The Heart in Sickle Cell Disease, a Model for Heart Failure with Preserved Ejection Fraction
Total Page:16
File Type:pdf, Size:1020Kb
COMMENTARY COMMENTARY The heart in sickle cell disease, a model for heart failure with preserved ejection fraction John C. Wooda,1 Sickle cell disease (SCD) is caused by a point mutation production, favoring smooth muscle cell proliferation in the β-subunit of hemoglobin. The abnormal he- and collagen deposition (7). moglobin polymerizes after releasing oxygen to the Cell-free hemoglobin is also toxic. Circulating hap- tissues, producing long filaments that contort the toglobin binds and clears free hemoglobin efficiently erythrocyte into a characteristic sickle-shaped confor- in normal subjects, but this buffering system is over- mation. Cardiopulmonary complications, including pul- whelmed in SCD (8). Free hemoglobin avidly binds monary hypertension, diastolic dysfunction, and sudden nitric oxide and can scavenge NO close to the vessel death (1), are the most common causes of death in SCD. wall, unlike hemoglobin contained in red blood cells (9). In PNAS, Bakeer et al. (2) delve into the pathobiology of Acute hemolysis promotes vascular inflammation that these changes. can be blocked by nitric oxide donors (10). In SCD The link between sickle hemoglobin and cardiac patients, circulating cell-free hemoglobin levels corre- disease is fairly convoluted. Chronic vascular damage late with endothelial reactivity and pulmonary artery occurs through parallel pathways (Fig. 1, Left). Red cells pressures (11). are more rigid, have increased adhesion molecule Circulating free hemoglobin can also be stripped expression, and become progressively dehydrated, of its protein shell, leaving free heme groups. Heme promoting microvascular obstruction. Direct vaso- moities freely participate in redox cycling, causing pro- occlusion forms the basis for classic SCD complica- found endothelial damage, and can activate the innate tions, such as dactylitis, pain crisis, osteonecrosis, splenic immune system through Toll-like receptor 4 (12). A sec- infarction, and acute chest syndrome. ond buffering protein, hemopexin, is present to remove However, vascular damage in SCD also occurs circulating free heme but is depleted in SCD patients more insidiously. Red and white blood cell adhesion (13). As a result, micromolar heme injections can produce to vascular endothelia promotes inflammation, and lethal acute chest syndrome in sickle mice, recapitulating there is a background of recurrent microscopic ische- human disease (14). Hemopexin depletion and heme mia-reperfusion injury in many organs (3). Repeated toxicity are not unique to SCD, but are also critically conformational changes with oxygenation and deoxy- important in septic shock (15). genation prematurely age red blood cells, shortening Vaso-occlusive and hemolytic vascular stressors their lifespan 5- to 10-fold, causing anemia and tissue reinforce one another to compound vascular damage. hypoxia. Rapid red cell turnover is partially buffered by Hence, it is not surprising that all major organs of the increased erythropoiesis. Unfortunately, erythropoiesis body are affected (Fig. 1, Right). Pulmonary arterial is metabolically demanding, and increases procoagulant hypertension is caused by pulmonary vascular con- autophagic vesicles (4). In addition, developing erythro- striction and increased left ventricular filling pressures, blasts produce a number of vasoactive signaling mol- as well as the interaction of both mechanisms (16). ecules, including placental growth factor, that are Whereas hemolysis-induced pulmonary arterial hyper- associated with airway hyperactivity, monocyte activa- tension has been relatively well studied, mechanisms tion, and pulmonary hypertension (5). for the diastolic dysfunction and sudden death are Some of the red blood cells escape clearance by the poorly understood in SCD patients. reticuloendothelial system and rupture intravascularly, In PNAS, Bakeer et al. step firmly into this void (2). releasing arginase and erythroid damage-associated Using the Berk mouse model of SCD (a double murine molecular pattern molecules, including cell-free hemo- knockout which completely expresses human sickle he- globin (6). Arginase depletes an important substrate moglobin via a transgene), the authors demonstrate a of endothelial nitric oxide synthase, promoting eNOS novel, progressive cardiomyopathy with preserved sys- uncoupling, contributing to decreased nitric oxide bio- tolic function, concentric ventricular hypertrophy, im- availability. In addition, arginase promotes L-ornithine paired diastolic function, left atrial hypertrophy, and aDivision of Cardiology, Children’s Hospital Los Angeles, Los Angeles, CA 90027 Author contributions: J.C.W. wrote the paper. The author declares no conflict of interest. See companion article on page E5182. 1Email: [email protected]. 9670–9672 | PNAS | August 30, 2016 | vol. 113 | no. 35 www.pnas.org/cgi/doi/10.1073/pnas.1611899113 Downloaded by guest on September 29, 2021 Fig. 1. (Left) Schematic linking sickle hemoglobin to the multiple mechanisms by which it produces vascular damage. Broadly, these are divided into those caused by direct mechanical interactions of the red cell and endothelium and those caused by premature red cell destruction. (Right) Major organ systems affected in sickle cell disease; many organs experience both infarctions and interstitial fibrosis. increased interstitial fibrosis on histology and MRI. Using an ac- The cardiac changes in the Berk mouse bear striking similarity quired anemia model (iron deficiency) as a control, Bakeer et al. to those observed in heart failure with preserved ejection (HFpEF), demonstrate that these changes are not a consequence of chron- including concentric hypertrophy, diastolic dysfunction, systolic ically elevated cardiac output. Transplantation of sickle bone mar- hypertension, sudden death, microvascular destruction, and a row into a different background mouse strain recapitulated the proinflammatory/profibrotic mileau (20). HFpEF accounts for 50% phenotype, supporting generalizability of the pathophysiology. of heart failure in the general population (20). Loss of arterial On microscopic analysis, multiple, randomly distributed areas compliance, chronic oxidative stress, systolic hypertension, auto- of microvascular occlusion, ischemic cell death, and fibrosis were nomic system overstimulation, and chronic vascular inflammation present. Sarcomeres were shortened, suggesting impaired relaxa- play critical roles in both diseases (20). However, whereas insulin tion. Mitochondria were more abundant, hypertrophied, rounded, resistance and hyperlipidemia represent the primary vascular swollen, and had disrupted cristae, consistent with chronic ischemia. stressors in HFpEF, abnormal red cell mechanics and erythroid Myocardial transcriptome analysis demonstrated a marked up- damage-associated molecular pattern molecules are responsible regulation of genes associated with extracellular matrix produc- for chronic vascular inflammation and destruction in SCD. None- tion and angiogenesis. Transcriptome profile also suggested a theless, the two diseases appear to have convergent downstream metabolic shift from glucose to fatty acid metabolism, a change pathways and may benefit from discoveries in the other arena. commonly noted in type II diabetes and acute ischemia. CPT1 and Taken together then, the Bakeer et al. (2) data suggest that the CPT2, genes critical for very long-chain fatty acid production and heart suffers from similar progressive microvascular damage as membrane stabilization, were down-regulated. Congenital defi- observed in the brain, lungs, liver, and kidneys of SCD patients. ciencies in these genes are associated with hypertrophic cardio- Even though global cardiac oxygen delivery may be preserved, myopathy, diastolic dysfunction, and arrhythmias (17). loss of capillary cross-sectional area increases oxygen diffusion More importantly, Bakeer et al. (2) reproduced the repolarization distances and regional ischemia. Baseline cardiac blood flow is abnormalities and sudden-death risk that have baffled clinicians for elevated to compensate for anemia, blunting the vasodilatory decades. A shocking percentage of SCD patients die from sudden reserve in response to cardiovascular stress (such as pain crises death, despite grossly normal-appearing echocardiograms (1). The or acute hemolysis). Greater oxygen extraction increases the risk Berk mice exhibited both QRS and QT prolongation, consistent with of vaso-occlusion and further capillary destruction, completing a the observed replacement fibrosis. Roughly 40% of their mice died vicious cycle. Recent cardiac MRI studies suggest that microvas- suddenly. Electron microscopy demonstrated chaotic cardiomyo- cularmyocardialinfarctionisnotuncommoninSCDpatients cyteremodelingthatwouldexplaindispersioninconductionand during pain crisis and can be easily missed without a high index repolarization velocity. Such islands of impaired electrical conductiv- of suspicion. ity create potential substrates for unstable ventricular tachycardias Therefore, if we accept this model for cardiac disease in (Torsades de Pointes). Transcriptome analysis demonstrated de- humans with SCD, how might it change our clinical practice? The creased expression of SCN4B, AKAP9, CACNA1S, KCNJ2,and mainstay of therapy remains the amelioration of the primary hema- SCN4A, which have all been linked to long QT syndromes (18). tologic defect, including bone marrow transplantation, chronic Transcriptome profile also suggested disruptions in circadian transfusion therapy,