Pharmacologic augmentation of fetal hemoglobin (HbF, γ-globin) production, to replace diminished β-globin chains in the β-thalassemias and to inhibit HbS polymerization in sickle cell disease, is a definitive therapeutic modality. Despite long-term efforts, regulatory approval has been obtained for only one chemotherapeutic agent. Pharmacologic reactivation of high-level HbF expression with non-cytotoxic, tolerable therapeutics is still an unmet medical need for this global health burden. To investigate potential therapeutic libraries for unrecognized HbF inducers, we developed a high-throughput screening (HTS) program to interrogate diverse chemical libraries, including a library of FDA-approved and clinical stage drugs. This program has identified unexpected new and highly potent HbF-inducing drugs, some of which are already in clinical use for other medical indications and have established safety profiles. A human cell-based assay which was previously used in low throughput assays, utilizing a 1.4-kilobase (kb) KpnI-BglII fragment of the HS2 of the locus control region (LCR) linked to the γ-globin gene promoter and the enhanced green fluorescent protein (EGFP) reporter gene, was adapted for high throughput screening and employed as the primary screen. Cytotoxic activity was assayed in a simultaneous counter screen. A number of hits were identified as being more potent than positive controls (such as butyrate). Several hits were immediately eliminated from further development as potential hemoglobinopathy therapeutics because of cytotoxicity (e.g., Idarubicin) or undesirable off-target effects, but nonetheless validated the HTS itself and were validated in secondary confirmatory assays as highly-potent HbF-inducers. The HTS assay identified 8 FDA-approved drugs as potent inducers of γ-globin gene expression, with activity at 1–2 logs lower concentrations (1000-fold higher potency) than prior generation therapeutic candidates. The γ-globin-specificity of hits was determined in a secondary assay employing a stably-transfected dual-luciferase reporter construct containing the LCR and the β-globin promoter linked to renilla luciferase and the Aγ-globin promoter linked to firefly luciferase (μLCRβprRlucAγprFluc cassette). Clinical-stage or clinically-approved agents, including Ambroxol at 1 μM, Desloratadine at 1 μM, Resveratrol at 10 μM, Benserazide at 5 μM, the HDAC inhibitor MS-275 at 5 μM, and an established bioactive, NSC-95397, at 1 μM were all significantly more active in this assay than Butyrate at 2000 μM, with MS-275 and Resveratrol being the most active. These drugs were then assayed for their ability to induce γ-globin mRNA expression in cultured primary human erythroid progenitors, at concentrations which are pharmacologically achievable in humans. Drugs significantly more active in γ -globin mRNA induction than the positive control (2-fold induction) in this system included Ambroxol (3-fold), Desloratadine (up to 6-fold), Resveratrol (up to 3-fold), Benserazide (up to 5-fold), and MS-275 (up to 3.7-fold). Two agents were subsequently studied in anemic baboons, and demonstrated in vivo induction of γ-globin mRNA, HbF, and F-reticulocytes. Unexpectedly, rises in total hemoglobin (>1 gm/dL) also occurred with 2 agents. Thus, a panel of structurally- and functionally-unrelated compounds demonstrate greater HbF-inducing activity, with up to 1000-fold higher potency, than current HbF-inducers which have significant activity in clinical trials. Some of the drugs identified by HTS have entirely benign safety profiles. These candidates could be clinically evaluated rapidly and at significantly less cost than new chemical entities, which require extensive toxicology, manufacturing, and clinical evaluation. These findings demonstrate the utility of a high-throughput screening program based on γ-globin gene promoter induction.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.