Introduction

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

Clomiphene, or 1-[p-(diethylaminoethoxy)phenyl]-1,2-diphenylchloroethylene (Clomid, Serophene, and Milophene; CLM), is a nonsteroidal triphenylethylene antiestrogen agent that has been available on the U.S. market since late 1960s. Despite its competitive antagonistic action to estrogen, CLM displays some estrogenic properties, depending upon species and tissue (McKenna and Pepperell, 1988). Since the original synthesis of CLM in 1956 (Allen et al., 1959), CLM has been used to 1) induce ovulation in infertile women, 2) treat oligospermia in men, 3) help in diagnosis of impaired hypothalamic-pituitary-gonadal axis function, and 4) help in diagnosis of impaired ovarian function (Foss et al., 1973; McKenna and Pepperell, 1988).

Most of the antiestrogenic effects of CLM and other antiestrogens are mediated by binding to the estrogen receptor. Nevertheless, evidence is accumulating that some actions assigned to antiestrogens do not involve an interaction with the estrogen receptor (Zhang et al., 1994; Voets et al., 1995; Manolopoulos et al., 2000; Dodds et al., 2001). Recently, Maertens et al. (2001) have shown that both the cis- and trans-isomers of CLM inhibit both volume-regulated anion channels (VRACs) and cell proliferation in cultured pulmonary artery endothelial cells (CPAEs).

VRACs have been found to have a ubiquitous distribution in most mammalian cell lines and are important regulators of cell volume, intracellular pH, amino acid transport, and other metabolic functions (for review, see Nilius et al., 1997a). In the heart, blockers of Cl currents [such as the volume-regulated Cl current (I_Cl,vol)] may exert a class III antiarrhythmic action, because the inhibition of outward current due to Cl influx should prolong the action potential and refractoriness (Mulvaney et al., 2000). The identification of selective VRAC-specific blockers is therefore of potential importance to the development of novel antiarrhythmic agents.

Thus, the primary aims of this study were to determine 1) whether CLM inhibits the cardiac I_Cl,vol; and 2) if so, whether the effects of CLM on I_Cl,vol were selective over other chloride conductances in cardiac myocytes, thereby indicating whether CLM would be a useful pharmacophore to study I_Cl,vol. Because most compounds presently used as pharmacological tools in cardiac chloride channel research possess a range of effects on other conductances in the heart (Liu et al., 1998a; Dick et al., 1999; Sitsapesan, 1999; Kargacin et al., 2000), an additional aim of this study was to test whether CLM can modulate major cation conductances present in cardiac myocytes. This study therefore 1) addresses the lack of information available regarding the effects of CLM on cardiac sarcolemmal cationic currents and 2) shows that CLM selectively blocks I_Cl,vol over other cardiac Cl conductances having marked inhibitory effects on cardiac cationic conductances.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

Cell Isolation

Guinea Pig Isolated Ventricular Myocytes. Ventricular myocytes were isolated from male guinea pig hearts (400-500 g) using an enzymatic procedure similar to that described by Levi and Issberner (1996). In brief, guinea pigs were cervically dislocated using a Home Office-approved procedure (schedule 1) and thereafter a thoracotomy was performed. The heart was rapidly excised, cannulated via the ascending aorta, and perfused in a retrograde manner at 37°C at 6 ml·min¹, initially with oxygenated solution (solution A; Table 1) + 750 µM Ca²⁺ for 2 min, and then for 5 min with solution A + 8 mM EGTA. Finally, it was perfused for 12 to 16 min (depending on heart size/animal weight ratio) with an enzyme-containing solution consisting of solution A + 150 µM Ca²⁺ with 1 mg·ml¹ collagenase (type 1; Worthington Biochemicals, Freehold, NJ) and 0.1 mg·ml¹ protease (type XIV; Sigma-Aldrich). The enzyme was then washed out by perfusing with solution A + 150 µM Ca²⁺ for a further 5 min. Cells were released from enzyme-digested ventricles by gentle shaking in solution A + 150 µM Ca²⁺ for 6 min. After isolation, the cell suspension was filtered through nylon gauze (200-µm mesh), sedimented for 4 min, and then the supernatant was replaced with a high K⁺, Ca²⁺-free Kraft-Brühe solution; in which cells were stored at 4°C and used within 8 h.

Rat Isolated Ventricular Myocytes. Myocytes were also isolated from male Wistar rats [175-225 g; killed by an intraperitoneal injection of pentobarbitone sodium (200 mg/kg)]. Hearts were digested using collagenase and protease enzymes in an enzymatic dispersion method broadly similar to that described above (Spencer et al., 2000b). The rat ventricular myocytes were stored at 4°C in a solution containing 0.2 mM Ca²⁺ until use. Cells remained viable for up to 8 h after isolation.

Electrophysiological Recording

Membrane currents and potentials were acquired using an Axopatch 200A or Axopatch 200B amplifier (Axon Instruments, Union City, CA). Signals were digitized at a rate of 10 kHz and stored on digital audiotape (DTC 1000; Sony, Tokyo, Japan). Digitized membrane currents were acquired by computer from tape with low-pass filtering (frequencies conforming to Nyquist criterion) and subsequently signal-averaged over 10 stimulations using pClamp software, version 6 (Axon Instruments). Hyperpolarizing voltage steps of 20 mV and 5 ms duration were applied at 20 Hz to record the capacitance transients required for direct integration and the calculation of cell capacitance. Series resistance values were in the range of 2 to 4 M. Typically, 75 to 80% of series resistance could be compensated. During whole-cell anion current recordings, junction potential changes were minimized by using a continuous agar bridge (4% agar in 3 M KCl) where the reference Ag/AgCl electrode was immersed in a 3 M KCl solution. During macroscopic cation current recordings the Ag/AgCl electrode was immersed directly into the perfusate in the recording bath. Borosilicate glass pipettes (Harvard Apparatus, Edenbridge, Kent, UK) were pulled using a vertical two-step PP-830 microelectrode puller (Narishige, Tokyo, Japan) and had a tip resistance of 2.5 to 5 M when filled with the various pipette solutions. For selective I_Na recordings patch pipettes (Corning 7052 glass; AM Systems, Inc., Everett, Calsborg, UK) were pulled (P-87; Sutter Instrument Co., Novato, CA) and polished (MF-83 microforge; Narishige) to resistances between 2 and 3 M when filled with the appropriate intracellular dialysis solution.

Isolated myocytes used for whole-cell voltage-clamp experiments were placed in a Perspex chamber mounted on an inverted microscope (Diaphot 300; Nikon, Tokyo, Japan), allowed to settle, and then superfused at 20-25°C with a standard HEPES-buffered Tyrode's solution (solution B) until the whole-cell recording configuration had been obtained. After the cell interior was equilibrated with the pipette solution at the relevant holding potential (40, 50, or 80 mV), the bath solution was changed from solution B to the solutions used to isolate the different currents.

Voltage-Clamp Recordings from Guinea Pig Ventricular Myocytes. Guinea pig ventricular myocytes were chosen for use in this study because 1) I_Cl,vol has been well characterized and is more easily recorded in guinea pigs than in adult rat ventricular myocytes, and 2) quantitative measurement of cardiac I_Na from this species has been well established in our laboratories (Yuill et al., 2000; Spencer et al., 2001). To activate I_Cl,vol, a K⁺-free isoosmotic extracellular solution (ISO; solution C) was replaced by hyposmotic solution (HTS; solution D) prepared by simply omitting 140 mM sucrose from solution C. Internal solution F (which contained Cs⁺ to block outward K⁺ currents) was used in these experiments. I_Cl,cAMP was activated by forskolin (FSK; 1 µM) added to solution E. A different intracellular Cs⁺-based solution (solution G) was used in these I_Cl,cAMP recordings. To record signature currents (Spencer et al., 2000b), myocytes were continuously superfused with solution B and a simple intracellular K⁺-based solution (solution H) was used.

1

1

Voltage-Clamp Recordings from Rat Ventricular Myocytes. Rat ventricular myocytes were used to study 1) the effect of CLM on I_AB, because during the time when this study was conducted I_AB had been well characterized in rat ventricular myocytes (Spencer et al., 2000a); and 2) the effects of CLM on net cationic currents, using the signature current technique (Spencer et al., 2000b). For isolation of I_AB a sodium-free Tyrode's solution where NaCl was replaced by N-methyl-D-glucamine-NO₃ (NMDG-NO₃; solution I) was used, whereas solution F was used as the intracellular pipette solution. Because I_AB has been previously shown to be outwardly rectifying and highly permeable to NO₃ (Spencer et al., 2000a; Borg et al., 2002), effects of CLM on I_AB were investigated using NO₃ as the charge carrier to accentuate the current profile of I_AB. This current was elicited by continuous trains of depolarizing ramps from 90 to +70 mV from a holding potential of 50 mV (ramp rate of 0.32 Vs¹) at a stimulation frequency of 0.33 Hz. Capitative current was eliminated by subtraction.

Mathematical Equations. The following equations have been used to analyze data. Concentration-response relations were fitted by a Hill equation of the following form:

(1)

where I_max is the maximum degree of inhibition expressed as a percentage (I_max = 100), IC₅₀ is the concentration of CLM causing half-maximal inhibition, and n_H is the Hill coefficient.

(2)

(3)

(4)

Statistical Analysis. One-way analysis of variance was used to test whether the inhibition of CLM of I_Cl,vol was voltage-dependent. All other comparisons between control and CLM superfusion were made using a paired Student's t test. Significance refers to the 95% level of confidence (P < 0.05) unless otherwise stated. All data for statistical analysis were analyzed using Microsoft Excel 97. The concentration-response curve for inhibition of I_Cl,vol by CLM was plotted using GraphPad Prism, version 3 (GraphPad Software, San Diego, CA). All other graphs were drawn/fitted by use of Origin, version 3.5 (MicroCal Software, Northampton, MA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

Inhibition of I_Cl,vol by Clomiphene. The effects of CLM on I_Cl,vol were studied as shown in Fig. 1. Each myocyte was initially exposed to the hyposmotic external solution to induce I_Cl,vol and then to cumulative increases in CLM concentrations (Fig. 1A). Figure 1B shows representative I-V relationships before (a) and during exposure to hyposmotic solution (b), and after addition of CLM at concentrations of 3, 10, and 30 µM (c-e). The current activated during exposure to hyposmotic solution (Fig. 1C, b-a) was obtained by digital subtraction of the current trace recorded in control isoosmotic solution from that in hyposmotic solution. The current exhibited outward rectification and a reversal potential of 20.33 ± 1.35 mV (n = 6, mean ± standard error of the mean; throughout): a value slightly depolarized to the theoretical E_Cl (33.05 mV) under these recording conditions (due to permeation of anions present in the pipette; Nilius et al., 1997a). Figure 1C also shows for the same cell the fraction of steady-state I_Cl,vol inhibited by each concentration of CLM: derived by digitally subtracting the current recorded in the presence (c, d, and e) from that in the absence of CLM (a). The component of I_Cl,vol inhibited by each concentration of CLM tested (Fig. 1C) was also outwardly rectifying, with a reversal potential of 20.62 ± 1.32 mV (n = 6). This indicated that only I_Cl,vol was affected by CLM under these conditions. The data in Fig. 1, A and B, show that CLM induced a reversible concentration-dependent block of I_Cl,vol with both inward as well as outward currents being inhibited. To verify the presence of I_Cl,vol the current activated by the hypotonic solution was shown to be blocked by DIDS (50 µM, n = 4; Fig. 1D). The steady-state inhibition of I_Cl,vol by CLM was independent of membrane voltage using continuous voltage-ramp protocols between 60 and +60 mV (data not shown). Figure 1E shows the concentration dependence of the inhibition by CLM of I_Cl,vol at +60 mV derived from five different concentrations (each data point for CLM was obtained from three to six cells) and fitted to eq. 1. The estimated IC₅₀ and Hill coefficient for I_Cl,vol block by CLM were 9.67 ± 0.02 µM and 1.08 ± 0.06, respectively.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Inhibition of ICl,vol by clomiphene. A, time course of activation and inhibition of ICl,vol measured from guinea pig ventricular myocytes at +60 and 60 mV (during voltage ramps) during superfusion with control ISO solution (a; solution C), after 5 min of superfusion with HTS (b; solution D), and subsequent superfusion with 3 µM CLM (c), 10 µM CLM (d), and with 30 µM CLM (e) in solution D, f, return to control conditions during application of HTS without CLM. Horizontal bars mark respective interventions. B, representative I-V relations for net whole-cell current obtained at different phases of experiment shown in A recorded as follows: 5 min of superfusion with solution D (b), 2 min of superfusion with 3 µM CLM (c), 4 min of superfusion with 10 µM CLM (d), 2 min of superfusion with 30 µM CLM (e), and 4.5 min of superfusion with solution D (f). C, membrane current activated by hypotonic solution (b-a) and its components inhibited by CLM at a concentration of 3 (b-c), 10 (b-d), and 30 µM (b-e) obtained by digital subtraction. The traces in C are from the same cell as in B. D, histogram showing mean amplitude of the current induced by HTS at a test potential of +40 mV during superfusion with ISO solution (solution C), after superfusion with HTS (solution D) and after 5 min of superfusion with 50 µM DIDS (n = 4) in solution D. Solution F was used as the intracellular solution in A to D). E, dose-response curve for inhibition of ICl,vol by CLM. Each data point represents mean ± S.E.M. of n cells as indicated. The filled line represents the best fit of the data to the Hill equation (eq. 1).

Effects of Clomiphene on I_Cl,cAMP and I_AB. Figure 2A shows the effect of 100 µM CLM on the I-V relationship of I_Cl,cAMP (activated by 1 µM FSK) recorded from a guinea pig ventricular myocyte. The Cl current activated by FSK (1 µM) was slightly outwardly rectifying under the transmembrane Cl gradient (intracellular, 21 mM; extracellular, 153 mM), with a reversal potential of 24.75 ± 1.6 mV. The estimated Cl equilibrium potential under these recording conditions is 53 mV [solution E ([Cl]₀, 153 mM intracellular]; solution G ([Cl]_i, 21 mM)]. This deviation is consistent with predictions for the ionic gradients (Nilius et al., 1997a). The current in the presence of FSK (FSK added to solution E) was found to be insensitive to 5-min extracellular application of 100 µM CLM [Fig. 2, A and B (b)] (solution E and 1 µM FSK). Removal of FSK resulted in a return to control conditions.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Lack of inhibition ICl,cAMP and IAB by clomiphene. A, representative I-V relations for whole-cell currents (Im; normalized to membrane capacitance) obtained from guinea pig ventricular myocytes during superfusion with solution E (a), after adding 1 µM FSK to solution E for 5 min (b), after 5 min of superfusion with 100 µM CLM in the presence of 1 µM FSK in solution E (c), and after 3 min of superfusion with solution E (d). B, effect on mean outward current amplitude of the FSK-induced current at a test potential of +60 mV under conditions described for a to d in A (n = 4). Solution G was used as the intracellular solution in all experiments shown in A and B. C, representative net anion whole-cell current recorded from rat ventricular myocytes in the presence of extracellular nitrate (solution I; NMDG-NO3) (a) and after 5 min with 100 µM CLM in solution I (b). Recordings were carried out in the absence of cell swelling or PKA, PKC stimulation. The residual current in the presence of anion is attributable to an anionic background current that is larger with NO3 than Cl, we therefore used NO3 to enhance the contribution of this current to the net current. Solution F was used as the intracellular solution in Fig. 2C. Insets in A and C depict the voltage protocol used.

Effects of Clomiphene on Macroscopic Cation Conductances in Signature Current Profiles Recorded from Rat and Guinea Pig Ventricular Myocytes. The selectivity of CLM for I_Cl,vol over I_Cl,cAMP and I_AB raised the possibility that CLM may provide a useful tool for pharmacological dissection of Cl current responses. However, for this to be the case, CLM would ideally also be selective for I_Cl,vol over major cardiac cationic conductances. We therefore performed additional experiments to establish whether CLM had any effects on major cationic conductances in ventricular myocytes. Voltage-clamp experiments using rat and guinea pig ventricular myocytes were undertaken to record ionic currents during membrane potential ramps as described previously (Spencer et al., 2000b, 2001). Our previous results have shown the utility of this method (the use of depolarizing voltage ramps over the range of potentials encountered in the cardiac action potential) for qualitatively determining which ionic currents are modified by experimental compounds. Signature currents (Spencer et al., 2000b) were continuously recorded from rat ventricular myocytes before (Fig. 3A) and during superfusion with 10 µM CLM (Fig. 3, B and C). Figure 3B shows that CLM blocked the components of the signature current identified previously (Spencer et al., 2000b) as I_K1, I_Na, I_Ca,L, and I_Kv.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Effects of clomiphene on membrane signature currents obtained from rat ventricular myocytes. Representative membrane signature currents plotted against ramp voltage (Vm) during depolarizing voltage ramps at a stimulation frequency of 0.33 Hz recorded from a rat ventricular myocyte during superfusion with solution B. Solution H was used as the intracellular solution for this experiment. A, membrane signature current response elicited after 2 min of superfusion with solution B. Four main current components were evident, as previously identified by Spencer et al. (2000a). Voltages shown in parentheses indicate the potential at which the four main labeled current components were observed. B, membrane signature current from the same cell after 2 min of superfusion with 10 µM CLM in solution B. Arrows indicate that IK1 (1), INa (2), ICa,L (3), and IKv (4) were sensitive to extracellular superfusion of 10 µM CLM. Note that no further inhibitory effects on IK1, INa, ICa,L, and IKv were obtained after 2 min of exposure with 10 µM CLM (data not shown). In A and B, 500-ms depolarizing voltage ramps were applied from 90 to +70 mV after a 10-ms voltage step to 90 mV, from a holding potential of 40 mV at a frequency of 0.33 Hz. C, membrane signature current recorded after changing the holding potential of the depolarizing ramp protocol used in B from 40 to 70 mV. The myocyte was still being superfused with 10 µM CLM in solution B. Note that INa (2) was reactivated during this applied voltage protocol. Membrane current traces in A and B were constructed from continuously applied voltage ramps from a holding potential of 40 mV. During the lag phase, required for changing to the second voltage-ramp protocol, the membrane potential was held at 40 mV. The total time taken to change between voltage-ramp protocol 1 and 2 was less than 5 s. All traces shown in A to C were obtained from the same cell. Time scale bar applies to all panels.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Inhibition of IK1 and IKv by clomiphene. A and B, representative time course of inhibition of IK1 (A) and IKv (B) by CLM at 90 and +70 mV, respectively, produced by 10 µM CLM (in solution B) during continuous trains of membrane depolarizing voltage ramps applied at a basic stimulation frequency of 0.33 Hz. A and B represent the same cell. Horizontal bars mark records of CLM superfusion. Solution H was used as the intracellular solution.

Effects of Clomiphene on Selective I_Na in Guinea Pig Ventricular Myocytes and Selective I_Ca,L Recordings in Rat Ventricular Myocytes. Although the signature current data in Fig. 3 suggest that CLM inhibits cardiac I_Na, it was not possible to make quantitative conclusions regarding the extent of this action from signature current measurements due to the fast and large nature of I_Na under these conditions. Therefore, additional experiments were performed under conditions suitable for making quantitative I_Na measurements. Selective and quantitative recordings of I_Na were made using conventional "square pulse" voltage commands using recording conditions recently validated for guinea pig ventricular myocytes (Yuill et al., 2000; Spencer et al., 2001). Selective I_Na measurements in the present study therefore focused on guinea pig ventricular myocytes.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Clomiphene inhibits INa recorded selectively. A, representative INa traces elicited by voltage commands to 30 mV from prepulse potentials of 140 mV (left column; bottom trace shows voltage command) and 80 mV (right column; bottom trace shows voltage command). Each superimposed pair of current traces represents INa recorded in control solution and in the steady state, in the presence of the CLM concentrations marked. All INa traces are shown as TTX-sensitive current (30 µM TTX). Pulses were applied at a frequency of 0.2 Hz. CLM produced a concentration-dependent inhibition of INa that was influenced by prepulse potential. B, concentration-response relations for peak INa inhibition by CLM, for prepulse potentials of both 140 and 80 mV. Number of replicates for each data point are as follows: for 140 mV: 0.1 µM, n = 5; 1 µM, n = 5; 10 µM, n = 7; and 100 µM, n = 5; for 80 mV: 0.1 µM, n = 6; 1 µM, n = 5; 10 µM, n = 6; and 100 µM, n = 5. Points on each data plot were joined by a continuous smooth curve in B. Data points were also converted to fractional block levels and fitted by eq. 1 to give IC50 values at each potential. For a prepulse potential of 140 mV, IC50=5.31 ± 0.28 µM; nH = 0.64 ± 0.02. For a prepulse potential of 80 mV, IC50 = 0.32 ± 0.11 µM; nH = 0.39 ± 0.06.

View larger version (12K): [in this window] [in a new window]   Fig. 6.   Clomiphene inhibits ventricular ICa,L recorded selectively with a Cs+-based internal dialysis solution (solution F). In A, 500-ms duration voltage-clamp test pulses were applied from 40 to 0 mV. Pulse frequency of 0.33 Hz. A, representative control ICa,L (left) in solution J and 2 min after exposure to 10 µM CLM (right). The amplitude of ICa,L in the presence of 10 µM CLM (right) was reduced compared with that in control (left). B, I-V (data were fitted manually with a curve of best fit) relation for ICa,L in solution J () and in the presence of 10 µM CLM (), obtained by applying a preconditioning pulse to 40 mV for 100 ms (to inactivate INa) from a holding potential of 80 mV and subsequently membrane potential steps of 500 ms, duration at 10-mV increments, were applied from 40 to +40 mV (n = 4). C, steady-state activation data for ICa,L in a rat ventricular myocytes in solution J () and in the presence of 10 µM CLM () (curves fitted by use of equation described under Results). No significant effect was observed in each of four cells (P > 0.05). The paired t test for two sample means was used to calculate any statistical significance between drug free periods and specific interventions with CLM.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

Two previous studies have reported that CLM inhibits 1) gap junction communication between cardiomyocytes (Verrecchia and Herve, 1997) and 2) Ca²⁺ uptake by cardiac sarcoplasmic reticulum vesicles (Dodds et al., 2001). However, despite the clinical importance of this drug, there have been no studies until now that examined the effect of this compound on cardiac sarcolemmal ion channels. Our principal findings are that CLM inhibits cardiac I_Cl,vol with selectivity over other anionic conductances and that, in addition, major cation conductances (I_Na, I_Ca,L, I_K1, and I_Kv) can also be inhibited by CLM.

Our concentration-response data for inhibition of I_Cl,vol (Fig. 1F) were described by a relation with a Hill coefficient for inhibition of 1.08 ± 0.06. This suggests that CLM molecules bind to the VRACs with a simple 1:1 stoichiometry. The IC₅₀ value for I_Cl,vol that we obtained was ~10.0 µM. In a recent study of effects of CLM on CPAE cells, CLM was found to inhibit I_Cl,vol with an IC₅₀ and Hill coefficient of ~1.0 µM and ~1.4, respectively (Maertens et al., 2001). Although the IC₅₀ for I_Cl,vol block by CLM in guinea pig ventricular myocytes is a little higher than that reported for I_Cl,vol in CPAE cells, there is a great deal of evidence that chloride channel blockers have different IC₅₀ values for I_Cl,vol in different cell types and species (Lewis et al., 1993; Ehring et al., 1994; Gosling et al., 1995; Nilius et al., 1997a; Greenwood and Large, 1998; Wondergem et al., 2001). This situation may be due to different anionic channels underlying VRACs (Nilius et al., 1997a) but may equally reflect experimental differences between studies from different laboratories (Tominaga et al., 1995; Yamazaki and Hume, 1997).

A number of compounds structurally unrelated to clomiphene have been reported to inhibit I_Cl,vol, in different tissues with IC₅₀ values of 0.84 µM (DIDS; Dick et al., 1999), 5.4 µM (mibefradil; Nilius et al., 1997b), 6.0 µM (fluoxetine; Maertens et al., 1999), 23.0 µM (Gd³⁺; Dick et al., 1999), >100 µM (La³⁺; Dick et al., 1999), and 226.0 µM (4-acetamido-4'-isothiocyano-2,2'-disulfonic stilbene; Dick et al., 1999). Other antiestrogens have also been reported to inhibit I_Cl,vol with IC₅₀ values in the low micromolar range (tamoxifen, IC₅₀ = 0.57-3.8 µM; Voets et al., 1995; Dick et al., 1999) and nafoxidine (IC₅₀ = 1.61 µM; Maertens et al., 2001). Similar to CLM's inhibition on I_Cl,vol in CPAE cells (Maertens et al., 2001), CLM's inhibition of I_Cl,vol in guinea pig ventricular myocytes was obtained within minutes of exposure with fast reversal of inhibition during washout. The structurally related antiestrogen tamoxifen has also been observed to produce selective inhibition of VRAC Cl currents from intestinal cells over other Cl currents (Valverde et al., 1993), and this parallels our findings with CLM. Tamoxifen has also been observed to block I_Cl,vol from canine colonic myocytes (Dick et al., 1999) in a voltage-independent manner, whereas CLM's inhibition of I_Cl,vol from CPAE cells (Maertens et al., 2001) is also voltage-independent. These findings are consistent with ours and collectively, may be interpreted to suggest that the mechanism of inhibition of I_Cl,vol by this class of compounds differs from that of stilbene derivatives (DIDS and 4-acetamido-4'-isothiocyano-2,2'-disulfonic stilbene), which exert a voltage-dependent inhibition of I_Cl,vol (Sorota, 1994; Dick et al., 1999).

To date, compounds identified to block I_Cl,vol with selectivity over other chloride conductances have been found also to affect cationic conductances. Examples include tamoxifen (L-type Ca²⁺ and delayed rectifier K⁺ currents; Duan et al., 1997; Liu et al., 1998b), mibefradil (T-type Ca²⁺ channels; Nilius et al., 1997b), and DIDS (voltage-dependent fast sodium, L-type Ca²⁺ and delayed rectifier K⁺ currents; Liu et al., 1998a; Dick et al., 1999). Our data indicate that this situation also applies to CLM and the heart. CLM's inhibitory effects on cardiac cation conductances are consistent with published literature on the effects of CLM on cells from a neuroblastoma cell line (N1E-115). In N1E-115 cells, CLM produced a fast inhibition of 1) [¹⁴C]guanidinium influx through 5-hydroxytryptaime₃ receptors with an IC₅₀ of 2.6 µM, and 2) [¹⁴C]guanidinium influx through TTX-sensitive sodium channels with an IC₅₀ value of 13 µM (Barann et al., 1999). Our experiments suggest that the inhibitory potency of CLM against cardiac I_Na is comparable to that against I_Cl,vol, and may even be greater under conditions in which binding to the inactivated channel state for I_Na is favored. We also found CLM to inhibit cardiac I_Ca,L in rat (10 µM by 82 ± 2%) without significantly affecting the voltage dependence of I_Ca,L activation. This inhibitory action is consistent with blockade by tamoxifen of I_Ca,L from colonic myocytes (10 µM by 87 ± 7%; Dick et al., 1999). One conclusion from these data is that CLM could not be expected to discriminate between cardiac I_Cl,vol and I_Ca,L or I_Na under recording conditions in which each of these currents is present.

Tamoxifen has been shown to block VRACs and voltage-gated K⁺ currents in cardiac and colonic myocytes (Duan et al., 1997; Liu et al., 1998b; Dick et al., 1999). This also parallels our findings with CLM, because CLM partially inhibited the "outward" current (I_Kv) at +70 mV in rat ventricular myocytes. CLM also inhibited the current at 90 mV (I_K1). However, the degree of inhibition of these current components by 10 µM CLM was less than for I_Na, I_Ca,L, and I_Cl,vol, suggesting that CLM is a comparatively weaker blocker of I_K1 and I_Kv.

CLM's range of inhibitory effects on sarcolemmal ion channels might feasibly be considered important because CLM has a number of adverse effects that could be related to the nonspecific actions of the compounds on various ion channels. Described adverse effects of this widely used drug include various ocular adverse effects, bloating, stomach or pelvic pain, hot flashes, breast discomfort, dizziness or lightheadedness, headache, heavy menstrual periods or bleeding between periods, mental depression, nausea or vomiting, nervousness, restlessness, and tiredness and trouble sleeping (Siedentopf et al., 1997). However, despite the sensitivity of cardiac cation and anionic currents to CLM observed in the present study, there is a lack of reported cardiac adverse effects in patients taking this drug. Although it is difficult to extrapolate concentration-response data from the in vitro to the in vivo situation, it should be noted that serum concentrations of CLM in patients are rather low (0.1-0.4 µM; Young et al., 1999), which may account for the lack of cardiac adverse effects of CLM in adult patients.

	Conclusion

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

We conclude that the antiestrogen agent CLM shows selectivity for cardiac I_Cl,vol over I_AB and I_Cl,cAMP. However, it also inhibits a number of cationic conductances, which preclude its use a pharmacological tool to investigate the physiological and pathophysiological role of I_Cl,vol in both in vivo and in vitro experiments. However, CLM may have an application for the dissection of cardiac I_Cl,vol responses from those of other anion conductances, under conditions in which potentially interfering cation currents are blocked. In addition, our findings illustrate the utility of signature currents for rapidly studying the specificity of test compounds. In each case, the electrophysiological effects of CLM observed in selective recordings of individual ionic currents (I_Na and I_Ca,L) were successfully predicted from the signature current methodology in both rat and guinea pig ventricular myocytes.