Identification of Pharmacological Autophagy Regulators of Active Ulcerative Colitis

Abstract

Background: Ulcerative colitis (UC) is a chronic recurrent disease of unknown etiology. Recently, it has been reported that autophagy-related gene polymorphism is closely associated with increased risk of UC, and the therapeutic effect of some UC drugs is mediated by regulating autophagy pathways. This study aims to identify pivotal autophagy-related regulators in UC pathogenesis and provide novel molecular targets for the treatment of active UC. Methods: Gene expression profiles and clinical information of active UC patients were obtained from GEO databases. CIBERSORT was adopted to evaluate the immune cell infiltration. We used weighted gene co-expression network analysis (WGCNA) and differential expression analysis to identify the pivotal modules and genes associated with active UC. Subsequently, we conducted validation in the validation set and explored its relationship with commonly used UC therapeutics. Results: 36 healthy controls and 46 active UC patients have been obtained from the training set of GSE53306, GSE87466, and GSE134025. There were 423 differentially expressed genes (DEGs) found, which dramatically enriched in autophagy-related pathways. And more infiltration of mast cells, activated T cells, dendritic cells, and M1 macrophages were observed in the intestinal mucosa of active UC, while more infiltration of resting immune cells and M2 macrophages in healthy controls. WGCNA indicated that the turquoise and blue modules were the critical modules. CASP1, SERPINA1, and CCL2 have been identified as the hub autophagy-related genes of active UC, after combining DEGs and 232 autophagy-related genes from HADb with the genes of turquoise and blue modules, respectively. We further verified that CASP1, SERPINA1, and CCL2 were positively associated with active UC and served as an autophagy-related biomarker for active UC. Moreover, increased SERPINA1 in the involved intestinal mucosa was reduced in patients with active UC who responded to golimumab or glucocorticoid therapy. But, neither CASP1, SERPINA1, and CCL2 were changed by treatment of 5-aminosalicylic acid (5-ASA) and azathioprine. Conclusion: CASP1, SERPINA1, and CCL2 are autophagy-related hub genes of active UC. And SERPINA1 may serve as a new pharmacological autophagy regulator of UC, which provides a new target for the use of small molecules targeting autophagy in the treatment of active UC.

Keywords: WGCNA; autophagy; biomarkers; pharmacology; ulcerative colitis.

FIGURE 1

Research flow chart of this study. PCA, principal component analysis. UC, ulcerative colitis. WGCNA, weighted gene co-expression network analysis. HADb, the Human Autophagy Database. DEGs, differentially expressed genes. GO, the Gene Ontology. KEGG, the Kyoto Encyclopedia of Genes and Genomes analyses. GC, glucocorticoid. 5-ASA, 5-aminosalicylic acid.

FIGURE 2

Principal component analysis (PCA) and immune-infiltrating landscape of active UC. The PCA of before (A) and after (B) batch correction of all samples. (C) The distribution of 22 immune cells in all active UC patients and healthy control samples in the total set. (D) The heat map delineated the abundance of the immune cell populations in each sample of active UC and healthy controls. (E) Correlation matrix of different infiltrating immune cells. The darker the blue, the stronger the negative correlation, and the darker the red, the stronger the positive correlation. (F) Comparison of immune cell infiltration in the intestinal mucosa between active UC patients and healthy control. Blue represents the healthy control group, and red represents the active UC group.

FIGURE 3

Identification of DEGs and modules associated with active UC. (A) Volcano map of DEGs in the intestinal mucosa between active UC patients and healthy control. Red represents up-regulated genes, green represents down-regulated genes, and black represents no significant difference genes. (B) The expression profiles of top 50 DEGs between active UC patients and healthy controls. The red to green colors represent the change from high to low expression. (C) GO enrichment analysis of DEGs. (D) KEGG pathway enrichment analysis of DEGs. (E) Module clustering dendrogram based on a dissimilarity measure (1-TOM). (F) Heatmap of the correlation between module eigengenes and active UC. DEG, differentially expressed genes. UC, ulcerative colitis. GO, the Gene Ontology. KEGG, the Kyoto Encyclopedia of Genes and Genomes analyses. TOM: topological overlap matrix.

FIGURE 4

Identification and verification of autophagy-related hub genes. (A) Venn diagram shows the overlapping genes between differentially expressed genes, autophagy-related genes and the genes of blue module. (B) Venn diagram shows the overlapping genes between differentially expressed genes, autophagy-related genes and the genes of turquoise module. (C) The expression of CASP1, SERPINA1, CCL2, IL24, NAMPT, and CASP4 in the lesional and non-lesional intestinal mucosa of active UC patients in GSE107499. (D) The expression of CASP1, SERPINA1, CCL2, IL24, NAMPT, and CASP4 in the intestinal mucosa of healthy controls, inactive and active UC patients in GSE59071. UC, ulcerative colitis. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, ns, not statistically significant.

FIGURE 5

Intestinal mucosal invasion of active UC is associated with autophagy defects. (A) The Volcano map and expression profiles of top 20 DEGs between high and low CASP1 expression groups in active UC patients. (B) The Volcano map and expression profiles of top 4 DEGs between high and low SERPINA1 expression groups in active UC patients. (C) The Volcano map and expression profiles of top 20 DEGs between high and low CCL2 expression groups in active UC patients. (D) The relative expression levels of CASP1, SERPINA1, and CCL2 in the involved intestinal mucosa of active UC patients treated 5-ASA or azathioprine. (E) The relative expression levels of CASP1, SERPINA1 and CCL2 in involved and non-involved intestinal mucosa of active UC patients treated with 5-ASA. (F) The relative expression levels of CASP1, SERPINA1 and CCL2 in involved and non-involved intestinal mucosa of active UC patients treated with azathioprine. UC, ulcerative colitis. DEGs, differentially expressed genes. 5-ASA, 5-aminosalicylic acid. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

FIGURE 6

GC or golimumab responders improve impaired autophagy of intestinal mucosal in patients with active UC by reducing SERPINA1. (A) The relative expression levels of CASP1, SERPINA1, and CCL2 in the intestinal mucosa of active UC patients in responding and non-responding groups before golimumab treatment. (B) The relative expression levels of CASP1, SERPINA1, and CCL2 in the intestinal mucosa of active UC patients in responding and non-responding groups after 6 weeks of golimumab treatment. (C) The relative expression levels of CASP1, SERPINA1, and CCL2 in the intestinal mucosa of active UC patients in responding and non-responding groups before and after 6 weeks of golimumab treatment. (D) The relative expression levels of CASP1, SERPINA1, and CCL2 in the intestinal mucosa of active UC patients in responding and non-responding groups before and after GC treatment for 3 days. UC, ulcerative colitis. GC, glucocorticoid. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.