Introduction: While there are numerous and major roadblocks to the development of Foxp3+ T-regulatory (Treg) cells for cellular therapy in allograft recipients, pharmacologic approaches to expand the endogenous Treg population have received far less attention and typically been confined to use of IL-2 complexes that have only very transient actions. In vitro, CD3/CD28 stimulation of conventional T cells in the presence of IL-2 and TGF-beta (TGFb) is well known to promote the development of Foxp3+ inducible Treg (iTreg) cells. However, the Foxp3 expression and the suppressive activity of iTregs is typically labile. We sort to remedy this situation.
Materials and Methods: We undertook in vitro studies of murine T cells that were activated and cultured under polarizing conditions, and in vivo studies using a cardiac allograft model.
Results: Culture of conventional CD4+CD25-YFP- T cells isolated from YFP+Foxp3+ mice for 4 d with CD3/CD28 plus IL-2 did not induce Foxp3, whereas addition of TGFb (2 ng/ml) led to ~45% Foxp3+ cells (flow cytometry). While use of rapamycin (RPM, 10-20 nM) instead of TGFb led to <10% Foxp3+ cells, their combination resulted in ~85% Foxp3+ cells. qPCR showed the combination increased Foxp3, TGFb, IL-10, Eos and Helios mRNA, and ChIP studies showed increased Smad2/3 binding at the Foxp3 promoter and CNS1 site, and increased H3K4me3 and Foxp3 binding at the CNS2 enhancer. Western blotting showed that TGFb activated the Smad pathway but also increased Akt phosphorylation. It is known that Akt activation leading to phosphorylation and acetylation of FOXO1 results in diminished Treg function, and RPM significantly decreased phosphorylation of Akt in our studies. Moreover, combination of TGFb and RPM led to a 2-fold increase in Foxp3 protein compared to either agent alone, was accompanied by increased phospho-Stat5, and the associated iTregs were fully suppressive in vitro.
In vivo, daily injection of TGF beta1 (1 µg/kg/d, 7 d) or use of RPM (0.5 mg/kg/d, 28 d Alzet pump) in C57BL/6 recipients of BALB/c cardiac allografts did not increase Foxp3+ cells or reduce CD4 or CD8 T cell activation (CD44hiCD62Llo) at 7 days post-Tx, or prolong allograft survival. However, combined TGFb/RPM led to a doubling of Foxp3+ Tregs when assessed at day 14 post-Tx, suppressed T cell activation, and resulted in permanent cardiac allograft survival (>100 days, p<0.01). The effects of combined TGFb/RPM on allograft survival were abrogated by use of a neutralizing mAb to TGFb.
Conclusion: The combination of TGFb and RPM has a powerful effect on iTreg development and induces permanent allograft survival in fully MHC-disparate mice. Much yet remains to be learnt regarding the pathways that suppress Foxp3 expression in conventional T cells, and how Foxp3 expression can be induced and stabilized to create functional Tregs. However, our work shows progress in this regard, and may ultimately have important consequences for clinical allograft recipients.