Sialyllactose and Galactooligosaccharides Promote Epithelial Barrier Functioning and Distinctly Modulate Microbiota Composition and Short Chain Fatty Acid Production In Vitro

Abstract

Human milk oligosaccharides (HMO) and prebiotic oligosaccharides are proposed to confer several health benefits to the infant. They shape the microbiota, have anti-inflammatory properties, and support epithelial barrier functioning. However, in order to select the best oligosaccharides for inclusion in infant formulas, there is a need to increase our understanding of the specific effects of HMO and prebiotics on the host immune system. Therefore, we investigated the effects of the HMO sialyllactose (SL), and galactooligosaccharides (GOS) on epithelial barrier functioning, microbiota composition, and SCFA production. The effect of GOS and SL on epithelial barrier functioning and microbiota composition was investigated using in vitro models. Epithelial barrier function was investigated by transcriptome analysis of fully polarized Caco-2 cells exposed for 6 h to SL or GOS. In addition, epithelial cell growth, alkaline phosphatase production, and re-epithelization was studied. Further, we investigated the effect of SL and GOS on microbiota composition and SCFA production using in vitro fecal batch cultures. Transcriptome analysis showed that SL and GOS both induced pathways that regulate cell cycle control. This gene-expression profile translated to a phenotype of halted proliferation and included the induction of alkaline phosphatase activity, a marker of epithelial cell differentiation. SL and GOS also promoted re-epithelialization in an in vitro epithelial wound repair assay. SL and GOS did show distinct modulation of microbiota composition, promoting the outgrowth of Bacteroides and bifidobacteria, respectively, which resulted in distinct changes in SCFA production profiles. Our results show that SL and GOS can both modulate epithelial barrier function by inducing differentiation and epithelial wound repair, but differentially promote the growth of specific genera in the microbiota, which is associated with differential changes in SCFA profiles.

Keywords: epithelium; galactooligosaccharides; microbiota; short chain fatty acids; sialyllactose.

Figure 1

SL and GOS modulate cell proliferation and induce differentiation. A fully polarized epithelial layer of Caco-2 cells cultured in DMEM +10% FCS was exposed to 10 mg/ml SL or GOS for 6 h. After 6 h the cells were lysed and RNA was isolated for microarray analysis and subsequent IPA analysis. The top five most significantly, excluding ERK5 (p = 0.07), regulated pathways by (A) SL and (B) GOS (IPA analysis results) were shown. Next, phenotypic changes were assessed by culturing Caco-2 cells for 4 days in the presence of different concentrations of GOS and SL (0.1–20 mg/ml), followed by (C) cell counting by flow cytometry (n = 2) and (D) alkaline phosphatase activity determination in culture supernatants (n = 4). Data was represented as mean ± SEM.

Figure 2

SL and GOS promote re-epithelialization, one aspect of wound healing. Longitudinal scratches were applied to a confluent layer of labeled Ca9-22 epithelial cells. The medium with cell debris was replaced by fresh medium containing mixtures of 3′SL and 6′SL as present in cow's milk (0.1 mg/ml), human milk (0.5 mg/ml) and human colostrum (3 mg/ml) or matching concentrations of GOS or inhibitors or TGFα that were used as negative and positive controls, respectively. (A) The total number of cells at the end of the measurement (t = 280min) and (B) the increase in cell numbers over time (i.e., repair rate) was calculated using a non-linear model using the Gompertz equation. Data of three independent experiments (n = 7–11) was represented as mean ± SEM. (C) An image of one representative well was shown (0.5 mg/ml SL or GOS). Significant differences compared to DMEM control were indicated by ^*P < 0.05; ^**P < 0.01 and ^***P < 0.001.

Figure 3

SL and GOS differentially modulate microbiota composition. Batch cultures of adult and infant pooled fecal samples cultured in growth medium were supplemented with or without SL or GOS in duplo. Fecal samples were collected at the start of the batch culture and after 3, 6, 9, and 24 h. Microbiota composition on genus level (A–F) and on species level (G,H) was determined by qPCR and chip analysis, respectively. Bacterial numbers were shown as mean ± SEM of two independent batch cultures. Raw fluorescence data are shown for both individual runs for chip analysis.

Figure 4

SL and GOS differentially stimulate SCFA production. Batch cultures of (A–F) adults and (G–L) infants pooled fecal samples were inoculated with SL or GOS. Fecal samples were collected at the start (0 h) and after a culture time of 3, 6, 9, and 24 h. SCFA levels were measured in the fecal samples by HPLC. SCFA levels of two independent batch cultures were represented as mean ± SEM.