Novel Gut Microbiota Patterns Involved in the Attenuation of Dextran Sodium Sulfate-Induced Mouse Colitis Mediated by Glycerol Monolaurate via Inducing Anti-inflammatory Responses

Abstract

Inflammatory bowel disease (IBD) is a type of immune-mediated chronic and relapsing inflammatory gastrointestinal symptoms. IBD cannot be completely cured because of the complex pathogenesis. Glycerol monolaurate (GML), naturally found in breast milk and coconut oil, has excellent antimicrobial, anti-inflammatory, and immunoregulatory functions. Here, the protective effect of GML on dextran sodium sulfate (DSS)-induced mouse colitis and the underlying gut microbiota-dependent mechanism were assessed in C57BL/6 mice pretreated or cotreated with GML and in antibiotic-treated mice transplanted with GML-modulated microbiota. Results showed that GML pretreatment has an advantage over GML cotreatment in alleviating weight loss and reducing disease activity index (DAI), colonic histological scores, and proinflammatory responses. Moreover, the amounts of Lactobacillus and Bifidobacterium and fecal propionic acid and butyric acid were elevated only in mice pretreated with GML upon DSS induction. Of note, fecal microbiota transplantation (FMT) from GML-pretreated mice achieved faster and more significant remission of DSS-induced colitis, manifested as reduced DAI, longer colon, decreased histological scores, and enhanced colonic Foxp3⁺ regulatory T cells (Tregs) and ratio of serum anti-inflammatory/proinflammatory cytokines, as well as the reconstruction of microbial communities, including elevated Helicobacter ganmani and decreased pathogenic microbes. In conclusion, GML-mediated enhancement of Bifidobacterium and fecal short-chain fatty acids (SCFAs) could be responsible for the anticolitis effect. FMT assay confirmed that gut microbiota modulated by GML was more resistant to DSS-induced colitis via elevating beneficial H. ganmani and establishing Treg tolerant phenotype. Importantly, colitis remission induced by GML is associated with novel gut microbiota patterns, even though different microbial contexts were involved. IMPORTANCE The gut microbiota, which can be highly and dynamically affected by dietary components, is closely related to IBD pathogenesis. Here, we demonstrated that food-grade glycerol monolaurate (GML)-mediated enhancement of Bifidobacterium and fecal SCFAs could be responsible for the anticolitis effect. FMT assay confirmed that gut microbiota modulated by GML was more resistant to DSS-induced colitis via elevating beneficial H. ganmani and establishing Treg tolerant phenotype. Collectively, colitis remission induced by GML is associated with novel gut microbiota patterns, even though different microbial contexts were involved, which further provided a perspective to identify specific microbial members and those responsible for the anticolitis effect, such as Bifidobacterium and Helicobacter.

Keywords: Bifidobacterium; Helicobacter; IBD; Tregs; bifidobacterium; glycerol monolaurate; inflammatory bowel disease; microbiota pattern.

FIG 1

GML pretreatment prevents DSS-induced acute colitis. (A) The schematic diagram of experimental design and procedure. (B) Body weight changes. (C) The disease activity index (DAI) assessment. (D and E) Colon length at day 7. (F and G) H&E staining analysis (F) and histology score (G) of colonic tissues. Each dot represents an individual mouse. (B to G) n = 6 to 15 mice/group. *, P < 0.05; **, P < 0.01; ***, P < 0.001, versus Veh group and #, P < 0.05; ##, P < 0.01, versus DSS group, determined by a one-way ANOVA followed by Bonferroni’s post hoc test.

FIG 2

GML pretreatment reduced DSS-induced proinflammatory status. (A to F) Colonic mRNA levels of Il6 (A), Tnfa (B), Il23p19 (C), Il10 (D), Occludin (E), and Jam-1 (F) assessed by RT-qPCR. (G) In vivo intestinal permeability characterized as median fluorescence intensity (MFI) of FITC-dextran. (H) Serum levels of IL-6 assessed by enzyme-linked immunosorbent assay (ELISA). (I and J) Volcano plot of differentially expressed genes (DEGs) (I) and heatmap representation of DEG levels from colon tissue between pre-GML+DSS and DSS groups as measured by RNA-Seq technology (J). Twenty-three genes were upregulated and 8 genes were downregulated (log₂|fold change| ≥ 1.45, adjusted P < 0.05) between pre-GML+DSS and DSS groups. Expression level was represented as each gene’s fragments per kilobase million (FPKM) value. *, P < 0.05; **, P < 0.01; ***, P < 0.001, versus Veh group and #, P < 0.05, ##, P < 0.01, versus DSS group for panels A to H, determined by a one-way ANOVA followed by Bonferroni’s post hoc test.

FIG 3

GML pretreatment prevents DSS-induced microbial dysbiosis. (A) α-Diversity of gut microbiota represented by Observed_species, Chao1, Shannon, and Simpson indices. (B) β-Diversity of gut microbiota based on weighted UniFrac distance and unweighted UniFrac distance. (C) Relative abundances of gut microbiota at the phylum level. (D) Heatmap depicting the log₁₀|relative abundance| of top 30 microbial genera, when the taxa with the relative abundance as zero were expressed as −4. (E) Relative abundance of five differential microbial genera. (F) The production of microbial metabolite SCFAs from feces among the four different groups. Each dot represents an individual mouse (n = 9 to 10 mice/group). *, P < 0.05; **, P < 0.01; ***, P < 0.001, versus Veh group; #, P < 0.05; ##, P < 0.01, versus DSS group; †, P < 0.05; ††, P < 0.01, versus pre-GML+DSS group, determined by a nonparametric factorial Kruskal-Wallis test followed by the Mann-Whitney U test when P was <0.05.

FIG 4

FMT from GML-treated mice prevented DSS-induced colitis in antibiotic-treated mice. (A) The schematic diagram of experimental design and procedure of FMT experiment. (B) Body weight changes. (C) DAI assessment. (D and E) Colon length measurement. (F and G) H&E staining analysis (F) and histology score (G) of colonic tissues. (B to E) n = 12 to 15 mice/group; (F and G) n = 8 mice/group. §, P < 0.05; §§, P < 0.01; §§§, P < 0.001, versus V-FMT+DSS group, determined by a two-tailed Student t test.

FIG 5

FMT from GML-treated mice promoted anti-inflammatory response upon colitis induction. (A) CD4⁺ CD25⁺ Foxp3⁺ lymphocytes of colonic lamina propria were analyzed by flow cytometry. (B) Indicated percentages are from viable cells gated on FVD (n = 4 mice/group). (C) Colonic mRNA levels of Foxp3, Rorgt, Il6, Il23p19, and Il10 assessed by RT-qPCR (n = 6 to 8 mice/group). (D) Serum levels of IL-6, IL-17, IL-1β, TGF-β1, IL-10, and IL-22 assessed by ELISA (n = 9 to 12 mice/group). §, P < 0.05; §§, P < 0.01; §§§, P < 0.001, versus V-FMT+DSS group, determined by a two-tailed Student t test.

FIG 6

FMT of GML-treated mice reshaped the gut microbiota in antibiotic-treated mice upon colitis induction. (A) Three-dimensional (3D) PCoA plot based on OTU-level Bray-Curtis dissimilarity matrix and PERMANOVA. (B) Relative abundances of gut microbiota at the phylum level. (C) Heatmap depicting the log₁₀|relative abundance| of top 30 microbial genera, when the taxa with the relative abundance as zero were expressed as −4. (D) Relative abundance of two differential microbial genera between V-donor group and G-donor group. (E) Relative abundance of six differential microbial genera between V-FMT+DSS group and G-FMT+DSS group. (F) Fecal SCFA quantification at day 15 between donor mice, as well as between V-FMT+DSS and G-FMT+DSS groups. Each dot represents an individual mouse (n = 6 mice/group for donor mice, n = 11 mice/group for recipient mice). $, P < 0.05, versus V-donor group. §, P < 0.05, and §§§, P < 0.001, versus V-FMT+DSS group, determined by a nonparametric factorial Kruskal-Wallis test followed by the Mann-Whitney U test when P was <0.05.

FIG 7

The Spearman correlation between the specific or differential microbial taxa and the colitis-related indicators among Veh, DSS, pre-GML+DSS, and co-GML+DSS groups (A) and between V-FMT+DSS and G-FMT+DSS groups (B). Red squares indicate positive correlations, and green squares indicate negative correlations. ▴, 0.05 < P < 0.1, means that correlations remained trending significant, and *, P < 0.05, and **, P < 0.01, mean statistically significant after correction for false-discovery rate.