Mesenchymal stem cells (MSCs) are attractive source for regenerative therapy as they have been shown to be capable of differentiating into adipocytes, chondrocytes, osteoblasts, myocytes, cardiomyocytes, and neural precursors. MSCs have also been shown to suppress T cell proliferation in vitro and were reported to be effective as a treatment for acute graft-versus-host disease (GVHD) but the underlying molecular mechanisms for T cell suppression are uncertain. So far, TGF-β, HGF, and PGE2 were shown to be candidates as molecules causing the suppression. To address the molecular mechanisms, we used primary mouse MSCs derived from bone marrow cells and CFSE (carboxyfluorescein diacetate succinimidyl ester) or thymidine uptake for T cell proliferation assay. Co-culture of MSCs inhibited T cell proliferation induced by PMA plus Ionomycin, suggesting that TCR and signaling molecules interacting with TCR such as Lck and ZAP70 are not involved and that downstream signals of PMA plus Ionomycin are essential for the suppression by MSCs. The proliferation of either purified CD4 or CD8 cells induced by PMA plus Ionomycin was also inhibited by co-culture with MSCs, indicating MSCs suppression is active on both CD4 and CD8 cells. Stat5 phosphorylation in activated T cells was suppressed by co-culture with MSCs. Induction of cell-cycle promoting proteins such as CDK6, Cyclin D2, and Cyclin E by mitogenic stimulation were inhibited and suppression of a cell-cycle inhibitor, Kip1, was abolished. A previous report showed that T cells from stat5 deficient mice failed to induce cell-cycle promoting proteins and were not be able to proliferate on the stimulation through TCR. It was also reported that Nitric Oxide (NO) suppressed stat5 phosphorylation. Taken together with these reports, we hypothesized that NO is another candidate for the cause of suppression. In fact, NO synthase inhibitor (N-nitro-L-arginine methyl ester) recovered T cell proliferation from the suppression by MSCs in a dose-dependent manner. The amount of NO production and the strength of T cell suppression were parallel and dependent on the number of MSCs. MSCs blocked production of IFNγ but induction of T cell activation markers such as CD25 and CD69 and production of IL-2 were unaffected as reported. Our data suggest that MSCs block stat5 phosphorylation by production of NO, resulting in that T cells can neither proliferate nor produce high level of IFNγ. Here we demonstrate a new critical NO-stat5 dependent mechanism for T cell suppression by MSCs.

Author notes

Corresponding author