THE COMPLEX ACTIONS OF SUMATRIPTAN ON RAT DURAL AFFERENTS

Abstract

Aim

To test the hypothesis that the clinical efficacy of triptans reflects convergent modulation of ion channels also involved in inflammatory mediator (IM)-induced sensitization of dural afferents.

Methods

Acutely dissociated retrogradely labeled dural afferents were studied with whole cell and perforated patch techniques in the absence and presence of sumatriptan and/or IM (prostaglandin E2, bradykinin, and histamine).

Results

Sumatriptan dose-dependently suppressed voltage-gated Ca²⁺ currents. Acute (2 min) sumatriptan application increased dural afferent excitability and occluded further IM-induced sensitization. In contrast, pre-incubation (30 min) with sumatriptan had no influence on dural afferent excitability and partially prevented IM-induced sensitization of dural afferents. The sumatriptan-induced suppression of voltage-gated Ca²⁺ currents, acute sensitization and pre-incubation-induced block of IM-induced sensitization were blocked by the 5-HT_1D antagonist, BRL 15572. Pre-incubation failed to suppress the IM-induced decrease in action potential threshold and overshoot (which results from modulation of voltage-gated Na+ currents) and activation of Cl⁻ current, and had no influence on the Cl⁻ reversal potential. However, pre-incubation with sumatriptan caused a dramatic hyperpolarizing shift in the voltage dependence of K⁺ current activation.

Discussion

These results indicate that while the actions of sumatriptan on dural afferents are complex, at least two distinct mechanisms underlie the antinociceptive actions of this compound. One of these mechanisms, the shift in the voltage-dependence of K⁺ channel activation may suggest a novel strategy for future development of anti-migraine agents.

INTRODUCTION

Migraine is a debilitating neurological disorder that impacts a large percentage of the population (1). Furthermore, the social and economic burden of this disorder remains a major concern despite the prophylactic and abortive agents used to treat migraine pain (2, 3). Previous data indicate that the release of inflammatory mediators in the dura and subsequent dural afferent sensitization are important for initiating migraine pain (4, 5). Furthermore, we have recently demonstrated that inflammatory mediators (IM) not only sensitize the vast majority of dural afferents (6), but that this sensitization reflects the modulation of a number of different ion channels, at least one of which appears to be unique to dural afferents (7).

Triptans, one of the most effective classes of drugs for the treatment of migraine pain, are serotonin 1B/1D (5-HT_1B/1D) receptor agonists. While the 5-HT_1B receptors appear to be primarily post-synaptic, located on vascular smooth muscle, the 5-HT_1D receptors are located on the peripheral and central terminals of dural afferents (8, 9). Interestingly, despite widespread distribution of 5-HT_1D receptors in trigeminal (TG) and dorsal root ganglion (DRG) neurons (9), clinical data indicate that these compounds have little, if any utility in the treatment of anything but migraine pain (10). And while pre-clinical data suggest that these compounds may have anti-inflammatory efficacy (11) and may be analgesic when directly applied to the CNS (12, 13), systemic administration of triptans selectively inhibit nociceptive behavior (14), neuropathic pain behavior (15), and evoked activity in trigeminal dorsal horn neurons (16) in response to noxious stimulation of trigeminal targets.

These observations raised the possibility that the clinical selectivity and efficacy of triptans reflect a unique mechanism of action on dural afferents. To begin to assess this possibility, we examined the effect of sumatriptan on the excitability of dural afferents as well as the influence of this compound on IM-induced sensitization of dural afferents.

MATERIALS AND METHODS

Animals

Adult female Sprague Dawley rats (Harlan, Indianapolis, IN) weighing between 180–290 g were used for all experiments. Rats were housed two per cage at the University of Pittsburgh animal facility on a 12:12 light: dark schedule with food and water freely available. Prior to all procedures, animals were deeply anesthetized with an i.p. injection (1 ml/kg) of rat cocktail containing ketamine (55mg/kg), xylazine (5.5 mg/kg) and acepromazine (1.1 mg/kg). Experiments were approved by the University of Pittsburgh Institutional Animal Care and Use Committee and performed in accordance with National Institutes of Health guidelines for the use of laboratory animals in research. All efforts were employed to minimize the total number of animals used.

Retrograde labeling

Afferents innervating the dura were identified as previously described following labeling with the retrograde tracer 1,1'-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine perchlorate (DiI, Invitrogen, Carlsbad, CA) (8). Immediately post-operatively, animals received a single i.m. injection of penicillin G (10,000 units/kg) and a single injection of buprenorphine (0.03 mg/kg) to minimize infection and discomfort. Subsequent administration of ketoprofen was provided if evidence of hypersensitivity persisted over subsequent days post-labeling.

Tissue Preparation

Ten to fourteen days following DiI application, trigeminal ganglia (TG) were removed, enzymatically treated and mechanically dissociated as previously described (7). Acutely dissociated cells were plated on laminin/ornithine coated glass coverslips. Changes in current and excitability were measured 2–8 hours after cells were plated.

Electrophysiology

All whole cell and perforated patch-clamp recordings were performed with a HEKA EPC10 amplifier (HEKA Electronik, Lambrecht/Rhineland-Pfalz, Germany). Data were low-pass filtered at 5–10 kHz with a four-pole Bessel filter and digitally sampled at 25–100 kHz.

Current Clamp

To assess changes in excitability, borosilicate glass electrodes were filled with (mM) K-methanesulfonate 110, KCl 30, NaCl 5, CaCl₂ 1, MgCl₂ 2, HEPES 10, EGTA 11, Mg-ATP 2, Li-GTP 1, pH 7.2 (adjusted with Tris-base), 310 mOsm (adjusted with sucrose). Bath solution contained (mM) KCl 3, NaCl 130, CaCl₂ 2.5, MgCl₂ 0.6, HEPES 10, glucose 10, pH 7.4 (adjusted with Tris-base), 325mOsm (adjusted with sucrose) and either vehicle (0.01% ETOH and 0.1% acetic acid) or test compounds: IM [(μM) bradykinin 10, histamine 1, and prostaglandin E₂ 1]; and/or sumatriptan (1μM). Excitability was assessed with three parameters as previously described (6): rheobase, action potential threshold, and the response to suprathreshold stimulation. A neuron was considered sensitized if application of a test solution resulted in a hyperpolarization of action potential threshold, decrease in rheobase, and/or an increase in the response to suprathreshold stimulation greater than 2 SD's from the baseline mean.

Passive properties measured were resting membrane potential (E_m), and input resistance (R_in). R_in was assessed with five 750-ms hyperpolarizing current injections (2–5 pA) from E_m immediately before and 90 s after the application of sumatriptan alone, IM alone, or sumatriptan and IM. Active electrophysiological properties were assessed with an action potential (AP) evoked with a 4-ms depolarizing current pulse. These included: AP duration at 0 mV, magnitude of AP overshoot, magnitude of the after-hyperpolarization (AHP), AHP decay (τ AHP). The magnitude of the overshoot was measured from 0 mV. The magnitude of the AHP was measured from the E_m. Decay of the AHP was estimated by fitting the decay phase of the AHP with a single exponential function.

Voltage Clamp

To isolate Ca²⁺ currents, electrodes were 1 – 4 MΩ when filled with and electrode solution containing (mM): Cs-methanesulfonate 100, Na-methanesulfonate 5, TEA-Cl 40, CaCl₂ 1, MgCl₂ 2, EGTA 11, HEPES 10, pH 7.2 (adjusted with Tris-base), 310 mOsm (adjusted with sucrose). The bath solution contained (mM): choline-Cl 100, TEA-Cl 30, CaCl₂ 2.5, MgCl₂ 0.6, NFA 0.1, HEPES 10, glucose 10, pH 7.4 (adjusted with Tris-base), 325 mOsm (adjusted with sucrose).

IM-induced Cl⁻ currents (I_IM-Cl) were isolated with electrode solutions containing (mM) Cs-methanesulfonate 100, CsCl 30, CaCl₂ 1, MgCl₂ 2, HEPES 10, EGTA 11, Mg-ATP 2, Li-GTP 1, pH 7.2 (adjusted with Tris-base), 310 mOsm (adjusted with sucrose) and bath solution containing (mM) Choline-Cl 130, CaCl₂ 2.5, MgCl₂ 0.6, HEPES 10, glucose 10, pH 7.4 (adjusted with Tris-base), 325mOsm (adjusted with sucrose). I_IM-Cl was elicited with 100 ms test pulses from −70 to +50mV following a 40ms prepulse to 0 mV to evoke Ca²⁺ currents in the presence of sumatriptan with and without IM. I_IM-Cl was also recorded with Ca²⁺ artificially buffered to 622nM with an electrode solution containing EGTA (1.2 mM), Ca²⁺ (1 mM) and Mg²⁺ (2 mM) and influx via voltage-gated Ca²⁺ channels was also blocked by the addition of Cd²⁺ (50 μM) to the bath solution. MaxChelator was used to generate estimates of resting free intracellular Ca²⁺.

To isolate K⁺ currents, electrodes were 1–4 MΩ when filled with (mM) K-methanesulfonate 110, KCl 30, NaCl 5, CaCl₂ 1, MgCl₂ 2, HEPES 10, EGTA 11, Mg-ATP 2, Li-GTP 1, pH 7.2 (adjusted with Tris-base), 310 mOsm (adjusted with sucrose). Bath solution contained (mM) KCl 3, Choline-Cl 130, CaCl₂ 2.5, MgCl₂ 0.6, NFA 0.1, HEPES 10, glucose 10, pH 7.4 (adjusted with Tris-base), 325 mOsm (adjusted with sucrose). Because the bath solution contained Ca²⁺, total K⁺ current consisted of both voltage-gated K⁺ and Ca²⁺ modulated K⁺ currents (7).

Drugs

All salts and reagents were obtained from Sigma-Aldrich (St. Louis, MO), unless indicated below. Bradykinin was dissolved in 1% acetic acid (23.58mM stock concentration), PGE₂ was dissolved in 100% ETOH (10mM stock concentration), and histamine was dissolved in water (100mM stock concentration). All stock solutions were stored at −20°C until the day of use. IM-vehicle bath containing the final concentration of ETOH (0.01%) and acetic acid (0.001%) was used as a control. Niflumic acid (NFA) was dissolved in 100% ETOH. Sumatriptan was a generous gift from Glaxo Smith Kline. Sumatriptan was dissolved as a 10mM stock solution in water and subsequently diluted in bath solution. The 5-HT_1D receptor antagonist BRL 15572 was obtained from Tocris Biosciences (R & D Systems, Minneapolis, MN), was dissolved as a 10 mM stock in 100% ethanol and diluted in bath solution.

Data Analysis

Data were analyzed with PulseFit (HEKA), Sigma Plot and Sigma Stat software (Systat Software Inc., Richmond, CA). Conductance-voltage (G-V) curves were constructed from I–V curves by dividing the evoked current by the driving force on the current, such that G = I/(V_m - V_rev) where V_m is the potential at which current was evoked and V_rev is the reversal potential for the current was measured directly (for K and Ca²⁺). Instantaneous I–V data was obtained from the tail currents measured following activation of voltage-gated Ca²⁺ currents.

Statistical Analysis

For comparisons of data collected before and after IM application, either a paired t-test or Repeated Measures ANOVA was used if data were parametric. Otherwise, a Wilcoxin or Friedman test was used for nonparametric analysis. For unpaired comparisons, Student's t-test, one- and two-way ANOVA were used for parametric data and a Mann Whitney U for nonparametric analysis. Data were considered statistically significant when p < 0.05. All data are represented as mean ± standard error.

RESULTS

Data was collected from 78 dural afferents acutely dissociated from 15 female Sprague Dawley rats. Of these, 53 were studied in voltage-clamp and 25 were studied in current clamp.

Sumatriptan dose-dependently inhibits voltage-gated calcium currents (VGCC)

Previous data suggests a primary mechanism of triptan action is a G protein-mediated inhibition of voltage-gated Ca²⁺ currents (17, 18). To determine if such a mechanism exists in dural afferents and to determine the appropriate concentration of sumatriptan for subsequent experiments, VGCC in dural afferents (n=7) were studied with increasing concentrations (0.001 to 10 μM) of sumatriptan. Currents were evoked with 50ms pulses from −60 to +80mV following a 100ms prepulse to −100mV.

A concentration-dependent inhibition of VGCC was observed in 7 of 7 dural afferents studied (Fig 1A, B). Inhibition of peak current evoked at 10 mV was converted to percent inhibition and data from all 7 neurons were pooled, plotted as a function of the concentration of sumatriptan and fitted with a modified Hill equation (Fig 1C). The IC₅₀ for sumatriptan-induced inhibition of VGCC was 142 nM, with a maximal fractional inhibition of 20 ± 2%. Interestingly, there was no evidence of a low threshold VGCC in any of the dural afferents studied.

Figure 1. Sumatriptan inhibits high threshold voltage gated Ca²⁺ currents (I_Ca) in dural afferents.

A. Example of sumatriptan mediated inhibition of I_Ca. B. Bath application of increasing concentrations of sumatriptan from 0.001 to 10μM suppressed I_Ca amplitude recorded with a single pulse to 10mV in 7 of 7 dural afferents studied. C. The IC₅₀ for sumatriptan was determined with percent inhibition plotted against sumatriptan concentration. D. To examine the voltage dependence of inhibition, currents were elicited with a test pulse to +10mV following pre-pulses to −60 and +80mV before (Baseline) and after sumatriptan (10 μM Suma) application (n=5). The ratio of the current amplitude following a prepulse to +80 divided by the current amplitude following a prepulse to −60 was determined before and after sumatriptan application. Following sumatriptan application, there was no significant difference in the current ratio. E. Instantaneous I–V data were plotted from tail currents. Sumatriptan decreased the amplitude of the tail currents but did not produce a shift in their voltage dependence of activation.

To determine if a membrane delimited displacement of the N-type Ca²⁺ channel β-subunit by the G-protein βγ subunits (19) could also be mediating the decrease in VGCC with sumatriptan, Ca²⁺ currents were elicited with a two-pulse protocol in which a test pulse to +10mV was preceded by a conditioning pulse to either −60 mV or +80 mV (20). A 50 ms step to −60 mV between the conditioning and test pulses was used to enable channel deactivation following the step to +80 mV (n=5). Consistent with the absence of a detectable shift in the VGCC instantaneous I–V curve, there was no evidence of pre-pulse potentiation as the ratio of the currents elicited before sumatriptan application (1.07 ± 0.03) were comparable to that after application (1.12 ± 0.07, Fig 1D). To confirm that sumatriptan-induced inhibition of voltage-gated Ca²⁺ currents in dural afferents was mediated by the 5-HT_1D receptor, sumatriptan (1 μM) was co-applied to 3 dural afferents with the 5-HT_1D receptor selective antagonist BRL 15572 (1 μM). Two minutes after the application of the combination of sumatriptan and BRL 15572, the decrease in maximal conductance (7.0 ± 0.1% of baseline) was significantly (p < 0.01, Student's t-test) less than that observed with sumatriptan alone.

Acute sumatriptan increases baseline dural afferent excitability

Acute (2 min) application of 1μM sumatriptan alone produced a significant increase in excitability of dural afferents (n = 7) as evidenced by changes in rheobase (p < 0.01, Fig 2A), action potential threshold (p < 0.01, Fig 2B), and the response to suprathreshold current injection (p < 0.01, Fig 2C). These changes were associated with a significant (p < 0.01, paired t test) depolarization of E_m from −71.3 ± 1.6 mV to −54.0 ± 3.6 mV. These sumatriptan-induced changes in excitability were blocked by the co-application of the 5-HT_1D receptor antagonist BRL 15572 (1 μM, n = 5, Fig 2A, B and C).

Figure 2. Acute sumatriptan increases dural afferent excitability.

A. Acute bath application of 1μM sumatriptan resulted in a significant reduction in rheobase in dural afferents (n=7). This effect was blocked when sumatriptan was co-applied with the 5-HT_1D receptor antagonist BRL 15572 (Antag, 1 μM, n = 6). When IM were applied to dural afferents in the presence of sumatriptan, there was no further decrease in rheobase. B. Acute sumatriptan application also significantly hyperpolarized the AP threshold. This change was also blocked when sumatriptan was co-applied with BRL 15572. There was no further change in AP threshold following IM application. Data in A and B were analyzed with a one way ANOVA with a Holm-Sidak test used for post-hoc analysis. The most relevant comparisons are illustrated for clarity where * is p < 0.05. C. The stimulus response function data in C for neurons treated with sumatriptan alone (Suma) or sumatriptan + BRL 15572 (Suma + Antag) were analyzed with a Fisher Exact test. The proportion of neurons treated with Suma alone (7 of 7) with a left shift in the stimulus response function (relative to baseline) was significantly (p < 0.05) greater than that for the Suma + Antag group (2 of 6). There was no further shift in the stimulus response function in the Suma group (0 of 7) following application of IM. Baseline data are plotted for comparison.

To determine whether sensitization of dural afferents via sumatriptan or IM involve comparable mechanisms, IM were applied following sumatriptan. No further increase in excitability was detected in these neurons (Fig 2). These results suggest that either sumatriptan-induced sensitization shares common mechanisms with those of IM, or this drug has blocked the actions of IM.

Prolonged sumatriptan exposure has no influence on baseline excitability and attenuates IM-induced sensitization of dural afferents

There is evidence that triptan analgesia does not occur immediately after administration. Instead, pain relief is experienced typically 20–30 min after taking the drug (21, 22). Therefore, we examined the possibility that with a longer exposure time, sumatriptan may switch from being excitatory to inhibitory.

Following 30 minute pre-incubation with sumatriptan, in which neurons (n = 8) were incubated in sumatriptan prior to recording, there was no significant (p > 0.05, Student's t test) difference in rheobase compared to that in control (vehicle) neurons (n=8) indicating that the decrease in rheobase following acute sumatriptan application returns to baseline levels with longer incubation times: rheobase normalized to membrane capacitance was 8.4 ± 1.3 pA/pF and 9.5 ± 1.5 pA/pF in neurons from vehicle and sumatriptan treated groups, respectively. Similarly, sumatriptan pre-incubation produced no significant (p > 0.05, Student's t test) changes in AP threshold compared to control: AP threshold was −29.5 ± 1.6 mV and −23.6 ± 4.3mV in neurons from vehicle and sumatriptan treated groups, respectively. There was also no significant (p < 0.05, two-way repeated measure ANOVA) influence of sumatriptan pre-incubation on the response to suprathreshold current injection (Fig 3C). Furthermore, pre-incubation of sumatriptan with BRL 15572 (n = 5) had no detectable influence on rheobase (which was 7.3 ± 2.1 pA/pF), action potential threshold (which was −25.4 ± 4.0), the or the response to suprathreshold current injection (Fig 3C).

Figure 3. Prolonged sumatriptan exposure has no influence on excitability and attenuates IM-induced sensitization of dural afferents.

A. Following 30 minute pre-incubation with sumatriptan, IM application (Suma + IM) had little influence on rheobase (expressed as a % of baseline determined prior to the application of IM for each neuron). However, the IM-induced decrease in rheobase in neurons pre-incubated with the combination of sumatriptan and BRL 15572 (Suma + antag + IM) was significantly (Student's t test) greater than the change in observed in neurons treated with sumatriptan alone. B. In contrast to rheobase, application of IM resulted in a decrease in AP threshold in neurons preincubated with sumatriptan alone as well as the combination of sumatriptan and BRL 15572. There was no significant difference between these groups. C. Sumatriptan pre-incubation (Suma) had no significant influence on the baseline response to suprathreshold stimulation. Nor was there an influence of pre-incubating neurons with the combination of sumatriptan and BRL 15572 (Suma + Antag). Data were analyzed with a two-way repeated measures ANOVA and compared to control neurons incubated in vehicle for 30 minutes (Vehicle). Furthermore, the application of IM to neurons pre-incubated for 30 minutes with sumatriptan (Suma + IM) had no significant influence on the stimulus response function as determined with a one-way repeated measures ANOVA. However, application of IM to neurons pre-incubated with the combination of sumatriptan and BRL 15572 (Suma + Antag +IM) resulted in a significant leftward shift in the stimulus response function as determined by both the increase in the number of evoked action potentials at 2× and 3× rheobase (relative to the response prior to the application of IM), as well as the proportion of neurons in which IM produced a change (4 of 4) relative to the Suma + IM group (0 of 7, p < 0.01, Fisher Exact test). * Indicates a significant difference between groups in A and before and after IM application in C where p<0.05.

To determine the effects of sumatriptan pre-incubation on IM-induced sensitization of dural afferents, changes in excitability were recorded with IM in the presence of sumatriptan. In contrast to our previous observations in which application of IM to dural afferents resulted in a significant decrease in rheobase and leftward shift in the response to suprathreshold stimulation (6), IM had no significant influence on rheobase (Fig 3A) or the response to suprathreshold stimulation (Fig 3C) in dural afferents pre-incubated with sumatriptan as compared to vehicle treated dural afferents. However, pre-incubation with sumatriptan did not prevent IM-induced hyperpolarization of AP threshold (Fig 3B). The suppressive effects of sumatriptan pre-incubation on IM-induced changes in rheobase and the response to suprathreshold current injection were blocked by the presence of BRL 15572 during the 30 min pre-incubation (n = 5, Fig 3A and C).

Sumatriptan Modulates Active and Passive Electrophysiological Properties

To begin to determine the basis for the sumatriptan-induced decrease in dural afferent excitability as well as the inhibition of IM-induced sensitization, changes in passive and active electrophysiological properties were examined. Thirty minute sumatriptan pre-incubation had no effect on baseline passive electrophysiological properties as assessed by the resting membrane potential and input resistance as these values, −69.0 ± 1.7 mV and 635 ± 146 MΩ (n = 8), were comparable to values previously reported (i.e., −71.3 ± 1.6 mV and 473 ± 57.3 MΩ (6)).

We previously demonstrated that IM produce significant changes in passive and active electrophysiological properties of dural afferents (6). These IM-induced changes included a ~10 mV membrane depolarization that was accompanied by a decrease in R_in subsequent to activation of I_IM-Cl (6). While sumatriptan pre-incubation did not prevent the IM-induced decrease in R_in, it blocked the IM-induced membrane depolarization as the IM-induced depolarization in neurons pre-incubated with the combination of sumatriptan and BRL 15572 (n = 5) was significantly (p < 0.01) larger than that in neurons pre-incubated with sumatriptan alone (n = 8, Table 1). We also previously demonstrated an IM-induced increase the AP overshoot subsequent to modulation of voltage gated Na⁺ currents (VGSC) in dural afferents (6). Sumatriptan pre-incubation did not prevent the IM-induced increase in AP overshoot, either, suggesting that sumatriptan did not prevent IM modulation of VGSC.

Table 1.

Sumatriptan blocks IM-induced changes in passive and active electrophysiological properties

Group	N	Δ Em (mV)	% Rin (MΩ)	Δ AP Duration (ms)	Δ AP overshoot (mV)	Δ AHP Magnitude (mV)	Δ τ AHP (ms)
30 min Suma + IM	8	1.0 ± 0.9	41.2 ± 14.6	0.38 ± 0.4	2.9 ± 5.8	2.3 ± 1.6	17.9 ± 26.5
30 min Suma + Antag + IM	5	7.5 ± 1.7	42.3 ± 9.0	0.34 ± 0.5	2.6 ± 1.4	3.1 ± 1.6	24.5 ± 11.4

Neurons were pre-incubated with sumatriptan (Suma, 1 μM) for 30 minutes alone or with the 5-HT_1D receptor antagonist (Antag) BRL 15572 (1 μM) prior to electrophysiological analysis. The IM-induced change in resting membrane potential (Δ E_m) calculated as the difference between E_m after IM and E_m before IM is significantly (p < 0.01, Student's t test) greater in the antagonist group. All other IM-induced changes were comparable between the two groups. N is the number of neurons studied in each group. R_in is input resistance. AP Duration is the duration of the action potential at 0 mV. AP overshoot is the amplitude of the action potential over 0 mV. AHP magnitude is the magnitude of the afterhyperpolarization following the AP relative to E_m. τ AHP is the time constant of decay of the AHP.

Sumatriptan does not Prevent IM-induced Activation of I_IM-Cl

While data from the AP waveform suggest that VGSC are probably not a convergent target of sumatriptan, the observation that sumatriptan was able to block the IM-induced depolarization suggests that these drugs may block IM-induced activation of I_IM-Cl. To test this possibility, IM-induced changes in I_IM-Cl were monitored in dural afferents (n = 7) with a protocol in which I_IM-Cl was evoked with 100 ms test pulses from −70mV to +50mV following a 40ms pre-pulse to 0mV to evoke Ca²⁺ currents (Fig 4A). The currents reversed at −30mV close to the predicted reversal potential for Cl⁻ (−34 mV) based on the composition of our intracellular and extracellular solutions. Pre-incubation with sumatriptan had no detectable influence on the peak density or rectification of I_IM-Cl (Fig 4A).

Figure 4. Sumatriptan does not prevent IM-induced activation of I_IM-Cl.

I_IM-Cl was activated by IM application and elicited with 100 ms test pulses from −70mV to +50mV following a 40ms pre-pulse to 0mV to evoke Ca²⁺ currents (n=7) and isolated as the difference between current evoked before and after application of IM (I_IM-Cl Difference Current). A. Pre-incubation with sumatriptan had no significant (p > 0.05, two-way repeated measures ANOVA) influence on peak I_IM-Cl density (at any voltage tested). B. To determine if sumatriptan may change the sensitivity of I_IM-Cl to high intracellular Ca²⁺, I_IM-Cl was recorded in the presence of Cd²⁺ and low intracellular EGTA to buffer intracellular Ca²⁺ at 622nM (n=5). Sumatriptan had no significant (p > 0.05, two-way repeated measures ANOVA) influence on the amplitude of I_IM-Cl at any potential under these conditions. Currents were blocked with 100μM niflumic acid (NFA). C. The reversal potential for Cl- was recorded in response to a ramp voltage protocol from +50mV to −100mV using the gramicidin perforated patch configuration (n=5). Sumatriptan pre-incubation had no significant (p > 0.05, Student's t test) influence on the reversal potential of the IM-induced current.

Given the influence of both sumatriptan and IM on VGCC in dural afferents (7), to rule out a potential interaction between changes in Ca²⁺ influx and I_IM-Cl activation, this experiment was repeated in the presence of Cd²⁺ to block VGCC and low intracellular EGTA (1.2mM) to buffer intracellular Ca²⁺ at a high concentration (622nM). I_IM-Cl was again recorded with test pulses from −70mV to +50mV. Sumatriptan did not produce any change in the amplitude of I_IM-Cl (n=5) at any potential under these conditions (Fig 4B). Furthermore, I_IM-Cl recorded in the presence of sumatriptan was blocked by Cl⁻ channel blocker niflumic acid (100μM, Fig 4B) as previously demonstrated (7).

Because the excitatory influence of IM-induced activation of I_IM-Cl on dural afferents appears to reflect a depolarized Cl⁻ equilibrium potential (E_Cl) in these neurons (7), we also examined the effects of sumatriptan on the reversal potential of I_IM-Cl. Cl⁻ currents were recorded in response to a ramp voltage protocol from +50mV to −100mV using gramicidin perforated patch to prevent dialysis of intracellular Cl⁻ (23). Sumatriptan pre-incubation (n=5) did not shift the reversal potential of I_IM-Cl (Fig 4C).

Sumatriptan Modulates K⁺ Currents and Inhibits IM-induced Suppression of K⁺ Currents

The decrease in R_in observed in the absence of an IM-induced depolarization of E_m suggests that pre-incubation with sumatriptan may result in the activation of a K⁺ current (I_K) that counters the depolarization driven by the activation of I_IM-Cl. To test this possibility, I_K was evoked with voltage protocols described in Methods, in the absence (n=7) and presence (n=6) of 30 min pre-incubation with sumatriptan and IM (Fig 5A). From these data, changes in the voltage dependence of activation and maximal conductance (G_max) were determined. Consistent with the decrease in R_in in the absence of E_m depolarization, 30 min sumatriptan pre-incubation resulted in a dramatic leftward shift in the voltage dependence of I_K activation. There was a significant (p < 0.01) hyperpolarization of the V_0.05 of activation (Fig 5B) following sumatriptan pre-incubation (−27.3 ± 4.7mV) as compared to control (−11.5± 2.4mV). IM application alone produced no change in the voltage dependence of activation of I_K (Fig 5B). However, IM (n=7) significantly reduced the maximal conductance (Fig 5C). This effect was completely blocked by sumatriptan (n=8) pre-incubation (Fig 5C).

Figure 5. Sumatriptan both modulates K⁺ currents and blocks IM-induced suppression of K⁺ currents.

A. The voltage-dependence of K⁺ current activation was determined with current-voltage protocols consisting of 10mV, 500ms voltage-steps between −60 and +60mV following a 500ms pre-pulse to −120mV. B. 30 minutes of sumatriptan (Suma) pre-incubation (n=6) resulted in a significant (p < 0.01, Student's t test) left shift in the voltage dependence of K⁺ current activation compared to vehicle (V) control (n=7): the V0.5 of current activation was shifted from −11.5± 2.4mV to −27.3 ± 4.7mV. IM application produced no significant (p > 0.05, one way ANOVA with Holm-Sidak post-hoc) change in the voltage dependence of K⁺ current activation in the presence or absence of sumatriptan. C. IM resulted in a significant (p > 0.05, one-way ANOVA with Holm-Sidak post-hoc test) reduction maximal K⁺ conductance (normalized by membrane capacitance), compared to vehicle treated neurons. However, there was no significant influence of IM on the maximal K⁺ conductance when applied to neurons pre-incubated with sumatriptan. Inset: When analyzed as a change from baseline, the IM-induced decrease in K⁺ conductance observed in vehicle treated neurons (V) was significantly greater (p < 0.05, Student's t test) than that observed in neurons pre-incubated with sumatriptan (Suma).

Because the shift in I_K activation should have attenuated the initial sumatriptan-induced sensitization of dural afferents, our current clamp results suggested that this shift takes time to develop. To begin to test this suggestion, we recorded I_K in dural afferents before and after the application of sumatriptan. Results of this analysis confirmed that this shift takes time to develop as the change in the V_0.05 of activation was −2.5 ± 1.8 mV after 5 minutes of incubation and −3.7 ± 3.7 mV after 10 minutes (n = 3).

DISCUSSION

The purpose of this study was to identify the ionic mechanism(s) underlying the actions of sumatriptan on dural afferents. Our results indicate that 1) acute sumatriptan application produces an increase in baseline dural afferent excitability that is blocked by the 5-HT_1D receptor antagonist BRL 15572. No further increase in excitability was observed following subsequent application of IM, 2) 30 min sumatriptan pre-incubation has no detectable influence on dural afferent excitability but attenuates IM induced-sensitization in a 5-HT_1D receptor antagonist-dependent manner, 3) while sumatriptan produced an expected inhibition of VGCC, pre-incubation with sumatriptan did not attenuate the IM-induced decrease in AP threshold or action potential overshoot (changes that appear to depend on an increase in TTX-R Na⁺ currents(7)) or I_IM-Cl, and 4) Sumatriptan both increases K⁺ currents in dural afferents via a leftward shift in the voltage-dependence of activation, and attenuates, IM-induced suppression of total K⁺ current.

Sumatriptan Mediated Inhibition of VGCC

Our data demonstrate that sumatriptan concentration-dependently inhibits VGCC in dural afferents. One of the most dramatic mechanisms of G-protein coupled receptor mediated inhibition of VGCC involves a rapid displacement of the VGCC β subunit via the G-protein βγ subunit (19). A unique feature of this form of inhibition is that it can be overcome with a strong depolarizing pre-pulse (20). However, following a depolarizing pre-pulse to +80mV, Ca²⁺ currents did not recover from sumatriptan inhibition. Furthermore, a lack of depolarizing shift in the voltage dependence of activation with increasing concentrations of sumatriptan suggests that this inhibition is via an as of yet unidentified intracellular second messenger. These conclusions are consistent with previous results from a study of Xenopus larvae spinal neurons indicating that 5-HT_1B/1D receptor agonists (L694 247) reduce high voltage activated N and P/Q type currents by a G-protein activated diffusible second messenger pathway (18). More relevantly, these data are also consistent with previous reports that zolmitriptan can block P/Q and possibly R type currents in dissociated TG neurons. This effect was pertussis toxin sensitive indicating the activation of Gi/Go class of G proteins (17). Such a mechanism was recently suggested to account for the sumatriptan-induced suppression of capsaicin-evoked currents in dural afferents (24).

Acute Sumatriptan Application Increases Dural Afferent Excitability

Acute application of sumatriptan produced an increase in dural afferent excitability. These data may explain the clinical observation that triptans transiently aggravate headache. Within 5–15 min of taking sumatriptan, approximately 50% of patients experience exacerbated pain that lasts for about 10 – 15 min (25) before the onset of pain relief. Our data are also consistent with previous observations that sumatriptan can drive a Ca²⁺ dependent discharge (26), an increase the firing rate of C and Aδ meningeal nociceptors, and increase their mechanical sensitivity (25).

Multiple mechanisms are likely involved in this sumatriptan-induced transient increase in dural afferent excitability. However, the only change in active or passive electrophysiological properties observed in this study following acute sumatriptan was a significant depolarization in the membrane potential from −71.3mV to −54.0mV. That the depolarization was not accompanied by a significant change in R_in suggests that there was no net change in the number of open channels, only a shift in the proportion of the various types of channels that were open.

Pre-Incubation with Sumatriptan has no Influence on Dural Afferent Excitability

There was no significant influence of pre-incubation with sumatriptan on baseline dural afferent excitability. This is somewhat surprising, in retrospect, given the dramatic leftward shift in the activation of I_K. Multiple K⁺ currents are expressed in sensory neurons and are critically involved in regulating their excitability (27). K⁺ channels regulate the timing between APs and therefore impact AP frequency. Thus, the dramatic shift in the voltage dependence of activation of K⁺ currents should have resulted in an increase in rheobase and/or a decrease in the response to suprathreshold current injection. The failure to detect such changes in excitability suggests that the shift in the voltage-dependence of K⁺ current activation is compensated, at least in part, by excitatory changes that persist following acute application of summatriptan. One such mechanism would include a suppression of Ca²⁺ dependent K⁺ channels secondary to the sumatriptan-induced inhibition of VGCC. We have recently demonstrated that such a channel is present in a subpopulation of cutaneous neurons where it plays a significant role in the regulation of afferent excitability (28) and appears to be tightly coupled to the Ca²⁺ influx via VGCC (29). As we have also demonstrated that a Ca²⁺ dependent K⁺ channel is present in dural afferents and suppressed following IM application (7), sumatriptan-induced suppression of such a current could also account for the apparent block of the IM-induced suppression of total K⁺ current following sumatriptan pre-incubation. Such an explanation would suggest that the shift in the voltage-dependence of K⁺ current activation is associated with an increase in K⁺ channel density. That is, an increase in one K⁺ channel type associated with the shift in the voltage-dependence of activation would compensate for a decrease in Ca²⁺-dependent K⁺ channel activity resulting in the observed no net change in peak K⁺ conductance. K⁺ channel subunits present in sensory neurons that could undergo such dramatic shifts in the voltage-dependence of activation include Kv2.1 (30), which can undergo a ~26 mV hyperpolarizing shift in the G-V following Ca²⁺/calcineurin dependent dephosphorylation (31). Future studies will be needed to identify the K⁺ channel subunit(s) that underlie the actions of sumatriptan in dural afferents.

Sumatriptan Selectivity

The observation that both the sumatriptan-induced acute sensitization and the subsequent inhibition of IM-induced sensitization of dural afferents were blocked by BRL 15572 indicates that both processes are mediated by the 5-HT_1D receptor. This is consistent with previous data suggesting that while both 5-HT_1B and 5-HT_1D receptors are present on trigeminal ganglion neurons (32, 33), the vasoconstrictive effects of triptans are due to the 5-HT_1B receptors on the dural vasculature (34), while the selective therapeutic efficacy of triptans for migraine is due to 5-HT_1D receptors in dural afferents (35). However, evidence that the 5-HT_1D receptor is present on subpopulations of afferents throughout the body (9) and that triptans have analgesic efficacy in other preclinical pain models (12, 13), still begs the question as to the basis of the selective clinical profile of this class of drugs. Our recent observation that the higher density of the 5-HT_1DR in nerve fibers preferentially involved in signaling migraine pain may partially explain the selectivity of these drugs (8). However, in light of the fact that a receptor for these drugs is present in other afferent populations, albeit at lower densities, we proposed that other mechanisms likely contribute to efficacy and selectivity. Given evidence that IM-induced activation of I_IM-Cl appears to be a relatively unique mechanism underlying the sensitization of dural afferents, sumatriptan-induced inhibition of I_IM-Cl would provide another mechanism to account for the therapeutic selectivity of this compound. The observations that sumatriptan neither blocked the activation of I_IM-Cl, nor shifted the equilibrium potential for Cl⁻ indicates that this channel cannot account for the therapeutic actions of triptans. However, the modulation of I_K could account for the therapeutic selectivity of this compound if data from subsequent studies confirm that this modulation is only observed in dural afferents.

The complex actions of sumatriptan on dural afferents raise at least 3 questions. One question is how could a decrease in VGCC contribute to the antinociceptive efficacy of triptans at the same time triptans have increased excitability of dural afferents. VGCCs are largely responsible for the influx of Ca²⁺ necessary to enable transmitter release from pre-synaptic terminals. The suppression of VGCC on the central terminals of dural afferents should contribute to the antinociceptive efficacy of triptans and account for the normalization dural stimulation-induced activity in trigeminal dorsal horn neurons following IM-induced sensitization (36). While there is evidence that low threshold or T-type VGCC may contribute to afferent sensitization (37), the high threshold channels described in the present study that mediate transmitter release have a minimal direct contribution to action potential generation (38). However, Ca²⁺ influx through these channels may contribute to the activation and/or modulation of a number of channels including 2-pore K⁺ channels (39) and Ca²⁺ modulated K⁺ channels (28). Thus, as noted above, while the triptan-induced suppression of VGCC in dural afferents may occur in parallel with the increase in excitability, the two may be causally linked. A second question pertains to the differential time course of the sumatriptan-induced excitation and inhibition of IM-induced sensitization, particularly if both processes are mediated by the same receptor. While pharmacokinetics could explain the relatively slow onset of triptan-induced pain relief observed clinically, the present results suggest an alternative possibility: distinct cellular processes underlie excitatory and inhibitory actions of the drug where those underlying inhibition develop far more slowly than those underlying excitation. Additional work will be needed to tease apart the specific mechanisms underlying the actions of sumatriptan in dural afferents, but the literature is now full of examples of receptor mediated processes, in particular those like the 5-HT_1D receptor that are coupled to G-proteins, that develop over very different time scales. For example, the membrane delimited form of G-protein-mediated suppression of VGCC can occur within tens of milliseconds in sensory neurons (40), while there is evidence that metabotropic glutamate receptor-mediated decrease in membrane ionotropic glutamate receptors develops over 10s of minutes (41). A third question is why triptans fail alleviate migraine pain once it is already established. Our results indicating the pre-incubation with sumatriptan blocked IM-induced sensitization are consistent with the evidence that triptans administered prior to the development of migraine pain can abort a migraine. However, the leftward shift in the activation of K⁺ currents should enable sumatriptan to reverse afferent sensitization, even after it is established. The observation that triptans fail to reverse IM-induced sensitization of dural afferents (36) suggests that the second messenger pathways activated by IM block the actions of sumatriptan, at least those underlying the modulation of K⁺ currents. Ongoing experiments are designed to identify the point(s) of convergence of the underlying second messenger pathways.

SUMMARY

We have described both excitatory and inhibitory effects of sumatriptan that follow a time course that may explain why some migraineurs experience increases in pain sensitivity before the onset of pain relief. Additional work is needed to identify the ionic mechanisms underlying the excitatory effects of sumatriptan, as the ability to block these effects may ultimately increase the efficacy of these compounds. We have ruled out two important targets for the therapeutic actions of sumatriptan, TTX-R I_Na and I_IM-Cl. The implication of this observation is that there is a balance between the excitatory actions of inflammatory mediators and the inhibitory actions of triptans which appear to be acting on different targets. Relatively more excitation and/or less inhibition in a subpopulation of patients would result in a population unresponsive to triptans. Differences in the relative balance between excitation and inhibition may suggest an explanation for why triptans are only effective in ~70% of migraineurs (3) (although those pain free at 2 hrs may be considerably lower (42)). More importantly, in addition to voltage-gated Ca²⁺ channels previously identified by others, we have identified a novel target that may account of the therapeutic actions of triptans. Maximizing the hyperpolarizing shift in I_K may provide a novel approach for the treatment of migraine.