Modulation of natural killer cell and T-cell functions by CD94/NKG2A receptors

Sammanfattning: Natural Killer (NK) cells constitute a subset of lymphocytes capable of cytokine production and/or cellular cytotoxicity. Although NK cells can be activated experimentally by a multitude of agents, the in vivo mechanisms of activation and regulation are not fully elucidated. CD94/NKG2A is a surface receptor expressed on NK cells, which has been shown to deliver an inhibitory signal (preventing activation) upon binding the nonpolymorphic and ubiquitously expressed HLA-E molecule. HLA-E is coexpressed with other MHC class I molecules, as it is dependent on peptides derived from these for its stability. Physiologically it may confer protection from NK cell mediated autoimmunity. The work presented in this thesis is focused on the ligand-receptor interaction of HLA-E and CD94/NKG2A and the role of this in regulation mainly of NK cells, but also of cytotoxic Tcells. Heat shock protein (hsp) 60 is a molecule that has been implicated in autoimmune phenomena. We searched for putative HLA-E binding peptides in the human hsp60 amino acid sequence, which might function as alternatives to M14C class I derived peptides. We identified four such peptide sequences. We hypothesized that these HLAElhsp60 signal peptide (sp) complexes might modulate NK cell cytotoxicity through the inhibitory CD94/NKG2A receptor or the activating CD94/NKG2C receptor. The first paper describes experiments in tumour cell lines transfected with HLA-E and loaded with hsp60sp. Such HLA-E molecules were not permissive for protection through CD94/NKG2A. On the basis of our results, we propose a novel mechanistic model by which NK cell activity is triggered when the inhibitory interaction of HLA-E with CD94/NKG2A is cancelled by HLA-E complexed with hsp60-derived peptides. The second paper is a study of NK cell phenotype and function in a group of patients affected by chronic arthritides. NK cells isolated from the site of inflammation are activated, and have the phenotype (CD56bright) characteristically seen in a minor blood subpopulation of cytokine producing, KIR-negative but strongly CD94/NKG2A positive NK cells. Functional assays on these cells indicated that their cytokine production could be modulated via the CD94/NKG2A receptor, with the same peptide specificity as observed for blood NK cells in the first paper. In order to further understand immunomodulation by NK cells, we investigated their role in dendritic cell (DC) differentiation. In the third paper we showed that monocytes rapidly differentiate into DC when co-cultured with IL-15 activated NK cells. The cytokine producing CD56bright NK cell subset was particularly potent at inducing DC differentiation. SF NK cells were able to mediate DC development in the absence of in vitro activation. IFN-? acts as an important regulator of inflammation, partly through up-regulation of M11C class I expression, which implies an increased susceptibility to T-cell receptor mediated cytotoxicity. In the fourth paper we described the paradoxical protection of short-term ovarian carcinoma cell lines from cytotoxic lysis following IFN-? treatment. By antibody blocking of the CD94/NKG2A inhibitory receptor on cytotoxic T-cells, we demonstrated that the interaction of HLA-E with this receptor provides protection from cellular cytotoxicity. In summary these data further our knowledge of the relation between the phenotype and the effector functions of NK cells, as well as of the interactions between HLA-E and CD94/NKG2 family receptors expressed by NK cells and T-cells. Manipulation of this receptor-ligand interaction may be exploited in immunotherapy.