Open access journal of the Ferrata-Storti Foundation, a non-profit organization
Articles

Metabolic signatures of cardiorenal dysfunction in plasma from sickle cell patients as a function of therapeutic transfusion and hydroxyurea treatment

Department of Biochemistry and Molecular Genetics, University of Colorado Denver – Anschutz Medical Campus, Aurora, CO, USA; Department of Medicine – Division of Hematology, University of Colorado Denver – Anschutz Medical Campus, Aurora, CO
Division of Pulmonary, Allergy and Critical Care Medicine, Department of Medicine, University of Pittsburgh, Pennsylvania
Division of Pulmonary, Allergy and Critical Care Medicine, Department of Medicine, University of Pittsburgh, Pennsylvania
Department of Biochemistry and Molecular Genetics, University of Colorado Denver – Anschutz Medical Campus, Aurora, CO
Department of Biochemistry and Molecular Genetics, University of Colorado Denver – Anschutz Medical Campus, Aurora, CO
Department of Biochemistry and Molecular Genetics, University of Colorado Denver – Anschutz Medical Campus, Aurora, CO
Department of Medicine, University of Illinois at Chicago, Chicago, Illinois
Linda Crnic Institute for Down Syndrome, University of Colorado – Anschutz Medical Campus, Aurora, CO, USA; Department of Pharmacology, University of Colorado Anschutz Medical Campus
Linda Crnic Institute for Down Syndrome, University of Colorado – Anschutz Medical Campus, Aurora, CO, USA; Department of Pharmacology, University of Colorado Anschutz Medical Campus
Linda Crnic Institute for Down Syndrome, University of Colorado – Anschutz Medical Campus, Aurora, CO, USA; Department of Pharmacology, University of Colorado Anschutz Medical Campus; School of Medicine Information Services, University of Colorado Anschutz Medical Campus
Department of Medicine, University of Illinois at Chicago, Chicago, Illinois
University of Maryland School of Medicine, University of Maryland, Baltimore, MD
Vol. 108 No. 12 (2023): December, 2023 https://doi.org/10.3324/haematol.2023.283288

Abstract

Metabolomics studies in sickle cell disease (SCD) have been so far limited to tens of samples, owing to technical and experimental limitations. To overcome these limitations, we performed plasma metabolomics analyses on 596 samples from patients with SCD enrolled in the WALK-PHaSST study (clinicaltrials gov. Identifier: NCT00492531). Clinical covariates informed the biological interpretation of metabolomics data, including genotypes (hemoglobin [Hb] SS, hemoglobin SC), history of recent transfusion (HbA%), response to hydroxyurea treatment (fetal Hb%). We investigated metabolic correlates to the degree of intravascular hemolysis, cardiorenal function, as determined by tricuspid regurgitation velocity (TRV), estimated glomerular filtration rate (eGFR), and overall hazard ratio (unadjusted or adjusted by age). Recent transfusion events or hydroxyurea treatment were associated with elevation in plasma-free fatty acids and decreases in acyl-carnitines, urate, kynurenine, indoles, carboxylic acids, and glycine- or taurine-conjugated bile acids. High levels of these metabolites, along with low levels of plasma S1P and L-arginine were identified as top markers of hemolysis, cardiorenal function (TRV, eGFR), and overall hazard ratio. We thus uploaded all omics and clinical data on a novel online portal that we used to identify a potential mechanism of dysregulated red cell S1P synthesis and export as a contributor to the more severe clinical manifestations in patients with the SS genotype compared to SC. In conclusion, plasma metabolic signatures - including low S1P, arginine and elevated kynurenine, acyl-carnitines and bile acids - are associated with clinical manifestation and therapeutic efficacy in SCD patients, suggesting new avenues for metabolic interventions in this patient population.

Introduction

Sickle cell disease (SCD) encompasses a group of inherited hemoglobinopathies caused by mutations in the gene coding for the β globin chain (HBB) in the hemoglobin tetramer.1,2 These mutations result in the substitution of single amino acid residue, E6 to V in sickle hemoglobin (HbS) allele βS, or E6 to K in HbC. Under deoxygenated conditions, sickling of red blood cells (RBC) occurs following crystallization of mutant globins into macro-fibers,3-5 which in turn disrupts erythrocyte morphology, rheology6 and metabolism.7 Clinically, these molecular events manifest themselves in a complex disease characterized by hemolytic anemia, inflammation and recurrent vaso-occlusive episodes.8 While such manifestations tend to be more severe in patients with the SS genotype, vaso-occlusive complications are still present in patients with the SC genotype.2,9

Lifelong transfusion represent a mainstay in the management of SCD in children and adults.8 Increases in circulating levels of heme and iron secondary to the elevated hemolytic propensity of sickle RBC triggers recurrent episodes of vaso-occlusion and inflammation, which progressively cause damage to most organs, including the kidneys, lungs, and the cardiovascular system, and overall reduced life expectancy.8 Interventions based on treatment with hydroxyurea (hydroxycarbamide) have shown a certain degree of efficacy in upregulating the expression of fetal hemoglobin (HbF) and increasing mean corpuscular volume,10 while decreasing the frequency of pain episodes and other acute complications in adults and children with sickle cell anemia of SS or Sβºthal genotypes, reducing the incidence of life-threatening neurological events such as primary stroke by maintaining transcranial Doppler velocities.11

The circulating metabolome offers a window not just on chemical individuality under healthy conditions,12 but also onto dysregulation of systems homeostasis under pathological conditions, such as in the case of SCD.7 Studies on the circulating metabolome in SCD have identified potential new avenues for metabolic interventions. For example, arginine supplementation13 has been proposed as a strategy to boost nitric oxide (NO) synthesis and NO-dependent vasodilation, thus counteracting hemoglobin-dependent NO scavenging secondary to intravascular hemolysis, and release of RBC arginase which has been shown to catabolize L-arginine.13,14 Similarly, hemolysis-derived free hemoglobin, heme and iron have been associated with increased renal, cardiopulmonary and infectious complications in this and other patient populations.15,16 An alternative metabolic intervention, glutamine supplementation has been proposed as a strategy to boost glutathione synthesis, thus counteracting oxidant stress in the sickle RBC.1 7,1 8 More recently, metabolomics interventions have been designed to boost the rate of late, payoff steps enzymes of glycolysis such as pyruvate kinase. This strategy promotes the synthesis of adenosine triphosphate and energy metabolism in the sickle cell, while depleting the reservoirs of allosteric modulators like 2,3-diphosphoglycerate that promote sickling by stabilizing the tense deoxygenated state of hemoglobin.19 Other metabolomics studies have identified dysregulated adenosine signaling20,21 and sphingolipid metabolism22 as potential targets for therapeutic interventions, though studies have not progressed beyond promising preclinical animal models. One major limitations of almost all metabolomics studies of SCD plasma to date is that limited sample size and lack of extensively curated clinical data hampered the extrapolation of the translational relevance of the reported findings. To bridge this gap, here we performed a metabolomics study of plasma from 596 SCD patients, for which extensive information on patients’ demographics (sex, age, body mass index [BMI]), genotypes, hematological and clinical characteristics, as well as therapeutic interventions (transfusion events, as inferred from HbA%, hydroxyurea treatment) were available. Results from analyses in plasma were compared against the recently characterized RBC metabolome in this cohort,23 identifying compartment-specific metabolic signatures associated with potential novel strategies for metabolic interventions in SCD.

Methods

An extended version of this section is provided in the Online Supplementary Appendix, and our previous meta-bolomics report on RBC samples from the same cohort.23

Ethical statement, blood collection, processing and storage

A total of 596 SCD patients with available plasma samples were recruited as part of the Walk-PHaSST clinical trial (clinicaltrials gov identifier: NCT00492531 and BioLINCC study: HLB01371616a) under protocols approved by the University of Pittsburgh.9,24-28 Clinical covariates were collected prior to plasma metabolomics analysis, including intravascular hemolysis, tricuspid regurgitation velocity (TRV),29 parameters indicative of kidney function such as estimated glomerular filtration rate (eGFR), creatinine.9,24-28 Standard high-pressure liquid chromotography (HPLC) methods were used to determine the percentage of sickle HbS, HbC, HbF and normal HbA.9,24-28

Ultra-high pressure liquid chromatography-mass spectrometry metabolomics

Metabolomics extractions,30,31 analyses32,33 and elaborations34 were performed as previously described.

Statistical analyses

Multivariate and multivariable analyses, including principal component analyses (PCA), hierarchical clustering analyses, two-way ANOVA, correlation analyses (Spearman) and calculation of receiver operating characteristic curves were performed using MetaboAnalyst 5.0.35 For survival analysis, we used time to right censorship (including death or last follow-up) as time to event and vital status (death or alive) was the studied outcome. PCA was used to derive a hemolytic component from lactate dehydrogenase, aspartate aminotransferase, total bilirubin and reticulocyte percent.27,36,37 We applied Cox proportional hazard models to calculate the hazard ratios and P value for each metabolite. Cox analysis and adjusted regressions of metabolites to clinical outcomes were performed in R (R Core Team (2022), https://www.r-project.org/).

Sickle cell disease ShinyApp portal

A portal for online sharing of all the data generated in this study, based on code from the COVID-Ome Explorer portal.38

Results

Metabolomics of the WALK-PHASST sickle cell disease cohort

Metabolomics analyses were performed on plasma samples from 596 sickle cell study participants in the screening phase of Walk-PHASST cohort (Figure 1A). Results are extensively reported in the Online Supplementary Table S1, which also includes anonymized patients’ biological data, including genotype, sex, age, BMI (Figure 1A). An unsupervised analysis of the plasma metabolomics results is provided in the Online Supplementary Figure S1. PCA and hierarchical clustering analyses (HCA) of all the metabolomics data showed a significant separation of plasma samples from individuals with SS genotypes compared to the other groups (Online Supplementary Figure S1A, B). As previously noted,23 despite double randomization of the samples and blinding of the analytical team, plasma creatinine levels measured via mass spectrometry significantly correlated with clinical measurement for this metabolite (Online Supplementary Figure S1C).

Plasma metabolic profiles in SS versus SC patients who have not been recently transfused

Individuals with SS genotypes represented approximetely 73% of the cohort in this study and showed distinct clinical9,24-28 and metabolic phenotypes (Online Supplementary Figure S1) compared to patients with other genotypes. As such, the first analysis we performed focused on the SS patient population. Clinically, many of these patients are routinely transfused, which could be herein assessed by screening of HbA percentages. We used this intervention to all for an in vivo assessment of the metabolome of plasma from patients with high levels of HbS compared with patients with SS genotype who have very high levels of transfused AA blood as a control. For this first analysis, other treatments (e.g., hydroxyurea - see below) were excluded. HCA of plasma metabolites significantly affected by HbA and HbS levels in SS patients identified a significant increase in the levels of free fatty acids (palmitoleic, myristoleic, myristic, stearic, docosapentaenoic, arachidonic, dihomogammalinolenic and adrenic acid) as a function of high levels of HbA, as opposed to declines in a series of acyl-carnitines (5:0, 5:0 OH, 14:0, 16:1; Figure 1B). Elevated HbS levels were correlated (Spearman; P<0.05) with increases in urate, gluconolactone phosphate (a metabolite of the oxidative phase of the pentose phosphate pathway [PPP]) and cystine, with decreases in the levels of pyruvate, choline and acetylcholine, on top of the aforementioned free fatty acids (Figure 1C). In order to further evaluate the impact of recent transfusion events on the plasma metabolome of SS patients, data were binned into extremes HbA phenotypes, with two groups identified based on “no recent transfusion events” (HbA <5%) or “recent transfusion events” (HbA >80%) that showed distinct metabolic phenotypes by partial least square-discriminant analysis (PLS-DA; Figure 1D). Volcano plots of significant metabolites in these two groups further identified transfusion-associated elevation in pyruvate and other pentose or hexose phosphate compounds (fructose bisphosphate and ribose phosphate) - usually associated with hemolysis when observed in plasma, since they should only be detected in intracellular compartments - and decreases in indoles, acyl-carnitines (especially those derived from branched chain amino acid catabolism - 5:0 and 5:0 OH) and pyridoxamine (Figure 1E).

Impact of hydroxyurea treatment on the plasma metabolome

Of the 433 patients with SS genotypes enrolled in this study, 306 had not been recently transfused (HbA <20%), of which 147 had received treatment with hydroxyurea (Online Supplementary Figure S2A). Unsupervised analyses including PCA and HCA identified a modest impact of hydroxyurea treatment on the plasma metabolome, with the strongest impact on the levels of acyl-carnitines, urate, L-arginine and amino acids (lower in treated patients compared to untreated ones; Online Supplementary Figure S2B, C). These effects were corroborated by correlation with HbF %, which increased after hydroxyurea treatment and was also positively associated with elevated tryptophan and indole metabolites (indole, indole acetate; Online Supplementary Figure S2D). Linear models, unadjusted or adjusted by HbF% confirmed a strong impact of hydroxyurea on plasma L-arginine, acyl-carnitines and amino acids (Online Supplementary Figure S2E). Acknowledging the limited signal emerged from this analysis, perhaps as a result of uncertain adherence to the hydroxyurea regimen, we performed an additional analysis by focusing on SS patients on hydroxyurea with mean corpuscular volumes >105 (Online Supplementary Figure S2F), owing to the well-established effect on this hematological parameter of hydroxyurea treatments.10 This analysis unmasked a signal related with hydroxyurea-dependent decrease in circulating levels of some long-chain acyl-carnitines, but not conjugated bile acids – still elevated in this subgroup of patients compared to the untreated group (Online Supplementary Figure S2G).

Comparison of SS patients to less clinically severe SC genotypes show trends consistent with beneficial effects of transfusion and hydroxyurea

We then performed a direct comparison of non-recently transfused SS patients to the metabolome of patients with less clinically severe genotype of SC (Figure 2A). Linear models identified alterations of sphingolipid metabolism (especially sphingosine 1-phosphate [S1P] and sphinganine 1-phosphate), methionine sulfoxide, phosphocreatine and PPP metabolites (gluconolactone phosphate and ribose diphosphate) as the top markers distinguishing SS and SC patients, even upon adjustment for HbA% (Figure 2B). These findings were corroborated by HCA, which also showed enrichment in SS plasma of acyl-carnitines, taurine- and glycine-conjugated bile acids (taurodexycholate, taurocholate, taurochenodeoxycholate, taurolithocholate, glycocholate, glycochenodeoxycholate), carboxylic acids (succinate, malate, oxaloacetate), purine deamination products (urate, hydroxisourate), as opposed to depletion in L-arginine and S1P (Figure 2C). These findings were further highlighted by the identification of metabolic correlates to HbC % (Figure 2D), the top of which presented in the form of violin plots in Figure 2E.

Figure 1.Plasma metabolomics of the WALK-PHASST SCD cohort. (A) The plasma metabolome of patients with SS genotype -who did not receive other treatment (e.g., hydroxyurea) was evaluated as a function of recent transfusion events, as determined by hemoglobin A (HbA)% (N=123 males and N=120 females). In (B), hierarchical clustering analysis of plasma metabolomics data as a function of HbS and HbA% (top 50 significant metabolites are shown). In (C), smile plot of plasma metabolic correlates to HbS%. In (D), partial least-square discriminant analysis of plasma metabolomics data in SS patients with recent (HbA >80%) or non-recent (HbA <5%) transfusion events. In (E), volcano plot of plasma metabolites significantly higher in patients with no recent transfusion event (HbA <5%) or with recent exchange transfusion (HbA >80%). PC: principal component; FC: fold change; UHPLC-MS: ultra-high pressure liquid chromatography-mass spectometry.

Plasma metabolic markers of intravascular hemolysis in the SS sickle cell disease population

Degree of intravascular hemolysis was most elevated in SS patients with the more severe clinical manifestations. Analysis of metabolomics data as a function of the degree of hemolysis were thus performed in non-recently transfused (HbA <20%) SS patients via PCA, HCA and correlation analysis (Figure 3A-C). Results indicate a strong positive correlation of hemolysis with circulating levels of conjugated bile acids, urate, acyl-carnitines, free fatty acids, pentose and hexose phosphate metabolites, carboxylic acids (succinate, malate) and a negative correlation with free amino acids, especially L-arginine and tryptophan metabolites (e.g., kynurenine). Similar results were even more evident in the Online Supplementary Figure S3, in which we focused on the top 30 patients with highest and lowest overall hemolysis, without constraining for patients’ genotypes nor for recent transfusion events.

Plasma metabolic correlates to Doppler-echocardiographic tricuspid regurgitation jet velocity in the WALK-PHaSST cohort

Cardiorenal dysfunction secondary to elevated hemolysis are common comorbidities in SCD, especially in patients with SS genotypes.13,28 Altered TRV is a hallmark of pulmonary dysfunction in this patient population, though plasma metabolic markers of aberrant TRV have not yet been reported, and is the focus of the analysis in Figure 4. Focusing on non-recently transfused (HbA <20%) SS patients, PCA, HCA, linear models (unadjusted, or adjusted for relevant covariates of patients’ age and sex) and correlation analyses all highlighted a significant correlation to TRV for plasma levels of S1P, L-arginine and tryptophan (depleted with high TRV) and their catabolites (citrulline, creatinine, kynurenine), acyl-carnitines (elevated with high TRV; Figure 4A-D). Similar results were even more evident in the Online Supplementary Figure S4, in which we focused on the top 30 patients with highest and lowest overall TRV, without constraining for patients’ genotypes nor for recent transfusion events. A snapshot of this analysis is provided in the volcano plot in Figure 4E, which focuses on SS patients with high versus low TRV - further highlighting depletion of glutamine and 5-oxoproline (a product of the γ-glutamyl-cycle) in patients with elevated TRV.

Figure 2.Plasma metabolic differences between sickle cell disease patients with SS and SC genotypes. (A) Plasma metabolomics analyses were performed on plasma from patients with SS (N=433) or SC genotype (N=107) who were not recently transfused (HbA <20% - N=306 and N=101 for SS and SC, respectively). (B) Linear models were used to identify the main differences between SS and SC patients, either unadjusted (x axis) or adjusted by HbA% (y axis). In (C), hierarchical clustering analysis of the top 50 metabolites by t test, and Spearman correlation analyses between plasma metabolites and HbC % (D). In (E), focus on selected metabolites, including sphingosine 1-phosphate (S1P) and pentose phosphate pathway metabolites that differed significantly between plasma of SC and SS patients.

Figure 3.Plasma metabolic markers of hemolysis in SS sickle cell disease patients who were not recently transfused (hemoglobin A <20%). In (A), principal component analysis (PCA) of the plasma metabolomics data from this cohort, color-coded based on the degree of hemolysis. In (B-D) heat map, linear model (unadjusted or adjusted by subject sex and age on the y axis) and volcano plot of the plasma metabolites most significantly associated with hemolytic propensity. In (E), volcano plot of the most significant plasma metabolic changes between the 25 patients with the highest and lowest degree of hemolysis (F). Hb: hemoglobin; RBC: red blood cell.

Figure 4.Plasma metabolic markers of cardiovascular dysfunction as gleaned by tricuspid jet velocity in SS patients. In (A), principal component (PC) analysis of the plasma metabolomics data from sickle cell (SS) patients with no recent transfusion event (hemoglobin A [HbA] <20%), color-coded based on the tricuspid regurgitation velocity (TRV). In (B-D), heat map, linear model (unadjusted or adjusted by subject sex and age on the y axis) and volcano plot of the metabolites most significantly associated with tricuspid regurgitation velocity (TRV). In (E), volcano plot and heat map of the most significant metabolic changes between the 25 patients with the highest and lowest degree of TRV.

Plasma metabolic correlates to renal function in the WALK-PHaSST cohort

Plasma metabolic markers of renal function in SS patients who had not been transfused recently (HbA <20%) were here evaluated as a function of eGFR, via PCA, HCA, linear models (unadjusted, or adjusted by patients’ sex and age) and correlation analyses (Figure 5A-D). Results confirmed a significant negative correlation between eGFR and creatinine, with concomitant accumulation of acyl-carnitines, depletion of arginine and accumulation of its catabolites (creatinine, citrulline), elevation in the levels of pyridoxal, n-acetylneuraminate, pro-inflammatory carboxylic acids (succinate), kynurenine, choline, acetyl-choline, urate, indoles. Similar results were even more evident in the Online Supplementary Figures S5 and S6, in which we focused on the top 25 patients with highest and lowest overall eGFR as a function of patients’ age in SS patients only with no recent transfusion event (HbA <20% - volcano plot in Figure 5E and heat map in the Online Supplementary Figure S5) or the top 30 patients by high versus low eGFR overall, without constraining for patients’ genotypes nor for recent transfusion events (Online Supplementary Figure S6).

Plasma metabolic markers of mortality, time to event and hazard ratio in the WALK-PHaSST cohort

Finally, we performed biomarker analyses to identify plasma metabolites associated with signatures of mortality, as gleaned by volcano plot of dichotomy outcomes (dead vs. alive; Figure 6A), receiver operating characteristics curve determination (Online Supplementary Figure S7) and a more formal hazard ratio of metabolomics data adjusted by patients’ age (Figure 6B). Results confirmed a strong association between conjugated bile acids, arginine and tryptophan metabolism (creatinine, citrulline, kynurenine, anthranilate, indoles), carboxylic acids (2-hydroxyglutarate, succinate), acyl-carnitines and untoward clinical outcomes. Correlation of clinical and metabolic covariates were thus plotted in the form of a network analysis in Figure 6C, which identified hubs of conjugated bile acids, free fatty acids (especially very-long chain poly- and highly-unsaturated fatty acids), glutamine metabolites (glutamine, glutamate, 2-oxoglutarate), pantothenate and acyl-carnitines associated with hematological (HbS, HbA, HbC, HbF %; RBC, white blood cell [WBC] counts) and cardiorenal function parameters (eGFR, creatinine, siten, BMI, cardiovascular risk, hemolysis). Further correlation analyses are reported for these clinical covariates to plasma metabolite levels in Online Supplementary Figure S8.

The sickle cell disease online portal for real time processing of plasma and red blood cell metabolomics data in sickle cell disease against clinical covariates, and cross-matrix correlations: impact on pain crises

Clinical and plasma metabolomics data in this study, and RBC metabolomics data from our previous study23 were collated into an online portal for real time data processing and figure generation. The portal, open and accessible at https://mirages.shinyapps.io/SCD2023/ serves as a repository for the data generated as part of this study, and as a tool to perform real time analyses and export of results upon correlation analyses between metabolites and clinical covariates, with the opportunity to modify filters (e.g., age, sex) or cross-matrix (plasma vs. RBC) correlations. The portal offers the opportunity to correct for covariates of interest, for example by adjusting by HbS, HbA, HbF and HbC% (Online Supplementary Figure S9). Leveraging this portal we performed cross-matrix analysis of plasma and RBC metabolomics data from the Walk-PHASST study. An overview of the matrix of plasma and RBC intra- and cross-matrix metabolic correlates is provided in Figure 6D, where metabolite pairs (x axis) are sorted by Spearman correlation coefficients (from lowest to highest). These analyses are further presented in the correlation matrix in the Online Supplementary Figure S10. Notably, most metabolic markers of cardiorenal dysfunction (especially creatinine, citrulline and tryptophan metabolites, serotonin and kynurenine) showed significant (P<0.0001) positive correlations (r>0.5) between plasma and RBC levels. Arginine and ornithine levels, like pyruvate, poorly correlated between plasma and RBC, suggesting cell-intrinsic mechanisms affecting the levels of these metabolites in SCD. However, these analyses also identified a discrepancy between elevation of intracellular RBC S1P levels and depletion of circulating S1P in plasma of patients with SS genotype compared to any other genotype tested in this study (Figure 6E; Online Supplementary Figure S11). Similarly, increased arginine uptake from plasma to RBC could explain the observed compartment and genotype-specific (SS vs. SC) changes in arginine levels (Figure 6E; Online Supplementary Figure S11).

Compartment-specific responses result in differential matrix-specific correlations to clinically-relevant outcomes, such as self-reported pain crises events in the year preceding the analysis. Correlation analysis of plasma and RBC metabolomics data to clinical measurements of pain perception showed a limited number of metabolites significantly associated with this parameter, mostly plasma metabolites that were positively associated (Online Supplementary Figure S12A). Of note, AcCa 10:1 in plasma and AcCa 2:0 in RBC were the top positive and negative correlates to pain crisis events (Online Supplementary Figure S12B, C, respectively). In keeping with compartment-specific levels of ornithine and S1P, plasma - but not RBC – ornithine ranked amongst the top positive correlates to pain (Online Supplementary Figure S12D). Plasma S1P and RBC S1P were positively and negatively associated to pain, respectively (Online Supplementary Figure S12E).

Discussion

Here we performed an extensive plasma metabolomics analysis of a clinically well-curated cohort of 596 SCD patients. Availability of detailed clinical measurements of HbS, A, C and F percentages allowed us to identify the impact of recent transfusion events, genotype (with an emphasis on the more severe SS vs. the less clinically severe SC genotype). As a result, we identified multiple signatures associated with cardiorenal dysfunction and hazard ratios, including alterations in tryptophan/kynurenine/indole metabolism, bile acid deconjugation, acyl-carnitine and S1P metabolism, glutaminolysis and mitochondrial dysfunction and L-arginine metabolism to citrulline and creatinine (Figure 7). Results are relevant in that they confirm and expand upon previous findings in the literature, linking metabolic differences23 to clinically relevant outcomes and providing a metabolic window on clinical efficacy of common interventions like transfusion of packed RBC and treatment with hydroxyurea.39 For example, previous smaller scale studies had documented a beneficial effect of transfusion events in SCD in the circulating plasma and RBC metabolome,40 though the scale of these studies was limited to a handful of recipients with no characterization of clinical relevant covariates.

Figure 5.Plasma metabolic markers of renal dysfunction as gleaned by estimated glomerular filtration rate and creatinine in SS patients. In (A), principal component analysis of the metabolomics data from this cohort, color-coded based on the estimated glomerular filtration rate (eGFR). In (B-D), heat map, linear model (unadjusted or adjusted by subject sex and age on the y axis) and volcano plot of the metabolites most significantly associated with eGFR. In (E), volcano plot and heat map of the most significant metabolic changes between the 25 patients with the highest and lowest eGFR. Tr: tricuspid; Hb: hemoglobin.

Figure 6.Plasma metabolic signatures of mortality, hazard ratio and plasma metabolic correlates to previously published red blood cell metabolomics data in the WALK-PHaSST cohort. (A) Plasma metabolic correlates to mortality in the WALK-PHaSST cohort. In (B), results from the hazard ratio analysis of plasma metabolomics data adjusted by patients’ age. In (C), plasma metabolite levels from the present analysis of 596 samples were correlated with all available clinical and hematological covariates and plotted in the form of a network, where each node represents a covariate, and edges indicate correlation between two covariates (thickness proportional to the module and significance of this association). In (D), plasma metabolite levels from the WALK-PHaSST cohort were matched to red blood cell (RBC) metabolomics measurements (D’Alessandro et al.23). Results indicated a strong positive cross-matrix correlation for almost all metabolic markers of cardiorenal dysfunction, except for sphingosine 1-phosphate (S1P), whose levels were elevated in SS RBC and sickle cell (SC) plasma, suggesting an increased synthesis or decreased export in the more clinically severe SS genotype (E).

Previous studies had identified dysregulated arginine metabolism41 via hemolysis-triggered increases in arginase activity as factor limiting nitric oxide synthesis and vasodilation in SCD patients, ultimately contributing to endothelial and cardiovascular dysfunction.13 Here we expand on these findings identifying a role for transfusion events or hydroxyurea treatment in the amelioration of dysregulated metabolic signatures in these pathways, effects that result in the mitigation of arginine consumption and accumulation of creatinine, ornithine, citrulline and guanidinoacetate. Of note, arginine catabolism to ornithine by arginase 1 was here associated with higher incidence of pain crises events in the year preceding the analysis. Altogether, our findings are supportive of a role for the modulation of this pathway in the management of clinical manifestations in SCD, from pain crises to hemolysis-associated cardiovascular and renal dysfunction.

Similarly, previous studies had focused on glutamine consumption in the context of SCD, and glutamine supplementation as a therapeutic strategy to fuel de novo glutathione synthesis as a strategy to counteract oxidant stress in the sickle RBC, though findings are controversial as to whether glutamine supplementation aggravates or mitigates the incidence of pain and vaso-occlusive crises in this patient population.17,18,42 Our findings show that accumulation of carboxylic acids - especially succinate and malate - are observed in plasma concomitantly to glutamine consumption and generation of 5-oxoproline, a byproduct of glutathione catabolism via the y-glutamyl-cycle and a metabolic endproduct of this cycle in the mature RBC, owing to the lack of the enzyme oxoprolinase. This observation is relevant in light of the role of these carboxylic acids, especially succinate, in the stabilization of hypoxia inducible factor 1a,43 and activation of its transcriptional targets, including interleukin 1|3.44 Of note, accumulation of amino and carboxylic acids downstream to glutamine catabolism (glutamate, 2-oxoglutarate, 2-hydroxyglutarate, succinate - in particular) was linked to signatures of cardiorenal dysfunction and overall hazard ratio, suggesting that fueling of glutathione synthesis via oral supplementation not only promotes anaplerotic reactions, but also likely fuels catabolism towards the accumulation of pro-inflammatory small molecule mediators. Further studies with stable isotope-labeled tracers will be necessary to further test this hypothesis and the ensuing clinical ramifications. Glycine and taurine conjugated bile acids are synthesized in the liver and deconjugated by the gut microbiome, and their accumulation in plasma is consistent with microbiome dysbiosis.45,46 These metabolites have been implicated in the pro-inflammatory activation of macrophages,47 as well as in coagulopathic complications (e.g., after trauma and hemorrhage),48 both common comorbidities in sickle cell patients. Conjugated bile acids had been previously identified as markers of cholecystectomy in sickle cell patients,42 though the impact of patients’ genotypes, transfusion events or hydroxyurea treatment on the circulating levels of these metabolites had not been reported. Of note, indole metabolites derived from tryptophan catabolism are also derived from microbial catabolism,49 and participate in systems homeostasis.50 Like kynurenines,51 other endogenous tryptophan metabolites, indoles participate in immunomodulatory effects on macrophages through triggering of aryl hydrocarbon receptors.52 Of note, kynurenine was here associated with high hemolysis, increased TRV and hazard ratios. This finding is interesting in that elevated kynurenine was previously reported in individuals suffering with elevated cardiopulmonary dysfunction, pulmonary hypertension and lung fibrosis, such as subjects with Down syndrome.53 Elevation of circulating levels of kynurenine is a hallmark of viral infection with pulmonary complications, such as COVID-19, whereby this metabolite correlates with interleukin 6 levels, clinical severity of SARS-CoV-2 infection and, ultimately, clinical outcomes.54,55 Of note, in both cases elevation of kynurenine is associated with the activation of the interferon-indoleamine 2,3-dioxygenase axis, in the latter case stimulated by sensing of viral nucleosides by the cGAS-STING system.56 It is interesting to note that mitochondrial DNA and RNA (e.g., in the context of aging) have been reported to contribute to activation of the cGAS-STING responses.57 Since mature sickle RBC have been reported to contain up to five to six residual mitochondria per cell,58,59 in light of the elevated degree of sickle RBC hemolysis, it is tempting to speculate that the elevated kynurenine levels as a function of high degree of hemolysis reported here participate in a similar circulating mitochondrial DNA sensing cascade, which could pave the way for potential novel therapeutic interventions. Finally, it is worth noting that elevation of kynurenine and indoles subtracts a substrate for the synthesis of serotonin, which is an essential component of platelet dense granules60 and a neurotransmitter that could contribute to counteracting nociceptive signaling.61 Here we performed network analyses of combined metabolomics and clinical data, which showed a strong association between hematological (e.g., HbS%) and clinical parameters with all metabolites mentioned in this discussion, as well as free fatty acids and acyl-carnitines. Of note, elevated kynurenine levels had been previously associated with increased susceptibility to hemolysis in the context of obesity,62 whereby depletion in circulating free fatty acids (previously also reported in pediatric patients with SCD63) and elevation acyl-carnitines had been identified as a marker of increased membrane RBC lipid damage and remodeling. This observation was in keeping with previous reports on the activation of the so-called Lands cycle in the context of SCD64 (or even just sickle cell trait),65 or other osmotic, mechanical or oxidative stressors to RBC, such as blood storage,66 exercise,67 chronic68 or acute kidney dysfunction.69 Of note, our findings expand on these observations, by identifying acyl-carnitine accumulation (and correction thereof by transfusion more than hydroxyurea treatment) as a hallmark of cardiorenal dysfunction in the sickle cell patient population. Interventions with carnitine supplementation in the context of SCD70 may thus be beneficial - especially in combination with hydroxyurea71 - to fuel the Lands cycle and contribute to prevent or mitigate RBC membrane lipid damage. Here we report a significant association between S1P depletion in plasma and genotypes with distinct severity of clinical manifestations (SS vs. SC), including pain crisis events. In previous studies, we and others had reported an elevation of RBC levels of the sphingolipid S1P in association with disease severity22 or increased (normal and sickle) erythrocyte susceptibility to extravascular hemolysis.72 While the mechanism of action is unclear, it has been posited that S1P may contribute to 2,3-diphosphoglycerate-dependent modulation of oxygen off-loading.22 By contributing to deoxyhemoglobin stabilization, hypoxia-induced elevation of S1P73 may cooperate with DPG to impact the so-called oxygen-dependent metabolic modulation that involves glycolytic enzyme sequestration to the membrane via binding to band 3, which in turn regulates RBC capacity to activate the PPP to generate reducing equivalents to sustain antioxidant challenges.74 ,7 5 However, here we observe a significant depletion of circulating S1P in plasma from subjects with SS genotypes, especially upon correction for transfusion events or hydroxyurea treatment. These results are suggestive of a role for decreased export, rather than (or in combination with) elevated synthesis may be responsible for dysregulated S1P metabolism in SCD. These results could be perhaps explained by polymorphisms in the S1P transporter Mfsd2b76 - previously associated with elevated susceptibility to osmotic fragility in healthy blood donor volunteers and sickle cell patients.37

Figure 7.Summary of the main metabolic pathways affected in plasma of sickle cell disease patients that correlate with relevant clinical outcomes. Plasma metabolomics analyses identified dysregulation of (A) tryptophan/kynurenine/indole metabolism, (B) conjugated bile acids, (C) elevation of acyl-carnitines, (D) mitochondrial dysfunction, (E) arginine metabolism as critically altered metabolic pathways associated with cardiorenal dysfunction and hazard ratios in this patient population.

In brief, while focusing only on central carbon and nitrogen metabolism with limited coverage of the lipidome or exposome,77 our study expands on existing literature and provides an extensive overview of the plasma metabolome in SCD and its clinical correlates, to the extent they are affected by common interventions like blood transfusion or treatment with hydroxyurea. The generation of a novel online portal as a repository for all the data described in this study will facilitate further dissemination and elaborations of the data presented herein in the future.

Footnotes

  • Received April 5, 2023
  • Accepted June 30, 2023

Correspondence

A. D’Alessandro angelo.dalessandro@cuanschutz.edu

M.T. Gladwin mgladwin@som.umaryland.edu

Disclosures

AD is a founder of Omix Technologies Inc. and Altis Biosciences LLC. AD is also a consultant for Hemanext Inc and Macopharma Inc. MTG is a co-inventor of patents and patent applications directed to the use of recombinant neuroglobin and heme-based molecules as antidotes for CO poisoning, which have been licensed by Globin Solutions, Inc. MTG is a shareholder, advisor, and director in Globin Solutions, Inc. MTG is also co-inventor on patents directed to the use of nitrite salts in cardiovascular diseases, which were previously licensed to United Therapeutics, and is now licensed to Globin Solutions and Hope Pharmaceuticals. MTG is a principal investigator in this investigator-initiated research study with Bayer Pharmaceuticals to evaluate riociguate as a treatment for patients with SCD. MTG has previously served on Bayer HealthCare LLC’s Heart and Vascular Disease Research Advisory Board and previously on the Forma Therapeutics scientific advisory board. All the other authors disclose no conflicts of interest relevant to this study. JME has provided consulting services for Elli Lily and Gilead Sciences and serves in the advisory board of Perha Pharmaceuticals.

Contributions

SMN, YZ, VG, MTG designed and executed the WALKPHASST study, collected and stored the samples, performed clinical measurements. FC, FG, JAR and AD performed metabolomics analyses (untargeted and targeted quantitative). KB, MG and JME built the SCD metabolome portal. AD performed data analysis and prepared figures and tables. AD wrote the first draft of the manuscript, which was revised by all the other authors. All the authors contributed to finalizing the manuscript.

Data-sharing statement

All the data generated in this study are available in the Online Supplementary Table S1 and interactively accessible through the ad hoc data portal https://mirages.shinyapps.io/SCD2023/.

Funding

AD was supported by funds by the National Institute of General and Medical Sciences (RM1GM131968), and from the National Heart, Lung, and Blood Institute (R01HL146442, R01HL149714, R01HL148151, R01HL161004). MTG receives research support from NIH grants 5R01HL098032, 2R01HL125886, 5P01HL103455, 5T32HL110849, and UH3HL143192. This work was supported by UH3HL143192. JME was supported by the National Institute of Allergy and Infectious Diseases (R01AI50305) and the Global Down Syndrome Foundation. The content is solely the responsibility of the authors and does not necessarily represent the ofcial views of the National Institutes of Health.

References

  1. Kato GJ, Piel FB, Reid CD. P. Sickle cell disease. Nat Rev Dis Primers. 2018; 4(1):18010. https://doi.org/10.1038/nrdp.2018.10PubMedGoogle Scholar
  2. da Guarda CC, Yahouédéhou S, Santiago RP. Sickle cell disease: a distinction of two most frequent genotypes (HbSS and HbSC). PLoS One. 2020; 15(1):e0228399. https://doi.org/10.1371/journal.pone.0228399PubMedPubMed CentralGoogle Scholar
  3. Rab MAE, van Oirschot BA, Bos J. Rapid and reproducible characterization of sickling during automated deoxygenation in sickle cell disease patients. Am J Hematol. 2019; 94(5):575-584. https://doi.org/10.1002/ajh.25443PubMedPubMed CentralGoogle Scholar
  4. Henry ER, Cellmer T, Dunkelberger EB. Allosteric control of hemoglobin S fiber formation by oxygen and its relation to the pathophysiology of sickle cell disease. Proc Natl Acand Sci U S A. 2020; 117(26):15018-15027. https://doi.org/10.1073/pnas.1922004117PubMedPubMed CentralGoogle Scholar
  5. McDade WA, Josephs R. On the formation and crystallization of sickle hemoglobin macrofibers. J Struct Biol. 1993; 110(1):90-97. https://doi.org/10.1006/jsbi.1993.1008PubMedGoogle Scholar
  6. Connes P, Alexy T, Detterich J, Romana M, Hardy-Dessources MD, Ballas SK. The role of blood rheology in sickle cell disease. Blood Rev. 2016; 30(2):111-118. https://doi.org/10.1016/j.blre.2015.08.005PubMedPubMed CentralGoogle Scholar
  7. Adebiyi MG, Manalo JM, Xia Y. Metabolomic and molecular insights into sickle cell disease and innovative therapies. Blood Adv. 2019; 3(8):1347-1355. https://doi.org/10.1182/bloodadvances.2018030619PubMedPubMed CentralGoogle Scholar
  8. Rees DC, Williams TN, Gladwin MT. Sickle-cell disease. Lancet. 2010; 376(9757):2018-2031. https://doi.org/10.1016/S0140-6736(10)61029-XPubMedGoogle Scholar
  9. Saraf SL, Molokie RE, Nouraie M. Differences in the clinical and genotypic presentation of sickle cell disease around the world. Paediatr Respir Rev. 2014; 15(1):4-12. https://doi.org/10.1016/j.prrv.2013.11.003PubMedPubMed CentralGoogle Scholar
  10. Keikhaei B, Yousefi H, Bahadoram M. Hydroxyurea: clinical and hematological effects in patients with sickle cell anemia. Glob J Health Sci. 2015; 8(3):252-256. https://doi.org/10.5539/gjhs.v8n3p252PubMedPubMed CentralGoogle Scholar
  11. Rankine-Mullings AE, Nevitt SJ. Hydroxyurea (hydroxycarbamide) for sickle cell disease. Cochrane Database Syst Rev. 2022; 9(9):CD002202. https://doi.org/10.1002/14651858.CD002202.pub3PubMedPubMed CentralGoogle Scholar
  12. Houten SM. Metabolomics: unraveling the chemical individuality of common human diseases. Ann Med. 2009; 41(6):402-407. https://doi.org/10.1080/07853890902729794PubMedGoogle Scholar
  13. Morris CR, Kato GJ, Poljakovic M. Dysregulated arginine metabolism, hemolysis-associated pulmonary hypertension, and mortality in sickle cell disease. JAMA. 2005; 294(1):81-90. https://doi.org/10.1001/jama.294.1.81PubMedPubMed CentralGoogle Scholar
  14. Kato GJ, Steinberg MH, Gladwin MT. Intravascular hemolysis and the pathophysiology of sickle cell disease. J Clin Invest. 2017; 127(3):750-760. https://doi.org/10.1172/JCI89741PubMedPubMed CentralGoogle Scholar
  15. Vallelian F, Buehler PW, Schaer DJ. Hemolysis, free hemoglobin toxicity, and scavenger protein therapeutics. Blood. 2022; 140(17):1837-1844. https://doi.org/10.1182/blood.2022015596PubMedPubMed CentralGoogle Scholar
  16. Buehler PW, Swindle D, Pak DI. Murine models of sickle cell disease and beta-thalassemia demonstrate pulmonary hypertension with distinctive features. Pulm Circ. 2021; 11(4):20458940211055996. https://doi.org/10.1177/20458940211055996PubMedPubMed CentralGoogle Scholar
  17. Niihara Y, Miller ST, Kanter J. A phase 3 trial of l-glutamine in sickle cell disease. N Angl J Med. 2018; 379(3):226-235. https://doi.org/10.1056/NEJMoa1715971PubMedGoogle Scholar
  18. Niihara Y, Zerez CR, Akiyama DS, Tanaka KR. Oral L-glutamine therapy for sickle cell anemia: I. subjective clinical improvement and favorable change in red cell NAD redox potential. American J Hematol. 1998; 58(2):117-121. https://doi.org/10.1002/(SICI)1096-8652(199806)58:2<117::AID-AJH5>3.0.CO;2-VGoogle Scholar
  19. Rab MAE, Van Oirschot BA, Kosinski PA. AG-348 (Mitapivat), an allosteric activator of red blood cell pyruvate kinase, increases enzymatic activity, protein stability, and ATP levels over a broad range of PKLR genotypes. Haematologica. 2021; 106(1):238-249. https://doi.org/10.3324/haematol.2019.238865PubMedPubMed CentralGoogle Scholar
  20. Song A, Wen AQ, Wen YE. p97 dysfunction underlies a loss of quality control of damaged membrane proteins and promotes oxidative stress and sickling in sickle cell disease. FASEB J. 2022; 36(5):e22246. https://doi.org/10.1096/fj.202101500RRPubMedGoogle Scholar
  21. Zhang Y, Dai Y, Wen J. Detrimental effects of adenosine signaling in sickle cell disease. Nat Med. 2011; 17(1):79-86. https://doi.org/10.1038/nm.2280PubMedPubMed CentralGoogle Scholar
  22. Sun K, D'Alessandro A, Ahmed MH. Structural and functional insight of sphingosine 1-phosphate-mediated pathogenic metabolic reprogramming in sickle cell disease. Sci Rep. 2017; 7(1):15281. https://doi.org/10.1038/s41598-017-13667-8PubMedPubMed CentralGoogle Scholar
  23. D'Alessandro A, Nouraie SM, Zhang Y. In vivo evaluation of the effect of sickle cell hemoglobin S, C and therapeutic transfusion on erythrocyte metabolism and cardiorenal dysfunction. Am J Hematol. 2023; 98(7):1017-1028. https://doi.org/10.1002/ajh.26923PubMedPubMed CentralGoogle Scholar
  24. Gladwin MT, Barst RJ, Gibbs JS. Risk factors for death in 632 patients with sickle cell disease in the United States and United Kingdom. PLoS One. 2014; 9(7):e99489. https://doi.org/10.1371/journal.pone.0099489PubMedPubMed CentralGoogle Scholar
  25. Machado RF, Barst RJ, Yovetich NA. Hospitalization for pain in patients with sickle cell disease treated with sildenafil for elevated TRV and low exercise capacity. Blood. 2011; 118(4):855-64. https://doi.org/10.1182/blood-2010-09-306167PubMedPubMed CentralGoogle Scholar
  26. Milton JN, Rooks H, Drasar E. Genetic determinants of haemolysis in sickle cell anaemia. Br J Haematol. 2013; 161(2):270-278. https://doi.org/10.1111/bjh.12245PubMedPubMed CentralGoogle Scholar
  27. Nouraie M, Lee JS, Zhang Y. The relationship between the severity of hemolysis, clinical manifestations and risk of death in 415 patients with sickle cell anemia in the US and Europe. Haematologica. 2013; 98(3):464-472. https://doi.org/10.3324/haematol.2012.068965PubMedPubMed CentralGoogle Scholar
  28. Sachdev V, Kato GJ, Gibbs JS. Echocardiographic markers of elevated pulmonary pressure and left ventricular diastolic dysfunction are associated with exercise intolerance in adults and adolescents with homozygous sickle cell anemia in the United States and United Kingdom. Circulation. 2011; 124(13):1452-1460. https://doi.org/10.1161/CIRCULATIONAHA.111.032920PubMedPubMed CentralGoogle Scholar
  29. Kato GJ. TRV: a physiological biomarker in sickle cell disease. Pediatr Blood Cancer. 2012; 58(6):831-832. https://doi.org/10.1002/pbc.23399PubMedPubMed CentralGoogle Scholar
  30. Bertolone L, Shin HK, Stefanoni D. ZOOMICS: comparative metabolomics of red blood cells from Old World monkeys and humans. Front Physiol. 2020; 11:593841. https://doi.org/10.3389/fphys.2020.593841PubMedPubMed CentralGoogle Scholar
  31. Stefanoni D, Shin HKH, Baek JH. Red blood cell metabolism in Rhesus macaques and humans: comparative biology of blood storage. Haematologica. 2020; 105(8):2174-2186. https://doi.org/10.3324/haematol.2019.229930PubMedPubMed CentralGoogle Scholar
  32. Nemkov T, Hansen KC, D'Alessandro A. A three-minute method for high-throughput quantitative metabolomics and quantitative tracing experiments of central carbon and nitrogen pathways. Rapid Commun Mass Spectrom. 2017; 31(8):663-673. https://doi.org/10.1002/rcm.7834PubMedPubMed CentralGoogle Scholar
  33. Nemkov T, Reisz JA, Gehrke S, Hansen KC, D'Alessandro A. High-throughput metabolomics: isocratic and gradient mass spectrometry-based methods. Methods Mol Biol. 2019; 1978:13-26. https://doi.org/10.1007/978-1-4939-9236-2_2PubMedGoogle Scholar
  34. D'Alessandro A, Culp-Hill R, Reisz JA. Heterogeneity of blood processing and storage additives in different centers impacts stored red blood cell metabolism as much as storage time: lessons from REDS-III-Omics. Transfusion. 2019; 59(1):89-100. https://doi.org/10.1111/trf.14979PubMedPubMed CentralGoogle Scholar
  35. Pang Z, Chong J, Zhou G. MetaboAnalyst 5.0: narrowing the gap between raw spectra and functional insights. Nucleic Acids Res. 2021; 49(W1):W388-W396. https://doi.org/10.1093/nar/gkab382PubMedPubMed CentralGoogle Scholar
  36. Minniti CP, Sable C, Campbell A. Elevated tricuspid regurgitant jet velocity in children and adolescents with sickle cell disease: association with hemolysis and hemoglobin oxygen desaturation. Haematologica. 2009; 94(3):340-347. https://doi.org/10.3324/haematol.13812PubMedPubMed CentralGoogle Scholar
  37. Page GP, Kanias T, Guo YJ. Multiple-ancestry genome-wide association study identifies 27 loci associated with measures of hemolysis following blood storage. J Clin Invest. 2021; 131(13):e146077. https://doi.org/10.1172/JCI146077PubMedPubMed CentralGoogle Scholar
  38. Sullivan KD, Galbraith MD, Kinning KT. The COVIDome Explorer researcher portal. Cell Rep. 2021; 36(7):109527. https://doi.org/10.1016/j.celrep.2021.109527PubMedPubMed CentralGoogle Scholar
  39. Ribeiro PR, Teixeira RdS, Souza AR. Blood plasma metabolomics of children and adolescents with sickle cell anaemia treated with hydroxycarbamide: a new tool for uncovering biochemical alterations. Br J Haematol. 2021; 192(5):922-931. https://doi.org/10.1111/bjh.17315PubMedGoogle Scholar
  40. Culp-Hill R, Srinivasan AJ, Gehrke S. Effects of red blood cell (RBC) transfusion on sickle cell disease recipient plasma and RBC metabolism. Transfusion. 2018; 58(12):2797-2806. https://doi.org/10.1111/trf.14931PubMedPubMed CentralGoogle Scholar
  41. Darghouth D, Koehl B, Madalinski G. Pathophysiology of sickle cell disease is mirrored by the red blood cell metabolome. Blood. 2011; 117(6):e57-e66. https://doi.org/10.1182/blood-2010-07-299636PubMedGoogle Scholar
  42. Ilboudo Y, Garrett ME, Bartolucci P. Potential causal role of l-glutamine in sickle cell disease painful crises: A Mendelian randomization analysis. Blood Cells Mol Dis. 2021; 86:102504. https://doi.org/10.1016/j.bcmd.2020.102504PubMedPubMed CentralGoogle Scholar
  43. Selak MA, Armour SM, MacKenzie ED. Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-α prolyl hydroxylase. Cancer Cell. 2005; 7(1):77-85. https://doi.org/10.1016/j.ccr.2004.11.022PubMedGoogle Scholar
  44. Tannahill GM, Curtis AM, Adamik J. Succinate is an inflammatory signal that induces IL-1β through HIF-1α. Nature. 2013; 496(7444):238-242. https://doi.org/10.1038/nature11986PubMedPubMed CentralGoogle Scholar
  45. Ridlon JM, Kang DJ, Hylemon PB, Bajaj JS. Bile acids and the gut microbiome. Curr Opin Gastroenterol. 2014; 30(3):332-338. https://doi.org/10.1097/MOG.0000000000000057PubMedPubMed CentralGoogle Scholar
  46. Soderborg TK, Clark SE, Mulligan CE. The gut microbiota in infants of obese mothers increases inflammation and susceptibility to NAFLD. Nat Commun. 2018; 9(1):4462. https://doi.org/10.1038/s41467-018-06929-0PubMedPubMed CentralGoogle Scholar
  47. El Kasmi KC, Vue PM, Anderson AL. Macrophage-derived IL-1beta/NF-kappaB signaling mediates parenteral nutrition-associated cholestasis. Nat Commun. 2018; 9(1):1393. https://doi.org/10.1038/s41467-018-03764-1PubMedPubMed CentralGoogle Scholar
  48. Wiener G, Moore HB, Moore EE. Shock releases bile acid inducing platelet inhibition and fibrinolysis. J Surg Res. 2015; 195(2):390-395. https://doi.org/10.1016/j.jss.2015.01.046PubMedPubMed CentralGoogle Scholar
  49. Roager HM, Licht TR. Microbial tryptophan catabolites in health and disease. Nat Commun. 2018; 9(1):3294. https://doi.org/10.1038/s41467-018-05470-4PubMedPubMed CentralGoogle Scholar
  50. Su X, Gao Y, Yang R. Gut microbiota-derived tryptophan metabolites maintain gut and systemic homeostasis. Cells. 2022; 11(15):2296. https://doi.org/10.3390/cells11152296PubMedPubMed CentralGoogle Scholar
  51. Clement CC, D'Alessandro A, Thangaswamy S. 3-hydroxyL-kynurenamine is an immunomodulatory biogenic amine. Nat Commun. 2021; 12(1):4447. https://doi.org/10.1038/s41467-021-24785-3PubMedPubMed CentralGoogle Scholar
  52. Dvořák Z, Poulíková K, Mani S. Indole scaffolds as a promising class of the aryl hydrocarbon receptor ligands. Eur J Med Chem. 2021; 215:113231. https://doi.org/10.1016/j.ejmech.2021.113231PubMedGoogle Scholar
  53. Powers RK, Culp-Hill R, Ludwig MP. Trisomy 21 activates the kynurenine pathway via increased dosage of interferon receptors. Nat Commun. 2019; 10(1):4766. https://doi.org/10.1038/s41467-019-12739-9PubMedPubMed CentralGoogle Scholar
  54. D'Alessandro A, Thomas T, Akpan IJ. Biological and clinical factors contributing to the metabolic heterogeneity of hospitalized patients with and without COVID-19. Cells. 2021; 10(9):2293. https://doi.org/10.3390/cells10092293PubMedPubMed CentralGoogle Scholar
  55. Thomas T, Stefanoni D, Reisz JA. COVID-19 infection alters kynurenine and fatty acid metabolism, correlating with IL-6 levels and renal status. JCI Insight. 2020; 5(14):e140327. https://doi.org/10.1172/jci.insight.140327PubMedPubMed CentralGoogle Scholar
  56. Domizio JD, Gulen MF, Saidoune F. The cGAS–STING pathway drives type I IFN immunopathology in COVID-19. Nature. 2022; 603(7899):145-151. https://doi.org/10.1038/s41586-022-04421-wPubMedPubMed CentralGoogle Scholar
  57. Huang LS, Hong Z, Wu W. mtDNA activates cGAS signaling and suppresses the YAP-mediated endothelial cell proliferation program to promote inflammatory injury. Immunity. 2020; 52(3):475-486. https://doi.org/10.1016/j.immuni.2020.02.002PubMedPubMed CentralGoogle Scholar
  58. Jagadeeswaran R, Lenny H, Vazquez B. The abnormal presence of mitochondria in circulating red blood cells cause an increased oxygen consumption rate, ROS generation and hemolysis in patients with sickle cell disease. Blood. 2017; 130Google Scholar
  59. Moriconi C, Dzieciatkowska M, Roy M. Retention of functional mitochondria in mature red blood cells from patients with sickle cell disease. Br J Haematol. 2022; 198(3):574-586. https://doi.org/10.1111/bjh.18287PubMedPubMed CentralGoogle Scholar
  60. Hanby HA, Bao J, Noh J-Y. Platelet dense granules begin to selectively accumulate mepacrine during proplatelet formation. Blood Adv. 2017; 1(19):1478-1490. https://doi.org/10.1182/bloodadvances.2017006726PubMedPubMed CentralGoogle Scholar
  61. Dembélé KC, Mintz T, Veyrat-Durebex C. Metabolomic profiling of plasma and erythrocytes in sickle mice points to altered nociceptive pathways. Cells. 2020; 9(6):1334. https://doi.org/10.3390/cells9061334PubMedPubMed CentralGoogle Scholar
  62. Hazegh K, Fang F, Bravo MD. Blood donor obesity is associated with changes in red blood cell metabolism and susceptibility to hemolysis in cold storage and in response to osmotic and oxidative stress. Transfusion. 2021; 61(2):435-448. https://doi.org/10.1111/trf.16168PubMedPubMed CentralGoogle Scholar
  63. Qu H-Q, Glessner J, Qu J. Metabolomic profiling for dyslipidemia in pediatric patients with sickle cell disease, on behalf of the IHCC consortium. Metabolomics. 2022; 18(12):101. https://doi.org/10.1007/s11306-022-01954-zPubMedPubMed CentralGoogle Scholar
  64. Wu H, Bogdanov M, Zhang Y. Hypoxia-mediated impaired erythrocyte Lands’ Cycle is pathogenic for sickle cell disease. Sci Rep. 2016; 6(1):29637. https://doi.org/10.1038/srep29637PubMedPubMed CentralGoogle Scholar
  65. Nemkov T, Skinner S, Diaw M. Plasma levels of acylcarnitines and carboxylic acids correlate with cardiovascular and kidney function in subjects with sickle cell trait. Front Physiol. 2022; 13:916197. https://doi.org/10.3389/fphys.2022.916197PubMedPubMed CentralGoogle Scholar
  66. Yoshida T, Prudent M, D'Alessandro A. Red blood cell storage lesion: causes and potential clinical consequences. Blood Transfus. 2019; 17(1):27-52. Google Scholar
  67. Nemkov T, Skinner SC, Nader E. Acute cycling exercise induces changes in red blood cell deformability and membrane lipid remodeling. Int J Mol Sci. 2021; 22(2):896. https://doi.org/10.3390/ijms22020896PubMedPubMed CentralGoogle Scholar
  68. Bissinger R, Nemkov T, D'Alessandro A. Proteinuric chronic kidney disease is associated with altered red blood cell lifespan, deformability and metabolism. Kidney Int. 2021; 100(6):1227-1239. https://doi.org/10.1016/j.kint.2021.08.024PubMedGoogle Scholar
  69. Xu P, Chen C, Zhang Y. Erythrocyte transglutaminase-2 combats hypoxia and chronic kidney disease by promoting oxygen delivery and carnitine homeostasis. Cell Metab. 2022; 34(2):299-316. https://doi.org/10.1016/j.cmet.2021.12.019PubMedPubMed CentralGoogle Scholar
  70. El-Beshlawy A, Abd El Raouf E. Diastolic dysfunction and pulmonary hypertension in sickle cell anemia: is there a role for L-carnitine treatment?. Acta Haematol. 2006; 115(1-2):91-96. https://doi.org/10.1159/000089472PubMedGoogle Scholar
  71. A. Khaled S, M. Ashry I. Drug therapy in patients with severe forms of sickle cell anemia: a nonrandomized clinical trial of combining l-carnitine with hydroxycarbamide therapy. J Appl Hematol. 2022; 13(4):237-248. https://doi.org/10.4103/joah.joah_23_22Google Scholar
  72. Hay AM, Nemkov T, Gamboni F. S1P has a negative effect on RBC storage quality. Blood Adv. 2023; 7(8):1379-1393. https://doi.org/10.1182/bloodadvances.2022008936PubMedPubMed CentralGoogle Scholar
  73. Sun K, Zhang Y, D'Alessandro A. Sphingosine-1-phosphate promotes erythrocyte glycolysis and oxygen release for adaptation to high-altitude hypoxia. Nat Commun. 2016; 7:12086. https://doi.org/10.1038/ncomms12086PubMedPubMed CentralGoogle Scholar
  74. Lewis IA, Campanella ME, Markley JL, Low PS. Role of band 3 in regulating metabolic flux of red blood cells. Proc Natl Acad Sci U S A. 2009; 106(44):18515-18520. https://doi.org/10.1073/pnas.0905999106PubMedPubMed CentralGoogle Scholar
  75. Issaian A, Hay A, Dzieciatkowska M. The interactome of the N-terminus of band 3 regulates red blood cell metabolism and storage quality. Haematologica. 2021; 106(11):2971-2985. https://doi.org/10.3324/haematol.2020.278252PubMedPubMed CentralGoogle Scholar
  76. Vu TM, Ishizu AN, Foo JC. Mfsd2b is essential for the sphingosine-1-phosphate export in erythrocytes and platelets. Nature. 2017; 550(7677):524-528. https://doi.org/10.1038/nature24053PubMedGoogle Scholar
  77. Nemkov T, Stefanoni D, Bordbar A. Blood donor exposome and impact of common drugs on red blood cell metabolism. JCI Insight. 2021; 6(3):e146175. https://doi.org/10.1172/jci.insight.146175PubMedPubMed CentralGoogle Scholar

Data Supplements

Figures & Tables

Article Information

Vol. 108 No. 12 (2023): December, 2023 : Articles
 
DOI
https://doi.org/10.3324/haematol.2023.283288
Pubmed
37439373
 
Published
2023-07-13
Published By
Ferrata Storti Foundation, Pavia, Italy
Print ISSN
0390-6078
Online ISSN
1592-8721
 
Copyright & Usage
Copyright (c) 2023 Ferrata Storti Foundation
Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Article Usage

Online Views
1444
PDF Downloads
681

Statistics from Altmetric.com

No Data
Navigate
For Authors
For Reviewers
For Advertisers
Education
Privacy
More
Copyright © 2024 by the Ferrata Storti Foundation | Web design ￫ | Development ￫
ISSN 0390-6078 print | ISSN 1592-8721 online