Hyperpolarization-activated cyclic nucleotide gated (HCN) ion channels regulate membrane potential, neurotransmitter release, and patterning of synchronized neuronal activity. Currently, there is an intense debate as to whether or not these ion channels play a pro- or anticonvulsant role in vivo. To gain an insight into this question, we have examined how inhibitors of the response mediated by HCN channels (referred to as I-h) affect epileptiform activity induced in adult hippocampal slices. The archetypal Ih blocker ZD-7288 produced a concentration-dependent inhibition of both nonsynaptic- (low Ca2+/elevated K+ aCSF) and synaptic- (low Mg2+ aCSF, elevated K+ aCSF or convulsant application (bicuculline or pentylenetetrazol)) based epileptiform. activities. The IC50 value for ZD-7288 induced inhibition of epileptiform. activity was similar across all forms of epileptiform response and was below concentrations producing nonspecific inhibition of glutamatergic synaptic transmission. Furthermore, capsazepine, which exhibits similar potency to ZD-7288 at inhibiting Ih, failed to inhibit glutamatergic synaptic transmission per se but produced a significant inhibition of bicuculline-induced epileptiform activity. These data suggest that broad spectrum inhibition Of Ih reduces neuronal hyperexcitability in the hippocampus. Synapse

We examined if a range of carbonic anhydrase inhibitors (CAIs) interacted with the high-voltage activated voltage-sensitive calcium channels (VSCCs) encoded by the human alpha(1E) subunit. Whole-cell recordings were made from HEK293 cells stably expressing human alpha(1E)beta(3)-mediated calcium channels. SNX-482 (an alpha(1E) inhibitor) blocked alpha(1E)-mediated VSCCs with an IC(50) close to 10 nM. The anticonvulsant CAI ethoxyzolamide also inhibited these currents, with an IC(50) close to 1 microM, and produced an accompanying 20-mV hyperpolarizing shift in the steady-state inactivation profile. Other structurally diverse CAIs (e.g., acetazolamide and benzolamide) produced approximately 30 - 40% inhibition of alpha(1E)beta(3)-mediated Ca(2+) currents at 10 microM. Topiramate (10 microM), an anticonvulsant with CAI activity, inhibited these currents by 68 +/- 7%. This off-target activity of CAIs at VSCCs may contribute to some of the effects they produce both in vitro and in vivo.

Relatively little has been published on the pharmacology of R-type and T-type Ca(2+) channels. Here, whole-cell Ca(2+) channel currents were recorded from human embryonic kidney 293 cell-lines transfected with either alpha1E subunits (R-type currents) or alpha1G or alpha1I subunits (T-type currents). R-type currents were inhibited by sipatrigine and the related compound 202W92 (R-(-)-2,4-diamino-6-(fluromethyl)-5-(2,3,5-trichlorophenyl)pyrimidine) with IC(50) 10 and 56 microM, respectively. A therapeutic concentration of lamotrigine (10 microM) inhibited R-type currents (30%) but was without effect on alpha1I-mediated T-type currents. Lamotrigine was also a weak inhibitor of T-type currents mediated by alpha1G subunits (<10% inhibition by 100 microM).

Electrophysiological analysis of human embryonic kidney 293 cells stably expressing recombinant channels was used to compare how the biophysical properties of the low-voltage-activated Ca(2+) channels encoded by the alpha(1G) (Ca(V)3.1) or alpha(1I) (Ca(V)3.3) subunits shape their responses to excitatory synaptic potentials. In medium containing 2 mM extracellular Ca(2+) standard current-voltage relationships demonstrated both channel types to be clearly low-voltage activated with significant slowly activating current responses being observed at -66 mV. At all test potentials examined, activation of Ca(V)3.3 was substantially slower than that of Ca(V)3.1. To probe how these different T-type channels might respond to excitatory postsynaptic potentials (EPSPs), mock EPSPs with different kinetic profiles were created from the sum of exponentials. These waveforms were then used as command templates in voltage-clamp experiments. Ca(V)3.1-mediated channels responded effectively to both rapidly decaying mock EPSPs and slowly decaying EPSPs. In contrast, Ca(V)3.3-mediated channels were poorly gated by rapidly decaying EPSPs but were effectively activated by the more prolonged synaptic response. When activated with mock EPSPs Ca(V)3.3-mediated currents were more resistant to steady-state depolarisation of the pre-stimulus holding potential. Ca(V)3.3 currents were also more resistant to repetitive application of prolonged EPSPs, which caused substantial inactivation of Ca(V)3.1-mediated currents. The addition of a single mock action potential to the peak of a rapidly decaying EPSP voltage-clamp template greatly enhanced the currents produced by either Ca(V)3.1 or Ca(V)3.3-expressing cells. This facilitatory effect was considerably greater for Ca(V)3.3-mediated channels. From these data we suggest that the slow activation kinetics of Ca(V)3.3-mediated T-type channels enable them to respond selectively to either slow or suprathreshold synaptic potentials.

1. Voltage-sensitive Ca(2+) channels (VSCCs) are often heteromultimeric complexes. The VSCC subtype specifically expressed by skeletal muscle has long been known to contain a gamma subunit, gamma(1), that is only expressed in this tissue. Recent work, initiated by the identification of the mutation present in the stargazer mouse, has led to the identification of a series of novel potential Ca(2+) channel gamma subunits expressed in the CNS. 2. Based on bioinformatic techniques we identified and cloned the human gamma(2), gamma(3) and gamma(4) subunits. 3. TaqMan analysis was used to quantitatively characterise the mRNA expression patterns of all the gamma subunits. All three subunits were extensively expressed in adult brain with overlapping but subunit-specific distributions. gamma(2) and gamma(3) were almost entirely restricted to the brain, but gamma(4) expression was seen in a broad range of peripheral tissues. 4. Using a myc epitope the gamma(2) subunit was tagged both intracellularly at the C-terminus and on a predicted extracellular site between the first and second transmembrane domains. The cellular distribution was then examined immunocytochemically, which indicated that a substantial proportion of the cellular pool of the gamma(2) subunit was present on the plasma membrane and provided initial evidence for the predicted transmembrane topology of the gamma subunits. 5. Using co-transfection techniques we investigated the functional effects of each of the gamma subunits on the biophysics of the T-type VSCC encoded by the alpha(1I) subunit. This revealed a substantially slowed rate of deactivation in the presence of gamma(2). In contrast, there was no significant corresponding effect of either gamma(3) or gamma(4) on alpha(1I) subunit-mediated currents.

202W92 (R-(−)-2,4-diamino-6-(fluromethyl)-5-(2,3,5-trichlorophenyl)pyrimidine) is a novel compound in the same chemical series as the antiepileptic drug lamotrigine and the neuroprotective sipatrigine. Here 202W92 was quantitatively assessed as a neuroprotective agent in focal cerebral ischaemia, and as an inhibitor of sodium and calcium channels and of synaptic transmission. In the rat permanent middle cerebral artery occlusion (MCAO) model of acute focal ischaemia, 202W92 reduced infarct volume by 75% in cortex and by 80% in basal ganglia, with ED50 approximately 2 mg/kg (single i.v. dose, 10 min post-occlusion). In whole-cell current recordings from single cells, 202W92 completely and reversibly inhibited voltage gated sodium channels (IC50 3×10−6 M) in rat freshly-isolated cortical neurons and in the GH3 pituitary cell line. 202W92 also inhibited a nifedipine-sensitive fraction (approximately 35%) of native high-voltage-activated (HVA) calcium current in rat cortical neurons (IC50 15×10−6 M) and weakly inhibited low-voltage-activated (LVA) calcium currents of the recombinant α1I-mediated T-type (IC50>100×10−6 M). The drug inhibited the amplitude and frequency of 4-aminopyridine-evoked glutamatergic excitatory post-synaptic currents (EPSCs). In conclusion, 202W92 is an effective neuroprotective agent when administered post-ischaemia and a potent sodium channel inhibitor in vitro.

Chlorpromazine is a neuroleptic antipsychotic agent with a long history of clinical use. Its primary mode of action is thought to be through modulation of monoaminergic inter-neuronal communication; however, its side-effect profile indicates substantial activities in other systems. Recent work has begun to uncover actions of this compound on ion channels. In this light we have investigated the actions of chlorpromazine on the recombinant alpha1E subunit-encoded voltage-sensitive Ca2+ channel (VSCC) that is believed to encode drug-resistant R-type currents found in neurones and other cells. Chlorpromazine produced a dose-dependent antagonism of these channels that was reversed on drug removal. The mean IC50 was close to 10 microM. At this concentration, the level of antagonism observed was dependent on the membrane potential, with greater inhibition being observed at more negative test potentials. Furthermore, chlorpromazine induced substantial changes in the steady-state inactivation properties of alpha1Ebeta3-mediated currents, although it was not seen to elicit a corresponding change in inactivation kinetics. These results are discussed with regard to the possible clinical mechanisms of chlorpromazine actions.

T-type Ca(2+) currents were recorded in 2 mM Ca(2+) from HEK 293 cells stably expressing recombinant low-voltage-activated Ca(2+) channel subunits. Current-voltage relationships revealed that these currents were low-voltage activated in nature and could be reversibly antagonised by mibefradil, a known T-type channel blocker. At a test potential of -25 mV alpha(1I)-mediated Ca(2+) currents were rapidly and reversibly inhibited by 1-100 microM BW619C89 (IC(50)=14 microM, Hill coefficient 1.3). In contrast to its actions on N-type Ca(2+) channels, a near IC(50) dose (10 microM) of BW619C89 produced no alterations in either the kinetics or voltage-dependence of T-type currents. In additional single dose experiments, currents mediated by rat alpha(1G), human alpha(1H) or human alpha(1I) channel subunits were also inhibited by BW619C89. Overall our data indicate that T-type Ca(2+) channels are more potently blocked by BW619C89 than either type-II Na(+) channels or N-type Ca(2+) channels. It seems, therefore, that inhibition of low-voltage-activated Ca(2+) channels is likely to contribute to the anticonvulsant and neuroprotective actions of this and related compounds.

T-type Ca(2+) currents were recorded in 2 mM Ca(2+) from HEK293 cells stably expressing the low voltage-activated Ca(2+) channel sub-unit alpha(1I). These currents were inhibited by the known Ca(2+) channel antagonist mibefradil with an IC(50) close to 1 microM. SB-209712 (1,6,bis¿1-[4-(3-phenylpropyl)piperidinyl]¿hexane), a compound originally developed as a high voltage-activated Ca(2+) channel blocker, proved to be a more potent T-type channel antagonist, exhibiting an IC(50) in the region of 500 nM. The antagonism produced by SB-209712 was reversed following drug removal and the observed antagonism exhibited little or no voltage-dependence with respect to either holding or test potential. These data indicate that SB-209712 is amongst the most potent known non-peptide T-type channel antagonists and thus may have some use in understanding the role of these channels in cellular function.

The effects of a novel synthetic macrocyclic polyamine (LY310315) were investigated on recombinant human N-type Ca2+ channels stabley expressed in HEK293 cells. LY310315 proved to be a potent and reversible N-type Ca2+ channel antagonist. Inhibition by this compound was dose-dependent with an IC50 of ∼0.4 μM at pH 7.35. LY310315 blocked very rapidly at all concentrations tested. Upon washout, recovery of the Ca2+ current developed with a time constant of ∼30 s. Use-dependence in the development of block indicated that voltage-dependent transitions in the channel protein were required to permit significant inhibition. Application of >100 times the IC50 dose of LY310315 to the interior of the cell produced no detectable Ca2+ current inhibition. LY310315 had no effects on the kinetics of channel activation or deactivation but did slightly slow the rate of macroscopic inactivation observed during a 300 ms test depolarisation. In the presence of LY310315 the activation curve was significantly shallower. This resulted in a shift in the activation midpoint voltage to a more depolarised levels. LY310315-induced inhibition of human N-type channels was strongly dependent on the extracellular pH, with increased potency seen upon extracellular acidification. Although most effective against N-type Ca2+ channels, LY310315 was also found to inhibit both P-type and L-type Ca2+ channels. LY310315 proved to be a weak blocker of Na+ currents, but produced ∼50% of the K+ currents of AtT20 cells at a concentration of 0.5 μM.

In a cell line (C2D7) stably expressing the human N-type calcium channel encoded by the subunits α1B-a, β1b, α2bδ, we have analysed the Ca2+ currents produced by a range of action potential-like voltage protocols (APVPs). Such protocols consistently produced robust inward currents that could be eliminated by co-application of the Ca2+ channel blocking ions Cd2+ and La3+. The amplitude, latency to peak and area of the current produced by APVPs was dependent on the precise waveform of voltage protocol employed and the temperature. Short bursts of APVPs applied at 100 Hz produced a depression of the Ca2+ current amplitude which was dependent on the half-width of the APVP employed. In contrast, no frequency-dependent changes in the evoked current kinetics were detected. The amount of current depression seen during an 100 Hz 8 APVP burst was greatly enhanced by increasing the temperature from 22 to 37°C. Alterations to the intracellular Ca2+ buffering capacity suggested that the Ca2+ current depression produced during an APVP train arose, at least in part, from a Ca2+-dependent inactivation of the human N-type Ca2+ channel.

Human N-type Ca2+ channels were rapidly and reversibly inhibited by 5–100 μM BW619C89 (ic50 = 16.4 μM at Vtest = + 10 mV and Vhold = − 90 mV). In the presence of 20 μM BW619C89, activation kinetics were significantly faster. The degree of inhibition observed was affected by both test and holding potential, indicating state-dependent interactions with the N-type Ca2+ channel.

We have characterized the properties of the human N-type Ca2+ channel produced by the stable co-expression of the alpha(1B-1), alpha(2b)delta and beta(1b) subunits. The channel displayed the expected pharmacology with respect to the toxins omega-CTx-GVIA and omega-CTx-MVIIC, which depressed currents in a voltage-independent fashion. We characterized a variety of biophysical properties of the channel under conditions in which either Ca2+, Ba2+ or Sr2+ was the sole extracellular divalent ion. In all three ions, current-voltage relationships revealed that the channel was clearly high-voltage activated. Current activation was significantly slower in Ca2+ than either Sr2+ or Ba2+. Construction of conductance-voltage relationships from tail current measurements indicated that the channel was more high-voltage activated in Ca2+ than in either Sr2+ or Ba2+. The rank order of current amplitude at +4 mV was Ba2+ > Sr2+ > or = Ca2+. Elevation of the extracellular concentration of Ba2+ increased maximal current amplitude and shifted the current-voltage relationship to the right. In all three ions channel inactivation was complex consisting of three distinct exponentials. Recovery from inactivation was slow taking several seconds to reach completion. Steady-state inactivation curves revealed that channel inactivation became detectable at holding potentials of between -101 and -91 mV depending on the permeating species. The rank order of mid-points of steady state inactivation was (most negative) Sr2+ > Ca2+ > Ba2+ (most positive). Deactivation of the N-type Ca2+ channel was voltage-dependent and very fast in all three ions. The deactivation rate in Ba2+ was significantly slower than that in both Ca2+ and Sr2+, however the voltage-dependence of deactivation rate was indistinguishable in all three ions.