TNFR2 Is Expressed Preferentially By Late Differentiated CD8 T-Cells and Can be Triggered By TNFR2-Specific Ligand to Induce Cell Death of Recently Activated Antigen-Specific T Cells: A Possible Role of TNFR2 in T-Cell Deflation

Recent evidence suggests that Tumor Necrosis Factor (TNF) can selectively kill antigen-specific autoreactive CD8+ T-cells through engagement with TNF Receptor 2 (TNFR2) (1). Within the immune system, TNFR2 expression is restricted to subsets of T-cells, a profile which is in marked contrast to the ubiquitous pattern of expression of TNFR1. However, the spectrum and physiological significance of TNFR2 expression by CD8+ T-cell subpopulations is unknown. In this study we analysed the expression of TNFR2 by CD8 T-cell subsets isolated from normal healthy donors by flow cytometry. In addition, in order to understand the physiological significance of TNFR2 expression on recently activated T cells, we further studied expression on CMV-specific CD8 T-cells which expanded in stem cell transplant patients in response to episodes of CMV reactivation. The expression of TNFR2 was compared to that of other common gamma chain receptors including IL2R and IL7R, and to the expression of a receptor for inflammatory cytokine IL6. TNFR2 expression was found to increase during differentiation of CD8+ T cells. In particular, TNFR2 expression was seen on 6.5% of naïve, 14.6% of central memory, 37.9% of effector memory and 45.2% of CD45RA-revertant effector memory (TEMRA) CD8+ T cells. In contrast, common gamma chain cytokine receptor expression was skewed towards less differentiated T-cell subsets. For example, IL-7R was expressed by 63% of central memory populations but only 18.4% of the TEMRA subset. Comparable expression of IL2R was 12.1% on TCM and 2% on TEMRA. Of interest, IL-6 receptor expression was predominantly expressed by naïve CD8 T-cells (69.5%). In support of these results, we went on to show that expression of TNFR2 was inducible on primary T cells following activation with anti-CD3 and IL-2 in vitro. Healthy CMV seropositive donors had a larger median number of CD8+ T cells expressing TNFR2 (53%) in comparison to CMV seronegative donors (15%), (p<0.0001), consistent with the known accumulation of differentiated T-cells within CMV seropositive individuals.The expression of TNFR2 was then examined on CMV-specific CD8 T-cells which were undergoing acute expansion in response to viremia in six haemopoietic stem cell transplant patients. The expansion of CMV-specific CD8 T-cells was accompanied by an increase in the intensity of TNFR2 expression which later decreased during the retraction of antigen-specific T-cells during resolution of viremia. In order to explore the functional significance of TNFR2 expression, T-cells isolated from healthy donors were treated with recombinant TNFR2-specific ligand. This induced cell loss ranging from 13% to 60% of all CD8 T-cells in relation to untreated control cells, with selective depletion of the TNFR2+ population. A similar proportion of CMV-specific T-cells from transplant patients were eliminated by ex vivo stimulation of TNFR2. In conclusion our work shows that TNFR2 expression increases during differentiation of CD8+ T cells. In addition, we were able to utilize virus-specific T cells from SCT patients to show that expression is increased during the acute response to stimulation with antigen. We also provide evidence that TNFR2 activation can lead to the partial elimination of antigen-specific CMV-specific T-cells and it may thus play an important role in the ‘deflation’ of a pathogen-specific T-cell immune response following resolution of infection. These data suggest that TNFR2 expression may act as a ligand to signal activation-induced cell death in late differentiated populations of CD8+ T cells. Further investigations are required to assess the molecular pathways of TNFR2 signalling that are activated following receptor ligation in vivoand whether or not these are disrupted in disorders associated with chronic CD8+ T cell lymphproliferation. (1) L. Ban et al, PNAS 2008, 105: 3644 Disclosures No relevant conflicts of interest to declare.