Memory depending on an intimate relationship

Accumulating evidence suggests a critical role for interleukin-15 (IL-15) in the development and maintenance of CD8 memory T cells. Nevertheless, there are some unsolved issues as to how IL-15 acts on target cells. Based on in vitro studies, it was postulated that nonlymphoid cells secrete IL-15, and T/natural killer (NK) cells expressing all 3 IL-15 receptor (IL-15R) components (α, β, and γ chain [γC]) bind it and proliferate.

However, Lodolce et al¹  challenged this paradigm. Using adoptive transfer experiments involving IL-15Rα^–/– and wild-type (wt) mice, they demonstrated that although the expression of IL-15Rα by radiation-resistant “environmental” cells is crucial, the expression of IL-15Rα by T cells is dispensable for the survival of CD8 T cells. But how do T cells respond to IL-15 in vivo without IL-15Rα?

In parallel, Dubois et al²  demonstrated that activated monocytes bear membrane-associated IL-15 bound by IL-15Rα and present IL-15 in trans to neighboring T cells. This IL-15 transpresentation model presented a reasonable solution to the paradox presented by Lodolce et al.¹  Taken together, the suggested scenario would be that IL-15 is first captured by environmental cells expressing IL-15Rα, and then it is transpresented to CD8 memory T cells or NK cells to support their proliferation.

In this issue, Schluns and colleagues (page 988) further advance the model by defining the specific cell types that are capable of presenting IL-15 in trans to CD8 memory T cells. Using bone marrow (BM) chimeras generated from IL-15Rα^–/– and wt mice, they first confirmed the observation made by Burkett et al³  that memory CD8 cells proliferate in wt mice even if the cells lack the autologous expression of IL-15Rα, but their proliferation critically depends on IL-15Rα expression by bone marrow–derived cells.

The most intriguing observation of their study is that CD8 memory T cells proliferated in the spleen of chimeras expressing IL-15Rα only on BM-derived nonlymphoid cells, but not in chimeras expressing IL-15Rα on parenchymal cells. Interestingly, memory CD8 T cells that reside in the lungs did not show such preference in their cell-type dependency on IL-15Rα expression, as lung memory CD8 T cells proliferated, albeit slightly less robustly, in chimeras expressing IL-15Rα on either BM-derived or parenchymal cells.

This observation suggests that a cell-specific molecule may be required in addition to IL-15Rα for the IL-15 transpresentation to function. In addition, memory CD8 T cells from different tissues may have different characteristics, which may account for the tissue-specific occurrence of various immunologic events, including the development of autoimmune diseases. Last, though it may sound paradoxical, we should not be so hasty to ignore the relevance of autologous IL-15Rα expression on T cells, as the effect of IL-15 on the long-term survival of memory CD8 T cells is not fully understood.

Gett et al⁴  reported that the expression levels of IL-15Rα, determined by the strength of the initial antigen stimulation, define the fitness status of CD8 T cells and affect their long-term survival in response to IL-15. Nonetheless, the IL-15 transpresentation paradigm seems to shed a new light on our understanding of the involvement of IL-15 in the maintenance of CD8 memory T cells.