Daan Pieren

103 Co-expression of TIGIT and Helios by CD8+ T cells In addition, the proportion of the TIGIT + Helios + subset within the CD8 + T-cell population negatively correlated with proliferation of the total CD8 + T-cell population ( Figure 1E ), indicating a significant contribution of this subset to decline in proliferative capacity of the overall CD8 + T-cell population. Finally, TIGIT + Helios + cells characteristically showed reduced expression of the co-stimulatory receptor CD226 ( Figure 2A-C ), which is the functional counterpart of the co-inhibitory TIGIT and has recently been shown to mark senescent T cells in mice [24]. Moreover, TIGIT + Helios + cells showed a trend towards slightly higher levels of CD57 and killer cell lectin-like receptor G1 (KLRG1), which are markers that have previously been associated with cellular senescence of T cells [25,26] ( Figure 2D, E ). Overall, these findings show that Helios refines the description of immunosenescence among the previously reported TIGIT + population that increases with age [13], as Helios more accurately defines immunosenescent cells among the TIGIT + CD8 + T-cell population. Therefore, co-expression of TIGIT and Helios can be used to accurately define functionally senescent CD8 + T cells. Late-differentiated CD8 + T cells accumulate with age at the cost of early- differentiated T cells CD8 + T cells downregulate CD27 and CD28 expression during aging by conversion from CD27 + CD28 + early-differentiated cells to CD27 + CD28 - intermediate-differentiated cells and, finally, into CD27 - CD28 - late-differentiated T cells leading to accumulation of the latter subset during aging [8,9,12,27]. Accumulation of CD27 - CD28 - cells has often been ascribed to latent infection by cytomegalovirus (CMV) as a result of repetitive T-cell stimulation [9,28,29]. However, we hypothesized that such T-cell differentiation would be a more general phenomenon and also occurs in the absence of CMV infection. We therefore assessed the association of differentiation status of CD8 + T cells based on expression of CD27 and CD28 with age in PBMC of CMV seronegative individuals (n=50, 21-82 years). We found that early-differentiated cells were most abundant amongst CD8 + T cells of these donors, compared to the presence of intermediate- and late-differentiated cells ( Figure 3A, B ). In relationship to age, the proportion of late-differentiated cells increased, whereas the proportion of early-differentiated cells declined ( Figure 3C ) . The proportion of intermediate-differentiated cells did not change with age. These findings show that late-differentiated cells accumulate with age at the expense of early- differentiated cells and indicate that such accumulation does not depend on CMV. 4

RkJQdWJsaXNoZXIy ODAyMDc0