Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Voltage-gated calcium channels are important regulators of calcium influx in a number of cell types. Calcium entry through these channels activates a plethora of intracellular events, from the broad stimulation of gene expression, calcium-dependent second messenger cascades, and cell proliferation to the specific release of neurotransmitter within the nervous system and contraction in smooth and cardiac muscle (Tsien et al., 1991; Wheeler et al., 1994; Dunlap et al., 1995). A number of different types of calcium channels have been identified in native tissues and divided based on their biophysical profiles into low-voltage-activated (LVA) and high-voltage-activated (HVA) channels (Nowycky et al., 1985; Tsien et al., 1991). LVA channels first activate at relatively hyperpolarized potentials and rapidly inactivate (Akaike et al., 1989; Takahashi et al., 1991). By contrast, HVA channels require stronger membrane depolarizations to activate and can be divided further into N-, P/Q-, R- and L-types based on their pharmacological properties (for review, see Stea et al., 1995; Zamponi, 1997). Molecular cloning has revealed that HVA channels are heteromultimers composed of a pore-forming ₁ subunit plus ancillary ₂-, , and possibly  subunits (Pragnell et al., 1994; Klugbauer et al., 1999, 2000; for review, see Catterall, 2000), whereas LVA channels seem to contain only the ₁ subunit (Lacinova et al., 2000). So far, 10 different types of calcium channel ₁ subunits have been identified and shown to encode the previously identified native calcium channel isoforms. Expression studies show that alternative splicing of _1A generates both P- and Q-type Ca²⁺ channels (Bourinet et al., 1999); _1B encodes N-type channels (Dubel et al., 1992); _1C, _1D, and _1F are L-type channels (Williams et al., 1992; Bech-Hansen et al., 1998); _1G, _1H, and _1I form T-type channels (i.e., McRory et al., 2001); _1E may encode R-type channels (Soong et al., 1993; Tottene et al., 1996); and _1S encodes the skeletal muscle L-type channel isoform (Tanabe et al., 1987).

Dihydropyridine (DHP) antagonists of L-type calcium channels are widely used therapeutics in the treatment of hypertension, angina, arrhythmias, congestive heart failure, cardiomyopathy, atherosclerosis, and cerebral and peripheral vascular disorders (Janis and Triggle, 1990). The ability of DHPs to both block and enhance native L-type calcium currents has been well documented (Bean, 1984; Bechem and Hoffmann, 1993; Bangalore et al., 1994; Peterson and Catterall, 1995). Although they are considered to be selective inhibitors of the L-type, a number of reports have suggested that native T-type channels may also show sensitivity to certain commonly used DHPs (Akaike et al., 1989; Romanin et al., 1992; Formenti et al., 1993; Santi et al., 1996). In contrast, for cloned calcium channels, only _1S, _1C, and _1D have been shown to be sensitive to the most commonly used DHPs (Williams et al., 1992; Tomlinson et al., 1993; Grabner et al., 1996; Berjukow et al., 2000; Koschak et al., 2001). None of the remaining calcium channel subtypes cloned from brain exhibit significant DHP sensitivity; however, information on DHP block of cloned T-type calcium channels is still sparse.

Here, we report the calcium channel blocking action of a novel series of DHP derivatives [dialkyl 1,4-dihydro-4-(2'alkoxy-6'-pentadecylphenyl)-2,6-dimethyl-3,5 pyridine dicarboxylates] derived from anacardic acid, a phenolic constituent present in cashew nut shell liquid (Paul and Yeddanapalli, 1956). Our data show that these novel DHPs have the propensity to mediate high-affinity inhibition of cloned L-type calcium channels. However, more interestingly, these compounds also effectively block T-type calcium channels, in some cases with a substantial unparalleled selectivity over the L-type channel.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemistry. Anacardic acid was isolated from cashew nut shell liquid by a novel method reported by Paramashivappa et al. (2001). As illustrated in Fig. 1, the ene mixture of anacardic acid obtained by this method was hydrogenated using Pd/C to obtain saturated anacardic acid (Fig. 1, scheme 1.1), which after alkylation with dimethyl or diethyl sulfate using potassium carbonate gave the dialkylated derivative (Fig. 1, scheme 1.2). Di isopropyl anacardic acid was obtained by using isopropyl bromide (Fig. 1 , scheme 2). Dialkylated anacardic acid (Fig. 1, scheme 1.2) was reduced to alcohol (Fig. 1, scheme 1.3) using lithium aluminum hydride and oxidized to corresponding aldehyde (Fig. 1, scheme 1.4) using pyridinium chloro chromate. The aldehyde is then converted to 1,4-dihydropyridine using the modified Hantzsch procedure. All the 1,4-dihydro pyridines were converted to pyridine derivatives (Yadav et al., 2000; Fig. 1, scheme 3) for further characterization of the compounds.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Synthesis of the series of dialkyl 1,4-dihydro-4-(2'alkoxy-6'-pentadecylphenyl)-2,6-dimethyl-3,5 pyridine dicarboxylates from anacardic acid as described in detail under Materials and Methods section. In scheme 1, the reagents used in each step were as follows: (CH3)2SO4/K2CO3, acetone, reflux 3 h (a); LiAlH4, Tetrahydrofuran, reflux 3 h (b); PCC, dichloromethane, rt. 3 h (c); CH3COCH2COOR1/piperidine/acetic acid, n-butanol, rt. 3 h (d); and (CH3)(NH2)C = CH(COOR2), n-butanol, reflux 30 h (e).

Preparation of Dialkyl 1,4-Dihydro-4-(2'alkoxy-6'-pentadecylphenyl)-2,6-dimethyl-3,5 Pyridine Dicarboxylates. 2-Ethoxy-6-pentadecyl-2-benzaldehyde (3 g, 8.3 mmol) and ethylacetoacetate (1.08 g, 8.3 mmol) were dissolved in n-butanol (20 ml). Acetic acid (0.5 g, 8.3 mmol) and piperidine (0.7 g, 8.3 mmol) were added and stirred at room temperature for 3-4 h. Ethyl-3-amino crotonate (1.1 g, 8.3 mmol) was then added and refluxed for 30 h. 2-n-Butanol was distilled off under vacuum and product was purified by column chromatography using silicagel (100-200 mesh) with hexane/EtoAc (94:6) solvent system to give dialkyl 1,4-dihydro-4-(2'alkoxy-6'-pentadecylphenyl)-2,6-dimethyl-3,5 pyridine dicarboxylates (PPK4) as white powder (0.69 g, 14.2%).

Cell Culture and Transfections. The human embryonic kidney tSA-201 cell line was used to transiently express cloned rat _1C and human _1G) calcium channels. The cells were grown in standard Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum, and 0.5 mg/ml penicillin streptomycin. Once the cells were at 90% confluence, they were split with trypsin-EGTA, and plated on glass coverslips at 10% confluence 12 h before transfection. Immediately before transfection, fresh media was given to the cells. A standard calcium phosphate protocol was used to transiently transfect the cells with cDNA constructs encoding for the calcium channel ₁ subunit (_1C or _1G) and the green fluorescent protein EGFP (7 and 4 µg of DNA, respectively). DNA constructs encoding ancillary _1b and ₂- subunits were added to the _1C mix (7 µg of each DNA). After a 12-h incubation at 36°C, the cells were washed with fresh DMEM and allowed to recover for an additional 12 h. Subsequently, the cells were incubated at 28°C in 5% CO₂ for 1 to 3 days before recording. The cDNA constructs encoding calcium channel _1C, ₂-, and _1b subunits were kindly donated by Dr. Terry Snutch, and the "b" splice variant of the human _1G construct was isolated as we have described elsewhere (A. M. Beedle, J. Hamid, and G. W. Zamponi, submitted). The electrophysiological and biophysical properties of our human _1G clone (i.e., half-activation potential of 47 mV, half-inactivation potential of 79 mV in 2 mM barium) closely parallel those described for the "b" splice variant by Monteil et al. (2000).

Electrophysiological Recordings. Expressed calcium channels were screened at room temperature for macroscopic currents using an Axopatch 200B amplifier (Axon Instruments, Union City, CA) linked to a personal computer equipped with pCLAMP version 6.0. Patch pipettes (Sutter borosilicate glass, BF150-86-15), pulled using a microelectrode puller (P87; Sutter Instruments, Novato, CA) to a typical resistance of ~4 M and fire-polished with a Narashige microforge, were used for whole-cell patch clamp recordings. Typically, series resistance and capacitance were compensated by at least 85% to minimize contamination of records by voltage errors caused by insufficient voltage clamp. The internal pipette solution consisted of 108 mM cesium methanesulfonate, 4 mM MgCl₂, 9 mM EGTA, and 9 mM HEPES, pH 7.2 with cesium hydroxide. The external bath solution consisted of 5 mM BaCl₂, 1 mM MgCl₂, 10 mM HEPES, 40 mM tetraethylammonium chloride, 10 mM glucose, and 90 mM CsCl, pH 7.2 with tetraethylammonium hydroxide. The DHP derivatives were solubilized in dimethyl sulfoxide to a concentration of 100 mM. From the stock, the compounds were diluted to the appropriate concentration in external solution. The final concentration of dimethyl sulfoxide in the recording solution was kept below 1:1000, which did not affect calcium channel activity. External solutions were applied to cells expressing calcium channels through a solenoid-driven perfusion system. In our experiments, rundown of _1C and _1G currents was typically minimal (<10% over the course of a typical experiment), and the few cells that exhibited significant rundown were discarded. To minimize the possibility of contamination from rundown during prolonged drug applications, cells were repeatedly pulsed in the presence of 5 mM barium control solution until a stable baseline was obtained. Only after currents had reached equilibrium were the compounds bath-applied at the concentrations detailed in the figures. The time course of the effects of the various PPK-compounds was determined at a test potential of 0 mV (_1C) or 20 mV (_1G) applied every 10 s for 200 ms from a holding potential of 80 mV. After the drug effect had stabilized, the cells were washed with control external solution for up to 10 min to determine whether any effects were reversible. To determine the effect of the holding potential on drug affinity, the holding potential was changed to either 60 mV or 40 mV and the duration of the test potential was reduced to 30 ms and 15 ms, respectively. Steady-state inactivation curves were obtained in the presence and absence of compounds by providing conditioning pulses between 100 mV and + 20 mV in 10 mV increments for 5 s and measuring the corresponding currents at a standard potential of 10 mV (_1C) or 20 mV (_1G).

Data Analysis. All data were analyzed using Clampfit (Axon Instruments) and Sigmaplot 4.0 (Jandel Scientific, Costa Madre, CA). Steady-state inactivation curves were fit with the Boltzmann equation to obtain half inactivation potentials. The dose-response curves were fit with the Hill equation to determine the IC₅₀ values of the compounds. The data are given as mean ± S.E. and the numbers in parentheses reflect the number of experiments per group. Student's t tests were carried out to ascertain the statistical significance of the results (P < 0.05).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Structure Activity Relationship of the Dialkyl 1,4-Dihydro-4-(2'alkoxy-6'-pentadecylphenyl)-2,6-dimethyl-3,5 Pyridine Dicarboxylates. Anacardic acid is a phenolic constituent present in cashew nut shell liquid (Paul and Yeddanapalli, 1956) and is reported to exhibit antibacterial (Kubo et al., 1993a), antitumor (Kubo et al., 1993b) and antiacne (Kubo et al., 1994a) properties. It is also known to inhibit such medicinally important enzymes as prostaglandin synthase (Grazzini et al., 1991), lipoxygenase (Shobha et al., 1994), and tyrosinase (Kubo et al., 1994b). We have used saturated anacardic acid (2-hydroxy-6-pentadecyl benzoic acid) to prepare a series of 18 novel dialkyl 1,4-dihydro-4-(2'alkoxy-6'-pentadecylphenyl)-2,6-dimethyl-3,5 pyridine dicarboxylates (termed PPK-1 through 18; Table 1). Because these derivatives are structurally related to DHP compounds of the nifedipine class, we investigated their abilities to block transiently expressed L-type calcium channels via whole-cell patch-clamp recordings. In these experiments, cells were held at 80 mV to approximate the membrane potential of native excitable cells and to isolate tonic block from any putative inactivation state dependent effects. As seen in Fig. 2, under these conditions, the application of a standard concentration for each compound (10 µM) results in varying degrees of inhibition of _1C + ₂-  + _1b calcium channels transiently expressed in tSA-201 cells ranging from about 25% (i.e., PPK-2) to virtually complete block (PPK-12). Examination of the R group structures of these compounds reveals a moderate structure activity relationship: increasing the size of alkoxy group on 4-phenyl ring and ester substituent in the 3,5 positions increases the efficacy of these compounds, where the role of the C₁₅ alkyl chain is to ensure perpendicularity between 4-aryl and dihydropyridine rings. Moreover, modification of the dihydropyridine ring of derivatives with inherent symmetry to nonsymmetric analogs seems to result in more potent block, as suggested by earlier work (Rovnyak et al., 1992) on nifedipine analogs. Hence, dialkyl 1,4-dihydro-4-(2'alkoxy-6'-pentadecylphenyl)-2,6-dimethyl-3,5 pyridine dicarboxylates can yield effective blockers of L-type calcium channels with SAR in both the phenyl and DHP rings.

View larger version (60K): [in this window] [in a new window]   Fig. 2.   Inhibitory effects of the PPK series on L-type (1C +2-  + 1b) calcium channels transiently expressed in tSA-201 cells. Whole-cell currents were recorded in 5 mM external barium by stepping from a holding potential of 80 mV to a test depolarization of 0 mV. The degree of block was determined from the percentage of peak current inhibition obtained in the presence of each of the compounds. The bars reflect means, error bars are S.D., and numbers in parentheses indicate the number of experiments.

Analysis of PPK-12 Block of L-Type Calcium Channels. Figs. 3 and 4 more precisely characterize the effects of the most efficient blocker identified in the initial screen, PPK-12. Figure 3A illustrates a typical time course of the effects of PPK-12 on transiently expressed L-type calcium channels. Block developed rapidly, but was not reversible after washout, indicating either a tight interaction with the blocking site, or accumulation of the compounds in the plasma membrane. Figure 3B displays current records obtained with transiently expressed L-type calcium channels in the presence and the absence of PPK-12. As seen from the records, significant block already occurs at 300 nM concentrations, whereas block is greater than half-maximal at 1 µM PPK-12. Upon inspection of the current records, a significant speeding of the macroscopic time constant for current inactivation becomes apparent. This could reflect additional open channel block developing during the time course of the membrane depolarization or, alternatively, a drug-induced inactivation of the currents similar to what has been proposed to occur for certain DHPs (Berjukow et al., 2000). Irrespective of the mechanism, this speeding would serve to further depress calcium influx during prolonged membrane depolarization. Figure 3C displays a complete dose-response curve for PPK-12 block obtained at a holding potential of 80 mV. The data are nicely described by the Michaelis-Menten equation, yielding an apparent IC₅₀ of 540 nM. For comparison, the IC₅₀ obtained for nifedipine block under identical conditions was 170 nM (see Fig. 6), indicating that PPK-12 is only about 3-fold less effective in inhibiting L-type currents compared with established DHP compounds.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Effects of PPK-12 on transiently expressed L-type channels at a holding potential of 80 mV. A, representative time course of PPK-12 block of transiently expressed L-type calcium channels. Symbols reflect peak current amplitudes in the presence and absence of 10 µM PPK-12. B, representative current records obtained before and after application of PPK-12. Note the inhibition of peak current amplitude and the speeding of the rate of current decay. C, ensemble dose-response curve obtained from 10 experiments. The data were fitted with the Hill equation (solid line). The parameters obtained from the fit were as follows; IC50 = 0.54 µM, nH=0.95). Error bars reflect S.E. and the symbols are means of 7 to 10 independent determinations of drug effect.

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Dependence of PPK-12 block of L-type calcium channel on holding potential. A, ensemble steady-state inactivation curves obtained in the absence () and presence () of 1 µM PPK-12. The data were fitted with the Boltzmann equation (solid lines). The half inactivation potentials obtained from the fits were 36.5 mV and 56.8 mV in the absence and the presence of 1 µM PPK-12, respectively. In each case, steady-state inactivation curves in the absence and presence of PPK-12 were recorded from the same cell for a total of 10 cells. Error bars are S.E. B, dependence of half-inactivation potential on the concentration of PPK-12. The asterisks indicate statistical significance relative to control (P < 0.05; Student's t test). C, dependence of the blocking effects of 300 nM PPK-12 on the holding potential. Note the significant increase in the degree of block (P < 0.05) at more depolarized membrane potentials. The asterisks indicate significance relative to the data obtained at 80 mV (p < 0.05).

PPK-12 and PPK-5 Block T-Type Calcium Channels. In native cells, a number of DHPs have been shown to exert a low-affinity block of T-type calcium channels. To determine whether this occurred for dialkyl 1,4-dihydro-4-(2'alkoxy-6'-pentadecylphenyl)-2,6-dimethyl-3,5 pyridine dicarboxylates, we examined the effects of PPK-12 on _1G (Cav 3.1) T-type calcium channels transiently expressed in tSA-201 cells. Application of 3 µM PPK-12 mediated a dramatic reduction in _1G peak current amplitude without any additional speeding of the macroscopic time course of inactivation that was typically irreversible after washout (Figs. 5, A and B). Interestingly, the dose dependence of PPK-12 block of _1G was surprisingly steep (n_H=1.88; Fig. 5C). Given the negative inactivation range of T-type calcium channels and our standard holding potential of 80 mV, a drug induced hyperpolarizing shift in the midpoint of the steady-state inactivation curve could result in additional inhibition, thus increasing the slope of the dose-response curve. To examine this possibility, we assessed the effects of PPK-12 on half-inactivation potential. As seen in Fig. 5D, 1 µM PPK-12 resulted in a small (~5 mV) but statistically significant (p < 0.05, paired t test) hyperpolarizing shift in the midpoint of the steady state inactivation curve (Fig. 5D). Although the magnitude of this effect did not approach that seen with the L-type calcium channel isoform, it may contribute to the shape of the dose-response curve shown in Fig. 5C. Overall, these data indicate that at hyperpolarized membrane potentials, PPK-12 blocks L- and T-type calcium channels with almost equal efficacy.

View larger version (25K): [in this window] [in a new window]   Fig. 5.   Effects of PPK-12 on transiently expressed T-type calcium channels. A, representative time course of peak current block induced by 10 µM PPK-12. Note that the inhibition is virtually complete and not reversible after washout. B, effect of 3 µM PPK-12 on whole-cell current records elicited by stepping from 80 mV to a test potential of 20 mV. C, dose dependence of the effects of PPK-12 on transiently expressed T-type calcium channels. Data from 12 experiments are included in the figure. Symbols reflect means from 8 to 12 individual determinations, the solid line is a fit according to the Hill equation (IC50 = 1.65 µM, nH=1.88), error bars reflect S.E. D, comparison of the half-inactivation potentials obtained in the absence and presence of 1 µM PPK-12. The asterisk indicates statistical significance (P < 0.005, paired Student's t test).

View larger version (26K): [in this window] [in a new window]   Fig. 6.   Block and selectivity of PPK-5 for T-type calcium channels. A, representative time course of block of T-type calcium channel by 3 µM PPK-5. Note the lack of reversibility. B, current records obtained in the absence and presence of PPK-5 under the same conditions as those in Fig. 5B. C, dose dependence of PPK-5 action on transiently expressed T-type channels obtained from 11 different cells. The experimental conditions were the same as in Fig. 5C, the IC50 and Hill coefficient obtained from the fit were 1.14 µM and 2.06, respectively. D, relative selectivities of nifedipine, PPK-12, and PPK-5 among L- and T-type calcium channels. Note that the IC50 values shown on the ordinate are on a logarithmic scale. Error bars are S.E., numbers in parentheses reflect the number of cells used to calculate the means.

Nifedipine Antagonizes PPK-5 Action. In view of the overall structural similarity between our series of compounds and nifedipine, it is possible that nifedipine may bind to T-type calcium channels without significantly affecting current activity. To examine this possibility, we compared the time course of development of PPK-5 block with and without prior application of 30 µM nifedipine. Figure 7 shows that although channels that were exposed to nifedipine could still be completely blocked by 3 µM PPK-5, the time constant for development of PPK-5 block was slowed 2-fold [from 70.6 ± 11.6 s (n = 5) to 138.9 ± 12.6 s (n = 4), p < 0.05]. The simplest explanation for this effect is that nifedipine may indeed occupy part of the PPK-5 binding site without actually blocking current activity and that for PPK-5 to mediate its blocking effects, nifedipine may have to first dissociate from the channel. If so, these would suggest that T-type calcium channels contain a binding pocket for the DHP pharmacophore, but that channel block requires the presence of specific R groups on the DHP molecule.

View larger version (20K): [in this window] [in a new window]   Fig. 7.   Slowing of the development of PPK-5 block of T-type calcium channels by prior application of nifedipine. A, whole-cell current records obtained from 1G calcium channels with (right) and without (left) prior application of 30 µM nifedipine (for 3 min), measured 90 s after application of 3 µM PPK-5. The currents obtained in the absence of PPK-5 were normalized relative to each other to facilitate comparison of the PPK-5 effect. Note that in the absence of pretreatment with nifedipine, a 90-s application of PPK-5 mediates 65% block, whereas the same exposure in cells pretreated with nifedipine results in a much smaller degree of block (~20%), and longer exposure times are required achieve a comparable level of inhibition (in this example, 170 s is required for 3 µM PPK-5 to block 70% of the current). B, quantitative comparison of the effects illustrated in A. Note that nifedipine dramatically increases the time constant for development of PPK-5 block. A total of five experiments is included in the figure.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Comparison with Previous Work. Of the DHP blockers, dialkyl 1,4-dihydro-4-aryl-2,6-dimethyl-3,5-pyridine dicarboxylates of the nifedipine class have been found to offer longer bioavailability and greater tissue selectivity (Goldmann and Stoltefuss, 1991). The superior blocking efficacy of this class of compounds containing bulky ortho substituents on the 4-aryl ring has been attributed to a forced perpendicular orientation between 4-aryl ring and 1,4-dihydropyridine ring (Loev et al., 1974). Although 4-aryl(2',6'-di substituted)1,4 DHPs were predicted to possess high affinity for the L-type calcium channel (Loev et al., 1974; Coburn et al., 1988), these derivatives received less attention because of low product yield obtained even with modified Hantzsch synthesis (Loev et al., 1974). For this study, we prepared a series of dialkyl 1,4-dihydro-4-(2'alkoxy-6'-pentadecylphenyl)-2,6-dimethyl-3,5 pyridine dicarboxylates starting from saturated anacardic acid (2-hydroxy-6-pentadecyl benzoic acid). These compounds are structurally related to nifedipine (see Table 1), and it is thus not surprising that they are able to inhibit L-type calcium channel activity, albeit to a somewhat lesser extent compared with nifedipine per se. The most efficacious L-type channel blocker of the series, PPK-12, showed several similarities to nifedipine action: first, like nifedipine, PPK-12 accelerated the time course of current decay (Lee and Tsien, 1983), which may reflect either a true effect on inactivation of the channel (Berjukow et al., 2000) or additional open channel block that develops during the test depolarization. Second, like nifedipine and several other DHPs (Lee and Tsien, 1983; Shen et al., 2000), PPK-12 mediated a robust shift in the midpoint of the steady-state inactivation curve toward more hyperpolarized potentials, indicative of strong inactivated channel block. Unlike nifedipine, however, the action of PPK-12 or any of the other derivatives examined, could not be reversed during the time course of a typical experiment. In view of evidence that DHPs must at least partially partition into the plasma membrane to mediate L-type channel block (Kass et al., 1991; Strubing et al., 1993; Bangalore et al., 1994), it is possible that the added hydrophobicity arising from the pentadecyl side chain results in accumulation of these compounds in the plasma membrane, thus precluding rapid reversal of the blocking action.

Mechanism of T-Type Calcium Channel Block. Perhaps the most striking result from our study is the strong inhibition of T-type calcium channel activity by PPK-5 and PPK-12. Moreover, the 40-fold selectivity of PPK-5 for T-type over L-type calcium channels is particularly interesting, although one must be aware that this value was obtained at a holding potential of 80 mV, where little inactivated channel block of the L-type isoform is expected. A large (~20 mV) hyperpolarizing shift in the midpoint of the voltage dependence of L-type channel inactivation (such as that observed with PPK-12, Fig. 4A) would be expected to reduce the degree of selectivity for T-type channels at very depolarized holding potentials; at membrane potentials more negative than 40 mV, however, significant T-type channel selectivity would remain (for example, the apparent blocking affinity for PPK-12 increased by less than 5-fold when the holding potential was switched from 80 mV to 40 mV).