Impairment of HERG K(+) channel function by tumor necrosis factor-alpha: role of reactive oxygen species as a mediator.

Congestive heart failure (CHF) is associated with susceptibility to lethal arrhythmias and typically increases levels of tumor necrosis factor-alpha (TNF-alpha) and its receptor, TNFR1. CHF down-regulates rapid delayed-rectifier K(+) current (I(Kr)) and delays cardiac repolarization. We studied the effects of TNF-alpha on cloned HERG K(+) channel (human ether-a-go-go-related gene) in HEK293 cells and native I(Kr) in canine cardiomyocytes with whole-cell patch clamp techniques. TNF-alpha consistently and reversibly decreased HERG current (I(HERG)). Effects of TNF-alpha were concentration-dependent, increased with longer incubation period, and occurred at clinically relevant concentrations. TNF-alpha had similar inhibitory effects on I(Kr) and markedly prolonged action potential duration (APD) in canine cardiomyocytes. Immunoblotting analysis demonstrated that HERG protein level was slightly higher in canine hearts with tachypacing-induced CHF than in healthy hearts, and TNF-alpha slightly increased HERG protein level in CHF but not in healthy hearts. In cells pretreated with the inhibitory anti-TNFR1 antibody, TNF-alpha lost its ability to suppress I(HERG), indicating a requirement of TNFR1 activation for HERG suppression. Vitamin E or MnTBAP (Mn(III) tetrakis(4-benzoic acid) porphyrin chloride), a superoxide dismutase mimic) prevented, whereas the superoxide anion generating system xanthine/xanthine oxidase mimicked, TNF-alpha-induced I(HERG) depression. TNF-alpha caused robust increases in intracellular reactive oxygen species, and vitamin E and MnTBAP abolished the increases, in both HEK293 cells and canine ventricular myocytes. We conclude that the TNF-alpha/TNFR1 system impairs HERG/I(Kr) function mainly by stimulating reactive oxygen species, particularly superoxide anion, but not by altering HERG expression; the effect may contribute to APD prolongation by TNF-alpha and may be a novel mechanism for electrophysiological abnormalities and sudden death in CHF.

ating system xanthine/xanthine oxidase mimicked, TNF-␣-induced I HERG depression. TNF-␣ caused robust increases in intracellular reactive oxygen species, and vitamin E and MnTBAP abolished the increases, in both HEK293 cells and canine ventricular myocytes. We conclude that the TNF-␣/TNFR1 system impairs HERG/I Kr function mainly by stimulating reactive oxygen species, particularly superoxide anion, but not by altering HERG expression; the effect may contribute to APD prolongation by TNF-␣ and may be a novel mechanism for electrophysiological abnormalities and sudden death in CHF.
TNF-␣ 1 is a potent inducible cytokine with pleiotropic biological effects (1). Up-regulation of TNF-␣ is a consistent finding in clinical (2) and experimental CHF (3). Circulating concentrations of TNF-␣ and soluble TNF receptors are independent predictors of mortality in CHF (4).
Patients with CHF are at increased risk of sudden death due to cardiac arrhythmias. CHF increases action potential duration (APD) (5), leading to early afterdepolarizations (EADs) and lethal ventricular tachyarrhythmias (6). Polymorphic ventricular tachycardias, likely related to arrhythmogenic afterdepolarizations, are common in CHF (6,7). The molecular mechanisms underlying APD prolongation in CHF remain incompletely understood.
The rapid delayed rectifier K ϩ current (I Kr ) is crucial in cardiac repolarization. The human ether-a-go-go-related gene (HERG) encodes the pore-forming ␣-subunit of I Kr and congenital or drug-induced abnormalities in HERG protein function are a common cause of the long QT syndrome. Simulations of cellular electrophysiology predict I Kr inhibition to cause EADs in failing, but not nonfailing, myocytes (8). A recent study demonstrated that transgenic mice overexpressing TNF-␣ with heart failure had significantly prolonged APD (9). It is unknown whether TNF-␣ affects cardiac K ϩ channels. We therefore examined the hypothesis that TNF-␣ might affect HERG/I Kr , thereby potentially contributing to CHF-related repolarization abnormalities.

EXPERIMENTAL PROCEDURES
Cell Disposition-HEK293 cells stably expressing HERG were a kind gift from Drs. Zhou and January. Cell culture and handling procedures have been described previously (10). Cardiomyocytes were isolated from healthy adult mongrel dogs as described in detail previously (11,12). The procedures for animal use were in accordance with institutional guidelines.
Whole-cell Patch Clamp Recording-Patch clamp techniques have been described in detail elsewhere (13)(14)(15)(16). Experiments were conducted at 36 Ϯ 1°C. For current recordings in canine myocyte studies, the following were included in the bath to block contaminating currents: CdCl 2 (200-mol/liter, L-type Ca 2ϩ current), 4-aminopyridine (1 mmol/liter, transient outward K ϩ currents), glyburide (10 mol/liter, ATP-sensitive K ϩ current), and 293B (10 mol/liter, slow delayedrectifier K ϩ current). Action potentials were recorded in the current clamp mode with Tyrode solution free of ion channel blockers. TNF-␣ was either added to the extracellular solution 10 min after formation of * This work was supported in part by the Heart and Stroke Foundation of Canada and Fonds de la Recherche de l'Institut de Cardiologie de Montreal (awarded to Z. W.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Western Blot-The procedures were similar to those described previously (15). Polyclonal anti-HERG raised in rabbit against highly purified peptide (CY)EEL PAGAPELPQD GPT, corresponding to residues 1118 -1133 of human HERG was purchased from Alomone Laboratories (Jerusalem, Israel).
Data Analysis-Group data are mean Ϯ S.E. Paired t tests were used for single comparisons. Kinetics were analyzed with CLAMPFIT (pCLAMP 8.0) or Graphpad Prism.

RESULTS
I HERG was elicited by 2-s depolarizations followed by 2-s repolarizing steps (Fig. 1, inset). Currents were recorded immediately after formation of whole-cell configuration and series resistance compensation. Comparisons were made between control cells (without TNF-␣) and cells incubated for 10 h with various TNF-␣ concentrations from 0.01 to 10 ng/ml, which are within the pathophysiological range of TNF-␣ levels (ϳ0.1 ng/ ml) (17)(18)(19). I HERG density was reduced by TNF-␣, with effects that were concentration-and voltage-dependent, being larger at more negative potentials (Fig. 1A). I HERG kinetics were unaltered by TNF-␣.
Exposure to TNF-␣ for 15 min concentration-dependently decreased I HERG . I HERG amplitude was decreased by 9, 16, and 35% by TNF-␣ at 1, 10, and 100 ng/ml, respectively. Results at 100 ng/ml are shown in Fig. 1B. Depression of I Kr by TNF-␣ was reproduced in both dog atrial and ventricular myocytes (Fig. 1C). APD 50 and APD 90 , duration at 50 and 90% repolarization, respectively, were both significantly longer in single ventricular cells preincubated with TNF-␣ at 10 ng/ml in Tyrode solution for 10 h relative to control cells (Fig. 1D).
Western blot analysis of HERG protein levels in the membrane preparations extracted from HERG-expressing HEK293 cells and from the ventricular myocytes of healthy dogs or dogs with tachypacing induced CHF was performed. A band of around 135 kDa was identified by anti-HERG antibody, and the band was abolished after the antibody had been neutralized by its antigenic peptide. TNF-␣ treatment neither significantly alter HERG protein level in HEK293 cells nor in healthy dogs. HERG protein level was slightly higher in CHF than in healthy dogs and was slightly increased by TNF-␣ in CHF dogs ( Fig. 2A).
To clarify whether TNF-␣ acts on I HERG via activation of TNF receptor I (TNFR1), we incubated HEK293 cells with H389 (an inhibitory anti-TNFR1 antibody) for 1 h before patch clamp recording upon acute exposure to 100 ng/ml TNF-␣ or beginning 1 h before prolonged (10 h) exposure to 1 ng/ml TNF-␣. H389 prevented suppression of I HERG by subsequent acute or prolonged application of TNF-␣. Data from prolonged exposure experiments are shown in Fig. 2B.
To confirm that intracellular ROS production was indeed stimulated by TNF-␣, we detected ROS level using CM-H2DFDA fluorescence dye to stain the cells. The cells stained with fluorescence intensity Ն5 times the background were defined as positive staining, and the number of cells with positive staining was pooled from five fields. The intensity of staining was analyzed by densitometric scanning using the LSM pro-

Cellular Mechanisms for HERG Modulation by TNF-␣ 13290
gram, and the data were normalized to the control values without TNF-␣ (0.1 and 10 ng/ml) treatment (16). Under control conditions, cells stained by CM-H2DCFDA were sparse, and the staining was weak. Yet with TNF-␣ treatment, the number of the cells with positive staining was considerably higher and the cells were stained evenly throughout the cytoplasm. Pretreatment with VitE or MnTBAP drastically diminished the number and the intensity of staining (Fig. 2F). Similar results were obtained with isolated canine ventricular myocytes; TNF-␣ (0.1 ng/ml) markedly increased ROS level and co-application with ViTE (100 M) or MnTBAP (5 M) prevented the effects of TNF-␣ (Fig. 2G). DISCUSSION Heart failure is associated with APD and QT interval prolongation, believed to contribute to the occurrence of sudden cardiac death (6, 7). We show here that TNF-␣ suppresses I HERG in HEK293 cells and I Kr in dog cardiomyocytes and prolonged APD. Depression of I HERG /I Kr , as produced by TNF-␣ in this study, may contribute to delayed repolarization and associated malignant ventricular tachyarrhythmias with increased TNF-␣ level in patients with CHF.
Ionic remodeling in CHF has been studied (21). L-type Ca 2ϩ current density appears to be unaltered (20). The inward-rectifier K ϩ current is consistently reduced (5). The transient outward K ϩ current (I to ) is also reduced, potentially causing APD prolongation (5,22). However, inhibition of I to reduces APD in human atrial cells (23), canine atrial cells (12), and dog Purkinje fibers (24). The effect of I to on the AP depends largely on the magnitude of I K (25). Tsuji et al. (26) showed I Kr , measured as E-4031-sensitive tail current, to be ϳ36% smaller in rabbits with ventricular tachypacing-induced CHF than in healthy rabbits. Lodge and Normandin (27) demonstrated earlier that I Kr , measured as dofetilide-sensitive tail current, reduced by ϳ45% in the BIO TO-2 strain of cardiomyopathic hamster of 10 months old, derived from the BIO 53.58 animals and providing a model of dilated low output heart failure, compared with the 10-month-old control (BIO F1B) hamsters. A recent study by London et al. (9) showed significant APD prolongation in transgenic mice which overexpressed TNF-␣ and developed heart failure. Our study suggests that TNF-␣ may be an important mediator of CHF-induced I Kr reduction and is the first to demonstrate that TNF-␣ can modulate cardiac K ϩ channels.
We further demonstrated that pretreatment with VitE or MnTBAP prevented, whereas X/XO mimicked, TNF-␣-induced I HERG depression. The effects of VitE and MnTBAP are likely due to their antioxidant actions because TNF-␣ increased the intracellular ROS level in a concentration-dependent manner in both HEK293 cells and canine ventricular myocytes, more specifically O 2 . level because VitE or MnTBAP effectively prevented the increase. In line with our finding, a recent study published during the course of this study clearly demonstrated the ability of TNF-␣ to stimulate mitochondrial production of ROS in cardiomyocytes (20). It has also been shown that ROS is one of the key deleterious factors in failing heart (28,29). Our data therefore indicate that TNF-␣-induced HERG depression occurs at the functional level, but not at the expression levels (TNF-␣ did not alter HERG protein content), and the functional impairment of HERG channels by TNF-␣ is mediated by ROS, particularly O 2 . .
Circulating TNF-␣ levels predict mortality in CHF, and therapies directed against TNF-␣ may limit the pathophysiologic consequences (1). In healthy human subjects, the TNF-␣ level is below 0.01 ng/ml, but in patients with heart failure, it can increase to over 0.1 ng/ml (17)(18)(19). TNF-␣ significantly inhibited I HERG over this concentration range (e.g. by ϳ35% at plateau voltages from Ϫ10 to ϩ10 mV in cells exposed to 0.1 ng/ml TNF-␣ for 10 h). Our study might have underestimated the effects of TNF-␣ on APD because the myocytes were incubated with TNF-␣ at 4°C to maintain good quality of the cells. Our observations provide new insights into the potential molecular mechanisms underlying electrophysiological abnormalities and sudden arrhythmic death in patients with CHF.