Decitabine-Driven Foetal Haemoglobin Induction in Townes Mice and Human Erythroblasts

Abstract

Background: Induction of foetal haemoglobin (HbF) is a clinically validated approach to modulate the severity of sickle cell disease (SCD). This manuscript evaluates the efficacy of decitabine, a DNA methyltransferase (DNMT) inhibitor, in inducing HbF in healthy human erythroblasts and Townes mice, which are well-established systems modelling SCD.

Methods: Healthy human erythroblasts were treated with decitabine, and HbF induction was measured. Townes sickle cell mice were administered decitabine for 12 weeks, and various haematological parameters were assessed.

Results: In healthy human erythroblasts, decitabine treatment resulted in a significant increase in the fraction of HbF-rich cells (F-cells), accompanied by elevated HbF protein levels. The HbF induction was superior to that achieved with hydroxyurea, the primary therapy for SCD. In Townes mice, the maximal response was observed after 12 weeks of dosing, with an increase in both HbF protein and F-cells, alongside reduced red blood cell and reticulocyte counts. Additionally, we observed changes in other haematological parameters, such as increased mean corpuscular volume and mean corpuscular haemoglobin. However, the HbF induction observed in the mice was modest relative to known human responses. No marked improvements in SCD-related biomarkers such as haemolysis or liver function were detected, suggesting that the mouse model may not fully capture the extent of phenotype improvement. Histopathological examination revealed no adverse effects on bone marrow cellularity or morphology and indicated a protective effect on liver tissue integrity.

Conclusion: Our results demonstrate that decitabine induces HbF in a dose-dependent manner in both in vitro and in vivo settings, highlighting the complexity of HbF induction as a treatment for SCD and underscoring the need for further refinement of this model for SCD therapy research. Trial Registration: The authors have confirmed clinical trial registration is not needed for this submission.

Keywords: haemoglobin disorders; haemoglobinopathies; haemolysis; red blood cell disorders; red cell disorders; sickle cell anaemia; sickle cell disease.

FIGURE 1

In vitro foetal haemoglobin (HbF) induction with decitabine. (A) Representative dose‐response curves of decitabine and hydroxyurea in two healthy donors. Data points for 3 µM decitabine and 30 µM hydroxyurea were excluded due to high cytotoxicity. Mean ± SD of technical triplicates are shown. (B) Baseline F‐cells, maximum F‐cells at 1 µM decitabine, and EC50 (half maximal effective concentration) of four healthy donors. (C) HbF protein levels by HPLC of two donors. (D) Immunoblot showing the expression of DNMT1 and BCL11A in erythroblasts at 48 h after decitabine treatment.

FIGURE 2

In vivo study design, exposure and growth. (A) Schematic of the in vivo experimental design depicting the treatment groups, dosing regimen, and timeline of the study. Townes SCD mice were divided into three treatment groups: vehicle (control), decitabine 0.4, and 0.6 mg/kg. Each group consisted of three male and three female mice. Treatments were administered subcutaneously three times a week, body weight was monitored thrice weekly and blood samples were collected at the beginning, midpoint, and end of the study, with organ collection conducted at the study's conclusion. (B) Decitabine exposure at Week 6 (ng/mL) in plasma; bars show the mean ± SD. (C) Body weight of the study mice over the study period. Lines represent the average body weight for each group: vehicle (control), decitabine 0.4 mg/kg, decitabine 0.6 mg/kg, and HbAA historic control 53; the shaded area indicates the SD for the HbAA controls. N.D. = not detectable. n = 6/group, *p < 0.05.

FIGURE 3

Impact of decitabine treatment on haematological parameters and spleen size in HbSS Townes mice. Each figure represents the three study groups: vehicle control, decitabine 0.4 and 0.6 mg/kg. Haematological parameters were assessed at the beginning of the study (Week 1), mid‐way (Week 6) and end of the study (Week 12). (A) Red blood cell count (RBC, 10¹²/L), (B) Haemoglobin concentration (Hb, mmol/L), (C) Reticulocyte counts (10¹²/L), (D) Mean corpuscular volume (MCV, fL), (E) Mean corpuscular haemoglobin (MCH, amol), (F) Mean corpuscular haemoglobin concentration (MCHC, mmol/L), (G) Platelet count (PLT, 10⁹/L), (H) White blood cell count (WBC, 10⁹/L), and (I) Spleen weight (g). Error bars indicate standard deviation. Data points represent individual mice. *p < 0.05, **p < 0.01 and ***p < 0.001 in black comparisons over time within the same group and in red compared to the vehicle group for the same timepoint, indicating statistical significance.

FIGURE 4

Foetal haemoglobin (HbF) induction post‐treatment with decitabine at Week 12 and effects of decitabine treatment on LDH, sternum and liver histology in HbSS Townes mice. (A) The left y‐axis (in black) shows the percentage of F‐cells (cells containing HbF) by flow cytometry, right y‐axis (in blue) shows the percentage of HbF protein levels by HPLC. (B) Correlation between HbF levels measured by HPLC and flow cytometry, displayed with its correlation coefficient (r = 0.9). (C) HbF levels calculated from HPLC data. (D) Lactate dehydrogenase (LDH) levels (U/L) for the three treatment groups (vehicle, decitabine 0.4, and 0.6 mg/kg) were measured at Weeks 1, 6 and 12. (E) Representative H&E histological sections of the sternum at 20× magnification for each treatment group. (F) Quantitative analysis of histological features classified by the liver classifier algorithm, affected by RBC clot, diseased tissue and necrotic tissue, presented as a percentage of the total area for each liver section from mice in each treatment group. (G) Representative H&E stained histological sections of the liver from each treatment group at 20× magnification with examples of RBC clots marked in pink, diseased tissue in brown and necrotic tissue in orange. Data points represent individual mice, with mean values indicated by bars and SD by error bars. *p < 0.05, ***p < 0.001 indicates significant differences from vehicle control. Scale bars in (E) and (G) indicate 100 µm.