L-Type Ca2+ Current Downregulation in Chronic Human Atrial Fibrillation Is Associated With Increased Activity of Protein Phosphatases
Background— Although downregulation of L-type Ca2+ current (ICa,L) in chronic atrial fibrillation (AF) is an important determinant of electrical remodeling, the molecular mechanisms are not fully understood. Here, we tested whether reduced ICa,L in AF is associated with alterations in phosphorylation-dependent channel regulation.
Methods and Results— We used whole-cell voltage-clamp technique and biochemical assays to study regulation and expression of ICa,L in myocytes and atrial tissue from 148 patients with sinus rhythm (SR) and chronic AF. Basal ICa,L at +10 mV was smaller in AF than in SR (−3.8±0.3 pA/pF, n=138/37 [myocytes/patients] and −7.6±0.4 pA/pF, n=276/86, respectively; P<0.001), though protein levels of the pore-forming α1c and regulatory β2a channel subunits were not different. In both groups, norepinephrine (0.01 to 10 μmol/L) increased ICa,L with a similar maximum effect and comparable potency. Selective blockers of kinases revealed that basal ICa,L was enhanced by Ca2+/calmodulin-dependent protein kinase II in SR but not in AF. Norepinephrine-activated ICa,L was larger with protein kinase C block in SR only, suggesting decreased channel phosphorylation in AF. The type 1 and type 2A phosphatase inhibitor okadaic acid increased basal ICa,L more effectively in AF than in SR, which was compatible with increased type 2A phosphatase but not type 1 phosphatase protein expression and higher phosphatase activity in AF.
Conclusions— In AF, increased protein phosphatase activity contributes to impaired basal ICa,L. We propose that protein phosphatases may be potential therapeutic targets for AF treatment.
Received February 26, 2004; de novo April 30, 2004; accepted June 2, 2004.
Atrial fibrillation (AF) induces alterations in atrial electrophysiology that promote its own perpetuation, which has led to the concept of electrical remodeling.1 AF-induced remodeling is associated with a decrease in the atrial effective refractory period and a loss of physiological rate adaptation.
In humans, electrical remodeling is associated with changes in activity of several ion currents.2 During experimental and clinical AF, the amplitude of L-type Ca2+ current (ICa,L) decreases3–8 with corresponding reductions in mRNA and protein levels of the pore-forming α1c subunit.3,8–12 However the hypothesis of transcriptionally mediated downregulation of ICa,L was challenged in humans by recent studies that failed to detect changes in mRNA and protein levels of α1c and the regulatory β2a subunits.13,14 Although reduced amplitude of ICa,L is a consistent finding in AF, the molecular mechanisms are not fully understood.
Functional regulation of Ca2+ channels relies on phosphorylation processes. Protein kinase A and C and the Ca2+/calmodulin-dependent protein kinase II (PKA, PKC, and CAMKII, respectively) affect ICa,L.15,16 In the heart, phosphorylation of ICa,L is counteracted by type 1 and type 2A phosphatases (PP1 and PP2A).17 Thus, the actual amplitude of basal ICa,L is determined by the balanced activity of kinases and phosphatases.
The present study tested the hypothesis that reduced ICa,L in patients with AF is associated with alterations in phosphorylation-dependent channel regulation.
The study was approved by the local ethics committee (No. EK114082202). Each patient gave written informed consent.
Right atrial appendages were obtained from 95 patients with SR and 53 patients with chronic AF (AF >6 months, Table). Significant differences between the groups were found for age, body mass index, incidence of coronary artery disease and/or valve disease, hyperlipidemia, and left atrial diameter. AF patients more often received digitalis, angiotensin receptor inhibitors, and diuretics, whereas β-blockers and lipid-lowering drugs were more frequently used in SR (Table).
Atrial myocytes were isolated with a standard protocol.18 The solution for cell storage contained (in mmol/L) KCl 20, KH2PO4 10, glucose 10, K-glutamate 70, β-hydroxybutyrate 10, taurine 10, EGTA 10, and albumin 1, pH=7.4. Whole-cell voltage-clamp ICa,L recordings were performed as previously described.19 ISO-2 software (MFK) was used for data acquisition and analysis. Borosilicate glass electrodes (Hilgenberg) had tip resistances of 2 to 5 MΩ when filled with (in mmol/L) Cs-methanesulfonate 90, CsCl 20, HEPES 10, Mg-ATP 4, Tris-GTP 0.4, EGTA 10, and CaCl2 3, pH =7.2. Cell capacitances averaged 88.1±1.9 and 106.1±3.8 pF for SR (n=276) and AF (n=138) myocytes, respectively (P<0.01). Seal resistances were 3 to 6 GΩ. Series resistance and system capacitance were compensated. The bath solution contained (in mmol/L) TEA-Cl 120, CsCl 10, HEPES 10, CaCl2 2, MgCl2 1, and glucose 10 (pH 7.4, with CsOH). Basal and norepinephrine (NE)- and isoproterenol (ISO)-activated ICa,L were measured at 37°C in the absence and presence of selective inhibitors of PKA (8-Br-Rp-cAMP, 100 μmol/L in the pipette), PKC (bisindolylmaleimide-I and its inactive form bisindolylmaleimide-V, 1 μmol/L each), CAMKII (KN-93 and its inactive form KN-92, 20 μmol/L each), and PP1/PP2A (okadaic acid, 1 μmol/L). All drugs were from Calbiochem and were applied via a rapid solution exchange system (ALA Scientific Instruments). The data were not corrected for the calculated liquid junction potential (−15 mV; JPCalc, version 2.2).
Reverse-Transcription Polymerase Chain Reaction Analysis
Total RNA isolated from right atrial homogenates was reverse transcribed (Invitrogen) in the presence of random hexanucleotides. PCR experiments were performed in a thermocycler (Master cycler, Eppendorf) using standard PCR reaction mixes (Applied Biosystems) and 3-μL cDNA aliquots. Oligonucleotide sequences for α-actin,20 α1c (forward primer, GCCCCGAAACATGAGCAT; reverse primer, GAAAATCACCAGCCAGTAGAAGA), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH; forward primer, AACAGCGACACCCACTCCTC; reverse primer, GGAGGGGAGATTCAGTGTGGT) were according to published sequences (GeneBank accession Nos. J00071, L29534, and J02642, respectively). The catalytic subunits of PP1 (α, β, and γ) and PP2A (α and β) were detected with isoform-specific PCR primers as previously described.21
Western Blot Analysis
The α1c Ca2+ channel subunit and GAPDH were detected with primary antibodies (Biotrend, Köln, Germany) and anti-rabbit IgG (DAKO, Hamburg, Germany) and anti-mouse IgG (Sigma, Taufkirchen, Germany) as previously described.8 Antibodies against the β2a-subunit were raised in rabbit against an epitope mapping near the C-terminus (KKRNEAGEWNRDVYIRQ) and were affinity purified on the respective antigen column. The protein bands were visualized by using enhanced chemiluminescence (Pharmacia Biotech) and quantified using Quantity One Software (Bio-Rad).
Protein expression of catalytic subunits of PP1α and PP2A was quantified as described.21 The structural subunit of PP2A (PP2A-A)17 was assessed by a goat polyclonal antibody against the carboxy terminus of PP2A-Aα of human origin (Santa Cruz Biotechnology, Inc, Santa Cruz, Calif) and alkaline phosphatase-conjugated rabbit-anti-goat IgG (Sigma, St. Louis, Mo). To demonstrate the specificity of bands, antibodies for α1c, β2a, PP1, and PP2A-C were incubated with corresponding immunizing peptides before blotting. Immunologic signals were visualized using enhanced chemifluorescence (Amersham Pharmacia Biotech) and quantified in Storm 820 using ImageQuaNT-Software (Molecular Dynamics).
Activity of phosphatases was measured in atrial homogenates as previously described.22
Univariate ANOVAs were applied to associate preoperative variables with expression and electrophysiological data (SPSS version 11.5). Concentration-response curves were fitted with software Prism (version 4.0). Differences between continuous data were compared by an unpaired Student t test. Frequency data were analyzed with χ2 statistics. Data are given as mean±SEM. P<0.05 was considered statistically significant.
Expression and activity of proteins and amplitudes of ICa,L were related to selected clinical variables. With univariate ANOVAs, AF was the only predictor of protein expression, phosphatase activity, and electrophysiological parameters (data not shown).
Properties of ICa,L in SR and AF
Basal peak ICa,L at +10 mV was smaller in AF than in SR (−3.8±0.3 pA/pF, n=138/37 [myocytes/patients] and −7.6±0.4 pA/pF, n=276/86, respectively; P<0.001). When expression and function of Ca2+ channels were compared in the same patients, reduced basal ICa,L was not paralleled by decreased α1c and β2a proteins (Figure 1, B through E). In addition, the maximum current was obtained at more positive potentials in AF than in SR (Figure 1F), suggesting phosphorylation-dependent changes in channel regulation rather than expression.23
Regulation of Basal ICa,L by Kinases and Phosphatases
The phosphorylation state of Ca2+ channels depends on the balance between kinases and phosphatases. Therefore, we measured ICa,L in the presence of selective enzyme inhibitors. The PKA blocker 8-Br-Rp-cAMP (100 μmol/L), the PKC blocker bisindolylmaleimide-I, and its inactive form bisindolylmaleimide-V (1 μmol/L each) did not modulate basal ICa,L in either group (data not shown). In contrast, the CAMKII inhibitor KN-93 (20 μmol/L), but not the inactive form KN-92, reduced basal ICa,L by ≈40% in SR but was ineffective in AF (Figure 2, A and B).
The lack of effect of KN-93 on basal ICa,L in AF may be due to increased phosphatase activity. Indeed, the PP1/PP2A inhibitor OA (1 μmol/L) increased basal ICa,L more effectively in AF than in SR (Figure 2, C and D). In the presence of the CAMKII blocker KN-93, OA failed to increase ICa,L in AF (−2.3±0.6 pA/pF before and −2.5±0.2 pA/pF after OA, n=8/2, NS) and SR myocytes (−8.2±0.9 pA/pF before and −8.5±0.5 pA/pF after OA, n=13/3; NS), indicating that the OA-mediated ICa,L increase involves activation by CAMKII.
Modulation of ICa,L by NE and ISO
In both groups, exposure of myocytes to NE (0.01 to 10 μmol/L) increased ICa,L with similar maximum effect (increase in peak ICa,L at 10 μmol/L NE: −6.8±1.3 pA/pF, n=13/5, AF versus −6.7±0.8 pA/pF, n=46/19, SR) and comparable potency (Figure 3). As expected from the higher phosphorylation state of the channels in the presence of NE,23 the maximum of the current-voltage relationship in SR and AF was shifted toward less positive potentials (Figure 3).
Contribution of kinases to NE-activated ICa,L differed between SR and AF. 8-Br-Rp-cAMP did not change NE-activated ICa,L in SR but blocked this current by 60% in AF (P<0.05, Figure 4, A and B). In SR, bisindolylmaleimide-I increased NE-activated ICa,L by ≈100% compared with control but was ineffective in AF. KN-93 inhibited the NE-activated ICa,L by 19% in SR and by 38% in AF (P<0.05), suggesting larger contribution of CAMKII to NE effects in AF (Figure 4, A and B). Bisindolylmaleimide-V and KN-92 had no significant effect. The ineffectiveness of 8-Br-Rp-cAMP on NE-activated ICa,L in SR may be due to opposite effects of α- and β-adrenoceptor-mediated signal transduction. Therefore, we repeated the experiments with ISO, which activates β-adrenoceptors only. In the presence of 8-Br-Rp-cAMP, ISO-activated ICa,L was 31% smaller in SR and 51% smaller in AF than in its absence (Figure 4C).
Expression and Activity of Phosphatases
Because of the evidence for enhanced phosphatase activity in AF, we also studied the expression of the catalytic subunits of PP1 and PP2A in their various isoforms. Unexpectedly, the mRNA levels of all isoforms were lower in AF than in SR, though statistical significance was reached for PP1α only (Figure 5). Because mRNA and protein levels do not always correlate, we also measured PP1 and PP2A proteins. In accordance with the higher impact of PP2A on channel regulation,17 protein expression of the catalytic subunit of this phosphatase was larger in AF than in SR, whereas protein levels were similar for PP1 and the structural PP2A subunit (Figure 6). Correspondingly, phosphatase activity was higher in AF than in SR (Figure 6B).
Here, we demonstrate that decreased basal ICa,L current density in chronic AF is not accompanied by altered expression of the corresponding α1c and β 2a channel subunits. We provide evidence for decreased channel phosphorylation in AF from several observations: (1) the rightward shift of the maximum of the ICa,L current-voltage curve in AF; (2) loss of effect of the CAMKII inhibitor KN-93 in AF; (3) larger increase of basal ICa,L in AF by block of phosphatases with OA; and (4) higher PP2A-C protein expression and phosphatase activity in AF. Additional evidence for impaired channel phosphorylation in AF was provided by the lack of block of NE-activated ICa,L by the PKC inhibitor bisindolylmaleimide-I. Our results suggest that in AF the ratio of protein kinase/phosphatase activity is altered in favor of increased phosphatase activity, resulting in lower basal ICa,L.
Comparison With Previous Studies
Several studies in human atria reported smaller ICa,L amplitude in AF than in SR. Although reduced α1c expression is an attractive molecular mechanism, 10–11 others did not confirm this hypothesis.13,14 In the present study, reduced ICa,L was not paralleled by decreased channel expression (Figure 1). Moreover, the maximum of the current-voltage relationship of ICa,L in AF was shifted to the right, suggesting altered phosphorylation-dependent ICa,L regulation.23
The α1c and β2a subunits possess several phosphorylation sites for PKA and PKC15 and possibly for CAMKII. Interestingly, the latter kinase is regulated by membrane voltage.16 Using selective kinase inhibitors, we found that in SR basal ICa,L is not modulated by PKA or PKC. Interestingly, the CAMKII blocker KN-93 reduced basal ICa,L in SR but not in AF, suggesting that CAMKII may modulate basal ICa,L in SR only. However, reduced CAMKII activity is not a likely explanation because protein levels of CAMKII were ≈90% higher in AF than in SR.24 Because KN-93 prevents the calmodulin binding to CAMKII,25 calmodulin abundance or activity may be affected in AF. Modulation of ICa,L by CAMKII requires the cytoskeleton.26 Thus, we cannot exclude the possibility that abnormalities of cytoskeletal proteins contribute to the lack of effect of CAMKII on ICa,L in AF. Further work is needed to verify these hypotheses.
Cardiac β- and α-adrenoceptors couple to different subtypes of G-proteins, thereby modulating PKA-, PKC-, and CAMKII-dependent processes that regulate ICa,L.15,16 Here, NE increased ICa,L with similar potency and efficacy in SR and AF. The same holds true for maximum effect of ISO, confirming previous results.5 However, the effects of the kinase inhibitors differed between SR and AF. In both groups, block of CAMKII with KN-93 reduced NE-activated ICa,L, whereas inhibition of PKC with bisindolylmaleimide-I strongly increased the current in SR but not in AF. Our data are consistent with results in human atrial myocytes, where activation of PKC inhibits ICa,L.27 In SR, the PKA blocker 8-Br-Rp-cAMP had no effect on NE-activated ICa,L, which may be due to opposite effects of the α- and β-adrenoceptor-mediated signal transduction on ICa,L, because 8-Br-Rp-cAMP reduced the ISO-activated ICa,L both in SR and in AF. In AF, the NE-activated ICa,L was inhibited by 8-Br-Rp-cAMP but not by bisindolylmaleimide-I, indicating impaired ability of PKC to modulate ICa,L in AF.
The lack of effects of CAMKII and PKC on ICa,L in AF suggests either limited kinase ability to phosphorylate the channels or increased phosphatase activity. Blockade of PP1 and PP2A with OA increased basal ICa,L to a greater extent in AF than in SR, and the difference in current density disappeared. The OA-induced ICa,L increase was absent after blocking CAMKII with KN-93, suggesting involvement of this kinase. The stronger effect of OA in AF was compatible with increased PP2A-C protein expression and higher phosphatase activity. Increased PP2A-C expression was not associated with alterations of the structural PP2A-A subunit, suggesting that the former may directly target the channels.28 Thus, increased PP2A activity in AF appears to counteract stimulatory effects of CAMKII on ICa,L, resulting in reduced basal ICa,L.
Here, we did not investigate channel phosphorylation, because measurement of the phosphorylation state of the channels was not feasible in cardiac myocytes.
β-Blockers increase PP1 and PP2A expression in dog hearts.29 However, their distributions were similar in the SR and AF subgroups. In rat hearts, expression and activity of PP1 and PP2A decline with age.22 Though patients were older in the AF group than in the SR group, age-related effects on phosphatase activity are unlikely because neither expression nor activity correlated with age.
Our results are not consistent with data at the single-channel level, where reduced α1c expression was associated with increased mean open time of the single channels.12 The authors suggested reduced PP2A activity as the underlying mechanism, though the protein levels of PP1 and PP2A were unchanged.12 The reason for the discrepant observations is currently unknown. Differences between cell-attached and ruptured whole-cell voltage-clamp methods, uncontrolled state of the cardiac diseases, and/or patients’ medications could contribute to these discrepant findings. Therefore, we cannot exclude the possibility that in a subset of patients with AF, a reduction of α1c expression and a compensatory increase in single-channel ICa,L activity occur.
Increased phosphatase activity may associate with enhanced dephosphorylation of other proteins, eg, those involved in excitation-contraction coupling.17 Because contractile dysfunction in AF may promote atrial thrombus formation,30 possible improvement of atrial contractility by blockade of phosphatases could reduce the risk of stroke in AF patients after cardioversion to SR. In addition, inhibition of phosphatases in AF may promote reversal of atrial remodeling.
Physiological function of cardiac potassium currents requires basal kinase activities.24,31 Thus, increased phosphatase activity may contribute to downregulation of potassium currents in human AF.
We conclude that in AF the ratio of protein kinase/phosphatase activity is altered in favor of increased phosphatase activity, resulting in lower basal ICa,L activity. Thus, protein phosphatases may be potential drug targets for treatment of chronic AF.
These studies were supported by Bundes Ministerium für Bildung und Forschung (“Atrial Fibrillation Network”; Interdisziplinäres Zentrum für Klinische Forschung, Tübingen), Deutsche Forschungsgemeinschaft (DO 769/1 and SFB556), and the MeDDrive-Program of Dresden University of Technology. The authors thank Annegrett Häntzschel, Manja Schöne, Trautlinde Thurm, and Ulrike Heinrich for excellent technical assistance.
Wijffels MCEF, Kirchhof CJHJ, Dorland R, et al. Atrial fibrillation begets atrial fibrillation: a study in awake, chronically instrumented conscious goats. Circulation. 1995; 92: 1954–1968.
Yue L, Feng J, Gaspo R, et al. 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, et al. Ionic mechanisms of electrical remodeling in human atrial fibrillation. Cardiovasc Res. 1999; 44: 121–131.
Van Wagoner DR, Pond AL, Lamorgese M, et al. Atrial L-type Ca2+ currents and human atrial fibrillation. Circ Res. 1999; 85: 428–436.
Workman AJ, Kane KA, Rankin AC. The contribution of ionic currents to changes in refractoriness of human atrial myocytes associated with chronic atrial fibrillation. Cardiovasc Res. 2001; 52: 226–235.
Gaspo R, Sun H, Fareh S, et al. Dihydropyridine and beta adrenergic receptor binding in dogs with tachycardia-induced atrial fibrillation. Cardiovasc Res. 1999; 42: 434–442.
Brundel BJJM, Van Gelder IC, Henning RH, et al. Ion channel remodeling is related to intraoperative atrial effective refractory periods in patients with paroxysmal and persistent atrial fibrillation. Circulation. 2001; 103: 684–690.
Klein G, Schroder F, Vogler D, et al. Increased open probability of single cardiac L-type calcium channels in patients with chronic atrial fibrillation: role of phosphatase 2A. Cardiovasc Res. 2003; 59: 37–45.
Kamp TJ, Hell JW. Regulation of cardiac L-type calcium channels by protein kinase A and protein kinase C. Circ Res. 2000; 87: 1095–1102.
Xiao RP, Cheng H, Lederer WJ, et al. Dual regulation of Ca2+/calmodulin-dependent kinase II activity by membrane voltage and by calcium influx. Proc Natl Acad Sci U S A. 1994; 91: 9659–9663.
Herzig S, Neumann J. Effects of serine/threonine protein phosphatases on ion channels in excitable membranes. Physiol Rev. 2000; 80: 173–210.
Dobrev D, Wettwer E, Himmel HM, et al. G-protein β3-subunit 825T-allele is associated with enhanced human atrial inward rectifier potassium currents. Circulation. 2000; 102: 692–697.
Dobrev D, Graf E, Wettwer E, et al. Molecular basis of downregulation of G-protein-coupled inward rectifying K+ current IK,ACh in chronic human atrial fibrillation: decrease in GIRK4 mRNA correlates with reduced IK,ACh and muscarinic receptor-mediated shortening of action potentials. Circulation. 2001; 104: 2551–2557.
Chen X, Piacentino V 3rd, Furukawa S, et al. L-type Ca2+ channel density and regulation are altered in failing human ventricular myocytes and recover after support with mechanical assist devices. Circ Res. 2002; 91: 517–524.
Tessier S, Karczewski P, Krause EG, et al. Regulation of the transient outward K+ current by Ca2+/calmodulin-dependent protein kinases II in human atrial myocytes. Circ Res. 1999; 85: 810–819.
Dzhura I, Wu YW, Colbran RJ, et al. Cytoskeletal disrupting agents prevent calmodulin kinase, IQ domain and voltage-dependent facilitation of L-type Ca2+ channels. J Physiol. 2002; 545.2: 399–406.
Boixel C, Tessier S, Pansard Y et. al. Tyrosine kinase and protein kinase C regulate L-type Ca2+ current cooperatively in human atrial myocytes. Am J Physiol. 2000; 278: H670–H676.
Davare MA, Avdonin V, Hall DD, et al. A β2 adrenergic receptor signaling complex assembled with the Ca2+ channel Cav1.2. Science. 2001; 293: 98–101.
Reiken S, Gaburjakova M, Gaburjakova J, et al. β-Adrenergic receptor blockers restore cardiac calcium release channel (ryanodine receptor) structure and function in heart failure. Circulation. 2001; 104: 2843–2848.
Schotten U, Ausma J, Stellbrink C, et al. Cellular mechanisms of depressed atrial contractility in patients with chronic atrial fibrillation. Circulation. 2001; 103: 691–698.