Increased Na(+)-Ca(2+) exchanger in the failing heart.

Heart failure (HF) is a significant problem, affecting more than 2 million people in the United States alone.1 When severe, it is associated with a 50% 2-year mortality rate attributable to either contractile dysfunction or sudden death from ventricular tachycardia and ventricular fibrillation.1 The underlying molecular mechanisms have not yet been defined, but there is considerable evidence from studies in experimental models of HF and in the failing human heart that alterations in intracellular Ca handling could play a role in both the contractile abnormalities and arrhythmogenicity.2,3 Specifically, HF cardiac myocytes exhibit Ca transients with decreased amplitude and prolonged Ca decline.4,5 These changes have been associated with altered expression and function of several Ca regulatory proteins. There has been considerable interest in alterations of sarcoplasmic reticulum (SR) function, particularly the SR Ca-ATPase (SERCA) and its inhibition by phospholamban. Reduced SERCA mRNA has been noted in animal models of HF6,7 and in the failing human heart,6,8,9 but alterations in SERCA protein levels and SR Ca uptake have been inconsistent and more controversial.10,11 The same is true for changes in phospholamban.9–12 More recently, additional attention has been focused on the Na-Ca exchanger (NCX). NCX is a transsarcolemmal protein that plays an important role in controlling levels of [Ca]i.13 NCX can operate in both a forward mode (Ca out, Na in) and a reverse mode (Na out, Ca in), and most evidence to date14 has suggested that it does so with a stoichiometry of 3:1 (ie, it exchanges 3 Na ions for every 1 Ca ion). As a result, NCX is electrogenic, producing an inward current (INCX) when Ca is extruded from the cell. This current is believed to underlie the development of an arrhythmogenic transient inward current that produces delayed afterdepolarizations (DADs) and triggered activity.15 NCX activity is regulated by levels of intracellular Ca and Na,16,17 although the magnitude and physiological significance of this regulation remains to be defined. NCX has been found to be upregulated (on both an mRNA level and a protein level) in the failing human heart18,19 and in some experimental models of HF.5,20 However, there is limited data on NCX function in HF. Litwin and Bridge21 demonstrated increased INCX in the infarcted rabbit heart. We recently showed a parallel 2-fold increase in NCX mRNA, protein, rate of [Ca]i decline, and INCX (both inward and outward) in an arrhythmogenic rabbit model of heart failure (combined aortic insufficiency and aortic constriction).20 NCX expression and function have been extensively studied in the pacing-induced HF model in dogs and rabbits. Although this model lacks myocardial hypertrophy and the heart failure is reversible with termination of pacing,5,22 it exhibits severe left ventricular systolic dysfunction, impaired relaxation, increased LV filling pressures, and alterations in ionic currents similar to those of human heart failure.5,23 Myocardium from dogs with pacing-induced HF exhibits significant downregulation of SERCA (mRNA and protein), and isolated HF myocytes exhibit Ca transients with decreased amplitude and prolonged Ca declines (consistent with the decreased SERCA expression).5 However, the results of studies of NCX function have been conflicting. O’Rourke et al5 studied dogs with pacing-induced HF and reported a 2-fold increase in NCX protein compared with controls. NCX function (assessed by the rate of Ca decline with SERCA inhibited by cyclopiazonic acid) was increased, but not significantly. Rose et al24 studied rabbits with pacinginduced HF and found a 44% increase in NCX protein versus control but no change in INCX density. In contrast, using a similar model, Yao et al25 found decreased NCX mRNA levels associated with a significant reduction in INCX. Thus, it remains unclear whether and how NCX function is altered in the pacing-induced HF model. In this issue of Circulation Research, Hobai and O’Rourke26 report on NCX function in the canine-pacing HF model and expand on their previous work.5 They assessed NCX function by numerous approaches and under various conditions. When [Ca]i was buffered to 200 nmol/L, INCX (measured as Ni-sensitive current) was similar in both HF and control myocytes. But when they allowed [Ca]i to rise freely (by buffering with only 50 mmol/L indo-1 in the pipette), they found that the decline of normalized caffeine transients (reflecting Ca efflux primarily by NCX) was 2.2-fold faster in HF. When NCX function was assessed with SR Ca uptake inhibited by thapsigargin (again with minimal Ca buffering), depolarizing pulses increased reversemode NCX activity (increased [Ca]i) and outward INCX (consistent with the 2-fold increased NCX protein levels).5,20 Repolarizing pulses increased inward INCX and rate of [Ca]i decline, but INCX was not different when normalized to [Ca]i. Thus, Ca transport and outward INCX seem to be upregulated, but not inward INCX. This presents an interesting paradox: how do more NCX molecules enhance Ca efflux without increasing inward INCX? This also raises several questions. The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association. From the Department of Medicine, Section of Cardiology, University of Illinois at Chicago. Correspondence to Steven M. Pogwizd, MD, Department of Medicine, Section of Cardiology, University of Illinois at Chicago, 840 S Wood St, M/C787, Chicago, IL 60612-7323. E-mail [email protected] (Circ Res. 2000;87:641-643.) © 2000 American Heart Association, Inc.