A new role for NKG2D signaling in CD8+ T cells and autoimmune diabetes

Abstract

The demands placed on the immune system are immense and highly complex. It must protect the body against untold threats while maintaining a balance between immune defense and autoimmune damage. One major player in immune recognition is the receptor Natural-Killer-Group-2-Member-D (NKG2D), best known for its expression on natural killer (NK) cells and CD8+ T cells, where it recognizes NKG2D ligands expressed by stressed cells following viral infection or cancerous transformation. NKG2D is most well studied for its role in tumor immunity, for which NKG2D based therapies are currently being developed clinically. Despite this, it is apparent that NKG2D has other poorly understood immune regulating functions, such as its implicated involvement in type 1diabetes and other autoimmune disorders. However, the mechanism by which NKG2D signaling affects diabetes has been unclear. We therefore sought to further clarify the role NKG2D plays in autoimmune diabetes development. Canonically, NKG2D engaging NKG2D ligand results in immune killing of the infected or damaged ligand-bearing cell by NK cells, and costimulation of CD8+ cytotoxic lymphocytes (CTL) augmenting CTL responses and target cell killing. However, we and others have observed robust expression of NKG2D ligands by seemingly healthy immune cells. In the work presented in this dissertation, I show the work I performed to investigate how this expression by seemingly healthy cells affects the immune response. In particular I focused on CD8+ T cells, which express both NKG2D and NKG2D ligands after activation, and play a key role in the development of autoimmune diabetes. Using the non-obese diabetic (NOD) mouse model of autoimmune diabetes, I found that engagement of the NKG2D ligand H60a, expressed by NOD T cells, during CD8+ T cell differentiation resulted in decreased cytokine production upon later antigen stimulation. This correlated with other findings from our lab showing decreased incidence of autoimmune diabetes in microbiota-depleted NKG2D sufficient versus NKG2D deficient NOD mice. Further, I showed that this correlated with NKG2D signaling driving an increase in CD8+ T cells with a central memory phenotype in both mouse and human cells. I then found significantly reduced transfer of NOD diabetes by these central memory phenotype CD8+ T cells compared to effector/effector memory CD8+ T cells. NKG2D on human CD8+ T cells increases generation of a subset of CD8+ central memory phenotype T cells that match the surface phenotype of a described regulatory CD8+ T cell population. Finally, I found that NKG2D stimulation increased expression of the inhibitory receptor PD-1 by NOD CD8+ T cells. I therefore suggest a previously undescribed role for NKG2D signaling between healthy cells in immunity and immune regulation, and take steps towards answering critical questions to determine whether modulation of NKG2D signaling could be used in intervention strategies in type 1 diabetes.