Sage Journals: Discover world-class research

Abstract

Cortex and periaqueductal grey (PAG) play a major role in the pathophysiology of migraine. Some antiepileptic drugs (AEDs) influence the activity of these structures by modulating high-voltage-activated (HVA) Ca²⁺ channels and are effective in migraine prevention. The aim of the present study was to investigate the expression of total HVA Ca²⁺ channels in cortical and PAG neurons and to study the differential action of AEDs on these channels. Isolated neurons were visually identified based on morphological criteria. HVA currents were recorded by whole-cell patch-clamp technique. The distribution ratio of L-, N-, P-, Q- and R-type HVA Ca²⁺ channels was different between cortical and PAG neurons. In particular, we found that P- and Q-type HVA Ca²⁺ channels were more expressed in PAG neurons than in cortical cells, whereas L- and R-type HVA Ca²⁺ channels showed an opposite distribution. Interestingly, N-type HVA Ca²⁺ channels were equally distributed in these two neuronal populations. A differential sensitivity to AEDs of HVA Ca²⁺ channels located on cortical and PAG neurons was observed for topiramate (TPM), but not for lamotrigine (LTG) or levetiracetam (LEV). In fact, whereas both LTG and LEV were equally effective and potent in inhibiting HVA Ca²⁺ currents in the two neuronal populations, TPM showed a much higher potency and efficacy in blocking these currents in PAG neurons than in cortical pyramidal cells. TPM, in fact, inhibited N-, P- and L-type channels in PAG neurons, whereas in cortical neurons this AED modulated only P- and L-type channels. Unlike the other AEDs investigated, valproic acid did not affect HVA Ca²⁺ currents in cortical and PAG neurons. The negative modulation of specific subtypes of HVA Ca²⁺ channels by various AEDs can restore normal electrical activity in target brain areas such as cortex and PAG, providing interesting therapeutic approaches in migraine prevention.

Keywords

Migraine AEDs periaqueductal grey cortex calcium channels

Introduction

Migraine and epilepsy share several clinical features (1). Both conditions have a paroxysmal onset and they are often strictly lateralized. They may be associated with transient motor and/or sensory symptoms. Ictal and interictal EEG abnormalities, reflecting increased neuronal excitability, can be found in migraine as well as in epilepsy.

Some antiepileptic drugs (AEDs) are effective in the prevention of migraine (2–5). A rationale for this use is the hypothesis that migraine and epilepsy share several pathogenic mechanisms (4, 5). Unbalanced activity between excitatory glutamatergic transmission and GABAergic inhibition has been postulated in these two pathological conditions. Similarly, abnormal activation of voltage-operated ionic channels such as sodium (Na⁺) and calcium (Ca²⁺) channels has been implicated in both migraine and epilepsy (5–8).

Ca²⁺ channels also play a key role in the physiology of the periaqueductal grey (PAG). PAG is a major nodal point of modulation of craniovascular nociception and, through its role in the descending pain modulation system, contributes to migraine pathophysiology (9–11). Ca²⁺ channels in the PAG modulate trigeminal nociception, and dysfunctional activity of these channels influences migraine pathophysiology (12). In fact, the pathophysiology of migraine seems to involve also the activation of trigeminal afferents (10). Pain in migraine results from the activation of trigeminovascular afferents from the meninges, and substantial evidence indicates that the trigeminovascular system is activated by cortical spreading depression, which results from neocortical hyperexcitability. The mechanisms that underlie cortical hyperexcitability are unknown, but the phenomenon could be related to excessive excitatory transmitter release resulting from alterations in Ca²⁺ channel function, as occurs in some forms of migraine (3).

Altered activity of Ca²⁺ channels in both cortical and PAG neurons can represent a target of AEDs showing efficacy in migraine prevention (5). For this reason, in the present study we have first analysed the differential expression of various subtypes of high-voltage-activated (HVA) Ca²⁺ channels in neurons isolated from cortex and PAG. In the second phase of the study, we have compared the efficacy and the potency of four different AEDs in inhibiting the various subtypes of Ca²⁺ channels in cortical and PAG neurons. In particular, we have focused our attention on topiramate (TPM) and valproic acid (VPA), whose efficacy in migraine prevention has been widely confirmed in several clinical studies (5). Moreover, we also have extended our analysis to lamotrigine (LTG), since preliminary clinical findings have raised the possibility that this AED is selectively effective in the prevention of migraine with aura (13–15). Finally, we have included in the study levetiracetam (LEV), a new AED selectively targeting HVA Ca²⁺ channels (16) and whose efficacy in migraine prevention has been reported in open-label studies (17).

Methods

Slice preparation

Preparation of brain slices has been described previously (16, 18, 19), in accordance with the guidelines of the European Union Council (86/609/EU) and the Animal Act (1986). Wistar rats (4–6 weeks of age), were sacrificed under ether anaesthesia by cervical dislocation. Coronal slices (5–10 slices per rat, 180–200 µm thick) were prepared from tissue blocks with a vibratome in oxygenated Krebs' solution (composition below) maintained at 33 °C. Slices were allowed to recover for 30–60 min. The composition of the Krebs' solution was (in mM): 126 NaCl, 2.5 KCl, 1.3 MgCl₂, 1.2 NaH₂PO₄, 2.4 CaCl₂, 10 glucose and 18 NaHCO₃. The pH of the extracellular solution was 7.4.

Preparation of acutely dissociated neurons

Slices from frontal cortex, and dorsolateral PAG were incubated in N-2-hydroxyethylpiperazine-N'-2-ethanesulphonic acid (HEPES)-buffered Hank's balanced salt solution (HBSS), bubbled with pure oxygen. Temperature was kept at 35°C. A single slice was then incubated in HBSS media with 0.5 mg/ml protease XIV added. After 30 min the tissue was repeatedly washed in HBSS and mechanically triturated with a graded series of fire-polished Pasteur pipettes. The obtained supernatant was placed in a Petri dish mounted on the stage of an inverted microscope (Nikon Diaphot, Japan). Healthy cells were allowed to settle for about 10 min. When examining dissociated neurons, only presumed pyramidal neurons, from cortex, and neurons from PAG were chosen for recordings.

Whole-cell patch-clamp recordings

Pyramidal neurons were identified by their peculiar shape and by the typical apical process spared by the enzymatic dissociation (Fig. 1). PAG neurons were identified (Fig. 1) using differential interference contrast optics on an upright microscope (Olympus, Tokyo, Japan), in accordance with previous works (20). In particular, we selected the grey matter near Silvius aqueduct, according to the stereotaxic coordinates (21).

Figure 1

Distinctive morphological and electrophysiological features of two neuronal subtypes acutely dissociated from rats. (A) Representative images of a cortical neuron and a periaqueductal grey (PAG) neuron. Left side shows a picture of a typical cell classified as a pyramidal neuron. Note the triangular shape with two large dendrites emerging from the cell body. Right side shows a picture of a PAG neuron. The size of this multipolar neuron is significantly less than that of pyramidal cortical neurons. (B) The histograms show the representative distribution of Ca²⁺ channel subtypes in pyramidal neurons and PAG neurons. Asterisk represents significance (∗P < 0.05, ∗∗P < 0.01). The pharmacological dissection of high-voltage-activated Ca²⁺ currents in cortical and PAG neurons was obtained by utilizing specific channel blockers: nifedipine (NIFE) for L-type, ω-conotoxin (Ctx) GVIA for N-type, ω-agatoxin (Atx) IVA for P-type, ω-conotoxin (Ctx) MVIIC for Q-type. (C) Representative time course showing the effect of toxins sequentially added. Drugs were applied for the duration indicated by the respective bars. Note that 200 µm cadmium (Cd²⁺) fully abolished the recorded current. A partial return to control levels was obtained after drug washout.

After enzymatic dissociation two different cell types were obtained, with distinct sizes: ≤ 15 µm and > 20 µm in diameter.

Patch-clamp recordings in the whole-cell configuration were performed utilizing fire-polished pipettes (WPI PG52165-4) pulled on a Sutter Flaming-Brown micropipette puller (Sutter Instruments, Novato, CA, USA). The pipettes had diameters of 1–2 µm and resistance ranging between 6 and 9 MΩ. The access resistance did not exceed the value of 15 MΩ (typically it was approximately 7–15 MΩ). Moreover, the resistance was monitored and compensated for (if necessary) throughout the experiment. Extracellular and dialysing solutions were prepared in order to separate effectively only barium-sensitive Ca²⁺ currents.

The internal solution to record HVA Ca²⁺ currents consisted of (in mm): N-methyl-d-glucamine 185, HEPES 40, ethyleneglycol tetraaceticacid 11, MgCl₂ 4, with the final daily addition in the working solution of phosphocreatine 20, adenosine triphosphate 2–3, guanosine triposphate 0–0.2 and leupeptin 0.2; the osmolarity was 275–280 mOsm/l. After obtaining cell access, the neuron was usually bathed in a medium composed of (in mm): triethanolamine– Cl 165, CsCl₂ 5, HEPES-Na⁺ 10 and BaCl₂ 5 as the charge carrier; pH was adjusted to 7.35 and the osmolarity to approximately 300 mOsm/l (22).

Recordings were made with an Axopatch 1D (Axon Instruments, Union City, CA, USA) at room temperature. Series resistance compensation was routinely employed (70–80%). Data were low-pass filtered (corner frequency = 5 kHz). For data acquisition and analysis pClamp 9 running on a PC was used. Barium currents were studied with both voltage steps and ramps (0.3–0.6 mV/ms). In some experiments, leak currents were subtracted online with a p/4 leak subtraction protocol. Control and drug solutions were applied with a linear array of six gravity-fed capillaries positioned within 500–600 µm of the patched neuron. Dose–response curves were achieved by adding sequentially different concentrations of AEDs. Data analysis was performed offline using Microcal Origin and Graphpad Prism software running on a PC. Values given in the text and in the figures are mean ± s.d. of changes in the respective cell populations. An anova test was used to determine if there was a significant difference between group means (P < 0.05). The Tukey post hoc test was used to perform a pair-wise comparison of the means to detect any significant difference. The α value was fixed to 0.001.

Drug source and handling

Nifedipine (NIFE), ω-conotoxin (Ctx) GVIA, ω-conotoxin (Ctx) MVIIC and ω-agatoxin (Atx) IVA were from Alomone Labs (Jerusalem, Israel). LTG was kindly provided by GlaxoSmithKline (Brentford, UK). VPA was purchased from Sigma (Milan, Italy), TPM was provided by Johnson & Johnson (Spring House, PA, USA) and LEV was obtained from UCB-Pharma (Brussels, Belgium).

Results

Distribution of HVA Ca²⁺ channels in cortex and PAG

Patch-clamp recordings were performed from acutely dissociated neurons from layer V of the frontal cortex, and from PAG, identified by means of optical microscopy (Fig. 1A). Barium-sensitive HVA Ca²⁺ currents were studied in isolation by omitting Na⁺ and K⁺ ions from intracellular and extracellular solutions (22). For this study, we operationally defined several HVA calcium current subtypes, based on the proposed nomenclature: L-type current was defined as that blocked by 5 µm NIFE, N-type current was that blocked by 1 µm Ctx GVIA, Q-type current was that blocked by 100 nm Ctx MVIIC, P-type current was that blocked by 20 nm Atx IVA. To determine subtypes of Ca²⁺ currents, we elicited the currents in control solution and again in the presence of selective organic Ca²⁺ current blockers. During the time course of these experiments, the effects of NIFE were reversible and those of Ctx GVIA, Ctx MVIIC and Atx IVA were essentially irreversible. To measure the effects of toxins, in the first part of the study toxins were used one by one in the sequence shown in Fig. 1.

Bath-application of nifedipine (5 µm), a selective L-type, dihydropyridine-sensitive HVA Ca²⁺ current blocker, reduced the total Ca²⁺ current by 22.0 ± 2.9% in cortical and by 12.6 ± 2.1% in PAG neurons (P < 0.05; n = 11 and n = 12, respectively).

Conversely, in the presence of Ctx GVIA (1 µm), a selective N-type HVA blocker, the total HVA Ca²⁺ current was inhibited by 32.2 ± 1.8% in cortical neurons (n = 13) and by 36.2 ± 3.0% in PAG neurons (n = 13) (P > 0.05, α = 0.01). Application of the selective P-type Ca²⁺ channel blocker, Atx IVA (20 nm), caused an inhibition of 17.2 ± 2.6% and 31.2 ± 3.6%, respectively in cortex (n = 12) and PAG (n = 12) (P < 0.05). Moreover, the blockade of Q-type HVA channels by Ctx MVIIC (100 nm) produced an inhibition of 15.4 ± 3.0% in cortex and 19.1 ± 4.2% in PAG (P < 0.05; n = 9 each). Finally, R-type current fraction was evaluated by subtraction of peak current amplitude in the presence of a solution of toxins (NIFE, Ctx GVIA, Ctx MVIIC and Atx IVA) from control peak current amplitude (14.2 ± 1.8% in cortex, n = 10; 7.9 ± 3.3% in PAG, n = 10; P < 005).

Effects of VPA on HVA Ca²⁺ channels in cortex and PAG

Ca²⁺ currents were activated by voltage ramps (Figs 2 and 3) or step pulses (not shown). The holding potentials ranged from −70 mV to +40 mV for the ramps and from −10 mV to +10 mV for the tests. Under these conditions, barium currents were dominated by HVA components. VPA (0.01–1 mM) was tested on 37 and 34 PAG and cortical neurons, respectively. Bath-application of VPA, at all the concentrations tested, failed to affect HVA Ca²⁺ currents (Fig. 2A; P > 0.05).

Figure 2

Effect of valproic acid (VPA) and lamotrigine (LTG) on high-voltage-activated (HVA) Ca²⁺ currents. (A) Dose–response curve of VPA on HVA currents (0.001–1 mm). Note the lack of effect of VPA on HVA Ca²⁺ currents in both neuronal subtypes. The inset shows a single experiment for each neuronal subtype. (B) Dose–response curve of LTG-mediated inhibition of barium sensitive currents. In the inset a single experiment is shown: application of 100 µm of LTG inhibited HVA Ca²⁺ currents in cortical and periaqueductal grey neurons.

Figure 3

Inhibitory action of levetiracetam (LEV) and topiramate (TPM) on high-voltage-activated (HVA) Ca²⁺ currents. (A) LEV dose-dependently inhibited HVA currents both on cortical and periaqueductal grey (PAG) neurons. The IC₅₀ was 14 µm in the cortex and 16 µm in the PAG. In the inset a single experiment for each neuronal subtype is shown. (B) Dose–response curve of TPM-mediated inhibition of barium sensitive currents. In the inset a single experiment from cortical and PAG neurons is shown: application of 50 µm of TPM reduced HVA Ca²⁺ currents in cortical and PAG neurons. Note the statistically significant difference between the IC₅₀ values of cortical and PAG neurons.

Effects of LTG on HVA Ca²⁺ channels in cortex and PAG

The modulatory effect of LTG was tested on 41 cortical pyramidal cells and 37 PAG neurons. LTG (0.1–200 µm) modulated barium-sensitive Ca²⁺ currents in a dose-dependent mode (Fig. 2B). This inhibition was fully reversible and the IC₅₀ was 13 µm in cortical neurons and 16 µm in PAG neurons (n = 11 for each neuronal subtype). In order to evaluate the Ca²⁺ channel subtype involved in this inhibitory effect, a new set of experiments was performed in the presence of selective HVA channel subtype blockers (Fig. 4A,B).

Figure 4

Characterization of the pharmacological modulation of high-voltage-activated (HVA) Ca²⁺ currents in cortical and periaqueductal grey (PAG) neurons by lamotrigine (LTG) and levetiracetam (LEV). (A) Ca²⁺currents were activated by step pulses ranging from −10 mV to +10 mV. The histograms shown on right and left side represent the mean ± s.d. of at least 11 independent recordings for each column. The inhibitory effect of saturating doses of LTG 100 µm on cortical pyramidal neurons was not fully occluded by Ctx GVIA, but only in presence of Ctx GVIA 1 µm plus Atx IVA 20 nm. Instead, the effect of LTG 100 µm on PAG neurons was fully occluded by Ctx GVIA 1 µm. (B) Representative time course shown on both sides describes occlusion experiments in pyramidal and PAG neurons. Time course is of a representative experiment. Note the LTG inhibitory effect (100 µm) was completely reversed in both cell subtypes. The LTG effect was fully occluded in the presence of Ctx GVIA only in PAG neurons, but not in cortical neurons, where Ctx GVIA 1 µm plus Atx IVA were necessary for the occlusion. (C) The histogram shows the occlusion experiments on LEV 100 µm in cortical neurons. Note that the inhibitory effect of LEV on HVA Ca²⁺ current was occluded in cortical pyramidal neurons by Ctx GVIA 1 µm plus Atx IVA 20 nm. Similar to the LTG effect, LEV 100 µm was occluded by Ctx GVIA 1 µm alone in PAG neurons. (D) The histogram shows the occlusion experiments on LEV 100 µm in PAG neurons. Representative traces showing the HVA Ca²⁺ current in control conditions (a) and in presence of Ctx GVIA 1 µm (b), Ctx GVIA 1 µm plus Atx IVA 20 nm (c), Ctx GVIA 1 µm plus Atx IVA 20 nm plus LEV 100 µm (d). Note that Ctx GVIA 1 µm alone is not sufficient to occlude the LEV inhibitory effects in pyramidal cortical neurons. Typical example of HVA Ca²⁺ current traces activated by step pulses ranging from −10 mV to +10 mV, in PAG neurons. Letters indicate control conditions (a), Ctx GVIA 1 µm effect (b) and occlusion of LEV effects by Ctx GVIA 1 µm (c).

A significant interaction was found between LTG and the N-type channel blocker. In the presence of 1 µm Ctx GVIA plus 20 nmω-agatoxin IVA, the inhibitory action of a maximal dose of LTG (100 µm) in cortical neurons was saturated (n = 11, 35.6 ± 3.8%, P < 0.05), suggesting that both N-type and P-type channels are primarily involved in the action of LTG in cortical neurons (Fig. 6).

Figure 6

Summary of pharmacological modulation of high-voltage-activated (HVA) Ca²⁺ currents by antiepileptic drugs in cortical and periaqueductal grey neurons. The histograms summarize the distinct effects of lamotrigine (LTG) (A), levetiracetam (LEV) (B) and topiramate (TPM) (C) on the Ca²⁺ channel subtypes pharmacologically isolated using selective channel blockers.

In contrast, Ctx GVIA fully occluded the inhibitory action of LTG (100 µm) in PAG neurons (n = 10, 36.0 ± 2.88% and 36.6 ± 3.0%, respectively).

Effects of LEV on HVA Ca²⁺ channels in cortex and PAG

Levetiracetam (0.1–300 µm) was tested on 36 cortical neurons and 54 PAG neurons. The saturating dose of 100 µm inhibited HVA currents by 40.11 ± 3.65% in cortical cells and 37.15 ± 2.45% in PAG neurons. The calculated IC₅₀ was 14 µm in the cortex and 16 µm in the PAG (n = 11 for each point in each group) (Fig. 3A).

In cortical neurons the effects of LEV were occluded by N-type and P-type selective blockers (Fig. 4C,D), whereas in PAG neurons LEV inhibition was saturated by Ctx GVIA 1 µm (n = 11, P < 0.05), suggesting a selective involvement of N-type channels (Fig. 6).

Effects of TPM on HVA Ca²⁺ channels in cortex and PAG

The effect of TPM (0.5–200 µm) was evaluated on 52 cortical and 58 PAG neurons. The inhibitory action of TPM was significantly different in the two regions investigated. In cortical neurons, the maximal inhibition produced by TPM (100 µm) was 33.2 ± 3.8%, with an IC₅₀ of 21 µm. Interestingly, in PAG neurons, 50 µm TPM produced an inhibition of 62.9 ± 2.7%, with an IC₅₀ of 6 µm (n = 16 for each point) (Fig. 3B).

In another set of experiments we tested the channel subtype involved in the inhibitory effect caused by TPM. In cortical neurons application of both NIFE and Atx IVA totally occluded the effects of TPM (33.1 ± 3.3% and 33.4 ± 4.1%, respectively), suggesting that in cortical neurons the modulation of TPM was mediated by L- and P-type Ca²⁺ channels (n = 7 for each group) (Fig. 5A).

Figure 5

Characterization of the pharmacological modulation of high-voltage-activated (HVA) Ca²⁺ currents in cortical and periaqueductal grey (PAG) neurons by topiramate (TPM). (A) The selective L-type Ca²⁺ channel blocker nifedipine (NIFE, 5 µm) largely suppressed the total Ca²⁺ current in pyramidal cortical neurons. In the presence of NIFE plus Atx IVA 20 nm the inhibitory effect of saturating concentrations of TPM was occluded (left side). Traces on the right of the figure show test pulses ranging from −10 mV to +10 mV. These pulses were applied to activate Ca²⁺ currents recorded from pyramidal neurons in control conditions (a). The selective L-type Ca²⁺ channel antagonist NIFE (5 µm) reduced the total Ca²⁺ current (b). NIFE plus Atx IVA significantly reduced the total Ca²⁺ current (c) and occluded the inhibitory effect of TPM on cortical neurons (e). (B) The histogram shows HVA Ca²⁺ currents recorded from PAG neurons following the application of various neurotoxins and TPM. In the right part of the figure representative traces of Ca²⁺ currents during occlusion experiments are shown. Control condition is represented in (a). NIFE (5 µm) inhibited the Ca²⁺ currents (b). NIFE plus Atx IVA 20 nm caused additional inhibition of Ca²⁺ currents (c). However, the concomitant application of NIFE plus Atx IVA and Ctx GVIA induced a greater inhibition (d) and fully occluded the TPM effect in PAG neurons (e). Histograms represent the mean ± s.d. of at least seven experiments for each group.

Conversely, in PAG neurons, a cocktail with NIFE, Ctx and Atx was necessary to obtain full occlusion of the TPM effect (n = 9, 65.8 ± 4.8% of inhibition) (Fig. 5B). These data suggest the involvement of N-, L- and P-type channels in the modulation of TPM on PAG neurons (Fig. 6).

Discussion

In the present study, the expression of total HVA Ca²⁺ channels and the specific action of AEDs on these channels have been evaluated in cortical pyramidal cells and compared with the effect on PAG neurons. The expression and pharmacological modulation of HVA Ca²⁺ channels in cortical and PAG neurons may differentially affect the pathophysiology of pain in migraine. Direct evidence of a prominent role for PAG in migraine has been provided by functional imaging studies of patients (11, 23–25). Stimulation or lesions in the PAG can produce migraine-like headache in non-migraineurs (26–28).

The PAG has been shown to modulate nociception in various experimental animal models of pain (29, 30). Moreover, it has been reported that P- and Q-type Ca²⁺ channels in the PAG facilitate trigeminal nociception (31). The existence of L-, N-, P-, Q- and R-type HVA Ca²⁺ channels in acutely dissociated PAG neurons has been previously reported (32).

We found that P- and Q-type HVA Ca²⁺ channels are more significantly expressed in PAG neurons than in cortical cells, whereas L- and R-type HVA Ca²⁺ channels show an opposite distribution. Interestingly, our findings suggest that N-type HVA Ca²⁺ channels are equally distributed in these two neuronal populations.

Although in the present study we made a major effort to dissect distinct Ca²⁺ channels properly using the best available blockers, it is possible that partial overlapping in the effects of some blockers could influence the obtained results (in particular those dealing with the pharmacological distinction between P and Q channels). We hope that in the future more selective drugs will allow us to achieve an even more accurate pharmacological dissection of Ca²⁺ channels.

The differential sensitivity to AEDs of HVA Ca²⁺ channels located on cortical and PAG neurons is one of the major findings of the present study. Whereas LTG and LEV were equally effective and potent in inhibiting HVA Ca²⁺ currents, TPM showed a much higher potency and efficacy in blocking these currents in PAG neuron than in cortical pyramidal cells. As suggested by the pharmacological analysis obtained using selective toxins for each HVA Ca²⁺ channel subtype, this difference was achieved because in PAG neurons TPM inhibited N-, P- and L-type channels, whereas in cortical neurons this AED modulated only P- and L-type channels.

It has been reported that the therapeutic action of TPM in migraine possibly involves various cellular mechanisms such as the reduction of glutamatergic transmission, the increase of GABAergic responses and inhibition of Na⁺ currents (5). However, our findings support the hypothesis that the inhibition of Ca²⁺ channels in the PAG represents a possible mechanism to explain the therapeutic action of this drug in migraine.

Abnormal activity of Ca²⁺ channels in the cortex can trigger the induction of spreading depression (SD) (33–35). SD represents a pathophysiological signal that has been associated with the induction of migraine. In particular, it can represent an electrical correlate of the neurological symptoms in migraine with aura (33–35). AEDs could selectively increase the threshold for the induction and reduce the progression of SD to counteract the frequency and intensity of migraine attacks. Accordingly, it has been recently reported that both TPM and VPA inhibit cortical SD (36, 37).

Our study has shown that VPA, unlike TPM, did not affect Ca²⁺ channels in cortical and PAG neurons. Thus, we must assume that alternative mechanisms are involved in VPA-induced inhibition of SD. VPA may suppress cortical excitability via actions on multiple sites, including ion channels and neurotransmitter receptors. Importantly, long-term treatment with VPA alters whole-brain expression of a variety of genes, including ion channels, presynaptic Ca²⁺-binding proteins and neurotransmitter receptors (37).

The similar inhibitory action of LTG and LEV on HVA currents in both PAG and cortical neurons seems to provide a cellular mechanism for their possible use in migraine prevention. However, clinically available data for this use are still scattered. In fact, although double-blind, randomized trials have failed to show a significant therapeutic effect of LTG in migraine prophylaxis (13), two small open clinical studies have found that LTG is effective in preventing migraine aura symptoms (14, 38). Accordingly, a controlled 3-year prospective open study has confirmed that LTG significantly reduced both frequency of migraine aura and aura duration (15). Similarly, an open study has suggested that LEV is a safe and effective treatment for migraine with aura (17). However, controlled trials are needed to confirm the observed results. If applied in a pathophysiological model of migraine, our electrophysiological data suggest the possibility that both LTG and LEV might selectively inhibit cortical SD via inhibition of Ca²⁺ channels and potentially normalize pathological neuronal activity in the PAG. Further pathophysiological studies are required to confirm these possible mechanisms.

Although the cortex and PAG are major sites in migraine pathophysiology, another major area of interest in migraine is the trigeminocervical complex. Interestingly, there is evidence that Ca²⁺ channel blockers, including AEDs such as topiramate, are able to inhibit trigeminovascular responses (39–42). Moreover, it should be stressed that PAG also has a critical role in inhibiting afferent trigeminal nociceptive traffic (43). Thus, we can assume that brainstem dysfunction might lead to disinhibition of trigeminal afferents and be important in the pain process of migraine. This consideration supports the possibility that AEDs prevent migraine by acting also on the trigeminocervical complex via either direct or indirect action.

Abnormal neuronal activity of Ca²⁺ channels can represent one of the major targets of AEDs in migraine (5). Accordingly, recent genetic studies on familial hemiplegic migraine (FHM) suggest that abnormalities in Ca²⁺ channels also play a critical role in migraine pathogenesis. In fact, a missense mutation of a gene, which encodes the α1A subunit of the P/Q-type Ca²⁺ channel, has been discovered in patients suffering from FHM (44). This finding raises the possibility that sporadic forms of migraine could also be considered channelopathies. Genetic migraine models will be useful in examining the critical mechanisms of migraine attacks and in identifying novel migraine prophylactic targets and therapies (45).

Thus, as suggested in the present study, the negative modulation of specific subtypes of HVA Ca²⁺ channels by various AEDs can restore normal electrical activity in target brain areas such as the cortex and PAG, providing interesting therapeutic approaches in migraine prevention.

Acknowledgements

We thank Massimo Tolu, Franco Lavaroni and Cristiano Spaccatini for their excellent technical support.

References

1 Post

Silberstein

. Shared mechanisms in affective illness, epilepsy, and migraine. Neurology 1994; 44:S37–47.

2 Cutrer

. Antiepileptic drugs: how they work in headache. Headache 2001; 41 (Suppl. 1):S3–10.

3 Rogawski

Loscher

. The neurobiology of antiepileptic drugs for the treatment of nonepileptic conditions. Nat Med 2004; 10:685–92.

4 Welch

. Brain hyperexcitability: the basis for antiepileptic drugs in migraine prevention. Headache 2005; 45 (Suppl. 1):S25–32.

5 Calabresi

Galletti

Rossi

Sarchielli

Cupini

. Antiepileptic drugs in migraine: from clinical aspects to cellular mechanisms. Trends Pharmacol Sci 2007; 28:188–95.

6 Avanzini

Franceschetti

. Prospects for novel antiepileptic drugs. Curr Opin Investig Drugs 2003; 4:805–14.

7 Pietrobon

Striessnig

. Neurobiology of migraine. Nat Rev Neurosci 2003; 4:386–98.

8 Pietrobon

. Migraine: new molecular mechanisms. Neuroscientist 2005; 11:373–86.

9 Moskowitz

. The neurobiology of vascular head pain. Ann Neurol 1984; 16:157–68.

10.

10 Goadsby

Edvinsson

. The trigeminovascular system and migraine: studies characterizing cerebrovascular and neuropeptide changes seen in humans and cats. Ann Neurol 1993; 33:48–56.

11.

11 Welch

KMA

Nagesh

Sheena

Gelman

. Periaqueductal gray matter dysfunction in migraine: cause or the burden of illness? Headache 2001; 41:629–37.

12.

12 Silberstein

. Preventive treatment of migraine. Trends Pharmacol Sci 2006; 27:410–15.

13.

13 Steiner

Findley

Yuen

. Lamotrigine versus placebo in the prophylaxis of migraine with and without aura. Cephalalgia 1997; 17:109–12.

14.

14 D'Andrea

Granella

Cadaldini

Manzoni

. Effectiveness of lamotrigine in the prophylaxis of migraine with aura: an open pilot study. Cephalalgia 1999; 19:64–6.

15.

15 Lampl

Katsarava

Diener

Limmroth

. Lamotrigine reduces migraine aura and migraine attacks in patients with migraine with aura. J Neurol Neurosurg Psychiatry 2005; 76:1730–2.

16.

16 Costa

Martella

Pisani

Bernardi

Calabresi

. Multiple mechanisms underlying the neuroprotective effects of antiepileptic drugs against in vitro ischemia. Stroke 2006; 37:1319–26.

17.

17 Brighina

Palermo

Aloisio

Francolini

Giglia

Fierro

. Levetiracetam in the prophylaxis of migraine with aura: a 6-month open-label study. Clin Neuropharmacol 2006; 29:338–42.

18.

18 Wheeler

Randall

Tsien

. Roles of N-type and Q-type Ca²⁺ channels in supporting hippocampal synaptic transmission. Science 1994; 264:107–11.

19.

Martella

De Persis

Bonsi

Natoli

Cuomo

Bernardi

Inhibition of persistent sodium current fraction and voltage-gated L-type calcium current by propofol in cortical neurons: implications for its antiepileptic activity. Epilepsia 2005; 46:624–35.

20.

20 Connor

Christie

. Modulation of Ca²⁺ channel currents of acutely dissociated rat periaqueductal grey neurons. J. Physiol 1998; 15:47–58.

21.

21 Paxinos

Watson

. The rat brain in stereotaxic coordinates, 3rd edn. San Diego: Academic Press 1997.

22.

Martella

De Persis

Bonsi

Natoli

Cuomo

Bernardi

Inhibition of persistent sodium current fraction and voltage-gated L-type calcium current by propofol in cortical neurons: implications for its antiepileptic activity. Epilepsia 2005; 46:624–35.

23.

Weiller

May

Limmroth

Juptner

Kaube

Schayck

Brain stem activation in spontaneous human migraine attacks. Nat Med 1995; 1:658–60.

24.

May

Ashburner

Buchel

McGonigle

Friston

Frackowiak

RSJ

Correlation between structural and functional changes in brain in an idiopathic headache syndrome. Nat Med 1999; 5:836–8.

25.

25 Bahra

Matharu

Buchel

Frackowiak

Goadsby

. Brainstem activation specific to migraine headache. Lancet 2001; 357:1016–17.

26.

26 Raskin

Hosobuchi

Lamb

. Headache may arise from perturbation of brain. Headache 1987; 27:416–20.

27.

27 Haas

Kent

Friedman

. Headache caused by a single lesion of multiple sclerosis in the periaqueductal gray area. Headache 1993; 33:452–5.

28.

28 Veloso

Kumar

Toth

. Headache secondary to deep brain implantation. Headache 1998; 38:507–15.

29.

29 Reynolds

. Surgery in the rat during electrical analgesia induced by focal brain stimulation. Science 1969; 164:444–5.

30.

30 Behbehani

. Functional characteristics of the midbrain periaqueductal gray. Prog Neurobiol 1995; 46:575–605.

31.

31 Knight

Bartsch

Kaube

Goadsby

. P/Q-type calcium-channel blockade in the periaqueductal gray facilitates trigeminal nociception: a functional genetic link for migraine? J Neurosci 2002; 22:RC213.

32.

32 Kim

Rhee

Akaike

. Modulation of high-voltage activated Ca²⁺ channels in the rat periaqueductal gray neurons by mu-type opioid agonist. J Neurophysiol 1997; 77:1418–24.

33.

33 Lauritzen

. Pathophysiology of the migraine aura. The spreading depression theory. Brain 1994; 117:199–210.

34.

34 James

Smith

Boniface

Huang

CLH

Leslie

. Cortical spreading depression and migraine: new insights from imaging? Trends Neurosci 2001; 24:266–71.

35.

35 Somjen

. Mechanisms of spreading depression and hypoxic spreading depression-like depolarization. Physiol Rev 2001; 81:1065–96.

36.

36 Akerman

Goadsby

. Topiramate inhibits cortical spreading depression in rat and cat: impact in migraine aura. Neuroreport 2005; 22:1383–7.

37.

37 Ayata

Jin

Kudo

Dalkara

Moskowitz

. Suppression of cortical spreading depression in migraine prophylaxis. Ann Neurol 2006; 59:652–61.

38.

38 Lampl

Buzath

Klinger

Neumann

. Lamotrigine in the prophylactic treatment of migraine aura—a pilot study. Cephalalgia 1999; 19:58–63.

39.

39 Akerman

Williamson

Goadsby

. Voltage-dependent calcium channels are involved in neurogenic dural vasodilatation via a presynaptic transmitter release mechanism. Br J Pharmacol 2003; 140:558–66.

40.

40 Storer

Goadsby

. Topiramate inhibits trigeminovascular neurons in the cat. Cephalalgia 2004; 24:1049–56.

41.

41 Akerman

Goadsby

. Topiramate inhibits trigeminovascular activation: an intravital microscopy study. Br J Pharmacol 2005; 146:7–14.

42.

42 Shields

Storer

Akerman

Goadsby

. Calcium channels modulate nociceptive transmission in the trigeminal nucleus of the cat. Neuroscience 2005; 135:203–12.

43.

43 Knight

Goadsby

. The periaqueductal grey matter modulates trigeminovascular input: a role in migraine? Neuroscience 2001; 106:793–800.

44.

Ophoff

Terwindt

Vergouwe

van Eijk

Oefner

Hoffman

Familial hemiplegic migraine and episodic ataxia type-2 are caused by mutations in the Ca²⁺ channel gene CACNL1A4. Cell 1996; 87:543–52.

45.

45 van de Ven

Kaja

Plomp

Frants

van den Maagdenberg

Ferrari

. Genetic models of migraine. Arch Neurol 2007; 64:643–6.

Antiepileptic Drugs on Calcium Currents Recorded from Cortical and PAG Neurons: Therapeutic Implications for Migraine

Abstract

Keywords

Introduction

Methods

Slice preparation

Preparation of acutely dissociated neurons

Whole-cell patch-clamp recordings

Drug source and handling

Results

Distribution of HVA Ca2+ channels in cortex and PAG

Effects of VPA on HVA Ca2+ channels in cortex and PAG

Effects of LTG on HVA Ca2+ channels in cortex and PAG

Effects of LEV on HVA Ca2+ channels in cortex and PAG

Effects of TPM on HVA Ca2+ channels in cortex and PAG

Discussion

Acknowledgements

References

Distribution of HVA Ca²⁺ channels in cortex and PAG

Effects of VPA on HVA Ca²⁺ channels in cortex and PAG

Effects of LTG on HVA Ca²⁺ channels in cortex and PAG

Effects of LEV on HVA Ca²⁺ channels in cortex and PAG

Effects of TPM on HVA Ca²⁺ channels in cortex and PAG