CD8 T cell memory is commonly divided into two types of memory cells: central memory T cells residing in lymph nodes and effector memory T cells residing in peripheral tissues. Relative contribution of each type of memory CD8 T cells in protection against infection depends on many factors, for example, nature of infection. In general, it is assumed that central memory T cells (Tcm) are more potent compared to effector memory T cells (Tem) due to their higher self-renewal capacity.
Recently, a paper published in Nature added another twist to this story. In this study by Jiang X et al. (1), the authors examined CD8 T cell response to localized vaccinia virus (VV) skin infection. First, CD8 T cell response to VV skin infection appears to be CD4 T cell-independent. Activated VV-specific CD8 T cells accumulated in the draining lymph nodes and in the skin. Then, the authors analyzed recirculation pattern of VV-specific memory CD8 T cells by surgically creating parabiotic pairs between infected (30-days post infection) and uninfected mice. As expected, 8 weeks post joining, VV-specific CD8 T cells numbers in the lymph nodes of previously-infected and uninfected mice became equalized. However, to their surprise, even after 24 weeks of parabiotic state they could not detect VV-specific CD8 T cells in the skin of uninfected mouse suggesting that skin resident memory CD8 T cells did not recirculate. For the further analysis, the authors separated co-joined mice after 8 weeks and challenged them with the second VV skin infection (recall response). Again, to the authors surprise, previously parabiotic-uninfected mouse that had the lymph node resident VV-specific Tcm CD8 T cells but not skin-resident VV-specific CD8 T cells, cleared VV skin infection as if it was completely naive mouse, while previously infected mouse with skin-resident VV-specific CD8 T cells cleared VV skin infection with the significantly accelerated manner. Interestingly, if primary VV infection was through intraperitoneal (i.p.) route, then recall response to the secondary VV skin infection was weaker (at least 1000-fold weaker). These experiments indicate that
(a) VV skin infection generates very potent skin-resident VV-specific memory T cells that will localize throughout body skin
(b) VV i.p infection does not generate skin-resident VV-specific memory CD8 T cells
(c) At the end primary skin infection, skin-resident memory CD8 T cells stop to recirculate
(d) Tcm do not play any substantial role in protection against VV skin infection
The most surprising observation was the fact that VV-specific Tcm played almost no role in protection of naive mouse against VV skin infection. This implies that the efficacy of vaccination may be improved by mimicking the route of infection, for example, flu-vaccines may be more potent if introduced through upper-respiratory tract epithelial tissue rather than through skin needle-injection. From tissue point of view it may make sense to limit T cell accumulation when there is no “danger signature” associated with any prior skin infection. Thus, naive mouse with healthy skin may not send out any “come in” signals to T cells. In this regard, it would have been interesting if the authors would had analyzed T cell recirculation pattern during first 30-days of primary VV skin infection to know whether non-circulation pattern of resident memory T cells was because of healthy skin in uninfected mouse or because of change in T cell behavior itself.
David
P.S See second paper from the same group published in Science Translational Medicine that support their conclusion.
