Pharmacological electrical remodelling in human atria induced by chronic β-blockade

Time for primary review 3 days. See article by Workman et al. [5] (pages 518–525) in this issue. Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia found in clinical practice. AF is a supraventricular tachyarrhythmia characterised by uncoordinated atrial activation with consequent deterioration of atrial mechanical function [1]. The ventricular response to AF depends on the electrophysiological properties of the atrio-ventricular (AV) node, the level of vagal or sympathetic tone, and the action of drugs with which the patient is being treated [2]. Very often, AF occurs in conjunction with other cardiovascular diseases, such as hypertension, ischaemic heart disease, valve disease, or cardiac failure. However, in a certain percentage of patients (20–50%) AF is not associated with any underlying pathology [3]. Treatment with β-blockers is considered to be efficacious for controlling the ventricular heart rate during AF, which is likely due to the ability of these drugs to depress the conduction velocity through the AV node. Moreover, in patients with adrenergically mediated AF, β-blockers represent a first-line treatment [4]. However, the underlying mechanism of action of β-blockers in human atrial fibrillation has not been established. In fact, the mechanisms by which most cardioactive drugs are effective in clinical practice are unknown, despite the knowledge of their effects on ionic currents, receptors, signal transduction, etc. Indeed, there is an important gap of knowledge that involves drug effects on the target cells; atrial myocytes from patients with AF, in this case. In this regard, different ethical and technical problems have limited these studies. In the present issue of Cardiovascular Research, Workman et al. [5] present experimental evidence of an electrical remodelling in human atrial cells from patients chronically treated with β-blockers, the so-called ‘pharmacological remodelling’, consistent with a partial reversion of the electrical AF remodelling. This study represents the first one dealing with the effects of chronic β-blockade in AF patients at the cellular level. An abrupt increase in heart frequency, like that observed in AF, induces an immediate (within one action potential) and then a gradual (that arises steady state after several minutes) decrease in the action potential duration (APD) [6]. This shortening of the APD leads to a parallel decrease in the atrial effective refractory period (ERP), shortening the wavelength for re-entry and thus facilitating the occurrence and maintenance of re-entrant arrhythmias like AF. This coincides with functional changes in the L-type Ca2+ channel following calcium overload [7–9]. During human AF an electrical remodelling has been described [10,11] that involves changes in several ionic channels and leading to the shortening of the atrial ERP. In fact, patients with paroxysmal and persistent AF show a marked reduction in L-type Ca2+ channel expression [12]. The decrease in L-type Ca2+ current occurs much faster in experimental animal models than in human beings suffering from AF, indicating the existence of other (likely protective) adaptation mechanisms in human AF [13]. Since the observed shortening of the APD can also be explained by an increased K+ conductance, there have been a number of studies performed that have focused on the analysis of a possible electrical remodelling during human AF involving different K+ channels [14–17]. A result in contradiction with a shortening of the APD observed during the AF is the finding that both mRNA and protein levels of K+ channels are reduced in human atrial cells with persistent AF. Remodelling is likely composed of a series of responses to maintain cell integrity in the face of tachycardia-induced Ca2+ overload, starting immediately upon tachycardia onset and developing in different time domains as long as the arrhythmia persists [1]. An acute increase in rate is associated with a significant increase in intracellular Ca2+ concentration that persists for up to 50 min after the onset of rapid stimulation [18]. This increase in [Ca2+]i decreases ICa by a Ca2+-dependent inactivation [19,20], thus preventing Ca2+ overload. With regard to K+ channels, it has been described that in cells from AF patients ITO is decreased in a parallel manner with a decreased transcription of Kv4.3, but not of Kv1.4 subunits. Unfortunately, there is very limited information available concerning the effects of AF on IKur, IKr, and IKs. It has been reported that IK1 and the acetylcholine-activated inward-rectifier IKACh are increased in patients with AF [21]. The outward component of these currents is important in the late phase of repolarisation and, thus, a decrease in IK1 and IKACh could contribute to the observed shortening of the action potential. In the present issue of this journal, Workman et al. [5] demonstrate that chronic β-blockade of patients with AF leads to an adaptive response concomitant with atrial electrophysiological changes. At the action potential level, the authors observed a lengthening of both APD and ERP in atrial cells from patients chronically treated with β-blockers in comparison with that observed in cells from untreated patients. They also observed a reduced action potential phase 1 velocity as well as a reduction of the heart rate in the atrial cells obtained from the first group of patients. All these electrophysiological changes were found in the absence of any other modification of the overshoot, amplitude or maximum upstroke velocity (Vmax). The increase observed in ERP in cells obtained from atria of patients chronically treated with β-blockers was not observed in those obtained from patients with AF and treated with calcium channel antagonists or angiotensin converting enzyme (ACE) inhibitors. Workman et al. [5] also report the effects of chronic β-blockade on ionic currents recorded in atrial cells obtained from patients suffering from AF. They show a decrease in the amplitude and density of the atrial ITO in atrial myocytes obtained from patients under chronic β-blockade. From records of ITO, the authors conclude that chronic β-blockade does not modify the amplitude of IKsus, which has been mostly attributed to IKur carried by hKv1.5 channels. Workman et al. also compare the effects of chronic β-blockade on the phase 1 Vmax and the ERP with the effects induced by 4-aminopyridine (4-AP) in order to correlate these effects with those observed on ITO. The results obtained with 4-AP mimic those obtained in cells from chronically β-blocked patients. However, although 4-AP is an ITO blocker, it is also a potent IKur antagonist and, therefore, these results have to be taken with caution. Additionally, the authors find a higher R1 in atrial myocytes from patients who underwent β-blockade than in untreated patients, occurring in parallel with an apparent decrease in the amplitude of IK1. Thus, the observed changes in the amplitude and density of ITO, as well as those in R1, appeared in the absence of changes in ICa and IKsus. In contrast to that observed after chronic treatment with β-blockers, chronic administration of calcium channel blockers or ACE inhibitors did not modify APD (both measured at the 50% or 90% repolarization) or ERP. The underlying molecular mechanisms by which the chronic β-blockade induces the electrophysiological effects reported in the present study are unknown at the present time. These changes could be due to modifications either: (1) in the transcription or RNA processing (involving pre RNA splicing, pre RNA editing and RNA destabilisation), or (2) in the translation and post-translational processing (including folding, trafficking, subunit assembly or degradation at the endoplasmic reticulum or Golgi apparatus level, as well as phosphorylation or protein–protein interactions). These modifications could affect the level of functional ion channels in the plasma membrane and, thus, the magnitude of the ionic current [22]. The authors speculate with the notion that if chronic β-stimulation has been shown to alter ion channels via mechanisms including cAMP-regulated modification of transcription [23–25], chronic β-blockade could induce the opposite effects. Whether or not this is the mechanism by which chronic β-blockade induces such electrical remodelling remains to be elucidated in future studies. The analysis of the electrophysiological consequences of pharmacological treatment of AF that have proven its efficacy in the clinical practice is undoubtedly of interest. These kinds of studies will help to unravel the mechanisms that underlie the advantages of the use of a given drug. The distinct electrophysiological profiles observed in cells from patients treated with Ca2+ channel antagonists, ACE inhibitors, and β-blockers are of crucial interest in the development of new strategies of AF treatment not only based on clinical criteria, but also on molecular and cellular electrophysiological aspects. Supported by CICYT SAF2002-02160 and FIS 01/1130 Grants. The author thanks Drs Teresa González and Leandro Sastre for critical reading of the manuscript and useful comments.