- 9:06 AM

Genomic-Scale Transcriptional Analysis of T-Cell Activation Reveals Novel Genes and Signaling Programs in the Context of Ex Vivo Expansion

Min Wang, Department of Chemical and Biological Engineering, Northwestern University, 15 Innovation Way room288, Newark, DE 19711 and Eleftherios Terry Papoutsakis, Dept. of Chemical Engineering & Delaware Biotechnology Institute, University of Delaware, 15 Innovation Way, Newark, DE 19711.

Introduction. Ex vivo expanded T cells are increasingly employed in a variety of immunotherapy protocols. The success of such protocols requires large numbers of biologically active T cells. Several activation and culture protocols have been examined, but none in significant or molecular detail to allow more targeted protocol interventions. In this study, our aim was to better understand T-cell activation using a comprehensive, microarray-based comparative transcriptional analysis, in conjunction with targeted protein-level and functional analyses.

Methods. Six independent biological experiments were transcriptionally analyzed. In three biological experiments, CD3+ T cells were activated with anti-CD3/anti-CD28 Mab, cultured for 96 hours, and samples collected at 0, 4, 10, 48 and 96 hours. To cover earlier timepoints, CD4+ and CD8+ T cells of another three biological experiments were activated with anti-CD3/anti-CD28 Mab, cultured for 72 hours, and samples collected at 0, 6, 12, 48 and 72 hours. Multi-class SAM (Significance Analysis of Microarrays) with a false discovery rate <1% was used to select genes that show statistically different expression between groups. A group is defined as all the samples belonging to the same timepoint regardless of the donors. Briefly, there were 5 groups (0 hour, 4, 10, 48 and 96 hours) in the set of CD3+ experiments, and 6 groups (0 hour, 6, 12, 24, 48 and 72 hours) in the set of CD4+ experiments and CD8+ experiments. The three samples from biological experiments in each group were treated as replicates. To focus on the expression change, gene expression at each time point was compared to that of 0 hour in each experiment. Gene Ontology annotations, as curated by European Bioinformatics Institute, were retrieved from the Gene Ontology Consortium website. Interferon-γ (IFNG) and CCL20 ELISA assays, CD69, TNFSF4, CD40LG (TNFSF5), TNFRSF9, KLRD1, CD48, GZMB, CASP3, phospho-NFκB-p65, phospho-p38 (MAPK14), phospho-ERK1 (MAPK3), PUMA (BBC3), BCL2A1 protein abundance assay by flow cytometry, and AP-1 activity assay were carried out to confirm select findings of the transcriptional analysis.

Results. Comparison of the microarray-based gene expression patterns between CD3+ T cells, and the CD4+ and CD8+ subsets revealed largely conserved, but not identical, transcriptional patterns. We employed a Gene-Ontology-driven transcriptional analysis coupled with protein abundance assays in order to identify novel T-cell activation genes and cell-type-specific genes associated with the immune response. We identified potential genes involved in the communication between the two subsets (including IL23A, NR4A2, CD83, PSMB2, -8, MIF, IFI16, TNFAIP1, POU2AF1, and OTUB1) and would-be effector-function-specific genes (XCL2, SLAMF7, TNFSF4, -5, -9, CSF3, CD48 and CD244). Chemokines induced during T-cell activation, but not previously identified in T cells, include CCL20, CXCL9, -10, -11 (in all three populations), and XCL2 (preferentially in CD8+ T cells). Increased expression of other unexpected cytokines (GPI, OSM and MIF) suggests their involvement in T-cell activation with their functions yet to be examined. Differential expression of many receptors, not previously reported in the context of T-cell activation, including CCR5, CCR7, IL1R2, IL1RAP, IL6R, TNFRSF25 and TNFRSF1A, suggests their role in this immune process. Several receptors involved in TCR activation (CD3D, CD3G, TRAT1, ITGAL, ITGB1, ITGB2, CD8A and B (CD8+ T-cell specific) along with LCK, ZAP70 and TYROBP were synchronously downregulated. Members of cell-surface receptors (HLA-Ds and KLRs), none previously identified in the context of T-cell activation, were also downregulated. We also identified significantly regulated apoptotic genes in several protein families and detailed their transcriptional kinetics during the T-cell activation process. Transcription patterns of some selected genes (BCL2A1, BBC3 and CASP3) were validated at the protein level.

Conclusion. Our data and analysis provide a better understanding of T-cell activation and identify several potential targets that can be explored to benefit T-cell expansion protocols for immunotherapy applications. Although there is much known about T-cell activation, modern genomic tools provide an extraordinary opportunity to verify, extend and enrich prior knowledge, and, discover new players and processes not previously associated with T-cell activation.