ABSTRACT
Pyruvate kinase (PK) is a key enzyme involved in the final step of glycolysis, essential to produce adenosine triphosphate (ATP). Relatively decreased red blood cell (RBC) PK activity (reflected by a lower PK/hexokinase [HK] ratio) and PK thermostability (PK activity after exposure to heat) were recently identified as pathophysiological features of sickle cell disease (SCD). In this study, we investigated whether impaired PK function is associated with sickle RBC properties and SCD‐related clinical manifestations. This study included 97 non‐transfused patients with SCD (88 HbSS, 9 HbS/β0 thalassemia). PK thermostability was correlated with RBC parameters such as reticulocyte count (r = −0.402, p < 0.0001) and hemoglobin F (r = 0.394, p < 0.0001), and indicators of impaired functional properties of sickle RBCs, such as the point of sickling (r = −0.417, p < 0.0001), oxygen affinity (r = 0.408, p < 0.001) and RBC adhesion to laminin (r = −0.322, p = 0.024). Additionally, a low PK/HK ratio correlated with decreased PK thermostability (r = 0.308, p = 0.002), decreased RBC deformability (r = 0.268, p = 0.009), and elevated 2,3‐diphosphoglycerate levels (r = −0.244, p = 0.016). Multivariate Poisson regression analysis demonstrated that reduced PK thermostability and PK/HK ratio were associated with a higher incidence of SCD‐related clinical complications. For every 10‐unit decrease in PK thermostability and 1‐unit decrease in PK/HK ratio, the incidence of SCD‐related clinical complications increased by 11% (p = 0.012) and 10% (p = 0.019), respectively. Altogether, these findings indicate that impaired PK function is related to compromised sickle RBC properties and SCD‐related clinical manifestations. This supports the relevance and underlines the potential of PK activation therapy.
Keywords: pyruvate kinase, red cells metabolism, sickle cell disease
1. Introduction
Sickle cell disease (SCD) is caused by a mutation in the β‐globin gene that leads to the formation of abnormal hemoglobin S (HbS). Upon deoxygenation, red blood cells (RBCs) from patients with SCD sickle due to polymerization and are less deformable. This results in multiple downstream effects, including increased RBC adhesion, microvascular occlusion, and chronic hemolysis [1]. SCD patients experience a broad range of clinical complications, including vaso‐occlusive episodes (VOEs) causing acute pain, as well as chronic multiorgan damage, such as cerebral infarction, retinopathy, and nephropathy [1, 2, 3]. RBCs rely solely on anaerobic glycolysis to produce adenosine triphosphate (ATP), as they lack a nucleus. ATP is crucial for the RBC to maintain the cell's structure and function as less ATP content in RBCs leads to increased hemolysis [4, 5, 6, 7]. Pyruvate kinase (PK) is a key enzyme involved in the final step of glycolysis and responsible for approximately 50% of the cell's ATP production. Studies show that modulation of ATP content enhances RBC health by improving deformability and ATP‐dependent antioxidant mechanisms [8, 9]. PK deficiency (PKD), caused by a mutation in the PKLR gene, is the most common enzyme deficiency leading to nonspherocytic hemolytic anemia. The mutated PKR decreases PK activity and stability, which leads to ineffective glycolysis, causing insufficient ATP production [4]. Furthermore, the decrease in PK activity is accompanied by elevated levels of 2,3‐diphosphoglycerate (2,3‐DPG): a glycolytic intermediate upstream of PK, which plays a pivotal role in the regulation of the oxygen affinity of hemoglobin. Patients with sickle cell trait (HbAS) typically do not experience clinical complications. However, when PKD is co‐inherited with sickle cell trait, it may induce RBC sickling, resulting in a phenotype resembling SCD [10, 11]. Elevated levels of 2,3‐DPG decrease the oxygen affinity of hemoglobin and thus increase the polymerization rate of HbS, thereby facilitating RBC sickling [12]. A decrease in relative PK activity (reflected by a lower PK/hexokinase [HK] ratio) and stability, with the absence of PKLR gene mutations, has been recently observed in patients with SCD, establishing these PK properties as a pathophysiological feature [13].
Currently, activation of PK is being explored as a new therapeutic approach for patients with SCD [14]. Several ongoing clinical trials investigate the safety and efficacy of PK activation by small molecules such as mitapivat (NL8517, NCT04000165, NCT05031780), AG‐946 (NCT04536792), and etavopivat (NCT04624659, NCT04987489). Results from clinical trials on mitapivat and etavopivat show an increase in hemoglobin levels, a decrease in hemolytic markers, and an increased ATP/2,3‐DPG ratio, which is accompanied by an improvement in oxygen affinity [15, 16, 17]. In a phase 2 study of 10 patients on mitapivat, a significant reduction in VOE frequency was observed when combining results of the 8‐week dose‐finding period and the 1‐year fixed‐dose extension period compared to baseline (from 1.33 to 0.64 events, p = 0.049) [15]. Additionally, a non‐significant trend toward a reduction in annualized VOE frequency was observed in the randomized phase 2 study involving 79 patients [18]. A phase 3 study is currently ongoing to further investigate mitapivat's effectiveness (NCT05031780). The results from clinical trials underline that targeting PK is a potentially important approach for the treatment of SCD. In this study, we aim to correlate the PK activity and stability with RBC parameters and functional properties of the sickle RBCs as well as the history of acute and chronic SCD‐related clinical manifestations in order to enhance our understanding of the clinical significance of PK activators in the treatment of SCD.
2. Materials and Methods
2.1. Study Population
Homozygous HbS (HbSS) and HbS/β0 thalassemia patients, who visited outpatient clinics as part of the Sickle Cell Outcome Research (SCORE) consortium in the Netherlands (Erasmus University Medical Center [UMC], Amsterdam UMC, Utrecht UMC, and Radboud UMC), were eligible to participate. Both children > 3 years and adults were eligible. Patients who received a blood transfusion within 3 months of inclusion were excluded. Patients were allowed to use hydroxyurea treatment or other disease‐modifying drugs, such as voxelotor or crizanlizumab. EDTA blood was collected from patients in a steady state condition (i.e., not during a hospital admission and/or VOE, and the dosages of SCD‐related drugs remained unchanged for 6 months prior to inclusion). Study procedures were approved by the local medical ethical committees in accordance with the Declaration of Helsinki (protocol number: 17/392). Blood samples from healthy controls were obtained by the Mini Donor Service, a blood donation facility at the University Medical Center Utrecht, the Netherlands (protocol number: 18/774). Blood samples were stored at 4°C, and laboratory measurements (Figure S1) were performed within 24 h.
Medical records were reviewed to obtain information on patient characteristics, current treatment status, and history of SCD‐related complications (e.g., VOEs, acute chest syndrome [ACS], cerebrovascular accident [CVA], proliferative retinopathy, osteonecrosis, chronic kidney disease [CKD], leg ulcers, and cholelithiasis) and need for acute exchange transfusion(s) or intensive care unit admission due to a SCD‐related complication. VOEs were defined as acute episodes of pain, with no other medically determined cause than a VOE resulting in a visit to a medical facility (e.g., hospital admission, emergency department or outpatient clinic visit). VOEs were only included if they occurred within the last 2 years. All other complications over the patient's lifetime were included. Osteonecrosis was diagnosed based on the combination of physical examination and magnetic resonance imaging results. Patients with a history of cholecystectomy for cholelithiasis were also classified as having cholelithiasis. An increased risk of CKD was defined based on the combination of the urine albumin to creatinine ratio (ACR, > 3.4 mg/mmol) and estimated glomerular filtration rate (eGFR) calculated using the CKD‐EPI formula [19]. Figure S2 provided an overview of the complications. The clinical complication rate was defined as the total number of different SCD‐related complications, with each complication counting as one point (e.g., three VOEs in the past 24 months would equal point). Acute complications included VOEs, ACS, CVA, the need for acute exchange transfusion, or intensive care unit admission. Chronic complications included proliferative retinopathy, osteonecrosis, increased risk of CKD, leg ulcers, and cholelithiasis.
2.2. Hematological Parameters
Routine laboratory parameters (Figure S1A), including complete blood count, were measured using the Abbott Cell‐Dyn Sapphire hematology analyzer (Abbott Diagnostics). The percentage of total hypochromic RBCs (RBCs with hemoglobin concentration < 28 g/dL and mean corpuscular volume [MCV] < 120 fL) and total dense RBCs (RBCs with hemoglobin concentration > 41 g/dL MCV < 120 fL) were obtained using the Advia 120/2120 hematology analyzer (Siemens Healthcare diagnostics). High‐performance liquid chromatography was performed to analyze the levels of normal hemoglobin (HbA), fetal hemoglobin (HbF), and HbS (Tosoh G8).
2.3. Pyruvate Kinase Measurements
RBCs were isolated from whole blood using microcrystalline cellulose α‐cellulose columns [20]. PK and HK activity and PK thermostability were measured on purified RBCs (Figure S1B,C) [21]. PK thermostability (expressed as % residual PK activity) was performed as described on lysates after 1 h of incubation (53°C) [22]. Protein levels of PKR and PKM2 were determined by Western Blot analysis (Figure S1D) using polyclonal antibodies against PKRL (gifted by the late Dr. Kahn) and monoclonal antibodies against PKM2 (PodiCeps). Actin was used as a loading control (Calbiochem). Alexa Fluor 680‐conjugated goat anti‐mouse IgG/IgM (Life) and IRDye 800‐conjugated goat anti‐rabbit IgG (Licor) were used as detection antibodies. THP1 cell lysate (sc‐2238) was used as a positive control for PKM2 protein detection, and healthy control RBC lysate was used as a positive control for PKR protein detection. PKR, PKM2, and actin bands were quantified using Odyssey M and Empiria Studio 2.1.
2.4. Oxygen Affinity of Hemoglobin (p50) and 2,3‐DPG/ATP Levels
Oxygen affinity (p50) was measured with the Hemox‐Analyzer (TCS) and reflects the oxygen tension at which 50% of the measurable reach of hemoglobin is saturated with oxygen (Figure S1D). Levels of 2,3‐DPG and ATP were measured using liquid chromatography tandem mass spectrometry after snap‐freezing whole blood in liquid nitrogen (Figure S1E) [23]. Metabolite levels were normalized to hematocrit.
2.5. Oxygen and Osmotic Gradient Ektacytometry
Oxygen and osmotic gradient ektacytometry (oxygenscan and osmoscan, respectively) were performed to measure RBC deformability by the Laser‐Optical Rotational Red Cell Analyzer (Lorrca, RR Mechatronics) as previously described [24, 25, 26]. Figure S1F shows representative curves of the oxygenscan and osmoscan. Oxygen gradient ektacytometry measures RBC deformability upon deoxygenation. The point of sickling (PoS) is the partial oxygen pressure (pO2) at a 5% decrease from the EImax reflecting the RBC sickling tendency. The minimum EI (EImin) represents the minimal deformability upon deoxygenation. Osmotic gradient ektacytometry measures RBC deformability, expressed as the elongation index (EI), under an osmotic gradient. Key parameters derived from osmotic gradient ektacytometry are the maximum EI (EImax), reflecting the maximal deformability, and the Ohyper (osmolality in the hyper‐osmolar region at 50% of the EImax) reflecting the RBCs hydration status.
2.6. RBC Adhesion to Laminin Assay
RBC adhesion to laminin was measured using a microfluidic device (IBIDI μ‐Slide I 0.4 uncoated, Figure S1G). IBIDI slides were incubated for 2 h at 37°C with 6.7 μg/mL laminin (BioLamina). Whole blood (with a standardized concentration of 64 × 106 RBCs/mL) was diluted in N‐2‐hydroxyethylpiperazine‐N′‐2‐ethanesulfonic acid (HEPES) buffer (osmolality between 280 and 295 mOsm/kg and final pH of 7.4) with bovine serum albumin (0.5% final concentration) and glucose (5.55 mM final concentration). The sample was perfused at room temperature using a syringe pump (Harvard Apparatus PHD Ultra) at a shear stress of 0.5 dyn/cm2. After 10 min, the IBIDI slide was washed with HEPES buffer for 2 min, and images were taken at nine different places using the Zeiss Observer Z1 microscope (×20 objective). The number of adherent cells was counted using ImageJ (Fiji).
2.7. Density Separation
To explore whether RBCs with varying densities exhibit different properties, density separation with Percoll (GE Healthcare) was performed in a subset of patients with SCD [27]. Isolated RBCs were added to the columns containing the three layers with different percentages of Percoll (59%, 70%, and 78%) including HEPES, NaCl, KCl, and NaOH [27]. After centrifugation (15 min, 1665g, no brake) the four different RBC fractions were collected and washed two times with sodium chloride 0.9% to remove the Percoll. Fraction 1, on top of the 59% layer, contains RBCs with the lowest density, and is therefore usually enriched with reticulocytes. Fraction 2 was collected on top of the 70% layer, and Fraction 3 on top of the 78% layer. Fraction 4, on the bottom of the 78% layer, contains the most dense RBCs. Afterwards, the fractions were measured with a hematology analyzer as described above, and enzymatic and RBC functional measurements were performed.
2.8. Statistical Analysis
Statistical analysis was performed in R Studio (v4.4.0) and GraphPad Prism (v10.3.1). Continuous laboratory parameters were presented as the means with standard deviation (SD) or median with interquartile range (IQR), depending on the data distribution. The Shapiro–Wilk normality test was performed to check if the residuals were normally distributed. Continuous values between patients with SCD and the control group were compared using the unpaired t‐test or Mann–Whitney test, depending on the data distribution. Friedman test or mixed effects analysis (in case of random missing values) with respectively Dunn's or Tukey's multiple comparisons test were used to compare results of the RBC density separation. A Spearman rank correlation coefficient was used for the correlation between laboratory parameters and enzyme activity measurements. Multivariate Poisson regression analysis was used for the complication rate (count data) to calculate the rate ratio of PK properties, adjusted for age, sex, and hydroxyurea treatment, in relation to the complication rate. Logistic regression analysis was performed to assess odds ratios (ORs) for complications associated with the PK/HK ratio and PK thermostability, adjusted for age, sex, and hydroxyurea treatment. A p ≤ 0.05 was considered statistically significant.
3. Results
Ninety‐seven patients with SCD (88 HbSS, 9 HbS/β0 thalassemia; n = 66 Erasmus UMC, n = 15 Amsterdam UMC, n = 9 Utrecht UMC, and n = 7 Radboud UMC) were enrolled and blood samples were retrieved from 20 healthy controls. Of the SCD patients, 22 were children (median age 13 years [range 6–17]) and 75 were adults (median age 31 years [range 18–58]). Sixty‐seven (69.1%) patients were treated with hydroxyurea. Four patients received crizanlizumab, and two patients received voxelotor in addition to hydroxyurea. One patient was treated with voxelotor only. In the past 24 months prior to inclusion, 55 patients with SCD (56.7%) experienced at least one VOE, with a mean of 3.3 VOEs per patient (SD 3.1). ACS occurred at least once in 26 patients (26.8%) during their lifetime, with a mean of 1.4 events (SD 0.5). Further details on patient characteristics, SCD‐related treatments, and SCD‐related complications are listed in Table S1.
3.1. Pyruvate Kinase Activity and Stability Are Compromised in RBCs From Patients With SCD
Patients with SCD showed an increased absolute PK (median of 13.2 [IQR 11.1–15.0] vs. 7.7 [IQR 6.7–8.5]) and HK activity (median 2.4 [IQR 1.9–2.7] vs. 0.8 [IQR 0.7–0.9]) compared to healthy controls (Table 1). This increase can most likely be attributed to reticulocytosis (median 225.0 × 109/L in SCD patients compared to 59.8 × 109/L in healthy controls), since PK and HK are both RBC age dependent [28]. When evaluating PK activity to that of HK activity, a decreased PK/HK ratio was observed (median of 5.7 [IQR 5.0–6.4] in SCD patients compared to a median of 9.4 [IQR 8.5–10.2] in healthy controls), indicating that the PK activity is decreased in light of the mean RBC age. Median PK thermostability was 74.8% (IQR 66.6–81.5) in patients with SCD compared to 92.2% (IQR 88.7–95.7) in healthy controls. These results confirm our previous findings of decreased relative PK activity and compromised PK thermostability in patients with SCD [13]. When stratifying patients based on hydroxyurea treatment, those who were treated with hydroxyurea demonstrated significantly higher absolute PK activity (Table S2). However, no significant differences were observed in the PK/HK ratio and PK thermostability between SCD patients treated with hydroxyurea and those who were not.
TABLE 1.
Laboratory characteristics of SCD patients compared to healthy controls.
| SCD patients (n = 97) | Healthy controls (n = 20) | p | |
|---|---|---|---|
| RBC parameters | |||
| Hemoglobin (g/dL) | 9.1 (8.0–10.2) | 13.8 (13.4–14.7) | < 0.001 |
| RBC count (1012/L) | 3.0 (2.5–3.5) | 4.7 (4.5–4.9) | < 0.001 |
| Hematocrit (%) | 27.3 (23.7–30.6) | 41.6 (40.1–42.9) | < 0.001 |
| MCV (fL) | 89.4 (78.1–99.3) | 88.9 (87.2–92.0) | 0.871 |
| MCHC (g/dL) | 34.0 (32.9–35.1) | 33.6 (32.9–34.2) | 0.164 |
| Reticulocyte count (109/L) | 225.0 (156.0–309.0) | 59.8 (51.5–75.1) | < 0.001 |
| Hypochromic RBCs (%) a | 3.5 (2.2–6.2) | 0.2 (0.1–0.4) | < 0.001 |
| Dense RBCs (%) a | 4.4 (2.9–6.5) | 0.7 (0.4–1.1) | < 0.001 |
| Hemoglobin F (%) | 9.4 (5.9–18.3) | 0.6 (0.5–0.7) | < 0.001 |
| Hemoglobin S (%) | 77.1 (68.8–81.6) | 0.0 (0.0–0.0) | < 0.001 |
| Enzymatic properties | |||
| PK activity (U/gHb) | 13.2 (11.1–15.0) | 7.7 (6.7–8.5) | < 0.001 |
| HK activity (U/gHb) | 2.4 (1.9–2.7) | 0.8 (0.7–0.9) | < 0.001 |
| PK/HK ratio | 5.7 (5.0–6.4) | 9.4 (8.5–10.2) | < 0.001 |
| PK thermostability (% residual activity) | 74.8 (66.6–81.5) | 92.2 (88.7–95.7) | < 0.001 |
| Metabolic properties | |||
| ATP (mg/L RBCs) | 916.5 (836.5–1033.5) | 850.6 (790.9–910.2) | 0.009 |
| 2,3‐DPG (mg/L RBCs) | 3460 (3275–3782) | 2434 (2316–2676) | < 0.001 |
| ATP/2,3‐DPG ratio | 0.27 (0.23–0.29) | 0.34 (0.27–0.40) | < 0.001 |
| Functional RBC parameters | |||
| p50 (mmHg) b | 25.7 (24.3–27.4) | 24.2 (23.5–25.1) | 0.001 |
| EImin (EI) c | 0.04 (0.02–0.10) | 0.62 (0.61–0.62) | < 0.001 |
| EImax (EI) c | 0.51 (0.48–0.54) | 0.59 (0.58–0.59) | < 0.001 |
| Point of sickling (mmHg) | 40.7 (33.6–47.8) | NA | NA |
| Ohyper (mOsm/kg) | 380.0 (368.0–393.0) | 444.5 (441.0–453.3) | < 0.001 |
| RBC adhesion to laminin (n) d | 12.0 (9.0–18.9) | 4.9 (3.8–5.3) | < 0.001 |
Note: Numbers represent median (IQR).
Abbreviations: 2,3‐DPG, 2,3‐diphosphoglycerate; ATP, adenosine triphosphate; HK, hexokinase; MCHC, mean corpuscular hemoglobin concentration; MCV, mean corpuscular volume; PK, pyruvate kinase; RBC, red blood cell.
Percentage of hypochromic and dense RBCs is missing in 12 (12.4%) of the SCD patients and 3 (15.0%) of the healthy controls.
p50 is missing in 20 (20.6%) of the SCD patients.
EImin is derived from the oxygenscan, and the EImax is derived from osmoscan.
RBC adhesion to laminin is missing in 48 (49.5%) of the SCD patients and 6 (30.0%) of the healthy controls.
PKR protein was detected in all patients with SCD, while PKM2 was either absent or expressed at very low levels (Figure S3). PKR protein levels (assessed as the normalized PKR‐to‐actin ratio) in SCD patients showed a significant positive, though weak, correlation with PK activity (r = 0.25, p = 0.02) and PK HK ratio (r = 0.24, p = 0.03), not with PK thermostability (r = 0.11, p = 0.303).
3.2. RBCs From SCD Patients Have a Lower Oxygen Affinity, Reduced Deformability, and Increased Dehydration and Adhesion Compared to RBCs From Healthy Controls
SCD patients have significantly increased ATP and 2,3‐DPG levels (Table 1). Median ATP levels were 7.7% higher compared to healthy controls (916.5 [IQR 836.5–1033.5] vs. 850.6 [790.9–910.2]), whereas median 2,3‐DPG levels were 42.2% higher compared to healthy controls (3460 [3275–3782] vs. 2434 [2316–2676]), resulting in a decreased ATP/2,3‐DPG ratio in SCD patients (median 0.27 [IQR 0.23–0.29] vs. 0.34 [0.27–0.40]). In line with this increase in 2,3‐DPG, increased p50 values were observed in patients with SCD reflecting a lower oxygen affinity: median 25.7 (IQR 24.3–27.4) versus 24.2 (IQR 23.5–25.1) in controls. RBCs from patients with SCD showed decreased deformability in both oxygenscan and osmoscan analyses (Table 1 and Figure S4), with lower EImin in the oxygenscan (median 0.04 [IQR 0.02–0.10] vs. 0.62 [0.61–0.62], p < 0.001) and lower EImax in the osmoscan (median 0.51 [IQR 0.48–0.54] vs. 0.59 [0.58–0.59], p < 0.001) compared to RBCs of healthy controls. Upon deoxygenation, the median PoS was 40.70 mmHg (IQR 33.6–47.8). Furthermore, RBCs from patients with SCD are more dehydrated, as indicated by a significantly decreased Ohyper (median 380.0 [IQR 368.0–393.0] vs. 444.5 [IQR 441.0–453.3], p < 0.001). Additionally, these RBCs exhibit increased adhesion to laminin (Table 1): 12.0 RBCs (IQR 9.0–18.9) in SCD versus 5.4 RBCs (SD 2.8) in healthy controls, p < 0.0001.
3.3. Pyruvate Kinase Thermostability and PK/HK Ratio Are Associated With RBC Properties
When examining PK thermostability in the correlation analysis (Figure 1, Figure S5), we observed that patients with SCD who had decreased PK thermostability exhibited significantly higher absolute reticulocyte counts (r = −0.402, p < 0.0001) and % dense RBCs (r = 0.368, p = 0.0005), lower hemoglobin (r = 0.262, p = 0.010) and HbF (r = 0.394, p < 0.0001) levels, and increased RBC adhesion to laminin (r = −0.322, p = 0.024). Further, an inverse correlation of PK thermostability with p50 (r = −0.408, p < 0.001) and PoS (r = −0.417, p < 0.0001), and a positive correlation with EImax (r = 0.344, p = 0.001) and EImin (r = 0.449, p < 0.0001) was observed. These results imply that patients with decreased PK thermostability had less deformable RBCs, which sickled at a higher oxygen tension. This suggests that an increase in PK thermostability is beneficial for sickle RBCs. We found no significant correlation between PK thermostability and 2,3‐DPG (r = −0.155, p = 0.131). Further, no correlation of Ohyper (marker of RBC dehydration) with PK thermostability (r = 0.093, p = 0.373) was observed, which suggests that RBCs with reduced PK thermostability are not more dehydrated, as assessed by osmotic gradient ektacytometry.
FIGURE 1.
Linear regression with Spearman correlation coefficient of PK thermostability measurements and laboratory parameters. PK thermostability, expressed as the residual PK activity after 60 min of incubation at 53°C, is significantly correlated with reticulocyte count, hemoglobin, dense RBCs, and hemoglobin F (HbF). Additionally, patients with low PK thermostability showed increased RBC adhesion to laminin, reduced oxygen affinity (reflected by a higher p50), a tendency to sickle at higher oxygen tensions upon deoxygenation (reflected by a higher point of sickling) and decreased deformability (reflected by lower EImax and EImin values).
Since PK and HK activity are correlated with red cell age (correlation PK and HK activity and reticulocytes: r = 0.452 and r = 0.505, respectively, both p < 0.0001), we investigated whether the PK/HK ratio was correlated with RBC properties. Only significant correlations with PK thermostability (r = 0.308, p = 0.002), 2,3‐DPG (r = −0.244, p = 0.016) and EImax (r = 0.268, p = 0.009) were observed, indicating that patients with a higher PK/HK ratio had increased PK thermostability and RBC deformability and lower 2,3‐DPG levels. No significant correlation was observed between the PK/HK ratio and hemoglobin levels (r = 0.167, p = 0.101) or reticulocyte count (r = −0.169, p = 0.098).
3.4. Increase in RBC Density Is Associated With a Decrease in Enzyme Activity
Since PK activity and PK thermostability are differentially correlated with reticulocyte count and dense RBCs (Figure S5), we performed RBC density separation of blood samples from five HbSS patients to explore these findings in more detail (Figure 2). Figure 2A displays the density separation columns of five patients with HbSS showing a heterogeneous RBC distribution among patients. The increase in RBC density is reflected by an increase in mean corpuscular hemoglobin concentration (MCHC) (Figure 2B) and a decrease in reticulocyte levels (Figure 2C). In line with the decrease in reticulocytes, we observed a significant decrease in PK and HK activity (Figure 2D,E). The PK/HK ratio was significantly higher in Fraction 4 compared to the total RBCs, primarily because HK decreased more substantially than PK with increasing density (Figure 2F). PK thermostability was not significantly different compared to the total of all fractions (Figure 2G). Further, the densest RBCs (Fraction 4) exhibited the most strongly reduced deformability (EI) and highest PoS, indicating that they sickle earlier upon deoxygenation (Figure 2H,I).
FIGURE 2.
Enzymatic measurements and oxygen gradient ektacytometry curves of RBCs after density separation. Bars represent median values with interquartile range. Only significant comparisons were shown. All other comparisons between fractions were nonsignificant. Enzyme activity measurements in Fraction 1 were not feasible due to an insufficient volume. (A) Percoll gradient density separation of five different patients with HbSS. Fraction 1 contains RBCs with the lowest density, and Fraction 4 contains RBCs with the highest density. The distribution of fractions differs between patients. (B) The mean corpuscular hemoglobin concentration (MCHC) increase reflects the increase in density. (C) Reticulocyte levels were the highest in the low dense fraction; however, these results were not statistically significant. (D, E) Enzymatic pyruvate kinase (PK) and hexokinase (HK) activity decreases with increasing density. (F) PK/HK ratio is significantly higher in Fraction 4 compared to the PK/HK ratio of the total RBCs. (G) PK thermostability is not significantly different compared to the total of all fractions. (H, I) RBCs with the highest density have the lowest deformability and sickle earlier upon deoxygenation. [Color figure can be viewed at wileyonlinelibrary.com]
3.5. Patients With SCD and Impaired Pyruvate Kinase Function Have More Clinical Complications
To identify whether PK thermostability and PK/HK ratio are related to clinical severity, a clinical complication rate was determined. Results of multivariate Poisson regression analysis (Table 2) showed a significant relationship between PK thermostability and total complication rate (rate ratio = 0.89, p = 0.012) indicating that for every 10‐unit increase in PK thermostability, the complication rate decreases by 11%. Furthermore, this analysis showed that for every unit increase in PK/HK ratio, the total complication rate decreases by 10% (rate ratio = 0.90, p = 0.019). When complications were categorized into acute and chronic, significant associations were observed between PK thermostability and the chronic complication rate (rate ratio 0.84, p = 0.014) and PK/HK ratio and the acute complication rate (rate ratio 0.85, p = 0.016).
TABLE 2.
Univariate and multivariate Poisson and logistic regression analysis show significant effect of PK thermostability and PK/HK ratio on clinical complications in SCD.
| PK thermostability, for every 10‐unit increase | ||||||
|---|---|---|---|---|---|---|
| Univariate Poisson's regression | Multivariate Poisson's regressiona | |||||
| RR | 95% CI | p | RR | 95% CI | p | |
| Total complication rate | 0.85 | (0.77–0.94) | 0.001 | 0.89 | (0.81–0.97) | 0.012 |
| Acute complication rate | 0.89 | (0.78–1.02) | 0.081 | 0.92 | (0.80–1.05) | 0.194 |
| Chronic complication rate | 0.80 | (0.70–0.93) | 0.003 | 0.84 | (0.73–0.97) | 0.014 |
| Univariate logistic regression | Multivariate logistic regressiona | |||||
|---|---|---|---|---|---|---|
| OR | 95% CI | p | OR | 95% CI | p | |
| > 1 any complication | 0.45 | (0.28–0.67) | < 0.001 | 0.39 | (0.22–0.64) | < 0.001 |
| > 1 acute complications | 0.78 | (0.58–1.09) | 0.147 | 0.81 | (0.56–1.14) | 0.226 |
| > 1 chronic complication | 0.58 | (0.39–0.82) | 0.004 | 0.59 | (0.39–0.86) | 0.007 |
| VOE ≥ 2 < 24 months | 0.69 | (0.49–0.96) | 0.033 | 0.71 | (0.49–0.99) | 0.048 |
| ACS | 0.88 | (0.62–1.25) | 0.463 | 0.91 | (0.63–1.31) | 0.622 |
| CVA | 0.64 | (0.37–1.09) | 0.100 | 0.72 | (0.41–1.23) | 0.239 |
| Acute exchange transfusion | 0.71 | (0.49–1.00) | 0.057 | 0.73 | (0.50–1.05) | 0.096 |
| Intensive care unit admission | 1.04 | (0.65–1.71) | 0.875 | 1.11 | (0.68–1.85) | 0.683 |
| Proliferative retinopathy | 0.83 | (0.56–1.22) | 0.342 | 0.92 | (0.61–1.39) | 0.686 |
| Osteonecrosis | 0.76 | (0.50–1.14) | 0.180 | 0.78 | (0.51–1.16) | 0.219 |
| Increased risk CKD | 0.76 | (0.52–1.10) | 0.150 | 0.82 | (0.55–1.23) | 0.343 |
| Leg ulcer | 0.93 | (0.43–2.11) | 0.844 | 0.91 | (0.44–1.93) | 0.795 |
| Cholelithiasis | 0.49 | (0.31–0.74) | 0.001 | 0.47 | (0.29–0.73) | 0.001 |
| PK/HK ratio, for every unit increase | ||||||
|---|---|---|---|---|---|---|
| Univariate Poisson's regression | Multivariate Poisson's regression a | |||||
| RR | 95% CI | p | RR | 95% CI | p | |
| Total complication rate | 0.91 | (0.83–1.00) | 0.050 | 0.90 | (0.82–0.98) | 0.019 |
| Acute complication rate | 0.86 | (0.75–0.97) | 0.020 | 0.85 | (0.75–0.97) | 0.016 |
| Chronic complication rate | 0.95 | (0.83–1.08) | 0.445 | 0.92 | (0.81–1.04) | 0.216 |
| Univariate logistic regression | Multivariate logistic regression a | |||||
|---|---|---|---|---|---|---|
| OR | 95% CI | p | OR | 95% CI | p | |
| > 1 any complication | 0.82 | (0.62–1.08 | 0.165 | 0.70 | (0.50–0.96) | 0.030 |
| > 1 acute complications | 0.64 | (0.43–0.90) | 0.018 | 0.60 | (0.39–0.86) | 0.011 |
| > 1 chronic complication | 1.00 | (0.75–1.31) | 0.999 | 0.93 | (0.68–1.23) | 0.602 |
| VOE ≥ 2 < 24 months | 0.66 | (0.45–0.91) | 0.021 | 0.64 | (0.43–0.89) | 0.015 |
| ACS | 0.94 | (0.68–1.25) | 0.690 | 0.92 | (0.66–1.24) | 0.610 |
| CVA | 0.47 | (0.22–0.88) | 0.029 | 0.47 | (0.21–0.90) | 0.042 |
| Acute exchange transfusion | 0.88 | (0.62–1.19) | 0.435 | 0.86 | (0.60–1.16) | 0.353 |
| Intensive care unit admission | 0.80 | (0.48–1.22) | 0.360 | 0.79 | (0.47–1.20) | 0.335 |
| Proliferative retinopathy | 1.25 | (0.92–1.71) | 0.142 | 1.21 | (0.87–1.71) | 0.248 |
| Osteonecrosis | 0.93 | (0.63–1.30) | 0.711 | 0.91 | (0.61–1.27) | 0.597 |
| Increased risk CKD | 0.86 | (0.60–1.16) | 0.362 | 0.79 | (0.54–1.10) | 0.197 |
| Leg ulcer | 0.56 | (0.20–1.23) | 0.209 | 0.47 | (0.15–1.10) | 0.135 |
| Cholelithiasis | 0.78 | (0.55–1.07) | 0.137 | 0.75 | (0.51–1.04) | 0.105 |
Note: Bold values represent statistical significance (p ≤ 0.05).
In the multivariate regression analysis, the variables were adjusted for age, sex, and hydroxyurea treatment. Total complication rate is a combination of all acute and chronic complications. Acute complication rate includes vaso‐occlusive episodes, acute chest syndrome, cerebrovascular accident, the need for acute exchange transfusion, and intensive care unit admission. Chronic complication rate includes proliferative retinopathy, osteonecrosis, increased risk of chronic kidney disease, leg ulcer, and cholelithiasis. Rate ratios (RR) and odds ratios (OR) were calculated for every 10‐unit increase in PK thermostability and every unit increase in pyruvate kinase/hexokinase (PK/HK) ratio.
Logistic regression analysis demonstrated that a 10‐unit increase in PK thermostability was associated with a 61% decrease in the adjusted odds of experiencing more than one complication (p ≤ 0.001), while a 1‐unit increase in PK/HK ratio reduced these odds by 30% (p = 0.030). Specifically, each 10‐unit increase in PK thermostability lowered the adjusted odds of experiencing more than one chronic complication by 41% (p = 0.007) and each 1‐unit increase in PK/HK ratio lowered the adjusted odds of experiencing more than one acute complication by 40% (p = 0.011). Additionally, the adjusted odds of having two or more VOEs or cholelithiasis decreased by about 29% (p = 0.048) and 53% (p = 0.001), respectively, with a 10‐unit increase in PK thermostability. For each unit increase in PK/HK ratio, the adjusted odds of experiencing two or more VOEs or a CVA decreased by 36% (p = 0.015) and 53% (p = 0.042), respectively.
4. Discussion
In this study, we demonstrated decreased PK thermostability and decreased relative PK activity (PK/HK ratio) in patients with SCD. Furthermore, PK thermostability and PK/HK ratio were related to sickle RBC properties and clinical manifestations of SCD. Decreased PK thermostability is associated with an increase in reticulocyte count as well as impaired functional properties of RBCs, such as a decrease in RBC deformability and oxygen affinity, and an increase in RBC adhesion. The PK/HK ratio shows a negative association with 2,3‐DPG levels. This suggests that a modest reduction in PK activity, which is downstream of 2,3‐DPG in glycolysis, compared to HK activity, which is upstream of 2,3‐DPG, results in increased levels of 2,3‐DPG. We also showed that a decreased PK thermostability and PK/HK ratio are associated with the occurrence of SCD‐related complications such as VOEs [29]. These results imply that decreased PK thermostability and PK/HK ratio are pathophysiological features of SCD that are related to clinical severity. Hence, these findings are of importance in light of PK activator therapies that are currently under investigation in clinical trials.
An elevated ATP content was observed in sickle RBCs compared to healthy controls. This is likely attributed to reticulocytosis, accompanied by increased absolute PK activity and oxidative phosphorylation in the mitochondria for ATP production [30]. The increased PK and HK activity levels in patients with SCD are both attributed to younger RBC age [31], as confirmed by our density separation data (Figure 2). When adjusted for RBC age, SCD patients exhibit a reduced PK/HK ratio compared to healthy controls. This reduced ratio can result from either decreased PK activity or increased HK activity. Increased HK activity could, theoretically, enhance glycolytic flux, leading to elevated ATP production. However, patients with SCD also have elevated 2,3‐DPG levels, indicating impaired glycolysis downstream in the Embden‐Meyerhof pathway, similar to observations in patients with PKD. Therefore, we believe that the PK/HK ratio reflects decreased PK activity rather than abnormally elevated HK activity in patients with SCD. Clinical trials demonstrated that PK activation therapy ameliorates pathophysiological effects in SCD. Phase 1/2 studies of mitapivat and etavopivat in patients with SCD have demonstrated increased ATP production and reduced intracellular 2,3‐DPG levels, leading to higher hemoglobin levels [15, 16, 17]. Moreover, improvements in markers of RBC sickling and oxygen affinity were observed, and treatment with PK activator therapy resulted in a reduction of reticulocytes and hemolytic parameters, such as lactate dehydrogenase and bilirubin [15, 16, 17]. Reduced hemolysis and, hence, increased RBC survival alters the distribution of RBC subpopulations. We, therefore, performed an RBC density separation in a subset of patients to investigate whether the PK and sickling properties are RBC age/density dependent. Importantly, RBC density is not identical to RBC age, but reticulocytes are less dense with increased MCV and decreased MCHC. During cell aging, RBCs shed parts of their membrane resulting in smaller and denser cells [27, 32]. RBC fractions with the highest density had the lowest enzymatic PK and HK activity, lowest RBC deformability, and sickled earlier upon deoxygenation. Understanding the PK properties in the RBC fractions can help in evaluating the treatment responses, as the distribution of these fractions differs between patients and may also be affected by (PK activator) therapy. Measuring the PK and HK activity and PK thermostability exclusively in Fraction 1 was not feasible. This is because this fraction contained an excessive number of leukocytes and platelets, which interfere with enzyme activity, and an insufficient volume to perform RBC isolation and enzyme activity measurements. Overall, the results of the density separation experiments in a subset of SCD patients show that each RBC fraction possesses unique characteristics, with a decrease in PK and HK activity but no difference in PK thermostability according to density. The varying proportions of these RBC fractions among patients could explain the heterogeneity in RBC characteristics in SCD. To accurately assess potential PK impairment, PK activity in SCD patients could be compared to that of individuals with a similar reticulocyte count. Future studies should aim to investigate a population with high reticulocytes but without RBC defects to provide clearer insights into PK.
Four PK isoforms are encoded by two genes: PKLR (PKL in the liver, PKR in RBCs) and PKM (PKM1 in musle and brain, PKM2 in embryonic cells and various adult tissues, including erythroblasts) [4, 33]. The PK thermostability and PK/HK ratio assays were initially developed as a diagnostic tool for PKD [34]. The PK thermostability test measures the ability of the PK enzyme to maintain its structure and function when exposed to heat [22]. In PKD the PKLR gene is mutated, often resulting in decreased PK activity and thermostability of the variant PK. However, why the PK enzyme in patients with SCD is less active and less resistant to heat exposure despite the absence of a mutation in the PKLR gene is yet unknown. A decreased relative PK activity is not limited to PKD and SCD but has also been reported in other hereditary hemolytic anemias, including hereditary spherocytosis, beta thalassemia, and hereditary xerocytosis (also known as dehydrated stomatocytosis), as well as in patients with myelodysplastic syndrome [35, 36, 37]. In hereditary spherocytosis, it was hypothesized that membrane instability causes the loss of glycolytic enzyme complexes from the RBC membrane, including PK, thereby reducing the local availability of PK [36]. In SCD and thalassemia, oxidative stress could affect the structure and/or function of the PK enzyme [38]. This hypothesis is supported by data showing inhibition of PKM2 by reactive oxygen species [39]. While most research on PKM2 had focused on its role in cancer cells, the mechanism observed in PKM2 may also be relevant to PKR. Currently, it is unclear whether the relatively decreased PK activity observed in the above conditions is primarily caused by oxidative stress, the loss of glycolytic enzyme complexes due to membrane instability, and/or another underlying factor. However, impaired PK activity is a rationale for PK activator therapy in all these red cell disorders and is currently under investigation [40, 41, 42].
Although SCD is caused by a single nucleotide mutation, the pathophysiology is highly complex, leading to significant phenotype variability [43]. Many biomarkers have been studied in SCD, each related to different aspects of disease pathophysiology. For decades, HbF has been a well‐known biomarker and key modulator of SCD severity [44]. HbF plays a protective role in SCD by reducing the HbS concentration and thus inhibiting RBC sickling. This protective mechanism can be induced with HbF induction therapies, such as hydroxyurea [45]. Patients with higher HbF levels generally experience a milder clinical disease course. However, even patients with elevated HbF levels as a result of hydroxyurea treatment may still suffer from clinical complications [45]. Therefore, other pathophysiological targets are needed that can be addressed by therapy, such as PK thermostability and the PK/HK ratio. The correlation between PK thermostability and the PK/HK ratio with both laboratory and SCD‐related clinical characteristics is a rationale for PK activation therapy. Future studies should investigate whether patients with reduced PK thermostability and/or a low PK/HK ratio will particularly benefit from PK activator therapy by measuring these specific biomarkers before initiation of therapy.
In conclusion, decreased PK thermostability and PK/HK ratio are associated with compromised RBC properties and a higher prevalence of SCD‐related clinical complications in patients with SCD. Enhancing PK activity and stability with PK activator therapy may therefore improve other pathophysiological features of SCD beyond RBC metabolism. As the development of disease‐modifying therapies for patients with SCD is urgently needed, identification and further research into new biomarkers and promising approaches such as PK activators are essential.
Author Contributions
M.J.M.T., R.v.W., and M.A.E.R. designed the study. M.J.M.T., J.F.B., L.v.P., M.J.v.D., B.A.v.O., and J.R.A.d.W. developed laboratory methodology and performed laboratory experiments. M.J.M.T., S.v.d.V., S.E.M.S., M.N.L., M.H.C., E.N., B.J.B., A.W.R., E.J.v.B., and M.A.E.R. recruited and included patients. M.J.M.T. collected clinical data and wrote the manuscript. R.v.W. and M.A.E.R. supervised the research. All authors revised the manuscript and approved the final version.
Ethics Statement
Study procedures were approved by the local medical ethical committees in accordance with the Declaration of Helsinki (protocol number: 17/392). Blood samples from healthy controls were obtained by the Mini Donor Service, a blood donation facility at the University Medical Center Utrecht, the Netherlands (protocol number: 18/774).
Consent
All patients provided verbal and written consent for participation in this study.
Conflicts of Interest
M.J.M.T., M.J.v.D., and J.R.A.d.W. received research funding from Agios Pharmaceuticals Inc. M.N.L. received research funding from ZonMW, LeoPharma, and INVENT‐VTE; reports consultancy fees (all to institute) from BMS/Pfizer, Amgen, Inari, Viatris, Servier, and AbbVie. M.H.C. received research funding from Takeda, Pfizer, Bayer, CSL Behring, Novo Nordisk, Novartis, Nordic Pharma; reports advisory or consulting role with Roche, Bayer, and Novartis. E.N. received research funding from Novartis; reports consulting role with Novartis and Vertex. B.J.B. received research funding from Sanquin, BMS, Pfizer, and Novartis; received honoraria from Sanofi; reports advisory board participation with Celgene, CSL Behring, Pfizer, and Novo Nordisk. E.J.v.B. received research funding from Agios Pharmaceuticals Inc.; reports consulting role with Agios Pharmaceuticals Inc. and Pfizer. A.W.R. received honoraria from Pfizer, Vertex pharmaceuticals, BMS, and Servier. R.v.W. received research funding from Agios Pharmaceuticals Inc. and Pfizer; reports consulting role with Agios Pharmaceuticals Inc., Pfizer, and RR Mechatronics. M.A.E.R. received research funding from Agios Pharmaceuticals Inc. and Pfizer; reports consulting role with RR Mechatronics. The other authors declare no conflicts of interest.
Supporting information
Data S1. Supporting Information.
Acknowledgments
The authors sincerely wish to thank all the patients and their families who participated in this study. This study was performed in collaboration with the Sickle Cell Outcome Research (SCORE) consortium in the Netherlands. We thank Aida Kidane Gembremeskel for the recruitment and enrollment of the children in this study. We thank Johan Gerrits for the ATP and 2,3‐DPG measurements.
Funding: This study was supported by Eurostars (E115561) and in part by research funding from Agios Pharmaceuticals Inc.
Data Availability Statement
Data and protocols are available upon request (e‐mail M.A.E.Rab@umcutrecht.nl). Data will be shared as is compliant with the General Data Protection Regulation and European Union privacy laws.
References
- 1. Ware R. E., de Montalembert M., Tshilolo L., and Abboud M. R., “Sickle Cell Disease,” Lancet 390, no. 10091 (2017): 311–323. [DOI] [PubMed] [Google Scholar]
- 2. Rees D. C., Williams T. N., and Gladwin M. T., “Sickle‐Cell Disease,” Lancet 376, no. 9757 (2010): 2018–2031. [DOI] [PubMed] [Google Scholar]
- 3. Piel F. B., Steinberg M. H., and Rees D. C., “Sickle Cell Disease,” New England Journal of Medicine 376, no. 16 (2017): 1561–1573. [DOI] [PubMed] [Google Scholar]
- 4. van Dijk M. J., de Wilde J. R. A., Bartels M., et al., “Activation of Pyruvate Kinase as Therapeutic Option for Rare Hemolytic Anemias: Shedding New Light on an Old Enzyme,” Blood Reviews 61 (2023): 101103. [DOI] [PubMed] [Google Scholar]
- 5. Jensen M., Shohet S. B., and Nathan D. G., “The Role of Red Cell Energy Metabolism in the Generation of Irreversibly Sickled Cells in Vitro,” Blood 42, no. 6 (1973): 835–842. [PubMed] [Google Scholar]
- 6. Thiagarajan P., Parker C. J., and Prchal J. T., “How Do Red Blood Cells Die?,” Frontiers in Physiology 12 (2021): 655393. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Xu Z., Dou W., Wang C., and Sun Y., “Stiffness and ATP Recovery of Stored Red Blood Cells in Serum,” Microsystems & Nanoengineering 5 (2019): 51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. McMahon T. J., Darrow C. C., Hoehn B. A., and Zhu H., “Generation and Export of Red Blood Cell ATP in Health and Disease,” Frontiers in Physiology 12 (2021): 754638. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Orrico F., Laurance S., Lopez A. C., et al., “Oxidative Stress in Healthy and Pathological Red Blood Cells,” Biomolecules 13, no. 8 (2023): 1262. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Alli N., Coetzee M., Louw V., et al., “Sickle Cell Disease in a Carrier With Pyruvate Kinase Deficiency,” Hematology 13, no. 6 (2008): 369–372. [DOI] [PubMed] [Google Scholar]
- 11. Cohen‐Solal M., Preúhu C., Wajcman H., et al., “A New Sickle Cell Disease Phenotype Associating Hb S Trait, Severe Pyruvate Kinase Deficiency (PK Conakry), and an Alpha2 Globin Gene Variant (Hb Conakry),” British Journal of Haematology 103, no. 4 (1998): 950–956. [DOI] [PubMed] [Google Scholar]
- 12. Charache S., Grisolia S., Fiedler A. J., and Hellegers A. E., “Effect of 2,3‐Diphosphoglycerate on Oxygen Affinity of Blood in Sickle Cell Anemia,” Journal of Clinical Investigation 49, no. 4 (1970): 806–812. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Rab M. A. E., Bos J., van Oirschot B. A., et al., “Decreased Activity and Stability of Pyruvate Kinase in Sickle Cell Disease: A Novel Target for Mitapivat Therapy,” Blood 137, no. 21 (2021): 2992–2997. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Xu J. Z. and Vercellotti G. M., “Pyruvate Kinase Activators: Targeting Red Cell Metabolism in Sickle Cell Disease,” Hematology 2023, no. 1 (2023): 107–113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Van Dijk M. J., Rab M. A. E., Van Oirschot B. A., et al., “One‐Year Safety and Efficacy of Mitapivat in Sickle Cell Disease: Follow‐Up Results of a Phase 2, Open‐Label Study,” Blood Advances 7, no. 24 (2022): 7539–7550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Xu J. Z., Conrey A., Frey I., et al., “A Phase 1 Dose Escalation Study of the Pyruvate Kinase Activator Mitapivat (AG‐348) in Sickle Cell Disease,” Blood 140, no. 19 (2022): 2053–2062. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Saraf S. L., Hagar R. W., Idowu M., et al., “Multicenter, Phase 1 Study of Etavopivat (FT‐4202) Treatment for up to 12 Weeks in Patients With Sickle Cell Disease,” Blood Advances 8, no. 16 (2024): 4459–4475. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Idowu M., Otieno L., Dumitriu B., et al., “A Phase 2/3, Double‐Blind, Randomized, Placebo‐Controlled, Multicenter Study of Mitapivat in Patients With Sickle Cell Disease: RISE UP Phase 2 Results,” Blood 142, no. 1 (2023): 271. [Google Scholar]
- 19. Vaidya S. R. and Aeddula N. R., “Chronic Kidney Disease,” in The Scientific Basis of Urology, ed. Shumway K. (StatPearls, 2024), 257–264. [Google Scholar]
- 20. Beutler E., “Chapter 3. The Preparation of Red Cells for Assay,” in Red Cell Metabolism: A Manual of Biochemical Methods, ed. Beutler E. (Grune & Stratton, 1984). [Google Scholar]
- 21. Blume K. G., Kaplan J. C., Löhr G. W., Ramot B., and Valentine W. N., “International Committee for Standardization in Haematology: Recommended Methods for Red‐Cell Enzyme Analysis,” British Journal of Haematology 35, no. 2 (1977): 331–340. [DOI] [PubMed] [Google Scholar]
- 22. Blume K. G., Arnold H., Lohr G. W., and Beutler E., “Additional Diagnostic Procedures for the Detection of Abnormal Red Cell Pyruvate Kinase,” Clinica Chimica Acta 43, no. 3 (1973): 443–446. [DOI] [PubMed] [Google Scholar]
- 23. Kim H., Kosinski P., Kung C., et al., “A Fit‐For‐Purpose LC–MS/MS Method for the Simultaneous Quantitation of ATP and 2,3‐DPG in Human K2EDTA Whole Blood,” Journal of Chromatography B: Analytical Technologies in the Biomedical and Life Sciences 1061–1062 (2017): 89–96. [DOI] [PubMed] [Google Scholar]
- 24. Da Costa L., Suner L., Galimand J., et al., “Diagnostic Tool for Red Blood Cell Membrane Disorders: Assessment of a New Generation Ektacytometer,” Blood Cells, Molecules & Diseases 56, no. 1 (2016): 9–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Lazarova E., Gulbis B., Oirschot B., and van Wijk R., “Next‐Generation Osmotic Gradient Ektacytometry for the Diagnosis of Hereditary Spherocytosis: Interlaboratory Method Validation and Experience,” Clinical Chemistry and Laboratory Medicine 55, no. 3 (2017): 394–402. [DOI] [PubMed] [Google Scholar]
- 26. Rab M. A. E., van Oirschot B. A., Bos J., et al., “Characterization of Sickling During Controlled Automated Deoxygenation With Oxygen Gradient Ektacytometry,” Journal of Visualized Experiments 2019, no. 153 (2019): 1–10. [DOI] [PubMed] [Google Scholar]
- 27. Rennie C. M., Thompson S., Parker A. C., and Maddy A., “Human Erythrocyte Fraction in “Percoll” Density Gradients,” Clinica Chimica Acta 98, no. 1–2 (1979): 119–125. [DOI] [PubMed] [Google Scholar]
- 28. Bianchi P., Fermo E., Glader B., et al., “Addressing the Diagnostic Gaps in Pyruvate Kinase Deficiency: Consensus Recommendations on the Diagnosis of Pyruvate Kinase Deficiency,” American Journal of Hematology 94 (2019): 149–161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Quinn C. T., “Minireview: Clinical Severity in Sickle Cell Disease: The Challenges of Definition and Prognostication,” Experimental Biology and Medicine 241, no. 7 (2016): 679–688. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Cooper G. M., “The Mechanism of Oxidative Phosphorylation,” in The Cell: A Molecular Approach, ed. Cooper G. M. (Sinauer Associates, 2000). [Google Scholar]
- 31. Jansen G., Koenderman L., Rijksen G., Punt K., Dekker A. W., and Staal G. E. J., “Age Dependent Behaviour of Red Cell Glycolytic Enzymes in Haematological Disorders,” British Journal of Haematology 61, no. 1 (1985): 51–59. [DOI] [PubMed] [Google Scholar]
- 32. Wei Q., Wang X., Zhang C., Dao M., and Gong X., “Evolution of Surface Area and Membrane Shear Modulus of Matured Human Red Blood Cells During Mechanical Fatigue,” Scientific Reports 13 (2023): 8563. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Gupta V. and Bamezai R. N. K., “Human Pyruvate Kinase M2: A Multifunctional Protein,” Protein Science 19, no. 11 (2010): 2031–2044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Miwa S., “Recommended Methods for the Characterization of Red Cell Pyruvate Kinase Variants: International Committee for Standardization in Haematology,” British Journal of Haematology 43, no. 2 (1979): 275–286. [DOI] [PubMed] [Google Scholar]
- 35. Rab M. A. E., van Oirschot B. A., Straaten S., et al., “Decreased Activity and Stability of Pyruvate Kinase in Hereditary Hemolytic Anemia: A Potential Target for Therapy by AG‐348 (Mitapivat), an Allosteric Activator of Red Blood Cell Pyruvate Kinase,” Blood 134, no. 1 (2019): 3506. [Google Scholar]
- 36. Andres O., Loewecke F., Morbach H., et al., “Hereditary Spherocytosis Is Associated With Decreased Pyruvate Kinase Activity Due to Impaired Structural Integrity of the Red Blood Cell Membrane,” British Journal of Haematology 187, no. 3 (2019): 386–395. [DOI] [PubMed] [Google Scholar]
- 37. Fattizzo B., Vercellati C., Marcello A., et al., “Glycolytic Activity and In Vitro Effect of the Pyruvate Kinase Activator AG‐946 in Red Blood Cells From Low‐Risk Myelodysplastic Syndromes Patients: A Proof‐Of‐Concept Study,” American Journal of Hematology 99, no. 6 (2024): 1201–1204. [DOI] [PubMed] [Google Scholar]
- 38. Sprengers E. D., Van Berkel T. J. C., Koster J. F., and Staal G. E. J., “Influence of the Redox State of Glutathione Upon Pyruvate Kinase in the Intact Erythrocyte,” Clinica Chimica Acta 80, no. 3 (1977): 495–502. [DOI] [PubMed] [Google Scholar]
- 39. Anastasiou D., Poulogiannis G., Asara J. M., et al., “Inhibition of Pyruvate Kinase M2 by Reactive Oxygen Species Contributes to Cellular Antioxidant Responses,” Science 334, no. 6060 (2011): 1278–1283. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Glenthøj A., van Beers E. J., van Wijk R., et al., “Designing a Single‐Arm Phase 2 Clinical Trial of Mitapivat for Adult Patients With Erythrocyte Membranopathies (SATISFY): A Framework for Interventional Trials in Rare Anaemias—Pilot Study Protocol,” BMJ Open 14, no. 7 (2024): e083691. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Kuo K., Layton D., Al‐Samkari H., et al., “P112: ENERGIZE and ENERGIZE‐T: Two Phase 3, Randomized, Double‐Blind, Placebo‐Controlled Studies of Mitapivat in Adults With Non‐Transfusion‐Dependent or Transfusion‐Dependent Alfa or Beta Thalassemia,” HemaSphere 6 (2022): 23–24. [Google Scholar]
- 42. Zeidan A. M., Sekeres M. A., Al‐Samkari H., et al., “A Phase 2B, Open‐Label Multicenter Study of Tebapivat (AG‐946), a Potent Pyruvate Kinase Activator, in Patients With Anemia Due to Lower‐Risk Myelodysplastic Syndromes,” Blood 144, no. 1 (2024): 6708. [Google Scholar]
- 43. Saah E., Fadaei P., Gurkan U. A., and Sheehan V., “Sickle Cell Disease Pathophysiology and Related Molecular and Biophysical Biomarkers,” Hematology/Oncology Clinics of North America 36, no. 6 (2022): 1077–1095. [DOI] [PubMed] [Google Scholar]
- 44. Jackson J. F., Odom J. L., and Bell W. N., “Amelioration of Sickle Cell Disease by Persistent Fetal Hemoglobin,” JAMA 177, no. 12 (1961): 867–869. [DOI] [PubMed] [Google Scholar]
- 45. Akinsheye I., Alsultan A., Solovieff N., et al., “Fetal hemoglobin in Sickle Cell Anemia,” Blood 118, no. 1 (2011): 19–27. [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
Data S1. Supporting Information.
Data Availability Statement
Data and protocols are available upon request (e‐mail M.A.E.Rab@umcutrecht.nl). Data will be shared as is compliant with the General Data Protection Regulation and European Union privacy laws.