Dynamics of MHC class 1 recognition by natural killer cells : from receptor modulation to ligand acquisition

Abstract

Natural killer (NK) cells are part of the innate immune system and can kill tumor cells and virus-infected cells without prior sensitization or clonal expansion. NK cells do not, in contrast to T and B cells, express any rearranged antigen-specific receptors. Instead, NK cells express an array of activating and inhibitory receptors, and a balance between these two opposing receptor groups decides the outcome of a target cell interaction. The major group of inhibitory receptors expressed by murine NK cells is the Ly49 family of receptors, which contains both activating and inhibitory members. The Ly49 receptors bind to MHC class I molecules on target cells, and are expressed on partially overlapping subsets of NK cells.

The first part of this thesis deals primarily, but not exclusively, with the Ly49A receptor, an inhibitory receptor expressed on 20 % of the NK cells that is specific for the MHC class I molecule H-2Dd . When this work was started, it was known that when the Ly49A receptor on NK cells engaged H-2Dd molecules on other cells, lysis was inhibited. Thus, interactions with "external" MHC class I molecules by Ly49 receptors were well established. The major conclusion from the first part of this thesis is that, in parallel with "external" interactions, "internal" MHC class I ligands play an important role in regulation of the Ly49A receptor expression. In the absence of "internal" MHC class I molecules, Ly49A receptor levels were rapidly upregulated on NK cells despite an abundance of cells expressing "external" H-2Dd molecules. Of equal importance was that we were the first to show that mature NK cells could regulate their Ly49 receptor levels in a flexible, almost adaptive, way when the MHC class I phenotype of the environment changed.

As the study continued, we observed that mature Ly49A+ NK cells not only changed their expression levels of Ly49A receptors upon interactions with "external" H-2Dd molecules, but at the same time acquired these H2-Dd molecules from surrounding cells and incorporated them into their own cell membrane. This novel finding was confirmed in vivo as well as in two independent in vitro-assays. We also showed that MHC class I acquisition was not a selective property of Ly49A+ NK cells but occurred also when Ly49C+ NK cells interacted with surrounding cells expressing the Ly49C ligand H-2Kb. Interestingly, NK d cells expressing Ly49A and Ly49C simultaneously showed an unexpected synergy in H-2Dd acquisition, but only when the strong Ly49C ligand H-2Kb was also present on surrounding cells. These data suggest that it may not be possible to correctly predict the biology of Ly49 receptors simply by analysing affinity and specificity in cell-free systems, but that the molecular context under which the receptor is expressed may be critical. MHC class I acquisition had different functional consequences for the NK cell that depended on the origin of the NK cell under study.

Of particular interest was the finding that NK cells with acquired MHC class I molecules were less efficiently inhibited by "external" MHC class I molecules suggesting that aqcuired MHC class I molecules became "internal" and interfered with Ly49 receptor biology. In this thesis, I have also studied development of self tolerance in NK cells of mixed bone marrow chimeras, a process containing adaptive and flexible components strikingly similar to Ly49 receptor regulation. The data presented here confirm previous findings that tolerance to the lack of MHC class I is not due to clonal deletion of potentially autoreactive NK cell subsets, but rather to a nonresponsiveness that is reversible when the tolerizing cells are removed by cell sorting. We also show that tolerance is not due to increases in expression of additional Ly49 receptors, suggesting alterations in activating pathways.

Finally, tolerance is variable when less than every 10:th hematopoietic cell is MHC class I-deficient, suggesting a model in which NK cells need to continuously monitor the environment to adapt fully to self. In summary, this thesis provides evidence for a previously unappreciated flexibility and adaptability in NK cells, governing diverse processes such as functional tolerance and regulation of Ly49 receptor levels at the cell surface, providing a basis for Ly49-mediated MHC class I acquisition by mature NK cells.