Action Potential Characteristics

The human atrial action potential typically demonstrates a triangular morphology (compared with a spike-and-dome shape with a prominent plateau phase of the ventricular counterpart).1,27 The human atrial action potential duration at 90% repolarization (APD₉₀) at 1 Hz shows large variations of between 150 and 500 msec, possibly influenced by recording conditions and ionic concentrations.^2,4,27–31 The atrial resting membrane potential (V_rest) has been found to vary between −65 and −80 mV,^2,4,27–31 and is more depolarized than in the human ventricle^26,27 (also see Figure 31-1). The depolarized V_rest in atrial cells compared with ventricular cells is mainly attributed to differences in density of the inward rectifier K⁺ current, I_K1^17,32,33 (also see later). Maximum upstroke velocities (V_max) for atrial action potentials have been experimentally reported to vary between ≈150 and 300 V/s,^2,27–31 in contrast to higher values of 300 to 400 V/s for human ventricular cells.^26,33 Characteristic human atrioventricular action potential property/shape differences are replicated in most mammalian species,¹⁹ except for the murine myocardium (rat/mouse), where both atrial and ventricular cells display short, triangular action potentials.^34,35

The Inward, Depolarizing Currents: (Na⁺, Ca²⁺)

Initiation of the atrial action potential is due to rapid activation of the inward Na⁺ current, I_Na, which is also the principal determinant of V_max. Functional differences in human ventricular and atrial I_Na biophysical properties were reported to be minimal,36–39 but recent studies have shown that the molecular correlates of I_Na in the human atrial and ventricular myocardium are different. Although Nav1.5 was found to be the main α-subunit encoding for atrial/ventricular I_Na, the transcript for the β-subunit, Navβ1, was more prominently expressed in the human atrium.⁴⁰ In addition, recent reports in canine atrial cells suggest that atrial I_Na has higher current density and a more negative steady-state half-inactivation voltage value, when compared with ventricular cells.⁴¹ Similar differences were reported in guinea pig atrial and ventricular cells.⁴² Further, because of the more depolarized V_rest, the availability of atrial I_Na is less, which results in a smaller I_Na underlying the atrial action potential, compared with the ventricular counterpart, as can be seen in Figure 31-1. The L-type Ca²⁺ current (I_CaL) is mainly responsible for the plateau phase of the atrial action potential, as well as the Ca²⁺-induced Ca²⁺ release in human atrial cells.⁴³ Patch-clamp experiments suggested that the T-type Ca²⁺ current (I_CaT) was not present in human atrial cells.⁴³ Experimental studies conducted in the Lederer lab reported differences in the biophysical properties of I_CaL between human atrial and ventricular cells in current density and steady-state inactivation properties.⁴⁴ At the transcript level, greater expression of the α-subunits Cav1.3 and Cav3.1 has been reported in the atrium, compared with the ventricle.⁴⁰ Moreover, a more “negative voltage” of the triangular plateau phase causes a larger driving force (V_m − E_Ca; where V_m represents membrane voltage, and E_Ca represents reversal potential for I_CaL), resulting in an I_CaL current with a larger magnitude underlying the human atrial cell action potential (compared with its ventricular counterpart), as can be seen in Figure 31-1.¹⁷ Differences in regulation of human atrial (but not ventricular) I_CaL by second messengers such as serotonin⁴⁵ and phosphodiesterases⁴⁶ (PDEs) have been reported and may further influence action potential morphology.⁴⁷

The Outward, Repolarizing K⁺ Currents (K⁺)

The key current that sets the V_rest (i.e., the inward rectifier K⁺ current I_K1) has a much smaller density (≈5-6-fold less) in human atrial cells compared with its ventricular counterpart,17,32 and this partially explains the depolarized V_rest in atrial as opposed to ventricular cells (see Figure 31-1). The smaller density of I_K1 is also responsible in part for the slower late phase of repolarization in the atrium. The main molecular correlate of I_K1, Kir2.1, is more robustly expressed in the human ventricle than in the atrium.⁴⁰ The human atrium also expresses the fast, depolarization-activated, 4-AP (aminopyridine)-sensitive, Ca²⁺-independent transient outward K⁺ current, I_to1.^4,27 This current is responsible for the initial repolarization phase of the atrial action potential^{4,17,27,48,49} (see Figure 31-1). It is interesting to note that despite having almost a 2-fold higher density in the atrium than in the ventricle,²⁷ the magnitude of I_to1 underlying the atrial and ventricular action potentials is similar, likely as the result of action potential morphologies (see Figure 31-1). The main molecular correlates of human atrial I_to1 are Kv4.3/KChiP2 (α/β-subunits, respectively).⁴⁰ The slow component of I_to1,slow (mainly encoded by Kv1.4) is absent in the human atrium, but present in the ventricle.^17,26 A Ca²⁺-dependent transient outward K⁺ current, I_to2, has been reported in human atrial cells, but both its molecular correlate and its contribution to the action potential remain unclear.⁴⁸ Human atrial cells also express three functional delayed rectifier K⁺ currents, which contribute to repolarization: the ultra-rapid delayed rectifier K⁺ current, I_Kur (molecular correlate Kv1.5)^6,50 and the rapid and slow delayed rectifier K⁺ currents, viz, I_Kr, and I_Ks (molecular correlates HERG, KvLQT1).^7,51,52 I_Kur is present only in the atrium (not ventricle), is active during the plateau phase (see Figure 31-1), and is blocked by low concentrations of 4-AP (100 µM).^5,6,50 Its atrial-selective nature has made it an attractive target for many antiarrhythmic drugs in development for terminating and/or preventing recurrence of AF.⁵³ Compared with I_Kur, the contribution of I_Kr and I_Ks to the human atrial action potential is small (see Figure 31-1),¹⁷ in part because of their small density, and in part because of their triangular shape and plateau phase at relatively negative membrane voltages, which preclude both I_Kr and I_Ks from activating fully. However, considerable variability in the shape of the human atrial action potentials has been noted, and I_Kr/I_Ks are likely to contribute more in cells showing a prominent plateau phase and dome (the so-called type 1 action potentials).⁷

[Ca²⁺]_i, Electrogenic Pumps and Exchangers

The human atrial action potential is also modulated by [Ca²⁺]_i, which directly influences the inactivation of I_CaL⁵⁴ and the magnitude and temporal profile of the electrogenic Na⁺/Ca²⁺ exchanger current, I_NCX^10,55 (molecular transcript NCX1⁴⁰), and has an indirect influence on the Na⁺/K⁺ pump current, I_NaK, by influencing intracellular Na⁺ ion accumulation¹⁷ (molecular transcripts Na/K ATPase α1, α3, β1⁴⁰). The Na⁺/Ca²⁺ exchanger current (I_NCX) is the main Ca²⁺ extrusion and Na⁺ influx pathway in cardiac myocytes. It extrudes 1 Ca⁺ in exchange for 3 Na⁺, thus generating an inward current that influences cardiac repolarization and arrhythmogenesis.⁵⁶ The Na⁺/K⁺ pump (NKA) is the main route of Na⁺ efflux in cardiac cells, thus regulating intracellular [Na⁺]. By extruding 3 Na⁺ in exchange for 2 K⁺, it generates an outward current that is known to influence both resting membrane potential and repolarization.⁵⁶ We recently simulated the [Ca²⁺]_i homeostasis in human atrial cells, and our results showed that whereas I_NaK is primarily a repolarizing outward current, I_NCX can be both outward and inward; it contributes to repolarization in the early phase of the action potential, and is a negative current (depolarizing influence) during the later phase of the action potential, when it extrudes Ca²⁺ ions from the atrial cells (see Figure 31-1).¹⁷ I_NaK and I_NCX also influence the frequency dependence of the human atrial action potential duration, which is discussed in later sections.

Other Ion Channels

Besides the currents discussed earlier, other ion channels are activated under specific conditions and can influence the human atrial action potential. These include the acetylcholine-activated K⁺ current, I_KACh28,57,58 (molecular correlates Kir3.1/3.4⁴⁰), the adenosine triphosphate (ATP)-sensitive K⁺ current, I_KATP^59,60 (molecular correlate, Kir6.1/6.2/SuR2.X⁴⁰), the Ca²⁺-dependent nonselective cation current^61,62 (molecular correlates unknown), and the hyperpolarization-activated funny current, I_f^63,78,93 (molecular correlates HCN1/HCN4⁴⁰). Some reports indicate that a Ca²⁺-activated K⁺ current I_KCa (putative molecular correlate SK2⁴⁰) may modulate the human atrial action potential^64,65; however, its presence and its contribution to repolarization remain controversial, and species-specific variations have been suggested.⁶⁶ Stretch-activated ion channels (I_SAC), which have been reported in human atrial cells,⁶⁷ influence both V_rest and APD; however, their molecular correlates remain unknown. Swelling-induced, outwardly rectifying chloride channels, such as I_Cl⁻⁶⁸ (putative molecular correlates Cl3,Cl6,Cl7⁴⁰), have been reported in human atrial cells. The transcripts of two-pore channels (TWIK1, TASK1) have also been reported to be present in the human atrium.⁴⁰ However, no information is available on the contribution of these currents or that of I_Cl⁻ to the APD.

Ionic Bases of Regional Heterogeneity in Atrial Action Potentials

The atrium is a complicated three-dimensional (3D) structure, and the heterogeneities in electrical properties between different regions, such as the left atrium (LA), the right atrium (RA), and the pulmonary veins (PV) and their ionic basis have been extensively studied in many species, including mouse,69,70 rabbit,^71,72 and dog.^73,74 Information regarding such heterogeneity in humans is limited, but recent studies have begun to address the putative ionic/molecular basis for these differences. The frequencies of excitation in AF have been reported to show regional gradients during paroxysmal AF and postoperative AF, with LA-PV regions displaying the highest frequencies.^75–77 This is highly indicative of underlying differences in biophysical properties of some ionic currents, and these variations, which are more prominent in disease conditions, are discussed later. In patients in normal sinus rhythm, Caballero and colleagues reported a higher density of I_Kur in the right than in the left atrial cardiomyocytes.⁵² Incorporating such data in human atrial mathematical models did not result in appreciable differences between the LA/RA action potentials in normal sinus rhythm.¹⁷ However, large variations in human atrial action potential shapes (mainly triangular, some dome shaped) have been reported in cells isolated from the right atrial appendage, and this has been attributed to differences in current density ratios of I_to1/I_Kr in these cells.⁷ Recently, the carbachol-activated K⁺ current (I_KACh) was reported to be 70% larger in RA than in LA, in sinus rhythm patients, with correspondingly higher protein expressions of its molecular correlates (Kir3.1/Kir3.4) in RA versus LA.⁷⁹ To the best of our knowledge, no report has described the ionic mechanisms underlying human PV cells. Thus available data regarding the regional variations in human atrial electrophysiology/underlying ionic mechanisms under healthy conditions are very limited, and further studies are needed in this regard. In the next paragraph, we briefly review the atrial action potential heterogeneity/ionic mechanisms in other animal species.

In healthy hearts of dogs and mice, the APD is shorter in LA than in RA, primarily on account of differences in current densities of the delayed rectifier K⁺ currents, that is, I_Kr in dogs and I_Kur and the steady-state K⁺ current I_ss in mice display larger current densities in LA than in RA.70,80,81 In sheep atrial cells, the I_KACh current has been shown to have a larger density in LA than in RA, whereas the density of I_K1 was similar.⁸² In contrast, I_KACh current density was larger in RA than in LA in mouse,⁸³ similar to humans in sinus rhythm.⁷⁹ The density of I_Na has been reported to be larger in LA than in RA in dogs.⁸⁴ Besides LA and RA, differences within RA have also been reported. For example, the action potential properties of the pectinate muscles and the crista terminalis in the rabbit RA are different⁸⁵; such differences have been attributed to differences in various ion channels (Na⁺, Ca²⁺, K⁺) and have been integrated into a mathematical model.⁸⁶ Furthermore, differences in the electrophysiological properties of cells isolated from the LA and PV regions have been extensively studied by the Nattel group in dogs^73,74 and by the Chen group in rabbits.^71,72 Canine PV cells exhibit a depolarized V_rest, a smaller V_max, and shorter APD compared with LA cells.⁷³ This has been attributed to a greater density of I_Kr, I_Ks, and a smaller density of I_CaL, I_to1 in PV than in LA cells, whereas the densities of I_NCX and I_CaT were found to be similar between PV and LA cells.⁷³ It is interesting to note that Ca²⁺-handling properties were not different between canine PV and LA cells.⁸⁷ In rabbit PV cells, most cells demonstrated pacemaker activity.⁷² As reported in the ventricle, epi-endocardial differences also exist within the atrium. A recent study in pigs reported a shorter refractory period in the atrial epicardium compared with the endocardium.⁸⁸ This is in line with a shorter atrial epicardial APD recorded in the canine right atrial free wall compared with the endocardium.⁸⁹ The canine atrial endocardium was also more sensitive to APD shortening after the addition of acetylcholine.⁸⁹ However, the ionic mechanisms underlying atrial epicardium-endocardium differences remain unexplored.

Ionic Bases of Atrial Action Potential Variations With Age

Age-related changes in atrial action potential differences in humans have been most systematically investigated between infants (neonatal) and adults.29,30 Adult cells displayed a more prominent initial notch compared with neonatal atrial cells.^29,30 The properties of I_to1 were significantly different as well; I_to1 displayed significantly higher density, faster inactivation, and a slower recovery from inactivation in neonatal compared with adult human atrial cells.²⁹ Correspondingly, the protein density of Kv4.3 was higher and KChiP2 was lower in neonatal compared with adult cells.²⁹ The properties of I_CaL were also different: The basal density was smaller and the protein density of Giα3 was larger in neonatal compared with adult cells, resulting in a different response to a lower dosage of β-adrenergic stimulation.⁹⁰ In addition, Ca²⁺ transients, which influence both adult and neonatal human atrial action potentials,^14,17,91 may display inherently different biophysical properties but remain unexplored. It is well known that the propensity for AF increases with age,⁹² but very few studies have been conducted to explore potential differences between adult and aged human atrial electrophysiology, and most of the knowledge in this area has been derived from animal models (rat, rabbit, dog). This topic was reviewed comprehensively in a recent article. We briefly describe the salient points.²⁰ V_rest was found to be more depolarized in rats/dogs,^94,95 whereas APD₉₀ was more prolonged in rats/rabbits/dogs in aged compared with adult atria.^20,96,97 The density of I_Na was not different,⁸⁴ but the density of I_CaL was reported to be smaller in canine atrial cells.⁹⁸ Further, the density of I_to1 was higher, its steady-state half-inactivation value showed more positive values, and recovery from inactivation was slower in canine aged atrial cells compared with their adult counterparts.⁹⁸ The density of the sustained outward K⁺ current, I_sus, was also reported to be larger in aged canine atrial cells.⁹⁸ In addition to the changes in ionic currents and action potentials, an increased level of interstitial fibrosis and modifications in intercellular coupling have been reported in aged compared with adult tissue,^99,100 and this may further enhance the susceptibility to AF with age. It is likely that Ca²⁺ handling is different between aged and adult atrial tissue, but this issue remains unexplored, especially in humans.

Effects of AF on AP and [Ca²⁺]_i

Atrial cells from patients in chronic AF (cAF) display shorter APD compared with healthy patients in normal sinus rhythm (Figure 31-2, A).^28,107–110 Furthermore, the normal human atrial APD₉₀ shortens when paced at progressively faster frequencies, but in cAF, this shortening is severely attenuated.¹⁷ The peak amplitude of [Ca²⁺]_i is reduced in atrial myocytes from cAF patients compared with those of healthy patients (Figure 31-2, B), although the SR Ca²⁺ content is unaltered.¹⁷ [Ca²⁺]_i decays more slowly in cAF compared with sinus rhythm.^17,112 Our recently published mathematical model of the human atrial cell provided novel insights into the ionic mechanisms underlying the altered APD/[Ca²⁺]_i in cAF (see Figure 31-2, A, B; black: sinus rhythm, red: cAF); the salient points are discussed here.

Ionic Remodeling in cAF

Sarcolemmal Ion Channels

I_Na

Bosch et al reported that the density of I_Na and the voltage dependence of steady-state activation were not altered in cAF in humans, whereas steady-state inactivation was shifted to the right by ≈10 mV,¹¹⁰ and no changes were detected in mRNA levels of the Na⁺ channel gene SCN5A.¹¹⁴ In contrast, data from Sossalla et al show that both expression of Nav1.5 and peak I_Na density are decreased (slightly) in the atrial myocardium of patients with cAF.¹¹⁵ Additionally, the late Na⁺ current component, I_NaL (see inset, Figure 31-2), was reported to be significantly increased in cAF patients.¹¹⁵ Sossalla et al proposed that this increase could be due to the increase in neuronal Na⁺ channel isoforms (Nav1.1 expression is increased),¹¹⁵ or it could be mediated by CaMKII, which is increased in AF^111,116 and is known to regulate I_NaL,¹¹⁷ or to be caused by oxidative stress.^118,119

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Exercise-Induced Arrhythmias
2. Calcium Signaling in Cardiomyocyte Models With Realistic Geometries
3. Ventricular Arrhythmias in Congenital Heart Disease
4. Implantable Cardioverter Defibrillator: Clinical Aspects

Stay updated, free articles. Join our Telegram channel

Join