Galacto-oligosaccharide preconditioning improves metabolic activity and engraftment of Limosilactobacillus reuteri and stimulates osteoblastogenesis ex vivo

Abstract

A probiotic-related benefit for the host is inherently linked to metabolic activity and integration in the gut ecosystem. To facilitate these, probiotics are often combined with specific prebiotics in a synbiotic formulation. Here, we propose an approach for improving probiotic metabolic activity and engraftment. By cultivating the probiotic strain in the presence of a specific prebiotic (preconditioning), the bacterial enzymatic machinery is geared towards prebiotic consumption. Today, it is not known if preconditioning constitutes an advantage for the synbiotic concept. Therefore, we assessed the effects galacto-oligosaccharide (GOS) addition and preconditioning on GOS of Limosilactobacillus reuteri DSM 17938 on ex vivo colonic metabolic profiles, microbial community dynamics, and osteoblastogenesis. We show that adding GOS and preconditioning L. reuteri DSM 17938 act on different scales, yet both increase ex vivo short-chain fatty acid (SCFA) production and engraftment within the microbial community. Furthermore, preconditioned supernatants or SCFA cocktails mirroring these profiles decrease the migration speed of MC3T3-E1 osteoblasts, increase several osteogenic differentiation markers, and stimulate bone mineralization. Thus, our results demonstrate that preconditioning of L. reuteri with GOS may represent an incremental advantage for synbiotics by optimizing metabolite production, microbial engraftment, microbiome profile, and increased osteoblastogenesis.

Keywords: Galacto-oligosaccharides; Limosilactobacillus reuteri DSM 17938; Osteoblastogenesis; Probiotic engraftment.

Figure 1

Overview of the experiments and analyses performed. (1) Limosilactobacillus reuteri DSM 17938 is produced through the normal production process, in which glucose (Glc) and fructose (Frc) are available as carbon sources. The probiotic strain is also produced through the preconditioning process, in which galacto-oligosaccharides (GOS), glucose (Glc), and fructose (Frc) are available as carbon sources. (2) The six experimental conditions we assessed consist of selective additions of the components galacto-oligosaccharide (GOS), normal (non-preconditioned) L. reuteri DSM 17938, and preconditioned L. reuteri DSM 17938. The different combinations of these components allow to assess each component individually. Infant formula was added in all cases. (3) Selected component mixes and cryopreserved toddler fecal inoculum were added to controlled ex vivo colonic incubators and sampled at 0 and 48 h of incubation time. Every condition was performed in biological triplicate. (4) Samples were analyzed through a multiphasic pipeline consisting of a microbiological and osteal part.

Figure 2

Microbial metabolite profiles after 48 h of ex vivo colonic incubation are affected by GOS and preconditioning. Dots denote individual replicates (n = 3) and bars their average. The effect of galacto-oligosaccharide (GOS) addition is apparent through strong increases in lactate and acetate upon addition. The effect of Limosilactobacillus reuteri DSM 17938 preconditioning is apparent to a lesser extent. Lactate and acetate (key metabolites of L. reuteri DSM 17938) show upward trends when moving from non-preconditioned to preconditioned L. reuteri DSM 17938 conditions. The same trend is observed for acidification. Notably, preconditioning L. reuteri DSM 17938 significantly increases lactate concentration and acidification, compared to the non-preconditioned strain. Pairwise p-values are denoted with a bracket. Differences due to GOS addition are significant for all conditions and thus omitted from the plot for clarity.

Figure 3

Microbial diversity in the SHIME ex vivo model displayed via non-metric multidimensional scaling shows how microbial dynamics are affected by incubation time, GOS, and preconditioning. Dots represent individual samples and are colored by different metadata attributes on each subplot A-C. Cluster separations were confirmed by permutational analysis of variance (PERMANOVA) at α = 0.05. Distance between dots represents the similarity of the underlying microbial community composition. Samples having a more similar microbial community composition are closer to each other and vice versa. (A) Diversity is shaped foremost by incubation time, as the distinct clusters are translated in the horizontal direction after 48 h. (B) For long incubation times, galacto-oligosaccharide (GOS) addition becomes a major factor determining microbial composition and divides communities that are associated with GOS addition and those who are not. Indeed, after 48 h incubation, clusters split solely based on GOS addition. A similar pattern is found for the microbial metabolite profiles (Fig. 2A). (C) Community composition is determined by Limosilactobacillus reuteri DSM 17938 addition but not by preconditioning. There are distinct clusters based on whether L. reuteri was added or not. This shows that L. reuteri DSM 17938 preconditioning does not affect the underlying microbial community composition differently from non-preconditioned L. reuteri DSM 17938.

Figure 4

Limosilactobacillus reuteri levels after 48 h of ex vivo colonic incubation replicates (n = 3) are highest in presence of GOS. The taxon detected corresponds with high confidence to Limosilactobacillus reuteri DSM 17938 (see Sect. 2.2 “Galacto-oligosaccharide addition affects microbial overall community composition, while preconditioning enhances engraftment of L. reuteri DSM 17938” for a detailed explanation). The taxon Limosilactobacillus reuteri is not detected when L. reuteri DSM 17938 is not added (Infant formula only and Infant formula + GOS conditions), which validates the pipeline used. Relative abundances as determined by combining 16S rRNA sequencing and flow cytometry. Limosilactobacillus reuteri is detectable after 48 h only when adding galacto-oligosaccharides (GOS), showing that GOS addition enhances survival and successful engraftment of the strain. Preconditioning the strain results in higher abundances after 48 h, showing that preconditioning boosts engraftment when combined with GOS. Pairwise p-values are denoted with a bracket. All differences between normal L. reuteri DSM 17938 with GOS or preconditioned L. reuteri DSM 17938 with GOS and any other condition is significant. Notably, the preconditioned L. reuteri + GOS condition is significantly different from all other conditions, including the corresponding condition with non-preconditioned L. reuteri DSM 17938.

Figure 5

Galacto-oligosaccharide (GOS) addition and Limosilactobacillus reuteri DSM 17938 preconditioning enhance osteoblastogenesis. MC3T3-E1 cells were exposed to end products obtained after ex vivo colonic incubation. Error bars represent standard errors of replicate means. Pairwise p-values are denoted with a bracket. (A) Osteoblast motility assessed by the average speed of scratch closure assays (n = 16). (B) Osteoblast differentiation assessed by alkaline phosphatase activity normalized by infant formula condition (n = 18). (C) Osteoblast mineralization assessed Alizarin Red S optical density absorption (n = 9). (D) Osteoblast differentiation assessed by marker gene expression (Ctnnb1, Atf4, and Ocn) when exposed to short-chain fatty acids cocktails (n = 12). Cocktails mirror the ratios of acetate, propionate and butyrate obtained with the ex vivo colonic incubation.