Experimental Physiology

Transcription

1 84 Exp Physiol 12.7 (217) pp Research Paper Research Paper Sex and regional differences in rabbit right ventricular L-type calcium current levels and mathematical modelling of arrhythmia vulnerability Zane M. Kalik 1,JoshuaL.Mike 2, Cassandra Slipski 1,MoriahWright 2, Jozsi Z. Jalics 2 and Mark D. Womble 1 1 Department of Biological Sciences, Youngstown State University, Youngstown, OH, USA 2 Department of Mathematics and Statistics, Youngstown State University, Youngstown, OH, USA Edited by: Ming Lei Experimental Physiology New Findings What is the central question of this study? Regional variations of ventricular L-type calcium current (I Ca-L ) amplitude may underlie the increased arrhythmia risk in adult females. Current amplitude variations have been described for the left ventricle but not for the right ventricle. What is the main finding and its importance? Adult female rabbit right ventricular base myocytes exhibit elevated I Ca-L compared with female apex or male myocytes. Oestrogen upregulated I Ca-L in cultured female myocytes. Mathematical simulations modelling long QT syndrome type 2 demonstrated that elevated I Ca-L prolonged action potentials and induced early after-depolarizations. Thus, ventricular arrhythmias in adult females may be associated with an oestrogen-induced upregulation of I Ca-L. Previous studies have shown that adult rabbit left ventricular myocytes exhibit sex and regional differences in L-type calcium current (I Ca-L ) levels that contribute to increased female susceptibility to arrhythmogenic early after-depolarizations (EADs). We used patch-clamp recordings from isolated adult male and female rabbit right ventricular myocytes to determine apex base differences in I Ca-L density and used mathematical modelling to examine the contribution of I Ca-L to EAD formation. Current density measured at mv in female base myocytes was 67% higher than in male base myocytes and 55% higher than in female apex myocytes. No differences were observed between male and female apex myocytes, between male apex and base myocytes, or in the voltage dependences of I Ca-L activation or inactivation. The role of oestrogen was investigated using cultured adult female right ventricular base myocytes. After 2 days, 17β-estradiol (1 nm) produced a 65% increase in I Ca-L density compared with untreated control myocytes, suggesting an oestrogen-induced upregulation of I Ca-L.Action potential simulations using a modified Luo Rudy cardiomyocyte model showed that increased I Ca-L density, at the level observed in female base myocytes, resulted in longer duration action potentials, and when combined with a 5% reduction of the rapidly inactivating delayed rectifier potassium current conductance to model long QT syndrome type 2, the action potential was accompanied by one or more EADs. Thus, we found higher levels of I Ca-L in adult female right DOI: /EP85977 C 217 The Authors. Experimental Physiology C 217 The Physiological Society

2 Exp Physiol 12.7 (217) pp Sex differences in right ventricular L-type Ca 2+ current levels 85 ventricle base myocytes and the upregulation of this current by oestrogen. Simulations of long QT syndrome type 2 showed that elevated I Ca-L contributed to genesis of EADs. (Received 25 July 216; accepted after revision 18 April 217; first published online 24 April 217) Corresponding author M. D. Womble: Department of Biological Sciences, Youngstown State University, One University Plaza, Youngstown, OH USA. mdwomble@ysu.edu Introduction Adult women have an increased action potential duration (APD) and a longer rate-corrected Q T interval than men (Makkar et al. 1993; Pham & Rosen, 22; Salama & Bett, 214). If the Q T interval is longer than ms in women, this condition is considered pathological and is known as long QT syndrome (LQTS), a disorder that increases susceptibility to early after-depolarization (EAD) formation, the polymorphic ventricular tachycardia known as torsade de pointes (TdP), and sudden cardiac death (Makkar et al. 1993; Abi-Gerges et al. 24; Jameset al. 27; Webster & Berul, 28). In women with LQTS and TdP, action potential (AP) recordings from the right ventricle found prolonged APDs, the longest of which demonstrated EADs during the repolarizing phase, resulting in focal re-excitations (Bonatti et al. 1983). Women are also more prone to ventricular tachycardias initiated by premature extrasystoles originating in the right ventricular outflow tract (Buxton et al. 1983; Noda et al. 25; Brugada & Diez, 21). In long QT type 2 (LQT2), reduction of the repolarizing rapid delayed rectifier potassium current (I Kr ) limits myocyte repolarization capacity and thus prolongs APD and promotes EAD formation (Pham & Rosen, 22; Abi-Gerges et al. 24; James et al. 27; Webster & Berul, 28; Salama & Bett, 214; Roden, 216). The congenital form of LQT2 results from a loss-of-function mutation within the gene that codes for the I Kr ion channel (Webster & Berul, 28; Ackerman & Mohler, 21), whereas drug-induced (acquired) LQT2 can be caused by a wide array of cardiac and non-cardiac drugs that inhibit I Kr (Makkar et al. 1993; Pham et al. 21; Verkerk et al. 25; Roden, 216). Women have a greater susceptibility to developing LQT2 owing to their smaller repolarizing reserve compared with men (Makkar et al. 1993; Locati et al. 1998; Abi-Gerges et al. 24; James et al. 27; Gaborit et al. 21; Yang & Clancy, 21; Maruyama et al. 211; Salama & Bett, 214; Roden, 216). This difference has been attributed to the downregulation of repolarizing potassium channels by female sex hormones (Drici et al. 1996; Drici & Clement, 21; Gaborit et al. 21; Yang & Clancy, 21) or to the testosterone-induced upregulation of I Kr during puberty in males (Pham et al. 22). Using left ventricular myocytes and the potassium channel inhibitor drug E431 to model acquired LQT2, Sims et al. (28) showed that adult female rabbit myocytes exhibit an increased susceptibility to EADs and arrhythmia formation compared with male myocytes. However, the reverse response is seen in prepubertal (PP) males and females, where males are more susceptible to arrhythmia generation after partial I Kr inhibition (Liu et al. 25; Sims et al. 28). Given that before puberty no sex differences in potassium channel expression are present (Liu et al. 25), this observation suggests that factors other than I Kr levels might influence arrhythmia sensitivity in adult females. Oestrogen also promotes EAD production (Drici et al. 1996; Hara et al. 1998; Pham et al. 21) by upregulating protein expressions of the L-type calcium channel (Cav1.2α) andna + Ca 2+ exchanger (NCX1) in adult female myocytes (Chen et al. 211; Yang et al. 212). Previous studies of left ventricular myocytes have demonstrated larger L-type calcium currents (I Ca-L )in adult females compared with adult males (Pham & Rosen, 22; Verkerk et al. 25; Sims et al. 28; Yang et al. 212). In adult rabbits, regional differences were also found among left ventricular myocytes, with female base myocytes exhibiting elevated I Ca-L density and Cav1.2α expression compared with female apex or male base or apex myocytes (Sims et al. 28; Yang et al. 212). The greater I Ca-L levels in adult female base myocytes were associated with prolonged APDs and the appearance of arrhythmogenic EADs in a drug-induced model of LQT2 (Sims et al. 28). In contrast, opposite findings were observed for PP rabbits, where higher I Ca-L levels are found in PP male base myocytes, thus making these myocytes more likely than female PP cells to exhibit EADs and arrhythmias during drug-induced LQT2 (Sims et al. 28). Formation of EADs and the initiation of TdP results from reactivation of I Ca-L (January & Riddle, 1989; Volders et al. 2; Antoons et al. 27). During a prolonged AP, enhanced calcium influx via L-type calcium channels results in calcium overload within the sarcoplasmic reticulum (SR). This subsequently induces spontaneous SR calcium release, followed by activation of the depolarizing Na + Ca 2+ exchanger current (I NCX ), leading to I Ca-L reactivation and EAD formation (Choi & Salama, 2; Volders et al. 2; Choi et al. 22; Weiss et al. 21). Thus, EADs are preferentially initiated in base myocytes of the adult female left ventricle because of their higher levels of both I Ca-L and I NCX (Sims et al. 28; Chen et al. 211; Kim et al. 213). C 217 The Authors. Experimental Physiology C 217 The Physiological Society

3 86 Z. M. Kalik and others Exp Physiol 12.7 (217) pp Previous studies using adult rabbits have focused largely on arrhythmia initiation in the left ventricle. In the right ventricle, female myocytes have a longer APD than male myocytes (Pham et al. 21), and EAD initiation can occur in the base region of the female right ventricle (Choi et al. 22; Sims et al. 28). However, regional (base apex) variations in I Ca-L amplitude and the relationship of I Ca-L to EAD initiation have not previously been investigated in female right ventricular myocytes. This report investigates the sex and regional differences in I Ca-L density in adult rabbit right ventricular myocytes and its regulation by oestrogen. New Zealand White rabbits were used for this study because myocytes from these animals share similar electrophysiological properties and sex differences in arrhythmia phenotype with human hearts (Simset al. 28; Salama & Bett, 214). Numerical simulations with a modified Luo Rudy model of the cardiac AP were also used to model EAD vulnerability. Methods Ethical approval All animal use and housing protocols were approved by the Northeastern Ohio Medical University (NEOMED) Institutional Animal Care and Use Committee (approval no. 8-21) and were in accordance with the current US National Institutes of Health GuidefortheCareand Use of Laboratory Animals (8th edition, 211). Animals were housed at the NEOMED Animal Care Facility and provided with food and water ad libitum. Isolation of ventricular myocytes Adult New Zealand White rabbits (Myrtle s Rabbitry, Thompsons Station, TN, USA) were used for this study. Animals (obtained at 12 weeks of age) were allowed to acclimate for 1 week before killing. This age (91 days) is well after the sharp increases in sex steroids, peaking around day 6, that mark puberty (Beyer & McDonald, 1973; de Turckheim et al. 1983; Chen et al. 211). Cardiac myocyte isolation and electrophysiological measurements were performed as previously reported (Sims et al. 28; Chen et al. 211). Briefly, animals were anaesthetized and killed with an I.V. injection via the marginal ear vein of sodium pentobarbital (5 mg kg 1 ), and co-injected with heparin (2 U kg 1 I.V.). After ensuring there was no corneal reflex response, a thoracotomy was performed and the heart removed. The heart was then perfused via the aorta for 5 min in a Langendorff apparatus with an oxygenated physiological salt solution (PSS) containing (mm): 14 NaCl, 5.4 CsCl, 2.5 CaCl 2,.5MgCl 2,11glucose and 5.5 Hepes (ph 7.4). This was followed by a 1 min perfusion with Ca 2+ -free PSS, perfusion with collagenase type 2 (Worthington Biochemical, Lakewood, NJ, USA; at.6 mg ml 1 in Ca 2+ -free PSS) for 15 min, and Ca 2+ -free PSS perfusion for 5 min. After perfusion, the heart was placed in a highpotassium buffer containing (mm): 11 potassium glutamate, 1 KH 2 PO 4, 25 KCl, 2 MgSO 4,2taurine,5 creatinine,.5 EGTA, 2 glucose and 5 Hepes (ph 7.4). The right ventricle was isolated and separated into apex and base regions, which were minced using surgical scissors and filtered using 2 μm nylon mesh. Isolated myocytes were stored in a high-potassium buffer until needed. The heart, myocytes and all solutions were maintained at 37 C throughout the perfusion and myocyte isolation process. The ph for all physiological solutions was adjusted to their final ph of 7.4 with the addition of NaOH or HCl as needed while the solution was stirring and monitored using a ph meter. Electrophysiology The whole-cell configuration of the patch-clamp technique was used to measure I Ca-L from individual myocytes. Patch pipettes had resistances of 1 3 M when filled with (mm): 13 CsCl, 2 tetraethylammonium chloride (TEA-Cl), 5 MgATP, 5 EGTA,.1 Tris-GTP and 5. Hepes (ph 7.4). Macroscopic currents were recorded using an Axopatch 2B voltage-clamp amplifier, filtered at 5 khz, and sampled at 1 khz using a Digidata 144A interface and pclamp (version 1.2) software (Molecular Devices, Sunnyvale, CA, USA). Cells were bathed in K + -free extracellular solution (35 C) containing (mm): 14 NaCl, 5.4 CsCl, 2.5 CaCl 2,.5 MgCl 2,11glucoseand 5.5 Hepes (ph 7.4). The magnitude of I Ca-L was measured as the peak inward current recorded during a 1 ms voltage-clamp step to mv, applied following a 5 ms prepulse to 3 mv from a holding potential of 8 mv every 6s.TheI Ca-L was isolated by blocking K + channels with Cs + and TEA-Cl, inactivating Na + channels with the 3 mv prepulse step, and eliminating the driving force for Cl currents by measuring I Ca-L close to the predicted Cl equilibrium potential ( mv). Current voltage (I V) relationships were determined using a similar protocol, with a series of 1 ms voltage steps (from 3 to +6 mv) to obtain the peak calcium current at each voltage. All recordings were made 3 5 min after gaining whole-cell access and after I Ca-L had stabilized. Series resistance was partly compensated to achieve values of 3. M so as to prevent large voltage errors when measuring larger (1.5 na) whole-cell I Ca-L. Capacitance measurements were obtained from membrane test parameters using Axon software, and I Ca-L was normalized to cell capacitance (C) byi Ca-L /C. The data are expressed as I Ca-L density in picoamperes per picofarad. L-Type calcium channel activation curves were constructed by plotting normalized conductance as a function C 217 The Authors. Experimental Physiology C 217 The Physiological Society

4 Exp Physiol 12.7 (217) pp Sex differences in right ventricular L-type Ca 2+ current levels 87 of test potential. Parameters for the voltage dependence of activation were obtained from the least-squares fit of data points to the following equation: g/g max = 1/{1 + exp [(V T V.5 )/b]} where g/g max represents the normalized Ca 2+ conductance, V T are test voltage steps from 3 to + 3 mv, V.5 is the voltage at which current activation is half-maximal, and b is the slope. Parameters for the voltage dependence of inactivation were obtained from the following equation: I = I ir + (1 I ir )/{1 + exp [(V c V.5 )/b]} where I is the normalized peak inward current measured during a test pulse to mv following a prolonged (5 s) conditioning pulse (V c ) to membrane potentials between 5 and +2 mv, I ir is the inactivation-resistant current, V.5 is the potential at which current inactivation was half-maximal, and b is the slope. To correct for possible current rundown, the current elicited during the test pulse was normalized to the magnitude of the current recorded during a pretest pulse from 8 to mv, which preceded each conditioning pulse. Cell culture of isolated myocytes To investigate the effects of oestrogen on I Ca-L, isolated female ventricular base myocytes cells were maintained in vitro for 2 days (Chen et al. 211; Yang et al. 212). For this, isolated myocytes in a high-k + buffer solution were allowed to settle, and the pellet was washed once and resuspended in an incubation medium consisting of Phenol Red-free medium 199 supplemented with 5% fetal bovine serum and 1 μg ml 1 primocin (Gibco-Life Technologies, Grand Island, NY, USA). Cells were incubated (in air supplemented with 5% CO 2 )at 37 C for 2 days on laminin-coated (2 μg ml 1 in serum-free medium 199) culture dishes in the presence or absence of 17β-estradiol (E 2 ;1nM; Sigma-Aldrich). The E 2 was added to the cultures from a stock solution in DMSO (final concentration of 1:1,). Control cells were incubated without E 2, but with the same final concentration of DMSO. Mathematical model To investigate the effects on the cardiac AP of altering L-type calcium current magnitude, we used a modification of the Luo Rudy 2 (LRd) mathematical model of the mammalian ventricular AP (Luo & Rudy, 1994a,b; Faber & Rudy, 2). This model includes voltage-gated ionic, exchanger and background currents. The membrane voltage equation is as follows: C dv dt + I Na + I Ca-L + I Ca-T + I Kr + I Ks + I Kl + I to + I NCX + I Na-K + I Na-b + I Ca-b + I Kp + I p(ca) =+I stim Voltage-gated ionic currents incorporated into the model include the sodium (I Na ), L-type and T-type calcium (I Ca-L and I Ca-T ), rapid and slow delayed rectifier potassium (I Kr and I Ks ), inward rectifier potassium (I K1 ) and transient outward (I to ) currents. Additional voltage-independent currents that function continuously include the sodium calcium exchanger (I NCX ), sodium potassium pump (I Na-K ), background sodium, background calcium and plateau potassium currents (I Na-b, I Ca-b and I Kp ) and the sarcolemmal calcium pump (I p(ca) ). A periodic stimulus current (I stim ) is used to model the signal from pacemaker cells that trigger each AP. The LRd model also tracks the intracellular ionic concentrations of sodium, potassium and calcium within three different intracellular compartments, namely the myoplasm, the network SR and the junctional SR. It updates ion concentrations at each time step and also accounts for the dynamics of intracellular calcium, including release, uptake, leak and translocation currents between the SR and myoplasm, as well as calcium buffering in the myoplasm and junctional SR. The model is coded in C++ and uses the code to solve the system of differential equations. The voltage equation is solved using the exponential Euler method, and the approximated voltage is used to solve for the gating variables (Rush & Larsen, 1978; Victorri et al. 1985). Buffering is modelled via a set of algebraic equations updated at each time step. The model is run for 3 cycles to approach a stimulus-driven steady state. Although LRd was designed to model guinea-pig ventricular myocytes, we adapted it for rabbit myocytes by the addition of the transient outward current (Dumaine et al. 1999). To mimic LQT2, the model equations were modified to reduce I Kr conductance by 5% (Makkar et al. 1993; Viswanathan et al. 1999; Sims et al. 28; Maruyama et al. 211). Statistical analysis All results are reported as the mean ± SD of at least three or more independent experiments. For I Ca-L density comparisons between male/female base/apex myocytes, statistical analysis used a 2 2 two-factor ANOVA, which showed significant interaction, followed by a one-factor test with Tukey s post hoc tests to identify significant pairings. Statistical comparisons between two groups of experimental data used Student s two-tailed t test. For all tests, statistical difference was defined as P <.5. Results Sex differences in I Ca-L density Whole-cell patch recordings of I Ca-L were obtained from acutely isolated right ventricular base and apex myocytes C 217 The Authors. Experimental Physiology C 217 The Physiological Society

5 88 Z. M. Kalik and others Exp Physiol 12.7 (217) pp taken from adult rabbits of both sexes. Peak current was normalized to cell capacitance to yield current density. During voltage steps to mv, I Ca-L activated rapidly, reaching its peak amplitude within 1 12 ms, then slowly inactivated (Fig. 1). Female base myocytes displayed a 67% larger I Ca-L density (6.5 ± 1.9 pa pf 1, n = 7) compared with male base myocytes (3.9 ± 1.4 pa pf 1, n = 12; P <.2). In addition, female base myocytes exhibited a 55% larger I Ca-L density compared with female apex myocytes (4.2 ±.8 pa pf 1, n = 8; P <.25). In contrast, there were no significant differences in I Ca-L density between male base and male apex myocytes (4.3 ± 1.2 pa pf 1, n = 4) or between male and female apex myocytes. Thus, I Ca-L showed both sex and regional differences. Similar differences were seen in the I V relationships for peak I Ca-L (Fig. 2). For all cells, the current began to activate near 3 mv, and the maximal current was observed at +1 mv. Peak I Ca-L densities for female base myocytes (n = 4) were significantly greater at voltages between 1 and +4 mv compared with male base myocytes (n = 8; P <.1). Female base myocytes also exhibited significantly greater I Ca-L densities at voltages ofto +1 mv compared with female apex myocytes (n = 5; P <.1). No differences were observed at any voltage in I Ca-L densities between female and male apex myocytes (n = 5) or between male apex and base myocytes. Thus, I Ca-L density was significantly larger in female base myocytes versus female apex, male apex or male base myocytes. To test whether differences in the voltage dependence of I Ca-L might underlie the observed differences in current density, activation and inactivation properties of I Ca-L were measured directly. In apex and base myocytes of both sexes, I Ca-L began to activate at approximately 3 mv and reached its maximal conductance at +1 mv (Fig. 3A), in agreement with the I V measurements (Fig. 2). Activation curves for all myocytes yielded similar half-activation voltages and slopes (Table 1). For current inactivation (Fig. 3B), conditioning voltage steps of 6 mv or more negative did not decrease the subsequent maximal amplitude of I Ca-L.However,less negative conditioning steps decreased I Ca-L amplitude, with voltages above 3 mv completely inactivating I Ca-L in all myocytes. Myocyte inactivation curves showed no differences in half-inactivation voltages or slopes (Table 1). Together, these data indicate that differences in the voltage dependence of I Ca-L activation or inactivation cannot explain the larger current density observed in female base myocytes. Effects of oestrogen on I Ca-L Previous studies have demonstrated that oestrogen can enhance APD and EAD production and upregulate I Ca-L (Drici et al. 1996; Hara et al. 1998; Pham et al. 21; Yang et al. 212). To test the effects of oestrogen on I Ca-L density, adult female myocytes isolated from the base region of the right ventricle were cultured for 2 days in the absence or presence of 17β-estradiol (E 2 ;1nM). The E 2 treatment significantly enhanced I Ca-L density in female base myocytes (Fig. 4). After 1 day of culture, I Ca-L density measured at mv was increased by 56%, from 7.5 ± 3.2 (control, n = 7) to 11.7 ± 2.9 pa pf 1 (E 2 treated, n = 5; P =.4). After 2 days, I Ca-L was 65% higher in E 2 -treated myocytes (E 2, 18.3 ± 3.2 pa pf 1, n = 4; control, 11.1 ± 2.4 pa pf 1, n = 4; P =.1). Mathematical simulations We used our modified LRd mathematical model to investigate the effects of I Ca-L current density on the cardiac AP. Simulations were run with varying levels of I Ca-L density to simulate male or female right ventricular base myocyte responses (Fig. 5; cardiac APs, top traces; I Ca-L density, lower traces). Continuous lines represent control conditions. To simulate LQTS type 2 (dotted lines), the maximal I Kr conductance was reduced to 5% (Makkar et al. 1993; Viswanathan et al. 1999; Sims et al. 28; Maruyama et al. 211). The APDs were measured from AP peak to 9% of repolarization (APD9). The initial simulation (Fig. 5A; top panel, continuous line) used the LRd default I Ca-L current density permeability (1. ) to represent normal male base myocytes. As female right ventricular base myocytes exhibit a 67% larger I Ca-L density (Fig. 1E), these cells were simulated by increasing the I Ca-L density permeability to 17% (1.7 ) of the basal level (Fig. 5B; top panel, continuous line). In control conditions, this increase in calcium current magnitude prolonged APD9 by 19%, from 145 (basal I Ca-L ) to 172 ms (1.7 I Ca-L ). Additional increases in I Ca-L density permeability further prolonged APD9, increasing it by 23% to 178 ms at 185% I Ca-L density permeability (1.85 ;Fig.5C; top panel, continuous line), and by 27% to 184 ms at 2% I Ca-L density permeability (2. ;Fig.5D; top panel, continuous line). In all of these simulations, the underlying calcium current (Fig. 5A D; bottom panels, continuous lines) followed the same time course as the AP, with the peak inward current corresponding to production of the AP peak. With a 5% reduction in the repolarizing I Kr conductance to mimic LQT2, APD9 lengthened in the male base myocyte simulation (1. I Ca-L density) by 19%, from 145 to 173 ms (Fig. 5A; top panel, dotted line). At higher I Ca-L density permeabilities (from 11 to 16% of basal), APD9 gradually increased (data not shown). However, at 17%, similar to the I Ca-L density observed in female base myocytes, a 5% I Kr conductance decrease resulted in an initial AP peak followed by a second C 217 The Authors. Experimental Physiology C 217 The Physiological Society

6 Exp Physiol 12.7 (217) pp Sex differences in right ventricular L-type Ca 2+ current levels 89 A 1 B 1 I Ca-L Density (pa/pf) 5 5 Male Base I Ca-L Density (pa/pf) 5 5 Female Apex C 1 D 1 I Ca-L Density (pa/pf) Female Apex I Ca-L Density (pa/pf) Male Base E 1 8 Peak I Ca-L Density (pa/pf) Male Base Female Apex Figure 1. Sex and regional differences in L-type calcium current (I Ca-L ) Whole-cell I Ca-L current densities were measured at mv in acutely isolated male and female right ventricular myocytes. A D, representative examples of I Ca-L recorded from individual myocytes. Female base myocytes exhibited larger current densities than those found in male base myocytes (A) or female apex myocytes (C). Currents from male and female apex myocytes (B) or from male base and apex myocytes (D) displayed similar peak amplitudes. E, summary of differences in I Ca-L density. The I Ca-L density was 67% higher in female base myocytes (6.5 ± 1.9 pa pf 1, n = 7) compared with male base myocytes (3.9 ± 1.4 pa pf 1, n = 12, P <.2), and 55% higher than female apex myocytes (4.2 ±.8 pa pf 1, n = 8; P <.25). There were no significant differences between male apex (4.3 ± 1.2pApF 1, n = 4) and female apex myocytes or between male base and apex myocytes. C 217 The Authors. Experimental Physiology C 217 The Physiological Society

7 81 Z. M. Kalik and others Exp Physiol 12.7 (217) pp peak (Fig. 5B; top panel, dotted line). This second peak is an EAD, initiated by reactivation of the underlying I Ca-L (Fig. 5B; bottom panel, dotted line), which increased APD9 by 12%, from 172 to 348 ms. Additional LQT2 simulations, using I Ca-L density permeabilities of 185 or 2% of basal, resulted in the appearance of multiple EADs (Fig. 5C and D; top panel, dotted lines), with each EAD associated with a reactivation of I Ca-L (Fig. 5C and D; bottom panels, dotted lines). The EADs greatly increased APD9 to 436 (1.85 ) or 441 ms (2. ). At these higher I Ca-L levels (above 1.7 ), the number of EADs per AP cycled in short patterns, with patterns of one then two EADs per AP, or two, two, then three EADs per AP. Similar variations in EAD production have also been observed in isolated myocytes and numerical models (Sridhar et al. 28; Weiss et al. 21). Our simulation results are summarized in Fig. 6, which shows the relationship between APD and changes in I Ca-L density permeability during different magnitudes of I Kr blockage. In the absence of I Kr inhibition (%), doubling I Ca-L density from the baseline level of 1. up to 2. results in only small increases in APD (Fig. 6A, contiuous line). A small decrease in I Kr (3% inhibition) produces similar responses. However, with increasing levels of I Kr blockage (4 8% I Kr inhibition), elevation of I Ca-L density above a threshold level results in a simulated AP that is accompanied by one or more EADs, causing the APD to increase dramatically. These same effects are exhibited in a two-dimensional contour plot (Fig. 6B), in which APD is displayed using a colour map for a given pair of parameter values of percentage I Kr inhibition and I Ca-L density permeability scaling. The region in the parameter space in which there arenoeadsisshowninblue.inthisregion,thelevel curves of equal APD are approximately linear. As the percentage I Kr inhibition and I Ca-L density permeability scaling increase, there are transitions to regions of one EAD (light green), two EADs, (yellow) and three EADs (red). A C I Ca-L Density (pa/pf) Male Base 2 Male Base Female Apex 2 I Ca-L Density (pa/pf) B Female Apex 2 Female Apex D I Ca-L Density (pa/pf) I Ca-L Density (pa/pf) Male Base Figure 2. Sex and regional differences in I Ca-L I V relationships Peak I Ca-L densities obtained during voltage steps over the range of 3 to +6 mv. Maximal current densities were significantly larger in female base myocytes (n = 4; open circles) compared with male base myocytes (n = 8; filled circles; P <.1) or with female apex myocytes (n = 5; open triangles; P <.1). The I Ca-L densities were similar over all voltages between female apex (n = 5) and male apex myocytes (n = 4; filled triangles) and between male base and male apex myocytes. C 217 The Authors. Experimental Physiology C 217 The Physiological Society

8 Exp Physiol 12.7 (217) pp Sex differences in right ventricular L-type Ca 2+ current levels 811 Our electrophysiological recordings from isolated myocytes showed that I Ca-L of female base myocytes is 67% larger (1.67 ) thanthatofmalebasemyocytes. Therefore, within the two-dimensional parameter space (Fig. 6B), the I Ca-L density permeability scalings for female base myocytes is 1.67, compared with 1. for male base myocytes, 1.1 for male apex myocytes, and 1.8 for female apex myocytes. Owing to the low I Ca-L density levels in male base, male apex and female apex myocytes, EAD initiation would require a near-total blockage of I Kr. In contrast, the threshold for EAD initiation in female base myocytes is 55% I Kr inhibition. The simulated appearance of an EAD at the I Ca-L density of female base myocytes with only partial I Kr blockage suggests that congenital or acquired reductions in repolarization reserve will make females more susceptible to the appearance of arrhythmogenic EADs. A Normalized Conductance B Normalized Availability Male Base Female Apex Male Base Female Apex Conditioning Potential (mv) Figure 3. Activation and inactivation properties of I Ca-L Voltage dependence of I Ca-L activation (A) and steady-state inactivation (B) in male and female right ventricle base and apex myocytes. Insets show voltage protocols used to generate the experimental data. See Table 1 for details. Discussion Whole-cell patch-clamp recordings obtained from right ventricular myocytes acutely isolated from the apex and base regions of adult male and female rabbits have demonstrated sex and regional differences in L-type calcium current density. We found significantly higher levels of I Ca-L in female base myocytes compared with female apex, male base or male apex myocytes. No differences in current density were found among female apex, male apex and male base myocytes. Currents recorded from base and apex myocytes of both sexes demonstrated similar I V relationships, with no differences in voltage dependences of current activation or inactivation. As female rabbit right ventricular base myocytes express significantly higher levels of L-type calcium channel protein (Cav1.2α; Kim et al. 213), it is likely that the elevated peak I Ca-L amplitude we observed in female base myocytes is attributable to the increased expression of calcium channels, rather than to variations in voltage-dependent channel properties. A similar conclusion was reached in studies of base myocytes from the left ventricle of adult female rabbits, which also display high Cav1.2α expression levels correlated with elevated I Ca-L density (Sims et al. 28; Yang et al. 212). Our results demonstrated that I Ca-L densities in adult female base myocytes from the right ventricle were 67 and 55% larger than current levels seen in male base or female apex myocytes, respectively. These differentials are larger than those observed for I Ca-L density in left ventricular myocytes (48% difference for female versus male base myocytes; 38% for female base versus apex myocytes; Sims et al. 28). However, when compared with right ventricular myocytes, both base and apex myocytes from the female left ventricle (Sims et al. 28) exhibited higher peak I Ca-L densities, with left ventricle base myocytes having a 46% larger I Ca-L density compared with right ventricle base myocytes (9.5 versus 6.5 pa pf 1 )and a 64% higher current density for apex myocytes from the left versus right ventricle (6.9 versus 4.2 pa pf 1 ). Thus, the right ventricle shows greater sex and regional variations in I Ca-L densities when compared with the left ventricle, but the magnitudes of the I Ca-L densities found in left ventricular myocytes are much larger than those of comparable myocytes from the right ventricle. As our simulations showed, increasing I Ca-L density prolongs the duration of the AP. Thus, it is likely that the larger I Ca-L amplitude of left ventricular myocytes will contribute to a longer APD in these cells. Itiswellknownthatadultfemalesareatincreasedrisk for LQT2 and drug-induced arrhythmias, such as TdP (Makkar et al. 1993; Locati et al. 1998; Drici & Clement, 21; Salama & Bett, 214; Roden, 216). Oestrogen may enhance this risk by reducing repolarization reserve with a partial suppression of the rapid delayed rectifier C 217 The Authors. Experimental Physiology C 217 The Physiological Society

9 812 Z. M. Kalik and others Exp Physiol 12.7 (217) pp Table 1. Activation and inactivation parameters for I Ca-L Activation Inactivation Myocyte V r (mv) V.5,act (mv) Slope n V.5,inact (mv) Slope n Male apex 54 ± ± ± ± ±.4 4 Male base 52 ± ± ± ± ±.4 4 Female apex 53 ± ± ± ± ±.2 4 Female base 52 ± ± ± ± ±.3 3 Abbreviations: V r, reversal potential; V.5,act, half-activation voltage; and V.5,inact, half-inactivation voltage. No significant sex or regional differences were found among the parameters. potassium current (I Kr ;Kurokawaet al. 28). In addition, we found that oestrogen upregulated I Ca-L in cultured female right ventricle base myocytes. Previous studies of female left ventricular rabbit myocytes have found similar oestrogen-induced upregulations of I Ca-L and the depolarizing sodium calcium exchange current (I NCX )in base but not apex myocytes (Chen et al. 211; Yang et al. 212). These findings suggest that oestrogen plays a role in base apex ionic current heterogeneity, thus enhancing the excitability of base region myocytes and increasing the vulnerability of adult females to arrhythmias. Numerical simulations were performed using a modified version of the Luo Rudy model of the cardiac AP to predict sex and regional differences in arrhythmia Peak l Ca-L Density (pa/pa) Control (day 1) E2 (day 1) Control (day 2) E2 (day 2) Figure 4. Effects of oestrogen on I Ca-L density of cultured adult female right ventricle base myocytes Myocytes were cultured for 1 or 2 days in the absence (control) or presence (E 2 )of1nm 17β-estradiol. Significantly higher I Ca-L levels were observed in E 2 -treated myocytes when compared with control myocytes ( P =.4; P =.1). susceptibility. We found that increasing the magnitude of I Ca-L lengthened APDs, suggesting that the elevated I Ca-L density observed in female base myocytes could be an important contributing factor to the longer APDs and rate-corrected QT intervals of adult women (Makkar et al. 1993; Pham & Rosen, 22). However, even at the highest simulated levels of I Ca-L,noEADswereobserved, suggesting that elevated I Ca-L alone is not arrhythmogenic. Women have less repolarizing capacity than men, which also contributes to the longer APD of women, making them more susceptible to LQT2, a potentially fatal prolongation of the cardiac AP resulting from a congenital or drug-induced decrease in I Kr (Makkar et al. 1993; Pham & Rosen, 22; James et al. 27; Webster & Berul, 28; Salama & Bett, 214; Roden, 216). To model LQT2 and examine the effects of elevated I Ca-L and reduced repolarization, AP simulations were run in which I Kr conductance was decreased by various amounts. Our simulations showed that the appearance of arrhythmogenic EADs does not occur immediately with AP prolongations resulting from increased I Ca-L density ( ), even when combined with a partial reduction (3%) in I Kr.However,athigherlevelsofI Kr inhibition, elevation of I Ca-L above a threshold level yielded an AP that was greatly prolonged because of the appearance of a large EAD. The underlying simulated calcium current indicated that the EAD was initiated by I Ca-L reactivation. When the I Ca-L density was set at 7% above the baseline level (1.7 ) to mimic the 67% larger I Ca-L that we observed in female compared with male right ventricular base myocytes, simulations using a 5% inhibition of I Kr to model LQT2 resulted in the appearance of an EAD. Additional simulations demonstrated the dependence of EAD initiation on the interplay between I Ca-L density and I Kr inhibition over broad ranges of these parameters. These findings lend support for the hypothesis that sex differences in I Ca-L magnitude can be an important contributor to arrhythmia susceptibility in women. Our findings are similar to those of a previous study (Sims et al. 28), which showed that adult rabbit female base myocytes isolated from the left ventricle exhibited elevated I Ca-L density when compared with male base, C 217 The Authors. Experimental Physiology C 217 The Physiological Society

10 Exp Physiol 12.7 (217) pp Sex differences in right ventricular L-type Ca 2+ current levels 813 A x B x Current Density (pa/pf) Current Density (pa/pf) C x D 4 2.x Current Density (pa/pf) Current Density (pa/pf) Figure 5. Myocyte action potential and I Ca-L simulations Numerical simulations obtained from the Luo Rudy 2 cardiac myocyte model, showing the voltage response (above) and the corresponding I Ca-L response (below). The control responses (continuous lines) are compared with responses obtained with a 5% reduction in rapid delayed rectifier potassium current (I Kr ) conductance to simulate long QT syndrome type 2 (LQT2; dotted lines). A, the default I Ca-L density permeability (1. ) of the model yielded an action potential duration (APD) of 145 ms (control) and 173 ms (LQT2). B, increasing I Ca-L density permeability to 17% (1.7 ) of the basal level resulted in the appearance of an early after-depolarization (EAD; second peak). The EAD increased overall APD from 172 (control) to 348 ms (LQT2). Note that in the LQT2 simulation, appearance of the EAD corresponds to reactivation of the underlying I Ca-L. C, atani Ca-L density permeability of 185% (1.85 ), APD was 178 (control) or 436 ms (LQT2), with two EADs. D, at 2% (2. ) I Ca-L density permeability, APD was 184 (control) or 441 ms (LQT2), with two EADs. All figures show the 3th action potential, to ensure that the simulation system had reached a stimulus-driven steady state. C 217 The Authors. Experimental Physiology C 217 The Physiological Society

11 814 Z. M. Kalik and others Exp Physiol 12.7 (217) pp female apex or male apex myocytes, and in the presence of E431 (a I Kr blocker) to simulate LQT2, EADs were observed only with female base myocytes. That study also showed that simulations of left ventricular APs using an elevated I Ca-L to model female base myocytes resulted in EAD formation when combined with a 5% reduction in I Kr repolarization (Sims et al. 28). In addition, optical voltage mapping of the anterior ventricular wall of isolated hearts in the presence of E431 found that the earliest EADs and their progression to TdP preferentially occurred in females at the base regions of both the left and right ventricles (Choi et al. 22; Sims et al. 28). Thus, Sims A AP Duration % I Kr Inhibition 8% 7% 6% 5% 4% 3% % I Ca-L Density Permeability Scaling B I Ca-L Density Permeability Scaling APD (ms) % I Kr Inhibition Figure 6. Relationship between action potential duration (APD) and I Ca-L density permeability during different levels of I Kr inhibition A, with no reduction in I Kr (% I Kr inhibition) or a low level (3%) of I Kr inhibition, increasing I Ca-L density over the range of results in only small increases in APD. At higher levels of I Kr blockage (4 8% I Kr inhibition), elevations of I Ca-L above a threshold level yield substantial increases in APD owing to the initiation of one, two or three EADs (seen as changes in the plateau levels of the response curves). B, these same effects are shown in a two-dimensional contour plot, in which the APD for a pair of values of percentage I Kr inhibition and I Ca-L density permeability scaling is displayed as a colour map. The blue region in the parameter space corresponds to the absence of EADs, while the light green, yellow and red regions correspond to one, two and three EADs, respectively. C 217 The Authors. Experimental Physiology C 217 The Physiological Society

12 Exp Physiol 12.7 (217) pp Sex differences in right ventricular L-type Ca 2+ current levels 815 et al. (28) concluded that the elevated I Ca-L of female left ventricular base myocytes promoted EAD initiation. The observation that the base region of the right ventricle was also a site for EAD initiation (Choi et al. 22; Sims et al. 28), combined with our measurements of elevated I Ca-L density in right ventricular myocytes and LQT2 AP simulations, strongly supports the hypothesis that the higher levels of I Ca-L in both right and left female base ventricular myocytes can contribute to arrhythmogenic EAD genesis. Reactivation of I Ca-L has been implicated as the trigger for EAD production in LQT2 or during drug-induced repolarization inhibition (January & Riddle, 1989; Volders et al. 2; Antoons et al. 27; Sims et al. 28). It has been postulated that the prolonged AP depolarization was directly responsible for I Ca-L reactivation and EAD initiation (January & Riddle, 1989), suggesting that an elevated I Ca-L density might be sufficient to extend APD to the point that allows for I Ca-L reactivation. However, we did not see evidence for this in simulations run without a reduction in I Kr. Thus, no EADs were generated in these conditions, even when APD was prolonged by 32% after a doubling of the basal I Ca-L density. More recent studies have hypothesized that AP prolongation promotes SR calcium overload, followed by spontaneous SR calcium release and enhancement of the depolarizing I NCX,thus leading to I Ca-L reactivation (Choi & Salama, 2; Volders et al. 2; Choi et al. 22; Weiss et al. 21; Ginsburg et al. 213). In support of this, NCX1 and I NCX have been found to be significantly elevated in adult female left and right ventricular base myocytes compared with female apex or male base or apex myocytes (Chen et al. 211; Kim et al. 213). Thus, elevated I Ca-L and I NCX densities are found in both right and left female base myocytes, a combination that would greatly enhance the propensity of these cells to develop arrhythmogenic EADs. Although the appearance of EADs and polymorphic ventricular tachycardias in LQT2 is often associated with sympathetic stimulation, in the present study we measured I Ca-L densities only in baseline conditions, and thus further studies are needed to determine the effects of isoprenaline or sympathetic stimulation on I Ca-L current densities and APD in right ventricular base and apex myocytes. In addition, the mathematical simulations did not model sympathetic stimulation, nor did they address the role of intracellular calcium release and any potential effect this might have for EAD formation in apex versus base myocytes. The rabbit myocyte model by Mahajan et al. (28), with more sophisticated intracellular calcium cycling, could be used for such a study. Conclusion This study demonstrated that base myocytes from the female rabbit right ventricle display significantly higher I Ca-L density, and this current is upregulated by oestrogen. Mathematical simulations found that this level of I Ca-L, combined with a partial reduction of I Kr repolarization, resulted in EAD initiation. Thus, our findings support the hypothesis that in women, the elevated I Ca-L of base myocytes, coupled with a partial reduction in repolarizing ability attributable to a congenital or drug-induced reduction of I Kr as seen in LQT2, may play a significant role in an increased susceptibility for ventricular arrhythmia initiation. References Abi-Gerges N, Philip K, Pollard C, Wakefield I, Hammond TG & Valentin JP (24). Sex differences in ventricular repolarization: from cardiac electrophysiology to Torsades de Pointes. Fundam Clin Pharmacol 18, Ackerman MJ & Mohler PJ (21). Defining a new paradigm for human arrhythmia syndromes: phenotypic manifestations of gene mutations in ion channel- and transporter-associated proteins. Circ Res 17, Antoons G, Volders PGA, Stankovicova T, Bito V, Stengl M, Vos MA & Sipido KR (27). Window Ca 2+ current and its modulation by Ca 2+ release in hypertrophied cardiac myocytes from dogs with chronic atrioventricular block. JPhysiol579, Beyer C & McDonald P (1973). Hormonal control of sexual behavior in the female rabbit. Adv Reproduct Physiol 6, Bonatti V, Rolli A & Botti G (1983). Recording of monophasic action potentials of the right ventricle in long QT syndromes complicated by severe ventricular arrhythmias. Eur Heart J 4, Brugada J & Diez DP (21). How to recognize and manage idiopathic ventricular tachycardia. E-Journal of Cardiology Practice 8, N26. Buxton A, Waxman HL, Marchlinski FE, Simson MB, Cassidy D & Josephson ME (1983). Right ventricular tachycardia: clinical and electrophysiologic characteristics. Circulation 68, Chen G, Yang X, Alber S, Shusterman V & Salama G (211). Regional genomic regulation of cardiac sodium calcium exchanger by oestrogen. JPhysiol589, Choi BR, Burton F & Salama G (22). Cytosolic Ca 2+ triggers early afterdepolarizations and Torsade de Pointes in rabbit hearts with type 2 long QT syndrome. JPhysiol543, Choi BR & Salama G (2). Simultaneous maps of optical action potentials and calcium transients in guinea-pig hearts: mechanisms underlying concordant alternans. JPhysiol529, de Turckheim M, Berger M, Jean-Faucher CH, Veyssière G & Jean Cl (1983). Changes in ovarian oestrogens and in plasma gonadotrophins in female rabbits from birth to adulthood. Acta Endocrinol (Copenh) 13, Drici MD, Burklow TR, Haridasse V, Glazer RI & Woosley RL (1996). Sex hormones prolong the QT interval and downregulate potassium channel expression in the rabbit heart. Circulation 94, C 217 The Authors. Experimental Physiology C 217 The Physiological Society

13 816 Z. M. Kalik and others Exp Physiol 12.7 (217) pp Drici MD & Clement N (21). Is gender a risk factor for adverse drug reactions? The example of drug-induced long QT syndrome. Drug Saf 24, Dumaine R, Towbin JA, Brugada P, Vatta M, Nesterenko DV, Brugada J, Brugada R & Antzelevitch C (1999). Ionic mechanisms responsible for the electrocardiograph phenotype of the Brugada syndrome are temperature dependent. Circ Res 85, Faber GM & Rudy Y (2). Action potential and contractility changes in [Na + ] i overloaded cardiac myocytes: a simulation study. Biophys J 78, Gaborit N, Varro A, LeBouter S, Szuts V, Escande D, Nattel S & Demolombe S (21). Gender-related differences in ion-channel and transporter subunit expression in non-diseased human hearts. JMolCellCardiol49, Ginsburg KS, Weber CR & Bers DM (213). Cardiac Na + Ca 2+ exchanger: dynamics of Ca 2+ -dependent activation and deactivation in intact myocytes. JPhysiol591, Hara M, Danilo P Jr & Rosen MR (1998). Effects of gonadal steroids on ventricular repolarization and on the response to E431. J Pharmacol Exp Ther 285, James AF, Choisy SCM & Hancox JC (27). Recent advances in understanding sex differences in cardiac repolarization. Prog Biophys Mol Biol 94, January CT & Riddle JM (1989). Early afterdepolarizations: mechanism of induction and block. A role for L-type Ca 2+ current. Circ Res 64, Kim JJ, Nemec J, Pap R, Strongin R, Abramson JJ & Salama G (213). Bradycardia alters Ca 2+ dynamics enhancing dispersion of repolarization and arrhythmia risk. Am J Physiol Heart Circ Physiol 34, H848 H86. Kurokawa J, Tamagawa M, Harada N, Honda S-I, Bai C-X & Nakaya H, Furukawa T (28). Acute effects of oestrogen on the guinea pig and human I Kr channels and drug-induced prolongation of cardiac repolarization. JPhysiol586, Liu T, Choi BR, Drici MD & Salama G (25). Sex modulates the arrhythmogenic substrate in prepubertal rabbit hearts with Long QT 2. J Cardiovasc Electrophysiol 16, Locati EH, Zareba W, Moss AJ, Schwartz PJ, Vincent GM, Lehmann MH, Towbin JA, Priori SG, Napolitano C, Robinson JL, Andrews M, Timothy K & Hall WJ (1998). Age- and sex-related differences in clinical manifestations in patients with congenital Long-QT Syndrome: findings from the international LQTS registry. Circulation 97, Luo CH & Rudy Y (1994a). A dynamic model of the cardiac ventricular action potential. I. Simulations of ionic currents and concentration changes. Circ Res 74, Luo CH & Rudy Y (1994b). A dynamic model of the cardiac ventricular action potential. II. Afterdepolarizations, triggered activity, and potentiation. Circ Res 74, Mahajan A, Shiferaw Y, Sato D, Baher A, Olcese R, Xie LH, Yang MJ, Chen PS, Restrepo JG, Karma A, Garfinkel A, Qu Z & Weiss, JN (28). A rabbit ventricular action potential model replicating cardiac dynamics at rapid heart rates. Biophys J 94, Makkar RR, Fromm BS, Steinman RT, Meissner MD & Lehmann MH (1993). Female gender as a risk factor for torsades de pointes associated with cardiovascular drugs. JAMA 27, Maruyama M, Lin SF, Xie Y, Chua SK, Joung B, Han S, Shinohara T, Shen MJ, Qu Z, Weiss JN & Chen PS (211). Genesis of Phase 3 early afterdepolarizations and triggered activity in acquired Long-QT syndrome. Circ Arrhythm Electrophysiol 4, Noda T, Shimizu W, Taguchi A, Aiba T, Satomi K, Suyama K, Kurita T, Aihara N & Kamakura S (25). Malignant entity of idiopathic ventricular fibrillation and polymorphic ventricular tachycardia initiated by premature extrasystoles originating from the right ventricular outflow tract. JAm Coll Cardiol 46, Pham TV, Robinson RB, Danilo P Jr & Rosen MR (22). Effects of gonadal steroid on gender-related differences in transmural dispersion of L-type calcium current. Cardiovasc Res 53, Pham TV & Rosen MR (22). Sex, hormones, and repolarization. Cardiovasc Res 53, Pham TV, Sosunov EA, Gainullin RZ, Danilo P Jr & Rosen MR (21). Impact of sex and gonadal steroids on prolongation of ventricular repolarization and arrhythmias induced by I K -blocking drugs. Circulation 13, Roden DM (216). Predicting drug-induced QT prolongation and torsades de pointes. JPhysiol594, Rush S & Larsen H (1978). A practical algorithm for solving dynamic membrane equations. IEEE Trans Biomed Eng 25, Salama G & Bett GCL (214). Sex differences in the mechanisms underlying long QT syndrome. Am J Physiol Heart Circ Physiol 37, H64 H648. Sims C, Reisenweber S, Viswanathan PC, Choi B, Walker WH & Salama G (28). Sex, age, and regional differences in L-type calcium current are important determinates of arrhythmia phenotype in rabbit hearts with drug-induced long QT type 2. Circ Res 12, e86 e1. Sridhar A, Nishijima Y, Terentyev, D, Terentyeva R, Uelmen R, Kukielka M, Bonilla IM, Robertson GA, Györke S, Billman GE & Carnes CA (28). Repolarization abnormalities and afterdepolarizations in a canine model of sudden cardiac death. Am J Physiol Regul Integr Comp Physiol 295, R1463 R1472. Verkerk AO, Wilders R, Veldkamp MW, de Geringel W, Kirkels JH & Tan HL (25). Gender disparities in cardiac cellular electrophysiology and arrhythmia susceptibility in human failing ventricular myocytes. Int Heart J 46, Victorri B, Vinet A, Roberge FA & Drouhard JP (1985). Numerical integration in the reconstruction of cardiac action potentials using Hodgkin-Huxley-type models. Comput Biomed Res 18, Viswanathan PC, Shaw RM & Rudy Y (1999). Effects of I Kr and I Ks heterogeneity on action potential duration and its rate dependence: a simulation study. Circulation 99, Volders PG, Vos MA, Szabo B, Sipido KR, de Groot SH, Gorgels AP, Wellens HJ & Lazzara R (2). Progress in the understanding of cardiac early afterdepolarizations and torsades de pointes: time to revise current concepts. Cardiovasc Res 46, Webster G & Berul CI (28). Congential long-qt syndromes: a clinical and genetic update from infancy through adulthood. Trends Cardiovasc Med 18, C 217 The Authors. Experimental Physiology C 217 The Physiological Society