Calcium-dependent arrhythmias in transgenic mice with heart failure

¹ Cardiovascular Institute and ² Department of Cell Biology and Physiology, University of Pittsburgh, Pittsburgh 15213; and ³ Department of Medicine, Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania 19107

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Transgenic mice overexpressing the inflammatory cytokine tumor necrosis factor (TNF)- (TNF- mice) in the heart develop a progressive heart failure syndrome characterized by biventricular dilatation, decreased ejection fraction, atrial and ventricular arrhythmias on ambulatory telemetry monitoring, and decreased survival compared with nontransgenic littermates. Programmed stimulation in vitro with single extra beats elicits reentrant ventricular arrhythmias in TNF- (n = 12 of 13 hearts) but not in control hearts. We performed optical mapping of voltage and Ca²⁺ in isolated perfused ventricles of TNF- mice to study the mechanisms that lead to the initiation and maintenance of the arrhythmias. When compared with controls, hearts from TNF- mice have prolonged of action potential durations (action potential duration at 90% repolarization: 23 ± 2 ms, n = 7, vs. 18 ± 1 ms, n = 5; P < 0.05), no increased dispersion of refractoriness between apex and base, elevated diastolic and depressed systolic [Ca²⁺], and prolonged Ca²⁺ transients (72 ± 6 ms, n = 10, vs. 54 ± 5 ms, n = 8; P < 0.01). Premature beats have diminished action potential amplitudes and conduct in a slow, heterogeneous manner. Lowering extracellular [Ca²⁺] normalizes conduction and prevents inducible arrhythmias. Thus both action potential prolongation and abnormal Ca²⁺ handling may contribute to the initiation of reentrant arrhythmias in this heart failure model by mechanisms distinct from enhanced dispersion of refractoriness or triggered activity.

optical mapping; genetically engineered mice

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

ARRHYTHMIAS remain a major health problem in cardiomyopathies of both ischemic and nonischemic origin. As many as 50% of patients with congestive heart failure (CHF) die suddenly, and this accounts for at least 250,000 annual deaths in the United States (42). Pharmacological treatments of arrhythmias can do more harm than good, and device therapies are limited by high cost and the limitations that they bring to quality of life (5). Prolonged action potential duration (APD) and downregulation of the repolarizing transient outward K⁺ current (I_to) and inward rectifier K⁺ currents (I_K1) are present in tissue and cardiac myocytes isolated from patients and animal models with CHF (3, 16, 25, 39). This delayed repolarization, along with enhanced dispersion of repolarization, may contribute to arrhythmias and sudden death (10, 12, 40). Altered intracellular Ca²⁺ handling in the heart, including decreased peak systolic Ca²⁺, elevated diastolic Ca²⁺, and prolongation of the Ca²⁺ transient are also present in CHF (4, 15, 27, 29). These alterations in Ca²⁺ handling can lead to triggered activity and arrhythmias (11). The relative contributions of abnormalities in depolarization and Ca²⁺ handling to the genesis of arrhythmias has yet to be determined.

Inflammatory cytokines, including TNF-, are increased in the serum and hearts of patients with CHF and may contribute to the pathophysiology of the disease (22, 37, 38). We recently engineered mice that overexpress TNF- in the heart under the control of the -myosin heavy chain promoter (TNF- mice) and develop a cardiomyopathy characterized by biventricular dilatation, decreased left ventricular ejection fraction, and decreased survival compared with nontransgenic littermates (19). Most of the mice exhibit symptoms of CHF before death (tachypnea, cyanosis, and ascites) and show evidence of decompensated heart failure at autopsy (pleural effusions, hepatic congestion, and severe atrial and ventricular dilatation). We now show that these mice develop arrhythmias on ambulatory telemetry monitoring and use optical mapping of voltage and Ca²⁺ in hearts isolated from the mice to demonstrate that abnormalities in both voltage and Ca²⁺ may contribute to the initiation of reentrant arrhythmias by mechanisms distinct from enhanced dispersion of refractoriness or triggered activity.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Breeding of TNF- transgenic mice. All studies were approved by the Institutional Animal Care and Use Committee at the University of Pittsburgh.

Heterozygous TNF- transgenic mice were bred with FVB controls to generate TNF- mice and wild-type littermate controls. Mice aged 3 to 9 mo were used for the current studies. For all studies, TNF- and control mice were age, sex, and strain matched.

Ambulatory recording of arrhythmias in mice. Radiotelemetry electrocardiogram monitors (Data Sciences) were implanted subcutaneously on the backs of TNF- and control mice with the use of tribromoethanol (Avertin) anesthesia as previously described (23). The mice were allowed to recover for at least 6 days, at which point 24 h of telemetry was recorded and saved on disk at 400 Hz. Monitoring was then continued for up to 4 mo.

All telemetry data were scanned by hand for atrial and ventricular arrhythmias. At least 95% of the data for each mouse was adequate for analysis. A computer-based arrhythmia detection system was simultaneously developed and tested using this telemetry.

Optical mapping of action potentials. Mice were anesthetized with fluothane, heparinized (50 units ip), and then euthanized. The heart was then rapidly excised, cannulated, and placed in a chamber specifically designed to immobilize, pace, and focus an image of the left ventricular free wall on a photodiode array as previously described (2). The perfusate contained (in mM) 141 NaCl, 25 NaHCO₃, 5 HEPES, 1.2 NaH₂PO₄, 1.0 MgSO₄, 5.0 KCl, 50 dextrose, and 1.8 CaCl₂ (pH 7.4), bubbled with 95% O₂-5% CO₂. Perfusion pressure was maintained at 60-80 mmHg, and the temperature of the perfusate was maintained at 37°C by feedback control as previously described (2). Hearts were stained with the voltage-sensitive dye 1-(3-sulfonatopropyl)-8-[-[2-(di-n-butylamino)-6-naphthyl]vinyl]pyridium betaine (di-4-ANEPPS, 10-15 µl of a 3 mM stock solution in DMSO) delivered as a bolus through the port of a bubble trap, which resulted in homogeneous dye loading throughout the heart and action potentials that were stable for up to 4 h (2). Atrial and ventricular contraction were initiated by programmed stimulation from one of the Teflon-coated silver wires brought into contact with the heart on the epicardial surface of the atrium or the ventricle at the apex, midventricle, and base. The electrodes were also used to record electrograms.

Light from a tungsten halogen lamp was collimated, passed through an interference filter (520 ± 20 nm), and focused on the surface of the stained heart. The fluorescence from the dye was passed through a cutoff filter (>630 nm) and focused on the 124-element photodiode array. The array viewed an area of ~4 × 4 mm on the free wall of the left ventricle, meaning that each element recorded electrical activity from a region of ~310 × 310 µm with a depth of field of ~40 µm. The photocurrent of each photodiode was passed through a current to a voltage converter and a second stage amplifier, digitized with temporal resolution >1 kHz, and stored in computer memory.

All analysis was automated using custom software written in Interactive Data Language. Activation times were calculated from maximal rate of force development (dF/dt_max) of each action potential upstroke, and activation patterns were displayed as isochronal maps as previously described (2, 34). An activation time was accepted if dF/dt_max was more than the SD of the background noise. Repolarization times taken at 75% or 90% recovery to baseline were used to determine action potential durations (APD) at 75% and 90% repolarization (APD₇₅ and APD₉₀, respectively). Action potentials with signal-to-noise ratios of <10 or excessive movement artifacts were eliminated, and the remaining activation times were triangulated with the use of Delaunay's triangulation algorithm to overcome spatially irregular data sets. Isochronal lines were drawn from these triangles with linear interpolation and connecting points by lines.

Local conduction velocities were calculated from dF/dt_max at 124 sites as previously described (2, 34). For each diode, a gradient vector was calculated to each of the eight adjacent diodes. The distance between adjacent diodes was used to determine local velocity vectors. Average velocity was calculated from the vectoral average of 124 local velocity vectors oriented between 0 and  radians.

Optical mapping of intracellular [Ca²+]. For mapping of Ca²⁺ transients, hearts were stained with the Ca²⁺-sensitive dye Rhod 2-AM (25 µg dissolved in 25 µl of DMSO added to the coronary perfusate). Rhod 2-AM is membrane permeable and becomes Ca²⁺ sensitive and trapped in the cytosol when esterified to Rhod 2 intracellularly. With excitation light (_ex = 520 ± 20 nm), Rhod 2 exhibits a more than 100-fold increase in fluorescence at its emission wavelength (_em = 585 nm) on binding Ca²⁺ and is typically used as a single excitation and single emission wavelength dye.

We have shown that the total dye concentration in myocytes can be determined from the absorption of the dye and dye-Ca²⁺ concentration ([dye-Ca²⁺]) through the fluorescence emission (8). Thus a single emission dye can be accurately calibrated by simultaneous measurements of the dye's absorption (total dye content) and fluorescence [dye-Ca²⁺]. Here, [Ca²⁺] was calibrated by measuring the Rhod 2 fluorescence when all the dye is bound to Ca²⁺ (F_max) and the fluorescence when all the dye is free (F_min) as previously described (7). In heart muscle homogenates, disocciation constant equaled 720 nM for Rhod 2/Ca²⁺. Calibration was performed in a dedicated subset of hearts because of concerns that F_max would decrease during prolonged experiments.

Data analysis. Data are presented as means ± SD. The number of experiments (n) indicates the number of hearts (or mice) used. APD and calcium transient parameters for each heart were measured as the average of several contiguous diodes. Arrhythmia and optical mapping data for TNF- mice and controls are compared by Student's t-test for paired and unpaired data as appropriate. Continuous variables were compared by ANOVA with Scheffé's multiple-range test. Isochronal activation maps of basic and premature beats are constructed with contour lines drawn at 1-ms apart. The activation time point is determined by the peak time derivative of the action potential upstroke. Gradients of APD are compared with gradients of conduction, and the locations of lines of block are determined as a function of the coupling intervals during programmed premature stimulation. Data before and after an intervention and comparison of rate-dependent changes are compared using analysis of variance using repeated measures.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Arrhythmias in TNF- transgenic mice. Ambulatory telemetry monitoring of the TNF- mice showed atrial and ventricular arrhythmias not found in FVB controls (Table 1). Examples of premature atrial beats, runs of atrial fibrillation or flutter, premature ventricular beats (PVBs), and runs of ventricular tachycardia (VT) are shown in Fig. 1. Couplets and runs of ventricular tachycardia were more common in male TNF- mice (4 of 7) than in female TNF- mice (1 of 8), consistent with the prior findings of more severe CHF symptoms and decreased survival in males (17).

View this table: [in this window] [in a new window]   Table 1.   Frequency of arrhythmias in TNF- transgenic mice

View larger version (33K): [in this window] [in a new window]   Fig. 1.   Arrhythmias in tumor necrosis factor (TNF)- transgenic mice. A: telemetry traces from a wild-type FVB and a TNF- mouse. Note clear P wave, QRS complex, and ST segment in these mice in sinus rhythms. B: atrial arrhythmias, including a run of a supraventricular tachycardia at a rate of 850 beats/min (a), and an irregularly irregular rhythm that appears to be atrial fibrillation (b). C: ventricular arrhythmias, including a ventricular couplet (a) and an 8-beat run of nonsustained ventricular tachycardia (b). D: death of a TNF- mouse. Baseline heart rhythm with a rate of 540 beats/min several weeks before death is shown (a). With symptomatic heart failure, rhythm became junctional at a rate of 280 beats/min, slow for the mouse (b). Near the time of death, the rhythm became agonal and wider complex, and ventricular ectopy was present (c).

In contrast to human heart failure, mean heart rate of age- and sex-matched mice was lower in TNF- mice than in controls (620 ± 64 beats/min, n = 15, vs. 679 ± 32 beats/min, n = 10; P = 0.01). This finding was less pronounced if mice not in sinus rhythm were excluded (632 ± 59, n = 12; P = 0.04). Serial 24-h telemetry recordings on three TNF- mice showed a progressive decrease in heart rate with increasing age. Prolonged episodes of bradycardia with junctional rhythms preceded and accompanied death (Fig. 1D). No episodes of tachyarrhythmias resulted in death during continuous monitoring of mice for up to 4 mo.

Optical mapping of ventricular action potentials in TNF- transgenic hearts. We used the voltage-sensitive dye di-4ANEPPS to record action potentials from a 4-mm × 4-mm area on the epicardial surface of the left ventricle of Langendorff-perfused isolated mouse hearts paced near the apex at a cycle length (CL) of 200 ms (Fig. 2A). The output of the 124-element array, representative optical action potentials from several channels, and isochronal activation maps are shown for hearts isolated from a TNF- mouse and a littermate control. APD₇₅ and APD₉₀ were prolonged by 14% and 28% in TNF- transgenic hearts (17 ± 1 and 23 ± 2 ms, n = 7) compared with wild-type hearts (15 ± 1 and 18 ± 1 ms, n = 5; P < 0.05 for each; Fig. 2B). Mean conduction velocity was similar for wild-type and transgenic hearts (0.51 ± 0.03 vs. 0.50 ± 0.05 m/s; n = 3 each) and decreased with faster stimulation frequencies (CL = 100 ms; 0.40 ± 0.02 vs. 0.40 ± 0.03 m/s; n = 3 each; Fig. 2C).

View larger version (65K): [in this window] [in a new window]   Fig. 2.   Optical mapping of action potentials from the free wall of the left ventricle. A,top: optical action potentials from the 124-element photodiode array (inset) for a control (a) and a TNF- transgenic mouse heart (b). Each action potential detects the sum of signals from cells in an area of ~310 × 310 µm × ~40 µm in depth. Action potentials represent a decrease in fluorescence (F/F) of ~10%. Note the motion artifact at the lower right edge of the diode for the TNF- heart. Middle, output from one channel for each heart is shown at a faster sweep speed. Bottom, maps showing isochronal lines of activation (1 ms apart), which spreads from the pacing site at the apex toward the base of the heart. Lighter colors indicate earlier activation, and a grayscale is included. B: action potential duration (APD) measured as the time from activation to 75% (APD75) and 90% repolarization (APD90). For each heart, data are the mean of 6 channels near the apex. Error bars on all figures show SD. *P < 0.05. C: conduction velocity in control and TNF- mouse hearts at cycle lengths (CL) of 100 and 200 ms. Conduction velocity is determined as described previously and in the text (2, 34).

Premature stimuli were applied to the ventricle following a train of 10 stimuli at a CL of 200 ms (Fig. 3, A and B). The refractory periods measured at the base (53 ± 6 ms) and at the apex (45 ± 7 ms) of hearts from the TNF- mice (n = 7) were nearly identical to those measured at the base (52 ± 9 ms) and at the apex (44 ± 10 ms) of hearts from wild-type mice (n = 7). In contrast, restitution kinetics for action potential amplitude (APA) showed significant differences between TNF- and wild-type hearts (Fig. 3C). The APAs of premature S2 beats normalized to the preceding S1 beat and plotted as a function of the S1-S2 interval were significantly smaller for S1-S2 intervals <100 ms in TNF- hearts compared with controls (n = 5 mice each; P < 0.001). Similarly, the conduction velocities of the premature beats were markedly slower at S1-S2 intervals <80 ms in TNF- hearts (P < 0.0001; Fig. 3, B, D, and E) and areas with very slow conduction were apparent.

View larger version (33K): [in this window] [in a new window]   Fig. 3.   A: optical action potentials of two control beats (S1, CL = 200 ms) and one premature beat (S2, CL = 50 ms) from the left ventricular free wall of a control and a TNF- heart. Note that the amplitude of the action potential from the TNF- heart is smaller than in the control. B: activation maps of the premature beats (S2) from the hearts shown in A. Numbers represent the time (in ms) from capture of the S2 pulse. Note that the spread of activation from apex to base is slower in the TNF- transgenic mouse than in the control mouse. Also note that the areas of marked conduction slowing and conduction failure are more prominent in the TNF- heart (e.g., near the apex). C: restitution kinetics plotted as action potential amplitude (APA) normalized to its stable value at CL = 200 ms as a function of S1-S2 interval for 5 control and 5 TNF- mice paced at the apex. Each data point represents the average of data from 4 contiguous diodes. For each heart, the S1-S2 interval was decreased by 10 ms from 100 to 70 ms and then by 5 ms to the refractory period. Actual S1-S2 interval was measured at each site. A best-fit line is shown for each data set. Statistical comparison was performed using all points with the Mann-Whitney U-test. D: conduction velocity of the S2 premature beat as a function of S1-S2 interval for a representative control and TNF- heart paced at an S1-S1 interval of 200 ms. Protocol is identical to that described in C. Each data point is an average of signals from 7 hearts. Note that conduction velocity falls in TNF- hearts to a greater extent than in control hearts before conduction failure at the refractory period. Error bars indicate SD. *P < 0.001; #P < 0.05. E: conduction velocity during the earliest premature beat (S2) that propagated normalized to the velocity of the last paced (S1) beat (baseline CL = 200 ms). Conduction velocity of the S2 beat is lower in TNF- hearts than in controls (P < 0.0001). In a subset of TNF- hearts, the extracellular [Ca2+] was decreased from 1.8 to 1.0 mM, and the conduction velocity increased significantly toward baseline (n = 3, P = 0.02).

Programmed stimulation of ventricular arrhythmias in TNF- hearts. Single extra stimuli elicited VT when applied at the apex of 12 of 13 and at the base of 2 of 4 transgenic mouse hearts compared with only 1 of 7 wild-type FVB hearts (Fig. 4A). The first tachycardia in any given heart usually lasted 12-15 beats and self- terminated, whereas subsequent inductions typically lasted for ~15 min. Some of the tachycardias were monomorphic, whereas others showed varying directions and velocities indicative of polymorphic arrhythmias. Activation maps showed that the slow, heterogeneous conduction of the premature impulse leads to functional lines of block and the initiation of a reentrant ventricular tachycardia (Fig. 4B). The tachycardias then propagated around the perimeter of the mouse heart as previously described (2).

View larger version (40K): [in this window] [in a new window]   Fig. 4.   Optical mapping of ventricular arrhythmias. A: induction of ventricular tachycardia (VT) using a single premature stimulus (S2) applied to the posteroapical wall of a TNF- transgenic heart during pacing at a CL of 200 ms. Note the rate of the tachycardia exceeds 1,000 beats/min. B: map showing isochronal lines of activation for the last paced S1 beat (left), the premature S2 beat that leads to the initiation of VT (middle), and the first two beats (VT1 and VT2) during the tachycardia (right). Note that slow, heterogeneous conduction of the premature beat (S2) leads to a functional line of conduction block. This leads to the initiation of a reentrant circuit (arrow) that propagates around the outer perimeter of the heart (beats VT1 and VT2). As in the previous figures, the heart is viewed from the left anterior oblique position, the apex of each heart is at the bottom, and isochronal lines of activation are 1-ms apart.

Optical mapping of Ca²+ transients in TNF- mice. Hearts from TNF- mice and FVB controls were stained with the Ca²⁺-sensitive dye Rhod 2-AM, and parameters of the calcium transient were measured (Figs. 5 and 6A). At a CL = 200 ms, peak systolic [Ca²⁺] was decreased in TNF- mice compared with controls (635 ± 21 vs. 746 ± 46 nM; n = 4 each; P < 0.05), diastolic [Ca²⁺] was increased (334 ± 37 vs. 257 ± 30 nM; n = 4 each; P < 0.01), and the Ca²⁺ transient was markedly prolonged (72 ± 6 ms, n = 10, vs. 54 ± 5 ms, n = 8; P < 0.01). The time from stimulation to the onset of the Ca²⁺ transient was not different in the TNF- hearts compared with wild-type hearts (16 ± 2 vs. 14 ± 1 ms; n = 5 each; P = not significant). Hearts from TNF- mice developed Ca²⁺ alternans at longer CLs than control hearts (Fig. 6B), and periods of rapid pacing (CL = 80 ms) raised diastolic Ca²⁺ and elicited runs of nonsustained polymorphic VT in transgenic but not control hearts (n = 3 each; Fig. 6C).

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Calcium transients in TNF- and control mice. A: systolic [Ca2+] is lower and diastolic [Ca2+] is higher in hearts from TNF- compared with control mice. *P < 0.05. **P < 0.01. B: calcium transient duration measured from the upstroke to 75% return to baseline. **P < 0.01.

View larger version (46K): [in this window] [in a new window]   Fig. 6.   Optical maps of calcium transients. A: representative Ca2+ signals from a wild-type heart (a) and a TNF- transgenic heart (b) paced at a CL = 180 ms. Transients from the TNF- heart are markedly prolonged. B: Ca2+ alternans in a TNF- heart (a) but not in a control heart (b) paced at a CL = 140 ms. C: period of rapid pacing (CL = 80 ms) elicits a nonsustained chaotic ventricular rhythm in TNF- transgenic heart (b) but not in a control heart (a).

Lower [Ca²+] prevents arrhythmias in TNF- mice. Single premature ventricular stimuli in hearts loaded with Rhod 2 did not elicit VT (n = 0/6). We speculated that the difference between voltage- and Ca²⁺-sensitive dyes might be due to the Ca²⁺-buffering effect of Rhod 2.

To test the dependence of arrhythmias on elevated Ca²⁺, we repeated the voltage- and calcium-mapping experiments and lowered extracellular [Ca²⁺] from 1.8 to 1.0 mM. Lowering extracellular [Ca²⁺] prolonged APD₉₀ from 20 ± 2 to 26 ± 0.3 ms in wild-type hearts and from 24 ± 2 to 34 ± 2 ms in TNF- hearts (n = 3 each; Fig. 7, A and C). Similarly, APD₇₅ lengthened from 18 ± 2 to 21 ± 2 ms in wild-type hearts and from 20 ± 1 to 25 ± 0.3 ms in TNF- hearts. Lowering extracellular [Ca²⁺] decreased the magnitude of the intracellular Ca²⁺ transient to a similar extent in control and TNF- hearts (20 ± 3% vs. 24 ± 3%, n = 3 each; Fig. 7B). The duration of the Ca²⁺ transient decreased with lowering of the extracellular [Ca²⁺] to a somewhat greater extent in hearts from TNF- mice versus wild-type mice (14 vs. 9 ms; Fig. 7D).

View larger version (31K): [in this window] [in a new window]   Fig. 7.   Effect of lowering extracellular [Ca2+] on action potentials and Ca2+ transients. A: superimposed representative action potentials from a wild-type (a) and a TNF- transgenic heart at baseline (b) (1.8 mM extracellular Ca2+, solid line) and after a decrease of the extracellular Ca2+ (1.0 mM, dotted line). We cannot exclude a small change in resting membrane potential with the change in extracellular [Ca2+]. B: superimposed representative Ca2+ transients using Rhod 2 from a wild-type (a) and a TNF- transgenic heart (b) at baseline (1.8 mM extracellular Ca2+, solid line) and after a decrease of the extracellular Ca2+ to 1.0 mM (dotted line). Ca2+ concentrations were not calibrated. C: APD90 decreased in both wild-type and TNF- mice following the decrease in extracellular Ca2+ concentration (n = 3 each). D: duration of the Ca2+ transient (to 75% baseline) decreased significantly in hearts from TNF- mice (n = 3; P = 0.03). In hearts from wild-type mice, duration of the Ca2+ transient also decreased, but did not reach statistical significance (P = 0.19).

No arrhythmias could be induced in the control hearts at the lower extracellular [Ca²⁺] despite the APD prolongation. Inducible ventricular arrhythmias in the TNF- hearts were eliminated by the lower [Ca²⁺] in the extracellular solution (n = 3 each). Of note, the fall of conduction velocity seen with ventricular PVBs was also attenuated by the low Ca²⁺ solution (n = 3, P = 0.02; Fig. 3E). The lower extracellular [Ca²⁺] solution did not lead to a significant change in conduction velocity at the basal CL of 200 ms (data not shown).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Repolarization abnormalities, cellular coupling, and arrhythmias. Abnormalities in repolarization are present in patients with CHF and in a number of large animal models (3, 16, 25, 29). Decreases in the repolarizing K⁺ currents I_to and I_K1 have been documented. The prolongation of APD and QT interval may promote arrhythmias directly via early afterdepolarizations or indirectly through enhanced dispersion of repolarization (10, 12, 41). Abnormalities in conduction are also present in the hearts of patients with CHF and result from structural changes such as fibrosis and decreased connexin43 gap junction protein (30, 31).

Similar findings are present in genetically engineered mice. Mice lacking both fast and slow I_to (I_to,f and I_to,s, respectively) have marked prolongation of APD, early afterdepolarizations, and ventricular arrhythmias (13). Transgenic mice overexpressing a truncated K⁺ channel fragment have increased vulnerability to reentrant VT due to enhanced dispersion of repolarization and refractoriness (2, 24). Disruption of the K⁺ channel-interacting protein 2 leads to loss of I_to and susceptibility to ventricular arrhythmias (20). Mice homozygous for disruption of the transcription factor HF-1b have conduction defects and arrhythmias, and those homozygous for a cardiac-restricted connexin43 knockout also have enhanced arrhythmogenesis and sudden death from tachyarrhythmias (14, 26).

The TNF- mice described here have a modest prolongation of APD compared with mice with disrupted K⁺ channel expression, without any significant increase in refractory periods. Lowering extracellular [Ca²⁺] further prolonged APD, possibly due to a decrease in Ca²⁺-dependent inactivation of the L-type Ca²⁺ channel (16). Of note, arrhythmias were suppressed under these conditions. In addition, conduction velocity is similar for transgenic and control mice at CLs down to 100 ms. Thus changes in repolarization and refractoriness alone do not seem sufficient to explain the increase in reentrant arrhythmias that result from single premature beats.

Calcium abnormalities and arrhythmias. Abnormalities of Ca²⁺ handling may contribute to arrhythmias. Human heart failure is usually characterized by decreased expression of the sarco(endo)plasmic reticulum Ca²⁺-ATPase (SERCA), variable decreases in phospholamban, and increased expression of the Na/Ca exchanger (1). The resulting prolongation of the Ca²⁺ transient and intracellular Ca²⁺ overload could lead to premature sarcoplasmic reticulum Ca²⁺ release and delayed afterdepolarizations (11). Mutations in the cardiac sarcoplasmic reticulum Ca²⁺ release channel (hRyR2) have recently been described in patients with cathecholaminergic polymorphic ventricular tachycardia and in patients with arrhythmogenic right ventricular dysplasia (32, 36). These genetic syndromes further suggest a direct link between abnormalities in Ca²⁺ handling and arrhythmias.

TNF- mice have abnormal expression of the transcripts encoding Ca²⁺ regulatory proteins (18), possibly as a direct effect of cytokine overexpression (42). The decreased expression of SERCA and of the SERCA-to-phospholamban ratio would be predicted to decrease sarcoplasmic reticulum Ca²⁺ uptake. Here, we show that hearts from wild-type FVB mice have systolic and diastolic intracellular [Ca²⁺] similar to those reported previously in mice and in other species (7, 9). In contrast, hearts from TNF- mice have an elevated diastolic [Ca²⁺], a decreased systolic [Ca²⁺], and marked prolongation of the Ca²⁺ transient. In addition, 1) premature beats triggered during the tail of the prior Ca²⁺ transient propagate slowly and heterogeneously, leading to functional lines of conduction block and the initiation of reentrant arrhythmias; 2) rapid pacing of the TNF- mouse hearts leads to an elevated diastolic [Ca²⁺], Ca²⁺ alternans, and arrhythmias; and 3) interventions that lower intracellular [Ca²⁺], including buffering by Rhod 2 and lowering extracellular [Ca²⁺], normalize conduction and prevented the inducible arrhythmias.

Ca²⁺ sparks result from the quantal release on Ca²⁺ from the sarcoplasmic reticulum (6). Spontaneous, asynchronous Ca²⁺ release from the sarcoplasmic reticulum is more common in the setting of Ca²⁺ overload and can lead to the propagated diastolic Ca²⁺ oscillations and contractions seen in isolated myocytes and intact tissue (21). Synchronization of the sparks and Ca²⁺ release can lead to delayed afterdepolarizations and triggered arrhythmias, at least in part through activation of the electrogenic Na/Ca exchange current. Diastolic Ca²⁺ oscillations could contribute to the measured increase in global diastolic [Ca²⁺] that we see in the TNF- mice. The nonuniform elevations in intracellular [Ca²⁺] and membrane potential could also contribute to the initiation and maintenance of reentrant arrhythmias by causing slow, heterogeneous conduction of premature beats. If true, this demonstrates an additional mechanism by which Ca²⁺ overload in heart failure can promote reentrant arrhythmias, independent of increasing the frequency of premature beats via initiation of triggered activity (afterdepolarizations).

Previous studies on transgenic mice with elevated intracellular [Ca²⁺] have shown diminished responsiveness to -adrenergic stimulation (35). Lowering the extracellular [Ca²⁺] reversed the defect. It is tempting to speculate that similar mechanisms may lead to both the abnormalities in -receptor responsiveness and arrhythmias.

How do elevations in intracellular [Ca²⁺] lead to changes in the conduction velocity of premature beats but not in the refractory period? Elevated intracellular [Ca²⁺] following a premature beat or local spontaneous sarcoplasmic reticulum Ca²⁺ release could depolarize the membrane potential through Ca²⁺-dependent leak currents, Ca²⁺-dependent Cl currents, or via the electrogenic Na/Ca exchange. A small change in membrane potential could affect the time- and voltage-dependent recovery from inactivation of fast Na⁺ channels, leading to slow propagation of premature impulses but not conduction failure. In support of this hypothesis, APA restitution was markedly abnormal during premature beats in TNF- mice (Fig. 3C). Thus interactions between voltage- and Ca²⁺-dependent abnormalities may contribute to arrhythmias in this mouse heart failure model.

It is also possible that dynamic alterations in cell-to-cell coupling, such as reduced gap junction conduction in the presence of high intracellular [Ca²⁺], could explain the initiation of reentrant VT in these mice (28, 40). These seem less likely to occur within the time course of the action potential, however. Alternatively, changes in Ca²⁺-dependent intracellular processes via cAMP- or calmodulin-dependent kinases could affect arrhythmogenesis (35).

In this isolated perfused mouse heart model, we found that staining with Rhod 2 suppressed arrhythmias. Lowering extracellular [Ca²⁺] also suppressed arrhythmias, suggesting Ca²⁺ buffering as the mechanism. In a guinea pig model with long QT, staining with equivalent concentrations of Rhod 2 did not suppress ventricular arrhythmias (7). Thus the potential effects of cytoplasmic buffering by Ca²⁺-sensitive dyes on cardiac electrophysiology are model dependent and must be considered. We also cannot exclude other differences in the handling of Rhod 2 in the TNF- mice compared with controls.

The mouse as a model for arrhythmias. Many structural and electrophysiological differences exist between the mouse and human heart, and findings in the mouse must be interpreted with caution. The slower basal heart rate of the TNF- mice compared with controls is one example. However, the TNF- mice do develop many of the structural and biochemical findings present in human heart failure. Here, we show that these mice also develop atrial and ventricular arrhythmias and have used optical mapping with voltage- and Ca²⁺-sensitive dyes to dissect the mechanisms of the arrhythmias. Our findings implicate changes in intracellular Ca²⁺ handling as critical to the genesis of arrhythmias. These findings complement those of other transgenic mice with alterations in electrophysiological pathways that lead to arrhythmias (2, 13, 14, 20, 24, 26, 33). Further exploration of the pathways that couple changes in intracellular [Ca²⁺] to arrhythmias is warranted.

	ACKNOWLEDGEMENTS

We thank William Hughes and Scott McPherson for technical assistance.

	FOOTNOTES

This study was supported, in part, by National Heart, Lung, and Blood Institute Grants HL-58030 and HL-66096 (to B. London), HL-60032 (to A. M. Feldman), and HL-59614 (to G. Salama) and by predoctoral fellowships from the American Heart Association, Western Pennsylvania Affiliate (to L. C. Baker and B.-R. Choi).

Address for reprint requests and other correspondence: B. London, Cardiovascular Institute, Univ. of Pittsburgh Medical Center, BST 1704, 200 Lothrop St., Pittsburgh, PA 15213 (E-mail: londonb{at}msx.upmc.edu).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

First published October 17, 2002;10.1152/ajpheart.00431.2002

Received 21 May 2002; accepted in final form 8 October 2002.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

1.   Arai, M, Alpert NR, MacLennan DH, Barton P, and Periasamy M. Alterations in sarcoplasmic reticulum gene expression in human heart failure: a possible mechanism for alterations in systolic and diastolic properties of the failing myocardium. Circ Res 72: 463-469, 1993[Abstract/Free Full Text].

2.   Baker, LC, London B, Choi BR, Koren G, and Salama G. Enhanced dispersion of repolarization and refractoriness in transgenic mouse hearts promotes reentrant ventricular tachycardia. Circ Res 86: 396-407, 2000[Abstract/Free Full Text].

3.   Beuckelmann, DJ, Nabauer M, and Erdmann E. Alterations of K⁺ currents in isolated human ventricular myocytes from patients with terminal heart failure. Circ Res 73: 379-385, 1993[Abstract/Free Full Text].

4.   Beuckelmann, DJ, Nabauer M, and Erdmann E. Intracellular calcium handling in isolated ventricular myocytes from patients with terminal heart failure. Circulation 85: 1046-1055, 1992[Abstract/Free Full Text].

5.   CAST, The Cardiac Arrhythmia Suppression Trial Investigators. Preliminary reporte of encainide and flecainide on mortality in a randomized trial of arrhythmia suppression after myocardial infarction. N Engl J Med 321: 406-412, 1989[Abstract].

6.   Cheng, H, Lederer MR, Lederer WJ, and Cannell MB. Calcium sparks and [Ca²⁺]_i waves in cardiac myocytes. Am J Physiol Cell Physiol 270: C148-C159, 1996[Abstract/Free Full Text].

7.   Choi, BR, and Salama G. Simultaneous maps of optical action potentials and calcium transients in guinea-pig hearts: mechanisms underlying concordant alternans. J Physiol 529: 171-188, 2000[Abstract/Free Full Text].

8.   Del Nido, P, Glynn P, Buenaventura P, Salama G, and Koretsky AP. Fluorescent measurements of calcium transients in perfused rabbit heart using Rhod-2. Am J Physiol Heart Circ Physiol 274: H728-H741, 1998[Abstract/Free Full Text].

9.   Du, C, MacGowan GA, Farkas DL, and Koretsky AP. Calibration of the calcium dissociation constant of Rhod₂ in the perfused mouse heart using manganese quenching. Cell Calcium 29: 217-224, 2001[Web of Science][Medline].

10.   El-Sherif, N, Chinushi M, Caref EB, and Restivo M. Electrophysiological mechanism of the characteristic electrocardiographic morphology of torsade de pointes tachyarrhythmias in the long-QT syndrome: detailed analysis of ventricular tridimensional activation patterns. Circulation 96: 4392-4399, 1997[Abstract/Free Full Text].

11.   Fozzard, HA. Afterdepolarizations and triggered activity. Basic Res Cardiol 87, Suppl 2: 105-113, 1992.

12.   Frazier, DW, Wolf PD, Wharton JM, Tang ASL, Smith WM, and Ideker RE. Stimulus-induced critical point: a mechanism for electrical initiation of re-entry in normal canine myocardium. J Clin Invest 83: 1039-1052, 1989[Web of Science][Medline].

13.   Guo, W, Li H, London B, and Nerbonne JM. Functional consequences of elimination of I_to,f and I_to,s: early afterdepolarizations, atrioventricular block and ventricular arrhythmias in mice lacking Kv1.4 and expressing a dominant-negative Kv4  subunit. Circ Res 87: 73-79, 2001[Abstract/Free Full Text].

14.   Gutstein, DE, Morley GE, Tamaddon H, Vaidya D, Schneider MD, Chen J, Chien KR, Stuhlmann H, and Fishman GI. Conduction slowing and sudden arrhythmic death in mice with cardiac-restricted inactivation of connexin43. Circ Res 88: 333-339, 2001[Abstract/Free Full Text].

15.   Gwathmey, JK, Copelas L, MacKinnon R, Schoen FJ, Feldman MD, Grossman W, and Morgan JP. Abnormal intracellular calcium handling in myocardium from patients with end-stage heart failure. Circ Res 61: 70-76, 1987[Abstract/Free Full Text].

16.   Kaab, S, Nuss HB, Chiamvimonvat N, O'Rourke B, Pak PH, Kass DH, Marban E, and Tomaselli GF. Ionic mechanism of action potential prolongation in ventricular myocytes from dogs with pacing-induced heart failure. Circ Res 78: 262-273, 1996[Abstract/Free Full Text].

17.   Kadokami, T, McTiernan CF, Kubota T, Frye CS, and Feldman AF. Sex-related survival differences in murine cardiomyopathy are associated with differences in TNF-receptor expression. J Clin Invest 106: 589-597, 2000[Web of Science][Medline].

18.   Kubota, T, Bounoutas GS, Miyagishima M, Kadokami T, Sanders VJ, Bruton C, Robbins PD, McTiernan CF, and Feldman AM. Soluble tumor necrosis factor receptor abrogates myocardial inflammation but not hypertrophy in cytokine-induced cardiomyopathy. Circulation 101: 2518-2525, 2000[Abstract/Free Full Text].

19.   Kubota, T, McTiernan CF, Frye CS, Slawson SE, Semster BH, Koretsky AP, Demetris AJ, and Feldman AM. Dilated cardiomyopathy in transgenic mice with cardiac-specific overexpression of tumor necrosis factor-. Circ Res 81: 627-635, 1997[Abstract/Free Full Text].

20.   Kuo, HC, Cheng CF, Clark RB, Lin JJC, Lin JLC, Hoshijima M, Nguyen-Tran VTB, Gu Y, Ikeda Y, Chu PH, Ross JJr Giles WR, and Chein KR. A defect in the Kv channel-interacting protein 2 (KChIP2) gene leads to a complete loss of I_to and confers susceptibility to ventricular tachycardia. Cell 107: 801-813, 2001[Web of Science][Medline].

21.   Lakatta, EG. Functional implications of spontaneous sarcoplasmic Ca²⁺ release in the heart. Cardiovasc Res 26: 193-214, 1992[Free Full Text].

22.   Levine, B, Kalman J, Mayer L, Fillit HM, and Packer M. Elevated circulating levels of tumor necrosis factor in severe chronic heart failure. N Engl J Med 223: 236-24, 1990.

23.   London, B, Guo W, Pan XH, Lee JS, Shusterman V, Rocco CJ, Logothetis DE, Nerbonne JM, and Hill JA. Targeted replacement of Kv1.5 in the mouse leads to loss of the 4-aminopyridine-sensitive component of I_K,slow and resistance to drug-induced QT prolongation. Circ Res 88: 940-946, 2001[Abstract/Free Full Text].

24.   London, B, Jeron A, Zhou J, Buckett P, Han X, Mitchell GF, and Koren G. Long QT and ventricular arrhythmias in transgenic mice expressing the N-terminus and first transmembrane segment of a voltage-gated potassium channel. Proc Natl Acad Sci USA 95: 2926-2931, 1998[Abstract/Free Full Text].

25.   Nabauer, M, Beuckelmann DJ, and Erdmann E. Characteristics of transient outward current in human ventricular myocytes from patients with terminal heart failure. Circ Res 73: 386-394, 1993[Abstract/Free Full Text].

26.   Nguyen-Tran, VT, Kubalak SW, Minamisawa S, Fiset C, Wollert KC, Brown AB, Ruiz-Lozano P, Barrere-Lemaire S, Kondo R, Norman LW, Gourdie RG, Rahme MM, Feld GK, Clark RB, Giles WR, and Chien KR. A novel genetic pathway for sudden cardiac death via defects in the transition between ventricular and conduction system cell linkages. Cell 102: 671-682, 2000[Web of Science][Medline].

27.   O'Rourke, B, Kass DA, Tomaselli GF, Kaab S, Tunin R, and Marban E. Mechanisms of altered excitation-contraction coupling in canine tachycardia-induced heart failure. I. Experimental Studies. Circ Res 84: 562-570, 1999[Abstract/Free Full Text].

28.   Owens, LM, Fralix TA, Murphy E, Cascio WE, Gettes LS, and and the Experimental Cardiology Group Correlation of ischemia-induced extracellular and intracellular ion changes to cell-to-cell electrical uncoupling in isolated blood-perfused rabbit hearts. Circulation 94: 10-13, 1996[Abstract/Free Full Text].

29.   Perreault, CL, Shannon RP, Komamura K, Vatner S, and Morgan JP. Abnormalities in intracellular calcium regulation and contractile function in myocardium from dogs with pacing-induced heart failure. J Clin Invest 89: 932-938, 1992[Web of Science][Medline].

30.   Peters, NS, and Wit AL. Myocardial architecture and ventricular arrhythmogenesis. Circulation 97: 1746-1754, 1998[Free Full Text].

31.   Peters, NS, Green CR, Poole-Wilson PA, and Severs NJ. Reduced content of connexin43 gap junctions in ventricular myocardium from hypertrophied and ischemic human hearts. Circulation 88: 864-875, 1993[Abstract/Free Full Text].

32.   Priori, SG, Napolitano C, Tiso N, Memmi M, Vignati G, Bloise R, Sorrentino V, and Danieli GA. Mutations in the cardiac ryanodine receptor gene (hRyR2) underlie catecholaminergic polymorphic ventricular tachycardia. Circulation 103: 196-200, 2001[Abstract/Free Full Text].

33.   Sah, VP, Minamisawa S, Tam SP, Wu TH, Dorn GW, II, Ross J, Jr, Chien KR, and Brown JH. Cardiac-specific overexpression of RhoA results in sinus and atrioventricular nodal dysfunction and contractile failure. J Clin Invest 103: 1627-1634, 1999[Web of Science][Medline].

34.   Salama, G, Kanai A, and Effimov IR. Subthreshold stimulation of purkinje fibers interrupts ventricular tachycardia in intact hearts. Experimental study with voltage-sensitive dyes and imaging techniques. Circ Res 74: 604-619, 1994[Abstract/Free Full Text].

35.   Serikov, VB, Petrashevskaya NN, Canning AM, and Schwartz A. Reduction of [Ca²⁺]_i restores uncoupled -adrenergic signaling in isolated perfused transgenic mouse hearts. Circ Res 88: 9-11, 2001[Abstract/Free Full Text].

36.   Tiso, N, Stephan DA, Nava A, Bagattin A, Devaney JM, Stanchi F, Larderet G, Brahmbhatt B, Brown K, Bauce B, Muriago M, Basso C, Thiene G, Danieli GA, and Rampazzo A. Identification of mutations in the cardiac ryanodine receptor gene in families with arrhythmogenic right ventricular cardiomyopathy type 2 (ARVD2). Hum Mol Genet 10: 189-194, 2001[Abstract/Free Full Text].

37.   Torre-Amione, G, Kapadia S, Benedict C, Oral H, Young JB, and Mann DL. Proinflammatory cytokine levels in patients with depressed left ventricular ejection fraction: a report from the studies of left ventricular dysfunction (SOLVD). J Am Coll Cardiol 27: 1201-1206, 1996[Abstract].

38.   Torre-Amione, G, Kapadia S, Lee J, Durand JB, Bies RD, Young JB, and Mann DL. Tumor necrosis factor- and tumor necrosis factor receptors in the failing human heart. Circulation 93: 704-771, 1996[Abstract/Free Full Text].

39.   Vermeulen, JT, McGuire MA, Opthof T, Coronel de BJM R, Klopping C, and Janse MJ. Triggered activity and automaticity in ventricular trabeculae of failing human and rabbit hearts. Cardiovasc Res 28: 1547-1554, 1994[Web of Science][Medline].

40.   White, RL, Doeller JE, Verselis VK, and Wittenberg BA. Gap junctional conductance between pairs of ventricular myocytes is modulated synergistically by H⁺ and Ca⁺⁺. J Gen Physiol 95: 1061-1075, 1990[Abstract/Free Full Text].

41.   Witt, AL, Alessie MA, Bonke FIM, Lammers W, Smeets J, and Fenolio JJ, Jr. Electrophysiologic mapping to determine the mechanism of experimental ventricular tachycardia induced by premature impulses. Am J Cardiol 49: 166-185, 1982[Web of Science][Medline].

42.   Yokoyama, T, Vaca L, Rossen RD, Durante W, Hazarika P, and Mann DL. Cellular basis for the negative inotropic effects of tumor necrosis factor- in the adult mammalian heart. J Clin Invest 92: 2303-2312, 1993[Web of Science][Medline].

43.   Zipes, D. Genesis of cardiac arrhythmias: Electrophysiological considerations. Management of cardiac arrhythmias: Pharmacological, electrical, and surgical techniques. Specific arrhythmias: diagnosis and treatment. In: Heart Disease: A Textbook of Cardiovascular Medicine, edited by Braumwald E.. Philadelphia, PA: Saunders, 1997, p. 548-704.