RESUMEN
BACKGROUND: Colitis-associated cancer (CAC) is one of the prognostic factors in inflammatory bowel disease (IBD), and prevention of CAC is a critical concern for patients with IBD. Component cells of the microenvironment, especially myofibroblasts, are known to affect tumor development, but the role of intestinal myofibroblasts (IMFs) in CAC has not been clarified. Here, we explored the role of IMFs in CAC and sought to identify candidate genes as novel therapeutic targets for the prevention of CAC. METHODS: We used the azoxymethane (AOM)/dextran sodium sulfate (DSS) model for dysplasia and CAC. Flow cytometry and RNA sequencing (RNA-seq) were performed to obtain an unbiased gene expression profile of IMFs. The transcriptome of significantly differentially expressed genes was analyzed by RNA-seq, quantitative reverse transcriptase polymerase chain reaction, and immunohistochemistry. RESULTS: Comparison of normal intestinal fibroblasts and IMFs revealed 1045 genes with significantly differential expression. Among them, we focused on tenascin-C (TNC; q = 0.00232, Log2(Fold Change) = 3.87). Tenascin-C gene expression was markedly increased in the dysplasia model compared with control and CAC model (P < 0.05). Tenascin-C protein was barely expressed in normal and nondysplastic mucosa but strongly expressed in the stroma around dysplastic lesions. Moreover, TNC surrounded and enclosed integrin αvß3-positive microvessels. Administration of ATN-161, an antagonist of αvß3-integrin, significantly suppressed tumorigenesis of CAC through inhibition of angiogenesis (P < 0.05). CONCLUSIONS: In the early stages of CAC, TNC produced by IMFs affects tumor development via integrin αvß3-mediated angiogenesis. Intestinal myofibroblasts might be a novel therapeutic target for preventing CAC.
