Colchicine modulates calcium homeostasis and electrical property of HL‐1 cells

Abstract Colchicine is a microtubule disruptor that reduces the occurrence of atrial fibrillation (AF) after an operation or ablation. However, knowledge of the effects of colchicine on atrial myocytes is limited. The aim of this study was to determine if colchicine can regulate calcium (Ca2+) homeostasis and attenuate the electrical effects of the extracellular matrix on atrial myocytes. Whole‐cell clamp, confocal microscopy with fluorescence, and western blotting were used to evaluate the action potential and ionic currents of HL‐1 cells treated with and without (control) colchicine (3 nM) for 24 hrs. Compared with control cells, colchicine‐treated HL‐1 cells had a longer action potential duration with smaller intracellular Ca2+ transients and sarcoplasmic reticulum (SR) Ca2+ content by 10% and 47%, respectively. Colchicine‐treated HL‐1 cells showed a smaller L‐type Ca2+ current, reverse mode sodium–calcium exchanger (NCX) current and transient outward potassium current than control cells, but had a similar ultra‐rapid activating outward potassium current and apamin‐sensitive small‐conductance Ca2+‐activated potassium current compared with control cells. Colchicine‐treated HL‐1 cells expressed less SERCA2a, total, Thr17‐phosphorylated phospholamban, Cav1.2, CaMKII, NCX, Kv1.4 and Kv1.5, but they expressed similar levels of the ryanodine receptor, Ser16‐phosphorylated phospholamban and Kv4.2. Colchicine attenuated the shortening of the collagen‐induced action potential duration in HL‐1 cells. These findings suggest that colchicine modulates the atrial electrical activity and Ca2+ regulation and attenuates the electrical effects of collagen, which may contribute to its anti‐AF activity.


Introduction
Atrial fibrillation (AF) is the commonest form of sustained arrhythmia and increases the risk of stroke, heart failure and mortality [1,2]. However, AF rhythm control agents are not very useful due to a high recurrence of AF, low adherence and common adverse effects [3,4]. Colchicine, a drug widely used to treat gouty arthritis, was found to reduce the occurrence of post-operative AF or early AF recurrence after pulmonary vein isolation in the absence of antiarrhythmic drug treatment [5,6]. Colchicine has a potent anti-inflammatory effect and decreases the occurrence of pericarditis [7][8][9], both critical in the pathogenesis of AF. Moreover, colchicine can directly modulate calcium (Ca 2+ ) homeostasis in cardiomyocytes [10]. However, it is not clear if colchicine can regulate atrial electrical activity to reduce AF genesis [11].
Colchicine is a microtubule disruptor that inhibits microtubule assembly in cells and inhibits neutrophil and endothelial cell adhesion molecules [5,12], which play roles in the pathophysiology of AF [13]. Microtubules, a major cytoskeletal component of cardiomyocytes, are linked to various pathological conditions and may play a role in modulating both the electrical and mechanical activities of the heart. Moreover, microtubule assembly increases the deposition of extracellular matrix (ECM) proteins, especially type I collagen [14,15]. Higher concentrations of ECM proteins enhance the occurrence of AF through electrical and structural remodelling [16,17]. Colchicine decreases interstitial myocardial fibrosis by interfering with collagen accumulation and reverses the contractile function in failing hearts [18,19]. Therefore, colchicine may attenuate myocardial remodelling through decreasing ECM accumulation and prevent AF occurrence. The purpose of this study was to investigate the effects of colchicine on Ca 2+ homeostasis and electrophysiological characteristics in atrial myocytes and find out if colchicine can modulate collagen-induced action potential (AP) changes.

Materials and methods
Cell culture HL-1 cells derived from mouse atrial cardiac muscle cells (kindly provided by Dr. Claycomb, Louisiana State University Health Sciences Center, New Orleans, LA, USA) were cultured in a humidified atmosphere of 5% CO 2 at 37°C in Claycomb medium (JRH Biosciences, Lenexa, KS, USA). HL-1 cells were seeded in fibronectin/gelatin-precoated dishes as described previously. Cells were plated at a density of 5 9 10 5 cells/well in 6-well culture plates. At 24 hrs after cell seeding, we added colchicine (3 nM, which is within the therapeutic concentration) [20] or a vehicle solution to cells for a further 24 hrs as colchicine-treated or control cells, respectively. We also studied the effect of colchicine on collagen-treated HL-1 cells. We incubated HL-1 cells with collagen (10 mg/ml) and colchicine (3 nM) for a further 24 hrs after cell seeding.

Measurement of intracellular Ca 2+
As described previously [21], HL-1 cells were loaded with fluorescent Ca 2+ (10 lmol/l) fluo-3/acetoxymethyl (AM) ester for 30 min. at room temperature. Excess extracellular dye was removed by changing the bath solution, and intracellular hydrolysis of fluo-3/AM occurred at 35 AE 1°C after 30 min. Fluo-3 fluorescence was excited with a 488-nm line of an argon ion laser. The emission was recorded at >515 nm. Cells were repetitively scanned at 3-ms intervals for a total duration of 6 sec. Fluorescence imaging was performed with a laser scanning confocal microscope (Zeiss LSM 510, Carl Zeiss, Jena, Germany) and an inverted microscope (Axiovert 100, Carl Zeiss, Jena, Germany). The fluorescent signals were corrected for variations in the dye concentration by normalizing the fluorescence (F) against the baseline fluorescence (F 0 ) to obtain reliable information about intracellular (cytoplasm) Ca 2+ ([Ca 2+ ] i ) transient changes from baseline values (F/F 0 ) and to exclude variations in the fluorescence intensity with different volumes of injected dye. [Ca 2+ ] i transients and peak systolic and diastolic [Ca 2+ ] i values were measured during 2-Hz field stimulation with 10-ms twice-threshold strength square-wave pulses.

APs and ionic currents
A whole-cell patch clamp was used for single isolated cardiomyocytes with an Axopatch 1D amplifier (Axon Instruments, Foster City, CA, USA) at 35 AE 1°C [26]. Borosilicate glass electrodes (o.d., 1.8 mm) with tip resistances of 3~5 MΩ were used. Before the formation of the membrane-pipette seal, the tip potentials were zeroed in Tyrode's solution. Junction potentials between the bath and pipette solution (9 mV) were corrected for AP recordings. APs were recorded in the current-clamp mode, and ionic currents were measured in the voltage-clamp mode. The ionic currents and AP were recorded at approximately similar period (3-5 min.) after membrane rupture or perforation by amphotericin B (for I Ca-L ) to avoid decay of ion channel activity over time. A small hyperpolarizing step from a holding potential of À50 mV to a test potential of À55 mV for 80 ms was delivered at the beginning of each experiment. The area under the capacitative current was divided by the applied voltage step to obtain the whole-cell capacitance. The series resistance (R s ) was compensated by 60~80%. APs were elicited from isolated cardiomyocytes without spontaneous activity at a driven rate of 1 Hz for 20 beats. The resting membrane potential (RMP) was measured during the period between the last repolarization and the onset of the subsequent AP. The AP amplitude (APA) was obtained from the RMP to the peak of AP depolarization. AP durations (APDs) at 90%, 50% and 20% repolarization were, respectively, measured as the APD 90 , APD 50 and APD 20 . Micropipettes were filled with a solution containing (in mM) KCl 20, K aspartate 110, MgCl 2  The I Ca-L was recorded using 300-ms pulses from a holding potential of À50 mV to inactivate the T-type Ca 2+ current [27], to test potentials that varied between À40 and +60 mV in 10-mV increments at a frequency of 0.1 Hz using a perforated patch clamp with amphotericin B. In the external solution, NaCl and KCl of the normal Tyrode's solution were respectively replaced with TEACl and CsCl.
The NCX current was obtained as the nickel-sensitive current by subtracting the current in the presence of 10 mM NiCl 2 from that of the control. The recording protocol consisted of 300-ms pulses ranging from À100 to +100 mV from a holding potential of À40 mV at a frequency of 0.1 Hz. The external solution for measuring the NCX contained (in mM): NaCl 140, CaCl 2 2, MgCl 2 1, HEPES 5 and glucose 10 (pH of 7.4). It was supplemented with strophanthidin (10 lM), nitrendipine (10 lM) and niflumic acid (100 lM).
The transient outward potassium current (I to ) was studied using a protocol consisting of a 30-ms pre-pulse from a holding potential of À80 to À40 mV to inactivate sodium channels followed by a 300-ms test pulse to +60 mV in 10-mV increments at a frequency of 0.1 Hz in the presence of 200 lM CdCl 2 in Ca 2+ -free normal Tyrode's solution as described previously [28]. The I to was measured as the difference between the peak outward current and the steady-state current [29], and the sustained outward potassium current (I Ksus ) was evaluated as the difference between the holding current and the end of the steady-state current.
The ultra-rapid delayed rectifier potassium current (I Kur ) was measured as 4-aminopyridine-sensitive currents. To record I Kur without contamination by I to , I Kur was dissected by 1 mM 4-aminopyridine with a double-pulse protocol, consisting of a 100-ms depolarizing pre-pulse to +40 mV from a holding potential of À50 mV to inactivate I to , followed by 150-ms voltage steps from À40 to +60 mV in 10-mV increments at room temperature to provide adequate temporal resolution [30,31].
The I KAS was recorded using the whole-cell mode of the patch-clamp technique. Stability of the total potassium current (I K ) was monitored with a step-pulse protocol (with a holding potential of À50 mV and test potentials from À120 mV to +70 mV for 300 ms at a frequency of 0.1 Hz). Once the I K became stable (usually for 5 min.), apamin (100 nM) was applied, and a step-pulse protocol was performed when the I K reached a steady state. The I KAS was calculated as the difference between the absence and presence of apamin. All experiments were performed at 36°C. The external solution contained (in mM): N-methylglucamine, 140; KCl, 4; MgCl 2 , 1; glucose, 5; and HEPES, 10 (pH 7.4 with HCl) [32].

Western blot analysis
Control and colchicine-treated HL-1 cells were centrifuged and lysed using radioimmunoprecipitation (RIPA) buffer. The protein concentration was determined with a Bio-Rad protein assay reagent (Hercules, CA, USA). For the analysis, equal amounts of protein from each sample were separated using 4~16% Tris-acetate polyacrylamide gradient gel electrophoresis. After electrophoresis, protein samples were transferred onto equilibrated polyvinylidene difluoride membranes (Amersham Biosciences, Buckinghamshire, UK). Blots were probed with primary antibodies against ryanodine receptor type 2 (RyR; Affinity BioReagent, Golden, CO, USA), RyR phosphorylation at S2808 and S2814 (RyR

Statistical analysis
All quantitative data are expressed as the mean AE standard error of the mean (S.E.M.). A paired t-test was used to compare differences between control and colchicine-treated HL-1 cells and collagen-treated HL-1 cells with or without colchicine. Nominal variables were compared by a Pearson's chi-square test or Fisher's exact test. A P value of <0.05 was considered statistically significant.

Effects of colchicine on Ca 2+ regulatory proteins and Kv channel subunits
The expression of SERCA2a and Cav1.2 in the colchicine (3 nM)treated HL-1 cells was lower by 14% and 11% than in the controls. Moreover, the expression of CaMKII, a multifunctional serine/threonine protein kinase that mediates Ca 2+ handling, had also decreased by 15% in colchicine (3 nM)-treated HL-1 cells. The colchicine (3 nM)-treated HL-1 cells also had lower expressions of total PLB and PLB-Thr17 by 18% and 23%, but had similar expressions of PLB-Ser16 (Fig. 4A). The colchicine (3 nM)-treated HL-1 cells had lower expression of NCX by 9%, but had similar expressions of RyR, RyR-2808, and RyR-2814 compared with control cells. The colchicine (3 nM)-treated HL-1 cells had lower The effect of colchicine on microtubule structure and the AP morphology in collagen-induced changes of HL-1 cells As shown in Figure 5A, a breakdown of the reticular microtubules was evident in the colchicine (3 nM)-treated HL-1 cells. When compared with the controls, the colchicine (3 nM)-treated HL-1 cells had longer APD 90 , APD 50 and APD 20 , but had similar values for the APA and RMP. In contrast to the effects of colchicine, collagen (10 lg/ml) shortened the APD 90 in HL-1 cells and had more delayed afterdepolarizations (DADs; 20% versus 66.7%, P < 0.005). Furthermore, colchicine (3 nM) ameliorated APD changes in collagen (10 mg/ml)-treated HL-1 cells as shown in Figure 5B, and no DADs were noted.

Discussion
Ca 2+ homeostasis is essential for maintaining normal cardiac function and electrical activity. There is tight coupling between Ca 2+ influx and Ca 2+ release. Dysregulated Ca 2+ handling can generate the triggered activity, commonly associated with AF development [33]. Ca 2+ overload may potentially produce arrhythmogenesis in atrial myocytes [34]. In this study, colchicine was shown to reduce the amplitude of [Ca 2+ ] i transients and SR Ca 2+ content, which suggests colchicine decreases Ca 2+ concentration in both cytoplasm and the SR. Accordingly, colchicine might directly modulate Ca 2+ homeostasis in atrial myocytes and result in a decrease in the occurrence of AF. In contrast, previous studies on rat or chick embryos ventricular myocytes showed that microtubule disruption by colchicine (incubation for 10 min. or 2 hrs, 1~10 lM) increased the I Ca-L and [Ca 2+ ] i transients, but did not modify SR Ca 2+ content [10,[35][36][37]. Additionally, other studies showed colchicine had no effect on APD and I to in isolated mouse or dog ventricular myocytes [38,39]. This disparity may arise from different experimental settings (animal species, colchicine concentrations and incubation period) or the variance between HL-1 cells and ventricular myocytes. As shown in Figure 5A, we have found that colchicine (3 nM) had significant effects on microtubule disruption, which supports the possibility that colchicine may modulate HL-1 cell calcium homeostasis through microtubule disruption. A previous study showed that patients receiving colchicine had been given therapeutic doses that ranged from 0.5 to 3 lg/l (0.8~7 nM) [20]. Therefore, the concentration used in this study was clinically relevant.
In addition to reductions in I Ca-L and reverse mode NCX in HL-1 cells, colchicine can down-regulate SERCA2a, total and PLB-Thr17. A decrease in PLB phosphorylation at Thr17 can functionally reduce SERCA2a Ca 2+ reuptake by increasing the effects of PLB on the inhibition of SERCA2a activity. All these changes can result in a decrease in SR Ca 2+ content and Ca 2+ transients. Colchicine decreased CaMKII expression, which is a key regulator of excitation-contraction coupling in cardiomyocytes. CaMKII activation may enhance I Ca-L facilitation and increase Ca 2+ influx, which accompanies Ca 2+ overload. In addition, CaMKII can enhance Ca 2+ leakage to generate triggered activity. The potential of CaMKII inhibition to reduce an arrhythmogenic outcome was demonstrated in AF [40]. This action may result in the known effect of colchicine in decreasing Ca 2+ sparks from a previous study [10]. Therefore, colchicine decreases calcium overload through modulation of Ca 2+ regulatory proteins, which may inhibit the occurrence of AF.
We found that colchicine can increase the APD, which may reduce the generation of re-entrant circuits and AF maintenance. This prolongation of the APD may be caused by a reduction of reverse mode NCX, I to , and I Ksus , which can overcome the APD-shortening effect by I Ca-L reduction. Thus, colchicine modulates atrial electrophysiological property by interfering Ca 2+ and potassium currents, which may ameliorate AF occurrence. Similarly, cytoskeletal disruption was also reported to increase the APD in ventricular myocytes by decreasing the I to [41]. These findings suggest that interactions exist between the potassium channels and microtubules or associated cytoskeletal components. The decreased expressions of Kv1.4 and Kv1.5 may contribute to decreases in the I to and I Ksus mediated by colchicine.
The ECM of collagen causes atrial myocytes to exhibit a shorter APD by increasing the I to and I Ksus [34]. In this study, it was found that colchicine reversed the APD-shortening effects of collagen. Moreover, collagen was also shown to induce Ca 2+ overload, which facilitates the occurrence of trigger activities in atrial and pulmonary vein myocytes. Colchicine normalized the APD in collagen-treated HL-1 cells, and this suggests that the ECM controls cytoskeletal mechanics in the development of cardiac remodelling, and colchicine may modulate the remodelling process and atrial arrhythmogenesis.
There are several limitations of our study that must be acknowledged. This study demonstrated the effect of colchicine on microtubule disruption in HL-1 cells, an atrial cell line from mice. However, the ultrastructure of HL-1 cells, including the cytoskeleton, and microtubule stability are different from adult na € ıve cardiomyocytes. [42,43] Because of the immature phenotype of HL-1 cells, our findings may not be fully applied to human atrial cells. In addition, we did not study the RyR or SERCA2a activity, and it is not clear whether fac-tors other than those found in this experiment may also contribute to colchicine-reducing SR content. Sorcin, a penta-EF-hand protein, interacts with intracellular target proteins, such as NCX, SERCA2a and I Ca-L [44][45][46], and may influence these current activities. Moreover, pulmonary vein cardiomyocyte plays an important role for triggering AF [47]. Ca 2+ dysregulation may increase arrhythmogenesis in pulmonary vein cardiomyocytes. Colchicine may reduce AF recurrence after pulmonary vein ablation [6]. Theoretically, the electrophysiological effects of colchicine on pulmonary vein cardiomyocytes may possibly play a role in its anti-AF potential.
In conclusion, our findings suggest that colchicine modulates atrial electrical activity and Ca 2+ regulation, which may be responsible for preventing AF occurrence or recurrence. These changes also The concentration (1, 3 and 10 nM) effect of colchicine on microtubule structure was studied in HL-1 cells by confocal microscope. Upper panel shows that HL-1 cells were labelled with a-tubulin and then visualized with secondary antibodies conjugated to FITC (green), and lower panel shows that nuclei were stained with DAPI (blue). Colchicine (1 nM) did not have significant effect on microtubule structure, colchicine (3 nM) induced depolymerization with an uneven microtubule structure in HL-1 cells, and colchicine (10 nM) caused patch aggregations of microtubule in cytoplasm and a decrease in the cell number. (B) Action potential (AP) characteristics of HL-1 cells with and without colchicine (3 nM) or collagen (10 lg/ml) treatment. Examples and average data of APs from control (n = 12), colchicine-treated (n = 14), collagen-treated (n = 12) and colchicine and collagen-treated (n = 15) HL-1 cardiomyocytes. AP durations at 20%, 50% and 90% repolarization of the AP amplitude (APD 20 , APD 50 and APD 90 , respectively) were measured at 1 Hz. *P < 0.05, **P < 0.01, ***P < 0.005. AP, action potential. demonstrate that microtubule modulation may play an important role in the regulation of ionic channels and Ca 2+ homeostasis, which contributes to the pathogenesis of AF.