RNA sequencing to highlight the heterogeneity in circulating exosomes from patients with esophageal squamous cell carcinoma.

Abstract

e15546 Background: Dissecting tumor heterogeneity is crucial for understanding tumor prognosis, response to therapy, and metastasis. But current tissue biopsy-based strategies for characterizing molecular heterogeneity are invasive and may be confounded by intra-tumor heterogeneity. Here we explore whether exosomes that contain bioactive molecules from the cell of origin can provide new noninvasive means to delineate the heterogeneity of human cancers. Methods: We used RNA-sequencing (RNA-seq) to perform unbiased profiling of mRNAs and long noncoding RNAs (lncRNAs) in plasma exosomes isolated from patients with esophageal squamous cell carcinoma (ESCC, n = 6), patients with esophagitis (n = 6), and healthy controls (n = 6). Results: The number of expressed genes detected in our data set is 63355, including 29615 lncRNAs. We found that exosomes from ESCC have dramatically distinct transcriptome and lncRNA landscapes from that of esophagitis and healthy controls, with 2278 genes and 584 lncRNAs showing differential expression between ESCC and controls; and 854 genes and 126 lncRNAs displaying differential expression between ESCC and esophagitis. We also observed variable expression of diverse transcriptional patterns related to immune response, signal transduction, cell mobility, and transmembrane protein binding, as well as differentially expressed 953 lncRNAs between Stage I and Stage II ESCC samples. Finally, we discovered that both gene and lncRNA expression profiles are variably across exosomal samples from different ESCC patients. Conclusions: Our data reveals that exosomes from ESCC contain distinct transcriptional and lncRNA profiles that separate ESCC from benign esophagitis and healthy controls. Our analysis also identifies unappreciated molecular heterogeneity in exosomes of ESCC, which may pave the way for using exosomal RNA-seq to decode molecular heterogeneity in cancers.