Salubrinal Mediated Fetal Hemoglobin Induction through the eIF2α-ATF4 Signaling Pathway

Sickle cell disease (SCD) is a genetic disorder caused by a mutation in the adult β-globin gene, affecting ~100,000 people in the United States and millions worldwide. Clinical symptoms of SCD include anemia, pain, and progressive organ damage creating a great burden to annual healthcare costs. An effective therapeutic intervention for SCD is fetal hemoglobin (HbF) induction by pharmacologic agents to ameliorate clinical symptoms. Hydroxyurea (HU) is the only FDA-approved drug used to induce HbF in SCD, however, it is not effective in all people. Therefore, the goal of this study is to determine the ability of Salubrinal (SAL), to induce HbF. Salubrinal is a selective inhibitor of protein phosphatase 1 leading to increased levels of p-eIF2α (phosphorylated eukaryotic initiation factor 2α) and inhibition of global protein synthesis. Activating transcription factor 4 (ATF4) is a downstream target of p-eIF2α activated during oxidative stress. The main function of these signaling events is to attenuate stress to the endoplasmic reticulum. Previously we identified a Gγ-globin cAMP response element (G-CRE) that binds ATF2, a binding partner of ATF4 involved in HbF induction (Sangerman J et al. Blood 2006). Furthermore, ENCODE analysis showed ATF4 sites at -822Gγ and β-globin gene second intron. Thus, studies were performed to determine if p-eIF2α-ATF4 signaling is involved in mechanisms of HbF induction by SAL.

Initial experiments involved the use of day 8 erythroid progenitors generated from human CD34⁺ stem cells; treatments included SAL 12, 18 and 24 µM, and 0.5% DMSO (vehicle control) for 48 h; cell viability remained >90% in all drug conditions. The level of γ-globin mRNA increased 1.2-fold and 1.3-fold at SAL 18 and 24 µM respectively (p<0.05). Comparable, HbF was induced by SAL 24 µM alone and combined SAL/HU treatments to 1.8-fold. To gain insights into mechanisms of HbF induction by SAL, we next quantified the level of p-eIF2α. We observed a 1.7-fold increase in p-eIF2α with SAL 12 and 24 µM and parallel increase in ATF4 (4.8-fold). Flow cytometry revealed SAL increased F-cells (HbF positive cells) from 30.9% (DMSO treated) to 90.6%. Similarly, studies were performed using sickle erythroid progenitors generated from peripheral blood mononuclear cells. On day 8, SAL (9, 18, 24 µM) dissolved in water was added for 48 h; cell viability remained >90% for all drug conditions. SAL (18 μM) increased γ-globin mRNA 3.2-fold and F-cells 2.5-fold (p<0.05) compared to the untreated control. We used mean fluorescence intensity (MFI) to quantify HbF protein per cells, which showed a dose-dependent increase with SAL treatment. Since sickle red blood cells are under oxidative stress, we measured the levels of reactive oxygen species (ROS) by flow cytometry. SAL 12, 18, 24 µM decreased ROS levels in a dose-dependent manner by 7.6%, 8.7% and 10% respectively. Interestingly, SAL/HU treatment decreased ROS levels by 10.2% compared to a 4.3% mediated by the nitric oxide donor HU. Western blot analysis showed a dose-dependent increase in HbF and a 3.3-fold increase in p-eIF2α (p<0.05) and ATF4, without changing HbS expression.

To generate data for clinical development, we utilized the Townes SCD mouse model. SCD mice (n=5 per group) were treated with SAL (3 and 5mg/kg), HU (100mg/kg; positive control) or water control (vehicle), 5 days a week for 4 weeks. Blood was drawn at week 0 (baseline), 2 and 4 at treatment completion. All data were normalized for each group and treatment response at week 2 and 4 compared to week 0 using a paired t-test and ANOVA to compare across treatment groups; statistical significance set at p<0.05. All groups showed normal weight gain and no significant changes in complete blood counts, differential or reticulocyte counts. Flow cytometry of peripheral blood showed that SAL (3mg/kg) produced a 2-fold increase in F-cells by 2 and 4 weeks while SAL (5mg/kg) produced a further 3.1-fold increase in F-cells by week 4 (p<0.05) without toxicity.

Our initial in vitro findings, supports HbF induction by SAL involving p-eIF2α-ATF4 signaling. The interaction of ATF4 in the G-CRE and/or other predicted binding sites will be investigated. To support clinical trials, studies in the SCD preclinical model support the ability of SAL to induce HbF in vivo; additional studies are underway. Defining the mechanism of HbF induction by SAL has the potential to impact treatment for SCD.