PD-1 expression is upregulated on adapted T cells in experimental autoimmune encephalomyelitis but is not required to maintain a hyporesponsive state.

T cell adaptation is an important peripheral tolerogenic process which ensures that the T cell population can respond effectively to pathogens but remains tolerant to self-antigens. We probed the mechanisms of T cell adaptation using an experimental autoimmune encephalomyelitis (EAE) model in which the fate of autopathogenic T cells could be followed. We demonstrated that immunisation with a high dose of myelin basic protein (MBP) peptide and complete Freund's adjuvant failed to effectively initiate EAE, in contrast to low dose MBP peptide immunisation which readily induced disease. The proportion of autopathogenic CD4+ T cells in the central nervous system (CNS) of mice immunised with a high dose of MBP peptide was not significantly different to mice immunised with a low dose. However, autopathogenic T cells in mice immunised with high dose MBP peptide had an unresponsive phenotype in ex vivo recall assays. Importantly, whilst expression of PD-1 was increased on adapted CD4+ T cells within the CNS, loss of PD-1 function did not prevent the development of the unresponsive state. The lack of a role for PD-1 in the acquisition of the adapted state stands in striking contrast to the reported functional importance of PD-1 in T cell unresponsiveness in other disease models.

Death, adaptation and regulation: the three pillars of immune tolerance restrict the risk of autoimmune disease caused by molecular mimicry.

Extensive cross-reactivity in T cell receptor (TCR) recognition of peptide-MHC (pMHC) complexes seems to be essential to give sufficient immune surveillance against invading pathogens. This carries with it an inherent risk that T cells activated during a response to clear an infection can, perhaps years later, respond to a self pMHC of sufficient similarity. This lies at the heart of the molecular mimicry theory. Here we discuss our studies on the disease-causing potential of altered peptide ligands (APL) based on the sequence of a single autoantigenic epitope, the Ac1-9 peptide of myelin basic protein that induces experimental autoimmune encephalomyelitis in mice. These show that the window of similarity to self for induction of disease by cross-reactive non-self peptides is actually quite restricted. We show that each of the three pillars of immune tolerance (death, anergy/adaptation and regulation) has a role in limiting the risk of molecular mimicry by maintaining a threshold for harm.