Functional Roles of Cav1.3(�1D) Calcium Channels in AtriaInsights Gained From Gene-Targeted Null Mutant Mice

Zhao Zhang, MD, PhD; Yuxia He, MD; Dipika Tuteja, PhD; Danyan Xu, MD;Valeriy Timofeyev, PhD; Qian Zhang, MD; Kathryn A. Glatter, MD; Yanfang Xu, MD, PhD;

Hee-Sup Shin, PhD; Reginald Low, MD; Nipavan Chiamvimonvat, MD

Background—Previous data suggest that L-type Ca2� channels containing the Cav1.3(�1D) subunit are expressed mainly inneurons and neuroendocrine cells, whereas those containing the Cav1.2(�1C) subunit are found in the brain, vascularsmooth muscle, and cardiac tissue. However, our previous report as well as others have shown that Cav1.3 Ca2�

channel–deficient mice (Cav1.3�/�) demonstrate sinus bradycardia with a prolonged PR interval. In the present study,we extended our study to examine the role of the Cav1.3(�1D) Ca2� channel in the atria of Cav1.3�/� mice.

Methods and Results—We obtained new evidence to demonstrate that there is significant expression of Cav1.3 Ca2�

channels predominantly in the atria compared with ventricular tissues. Whole-cell L-type Ca2� currents (ICa,L) recordedfrom single, isolated atrial myocytes from Cav1.3�/� mice showed a significant depolarizing shift in voltage-dependentactivation. In contrast, there were no significant differences in the ICa,L recorded from ventricular myocytes fromwild-type and null mutant mice. We previously documented the hyperpolarizing shift in the voltage-dependentactivation of Cav1.3 compared with Cav1.2 Ca2� channel subunits in a heterologous expression system. The lack ofCav1.3 Ca2� channels in null mutant mice would result in a depolarizing shift in the voltage-dependent activation of ICa,L

in atrial myocytes. In addition, the Cav1.3-null mutant mice showed evidence of atrial arrhythmias, with inducible atrialflutter and fibrillation. We further confirmed the isoform-specific differential expression of Cav1.3 versus Cav1.2 by insitu hybridization and immunofluorescence confocal microscopy.

Conclusions—Using gene-targeted deletion of the Cav1.3 Ca2� channel, we established the differential distribution ofCav1.3 Ca2� channels in atrial myocytes compared with ventricles. Our data represent the first report demonstratingimportant functional roles for Cav1.3 Ca2� channel in atrial tissues. (Circulation. 2005;112:1936-1944.)

Key Words: arrhythmias � ion channels � atrial fibrillation � calcium � atrium

Voltage-gated Ca2� channels are heteromultimeric com-plexes of a pore-forming, transmembrane-spanning �1-

subunit, a disulfide-linked complex of �2- and �-subunits, andan intracellular �- and �-subunit.1,2 The �1-subunit is thelargest and incorporates the conduction pore, the voltagesensor, gating apparatus, and the known sites of channelregulation by second messengers, drugs, and toxins. Mam-malian �1-subunits of voltage-gated Ca2� channels are en-coded by at least 10 distinct genes.3 Previous data suggest thatL-type Ca2� channels (LTCCs) containing the Cav1.3(�1D)subunit (D-LTCC) are expressed mainly in neurons andneuroendocrine cells, whereas those containing theCav1.2(�1C) subunit (C-LTCCs) are found in the brain, vas-cular smooth muscle, and cardiac tissue. Recently, we andothers, have shown that the Cav1.3 Ca2� channel is highlyexpressed in cardiac pacemaking tissue and plays an impor-

tant role in the spontaneous diastolic depolarization andfrequency of beating in sinoatrial (SA) node cells.4–6 Specif-ically, using a mouse model of gene-targeted deletion ofCav1.3 Ca2� channel, we established a role for the Cav1.3 Ca2�

channel in the generation of spontaneous action potential inSA node cells.4 The Cav1.3-null mutant mouse shows evi-dence of profound SA and atrioventricular (AV) node dys-function. We observed that D-LTCCs show a low activationthreshold compared with that of C-LTCCs. The hyperpolar-izing shift in the activation threshold of the Cav1.3 Ca2�

channel can be directly documented in isolated SA node cellsas well as in a nonexcitable expression system, wherein theCav1.3 subunit can be expressed and studied alone.4 Thisgating property of D-LTCCs contributes importantly to thegeneration of spontaneous action potential and pacemakingactivities within SA node cells.

Received January 31, 2005; revision received June 15, 2005; accepted June 27, 2005.From the Division of Cardiovascular Medicine (Z.Z., X.H., D.T., D.X., V.T., Q.Z., K.A.G., Y.X., R.L., N.C.), Department of Internal Medicine,

University of California, Davis; the Department of Veterans Affairs (N.C.), Northern California Health Care System, Mather, Calif; the Center forCalcium and Learning (H.-S.S.), Division of Life Sciences, Korea Institute of Science and Technology, Seoul, Korea; and the Department of Physiology(Z.Z.), Henan Medical University, Zhingzhou, China.

Correspondence to Nipavan Chiamvimonvat, Division of Cardiovascular Medicine, University of California, Davis, One Shields Ave, GBSF 6315,Davis, CA 95616. E-mail nchiamvimonvat@ucdavis.edu

© 2005 American Heart Association, Inc.

Circulation is available at http://www.circulationaha.org DOI: 10.1161/CIRCULATIONAHA.105.540070

1936 by guest on August 18, 2015http://circ.ahajournals.org/Downloaded from

In the present report, we present new evidence that dem-onstrates that there is significant expression of Cav1.3 Ca2�

channels in atrial but not in ventricular tissue. Specifically,using in situ hybridization and immunocytochemistry, weshow that there is robust expression of Cav1.3 Ca2� channelsin mouse atrial but not ventricular tissue. Because bothisoforms have similar pharmacological properties, it is diffi-cult to isolate one current from the other with conventionalelectrophysiology. The Cav1.3-null mutant mouse modelprovides a unique opportunity to directly determine thecontribution of D- versus C-LTCCs in atrial versus ventric-ular tissues. Here, using in vitro and in vivo electrophysio-logical recordings, we document for the first time the func-tional roles of the Cav1.3 Ca2� channel in mouse atrialmyocytes.

MethodsAnimals and ProtocolsThe present investigation conformed to the Guide for the Care andUse of Laboratory Animals published by the US National Institutesof Health (NIH publication 85-23, revised 1985) and was performedin accordance with the guidelines of the Animal Care and UseCommittee of the University of California, Davis. Generation ofCav1.3-null mutant mice has been previously described.4,7

Electrophysiological RecordingsSingle atrial and ventricular myocytes were isolated from Cav1.3�/�,Cav1.3�/�, and wild-type (WT, Cav1.3�/�) littermates on a C57BL/6Jbackground, as previously described.8 Whole-cell ICa was recorded atroom temperature with patch-clamp techniques.9,10 The externalsolution contained (in mmol/L) N-methyl glucamine (NMG) 140,CsCl 5, MgCl2 0.5, CaCl2 2, 4-amino pyridine 2, glucose 10, andHEPES 10, and the internal solution contained NMG 135, tetraeth-ylammonium chloride 20, disodium ATP 4, EGTA 1, and HEPES10. All chemicals were purchased from Sigma Chemical unlessstated otherwise. Cell capacitance was calculated by integrating thearea under a curve of an uncompensated capacitative transientelicited by a 20-mV hyperpolarizing pulse from a holding potentialof �40 mV. Whole-cell current records were filtered at 2 kHz andsampled at 10 kHz. Liquid junction potentials were measured aspreviously described,11 and all data were corrected for liquidjunction potentials. Curve fitting and data analysis were performedwith Origin software (MicroCal Inc).

Reverse Transcription–Polymerase Chain ReactionTotal RNA was prepared from the atria and ventricles of wild-type(WT) C57BL/6J mice with TRIzol Reagent (Invitrogen). cDNA wassynthesized from total RNA samples by oligo(dT)-primed reversetranscription (RT) (Superscript II RNase H-reverse transcriptase,Invitrogen). cDNA was then subjected to polymerase chain reaction(PCR) amplification with HotStarTaq DNA polymerase (Qiagen).Primers used in the PCR were designed from mutually uniqueregions of Cav1.2 and Cav1.3 channels as follows: (1) for Cav1.3,5�-ATGAACCTTCCGACATTTTC-3� (forward) and 5�-GTGCTCATAGTCTGGGCGGC-3� (reverse), according to the pub-lished sequence of mouse Cav1.3 (accession No. NM_028981) and(2) for Cav1.2, 5�-ATGGTCAATGAAAACACGA-3� (forward) and5�- ACTGACGGTAGAGATGGTTG-3� (reverse), according to thepublished sequence of mouse Cav1.2 (accession No. NM_009781).The absence of genomic contamination in the RNA samples wasconfirmed by RT-negative controls for each experiment.

In Situ HybridizationCav1.2- and Cav1.3-specific cDNA fragments were subcloned into aTA cloning vector (Invitrogen). All clones were sequenced. Thesense and antisense riboprobes were synthesized in the presence of

UTP-digoxigenin label with use of a DIG RNA labeling kit (Roche).Mouse hearts procured from 8- to 10-week-old WT mice weredissected and perfused first with Rnase-free phosphate-bufferedsaline and later with 4% paraformaldehyde (made in phosphate-buffered saline). Cav1.3�/� mouse hearts were also used as negativecontrols for the Cav1.3 probe. Perfused hearts were fixed overnight at4°C in 4% paraformaldehyde. Infiltration of hearts was done with amixture of 10% and 30% sucrose in ratios of 2:1, 1:1, and 1:2 atroom temperature for 30 minutes (each) with gentle rotation. Heartswere transferred to 30% sucrose and allowed to settle at 4°C. Heartswere then transferred to degassed OCT medium (Tissue-Tek) andmaintained at 4°C overnight with rotation. Embedding was done inOCT medium, and the samples were frozen on a dry ice/ethanol bath.Cryosectioning was completed, and sections were laid on gelatin-coated slides (Fisher Scientific). After air-drying, in situ hybridiza-tion was performed with the anti-sense as well as the correspondingsense cRNA probes on adjacent sections. After hybridization andwashes, the sections were subjected to immunologic detection withanti-digoxigenin Fab fragments conjugated to alkaline phosphataseby using a DIG nucleic acid detection kit (Roche). The signals weredeveloped with nitro blue tetrazolium and bromochloroindolyl phos-phate (Roche) added in alkaline phosphatase buffer in the presenceof levamisole (Sigma) to inhibit endogenous alkaline phosphatase.The specimens were inspected for development of purple precipitateby bright-field microscopy (Carl Zeiss Vision). Digitized imageswere obtained with AxioVision 4 (Zeiss).

Immunofluorescence Confocal MicroscopyImmunofluorescence labeling was performed as described previous-ly.12 The following primary antibodies were used: (1) anti-Cav1.3(Santa Cruz Biotechnology, Inc), a polyclonal antibody raised in goatagainst a purified peptide corresponding to amino acid residues 859to 875 of rat Cav1.3 (accession No. P27732)13 and (2) anti-Cav1.2(Alomone Labs), a polyclonal antibody raised in rabbit against aglutathione-S-transferase fusion protein with residues 1 to 46 ofrabbit Cav1.2 (accession No. P15381).14 The cells were treated withanti-Cav1.3 or anti-Cav1.2 antibodies (1:200 dilution for 1 hour).Immunofluorescence labeling for confocal microscopy was per-formed by treatment with Texas red–conjugated goat anti-rabbitantibody or rabbit anti-goat antibody (Calbiochem, 1:500 dilution).Immunofluorescence-labeled samples were examined with a PascalZeiss confocal laser scanning microscope. Control experimentsperformed by incubation with secondary antibody only did not showpositive staining under the same experimental conditions. Identicalsettings were used for all specimens.

Transient Transfection of Cav1.2 and Cav1.3 Ca2�

Channels in HEK CellsHEK 293 cells were maintained in Dulbecco’s modified Eagle’smedium containing 10% fetal bovine serum, 2 mmol/L L-glutamine,and 1% penicillin/streptomycin (Invitrogen) and kept at 37°C in a5% CO2 incubator. Cells were transiently transfected with thecalcium phosphate precipitation procedure (Invitrogen) as describedpreviously.4 Channel subunits to be studied were subcloned intopGW1H, an expression vector with a cytomegalovirus promoter(British Biotechnology). Cells were transiently transfected with 7.5�g of plasmid containing the Cav1.3 Ca2� channel (a gift from Dr S.Seino, Kobe University, Kobe, Japan) or the Cav1.2 Ca2� channeland coexpressed with 5 �g of plasmid containing the gene thatencodes a �1A-subunit (derived from skeletal muscle).

In Vivo Electrophysiological Studies in MiceIn vivo electrophysiological studies were performed as previouslydescribed.15 Standard pacing protocols were used to determine theelectrophysiological parameters, including sinus node recovery time;atrial, AV nodal, and ventricular refractory periods; and AV nodalconduction properties. Each animal underwent an identical pacingand programmed stimulation protocol. The Q-T interval was deter-mined manually by placing cursors on the beginning of the QRS andthe end of the T wave. The rate-corrected QT interval was calculated

Zhang et al Cav1.3(�1D) Ca2� Channel in Atria 1937

by guest on August 18, 2015http://circ.ahajournals.org/Downloaded from

with a modified Bazett’s formula as reported by Mitchell et al,16

whereby the RR interval was first expressed as a unitless ratio (RRin ms/100 ms). The rate-corrected QT interval was defined as QTinterval (in ms)/(RR/100)1/2.

To induce atrial and ventricular tachycardia and fibrillation,programmed extrastimulation techniques and burst pacing wereused. Programmed right atrial and right ventricular double and tripleextrastimulation techniques were performed at a 100-ms drive cyclelength, down to a minimum coupling interval of 10 ms. Right atrialand right ventricular burst pacing was performed as eight 50-ms andfour 30-ms cycle-length train episodes repeated several times, up toa maximum 1-minute time limit of total stimulation. For comparisonof inducibility, programmed extrastimulation techniques and stimu-lation duration of atrial and ventricular burst pacing were the same inall mice. Sustained atrial or ventricular arrhythmias were defined asatrial arrhythmias lasting �30 seconds. Reproducibility was definedas �1 episode of induced atrial or ventricular tachycardia.

StatisticsData are presented as mean�SEM. Comparison among the 3genotypes was performed with SigmaStat with ANOVA and pair-wise multiple comparison procedures (Holm-Sidak method). Weanalyzed each litter separately and did not find significant outliersfrom the different litters. Because each litter was too small to allowfor statistical analysis, we combined the data from all litters for thefinal statistical analysis. Comparison of the occurrence of atrialarrhythmias was performed with Fisher’s exact test.

ResultsCav1.3 Transcripts Are Highly Expressed inMouse Atria Compared With VentriclesWe directly probed for the existence of the Cav1.3 Ca2�

channel in mouse cardiac myocytes by RT-PCR. Figure 1Ashows representative RT-PCR–amplified products with prim-ers specific for Cav1.2 versus Cav1.3 Ca2� channels andprimers specific for glyceraldehyde 3-phosphate dehydroge-nase as a positive control from total RNA from mouse rightatria, left atria, right ventricles, left ventricles, septum, andbrain. Primers designed from mutually unique regions ofCav1.2 and Cav1.3 channels in the N-termini are shown inFigure 1B. Whereas Cav1.2 transcripts are expressed through-out the different regions in atria and ventricles, Cav1.3transcripts are highly expressed in the atria compared with theventricles. The signals obtained from the right and leftventricles and interventricular septum were very low com-pared with the atria.

To further confirm that the RT-PCR products generatedwith primers specific to the Cav1.3 Ca2� channel were indeedamplified from cardiac myocytes and not other cell types inthe cardiac homogenate (eg, vascular smooth muscle cells),we further generated sense and antisense riboprobes in thepresence of UTP-digoxigenin label for in situ hybridization.Figure 1C is a photomicrograph comparing the distribution ofCav1.2 versus Cav1.3 transcripts in mouse atria and ventricles.Whereas Cav1.2 transcripts are present in both the atria andventricles, Cav1.3 transcripts are present mainly in the atria.Sense riboprobes were used as negative controls from con-secutive sections (labeled as sense).

Immunodetection of Cav1.2 Versus Cav1.3 Ca2�

Channels in Dissociated Mouse Atrial andVentricular MyocytesTo further examine the regional distribution of the Cav1.2 andCav1.3 Ca2� channels at the protein level, we performed an

immunofluorescence confocal microscopy study of single,isolated, mouse atrial and ventricular myocytes. The speci-ficity of the antibodies and the lack of cross reactivity at thedilutions used for the 2 different isoforms of the Ca2�

channels were first tested in expressed Cav1.2 and Cav1.3Ca2� channels in HEK 293 cells (Figure 2A, 2B, 2D, and 2E)compared with nontransfected cells (Figure 2G and 2H).Figure 2C and 2F represent negative controls treated withsecondary antibodies only.

Figure 3A and 3C shows specific labeling with anti-Cav1.2antibody in isolated mouse atrial and ventricular myocytes,respectively. In contrast, only atrial myocytes show specificlabeling with the anti-Cav1.3 antibody (Figure 3B). Only alow level of staining with the anti-Cav1.3 antibody wasobserved in ventricular myocytes (Figure 3D). We furtherdocumented the specificity of the anti-Cav1.3 antibody inisolated atrial myocytes from Cav1.3�/� mutant mice (Figure3E and 3F). Whereas atrial myocytes from Cav1.3�/� mutantmice showed specific labeling with the anti-Cav1.2 antibody(Figure 3E), no staining was observed with the anti-Cav1.3antibody (Figure 3F). Figure 3G shows additional images ofnegative controls with secondary antibody only.

Functional Roles of the Cav1.3 Ca2� Channel inthe Heart Assessed in Cav1.3�/� Mutant MiceOur data on the regional localization of the Cav1.3 Ca2�

channel transcript and protein with in situ hybridization andimmunofluorescence confocal microscopy are consistent withexpression of the Cav1.3 Ca2� channel mainly in the atria.However, the functional roles of the differential expression ofthe Cav1.3 Ca2� channel are unknown. Because Cav1.2 andCav1.3 Ca2� channels have similar pharmacological proper-ties, we reasoned that Cav1.3�/� mutant mice would be anideal model to study the functional role of Cav1.3 Ca2�

channels in the atria. We undertook in vivo electrophysiolog-ical studies comparing mutant mice with heterozygous andWT animals. All mutant mice showed evidence of SA andAV nodes dysfunction, as assessed by sinus cycle length,sinus node recovery time, PR interval, and Wenckebach cyclelength (see the Table). Furthermore, atrial arrhythmias,mainly atrial fibrillation, were induced in all mutant mice anda small number of heterozygous littermates. In contrast, atrialarrhythmias were induced in none of the WT littermates(P�0.01 comparing WT and mutant animals by Fisher’sexact test). Indeed, previous studies of the same backgroundmouse model have shown that WT mice are not inducible foratrial arrhythmias in the absence of carbachol.17 Figure 4shows examples of atrial fibrillation and atrial flutter thatwere induced in a Cav1.3�/�-null mutant mouse. In contrast,ventricular arrhythmias were not induced in either the WT,heterozygous or the homozygous mutant mice. The in vivoelectrophysiological parameters are summarized in the Table.

Whole-Cell ICa,L Recorded From Cav1.3�/� Atrialand Ventricular Myocytes Compared With ThoseFrom WT LittermatesTo further corroborate the findings from the in vivo func-tional studies described earlier, we directly recorded ICa,L fromatrial and ventricular myocytes from Cav1.3�/� and compared

1938 Circulation September 27, 2005

by guest on August 18, 2015http://circ.ahajournals.org/Downloaded from

Figure 1. A, Representative agarose gels of RT-PCR–amplified products with the use of primers specific for Cav1.2 vs Cav1.3 Ca2�

channels and primers specific for glyceraldehyde 3-phosphate dehydrogenase as a positive control from total RNA from mouse RA(right atria), LA (left atria), RV (right ventricles), LV (left ventricles), S (septa), and B (brains). �ve refers to a negative control (PCR-amplified product without RT to ensure that there was no genomic contamination of the RNA samples). Lane 1 is the HI-LO DNA mark-ers (Bioscience, Inc). B, Nucleotide sequence alignment (ClustalW) of the N-termini of mouse Cav1.2 and Cav1.3 Ca2� channels. High-lighted regions refer to the primers used for the PCRs to generate the sense and antisense riboprobes. *Conserved nucleotide betweenthe 2 isoforms. Nucleotide sequence numbers are given on the right. C, Photomicrographs comparing the distribution of Cav1.2 vsCav1.3 transcripts in mouse atria and ventricles. Sections obtained with the corresponding sense riboprobes are shown to the right asthe negative controls. RA and LV refer to right atrium and left ventricle, respectively.

Zhang et al Cav1.3(�1D) Ca2� Channel in Atria 1939

by guest on August 18, 2015http://circ.ahajournals.org/Downloaded from

them with those of heterozygous and WT littermates. Whole-cell ICa,L was recorded at a holding potential of �55 mV.Figure 5A shows examples of ICa,L current traces elicited atthe various step potentials from atrial myocytes isolated fromCav1.3�/� and Cav1.3�/� littermates compared with Cav1.3�/�.The current-voltage relations are summarized in Figure 5B.Even though there were no significant differences in currentdensity among the 3 different groups of animals, there was adepolarizing shift in the voltage-dependent activation of thecurrent when we compared Cav1.3�/� to Cav1.3�/� andCav1.3�/� with Cav1.3�/� mice. We further confirmed thisinitial impression by generating activation curves from WT,heterozygous, and mutant animals (Figure 5C). ICa,L recordedfrom Cav1.3�/� atrial myocytes showed an �5-mV depolar-izing shift at the midpoint of activation compared with WTanimals, whereas current from Cav1.3�/� mice showed afurther depolarizing shift of �7 mV compared with theheterozygous animals. Figure 5D shows data obtained with a2-pulse protocol to examine the voltage- and Ca2�-dependentinactivation of ICa,L in WT, heterozygous, and mutant animals.The curves appear nearly superimposed, with no significantdifferences in the half-inactivation voltages. In addition, thecurves show the typical U-shape configuration for Ca2�-dependent inactivation of L-type Ca2� current. Typical traceselicited with the test pulse are shown in the insert. Prepulses

more positive than �20 mV elicited a progressively smallerinward current as the command voltages approach the rever-sal potential, leading to partial recovery of the L-type Ca2�

current elicited with the test pulse, owing to a decrease inCa2�-dependent inactivation. There were no significant dif-ferences in voltage dependence of the inactivation profileamong the 3 groups of animals.

Figure 6 shows the same set of experiments obtained fromfree-wall left ventricular myocytes, comparing the 3 groupsof animals. In contrast with the data obtained from atrialmyocytes, there were no significant differences in ICa,L rec-orded from the left ventricular myocytes among the 3 groupsof animals. These functional data are consistent with our insitu hybridization and immunofluorescence studies showingthat the Cav1.3 transcript and protein are present predomi-nantly in atrial tissues.

Previous data provide important clues that the differencesin biophysical properties of Cav1.2 versus Cav1.3 Ca2� chan-nels may be directly responsible for the observed findings inthe atria.18,19 ICa,L recorded from atrial myocytes isolated fromCav1.3�/� mutant animals were activated at more depolarizingpotentials compared with those from Cav1.3�/� or Cav1.3�/�,which expressed both Cav1.2 and Cav1.3 Ca2� channels(Figure 5C). Indeed, we have previously documented in aheterologous expression system that there is a significant

Figure 2. Confocal photomicrographs of HEK 293 cells expressing Cav1.2 (A–C) vs Cav1.3 (D–F) Ca2� channels. Anti-Cav1.2 (A and D)and anti-Cav1.3 (B and E) antibodies were used. Immunofluorescence labeling was done by treatment with secondary antibodies (Texasred–conjugated anti-rabbit antibodies). C and F show samples treated with secondary antibodies only as negative controls. G and Hincluded nontransfected cells treated with anti-Cav1.2 vs anti-Cav1.3, respectively. Corresponding differential interference contrast (DIC)images are shown in the right panels. The scale bar is 20 �m.

1940 Circulation September 27, 2005

by guest on August 18, 2015http://circ.ahajournals.org/Downloaded from

depolarizing shift in steady-state activation in Cav1.2 com-pared with Cav1.3 Ca2� currents, consistent with findings inthe Cav1.3�/� mice, which express only the Cav1.2 subunit.4

DiscussionIn this study, we directly tested the role of the Cav1.3 Ca2�

channel in atrial myocytes in Cav1.3-null mutant mice.

Whole-cell ICa,L recordings from atrial myocytes isolated fromthe null mutant mice showed a depolarizing shift in thevoltage-dependent activation compared with WT. In contrast,there were no significant differences in whole-cell ICa,L

recorded from ventricular myocytes from WT or null mutantmice. Consistent with these findings, we previously docu-mented the hyperpolarizing shift in voltage-dependent acti-vation of Cav1.3 compared with Cav1.2 Ca2� channel subunitsin a heterologous expression system.4 The lack of Cav1.3 Ca2�

channels in the null mutant mice would result in a depolar-izing shift in the voltage-dependent activation of ICa,L in atrialmyocytes. We further confirmed the isoform-specific differ-ential expression of Cav1.3 in atrial myocytes by in situhybridization and immunofluorescence confocal microscopy.

Voltage-Gated Ca2� Channel Subtypes in the HeartThe molecular basis for ICa in the heart has previously beeninvestigated. By in situ hybridization, it was found that themost prominently expressed low-voltage activated Ca2� chan-nel in the SA node was Cav3.1(�1G), whereas Cav3.2(�1H) ispresent at moderate levels.20 In addition, we and others havepreviously documented the critical role of Cav1.3 in the SAnode by using mutant mouse models.4–6 The dominanthigh-voltage activated Ca2� channel was Cav1.2, whereasonly a small amount of Cav1.3 mRNA was detected in SAnode myocytes of mice.20 The existence the Cav1.3 Ca2�

Figure 3. Confocal photomicrographs from freshly isolated WT mouse atrial (A and B) and ventricular (C and D) myocytes treated withanti-Cav1.2 (A and C) vs anti-Cav1.3 (B and D) antibodies. Immunofluorescence labeling was done by treatment with secondary anti-bodies (Texas red–conjugated antibodies). E and F represent mouse atrial myocytes isolated from Cav1.3�/� mutant mice. G, Mouseventricular myocytes isolated from WT mice treated with secondary antibodies only as negative controls. Corresponding DIC imagesare shown in the right panels. The scale bar is 20 �m.

In Vivo Electrophysiological Studies in Cav1.3�/� and Cav1.3�/�

Mice Compared With WT Littermates

Cav1.3�/�

(n�9)Cav1.3�/�

(n�11)Cav1.3�/�

(n�8)

Sinus cycle length 149.3�5.6 138.6�8.8 287.9�38.1*

PR interval 44.4�1.5 42.0�2.5 61.0�4.7*

QTc interval 32.6�1.3 36.0�1.8 33.2�2.1

Sinus node recovery time 201.9�17 205.8�14.9 329.6�37.7†

Wenckebach cycle length 85.6�2.4 85.0�3.5 123.8�9.1*

AV node ERP 74.4�2.9 69.1�2.5 85.0�3.3†

Atrial ERP 23.2�2.4 28.8�3.0 21.8�2.3

Ventricular ERP 33.3�2.9 37.8�4.9 35.0�3.9

Atrial arrhythmias 0/9 2/11 8/8*

Ventricular arrhythmias 0/9 0/11 0/8

Data shown are mean�SEM. Measurement of AV node, atrial, andventricular effective refractory periods (ERPs) were performed with a basiccycle length of 100 ms. n refers to the No. of animals in the studies.

*P�0.01, †P�0.05, Cav1.3�/� vs Cav1.3�/� and Cav1.3�/�.

Zhang et al Cav1.3(�1D) Ca2� Channel in Atria 1941

by guest on August 18, 2015http://circ.ahajournals.org/Downloaded from

channel in different regions of the heart has been furtherdocumented in a recent study by Marionneau et al21 and isconsistent with our findings: Cav1.3 was found to be moreprominently expressed in the atria compared with the ventri-cles. However, the functional roles of Cav1.3 in the atria orventricular tissues have never been documented.

Role of Cav1.3 Ca2� Channels in Atrial MyocytesHere, using gene-targeted deletion of the Cav1.3 isoform, wewere able to document that genetic ablation of the Cav1.3isoform results in the occurrence of atrial arrhythmias. In-deed, this mutant mouse model represents one of the fewgenetic models of atrial fibrillation. Even though Cav1.3represents only a small amount of the LTCC transcript in theatria, owing to the significant differences in biophysicalproperties of the Cav1.3 isoform, the channel contributessignificantly to the overall function in the atria. Our in vivoelectrophysiological data as well as patch-clamp recordingsare consistent with the notion that Cav1.3 LTCCs are ex-pressed and contribute functionally to atrial cardiac myocytesin contrast to ventricular myocytes.

Atrial FibrillationAtrial fibrillation is the most common clinical arrhythmia andis associated with a significant increase in morbidity andmortality.22 The underlying mechanisms of atrial fibrillationare very heterogeneous and are often related to underlyingheart or pulmonary diseases. However, more recently, severalstudies have identified mutations in ion channels as possiblecauses of inherited atrial fibrillation.23–26 The first gene for aninherited form of atrial fibrillation was identified in a familywith autosomal-dominant transmission.24 A mutation wasfound in the K� channel gene KCNQ1, resulting in again-of-function mutation. This is in contrast with the reduc-tion in current density seen with mutations in other residuesin this gene causing long QT syndrome type 1. The gain-of-function mutation is consistent with the decrease in action

potential duration and effective refractory period, which arethought to be the mechanisms of atrial fibrillation. On theother hand, a number of patients with the mutation also had aprolonged QT interval,24 emphasizing the fact that our under-standing of repolarization is incomplete. In addition, recentdata also suggest a role for genetic modifiers, or incompletelypenetrant disease genes, as mechanisms for the developmentof atrial fibrillation. Our data showing the development ofatrial fibrillation in null mutant mice suggest an importantfunctional role for Cav1.3 in the atria. Ablation of the Cav1.3Ca2� channel did not alter the atrial effective refractory periodbut might nonetheless alter atrial action potential duration orCa2�-activated repolarizing currents. Direct measurements ofaction potential duration are required to establish this. Addi-tional studies are also needed to further examine the effects ofthe Cav1.3 Ca2� channel on Ca2� transients. Finally, therelevance of this model to human atrial fibrillation remainsonly speculative at this time.

Compensatory Changes in Mutant MiceBecause the relative contribution of the Cav1.3 Ca2� channelto total ICa,L in atrial myocytes is unknown, we directlycompared the maximum ICa,L density between mutant miceand their WT littermates (Figure 5B). The current wasnormalized to cell capacity. Cell capacitance of single,isolated, atrial myocytes from the 3 groups of animals was53.9�2.2, 41.9�2.4, and 52.3�5.0 pF for Cav1.3�/�,Cav1.3�/�, and Cav1.3�/� mice, respectively (n�8, P�NS).There was a �12-mV depolarizating shift in the peak ICa,L inthe Cav1.3�/� mice compared with their WT littermates;however, the current density was not significantly differentbetween the WT and homozygous mutant animals. This mayrepresent a compensatory change, with upregulation ofCav1.2 ICa,L in the mutant animals.

In summary, using gene-targeted deletion of the Cav1.3Ca2� channel, we established the important functional roles ofCav1.3 Ca2� channels in atrial myocytes in addition to its

Figure 4. In vivo electrophysiological studies in Cav1.3-null mutant mice, showing evidence of inducible atrial fibrillation (A) and atrialflutter (B) after atrial extrastimuli. Upper tracings are surface ECG (leads I and II). Lower traces are intracardiac electrograms showingatrial and ventricular electrograms, with induced sustained rapid atrial fibrillation and atrial flutter with relatively slow ventricularresponse.

1942 Circulation September 27, 2005

by guest on August 18, 2015http://circ.ahajournals.org/Downloaded from

previously documented role in pacemaking cells. The hyper-polarizing shift in the activation threshold of the Cav1.3 Ca2�

channel can be directly documented by gene-targeted deletionin the Cav1.3 mutant mouse model. Our in vivo electrophys-iological data support important roles for Cav1.3 in atrialmyocytes; the Cav1.2 subunit cannot functionally substitutefor the Cav1.3 subunit. Important phenotypes of atrial fibril-lation and atrial flutter were documented in the null mutantmice. Similar to that in neuronal systems, the expression of

multiple Ca2� channel subtypes appears to be important incoordinating the different physiological functions in atrialmyocytes in addition to cardiac pacemaking cells. Takentogether, our data represent the first report on the functionalroles for Cav1.3 Ca2� channels in atrial cardiomyocytes.Importantly, the differential expression of the 2 differentisoforms of the LTCC, with predominant expression of theCav1.3 channel in the atria compared with the ventricles, mayoffer a unique therapeutic opportunity to directly modify theatrial cells without interfering with ventricular myocytes.

AcknowledgmentsThis study was supported by NIH/NHLBI grants (RO1, HL67737,and HL75274) and the Nora Ecceles Treadwell Foundation Award

Figure 5. A, Example of whole-cell ICa,L recorded from a holdingpotential of �55 mV elicited at various step potentials (�10,�20, and �30 mV) from atrial myocytes isolated from Cav1.3�/�

and littermates Cav1.3�/� compared with Cav1.3�/� mice. Thetest potentials used are shown to the left of the current traces.The current-voltage relations are summarized in B (n�10 foreach group). C, Voltage-dependent activation curves showingthe normalized conductances (g/gmax) from WT, heterozygous,and mutant animals. The solid lines represent fits to the Boltz-mann function yielding half-activation voltages (V1/2) of0.95�0.40, 6.30�0.40, and 13.75�0.38 mV for Cav1.3�/�,Cav1.3�/�, Cav1.3�/�, and respectively (P�0.05 comparingCav1.3�/� with Cav1.3�/�, Cav1.3�/� with Cav1.3�/�, and Cav1.3�/�

with Cav1.3�/�) and slope factors of 6.8, 7.2, and 7.6 mV(P�NS) for Cav1.3�/�, Cav1.3�/�, and Cav1.3�/�, respectively(n�9 for each group). D, Voltage-dependent inactivation accord-ing to a 2-pulse protocol to examine the voltage- and Ca2�-dependent inactivation of ICa,L in WT, heterozygous, and mutantanimals. Normalized current traces at a test potential of �20mV, comparing ICa,L recorded from Cav1.3�/�, Cav1.3�/�, andCav1.3�/� atrial myocytes. Examples of traces obtained from thetest pulse are shown in the insert. Prepulses more positive than�20 mV elicited a progressively smaller inward current as thecommand voltages approach the reversal potential, leading topartial recovery of the LTCC elicited with the test pulse owing toa decrease in the Ca2�-dependent inactivation.

Figure 6. A, Example of whole-cell ICa,L recorded from a holdingpotential of �55 mV in the same set of experiments as in Figure5 from free-wall left ventricular myocytes, comparing the 3groups of animals. In contrast to the data obtained from theatrial myocytes, there were no significant differences in ICa,L rec-orded from the left ventricular myocytes among the 3 groups ofanimals. The current-voltage relations are summarized in B(n�10 for each group). C shows the corresponding activationcurves and the normalized conductances (g/gmax) fromCav1.3�/�, Cav1.3�/�, and Cav1.3�/� ventricular myocytes. Thesolid lines represent fits to the Boltzmann function yielding half-activation voltages (V1/2) of 9.22�0.42, 9.41�0.60, and 9.5�0.48mV (P�NS) and slope factors of 5.9�0.38, 6.1�0.54, and6.4�0.43 mV for Cav1.3�/�, Cav1.3�/�, and Cav1.3�/�, respec-tively (P�NS, n�9 to 11 for each group). D, Voltage-dependentinactivation according to a 2-pulse protocol to examine thevoltage- and Ca2�-dependent inactivation of ICa,L in WT, het-erozygous, and mutant animals. Normalized current traces at atest potential of �20 mV, comparing ICa,L recorded fromCav1.3�/�, Cav1.3�/�, and Cav1.3�/� ventricular myocytes. Exam-ples of traces obtained from the test pulse are shown in theinsert.

Zhang et al Cav1.3(�1D) Ca2� Channel in Atria 1943

by guest on August 18, 2015http://circ.ahajournals.org/Downloaded from

(Dr Chiamvimonvat). The authors thank Dr E.N. Yamoah for helpfulsuggestions and comments and the UC, Davis, Health SystemConfocal Microscopy Facility.

References1. Catterall WA. Structure and function of voltage-gated ion channels. Annu

Rev Biochem. 1995;64:493–531.2. Walker D, De Waard M. Subunit interaction sites in voltage-dependent

Ca2� channels: role in channel function. Trends Neurosci. 1998;21:148–154.

3. Ertel EA, Campbell KP, Harpold MM, Hofmann F, Mori Y, Perez-ReyesE, Schwartz A, Snutch TP, Tanabe T, Birnbaumer L, Tsien RW, CatterallWA. Nomenclature of voltage-gated calcium channels. Neuron. 2000;25:533–535.

4. Zhang Z, Xu Y, Song H, Rodriguez J, Tuteja D, Namkung Y, Shin HS,Chiamvimonvat N. Functional roles of Cav1.3 (�1D) calcium channel insinoatrial nodes: insight gained using gene-targeted null mutant mice.Circ Res. 2002;90:981–987.

5. Platzer J, Engel J, Schrott-Fischer A, Stephan K, Bova S, Chen H, ZhengH, Striessnig J. Congenital deafness and sinoatrial node dysfunction inmice lacking class D L-type Ca2� channels. Cell. 2000;102:89–97.

6. Mangoni ME, Couette B, Bourinet E, Platzer J, Reimer D, Striessnig J,Nargeot J. Functional role of L-type Cav1.3 Ca2� channels in cardiacpacemaker activity. Proc Natl Acad Sci U S A. 2003;100:5543–5548.

7. Namkung Y, Skrypnyk N, Jeong MJ, Lee T, Lee MS, Kim HL, Chin H,Suh PG, Kim SS, Shin HS. Requirement for the L-type Ca2� channel �1D

subunit in postnatal pancreatic �-cell generation. J Clin Invest. 2001;108:1015–1022.

8. Xu Y, Dong PH, Zhang Z, Ahmmed GU, Chiamvimonvat N. Presence ofa calcium-activated chloride current in mouse ventricular myocytes.Am J Physiol Heart Circ Physiol. 2002;283:H302–H314.

9. Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ. Improvedpatch-clamp techniques for high-resolution current recording from cellsand cell-free membrane patches. Pflugers Arch. 1981;391:85–100.

10. Chiamvimonvat N, O’Rourke B, Kamp TJ, Kallen RG, Hofmann F,Flockerzi V, Marban E. Functional consequences of sulfhydryl modifi-cation in the pore-forming subunits of cardiovascular Ca2� and Na�

channels. Circ Res. 1995;76:325–334.11. Neher E. Correction for liquid junction potentials in patch clamp exper-

iments. Methods Enzymol. 1992;207:123–131.12. Xu Y, Tuteja D, Zhang Z, Xu D, Zhang Y, Rodriguez J, Nie L, Tuxson

HR, Young JN, Glatter KA, Vazquez AE, Yamoah EN, ChiamvimonvatN. Molecular identification and functional roles of a Ca2�-activated K�

channel in human and mouse hearts. J Biol Chem. 2003;278:49085–49094.

13. Ihara Y, Yamada Y, Fujii Y, Gonoi T, Yano H, Yasuda K, Inagaki N,Seino Y, Seino S. Molecular diversity and functional characterization ofvoltage-dependent calcium channels (CACN4) expressed in pancreatic�-cells. Mol Endocrinol. 1995;9:121–130.

14. Mikami A, Imoto K, Tanabe T, Niidome T, Mori Y, Takeshima H,Narumiya S, Numa S. Primary structure and functional expression of thecardiac dihydropyridine-sensitive calcium channel. Nature. 1989;340:230–233.

15. Berul CI, Aronovitz MJ, Wang PJ, Mendelsohn ME. In vivo cardiacelectrophysiology studies in the mouse. Circulation. 1996;94:2641–2648.

16. Mitchell GF, Jeron A, Koren G. Measurement of heart rate and Q-Tinterval in the conscious mouse. Am J Physiol. 1998;274:H747–H751.

17. Kovoor P, Wickman K, Maguire CT, Pu W, Gehrmann J, Berul CI,Clapham DE. Evaluation of the role of IK,ACh in atrial fibrillation using amouse knockout model. J Am Coll Cardiol. 2001;37:2136–2143.

18. Koschak A, Reimer D, Huber I, Grabner M, Glossmann H, Engel J,Striessnig J. �1D (Cav1.3) subunits can form L-type Ca2� channelsactivating at negative voltages. J Biol Chem. 2001;276:22100–22106.

19. Bell DC, Butcher AJ, Berrow NS, Page KM, Brust PF, Nesterova A,Stauderman KA, Seabrook GR, Nurnberg B, Dolphin AC. Biophysicalproperties, pharmacology, and modulation of human, neuronal L-type(�1D, CaV1.3) voltage-dependent calcium currents. J Neurophysiol. 2001;85:816–827.

20. Bohn G, Moosmang S, Conrad H, Ludwig A, Hofmann F, Klugbauer N.Expression of T- and L-type calcium channel mRNA in murine sinoatrialnode. FEBS Lett. 2000;481:73–76.

21. Marionneau C, Couette B, Liu J, Li H, Mangoni ME, Nargeot J, Lei M,Escande D, Demolombe S. Specific pattern of ionic channel geneexpression associated with pacemaker activity in the mouse heart.J Physiol. 2005;562:223–234.

22. Chugh SS, Blackshear JL, Shen WK, Hammill SC, Gersh BJ. Epidemi-ology and natural history of atrial fibrillation: clinical implications. J AmColl Cardiol. 2001;37:371–378.

23. Brugada R, Tapscott T, Czernuszewicz GZ, Marian AJ, Iglesias A, MontL, Brugada J, Girona J, Domingo A, Bachinski LL, Roberts R. Identifi-cation of a genetic locus for familial atrial fibrillation. N Engl J Med.1997;336:905–911.

24. Chen YH, Xu SJ, Bendahhou S, Wang XL, Wang Y, Xu WY, Jin HW,Sun H, Su XY, Zhuang QN, Yang YQ, Li YB, Liu Y, Xu HJ, Li XF, MaN, Mou CP, Chen Z, Barhanin J, Huang W. KCNQ1 gain-of-functionmutation in familial atrial fibrillation. Science. 2003;299:251–254.

25. Ellinor PT, Macrae CA. The genetics of atrial fibrillation. J CardiovascElectrophysiol. 2003;14:1007–1009.

26. Hong K, Bjerregaard P, Gussak I, Brugada R. Short QT syndrome andatrial fibrillation caused by mutation in KCNH2. J Cardiovasc Electro-physiol. 2005;16:394–396.

CLINICAL PERSPECTIVEAtrial fibrillation is the most common atrial arrhythmia affecting the American population and is associated with asignificant risk of embolism and stroke. The problem is further exacerbated by the fact that treatment strategies have provenlargely inadequate. An alteration in ion channel expression (electrical remodeling) has been implicated in the maintenanceof the arrhythmias. Importantly, a decrease in Ca2� current density in atrial myocytes of patients with persistent atrialfibrillation has been documented. Previous data suggest that L-type Ca2� channels containing the Cav1.3(�1D) subunit areexpressed mainly in neurons and neuroendocrine cells, whereas those containing the Cav1.2(�1C) subunit are found in thebrain, vascular smooth muscle, and cardiac tissue. We and others have shown that Cav1.3 Ca2� channel–deficient mice(Cav1.3�/�) demonstrate sinus bradycardia with a prolonged PR interval. This study examined the role of the Cav1.3 (�1D)Ca2� channel in the atria of Cav1.3�/� mice. Here, we demonstrate that there is significant expression of Cav1.3 Ca2�

channels predominantly in atrial compared with ventricular tissues. In addition, Cav1.3-null mutant mice have an increasedsusceptibility to atrial arrhythmias, indicated by inducible atrial flutter and fibrillation. The relevance of this model tohuman atrial fibrillation is speculative at this time, but the important functional role of the Cav1.3 Ca2� channel in atrialtissues in this model warrants additional study to assess its role in humans.

1944 Circulation September 27, 2005

by guest on August 18, 2015http://circ.ahajournals.org/Downloaded from

Glatter, Yanfang Xu, Hee-Sup Shin, Reginald Low and Nipavan ChiamvimonvatZhao Zhang, Yuxia He, Dipika Tuteja, Danyan Xu, Valeriy Timofeyev, Qian Zhang, Kathryn A.

Gene-Targeted Null Mutant Mice) Calcium Channels in Atria: Insights Gained From1Dα1.3(vFunctional Roles of Ca

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 2005 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/CIRCULATIONAHA.105.540070

2005;112:1936-1944; originally published online September 19, 2005;Circulation. 

http://circ.ahajournals.org/content/112/13/1936World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer this process is available in the

click Request Permissions in the middle column of the Web page under Services. Further information aboutOffice. Once the online version of the published article for which permission is being requested is located,

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the EditorialCirculationin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on August 18, 2015http://circ.ahajournals.org/Downloaded from