Adoptive immunotherapy with regulatory T cells (Tregs) is a promising treatment for allograft rejection and graft-versus-host disease (GVHD). Emerging data indicate that, compared with polyclonal Tregs, disease-relevant antigen-specific Tregs may have numerous advantages, such as a need for fewer cells and reduced risk of nonspecific immune suppression. Current methods to generate alloantigen-specific Tregs rely on expansion with allogeneic antigen-presenting cells, which requires access to donor and recipient cells and multiple MHC mismatches. The successful use of chimeric antigen receptors (CARs) for the generation of antigen-specific effector T cells suggests that a similar approach could be used to generate alloantigen-specific Tregs. Here, we have described the creation of an HLA-A2-specific CAR (A2-CAR) and its application in the generation of alloantigen-specific human Tregs. In vitro, A2-CAR-expressing Tregs maintained their expected phenotype and suppressive function before, during, and after A2-CAR-mediated stimulation. In mouse models, human A2-CAR-expressing Tregs were superior to Tregs expressing an irrelevant CAR at preventing xenogeneic GVHD caused by HLA-A2+ T cells. Together, our results demonstrate that use of CAR technology to generate potent, functional, and stable alloantigen-specific human Tregs markedly enhances their therapeutic potential in transplantation and sets the stage for using this approach for making antigen-specific Tregs for therapy of multiple diseases.

FOXP3-expressing T regulatory cells (Tregs) can be divided into two distinct subsets: naturally occurring Tregs (nTregs) that develop in the thymus, and induced Tregs (iTregs) that differentiate in peripheral tissues upon exposure to Ag in a tolerogenic environment. Recently it has been proposed that expression of Helios, an Ikaros family transcription factor, may specifically identify nTregs, allowing specific tracking of Tregs from different origins in health and disease. Surprisingly, we found that Helios- cells can be readily identified within naive (CD45RA+CD31+CCR7+CD62L+) FOXP3+ Tregs, a finding inconsistent with the notion that lack of Helios expression identifies Ag-experienced iTregs that should express memory markers. To investigate the phenotype and function of naive Helios+ and Helios− Tregs within the nTreg population, we isolated single-cell clones from each subset. We found that both Helios+ and Helios− nTreg clones have a similar suppressive capacity, as well as expression of FOXP3 and cell surface proteins, including CD39 and CTLA-4. Helios− nTregs, however, produced significantly more CCL3 and IFN-γ compared with Helios+ nTregs. Despite increased cytokine/chemokine production, Helios− FOXP3+ nTreg clones were demethylated at the FOXP3 Treg-specific demethylated region, indicative of Treg lineage stability. When cultured under Th1-polarizing conditions, Helios+ and Helios− nTreg clones had an equal ability to produce IFN-γ. Collectively, these data show that a lack of Helios expression does not exclusively identify human iTregs, and, to our knowledge, the data provide the first evidence for the coexistence of Helios+ and Helios− nTregs in human peripheral blood.

No abstract