Lack of IL-21 Signal Attenuates Graft-Versus-Leukemia Effect In the Absence of CD8 T-Cells.
Abstract Abstract 3739 Introduction: IL-21 is a pleiotropic cytokine belongs to a common cytokine receptor g chain family of cytokines. IL-21 is mainly produced by activated CD4+ T cells and acts on T-, B-, NK-cells, and other lineages. IL-21 can drive Th17 differentiation and contributes to the development of autoimmune disease. Previously, we have shown that IL-21R−/− (KO) splenocytes ameliorate GVHD as compared to wild type (WT) splenocytes (BMT 2010), indicating a critical role of IL-21 in GVHD. Bucher et al. reported that IL-21 neutralization resulted in the same results as ours and it did not attenuate GVL effect (Blood 2009). However, they did not titrate the required T-cells for GVL effect, making it impossible to determine if the GVL strength was similar in the normal and the IL-21 neutralized conditions. Here, we sought to quantify and compare the strength of GVL effect between WT and KO splenocytes (SP), and moreover, analyze the contributions of CD4 and CD8 cells to GVL effect. Methods: GVL experiments were performed by co-transplantation of P815 leukemic cell line, T-cell depleted wild type bone marrow cells (TCD-BM), and either KO-SP vs. WT-SP. C57BL/6-DBA F1 strain mice were used as recipients. To make the leukemia visible in alive mice, we used luciferase transduced P815 cell line and IVIS® imaging system. Results: Previously (ASH 2008), we have shown that IL-21R−/− splenocytes (KO-SP) retained GVL effect and that IL-21 decoy receptor treatment retained GVL effect. To confirm previous results, here we decided to perform dose-reduction experiment to determined the number of splenocytes required to eliminate leukemic cells after transplantation, as otherwise it is impossible to compare the strength of GVL effect between the WT and KO cells. We first started with 1 × 107 splenocytes. Without co-infusion of splenocytes, control mice died around 30–40 days after transplantation with a marked increase of luciferase activity, whereas recipients of both WT-SP and KO-SP demonstrated an eradication of P815 leukemic cells. In addition, more mice died in recipients of WT because of more severe GVHD. Secondly, we used 5 ×106 splenocytes; with this number, almost no mice died due to GVHD anymore, but still graft eradicated leukemic cells completely. Thirdly, we used 5 × 105 and 5 × 104; with these numbers, graft could not eliminate leukemic cells anymore and mice died due to leukemia. Taken together, the threshold to eradicate P815 cells in our experimental conditions was between 5 × 105 and 5 × 106 of splenocytes, regardless of genotype, WT or KO. In the experiments above with bulk splenocyte, GVL effect with KO-SP was almost comparable to that with WT-SP even throughout in the titration. However, our previous experiments demonstrated that IL-21R−/− CD4+ T-cells are defective in GVH reaction after transplantation (J Immunol. 2010). It was therefore of great interest to determine whether IL-21R−/− CD4+ T-cells are also defective in GVL effect. To evaluate the contribution of CD4+ T-cells to GVL effect, we performed transplantation with CD8-depleted splenocytes with dose-reduction as above. Because CD8 T-cells compose of only ≂f10% of splenocyte, we chose CD8-depletion rather than CD4 purification, so we could use the same dose of CD8-depleted splenocytes as in the case of bulk splenocytes. CD8-depleted KO-SP at the dose of 5 × 107 demonstrated attenuated GVHD and prolonged survival, consistent with our previous results. However, interestingly, CD8-depleted KO-SP at the dose of 5 × 106 demonstrated diminished GVL effect. Higher luciferase activity and more deaths in the KO-SP group indicated leukemic deaths. According to luciferase activity, with CD8-depleted KO-SP at the dose of 5 × 106 cells, 10 out of 21 recipients showed leukemic growth at day 21, whereas for CD8-depleted WT-SP, only 3 out of 21 mice showed leukemic growth at day 21; only 7 out of 21 mice survived in recipients of KO-SP but 13 out of 21 mice survived in recipients of WT-SP at day 100 after transplantation. Conclusion: Here we demonstrated that IL-21R−/− splenocytes (KO-SP), which ameliorate GVHD, do not attenuate GVL effect even in the splenocyte-dose-titration. In the further detailed analysis with CD8-depleted splenocytes, GVL effect with IL-21R−/− CD8-depleted splenocytes was significantly diminished compared to that with wild type, suggesting that IL-21R−/− CD4 cells have lower GVL activity than wild type cells. Disclosures: Ozawa: Nippon Shinyaku Co., Ltd.: Research Funding.