Cell type-specific transcriptomics of esophageal adenocarcinoma as a scalable alternative for single cell...

Single-cell transcriptomics have revolutionized our understanding of the cell composition of tumors and allowed us to identify new subtypes of cells. Despite rapid technological advancements, single-cell analysis remains resource-intense hampering the scalability that is required to profile a sufficient number of samples for clinical associations. Therefore, more scalable approaches are needed to understand the contribution of individual cell types to the development and treatment response of solid tumors such as esophageal adenocarcinoma where comprehensive genomic studies have only led to a small number of targeted therapies. Due to the limited treatment options and late diagnosis, esophageal adenocarcinoma has a poor prognosis. Understanding the interaction between and dysfunction of individual cell populations provides an opportunity for the development of new interventions. In an attempt to address the technological and clinical needs, we developed a protocol for the separation of esophageal carcinoma tissue into leukocytes (CD45+), epithelial cells (EpCAM+), and fibroblasts (two out of PDGFRα, CD90, anti-fibroblast) by fluorescence-activated cell sorting and subsequent RNA sequencing. We confirm successful separation of the three cell populations by mapping their transcriptomic profiles to reference cell lineage expression data. Gene-level analysis further supports the isolation of individual cell populations with high expression of CD3, CD4, CD8, CD19, and CD20 for leukocytes, CDH1 and MUC1 for epithelial cells, and FAP, SMA, COL1A1, and COL3A1 for fibroblasts. As a proof of concept, we profiled tumor samples of nine patients and explored expression differences in the three cell populations between tumor and normal tissue. Interestingly, we found that angiogenesis-related genes were upregulated in fibroblasts isolated from tumors compared with normal tissue. Overall, we suggest our protocol as a complementary and more scalable approach compared with single-cell RNA sequencing to investigate associations between clinical parameters and transcriptomic alterations of specific cell populations in esophageal adenocarcinoma.