Single Cell Hemoglobin Quantification Reveals HbF and HbS Increases in Sickle Erythrocytes from Patients Treated with Hydroxyurea

Sickle cell disease (SCD) is caused by a point mutation in the β-globin gene, leading to the formation of hemoglobin S (HbS). Polymerization of deoxygenated HbS causes red blood cell (RBC) sickling and hemolysis. Because of its anti-polymerization effect, fetal hemoglobin (HbF) is a major regulator of disease severity. Hydroxyurea (HU) treatment increases the global HbF percentage and the proportion of cells expressing HbF. However, the anti-polymerizing effect of HbF depends on the ratio of HbF to HbS in each cell. We developed a method to measure HbF level (in pg.) in each RBC by flow cytometry (FC) showing that RBCs containing less than 2 pg. of HbF were statistically decreased in patients treated with HU, while RBCs with more than 2 pg. were increased. HU treatment also increases mean corpuscular hemoglobin (MCH) and mean corpuscular volume (MCV), perhaps by enhancing stress erythropoiesis. To understand how HU treatment modulates HbF and HbS at the cellular level, HbF and HbS were quantified by FC in SCD patients.

HbF/RBC was quantified in 189 individuals (n=159, HU⁺ group, n=30, HU^- group) homozygous for the HbS gene or with HbS-β⁰ thalassemia, >1 mo. from a vaso-occlusive episode, >3 mos. from transfusion, and on a stable dose of HU for 3 mos. RBCs were stained with an anti-HbF antibody and analyzed by FC. Fluorescence was converted into pg. of HbF/cell using a standard curve (Hebert et al. Am J Hematol, 2020). Cellular HbS was calculated as: HbS (pg.) = MCH (pg.) – measured HbF (pg.) – MCHbA₂ (pg.). MCH was obtained from CBC, and MCHbA₂ was calculated from the percentage of HbA₂ determined by HPLC (MCH*%HbA₂). The ratio of HbF/HbS was calculated for each cell. RBCs were then classified into HbF, HbS, and HbF/HbS ranges. HU⁺ and HU^- groups were compared with a Mann-Whitney test, interaction between HU treatment and ranges was determined with a two-way ANOVA.

The mean age of our cohort was 38 yrs. (range 19-63 yrs.) with 42% of patients being male. Hb level was similar between groups (8.76±1.25 g/dL in the HU^- group vs 9.1±1.32 g/dL in the HU⁺ group). Percent HbF was lower in HU^- group (6.4±5.3% vs 16.7±10.7% in HU⁺ group, p<0.0001), and MCH was higher in the HU⁺ group (26.62±4.14 pg. HU^- vs 34.16±5.97 pg. HU⁺, p<0.0001). Percent HbS was higher in the HU^- group (84±4.9% vs 74.1±10.6% in the HU⁺ group, p<0.0001). HU treatment significantly modified RBC distribution in all HbF ranges (p<0.0001), with a lower %RBC in the [0-2 pg.] category and an increase in all other ranges from 2 to >20 pg. in accordance with increased %HbF with HU treatment. With HbF quantified for each patient and using MCH, we calculated HbS content for each cell, and classified RBCs into ranges of HbS. HU treatment significantly modified RBC distribution in all HbS ranges (p<0.0001), with a lower %RBC in the [15-20] and [20-25 pg.] ranges (respectively 11.3±23.2% and 37±39.2 % in the HU^- group vs 6.5±8.2% and 16±22.6% in the HU⁺ group) and a higher %RBC in the [30-35], [35-40] and ≥ 40 pg. ranges (respectively 13.7±27.2%, 1.4±7.9% and 0% in the HU^- group vs 26.8±27.9%, 11.51±9.1% and 5.9±14.9% in the HU⁺ group). This suggests that both HbF and HbS are augmented with HU treatment, both contributing to MCH increase. Despite this HbS induction, the proportion of RBCs with a HbF/HbS ratio <0.1 was significantly lower in patients treated with HU while the proportion of RBCs with a ratio >0.3, >0.5 and ≥1 was higher, showing a greater HbF induction compared to HbS.

Patients treated with HU have an increased HbF and HbS/RBC. Despite HbS augmentation, HbF/HbS ratio was higher in treated than in untreated patients. The high variability in HbF and HbS single cell expression in both HU-treated and non-treated patients might account for differences in the patient’s severity.