QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: mol.104.000893v1 66/5/1236    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Dilmac, N. Articles by Hockerman, G. H. Search for Related Content PubMed PubMed Citation Articles by Dilmac, N. Articles by Hockerman, G. H.

Molecular Determinants of Frequency Dependence and Ca²⁺ Potentiation of Verapamil Block in the Pore Region of Ca_v1.2

Department of Medicinal Chemistry and Molecular Pharmacology, School of Pharmacy and Pharmacal Sciences (N.H., G.H.H.) and Graduate Program in Biochemistry and Molecular Biology (N.D.), Purdue University, West Lafayette, Indiana

Received March 29, 2004; accepted July 28, 2004

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
Verapamil block of Ca_v1.2 is frequency-dependent and potentiated by Ca²⁺. We examined the molecular determinants of these characteristics using mutations that effect Ca²⁺ interactions with Ca_v1.2. Mutant and wild-type Ca_v1.2 channels were transiently expressed in tsA 201 cells with _1b and ₂ subunits. The four conserved glutamates that compose the Ca²⁺ selectivity filter in Ca_v1.2 were mutated to Gln (E363Q, E709Q, E1118Q, E1419Q) and the adjacent conserved threonine in each domain was mutated to Ala (T361A, T707A, T1116A, T1417A). The L-type-specific residues in the domain III pore region (F1117G) and the C-terminal tail (I1627A) were also mutated and assayed for block by verapamil using whole-cell voltage-clamp recordings in 10 mM Ba²⁺ or 10 mM Ca²⁺. In Ba²⁺, none of the pore-region mutations reduced the fraction of current blocked by 30 µM verapamil at 0.05 Hz stimulation. However, all of the pore-region mutations abolished Ca²⁺ potentiation of verapamil block at 0.05 Hz. The T1116A, F1117G, E1118Q, and E1419Q mutations all significantly reduced frequency-dependent verapamil block (1-Hz stimulation) in both Ba²⁺ and Ca²⁺. The I1627A mutation, which disrupts Ca²⁺-dependent inactivation, increased the fraction of closed channels blocked by 30 µM verapamil in Ba²⁺ but did not affect frequency-dependent block in Ba²⁺ or Ca²⁺. Our data suggest that the pore region of domain III may contribute to a high affinity verapamil binding site accessed during 1-Hz stimulation and that Ca²⁺ binding to multiple sites may be required for potentiation of verapamil block of closed channels.

The PAA verapamil blocks Ca_v1.2 at low micromolar concentrations in both primary cardiac myocytes (Lee and Tsien, 1983) and heterologous expression systems (Johnson et al., 1996). Quaternary amine derivatives of verapamil and methoxyverapamil block Ca_v1.2 channels in smooth muscle cells only when applied to the intracellular side of the membrane, whereas a quaternary amine derivative of desmethoxyverapmil blocks Ca_v1.2 from either side of the membrane (Berjukov et al., 1996). Block of closed channels by desmethoxyverapamil involves specific amino acid residues in transmembrane segments IIIS6 and IVS6 that are unique to L-type channels (Hockerman et al., 1995, 1997a; Doring et al., 1996). PAAs preferentially block Ca_v1.2 channels undergoing high-frequency depolarizations, a property called frequency-dependent block (Lee and Tsien, 1983), which is the result of a higher drug affinity for the inactivated state of the channel (Johnson et al., 1996; Nawrath and Wegener, 1997).

Verapamil is not highly selective for L-type channels, because Ca_v1.2, Ca_v2.1, and Ca_v2.3 channels show little difference in sensitivity to verapamil (Cai et al., 1997). Thus, the binding site for PAAs may be substantially conserved across voltage-gated Ca²⁺ channels. Indeed, mutation of two of the conserved Glu residues to Gln in the selectivity filter of Ca_v1.2 (EIIIQ or EIVQ) reduces the affinity of desmethoxyverapamil for closed channels in Ba²⁺ (Hockerman et al., 1997a). We assayed Ca_v1.2 channels in which the conserved Glu residues were mutated to Gln [E363Q (EIQ), E709Q (EIIQ), E1118Q (EIIIQ), E1419Q (EIVQ)], or the adjacent conserved Thr residues to Ala [T361A (TIA), T707A (TIIA), T1116A (TIIIA), T1417A (TIVA)] for closed-channel block and frequency-dependent block by verapamil in Ba²⁺. We found that none of the mutants tested significantly reduced the sensitivity of closed channels to block by verapamil but that EIIIQ, EIVQ, and TIIIA displayed a significantly lower extent of frequency-dependent block than did WT Ca_v1.2.

The conserved Glu residues that compose the selectivity filter of Ca_v1.2 bind Ca²⁺ ions as they pass through the channel pore, thus largely excluding monovalent cations when Ca²⁺ ions are present (Yang et al., 1993). However, the Ba²⁺ conductance of Ca_v1.2 is higher than that for Ca²⁺ because Ca²⁺ binds more tightly to the pore Glu residues (Almers and McCleskey, 1984). Ca²⁺ binding by Ca_v1.2 also modulates drug binding. All three chemical classes of L-type Ca²⁺ channel blockers are more potent when Ca²⁺, rather than Ba²⁺, is used as the charge carrier (Lee and Tsien, 1983). The conserved Glu residues in domains III and IV, as well as an adjacent Phe residue in the pore region of domain III in Ca_v1.1, mediate the Ca²⁺ potentiation of DHP (Peterson and Catterall, 1995) and diltiazem affinity (Dilmac et al., 2003). Furthermore, the IQ domain mutant I1627A potentiates diltiazem block of closed channels in Ba²⁺, and Ca²⁺-dependent inactivation is not required for Ca²⁺ potentiation of diltiazem block (Dilmac et al., 2003). We report here that Ca²⁺ potentiation of verapamil block of closed Ca_v1.2 channels is disrupted by any of the EQ or TA mutations. Furthermore, frequency-dependent block of Ca_v1.2 by verapamil in either Ba²⁺ or Ca²⁺ is also reduced in mutants with single amino acid substitutions at three adjacent positions in domain III (TIIIA, F1117G, EIIIQ) and in EIVQ. The C-terminal tail mutant I1627A increased the sensitivity of closed Ca_v1.2 channels to block by verapamil in Ba²⁺ but did not affect frequency-dependent block in either Ba²⁺ or Ca²⁺.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Construction of Wild-Type and Mutant Ca²⁺ Channels. The Ca_v1.2 (Snutch et al., 1991) and mutant channels were subcloned into pcDNA3 (Invitrogen, Carlsbad, CA). Mutations were introduced as described previously (Hockerman et al., 1997a) or using a modification of the QuikChange site-directed mutagenesis procedure (Stratagene, La Jolla, CA). The desired mutations were verified and the integrity of the clones was confirmed by cDNA sequencing and extensive restriction digest analysis.

Cell Culture. Human tsA 201 cells, a simian virus 40 T-antigen-expressing derivative of human embryonic kidney 293 cells, were maintained in Dulbecco's modified Eagle's medium/Nutrient Mixture F-12 (Invitrogen) enriched with 10% fetal bovine serum.

Expression of Ca²⁺ Channels. Human tsA 201 cells were co-transfected with WT and mutant Ca_v1.2 ₁ subunits, _1b (Pragnell et al., 1991) and ₂ (Ellis et al., 1988), and enhanced green fluorescent protein (GFP) (BD Biosciences Clontech, Palo Alto, CA) such that the molar ratio of the plasmids was 1:1:1:0.8. Cells were transfected using LipofectAMINE 2000 per the manufacturer's protocol (Invitrogen); cells were replated at low density for electrophysiological recording 20 to 24 h after transfection. Experiments were conducted 20 to 48 h after replating.

Electrophysiology. Transfected cells were recognized by GFP fluorescence at 510 nm with excitation at 480 nm. Barium and calcium currents through Ca²⁺ channels were recorded using the whole-cell configuration of the patch-clamp technique. Patch electrodes were pulled from VWR micropipettes (VWR, West Chester, PA) and fire-polished to produce an inner tip diameter of 4 to 6 µm. Currents were recorded using an Axopatch 200B patch-clamp amplifier (Axon Instruments Inc., Union City, CA) and filtered at 1 or 2 kHz (six-pole Bessel filter, -3 dB). Voltage pulses were applied and data were acquired using pClamp8 software (Axon Instruments Inc.). Voltage-dependent leak currents were subtracted using an on-line P/-4 subtraction paradigm. Verapamil, dissolved in bath saline, was applied to cells using an RSC 160 fast perfusion system (Bio-Logic, Claix, France) with constant exchange of the bath solution. I_Ba current was measured in the bath saline containing Tris (150 mM), MgCl₂ (4 mM), and BaCl₂ (10 mM). I_Ca current was measured in the same bath solution, except that Ca²⁺ (10 mM) replaced Ba²⁺. The intracellular saline buffer contained N-methyl-D-glucamine (130 mM), EGTA (10 mM), HEPES (60 mM), MgATP (2 mM), and MgCl₂ (1 mM). (±)-Verapamil (Sigma/RBI, Natick, MA) concentrations were made from 100 mM stock solution in 70% ethanol. The pH of both solutions was adjusted to 7.3 with methanesulfonic acid. All experiments were performed at room temperature (20–23°C).

Data Analysis. Data were analyzed using Clampfit (Axon Instruments, Inc.) and SigmaPlot and SigmaStat software (SPSS Inc., Chicago, IL).

	Results

Top Abstract Materials and Methods Results Discussion References

 
Frequency-Dependent Drug Block and Ca²⁺ Potentiation of Drug Block in Ca_v1.2 Channels. Previous studies (Lee and Tsien, 1983) demonstrated that BZPs and PAAs, but not DHPs, show frequency-dependent block properties. Frequency-dependent block of the DHP nicardipine, the BZP diltiazem, and the PAA verapamil was assessed in our experimental system (Fig. 1A). Verapamil and diltiazem show strong frequency-dependent block of Ca_v1.2 channels, whereas the charged DHP nicardipine does not. Frequency-dependent drug block was measured in the presence of approximately IC₅₀ concentrations of the indicated drugs in 10 mM Ba²⁺. Whole-cell currents through WT Ca_v1.2 channels were measured using 100-ms depolarizations to +10 mV from a holding potential of -60 mV in the absence or presence of the indicated drug concentrations at 0.05 Hz. In cells to which drug was applied, block was allowed to reach equilibrium for several minutes at 0.05-Hz stimulation. The stimulation frequency was then increased to 1 Hz for 20 pulses. This procedure was applied in all subsequent experiments measuring frequency-dependent drug block.

View larger version (32K): [in this window] [in a new window]   Fig. 1. Calcium potentiation of verapamil block and frequency dependence of block of WT Cav1.2 by structurally distinct drugs. A, frequency-dependent block of WT channels by approximately IC50 concentrations of the calcium channel blockers. Frequency-dependent block of WT channels was measured in the absence (closed symbols) or presence (open symbols) of 500 nM nicardipine, 50 µM diltiazem, and 30 µM verapamil. A frequency-dependent protocol was applied in which cells were depolarized from a holding potential of -60 mV to +10 mV for 100 ms every second. Note that, although nicardipine (charged DHP) did not show any frequency-dependent block accumulation, diltiazem (BZP) and verapamil (PAA) showed significant frequency-dependent drug block accumulation at the end of a 20-pulse stimulation. Block was brought to equilibrium with the indicated concentration of each drug at 0.05 Hz followed by a 20-pulse train of depolarizations given at 1 Hz. The asterisks indicate significant differences between the current remaining at the end of the train in the absence and presence of diltiazem or verapamil. B, verapamil block of WT channels at 0.05 Hz in Ba2+ or Ca2+. Representative Ba2+ or Ca2+ currents recorded from tsA 201 cells expressing WT channels in the absence (control) or presence of 30 µM verapamil. Current was elicited using depolarizations to +10 mV from a holding potential of -60 mV for 100 ms every 20 s (0.05 Hz) (Student's t test; , P < 0.001) (mean ± S.E, n = 3–8).

Block of DHPs, BZPs, and PAAs is also potentiated by Ca²⁺ (Lee and Tsien, 1983). That is, the same concentration of drug will block a larger fraction of Ca_v1.2 channels when Ca²⁺, rather than Ba²⁺, is used as the charge carrier. Figure 1B shows Ca²⁺ potentiation of 30 µM verapamil block of WT Ca_v1.2. To determine Ca²⁺ potentiation of drug block, closed-channel block was measured as described at 0.05 Hz in 10 mM Ba²⁺ or 10 mM Ca²⁺. Verapamil (30 µM) blocked a greater fraction of Ca²⁺ current in Ca²⁺ than in Ba²⁺ (62% versus 34%). Thus, these two properties of verapamil block, frequency dependence and Ca²⁺ potentiation, are reconstituted in our experimental system.

Structural Features and Sites of Mutation in Ca_v1.2. The mutations in the pore region and C-terminal tail of L-type Ca_v1.2 channels used in this study are indicated and highlighted in Fig. 2, A and B. We used single amino acid mutants at each of the four highly conserved glutamates that form the selectivity filter. Negatively charged Glu residues were replaced with neutral Gln residues, individually, at the positions E363Q, E709Q, E1118Q, and E1419Q (EIQ, EIIQ, EIIIQ, and EIVQ) in each domain. Note that the structure of verapamil includes a tertiary amino group that is predominantly charged at physiological pH and could potentially interact with the acidic Glu side chains (Fig. 2C). Replacing glutamates with glutamines could potentially disrupt electrostatic interactions between the pore glutamates and both verapamil and Ca²⁺ ions. It is proposed that the negatively charged Glu1118 and Glu1419 residues may interact with the positively charged PAA desmethoxyverapamil via electrostatic interactions (Hockerman et al., 1997a). We also mutated the four highly conserved threonine residues adjacent to the critical Glu residues. These hydrophilic residues are conserved across all high-voltage-activated Ca²⁺ channels and are invariably located two positions upstream of the conserved Glu residues in each domain. Therefore, we singly mutated amino acids Thr361, Thr707, Thr1116, and Thr1417 to Ala (TIA, TIIA, TIIIA, and TIVA). In addition, we studied a mutant in which the L-type-specific residue in domain III, Phe1117, was mutated to Gly (F1117G). We previously demonstrated that the F1117G mutant disrupts frequency dependence and Ca²⁺ potentiation of Ca_v1.2 by diltiazem (Dilmac et al., 2003). Finally we studied the IQ motif mutant, I1627A, in the C-terminal tail of Ca_v1.2, because we previously reported that this mutant increases the affinity of the channel for diltiazem. Of all of these mutants, only TIA did not express functional channels.

View larger version (29K): [in this window] [in a new window]   Fig. 2. Structural features of the 1 subunit of Cav1.2. A, topology of Cav1.2 (1C). Cylinders represent transmembrane segments (1–6) organized into four homologous domains (I–IV). The C- and N-terminal domains are intracellular. Boxes highlight the putative pore-lining regions that contain the elements of the Cav1.2 selectivity filter. B, the amino acid sequences surrounding the Glu residues in each homologous domain that compose the selectivity filter. The conserved Glu and Thr residues are in bold italic type, as is the L-type-specific Phe residue directly adjacent to the conserved Glu residue in domain III (box). The approximate location of Ile 1627 and elements of the IQ Ca2+/calmodulin binding domain are shown. C, chemical structure of verapamil. Note the ionizable tertiary alkylamino group.

Closed-Channel Block of the Glu-to-Gln Mutants by Verapamil in Ba²⁺. WT and mutant EIQ, EIIQ, EIIIQ, and EIVQ Ca_v1.2 channels were coexpressed along with the ₂, _1b, and GFP in tsA 201 cells by transient transfection. Forty-eight hours after transfection, whole-cell Ba²⁺ currents were measured in cells expressing the GFP marker. Closed-channel block was measured from a holding potential of -60 mV using 100-ms depolarization to +10 mV at a frequency of 0.05 Hz. Increasing concentrations of verapamil (1, 5, 10, 50, or 100 µM) were applied to cells under voltage clamp. Representative traces recorded from WT in increasing verapamil concentrations in Ba²⁺ are shown in Fig. 3A. Drug block was brought to equilibrium for each dose, and current amplitude in the presence of drug was normalized to current amplitude in the absence of drug (Fig. 3B). Normalized current amplitudes at each verapamil concentration for several cells were averaged and plotted against drug concentration (Fig. 3B, inset). The verapamil IC₅₀ value for WT was 40.5 ± 0.9 µM. The IC₅₀ value for EIQ was not significantly different from that for WT channels (data not shown). Figure 3C shows the percentage of control current amplitude remaining for WT Ca_v1.2 and each of the EQ mutants in 30 µM verapamil at 0.05 Hz. The percentage of current remaining for all EQ mutants was not significantly different from that of WT Ca_v1.2. Thus, the mutation of any of the conserved glutamate residues does not change the sensitivity of the channel to closed-channel block by verapamil.

View larger version (19K): [in this window] [in a new window]   Fig. 3. Verapamil block of WT and the mutant channels EIQ, EIIQ, EIIIQ, and EIVQ at 0.05 Hz in Ba2+. A, representative Ba2+ current recorded from a tsA 201 cell expressing transiently transfected WT channels in the absence (control) or presence of the indicated concentrations of verapamil. Pulses were from a holding potential of -60 mV to +10 mV for 100 ms every 20 s. B, dose-response relationships for WT channels. The averaged, normalized current amplitudes at 1, 5, 10, 50, and 100 µM verapamil were plotted against the corresponding drug concentration, and the IC50 value was determined by fitting the averaged relative current values at each verapamil concentration to the equation, relative current = 1 - {1/[1 - (IC50/[verapamil])]} (inset). The IC50 value of the WT channels was 40.5 ± 0.9 µM. C, percentage of remaining currents of WT and the mutant channels in 30 µM verapamil. The remaining current values for each case (mean ± S.E. and n = 5–7) were: WT = 58.6 ± 2.6%; EIQ = 58.3 ± 6.5%; EIIQ = 56.3 ± 4.6%; EIIIQ = 64.7 ± 3.4%; EIVQ = 49.2 ± 1.9%. None of the mutant channels were significantly different from the WT channel in sensitivity to block by verapamil under these conditions (one-way ANOVA).

Frequency-Dependent Block of Glu-to-Gln Mutants in Ba²⁺. Although none of the Glu-to-Gln mutants affected verapamil block of closed channels, we examined each mutant for frequency-dependent block by verapamil. Current was measured in the absence and presence of 30 µM verapamil in 10 mM Ba²⁺. In the presence of drug, block was brought to equilibrium at 0.05 Hz followed by a train of 20 depolarizations at 1 Hz as described for Fig. 1. The peak amplitudes of the resulting current traces were normalized to the peak amplitude of pulse 1 of the 20-pulse train, and the averaged data were plotted against pulse number. Mutation of the conserved glutamate residues exhibited distinct effects on the frequency-dependent block of verapamil. Although EIQ and EIIQ did not significantly affect frequency-dependent block (Fig. 4A), accumulation of frequency-dependent block by verapamil was significantly reduced in EIIIQ and EIVQ (Fig. 4, B and C).

View larger version (24K): [in this window] [in a new window]   Fig. 4. Frequency-dependent verapamil block of WT and the mutant EIQ, EIIQ, EIIIQ, and EIVQ Cav1.2 channels. A to C, whole-cell Ba2+ currents were recorded in the absence and presence of 30 µM verapamil using depolarizations to +10 mV for 100 ms from a holding potential of -60 mV, at a frequency of 1 Hz for 20 pulses. Current measured in the presence of verapamil followed equilibration of block by 30 µM verapamil at 0.05 Hz. Relative peak current (mean ± S.E., n = 3–8) in each successive depolarizing pulse is plotted against pulse number in the absence (closed symbols) or presence (open symbols) of 30 µM verapamil for WT and the indicated mutant channels. Smooth lines are fits of the data to a single exponential equation, and these fits indicate that block reaches equilibrium at 25, 51, and 34% of control current for WT, EIIIQ, and EIVQ, respectively. The asterisks indicate that the fraction of current remaining at the end of the 20th pulse was statistically different from that of WT Cav1.2 channels (Student's t test; , P < 0.05; , P < 0.001).

Ca²⁺ Potentiaion of Verapamil Block in WT and EQ mutants. We next assessed verapamil block of WT Ca_v1.2 and the Glu-to-Gln mutant channels using 10 mM Ca²⁺ as the charge carrier. First, we measured the percentage of control current remaining in the presence of 30 µM verapamil at 0.05 Hz (closed-channel block) and compared these values for each channel to those measured in 10 mM Ba²⁺ (Fig. 5A). Verapamil block of WT Ca_v1.2 was potentiated in Ca²⁺ relative to Ba²⁺ such that only 37.8 ± 1.5% of current remained in the presence of 30 µM verapamil in 10 mM Ca²⁺, whereas 58.6 ± 2.6% of current remained in 10 mM Ba²⁺. In contrast, the percentage of current remaining in the presence of 30 µM verapamil was not different in 10 mM Ba²⁺ or 10 mM Ca²⁺ in each of the Glu-to-Gln mutants. Thus, Ca²⁺ potentiation of verapamil block was eliminated in all of the Glu-to-Gln mutants, indicating that neutralization of any of the selectivity filter Glu residues uncouples Ca²⁺ binding from an increase in verapamil affinity.

View larger version (23K): [in this window] [in a new window]   Fig. 5. Verapamil block of WT and Glu-to-Gln mutant Cav1.2 channels in Ca2+. A, Ca2+ potentiation of verapamil block at 0.05 Hz. Verapamil (30 µM) was applied to tsA 201 cells expressing WT and the mutant channels under voltage clamp, as in Fig. 3C, using either 10 mM Ba2+ (closed bars) or 10 mM Ca2+ (open bars) as the charge carrier. The percentage of current remaining in each case was: WTBa = 58.6 ± 2.6; WTCa = 37.8 ± 1.5; EIQBa = 58.3 ± 6.5; EIQCa = 58.4 ± 0.6; EIIQBa = 56.3 ± 4.6; EIIQCa = 59.9 ± 4.7; EIIIQBa = 64.7 ± 3.4; EIIIQCa = 68.7 ± 6.4; EIVQBa = 49.2 ± 1.9; and EIVQCa = 59.6 ± 4.7. The asterisks indicate significant differences between current remaining in Ba2+ and Ca2+. B and C, frequency-dependent block of the indicated channels in the absence (closed symbols) and presence (open symbols) of 30 µM verapamil was measured as described in Fig. 4 except that Ba2+ was replaced with 10 mM Ca2+. Smooth lines are fits of the data to a single exponential equation, and these fits indicate that block reaches equilibrium at 17, 40, and 30% of control current for WT, EIIIQ, and EIVQ, respectively. EIIIQ and EIVQ channels were blocked to a significantly lesser extent than WT channels under these conditions (mean ± S.E, n = 3–8; Student's t test; , P < 0.05; , P < 0.001).

We also examined frequency-dependent verapamil block in the Ca_v1.2 WT and Glu-to-Gln mutations using 10 mM Ca²⁺ as the charge carrier (Fig. 5, B and C). As we observed in Ba²⁺, the EIQ and EIIQ mutants were not different from WT Ca_v1.2 in the extent of verapamil block accumulation by the end of a 1-Hz, 20-pulse train of depolarizations. However, the EIIIQ and EIVQ mutations both exhibited less accumulation of verapamil block under the same conditions. Surprisingly, we also observed that the extent of frequency-dependent verapamil block was not significantly different in the WT Ca_v1.2 channel whether we used 10 mM Ba²⁺ (23 ± 1.6% current remaining) or 10 mM Ca²⁺ (19 ± 2.0%; P = 0.09) as the charge carrier (see Figs. 4 and 5).

Kinetic Analysis of Depolarized Channel Block of WT and EIIIQ. To better understand the reduction in the frequency-dependent block of EIIIQ by verapamil, we compared the verapamil modulation of the voltage dependence and kinetics of inactivation in WT and EIIIQ. The voltage dependence of steady-state inactivation of WT and EIIIQ channels was measured in the absence and presence of 30 µM verapamil in Ba²⁺ (Fig. 6A). From a holding potential of -80 mV, cells were held at the indicated conditioning voltages for 10 s, followed immediately by a 100-ms test pulse to +10 mV. Normalized current amplitudes during the test pulses were plotted against the conditioning voltages. The data were fit to a Boltzman equation, and the voltage at which half the channels were inactivated during the conditioning pulse (V_1/2) was calculated. For WT and EIIIQ channels, V_1/2 was not significantly different in the absence of verapamil (-13.7 ± 0.5 mV versus -15.9 ± 0.6 mV, respectively). However, the leftward shift in V_1/2 induced by 30 µM verapamil was significantly greater in WT channels than in EIIIQ (-43.7 ± 0.8 mV versus -32.9 ± 0.4 mV, respectively) indicating that the EIIIQ mutant disrupted verapamil interaction with the inactivated state of the channel.

View larger version (20K): [in this window] [in a new window]   Fig. 6. Verapamil modulation of inactivation in WT and EIIIQ Cav1.2 channels. A, steady-state inactivation was measured by depolarization of tsA 201 cells to +10 mV for a 100-ms test pulse immediately after a 10-s conditioning pulse at the indicated potentials from a holding potential of -80 mV. Measured Ba2+ current was plotted against the conditioning pulse voltage in the absence (closed symbols) or presence (open symbols) of 30 µM verapamil. The data were fit to the equation, relative current = 1/{1 - exp[(V - V1/2)/k]}, where V is the conditioning potential, V1/2 is the voltage at which half of the channels are inactivated, and k is a slope factor (potential required for an e-fold change). V1/2 values for WT channels in the absence and presence of 30 µM verapamil were -13.7 ± 0.5 and -43.5 ± 0.8 mV, respectively. V1/2 values for EIIIQ channels V1/2)/k]}, where V is in the absence and presence of 30 µM verapamil were -15.9 ± 0.6 and -32.9 ± 0.4 mV, respectively. For WT channels, k values were -7.2 and -10.3 in the absence and presence of verapamil, respectively. For EIIIQ channels, k values were -10.8 and -8.4 in the absence and presence of verapamil, respectively. Time course of recovery from inactivation for WT (B) and EIIIQ (C) channels in the absence (closed symbols) and presence (open symbols) of 30 µM verapamil. Recovery from inactivation produced by a 1-s depolarization to +10 mV from a holding potential of -60 mV was measured using 50-ms test pulses to +10 mV after recovery intervals of 0.01, 0.1, 0.5, 1, 10, 20, and 30 s at -60 mV. The fraction of current recovered was plotted against the recovery interval. In both cases, the time course of recovery was fit to a double-exponential equation, representing fast and slow phases of recovery. For WT channels in the absence of verapamil, the time constants of recovery were fast = 0.28 s and slow = 6.88s, and the fraction of channels recovering with the fast time constant (ffast) was 0.60. In the presence of 30 µM verapamil, fast = 1.83 s, slow = 41.0 s, and ffast = 0.54. In the absence of verapamil, the time constants and the fraction of channels recovering for EIIIQ channels were fast = 0.21 s, slow = 16.9 s, and ffast = 0.46. In the presence of 30 µM verapamil, the time constants for EIIIQ recovery were fast = 1.91 s, slow = 8.07 s, and ffast = 0.61.

We also examined the time course of WT and EIIIQ channel recovery from depolarized channel verapamil block. Whole-cell Ba²⁺ current was equilibrated with 30 µM verapamil using 100-ms depolarizations to +10 mV from a holding potential of -60 mV at 0.05 Hz. After the current reached steady state, a 1-s depolarization to +10 mV was applied in the continued presence of drug, followed by a return to -60 mV. The recovery of the current after a 1-s depolarization to +10 mV in the presence or absence of drug was measured by depolarizing cells to +10 mV for 50 ms from -60 mV at intervals of 0.01, 0.1, 0.5, 1, 10, 20, and 30 s. To determine the fraction of WT and EIIIQ channels recovered from verapamil block, recovered current was expressed as a fraction of the peak current during the 1-s depolarization, plotted against recovery interval in the absence or presence of 30 µM verapamil, and fit to a double-exponential function (Fig. 6, B and C). The fast phase of recovery from inactivation (i.e., in the absence of drug) was similar in both WT and EIIIQ channels (WT: _fast = 283 ms, f_fast = 0.60; EIIIQ: _fast = 213 ms, f_fast = 0.46), whereas the slow phase of recovery was slower for EIIIQ (WT: _slow = 6.88 s; EIIIQ: _slow = 16.9 s). For the WT channel in the presence of 30 µM verapamil, both time constants for recovery were increased (WT: _fast = 1.83 s, _slow = 41.0; s f_fast = 0.54). For EIIIQ, 30 µM verapamil slowed the fast phase of recovery (_fast = 1.91 s; f_fast = 0.46), but the time constant for the slow phase of recovery was actually accelerated in comparison to EIIIQ in the absence of drug (_slow = 8.07 s). Thus, our results suggest that the marked decrease in frequency-dependent block of EIIIQ by verapamil compared with WT could be the result of faster dissociation of verapamil from EIIIQ at -60 mV. Recovery of current after a 1-s depolarization to +10 mV for EIIQ was not different from that for WT in either the absence or presence of 30 µ verapamil (data not shown).

Verapamil blocks Ca_v1.2 channels with higher affinity depolarized potentials. As shown in Fig. 7, A and B, even after bringing current to steady state in the presence of µM verapamil with depolarizations to +10 mV at 0.05 both WT and EIIIQ channels are rapidly and completely blocked during a 1-s depolarization to +10 mV. We compared the rate of current decay during a 1-s depolarization to + mV, from a holding potential of -60 mV, in the absence presence of 30 µM verapamil for the WT and EIIIQ channels. Current traces were normalized to peak current and fit with a double-exponential function. In the absence of drug (Fig. 7B, filled bars), both WT and EIIIQ channels inactivated with virtually identical fast and slow time constants, and fractions of channels inactivating with each time constant (WT: _fast = 0.16 ± 0.02 s, _slow = 1.78 ± 0.11 s, f_fast = 0.10 0.02, f_slow = 0.70 ± 0.13; EIIIQ: _fast = 0.15 ± 0.06 s, _slow 1.46 ± 0.53 s, f_fast = 0.12 ± 0.04, f_slow = 0.62 ± 0.07). In presence of 30 µM verapamil (Fig. 7B, open bars), WT and EIIIQ channel inactivation followed a single time constant and was virtually complete (WT: = 0.29 ± 0.02 s, f = 1.02 0.02; EIIIQ = 0.21 ± 0.02 s, f = 1.05 ± 0.01 s). Thus, in both WT and EIIIQ channels, 30 µM verapamil blocks virtually channels with a time course that resembles the fast-inactivating state, leading to complete inactivation by the end of 1-s depolarization.

View larger version (19K): [in this window] [in a new window]   Fig. 7. Onset of verapamil block of depolarized WT and EIIIQ channels. A, representative traces of Ba2+ current through WT (left) and EIIIQ (right) channels measured during a 1-s depolarization to +10 mV, from a holding potential of -60 mV, in the absence (control) or presence of 30 µM verapamil. B, time course of inactivation in the absence (filled bars) and presence (open bars) of 30 µM verapamil for WT Cav1.2 (left) and EIIIQ (right). In the absence of verapamil, WT inactivation was fit by a double-exponential equation with a fast time constant (F = 0.16 ± 0.02 s) and a slow time constant (S = 1.78 ± 0.11 s) corresponding to the fraction fast (fF = 0.10 ± 0.02) and the fraction slow (fS = 0.70 ± 0.13). In the presence of 30 µM verapamil, WT inactivation was single-exponential with a relatively fast time constant (F = 0.29 ± 0.02 s). In the absence of drug, EIIIQ channel inactivation was fit to a double-exponential equation with a fast time constant similar to that of WT channels (F = 0.15 ± 0.06 s; fF = 0.12 ± 0.04) and a slow time constant (S = 1.46 ± 0.53 s; fS = 0.62 ± 0.07). In the presence of 30 µM verapamil, EIIIQ channels displayed a single fast component of inactivation (F = 0.21 ± 0.01 s). The asterisks indicate significant differences between the indicated parameter in the presence and absence of 30 µM verapamil (Student's t test; , P < 0.01) (mean ± S.E., n = 3–6; #, not detected).

Effect of the F1117G and I1627A Mutants on Closed-Channel and Frequency-Dependent Block. Because both Phe1117 and Ile1627 have been implicated in modulation of Ca_v1.2 channel function by Ca²⁺ (Peterson and Catterall, 1995; Zuhlke et al., 1999), we examined their roles both Ca²⁺ potentiation and frequency dependence of verapamil block. Block of closed F1117G and I1627A channels was measured at 0.05 Hz in 30 µM verapamil in Ba²⁺ Ca²⁺ (Fig. 8A). Like the conserved pore glutamate mutants, the F1117G mutant disrupted Ca²⁺ potentiation of verapamil block such that current remaining in the presence of µM verapamil was the same whether Ca²⁺ or Ba²⁺ was charge carrier. Likewise, the extent of closed-channel verapamil block of the I1627A mutant was not different when measured in Ba²⁺ or Ca²⁺ from a holding potential of -60 mV. However, the percentage of I1627A current remaining the presence of 30 µM verapamil in either Ba²⁺ or Ca²⁺ was not different from that of WT in Ca²⁺. From a holding potential of -80 mV, the percentage of I1627A current remaining in 30 µM verapamil was not different from that of WT either Ba²⁺ (70.1 ± 4.7) or Ca²⁺ (71.0 ± 2.4). Frequency-dependent block was measured in Ba²⁺ and Ca²⁺ in 30 µ verapamil at 1 Hz as described in Figs. 4 and 5. Although I1627A mutant did not significantly affect accumulation frequency-dependent block in either Ba²⁺ or Ca²⁺, F1117G mutant significantly reduced frequency-dependent block by verapamil at the end of a 20-pulse, 1-Hz train depolarizations in either Ba²⁺ or Ca²⁺ (Fig. 8, B and Thus, whereas F1117G disrupts both Ca²⁺ potentiation and frequency dependence of verapamil block, I1627A enhances closed-channel verapamil block in Ba²⁺ to the level of that seen in Ca²⁺ with WT Ca_v1.2.

View larger version (35K): [in this window] [in a new window]   Fig. 8. Frequency and Ca2+ modulation of verapamil block in F1117G and I1627A. A, verapamil (30 µM) block of WT, F1117G, and I1627A Cav1.2 channels in 10 mM Ba2+ (closed bars) or 10 mM Ca2+ (open bars) at 0.05 Hz. The percentage of current remaining (mean ± S.E., n = 3–9) in each case was WTBa = 58.6 ± 2.6; WTCa = 37.8 ± 1.5; F1117GBa = 61.4 ± 6.2; F1117GCa = 65.5 ± 3.3.; I1627ABa = 36.4 ± 2.4; and I1627ACa = 31.0 ± 0.3. The percentage of current remaining for I1627A when measured from a holding potential of -80 mV was 70.1 ± 4.7 (n = 5) in 10 mM Ba2+, and 71.0 ± 2.3 (n = 4) in 10 mM Ca2+. Asterisks indicate a significant difference between current remaining in Ba2+ and Ca2+ (Student's t test; , P < 0.001). The percentage of current remaining in Ba2+ for I1627A was significantly different from both WT and F1117G when measured from a holding potential of -60 mV (one-way ANOVA with Tukey's post hoc test; ++, P < 0.01). The percentage of current remaining for F1117G in Ca2+ was significantly different from both WT and I1627A (one-way ANOVA with Tukey's post hoc test; # # #, P < 0.001). B and C, frequency-dependent block of the indicated channels (mean ± S.E., n = 3–8) was measured as described in Figs. 4 and 5 in Ba2+ or Ca2+ in the absence (closed symbols) and presence (open symbols) of 30 µM verapamil. Smooth lines are fits of the data to a single-exponential equation, and these fits indicate that block reaches equilibrium at 25, 28, and 22% of control current in Ba2+, and at 17, 43, and 21% of control current for WT, F1117G, and I1627A, respectively. In either Ba2+ or Ca2+, only F1117G was significantly different from WT in the extent of block at the end of the 20-pulse, 1-Hz train of depolarizations (Student's t test; , P < 0.001). D, the voltage dependence of inactivation was measured as described in Fig. 6A. Measured Ba2+ currents were plotted against the conditioning pulse voltage in the absence (closed symbols) or presence (open symbols) of 30 µM verapamil. The data were fit to a Boltzman equation as in Fig. 6A. V1/2 values for WT channels in the absence and presence of 30 µM verapamil were -13.7 ± 0.5 and -43.5 ± 0.8 mV, respectively. V1/2 values for F1117G channels in the absence and presence of 30 µM verapamil were -23.4 ± 0.5 and -35.4 ± 0.3 mV, respectively. V1/2 values for I1627A channels in the absence and presence of 30 µM verapamil were -28.4 ± 0.8 and -57.4 ± 0.9 mV, respectively. Slope (k) values for WT were -7.2 and -10.3 in the absence and presence of verapamil, respectively. Slope (k) values for F1117G were -5.8 and -6.0 in the absence and presence of verapamil, respectively. Slope (k) values for I1627A channels were -11.52 and -10.39 in the absence and presence of verapamil, respectively.

To better understand the basis for the observed differences in verapamil block of F1117G and I1627A, we measured voltage dependence of inactivation for these two mutants in the absence and presence of 30 µM verapamil in 10 mM Ba²⁺ (Fig. 8D). Although the V_1/2 values in the absence of drug for both channels were shifted to more negative potentials compared with WT (F1117G, -23.4 ± 0.5 mV; I1627A, -28.4 ± 0.8 mV), the leftward shift in the presence of 30 µM verapamil was 30 mV for both WT and I1627A channels but only 12 mV for F1117G. Thus, despite the observation that both mutants cause a leftward shift in V_1/2, they are markedly different in the manner in which they interact with verapamil.

Frequency Dependence and Ca²⁺ Potentiation in the TA Mutants. Because mutation of two of the four selectivity filter Glu residues and the L-type-specific Phe1117 residue markedly reduced frequency dependence and Ca²⁺ potentiation of verapamil block, we examined the role of the adjacent, conserved Thr residues in each domain. We found that mutation of Thr361 in domain I to Ala is apparently not tolerated, because neither this single point mutant, nor a quadruple mutant, with the conserved Thr in each domain changed to Ala, yielded functional channels. A hydrogen bond donor/acceptor in this position may be a critical channel function because the T361C mutant channel was found to be functional (Wu et al., 2000). Closed-channel verapamil block of TIIA, TIIIA, and TIVA was measured in Ba²⁺ and Ca²⁺ at 0.05 Hz as described above (Fig. 9A). The TIIA, TIIIA, and TIVA mutants did not differ from WT channels in the fraction of current remaining in the presence of 30 µM verapamil when 10 mM Ba²⁺ was used as the charge carrier. In Ca²⁺, TIIA, TIIIA, and TIVA were all less sensitive to verapamil block at 0.05 Hz than WT channel (i.e., Ca²⁺ potentiation of verapamil block was disrupted). In Ba²⁺, only TIIIA among the TA mutants accumulated significantly less frequency-dependent verapamil block at the end of a 20-pulse, 1-Hz train of depolarizations (Fig. 9B). In Ca²⁺, both TIIA and TIIIA significantly reduced the accumulation of verapamil block at 1 Hz (Fig. 9C). Thus, whereas TIIA, TIIIA, and TIVA all reduced the sensitivity of closed channels to verapamil in Ca²⁺, only TIIA and TIIIA disrupted frequency-dependent block.

View larger version (26K): [in this window] [in a new window]   Fig. 9. Frequency and Ca2+ modulation of verapamil block of TIIA, TIIIA, and TIVA. A, Ca2+ potentiation of verapamil block at 0.05 Hz in 30 µM verapamil was measured as described in Fig. 5A, using either 10 mM Ba2+ (closed bars) or 10 mM Ca2+ (open bars) as the charge carrier. The percentage of current remaining in each case was (mean ± S.E.): WTBa = 58.6 ± 2.6; WTCa = 37.8 ± 1.5; TIIABa = 45.3 ± 3.2; TIIACa = 80.2 ± 4.2; TIIIABa = 73.0 ± 6.1; TIIIACa = 73.5 ± 3.6; TIVABa = 56.5 ± 3.4; and TIVACa = 65.7 ± 6.8. The percentage of current remaining in Ba2+ was not significantly different among WT and all of the mutants (one-way ANOVA). The percentage of current remaining in Ca2+ in each of the mutant channels was significantly different from that of WT (one-way ANOVA with Tukey's post hoc test; ##, P < 0.01; # # #, P < 0.001). The asterisks indicate significant differences between current remaining in Ba2+ and Ca2+ for WT and TIIA (Student's t test; , P < 0.001). B and C, frequency-dependent verapamil block of WT and the Thr-to-Ala mutant channels. Frequency-dependent verapamil block (percentage of current remaining; mean ± S.E, n = 3–8) of the indicated channels was measured as described in Figs. 4 and 5 in Ba2+ or Ca2+ in the absence (closed symbols) and presence (open symbols) of 30 µM verapamil. Smooth lines are fits of the data to a single-exponential equation, and these fits indicate that block reaches equilibrium at 25, 39, 22, and 19% of control Ba2+ current and 17, 35, 27, and 25% of control Ca2+ current for WT, TIIIA, TIIA, and TIVA, respectively. The TIIIA mutant was blocked to a significantly lesser extent than WT channels in both Ba2+ and Ca2+, whereas the TIIA mutant was blocked to a significantly lesser extent than WT in Ca2+ (Student's t test; , P < 0.05).

Inactivation and Ion Permeability in the Thr-to-Ala Mutants. Since the Thr-to-Ala mutations had the unexpected effect of disrupting Ca²⁺ potentiation of closed-channel verapamil block, we examined the effect of these mutations on Ca²⁺ acceleration of inactivation kinetics and ion permeability. Figure 10A shows normalized, averaged current traces for WT and the Thr-to-Ala mutants in 10 mM Ba²⁺ or Ca²⁺. We compared the extent of inactivation in these channels by measuring the percentage of peak current inactivated at the end of a 1-s depolarization to +10 mV from a holding potential of -60 mV (Fig. 10B). No significant difference in the extent of inactivation was observed among WT, TIIA, TIIIA, TIVA, and the domain III mutant F1117G in Ba²⁺. In Ca²⁺, only the TIIA mutant exhibited a decrease in the percentage of current inactivated at 1 s compared with WT. However, Ca²⁺ significantly increased the extent of inactivation at 1 s compared with Ba²⁺ in all three Thr-to-Ala mutants (i.e., Ca²⁺-dependent inactivation was intact).

View larger version (25K): [in this window] [in a new window]   Fig. 10. Inactivation properties of WT and the mutant channels TIIA, TIIIA, and TIVA in Ba2+ and Ca2+. A, averaged, normalized current traces for WT channels and each of the indicated mutant channels elicited during a 1-s depolarization to +10 mV from a holding potential of -60 mV in extracellular solution containing 10 mM Ba2+ (left) or 10 mM Ca2+ (right) are shown. B, summary of the percentage of channels inactivated at the end of a 1-s depolarization in Ba2+ or Ca2+. All channels tested displayed significant increases in the percentage of channels inactivated at the end of a 1-s depolarization in Ca2+ compared with Ba2+ (mean ± S.E, n = 3–7; Student's t test; , P < 0.05; , P < 0.001). There was no significant difference in the percentage of inactivated channels in Ba2+ among WT and mutant channels (one-way ANOVA). In Ca2+, the percentage of channels inactivated at 1 s was significantly less in TIIA than in WT channels (WTCa = 91.1 ± 1.7%, n = 7; TIIACa = 77.0 ± 1.6%, n = 5; # # #, P < 0.001; one-way ANOVA with Tukey's post hoc test).

We also examined the permeability of the Thr-to-Ala mutants for Ba²⁺ relative to Ca²⁺. Figure 11A shows whole-cell currents recorded in the same cell in the presence of 10 mM Ba²⁺ or Ca²⁺ in tsA 201 cells expressing WT and the indicated mutant Ca_v1.2 channels. Current was elicited by 100-ms depolarizations to +10 mV at 0.05 Hz, from a holding potential of -60 mV while the extracellular solution was switched between 10 mM Ba²⁺ and 10 mM Ca²⁺. With Ca²⁺ as the charge carrier, the peak amplitude current measured in WT and TIVA channels was reduced relative to the peak current measured in Ba²⁺. In contrast, the peak current amplitudes measured in Ca²⁺ were greater than those measured in Ba²⁺ for the TIIA and TIIIA mutants. Figure 11B shows the relative change in current amplitude, for each of the channels tested, upon switching from Ba²⁺-to Ca²⁺-containing extracellular solution. Thus, whereas the Thr-to-Ala mutations disrupt Ca²⁺ potentiation of verapamil block of closed channels, they do not eliminate Ca²⁺-dependent inactivation. Furthermore, the TIIA and TIIIA mutations mediate an increased permeability of Ca_v1.2 for Ca²⁺ relative to Ba²⁺.

View larger version (16K): [in this window] [in a new window]   Fig. 11. Permeability of Ca2+ relative to Ba2+ in WT and mutant Cav1.2 channels. A, current traces recorded from single tsA 201 cells expressing WT, TIIA, TIIIA, and TIVA mutant channels recorded in 10 mM Ba2+ or 10 mM Ca2+. Current was elicited using a 100-ms depolarization to +10 mV from a holding potential of -60 mV. Note the increase in peak current in Ca2+ relative to peak current in Ba2+ in the TIIA and TIIIA mutants. B, change in peak current amplitude measured at +10 mV from a holding potential of -60 mV when the extracellular solution is switched from 10 mM Ba2+ to 10 mM Ca2+. The values shown are mean ± S.E. (n = 3–10) for WT Cav1.2 channels and each of the indicated mutants.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
Distinct Effects of Selectivity Filter Mutants on Ca²⁺ Potentiation of Closed-Channel Block and Frequency-Dependent Block
We have shown that mutation of any of the selectivity filter glutamates does not reduce the sensitivity of Ca_v1.2 channels to closed-channel verapamil block in Ba²⁺ (Fig. 3). In a previous study, EIIIQ and EIVQ increased the IC₅₀ of the nearly identical compound, desmethoxyverapamil (D888), by 15-and 20-fold, respectively, at a stimulation frequency of 0.1 Hz in 10 mM Ba²⁺. D888 is more potent than verapamil in blocking Ca_v1.2 (Hockerman et al., 1995) and binding to Ca_v1.1 (Goll et al., 1984). A binding site between IIIS6 (Hockerman et al., 1997a) and IVS6 (Hockerman et al., 1995) was shown to account for the higher affinity of D888 for Ca_v1.2 at low-frequency stimulation. Furthermore, mutation of the three amino acids in IVS6 critical for high-affinity D888 block of Ca_v1.2 did not affect the potency of verapamil block at 0.1 Hz (Johnson et al., 1996). Thus, the present study supports previous work suggesting that D888 and verapamil interact with distinct binding determinants on closed Ca_v1.2 channels (i.e., at low-frequency stimulation).

I1627A is unique among the mutants used in this study because it is in the cytoplasmic C-terminal tail, not, apparently, in the vicinity of the transmembrane segments composing the small-molecule drug binding sites. However, as we previously observed with diltiazem (Dilmac et al., 2003), I1627A increased the sensitivity of closed Ca_v1.2 channels to verapamil block in Ba²⁺, whereas frequency-dependent block was not different from WT in Ba²⁺ or Ca²⁺. The I1627A mutation shifted the voltage dependence of inactivation by approximately -20 mV compared with WT. Because we found that closed-channel block of I1627A by 30 µM verapamil from a holding potential of -80 mV was not different from that of WT measured from a holding potential of -60 mV, it seems that the increase in sensitivity to verapamil in I1627A at -60 mV results from the shift in voltage dependence of inactivation. However, Ca²⁺ potentiation of closed-channel block of I1627A by verapamil is abolished at both potentials. These results suggest that Ca²⁺-calmodulin binding to the C-terminal IQ motif, which is disrupted in I1627A (Zuhlke et al., 1999), may be required for Ca²⁺ potentiation of closed-channel verapamil block. The loss of Ca²⁺ potentiation of closed-channel block in all of the EQ mutations suggests that Ca²⁺ potentiation also requires Ca²⁺ binding in the channel pore. Taken together with our previous observation that Ca²⁺-dependent inactivation is disrupted in the EQ mutations (Dilmac et al., 2003), these results also suggest that the consequences of Ca²⁺-calmodulin binding to the C-terminal tail of Ca_v1.2 may include conformational changes in the pore region of the channel.

Ca²⁺ potentiation of closed-channel verapamil block was disrupted by mutation of any of the Glu-to-Gln mutants, F1117G, and all of the Thr-to-Ala mutants. The modulation of verapamil block by Ca²⁺ is distinct from what we previously observed for diltiazem (Dilmac et al., 2003), where EIIIQ, EIVQ, and F1117G, but not EIQ or EIIQ, disrupted Ca²⁺ potentiation of closed-channel diltiazem block. However, the effect of the Thr-to-Ala mutations on this property of diltiazem block was not examined in that study. Similar to diltiazem, DHPs bind in a pocket composed of IIIS5, IIIS6, and IVS6 (Hockerman et al., 1997c). Moreover, Ca²⁺ potentiation of DHP binding also requires Glu1118, Glu1419, and Phe1117 but not Glu363 or Glu709 (Peterson and Catterall, 1995). Taken together, these data suggest that Ca²⁺ binding to a site in domains III and IV modulates the diltiazem (Hockerman et al., 2000) and DHP (Hockerman et al., 1997c) binding sites, which include IIIS6 and IVS6, whereas Ca²⁺ binding to multiple sites in the selectivity filter is required for high-affinity verapamil block of closed channels. This scenario is consistent with a binding site for verapamil on closed channels that is distinct from that for DHPs and diltiazem.

Frequency Dependence of Verapamil Block in Ba²⁺ or Ca²⁺. In contrast to our results using low-frequency stimulation (0.05 Hz), we found that multiple mutations clustered in the domain III region of the selectivity filter markedly disrupted frequency-dependent verapamil block (Figs. 4, 5, 8, and 9). Disruption of block was similar whether Ba²⁺ or Ca²⁺ was used as the charge carrier. Surprisingly, we found that, unlike our previous observations with diltiazem (Dilmac et al., 2003), frequency-dependent block of Ca_v1.2 is not appreciably potentiated by Ca²⁺ in our experimental system. In addition, the F1117G and EIIIQ mutations both significantly decreased the leftward shift in V_1/2 induced by 30 µM verapamil [F1117G, = 12 mV (Fig. 8D); EIIIQ, = 17 mV (Fig. 6A)] compared with that observed in WT Ca_v1.2 [ = 30 mV (Fig. 6A)]. Furthermore, the EIIIQ mutation increases the rate of recovery from depolarized-channel verapamil block compared with WT Ca_v1.2 (Fig. 6, B and C) but does not affect the shift of channels to the fast-inactivating state induced by verapamil (Fig. 7B). Our finding that verapamil accelerates the inactivation of Ca_v1.2 channels and increases the fraction of channels inactivating is consistent with previous studies (Johnson et al., 1996; Sokolov et al., 2001). However, our results with EIIIQ are in contrast to observations with mutations in IVS6 which accelerated both verapamil block and recovery from verapamil block at +10 mV (Johnson et al., 1996). The effect of the F1117G mutation on frequency-dependent verapamil block seems to be Ca²⁺-dependent, because a fit of the data to a single exponential function indicates that steady-state block of F1117G in Ba²⁺ is not different from that of WT Ca_v1.2, although the current remaining at the 20th pulse is clearly different from that of WT (Fig. 8B). We examined the inactivation rate of F1117G in the presence and absence of 30 µM verapamil in Ba²⁺ and did not find any significant difference from WT (N. Dilmac and G. H. Hockerman, unpublished data). It is not clear how the F1117G mutant is specifically slowing development of frequency-dependent verapamil block in Ba²⁺. However, it is clear that F1117G decreases affinity for verapamil in Ca²⁺, because both the steady-state block determined from a fit of the data and the current remaining at the 20th pulse are clearly different from that of WT (Fig. 8C). Likewise, TIIA disrupts both closed-channel and frequency-dependent verapamil in Ca²⁺ but not in Ba²⁺ (Fig. 9). Our data support a scenario in which frequency-dependent verapamil block of Ca_v1.2 channels (i.e., channels in the open and/or inactivated state) involves binding to a site that includes Thr1116, F1117G, Glu1118, and Glu1419, which may be accessed from the intracellular mouth of the pore. Taken together, our results suggest that verapamil may block closed channels and open/inactivated channels by binding to distinct sites, which include TIIA as a common determinant in Ca²⁺. This scenario was proposed previously based upon experiments using a membrane-impermeable, quaternary ammonium derivative of D888. These experiments suggested that closed-channel block was mediated by a binding site accessible from the extracellular surface of the channel, whereas frequency-dependent block was mediated by a binding site accessible from the intracellular side of the membrane (Berjukov et al., 1996). In addition, our finding that, with the exception of Phe1117, the molecular determinants for frequency-dependent verapamil block are conserved among voltage-gated Ca²⁺ channels (i.e., Thr1116, Glu1118, Glu1419) may explain the relatively low discrimination of L-type and non-L-type Ca²⁺ channels by verapamil (Diochot et al., 1995; Ishibashi et al., 1995).

Conserved Residues Thr707, Thr1116, and Thr1417 and Ca_v1.2 Activity in Ca²⁺. Our observation that the conserved Thr residues adjacent to the selectivity filter glutamates contribute to Ca²⁺ modulation of verapamil block of closed channels was unexpected. We had previously shown that the EIIQ, EIIIQ, and EIVQ mutations disrupt Ca²⁺-dependent inactivation, and EIIIQ and EIVQ increase the permeability of Ca²⁺ relative to Ba²⁺ (Dilmac et al., 2003). Therefore, we examined these properties of channel activity in the Thr-to-Ala mutants. Our finding that none of the Thr-to-Ala mutations disrupts Ca²⁺-dependent inactivation (Fig. 10) is consistent with our observations with the F1117G mutant, which disrupts Ca²⁺ modulation of diltiazem and verapamil block but does not disrupt Ca²⁺-dependent inactivation. Our observations with the Thr-to-Ala mutants also further highlight the specific role of the selectivity filter Glu residues, among other adjacent residues, in mediating Ca²⁺-dependent inactivation. However, Thr707 and Thr1116, as well as Phe1117, apparently play a role in forming the ion permeation pathway, because mutation of these residues increases the permeability of Ca²⁺ relative to Ba²⁺ (Fig. 11).

In summary, we have shown that Ca²⁺ potentiation of closed-channel block of Ca_v1.2 by verapamil requires all four selectivity filter glutamates, whereas amino acid residues adjacent to the selectivity filter in domain III contribute to the high-affinity binding of verapamil to the depolarized state of the channel. Furthermore, because Ca²⁺ does not potentiate frequency-dependent verapamil block, Ca²⁺ is not likely to potentiate binding of verapamil to the depolarized state of the channel. Our results further differentiate the molecular mechanisms of verapamil and diltiazem block of Ca_v1.2, suggest that distinct binding sites may mediate verapamil block of closed and depolarized channels, and suggest that common molecular determinants may mediate the block of non-L-type voltage-gated Ca²⁺ channels by verapamil.

	Footnotes

 
This work was supported by Scientist Development Grant 9930016N from the American Heart Association (to G.H.H.).

Article, publication date, and citation information can be found at http://molpharm.aspetjournals.org.

doi:10.1124/mol.104.000893.

ABBREVIATIONS: DHP, dihydropyridine; PAA, phenylalkylamine; BZP, benzothiazepine; WT, wild-type; GFP, green fluorescent protein; D888, desmethoxyverapamil; ANOVA, analysis of variance.

Address correspondence to: Gregory Hockerman, 575 Stadium Mall Dr., West Lafayette, IN 47907-2091. E-mail: gregh{at}pharmacy.purdue.edu

	References

Top Abstract Materials and Methods Results Discussion References

 
Almers W and McCleskey EW (1984) Non-selective conductance in calcium channels of frog muscle: calcium selectivity in a single-file pore. J Physiol 353: 585-608.[Abstract/Free Full Text]

Berjukov S, Aczel S, Beyer B, Kimball SD, Dichtl M, Hering S, and Striessnig J (1996) Extra- and intracellular action of quaternary devapamil on muscle L-type Ca²⁺-channels. Br J Pharmacol 119: 1197-1202.[Medline]

Bers DM and Perez-Reyes E (1999) Ca channels in cardiac myocytes: structure and function in Ca influx and intracellular Ca release. Cardiovasc Res 42: 339-360.[Abstract/Free Full Text]

Cai D, Mulle JG, and Yue DT (1997) Inhibition of recombinant Ca²⁺ channels by benzothiazepines and phenylalkylamines: class-specific pharmacology and underlying molecular determinants. Mol Pharmacol 51: 872-881.[Abstract/Free Full Text]

Dilmac N, Hilliard N, and Hockerman GH (2003) Molecular determinants of Ca²⁺ potentiation of diltiazem block and Ca²⁺-dependent inactivation in the pore region of Ca_v1.2. Mol Pharmacol 64: 491-501.[Abstract/Free Full Text]

Diochot S, Richard S, Baldy-Moulinier M, Nargeot J, and Valmier J (1995) Dihydropyridines, phenylalkylamines, and benzothiazepines block N-, P/Q- and R-type calcium currents. Pflueg Arch Eur J Physiol 431: 10-19.[CrossRef][Medline]

Doring F, Degtiar VE, Grabner M, Striessnig J, Hering S, and Glossman H (1996) Transfer of L-type calcium channel IVS6 segment increases phenylalkylamine sensitivity of 1a. J Biol Chem 271: 11745-11749.[Abstract/Free Full Text]

Ellis SB, Williams ME, Ways NR, Brenner R, Sharp AH, Leung AT, Campbell KP, McKenna E, Koch WJ, Hui A, et al. (1988) Sequence and expression of mRNAs encoding the alpha 1 and alpha 2 subunits of a DHP-sensitive calcium channel. Science (Wash DC) 241: 1661-1664.[Abstract/Free Full Text]

Fleckenstein A and Fleckenstein-Grun G (1980) Cardiovascular protection by Ca antagonists. Eur Heart J 1 (Suppl B): 15-21.[Medline]

Goll A, Ferry DR, Striessnig J, Schober M, and Glossmann H (1984) (-)-[³H]Desmethoxyverapamil, a novel Ca²⁺ channel probe. Binding characteristics and target size analysis of its receptor in skeletal muscle. FEBS Lett 176: 371-377.[CrossRef][Medline]

Hockerman GH, Dilmac N, Scheuer T, and Catterall WA (2000) Molecular determinants of diltiazem block in domains IIIS6 and IVS6 of L-type Ca²⁺ channels. Mol Pharmacol 58: 1264-1270.[Medline]

Hockerman GH, Johnson BD, Abbott MR, Scheuer T, and Catterall WA (1997a) Molecular determinants of high affinity phenylalkylamine block of L-type calcium channels in transmembrane segment IIIS6 and the pore region of the 1 subunit. J Biol Chem 272: 18759-18765.[Abstract/Free Full Text]

Hockerman GH, Johnson BD, Scheuer T, and Catterall WA (1995) Molecular determinants of high affinity phenylalkylamine block of L-type calcium channels. J Biol Chem 270: 22119-22122.[Abstract/Free Full Text]

Hockerman GH, Peterson BZ, Johnson BD, and Catterall WA (1997b) Molecular determinants of drug binding and action on L-type calcium channels. Annu Rev Pharmacol Toxicol 37: 361-396.[CrossRef][Medline]

Hockerman GH, Peterson BZ, Sharp E, Tanada TN, Scheuer T, and Catterall WA (1997c) Construction of a high-affinity receptor site for dihydropyridine agonists and antagonists by single amino acid substitutions in a non-L-type Ca²⁺ channel. Proc Natl Acad Sci USA 94: 14906-14911.[Abstract/Free Full Text]

Ishibashi H, Yatani A, and Akaike N (1995) Block of P-type Ca²⁺ channels in freshly dissociated rat cerebellar Purkinje neurons by diltiazem and verapamil. Brain Res 695: 88-91.[CrossRef][Medline]

Johnson BD, Hockerman GH, Scheuer T, and Catterall WA (1996) Distinct effects of mutations in transmembrane segment IVS6 on block of L-type calcium channels by structurally similar phenylalkylamines. Mol Pharmacol 50: 1388-1400.[Abstract]

Jones SW (1998) Overview of voltage-dependent calcium channels. J Bioenerg Biomembr 30: 299-312.[CrossRef][Medline]

Lee KS and Tsien RW (1983) Mechanism of calcium channel blockade by verapamil, D600, diltiazem and nitrendipine in single dialysed heart cells. Nature (Lond) 302: 790-794.[CrossRef][Medline]

Nawrath H and Wegener JW (1997) Kinetics and state-dependent effects of verapamil on cardiac L-type calcium channels. Naunyn Schmiedeberg's Arch Pharmacol 355: 79-86.[CrossRef][Medline]

Peterson BZ and Catterall WA (1995) Calcium binding in the pore of L-type calcium channels modulates high affinity dihydropyridine binding. J Biol Chem 270: 18201-18204.[Abstract/Free Full Text]

Peterson BZ, DeMaria CD, Adelman JP, and Yue DT (1999) Calmodulin is the Ca²⁺ sensor for Ca²⁺-dependent inactivation of L-type calcium channels. Neuron 22: 549-558.[CrossRef][Medline]

Pragnell M, Sakamoto J, Jay SD, and Campbell KP (1991) Cloning and tissue-specific expression of the brain calcium channel beta-subunit. FEBS Lett 291: 253-258.[CrossRef][Medline]

Snutch TP, Tomlinson WJ, Leonard JP, and Gilbert MM (1991) Distinct calcium channels are generated by alternative splicing and are differentially expressed in the mammalian CNS. Neuron 7: 45-57.[CrossRef][Medline]

Sokolov S, Timin E, and Hering S (2001) On the role of Ca²⁺- and voltage-dependent inactivation in Ca(v)1.2 sensitivity for the phenylalkylamine (-)gallopamil. Circ Res 89: 700-708.[Abstract/Free Full Text]

Tanabe T, Takeshima H, Mikami A, Flockerzi V, Takahashi H, Kangawa K, Kojima M, Matsuo H, Hirose T, and Numa S (1987) Primary structure of the receptor for calcium channel blockers from skeletal muscle. Nature (Lond) 328: 313-318.[CrossRef][Medline]

Wu XS, Edwards HD, and Sather WA (2000) Side chain orientation in the selectivity filter of a voltage-gated Ca²⁺ channel. J Biol Chem 275: 31778-31785.[Abstract/Free Full Text]

Yang J, Ellinor PT, Sather WA, Zhang JF, and Tsien RW (1993) Molecular determinants of Ca²⁺ selectivity and ion permeation in L-type Ca²⁺ channels. Nature (Lond) 366: 158-161.[CrossRef][Medline]

Zuhlke RD, Pitt GS, Deisseroth K, Tsien RW, and Reuter H (1999) Calmodulin supports both inactivation and facilitation of L-type calcium channels. Nature (Lond) 399: 159-162.[CrossRef][Medline]