Sara van den Berg

158 Chapter 6 markers or functionalities govern cellular production and lifespan. In conclusion, although we cannot rule out the possibility that small differences in production rates have been missed in our in vivo deuterium labelling study, our data suggest that no substantial differences in underlying dynamics are required to maintain high CD8 + CMV-specific T-cell numbers. Finally, we found that the estimated in vivo production rates of memory T-cells based on deuterium labelling correlated with the expression of late-stage differentiation markers and the proliferation marker Ki-67. Expression of Ki-67 correlated weakly with the in vivo production rate, possibly due to the ‘snapshot’ nature of this marker, which makes it more sensitive to day to day variations and ongoing immune activations. The expression of the late-stage differentiation markers CD57 and KLRG-1 correlated negatively with the in vivo production rates. This is in line with in vitro results, which showed that CD57 + and KLRG-1 + T-cells are less able to proliferate upon in vitro T-cell receptor (TCR) stimulation [32, 33]. However, in mice, it was shown that CD57 + as well as CD57 - antigen-specific T-cells are able to divide upon IL-7 stimulation, a measure for homeostatic proliferation, and are therefore not hampered in their maintenance [44]. To give more clarity into this matter, a follow-up study would be needed where the in vivo production rate of sorted CD57 + and CD57 - cells would be directly compared. Altogether, this suggests that a combination of senescence markers could potentially be used as a proxy marker for cellular lifespan. In the context of ageing of the immune system, or immuno-senescence, potential differences in T-cell maintenance between CMV- and CMV+ individuals are of special interest. CMV has been suggested to be a driving force behind accelerated ageing of the immune system [45-47]. Previously, we did not find a difference in memory T-cell maintenance between aged and young individuals [22]. Here we expand on this finding by showing that infection with CMV has only minor effects on T-cell dynamics. Together, this suggests that during healthy ageing, with or without CMV infection, no substantial changes in memory T-cell dynamics arise. Secondly, our work shows that maintenance of CMV-specific memory T-cells, despite their high abundance, does not necessarily require altered cellular dynamics. This suggests that the characteristics of large memory T-cell responses may be set at the start of the T-cell response. This work thereby also offers important implications for vaccination strategies that aim to induce long-term and high immunological memory responses, for example using CMV-vector based vaccines. ACKNOWLEDGEMENTS: We thank René van Boxtel and Inge Pronk for help in designing the study and Jeroen van Velzen, Pien van der Burght and Sebastian van Burgh for help with the flow cytometry sorting.