Increased levels of fetal hemoglobin (HbF) decrease symptoms and increase life expectancy in Sickle Cell Disease (SCD). DNA Methyltransferase 1 (DNMT1) and Lysine Specific Demethylase-1, (LSD-1), a mono- and dimethyl-histone H3 K4 demethylating enzyme, are components of the DRED multiprotein complex, a repressor of γ-globin expression (Cui et al Mol Cell Biol 31:3298, 2011). While DNMT1 inhibitors are well known to induce high levels of HbF in non-human primates and SCD patients, it was recently shown that the LSD1 inhibitor tranylcypromine (TC) increased γ-globin expression in cultured human erythroid progenitors and human βYAC mice (Shi et al Nat Med 19:291, 2013). We tested the effect of TC in non-human primates by administering varying doses of TC (2-6mg/kg/d; 10d; sc; n=4) to anemic baboons. Small increases in HbF were observed at the highest TC dose (ΔHbF=4%; n=2) that were 10-15 fold less than in baboons treated with decitabine (DAC; 0.5mg/kg/d; 10d). To identify LSD1 inhibitors with increased potency that may be more powerful in vivo inducers of HbF, a screening assay for LSD1 activity was developed based on the recent report that TC in combination with low-dose ATRA induced differentiation of AML cells (Shenk et al Nat Med 18:605, 2012). The U9367 cell line was treated with various LSD1 inhibitors in combination with low-dose ATRA. Effects on differentiation were assessed by measurement of the differentiation marker CD11b by flow cytometry. The TC derivative RN-1 induced differentiation with a 64 fold increased potency compared to TC. Experiments were then conducted to compare the effects of TC and RN-1 on γ-globin expression in cultured baboon erythroid progenitors. Cultures (n=2) were treated with varying doses of either RN-1 or TC on d7 and d10 and globin chain expression measured by HPLC. Low level γ-globin expression (0.021, 0.048 γ/γ+β) was observed in controls. TC (5μM) increased γ-globin expression (0.32 γ/γ+β) in the absence of cytotoxicity while similar levels of γ-globin (0.24, 0.36 γ/γ+β) were induced by non-cytotoxic doses of RN-1 that were 20-70 fold lower (0.25, 0.07μM). We next compared the effects of RN-1, TC, and DAC on HbF induction in a knock-in humanized mouse model. Humanized B6;129-Hbatm1(HBA)TowHbbtm3(HBG1,HBB)Tow/J mice were treated with phenylhydrazine for 2 days followed by varying doses of DAC, TC or RN-1 for 3 days. Analysis of F cells and γ-globin mRNA performed on d8 and 10 showed that F cells were increased 2-3 fold (p<0.002) in mice treated with 10mg/kg RN-1 (9.70+1.96%) and 20mg/kg RN-1 (9.55+1.20%) compared to DMSO-treated controls (3.56+1.70%) while γ-globin mRNA levels (γ/γ+β) were increased 5-8 fold in RN-1 treated mice (p<0.002). Reticulocyte counts did not differ significantly between control and RN-1-treated mice (10mg/kg). Higher doses of RN-1 (30mg/kg) induced greater increases in γ-globin mRNA (14 fold; p<0.023) but were associated with decreased reticulocyte counts. In contrast, no changes F cells or γ-globin mRNA were observed in mice treated with TC (6mg/kg/ip; n=3 and 10mg/kg/ip; n=3). DAC treatment resulted in dose-dependent increases in F cells and γ-globin mRNA. F cells were increased 2-3 fold (p<0.01) and γ-globin mRNA (γ/γ+β) 7 fold (p<0.004) in mice treated with 0.25mg/kg DAC in the absence of significant cytotoxicity while higher doses of DAC were associated with decreased reticulocyte counts. Therefore, in this model, doses of either RN-1 (10mg/kg) or DAC (0.25mg/kg) that are non-toxic induce similar increases in F cells and γ-globin mRNA. Experiments were then performed to test the ability of RN-1 to increase F cells and γ-globin mRNA in the sickle cell mouse B6;129-Hbatm1(HBA)TowHbbtm2(HBG1,HBB*)Tow/J model. Mice were injected for four days with RN-1 (10mg/kg). F cells in RN-1 treated mice were increased 2.6 fold (p<0.0001) compared to controls while γ-globin mRNA levels (γ/γ+β) were increased 7.3 fold (p<0.004). These results demonstrate that RN-1, a recently developed LSD-1 inhibitor with increased potency and selectivity compared to TC, increases γ-globin expression in 1) cultured baboon erythroid progenitors, 2) a phenylhydrazine-treated humanized transgenic mouse model, and 3) sickle cell mice. We conclude that further studies to determine the effect of RN-1 in the non-human primate model be performed to evaluate its potential use in the treatment of sickle cell disease.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution