Given the relatively high level of coexistence, a nonlinear model is more suitable for mutual exclusion than a linear model. Based on mutual exclusion, negative markers were identified for CD4+ cytotoxic T cells, particularly CD126. However, according to FACS analysis, CD28 is the most effective cell surface negative marker.
ABSTRACT
CD4 cytotoxic T lymphocytes (CTLs) are a distinct subset of CD4+ T cells that can mediate antigen-specific cytotoxicity. However, their specific surface markers have yet to be defined. In this study, we used gene-plasticity-based mutual exclusion analysis to identify negative biomarkers for CD4 CTLs, using granzyme B (GZMB) expression as the pivotal reference. We analyzed bulk RNA-seq data from 768 human CD4+ T cell samples, calculated gene plasticity scores, and applied cosine similarity to detect genes that are anti-correlated with GZMB. Several genes encoding cell surface membrane proteins were identified as important candidates. ITGA6 (CD49f) and IL6R (CD126) were validated by flow cytometry in healthy individuals and primary Sjögren’s syndrome (pSS) patients. The expression profile confirmed mutually exclusive between GZMB and both CD49f and CD126, with CD49f−CD126−GPR56+ CD4+ T cells exhibiting the highest GZMB levels. In pSS patients, CD49f−CD126−GPR56+ cells in CD8+ T cells were significantly negatively correlated with clinical parameters (IgA, anti-SSB). ROC analysis revealed the percentages of GZMB in CD4+ T cells (AUC = 0.766) and GPR56/GZMB in CD8+ T cells (AUC = 0.76) as diagnostic biomarkers.