Intestinal mucosa-derived DNA methylation signatures in the penetrating intestinal mucosal lesions of Crohn's disease

Abstract

The purpose of this study was to evaluate genome-wide DNA methylation changes in intestinal mucosa tissue of adult patients with Crohn's disease comprehensively. DNA methylation chip was used to analyze abnormal methylation sites among penetrating and non-penetrating intestinal mucosa tissue of Crohn's disease and normal intestinal mucosa tissue of healthy controls. Methylation abnormalities of different locus were verified by pyrosequencing and quantitative polymerase chain reaction. Differential DNA methylation sites were participated in the positive regulation of apoptosis and the positive regulation of IL-8 production and were enriched in signaling pathways related to inflammatory bowel disease and extracellular matrix receptor interaction signaling pathways. Correlation analysis showed that the methylation abnormalities of HLA-DRB1 (r = - 0.62, P < 0.001), MUC1 (r = - 0.45, P = 0.01), YPEL5 (r = - 0.55, P = 0.001) and CBLB (r = - 0.62, P < 0.001) were significantly negatively correlated with their relative expression levels. The degree of methylation abnormality of MUC1 was negatively correlated with the disease activity score of Crohn's disease (r = - 0.50, P = 0.01). Apoptosis, interleukin-8 production and abnormal extracellular matrix might be involved in the mechanism of penetrating intestinal mucosal lesions in Crohn's disease. The degree of abnormal methylation of MUC1 was negatively correlated with the disease activity of Crohn's disease.

Figure 1

Volcano Plots. (A) Comparisons of differential DNA methylation sites between CD penetrating intestinal mucosal tissue with normal intestinal mucosal tissue; (B) Comparisons of differential DNA methylation sites between CD penetrating and non-penetrating intestinal mucosal tissue. Figures were performed with affy R package (Version 3.6.1, https://www.r-project.org).

Figure 2

Cluster analysis of CD penetrating intestinal mucosal tissue and normal intestinal mucosal tissue. (A) The overall; (B) Top 20 highest degree of hypermethylation and demethylation sites; Cluster analysis of CD penetrating and non-penetrating intestinal mucosal tissue. (C) The overall; (D) Top 20 highest degree of hypermethylation and demethylation sites. The serial numbers at the bottom of (A) and (C) presented: (A) CD penetrating intestinal mucosal tissue; (B) CD non-penetrating intestinal mucosal tissue; X: normal intestinal mucosal tissue. Figures were performed with affy R package (Version 3.6.1, https://www.r-project.org).

Figure 3

Go analysis. (A) Comparisons of differential DNA methylation sites between CD penetrating intestinal mucosal tissue with normal intestinal mucosal tissue; (B) Comparisons of differential DNA methylation sites between CD penetrating and non-penetrating intestinal mucosal tissue. GO data were downloaded from GO website (http://geneontology.org).

Figure 4

KEGG analysis. (A) Comparisons of differential DNA methylation sites between CD penetrating intestinal mucosal tissue with normal intestinal mucosal tissue; (B) Comparisons of differential DNA methylation sites between CD penetrating and non-penetrating intestinal mucosal tissue. KEGG pathways were downloaded from KEGG website (https://www.kegg.jp). The permission was provided by Kanehisa laboratory.

Figure 5

(A) The methylation abnormality degrees of HLA-DRB1, YPEL5, CBLB and MUC1 in the intestinal mucosal tissue of the CD group were verified by pyrosequencing. (B) The relative expression levels of HLA-DRB1, YPEL5, CBLB and MUC1in the intestinal mucosal tissue of CD group were tested by qPCR. (C) Correlation between methylation abnormality degrees of HLA-DRB1, YPEL5, CBLB and MUC1 and gene expression level. (D) Correlation between methylation degree of MUC1 and disease activity of CD. Figures were performed with Prism (Version 6.0c, https://www.graphpad.com/scientific-software/prism/).

Figure 6

The fold changes of DEGs among GSE95095, GSE83448, GSE103027 and GSE102133. The GEO databases were obtained from NCBI-GEO databases (https://www.ncbi.nlm.nih.gov/geo). Figures were performed with affy R package (Version 3.6.1, https://www.r-project.org).