Regional expression of cardiac ion channels and cardiac electrical activity

Introduction

Important differences in electrophysiological properties have been noted between different regions of the heart. Electrophysiological heterogeneity has also been detected within different parts of a given tissue, such as the ventricular subendocardium, midmyocardium and subepicardium. Although many molecular candidates for native ionic currents have been identified, the molecular basis of most currents is not completely understood. Heterogeneity of channel protein composition might well underlie the differences observed among the properties of ionic currents or the shape of the action potential in different regions of the heart. Techniques such as immunocytochemistry, immunohistochemistry and Western blotting have played an important role in identifying tissue expression of channel proteins as well as their cellular localization. This review summarizes the electrophysiological differences observed in different regions of the heart and correlates them with information regarding expression patterns of ion channel subunits that might account for regional functional variation.

The inward rectifier current, I_K1

I_K1 is responsible for the maintenance of the cardiac resting potential and the final phase of cardiac repolarization. It is believed that differences between atrial and ventricular I_K1 are responsible for the differences observed between the atrial and ventricular action potential. The atrial action potential has a short plateau phase and relatively slow repolarization, whereas the ventricular action potential has a long plateau and a much faster repolarization phase¹. Patch clamp studies have revealed significant differences between atrial and ventricular I_K1. Atrial I_K1 has a 4-10 fold lower current density and conducts less outward current than ventricular I_K1^1-3. It is believed that a smaller I_K1 contributes to the less negative resting potential and slower terminal repolarization in atrial cells².

Furukawa et al. reported a larger outward component of I_K1 in endocardial cells than in epicardial cells in the cat. No differences in single channel conductance and open probability were observed, suggesting that observed differences are due to differences in current density and not to differences in molecular composition⁴. However, no differences in I_K1 current density or kinetics have been found across the left ventricular wall in myocytes isolated from guinea pig⁵ or dog⁶. The observed discrepancies might be due to species-specific variations.

Clones of the K_ir2 family are believed to underlie I_K1. Four different α subunits have been cloned and shown to be present in the human heart^{2, 7-9}. K_ir2.1 mRNA is the most abundant subunit, with similar concentrations in atrium and ventricle². K_ir2.2 mRNA expression was relatively weak in both atrium and ventricle. K_ir2.3 concentration is approximately 12-fold higher in atrium than in ventricle, but about 10-fold lower than K_ir2.1². The abundance of K_ir2.4 in human heart has not been determined. It has been speculated that the comparatively high abundance of K_ir2.3, a lower-conductance subunit, in atrium might contribute to the smaller current density of atrial I_K1². Recent data suggests that K_ir5.1 can act like an endogenous dominant negative subunit, co-assemble with K_ir2.1 and lead to the formation of nonconducting inward rectifier channels in the brain¹⁰. It is unclear if this mechanism operates in the heart. Nothing is known at present about the cellular localization of K_ir2 subunits.

The delayed rectifier current, I_K

I_K consists of two distinct components, the rapidly activating I_Kr and the slowly activating I_Ks¹¹. I_K is the main phase 3 repolarizing current in the heart. Comparison of I_K between canine right and left atrium showed a consistently higher current density of I_Kr in the left atrium¹². No differences in current kinetics were observed. Western blot experiments showed a significantly greater expression of human ethera-go-go-related gene (Herg) protein in the left atrium, consistent with the electrophysiological data. No significant differences were found in I_Ks current densities¹². These results are significant, given the prominent role of the left atrium in AF maintenance¹³, which may in part be due to abbreviated left atrial refractoriness due to larger I_Kr.

Comparison of Herg protein expression with functional I_Kr expression in rat found Herg protein and I_Kr to be more prominent in atria than in ventricles, whereas Herg protein expression in mice and humans was higher in the ventricle¹⁴. RNase protection assays found similar levels of Herg mRNA in rabbit left atrium, left ventricle, SA node and in canine Purkinje fibres, left and right ventricle suggesting that either interspecies differences or post-translational mechanisms may lead to differences in the level of expressed protein. Herg mRNA levels are ~1.5-fold more abundant than those of K_v4.3 in canine right ventricle¹⁵. Voltage clamp studies using left ventricular guinea pig myocytes revealed a higher I_K current density in sub-epicardial and mid-myocardial than in endocardial myocytes.

Both components, I_Kr and I_Ks, were significantly smaller in sub-endocardial than in midmyocardial or sub-epicardial cells¹⁶. Furukawa et al. reported a higher current density of I_K, with faster activation and delayed deactivation, in left ventricular epicardial compared to endocardial cells of the cat. No changes were found in open probability and single channel conductance leading to the suggestion that the observed differences in current density are due to a higher expression of channel protein and not to differences in molecular composition⁴.

Liu and Antzelevitch described a smaller I_K in mid-myocardial cells compared to cells isolated from the endocardium or epicardium in dogs. Further evaluation showed that the observed difference was due to a significantly smaller I_Ks in mid-myocardium. No changes were found in the current density of I_Kr nor the kinetics of I_Kr or I_Ks in the three cell types¹⁷. In ferret, Herg mRNA transcripts and protein levels were found to be higher in epicardial cell than endocardial cell layers¹⁸.

To further investigate the lower I_Ks density in mid-myocardial cells Péréon and colleagues used RNase protection assays to determine K_vLQT1 message expression across the human right and left ventricular wall. Overall K_vLQT1 expression was similar throughout the ventricular wall. Midmyocardial cells, however, were found to contain a higher percentage of an endogenous N-terminal truncated K_vLQT1 splice variant, referred to as K_vLQT1 isoform 2, which acts as a dominant negative.

Co-expression of isoform 1 with isoform 2 in COS 7 cells in a stoichiometry mimicking that in midmyocardium resulted in a 75% reduction in current amplitude, consistent with the data provided by Liu & Antzelevitch. Current kinetics were not changed by co-expression with K_vLQT1 isoform 2. K_vLQT1 concentrations in atrium were found to be similar to those in endocardial and epicardial left and right ventricular cells. The regulatory subunit minK was found to be similarly expressed in midmyocardium, endocardial and epicardial left and right ventricular cells.

The transient outward K⁺ current, I_to

Regional differences in I_to are well described and contribute to the observed differences in action potential configuration in various species^19-23. I_to density has been found to be greater in epicardial cells than in cells isolated from the endocardium in both dog²⁰ and rat^21,22. In human ventricle, I_to is larger in sub-epicardial than in subendocardial myocytes²⁴. Differences in inactivation kinetics and 4-AP sensitivity have been found between atrial and ventricular human myocytes suggesting potential differences in their molecular basis²³. Action potentials recorded from regions with a high I_to current density show a typical “spike and dome” configuration²⁰.

Based upon differences in activation and inactivation kinetics as well as recovery from inactivation, I_to1 has been divided into two types, fast (I_to,f) and slow (I_to,s)^25-31. I_to,s density is high in left ventricular epicardial myocytes in human²⁸ and ferret²⁷. It is believed that K_v4.2 and K_v4.3 contribute to the rapid component, I_to,f , and k_v1.4 appears to underlie the slow component, I_to,s²². Important differences among various species as well as differences in the distribution of these subunits across the ventricular wall are well documented. The distribution of various subunits contributing to I_to has been most extensively studied in rat. In the atrium, only K_v4.2 has been found³², whereas both K_v4.2 and K_v4.3 have been identified in ventricle³³. K_v4.3 and K_v1.4 mRNA expression across the ventricular wall of rat heart is almost uniform, whereas K_v4.2 shows a marked gradient with higher expression levels in epicardial than endocardial cells^{34, 35}. K_v4.2 expression has been shown to parallel I_to current density across the left ventricular wall³⁴ suggesting that K_v4.2 is the main subunit underlying I_to in rat.

K_v4.2 expression predominates in the right ventricular wall²². Confocal and immunohistochemical studies show that K_v4.2 is localized in the transverse-axial tubular system of the rat myocytes³⁶. K_v4.3 and K_v1.4 may be proportionately more important in the septum²². Important differences have been found among various species. In the mouse, K_v1.4 has been found in high concentration in the septum and is believed to underlie I_to,s, whereas α subunits of the K_v4 subfamily underlie I_to,f in the ventricular apex and septum³⁷. The properties of I_to in rabbit myocytes are similar to those of K_v1.4, suggesting an important role of K_v1.4 in rabbit I_to^38-40. In situ hybridization and immunohistochemistry has shown regional differences in the expression of K_v1.4 and K_v4.2/4.3 in ferret heart, suggesting that K_v1.4 and K_v4.2/4.3 underlie I_to in ferret left ventricular endocardial and epic ardial myocytes, respectively²⁷. The molecular basis for canine and human I_to has been attributed to K_v4.3³⁵. RNase protection assays showed that neither K_v4.1 nor K_v4.2 mRNA is expressed at detectable levels in canine ventricular muscle, whereas K_v4.3 is abundant. K_v4.3 mRNA is expressed in human ventricle at similar levels to those found in canine ventricle³⁵. Han et al. reported the presence of TEA sensitive I_to in canine Purkinje fibers. Their data suggest that the molecular basis of canine Purkinje I_to is different from K_v4.2/4.3 and K_v1.4, since these channels are TEA-insensitive. It has been proposed that K_v3.3 and K_v3.4 subunits might play an important role in Purkinje I_to, but the exact molecular basis still remains to be determined⁴¹.

The Na⁺-Ca²⁺ exchanger (NCX)

The Na⁺-Ca²⁺ exchanger (NCX) catalyzes the exchange of three Na⁺ for one Ca²⁺ across the plasma membrane in many mammalian cells⁴². The transport is reversible and can facilitate Ca²⁺ entry, leading to Ca²⁺ release from the sarcoplasmic reticulum⁴³. The exchange activity is especially high in cardiomyocytes and plays a key role in the maintenance of Ca²⁺ homeostasis and relaxation of Ca²⁺ muscle⁴². The exchange of Ca²⁺ for sodium was first observed in guinea pig atria⁴⁴. Western blotting and Northern blot experiments have shown increased expression levels of the NCX1 isoform in heart failure^45-48.

NCX proteins in mammals are encoded by three genes: NCX1, NCX2 and NCX3^49-51. NCX2 and NCX3 are found only in the skeletal muscle and in the brain⁵². NCX1 transcripts undergo alternative splicing of 6 internal exons (A,B,C,D,E and F) to produce tissue-specific isoforms⁵³. This splicing confers distinct functional characteristics to tissue-specific isoforms of the Na⁺-Ca²⁺ exchanger^53-56. For example, the cardiac isoform (NCX1.1) is less sensitive to depolarizing voltages and to activation by [Ca²⁺]_i than the renal isoform. In addition, NCX1.1 is more sensitive to PKA activation than the renal form⁵⁷. Komuro et al. cloned and characterized the human cardiac Na⁺-Ca²⁺ exchanger in 1992⁵⁸.

Confocal microscopy of adult guinea pig and rat heart cells has shown that the NCX is present in all membranes of the myocytes that face the extracellular space⁵⁹. Dilly et al. showed NCX to be present on the surface and in T-tubule membranes on all cardiac myocytes⁶⁰. Immunohistochemistry has revealed a specific localization in some types of cells⁵⁹. Confocal microscopy has shown that in guinea pig myocytes, the NCX is located at the intercalated disks, the transverse tubules and exterior surface of the membrane⁶¹. In rabbit myocytes, the NCX appears to be more prominent in T-tubule membranes than in peripheral sarcolemma⁶².

The calcium current, I_Ca

Two principal Ca²⁺ currents have been described. The voltage-gated L-type (”long lasting” ) I_Ca is responsible for triggering sarcoplasmic reticulum Ca²⁺ release and consequently the initiation of excitation contraction coupling in heart^63-65 and for the plateau phase of the action potential⁶⁶. L-type Ca²⁺ channels cluster in the surface plasma membrane overlying junctional sarcoplasmic reticulum in guinea pig myocytes⁶⁷. N, P, Q and R-type channels have also been identified^66,68,69 as high voltage activated calcium channels⁷⁰ but appear not to be expressed in cardiomyocytes.

T-type (”transient”) I_Ca inactivates very rapidly and at relatively negative potentials⁶⁴. It is found mainly in sinoatrial node, Purkinje or atrial cells, but its expression is also species dependent⁶⁴. The fact that I_Ca.T is at high density in nodal cells⁷¹ and embryonic cardiomyocytes^{67 ,72} suggests that it is important in pacemaker function^65,67. T-type current is present in the guinea pig heart but has not been found in rat or in human adult myocytes^67,73.

L-type channels were initially purified from skeletal muscle⁷⁴. They are heteromultimers of at least 3 different subunits: α_1C, β and α₂δ. The α_1C subunit encodes the basic pore-forming protein of the channel, whereas the auxiliary β subunit modulates the expression, the open probability, activation and inactivation^67,75-77. The α₂δ subunit is a disulfide-linked dimer^78-80 and is ubiquitously expressed in all types of high voltage-dependent Ca²⁺ channels⁸¹. The α_1C subunit is encoded by three different genes (CaCh1-3). Only the product of CaCh2a is present in the heart. Four β subunit genes have been identified (β₁ to β₄). Genes encoding I_Ca.T were identified in human heart in 1998^70,82. The molecular basis for T-type calcium channels has been associated with α_1G, α_1H and α_1I genes^70,82,83. α_1G and α_1H are expressed in heart and brain at different levels⁷⁰. The cell or tissue specificity hasn’t yet been established. Immunoconfocal and immunogold electron microscopy labelling were used⁸¹ to show the distribution of L-type Ca²⁺ channels in cardiac myocytes isolated from rabbit and rat ventricle. The channels were localized on the surface of the plasma membrane and transverse T-tubules in rabbit myocytes, whereas the labelling was more intense in T-tubules than in the surface sarcolemma in rat myocytes.

Connexins/Gap Junctions

Gap junctions are more than simple conduits allowing for enhanced current flow between adjacent cardiac myocytes: they allow for the passage of various cations, anions and small, non-charged molecules from one cell to another. Though gap junctions are not directly affected by membrane potential, they can respond to differences in transcellular voltage by altering their conductance⁸⁴.

As determined by immunocytochemistry, among the channel proteins expressed in cardiac gap junctional plaques are three principal connexins: connexin (Cx)40, Cx43 and Cx45⁸⁵. Two other connexins – Cx37 and Cx46 – have also been identified in the heart, albeit in trace amounts⁸⁶. Different ratios of these connexins give rise to multiple gap junction phenotypes. These different phenotypes appear to have a deliberate and functionally important distribution pattern throughout the heart. Electron microscopy has shown SA nodal myocytes to have small, scattered gap junctions⁸⁷. This finding has implications regarding conduction velocity in this region, which is very slow⁸⁸. The number and distribution of gap junctions, as well as their composition, are important factors in modulating conduction propagation velocity. In a study of mice heterozygous for a null mutation in the Cx43 gene (Cx43 +/- mice), it was determined that this mutation was responsible for a reduction in the total number of gap junctions while the remaining gap junctions were unaffected in terms of size⁸⁹. These mice have a ~40% slower ventricular epicardial conduction velocity⁹⁰.

Cx43 is the most abundant gap junctional channel in human⁹¹ and rat⁸⁹ ventricles and atria. The expression level of Cx43 in the four chambers of the human heart is more or less uniform⁹¹. Studies have shown that the human SA node has little⁹² or no Cx43⁸⁶. Western blotting of the AV node reveals a weak signal found primarily in the middle of the region⁸⁶, consistent with the important slowly-conducting properties of the AV node. When Purkinje fibres are probed with an antibody for Cx43, the signal is more intense at longitudinal ends than transversely⁸⁶. This distribution pattern is likely important in the known functionally-important anisotropy of conduction.

Cx40 is expressed in human atrium⁹¹ and to a much lesser degree in ventricular subendocardium⁸⁶ A study of Cx40 expression in dog also showed that there was a much higher level of this channel in the crista terminalis compared with left ventricular subepicardium⁹³. In both man⁸⁶ and dog⁸⁵, Purkinje fibres show a far more robust signal for Cx40 than ventricular myocardium. Canine⁹² and human⁸⁶ SA nodal cells clearly express Cx40. Human AV nodal cells also express Cx40⁸⁶.

Low levels of Cx45 can be detected in human ventricles, with more being seen in the atria⁹¹. Both rat⁹⁴ and human⁸⁶ conduction systems (including the SA node, AV node and Purkinje fibres) have an abundant expression of Cx45. In a recent study of Cx45 knockout mice, it was found that these mice die of heart failure at embryonic day 10. Though the hearts of these mice do contract, conduction block through the AV node region is observed⁹⁵.

The pacemaker current I_f

A slow membrane depolarization phase between action potentials is responsible for the rhythmic activity of the heart^{96, 97}. The principal current underlying this phenomenon is referred to as I_f or I_h. Both Na⁺ and K⁺ ions carry this current with the selectivity being fourfold higher for K⁺.

I_f is activated on membrane hyperpolarization. The hyperpolarization-activated cyclic nucleotide gated cation (HCN) family of channels have many of the characteristics of I_f channels. Four members (HCN1-4) have been isolated from mouse, rabbit and human tissues^98-101, and share a homology of ~60% on the amino acid level¹⁰¹. Northern blot showed that only HCN2 and HCN4 are expressed in the human heart. HCN2 currents activate faster than HCN4. Two different native currents with distinct kinetics have been identified in the heart^102-104. Semi quantitative RT-PCR showed that human HCN2 and HCN4 mRNA is found to be approximately equally abundant throughout the atria and ventricles¹⁰⁵. Northern blotting revealed that in rabbit, HCN4 mRNA is more abundant in SA nodal cells than in ventricle or atrium¹⁰⁰. SA nodal total HCN message in general is 140 times the HCN message for the ventricle and 25 times that of Purkinje fibres¹⁰⁶, corresponding to the primary pacemaking role of the SA node and the greater pacemaking activity in Purkinje fibres compared to working muscle.

The sodium current, I_Na

I_Na is responsible for the rising phase (phase 0 upstroke) of action potentials in electrically excitable cells^{78, 107, 108} and for rapid impulse conduction through cardiac tissue¹⁰⁷.

The functional channel consists of a principal α pore-forming subunit composed of four homologous domains ( I- IV) of six transmembrane segments S1-S6¹⁰⁷ and 2 distinct β-subunits involved in the modulation of channel gating and cell surface expression. Ten genes encoding Na⁺ channel α-subunits have been described (Na_V1.1 to Na_V1.9, and Na_X)108 and 3 β-subunits (β1 to β3) have been identified to date^108-110. Most of the subunits in the Na_V1.x family have been studied in heterologous expression systems and, therefore, the most detailed observations are for this family. At least 2 α- subunit mRNAs are expressed in human heart: Na_V1.5 and Na_V2.1^108,111,112. β1- Subunit mRNA is expressed in rat and human heart^113,114 but not found in mouse heart¹¹⁵. β1 and β2 subunits have been localized in rat and mouse heart by immunofluorescence¹¹¹. Both proteins are expressed in cardiac muscle along the Z lines¹¹¹. The differences in the functional properties of Na⁺ channel isoforms result in unique conductances in specific cell types¹⁰⁸. Immunocytochemical studies have shown that β1, β2, Na_V1.1 and Na_V1.5 localize along Z lines in adult rat and mouse cardiac myocytes^111,116. Na_V1.5 has been localized to the surface and T-tubular system of rat hearts and does not display a large variation other than a somewhat enhanced labelling at the intercalated disks of ventricular myocytes¹¹⁶. It is possible that this enhanced localization is related to fast conduction in ventricular myocardium. In the compact rabbit AV node, a gradient of Na⁺ channel expression exists, with peripherally-located cells having stronger signals compared with centrally located ones¹¹⁷. This gradient is likely important in the slow conduction through this region.

Functional importance

The cardiac action potential consists of a fine interplay between a large number of inward and outward currents of which the main ones are carried by K⁺, Na⁺ or Ca²⁺ ions. Characteristic action potential shapes have been found in different regions of the heart as well as in different layers of the ventricular wall^26,118. Clones encoding a large number of channel proteins have been identified, some of which clearly reconstitute native channel properties^{35, 34, 94, 119, 120}. However, many discrepancies between native currents and putative underlying clones remain to be resolved.

Important regional differences in I_K1¹ are not readily explained by the mRNA distribution of corresponding K_ir2 family subunits².

Future studies using techniques like Western blotting, immunocytochemistry and immunohistochemistry may help to determine K_ir2 protein expression levels, tissue expression pattern and cellular localization of K_ir2 subunits giving further insight into the molecular basis of I_K1. Transmural gradients in I_to are likely important in governing action potential shape and explaining many important ECG phenomena^{6, 118, 120-123}. M-cells have longer action potentials than in other ventricular regions, playing an important role in arrhythmias due to excessive prolongation of repolarization (long QT syndrome). Transmural gradients in I_Ks¹⁷, likely related to differential distribution of K_vLQT1 isoforms¹²⁴, is probably responsible. I_to downregulation may be important in ventricular arrhythmogenesis in heart failure and appears to parallel a decrease in K_v4.3 mRNA¹²⁵, although the nature of any regional or transmural differences is not well-known. Changes in connexin expression may also contribute to arrhythmogenesis. For example, Peters et al. observed a reduction in ventricular Cx43 signal from hypertrophied and infarcted human hearts¹²⁶.

The above are only limited examples of the importance of regional variations in ion channel expression in physiological and pathological states. In addition to differential expression of pore- forming subunits, the formation of functionally distinct heterotetramers and the presence of regulatory subunits likely contribute importantly to regional diversity in channel function, and relatively little is known about them. Much more work clearly remains to be done in this important area.

References

1. Giles, W.R. and Imaizumi Y. (1988) J. Physiol. 405, 123.
2. Wang, Z. et al. (1998) Circulation. 98, 2422.
3. Varro, A. et al. (1993) Acta Physiol Scand. 149, 133.
4. Furukawa, T. et al. (1992) Circ. Res. 70, 91.
5. Main, M.C. et al. (1998) Exp. Physiol. 83, 747.
6. Liu, D.W. et al. (1993) Circ. Res. 72, 671.
7. Raab-Graham, K.F. et al. (1994) Neuroreport 5, 2501.
8. Wible, B.A. et al. (1995) Circ. Res. 76, 343.
9. Schram, G., Wang, Z., Nattel, S. (1999) Circulation 100, I633.
10. Derst, C. et al. (2001) FEBS Lett. 491, 305.
11. Sanguinetti, M.C. and Jurkiewicz, N.K. (1991) Am J Physiol. 260, H393.
12. Li, D., Zhang, L., Kneller, J., Nattel, S. (2001) Circulation, in press.
13. Mandapati, R. et al. (2000) J. Circulation 101, 194.
14. Pond, A.L. et al. (2000) J Biol Chem 275, 5997.
15. Wymore, R.S. et al. (1997) Circ Res 80, 261.
16. Bryant, S.M. et al. (1998) Cardiovasc Res 40, 322.
17. Liu, D.W. and Antzelevitch, C. (1995) Circ Res 76, 351.
18. Brahmajothi, M.V. et al. (1996) Circ Res 78, 1083.
19. Furukawa, T. et al. (1990) Circ. Res. 67, 1287.
20. Litovsky, S.H. and Antzelevitch, C. (1988) Circ Res 62, 116.
21. Clark, R.B. et al. (1993) Cardiovasc Res 27, 1795.
22. Wickenden, A.D. et al. (1999) Am J Physiol 276, H1599.
23. Amos, G.J. et al. (1996) J Physiol 491, 31.
24. Wettwer, E. et al. (1994) Circ Res 75, 473.
25. Guo, W. et al. (2000) Circ Res 87, 73.
26. Barry, D.M. and Nerbonne, J.M. (1996) Annu Rev Physiol 58, 363.
27. Brahmajothi, M.V. et al. (1999) J Gen Physiol 113, 581.
28. Nabauer, M. et al. (1996) Circulation 93, 168.
29. Nerbonne, J.M. (2000) J Physiol 525, 285.
30. Wei, J. et al. (1999) Circulation 99, 3165.
31. Xu, H. et al. (1999) J Gen Physiol 113, 661.
32. Bou-Abboud, E. and Nerbonne, J.M. (1999) J Physiol 517, 407.
33. Fiset, C. et al. (1997) J Physiol (Lond.) 500, 51.
34. Dixon, J.E. and McKinnon, D. (1994) Circ Res 75, 252.
35. Dixon, J.E. et al. (1996) Circ Res 79, 659.
36. Takeuchi, S. et al. (2000) J Mol Cell Cardiol 32, 1361.
37. Guo, W. et al. (1999) J Physiol 521, 587.
38. Tseng-Crank, J.C. et al. (1990) FEBS Lett 268, 63.
39. Petersen, K.R. and Nerbonne, J.M. (1999) Pflugers Arch 437, 381.
40. Wang, Z. et al. (1999) Circ Res 84, 551.
41. Han, W. et al. (2000) Am J Physiol Heart Circ Physiol 279, H466.
42. Philipson, K.D. and Nicoll, D.A. (2000) Annu Rev Physiol 62, 111.
43. Kimura, J. et al. (1986) Nature 319, 596.
44. Reuter, H. and Seitz, N.J. (1968) Physiol 195, 451.
45. Dipla, K. et al. (1999) Circ Res 84, 435.
46. Flesch, M. et al. (1996) Circulation 94, 992.
47. Reinecke, H. et al. (1996) Cardiovasc Res 31, 48.
48. Studer, R. et al. (1997) Basic Res Cardiol 92, 53.
49. Nicoll, D.A. et al. (1990) Science 250, 562.
50. Nicoll, D.A. et al. (1996) Ann N Y Acad Sci 779, 86.
51. Li, Z. et al. (1994) J Biol Chem 269, 17434.
52. Quednau, B.D. et al. (1997) J Physiol 272, C1250.
53. Ruknudin, A. et al. (2000). J Physiol 529, 599.
54. Dyck, C. et al. (1999) J Gen Physiol 114, 701.
55. Kofuji, P. et al. (1993) Am J Physiol 265, F598.
56. Nakasaki, Y. et al. (1993) J Biochem (Tokyo) 114, 528.
57. Ruknudin, A. et al. (1997) Am J Physiol 273, C257.
58. Komuro, I. et al. (1992) Proc Natl Acad Sci USA 89, 4769.
59. Blaustein, M.P. and Lederer, W.J. (1999) Physiol Rev 79, 763.
60. Dilly, K. et al. (1997) Biophys. J. 72, A66.
61. Kieval, R.S. et al. (1992) Am J Physiol 263, C545.
62. Frank, J.S. et al. (1992) J Cell Biol 117, 337.
63. Fabiato, A. and Fabiato, F. (1979) Annu Rev Physiol 41, 473.
64. Nargeot, J. et al. (1997) Eur Heart J 18, A15.
65. Martinez, M.L. and Heredia, M.P. (1999) J Mol Cell Cardiol 31, 1617.
66. Bean, B.P. (1989) Annu Rev Physiol 51, 367.
67. Gathercole, D.V. et al. (2000) J Mol Cell Cardiol 32, 1981.
68. Hess, P. (1990) Annu Rev Neurosci 13, 337.
69. Tsien, R.W. et al. (1991) Trends Pharmacol Sci 12, 349.
70. Cribbs, L.L. et al. (1998) Circ Res 83, 103.
71. Hagiwara, N. et al. (1988) J Physiol 395, 233.
72. Wetzel, G.T. et al. (1993) Circ Res 72, 1065.
73. Ouadid, H. et al. (1991) J Mol Cell Cardiol 23, 41.
74. Hofmann F, Biel M, Bosse E, Flockerzi V, Ruth P, Welling A. In: Spooner P, Brown AM, Catterall WA, Kaczorowski G, Strauss HC, editors. Ion Channels in the Cardiovascular System. Function and dysfunction. Futura Publishing Company, Inc. Armonk, New York, 1994: 369-381.
75. De Waard, M. et al. (1996) Ion Channels 4, 41.
76. Perez-Reyes, E. et al. (1992) J Biol Chem 267, 1792.
77. Walker, D. and De Waard, M. (1998) Trends Neurosci 21, 148.
78. Catterall, W.A. (2000) Annu Rev Cell Dev Biol 16, 521.
79. Glossmann, H. and Striessnig, J. (1988) Vitam Horm 44, 155.
80. Hofmann, F. et al. (1990) Curr Top Cell Regul 31, 223.
81. Takagishi, Y. et al. (2000) Am J Physiol Cell Physiol 279, C1963.
82. Perez-Reyes E. J. (1998) Bioenerg Biomembr. 30, 313.
83. Perez-Reyes, E. et al. (1998) Nature 391, 896.
84. Spray, D. et al. In: Spooner P, Brown A, Catterall WKG, Strauss H, editors. Ion channels in the cardiovascularsystem. Futura Publishing Company, Inc., Armonk, New York, 1994: 185.
85. Kanter, H.L. et al. (1993) Circ Res. 72, 1124.
86. Davis, L.M. et al. (1995) J Cardiovasc Electrophysiol. 6, 813.
87. Saffitz, J.E. et al. (1997) J Cardiovasc Electrophysiol. 8, 738.
88. Davis, L.M. et al. (1995) J Cardiovasc Electrophysiol. 6, 103.
89. Saffitz, J.E. et al. (2000) Am J Physiol Heart Circ Physiol. 278, H1662.
90. Guerrero, P.A. et al. (1997) J Clin Invest. 99, 1991.
91. Vozzi, C. et al. (1999) J Mol Cell Cardiol. 31, 991.
92. Kwong, K.F. et al. (1998) Circ Res. 82, 604.
93. Saffitz, J.E. et al. (1994) Circ Res. 74, 1065.
94. Coppen, S.R. et al. (1998) Circ Res. 82, 232.
95. Kumai, M. et al. (2000) Development 127, 3501.
96. DiFrancesco, D. (1993) Annu Rev Physiol. 55, 455.
97. Irisawa, H. et al. (1993) Physiol Rev. 73, 197.
98. Ludwig, A. et al. (1998) Nature 393, 587.
99. Santoro, B. et al. (1998) Cell 93, 717.
100. Ishii, T.M. et al. (1999) J Biol Chem. 274, 12835.
101. Ludwig, A. et al. (1999) EMBO J. 18, 2323.
102. DiFrancesco, D. (1986) J Physiol. 377, 61.
103. Maruoka, F. et al. (1994) J Physiol. 477, 423.
104. Liu, Z.W. et al. (1996). J Mol Cell Cardiol. 28, 2523.
105. Ludwig, A. et al. (1999) Cell Physiol Biochem 9, 179.
106. Shi, W. et al. (1999) Circ Res. 85, E1.
107. Balser, J.R. (1999) Cardiovasc Res. 42, 327.
108. Goldin, A. (2001) Annu Rev Physiol. 63, 871.
109. Kazen-Gillespie, K.A. et al. (2000) J Biol Chem 275, 1079.
110. Morgan, K. et al. (2000) Proc Natl Acad Sci USA 97, 2308.
111. Malhotra, J.D. et al. (2001) Circulation 103, 1303.
112. Rogart, R.B. et al. (1989) Proc Natl Acad Sci USA. 86, 8170.
113. Isom, L.L. et al. (1992) Science 256, 839.
114. Makita, N. et al. (1994) J Biol Chem. 269, 7571.
115. Grosson, C.L. et al. (1996) Brain Res Mol Brain Res. 42, 222.
116. Cohen, SA. (1996) Circulation 94, 3083.
117. Petrecca, K. et al. (1997) J Physiol. 501, 263.
118. Antzelevitch, C. et al. (1991) Circ Res. 69, 1427.
119. Barhanin, J. et al. (1996) Nature 384, 78.
120. Coppen, S.R. et al. (1999) Dev Genet 24, 82.
121. Fedida, D. and Giles, W.R. (1991) J Physiol. 442, 191.
122. Kimura, S. et al. (1990) Circ Res 66, 469.
123. Tseng, G.N. and Hoffman, B.F. (1989) Circ Res. 64, 633.
124. Pereon, Y. et al. (2000) Am J Physiol Heart Circ Physiol. 278, H1908.
125. Kaab, S. et al. (1998) Circulation 98, 1383.
126. Peters, N.S. et al. (1993) Circulation 88, 864.