Increased hemoglobin affinity for oxygen with GBT1118 improves hypoxia tolerance in sickle cell mice

Abstract

Therapeutic agents that increase the Hb affinity for oxygen (O₂) could, in theory, lead to decreased O₂ release from Hb and impose a hypoxic risk to tissues. In this study, GBT1118, an allosteric modifier of Hb affinity for O₂, was used to assess the impact of increasing Hb affinity for O₂ on brain tissue oxygenation, blood pressure, heart rate, O₂ delivery, and tolerance to hypoxia in Townes transgenic sickle cell disease (SCD) mice. Brain oxygenation and O₂ delivery were studied during normoxia and severe hypoxic challenges. Chronic treatment with GBT1118 increased Hb affinity for O₂, reducing the Po₂ for 50% HbO₂ saturation (P50) in SCD mice from 31 mmHg to 18 mmHg. This treatment significantly reduced anemia, increasing hematocrit by 33%, improved cardiac output (CO), and O₂ delivery and extraction. Chronically increasing Hb affinity for O₂ with GBT1118 preserved cortical O₂ tension during normoxia, improved cortical O₂ tension during hypoxia, and increased tolerance to severe hypoxia in SCD mice. Independent of hematological changes induced by chronic treatment, a single dose of GBT1118 significantly improved tolerance to hypoxia, highlighting the benefits of increasing Hb affinity for O₂ and consequently reducing sickling of RBCs in blood during hypoxia in SCD.

NEW & NOTEWORTHY Chronic pharmacologically increased hemoglobin affinity for oxygen in sickle cell disease mice alleviated hematological consequences of sickle cell disease, increasing RBC half-life, hematocrit, and hemoglobin concentration, while also decreasing reticulocyte count. Additionally, chronically increased hemoglobin affinity for oxygen significantly improved survival as well as cortical tissue oxygenation in sickle cell disease mice during hypoxia, suggesting that oxygen delivery and utilization is improved by increased hemoglobin affinity for oxygen.

INTRODUCTION

Sickle cell disease (SCD) is caused by a single point mutation in the DNA encoding the sixth amino acid of the hemoglobin (Hb) β-chain resulting in the production of sickle hemoglobin (HbS) (1). Upon deoxygenation, HbS aggregates due to the formation of multistranded helical polymers, which results in rigid and deformed red blood cells (RBCs) that are prone to lysis. Sickled RBCs can rapidly and permanently obstruct the microvasculature thus decreasing tissue oxygen (O₂) delivery, leading to ischemic injury and painful vasoocclusive crises. The resulting organ damage is a major cause of pain, morbidity, and mortality associated with SCD (1–6). Central nervous system vascular complications are one of the most devastating problems in SCD, where overt stroke or repeated silent cerebral infarcts lead to significant physical and neurocognitive consequences (7–9). In addition, anemia (low Hb concentration) and increased cerebral blood flow in SCD are associated with increased stroke risk, suggesting that cerebral hypoxia is an important factor contributing to SCD morbidity (10).

As polymerization of deoxy-HbS in RBCs drives the pathogenesis of SCD, inhibiting HbS polymerization has proven to reduce the burden of the disease. Therapeutic strategies to reduce HbS polymerization in SCD have been based on increasing the RBCs’ concentration of fetal hemoglobin (HbF), reducing the cellular concentration of HbS, or by chemically modifying HbS (1, 3). Because oxygenated Hb (oxyHb) is a potent inhibitor of deoxygenated HbS (deoxyHbS) polymerization, allosteric modification of Hb to increase the proportion of oxyHb in RBCs is another promising strategy to inhibit HbS polymerization (11). This approach led to the development of voxelotor (GBT440; Oxbryta), which was approved in 2019 by the Food and Drug Administration for the treatment of SCD in patients ages 12 years and older. Voxelotor reversibly binds to Hb and allosterically increases Hb affinity for O₂ (12, 13). Consequently, voxelotor increases the concentration of oxyHb and thereby inhibits deoxyHbS polymerization. Oral administration of voxelotor reduces RBC sickling, extends RBC half-life, and reduces anemia and hemolysis in vivo (12–15). In the Phase III HOPE trial, voxelotor significantly increased Hb levels and reduced markers of hemolysis in patients with SCD compared with placebo (16).

An excessive increase in Hb affinity for O₂ could, in theory, lead to decreased O₂ release from Hb and impose a hypoxic risk to tissues. This would be especially deleterious in the brain, as it is the most energy-demanding and metabolically active organ of the body. Compared with healthy individuals, this theoretical hypoxic risk may be increased in patients with SCD who already suffer from impaired arteriolar vasoregulation and hence cannot compensate by increasing cerebral blood flow (17–19). The Townes transgenic SCD mouse represents a well-established model, whereby murine α and β globin genes have been replaced by human α and β^s globin genes (20). The Townes mice develop hemolytic anemia and severe organ injury consistent with human SCD (20). The purpose of this study was to assess the impact of a pharmacologically mediated increase in Hb affinity for O₂ on O₂ delivery and extraction, cerebral tissue oxygenation during normoxia and hypoxia, and tolerance to severe hypoxia in Townes transgenic SCD mice. To increase Hb affinity for O₂, mice were dosed with 2-hydroxy-6-[(2S)-1-(pyridine-3-carbonyl)piperidin-2yl]methoxy (GBT1118), an analog of voxelotor with the same mechanism of action but with improved pharmacokinetic properties that allow it to achieve in SCD mice the same degree of Hb modification voxelotor targets clinically.

METHODS

Animal Model

Studies were performed in male 8- to 12-wk-old homozygous Townes [B6; 129-Hbatm1(HBA)Tow Hbbtm2(HBG1,HBB)Tow/Hbbtm3(HBG1,HBB)Tow/J] transgenic SCD mice of C57BL/6J background [with both human α and β (β^A and β^s form) globin genes knocked into the mouse locus] obtained from the Jackson Laboratory (Bar Harbor, ME). This SCD mice model mimics the morbidities associated with human homozygous SCD (20). Healthy male 8- to 12-wk-old C57BL/6J mice (The Jackson Laboratory) were used as wild-type (WT) controls. Research was conducted in compliance with the United States Animal Welfare Act and other federal statutes and regulations relating to animals and experiments involving animals and with principles stated in the National Institutes of Health’s Guide for the Care and Use of Laboratory Animals (National Institutes of Health, 2011), and all the experimental protocols were approved by the UC San Diego Institutional Animal Care and Use Committee. The SCD mice were randomly assigned to receive treatment with drug or vehicle.

Drug Treatment

GBT1118 or voxelotor was formulated in 0.5% methylcellulose/phosphate buffer pH 7.4/0.01% Tween-80 at 10 mg/mL. To establish the time–concentration profiles, SCD mice were treated with a single oral dose of 10 mg/kg of either GBT1118 or voxelotor, followed by quantification of compound concentrations in blood sampled at the indicated time points in Fig. 2B. For SCD pathophysiology and tolerance to hypoxia experiments, SCD mice were treated with 100 mg/kg GBT1118 PO twice a day either for 14 or 24 days (as specified in results) before subjected to any procedures. The final dose of GBT1118 was given ∼1 h before terminal procedures (hypoxia tolerance and blood gas measurements). Acutely treated SCD mice were subjected to a single PO dose of 100 mg/kg GBT1118 ∼1 h before being subjected to any procedures. All control mice (SCD and WT) received oral treatment with 10 mL/kg vehicle following the same dosing schedule as their experimental counterparts.

Figure 2.

Comparison of in vitro Hb affinity for O₂-modifying properties and in vivo pharmacokinetic properties of voxelotor and GBT1118. A: in vitro modification of human SCD Hb with voxelotor or GBT1118 results in similar changes in Hb affinity for O₂ at equivalent concentrations. B: time-concentration profiles following single oral doses (10 mg/kg) in SCD mice indicate that GBT1118 has higher exposure and longer half-life compared with voxelotor in SCD mice. C: GBT1118 achieved higher blood concentrations than voxelotor following repeated oral dosing in SCD mice; the dashed line represents the target occupancy. Mean ± SD for n = 20 and n = 5 SCD mice treated with 200 mg/kg PO for 14 days with voxelotor and GBT1118, respectively, and P value was calculated via unpaired Welch’s t test. GBT1118, 2-hydroxy-6-[(2S)-1-(pyridine-3-carbonyl)piperidin-2yl]methoxy; SCD, sickle cell disease.

PK Processing and Data Analysis

GBT1118 and voxelotor concentrations in blood samples were measured using liquid chromatography-tandem mass spectrometry (LC-MS/MS), as previously described (13, 21).

Hb Occupancy

Hb occupancy represents the percentage of Hb molecules occupied by compound, estimated as a molar ratio of test compound to Hb concentrations, assuming a 1:1 Hb to compound binding relationship. Percent Hb (%Hb) occupancy was calculated by dividing the concentration of compound (GBT1118 or voxelotor) in blood by the concentration of Hb in blood (as determined by the percent Hct) multiplied by 100.

Blood O₂ Equilibrium Curve

A Hemox Analyzer (TCS Scientific, New Hope, PA) was used to measure blood O₂ equilibrium curves (OECs). For determination of OECs in human blood collected from subjects with SCD, blood was obtained in sodium citrate vacutainers (3.2% citrate; 15–25 mL) from the University of North Carolina [UNC, Chapel Hill, NC (IRB No. 88-034)]. Blood samples (1 mL each) at 20% Hct were mixed with GBT1118 or voxelotor (100 mM DMSO stocks) in a borosilicate vial to achieve the indicated concentrations shown in Fig. 1A and incubated for 1 h at 37°C under gentle shaking conditions. The samples were then diluted 50-fold into 37°C TES buffer before introduction into the Hemox Analyzer. Similarly, blood samples collected from SCD mice that had been orally treated with GBT1118 at the end of the experimental protocol were diluted 50- to 100-fold into 37°C TES buffer before sampling into the Hemox Analyzer. To measure OECs, the diluted blood samples were first saturated with compressed air (21% O₂) and deoxygenated with pure N₂. Absorbance at isosbestic points for Hb (570 nm) and deoxyHb (560 nm) were recorded as a function of sample Po₂. Sample Po₂ and Hb saturation (sO₂) were recorded to obtain the OEC from which the P50 value was determined.

Figure 1.

Transgenic sickle cell disease (SCD) mice are significantly less tolerant to hypoxia than wild-type (WT) mice. A: the transgenic SCD mice express only human Hb and have a lower P50 than their WT controls. Tolerance to progressive hypoxia (B), changes in mean arterial pressure (MAP; C), and heart rate (HR; D) during tolerance to progressive hypoxia. The number of animals surviving per group, per time point, are displayed underneath the respective boxplots in parentheses at the bottom of the graphs. Survival P value was calculated via the log-rank test. **P < 0.01, ***P < 0.001, ****P < 0.0001, †P < 0.05 vs. 21% ${\text{F}\text{I}}_{{\text{O}}_2}$, ‡P < 0.05 vs. 15% ${\text{F}\text{I}}_{{\text{O}}_2}$, and §P < 0.05 vs. 10% ${\text{F}\text{I}}_{{\text{O}}_2}$ for two-way ANOVA with Tukey’s multiple comparisons test. BPM, beats per minute; HbO2, oxyhemoglobin; P50, Po₂ for 50% HbO₂ saturation.

RBC Parameters and Ex Vivo RBC Sickling

Hb, Hct, and RBC counts were determined using a HemaTrue Hematology Analyzer (Heska, Loveland, CO). To determine percent retics, cells from blood samples collected from SCD mice were washed twice with stain buffer with bovine serum albumin (BSA, No. 554657; BD Biosciences, Franklin Lakes, NJ), stained with Retic-Count (BD Biosciences) and analyzed by flow cytometry.

To determine RBC half-life, 50 mg/kg N-hydroxysuccinimide biotin (ThermoScientific, Waltham, MA) was intravenously injected into SCD mice via tail vein on day 17, producing a pulse label on RBCs. Mice were bled each subsequent day, and blood samples were analyzed by multicolor flow cytometry. Biotinylated Ter119⁺ live singlet scattergated RBCs were identified with fluorescently labeled streptavidin using an LSRII flow cytometer (BD Biosciences). Flow cytometry data for RBC half-life measurements were analyzed using FCS Express (DeNovo Software, Pasadena, CA), and the percentage of biotinylated RBCs was calculated and subsequently plotted using Prism 7 (GraphPad Software, San Diego, CA) to identify the percentage of biotinylated cells remaining over time. RBC half-life was calculated using a plateau followed by an exponential decay model.

For ex vivo sickling measurements, blood samples were collected during the deoxygenation phase in the Hemox Analyzer at 20 mmHg. They were fixed at room temperature for 1 h in 1 mL of 2% glutaraldehyde in PBS previously deoxygenated (by bubbling N₂ for 20–30 s). To determine the percentage of sickled cells, bright-field images (×40 magnification) of sickled RBCs suspended in 2% glutaraldehyde-PBS were quantitated by manually counting the number of sickled cells. Elongated RBCs with tapering of opposite ends that culminated in a point and nondiscoid RBCs were counted as sickled (22). Results were reported as percent sickled, calculated as the number of sickled RBCs divided by the total number of RBCs multiplied by 100. Bright-field images were acquired at 25°C on a Nikon Eclipse TS100 microscope (Nikon, Melville, NY) fitted with an Infinity Lite camera using the Infinity Capture software (both Lumenera Corp., Nepean, ON, Canada).

Hypoxia Protocols

Mice were randomly assigned to one of the two protocols. O₂ concentrations were confirmed during the protocols using a calibrated O₂ sensor (Maxtec, Valley City, UT) and adjusted to maintain desired O₂ concentration within 0.5% of the goal target. For protocol 1, mice were subjected to 15 min of normoxia (21% ${\text{F}\text{I}}_{{\text{O}}_2}$) and hypoxia (10% ${\text{F}\text{I}}_{{\text{O}}_2}$) before measurements were taken. This hypoxia protocol was used to assess changes in CO, blood gases and systemic oxygenation (catheterization), and cerebral oxygenation (microelectrodes). These methods are described in detail below. For protocol 2, mice were exposed to 5% decrements of ${\text{F}\text{I}}_{{\text{O}}_2}$ (normoxia, 15%, 10%, and 5% O₂) for 15 min before measurements were taken. This protocol was used to assess tolerance to hypoxia in catheterized mice. Tolerance to extreme hypoxia (5% O₂) was evaluated over 15 min. Tolerance to hypoxia was defined based on the animal’s ability to maintain MAP above 35 mmHg for 1 min, to ensure humane treatment of animals and following previous studies (21, 23, 24). When pressure drops below, the hypoxia was quickly removed, the experiment was terminated, and mice were euthanized.

Inclusion Criteria

Animals were suitable for the experiments if MAP was >65 mmHg at baseline and HR was >300 beats/min at baseline.

Catheter Implantation

Briefly, mice were anesthetized with isoflurane (5%/volume for induction, 2% for surgical maintenance; Drägerwerk, AG, Lübeck, Germany). After the neck was shaved and cleaned with 70% ethanol and betadine, the mouse was placed on a warm surgical stage and the carotid artery and jugular veins were catheterized. Arterial and venous catheters were filled with a heparinized saline solution (30 IU/mL) to ensure patency. During surgery, animals spontaneously breathed 40% O₂ to ensure appropriate oxygenation during anesthesia and postsurgery. After surgery, isoflurane was reduced to 1.5% for measurements.

Cardiac Output

The cardiac output (CO) was measured using a high-frequency ultrasound imaging system (Vevo 2100, VisualSonics, Inc., Toronto, Canada) with a 30 MHz linear array transducer. During ultrasound imaging, mice were anesthetized using isoflurane at 1%, and body temperature was maintained. Continuous left ventricles and aortic images were used to derive left ventricular stroke volume and cardiac output. The CO was averaged for three to five cardiac cycles.

Cranial Window for Cortical O₂ Measurements

Briefly, a subset of mice (eight mice per group) who were chronically treated for 14 days was used for cortical Po₂ measurements. They were anesthetized with isoflurane (5%/volume for induction, 2% for surgical maintenance). After shaving the head and cleansing with 70% ethanol and betadine, the mice were placed on a stereotaxic frame and the head immobilized. The scalp was removed and lidocaine and epinephrine were applied on the periosteum, which was then retracted exposing the skull. A 4-mm-diameter skull opening was made in the left parietal bone using a surgical drill. Under a drop of saline, the craniotomy was lifted away from the skull with very thin tip forceps and gelfoam previously soaked in saline applied to the dura mater to stop any eventual small bleeding. The pial surface was superfused with physiological solution buffered (including NaCl 118 mM, KCl 4.5 mM, CaCl₂ 2.5 mM, KH₂PO₄ 1.0 mM, MgSO₄ 1.0 mM, NaHCO₃ 25 mM, glucose 6 mM). The solution was saturated with a gas mixture of 5% O₂ and 5% CO₂, balanced with nitrogen the remainder, at a pH of 7.35 at 37°C. After surgery, isoflurane was reduced to 1.5%/volume for cortical O₂ measurements.

Cortical O₂ Measurements

The cortical Po₂ was measured using microelectrodes (OX-10, Unisense, Aarhus, Denmark). Microelectrodes were polarized at −0.8 V relative to a silver-silver chloride reference electrode (Oxy-Meter 1 Ch, Unisense). Po₂ measurements were determined based on the plateau current generated within 20 s. The microelectrodes were calibrated at 37°C with 0%, 5%, 10%, and 21% O₂ gases balanced with N₂ before and after the in vivo measurements. Prior to measurements, the microelectrode tip was placed 500 μm above the tissue to obtain a known Po₂. The cortical Po₂ measurements were made by penetrating a few microns into the tissue with the tip of the microelectrode. A long-working distance ×10 Leitz objective was used to visualize the position of the electrode. All measurements were performed away from microvessels.

Brain Pimonidazole Staining

Hypoxic area staining via pimonidazole was studied in a separate subset of SCD animals (eight mice per group) chronically treated with GBT1118 or vehicle for 14 days. Animals were kept under anesthesia using isoflurane (at 1%) and subjected to normoxia for 15 min, and then subjected to hypoxia at 10% O₂ ${\text{F}\text{I}}_{{\text{O}}_2}$. Pimonidazole was then administered at 10% O₂, and the hypoxia was maintained for 45–60 min before euthanasia. Brains were quickly harvested and snap frozen in liquid nitrogen. Immunohistochemical staining of pimonidazole bound to hypoxic zones in the brain was then performed as previously described (21).

Systemic Parameters and Blood Gases

The MAP and HR were continuously measured from the carotid artery catheter (PowerLab 8/30, AD Instruments, Colorado Springs, CO). Microhematocrit tubes collected from the catheter were used to determine arterial and venous blood gases, total Hb content, lactate, and electrolytes (ABL-90 FLEX, Radiometer, Denmark). The O₂ delivery (DO₂, Eq. 1) and O₂ consumption (V̇o₂, Eq. 2) were calculated from the CO, total Hb, and $\mathrm{S}{\mathrm{a}}_{{\mathrm{O}}_2}$ and $\mathrm{S}{\mathrm{v}}_{{\mathrm{O}}_2}$ as:

$${\text{DO}}_2\text{\hspace{0.17em}=\hspace{0.17em}CO·α·tHb\hspace{0.17em}·\hspace{0.17em}}\mathrm{S}{\mathrm{a}}_{{\mathrm{O}}_2}$$ (1)

$${\dot{\text{V}}}_{{\mathrm{O}}_2}\hspace{0.17em}=\hspace{0.17em}\text{CO}·\alpha ·\text{tHb}·(\mathrm{S}{\mathrm{a}}_{{\mathrm{O}}_2}-\mathrm{S}{\mathrm{v}}_{{\mathrm{O}}_2})$$ (2)

where α is the empirical HbO₂ binding capacity, defined as 1.34 mL of O₂ per gram Hb; tHb is the total Hb; $\mathrm{S}{\mathrm{a}}_{{\mathrm{O}}_2}$ and $\mathrm{S}{\mathrm{v}}_{{\mathrm{O}}_2}$ are, respectively, the measured arterial and venular HbO₂ saturations; and CO is the measured cardiac output. The O₂ extraction ratio was calculated as the ratio between the V̇o₂ and < DO₂ (V̇o₂/DO₂).

Statistics

Results are presented as Tukey’s box-and-whisker plots. Unless otherwise noted, data in the text are presented as means ± SD. All animals included in the analysis passed Grubbs test to confirm closeness for all measured parameters at baseline as inclusion criteria. Data analysis between groups and time points was performed via two-way analysis of variance (ANOVA), with Tukey’s post hoc test when appropriate. Survival analysis was performed by the Kaplan–Meier log-rank test. Analyses between only two groups were performed with Welch’s unpaired t test. All data analysis was performed in R, version 3.6.3. Before experiments were initiated, sample sizes were calculated based on α = 0.05, and power = 0.9 to detect differences between primary end points (MAP, HR, $\mathrm{S}{\mathrm{a}}_{{\mathrm{O}}_2}$). Results were considered statistically significant if P < 0.05. Supplemental figures are available at https://doi.org/10.6084/m9.figshare.13669076.

Study Approval

All studies in mice were approved by the UC San Diego Institutional Animal Use and Care Committee (protocol no. S11306). All studies using human SCD blood were approved by UNC Chapel Hill (IRB No. 88-034).

RESULTS

Tolerance to Hypoxia in SCD Mice and Healthy Controls

As shown in Fig. 1A, Townes transgenic SCD mice have a lower P50 in blood than WT mice (31 mmHg in SCD mice versus 43 mmHg in C57BL/6 mice) (13, 21). In addition, SCD mice are anemic with a hematocrit (Hct) of 35 ± 3.3% (25) and thus have reduced O₂-carrying capacity relative to WT mice, which have an Hct of ∼45% (26). To study the impact of the SCD pathophysiology on tolerance to hypoxia, we subjected Townes SCD mice and their WT counterparts to progressive hypoxia (Fig. 1, B–D). SCD mice were significantly less tolerant to severe (5% O₂) hypoxia compared with WT mice (Fig. 1B). During the hypoxic challenge protocol, although all WT mice survived the full duration of the 10% O₂ hypoxic challenge stage, a significant number of SCD mice succumbed at 10% O₂ hypoxia. Measurements of mean arterial blood pressure (MAP) during the hypoxic challenge indicated that SCD mice possessed a significantly lower MAP at normoxia (21% O₂) and at 15% O₂ hypoxia compared with their WT counterparts (Fig. 1C). In addition, both WT and SCD mice experienced a significant decrease in MAP at 10% and 5% O₂ hypoxia. Similarly, the heart rate (HR) in SCD mice was significantly lower than that in WT mice at all hypoxia levels (Fig. 1D). HR decreased significantly in WT mice at 5% O₂ hypoxia. In contrast, there were not enough SCD mice tolerant to 5% O₂ hypoxia to detect a significant difference in HR (Fig. 1D). These results confirm that the SCD pathophysiology reduces tolerance to hypoxia in SCD mice.

Selection of Compound to Increase Hb Affinity for O₂ in SCD Mice

The clinical application of voxelotor is targeted to an Hb occupancy of ∼30% to ensure protection from RBC sickling based on clinical studies in patients with SCD (15, 16). Voxelotor and its analog GBT1118 increase human Hb affinity for O₂ with similar potency (Fig. 2A). However, unlike in patients with SCD, the pharmacokinetic characteristics of voxelotor in SCD mice indicated that voxelotor could not consistently achieve a targeted Hb occupancy of 30% (Fig. 2B). During chronic oral treatment (100 mg/kg twice a day for 14 days), voxelotor concentration in the blood was only 25% of the concentration of GBT1118 at equal dose (Fig. 2C). This observation is consistent with the relatively short T1/2 of voxelotor (9 h) driven by its rapid clearance in this murine model. As GBT1118 is able to consistently achieve >30% Hb occupancy following oral chronic treatment in SCD mice, it was selected to pharmacologically increase the Hb affinity for O₂ in Townes SCD mice for our studies.

Chronic Treatment of SCD Mice with GBT1118

Effects on SCD pathophysiology in SCD mice.

The hematological impact of chronic GBT1118 treatment in SCD mice is presented in Table 1. Chronic oral treatment with GBT1118 for 24 days achieved blood concentrations equivalent to an Hb occupancy of 43 ± 6.5% (median ± SD) and reduced the P50 of blood from 31 to 18 mmHg, indicating increased Hb affinity for O₂ (Table 1). As a result, GBT1118 reduced ex vivo sickling under hypoxic conditions (20 mmHg), suggesting that by increasing the concentration of oxyHb in RBCs, GBT1118 reduces HbS polymerization. In accordance with its antipolymerization and antisickling activities, GBT1118 reduced hemolysis as demonstrated by the concurrent increase in total Hb, RBC counts, and Hct, as well as reduced reticulocyte counts (Table 1). Consistent with an improvement in RBC health, GBT1118 increased RBC half-life from a median of 1.9 to 3.9 days (Table 1). Together, these results demonstrate that chronic modification of HbS with GBT1118 leads to inhibition of HbS polymerization, while reducing anemia in SCD mice, all consistent with the outcomes observed in patients with SCD chronically treated with voxelotor (15, 16).

Table 1.

Impact of GBT1118 on RBC parameters in SCD mice following chronic dosing

	Vehicle		GBT1118, 200 mg/kg
Parameter	Median	n	Median	N
Blood concentration, µM	N/A		750.5 (100)	4
Hb occupancy, %	N/A		42.9 (6.5)	4
Blood P50, mmHg	30.6		17.6
Ex vivo sickling, %	36.3		18.0
RBC half-life, days	1.9 (0.2)	4	3.9 (0.5)*	3
Hematocrit, %	29.3 (2.8)	3	36.6 (4.7)*	4
Hemoglobin, g/dL	10.6 (0.7)	3	12.9 (1.2)*	4
RBC count, millions/µL	6.6 (1)	3	8.8 (1.2)*	4
Reticulocytes, %	53.3 (6.5)	4	37.8 (0.7)*	4

Values are medians (SD). Transgenic sickle cell disease (SCD) mice were treated with GBT1118 for 24 days. GBT1118, 2-hydroxy-6-[(2S)-1-(pyridine-3-carbonyl)piperidin-2yl]methoxy; n, number of mice; N/A, not applicable; RBC, red blood cell.

*P < 0.05 vs. vehicle for Welch’s t test.

Effects on the tolerance to hypoxia, oxygenation, and acid-base balance.

To determine the effects of chronic dosing of GBT1118 on tolerance to hypoxia, oxygenation and acid-base balance, SCD mice were dosed with GBT1118 for 14 days followed by an acute hypoxic challenge protocol (see methods, Hypoxia Protocols). SCD mice achieved GBT1118 blood concentrations of 802 ± 81 µM corresponding to an Hb occupancy of 44 ± 5% (Fig. 2C), which reduced the P50 of blood from ∼31 mmHg in vehicle-treated SCD mice to 18 mmHg in GBT1118-treated SCD mice (Fig. 3A). Notably, the same Hb occupancy (∼ 44%) was achieved when SCD mice were dosed with GBT1118 for 14 or 24 days, indicating that stable Hb modification with GBT1118 was attained by day 14 of chronic dosing.

Figure 3.

Chronic treatment with GBT1118 significantly increases the Hb affinity for O₂ and improves tolerance to progressive hypoxia in SCD mice. A: chronic treatment with GBT1118 increases the affinity of Hb for O₂. B: chronic treatment with GBT1118 significantly improves survival to extreme hypoxia in SCD mice, beyond the survival level of even WT mice. For reference, the WT group is shown on the survival curve as the red dashed data set. C: mean arterial pressure (MAP) is preserved at lower ${\text{F}\text{I}}_{{\text{O}}_2}$s for SCD mice chronically dosed with GBT1118. D: heart rate (HR) is significantly decreased for vehicle-treated SCD mice but not for GBT1118-treated SCD mice. The number of animals surviving per group, per time point, are displayed underneath the respective boxplots in parentheses at the bottom of the graph. Survival P value was calculated via the log-rank test. *P < 0.05, †P < 0.05 vs. 21% ${\text{F}\text{I}}_{{\text{O}}_2}$, ‡P < 0.05 vs. 15% ${\text{F}\text{I}}_{{\text{O}}_2}$, and §P < 0.05 vs. 10% ${\text{F}\text{I}}_{{\text{O}}_2}$ for two-way ANOVA with Tukey’s multiple comparisons test. BPM, beats per minute; GBT1118, 2-hydroxy-6-[(2S)-1-(pyridine-3-carbonyl)piperidin-2yl]methoxy; HbO₂, oxyhemoglobin; SCD, sickle cell disease; WT, wild type.

Tolerance to hypoxia.

The impact of chronic GBT1118 treatment on tolerance to hypoxia, MAP, and HR in SCD mice are shown in Fig. 3, B–D. Chronic GBT1118 treatment significantly improved tolerance to hypoxia in SCD mice. Although only two vehicle-treated SCD mice survived 120 s of exposure to 5% O₂ hypoxia, all GBT1118-treated SCD mice survived for at least 2 min during the 5% O₂ hypoxic challenge (Fig. 3B). Additionally, GBT1118-treated SCD mice were more tolerant to the hypoxic challenge compared with their WT counterparts (shown as the red dashed data set on the survival curve; Fig. 3B). The MAP of vehicle-treated and GBT1118-treated SCD mice decreased significantly during 10% and 5% O₂ hypoxia compared with normoxia; however, GBT1118-treated SCD mice maintained significantly higher MAP than vehicle-treated controls during 10% and 5% O₂ hypoxia (Fig. 3C). Additionally, although HR was preserved in GBT1118-treated SCD mice during the entire hypoxia protocol, HR significantly decreased for control animals at 5% O₂ hypoxia relative to normoxia (Fig. 3D). These results indicate that increasing the Hb affinity for O₂ with GBT1118 in SCD mice improves tolerance to severe hypoxia and is cardioprotective during severe hypoxia.

Oxygenation.

The impact of chronic GBT1118 treatment on systemic and cerebral oxygenation under normoxia (21% O₂) and 10% O₂ hypoxia is shown in Figs. 4 and 5. Chronic treatment with GBT1118 significantly increased total Hb levels in SCD mice (Fig. 4A). Differences in total Hb between normoxia and hypoxia are likely due to blood sampling (Fig. 4A). GBT1118 treatment increased cardiac output (CO) at normoxia and 10% O₂ hypoxia (Fig. 4B). Arterial and venous Po₂ ($\mathrm{P}{\mathrm{a}}_{\mathrm{O}}{}_{{}_2}$ and ${\text{Pv}}_{{\text{O}}_2}$, respectively) values were similar for GBT1118- and vehicle-treated SCD mice at normoxia, but GBT1118-treated SCD mice maintained higher Pao₂ and ${\text{Pv}}_{{\text{O}}_2}$ at 10% O₂ hypoxia. Additionally, $\mathrm{P}{\mathrm{a}}_{\mathrm{O}}{}_{{}_2}$ and ${\text{Pv}}_{{\text{O}}_2}$ significantly decreased during exposure to 10% O₂ hypoxia for both GBT1118- and vehicle-treated SCD mice (Fig. 4, C and E). GBT1118-treated SCD mice had significantly higher arterial and venous HbO₂ saturation ($\mathrm{S}{\mathrm{a}}_{{\mathrm{O}}_2}$ and $\mathrm{S}{\mathrm{v}}_{{\mathrm{O}}_2}$, respectively) compared with vehicle controls at normoxia and 10% O₂ hypoxia. Although $\mathrm{S}{\mathrm{a}}_{{\mathrm{O}}_2}$ decreased significantly for both groups during 10% O₂ hypoxia, GBT1118-treated animals maintained $\mathrm{S}{\mathrm{a}}_{{\mathrm{O}}_2}$ above 90% even at 10% O₂ hypoxia (Fig. 4D). Additionally, $\mathrm{S}{\mathrm{v}}_{{\mathrm{O}}_2}$ only decreased from 78% to 69% for GBT1118-treated SCD mice during hypoxia, but the $\mathrm{S}{\mathrm{v}}_{{\mathrm{O}}_2}$ in vehicle-treated mice dropped from 46% to 11% during hypoxia (Fig. 4F).

Figure 4.

Chronic treatment with GBT1118 improves blood oxygenation during hypoxia in SCD mice. A: chronic treatment with GBT1118 significantly increases total Hb levels by increasing RBC half-life and reduce sickling. B: GBT1118 significantly increases cardiac output (CO), but CO remains the same during hypoxia. Arterial Po₂ (PaO₂; C) and venous Po₂ (${\text{Pv}}_{{\text{O}}_2}$; E) are similar for both vehicle-treated and GBT1118-treated SCD mice during normoxia, but higher for GBT1118-treated SCD during hypoxia. Additionally, GBT1118 increases arterial SO₂ ($\mathrm{S}{\text{a}}_{{\mathrm{O}}_2}$; D) and venous sO₂ ($\mathrm{S}{\mathrm{v}}_{{\mathrm{O}}_2}$; F) during normoxia and hypoxia. n = 10 animals/group; *P < 0.05, ***P < 0.001, ****P < 0.0001, and †P < 0.05 vs. 21% ${\text{F}\text{I}}_{{\text{O}}_2}$ for two-way ANOVA with Tukey’s multiple comparisons test. GBT1118, 2-hydroxy-6-[(2S)-1-(pyridine-3-carbonyl)piperidin-2yl]methoxy; RBC, red blood cell; SCD, sickle cell disease.

Resulting from the increased total Hb, CO, and $\mathrm{S}{\mathrm{a}}_{{\mathrm{O}}_2}$, GBT1118-treated SCD mice possessed a significantly higher O₂ delivery (DO₂) at normoxia and at 10% O₂ hypoxia compared with vehicle-treated animals (Fig. 5A). GBT1118-treated SCD mice maintained O₂ extraction (V̇o₂) at normoxia and 10% O₂ hypoxia (Fig. 5B). The increase in DO₂ for GBT1118-treated SCD mice allowed for a significantly lower (i.e., more efficient) arterial to venous extraction ratio for GBT1118-treated SCD mice compared with vehicle-treated SCD mice (Fig. 5C). The preserved O₂ delivery:extraction ratio during hypoxia could explain the improved tolerance to hypoxia of GBT1118-treated over vehicle-treated SCD mice, as a higher percentage of HbS remains oxygenated, thus preventing sickling during exposure to hypoxia. Notably, the increased Hb affinity for O₂ induced by GBT1118 did not impair vital tissue Po₂ levels at normoxia and hypoxia, as indicated by cortical Po₂ measurements (Fig. 5, D and E). Although hypoxia significantly decreased cortical Po₂ levels for both GBT1118- and vehicle-treated SCD mice, the GBT1118 group presented a significantly higher cortical Po₂ during hypoxia compared with the control group (Fig. 5, D and E). Additionally, GBT1118 treatment decreased the percentage of hypoxic areas within the brain compared with the control group, as observed by the reduced pimonidazole staining in the treated group compared with the control group (Fig. 5F).

Figure 5.

Chronic treatment with GBT1118 improves oxygen delivery and does not impact tissue oxygen extraction or reduce tissue Po₂ in SCD mice. A: oxygen delivery (DO₂) during normoxia and hypoxia is significantly increased for SCD mice chronically treated with GBT1118 compared with control. B: oxygen extraction (V̇o₂) is not reduced by increasing the Hb affinity for O₂ of over 40% of the Hb. C: the O₂ extraction ratio indicates that most O₂ delivered is extracted in vehicle-treated SCD mice during hypoxia (10% ${\text{F}\text{I}}_{{\text{O}}_2}$), whereas only ∼30% of the O₂ delivered was extracted for the GBT1118 chronically treated SCD mice. D and E: cortical Po₂ levels are significantly decreased during hypoxia but increased Hb affinity for O₂ as chronic GBT1118 treatment does not impair cortical Po₂ during normoxia and helps preserve cortical Po₂ during hypoxia. F: vehicle-treated SCD mice show significantly higher tissue hypoxia in the brain than GBT1118-treated SCD mice following extended hypoxia, as indicated by pimonidazole staining. Data are shown as the percentage of cells positively labeled for pimonidazole in a microscopic field; individual values, as well as the median and IQR are displayed; P value was calculated via unpaired Welch’s t test. n = 10 animals/group (A–C); n = 8 animals/group (D and E); n = 8 animals/group, 40 microscopic fields/animal (F). *P < 0.05, ****P < 0.0001, and †P < 0.05 vs. 21% ${\text{F}\text{I}}_{{\text{O}}_2}$ for two-way ANOVA with Tukey’s multiple comparisons test. GBT1118, 2-hydroxy-6-[(2S)-1-(pyridine-3-carbonyl)piperidin-2yl]methoxy; IQR, interquartile range; SCD, sickle cell disease.

Acid-base and electrolyte balance.

The effects of chronic GBT1118 treatment on blood acid-base balance during normoxia and hypoxia are shown in Supplemental Fig. S1. The reduced arterial pCO₂ in the GBT1118 group reduced the range of systemic respiratory acidosis in the SCD mice via normalizing arterial pH at normoxia and 10% O₂ hypoxia (Supplemental Fig. S1, A and B). Lactate was not different between GBT1118- and vehicle-treated SCD mice at normoxia; however, vehicle-treated SCD mice experienced a significant increase in lactate during hypoxia that GBT1118-treated SCD mice did not experience (Supplemental Fig. S1C). This finding is consistent with the improved DO₂ afforded by GBT1118 treatment leading to reduction in anaerobic metabolism during the hypoxic challenge. Interestingly, GBT1118 prevented changes in acid-base balance and potassium compared with vehicle treatment. Healthy and SCD RBCs exhibit increased electrogenic cation permeability, particularly following deoxygenation, which results in increased plasma potassium (27). Consistent with the increase in Hb affinity for O₂ and the reduction of anaerobic metabolism (Supplemental Fig. S1C), GBT1118 treatment reduced RBC deoxygenation and sickling, preserved DO₂, and associated acid-base and electrolytic disturbances (Supplemental Fig. S1, D–F).

Single-Dose Treatment of SCD Mice with GBT1118

We examined the impact of an increase in Hb affinity for O₂ without increased O₂-carrying capacity by treating SCD mice with a single oral dose of GBT1118 (100 mg/kg) followed by hypoxic challenge. The single dose of GBT1118 had no impact on Hb or Hct but significantly increased Hb affinity for O₂ for at least 4 h (Fig. 6A) and improved survival compared with vehicle controls within this period (Fig. 6B). Although the single dose of GBT1118 slightly increased $\mathrm{S}{\mathrm{a}}_{{\mathrm{O}}_2}$, it significantly improved $\mathrm{S}{\mathrm{v}}_{{\mathrm{O}}_2}$ (Supplemental Fig. S2, A and B) without a significant impact on hemodynamics during progressive hypoxia (Fig. 6, C and D). A single oral dose of GBT1118 did not increase arterial O₂ tension, but it increased venular O₂ tension during normoxia and 10% O₂ hypoxia (Supplemental Fig. S2, C and D). These results suggest that the protective effects of GBT1118 are more immediately related to minimizing hypoxemia and preserving $\mathrm{S}{\mathrm{v}}_{{\mathrm{O}}_2}$, and during chronic treatment, these effects are complemented by the hematological downstream pharmacodynamic effects induced by preventing RBC sickling.

Figure 6.

Acute (single-dose) treatment with GBT1118 improves survival in SCD mice but has minimal effect on hemodynamics. A: a single dose of GBT1118 significantly increases the Hb affinity for O₂ for over 4 h. The P50 decreases from 31.4 ± 0.4 mmHg pretreatment to 21.3 ± 1.1 mmHg 1-h posttreatment, and then slowly increases over 4 h (P50 = 24.1 ± 0.8, 26.9 ± 0.9, and 27.7 ± 0.7 mmHg for 2, 3, and 4 h posttreatment, respectively; data represented as mean ± SD). All comparisons of P50, other than 3 vs. 4 h, were statistically significant following an ANOVA with Tukey’s post hoc test (n = 3 animals/curve). B: the acute dosage of GBT1118 mildly improves survival to extreme hypoxia. C and D: the acute dosage of GBT1118 has minimal impact on hemodynamics. The numbers of animals surviving per group, per time point, are displayed underneath the respective boxplots in parentheses at the bottom of the graphs. Survival P value was calculated via the log-rank test. †P < 0.05 vs. 21% ${\text{F}\text{I}}_{{\text{O}}_2}$, ‡P < 0.05 vs. 15% ${\text{F}\text{I}}_{{\text{O}}_2}$, and §P < 0.05 vs. 10% ${\text{F}\text{I}}_{{\text{O}}_2}$ for two-way ANOVA with Tukey’s multiple comparisons test. GBT1118, 2-hydroxy-6-[(2S)-1-(pyridine-3-carbonyl)piperidin-2yl]methoxy; HbO₂, oxyhemoglobin; HR, heart rate; MAP, mean arterial pressure; P50, Po₂ for 50% HbO₂ saturation; SCD, sickle cell disease.

DISCUSSION

It has been previously reported that pharmacologically increasing Hb affinity for O₂ improves O₂ delivery, reduces tissue hypoxia, decreases lactate and acidemia, and improves survival during hypoxia in healthy animals (21, 23, 24). The impact of increasing Hb affinity for O₂ on O₂ transport, hemodynamics, and tissue oxygenation is poorly understood in SCD due to the pathophysiological changes induced by the disease, where both O₂-carrying capacity and O₂ delivery are compromised. Questions have been raised about whether drugs that increase Hb affinity for O₂ impair O₂ off-loading in tissues in SCD. As expected, chronic treatment with GBT1118 increased RBC count, total Hb, and RBC half-life, while simultaneously decreasing reticulocytes and preventing the sickling of RBCs ex vivo. Increasing Hb affinity for O₂ in a fraction of the Hb decreases polymerization of HbS and relieves SCD mice from hematological consequences resulting from lingering sickling. Importantly, results from this study suggest that the GBT1118-mediated increase in Hb affinity for O₂ at target Hb modification of ∼44% does not affect the release of O₂ or the tissue O₂ tension required to offload physiologically needed O₂. Although a single dose of GBT1118 did not produce measurable hematological changes, it did increase Hb affinity for O₂ and significantly improved tolerance to extreme hypoxia. Lastly, our results demonstrate that pharmacologically increasing the Hb affinity for O₂ does not decrease tissue Po₂ in areas with high physiological O₂ demands, such as the brain, during normoxia and hypoxia. These various pharmacodynamic effects resulted in improved hemodynamics and increased survival during extreme hypoxia. Taken together, these data indicate that the theoretical concern that a drug like voxelotor would compromise tissue O₂ delivery is not supported by the evidence obtained in a SCD mouse model treated with GBT1118 and, in fact, suggest that this mechanism may confer protective effects and a hypoxic protection.

Pharmacologically increasing the Hb affinity for O₂ by treatment with GBT1118 improves the tolerance of SCD mice to hypoxic conditions to a greater extent than untreated WT animals. In addition, pharmacologically increasing the O₂ affinity of 44% of the Hb in SCD mice did not impair cortical tissue Po₂ under normoxia but rather improved cortical tissue Po₂ during hypoxia relative to vehicle-treated SCD mice. GBT1118 treatment reduced metabolic acid-base disruption and prevented electrolytic disturbances, which are observed during deoxygenation of sickle cells by increasing cation permeabilities (K⁺ and Ca²⁺). Thus, GBT1118 prevented red cell dehydration and supported the functioning of cells and organs as these cations determine electrical communication in many systems especially the nervous, cardiovascular, and muscular systems. Consistent with this observation, animals chronically treated with GBT1118 also demonstrated evidence of reduced hypoxia, measured by pimonidazole staining of hypoxic areas within the brain of SCD mice. Additionally, the increased Hb affinity for O₂ led to significantly improved O₂ loading in the lungs during hypoxia increasing $\mathrm{S}{\mathrm{a}}_{{\mathrm{O}}_2}$, and consequently increased the O₂ delivery in SCD mice under hypoxia.

Similar to observations made with voxelotor treatment in patients with SCD (15, 16), GBT1118 treatment in SCD mice prevents sickling of RBCs by decreasing the overall concentration of deoxyHb in the RBCs, thus preventing the formation of HbS polymers. This is known to have a multitude of downstream effects that ultimately improve hematological parameters, blood flow, inflammation, and organ function. For example, the significant increase in Hb (>2 g/dL) observed in this study was a large contributor to the observed increase in O₂ delivery. In addition to the Hb increase, the concurrent reduction in reticulocyte counts and increase in RBC half-life associated with GBT1118 treatment indicates decreased hemolysis and therefore reduced tissue exposure to heme, a known prooxidant. Together, these effects improved $\mathrm{S}{\mathrm{a}}_{{\mathrm{O}}_2}$, blood flow, and O₂ delivery to tissues; however, the increase in Hb and hematocrit cannot entirely explain the improvements in tissue oxygenation and tolerance to hypoxia.

In acute single-dose studies with GBT1118, the increase in the fraction of oxyHb during hypoxia and thus, the consequent inhibition of polymerization is the primary mechanism that improved hypoxia tolerance in SCD mice, because in the acute study, there were no measurable changes in the hematology between GBT1118-treated and vehicle-treated SCD mice. GBT1118 treatment may have also prevented the abnormal activation of the potassium and chloride cotransport system (Gardos channel) in RBCs, which has been proposed to be involved in RBC permeability changes during RBC sickling, as these permeability changes increase RBC dehydration in SCD and aggravate clinical complications (28). This change in RBC permeability may have acted as a secondary mechanism that improved hypoxia tolerance in SCD mice. Chronic increase of Hb affinity for oxygen in SCD mice has additional benefits related to the prevention of RBC sickling and damage to RBC structure and function.

SCD therapies that increase Hb affinity for O₂ reduce RBC sickling and hemolysis and alleviate the comorbidities associated with plasma Hb, heme, and iron, such as vascular inflammation, coagulopathies, and subsequent pulmonary hypertension and infection (29, 30). In this study, we see evidence of decreased pulmonary hypertension, as animals chronically treated with GBT1118 showed significantly increased CO compared with controls during normoxia and hypoxia (31). Novel treatments reducing Hb, heme, and iron toxicity based on hepcidin, hemopexin, and haptoglobin have been experimentally validated to reduce some side effects of SCDs but fail to address the underlying problem that drives hemolysis in SCD (32). Increasing Hb affinity for O₂ in SCD prevents HbS polymerization, thus inhibiting RBC sickling and lysis, which addresses both the underlying origin of SCD comorbidities and prevents the downstream sequalae associated with chronic hemolysis, anemia, and vasoocclusion.

These experimental results have several potential limitations. Results in SCD mice cannot be directly extrapolated to humans, and GBT1118 pharmacokinetics in the SCD mice is different compared with voxelotor’s. In addition, relatively young and healthy SCD mice may be able to adapt and tolerate modification of over 40% of their Hb by GBT1118, but older and sicker animals may have a different response. Cardiovascular and respiratory response to hypoxic hypoxia in rodents allow for increases in cardiac output and respiratory rates and leads to alterations in arterial blood gases. Therefore, direct extrapolation of the degree of protection to hypoxia provided by GBT1118 still needs to be evaluated in larger animals and humans. Arterial blood Pco₂ and ${\text{HCO}}_3^{-}$ were higher in GBT1118-treated groups than in the control group during hypoxia. Analysis of arterial blood gases indicates that the GBT1118 prevented metabolic acidosis and accumulation of lactate. It is important to recognize the measurement of O₂ affinity from blood samples presents the overall affinity for O₂, without distinguishing between the fraction of Hb modified by GBT1118 (or Hb with high affinity) and the unmodified Hb (with normal Hb affinity for O₂). In addition, the measurement of O₂ affinity requires dilution in a strongly buffered solution (TCS Hemox solution, pH 7.3) and deoxygenation is performed in the absence of CO₂, so these measurements do not reflect the Bohr and Haldane effects. Another limitation of the study is that WT mice have a significantly higher P50 (∼42–43 mmHg) compared with SCD mice (p50 of 32 mmHg), challenging the comparison between SCD and WT mice. Although the HbAA-Townes (mice expressing human Hb) mice were created similarly to the HbSS-Townes mice, by replacing murine globin genes with the human globin genes, HbAA-Townes mice do not represent healthy normal mice. To different degrees, the HbAA-Townes and HbSS-Townes mice have differences in their hematology, kidney and liver function, inflammatory markers, haptoglobin and hemopexin levels, red cell half-lives, and organ histopathology compared with healthy mice (33). For example, the HbAA-Townes mice have lower Hb levels, higher reticulocyte counts, and their livers and spleens are significantly enlarged compared with healthy WT mice. In these studies, we used WT mice as control to establish the implications of changes in O₂ affinity in HbSS to physiological responses to hypoxia. It may be valuable to include studies in the HbAA-Townes mouse models, but their results will need to be interpreted within the limitations of the model, further complicating the understanding of results. Lastly, the results presented here are specific to GBT1118 in SCD mice.

Conclusions

The main finding of this study is that in a murine model of SCD, chronic pharmacological increases of Hb affinity for O₂ increases O₂-carrying capacity by raising Hb levels and minimizes hypoxemia at low O₂ concentrations. In addition, it also improves tolerance to hypoxia by preventing RBC sickling, well before inducing any hematological changes. Although SCD mice have substantial differences from patients with SCD, SCD mice provide opportunities to further explore the mechanisms of this disease, and previous clinical studies with voxelotor confirm the pharmacodynamic effects of increasing Hb affinity for O₂ in patients similar to those observed in this SCD mouse study (15, 16). In summary, chronic modification of total Hb resulting in ∼44% high-affinity form not only led to increased O₂-carrying capacity (i.e., increased total Hb) and O₂ delivery but also preserved O₂ extraction by tissues and ultimately improved hypoxia tolerance in SCD mice. The results from these studies support the accumulated evidence to date that in SCD, agents such as voxelotor that increase Hb affinity for O₂ can improve the pathophysiology of SCD by inhibiting HbS polymerization and RBC sickling without compromising tissue oxygenation. Lastly, the physiological balance between O₂ supply and extraction in cerebral, gastrointestinal, coronary, and skeletal muscle tissues is importantly determined by blood flow, and it appears that modifying less than 50% of the Hb with GBT1118 in the SCD mice seems to enable hematological and hemodynamics improvements that preserve oxygenation. We cannot speculate that these conditions can be reproduced in humans and carefully advise not to draw this conclusion for the presented results.

SUPPLEMENTAL DATA

Supplemental Figs. S1-S2: https://doi.org/10.6084/m9.figshare.13669076.

GRANTS

This work was supported by the National Heart, Lung, and Blood Institute Grants R01-HL126945, R01-HL138116, and R01-HL159862. Financial support was also received from Global Blood Therapeutics Incorporated for the completion of the study.

DISCLOSURES

K. Dufu, C. Alt, B. E. Cathers, and D. Oksenberg are employees and shareholders of Global Blood Therapeutics Incorporated. All other authors declare no competing financial interests by the results presented in this report.

AUTHOR CONTRIBUTIONS

K.D., P.C., A.T.W., C.R.M., C.M.W., C.A., B.E.C., and D.O. conceived and designed research; K.D., P.C., A.T.W., C.R.M., C.M.W., A.L., and A.M.E. performed experiments; K.D., P.C., A.T.W., C.R.M., A.L., A.M.E., C.A., B.E.C., and D.O. analyzed data; K.D., P.C., A.T.W., C.R.M., C.A., B.E.C., and D.O. interpreted results of experiments; K.D., P.C., A.T.W., C.R.M., C.A., B.E.C., and D.O. prepared figures; K.D., P.C., A.T.W., C.R.M., A.L., C.A., B.E.C., and D.O. drafted manuscript; K.D., P.C., A.T.W., C.R.M., C.M.W., A.L., A.M.E., C.A., B.E.C., and D.O. edited and revised manuscript; K.D., P.C., A.T.W., C.R.M., C.M.W., A.L., A.M.E., C.A., B.E.C., and D.O. approved final version of manuscript.