Fang QZ et al / Acta Pharmacol Sin 2004 Mar; 25 (3): 327-333

Tetrandrine inhibits Ca²⁺-activated chloride channel in cultured human umbilical vein endothelial cells

Qi-zhi FANG^1,2, Ning ZHONG^1,3, Yi ZHANG¹, Zhao-nian ZHOU^{1, 4}

¹Physiological Lab of Hypoxia, Shanghai Institute of Biological Sciences, Chinese Academy of Sciences, Shanghai 200031, China; ³Department of Molecular and Cell Biology, University of California at Berkeley, Berkeley, CA 94720, USA

² Now in Cardiovascular Research Institute, University of California at San Francisco, San Francisco, CA 94143, USA

⁴ Correspondence to Prof Zhao-nian ZHOU. Phn 86-21-5492-0305. Fax 86-21-5492-0306. E-mail znzhou@server.shcnc.ac.cn

Received 2003-03-08 Accepted 2003-09-12

KEY WORDS chloride channels; endothelium; tetrandrine; patch-clamp techniques; calcium

ABSTRACT

AIM: To characterize the electrophysiological and kinetic properties of Ca²⁺-activated chloride channel (CaCC) in cultured human umbilical vein endothelial cell line (HUVEC), and test the inhibitory effects of tetrandrine (Tet) on CaCC. METHODS: Ca²⁺-activated Cl^- currents (I_Cl,Ca) were recorded by patch-clamp whole cell configurations. [Ca²⁺]_i was measured via intracellular Fura-2 fluorescence intensities. RESULTS: I_Cl,Ca was activated by increasing [Ca²⁺]_i via direct elevation of intracellular calcium. I_Cl,Ca showed an apparent outward rectification properties, and it was activated in a voltage- and calcium-dependent mode. Tet dose-dependently inhibited I_Cl,Ca, the IC₅₀ was (5.2±0.4) mmol/L (n=8 cells). Tet suppressed both voltage-dependent and calcium-dependent activation of I_Cl,Ca. The activation time constant was (326±12) ms [in the presence of 10 mmol/L Tet, compared to control (175±17) ms, at +100 mV], and Ca²⁺ concentration for half maximal activation was (387±61) nmol/L for Tet (compared to control (287±36) nmol/L. CONCLUSIONS: Tet effectively blocked I_Cl,Ca, and such effects might be due to its inhibitory effects on the activation process of Ca²⁺-activated chloride channel.

INTRODUCTION

Tetrandrine (Tet, 6,6',7,12-tetramethoxy-2,2'- dimethyl-berbaman) is a purified bis-benzylisoquinoline alkaloid derived from the root of a Chinese herb (Stephania tetrandra S Moore)^[1,2]. It was first shown as an anti-hypertensive agent in both normal and hypertensive subjects in 1950s^[3,4]. The primary anti-hypertensive action of Tet is presumably due to its vasodilatory property, which was confirmed both in vivo (15 mg/kg in conscious rats) and in vitro (1­100 mmol/L, effectivly decreased noradrenaline and high K⁺-evoked contraction of isolated aortic rings)^[5-8]. It was widely accepted that such vasodilatory effects principally relied on its direct Ca²⁺ antagonism (blocking voltage-dependent Ca²⁺ channels, mainly L-type Ca²⁺ channel, at 1-300 mmol/L with IC₅₀ of 14.8 mmol/L) and partially on its interaction with a-adrenoceptor in vascular smooth muscle cells^[6,9,10]. Recent studies showed that Tet also inhibited calcium-depletion-induced calcium entry in human and bovine endothelial cells, which might be through its blockade of a Ca²⁺ release-activated Ca²⁺ channels in such cells^[11,12]. In addition to the effects on Ca²⁺ release-activated Ca²⁺ channels, Wu et al reported that Tet (0.5-50 mmol/L) effectively suppressed large conductance Ca²⁺-activated K⁺ current in human endothelial cell line, and another group found that Tet depressed inwardly rectifying K⁺ current in cultured aortic endothelial cells^[13,14]. But Kwan et al showed that tetrandrine (30 mmol/L) did not affect either outward K⁺ current or inwardly rectifying K⁺ current in cultured bovine pulmonary artery endothelial cells, but Tet abolished thapsigargin-stimulated increase in outward K⁺ current^[11].

Endothelium plays an important role in modulating vascular smooth muscle tone by releasing many vasoactive factors, nitric oxide and PGI₂, which are controlled by intracellular Ca²⁺ signals^[15]. Since voltage-dependent Ca²⁺ channel is absent in endothelial cells, several types of Ca²⁺ entry channels are major influx route of Ca²⁺ (Ca²⁺ release-activated Ca²⁺ channels, nonselective cation channels etc). However after activation of these entry channels, Ca²⁺ influx depends on the driving force- electrochemical gradient for Ca²⁺. Endothelial K⁺ channels (Ca²⁺-activated K⁺ channels and inwardly rectifying K⁺ channel) and Cl^- channels (volume-regulated chloride channels and Ca²⁺-activated chloride channel) are two major channel types in regulating the driving force for Ca²⁺ entry^[16]. In contrast to the fast growing knowledge of Ca²⁺-activated K⁺ channels, the endothelial Ca²⁺-activated Cl^- channel (CaCC) is not very well studied and only a few papers reported its electrophysiological and pharmacological properties. Besides its role as a modulator in Ca²⁺ signaling, CaCC might also be important for regulating secretion, intracellular pH, and cell proliferation^[16-19].

Since both K⁺ and Cl^- channels are involved in maintaining normal function of endothelial cells, it is worthwhile to learn the effects of Tet on endothelial Cl^- channels besides its known action on various endothelial K⁺ channels. The results might be an important supplement to a clear understanding toward the mechanism of Tet. In our previous report, two types of chloride channels (volume-regulated and Ca²⁺-activated chloride channels) have been identified in cultured human umbilical vein endothelial cells^[20]. In this study, we focused on CaCC. We conducted whole-cell current recording, and in some experiments combined with fluorescence measurement of [Ca²⁺]_i in order to further characterize the kinetics and electrophysiological properties of CaCC. And the inhibitory effect of Tet on CaCC was tested.

MATERIALS AND METHODS

Cells Cells were recovered and cultured from a human umbilical vein endothelial cells line (cell line HUVEC, VEC304, obtained from the Cell Bank of Chinese Academy of Sciences, Shanghai). The cells were grown in Medium 199 containing 10 % calf serum, 0.03 % L-glutamine, benzylpenicillin 2 g/L and streptomycin 2 g/L. Cultures were maintained at 37 ºC in a fully humidified atmosphere of 5 % CO₂ in room air. The culture medium was exchanged every 48 h. Cells were detached by exposure to 0.1 % trypsin in Ca²⁺-and Mg²⁺-free solution and plated on gelatin-coated coverslips, and kept in culture for 2-4 d before use. For electrophysiological measurements, only non-confluent, single endothelial cells were used.

Electrophysiology Whole cell membrane currents were monitored with a patch clamp amplifier (CEZ2300, Nihon Kohden) and recorded in an IBM/PC via a D/A converter (Axon LM-1). Voltage clamping, signal acquisition, and analysis of membrane currents were achieved by computer program pCLAMP 5.55. All experiments were carried out at room temperature (22-25 ºC).

Two voltage protocols were applied to study the Ca²⁺-activated Cl^- current (I_Cl,Ca). Ramp current recording was designed to determine current-voltage (I-V) relationship. Holding potential was 0 mV, a step to -80 mV for 0.6 s was followed by a step to -150 mV for 0.2 s and a 2.6 s linear voltage ramp to +100 mV, thereafter a step to 0 mV. Another voltage steps were applied from a holding potential of -50 mV to a test potential of -100 to +100 mV in 20-mV increments. The duration of the step is 2 s. Junction potential between electrode and bath solution was corrected before experiment. In order to compare data obtained from different cells, the current amplitudes were normalized as per unit of membrane capacitance. Cell capacitance for the whole experiment is (29±9) pF (n=23 cells).

[Ca²⁺]_i measurement The cells were incubated in the medium containing Fura-2 AM (2 ¦̀mol/L) for 30 min at 37 ºC. Then the cells were washed by modified Krebs' solution for another 30 min, allowing the optimal hydrolysis of the AM ester. The dye was excited at wavelength of 350 and 380 nm via a rotating filter wheel. The fluorescence was measured at 510 nm and corrected for autofluorescence. The fluorescent signals were detected by an R928 photo-multiplier tube (PMT; Hamamatsu, Shizuoka, Japan). Calibration and calculation of the free Ca²⁺ concentration from the fluorescence ratio were achieved according to detailed methods described elsewhere^[21]. Briefly, [Ca²⁺]_i was determined from the ratio (R) of the fluorescence signals at both wavelengths, according to: [Ca²⁺]_i=K_dApp [(R-R_min)/(R_max -R)], where K_dApp is the apparent dissociation constant, R_min is the minimum ratio in zero Ca²⁺, and R_max is the maximum ratio at saturated Ca²⁺. These calibration constants were obtained from in-cuvette calibrations. Calibrations used values of R_max/R_min=22.9 and K_d=523 nmol/L

Solutions and drugs At the beginning of each experiments, cells were perfused with modified Krebs' solution, containing (mmol/L): NaCl 150, KCl 6, MgCl₂ 1, CaCl₂ 1.5, glucose 10, HEPES 10, adjusted to pH 7.4 with NaOH 1 mmol/L. Then through the whole experiment, the cells were superfused with a Cs-Krebs' solution in which Cs⁺ substituted K⁺ in the Krebs' solution to avoid the influence of any K⁺ current.

The pipette solution contained (mmol/L): CsCl 40, Cs-aspartate 100, MgCl₂ 1, EGTA 5, CaCl₂ 1.93, Na₂ATP 4, HEPES 10, adjusted pH to7.2 by CsOH 1 mmol/L.

To activate the Cl^- current directly by Ca²⁺, the free Ca²⁺ concentration was adjusted to 500 nmol/L in pipette solutions by modifying concentration of CaCl₂ and EGTA in pipette solution (EGTA 5, CaCl₂ 3.79, and MgCl₂ 1 mmol/L, pH=7.2). To test the Ca²⁺-dependent activation, free Ca²⁺ was adjusted to 100, 200, 500, and 1000 nmol/L by adding the appropriate CaCl₂ (1.94, 2.80, 3.79, 4.32 mmol/L), respectively and calculated by the program CHELATOR, provided by University of California at Berkeley, USA.

Tet (purity >98 %) obtained from Jin-Hua Pharmaceutical Co, Zhejiang, China. 4,4'-Diisocyanato-stilbene 2,2'-disulphonic acid (DIDS), verapamil were purchased from Sigma Co (USA).

Data analysis Inhibitory effects of Tet were defined as percentage of inhibition.

Inhibition=[(I_max-I _blocker)/(I_max-I _rest)]×100 %

where I_max and I_blocker were the currents at +100 mV testing potential before and during application of Tet respectively. I_rest was the membrane current without any stimulation just after the cell membrane was broken. Data were expressed as mean±SD and compared with the t-test.

All data and curve fit were achieved via software SigmaPlot2000 (SPSS Inc, USA) or Origin 7.02 (OriginLab Co, USA). Concentration-response curve of the inhibitory action of Tet is fitted by the modified Hill-equation: Y=(a-d)/[1+(X/c)^b]+d using Marquardt-Levenberg methods of nonlinear regression analysis, where a and d are the effect of concentration at maximum and minimum. c refers IC₅₀. b is the Hill coefficient^[22].

The Ca²⁺-dependent activation kinetics of I_Cl,Ca, were fitted by a single exponential function according to the following equation: I(t)=A₀+A ₁exp(-t/t). A is the amplitude; t is the time constant of activation (ms). Data of Ca²⁺-dependent activation of I_Cl,Ca were fitted with Hill equations: I=I_max/[1+(K_Ca/[Ca ²⁺])ⁿ], where I is the value of currents, I_max is the maximum current, K_Ca is the Ca²⁺ concentration for half-maximal activation, n is the Hill coefficient^[19].

RESULTS

Electrophysiological kinetic properties of the Ca²⁺-activated Cl^- current I_Cl,Ca was directly activated by loading endothelial cells with high concentration of Ca²⁺ (500 nmol/L) via patch electrode. After the membrane was ruptured and formed the whole-cell state, [Ca²⁺]_i was equilibrated slowly with the pipette solution [from resting level (58±11) nmol/L to the stable level (336±24) nmol/L, n=5 cells, Fig 1A], and in the meantime the membrane current slowly increased. The [Ca²⁺]_i and current amplitude reached to a stable level around 10 min thereafter. Ramp protocol testing of I_Cl,Ca showed a significant outward rectifying property (Fig 1B-b), and the reversal potential did not change after I_Cl,Ca being activated [(-25±3) mV, compared with (24±5) mV at rest]. The current recorded right after breaking into the cells (when there was no notable increase in [Ca²⁺]_i) was small and used as resting level of the current [I_rest, (-0.5±0.4) pA/pF and (3.6±1.0) pA/pF at -80 mV and +100 mV testing potential, respectively, n=8 cells, Fig 1C-a]. And the current at the testing potential +100 mV was used to evaluate the effects of channel blockers. The amplitude of I_Cl,Ca was (-3.2±1.4) pA/pF and (27±5) pA/pF at -80 mV and +100 mV, respectively (n=8 cells, Fig 1C-b). I_Cl,Ca was activated slowly at positive holding potential, and inactivated quickly at negative potentials. The activation of I_Cl,Ca at positive potential was voltage-dependent and mono-exponential (Fig 2, opened circle). The activation time constant (t_act) was (175±17) ms at +100 mV testing potential (n=6 cells).

Fig 1. Ca²⁺-activated Cl^- current (I_Cl,Ca) in HUVEC and inhibition by Tet 10 µmol/L. A) [Ca²⁺]_i measurement, "break" and arrow indicate the time point that patch electrode rupture the cell membrane. A-b, A-c is the time point sample current recorded. I-V relationship tested via a ramp protocol (B) and current trace tested via a step protocol (C) of I_Cl,Ca. B,C-a: current at resting level; B,C-b: I_Cl,Ca activated; B,C-c: current suppressed by Tet 10 µmol/L; B,C-d: current recovered from washout.

Fig 2. Voltage-dependent activation of I_Cl,Ca at positive testing potential. Activation time constants are plotted as function of testing potential. Open circle: control; closed circle: Tet 10 µmol/L.

DIDS (100 mmol/L), a specific chloride channel blocker, reduced the current amplitude by 50.1 %±3.6 % [(12±4) pA/pF at +100 mV, n = 4 cells]. Verapamil, acted as non-selective chloride channel blocker, inhibited I_Cl,Ca in a concentration-dependent manner. The IC₅₀ for its blockade was (120±10)  ¦̀mol/L (n=5 cells).

Quantitative analysis of Ca²⁺ dependence of I_Cl,Ca Loading of the endothelial cells with different concentration of Ca²⁺ enabled us to measure the relation between current amplitude and [Ca²⁺]_i. Fig 3 showed the steady state current densities at +100 mV as a function of [Ca²⁺]_i (Fig 3 opened circle). The steady state current-[Ca²⁺]_i relationship was fitted with Hill equations. At +100 mV testing potential, the maximal current was (39±8) pA/pF and the concentration for half maximal activation (K_Ca) was (287±36) nmol/L (n=5 cells). The Hill coefficient n_H was 1.89.

Fig 3. Calcium-dependent activation of I_Cl,Ca. Open circle: control; closed circle: Tet 10 µmol/L.

Tetrandrine inhibits Ca²⁺-activated Cl^- current After the amplitude of I_Cl,Ca reached to the stable level, Tet (1-100 ¦̀mol/L) was bath perfused. The outward current (testing at +100 mV) was decreased by 15 %± 11 %, 35 %±12 %, 71 %±9 %, 88 %±7 %, and 96 %±4 % by tetrandrine 1, 3, 10, 30, and 100 ¦̀mol/L, respec-tively. The inward current was decreased to a similar extent (Fig 1B-c, Fig 1C-c). The inhibition by Tet was reversible after washout by perfusion with Cs⁺- Krebs' solution (Fig 1B-d, 1C-d). [Ca²⁺]_i were monitored in some experiments, slight decrease of [Ca²⁺]_i was noticed after perfusion with Tet [(293±13) nmol/L, n=3 cells]. Fig 4 showed a summary of the concentration-dependent blockade by Tet. Dose-response curve was fitted by modified Hill equation. The half-maximal blockade by Tet was observed at around 5 mmol/L [(5.2±
0.4) ¦̀mol/L, n=8 cells].

Fig 4. Concentration-response curve of blockade of I_Cl,Ca by tetrandrine.

Tet did not affect the reversal potential of I_Cl,Ca (25±6 mV, Fig 1B-c). But Tet significantly affected voltage-dependent activation of I_Cl,Ca at positive testing potential (Fig 2 closed circle). When applying Tet 10 mmol/L, t_act was increased to (392±25), (381±39), (339±28), and (326±12) ms (n=8 cells) at +40, +60, +80, +100 mV, respectively [compared with control: (379±31) ms, (292±11) ms, (253±22) ms, and (175±17) ms]. Apparently, the blocking effects were more significant at more positive testing potential.

The effect of Tet on calcium-dependent activation of I_Cl,Ca was also tested (Fig 3, closed circle). Tet (10 ¦̀mol/L) significantly reduced the maximal current to (15±3) pA/pF (n=4 cells), and the K_Ca was increased to (387±61) nmol/L. But no apparent change of n_H (1.87).

DISCUSSION

Ca²⁺-activated chloride channel are less well studied than other chloride channels, such as cystic fibrosis transmembrane conductance regulator (CFTR) chloride channels, or volume-regulated chloride channel. Nevertheless, CaCC, in diverse cell types, is clearly an important channel type involved in various physiological functions (cell secretion, anion transport, electrode flux, cell adhesion etc)^[23,24]. The first distinct CaCC was identified in the bovine airway, with an ion selectivity of I>Cl, and sensitive to DIDS and DTT^[25]. To date, at least ten isoforms of CaCC (found from bovine, human, mouse and porcine) have been identified and published on NCBI GenBank database^[26]. The electrophysiological features of CaCC are very similar in the various cell types^[27]. The current-voltage relationship described here had the same outward rectification as in other cells^[24], and the activation kinetics of CaCC in HUVEC cells was similar to the time course of the Ca²⁺-activated Cl^- currents demonstrated in cultured calf pulmonary artery endothelial cells and in rat parotid acinar cells^[17,28]. Thus, CaCC presently tested belongs to general CaCC family. In endothelial cells, the speculated roles of CaCC are control of membrane potential and modulation of agonist-induced or store-depletion-dependent intracellular calcium signaling, including regulation of calcium influx. The calcium influx sequentially regulates the secretion of vasoactive compounds, which plays an essential role in controling vascular tone^[24].

Tet, in the concentration range from 0.1 to 100 mmol/L, not only shows a prominent blocking effect on L- and T-type channels, but also inhibits other ion channels, such as Ca²⁺-activated K⁺ channel, inward rectifying K⁺ channel, and Ca²⁺ release-activated Ca²⁺ channel in different cells or tissues^[11,29]. It suggests that Tet is not a selective Ca²⁺ channels, but display a rather wide spectrum effects on many ion channels. Our results demonstrated a concentration-dependent blockade of CaCC by Tet. The IC₅₀ was 5 mmol/L, which was at the therapeutically relevant concentration range. And compared with verapamil, a calcium channel blocker that exhibits a non-specific inhibitory effect on CaCC (IC₅₀ around 100 mmol/L), Tet is rather a potent blocker for endothelial CaCC. We also showed that Tet effectively suppressed both voltage- and calcium-dependent activation of CaCC. With the quantitative methods, we fitted the calcium-dependent activation curve with Hill equation. Hill coefficient >1 (n_H:
1.89) suggested that more than one Ca²⁺ ion bound to the channel to activate it. This suggests the existence of multiple Ca²⁺-liganded closed states in a voltage-dependent equilibrium with Ca²⁺-liganded open states. Tet did not affect n_H (1.87). It strongly indicates that Tet does not affect the Ca²⁺ binding sites of the channel. We presume that the mechanism of Tet blocking CaCC might be due to its effects on the activation process. However, the possibility of Tet directly binding to channel can not totally be excluded. The novel finding of the inhibitory effects of Tet on CaCC might be one of the underlying ionic mechanism of action of Tet on endothelial cells. Recent report showed that Tet exerted dual effects on vascular activities. As a vasodilator, it directly relaxes vascular smooth muscle. But since it blocks endothelial Ca²⁺-activated K⁺ channel and Ca²⁺-release activated Ca²⁺ channel, Tet might impede the NO production in endothelial cells, which leading to a transient vasoconstriction^[30]. Kwan pointed out, in vivo, such effects of Tet on endothelial cells might be overwhelmed by its sustained direct vasorelaxation on vascular smooth muscle cells^[31]. It is worth looking into the physiological role of its inhibitory effects on CaCC. In a separate experiment, we found that in the concentration range similar to that of blocking CaCC, Tet effectively blocked the volume-regulated chloride channel and endothelial cell proliferation (our unpublished data). The blocking CaCC by Tet might also take part in such regulating process since the anti-angiogenic effects of Tet may be related to its ability to inhibit the non-voltage dependent Ca²⁺-entry pathway.

REFERENCES

• 1 Chen K, Chen A. The alkaloids of han-fang-chi. J Biol Chem 1935; 109: 681-5.
• 2 Chen K, Chen A, Anderson R, Rose C. The pharmacological action of tetrandrine, an alkaloid of han-fang-chi. Chin J Physiol 1937; 11: 13-34.
• 3 Chang T, Chao K, Lue F. The cardiovascular effects of tetrandrine and dimethyltetrandrine. Acta Pharm Sin 1958; 6: 147-51.
• 4 Gao Y, Chang M, Mao H, Chao H, Chen D. The clinical observation of tetrandrine in the treatment of 270 cases of hypertensive patients and hypertensive crisis. Chin J Int Med 1965; 13: 504-8.
• 5 Kawashima K, Hayakawa T, Miwa Y, Oohata H, Suzuki T, Fujimoto K, et al. Structure and hypotensive activity relationships of tetrandrine derivatives in stroke-prone spontaneously hypertensive rats. Gen Pharmacol 1990; 21: 343-7.
• 6 Qian JQ, Thoolen MJ, van Meel JC, Timmermans PB, van Zwieten PA. Hypotensive activity of tetrandrine in rats. Investigation into its mode of action. Pharmacology 1983; 26: 187-97.
• 7 Jia JF, Gao NN, Xia GJ, Luo QF, Fang DC, Jiang MX. Studies on the calcium antagonistic action of tetrandrine: IX. Effects of tetrandrine on the contractility of isolated swine coronary artery strips. Acta Acad Med Wuhan 1985; 5: 15-9.
• 8 Hu WS, Pang XB, Wang Y, Hu CJ, Lu FH. Mode of action of tetrandrine on vascular smooth muscle. J Tradit Chin Med 1983; 3: 7-12.
• 9 Fang DC, Jiang MX. Studies on tetrandrine calcium antagonistic action. Chin Med J (Engl) 1986; 99: 638-44.
• 10 Liu QY, Li B, Gang JM, Karpinski E, Pang PK. Tetrandrine, a Ca⁺⁺ antagonist: effects and mechanisms of action in vascular smooth muscle cells. J Pharmacol Exp Ther 1995; 273: 32-9.
• 11 Kwan CY, Leung YM, Kwan TK, Daniel EE. Tetrandrine inhibits Ca²⁺ release-activated Ca²⁺ channels in vascular endothelial cells. Life Sci 2001; 68: 841-7.
• 12 Low AM, Berdik M, Sormaz L, Gataiance S, Buchanan MR, Kwan CY, et al. Plant alkaloids, tetrandrine and hernandezine, inhibit calcium-depletion stimulated calcium entry in human and bovine endothelial cells. Life Sci 1996; 58: 2327-35.
• 13 Wu SN, Li HF, Lo YC. Characterization of tetrandrine-induced inhibition of large-conductance calcium-activated potassium channels in a human endothelial cell line (HUV-EC-C). J Pharmacol Exp Ther 2000; 292: 188-95.
• 14 Liu WB, Liu GQ, Xiao H, Mao X, Shi Y, Wu JP. Tetrandrine inhibits inward rectifying potassium current in cultured bovine aortic endothelial cells. Acta Pharmacol Sin 2000; 21: 1115-8.
• 15 Nilius B. Signal transduction in vascular endothelium: the role of intracellular calcium and ion channels. Verh K Acad Geneeskd Belg 1998; 60: 215-50.
• 16 Nilius B, Droogmans G. Ion channels and their functional role in vascular endothelium. Physiol Rev 2001; 81: 1415-59.
• 17 Nilius B, Prenen J, Szucs G, Wei L, Tanzi F, Voets T, et al. Calcium-activated chloride channels in bovine pulmonary artery endothelial cells. J Physiol 1997; 498 (Pt 2): 381-96.
• 18 Nilius B, Szucs G, Heinke S, Voets T, Droogmans G. Multiple types of chloride channels in bovine pulmonary artery endothelial cells. J Vasc Res 1997; 34: 220-8.
• 19 Nilius B, Prenen J, Voets T, Van den Bremt K, Eggermont J, Droogmans G. Kinetic and pharmacological properties of the calcium-activated chloride-current in macrovascular endothelial cells. Cell Calcium 1997; 22: 53-63.
• 20 Zhong N, Fang QZ, Zhang Y, Zhou ZN. Volume- and calcium-activated chloride channels in human umbilical vein endothelial cells. Acta Pharmacol Sin 2000; 21: 215-20.
• 21 Zhong N, Beaumont V, Zucker RS. Roles for mitochondrial and reverse mode of Na⁺/Ca²⁺ exchange and the plasmalemma Ca²⁺ ATPase in post-tetanic potentiation at crayfish neuromuscular junctions. J Neurosci 2001; 21: 9598-607.
• 22 Maertens C, Wei L, Droogmans G, Nilius B. Inhibition of volume-regulated and calcium-activated chloride channels by the antimalarial mefloquine. J Pharmacol Exp Ther 2000; 295: 29-36.
• 23 Fuller CM, Ji HL, Tousson A, Elble RC, Pauli BU, Benos DJ. Ca(2+)-activated Cl(-) channels: a newly emerging anion transport family. Pflugers Arch 2001; 443 Suppl 1: S107-10.
• 24 Nilius B, Droogmans G. Amazing chloride channels: an overview. Acta Physiol Scand 2003; 177: 119-47.
• 25 Ran S, Fuller CM, Arrate MP, Latorre R, Benos DJ. Functional reconstitution of a chloride channel protein from bovine trachea. J Biol Chem 1992; 267: 20630-7.
• 26 Pauli BU, Abdel-Ghany M, Cheng HC, Gruber AD, Archibald HA, Elble RC. Molecular characteristics and functional diversity of CLCA family members. Clin Exp Pharmacol Physiol 2000; 27: 901-5.
• 27 Fuller CM, Benos DJ. Electrophysiological characteristics of the Ca²⁺-activated Cl^- channel family of anion transport proteins. Clin Exp Pharmacol Physiol 2000; 27: 906-10.
• 28 Arreola J, Melvin JE, Begenisich T. Activation of calcium-dependent chloride channels in rat parotid acinar cells. J Gen Physiol 1996; 108: 35-47.
• 29 Wang G, Lemos JR. Tetrandrine: a new ligand to block voltage-dependent Ca²⁺ and Ca(+)-activated K⁺ channels. Life Sci 1995; 56: 295-306.
• 30 Yao WX, Jiang MX. Effects of tetrandrine on cardiovascular electrophysiologic properties. Acta Pharmacol Sin 2002; 23: 1069-74.
• 31 Kwan CY, Achike F. Tetrandrine and related bis-benzyliso-quinoline alkaloids from medicinal herbs: cardiovascular effects and mechanisms of action. Acta Pharmacol Sin 2002; 23: 1057-68.