An Examination of Calcium Current Function on Heterotopic Neurons in Hippocampal Slices from Rats Exposed to Methylazoxymethanol

Summary:  Purpose: To study voltage‐dependent calcium currents (VDCCs) on hippocampal heterotopic neurons by using whole‐cell patch‐clamp techniques in brain slices prepared from methylaxozymethanol (MAM)‐exposed rats.

Voltage-dependent calcium channels are an essential pathway for Ca 2+ influx and play critical roles in regulating neuronal excitability (1). Several voltage-dependent calcium channel subtypes have been recognized in the central nervous system (CNS) (2)(3)(4)(5), each with characteristic voltage dependence, pharmacology, and single-channel kinetic properties. Often more than one voltage-dependent calcium current (VDCC) type coexists in the same neuron (6,7). Based on differences in kinetic-and voltage-dependent activation and inactivation properties, VDCC are subdivided into low voltageactivated (LVA) and high voltage-activated (HVA) components (8). LVA currents activate at relatively hyperpolarized potentials and inactivate more rapidly than HVA currents (9). In hippocampal CA1 pyramidal neurons, LVA currents are activated close to the resting membrane potential (near -65 mV) and show fast and complete inactivation kinetics. HVA currents are acti-vated at a threshold higher than -30 mV (10). HVA currents are further subdivided, based largely on pharmacologic profiles, into N-type, P/Q-type, L-type, and residual (R-type) components (11). Each Ca 2+ channel subtype has specialized physiologic roles in the CNS. For example, N-and P/Q-type HVA currents are essential for neurotransmitter release from presynaptic terminals (6,11,12), whereas L-type HVA channels mediate the effects of Ca 2+ on gene expression (13); and finally, LVA currents, because they operate at subthreshold levels of membrane potential, are uniquely poised to control rhythmic firing activities (9,14).
Given that VDCCs play such varied, yet important, roles in regulating neuronal excitability, it is not surprising that they have received considerable attention in the field of epilepsy research. In recent years, specific Ca 2+ channel defects have been identified in mutant mice exhibiting an absence-type epileptic phenotype (15)(16)(17)(18). VDCC expression and function also is enhanced in several rat models of epilepsy (19)(20)(21), and Ca 2+ entry blockers can prevent (or reduce) epileptic activity (22)(23)(24). Here we examined Ca 2+ channel function in an animal model of early-onset epilepsy associated with a neuronal heterotopia [e.g., rats exposed to methylazoxymethanol (MAM) in utero]. Cells within a neuronal heterotopia, in both humans and the MAM model, receive abundant catecholaminergic innervation (25)(26)(27). VDCCs are modulated by catecholamines (28)(29)(30) and are believed to be a source of independent seizure generation (31,32). Although potassium channels and ␥-aminobutyric acid (GABA) transmission have been studied on heterotopic neurons (33,34), virtually nothing is known regarding calcium channels. To examine VDCC function on heterotopic neurons and its potential modulation by catecholamines, whole-cell patch-clamp recordings and pharmacologic studies were performed on tissue slices obtained from MAM-exposed and control rats.

Prenatal methylazoxymethanol injection
Pregnant Sprague-Dawley rats were injected with either 0.9% physiologic saline (Control) or 25 mg/kg MAM. MAM was purchased from NCI Chemical Carcinogen (Kansas City, MO, U.S.A.). Intraperitoneal injections (0.3 ml, 15% DMSO) were made on day 15 of gestation (E15). All animal care and use conformed to the NIH Guide for Care and Use of Laboratory Animals and approved by the UCSF Committee on Animal Research.

Whole-cell recording
Whole-cell voltage-clamp recordings were obtained from visually identified neurons by using an infrared differential interference contrast (IR-DIC) video microscopy system (35). Conventional whole-cell patch recordings were obtained from identified neurons within 75 m of the slice surface. Patch electrodes (3-7 M⍀) were pulled from 1.5-mm o.d. borosilicate glass capillary tubing (WPI) by using a micropipette puller (Sutter P-87), coated with Sylgard (Dow Chemical), and fire polished. Intracellular patch pipette solution for wholecell recordings contained (in mM) 30 tetraethylammonium chloride (TEA), 100 CsCl, 10 HEPES, 10 EGTA, 4 NaCl, 1 MgCl 2 , 0.5 CaCl 2 , 3 Na 2 -ATP, 0.3 Na 2 -GTP (pH 7.25; 285-290 mOsm). To isolate Ca 2+ currents, slices were perfused with low sodium solution containing in (mM) 100 NaCl, 3 KCl, 1.25 NaH 2 PO 4 , 2 MgSO 4 , 26 NaHCO 3 , 2 CaCl 2 , 10 dextrose 0.001, tetrodotoxin (TTX), 5 4-aminopyridine (4-AP), 20 TEA (295-305 mOsm). VDCCs were recorded at a holding potential of -60 mV. Ca 2+ currents were activated by 100-ms depolarizing voltage steps between -70 mV and +20/+40 mV after a 50-ms hyperpolarizing prepulse to -130 mV (see inset in Fig. 2B). Admittedly, space-clamp problems can arise during Ca 2+ current activation in acute brain slices, resulting in a partially clamped action current at the outset of the voltage step. However, because the goal of these studies was to record VDCCs on heterotopic neurons, our protocol necessitates the use of acute hippocampal slices with identifiable regions of dysplasia rather than a dissociated cell preparation. As such, all data used for comparison of peak current amplitudes and generation of I-V curves were obtained at time points in the peak of activation step where voltage control problems are less significant. Additionally, VDCC protocols chosen for these studies are identical to those previously published for acute hippocampal slices (20,21,24,30). Current was recorded with an Axopatch 1D amplifier (Axon Instruments), and monitored on an oscilloscope (Tektronix). Whole-cell voltage-clamp data were lowpass filtered at 1 kHz (-3 dB, eight-pole Bessel), digitally sampled at 10 kHz, and monitored with pCLAMP software (Axon Instruments) running on a PC Pentium computer (Dell Computers). Whole-cell access resistance was carefully monitored throughout the recording, and cells were rejected if values changed by >25% (or exceeded 20 M⍀); only recordings with stable series resistance of <20 M⍀ were used for VDCC analysis.

Drugs
In some voltage-clamp experiments, pharmacologic agents were added to the perfusion medium: cadmium M) (from Alomone labs). In experiments using CdCl 2 , phosphate was omitted and MgSO 4 was substituted for MgCl 2 to prevent the precipitation of cadmium. Toxin stock solutions were prepared according to the supplier's specifications (i.e., using 0.1% BSA, 100 ml NaCl, 10 mM Tris (pH 7.5) and 1 mM EDTA as a diluent) and stored at -20°C. Toxin stock solutions were diluted to final working concentrations on the experimental day. CgTX and AgaTX were added to perfusate containing 0.1% cytochrome C (Sigma) to prevent adherence to connecting tubing and glass walls. Amiloride and nifedipine were prepared fresh and applied in the dark, as they are light-sensitive compounds. NE and adrenergic agonists were prepared fresh and protected from both light and oxidation (40 M ascorbic acid in ACSF). All other drugs were prepared from stock solutions, storage at -20°C. Drugs were tested by using a 5-min bathapplication protocol at a flow rate of ∼3 ml/min, unless otherwise indicated. Each cell was exposed to only one drug challenge, and only one cell was recorded per slice when drugs were applied.

Statistical analysis
VDCCs were analyzed off-line by using Clampfit software (Axon Instruments). Kinetic analysis of the VDCCs was performed with a single-exponential function. Results are presented as mean ± SEM. To compare results between different cell types, we used a one-way analysis of variance (ANOVA) on the SigmaStat program (Jandel Scientific). Significance level was taken as p < 0.05.

RESULTS
Hippocampal slices from MAM-exposed rats contained distinct clusters of displaced neurons (heterotopia) and loss of lamination (Fig. 1A), as described previously (32,36). Neurons were selected for whole-cell voltageclamp studies based on their location and morphology under direct IR-DIC visualization (Fig. 1B). Experimental data were obtained from hippocampal heterotopic pyramidal-like neurons (n ‫ס‬ 91). For comparison, control data were obtained from normotopic pyramidal neurons in hippocampal slices from MAM-exposed rats (e.g., pyramidal cells located within the normal CA1 laminar; n ‫ס‬ 40) and CA1 pyramidal neurons in slices from age-matched control rats (n ‫ס‬ 92).

Calcium current on heterotopic neurons
To study Ca 2+ channel function on hippocampal heterotopic neurons, we examined whole-cell Ca 2+ currents by using visualized patch-clamp recording techniques (35). Whole-cell VDCC was recorded in the presence of 1 M tetrodotoxin (Na + channel blocker), 20 M tetraethylammonium chloride, and 5 M 4-aminopyridine (K + channel blockers). VDCC displayed a current peak maximum, for CA1 pyramidal cells, during depolarizing steps to approximately -30 mV (n ‫ס‬ 92). By using the same voltage-clamp protocol and identical recording conditions for slices from MAM-exposed rats, we observed a current peak maximum for normotopic (n ‫ס‬ 40) and heterotopic pyramidal (n ‫ס‬ 91) neurons during depolarizing steps similar to the values observed in CA1 pyramidal cells [i.e., around -30 mV ( Fig. 2A-C)]. The inactivation time constant for VDCC evoked at -30 mV (depolarizing step for the maximum peak current value) was 49.9 ± 2.6 ms for control CA1 pyramidal cells (n ‫ס‬ 45). In MAM-exposed rats, the inactivation time constant at the same depolarizing step was 43.8 ± 2.3 ms for normotopic cells (n ‫ס‬ 21; p > 0.1) and 48.2 ± 2.4 ms for heterotopic pyramidal neurons (n ‫ס‬ 38; p > 0.5). Qualitative properties of VDCC ( Fig. 2A) and current-voltage plots (Fig. 2B) also failed to reveal significant differences between these three cell types.

Pharmacology of calcium current on heterotopic cells
Further to characterize VDCC on heterotopic neurons, we used a variety of pharmacologic manipulations designed to block Ca 2+ channels. First, slices were bathed in a low-Na + ACSF supplemented with the dihydropyri- dine L-type Ca 2+ channel blocker, nifedipine (10 M) (37,38). An example of the effect of 10 M nifedipine on VDCC in a heterotopic cell is illustrated in Fig. 3A. Second, slices were perfused in low Na + ACSF supplemented with a T-type Ca 2+ channel blocker, amiloride (1 mM) (39). Third, slices were bathed in a low-Na + ACSF containing inorganic Ca 2+ channel blockers (200 M CdCl 2 or 250 M NiCl 2 ). The level of inhibition observed with normotopic, heterotopic, and CA1 pyramidal cells was similar for all of these manipulations ( Fig. 3B; p > 0.2). In an additional set of pharmacologic studies, slices were perfused in low-Na + ACSF containing Ca 2+ channel subunit-specific toxins: 0.1 M AgaTX (P/Qtype), 1 M CgTX (N-type), or 1 M sFTX-3.3 (P-type). An example of the effect of sFTX-3.3 on VDCC in a heterotopic cell is illustrated in Fig. 4A. In agreement with data using less-specific VDCC blockers (e.g., nifedipine, amiloride, nickel, and cadmium), the level of inhibition observed with channel subunit-specific toxins was similar for all three cell types ( Fig. 4C; p > 0.5).

Adrenergic modulation of calcium current on heterotopic cells
Because hippocampal heterotopic neurons in the MAM model receive excessive innervation by catecholaminergic fibers (25)(26)(27), and it is well established that catecholamines modulate Ca 2+ channel activity (40) or hippocampal neuronal excitability (28,40,41,42), we investigated adrenergic modulation of VDCC on heterotopic neurons. Bath application of norepinephrine (NE, 10-100 M), a potent agonist at ␣ 1, ␣ 2 , and ␤-adrenergic receptors, significantly reduced VDCC peak current amplitude in a dose-dependent manner on both CA1 pyramidal control neurons and heterotopic cells (Fig. 5). An example of the effect of 10 M NE on VDCC in a heterotopic cell is illustrated in Fig. 5A. NE produced a similar level of VDCC inhibition for both cell types ( Fig.  5C and D). In additional experiments, we tested noradrenergic receptor-specific agonists dissolved in low-Na + ACSF: clonidine, 5 M (␣ 2 ); phenylephrine, 5 M (␣ 1 ); and isoproterenol, 5 M (␤). Among these agonists, the ␣ 2 -adrenergic agonist clonidine caused the most significant inhibition of VDCC, whereas the ␤-agonist, isoproterenol, produced very little effect. Again, NE-induced inhibition of VDCC by using receptor-specific agonists was similar for all cell types and all drugs tested. Despite the massive catecholaminergic innervation of hippocampal heterotopia (25,43), inhibition of VDCC in heterotopic cells by adrenergic agonists was comparable to control cells (Fig. 5D; p > 0.4)

DISCUSSION
We described the properties of voltage-dependent Ca 2+ currents on heterotopic neurons in the MAM model of brain malformation-associated epilepsy. HVA-and LVA-type voltage-dependent Ca 2+ currents were ob-  served on all heterotopic neurons. VDCC currentactivation thresholds and inactivation time constant values were not different between hippocampal heterotopic and control CA1 pyramidal neurons. No differences in VDCC peak current amplitude, voltage dependence, or pharmacology were observed. The response to exogenously applied catecholamines also was similar for VDCCs recorded on heterotopic and control cells. Thus our main finding is that heterotopic neurons in the hippocampus of MAM-exposed rats exhibit VDCC with "normal" physiologic and pharmacologic properties.

Pharmacology of VDCC on heterotopic neurons
A number of pharmacologic agents acting on VDCCs have been described. For example, dihydropyridines, such as nifedipine, block L-type Ca 2+ channels (3,6,38,44,45), and amiloride blocks T-type Ca 2+ channels (39). Therefore we used these channel blockers in an attempt to distinguish between the types of Ca 2+ channels expressed on heterotopic cells, normotopic cells, and CA1 control cells. Nifedipine blocked ∼70% of the peak Ca 2+ current on heterotopic cells. Similar blocking effects (∼70%) were obtained with amiloride. The response to these agents was comparable for all three cell types, suggesting that the expression of functional L-and T-type channels on these cells is similar. Several "N-like" HVA neuronal Ca 2+ channels also can be distinguished by using peptide toxins obtained from the venom of predatory invertebrates (snails and spiders). These toxins have high-affinity for Ca 2+ channels, and it is well established that N-type Ca 2+ channels are sensitive to micromolar levels of -conotoxin GVIA (6,46); P/Q-type channels are sensitive to -agatoxin IVA (47); P-type are sensitive to FTX-3.3 (48), and R-type channels resistant to these toxins (49). Except for their pharmacology, N-P/Q-, and R-types of Ca 2+ channels appear functionally similar. The cell body of vertebrate neurons also can express a specific mixture of various HVA Ca 2+ channels, including L type (50). We found that -conotoxin GVIA, -agatoxin KT, and sFTX-3.3 (the synthetic analogue of FTX-3.3) inhibited VDCC on heterotopic cells (∼30%, ∼30%, and ∼50%, respectively). Again, we observed nearly identical responses to these toxins in normotopic and control CA1 pyramidal cells, suggesting that the specific distribution of Ca 2+ channel subtypes is similar for all three hippocampal cell types. Because significant differences were not observed with any of the calcium FIG. 5. Adrenergic modulation of voltage-dependent calcium currents (VDCCs). A: Superimposed traces of a maximum peak current amplitude (evoked at -20 mV) from a heterotopic neuron before (black trace) and ∼5 min after application of norepinephrine (NE), 10 µM (gray trace). B: I/V curve showing the peak Ca 2+ current amplitudes of heterotopic cells before (᭹) and after (᭺) NE. C: Dose-effect curve of NE on peak Ca 2+ current amplitude for Control CA1 (ࡗ) and heterotopic cells (). D: Percentage of peak current amplitude after application of NE, 10 µM; clonidine, 5 µM; phenylephrine, 5 µM; or isoproterenol, 5 µM; for CA1 control (black) and heterotopic (white) cells. N = 8 cells for each experimental manipulation. Note that the adrenergic modulation of VDCCs was equivalent in both cell types.

FIG. 4.
Effect of Ca 2+ channel subunit-specific toxins on voltage-dependent calcium channels. A: Superimposed traces of a maximum peak current amplitude (evoked at -30 mV) from a heterotopic neuron before (black trace) and ∼5 min after application of sFTX-3.3 1 µM (gray trace) and its respective I/V curve before (᭹) and after (᭺) sFTX-3.3. B: Plot illustrating the percentage of peak current amplitude after application of AgaTX, 0.1 µM; CgTX, 1 µM; or sFTX-3.3, 1 µM; for CA1 control (black), MAM CA1 normotopic (gray), and heterotopic (white) cells. N = 3 cells for each experimental manipulation. Note the similarity of effect of Ca 2+ channel blockers for all cell types here and in Fig. 3. channel blockers tested, further isolation and analysis of specific calcium channel components was not pursued.

Catecholaminergic modulation of heterotopic neurons
Neurotransmitters that affect intracellular secondmessenger systems modulate Ca 2+ channel function and thus influence neuronal output. Norepinephrine, for example, decreases N-, P/Q-, and R-type VDCCs and therefore the amount of neurotransmitter released at a given synapse (51). In neurons, these modulatory actions can significantly reduce neurotransmitter release and depress fast synaptic transmission. Because release of neurotransmitters is a steep function of presynaptic Ca 2+ entry, a depression of Ca 2+ current would significantly reduce synaptic output. Such presynaptic inhibition could explain how some neurotransmitters block excitatory synaptic transmission in the hippocampus [e.g., direct inhibition of presynaptic voltage-dependent Ca 2+ channels (52)]. Because of the importance of NE-mediated modulation of excitability and anatomic studies in human tissue from patients with cortical dysplasia (53,54) or tissue sections from MAM-exposed rats (25,26,55) demonstrating that dysplastic or heterotopic cell regions were heavily innervated by catecholaminergic fibers, we were interested in determining whether VDCC modulation by exogenous NE was altered in the MAM model.
We observed that VDCC on heterotopic neurons is depressed by adrenergic receptor activation in a dosedependent manner. Although immunohistochemistry revealed an abundance of catecholaminergic fibers innervating cortical/hippocampal regions in the MAM brain (26,55), an altered inhibitory catecholaminergic effect to exogenously applied agonists, which could potentially contribute to abnormal modulation of excitatory synaptic transmission in the malformed hippocampus, was not observed. Responsiveness of VDCC on heterotopic neurons to NE application was similar to that measured for normal CA1 pyramidal cells. Similar findings were observed by using adrenergic receptor-specific agonists. Previously NE was reported to modulate hippocampal excitability via activation of ␣ 1 -(28) and ␣ 2adrenoreceptors (56) but not ␤-adrenoreceptors. Our results in both dysplastic and normal hippocampus are consistent with NE-mediated modulation of VDCCs via ␣-adrenergic receptors, more specifically ␣ 2 -adrenoreceptors.

Conclusion
We described, for the first time, the physiologic and pharmacologic properties of VDCC in an animal model featuring nodular heterotopia. We detected no differences in VDCCs between CA1 pyramidal cells, normotopic, and hippocampal heterotopic cells. Heterotopic neurons, therefore, do not appear to exhibit Ca 2+ channel abnormalities that could contribute to the observed hyperexcitability in the MAM model. Our results suggest that some other property of heterotopic neurons accounts for the differences in intrinsic firing properties reported (32,57)-a lack of functional A-type Kv4.2 potassium channels is one possibility (33)-and that VDCC responsiveness to exogenous adrenergic agonists is unchanged in these animals. Nonetheless, these findings add to our growing understanding of how heterotopic neurons function in the MAM model of malformation-associated epilepsy and could yield insights into the human condition.