMTGH group (n = 10/group) and orally administered the collected suspension from the donor CON group, DSS group and DSS + GH group for seven continuous days at a concentration of 0.1 ml/10 g body weight. Gut microbiota depletion preparation and UC induction were the same as in Part Ⅱ. The experimental timeline is shown in Figure 3A.

The body weight, stool consistency, and hematochezia of each mouse were measured daily, and the DAI scores were calculated. For the parameters used for the calculation, refer to previous reports (Kihara et al., 2003).

The 1.5-cm distal colon was fixed in 10% neutral-buffered formalin, and 4-μm-thick sections were stained with hematoxylin and eosin (HE) according to standard procedures. Colonic pathology was evaluated based on a previously studied histological scoring system (Kihara et al., 2003).

For Th17 cell staining, monocytes isolated from the spleens and MLNs were incubated with a cell stimulation cocktail (plus protein transport inhibitors) (Thermo Fisher Scientific, Waltham, MA, United States) at 37°C and 5% CO₂ for 6–8 h. Surface staining was performed using FITC-anti-mouse CD4 (BioLegend Inc, San Diego, CA, United States). For intracellular staining, cells were fixed and permeabilized with fixation/permeabilization solution (BD Biosciences, CA, United States), and then antibodies against PE-anti-IL-17A antibody (BioLegend Inc, San Diego, CA, United States) were used. Similarly, for Treg cell staining, the cell surface markers FITC-anti-CD4⁺ (BioLegend Inc, San Diego, CA, United States) and PerCP-Cy5.5-anti-CD25⁺ (BioLegend Inc, San Diego, CA, United States) and the intranuclear staining marker AF647-anti-Foxp3+ (BioLegend Inc, San Diego, CA, United States) were used to specifically label the Treg cells. Finally, the stained cells were analyzed by flow cytometry.

The levels of IL-17A, IL-21, IL-4 and TGF-β in the colonic tissues were detected using ELISA kits according to the manufacturer’s instructions (MultiSciences Biotech Co., Ltd, Hangzhou, China).

According to the E. Z.N.A.^® Soil DNA Kit (Omega Bio-Tek, Norcross, GA, United States) protocol, fecal genomic DNA was extracted from 0.1 g of frozen colon contents. A NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, United States) was used to measure the DNA concentration and purity, and DNA integrity was detected by 1% agarose gel electrophoresis. The V3-V4 region was amplified with the 16S rDNA gene universal primers 338 F (5′-ACT​CCT​ACG​GGA​GGC​AGC​AG-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′). The PCR amplification system was 20 μL, the PCR product was purified, and the concentration was adjusted. Sequencing was performed on an Illumina MiSeq PE300 system (MajorBio Co., Ltd, Shanghai, China).

The SCFA standard used was a mixture of standard acetic acid, propanoic acid, isobutyric acid, butanoic acid, isovaleric acid, valeric acid, hexanoic acid and isohexanoic acid. All standards except propionic acid (Shanghai Ok-Kasei Co., Ltd.) were purchased from Sigma-Aldrich (Shanghai, China). The SCFAs were then extracted according to the manufacturer’s protocol (Majorbio Bio-Pharm Technology Co., Ltd, Shanghai, China). The analytical instrument used in this experiment to detect the extracted samples was an 8890B–5977 B GC/MSD (Agilent Technologies, Santa Clara, CA). Masshunter quantitative software (Agilent Inc, United States, version No. V10.0.707.0) was used to automatically identify and integrate the ion fragments of target short-chain fatty acids with default parameters, and manual inspection was also performed. The actual contents of SCFAs in the samples were calculated by calculating the detection concentration of each sample using the standard curve.

Usearch (version 7.0) clustered operational taxon units (OTUs) with 97% identity and identified and deleted chimeric sequences. Alpha diversity index calculation were conducted using Mothur (version 1.30.2). Beta analysis was analyzed based on the Bray-Curtis, unweighted UniFrac and weighted UniFrac distance algorithms for principal coordinate analysis (PCoA). Microbial distribution was visualized using the R package (version 2.15.3) based on community composition information at the taxonomic level. The Kruskal-Wallis H test was used to detect the dominant bacterial community differences between the groups. Linear discriminant analysis (LDA) effect size (LEfSe) was applied to determine taxa, which can characterize each population (LDA score >3.6) to discover biomarkers (Liu et al., 2020).

The significance level was determined by appropriate statistical analysis. GraphPad Prism 8.01 software was used for statistical analysis. The results were expressed as the means ± SD. Unpaired Student’s t-test was used for statistical analysis between the two groups. For more than two groups, one-way ANOVA and Bonferroni’s multiple comparison post test were used for statistical analysis. A p value <0.05 was recognized as statistically significant.

The HPLC fingerprint of high concentration of MGQD extract (2 g crude herbs/mL) at the wavelength of 275 nm is shown in Supplementary Figure S1. According to the retention time and peak area of the reference standard, five components (puerarin, baicalin, wogonin, berberine chloride, and palmatine chloride) in MGQD were identified. After calculation, the contents of puerarin, baicalin, wogonin, berberine chloride and palmatine chloride were about 12.21, 17.18, 0.76, 10.25 and 2.67 mg/ml in high concentration of MGQD extract (2 g crude herbs/mL), respectively.

In this study, the therapeutic efficacy of MGQD was evaluated in DSS-induced colitis mice. The experiment was carried out according to the timeline shown in Figure 1A. After 1 week of adaptive feeding, mice in the GL, GM, GH groups and 5-ASA groups were administered the corresponding treatment intragastrically from day 1 to day 7, and the process of DSS induction was synchronized. The normal group and the DSS model group were given the same dose of distilled water orally. Our results showed that in comparison with the normal group, the DSS group suffered significant weight loss and gross bloody stool. In contrast, the reduced body weight percentages of mice induced by DSS recovered significantly in the mice treated with GL, GM, GH, and 5-ASA (Figure 1B). In addition, the colon length in the DSS model group was significantly shortened compared to that in the control group. However, treatment with MGQD and 5-ASA significantly alleviated colon shortening (Figures 1C,D). The DAI score, an indicator of colitis severity calculated from weight loss, stool consistency, and gross bleeding, was dramatically elevated in the DSS group but significantly decreased in all dosages of MGQD and 5-ASA groups (Figure 1E). Furthermore, all doses of MGQD and 5-ASA remarkably improved histopathological changes in the colon, such as crypts destruction, loss of epithelial and goblet cells, and infiltration of a large number of inflammatory cells, and displayed significantly lower histopathological scores in comparison with the DSS model group (Figures 1F,G), indicating that both MGQD and 5-ASA showed profound effects on alleviating symptoms of colitis in the mice. Furthermore, the therapeutic effect of MGQD at the three doses was almost comparable to that of the 5-ASA group, and there was no remarkable discrepancy in the abovementioned parameters between the GL, GM, and GH groups.

To investigate whether the ameliorative effect of MGQD on DSS-induced colitis is related to intestinal microbiota, we utilized a mixture of quadruple antibiotics to eliminate the intestinal microflora of mice for 5 days in advance according to previous literature (Wu et al., 2020b; Li et al., 2021) (Figure 2A). After antibiotic exposure, significantly lower total DNA levels of the intestinal microbiota were observed in the feces of antibiotic-exposed mice, implying that most of the intestinal microbiota was essentially eliminated (Figure 2B). Interestingly, the mitigative effect of MGQD seemed to be abrogated when the intestinal flora was depleted in mice, as evidenced by the ABX (DSS) and ABX (DSS + GH) groups showing striking similarities in weight loss percentage (Figure 2C), DAI score (Figure 2D), colon length and gross lesions (Figures 2E, F), and colon histological changes (Figures 2G, H), indicating no significant difference between the two groups. The above results revealed that MGQD treatment ameliorated the inflammatory damage of the colonic mucosa resulting from DSS in a manner dependent on the gut microbiota.

To further illustrate the protective effect of MGQD on DSS-induced colitis mediated by the gut microbiota, the FMT method was employed (Figure 3A). For better colonization, all recipient mice were given a cocktail of antibiotics prior to FMT to create conditions for intestinal microbiome depletion. Transferring fecal microbiota from control donors and GH-treated donors to DSS mice, namely, the FMTCON and FMTGH groups, both showed significantly attenuated colonic inflammation compared with mice receiving fecal microbiota from DSS-treated donors (FMTDSS), as manifested by the decreased weight loss percentage and DAI score and increased colon length. HE staining of the colon in the FMTGH and FMTCON groups both displayed less inflammatory cell infiltration, a relatively complete colon structure, less mucosal damage and a lower histological score than those in the FMTDSS group (Figures 3B–G). In conclusion, our results proved that the microbiota altered by MGQD effectively alleviated colitis induced by DSS, and fecal transplantation of MGQD-altered microbiota recreated the protective effect of MGQD, suggesting that the intestinal microbiota was responsible for the relieving effect of MGQD on intestinal inflammation.

Given that dysregulation of the intestinal microbiota constitutes a central feature of IBD, we conducted 16S rRNA sequencing analysis in fecal bacterial DNA extracted from Con, DSS, GL, GM, GH and 5-ASA group mice to explore whether MGQD treatment changed the profile of the fecal microbiome challenged by DSS. We first noted that mice treated with all doses of MGQD and 5-ASA had a significantly higher Shannon index, an indicator of α diversity in the gut microbiome, compared with that of the DSS group (Figure 4A). Additionally, β-diversity analysis was used to assess the discrepancies between microbial communities. PCoA based on Weighted UniFrac distance showed that the DSS model group and the control group were aggregated separately, indicating that the composition and structure between the control group samples and that of the DSS model group were apparently different, whereas the distance of all treatment groups was closer to that of the control group in comparison with that of the DSS model group (Figures 4B, C). These results suggested that MGQD altered the diversity reduction of intestinal microorganisms and the structure of microbial communities challenged by DSS.

To further clarify the landscape of the bacterial community among the six groups, we then explored the impact of MQGD on the gut microbiome and its relative abundance. In terms of differences in species composition at the genus level, we found that some taxonomic microbiota exhibited a relatively higher abundance in the DSS-treated group, including Bacteroides, Turicibacter, Escherichia-Shigella, Alistipes, and Clostridium_sensu_stricto_1, while displaying a lower abundance of SCFA-producing bacteria, such as Lachnospiraceae_NK4A136_group and Lactobacillus than in the control group. Remarkably, all MGQD-treated and 5-ASA groups significantly reversed the relative abundances of Escherichia-Shigella and Clostridium_Sensu_Stricto_1 to normal levels. Sequences that could not be classified were designated as “no rank” (Figure 4D). The above results were also confirmed by the Kruskal-Wallis H test (Supplementary Figure S2). The taxonomic composition of the six groups were also compared at the level of phylum, class, order and family (Supplementary Figures S3, 4).

Subsequently, LEfSe was performed to identify significant differences in the predominant bacterial community from phylum to genus among the six groups that may serve as a biomarker for colitis (LDA>3.6, p < 0.05). As shown in Figures 5A, B, Bacteroidaceae (from the family to the genus Bacteroides), Erysipelotrichaceae (from the class to the family) and Escherichia coli-Shigella (from the phylum Proteobacteria to the genus) were the core strains leading to the imbalance of intestinal flora in the DSS group. Nevertheless, the family Ruminococcaceae, the g_norank_f__Lachnospiraceae, and the family Lachnospiraceae exhibited comparative enrichment in the GL, GM, and GH groups, respectively, which might be related to the effect of anticolitis mediated by MGQD. In conclusion, the application of MGQD strikingly influenced the diversity and structure of the gut microflora.

Emerging evidence has shown that FMT can prevent UC through modification of the gut microbiota (Moayyedi et al., 2015; Goyal et al., 2018; Burrello et al., 2019). To obtain stronger evidence, we similarly clarified the intestinal microbiome profile by sequencing the V3-V4 region of bacterial 16S rRNA from colon contents isolated from FMT receptor mice. Consistent with the alterations in the diversity and composition of the microbial community in the control and MGQD intervention groups, mice in the FMTCON and FMTGH groups both harbored significantly higher microbial diversity than those in the FMTDSS group, as evidenced by increased Sobs, Shannon, ACE and Chao indices on OTU abundance (Figures 6A–D). To examine alterations in intestinal microbiota composition, PCoA based on the Bray-Curtis distance, Unweighted UniFrac and Weighted UniFrac algorithm analysis showed a statistically significant separation of microbiota composition between the FMTCON and FMTDSS groups and a closer distance between the FMTGH group and the FMTCON group (Figures 6E–G).

Taxa-based analysis revealed significant differences in intestinal microbial composition at the phylum and genus levels. At the phylum level, the dominant species of colonic microbiota in the three groups were Firmicutes, Bacteroidetes, Verrucobacteria, Proteobacteria, and Actinobacteria. Compared with the FMTCON group, the relative abundance of Firmicutes and Actinobacteria decreased significantly, and the relative abundance of Proteobacteria and Bacteroidetes was augmented significantly in the FMTDSS group, but the FMTGH group could significantly reverse this phenomenon (Figures 7A, B). Detailed information on the relative abundance of class, order, and family microbial communities is given in Supplementary Figure S5. At the genus level, Bacteroides, Akkermansia, Blautia, Enterobacter and Escherichia-Shigella levels were relatively higher in the FMTDSS group, while norank_f_Muribaculaceae, Lachnospiraceae_NK4A136_group, Turicibacter, Lactobacillus, Eubacterium_fissicatena_group, and unclassified_f__Lachnospiraceae levels were relatively lower than those observed in the FMTCON group. Pleasingly, the FMTGH group normalized the relative abundance of all the bacteria mentioned above to a large extent that were induced by DSS (Figures 7C, D).

Subsequently, we also used LEfSe to compare high-dimensional categories and identify the predominant bacterial communities from the three groups. Our results showed that Bacteroides (from the phylum to the genus) was the core bacterial species that caused the imbalance of the intestinal flora in the FMTDSS group. In the FMTCON group, genus Lachnospiraceae_NK4A136_group and family Muribaculaceae were the most enriched with an LDA score of 4.72 (p = 0.007). Firmicutes was the most enriched in the FMTGH group with the highest LDA score of 5.163 (p = 0.003) (Figure 8). In general, transfer of MGQD-treated fecal bacteria significantly compensated for the decreased diversity and discordance of the intestinal flora caused by DSS.

In view of the key role of SCFAs in microbial-host interactions and IBD pathogenesis, we next explored whether MGQD had an impact on the production of SCFAs in colitis mice. The level of SCFAs in the cecal contents was measured by targeting metabolomics. As shown in Figure 9A, compared with the control group, the contents of acetic acid, butanoic acid, valeric acid, hexanoic acid and total SCFAs in the cecal contents of the DSS model group mice were sharply decreased. Compared with the control group, the contents of propanoic acid, isobutyric acid and isovaleric acid were decreased in the DSS group, but there was no significant difference. In comparison with DSS treatment alone, GL significantly improved the levels of the seven SCFAs mentioned above and the total SCFAs. Compared with the DSS group, the contents of these seven SCFAs and total SCFAs were increased in the GM group, but there were no statistically significant differences in butyric acid, hexanoic acid or total SCFAs. Compared with the DSS model group, the GH group had significantly increased contents of the other 6 SCFAs and total SCFAs except for hexanoic acid. 5-ASA dramatically enhanced the contents of acetic acid, isobutyric acid and isovaleric acid, and the contents of propanoic acid, butyric acid, valeric acid, hexanoic acid and total SCFAs were also observed to increase, but there was no significant difference. No significant difference was observed in isohexanoic acid among the 6 groups (data not shown).

To further clarify whether the enhancement of metabolite SCFAs was due to alterations in intestinal microbiota after MGQD administration, targeted metabolomics analysis was also carried out among the FMT groups. The cecal contents of the mice treated with FMTDSS showed decreased levels of acetic acid, propanoic acid, butyric acid, isobutyric acid, and total SCFAs compared with those of the FMTCON group, but there was no significant difference in acetic acid between the FMTCON and FMTDSS groups. Similarly, compared with the FMTDSS group, the FMTGH group exhibited augmented contents of acetic acid, propanoic acid, butanoic acid, isobutyric acid and total SCFAs, but there was no significant difference in acetic acid between the FMTGH and FMTDSS groups (Figure 9B). No significant differences were observed among the three groups in terms of hexanoic acid, isohexanoic acid, isovaleric acid and valeric acid (data not shown). In summary, these results indicated that the anti-colitis properties of MGQD may be associated with regulating the structure and composition of the intestinal bacterial community, which were related to the increased level of microbial metabolite SCFAs.

The expression levels of the inflammatory cytokines IL-4, TGF-β, IL-17A and IL-21 in the colon tissues of each group were quantitatively analyzed by the corresponding ELISA kits to investigate the inflammatory response to MGQD treatment. As shown in Figure 10A, compared with the normal group, the expression of the anti-inflammatory cytokines IL-4 and TGF-β was dramatically reduced in the DSS model group, while the expression of the proinflammatory cytokines IL-17A and IL-21 was significantly increased. Compared with the DSS model group, the content of TGF-β was significantly increased in all dosages of MGQD groups, and the level of IL-4 was significantly increased only in the GH and 5-ASA groups. The content of IL-17A in all dosages of MGQD groups and in the 5-ASA group was significantly decreased, and the concentrations of IL-21 in the GL, GH, and 5-ASA groups were significantly decreased in comparison with that of the DSS group. The abovementioned cytokines in colon tissue were also detected in the intestinal microbial depletion groups. Notably, IL-17A, IL-21, IL-4 and TGF-β showed similar levels, with no significant difference between the ABX (DSS) and ABX (DSS + GH) groups (p > 0.05) (Figure 10B).

Targeted metabonomics showed that FMTGH increased the metabolite SCFAs; thus, we hypothesized that FMT could also have an impact on the expression of cytokines. As expected, colonic tissues in the FMTCON and FMTGH groups both had lower levels of IL-21 and IL-17A but higher levels of TGF-β and IL-4 than colonic tissues in the FMTDSS group with a significant difference (Figure 10C). In conclusion, MGQD significantly alleviated colitis by downregulating the levels of proinflammatory cytokines and upregulating anti-inflammatory cytokines.

Considering that the Treg/Th17 imbalance plays an important role in the maintenance of intestinal microecosystem homeostasis and intestinal immune function, flow cytometry was used to detect T cell response phenotypes in spleens and MLNs to explore the effect of MGQD intervention on the Treg/Th17 balance. Compared with the normal group, the ratio of CD4+IL-17A+/CD4+ lymphocytes in the spleens and MLNs in the DSS model group was significantly increased, while the ratio of CD4⁺CD25 + Foxp3+/CD4+ lymphocytes was significantly decreased, but all dosages of MGQD treatment groups and the 5-ASA group significantly reversed this phenomenon (Figure 11). In conclusion, MGQD administration selectively promoted the frequency of Treg cells and suppressed the proportion of Th17 cells against DSS-induced colitis to maintain intestinal homeostasis.

To further confirm whether the restoration of the Treg/Th17 balance after MGQD administration relied on the intestinal microbiota, the phenotypes of single-cell suspends isolated from the spleens and the MLNs of mice in the ABX model experiment were also analyzed by flow cytometry. Interestingly, neither the percentage of Treg cells nor the percentage of Th17 cells in the spleens and MLNs showed significant differences between the ABX (DSS + GH) and ABX (DSS) groups (Figures 12A,B). Phenotypic analysis of the frequency of Treg and Th17 cells in the spleens and MLNs was also performed in the FMT experiment. Compared with the FMTCON group, the spleens and MLNs of the FMTDSS group showed a decreased frequency of Treg cells but a higher percentage of Th17 cells. In sharp contrast, the FMTGH group had a higher percentage of Treg cells but a lower proportion of Th17 cells than the FMTDSS group. The results showed that transferring MGQD-treated fecal microbiota reversed the imbalance of Th17 and Treg cells in UC mice (Figures 12C,D). These results demonstrated that the intestinal microflora modified by MGQD administration was associated with the modulation of the Treg/Th17 balance. In conclusion, MGQD regulated the Treg/Th17 balance in a gut microbiota-dependent manner.