Peripheral but Not Central Tolerance Regulates RBC Autoreactive CD4+ T Cells in a Novel Mouse

Background:  Elucidating the mechanisms of T cell tolerance towards autoantigens can be critical for therapeutic efforts to re-establish tolerance in autoimmune diseases or break tolerance in tumor immunotherapy. Tolerance mechanisms have been well described towards tissue-restricted antigens and include deletion, anergy, or regulatory cells; however, T cell tolerance mechanisms against RBC-restricted antigens are poorly understood.  Herein, we utilized the HOD mouse to investigate T cell tolerance mechanisms to RBC-specific antigens. The HOD mouse expresses a triple fusion protein consisting of hen egg lysosyme (HEL), ovalbumin (OVA), and human blood group molecule, Duffy (HOD) driven by a RBC-specific promoter. A TCR transgenic mouse that recognizes an OVA peptide presented by MHI II (OT-II mouse) was crossed with the HOD mouse to allow analysis of T cells autoreactive to an RBC specific antigen.  

Methods: To evaluate central tolerance, OVA specific T cells were analyzed from thymii from 6-8 weeks old OT-II⁺HOD⁺ mice (OT-II⁺HOD^- littermates were used as controls for T cells in absence of autoantigen).  Splenic CD4⁺ T cells from both OT-II⁺HOD⁺ and control OT-II⁺HOD^- mice were CFSE-labeled and analyzed 1) in vitro by co-culturing with splenocytes pulsed with whole soluble OVA or 2) in vivo by adoptively transferring into B6.Thy1.1 recipients, followed by transfusion of HOD RBC.  Proliferative responses of cells were measured by CFSE dilution analyzed with flow cytometry.  

Results: There is no detectable autoantibody production against the HOD antigen in OT-II⁺HOD⁺ mice.  In the thymus, similar frequencies and total nbers of OVA-specific CD4⁺ single positive (SP) T cells were observed between OT-II⁺HOD⁺ and OT-II⁺HOD^- mice.  This was not because OT-II⁺HOD⁺ mice were incapable of deleting autoreactive T cells, since OT-II⁺HOD⁺ mice i.v injected with soluble whole OVA, known to induce deletion, had a significant decrease in the CD4⁺ SP T cells.  Phenotypic analysis of mature CD4⁺ T cells in the spleen revealed fewer OVA-specific CD4⁺ T cells in OT-II⁺HOD⁺ compared to OT-II⁺HOD^- mice (p=0.009).  The CD4⁺ T cells present in OT-II⁺HOD⁺ expressed higher levels of markers associated with T cell activation and proliferation such as CD69, CD44, Ki-67 and also decreased levels of L-selection.  Moreover, these cells expressed elevated levels of markers associated with T cell anergy and tolerance, including PD-1, CD5 and LAG-3.  Finally, OT-II⁺HOD⁺ mice exhibited increased numbers of CD25⁺Foxp3⁺ T_regs (p=0.007).  In 2 of 2 experiments (3 mice/group), adoptively transferred CD4⁺ cells from OT-II⁺HOD⁺ did not proliferate in response to HOD RBCs while OT-II⁺HOD^- exhibited robust proliferation.  In contrast to the in vivo results, CD4⁺ cells from OT-II⁺HOD⁺ proliferated when co-cultured with splenocytes + whole soluble OVA in vitro.

Conclusions: We describe a model in which regulation of CD4⁺ T cells autoreactive for self-RBC antigen can be studied.   Although a significant number of RBC autoreactive T cells are present, and appear to have seen antigen, been activated (i.e. increased CD69 and CD44), and proliferated (increased Ki-67); these cells are nevertheless anergic as indicated by adoptive transfer studies.  The ability to proliferate in vitro but not in vivo in response to HOD RBC suggests that removal from the normal in vivo environment removes inhibitory factors, consistent with peripheral tolerance mechanisms in vivo.  Indeed, increased Tregs were observed in autoreactive mice compared to control mice.   In contrast, thymic deletion seems to play little or no role in this case.  Together, these findings identify peripheral and not central tolerance as major control mechanisms of RBC autoreactive CD4⁺ T cells.  Mechanisms may include alterations in expression of PD-1, CD5, and LAG-3 by autoreactive cells.