Partial signaling by CD8(+) T cells in response to antagonist ligands

Structural variants of art agonist peptide-major histocompatibility complex (MHC) molecule ligand can show partial agonist and/or antagonist properties. A number of such altered ligands appear to act as pure antagonists. They lack any detectable ability to induce T cell effector function and have been described as unable to induce calcium transients and turnover of inositol phosphates. This has been interpreted as an inability of these ligands to initiate any T cell receptor (TCR)-dependent signal transduction, with their antagonist properties ascribed to competition with offered agonist for TCR occupancy. Yet antagonists for mature CD8(+) T cells can induce positive selection of thymocytes, implying active induction of T cell differentiation events, and partial agonists or agonist/antagonist combinations elicit a distinctive pattern of early TCR-associated tyrosine phosphorylation events in CD4(+) T cells. We have therefore directly examined proximal TCR signaling in a CD8(+) T cell line in response to various related Ligands. TCR engagement with natural peptide-MHC class I agonist resulted in the same pattern of early TCR-associated tyrosine phosphorylation events as seen with CD4(+) cells, including accumulation of both the p21 and p23 forms of phosphorylated zeta, phosphorylation of CD3 epsilon, and association of phosphorylated ZAP-70 with the TCR. Two antagonists that lacked the ability to induce any detectable CTL effector response (cytolysis, esterase release, gamma interferon secretion, interleukin-2 receptor alpha upregulation) were nevertheless found to also induce TCR-dependent phosphorylation events. In these cases, there was preferential accumulation of the p21 form of phospho-zeta: without net phosphorylation of CD3 epsilon, as well as the association of nonphosphorylated ZAP-70 kinase with the receptor. These data show that variant ligands induce similar TCR-dependent phosphorylation events in CD8(+) T cells as first observed in CD4(+) cells. More importantly, they demonstrate that some putatively pure antagonists are actually a subset of partial agonists able to induce intracellular biochemical changes through the TCR. This delivery of a partial signal by antagonists raises the possibility that antagonism in some cases may result from active interference with stimulation of effector activity by agonist in mature T cells, while the same variant signal could selectively trigger intracellular events that allow positive without negative selection in thymocytes.