Pharmacologic Inhibition of XPO1 By Selinexor Improves Late-Stage Erythropoiesis in Severely Affected β0-Thalassemia/Hemoglobin E

Ineffective erythropoiesis due to accumulation of excess α-globin chains profounds unwarranted consequences in β-thalassemia. In severe β-thalassemia, the excess α-globin chains trap HSP70—the guardian of GATA1—within the cytoplasm. Thus, in the nucleus, GATA1 is degraded and leads to late-stage erythropoiesis arrest.

XPO1 is a nuclear exporter that exports HSP70 from the nucleus to the cytoplasm. Previous study showed that chemical downregulation of XPO1 in β-thalassemia major using an old generation of XPO1 inhibitor (KPT-251) could improve late-stage in vitro erythropoiesis. Moreover, XPO1 inhibitor treatments in clinical trials revealed improvement of transfusion burden and hemoglobin levels in some patients with myelofibrosis. Together suggested that inhibition of XPO1 might be considered as a possible option for β-thalassemia treatment.

Here, we investigated the effects of pharmacological XPO1 inhibition using old and new generation of XPO1 inhibitors, and natural products in human CD34⁺ hematopoietic stem and progenitor cells (HSPCs) derived from β⁰-thalassemia/HbE with severe phenotypes. Coinheritance of common α-thalassemia were ruled out in all patients and severity score was computed using Sripichai scoring system (Sripichai et al. Am J Hematol 2008). HSPCs were purified from peripheral blood and subjected to 3-phase erythroid culture system. Nine compounds with known XPO1 inhibitory effect including five synthetic (KPT-185, KPT-276, KPT-330 or selinexor, KPT-335 or verdinexor and KPT-8602 or eltanexor) and four natural active compounds (curcumin, oridonin, piperlongumine and plumbagin) were evaluated in patient-derived HSPCs (n=1) for their potency and efficacy. Of these, selinexor exhibited the maximal efficacy in which the accumulation of CD71^Low/GPA^High cell population was predominate. To affirm this finding, HSPCs derived from six independent β⁰-thalassemia/HbE patients were treated with selinexor (5 and 10 nM) at early-stage of the in vitro erythropoiesis and maintained throughout the culture system. After treatments (n=3), XPO1 was downregulated in both cytoplasmic (5 nM; 1.5±0.1-fold, p < 0.005 and 10 nM; 1.7±0.5-fold, p < 0.05) and nuclear compartments (5 nM; 1.1±0.3-fold and 10 nM; 1.3±0.2-fold, p < 0.05). No significant cytotoxic effects were observed. In addition, effects of selinexor on erythroid differentiation and maturation were assessed by flow cytometry using CD71/GPA immunostaining and Wright-Giemsa staining. The results revealed that selinexor treatments could augment late-stage in vitro erythropoiesis (orthochromatic+enucleated/basophilic+polychromatic erythroid cell) in dose-dependent manner (5 nM; 1.4±0.4-fold and 10 nM; 3.5±1.8-fold, p < 0.05). This finding was correlated with erythroid cell morphology under a light microscope. We further explored whether XPO1 inhibition by selinexor (n=3) affects HSP70 and GATA1 expression. Homogenous erythroid cell population (CD71^High/GPA^High) were collected and subjected for western blot analysis. We found that HSP70 was upregulated in the cytoplasm (5 nM; 1.8±1.0-fold and 10 nM; 2.1±1.1-fold). However, HSP70 and GATA1 in the nucleus were insignificant changed. Our finding demonstrated that manipulations of XPO1-HSP70 axis in the cytoplasm by selinexor were associated with accelerated late-stage erythropoiesis in severely affected β⁰-thalassemia/HbE HSPCs. We then hypothesized whether selinexor functioned through an accumulation of HbF. Therefore, Hb profiles were investigated using HPLC. However, non-significant changes of Hb profiles and HbF levels were observed. With this finding, we purposed that selinexor functioned independently of HbF regulatory pathways and precise mechanisms underlying the regulation of late-erythropoiesis by selinexor are required.

According to our preliminary data, an FDA approval compounds for multiple myeloma—selinexor—is a potent compound for improving in vitro late-stage erythropoiesis in HSPCs derived from severe β⁰-thalassemia/HbE. Therefore, repurposing of selinexor to improve ineffective erythropoiesis in β-thalassemia is compelling. Moreover, combination treatments of selinexor with HbF inducers and/or SMAD2/3 blockers could be further validated to possibly achieve greater therapeutic benefits for β-thalassemia treatments.