The current in vitro or in vivo intestinal fibrosis models have many limitations. Recent advancements in the isolation and culturing of organoids has led to development of various three-dimensional (3D) intestinal disease models with in vivo physiology. In this study, we generated an organoid-based epithelial to mesenchymal transition (OEMT) model, which could be used as a novel intestinal fibrosis model. Intestinal epithelial organoids (IEOs) were isolated and cultured from the small intestines of normal mice. IEOs were treated with transforming growth factor- β1 (TGF-β1) or Tumor necrosis factor-α (TNF-α) to evaluate their phenotypic change. Raw 264.7 cells (macrophage) stimulated with lipopolysaccharide were co-cultured with IEOs in growth media with or without TGF-β1. TGF-β1 alone slightly induced epithelial to mesenchymal transition (EMT) in the IEOs but mainly disrupted them. Macrophage released cytokines synergistically induced mesenchymal phenotypic changes in TGF-β1 stimulated intestinal organoids. TNF-α and TGF-β1 synergistically induced proliferation of mesenchymal cells as well as EMT in the IEOs. We generated a novel OEMT model based on our finding that TNF-α and TGF-β synergistically induce type 2 EMT in IEOs. This 3D EMT model with in vivo physiology could be used to study EMT associated intestinal fibrosis.

BackgroundIntestinal fibrosis is a serious complication of inflammatory bowel disease, including Crohn’s disease and ulcerative colitis. There is no specific treatment for intestinal fibrosis. Studies have indicated that peroxisome proliferator-activated receptor- γ (PPAR-γ) agonists have anti-fibrogenic properties in organs besides the gut; however, their effects on human intestinal fibrosis are poorly understood. This study investigated the anti-fibrogenic properties and mechanisms of PPAR-γ agonists on human primary intestinal myofibroblasts (HIFs).MethodsHIFs were isolated from normal colonic tissue of patients undergoing resection due to colorectal cancer. HIFs were treated with TGF-β1 and co-incubated with or without one of two synthetic PPAR-γ agonists, troglitazone or rosiglitazone. mRNA and protein expression of procollagen1A1, fibronectin, and α-smooth muscle actin were determined by semiquantitative reverse transcription-polymerase chain reaction and Western blot. LY294002 (Akt inhibitor) was used to examine whether Akt phosphorylation was a downstream mechanism of TGF-β1 induced expression of procollagen1A1, fibronectin, and α-smooth muscle actin in HIFs. The irreversible PPAR-γ antagonist GW9662 was used to investigate whether the effect of PPAR-γ agonists was PPAR-γ dependent.ResultsBoth PPAR-γ agonists reduced the TGF-β1-induced expression of α-smooth muscle actin which was integrated into stress fibers in HIFs, as determined by actin microfilaments fluorescent staining and α-smooth muscle actin-specific immunocytochemistry. PPAR-γ agonists also inhibited TGF-β1-induced mRNA and protein expressions of procollagen1A1, fibronectin, and α-smooth muscle actin. TGF-β1 stimulation increased phosphorylation of downstream signaling molecules Smad2, Akt, and ERK. TGF-β1 induced synthesis of procollagen1A1, fibronectin, and α-smooth muscle actin through a phosphatidylinositol 3-kinase/Akt-dependent mechanism. PPAR-γ agonists down regulated fibrogenesis, as shown by inhibition of Akt and Smad2 phosphorylation. This anti-fibrogenic effect was PPAR-γ independent.ConclusionsTroglitazone and rosiglitazone suppress TGF-β1-induced synthesis of procollagen1A1, fibronectin, and α-smooth muscle actin in HIFs and may be useful in treating intestinal fibrosis.Electronic supplementary materialThe online version of this article (doi:10.1186/s12876-017-0627-4) contains supplementary material, which is available to authorized users.

Organoids, a multi-cellular and organ-like structure cultured in vitro, can be used in a variety of fields such as disease modeling, drug discovery, or cell therapy development. When organoids derived from Lgr5 stem cells are cultured ex vivo, recombinant R-spondin-1 protein should be added at a high concentration for the initiation and maintenance of the organoids. Because the addition of large amounts of R-spondin-1 greatly increases the cost of organoids, the organoids grown with R-spondin-1 are not practical for large-scale drug screening and for the development of therapeutic agents. In this study, we tried to find a R-spondin-1 substitute compound that is able initiate small intestinal organoids without the use of the R-spondin-1 protein; thus, using organoid media that each included one compound from among an 8,364 compound library instead of R-spondin-1, we observed whether organoids were established from the crypts of the small intestine. As a result, we found one compound that could promote the initial formation and growth of enteroids in the medium without R-spondin-1 and named it RS-246204. The enteroids grown with RS-246204 had a similar differentiation capacity as well as self-renewal capacity as the enteroids grown with R-spondin-1. Furthermore, the RS-246204-derived enteroids could successfully produce the forskolin induced swelling and the organoid based epithelial to mesenchymal transition model. This compound could be used for developing a cost-efficient culturing method for intestinal organoids as well as for exploring Lgr5 signaling, intestinal stem cell physiology and therapeutics for GI tract diseases.

A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has not been fixed in the paper.