Limited efforts were made to

Limited efforts were made to exploit the naturally-occurring antigen-specific Treg for ex vivo expansion. This study sets up a reproducible protocol for the expansion of insulin-specific Treg isolated from NOD mice. These PPADS tetrasodium salt exhibited higher suppressive capacity compared to the polyclonal Treg and was mediated both by cell contact (GITR) and soluble mediators (IL-10 and TGF-β).
Material and methods
Results
Discussion Autoantigen-specific Treg are extremely rarely found in the blood and lymphoid tissues, but have been shown to possess significant potential for inhibition of autoantigen-specific T effector cells. For example, as low as 5000 BDC 2.5-specific Treg were enough to delay the T1D transfer in BDC 2.5 transgenic NOD mice. All autoantigen-specific Treg that have been investigated for T1D treatment were transgenic, either they come from the animal that expresses the specific TCR for autoantigen, or T cells were engineered via transfection of viral vectors encoding specific T cell receptors (TCRs) or chimeric antigen receptors (CARs) (Zhang et al., 2018). We have chosen to try to expand insulin-specific Treg because insulin is the key autoantigen for driving the initiation and progression of autoimmune beta cell destruction (Nakayama et al., 2005; Nakayama et al., 2007; Jaeckel et al., 2004). The potency of innate insulin-specific Treg in the combat against T1D is represented by the fact that high frequency of these cells were found in the blood of prediabetic subjects that have long history of anti-islet autoimmunity compared to healthy subjects and those with diabetes onset at a young age (Serr et al., 2016). Our results suggest that insulin-specific Treg can be found in all lymphoid tissues in NOD mice. Also, the observations made by Spence et al. about the higher frequency insulin-specific Treg compared to the conventional T cells within the pancreatic infiltrates was confirmed in this study (Spence et al., 2018). The kinetics of insulin-specific Treg proliferation during encounter with the InsB9:23-loaded DC showed that these cells were significantly more expanding compared to the conventional insulin-specific T lymphocytes. This preference for Treg expansion was not surprising, as previously published observations suggest that mature DCs were able to stimulate Treg differentiation from naïve CD4+CD25− cells (Banerjee et al., 2006). Surprisingly, tolDCs were equally efficient in promoting insulin-specific Treg as were mature DCs. Hypothetically, this can be due to the fact that in the “normal” system, the proliferation of effector T cells to the self- antigen is limited, and even mature DC have the capacity to promote autoantigen-specific Treg. Treg obtained by in vitro enrichment were fully functional since they exhibited suppressive effect on T efector cells. Most importantly, these Treg had higher inhibitory influence compared to polyclonal, freshly isolated CD4+CD25high Treg cells suggesting that their action is specific and more focused. Also, the fact that smaller number of insulin-specific Treg was effective suggests their higher potency. This was also documented in the literature when as little as 5000 BDC2.5-specific Treg were used to halt T1D in the transfer model of the disease, compared to 105 polyclonal Treg (Tarbell et al., 2004). Treg exhibit their inhibitory effect on T cell proliferation either through cell contact (by expression of various surface molecules including GITR and CTLA-4) (Kumar et al., 2018) or by secretion of immunosuppressive cytokines (Yu et al., 2018). Also, Treg express on their surface ecto-nucleotidase CD39 that converts ATP to adenosine, an immunosuppressive molecule (Dwyer et al., 2007). The expanded insulin-specific Treg probably used versatile ways for suppression: inhibition by cell-to-cell contact through GITR (and to smaller extent through CTLA-4), soluble factors (IL-10 and TGF-β) and through generation of adenosine in the extracellular milieu.