Liposomal KRN7000 (RGI-2001) Synergizes with Low-Dose Rapamycin in Preventing GvHD by promoting expansion of naturally-Occurring CD4⁺FOXP3⁺ Regulatory T Cells.

RGI-2001 is a liposomal formulation of KRN7000 (a synthetic derivative of a-galactosylceramide). We previously demonstrated that RGI-2001 potently reduces GvHD lethality in a mouse model by inducing host-specific tolerance through the expansion of donor-derived CD4⁺Foxp3⁺ Regulatory T cells (Tregs) (ASH 2008 Poster III-582). In the present study, we examined if the combination with low-dose rapamycin (LD-RAPA) may further improve the efficacy of RGI-2001. In addition, the origin of Tregs induced by RGI-2001 was investigated.

A fully MHC mismatched GvHD mouse model using lethally-irradiated Balb/c (H-2d) recipients transplanted with C57BL/6 (H-2b) T cell depleted bone marrow cells with total spleen cells were used. In some experiments, Foxp3-GFP reporter mice (C57BL/6 background) were used as the donor. RGI-2001 (0.01 ug/kg to 1 ug/kg) was given on day 0 via single intravenous administration. LD-RAPA (0.1 mg/kg) was given daily from day 0 to day 14 intraperitoneally. Treg frequency was determined on day 15 post-BMT by flow cytometry either by using an antibody for Foxp3 or by detecting GFP as a surrogate marker for Foxp3.

First, the pharmacological dose induction of Tregs by RGI-2001 was investigated. Day 15 post-BMT analysis of mice treated with RGI-2001 demonstrated a dose-dependent increase in the % CD4⁺Foxp3⁺ Treg cells in the spleen, lymph nodes, and bone marrow. As shown in the Table, frequencies of CD4⁺Foxp3⁺ Tregs in the spleen increased in a dose dependent manner, from 3.0 ± 1.4 % in untreated mice to 14.9 ± 5.8% in 1 ug/kg RGI-2001 treated mice. The total Treg cell counts in the spleen also increased in a dose dependent manner with 1 ug/kg RGI-2001 consisting of 4.2×10e5 Tregs as compared with 0.3×10e5 Tregs in untreated mice. The frequency of CD8⁺ T cell population decreased in a dose dependent manner, suggesting an inverse relationship with frequencies of Tregs. When LD-RAPA was combined, a robust synergy between the two compounds was noted. The frequency of Treg in mice that received combination treatment was significantly higher than that in each monotherapy group (p = .0005 vs. RGI-2001 only, p< .0001 vs. LD-RAPA only). Absolute Treg counts in the spleen were markedly increased in mice received the combination treatment. In three independent experiments, Tregs in combination treatment group were consistently ∼3-fold higher than that in RGI-2001 monotherapy group and ∼9-fold higher than that in LD-RAPA monotherapy group. In contrast, CD8⁺ T cell frequencies decreased in mice treated with LD-RAPA containing regimen. Efficacy in survival was also improved in mice that received the combination treatment with 80-100% survival over 100 days.

Next, the origin of the Tregs induced by RGI-2001 was investigated using the Foxp3-GFP reporter mice as the spleen cell donor. While 2.5×10e6 total spleen cells from Foxp3-GFP mice gave rise to 7.2 ± 2.4 % Tregs (n= 7) when treated with RGI-2001 (1 ug/kg), equivalent number of GFP⁺ Treg-depleted spleen cells gave rise to 1.4 ± 0.3 % Tregs (n=5, p < .0006). Likewise, when mice were treated with the combination of RGI-2001 (1 ug/kg) and LD-RAPA, Treg frequencies in mice that received total spleen cells and GFP⁺ Treg -depleted spleen cells were 17.8 ± 3.8 % (n=7) and 1.5 ± 0.4 % (n=7), respectively (p < .0001). These results together suggest that the majority of Tregs induced by RGI-2001 with or without LD-RAPA were derived from the pre-existing naturally-occurring Treg population, and the contribution of de-novo Tregs was minimal in the GvHD setting.

RGI-2001 prevented GvHD by promoting the expansion of naturally-occurring Tregs. LD-RAPA synergistically enhanced the efficacy of RGI-2001 by significantly increasing Treg expansion while suppressing effector T cells.

REGiMMUNE employees (OD, AL, JL, TT), Founder (YI), and consultant (RN). YI and RN are stockholders.

No relevant conflicts of interest to declare.