Propionate-Producing Consortium Restores Antibiotic-Induced Dysbiosis in a Dynamic in vitro Model of the Human Intestinal Microbial Ecosystem

Abstract

Metabolic syndrome is a growing public health concern. Efforts at searching for links with the gut microbiome have revealed that propionate is a major fermentation product in the gut with several health benefits toward energy homeostasis. For instance, propionate stimulates satiety-inducing hormones, leading to lower energy intake and reducing weight gain and associated risk factors. In (disease) scenarios where microbial dysbiosis is apparent, gut microbial production of propionate may be decreased. Here, we investigated the effect of a propionogenic bacterial consortium composed of Lactobacillus plantarum , Bacteroides thetaiotaomicron , Ruminococcus obeum , Coprococcus catus , Bacteroides vulgatus , Akkermansia muciniphila , and Veillonella parvula for its potential to restore in vitro propionate concentrations upon antibiotic-induced microbial dysbiosis. Using the mucosal simulator of the human intestinal microbial ecosystem (M-SHIME), we challenged the simulated colon microbiome with clindamycin. Addition of the propionogenic consortium resulted in successful colonization and subsequent restoration of propionate levels, while a positive effect on the mitochondrial membrane potential (ΔΨ m ) was observed in comparison with the controls. Our results support the development and application of next generation probiotics, which are composed of multiple bacterial strains with diverse functionality and phylogenetic background.

Short term supplementation of celecoxib shifted butyrate production on a simulated model of the gut microbial ecosystem and amelioratedin vitroinflammation

“…Values of the similarity matrix were computed as the correlation between relative abundances of bacterial genera and metabolic variables, projected onto the space spanned by the first 3 components retained in the analysis (r ≥ 0.75). Genera in the plot were close to correlated variables in the treatment where they were more abundant (49).…”

“…VFA were used as benchmarks of community activity and extracted as per Truchado et al (48). Differences in VFA concentrations among treatments were compared using a repeated measures mixed model (49), with the lsmeans adjustment and Bonferroni correction for multiple comparisons (SAS, 2012). Statistical significance was assumed at P < 0.05.…”

“…Differences in alpha diversity and evenness measures were compared among treatments using a repeated measures mixed model in SAS (version 9.4, SAS Institute, Cary, USA). Beta diversity based on Chao and Bray-Curtis indices was used to examine dissimilarity and determine the impact of treatment and time on microbial community structure (49). PCoA was employed to visualize the differences among samples, using the vegan package in R (54).…”

“…The model was selected based on the Akaike Information Criterion (AIC). Differences among library size sample were accounted for with the offset option in proc GLIMMIX in SAS (49).…”

“…Bipartite networks highlighted functional associations among bacterial genera and metabolites (49), using a pair-wise similarity matrix obtained from a Regularised Canonical Correlation Analysis (57). Values of the similarity matrix were computed as the correlation between relative abundances of bacterial genera and metabolic variables, projected onto the space spanned by the first 3 components retained in the analysis (r ≥ 0.75).…”