The immune system is a highly balanced network of different cell types. This balance is disturbed in the setting of organ or stem cell transplantation, which can lead to graft rejection or "Graft versus host disease" (GvHD). Conventional pharmacological treatment by broad immune suppression is restricted by dose-limiting side effects. A novel strategy for prevention and control is cell therapy. This applies particularly to GvHD. A number of phase I trials have already been launched. The most appropriate cell type appears to be the regulatory T (Treg) cell as it is a natural "suppressor" of the immune system. Treg cells are able to inhibit various effector cells including CD4+ and CD8+ T cells, the main drivers of GvHD. Like other T cells, also Treg cells can be divided into naïve and memory-type cells. We have previously identified effector/memory Treg cells (T(REM)), the regulatory counterparts of CD4+ effector/memory T cells (T(EM)). T(REM) may be particularly suited to inhibit proinflammatory reactions in peripheral tissues as they express the chemokine receptor CCR6, a feature they share with proinflammatory Th17 cells. As specific marker, they also express CD39 but lack the expression of CD49d and CD127. We could show that a simple depletion of CD49d and CD127 expressing cells yields a population of "untouched" Treg cells that is highly pure and largely consist of highly suppressive T(REM) cells. Mouse models have confirmed the efficacy of Treg cells in controlling GvHD but the translation has been lagging. First clinical trials suggesting safety of adoptive Treg transfer increase the need for methods that allow obtaining clinical-grade Treg cells in sufficient amounts. The new approach may therefore provide a promising new alternative to facilitate a simple access to these cells.