Genetic predictors of gene expression associated with psychiatric comorbidity in patients with inflammatory bowel disease - A pilot study

Transcriptome analysis identifies genetic risk markers and explores the pathogenesis for inflammatory bowel disease

Inflammatory bowel disease (IBD) is an incurable and disabling bowel disease driven by multiple risk factors that severely limit patients' quality of life. We integrated the RNA-sequencing data of 1238 IBD patients, and investigated the pathogenesis of IBD by combining transcriptional element prediction analysis and immune-related analysis. Here, we first determined that KIAA1109 is inhibited in IBD patients. The expression of KIAA1109 and NOD2, the key receptor of NOD-like receptors, showed a negative correlation. The NOD-like receptor signaling pathway is activated and exerts transcriptional regulation on the chemokines CXCL1 and CXCL2 through the activation of the transcription factors NFκB and AP1. Analysis of immune infiltration revealed that the expression of chemokines CXCL1 and CXCL2 may regulate the inflammatory response induced by immune cells. These findings suggest that the KIAA1109-NOD2-NFκB/AP1-CXCL1/CXCL2 regulatory axis is the molecular mechanism of IBD pathogenesis, which will provide a new perspective for the diagnosis, treatment and management of IBD patients.

Identification and Validation of the Diagnostic Markers for Inflammatory Bowel Disease by Bioinformatics Analysis and Machine Learning

Inflammatory bowel disease (IBD) is a chronic inflammatory disorder of the gastrointestinal tract which is mediated by the inappropriate immune responses. This study was aimed to identify novel diagnostic biomarkers for diagnosis of IBD and explore the relationship between the diagnostic biomarkers and infiltrated immune cells. GSE38713, GSE53306, and GSE75214 downloaded from the Gene Expression Omnibus (GEO) database were split into training and testing sets. Differentially expressed genes (DEGs) were screened using the "limma" package. Gene Ontology (GO) and KEGG pathway enrichment analysis of DEGs were performed by clusterProfiler package. The LASSO regression and support vector machine recursive feature elimination (SVM-RFE) algorithms were conducted to identify novel diagnostic biomarkers. The receiver operating characteristic (ROC) curve was applied to evaluate the diagnostic value of the candidate biomarkers. The relationship of the candidate biomarkers and infiltrating immune cells in IBD were evaluated by CIBERSOTR. Quantitative Real-Time PCR (qRT-PCR) was applied to measure the expression level of the biomarkers in IBD. A total of 289 dysregulated genes were identified as DEGs in IBD. These DEGs were significantly enriched in chemokine signaling pathway and cytokine-cytokine receptor interaction. RHOU was identified as a critical diagnostic gene in IBD, which was confirmed using ROC curve and qRT-PCR assays. Immune cell infiltration analysis showed that RHOU was correlated with macrophages M2, dendritic cells resting, mast cells resting, T cells CD4 memory resting, macrophages M0, and mast cells activated. Our results imply that RHOU may be a potential diagnostic biomarker for IBD.

Polygenic risk and causal inference of psychiatric comorbidity in inflammatory bowel disease among patients with European ancestry

BACKGROUND

Approximately 40% of persons with inflammatory bowel disease (IBD) experience psychiatric comorbidities (PC). Previous studies demonstrated the polygenetic effect on both IBD and PC. In this study, we evaluated the contribution of genetic variants to PC among the IBD population. Additionally, we evaluated whether this effect is mediated by the expression level of the RBPMS gene, which was identified in our previous studies as a potential risk factor of PC in persons with IBD.

MATERIALS AND METHODS

The polygenic risk score (PRS) was estimated among persons with IBD of European ancestry (n = 240) from the Manitoba IBD Cohort Study by using external genome-wide association studies (GWAS). The association and prediction performance were examined between the estimated PRS and PC status among persons with IBD. Finally, regression-based models were applied to explore whether the imputed expression level of the RBPMS gene is a mediator between estimated PRS and PC status in IBD.

RESULTS

The estimated PRS had a significantly positive association with PC status (for the highest effect: P-value threshold = 5 × 10^-3, odds ratio = 2.0, P-value = 1.5 × 10^-5). Around 13% of the causal effect between the PRS and PC status in IBD was mediated by the expression level of the RBPMS gene. The area under the curve of the PRS-based PC prediction model is around 0.7 at the threshold of 5 × 10^-4.

CONCLUSION

PC status in IBD depends on genetic influences among persons with European ancestry. The PRS could potentially be applied to PC risk screening to identify persons with IBD at a high risk of PC. Around 13% of this genetic influence could be explained by the expression level of the RBPMS gene.