Acquisition and function of NK cell-associated molecules on T cells

Sammanfattning: One challenge for multicellular organisms, including humans, is to cope with a broad variety of microorganisms and their rapid replication and alterations. To do this, both fast and specific defense mechanisms are needed that can control threatening infections. Our immune system consists of two major parts: innate immunity, which is rapid and rather non-specific, and adaptive immunity, which is highly specific but requires more time for its development. Both innate (e.g. phagocytic cells and NK cells) and adaptive immune cells (B cells and T cells) are regulated by cellular interactions as well as by soluble factors. Each cell expresses a number of receptors, which can facilitate adhesion with other cells and/or mediate downstream signalling events that determine the outcome of immune reactions. In this thesis, I have focused on the expression of different NK cell-associated molecules on CD8+ T cells and the functional consequences of this. NKT cells have been classically defined as T cells expressing NK 1.1. These NKT cells are restricted to the MHC class I-like molecule CD1d and express an invariant T cell receptor, Jalpha281 -Valpha 14. My initial finding was that IL-2-activated spleen cell cultures derived from mice deficient in classical NKT cells contained NK 1.1+ T cells (the classical definition of NKT cells). We found that these cells were derived from conventional IL-2 receptor-beta+CD8+ T cells that had acquired NK1.1 upon activation. Subsequently, we observed that similar cells also appeared in the lungs of influenza A virus-infected mice. During the peak of infection, up to 10% of the CD8+ T cells in the lungs coexpressed NK 1.1. Around one third of the NK1.1+CD8+ T cells were specific for a single influenza A virus-derived epitope (ASNENMDAM) as determined by tetramer stainings and antigen-induced IFNgamma production. Interestingly, these NK1.1+CD8+ T cells exhibited a memory phenotype and coexpressed other NK cell-associated molecules including inhibitory Ly49 receptors and 2B4. This showed that all NK1.1+ T cells do not belong to a certain lineage and that the expression of NK cell-associated molecules may represent a state of activation for T cells. We found that, similar to NK 1.1, the NK cell receptor 2B4 was expressed on a subset of memory- like CD8+ T cells and induced upon IL-2 stimulation and during influenza A virus infection. Therefore, we decided to address the function of 2B4 when expressed on CD8+ T cells. To our surprise, we found that 2B4 acted as a ligand, rather than a receptor, for CD48 expressed on neighboring T cells and augmented both antigen-driven and IL-2-induced proliferation. These results prompted us to study whether other 2B4-expressing cells could costimulate T cells through 2B4/CD48 interactions. It was observed that NK cells as well as 2B4-expressing tumor cells enhanced both activation and proliferation of the T cells in response to IL-2 or CD3-crosslinking. NK cells can regulate adaptive immune responses by shaping the cytokine milieu during infection or autoimmune conditions. These data suggest that in addition to cytokine production, direct physical interactions between NK cells and T cells influence both activation and proliferation of T cells. We speculate that 2B4/CD48 interactions between immune cells may facilitate maintenance and reactivation of CD8+ memory T cells and regulate adaptive immune responses.