Effects of cRG-I Prebiotic Treatment on Gut Microbiota Composition and Metabolic Activity in Dogs In Vitro

Abstract

Low-dose carrot rhamnogalacturonan-I (cRG-I) has shown consistent modulatory effects on the gut microbiota and immune function in humans. In this study we investigated its effects on the microbial composition and metabolite production of the gut microbiota of small (5-10 kg), medium-sized (10-27 kg), and large (27-45 kg) dogs, using inulin and xanthan as comparators. Fecal samples from six dogs of each size group were evaluated. Overall microbiome composition, assessed using metagenomic sequencing, was shown to be driven mostly by dog size and not treatment. There was a clear segregation in the metabolic profile of the gut microbiota of small dogs versus medium-sized and large dogs. The fermentation of cRG-I specifically increased the levels of acetate/propionate-producing Phocaeicola vulgatus. cRG-I and inulin were fermented by all donors, while xanthan fermentation was donor-dependent. cRG-I and inulin increased acetate and propionate levels. The responses of the gut microbiota of different sized dogs to cRG-I were generally consistent across donors, and interindividual differences were reduced. This, together with the significant increase in P. vulgatus during fermentation in both this study and an earlier human ex vivo study, suggests that this abundant and prevalent commensal species has a core capacity to selectively utilize cRG-I.

Keywords: cRG-I; canine gut microbiome; carrot rhamnogalacturonan-I; dietary fiber; metagenomics; prebiotic.

Figure 1

Elements of study design. (A) Chemical structures of test fibers. (B) Experimental setup. cRG-I, carrot rhamnogalacturonan-I; IN, inulin; XA, xanthan.

Figure 2

Changes in microbial community diversity. (A) Alpha diversity (Shannon index). (B) Beta diversity (Bray–Curtis index) within cRG-I treatment group was significantly smaller than that within other conditions (t-test, p-values < 0.01). Black dots in the figure represent outliers. cRG-I, carrot rhamnogalacturonan-I.

Figure 3

Changes in microbial community composition. (A) PCA plots by test article. (B) PCA plots by dog size. cRG-I, carrot rhamnogalacturonan-I.

Figure 4

Heatmaps depicting microbial abundance. (A) The relative abundance of the top 25 most abundant genera. (B) The relative abundance of key families. Kruskal–Wallis, p < 0.05, size_treatment (e.g., cRG-I_small vs. cRG-I_large). cRG-I, carrot rhamnogalacturonan-I.

Figure 5

Relative abundance of (A) Phocaeicola vulgatus per treatment group and (B) Faecalimonas umbilicata per treatment group. Black dots in the figure represent outliers. cRG-I, carrot rhamnogalacturonan-I.

Figure 6

SCFAs acetate, propionate, and butyrate and total SCFA levels at 48 h in small (top row), medium-sized (middle row), and large (bottom row) dogs. * p-value < 0.05 versus blank (paired two-sided Student’s t-test). Orange dot in the figure represents an outlier. cRG-I, carrot rhamnogalacturonan-I; SCFA, short-chain fatty acid.

Figure 7

PCA-X-score plot based on LA-REIMS data (24 h and 48 h). (A) All biological samples (n = 144). (B) Blank versus inulin. (C) Blank versus cRG-I. (D) Blank versus xanthan. Associated iQC-samples were excluded from analysis. LA-REIMS metabolomic data were generated in negative ionization mode. OPLS-DA validation parameters for evaluation of various dog sizes and different treatments based on LA-REIMS metabolic fingerprints are also shown. ANCOVA, analysis of covariance; cRG-I, carrot rhamnogalacturonan-I; iQC, internal quality control; LA-REIMS, laser-assisted rapid evaporative ionization mass spectrometry; OPLS-DA, orthogonal partial least squares discriminant analysis; PCA-X, unsupervised principal component analysis, t time; Q²Y quality parameter; R²X the model performance; R²Y the ability to predict the Y-data for the specifically used dataset.