Identification of HnRNP Family as Prognostic Biomarkers in Five Major Types of Gastrointestinal Cancer.

Abstract

Heterogeneous nuclear ribonucleoproteins (hnRNPs), a large family of RNAbinding proteins, have been implicated in tumor progression in multiple cancer types. However, the expression pattern and prognostic value of hnRNPs in five gastrointestinal (GI) cancers, including gastric, colorectal, esophageal, liver, and pancreatic cancer, remain to be investigated. The current research aimed to identify prognostic biomarkers of the hnRNP family in five major types of gastrointestinal cancer. Oncomine, Gene Expression Profiling Interactive Analysis (GEPIA), and Kaplan-Meier Plotter were used to explore the hnRNPs expression levels concerning clinicopathological parameters and prognostic values. The protein level of hnRNPU was validated by immunohistochemistry (IHC) in human tissue specimens. Genetic alterations of hnRNPs were analyzed using cBioportal, and Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed to illustrate the biological functions of co-expressed genes of hnRNPs. The vast majority of hnRNPs were highly expressed in five types of GI cancer tissues compared to their adjacent normal tissues, and mRNA levels of hnRNPA2B1, D, Q, R, and U were significantly different in various GI cancer types at different stages. In addition, Kaplan-Meier analysis revealed that the increased hnRNPs expression levels were correlated with better prognosis in gastric and rectal cancer patients (log-rank p < 0.05). In contrast, patients with high levels of hnRNPs exhibited a worse prognosis in esophageal and liver cancer (log-rank p < 0.05). Using immunohistochemistry, we further confirmed that hnRNPU was overexpressed in gastric, rectal, and liver cancers. In addition, hnRNPs genes were altered in patients with GI cancers, and RNA-related processing was correlated with hnRNPs alterations. We identified differentially expressed genes of hnRNPs in tumor tissues versus adjacent normal tissues, which might contribute to predicting tumor types, early diagnosis, and targeted therapies in five major types of GI cancer.