3A). In addition, KLRG1 expression was increased in IFN-γ secreting P14 cells but decreased in cells producing

IL-2 after stimulation (Fig. 3B). Thus, KLRG1 was preferentially expressed by CD8+ T cells with a “late” differentiation phenotype. To determine whether KLRG1 played a causal role in CD8+ T-cell differentiation, expression of the T-cell differentiation markers used above was compared in P14 T cells from KLRG1 KO and WT mice at the acute and at the memory phase of the LCMV infection. Adoptively transferred P14 T cells from KLRG1 KO and WT mice proliferated to the same extent in recipient mice after LCMV infection and gave rise to similar numbers of memory T cells (Fig. 4, left). In addition, expression of CD5, CD27, CD62L and CD127 Luminespib manufacturer on effector and memory P14 T cells and their capacity to secrete IFN-γ and IL-2 after antigen stimulation did not differ between KO and WT cells (Fig. 4, right). Thus,

these data indicate that the differentiation pathways of P14 T cells after LCMV infection were not altered in STI571 the absence of KLRG1. We and others have previously demonstrated that repetitively stimulated P14 memory T cells express high levels of KLRG1 and are impaired in their proliferation capacity after antigen stimulation 11, 29. In addition, recent data in the human system indicate that KLRG1 signaling induces defective Akt phosphorylation and proliferative dysfunction of highly differentiated CD8+ T cells 14. To determine whether KLRG1 is causally linked to impaired proliferation, P14 T cells from KLRG1 KO and WT mice were used in consecutive adoptive T-cell transfer experiment as outlined in Fig. 5A. Confirming previous findings 11, 29, “tertiary” P14 memory T cells from WT mice were mostly KLRG1+ and expanded only marginally after antigen stimulation in vivo when compared with naïve or primary Carbohydrate memory P14 cells (Fig. 5B and C). However, “tertiary” P14 memory T cells from KLRG1

KO mice also proliferated poorly, demonstrating that the impaired proliferative capacity of these cells was not due to KLRG1 expression. Infection of mice with MCMV leads to CD8+ T-cell memory inflation whereby the magnitude of the response to some epitopes (i.e. M38 or m139 in B6 mice) increases with time, whereas T-cell reactivity to other epitopes (i.e. M45 in B6 mice) contracts after the peak of the acute phase 30, 31. Interestingly, KLRG1 expression by M38- or m139-specific CD8+ T cells also increased in the course of the infection whereas the portion of KLRG1+ cells within the pool of M45-specific CD8+ T cells decreased (Fig. 6A). This observation prompted us to examine epitope-specific CD8+ T cells in MCMV-infected KLRG1 KO mice.