Additional file 1 of Individualized microbiotas dictate the impact of dietary fiber on colitis sensitivity

Additional file 1: Table S1. Composition of the three purified diets used in this study. Diets used were composed by 10 kcal % of fat and supplemented with 200g per kg of cellulose (cellulose diet), 50g per kg of cellulose +150g per kg of inulin (inulin diet) and 50g per kg on cellulose + 150g per kg of psyllium (psyllium diet). Table S2. Composition of the BRM medium used in the in vitro MBRA system. Figure S1. Presentation of the MBRA system and schematic outline of the experimental plan used. (A) Overview of the MBRA system installed within an anaerobic chamber and inoculated with human microbiota. (B) Schematic representation of timeline used, samples collected, and analysis performed. Figure S2. Efficacy of the MBRA system to reproduce inter-individual variations in microbiota composition. (A) DNA was extracted from MBRA-generated samples collected at the 72h timepoint from chambers inoculated with the 6 human healthy donors used in the study. Microbiota composition was analysed through Illumina-based 16S rRNA gene sequencing. Principal coordinates analysis (PCoA) of the Bray Curtis matrix was computed through the QIIME2 pipeline. Dots are coloured by donor (N=9). Significance was determined using non-parametric multivariate analysis of variance (Permanova). (B) Taxonomical composition at the class level of samples collected at the 72h timepoint from the in vitro microbiota MBRA system inoculated with the 6 human healthy donors used in the study, with the 15 most abundant class being represented (N=9). (C) Taxonomical composition at the genus level of samples collected at the 72h timepoint from the in vitro microbiota MBRA system inoculated with the 6 human healthy donors used in the study represented (N=9). Figure S3. Inter-individual variations in fibre-induced metabolomic alterations. The in vitro microbiota MBRA system was inoculated with fecal slurry from 6 healthy donors and stabilized for 72h, at which point fibre treatment was applied using Cellulose, Inulin, or Psyllium. MBRA samples collected during the treatment phase (144h) were used for metabolomic analysis. Principal coordinate analysis of the Bray Curtis distance computed on metabolomic analysis performed on samples collected 72h after the initiation of fibres-treatment are presented. In A, all donors are included, with dots coloured by donor. In B-G, individual donors are represented every 6 donors, with dots coloured by treatment (N=3). Significance was determined using non-parametric multivariate analysis of variance (Permanova). Figure S4. Inter-individual variations in fibre-induced microbiota metabolomic alterations. The in vitro microbiota MBRA system was inoculated with fecal slurry from 6 healthy donors and stabilized for 72h, at which point fibre treatment was applied using Cellulose, Inulin, or Psyllium. Nineteen metabolites were quantified by HPLC on samples collected 72h after the initiation of fibres-treatment. Data are the means +/- S.E.M, with individual data points being represented (N=3). Significance was determined using a one-way ANOVA followed by Tukey’s multiple comparison test and significant differences were recorded as follow: *p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001. Figure S5. Inter-individual variations in metagenomic and metatranscriptomic based on the fibre sensitivity status. (A-C) Fecal samples from the donors were used for metagenomic analysis through shotgun sequencing. Obtained quality-filtered reads were grouped via MetaPhlAn 2.0 into taxonomical categories and via HUMAnN3 into functional categories. (A) Taxonomical features with a statistically significant difference between resistant donors and sensitive donors are presented. (B) CAZymes features with a statistically significant difference between resistant donors and sensitive donors are presented. (C) Relative abundances, for each donor used in this study, of 15 well-known fibre fermenting bacteria. Values are expressed as percentage, and dark blue indicates bacteria that are present in relatively high amount. (D-E) The in vitro microbiota MBRA system was inoculated with fecal slurry from 6 healthy donors and stabilized for 72h, at which point fibre treatment was applied using Cellulose, Inulin, or Psyllium. Total RNAs were extracted from MBRA samples collected during the treatment phase (120h - 144h) and subjected to metatranscriptomic analysis through shotgun sequencing. Obtained quality-filtered reads were grouped via HUMAnN3 into functional categories. (C) Principal coordinates analysis (PCoA) of the Bray Curtis distance computed on the generated HUMAnN3 table. All donors are included, and dots are coloured by donor (upper panel) or by treatment (lower panel). (D) HUMAnN3 identified pathway with a statistically significant difference between resistant donors and sensitive donors are presented. Figure S6. Schematic representation of the experimental design used for the mice experiement. (A) Upon arrival, germfree C57BL6/J WT mice undergoes fecal microbial transplantation with fecal suspension from donor 1 (fibres-resistant) or donor 2 (fibres-sensitive) (N=15 mice per donor). After one week of microbiota stabilization, mice were subsequently divided into three experimental groups and exposed to either cellulose- (grey), inulin- (purple) or psyllium- (green) supplemented diets for 25 days (N=5 mice per experimental group). On day 19 and for 6-days, Dextran Sulphate Sodium was added to the drinking water (2.5% w/v) to induce intestinal inflammation. (B) Schematic representation of timeline used, samples collected, and analysis performed. Figure S7. Impact of fibres consumption on intestinal microbiota bacterial load over time. Bacterial DNA was extracted from mice fecal samples and qPCR were performed on 16S rRNA in order to estimate bacterial density. For each donor, bacterial load is expressed as relative value compared to cellulose-treated chambers. Panels A-B represent all the experimental groups in mice transplanted with either donor 1(A) or donor 2 (B) microbiota. Panels C-D represent only inulin-treated (C) or psyllium-treated (D) groups in mice transplanted with either donor. Data presented are the means +/- S.E.M (N=5). Significance was determined using 2-way group ANOVA corrected for multiple comparisons with Bonferroni test compared to control group (Cellulose-treated chambers). Significance differences were recorded as follow: **p<0.01 and ****p<0.0001. Figure S8. Microbiota composition is differentially impacted by fibre treatment in mice colonized by fibre-resistant and fibre-sensitive donors. Bacterial DNA was extracted from mice fecal samples and subjected to Illumina-based 16S rRNA gene sequencing. (A-D) Beta diversity evolution computed through the QIIME2 pipeline using the Bray Curtis distance matrix. For each donor, evolution of microbiota composition is represented using distances expressed as relative value compared to cellulose-treated mice, defined as 1. Panels A-B represent all the experimental groups in mice transplanted with either donor 1(A) or donor 2 (B) microbiota. Panels C-D represent only inulin-treated (C) or psyllium-treated (D) groups in mice transplanted with either donor. (E-H) Alpha diversity evolution computed through the QIIME2 pipeline using the Evenness index. For each donor, evolution of microbiota composition is represented using distances expressed as relative value compared to cellulose-treated mice, defined as 1. Panels E-F represent all the experimental groups in mice transplanted with either donor 1(E) or donor 2 (F) microbiota. Panels G-H represent only inulin-treated (G) or psyllium-treated (H) groups in mice transplanted with either donor. Data presented are the means +/- S.E.M (N=5). Significance was determined using 2-way group ANOVA corrected for multiple comparisons with Bonferroni test compared to control group (Cellulose-treated chambers). Statistical differences were recorded as follow: *p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001. Figure S9. Bacterial DNA was extracted from mice fecal samples at day 19 and subjected to Illumina-based 16S rRNA gene sequencing. (A) Taxonomical composition, at the family level, of the fecal microbiota from mice transplanted with donor 1 and treated with either cellulose, inulin or psyllium (N=9). (B) Taxonomical composition, at the family level, of the fecal microbiota from mice transplanted with donor 2 and treated with either cellulose, inulin or psyllium (N=9). Data are represented as relative abundances (%). The most abundant families are represented, and the terms “others” refers to all families that represented less than 1% of the microbial communities. Figure S10. Microbiota pro-inflammatory potential is differentially impacted by fibre treatment in mice colonized by fibre-resistant and fibre-sensitive donors. Microbiota-derived expression of pro-inflammatory molecules lipopolysaccharide (A-D) and flagellin (E-H) were quantified using HEK reporter cells expressing TLR4 or TLR5, respectively. (A-D). Microbiota-derived expression of pro-inflammatory molecules lipopolysaccharide. For each donor, evolution of fecal lipopolysaccharide level is represented as relative value compared to cellulose-treated mice, defined as 1. Panels A-B represent all the experimental groups in mice transplanted with either donor 1(A) or donor 2 (B) microbiota. Panels C-D represent only inulin-treated (C) or psyllium-treated (D) groups in mice transplanted with either donor. (E-H) Microbiota-derived expression of pro-inflammatory molecules flagellin. For each donor, evolution of fecal flagellin level is represented as relative value compared to cellulose-treated mice, defined as 1. Panels E-F represent all the experimental groups in mice transplanted with either donor 1(E) or donor 2 (F) microbiota. Panels G-H represent only inulin-treated (G) or psyllium-treated (H) groups in mice transplanted with either donor. Data presented are the means +/- S.E.M (N=5). Significance was determined using 2-way group ANOVA corrected for multiple comparisons with Bonferroni test (# indicates p<0.05) compared to control group (Cellulose-treated chambers). Statistical differences were recorded as follow: *p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001. Figure S11. Inter-individual variations in fibre-induced body weight modulation. (A-D) Body weight evolution over time expressed as percentage compared to day 0 (start of the fibres treatment), defined as 100%. For each donor, body weight evolution over time is expressed as relative compared to cellulose-treated mice. Panels A-B represent all the experimental groups in mice transplanted with either donor 1(A) or donor 2 (B) microbiota. Panels C-D represent only inulin-treated (C) or psyllium-treated (D) groups in mice transplanted with either donor. Similar representations were used in panels E-H, but only for the DSS-treatment phase. Data presented are the means +/- S.E.M (N=5). Significance was determined using 2-way group ANOVA corrected for multiple comparisons with Bonferroni test compared to control group (Cellulose-treated chambers). Statistical differences were recorded as follow: *p<0.05, **p<0.01 and ***p<0.001.