Lactobacillus paracasei CNCM I-4034 and its culture supernatant modulate Salmonella-induced inflammation in a novel transwell co-culture of human intestinal-like dendritic and Caco-2 cells

Background: The action of probiotics has been studied in vitro in cells isolated from both mice and humans, particularly enterocytes (IECs), dendritic cells (DCs) and co-cultures of peripheral DCs and IECs. Peripheral DCs and murine DCs differ from human gut DCs, and to date there are no data on the action of any probiotic on co-cultured human IECs and human intestinal DCs. To address this issue, a novel transwell model was used. Human IECs (Caco-2 cells) grown in the upper chamber of transwell filters were co-cultured with intestinal-like human DCs grown in the basolateral compartment of the transwells. The system was apically exposed for 4 h to live probiotic L. paracasei CNCM I-4034 obtained from the faeces of breastfed infants or to its cell-free culture supernatant (CFS) and challenged with Salmonella typhi. The secretion of pro-and anti-inflammatory cytokines in the basolateral compartment was determined by immunoassay, and the DC expression pattern of 20 TLR signaling pathway genes was analysed by PCR array. Results: The presence of the live probiotic alone significantly increased IL-1 beta, IL-6, IL-8, TGF-beta 2, RANTES and IP-10 levels and decreased IL-12p40, IL-10, TGF-beta 1 and MIP-1 alpha levels. This release was correlated with a significant increase in the expression of almost all TLR signaling genes. By contrast, incubation of the co-culture with CFS increased IL-1 beta, IL-6, TGF-beta 2 and IP-10 production only when Salmonella was present. This induction was correlated with an overall decrease in the expression of all TLR genes except TLR9, which was strongly up-regulated. Conclusions: The data presented here clearly indicate that L. paracasei CNCM I-4034 significantly increases the release of pro-inflammatory cytokines, enhances TLR signaling pathway activation and stimulates rather than suppresses the innate immune system. Furthermore, our findings provide evidence that the effects of probiotics in the presence of IECs and DCs differ from the effects of probiotics on cultures of each cell type alone, as reported by us earlier. Thus, co-culture systems such as the one described here are needed to characterise the effects of probiotics in vitro, highlighting the potential utility of such co-cultures as a model system.