Isolation of Human Treg for Allogeneic Transplantation: Evaluation of Several Cell Sources.

Human regulatory T cells (Treg) have been shown to suppress activation and proliferation of alloreactive T cells in vitro. Animal studies suggest a role for Treg in prevention of GvHD together with sustained GvL-activity in allogeneic hematopoietic stem cell transplantation (HSCT). Clinical application of Treg in human HSCT requires substantial cell numbers isolated according to GMP-regulations. We examined several human blood products as cell sources for isolation of Treg using magnetic anti-CD25 coated beads together with AutoMACS and CliniMACS devices. Functionality of isolated Tregs were determined by MLR (1:1 ratio of CD4+CD25+ to CD4+CD25−) using allogeneic PBMC or OKT3 as stimulators. Single separation of buffy coats, leukapheresis products w/o prior G-CSF mobilisation on AutoMACS resulted in populations containing 70–80% CD4+CD25+ (52–64% CD25high) T cells. Despite co-separated CD8+ and CD19+ cells, this CD25+ population did not proliferate upon activation by cellular or antibody stimuli. However, it did suppress proliferation of syngeneic, alloreactive CD4+CD25− T cells stimulated by allogeneic PBMC or OKT3. These results can be obtained by Treg separated from buffy coats and unmobilized leukapheresis products, while Treg from G-CSF leukapheresis products still remain anergic upon stimulation but do supress much less efficiently the proliferation of CD4+CD25- T cells from the same cell product in an allogeneic MLR or by OKT3 activation. Interestingly, further preliminary results suggest that G-CSF Treg do completely suppress proliferation of naive CD4+CD25− T cells from the same donor upon activation. In additional experiments we tested buffy coats and unmobilized leukapheresis products for CD25+ enrichment by magnetic beads using the CliniMACS device in an attempt to isolate sufficient numbers of human Treg for allogeneic HSCT without in vitro expansion. Separation of cells from both sources by single positive enrichment resulted in CD25+ fractions containing 70–80% CD4+CD25+ (50–60% CD25high) T cells. Total isolated Treg cell numbers ranged from 87X10e6 (leukapheresis) to 3–5x10e6 (buffy coat) CD25high cells. Despite very similar results as obtained by small scale separation, the CliniMACS enriched CD25+ cells only modestly suppressed the proliferation of alloreactive CD4+CD25− T cells in MLR. But the anergic phenotype upon OKT3 stimulation was preserved. Further preliminary experiments revealed that CliniMACS separated leukapheresis Treg are still functional in completely inhibiting proliferation of CD4+CD25− T cells from the same donor separated by alternative methods. We conclude that CD25+ single separation is a feasible way to achieve human Treg populations with suppressive capacity in sufficient cell numbers from buffy coats and unmobilized leukapheresis products. Highly pure enrichment of CD25high T cells is not critical for the suppressive phenotype in CD25+ single separated cell fractions. Similar to G-CSF Treg, CD4+CD25+ T cells separated from CliniMACS appear to be functional in suppression of naive syngeneic alloreactive CD4+CD25− T cells and can be considered as useable for clinical application.