Daprodustat augments red cell and γ-globin production in models of sickle cell disease

Abstract

Sickle cell disease (SCD) causes erythrocyte sickling and downstream sequelae, including hemolysis, iron mishandling, and subclinical renal dysfunction, thereby reducing erythropoietic drive. These effects result in chronic anemia that drives SCD-related complications. Treatment of severe anemia in SCD with frequent subcutaneous injections of erythropoiesis-stimulating agents is often poorly tolerated by patients, underscoring the need for oral treatments. Here, we provide key insights into how daprodustat, an oral hypoxia-inducible factor prolyl hydroxylase inhibitor (HIF-PHI) that stimulates red cell production, may serve as a targeted treatment for anemia in SCD. In a mouse model of SCD, daprodustat increases red blood cell production, reduces sickling, and improves iron mobilization by suppressing hepatic hepcidin expression, thereby enhancing iron availability for erythropoiesis and ameliorating chronic anemia. In addition to these systemic, non-cell-autonomous effects on iron homeostasis, daprodustat directly induces γ-globin expression and increases the production of fetal hemoglobin-containing erythroid cells (F-cells) derived from CD34+ hematopoietic stem and progenitor cells of patients with SCD, conferring cell-autonomous therapeutic benefits. These findings demonstrate that HIF-PHIs act on multiple therapeutic targets in SCD and represent a promising oral treatment option for chronic anemia in this disease.

Introduction

Sickle cell disease (SCD) is caused by an abnormal hemoglobin (Hb) variant resulting from a point mutation in the β-globin gene. Deoxygenated sickled hemoglobin (HbS) polymerizes, causing erythrocyte sickling, chronic hemolytic anemia, and recurrent vaso-occlusive events (VOEs), leading to multiple organ system complications and decreased life expectancy.^1,2 SCD is also characterized by subclinical renal dysfunction causing inadequate erythropoietin (EPO) production³, chronic inflammation⁴, disrupted iron homeostasis⁵, and ineffective erythropoiesis⁶, limiting erythropoietic compensation in the bone marrow for the increased red blood cell (RBC) turnover. The resultant chronic anemia is a key driver of SCD-related complications^7,8. Erythropoiesis-stimulating agents (recombinant human EPO) have been used ad hoc to treat severe anemia but lack prospective data on efficacy and safety in SCD^9–11. Our recent Phase 1b/2 study of EPO for patients with SCD on stable-dose hydroxyurea (ACHiEvE-SCD; ClinicalTrials.gov NCT05451940) demonstrated significant improvements in Hb levels, six-minute walk distance, and key patient-related outcomes, including pain and fatigue, without increasing thrombosis or VOEs^12,13. However, the need for subcutaneous injection, often requiring transportation to a clinic setting for administration, limits EPO’s acceptability, accessibility, and feasibility in the SCD population, where medication uptake and adherence already present major challenges. Hence, oral therapeutics to improve erythropoietic response are urgently needed.

The current standard of care, hydroxyurea, inhibits HbS polymerization by inducing fetal hemoglobin (HbF) expression, preventing vaso-occlusive complications. However, hydroxyurea only modestly improves anemia¹⁴. Hypoxia-inducible factor prolyl hydroxylase inhibitors (HIF-PHIs) are a novel class of orally bioavailable drugs that stimulate endogenous EPO expression by stabilizing HIF1/2α¹⁵ and improve iron mobilization by suppressing hepcidin^16,17. This dual mechanism of action may yield a more robust erythropoietic response than EPO supplementation alone, especially in patients with high hepcidin levels¹⁸, though careful hemoglobin monitoring may be required to avoid vaso-occlusive or hyperviscosity-related complications¹⁹. Preclinical work with HIF-PHIs roxadustat and its prior generations (FG-2216) suggests that HIF-PHIs also increase HbF expression^15,20, offering an anti-sickling mechanism. Together, these mechanisms target multiple pathological processes contributing to anemia in SCD and address unmet needs in SCD treatment (Figure S1). Here, we demonstrate that daprodustat, an FDA-approved HIF-PHI, induces transcription of HIF targets, increases RBC production, reduces sickling, improves iron mobilization, and upregulates γ-globin expression in models of SCD.

Study design

Mice

Sickle cell disease mice (SCD, strain number: 013071) expressing human HbβS (SS) were procured from The Jackson Laboratory. All animal studies were approved by the Institutional Animal Care and Use Committee, University of Pittsburgh (protocol numbers: 22101947 and 202300066).

Preparation and treatment of daprodustat

Freshly prepared daprodustat was administered by oral gavage at a concentration of 30mg/kg body weight. 5.25 mg daprodustat was dissolved in 700μl of 0.9% NaCl using a sonicator and dosed at 100 μl/25g mouse weight every day for 4 weeks. NaCl was used as a vehicle control.

CD34⁺ HSPC Isolation, Differentiation, and Treatment

Control human CD34⁺ HSPCs were purchased from Fred Hutchinson Cancer Center. Sickle CD34⁺ HSPCs were isolated from deidentified, discarded peripheral blood from SCD donors undergoing RBC exchange transfusion at UPMC. Sickle CD34⁺ HSPCs were isolated using the CD34⁺ MACS Microbead kit (130-100-453, Miltenyi Biotec). Details on culture and differentiation of HSPCs are provided in the Supplemental Methods.

Cells were treated with vehicle (DMSO), 30μM daprodustat, or 50μM hydroxyurea starting on D6 and replenished every 3 days. RNA was isolated from cells on D12 of culture; F-cells were measured by flow cytometry on D15.

Blood, serum and transcriptional analyses

Blood, serum and mRNA were obtained by standard methods. Details are provided in the Supplemental Methods.

Statistical analysis

Results are presented as mean ± standard error of the mean (SEM). Data were analyzed using either Student’s t-test or one-way ANOVA to determine statistical significance between treatment groups and among time points. All p-values were calculated by Fisher LSD. Statistical significance was set at p < 0.05.

Results and Discussion

To assess the hematological effect of daprodustat in SCD, we used the Townes SS mouse model, a well-established SCD model engineered to express human sickle β-globin.²¹ 8-week-old SS mice were orally gavaged for 4 weeks with either 0.9% NaCl (vehicle) or daprodustat at 30mg/kg dosing. We evaluated hematological parameters before and 4 weeks after treatment (Figure 1). Sex-disaggregated baseline hematological parameters are shown in Figure S2.

Figure 1. Daprodustat increases red blood cell (RBC) count and ameliorates anemia from progressive sickle cell disease.

This is observed in an (A) increased RBC count, (B) arrested decrease in Hb, (C) increased hematocrit and (D) reduced reticulocytosis over the 4 weeks of daprodustat treatment. Other RBC indices including (E) MCV, (F) MCH, (G) MCHC and (H) RDW are unchanged, indicating no gross alterations in RBC formation. (I) Platelet and (J) WBC counts are unchanged, suggesting minimal toxicity. Markers of hemolysis, including (K) total plasma heme (TPH) and (L) LDH activity, remain unchanged with daprodustat treatment. n=5 per group; statistical significance (*) is set at P<0.05.

Daprodustat-treated mice had significantly increased RBC count and hematocrit after treatment, whereas untreated mice had significant decreases in RBC count, hemoglobin, and hematocrit, reflecting age-related progression of SCD²² (Figure 1A–C). At the same time, daprodustat decreased reticulocytosis (Figure 1D), suggesting a reduction in compensatory stress erythropoiesis and/or amelioration of hemolysis. We did not observe a decrease in other markers of hemolysis (total plasma heme and lactate dehydrogenase) after daprodustat treatment (Figure 1K,L). Additional hematological parameters, including MCV, MCH, MCHC, RDW, white blood cell count, and platelet count, were also unchanged by daprodustat treatment (Figure 1E–J). Similar hematological effects of increased hemoglobin, hematocrit, and RBC count and decreased reticulocyte percentage were observed in a transplant mouse model of SCD treated with daprodustat (Figure S3).

To confirm the mechanisms of action of daprodustat in SS mice, we measured serum levels of EPO, hepcidin, iron, and ERFE, as well as gene expression in erythropoietic and target organs (kidney, liver, bone marrow, and spleen). We did not observe an increase in serum EPO levels or renal Epo expression (Figure 2A,B). Hepatic Epo expression was significantly increased (Figure 2C), an effect specific to daprodustat relative to other HIF-PHIs²³. Daprodustat treatment decreased serum hepcidin, increased serum iron, decreased hepatic Hamp expression, and increased serum erythroferrone (ERFE; Figures 2D–G), though the latter was not significant (p=0.0625), consistent with regulation of iron mobilization by ERFE, a hormone secreted by erythroid progenitors to increase iron availability for erythropoiesis. However, FACS analysis of bone marrow and splenic erythroid progenitor populations revealed no alteration of erythroid maturation in daprodustat-treated mice (Figure S4).

Figure 2. Daprodustat improves iron mobilization and decreases sickling in vivo.

(A) Daprodustat does not alter overall concentrations of serum EPO. (B) Daprodustat does not significantly change renal Epo expression but increases (C) hepatic Epo expression. (D) Serum hepcidin and (E) expression of hepatic Hamp decrease in daprodustat-treated mice, and (F) serum ERFE trends towards an increase. (G) Serum iron increases in daprodustat-treated mice, reflecting iron mobilization. n=5 per group; statistical significance (*) is set at P<0.05. (H) Peripheral blood smears demonstrating that daprodustat treatment has an anti-sickling effect. These smears were taken at baseline and after 4 weeks of daprodustat treatment from the same mice (I and II are distinct mice).

To confirm HIF activation in target organs, we measured expression of other downstream HIF target genes and observed increased Bnip3 expression in the bone marrow and liver and increased Ldha expression in the bone marrow and kidney (Figure 3A–D). These results suggest that renal Epo expression is impaired and that iron mobilization via hepcidin suppression may be the dominant mechanism of action of daprodustat in the SCD mouse model.

Figure 3. Daprodustat increases expression of HIF transcriptional targets in the liver, kidney, and bone marrow, as well as BGLT3 and other HIF targets in human CD34⁺ HSPC derived erythroid cells, resulting in increased HBG expression and F-cell production in SCD patient-derived erythroid cells.

(A-D) Increased expression of hepatic Bnip (A), bone marrow Bnip (B), bone marrow Ldha (C), and renal Ldha (D) in daprodustat-treated Townes SS mice. (E-G) Increased expression of BGLT3 (E), LDHA (F), and BNIP3 (G) in HSPC-derived erythroid cells obtained from healthy donors. (H) HBA was unaltered, but (I) HBB and (J) HBG expression increased in control CD34⁺ erythroid cells treated with daprodustat. In HSPC-derived erythroid cells obtained from SCD patient donors, daprodustat did not increase HBA (K) and HBB (L) expression but significantly increased HBG expression (M) and F-cell production (N). (H-N) Hydroxyurea was used as a positive control for induction of HBG expression. n=5 per group for panels A-D; n=3 for panels E-J; n=4 for panels K-N. statistical significance (*) is set at P<0.05.

Finally, we assessed the anti-sickling potential of daprodustat, which we observed morphologically in the peripheral blood smears from the SS mice before and after daprodustat treatment (Figure 2H). γ-globin expression is controlled by HIF1 via its binding and transcriptional activation of the BGLT3 gene, which codes for a long noncoding RNA proximal to the γ-globin genes¹⁵. BGLT3 facilitates recruitment of transcription activator complexes, increasing interactions between the γ-globin genes and their upstream enhancer¹⁵ (Supplemental Figure 1). As mouse models lack BGLT3²⁴, we evaluated the effect of daprodustat on γ-globin expression in human CD34⁺ hematopoietic stem and progenitor cells (HSPCs) differentiated toward the erythroid lineage²⁵. CD34⁺ HSPCs from healthy controls and SCD patients were treated with daprodustat (5–25 μM) or 50 μM hydroxyurea 6 days after induction of differentiation. We analyzed transcriptional regulation of HIF target genes LDHA^26,27, BNIP3, and BGLT3^15,28 in control CD34⁺ HSPCs and found that daprodustat robustly activates their transcription in a dose-dependent manner, demonstrating daprodustat’s ability to activate multiple independent HIF pathways in this model system (Figure 3E–G). Using hydroxyurea treatment as a positive control for HbF induction, we evaluated the effect of daprodustat treatment on globin expression. Daprodustat increased expression of β- and γ-globin genes in control erythroid cells, but not α-globin (Figure 3H–J). In SCD erythroid cells, daprodustat showed specificity for increasing γ-globin expression (Figure 3M). Induction of HbF was confirmed by flow cytometry, which showed that daprodustat approximately doubled the percentage of F-cells (F-cell%) in SCD CD34+ HSCPs (Figure 3N).

Our studies demonstrate that daprodustat increases RBC production and decreases RBC sickling in a murine SCD model and increases γ-globin expression and F-cell% in an in vitro SCD erythroid cell model. Though daprodustat activated transcription of HIF-dependent genes in both SCD models, we observed no increase in serum EPO levels or renal Epo expression in the SCD mouse model, signaling impaired renal EPO production. Daprodustat had strong effects on the iron axis in our mouse model, suggesting the increase in RBC production may be primarily mediated by increased iron availability via control of serum hepcidin and hepatic Hamp expression, highlighting the critical role of non-cell-autonomous iron regulatory mechanisms in the pathophysiology of chronic anemia in SCD. These data demonstrate multiple potential therapeutic benefits of daprodustat for patients with SCD and provide a strong argument for further mechanistic and clinical studies of the HIF-PHI drug class as an oral treatment approach for chronic anemia in SCD.

Key points:

1. In mice, daprodustat increases red cell production, reduces sickling, and increases iron availability for erythropoiesis.

2. Daprodustat increases F-cell production in cultured SCD erythroid cells, demonstrating its potential as an oral therapy.

Data sharing statement:

For original data, please contact xujz2@upmc.edu, yieny@pitt.edu, and sag130@pitt.edu