In vitro Selection of Probiotics for Microbiota Modulation in Normal-Weight and Severely Obese Individuals: Focus on Gas Production and Interaction With Intestinal Epithelial Cells

Abstract

The intestinal microbiota plays important roles in the maintenance of health. Strategies aiming at its modulation, such as probiotics, have received a deal of attention. Several strains have been studied in different in vitro models; however, the correlation of results obtained with the in vivo data has been limited. This questions the usefulness of such in vitro selection models, traditionally relying on over-simplified tests, not considering the influence of the accompanying microbiota or focusing on microbiota composition without considering functional traits. Here we assess the potential of six Bifidobacterium, Lactobacillus and Lacticaseibacillus strains in an in vitro model to determine their impact on the microbiota not just in terms of composition but also of functionality. Moreover, we compared the responses obtained in two different population groups: normal-weight and severely obese subjects. Fecal cultures were conducted to evaluate the impact of the strains on specific intestinal microbial groups, on the production of short-chain fatty acids, and on two functional responses: the production of gas and the interaction with human intestinal epithelial cells. The response to the different probiotics differed between both human groups. The addition of the probiotic strains did not induce major changes on the microbiota composition, with significant increases detected almost exclusively for the species added. Higher levels of gas production were observed in cultures from normal-weight subjects than in the obese population, with some strains being able to significantly reduce gas production in the latter group. Moreover, in obese subjects all the Bifidobacterium strains tested and Lacticaseibacillus rhamnosus GG were able to modify the response of the intestinal cells, restoring values similar to those obtained with the microbiotas of normal-weight subjects. Our results underline the need for the screening and selection of probiotics in a target-population specific manner by using appropriate in vitro models before enrolling in clinical intervention trials.

Keywords: Bifidobacterium; Lactobacillus; SCFA; gas production; gut microbiota; in vitro model; probiotics; severe obesity.

FIGURE 1

Absolute levels (Log₁₀ CFU/mL) of intestinal microbial groups determined by qPCR in fecal cultures of (A) OB and (B) NW subjects. For each condition, the box and whiskers plot represent median, interquartile range and minimum and maximum values obtained in each human group (NW or OB). Different letters above the boxes indicate significant differences (p-value <0.05) among probiotic strains for the microbial groups considered.

FIGURE 2

Increments in ascending order, with respect to time 0, in the concentration (mM) of the major short-chain fatty acids (acetic, propionic, and butyric acids) after 24 h of incubation with different probiotic strains in fecal cultures from OB (A) and NW (B) groups. Differences are shown for each SCFA, columns that do not share the same letter are significantly different (p < 0.05).

FIGURE 3

Real-time monitoring the interaction with HT29 intestinal epithelial cells between (A) supernatants obtained after fecal culture with probiotic and (B) a mixture of a probiotic strain with the gut microbiota from NW and OB population groups. Values (media ± SD) correspond to the AUC resulting from monitoring Cell Index (CI) during 10 h. Significant differences (p-value <0.05) represent the comparison of results before and after probiotic addition in each condition.