Identification of Human Secretome and Membrane Proteome-Based Cancer Biomarkers Utilizing Bioinformatics

• Published in: The Journal of membrane biology Vol. 253; no. 3; pp. 257 - 270
• Main Authors: Rehman, Adeel ur, Olof Olsson, P., Khan, Naveed, Khan, Khalid
• Format: Journal Article
• Language: English
• Published: New York Springer US 01-06-2020; Springer Nature B.V
• Subjects: Algorithms; Biochemistry; Bioinformatics; Biomarkers; Biomedical and Life Sciences; Cancer; Cancer immunotherapy; Cell membranes; Data mining; Diagnostic systems; Functional analysis; Gene expression; Genes; Human Physiology; Immunotherapy; Isoforms; Life Sciences; Membrane proteins; Proteins; Proteomes; Secretome; Cancer biomarkers; Membrane protein; RNA-Sequencing; Secretory protein; Hematopoietic lineage
• ISSN: 0022-2631; 1432-1424
• DOI: 10.1007/s00232-020-00122-5

Cellular secreted proteins (secretome), together with cellular membrane proteins, collectively referred to as secretory and membrane proteins (SMPs) are a large potential source of biomarkers as they can be used to indicate cell types and conditions. SMPs have been shown to be ideal candidates for several clinically approved drug regimens including for cancer. This study aimed at performing a functional analysis of SMPs within different cancer subtypes to provide great clinical targets for potential prognostic, diagnostic and the therapeutics use. Using an innovative majority decision-based algorithm and transcriptomic data spanning 5 cancer types and over 3000 samples, we quantified the relative difference in SMPs gene expression compared to normal adjacent tissue. A detailed deep data mining analysis revealed a consistent group of downregulated SMP isoforms, enriched in hematopoietic cell lineages (HCL), in multiple cancer types. HCL-associated genes were frequently downregulated in successive cancer stages and high expression was associated with good patient prognosis. In addition, we suggest a potential mechanism by which cancer cells suppress HCL signaling by reducing the expression of immune-related genes. Our data identified potential biomarkers for the cancer immunotherapy. We conclude that our approach may be applicable for the delineation of other types of cancer and illuminate specific targets for therapeutics and diagnostics.