Team Tolerance

PD-1: Preventing T cell pathogenesis

CD8 T cells develop a T cell receptor (TCR) in the thymus that has the ability to both recognize the major histocompatibility complex 1 (MHC-I; also known as HLA) protein and the peptide contained within that molecule (antigen). CD8 T cells undergo deletion for binding of the MHC (positive section) and absence of recognition of self-antigens (negative selection) in the thymus, but this is an imperfect process.

Studies in humans have found that self-antigen reactive T cells are present in circulation, but it is thought that these cells have been inactivated by mechanisms of peripheral tolerance. Previous work has focused on the role of two processes: Anergy and Deletion in preventing autoimmunity from self-reactive CD8 T cells, which should result in irreversible silencing and protection from disease. This included a putative role for PD-1 in promoting T cell tolerance via anergy.

The advent of immune checkpoint blocking therapies revealed the common occurrence of immune related adverse events, which are autoimmune like events that are thought to be mediated by self-reactive T cells. Yet this raised the question of why these self-reactive T cells were not previously anergized or deleted, which led us to re-examine the mechanisms for peripheral tolerance in skin using our NINJA model.

Using the NINJA mouse, we identified that skin-reactive CD8 T cells do not become anergic, but rather these cells migrate into the skin where they become enriched in the dermis. Moreover, these “tolerant” T cells recruit CD16+ CD14+ CD11b+ CD11c+ macrophages, which we hypothesize prevent the T cells from reaching the epidermis, where their antigen is expressed. In the context of anti-PD-1 blockade, the tolerant T cells become pathogenic, but contrary to our expectation, this occurs at the post-transcriptional level. Thus, peripheral tolerance at this model is regulated within the tissue rather than by a more canonical differentiation towards anergy or deletion.

This work has several practical implications. The first is the idea that healthy people may have self-reactive CD8 T cells in their tissues that would be pathogenic, if not for the sustained function of PD-1 and other checkpoint receptors. This we believe is the origin of the skin irAEs we and others looked at in patients. Moreover, we believe that the mechanisms of regulation of self-reactive T cells may differ between tissues, which reflects our ongoing focus in the lab. Finally, we remain intrigued by the role of T cells in recruiting myeloid cells into tissue and the potential for T cells to regulate innate immune cells within tissues. We hypothesize that (under certain conditions) T cells may require interactions with myeloid cells in tissues for their proper function, and that this state will be established by the T cells entering the tissue. This system would allow myeloid cells to regulate immune response in healthy tissue by reading out the presence of inflammatory mediators, while providing a mechanism for regulation of the T cells under homeostatic conditions.