Is Epigenetic Modulation to Induce Fetal Hemoglobin Translatable in Sickle Cell Mice?

Sickle cell disease (SCD) is a genetic disorder caused by a mutation in the β-globin subunit of hemoglobin, resulting in the formation of HbS. Under deoxygenation, HbS polymerizes leading to red blood cell sickling, increased hemolysis and multi-organ damage.

During human fetal development, due to expression of the γ-globin gene, the predominant type of hemoglobin produced is fetal hemoglobin (HbF). However, after birth, this shifts to adult globin (HbA or HbS in SCD). This is driven by epigenetic repression of γ-globin due to an increased methylation of its gene promoters. DNA methyltransferase 1 (DNMT1) is an important enzyme for DNA methylation and is known to be pivotal in regulation of HbF expression.

Reactivation of HbF provides a functional hemoglobin, and therefore has the potential to treat SCD symptoms ( Steinberg et al. JAMA 2003;289:1645-51). HbF induction is an accepted treatment strategy in SCD; even small increases result in decreased vaso-occlusive crises and mortality ( Kato et al. Nat Rev Dis Primers 2018;4:18010). Decitabine inhibits DNMT1 activity, thus promoting γ-globin expression and thereby increasing HbF levels.

While inhibition of DNA methylation is known to cause HbF induction in humans and non-human primates, the effects observed in SCD mouse models are more ambiguous. This study investigated to which extent decitabine treatment could induce HbF expression in the Townes SCD model.

In Townes mice, the murine globin genes have been substituted for their sickle human counterparts, HBA, HBB ^sickle and HBG1, and mimic many features of human SCD. Mice aged 4-5 weeks were dosed subcutaneously 3 times a week for 12 weeks with research-grade decitabine (0.6 mg/kg or 0.4 mg/kg) or vehicle. Blood samples were collected for analysis of HbF protein levels by HPLC (represented as % of total Hb) and F-cells by flow cytometry (represented as % of the total RBC population positive for HbF), complete blood counts, histology, and other markers of disease and organ damage. Results are presented as mean ± SD. Groups were compared using an ANCOVA model, adjusting for multiple comparisons (Dunnett) with p<0.05 considered statistically significant.

For both treatment groups, HbF was significantly elevated compared to the control group, with the maximal response observed at 12 weeks (Figure 1). F-cells at the end of the study were increased in both 0.4 mg/kg (11.4±1.5%) and 0.6 mg/kg (11.8±1.4%) groups compared to vehicle (4.5±0.6%). The same was observed for protein levels: 0.4 mg/kg group (1.2±0.3%), 0.6 mg/kg (1.4±0.3%) compared to vehicle (0.4±0.1%).

Interestingly, after 12 weeks RBC and reticulocyte counts were significantly decreased in treatment groups and mean corpuscular volume (MCV) was significantly increased. No changes on LDH, bilirubin and liver markers (ALP, ALT, AST) were observed at 6 or 12 weeks.

The results showed that treatment groups had significantly higher HbF levels than the vehicle group. However, the level of induction was lower than observed for inhibition of DNA methylation in humans and primates. The decrease in RBCs and reticulocytes could be partially explained by a combination of HbF induction, reduction of sickle RBC, and inhibition of erythropoiesis. The lack of changes in disease markers suggests that the increase in HbF levels in Townes mice might not be sufficient to alleviate the disease. This observation is consistent with recently published data (Woodard et al. Dis Model Mech. 2022;15(6)) that indicates that Townes mice are less suitable for studying this mechanism of action since, even though they have all human globin genes and some proximal regulatory elements, they lack key cis-acting distant DNA elements that regulate the globin genes. However, in primates the globin structure and regulation mechanisms are similar to human (Hardison. Cold Spring Harb Perspect Med. 2012;2(12)).

In conclusion, although the measured levels of HbF protein were no higher than 2%, we found that decitabine can induce HbF expression in the Townes mouse model of SCD. However, within the 12-weeks, there was no clear beneficial effect on the disease markers studied. The successful induction of HbF suggests that better outcomes might be achieved with further optimization of the dosing and duration of the study. This indicates that SCD mice may be useful for preclinical studies of HbF induction, but caution is needed when extrapolating results to humans.