Impaired Ca²⁺ homeostasis is associated with atrial fibrillation in the α_1D L-type Ca²⁺ channel KO mouse

Abstract

The novel α_1D Ca²⁺ channel together with α_1C Ca²⁺ channel contribute to the L-type Ca²⁺ current (I_Ca-L) in the mouse supraventricular tissue. However, its functional role in the heart is just emerging. We used the α_1D gene knockout (KO) mouse to investigate the electrophysiological features, the relative contribution of the α_1D Ca²⁺ channel to the global I_Ca-L, the intracellular Ca²⁺ transient, the Ca²⁺ handling by the sarcoplasmic reticulum (SR), and the inducibility of atrial fibrillation (AF). In vivo and ex vivo ECG recordings from α_1D KO mice demonstrated significant sinus bradycardia, atrioventricular block, and vulnerability to AF. The wild-type mice showed no ECG abnormalities and no AF. Patch-clamp recordings from isolated α_1D KO atrial myocytes revealed a significant reduction of I_Ca-L (24.5%; P < 0.05). However, there were no changes in other currents such as I_Na, I_Ca-T, I_K, I_f, and I_to and no changes in α_1C mRNA levels of α_1D KO atria. Fura 2-loaded atrial myocytes showed reduced intracellular Ca²⁺ transient (∼40%; P < 0.05) and rapid caffeine application caused a 17% reduction of the SR Ca²⁺ content (P < 0.05) and a 28% reduction (P < 0.05) of fractional SR Ca²⁺ release in α_1D KO atria. In conclusion, genetic deletion of α_1D Ca²⁺ channel in mice results in atrial electrocardiographic abnormalities and AF vulnerability. The electrical abnormalities in the α_1D KO mice were associated with a decrease in the total I_Ca-L density, a reduction in intracellular Ca²⁺ transient, and impaired intracellular Ca²⁺ handling. These findings provide new insights into the mechanism leading to atrial electrical dysfunction in the α_1D KO mice.

the voltage-gated α_1d Ca²⁺ channel was previously considered to be expressed only in cells of neuroendocrine origin (10). Recent studies (11, 15, 16, 19, 26) reported that the α_1D deletion in mice causes sinus bradycardia and various degrees of atrioventricular (AV) block. The evidence presented suggests a critical role of the α_1D Ca²⁺ channel in diastolic depolarization and in the rate of discharge of the sinoatrial (SA) node. The expression of α_1D Ca²⁺ channel was demonstrated in the SA node, AV node, and atria, but not in the ventricles of adult hearts (15, 20, 26). Therefore, deletion of α_1D Ca²⁺ channel could play an important role in altering the normal activity of the atrium and might also contribute to atrial arrhythmias, such as atrial fibrillation (AF).

AF is the most common cardiac arrhythmia, and it is associated with a four- to fivefold increase in the risk of stroke compared with the general population. It affects an estimated 2.2 million adults in the United States (4). Because of its clinical relevance and the lack of an effective therapy, many experimental models of AF have been tested to elucidate the mechanisms of initiation and maintenance of AF. A decreased L-type Ca²⁺ current (I_Ca-L) density is a common finding and has been reported in several models of AF (23, 25).

Unlike ventricular cells, a unique property of atrial cells is that both α_1C and α_1D Ca²⁺ channels contribute to the total I_Ca-L (26). While a decrease in I_Ca-L density and a reduction in mRNA and protein level of the α_1C Ca²⁺ channel in AF have been reported (2, 12), the role of α_1D Ca²⁺ channel in AF remains unclear. However, one study (5) using gene microarray approach demonstrated a significant reduction in α_1D Ca²⁺ channel mRNA (as well as a reduction of α_1C Ca²⁺ channel mRNA) in atrial samples from patients with AF, supporting the potential role of α_1D L-type Ca²⁺ channel in the genesis of AF.

Because α_1C and α_1D Ca²⁺ currents cannot be separated in the native atrial myocytes using conventional electrophysiological and/or pharmacological tools, the α_1D Ca²⁺ channel knockout (KO) mouse thus represents a unique model to assess the relative contribution of α_1D Ca²⁺ channel to atrial myocyte electrophysiology. In this study, we used the α_1D Ca²⁺ channel KO mouse model, to investigate the electrophysiological features, the relative contribution of α_1D Ca²⁺ channel to the total I_Ca-L, the intracellular Ca²⁺ transient ([Ca²⁺]_iT), and the Ca²⁺ handling by the sarcoplasmic reticulum (SR) to ultimately gain insight into the mechanism leading to atrial dysfunction in this mouse model.

MATERIALS AND METHODS

The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (National Institutes of Health Publication No. 85-23, revised 1996). The experimental protocols were approved by the Animal Care and Use Committee at the Veterans Affairs New York Harbor Healthcare System (New York, NY).

Six- to ten-week-old α_1D Ca²⁺ channel KO mice and age-matched wild-type (WT) littermates were used in this study. The α_1D KO mice were kindly provided by Dr. R. Flavell (Yale University, School of Medicine, New Haven, CN) with the concurrence of Dr. J. Striessnig (Peter Mayr Strasse, Innsbruck, Austria). The generation of α_1D KO has been previously described (19).

Surface ECG recordings.

High resolution surface ECG was recorded and analyzed using a digital acquisition and analysis system (Biopac Systems, Santa Barbara, CA). Mice were placed on a warm pad and subjected to anesthetic inhalation; gauze pledged was moistened with isoflurane (3–4%) and placed in the bottom of a drop jar under a fume hood. After the mouse was anesthetized, it was removed and a mask was used to maintain anesthesia with 2% isoflurane (mix of isoflurane and oxygen). Electrodes were positioned on the sole of each mouse foot. Electrical signals were recorded at 1,200 Hz stored in a computer hard disk and analyzed off-line (AcqKnowledge software, Biopac Systems). Tracings were analyzed for heart rate, P-wave duration, P-wave amplitude, P-R interval, QRS duration, and conduction abnormalities including first-, second-, and third-degree AV block.

Gene expression analysis for α_1C and α_1D transcripts using real-time RT-PCR.

Total cellular RNA was isolated from atrial tissue of WT and α_1D KO mice using Trizol LS Reagent (Invitrogen) according to the manufacturer's recommendations. RNA was quantified by spectrophotometry at 260 nm, and the ratio of absorbance at 260 nm to that of 280 nm was >1.8 for all samples. Degradation of RNA was monitored by the observation of appropriate 28S to 18S ribosomal RNA ratios as determined by ethidium bromide staining of agarose gels. First-strand cDNA synthesis was carried out using RETROscript reverse transcriptase kit (Ambion). Gene-specific intron-spanning Taqman primers (Applied Biosystems) were used to quantify the relative changes in mRNA levels of α_1C and α_1D in the WT and α_1D KO mice (Hs99999901_s1, Mm00437917_m1, Mm00551384_m1) using ABI PRISM 7500 real-time PCR (Applied Biosystems). All real-time PCR reactions were done in triplicates, and the cycle threshold (C_T) values were normalized against the endogenous control 18S.

Epicardial electrograms from isolated hearts.

Hearts were removed and quickly cannulated through the aorta and Langendorff-perfused with Krebs-Henseleit buffer solution containing the following (in mM): 118 NaCl, 4.7 KCl, 1.2 KH₂PO₄, 1.2 MgSO₄, 1.8 CaCl₂, 25 NaHCO₃, and 11.1 glucose, (pH 7.4), bubbled with 95% O₂-5% CO₂ at 35 ± 1 ^°C. A bimodular amplifier (Biopac MP System) was used to record atrial and ventriclar epicardial electrical activity from a total of six chlorinated electrodes placed on the heart, as previously described (9). A third module was a stimulator controlled by a computer to deliver programmed fast pacing stimuli with S₁ cycle lengths ranging from 80 to 40 ms followed by three extra-stimuli S₁-S₂-S₃ delivered at a coupling interval of 30–20 ms. Stimuli were delivered to the right atrium to induce AF (1).

Atrial myocytes isolation.

Atrial cells were isolated from Langendorff-perfused hearts. The heart was perfused with nominally Ca²⁺-free Tyrode solution containing the following (in mM): 137 NaCl, 5.4 KCl, 1 MgCl₂, 0.33 NaH₂PO₄, 10 HEPES, and 10 glucose (pH 7.4) equilibrated with 100% O₂ (35 ± 1^°C). After the blood was washed out, collagenase type-2 (1 mg/ml; Worthington, Biochemical), protease type XIV (0.02 mg/ml; Sigma), and elastase (0.2 mg/ml; Worthington, Biochemical) were added to the perfusion solution. When the hearts became dilated, the left ventricle was removed and the myocytes were allowed to disperse in solution. Ventricular myocytes were subjected to Ca²⁺ readaptation and finally resuspended in Tyrode solution (1 mM CaCl). The right atrium was quickly removed and allowed to digest for an additional 5–15 min in the solution containing the enzymes. The tissue was transferred into the KB-recovery solution containing the following: 70 mM/l l-glutamic acid, 20 mM/l KCl, 80 mM/l KOH, 10 mM/l (±)-β-OH-butyric acid, 10 mM/l KH₂PO₄, 10 mM/l taurine, 1 mg/ml BSA, and 10 mM/l HEPES-K (adjusted to pH 7.4 with KOH). After 40 min, cells were dispersed in the same solution. Na⁺ and Ca²⁺ were reintroduced by the addition of a solution containing the following: 10 mM NaCl, 1.8 mM CaCl₂, and 1 mg/ml BSA. Cells were finally stored in a solution containing the following: 100 mM NaCl, 35 mM KCl, 1.3 mM CaCl₂, 0.7 mM MgCl₂, 14 mM l-glutamic acid, 2 mM β-OH butyric acid, 2 mM KH₂PO₄, 2 mM taurine, and 1 mg/ml BSA (pH 7.4) and were used within 4 h after the isolation (15).

Patch-clamp recordings.

Whole cell configuration of the patch-clamp technique was used for ion current recordings. Cells were first superperfused with Tyrode solution, and then the perfusion was switched to the appropriate solution for each current studied as previously described (7). All experiments were conducted at room temperature. Briefly, for I_Ca-L and I_Ca-T, Na⁺ and K⁺ were substituted by equimolar concentration of tetraethylammonium (TEA⁻) and cesium (Cs⁺) respectively. Niflumic acid (30 μM) and 4-aminopyridine (4-AP; 1 mM) were added to block chloride and K⁺ currents; cobalt (5 mM) was added during I_Ca-T recording (27). Pipette solution for I_Ca-L and I_Ca-T recordings contained (in mM): 110 Cs-aspartate, 20 CsCl, 1 MgCl₂, 10 EGTA, 10 HEPES, 5 Mg-AT, and 0.5 Tris-GTP (pH 7.4) with CsOH. The I_Ca-L was activated by a series of 250-ms depolarization pulses from −90 mV holding potential (HP) to test potentials ranging from −50 mV to +60 (10-mV step) at 10-s intervals. The I_Ca-T was elicited by 250-ms depolarizing pluses (10-mV step) from a HP of −90 mV to +60 mV. The external solution for I_Na recordings contained (in mM): 60 N-methyl glucosamine, 60 NaCl, 10 CsCl, 1.8 CaCl₂, 1 MgCl₂, 10 HEPES, and 10 glucose (pH 7.4, with CsOH). I_Ca-L and I_Ca-T were blocked by CoCl₂ (5 mM) and NiCl₂ (1 mM), respectively. The internal solution contained (in mM): 135 CsOH, 135 l-aspartic acid, 1 MgCl₂, 10 EGTA, 10 HEPES, 5 Mg-ATP, and 0.1 Na-GTP (pH 7.2, CsOH). I_Na was evoked with 30-ms duration pulses to +30 mV from a HP of −90 mV at 5-s intervals.

For I_to, I_f, and I_K current recording, the extracellular solution was adjusted as follows. The I_K and the I_to were blocked with 4-AP (0.5 mM); the I_Ca-L was blocked with verapamil (10 μM). I_K was elicited by step membrane depolarization from −50 mV to +60 mV. I_to was studied in the presence of TEA-Cl and CdCl₂ (200 μM), and current was elicited by step depolarization from −50 mV to +60 mV. The hyperpolarization-activated current (I_f) was studied by adding BaCl₂ (1 mM) to the external solution, and the membrane was then hyperpolarized from −20 mV to −130 mV. The standard pipette solution contained (in mM): 110 K-aspartate, 20 KCl, 2 CaCl₂, 1 MgCl₂, 0.1 GTP, 5 Mg₂-ATP, 5 Na₂-phosphocreatine, 10 EGTA, and 10 HEPES (pH 7.3, with KOH). Data were sampled with an A/D converter (Digital 1320A, Axon Instruments) and stored on the hard disk of a computer for subsequent analysis.

Measurement of [Ca²⁺]_iT and SR Ca²⁺ loading.

The [Ca²⁺]_iT was recorded in isolated right atrial myocytes loaded with fura 2-AM (5 μM, 15 min; Invitrogen). Measurements started at least 30 min after fura 2 was removed to allow the complete dye deesterification. Drops containing cells were placed in a perfusion chamber mounted in an inverted microscope (Nikon) and were continuously superfused with Tyrode solution at 35^°C. The [Ca²⁺]_iT was evoked by electrical field-stimulation (square pulses 3–5 ms duration at rate of 0.5 Hz). Fluorescence was measured using a photomultiplier (DeltaRam, Photon Technology International).

Caffeine was used to release SR Ca²⁺ and thus estimate SR Ca²⁺ content. The Na⁺ and Ca²⁺ were not included in the caffeine solution to minimize extrusion of Ca²⁺ by the Na⁺/Ca²⁺ exchanger. Caffeine was applied locally by pressure ejection through a pipette positioned near the cell (200 μm) after the cells were electrically stimulated at steady state to allow homogeneous (SR) Ca²⁺ loading.

Statistical analyses.

Data are means ± SD. Experimental data from the α_1D KO mice and WT controls were compared by unpaired Student's t-test. A value of P < 0.05 was considered to be statistically significant; n represents the number of cardiomyocytes or hearts.

RESULTS

ECG recordings.

Surface ECGs were obtained from a total of 15 α_1D KO and 15 WT anesthetized mice. Representative ECG recordings are illustrated in Fig. 1. The recording from the WT mouse shows a sinus rhythm with regular P-QRS complex (Fig. 1A). However, the ECG from the α_1D KO mice shows sinus bradycardia and blocked atrial impulses, which failed to propagate to the ventricle (AV block; Fig. 1B). Averaged data demonstrated significant sinus bradycardia in the α_1D KO mice (46.1% ± 15.4 reduction in beats/min, P < 0.01). In the α_1D KO mice, ECG analysis shows that P-wave amplitude was depressed and prolonged and that P-R interval was also prolonged (Table 1). The mean QRS duration was not different in the two groups. Three out of 15 α_1D Ca channel KO mice showed spontaneous episodes of atrial arrhythmia. These spontaneous episodes were not seen in any of the WT mice.

Fig. 1.

Representative surface ECG recordings from anesthetized mice. A: wild-type (WT) mouse ECG, showing sinus rhythm (heart rate 571 beats/min) with a regular P-QRS complex. B: ECG from α_1D Ca²⁺ channel knockout (KO) mouse with pronounced sinus bradycardia (375 beats/min). Arrowheads illustrate atrial activity failing to propagate into the ventricle [atrioventricular (AV) block].

Table 1.

Electrocardiographic parameters in anesthetized WT and α_1D channel KO mice

Parameters	WT	KO
Heart rate, beats/min	551.2+58.6	303.5+72.4*
P-R interval, ms	33.5+2.02	45.1+2*
P-wave duration, ms	25.3+1.0	30.8+1.0*
P-wave amplitude, mV	2.24+0.03	1.26+0.06*

Data are means ± SD. WT, wild type; KO, knockout.

^*P < 0.05.

Electrograms from isolated hearts and induction of AF in α_1D KO mice.

To rule out potential interferences from neural innervations and/or anesthesia, we conducted studies in isolated Langendorff-perfused hearts. All the atrial abnormalities observed in vivo were still present in isolated α_1D Ca²⁺ channel KO mice hearts (n = 5), but none were found in WT mice hearts (n = 5), suggesting that the electrical abnormalities are intrinsic to the heart and reinforcing the hypothesis that α_1D Ca²⁺ channel is necessary for the normal genesis and conduction of the electrical impulse. Sinus bradycardia was detected in the α_1D Ca²⁺ KO hearts (205 ± 17 beats/min; n = 5) compared with WT mice hearts (309 ± 20 beats/min; n = 8; P < 0.01).

Programmed electrical stimulation (burst stimulation and extrasystolic beats, combined) was delivered in the right atria to induce AF. AF was readily inducible in all α_1D Ca²⁺ channel KO hearts but in none of the WT mice hearts. Representative ECG traces from a WT mouse heart in sinus rhythm before and after burst stimulation is shown in Fig. 2A. In contrast, application of the same burst stimulation protocol resulted in inducible AF/tachycardia (Fig. 2B) in α_1D Ca²⁺ channel KO mice hearts (5 out of 5). During AF, the atrial rate was much faster than the ventricular rate, with a median fibrillation interval of ∼60 milliseconds (Fig. 2B). Irregular and slow ventricular rhythm can be also observed during AF (Fig. 2B). The episodes of AF spontaneously converted after 125 ± 66 s to sinus rhythm.

Fig. 2.

Epicardial ECG and rapid pacing (burst) in isolated hearts applied at the S1 cycle length of 40 ms followed by 3 extra-stimuli S1-S2-S3 at 30 ms. Recordings electrodes were placed in each heart chamber free walls, and stimulator electrodes are placed in the right atria. A: representative epicardial ECG recording from a WT mouse heart, sinus rhythm is evident before and after the burst stimulation. B: burst stimulation of the heart from α_1D Ca²⁺ channel KO mouse resulted in induced AF and irregular ventricular response. Expanded view of the uncoordinated atrial activities is shown in inset.

Patch-clamp recordings from atrial cells.

To determine the relative contribution of the α_1D Ca²⁺ channel to the total I_Ca-L in the atria, right atrial cells were isolated and the whole cell recording of I_Ca-L was performed as shown in Fig. 3A. The capacitance of atrial cells from the KO mice cells was smaller compared with the WT cells [42.13 ± 2.22 pA/pF in KO (n = 26) vs. 50.75 ± 3.25 pA/pF in WT (n = 26); P = 0.05]. The I-V relationship for I_Ca-L revealed a depolarization shift (13 mV) in the I_Ca-L peak density in the KO mice cells (Fig. 3B). Atrial myocytes from the KO mice demonstrated a reduced I_Ca-L peak density (at +10 mV) compared with WT [3.51 ± 0.3 pA/pF in KO (n = 18) vs. 4.65 ± 0.2 pA/pF in WT (n = 25); P < 0.05]. This finding indicates that the I-V relationship generated from α_1D Ca²⁺ channel KO atrial cell is represented only by the α_1C subunit of the L-type Ca²⁺ channel. Interestingly, the I-V relationship from ventricular cells did not show any changes in the I_Ca-L density between the WT and the α_1D Ca²⁺ channel KO mice (Fig. 3C).

Fig. 3.

Membrane ion currents in single cardiomyocytes isolated from the right atria of WT and KO mice. A: selected I_Ca-L tracings of atrial myocytes from WT (left) and α_1D Ca²⁺ channel KO (right) mice. B: current-voltage (I-V) relationship showing a reduction of I_Ca-L density in α_1D Ca²⁺ channel KO (▪) atrial myocytes compared with the WT (○). The I-V peak is shifted rightward in α_1D Ca²⁺ channel KO cells (∼13 mV). C: I-V relationship of α_1D Ca²⁺ channel I_Ca-L density from ventricular myocytes of the WT (○) and KO (▪) mice. The voltage protocol used to generate the I-V relationship is shown in the inset. *P < 0.05.

To investigate whether the deletion of α_1D Ca²⁺ channel may have resulted in changes (i.e., compensatory mechanisms) of other major atrial ion currents (17, 21), we recorded I_Na, I_Ca-T, I_K, I_f, and I_to from both groups (Table 2). Current densities were not significantly different between WT and α_1D Ca²⁺ channel KO mice, indicating that deletion of α_1D Ca²⁺ channel did not result in compensation from the major currents we measured (Fig. 4A). Because the deletion of α_1D Ca²⁺ channel might result in altered expression of the α_1C Ca²⁺ channel (16), we used real-time PCR to assess the levels of α_1C Ca²⁺ channel mRNA. The transcript for α_1D was detected in the atria of the WT but not in the α_1D Ca²⁺ channel KO mice. The deletion of the α_1D Ca²⁺ channel gene did not result in any significant changes in the α_1C Ca²⁺ channel mRNA level in α_1D Ca²⁺ channel KO mice (Fig. 4B). This further supports the conclusion that the decrease in I_Ca-L observed in the α_1D Ca²⁺ channel mice was attributable to the deletion of α_1D Ca²⁺ channel.

Table 2.

Averaged ion current densities of atrial cells from WT and α_1D Ca²⁺ channel KO mice

Density Peak at mV	WT	KO	P Value
ICaL at +10	4.65+0.2 (25)	3.51+0.3*(18)	P=0.039
ICaT at −10	1.24+0.2 (5)	1.1+0.3 (4)	P=0.95
INa at −20	33.8+3.6 (6)	29.80+3.2 (6)	P=0.86
IK at +60	5.32+0.54 (8)	4.17+0.44 (8)	P=0.50
Ito at +60	8.12+0.85 (9)	8.52+1.85 (8)	P=0.96
If at −110	3.31+0.73 (3)	4.30+0.88 (4)	P=0.71

Data are means ± SD; no. in parenthesis are no. of animals.

^*P < 0.05.

Fig. 4.

Bar graphs representing averaged current densities and mRNA levels in right atrial cells from α_1D Ca²⁺ channel (KO) and WT mice. A: summary of several ion current densities from atrial myocytes of KO and WT mice. Only I_Ca-L density is significantly (P < 0.05) reduced. B: real-time PCR analysis for the relative quantification of α_1D and α_1C Ca²⁺ channel mRNA in WT and α_1D Ca²⁺ channel KO mice atria. α_1D mRNA level is minimal in the α_1D Ca²⁺ channel KO mice compared with the WT mice. There were no significant changes in α_1C mRNA level in α_1D Ca²⁺ channel KO compared with the WT. *P < 0.05.

[Ca²⁺]_i measurements in atrial myocytes.

The [Ca²⁺]_iT was monitored in single right atrial cells loaded with fura 2, and [Ca²⁺]_iT was evoked by electrical field stimulation. Representative traces of a single atrial cell from a WT and KO stimulated until steady state are shown (Fig. 5A). A significant reduction of the [Ca²⁺]_iT peak amplitude was observed in all of the α_1D KO mice atrial cells compared with the WT atrial cells (40.48 ± 5.02%; n = 13; P < 0.05). This reduction was associated with an increase of the time to peak without alteration of the [Ca²⁺]_iT decay time (Fig. 5C).

Fig. 5.

Representative intracellular Ca² ([Ca²⁺]_i) transient from fura 2 loaded cells electrically stimulated at 0.5 Hz. A: [Ca²⁺]_i transient original records from WT and α_1D Ca²⁺ channel KO atrial cells. B: normalized and superimposed individual [Ca²⁺]_i transients from a WT and from α_1D Ca²⁺ channel KO show a reduced systolic [Ca²⁺]_i peak in α_1D Ca²⁺ channel KO atrial cells. C: averaged data for [Ca²⁺]_i transient amplitude, time to peak, and time to 50% relaxation. *P < 0.05.

SR Ca²⁺ load and function in atrial myocytes.

Since the SR Ca²⁺ store is the main Ca²⁺ contributor to the [Ca²⁺]_iT in mice (22), we investigated whether the decrease in systolic Ca²⁺ might reflect a decrease in SR Ca²⁺ content. Rapid caffeine application (10 mM) to empty the intracellular Ca²⁺ store in single atrial cells revealed a significant reduction of the SR Ca²⁺ content in the α_1D Ca²⁺ channel KO cells compared with the WT cells (17 ± 2% reduction; n = 8; P < 0.05; Fig. 6A). Superimposed individual caffeine-induced [Ca²⁺]_iT from a WT and from a KO cell indicates that the SR Ca²⁺ content in KO cells is less than in WT (Fig. 6B). The average data of caffeine-induced Ca²⁺ release peak are shown in Fig. 6C. To assess whether the Ca²⁺-induced Ca²⁺-release (CICR) might be affected by the α_1D Ca²⁺ channel gene deletion, we measured fractional SR Ca²⁺ release, defined as the electrically induced Ca²⁺ released during a single twitch and the total Ca²⁺ released by caffeine application ([Ca²⁺]_iT twitch/[Ca²⁺]_iT,caffeine). The average fractional Ca²⁺ release was significantly smaller in the α_1D KO cells (28 ± 3%; n = 6; P < 0.05; Fig. 6D). This result suggests an impairment of the CICR in the α_1D Ca²⁺ channel KO atrial cells.

Fig. 6.

Sarcoplasmic (SR) Ca²⁺ content measured by caffeine application. A: original traces from WT and α_1D Ca²⁺ channel KO atrial cells showing SR Ca²⁺ content measured by caffeine induced Ca²⁺ release. Cells were electrically field-stimulated followed by local caffeine pulse (10 mM). B: superimposed individual caffeine-induced Ca²⁺ transient from WT and α_1D Ca²⁺ channel KO cells showing that the SR Ca²⁺ content in KO cells is less than in WT. C: average data of peak caffeine-induced Ca²⁺ release. D: SR fractional Ca²⁺ release. Data are normalized to facilitate comparison. *P < 0.05.

DISCUSSION

Our results show that α_1D Ca²⁺ channel deletion leads to abnormal electrical activity and AF vulnerability. Furthermore, the results show a reduction of I_Ca-L density, a reduced intracellular [Ca²⁺]_iT amplitude, and impaired CICR in the α_1D Ca²⁺ channel KO mouse atrial myocytes. These findings may account for the electrocardiographic abnormalities, including AF vulnerability found in these mice.

Electrocardiographic abnormalities of the α_1D Ca²⁺ channel KO mice.

During ECG recordings in KO mice, electrocardiographic abnormalities such as sinus bradycardia, P-wave changes, and AV block were present. The reduced P-wave amplitude and prolonged P-wave duration suggest slower atrial impulse propagation in the α_1D Ca²⁺ channel KO mice atria. Similar ECG abnormalities were found in isolated Langendorff-perfused hearts. Furthermore, isolated hearts from α_1D KO mice were prone to AF/tachycardia upon burst stimulation of the right atrium. This evidence suggests that the increased incidence of AF in the α_1D Ca²⁺ channel KO paced heart may be related to intrinsic abnormalities of atrial electrical properties such as intra-atrial or interatrial conduction disturbances. Altogether, the present findings suggest that the α_1D Ca²⁺ channel plays an important role in the maintenance of normal atrial electrical activity.

Ion currents in the α_1D Ca²⁺ channel KO atrial cells.

Reduction of the global I_Ca-L, which is represented only by α_1C in the α_1D Ca²⁺ channel KO mouse, indirectly shows that the α_1D Ca²⁺ channel actively contributes to the Ca²⁺ influx in atrial myocytes and the CICR mechanism. The patch-clamp results unmasked the different biophysical characteristics between α_1C and α_1D channels. The latter activates at more negative potentials, consistent with the properties of α_1D Ca²⁺ channel characterized in heterologous expression systems (11, 24). The α_1C Ca²⁺ channel mRNA expression is not altered in the α_1D Ca²⁺ channel KO mice atria, and as indicated by real-time PCR data (Fig. 4B), deletion of α_1D Ca²⁺ channel did not result in α_1C transcript level changes. Moreover I_Na, I_Ca-T, I_K, I_f, and I_to current densities are not affected by the deletion of the α_1D Ca²⁺ channel in the atrial cells, suggesting no compensatory mechanisms. Altogether, the patch-clamp data further support the unique role of α_1D Ca²⁺ channel in the atria. The reduction of I_Ca-L density in the α_1D Ca²⁺ channel KO mice is consistent with a reduced total I_Ca-L density in the SA node cells of the same α_1D Ca²⁺ channel KO mice (15) with no compensatory changes from other currents but is in contrast with others (26) showing no changes in total I_Ca-L of the α_1D Ca²⁺ channel in KO mice atrial cells. The use of different stimulation/conditioning voltage-clamp protocols used to evoke I_Ca-L and the differences in the genetic background (18) of the mice used between the two studies may account for the different findings.

[Ca²⁺]_iT in the α_1D Ca²⁺ channel KO atrial cells.

To date, there are no data available regarding the potential contribution of α_1D Ca²⁺ channel to the [Ca²⁺]_iT in single atrial myocytes. The I_Ca-L reduction in the α_1D Ca²⁺ channel KO mice may account, at least in part, for the significant reduction of the [Ca²⁺]_iT in single atrial cells. Since the main source of Ca²⁺ during CICR is the SR, the smaller I_Ca-L cannot be solely responsible for the large [Ca²⁺]_iT decrease. In fact, our results show a concomitant SR Ca²⁺ load reduction and a smaller fractional SR Ca²⁺ release. The results obtained indicate that the α_1D Ca²⁺ channel deletion contributes significantly to the [Ca²⁺]_iT and that perturbations of α_1D Ca²⁺ channel can lead to abnormal electrophysiological function of the atria. Moreover, the absence of the α_1D Ca²⁺ channel may result in an inefficient CICR. Giving the interplay between Ca²⁺ cycling and cellular electrophysiology in arrhythmogenesis, this animal model serves as a valuable research tool for further investigations.

Pathophysiological significance.

Unlike ventricular myocytes, atrial cells do not possess an extensive T-tubule system (8, 13) but exhibit additional ryanodine receptor clusters around the periphery of the atrial cells (3, 6, 14). Two types of ryanodine receptors are found in atrial myocytes: junctional ryanodine receptors, which are located just beneath the sarcolemma and in close proximity to L-type Ca²⁺ channels, and nonjunctional ryanodine receptors, which are deeper inside the cell. As a consequence (the close apposition of L-type Ca²⁺ channels and ryanodine receptors), triggering of CICR occurs only at the cell periphery. Because cardiac ryanodine receptors are under local control of a single L-type Ca²⁺ channel, a change in properties of the L-type channel may have profound consequences on the normal function of CICR. In this study, the important role of the sarcolemmal α_1D Ca²⁺ channel to the atrial Ca²⁺ homeostasis and to the electrical activity of the atria has been emphasized. The genetic deletion of α_1D Ca²⁺ channel affected time to peak (slower time to peak) of [Ca²⁺]_iT, the amplitude of [Ca²⁺]_iT (reduced amplitude), the Ca²⁺ content of the SR (reduced SR Ca²⁺ content), and the fractional Ca²⁺ release (reduced), all of which are expected to affect Ca²⁺ homeostasis. The reduced SR Ca²⁺ load and unchanged decay of the caffeine-induced Ca²⁺ transient suggest that triggered activity/automaticity are unlikely mechanisms for the increased propensity to atrial arrhythmias. The reduced P-wave amplitude and prolonged P-wave duration, which suggest slower conduction in the α_1D Ca²⁺ channel KO mice atria, and the reduced I_Ca,L would suggest that reentry susceptibility may be a likely mechanism. Unlike the commonly reported reduction of I_Ca,L current density due to remodeling after AF has taken place, the present data support the hypothesis that an a priori reduction in I_Ca,L current density may be involved in the initiation of AF, since deletion of the α_1D Ca²⁺ channel makes the mice prone to AF.

Altogether, the data from the present study open new directions into the characteristic features of the α_1D Ca²⁺ channel in the supraventricular tissue, which may be useful for the development of potential atrial specific therapeutics targeted towards the management of atrial arrhythmias especially AF.