Accelerated inactivation of cardiac L-type calcium channels triggered by anaesthetic-induced preconditioning

Abstract

Background and purpose:

Cardioprotection against ischaemia by anaesthetic-induced preconditioning (APC) is well established. However, the mechanism underlying Ca²⁺ overload attenuation by APC is unknown. The effects of APC by isoflurane on the cardiac L-type Ca channel were investigated.

Experimental approach:

In a model of in vivo APC, Wistar rats were exposed to isoflurane (1.4%), delivered via a vaporizer in an enclosure, prior to thoracotomy. The Dahl S rats were similarly preconditioned to determine strain-dependent effects. Whole-cell patch clamp using cardiac ventricular myocytes was used to determine the L-type Ca²⁺ current (I_Ca,L) characteristics and calmodulin (CaM) levels were determined by Western blot analysis. Cytosolic Ca²⁺ levels were monitored using fluo-4-AM. Action potential (AP) simulations examined the effects of APC.

Key results:

In Wistar rats, APC significantly accelerated I_Ca,L inactivation kinetics. This was abolished when external Ca²⁺ was replaced with Ba²⁺, suggesting that Ca²⁺-dependent inactivation of I_Ca,L was modulated by APC. Expression levels of CaM, a determinant of I_Ca,L inactivation, were not affected. Attenuation of cytosolic Ca²⁺ accumulation following oxidative stress was observed in the APC group. Simulations showed that the accelerated inactivation of I_Ca,L resulted in a shortening of the AP duration. The Dahl S rat strain was resistant to APC and changes in I_Ca,L inactivation were not observed in cardiomyocytes prepared from these rats.

Conclusions and implications:

APC triggered persistent changes in the inactivation of cardiac L-type Ca channels. This can potentially lead to a reduction in Ca²⁺ influx and attenuation of Ca²⁺ overload during ischaemia/reperfusion.

Introduction

Ischemic preconditioning (IPC) confers cardioprotective effects whereby short, non-lethal ischemic episodes result in a significant reduction in infarct size following a subsequent prolonged ischaemia (Murry et al., 1986). Since IPC is invoked by a short, non-fatal ischemic period, its potential clinical application may be limited due to the difficulty in precisely timing the preconditioning ischemic stimulus. However, several pharmacological agents including volatile anaesthetics are known to mimic the cardioprotective effects of IPC (Cason et al., 1997; Kersten et al., 1997a). The efficacy of anaesthetic-induced preconditioning (APC) is similar to that of IPC, and mechanisms underlying its cardioprotective effects have been intensely investigated. The key components of the intracellular signal transduction pathways include G-proteins, protein kinase C (PKC)-δ and ε isoforms, protein tyrosine kinase, mitogen-activated protein kinase, reactive oxygen species, sarcolemmal and mitochondrial ATP-sensitive potassium channels (sarcK_ATP and mitoK_ATP channels respectively) (Kersten et al., 1997b; Tanaka et al., 2002; Zaugg et al., 2002; Ludwig et al., 2004; da Silva et al., 2004; Marinovic et al., 2006). Recent evidence suggests that the signalling pathways converge at the mitochondria and preserve their function, resulting in the cardioprotective effects afforded by preconditioning (Krolikowski et al., 2005; Riess et al., 2005; Ljubkovic et al., 2007). Nevertheless, key questions still remain unanswered on how the converging signalling cascades affect mitochondrial function and ultimately lead to cardioprotection.

In addition, whether there is a genetic component to APC has not been clearly established. Previous studies have reported species-dependent differences in response to myocardial ischaemia (Galinanes and Hearse, 1990). Furthermore, differential responses to resistance to myocardial ischaemia have been shown in different rat strains (Baker et al., 2000). Thus, a genetic component may also underlie the efficacy of cardioprotection by APC.

A physiologically beneficial outcome of APC that confers cardioprotection appears to be the attenuation of Ca²⁺ overload following ischaemia/reperfusion (Hoka et al., 1987; An et al., 2001; Varadarajan et al., 2002). However, the mechanism underlying this attenuation is unknown. A major source of myocardial Ca²⁺ entry is via the L-type Ca channel. Consequently, inhibition of the L-type Ca channel can result in reduced Ca²⁺ entry, and thus lead to diminished Ca²⁺ overload. On the other hand, Ca²⁺ entry via the L-type Ca channel may also contribute to preconditioning. A small increase in intracellular Ca²⁺ during IPC or Ca²⁺ preconditioning has been reported to trigger PKC-dependent pathways that contributed to the induction of cardioprotection (Miyawaki and Ashraf, 1997). This was abolished by blockade of Ca²⁺ influx via the reverse mode of the Na⁺/Ca²⁺ exchanger and the L-type Ca channel. Furthermore, atrial trabeculae extracted from patients taking L-type Ca channel blockers were resistant to cardioprotection by IPC (Cain et al., 1999). Though an increase in cytosolic Ca²⁺ by IPC would appear to counteract its cardioprotective effect, a rise in cytosolic Ca²⁺ during an initial episode of brief ischaemia can become attenuated in subsequent episodes of ischaemia (Smith et al., 1996). Thus, a transient ischaemic event can lead to a rapid adaptation of Ca²⁺ homeostasis during subsequent episodes.

Based on these previous reports, the L-type Ca channel may potentially play diverse but pivotal functional roles in protecting the myocardium during preconditioning. Ca²⁺ influx may contribute to the underlying cardioprotective signalling cascade, while its attenuation may contribute to a reduction in reperfusion injury. The consequence of a preconditioning stimulus on this channel has not been established. Furthermore, whether APC affects the molecular modulators of Ca channel function has not been investigated. For example, the effect of APC on CaM is unknown even though a major aspect of Ca²⁺-dependent inactivation is the involvement of CaM, through its association with the channel (Dick et al., 2008).

The purpose of this study was to investigate the effect of preconditioning on the cardiac L-type Ca channel current, I_Ca,L. We hypothesized that preconditioning triggers persistent changes in the biophysical profile of the L-type Ca channel and a physiologically relevant model of in vivo APC with isoflurane was utilized to test this hypothesis. Our results showed that APC triggered a persistent change in the inactivation profile of the L-type Ca channel, with a specific acceleration of Ca²⁺-dependent inactivation.

Methods

In vivo anaesthetic-induced preconditioning

This study was approved by the Institutional Animal Care and Use Committee of the Medical College of Wisconsin. Adult male Wistar and Dahl S rats (Harlan, Indianapolis, IN) were used for this study. Two strains of rats were chosen to determine whether the efficacy of APC was dependent on strain, which would be indicative of a genetic component. The rats were divided into two groups, non-APC (control) and APC (in vivo preconditioning). For in vivo APC, rats were exposed to 1.4% (1.0 minimum alveolar concentration) isoflurane (Baxter, Deerfield, IL) delivered via a vaporizer in an enclosure for 30 min followed by a 30 min recovery period prior to thoracotomy (Stadnicka et al., 2006). In the non-APC group, rats were not exposed to isoflurane prior to thoracotomy. This was a physiologically relevant model of preconditioning by APC. Since the rats were exposed to isoflurane in an enclosure, the animals were subjected to all the triggering events induced by the volatile anaesthetic and not only to those confined to the heart. This included any changes in the systemic input and contributions from the vasculature and the endothelium. This contrasts with the method of preconditioning using an isolated heart on a Langendorff apparatus where the preconditioning stimulus is applied at the level of the organ, in the absence of systemic input.

Isolation of cardiac ventricular myocytes

Cardiac ventricular myocytes were enzymatically isolated from adult male Wistar and Dahl S rats weighing 180–280 g as previously described (Aizawa et al., 2004). Briefly, rats were anaesthetized with thiobutabarbital (Inactin, 200 mg·kg⁻¹, i.p.) and heparinized with 1000 U heparin (Baxter). Under anaesthesia, the heart was quickly removed and mounted on a Langendorff apparatus followed by retrograde perfusion through the aorta with a solution containing Joklik medium (Sigma-Aldrich, St. Louis, MO) and heparin at pH 7.23. After allowing for the blood to clear from the heart (for approximately 1–2 min), the perfusate was replaced with an enzyme solution containing Joklik medium, collagenase (Type II; Invitrogen, Carlsberg, CA), protease (Type XIV) and bovine serum albumin (Serologicals Proteins, Kankakee, IL) at pH 7.23. The solutions were gassed with 95% O₂ and 5% CO₂, and the temperature was maintained at 37°C. Following 15–20 min of perfusion with the enzyme solution, the ventricles were cut from the heart, minced and incubated in a shaking bath with the enzyme solution for 5–8 min. The resulting myocyte suspension was then filtered, centrifuged and stored in a modified Tyrode solution at room temperature. The myocytes were used for experiments within 10 h following the isolation procedure.

Solutions

The isolated cardiomyocytes were stored in a modified Tyrode solution that contained (in mmol·L⁻¹): 132 NaCl, 4.8 KCl, 1 CaCl₂, 1.2 MgCl₂, 10 HEPES, 5 dextrose, and pH was adjusted to 7.4 with NaOH. For the recording of whole-cell I_Ca,L, the pipette solution contained (in mmol·L⁻¹): 110 CsCl (substitute for potassium), 1 CaCl₂, 1 MgCl₂, 11 EGTA, 10 HEPES, 5 Mg-ATP, at pH 7.3 adjusted with CsOH. To isolate for I_Ca,L, the bath (external) solution contained (in mmol·L⁻¹): 132 N-methyl-D-glucamine (substitute for sodium), 4.8 CsCl, 2 CaCl₂, 2 MgCl₂, 10 HEPES, 5 dextrose and 5 mmol·L⁻¹ of 4-aminopyridine (transient outward potassium current antagonist), at pH 7.4 adjusted with HCl. For the investigation of the voltage-dependent inactivation and recovery from slow inactivation, CaCl₂ was replaced with BaCl₂, so Ba²⁺ was the charge carrier to remove Ca²⁺-dependent inactivation. All chemicals used in this study were obtained from Sigma-Aldrich (St. Louis, MO), unless otherwise noted.

Oxidative stress

A cell survival study was conducted to validate the in vivo APC model by comparing the tolerance against oxidative stress between the non-APC and APC myocytes (Marinovic et al., 2006). The cell suspension was transferred to a chamber on an inverted microscope (Eclipse TE2000-U; Nikon, Tokyo, Japan) and the myocytes were allowed to settle for 10 min, after which they eventually attached to the glass coverslip. They were then exposed to 0.4% Trypan blue solution for 2 min followed by washout with glucose-free Tyrode solution. Cells that were rod-shaped and not stained were considered to be viable and this number was counted. Following a 5 min perfusion with the glucose-free Tyrode solution, the myocytes were subjected to oxidative stress by perfusion with 200 µmol·L⁻¹ H₂O₂ and 100 µmol·L⁻¹ FeSO₄ for 17 min. Subsequently, the myocytes were reperfused with glucose-free Tyrode solution for another 20 min. The myocytes were exposed to Trypan blue again and the living myocytes were counted to determine the percentage of cell death. For each group, 3–4 rats were utilized, and 1–3 experiments were conducted from myocytes isolated from each animal. The percentage of cell death was calculated from at least 200 myocytes for each experiment. For the survival study, the n number denotes the number of experiments that were conducted.

Electrophysiology

The whole-cell configuration of the patch clamp technique was used to record I_Ca,L. Patch pipettes were pulled from borosilicate glass capillary tubes (Garner Glass, Claremont, CA) using a horizontal micropipette puller (P-97; Sutter Instrument, Novato, CA), and heat polished using a microforge (MF-830; Narishige, Tokyo, Japan). The pipette resistances were in the range of 2–5 MΩ. I_Ca,L was monitored using a patch clamp amplifier (Axopatch 200B; Molecular Devices, Sunnyvale, CA) interfaced with a digitizer (Digidata 1322A; Molecular Devices) to a computer. Data acquisition and analysis were conducted using pClamp 9.2 (Molecular Devices), and additional analyses were performed using Origin 7 (OriginLab, Northampton, MA). All the experiments were performed under room temperature. For the patch clamp studies, 38 rats were utilized (19 each for the non-APC and APC groups). The n number denotes the number of myocytes from which the current recordings were obtained.

L-type Ca²⁺ current measurement

I_Ca,L was elicited by 400 ms duration test pulses from a holding potential of −80 to +50 mV in 10 mV increments at a rate of 0.33 Hz. A conditioning pulse (to −50 mV; 50 ms) was used to inactivate any residual cardiac sodium current before each test pulse. I_Ca,L was normalized to cell capacitance and the resultant current density was plotted against the test potentials to yield the current-voltage relationship.

Steady-state activation curve

The conductance was calculated with the following equation, G = I/(V_m − V_rev), where I is the current amplitude, V_m is the test potential and V_rev is the reversal potential. The corresponding steady-state activation curves were obtained by normalizing the conductance to the peak conductance. The normalized conductances (G/G_max) were then plotted against the test potentials and fitted with the Boltzmann function,

where V_1/2 is the voltage at half-maximal conductance and k is the slope factor.

Steady-state inactivation curve

The steady-state inactivation curves were obtained using a standard protocol, in which 500 ms conditioning pulses from a holding potential of −80 mV to +20 mV in 10 mV increments were followed by a 600 ms test pulse to 0 mV. Current amplitude was normalized to the peak current (I/I_max), and plotted against each conditioning potential. The steady-state inactivation curves were also fitted with the Boltzmann function,

to obtain V_1/2, the voltage at which half the channels are available for opening, and k, the slope factor.

Recovery from fast and slow inactivation

A standard two-pulse protocol was used to monitor the recovery of I_Ca,L from fast inactivation. A 500 ms conditioning pulse to 0 mV was followed by a 600 ms test pulse to 0 mV in varying intervals from 0 to 550 ms in 50 ms increments. The holding potential was set at −80 mV. Current amplitude elicited by the test pulses were normalized to the peak current elicited by the conditioning pulse. The normalized currents were plotted against each time interval, and fitted with a single exponential function that yielded the time constant for recovery from fast inactivation.

To investigate the recovery from slow inactivation, Ba²⁺ was used as the charge carrier. The Ba²⁺ current, I_Ba, was elicited using a two-pulse protocol, in which a 5 s conditioning pulse to 0 mV was followed by a 50 ms test pulse to 0 mV in varying intervals from 1 to 15 s in 1 s increments from a holding potential of −80 mV. Current amplitude obtained during the test pulse was normalized to the peak current obtained during the conditioning pulse and plotted against the varying time intervals. The resultant plot was then fitted with a double exponential function to obtain the fast and slow time constants (τ_fast and τ_slow respectively) for the recovery from slow inactivation.

Ca²⁺ current inactivation kinetics and transmembrane charge transfer

I_Ca,L inactivation kinetics were quantified by fitting the current traces with a double exponential function of the form

where τ_fast and τ_slow are the fast and slow time constants, respectively, A_fast and A_slow are the respective amplitudes of the time constants, and C is the final amplitude following the inactivation. The fraction of channels that gated either fast or slow was quantified by the respective amplitudes with the following equations:

In experiments where the voltage-dependent inactivation of I_Ca,L was investigated, Ba²⁺ substituted for Ca²⁺ as the charge carrier in the external solution to remove Ca²⁺-dependent inactivation and the Ba²⁺ current (I_Ba) was recorded.

To correlate the changes in Ca²⁺ influx to the changes in I_Ca,L inactivation kinetics, the transmembrane charge transfer was determined. This was calculated by integrating the area of I_Ca,L traces after the capacitive transients to the time point of 100 ms. This temporal duration was chosen to mimic the time course of a rat action potential at room temperature (Linz and Meyer, 2000).

Western blotting

Western blot analysis was carried out to determine the cytosolic levels of calmodulin (CaM). Ventricles were excised from both non-APC and APC hearts, and homogenized on ice in a solution that contained (in mmol·L⁻¹): 20 Tris-HCl, 150 NaCl, 1 EDTA, 1 EGTA, 1% Triton X-100, and cocktails of protease inhibitors (complete mini; Roche, Indianapolis, IN) and phosphatase inhibitors (cocktail set II; Calbiochem, San Diego, CA), pH 7.4. The homogenate was centrifuged for 10 min at 10 000× g (4°C), and the resulting supernatant was centrifuged for 1 h at 100 000× g (4°C) to separate the cytosolic and membrane fractions. The total protein concentration was determined by a modified Lowry method (Bio-Rad, Hercules, CA). Equal amounts of proteins (100 µg) were separated on a 15% polyacrylamide gel and transferred to the nitrocellulose membrane (Bio-Rad). The membrane was blocked with 5% fat-free milk and incubated overnight at 4°C with a rabbit monoclonal CaM antibody (Epitomics, Burlingame, CA) at 1:1000 dilution. After being washed, the membrane was incubated for 1 h at room temperature with a horseradish peroxidase-conjugated anti-rabbit immunoglobulin G (Bio-Rad) at 1:10 000 dilution. Chemiluminescence was detected after incubation with Luminol/Enhancer solution (Pierce, Rockford, IL) using radiographic film. To verify equal protein loading, the membrane was re-probed with a mouse monoclonal anti-β-actin antibody at 1:1000 dilution followed by incubation with a horseradish peroxidase-conjugated anti-mouse immunoglobulin G (Bio-Rad) at 1:5000 dilution and developed on film. CaM/β-actin ratio was quantified using the ImageJ software (National Institutes of Health, Bethesda, MD).

Cytosolic Ca²⁺ measurement

A Ca²⁺-sensitive indicator, fluo-4-AM (2 µmol·L⁻¹; Invitrogen, Carlsbad, CA) that preferentially loaded into the cytosol was used to assess cytosolic Ca²⁺. Images were obtained with a confocal microscope (Eclipse TE2000-U) equipped with a 40×/1.3 oil-immersion objective. Excitation at 488 nm was achieved with an argon laser. After 20 min of dye loading, followed by a 5 min dye washout, cells were exposed to an oxidative stress protocol identical to that used in the cell survival study described above. Images were taken every 5 min. Data were analysed with the Metamorph 6.1 software (Universal Imaging, West Chester, PA).

Simulation of the action potential

Simulations of the cardiac AP were conducted using Cellular Open Resource (COR), a modelling environment developed by the Oxford Cardiac Electrophysiology Group in the Department of Physiology at the University of Oxford (Great Britain) (Garny et al., 2003). Specifically, simulations were run using the Fox-McHarg-Gilmour (FMG) model on COR (Fox et al., 2002). This model was chosen for its ease of use in determining the effect of changes in the L-type Ca channel's Ca²⁺-dependent inactivation on the AP profile. To simulate an AP with the time course that mimicked that of a rat cardiac AP at room temperature, the FMG model was modified by adjusting the amplitudes of the following potassium (K) currents: I_to (transient outward K current), I_Kr (rapidly activating delayed rectifier K current), I_Kp (plateau K current) and I_K1 (inward rectifier K current). Contributions from I_Ks (slowly activating delayed rectifier K current) were eliminated. In addition, since the original FMG model included an enhanced Ca²⁺-dependent inactivation of the L-type Ca channel, modifications were made to incorporate the ‘basal’ rate of Ca²⁺-dependent inactivation as described by the phase-2 Luo-Rudy model (Luo and Rudy, 1994). Specifically, the Ca²⁺-dependent inactivation gate at steady state was defined by

where [Ca²⁺]_i is the intracellular Ca²⁺ concentration, and K_mfCa is the half-saturation constant. For the initial conditions, K_mfCa was set at 0.60 µmol·L⁻¹ (Luo and Rudy, 1994).

Statistical analysis

Data are shown as mean ± SEM. One-way ANOVA followed by Scheffe's test was used to compare the data among the three groups in the cell survival study. The unpaired Student's t-test was used for comparisons between the non-APC and APC groups in the patch clamp studies and Ca²⁺ measurements. Statistical analyses were performed using the Origin 7 software and P < 0.05 was considered significantly different.

Results

Validation of the in vivo APC model: protection against oxidative stress

To validate the effectiveness of the in vivo APC model, a cell survival study was conducted to determine the efficacy of in vivo APC in inducing tolerance against oxidative stress injury on isolated myocytes obtained from the Wistar rats. Myocytes from the non-APC and APC groups were subjected to oxidative stress as described in Methods. As a time control, myocytes were perfused for 42 min with glucose-free Tyrode solution. As shown in Figure 1, in the time control group, the per cent cell death was low after this perfusion. In the non-APC group, oxidative stress resulted in a markedly increased proportion of dead cells and exposure to the APC protocol significantly attenuated this high percentage of cell death in myocytes following oxidative stress. These results using the Wistar rats confirmed that in vivo APC was very effective in decreasing cell death following oxidative stress. This cardioprotection persisted past the cell isolation procedure.

Figure 1.

Effects of in vivo APC on cardiomyocyte survival after oxidative stress. Myocytes were isolated from Wistar rat hearts. In the non-APC group, oxidative stress significantly increased the percentage of cell death compared with the time control (TC) group. This effect of oxidative stress was significantly attenuated by in vivo APC. *P < 0.05 vs. TC, ^†P < 0.05 vs. non-APC. n = 7–10 per group.

Effects of in vivo APC on I_Ca,L steady-state activation, inactivation and recovery from inactivation

To characterize the effects of APC on I_Ca,L, whole-cell current recordings were made from isolated myocytes obtained from Wistar rats that had been previously preconditioned by isoflurane via the in vivo APC protocol. The I_Ca,L current-voltage relationship, the steady-state activation and inactivation curves, and the recovery from fast and slow inactivation were determined. Figure 2A depicts representative I_Ca,L traces recorded from ventricular myocytes isolated from a control (non-APC) rat heart. The current-voltage relationships obtained from the non-APC and APC groups suggested that the voltage dependence of activation was unaltered by APC (Figure 2B). The peak I_Ca,L densities were also not significantly different between the two groups with −7.6 ± 0.4 pA·pF^–1 in the non-APC and −8.8 ± 0.5 pA·pF^–1 in the APC groups (n = 21–22 per group). The absence of a significant shift in the voltage dependence of activation was confirmed by constructing the steady-state activation curve (Figure 2C, Table 1). In addition, the steady-state inactivation curves obtained from the non-APC and APC groups were not significantly different (Figure 2D, Table 1). Moreover, recovery from fast inactivation of I_Ca,L was not affected by in vivo APC (Figure 2E, Table 1). As I_Ca,L also exhibits slow inactivation, the effect of APC on recovery from slow inactivation was determined. As was the case for recovery from fast inactivation, APC had no significant effect on recovery from slow inactivation (Figure 2F, Table 1).

Figure 2.

Effects of in vivo APC on the biophysical profile of I_Ca,L obtained in myocytes from non-APC and APC hearts. (A) Representative whole-cell I_Ca,L recorded from an isolated cardiomyocyte obtained from a non-APC heart. Arrow indicates zero current level. (B) Current-voltage relationship. The current-voltage relationship for the non-APC and APC groups are shown. No significant differences were observed (n = 21–22 per group). (C) Steady-state activation curves. Data were fitted with the Boltzmann function. No significant shifts in the steady-state activation curves were observed between the two groups (n = 21–22 per group). (D) Steady-state inactivation curves. Similarly to (C), data were fitted with the Boltzmann function. No significant shifts in the steady-state inactivation curves were observed between the two groups (n = 22 per group). (E) Recovery from fast inactivation. The inset depicts the standard two-pulse protocol that was utilized as described in Methods (n = 19 per group). (F) Recovery from slow inactivation. The inset depicts the two pulse protocol that was utilized as described in Methods (n = 4 per group). There were no significant differences in the recoveries from fast and slow inactivations between the non-APC and APC groups.

Table 1.

The effects of in vivo APC on the steady-state activation and inactivation curves, and recovery from fast and slow inactivation of I_Ca,L

	non-APC	APC	N per group
Steady-state activation
V1/2 (mV)	−16.4 ± 0.6	−17.8 ± 0.9	21–22
k	4.9 ± 0.2	4.6 ± 0.2
Steady-state inactivation
V1/2 (mV)	−31.0 ± 0.5	−31.7 ± 0.5	22
k	5.1 ± 0.1	4.8 ± 0.1
Recovery from fast inactivation
τ (ms)	91.8 ± 3.4	86.7 ± 4.0	19
Recovery from slow inactivation
τfast (s)	0.47 ± 0.03	0.42 ± 0.04	4
τslow (s)	6.4 ± 1.8	6.5 ± 1.1

Data are presented as mean ± SEM.

Effects of in vivo APC on I_Ca,L inactivation kinetics and transmembrane charge transfer

Though in vivo APC did not result in any significant persistent changes in the steady-state activation and inactivation curves, and in the recovery from fast and slow inactivation, it did significantly accelerate I_Ca,L inactivation kinetics, as demonstrated in Figure 3A. Specifically, τ_fast in the voltage range of −20 to +10 mV, and τ_slow in the voltage range from −20 to +40 mV were significantly decreased compared with the non-APC group (n = 21–22 per group; Figure 3B). Additional analysis of I_Ca,L inactivation, fitted with the double exponential function (see Methods), revealed that APC induced a greater fraction of channels to favour fast inactivation as determined by the relative fraction of channels that gated fast (Figure 3c). This was accompanied by a corresponding decrease in the relative fraction of channels that gated slow (data not shown). These results were not affected by the time interval between cell isolation and the electrophysiological recordings, and persisted during the 10 h period the myocytes were used.

Figure 3.

Effects of in vivo APC on voltage- and Ca²⁺-dependent inactivation of the cardiac L-type Ca channel. (A) Sample whole-cell I_Ca,L traces recorded from myocytes in the non-APC and APC groups monitored at 0 mV from a holding potential of −80 mV are shown. The current traces were superimposed. Arrow indicates zero current level. (B) The fast and slow time constants of the inactivation kinetics of I_Ca,L in myocytes from the non-APC and APC groups are plotted against test potentials. Compared with the non-APC group, APC significantly decreased τ_fast in the range from −20 to +10 mV and τ_slow in the range from −20 to +40 mV. (C) The fraction of channels that ‘fast inactivated’ was determined from the double exponential fit to the inactivating I_Ca,L traces. In the APC group, the fraction in the fast inactivated state was significantly increased compared to the non-APC group. (D) Changes in I_Ca,L inactivation was correlated to changes in the total influx of charge by integrating the area of I_Ca,L traces during a 100 ms period. The charge transfer was significantly decreased by APC in the range from −20 to +30 mV. *P < 0.05 vs. non-APC. n = 21–22 per group.

In order to quantify the change in ionic influx due to the acceleration of I_Ca,L inactivation, the transmembrane charge transfer was determined. To correlate Ca²⁺ influx during a rat ventricular AP, the charge transfer was calculated during 100 ms. As summarized in Figure 3D, a significant decrease in charge transfer resulted from the APC-triggered acceleration of I_Ca,L inactivation.

As the inactivation of I_Ca,L consists of voltage- and Ca²⁺-dependent components, experiments were conducted to determine which of the components, if not both, were affected by APC. The result above showed an increase in the fraction of channels in the fast inactivated state, which suggested that the effect of APC was likely to be on the Ca²⁺-dependent inactivation of the channel. Consequently, experiments were conducted to test this hypothesis. Ba²⁺ was utilized as the charge carrier to remove Ca²⁺-dependent inactivation. Under this condition, the inactivation of I_Ca,L was due to the voltage-dependent mechanism. As shown in Figure 4A, I_Ba exhibited markedly slower inactivation kinetics as expected. In contrast to its effect on I_Ca,L, APC did not affect I_Ba inactivation kinetics (Figure 4B). Under our recording conditions, inactivation of I_Ba was best fitted by a double exponential function (Findlay, 2002). This indicated that APC accelerated the Ca²⁺-dependent inactivation of I_Ca,L, but had no effect on voltage-dependent inactivation.

Figure 4.

Effects of in vivo APC on the voltage-dependent inactivation of the cardiac L-type Ca channel. (A) Representative traces of I_Ca,L and I_Ba recorded at 0 mV from a holding potential of −80 mV using Ca²⁺ and Ba²⁺ as charge carriers, respectively, are shown. The currents were scaled and superimposed. Arrow indicates zero current level. Ba²⁺ as the charge carrier removed Ca²⁺-dependent inactivation of the L-type Ca channel. (B) The fast and slow time constants of the inactivation kinetics of I_Ba in the non-APC and APC groups are plotted against test potentials. No significant differences were observed in τ_fast and τ_slow between the non-APC and APC groups (n = 10 per group).

Cytosolic CaM levels

A major aspect of Ca²⁺-dependent inactivation is the involvement of CaM which associates with the channel. To investigate the mechanism of APC-induced changes in I_Ca,L inactivation kinetics, changes in the expression of cytosolic CaM were determined by Western blot analysis. CaM expression levels were compared between the non-APC and APC groups (n = 6 hearts per group). As shown in Figure 5, the cytosolic CaM expression levels, normalized to β-actin, were unchanged by APC. This suggested that APC did not induce a change in the expression levels of CaM.

Figure 5.

Cytosolic calmodulin (CaM) levels. Effects of in vivo APC on cytosolic CaM levels were determined by Western blot. (A) Representative results from one heart each from the non-APC and APC groups are shown. β-actin was used as the loading control. (B) Summary of the CaM/β-actin ratio in the non-APC and APC groups are shown. No significant differences were observed between two groups (n = 6 hearts per group).

Attenuation of cytosolic Ca²⁺ accumulation

Based on our results, in vivo APC significantly improved cell survival and also triggered the acceleration of I_Ca,L inactivation kinetics that could lead to diminished Ca²⁺ influx. However, the question of whether APC did indeed result in the attenuation of cytosolic Ca²⁺ accumulation in our model still remained. This was addressed by monitoring cytosolic Ca²⁺ with confocal microscopy. Using an identical oxidative stress protocol as in the cell survival study described above, cytosolic Ca²⁺ was monitored by loading isolated myocytes with the fluo-4-AM fluorescent indicator and monitored over time. Rats in the in vivo APC group were exposed to isoflurane prior to thoracotomy as described in Methods. As shown in Figure 6, cytosolic Ca²⁺ accumulation was observed in both the non-APC and APC groups following oxidative stress. However, in myocytes isolated from the APC group, this accumulation of Ca²⁺ was significantly attenuated as compared with that of the non-APC group. These results demonstrated that myocytes from the APC group were better protected against stress-induced Ca²⁺ accumulation.

Figure 6.

Cytosolic Ca²⁺ accumulation under stress conditions. Isolated cardiomyocytes were loaded with the cytosolic Ca²⁺ indicator, fluo-4-AM, and exposed to oxidative stress, followed by perfusion with glucose-free Tyrode solution. Fluo-4 fluorescence was monitored over time. Myocytes from both the non-APC and APC groups showed increases in cytosolic Ca²⁺ following stress as indicated by the increases in fluo-4 fluorescence. However, the increase in fluo-4 fluorescence was significantly reduced in the APC-treated myocytes. *P < 0.05 vs. non-APC. n = 10–12 per group.

Action potential simulation

To characterize the effects of accelerated I_Ca,L inactivation kinetics on the AP, simulations were conducted. Simulations were run using a modified FMG model of the cardiac AP as described in Methods. The resultant simulation is depicted in Figure 7A. As expected, the I_Ca,L activation and inactivation kinetics during an AP with dynamic changes in the membrane potential differed from those observed during a rectangular test pulse. In the modified FMG model, based on our experimental results, the descending phase of the I_Ca,L inactivation was accelerated by 23% in the APC-simulated I_Ca,L compared with the control-simulated I_Ca,L. This change in I_Ca,L inactivation kinetics resulted in a limited shortening of the APD, as shown in Figure 7B. In the control-simulated AP, the APD₅₀ and APD₉₀ were 83.0 and 98.1 ms, respectively, while in the APC-simulated AP, they were 77.8 and 92.9 ms.

Figure 7.

Simulation of the action potential. (A) Using a modified FMG model as described in Methods, inactivation kinetics were accelerated by 23% in the simulated APC I_Ca,L compared with the Control I_Ca,L. (B) Simulated changes in the AP (control vs. APC) are shown as the corresponding consequence of the accelerated I_Ca,L inactivation. The APD₅₀ and APD₉₀ were 83.0 and 98.1 ms, respectively, in the control-simulated AP, and 77.8 and 92.9 ms in the APC-simulated AP.

Although changes in CaM levels did not appear to account for the accelerated inactivation kinetics based on our Western blot analysis, AP simulations were run to test whether changes in Ca²⁺ sensitivity of the L-type Ca channel can lead to the observed changes in inactivation. To accommodate this change in Ca²⁺ sensitivity, the half-saturation constant for Ca²⁺ in the modified FMG model was adjusted (see Methods). We found that arbitrarily decreasing the half-saturation constant by approximately 28%, from 0.60 to 0.43 µmol·L⁻¹, accelerated the inactivation of I_Ca,L by only 5% (Figure 8).

Figure 8.

Effect of Ca²⁺ sensitivity on I_Ca,L inactivation. The effect of altering the sensitivity of L-type Ca channels to intracellular Ca²⁺ on I_Ca,L inactivation was simulated using the modified FMG model. Ca²⁺ sensitivity was enhanced by 28% (dashed lines) relative to the control. The exponential fits are depicted by the dotted lines. This enhancement of Ca²⁺ sensitivity induced only 5% change in the acceleration of the I_Ca,L inactivation.

Rat strain-dependent effects

Previous studies have reported on rat strain-dependent differences in resistance to myocardial ischaemia, indicative of a genetic component to cardioprotection (Baker et al., 2000). However, rat strain-dependent differences for APC have not been established. We conducted survival studies on cardiomyocytes from the Dahl S strain of rats and found that they showed resistance to the cardioprotective effects of in vivo APC (n = 5–9 per group; Figure 9a). This contrasted the effects of APC on the Wistar rats (Figure 1). Subsequently, to confirm that the observed changes in I_Ca,L inactivation were correlated with APC, we profiled the I_Ca,L inactivation in myocytes isolated from the Dahl S rats. As shown in Figure 9 (panels B and C), APC failed to alter I_Ca,L inactivation kinetics in myocytes obtained from the Dahl S rats (n = 9 per group). Thus, in a rat strain that is resistant to APC, changes in I_Ca,L inactivation were not evident.

Figure 9.

Effects of in vivo APC on the Dahl S strain of rats. (A) Cell survival study. Oxidative stress significantly increased the percentage of cell death in the non-APC group compared with the time control (TC) group. In the APC group, cell survival following oxidative stress was not significantly different from the APC group, indicating resistance to cardioprotection in the Dahl S rats. *P < 0.05 vs. TC. n = 5–9 per group. (B) Representative whole-cell I_Ca,L traces from Dahl S myocytes in the non-APC and APC groups. Currents were recorded at a test potential of 0 mV from a holding potential of −80 mV. The traces shown were superimposed. Arrow indicates zero current level. (C) I_Ca,L inactivation kinetics. The fast and slow time constants of current inactivation in myocytes from the non-APC and APC groups showed no significant differences in the Dahl S rats (n = 9 per group).

Discussion and conclusions

In the present study, we found that in vivo APC by isoflurane accelerated I_Ca,L inactivation kinetics, while all other measured biophysical characteristics of the L-type Ca channel remained unchanged. Although the acute inhibitory effects of volatile anaesthetics on the cardiac L-type Ca channel have been previously reported (Bosnjak et al., 1991; Pancrazio, 1996; Camara et al., 2001; Suzuki et al., 2002), this is the first report on an APC-triggered persistent change in a biophysical property of the channel. The voltage-dependent inactivation of the channel was not affected by in vivo APC, suggesting that the induced change in inactivation kinetics was due to the effects of APC on Ca²⁺-dependent inactivation. In addition, these observed changes in I_Ca,L inactivation in the Wistar rats were not evident in the Dahl S rats that were resistant to APC. Thus, the changes in I_Ca,L inactivation observed in the Wistar rats are a consequence of APC.

The mechanism underlying this observation is unknown. Based on the biophysical analysis, APC induced the L-type Ca channels to favour the fast inactivated state. Key components of Ca²⁺-dependent inactivation of the L-type Ca channel are modulated by CaM, whereby binding of Ca²⁺ to CaM that is bound to the carboxy-terminal tail of the α-subunit of the channel is a likely necessary step (Peterson et al., 1999; Van Petegem et al., 2005; Zhou et al., 2005). Using a general approach, we hypothesized that APC triggered changes in the CaM levels that can be linked to changes in the L-type Ca channel inactivation. Our results showed that APC did not induce changes in cytosolic CaM levels. Consequently, the accelerated inactivation of I_Ca,L was not due to changes in the expression levels of CaM. However, this does not exclude potential changes in the functional role of CaM in the inactivation of I_Ca,L. For example, changes in the Ca²⁺ binding to the C-lobe of CaM, purported to be a required step for Ca²⁺-dependent inactivation (Peterson et al., 1999; Alseikhan et al., 2002), can potentially be modified by APC. Additionally, since binding of apoCaM to the IQ motif of the L-type Ca channel is also a critical component of inactivation (Tang et al., 2003), changes in this interaction could also influence Ca²⁺-dependent inactivation.

Although the net physiological consequence of the APC-triggered change in I_Ca,L inactivation was not established, it can potentially contribute to reduced Ca²⁺ influx and to changes in cardiac electrophysiology. Regarding the former, since the L-type Ca channel is a major route of Ca²⁺ influx into the cytosol, the acceleration of I_Ca,L inactivation kinetics can potentially contribute to a decrease in Ca²⁺ overload, and subsequently reduce mitochondrial Ca²⁺ overload, and prevent or delay apoptosis. Our results showed a significant decrease in the total charge transfer as a result of the accelerated I_Ca,L inactivation by APC. Studies have indeed demonstrated that APC significantly reduced cytosolic Ca²⁺ concentration, relative to non-preconditioned hearts following global ischaemia (An et al., 2001; Varadarajan et al., 2002), and attenuated cytochrome c release from mitochondria (Qian et al., 2005). These effects can be correlated with a reduction in infarct size and with improved functional recovery following ischaemia/reperfusion.

In addition to the L-type Ca channel, contributions from other sources of Ca²⁺, for example, the roles of the Na⁺/Ca²⁺ exchanger and the ryanodine receptors on the sarcoplasmic reticulum, and the buffering capacity of the mitochondria, must also be considered in determining the precise mechanism(s) underlying Ca²⁺ overload attenuation, induced by APC. Our results showed that cytosolic Ca²⁺ accumulation was decreased following oxidative stress in myocytes obtained from preconditioned hearts. Although this demonstrated the functional efficacy of our APC model, the contribution from the L-type Ca channel in attenuating Ca²⁺ overload was uncertain as the Ca²⁺ measurements were conducted under in vitro conditions in non-contracting myocytes. Consequently, other mechanisms are likely to be involved. On the other hand, in a beating heart, cytosolic Ca²⁺ accumulation following stress might be greater due to influx of Ca²⁺ via the L-type Ca channel, which is a major source of Ca²⁺ entry into the cytosol. Thus, one can speculate that the acceleration of I_Ca,L inactivation kinetics triggered by APC would be likely to contribute to decreased Ca²⁺ accumulation in contracting myocytes. Overall, our results suggested that the preconditioned myocytes were better able to handle Ca²⁺ accumulation following stress.

With regard to cardiac electrophysiology, though the cardioprotective effect of APC is well established, in terms of reduction in infarct size, its effect against arrhythmias has not been firmly documented. For example, studies using the isolated heart model of APC showed that sevoflurane can delay the onset of ventricular fibrillation or have no effect on this type of arrhythmia (Novalija et al., 1999; Kevin et al., 2003). Based on our simulation studies, the accelerated I_Ca,L inactivation would result in the shortening of the cardiac APD. It is well established that APC enhances the opening of the sarcK_ATP channel and this effect can also lead to the shortening of the APD. Consequently, the acceleration of I_Ca,L inactivation could potentiate this effect, resulting in a further shortening of the APD and triggering re-entrant arrhythmias. Hence, a critical balance between the sarcK_ATP channel opening and acceleration of I_Ca,L inactivation may underlie some of the confusion regarding the effects of APC on cardiac rhythm. The overall effect on cardiac electrophysiology, thus, will depend on the cumulative effects the APC-triggered signalling cascade would have on the various sarcolemmal ion channels. Furthermore, it may also depend on the timing of the opening of the sarcK_ATP channel. Currently, the magnitude and duration of the activation of the sarcK_ATP channel during the various phases of preconditioning (stimulus, washout, ischaemia and reperfusion) are unknown, although a recent study suggested that the sarcK_ATP channels were more likely to open during the ischaemia/reperfusion period and contribute to cardioprotection (Marinovic et al., 2006; Stadnicka et al., 2006).

One intriguing but remarkable aspect of our findings is that only one biophysical characteristic of the L-type Ca channel, the rate of inactivation, was affected by APC. All other measured parameters remained unchanged. The direct, acute application of volatile anaesthetics is known to inhibit I_Ca,L amplitude and thus to depress cardiac contractility (Hanley et al., 2004). However, this inhibition is absent in the persistent effect triggered by isoflurane preconditioning. This change in the inactivation kinetics may allow for the fine tuning of Ca²⁺ entry without affecting the contractile ability of the cardiac myocytes. Furthermore, a recent study proposed that accelerating I_Ca,L inactivation while preserving peak I_Ca,L may be an effective way to prevent arrhythmias (Mahajan et al., 2008).

To date, studies systematically exploring the effect of APC on the cardiac sarcolemmal voltage-gated ion channels have been limited, despite the strong probability that the signalling mechanism triggered by APC can modulate the function of these proteins. In pursuit of the mechanism underlying APC, studies have focused on the intracellular signalling cascade and on mitochondrial function (Bienengraeber et al., 2005). With respect to the role of ion channels in APC, those not gated by voltage, specifically the sarcK_ATP and mitoK_ATP channels, have been intensely investigated. The roles of other ion channels, particularly those on the mitochondria such as the voltage-dependent anion and the calcium activated potassium channels and the adenine nucleotide translocase, have yet to be established.

In conclusion, the results of this study showed that in vivo APC triggered the acceleration of I_Ca,L inactivation kinetics in cardiomyocytes. Voltage-dependent inactivation of I_Ca,L was not affected, suggesting the involvement of Ca²⁺-dependent inactivation. Although a direct physiological consequence of this change in inactivation has yet to be determined, this alteration in the L-type Ca channel may contribute to the preservation of cardiomyocytes against Ca²⁺ overload and subsequent cell death.

Glossary

Abbreviations:

AP

action potential

APC

anaesthetic-induced preconditioning

APD

action potential duration

CaM

calmodulin

I_Ca,L

L-type calcium current

IPC

ischemic preconditioning

τ

time constant

Conflicts of interest

None.