NKT Cells Are Potent Regulators of GVHD Following Adoptive Transfer in Allogeneic BMT.

NKT cells, which are CD1d reactive and express an invariant TCR, are thought to play an immunoregulatory role in suppressing dysfunctional immune reactions including graft vs. host disease (GVHD). Whether non-manipulated donor-type NKT can suppress GVHD following adoptive transfer has not been addressed, nor has the trafficking pattern of NKT in a hematopoetic transplantation (HCT) setting. To determine how effectively NKT proliferate and traffic in a HCT setting, 5.5x10⁵ highly purified (>95%) NKT (DX5⁺TCRβ⁺CD4⁺) from luciferase positive (luc⁺) C57BL/6 (H-2b) mice were transferred into lethally irradiated Balb/c (H-2d) recipients along with 5x10⁶ T-cell depleted bone marrow (TCD-BM) from wild-type C57BL/6 mice. Proliferation and migration of luc⁺NKT was monitored by bioluminescence imaging (BLI). By day 4 after transfer, a prominent signal was observed in the spleen and lymph node (LN) sites. Between days 7 and 10, the NKT migrated to skin, while still remaining present in high numbers in LNs, but decreasing to low levels in spleen. The total photons emitted per mouse reached a peak at approximately 25 days after transplantation, followed by a steady decline. The NKT expansion was more vigorous than that of luc⁺CD4⁺CD25⁺ regulatory T cells (Tregs), which also peak around day 25, but do not show extensive migration to skin. Expansion of NKT was far less than conventional (CD4⁺ and CD8⁺) T cells (Tcon), with approximately 10 times more photons/mouse being emitted from 5x10⁵ luc⁺Tcon as from 5.5x10⁵ NKT. The NKT did not cause GVHD where Tcon rapidly resulted in acute GVHD and animal mortality. To assess the impact of donor-type NKT on GVHD induction by Tcon, we co-transferred various doses of highly purified wt NKT at day 0 with 5x10⁶ TCD-BM, followed by 5x10⁵ luc⁺Tcon at day 2. Weight and survival of groups were monitored, as well as the proliferation of Tcon by BLI. In groups receiving only Tcon, 50% of the mice died within 2 weeks, and 90% died by day 80. Remarkably, if 2.5x10⁴ sorted NKT were transferred, 100% of the mice survived past day 100. To achieve the same effect with Tregs, 2.5x10⁵ Tregs were required. Mice treated with 2.5x10⁴ NKT lost more weight at early time points than those receiving 2.5x10⁵ Tregs, but both groups recovered from this weight loss and did not exhibit other signs of GVHD (hair loss, hunched back, diarrhea, etc). Interestingly, 2.5x10⁴ NKT caused only a slight reduction in Tcon proliferation, whereas 2.5x10⁵ Treg strongly reduced Tcon proliferation. Surprisingly, when the dose of NKT was increased to 5x10⁴, survival was only 62%, and when increased to 1x10⁵ cells, only 50% of mice survived past day 100. To determine how NKT reduce GVHD, we examined intracellular levels of various cytokines in Tcon with or without 2.5x10⁴ NKT, following HCT. At 8 days after HCT, mice receiving NKT had reductions in the number of IL-17-positive cells (CD4: 2.6% to 0.84%; CD8: 2.5% to 0.66%), and TNFα⁺ cells (CD4: 45% to 27%; CD8 36% to 24%) in cells from LNs. By day 11, IL-17-positive cells had declined to undetectable levels and TNF levels between groups were equivalent. IFNg levels, which were high in both NKT treated and untreated groups at day 8 (85%–95%), decreased significantly in NKT treated mice by day 11 (CD4: 40%; CD8: 43%), but were still abundant in Tcon only mice (CD4: 78%; CD8: 80%). Together these data indicate that NKT cells persist in vivo upon adoptive transfer and suppress GVHD by decreasing the percentage of Tcon secreting pro-inflammatory cytokines.