Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Jul 1.
Published in final edited form as: Blood Cells Mol Dis. 2022 Mar 12;95:102660. doi: 10.1016/j.bcmd.2022.102660

Mitapivat increases ATP and decreases oxidative stress and erythrocyte mitochondria retention in a SCD mouse model

Zenaide MN Quezado a,b, Sayuri Kamimura a, Meghann Smith a, Xunde Wang b, Michael R Heaven c, Sirsendu Jana c, Sebastian Vogel a, Patricia Zerfas d, Christian A Combs e, Luis EF Almeida a, Quan Li f, Martha Quezado g, Iren Horkayne-Szakaly h, Penelope A Kosinski i, Shaoxia Yu i, Unnati Kapadnis i, Charles Kung j, Lenny Dang j, Paul Wakim j, William A Eaton f, Abdu I Alayash c, Swee Lay Thein b
PMCID: PMC9148618  NIHMSID: NIHMS1795098  PMID: 35366607

Abstract

Polymerization of deoxygenated sickle hemoglobin (HbS) leads to erythrocyte sickling. Enhancing activity of the erythrocyte glycolytic pathway has anti-sickling potential as this reduces 2,3-diphosphoglycerate (2,3-DPG) and increases ATP, factors that decrease HbS polymerization and improve erythrocyte membrane integrity. These factors can be modulated by mitapivat, which activates erythrocyte pyruvate kinase (PKR) and improves sickling kinetics in SCD patients. We investigated mechanisms by which mitapivat may impact SCD by examining its effects in the Townes SCD mouse model. Control (HbAA) and sickle (HbSS) mice were treated with mitapivat or vehicle. Surprisingly, HbSS had higher PKR protein, higher ATP, and lower 2,3-DPG levels, compared to HbAA mice, in contrast with humans with SCD, in whom 2,3-DPG is elevated compared to healthy subjects. Despite our inability to investigate 2,3-DPG-mediated sickling and hemoglobin effects, mitapivat yielded potential benefits in HbSS mice. Mitapivat further increased ATP without significantly changing 2,3-DPG or hemoglobin levels, and decreased levels of leukocytosis, erythrocyte oxidative stress, and the percentage of erythrocytes that retained mitochondria in HbSS mice. These data suggest that, even though Townes HbSS mice have increased PKR activity, further activation of PKR with mitapivat yields potentially beneficial effects that are independent of changes in sickling or hemoglobin levels.

Keywords: sickle cell, anemia, pyruvate kinase, oxidative stress, proteomics

INTRODUCTION

Sickle cell disease (SCD) is an inherited hemoglobinopathy that affects approximately 100,000 Americans and millions of individuals of predominantly African heritage worldwide[1, 2]. The substitution of valine for glutamic acid at the 6th amino-acid of beta globin results in an abnormal hemoglobin variant (HbS), which polymerizes when deoxygenated to form stiff fibers, altering the structure and function of erythrocytes and distorting them into a ‘sickle’ shape. The ensuing recurrent micro-vaso-occlusion and hemolytic anemia lead to acute pain episodes, chronic vascular endothelial injury, inflammation, and multisystem organ damage[2]. Given that HbS polymerization is the root cause of the pathobiology of SCD, therapies aimed at decreasing HbS fiber formation should be beneficial for SCD patients[3].

2,3-diphosphoglycerate (2,3-DPG) is a metabolite of the erythrocyte glycolytic pathway and the major allosteric modulator of hemoglobin’s affinity for oxygen[4–6]. 2,3-DPG promotes sickling because it decreases HbS oxygen affinity by preferentially binding to the low-affinity deoxyhemoglobin conformation (T) and stabilizes the sickle fiber[4–6]. Erythrocytes from SCD patients have elevated levels of 2,3-DPG compared to those from control subjects[7, 8], a finding also reported in SCD mice[9]. The clinical relevance of 2,3-DPG is illustrated by cases of patients who have sickle cell trait and developed a severe phenotype similar to patients with SCD because of co-inheritance of pyruvate kinase (PK) deficiency, which can lead to elevated levels of 2,3-DPG[10, 11]. Conversely, lowering intracellular 2,3-DPG levels decreases sickling of erythrocytes from SCD patients[6, 12]. One strategy to lower intracellular 2,3-DPG levels is to enhance the activity of erythrocyte pyruvate kinase (PKR), a key enzyme in the glycolytic pathway. Mitapivat (AG-348), an allosteric modulator of PKR, has been shown to restore the thermostability of PKR, increasing its activity, and decreasing 2,3-DPG levels in erythrocytes from SCD patients [8, 13–15]. In an effort to delineate how mitapivat might be affecting functional metabolism of sickle erythrocytes, we evaluated the drug’s effects in the Townes SCD mouse model.

METHODS

The supplemental materials provide details on mouse strains, protocol, and analytical methodology.

Sickle cell mice and study design

The NIH Clinical Center Animal Care and Use Committee approved the study. We used the B6;129-Hbatm1(HBA)TowHbbtm2(HBG1,HBB*)Tow/ Hbbtm3(HBG1,HBB)Tow/J strain, here referred to as the Townes strain[16, 17]. We also included C57BL/6J, a strain in which most preclinical studies of mitapivat have been conducted[13]. Breeding and genotyping were conducted as described[18, 19].

We studied two Townes mice cohorts each including all four experimental groups (HbAA and HbSS mice treated with vehicle or mitapivat). Each experimental group in both cohorts contained 6 to 7 age- (10 to 12 weeks at time of entry) and sex-matched male and female HbSS, HbAA, and C57BL/6J mice. C57BL/6J were included in the first cohort only. Sixty-three mice were included in the pre-clinical study [N=12–13 for Townes mice (HbAA or HbSS) and N=6 for C57BL/6J per genotype and per treatment arm (vehicle vs. mitapivat)].

For 4 weeks, mice were fed vehicle (standard) or mitapivat mouse chow (200 mg/kg/day) provided by Agios, Inc. The mitapivat dose was based on preliminary experiments examining pharmacokinetics and pharmacodynamics of the drug in Townes mice. Hematological and biochemical outcome measurements were obtained before (baseline) and after treatment.

Mitapivat quantification in plasma

Mitapivat plasma concentration was measured using liquid chromatography tandem mass spectrometry (LC-MS/MS) following a simple protein precipitation method. A 20-μl aliquot of the plasma (K2-EDTA) from study samples was mixed with 200μl acetonitrile (with 0.1% FA) and the internal standard (mitapivat d8, 200 ng/mL). The mixture was vortexed at 1500 rpm for 2 min and centrifuged at 4000 rpm for 5 min. Subsequently, 20-µl aliquot of supernatant was diluted with 180-μL water (with 0.1% FA). The contents were mixed and 10-µL samples was injected for analysis by LC-MS/MS. A reversed-phase gradient method utilizing Water (with 0.1% FA, mobile phase A) and Acetonitrile (with 0.1% FA, mobile phase B) was used for chromatographic separation on an Hypersil Gold™ C18 (3 µm, 2.1× 50 mm, ThermoFisher Scientific, Framingham, MA). The detection of eluting analytes was carried on a Triple Quad 6500 mass spectrometer (SCIEX. Framingham, MA) in positive electrospray ionization mode. The total run time of the assay was 3.5 min and analytical column was kept at 40°C throughout the analysis while sample manager was maintained at 7°C. Supplemental Table 1 provides additional details.

ATP and 2,3-DPG quantification in whole blood

Whole blood levels of ATP and 2,3-DPG were determined using liquid chromatography tandem mass spectrometry (LC-MS/MS). A 5-µL of whole blood sample was mixed with 10μl of water containing the internal standards 13C3-2,3-diphosphoglycerate (200 μg/mL) and 13C10,15N5-ATP (200μg/mL). Samples were then vortex mixed and 40µL water was added. Subsequently, samples were extracted with 320μL of methanol, mixed for 10 min and then centrifuged at 4000 rpm for 10 minutes. A 400-µL supernatant was transferred to another plate and 320μl of acetonitrile was added. The mixture was mixed for 2 minutes, followed by centrifugation at 4000 rpm for 5 minutes. The resulting supernatant was transferred to an injection plate for LC-MS analysis. A 5-µL aliquot of supernatant was injected for analysis by LC-MS/MS. An ion exchange method was used for separation of ATP and 2,3 DPG on a BioBasic AX analytical column (2.1 X 50 mm, 5 µM, Thermo Fisher Scientific, Waltham, MA). Detection was done on a SCIEX Triple Quad 6500 mass spectrometer (SCIEX. Framingham, MA) in negative electrospray ionization mode. Calibration curves were established using ATP and 2,3 DPG authentic standards (50 to 2000 µg/mL). The peak area ratios of the analyte relative to the internal standard were used for quantitation. Supplemental Table 2 provides additional details.

Sickling assay

When HbS polymerizes upon deoxygenation in erythrocytes, the cell shape is simultaneously distorted[20], which is readily detected using an optical microscope[21]. Images of SS mouse erythrocytes suspended in a pH 7.3 300 mOsM phosphate buffered saline solution (32 mM Na2HPO4, 8 mM KH2PO4, 130 mM NaCl, 1 mg/mL BSA, 1 mg/mL dextrose) were collected following the initiation of deoxygenation using a Lionheart FX automated microscope system (BioTek Instruments, Inc.) in a 37°C humidified chamber; nitrogen flow was regulated to reach and maintain 5% oxygen inside the instrument. Complete deoxygenation of the cells requires about 30 min. Several metrics were used to determine the time at which a cell sickles, loss of circular shape, loss of more transparent center characteristic of a biconcave disc, and a decrease in area of the cell. The output of the assay is the fraction of cells sickled as a function of time after the initiation of deoxygenation, from which the time required for 50% of the cells to sickle (t50) is obtained. An increase in t50 corresponds to a decrease in sickling. More details of the assay have been previously reported[21].

Measurement of reactive oxygen species (ROS) in erythrocytes.

Intracellular ROS was measured using a cell-permeant fluorometric probe 2’,7’-dichlorofluorescin diacetate (DCFDA) that detect mainly hydroxyl, peroxyl and some other ROS within the cell. Upon oxidation, fluorescent DCF was detected by fluorescence spectroscopy with excitation/emission at 495nm/530nm[22]. Erythrocytes (995μL 10%v/v suspension in PBS) were incubated with 5μL of DCFDA (10mM, prepared in DMSO) at 37°C for 30 min. Following incubation, the suspension was diluted 20 times in PBS and the fluorescence measured in Synergy-HTX 96-well plate fluorimeter (Biotek Instruments, VT). ROS formation was expressed as relative fluorescence units (RFU)/µg of total protein. To eliminate any autofluorescence from hemoglobin, blank erythrocyte controls from each group were prepared with DMSO and without DCFDA.

Whole blood transmission electron microscopy (TEM) and quantification of mitochondria in erythrocytes

TEM images of whole blood were obtained as previously described[23, 24]. Investigators unaware of animals’ genotypes and treatment, quantified erythrocytes and mitochondria using a commercial machine learning software (Aivia v9.0, SVision LLC, Bellevue, WA)[25, 26]. In TEM images, individual anuclear cells with curved and generally smooth elliptical shape were interpreted as to represent the cross-section images of individual erythrocytes. Cells with irregular surface contour, numerous surface indentations, containing endocytic vesicles and intracellular vacuoles were interpreted as reticulocytes[27]. Intracellular round structures with a smooth outer membrane containing an inner membrane and broad folds forming cristae, were interpreted as to represent mitochondria. We analyzed four images per mouse and calculated the average fraction of erythrocytes containing mitochondria and number of mitochondria per erythrocytes[28].

Micro data-independent acquisition (µDIA) mass spectrometry analysis

Thirty µg of protein from erythrocytes lysates were subjected to SDS-PAGE electrophoresis for 5 min to permit performing an in-gel trypsin digestion with iodoacetamide alkylation as previously described[29]. Digested peptides from each sample were analyzed by LC-MS/MS on a Q-Exactive Orbitrap mass spectrometer (Thermo Scientific, Waltham, MA) in conjunction with an Easy nLC II Proxeon nanoflow HPLC and nanoelectrospray ionization source operating in positive ionization mode.

Statistical Analysis

For each outcome (dependent) variable, a multi-factor analysis of variance (ANOVA) model was fit with the following explanatory (independent) variables: mouse cohort, treatment, genotype, sex, outcome baseline value, and the following interaction terms: treatment-by-genotype, treatment-by-sex, and treatment-by-genotype-by-sex. Least-squares means and corresponding 95% confidence intervals were obtained from the models. Model fit diagnostics were examined to determine whether model assumptions were met. In the result section, when reporting overall genotype effect, those were controlled for treatment and available baseline levels. Similarly, when reporting overall effect of treament, those were controlled for genotype and available baseline levels. Results of whether there is a difference between groups have been interpreted and stated based on unadjusted p-values, while keeping in mind that multiple statistical tests have been performed.

RESULTS

Mitapivat levels varied according to mouse genotype and sex

During the study, there was no mortality in any treatment group. Mitapivat-treated HbAA and HbSS mice had similar body weight, whereas mitapivat-treated C57BL/6J had higher body weight compared to respective vehicle-treated mice (supplemental Figure 1, p=0.0002).

At week 4, treatment was associated with lower mitapivat plasma levels in HbAA compared to HbSS (p<0.0001) and C57BL/6J mice (p=0.016, Figure 1A). Mitapivat levels varied according to animal sex as controlling for genotype, mitapivat-treated females had higher plasma levels than mitapivat-treated male mice at week 4 (p=0.0001, Figure 1B).

Figure 1. Mitapivat plasma levels varied by mouse genotype and sex.

Figure 1.

Open in a new tab

A. At the end of the 4-week treatment, mitapivat plasma levels varied according to genotype (p=0.0057 for treatment-by genotype interaction). Controlling for sex, plasma levels of mitapivat were lower in HbAA mice compared to HbSS (p<0.0001) and C57BL/6J animals (p=0.016). B. Mitapivat plasma levels also varied according to sex (p=0.0057 for treatment-by-sex interaction) in that, controlling for genotype, mitapivat-treated females had higher plasma levels compared to mitapivat-treated male mice (p=0.0001). Data are shown as least-square mean estimates (bars) and 95% confidence interval (error bars) for each genotype, sex, and treament.

HbSS have lower 2,3-DPG levels compared to HbAA mice and administration of mitapivat increases ATP without significantly changing 2,3-DPG levels in Townes mice

ATP and 2,3-DPG, normalized to hematocrit, served as a read-out of PKR activity and evidence of target engagement (Figure 2). At week 4, across treatment groups, HbSS mice had higher levels of ATP compared to HbAA and C57BL/6J mice (p=0.010, for overall genotype effect, Figure 1C). Mitapivat treament was associated with higher ATP levels compared to vehicle in animals of all genotypes (p=0.0004, for overall treatment effect, Figure 2A). Across treament groups, at week 4, HbSS mice had lower 2,3-DPG levels compared to HbAA and C57BL/6J mice (p=0.0003, for overall genotype effect, Figure 2B), which suggests that HbSS mice have higher net PKR activity compared to control mice. Mitapivat administration was associated with lower 2,3-DPG levels in C57BL/6J (p=0.008) but had minimal effect on 2,3-DPG levels in HbAA (p=0.156) or HbSS (p=0.827), compared to respective vehicle-treated mice, Figure 2B. Overall, across treatment groups, HbSS mice had higher ATP/2,3-DPG ratio compared to HbAA and C57BL/6J mice (p<0.0001, for overall genotype effect, Figure 2C). Mitapivat administration further increased ATP/2,3-DPG ratio compared to vehicle (p<0.0001, for overall treatment effect, Figure 2C).

Figure 2. Oral administration of mitapivat further increases whole blood ATP without significantly changing 2,3-DPG levels in sickle cell mice.

Figure 2.

Open in a new tab

A. Controlling for treatment, sex, and baseline measurements, HbSS mice had higher levels of ATP compared to HbAA and C57BL/6J mice (p=0.010, for overall genotype effect). At week 4 of administration, mitapivat-treated mice had higher levels of ATP compared to vehicle-treated animals (p=0.0004, for overall treatment effect). B. Controlling for treatment, sex, and baseline measurements, HbSS mice had lower levels of 2,3-DPG compared to HbAA and C57BL/6J mice (p=0.0003, for overall genotype effect). At week 4 of administration, mitapivat-treated C57BL/6J (p=0.008), but not HbAA (p=0.156) or HbSS (p=0.827), had lower levels of 2,3-DPG compared to respective vehicle-treated mice. C. Controlling for treatment, sex, and baseline measurements, HbSS mice had higher ATP/2,3-DPG ratio compared to HbAA and C57BL/6J mice (p<0.0001, for overall genotype effect). Mitapivat-treated mice had higher ATP/2,3-DPG ratio compared to vehicle-treated animals (p<0.0001, for overall treatment effect). Data are shown as least-square mean estimates (bars) and 95% confidence interval (error bars) for each genotype, sex, and treament.

Mitapivat decreased spleen size and white blood cell count without significantly changing red blood cell count, hemoglobin, or sickling kinetics in HbSS mice

As the spleen is a site of extramedullary hematopoiesis in SCD mice, HbSS mice had a higher spleen/body weight ratio compared to HbAA and C57BL/6J mice (p<0.0001 for overall genotype effect, Figure 3A). Mitapivat administration resulted in decreased spleen/body weight ratio in HbSS (p=0.042), but not HbAA or C57BL/6J mice, compared with vehicle. White blood cell counts were significantly higher in HbSS compared to HbAA and C57BL/6J mice (p=0.0002 for overall genotype effect, Figure 3B). Mitapivat treatment in HbSS mice, but not in HbAA or C57BL/6J, resulted in reductions in white blood cell counts (p=0.002, Figure 3B). In contrast, mitapivat treatment did not yield significant changes in red blood cell counts (p=0.873 for overall treatment effect, Figure 3C) or hemoglobin (p=0.895 for overall treatment effect, Figure 3D). Across treatments and in concert with prior reports [16–18], HbSS mice had higher mean corpuscular volume (MCV, p=0.028), red blood cell distribution width (RDW, p=0.036), mean corpuscular hemoglobin (MCH, p=0.0005) and lower mean corpuscular hemoglobin concentration (MCHC, p=0.0005) compared to HbAA mice. Overall, mitapivat administration had no effect on any of those indices (MCV, p=0.685; RDW, p=0.440; MCH, p=0.369; and MCHC, p=0.759). However, post hoc analyses showed that in HbSS mice, mitapivat treatment led to lower mean corpuscular hemoglobin (p=0.039, supplemental Figure 2) compared to vehicle. We also examined sickling kinetics and found that mitapivat-treated HbSS mice showed a trend towards an increase in the time for 50% of the cells to sickle following the initiation of deoxygenation with nitrogen to 5% oxygen (t50, p=0.092, Figure 3E), suggesting an anti-sickling effect.

Figure 3. Mitapivat decreases spleen size and white blood cell count in HbSS mice without altering red blood cells count, hemoglobin or sickling kinetics.

Figure 3.

Open in a new tab

A. Controlling for treatment and sex, HbSS mice had higher spleen/body weight ratio compared to HbAA and C57BL/6J mice (p<0.0001 for overall genotype effect). At week 4 of treatment, mitapivat-treated HbSS (p=0.042), but not HbAA or C57BL/6J mice, had lower spleen/body weight ratio compared to vehicle-treated animals. B. Controlling for treatment, sex, and baseline measurements, HbSS mice had higher white blood cell count compared to HbAA and C57BL/6J mice (p=0.0002 for overall genotype effect). Mitapivat-treated HbSS (p=0.0023), but not HbAA or C57BL/6J mice, had lower white blood cell count compared to vehicle-treated animals. C and D. At week 4, mitapivat treatment did not yield significant changes in red blood cell counts (p=0.873 for overall treatment effect, C) or hemoglobin (p=0.895 for overall treatment effect, D). E. At week 4 of treatment, mitapivat-treated HbSS showed a trend towards increase in t50 (p=0.092). F. Among vehicle-treated mice, we observed an approximately two-fold increase in dichlorodihydrofluorescein diacetate (DCFDA)-fluorescence (a marker for reactive oxygen species) in HbSS erythrocytes compared to HbAA (p<0.001). At week 4 of treatment, mitapivat-treated HbSS (p=0.021), but not HbAA (p=0.94), had lower erythrocyte reactive oxygen species levels compared to respective vehicle-treated mice. Data are shown as least-square mean estimates (bars) and 95% confidence interval (error bars) for each genotype and treament.

Mitapivat reduces ROS levels in erythrocytes in SS mice.

Erythrocytes from HbSS mice are known to have higher oxidative burden compared with HbAA mice[30, 31]. We measured ROS in erythrocytes after mitapivat treatment using a ROS specific fluorescent probe, DCFDA. Among vehicle-treated mice, we observed a higher DCFDA-fluorescence in HbSS compared to HbAA mice (p=0.001, Figure 3F). Notably, compared with vehicle, mitapivat treatment was associated with lower erythrocytic ROS levels in HbSS mice (p=0.021), but not in HbAA or C57BL/6J (Figure 3F).

Mitapivat reduces mitochondria retention in sickle cell mice erythrocytes

We and others have reported that erythrocytes from SCD patients and mice retain mitochondria[32, 33]. Using TEM and a machine learning software, we examined the effect of mitapivat treatment on mitochondria retention by erythrocytes (Figure 4). Akin to SCD patients, Townes HbSS mice had a higher percentage of erythrocytes containing mitochondria (p<0.001, for overall genotype effect, Figures 4A-E) and higher number of mitochondria per erythrocyte (p<0.001, for overall genotype effect, Figure 4F) compared to HbAA mice. Compared with vehicle, mitapivat treatment in HbSS mice decreased the percentage of erythrocytes retaining mitochondria (p=0.002, Figure 4E) and reduced the number of mitochondria per erythrocyte (p=0.027, Figure 4F).

Figure 4. Mitapivat reduces mitochondria retention in sickle cell mice erythrocytes.

Figure 4.

Open in a new tab

We examined mitochondria retention in erythrocytes in whole blood using transmission electron microscopy and a machine learning software for image segmentation (for details see supplemental methods). Panels A through D show representative electron microscopy images and corresponding machine learning software generated images of erythrocytes (green) and mitochondria (white). The scale bar represents 250nm. Panel A shows a vehicle-treated HbAA mouse, panel B a mitapivat-treated HbAA, panel C a vehicle-treated HbSS, and panel D a mitapivat-treated HbSS mouse. HbSS mice had a higher percentage of erythrocytes containing mitochondria (p<0.001 for genotype effect, E) and a higher number of mitochondria per erythrocyte that contained mitochondria (p<0.001 for genotype effect, F) compared to HbAA mice. Mitapivat-treated HbSS had a lower percentage of red blood cells that retained mitochondria (p=0.002, E) and reduced number of mitochondria per erythrocyte that contained mitochondria (p=0.027, F). Data are shown as least-square mean estimates (bars) and 95% confidence interval (error bars) for each genotype and treament. Twenty-four mice were included in this experiment, N=6 for each experimental group including balanced number of males and females.

Mitapivat alters the proteomic profile in HbSS mice

To identify proteins with altered abundance among experimental groups (mitapivat treated HbSS, vehicle-treated HbSS, and vehicle-treated HbAA mice), we applied untargeted µDIA mass spectrometry on tryptic digests generated from RBC lysates. In total, this approach measured the relative abundance of 1,262 proteins, with 592 protein abundance differences (p<0.05) comparing vehicle-treated HbSS and HbAA mice and thirteen differences (p<0.05) comparing mitapivat-treated and vehicle-treated HbSS mice. The protein quantification results are displayed as a volcano plot (Figure 5 and Supplemental Table 3 for fold-change (f.c.) and p-values for all quantified proteins).

Figure 5. Mitapivat alters the proteomic profile in HbSS mice.

Figure 5.

Open in a new tab

Volcano plots of all the proteins assayed by mass spectrometry. The y-axis shows the p-value for the comparison of protein-level relative abundance and the x-axis the fold-change ratio comparing vehicle-treated HbSS mice with vehicle-treated HbAA animals (A) and mitapivat-treated with vehicle-treated HbSS mice (B). The Swiss-Prot gene names shown in red indicate proteins that were confirmed altered by targeted mass spectrometry. White circles represent p<0.05, black circles p<0.05, and red circles proteins confirmed altered by targeted mass spectrometry. Eighteen mice were included in this experiment, n=6 for each experimental group including balanced number of males and females.

The vehicle-treated HbSS versus AA mice comparison recapitulated previously reported observations in SCD mice[31], Figure 5A. Vehicle-treated HbSS mice had increases in superoxide dismutase (Sod2, f.c.+2.6, p=5.8E-4) and thioredoxin-dependent peroxide reductase (Prdx3, f.c.+2.1, p=0.11) compared to vehicle-treated HbAA in keeping with the observed elevated ROS levels. Among vehicle-treated animals, HbSS mice had upregulation of mitochondrial proteins including ATP synthase alpha (Atp5a1, f.c.+10.3, p=1.9E-6) and beta (Atp5b, f.c.+8.9, p=2E-7) as well as NADH-ubiquinone oxidoreductase 75 kDa subunit (Ndufs1, f.c.+18.9, p=6.1E-6) compared with HbAA. Additionally, among vehicle-treated animals, HbSS mice had highly upregulated levels of cytochrome c oxidase subunit 2 (Mtco2, f.c.+9.2, p=4.5E-7) compared with HbAA animals. When we examined the levels of the erythrocyte glycolytic pathway enzymes, we found that vehicle-treated HbSS mice had increased levels of PKR (f.c.+1.8, p=0.006) and hexokinase (f.c.+3.4, p=0.012) compared to HbAA, Figure 5A.

As for proteins altered by mitapivat treatment (f.c.>1.3 or f.c.<−1.3, p< 0.05), comparing mitapivat- with vehicle-treated HbSS (Figure 5B), we found increased levels of retinal dehydrogenase 1 (Aldh1a1, f.c.+1.7, p=0.008) and the translation initiation factor (Eif2b1, f.c. not determined because it was only detectable in mitapivat-treated HbSS mice, p=0.03) and decreased levels of Aldh3a2 (f.c.−1.6, p=0.004), Acd (f.c.−31, p= 0.004), Rab21 (f.c.−5.4, p=0.007), Arpc3 (f.c.−1.7, p=0.007), Acta1 (f.c.−1.7, p=0.02), Chil3 (f.c.−3.0, p=0.02), Serpinf2 (f.c.−4.4, p=0.03), Kpna3 (f.c-2.3, p=0.03), and Tubb6 (f.c.−1.4, p=0.05) (Supplemental Table 3). Numerous serine proteases (including Serpinf2 above) were downregulated in mitapivat-treated versus vehicle-treated HbSS mice, but these did not reach statistical significance Serpina1d, (f.c.−2.4, p=0.06), Serpina1a (f.c.−1.5, p=0.07), Serpina3n (f.c.−4.9, p=0.14). We also found that the levels of hemoglobin subunit beta (HBB) were increased in mitapivat-treated compared to vehicle-treated HbSS mice, although the difference did not reach statistical significance (f.c.+1.8, p = 0.18).

Targeted PRM-MS Confirmation of PK, Aldh1a1, Serpina1d, Itga2b, and Mtco2.

To determine if the differentially abundant proteins identified in the comparisons of mitapivat treated HbSS relative to vehicle-treated HbSS and of vehicle-treated HbSS and HbAA mice could be validated, we applied a targeted PRM-MS assay (Supplemental Table 4). We chose to validate PK, Aldh1a1, Serpina1d, Itga2b, and Mtco2, as well as proteasome inhibitor PI31 subunit (Psmf1) and complement factor I (Cfi). Additionally, we included tryptic autolysis peptides from the protease used for in-gel digestion as a negative control, since porcine trypsin was added equally to each sample analyzed. The relative protein fold-change ratios for 6/8 (75%) proteins selected for measurement had the same up or down quantification result in mitapivat-treated versus vehicle-treated HbSS and in vehicle-treated HbSS compared with vehicle-treated HbAA mice RBC lysates (Table 1). Further, the PRM-MS assay provided similar protein quantification ratios for the porcine trypsin autolysis negative control analyzed in samples from mitapivat T-treated HbSS, vehicle-treated HbSS, and vehicle-treated HbAA samples. The PRM-MS reproduced the same quantification finding in the targeted PRM-MS assay for PK, Aldh1a1, Serpina1d, Itga2b, and Mtco2 as the initial discovery proteomics experiment.

Table 1.

Verification of µDIA untargeted proteomics relative quantification with targeted PRM-MS*

Protein Vehicle-treated SS/AA µDIA fold-change Vehicle-treated SS/AA PRM fold-change Mitapivat-treated SS/vehicle-treated SS µDIA fold-change Mitapivat-treated SS/vehicle-treated SS PRM fold-change
Pklr 1.80 2.56 1.01 1.01
Aldh1a1 −1.81 −1.10 1.65 1.55
Serpina1d 1.14 −1.58 −2.37 −1.49
Itga2b −1.86 −1.32 1.95 1.63
Mtco2 9.22 5.74 −1.38 −1.42
Psmf1 2.37 4.89 1.32 −1.35
Cfi 2.35 4.18 −1.98 1.18
** 1.04 1.29 1.08 −1.22
Open in a new tab
*

Comparison of the average fold change ratio of proteins using micro data-independent acquisition (µDIA) versus parallel reaction monitoring-mass spectrometry (PRM-MS) experiments for eight proteins selected for confirmation (n=6 per group in µDIA and PRM-MS experiments). The ‘protein column’ shows the Swiss-prot gene abbreviation for each protein.

**

indicates the porcine trypsin negative control that was spiked equally into each sample for protein digestion

Discussion

We hypothesized that mitapivat, an allosteric activator of PKR, would decrease sickling and ameliorate the hematologic parameters in the Townes SCD model by reducing 2,3-DPG and elevating ATP. This hypothesis is supported by reports that erythrocytes from Berkeley SCD mice have lower ATP[34] and higher 2,3-DPG levels[9] compared to controls, suggesting that PKR activity is diminished in SCD mice. That PKR activity is diminished in SCD is also supported by humans studies showing significantly elevated 2,3-DPG levels and reduced PKR activity in erythrocytes from sickle-cell anemia patients[7, 8]. In vitro, enhancement of PKR stability and activity with mitapivat in erythrocytes from SCD patients, decreased 2,3-DPG, p50 (oxygen tension when hemoglobin is 50% saturated with oxygen) and the point of sickling (oxygen tension at which erythrocytes start to sickle)[8]. Together, these reports support the hypothesis that PKR activity is compromised in SCD and that enhancement of PKR activity with mitapivat would decrease sickling and ameliorate the SCD phenotype.

Indeed, a recent study showed that, while the levels of PKR and hexokinase (the first enzyme in the glycolytic pathway) were elevated compared to controls, the PKR/hexokinase ratio was decreased, indicating that SCD erythrocytes have a relative deficiency in PKR activity[8]. In that study, the investigators also showed that this deficiency in PKR is associated with decreased PKR thermostability and activity, which in turn, leads to elevated levels of erythrocyte 2,3-DPG[8]. Here we found that, contrary to these findings in SCD patients[8], HbSS Townes mice appear to have increased PKR activity as indicated by higher ATP and lower 2,3DPG levels. Therefore, these surprising and contradictory results suggest that erythrocytes from the Townes SCD model do not recapitulate the PKR activity profile observed in erythrocytes from SCD patients.

In keeping with prior reports[13], mitapivat treatment decreased 2,3-DPG and increased ATP levels in C57BL/6J mice, suggesting that the levels of drug exposure were adequate. Conversely, in HbAA and HbSS mice, mitapivat treatment, while yielding further increases in ATP, did not significantly change 2,3-DPG levels. These effects in erythrocyte metabolites were associated with a modest trend towards decreases in sickling but not with significant changes in hemoglobin or number of red blood cells. These results are contrary to the preliminary findings of an ongoing clinical trial of mitapivat in SCD patients, where the drug yielded increases in hemoglobin[15]. Even though mitapivat administration to HbSS mice did not change hemoglobin levels, it decreased leukocytosis, erythrocyte oxidative stress, and the percentage of erythrocytes that retained mitochondria. One could postulate that this reduction in erythrocyte mitochondria retention could be related to an improvement in erythrocyte stability derived from increases in ATP levels. Together these data suggest that, while HbSS Townes intrinsically appear to have elevated PKR activity, further enhancement of PKR activity with mitapivat administration yields potentially beneficial effects independent of changes in sickling kinetics or hemoglobin levels.

Mature erythrocytes from humans and mice with SCD abnormally retain mitochondria[32, 33]. We have recently shown that mitochondria retained in mature erythrocytes are the source of cell free mitochondrial DNA in plasma, which can trigger the formation of neutrophil extracellular traps and increase inflammatory activity[33]. Additionally, in Townes SCD mice, the proportion of mature erythrocytes that retain mitochondria directly correlates with the proportion of erythrocytes with high ROS levels[32]. Here we found that mitapivat treatment reduced the fraction of erythrocytes retaining mitochondria, the number of mitochondria retained per erythrocyte, and the levels of cytochrome c oxidase subunit 2 protein. Importantly, these changes were coupled with significant decreases in erythrocyte ROS levels. Overall, our findings in SCD mice are congruent with those in a mouse model of β-thalassemia showing that mitapivat improves mitochondrial biogenesis, ameliorates erythropoietic mitochondrial dynamics and reduces ROS production[35]. Together, our results suggest that in SCD mice, mitapivat treatment improves mitochondrial dynamics and the metabolic milieu, leading to improvement in erythrocyte redox status.

Mitapivat administration to Townes SCD mice yielded a decrease in spleen/body weight ratio, suggesting an improvement in extramedullary erythropoiesis. This finding was associated with decreases in leukocytosis and mean corpuscular hemoglobin, a trend towards an increase in the beta subunit of hemoglobin level, and a significant reduction in erythrocyte mitochondria retention. In a β-thalassemia model, mitapivat treatment was associated with increases in hemoglobin and mean corpuscular volume and an improvement in the ineffective erythropoiesis seen in β-thalassemia[35]. Additionally, in the β-thalassemia model, mitapivat administration yielded a reduction in spleen/body weight ratio, thus suggesting a decrease in extramedullary erythropoiesis[35]. These findings that mitapivat administration decreased spleen/body weight ratio suggest that mitapivat improves the effectiveness of hematopoiesis decreasing extra-medullary hematopoiesis both in β-thalassemia and SCD mice.

Mitapivat administration yielded changes in erythrocyte lysate proteomics that are potentially relevant in SCD. The finding that Aldh1a1 is increased by mitapivat administration in SS is noteworthy given that vitamin A increases iron incorporation into hemoglobin in developing erythrocytes[36],[37]. Mitapivat-induced downregulation of serine proteases might be relevant as these proteins are decreased in hematopoietic progenitor cells during mobilization from bone marrow into the blood stream[38]. Another potentially relevant mitapivat-associated change is the upregulation of the receptor for soluble fibrinogen, Itga2b, which is involved in platelet-to-platelet interactions, platelet aggregation, and plugging of ruptured endothelial cell surfaces[39]. Therefore, it is conceivable that mitapivat could impact platelet function in SCD. Lastly, the downregulation of Mtco2, the last enzyme in the electron transport chain that drives oxidative phosphorylation, is consistent with an attenuated energy demand following additional PK activation and decrease in mitochondria retention by erythrocytes. Together, these data suggest that mitapivat administration might change the erythrocyte proteomic profile of SCD erythrocytes.

Our findings that Townes HbSS erythrocytes have low 2,3DPG and elevated ATP are contrary to those reported in Berkeley SCD mice [9, 34]. Our findings are also contrary to those from a study of another PKR activator, where high dose FT-4202 treatment decreased 2,3-DPG, increased hemoglobin, prolonged erythrocyte half-life, and decreased the point of sickling in Berkeley SCD mice[40]. In that study, animals aged 5 to 15 weeks were treated with FT-4202 for two weeks, and ATP and 2,3-DPG levels were not corrected for hematocrit[40]. Moreover, in that pre-clinical study[40], control mice were not included therefore an estimation of PKR activity in Berkeley mice could not be made. It is possible that the baseline activity in PKR differs between Berkeley and Townes SCD mice, however, in our hands, Berkeley mice showed low levels of 2,3-DPG compared to control mice (supplemental Figure 3). While the reasons for the contrasting results are unclear, genetic differences between the two strains of SCD mice and possible differences in baseline PKR activity might explain some of these inconsistent results.

In conclusion, our findings suggest that the Townes SCD mouse intrinsically have elevated PKR activity as shown by higher ATP and lower 2,3-DPG levels in HbSS erythrocytes. Activating PKR with mitapivat in the Townes SCD model yielded further increases in ATP, which were associated with beneficial reduction in extramedullary hematopoiesis, leukocytosis and mitochondria retention and oxidative stress in erythrocytes - all desirable and salutary effects to ameliorate SCD pathology.

Supplementary Material

1
2
3

Highlights.

  • Mitapivat further increased ATP without changing 2,3-DPG or hemoglobin levels

  • Mitapivat decreased extra-medullary hematopoiesis in SCD mice

  • Mitapivat decreased leukocytosis and erythrocyte oxidative stress in SCD mice

  • Mitapivat decreased percentage of erythrocytes retaining mitochondria in SCD mice

Acknowledgements

The authors are grateful to Jim Paladino and Hoyin Lai from Leica Microsystems, for their invaluable advice on the use of the Aivia software.

Funding

This work was supported by the Intramural Research Program of the National Institutes of Health Clinical Center, National Heart Lung and Blood Institute, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, and U.S. Food and Drug Administration. This research also has been funded, in part, by Agios Pharmaceuticals Inc.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Publisher's Disclaimer: Disclaimer: This article reflects the views of the authors and should not be construed to represent FDAs views or policies

Declaration of interest: P.A.K., S.Y., U.K., and L.D. are current employees and C.K. a former employee of Agios Pharmaceuticals Inc. The remaining authors declare no competing interests

References

  • [1].Hassell KL, Population estimates of sickle cell disease in the U.S, Am J Prev Med, 38 (2010) S512–521. [DOI] [PubMed] [Google Scholar]
  • [2].Piel FB, Steinberg MH, Rees DC, Sickle Cell Disease, N Engl J Med, 376 (2017) 1561–1573. [DOI] [PubMed] [Google Scholar]
  • [3].Eaton WA, Bunn HF, Treating sickle cell disease by targeting HbS polymerization, Blood, 129 (2017) 2719–2726. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [4].Bunn HF, Briehl RW, The interaction of 2,3-diphosphoglycerate with various human hemoglobins, J Clin Invest, 49 (1970) 1088–1095. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [5].Poillon WN, Kim BC, 2,3-Diphosphoglycerate and intracellular pH as interdependent determinants of the physiologic solubility of deoxyhemoglobin S, Blood, 76 (1990) 1028–1036. [PubMed] [Google Scholar]
  • [6].Poillon WN, Kim BC, Labotka RJ, Hicks CU, Kark JA, Antisickling effects of 2,3-diphosphoglycerate depletion, Blood, 85 (1995) 3289–3296. [PubMed] [Google Scholar]
  • [7].Charache S, Grisolia S, Fiedler AJ, Hellegers AE, Effect of 2,3-diphosphoglycerate on oxygen affinity of blood in sickle cell anemia, J Clin Invest, 49 (1970) 806–812. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [8].Rab MAE, Bos J, van Oirschot BA, van Straaten S, Kosinski PA, Chubukov V, Kim H, Mangus H, Schutgens REG, Pasterkamp G, Dang L, Kung C, van Beers EJ, van Wijk R, Decreased activity and stability of pyruvate kinase in sickle cell disease: a novel target for mitapivat therapy, Blood, 137 (2021) 2997–3001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [9].Zhang Y, Dai Y, Wen J, Zhang W, Grenz A, Sun H, Tao L, Lu G, Alexander DC, Milburn MV, Carter-Dawson L, Lewis DE, Eltzschig HK, Kellems RE, Blackburn MR, Juneja HS, Xia Y, Detrimental effects of adenosine signaling in sickle cell disease, Nat Med, 17 (2011) 79–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [10].Cohen-Solal M, Préhu C, Wajcman H, Poyart C, Bardakdjian-Michau J, Kister J, Promé D, Valentin C, Bachir D, Galactéros F, A new sickle cell disease phenotype associating Hb S trait, severe pyruvate kinase deficiency (PK Conakry), and an alpha2 globin gene variant (Hb Conakry), Br J Haematol, 103 (1998) 950–956. [DOI] [PubMed] [Google Scholar]
  • [11].Alli N, Coetzee M, Louw V, van Rensburg B, Rossouw G, Thompson L, Pissard S, Thein SL, Sickle cell disease in a carrier with pyruvate kinase deficiency, Hematology, 13 (2008) 369–372. [DOI] [PubMed] [Google Scholar]
  • [12].Rose ZB, A procedure for decreasing the level of 2,3-bisphosphoglycerate in red cells in vitro, Biochem Biophys Res Commun, 73 (1976) 1011–1017. [DOI] [PubMed] [Google Scholar]
  • [13].Kung C, Hixon J, Kosinski PA, Cianchetta G, Histen G, Chen Y, Hill C, Gross S, Si Y, Johnson K, DeLaBarre B, Luo Z, Gu Z, Yao G, Tang H, Fang C, Xu Y, Lv X, Biller S, Su SM, Yang H, Popovici-Muller J, Salituro F, Silverman L, Dang L, AG-348 enhances pyruvate kinase activity in red blood cells from patients with pyruvate kinase deficiency, Blood, 130 (2017) 1347–1356. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Alayash AI, Targeting the red cell enzyme pyruvate kinase with a small allosteric molecule AG-348 may correct underlying pathology of a glycolytic enzymopathy, Haematologica, 106 (2021) 9–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [15].Xu JZ, Conrey A, Frey I, Nichols J, Menapace LA, Tumburu L, Lequang T, Li Q, Dunkelberger E, Henry E, Cellmer T, Iyer V, Mangus H, Kung C, Dang L, Kosinski PA, Hawkins P, Jeffries N, Eaton W, Thein SL, Phase 1 Multiple Ascending Dose Study of Safety, Tolerability, and Pharmacokinetics/Pharmacodynamics of Mitapivat (AG-348) in Subjects with Sickle Cell Disease, Blood, 136 (2020) 21–22. [Google Scholar]
  • [16].Wu L-C, Sun C-W, Ryan T, Pawlik K, Ren J, Townes T, Wu LC, Sun CW, Ryan TM, Pawlik KM, Ren J and Townes TM. Correction of sickle cell disease by homologous recombination in embryonic stem cells. Blood 108: 1183–1188, Blood, 108 (2006) 1183–1188. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Hanna J, Wernig M, Markoulaki S, Sun CW, Meissner A, Cassady JP, Beard C, Brambrink T, Wu LC, Townes TM, Jaenisch R, Treatment of sickle cell anemia mouse model with iPS cells generated from autologous skin, Science, 318 (2007) 1920–1923. [DOI] [PubMed] [Google Scholar]
  • [18].Khaibullina A, Almeida LE, Wang L, Kamimura S, Wong EC, Nouraie M, Maric I, Albani S, Finkel J, Quezado ZM, Rapamycin increases fetal hemoglobin and ameliorates the nociception phenotype in sickle cell mice, Blood Cells Mol Dis, 55 (2015) 363–372. [DOI] [PubMed] [Google Scholar]
  • [19].Wang L, Almeida LEF, de Souza Batista CM, Khaibullina A, Xu N, Albani S, Guth KA, Seo JS, Quezado M, Quezado ZMN, Cognitive and behavior deficits in sickle cell mice are associated with profound neuropathologic changes in hippocampus and cerebellum, Neurobiol Dis, 85 (2016) 60–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [20].Mozzarelli A, Hofrichter J, Eaton WA, Delay time of hemoglobin S polymerization prevents most cells from sickling in vivo, Science, 237 (1987) 500–506. [DOI] [PubMed] [Google Scholar]
  • [21].Dunkelberger EB, Metaferia B, Cellmer T, Henry ER, Theoretical Simulation of Red Cell Sickling Upon Deoxygenation Based on the Physical Chemistry of Sickle Hemoglobin Fiber Formation, J Phys Chem B, 122 (2018) 11579–11590. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [22].Hermann PB, Pianovski MAD, Henneberg R, Nascimento AJ, Leonart MSS, Erythrocyte oxidative stress markers in children with sickle cell disease, Jornal de Pediatria [DOI] [PubMed]
  • [23].Khaibullina A, Almeida LEF, Kamimura S, Zerfas PM, Smith ML, Vogel S, Wakim P, Vasconcelos OM, Quezado MM, Horkayne-Szakaly I, Quezado ZMN, Sickle cell disease mice have cerebral oxidative stress and vascular and white matter abnormalities, Blood Cells Mol Dis, 86 (2021) 102493. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Almeida LEF, Wang L, Kamimura S, Zerfas PM, Smith ML, Neto OLA, Vale T, Quezado MM, Horkayne-Szakaly I, Wakim P, Quezado ZMN, Locomotor mal-performance and gait adaptability deficits in sickle cell mice are associated with vascular and white matter abnormalities and oxidative stress in cerebellum, Brain Res, 1746 (2020) 146968. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [25].Coto Hernández I, Yang W, Mohan S, Jowett N, Label-free histomorphometry of peripheral nerve by stimulated Raman spectroscopy, Muscle Nerve, 62 (2020) 137–142. [DOI] [PubMed] [Google Scholar]
  • [26].Vogel S, Kamimura S, Arora T, Smith ML, Almeida LEF, Combs CA, Thein SL, Quezado ZMN, NLRP3 inflammasome and bruton tyrosine kinase inhibition interferes with upregulated platelet aggregation and in vitro thrombus formation in sickle cell mice, Biochem Biophys Res Commun, 555 (2021) 196–201. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [27].Koury MJ, Koury ST, Kopsombut P, Bondurant MC, In vitro maturation of nascent reticulocytes to erythrocytes, Blood, 105 (2005) 2168–2174. [DOI] [PubMed] [Google Scholar]
  • [28].Sbardella D, Tundo GR, Campagnolo L, Valacchi G, Orlandi A, Curatolo P, Borsellino G, D’Esposito M, Ciaccio C, Cesare SD, Pierro DD, Galasso C, Santarone ME, Hayek J, Coletta M, Marini S, Retention of Mitochondria in Mature Human Red Blood Cells as the Result of Autophagy Impairment in Rett Syndrome, Sci Rep, 7 (2017) 12297. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [29].Strader MB, Jana S, Meng F, Heaven MR, Shet AS, Thein SL, Alayash AI, Post-translational modification as a response to cellular stress induced by hemoglobin oxidation in sickle cell disease, Sci Rep, 10 (2020) 14218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [30].Belcher JD, Chen C, Nguyen J, Zhang P, Abdulla F, Nguyen P, Killeen T, Xu P, O’Sullivan G, Nath KA, Vercellotti GM, Control of Oxidative Stress and Inflammation in Sickle Cell Disease with the Nrf2 Activator Dimethyl Fumarate, Antioxidants & redox signaling, 26 (2017) 748–762. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [31].Jana S, Strader MB, Meng F, Hicks W, Kassa T, Tarandovskiy I, De Paoli S, Simak J, Heaven MR, Belcher JD, Vercellotti GM, Alayash AI, Hemoglobin oxidation-dependent reactions promote interactions with band 3 and oxidative changes in sickle cell-derived microparticles, JCI insight, 3 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [32].Jagadeeswaran R, Vazquez BA, Thiruppathi M, Ganesh BB, Ibanez V, Cui S, Engel JD, Diamond AM, Molokie RE, DeSimone J, Lavelle D, Rivers A, Pharmacological inhibition of LSD1 and mTOR reduces mitochondrial retention and associated ROS levels in the red blood cells of sickle cell disease, Exp Hematol, 50 (2017) 46–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [33].Tumburu L, Ghosh-Choudhary S, Seifuddin FT, Barbu EA, Yang S, Ahmad MM, Wilkins LHW, Tunc I, Sivakumar I, Nichols JS, Dagur PK, Yang S, Almeida LEF, Quezado ZMN, Combs CA, Lindberg E, Bleck CKE, Zhu J, Shet AS, Chung JH, Pirooznia M, Thein SL, Circulating mitochondrial DNA is a proinflammatory DAMP in sickle cell disease, Blood, 137 (2021) 3116–3126. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [34].Banerjee T, Kuypers FA, Reactive oxygen species and phosphatidylserine externalization in murine sickle red cells, Br J Haematol, 124 (2004) 391–402. [DOI] [PubMed] [Google Scholar]
  • [35].Matte A, Federti E, Kung C, Kosinski PA, Narayanaswamy R, Russo R, Federico G, Carlomagno F, Desbats MA, Salviati L, Leboeuf C, Valenti MT, Turrini F, Janin A, Yu S, Beneduce E, Ronseaux S, Iatcenko I, Dang L, Ganz T, Jung CL, Iolascon A, Brugnara C, De Franceschi L, The pyruvate kinase activator mitapivat reduces hemolysis and improves anemia in a β-thalassemia mouse model, J Clin Invest, 131 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [36].Hodges RE, Sauberlich HE, Canham JE, Wallace DL, Rucker RB, Mejia LA, Mohanram M, Hematopoietic studies in vitamin A deficiency, Am J Clin Nutr, 31 (1978) 876–885. [DOI] [PubMed] [Google Scholar]
  • [37].Allen LH, Iron supplements: scientific issues concerning efficacy and implications for research and programs, J Nutr, 132 (2002) 813s–819s. [DOI] [PubMed] [Google Scholar]
  • [38].Winkler IG, Hendy J, Coughlin P, Horvath A, Lévesque JP, Serine protease inhibitors serpina1 and serpina3 are down-regulated in bone marrow during hematopoietic progenitor mobilization, J Exp Med, 201 (2005) 1077–1088. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [39].Botero JP, Lee K, Branchford BR, Bray PF, Freson K, Lambert MP, Luo M, Mohan S, Ross JE, Bergmeier W, Di Paola J, Glanzmann thrombasthenia: genetic basis and clinical correlates, Haematologica, 105 (2020) 888–894. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [40].Shrestha A, Chi M, Wagner K, Malik A, Korpik J, Drake A, Fulzele K, Guichard S, Malik P, FT-4202, an oral PKR activator, has potent antisickling effects and improves RBC survival and Hb levels in SCA mice, Blood Adv, 5 (2021) 2385–2390. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1795098-supplement-1.pdf (400KB, pdf)
2
NIHMS1795098-supplement-2.xlsx (381.1KB, xlsx)
3
NIHMS1795098-supplement-3.xlsx (10.8KB, xlsx)

RESOURCES