Society: AGA

Background: Mesenteric fat wrapping around the intestinal wall, so called ‘creeping fat’ (CF), is spatially linked with stricture formation in Crohn’s disease (CD). Intestinal muscularis propria (MP) smooth muscle cell (HIMC) hyperplasia is a major contributor to luminal narrowing in stricturing CD. We investigated CF derived factors and their effect on HIMC hyperplasia in vitro and in vivo.

Methods: Secretion of free fatty acids (FFA) by mesenteric fat (MF) or CF organ cultures was determined via lipidomic mass spectrometry. Effects of different length and types of FFA as well as CF and MF conditioned medium on proliferation of primary HIMF was assessed. Next generation sequencing (NGS) and lipidomics was performed on HIMF and relevant pathways were inhibited with small molecules or siRNA knockdown. In the dextran sodium sulfate (DSS) induced colitis model fat deletion was accomplished using the transgenic FAT-ATTAC mouse and CPT-1 blockade was achieved via the small molecule etomoxir.

Results: Histopathology of intestinal resection tissues revealed CD CF being located in the subserosa and its presence was linked with dramatic thickening of the MP. MP thickening was also observed in the acute DSS colitis model compared to controls and the thickening was reduced by fat deletion in the FAT-ATTAC mouse. CF conditioned medium markedly upregulated HIMC proliferation compared to mesenteric fat from CD, UC and normal controls. CF released higher amounts of total, saturated and poly-unsaturated FFA with elevated levels of five long-chain (LC-)FFA, including palmitate. LC, but not medium or short chain FFA selectively increased proliferation in HIMC and fibroblasts, but not other intestinal cell types. NGS revealed gene regulation suggesting LC-FFA transport as a putative mechanism. Lipidomic analysis indicated the majority of palmitate being converted into phospholipids, predominantly phosphatidylcholine. Inhibition of LC-FFA uptake into cells via CD36, metabolism through inhibition of acyl-CoA synthetase, choline-phosphate cytidylyltransferase & choline kinase or blockade of LC-FFA uptake into mitochondria through CPT-1A reduced palmitate and CF conditioned medium induced HIMC proliferation. Prophylactic inhibition of CPT-1 with etomoxir in acute DSS colitis did not reduce histopathologic inflammation or inflammatory cytokines gene expression, but reduced MP thickness and gene expression of the smooth muscle cell genes Desmin and Sm22.

Conclusion: Subserosal CF releases LC-FFA inducing a selective proliferative response by HIMC. This effect was dependent on CD36, acyl-CoA synthetase and CPT-1. LC-FFA taken up by HIMC are converted into phospholipids. Deletion of fat or inhibition of CPT-1 in acute DSS colitis reduced the increased MP thickness in proximity to mesenteric fat. These results point to CF as a novel contributor to stricture formation in CD.

Intestinal fibrosis is a common complication in patients suffering from Crohn’s disease (CD). Dysbalanced extracellular matrix deposition leads to thickening of the intestinal wall resulting in stenotic strictures that require surgical interventions, whereas medical therapy of stricturing CD represents a high unmet need. Previous work provided evidence that Interleukin 36 receptor (IL36R) activation is connected with intestinal inflammation and fibrosis. Correspondingly, the targeting of IL36R signaling as therapy option in IBD including CD patients with stenosis is currently investigated. However, the mechanisms mediated by IL36R signaling in the context of intestinal fibrosis are only partly understood. Here, we studied the contribution of the IL36-inducible matrix metalloproteinase 13 (MMP13, collagenase 3) during intestinal fibrosis in mice and in human tissue specimens.
We performed bulk RNA sequencing of paired intestinal biopsies taken from nonstenotic and stenotic areas of patients with CD (n=8). Corresponding tissue samples from healthy controls and CD patients with stenosis were used for immunofluorescence (IF) stainings. MMP13 gene expression was analyzed in cDNA of intestinal biopsies from healthy controls (n=51), patients with CD (n=137) and patients with ulcerative colitis (UC, n=57). In addition, gene regulation on RNA and protein level was studied in purified intestinal fibroblasts from mice upon IL36R stimulation. Finally, in vivo studies were performed with MMP13 deficient mice and wildtype controls in an experimental model of chronic intestinal inflammation and fibrosis. Ex vivo tissue analysis included stainings by Sirius Red and for intestinal cell populations and Collagen type VI by IF.
We observed that intestinal biopsies from stenotic areas as compared to nonstenotic regions of patients with CD showed increased levels of multiple proinflammatory cytokines including IL36G. Interestingly, MMP13 was detected within the 20 highest induced DEGs. IF analysis confirmed higher levels of MMP13 in stenotic tissue sections of CD patients as well as MMP13 expression by Pdpn+ and aSMA+ fibroblasts. Mechanistic experiments in vitro revealed that MMP13 expression was regulated by IL36R signaling, which was further confirmed by in vivo findings including studies using neutralizing anti-IL36R antibodies. Finally, MMP13 deficient mice as compared to wildtype controls developed less inflammation and fibrosis in an experimental model of chronic colitis characterized by reduced activation of myofibroblasts and Collagen type VI accumulation. These findings are consistent with a model suggesting a molecular axis involving IL36R activation and MMP13 expression during the pathogenesis of intestinal fibrosis.
Targeting IL36R-induced MMP13 expression could evolve as a promising strategy to interfere with the development of intestinal fibrosis.

Claudin-1(CLDN-1) plays a key role in the regulation of colonic epithelial homeostasis. We have recently demonstrated using villin-CLDN-1Tg mice, CLDN-1 renders susceptibility to colitis and causes impaired recovery following DSS-induced colitis. This is achieved through the regulation of Notch activation, MMP-9, and p-ERK signaling. An upregulated claudin-1 expression induces MMP-9 and p-ERK signaling to activate Notch-signaling, which in turn inhibits goblet cell differentiation. However, the details of this mechanism/s are not clearly known. In the current study, we investigated the role of MMP-9 in CLDN-1-mediated regulation of intestinal epithelial homeostasis, colitis, and recovery in vivo by crossing the CLDN-1Tg mouse with MMP-9 KO (MMP-9 -/-). CLDN-1/MMP-9-/- mice abrogated the loss of goblet cells, increased MUC-2 expression, and less proliferative epithelium compared to Claudin-1Tg mice alone. During acute colitis, CLDN-1Tg/MMP-9-/- mice had less severity of colitis as compared to CLDN-1Tg mice alone as judged by parameters including disease activity index (DAI), body weight loss, inflammatory cytokines, and immune infiltration, in the colon tissue. We also observed less DNA damage and apoptosis in CLDN-1Tg/MMP-9-/- compared to CLDN-1 Tg mice under the DSS scheme. In addition, unlike CLDN-1 Tg mice who developed dysplastic crypts during recovery, the CLDN-1Tg/MMP-9-/- demonstrated normal regenerative crypts during recovery from DSS insults. Mechanistically, CLDN-1 overexpression associates with and increases the activity of Src which is known to phosphorylate and activate MT1-MMP (Tyr⁵⁷³) to further activate MMP-9. Activated MMP-9 regulates the Notch cascade to suppress goblet cells and increase proliferation thereby enhancing the dysplasia in CLDN-1Tg mice. Together, our data indicate that MMP-9 plays a crucial role in CLDN-1-mediated regulation of intestinal epithelial homeostasis. Our findings may help in providing a mechanistic understanding of the role of CLDN-1 in causing severe colitis and impaired recovery and thus highlighting its importance as a potential therapeutic target for intervention in IBD and associated cancer.

Background: MSCs are a novel therapeutic approved in the European Union (EU) for the treatment of Crohn’s disease (CD) perianal fistulas, however, their mechanism of action in CD is unclear. Our earlier study in SAMP, a chronic and spontaneous murine model of the small intestine (SI) inflammation, had shown that human bone marrow-derived MSC (hMSC) modulates T lymphocytes in MLR through PGE₂ secretion, reprogrammed macrophages (MΦ) to an anti-inflammatory phenotype, and results in histologic, mucosal, immunological, and radiologic healing in SI by day 28 (M Dave, Inflammatory Bowel Diseases, 25 (1), 2019, S62). In human studies, MSCs do not engraft and yet have long-term effects; therefore, here we studied the mechanism of hMSCs’ long-term therapeutic efficacy in SAMP.
Methods: hMSC therapeutic efficacy and mechanism in SAMP were studied using flow cytometry, RT-qPCR, small animal imaging (bioluminescence and epifluorescence), single-cell RNA sequencing (Sc-RNAseq), bright field, and fluorescence microscopy.
Results: Our small animal imaging data showed that intraperitoneally administered hMSC survived until day 9 and did not home to SI. During the early treatment phase at day 9, hMSC showed mucosal healing (P=0.01), T cell suppression in mesenteric lymph node (mLN), and modulation of stromal vascular fraction (SVF) and mLN MΦ to anti-inflammatory phenotype. At day 28, during the late treatment phase, a therapeutic effect was observed even in the absence of live hMSCs. As hMSC mediate their effect via the modulation of T cells and MΦ in the mesentery and mesenteric inflammation is implicated in CD, we studied mesentery using Sc-RNAseq. Sc-RNAseq data, at day 9 of hMSC treatment showed that MΦ had marked increase in the average expression of genes representing anti-inflammatory phenotype, however, the numbers of MΦ expressing these genes did not change. At day 28, MΦ showed significant (logFC >1.5; FDR <0.05) decrease in the expression of gene involved in inflammation (cdk8), an increase in expression of gene contributing to phagocytosis (Gas7) along with increase in the number of anti-inflammatory MΦ implicating efferocytosis; a process of rapid and efficient clearance of apoptotic cells by phagocytes. In vitro co-culture assay on SAMP MΦ with apoptotic hMSCs showed active phagocytosis and uptake of annexin-labeled hMSCs. These results were replicated in vivo by injecting cellTracker^TM labeled hMSC in the SAMP peritoneum cavity; flow cytometry data confirmed that peritoneal MΦ(P=0.04) and mesenteric MΦ (P=0.03) internalize hMSCs through efferocytosis.
Conclusion: hMSC are short-lived which contributes to their safety profile noted in human clinical studies but exert immunosuppressive and tissue regenerative properties that outlast their presence in the body, our study provides a mechanism for this observation.

Jan Nolta

Appearances