Voltage-gated calcium channels in adult rat inferior colliculus neurons

Neuroscience 120 (2003) 815– 826

VOLTAGE-GATED CALCIUM CHANNELS IN ADULT RAT INFERIOR COLLICULUS NEURONS P. N'GOUEMO AND M. MORAD*

other related models (Frye et al., 1986; Browning, 1986; McCown et al., 1984; Faingold and Boersma Anderson, 1991; Chakravarty and Faingold, 1998; Wada et al., 1970). Intrinsic cellular membrane properties and the level of the expression of different ion channels are thought to be critical in normal and abnormal processing of the auditory signal in adult IC neurons. Although there are significant data on the ionic currents of adult IC neurons (Smith, 1992; Li et al., 1998; Sivaramakrishnan and Oliver, 2001), little is known as to the expression of different types of Ca channels or their biophysical, and pharmacological properties. In neonatal IC neurons, we have previously identified the presence of low (LVA) and high (HVA) threshold voltageactivated Ca channels (N⬘Gouemo and Rittenhouse, 2000). These studies, however, were confined to very young animals, when the auditory function and the susceptibility to audiogenic seizures appear not to have been fully developed (Reigel et al., 1989). In this report, using a variety of pharmacological blockers and molecular techniques, we have identified the type and the fraction of currents carried by various Ca channels and determined their biophysical properties in freshly isolated, adult IC neurons.

Department of Pharmacology, Georgetown University Medical Center, 3900 Reservoir Road NW, Washington, DC 20007, USA

Abstract—The inferior colliculus (IC) plays a key role in the processing of auditory information and is thought to be an important site for genesis of wild running seizures that evolve into tonic– clonic seizures. IC neurons are known to have Ca channels but neither their types nor their pharmacological properties have been as yet characterized. Here, we report on biophysical and pharmacological properties of Ca channel currents in acutely dissociated neurons of adult rat IC, using electrophysiological and molecular techniques. Ca channels were activated by depolarizing pulses from a holding potential of ⴚ90 mV in 10 mV increments using 5 mM barium (Ba) as the charge carrier. Both low (T-type, VA) and high (HVA) threshold Ca channel currents that could be blocked by 50 ␮M cadmium, were recorded. Pharmacological dissection of HVA currents showed that nifedipine (10 ␮M, L-type channel blocker), ␻-conotoxin GVIA (1 ␮M, N-type channel blocker), and ␻-agatoxin TK (30 nM, P-type channel blocker) partially suppressed the current by 21%, 29% and 22%, respectively. Since at higher concentration (200 nM) ␻-agatoxin TK also blocks Q-type channels, the data suggest that Q-type Ca channels carry approximately 16% of HVA current. The fraction of current (approximately 12%) resistant to the above blockers, which was blocked by 30 ␮M nickel and inactivated with ␶ of 15–50 ms, was considered as R-type Ca channel current. Consistent with the pharmacological evidences, Western blot analysis using selective Ca channel antibodies showed that IC neurons express Ca channel ␣1A, ␣1B, ␣1C, ␣1D, and ␣1E subunits. We conclude that IC neurons express functionally all members of HVA Ca channels, but only a subset of these neurons appear to have developed functional LVA channels. © 2003 IBRO. Published by Elsevier Science Ltd. All rights reserved.

EXPERIMENTAL PROCEDURES Cell preparation Male Sprague–Dawley rats were anesthetized with pentobarbital (50 mg/kg, i.p.). The brains were perfused with a solution containing (in mM): 110 choline chloride, 2.5 KCl, 1.2 NaH2PO4, 26 NaHCO3, 2.4 sodium/pyruvate, 1.3 L-ascorbic acid, 20 dextrose, 0.5 CaCl2 and 7 MgCl2 (osmolarity 290 –300 mOsm with sucrose), and bubbled with 95% O2 and 5% CO2. The brains were removed and immersed in sucrose solution containing (in mM): 205 sucrose, 5 KCl, 1.2 NaH2PO4, 26 NaHCO3, 2.4 sodium/pyruvate 1.3 L-ascorbic acid, 20 dextrose, 0.2 CaCl2, and 1.3 MgSO4 (osmolarity 290 –300 mOsm with sucrose), and bubbled with 95% O2 and 5% CO2. Coronal brainstem slices (500 ␮M thick) at the level of the IC were sectioned using a Vibratome. The IC external cortex was microdissected and placed in Hibernate-A or L-15 (Life Technologies, Gaithersburg, MD, USA) medium containing papain (20 U/ml; Worthington, Lakewood, NJ, USA) and bubbled with 100% oxygen for 60 min at 30 –32 °C. The enzyme was washed with Hibernate-A or Leibovitz’s L-15 medium containing trypsin inhibitor (1 mg/ml) and albumin bovine serum (1 mg/ml). Neurons were then dissociated by gentle trituration with a fire-polished Pasteur pipette in Neurobasal-A (Life Technologies, Gaithersburg, MD, USA) medium supplemented with 2% B27 (Life Technologies, Gaithersburg, MD, USA), 5% horse serum, 5% fetal bovine serum, and penicillin (100 U/ml)-streptomycin (0.1 mg/ml). Dissociated IC neurons were then plated onto poly-D-lysine (50 ␮g/ml) coated glass coverslips for at least 1 h before whole cell recordings. All experimental procedures were approved by the Animal Care

Key words: calcium channel subunit, nifedipine, ␻-agatoxin TK, ␻-conotoxin GVIA, and window current.

Inferior colliculus (IC) is thought to process auditory information and serve as a gateway to the auditory motor integration in the mammalian nervous system (Caird, 1991). IC neurons also play a critical role in initiation of audiogenic seizures in genetically epilepsy-prone rats and *Corresponding author. Tel: ⫹1-202-687-8453; fax: ⫹1-202-6878458. E-mail address: [email protected] (M. Morad). Abbreviations: EDTA, ethylenediaminetetraacetic acid; EGTA, ethyleneglycol-bis(␤-aminoethylether)-N,N,N',N'-tetraacetic acid; HEPES, (N-[2-hydroxyethyl]piperazine-N'-[2-ethanesulfonic acid]); HVA, high threshold voltage-activated calcium channel; IC, inferior colliculus; LVA, low threshold voltage-activated calcium channel; TBST, 20 mM Tris–HCl, pH 7.4, 140 mM NaCl and 0.1% v/v Tween 20; TTX, tetrodotoxin.

0306-4522/03$30.00⫹0.00 © 2003 IBRO. Published by Elsevier Science Ltd. All rights reserved. doi:10.1016/S0306-4522(03)00323-3

815

816

P. N'Gouemo and M. Morad / Neuroscience 120 (2003) 815– 826

Committee of the institution. All efforts were made to minimize the number of animals used and their suffering.

Electrophysiology The currents through Ca channels were recorded at room temperature using the whole cell configuration of the patch clamp techniques (Hamill et al., 1981). The patch electrodes were made from borosilicate glass capillaries and had resistances which ranged from 2 to 4 ⍀M when filled with a solution containing (in mM): 90 cesium methanesulfonate, 10 EGTA, 14 phosphocreatine, 10 HEPES, 10 glucose, 4 disodium adenosine 5⬘-triphosphate, 5 MgCl2 and 0.4 sodium guanosine 5⬘-triphosphate (pH 7.3 with CsOH). Whole cell configuration was established in tyrode solution containing (in mM): 145 NaCl, 5.4 KCl, 5 CaCl2, 10 HEPES (pH 7.4 with NaOH), and the extracellular recording solution for isolating Ba currents contained (in mM): 5 BaCl2, 130 NaCl, 10 HEPES, 1 MgCl2, 20 CsCl2 10 glucose, and 0.001 tetrodotoxin (TTX; pH 7.4 with NaOH). Although Ca is the biological charge carrier through Ca channels, Ba was used as a more reliable method for characterization of Ca currents because Ba permeates as well as Ca, but does not inactivate the channel as it permeates through and in addition suppresses the residual K currents. Voltage clamp experiments were performed with a Dagan 8900 patch clamp amplifier. Currents were filtered at 10 kHz and normalized relative to the membrane capacitance as an estimate of current density. Whole cell capacitance and series resistance were compensated. Leak and residual capacitance currents were subtracted off-line. Data were collected and analyzed off line using the PClamp program (Axon Instruments, Inc., Union City). Currents were measured by using depolarization pulses of 50 ms from a holding potential of ⫺80 mV (depolarization range: ⫺90 mV to ⫹60 mV by 10 mV increments) using Ba as a charge carrier for inward currents through Ca channels. A 0.5 s conditioning pulse to ⫺60 mV was applied to inactivate the LVA Ca channels and Ba currents were evoked by applying voltage steps to 0 mV, at 10 s interval until a steady block was obtained. The amplitudes of Ba currents were measured at 10 ms into the test pulse corresponding to peak current. For pharmacological tests, rapid solution exchanges were performed to deliver Ca channel blockers (Cleemann and Morad, 1991). The effects of Ca channel blockers were tested after obtaining 2–5 min of steady control recordings. Ca channel blockers ␻-agatoxin TK and ␻-conotoxin GIVA were purchased from Alomone laboratories (Jerusalem, Israel), while nifedipine was obtained from Calbiochem (San Diego, CA, USA). The solutions of Ca channel blockers were prepared as concentrated stock solutions in distilled water and stored (for less than 4 weeks) in 1 ml aliquots at ⫺20 °C. Stock solutions were diluted ⬎1:1000 with the extracellular recording solution before each experiment. Cytochrome C (0.01%) was added to the ␻-agatoxin TK solutions to saturate any nonspecific binding sites located on the walls of the tubing and chamber, which, when applied alone had no effect on Ba currents. Nifedipine was solubilized in 100% ethanol as concentrated stock solutions and protected from ambient light at all times and stored in 1 ml aliquots at ⫺20 °C. Application of ethanol at the same final concentration has no effect on Ba currents. Gel electrophoresis and immunoblotting. Rats were deeply anesthetized with pentobarbital (100 mg/kg; i.p.) and the IC was quickly dissected and the area corresponding to the periaqueductal gray was discarded. The IC tissues were stored at ⫺70 °C until processed. Tissue homogenates from each rat were prepared in TE buffer (10 mM Tris–HCl, pH 7.4; 1 mM EDTA). Protein concentrations in the supernatant were determined using bicinchoninic acid Assay Reagent (Pierce, Rockford, IL, USA) and each sample was diluted to a protein concentration of 5 or 10 mg/ml in TE buffer. Homogenate proteins were separated using sodium dodecyl sulfate-polyacrylamide gel (containing 7.5% polyacryl-

amide; Invitrogen, Carlsbad, CA, USA) electrophoresis and transferred to polyscreen polyvinylidene fluoride membrane (New England Nuclear, Boston, MA, USA) in blotting buffer (25 mM Tris– HCl, 192 mM glycine, 20% methanol). The blots were processed using methods described by Towbin et al. (1979). Briefly, the blots were incubated in TBST buffer (20 mM Tris–HCl, pH 7.4; 140 mM NaCl and 0.1% v/v Tween 20) containing 5% nonfat dry milk (BLOTTO buffer) with Ca channel subunit antibodies ␣1A, ␣1B, ␣1C, ␣1D, ␣1E and ␤3 (Alomone laboratories, Jerusalem, Israel) at concentrations of 1.5, 1.5, 0.5, 1.5, 0.75 and 1.0 ␮g/ml, respectively overnight at 4 °C. The following day, the blots were washed (30 min) with a TBST buffer and incubated with a donkey antirabbit antibody (Amersham, Arlington Heights, IL, USA) at a dilution of 1:2000 in BLOTTO buffer for 30 min at room temperature. After several washes of the membrane, individual bands were visualized on film using the enhanced chemiluminescence Super Signal West Pico reagents (Pierce, Rockford, IL, USA) and analyzed.

Data analysis To evaluate the effects of Ca channel blockers on whole cell currents, the percentage of Ba current (I) inhibition was expressed as 100 [1⫺(Idrug/Icontrol)] and used for further analysis. The data were fitted with the Boltzmann equation: I/Imax⫽[I1⫺I2/1⫹e⫹I2] where I/Imax is normalized Ba currents, I1 and I2 are the minimum and maximum of current, respectively, V is voltage, V1/2 is the voltage at half maximal for current, and k is the slope of activation in mV per e-fold charge in current. Data from neurons were grouped and compared using one-way analysis of variance or ␹ on incidences. Differences were considered significant if P⬍0.05. The illustrated current traces represent the average of three to five consecutive traces. All data are presented as mean⫾S.E.M.

RESULTS Biophysical properties of Ca channel currents Acutely dissociated large IC neurons (cell capacitances of 77.3⫾2.8; n⫽108) from 6 – 8-week-old rats (n⫽72) were used in the electrophysiological studies. Depolarizing test pulses from the holding potential of ⫺90 mV in 10 mV increments were applied to determine the voltage dependence and kinetics of Ba currents (n⫽51, 44 rats). We found that the expression pattern of Ba current was heterogeneous among IC neurons, as they expressed the LVA and HVA currents. LVA currents consisted of a small, slowly activating, current at voltages negative to ⫺40 mV when using ⫺90 mV holding potential (Fig. 1A, left). At voltages positive to ⫺50 mV, the current exhibited both slow activation and inactivation kinetics (Fig. 1A, left). The expression of LVA current was quite variable among IC neurons such that only 14 (13 rats) of 51 IC neurons appeared to express this current (Fig. 1A). In sharp contrast, HVA currents were recorded in all IC neurons tested but could be grouped in those where the current had both a transient and a sustained component and those without a transient component. The HVA current with both transient and sustained components, with maximal current at about ⫺20 mV, was found only in a subset of IC neurons (n⫽5 from four rats; Fig. 1B). The HVA current without a transient component activated around ⫺40 mV, was strongly activated between ⫺20 mV and ⫺10 mV, and reached its a maximum at 0 mV (Fig. 1C).

P. N'Gouemo and M. Morad / Neuroscience 120 (2003) 815– 826

817

Fig. 1. LVA and HVA whole cell Ba2⫹ currents recorded in IC neurons. (A) Left, LVA currents showing slow activation and inactivation. Right, Averaged current-voltage (I–V) exhibiting LVA currents (n⫽14 neurons, 13 rats) activated around ⫺60 mV. (B) Left, HVA currents with a rapid transient (T) and sustained (S) component. Right, Averaged I–V relations (n⫽5 neurons, 4 rats) exhibiting both transient and sustained components peaked at ⫺10 mV. (C) Left, HVA current without a transient component. Right, Averaged I–V relation (n⫽32 neurons, 27 rats) exhibiting HVA current without a transient component that activates around ⫺40 mV.

818

P. N'Gouemo and M. Morad / Neuroscience 120 (2003) 815– 826

This HVA current was observed in 32 (27 rats) of 51 IC neurons and was subjected to pharmacological analysis. To examine whether IC neurons exhibit time and voltage dependent inactivation, Ba currents were measured from ⫺90 mV and ⫺50 mV holding potentials using 50 ms depolarizing pulses. Fig. 2 shows that changing the holding potential from ⫺90 to ⫺50 mV for 3– 6 min caused a 44⫾3% (n⫽4 rats; range: 36 –51%) reduction in the peak current density (activated at 0 mV) suggesting a slow voltage-dependent inactivation of Ca channels. Full recovery from this type of inactivation generally required 2–3 min on returning the holding potential to ⫺90 mV (Fig. 2A). Current traces activated from the holding potential of ⫺50 mV exhibited somewhat slower inactivation kinetics compared with ⫺90 mV (Fig. 2B), but tail currents evoked at ⫺90 mV exhibited a slow component of deactivation compared with ⫺50 mV. The slow component of tail current was also observed in neonatal IC neurons and appeared to be selectively sensitive to L-type channel blocker (N⬘Gouemo and Rittenhouse, 2000). No shifts in the voltage dependence of current were observed when changing the holding potential from ⫺90 mV to ⫺50 mV (Fig. 2C). The voltage dependence of difference current obtained by subtracting the current recorded from the holding potential of ⫺50 mV from ⫺90 mV also peaked at 0 mV (Fig. 2C). Fig. 3 quantifies the steady-state, voltage dependent of availability of the current. Ba currents were measured using a 0.5 s conditioning pulses ranging between ⫺100 mV and ⫹30 mV followed by test pulse to 0 mV from the holding potential of ⫺80 mV. The normalized peak current density was plotted versus conditioning potential. The data were fit with the Boltzmann equation yielding a half-inactivation potential of ⫺25.4⫾1.3 mV (n⫽5 rats; range: 23–30 mV) and a slope factor k of 9.9⫾1.2 (n⫽5; range: 8 –14). The currents evoked by the test pulse to 0 mV progressively decreased, but consistently showed a noninactivating component of the current (Fig. 3A, B). On average, the noninactivating component represented 40ⴞ8% (n⫽5; range: 10 –57%) of the Ba currents. Conditioning pulses as long as 5–20 s were also used in an attempt to inactivate Ba currents completely, but we found little or no additional inactivation in response to the longer pulses (data not shown). Ten second depolarizing pulses were used to probe the kinetics of inactivation of Ba currents. On average 20% of currents failed to fully inactivate even at the end of the 10 s pulses, consistent with the existence of multiple types of HVA Ca channels (Fig. 3C). The time course of inactivation could be approximated by three time constants (n⫽3 rats): very slow ␶⫽10⫾0.5 s (range: 10 –11.5 s); slow ␶⫽1⫾0.1 s (range: 0.8 –1.3 s); and fast inactivation ␶⫽0.2⫾0.03 s (range: 0.2– 0.3 s). The voltage dependence of activation parameters was determined by normalizing Ba currents measured at different potentials to the maximal current. The data were fitted with a Boltzmann function yielding a half activation voltage of ⫺17.0⫾2.2 mV (n⫽5 rats; range: 12–24 mV) and a slope factor of 4.3⫾0.2 (n⫽5; range: 4 –5; Fig. 3B). The steady-state inactivation and activation curves overlapped

Fig. 2. Voltage-dependent rapid “rundown” of HVA current. (A) Time course decrease and recovery of Ba2⫹ currents activated at 0 mV on changing the holding potential from ⫺90 mV to ⫺50 mV and back to ⫺90 mV. Ba2⫹ currents decrease by approximately 50% and fully recovered after about 3 min when holding potential was returned to ⫺90 mV. (B) Representative current traces from the same neuron as in A show less inactivation at holding potential of ⫺50 mV (a) compared with ⫺90 mV (b). (C) Current-voltage (I–V) relation of Ba2⫹ currents recorded from holding potential of ⫺90 mV (filled squares) and -50 mV (filled circles) showed a partial voltage-dependent inactivation of high threshold voltage-activated channel current. The difference current (Ba2⫹ currents at ⫺90 mV minus Ba2⫹ currents at ⫺50 mV; filled triangles) activates at ⫺40 mV. Each point represents mean⫾S.E.M (n⫽4 rats).

P. N'Gouemo and M. Morad / Neuroscience 120 (2003) 815– 826

Fig. 3. Activation and inactivation parameters of HVA Ca2⫹ channel currents. (A) Superimposed tracing of Ba2⫹ currents activated at 0 mV following 0.5 s conditioning depolarizing pulses from ⫺100 to ⫹30 mV (see schematic in A). Ba2⫹ currents activated at 0 mV decreased progressively from ⫺100 mV to 0 mV, but about 40% of the current continued to be activated even at ⫹30 mV. (B) Voltage dependence of activation (filled circles) and inactivation (filled squares) normalized for five IC neurons. Ba2⫹ currents did not completely inactivate using 500 ms conditioning pulse. Inactivation and activation curves overlap around ⫺10 mV, indicating a persistent or “window” current. (C) A Ba2⫹ current elicited by a 10 s voltage step to 0 mV from a holding potential of ⫺80 mV, showing significant noninactivating current component (n⫽3 rats). About 20% of the current remained at the end of the pulse.

between ⫺20 mV and 0 mV (Fig. 3B), indicating that IC neurons exhibited persistent “window current” (Johnston and Wu, 1995). Pharmacological profiles of Ca channel currents To examine the HVA currents, a 0.5 s conditioning pulse was applied to inactivate LVA Ca channels prior to activa-

819

tion of HVA current at 0 mV (Fig. 4). HVA currents were blocked effectively by 50 ␮M cadmium (data not shown), indicating that these currents were transported through voltage-gated Ca channels. To determine the possible expression of different Ca channel types in IC neurons, we used relatively high concentrations of a number of specific HVA Ca channel blockers. Nifedipine (10 ␮M), a potent L-type channel blocker, reduced the total HVA current on average by 24⫾2% (range: 13– 41%; n⫽16 from 12 rats; Fig. 4A–C). The inhibitory effect of nifedipine was reversible (data not shown). The nifedipine-sensitive current obtained by subtracting the current recorded in nifedipine from control exhibited a slower activation phase with little or no inactivation (Fig. 4A). The voltage-dependence of nifedipine-sensitive current was similar to that of control, showing activation around ⫺40 mV with a maximum at 0 mV (Fig. 4C). Fig. 4 also illustrates the effects of 1 ␮M ␻-conotoxin GVIA, a potent blocker of N-type channel, on total HVA current. A 33⫾2% (range: 21– 48%; n⫽16 from 11 rats) reduction in the mean HVA current was measured on application of 1 ␮M ␻-conotoxin GVIA (Fig. 4D–F). In six of 14 IC neurons tested, the block following application of ␻-conotoxin GVIA was partially or fully reversible (data not shown). The ␻-conotoxin GVIA-sensitive current, obtained as the difference current recorded in the presence and absence of ␻-conotoxin GVIA, had a noninactivating kinetics (Fig. 4D). Fig. 5 compares the effects of two concentrations of ␻-agatoxin TK on the total HVA currents. At 30 nM, a concentration of ␻-agatoxin TK reported to block selectively the P-type Ca channels (Teramoto et al., 1995), the current was suppressed by 25⫾2% (range: 14 –32%; n⫽11 from seven rats). At 200 nM ␻-agatoxin TK, a concentration reported to block also the Q-type current (Teramoto et al., 1995), the current was reduced by 43⫾3% (range: 22– 57%; n⫽13 from seven rats; Fig. 5A, B). Since there are no selective blockers of the Q-type Ca channel, this current was isolated by subtracting the 30 nM ␻-agatoxin TKsensitive current from that of 200 nM ␻-agatoxin TK sensitive current. This approach suggests that 18% of total HVA current was carried by Q-type channel. The 30 nM ␻-agatoxin TK-sensitive current was noninactivating (Fig. 5B), consistent with the presence of P-type current, while the 200 nM ␻-agatoxin TK-sensitive current exhibited a rapidly inactivating component (Fig. 5B) associated with Q-type current in three of six neurons tested (Mermelstein et al., 1999; Randall and Tsien 1995; Teramoto et al., 1995). The inhibition by ␻-agatoxin TK was partially or completely reversible (data not shown). The voltage-dependence of the 30 nM and 200 nM ␻-agatoxin TK-sensitive current activated around ⫺10 and ⫺20 mV, respectively (Fig. 5C,D). The I–V relation shows that 30 nM ␻-agatoxin TK-sensitive current peaked at ⫹10 mV (Fig. 5C). Quite similar to the nifedipine- and ␻-conotoxin GVIAsensitive currents, the voltage dependence of 200 nM ␻-agatoxin TK-sensitive current (Q-type) had a maximum at 0 mV (Fig. 5D), suggesting that these currents may have similar voltage dependence.

820

P. N'Gouemo and M. Morad / Neuroscience 120 (2003) 815– 826

Fig. 4. Effects of 10 ␮M nifedipine and 1 ␮M ␻-conotoxin GVIA on HVA currents in IC neurons. Ba2⫹ currents were activated by a voltage step to 0 mV, using a 0.5 s conditioning pulse at ⫺60 mV from a holding potential of ⫺80 mV to inactivate LVA currents. (A) Representative Ba2⫹ current in absence (a) and presence of 10 ␮M nifedipine (b). Nifedipine (10 ␮M) reduced the peak current density typically by 25%. The nifedipine-sensitive current (difference current in the presence and absence of nifedipine) showed slow activation kinetics. (B) Time course of the suppressive effect of 10 ␮M nifedipine on HVA currents. (C) Examples of current-voltage (I–V) relation in the absence and presence of nifedipine. The I–V relation of the nifedipine-sensitive current typically activated around ⫺40 and peaked at 0 mV. (D) Current traces in the absence (a) and presence (b) of 1 ␮M ␻-conotoxin GVIA. The ␻-conotoxin GVIA-sensitive current obtained as the difference current showed little inactivation. (E) Time course of the effect of 1 ␮M ␻-conotoxin GVIA on HVA currents, showing a 35% reduction in the current density. (F) Examples of I–V relation in the absence and presence of 1 ␮M ␻-conotoxin GVIA shows that ␻-conotoxin GVIA-sensitive current, obtained as the difference current, activated around ⫺30 mV.

P. N'Gouemo and M. Morad / Neuroscience 120 (2003) 815– 826

821

Fig. 5. Effects of ␻-agatoxin TK on HVA currents. Ba2⫹ currents were activated at 0 mV (schematic B; Figure legend 4). (A) Time course of effect of sequential application of 30 and 200 nM ␻-agatoxin TK on HVA currents. Application of 30 nM ␻-agatoxin TK reduced Ba2⫹ currents by 24%, while 200 nM ␻-agatoxin TK suppressed an additional fraction of the current. (B) Ba2⫹ current in the absence (a) and presence of 30 nM (b) and 200 nM (c) ␻-agatoxin TK. The 30 nM ␻-agatoxin TK-sensitive current (obtained as described in Fig. 4A, lower traces) exhibited little inactivation, while the 200 nM ␻-agatoxin TK-sensitive current (Fig. 5B, lower traces) showed a fast inactivating component. (C) Examples of current-voltage (I–V) relation in the absence and presence of 30 nM ␻-agatoxin TK shows that the blocker reduces the current at voltages positive to ⫺20 mV. The I–V relation of the 30 nM ␻-agatoxin TK-sensitive current, activates at ⫺20 mV. (D) Examples of I–V relation in the absence and presence of 200 nM ␻-agatoxin TK shows that the higher concentration of this blocker reduces the current at voltages positive to ⫺30 mV. The difference 200 nM ␻-agatoxin TK-sensitive current, shows a symmetrical suppression of Ba2⫹ currents positive to ⫺30 mV.

822

P. N'Gouemo and M. Morad / Neuroscience 120 (2003) 815– 826

Fig. 6. Sequential block of HVA current by sub-type specific Ca2⫹ channel blockers. (A) Sequential application of 1 ␮M ␻-conotoxin GVIA (b), 10 ␮M nifedipine (c), and 200 nM ␻-agatoxin TK (d) shows that the three blockers reduce but do not completely suppress HVA currents activated at 0 mV (schematic in B). (B) Representative current traces of sequential application of blockers as indicated in A. Ba2⫹ current resistant to the blockers (lower trace, B) shows inactivation kinetics with ␶ of 50 ms. (C) Current-voltage (I–V) relation in the absence and presence of 200 nM ␻-agatoxin TK, 1 ␮M ␻-conotoxin GVIA and 10 ␮M nifedipine shows that R-type current activates around ⫺50 mV.

Fig. 6 shows that IC neurons also express an HVA type current that is not suppressed following simultaneous application of 10 ␮M nifedipine, 1 ␮M ␻-conotoxin GVIA, and 200 nM agatoxin TK. Such drug-insensitive current (Rtype) was seen in only three of 11 IC neurons (n⫽8 rats) and had fairly rapid inactivation kinetics (␶ approximately 14 –50 ms). The voltage-dependence of R-type current showed activation at ⫺50 mV with a maximum at ⫹10 mV. R-type current averaged 12⫾1% (range: 10 –17%; n⫽3 rats) of the HVA Ba currents and could be blocked by 30 ␮M nickel (data not shown). In IC neurons where an R-type current could not be identified using the above procedures, it was not clear whether such cells did not express R-type channel or whether the current was more sensitive to the cocktail of Ca channel blockers. We also examined the specificity of the above blockers on total HVA current by comparing the magnitude of the inhibition when a given blocker was applied alone or in combination with another blocker. For instance, 1 ␮M ␻-conotoxin GVIA reduced the total HVA current density by 33% when applied alone, compared with 29⫾3% (range: 26 –31%; n⫽3 rats) when the blocker was applied following 10 ␮M nifedipine. Similarly, when cells were first exposed to 1 ␮M ␻-conotoxin GVIA, application of 10 ␮M nifedipine reduced the current density by 21⫾4% (range: 11–32%; n⫽7 rats; range: 11–32%) compared with 24% when the blocker was applied alone. Simultaneous application of nifedipine (10 ␮M) and ␻-conotoxin GVIA (1 ␮M) reduced the peak current density by 56⫾3% (range: 54 –56%; n⫽5 rats) suggesting that the effects of these blockers were additive. When the cells were first incubated with both nifedipine (10 ␮M) and ␻-conotoxin GVIA (1 ␮M), application of 200 nM ␻-agatoxin TK reduced the current by 41⫾1% (range: 32–51%; n⫽4 rats) compared with 43% when the blocker was applied alone. Similarly, when the cells were fist incubated with 200 nM ␻-agatoxin TK, application of both nifedipine (10 ␮M) and ␻-conotoxin GVIA (1 ␮M) suppressed the current by 55⫾1% (range: 54 – 56%; n⫽3 rats) compared with 56% when given alone. Thus, it appears that the efficacy of the blockers whether added alone or in tandem is quite similar. In IC neurons the sum contribution of all Ca channel currents, blocked by the drugs, was often over 100%, which may reflect the redundancy of blocker specificity previously reported (McDonough et al., 2002). Because no significant difference was found when a channel blocker was applied alone or following another, or in combination with other channel blockers, normalization (to 100%) allowed estimation of the distribution of L-, N-, P-, Q- and R-type current distribution with the following order: 21%, 29%, 22%, 16% and 12%, respectively. Thus, it appears that P/Q-type channels account for 38% of total HVA current. Molecular profiles of the ␣1 subunits of Cachannel Since the single cell reverse transcription polymerase chain reaction methodology that allows correlation of the

P. N'Gouemo and M. Morad / Neuroscience 120 (2003) 815– 826

823

Fig. 7. Protein expression of Ca2⫹ channel subunits in IC neurons. IC tissues were solubilized and proteins separated by sodium dodecylsulfate polyacrylamide gel electrophoresis. The resulting immunoblots using 10 ␮g of protein per lane were probed with Ca2⫹ channel antibodies. Protein markers (in kDa) are shown at the right of the figure. IC neurons express Ca2⫹ channel ␣1A, ␣1B, ␣1C, ␣1D and ␣1E subunit protein, the pore-forming subunits of P/Q-, N-, L- and R-type Ca2⫹ channel, respectively, as well as the regulatory ␤3 subunit protein.

presence of specific Ca channels with their electrophysiological characteristics is not as yet well developed in IC neurons, we examined the protein expression of the HVA family of Ca channels using rabbit polyclonal antibodies to ␣1A, ␣1B, ␣1C, ␣1D, and ␣1E pore-forming subunits of Ca channels, as well as the structural/regulatory ␤3 subunit (Fig. 7). The antibody to ␣1A subunit reveals an immunoreactive band with a molecular mass of protein of about 190 kDa, suggesting the presence of P/Q-type channels (Westenbroek et al., 1995). The ␣1B subunit was observed at about 240 kDa indicating the presence of N-type channels (Westenbroek et al., 1992). The antibody to ␣1C and ␣1D subunit reveals immunoreactive bands at about 207 kDa suggesting the presence of L-type Ca channels (Hell et al., 1993). The ␣1E subunit was seen at about 255 kDa indicating the presence of R-type channel while the structural/regulatory ␤3 subunit was observed at about 60 kDa (Scott et al., 1996; Yokoyama et al., 1995). Thus, it appears that IC neurons express all subtype of HVA Ca channel protein.

DISCUSSION The main findings of this study are that adult IC neurons express HVA Ca channel currents that are composed primarily of five pharmacologically distinct types that include: L-type nifedipine-sensitive, N-type ␻-conotoxin GVIA-sensitive, P/Q-types ␻-agatoxin TK-sensitive, and a residual current not sensitive to the combination of these blockers. These electrophysiological findings are supported by the expression of Ca channel ␣1A, ␣1B, ␣1C, ␣1D, and ␣1E subunit proteins in adult IC neurons. Since we found significant expression of both ␣1C and ␣1D subunit proteins, it

is likely that both genes encoded for the L-type Ca channel in IC neurons. High threshold voltage-activated Ca channel types Ca channel currents were recorded from acutely dissociated IC neurons, but the cell dissociation procedures used in this study did not allow a precise identification of the phenotype of neurons tested. The type of neurons used may be critical as IC is anatomically subdivided into the central nucleus, external cortex, and dorsal cortex (FayeLund and Osen, 1985), each with distinct functions in normal and pathological processing of auditory information (Caird, 1991). Our findings suggest that all adult neurons from the IC external cortex express L-, N-, and P-type currents, but only a subset of neurons express Q- and R-type currents. The relatively low incidence of Q-and R-type currents in adult IC neurons may be related to their preferential expression on the dendritic processes, but not on the soma where whole cell currents are recorded. Pharmacological quantification in neonatal IC neurons has shown that L-, N-, P/Q- and R-type currents contribute on average 25, 28, 27 and 38% to the total Ca current, respectively (N⬘Gouemo and Rittenhouse, 2000). Although it is tempting to consider that the contribution of each HVA Ca channel subtype changes with age, such assumptions are problematic because the cell dissociation procedures used in neonatal studies do not allow identification of cellular location within the IC. Nevertheless, if we assume that the majority of neonatal IC neurons studied were located in the external nucleus of the IC, the contribution of N- and L-type current to the total HVA current appears not to have been significantly changed with age. By the same token,

824

P. N'Gouemo and M. Morad / Neuroscience 120 (2003) 815– 826

the fractional increase in expression of P/Q-type currents in adult (approximately 43%) compared with neonatal (approximately 27%) IC neurons may reflect age-related upregulation. Whether up-regulation of P/Q-type current is due to an increase in the expression P- and/or Q-type proteins remains to be determined, since these currents were not separated and quantified in neonatal neurons. Interestingly, the relative contribution of R-type current, within individual IC neuron to the total HVA current appears to decrease from approximately 38% to approximately 12% from neonatal to adult neurons, suggestive of a significant down-regulation of this current with age. The up-regulation of P/Q- type current and down-regulation of R-type current may reflect developmental changes of Ca channels as reported for other parts of the brain (Gray et al., 1992; Rossi et al., 1994). A subset of adult IC neurons also expressed HVA Ca channel currents with a transient component that peaked around ⫺20 mV. Such transient HVA currents were observed in the presence of TTX ruling out contaminating TTX-sensitive sodium (Na) currents. Although it could be argued that these transient currents are in fact TTX-resistant Na currents, similar transient currents were also recorded by us in neonatal IC neurons where external Na was completely replaced by tetraethylammonium chloride in TTX containing solutions (N⬘Gouemo and Rittenhouse, 2000). It should be noted that such currents were sensitive to ␻-agatoxin IVA but were not affected by ␻-conotoxin GIVA and nifedipine, indicating that these currents were indeed Ca channel currents rather than TTX-resistant Na currents (N⬘Gouemo and Rittenhouse, 2000). The expression of such transient HVA currents appears to be significantly reduced in adult IC neurons (approximately 9%) compared with neonates (approximately 41%; N⬘Gouemo and Rittenhouse, 2000), consistent with possible developmental role of this channel. Low threshold voltage-activated Ca channels Ca channel currents in a subset of IC neurons activated at more negative voltages and were as such identified as LVA T-type Ca channels (Huguenard, 1996). The low expression of LVA currents in IC neurons is consistent with other reports that show only low to moderate levels of expression of ␣1G, ␣1H, and ␣1I transcripts encoding for LVA T-type Ca channels in the IC (Talley et al., 1999). The expression of LVA currents appears slightly reduced in adult IC neurons (approximately 27%, present study) compared with neonates (approximately 33%; N⬘Gouemo and Rittenhouse, 2000). The lack of significant change in the expression of LVA currents suggests that the contribution of these channels to the generation of action potentials and neuronal firing is unchanged with age. Quantification of Ca channel currents in adult neurons of the external cortex of the IC may be somewhat flawed as cell isolation procedures result in the loss of dendritic processes, as compared with the intact in situ neurons. In addition, IC external cortex contains diverse cell types that may be at different developmental stage that were not

morphologically or physiologically identified (Faye-Lund and Osen, 1985; Li et al., 1998). Whether there are distinct patterns of differential expression of Ca channel types in different adult neuron of IC external nucleus remains to be critically evaluated. In evaluating the functional significance of Ca channels and their various subtypes in IC neurons, it has to be remembered that although the use of Ba as the charge carrier allows clear and reliable identification of various channel subtypes, it minimizes the role of kinetics of channels current when Ca is the charge carrier. Ca-dependent inactivation is an important negative feedback mechanism that limits the influx of Ca into neurons. Ca-sensing and calmodulin binding domains on C-terminus of ␣1A and ␣1C subunits of Ca channel have been recently shown to be critical in the Ca mediated inactivation of the channel (Lee et al., 1999, 2000; Peterson et al., 1999; Soldatov et al., 1998; Zu¨hlke and Reuter, 1998; Zu¨hlke et al., 1999). Even though Ba currents did not completely inactivate using a 0.5–20 s inactivating conditioning pulses, it is not clear whether such noninactivating behavior would also be present if Ca were the charge carrier in these neurons. Although P-type channel could be responsible for such current (Mintz et al., 1992), the major fraction of current in the IC neurons appears to be carried by the inactivating L-, N- and P/Q-type channels which have significant Ca mediated inactivation (Meuth et al., 2001; Lee et al., 1999, 2000; Zeilhofer et al., 2000). A critical evaluation of the kinetics of inactivation of HVA currents using Ca as a charge carrier is required to address this issue. Functional significance Ca influx can influence IC neuronal excitability in at least two ways: by causing depolarization through direct influx of Ca and/or by causing hyperpolarization through activation of Ca-dependent K channels (thereby increasing interspike interval). Ca-dependent slow depolarizations and spikes have been already reported in adult IC neurons (Li et al., 1998; Smith, 1992). Adult IC neurons also exhibited Cadependent spike frequency adaptation and afterhyperpolarization indicating the presence of Ca-activated K channels (Li et al., 1999; Smith, 1992). Consistent with these observations, Ca-activated K currents and significant level of expression of Ca-activated K channel protein was found in IC neurons (N'Gouemo et al., 2001). Further, since we find significant levels of expression of Q-type Ca currents, these channels could be involved in the regulation of afterhyperpolarization and spike frequency adaptation, as reported for the pyramidal cells (Pineda et al., 1998). Interestingly, we also find a significant up-regulation of P/Qtype and down-regulation of R-type currents in adult IC neurons, suggesting that these channels play an important role in the maturation of acousticomotor reflex pathway. Whether there is a direct link between the up-regulation of P/Q-type and down-regulation of R-type or the transient HVA currents remains to be determined. Voltage-gated Ca channels are known to contribute to the release of various neurotransmitters including GABA, which play an important role in the modulation of IC neu-

P. N'Gouemo and M. Morad / Neuroscience 120 (2003) 815– 826

ronal responses (Faingold et al., 1991; Li et al., 1999; Smith, 1992). IC neurons are known to be immunoreactive for GABA antibodies and many terminals in the IC have flattened or pleomorphic vesicles, suggestive of inhibitory synapses as seen in other parts of the brain (Caspary et al., 1990; Moore and Moore, 1987; Roberts et al., 1985). Because N and P/Q-type Ca channels contribute to approximately 68% of total HVA current, it is likely that these channels may initiate GABA release in adult IC neurons as reported in various types of neurons in the brain (Horne and Kemp, 1991; Ohno-Shosaku et al., 1994; Rhee et al., 1999; Sitges and Chiu, 1995; Timmermann et al., 2002). Adult IC neurons are known to exhibit spontaneous action potential firing at resting membrane potential and exhibit epileptiform responses evoked by synaptic stimulation, in vitro (Li et al., 1998; Smith, 1992). The observed presence of window Ca current (Fig. 3) and significant expression of LVA and transient HVA Ca currents may contribute to such spontaneous firing and/or epileptogenesis. HVA Ca currents may also participate in the control of IC neuronal excitability since pharmacological blockade of these channels suppressed the inherited or induced susceptibility to audiogenic seizures (Jackson and Scheideler, 1996; Little et al., 1986). Acknowledgements—This publication was made possible by Public Health Service Grant HL62525 (M.M.) and AA05497 (P.N.) from the National Institutes of Health and its contents are the responsibility of the authors and do not necessarily represent the official views of the NHLBI and NIAAA. We are grateful to Dr. Robert P. Yasuda for his critical input in the Western blot studies and to Dr. Robert C. Foehring for his critique of the earliest version of the manuscript.

REFERENCES Browning RA (1986) Neuroanatomical localization of structures responsible for seizures in the GEPR: lesion studies. Life Sci 39:857–867. Caird D (1991) Neurobiology of hearing: the central auditory system. New York: Raven Press. Caspary DM, Raza A, Lawhorn Armour BA, Pippin J, Arneric SP (1990) Immunocytochemical and neurochemical evidence for agerelated loss of GABA in the inferior colliculus: implications for neural presbycusis. J Neurosci 10:2363–2372. Chakravarty DN, Faingold CL (1998) Comparison of neuronal response patterns in the external and central nuclei of inferior colliculus during ethanol administration and ethanol withdrawal. Brain Res 783:102–108. Cleemann L, Morad M (1991) Role of Ca channel in cardiac excitationcontraction coupling in the rat: evidence from Ca transients and contraction. J Physiol 432:283–312. Faingold CL, Boersma Anderson CA (1991) Loss of intensity-induced inhibition in inferior colliculus neurons leads to audiogenic seizure susceptibility in behaving genetically epilepsy-prone rats. Exp Neurol 113:354 –363. Faingold CL, Boersma Anderson CA, Caspary DM (1991) Involvement of GABA in acoustically-evoked inhibition in inferior colliculus neurons. Hear Res 52:201–216. Faye-Lund H, Osen KK (1985) Anatomy of the inferior colliculus in rat. Anat Embryol 171:1–20. Frye GD, McCown TJ, Breese GR, Peterson SL (1986) GABAergic modulation of inferior colliculus excitability: role in the ethanol withdrawal seizures. J Pharmacol Exp Ther 237:478 –485. Gray DB, Bruses JL, Pilar GR (1992) Developmental switch in the

825

pharmacological of Ca channel coupled to acetylcholine release. Neuron 8:715–724. Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ (1981) Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflu¨ger Arch 391: 85–100. Hell JW, Westenbroek RE, Warner C, Ahlijanian MK, Prystay W, Gilbert MM, Snutch TP, Catterall WA (1993) Identification and differential subcellular localization of the neuronal class C and class D L-type calcium channel ␣1 subunits. J Cell Biol 123:949 –962. Horne AL, Kemp JA (1991) The effect of omega-conotoxin GVIA on synaptic transmission within the nucleus accumbens and hippocampus of the rat in vivo. Br J Pharmacol 103:1733–1739. Huguenard JR (1996) Low-threshold calcium currents in central nervous system neurons. Annu Rev Physiol 58:329 –348. Jackson HC, Scheideler MA (1996) Behavioral and anticonvulsant effects of Ca channel toxins in DBA/2 mice. Psychopharmacologia 126:85–90. Johnston D, Wu SMS (1995) Foundation of cellular neurophysiology. MIT Press: Cambridge. Lee A, Scheuer T, Catterall WA (2000) Ca/calmodulin-dependent facilitation and inactivation of P/Q-type Ca channels. J Neurosci 20:6830 –6838. Lee A, Wong ST, Gallagher D, Li B, Storm DR, Scheuer T, Catterall WA (1999) Ca/calmodulin binds to and modulates P/Q-type calcium channels. Nature 339:155–159. Li Y, Evans MS, Faingold CL (1998) In vitro electrophysiology of neurons in subnuclei of rat inferior colliculus. Hear Res 121:1–10. Li Y, Evans MS, Faingold CL (1999) Synaptic response patterns of the cortex of rat inferior colliculus. Hear Res 137:15–28. Little HJ, Dolin SJ, Halsey MJ (1986) Calcium channel antagonists decrease the ethanol withdrawal syndrome. Life Sci 30:2059–2065. McCown TJ, Greenwood RS, Frye GD, Breese GR (1984) Electrically elicited seizures from the inferior colliculus: a potential site for the genesis of epilepsy? Exp Neurol 86:527–542. McDonough SI, Boland LM, Mintz IM, Bean BP (2002) Interactions among toxins that inhibit N-type and P-type calcium channels. J Gen Physiol 119:313–328. Mermelstein PG, Foehring RC, Tkatch T, Song W-J, Baranauskas G, Surmeier DJ (1999) Properties of Q-type calcium channels in neostriatal and cortical neurons are correlated with ␤ subunit expression. J Neurosci 19:7268 –7277. Meuth S, Budde T, Pape HC (2001) Differential control of high voltageactivated Ca current components by a Ca-dependent inactivation mechanism in thalamic relay neurons. Thalamus Relat Syst 1:31– 38. Mintz IM, Adams ME, Bean BP (1992) P-type calcium channels in central and peripheral neurons. Neuron 9:85–95. Moore JK, Moore RY (1987) Glutamic acid decarboxylase-like immunoreactivity in brainstem auditory nuclei of the rat. J Comp Neurol 260:157–174. N’Gouemo P, Rittenhouse AR (2000) Biophysical and pharmacological characterization of voltage-sensitive calcium current in neonate rat inferior colliculus neurons. Neuroscience 96:753–765. N’Gouemo P, Yasuda RP, Morad M (2001) Upregulation of voltagegated Ca2⫹ channel and downregulation of Ca2⫹-activated channel proteins in rat inferior colliculus associated with ethanol withdrawal. Abs Soc Neurosci Program 24:980.16. Ohno-Shosaku T, Hirata K, Sawada S, Yamamoto C (1994) Contribution of multiple calcium channels types to GABAergic transmission in rat cultured hippocampal neurons. Neurosci Lett 181:145–148. Peterson BZ, DeMaria CD, Yue DT (1999) Calmodulin is the Ca sensor for Ca-dependent inactivation of L-type calcium channels. Neuron 22:549 –558. Pineda JC, Walters RS, Foehring RC (1998) Specificity in the interaction of HVA Ca channel types with Ca-dependent AHPs and firing behavior in neocortical pyramidal neurons. J Neurophysiol 79: 2522–2534.

826

P. N'Gouemo and M. Morad / Neuroscience 120 (2003) 815– 826

Randall A, Tsien RW (1995) Pharmacological dissection of multiple type of Ca channels in rat cerebellar granule neurons. J Neurosci 15:2995–3012. Reigel CE, Jobe PC, Dailey JW, Savage DD (1989) Ontogeny of sound-induced seizures in the genetically epilepsy-prone rat. Epilepsy Res 4:63–71. Rhee JS, Ishibashi H, Akaike N (1999) Calcium channels in the GABAergic presynaptic terminal nerve projecting to Meynert neurons of the rat. J Neurochem 72:800 –807. Roberts RC, Ribak CE, Oertel H (1985) Increased numbers of GABAergic neurons occur in the inferior colliculus neurons of an audiogenic model of genetic epilepsy. Brain Res 361:324 –338. Rossi P, D’Angelo E, Magistretti J, Toselli M, Taglietti V (1994) Agedependent expression of high-voltage activated calcium currents during cerebellar granule cell development in situ. Pflu¨gers Arch 429:107–116. Scott VES, De Waard M, Liu H, Gurnett AC, Venzke DP, Lennon VA, Campbell KP (1996) ␤ Subunit heterogeneity in N-type Ca channels. J Biol Chem 271:3207–3212. Sitges M, Chiu LM (1995) Characterization of the types of calcium channel primary regulating GABA exocytosis from brain nerve endings. Neurochem Res 20:1073–1080. Sivaramakrishnan S, Oliver DL (2001) Distinct K currents result in physiologically distinct cell types in the inferior colliculus of the rat. J Neurosci 21:2861–2877. Smith PH (1992) Anatomy and physiology of multipolar cells in the rat inferior collicular cortex using the in vitro brain slice technique. J Neurosci 12:3700 –3715. Soldatov NM, Oz M, O’Brien KA, Abernathy DR, Morad M (1998) Molecular determinants of L-type Ca channel inactivation: segment exchange analysis of the carboxyl-terminal cytoplasmic motif encoded by exons 40 – 42 of the human alpha 1C subunit gene. J Biol Chem 273:957–863. Talley EM, Cribbs LL, Lee JH, Daud A, Perez-Reyes E, Bayliss DA (1999) Differential distribution of the three members of a gene

family encoding low voltage-activated (T-type) calcium channels. J Neurosci 19:1895–1911. Teramoto T, Niidome T, Miyagawa T, Nishizawa Y, Katayama K, Sawada K (1995) Two types of calcium channels sensitive to omega-agatoxin-TK in cultured rat hippocampal neurons. Neuroreport 6:1684 –1688. Timmermann DB, Westenbroek RU, Schousboe A, Catterall WA (2002) Distribution of high-voltage-activated calcium channels in cultured ␥-aminobutyric acidergic neurons from mouse cerebral cortex. J Neurosci Res 67:48 –61. Towbin H, Staehelin T, Gordon J (1979) Electrophoretic transfer of protein from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci USA 76:4350–4354. Wada JA, Terao A, White B, Jung W (1970) Inferior colliculus lesion and audiogenic seizure susceptibility. Exp Neurol 28:326 –332. Westenbroek RE, Hell JW, Warner C, Dubel SJ, Snutch TP, Catteral WA (1992) Biochemical properties and distribution of N-type calcium channel alpha 1 subunit. Neuron 9:1099 –1115. Westenbroek RE, Sakurai T, Elliott EM, Hell JW, Starr TVB, Snutch TP, Catteral WA (1995) Immunochemical identification and subcellular distribution of the ␣1A subunits of brain calcium channels. J Neurosci 15:6403–6418. Yokoyama CT, Westenbroek RE, Hell JW, Soong TW, Snutch TP, Catteral WA (1995) Biochemical properties and subcellular distribution of the neuronal class E calcium channel ␣1 subunit. J Neurosci 15:6419 –6432. Zeilhofer HU, Blank NM, Neuhuber WL, Swandulla D (2000) Calciumdependent inactivation of neuronal calcium channel current is independent of calcineurin. Neuroscience 95:235–241. Zu¨hlke RD, Pitts GS, Deisseroth K, Tsien RW, Reuter H (1999) Calmodulin supports both inactivation and facilitation of L-type calcium channels. Nature 399:159 –162. Zu¨hlke RD, Reuter H (1998) Ca-sensitive inactivation of L-type Ca channels depends on multiple cytoplasmic amino acid sequences of the ␣1C subunit. Proc Natl Acad Sci USA 95:3287–3294.

(Accepted 14 April 2003)