Sage Journals: Discover world-class research

Abstract

Background

Recently, the adenosine triphosphate (ATP) sensitive potassium channel opener levcromakalim was shown to induce migraine attacks with a far higher incidence than any previous provoking agent such as calcitonin gene-related peptide. Here, we show efficacy of ATP sensitive potassium channel inhibitors in two validated rodent models of migraine.

Methods

In female spontaneous trigeminal allodynic rats, the sensitivity of the frontal region of the head was tested by an electronic von Frey filament device. In mice, cutaneous hypersensitivity was induced by repeated glyceryl trinitrate or levcromakalim injections over nine days, as measured with von Frey filaments in the hindpaw. Release of calcitonin gene-related peptide from dura mater and trigeminal ganglion was studied ex vivo.

Results

The ATP sensitive potassium channel inhibitor glibenclamide attenuated the spontaneous cephalic hypersensitivity in spontaneous trigeminal allodynic rats and glyceryl trinitrate-induced hypersensitivity of the hindpaw in mice. It also inhibited CGRP release from dura mater and the trigeminal ganglion isolated from spontaneous trigeminal allodynic rats. The hypersensitivity was also diminished by the structurally different ATP sensitive potassium channel inhibitor gliquidone. Mice injected with the ATP sensitive potassium channel opener levcromakalim developed a progressive hypersensitivity that was completely blocked by glibenclamide, confirming target engagement.

Conclusion

The results suggest that ATP sensitive potassium channel inhibitors could be novel and highly effective drugs in the treatment of migraine.

Keywords

Migraine headache trigeminal pain potassium channels calcitonin gene-related peptide

Abbreviations:

ATP: adenosine triphosphate; cAMP: cyclic adenosine monophosphate; cGMP: cyclic guanosine monophosphate; CGRP: calcitonin gene-related peptide; DMSO: dimethyl sulfoxide; EIA: enzyme immunoassay; GTN: glyceryl trinitrate; KATP: ATP-sensitive potassium; KIR: inwardly-rectifying potassium channels; NO: nitric oxide; PACAP: pituitary adenylate cyclase activating peptide; PKA: protein kinase A; PKG: protein kinase G; sAC: soluble adenylyl cyclase; SD: Sprague Dawley; sGC: soluble guanylyl cyclase; SIF: synthetic interstitial fluid; STA: spontaneous trigeminal allodynia; SUR: sulfonylurea receptor; TG: trigeminal ganglia

Introduction

Migraine affects one billion people worldwide. As a cause of disability, it is number two surpassed only by low back pain according to the Global Burden of Disease study (1). It has an estimated economic cost of €50 – €111 billion per year in Europe (2).

Treatment possibilities have expanded, most recently with the advent of antibodies directed against calcitonin gene-related peptide (CGRP) or its receptor. Small molecule CGRP receptor antagonists recently reached the market. However, none of these are effective in more than 50–60% of patients (3). Thus, the unmet need for better migraine treatments is still huge.

Potassium channels are found in most living organisms, where they serve a variety of crucial functions. Within the trigeminovascular system, a number of K+ channel subtypes such as two-pore potassium channels (K2Ps), big calcium activated potassium channels (BK_Ca) and ATP-sensitive potassium (K_ATP) channels have been implicated in the aetiology of migraine (4 –7). K_ATP channels are expressed in various tissues and serve to couple cellular metabolism, via adenine nucleotides to membrane potential and excitability (8 –11). This enables K_ATP channels to regulate a diverse range of physiological processes including maintenance of blood glucose homeostasis, control of excitation-contraction coupling in muscular tissue, and regulation of vascular tone (12 –14).

With the use of a human migraine model, it was shown that the K_ATP channel opener levcromakalim appears to be a potent migraine trigger (15). The use of this model has proven beneficial in identifying signaling molecules, such as nitric oxide (NO) (16 –18), pituitary adenylate cyclase activating peptide (PACAP) (19,20) and CGRP (21), with a role in migraine pathophysiology (22). Thus, migraine provocation by K_ATP channel opening immediately suggests the possibility that inhibitors of K_ATP channels may be effective in migraine. However, that a naturally occurring substance or mechanism can induce migraine does not necessarily mean that their reciprocal modulators are effective in the treatment of migraine attacks (23,24).

There are no human studies of K_ATP channel inhibitors in relation to migraine treatment. Therefore, it is pertinent to obtain as good evidence of efficacy in animal models as possible before recommending industry to pursue K_ATP channel inhibition as a drug target. Here, we have used two validated rodent models of migraine. One model is a genetically inbred strain, the spontaneous trigeminal allodynic (STA) rat, which exhibits tactile hypersensitivity in the head but not in the hindpaws (25,26). The other model uses repeated injection of glyceryl trinitrate (GTN) to provoke a general state of hyperalgesia in mice (27,28). Both animal models respond to migraine-specific drugs including the 5-HT_1B/D receptor agonist sumatriptan, the CGRP receptor antagonist olcegepant and a CGRP antibody (25 –28). In addition, a validated ex vivo model of capsaicin-induced CGRP release from migraine-relevant tissue was used (29,30). Our results show that K_ATP channel inhibition is a relevant future target for novel migraine treatments.

Materials and methods

Animals

Female STA rats (body weight 216–255 g; age 19 weeks, n = 24) were derived from a single breeding pair originally sourced from the Thomas Jefferson University under license (26) and subsequently bred in house. Periorbital thresholds for this breeding pair, measured via manual von Frey filament testing, have been described previously (25). For experiments with control rats we used wild type Sprague-Dawley rats (body weight 196–280 g; age 8–13 weeks, n = 22) obtained from Charles River (Sulzfeld, Germany). All rats were group housed in Tecniplast 1354G Eurostandard type IV polycarbonate cages (L × W ×H: 60 × 38 × 20 cm; Brogaarden, Denmark) using a 12-h light/dark cycle with lights on at 4 am. Individual opaque red polycarbonate shelters (20 ×11.5 × 16 and 15 × 9 × 9 cm respectively), together with an aspen biting stick (10 × 2 × 2 cm; Tapvei, Estonia) and piece of hemp rope suspended from the cage lid were provided in each homecage for retreat and enrichment purposes. Bedding consisted of Enviro-Dri nesting material (Brogaarden, Denmark). Standard rat chow (Altromin) and tap water were available ad libitum in the animals’ home cage environment. Humidity ranged from 45–65%. All in vivo experiments were performed in a laboratory in the animal housing facilities between 8 am and 4 pm.

When cross-over studies were performed, at least 3–4 days were allowed for drug washout before rats were allocated to a different treatment. Importantly, low periorbital threshold responses at baseline was confirmed prior to treatment. Experiments were approved by the Danish Animal Experiments Inspectorate and performed in accordance with the relevant guidelines and regulations of approval number 2014-15-0201-00256.

Male C57Bl/6J mice (Taconic, Denmark) of 8–9 weeks (22–31 g) at start of experiment were used following 7 days acclimatisation to the animal facility. In total, 107 mice were used to complete the study. Mice were housed in IVC type II cages (Tecniplast) in groups of six with ad libitum access to food (Altromin 1314, Altromin GmbH & Co., Seelenkamp, Germany) and tap water. The cages were floored with sawdust (Aspen chip, Tapvei, Finland) and enriched with a red-tinted polycarbonate shelter (Molytex, Denmark), nesting material (Happi-Mats, Scanbur, Denmark) and biting sticks (Tapvei, Finland). The light cycle was 12/12, light on at 6 am. All handling of animals was done between 7.30 am and 4 pm in a dedicated procedure room. The experiments were approved by the Danish Animal Experiments Inspectorate and performed in accordance with the relevant guidelines and regulations of approval number 2017-15-0201-01358.

In both species, one experimental unit represents one animal.

Measurement of glucose levels

The rat was gently restrained by one investigator, while another investigator used a 1 ml plastic insulin syringe (29 gauge) to obtain a venous blood sample (approximately 50 µl) from the dorsal aspect of the hindpaw. A new syringe was used for each blood sample. Blood glucose was measured immediately according to the manufacturer’s instructions using the Accu-Chek, Aviva kit (Roche Diabetes Care, Mannheim, Germany). Blood glucose was either measured in a separate group of rats or at the end of the experiment once all behavioral testing was completed. Accordingly, there would have been no effect of the blood sampling protocol on the sensory threshold responses.

Periorbital pain thresholds in STA rats

The sensitivity of the frontal region of the head (V1 ophthalmic trigeminal dermatome), to static mechanical stimulation was measured using an electronic von Frey device fitted with a rigid plastic tip (IITC LifeScience Inc, USA). The rat was gently restrained using a cotton towel and held in a prone position in the lap of the investigator with its head and neck region left exposed and unrestricted. The tip of the device was then applied with increasing force (maximum of 450 g) to the right periorbital area above the eyes until the rat withdrew its head, laterally rotated its head and/or vocalized. The soft tissue around the eye was carefully avoided. The process was then repeated over the midline, the left and then the right periorbital areas above the eye. The average of these latter three measurements was considered the withdrawal threshold (g). If a stimulus had to be reapplied due to inappropriate application, care was taken not to apply the probe tip to the exact same location within the region being measured. The whole procedure typically took less than 2 minutes to perform.

Hindpaw pain thresholds in STA rats

Hindpaw sensitivity to mechanical stimulation was assessed using an electronic Randall Selitto paw pressure device (IITC LifeScience Inc, USA) as described previously (25). Briefly, the rat was gently restrained within the hand and lower arm of the investigator and the right hindleg extended and the hindpaw inverted. The probe tip was then placed on the paw between the foot pads and increasing force uniformly applied (maximum of 450 g) until the rat withdrew the hindpaw and/or vocalized. The average of three measurements obtained from defined parts of the hindpaw excluding the footpads was considered the paw pressure withdrawal threshold (g).

To assess hindpaw thresholds to noxious thermal stimulation, we used a hot/cold plate (IITC LifeScience Inc, USA) with a pre-set plate temperature of 48°C. The latency (s) to respond to the thermal stimulus which consisted of either licking, rapid shaking, or stepping of the hindpaws, was recorded from the moment the rat was placed onto the surface of the plate. A cut-off time was set at 60 sec. No rats showed signs of thermally-induced damage to the paws throughout the duration of the study.

Assessment of motor function in rats

To assess if motor function was putatively impaired by drug treatment, for each rat a simple observational analysis was made according to methods modified from Zahn and co-workers (1997) and Hao & Xu (1996) (31 –33). The rat was placed onto a clean tabletop and hindlimb gait scored as (2, normal; 1, moderate hindlimb weakness/splaying; 0, complete hindlimb weakness/splaying). The ability to walk was then scored as (2, normal; 1, walks with mild deficit; 0, no attempt to walk). Next, the dorsum of the right hindpaw was drawn across the edge of the table, a stimulus which elicits a lifting of the paw onto the surface of the table and scored as (2, normal; delay of 1–2 s, 1; 0, delay >2 s). Finally, the whole-body righting reflex was assessed by placing the rat horizontally with its back on the table which normally gives rise to an immediate, coordinated twisting of the body to an upright position (2, normal; delay of 1–2 s, 1; 0, delay >2 s). Each test was performed once, and the complete battery of tests took less than 1 minute to perform for each rat. The total score was summed for each rat and statistical comparisons between treatment groups made using a non-parametric ANOVA.

Induced hindpaw hyperalgesia in mice

Sensitivity towards mechanical stimulation to a series of von Frey filaments (Ugo Basile, Italy) was evaluated at the plantar surface of the left hindpaw using the up-down test paradigm (34). Analysis of the data was performed with an improved algorithm for calculation of 50% thresholds based on exact between-filament intervals (35). Filaments were applied until withdrawal of the foot or up to 3 sec in case of non-withdrawal. Rapid removal of the paw during or in relation to the end of stimulation was considered a positive response. For testing, animals were kept in individual clear plexiglass cages (L × W × H: 10 × 10 × 12 cm) with a mesh floor (IITC Life Science, US) (27). Measures were taken before and after drug administrations.

GTN and levcromakalim were administered every other day for a total of five times. Using this paradigm, GTN mice present hindpaw and cephalic hyperalgesia that becomes more severe with every injection (27,28). We tested if glibenclamide could prevent the development of GTN-induced hyperalgesia by injecting glibenclamide 15 min prior to GTN on each of the five test days. The same protocol was applied to investigate if levcromakalim, like GTN, could induce hyperalgesia. First, we did a small experiment for dose-finding purposes and in a subsequent experiment the effect of levcromakalim was antagonised by prior treatment with glibenclamide.

All pharmacology experiments were performed with the investigator blinded to treatment, with rats and mice randomly allocated to vehicle or treatment groups.

CGRP release ex vivo

Female STA (215–318 g, n = 16) rats were used for ex vivo experiments. Origin of STA rats, housing and approval numbers by the Danish Animal Experiments Inspectorate are the same as described above. On the day of experiment, the rats were deeply anaesthetised by i.p. injection of a mixture of pentobarbital (90 mg/kg) and lidocaine (9 mg/kg) followed by decapitation and dissection of tissues as previously described in detail (29,36). Briefly, the skull was cut mid-sagittally and the brain halves were carefully removed while the dura mater was left attached to the skull. Then the skull halves (with intact dura mater and without the trigeminal ganglion (TG)) and the carefully dissected TG were immersed in 25 ml synthetic interstitial fluid (SIF) at room temperature and repeatedly washed (every 5 min) for 30 minutes with SIF (in mM: 108 NaCl, 3.5 KCl, 3.5 MgSO₄, 26 NaHCO₃, 11.7 NaH₂PO₄, 1.5 CaCl₂, 9.6 Na+ gluconate, 5.6 glucose and 7.6 sucrose).

The two TGs from each rat were individually placed in 350 µl SIF in the cap of an Eppendorf tube and the two skull halves were placed on separate platforms. The tissues were then placed in a humid chamber at 37ºC in an incubator. They were washed with SIF five times without touching the tissue. The basal CGRP release was determined after 10 minutes of incubation in SIF after the fifth wash.

The experiments were performed according to two different protocols, as shown in Figure 1. In the first protocol, sampling was performed after each 10 min incubation with increasing concentrations of capsaicin (10 nM to 10 µM) or its vehicle. In the second protocol, the tissues were incubated for 10 min with vehicle or 3 µM glibenclamide; after sampling, the tissues were incubated with a mixture of 1 µM capsaicin and 3 µM glibenclamide. After sampling of each concentration, the residual SIF was discarded and new SIF containing the next concentration of capsaicin or vehicle was added.

Figure 1.

Schematic overview of experimental protocol. The upper panel shows the experimental protocol for concentration-response curves of CGRP release after administration of capsaicin in increasing concentrations. The lower panel represents the experimental protocol for studies investigating the effect of glibenclamide on capsaicin-induced CGRP release. The hemisected skull with dura and TG from one side served as control to the treatment in the skull and TG from the other side.

For measurement of CGRP, samples were processed with commercial EIA (Enzyme–linked immunoassay) kits (SPIbio, Paris, France). The antibody in the CGRP-EIA kit is directed against human-CGRP but has 120% cross-reactivity to rat CGRP. The CGRP detection level is about 0.7 pg/ml. Previous studies have shown that the EIA assay is strictly specific for the entire sequence of CGRP, making no cross reactivity with structurally matched peptides; that is, calcitonin, amylin or with CGRP degradation fragments. We have matched the signal by adding the known concentration of synthetic α-CGRP in this assay. The protocol provided with the kit was followed in detail. The optical density was measured at 410 nm using a micro plate photometer (Tecan, infinite M200, software SW Magellan v.6.3).

Drugs

For use in rats, buprenorphine (Indivior UK, Temgesic 0.3 mg/ml) was diluted to 0.1 mg/ml in saline and administered s.c. 0.1 mg/kg. Glibenclamide (Tocris/Bio-Techne, UK) was dissolved to 0.1, 1 and 10 mg/ml in vehicle (20% dimethyl sulfoxide (DMSO)-50% alcohol (96%)-30% saline) and administered i.p. at 1 ml/kg. Suspensions of gliquidone (Sigma-Aldrich Denmark A/S, Denmark) were made to 0.5, 5 and 50 mg/ml in milli-Q water with 25% (w/v) 2-hydroxypropyl-beta-cyclodextrin and subsequently administered p.o. at 2 ml/kg.

For use in mice, glibenclamide (Tocris/Bio-Techne, UK) was diluted to 0.1 mg/ml in a 5% DMSO solution and administered at 1 mg/kg. Levcromakalim (Tocris/Bio-Techne, UK) was dissolved in DMSO and diluted in saline to the desired concentration for injection of 0.1, 0.5 and 1 mg/kg in a volume of 10 ml/kg resulting in a maximum concentration of DMSO of 2%. GTN 7.89 mg/ml in ethanol (Capital Region, Hospital Pharmacy) was diluted in saline to 1 mg/ml (12.8% ethanol) and administered at 10 mg/kg. All drugs were administered i.p. to mice in a volume of 10 ml/kg. For all drugs, the vehicle was appropriate to the injected drug.

For the ex vivo assay, capsaicin and glibenclamide were purchased from Tocris (Bio-Techne Ltd., Abingdon, UK). The stock of capsaicin was prepared in ethanol to 10 mM and was further diluted in SIF for working solutions. Glibenclamide was prepared in ethanol to a master stock of 1 mM and was further diluted in SIF to 3 µM, as used in the experiments.

Statistical analyses

All statistical analyses were performed using GraphPad Prism 8 (Graph Pad Software Inc., San Diego, CA). Group sizes for the pharmacology experiments on cutaneous threshold values were estimated as a function of the desired effect size (approximately 50% change vs. corresponding vehicle treatment) where we assumed a significance level of 5% and a power of 90% (37). All data are presented as mean ± SEM, p < 0.05 was considered statistically significant.

Rat behavioral experiments were performed as cross-over experiments where the individual rat received all treatments on different test days in a randomized manner. Repeated measure analysis of variance (ANOVA) or mixed effects model for data sets with missing values were used to analyze the overall effects of treatments followed by Bonferroni’s multiple comparisons test. For comparisons between only two groups/treatments, paired or unpaired t-tests were performed as appropriate. Mouse 50% withdrawal thresholds were square root transformed to obtain normally distributed data, and data were analyzed by repeated measure ANOVA followed by Bonferroni’s post hoc test for comparison between treatment groups. For in vitro assays, the release of CGRP (measured as immunoreactive CGRP in the samples) was expressed in pg/ml, given as mean values ± SEM. Absorbance was recorded, and values were calculated through an interpolation method using an equation derived from the standard curve. The value of the diluted samples was multiplied by 2. Paired data (control site and test site) were analyzed for overall effect of treatment by two-way ANOVA, one-way ANOVA or mixed effects analysis for data sets with missing values followed by Bonferroni’s multiple comparisons test.

Results

K_ATP channel inhibition in STA rats

The non-selective K_ATP channel inhibitor glibenclamide was first tested in STA (spontaneous trigeminal allodynia) rats and blood glucose levels measured to determine doses of glibenclamide that actively engage pancreatic K_ATP (e.g. Kir6.2/SUR1) channels. Accordingly, glibenclamide (0.1–10 mg/kg, i.p.) treatment lowered blood glucose levels (F(4,20) = 116.1, p < 0.0001, n = 4–6 per group). Note that we observed that the applied vehicle (20% DMSO-50% alcohol (96%)-30% saline) increased blood glucose levels compared with baseline plasma glucose levels in non-injected naïve rats (Figure 2(a)). The 2 h post-treatment analysis revealed that the 1 mg/kg dose of glibenclamide lowered blood glucose levels by 55% compared with vehicle treatment (p < 0.001) and this was the lowest active dose of glibenclamide.

Figure 2.

Effects of glibenclamide on blood glucose and sensory nociceptive thresholds. (a) Effects of glibenclamide (0.1–10 mg/kg, i.p.) on blood glucose levels (mmol/l) in four to six STA rats. (b) Cutaneous periorbital and hindpaw withdrawal thresholds (g) were increased after buprenorphine (0.1 mg/kg, s.c.; n = 12), but not after glibenclamide (1 mg/kg, i.p.; n = 10) in wild type Sprague Dawley (SD) rats. (c) Latency responses (s) to thermal hindpaw stimulation after buprenorphine 0.1 mg/kg (n = 12) and glibenclamide 1 mg/kg (n = 12) as compared to vehicle treatment in SD rats. (d) Periorbital baseline withdrawal threshold response to cutaneous mechanical stimulation in SD (n = 12) and STA rats (n = 12). Thereafter, glibenclamide (0.1–10 mg/kg, i.p.; n = 12–15) was administered and the effects on (e) periorbital and (f) hindpaw post-treatment thresholds were evaluated. Data represent mean ± SEM.

Next, we assessed the effects of glibenclamide on nociceptive thresholds in naïve Sprague-Dawley (SD) rats. Systemic administration of glibenclamide (1 mg/kg i.p.) had no effect on cephalic or hindpaw thresholds in response to cutaneous mechanical stimulation (p = 0.541 and p = 0.564 vs. vehicle, respectively, n = 10), (Figure 2(b)). In contrast, the partial µ-opioid receptor agonist buprenorphine increased both cephalic and hindpaw thresholds by 107% and 59% (p = 0.0001 and p = 0.001 vs. vehicle respectively, n = 12). Similarly, glibenclamide had no effect on thermal nociceptive thresholds in naïve rats whereas buprenorphine administration increased the latency response to thermal stimulation by 185% (p = 0.001 vs. vehicle, n = 12), (Figure 2(c)). Thus, glibenclamide has no effect on normal nociceptive transmission.

Figure 2(d) shows that periorbital thresholds to cutaneous mechanical stimulation were 51% lower in STA rats compared with control SD rats (p < 0.0001, n = 12), indicative of sensitization of trigeminal nociceptive signalling in STA rats. To better understand a potential role for K_ATP channels in pain transmission relevant to migraine, we administered glibenclamide to female STA rats. Glibenclamide (0.1–10 mg/kg, i.p.) increased the periorbital threshold to mechanical stimulation (treatment × time interaction, F(6, 57) = 8860, p < 0.0001) in STA rats (Figure 2(e)). Post-treatment analysis revealed that this effect was significant for both the 1 and 10 mg/kg doses of glibenclamide (p < 0.001 vs. vehicle, n = 12–15) at both the 1 and 4 h timepoints post-treatment. Notably, glibenclamide had no effect on the normosensitive hindpaw threshold to mechanical stimulation when comparing to vehicle treatment responses at the same time-points (treatment × time interaction, F(3.7, 35.2) = 1572, p = 0.2063, n = 12–15, thereby confirming specificity for trigeminal versus somatic pain circuits (Figure 2(f)).

Glibenclamide has been reported to be an agonist at human transient receptor potential ankyrin 1 channels in contrast to other sulphonylurea compounds (38). Thus, to help confirm that glibenclamide-mediated effects on cephalic threshold responses in STA rats were not simply compound specific, we assessed the effects of the structurally distinct K_ATP channel inhibitor gliquidone (39). A dose range finding pilot study on five rats showed a significant fall in glucose levels after treatment with 10 and 100 mg/kg p.o. of gliquidone of 40% and 63%, respectively. The lowest dose of 1 mg/kg only caused a non-significant 3% fall in glucose levels (p = 0.988). Therefore, we decided to continue with the two highest doses in the main experiment, in which 11 female STA rats were used. In these rats, gliquidone caused a significant (F(2,15) = 53.22, p < 0.0001) decrease in blood glucose levels of 43% versus vehicle in both the 10 mg/kg and 100 mg/kg groups (n = 11) (Figure 3(a)). Figure 3(b) illustrates that gliquidone (10–100 mg/kg, p.o.) caused a significant increase in periorbital threshold to mechanical stimulation when administered to STA rats (treatment × time interaction, F(2,23) = 3.77, p = 0.036). Gliquidone had no effect on hindpaw threshold to mechanical stimulation (Figure 3(c)), thereby confirming specificity for trigeminal versus somatic pain circuits. Finally, accumulated motor scores were tested in STA rats treated with glibenclamide (n = 12) and gliquidone (n = 12) as compared to vehicle responses in the same individual rats. In all STA rats and treatments, the accumulated motor score reached a maximum value of eight, indicating a normal motor function. This supports that the reversal of cutaneous hypersensitivity did not simply occur because of non-specific effects on efferent motor function.

Figure 3.

Cephalic hypersensitivity in STA rats is attenuated by gliquidone. (a) Effects of gliquidone (10–100 mg/kg, p.o.) on blood glucose levels (mmol/l) in STA rats. Cutaneous (b) periorbital and (c) hindpaw threshold responses (g) to mechanical stimulation 2 and 4 h after treatment with the selective K_ATP channel inhibitor gliquidone (10–100 mg/kg, p.o.; n = 11). Data represent mean ± S.E.M.

K_ATP channel inhibition in mice

As previously described, repeated administration of GTN (10 mg/kg, i.p.) produced hindpaw and cephalic allodynia in mice (27,28,40,41). Treatment with glibenclamide (1 mg/kg, i.p.) 15 min before GTN on every test day (1, 3, 5, 7 and 9) resulted in significantly higher 2 h post-drug withdrawal thresholds in response to GTN than in mice administered vehicle before GTN treatment (overall effect of treatment, F(2,32) = 41.38, p < 0.0001) (Figure 4(b)). The basal response, measured before injection of GTN on each test day, was significantly (overall effect of treatment, F(2,32) = 39.54, p < 0.0001) reduced from the third test day (Figure 4(a)). Glibenclamide had no significant (p >0.05) effect on the basal responses found after GTN treatment (Figure 4(a)).

Figure 4.

Hindpaw hyperalgesia in GTN-sensitized mice is attenuated by glibenclamide. Glibenclamide (1 mg/kg, i.p.) or vehicle (5% DMSO, 10 ml/kg, i.p.) was given 15 min prior to GTN (10 mg/kg, i.p.) on each test day (1, 3, 5, 7 and 9). The sensitivity to mechanical stimulation with von Frey filaments was measured as basal values before (a) and 2 h after GTN injection. (b) Data are square root transformed 50% withdrawal thresholds and depicted as mean ± SEM.

The 2 h post-drug response to the K_ATP channel opener levcromakalim dose-dependently (0.1–1 mg/kg, i.p.) induced hyperalgesia (Figure 5(b)). Whereas the acute hyperalgesia to GTN injection occurred after just a single injection, the effect of levcromakalim was not robust until the fifth test day (third dose). But, as observed for GTN-induced hyperalgesia, the response became stronger throughout the protocol (Figure 5(b)) (treatment effect F(3,20) = 9.71, p = 0.0004). The levcromakalim-induced hyperalgesia was most pronounced 2 h after administration (Figure 5(c)). The basal response to levcromakalim was significantly reduced (overall effect of treatment, F(3,20) = 2.82, p < 0.065) at the ninth test day in the two highest doses (0.5 (p = 0.025) and 1 mg/kg (p = 0.027)) (Figure 5(a)), but was reduced from the fifth day.

Figure 5.

Hyperalgesia mediated by the K_ATP channel opener levcromakalim in mice. On every test day (1, 3, 5, 7 and 9) sensitivity to mechanical stimulation with von Frey filaments was measured at baseline and every hour for 4 h following injection of levcromakalim (0.1, 0.5 and 1.0 mg/kg, i.p.) or vehicle (2% DMSO, 10 ml/kg). (a) Baseline thresholds (g) measured before administration of levcromakalim. (b) Effect of levcromakalim on every test day measured 2 h after dosing. (c) Here, data are shown for test day 5 wherein the maximal hyperalgesia mediated by levcromakalim occurs 2 h after injection, thereby supporting the choice of this timepoint for subsequent experiments. Data are square root transformed 50% withdrawal thresholds (g) and depicted as mean ± SEM.

Based on this initial dose-finding experiment, we decided to assess the ability of glibenclamide to block levcromakalim (1 mg/kg, i.p.) induced hyperalgesia in a fully powered experiment (Figure 6). In this set of experiments, levcromakalim induced a significant basal hyperalgesia from the fifth day to the end of the experimental period on the ninth day (overall effect of treatment, F(3,44) = 24.55, p < 0.0001). This basal effect was on each of these days significantly (day 5, p = 0.0162; day 7 and 9, p < 0.0001) inhibited by glibenclamide (Figure 6(a)). We found that glibenclamide completely inhibited the development of 2 h post-drug hyperalgesia on every test day (overall effect of treatment, F(3, 44) = 24.93, p < 0.0001) (Figure 6(b)). This experiment also included a glibenclamide control group to test effects of this drug alone on hindpaw hyperalgesia. This group was not different from the vehicle control group (p >0.99 on all test days) (Figure 6(b)).

Figure 6.

Hindpaw hyperalgesia in levcromakalim-sensitized mice is inhibited by glibenclamide. Levcromakalim (1 mg/kg, i.p.) or vehicle (2% DMSO, 10 ml/kg) was administered on every test day (1, 3, 5, 7 and 9) 15 min after an injection of glibenclamide (1 mg/kg, i.p.) or vehicle (5% DMSO, 10 ml/kg, i.p.). The sensitivity to mechanical stimulation was measured before (a) and 2 hrs after (b) levcromakalim dosing with von Frey filaments. Glibenclamide attenuated the basal hypersensitivity from the fifth day (a) and completely inhibited levcromakalim-induced hyperalgesia but had no effect on withdrawal thresholds on its own (b). Data are square root transformed 50% withdrawal thresholds (g) and depicted as mean ± SEM.

CGRP release from dura and TG

CGRP release has for a long time been suggested to be involved in the pathophysiology of migraine and the most recent treatment strategies of migraine (triptans) inhibit the release of CGRP or directly block the action of CGRP.

To find the most optimal concentration of capsaicin for the subsequent blockade experiments with glibenclamide, the release of CGRP was measured after cumulative application of capsaicin and its vehicle in tissue from four rats. The results revealed a significant concentration-dependent CGRP release from dura mater (F(4,22) = 39.67, p < 0.0001) and TG (F(4, 22) = 13.39, p < 0.0001) in relation to vehicle (Figure 7(b) and (c)). The maximum release of CGRP from dura mater was at 1 µM capsaicin, resulting in a 10-fold increase (n = 4) compared to vehicle (Figure 7(a)). In TG, the maximum response was measured at 10 µM capsaicin and increased 1.85-fold compared to vehicle (n = 4) (Figure 7(b)). Based on the above results, we performed blockade experiments with glibenclamide at a capsaicin concentration of 1 µM as it evoked a maximal or submaximal response in dura and TG, respectively. We used glibenclamide in a concentration of 3 µM, as it was previously shown to inhibit levcromakalim-induced rat basilar artery relaxation (42) and to sufficiently block K_ATP channels comprised of Kir6.2 and SUR1 subunits located in peripheral tissue (43,44). There was no significant difference in CGRP release between treatment with vehicle and 1 µM glibenclamide in the two tissues (dura p = 0.441; TG p = 0.881) (Figure 7(c) and (d)). Thus, glibenclamide had no effect on baseline CGRP release.

Figure 7.

Ex vivo CGRP release is inhibited by glibenclamide. ((a) and (b)) CGRP release after increasing concentrations of capsaicin in dura mater and trigeminal ganglion (TG), n = 4. ((c) and (d)) The amount of CGRP released during 10 min incubation with vehicle, 1 µM capsaicin or 1 µM capsaicin + 3µM glibenclamide in (c) dura mater and (d) trigeminal ganglion from STA rats.

In dura mater, a single concentration of 1 µM capsaicin induced a significant (p < 0.0001) release of CGRP amounting to 123 + 15 pg/ml (n = 10). In the presence of glibenclamide, the capsaicin-induced CGRP release was significantly (p < 0.031) decreased by 40% of the response to capsaicin alone (Figure 7 (c)). In TG, 1 µM capsaicin significantly (p < 0.0011) increased the CGRP release to 99 + 10 pg/ml (n = 8). In the presence of glibenclamide, the response was significantly (p < 0.003) reduced by 39% (n = 8) (Figure 7(d)). The inhibitory effect of glibenclamide on CGRP release further supports a role of K_ATP channels in migraine pathophysiology.

Discussion

The main results of the present study are that non-selective inhibitors of K_ATP channels have a significant effect in two validated animal models of migraine, the spontaneous trigeminal allodynic (STA) rat and mice sensitized by repeated injections of GTN. In the rat model, glibenclamide inhibited a chronic state of cephalic hypersensitivity while in the mouse model the 2 h post-drug effect of GTN was prevented whereas basal hyperalgesia progressed. In addition, the K_ATP channel opener levcromakalim induced hypersensitivity similar to that induced by GTN. Ex vivo studies showed that glibenclamide inhibits CGRP release from dura mater and TG. Our data suggest that K_ATP channel inhibition is a promising new target for migraine drugs.

Role of blood glucose levels for pain responses

Decreased levels of blood glucose after insulin administration were recently shown to attenuate dural-evoked neuronal firing (45). In our studies with glibenclamide, blood glucose was decreased by 55% and it could therefore be expected to have an impact on the measured mechanical thresholds. However, in experiments performed on wild type SD rats we saw no decrease in sensitivity after glibenclamide administration (Figure 2(b)). The mechanical thresholds in our study were measured at 2 h after glibenclamide administration. In the study by Martins-Oliveira (45), blood glucose levels were shown to be stably decreased at 1 h after insulin treatment. However, at 30 min after the treatment, the dural-evoked firing had already started to decline, which might be indicating the initiation of unknown adapting mechanisms. It could be speculated that the lack of an effect on mechanical thresholds in spite of decreased blood glucose levels is due to adaptive mechanisms being fully activated 2 h after administration of glibenclamide. A key point is provided in Figure 3, where it is shown that 10 and 100 mg/kg of gliquidone lowered blood glucose to the same level. However, only the 100 mg/kg dose significantly increased withdrawal responses in STA rats. Accordingly, we believe that the anti-allodynic effect observed is most likely to be mediated via K_ATP channel mediated effects within the trigeminovascular system rather than simply via lowering of blood glucose.

K_ATP channel activation causes opposite pain effects in the periphery compared to the brain

Patch clamp recordings made from cultured nociceptive neuronal preparations exemplify the role of K+ channels within neuronal pain circuits, wherein they act to dampen cellular excitability. Accordingly, the Kv7 channel opener retigabine and the K_ATP channel opener pinacidil hyperpolarize the somatic membrane potential of cultured dorsal root ganglion neurons (46). Local injection of these substances into the dorsal root ganglion diminishes hindpaw inflammatory hyperalgesia in rats produced by bradykinin (46). A further extensive literature indicates that intracerebroventricular administration of K+ channel openers mediates robust antinociception in an array of inflammatory, post-operative and neuropathic pain models (47). These data are at odds with the current data in both human and rodent models, where levcromakalim induced migraine and hyperalgesia, respectively.

So how can we explain their apparent paradoxical effects on trigeminal nociceptive processing? At this point it is impossible to distill the mechanisms evoked by levcromakalim and glibenclamide as being mediated by only one cell type. The downstream consequences mediated via these channels in, for example, vascular cells versus neurons, is ultimately different. This is a network and ultimately it is the balance of excitation versus inhibition that is measured as glibenclamide efficacy in the animal models. This probably also explains the differences between the trigeminal and spinal pain processing systems in relation to K_ATP channels. K_ATP channels are widely expressed within the brain, which theoretically could be a target site (48,49). However, there is no evidence that either levcromakalim ⁵⁰ or glibenclamide cross the blood-brain barrier. Rats implanted subcutaneously with 200 mg glibenclamide pellets (dose ∼ 50 mg/kg/day) attain a mean plasma concentration in the low micromolar range within 7–10 days, whereas levels within the brain and CSF were undetectable (51). A concentration in the micromolar range is easily sufficient to block inwardly rectifying K⁺ currents mediated via K_ATP channels comprised of Kir6.2 and SUR1 subunits located in peripheral tissues (43,44). Moreover, even when given as a bolus injection (50 mg/kg, i.p.), similar to the administration method used here, CSF levels are 1000-fold lower than in plasma (34 µg/ml) (51). Thus, it is likely that drugs such as glibenclamide act at peripherally localized K_ATP channels to diminish migraine-like pain behaviors of the cephalic pain pathways.

Vascular effects of K_ATP channel agents

Using a closed cranial window model, we have previously reported that glibenclamide attenuates dural dilatation induced by CGRP and transcranial electrical stimulation in anesthetized rats (52). This is consistent with the extensive localization of K_ATP channels throughout vascular tissues (53), wherein channel opening within smooth muscle results in membrane hyperpolarization and closure of voltage-operated Ca²⁺channels. Accordingly, the associated relaxation induces a vasodilation, which we believe can be a primary functional mechanism that promotes hyperalgesia within the trigeminovascular system. It is also consistent with human studies showing extracerebral but no cerebral vasodilation after levcromakalim infusion (50). Crucially, this mechanism appears to be variously conserved for all vasodilatory molecules irrespective of the target and/or signal transduction pathway(s) they engage.

Cellular mechanisms of action in relation to clinical effects

When administered to humans, a single dose of levcromakalim was 100% effective in migraine provocation. In contrast, in mice hypersensitivity was not observed until the third or fifth day. How can this be explained? The studies are performed in two different species and with different routes of administration. Furthermore, the endpoint values are not the same. In the human study, patients report on their headache intensity, while in mice, we investigate changes in mechanical sensitivity. Thus, the results in mice cannot be directly compared to human experiments. The time of onset of the increased sensitivity is different in mice treated with levcromakalim versus GTN. However, GTN releases the gaseous transmitter NO that rapidly diffuses into the cells to activate guanylyl cyclase and forms oxidative radicals, while levcromakalim has to diffuse to its site of action.

It is also interesting that K_ATP channel opening in man is more effective in provoking migraine than other provoking agents and likewise K_ATP channel inhibition was very effective in the animal models used in the present study. What are the possible cellular and molecular mechanisms? We hypothesize that the least effective provoking agents act on receptors in the cellular membrane (Figure 8); next come agents like NO, sildenafil and cilostazol that directly affect intracellular second messenger concentrations and, as the most efficient, comes the K_ATP channel opener levcromakalim. In the cascade of events of at least some pathways, K_ATP channels are the final common mechanism (54).

Figure 8.

K_ATP channels as the common denominator target for migraine provoking agents. Schematic representation of membrane bound and intracellular targets for migraine provoking agents. Glyceryl trinitrate (GTN) releases nitric oxide (NO), which activates soluble guanylyl cyclase (sGC) catalysing the formation of cyclic guanosine monophosphate (cGMP) that binds to protein kinase G (PKG) causing release of active catalytic subunits of PKG leading to phosphorylation and opening of K_ATP channels. This system is also affected by the phosphodiesterase 5 (not shown in the figure) inhibitor sildenafil, resulting in accumulation of cGMP. Binding of the migraine provokers calcitonin gene-related peptide (CGRP) and pituitary adenylate cyclase activating peptide (PACAP) to their receptors leads to activation of soluble adenylyl cyclase (sAC). This causes increased formation of cyclic adenosine monophosphate (cAMP) that activates protein kinase A (PKA) followed by phosphorylation of K_ATP channels. The migraine provoker cilostazol increases accumulation of cAMP by inhibition of phosphodiesterase 3 (not shown). The K_ATP channel is directly activated by levcromakalim and inhibited by glibenclamide.

K_ATP channel inhibition as a possible new treatment for migraine

Given that K_ATP channel inhibition may be effective in spontaneous migraine, it is worth speculating about the possibilities of propagating this target. As discussed above, there are several isoforms of the K_ATP channel. Glibenclamide and gliquidone are relatively unspecific and inhibit most if not all isoforms of the channel. However, gliquidone has a lower selectivity for cardiomyocytes and vascular smooth muscle cells than glibenclamide (33). This might be why gliquidone was less effective in the present study. Both glibenclamide and gliquidone have potent effects on pancreatic β-cells which are associated to SUR1 and induced hypoglycemia makes them unsuitable for migraine therapy (55). Other isoforms such as SUR2A may have cardiac side effects that should preferably be avoided (56). In previous studies, we have analyzed the presence of K_ATP channel isoforms in the cephalic vasculature and came to the result that inhibitors of the SUR2B channel subunit most likely would be preferred (42,53,57 –59). Unfortunately, no such compound with suitable in vivo pharmacokinetic properties was available for testing in the present study. Subtype specific K_ATP channel inhibitors, preferably targeting SUR2B/Kir6.1, should be developed and tested in our models and in other migraine models before clinical trials can be initiated.

Key findings

K_ATP channel inhibitors glibenclamide and gliquidone attenuate periorbital hypersensitivity in STA rats.

K_ATP channel opener levcromakalim induces hypersensitivity in mice.

Glibenclamide blocks GTN and levcromakalim induced hypersensitivity in mice.

Glibenclamide inhibits CGRP release from isolated dura mater and trigeminal ganglion.

Inhibition of K_ATP channels is suggested to be a promising target for development of new migraine drugs.

Footnotes

Author contributions

GM, DMK and SLC designed experiments with advice and guidance from JO and IJ-O. AS, SP, GM and SLC performed the experiments and analysed the data. All authors were involved in the interpretation of the data. GM, JO and IJ-O wrote the paper with contributions from all the other authors.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Candys Foundation.

References

Stovner

Nichols

Steiner

, et al. Global, regional, and national burden of migraine and tension-type headache, 1990–2016: A systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol 2018; 17: 954–976.

Linde

Gustavsson

Stovner

, et al. The cost of headache disorders in Europe: The Eurolight project. Eur J Neurol 2012; 19: 703–711.

Khan

Olesen

Ashina

CGRP, a target for preventive therapy in migraine and cluster headache: Systematic review of clinical data. Cephalalgia 2019; 39: 374–389.

Akerman

Holland

Lasalandra

, et al. Inhibition of trigeminovascular dural nociceptive afferents by Ca(2+)-activated K(+) (MaxiK/BK(Ca)) channel opening. Pain 2010; 151: 128–136.

Guo

Cao

YQ.

Over-expression of TRESK K(+) channels reduces the excitability of trigeminal ganglion nociceptors. PLoS One 2014; 9: e87029.

Lafreniere

Cader

Poulin

, et al. A dominant-negative mutation in the TRESK potassium channel is linked to familial migraine with aura. Nat Med 2010; 16: 1157–1160.

Wulf-Johansson

Amrutkar

Hay-Schmidt

, et al. Localization of large conductance calcium-activated potassium channels and their effect on calcitonin gene-related peptide release in the rat trigemino-neuronal pathway. Neuroscience 2010; 167: 1091–1102.

Belloni

Hintze

TH.

Glibenclamide attenuates adenosine-induced bradycardia and coronary vasodilatation. Am J Physiol 1991; 261: H720–H727.

Farzaneh

Tinker

Differences in the mechanism of metabolic regulation of ATP-sensitive K+ channels containing Kir6. 1 and Kir6. 2 subunits. Cardiovascul Res 2008; 79: 621–631.

10.

Hibino

Inanobe

Furutani

, et al. Inwardly rectifying potassium channels: Their structure, function, and physiological roles. Physiol Rev 2010; 90: 291–366.

11.

Kleppisch

Nelson

MT.

Adenosine activates ATP-sensitive potassium channels in arterial myocytes via A2 receptors and cAMP-dependent protein kinase. Proc Natl Acad Sci USA 1995; 92: 12441–12445.

12.

Aguilar-Bryan

Bryan

Molecular biology of adenosine triphosphate-sensitive potassium channels. Endocr Rev 1999; 20: 101–135.

13.

Brayden

JE.

Functional roles of KATP channels in vascular smooth muscle. Clin Exp Pharmacol Physiol 2002; 29: 312–316.

14.

Yamada

Inagaki

Neuroprotection by KATP channels. J Mol Cell Cardiol 2005; 38: 945–949.

15.

Al-Karagholi

Hansen

Guo

, et al. Opening of ATP-sensitive potassium channels causes migraine attacks: A new target for the treatment of migraine. Brain 2019; 142: 2644–2654.

16.

Lassen

Ashina

Christiansen

, et al. Nitric oxide synthase inhibition in migraine. Lancet 1997; 349: 401–402.

17.

Olesen

Thomsen

Lassen

, et al. The nitric oxide hypothesis of migraine and other vascular headaches. Cephalalgia 1995; 15: 94–100.

18.

Schytz

Birk

Wienecke

, et al. PACAP38 induces migraine-like attacks in patients with migraine without aura. Brain 2008; 132: 16–25.

19.

Lassen

Haderslev

Jacobsen

, et al. CGRP may play a causative role in migraine. Cephalalgia 2002; 22: 54–61.

20.

Schytz

Olesen

Ashina

The PACAP receptor: A novel target for migraine treatment. Neurotherapeutics 2010; 7: 191–196.

21.

Olesen

Diener

Husstedt

, et al. Calcitonin gene-related peptide receptor antagonist BIBN 4096 BS for the acute treatment of migraine. N Engl J Med 2004; 350: 1104–1110.

22.

Hargreaves

Olesen

Calcitonin gene-related peptide modulators – the history and renaissance of a new migraine drug class. Headache 2019; 59: 951–970.

23.

Lassen

Thomsen

Olesen

Histamine induces migraine via the H1-receptor. Support for the NO hypothesis of migraine. Neuroreport 1995; 6: 1475–1479.

24.

Worm

Falkenberg

Olesen

Histamine and migraine revisited: Mechanisms and possible drug targets. J Headache Pain 2019; 20: 30.

25.

Munro

Petersen

Jansen-Olesen

, et al. A unique inbred rat strain with sustained cephalic hypersensitivity as a model of chronic migraine-like pain. Sci Rep 2018; 8: 1836.

26.

Oshinsky

Sanghvi

Maxwell

, et al. Spontaneous trigeminal allodynia in rats: A model of primary headache. Headache 2012; 52: 1336–1349.

27.

Christensen

Petersen

Kristensen

, et al. Targeting CGRP via receptor antagonism and antibody neutralisation in two distinct rodent models of migraine-like pain. Cephalalgia 2019; 39: 1827–1837.

28.

Pradhan

Smith

McGuire

, et al. Characterization of a novel model of chronic migraine. Pain 2014; 155: 269–274.

29.

Amrutkar

Ploug

Olesen

, et al. Role for voltage gated calcium channels in calcitonin gene-related peptide release in the rat trigeminovascular system. Neuroscience 2011; 172: 510–517.

30.

Ebersberger

Averbeck

Messlinger

, et al. Release of substance P, calcitonin gene-related peptide and prostaglandin E2 from rat dura mater encephali following electrical and chemical stimulation in vitro. Neuroscience 1999; 89: 901–907.

31.

Hao

XJ.

Treatment of a chronic allodynia-like response in spinally injured rats: Effects of systemically administered excitatory amino acid receptor antagonists. Pain 1996; 66: 279–285.

32.

Zahn

Brennan

TJ.

Lack of effect of intrathecally administered N-methyl-D-aspartate receptor antagonists in a rat model for postoperative pain. Anesthesiology 1998; 88: 143–156.

33.

Zahn

Gysbers

Brennan

TJ.

Effect of systemic and intrathecal morphine in a rat model of postoperative pain. Anesthesiology 1997; 86: 1066–1077.

34.

Chaplan

Bach

Pogrel

, et al. Quantitative assessment of tactile allodynia in the rat paw. J Neurosci Methods 1994; 53: 55–63.

35.

Christensen

Hansen

Storm

, et al. Von Frey testing revisited – provision of an online algorithm for improved accuracy of 50% thresholds. Eur J Pain 2020; 24: 783–790.

36.

Jansen-Olesen

Baun

Amrutkar

, et al. PACAP-38 but not VIP induces release of CGRP from trigeminal nucleus caudalis via a receptor distinct from the PAC1 receptor. Neuropeptides 2014; 48: 53–64.

37.

Festing

Altman

DG.

Guidelines for the design and statistical analysis of experiments using laboratory animals. ILAR J 2002; 43: 244–258.

38.

Babes

Fischer

Filipovic

, et al. The anti-diabetic drug glibenclamide is an agonist of the transient receptor potential Ankyrin 1 (TRPA1) ion channel. Eur J Pharmacol 2013; 704: 15–22.

39.

Liu

Tian

Liao

, et al. The effect of gliquidone on KATP channels in pancreatic beta-cells, cardiomyocytes, and vascular smooth muscle cells. Diabetes Res Clin Pract 2015; 109: 334–339.

40.

Bates

Nikai

Brennan

, et al. Sumatriptan alleviates nitroglycerin-induced mechanical and thermal allodynia in mice. Cephalalgia 2010; 30: 170–178.

41.

Moye

Tipton

Dripps

, et al. Delta opioid receptor agonists are effective for multiple types of headache disorders. Neuropharmacology 2019; 148: 77–86.

42.

Jansen-Olesen

Mortensen

El-Bariaki

, et al. Characterization of KATP-channels in rat basilar and middle cerebral arteries: Studies of vasomotor responses and mRNA expression. Eur J Pharmacol 2005; 523: 109–118.

43.

Gribble

Ashfield

Ammala

, et al. Properties of cloned ATP-sensitive K+ currents expressed in Xenopus oocytes. J Physiol 1997; 498: 87–98.

44.

Proks

de Wet

Ashcroft

FM.

Molecular mechanism of sulphonylurea block of K(ATP) channels carrying mutations that impair ATP inhibition and cause neonatal diabetes. Diabetes 2013; 62: 3909–3919.

45.

Martins-Oliveira

Akerman

Holland

, et al. Neuroendocrine signaling modulates specific neural networks relevant to migraine. Neurobiol Dis 2017; 101: 16–26.

46.

Hao

Gigout

, et al. Control of somatic membrane potential in nociceptive neurons and its implications for peripheral nociceptive transmission. Pain 2014; 155: 2306–2322.

47.

Tsantoulas

McMahon

SB.

Opening paths to novel analgesics: The role of potassium channels in chronic pain. Trends Neurosci 2014; 37: 146–158.

48.

Thomzig

Laube

Pruss

, et al. Pore-forming subunits of K-ATP channels, Kir6.1 and Kir6.2, display prominent differences in regional and cellular distribution in the rat brain. J Comp Neurol 2005; 484: 313–330.

49.

Zhou

Tanaka

, et al. Localization of the ATP-sensitive K(+) channel regulatory subunits SUR2A and SUR2B in the rat brain. Neurosci Res 2012; 74: 91–105.

50.

Al‐Karagholi

MAM

Ghanizada

Hansen

, et al. Levcromakalim, an adenosine triphosphate‐sensitive potassium channel opener, dilates extracerebral but not cerebral arteries. Headache 2019; 59: 1468–1480.

51.

Lahmann

Kramer

Ashcroft

FM.

Systemic administration of glibenclamide fails to achieve therapeutic levels in the brain and cerebrospinal fluid of rodents. PLoS One 2015; 10: e0134476.

52.

Gozalov

Jansen-Olesen

Klaerke

, et al. Role of K ATP channels in cephalic vasodilatation induced by calcitonin gene-related peptide, nitric oxide, and transcranial electrical stimulation in the rat. Headache 2008; 48: 1202–1213.

53.

Ploug

Sorensen

Strobech

, et al. K(ATP) channels in pig and human intracranial arteries. Eur J Pharmacol 2008; 601: 43–49.

54.

Al-Karagholi

MA-M

Hansen

Severinsen

, et al. The K ATP channel in migraine pathophysiology: A novel therapeutic target for migraine. J Headache Pain 2017; 18: 90.

55.

Gribble

Ashcroft

FM.

Sulfonylurea sensitivity of adenosine triphosphate-sensitive potassium channels from beta cells and extrapancreatic tissues. Metabolism 2000; 49: 3–6.

56.

Englert

Heitsch

Gerlach

, et al. Blockers of the ATP-sensitive potassium channel SUR2A/Kir6.2: A new approach to prevent sudden cardiac death. Curr Med Chem Cardiovasc Hematol Agents 2003; 1: 253–271.

57.

Ploug

Amrutkar

Baun

, et al. K(ATP) channel openers in the trigeminovascular system. Cephalalgia 2012; 32: 55–65.

58.

Ploug

Baun

Hay-Schmidt

, et al. Presence and vascular pharmacology of KATP channel subtypes in rat central and peripheral tissues. Eur J Pharmacol 2010; 637: 10–117.

59.

Ploug

Boni

Baun

, et al. K(ATP) channel expression and pharmacological in vivo and in vitro studies of the K(ATP) channel blocker PNU-37883A in rat middle meningeal arteries. Br J Pharmacol 2008; 154: 72–81.

ATP sensitive potassium (K ATP ) channel inhibition: A promising new drug target for migraine

Abstract

Background

Methods

Results

Conclusion

Keywords

Abbreviations:

Introduction

Materials and methods

Animals

Measurement of glucose levels

Periorbital pain thresholds in STA rats

Hindpaw pain thresholds in STA rats

Assessment of motor function in rats

Induced hindpaw hyperalgesia in mice

CGRP release ex vivo

Drugs

Statistical analyses

Results

KATP channel inhibition in STA rats

KATP channel inhibition in mice

CGRP release from dura and TG

Discussion

Role of blood glucose levels for pain responses

KATP channel activation causes opposite pain effects in the periphery compared to the brain

Vascular effects of KATP channel agents

Cellular mechanisms of action in relation to clinical effects

KATP channel inhibition as a possible new treatment for migraine

Key findings

Footnotes

Author contributions

Declaration of conflicting interests

Funding

References

K_ATP channel inhibition in STA rats

K_ATP channel inhibition in mice

K_ATP channel activation causes opposite pain effects in the periphery compared to the brain

Vascular effects of K_ATP channel agents

K_ATP channel inhibition as a possible new treatment for migraine