Enantiomer-Specific Transformation by 2HG Is Linked to Opposing Effects on α-Ketoglutarate-Dependent Dioxygenases

LBA-4 Somatic mutations in IDH1 and IDH2 occur frequently in clonal myeloid disorders and result in the neomorphic ability of IDH to convert α-ketoglutarate (2-OG) to the R-enantiomer of 2-hydroxyglutarate (R-2HG) (Dang, et al Nature 462: 739, 2009). 2OG is an essential cofactor for many metabolic enzymes, including the TET family of 5-methylcytosine hydroxylases and the EglN family of prolyl-4-hydroxylases, and 2HG has been shown to inhibit several 2OG-dependent dioxygenases in vitro, including TET2 (Xu, et al Cancer Cell 19: 17, 2011; Figueroa, et al Cancer Cell 18: 1, 2010). We recently showed that the (S) enantiomer of 2HG (S-2HG), but not the (R) enantiomer of 2HG (R-2HG), inhibits the EglN prolyl-4-hydroxylases (Koivunen, et al. Submitted for publication). Moreover, we found that R-2HG can act as a cofactor to promote the hydroxylase activity of EglN1, EglN2 and EglN3. We hypothesized that the qualitatively different effects of R- and S-2HG on the EglN prolyl-4-hydroxylases might influence their transforming activities. In order to elucidate the role of mutant IDH, and R- and S-2HG, in myeloid leukemia, we developed a myeloid transformation assay using TF-1 cells. TF-1 is a human erythroleukemia cell line that requires GM-CSF for growth and undergoes erythrocytic differentiation when stimulated with erythropoietin (EPO). We expressed wild-type IDH1 (WTIDH1), a tumor-derived mutant IDH1 (IDH1R132H), or a catalytically inactive IDH1R132H variant (IDH1R132H/3DN) in TF-1 cells. As expected, cells expressing IDH1R132H, but not cells expressing WTIDH1 or IDH1R132H/3DN, had dramatically elevated levels of 2HG. Furthermore, we found that expression of IDH1R132H, but not WTIDH1 or IDH1R132H/3DN, conferred growth factor-independence to TF-1 cells (Figure 1a), and blocked their EPO-induced differentiation (Figure 1b). In order to determine whether transformation of TF-1 cells by IDH1R132H is mediated by 2HG, we treated TF-1 cells with cell-permeable esterified R-2HG or S-2HG. R-2HG recapitulated the growth and differentiation phenotypes of IDH1R132H expression in a dose-dependent manner. In contrast, S-2HG did not induce these phenotypes at any concentration tested. Next, we examined the effect of loss of TET2 on TF-1 cells. We infected TF-1 cells with shRNAs targeting TET1 or TET2 and found that knockdown of TET2, but not TET1, induced growth factor-independence and blocked EPO-induced differentiation similarly to expression of IDH1R132H or treatment with R-2HG. Interestingly, we found that transformation by IDH1R132H and TET2 knockdown were reversed by inhibition of EglN1 (Figure 2), suggesting that R-2HG, but not S-2HG, transforms leukemic cells by inhibiting targets such as TET2 while preserving, and possibly enhancing, EglN activity. These findings further suggest that therapeutic targeting of EglN prolyl-4-hydroxylase activity might be effective in the treatment of IDH1-mutant and TET2-mutant myeloid leukemias. Disclosures: Kaelin: Fibrogen: Consultancy, Equity Ownership.