Search

Background: Extensive evidence has shown that immune cell infiltration is associated with the pathogenesis of Crohn’s disease (CD). Methods: Differentially expressed genes (DEGs) from the GSE179285 dataset in the intestinal mucosa of CD patients and healthy individuals were then identified. The infiltration pattern of 22 immune cell types was assessed using the CIBERSORT algorithm. The DEGs and 22 immune cell types were combined to find the key gene network using weighted gene co-expression network analysis (WGCNA). A linear regression model for the relationship between the expression of the hub genes in CD patients and infiltration of immune cells was also developed. The utility and accuracy of the hub genes for CD diagnosis were assessed using receiver operating characteristic (ROC) analysis. The accuracy of the model was validated using the GSE20881 dataset. Results: There were 1135 DEGs between the intestinal mucosal tissue of CD patients and healthy individuals. Of these DEGs, 711 genes were upregulated, whereas 424 of them were downregulated. There was also a significant difference in the infiltration of immune cells to the intestinal mucosal between the CD patients and healthy individuals. WGCNA revealed that the turquoise module genes were strongly correlated with the infiltration of M1 macrophages (cor =0.68, p = 10 −16 ). Finally, the expression of GBP4, the identified hub gene, strongly correlated with the infiltration of M1 macrophages (adjusted r-squared =0.661, p < 2×10 −16 ), and is a relatively good marker for CD diagnostic prediction (AUC =0.736). The relationship between GBP4 expression and infiltration of M1 macrophages (adjusted r-squared =0.435, p < 2×10 −16 ) and diagnostic value of the gene (AUC =0.702) were verified using the GSE20881 validation dataset. Conclusion: The expression of GBP4 is associated with the infiltration of M1 macrophages to the intestinal mucosa of CD patients.