Loss of Cyb5R3 Function Silences Fetal Hemoglobin and Hematocrit Responses to Hydroxyurea in Patients with Sickle Cell Anemia

Abstract

BACKGROUND: Hydroxyurea (HU) is the first FDA-approved and most widely prescribed drug for sickle cell (SS) patients due to its beneficial effects (decreased Hb polymerization, pain crises and hospitalization) and relative safety profile. While the clinical benefits of HU have been found to associate with increased nitric oxide (NO) and fetal hemoglobin (HbF) in most SS patients, there has been considerable interpatient variability in HbF response to the drug. From this standpoint, HU is imperfectly understood mechanistically. Cytochrome b5 reductase 3 (Cyb5R3), also known as methemoglobin reductase, is an enzyme known to modulate redox signaling in erythrocytes and vascular cells by catalyzing one electron transfer reactions using reduced nicotinamide adenine dinucleotide (NADH) as an electron donor. Functionally, membrane-associated CYB5R3 exerts antioxidant properties by modulating heme iron reduction, plasma membrane coenzyme Q reduction, elongation and unsaturation of fatty acids, cholesterol biosynthesis and drug metabolism, while the soluble form of CYB5R3 reduces red blood cell methemoglobin. Unremitting oxidation of the soluble guanylate cyclase (sGC) heme iron in the absence of sufficient Cyb5R3 activity in vascular smooth muscle has been shown to explain some NO resistance that characterizes pulmonary arterial hypertension and other vasculopathy associated with SCD. More than 40 variants of Cyb5R3 have been identified. Of particular interest is a Cyb5R3 variant (T117S) that contains a threonine to serine substitution in the flavoprotein, which lessens the reductase's activity by roughly 50% compared to wild-type. This variant occurs with high frequency (0.23 minor allele) in persons of African ancestry but is rare (<1%) in all other races. Nothing is known about the role of Cyb5R3 in the anemia of sickle cell disease. We hypothesized that Cyb5R3 may act as a modifier of HbF induction by HU in sickle cell anemia. METHODS: Male and female SS patients (n=633, 12-70 years of age) from the Treatment of Pulmonary Hypertension and Sickle Cell Disease with Sildenafil Therapy (walk-PHaSST) trial were recruited for genetic analysis (University of Pittsburgh IRB Protocol Study 19060255). Genotyping of rs1800457 (T117S SNP) was performed using Taqman genotyping assay. One-way ANOVA and Holms Sidak were used for multiple comparisons of HbF levels by Cyb5R3 genotype and HU treatment status. Intrapatient correlation analyses of HbF and hematocrit (HCT) were conducted using Pearsons r. RESULTS: Comparisons of HbF levels by Cyb5R3 genotype (CC, wild-type; CG, heterozygous wild-type and T117S; GG, homozygous T117S) and HU treatment status (treated vs. untreated) at time of study enrollment (baseline) show that HU is a less potent HbF inducer in the GG group (4.32 mean diff, P=0.344) versus the CC and CG patient groups (CC: 6.027 mean diff, P<0.0001 and CG: 6.914 mean diff, P<0.0001). Correlation analyses showed that as HbF levels increased among HU-untreated patients, the trend was for HCT to decrease. This was true for all SS patients, irrespective of Cyb5R3 genotype (CC: r = -0.336, P<0.0001, n=320; CG: r = -0.437 P<0.0001, n=120; GG: r = -0.253, P=0.136, n=36). However, with HU therapy, the inverse relationship between HbF and HCT was less negative in SS patients carrying at least one copy of the wild-type Cyb5R3 gene (CC or CG). In these individuals, increasing levels of HbF (up to 10%) correlated with positive gains in HCT at HbF greater than 5-6% (CC: r = -0.161, P=0.097, n=107) and 6-7% (CG: r = -0.186, P=0.25, n=40). In patients homozygous for T117S, HCT continued trending downward as HbF increased (GG: r = -0.419, P=0.228, n=10). These data indicate that the positive relationship between HbF and HCT is subject to uncoupling by loss of Cyb5R3 function. In vitro studies with human erythroleukemic K562 cells reveal loss of gamma globin induction by HU treatment when wild-type Cyb5R3 is knocked down by siRNA, indicating a requirement for Cyb5R3 in HU-stimulated HbF induction. SUMMARY andCONCLUSION: Loss of Cyb5R3 activity negatively impacts HbF and HCT in HU treated SS patients. We conclude that the T117S variant may be a useful marker for predicting which SS patients will incur beneficial HbF and HCT responses to HU, thereby improving clinical decision-making about best drug choices for preventing HbS polymerization and treating anemia in SS patients.