Ultrastructural and Functional Remodeling of the Coupling Between Ca2+ Influx and Sarcoplasmic Reticulum Ca2+ Release in Right Atrial Myocytes From Experimental Persistent Atrial Fibrillation
Rationale: Persistent atrial fibrillation (AF) has been associated with structural and electric remodeling and reduced contractile function.
Objective: To unravel mechanisms underlying reduced sarcoplasmic reticulum (SR) Ca2+ release in persistent AF.
Methods: We studied cell shortening, membrane currents, and [Ca2+]i in right atrial myocytes isolated from sheep with persistent AF (duration 129±39 days, N=16), compared to matched control animals (N=21). T-tubule density, ryanodine receptor (RyR) distribution, and local [Ca2+]i transients were examined in confocal imaging.
Results: Myocyte shortening and underlying [Ca2+]i transients were profoundly reduced in AF (by 54.8% and 62%, P<0.01). This reduced cell shortening could be corrected by increasing [Ca2+]i. SR Ca2+ content was not different. Calculated fractional SR Ca2+ release was reduced in AF (by 20.6%, P<0.05). Peak Ca2+ current density was modestly decreased (by 23.9%, P<0.01). T-tubules were present in the control atrial myocytes at low density and strongly reduced in AF (by 45%, P<0.01), whereas the regular distribution of RyR was unchanged. Synchrony of SR Ca2+ release in AF was significantly reduced with increased areas of delayed Ca2+ release. Propagation between RyR was unaffected but Ca2+ release at subsarcolemmal sites was reduced. Rate of Ca2+ extrusion by Na+/Ca2+ exchanger was increased.
Conclusions: In persistent AF, reduced SR Ca2+ release despite preserved SR Ca2+ content is a major factor in contractile dysfunction. Fewer Ca2+ channel–RyR couplings and reduced efficiency of the coupling at subsarcolemmal sites, possibly related to increased Na+/Ca2+ exchanger, underlie the reduction in Ca2+ release.
Atrial fibrillation (AF) is the most common arrhythmia with a life time risk of developing AF of 1 out of 4 in people older than 40 years.1 AF is associated with a significant morbidity. Stroke resulting from embolization of atrial thrombi is the major determinant of this morbidity,2 with loss of atrial contractility as major cause of thrombus formation. The atrial mechanical dysfunction can persist for more than one month after cardioversion to sinus rhythm in patients in persistent AF, a phenomenon called “atrial stunning.”3 The contractile dysfunction is due to remodeling of the muscle. Contractile force of atrial tissue strips from patients with AF was reduced by 75%.4 Exposure to high extracellular Ca2+ restored atrial function, implicating a reduced Ca2+ availability. Studies on animal models and human myocytes isolated from atrial appendages removed at the time of surgery identified a loss of Ca2+ current density as a hallmark of AF.5–8A decrease in mRNA and protein expression of the L-type Ca2+ channel, as well as modulation by phosphorylation and redox potential, contributes to the decrease in Ca2+ current density.9–12 A decrease in the trigger for the Ca2+ release from the sarcoplasmic reticulum (SR) can be a major factor in the contractile dysfunction in AF but may not fully explain reduced contractility.13 Ca2+ release and propagation also depend on properties of the ryanodine receptor (RyR), the extent of a T-tubular system, buffering of Ca2+ by mitochondria, sarcoplasmic reticulum Ca2+ ATPase (SERCA), or myofilaments, and on the SR Ca2+ content. Hyperphosphorylation of RyR and increased SR Ca2+ leak have been implicated in an increased spontaneous Ca2+ release from the SR in AF.14,15 Whether the Ca2+ channel–RyR coupling is intact or whether changes contribute to the contractile dysfunction in chronic AF remains uncertain. Although ideally one would like to study human atrial cells to identify the pathophysiological mechanisms of AF, this approach is limited by the many confounding factors in patient studies, such as sex, age, medication, and concurrent disease. The nature of AF, requiring a substantial mass of cardiac tissue, has therefore early on prompted researchers to use large animals, like dogs or goats for experimental studies.5,16 A prime requirement for these models is that they approach as closely as possible the human condition. Several clinically relevant aspects, such as the time course and nature of the electric remodeling or the link to heart failure, have been successfully implemented in these models, but AF is mostly short-lived. In humans, alterations in gene expression of proteins involved in Ca2+ homeostasis are only observed in patients with persistent AF and not with paroxysmal AF.17 We therefore used a sheep model of persistent AF18,19 to further investigate the mechanisms of atrial contractile remodeling during chronic AF. We focus on the relation between Ca2+ and contraction, on alterations in SR Ca2+ release and the underlying mechanisms.
Animal Characteristics and Instrumentation
A detailed description is provided in the Online Data Supplement, available at http://circres.ahajournals.org. Figure 1 summarizes the salient features of the sheep model for persistent AF induced by long-term intraatrial pacing.18,19 At regular time intervals, pacing was interrupted to asses the underlying rhythm (Figure 1A and 1B). The sheep were considered to be in persistent AF if no sinus rhythm could be demonstrated anymore in the follow-up period. Electric remodeling was present, with shortening of the atrial effective refractory period in vivo and shortening of the action potential in vitro (from 111±6 to 84±12 ms at 75% repolarization; stimulation rate, 1 Hz; AF, n=6; versus control, n=19; P<0.05; Figure 1C) with loss of the normal rate adaption. Persistent AF also induced hemodynamic and structural changes, like atrial dilatation and atrial fibrosis.19 A component of left ventricular dysfunction was present as documented by the reduction of stroke volume and trend to increase of pulmonary capillary wedge pressure (Figure 1D and 1E).
In the present study, sheep were euthanized after being in persistent AF for a mean of 129±39 days. All data were obtained in myocytes isolated from the right atrium. The experimental protocol was approved by the ethical committee on animal handling of the University of Leuven.
Methods for cell isolation, measurements of cell shortening and [Ca2+]i, confocal imaging and image analysis, immunofluorescence, and molecular biology are detailed in the Online Data Supplement.
Decreased Cell Contraction and [Ca2+]i Transients in AF During Field Stimulation
Figure 2A (a) shows typical recordings of cell shortening (ΔL/L0) in control (CTRL) and AF. In AF cells, fractional shortening was significantly decreased, on average by 54.8% (Figure 2A, b). Time to peak of contraction was comparable within the groups (Figure 2A, c), but the rate of relaxation, evaluated from the time constant of the exponential decline, was significantly decreased in AF cells (Figure 2A, d). Figure 2B (a) shows typical examples of [Ca2+]i transients. The amplitude was significantly decreased in the AF group (Figure 2B, b). Time to peak was increased in AF (Figure 2B, c), without change in relaxation rate (Figure 2B, d).
The data suggest that the reduced [Ca2+]i transient could be responsible for the observed depressed cell contraction. Therefore contraction would potentially normalize with a larger [Ca2+]i transient as during maximal SR Ca2+ release obtained during a 10 mmol/L caffeine application. The amplitude of the [Ca2+]i transient during caffeine application was 2- to 3-fold larger than during a depolarizing step and was not significantly different between AF and CTRL (Figure 2C). With this large [Ca2+]i transient, cell shortening was increased 4- to 5-fold and not different between the 2 groups. These data indicate that a defect in Ca2+ handling is a major factor in the reduced contraction of atrial myocytes from persistent AF. They also suggest that the amount of Ca2+ available in the SR is not decreased in AF myocytes.
We therefore further examined the mechanisms underlying the reduced SR Ca2+ release during depolarization.
Decreased L-type Ca2+ Current Density and Underlying Mechanisms
[Ca2+]i transients evoked during voltage clamp to different levels had a similar bell-shaped voltage dependence in AF as in CTRL, indicating that ICaL is the main trigger for SR Ca2+ release (Figure 3A). The amplitude was however profoundly reduced in AF; ICaL was also significantly reduced in AF, but the difference was small (Figure 3B).
When we used a pipette solution with high EGTA to record ICaL, this difference was much larger (Figure 3C) and the voltage dependence of inactivation was shifted to the right (Figure 3D), as was the voltage dependence of activation (Figure 3E). Isoproterenol increased ICaL in both groups with a trend to larger increase in AF; using forskolin to bypass possible changes in β-receptor density, the response in AF was much larger than in CTRL (Figure 3F). Protein expression of α1c subunit of L-type Ca2+ channel, measured by immunoblotting in left atrium (LA) tissue, was significantly decreased in the AF group by 68% (Online Figure I, A).
Reduced Coupling Between Sarcolemmal Ca2+ Influx and SR Ca2+ Release
The data of Figure 2C suggest that SR Ca2+ content was not changed in AF. This was further quantified from the integrated Na+/Ca2+ exchanger (NCX) current (INCX) during fast application of 10 mmol/L caffeine, as illustrated in Figure 4A (a). The integrated current reflecting SR Ca2+ content was similar in both groups. Peak [Ca2+]i of the caffeine-induced transient was also not different (Figure 4A, b), although there was a trend to a decrease, possibly related to slower kinetics of release, as evidenced by the increased time to peak (Figure 4A, c). If SR content is unchanged and Ca2+ influx is also comparable, then fractional release of available SR content during a depolarizing step gives an estimate of global coupling. We recorded [Ca2+]i with the last depolarizing pulse to +10 mV of the conditioning train and the caffeine-induced [Ca2+]i transient (Figure 4B). Fractional release, the ratio between peak [Ca2+]i of the last conditioning pulse and the peak of the caffeine-induced [Ca2+]i transient, was significantly reduced in AF.
From the caffeine-induced Ca2+ transient and NCX current, we evaluated NCX function.
Increased Activity of the NCX in AF
The relaxation of the caffeine-induced [Ca2+]i transient in myocytes from AF was increased (Figure 4C, a), and peak INCX density was significantly increased as well (Figure 4C, b). To allow for differences in [Ca2+]i, we plotted INCX density as a function of [Ca2+]i (Figure 4D, a). With AF, the loop shifted downwards and the slope calculated during the decline of the [Ca2+]i transient was steeper in the AF group, confirming an increased NCX activity. Immunoblot analysis showed a higher protein expression in AF (Figure 4D, b). Protein levels of total SERCA, phospholamban, and Ser16-phosphorylated phospholamban were unchanged (Online Figures I through III).
Structural Changes in the Coupling Between L-type Ca2+ Channel and RyR
Atrial cells from small mammals have no or only a rudimentary T-tubular system,20,21 but this may be different in larger mammals.22 The reduction in L-type Ca2+ current density and α1c protein expression could reflect a decrease in T-tubule surface, where L-type Ca2+ channels are mainly located. After staining the cells with di-8-ANEPPS, confocal images clearly identified tubular structures in a pattern much like that of ventricular myocytes but at lower density (Figure 5A; Online Figure IV). Quantification confirmed that the density was much lower than in ventricular cells obtained from the same sheep (Online Figure V). There was a marked reduction in T-tubule density in the AF group to 44.5% of control levels (Figure 5B). Fourier analysis also indicated a loss of T-tubule organization (Online Figure VI). Decrease in atrial T-tubule density was accompanied, however, by an increase in cell capacitance (AF: 135±9 pF, n=32; versus CTRL: 84±4 pF, n=38; P<0.05), suggesting an increase in cell size. Cell volumes were indeed almost 3-fold larger in AF (Figure 5C, a) with reduced surface-to-volume (S/V) ratio, consistent with loss of T-tubules (Figure 5C, b). Independent measurement of the morphology of isolated atrial myocytes confirmed the cell hypertrophy (cell length and width were, respectively, 188±9 and 26±1 μm in AF versus 140±5 and 18±1 μm in CTRL; P<0.05). Sarcomere length was comparable between both groups (AF: 1.87±0.04 μm; versus CTRL: 1.81±0.08 μm).
Overall RyR distribution with immunofluorescent labeling in isolated myocytes appeared normal (Figure 5D), although density showed a tendency to decrease in AF. An analysis of RyR distribution by measuring the variance of signal density in a grid analysis23 showed no differences between AF and CTRL. Immunoblotting showed a small but significant reduction of RyR expression (Online Figure I, C), but inositol 1,4,5-triphosphate receptor was unchanged (Online Figure III).
Mechanisms of Reduced RyR-Ca Channel Coupling
The data above suggest that loss of T-tubules could reduce the number of functional couplons with more uncoupled or “orphaned” RyR.21,23,24 We therefore examined subcellular patterns of SR Ca2+ release in confocal line scan imaging. In the longitudinal scan direction the Ca2+ release pattern was somewhat inhomogeneous in CTRL myocytes, but Ca2+ release occurred at many sites simultaneously and within a short time frame, consistent with the presence of a T-tubular system (Figure 6A). We analyzed the time course and distribution of Ca2+ release by defining early and delayed areas.23 In AF, a larger number of areas with delayed Ca2+ release were present and these regions were also increased in width when compared to the control group (Figure 6B), consistent with a decreased T-tubule density. This leads to a slower upstroke of the Ca2+ transient in AF, which was quantified as the fraction of the line that had a fluorescence larger than 50% of the maximal (F>F50%, Figure 6C). Ca2+ release remained below F50% along 32.1±3.7% of the line in AF versus 21.3±3.3% in CTRL (P<0.05).
In the transversal scan, the Ca2+ release pattern was much more irregular with often a horse-shoe appearance, although the pattern could be interrupted by areas of early release. We always observed early release at the subsarcolemma (Figure 7A). Ca2+ release propagation in the transversal direction thus appeared less dependent of T-tubule presence and mostly driven by propagation between RyR following the SS release, as inferred for human and small animal atrial myocytes.21,25,26 Consistent with this type of propagation, in spatial average, Ca2+ transients from transversal scans had longer time to peak than the longitudinal transients and lower amplitudes.
From these recordings, we can derive information on the coupling efficiency between Ca2+ channels and RyR, and on propagation between RyR. The SS Ca2+ transients can be assumed to result from couplons of Ca2+ channels and RyR. In the SS transients of the transversal line scan, amplitude in AF tended to be smaller than in control and tended to be abbreviated (Figure 7B). Early release transients in the longitudinal scan that most likely also result from couplons at T-tubular junctions also had smaller amplitude in AF (Figure 7C). In the central regions of the transversal scans, we quantified the propagation between RyR (Figure 7D). The average propagation speed was not altered in AF. Propagation within delayed areas of the longitudinal scan should also reflect the same process23 and was indeed comparable and not different between the groups (Figure 7D).
Increased Glycogen Accumulation but Limited Myolysis
Quantification of periodic acid–Schiff (PAS) positivity in paraffin sections of atrial tissue revealed a significant increase of glycogen deposition in AF (Figure 8A). PAS staining was also performed on epoxy resin–embedded atrial tissue to obtain detailed information on the specific localization of glycogen (Figure 8B). Longitudinally sectioned cardiac myocytes from CTRL sheep mostly showed regular patterns of myofilaments with parallel rows of mitochondria (Figure 8B, 1). Occasionally, the presence of some distributed PAS-positive glycogen granules around the nuclei was observed (Figure 8B, 2). Longitudinal sectioned cardiac myocytes from sheep in AF often revealed increased depositions of PAS-positive glycogen granules interspersed between the intermyofibrillar spaces (Figure 8B, 3). Only a few myocytes were characterized by abundant accumulation of glycogen granules and a concomitant loss of myofilaments or myolysis (Figure 8B, 4), and these were mostly seen in AF. Semiquantitative analysis of the incidence of these distinctive cellular phenotypes in AF and CTRL confirmed these patterns (Figure 8C), consistent with an increased glycogen deposit in AF, with, however, few myocytes with extensive deposits and myolysis.
Cellular Ultrastructural Remodeling in Persistent AF
Persistent AF leads to global remodeling of the atria with dilatation. This dilatation is accompanied by changes in extracellular matrix, which presumably allow cell slippage and rearrangement. Dilatation also results from cellular hypertrophy and increase in cell size,4 related to a certain extent to ultrastructural changes with glycogen accumulation and cells swelling.27 The degree of hypertrophy and of glycogen accumulation and myolysis depends on the chronicity.28
In the present study of persistent AF cells are hypertrophied with glycogen deposition but limited myolysis; S/V is substantially decreased. The latter could in principle simply be the consequence of the increase in cell volume, as for a brick-like structure with a simple unfolded surface, such a volume increase would result in a reduction of S/V. In the more complex organization of the sarcolemma in ventricular myocytes however, T-tubules are essential in increasing S/V, contributing to synchronization of excitation–contraction coupling.29 Recent animal studies have demonstrated reduction of T-tubules and reduced S/V in hypertrophy and failure.23,24,30 Our current data show that in atrial myocytes of a larger animal T-tubules are present and subject to remodeling. This is consistent with very recent data of atrial T-tubular remodeling in tachy-pacing induced heart failure.22 A common feature of the conditions in which this has been observed to date is hypertrophy of the myocytes. It is therefore conceivable that the remodeling of T-tubules is part of the general maladaptive hypertrophy response.
Reduced Efficiency of Coupling Between Ca2+ Influx and SR Ca2+ Release
A reduction in the [Ca2+]i transient amplitude appears to be the major change accounting for the reduced contraction of atrial myocytes in persistent AF. Our data identify that multiple changes are involved.
The first is the structural organization, density, and function of the sarcolemmal L-type Ca2+ channel. As a general concept, this in line with the earlier data from human studies, as well as animal models of chronic AF which reported reduction in Ca2+ current density,6–8 with9 or without31 reduction in channel expression. Our data are consistent with a reduction in protein expression of the pore unit, but reduced baseline phosphorylation could also contribute. Reduced src kinase activity and enhanced protein phosphatase activity have also been implicated in human AF.10,12 In the absence of intracellular Ca2+ buffering, however, the difference between AF and control is much less pronounced, presumably because the larger Ca2+ transients in control myocytes reduce ICaL to a much larger extent than in AF. In the present study, we also identified an additional and novel mechanism, ie, the reduction in the density of the structures where L-type Ca2+ channels are localized, the T-tubules. Taken together, our data indicate that the reduction of Ca2+ influx via L-type channels is a combination of structural and functional alterations.
In ventricular myocytes, loss of T-tubules, even in the absence of any changes in the RyR, leads to a loss of synchrony of SR Ca2+ release and a reduced amplitude of the [Ca2+]i transient.23,24,30 A similar mechanism appears to be present in atrial myocytes from chronic AF. In atrial myocytes, Ca2+ release along the long axis, the contraction axis is remarkably synchronized given the sparse T-tubular system, but this is significantly reduced in AF. Earlier studies on synchrony of Ca2+ release performed in atrial myocytes have emphasized the slow propagation from sarcolemma to center when examining short-axis line scans, resulting in a horse-shoe appearance of the [Ca2+]i transient (eg, see Sheehan et al26). In the short axis, the [Ca2+]i transients we observed did not fully match this pattern, presumably because there are some intersections with T-tubules, but, overall, the propagation was much slower than in the longitudinal direction, both in CTRL and AF. The transversal line recording allows studying Ca2+ release at SS sites of Ca2+ channel–RyR coupling. In AF, Ca2+ release at SS sites tended to be reduced and abbreviated. Possible underlying mechanisms are a reduction in number and/or activity of Ca2+ channels in the couplons or a reduction in the local trigger Ca2+ due to the increased NCX activity. In contrast, the rate of Ca2+ release propagation between RyR was unaffected by chronic AF.
Altered function and distribution of RyR could not be identified in our study, although we found a modest reduction in expression at the molecular level, as reported before in human samples,32 whereas other groups described no change in AF.33 Even with the function of RyR unchanged, the increase in width of the atrial myocytes and the increased number of noncoupled RyR can be expected to prolong the time needed for full propagation of SR Ca2+ release throughout the cell.
There is no apparent decrease in the availability of Ca2+ in the SR (Figure 4A) and expression of SERCA and phospholamban. Yet, taking into account the reduced S/V in AF, this may still result in a reduced increase of cytosolic Ca2+ during release.
Taken together, our data indicate that in chronic AF a reduced number of Ca2+channel–RyR couplings due to loss of T-tubules and a decrease in efficiency of the SS couplings both contribute to the reduction in Ca2+ release from the sarcoplasmic reticulum. The larger cell volume increases the time needed for propagation of Ca2+ release and may further reduce the actual availability of Ca2+ to the myofilaments.
An increase in NCX function is a common, although not general, observation in ventricular remodeling. In AF either no changes in NCX expression or an increased NCX protein expression has been reported, which may depend on the duration of AF.9,34,35 However, there are few functional data available. In this study of persistent AF, the removal of Ca2+ by NCX during caffeine-induced SR Ca2+ release is faster, a measurement that takes into account differences in S/V. The increased NCX capacity in AF could compensate for the high cellular Ca2+ load as consequence of the high and irregular rates of atrial depolarization. We see no change in relaxation of the [Ca2+]i transient during experiments under membrane voltage control, indicating that despite the higher activity, the contribution of NCX to Ca2+ removal on a beat-to-beat basis is modest and that the dominant mechanism of fast Ca2+ removal after SR Ca2+ release is SERCA. Increased NCX activity in the dyad could, however, reduce the local Ca2+ trigger for activation of RyR.
Although we have focused in the present study on the role of Ca2+ handling in the contractile dysfunction, there may also be a role for changes in myofilament Ca2+ response and myofilament organization. It has been reported previously that myocytes from patients with AF show myolysis and glycogen accumulation as observed in hibernating human myocardium.27 In our present model, we saw significant glycogen accumulation but the extent of myolysis was limited. This may well be dependent on the duration of AF.28 It is intriguing that in AF the cell-shortening deficit could be corrected by increasing available Ca2+. The prolonged relaxation despite normal decline of the [Ca2+]i transient may indicate some degree of sensitization, as also suggested by the finding of a lower level of myosin binding protein C phosphorylation.35 Therefore myofilament function needs to be further investigated.
We have restricted our study to the right atrium. Although molecular probing of changes in Ca2+ channel expression and NCX in the LA suggest that similar changes occur in the LA, the functional consequences need to be confirmed, in particular with regard to perspectives for arrhythmogenesis. Several studies have emphasized the predominance of the LA in initiating AF36 and regional differences in electric remodeling.37 Changes in Ca2+ handling and NCX in the LA with focus on a potential link to arrhythmogenesis will be the subject of a future investigation.
In persistent AF, reduced fractional SR Ca2+ release despite preserved SR Ca2+ content is a major factor in the reduced cell contraction. We identified novel mechanisms underlying the reduced SR Ca2+ release, namely a loss of T-tubules with Ca2+ channel–RyR couplings, and a reduced efficiency of coupling of SS Ca2+ channels to RyR. The latter may be related to altered function of the Ca2+ channels, reduced density and/or the increase in NCX activity reducing dyadic Ca2+.
Sources of Funding
This work is supported by grants from the Fund for Scientific Research Flanders (to R.W. and K.R.S.); the Fund for Cardiosurgery (to R.W.); the European Union, LSHM-CT-2005-018833, EUGeneHeart (to K.R.S.); and the Belgian Science Program (IAP6/31 to K.R.S.).
Original received August 28, 2008; first resubmission received March 20, 2009; second resubmission received July 30, 2009; revised second resubmission received August 25, 2009; accepted September 8, 2009.
Lloyd-Jones DM, Larson MG, Leip EP, Beiser A, D'Agostino RB, Kannel WB, Murabito JM, Vasan RS, Benjamin EJ, Levy D. Lifetime risk for developing congestive heart failure: the Framingham Heart Study. Circulation. 2002; 106: 3068–3072.
Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke. 1991; 22: 983–988.
Schotten U, Ausma J, Stellbrink C, Sabatschus I, Vogel M, Frechen D, Schoendube F, Hanrath P, Allessie MA. Cellular mechanisms of depressed atrial contractility in patients with chronic atrial fibrillation. Circulation. 2001; 103: 691–698.
Yue L, Feng J, Gaspo R, Li GR, Wang Z, Nattel S. Ionic remodeling underlying action potential changes in a canine model of atrial fibrillation. Circ Res. 1997; 81: 512–525.
Bosch RF, Zeng X, Grammer JB, Popovic K, Mewis C, Kuhlkamp V. Ionic mechanisms of electrical remodeling in human atrial fibrillation. Cardiovasc Res. 1999; 44: 121–131.
Van Wagoner DR, Pond AL, Lamorgese M, Rossie SS, McCarthy PM, Nerbonne JM. Atrial L-type Ca2+ currents and human atrial fibrillation. Circ Res. 1999; 85: 428–436.
Yagi T, Pu J, Chandra P, Hara M, Danilo PJ, Rosen MR, Boyden PA. Density and function of inward currents in right atrial cells from chronically fibrillating canine atria. Cardiovasc Res. 2002; 54: 405–415.
Yue L, Melnyk P, Gaspo R, Wang Z, Nattel S. Molecular mechanisms underlying ionic remodeling in a dog model of atrial fibrillation. Circ Res. 1999; 84: 776–784.
Christ T, Boknik P, Wohrl S, Wettwer E, Graf EM, Bosch RF, Knaut M, Schmitz W, Ravens U, Dobrev D. L-type Ca2+ current downregulation in chronic human atrial fibrillation is associated with increased activity of protein phosphatases. Circulation. 2004; 110: 2651–2657.
Carnes CA, Janssen PM, Ruehr ML, Nakayama H, Nakayama T, Haase H, Bauer JA, Chung MK, Fearon IM, Gillinov AM, Hamlin RL, Van Wagoner DR. Atrial glutathione content, calcium current, and contractility. J Biol Chem. 2007; 282: 28063–28073.
Sun H, Gaspo R, Leblanc N, Nattel S. Cellular mechanisms of atrial contractile dysfunction caused by sustained atrial tachycardia. Circulation. 1998; 98: 719–727.
Hove-Madsen L, Llach A, Bayes-Genis A, Roura S, Rodriguez FE, Aris A, Cinca J. Atrial fibrillation is associated with increased spontaneous calcium release from the sarcoplasmic reticulum in human atrial myocytes. Circulation. 2004; 110: 1358–1363.
Vest JA, Wehrens XH, Reiken SR, Lehnart SE, Dobrev D, Chandra P, Danilo P, Ravens U, Rosen MR, Marks AR. Defective cardiac ryanodine receptor regulation during atrial fibrillation. Circulation. 2005; 111: 2025–2032.
Wijffels MC, Kirchhof CJ, Dorland R, Allessie MA. Atrial fibrillation begets atrial fibrillation. A study in awake chronically instrumented goats. Circulation. 1995; 92: 1954–1968.
Brundel BJ, Van Gelder IC, Henning RH, Tuinenburg AE, Deelman LE, Tieleman RG, Grandjean JG, Van Gilst WH, Crijns HJ. Gene expression of proteins influencing the calcium homeostasis in patients with persistent and paroxysmal atrial fibrillation. Cardiovasc Res. 1999; 42: 443–454.
Anne W, Willems R, Holemans P, Beckers F, Roskams T, Lenaerts I, Ector H, Heidbuchel H. Self-terminating AF depends on electrical remodeling while persistent AF depends on additional structural changes in a rapid atrially paced sheep model. J Mol Cell Cardiol. 2007; 43: 148–158.
Dibb KM, Clarke JD, Horn MA, Richards MA, Graham HK, Eisner DA, Trafford AW. Characterization of an extensive transverse tubular network in sheep atrial myocytes and its depletion in heart failure. Circulation Heart Failure. 2009.
Heinzel FR, Bito V, Biesmans L, Wu M, Detre E, von Wegner F, Claus P, Dymarkowski S, Maes F, Bogaert J, Rademakers F, D'hooge J, Sipido KR. Remodeling of T-tubules and reduced synchrony of Ca2+ release in myocytes from chronically ischemic myocardium. Circ Res. 2008; 102: 338–346.
Song LS, Sobie EA, McCulle S, Lederer WJ, Balke CW, Cheng H. Orphaned ryanodine receptors in the failing heart. Proc Natl Acad Sci U S A. 2006; 103: 4305–4310.
Hatem SN, Benardeau A, Rucker Martin C, Marty I, de Chamisso P, Villaz M, Mercadier JJ. Different compartments of sarcoplasmic reticulum participate in the excitation-contraction coupling process in human atrial myocytes. Circ Res. 1997; 80: 345–353.
Brette F, Orchard C. T-tubule function in mammalian cardiac myocytes. Circ Res. 2003; 92: 1182–1192.
Schotten U, Haase H, Frechen D, Greiser M, Stellbrink C, Vazquez-Jimenez JF, Morano I, Allessie MA, Hanrath P. The L-type Ca2+-channel subunits alpha1C and beta2 are not downregulated in atrial myocardium of patients with chronic atrial fibrillation. J Mol Cell Cardiol. 2003; 35: 437–443.
Lai LP, Su MJ, Lin JL, Lin FY, Tsai CH, Chen YS, Huang SK, Tseng YZ, Lien WP. Down-regulation of L-type calcium channel and sarcoplasmic reticular Ca(2+)-ATPase mRNA in human atrial fibrillation without significant change in the mRNA of ryanodine receptor, calsequestrin and phospholamban: an insight into the mechanism of atrial electrical remodeling. J Am Coll Cardiol. 1999; 33: 1231–1237.
Schotten U, Greiser M, Benke D, Buerkel K, Ehrenteidt B, Stellbrink C, Vazquez-Jimenez JF, Schoendube F, Hanrath P, Allessie M. Atrial fibrillation-induced atrial contractile dysfunction: a tachycardiomyopathy of a different sort. Cardiovasc Res. 2002; 53: 192–201.
El Armouche A, Boknik P, Eschenhagen T, Carrier L, Knaut M, Ravens U, Dobrev D. Molecular determinants of altered Ca2+ handling in human chronic atrial fibrillation. Circulation. 2006; 114: 670–680.