Human CD4 Negative NKT Cells Potently Suppress Alloreactive Responses in a TGF-β and IL-10-Independent Manner.

Acute graft versus host disease (aGVHD) remains a major source of morbidity and mortality in allogeneic stem cell transplantation. There is a need for therapeutic interventions to ameliorate its severity. NKT cells, a novel class of regulatory T cells, are restricted by the glycolipid presenting MHC-like molecule CD1d. They constitute <0.1% of T cells and modulate responses of the innate and adaptive immune system in autoreactivity and against pathogens and tumours through interactions with antigen presenting cells (APC). The regulation of alloreactivity and aGVHD by NKT cells has not been studied. We tested the effect of expanded human NKT cells on the mixed lymphocyte reaction (MLR). First, NKT cells were expanded ex vivo from peripheral blood mononuclear cells (PBMC) of normal donors in the presence of the CD1d-presented glycolipid α-galactosylceramide. Under these conditions NKT cells (n=7), identified by staining with mAbs specific to their TCR α and β chains, expanded by 238-fold±102 by day 10. Total, CD4+ (69%±17 of total) and CD4- (30%±17 of total) NKT were purified to >95% and tested for their effect on the MLR using ³H incorporation assays. Total and NKT subsets from a panel of donors were added to the MLR at a ratio of NKT:Responder (consisting of autologous T cells) of 1:10 and a panel of allogeneic irradiated PBMC were used as stimulators. Co-culture with total or CD4+ NKT had a variable effect on the MLR. By contrast, in all MLR tested using 7 different responders against a panel of 3–5 stimulators, the presence of CD4- NKT had a profound suppressive effect (mean suppression from baseline MLR: 67%±18%). Titration of the NKT numbers revealed that the magnitude of the suppressive effect was inversely proportional to the NKT:R ratio, suggesting that NKT can exert opposing effects of differential magnitude and the maximum suppression depends on an optimal NKT:R ratio. In addition we demonstrated that CD4- NKT cell lines exert linked suppression, i.e., that they have suppressive effect (mean 52%±16.5, n=2) in MLR when the responding T cells are allogeneic to the NKT themselves. In transwell experiments we determined that the suppressive effect is at least partially contact-dependent and it is not mediated by either IL-10 or TGFβ as these anti-proliferative cytokines were not detected in the supernatants of the MLR and the presence of the corresponding neutralising mAbs did not reverse the suppressive effect. These findings indicate that the suppressive effect of NKT is not due to induction of anergy on the alloresponsive T cells. Instead, it is likely that it is mediated by a modulatory effect on the APC. The mechanisms of this modulation are the subject of current investigation. Finally, all CD4- NKT cells lines tested (n=4) had markers of activation (>90% were CD25⁺ CD69⁺), demonstrated potential for homing to peripheral tissues (mean expression 52%±23.9 were CD62L⁻CCR7⁻), especially to mucosal tissues (mean expression of integrin β7 84%±8) but not to the skin (mean expression of CLA <1%). In summary, we have identified a subset of NKT cells that when expanded ex vivo are very efficient in suppressing the alloreactive response. These findings form the basis for the use of such cells at a clinical level for the treatment of aGVHD, especially of mucosal localisation, through their adoptive transfer. The procedure of expanding suppressive NKT can be easily adapted for clinical purposes as using autologous PBMC as APC can result in a >200-fold expansion after just one round of stimulation.